Annual Plant Reviews, Plant Polysaccharides: Biosynthesis and Bioengineering (Volume 41)

ANNUAL PLANT REVIEWS VOLUME 41 Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering ©...

Author: Peter Ulvskov

39 downloads 1016 Views 80MB 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

ANNUAL PLANT REVIEWS VOLUME 41

Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

ANNUAL PLANT REVIEWS VOLUME 41 Plant Polysaccharides, Biosynthesis and Bioengineering

Edited by

Peter Ulvskov Department of Plant Biology & Biotechnology Copenhagen University Denmark

A John Wiley & Sons, Inc., Publication

This edition first published 2011 © 2011 Blackwell Publishing Ltd. Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing programme has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Registered office John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK Editorial offices 9600 Garsington Road, Oxford, OX4 2DQ, UK The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK 2121 State Avenue, Ames, Iowa 50014-8300, USA For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell. The right of the author to be identified as the author of this work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988. 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, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. Library of Congress Cataloging-in-Publication Data Plant polysaccharides / edited by Peter Ulvskov. p. cm. – (Annual plant reviews ; v. 41) Includes bibliographical references and index. ISBN 978-1-4051-8172-3 (hardcover : alk. paper) 1. Polysaccharides. chemistry. I. Ulvskov, Peter. II. Series: Annual plant reviews ; v. 41. QK898.P77P53 2011 572’.5662–dc22 2010027203

2. Botanical

A catalogue record for this book is available from the British Library. This book is published in the following electronic formats: ePDF (9781444390995); Wiley Online Library (9781444391015); ePub (9781444391008) Set in 10/12 pt Palatino by Toppan Best-set Premedia Limited

1

2011

Annual Plant Reviews A series for researchers and postgraduates in the plant sciences. Each volume in this series focuses on a theme of topical importance and emphasis is placed on rapid publication. Editorial Board: Prof. Jeremy A. Roberts (Editor-in-Chief), Plant Science Division, School of Biosciences, University of Nottingham, Sutton Bonington Campus, Loughborough, Leicestershire, LE12 5RD, UK; Dr David Evans, School of Biological and Molecular Sciences, Oxford Brookes University, Headington, Oxford, OX3 0BP; Prof. Hidemasa Imaseki, Obata-Minami 2419, Moriyama-ku, Nagoya 463, Japan; Dr Michael T. McManus, Institute of Molecular BioSciences, Massey University, Palmerston North, New Zealand; Dr Jocelyn K.C. Rose, Department of Plant Biology, Cornell University, Ithaca, New York 14853, USA. Titles in the series: 1. Arabidopsis Edited by M. Anderson and J.A. Roberts 2. Biochemistry of Plant Secondary Metabolism Edited by M.Wink 3. Functions of Plant Secondary Metabolites and their Exploitation in Biotechnology Edited by M.Wink 4. Molecular Plant Pathology Edited by M. Dickinson and J. Beynon 5. Vacuolar Compartments Edited by D.G. Robinson and J.C. Rogers 6. Plant Reproduction Edited by S.D. O’Neill and J.A. Roberts 7. Protein–Protein Interactions in Plant Biology Edited by M.T. McManus, W.A. Laing and A.C. Allan 8. The Plant CellWall Edited by J.K.C. Rose 9. The Golgi Apparatus and the Plant Secretory Pathway Edited by D.G. Robinson 10. The Plant Cytoskeleton in Cell Differentiation and Development Edited by P.J. Hussey 11. Plant–Pathogen Interactions Edited by N.J. Talbot 12. Polarity in Plants Edited by K. Lindsey 13. Plastids Edited by S.G. Moller 14. Plant Pigments and their Manipulation

15. Membrane Transport in Plants Edited by M.R. Blatt 16. Intercellular Communication in Plants Edited by A.J. Fleming 17. Plant Architecture and Its Manipulation Edited by C.G.N. Turnbull 18. Plasmodeomata Edited by K.J. Oparka 19. Plant Epigenetics Edited by P. Meyer 20. Flowering and Its Manipulation Edited by C. Ainsworth 21. Endogenous Plant Rhythms Edited by A. Hall and H. McWatters 22. Control of Primary Metabolism in Plants Edited by W.C. Plaxton and M.T. McManus 23. Biology of the Plant Cuticle Edited by M. Riederer 24. Plant Hormone Signaling Edited by P. Hadden and S.G. Thomas 25. Plant Cell Separation and Adhesion Edited by J.R. Roberts and Z. Gonzalez-Carranza 26. Senescence Processes in Plants Edited by S. Gan 27. Seed Development, Dormancy and Germination Edited by K.J. Bradford and H. Nonogaki 28. Plant Proteomics Edited by C. Finnie 29. Regulation of Transcription in Plants Edited by K. Grasser 30. Light and Plant Development Edited by G. Whitelam 31. Plant Mitochondria Edited by D.C. Logan 32. Cell Cycle Control and Plant Development Edited by D. Inzé 33. Intracellular Signaling in Plants Edited by Z. Yang 34. Molecular Aspects of Plant Disease Resistance Edited by J. Parker 35. Plant Systems Biology Edited by G.M. Coruzzi and R.A. Gutiérrez 36. The Moss Physcomitrella patens Edited by C.D. Knight, P-F. Perroud and D.J. Cove 37. Root Development Edited by T. Beeckman 38. Fruit Development and Seed Dispersal Edited by L. Østergaard

39. Functions and Biotechnology of Plant Secondary Metabolites Edited by M. Wink 40. Biochemistry of Plant Secondary Metabolism Edited by M. Wink 41. Plant Polysaccharides Edited by P. Ulvskov 42. Nitrogen Metabolism in Plants in the Post-genomic Era Edited by C. Foyer and H. Zhang 43. Biology of Plant Metabolomics Edited by R.D. Hall

CONTENTS

Preface Dedication Contributors 1

Cell Wall Polysaccharide Composition and Covalent Crosslinking Stephen C. Fry 1.1 Remit 1.1.1 Some definitions 1.1.1.1 Pectins 1.1.1.2 Hemicelluloses 1.1.1.3 Crosslinks 1.1.1.4 Non-polysaccharide components 1.1.1.5 Primary wall 1.1.1.6 Secondary wall 1.2 The classic primary cell walls of dicots 1.2.1 Pectins 1.2.1.1 Homogalacturonan domains 1.2.1.2 Rhamnogalacturonan-I domains 1.2.1.3 Rhamnogalacturonan-II domains 1.2.1.4 Xylogalacturonan domains 1.2.2 Hemicelluloses 1.2.2.1 Xyloglucans 1.2.2.2 Xylans 1.2.2.3 Mannans 1.2.2.4 Glucuronomannans 1.2.3 Cellulose 1.3 Secondary cell walls 1.4 Taxonomic consideration of primary cell walls 1.4.1 Poalean primary cell walls 1.4.1.1 Poalean xyloglucans 1.4.1.2 Poalean (feruloylated) xylans 1.4.1.3 Poalean mixed-linkage glucans 1.4.1.4 Other poalean polysaccharides 1.4.2 Taxonomically restricted features of non-poalean angiosperm walls 1.4.3 Cell walls of non-angiosperms 1.4.3.1 Charophytic algae 1.4.3.2 Bryophytes

xix xxiii xxv 1 2 2 2 3 3 3 3 6 6 6 7 8 12 12 12 13 15 15 15 16 16 18 18 18 19 19 20 21 21 21 22 ix

x

■

Contents

1.5

1.6

1.7

2

1.4.3.3 Lycopodiophytes 1.4.3.4 Euphyllophytic pteridophytes 1.4.3.5 Gymnosperms Covalent bonds between wall polysaccharides 1.5.1 Glycosidic bonds joining polysaccharides into molecular ‘trees’ 1.5.2 Glycosidic bonds forming true (‘lateral’) crosslinks between polysaccharides? 1.5.3 Oxidative coupling products as crosslinks or intrapolymeric loops 1.5.4 Uronoyl esters and uronoyl amides 1.5.5 Borate diesters Methodology 1.6.1 Specific and non-specific radiolabelling in vivo 1.6.2 Chemical and enzymic ‘dissection’ of wall polysaccharides 1.6.3 Fractionation and characterization of mono- and oligosaccharides 1.6.3.1 Paper chromatography (PC) 1.6.3.2 Thin-layer chromatography (TLC) 1.6.3.3 Paper electrophoresis (PE) 1.6.3.4 High-pressure liquid chromatography (HPLC) 1.6.3.5 Methylation analysis and gas chromatography (GC) 1.6.3.6 Mass spectrometry (MS) Conclusions Acknowledgements References

Dissection of Plant Cell Walls by High-throughput Methods Staffan Persson, Iben Sørensen, Isabel Moller, William Willats and Markus Pauly 2.1 Introduction 2.2 Enzyme fingerprinting 2.3 Structural determination of oligosaccharides 2.4 Fourier transform infrared spectroscopy (FTIR) 2.5 Microarray-based polymer profiling 2.6 Additional high-throughput methods 2.7 Future perspectives References

3 Approaches to Chemical Synthesis of Pectic Oligosaccharides Sergey A. Nepogodiev, Robert A. Field and Iben Damager 3.1 Introduction

23 24 24 24 25 26 27 27 28 29 29 32 33 33 34 34 35 35 35 36 36 36 43

44 44 47 50 52 55 57 58 65 66

Contents

3.2

3.3 3.4

3.5

3.6 3.7 3.8

Pectic polysaccharides: structures and availability of fragments from natural sources 3.2.1 Structures 3.2.2 Preparation of pectic oligosaccharides by cleavage of polysaccharides 3.2.2.1 Homogalacturonan fragments 3.2.2.2 Rhamnogalacturonan-I fragments Reported preparations of pectic oligosaccharides by chemical synthesis Oligosaccharide synthesis – basic principles and key features 3.4.1 General points 3.4.2 Main considerations in chemical synthesis of complex oligosaccharides and polysaccharides Synthesis of homogalacturonan fragments 3.5.1 Synthesis of oligogalacturonides by direct glycosylation with galacturonic acid derivatives 3.5.2 Synthesis of oligogalacturonides by a late stage oxidation approach 3.5.2.1 The convergent block synthesis approach 3.5.2.2 The reiterative synthesis strategy 3.3.2.3 Synthesis of selectively methyl-esterified oligogalacturonic acids Rhamnogalacturonan-II fragments Rhamnogalacturonan-I fragments Future perspective References

4 Annotating Carbohydrate-active Enzymes in Plant Genomes: Present Challenges Pedro M. Coutinho and Bernard Henrissat 4.1 Introduction 4.2 CAZy: what’s behind the name? 4.3 Plant CAZymes: the quest for ‘function’ 4.4 Plant CAZymes: problems in functional annotation References 5

Biosynthesis of Plant Cell Wall and Related Polysaccharides by Enzymes of the GT2 and GT48 Families Bruce A. Stone, Andrew K. Jacobs, Maria Hrmova, Rachel A. Burton and Geoffrey B. Fincher 5.1 Introduction 5.2 Structures and distribution of β-d-glucans synthesized by GT2 and GT48 enzymes

■

xi

66 66 67 67 68 69 71 71 71 74 75 76 76 78 78 81 86 88 89

93 93 96 97 103 105

109

110 112

xii

■

Contents

Early biochemical approaches to plant β-d-glucan synthases 5.4 Functional genomics and the identification of GT2 cellulose synthases 5.4.1 Cellulose synthesis by embryophytes 5.4.2 Cellulose synthesis by Gluconoacetobacter xylinus (formerly Acetobacter xylinus) 5.4.3 Cellulose synthesis by Agrobacterium tumefaciens 5.5 Identification of the functions of other GT2 enzymes from plants 5.6 Comparative genomics and the identification of GT2 (1,3;1,4)-β-d-glucan synthases 5.7 Genes for GT2 synthases for bacterial (1,3)-β-d-glucans and related polysaccharides 5.7.1 Curdlan 5.7.2 Cyclic (1,3;1,6)-β-d-glucan from Bradyrhizobium japonicum 5.7.3 Capsular polysaccharides from Pasteurella multocida 5.7.4 Chitin 5.8 Enzymic properties and catalytic mechanisms of the GT2 proteins 5.8.1 Topology 5.8.2 The catalytic region 5.8.3 3D structures 5.8.4 Catalytic mechanisms 5.8.5 Specification of linkage position in β-glycans 5.8.6 Bifunctional GT2 β-Glycan Synthases 5.8.7 Chain initiation, direction of chain elongation, and chain termination 5.9 Subcellular locations of GT2 enzymes in plants 5.10 Proteomics and biochemical approaches to the identification of GT48 (1,3)-β-d-glucan synthases from plants 5.11 Enzymic properties of the GT48 proteins 5.11.1 Topology 5.11.2 Catalytic mechanisms 5.12 Future role of biochemistry in the characterization of GT2 and GT48 enzymes 5.12.1 Biochemistry and the definition of gene function 5.12.2 Post-translational modifications of glycosyl transferases 5.12.3 Protein–protein interactions 5.13 Applications of modified levels of plant β-d-glucans Acknowledgements References 5.3

119 121 121 124 125 125 127 130 130 131 131 131 133 133 133 134 137 138 139 140 142

142 146 146 147 147 148 149 150 151 153 153

Contents

6

7

■

Glycosyltransferases of the GT8 Family Yanbin Yin, Debra Mohnen, Ivana Gelineo-Albersheim, Ying Xu and Michael G. Hahn 6.1 Introduction 6.2 Phylogeny of family GT8 6.3 GT8 clades related to plant cell wall polysaccharide synthesis 6.3.1 Galacturonosyltransferase (GAUT) clade 6.3.1.1 Phylogeny of GAUT clade 6.3.1.2 Function of GAUT proteins 6.3.2 Galacturonosyl transferase-like (GATL) clade 6.3.2.1 Phylogeny of GATL clade 6.3.2.2 Function of GATL proteins 6.4 GT8 clades not related to cell wall synthesis 6.4.1 Plant glycogenin-like (PGSIP) clades 6.4.1.1 Phylogeny of PGSIP clades 6.4.1.2 Function of PGSIP proteins 6.4.2 Galactinol synthase (GolS) clade 6.4.2.1 Phylogeny of GolS clade 6.4.2.2 Function of GolS proteins 6.5 Conclusions Acknowledgements References Genes and Enzymes of the GT31 Family: Towards Unravelling the Function(s) of the Plant Glycosyltransferase Family Members Jack Egelund, Miriam Ellis, Monika Doblin, Yongmei Qu and Antony Bacic 7.1 Introduction 7.2 Identification and characterization of the first β-(1,3)-GalTs 7.2.1 The first β-(1,3)-GT family members of non-plant origin 7.2.2 The first β-(1,3)-GalTs of plant origin 7.3 Grouping of accessions based on their phylogenetic relationship 7.4 Conserved motifs and implications for catalysis 7.4.1 Motif of the catalytic domain belongs to the GT-A fold superfamily 7.4.2 Sequence analysis and functional assignment of conserved motifs 7.4.3 Implication of different substrate specificity for β-(1,3)-GTs 7.5 Domains conserved within the plant-specific clades 7.5.1 Galactosyltransferase domain-containing clades 7 and 10

xiii 167

168 170 188 188 189 190 192 193 193 196 197 198 200 201 202 204 205 205 205

213

214 215 215 216 216 221 221 222 223 223 225

xiv

■

Contents

7.5.2

7.6

8

9

10

Galactoside-binding lectin (galectin) domain specific to clade 7 7.5.3 Domain of unknown function (DUF604) in clade 1 7.5.4 No identifiable domains in accessions of clade 11 Conclusions Acknowledgements References

Glycosyltransferases of the GT34 and GT37 Families Kenneth Keegstra and David Cavalier 8.1 Introduction 8.2 Family GT37 enzymes 8.2.1 Xyloglucan fucosyltransferase (FUT1) 8.2.2 Other FUT genes 8.3 Family GT34 enzymes 8.3.1 Galactomannan galactosyltransferase 8.3.2 Xyloglucan xylosyltransferase 8.4 Concluding comments References Glycosyltransferases of the GT43 Family Nadine Anders and Paul Dupree 9.1 Introduction 9.2 GT43 glycosyltransferases in plants – putative β-1,4-xylosyltransferases 9.3 GT43 glycosyltransferases in animals – β-1,3-glucuronosyltransferases 9.4 Structural characteristics of GT43 proteins 9.5 Concluding remarks References Glycosyltransferases of the GT47 Family Naomi Geshi, Jesper Harholt, Yumiko Sakuragi, Jacob Krüger Jensen and Henrik Vibe Scheller 10.1 Introduction 10.2 Phylogenetic analysis of CAZy GT47 10.3 Group A 10.4 Group D 10.5 Group B 10.6 Group C 10.7 Subcellular localization and protein–protein interactions 10.8 Conclusion References

225 227 228 228 229 229 235 235 236 236 240 241 241 243 246 247 251 251 255 256 257 259 260 265

266 266 269 272 275 277 278 280 280

Contents

11

12

13

The Plant Glycosyltransferase Family GT64: in Search of a Function Ellinor Edvardsson, Sunil Kumar Singh, Min-Soo Yun, Agata Mansfeld, Marie-Theres Hauser and Alan Marchant 11.1 Introduction 11.2 GT64 family members are found in a diverse range of species 11.3 The Arabidopsis GT64 family 11.3.1 Structure of the GT64 genes and proteins 11.3.2 Function of Arabidopsis GT64s proteins 11.3.2.1 At3g55830 EPC1 11.3.2.2 At1g80290 EPC-L1 11.3.2.3 At5g04500 EPC-L2 11.4 Possible activities of the plant GT64 enzymes 11.4.1 Could EPC1 function in AGP synthesis? 11.4.2 Could EPC1 play a role in FLA biosynthesis? 11.4.3 EPC1 as a negative regulator of ABA signalling 11.5 Concluding remarks References

■

xv

285

286 286 287 287 290 290 295 295 296 297 298 298 299 299

Glycosyltransferases of the GT77 Family Bent Larsen Petersen, Kirsten Faber and Peter Ulvskov 12.1 Introduction 12.2 The oldest cell wall 12.3 Pfam and fold prediction 12.4 Establishing GT77 12.4.1 The Dictyostelium discoideum (1,3)-α-d-galactosyltranferase 12.4.2 Clade B – rhamnogalacturonan-II biosynthesis 12.4.3 Clade A – the mixed algal and higher plant clade 12.4.4 Clade C and PfamB 13934 12.4.5 Clade D 12.4.6 Clade E – the youngest clade 12.5 Discussion Acknowledgments References

305

Hydroxyproline-rich Glycoproteins: Form and Function Marcia J. Kieliszewski, Derek T.A. Lamport, Li Tan and Maura C. Cannon 13.1 Introduction 13.1.1 Background 13.1.2 Definitions 13.2 Post-translational modifications 13.2.1 Rules for proline hydroxylation

321

305 306 307 311 312 312 315 317 317 317 318 318 319

322 322 322 323 323

xvi

■

Contents

13.2.2

13.3

13.4

13.5

14

15

O-Hyp-glycosylation codes – the Hyp contiguity hypothesis 13.2.3 Glycosylation enhances secretion of designer glycoproteins 13.2.4 Glycosylation anomalies 13.2.5 Crosslinking code involving tyrosine motifs Molecular function, biological role 13.3.1 Arabinogalactan proteins (AGPs) 13.3.2 Extensins as self-assembling amphiphiles 13.3.3 Role of multiple extensins Evolution 13.4.1 Conserved motifs 13.4.2 Evolution of the cell plate – nascent cell wall Epilogue Acknowledgments References

Plant Cell Wall Biology: Polysaccharides in Architectural and Developmental Contexts Maureen C. McCann and J. Paul Knox 14.1 Introduction 14.2 Plant cell wall biology basics 14.3 Analytical tools to study cell wall microstructures and the diversity of cell wall architectures 14.3.1 Molecular probes 14.3.2 Advances in microscopies 14.3.3 Spectroscopic and spectrometric technologies 14.4 Cell wall architectures: primary cell walls 14.4.1 Modulation of cellulose–hemicellulose networks and cell enlargement 14.4.2 Pectins: modular multifunctional polymers of the primary cell wall matrix 14.5 In vitro polysaccharide composites 14.6 Cell wall diversity 14.7 Cell wall architectures: secondary cell walls 14.8 Prospects for plant cell wall biology Acknowledgements References Enzymatic Modification of Plant Cell Wall Polysaccharides Jens Øbro, Takahisa Hayashi and Jørn Dalgaard Mikkelsen 15.1 Introduction 15.2 In vivo modifications 15.3 Post-harvest modifications 15.4 Perspectives

324 327 328 329 329 329 330 332 334 334 335 336 336 336

343 344 344 346 346 349 350 350 352 353 355 355 358 360 360 360 367 368 369 376 380

Contents

■

Acknowledgments References 16

17

Production of Heterologous Storage Polysaccharides in Potato Plants Xing-Feng Huang, Jean-Paul Vincken, Richard G.F. Visser and Luisa M. Trindade 16.1 Introduction 16.2 Starch: native and modified starch, consequences for its properties 16.2.1 Starch biosynthesis in storage organs 16.2.2 Alteration of starch composition through the modeling of plant genes 16.2.2.1 Amylose-free starch 16.2.2.2 High-amylose starch 16.2.3 Modification of starch properties by expression of bacterial proteins in plants 16.3 Production of novel storage polysaccharides in plants 16.3.1 Glucans synthesized in transgenic plants 16.3.1.1 Alternan 16.3.1.2 Dextran 16.3.1.3 Mutan 16.3.1.4 Other novel carbohydrates 16.3.2 Fructans synthesized heterologously in transgenic potato plants 16.3.2.1 Enzymes of fructan biosynthesis 16.3.2.2 Industrial applications of fructans 16.3.2.3 Fructans in transgenic plants 16.4 Final remarks References Glycan Engineering in Transgenic Plants Muriel Bardor, José A. Cremata and Patrice Lerouge 17.1 Introduction 17.2 N-glycosylation: a major post-translational modification of secreted proteins 17.2.1 N-Glycosylation of plant-derived pharmaceuticals: antibodies as a glycoprotein model 17.2.2 Differences in N-glycan structure may compromise in vivo use of plant-derived therapeutic proteins 17.3 Strategies for glycan engineering in transgenic plants 17.3.1 Retention in the ER 17.3.2 In planta reconstruction of human-like N-glycans by knock-in strategies

xvii 381 382

389

390 390 391 391 391 393 393 394 394 395 396 396 398 398 399 400 400 402 403 409 409 410 412

413 415 415 415

xviii

■

Contents

17.3.3

17.4

18

Removal of immunogenic plant epitopes by knock-out strategies 17.3.4 In planta sialylation of recombinant proteins Conclusions Acknowledgements References

Polysaccharide Nanobiotechnology: A Case Study of Dental Implant Coating Marco Morra, Clara Cassinelli, Giovanna Cascardo, Hanna Kokkonen, Juha Tuukkanen, Claudio Della Volpe, Stefano Siboni, Giordano Segatta, Marco Brugnara and Giacomo Ceccone 18.1 Introduction: titanium dental implants and surface modifications 18.2 Rationale for the surface modification of titanium dental implants by nanolayers of MHRs 18.2.1 Background 18.2.2 Modification of solid surfaces by MHRs: general principles 18.2.3 Experiments on osteogenic cell adhesion to MHR-coated surfaces 18.2.4 Wettability of MHR-coated surfaces 18.3 Surface modification of titanium dental implants by MHRs 18.3.1 Surface modification of titanium by MHR: surface chemical analysis 18.3.2 Surface modification of titanium by MHRs: wettability measurements on machined and rough surfaces 18.4 Reflections and conclusions Acknowledgments References

Index

417 418 419 420 421

425

426 428 428 429 431 434 438 438

443 444 446 446 451

PREFACE

This book introduces students and researchers to the polysaccharides and proteins that form the fundamental architecture of the plant cell wall, and the to the genes encoding the cellular machinery that synthesizes them. In assembling the chapters in the core of this book, we focus on the evolution of the many families of genes whose products are required to make a particular kind of polysaccharide, rather than describing all the gene products needed to make certain classes of polysaccharides. In so doing, we draw attention to the specific biochemical properties of the proteins, at the level of kinds of sugar linkages they make. Cell wall structure has, for the same reason, been split into chapters on composition, the building blocks to be synthesized, and architecture which relates to wall assembly. Several gene products from different families are often needed to make a complex polysaccharide, and the protein elements of the synthase complex interact to produce the characteristic product. This collection of articles is designed to encourage thinking as to how this occurs in the dynamic context of cell growth and development. As most of the cell wall is carbohydrate, most of the gene families annotated to possible function are involved in polysaccharide synthesis. Chapter 1 by Stephen Fry gives us an up-to-date comprehensive inventory of the classes of specific polysaccharides, drawing attention to the simple but effective methods to monitor their synthesis in vivo. The repertoire of covalent bonds in this inventory of cell wall structural components allows us to hypothesize about the kinds and minimum number of enzymes needed to make them. Persson and colleagues (Chapter 2) then provide a summary of numerous new tools to analyse these constituents, such as immunocytochemistry, oligosaccharide fingerprinting, and carbohydrate microarrays, which will fine-tune our capabilities to chemically phenotype subtle alterations in polysaccharide structures caused by mutation. Nepogodiev and colleagues (Chapter 3) describe the current knowledge in the chemical synthesis of pectic oligosaccharides, which will undoubtedly prove useful in the fingerprinting of mutant phenotypes and to provide novel substrates for biosynthesis. One of the most valuable resources to cell wall researchers has been CAZy – the Carbohydrate-Active Enzymes database (http://www.cazy.org/) – which defines the known world of glycosyl transferases, hydrolases, lyases, and carbohydrate binding modules across the tree of life. Of 92 distinct gene families of glycosyl tranferases described in CAZy, 42 are represented in plants, although many of these are not involved directly in cell wall xix

xx

■

Preface

biosynthesis. Coutinho and Henrissat (Chapter 4) describe their respective groups’ continuing efforts to annotate the genes involved in encoding the polysaccharide synthesis machinery that is essential to cell wall biogenesis. Moreover, this chapter defines the genes of cell wall synthesis in their evolutionary context of enzymatic function, and in so doing they provide an inventory of proteins that constitute polysaccharide synthases essential for systems approaches to integrate Golgi and plasma membrane synthase mechanisms. This book is not exhaustive of gene families, but several chapters cover the major families and the specific functions of their gene products in the synthesis of cell wall-related specific polysaccharides. The basic tenet is that no one family is responsible for the synthesis of a single class of complex polysaccharide, such as pectin or cross-linking glycan, but for the synthesis of a general class of anomeric linkages that are part of them. In the last article to be published by the esteemed Bruce Stone and his colleagues (Chapter 5), they describe the history of discovery of linkage structure, of biosynthesis, and of discovery of the genes encoding the catalytic components of the processive synthases of the (1→3)-β-d-glucan, callose, from family GT48, and the synthesis of (1→4)-β-d- and (1→3),(1→4)-β-d-glucans, in cellulose and the mixed-linkage glucans, from the cellulose- and cellulose-synthaselike superfamily within GT2. This chapter serves as a paradigm of discovery at all levels of biochemistry and molecular biology, and so it is fitting that we dedicate this book to Bruce, and also to Debby Delmer, a long-time biochemist who contributed a vast amount of knowledge on the biochemistry of cellulose synthesis and who discovered the first cellulose synthase gene in cotton (see dedication); they are among the leading pioneers who took polysaccharide chemistry and synthesis into the molecular genetics age. Two of the major families of non-processive glycosyl transferases are GT8 and GT47. As Yin et al. (Chapter 6) explain, the GT8 family encodes enzymes of the retaining type, i.e. enzymes that retain the α-d- or β-l- linkage of the nucleotide sugar as an α-d- or β-l-glycosyl linkage. Thus, it is not surprising that this family encodes GAUT1, the enzyme that synthesizes the α-d-GalA linkage found in homogalacturonan (HG) from UDP-α-d-GalA. In contrast, as Geshi et al. Chapter 10) explain, the GT47 family encodes enzymes of the inverting type, which means that β-d- and α-l- linked sugars are made from their respective α-d- and β-l- linked nucleotide sugars. Both of these chapters illustrate how cell wall phenotypes in mutants knocked out in these GTs are now being used as clues to infer the biochemical activity of members of the GT family. We are beginning to achieve success in defining the function of a few members of many other GT families as well, as illustrated in the chapters by Egelund and colleagues on GT31 (Chapter 7), whose members are thought to function in O-glycosylation of proteins, such as the arabinogalactan proteins, and by Keegstra and Cavalier (Chapter 8) on GT34, whose members include retaining xylosyl and galactosyltransferase involved in xyloglucan

Preface

■

xxi

and galactomannan synthesis, respectively, and GT37, whose members include a fucosyl transferase responsible for completing the characteristic trisaccharide sidegroup of xyloglucan but is otherwise largely unknown. Anders and Dupree (Chapter 9) describe recent knowledge of family GT43 and its potential role of at least some of its members in xylan biosynthesis, and Peterson and colleagues (Chapter 12) describe an ancient family, GT77, whose members extend from microalgae to flowering plants, possibly encoding xylosyl and arabinosyl transferases to produce complex pectins and Olinked structural glycoproteins. Edvardsson and colleagues report on family GT64 (Chapter 11), whose members, based on animal models, are transferases of aminosugars and their N-acetylated derivatives, but whose specific functions in plants are still to be resolved. The later chapters begin to bring together the polysaccharide synthases and glycosyl transferases into the big picture of how plant cell walls are made. Kieliszewski and colleagues (Chapter 13) focus on the structural proteins that function to interlock the cell wall through interaction with the carbohydrate components, and hence bring together the gene families hypothesized to be involved in protein O-glycosylation, GT31 and GT77. A group of proteins that carry the DUF266 domain are also hypothesized to be involved in O-glycosylation. No chapter is devoted to these, as they have not yet been included in the CAZy classification. With biosynthesis thus covered, McCann and Knox (Chapter 14) assemble a wall from the elements and review how these elements are dynamic during development, and they provide excellent insight into how cytochemical tools are used to define regional domain-specific walls, even in the wall surrounding a single cell. The book concludes with a short bioegineering section. Øbro and colleagues (Chapter 15) and Huang et al. (Chapter 16) provide the history of engineering cell wall polysaccharides and storage polysaccharides in planta by plant or microbial enzymes, as well as giving insights into the next generation of products that are possible. Researchers wanting to characterize an unannotated GT often cannot know whether they will end up working on polysaccharide biosynthesis, protein O-glycosylation or even on protein N-glycosylation — latter will take the researcher out of the cell wall area. Chapter 17 by Bardor and colleagues deals with N-glycosylation but does so from a bioengineering perspective, thus demonstrating that knowledge of plant GTs may allow us to engineer plant cells to produce protein-based pharmaceuticals with humanstyle glycosylation. Bioengineering of cell wall polysaccharides to target medical applications is an area in its infancy. Morra and his colleagues are pioneers in this field and in Chapter 18 they discuss the possibilities of using rhamnogalacturonan-I as a versatile starting material for engineering biocompatible coating materials for implants and medical devices; in doing so, they also touch upon analytical methods that could find broader uses in cell wall biology.

xxii

■

Preface

Until recently, plant cell wall biologists regarded their work as a means to understand growth and development and gain useful information about the polysaccharides themselves. This information is integral to our understanding of plant stature and form, but advances in this knowledge are now part of grander, more immediate challenges to meet not only the world’s needs for sustainable food and fibre, but also to provide new crops whose cell walls are the feedstock for efficient production of alternative carbonneutral fuels. Understanding the genes of cell wall formation is only the initial knowledge base. The next step will be to understand transcript and protein networks and learn how the protein products of these genes form functional complexes. Extending a systems-level understanding to cell wall formation will identify the bottlenecks to productivity at the metabolic and cellular level. Only one-half of the genes of cell wall biogenesis are even hazily understood, and new high-throughput tools in phenotyping, from mass spectrometry of protein complexes to molecular imaging, will be needed to define the complete set. We have a long way to go, and the challenge to researchers new to this field is to use this knowledge to move forward more briskly. Nicholas Carpita & Peter Ulvskov

DEDICATION

Cell wall giants Debbie Delmer and Bruce Stone at the cell wall meeting in Toulouse, 2001. Debbie Delmer pioneered the molecular approach to studying polysaccharide biosynthesis, a strategy that led to her discovery of the genes encoding cellulose synthases of family GT2. The GT2 chapter in this book is the last to be written by Bruce Stone (1928–2008) and thus caps a remarkable career in science with major achievements in elucidating both structure and biosynthesis of polysaccharides. His achievements in cereal mixed-linkage glucans research and contributions to the discovery of arabinogalactan proteins in particular stand out. Bruce Stone was an avid traveller and frequently visited laboratories of the cell wall community where he inspired young scientists and impressed both young and old with his total recall of decades of polysaccharide research.

xxiii

CONTRIBUTORS

Nadine Anders Department of Biochemistry University of Cambridge Building O Downing Site Cambridge CB2 1QW UK Antony Bacic Plant Cell Biology Research Centre School of Botany The University of Melbourne Victoria 3010 Australia Muriel Bardor UPRES-EA 4358 IFRMP 23 Université de Rouen 76821 Mont Saint Aignan Cédex France Marco Brugnara Department of Materials Engineering and Industrial Technologies University of Trento Via Mesiano 77 I-38050 Trento Italy Rachel A. Burton Australian Centre for Plant Functional Genomics University of Adelaide Waite Campus Glen Osmond SA 5064 Australia

xxv

xxvi

■

Contributors

Maura C. Cannon Department of Biochemistry and Molecular Biology 913 Lederle Graduate Research Tower University of Massachusetts Amherst MA 01003 USA Nicholas C. Carpita Purdue University Botany & Plant Pathology and Bindley Bioscience Center 915 West State Street West Lafayette, IN 47907-2054 USA Giovanna Cascardo Nobil Bio Ricerche Via Valcastellana 26 14037 Portacomaro (AT) Italy Clara Cassinelli Nobil Bio Ricerche Via Valcastellana 26 14037 Portacomaro (AT) Italy David Cavalier MSU-DOE Plant Research Laboratory Michigan State University Plant Biology Building East Lansing MI 48824 USA Giacomo Ceccone Institute for Health and Consumer Protection European Commission Joint Research Center Ispra Italy

Contributors

Pedro M. Coutinho Architecture et Fonction des Macromolécules Biologiques UMR6098, CNRS Universities of Aix-Marseille I and II Case 932 163 Avenue de Luminy 13288 Marseille Cédex 9 France José A. Cremata Department of Carbohydrate Chemistry CIGB Havana Cuba Jørn Dalgaard Mikkelsen Department of Chemical and Biochemical Engineering Technical University of Denmark Søltofts Plads Building 229 2800 Kgs. Lyngby Denmark Iben Damager Department of Plant Biology and Biotechnology Faculty of Life Sciences University of Copenhagen Thorvaldsensvej 40 1871 Frederiksberg Denmark Claudio Della Volpe Department of Materials Engineering and Industrial Technologies University of Trento Via Mesiano 77 I-38050 Trento Italy Monika Doblin Plant Cell Biology Research Centre School of Botany The University of Melbourne Victoria 3010 Australia

■

xxvii

xxviii

■

Contributors

Paul Dupree Department of Biochemistry University of Cambridge Building O Downing Site Cambridge CB2 1QW UK Ellinor Edvardsson Department of Forest Genetics and Plant Physiology Swedish University of Agricultural Sciences (SLU) SE-901 83 Umeå Sweden Jack Egelund University of Copenhagen Department of Molecular Biology Building 4-2-15 Ole Maaløes Vej 5 2200 Copenhagen Denmark Miriam Alexandra Ellis Plant Cell Biology Research Centre School of Botany The University of Melbourne Victoria 3010 Australia Kirsten Faber Department of Plant Biology and Biotechnology Faculty of Life Sciences University of Copenhagen Thorvaldsensvej 40 1871 Frederiksberg Denmark Robert A. Field Department of Biological Chemistry John Innes Centre Norwich Research Park Colney Norwich NR4 7UH UK

Contributors

■

xxix

Geoffrey B. Fincher Australian Centre for Plant Functional Genomics University of Adelaide Waite Campus Glen Osmond SA 5064 Australia Stephen C. Fry The Edinburgh Cell Wall Group Institute of Molecular Plant Sciences School of Biological Sciences Daniel Rutherford Building The King’s Buildings Edinburgh EH9 3JH UK Ivana Gelineo-Albersheim Complex Carbohydrate Research Center and BioEnergy Science Center The University of Georgia 315 Riverbend Road Athens GA 30602 USA Naomi Geshi Department of Plant Biology and Biotechnology Faculty of Life Sciences University of Copenhagen Thorvaldsensvej 40 1871 Frederiksberg Denmark Michael G. Hahn Complex Carbohydrate Research Center and BioEnergy Science Center Department of Plant Biology The University of Georgia Athens GA 30602 USA Jesper Harholt Department of Plant Biology and Biotechnology Faculty of Life Sciences University of Copenhagen Thorvaldsensvej 40 1871 Frederiksberg Denmark

xxx

■

Contributors

Marie-Theres Hauser Institute of Applied Genetics and Cell Biology BOKU – University of Natural Resources and Applied Life Sciences Muthgasse 18 A-1190 Vienna Austria Takahisa Hayashi Research Institute for Sustainable Humanosphere Kyoto University Uji Kyoto 611-0011 Japan Bernard Henrissat Architecture et Fonction des Macromolécules Biologiques UMR6098, CNRS Universities of Aix-Marseille I and II Case 932 163 Avenue de Luminy 13288 Marseille Cédex 9 France Maria Hrmova Australian Centre for Plant Functional Genomics University of Adelaide Waite Campus Glen Osmond SA 5064 Australia Xing-Feng Huang Graduate School Experimental Plant Sciences Wageningen UR – Plant Breeding Wageningen University and Research Center P.O. Box 386 6700 AJ Wageningen The Netherlands Andrew K. Jacobs Australian Centre for Plant Functional Genomics University of Adelaide Waite Campus Glen Osmond SA 5064 Australia

Contributors

Kenneth Keegstra MSU-DOE Plant Research Laboratory Michigan State University Plant Biology Building East Lansing MI 48824 USA Marcia J. Kieliszewski Department of Chemistry and Biochemistry Ohio University Athens OH 45701 USA J. Paul Knox Centre for Plant Sciences Biological Sciences University of Leeds Leeds LS2 9JT UK Hanna Kokkonen Department of Anatomy and Cell Biology University of Oulu PO BOX 5000 90014 Oulu Finland Jacob Krüger Jensen Department of Plant Biology and Biotechnology Faculty of Life Sciences University of Copenhagen Thorvaldsensvej 40 1871 Frederiksberg Denmark Present address: MSU-DOE Plant Research Laboratory Michigan State University Plant Biology Building East Lansing MI 48824 USA

■

xxxi

xxxii

■

Contributors

Derek T.A. Lamport Biology and Environmental Science University of Sussex Falmer Brighton BN1 9RH UK Bent Larsen Petersen Department of Plant Biology and Biotechnology Faculty of Life Sciences University of Copenhagen Thorvaldsensvej 40 1871 Frederiksberg Denmark Patrice Lerouge UPRES-EA 4358 IFRMP 23 Université de Rouen 76821 Mont Saint Aignan Cédex France Agata Mansfeld Institute of Applied Genetics and Cell Biology BOKU – University of Natural Resources and Applied Life Sciences Muthgasse 18 A-1190 Vienna Austria Alan Marchant School of Biological Sciences Life Sciences Building 85 University of Southampton Highfield Campus, Southampton SO17 1BJ UK Maureen C. McCann Biological Sciences Purdue University West Lafayette IN 47907 USA

Contributors

■

xxxiii

Debra Mohnen Complex Carbohydrate Research Center and BioEnergy Science Center Department of Biochemistry and Molecular Biology The University of Georgia Athens GA 30602 USA Isabel Moller Department of Biology University of Copenhagen Ole Maaløes vej 5 2200 Copenhagen N Denmark Marco Morra Nobil Bio Ricerche Via Valcastellana 26 14037 Portacomaro (AT) Italy Sergey A. Nepogodiev Department of Biological Chemistry John Innes Centre Norwich Research Park Colney Norwich NR4 7UH UK Jens Øbro Department of Biology University of Copenhagen Ole Maaløes vej 5 2200 Copenhagen N Denmark Present address: Novozymes A/S Krogshøjvej 36 2880 Bagsværd Denmark

xxxiv

■

Contributors

Markus Pauly Department of Plant and Microbial Biology Energy Biosciences Institute University of California, Berkeley Berkeley CA 94720 USA Staffan Persson Max-Planck-Institute for Molecular Plant Physiology Am Muehlenberg 1 14476 Potsdam Germany Yongmei Qu Plant Cell Biology Research Centre School of Botany The University of Melbourne Vic 3010 Australia Yumiko Sakuragi Department of Plant Biology and Biotechnology Faculty of Life Sciences University of Copenhagen Thorvaldsensvej 40 1871 Frederiksberg Denmark Henrik Vibe Scheller Department of Plant Biology and Biotechnology Faculty of Life Sciences University of Copenhagen Thorvaldsensvej 40 1871 Frederiksberg Denmark Present address: Joint BioEnergy Institute Lawrence Berkeley National Laboratory 5885 Hollis St Emeryville CA 94608 USA

Contributors

Giordano Segatta Department. of Materials Engineering and Industrial Technologies University of Trento Via Mesiano 77 I-38050 Trento Italy Stefano Siboni Department. of Materials Engineering and Industrial Technologies University of Trento Via Mesiano 77 I-38050 Trento Italy Sunil Kumar Singh Department of Forest Genetics and Plant Physiology SLU SE-901 83 Umeå Sweden Present address: Department of Plant Cell and Molecular Biology Indian Institute of Advanced Research (IIAR) Koba Gandhinagar-382007 India Iben Sørensen Department of Biology University of Copenhagen Ole Maaløes vej 5 2200 Copenhagen N Denmark Bruce A. Stone School of Biochemistry La Trobe University Bundoora Vic 3083 Australia Li Tan Department of Chemistry and Biochemistry Ohio University Athens OH 45701 USA

■

xxxv

xxxvi

■

Contributors

Luisa M. Trindade Graduate School Experimental Plant Sciences Wageningen UR – Plant Breeding Wageningen University and Research Center P.O. Box 386 6700 AJ Wageningen The Netherlands Juha Tuukkanen Department of Anatomy and Cell Biology University of Oulu PO BOX 5000 90014 Oulu Finland Peter Ulvskov Department of Plant Biology and Biotechnology Faculty of Life Sciences University of Copenhagen Thorvaldsensvej 40 1871 Frederiksberg Denmark Jean-Paul Vincken Graduate School Experimental Plant Sciences Wageningen UR – Plant Breeding Wageningen University and Research Center P.O. Box 386 6700 AJ Wageningen The Netherlands Present address: Laboratory of Food Chemistry Wageningen University P.O. Box 8129 6700 EV Wageningen The Netherlands Richard G.F. Visser Graduate School Experimental Plant Sciences Wageningen UR – Plant Breeding Wageningen University and Research Center P.O. Box 386 6700 AJ Wageningen The Netherlands

Contributors

William Willats Department of Biology University of Copenhagen Ole Maaløes vej 5 2200 Copenhagen N Denmark Ying Xu Computational Systems Biology Lab Institute of Bioinformatics and BioEnergy Science Center Department of Biochemistry and Molecular Biology The University of Georgia Athens GA 30602 USA Yanbin Yin Computational Systems Biology Lab Institute of Bioinformatics and BioEnergy Science Center Department of Biochemistry and Molecular Biology The University of Georgia Athens GA 30602 USA Min-Soo Yun Department of Forest Genetics and Plant Physiology Swedish University of Agricultural Sciences (SLU) SE-901 83 Umeå Sweden Present address: Plant Genetic Engineering Research Unit National Institute of Agrobiological Sciences 2-1-2 Kannondai Tsukuba Ibaraki 305-8602 Japan

■

xxxvii

Annual Plant Reviews (2011) 41, 1–42 doi: 10.1002/9781444391015.ch1

Chapter 1

CELL WALL POLYSACCHARIDE COMPOSITION AND COVALENT CROSSLINKING Stephen C. Fry The Edinburgh Cell Wall Group, Institute of Molecular Plant Sciences, The University of Edinburgh, The King’s Buildings, Edinburgh EH9 3JH, UK Manuscript received February 2008

Abstract: Genetics now potentially lets us modify the production, crosslinking and degradation of cell wall polysaccharides. There remains, however, the need to test experimentally whether intended modifications of polysaccharide metabolism have successfully been effected in vivo. Simple methods for this are described, including in-vivo radiolabelling, enzymic dissection (e.g. with Driselase) and chromatographic/electrophoretic fractionation of dissection products. After an overview of polysaccharide chemistry, I discuss the structures and taxonomic distribution of wall polysaccharides in charophytes and land plants. Primary and secondary walls are compared. The major wall polysaccharides are cellulose [microfibrillar β-(1→4)-d-glucan], pectins (α-d-galacturonate-rich) and hemicelluloses (lacking galacturonate; hydrogen-bonding to cellulose; extractable by 6 M NaOH at 37 °C). Land-plant pectins are anionic polymers built of about four glycosidically interconnected domains (homogalacturonan, rhamnogalacturonans I and II, xylogalacturonan). Hemicelluloses occurring in most/all land plants are α-xylo-β-glucans, β-xylans (including α-arabino-β-xylans, α-glucurono-β-xylans, etc.) and β-mannans (including α-galacto-β-mannans, β-gluco-β-mannans, etc.). Another hemicellulose [mixed-linkage β-(1→3)(1→4)-d-glucan) is confined to Equisetum and some Poales. Other taxonomically restricted features of angiosperm primary walls occur in Poales (xylose-poor xyloglucans; feruloylated arabinoxylans); Solanales and Lamiales (characteristic xyloglucans); Caryophyllales (feruloylated pectins); and Alismatales (apiogalacturonan). I also summarize characteristic wall features of charophytes, bryophytes, lycopodiophytes, fern-allies and gymnosperms. Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

1

2

■

Plant Polysaccharides

The making or breaking of a ‘crosslink’ (defined as an individual chemical bond, not a whole ‘tethering’ chain) may cause wall tightening/loosening. Covalent crosslinks include phenolic coupling products, uronoyl esters and amides, and borate diesters. Keywords: angiosperms; apiogalacturonan; bryophytes; cellulose; charophytes; chromatography; crosslinks (covalent); Driselase; electrophoresis (paper); gymnosperms; hemicellulose; homogalacturonan; hydrolysis; mannans; mixed-linkage glucans; pectins; polysaccharide chemistry; primary wall; pteridophytes; radiolabelling; rhamnogalacturonans; secondary wall; taxonomic variation; uronoyl esters/amides; xylans; xylogalacturonan; xyloglucan

1.1 Remit This chapter discusses the main cell wall polysaccharides of streptophytes – i.e. land plants (embryophytes, from liverworts to angiosperms) plus charophytes (a group of algae sharing many subcellular features with land plants). Primary structures (sequences of sugar residues) and covalent crosslinks are discussed. A brief introduction to the vocabulary of polysaccharide (‘glycan’) chemistry, and the sugar abbreviations used, is included in the legend to Fig. 1.1; these abbreviations are used throughout the text without further definition. Further details of secondary and tertiary structures of polysaccharides can be found in Chapter 14. A major theme in the present chapter is the application of simple analytical methods by which polysaccharides can be identified, characterized and quantified, and their metabolism monitored in vivo. Another theme is the taxonomic distribution of various wall polysaccharides. Works complementing this chapter include Brett & Waldron (1996), Fry (2000), Schols & Voragen (2002), Mort (2002), O’Neill & York (2003) and Obel et al. (2006). 1.1.1 Some definitions The best-known cell wall component is cellulose – a highly insoluble polysaccharide, of which the microfibrils (‘scaffolding’) of the wall are composed. Cellulose is, however, only one of many polysaccharides found in plant cell walls, usually accounting for less then half the wall’s dry mass. The other, non-cellulosic, wall polysaccharides (matrix components) are categorized into pectins and hemicelluloses. 1.1.1.1 Pectins These were traditionally defined by their extractability from the wall with chelating agents, often with the assistance of heating (though this inevitably causes partial degradation of pectins and should not be used if a determination of molecular weight is planned) and often followed by ice-cold aqueous Na2CO3. A more acceptable definition of pectins is wall polysaccharides rich in α-GalA residues.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

3

1.1.1.2 Hemicelluloses These are not extracted by chelating agents or ice-cold Na2CO3, but are by concentrated aqueous alkali. They generally share the property of hydrogenbonding to cellulose, at least in vitro; and some hemicelluloses probably also do this in muro, tethering adjacent microfibrils (Fry 1989; Hayashi 1989). For this reason, the term ‘crosslinking glycans’ was suggested for hemicelluloses. However, this term is not used here because the proposed inmuro tethering role remains largely hypothetical in many cases, and also because some non-hemicellulosic polysaccharides (e.g. rhamnogalacturonan II, RG-II) do crosslink. 1.1.1.3 Crosslinks A ‘crosslink’, as the term is used here, is an individual chemical bond, e.g. an ester linkage or a hydrogen bond, that joins together two otherwise separate polymers; it is not a whole molecular chain that joins together two structures (e.g. a xyloglucan chain tethering two microfibrils). Definitions of polysaccharide classes by their extractability from the wall are far from perfect. One problem is that some polysaccharides chemically identical to hemicelluloses but not hydrogen-bonded to cellulose can sometimes be solubilized with hot neutral water – for example, the bulk xyloglucan present in some seeds as ‘food reserves’. Another problem is that some hemicelluloses are covalently attached to pectins, resulting in hybrid polysaccharides that are difficult to classify. Nevertheless, the broad classification of wall polysaccharides into pectins, hemicelluloses and cellulose remains a useful convention. 1.1.1.4 Non-polysaccharide components Also important in cell walls are non-polysaccharide components. First among these is water, typically accounting for around 60% of the wall’s total fresh weight, and around 70% of the fresh weight of the wall matrix (Monro et al. 1976). Water confers important physical properties on (hydrated) wall polysaccharides, acts as a solvent for apoplastic solutes and enables the functioning of wall-located enzymes. Changes in the water content of the matrix may explain changes in wall extensibility (Ulvskov et al. 2005; Thompson 2008). Other wall constituents are highly variable between tissues, developmental stages and taxa: (glyco)proteins (e.g. extensins and arabinogalactan-proteins); the phenolic polymer lignin; the polyesters cutin and suberin; highly resistant cutan and sporopollenin; and silica (Fry 2001). 1.1.1.5 Primary wall A primary wall layer is one whose cellulosic microfibrils were laid down while the cell was still (capable of) growing. Once deposited and the cell has stopped growing, a primary wall layer will not acquire more cellulose, although in certain cell types it later becomes impregnated with for example lignin or cutin. Such a wall layer is still ‘primary’, even if lignified.

4

■

Plant Polysaccharides

COOH

CH2OH

O

HO

O

HO

OH

CH3

OH

O

HO OH

OH

OH

HOH2C

OH α-D-Galacturonic acid (α -GalA)

OH β-D-Galactose (β-Gal)

OH OH α-L -Rhamnose (α-Rha)

OH α-L -Arabinofuranose (α-Ara) COOH

CH2OH

CH2OH O

O

OH

OH

O

OH

HO OH β-D-Glucose (β-Glc)

O

OH

OH HO

OH

HO

OH

O

OH

OH

HO OH β-D-Xylose (β-Xyl)

OH

HO

OH α-D-Glucuronic acid (α-GlcA)

β-D-Mannose (β-Man)

COOH O

O

OH CH3

CH2OH HO

OH

O

HO

O

OH OH OH

OH

HO

HO

HO OH α-L -Galactose (α-L -Gal)

O CH2OH

OH α-L -Arabinopyranose (α-Ara p)

D-lyxo -5-Hexosulose

uronic acid (Hua)

CH2OH

OH

O H3C

COOH

HO

COOH HO

O

HO

COOH

O

C

O CH3

OH OH OH

Methyl α-Dgalacturonate [ester] (MeGalA)

O H3C

[2-keto]-3-Deoxy-β-Dlyxo -heptulosaric acid (β-Dha)

COOH

OH OH

4-O-Methyl- α-Dglucuronic acid [ether] (MeGlcA)

O

C

O

OH

O

HO

OH O

OH

COOH

OH

O

OH

OH

[2-Keto]-3-deoxy-α-Dmanno-octulosonic acid (α -Kdo)

COOH

O

HO

O

HO

OH OH α-L -Aceric acid (α-AceA)

OH OH β-D-Apiose (β-Api)

HO

HO OH α-L -Fucose (α-Fuc)

OH

OH

CH3

OH OH

3-O-Acetyl-α-Dgalacturonic acid [ester] (3AcGalA)

H2C O

OH

O 5-O-Feruloylα-L -arabinose [ester] (Fer-Ara)

O CH3 OH

Figure 1.1 Monosaccharide building blocks (shown as Haworth formulae) of plant cell wall polysaccharides. The figure shows all the known sugar residues of plant cell wall polysaccharides and a selection of their esters and ethers. Top row, major components of pectins; 2nd row, major components of hemicelluloses; 3rd row, minor sugars of various origins; 4th row, mainly or only known from RG-II; bottom row, a selection of sugars with non-carbohydrate substituents. Sugars with five and six C atoms are called pentoses and hexoses respectively; Rha and Fuc are deoxyhexoses. The monosaccharides are shown as hemiacetal or hemiketal rings. However, within a polysaccharide, each sugar (except one, the reducing terminus) is present as an acetal or ketal residue, the term ‘residue’ implying that it is ‘what remains’ after losing the –OH group (shown in blue) from the anomeric carbon (the anomeric carbon is the one with single-bonds to two oxygens; it is here drawn as the right-hand extremity of the hexagon or pentagon). In a sugar residue of a polysaccharide, this particular –OH group has departed (in the form of H2O), ‘taking with it’ one oxygen-linked H atom from the next sugar along the polysaccharide chain. The one sugar of the polysaccharide that is not strictly a residue is the reducing terminus, so called because it has not lost its anomeric –OH group and in aqueous solution can therefore equilibrate with the straight-chain form, which possesses an oxo group (C=O, which has reducing properties). All but two of the sugars shown are aldoses (i.e. the anomeric carbon has only one additional C atom attached to it), but Kdo and Dha are ketoses (the anomeric C is attached to two other carbons). (Hua has two anomeric carbons (C-1 and C-5) and is both an aldose and a ketose.) In aqueous solution, each illustrated hemiacetal and hemiketal equilibrates with a small percentage of a straight-chain form possessing an oxo group (an aldehyde or ketone, in aldoses and ketoses respectively) – hence the slightly redundant term ‘keto’ in the names of Kdo and Dha. Each named sugar could theoretically occur as two isomeric forms (enantiomers, designated D- and L-), distinguished by the orientation of the C–O bond of the penultimate C atom. Galactose is the only wall residue known to occur as both D- and L-enantiomer. Note that D- and L-Gal differ in orientation of the C–O bond at all four non-anomeric, chiral centres (= carbons 2, 3, 4 and 5; the difference at C-5 is indicated by the placement of the –CH2OH group). The linkage between a sugar residue and the next building-block along a polysaccharide chain can be in either of two isomeric forms (anomers, designated α- and β-) defined by the orientation of the bond between the anomeric C atom and the oxygen atom (shown in blue) that bridges the two sugars: if this C–O bond has the same orientation as that of the penultimate C atom, then the residue is α-; if opposite, β-. This means that, in these Haworth formulae, the –OH of the anomeric carbon points down in α-D- and β-L-sugars, and up in β-D- and α-L-sugars. The sugar ring can be 6-membered (pyranose; -p) or 5-membered (furanose; -f). Api and AceA must be -f because of the absence of an oxygen on a C-5, and MeGlcA can only be -p. Ara occurs in both forms. All the others could theoretically occur in either form, but in practice occur only in the -p form illustrated. Each sugar residue is attached, via its anomeric carbon, to an –OH group on the following sugar unit in the polysaccharide chain. Usually, there are several such –OH groups to choose from (e.g., in the case of Glcp, on carbons 2, 3, 4 or 6: the linkage is designated (1→2), (1→3), (1→4) or (1→6), accordingly). However, a given sugar unit (either a residue or the reducing terminus) can and often does have more than one sugar residue attached to it. Once it has become part of a polysaccharide chain, a given sugar residue is ‘locked’ in one of the four possible ring forms (α-p, β-p, α-f, or β-f). These ring forms have a huge impact on the polysaccharide, as is obvious from the enormous differences in physical, chemical and biological properties between amylose and cellulose (which are α-p and β-p, respectively, but otherwise identical). Although illustrated here in unionized form, the free carboxy groups (–COOH, shown in red) would often be negatively charged (–COO−) under physiological conditions of pH. Relatively hydrophobic (non-polar) groups are shown in green. Abbreviations: The diagrams show (in parentheses) the shorthand used throughout this chapter. Thus, unless otherwise stated in the text, the ring-form (-p or -f) and enantiomer (D- or L-) are assumed to be as illustrated here; for example, ‘β-Gal’ implies β-D-Galp unless specified as L-Gal. Other abbreviations used (not illustrated): MeXyl, 2-O-methyl-α-D-Xylp (ether); MeFuc, 2-O-methyl-α-L-Fucp (ether); MeRha, 3-O-methyl-α-L-Rhap (ether); MeGal, 3-O-methyl-D-Galp (ether); 5AcAra, 5-O-acetyl-L-Araf (ester); 6AcGal, 6-O-acetyl-D-Galp (ester); 6AcGlc, 6-O-acetylD-Glcp (ester); ΔUA, a 4,5-unsaturated, 4-deoxy derivative of GalA or GlcA.

6

■

Plant Polysaccharides

1.1.1.6 Secondary wall A secondary wall layer is one whose microfibrils were laid down after the cell had lost the ability to expand. Since microfibrils are laid down adjacent to the plasma membrane, it follows that the primary wall is external to the secondary wall and that the most recently deposited layer of secondary wall is innermost. Xylem vessel elements, tracheids, sclerenchyma and cork are cell types depositing much secondary wall material. Like primary wall layers, secondary layers sometimes later become impregnated with nonpolysaccharide substances, typically lignin and suberin.

1.2 The classic primary cell walls of dicots For many years, rapidly growing and dividing suspension-cultured cells of sycamore (Acer pseudoplatanus) were the paradigm for primary cell wall studies. They do, however, have certain unusual features, e.g. an exaggerated extensin content (∼20% of the dry weight) and – probably associated – resistance to digestion by ‘protoplasting’ enzymes. Not all primary walls are chemically identical. There is taxonomic variation, surveyed later in this chapter. In addition, there are distinct differences between different primary walls within a given plant. Celery parenchyma, for example, has an unusually low xyloglucan content (Thimm et al. 2002). Changes in wall polysaccharide biosynthesis also occur during the cell cycle (Amino et al. 1984). Another example is provided by the duckweed Spirodela, which has a high proportion of apiogalacturonan in the walls of its vegetative fronds but almost none in its turions (organs of perennation). The frond:turion ratios (of rates of de-novo synthesis of polysaccharide residues) were 165:1 for β-Api, 11.5:1 for α-Rha and 1.7:1 for α-GalA. Thus, compared with fronds, developing turions continued to synthesize pectin, but much less rhamnogalacturonan and almost no apiogalacturonan (Longland et al. 1989). 1.2.1 Pectins Much useful work on pectin characterization can be performed on whole isolated cell walls or alcohol-insoluble residue (AIR) (see Section 1.6). However, if necessary, pectins can be extracted from some tissues (especially ripe fruits) with chelating agents at neutral pH and 20 °C. This does not apply to most other plant tissues, especially actively growing ones. Extraction can be increased by heating, but with partial depolymerization. Heating at around pH 4 is least detrimental in this respect; at this pH, oxalate is a more effective chelator than EDTA or EGTA. Oxalate is also easier than EDTA, EGTA and hexametaphosphate to remove (e.g. by dialysis) after pectin extraction. Another agent increasing some pectins’ extractability,

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

7

and maintaining their solubility, is 0.5 M imidazole (pH 7) (Zhang et al. 2007). The ‘pectins’ described below (homogalacturonan, RG-I, RG-II, etc.) are probably conjoined domains, linked end-to-end by glycosidic bonds into a single polysaccharide; in muro each ‘pectin’ is not a separate polysaccharide in its own right. For example, high-Mr pectin extracted from sugar-beet roots with aqueous imidazole was degraded to RG-I, RG-II and free GalA by polygalacturonases (which hydrolyse homogalacturonan but not rhamnogalacturonans; Ishii & Matsunaga 2001). The simplest explanation is that the pectin was a long chain with RG-I and RG-II domains arranged like beads along a homogalacturonan ‘string’. Further support for the domain concept came from the characterization of oligosaccharides released by mild acid hydrolysis, including GalA-GalA-GalA-Rha-GalA-Rha, which appears to be a fragment straddling part of a homogalacturonan domain and part of an RG-I domain (Coenen et al. 2007). 1.2.1.1 Homogalacturonan domains These quantitatively major domains comprise an unbranched chain of anionic α-GalA and uncharged MeGalA (= methyl ester of α-GalA) residues joined by (1→4)-bonds. The MeGalA residues tend to occur contiguously, as neutral blocks, along the homogalacturonan chain; this arrangement probably arises by the progressive action of pectin methylesterase on a more fully methylesterified precursor chain. The backbone also includes some 2AcGalA and/or 3AcGalA residues, which may affect the polysaccharide’s solubility but not its charge. After de-esterification, e.g. by ice-cold dilute alkali, homogalacturonan can be cleaved by endo-polygalacturonase (EPG) to small fragments with degree of polymerization (DP) 1–3, which serve as a simple quantitative assay for this domain. Homogalacturonans are water soluble at neutral and alkaline pH, but insoluble at mildly acidic pH (especially after removal of any methyl ester groups) and in the presence of Ca2+. De-esterified homogalacturonans are digested by ‘Driselase’ (a mixture of numerous endo- and exo-hydrolases from the fungus Irpex lacteus; Fry 2000), giving a quantitative yield of free GalA. Driselase is preferred over acid hydrolysis for this purpose because (1) uronosyl linkages are only slowly acid-hydrolysed, and (2) the free GalA liberated is considerably less stable in hot acid than naturally occurring neutral monosaccharides, so recovery is low. Driselase also contains esterases that convert MeGalA to GalA unless a nearby O-acetyl group interferes. Driselase does not remove acetyl groups from homogalacturonan, and can thus yield structurally informative oligosaccharides containing AcGalA residue(s) (Perrone et al. 2002). Commercial homogalacturonan (‘polygalacturonic acid’) is a useful model, e.g. for practising analyses, but the commercial material is of much lower Mr than natural homogalacturonan.

8

■

Plant Polysaccharides

1.2.1.2 Rhamnogalacturonan-I domains EPG digestion of walls or AIR solubilizes rhamnogalacturonan-I (RG-I). This high-Mr pectic domain (typically DP 1000–2000) has a backbone of the repeating disaccharide [–4)-α-GalA-(1→2)-α-Rha-(1→–]. To O-4 of roughly half the Rha residues, diverse neutral side chains are attached, which are rich in β-Gal (more abundant close to the Rha residue) and α-Ara (more abundant at the side-chains’ non-reducing termini). Some very high-Mr side chains of RG-I (rich in Gal and/or Ara) may enable this pectic domain to hydrogen-bond to cellulose, a property not exhibited by most pectins (Zykwinska et al. 2005). Minor sugar residues also occur (Table 1.1). There are very many different side chains, their individual structures probably defined stochastically.

Table 1.1 Summary composition of plant cell wall polysaccharides

Linkage(s) within backbone

Main residue(s) of side chains

Main linkages between side chain and backbone

Polysaccharide

%a

Qb

Residue (s) of backbone

Microfibrillar Cellulose

30

0

βGlc

(1→4)

none

15

–

αGalA (±Me ester, ±Ac ester)

(1→4)

none

Rhamnogalacturonan- 10 I domain

–

αGalA (± Ac ester), αRha

[-GalA(1→2)-Rha(1→4)-]n

βGal, αAra > αFuc, βXyl, βGlcA

Gal-(1→4)-Rha, Ara-(1→4)-Rha

αGalA (no Me-esters)

(1→4)

αAceA, βApi, αAra, αArap, βAra, βDha, αFuc, βGal, l-Gal, αGalA, βGalA, βGlcA, α(?)Kdo, αMeFuc, αMeXyl, αRha, βRha

Api-(1→2)GalA, Kdo-(1→3)GalA, Dha-(2→3)GalA, Ara-(1→3)GalA

Matrix Pectins Homogalacturonan domain

Rhamnogalacturonan- 1–4 – II domain

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

9

Two useful enzymes will cleave the backbone of RG-I into structurally informative fragments: rhamnogalacturonan lyase attacks near the reducing end, releasing oligosaccharides such as ΔUA-Rha-GalA-Rha; rhamnogalacturonan hydrolase attacks near the non-reducing end, releasing oligosaccharides such as GalA-Rha-GalA-Rha-GalA (Mutter et al. 1998). RG-I lacks methyl esters, but it does have 2AcGalA, 2,3-Ac2GalA (Ishii 1997) and 3AcGalA residues (Perrone et al. 2002), which can be isolated as components of Rha-containing oligosaccharides after Driselase digestion. De-esterified RG-I is thoroughly hydrolysed by Driselase, giving an almost 100% yield of the constituent monosaccharides. RG-I preparations from potato and soyabean are available commercially (Megazyme); gum karaya is also rich in RG-I.

Frequent linkages within side chains

Major product(s) of Driselase digestion

Other enzymes that cleave backbone

Glc

Other notes Water-insoluble. Hydrogen-bonded within microfibril

GalA + MeOH, except where acetylated

Endo-polygalacturonase (EPG)

Very few (no?) Rha residues interrupting the backbone. Some GalA residues Meesterified, some 2- and/ or 3-O-acetylated, some both

Gal-(1→4)-Gal, Ara-(1→5)-Ara, Ara-(1→3)-Ara, Ara-(1→3)-Gal

GalA (except where acetylated), Rha, Gal, Ara, Fuc

RG lyase, RG hydrolase

Some GalA residues 2- or 3-O-acetylated, or 2,3-di-O-acetylated. EPG-resistant

complex; five main types of sidechain; see text

Trace of GalA and GlcA?; largely resisant to Driselase, with borate retained

none known

Some αRha present as 3-O-Me ether in some pteridophytes. MeFuc and AceA are Oacetylated. One Api residue can stably esterify with borate, crosslinking two RG-II molecules. EPG-resistant

10

■

Plant Polysaccharides

Table 1.1 Continued

Residue (s) of backbone

Linkage(s) within backbone

Main residue(s) of side chains

Main linkages between side chain and backbone

Polysaccharide

%a

Qb

Xylogalacturonan domain

±

–

αGalA (± Me-ester)

(1→4)

βXyl > αFuc

Xyl-(1→3)GalA

Apiogalacturonan (domain?)

0

–

αGalA

(1→4)

βApi, (Xyl)

Xyl-(1→2)GalA, Xyl-(1→3)GalA

Hemicelluloses Xyloglucans

20

0

βGlc (± Ac ester)

(1→4)

αXyl, βGal (± Ac ester), αFuc (±αAra, βAra, βXyl, α-l-Gal)

αXyl-(1→6)Glc, αAra-(1→2)Glc, βXyl-(1→2)-Glc

Xylans

8

–

βXyl (± Ac ester)

(1→4)

αAra, αGlcA, (±βXyl, β-d-Gal, ?-l-Gal, …)

Ara-(1→2)-Xyl, Ara-(1→3)-Xyl, GlcA-(1→2)-Xyl

Mannans

±

0

βMan

(1→4)

±αGal

Gal-(1→6)Man

Glucomannans

±

0

βMan (± Ac ester), βGlc

(1→4)

±αGal

Gal-(1→6)Man

Mixed-linkage glucans (MLGs)

0

0

βGlc

(1→4), (1→3)

none

Callose

±

0

βGlc

(1→3)

none?

Glucuronomannans

±

–

βGlcA, αMan

[-GlcAαAra, βGal, (1→2)-Man- αFuc, Xyl (1→4)-]n

To O-3 of Man and GlcA

Ac, acetyl; Me, methyl. See Fig. 1.1 for further abbreviations. Unless otherwise indicated, the sugar is the enantiomer (d- or l-) and ring-form (-p or -f ) illustrated in Fig. 1.1. ±, not always present; ∼, amount varies greatly. a Rough guide to amount of polymer present, as % of dry weight of a typical dicot primary cell wall from a rapidly growing cell culture; b Charge on polymer molecule (at physiological pH): –, negative; 0, uncharged.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

Frequent linkages within side chains

Major product(s) of Driselase digestion

Other enzymes that cleave backbone

■

11

Other notes

Some Xyl-Xyl and/ or Fuc-Xyl.

?

Xylogalacturonan hydrolase

Major side-chains = single Xyl; some disaccharides

Api-(1→3)-Api

?

?

Only in certain monocots. Major side-chains = single Api; some disaccharide

Gal-(1→2)-Xyl, Fuc-(1→2)-Gal, α-Ara-(1→2)-Xyl

Isoprimeverose, Glc, Gal, Fuc, Ara

Endo-β-glucanase (cellulase), xyloglucan endo-glucanase (XEG)

See text for repeat units. Some 6AcGlc, 6AcGal and 5AcAra

Xyl-(1→2)-Ara, Ara-(1→2)-Ara, Xyl-(1→3)-Xyl

Xylobiose, Xyl, Ara, Fer-Ara-Xyl, Fer-Ara-Xyl-Xyl. No free GlcA?

Xylanases

Some 2AcXyl or 3AcXyl. In Poales, some Fer-Ara and 2AcAra

Man, Gal, oligosaccharides.

Mannanase

Heavily galactosylated mannans in some seeds. Little studied in primary cell walls

Man-Glc, ManMan, Man, Glc

Mannanase

Well known in secondary walls of xylem. Some Man residues 2- or 3-O-acetylated

Glc

Lichenase, cellulase

Only in Poales and Equisetum.

Glc

Laminarinase

Mainly in wounded tissues, wallregenerating protoplasts, and phloem sieve-tubes

?

[Acid hydrolysis yields GlcA-Man disaccharide]

Some GlcA may be Me-esterified

12

■

Plant Polysaccharides

1.2.1.3 Rhamnogalacturonan-II domains RG-II is a small but exceedingly complex pectic domain. It has a backbone of at least eight α-GalA residues, to which are attached five different types of side chain: the first four are designated A–D, each with a precisely (not stochastically) defined primary structure (O’Neill et al. 2004). The side chains typically have the following composition: A (octasaccharide), α-l-Gal, β-GlcA, α-MeXyl, α-Fuc, β-Rha, α-GalA, β-GalA, β-Api(⊗) B (octa- or nonasaccharide), β-Ara, α-Rha (×1 or ×2), α-Arap, β-Gal, α-MeFuc, α-AceA, β-Rha, β-Api(⊗) C (disaccharide), α-Rha, α(?)-Kdo(⊗) D (disaccharide), β-Ara, β-Dha(⊗). where (⊗) indicates the residue attached to the oligo-GalA backbone. All four of these side chains are themselves acidic (unlike those of RG-I) and contain very unusual residues, including some known only from RG-II. The fifth side chain is a single α-Ara residue (Melton et al. 1986). Side chain B has acetyl groups on the AceA and MeFuc residues (O’Neill et al. 2004). There are no reports of methyl esters in RG-II. If RG-II has one copy of each side chain, it is DP 30 (∼5 kDa). The dimerization of RG-II by borate crosslinks is discussed later. The linkages between the side-chains and RG-II’s backbone are unusually acid-labile, especially the Api→GalA* and Kdo→GalA bonds; the side chains can therefore be pruned off the backbone by warm dilute acid. However, RG-II is largely resistant to Driselase, which provides a useful method for its purification. A convenient source from which to purify RG-II is red wine, since the yeasts used in its manufacture cannot hydrolyse grape RG-II (O’Neill et al. 2004). 1.2.1.4 Xylogalacturonan domains Another pectic domain, often quantitatively minor, is xylogalacturonan, briefly described in Table 1.1. A commercial preparation rich in xylogalacturonan is gum tragacanth (Zandleven et al. 2005). 1.2.2 Hemicelluloses Hemicelluloses are conventionally extracted from the cell wall with aqueous alkali. NaOH acts as a chaotropic agent, minimizing hydrogen-bonding between hemicelluloses and cellulose. The high pH ionizes the –OH groups of carbohydrates, * In this chapter, light arrows (e.g. →) indicate glycosidic bonds; heavy arrows (e.g. Î) indicate the direction of a chemical reaction.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

13

R−OH + OH − Î R− O − + H 2 O , so that the hemicellulose molecules become negatively charged and repel one another, helping to maintain their solubility. The alkali is usually supplemented with NaBH4 as a precaution against polysaccharide oxidation or ‘peeling’. Hemicelluloses differ in the alkali concentration required for extraction. Dilute alkali may extract xylans with many side chains; higher concentrations are required for xylans with fewer side chains; and the highest concentrations are required for MLG and xyloglucans (Carpita 1984). Complete extraction of xyloglucans requires 6 M NaOH at 37 °C (Edelmann & Fry 1992). The more commonly used 4 M alkali is not fully effective, and is not recommended. Inclusion of 4% H3BO3 with the 6 M NaOH may facilitate extraction of mannans. Once extracted from the wall with alkali, some hemicelluloses precipitate on neutralization (hemicellulose A), but usually the bulk do not (hemicellulose B). Alkali treatment strips off all ester-bonded (e.g. acetyl and feruloyl) groups and should not be used if such groups are of interest. 1.2.2.1 Xyloglucans The best-studied hemicelluloses of the primary cell wall are the xyloglucans. These possess a linear backbone of (1→4)-linked β-Glc residues (in this sense identical with cellulose). In many dicots, the backbone consists of a cellotetraose (G4G4G4G) repeat with the first three Glc residues (counting from the non-reducing end) carrying an α-Xyl residue on position 6. The repeating disaccharide, α-Xyl-(1→6)-Glc, is called isoprimeverose. The fourth Glc residue is unsubstituted. Some of the Xyl residues (especially the third from the non-reducing end) carry an additional β-Gal residue on position 2, and the β-Gal itself often carries α-Fuc on its 2-position. Xyloglucans can be hydrolysed by endo-(1→4)-β-d-glucanase (cellulase) or by a xyloglucan-specific endo-glucanase (XEG). These two enzymes attack most xyloglucans at the same sites (the unsubstituted Glc residues); however, XEG is more specific and does not cleave cellulose and xylans. XEG hydrolyses the (1→4)-β-Glc linkage adjacent to and on the non-reducing side of an isoprimeverose unit. Endo-(1→4)-β-d-glucanase is available commercially (e.g. Megazyme) at acceptable purity; XEG may be available on request from Novozymes (Bagsværd) but requires purification by the method of Pauly et al. (1999). Either of these enzymes cleaves xyloglucan to yield oligosaccharides (XGOs), which can be purified and characterized, thus indicating the structure of the xyloglucan (though without information on the order in which the constituent XGOs occur). The following code letters are used for describing concisely the sequence of side chains (and unbranched Glc residues) along the (1→4)-β-d-glucan backbone of xyloglucan:

14 G Gol G X L L F F A B C J S S T U

■

Plant Polysaccharides

a β-Glc residue of the backbone (or the reducing terminal Glc moiety) with no side-chain attached glucitol (the former reducing terminus, if reduced with NaBH4) β-6AcGlc α-Xyl-(1→6)-β-Glc (=β-isoprimeverose) β-Gal-(1→2)-α-Xyl-(1→6)-β-Glc β-6AcGal-(1→2)-α-Xyl-(1→6)-β-Glc α-Fuc-(1→2)-β-Gal-(1→2)-α-Xyl-(1→6)-β-Glc α-Fuc-(1→2)-β-6AcGal-(1→2)-α-Xyl-(1→6)-β-Glc α-Ara-(1→2)-[α-Xyl-(1→6)]-β-Glc β-Xyl-(1→2)-[α-Xyl-(1→6)]-β-Glc α-Ara-(1→3)-β-Xyl-(1→2)-[α-Xyl-(1→6)]-β-Glc α-l-Gal-(1→2)-β-Gal-(1→2)-α-Xyl-(1→6)-β-Glc α-5AcAra-(1→2)-α-Xyl-(1→6)-β-Glc α-Ara-(1→2)-α-Xyl-(1→6)-β-Glc β-Ara-(1→3)-α-Ara-(1→2)-α-Xyl-(1→6)-β-Glc β-Xyl-(1→2)-α-Xyl-(1→6)-β-Glc

The β-Glc in each structure is part of the (1→4)-β-glucan backbone xyloglucan. Square brackets indicate branching; for example S is Ara→Xyl→Glc, whereas A could be represented Ara→Glc←Xyl (both pentose residues being directly attached to the backbone glucose). The list of code letters will soon be extended by new XGOs discovered in bryophyte xyloglucans (Peña et al. 2007a). The primary cell walls of Acer, and many other plants, yield two very predominant XGOs: XXFG and XXXG. Other widely observed sequences include XXLG, XLXG and XLFG. Like most hemicelluloses, xyloglucans bind strongly to cellulose (e.g. filter paper in vitro) by hydrogen-bonding, especially after drying, but the XGOs mentioned above do not and can readily be washed off paper (even after drying) with water. In this sense, XGOs differ markedly from oligosaccharides of cellulose (e.g. GGGGGG), which have a very high affinity for cellulose. XXFG often contains 6AcGal in place of the Gal in the F unit (the oligosaccharide is then designated XXFG), although the acetyl group can migrate non-enzymically to other positions of the same Gal residue. The acetate groups are stable during digestion with cellulase or XEG, though rapidly removed by NaOH. Isoprimeverose [α-Xyl-(1→6)-Glc] is the most diagnostic repeat unit of xyloglucan, being unknown from any other polymer; it can be released quantitatively from xyloglucan by Driselase digestion, which thus serves as a simple assay for xyloglucan. The α-Xyl residues are completely stable to Driselase, which lacks detectable α-xylosidase activity. Essentially all the other residues are released as free monosaccharides; however, the 6AcGal residues appear to be released by Driselase as galactose + free acetate.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

15

Commercially available xyloglucan (Megazyme) is from tamarind seed. This polysaccharide broadly resembles Acer xyloglucan except that fucose is absent; the major XGOs released by XEG are XLLG > XXLG > XXXG > XLXG. A small proportion of Ara residues is present. 1.2.2.2 Xylans The second most abundant hemicelluloses in dicot primary cell walls are xylans, which have a backbone of (1→4)-linked β-Xyl residues. Dicot xylans carry side chains, especially α-Ara, α-GlcA and α-MeGlcA attached predominantly to position 2 of some Xyl residues – hence the use of fuller names such as glucuronoarabinoxylans (GAXs), but there is probably a continuum of compositions, and the term ‘xylans’ will generally be used here to cover all such hemicelluloses. O-Acetyl groups are also present, especially on the Xyl residues. Xylans can hydrogen-bond to cellulose, though generally more slowly and less strongly than do xyloglucans and MLGs. When cell walls or AIR are digested with Driselase, the xylan backbone is cleaved to yield a mixture containing xylose and xylobiose, the yield of which is a valuable indication of xylan content. Xylobiose is a particularly useful diagnostic fragment which enables assay of xylan – as with isoprimeverose for xyloglucan. The yield is maximized by prior treatment with NaOH or mild acid. Xylobiose is completely stable in the presence of Driselase. The endo-xylanase activity of Driselase cleaves xylan’s backbone to yield progressively smaller oligosaccharides, the last permitted reaction being hydrolysis of the trisaccharide: Xyl−Xyl−Xyl + H 2 O Î Xyl−Xyl + Xyl forming a stable disaccharide, xylobiose. Other enzymes in Driselase remove the Ara residues as monosaccharide. However, Driselase lacks α-d-glucuronidase; the GlcA and MeGlcA residues of xylans probably end up in oligosaccharides. 1.2.2.3 Mannans Dicot primary cell walls contain relatively small proportions of mannans, the broad term used here for polysaccharides with a backbone rich in (or composed entirely of) (1→4)-linked β-Man residues. Some mannans include β-Glc residues within the backbone and/or α-Gal residues as side chains (Table 1.1). Terms such as glucomannans, galactomannans and galactoglucomannans can specify this. Endo-mannanases digest the backbone of mannans yielding useful diagnostic oligosaccharides (McCleary & Matheson 1983; Handford et al. 2003); some are available commercially (Megazyme). 1.2.2.4 Glucuronomannans Perhaps better classed as mucilages than hemicelluloses, and completely distinct from the mannans described above, are glucuronomannans, which

16

■

Plant Polysaccharides

contain α-Man instead of β-Man. They have a backbone of the repeating disaccharide [–4)-β-GlcA-(1→2)-α-Man-(1→–]. To position 3 of many GlcA and Man residues are attached diverse side chains rich in α-Ara, β-Gal, α-Fuc (Redgwell et al. 1986) and sometimes Xyl. Some GlcA residues are reported to be methyl-esterified (Honda et al. 1996); if confirmed, this is a unique feature. The simplest tool for glucuronomannan identification is partial acid hydrolysis, which yields the relatively acid-stable disaccharide β-GlcA(1→2)-Man, detectable by chromatography or electrophoresis and further characterized by digestion with β-glucuronidase (Sotiriou et al. 2007). Preparations rich in glucuronomannan are gum ghatti and leiocarpan A (Aspinall et al. 1969) – useful sources of this disaccharide. 1.2.3 Cellulose Microfibrils are composed of cellulose, possibly with small amounts of entrapped hemicelluloses. In aerial organs (stems, hypocotyls, leaves), the orientation of microfibrils in the outer wall of the epidermis governs the direction of organ growth (length versus girth), this specific wall being much thicker than the others in a growing organ (Kutschera 2008). Cellulose can be partially purified as the ‘α-cellulose’ fraction – wall material left insoluble after exhaustive extraction of the matrix polysaccharides e.g. with 6 M NaOH at 37 °C. Prepared in this way, cellulose is contaminated by glycoproteins (especially extensins) and some pectic material (possibly covalently linked to the extensins). Cellulose is relatively resistant to hydrolysis in 2 M TFA at 120 °C for 1 h, although a small proportion can be released as glucose, especially if the crystallinity of the cellulose has been compromised by severe alkali pretreatments. Good conversion of cellulose to glucose can be achieved by a twostage acid hydrolysis (Saeman method: stirring in 72% w/w H2SO4 at room temperature to dissolve the polysaccharide, then heating in 2 M H2SO4 at 120 °C for 1 h to hydrolyse it; subsequently the H2SO4 is neutralized with a small excess of BaCO3). More convenient, however, is digestion of the αcellulose fraction with Driselase, which usually gives an excellent yield of glucose. A short-cut, omitting the preparation of α-cellulose, is sequential treatment of cell walls with 2 M TFA (120 °C, 1 h) and then Driselase, giving a useful estimate of non-cellulosic and cellulosic Glc residues respectively (O’Looney & Fry 2005; Fig. 1.2); note, however, that contaminating starch would be recorded as non-cellulosic Glc in this procedure.

1.3 Secondary cell walls Cells change their repertoire of polysaccharide synthesis during the differentiation of cambium into xylem (Thornber & Northcote 1962). Whereas

Cell Wall Polysaccharide Composition and Covalent Crosslinking Autoradiogram: cellulose

Markers

Autoradiogram: total AIR

a

■

Markers

b

Rib

Rha

Rha

17

Rib Rib

?

Fuc

Xyl

Xyl

Ara

Ara

Fuc Xyl

Ara Man Man

Man Glc Gal

Glc

Xyl2

Xyl2 IP GalA

Glc

Glc

Gal

Gal

GalA + IP

Xyl2 IP GalA

?

–

+ –

0.1 h

+

4h

–

+ –

8h

+

30 h

1 2 3

–

+ –

0.1 h

+

4h

–

+ –

8h

+

1 2 3

30 h

Figure 1.2 Exploring the effect of a new herbicide on cell wall polysaccharide biosynthesis. Maize cell-cultures were pretreated with (+) or without (−) 480-nM oxaziclomefone for 0.1–30 h and then pulse-labelled with [U-14C]glucose for a further 2 h. The results, showing no effect of the herbicide, illustrate the reproducibility of the analytical method. (a) Total polysaccharide biosynthesis. The radiolabelled AIR was treated with Driselase, and the resultant sugars were separated by PC and autoradiographed. The 14C-labelled products were Glc (cellulosic plus wall matrix-derived); Gal; Ara; Xyl and xylobiose (Xyl2, diagnostic of xylans); GalA and Rha (diagnostic of pectins); isoprimeverose (IP, diagnostic of xyloglucan); Man, and Rib. Ribose is not a cell wall sugar, and the polymeric component of AIR from which it was released by hydrolysis was probably RNA. (b) Cellulose biosynthesis. The radiolabelled AIR was freed of pectins and hemicelluloses by TFA hydrolysis (2 M TFA, 120 °C, 1 h) followed by washing. The remaining insoluble material was then Driselase-digested, yielding [14C]glucose, which indicates how much [14C]cellulose had been synthesized. From O’Looney & Fry (2005). 1, 2, 3: Marker lanes.

cambium has only primary walls, basically similar to those described in the above generalizations, the major polysaccharides of mature xylem cell walls are cellulose, xylans and mannans. Angiosperm xylem (hardwood) contains 33–50% w/w cellulose (dry weight basis; Willför et al. 2005). 4-OMethylglucuronoxylan and glucomannan (constituting roughly 20–30% and 0.5–5.0% of the dry weight of the wood, respectively; Willför et al. 2005) are both partially O-acetylated (Teleman et al. 2003). The glucomannans are acetylated on position 2 of some Man residues and position 3 of others; some Man residues and the Glc residues are non-acetylated. Recent work (Peña et al. 2007b) has confirmed the older report (Andersson et al. 1983) that xylans of secondary walls have a backbone whose reducing terminus is β-Xyl(1→3)-α-Rha-(1→2)-α-GalA-(1→4)-Xyl. In this sequence the Rha-GalA bond

18

■

Plant Polysaccharides

is (1→2), whereas in RG-I it is (1→4). It seems likely that the Xyl-Rha-GalAXyl moiety is a primer on which xylan backbone biosynthesis is initiated. In conifer xylem (softwood), the (galacto)-glucomannans are acetylated but the xylans are not (Teleman et al. 2003). Xyloglucans may be present in small proportions in xylem secondary walls (Nishikubo et al. 2007). Arabinogalactans are abundant in larch wood (Sigma catalogue), but these are water-soluble rather than structural wall components. Wood also contains pectic residues (GalA + Rha ≈ 1.5–3.5% of dry wood; Willför et al. 2005), though probably mainly in the primary walls of tracheary elements and ray cells. Cork (phellem) has thick secondary walls. Mature oak cork contains 26% w/w polysaccharide, which on Saeman hydrolysis yields 42 mol% Glc (mainly cellulosic), 30 mol% Xyl, 12 mol% uronic acids and 8 mol% Ara (suggesting xylans). There is only 3 mol% Man, 3 mol% Gal and 2 mol% Rha, indicating small amounts of mannans and pectins (Rocha et al. 2000). NaOHextractable hemicellulose contains xylans with MeGlcA and β-Xyl residues attached to O-2 of some of the Xyl residues (Asensio 1987). A very different secondary wall is that of cotton seed trichomes (‘fibres’). At maturity, this thick wall is almost pure cellulose.

1.4 Taxonomic consideration of primary cell walls The description of primary walls given above relates to dicots – the plants most intensively studied because of their agricultural importance. Similar primary walls, described as ‘type I’, are also found in many monocots. However, a substantially different primary wall, ‘type II’, occurs in the monocot order Poales. Other plant taxa also exhibit characteristic compositional features. A brief survey follows. 1.4.1 Poalean primary cell walls Type II primary walls are characterized by a lower content of pectins and xyloglucans, a correspondingly higher proportion of (feruloylated) xylans, and the presence of a hemicellulose called mixed-linkage (1→3,1→4)-β-dglucan (MLG), which is absent in dicots. 1.4.1.1 Poalean xyloglucans Poalean xyloglucan also differs qualitatively from that of dicots. Although possessing a backbone rich in G and X units (see above), the X:G ratio is substantially lower than in dicots (e.g. only ∼32 and ∼50% of the Glc residues bear a Xyl side chain in the xyloglucan of rice shoots (Kato et al. 1982) and barley coleoptiles (Gibeaut et al. 2005) respectively), and the Gal, Ara and Fuc contents are very low. Poalean xyloglucans do possess a few Fuc residues – readily detected if the cells are prelabelled by feeding with [3H]fucose (McDougall & Fry 1994). Digestion of alkali-extracted rice seedling xyloglucan with partially purified cellulase yielded XGOs from which Kato et al.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

19

(1982) deduced major sequences to have been XXGGG and XGGG. Gibeaut et al. (2005) isolated numerous XGOs from XEG digests of barley coleoptile cell walls. Because of the low X:G ratio, NaOH-extracted poalean xyloglucans have a limited water solubility; they may fall into the hemicellulose A class, precipitating upon neutralization. However, in muro many of the unbranched Glc residues in the backbone of poalean xyloglucans are 6AcGlc (designated G); this acetylation increases water solubility, like xylosylation. Major oligosaccharides obtained upon XEG digestion of alkali-untreated barley coleoptile walls include XXGGG, XXGG, XXGGGG, XXGGGG and XXGGG (Gibeaut et al. 2005). Indeed, (Xyl-free) cellulose that artificially carries 0.5–1.0 O-acetyl groups per Glc residue is water-soluble – unlike cellulose itself and unlike commercial cellulose acetate, which typically has approximately 2.5 O-acetyl groups per Glc residue. It is interesting that artificial water-soluble cellulose acetate is an effective donor substrate for xyloglucan endo-transglucosylase (XET) activity, in this sense evidently resembling xyloglucan (Fry et al., unpublished). 1.4.1.2 Poalean (feruloylated) xylans Poalean xylans differ from dicot xylans in having the α-Ara residues attached mainly to position 3, rather than 2, of some backbone β-Xyl residues. The Ara residues are frequently also further substituted with oligosaccharide side chains containing β-Xyl, d-Gal, l-Gal, and acetyl groups (Carpita 1996). The α-GlcA and/or MeGlcA are mainly on O-2 of the Xyl residues, as in dicots. Poalean xylans have feruloyl and 4-coumaroyl groups esterified to O-5 of some Ara residues (forming Fer-Ara; Fig. 1.1). This applies to very diverse members of the Poales (sensu lato), from grasses to pineapples (Smith & Harris 2001) and also to some related monocot orders (Commelinales, Zingiberales, Arecales – the commelinoid group), but not to the Asparagales or Liliales. The position of feruloylation can be investigated by isolation and characterization of feruloyl-oligosaccharides. Mild acid hydrolyses furanosyl more rapidly than pyranosyl linkages, so the Araf→xylan bond is selectively cleaved, yielding free arabinose and Fer-Ara; in addition, larger feruloylated side chain oligosaccharides are also obtained, always with Ara at the reducing terminus (Wende & Fry 1996). These structures show that some of the feruloyl groups were attached to Ara residues which themselves also bore additional sugar side chains. Xylans can also be hydrolysed with Driselase, releasing feruloylated oligosaccharides in which the reducing terminus is a Xyl that was formerly a backbone residue (Kato & Nevins 1985). 1.4.1.3 Poalean mixed-linkage glucans Mixed-linkage (1→3,1→4)-β-d-glucan (MLG) is a hemicellulose of poalean primary walls, being especially abundant during periods of rapid cell expansion. It often decreases in quantity per cell (indicating breakdown,

20

■

Plant Polysaccharides

rather than simply dilution by continued deposition of different polysaccharides) when cell expansion is complete. Once extracted in alkali, MLG is a flexible molecule, soluble in neutral water (Buliga et al. 1986) – unlike (1→4)-β-glucan (cellulose), which is insoluble. In growing cells, MLG is very firmly hydrogen-bonded to the cellulose (Carpita et al. 2001), and may tether microfibrils (Labavitch & Ray 1978; Wada & Ray 1978; Fry 1989). MLG also occurs abundantly in cereal grains, where it is largely not microfibril-bonded. Barley flour MLG is available commercially (Megazyme, Sigma). Poalean MLG is an unbranched chain of β-Glc residues linked by (1→3) and (1→4) bonds. The polysaccharide typically takes the form: … G 3 G 4 G 4 G 3 G 4 G 4 G 4 G 3 G 4 G 4 G 3 G 4 G 4 G 3 G… , reading from non-reducing to reducing terminus, where ‘G’ is β-Glc, and ‘3’ and ‘4’ indicate (1→3) and (1→4) bonds respectively. The underlined domains are effectively cellotriose and (fewer) cellotetraose units, interlinked by single (1→3) bonds (Meikle et al. 1994). MLG can be characterized by digestion with lichenase [(1→3),(1→4)-β-dglucan 4-glucanohydrolase; EC 3.2.1.73], which cleaves the (1→4) bond in the sequence …G3G4G…, yielding analytically informative oligosaccharides, always with a (1→3) bond at the reducing end (Meikle et al. 1994; Grishutin et al. 2006; Li et al. 2006). The major oligosaccharide released by lichenase digestion of poalean MLG is G4G3G, but appreciable amounts of G4G4G3G are also obtained. The trisaccharide:tetrasaccharide molar ratio varies between samples, usually within the range 1.5:1 to 4:1. Progressively smaller proportions of G4G4G4G3G, G4G4G4G4G3G etc. also occur, certain larger oligosaccharides being favoured, especially a nonasaccharide and a dodecasaccharide (Woodward et al. 1983). These short ‘cellulose-like’ domains within MLG probably help the MLG to hydrogen-bond to microfibrils. Very little (sometimes ‘no’) disaccharide (G3G; laminaribiose) is obtained, which would have indicated the sequence …G4G3G4G3G…, and there is no strong evidence for consecutive (1→3)-bonds. Cellulase digests MLG by cleaving mainly the (1→4) bond in …G4G3G…, yielding oligosaccharides with a (1→3) bond at the non-reducing terminus. However, cellulase also attacks some other hemicelluloses. 1.4.1.4 Other poalean polysaccharides Members of the Poales also possess (gluco)mannans, concentrated in the epidermal walls (Carpita et al. 2001) – a location playing a particularly important role in restraining organ growth. Most poalean cells contain relatively small proportions of pectin (thus a low GalA and Rha content). However, cultured maize and rice cells have RG-I that is remarkably similar to that of dicots, except for a lower Fuc content. Grass RG-IIs are typically present at less than 10% of the concentra-

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

21

tion seen in dicot primary walls, but are qualitatively indistinguishable from those of dicots (Thomas et al. 1989). 1.4.2 Taxonomically restricted features of non-poalean angiosperm walls Although all dicots have xyloglucan in their primary cell walls, there is some distinct taxonomic and developmental specificity in the repeat units liberated by XEG digestion (Hoffman et al. 2005). The classic (‘XXXG-type’) of xyloglucan has a backbone in which three contiguous xylosylated β-Glc residues are followed by a single non-xylosylated one. The polysaccharide is thus built of Glc4-based repeat-units, usually mainly XXXG and XXFG, with less XXLG, XLFG, XLLG and others. In some dicots – the Solanales and Lamiales – ‘XXXG-type’ xyloglucan is scarce or absent. In these two orders, both in the asterid clade of eudicots, most xyloglucan has two or three contiguous non-xylosylated Glc residues. Examples are Glc4-based (XXGG-type) repeat units, e.g. XXGG, XSGG, SXGG, SSGG, LXGG, LSGG, LLGG, LTGG; and Glc5-based (XXGGG-type) repeat units, e.g. XLGGG, XSGGG, SXGGG and SSGGG. However, xyloglucan in olive fruit (Lamiales) is ‘XXXG type’ (predominant repeat units being XXSG and XLSG). Some S units (designated S) have 5AcAra in place of Ara. Many of the XXGG- and XXGGG-type repeats are 6-O-acetylated, most commonly in the sequence XXGG and XXGGG, underlining indicating that an O-acetyl group that effectively replaces the ‘missing’ α-Xyl residue(s), as discussed for Poales. In the order Caryophyllales (e.g. Spinacia, Beta, Sagina, etc.), pectins carry feruloyl and 4-coumaroyl ester groups linked to certain α-Ara and β-Gal residues, predominantly at the non-reducing termini of the Ara- and/or Gal-rich side chains of a pectic domain, probably RG-I (Fry 1983). Diagnostic oligosaccharides (Fer-Gal-Gal and Fer-Ara-Ara) can be released from spinach pectin by Driselase digestion (Fry 1982). Another taxonomically restricted polysaccharide (domain?) is apiogalacturonan, found only in certain aquatic monocots of the order Alismatales, e.g. Lemna, Spirodela and Zostera. 1.4.3 Cell walls of non-angiosperms To date, most cell wall research has understandably been focused on angiosperm crops and coniferous wood. Much less detail is available about the wall polysaccharides of non-angiosperms, although progress is beginning towards a description of the evolutionary history of the plant cell wall. 1.4.3.1 Charophytic algae Charophytes are the closest living algal relatives of land plants. Indeed charophytes plus land plants are now often classified as a single taxon, the

22

■

Plant Polysaccharides

Streptophyta, well resolved from other green algae. There is debate as to which charophytic order (Charales or Coleochaetales) is most closely related to the land plants. Unfortunately, very little is known about charophyte wall composition. They contain cellulose; they are also usually rich in GalA residues, probably as partially methylesterified homogalacturonans (Cherno et al. 1976), much of which is readily extracted from the wall with oxalate buffer, pH 4.3. Acid hydrolysis yielded roughly equal amounts of Xyl and Glc from walls of Klebsormidium and Chara (Popper & Fry 2003). However, Driselase digestion of Chara, Coleochaete and Klebsormidium walls (or hemicelluloses) yielded no isoprimeverose, indicating the absence of conventional xyloglucan (Popper & Fry 2003). In fact, Coleochaete walls yielded very little Xyl on acid hydrolysis, precluding an appreciable xyloglucan content (Popper & Fry 2003). Lichenase digestion also failed to yield oligosaccharides from Chara or Coleochaete, indicating undetectable MLG. Charophytic walls have a moderate content of GlcA, Man and MeRha residues (from unidentified polysaccharides), but no detectable MeGal (Popper & Fry 2003). Much remains to be discovered about charophytic cell walls – information that would promote our understanding of the primordial features of the land-plant wall. 1.4.3.2 Bryophytes Bryophytes are non-vascular land plants, traditionally divided into liverworts, hornworts and mosses, although some ‘mosses’ such as Sphagnum and Andreaea seem sufficiently isolated that they should be placed in separate classes. Unlike all other land plants, the bryophytes have the gametophyte (haploid) phase of the life cycle dominant. The sporophyte (diploid) phase tends to be shorter-lived, smaller and dependent on a supply of nutrients from the gametophyte. Few chemical analyses of bryophyte sporophyte walls have been published; therefore, in comparisons between bryophytes and vascular plants, the effects of taxonomy and life-cycle stage are usually conflated. Like all land plants, bryophytes possess xyloglucan, albeit often in smaller proportions than in vascular plants. The simplest demonstration of this is the production of isoprimeverose on Driselase digestion of bryophyte AIR. Work is in progress (Peña et al. 2007a) to characterize the bryophyte XGOs. Intriguingly, XGOs from the moss Physcomitrella have branched side chains containing acidic sugar residues. Despite these major differences between bryophyte and angiosperm xyloglucans, AIR from cell cultures of the hornwort Anthoceros yields the nonasaccharide XXFG (unpublished observations), a highly specific structure that has evidently been preserved unchanged throughout land-plant evolution. All bryophyte walls investigated contain GalA and Rha, characteristic of pectins, and Driselase digestion yields xylobiose, indicative of xylans.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

23

Bryophytes usually give a very high yield of Man on acid hydrolysis, suggesting a high mannan content. They contain MeRha (approximately equalling Rha), but little or no MeGal. Driselase digestion of walls from bryophyte gametophytes yielded small quantities of a borate-containing polysaccharide (0.004–0.025% of the wall dry weight) that coelutes with RG-II dimer on gel-permeation chromatography (Matsunaga et al. 2004). The same bryophyte walls also contained traces of MeFuc and MeXyl, characteristic of RG-II; however, Api and AceA were below the limit of detection (Matsunaga et al. 2004). Thus, bryophytes may well contain RG-II or a similar polysaccharide, albeit about 100 times less than in flowering plants. Bryophyte sporophytes have not yet been tested in this respect. On acid hydrolysis, gametophyte walls of the hornwort Anthoceros caucasicus, unlike mosses and liverworts, give a very high yield of an unusual disaccharide, α-GlcA-(1→3)-l-Gal, from an unidentified polysaccharide (Popper et al. 2003). The same disaccharide was obtained from cultured cells of Anthoceros agrestis (unpublished), suggesting that it is taxonomically rather than developmentally stipulated. The existence of a unique polysaccharide in this evolutionarily isolated genus supports the view that major steps in plant evolution were accompanied by significant changes in wall composition. Extraction of bryophyte cell wall polysaccharides is often difficult by standard agents, e.g. chelating agents and alkali, possibly because of the presence of Hua residues (Fig. 1.1) acting as crosslinks – see later. A pretreatment with mild acid may be beneficial, as may acidified NaClO2 (Painter 1983).

1.4.3.3 Lycopodiophytes Lycopodiophytes, e.g. Lycopodium, Selaginella and Isoetes, are early-diverging vascular plants, traditionally treated as pteridophytes although clearly distinct from the euphyllophytic pteridophytes (see below). Modern lycopodiophytes are the few remaining progeny of a much more diverse and abundant flora that throve in the Devonian. They are split into homosporous (e.g. Lycopodium) and heterosporous plants (e.g. Selaginella). MeGal is abundant in all lycopodiophytes tested but not in bryophytes or euphyllophytes. MeRha is abundant in homosporous (but not heterosporous) lycopodiophytes. It is a component of their RG-II, in place of Rha (Matsunaga et al. 2004); however, the total MeRha content (Popper & Fry 2004) seriously exceeds what would be required for the modest RG-II content, suggesting that the MeRha of homosporous lycopodiophyte walls is not confined to RG-II. Many lycopodiophyte walls are rich in Man, although the polysaccharides involved have not been characterized.

24

■

Plant Polysaccharides

1.4.3.4 Euphyllophytic pteridophytes All vascular plants possessing megaphylls (leaves with true veins) instead of microphylls are called ‘euphyllophytes’. Living euphyllophytic pteridophytes are Equisetum, Psilotum and Tmesipteris (which are all eusporangiate) and ferns (both eusporangiate and leptosporangiate, the latter being laterdiverging). They contain RG-II remarkably similar to that of angiosperms, though in some groups (ferns and Psilotum but not Equisetum) Rha is replaced by MeRha (Matsunaga et al. 2004). No functional consequences of the Rha/ MeRha swap were detectable. It has recently been found that, apparently unique outside the Poales, horsetails (Equisetum) contain MLG (Fry et al. 2008b; Sørensen et al. 2008). Equisetum MLG broadly resembles that of the Poales, except that the major product of lichenase is the tetrasaccharide, not the trisaccharide, and that laminaribiose is also a noticeable lichenase product. Equisetum MLG resembles poalean MLG in having appreciable quantities of a nonasaccharide repeat unit, but appears to lack the dodecasaccharide. Equisetum possesses not only the polysaccharide MLG but also a novel endo-transglucosylase whose favoured action appears to be cleaving MLG and grafting a segment of it on to a xyloglucan (acceptor substrate) chain (Fry et al., unpublished). For such an enzyme to act in vivo, Equisetum would need also to possess xyloglucan. This it clearly does, on the basis of Driselase and XEG digestion products; however, these digests include several unusual oligosaccharides, indicating that Equisetum xyloglucan has unique structural features. Man residues are more abundant in eusporangiate pteridophyte walls (extreme in the case of Psilotum) than in leptosporangiate fern walls (Popper & Fry 2004). Conversely, proanthocyanidins are more abundant in leptosporangiate fern AIR than in eusporangiate species. 1.4.3.5 Gymnosperms Gymnosperms studied to date (mainly conifers) seem to have wall components resembling those of dicots, including xyloglucan with the repeat units XXXG, XXFG, XLFG, XXLG and XLLG. A high mannan content is a recurrent wall feature of conifers (Edashige & Ishii 1996) and cycads, though not gnetophytes (Popper & Fry 2004). Gnetophytes may be the gymnosperms closest to angiosperms. All four extant gymnosperm classes possess feruloyl esters in their primary walls (Carnachan & Harris 2000), but the feruloylated polysaccharide is unidentified.

1.5 Covalent bonds between wall polysaccharides The primary cell wall has the strength to resist expansion (and rupture) in the face of, typically, 0.5-MPa turgor pressures, while permitting controlled creep and thus cell growth. The ability of the wall to restrain irreversible expansion (i.e. the wall’s plastic extensibility) can be regulated in vivo, e.g.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

25

hormonally, and this is assumed to involve the making or breaking of interpolymeric crosslinks in the wall (see above regarding definition of crosslink). Decreasing wall extensibility may involve the formation of additional crosslinks and is termed wall tightening. This term is preferred over ‘stiffening’ or ‘rigidification’, which would wrongly imply that the inextensible wall becomes difficult to bend; it usually does not. Conversely, an increase in plastic extensibility is described as wall loosening. It is therefore important to document the polymer–polymer crosslinks existing in the primary cell wall. This chapter considers covalent crosslinks, non-covalent ones being discussed in Chapter 14. 1.5.1 Glycosidic bonds joining polysaccharides into molecular ‘trees’ Glycosidic bonds are not regarded as true ‘crosslinks’. A glycosidic bond has a well-defined directionality, indicated for example by the arrow in the systematic name of isoprimeverose: α-d-Xylp-(1→6)-d-Glc. This is described as Xyl attached to Glc (not Glc to Xyl) because the linkage involves the anomeric centre of the Xyl, not the Glc. With one postulated exception (Hua; see Section 1.5.2), each sugar residue can form only a single glycosidic bond to another sugar unit, whereas a given sugar unit can in principle have anything from zero up to three (in a C5 sugar), four (in a C6 sugar) or even five (in Kdo) additional sugar residues attached to it – potentially leading to a branched tree-like structure (e.g. RG-I, RG-II or xyloglucan); the reducing terminus is the base of the ‘tree trunk’ (ignore the tree’s roots!). It is chemically feasible for the reducing terminus of one polysaccharide to form a glycosidic bond to a residue within another (identical or non-identical) polysaccharide, thus covalently joining the two polysaccharides. However, this is not a true crosslink because the linkage does not differ in its fundamental nature from any glycosidic linkage within either ‘individual’ polysaccharide. The product is effectively a single larger polysaccharide, albeit maybe with qualitatively dissimilar domains. Examples of this type of structure are found in pectins, where the reducing terminus of xylogalacturonan or RG-I can be glycosidically linked to the non-reducing end of homogalacturonan. It is likely that pectins in muro have structures such as … RG-I →HGA → XGA →HGA →RG-II →HGA… where → denotes a glycosidic bond, HGA is homogalacturonan and XGA is xylogalacturonan (Ishii & Matsunaga 2001; Coenen et al. 2007). In suspension-cultured angiosperm cells, roughly half the xyloglucan is covalently bonded to acidic pectic domains, probably RG-I (Thompson & Fry 2000; Popper & Fry 2005). The precise chemistry of this xyloglucan–pectin bond is still uncertain; it is stable to prolonged treatment in concentrated

26

■

Plant Polysaccharides

NaOH and urea and a tenable model is a glycosidic bond between the reducing end of the xyloglucan and the Ara/Gal-rich side chains of RG-I. The xyloglucan–RG-I association is most easily demonstrated by ion-exchange chromatography or high-voltage electrophoresis on glass-fibre ‘paper ’: despite angiosperm xyloglucans containing no acidic residues, 44–75% of the 6 M NaOH-extractable xyloglucan (recognized by its conversion to isoprimeverose by Driselase) binds to an anion-exchange column (Popper & Fry 2005) or migrates towards the anode (Thompson & Fry 2000). The formation of these xyloglucan–RG-I bonds has been shown by in-vivo pulse-radiolabelling to occur cosynthetically (probably within the Golgi system), rather than by heterotransglycosylation in the wall (Popper & Fry 2008). Hrmová et al. (2007) have shown that a barley XTH can catalyse the formation of an MLG→xyloglucan bond by using MLG as donor substrate and a xyloglucan oligosaccharide (e.g. XXXGol) as acceptor substrate in a heterotransglycosylation reaction: … GGGGGGGGGGGGGGGGGGGGG… + XXXGol Ð … GGGGGGGGG → XXXGol + GGGGGGGGGGGG… where …GGGGGG… is MLG, → is a glycosidic bond, and Ð indicates the direction of the enzymic reaction. However, the barley enzyme catalysed this reaction at approximately 0.2% of the rate of the classic XET reaction (where the donor is a xyloglucan). This is slower than with any of various artificial substrates, e.g. hydroxyethylcellulose or cellulose sulphate; the biological significance of the heterotransglycosylation reaction therefore remains unclear. Recently, an enzyme activity was detected in Equisetum for which the formation of MLG→xyloglucan bonds is the preferred reaction, well exceeding the classic XET reaction rate (Fry et al. 2008a). Thus, at least in Equisetum, the formation of interpolysaccharide glycosidic bonds may be a major in-vivo reaction.

1.5.2 Glycosidic bonds forming true (‘lateral’) crosslinks between polysaccharides? The inextractability of Sphagnum (moss) polysaccharides may be due to the presence of d-lyxo-5-hexosulopyranuronic acid (Hua) residues. This is an unusual monosaccharide with two anomeric carbons, potentially capable of simultaneously forming glycosidic bonds to two different sugars and thus acting as a true glycosidic crosslink between polysaccharide chains (joining two trees through their ‘branches’) (Painter 1983). This intriguing hypothesis requires future evaluation.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

27

1.5.3 Oxidative coupling products as crosslinks or intrapolymeric loops The feruloyl residues of poalean xylans and caryophyllalean pectins undergo oxidative phenolic coupling in vivo, forming dimers (dehydrodiferuloyl residues), probably catalysed by peroxidases (Encina & Fry 2005). The linkages between the two participating feruloyl residues are C–O–C (ether) or C–C (e.g. biphenyl) bonds. Some of this coupling occurs well after the feruloylated polysaccharides have been deposited in the wall, but pulseradiolabelling experiments show that a substantial percentage of it occurs within 1–2 min of the feruloyl groups’ becoming attached to the polysaccharide, thus presumably still within the Golgi system (Fry et al. 2000). It is widely assumed that the two feruloyl residues that couple are from separate polysaccharides, so that the coupling reaction makes an interpolymeric crosslink. Recent data support this hypothesis (Lindsay & Fry 2008), which would provide a plausible mechanism for wall tightening, restraining further growth (Ortega et al. 2006). However, it is currently not possible to eliminate the additional possibility that sometimes both consenting feruloyl residues are on the same polysaccharide chain, coupling to form an intrapolymeric loop. Such loops are less likely in the case of ferulate trimers and larger coupling products, which in cultured maize cells considerably exceed the dimers (Fry et al. 2000). Some specific trimers and tetramers have been characterized (e.g. Funk et al. 2005), though this is challenging because of the steeply increasing number of structural permutations with increasing size. Coumaroyl esters are much less susceptible than feruloyl esters to oxidative coupling in vivo (Fry et al. 2000).

1.5.4 Uronoyl esters and uronoyl amides Theoretically, the –COOH group of an acidic sugar residue could become ester-bonded to an alcohol (e.g. another sugar residue) or amide-bonded to an amino compound (e.g. glucosamine, lysine or putrescine), thus potentially forming a crosslink. In the case of methylesterified GalA residues, a plausible crosslinking mechanism is transacylation whereby a GalA–methanol bond in homogalacturonan is cleaved and the ‘uronoyl’ residue transferred (still glycosidically linked within a homogalacturonan chain) on to an acceptor substrate (alcohol or amine). In principle, a pectinmethylesterase-like enzyme could catalyse such a reaction, if the enzyme could be coerced into favouring transacylation over hydrolysis; and it is intriguing that there are about 98 and 53 genes encoding putative pectinmethylesterases in Arabidopsis and rice respectively (few of which have been tested for enzymic activity). Authentic model O-galacturonoyl–sugar esters (Brown & Fry 1993a), Nεgalacturonoyl–lysine amide (Perrone et al. 1998) and N,N′-di-galacturonoyl– putrescine amide (Lenucci et al. 2005) have been synthesized chemically and their properties investigated. Notably, these ester and amide bonds are

28

■

Plant Polysaccharides

resistant to Driselase, suggesting an obvious method of documenting their presence in muro. Preliminary evidence was compatible with GalA–Lys bonds in cultured cells (Perrone et al. 1998). Corresponding evidence was originally interpreted as suggesting GalA–sugar ester bonds in muro (Brown & Fry 1993b) and was, with hindsight, probably due to the inability of Driselase to cleave the glycosidic bonds of methylesterified GalA residues that are also O-acetylated (Perrone et al. 2002). Independent evidence was obtained for non-methyl GalA esters in maize coleoptile cell walls: namely, the yield of methanol obtained on alkaline hydrolysis was only 50–75% of the total quantity of apparently esterified (NaB2H4-reducible at pH 7) GalA residues (Kim & Carpita 1992). However, it is difficult to imagine that, as these data suggest, cell walls could contain one to three ester crosslinks for every three MeGalA residues. Covalent crosslinks are expected to be strong but few (cf. weak but many in the case of non-covalent crosslinks). The nature of the putative non-methyl galacturonoyl esters remains an intriguing mystery. 1.5.5 Borate diesters A proportion (usually high) of the RG-II present in muro is dimerized via a crosslink formed by a single borate group that is esterified to four –OH groups: those at positions 2 and 3 of the Api residue of side chain A of each of two RG-II domains (O’Neill et al. 2004). This arrangement can be represented as RG-II >( B − )< RG-II. The formation of this structure is probably the main reason why plants require borate as a micronutrient. Most common sugars extremely rapidly form borate esters in the presence of inorganic borate – the basis for separating neutral sugars by electrophoresis in borate buffer. However, such electrophoresis is routinely performed at pH 9.4 (Fry 2000), and the esters are immediately hydrolysed at slightly lower pH values. Furanose residues possessing a diol group (e.g. the ribofuranose residue of RNA and the Apif residue of methyl apioside) form borate esters that are more acid-resistant than those of pyranose residues. However, the ester bonds in RG-II>(B−)(B−)
Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

29

mutant of Arabidopsis produces an RG-II with l-Gal and 2-O-Me-l-Gal in place of the deoxyhexose residues Fuc and MeFuc respectively: the mutant RG-II makes borate crosslinks significantly more slowly and less stably than the wild type (O’Neill et al. 2001).

1.6

Methodology

1.6.1 Specific and non-specific radiolabelling in vivo In-vivo radiolabelling offers access to much information that would be difficult or impossible to acquire without it (Fry 2000). By straightforward methods, radiolabelling renders detectable minute traces (e.g. 1–10 pg) of the polysaccharides of interest; it is non-destructive, so the sample can be further analysed later; and it enables studies of the dynamics of synthesis, modification, secretion, crosslinking and degradation of polysaccharides in living cells. Non-specific radiolabelling results in all newly synthesized organic components of the cell becoming radioactive (Fig. 1.3). This can be useful for the detection, quantification and characterization of novel or little-known residues or substituents, which will be detectable by their radioactivity whether or not their presence was anticipated. Specific radiolabelling permits attention to be focused on a chosen group of metabolically related residues without interference from a large excess of other accompanying residues (Table 1.2). Both specific and non-specific radiolabelling can give valuable information about the partitioning of metabolic flux between competing intermediary pathways (Sharples & Fry 2007), and between the diverse end-products that stem from a common pool of precursors. Thus, feeding of [14C]glucose can indicate the relative rates of synthesis (during the time-frame for which the exogenous [14C]glucose was available in the medium) of all organic components of the wall (Sotiriou et al. 2007). Similarly, feeding of [2-3H] mannose can indicate the relative rates of synthesis of Man, Fuc and l-Gal residues (three quantitatively minor hexoses), without interference from the large excess of the major hexoses (Glc and d-Gal). [2-3H]Mannose feeds 3H into the metabolic network upstream of the branch-point leading to polysaccharide-bound Man, Fuc and l-Gal residues (Fig. 1.3a), so these three residues can be expected to acquire the same specific radioactivity (e.g. measured in MBq μmol−1). This contrasts with the situation when [3H]arabinose is fed (Fig. 1.3b). [3H]Arabinose selectively labels Ara and Xyl residues, such that pentosan metabolism can be specifically examined; however, because of the tritium enters metabolism downstream of the branch point leading to polysaccharide-bound Xyl and Ara residues (the radioisotope is infiltrating the UDP-Xyl pool against the bulk metabolic flux, which is indicated by large arrows in the figure), the Ara residues of polysaccharides will acquire a higher specific radioactivity than the Xyl residues, so it is not possible to quantify the partitioning of material

– – – – – – –

+ + +

+ +

Specific radiolabelling [2-3H]Man (1,2) FUC (3) [6-14C]GlcA (4) [3H]GlcA (5) Ara (6)c,d 14 L-[Me- C]methionine (7) trans-Cinnamate (7,8)

Semi-specificc [14C]Man (2) [1-3H]Gal (9) Acetate (7)

Non-specificf Glc (9–11) D-Fructose (9) + +

+ ++ +

– – – – – – –

Gal

+ +

+ + +

– – + + – – –

GlcA

+ +

+ + +

– – + + – – –

GalA

+ +

+ + +

– – – + ++ – –

Ara

+ +

+ + +

– – – + + – –

Xyl

+ +

+ + +

– – – + – – –

Api

+ +

++ ± +

+ – – – – – –

Man

+ +

++ ± +

+ + – – – – –

Fuc

+ +

++ ± +

+ – – – – – –

L-Gal

+ +

+ ± +

– – – – – + –

Me

+ +

+ ± ++

– – – – – – –

Ac

+ +

+ ± +

– – – – – – +

Fer

b

Where no isotope or position of labelling is stated, these variables are thought to be inconsequential. No reliable method for specific radiolabelling of Rha or Glc residues has been found. c The number of +s is a guide to the specific radioactivity (e.g. measured in kBq/nmol) achieved, not activity (kBq). d In some plants (e.g. Arabidopsis), radiolabel from Ara can re-enter central metabolism via pentose phosphate pathways, leading to moderate labelling of hexoses. e References: (1) Baydoun & Fry 1988; (2) Sotiriou et al.(2007); (3) McDougall & Fry (1994); (4) Brown & Fry (1993b); (5) Longland et al.(1989); (6) Kerr & Fry (2003); (7) Miller et al.(1995); (8) Fry et al.(2000); (9) Sharples & Fry (2007); (10) Fry (1982); (11) O’Looney & Fry (2005; Fig. 1.2). f Note that inclusion of certain non-radioactive polysaccharide-precursors in culture media, particularly myo-inositol, can render radiolabelling with [14C]glucose and [14C]fructose semi-specific. Plant cells can utilize exogenous non-radioactive inositol to produce GlcA, GalA, Api, Ara and Xyl residues, thus decreasing their radiolabelling.

a

Glc, Rhab

Radioactive precursor feda (and referencese)

■

Wall polysaccharide residue(s) radiolabelled

Table 1.2 Precursors found useful in the author’s laboratory for radiolabelling cell wall polysaccharides in plant cell-suspension cultures

30 Plant Polysaccharides

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

31

Figure 1.3 Incorporation of radioisotopes for specific radiolabelling of polysaccharide residues in vivo. In (a), the isotope (dotted arrows) enters the network upstream of the branch point so the three radiolabelled residues acquire approximately equal specific radioactivities. In (b), the isotope enters downstream of the branch point: 3H enters the UDP-Xyl pool only by travelling against the bulk carbon flux, and in so doing decreases in specific radioactivity. Solid arrows denote flux from bulk carbon source, dotted arrows flux from radiolabelled tracer.

32

■

Plant Polysaccharides

between these two pentose residues. For example, when [1-3H]arabinose was fed to cultured maize cells, the polysaccharide-bound [3H]Ara and [3H] Xyl residues acquired specific radioactivities of 1.68 and 0.78 MBq μmol−1 respectively (Wende & Fry 1996). Nevertheless, it is reasonable to assume that all polysaccharide-bound Xyl residues will have approximately the same specific activity, so the partitioning of material between the Xyl residues of xyloglucan and the Xyl residues of xylans can be validly monitored by analysis of the [3H]isoprimeverose and [3H]xylobiose released from these two polysaccharides, respectively, by Driselase (Kerr & Fry 2003). 1.6.2 Chemical and enzymic ‘dissection’ of wall polysaccharides Much valuable work can be done without extraction of intact polysaccharides from the cell wall. For example, plant tissue can be homogenized in 75% ethanol (preferably acidified with 10% formic acid to inactivate endogenous wall-digesting enzymes) and washed to yield an AIR which is comprised of the total cellular polymers, of which polysaccharides usually predominate. Phenol/acetic acid/water (PAW, 2:1:1, w/v/v final composition; Fry 2000) is useful for denaturation and removal of proteins; it does not extract polysaccharides, and a PAW treatment often seems to render wall polysaccharides more vulnerable to subsequent digestion by deliberately added enzymes. Starch can be removed by prolonged stirring in 90% dimethylsulfoxide (though this may dissolve a small proportion of some wall polysaccharides, e.g. MLG and O-acetylated hemicelluloses) or in α-amylase solution. For analysis of total sugar residues, it is recommended that the walls (or extracted polysaccharides) are first hydrolysed in 2 M TFA at 120 °C for 1 h. This yields monosaccharides as the major products from most matrix polysaccharides; however, because of the relative acid stability of uronosyl linkages, uronic acid-rich polysaccharides will additionally yield disaccharides (e.g. GalA→GalA from homogalacturonan, GalA→Rha from RG-I, GlcA→Xyl from xylans, GlcA→Man from glucuronomannans, and GlcA→lGal from Anthoceros polysaccharides). TFA does not efficiently hydrolyse cellulose; therefore the TFA-insoluble residue is subsequently treated with Driselase, which yields free cellulosic glucose. For ‘dissection’ of wall polymers, retaining more structural information about sugar→sugar linkages, mild acid or enzymes are used. Mild acid, sufficient to hydrolyse furanose linkages but not the majority of pyranose residues (Kerr & Fry 2003), cleaves the Araf→Xyl bonds of xylans and thereby releases much free Ara and also intact side chains that had been linked to the xylan backbone via an Araf residue. Examples are Fer-Ara, Xyl(Fer)-Ara, and more complex feruloylated side chains (Wende & Fry 1996). Driselase digests all the common polysaccharides of the plant cell wall, and in several cases gives analytically useful ‘diagnostic’ fragments that define what polysaccharides were present – e.g. isoprimeverose from

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

33

xyloglucans, xylobiose from xylans (Kerr & Fry 2003), GalA-3AcGalA3AcGalA(Me-ester)-GalA from acetylated homogalacturonan, GalA-Rha3AcGalA-Rha-3AcGalA from RG-I (Perrone et al. 2002) and essentially intact RG-II from whole pectin (Matsunaga et al. 2004) (Table 1.1). Some substituents on polysaccharides may hinder Driselase digestibility, and this problem can be overcome by a preliminary treatment of the AIR with alkali and/or mild acid (Kerr & Fry 2003); TFA is the acid of choice because it can easily be removed, by drying, before Driselase addition. Purified endo-acting enzymes (Table 1.1) attack polysaccharides at specific sites along the backbone to yield larger fragments than does Driselase. Such enzymes are endo-hydrolases (e.g. XEG, EPG, RG hydrolase, lichenase) and RG lyase. Several of these are available commercially (e.g. Megazyme). An extensive battery of wall-digesting enzymes, produced in the fungus Pichia pastoris, is also now available (Bauer et al. 2006). It is important to distinguish the enantiomeric forms of monosaccharide residues. d-Gal is metabolically quite unrelated to l-Gal, and both occur in wall polysaccharides. Enantiomers can be distinguished if the monosaccharide is made to interact with another chiral molecule, which can be either an enzyme or a low-Mr chiral alcohol. For example, the Gal is treated with ATP + d-galactokinase (which phosphorylates d- but not l-Gal) or O2 + d-galactose oxidase (Popper et al. 2003). The enzymic technique is particularly useful when then the Gal in question is radiolabelled and available only in minute quantities: for example, [3H]Gal-1-P (formed by galactokinase) is readily separated from remaining unreacted [3H]Gal by chromatography or electrophoresis. Alternatively, Gal (either enantiomer) will react with hot (+)-butan-2-ol (containing 1 M acetyl chloride, which forms HCl on contact with butanol) to produce a series of galactosides, the d-Gal and l-Gal versions of which are not enantiomers (mirror images) of each other and can therefore be resolved chromatographically (Gerwig et al. 1979). 1.6.3 Fractionation and characterization of mono- and oligosaccharides Once the polysaccharides have been dissected into monosaccharides and oligosaccharides, these fragments can be fractionated and characterized. Depending on the aims of the research, various types of chromatography or electrophoresis are employed. Use the simplest technique that works. 1.6.3.1 Paper chromatography (PC) PC gives excellent resolution of most of the common monosaccharides and disaccharides obtained from wall polysaccharides (Fry 2000). They can be detected by staining, either with AgNO3 (which detects ∼0.1 μg Ara) or with aniline hydrogen phthalate (which detects ∼0.4 μg Ara). The latter gives different colours with different monosaccharide classes (aldohexoses, brown; aldopentoses, red; ketohexoses, yellow; uronic acids, orange); in

34

■

Plant Polysaccharides

disaccharides, the colour is dictated by the reducing terminus, so isoprimeverose (brown) is easily distinguished from xylobiose (red). PC is also suitable for larger oligosaccharides (e.g. XEG digestion products, and feruloyl oligosaccharides), which it effectively resolves into size-classes. Radioactive analytes can be located by autoradiography (for 14C) or fluorography (3H), or, if these film-based methods are not sensitive enough, the paper is cut into strips, which are assayed by scintillation counting. To exploit maximally the resolving power of PC, the sample is mixed with an internal authentic marker (radioactive if the sample is non-radioactive, and vice versa); any mismatch between the sample spot and the internal marker spot, even by as little as 1–2 mm, proves that the sample is not the same substance as the marker. A few important oligosaccharides, e.g. MLG-, cello- and mannooligosaccharides, cannot satisfactorily be run on PC if their DP exceeds about 3. This is because of their high affinity for paper (cellulose) – they fail to migrate from the origin, or do so only as an diffuse streak. Suitable papers for PC are Whatman 1 CHR for general-purpose rapid runs (typically 16–36 h, with about 60 samples per tank), the thicker Whatman 3mm Chr for preparative work, and the denser Schleicher and Schüll 2045B for better-resolution slow runs (2–9 days). The best general-purpose solvent for exploratory work with mono- and disaccharides is butan-1-ol/acetic acid/water (12:3:5 by volume, freshly prepared); for larger oligosaccharides it is ethyl acetate/acetic acid/water (10:5:6). These acidic solvents tolerate quite heavy contamination with salts and proteins. For better discrimination between different neutral mono- and disaccharides, ethyl acetate/pyridine/ water (8:2:1) is excellent (Fry 2000). 1.6.3.2 Thin-layer chromatography (TLC) TLC on silica gel gives similar resolution to PC but on a smaller area (typically 20 × 20 cm as opposed to 46 × 57 cm) (Fry 2000). TLC is more useful for large oligosaccharides, including those that bind paper (MLG-, cello- and manno-oligosaccharides). Suitable solvents include butan-1-ol/acetic acid/ water (2:1:1) and propan-1-ol/nitromethane/water (5:2:3). A useful stain for carbohydrates on silica gel is freshly prepared 0.5% (w/v) thymol in 96% ethanol/H2SO4 (20:1 v/v) – the plate is quickly dipped in this solution, dried, and then heated in an oven at 105 °C for 3–6 min. 1.6.3.3 Paper electrophoresis (PE) High-voltage PE (Fry 2000) is a surprisingly little-used technique that can, however, give excellent resolution of charged carbohydrates e.g. amino sugars, uronic acids, aldonic acids, aldaric acids, sugar phosphates and NDP sugars (e.g. Green & Fry 2005; Sharples & Fry 2007; Takeda et al. 2008; see also http://homepages.ed.ac.uk/sfry/service.html). Even neutral sugars can be analysed by PE if a borate buffer is used, with which the sugar reversibly forms an anionic borate ester (e.g. O’Looney & Fry 2005). Electrophoretic

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

35

mobility depends on both the charge (Q) of the compound (which at pH 6.5 will be about −1 in the case of GalA and GlcA→Man, and nearly −2 in the case of GalA→GalA) and its molecular weight (Mr). The migration rate towards the electrode (relative to a neutral marker) is proportional to Q/Mr2/3 (Offord 1966). Therefore, if several compounds with known Q and Mr are run on the same electrophoretogram, the mobility of an unknown radioactive spot can provide valuable data – if its charge is known, then its Mr can be estimated, and if its approximate Mr is known (e.g. that it is a disaccharide), then its charge can be estimated. 1.6.3.4 High-pressure liquid chromatography (HPLC) For details, see El Rassi (1995). Several forms of HPLC are useful for the resolution of mono- and oligosaccharides. In particular, a form of anionexchange chromatography developed by Dionex Inc. enables excellent separation of all the common monosaccharides and numerous diverse oligosaccharides. The analytes are readily quantified with a pulsed amperometric detector if it has been calibrated for the compounds under investigation. Samples require more cleaning up than for PC. Radioactivity monitors designed for HPLC eluates are not sensitive enough (compared with autoradiography or scintillation counting in vials) to be useful unless unusually high levels of radiolabelling are employed. 1.6.3.5 Methylation analysis and gas chromatography (GC) Methylation analysis is particularly useful when no suitable enzymes are available for ‘dissection’ of wall polysaccharides, or when detailed structural information is required about novel polysaccharides. Details are outside the scope of the present chapter (Waeghe et al. 1983; Harris et al. 1984; Carpita & Shea 1989). GC of sugars (or partially methylated sugars) requires them to be converted to volatile derivatives. Usually this involves reduction of the monosaccharide to an alditol with NaBH4 (or NaB2H4 so that the original reducing end can be followed during MS) followed by acetylation with acetic anhydride. Resolution of partially methylated alditol acetates by GC is excellent, as is quantification; GC is, however, rarely attempted with radiolabelled samples because of the difficulties involved in measuring and disposing of radiolabelled volatiles. 1.6.3.6 Mass spectrometry (MS) GC and HPLC are both frequently combined with MS, giving information about the masses of ions (strictly, the Mr/Q ratio, referred to in MS work as ‘m/z’) of the analytes and of the fragments formed from them inside the mass spectrometer. The m/z values can reveal the number and often the sequence of sugar residues of each class (e.g. pentose, hexose, deoxyhexose), but not the identity of the specific monosaccharide(s) (e.g. Glc, Gal and Man all have the same mass).

36

■

Plant Polysaccharides

1.7 Conclusions The recent development of convenient methods for genetic manipulation of plant cells, by which the genes for chosen wall-related enzymes can be targeted for up- or down-regulation in a spatially and temporally precise fashion, has given us the potential ability to alter artificially the production, modification, secretion, crosslinking and degradation of wall polysaccharides. There remains, however, the need to test experimentally whether the intended change in polysaccharide metabolism has actually been effected. Therefore, there is a need for simple methods by which wall polysaccharide composition and metabolism can be documented in vivo. A range of suitable methods for such endeavours has been compiled in this chapter. Of particular value are techniques of in-vivo radiolabelling (by which even quantitatively minor components can be detected, and the dynamics of their metabolism explored), enzymic dissection of wall polysaccharides, and separation of the digestion-products by chromatography and electrophoresis.

Acknowledgements I thank the UK BBSRC for funding in support of this work.

References Amino, S., Takeuchi, Y., Komamine, A. (1984) Changes in cell wall constituents during the cell cycle in a synchronous culture of Catharanthus roseus. Physiologia Plantarum, 60, 326–332. Andersson, S.I., Samuelson, O., Ishihara, M., Shimizu, K. (1983) Structure of the reducing end-groups in spruce xylan. Carbohydrate Research, 111, 283–288. Asensio, A. (1987) Quercus suber polysaccharides. 1. Structural studies of the hemicellulose-A from the cork of Quercus suber. Carbohydrate Research, 161, 167–170. Aspinall, G.O., Carlyle, J.J., McNab, J.M., Rudowski, A. (1969) Anogeissus leiocarpus gum. II. Fractionation of gum and partial hydrolysis of leiocarpan A. Journal of the Chemical Society C 1969, 840–845. Bauer, S., Vasu, P., Persson, S., Mort, A.J., Somerville, C.R. (2006) Development and application of a suite of polysaccharide-degrading enzymes for analyzing plant cell walls. Proceedings of the National Academy of Sciences of the U S A, 103, 11417–11422. Baydoun, E.A.-H., Fry, S.C. (1988) [2-3H]Mannose incorporation in cultured plant cells: investigation of l-galactose residues in the primary cell wall. Journal of Plant Physiology, 132, 484–490. Brett, C.T., Waldron, K.W. (1996) Physiology and Biochemistry of Plant Cell Walls, 2nd edn. Chapman & Hall, London. Brown, J.A., Fry, S.C. (1993a) The preparation and susceptibility to hydrolysis of novel O-galacturonoyl derivatives of carbohydrates. Carbohydrate Research, 240, 95–106.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

37

Brown, J.A., Fry, S.C. (1993b) Novel O-d-galacturonoyl esters in the pectic polysaccharides of suspension-cultured plant cells. Plant Physiology, 103, 993–999. Buliga, G.S., Brant, D.A., Fincher, G.B. (1986) The sequence statistics and solution conformation of a barley (1→3, 1→4)-β-d-glucan. Carbohydrate Research, 157, 139–156. Carnachan, S.M., Harris, P.J. (2000) Ferulic acid is bound to the primary cell walls of all gymnosperm families. Biochemical Systematics and Ecology, 28, 865–879. Carpita, N.C. (1984) Fractionation of hemicelluloses from maize cell-walls with increasing concentrations of alkali. Phytochemistry, 23, 1089–1093. Carpita, N.C. (1996) Structure and biogenesis of the cell walls of grasses. Annual Review of Plant Physiology and Plant Molecular Biology, 47, 445–76. Carpita, N.C., Shea, E.M. (1989) Linkage structure of carbohydrates by gas chromatography–mass spectrometry (GC–MS) of partially methylated alditol acetates. In: Analysis of Carbohydrates by GLC and MS (eds C.J Biermann, G.D. McGinnis). CRC Press, Boca Raton. Carpita, N.C., Defernez, M., Findlay, K., et al. (2001) Cell wall architecture of the elongating maize coleoptile. Plant Physiology, 127, 551–565. Cherno, N.K., Dudkin, M.S., Areshidze, I.V. (1976) Pectin substances of Chara aculeolata. Khimiya Prirodnykh Soedinenii, 6, 702–705. Coenen, G.J., Bakx, E.J., Verhoef, R.P., Schols, H.A., Voragen, A.G.J. (2007) Identification of the connecting linkage between homo- or xylogalacturonan and rhamnogalacturonan type I. Carbohydrate Polymers, 70, 224–235. Edashige, Y., Ishii, T. (1996) Structures of cell-wall polysaccharides from suspensioncultured cells of Cryptomeria japonica. Mokuzai Gakkaishi, 42, 895–900. Edelmann, H.G., Fry, S.C. (1992) Factors that affect the extraction of xyloglucan from the primary cell walls of suspension-cultured rose cells. Carbohydrate Research, 228, 423–431. El Rassi, Z. (ed.) (1995) Carbohydrate Analysis: High-Performance Liquid Chromatography and Capillary Electrophoresis. Journal of Chromatography Library, Vol. 58. Elsevier, Amsterdam. Encina, A., Fry, S.C. (2005) Oxidative coupling of a feruloyl-arabinoxylan trisaccharide (FAXX) in the walls of living maize cells requires endogenous hydrogen peroxide and is controlled by a low-Mr apoplastic inhibitor. Planta, 223, 77–89. Fry, S.C. (1982) Phenolic components of the primary cell wall: feruloylated disaccharides of d-galactose and l-arabinose from spinach polysaccharide. Biochemical Journal, 203, 493–504. Fry, S.C. (1983) Feruloylated pectins from the primary cell wall: their structure and possible functions. Planta, 157, 111–123. Fry, S.C. (1989) Cellulases, hemicelluloses and auxin-stimulated growth: a possible relationship. Physiologia Plantarum, 75, 532–536. Fry, S.C. (2000) The Growing Plant Cell Wall: Chemical and Metabolic Analysis, reprint edition, The Blackburn Press, Caldwell, NJ. Fry, S.C. (2001) Plant cell walls. In: Encyclopedia of Life Sciences. Nature Publishing Group, London (also electronic edition 1999, http://www.els.net). Fry, S.C., Mohler, K.E., Nesselrode, B.H.W.A., Franková, L. (2008a) Mixed-linkage β-glucan : xyloglucan endotransglucosylase, a novel wall-remodelling enzyme from Equisetum (horsetails) and charophytic algae. Plant Journal, 54, 451–521. Fry, S.C., Nesselrode, B.H.W.A., Miller, J.G., Mewburn, B. (2008b) Mixed-linkage (1→3,1→4)-β-d-glucan is a major hemicellulose of Equisetum (horsetail) cell walls. New Phytologist, 179, 104–115.

38

■

Plant Polysaccharides

Fry, S.C., Willis, S.C., Paterson, A.E.J. (2000) Intraprotoplasmic and wall-localised formation of arabinoxylan-bound diferulates and larger ferulate coupling-products in maize cell-suspension cultures. Planta, 211, 679–692. Funk, C., Ralph, J., Steinhart, H., Bunzel, M. (2005) Isolation and structural characterisation of 8–O–4/8–O–4-and 8–8/8–O–4-coupled dehydrotriferulic acids from maize bran. Phytochemistry, 66, 363–371. Gerwig, G.J., Kamerling, J.P., Vliegenthart, J.F.G. (1979) Determination of the absolute-configuration of monosaccharides in complex carbohydrates by capillary GLC. Carbohydrate Research, 77, 1–7. Gibeaut, D.M., Pauly, M., Bacic A., Fincher, G.B. (2005) Changes in cell wall polysaccharides in developing barley (Hordeum vulgare) coleoptiles. Planta, 221, 729–738. Green, M.A., Fry, S.C. (2005) Vitamin C degradation in plant cells via enzymatic hydrolysis of 4-O-oxalyl-l-threonate. Nature, 433, 83–88. Grishutin, S.G., Gusakov, A.V., Dzedzyulya, E.I., Sinitsyn, A.P. (2006) A lichenaselike family 12 endo-(l→4)-β-glucanase from Aspergillus japonicus: study of the substrate specificity and mode of action on β-glucans in comparison with other glycoside hydrolases. Carbohydrate Research, 341, 218–229. Handford, M.G., Baldwin, T.C., Goubet, F., et al. (2003) Localisation and characterisation of cell wall mannan polysaccharides in Arabidopsis thaliana. Planta 218, 27–36. Harris, P.J., Henry, R.J., Blakeney, A.B., Stone, B.A. (1984) An improved procedure for the methylation analysis of oligosaccharides and polysaccharides. Carbohydrate Research, 127, 59–74. Hayashi (1989) Xyloglucans in the primary cell wall. Annual Review of Plant Physiology and Plant Molecular Biology, 40, 139–168. Hoffman, M., Jia, Z., Peña, M.J., et al. (2005) Structural analysis of xyloglucans in the primary cell walls of plants in the subclass Asteridae. Carbohydrate Research, 340, 1826–1840. Honda, Y., Inaoka, H., Takei, A., Sugimura, Y., Otsuji, K. (1996) Extracellular polysaccharides produced by tuberose callus. Phytochemistry, 41, 1517–1521. Hrmová, M., Farkaš, V., Lahnstein J., Fincher, G.B. (2007) A barley xyloglucan xyloglucosyl transferase covalently links xyloglucan, cellulosic substrates, and (1,3;1,4)-β-d-glucans. Journal of Biological Chemistry, 282, 12951–12962. Ishii, T. (1997) O-Acetylated oligosaccharides from pectins of potato tuber cell walls. Plant Physiology, 113, 1265–1272. Ishii, T., Matsunaga, T. (2001) Pectic polysaccharide rhamnogalacturonan II is covalently linked to homogalacturonan. Phytochemistry, 57, 969–974. Kato, Y., Ito, S., Iki, K., Matsuda, K. (1982) Xyloglucan and β-d-glucan in cell walls in rice seedlings. Plant and Cell Physiology, 23, 351–364. Kato, Y., Nevins, D.J. (1985) Isolation and identification of O-(5-O-feruloyl-α-larabinofuranosyl)-(1→3)-O-β-d-xylopyranosyl-(1→4)-d-xylopyranose as a component of Zea shoot cell walls. Carbohydr Res, 137, 139–150. Kerr, E.M., Fry, S.C. (2003) Pre-formed xyloglucans and xylans increase in molecular weight in three distinct compartments of a maize cell-suspension culture. Planta, 217, 327–339. Kim J-B., Carpita, N.C. (1992) Changes in esterification of the uronic acid groups of cell wall polysaccharides during elongation of maize coleoptiles. Plant Physiology, 98, 646–653.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

39

Kutschera, U. (2008) The growing outer epidermal wall: design and physiological role of a composite structure. Annals of Botany 101, 615–621. Labavitch, J.M., Ray, P.M. (1978) Structure of hemicellulosic polysaccharides of Avena sativa coleoptile cell walls. Phytochemistry, 17, 933–937. Lenucci, M., Piro, G., Miller, J.G., Dalessandro, G., Fry, S.C. (2005) Do polyamines contribute to plant cell wall assembly by forming amide bonds with pectins? Phytochemistry, 66, 2581–2594. Li, W., Cui, S.W., Kakuda, Y. (2006) Extraction, fractionation, structural and physical characterization of wheat β-d-glucans. Carbohydrate Polymers, 63, 408–416. Lindsay, S.E., Fry, S.C. (2008) Control of diferulate formation in dicotyledonous and gramineous cell-suspension cultures. Planta, 227, 439–452. Longland, J.M., Fry, S.C., Trewavas, A.J. (1989) Developmental control of apiogalacturonan biosynthesis and UDP-apiose production in a duckweed. Plant Physiology, 90, 972–976. Matsunaga, T., Ishii, T., Matsumoto, S., et al. (2004) Occurrence of the primary cell wall polysaccharide rhamnogalacturonan II in pteridophytes, lycophytes, and bryophytes. implications for the evolution of vascular plants. Plant Physiology, 134, 339–351. McCleary, B.V., Matheson, N.K. (1983) Action patterns and substrate-binding requirements of β-d-mannanase with mannosaccharides and mannan-type polysaccharides. Carbohydrate Research, 119, 191–219. McDougall, G.J., Fry, S.C. (1994) Fucosylated xyloglucan in suspension-cultured cells of the gramineous monocotyledon, Festuca arundinacea. Journal of Plant Physiology, 143, 591–595. Meikle, P.J., Hoogenraad, N.J., Bonig, I., Clarke, A.E., Stone, B.A. (1994) A (1→3,1→4)-β-glucan-specific monoclonal antibody and its use in the quantification and immunocytochemical location of (1→3,1→4)-β-glucans. Plant Journal, 5, 1–9. Melton, L.D., McNeil, M., Darvill, A.G., Albersheim, P., Dell, A. (1986) Structural characterization of oligosaccharides isolated from the pectic polysaccharide rhamnogalacturonan II. Carbohydrate Research, 146, 279–305. Miller, J.G., Buchanan, C.J., Eastwood, M.A., Fry, S.C. (1995) Specific 14C-labelling of spinach cell walls in their methyl, acetyl or uronic acid residues. Carbohydrate Letters, 1, 129–136. Monro, J.A., Penny, D., Bailey, R.W. (1976) The organization and growth of primary cell walls of lupin hypocotyl. Phytochemistry, 15, 1193–1198. Mort, A.J. (2002) Interactions between pectins and other polymers. Pp 30–51 in Pectins and their Manipulation, ed. by G.B. Seymour and J.P. Knox, Blackwell Publishing Ltd., Oxford. Mutter, M., Renard, C.M.G.C., Beldman, G., Schols, H.A., Voragen, A.G.J. (1998) Mode of action of RG-hydrolase and RG-lyase toward rhamnogalacturonan oligomers. Characterization of degradation products using RG-rhamnohydrolase and RG-galacturonohydrolase. Carbohydrate Research, 311, 155–164. Nishikubo, N., Awano, T., Banasiak, A., et al. (2007) Xyloglucan endo-transglycosylase (XET) functions in gelatinous layers of tension wood fibers in poplar – a glimpse into the mechanism of the balancing act of trees. Plant and Cell Physiology, 48, 843–855. O’Looney, N., Fry, S.C. (2005) Oxaziclomefone, a new herbicide, inhibits wall expansion in maize cell-cultures without affecting polysaccharide biosynthesis,

40

■

Plant Polysaccharides

xyloglucan transglycosylation, peroxidase action or apoplastic ascorbate oxidation. Annals of Botany, 96, 1097–1107. O’Neill, M.A., York, W.S. (2003) The composition and structure of plant primary cell walls. In: The Plant Cell Wall (ed. J.K.C. Rose), pp. 1–54. Blackwell Publishing Ltd., Oxford. O’Neill, M.A., Eberhard, S., Albersheim, P., Darvill, A.G. (2001) Requirement of borate cross-linking of cell wall rhamnogalacturonan II for Arabidopsis growth. Science, 294, 846–849. O’Neill, M.A., Ishii, T., Albersheim, P., Darvill, A.G. (2004) Rhamnogalacturonan II: structure and function of a borate cross-linked cell wall pectic polysaccharide. Annual Review of Plant Biology, 55, 109–139. Obel, N., Neumetzler, L., Pauly, M. (2006) Hemicelluloses and cell expansion. In: The Expanding Cell (eds J.-P. Verbelen, K. Vissenberg), Plant Cell Monograph 5, pp. 57–88. Springer, Berlin. Offord, R.E. (1966) Electrophoretic mobilities of peptides on paper and their use in the determination of amide groups. Nature, 211, 591–593. Ortega, L.I., Fry, S.C., Taleisnik, E. (2006) Why are Chloris gayana leaves shorter in salt-affected plants? Analyses in the elongation zone. Journal of Experimental Botany, 57, 3945–3952. Painter, T.J. (1983) Residues of d-lyxo-5-hexosulopyranuronic acid in Sphagnum holocellulose and their role in cross-linking. Carbohydrate Research, 124, C18–C21. Pauly, M., Andersen, L.N., Kauppinen, S., et al. (1999) A xyloglucan-specific endoβ-1,4-glucanase from Aspergillus aculeatus: expression cloning in yeast, purification and characterization of the recombinant enzyme. Glycobiology, 9, 93–100. Peña, M.J., Darvill,A.G., York,W.S., O’Neill M.A. (2007a) Structural diversity of xyloglucans in the cell walls of land plants. Physiologia Plantarum, 130, Abstract 18, XIth Cell Wall Meeting. Peña, M.J., Zhong, R.Q., Zhou, G.K., et al. (2007b) Arabidopsis irregular xylem8 and irregular xylem9: implications for the complexity of glucuronoxylan biosynthesis. Plant Cell, 19, 549–563. Perrone, P., Hewage, C., Sadler, I.H., Fry, S.C. (1998) Nα- and Nε-d-galacturonoyl-llysine amides: properties and possible occurrence in plant cell walls. Phytochemistry, 49, 1879–90. Perrone, P., Hewage, C.M., Thomson, A.R., Bailey, K., Sadler, I.H., Fry, S.C. (2002) Patterns of methyl and O-acetyl esterification in spinach pectins: new complexity. Phytochemistry, 60, 67–77. Popper, Z.A., Fry, S.C. (2003) Primary cell wall composition of bryophytes and charophytes. Annals of Botany, 91, 1–12. Popper, Z.A., Fry, S.C. (2004) Primary cell wall composition of pteridophytes and spermatophytes. New Phytologist, 164, 165–174. Popper, Z.A., Fry, S.C. (2005) Widespread occurrence of a covalent linkage between xyloglucan and acidic polysaccharides in suspension-cultured angiosperm cells. Annals of Botany, 96, 91–99. Popper, Z.A., Fry, S.C. (2008) Xyloglucan–pectin linkages are formed intraprotoplasmically, contribute to wall-assembly, and remain stable in the cell wall. Planta, 227, 781–794. Popper, Z.A., Sadler, I.H., Fry, S.C. (2003) α-d-Glucuronosyl-(1→3)-l-galactose, an unusual disaccharide from polysaccharides of the hornwort Anthoceros caucasicus. Phytochemistry, 64, 325–335.

Cell Wall Polysaccharide Composition and Covalent Crosslinking

■

41

Redgwell, R.J., O’Neill, M.A., Selvendran, R.R., Parsley, K.J. (1986) Structural features of the mucilage from the stem pith of kiwifruit (Actinidia deliciosa). 1. structure of the inner core. Carbohydrate Research, 153, 97–106. Rocha, S.M., Coimbra, M.A., Delgadillo, I. (2000) Demonstration of pectic polysaccharides in cork cell wall from Quercus suber L. Journal of Agricultural and Food Chemistry, 48, 2003–2007. Schols, H.A., Voragen, A.G.J. (2002) The chemical nature of pectins. In: Pectins and their Manipulation (eds G.B. Seymour, J.P. Knox), pp. 1–29. Blackwell Publishing Ltd., Oxford. Sharples, S.C., Fry, S.C. (2007) Radioisotope ratios discriminate between competing pathways of cell wall polysaccharide and RNA biosynthesis in living plant cells. Plant Journal, 52, 252–262. Smith, B.G., Harris, P.J. (2001) Ferulic acid is esterified to glucuronoarabinoxylans in pineapple cell walls. Phytochemistry, 56, 513–519. Sørensen, I., Pettolino, F.A., Wilson, S.M., et al. (2008) Mixed linkage (1→3),(1→4)-βd-glucan is not unique to the Poales and is an abundant component of Equisetum arvense cell walls. Plant Journal, 54, 451–521. Sotiriou, P., Fry, S.C., Spyropoulos, C.G. (2007) Protoplast isolation and culture from carob (Ceratonia siliqua L.) hypocotyls: ability of regenerated protoplasts to produce mannose-containing polysaccharides. Physiologia Plantarum, 130, 11–22. Takeda, T., Miller, J.G., Fry, S.C. (2008) Anionic derivatives of xyloglucan function as acceptor but not donor substrates for xyloglucan endotransglucosylase activity. Planta, 227, 893–905. Teleman, A., Nordström, M., Tenkanen, M., Jacobs, A., Dahlmana, O. (2003) Isolation and characterization of O-acetylated glucomannans from aspen and birch wood. Carbohydrate Research, 338, 525–534. Thimm, J.C., Burritt, D.J., Sims, I.M., Newman, R.H., Ducker, W.A., Melton, L.D. (2002) Celery (Apium graveolens) parenchyma cell walls: cell walls with minimal xyloglucan. Physiologia Plantarum, 116, 164–171. Thomas, J.R., Darvill, A.G., Albersheim, P. (1989) The structure of plant cell walls XXIV. Isolation and structural characterization of the pectic polysaccharide rhamnogalacturonan II from walls of suspension-cultured rice cells. Plant Physiology, 185, 261–277. Thompson, D.S. (2008) Space and time in the plant cell wall: relationships between cell type, cell wall rheology and cell function. Annals of Botany, 101, 203–211. Thompson, J.E., Fry, S.C. (2000) Evidence for covalent linkage between xyloglucan and acidic pectins in suspension-cultured rose cells. Planta, 211, 275–286. Thornber, J.P., Northcote, D.H. (1962) Changes in the chemical composition of a cambial cell during its differentiation into xylem and phloem tissues in trees. 3. Xylan, glucomannan and α-cellulose fractions. Biochemical Journal, 82, 340–346. Ulvskov, P., Wium, H., Bruce, D., et al. (2005) Biophysical consequences of remodeling the neutral side chains of rhamnogalacturonan I in tubers of transgenic potatoes. Planta, 220, 609–620. Wada, S., Ray, P.M. (1978) Matrix polysaccharides of oat coleoptile cell walls. Phytochemistry, 17, 923–931. Waeghe, T.J., Darvill, A.G., McNeil, M., Albersheim, P. (1983) Determination by methylation analysis of the glycosyl-linkage compositions of microgram quantities of complex carbohydrates. Carbohydrate Research, 123, 281–304.

42

■

Plant Polysaccharides

Wende, G., Fry, S.C. (1996) 2-O-β-d-Xylopyranosyl-(5-O-feruloyl)-l-arabinose, a widespread component of grass cell walls. Phytochemistry, 44, 1019–1030. Willför, S., Sundberg, A., Pranivich, A., Holmbom, B. (2005) Polysaccharides in some industrially important hardwood species. Wood Science and Technology, 39, 601–617. Woodward, J.R., Fincher, G.B., Stone, B.A. (1983) Water-soluble (1→3),(1→4)-β-dglucans from barley (Hordeum vulgare) endosperm. II. Fine structure. Carbohydrate Polymers, 3, 207–225. Zandleven, J., Beldman, G., Bosveld, M., Benen, J., Voragen, A. (2005) Mode of action of xylogalacturonan hydrolase towards xylogalacturonan and xylogalacturonan oligosaccharides. Biochemical Journal, 387, 719–725. Zhang, Z.Q., Pierce, M.L., Mort, A.J. (2007) Changes in homogalacturonans and enzymes degrading them during cotton cotyledon expansion. Phytochemistry, 68, 1094–1103. Zykwinska, A.W., Ralet M-C.J., Garnier, C.D., Thibault J-F.J. (2005) Evidence for in vitro binding of pectin side chains to cellulose. Plant Physiology, 139, 397–407.

Annual Plant Reviews (2011) 41, 43–64 doi: 10.1002/9781444391015.ch2

Chapter 2

DISSECTION OF PLANT CELL WALLS BY HIGH-THROUGHPUT METHODS Staffan Persson1, Iben Sørensen2, Isabel Moller2, William Willats2 and Markus Pauly3 1

Max-Planck-Institute for Molecular Plant Physiology, Am Muehlenberg 1, 14476 Potsdam, Germany 2 Department of Biology, University of Copenhagen, Ole Maaløes vej 5, 2200 Copenhagen N, Denmark 3 Department of Plant and Microbial Biology, Energy Biosciences Institute, University of California, Berkeley, Berkeley, CA 94720, USA Manuscript received October 2008

Abstract: The structural complexity and heterogeneity of plant cell walls presents major obstacles to understanding the precise role of their constituent polysaccharides in plant growth and development. Further complexity is added as cell wall structures are modulated during plant development and in response to environmental cues. A variety of techniques have been developed to help advance our understanding of cell wall structures, and functions. Recent advances in immunocytochemistry, monosaccharide composition and linkage analysis, MALDI-TOF oligosaccharide fingerprinting and carbohydrate microarrays now provide cell wall biologists with powerful, robust and in some cases highthroughput tools for accurately determining cell wall structures. This chapter covers some of the latest developments in cell wall analysis, and also explores the prospects for effectively integrating different analytical platforms. Keywords: capillary electrophoresis; cell wall proteome; chemical genetics; coexpression; comprehensive microarray polymer profiling; Fourier transform infrared spectroscopy; MALDI-TOF; oligosaccharide; OLIMP; PACE; plant cell wall

Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

43

44

■

Plant Polysaccharides

2.1 Introduction Plant cell walls are complex and dynamic structures composed mainly of polysaccharides, polyphenols and highly glycosylated proteins. The structures of individual polysaccharides are very complex and their relative abundance and exact composition vary from one plant species to another, and also from tissue to tissue within a plant (for review see Carpita & McCann 2000; Somerville et al. 2004). The analysis of polysaccharide structure and content is therefore a challenging task and requires specialized techniques which, in most cases, differ from those used to characterize other types of biological macromolecules. These techniques include infrared spectroscopy, monosaccharide composition- and linkage analysis, immunohistochemistry, mass spectrometry and one- and two dimensional NMR of oligosaccharide fragments. These approaches are typically combined to elucidate alterations in cell wall structures between different species, for various tissues within a single plant, and for plants that have been genetically modified. However, approximately 2000 gene products, or one-tenth of the genome, in the model plant Arabidopsis thaliana are anticipated to be involved in the synthesis and/or modification of the cell wall polysaccharides (Somerville et al. 2004). It is therefore expected that high-throughput technologies for the individual techniques will be in strong demand, and several such efforts have been, and are currently being, developed. This chapter attempts to give an overview of several of these platforms (Fig. 2.1), and to give the reader a sense of how and when these approaches may be of use.

2.2 Enzyme fingerprinting The compositions of individual polysaccharides can be very complex, and analysis of their structures is therefore often challenging. Since the cell wall is made of different types of polymers, a first analytic step is usually to enrich for specific polysaccharides. This may be done through sequential extraction from the alcohol-insoluble residue of the plant material (York et al. 1985; MacDougall et al. 1997; Carpita & McCann 2000). These fractions can then be analysed by a range of techniques of which the most frequently used is a monosaccharide composition analysis. This can be done by hydrolysing the solubilized wall poly-/oligosaccharides with acids such as 2 M trifluoracetic acid (Albersheim et al. 1967). Certain monosaccharides can then be quantified by colorimetric assays such as phenol–sulfuric acid for hexoses and pentoses (Dubois et al. 1956), the anthrone assay for hexoses (Dische 1962), the m-hydroxybiphenyl assay for uronic acids (Blumenkrantz & Asboe-Hansen 1973), or by enzyme-coupled assays (McCleary & Codd 1991). Other techniques result first in the separation and subsequently in the quantification of the sugars, for example by the use of thin layer chromatography, paper electrophoresis (Fry 1988), high-performance anion exchange

Dissection of Plant Cell Walls by High-Throughput Methods

Genomic and proteomic analyses Chemical genetics

Microarray based analyses

■

45

Species comparisons

Proteomics

Genetic modifications of candidate genes

Tissue and organspecific comparisons

Cell wall analyses

Whole plant tissue Cell wall extract

CWB arrays

CoMPP

NMR

PACE

OLIMP

FTIR

Additional structural and compositional analyses of cell wall fractions Figure 2.1 Schematic overview of the ‘high-throughput’ plant cell wall analyses platforms. CoMPP, comprehensive microarray polymer profiling; CWB arrays, cell wall binding arrays; FTIR, Fourier transform infrared spectroscopy; NMR, nuclear magnetic resonance; OLIMP, oligosaccharide mass profiling; PAGE, polyacrylamide gel electrophoresis.

chromatography (HPAEC; Øbro et al. 2004) or by GC-MS analysis after monosaccharide derivatization to their corresponding alditol acetates or trimethylsilane derivatives (Albersheim et al. 1967; York et al. 1985). Additional information about the polysaccharide can be obtained by determination of the glycosidic linkage of the sugars present in the wall sample (Ciucanu & Luca 1990). Non-carbohydrate substituents such as acetyl or methyl esters can also be determined (see Klavons & Bennet 1986; Voragen et al. 1986). However, determination of the sequence and arrangement of the sugar moieties usually involves detailed characterization of smaller oligosaccharide fragments. The polysaccharides can be fragmented by somewhat selective chemical methods, such as partial acid hydrolysis, or by endo-hydrolytic enzymes (Mort & Pierce 1994). A crucial step in this type of analysis is the hydrolysis of the polysaccharides into oligomers that are used for subsequent analyses.

46

■

Plant Polysaccharides

Chemical fragmentations are not specific for any one glycosidic bond in a polymer and therefore result in a mixture of oligomers. These oligomers are generally difficult to purify and this also leads to relatively low yields of certain oligosaccharides. A more efficient approach is to selectively cleave polymers using endo-hydrolases. The specificity of such enzymes depends on purity, i.e. the removal of undesired enzymatic activities. Most plant saprophytes secrete multiple enzymes for degrading plant polysaccharides, and the isolation of single enzymes therefore involves extensive purification (Vries & Visser 2001). A more useful approach would be to clone the genes encoding different endo-hydrolases and then express them in a heterologous system. Enzymes with different substrate specificities that cleave bonds e.g. in xylan (Fernández-Espinar et al. 1992; Fernández-Espinar et al. 1993; Fernández-Espinar et al. 1994; Kumar & Ramón 1996; Pérez-González et al. 1996 Galagan et al. 2005), pectate (Pérez-González et al. 1998), arabinan (Ramón et al. 1993; Fernández-Espinar et al. 1994; Ho et al. 1995), glucans (Chikamatsu et al. 1999; Gielkens et al. 1999), and other polysaccharides (Lockington et al. 2002) have been cloned and characterized in such way. For example, a hydrolase cloned from Aspergillus aculeatus was biochemically characterized by heterologous expression in Aspergillus oryzae, and shown to encode a xyloglucan-specific endo-β-1,4-glucanase (Pauly et al. 1999). In another study, a xyloglucan endo-glucanase from Geotrichum sp. M128 was cloned and characterized by expressing it in E. coli (Yaoi & Mitsuishi 2004). It therefore appears that these enzymes may be of great use to hydrolyse specific bonds in certain polysaccharides. However, as exemplified above, most laboratories have their own preferred expression systems and protein tags, and it would therefore be quite time- and labour intensive to obtain a collection of pure hydrolases. Recent efforts have overcome this obstacle. A suite of 74 cell wall hydrolytic enzymes from Aspergillus nidulans (72), Aspergillus fumigatus (1) and Neurospora crassa (1) was cloned, fused to a 6×His tag and expressed in Pichia pastoris (Bauer et al. 2006). These enzymes may subsequently be expressed and purified using a single strategy. The collection included enzymes that are active on the major cell wall polymers such as cellulose, hemicelluloses (glucans, xyloglucans, xylans, mannans) and pectins (polygalacturonic acid, xylogalacturonans and rhamnogalacturonan I). Although substrates for most of the enzymes were identified, several did not exhibit activity on the most common polymers and may therefore be active on glycosidic bonds that are less common and/or characterized in the plant cell wall (Bauer et al. 2006). More extensive information about the enzymes may be obtained from http://www-ciwdpb.stanford.edu/research/csomerville/enzymes.php. However, it should also be noted that owing to the supramolecular structure of wall polymers, such as the formation of aggregates via covalent and non-covalent bonds, endo-hydrolases may not be able to solubilize a particular polysaccharide completely out of the wall. Only a certain proportion of the polymer may therefore be released. For example, it has been shown that the above-mentioned xyloglucan-specific endo-glucanase releases only

Dissection of Plant Cell Walls by High-Throughput Methods

■

47

Arbitrary abundance

XXXG

XXLG/ XLXG

XXFG

XXFG+Ac

XLFG XLFG+Ac

1000

1700 m/z

Figure 2.2 MALDI-TOF mass spectra of endo-glucanase-generated xyloglucan fragments from Arabidopsis seedlings. Structural features for the fragments are indicated; Dots: glucose, squares: xylose, circles: galactose, triangles: fucose, small dots: acetyl groups. Modified from Lerouxel et al. (2002).

approximately one-third of the xyloglucan present in a primary cell wall (Fig. 2.2; Pauly et al. 1999). The remaining xyloglucan portions/domains were released with either strong alkali and/or cellulases, which resulted in a concomitant solubilization of cellulose (Pauly et al. 1999).

2.3 Structural determination of oligosaccharides Once oligosaccharides have been released from the wall preparations they can be separated and quantified using a variety of chromatographic techniques, such as HPLC, capillary electrophoresis (CE), and polyacrylamide based gel-electrophoresis (Evangelista et al. 1995; Broberg et al. 2000; Stroop et al. 2002; Ray et al. 2004). The purified oligomers can then be analysed by mass spectrometric techniques (MS) such as electrospray (ESI), or matrixassisted laser-desorption ionization (MALDI; Fig. 2.2; Vierhuis et al. 2001; Strasser & Amado 2002; Harvey 2006) to obtain the corresponding mass of the different oligosaccharides. For complete molecular characterization of the oligomers, one- and two-dimensional NMR spectroscopy should be employed (York et al. 1990; Broberg et al. 2000; Cardoso et al. 2002). HPAEC in combination with a pulsed amperometric detector is a well established liquid chromatography procedure (Lee 1996). For this method oligosaccharides do not need to be derivatized. The separation of oligosaccharides using specialized columns is superior to standard columns, and the detection very sensitive (pmol range). Structural isomers can often be baseline separated, e.g. two xyloglucan octasaccharides containing single

48

■

Plant Polysaccharides

galactosyl residues at different positions could be separated (Lerouxel et al. 2002). Because of the strongly alkaline nature of the eluant, ester substituents on the oligosaccharides such as O-acetyl or methyl esters are generally lost. The equipment requires a substantial amount of capital investment, samples can only be run one at a time and the runs take rather a long time. For example, the separation of xyloglucan oligosaccharides normally takes 40 min (McDougall & Fry 1991), and the separation of glucuronoxylan oligosaccharides 80 min (Quemener et al. 2007). Once calibration curves have been established with standard compounds the method is quantitative and can therefore result in a quantitative oligosaccharide profile of the wall. The eluant can be desalted and collected; eluant fractions can then be further analysed via MS or NMR to determine their content (Daas et al. 1998; Ray et al. 2004). MS without pre-purification is a rapid technique for the structural analysis of enzyme-solubilized oligomers. Depending on the ionization mode, the technique allows the determination of the exact molecular weight of oligosaccharides present in a mixture in the range of 400–5000 Da. For MALDITOF oligosaccharides are co-dried on a sample plate with a light absorbing matrix chemical such as dihydroxybenzoic acid (DHB; Mohr et al. 1995). The advantage of using MALDI-TOF MS is that very small amounts of material can be used for analysis (100 ng), and that the time for analysis is in the minute range. The enzymatically released oligosaccharides usually exhibit similar ionization properties and thus the abundances of oligosaccharide results in quantitative information as the relative abundance ratios of the various oligosaccharide ions can be ascertained (Fig. 2.2; Lerouxel et al. 2002). However, the absolute amount of an oligosaccharide cannot be determined by MS. If the specificity (cleavage site) of the enzyme on the wall polysaccharide is known, detailed structural information about the sidechain distribution of the wall polysaccharide can be obtained. Because of the mild enzymatic release, labile substituents, such as ester substitutents, still remain on the oligosaccharides and their nature and abundance on an oligosaccharide can be determined (Mouille et al. 2006). The use of this technique to rapidly assess a large number of wall materials has been termed OLIMP (oligosaccharide mass profiling) and allows high-throughput analysis (Lerouxel et al. 2002). Examples of polysaccharides that can be analysed by OLIMP include xyloglucan, and also pectic polysaccharides such as xylogalacturonan (Zandleven et al. 2007), galacturonic acid oligomers (Obel et al. 2006; Lionetti et al. 2007) and xylan fragments (Zhong et al. 2005). One disadvantage of OLIMP is the capital investment required for a mass spectrometer. In addition, monosaccharides cannot be detected because of their low molecular weight, and obviously structural oligosaccharide isomers cannot be distinguished unless the ion is further fragmented and analysed by MS (Yamagaki et al. 1998). Other mass spectrometry ionization techniques for the analysis of carbohydrate oligosaccharides include fast atom bombardment (FAB; Kolli et al. 1998), electrospray ionization (ESI; Matsunaga et al.

Dissection of Plant Cell Walls by High-Throughput Methods

■

49

2004), and pyrolysis mass spectrometry combined with chemical ionization (Lomax et al. 1991). Another technique that is commonly used is polyacrylamide-based gel electrophoresis (PACE) analysis of oligosaccharides. This type of analysis is able to detect both monosaccharides and oligosaccharides quantitatively manner and does not require a major capital investment (Goubet et al. 2002; Lerouxel et al. 2002). After hydrolysis the resulting glycosides are derivatized with a charged or uncharged fluorophore (e.g. AMAC, 2-aminoacridone or ANTS, 8-aminonaphtalene-1,3,6-trisulfonic acid; Goubet et al. 2002). The separation of the derivatives is then performed by PACE using a fluorescence imager for detection. The use of the fluorophore means that minute amounts of material can be detected (100 fmol of oligosaccharides or 500 fmol of monosaccharides; Goubet et al. 2002). Detected oligosaccharide bands can be given an absolute quantification. In addition, at least two studies have reported that structural isomers can be separated (Goubet et al. 2002; Goubet et al. 2006). Depending on the severity of the derivatization conditions, labile substituents may be detected (Goubet et al. 2003). One drawback of polysaccharide analysis by carbohydrate gel electrophoresis (PACE) is that the labelled compounds on a gel can only be identified by retention time of standards, and have to be excised for additional analysis such as MS or NMR to ascertain their precise structure. A number of wall polymers have been analysed by PACE including oligogalacturonans (Goubet et al. 2003), galactans and arabinans (Barton et al. 2006), secondary wall glucomannans (Benová-Kákosová et al. 2006), and xylan oligosaccharides (Beaugrand et al. 2004). An alternative method for the separation of fluorescently labelled oligosaccharides is capillary electrophoresis (CE; Rassi & Mechref 1998). A commonly used technique is the use of laser-induced fluorescence (LIF; Guttman 1996). The results are in quite good agreement qualitatively and quantitatively with PACE (Goubet et al. 2005). Some advantages of CE over conventional chromatographic methods are the smaller amounts of material (attomol range) that can be used, high separation efficiency, and the relative short time required for the analysis. For example, a typical run for the separation of pectic oligosaccharides takes 15 min (Zhong et al. 1998; Williams et al. 2002). However, as with gel electrophoresis of fluorescently labelled compounds, only the retention time is obtained; again, standards are necessary to identify any oligosaccharide unambiguously. Alternatively, the coupling of CE to MS allows more specific and selective detection (Kabel et al. 2006). Since ions can be formed from liquid droplets emerging from the CE, electrospray ionization is most often used. Yet another approach of oligosaccharide profiling is by nuclear magnetic resonance (NMR). In this method a mixture of oligosaccharides are subjected to 1H-NMR analysis and the compounds of the mixture can be identified and quantified by their diagnostic reducing end anomeric reporter signal (Perrin et al. 2003). For this approach neither separation nor derivatization

50

■

Plant Polysaccharides

Table 2.1 Summary of oligosaccharide profiling techniques of cell wall polysaccharides

Equipment required

Sample amount

Analysis time per sample (min)

HPAEC

Auto sampler, pump,PAD detector

5–15 μg

40–120

OLIMP

MALDI-TOF

1 μg

1

PACE

Electrophoresis apparatus, fluorescence imager

100 fmol

45 (parallel samples)

CE

CE instrument, detector

attomol

15

NMR

NMR spectroscope

100 μg

60–120

Method

Data obtained Peaks have to be identified with standards, quantitative, structural isomers can be detected Exact mass of the fragments, not quantitative (only relative) Gel bands have to be identified with standards, quantitative, structural isomers can be detected Peaks have to be identified with standards, quantitative Quantitative, exact structures (if standards for calibration were available)

CE, capillary electrophoresis; HPAEC, high-performance anion exchange chromatography; MALDI, matrix-assisted laser-desorption ionization; NMR, nuclear magnetic resonance; OLIMP, oligosaccharide mass profiling; PACE, polyacrylamide gel electrophoresis. Modified from Immerzeel & Pauly (2006).

is necessary prior to analysis, nor is the method destructive to the compounds. However, relatively large amounts of material are needed (100 μg range), and reference spectra of all compounds present in mixtures have to be generated and interpreted. A comparative summary of most attributes of the methods described in this section is shown in Table 2.1.

2.4 Fourier transform infrared spectroscopy (FTIR) Fourier transform infrared spectroscopy (FTIR) is another spectroscopic technique commonly used for cell wall analysis. The technique is based on vibrations, or energy transitions, occurring in different chemical bonds. This means that different chemical bonds absorb infrared, or near-infrared, light

Dissection of Plant Cell Walls by High-Throughput Methods

■

51

at different wavelengths which is quite useful for simpler samples, where the absorption spectra give distinct patterns that may be matched with external references. However, the cell wall typically consists of a large variety of polymers and thus of chemical bonds. Considerable overlap in absorption and vibrational coupling between bonds are therefore common (Kemsley et al. 1994; Séné et al. 1994). FTIR spectroscopy is a very rapid and non-invasive method that can, in theory, generate quantitative estimates of various functional groups in different macromolecules, including carboxylic esters, protein amides and carboxylic acids. FTIR may also be used to obtain information about carbohydrate constituents, in particular the so-called fingerprint region which corresponds to the wavenumber region 900–2000 cm−1 (McCann et al. 1992; Kemsley et al. 1994). Vibrations in this region are largely due to the intramolecular, or skeletal, structures of the polymers and even very similar structures may be separated within this spectral region. The technique is rapid (a spectrum is obtained in less than a minute), non-destructive and does not require derivatization, which makes it a useful initial screen that may be viewed as a platform for subsequent cell wall analyses. Comparing spectra to each other in order to gain the most significant information is a challenging task. The intricacy and variability of different spectra makes it quite difficult to gain any relevant information from visual comparisons of spectra. A variety of statistical methods have therefore been utilized to compare two or more spectra (Kemsley et al. 1994; Chen et al. 1998; Mouille et al. 2003; McCann et al. 2007). The approach that has been most widely used is principal component analysis (PCA; Massart et al. 1990). This approach transforms data points to a new coordinate system in which any projection of the data that produces the greatest variance is referred to as the first coordinate. This approach has been proved quite useful and readily identified mutants that had significantly altered cell wall composition compared to control samples (Chen et al. 1998). Alternative approaches include hierarchical clustering based on the Mahalanobis distance between samples (Mouille et al. 2003) and artificial neural network analysis (McCann et al. 2007). The hierarchical clustering approach was undertaken on 51 Arabidopsis cell wall mutant lines, additional chemically treated wild type lines and controls (Mouille et al. 2003). Based on 39 selected wavenumbers, covering the range between 830 and 1800 cm−1, the study clustered known cellulose-deficient mutants with each other and showed that allelic mutants of similar strength also clustered together. The neural networks approach was applied to etiolated maize coleoptiles and was shown to discriminate between half-day intervals of growth (McCann et al. 2007). The study used two different network algorithms, a supervised and an unsupervised one, to assign unknown samples to spectral classes. The approach was judged to be superior to the PCA method, which did not resolve several of these samples from each other. However, the authors also point out that this approach may have limitations since the nature of the spectral classes must be known prior to analysis (McCann et al. 2007). A similar limitation is also

52

■

Plant Polysaccharides

evident for the clustering of FTIR spectra, since the dendrograms per se do not give any information about the cell wall deficiencies. Another interesting approach to the measurement of cell wall mechanical properties is the introduction of two-dimensional (2D) FTIR spectroscopy (Wilson et al. 2000; Kokot et al. 2002). This technique allows mounted tissue to be viewed with IR spectroscopy before, during and after mechanical stress. These observations may then be used to assess the orientation of cell wall polymers and also to predict interactions between different polymers (Wilson et al. 2000). For example, 2DFTIR was used to measure the mechanical characteristics of the wall polymers during stretching of plant tissue. The study concluded that the mechanical strain was first sensed by the pectic matrix, in particular homogalcturonan, and when the strain increased it was also sensed by the cellulosic part of the cell wall (Wilson et al. 2000).

2.5 Microarray-based polymer profiling Recent research has demonstrated that the specificity of monoclonal antibodies (mAbs) and carbohydrate binding modules (CBMs) can be combined with the high-throughput capacity of microarrays for cell wall research (Moller et al. 2007). Two basic approaches have been developed. In one, cell wall polymers are sequentially extracted, spotted as microarrays onto membranes and probed with mAbs and CBMs. This approach, known as comprehensive microarray polymer profiling (CoMPP) is used to rapidly profile cell wall composition and to compare cell walls, for example between different species or between mutant and wild type plants. With CoMPP, no attempt is made to purify cell wall polymers, which are spotted as complex mixtures. The relative abundance of individual glycans is assessed by the ability of the probes used to bind to specific structural features (epitopes). Established biochemical techniques, such as monosaccharide linkage and composition analysis are quantitative but low-throughput and are indirectly informative about polysaccharide levels. CoMPP is complementary to these techniques because it does provide information about polysaccharide levels, albeit semiquantitatively. As with FTIR, CoMPP is intended primarily as a high-throughout initial screening method that is useful for directing the application of subsequent low-throughput but quantitative analyses. Most CoMPP analyses are performed on the alcohol-insoluble residues (AIR) of different samples. The next steps typically involve weighing samples prior to extraction, microarray printing, probing and spot quantification. Each of these steps can be altered according to particular needs of the experiment. For example, a typical extraction procedure would involve sequentially extracting with CDTA (to release pectins), NaOH (to release non-cellulosic polysaccharides) and Cadoxen (diaminoethane with cadmium oxide, to solubilize cellulose). Supernatants from each step are stored until all the extractions are completed and then printed using a microarray robot

Dissection of Plant Cell Walls by High-Throughput Methods

■

53

on to a suitable slide or membrane surface. Since the supernatants contain complex mixtures of cell wall polymers it is important that the surface used is capable of reliably immobilizing chemically and structurally diverse glycans. Several surfaces can be employed, including polystyrene slides (MaxiSorp; NUNC; Willats et al. 2002), nitrocellulose-coated glass slides (FAST slides; Whatman; Wang et al. 2002) or nitrocellulose sheets. If polystyrene or nitrocellulose coated glass slides are used then antibody binding can be detected using either enzyme-based colorimetric detection or fluorescence detection systems. In the case of nitrocellulose sheets, enzyme-based colorimetric detection has the advantage that spot signals can be captured using a flat-bed scanner rather than a specialized slide scanner. This in turn means that, once printed, arrays can be processed by almost any laboratory. Nitrocellulose has been very widely used for the analysis of cell wall glycans and immobilization is generally reliable and stable. Furthermore, a larger printing area means there is scope for expanding arrays or changing array layouts. CoMPP is reliant on mAbs and CBMs with specificity for cell wall components. A wide range of suitable mAbs have been produced and are reviewed in Knox (1997) and elsewhere in this book (see Chapter 17). The mAbs currently available cover most classes of cell wall polymers: pectins (including homogalacturonan, arabinan, galactan and xylogalacturonan); non-cellulosic polysaccharides (including xylans, mannans, xyloglucans and mixed linkage glucans) and glycoproteins (including arabinogalactanproteins and hydroxyproline-rich glycoproteins). CBMs are the binding domains of carbohydrate degrading enzymes and can readily be produced as recombinant proteins with epitope tags (Blake et al. 2006; McCartney et al. 2006). CBMs with a range of specificities have been used for cell wall analysis and are especially valuable for the detection of cellulose, since anticellulose mAbs are not widely available. For both mAbs and CBMs there are considerable variations in the extent to which specificities have been characterized. mAbs that are raised using highly purified oligosaccharide targets usually show a high degree of specificity for that glycan structure. For example, mAbs LM6 and LM5 were produced using neo-glycoproteins of (1–5)-α-l-arabinan and (1–4)-β-d-galactan respectively and are highly specific for these epitopes. In contrast, mAb JIM13 binds to a glycan moiety of arabinogalactan proteins, but the epitope structure is not known. However, even when precise epitope structures are unknown, mAbs and CBMs may still be useful for CoMPP analysis because they may highlight differences per se in polysaccharide content between samples that suggest how subsequent analyses should be focused. CoMPP spot signals are quantified using microarray analysis software such as ImaGene (BioDiscovery, El Segundo, USA) which allows a range of different parameters to be simultaneously measured such as mean, median and mode signal values, as well as global and local background levels. The presentation of large microarray datasets can be problematic, and formats that are well established for transcriptomics data are routinely used for

54

■

Plant Polysaccharides

Arabinan (mAb LM6)

Galactan (mAb LM5)

Mannan (mAb 400-4)

HRGP (mAb LM1)

Figure 2.3 Examples of CoMPP cell wall polysaccharide arrays produced from Arabidopsis thaliana probed with a range of monoclonal antibodies.

CoMPP. Scatter plots of spot signals from two sample types, for example mutant versus wild type, are useful for visualizing the difference in the abundance of a particular polysaccharide. Heat maps where mean spot signals are related to colour intensity are valuable for providing a global overview of the relative levels of many different polysaccharides in a wide range of samples. CoMPP can also be used for comparative studies, such as mapping of the relative levels of cell wall components in different organs or tissues throughout a plant. For example, CoMPP was used to generate a comprehensive overview of cell wall composition in Arabidopsis and the moss Physcomitrella patens and examples of arrays from Arabidopsis are shown in Fig. 2.3. CoMPP can similarly be used for comparing mutant and wild type plants, assessing global changes in cell wall composition that arise from biotic or abiotic stresses, or analysis of cell walls across the plant kingdom. One disadvantage of CoMPP is that it is only semiquantitative and is therefore not intended as a tool for analysing absolute levels of cell wall components. There are several features of the procedure that preclude CoMPP from being quantitative. The extraction solvents and procedures used may not be equally effective across all sample types and epitopes may be destroyed, modified or masked during extraction. Immobilization of cell wall components is non-covalent and although polysaccharides adhere well to nitrocellulose, there may be differences between the immobilization efficiency of different polymers. Different mAb and CBM probes bind with differing avidities to their epitopes and this means that signals generated from different probes cannot be used to compare the levels of different polymers. A second microarray-based polymer profiling approach involves the construction of microarrays of defined cell wall components. The primary use of these arrays is for screening the specificities of mAbs and CBMs and for assaying the activities of enzymes that act on cell wall components. The

Dissection of Plant Cell Walls by High-Throughput Methods

■

55

arrays are typically about 20 mm × 20 mm and contain 50–100 different cell wall samples. Their relatively small size means that small quantities of probe or enzyme are required for screening, and this is one advantage of arrays over enzyme linked immunosorbent assays (ELISA). The utility of defined polymer arrays has been demonstrated by the screening of a population of new cell wall directed antibodies (Moller et al. 2007). Following immunization with a crude cell wall preparation and initial ELISA binding tests, the resulting complex mixture of antibodies was screened using arrays of cell wall polymers. Cluster analysis of signals from the newly produced antibodies with signals from previously characterized ones resulted in the identification of several new mAbs with specificities for pectic and glycoprotein epitopes. Defined polymer arrays can be used for screening enzyme activities since loss or creation of epitopes by enzymes can be detected using antibodies. For example, the in situ digestion of arrayed pectin polymers by pectate lyase was assessed by monitoring the reduction in binding of the anti-homogalacturonan mAbs JIM5 and JIM7 (Øbro et al. 2007). Currently the most serious limitation for developing defined polymer arrays for cell wall research is obtaining suitably pure samples. Unlike proteins, complex carbohydrates are extremely difficult to synthesize and almost all samples are obtained from cell wall extracts. This is not only time consuming and expensive but also usually results in only partially pure samples. An important future goal will be construct arrays of well-defined oligosaccharides. One attractive but ambitious goal would be to integrate structural analysis of oligosaccharides with the construction of microarrays. In this scenario, polysaccharides would be enzymatically degraded and the resulting oligosaccharides fractionated. One portion of each fraction would be subjected to biochemical analysis, for example monosaccharide linkage analysis or MALDI-TOF MS. Aliquots of the same fractions would be used to populate arrays which would then contain well-defined oligosaccharides and be valuable tools for mAb, CBM and enzyme characterization.

2.6

Additional high-throughput methods

In addition to the approaches discussed above, several general large-scale approaches have been applied to dissect cell wall associated processes. These include analyses of the plant cell wall proteome (Borner et al. 2003; Rose et al. 2004), transcriptional coordination of cell wall related genes (Brown et al. 2005; Persson et al. 2005; Usadel et al. 2005), and chemical genetics screens (Debolt et al. 2007; Yoneda et al. 2007). Approximately 10% of the cell wall mass is anticipated to be made up of extracellular proteins (Chivasa et al. 2002). Efforts have therefore been made to obtain proteomic profiles of the cell wall compartment from higher plants (Borner et al. 2003; Rose et al. 2004; Kwon et al. 2005; Bayer et al. 2006) down to unicellular algae (Frigeri et al. 2006). The typical approach includes extraction of cell wall associated proteins without disrupting the cell and then

56

■

Plant Polysaccharides

identification of peptides with the aid of two-dimensional gel electrophoresis and MS techniques. For example, Borner et al. (2003) used liquid chromatography–tandem MS of Arabidopsis callus cells to uncover glycosylphosphatidylinositol (GPI)-anchored proteins. This study identified several known cell wall proteins, such as COBRA homologues and arabinogalactan proteins, but also novel receptor-like proteins with putative GPI anchors. Interestingly, this approach was subsequently linked with a search algorithm for GPI-anchored proteins based on the proteomic results, and a genomic search for new GPI-anchored proteins was carried out resulting in 64 new candidate proteins (Borner et al. 2003). Several studies have utilized the notion that genes that are coordinately expressed may be involved in functionally related processes (Stuart et al. 2003; Brown et al. 2005; Persson et al. 2005; Usadel et al. 2005). Typically coexpression analyses are using several hundreds of normalized microarray datasets to assess transcriptionally coordinated genes. However, it has been estimated that only approximately 40 datasets should be sufficient to obtain similar confidence (Manfield et al. 2006). Generally, microarray signals for individual genes are plotted against each other to investigate the linear relationship between the genes (Persson et al. 2005; Wei et al. 2006). The level of coexpression is then assessed by linear regression using R-values or p-values to rank the genes to the query gene. Several web-based tools to assess coexpression between genes are currently available (Steinhauser et al. 2004; Zimmermann et al. 2004; Toufighi et al. 2005; Manfield et al. 2006; Obayashi et al. 2007). These tools may generate initial hypotheses for different processes and prioritize genes for biological testing. However, it should be noted that basic skills in statistics and programming are virtually essential for broader coexpression studies. Small molecules that perturb specific processes and/or result in phenotypes that are of interest have good potential for unravelling biological components and processes that may be masked in mutant analyses. At least two molecules that inhibit cellulose production, isoxaben and 2,6dichlorobenzonitrile (DCB), are commonly used to mimic cellulose deficiencies (Scheible et al. 2001; Bonetta et al. 2002; Debolt et al. 2007). Several screens for novel compounds that mimic certain phenotypic traits have also been undertaken. These screens are typically performed using several thousands of chemical compounds, and the targets of the drugs are hence not known. Recently two studies identified novel molecules that perturb the microtubule– cellulose relationship in such a screen (Debolt et al. 2007; Yoneda et al. 2007). These compounds may thus be valuable tools to explore the link between microtubules and the cellulose synthesis in a way that classical genetic studies do not allow. One potential drawback for chemical genetics screens is the rather expensive chemical library that must be acquired. In addition, to obtain information about the target for the drug one must again perform screens on mutagenized populations that are resistant to the drug. This type of analysis is most commonly done by map-based cloning strategies on

Dissection of Plant Cell Walls by High-Throughput Methods

■

57

EMS-mutagenized plants, which used to require several months to years to complete (Scheible et al. 2001), but currently require significantly less time thanks to recent advances in sequencing techniques.

2.7

Future perspectives

Despite the importance of different cell wall polymers for various industrial applications, basic knowledge regarding synthesis of the components and interactions among these is still largely lacking. It is therefore apparent that we need to revise the current approach in order to acquire useful information more rapidly. Most of the methods presented here were developed to measure cell wall changes on a sample-to-sample basis. However, the rapid increases in high-throughput platforms for these techniques, the number of cell wall mutants available, and the number of epitope-specific antibodies, suggest that we will witness a surge in cell wall related information in the near future. This information should be deposited into publicly available databases which scientists worldwide may explore to obtain crucial information for further analyses on particular mutants. Such efforts are currently being developed; for example, the cell wall genomics initiative at Purdue has made FTIR spectra for approximately 2000 different genotypes in both Arabidopsis and maize publicly available. These data may thus be used as a first step to determine what kind of downstream analyses to perform. Another important task is to try to coordinate the datasets that are being generated from the different platforms. This may be challenging, since different groups may generate data from different tissues and species. Joint efforts where groups run the different analyses on equivalent samples may therefore generate more useful data than the more or less random sampling which, unfortunately, is more common today. Data integration of these results may then facilitate hypotheses for specific processes and prioritize genes for biological testing. For example, Paul Dupree and co-workers have developed a tool termed LOPIT where the subcellular vicinities of proteins are determined using multiple protein profiles from sucrose gradients (Dunkley et al. 2004). These profiles may be combined with coexpression analysis to both assess subcellular localization and transcriptional coordination and may hence be used to predict protein–protein interactions. In addition, a tool named Map-O-Matic has been developed (available at genecat. mpg.de) that uses existing phenotypic information to predict candidate genes in forward genetics screens (Mutwil et al., 2008). The tool generates two datasets: one that contains genes that may correspond to a specific phenotype, identified for example through rough map-based cloning, and another that contains genes that, upon mutation, phenocopy the phenotype that is being mapped. Since genes that are functionally related tend to be coexpressed, the tool then performs a cross-wise coexpression analysis for

58

■

Plant Polysaccharides

the two pools and ranks the genes in the map-based region according to the coexpression results. Since this tool depends on available mutant data, it is clear that more information regarding phenotypic features, for example though cell wall deficiencies that are linked to particular genes, will increase the accuracy for this, and similar, tools.

References Albersheim, P., Nevins, D.J., English, P.D., Karr, A. (1967) A method for the analysis of sugars in plant cell wall polysaccharides by gas-liquid chromatography. Carbohydrate Research, 5, 340–345. Barton, C.J., Tailford, L.E., Welchman, H., et al. (2006) Enzymatic fingerprinting of Arabidopsis pectic polysaccharides using polysaccharide analysis by carbohydrate gel electrophoresis (PACE). Planta, 224, 163–174. Bauer, S., Vasu, P., Persson, S., Mort, A.J., Somerville C.R. (2006) Development and application of a suite of polysaccharide-degrading enzymes for analyzing plant cell walls. Proceedings of the National Academy of Sciences of the U S A, 103, 11417–11422. Bayer, E.M., Bottrill, A.R., Walshaw, J., et al. (2006) Arabidopsis cell wall proteome defined using multidimensional protein identification technology. Proteomics, 6, 301–311. Beaugrand, J., Chambat, G., Wong, V.W., et al. (2004) Impact and efficiency of GH10 and GH11 thermostable endoxylanases on wheat bran and alkali-extractable arabinoxylans. Carbohydrate Research, 339, 2529–2540. Benová-Kákosová, A., Digonnet, C., Goubet, F., et al. (2006) Galactoglucomannans increase cell population density and alter the protoxylem/metaxylem tracheary element ratio in xylogenic cultures of zinnia. Plant Physiology, 142, 696–709. Blake, A.W., McCartney, L., Flint, J.E., et al. (2006) Understanding the biological rationale for the diversity of cellulose-directed carbohydrate-binding modules in prokaryotic enzymes. Journal of Biological Chemistry, 281, 29321–29329. Blumenkrantz, N., Asboe-Hansen, G. (1973) New method for quantitative determination of uronic acids. Analytical Biochemistry, 54, 484–489. Bonetta, D.T., Facette, M., Raab, T.K., Somerville C.R. (2002) Genetic dissection of plant cell-wall biosynthesis. Biochemical Society Transactions, 30, 298–301. Borner, G.H., Lilley, K.S., Stevens, T.J., Dupree, P. (2003) Identification of glycosylphosphatidylinositol-anchored proteins in Arabidopsis. A proteomic and genomic analysis. Plant Physiology, 132, 568–577. Broberg, A., Thomsen, K.K., Duus, J.O. (2000) Application of nano-probe NMR for structure determination of low nanomole amounts of arabinoxylan oligosaccharides fractionated by analytical HPAEC-PAD Carbohydrate Research, 328, 375–382. Brown, D.M., Zeef, L.A., Ellis, J., Goodacre, R., Turner, S.R. (2005) Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics. Plant Cell, 17, 2281–2295. Cardoso, S.M., Silva, A.M., Coimbra, M.A. (2002) Structural characterisation of the olive pomace pectic polysaccharide arabinan side chains. Carbohydrate Research, 337, 917–924.

Dissection of Plant Cell Walls by High-Throughput Methods

■

59

Carpita, N., McCann, M. (2000) The plant cell wall. In: Biochemistry and Molecular Biology of Plants (eds. B. Buchanan, W. Gruissem, R. Jones). American Society of Plant Biologists, Rockville, MD. Chen, L., Carpita, N.C., Reiter, W.D., Wilson, R.H., Jeffries, C., McCann, M.C. (1998) A rapid method to screen for cell-wall mutants using discriminant analysis of Fourier transform infrared spectra. Plant Journal, 16, 385–392. Chikamatsu, G., Shirai, K., Kato, M., Kobayashi, T., Tsukagoshi, N. (1999) Structure and expression properties of the endo-beta-1,4-glucanase A gene from the filamentous fungus Aspergillus nidulans. FEMS Microbiology Letters, 175(2), 239–245 Chivasa, S., Ndimba, B.K., Simon, W.J., et al. (2002) Proteomic analysis of the Arabidopsis thaliana cell wall. Electrophoresis, 23, 1754–1765. Ciucanu, I., Luca, C. (1990) Avoidance of degradation during methylation of uronic acids. Carbohydrate Research, 206, 71–77. Daas, P.J.H., Arisz, P.W., Schols, H.A., De Ruiter, G.A., Voragen, A.G. (1998) Analysis of partially methyl-esterified galacturonic acid oligomers by high-performance anion-exchange chromatography and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Analytical Biochemistry, 257, 195–202. DeBolt, S., Gutierrez, R., Ehrhardt, D.W., et al. (2007) Morlin, an inhibitor of cortical microtubule dynamics and cellulose synthase movement. Proceedings of the National Academy of Sciences of the U S A, 104, 5854–5859. Dische, Z. (1962) General color reactions. In: Methods in Carbohydrate Chemistry (eds Whistler and Wolfson), pp. 478–481. Academic Press, New York. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F. (1956) Colorimatric method for determination of sugars and related substances. Analytical Biochemistry, 28, 350–356. Dunkley, T.P., Watson, R., Griffin, J.L., Dupree, P., Lilley, K.S. (2004) Localization of organelle proteins by isotope tagging (LOPIT). Molecular Cell Proteomics, 3, 1128–1134. Evangelista, R.A., Liu, M.S., Chen, F.T.A. (1995) Characterization of 9-aminopyrene1,4,6-trisulfonate derivatized sugars by capillary electrophoresis with laserinduced fluorescence detection. Analytical Chemistry, 67, 2239–2245. Fernández-Espinar, M.T., Ramón, D., Piñaga, F., Vallés, S. (1992) Xylanase production by Aspergillus nidulans. FEMS Microbiology Letters, 91, 91–96. Fernández-Espinar, M.T., Piñaga, F., Sanz, P., Ramón, D., Vallés, S. (1993) Purification and characterization of a neutral endoxylanase from Aspergillus nidulans. FEMS Microbiology Letters, 113, 223–228. Fernández-Espinar, M.T., Piñaga, F., de Graaff, L., Visser, J., Ramón, D., Vallés, S. (1994) Purification, characterisation and regulation of the synthesis of an Aspergillus nidulans acidic xylanase. Applied Microbiology and Biotechnology, 42, 555–562. Frigeri, L.G., Radabaugh, T.R., Haynes, P.A., Hildebrand, M. (2006) Identification of proteins from a cell wall fraction of the diatom Thalassiosira pseudonana: insights into silica structure formation. Molecular Cell Proteomics, 5, 182–193. Fry, S.C. (1988) The Growing Plant Cell: Chemical and Metabolic Analysis. Longman, London. Galagan, J.E., Calvo, S.E., Cuomo, C., et al. (2005) Sequencing of Aspergillus nidulans and comparative analysis with A. fumigatus and A. oryzae. Nature, 438, 1105–1115.

60

■

Plant Polysaccharides

Gielkens, M., González-Candelas, L., Sánchez-Torres, P., et al. (1999) The abfB gene encoding the major alpha-L-arabinofuranosidase of Aspergillus nidulans: nucleotide sequence, regulation and construction of a disrupted strain. Microbiology, 145, 735–751. Goubet, F., Jackson, P., Deery, M.J., Dupree, P. (2002) Polysaccharide analysis using carbohydrate gel electrophoresis: A method to study plant cell wall polysaccharides and polysaccharide hydrolases. Analytical Biochemistry, 300, 53–68. Goubet, F., Morriswood, B., Dupree, P. (2003) Analysis of methylated and unmethylated polygalacturonic acid structure by polysaccharide analysis using carbohydrate gel electrophoresis. Analytical Biochemistry, 321, 174–182. Goubet, F., Ström, A., Dupree, P., Williams, M.A. (2005) An investigation of pectin methylesterification patterns by two independent methods: capillary electrophoresis and polysaccharide analysis using carbohydrate gel electrophoresis. Carbohydrate Research, 340, 1193–1199. Goubet, F., Ström, A., Quéméner, B., et al. (2006) Resolution of the structural isomers of partially methylesterified oligogalacturonides by polysaccharide analysis using carbohydrate gel electrophoresis. Glycobiology, 16, 29–35. Guttman, A. (1996) High-resolution carbohydrate profiling by capillary gel electrophoresis. Nature, 380, 461–462. Harvey, D.J. (2006) Analysis of carbohydrates and glycoconjugates by matrix-assisted laser desorption/ionization mass spectrometry: An update covering the period 1999–2000. Mass Spectrometry Review, 25, 595–662. Ho, M.C., Whitehead, M.P., Cleveland, T.E., Dean, R.A. (1995) Sequence analysis of the Aspergillus nidulans pectate lyase pelA gene and evidence for binding of promoter regions to CREA, a regulator of carbon catabolite repression. Current Genetics, 27, 142–149. Immerzeel, P., Pauly, M. (2006) Profiling methods for the analysis of cell wall polysaccharides. New Zealand Journal of Forestry Research, 36, 145–156. Kabel, M.A., Heijnis, W.H., Bakx, E.J., Kuijpers, R., Voragen, A.G., Schols, H.A. (2006) Capillary electrophoresis fingerprinting, quantification and mass-identification of various 9-aminopyrene-1,4,6-trisulfonate-derivatized oligomers derived from plant polysaccharides. Journal of Chromatography A, 1137, 119–126. Kemsley, E.K., Belton, P.S., McCann, M.C., Ttofis, S., Wilson, R.H., Delgadillo, I. (1994) A rapid method for the authentication of vegetable matter using Fourier transform infrared spectroscopy. Food Control, 5, 241 243. Klavons, J.A., Bennet, R.D. (1986) Determination of methanol using alcohol oxidase and its application to methyl ester content of pectins. Journal of Agricultural and Food Chemistry, 34, 597–599. Knox, J.P. (1997) The use of antibodies to study the architecture and developmental regulation of plant cell walls. International Review of Cytology, 171, 79–120. Kokot, S., Czarnik-Matusewicz, B., Ozaki, Y. (2002) Two-dimensional correlation spectroscopy and principal component analysis studies of temperature-dependent IR spectra of cotton-cellulose. Biopolymers, 67, 456–469. Kolli, V.S.K., York, W.S., Orlando, R. (1998) Fast atom bombardment mass spectrometry of carbohydrates contaminated with inorganic salts using a crown ether. Journal of Mass Spectrometry, 33, 680–682. Kumar, S., Ramón, D. (1996) Purification and regulation of the synthesis of a βxylosidase from Aspergillus nidulans. FEMS Microbiology Letters, 135, 287–293.

Dissection of Plant Cell Walls by High-Throughput Methods

■

61

Kwon, H.K., Yokoyama, R., Nishitani, K. (2005) A proteomic approach to apoplastic proteins involved in cell wall regeneration in protoplasts of Arabidopsis suspension-cultured cells. Plant Cell Physiology, 46, 843–857. Lee, Y.C. (1996) Carbohydrate analyses with high-performance anion-exchange chromatography. Journal of Chromatography A, 720, 137–149. Lerouxel, O., Choo, T.S., Séveno, M., et al. (2002) Rapid structural phenotyping of plant cell wall mutants by enzymatic oligosaccharide fingerprinting. Plant Physiology, 130, 1754–1763. Lionetti, V., Raiola, A., Camardella, L., et al. (2007) Overexpression of pectin methylesterase inhibitors in Arabidopsis restricts fungal infection by Botrytis cinerea. Plant Physiology, 143, 1871–1880. Lockington, R.A., Rodbourn, L., Barnett, S., Carter, C.J., Kelly, J.M. (2002) Regulation by carbon and nitrogen sources of a family of cellulases in Aspergillus nidulans. Fungal Genetics and Biology, 37, 190–196. Lomax, J.A., Boon, J.J., Hoffmann, R.A. (1991) Characterization of polysaccharides by in-source pyrolysis positive- and negative-ion direct chemical ionisation-mass spectrometry. Carbohydrate Research, 221, 219–233. MacDougall, A.J., Brett, G.M., Morris, V.J., Rigby, N.M., Ridout, M.J., Ring, S.G. (1997) The effect of peptide–pectin interactions on the gelation behaviour of a plant cell wall pectin. Plant Physiology, 114, 353–362. Manfield, I.W., Jen, C.H., Pinney, J.W., et al. (2006) Arabidopsis Co-expression Tool (ACT): web server tools for microarray-based gene expression analysis. Nucleic Acids Research, 34(web server issue), W504–509. Massart, D.L., Vanden Driessche, N., Van Dessel, A. (1990) Databases and spreadsheets. In: Data Handling in Science and Technology, Vol. 5: PCs for chemists (ed. J. Zupan). Elsevier, Amsterdam. Matsunaga, T., Ishii, T., Matsumoto, S., et al. (2004) Occurrence of the primary cell wall polysaccharide rhamnogalacturonan II in pteridophytes, lycophytes, and bryophytes. Implications for the evolution of vascular plants. Plant Physiology, 134, 339–351. McCann, M.C., Hammouri, M., Wilson, R., Belton, P., Roberts, K. (1992) Fourier transform infrared microspectroscopy is a new way to look at plant cell walls. Plant Physiology, 100, 1940–1947. McCann, M.C., Defernez, M., Urbanowicz, B.R., et al. (2007) Neural network analyses of infrared spectra for classifying cell wall architectures. Plant Physiology, 143, 1314–1326. McCartney, L., Blake, A.W., Flint, J., et al. (2006) Differential recognition of plant cell walls by microbial xylan-specific carbohydrate-binding modules. Proceedings of the National Academy of Sciences of the U S A, 103, 4765–4770. McCleary, B.V., Codd, R. (1991) Measurement of (1→3),(1→4)-beta-d-glucan in barley and oats – a streamlined enzymatic procedure. Journal of the Science of Food and Agriculture, 55, 303–312. McDougall, G.J., Fry, S.C. (1991) Purification and analysis of growth-regulating xyloglucan-derived oligosaccharides by high-pressure liquid chromatography. Carbohydrate Research, 219, 123–132. Mohr, M.D., et al. (1995) Matrix-assisted laser desorption/ionization mass spectrometry: improved matrix for oligosaccharides. Rapid Communications in Mass Spectrometry, 9, 809–814.

62

■

Plant Polysaccharides

Moller, I., Sørensen, I., Bernal, A.J., et al. (2007) High-throughput mapping of cellwall polymers within and between plants using novel microarrays. Plant Journal, 50,1118–1128. Moller, I., Marcus, S.E., Haeger, A., et al. (2007) High-throughput screening of monoclonal antibodies against plant cell wall glycans by hierarchical clustering of their carbohydrate microarray binding profiles. Glycoconjugate Journal, 25, 37–48. Mort, A.J., Pierce, M.L. (1994) In: Carbohydrate Analysis: High Performance Liquid Chromatography and Capillary Electrophoresis (ed. Z. El Rassi), pp. 3–37. Elsevier, Amsterdam. Mouille, G., Robin, S., Lecomte, M., Pagant, S., Höfte, H. (2003) Classification and identification of Arabidopsis cell wall mutants using Fourier-transform infrared (FT-IR) microspectroscopy. Plant Journal, 35, 393–404. Mouille, G., Witucka-Wall, H., Bruyant, M.P., et al. (2006) Quantitative trait loci analysis of primary cell wall composition in arabidopsis. Plant Physiology, 141, 1035–1044. Mutwil, M., Obro, J., Willats, W.G., Persson, S. (2008) GeneCAT – novel webtools that combine BLAST and co-expression analyses. Nucleic Acids Research, 36(web Server issue), W320–326. Obayashi, T., Kinoshita, K., Nakai, K., et al. (2007) ATTED-II: a database of coexpressed genes and cis elements for identifying co-regulated gene groups in Arabidopsis. Nucleic Acids Research, 35(database issue), D863–869. Obel, N., Erben, V., Pauly, M. (2006) Functional wall glycomics through oligosaccharide mass profiling. In: The Science and Lore of the Plant Cell Wall: Biosynthesis, Structure and Function (ed. T. Hayashi), pp. 258–266. Brown Walker Press, Boca Raton, FL. Øbro, J., Harholt, J., Scheller, H.V., Orfila, C. (2004) Rhamnogalacturonan I in Solanum tuberosum tubers contains complex arabinogalactan structures. Phytochemistry, 65, 1429–1438. Øbro, J., Sørensen, I., Moller, I., Skjøt, M., Dalgaard Mikkelsen, J., Willats, W.G.T. (2007) High-throughput microarray analysis of pectic polymers by enzymatic epitope deletion. Carbohydrate Polymers, 70, 77–81. Pauly, M., Albersheim, P., Darvill, A., York, W.S. (1999) Molecular domains of the cellulose/xyloglucan network in the cell walls of higher plants. Plant Journal, 20, 629–639. Pérez-González, J.A., De Graaff, L.H., Visser, J., Ramón, D. (1996) Molecular cloning and expression in Saccharomyces cerevisiae of two Aspergillus nidulans xylanase genes. Applied Environmental Microbiology, 62, 2179–2182. Pérez-González, J.A., van Peij, N.N., Bezoen, A., MacCabe, A.P., Ramón, D., de Graaff, L.H. (1998) Molecular cloning and transcriptional regulation of the Aspergillus nidulans xlnD gene encoding a β-xylosidase. Applied Environmental Microbiology, 64, 1412–1419. Perrin, R.M., Jia, Z., Wagner, T.A., et al. (2003) Analysis of xyloglucan fucosylation in Arabidopsis. Plant Physiology, 132, 768–778. Persson, S., Wei, H., Milne, J., Page, G.P., Somerville, C.R. (2005) Identification of genes required for cellulose synthesis by regression analysis of public microarray data sets. Proceedings of the National Academy of Sciences of the U S A, 102, 8633–8638. Quemener, B., Bertrand, D., Marty, I., Causse, M., Lahaye, M. (2007) Fast data preprocessing for chromatographic fingerprints of tomato cell wall polysaccharides using chemometric methods. Journal of Chromatography A, 1141, 41–49.

Dissection of Plant Cell Walls by High-Throughput Methods

■

63

Rassi, Z.E., Mechref, Y.S. (1998) Analysis of carbohydrates. In: Capillary Electrophoresis, Theory and Practise (ed. P. Camilleri), pp. 273–362. CRC Press, Boca Raton, FL. Ramón, D., vd Veen, P., Visser, J. (1993) Arabinan degrading enzymes from Aspergillus nidulans: induction and purification. FEMS Microbiology Letters, 113, 15–22. Ray, B., Loutelier-Bourhis, C., Lange, C., Condamine, E., Driouich, A., Lerouge, P. (2004) Structural investigation of hemicellulosic polysaccharides from Argania spinosa: characterisation of a novel xyloglucan motif. Carbohydrate Research, 339, 201–208. Rose, J.K., Bashir, S., Giovannoni, J.J., Jahn, M.M., Saravanan, R.S. (2004) Tackling the plant proteome: practical approaches, hurdles and experimental tools. Plant Journal, 39, 715–733. Scheible, W.R., Eshed, R., Richmond, T., Delmer, D., Somerville, C. (2001) Modifications of cellulose synthase confer resistance to isoxaben and thiazolidinone herbicides in Arabidopsis Ixr1 mutants. Proceedings of the National Academy of Sciences of the U S A, 98, 10079–10084. Séné, C., McCann, M.C., Wilson, R.H., Grinter, R. (1994) Fourier-transform Raman and Fourier-transform infrared spectroscopy (an investigation of five higher plant cell walls and their components). Plant Physiology, 106, 1623–1631. Somerville, C., Bauer, S., Brininstool, G., et al. (2004) Toward a systems approach to understanding plant cell walls. Science, 306, 2206–2211. Steinhauser, D., Usadel, B., Luedemann, A., Thimm, O., Kopka, J. (2004) CSB.DB: a comprehensive systems-biology database. Bioinformatics, 12, 3647–3651. Strasser, G.R., Amado, R. (2002) Pectic substances from red beet (Beta vulgaris L. var. conditiva). Part II. Structural characterisation of rhamnogalacturonan II. Carbohydrate Polymers, 48, 263–269. Stroop, C.J., et al. (2002) Carbohydrate analysis of bacterial polysaccharides by highpH anion-exchange chromatography and online polarimetric determination of absolute configuration. Analytical Biochemistry, 15, 176–185. Stuart, J.M., Bush, C.A., Marple, R.L., LaCourse, W.R. (2003) A gene-coexpression network for global discovery of conserved genetic modules. Science, 302, 249–255. Toufighi, K., Brady, S.M., Austin, R., Ly, E., Provart, N.J. (2005) The Botany Array Resource: e-Northerns, Expression Angling, and promoter analyses. Plant Journal, 43, 153–163. Usadel, B., Nagel, A., Thimm, O., et al. (2005) Extension of the visualization tool MapMan to allow statistical analysis of arrays, display of corresponding genes, and comparison with known responses. Plant Physiology, 138,1195–1204. Vierhuis, E., York, W.S., Kolli, V.S., et al. (2001) Structural analyses of two arabinose containing oligosaccharides derived from olive fruit xyloglucan: XXSG and XLSG. Carbohydrate Research, 332, 285–297. Voragen, A.G.J., Schols, H.A., Pilnik, W. (1986) Determination of the degree of methylation and acetylation of pectins by h.p.l.c. Food Hydrocolloids, 1, 65–70. Vries, R.P., Visser, J. (2001) Aspergillus enzymes involved in degradation of plant cell wall polysaccharides. Microbiology and Molecular Biology Review, 65, 497–522. Wang, D., Liu, S., Trummer, B.J., Deng, C., Wang, A. (2002) Carbohydrate microarrays for the recognition of cross-reactive molecular markers of microbes and host cells. Nature Biotechnology, 20, 275–281. Wei, H., Persson, S., Mehta, T., et al. (2006) Transcriptional coordination of the metabolic network in Arabidopsis. Plant Physiology, 142, 762–774.

64

■

Plant Polysaccharides

Willats, W.G., Rasmussen, S.E., Kristensen, T., Mikkelsen, J.D., Knox, J.P. (2002) Sugar-coated microarrays: a novel slide surface for the high-throughput analysis of glycans. Proteomics, 2, 1666–1671. Williams, M.A.K., Buffet, G.M., Foster T.J. (2002) Analysis of partially methylesterified galacturonic acid oligomers by capillary electrophoresis. Analytical Biochemistry, 301, 117–122. Wilson, R.H., Smith, A.C., Kacuráková, M., Saunders, P.K., Wellner, N., Waldron, K.W. (2000) The mechanical properties and molecular dynamics of plant cell wall polysaccharides studied by Fourier-transform infrared spectroscopy. Plant Physiology, 124, 397–405. Yamagaki, T., Mitsuishi, Y., Nakanishi, H. (1998) Determination of structural isomers of xyloglucan octasaccharides using post-source decay fragment analysis in MALDI-TOF mass spectrometry. Tetrahedron Letters, 39, 4051–4054. Yaoi, K., Mitsuishi, Y. (2004) Purification, characterization, cDNA cloning, and expression of a xyloglucan endoglucanase from Geotrichum sp. M128. FEBS Lett, 560, 45–50. Yoneda, A., Higaki, T., Kutsuna, N., et al. (2007) Chemical genetic screening identifies a novel inhibitor of parallel alignment of cortical microtubules and cellulose microfibrils. Plant Cell Physiology, 48, 1393–1403. York, W.S., Darvill, A.G., McNeil, M., Stevenson, T.T., Albersheim, P. (1985) Isolation and characterization of plant cell walls and cell wall components. Methods in Enzymology, 118, 3–40. York, W.S., van Halbeek, H., Darvill, A.G., Albersheim, P. (1990) Structural analysis of xyloglucan oligosaccharides by H1 NMR-spectroscopy and fast-atom bombardment mass- spectrometry. Carbohydrate Research, 200, 9–31. Zandleven, J., Sørensen, S.O., Harholt, J., et al. (2007) Xylogalacturonan exists in cell walls from various tissues of Arabidopsis thaliana. Phytochemistry, 68, 1219–1226. Zhong, H.J., Williams MA.K., Goodall, D.M., Hansen ME. (1998) Capillary electrophoresis studies of pectins. Carbohydrate Research, 308, 1–8. Zhong, R.Q., Peña, M.J., Zhou, G.K., et al. (2005) Arabidopsis fragile fiber8, which encodes a putative glucuronyltransferase, is essential for normal secondary wall synthesis. Plant Cell, 17, 3390–3408. Zimmermann, P., Hirsch-Hoffmann, M., Hennig, L., Gruissem, W. (2004) GENEVESTIGATOR. Arabidopsis microarray database and analysis toolbox. Plant Physiology, 136, 2621–2632.

Annual Plant Reviews (2011) 41, 65–92 doi: 10.1002/9781444391015.ch3

Chapter 3

APPROACHES TO CHEMICAL SYNTHESIS OF PECTIC OLIGOSACCHARIDES Sergey A. Nepogodiev1, Robert A. Field1 and Iben Damager2 1

Department of Biological Chemistry, John Innes Centre, Norwich Research Park, Colney Lane, Norwich NR4 7UH, UK 2 Department of Plant Biology and Biotechnology, Faculty of Life Sciences, University of Copenhagen, Frederiksberg, Denmark Manuscript received April 2009

Abstract: Many aspects of pectin biosynthesis and the physical properties of pectic polysaccharides can be better understood with the aid of small, welldefined oligosaccharide fragments of these macromolecules. Synthetic chemists have contributed to the study of pectin by preparation of fragments representing all three major types of pectic polysaccharide: homogalacturonan (HG), rhamnogalacturonan-I (RG-I) and rhamnogalacturonan-II (RG-II). Such molecules have been synthesized by sequential coupling of building blocks, the so-called glycosyl donors (GD) and glycosyl acceptors (GA), which aimed at the formation of specific glycosidic linkages as they are present in the target oligosaccharides. Challenges in synthesis of pectic oligosaccharides are associated with often poor stereoselectivity of glycosylation reactions between GA and GD, in particular for the construction of 1,2-cis-glycosidic linkages, high degrees of branching of oligosaccharide chains of target molecules and the nature of many monosaccharide components of pectin, which are often acidic and sometimes rare branched-chain sugars. Preparation of carbohydrate building blocks, including de novo syntheses of unusual sugars, protecting group strategies for GA and GD, glycosylation methodologies and general strategies for oligosaccharide assembly are described with the focus on pectin fragments. Synthetic routes to fragments of each type of pectic polysaccharides are discussed in detail in separate sections and structures of all currently known synthetic pectin fragments are summarized. The unsolved problems and future prospects for improved access to synthetic pectin fragments are also discussed. Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

65

66

■

Plant Polysaccharides

Keywords: glycosylation methodologies; glycosylation strategies; homogalacturonan; pectic oligosaccharide synthesis; rhamnogalacturonan-I; rhamnogalacuronan-II

3.1 Introduction The metabolism of carbohydrates underpins some of the most fundamentally important processes in nature. The diverse structure and function of glycans is reflected in the vast array of enzymes involved in their synthesis (glycosyltransferases), modification (carbohydrate esterases and methyl transferases) and breakdown (glycoside hydrolases and polysaccharide lyases). However, our knowledge of plant cell wall structure and metabolism is still rather limited (Somerville et al. 2004). Even with the advent of genome sequencing, we are still confounded by the apparent wealth of genes for which no specific functional assignment can be made. Developments in informatics (Henrissat et al. 2001) have allowed us to dissect the 25 498 genes present in the genome of Arabidopsis thaliana (Walbot 2000) such that more than 420 genes have been assigned probable roles in pathways responsible for the synthesis and modification of cell wall polymers (Arabidopsis Genome Initiative 2000). The high degree of apparent redundancy evident from such analyses may well reflect subtle differences in enzyme substrate specificity. The need for new tools with which to dissect plant polysaccharide metabolism is clearly evident. At one level, such tools may come from molecular biology and genetics. However, in contrast to DNA, RNA and protein sequences, which are template-encoded, the structure of oligo- and polysaccharides is not directly genome-dependent in the same predictable fashion (Turnbull & Field 2007). For the foreseeable future, genomic information may therefore be less influential in glycobiology than elsewhere. Particularly in relation to plant cell wall research, the need for direct analysis of glycan structure and enzyme function is clear. Paying specific attention to pectic polysaccharides, this chapter addresses the opportunities and challenges that need to be addressed in the preparation of oligosaccharide probes that might find application in structural and biosynthetic studies.

3.2 Pectic polysaccharides: structures and availability of fragments from natural sources 3.2.1 Structures Pectin is composed of large domains of different polysaccharides (Fig. 3.1) that contribute in different ways to the biological and bulk physical properties of pectin (Carpita & McCann 2000; Ridley et al. 2001). Homogalacturonan (HG) – a major pectic polysaccharide – is a homopolymer containing as

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

67

Figure 3.1 Schematic representation of a model of the basic pectin structure (adapted from Scheller et al. 2007). Detailed structures of each domain of pectin are highlighted in Sections 3.5–3.7.

many as 200 galacturonic acid moieties with a distribution of methyl esters and acetyl groups. There are two kinds of structurally modified HG, namely xylogalacturonan (XGA) and rhamnogalacturonan-II (RG-II), which are minor components of pectin. XGA also has an HG backbone, but 25–75% of the galacturonic acid units are substituted with xylose. RG-II is a rather complex highly branched glycan. Attached to the short HG backbone of RG-II are four side chains (A, B, C and D), which in total consist of up to 29 monosaccharides. Rhamnogalacturonan-I (RG-I) is the second major component of pectic polysaccharides and, in contrast to the other three pectic polysaccharides described above, it has a backbone of alternating galacturonic acids and rhamnose units. This backbone is variously substituted with arabinan, galactan and arabinogalactan side chains. Detailed structures of HG, XGA, RG-I and RG-II are highlighted in Section 3.5. 3.2.2 Preparation of pectic oligosaccharides by cleavage of polysaccharides A number of studies have reported the preparation of pectic oligosaccharides through selective chemical or enzymatic degradation of pectins. The practicality of these approaches is limited mostly to linear backbone fragments: side chains are usually cleaved off before or during the degradation. 3.2.2.1 Homogalacturonan fragments Fragments of homogalacturonan can be produced by depolymerization of polygalacturonic acid (PGA) using various endo-polygalacturonases and

68

■

Plant Polysaccharides

pectate lyases, as well as acid hydrolysis and physical treatment (Spiro et al. 1993). Oligomers generated by either chemical or enzymatic degradation require extensive chromatographic purification. On a preparative scale, oligogalacturonic acids with degree of polymerization (DP) 2–7 can be separated efficiently using a combination of size-exclusion and ion-exchange chromatography (Fan et al. 2002), or weak ion-exchange HPLC on aminopropyl silica (Hotchkiss et al. 1991). Di- and tri-galacturonic acids are commercially available, and multi-milligram quantities of larger oligo- and polygalacturonic acids (DP 7–20) can be purified by high-performance anion-exchange chromatography (Hotchkiss et al. 2001; Guillaumie et al. 2006). 3.2.2.2 Rhamnogalacturonan-I fragments RG-I is the second major component of pectic polysaccharides. Fragments of RG-I backbone composed of repeating disaccharide [→4)-α-d-GalpA(1→2)-α-l-Rhap] can be released from commercial pectin by controlled acid hydrolysis, which allows the selective cleavage of rhamnopyranosyl linkages (Renard et al. 1998). Another method of chemical degradation of RG-I makes use of the ease of β-elimination of uronic acid esters. Heating fully methyl-esterified RG-I in mild base results in cleavage at galacturonide residues, releasing oligomers of 1–5 disaccharide repeat units terminated at the non-reducing end by 4-deoxy-β-l-threo-hex-4-enepyranosyluronic acid (Deng et al. 2006) (Fig. 3.2). Enzymatic hydrolysis is a powerful tool for analysis of RG-I structure, but it is also potentially useful for the generation of oligosaccharide fragments on a preparative scale. Several enzymes are known to be capable of breaking glycosidic linkages in the RG-I backbone including RG-hydrolases and RG-lyases (Mutter et al. 1998). The former catalyse hydrolysis of GalpA-(1→2)-Rhap linkages and the latter catalyse β-elimination (as in base-catalysed fragmentation – Fig. 3.2). However, the use of enzymes for producing RG-I backbone oligosaccharides is complicated by the fact that up to 80% of the Rhap residues in RG-I are substituted. Oligosaccharide side chains, which are attached at C-4 of Rhap residues and composed predominantly of β-d-Galp and α-l-Araf, need to be removed prior to the action of RG-hydrolase and to some extent RG-lyase. Some Galp residues remain in RG-I even after pretreatment with galactosidases; branched oligomers incorporating one or two β-d-Galp residues have been isolated (Schols et al. 1994). Although degradation techniques have been widely used for elucidation of RG-I structure, the oligosaccharide fragments that have been prepared in this way require extensive chromatographic purification and they are usually only obtained in analytical quantities. Limited availability has made comprehensive characterization of RG-I fragments difficult and their potential application as chemical or biochemical reagents is not generally practical.

Approaches to Chemical Synthesis of Pectic Oligosaccharides

Figure 3.2

■

69

Chemical fragmentation of RG-I.

3.3 Reported preparations of pectic oligosaccharides by chemical synthesis The limited repertoire of pectic oligosaccharide structures available from degradative methods (see above) has prompted chemical synthetic approaches to these compounds, albeit in a rather limited number of cases to date. The synthesis of fragments of HG, side chains of RG-II and RG-I backbone are summarized in Table 3.1. XGA-based glycans have not yet received attention from synthetic chemists. As is clear from Table 3.1, only a limited number of chemical syntheses of pectic oligosaccharides have been reported to date. Why? What are the challenges? How are they addressed? These points are illustrated below with representative examples of chemical syntheses of pectic oligosaccharides.

Table 3.1 Pectic saccharides that have been chemically synthesized Synthetic fragments of pectic polysaccharides Homogalacturonan fragments α-D-GalpA-(1→4)-D-GalpA Two monomethyl esterified isomers Protected mono- and dimethyl-esterified methyl α- and β-glycosides Protected dimethyl esterified allyl β-glycoside Protected dimethyl esterified allyl α-glycoside α-D-GalpA-(1→4)-α-D-GalpA-(1→4)-D-GalpA Three monomethyl esterified isomers Protected fully methyl esterified allyl β-glycoside α-D-GalpA-(1→4)-{(α-D-GalpA-(1→4)}4-D-GalpA Five partially methyl esterified compounds α-D-GalpA-(1→4)-{(α-D-GalpA-(1→4)}8-β-D-GalpA-OPr α-D-GalpA-(1→4)-{(α-D-GalpA-(1→4)}10-D-GalpA Rhamnogalacturonan-II side chain A fragments β-D-Apif-(1→2)-α-D-GalpA-OMe β-L-Rhap-(1→3′)-β-D-Apif-OMe β-L-Rhap-(1→3′)-β-D-Apif-(1→2)-α-D-GalpA-OMe α-L-Fucp-(1→4)-L-Rhap (free disaccharide and methyl α- and β-glycosides) β-D-GalpA-(1→3)-α-L-Rhap-OMe β-D-GalpA-(1→3)-[α-D-GalpA-1→2]-α-L-Rhap-OMe α-L-Fucp-(1→4)-[β-D-GalpA-(1→3)]-[α-D-GalpA-1→2]-α-L-Rhap-OMe Rhamnogalacturonan-II side chain B fragments Acef β-L-Acef-(1→3)-α-L-Rhap-OMe (partially protected) α-L-Rhap-(1→3)-α-L-Arap-(1→4)-[2-O-β-L-MeFucp-(1→2)]-β-D-Galp-O(CH2)3NH2 β-L-Araf-(1→3)-α-L-Rhap-(1→2)-[α-L-Rhap-(1→3)-]-α-L-Arap-(1→4)-[2-OMe-β-L-Fucp-(1→2)]-β-D-GalpO(CH2)3NH2 β-L-Araf-(1→3)-α-L-Rhap-(1→2)-[α-L-Rhap-(1→3)-]-α-L-Arap-O(CH2)3NH2 Rhamnogalacturonan-I fragments α-D-GalpA-(1→2)-α-L-Rhap-(1→4)-D-GalpA (dimethyl esterified and partially protected) α-L-Rhap-(1→4)-α-D-GalpA-(1→4)-β-D-GalpA-OAll (partially protected) α-L-Rhap-(1→4)-α-D-GalpA-(1→2)-α-L-Rhap-(1→4)-β-D-GalpA-OPr α-L-Rhap-(1→4)-α-D-GalpA-(1→2)-α-L-Rhap-(1→4)-α-D-GalpA-OMe (both with free and methyl esterified GalpA residues)

Reference

Magaud et al. (1998) Magaud et al. (2000) Kramer et al. (2000) Vogel et al. (1992) Clausen et al. (2001) Kramer et al. (2000) Clausen & Madsen (2003) Nakahara& Ogawa (1989) Nakahara & Ogawa (1990a) Buffet et al. (2004) Nepogodiev et al. (2008) Chauvin et al. (2004) Nepogodiev et al. (2010) Egelund et al. (2006) Chauvin et al. (2005) Chauvin et al. (2005) Chauvin et al. (2005)

Jones et al. (2005) Nepogodiev et al. (2007) Timmer et al. (2006) de Oliveira et al. (2008) (Rao & Boons 2007) (Rao & Boons 2007)

(Rao & Boons 2008)

Nolting et al. (2000) Nolting et al. (2001) Maruyama et al. (2000) Reiffarth & Reimer (2008)

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

71

3.4 Oligosaccharide synthesis – basic principles and key features 3.4.1 General points The lack of a glycan equivalent to site-directed mutagenesis with which to ‘dial up’ required sequences leaves chemical and enzymatic synthesis as the primary routes to obtaining specific glycan structures. Although enzymatic synthesis shows great promise, the rather modest range of enzymes currently available limits its impact. Mirroring Merrifield’s classic work on solid-phase peptide synthesis in the 1960s (Merrifield 1963), the recent development of an equivalent for oligosaccharide synthesis has proved incredibly powerful (Plante et al. 2001). However, this is a specialist art that has yet to find general application. The principle limitation of the approach is that, for example, one commercial alanine building block can be used for all alanine-containing peptides, but several different building blocks are required in order to account for the various regio- and stereo-isomeric linkages found for any given sugar residue in nature. Seeberger and colleagues have assessed the structural diversity found in the three major classes of mammalian carbohydrates, namely glycolipids, O- and N-linked glycans (Werz et al. 2007). Analysis of size, chain length and branching complexity revealed that the average mammalian oligosaccharide is composed of about 8 monosaccharide units, with just 11 different inter-sugar connections accounting for more than 75% of all mammalian glycosidic linkages. Thus, the number of glycan structural combinations found in mammalian glycospace is much smaller than expected. Although such an analysis has not been made for plant carbohydrates, it can be predicted that the number of unique inter-sugar connections will be much higher. Thus, the single structure of pectic oligosaccharide RG-II is composed of 12 distinct monosaccharides with 20 different inter-sugar connections (i.e. distinct disaccharide units). The enormity of the challenge facing those prepared to engage in the chemical synthesis of plant glycans is immediately clear.

3.4.2 Main considerations in chemical synthesis of complex oligosaccharides and polysaccharides The basic principles of oligosaccharide synthesis have changed little since Paulsen’s classic review of the subject (Paulsen 1982). A lucid account of the basic considerations in chemical synthesis of oligosaccharides was published by Kahn and Hindsgaul in 1994; again, the principles covered there are still relevant today. The progress in the field of oligosaccharide synthesis is regularly reviewed (Khan & O’Neill 1996; Boons 1998; Ernst et al. 2000; Fraser-Reid et al. 2001; Demchenko 2008). In essence, classical synthesis of oligosaccharides consists of building large structures in a selective manner

72

■

Plant Polysaccharides

Figure 3.3 General representation of the glycosylation reaction. LG denotes the anomeric leaving group, P and R protecting groups. The stereodirecting effect of 2-Oprotecting groups is presented in a simplified way as both 1,2-cis and 1,2-trans glycosides can be formed in each case. Note that common designations of glycosidic bonds as αand β- relate to absolute D- and L-configuration of sugars, e.g. α-D-galactopyranoside is a 1,2-cis- and α-L-arabinopyranoside is a 1,2-trans-glycoside.

through stepwise coupling of sugar partners to give glycosidic bonds (Fig. 3.3). These partners or building blocks are often called glycosyl donors and glycosyl acceptors. Anomeric centers (C-1) of glycosyl donors are usually functionalized with leaving groups, which help in generation of reactive intermediates. Glycosyl acceptors should possess at least one free (unprotected) OH group intended for coupling. The rest of the functional groups, such as OH and CO2H, that are present both in glycosyl donors and glycosyl acceptors need to be blocked with protecting groups in order to allow regioselectivity of glycosylation (only the correct alcohol reacts). The nature of a protecting group installed at O-2 of glycosyl donors is crucial for stereoselectivity of glycosylation (which stereoisomer, α- or β-, is formed). Glycosyl donors equipped with 2-O-acyl (acetyl, benzoyl, etc.) groups usually produce 1,2-trans-glycosides whereas construction of 1,2-cis-glycosides requires the presence of 2-O-benzyl or similar 2-O-alkyl groups. These groups are often referred as participating (2-O-acyl) and non-participating (2-O-alkyl) groups, which reflects their involvement in formation of reactive carbonium ion intermediates. Glycosylation has been a focus of research activity for carbohydrate synthetic chemists for a long time. This has led to the development of many glycosylation methods, although no single procedure suitable for synthesis of all kinds of oligosaccharides exists (in contrast to standardized peptide coupling methodology). These methods can be classified on the basis of the nature of a leaving group in glycosyl donors, and activating reagent. The procedures that are most often used for oligosaccharide assembly are listed in Table 3.2. All of these methods can employ glycosyl donors with

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

73

Table 3.2 Examples of glycosyl donors, leaving groups and activators commonly used in oligosaccharide synthesis Glycosyl donor

Leaving group (LG)

Typical activatora

Glycosyl bromide Glycosyl fluoride Alkyl (aryl) thioglycoside Glycosyl trichloroacetimidate n-Pentenyl glycoside

Br F SR (R = Et, Ph) OC(=NH)CCl3 O(CH2)3CH=CH2

AgOTf SnCl2-AgOTf N-iodosuccinimide-TfOH Me3SiOTf N-iodosuccinimide-TfOH

a

Tf denotes the trifluoromethanesulfonyl (CF3SO2) group.

Figure 3.4 Examples of glycosylation with glycosyl bromide (A) (Kramer et al. 2000), glycosyl trichloroacetimidate (B) (Nolting et al. 2001) and phenyl thioglycoside (C) (Magaud et al. 1998) leading to protected digalacturonic acid. Tf, trifluoromethanesulfonyl (CF3SO2) group.

2-O-participating or non-participating groups, thus being efficient in syntheses of both 1,2-cis- and 1,2-trans-glycosidic linkages. These methods have been applied to synthesis of a variety of pectic oligosaccharides (see above) that can be exemplified by synthesis of the GalpA-(1→4)-GalpA disaccharide fragment of HG (Fig. 3.4). The polyfunctionality of carbohydrates makes synthesis of large oligosaccharides rather complicated, and therefore the choice of optimal synthetic strategy is as important as choice of efficient glycosylation methodology. When there are more than two units in a target oligosaccharide, the question is in which order these unit interconnections should be made. How many different leaving groups and protecting groups need to be used to design

74

■

Plant Polysaccharides

building blocks that can be assembled into oligomers with a minimum number of synthetic steps? Several approaches have been developed which can be used in designing strategies. These can be divided into linear stepwise and convergent block synthesis. Certain leaving groups (thioalkyl, fluoride and pentenyl), which can withstand a wide range of protecting group manipulations and even glycosylation conditions are particularly useful in block synthesis. An additional opportunity to influence reactivity of both glycosyl donors and glycosyl acceptors is provided by the effects that are caused by protecting groups. Some of them can greatly deactivate both glycosyl donor and acceptor, which has many important implications. For example, electron-withdrawing (acyl) protecting groups can reduce reactivity of glycosyl donors 1000 times more than alkylated donors of the same type, so that the former can actually act as acceptors in reactions with the latter. Therefore protecting groups not only define regioselectivity and assist in controlling stereoselectivity of a single glycosylation act, but are also very important elements in designing optimal synthetic strategies. Total deblocking of protected functional groups in synthetic oligosaccharides is assumed to be standard and efficient, but caution should be exercised in some cases, in particularly for large structures. Further, purification of the final deprotected products is not considered to be problematic since all the efforts of synthetic chemists are focused on preparation of pure distinct structures. Unfortunately a significant part of these efforts can be taken up by the tedious purification of protected building blocks and intermediates. To overcome this shortcoming in the preparation of oligosaccharides, new approaches based on solid support synthesis have emerged (Plante et al. 2001).

3.5 Synthesis of homogalacturonan fragments HG is one of the major pectic polysaccharides of the plant primary cell wall. It is a homopolymer of α-(1→4) linked D-galacturonic acids, containing as many as 200 galacturonic acid units and measuring up to 100 nm long (Fig. 3.5). Methyl and acetyl ester are distributed throughout HG and contribute significantly to its gelling behaviour and the ripening process of fruit and vegetables. Efforts to synthesize fragments of HG, and selectively methylesterified versions thereof, have provided access to a variety of fragments ranging from di- to dodecasaccharides (Table 3.1).

Figure 3.5 Structure of homogalacturonan (HG). Methyl esters, which can be randomly positioned along the polysaccharide chain, are not shown.

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

75

Figure 3.6 (A) The ‘direct glycosylation with galacturonic acid derivatives’ strategy. (B) The ‘late stage oxidation of primary OH groups’ strategy. P, P1 and P2 denote protecting groups for hydroxyl and carboxyl functional groups; LG is a leaving group at anomeric centres of glycosyl donors.

From a synthetic point of view, two quite different strategies have been employed to obtain oligouronic acids, as shown in Fig. 3.6. In the ‘direct glycosylation with galacturonic acid derivatives’ strategy, the first step is to perform a glycosylation reaction in which suitably protected derivatives of esters of galacturonic acid are coupled. Subsequent deprotection provides the oligogalacturonide. In contrast, the first step in the ‘late stage oxidation of primary hydroxyl groups’ strategy is the glycosylation reaction in which suitable protected galactopyranosides are coupled. The synthesis proceeds with deprotection and selective oxidation of the primary hydroxyl groups (in the presence of secondary hydroxyl groups) to the carboxylic acid oxidation level of the required oligogalacturonide. These strategies are general for synthesis of all glycouronides and their principles have been reviewed (van den Bos et al. 2007). 3.5.1 Synthesis of oligogalacturonides by direct glycosylation with galacturonic acid derivatives It should be noted that, due to the electron-withdrawing carboxylic acid group, galacturonic acid derivatives used in glycan synthesis are substantially less reactive than the corresponding galactose derivatives, which impacts on the choice of protecting and leaving groups and promoter reagents used. The best results obtained in the synthesis of homogalacturonide disaccharides have been achieved by use of thiophenyl galacturonides as

76

■

Plant Polysaccharides

Figure 3.7 Synthesis of protected trigalacturonide.

glycosyl donors (Fig. 3.4C) (Magaud et al. 1997; Magaud et al. 1998), which is itself prepared by selective oxidation of galactose derivatives in two steps (Fig. 3.7). Syntheses of trigalacturonide applying the same thioglycosyl donors and a disaccharide acceptor proceeded with lower efficiency (Fig. 3.7) (Magaud et al. 1997; Kramer et al. 2000). Understandably, therefore, no syntheses of higher oligomers have been reported using this approach. 3.5.2 Synthesis of oligogalacturonides by a late stage oxidation approach The second strategy which has been applied to the construction of oligogalacturonides is based on post-glycosylation oxidation of the initially synthesized neutral α-(1→4)-galactooligosaccharides. Galactopyranose-based building blocks used for this synthesis possess a temporary protecting group at the primary hydroxyl group, which can be removed selectively and the free alcohol oxidized into a carboxyl group at a later stage of the oligosaccharide assembly. Two synthetic approaches have been employed with the post-glycosylation oxidation strategy: convergent block synthesis and reiterative monomer addition. Several important examples of oligogalacturonide synthesis based on the post-glycosylation oxidation strategy are outlined below. 3.5.2.1 The convergent block synthesis approach An impressive total synthesis of dodecagalacturonic acid employing construction of neutral galactododecaoside followed by oxidation of all 12 primary hydroxyl groups has been described (Fig. 3.8) (Nakahara & Ogawa 1990a; Nakahara & Ogawa 1990b). The common glycosylation methodology, based on glycosyl fluorides as glycosyl donors, was adopted throughout the whole synthesis. As well as imparting good overall reactivity, the use of non-participating benzyl protecting groups also helps to produce the required 1,2-cis-stereoselectivity during glycosylation. Only three types of

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

77

Figure 3.8 Synthesis of dodecagalacturonic acid 3.8.1 (Nakahara & Ogawa 1990a). Structural formulae are shown for starting monosaccharide building blocks (top), protected galacto-oligosaccharide and final product (bottom). The convergent block strategy is illustrated using cartoon representation.

78

■

Plant Polysaccharides

temporary protecting groups were employed for hydroxyl functionalities, including acetates for O-6 of galactose units. Temporary acetates allowed selective deprotection and oxidation of primary hydroxyl groups. Readily available starting monosaccharide building blocks were used in a highly convergent scheme to assemble a 12-mer galactooligosaccharide as shown in Fig. 3.8. Deacetylation of the 12-mer followed by two-step oxidation led to protected dodecagalacturonic acid, which was finally deprotected to give target 12-mer 3.8.1. 3.5.2.2 The reiterative synthesis strategy The 10-mer analogue of compound 3.8.1, decagalacturonic acid propyl glycoside 3.9.4, has been synthesized using a slightly different approach based on reiterative chain extension (Fig. 3.9) (Nakahara & Ogawa 1989). Three disaccharide building blocks were employed in this synthesis, with compounds 3.9.1 and 3.9.3 serving as glycosyl donors and compound 3.9.2 as the initial glycosyl acceptor. Donor 3.9.1 was attached to a growing oligosaccharide chain in a series of three consecutive glycosylation reactions, leading to an octasaccharide, which was finally capped with donor 3.9.3. The protecting group pattern in the resulting galactodecaoside has allowed transformation of the latter into decauronic acid 3.9.4 in essentially the same way as described above for the 12-mer synthesized by the convergent scheme (Fig. 3.8). Clearly the number of glycosylation steps required for the assembly of linear oligosaccharides by the reiterative scheme is higher than that in the convergent scheme, but the overall efficiencies of both approaches are comparable. 3.5.2.3 Synthesis of selectively methyl-esterified oligogalacturonic acids The examples of oligogalacturonide synthesis via late state oxidation described above do not allow preparation of partially methyl-esterified pectin fragments. Such fragments are of significant interest in studies on polygalacturonases, lyases and methyl esterases. In order to access oligogalacturonides with well-defined methyl-esterification patterns, synthetic route employing more elaborate building blocks have been developed (Clausen et al. 2001; Clausen & Madsen 2003). The methodology is based on repeated coupling of galactose mono- and disaccharide building blocks equipped with orthogonal protecting groups at O-6. By choosing a particular combination of the building blocks and the order of their coupling, it has been possible to synthesize a variety of di-, tri- and hexasaccharides differing in substitution at the primary hydroxyl groups. Removal of either set of O-6 protecting groups makes it possible to convert the corresponding galactopyranose into the corresponding methyl-esterified galacturonide residue. Subsequent deprotection and oxidation of the second set of selected galactopyranose residues leads to their transformation into free galacturonic acids. This strategy is exemplified by synthesis of the hexagalacturonic acid

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

79

Figure 3.9 Synthesis of decagalacturonic acid propyl glycoside (Nakahara & Ogawa 1989). Three disaccharide building blocks shown at the top are assembled into a 10-mer according to the reiterative strategy illustrated by cartoon presentation.

trimethyl esters 3.10.6 and 3.10.7 (Fig. 3.10). The key building blocks 3.10.1, 3.10.2 and 3.10.3 have incorporated either acetyl or p-methoxyphenyl (PMP) or both protecting groups at the primary hydroxyls. Three subsequent glycosylation reactions which were altered with selective deprotection steps furnished a key hexasaccharide which has 6-O-PMP-protected residues 2, 3 and 4, and free 6-OH groups in residues 1, 5 and 6. Functionalization of each set of these galactose residues using oxidation and esterification reactions

80

■

Plant Polysaccharides

Figure 3.10 Synthesis of partially methyl-esterified hexasaccharide (Clausen & Madsen 2003) as an example of the strategy used for preparation of methyl-esterified oligogalacturonic acids.

led to hexasaccharides 3.10.4 and 3.10.5. Removal of benzyl protecting groups using hydrogenolysis afforded a pair of methyl esterified hexagalacturonic acids 3.10.6 and 3.10.7. In total, eight different partially methylesterified oligogalacturonic acids have been synthesized by this approach (Clausen et al. 2001; Clausen & Madsen 2003) (Table 3.1).

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

81

Figure 3.11 Proposed structure of rhamnogalacturonan-II (RG-II) based on detailed knowledge of glycosyl sequences for each side chain. Variation in structure of side chains A and B, which may lack certain terminal residues, are not shown. It is established that chain B and D are attached to residues 5 and 6 of homogalacturonan backbone, whereas the exact branch points of side chains A and C are still unknown (Rodríguez-Carvajal et al. 2003).

3.6 Rhamnogalacturonan-II fragments RG-II can be considered as structurally modified HG, which in total consists of up to 29 monosaccharide residues (Fig. 3.11). Attached to RG-II HG backbone are four oligosaccharide side chains (so-called A, B C and D chains)

82

■

Plant Polysaccharides

Figure 3.12

Four possible cyclic forms for D-apiose.

(O’Neill et al. 2004; Scheller et al. 2007). The latter are derived from 12 different kinds of sugars, including some rare carbohydrates such as aceric acid (Ace) and 3-deoxy-d-lyxo-2-heptulosaric acid (Dha). The structures of RG-II side chains are well established and to date synthetic efforts have focused on fragments of these side chains, in particular chains A and B. Both of these oligosaccharide chains are attached to a polygalacturonan backbone through apiofuranose (Apif). Apiofuranose is a branched chain pentose that is common in the plant kingdom, occurring in polysaccharides as well as a variety of secondary metabolites. Its isolation from natural sources is tedious (Watson & Orenstein 1975) and, in addition, this branched chain sugar can exist in four different forms with only the β-d-erythro-form being naturally occurring (Fig. 3.12). Totally synthetic routes have been developed for preparation of apiofuranose in forms that are suitable for application in oligosaccharide synthesis (Hammerschmidt et al. 1995; Koóš et al. 2002). Thus 2,3-O-isopropylidene derivative 3.13.1 (Chauvin et al. 2004) has been used as glycosyl acceptor and thioglycoside 3.13.2 as donor in synthesis of RG-II disaccharide fragments associated with ‘reducing’ ends of both side chain A and B (Buffet et al. 2004) (Fig. 3.13). A similar strategy has been applied to the synthesis of the more complex trisaccharide fragment of chain A and B having apiofuranose as a middle unit (Fig. 3.14) (Nepogodiev et al. 2008). In this case, the apiofuranose derivative has a different protecting group pattern, which allowed its use first as a glycosyl donor and then, after selective removal of the chloroacetyl group, as a glycosyl acceptor. The tetrasaccharide fragment 3.15.6 of side chain A shown in Fig. 3.15, incorporating rhamnopyranose as a ‘reducing’ residue, has also been synthesized using readily available building blocks 3.15.1–3.15.4 and a late stage oxidation strategy (Chauvin et al. 2005). The latter has become possible as a result of application of regioselective TEMPO-assisted oxidation (Davis & Flitsch 1993), which has allowed convenient transformation of galactose to galacturonic acid residues in 3.15.5 without affecting 6-deoxy sugars present

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

83

Figure 3.13 Application of apiofuranose derivatives as glycosyl donor and acceptor in synthesis of RG-II fragments.

Figure 3.14 Synthesis of a common trisaccharide fragment of side chains A and B incorporating one galacturonic acid residue of the backbone HG.

in the molecule. Two trisaccharide fragments each carrying only one galacturonic acid residues attached to methyl α-l-fucopyranosyl-(1→4)-α-lrhamnopyranoside (Chauvin et al. 2005) as well as the disaccharide α-l-Fucp-(1→4)-α-l-Rhap itself (Egelund et al. 2006) have been also prepared (Table 3.1). The most unusual sugar present in RG-II is a branched chain furanose called aceric acid (Acef). It has not been found in any other plant polysaccharides or glycosides, although some analogous structures are known to be constituents of microbially derived antibiotics (Grisebach & Schmid 1972). In order to synthesize fragments of side chain B incorporating aceric acid, the latter should be prepared first. Several synthetic approaches have been developed to provide access to aceric acid or its derivatives, which can potentially serve as building blocks in synthesis of side chain B (Jones et al.

84

■

Plant Polysaccharides

Figure 3.15 Synthesis of side chain A tetrasaccharide fragment. The order in which three glycosyl residues have been attached to a central rhamnose unit is indicated by arrows.

2005). For example, l-xylose has been transformed into aceric acid in a multistep synthesis using two different routes (Fig. 3.16) with overall efficiency about 13% (Nepogodiev et al. 2007). Another reported approach to aceric acid has utilized d-arabinose as starting material (Timmer et al. 2006). In the latter case a six-step synthesis furnished the target saccharide in 31% overall yield. It is evident from these examples that synthesis of RG-II fragments is difficult not only on the level of larger oligomers but even at the monosaccharide level. Synthesis of larger fragments of side chain B incorporating aceric acid is associated with further synthetic challenges: construction of a 1,2-cis-glycofuranosidic linkage between aceric acid and the adjacent sugar residue, L-rhamnose, and glycosylation of a sterically crowded position at O-2 of aceric acid itself. Application of building blocks based directly on protected aceric acid seems to be problematic since formation of undesired 1,2-trans-furanosidic bond can be anticipated. To overcome this difficulty an indirect approach exploiting the high efficiency and stereoselectivity of the

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

85

Figure 3.16 Overview of two alternative routes to aceric acid synthesis showing key intermediate compounds (Jones et al. 2005; Nepogodiev et al. 2007).

Figure 3.17 Synthesis of an analogue of β-L-Acef-(1→3)-L-Rhap fragment of side chain B using C-2′ epimerization approach (de Oliveira et al. 2008b).

formation of 1,2-trans-furanosides has been proposed (de Oliveira et al. 2008b). In this approach C-2 epimers of functionalized aceric acid need to be used as glycosyl donors. This concept has been demonstrated by the preparation of a 5-hydroxy analogue 3.17.1 of aceric acid containing disaccharide β-l-Acef-(1→3)-l-Rhap (Fig. 3.17). Two oligosaccharides corresponding to the ‘outer ’ part of side chain B and incorporating terminal ‘non-reducing’ residues have been synthesized with the aim of creating an artificial antigen suitable for raising RG-II specific antibodies (Rao & Boons 2007). Synthesis of hexasaccharide 3.18.1 has been performed in a highly convergent and stereoselective manner by careful planning of coupling reactions between specially designed building blocks

86

■

Plant Polysaccharides

Figure 3.18 Side chain B fragment 3.18.1 and building blocks used for its synthesis (Rao & Boons 2007). Arrows indicate the order in which building blocks have been coupled using glycosylation reactions.

(Fig. 3.18). To assemble an oligosaccharide with such a dense architecture and a number of synthetically challenging glycosidic linkages, the reactivity of each of these building blocks had to be adjusted by introduction of a specific combination of protecting groups. Smaller versions of the same part of side chain B represented by tetrasaccharides composed of the I, II, III and IV residues (Rao & Boons 2007) or III, IY, Y and YI residues (Rao & Boons 2008) have been also synthesized (Table 3.1).

3.7 Rhamnogalacturonan-I fragments RG-I is a group of pectic polysaccharides which have a common backbone heteropolymer with repeating disaccharide unit α-d-GalpA-(1→2)-α-lRhap-(1→4). The diversity of this group is a result of the presence of heterogeneous side chain polysaccharides represented by linear and branched (1→5)-α-l-arabinofuranan, (1→4)-β-d-galactopyranan and arabinogalactan (Fig. 3.19). Chemical synthesis of RG-I backbone fragments has been performed using two approaches similar to those highlighted above for the synthesis of homogalacturonan fragments. The first approach employs galacturonic acid derivatives as constituents of glycosyl acceptors and donors (Reiffarth

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

87

Figure 3.19 Model of rhamnogalacturonan-I structure, adapted from Scheller et al. (2007). 20–80% of Rhap residues of RG-I are substituted at C-4 with oligosaccharide side chains composed of α-L-Araf and β-D-Galp residues. The GalpA residues of the RG-I backbone may be acetylated at C-2 and/or C-3.

& Reimer 2008) and the second adopts post-glycosylation oxidation of galactopyranose to galacturonic acid residues (Maruyama et al. 2000). According to the first approach, protected galacturonates have been applied directly as building blocks in the synthesis of tetrasaccharide RG-I backbone fragment 3.20.1. This oligosaccharide has galacturonate as the terminal ‘reducing’ residue and it has been assembled using a convergent scheme starting from a rhamnopyranosyl donor and a galacturonate acceptor (Fig. 3.20) (Reiffarth & Reimer 2008). The late stage oxidation approach takes advantage of the high efficiency with which neutral saccharides can be coupled compared to saccharides, which have uronic acids as their constituents. Thus in the synthesis of tetrasaccharide 3.21.1, which has been performed using the same convergent strategy as shown in Fig. 3.20, the disaccharide and tetrasaccharide intermediate building block were synthesized in high yields (Fig. 3.21). After removal of all protecting groups only two primary OH groups remain. These groups have been selectively oxidized forming carboxylic acids and furnishing RG-I tetrasaccharide synthesis (Maruyama et al. 2000).

88

■

Plant Polysaccharides

Figure 3.20 Application of galacturonic acid-based building blocks in convergent synthesis of RG-I tetrasaccharide fragments (Reiffarth & Reimer 2008).

Figure 3.21 Synthesis of RG-I fragment using the late stage oxidation approach (Maruyama et al. 2000).

3.8 Future perspective In the future, synthesis of pectin fragments will benefit from general advances in carbohydrate chemistry, which nowadays has methodologies for assembly of virtually any oligosaccharide. Having said that, one has to appreciate the fact that there is no routine technology for oligosaccharide synthesis that is comparable in its efficiency to those available in the fields of peptide and nucleotide chemistry. Synthesis of oligosaccharides with less common substitution patterns or incorporating unusual sugars, as is often the case in plant glycans, remains a specialist craft. Considerable time needs to be

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

89

dedicated to preparation of building blocks, which are specifically designed for each particular case. As is evident from the examples presented in Sections 3.5–3.6, plant glycan synthesis is dominated by extensive protecting group manipulations that are characteristic of classical carbohydrate chemistry. A step-change in methodology is needed in order for the chemical synthesis of plant glycans to move forward at pace. In recent years, the development of automated oligosaccharide synthesis has greatly enhanced productivity of research associated with mammalian and bacterial glycoconjugates (Seeberger 2008). Provided suitable building blocks become available, this technology is awaiting wide application in the chemical synthesis of plant oligosaccharides.

References Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature, 408, 796–815. Boons, G.J. (1998) Carbohydrate Chemistry. Blackie Academic, London. Buffet, M.A.J., Rich, J.R., McGavin, R.S., Reimer, K.B. (2004) Synthesis of a disaccharide fragment of rhamnogalacturonan II. Carbohydrate Research, 339, 2507–2513. Carpita, N., McCann, M. (2000) The cell wall. In: Biochemistry and Molecular Biology of Plants (eds B. Buchanan, W. Gruissem, and R.L. Jones), pp. 52–108. American Society of Plant Physiologists, Rockville, MD. Chauvin, A.L., Nepogodiev, S.A., Field, R.A. (2004) Synthesis of an apiose-containing disaccharide fragment of rhamnogalacturonan-II and some analogues. Carbohydrate Research, 339, 21–27. Chauvin, A.L., Nepogodiev, S.A., Field, R.A. (2005) Synthesis of a 2,3,4-triglycosylated rhamnoside fragment of rhamnogalacturonan-II side chain A using a late stage oxidation approach. Journal of Organic Chemistry, 70, 690–696. Clausen, M.H., Jorgensen, M.R., Thorsen, J., Madsen, R. (2001) A strategy for chemical synthesis of selectively methyl-esterified oligomers of galacturonic acid. Journal of the Chemical Society Perkin Transactions, 1, 543–551. Clausen, M.H., Madsen, R. (2003) Synthesis of hexasaccharide fragments of pectin. Chemistry – A European Journal, 9, 3821–3832. Davis, N.J., Flitsch, S.L. (1993) Selective oxidation of monosaccharide derivatives to uronic acids. Tetrahedron Letters, 34, 1181–1184. de Oliveira, M.T., Hughes, D.L., Nepogodiev, S.A., Field, R.A. (2008) Indirect approach to C-3 branched 1,2-cis-glycofuranosides: synthesis of aceric acid glycoside analogues. Carbohydrate Research, 343, 211–220. Demchenko, A.V. (2008) Handbook of Chemical Glycosylation: Advances in Stereoselectivity and Therapeutic Relevance. Wiley-VCH, Weinheim. Deng, C.H., O’Neill, M.A., York, W.S. (2006) Selective chemical depolymerization of rhamnogalacturonans. Carbohydrate Research, 341, 474–484. Egelund, J., Petersen, B.L., Motawia, M.S., et al. (2006) Arabidopsis thaliana RGXT1 and RGXT2 encode Golgi-localized (1,3)-alpha-D-xylosyltransferases involved in the synthesis of pectic rhamnogalacturonan-II. Plant Cell, 18, 2593–2607.

90

■

Plant Polysaccharides

Ernst, B., Hart, G.W., Sinaÿ, P. (2000) Carbohydrates in Chemistry and Biology, Vol. 1. Wiley-VCH, Weinheim. Fan, H.N., Liu, M.Z., Lee, Y. (2002) Large-scale preparation of α-D-(1–4)oligogalacturonic acids from pectic acid. Canadian Journal of Chemistry, 80, 900–903. Fraser-Reid, B., Tatsuta, K., Thiem, J. (2001) Glycoscience: Chemistry and Chemical Biology, Vol. 2. Springer-Verlag, Berlin. Grisebach, H., Schmid, R. (1972) Chemistry and biochemistry of branched-chain sugars. Angewandte Chemie International Edition, 11, 159–248. Guillaumie, F., Justesen, S.F.L., Mutenda, K.E., Roepstorff, P., Jensen, K.J., Thomas, O.R.T. (2006) Fractionation, solid-phase immobilization and chemical degradation of long pectin oligogalacturonides. Initial steps towards sequencing of oligosaccharides. Carbohydrate Research, 341, 118–129. Hammerschmidt, F., Oehler, E., Polsterer, J.-P., Zbiral, E., Balzarini, J., DeClercq, E. (1995) A convenient route to D-apio-β-D-furanosyl- and 2′-deoxyapio-β-Dfuranosyl nucleosides. Liebigs Annalen 551–558. Henrissat, B., Coutinho, P.M., Davies, G.J. (2001) A census of carbohydrateactive enzymes in the genome of Arabidopsis thaliana. Plant Molecular Biology, 47, 55–72. Hotchkiss, A.T., Hicks, K.B., Doner, L.W., Irwin, P.L. (1991) Isolation of oligogalacturonic acids in gram quantities by preparative HPLC. Carbohydrate Research, 215, 81–90. Hotchkiss, A.T., Lecrinier, S.L., Hicks, K.B. (2001) Isolation of oligogalacturonic acids up to DP 20 by preparative high-performance anion-exchange chromatography and pulsed amperometric detection. Carbohydrate Research, 334, 135–140. Jones, N.A., Nepogodiev, S.A., MacDonald, C.J., Hughes, D.L., Field, R.A. (2005) Synthesis of the branched-chain sugar aceric acid: A unique component of the pectic polysaccharide rhamnogalacturonan-II. Journal of Organic Chemistry, 70, 8556–8559. Khan, S.H., Hindsgaul, O. (1994) In: Molecular Glycobiology (eds. M. Fukuda, O. Hindsgaul), pp. 206–229. Oxford University Press, Oxford. Khan, S.H., O’Neill, R.A. (1996) Modern Methods in Carbohydrate Synthesis, Vol. 1. Harwood Academic, Amsterdam. Koóš, M., Micová, J., Steiner, B., Alföldi, J. (2002) An efficient and versatile synthesis of apiose and some C-1- aldehyde- and/or 2,3-O-protected derivatives. Tetrahedron Letters, 43, 5405–5406. Kramer, S., Nolting, B., Ott, A.J., Vogel, C. (2000) Synthesis of homogalacturonan fragments. Journal of Carbohydrate Chemistry, 19, 891–921. Magaud, D., Dolmazon, R., Anker, D., Doutheau, A., Dory, Y.L., Deslongchamps, P. (2000) Differential reactivity of α- and β-anomers of glycosyl accepters in glycosylations. A remote consequence of the endo-anomeric effect? Organic Letters, 2, 2275–2277. Magaud, D., Grandjean, C., Doutheau, A., et al. (1997) An efficient and highly stereoselective α(1–4) glycosylation between two D-galacturonic acid ester derivatives. Tetrahedron Letters, 38, 241–244. Magaud, D., Grandjean, C., Doutheau, A., et al. (1998) Synthesis of the two monomethyl esters of the disaccharide 4-O-α-D-galacturonosyl-D-galacturonic acid and of precursors for the preparation of higher oligomers methyl uronated in definite sequences. Carbohydrate Research, 314, 189–199.

Approaches to Chemical Synthesis of Pectic Oligosaccharides

■

91

Maruyama, M., Takeda, T., Shimizu, N., Hada, N., Yamada, H. (2000) Synthesis of a model compound related to an anti-ulcer pectic polysaccharide. Carbohydrate Research, 325, 83–92. Merrifield, R.B. (1963) Solid phase peptide synthesis. 1. Synthesis of a tetrapeptide. Journal of the American Chemical Society, 85, 2149–2154. Mutter, M., Renard, C.M.G.C., Beldman, G., Schols, H.A., Voragen, A.G.J. (1998) Mode of action of RG-hydrolase and RG-lyase toward rhamnogalacturonan oligomers. Characterization of degradation products using RG-rhamnohydrolase and RG-galacturonohydrolase. Carbohydrate Research, 311, 155–164. Nakahara, Y., Ogawa, T. (1989) A highly stereocontrolled synthesis of the propyl glycoside of a decagalacturonic acid, a model-compound for the endogenous phytoalexin elicitor-active oligogalacturonic acids. Carbohydrate Research, 194, 95–114. Nakahara, Y., Ogawa, T. (1990a) Stereoselective total synthesis of dodecagalacturonic acid, a phytoalexin elicitor of soybean. Carbohydrate Research, 205, 147–159. Nakahara, Y., Ogawa, T. (1990b) Synthesis of (1–4)-linked galacturonic acid trisaccharides, a proposed plant wound-hormone and a stereoisomer Carbohydrate Research, 200, 363–375. Nepogodiev, S.A., Fais, M., Hughes, D.L., Field, R.A. (2008) Synthesis of a common trisaccharide fragment of side chain A and B of pectic polysaccharide rhamnogalacturonan-II. In preparation. Nepogodiev, S.A., Jones, N.A., Field, R.A. (2007) Plant cell wall glycans: Chemical synthesis of the branched sugar aceric acid. In: Frontiers in Modern Carbohydrate Chemistry (ed. A.V. Demchenko), pp. 34–49. American Chemical Society, Washington, DC. Nolting, B., Boye, H., Vogel, C. (2000) Synthesis of rhamnogalacturonan I fragments. Journal of Carbohydrate Chemistry, 19, 923–938. Nolting, B., Boye, H., Vogel, C. (2001) Block synthesis with galacturonate trichloroacetimidates. Journal of Carbohydrate Chemistry, 20, 585–610. O’Neill, M.A., Ishii, T., Albersheim, P., Darvill, A.G. (2004) Rhamnogalacturonan II: Structure and function of a borate cross-linked cell wall pectic polysaccharide. Annual Review of Plant Biology, 55, 109–139. Paulsen, H. (1982) Advances in selective chemical syntheses of complex oligosaccharides. Angewandte Chemie International Edition, 21, 155–173. Plante, O.J., Palmacci, E.R., Seeberger, P.H. (2001) Automated solid-phase synthesis of oligosaccharides. Science, 291, 1523–1527. Rao, Y., Boons, G.J. (2007) Highly convergent chemical synthesis of conformational epitopes of rhamnogalacturonan II. Angewandte Chemie International Edition, 46, 6148–6151. Rao, Y., Buskas, T., Albert, A., O’Neill, M.A., Hahn, M.G., Boons, G.J. (2008) Synthesis and immunological properties of a tetrasaccharide portion of the B side chain of rhamnogalacturonan II (RG-II). Chembiochem, 9, 381–388. Reiffarth, D., Reimer, K.B. (2008) Synthesis of two repeat units corresponding to the backbone of pectic polysaccharide rhamnogalacturonan I. Carbohydrate Research, 343, 179–188. Renard, C.M.G.C., Lahaye, M., Mutter, M., Voragen, F.G.J., Thibault, J.F. (1998) Isolation and structural characterisation of rhamnogalacturonan oligomers generated by controlled acid hydrolysis of sugar-beet pulp. Carbohydrate Research, 305, 271–280.

92

■

Plant Polysaccharides

Ridley, B.L., O’Neill, M.A., Mohnen, D.A. (2001) Pectins: structure, biosynthesis, and oligogalacturonide-related signaling. Phytochemistry, 57, 929–967. Rodríguez-Carvajal, M.A., Hervé du Penhoat, C., Mazeau, K., Doco, T., Pérez, S. (2003) The three-dimensional structure of the mega-oligosaccharide rhamnogalacturonan II monomer: a combined molecular modeling and NMR investigation. Carbohydrate Research, 338, 651–671. Scheller, H.V., Jensen, J.K., Sorensen, S.O., Harholt, J., Geshi, N. (2007) Biosynthesis of pectin. Physiologia Plantarum, 129, 283–295. Schols, H.A., Voragen, A.G.J., Colquhoun, I.J. (1994) Isolation and characterization of rhamnogalacturonan oligomers, liberated during degradation of pectic hairy regions by rhamnogalacturonase. Carbohydrate Research, 256, 97–111. Seeberger, P.H. (2008) Automated oligosaccharide synthesis. Chemical Society Reviews, 37, 19–28. Somerville, C., Bauer, S., Brininstool, G., et al. (2004) Toward a systems approach to understanding plant cell walls. Science, 306, 2206–2211. Spiro, M.D., Kates, K.A., Koller, A.L., O’Neill, M.A., Albersheim, P., Darvill, A.G. (1993) Purification and characterization of biologically-active 1,4-linked α-Doligogalacturonides after partial digestion of polygalacturonic acid with endopolygalacturonase. Carbohydrate Research, 247, 9–20. Timmer, M.S.M., Stocker, B. Ch 4, L., Seeberger, P.H. (2006) De novo synthesis of aceric acid and an aceric acid building block. Journal of Organic Chemistry, 71, 8294–8297. Turnbull, J.E., Field, R.A. (2007) Emerging glycomics technologies. Nature Chemical Biology, 3, 74–77. van den Bos, L.J., Codee, J.D.C., Litjens, R.E.J.N., Dinkelaar, J., Overkleeft, H.S., van der Marel, G.A. (2007) Uronic acids in oligosaccharide synthesis. European Journal of Organic Chemistry 3963–3976. Vogel, C., Steffan, W., Ott, A.Y., Betaneli, V.I. (1992) D-Galacturonic acid derivatives as acceptors and donors in glycosylation reactions. Carbohydrate Research, 237, 115–129. Walbot, V. (2000) A green chapter in the book of life. Nature, 408, 794–795. Watson, R.R., Orenstein, N.S. (1975) Chemistry and biochemistry of apiose. Advances in Carbohydrate Chemistry and Biochemistry, 31, 135–184. Werz, D.B., Ranzinger, R., Herget, S., Adibekian, A., von der Lieth, C.W., Seeberger, P.H. (2007) Exploring the structural diversity of mammalian carbohydrates (‘Glycospace’) by statistical databank analysis. ACS Chemical Biology, 2, 685–691.

Annual Plant Reviews (2011) 41, 93–108 doi: 10.1002/9781444391015.ch4

Chapter 4

ANNOTATING CARBOHYDRATEACTIVE ENZYMES IN PLANT GENOMES: PRESENT CHALLENGES Pedro M. Coutinho and Bernard Henrissat Architecture et Fonction des Macromolécules Biologiques, UMR6098, CNRS, Universities of Aix-Marseille I and II, 13288 Marseille Cédex 9, France Manuscript received September 2008

Abstract: There are far more carbohydrate structures than protein folds and, not unexpectedly, sequence-based families of carbohydrate-active enzymes often group together enzymes of different specificity. While this fact illustrates the remarkable plasticity of proteins to evolve novel specificities from pre-existing folds, this is also the source of potential errors during the functional prediction of genes derived from whole genome sequencing efforts. This chapter reviews the problems and the advances associated with the sequence-based classification of plant carbohydrate-active enzymes as used in the CAZy database (www.cazy.org). Keywords: arabinosyltransferase; cell wall; fucosyltransferase; galactosyltransferase; glucuronyltransferase; glycosylphosphatidylinositol; hexuronyltransferase; lipopolysaccharide; N-acetylglucosaminyltransferase; sialyltransferase

4.1 Introduction Carbohydrates and glycoconjugates are essential components for the life of most cellular organisms (Varki 1999). Carbohydrates are essential in cellwall structure and protection and often play roles in energy and carbon storage and transfer. Glycoconjugates may occur as intermediates in the Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

93

94

■

Plant Polysaccharides

synthesis of many complex molecules or as final products in glycolipids, glycoproteins, proteoglycans, etc. These compounds present a complex structure arising from the multiplicity of possible glycosidic bonds, the large variety of natural and modified sugar monomers, and the unique capacity to generate branched structures (Laine 1994). Unlike protein or nucleotide synthesis, where sequence is directly governed by gene sequence, the biosynthesis of precise glycoconjugate structures arises indirectly from (1) the combined action of specific enzymes; (2) the availability of required precursors; and (3) their confinement of the reactions in specific cells, cellular locations, compartments or tissues (Koch 1996) which often result in heterogeneous mixtures. The sets of enzymes involved in the biosynthesis of carbohydrates and glycoconjugates, but also in their degradation and modification, are collectively designated as carbohydrate-active enzymes (hereafter designated as CAZymes) (Coutinho & Henrissat 1999). Because the differences between various carbohydrates often resides in subtle stereochemical changes, most CAZymes have evolved to become very specific in order to recognize their substrates and perform catalysis, higher catalytic activities being often associated to more restricted specificity (Kacser & Beeby 1984). Thus relatively minor modifications in active site residues may result in significant specificity changes, such as a new ability to act on new monomers or different glycosidic bonds, and these modifications will be stabilized and optimized if they provide a competitive advantage. This capacity is probably at the origin of the diversity of enzyme activities where evolutionary steps leading to the biosynthesis of a new polysaccharide may be followed by the adaptation of sets of endogenous or exogenous enzymes to digest or rearrange it. Increased complexity also arise from the fact the many polysaccharides exist in nature in diverse forms, such as gels or in amorphous and crystalline microfibrillar forms, or in different combinations, such as ‘pure’ or complex mixtures (Rees 1977). Depending on context, enzymes may evolve binding capabilities to target specific carbohydrate structures so that their catalytic action is performed on these or on other substrates present in their vicinity. Often the acquisition of such binding function occurs through gene fusion events resulting in the attachment of one or several carbohydrate-binding modules to the original catalytic domain (Boraston et al. 2004). Since the late 1950s, like all other enzymes, CAZymes have been systematically classified using the IUPAC/IUBMB enzyme classification (EC) scheme (http://www.chem.qmul.ac.uk/iubmb/enzyme/) (Webb 1992), a system that is based exclusively on ‘macroscopic’ observations of enzyme action where the structural and mechanistic features of the enzymes are not considered. With the advent of molecular biology in the 1980s, of structural biology in the 1990s and the emergence of genomics in the late 1990s, it became desirable to correlate the different EC classes with protein sequences. Many developments in bioinformatics have used such principles to extrapolate EC assignments from a few fully characterized enzymes, in order to

Annotating Carbohydrate-Active Enzymes in Plant Genomes

■

95

infer basic aspects of the metabolism of organisms with a completely sequenced genome. In the field of oligosaccharide, polysaccharide, and, to a limited extent, glycoconjugate metabolism, the literature on biochemical characterization was relatively abundant prior to the 2000s. This characterization resulted from the independent fundamental efforts of an emerging glycobiology community, and also included an accumulation of biotechnology-fuelled characterizations that began in the 1960s and were often associated with medium- to large-scale modification and degradation of oligo- and polysaccharides (Ward & Moo-Young 1989). Rationalization of enzyme knowledge led to detailed independent classifications of enzymes. For CAZymes, some early schemes were based on aspects arising from the observation of enzyme properties with more detail than that provided by the evolving EC classification, such as exo- versus endoactivity, pH range of action, capacity to bind specific saccharide forms, etc. Such empirical and often contradictory information was typically organized in encyclopedic form, accumulating scattered information from the literature. When sufficient protein sequence information became available, comparisons of sequences led to the definition of sequence-based families, the advantage being that new sequences could be classified even if their biochemical characterization was minimal or absent. For CAZymes, it became clear very early that the sequence-based classification did not match the specificity of the ‘traditional’ EC classification. This may be exemplified by the initial classification of cellulases into several families (Henrissat et al. 1989), where some of the enzymes included have activities on other substrates related to cellulose, or by the observation of sequence similarities among starch-related enzymes with different EC numbers (Svensson 1988; MacGregor & Svensson 1989). Such observations led to the establishment of a ‘modern’ sequence-based family classification of glycoside hydrolases (GHs), where all aspects limiting substrate specificity were removed (Henrissat 1991). Further developments of this classification (Henrissat & Bairoch 1993; Henrissat & Bairoch 1996) allowed the validation of a number of the basic conclusions: (1) catalytic residues and reaction mechanism are shared by family members; (2) the structural fold is conserved within a family; (3) the various substrates of the enzymes within a family have topographical similarities. The same principles of sequence-based family classifications have been applied to the numerous glycosyltransferases (GTs) (Campbell et al. 1997; Coutinho et al. 2003) and also to the less abundant polysaccharide lyases (PLs) and carbohydrate esterases (CEs). As many of these enzymes exhibit a modular structure bearing accessory domains that facilitate substrate targeting and therefore improve catalytic efficiency, a complementary classification of carbohydrate-binding modules (CBMs; for a review see Boraston et al. 2004) has been created. These classifications are available in the

96

■

Plant Polysaccharides

Carbohydrate-Active Enzymes database (CAZy; www.cazy.org), online since September 1998 (Cantarel et al. 2009; Coutinho & Henrissat 1999). Such classifications allow a unified understanding of these enzymes, where fold and mechanism, rather than specificity, govern both fundamental and applied knowledge (Lairson & Withers 2004). The CAZy family classification system covers all taxonomic groups, and provides the ground for common nomenclature of CAZymes across different glycobiologists (Berteau & Stenutz 2004) who generally specialize only in some specific groups of organisms. Day-today inspection of new enzyme characterizations reported in the literature has regularly led to the definition of new enzyme families, and continues to do so. Significantly, the CAZy families, originally described following hydrophobic cluster analysis in the 1990s from the very limited number of sequences available (Henrissat 1991; Henrissat & Bairoch 1993; Henrissat & Bairoch 1996; Campbell et al. 1997) later complemented by more ‘modern’ sequence similarity approaches, are all surviving the challenge of time in spite of a 100fold increase in number at the time of writing. Naturally some structural and mechanistic relationships between families have been identified, such as the definitions of GH clans (Henrissat & Bairoch 1996) and GT folds, (Coutinho et al. 2003), but the practice of strictly controlling the extension of families on the basis of the conclusions mentioned earlier has been fruitful.

4.2 CAZy: what’s behind the name? The database supporting the CAZy operations (CAZyDB) is presently structured around two core components: (1) the ‘CAZy’ component associated with the management of sequence and structure accessions in public databases associated with a single protein, and its biochemical characterization; (2) the modular organization (ModO) component describing the different segments found in a sequence and its relationships to CAZy family descriptions (Coutinho & Henrissat 1999). Additional database components incorporate external information to manage information on organisms, such as the identification of species whose genome has been fully sequenced or their positioning in the tree of life according to the NCBI taxonomy (Wheeler et al. 2000). All these complementary components are necessary for the regular updates of the public CAZy site. The present structure of CAZy has several advantages: 1 For a given organism, the multiple accession numbers of a given protein are combined into a single entry to remove redundancy. Information from different strains is not combined, in order to facilitate genome comparisons. 2 The database accommodates partial sequences (e.g. those from PDB structures) with the corresponding full-length proteins, without losing track of the families that are described.

Annotating Carbohydrate-Active Enzymes in Plant Genomes

■

97

3 Since 2006 the database organizes families according to the main taxonomical subgroups: archaea, bacteria, eurkaryota and viruses. 4 It provides a summary of the content of CAZymes in each genome by family class. Maintenance of the CAZyDB involves a number of regular and systematic analyses to include new publicly released sequences and three-dimensional structures. This makes it possible to either update or complete information on proteins already classified in CAZy, or to create new entries. An internal bibliographical layer allows the maintenance of referenced biochemical characterizations. Our present philosophy is that each CAZy entry should correspond to one sequence per gene per species strain. At present no effort is made to address the issues of alternative splicing issues of a given protein, and as far as possible the alternatively spliced variants are lumped together in a single entry.

4.3

Plant CAZymes: the quest for ‘function’

Plant enzymes have always been included in CAZy and the underlying classifications. Plant biology entered the genomic era in 2000 with the landmark completion of the genome of the model plant Arabidopsis thaliana (Arabidopsis Genome Initiative 2000). The availability of this genome led to a gradual shift from independent ‘classical’ small-scale studies yielding protein and CAZyme characterization on a number of plants to a more systematic comparison of multiple organisms to Arabidopsis. In silico identification and interpretation of the A. thaliana set(s) of CAZymes in light of the available knowledge provided a list of target proteins (Henrissat et al. 2001) for characterization by the Arabidopsis community. Subsequently the completion of the rice genome in 2002 (Yu et al. 2002) followed by that of the genome of the first tree, poplar, in 2006 (Tuskan et al. 2006), enlarged the set of model plant genomes that will constitute a reference for many years to come. The availability of our first list of Arabidopsis CAZymes in 2001 (Henrissat et al. 2001) helped researchers involved in plant polysaccharide research realize the extent of the formidable characterization effort to be undertaken in the years to come. As CAZymes are universal and our analysis is not limited taxonomically, family definition and initial ‘foundation’ often relies on a very limited set of biochemically characterized examples. In the case of plants in general and of Arabidopsis as a first example, it was not uncommon for many of the identified sequences to be found in families that initially did not have a single characterized plant protein. Although such a situation might have been expected, its extent came as a surprise for most. In order to tackle the necessary characterization efficiently, plant biologists, and especially those studying the cell wall, were, in a way, forced by

98

■

Plant Polysaccharides

circumstances to adopt CAZy principles as a common language in order to directly benefit from the progress made by other glycobiological subcommunities (for instance fungi vs mammalia vs insect vs bacteria). What changed from the earlier glimpse of 2001 to the situation at the beginning of 2008 in the CAZyme field for Arabidopsis and plants in general? During this fast-moving period, two major events took place: (1) Arabidopsis was rapidly adopted by many plant biologists as the major model organism in undertaking the massive characterization efforts required to assign a function to the thousands of genes of the genome and (2) the arrival of other genomes massively lengthened the lists of uncharacterized genes. Recent reviews highlight much of the characterization efforts already accomplished on plant glycoside hydrolases (GHs), mostly in A. thaliana (Minic et al. 2007). Similarly, accounts describing several interrelated plant glycosyltransferases (GT) families have also appeared in the literature (Scheible & Pauly 2004; Farrokhi et al. 2006), but it is clear that GTs are more difficult to characterize experimentally than GHs. Significantly, new approaches like large-scale analysis resulting from combining bioinformatics analysis with post-genomic data can provide clues for focusing experimental efforts (Geisler-Lee et al. 2006; Mitchell et al. 2007). To illustrate the extend of the required characterization efforts, an account of the glycosyltransferase (GT) families content in for each of the three available genomes Arabidopsis thaliana, Oryza sativa (japonica cultivar), and Populus trichocarpa (Arabidopsis Genome Initiative 2000; Yu et al. 2002; Tuskan et al. 2006) is given in Table 4.1. This table also attempts to summarize the current knowledge of plant GTs and provides means for genomic comparison. Higher plants have large genomes, resulting from several rounds of complete genome duplication. These genomes encode more GTs than other organisms sequenced to date, with approximately 450 in Arabidopsis, over 560 in rice and more than 800 in poplar. By comparison, out of a similar total number of genes, the human genome encodes approximately 230 GTs, i.e. only half as many as Arabidopsis. The sheer numbers of CAZymes present in different genomes may be used for the identification and analysis of features that are common or distinct among genomes. Changes in plant CAZyme research since 2001 have been marked by: 1 A leap forward in the number of characterized enzymes and families. Two typical approaches arose in parallel in this time: (1) a large scale analysis of members of a family, or of a group of isoforms; (2) the characterization of representative sequences in model or in closely related organisms. 2 A stronger awareness that differences in genome content, i.e. relative family size, might reflect the relative diversity or complexity of the inherent biological processes (Yokoyama & Nishitani 2004), and therefore the biology of the compared species. For instance, differences suggesting a more pronounced pectin metabolism in ‘dicot’ Arabidopsis versus

Annotating Carbohydrate-Active Enzymes in Plant Genomes

■

99

Table 4.1 Relative abundance of glycosyltransferases (GTs) in the genomes of Arabidopsis thaliana (Arabidopsis Genome Initiative 2000), Oryza sativa (Yu et al.2002), and Populus trichocarpa (Tuskan et al.2006) subdivided into major metabolic classes following analysis with CAZy (http://www.cazy.org). The known plant GT activities present in these families, the identified or inferred role in the metabolism, and source of indirect evidence are indicated A. tha

O. sat

P. tri a

Cell wall 121 GT1b

202

330

Family

GT2

42

47

76

GT47

39

35

77

GT8b

42

39

61

GT31b

33

40

52

GT77

19

16

17

GT48

12

11

19

GT61b

7

25

10

GT37

10

18

8

GT34

8

6

8

GT43b

4

10

7

Energy storage and transfer 42 39 61 GT8b GT4b

24

25

41

GT20

11

11

12

GT5 GT35

6 2

11 2

12 6

Metabolism CW? {Lignin?}; Otherb CW {Cellulose; (Gluco)Mannan; β-1,3-1,4Glucan; Other?} CW {Xyloglucan; Arabinan; Pectin; Arabinoxylan?; Other?} CW {Pectin; Other?}; Otherb CW {Unclear}; Otherb CW {Xylan; Pectin; Other?} CW {Callose; Other?} CW {Arabinoxylan?}; Otherb CW {Xyloglucan; Other?} CW {Xyloglucan; Galactomannan; Other?} CW {Arabinoxylan?}; Otherb Energy {Sucrose}; Otherb Energy {Sucrose}; Otherb Energy {Trehalose} Energy {Starch} Energy {Starch}

Known activities in class

Indirect evidence

Processive β-polysaccharide synthases

AraT; GalT; GlcAT

Bioinformatics/ Post-genomics

AraT; XylT Callose synthase Bioinformatics/ Post-genomics FucT α-1,6-XylT/GalT

Bioinformatics/ Post-genomics

Galactinose synthase

Sucrose synthase (Susy)

Trehalose-P synthase Starch synthase Starch phosphorylase

100

■

Plant Polysaccharides

Table 4.1 Continued A. tha

O. sat

P. tri a

Glycolipids 24 GT4b

25

Family

Metabolism

Known activities in class

41

Glycolipids; Otherb

Sulfoquinovosyltransferase; Digalactosyldiacylglycerol synthase Monogalactosyldiacylglycerol synthase (Mgd) Ceramide synthase

GT28

4

4

6

Glycolipids

GT21

1

1

2

Glyco(sphingo) lipids

N-,O-Glycosylation & GPI anchor 33 40 52 N-glycans; GT31b Otherb b GT61 7 25 10 N-glycans; Otherb b GT43 4 10 7 N-glycans: Otherb GT17 6 4 5 N-glycans GT10 3 3 4 N-glycans GT66 2 2 6 N-glycans GT57

2

2

4

N-glycans

GT13 GT24

1 1

1 1

3 2

N-glycans N-glycans

GT33

1

2

1

N-glycans

GT16 GT58

1 1

1 1

1 1

N-glycans N-glycans

GT59 GT14

1 11

1 12

1 14

GT32

6

3

6

GT41 GT68 GT65 GT4b GT22

2 3 1 24 3

3 1 1 25 4

5 1 2 41 5

GT50 GT76

1 1

1 1

1 0

Other GT1b

121

202

330

GT75

5

3

11

N-glycans O-glycans; Other? O-glycans; Other? O-glycans O-glycans O-glycans GPI; Otherb GPI

N-glycan GalT N-glycan XylT Animal GlcATs Animal ManTs N-glycan FucT OligosaccharylT

Animal/fungal GTs Animal/fungal GlcTs

N-glycan GlcNAcT Animal/fungal GlcTs Animal/fungal ManTs Animal GTs Animal/fungal ManTs Animal GlcTs Animal GTs Animal/fungal GTs GlcNAcT α-1,6-GlcNAcT

Animal FucT Animal FucTs Animal/fungal ManT Animal ManT Animal/fungal ManT

GPI GPI

Hormones; Otherb Reversible O-glycan

Indirect evidence

Aromatic GTs Reversibly glycosylated polypeptide

Annotating Carbohydrate-Active Enzymes in Plant Genomes

■

101

Table 4.1 Continued A. tha

O. sat

P. tri a

GT29 GT64

3 3

5 3

GT19

1

GT30

Family

Metabolism

Known activities in class

5 4

Unknown Unknown

SiaT

1

1

Unknown

1

1

1

Unknown

GT51

0

0

3

Unknown

GT9

0

0

2

Unknown

Indirect evidence Animal SiaT Animal heparin/ heparan HexAT Bacterial LPS GTs Bacterial LPS GTs Bacterial peptidoglycan GTs Bacterial LPS GTs

a

In P. trichocarpa the total number of gene models identified in the different families may include partial sequences. Families having members involved in different major metabolic classes.

b

‘monocot’ rice have been noted (Yokoyama & Nishitani 2004) as well as expected differences in cell-wall metabolism between short-lived annual Arabidopsis versus long-lived poplar tree have been suggested (Tuskan et al. 2006). 3 A new vision of CAZymes as significant components of carbohydratebased systems is now emerging with the advent of a variety of postgenomic techniques. Examples include: N- and O-glycosylation of proteins, starch metabolism, biosynthesis of the cell wall and its subcomponents. Geisler-Lee et al. (2006) have combined bioinformatics and transcriptome analysis of various poplar and Arabidopsis tissues and organs and have shown that CAZyme transcripts are particularly abundant in wood tissues. For starch, however, the situation was different since there was little expression of genes related to starch metabolism during xylogenesis, consistent with the preferential flux of carbon to the cell wall. Even though a picture is emerging of a number of systems based on large sets of redundant isoforms of CAZymes with exquisite levels of control, the important issue is that genomic differences are further complicated by differential tissue expression. Interestingly, such large-scale efforts can associate established and new plant sequence families to specific metabolic processes like arabionoxylan biosynthesis (Mitchell et al. 2007). And the story is only beginning, as the arrival of more plant genomes and/or sampling of genomics data will lead to a larger coverage of the plant sequence ‘space’, which will in turn allow a more detailed understanding

102

■

Plant Polysaccharides

of plants in general. This is likely to open the doors for a better usage of plant biotechnology for a number of challenges ranging from the food and biomass production, to the production of proteins and compounds of pharmaceutical interest. The massive number of GTs in plants is due to the expansion of several extremely well-populated GT families: for instance Arabidopsis, rice and poplar have approximately 42, 47 and 76 family GT2 genes, respectively (see Table 4.1). The analysis in Table 4.1 already provides some details on how the different families relate with the metabolism. Major broad metabolic groups can be already be assigned for most families, while a few large families are likely to contain members that intervene in different metabolic pathways (indicated in Table 4.1). Table 4.1 shows that a number of basic aspects of plant metabolism, namely those associated with protein N- and O-glycosylation and glycosylphosphatidylinositol (GPI) anchor synthesis, have been directly characterized in plants, including members of the N-glycan-related families GT31, GT61, GT10 and GT13, O-glycan-related family GT41, and GPI-related family GT4. By contrast, the knowledge in others families is indirect and has, to some extent, benefited from the results obtained with other eukaryotes such as animals and fungi, where the simplicity of some model organisms has allowed significant advances in recent decades. In particular, there a large number of enzymes potentially involved in protein N-glycosylation that can be deduced for plants with varying levels of confidence, such as in families GT66, GT57, GT24, GT33, GT16, GT58 and GT59. For enzymes involved in protein O-glycosylation, the list is also long and is likely to include members of families GT14, GT32, GT68 and GT65, while the families putatively involved in GPI-anchor synthesis such as GT22, GT50 and GT76 have a smaller number of members. Detailed knowledge has also been obtained in the fields of carbohydrate energy storage and of glycolipid metabolism. The first category includes enzymes involved in the metabolism of starch (e.g. families GT5 and GT35), sucrose (GT4 and GT8) and trehalose (GT20). The second group contains enzymes synthesizing plastid and other glycolipids (GT4 and GT28), and sphingolipids (family GT21). These two groups of GT families are characterized by a relatively small number of isoforms in plants – an indication of their basal role – which probably facilitated the identification of significant mutants and the subsequent biochemical characterization of proteins. One of the major observations of the genomic content in CAZymes is that cell-wall-related enzyme families in plants tend to have a large number of members. This naturally includes the abundant processive polysaccharide synthases of family GT2, known to be implicated in cellulose, (gluco)mannan and β-1,3–1,4-glucan biosynthesis, but also families having members involved in the metabolism of xyloglucan (GT47, GT37, GT34), pectin (GT8,

Annotating Carbohydrate-Active Enzymes in Plant Genomes

■

103

GT77), arabinan (G47), rhamnogalacturonan (GT77), callose (GT48), galactomannan (GT34), etc. It is likely that these multigene families are a consequence of the genome expansion that accompanied the emergence of land plants (Nedelcu et al. 2006) and the evolution of complex and highly differentiated tissues. The multiplicity and probable functional redundancy of these genes, associated with the practical difficulties in biochemical and genetic characterization of the activity of the encoded proteins, are serious hurdles that render the elucidation of the molecular functions of this class of enzymes in plants particularly difficult. In line with the observation that enzymes known to be involved in plant cell-wall polysaccharide biosynthesis form abundant multigene families, it is tempting to speculate that other abundant, but less characterized or uncharacterized families such as GT61 could be involved in the synthesis of some cell-wall glycosides. In fact, recent bioinformatics analysis suggests that arabinoxylan synthesis may be associated with a number of established and unknown GT families (including GT43, GT47 and GT61) (Mitchell et al. 2007). Among the other abundant CAZyme families, it is worth mentioning the extremely large size of families GT1 and GH1 (Henrissat et al. 2001), which together constitute a complementary set for the synthesis and degradation of different secondary metabolites (Gachon et al. 2005). The plant cell wall is not only made of polysaccharides; it also contains lignin. Interestingly, many plant GH1 monolignol β-glucosidases are involved in lignification by cleaving the β-glucosyl component of soluble forms of lignin precursors (Escamilla-Treviño et al. 2006), and likewise several GT1 enzymes are capable of conjugating glucose to lignin monomers (Lanot et al. 2006).

4.4

Plant CAZymes: problems in functional annotation

As with other organisms with fully or partially sequenced genomes, any newly obtained plant protein sequence should benefit from previous biochemical characterization(s) to derive accurate functional predictions. Unfortunately, this task is made extremely difficult by the massive pollution of sequence descriptors in general databases, which, sadly, serve for the functional predictions of novel genomes, and which therefore contribute in turn to the propagation of errors (Gilks et al. 2002). The reasons behind such a problem are multiple and include: (1) massive unsupervised functional annotation of genomes, (2) manual annotation of organisms by generalist annotators lacking expert knowledge; (3) insufficient coverage of the sequence-specificity space; (4) insufficient ‘percolation’ of the experimental data in general databases, a consequence of present-day annotation policies (and perhaps of insufficient funding): except for a few model organisms, the genomes are annotated at the time of publication and are rarely re-annotated subsequently in the light of subsequent biochemical advances.

104

■

Plant Polysaccharides

In CAZy, our present goal is to limit the extent of erroneous annotations of CAZymes, by relying systematically on biochemically based and, to some extent, genetically based molecular functional assignments from peerreviewed efforts and a regular interaction with the glycobiological community. This approach has allowed us to maintain consistent annotation that may constitute the basis for the required, experimentally supported, automated annotation (Valencia 2005), and can be used for genome annotation and re-annotation (Ouzounis & Karp 2002). Our present CAZy annotation scheme derives from our participation in a number of genome annotation efforts for organisms in all kingdoms, including in particular the poplar genome (Tuskan et al. 2006). Our approach involves two steps: the initial assignment of predicted protein models to CAZy families, followed by the more challenging task (by size and nature) of inferring as accurately as possible the biochemical function and specificity of hundreds to thousands of potential CAZymes per genome. Most of the functional annotation difficulties reside in the fact that small stereochemical differences between carbohydrates (e.g. d-glucose and d-mannose are just epimers at C-2) and the multiple ways of attaching them one to another are exploited by nature to achieve very different functions, and generate a colossal diversity of oligo- and polysaccharides. The consequent diversity of substrates and of reaction products requires an equivalent diversity of CAZymes. As the number of protein folds of CAZymes is much smaller than the number of existing specificities, the sequence-based families almost invariably contain enzymes of differing function. As a consequence, and as already highlighted in Table 4.1, a simple family assignment is not sufficient for an accurate prediction of substrate specificity. The solution to this problem requires the adequate combination of experimental approaches and bioinformatics. For CAZymes, the sequencebased families are progressively being further refined by the definition of appropriate subfamilies, in the expectation that more closely related sequenced are more likely to display identical substrate specificity. This can be exemplified by our analysis of glycoside hydrolase family GH13 (Stam et al. 2006), which showed that the majority of subfamilies have very narrow substrate specificity. We and others have started efforts that will eventually lead to the definition of hundreds if not thousands of subfamilies of CAZymes. Combining the results of such subfamily analyses to the massive advances in experimental investigations of proteins from Arabidopsis and other model plants will allow us to determine how far can a function be extended to related sequences. Although we anticipate that the thresholds may vary from one family to another, such an effort, combined with experimental characterization, is needed to harness the true extent of the plant ‘CAZomes’. En route to this exciting future, several chapters of this book give a detailed account of the current knowledge of various families of plant CAZymes involved in the biosynthesis of plant cell-wall polysaccharides.

Annotating Carbohydrate-Active Enzymes in Plant Genomes

■

105

References Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature, 408, 796–815. Berteau, O., Stenutz, R. (2004) Web resources for the carbohydrate chemist. Carbohydrate Research, 339(5), 929–936. Boraston, A.B., Bolam, D.N., Gilbert, H.J., Davies, G.J. (2004) Carbohydrate-binding modules: fine-tuning polysaccharide recognition. Biochemical Journal, 382, 769–781. Campbell, J.A., Davies, G.J., Bulone, V., Henrissat, B. (1997) A classification of nucleotide-diphospho-sugar glycosyltransferases based on amino acid sequence similarities. Biochemical Journal, 326, 929–939. Cantarel, B.L., Coutinho, P.M., Rancurel, C., Bernard, T., Lombard, V., Henrissat, B. (2009) The Carbohydrate-Active EnZymes database (CAZy): an expert resource for Glycogenomics. Nucleic Acids Res., 37, D233–D238. Coutinho, P.M., Henrissat, B. (1999) Carbohydrate-active enzymes: an integrated database approach. In: Recent Advances in Carbohydrate Bioengineering (eds H.J. Gilbert, G. Davies, B. Henrissat, B. Svensson), pp. 3–12. Royal Society of Chemistry, Cambridge. Coutinho, P.M., Deleury, E., Davies, G.J., Henrissat, B. (2003) An evolving hierarchical family classification for glycosyltransferases. Journal of Molecular Biology, 328, 307–317. Escamilla-Treviño, L.L., Chen, W., Card, M.L., Shih, M.C., Cheng, C.L., Poulton, J.E. (2006) Arabidopsis thaliana β-glucosidases BGLU45 and BGLU46 hydrolyse monolignol glucosides. Phytochemistry, 67, 1651–1660. Farrokhi, N., Burton, R.A., Brownfield, L., et al. (2006) Plant cell wall biosynthesis: genetic, biochemical and functional genomics approaches to the identification of key genes. Plant Biotechnology Journal, 4, 145–167. Gachon, C.M., Langlois-Meurinne, M., Saindrenan, P. (2005) Plant secondary metabolism glycosyltransferases: the emerging functional analysis. Trends in Plant Science, 10, 542–549. Geisler-Lee, J., Geisler, M., Coutinho, P.M., et al. (2006) Poplar carbohydrate-active enzymes. Gene identification and expression analyses. Plant Physiology, 140, 946–962. Gilks, W.R., Audit, B., De Angelis, D., Tsoka, S., Ouzounis, C.A. (2002) Modeling the percolation of annotation errors in a database of protein sequences. Bioinformatics, 18, 1641–1649. Henrissat, B. (1991) A classification of glycosyl hydrolases based on amino-acid sequence similarities. Biochemical Journal, 280, 309–316. Henrissat, B., Bairoch, A. (1993) New families in the classification of glycosyl hydrolases based on amino-acid sequence similarities. Biochemical Journal, 293, 781–788. Henrissat, B., Bairoch, A. (1996) Updating the sequence-based classification of glycosyl hydrolases. Biochemical Journal, 316, 695–696. Henrissat, B., Claeyssens, M., Tomme, P., Lemesle, L. & Mornon, J.P. (1989) Cellulase families revealed by hydrophobic cluster analysis. Gene, 81, 83–95. Henrissat, B., Coutinho, P.M., Davies, G.J. (2001) A census of carbohydrate-active enzymes in the genome of Arabidopsis thaliana. Plant Molecular Biology, 47(1–2), 55–72.

106

■

Plant Polysaccharides

Kacser, H., Beeby, R. (1984) Evolution of catalytic proteins or on the origin of enzyme species by means of natural selection. Journal of Molecular Evolution, 20, 38–51. Koch, K.E. (1996) Carbohydrate-modulated gene expression in plants. Annual Reviews in Plant Physiology and Plant Molecular Biology, 47, 509–540. Lairson, L.L., Withers, S.G. (2004) Mechanistic analogies amongst carbohydrate modifying enzymes. Chemical Communications (Camb.), 20, 2243–2248. Laine, R.A. (1994), A calculation of all possible oligosaccharide isomers both branched and linear yields 1.05 × 1012 structures for a reducing hexasaccharide: the Isomer Barrier to development of single-method saccharide sequencing or synthesis systems. Glycobiology, 4, 759–767. Lanot, A., Hodge, D., Jackson, R.G., et al. (2006) The glucosyltransferase UGT72E2 is responsible for monolignol 4-O-glucoside production in Arabidopsis thaliana. Plant Journal, 48, 286–295. MacGregor, E.A., Svensson, B. (1989) A super-secondary structure predicted to be common to several α-1,4-d-glucan-cleaving enzymes. Biochemical Journal, 259, 145–152. Minic, Z., Jamet, E., Négroni, L., Arsene der Garabedian, P., Zivy, M., Jouanin, L. (2007) A sub-proteome of Arabidopsis thaliana mature stems trapped on Concanavalin A is enriched in cell wall glycoside hydrolases. Journal of Experimental Botany, 58(10), 2503–2512. Mitchell, R.A., Dupree, P., Shewry, P.R. (2007) A novel bioinformatics approach identifies candidate genes for the synthesis and feruloylation of arabinoxylan. Plant Physiology, 144, 43–53. Nedelcu, A.M., Borza, T., Lee, R.W. (2006) A land plant-specific multigene family in the unicellular Mesostigma argues for its close relationship to Streptophyta. Molecular and Biological Evolution, 23, 1011–1015. Ouzounis, C.A., Karp, P.D. (2002) The past, present and future of genome-wide reannotation. Genome Biology, 3, comment2001.1–2001.6. Rees, D.A. (1977) Polysaccharide Shapes. Chapman & Hall, London. Scheible, W.R., Pauly, M. (2004) Glycosyltransferases and cell wall biosynthesis: novel players and insights. Current Opinion in Plant Biology, 7, 285–295. Stam, M.R., Danchin, E.G., Rancurel, C., Coutinho, P.M., Henrissat, B. (2006) Dividing the large glycoside hydrolase family 13 into subfamilies: towards improved functional annotations of α-amylase-related proteins. Protein Engineering, Design, Selection, 19, 555–562. Svensson, B. (1988) Regional distant sequence homology between amylases, αglucosidases and transglucanosylases. FEBS Letters, 230, 72–76. Tuskan, G.A., Difazio, S., Jansson, S., et al. (2006) The genome of black cottonwood, Populus trichocarpa (Torr. and Gray). Science, 313, 1596–1604. Valencia, A. (2005) Automatic annotation of protein function. Current Opinion in Structural Biology, 15, 267–274. Varki, A. (1999) Essentials in Glycobiology, CSHL Press, Cold Spring Harbour, NY. Ward, O.P., Moo-Young, M. (1989) Enzymatic degradation of cell wall and related plant polysaccharides. Critical Reviews in Biotechnology, 8, 237–274. Webb, E.C. (1992) Enzyme Nomenclature 1992: Recommendations of the Nomenclature Committee of the International Union of Biochemistry and Molecular Biology on Nomenclature and Classification of Enzymes by the Reactions they Catalyse. Academic Press, San Diego, CA.

Annotating Carbohydrate-Active Enzymes in Plant Genomes

■

107

Wheeler, D.L., Chappey, C., Lash, A.E., et al. (2000) Database resources of the National Center for Biotechnology Information. Nucleic Acids Research, 28, 10–14. Yokoyama, R., Nishitani, K. (2004) Genomic basis for cell-wall diversity in plants. A comparative approach to gene families in rice and Arabidopsis. Plant Cell Physiology, 45, 1111–1121. Yu, J., Hu, S., Wang, J., et al. (2002) A draft sequence of the rice genome (Oryza sativa L. ssp. indica). Science, 296, 79–92.

Annual Plant Reviews (2011) 41, 109–166 doi: 10.1002/9781444391015.ch5

Chapter 5

BIOSYNTHESIS OF PLANT CELL WALL AND RELATED POLYSACCHARIDES BY ENZYMES OF THE GT2 AND GT48 FAMILIES Bruce A. Stone1, Andrew K. Jacobs2, Maria Hrmova2, Rachel A. Burton2 and Geoffrey B. Fincher2 1

School of Biochemistry, La Trobe University, Bundoora, Vic 3083, Australia Australian Centre for Plant Functional Genomics, University of Adelaide, Waite Campus, Glen Osmond, SA 5064, Australia 2

Manuscript received August 2008

Abstract: In this chapter we outline the evolution of our understanding of the biological functions, genetics and regulation of enzymes of the GT2 and GT48 families of glycosyl transferases. The GT2 family is very large and includes enzymes encoded by the cellulose synthase gene superfamily, together with many other transferases with diverse substrate specificities that are distributed from the Archaea to humans. On the other hand, there are relatively few known members of the GT48 family, in which activities are limited to putative (1,3)-β-d-glucan synthases of embryophytes, fungi and yeasts. The review is focused on a number of individual case studies, which have been chosen on the basis of their biological and economical importance in plant biology. The history of our knowledge of (1,4)-β-d-glucan (cellulose) synthases and (1,3;1,4)-β-d-glucan synthases in plants will be reviewed as representatives of the GT2 family, while the (1,3)-β-d-glucan (or callose) synthases will be compared with these, as representatives of the GT48 family. These synthases have been extensively studied using biochemical techniques, but they are associated with membranes and are not easily purified. Emerging functional genomics technologies have been used to identify the genes encoding the cellulose synthases of the GT2 family, while new proteomics Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

109

110

■

Plant Polysaccharides

procedures have provided amino acid sequence that has in turn been applied to the identification of the GT48 (1,3)-β-d-glucan synthases of plants. The enzymes are involved in the synthesis of structurally similar polysaccharides, but appear to have evolved independently. Keywords: biochemistry; cellulose; functional genomics; (1,3;1,4)-β-d-glucan; molecular biology; structural biology

(1,3)-β-d-glucan;

5.1 Introduction The GT2 family of glycosyl transferases includes proteins encoded by the cellulose synthase gene superfamily. It also contains chitin and hyaluronan synthases, and many other transferases with diverse substrate specificities. There are more than 8000 entries for GT2 enzymes in the CAZY databases (Coutinho & Henrissat 1999; http://www.cazy.org/) and the family embodies enzymes that are widely distributed in nature, from the Archaea to humans. In contrast, the GT48 family has relatively few entries and known activities in the family are limited to putative (1,3)-β-d-glucan synthases, which are found in embryophytes, fungi and yeasts (Fig. 5.1). In this chapter we have combined discussion of the GT2 and GT48 families of glycosyl transferases, because in plants at least the two families share a common history with respect to our understanding of their biology and to the methods that have been used to define their characteristics. Thus, during the era of biochemistry in the 1970s and 1980s, plant cellulose synthases of the GT2 family and (1,3)-β-d-glucan synthases of the GT48 family were the subject of much attention and in many cases interpretation of results was complicated by the presence in plant tissue extracts of both enzyme types competing for the same sugar nucleotide substrate. The situation was further complicated in studies on the cereals and grasses, where GT2 (1,3;1,4)-β-d-glucan synthases were also present in many tissue extracts and also used UDP-Glc as a substrate. The enzymes are associated with membranes and are not easily purified. The attendant inability to generate amino acid sequence information through biochemical methods also meant that the early years of molecular biology shed little further light on the nature of plant enzymes of the GT2 and GT48 families, or on the genes that encoded them. At the beginning of the era of functional genomics, however, transcript analyses finally started to provide information on the genes encoding the cellulose synthases of the GT2 family and developing proteomics procedures were used to generate amino acid sequence information on the GT48 (1,3)-β-d-glucan synthases of plants. The subsequent growth of expressed sequence tag (EST) databases and the availability of plant genome sequences, coupled with developing bioinformatics and other comparative genomics technologies, have allowed the gene families of key members of the two groups to be identified and characterized in detail. However, the data so generated has not always provided us with insights into the function of the enzymes at the molecular and cellular levels. Furthermore, their precise

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

111

A GT2 family 2 85 14 27 352 18 69 247 57 61 630

Saccharomyces cerevisiae Fungi Caenorhabditis elegans Nematoda Metazoa Fruit fly Arthropoda Chordata Mouse Human Eukaryota

2 Unclassified Virus 20 Archaea 442 Bacteria 8604 Cyanobacteria 623 Synechocystis PCC 6803 24 72 Oryza sativa (rice) 25 Arabidopsis thaliana 125 Green plants 173 Plastid group 20 Other Eukaryota

B GT48 family 7 91

189

Saccharomyces cerevisiae Fungi Caenorhabditis elegans Nematoda Metazoa Fruit fly Arthropoda Chordata Mouse Human Eukaryota

Unclassified Virus Archaea Bacteria Cyanobacteria Synechocystis PCC 6803 Oryza sativa (rice) Arabidopsis thaliana Green plants Plastid group Other Eukaryota

45 22 98 98

Figure 5.1 Distribution of GT2 (A) and GT48 (B) glycosyl transferases. The figures show the taxonomic range of amino acid sequences from the major groups of organisms and the number of sequences associated with each lineage. The tree root is at the centre of the circular taxonomic display and model organisms are present in the outer circle. Nodes of the tree are labelled via radial lines, which extend to the exterior of the circle. The position of a node on the inner circle has no significance, although the program attempts to group the nodes. The nodes represent true taxonomy nodes with the exception of those labeled Unclassified, Other Eukaryota or Plastid Group which are artificial nodes created for simplicity. Images were generated at the InterPro web site (http:// www.ebi.ac.uk/interpro) by searching with the InterProScan tool and the sequences of the GT2 and GT48 families.

substrate specificities and kinetics, protein–protein interactions that might be necessary for their activity, or their regulation by metabolites or posttranslational modifications were difficult to assess. One might predict therefore that as these questions become more important in defining biological systems or in manipulating plant crops for enhanced performance, we will inevitably return to biochemical methods to complete our understanding of the genetics, cell biology, enzymology and three-dimensional (3D) structure of members of these two groups of glycosyl transferases. In this review, we track the history of techniques that have been available for the characterization of glycosyl transferases generally, but with an emphasis on their success or otherwise in the evolution of our understanding of the biological functions, genetics and control of the enzymes of the GT2 and GT48 families. Given the size and diversities of the two families, in particular the GT2 family, we will by necessity focus on a number of

112

■

Plant Polysaccharides

individual case studies, through which various experimental approaches have combined to give us a great deal of biological information, which is nevertheless incomplete at this stage. The individual cases for discussion have been chosen on the basis of their biological and economical importance in plant science and utilization. More specifically, the history of our knowledge of (1,4)-β-d-glucan (cellulose) synthases and (1,3;1,4)-β-d-glucan synthases in plants will be reviewed as representatives of the GT2 family, while the (1,3)-β-d-glucan (or callose) synthases will be compared with these, as representatives of the GT48 family. These three groups of enzymes mediate in the synthesis of polysaccharides that are closely related in their chemistry, but appear to have evolved along different routes and are now represented in two quite different families.

5.2 Structures and distribution of β-D-glucans synthesized by GT2 and GT48 enzymes The (1,4)-β-d-glucans (cellulose), (1,3)-β-d-glucans and (1,3;1,4)-β-d-glucans are all polysaccharides that are composed predominantly of β-linked glucosyl residues and, in plants, are generally linear polysaccharides with few or no branches or main chain substitutions. Despite these similarities, the different linkage configurations between the β-glucosyl residues have dramatic effects on the final conformations and hence biological functions of the three polysaccharides (Fig. 5.2). While the focus in this review will remain upon

Figure 5.2 Structural differences between (1,4)-β-D-glucans (cellulose), (1,3)-β-Dglucans and (1,3;1,4)-β-D-glucans of plants. The top section shows the structure of a typical (1,3;1,4)-β-D-glucan and indicates positions along the chain that are hydrolysed by family GH17 (1,3;1,4)-β-D-glucan endo-hydrolases. The cellotriosyl and cellotetraosyl resiudes, linked by single (1,3)-β-linkages, are also evident.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

113

higher plant β-d-glucans that are synthesized by enzymes from the GT2 and GT48 families, we nevertheless extend our considerations of the structure and function of these polysaccharides to other organisms, especially where it is possible to gain some insight into the biological strategies and enzymic mechanisms used by members of the GT2 and GT48 families. The structures and distributions of polysaccharides that are believed to be synthesized, in part at least, by GT2 and GT48 enzymes are summarized in Table 5.1. Cellulose is a linear homopolymer of β-d-glucopyranosyl residues, all of which are (1,4)-linked. Individual molecules have a degree of polymerization (DP) of up to 6000 in primary walls and 14 000 in secondary walls (Fincher & Stone 2004). These polysaccharides have an extended, ribbon-like conformation (Fig. 5.2) that allows parallel packing of chains into 3D, fibrillar aggregates stabilized by extensive intermolecular hydrogen bonding and van der Waals forces. The aggregates are referred to as microfibrils and contain approximately 35 individual cellulose molecules that are generally packed into parallel, highly ordered, crystalline arrays (Newman et al. 1996; Smith et al. 1998). The highly regular structure of cellulose allows individual molecules to align into very long, insoluble microfibrils. The physical properties, chemical reactivity and biological functions of cellulose are essentially determined by the regular packing of the cellulose molecules in the microfibrils. For its density (1.5 mg m−3) cellulose fibres are stiffer (Young modulus 50–130 GPa) and stronger (1 GPa), when measured along the polymer length, than nylon, silk, chitin, collagen, tendon or bone (Ashby et al. 1995). Cellulose microfibrils can also adopt higher level structures. For example, the chromistan haptophytes (single-celled green algae) have cell surface scales containing microfibrillar cellulose, in some xanthophytes (yellowgreen algae) the cyst and cell walls are cellulosic and some chrysophytes (golden algae) have cellulosic scales and loricae composed of a mesh of cellulosic fibres. Among the alveolates, some dinoflagellates have flattened vesicles underlying the plasma membrane, known as alveoli, and these contain cellulosic plates. The (1,3)-β-d-glucans of plants fulfil a number of roles and, inter alia, may function as specialized components of some cell walls. Certain plant (1,3)-βd-glucans are referred to as callose, on the basis of their intense yellow UV-induced fluorescence in the presence of the aniline blue fluorochrome. Like cellulose, the (1,3)-β-d-glucans are linear homopolymers of β-dglucopyranosyl residues, but in contrast to cellulose the β-d-glucopyranosyl residues are (1→3)-linked. The DP of linear (1,3)-β-d-glucans depends on the source. Brown algal laminarins have a DP of about 30 (Harada 1979), whereas bacterial curdlan chains have DPs of the order of 12 000 (Futatsuyama et al. 1999) but are generally polydisperse with respect to their molecular size distribution. Linear (1,3)-β-d-glucans are predicted to adopt a loose helical conformation (Fig. 5.2) and in the solid state adopt a triple helical structure in which three intertwining glucan chains form a triple helix. Each chain has a right handed, six-fold conformation and the chains in the helix

Glc

Glc

Glc

Glc

Glc

Glc

Glc

1,3

1,3

1,3

1,3

1,3;1,6

1,3;1,4

1,3; 1,4

Lichenin

cereal Glucan

Yeast glucan Fungal glucan

Callose

Paramylon

Laminarin

Curdlan

Glc

1,3

Name

Cellulose

Monosaccharide

Commelinid monocots (especially Poaceae) Lichens

Yeasts Fungi

Embryophytes

Chromistan oomycete fungi

Euglenoid protozoa

Bacteria, cyanobacteria, slime moulds, dinoflagellates, green, red and brown algae, chromistan comycetes and hypochytrids, chrysophytes, haptophytes, embryophytes, tunicate tests Rhizobiaceae and Cellulomonas flavigena Chromistan brown algae

Organism

Mycobiont cell walls

Cell walls

Cell walls Cell walls

Special cell wall deposits

Cell walls

Intracellular

Structural

Structural

Structural, wound response Structural Structural

Carbohydrate reserve Carbohydrate reserve Structural

Protective

Capsular Intracellular

Structural

Function

Cell walls, cell envelopes, sheaths, scales,plates and capsules

Distribution

Honegger & Haisch (2001)

Kollar et al. (1997) Bernard & Latge (2001) Harris (2005)

Sietsma et al. (1969) Stone (2006)

Stone (2006)

McIntosh et al. (2005) Stone (2006)

Stone (2005)

References

■

β-Glycan homopolysaccharides 1,4 Glc

Major linkage(s)

Table 5.1 Structure and distribution of β-glycans synthesized by GT2 and GT48 enzymes

114 Plant Polysaccharides

Glc

GlcNAc

Man (Gal)

Man, Glc

1,3; 1,2

1,4

1,4

1,4

Glc, GlcA

GlcA, GlcNac

GlcA, GalNac

1,3;1,4

1,4;1,4

1,3;1,4

β-Glycan heteropolysaccharides 1,4 Glc, Xyl, Gal, Fuc,(Ara) 1,3;1,4 GlcA, GlcNAc

Glc

Monosaccharide

1,3; 1,6 (cyclic)

Major linkage(s)

Chondroitin

Pasteurella multocida, E. coli K5 Pasteurella multocida, E. coli K4

Streptococcus pneumoniae

Streptococci, Pasteurella multocida, vertebrates

Hyaluronan

Pneumococcal type 3 polysaccharide Heparosan

Embryophytes

Embyophytes

Actinomycetes, Entamoebafungi, chromistan comycetes and hypochyrtrids, diatoms, haptophytes, invertebrates Embryophytes

Streptococcus pneumoniae

Bradyrhizobium, Rhizobium meliloti, Azospirillum brasilianse

Organism

Xyloglucan

Glucomannan

Mannan (galactomannan)

Pneumococcal type 37polysaccaride Chitin

Name

Capsules

Capsules

Capsules

Capsules, extracellular matrix

Cell walls

Cell walls

Cell walls

Cell walls, exoskeletons

Capsules

Extracellular Periplasmic

Distribution

Protective

Protective

Protective

Protective, structural

Structural

Structural and carbohydrate reserve Structural

Structural

Protective

? Osmotic

Function

Kass & Seastone (1944), Carter & Annau (1953), Fraser et al. (1997 Reeves & Goebel (1941) Rimler (1994), Vann et al. (1981) DeAngelis & Padgett-McCue (2000), Rodriguez et al. (1988)

Bacic et al. (1988)

Bacic et al. (1988)

Bacic et al. (1988)

Martínez & Gozalbo (2001)

Breedveld & Miller (1994) Pfeffer et al. (1996), Altabe et al. (1994) Knecht et al. (1970)

References Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families ■

115

116

■

Plant Polysaccharides

are linked together through triads of strong hydrogen bonds between C(O)2 hydroxyls. Linear (1,3)-β-d-glucans can also form microfibrils. Thus, curdlan in the capsule of Agrobacterium is present in the form of fine microfibrils and cell-free systems of Rubus (Pelosi et al. 2003) produce similar microfibrils in which the (1,3)-β-d-glucans chains are present as a mixture of single and multiple chains. Paramylon, the storage (1,3)-β-d-glucan in euglenids, is found in highly crystalline granules containing concentrically wound microfibrils. Curdlan and other high-DP (1,3;1,6)-β-d-glucans form gels from aqueous suspension or solutions. More generally, (1,3)-β-d-glucans and related (1,3;1,6)-β-d-glucans are widely represented in biological systems. Structural variants of the linear (1,3)-β-d-glucans include cyclic (1,3;1,6)-β-d-glucans, branch-on-branch (1,3;1,6)-β-d-glucans, side-chain-branched (1,3;1,6)-β-d-glucans, (1,3;1,6)-βd-glucans with both (1,3)- and (1,6)-linkages in the main chain, and side chain branched (1,3;1,2)-β-d-glucans (Table 5.1). Finally, the (1,3;1,4)-β-d-glucans, also referred to as mixed-linkage or cereal β-glucans, occur almost exclusively as major cell wall components in members of the monocotyledon family Poaceae, to which the cereals and grasses belong, and in related families of the order Poales (Trethewey et al. 2005). A structurally related (1,3;1,4)-β-d-glucan, lichenin, is found in the walls of the fungal component of the lichen, Iceland moss (Cetraria islandica) (Honegger & Haisch 2001) and is covalently attached to the non-reducing ends of the core of (1,3;1,6)-β-d-glucan of the cell wall polysaccharide complex in the yeast Saccharomyces cerevisiae (Kapteyn et al. 1997) and the fungus Aspergillus fumigatus (Bernard & Latgé 2001). (1,3;1,4)-β-d-Glucans have also been reported in a dinoflagellate (Stone & Clarke 1992). Again, (1,3;1,4)-β-d-glucans are linear, unbranched polysaccharides containing β-d-glucopyranosyl monomers. But in this case, the β-d-glucopyranosyl residues are polymerized through both (1,4)- and (1,3)-linkages (Fig. 5.2). The ratio of (1,4)- to (1,3)-linkages is usually in the range 2.2–2.6 (Fincher & Stone 2004), but the arrangements of these linkages are the key determinants of the polysaccharide’s shape and hence its function in the cell wall. The (1,3)- and (1,4)-linkages in (1,3;1,4)-β-d-glucans of the cereals and grasses are not arranged in regular repeating sequences. If this were the case, individual polysaccharide chains would have a degree of shape complementarity that would enable them to align into aggregates with limited solubility in aqueous media. While they are not arranged in a strict repeating structure, the (1,3)and (1,4)-linked β-d-glucopyranosyl residues are not randomly arranged either. Single (1,3)-linkages are generally separated by two or more adjacent (1,4)-linkages. Regions of two or three adjacent (1,4)-linkages predominate, but up to 10% of the chain may consist of longer stretches of 5–15 adjacent (1,4)-linkages (Woodward et al. 1983b). Adjacent (1,3)-linked β-dglucopyranosyl residues are rare or absent. In other words, the cereal (1,3;1,4)-β-d-glucans can be considered as polymers that consist mostly of (1,3)-β-linked cellotriosyl or cellotetraosyl residues. The ratio of tri- to tetrasaccharide units varies between species and Markov chain analysis indi-

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

117

cates that these oligosaccharide units are arranged at random along the chain, which means that the single (1,3)-linked β-d-glucopyranosyl residues are also distributed unevenly along the chain. This severely limits the ability of individual polysaccharide chains to interact over extended regions or to aggregate into insoluble, fibrillar material (Fincher & Stone 2004). Thus, the linear, unbranched (1,3;1,4)-β-d-glucans polysaccharides, which are polydisperse in nature and can have degree of polymerization (DP) values well in excess of 1000, are soluble in aqueous medium. The net effect of these properties is that the ‘cellulosic’ (1,4)-linked β-dglucopyranosyl residues in (1,3;1,4)-β-d-glucans impart some rigidity and an extended conformation to internal portions of the molecule, while the (1,3)-linked β-d-glucopyranosyl residues introduce relatively flexible molecular ‘kinks’ in the molecule that prevent extensive alignment of the chains. The result is an extended, worm-like chain that is highly asymmetrical (Fig. 5.2). The axial ratio of the barley (1,3;1,4)-β-d-glucan that is soluble in water at 40 °C is about 100 (Woodward et al. 1983a). High molecular weight, soluble and asymmetrical polysaccharides with these properties typically form aqueous solutions of high viscosity. Gelation of (1,3;1,4)-β-D-glucans occurs by interactions between pairs of consecutive cellotriosyl units, which predominate in wheat and lichenin (1,3;1,4)-β-d-glucan but are less frequent in those from barley and oats, and which allow the formation of junction zones between adjacent chains (Bohm & Kulicke 1999; Cui 2001a). The higher proportion of cellotriose units in the wheat (1,3;1,4)-β-d-glucan accounts for its relative water unextractibility compared with the (1,3;1,4)-β-d-glucans from barley or oats. As mentioned above, the 3D structures and chemical properties of the biologically and commercially important polysaccharides under examination here are the prime determinants of biological function. Thus, the extended, highly regular conformation of individual (1,4)-β-d-glucan molecules allows them to pack, through extensive intermolecular hydrogen bonding, into very long microfibrils that are perfectly adapted to the function of cellulose as a reinforcing element with high tensile strength in the cell walls of plants. The irregular structure and high DP of the (1,3;1,4)-β-d-glucans allows them to form the gel-like matrix of cell walls that is built around the microfibrillar phase. The matrix phase of primary walls of plants consists of other polysaccharides, including xyloglucans, heteroxylans and pectic polysaccharides. In each case, but using different chemistries, these polysaccharides form open gel-like structures that contribute to the strength and flexibility of the wall, but also allow the diffusion of water, nutrients, phytohormones and other small molecules across the walls during normal cell growth and development. In some cases, these wall polysaccharides might have a secondary function as a store of metabolizable energy. Thus, the (1,3;1,4)-β-d-glucans of barley leaves are degraded in the dark (Roulin et al. 2002), while xyloglucans and galactoglucomannans in the walls of storage tissues in various plants are depolymerized following germination as a source of monosaccharides for energy production in the developing seedling (Bacic et al. 1988).

118

■

Plant Polysaccharides

In some primary walls, for example in the endosperm of grasses, cellulose contributes as little as 2–3% of the dry weight of the wall, in other primary walls as much as 40%. In primary walls, the cellulose microfibrils are very narrow (2–3 nm) and are deposited in lamellae of which there may be only about four in the thin primary wall. The secondary wall layer in xylem fibres and tracheids, the principal cellular components of wood, is deposited inside the primary wall after the elongation phase of wall growth has been completed. Secondary walls are rich in cellulose (35–60% dry weight) and usually consist of three distinct layers of precisely parallel microfibrils, whose oriA

B

Cellulose microfibrils being firmly integrated into the wall

C

Extracellular space

Plasma membrane Microtubule anchored to plasma membrane Cytosol

Cellulose synthase complex (rosette)

Figure 5.3 (A) Parallel orientation of cellulose microfibrils in secondary walls (image courtesy of R. Malcolm Brown and Takeo Itoh). (B) Microscopic appearance of terminal complex rosettes in the plasma membrane (image courtesy of Candace Haigler and Mark Grimson). (C) Proposed guidance system for cellulose synthesis. Parallel arrangements of microtubules just beneath the surface of the plasma membrane are believed to limit the tracking of cellulose synthase complexes, such that newly synthesized cellulose microfibrils are forced into a parallel orientation (diagram kindly provided by Andrew Staehelin; copyright 2009 from Plant Cell Walls by Albersheim et al. Reproduced by permission of Garland Science/Taylor & Francis LLC).

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

119

entation is different in successive layers (Fig. 5.3A). Non-cellulosic polysaccharides such as xyloglucans and heteroxylans form the matrix of the secondary wall, and in some cell types during the later stages of wall differentiation, lignins are deposited in the matrix of both the primary and secondary wall, thereby enveloping the cellulosic microfibrils and conferring resistance to compression on the tissue concerned. The functions of the (1,3)-β-d-glucans in higher plants are more diverse (Stone & Clarke 1992). During normal growth and development, the (1,3)-βd-glucan known as callose is deposited as a transitory component of the cell plate in dividing cells and of the cell wall during endosperm cellularization in developing grain (Wilson et al. 2006). It is also found in pollen mother cell walls and pollen tubes, in plasmodesmatal canals, in abscission zones, at wound sites, and on sieve plates in dormant phloem (Stone & Clarke 1992). Deposition of callosic plugs, or papillae, at sites of fungal penetration has been observed as an early response of plants to microbial attack and has been linked with wound responses when plant surfaces are physically damaged. In other eukaryotic systems, (1,3)-β-d-glucans may function as storage polysaccharides, as observed in euglenoid protozoa and haptophytes (paramylon), chromistan brown algae and oomyctes, and some diatoms and chrysophytes (laminarin-type glucans). They are also wall components of yeasts and fungi [branch-on-branch (1,3;1,6)-β-d-glucans)], or act as fungal surface mucilages [side-chain-branched (1,3;1,6)-β-d-glucans]. In prokaryotes, linear (1,3)-β-d-glucan (curdlan), cyclic (1,3;1,6)-β-d-glucans and the side-chain-branched (1,3;1,2)-β-d-glucans act as capsular components with protective functions.

5.3 Early biochemical approaches to plant β-D-glucan synthases Now let us return to the early biochemical work on the characterization of the enzymes of the GT2 and GT48 families as participants in the biosynthesis of this set of plant β-d-glucans. It was established early on that the biosynthetic enzymes were membrane-associated. The usual experimental approach was therefore to isolate microsomal membrane preparations either from whole tissue extracts or from cellular fractions enriched in organelles such as Golgi. In many cases coleoptiles, hypocotyls or suspension-cultured cells were chosen as sources of membrane preparations, on the grounds that they were easy to obtain in quantity and because wall biosynthesis occurs rapidly in this material. The microsomal membrane preparation was incubated in vitro with labeled sugar nucleotides, such as UDP-[14C]Glc, and the incorporation of [14C] into ethanol-insoluble polysaccharides was monitored. Although the radioactivity-based assays were very sensitive, the values calculated for specific activity were usually very low, typically in the pmol min−1 mg protein−1 range (Raymond et al. 1978). Furthermore, the

120

■

Plant Polysaccharides

putative polysaccharide synthase activity in the membrane preparations rapidly decayed with time. Products of the in vitro assays were painstakingly characterized through gas–liquid chromatography (GLC) of partially methylated alditol acetates and/or with specific (1,3)-, (1,4)- or (1,3;1,4)-β-d-glucanases. In early work attempts to obtain cell-free synthesis of plant cellulose instead encountered substantial (1,3)-β-d-glucan synthesis. This was invariably observed in microsomal preparations from dicotyledonous plants, where (1,4)-β-dglucosyl residues could be detected at low (5 μM) concentrations of the UDP-Glc substrate, but at higher substrate concentrations (5 mM) most of the [14C] was incorporated into (1,3)-β-d-glucosyl residues (Ray et al. 1969; Heiniger & Delmer 1977; Raymond et al. 1978). It was also difficult to demonstrate that entire polysaccharide chains had been synthesized de novo, rather than the more likely addition of just a few glucosyl residues to existing polysaccharide chains that might be associated with the membrane preparation. In one study the addition of at least 30 glucosyl residues was measured and it was additionally shown that chain extension was occurring at the non-reducing end of the growing chain (Henry & Stone 1982). The prime target in most of these studies was to investigate cellulose biosynthesis in vitro, but the low levels of activity and amounts of reaction products precluded any confident conclusions that microfibrillar cellulose has actually been synthesized in vitro. Indeed, high-level in vitro cellulose biosynthesis by plant enzymes was convincingly demonstrated only relatively recently, when rosette-like complexes were observed at the termini of cellulose microfibrils synthesized in vitro by membrane extracts of suspension-cultured cells of Rubus fruticosus (Fairweather et al. 2004). The product of the (1,4)-βd-glucan synthase activity was microfibrillar in form, with a diffraction pattern typical of the cellulose II allomorph (Fairweather et al. 2004). The situation was even more complicated in monocotyledonous grasses, where in vitro synthesis of very low levels of β-glucans could be demonstrated, but where the products comprised a mixture of (1,3)-, (1,4)- and (1,3;1,4)-β-d-glucans (Smith & Stone 1973). Again, (1,4)-β-d-glucans were the major products at low substrate concentrations, and while the levels of (1,3;1,4)-β-d-glucan increased in the ethanol-insoluble products at higher concentrations, they were always much lower than (1,3)-β-d-glucans (Henry & Stone 1982). In the grasses the prime target was often the characterization of enzymes involved in (1,3;1,4)-β-d-glucan synthesis, but the results were usually disappointing (Becker et al. 1995; Vergara & Carpita 2001; Tsuchiya et al. 2005). The different linkage types of the (1,3;1,4)-β-glucans (Table 5.1) present a unique biosynthetic problem. In lichenin the molecules consist predominantly of blocks of 2–3 (1,4)-linked glucosyl residues joined by single (1→3)-linkages, whereas in cereal (1,3;1,4)-β-glucans in addition to blocks of 2–3 (1,4)-linked glucosyl residues some 10% of the molecule consists of blocks of up to 15 (1,4)-linked glucosyl residues joined by single (1,3)-linkages (Woodward et al. 1983b). Whether one or more enzymes are involved, and how the linkage sequence is specified, remains unanswered.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

121

In vitro, the linkage pattern is affected by the concentration of the UDPGlc substrate (Henry & Stone 1982; Buckeridge et al. 1999; Buckeridge et al. 2004) at concentrations in excess of in vivo levels. Several models have been proposed (Stone 1984), including one involving multiple UDPGlc binding sites (Buckeridge et al. 1999). However, Henry & Stone (1982) did show that the ratio of cellotriose to cellotetraose in the (1,3;1,4)-β-glucan product increased when the concentration of UDP-glucose in the reaction mixture increased from 10 μM to 1 mM. This served as a reminder that any enzymic mechanism proposed for (1,3;1,4)-β-glucan synthesis (Buckeridge et al. 2004) must account for the asymmetrical distribution of (1,3)-β-glucosyl residues in (1,3;1,4)-β-glucans. In bacteria, cell-free synthesis of cellulose by membrane preparations from Gluconacetobacter xylinus (formerly Acetobacter xylinus) was achieved by Aloni et al. (1982). Suitably stabilized membrane preparations of G. xylinus are capable of incorporating glucose from UDP-Glc into cellulose at a rate comparable with that found in living cells. Whether the polymerization of cellulose by G. xylinus is directly from UDP-Glc or via a lipid-oligosaccharide intermediate as implicated in Agrobacterium tumefaciens and embryophyte cellulose synthesis, remains undecided. In contrast the rapid loss of biosynthetic activity experienced in most of the biochemical studies on these enzymes from plants meant that it was not possible to purify them at that time. The (1,3)-β-glucan synthases did appear to be more robust than the cellulose synthases and the (1,3;1,4)-β-glucan synthases. Nevertheless, it was not possible to generate any amino acid sequence data for the (1,3)-, (1,4)- or (1,3;1,4)-β-d-glucan synthases and hence the genes and gene families encoding these GT2 and GT48 enzymes remained unidentified throughout the molecular biology era of the 1980s and early 1990s.

5.4 Functional genomics and the identification of GT2 cellulose synthases 5.4.1 Cellulose synthesis by embryophytes In the mid-1990s molecular biology and emerging functional genomics technologies were used to revisit the difficult question of cell wall polysaccharide biosynthesis in higher plants and, in particular, to identify the genes that control the process. It was well established at that time that cellulose biosynthesis in vascular plants was effected at the plasma membrane by rosette terminal complexes of proteins that contain catalytic cellulose synthase subunits (Mueller & Brown 1980). Pear et al. (1996) used transcript analyses to select and characterize highly expressed genes in cotton fibres at the onset of secondary wall synthesis, at a time when rapid biosynthesis of cellulose was occurring. They found highly expressed genes that were homologous to bacterial cellulose synthase (CelA) genes. Thus,

122

■

Plant Polysaccharides

computer-based database searches enabled them to tentatively identify a cellulose synthase gene (CesA) from cotton, for which the encoded protein was classified in the GT2 family of glycosyl transferases ((http://afmb.cnrsmrs.fr/CAZY/). Since then, mutational genetics, gene silencing and herbicide studies have provided strong evidence that the catalytic subunits of cellulose synthase enzymes in the rosettes are encoded by CesA genes (Pear et al. 1996; Arioli et al. 1998; Burton et al. 2000; Scheible et al. 2001; Burton et al. 2004). Genome sequencing programs and bioinformatical analyses of large EST databases have shown that plant CesA genes are members of multigene families. There are 12 CesA genes in rice (Oryza sativa; Richmond & Somerville 2000; http://cellwall.stanford.edu/), 10 in Arabidopsis, and at least nine in maize (Zea mays; Holland et al. 2000; Dhugga 2001). Mutations in individual Arabidopsis CesA genes are associated with reduced cellulose levels in certain cell walls (Arioli et al. 1998; Fagard et al. 2000; Scheible et al. 2001; Beeckman et al. 2002; Burn et al. 2002; Williamson et al. 2002; CanoDelgado et al. 2003; Gardiner et al. 2003; Taylor et al. 2003) and with resistance to herbicides that target cellulose biosynthesis (Delmer 1999; Scheible et al. 2001; Desprez et al. 2002). A major challenge in describing cellulose biosynthesis is to account for the enzymology of aggregate formation, the number of individual molecules in the aggregates and the synthesis of very long polysaccharide chains. The individual Arabidopsis CesA genes have evolved specialized functions, which require different genes for expression in different tissues, in primary or secondary wall synthesis, or as multiple components of the cellulose synthesizing rosettes. The availability of naturally occurring mutant lines and T-DNA insertional mutants of Arabidopsis has greatly assisted in the dissection of the functions of individual members of the CesA gene family. The genotypic and phenotypic characteristics of CesA and related mutants that apparently affect cellulose content of cell walls are summarized in Table 5.2. On current information, it appears likely that distinct CesA proteins are necessary for the correct assembly of rosettes in the plasma membrane of Arabidopsis cells (Doblin et al. 2002; Williamson et al. 2002; Taylor et al. 2003; Persson et al. 2007). In barley, coexpression of two groups of genes, namely HvCesA1, HvCesA2 and HvCesA6 in one group and HvCesA4, HvCesA7 and HvCesA8 in the other, is consistent with the participation of three CesA subunits in rosettes during cellulose synthesis and with the participation of distinct groups of CesA genes in primary and secondary wall assembly (Burton et al. 2004). These data, together with earlier microscopic evidence on the size of the rosettes (Fig. 5.3B), support the suggestion that each rosette is composed of a total of approximately 36 CesA molecules, arranged into 6 morphologically distinguishable subunits of the complete rosette (Doblin et al. 2002; Tanaka et al. 2003; Taylor et al. 2003). If such a complex exists, then one would expect cellulose microfibrils to consist of 36 individual cellulose chains, and this is close to the number of chains that can be estimated from the thickness of microfibrils observed by electron microscopy. However,

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

123

Table 5.2 Characteristics of mutant lines of arabidopsis in which cellulose synthesis is compromised Mutation

Gene name

Locus

Phenotype

rsw1

Cellulose synthase 1

At4g32410

rsw3 rsw10

Glucosidase II Ribose 5-phosphate isomerase Cellulose synthase 8

At5g63840 At1g71100 At4g18780

irx3, mur10

Cellulose synthase 7

At5g17420

irx5–4 irx8

Cellulose synthase 4 Galacturonosyl transferase 12

At5g44030 At5g54690

ixr1, eli1, cev1 ixr2, procuste

Cellulose synthase 3

At5g05170

Cellulose synthase 6

At5g64740

ixr9

Unknown

–

cob1 to 4

COBRA-like

At5g60920

kor, irx2, rsw2

Endo-1,4-βglucanase GH9

At5g4972

knf14

Glucosidase1 GH63

At1g67490

pnt1

Mannosyl transferase GT50

At5g22130

tbr1

Unknown

–

fra2, erh3, bot1, fr

Katanin small subunit p60

At1g80350

fra8

Glycosyl transferase GT47

At2g28110

kob1

Plasma membrane protein NAC transcription factor NST1

At3g08550

Radial swelling of various degrees of severity, depending on allele Radial swelling of root tips Radial swelling of roots and hypocotyls in seedlings Collapsed xylem, small dark green plant, prostrate stems Irregular xylem, more arabinose in wall, slower growing dark green plants Small plants with irregular xylem Collapsed xylem, uneven walls, reduced xylose, homozygote infertile Isoxaben resistant, enhanced resistance to pathogens Isoxaben resistant, radial swelling, irregular surface, incomplete walls Small plants with abnormal xylem Root radial swelling, bulging epidermal cells, microfibrils randomly oriented Dwarfed, irregular surface and xylem, radial swelling, abnormal flowers Embryo defective, radial swelling, elongated thin-walled cells Embryo defective with radial swelling, mature seeds peanut-shaped Trichome and leaf xylem birefringence reduced Abnormal orientation of microtubules, ectopic root hairs, abnormal trichomes Thin fibre wall, decreased stem strength, reduced xylan and cellulose Dwarf

NAC transcription factor SND1

At1g32770

irx1

nst1–1

nst3

At2g46770

Embryo defective, floppy inflorescence stems, secondary wall affected Floppy inflorescence stems, secondary wall affected

124

■

Plant Polysaccharides

given the variation in cellulose synthase morphology (Williamson et al. 2002) and observations that hexameric rosettes are not always present (Tsekos 1999), it is possible that some cellulose synthase complexes might not consist of 36 subunits arranged in hexameric rosettes. The development of sophisticated spinning disc confocal microscopy procedures has now allowed the cellulose synthase rosettes to be tracked in real time as they move through the plasma membrane. Selected CesA proteins tagged with a yellow fluorescent protein in transgenic Arabidopsis plants allowed the process of cellulose deposition to be observed in living cells (Paredez et al. 2006). The CesA complexes in the plasma membrane move in linear tracks that were aligned and were coincident with cortical microtubules. Within each track, the rosettes moved bidirectionally. These experiments indicated that the long-observed parallel orientation of cellulosic microfibrils in cell walls could be attributed to a guidance system through which the cellulose synthase complexes in the plasma membrane move between adjacent microtubules (Paredez et al. 2006; Wightman & Turner 2007; Fig. 5.3C). DeBolt et al. (2007) have shown that nonmotile cellulose synthase subunits repeatedly accumulate within localized regions at the plasma membrane in Arabidopsis hypocotyl cells following treatment with the herbicide 2,6-dichlorobenzonitrile. The advances in our knowledge of the genetics, enzymology and cell biology of cellulose synthesis in little over 10 years can justifiably be described as spectacular. The next 10 years are likely to see further advances in our understanding of the nature and composition of the cellulose synthase complexes and of the regulatory mechanisms, at both the gene and posttranslational modification levels, that control cellulose deposition in walls. 5.4.2 Cellulose synthesis by Gluconoacetobacter xylinus (formerly Acetobacter xylinus) Genes directly concerned with cellulose production in G. xylinus (Ross et al. 1991) have been identified in an operon (acsABCD/bcsABCD). The acsA/bcsA gene encodes the UDP-Glc-binding catalytic polypeptide, acsB/bcsB encodes a presumptive regulatory polypeptide that binds the activator c-di-GMP and the acsC/bcsC gene encodes a protein that forms a pore in the outer membrane through which the nascent cellulose molecule is secreted. Mutants in acsD/bcsD form reduced amounts of cellulose in vivo, but at similar levels to the wild type in vitro. In agitated cultures the mutants produce the cellulose II allomorph whereas wild type cells under the same conditions synthesize exclusively cellulose I. The acsD/bcsD gene may encode a polypeptide involved in microfibril crystallization, a step known to be rate limiting. Thus, the acsD/bcsD gene product is not essential for cellulose production but is required for maximum cellulose synthesis. Mutants of a fifth gene, not associated with the acsABCD/bcsABCD operon and encoding a proline-rich protein, do not produce cellulose. A gene for a secreted endocellulase appears to be associated only with cellulose-producing strains of G. xylinus.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

125

5.4.3 Cellulose synthesis by Agrobacterium tumefaciens Cellulose microfibrils produced by the plant pathogen A. tumefaciens appear to have a role in adhesion of the bacteria to root surfaces. Membrane preparations from the plant pathogen A. tumefaciens form cellulose from UDP-Glc but are not stimulated by c-di-GMP. Two operons, celABC and celDE, are required for cellulose production (Matthysse et al. 1995a; Matthysse et al. 1995b). The products of the celABC operon are active in membrane fractions, whereas the products of the celDE operon are active in soluble fractions. When combined, the membrane and soluble fractions incorporate glucose from UDP-Glc into cellulose. The celB, celD and celE gene products synthesize lipid-linked glucosyl intermediates, whereas the celA gene is believed to encode the cellulose synthase polypeptide. The membrane-associated protein encoded by celC shows homology with bacterial endocellulases but is proposed to act as an oligoglucantransferase using lipid-oligoglucosyl intermediates as substrates (Matthysse et al. 1995a, b). The notional steps in the assembly of cellulose molecules by A. tumefaciens as deduced by analysis of the products in mutants of the genes of the two operons, involve the formation of a glucosyl-lipid that is catalysed by the UDPGlc : lipid glucosyl transferase (CelE) or CelD. Next, cellulose synthase (CelA) adds additional glucosyl units to the glucosyl-lipid intermediate. In the following step, CelB catalyses the transfer of the oligoglucosyl chain from one lipid to another lipid. The final step involves elongation of the oligoglucosyl-lipid to form cellulose by an endocellulase (CelC) acting as a glycosyltransferase and/or the cellulose synthase (CelA).

5.5 Identification of the functions of other GT2 enzymes from plants Following the identification of the CesA gene family as the source of cellulose synthase enzymes, bioinformatic analyses of the CesA genes, EST databases and in the genome sequences of Arabidopsis and rice revealed that the CesA genes were in fact a subgroup within a larger, cellulose synthase gene superfamily (Richmond & Somerville 2000; Fig. 5.4). The other eight subgroups were referred to as cellulose synthase-like (Csl) genes and were assumed to encode polysaccharide synthases that mediated the synthesis of β-linked wall polysaccharides. All subgroups of the cellulose synthase gene superfamily encoded GT2 enzymes. Defining the functions of the Csl genes has proved to be somewhat more difficult than expected. It took several years before Dhugga et al. (2004) reported that the CslA subgroup encoded enzymes that synthesized (1,4)-βd-mannans. Dhugga et al. (2004) detected a relatively small number of CslA transcripts (15 from 15 000 ESTs) in developing guar seeds, at a stage when mannan synthesis in the seed was peaking. When the CslAs were transformed into soybean, the transgenic soybean plants showed

126

■

Plant Polysaccharides

greatly enhanced mannan synthase activity. The role of CslAs in mannan synthesis was confirmed in Arabidopsis, rice and the moss Physcomitrella, through heterologous expression of CslAs from these species in insect cells (Liepman et al. 2005; Liepman et al. 2007). The recombinant CslA proteins synthesized (1,4)-β-d-mannans when supplied with GDP-mannose, or (1,4)-β-d-glucomannans when supplied both GDP-mannose and GDPglucose; one of the CslA proteins synthesized (1,4)-β-d-glucan when supplied with GDP-glucose alone (Liepman et al. 2005). Thus, plant CslA genes are members of an extended multigene family and while many are involved in (1,4)-β-d-glucomannan synthesis, it is not yet known whether all CslA proteins are (1,4)-β-d-glucomannan synthases (Liepman et al. 2007). The CslF, CslH and CslJ groups of genes are the three so-called Poaceaespecific subfamilies of the large GT2 cellulose synthase-like gene family (Fig. 5.4). Members of the CslF and CslH gene families have been shown to be involved in the synthesis of (1,3;1,4)-β-d-glucans (Burton et al. 2006; Burton

67

69 69

0

68

5

1

6

12

9

2 OsCslF7 830 OsCslF3 868

602 A4 4 Csl 52 Os slA2 1 5 C Os lA3 5 9 49 Cs Os slA15 335 1 AtC CslA12 At 1 55 A 8 l s 44 AtC 3 lA 33 s 5 C 4 At lA9 53 Cs 2 527 At A l 1 s 9 C A 52 At Csl A1 0 0 l Os sCs 7 5 4 57 4 O slA 6 57 C CslA At CslA lA5 s Os sC O

OsCslF4 889

CslF (cereals)

OsCslF1 860 OsCslF2 889 OsCslH1 750

Sb Zm CslJ 1 C Ta slJ Cs 1 lJ1 Hv Cs lJ1 At C At slG Cs 2 At lG 73 Cs 1 0 lG 770 37 47

OsCslH2 762

CslG (dicots)

99 16 0 slE 1 73 AtC lE 45 Cs 7 Os lE2 Cs Os

CslJ (cereals) 0.1

81 5 11 36 slD 0 AtC slD1 1 1127 CslD 1 AtC sCslD 2 1170 O CslD 5 Os slD2 114 5 AtC lD3 114 7 AtCs CslD3 114 Os 1 1 1 1 4 AtCslD

59

C lC

CesA

4

9

lC

sl

Cs

lC

slC

sC Cs

8 69 lC2 5 Cs Os 3 74 lC 90 Cs 6 Os lC8 2 Cs 5 69 lC 3 Cs

Cs

sC

O At

At

At

O

At

At

CslC

CslH (cereals)

At AtCs Cs lB5 l At AtC AtC B2 7 806 Cs sl sl 5 7 lB B3 B1 4 75 712 748 5

CslB (dicots)

CslE

Figure 5.4 Phylogenetic tree for the family GT2 cellulose synthase gene superfamily from plants. The cellulose synthase (CesA) and cellulose synthase-like (Csl) groups of genes are shown. Modified from Richmond & Somerville (2000).

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

127

et al. 2008; Doblin et al. 2009), as outlined in detail in the following section, but it is not yet confirmed that all CslF or CslH genes encode (1,3;1,4)-β-dglucan synthases. It has recently been reported that the CslC gene family in Arabidopsis includes genes that encode polysaccharide synthases that are responsible for the synthesis of the (1,4)-β-d-glucan backbone of cell wall xyloglucans (Cocuron et al. 2007). These authors searched for genes that were expressed at relatively high levels during the last stages of nasturtium (Tropaeolum majus) seed development, when large amounts of xyloglucan are deposited as a storage polysaccharide. High levels of transcripts for a single member of the CslC subfamily detected and it was subsequently shown that the CslC gene was coexpressed with a xylosyl transferase gene. Heterologous expression of the CslC gene in Pichia pastoris resulted in the production of shortchain (1,4)-β-d-oligoglucosides, but not xyloglucan. On the basis of these results the authors concluded that the CslC gene family encodes proteins that synthesize the (1,4)-β-d-glucan backbone of xyloglucans. There have been a number of studies in which various mutants that carry lesions in other Csl genes have been examined. For example, there is evidence that members of the Arabidopsis and tobacco CslD gene subfamily are required for pollen tube formation and for normal root hair formation (Doblin et al. 2001; Bernal et al. 2007; Kim et al. 2007; Bernal et al. 2008 communication). This suggest that at least some members of the AtCslD subgroup are involved in the synthesis of a polymer that is required for tip growth (Bernal et al. 2008), but the nature of the polysaccharide is not yet known. In summary, there is some information available on the CslA, CslC, CslD and CslFgene subfamilies that encode GT2 enzymes, but direct evidence for involvement of the genes in the biosynthesis of clearly identified wall polysaccharides is not always strong. Little or no information has been published on the CslB, CslE, CslG or CslJ subgroups. Thus, it is probably too early to say with any confidence that the Csl and CesA clades of the cellulose synthase gene superfamily (Fig. 5.4) are ‘well-formed’ in the sense that all genes within a clade, or subfamily, are involved in the biosynthesis of a single class of wall polysaccharide. On the other hand, there is no evidence to date that this is not so.

5.6 Comparative genomics and the identification of GT2 (1,3;1,4)-β-D-glucan synthases Another biologically and commercially important group of GT2 enzymes that defied characterization until the emergence of functional genomics and associated technologies is the group containing the (1,3;1,4)-β-d-glucan synthases. These were expected to be GT2 enzymes and it was further anticipated that the corresponding genes would be found in one of the Csl

128

■

Plant Polysaccharides

subgroups, given the structural similarities between cellulose and (1,3;1,4)-βd-glucans. During the original compilation of the phylogenetic tree of the cellulose synthase gene superfamily (Richmond & Somerville 2000; Fig. 5.4), it was noted that two of the Csl subgroups, namely CslF and CslH, were found exclusively in the grasses. Given that the (1,3;1,4)-β-d-glucans are also distributed almost exclusively in the grasses (Trethewey et al. 2005), members of the CslF and CslH gene families immediately became candidate genes for (1,3;1,4)-β-d-glucan synthases (Hazen et al. 2002). Reverse genetic approaches were therefore adopted by a number of research groups, but progress towards proving the functions of the CslF and CslH genes was relatively slow. Using a combination of forward genetics and comparative genomics, Burton et al. (2006) showed that the CslF genes of rice mediate the synthesis of cell wall (1,3;1,4)-β-d-glucans. The forward genetic component had been undertaken previously by Han et al. (1995), who had explored natural variation in a barley mapping population to identify quantitative trait loci (QTLs) for the amount of (1,3;1,4)-β-d-glucan in ungerminated barley grain. The impetus for the analysis of QTLs for grain (1,3;1,4)-β-d-glucan content in barley had been provided by the importance of these polysaccharides in the malting and brewing processes (Bamforth 1993). Although it would have been possible at that stage to pursue the forward genetic approach further through fine mapping and eventually map-based cloning of the genes under the QTLs, Burton et al. (2006) chose to adopt a comparative genomic approach to short-cut the identification of the (1,3;1,4)-β-d-glucan synthase genes. One major QTL was located in a large chromosomal interval (more than 10 cM) on barley chromosome 2H (Han et al. 1995), but in the absence of a genome sequence for barley it was not possible to identify the genes under the QTL that might be involved in (1,3;1,4)-β-d-glucan synthesis. However, the common evolutionary origin of the cereals is reflected in a well-established property known as synteny (Gale & Devos 1998; Kota et al. 2007), which describes the conservation of gene order along corresponding sections of chromosomes in different cereal species. Syntenous regions in different genomes can be fragmented or located on different chromosomes, and in some instances the colinearity breaks down at the ‘micro’ level (Pourkheirandish et al. 2007). Nevertheless, it was possible to identify the region of the rice genome that was syntenous to the region of the barley chromosome 2H where the QTL for grain (1,3;1,4)-β-d-glucan was located, using the sequences of DNA molecular markers that flank the barley QTL. The sequences corresponding to those barley markers were found on rice chromosome 7, where they defined a region of about 3 Mb (Burton et al. 2006). The availability of the annotated rice genome sequence allowed this region of rice chromosome 7 to be scanned for candidate genes for (1,3;1,4)-βd-glucan synthesis. A cluster of six OsCslF genes and two truncated OsCslF pseudogenes was detected in a region of about 120 kb near one end of the 3 Mb section on rice chromosome 7 (Fig. 5.5). The observation that this gene

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

OsCslF4 OsCslF3 OsCslF cluster

129

OsCslF9

OsCslF2

OsCslF8

OsCslF1

22.00Mb

Rice 30Mb chr 7 (partial)

■

21.90Mb 25Mb

20Mb 6.0

LOD

Barley chr 2H

2.4 8.60

77.40 Adh8

155 cH

ABG019

Figure 5.5 The CslF family was identified as the prime candidate for genes encoding (1,3;1,4)-β-D-glucan synthases. A major QTL for (1,3;1,4)-β-D-glucan content of ungerminated barley grains was identified on barley chromosome 2H by Han et al. (1995) and the ‘likelihood of differences’ (LOD) scores shown here were derived from that work. Markers flanking the estimated position of the barley chromosome 2H QTL, (Adh8, ABG019 and Bmy2), were used to identify a syntenic region of about 3.5 Mb on rice chromosome 7. Examination of the rice genome sequence in this region revealed a group of six OsCslF genes, close to the Bmy2 marker; the six genes were clustered within an interval of about 118 kb. From Burton et al. (2006), reprinted with permission from AAAS.

cluster in rice was located in a position syntenous to the grain (1,3;1,4)-β-dglucan QTL on barley chromosome 2H, coupled with the fact that the CslF subgroup is a so-called monocot-specific group in the cellulose synthase gene superfamily (Richmond & Somerville 2000), immediately increased our confidence that these genes might indeed mediate (1,3;1,4)-β-d-glucan synthesis in cereals and grasses. Forward genetics and comparative genomics had delivered strong candidate genes, but it remained necessary to prove the functions of the CslF genes in (1,3;1,4)-β-d-glucan synthesis. The rice OsCslF genes were therefore inserted into Arabidopsis, which normally does not contain CslF genes and does not have (1,3;1,4)-β-d-glucans in its walls. (1,3;1,4)-β-d-Glucan was subsequently detected in walls of the transgenic plants, using specific monoclonal antibodies and enzymatic analyses (Burton et al. 2006). These experiments provided direct, gain-of-function evidence for the participation of rice OsCslF genes in (1,3;1,4)-β-d-glucan biosynthesis. However, Burton et al. (2006) pointed out that the observations did not preclude a requirement for other enzymes, proteins or cofactors in (1,3;1,4)-β-d-glucan synthesis. The

130

■

Plant Polysaccharides

relatively low levels of (1,3;1,4)-β-d-glucan in walls of the transgenic Arabidopsis plants, where OsCslF transcript levels were often high, was consistent with limiting levels of other components that might be required for high-level synthesis of the polysaccharide or its transfer to the cell wall (Burton et al. 2006). While there are two ‘monocot-specific’ subgroups (CslF and CslH) in the cellulose synthase gene superfamily, there is only type of polysaccharide, namely (1,3;1,4)-β-d-glucan, that is found almost exclusively in the commelinoid monocotyledons (Trethewey et al. 2005) and whether the CslH and CslF proteins might act in concert or mediate the synthesis of (1,3;1,4)-β-d-glucans with different fine structures remains to be demonstrated. Whether the CslH and CslF proteins might act in concert or mediate the synthesis of (1,3;1,4)-β-d-glucans with different fine structures remains to be demonstrated. It has been suggested that the biosynthesis of (1,3;1,4)-βd-glucans might require the concerted action of multiple enzymes that form large multimeric complexes (Buckeridge et al. 1999; Buckeridge et al. 2004), as observed for cellulose biosynthesis (Doblin et al. 2002). In describing the enzymic apparatus for (1,3;1,4)-β-d-glucan synthesis, one must also ask how the fine structure of the polysaccharide is effected, in particular with respect to the insertion of (1,3)-β-d-glucopyranosyl residues at irregular intervals along the backbone chain, and hence to the arrangement of the cellotriose, cellotetraose and longer (1,4)-β-d-oligoglucoside blocks described above.

5.7 Genes for GT2 synthases for bacterial (1,3)-β-Dglucans and related polysaccharides 5.7.1 Curdlan Curdlan is found as a protective (1,3)-β-d-glucan capsule around Agrobacterium and related rhizobia and the Gram-positive Cellulomonas spp. (Table 5.1). Four genes (crdA, S, C and R) that are essential for curdlan production were identified in Agrobacterium sp. ATCC31749 (Stasinopoulos et al. 1999) and a fifth gene encoding a phosphatidylserine synthase (pssAG) is required for maximal yields (Karnezis et al. 2002). Curdlan production also depends on the global nitrogen metabolism genes, ntrBC, and on several as yet uncharacterized genes (S. Aracic, A. Anguillesi, V. Stanisich, personal communication). The crdASC genes have an operon-like organization, while the remaining genes (pssAG, crdR and ntrBC) occur at separate loci that are not linked to the crdASC cluster. The CrdS gene shares no homology with the (1,3)- β-d-glucan synthase-related FSK1 and FSK2 genes from yeasts and fungi (Dijkgraaf et al. 2002) or with the plant callose synthase-related proteins (Li et al. 2003; Brownfield et al. 2007), which are classified as GT48 glycosyltransferases. In contrast to crdS, crdA (1539 bp) and crdC (1269 bp) encode proteins that have no counterparts in the gene/protein databases.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

131

The CrdA protein (molecular mass 48 kDa) is predicted to be membraneanchored with a large periplasmic C-terminal portion, while the CrdC protein (molecular mass 42 kDa) is predicted to be periplasmic, because it carries a cleavable signal sequence. 5.7.2 Cyclic (1,3;1,6)-β-d-glucan from Bradyrhizobium japonicum The cyclic (1,3,1,6)-β-d-glucan (Table 5.1) of Bradyrhizobium japonicum is produced by a 90-kDa inner membrane protein that uses UDP-Glc as the monosaccharide donor (Bhagwat & Keister 1992; Bhagwat et al. 1992; Inon de Iannino & Ugalde 1993). The cyclic (1,3,1,6)-β-d-glucan synthase gene (NdvB;blr4614) encodes a family GT2 enzyme (Bhagwat et al. 1992; Bhagwat & Keister 1995). Site-directed mutagenesis of NdvC gene (Bhagwat et al. 1993; Bhagwat et al. 1999) gave a mutant that was defective in motility, grew slowly at low osmolality, was ineffective in nodulation and produced a cyclic β-d-glucan lacking (1,6)-linked residues. A third gene, NdvD, is also involved in cyclic (1,3,1,6)-β-d-glucan synthesis (Chen et al. 2002). The NdvC and D proteins are not in the GT2 family (http://www.cazy.org/). 5.7.3 Capsular polysaccharides from Pasteurella multocida Pasteurella multocida is a pathogenic Gram-negative bacterium that produces capsular polysaccharides that resemble host connective tissue polysaccharides and serve as molecular camouflages to evade host defences. These include hylauronan produced by P. multocida type A that has a (1,3)-linked repeating unit of GlcApβ(1,4)GlcpNAcβ. Hyaluronic acid synthase activity has been obtained from cell-free membrane preparations of P. multocida. The enzyme utilizes UDP-sugar precursors (DeAngelis 1996). The synthase gene was cloned by DeAngelis et al. (1998) and encodes a 972-kDa bifunctional family GT2 enzyme. Two transferase activities have been dissected and two active sites exist in one polypeptide (Jing & DeAngelis 2000). A second capsular polysaccharide, heparosan, produced by P.multocida types A and D is synthesized by a 965-amino-acid residue family GT2 enzyme (DeAngelis & White 2001; DeAngelis & White 2004) and a third capsular polysaccharide, chondroitin, found on P. multocida type F is also synthesized by a GT2 enzyme (DeAngelis & Padgett-McCue 2000). 5.7.4 Chitin The first chitin synthase genes were cloned and sequenced from yeast and filamentous fungi (Bulawa et al. 1986; Silverman et al. 1988; Au-Young & Robbins 1990; Yarden & Yanofsky 1991). In Saccharomyces cerevisiae three chitin synthases have been described (Chs1p, Chs2p and Chs3p): the Chs1p enzyme repairs damaged chitin during cell separation, Chs2p is involved in the formation of the primary septum disc and Chs3p is responsible for the

132

■

Plant Polysaccharides

synthesis of the chitin ring laid down at the base of the emerging bud (Nagahashi et al. 1995). Later, cDNAs for insect (Tellam et al. 2000) and nematode chitin synthases (Veronico et al. 2001) were cloned. The genes encode proteins of the GT2 family and their evolutionary relationship with those from higher plants and other GT2 glycosyl transferases has been examined by Merzendorfer (2006) and is shown in Fig. 5.6.

CER-DmNP_651407 CHS-ScP29465 74

CHS-EdAAC35278 52

CHS-CeNP_493682 87 CHS-DmAAN13280

99

NodC-SmAAB95329 52

HAS-HsNP_001514 92

HAS-DrNP_775327

58

HAS-GgAAF14347 81 HAS-XIO57427 CES-ZmAAF89961

69

CES-DdAAF00200 95

CES-XaNP_643825 51

0.1

CES-CsAAR89623

Figure 5.6 Evolutionary relationship between members of processive family GT2 glycosyltransferases. Merzendorfer (2006) calculated the most likely phylogenetic tree with Tree-Puzzle (5000 puzzling steps, log L−6753.99; Schmidt et al. 2002). Percentage supports for clades using the maximum likelihood method (quartet puzzling support values) are shown beside each internal branch. The scale bar represents 0.1 substitutions per site. The tree is rooted with a Drosophila glucosylceramide synthase as an outgroup, which belongs to family 21 glycosyltransferases and is distantly related to family 2 glycosyltransferases. Glycosyltransferases are abbreviated as follows: CER glucosylceramide synthase; CES cellulose synthase; CHS chitin synthase; HAS hyaluronan synthase; NodC Rhizobium chito-oligomer synthase; Species names are abbreviated as follows: Ce, Caenorhabditis elegans; Cs, Ciona savignyi; Dd, Dictyostelium discoideum; Dm, Drosophila melanogaster; Dr, Danio rerio; Ed, Exophiala dermatitidis; Gg, Gallus gallus; Hs, Homo sapiens; Sc, Saccharomyces cerevisae; Sm, Sinorhizobium meliloti; Xa, Xanthomonas axonopodis; Xl, Xenopus laevis; Zm, Zea mays; Genbank/EMBL accession numbers follow. From Merzendorfer (2006).

133

■

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

5.8 Enzymic properties and catalytic mechanisms of the GT2 proteins 5.8.1 Topology The family GT2 proteins encoded by the CesA and CslF genes are both predicted to be integral, membrane-bound proteins with a total of eight transmembrane helices (TMHs) (Fig. 5.7), two of which are located towards the N-terminus while six are located towards the C-terminus (Doblin et al. 2002; Burton et al. 2008). The coding region of the cotton GhCesA1 gene encodes a 974-amino-acid polypeptide of approximately 110 kDa (Doblin et al. 2002), while the seven HvCslF genes encode glycosyl transferases with 810–947 amino acid residues (Burton et al. 2008). Experimental topological analysis of the Streptococcus pyogenes hyaluronan synmthase (Heldermon et al. 2001) and GT2 curdlan synthase (Karnezis et al. 2003) shows the disposition of TMHs and that the catalytic region is situated in a loop on the cytoplasmic face of the inner bacterial membrane. In the curdlan synthase a 42-aminoacid C-terminal domain lies in the cytoplasm. 5.8.2 The catalytic region More generally, a central cytoplasmic domain in both classes of the GT2 enzymes is likely to include the UDP-glucose substrate-binding site and it

HvCesA2 (GT2)

HvCslF9 (GT2)

Extracellular

HvGSL1 (GT48)

Extracellular

Extracellular

879

303

890

944

977

1033

K

L

T

G A R

K

E

F

R

S

L

V G G V G R L

A A P S A M N A D P L L A A V

P V W Y K D A D E R E I

272

861

321

908

927

L L P Y R T

1015 1061

A F

K L

A G A S

R G D Q E R V K G F

A

G D V V V

Q

I

E

L

L I

V R V VV

I L T R

N

L

L T

F W M

PV Y Y

M

D

G T

II

M

V V

V

A

Q E N

A

G

V M

A

S

F

S

Y

Q

A

I

F

V

R

L

P

M L

V

Y

I

A

R

V Y L L K K E K T L S

T

E

A

F

I

E

F

N

P

N

A

W

G

V

S

P

C

R

I

L

S

V

C

A

T

E P D K S P Q

E

D

Q

L

L

M

N

V

D

Q

Q

L

P

I

D

A

G

P

G

H

S

H

D

G

L

K

P

R

L

H G

L

K

D

A

V

P

F

K

I

I

N

W

T

T

V

G

D

Q

P V

V

F

Q

G

L

T

R

V

V

V

S

D

A

E

L

T

L

A

S

I

N

N

A

Y

P

Y

T

I N

G C G E

Y

D T A F

L

R

L

A

V

K

H

F

Y

E

E

E

G

Y

L

A

L

A

L

R M

T

R

A

R

E

H

F

D

A

D

A

V P F C R K

H

C

V

E

P

R

A

P

E

S

Y

F

W A

K

M R

G

E

Y

V R Q

S

R

A

A

R

P

S

S

A

A

S

L

A S E K I

D

T

Y

L L S

T E

Y

K

R

N

I V

W P

R A R

S K V F L A K D

A V

D V

T D

C N H N M F F

T L

L N G I Q G P S

N D S

T

G

C

M F

R

R

V

A

L

Y

S

A

D

P

P

R

L

L D M A G

A

L A D T H

F

E E F Y N D Q N M D I T Q

P

V

L

E

V

A

E

R

H

A

H

Q

V

A

V

D

E

L

Y

K

L

F

D

T

Q

Y

I

V

A

V

N

T

V

C

A

Q

G

I

K

Y

K

G V Q

R I

K D E Y M Q Y

E C

A

K

S

G

E

E

T

G

E

K

K

P

L

E

I

A

G

A D

F H S I K Y I

V

L

A

S N I

V L K A

L

I

D K E H

L W V Q R I

F D G I

Q E S I

S K N N I

Q S D I

S H F

A N G R T F I V A H N Q N

A

E

H S

A E

D D S

857 amino acid residues

K

A

M

L

S

W

M

S

D

E

W

K

G

R

E

V

D

L

R V

N

L

A

Q

T

A

F

L

D

S

S

K

G

E

Y

F

A

L

L

S

S

N

E

M

R

A

L F E K A

E

H

N I Q W D D I G

F

N P

D

S F E

S

K G

R

Q

L

L

N

S

Y

N

Y

D

I F

R

S

V

A

S

P

H

C

A

M

D

G

Q

D

P

D

K

L

M

A

A

M

R

G

L

F

L

E

E

L

K

A

S

R

F

P

K

N

L

L

G L I S

L

F

Q

F

Y

D

E

V S G T

V H E

V

M M

M

P

E

S

R

G

I

G

A

L S

T

K

L

A

A R V R

I

S

Y R G Q T L

L R

F

L W

H

L

I

F

M

N

E

E Q

V

T

T

R

P

A

S

N

K

I

N

P

F

N

E

V

N

P

V

N D I L E L

E

K

V

L

F S

L

A

F W N

G

T

L

G

1070 amino acid residues

L

Q

E

A

A

S

F

Q

T

N

L

F

N

L E

P

E

T

A

S

W

L D A S R F

R

I I

L

S

L

Q

P

R

K

D

N

F

K

R

F

L

G

V

E N A A

S

M

P

S

Q

G

S

R

E L N K

M

S

V

S

R

P

R

A

R I N

A

G

A

L

F

D

M

D

C

V

G

R

T

V

H

R

M

K

D

F

H

F

Y

P

W

R

F

I

D

G

V

P

P

I

H

R

H

A

P

D

F

A

I

R

L

L

L

F

T

S S G Q

G

M

G

H

E W K N F L

H

G

R

L

S

S

A

E

D

G

R W A N N H D N G R A A

L Q K I

D

N

G

G

Y N

G

S T L F Y

Y

A

A

S

D

A

A E L Q K K N E D G I

P

L

A

A

D

E

V Y

A

D

F

M

N

V

S

A

T V L Y S I Y C S E

P T

C

A

K

I

F

A

L

Y

C A E S R Q

V

L

W

E

V H V A G E

L

M

T

K I L S G N E P

D

D

T

A

P

GK

P

V

Q

S

M D K L

G

Y

Y

V

L

D E

T

K

L

D

N

M

Q

S

D

I

S

A

V

E

Q

E

I

R

P

M

K

Y

N

F

Y

L

R

L

G

A

D

S

K

P

L

Y

I

F

P

Y

R

H

T

S

V

I

S

E

M

R

A

V

W

V

K

W

E

G

V

Q

G

R F D Q K V

S E A W W

D E

G

F

V

D

L

C

S T F Q T R L L F N Q A F S R G G A L I L S I E L N Q N N A G I W H H P S F E S S P R C

I

H

L

Q

I

PI S F

F

R

K

F

F

M G

F

A I

I

I

F

S

C

W

G MI V

W

I

P

V

K

E

Y S

V D

N

E

G

E

N

D

F

I G T

A I

I

V

D

S

W T W

P G

A L

S

F

N

W

Q

T

P N S L K W P

P

W

L

N S

P V

S

L

F

F

F

A

I L

WG

K

Y

F

Y

A

S

L

V V

E

N

E

G

G

D

C

N

G

F

I

W

F

I

F

N

V

L A

A

G E E

G

V

S

A

A

V

AG V A A I

S K V

R

Y

T

K

L

N

F

L

Cytoplasmic

I

L

E

D G

A

A

L L G

R L G E I R S

V H R F

Y

H

S W

L A

Y D

W

I A

V

L V

T R

E

L

P K

S LL L

I

S

R

I

R L V K V

S

V

L

F

R

K

F

N L

VL L G

V A W S WV V L L A T

V

R L

P

LG

Y

M

P

A

H F Q

K

M L F

LA F N

L L S

A

A

R

V M A

S

V

I

S G A G Y

L F

F

N

V

V K

R

K T L W

L

F

G

Q

D

G

H

L

G

E

D

I

S

E

Y G

I I I

W

N

A

N P K S

R

V1624 A L

LL F

N

S

I

V

R

S

F A V

E G

L

T

E

K

D

D

L

E

H

G

I

Q

R

S

S

G

E

T

L

L

E

L

I

I

V D

M

S

S

G

P

N

R

W

E

A

N

F

N

T

H

I

S

N

H

I N F

Q

M

Y

V

W

D

R

G

S F

T L

L

Y

F

Q M

S

H

L

H

Q

L

K

V

L

E

H

A L V

S

E

E

H

C Y I F H H V A R E L E V I M Q K Q

S

T I

I

L

H

V VR

T G

VF C L

FQ

P I

G F

E

S

H R G L

K

N

G

R

G

M

E

V

S

I

Y

T I

R

Y

F

R

T F

Q VF

I

Y

L P

N

S

S

W

T

E H

F

V

R

G

V S W

V

N

S

A

R F

Y S

M

R

G

H

Q F

I

T A

Y

A A T F R I

Y T

LF

N

R E

G K

F

L T

G

Y

R G

N F

Q

N

D

D

M

L

C

S

A

K

H

A

S V L

Q

Q K

R

Y

K

G L

G

R

M

Y

V

S

Q

E

C

P

C

P

Y

S

H

D

E K R

P G

M

R

T

C

A

I

S

L

A

D

N V R F L P E A A GL

Y

F

V

F

Y

R

A

F

L

G

S

G

V

E

T

F

E

I

A

D

H

W

P

L

R

G

I

P

A

T

L

T

H

N

F

P

W

D

T

L

R

L

A

I

H

L

L

S

G

L

P

N

S

Y

N

N

G V

H

T

Q

D

V

I

I

F

R

D

A

A

V

Y

L

A

T

N E

L

K G

K

L I

G

A L

A

T

TV V T

T C F Y F G V T T V Y F

G G

H

I

I

Y

ML

L L D I H

I

Q

D

S

L

L

AP I

Y

W A K

H

A

S

S

F A M KW

A

V

W

C C

I

P

R

I S L L M S V P

I

S

D

G

T

S

A

Q

A

I Y

I T

L S L

S

A

V

E

T S R

G

V A

L

R

R

Y

Y

I

W

F

Q

I

V

Q

G

T F K

I V L

Y

V

W

I

L

T K

V C

S

D

L

L

L

A Y S

N

L

R

K

S

L

L

W

P

K

A

A

F

D

T

S

R

I

MM

H

L

L

G

K

G

S V L D I L

R

A A

Y

A

LQ Y

SF

I

L

N I

F

S

AG

H

L

S

L

A

F Q

G

H

P

V V

VI

R

V

S L L H

L

T F P E

Y

Y

F W C F WR I V R IT A

S

N

L R

L

TA

V

D F V

N

V

F K

Y

VI S A

Y C L

K

K

P

I

K I

S

P

A

A

P I F

I

G

H

T S

L

L S

T I

M E

M

G

G

996

Y A

Y

E

R

S

Cytoplasmic

V A T

A G LY

Q

Q

L

LG I

G V Q

D

V

L

N F I

L

L

W L W W

S V V G D L W

I V

A S V

V

K

T N

T

A

T

E

N

G Q F M L L F M W

E

W

I L R N

T D F S

LT

L

I LD

V L

VT

V L G A A

G A

A

S L

A

S

I

P

G

V A

YN

I L

L

A

LY L L L M W V N F

GA

F

M I L

F P

V N V A A I

G Y

W

G

Q W

L L V PT

A M

P Y T

E

R

E

S D A

ML

V A

G V

VP

H

I

F

M

V I L

I E

I

G

V

R L

L

A L

T V

F V

F

H R

I

Cytoplasmic

FV V

M

F T S

R

F M

Y F

L

I

F

S

T

Q F W

H

D

L

Q

G

Y R L

C

F

P Y

W

Y

G

F

S

L

I

P

F

K R

E

K

R

Y

Q S S G G G F I N

M N R A A E A N W E R L L R A A L R G D R M G G V Y G V I G S A P D I H T N N G L S S P V N G A V L E R A A D E I Q D E D P T V A R I L C E H A Y A N Q A L T K F Q L V G R G E S N P D L M S LV IG R Q K L A K R E G G A I D R S R D I A K L K Y F E Q E D E C L E D V K H K E R Y L K L M R E S G V F S G N L G E L E R K T L K R K K V K L T A L P S I E K T I D E I V S W L V D A E A N L I S E K M K E F M E K D A A R T E D

G

L

1042

Q N R

I K G L Q

F S I

K

A T

290

E P K G E G L K P N G P L K V S

I

Y

A

I

L

E

L

K

M

D

Q

V

R

N

N

P

E

L

A

K

L

K G H I D R I

A Y I

D V V E S I

K L V K K N G K S S T E Y Y S

A

1015 amino acid residues

Figure 5.7 Topologies of GT2 and GT48 glycosyl transferase enzymes. The topologies of HvCesA2 cellulose synthase (1070 amino acid residues), HvCslF9 (1,3;1,4)-β-D-glucan synthase (857 amino acid residues) and HvGSL1 (1,3)-β-D-glucan synthase (1015 amino acid residues) were predicted with the HMMTOP algorithm (Tusnády & Simon 2001). The graphics was generated with TMRpres2D software (Spyropoulos et al. 2004).

134

■

Plant Polysaccharides

is generally assumed that the nascent polysaccharide chain is extruded through the membrane, either directly into the cell wall from the plasma membrane or into the Golgi for subsequent transport to the wall. The TMHs of the protein are likely to align in such a way as to form a pore through the membrane, so that the glucosyl residues can be provided by UDP-Glc on the cytoplasmic side of the membrane and the newly synthesized polysaccharide chain is progressively extruded through to the other side of the membrane. The amino acid sequence motif DDDQXXRW, which is characteristic of family GT2 enzymes (Saxena et al. 1995a; Doblin et al. 2002; Charnock et al. 2001), is evident in the cytoplasmic regions of the enzymes. The amino acid sequences of family GT2 glycosyl transferases have been subjected to hydrophobic cluster analyses (HCA), which reveal common structural features, including two distinct regions that have been named domains A and B (Saxena et al. 1995). The DDDQXXRW motif extends over both domains and may be involved in the catalytic process, metal ion binding and repetitive action. Table 5.3 shows the motifs common to the putative catalytic site sequences of representative family GT2 β-glycan synthases. 5.8.3 3D structures The first published 3D structure for a GT2 protein was that of the catalytic domain of the spore coat polysaccharide biosynthesis protein, SpsA, from Bacillus subtilis (Charnock & Davies 1999; Tarbouriech et al. 2001). The SpsA is an α/β protein that adopts a GT-A fold. The high-resolution (1.5 Å) 3D structure provides information on how nucleotide diphosphates might be bound by interacting loops in the active site pocket, via Mn2+ or Mg2+ ions. The SpsA structure points to the identity of active site amino acid residues in the pocket and to the likely catalytic mechanism within the GT2 group of glycosyl transferases (Tarbouriech et al. 2001). Although the specificity of the enzyme is unknown, the crystal structures of SpsA are available both with and without bound UDP-Mn2+ or UDP-Mg2+, and it is assumed that a uridine nucleotide-monosaccharide is the substrate. The active site structure of SpsA synthase serves as a prototype for the organization of other family GT2 synthases. The 256-amino-acid protein has two domains, a nucleotide-binding domain and an acceptor-binding domain, and features a disordered loop spanning the active site. Both UDP-Mn2+ or UDP-Mg2+ bind in a deep cleft situated in the N-terminal domain (Fig. 5.8A). The hydrogen and ionic interactions of the enzyme with UDP-Mn2+ at the binding site are shown in Fig. 5.8B and Fig. 5.8C. The sequence of the family GT2 bacterial (1,3)-β-d-glucan synthase (CrdS) (Stasinopoulos et al. 1999), which contains the extended DDD35QXXRW motif, has been threaded on to the SpsA structure and shows a similar folding pattern in the UDP-Glc-binding site of the catalytic region of SpsA (Karnezis et al. 2003).

TVDIFIPTYDX16DWPPDKVNVYILDDG VVDVYVPSYNX16DWPADKLNVYILDDG TVDIFVPTYNX16DWPPEKVRVHILDDG TVDVFVPSYNX16DYPADRFTVWLLDDG - - - - - - - - - - -DYPVDKVSCYISDDG - - - - - - - - - - -DYPVEKISCYVSDDG - - - - - - - - - - -DYPVDKVACYVSDDG LVDVFICTYNX16DYP– RLRVFVCDDN - - - - - - - - - - -DYPSENLWIGLLDDS - - - - - - - - - - -DYPG-ELRVYVVDDG - - - - - - - - - - -HYP– FEVIVTDDG KVAAVIPSYNX16TYP– LSEIYIVDDG – – – KVSVIMTSYN- - - - - - - - - - -FIMDDN

Axy AcsA Axy AcsAII Axy BscA Atu CelA Ghi CelA1 Ath Irx3 Ath Rsw1 Asp CrdS Ddi DcsA Rme NodC Pmu HasA Spy HasA PCV HasA Spn Cap3B Sce Chs2 Bsu SpsA

YIGRVDS – SHAKAGNLN YIIRDQN – NHAKAGNLN YIARPTN – EHAKAGNLN YLTRERN – VHAKAGNLN YVSREKRPGYQHHKKAGAEN YVSREKRPGFQHHKKAGAMN YVSREKRPGFQHHKKAGAMN YVTRPDN – KHAKAGNLN YLRRRKPP – IPHNKAGNIN – – – – – – –

1

KAG LMQTPHHFYSP –D LMQTPHHFYSP –D LLQTPHHFYSP –D LVQTPHFFVNP –D YVQFPQRFDGI –D LDQFPKWFPIE –D LDQFPKWYPIN –D VVQTPQFYFNS –D FVQTPQFFSNIYPVD – – – – – – –

QTP 2

3

DX46DCDX94DX35QRVRW DX46DCDX94DX35QRVRW DX46DCDX94DX35QRMRW DX65DADX94DX35QRSRW DX165DCDX210DX37QVLRW DX165DCDX199DX37QVLRW DX165DCDC216DX37QVLRW DX52DADX94DX35QRTRW DX90DADX255DX35QRKRW DX51DSDX98DX35QQLRW – DX46DADX99DX35QQNRW DX51DSDX138DX35QQTRW DX43DSDX99DX35QQLRW DX103DMDX102DX38QRRRW –

1

D1 D2 D335 QXXRW

Axy, Acetobacter xylinus; Atu, Agrobacterium tumefaciens; Ghi, Gossypium hirsutum; Ath, Arabidopsis thaliana; Asp, Agrobacterium sp.; Ddi, Dictyostelium discoidium; Rme, Rhizobium meliloti; Pmu, Pasturella multocida; Spy, Streptococcus pyogenes; PCV, Chlorella virus; Spn, Streptococcus pneumoniae; Sce, Saccharomyces cerevisiae; Bsu, Bacillus subtilis. SwissProt accession numbers: AcsA (P19449), AcsII (Q59167), BscA (P21877), CelA (Q44418), CelA1 (P93155), Irx3 (AF088917), Rsw1 (AAC39334), CrdS (A5189370), DcsA (A500200), NodC (P0341), HasA (Pmu) (O68389), HasA (Spy) Q54865, HasA (PCV) Q84419, Cap3B (P72520), Chs2 (P14180), SpsA (P39621).

UDP-Glc

Protein

Motif

Table 5.3 Motifs common to the putative catalytic site sequences of representative family GT2 β-glycan synthases

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families ■

135

B

A

D191 Mg2+

D39

UDP

Mg2+

D158

D99

Arg 71

C

NH2

HN

NH2

Asp [9]

O

3.0A 2.5A

O

2.0A

NH

2.7A

HO OH 2.3A

O

OH O

2.5A

O

O

Mg

O

2+

O 2.7A

P

O

O2.1A

Mn2+ 2.6A 2.2A 5.0A O

O 2.0A

NH+3

O 99

OH

O

O

P O 2.1A

2.0A

Asp 158

N

O

YATDDN Asp 99 YLTDDT LDDDDD RMDADD NVDSDT ILDCDH

HO 2.0A

OH

3.1A 2.9A 3.2A

O Lys 13 OH

39 FINDDN FINDDN O IIVDDG Asp 39 IIINDG YVVDDG YILDDG

11 KVSVIMTSYN KVSIILTSYN LVSVVLPTYN KISVVMSVYN SVDVIVPCYN Thr 11 TVDIFIPTYD

O

O Asp 98

Thr 9

Figure 5.8 Structure of a GT2 glycosyl transferase SpsA from Bacillus subtilis. (A) Surface representation of SpsA (PDB accession code 1QGS) showing an (α/β)8 fold with secondary structure elements (Charnock & Davies 1999). Four conservative aspartic acid residues (D39, D99, D158 and D191), depicted as red patches on the surface of the SpsA, represent the key structural elements and with two Mg2+ ions (grey spheres) they collectively coordinate the position of the UDP molecule (orange/red sticks) in the active site pocket. (B) A view of the active site depression, where the four aspartic amino acids shown in cpk sticks bind the UDP molecule (orange/red sticks). The black dashed lines indicate interactions that are critical for UDP binding with the four aspartic acid residues and Mg2+ ions. Panels A and B were prepared with PyMol (http://pymol.sourceforge.net/). (C) Hydrogen bonding and ionic interactions between amino acid residues of the active site of the GT2 protein SpsA from B. subtilis and the substrate analogue UDP-Mn. Amino acid sequences adjacent to the residues involved in binding are also shown and correspond to the sequences from SpsA from B. subtilis, CgeD from B. subtilis, 334-amino acid hypothetical protein from Pyrococcus horikoshii, glycosyltransferase involved in O-antigen synthesis from Vibrio cholerea, root nodulation factor NodC from Rhizobium sp., and cellulose synthase from Acetobacter xylinus (A subunit), respectively (Charnock & Davies 1999.)

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

137

A second 3D structure of a putative glycosyl transferase from Bacteroides fragilis belonging to the GT2 group of enzymes was recently deposited in the Protein Data Bank (PDB accession code 3BCV; K. Palani, D. Kumaran, S.K. Burley, S. Swaminathan, New York Structural GenomiX Research Consortium). This structure shows a spatial conservation of aspartic acid residues D42 and D94, which correspond to D39 and D99 in the SpsA structure. These acidic residues most likely participate in coordination of sugar nucleotides in the active site pocket of the glycosyl transferase from Bacteroides fragilis (Fig. 5.8B). 5.8.4 Catalytic mechanisms The enzymic synthesis of glycosidic linkages from either NDP or polyprenol substrates is likely to involve general acid–base catalysis, as is the case for glycosidase-catalysed cleavage and glycosyl transfer reactions. The GT2 glycosyl transferases have an inverting reaction mechanism, through which the α-linked glucosyl residues of UDP-Glc are inverted to form β-linked glucosyl residues in the polysaccharide product (http://www.cazy.org/). The suggestion that the enzymes operate through a general acid–base catalytic mechanism is supported by the essential role of acidic amino acids, as shown by the kinetics of carbodiimide inhibition of the (1,3)-β-d-glucan synthase from Lolium multiflorum (Bulone et al. 1998). Further support for the role of acidic amino acids in catalysis is given by site-directed mutagenesis studies. In the Saccharomyces cerevisiae chitin synthase (Chs2p) and in Gluconoacetobacter xylinus cellulose synthase (AcsAB) (Saxena & Brown 1997), acidic and other amino acids in the DDD35QXXRW motif have been implicated in the catalytic mechanism. The first Asp (D1, Table 5.3) of the extended motif forms part of the nucleotide-binding site and is found in domain A, as is the second Asp (D2, Table 5.3). The remaining portion of the motif, DQXXRW, contains the third Asp (D3, Table 5.3) and is found in domain B. Repetitiveness has been associated with the QXXRW component as non-repetitive glycosyl transferases have domain A but not domain B. Two other motifs of unknown function, KAG and QTP (Stasinopoulos et al. 1999) (Table 5.3), are confined to (1,4)- and (1,3)-β-d-glucan synthase sequences. Two motifs (FFCGS and Rx2FLx2PL) in known or putative bacterial cellulose synthases are proposed to have a role in determining (1,4)-βlinkage specificity (Römling 2002). In AcsAB, replacement of residues D1 or D2 impaired in vitro enzyme activity, as did replacement in Chs2p of D2 or D3, or of Gln, Arg or Trp in the QXXRW sequence. In chitin synthase Chs2p from Saccharomyces cereviseae, Nagahashi et al. (1995) have shown that replacement of residues D2 or D3, or of Gln, Arg or Trp in the QXXRW sequence impaired in vitro enzyme activity. Even conservative replacements of D3, or of Gln, Arg and Trp in Chs2p were not tolerated, suggesting that these residues are involved in the catalytic action. The two residues flanking D3 in Chs2p were also essential

138

■

Plant Polysaccharides

for enzyme activity. This is of interest because the first of these (Glu) is conserved in bacterial cellulose and curdlan synthases and the NodC proteins (Table 5.3). It is likely that other components necessary for the regulation of enzymic activity and chain initiation, elongation and termination will also be identified in the future. A noteworthy component of the DDD35QXXRW motif of many Family GT2 synthases is a DDX sequence (often DDG) that incorporates D1 of the extended motif (Table 5.3). In SpsA, the corresponding sequence is DDN (Table 5.3). Here D1(Asp99) sits adjacent to the distal phosphate of UDP in the binding site and coordinates with the leaving group Mn2+ (Fig. 5.8C), providing direct support for the proposed role of this Asp in metal binding (Charnock & Davies 1999; Tarbouriech et al. 2001). An additional sequence, DXD, occurs in many, but not all, families of nucleotide monosaccharide-dependent glycosyl transferases, whether retaining or inverting, repetitive or non-repetitive. In the case of a Clostridium difficile glucosyl transferase (Busch et al. 1998) and S. cerevisiae mannosyl transferase (Wiggins & Munro 1998), mutational substitution of the Asp residues results in complete abolition of the respective enzyme activities. The first D of the DXD sequence most likely aligns with D2 of the extended motif (Table 5.3). The notional reaction mechanism for a family GT2 polysaccharide synthase, involves inversion of configuration, (a→e), at the anomeric centre and proceeds via an oxacarbenium ion intermediate (Charnock & Davies 1999). Consideration of the SpsA structure with bound UDP-Mn2+ suggests that the role of Asp99 may be to assist leaving group departure (Fig. 5.8C), but the identification of the general base is more problematic. On the available evidence, Asp191 is the most probable candidate (Charnock & Davies 1999). The mechanistic basis of repetitive glycosyl transfer is not understood. Such a reaction requires that after each transfer event, the polysaccharide chain with the newly inserted terminal glycosyl residue, translocates (processes) in the acceptor site so that the −1 site (Davies et al. 1997) is unoccupied, to allow acceptance of the incoming NDP-glycosyl donor. It has been suggested that the QXXRW sequence may be important for the translocation of the elongating polysaccharide in the active site (Saxena & Brown 1997). This sequence is lacking in the non-repetitive glycosyl transferases, and it is not present in all repetitive β-glycan synthases (Table 5.3 and Griffiths et al. 1998). 5.8.5 Specification of linkage position in β-glycans The β-glycan synthases in family GT2 have related sequences, folding patterns and very likely similar mechanisms, but nevertheless synthesize products with different linkage types, such as (1,3)-β-d-glucans and (1,4)-βd-glucans, and others with both (1,3)- and (1,4)-linkages, such as in hyaluronan. Linkage type is presumably determined by subtle features of the active

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

139

site that allow the presentation of the appropriate hydroxyl on the growing polysaccharide chain to orient in the correct position to accept the transferred glycosyl residue. The question is not resolved at the level of the global folding patterns reflected in hydrophobic cluster analyses (Campbell et al. 1997). By analogy with the situation with (1,3)- and (1,3;1,4)-β-d-glucan hydrolases (Høj & Fincher 1995) its elucidation will require comparative studies of active sites in crystalline glycan synthases or at least, construction of chimeric enzymes as described for the α-l-fucosyl transferases (Legault et al. 1995). 5.8.6 Bifunctional GT2 β-Glycan Synthases The products of several family GT2 β-glycan synthases have dual monosaccharide units and dual linkage types in repeating units in the polysaccharide chain. For example, the membrane bound synthase for the capsular type 3 pneumococcal polysaccharide with a repeating unit Glcβ(1,4)GlcAβ(1,3) is a 49-kDa protein encoded by the cap3B gene (Arrecubieta et al. 1996) and has a dual specificity, synthesizing both (1,3)- and (1,4)-linkages in the polysaccharide. In this system the chain growth can be terminated by withdrawal of the UDPGlcA needed for synthesis. In vivo UDPGlcA levels, controlled by UDPGlcA dehydrogenase activity, may control chain growth (Ventura et al. 2006). Synthesis of the type 37 (1,3;1,2)-β-d-glucan pneumococcal polysaccharide, in which every 1,3-linked glucosyl residue in the backbone is substituted by a (1,2)-β-linked glucosyl residue, is determined by a single gene (tts) located distant from the cap locus responsible for capsular formation in all other pneumococal types. The tts gene encodes a GT2 glycosyltransferase that is an integral membrane protein with a potentially cleavable signal sequence. Cell-free membrane preparations support the synthesis of the type 37 polysaccharide from UDP[14C]-glucose without the participation of a lipid intermediate. The synthase has a dual specificity, synthesizing both (1,3)and (1,2)-linkages in this side-chain-branched polymer, a feature shared with the synthases producing the type 3 pneumococcal polysaccharide, hyaluronan in S. pyogenes, the K5 capsular polysaccharide in Escherichia coli, and chondroitin and heparosan from P. multocida. These are different from the NdvB and NdvC synthases for the cyclic (1,3;1,6)-β-glucan. Similarly, the streptococcal and eukaryotic hyaluronan synthases are transmembrane proteins with large intracellular, central domains that catalyse the incorporation of both GlcpA and GlcpNAc from the respective uridine nucleotides to form a repeating unit. For the S. pyogenes synthase there is both biochemical and genetic evidence (DeAngelis & Weigel 1994; Tlapak-Simmons et al. 1998) that only the hyaluronan synthase gene product is required for hyaluronan synthesis. No primer is needed. Thus, this 48-kDa protein (Kumari & Weigel 1997) binds the two substrates and synthesizes both linkage types in the active site. Whether there are two substrate binding

140

■

Plant Polysaccharides

sites and two glycosyl transferase catalytic sites, or whether there is one general binding/catalytic site that alternates in binding and catalytic specificity, remains to be demonstrated. Finally, the E. coli K5 capsular polysaccharide heparosan, a homologue of heparan, has a repeating →4)GlcpAβ(1,4)GlcpNAcα(1→ unit and its synthase is a 60-kDa, bifunctional, repetitive glycosyl transferase. There are two transferase catalytic sites in the synthase protein; removal of 139 amino acids from the C-terminus of the protein results in loss of UDPGlcpA activity, but UDPGlcpNAc activity is retained (Griffiths et al. 1998). In each case the synthases are molecules of 48–60 × 106 kDa, their action patterns are repetitive and they show high sequence homology with one another. In E. coli K5 heparosan synthase there are two transferase catalytic sites; removal of 139 amino acids from the C-terminus of the protein results in loss of UDPGlcpA activity, but UDPGlcpNAc activity is retained (Griffiths et al. 1998). Mutational studies with P. multocida hylauronan and chondroitin synthases indicate that they also possess independent hexosamine and glucuronic acid transfer sites (Jing & DeAngelis 2000). Kinetic studies suggest that there are separate receptor binding pockets for each of these glycosyl transferase activities that apparently interact with three or four saccharide units of the nascent hyaluronan chain (Williams et al. 2006). Although GT2 enzymes with dual specificity have not yet been described in higher plants, the examples presented above suggest that their existence in higher plants is possible. The corollary to that possibility is that, under appropriate conditions, plant polysaccharides such as (1,3;1,4)-β-d-glucans might be synthesized by a single enzyme, rather than a multienzyme complex. 5.8.7 Chain initiation, direction of chain elongation, and chain termination The process of polymerization of β-glycan chains, as with all biological polymers, involves three steps: chain initiation, chain elongation and chain termination. Details of these events and their mechanisms are sketchy for all but a few polysaccharides, but an understanding of these events is an essential part of a description and control of these processes. There are two general patterns of enzyme-catalysed elongation of polysaccharides. In tailward growth, saccharides are added repetitively to the nonreducing end of the growing polysaccharide chain. In these reactions the saccharide transferred is activated at its reducing end, as for example, in a NDP-saccharide, and the activating moiety is released in the reaction. In headward growth, saccharides are inserted at the reducing end of the nascent polysaccharide. The saccharide transferred is activated at its reducing end by attachment to, for example, a polyprenol pyrophosphate. The events of chain initiation and chain growth are intimately connected and one or both have been identified for a number of β-glycan synthases.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

141

Unequivocal headward polymerization has been demonstrated only for those β-glycans that are dependent on undecaprenol diphosphooligosaccharide donors. Thus, in succinoglycan biosynthesis, the chains are initiated on an undecaprenol diphosphoheterosaccharide and elongation proceeds by repetitive transfer of the growing polysaccharide chain on an undecaprenol pyrophosphate to a saccharide unit on an acceptor undecaprenol pyrophosphate-heterosaccharide. Vertebrate hyaluronan synthases have been reported to add monosaccharides to the reducing end of the growing chain (headward) (Prehm 1983; Asplund et al. 1998) but the hyaluronan synthases from Xenopus levis are reported to extend by tailward growth (Bodevin-Authelet et al. 2005) as do the streptococcal hyaluronan synthases ( Bodevin-Authelet et al. 2005; Tlapak-Simmons et al. 2005). In the case of the streptococcal hyaluronan synthases, the chain is initiated on a UDPGlcNAc acceptor and the growing chains are terminated at their reducing end with UDP. This process is analogous to protein and lipid synthesis. The biosynthesis of cellulose by Gluconoacetobacter xylinus has been studied using pulse chase techniques (Han & Robyt 1998), which suggest that the growth is from the reducing end (headward). This might imply the involvement of polyprenol intermediates. Putative lipid-saccharide intermediates have been implicated in the pathway for cellulose biosynthesis in Agrobacterium tumefaciens (Matthysse et al. 1995b). However, Koyama et al. (1997) used labelling and crystallographic techniques to reach the opposite conclusion, namely that growth is from the non-reducing end (tailward). The molecular directionality of chitin biosynthesis has been investigated (Imai et al. 2003) by transmission electron microscopy (TEM) using electron crystallography methods applied to reducing-end-labelled β-chitin microcrystals from vestimentiferan Lamellibrachia satsuma tubes and nascent βchitin microfibrils from the diatom Thalassiosira weissflogii. The data confirmed that the microfibrils were extruded with their reducing end away from the biosynthetic loci, an orientation consistent only with elongation through polymerization at the non-reducing end of the growing chains. For the GT2 NodC chitosaccharide synthases, chain initiation is on a simple monosaccharide primer. A membrane preparation from E. coli expressing the nodC gene from Mesorhizobium loti C forms chitin oligosaccharides from UDPGlcpNAc (Kamst et al. 1999). The synthesis is initiated on GlcNAc or p-nitrophenyl β-GlcpNAc, and the chains grow by sequential addition of monosaccharides to the non-reducing ends. Added chitosaccharides are not incorporated into the newly synthesized products, indicating that the glycosyl transfer reaction is repetitive, and the growing chain does not dissociate from the enzyme during productive reactions. The mechanism of chain termination in β-glycan synthesis is not understood, although it may be inferred from the polydispersity of their chain lengths that chain termination is not a very precise process. For example, the size distribution of cellulose chains is broad but the differences between

142

■

Plant Polysaccharides

cellulose in primary and secondary walls implies that some control is exerted (Marx-Figini 1969). The possibility that membrane-bound endocellulases found in plants and the cellulose-producing bacteria Agrobacterium tumefaciens and Gluconoacetobacter xylinus could serve as terminators of nascent cellulose chains has been proposed (Delmer 1999), although other roles for these enzymes have also been suggested (Matthysse et al. 1995a, 1995b). Different isoforms of hyaluronan synthases produce polymers of different sizes. Thus, the mammalian HAS1 and HAS2 synthase products are between 2 × 105 and approximately 2 × 106 Da, whereas those of HAS3 are shorter, 1 × 105 to 1 × 106 Da (Itano et al. 1999). The average rates of chain lengthening also varies between HAS1, HAS2 and HAS3 at 1256 * 251, 1014 * 338 and 174 * 42 monosaccharides/min, respectively. The Lolium multiflorum (1,3)-β-d-glucan synthase forms products with at least 30 glucosyl residues and chain extension occurs at the non-reducing end of the growing chain (Henry & Stone 1982).

5.9 Subcellular locations of GT2 enzymes in plants Although plant cellulose synthase complexes operate in the plasma membrane, it is likely that they are assembled in the Golgi and later transported to the plasma membrane (Haigler & Brown 1986). As mentioned above, the cellulose synthase rosettes probably contain six subunits that themselves consist of six individual CesA proteins, and that together these synthesize a total of 36 cellulose chains that aggregate into microfibrils (Doblin et al. 2002). No such aggregates have been identified for the (1,3;1,4)-β-d-glucan synthases, and indeed the location of the CslF proteins within the cell remains unknown. It is generally assumed that (1,3;1,4)-β-d-glucans are synthesized in the Golgi and transported via Golgi-derived vesicles to the plasma membrane, where they are deposited into the wall. However, Wilson et al. (2006) were unable to detect any (1,3;1,4)-β-d-glucan in the Golgi of developing endosperm cells in barley, using specific monoclonal antibodies.

5.10 Proteomics and biochemical approaches to the identification of GT48 (1,3)-β-D-glucan synthases from plants Characterization of the GT48 proteins that have been implicated in higher plant (1,3)-β-d-glucan synthase has been effected through biochemical, proteomic and bioinformatic approaches. In contrast to the cellulose synthases and the (1,3;1,4)-β-d-glucan synthases, the (1,3)-β-d-glucan or callose synthases have been more amenable to purification from tissue extracts of plants (Bulone et al. 1995; Li et al. 2003; Pelosi et al. 2003). They appear to be more stable, and partial amino acid sequences have been obtained directly from

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

143

proteins that appear to have (1,3)-β-d-glucan synthase activity. However, there are still some anomalies and uncertainties regarding the identity of the genes encoding these enzymes. Callose is a plant (1,3)-β-d-glucan in some, but not all, cell walls. It is deposited in a number of specialized walls or wall-associated structures during normal plant development and also in response to wounding or pathogen challenge (Stone & Clarke 1992; Farrokhi et al. 2006). The identification of the gene family that encodes the callose synthases came via an indirect route, from work on the genes responsible for the synthesis of a (1,3)-β-d-glucan in fungi. In the yeast Saccharomyces cerevisiae, the Fks (FK506 supersensitive) genes are believed to be involved in (1,3)-β-d-glucan biosynthesis (Douglas et al. 1994; Cabib et al. 2001; Dijkgraaf et al. 2002), as shown by mutant analyses (Douglas et al. 1994; Inoue et al. 1995) and biochemical studies (Inoue et al. 1995). Searches of EST databases for plant genes with similarity to the Fks genes led to the identification of a family of candidate genes, called glucan synthase-like (GSL) genes (Saxena & Brown 2000; Cui et al. 2001b). The proteins encoded by these genes are classified in the GT48 family of glycosyl transferases (Coutinho & Henrissat 1999; http:// afmb.cnrs-mrs.fr/CAZY/). As discussed earlier, cell-free preparations of higher plants invariably produced (1,3)-β-d-glucan as the major product when UDP-Glc was used as the substrate (Feingold et al. 1958). The enzyme is located in the plasma membrane (Henry et al. 1983) but active enzyme preparations have been obtained by detergent solubilization (Henry & Stone 1982; Meikle et al. 1991; Kuribayashi et al. 1992; Kuribayashi et al. 1993; Bulone et al. 1995). By careful selection of detergents, Lai-Kai-Him et al. (2001) and Colombani et al. (2004) were able to obtain preparations that specifically synthesized either (1,3)-βd-glucan or (1,4)-β-d-glucan. The detergent-solubilized preparations produced microfibrillar (1,3)-β-d-glucan (Bulone et al. 1995; Colombani et al. 2004; Pelosi et al. 2006). Protein profiles of the detergent solubilized enzymes were invariably complex with prominent bands in the 20–70-kDa range (Fredrikson et al. 1991; Meikle et al. 1991). Feingold et al. (1958) showed that cellobiose and laminaribiose are activators of the (1,3)-β-d-glucan synthase and subsequently a wide range of β-d-glucosides (Callaghan et al. 1988a; Callaghan et al. 1988b; Ng et al. 1996) including cellobiose and laminaribiose (Feingold et al. 1958) and the naturally occurring furfuryl β-d-glucoside (Ohana et al. 1991; Ohana et al. 1992) were shown to be activators. There is no evidence that β-glucosides were incorporated into the polymeric product (Feingold et al. 1958; Henry & Stone 1982). The activation of the pollen tube (1,3)-β-d-glucan synthase by trypsin led to the suggestion that the enzyme was present as an inactive zymogen (Li et al. 1999). The identification of the gene family that encodes the callose synthases in plants came from the important observation that a 170-kDa polypeptide from mung bean (Phaseolus vulgaris), enriched in (1,3)-β-glucan synthase (Kudlicka & Brown 1997), yielded a tryptic peptide showing limited

144

■

Plant Polysaccharides

homology with the yeast Fks catalytic subunit. Searches of EST databases for plant genes with similarity to the Fks genes led to the identification of a family of candidate genes, the GSL genes (Saxena & Brown 2000; Cui et al. 2001b). Later, Li et al. (2003) purified a (1,3)-β-d-glucan synthase more than 60-fold from barley suspension-cultured cells by detergent extraction, CaCl2 treatment, sucrose density gradient centrifugation and non-denaturing gel electrophoresis. The enzyme preparation synthesized (1,3)-β-d-glucan in vitro. High-molecular-mass proteins within the preparation were recognized by antibodies raised against a 17-kDa protein generated by heterologous expression of a fragment of a HvGSL1 cDNA (Li et al. 2003). Furthermore, mass spectrometric analyses showed that tryptic peptides produced by in-gel digestion of the active enzyme exactly matched peptides predicted from the gene sequence. Thus, the amino acid sequence predicted from the HvGSL1 gene was linked with the actual amino acid sequence of an active (1,3)-β-d-glucan synthase fraction from barley (Li et al. 2003). Similar results were obtained with a partially purified fraction from pollen tubes of Nicotiana alata. The preparation had (1,3)-β-d-glucan synthase activity and sequences of tryptic peptides from the active protein fraction matched the nucleotide sequence of the Nicotiana GSL gene, which together provided proteomic and biochemical evidence to link the callose synthase in Nicotiana alata pollen tubes to the product of the NaGSL1 gene (Brownfield et al. 2007). Mutant analyses have also been used to link GSL function with (1,3)-β-dglucan deposition. Thus, the absence of callose from cell wall appositions (papillae) that form beneath fungal infection sites in Arabidopsis is associated with null mutations in the AtGSL5 gene (Jacobs et al. 2003; Nishimura et al. 2003), strongly implicating the AtGSL5 gene in wound-activated callose synthesis. Plant GT48 GSL genes are usually members of multigene families, with 12 genes in Arabidopsis (Hong et al. 2001a), 10 in rice, more than 9 in Populus trichocarpa (http://genome.jgi-psf.org/Poptr1/Poptr1.home.html), and at least 8 in barley (M. Schober, A.K. Jacobs, R.A. Burton and G.B. Fincher, unpublished data). A phylogenetic tree based on the GT48 GSL proteins from Arabidopsis is shown in Fig. 5.9. The GSL genes can be divided into two distinct groups, based on gene structure (Fig. 5.10, Table 5.4; Doblin et al. 2001). One group of genes contain 3 introns or fewer, while others are highly fragmented, with about 40 introns (Richmond & Somerville 2000; Doblin et al. 2001; Hong et al. 2001a; Verma & Hong 2001). The GSL genes from barley and rice have a similar diversity of structure (Richmond & Somerville 2000; Jacobs et al. 2003). It has been suggested that each of the 12 Arabidopsis genes is responsible for callose synthesis in a different location within the plant (Hong et al. 2001a) and the expression of particular GSL genes specifically in tissues that produce callose, such as the expression of GhGSL1 (Cui et al. 2001b) in cotton fibres and NaGSL1 in pollen tubes (Doblin et al. 2001), supports this suggestion. From an evolutionary point of view it is worth noting that the (1,3)-β-dglucan synthesis has been studied in some detail in the yeasts Saccharomyces

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

145

AtGSL1 AtGSL5 AtGSL8 AtGSL10 AtGSL4 AtGSL2 AtGSL3 AtGSL6 AtGSL9 AtGSL12 AtGSL7 AtGSL11 0.1

Figure 5.9 GT48 phylogenetic tree. The GT48 glucan-synthase-like (GSL) family of proteins in Arabidopsis thaliana consists of 12 members which form 3 major clades. The GT48 locus identifiers of sequences used are listed in Table 5.4. AtGSL1 AtGSL2 AtGSL3 AtGSL4 AtGSL5 AtGSL6 AtGSL7 AtGSL8 AtGSL9 AtGSL10 AtGSL11 AtGSL12 OsGSL1 OsGSL2 OsGSL3 5 kb

10 kb

15 kb

20 kb

Figure 5.10 Structures of GT48 genes in higher plants. Note the two distinct groups, with one group of genes containing three introns or fewer, while others are highly fragmented, with about 40 introns. Taken from http://cellwall.stanford.edu.

cerevisiae (Shematek & Cabib 1980; Cabib et al. 1982) and Candida albicans (Orlean 1982; Notario et al. 1982), and in other fungi. Cell-free preparations from these species incorporate glucose from UDP[14C]-α-glucose into an ethanol-insoluble (1,3)-β-d-glucan and the S. cerevisiae synthase complex has

146

■

Plant Polysaccharides Table 5.4 Arabidopsis thaliana GT48 locus identifiers of sequences used in Figs 5.9 and 5.10 AtGSL1 AtGSL2 AtGSL3 AtGSL4 AtGSL5 AtGSL6 AtGSL7 AtGSL8 AtGSL9 AtGSL10 AtGSL11 AtGSL12

At4g04970 At2g13680 At2g31960 At3g14570 At4g03550 At1g05570 At1g06490 At2g36850 At5g36870 At3g07160 At3g59100 At5g13000

been resolved into two fractions. A soluble fraction contains a 29-kDa Rho1p GTPase while a membrane fraction contains a pair of approximately 210kDa polytopic proteins encoded by the Fks1 and Fks2 genes. Proteins closely related to Fks1 and Fks2 are found in a number of fungi and have been classified as family GT48 glycosyltransferases (http://afmb.cnrs-mrs.fr/ CAZY/). However, there is no unequivocal evidence that they act as glycosyltransferases. Furthermore, unlike the catalytic subunit of the curdlan synthase, the Fks proteins show no significant sequence similarity with GT enzymes from other families. In particular, the conserved motifs of the GT2 family members, which are potentially involved in substrate binding and catalysis, are absent. The regulatory subunit of the (1,3)-β-d-glucan complex in yeasts, Rho1p, is a central regulator of yeast morphogenesis in its GTP-bound form, through binding to and activating the Fks1 and Fks2 proteins and at least two other morphogenesis effectors. In vivo, yeast glucan synthesis is sensitive to membrane perturbations caused by lipophilic compounds such as fungal lipopeptide, echinocandins and pneumocandins, and the liposaccharide, papulocandins. Mutations that lead to resistance to these fungicidal antibiotics map to Fks1 in both S. cerevisiae and C. albicans.

5.11 Enzymic properties of the GT48 proteins 5.11.1 Topology The putative (1,3)-β-d-glucan synthases encoded by the GSL genes are members of the GT48 family of glycosyl transferases and although they are also predicted to be integral, membrane-bound proteins with an inverting action pattern (http://www.cazy.org/), there are clearly a number of important differences between them and the GT2 synthases. Predicted GSL

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

147

proteins are very large, with about 1800–1950 amino acid residues (Saxena & Brown 2000; Hong et al. 2001a; Li et al. 2003). They are predicted to contain 13–16 TMHs (Fig. 5.7; Cui et al. 2001b; Doblin et al. 2001; Verma & Hong 2001; Østergaard et al. 2002; Li et al. 2003). An N-terminal domain of 350–500 amino acid residues is predicted to be cytoplasmic and is followed by a predicted membrane-associated region of six TMHs (Li et al. 2003). A large central domain of 620–780 amino acid residues is predicted to be cytoplasmic, and towards the C-terminal end of the protein there is another membrane-associated region predicted to contain between 8 and 10 TM helices (Fig. 5.7). 5.11.2 Catalytic mechanisms Although there is considerable evidence that the GSL proteins are involved in (1,3)-β-d-glucan synthesis and that they can use UDP-Glc as a source of glucosyl residues, they do not contain a recognizable UDP-Glc binding motif. There has been some debate as to whether the GSL proteins constitute the catalytic subunit itself, or whether they are simply a pore-forming unit in a larger (1,3)-β-d-glucan synthase complex. The GSL proteins have some structural similarity to bacterial and eukaryotic transporters (Douglas et al. 1994; Cui et al. 2001b; Hong et al. 2001a; Dijkgraaf et al. 2002). However, it remains possible that GSL proteins are catalytically active but use a novel motif for UDP-Glc binding. The (1,3)-β-d-glucan synthases of plants may consist of multiprotein complexes. Thus, the barley GSL1 protein has been identified in high-molecular-mass complexes on gels (Li et al. 2003), but other polypeptides were also present in the region containing the GSL proteins. Potential complex components have been identified by their interaction with GSL proteins. The GhGSL1 protein from cotton fibres binds calmodulin in vitro in the presence of Ca2+ (Cui et al. 2001b) and yeast-twohybrid analyses indicate that AtGSL6, the putative cell-plate callose synthase, interacts with phragmoplastin (a cell-plate-associated dynamin-like protein) and a novel UDP-Glc transferase (Hong et al. 2001b). Furthermore, (1,3)-β-d-glucans can form triple helices of parallel chains (Stone & Clarke 1992; Pelosi et al. 2003) and although these might form spontaneously, it is formally possible that they could result from the activity of associated catalytic subunits (Lai-Kai-Him et al. 2001).

5.12 Future role of biochemistry in the characterization of GT2 and GT48 enzymes Now that the characteristics of the gene families encoding higher plant GT2 and GT48 glycosyl transferases have been described, one can expect the roles of individual genes and enzymes to be defined in detail in the immediate future, especially with respect to the temporal and spatial aspects of

148

■

Plant Polysaccharides

individual gene transcription. One can also be confident that transcription factors that regulate the first stages of gene expression will be identified and the importance of other regulatory processes through small RNA molecules, mRNA turnover, alternative splicing of mRNAs, etc., will become clearer. However, although we suggested earlier in this review that biochemical approaches had met with limited success in identifying genes encoding GT2 and GT48 enzymes, it is evident now that a resurgence in biochemical approaches to defining cellular processes will occur in the near future. We would predict that biochemical approaches will be applied in three major areas, namely in the definition of cellular functions of gene products, in the regulation of protein activity through post-translational modifications, and in the definition of protein–protein interactions that occur widely in cellular processes. These applications are mentioned briefly below. 5.12.1 Biochemistry and the definition of gene function While genome sequencing programs are providing extensive lists of structural genes from selected organisms, attention is likely to shift towards the accurate assignment of functions to all individual genes in a genome. The Arabidopsis 2010 project had this as a central objective. Currently, the functions of about 50% of genes on plant genomes can be predicted from sequence similarities with proteins from other organisms. However, the evidence for function is usually circumstantial or inferred from a third source and in most cases direct biochemical evidence for function is not available (Hrmova & Fincher 2009). In many cases the identities of the genes and their annotated function are simply incorrect (Brown & Sjölander 2006). In the case of the GT2 CesA genes from plants, there is certainly evidence that some members of the subgroup encode enzymes that are involved in cellulose synthesis. However, on this basis it is often assumed that all the CesA genes encode cellulose synthases. This may be correct, but equally some of the CesA genes might encode synthases that are involved in the synthesis of a non-cellulosic polysaccharide that might have a chemical structure closely related to cellulose. Another key question in the context of this review is whether or not the GT48 protein product of an expressed GSL gene really does have catalytic activity with respect to (1,3)-β-d-glucan synthesis. One would envisage that, initially at least, the new biochemical approaches will focus on the heterologous expression of cDNAs and gene fragments, so that relatively pure proteins can be isolated from specific genes for functional analysis (Hrmova & Fincher 2007). Heterologous expression of membrane proteins with multipass TMHs of the type found in the GT2 and GT48 families is still not straightforward. For membrane-bound enzymes, it is becoming apparent that they depend upon associated lipids for correct folding, for structural integrity and for functional activity (Opekarová & Tanner 2003). The eukaryotic heterologous expression systems of Saccharomyces and Pichia pastoris are likely to become the systems of choice

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

149

for plant protein expression, because of the more appropriate posttranslational modifications and processing that occur in these systems. While the S. cerevisiae system often hyperglycosylates proteins, which is likely to impair the folding and/or biological activity of the recombinant protein (Malissard et al. 1999), P. pastoris is less prone to hyperglycosylating the expressed proteins (Eckart & Bussineau 1996; Malissard et al. 1999). In addition, yeasts perform other eukaryotic post-translational modifications, such as the removal of N-terminal methionine residues, the addition of acetyl groups to the N-terminus, the formation of C-terminal methylation complexes, myristoylation, palmitolyation and farnesylation (Eckart & Bussineau 1996), which can be important in intracellular targeting of expressed proteins. This might explain why yeast is a relatively good system for the expression of membrane proteins (Grisshammer & Tate 1995; Eckart & Bussineau 1996). In comparison to soluble proteins, a key limitation of heterologous expression systems for the production of membrane-associated enzymes is that purification of the expressed protein remains challenging. To overcome this, another experimental approach to the expression of functional membrane proteins is to use a range of cell-free in vitro systems (Endo & Sawasaki 2004; Klammt et al. 2006). A variety of preformed micelles can be added to the transcription and translation reaction mixture so that the membrane proteins can be synthesized directly into a defined hydrophobic environment (Hrmova & Fincher 2007). It is well known that detergents interact with membrane proteins non-specifically and therefore a large variety of detergents is usually tested to select conditions that are beneficial to membrane protein folding but are chemically neutral in reaction mixtures (Hrmova & Fincher 2007). 5.12.2 Post-translational modifications of glycosyl transferases Evidence is emerging to suggest that post-translational modifications of membrane-bound proteins are important in the regulation of their activity and function in the cell. In the case of GT2 and GT48 enzymes, the extent to which glycosylation or phosphorylation might affect activity is largely unknown. Taylor (2007) suggests that phosphorylation of the AtCesA7 (Irx3) catalytic subunits may target them for degradation via a proteasomedependent pathway. The (1,3)-β-d-glucan synthase from pollen tubes of Nicotiana alata can be activated in vitro by treatment with trypsin or certain zwitterionic detergents (Schlupmann et al. 1993; Li et al. 1997; Li et al. 1999). An important role for protein lipidation is also emerging from studies on membrane-bound proteins from mammalian and yeast systems. Proteins can be covalently modified with a variety of lipids, including fatty acids, isoprenoids, and cholesterol. Lipid modifications are believed to be involved in the localization and cellular trafficking of proteins, and also in their functional activity (Nadolski & Linder 2007). Commonly-observed lipid

150

■

Plant Polysaccharides

modifications include N-myristoylation, S-palmitoylation, and prenylation, as well as the attachment of glycosylphosphatidylinositol (GPI) anchors. For example, the reversible addition of palmitate and other long-chain fatty acids to proteins at cysteine residues has a number of functional consequences, including the facilitation of protein–membrane association, protein trafficking and the regulation of protein stability (Nadolski & Linder 2007). Thus, palmitolyation of the yeast GT2 enzyme chitin synthase has been associated with the normal transport of the enzyme from the ER to the cell surface (Lam et al. 2006). Given that the yeast chitin synthase is a member of the GT2 family of glycosyl transferases, it is quite likely that plant enzymes from the GT2 group might also be post-translationally modified through the addition of lipid molecules and that the modification could affect activity, stability and targeting. 5.12.3 Protein–protein interactions In the future, defining the functions of individual components of protein complexes and the nature of protein–protein interactions in the cell will become increasingly important as we attempt to define complex cellular processes. Membrane proteins are likely to be particularly important as we dissect processes such as signal transduction and the activation of membraneassociated transporters. An obvious early target in defining protein–protein interactions involving members of the GT2 family has been the composition and nature of the cellulose-synthesizing terminal rosette structures that have been observed in the plasma membrane (Haigler & Brown 1986; Doblin et al. 2002; Taylor et al. 2003). It will be important to define the chemistry of the putative interactions between the individual CesA proteins in the complex and whether other proteins might be involved. It is not known if the CslF proteins might be part of a larger complex during (1,3;1,4)-β-d-glucan synthesis (Buckeridge et al. 2004; Burton et al. 2006), although the GT2 curdlan synthase (CrdS) enzymes appear to require several associated proteins for activity (Stasinopoulos et al. 1999). In the CesAs in cotton, the enzyme binds two atoms of Zn2+ in the reduced state. Under oxidative conditions the N-terminal portion of the cotton CesA1 and CesA2 interact to form homo- or heterodimers via intermolecular disulfide bonds present in zinc finger domains (Kurek et al. 2002). Under reduced conditions, the zinc finger domain coordinates two Zn2+ ions and can interact with either lipid transfer protein (LTP), metallothionein (Mt), microtubules (MT) or cysteine protease. Under oxidized conditions, the CesA protein can dimerize with itself or another CesA protein. Proteins such as the Korrigan endocellulase (Kor), the plasma membrane-associated form of sucrose synthase (pm-SuSy), actin and the herbicide CGA binding protein (CGAbp), all of which have been implicated in cellulose synthesis, might also interact in a similar fashion (Doblin et al. 2002).

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

151

In the case of the GT48 proteins, there is also some evidence that the GSL proteins are components of a larger (1,3)-β-d-glucan synthase complex, based on observations that multiple polypeptides are present in fractions enriched in (1,3)-β-d-glucan synthase activity, on yeast-two-hybrid and on calmodulin-binding assays (Cui et al. 2001b; Hong et al. 2001a; Verma & Hong 2001).

5.13

Applications of modified levels of plant β-D-glucans

In addition to the biological importance of these polysaccharides, interest in the GT2 and GT48 enzymes that are believed to be involved in their biosynthesis has been stimulated by the commercial and nutritional properties of the polysaccharide products of the enzymes. Thus, these plant polysaccharides are of fundamental importance in plant growth and development, resistance to pathogen invasion, the quality of plant-based foods and the properties of plant fibres and fuels. Cellulose, for example, represents the world’s largest renewable carbon resource. Cellulose biogenesis by land plants and marine algae from photosynthetically-derived carbohydrate occurs at the prodigious rate of 8.5 × 1010 tonnes per year. One particularly promising agro-industrial application is in the replacement of fossil fuels with bioethanol. Currently, bioethanol production is increasing rapidly, and with recent advances in the catalytic efficiency of hydrolytic enzymes and non-enzymatic methods for depolymerizing wall polysaccharides, it is becoming apparent that lignocellulosic complexes and other crop residues that consist predominantly of cellulose and non-cellulosic polysaccharides of wall origin will become economical as a source of fermentable sugars for ethanol production. Attention so far has focused on the enzymic degradation of the cellulose and non-cellulosic polysaccharides, but increased knowledge of biosynthetic mechanisms will also provide opportunities to manipulate the fine structures of the polysaccharides, the interactions of cellulose and non-cellulosic polysaccharides within the wall and their relative abundance in walls during plant growth, such that major crop residues and other sources of plant biomass will be more amenable to rapid enzymic degradation and the products of hydrolysis might be modified to enhance the efficiency of the fermentation process. In addition, the cellulose content of cereal stems is related to strength (Appenzeller et al. 2004), and increased cellulose content would be expected to reduce the susceptibility of crops to losses that result from lodging prior to harvest. Similarly, the (1,3;1,4)-β-d-glucans of cereals and grasses have a number of important nutritional and industrial applications that have led to interest in the enzymes involved in their biosynthesis. A high proportion of our daily caloric intake is obtained from rice (Oryza sativa), wheat (Triticum aestivum), sorghum (Sorghum bicolor), barley (Hordeum vulgare), the millets (Panicum miliaceum and Pennisetum americanum) and sugar cane (Saccharum

152

■

Plant Polysaccharides

officinarum), while numerous forage and fodder grass species support the production of sheep, cattle and other domesticated livestock. Maize (Zea mays) is also used widely for animal feed, while switchgrass (Panicum virgatum) and other perennial grasses are showing considerable promise as future biomass energy crops in North America (McLaren 2005; Burton et al. 2006). In the areas of human health, the (1,3;1,4)-β-d-glucans are components of dietary fibre that are highly beneficial in the prevention and treatment of serious human health conditions, including colorectal cancer, high serum cholesterol and cardiovascular disease, obesity, and non-insulin-dependent (type 2) diabetes (Brennan & Cleary 2005). In contrast, (1,3;1,4)-β-d-glucans have antinutritive effects in monogastric animals such as pigs and poultry (Brennan & Cleary 2005), and are important in many cereal processing applications, including malting and brewing. In most of these applications, the key property of the (1,3;1,4)-β-d-glucans is their molecular asymmetry and the resultant high viscosity in aqueous solutions. Thus, highly viscous contents of the small intestine in monogastric animals slow the diffusion both of hydrolytic enzymes to substrates such as starch, and of hydrolytic products back to the intestinal epithelial layer for absorption. This reduces the metabolizable energy of animal feeds and hence growth rate, but is beneficial in human health because it slows the release of glucose into the bloodstream and flattens the insulin response curve. In the brewing industry, excessive levels of undegraded (1,3;1,4)-β-d-glucans in malt extracts will increase their viscosity and hence slow filtration steps in the process. Undegraded (1,3;1,4)-β-d-glucans can also lead to the formation of undesirable hazes in the final product (Bamforth 1993). In summary, therefore, there are commercial incentives to both increase and decrease levels of (1,3;1,4)-β-d-glucans in common cereal grains and vegetative tissues. The potential applications for (1,3)-β-d-glucan synthases in plants is likely to be related to their role in plant–pathogen interactions, during which plant host cells respond to microbial attack by rapidly synthesizing and depositing callose in close proximity to the invading pathogen (Ryals et al. 1996; Donofrio & Delaney 2001). These callosic deposits are commonly referred to as papillae and are thought to contain, in addition to (1,3)-β-d-glucan, minor amounts of other polysaccharides, phenolic compounds, reactive oxygen intermediates and proteins (Smart et al. 1986; Bestwick et al. 1997; Heath et al. 2002). It has been suggested that the papillae act as a physical barrier to microbial penetration, but no general agreement on the precise function of callosic papillae during microbial attack has been reached (Stone & Clarke 1992). If the deposited callose did indeed slow or immobilize the invading microorganisms, the host plant could focus upon them a number of antimicrobial compounds, such as wall-degrading enzymes, phytoalexins and active oxygen species, or initiate cascade responses involving racespecific resistance genes (Jacobs et al. 2003). In Arabidopsis T-DNA insertion lines in which wound callose formation is compromised, the absence of

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

153

callose in papillae or haustorial complexes correlates with growth cessation of several normally virulent fungal pathogens (Jacobs et al. 2003; Nishimura et al. 2003). This observation therefore offers a potential opportunity to increase crop plant resistance to certain fungal pathogens through downregulation of callose synthase activity. Manipulation of callose in plasmodesmata might also find applications in plant protection against viral diseases, through the inhibition of viral movement within the plant. As noted above, these commercial incentives, together with the broader importance of cellulose, (1,3;1,4)-β-d-glucans, (1,3)-β-d-glucans and related polysaccharides in biological terms, have resulted in an intense interest in the biosynthetic enzymes of the GT2 and GT48 families that are believed to mediate in the biosynthesis of these polysaccharides.

Acknowledgements This work has been supported by grants from the Australian Research Council, the Grains Research and Development Corporation and the South Australian State government.

References Aloni, Y., Delmer, D.P., Benziman, M. (1982) Achievement of high rates of in vitro synthesis of 1,4-β-d-glucan: activation by cooperative interaction of the Acetobacter xylinum enzyme system with GTP, polyethylene glycol, and a protein factor. Proceedings of the National Academy of Sciences of the U S A, 79, 6448–6452. Altabe, S.G., Inon De Iannino, N., De Mendoza, D., Ugalde, R.A. (1994) New osmoregulated b(1–3),b(1–6) glucosyltransferase(s) in Azospirillum brasilense. Journal of Bacteriology, 176, 4890–4898. Appenzeller, L., Doblin, M., Barreiro, R., et al. (2004) Cellulose synthesis in maize: isolation and expression analysis of the cellulose synthase (CesA) gene family. Cellulose, 11, 287–299. Arioli, T., Peng, L., Betzner, A., et al. (1998) Molecular analysis of cellulose biosynthesis in Arabidopsis. Science, 279, 717–720. Arrecubieta, C., López, R., García, E. (1996) Type 3-specific synthase of Streptococcus pneumoniae (Cap3B) directs type 3 polysaccharide biosynthesis in Escherichia coli and in pneumococcal strains of different serotypes. Journal of Experimental Medicine, 184, 449–455. Ashby, M.F., Gibson, L.J., Wegst, U., Olive, R. (1995) The mechanical properties of natural materials I. Material property charts Proceedings of the Royal Society of London Series A, 450, 123–140. Asplund, T., Brinck, J., Suzuki, M., Briskin, M.J., Heldin, P. (1998) Characterization of hyaluronan synthase from a human glioma cell line. Biochimica et Biophysica Acta, 1380, 377–388. Au-Young, J., Robbins, P. (1990) Isolation of a chitin synthase gene (CHS1) from Candida albicans by expression in Saccharomyces cerevisiae. Molecular Microbiology, 4, 197–207.

154

■

Plant Polysaccharides

Bacic, A., Harris, P.J., Stone, B.A. (1988) Structure and function of plant cell walls. In: The Biochemistry of Plants (ed. J. Preiss). Academic Press, San Diego. Bamforth, C.W. (1993) β-Glucans and β-glucanases in malting and brewing: practical aspects. Brewers Digest, 69, 12–16. Becker, M., Vincent, C., Reid, J.S.G. (1995) Biosynthesis of (1→3)(1→4)-beta-glucan and (1→3)-beta-glucan in barley Hordeum vulgare L. Planta, 195, 331–338. Beeckman, T., Przemeck, G.H.K., Stamatiou, G., et al. (2002) Genetic complexity of cellulose synthase A gene function in Arabidopsis embryogenesis. Plant Physiology, 130, 1883–1893. Bernal, A.J., Jensen, J.K., Harholt, J., et al. (2007) Disruption of AtCslD5 results in reduced growth, reduced xylan and homogalacturonan synthase activity and altered xylan occurrence in Arabidopsis. Plant Journal, 52, 791–802. Bernal, A.J., Yoo, C.-M., Mutwil, M., et al. (2008) Functional analysis of the cellulose synthase-like genes CSLD1, CSLD2 and CSLD4 in tip-growing Arabidopsis cells. Plant Phys, 148, 1238–1253. Bernard, M., Latgé, J.-P. (2001) Aspergillus fumigatus cell wall: Composition and biosynthesis. Medical Mycology, 39, 9–17. Bestwick, C.S., Brown, I.R., Bennett, M.H.R., Mansfield, J.W. (1997) Localization of hydrogen peroxide accumulation during the hypersensitive reaction of lettuce cells to Pseudomonas syringae pv phaseolicola. Plant Cell, 9, 209–221. Bhagwat, A.A., Keister, D.L. (1992) Synthesis of β-glucans by Bradyrhizobium japonicum and Rhizobium fredii. Canadian Journal of Microbiology, 38, 510–514. Bhagwat, A.A., Keister, D.L. (1995) Site-directed mutagenesis of the beta-(1→3),beta(1→6)-d-glucan synthesis locus of Bradyrhizobium japonicum. Molecular Plant Microbe Interactions, 8, 366–370. Bhagwat, A.A., Mithöfer, A., Pfeffer, P.E., et al. (1999) Further studies of the role of cyclic β-glucans in symbiosis. An ndvC mutant of Bradyrhizobium japonicum synthesizes cyclodecakis-(1–3)-β-glucosyl. Plant Physiology, 119, 1057–1064. Bhagwat, A.A., Tully, R.E., Keister, D.L. (1992) Isolation and characterization of an ndvB locus from Rhizobium fredii. Molecular Microbiology, 6, 2159–2165. Bhagwat, A.A., Tully, R.E., Keister, D.L. (1993) Identification and cloning of a cyclic beta-(1→3), beta-(1→6)-d-glucan synthesis locus from Bradyrhizobium japonicum. FEMS Microbiology Letters, 114, 139–144. Bodevin-Authelet, S., Kusche-Gullberg, M., Pummill, P.E., DeAngelis, P.L., Lindahl, U. (2005) Biosynthesis of hyaluronan: direction of chain elongation. Journal of Biological Chemistry, 280, 8813–8818. Bohm, N., Kulicke, W.-M. (1999) Rheological studies of barley (1→3),(1→4)-β-glucan in concentrated solution: Investigation of the viscoelastic flow behavior in the solstate. Carbohydrate Research, 315, 302–311. Breedveld, M.W., Miller, K.J. (1994) Cyclic beta-glucans of members of the family Rhizobiaceae. Microbiological Reviews, 58, 145–161. Brennan, C.S., Cleary, L.J. (2005) The potential use of (1→3,1→4)-β-d-glucans as functional food ingredients. Journal of Cereal Science, 42, 1–13. Brown, D., Sjölander, K. (2006) Functional classification using phylogenomic inference. PLoS Computational Biology, 30, 2(6):e77. Brownfield, L., Ford, K., Doblin, M.S., Newbigin, E., Read, S., Bacic, A. (2007) Proteomic and biochemical evidence links the callose synthase in Nicotiana alata pollen tubes to the product of the NaGSL1 gene. Plant Journal, 52, 147– 156.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

155

Buckeridge, M.S., Vergara, C.E., Carpita, N.C. (1999) The mechanism of synthesis of a mixed-linkage (1→3),(1→4)-β-d-glucan in maize. Evidence for multiple sites of glucosyl transfer in the synthase complex. Plant Physiology, 120, 1105–1116. Buckeridge, M.S.C., Rayon, B., Urbanowicz, M.A.S.T., Carpita, N. (2004) Mixed linkage glucan (1→3), (1→4)-β-d-glucans of grasses. Cereal Chemistry, 81, 115–127. Bulawa, C., Slater, M., Cabib, E., et al. (1986) The S. cerevisiae structural gene for chitin synthase is not required for chitin synthesis in vivo. Cell, 46, 213–225. Bulone, V., Fincher, G.B., Stone, B.A. (1995) In vitro synthesis of a microfibrillar (1,3)-β-glucan by a ryegrass (Lolium multiflorum) endosperm (1,3)-β-glucan synthase enriched by product entrapment. Plant Journal, 8, 213–25. Bulone, V., Lam, B.-T., Stone, B.A. (1998) The effect of amino acid modifying reagents on the activity of a (1,3)-β-glucan synthase from Italian ryegrass (Lolium multiflorum) endosperm. Phytochemistry, 50, 9–15. Burn, J.E., Hocart, C.H., Birch, R.J., Cork, A.C., Williamson, R.E. (2002) Functional analysis of the cellulose synthase genes CesA1, CesA2, and CesA3 in Arabidopsis. Plant Physiology, 129, 797–807. Burton, R.A., Gibeaut, D.M., Bacic, A., et al. (2000) Virus-induced silencing of a plant cellulose synthase gene. Plant Cell, 12, 691–706. Burton, R.A., Shirley, N.A., King, B.J., Harvey, A.J., Fincher, G.B. (2004 ) The CesA gene family of barley. Quantitative analysis of transcripts reveals two groups of co-expressed genes. Plant Physiology, 134, 224–236. Burton, R.A., Wilson, S.M., Hrmova, M., et al. (2006) Cellulose synthase-like CslF genes mediate the synthesis of cell wall (1,3;1,4)-beta-d-glucans. Science, 311, 1940–1942. Burton, R.A., Jobling, S.A., Harvey, A.J., et al. (2008) The genetics and transcriptional profiles of the cellulose synthase-like HvCslF gene family in barley (Hordeum vulgare L.). Plant Physiology, 146, 1821–1833. Busch, C., Hofmann, F., Selzer, J., Munro, S., Jeckel, D., Aktories, K. (1998) A common motif of eukaryotic glycosyltransferases is essential for the enzyme activity of large clostridial cytotoxins. Journal of Biological Chemistry, 273, 19566–19572. Cabib, E., Roberts, R., Bowers, B. (1982) Synthesis of the yeast cell wall and its regulation. Annual Review of Biochemistry, 51, 763–769. Cabib, E., Roh, D.H., Schmidt, M., Crotti, L.B., Varma, A. (2001) The yeast cell wall and septum as paradigms of cell growth and morphogenesis. Journal of Biological Chemistry, 276, 19679–19682. Callaghan, T., Ross, P., Weinberger-Ohana, P., Benziman, M. (1988a) β-Glucoside activators of mung bean UDP-glucose:β-glucan synthase. II. Comparison of effects of an endogenous β-linked glucolipid with synthetic n-alkyl β-Dmonoglucopyranosides. Plant Physiology, 86, 1104–1107. Callaghan, T., Ross, P., Weinberger-Ohana, P., Garden, G., Benziman, M. (1988b) βGlucoside activators of mung bean UDP-glucose:β-glucan synthase. I. Identification of an endogenous β-linked glucolipid activator. Plant Physiology, 86, 1099–1103. Campbell, J.A., Davies, G.J., Bulone, V., Henrissat, B. (1997) A classification of nucleotide-diphospho-sugar glycosyltransferases based on amino acid sequence similarities. Biuochemical Journal, 326, 929–939. Cano-Delgado, A., Penfield, S., Smith, C., Catley, M., Bevan, M. (2003) Reduced cellulose synthesis invokes lignificatand defense responses in Arabidopsis thaliana. Plant Journal, 34, 351–362.

156

■

Plant Polysaccharides

Carter, G.R., Annau, E. (1953) Isolation of capsular polysaccharides from colonial variants of Pasteurella multocida. American Journal of Veterinary Research, 14, 475–485. Charnock, S.J., Davies, G.J. (1999) Structure of the nucleotide-diphospho-sugar transferase, SpsA from Bacillus subtilis, in native and nucleotide-complexed forms. Biochemistry, 38, 6380–6385. Charnock, S.J., Henrissat, B., Davies, G. (2001) Three-dimensional structures of UDPsugar glycosyltransferases illuminate the biosynthesis of plant polysaccharides. Plant Physiology, 125, 527–531. Chen, R., Bhagwat, A.A., Yaklich, R., Keister, D.L. (2002) Characterisation of nvdD, the third gene involved in the synthesis of cyclic-β-(1→3),(1→6)-d-glucans in Bradyrhizobium japonicum. Canadian Journal of Microbiology, 48, 1008–1016. Cocuron, J.C., Lerouxel, O., Drakakaki, G., et al. (2007) A gene from the cellulose synthase-like C family encodes a beta-1,4 glucan synthase. Proceedings of the National Academy of Sciences of the U S A, 104, 8550–8555. Colombani, A., Djerbi, S., Bessueille, L., et al. (2004) In vitro synthesis of (1→3)-betad-glucan (callose) and cellulose by detergent extracts of membranes from cell suspension cultures of hybrid aspen. Cellulose, 11, 313–327. Coutinho, P.M., Henrissat, B. (1999) Carbohydrate Active Enzymes server. http:// afmb.cnrs-mrs.fr/∼pedro/CAZY/db.html Cui, S.W. (2001a) Cereal non-starch polysaccharides I: (1→3),(1→4)-β-d-glucans. In: Polysaccharide Gums from Agricultural Products Processing, Structures and Functionality (ed. S.W. Cui). Technomic Publishing Co., Lancaster, PA. Cui, X., Shin, H., Song, C., Laosinchai, W., Amano, Y., Brown, R.M., Jr. (2001b) A putative plant homolog of the yeast β-1,3-glucan synthase subunit FKS1 from cotton (Gossypium hirsutum L.) fibers. Planta, 213, 223–230. Davies, G.J., Wilson, K.S., Henrissat, B. (1997) Nomenclature for sugar-binding subsites in glycosyl hydrolases. Biochemical Journal, 321, 557–559. DeAngelis, P.L. (1996) Enzymological characterization of the Pasteurella multocida hyaluronic acid synthase. Biochemistry, 35, 9768–9771. DeAngelis, P.L., Jing, W., Drake, R.R., Achyuthan, A.M. (1998) Identification and molecular cloning of a unique hyaluronan synthase from Pasteurella multocida. Journal of Biological Chemistry, 273, 8454–8458. DeAngelis, P.L., Padgett-McCue, A.J. (2000) Identification and molecular cloning of a chondroitin synthase from Pasteurella multocida type F. Journal of Biological Chemistry, 275, 24124–24129. DeAngelis, P.L., Weigel, P.H. (1994) Immunochemical confirmation of the primary structure of Streptococcal hyaluronan synthase and synthesis of high molecular weight product by the recombinant enzyme. Biochemistry, 33, 9033–9039. DeAngelis, P.L., White, C.L. (2001) Identification and molecular cloning of a heparosan synthase from Pasteurella multocida type D. Journal of Biological Chemistry, 277, 7209–7213. DeAngelis, P.L., White, C.L. (2004) Identification of a distinct, cryptic heparosan synthase from Pasteurella multocida Types A, D, and F. Journal of Bacteriology, 186, 8529–8532. DeBolt, S., Gutierrez, R., Ehrhardt, D.W., Somerville, C. (2007) Nonmotile cellulose synthase subunits repeatedly accumulate within localized regions at the plasma membrane in Arabidopsis hypocotyl cells following 2,6-dichlorobenzonitrile treatment. Plant Physiology, 145, 334–338.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

157

Delmer, D.P. (1999) Cellulose biosynthesis: Exciting times for a difficult field of study. Annual Review of Plant Physiology and Plant Molecular Biology, 50, 245–276. Desprez, T., Vernhettes, S., Fagard, M., et al. (2002) Resistance against herbicide isoxaben and cellulose deficiency caused by distinct mutations in same cellulose synthase isoform CESA6. Plant Physiology, 128, 482–490. Dhugga, K.S. (2001) Building the wall: genes and enzyme complexes for polysaccharide synthases. Current Opinion in Plant Biology, 4, 488–493. Dhugga, K.S., Barreiro, R., Whitten, B., et al. (2004) Guar seed beta-mannan synthase is a member of the cellulose synthase super gene family. Science, 303, 363–366. Dijkgraaf, G.J., Abe, M., Ohya, Y., Bussey, H. (2002) Mutations in FKS1p affect the cell wall content of beta-1,3- and beta-1,6-glucan in Saccharomyces cerevisiae. Yeast, 19, 671–690. Doblin, M.S., De Melis, L., Newbigin, E., Bacic, A., Read, S.M. (2001) Pollen tubes of Nicotiana alata express two genes from different beta-glucan synthase families. Plant Physiology, 125, 2040–2052. Doblin, M.S., Kurek, I., Jacob-Wilk, D., Delmer, D.P. (2002) Cellulose biosynthesis in plants: from genes to rosettes. Plant and Cell Physiology, 43, 1407–1420. Donofrio, N.M., Delaney, T.P. (2001) Abnormal callose response phenotype and hypersusceptibility to Peronospora parasitica in defense-compromised Arabidopsis nim1-1 and salicylate hydroxylase plants. Molecular Plant-Microbe Interactions, 14, 439–450. Douglas, C.M., Foor, F., Marrinan, J.A., et al. (1994) The Saccharomyces cerevisiae FKS1 (ETG1) gene encodes an integral membrane protein which is a subunit of 1,3-betad-glucan synthase. Proceedings of the National Academy of Sciences of the U S A, 91, 12907–12911. Eckart, M.R., Bussineau, C.M. (1996) Quality and authenticity of heterologous proteins synthesized in yeast. Current Opinion in Biotechnology, 7, 525–530. Endo, Y., Sawasaki, T. (2004) High-throughput, genome-scale protein production method based on the wheat germ cell-free expression system. Journal of Structural and Functional Genomics, 5, 45–57. Fagard, M., Desnos, T., Desprez, T., et al. (2000) PROCUSTE1 encodes a cellulose synthase required for normal cell elongation specifically in roots and dark-grown hypocotyls of Arabidopsis. Plant Cell, 12, 2409–2423. Fairweather, J.K., Him, J.L.K., Heux, L., Driguez, H., Bulone, V. (2004) Structural characterization by C-13-NMR spectroscopy of products synthesized in vitro by polysaccharide synthases using C-13-enriched glycosyl donors: application to a UDP-glucose:(1,3)-beta-d-glucan synthase from blackberry (Rubus fruticosus). Glycobiology, 14, 775–781. Farrokhi, N., Burton, R.A., Brownfield, L., et al. (2006) Plant cell wall biosynthesis: genetic, biochemical and functional genomics approaches to the identification of key genes. Plant Biotechnology Journal, 4, 145–167. Feingold, D.S., Neufeld, E.F., Hassid, W.Z. (1958) Synthesis of a β-(1,3)-linked glucan by extracts of Phaseolus aureus seedlings. Joutrnal of Biological Chemistry, 233, 783–788. Fincher, G.B., Stone, B.A. (2004) Chemistry of nonstarch polysaccharides. In: Encyclopedia of Grain Science (ed. C. Wrigley, , H. Corke, C.E. Walker). Elsevier, Oxford. Fraser, J.R., Laurent, T.C., Laurent, U.B. (1997) Hyaluronan: its nature, distribution, functions and turnover. Journal of Internal Medicine, 242, 27–33.

158

■

Plant Polysaccharides

Fredrikson, K., Kjellbom, P., Larsson, C. (1991) Isolation and polypeptide composition of 1,3-β-glucan synthase from plasma membranes of Brassica oleracea. Physiologia Plantarum, 81, 289–294. Futatsuyama, H., Yui, T., Ogawa, K. (1999) Viscometry of curdlan, a linear (1,3)-betad-glucan, in DMSO or alkaline solutions. Bioscience Biotechnology and Biochemistry, 63, 1481–1483. Gale, M.D., Devos, K.M. (1998) Plant comparative genetics after 10 years. Science, 282, 656–659. Gardiner, J.C., Taylor, N.G., Turner, S.R. (2003) Control of cellulose synthase complex localization in developing xylem. Plant Cell, 15, 1740–1748. Griffiths, G., Cook, N.J., Gottfridson, E., Lind, T., Lidholt, K., Roberts, I.S. (1998) Characterization of the glycosyltransferase enzyme from the Escherichia coli K5 capsule gene cluster and identification and characterization of the glucuronyl active site. Journal of Biological Chemistry, 273, 11752–11757. Grisshammer, R., Tate, C.G. (1995) Overexpression of integral membrane proteins for structural studies. Quarterly Review of Biophysics, 28, 315–422. Haigler, C., Brown, R.M. (1986) Transport of rosettes from the Golgi apparatus to the plasma membrane in isolated mesophyll cells of Zinnia elegans during differentiation to tracheary elements in suspension culture. Protoplasma, 134, 111–120. Han, F., Ullrich, S., Chirat, S., et al. (1995) Mapping of beta-glucan content and betaglucanase activity loci in barley grain and malt. Theoretical and Applied Genetics, 91, 921–927. Han, N.S., Robyt, J.F. (1998) The mechanism of Acetobacter xylinum cellulose biosynthesis: direction of chain elongation and the role of lipid pyrophosphate intermediates in the cell membrane. Carbohydrate Research, 313, 125–133. Harada, T. (1979) Curdlan: a gel-forming β-1,3-glucan. In: Polysaccharides in Food (eds J.M.V. Blanshard, J.R. Mitchell), pp 282–300. Butterworths, London. Harris, P.J. (2005) Diversity in plant cell walls. In: Plant Diversity and Evolution: Genotypic and Phenotypic Variation in Higher Plants (ed. R.J. Henry). CAB International, Wallingford, Oxon. Hazen, S., Scott-Craig, J., Walton, J. (2002) Cellulose synthase-like genes of rice. Plant Physiology, 128, 336–340. Heath, L.S., Ramakrishnan, N., Sederoff, R.R., et al. (2002) Studying the functional genomics of stress responses in loblolly pine with the Expresso microarray experiment management system. Comparative and Functional Genomics, 3, 226–243. Heiniger, U., Delmer, D.P. (1977) UDP-glucose: glucan synthetase in developing cotton fibers. II. Structure of the reaction product. Plant Physiology, 59, 719–723. Heldermon, C., DeAngelis, P.L., Weigel, P.H. (2001) Topological organization of the hyaluronan synthase from Streptococcus pyogenes. Journal of Biological Chemistry, 276, 2037–2046. Henry, R.J., Stone, B.A. (1982) Solubilization of β-glucan synthases from the membranes of cultured ryegrass endosperm cells. Biochemical Journal, 203, 629–36. Henry, R.J., Schibeci, A., Stone, B.A. (1983) Localization of β-glucan synthases on the membranes of cultured Lolium multiflorum (ryegrass) endosperm cells. Biochemical Journal, 209, 627–633. Høj, P.B., Fincher, G.B. (1995) Molecular evolution of plant β-glucan endohydrolases. Plant Journal, 7, 367–379. Holland, N., Holland, D., Helentjaris, T., Al, E. (2000) A comparative analysis of the plant cellulose synthase (CesA) gene family Plant Physiology, 123, 1313–1323.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

159

Honegger, R., Haisch, A. (2001) Immunocytochemical location of the (1→3) (1→4)-beta-glucan lichenin in the lichen-forming ascomycete Cetraria islandica (Icelandic moss). New Phytologist, 150, 739–746. Hong, Z.L., Delauney, A.J., Verma, D.P.S. (2001a) A cell plate specific callose synthase and its interaction with phragmoplastin. Plant Cell, 13, 755–768. Hong, Z.L., Zhang, Z.M., Olson, J.M., Verma, D.P.S. (2001b) A novel UDP-glucose transferase is part of the callose synthase complex and interacts with phragmoplastin at the forming cell plate. Plant Cell, 13, 769–779. Hrmova, M., Fincher, G.B. (2009) Functional genomics and structural biology in the definition of gene function. In: Methods in Molecular Biology: Plant Genomics (eds D. Somers, P. Langridge, P. Gustafson). Humana Press, Totowa, NJ. Imai, T., Watanabe, T., Yui, T., Sugiyama, J. (2003) The directionality of chitin biosynthesis: a revisit Biochemical Journal, 374, 755–760. Inon De Iannino, N., Ugalde, R.A. (1993) Biosynthesis of cyclic β-(1→3),β-(1→6) glucan in Bradyrhizobium spp. Archives of Microbiology, 159, 30–38. Inoue, S.B., Takewaki, N., Takasuka, T., et al. (1995) Characterization and gene cloning of 1,3- β-d-glucan synthase from Saccharomyces cerevisiae. European Journal of Biochemistry, 231, 845–854. Itano, N., Sawai, T., Yoshida, M., et al. (1999) Three isoforms of mammalian hyaluronan synthases have distinct enzymatic properties. Journal of Biological Chemistry, 274, 25085–25092. Jacobs, A.K., Lipka, V., Burton, R.A., et al. (2003) An Arabidopsis callose synthase, GSL5, is required for wound and papillary callose formation. Plant Cell, 15, 2503–2513. Jing, W., DeAngelis, P.L. (2000) Dissection of the two transferase activities of the Pasteurella multocida hyaluronan synthase: two active sites exist in one polypeptide. Glycobiology, 10, 883–889. Kamst, E., Bakkers, J., Quaedvlieg, N.E.M., et al. (1999) Chitin oligosaccharide synthesis by Rhizobia and zebrafish embryos starts by glycosyl transfer to O4 of the reducing-terminal residue. Biochemistry, 38, 4045–4052. Kapteyn, J.C., Ram, A.F.J., Groos, E.M., et al. (1997) Altered extent of cross-linking of beta 1,6-glucosylated mannoproteins to chitin in Saccharomyces cerevisiae mutants with reduced cell wall beta 1,3-glucan content. Journal of Bacteriology, 179, 6279–6284. Karnezis, T., Epa, V.C., Stone, B.A., Stanisich, V.A. (2003) Topological characterization of an inner membrane (13)-β-d-glucan (curdlan) synthase from Agrobacterium sp. strain ATCC 31749. Glycobiology, 13, 693–706. Karnezis, T., Fisher, H.C., Neumann, G.M., Stone, B.A., Stanisich, V.A. (2002) Cloning and characterization of the phosphatidylserine synthase gene of Agrobacterium sp.strain ATTC31749 and effect of its inactivation on production of high molecularmass (1→3)-β-d-glucan (curdlan). Journal of Bacteriology, 184, 4114–4123. Kass, E.H., Seastone, C.V. (1944) Role of the mucoid polysaccharide (hyaluronic acid) in the virulence of group A hemolytic streptococci. Journal of Experimental Medicine, 79, 319–330. Kim, C.M., Park, S.H., Je, B.I., et al. (2007) OsCSLD1, a cellulose synthase-like D1 gene, is required for root hair morphogenesis in rice. Plant Physiology, 143, 1220–1230. Klammt, C., Schwarz, D., Lohr, F., et al. (2006) Cell-free expression as an emerging technique for the large scale production of integral membrane protein. FEBS Journal, 273, 4141–4153.

160

■

Plant Polysaccharides

Knecht, J.C., Schiffman, G., Austrian, R. (1970) Biological properties of pneumococcus type 37 and the chemistry of its capsular polysaccharide. Journal of Experimental Medicine, 132, 475–487. Kollar, R., Reinhold, B.B., Petrakova, E., (1997) Architecture of the yeast cell wall beta(1,6)-glucan interconnects mannoprotein, beta(1–3)-glucan, and chitin. Journal of Biological Chemistry, 272, 17762–17775. Kota, R., Varshney, R.K., Prasad, M., Zhang, H., Stein, N., Graner, A. (2007) ESTderived single nucleotide polymorphism markers for assembling genetic and physical maps of the barley genome. Functional and Integrative Genomics, 8, 223–233. Koyama, M., Helbert, W., Imai, T., Sugiyama, J., Henrissat, B. (1997) Parallel-up structure evidences the molecular directionality during biosynthesis of bacterial cellulose. Proceedings of the Natlional Academy of Sciences of the U S A, 94, 9091–9095. Kudlicka, K., Brown, R.M. (1997) Cellulose and callose biosynthesis in higher plants. 1. Solubilization and separation of (1→3)- and (1→4)-beta-glucan synthase activities from mung bean. Plant Physiology, 115, 643–656. Kumari, K., Weigel, P.H. (1997) Molecular cloning, expression, and characterization of the authentic hyaluronan synthase from group C Streptococcus equisimilis. Journal of Biological Chemistry, 272, 32539–32546. Kurek, I., Kawagoe, Y., Jacob-Wilk, D., Doblin, M.S., Delmer, D.P. (2002) Dimerization of cotton fiber cellulose synthase catalytic subunits occurs via oxidation of the zinc-binding domains. Proceedings of the National Academy of Sciences of the USA, 99, 11109–11114. Kuribayashi, I., Kimura, S., Morita, T., Igaue, I. (1992) Characterization and solubilization of β-glucan synthases from cultured rice cells. Bioscience, Biotechnology, and Biochemistry, 56, 388–393. Kuribayashi, I., Morita, T., Mitsui, T., Igaue, I. (1993) Purification of β-glucan synthase II from suspension-cultured rice cells. Bioscience, Biotechnology, and Biochemistry, 57, 682–684. Lai-Kai-Him, J., Pelosi, L., Chanzy, H., Putaux, J.L., Bulone, V. (2001) Biosynthesis of (1→3)-beta-d-glucan (callose) by detergent extracts of a microsomal fraction from Arabidopsis thaliana. European Journal of Biochemistry, 268, 4628–4638. Lam, K.K.Y., Davey, M., Sun, B.M., Roth, A.F., Davis, N.G., Conibear, E. (2006) Palmitoylation by the DHHC protein Pfa4 regulates the ER exit of Chs3. Journal of Cell Biology, 174, 19–25. Legault, D.J., Kelly, R.J., Natsuka, Y., Lowe, J.B. (1995) Human alpha(1→3/1→4)fucosyl-transferases discriminate between different oligosaccharide acceptor substrates through a discrete peptide fragment. Journal of Biological Chemistry, 270, 20987–20996. Li, H., Bacic, A., Read, S.M. (1999) Role of a callose synthase zymogen in regulating wall deposition in pollen tubes of Nicotiana alata. Planta, 208, 528–538. Li, H.J., Bacic, A., Read, S.M. (1997) Activation of pollen tube callose synthase by detergents - Evidence for different mechanisms of action. Plant Physiology, 114, 1255–1265. Li, J., Burton, R.A., Harvey, A.J., et al. (2003) Biochemical evidence linking a putative callose synthase gene with (1,3)-beta-d-glucan biosynthesis in barley. Plant Molecular Biology, 53, 213–225.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

161

Liepman, A.H., Wilkerson, C.G., Keegstra, K. (2005) Expression of cellulose synthaselike (Csl) genes in insect cells reveals that CslA family members encode mannan synthases. Proceedings of the National Academy of Sciences of the U S A, 102, 2221–2226. Liepman, A.H., Nairn, C.J., Willats, W.G., Sørensen, I., Roberts, A.W., Keegstra, K. (2007) Functional genomic analysis supports conservation of function among cellulose synthase-like a gene family members and suggests diverse roles of mannans in plants. Plant Physiology, 143, 1881–1893. Malissard, M., Zeng, S., Berger, E.G. (1999) The yeast expression system for recombinant glycosyltransferases. Glycoconjugate Journal, 16, 125–139. Martínez, J.P., Gozalbo, D. (2001) Chitin. In: Encyclopedia of Life Sciences. John Wiley & Sons, New York. Marx-Figini, M. (1969) Biosynthesis of cellulose in higher and lower plants. Journal of Polymer Science, Polymer Symposia No. 28, 57–67. Matthysse, A.G., Thomas, D.L., White, A.R. (1995a) Mechanism of cellulose synthesis in Agrobacterium tumefaciens. Journal of Bacteriology, 177, 1076–1081. Matthysse, A.G., White, S., Lightfoot, R. (1995b) Genes required for cellulose synthesis in Agrobacterium tumefaciens. Journal of Bacteriology, 177, 1069–1075. McIntosh, M., Stone, B.A., Stanisich, V.A. (2005) Curdlan and other bacterial (1→3)-βd-glucans. Applied Microbiology and Biotechnology, 68,163–173. McLaren, J.S. (2005) Crop biotechnology provides an opportunity to develop a sustainable future. Trends in Biotechnology, 23, 339–342. Meikle, P.J., Ng, K.F., Johnson, E., Hoogenraad, N.J., Stone, B.A. (1991) The β-glucan synthase from Lolium multiflorum. Detergent solubilization, purification using monoclonal antibodies, and photoaffinity labeling with a novel photoreactive pyrimidine analog of uridine 5′-diphosphoglucose. Journal of Biological Chemistry, 266, 22569–22581. Merzendorfer, H. (2006) Insect chitin synthases: a review. Journal of Comparative Physiology B, 176, 1–15. Mueller, S., Brown, R., Jr. (1980) Evidence for an intramembranous component associated with a cellulose microfibril synthesizing complex in higher plants. Journal of Cell Biology, 84, 315–326. Nadolski, M.J., Linder, M.E. (2007) Protein lipidation. FEBS Journal, 274, 5202–5210. Nagahashi, S., Sudoh, M., Ono, N., et al. (1995) Characterization of chitin synthase 2 of Saccharomyces cerevisiae. Implication of two highly conserved domains as possible catalytic sites. Journal of Biological Chemistry, 270, 13961–13967. Newman, R.H., Davies, L.M., Harris, P.J. (1996) Solid-state 13C nuclear magnetic resonance characterization of cellulose in the cell walls of Arabidopsis thaliana leaves. Plant Physiology, 111, 475–485. Ng, K., Johnson, E., Stone, B.A. (1996) Specificity of binding of β-glucoside activators of ryegrass (1→3)-β-glucan synthase and the synthesis of some potential photoaffinity activators. Plant Physiology, 111, 1227–1231. Nishimura, M.T., Stein, M., Hou, B.H., Vogel, J.P., Edwards, H., Somerville, S.C. (2003) Loss of callose synthase results in salicylic acid-dependent disease resistance. Science, 310, 969–972. Notario, V., Gale, E.F., Kerridge, D., Wayman, F. (1982) Phenotypic resistance to amphotericin B in Candida albicans: relationship to glucan metabolism. Journal of General Microbiology, 128, 761–777.

162

■

Plant Polysaccharides

Ohana, P., Delmer, D.P., Steffens, J.C., Matthews, D.E., Mayer, R., Benziman, M. (1991) β-Furfuryl-β-glucoside. An endogenous activator of higher plant UDP-glucose: (1→3)-β-glucan synthase. Journal of Biological Chemistry, 266, 13742–13745. Ohana, P., Delmer, D.P., Volman, G., Steffens, J.C., Matthews, D.E., Benziman, M. (1992) Beta-furfuryl-beta-glucoside – an endogenous activator of higher-plant UDP-glucose – (1→3)-beta-glucan synthase – biological activity, distribution, and in vitro synthesis. Plant Physiology, 98, 708–715. Opekarová, M., Tanner, W. (2003) Specific lipid requirements of membrane proteins: a putative bottleneck in heterologous expression. Biochimica et Biophysica Acta, 1610, 11–22. Orlean, P.A.B. (1982) (1,3)-β-D-Glucan synthase from budding and filamentous cultures of the dimorphic fungus Candida albicans. European Journal of Biochemistry, 127, 397–403. Østergaard, L., Petersen, M., Mattson, O., Mundy, J. (2002) An Arabidopsis callose synthase. Plant Molecular Biology, 49, 559–566. Paredez, A.R., Somerville, C.R., Ehrhardt, D.W. (2006) Visualization of cellulose synthase demonstrates functional association with microtubules. Science, 312, 1491–1495. Pear, J., Kawagoe, Y., Schreckengost, W., Delmer, D., Stalker, D. (1996) Higher plants contain homologs of the bacterial celA genes encoding the catalytic subunit of cellulose synthase. Proceedings of the National Academy of Sciences of the U S A, 93, 12637–12642. Pelosi, L., Bulone, V., Heux, L. (2006) Polymorphism of curdlan and (1→3)-beta-dglucans synthesized in vitro: A C-13 CP-MAS and X-ray diffraction analysis. Carbohydrate Polymers, 66, 199–207. Pelosi, L., Imai, T.H.C., Heux, L.E.B., Bulone, V. (2003) Structural and morphological diversity of (1→3)-beta-d-glucans synthesized in vitro by enzymes from Saprolegnia monoica. Comparison with a corresponding in vitro product from blackberry (Rubus fruticosus). Biochemistry, 42, 6264–2003. Persson, S., Paredez, A., Carroll, A., et al. (2007) Genetic evidence for three unique components in primary cell-wall cellulose synthase complexes in Arabidopsis. Proceedings of the National Academy of Sciences of the U S A, 104, 15566–15571. Pfeffer, P.E., Osman, S.F., Hotchkiss, A., Bhagwat, A.A., Keister, D.L., Valentine, K.M. (1996) Cyclolaminarinose. A new biologically active β-(1→3) cyclic glucan. Carbohydrate Research, 296, 23–37. Pourkheirandish, M., Wicker, T., Stein, N., Fujimura, T., Komatsuda, T. (2007) The barley six-rowed spike gene vrs1 reveals a breakdown of rice-barley microcollinearity by a transposition. Theoretical and Applied Genetics, 114, 1357–1365. Prehm, P. (1983) Synthesis of hyaluronate in differentiated teratocarcinoma cells. Mechanism of chain growth. Biochemical Journal, 211, 191–198. Ray, P.M., Shininger, T.L., Ray, M.M. (1969) Isolation of β-glucan synthetase particles from plant cells and identification with Golgi membranes. Proceedings of the National Academy of Sciences of the U S A, 64, 605–612. Raymond, Y., Fincher, G.B., Maclachlan, G.A. (1978) Tissue slice and particulate β-glucan synthetase activities from Pisum epicotyls. Plant Physiology, 61, 938–942. Reeves, R.E., Goebel, W.F. (1941) Chemoimmunological studies on the soluble specific substance of pneumococcus. V. The structure of the type III polysaccharide. Journal of Biological Chemistry, 139, 511–519.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

163

Richmond, T., Somerville, C. (2000) The cellulose synthase superfamily. Plant Physiology, 124, 495–498. Rimler, R.B. (1994) Presumptive identification of Pasteurella multocida serogroups A, D and F by capsule depolymerisation with mucopolysaccharidases. Veterinary Record, 134, 191–192. Rodriguez, M.L., Jann, B., Jann, K. (1988) Structure and serological characteristics of the capsular K4 antigen of Escherichia coli O5:K4:H4, a fructose-containing polysaccharide with a chondroitin backbone. European Journal of Biochemistry, 177, 117–124. Römling, U. (2002) Molecular biology of cellulose production in bacteria. Research in Microbiology, 153, 205–212. Ross, P., Mayer, R., Benziman, M. (1991) Cellulose biosynthesis and function in bacteria. Microbiological Reviews, 55, 35–58. Roulin, S., Buchala, A.J., Fincher, G.B. (2002) Induction of (1→3,1→4)-β-glucan hydrolases in leaves of dark-incubated barley seedlings. Planta, 215, 51–59. Ryals, J.A., Neuenschwander, U.H., Willits, M.G., Molina, A., Steiner, H.Y., Hunt, M.D. (1996) Systemic Acquired Resistance. Plant Cell, 8, 1809–1819. Saxena, I., Brown, R. (1997) Identification of cellulose synthase(s) in higher plants: Sequence analysis of processive beta-glycosyltransferases with the common motif ‘D, D, D35Q(R,Q)XRW’. Cellulose, 33–49. Saxena, I., Brown, R. (2000) Cellulose synthases and related enzymes. Current Opinion in Plant Biology, 3, 523–531. Saxena, I., Brown, R., Fevre, M., Geremia, R., Henrissat, B. (1995) Multidomain architecture of beta-glycosyl transferases – Implications for mechanism of action. Journal of Bacteriology, 177, 1419–1424. Scheible, W.-R., Eshed, R., Richmond, T., Delmer, D., Somerville, C. (2001) Modifications of cellulose synthase confer resistance to isoxaben and thiazolidinone herbicides in Arabidopsis ixr1 mutants. Proceedings of the National Academy of Sciences of the U S A, 98, 10079–10084. Schlupmann, H., Bacic, A., Read, S.M. (1993) A novel callose synthase from pollen tubes of Nicotiana. Planta, 191, 470–481. Schmidt, H.A., Strimmer, K., Vingron, M., von Haeseler, A. (2002) TREE-PUZZLE: maximum likelihood phylogenetic analysis using quartets and parallel computing. Bioinformatics, 18, 502–504. Shematek, E.M., Cabib, E. (1980) Biosynthesis of the yeast cell wall. II. Regulation of β-(1,3)glucan synthetase by ATP and GTP. Journal of Biological Chemistry, 255, 895–902. Sietsma, J.H., Eveleigh, D.E., Haskins, R.H. (1969) Cell wall composition and protoplast formation of some oomycete species. Biochimica et Biophysica Acta, 184, 306–317. Silverman, S., Sburlati, A., Slater, M., Cabib, E. (1988) Chitin synthase 2 is essential for septum formation and cell division in Saccharomyces cerevisiae. Proceedings of the National Academy of Sciences of the U S A, 85, 4735–4739. Smart, M.G., Aist, J.R., Israel, H.W. (1986) Structure and function of wall appositions: 1. General histochemistry of papillae in barley coleoptiles attacked by Erisiphe graminis f-sp hordei. Canadian Journal of Botany, 64, 793–801. Smith, B.G., Harris, P.J., Melton, L.D., Newman, R.H. (1998) Crystalline cellulose in hydrated primary cell walls of three monocotyledons and one dicotyledon. Plant and Cell Physiology, 39, 711–720.

164

■

Plant Polysaccharides

Smith, M.M., Stone, B.A. (1973) β-Glucan synthesis in cell-free extracts from Lolium multiflorum endosperm. Biochimica et Biophysica Acta, 313, 72–94. Spyropoulos, I.C., Liakopoulos, T.D., Bagos, P.G., Hamodrakas S.J. (2004) TMRPres2D: high quality visual representation of transmembrane protein models. Bioinformatics, 20, 3258–3260. Stasinopoulos, S.J., Fisher, P.R., Stone, B.A., Stanisich, V.A. (1999) Detection of two loci involved in (1→3)-β-glucan (curdlan) biosynthesis by Agrobacterium sp. ATCC31749, and comparative sequence analysis of the putative curdlan synthase gene. Glycobiology, 9, 31–41. Stone, B.A. (1984) Noncellulosic β-glucans in cell walls. In: Structure, Function and Biosynthesis of Plant Cell Walls, Proceedings of the 7th Annual Symposium in Botany (eds W.M. Dugger, S. Bartnicki-Garcia). American Society of Plant Physiologists, Rockville, MD. Stone, B.A. (2005) Cellulose: Structure and distribution. In: Encyclopedia of Life Sciences. John Wiley & Sons, New York. Stone, B.A. (2006) Callose and related glucans. In: Encyclopedia of Life Sciences. John Wiley & Sons, New York. Stone, B.A., Clarke, A.E. (1992) The Chemistry and Biochemistry of (1,3)-β-Glucans. La Trobe University Press, Melbourne. Tanaka, K., Murata, K., Yamazaki, M., Onosato, K., Miyao, A., Hirochika, H. (2003) Three distinct rice cellulose synthase catalytic subunit genes required for cellulose synthesis in the secondary wall. Plant Physiology, 133, 73–83. Tarbouriech, N., Charnock, S.J., Davies, G.J. (2001) Three-dimensional structures of the Mn and Mg dTDP complexes of the family GT-2 glycosyltransferase SpsA: A comparison with related NDP-sugar glycosyltransferases. Journal of Molecular Biology, 314, 655–661. Taylor, N.G. (2007) Identification of cellulose synthase AtCesA7 (IRX3) in vivo phosphorylation sites – a potential role in regulating protein degradation. Plant Molecular Biology, 64, 161–171. Taylor, N., Howells, R., Huttly, A., Vickers, K., Turner, S. (2003) Interactions among three distinct CesA proteins essential for cellulose synthesis. Proceedings of the National Academy of Sciences of the U S A, 100, 1450–1455. Tellam, R.L., Vuocolo, T., Johnson, S.E., Jarmey, J., Pearson, R.D. (2000) Insect chitin synthase cDNA sequence, gene organization and expression. European Journal of Biochemistry, 267, 6025–6043. Tlapak-Simmons, V., Kempner, E., Baggenstoss, B., Weigel, P.H. (1998) The active streptococcal hyaluronan synthases (HASs) contain a single HAS monomer and multiple cardiolipin molecules. Journal of Biological Chemistry, 273, 26100–26109. Tlapak-Simmons, V.L., Baron, C.A., Gotschall, R., Haque, D., Canfield, W.M., Weigel, P.H. (2005) Hyaluronan biosynthesis by class I streptococcal hyaluronan synthases occurs at the reducing end. Journal of Biological Chemistry, 280, 13012–13018. Trethewey, J.A.K., Campbell, L.M., Harris, P.J. (2005) (1→3),(1→4)-β-D-Glucans in the cell walls of the Poales (sensu lato): An immunogold labeling study using a monoclonal antibody. American Journal of Botany, 92, 1669–1683. Tsekos, I. (1999) The sites of cellulose synthesis in algae: diversity and evolution of cellulose-synthesizing enzyme complexes. Journal of Phycology, 35, 635–655.

Biosynthesis of Polysaccharides by Enzymes of the GT2 and GT48 Families

■

165

Tsuchiya, K., Urahara, T., Konishi, T., et al. (2005) Biosynthesis of (1→3),(1→4)-betaglucan in developing endosperms of barley (Hordeum vulgare). Physiologia Plantarum, 125, 181–191. Tusnády, G.E., Simon, I. (2001) The HMMTOP transmembrane topology prediction server. Bioinformatics, 17, 849–850. Vann, W.F., Schmidt, M.A., Jann, B., Jann, K. (1981) The structure of the capsular polysaccharide (K5 antigen) of urinary-tract-infective Escherichia coli 010:K5:H4. A polymer similar to desulfo-heparin. European Journal of Biochemistry, 116, 359–364. Ventura, C.L., Cartee, R.T., Forsee, W.T., Yother, J. (2006) Control of capsular polysaccharide chain length by UDP-sugar substrate concentrations in Streptococcus pneumonia. Molecular Microbiology, 61, 723–733. Vergara, C., Carpita, N. (2001) Beta-D-glycan synthases and the CesA gene family: lessons to be learned from the mixed-linkage (1,3),(1,4)-beta-d-glucan synthase. Plant Molecular Biology, 47, 145–160. Verma, D.P.S., Hong, Z. (2001) Plant callose synthase complexes. Plant Molecular Biology, 47, 693–701. Veronico, P., Gray, L.J., Jones, J.T., et al. (2001) Nematode chitin synthases: gene structure, expression and function in Caenorhabditis elegans and the plant parasitic nematode Meloidogyne artiellia. Molecular Genetics and Genomics, 266, 28–34. Wightman, R., Turner, S.R. (2007) Severing at sites of microtubule crossover contributes to microtubule alignment in cortical arrays. Plant Journal, 52, 742–751. Wiggins, C.A., Munro, S. (1998) Activity of the yeast MNN1 alpha-1,3mannosyltransferase requires a motif conserved in many other families of glycosyltransferases. Proceedings of the National Academy of Sciences of the U S A, 95, 7945–7950. Williams, K.J., Halkes, K.M., Kamerling, J.P., DeAngelis, P.L. (2006) Critical elements of oligosaccharide acceptor substrates for the Pasteurella multocida hyaluronan synthase. Journal of Biological Chemistry, 281, 5391–5397. Williamson, R.E., Burn, J.E., Hocart, C.H. (2002) Towards the mechanism of cellulose synthesis. Trendsd in Plant Science, 7, 461–467. Wilson, S.M., Burton, R.A., Doblin, M.S., et al. (2006) Temporal and spatial appearance of wall polysaccharides during cellularization of barley (Hordeum vulgare) endosperm. Planta, 224, 655–667. Woodward, J.R., Phillips, D.R., Fincher, G.B. (1983a) Water-soluble (1→3),(1→4)-β-dglucans from barley (Hordeum vulgare) endosperm. I. Physiochemical properties. Carbohydrate Polymers, 3, 143–156. Woodward, J.R., Fincher, G.B., Stone, B.A. (1983b) Water-soluble (1,3;1,4)-β-glucans from barley (Hordeum vulgare) endosperm. II. Fine structure. Carbohydrate Polymers, 3, 207–225. Yarden, O., Yanofsky, C. (1991) Chitin synthase 1 plays a major role in cell wall biogenesis in Neurospora crassa. Genes and Development, 5, 2420–2430.

Annual Plant Reviews (2011) 41, 167–212 doi: 10.1002/9781444391015.ch6

Chapter 6

GLYCOSYLTRANSFERASES OF THE GT8 FAMILY Yanbin Yin1,3,4, Debra Mohnen2,3,4, Ivana Gelineo-Albersheim2,3, Ying Xu1,3,4 and Michael G. Hahn2,3,5 1

Computational Systems Biology Lab, Institute of Bioinformatics, The University of Georgia, Athens, GA 30602, USA 2 Complex Carbohydrate Research Center, The University of Georgia, 315 Riverbend Road, Athens, GA 30602, USA 3 BioEnergy Science Center, The University of Georgia, Athens, GA 30602, USA 4 Department of Biochemistry and Molecular Biology, The University of Georgia, Athens, GA 30602, USA 5 Department of Plant Biology, The University of Georgia, Athens, GA 30602, USA Manuscript received November 2009

Abstract: The higher plant genomes sequenced to date include numerous genes encoding proteins classified as belonging to CAZy family GT8. The large number and diversity of GT8 proteins in higher plants, which currently constitute more than 65% of the identified eukaryotic GT8 genes, highlight the importance of these proteins in plants. Here we summarize a detailed phylogenetic study of GT8 proteins from three monocot and four dicot plant genomes that clearly divides higher plant GT8 proteins into two distantly related sets of clades, many of which are further divided into statistically well-supported subclades. One set, the GAUT1 (GAlactUronosylTransferase1)-related family, includes the GAUT and GAUT-Like (GATL) proteins, comprising one proven galacturonsyltransferase and multiple additional members strongly implicated in the synthesis of pectins and xylan, two major types of polysaccharides present in plant cell walls. The second set, which includes Plant Glycogenin-like Starch Initiation Proteins (PGSIPs) and Galactinol Synthases (GolSs), appears not to be directly involved in plant cell wall synthesis. The PGSIPs have been suggested to play a role in priming starch biosynthesis, while the GolSs are key enzymes in the synthesis of

Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

167

168

■

Plant Polysaccharides

the raffinose family of oligosaccharides that play important roles in plant responses to environmental stress. This chapter also summarizes data from biochemical, transcriptional, and mutational studies that provide additional insights into possible functions of higher plant GT8 proteins. However, much work is needed to fully define the roles of GT8 proteins in plants, particularly with respect to their enzymatic function in plant cell wall biosynthesis. Keywords: biosynthesis; cell wall; dicot; galactinol synthase; galacturonosyltransferase; GATL; GAUT; glycosyltransferase; GolS; GT8; monocot; pectin; PGSIP; retaining; starch; xylan

6.1 Introduction Glycosyltransferase family 8 (GT8) of the CAZy Carbohydrate Active Enzymes database (Cantarel et al. 2009 and see Chapter 4) is presently a family of 973 predicted retaining glycosyltransferases (GTs) with 486 bacterial, 451 eukaryote, 34 viral and 2 unclassified members. Of the eukaryote GT8 members currently listed, 297 are from higher plants with 41 of these from Arabidopsis. The enzyme activities that have been confirmed biochemically for selected members of this family include multiple bacterial α-glycosyltransferases, many of which synthesize part of the core region of lipopolysaccharide (LPS), as well as three classes of eukaryotic enzymes: glycogenin glucosyltransferases and plant glycogenin-like proteins; galactinol synthases; and a galacturonosyltransferase involved in the synthesis of the plant cell wall pectic polysaccharide, homogalacturonan. As a group, the GT8 enzymes catalyse the transfer of diverse sugars (Glc, Gal, GlcNAC, GalA) onto lipooligosaccharide, protein, inositol, oligosaccharide or polysaccharide acceptors using nucleotide sugar substrates. After a very brief overview of bacterial GT8 enzymes, this chapter focuses on the eukaryotic enzymes with particular emphasis on the GT8 genes present in plants. The initial overview of bacterial GT8 enzymes, which are involved in lipopolysaccharide biosynthesis, is provided in part because the structure of one of these bacterial GT8 proteins has been determined by X-ray crystallography (see below). Lipopolysaccharides are amphipathic glycoconjugates located in the outer membrane of Gram-negative bacteria, but whose synthesis begins on the cytoplasmic side of the inner membrane and includes flipping and transport across the periplasm (Raetz & Whitfield 2002). LPS has a protective and permeability barrier function and also provides information for interaction of bacteria with other cells. In this regard, LPS can also serve as a virulence factor. LPS consists of a hydrophobic so-called lipid A domain, a nonrepeating core domain, and a distal polysaccharide domain such as O-specific polysaccharide, enterobacterial common antigen (ECA) or capsular polysaccharide. Many of the bacterial GT8 enzymes, which include α1,2 and

Glycosyltransferases of the GT8 Family

■

169

α1,3 glucosyltransferases (GlcTs) and galactosyltransferases (GalTs), synthesize part of the core region of LPS. The LPS core is an oligosaccharide region that contains up to 15 glycosyl residues (Holst 2007). There is structural variation between the core oligosaccharide regions of LPS from different bacteria. Examples of GT8 LPS core GTs include the lipopolysaccharide glucosyltransferase 1 (EC 2.4.1.58) (also known as lipopolysaccharide glucosyltransferase I and UDPglucose:(heptosyl)lipopolysaccharide 1,3-glucosyltransferase) which transfers Glc from UDP-Glc onto the nonreducing end heptose of some LPS inner cores (Müller et al. 1972). Another example is the bacterial lipopolysaccharide (LPS) α1,3-galactosyltransferase (EC 2.4.1.44) which catalyses the transfer of Gal from UDP-Gal on to Glc in the partially completed outer LPS core (Holst 2007). The eukarotic GT8 proteins are divided into three groups based on the glycan synthesized. Proteins from each of these groups are found in plants. These three groups of enzymes are: 1 Glycogenin glucosyltransferase (EC 2.4.1.186) which, in glycogenproducing organisms such as mammals, is an α-glucosyltransferase that first glucosylates itself at a Tyr residue and then elongates the Glc to form an oligosaccharide of 5–13 α1,4-linked glucosyl resides attached to the protein glycogenin (Hurley et al. 2005). This oligosaccharide primer is then further glycosylated by glycogen synthase and branching enzyme to yield glycogen, an energy storage polysaccharide. 2 Inositol 1-α-galactosyltransferase (galactinol synthase) (EC 2.4.1.123) is another class of eukaryotic GT8 GTs which galactosylates myoinositol to form UDP and O-α-d-galactopyranosyl-[1→1]-l-myo-inositol, also known as galactinol. Galactinol serves as the donor for stachyose and raffinose synthesis. As such, galactinol synthase represents the first unique enzyme in the biosynthetic pathway that leads to these oligosaccharides, which are osmoprotectants in specialized plant tissues and serve as sources of energy during the germination of some seeds (Saravitz et al. 1987). Sequential transfer of αGal from galactinol on to sucrose yields the trisaccharide (raffinose: αGal-1,6-αGlc-1,2-βFru) or the tetrasaccharide (stachyose: αGal-1,6-αGal-1,6-αGlc-1,2-βFru). 3 The largest of the eukaryotic GT8 groups consists of plant cell wall biosynthetic glycosyltransferases belonging to the GAUT1-related gene family. Only one protein in this group has had its activity biochemically verified and that is the pectin biosynthetic enzyme homogalacturonan α1,4-gal acturonosyltransferase 1 (EC 2.4.1.43) (GAUT1) (Sterling et al. 2001). As a group the GT8 enzymes are predicted to be retaining glycosyltransferases and to have a GT-A fold, on the basis of information gained from the only two family members having a resolved crystal structure: UDPGal:α1,4-galactosyltransferase (LgtC) from Neisseria meningitides, which catalyses the transfer of d-Gal from UDP-Gal on to the Gal in a terminal

170

■

Plant Polysaccharides

lactose portion of LPS in Neisseria (Persson et al. 2001), and rabbit muscle glycogenin (Gyg), the self-glucosylating initiation protein that synthesizes the oligosaccharide primer for glycogen synthesis (Gibbons et al. 2002). Like all retaining glycosyltransferases, the GT8 GTs synthesize glycoconjugates with an α-anomeric configuration when d-nucleotide sugars (or other activated sugars with an α linkage) are used as donor substrates. Of course, retaining enzymes will synthesize β-linked glycoconjugates when l-nucleotide sugars are the donor substrates.

6.2 Phylogeny of family GT8 A detailed phylogenetic analysis of the complete CAZy Family GT8 has been carried out using genes identified from fully sequenced organisms utilizing the Pfam GT8 domain model (PF01501; http://pfam.sanger.ac.uk/ family?entry=PF01501) as the probe (Yin et al. 2010). We summarize the results of those analyses here as they pertain to GT8 proteins identified from those higher plants for which complete genome sequences have been published. An E-value cutoff of 1e−2 was used as a threshold to identify genes encoding GT8 family members. This threshold resulted in the identification of all 41 GT8 proteins previously identified in Arabidopsis (Sterling et al. 2006) with no false positives identified. The plants used for the analysis reported here include three monocots (Oryza sativa (Yu et al. 2002; Goff et al. 2002), Sorghum bicolor (Paterson et al. 2009), Brachypodium distachyon (http://www.phytozome.net/) and four dicots (Arabidopsis thaliana (Arabidopsis Genome Initiative 2000), Populus trichocarpa (Tuskan et al. 2006), Vitis vinifera (Jaillon et al. 2007), Glycine max (http://www. phytozome.net/)). During the analysis it became clear that some proteomes, such as those of Arabidopsis and rice, contain protein variants that arise from alternative splicing. For such genes, we kept only one protein sequence for each gene. The plant GT8 family members identified in these screens are listed in Table 6.1, together with useful information about each gene and its encoded protein, including the plant species, Gene ID, GT8 clade, clade subfamily (where identifiable), the number of ESTs found in GenBank, the presence of signal peptide (as predicted by SignalP v3.0; Bendtsen et al. 2004)), the number of transmembrane domains (as predicted by TMHMM v2.0; Krogh et al. 2001)), and the protein length. All listed proteins contain the Glyco_ transf_8 (PF01501) Pfam domain. The EST matches were found by a BLAST search of the full-length coding region sequences of the plant GT8 proteins against the GenBank EST sequences obtained from PlantGDB (Dong et al. 2005). Using an E-value cutoff of ≤1e−2, ESTs more than 98% identical to the query were kept and numerated. These results provide evidence that at least 75% of the GT8 proteins listed are expressed in at least one plant tissue.

XP_002279893 NP_191438 NP_200280 XP_002319802

AT5G54690.1 (AtGAUT12)

estExt_fgenesh4_pm.C_ LG_XIII0357 eugene3.00111083 GSVIVP00027429001 Glyma12g34280.1 Glyma13g36280.1 Glyma06g41630.1 Glyma12g16550.1 estExt_Genewise1Plus.C_ chr_85936 LOC_Os12g38930.1 Bradi4g03670.1

rice (os) Brachypodium (bd)

poplar (pt) grape (vv) soybean (gm) soybean (gm) soybean (gm) soybean (gm) sorghum (sb) NP_001067123 NA

XP_002317564 CAO39049 NA NA NA NA XP_002443426

NA NA NA XP_002321071

Glyma08g42280.1 Glyma14g03110.1 Glyma02g45720.1 estExt_Genewise1Plus.C_ LG_XIV2539 GSVIVP00018928001 AT3G58790.1 (AtGAUT15)

grape (vv) Arabidopsis (at) Arabidopsis (at) poplar (pt)

XP_002456278 NP_001044121 NA

Sb03g033400 LOC_Os01g52710.1 Bradi2g48710.1

sorghum (sb) rice (os) Brachypodium (bd) soybean (gm) soybean (gm) soybean (gm) poplar (pt)

GenBankID

GeneID

Species

GAUT GAUT

GAUT GAUT GAUT GAUT GAUT GAUT GAUT

GAUT

GAUT

GAUT GAUT

GAUT GAUT GAUT GAUT

GAUT GAUT GAUT

GT8 clade

GAUT-a GAUT-a

GAUT-a GAUT-a GAUT-a GAUT-a GAUT-a GAUT-a GAUT-a

GAUT-a

GAUT-a

GAUT-a GAUT-a

GAUT-a GAUT-a GAUT-a GAUT-a

GAUT-a GAUT-a GAUT-a

GT8 subclade

8 0

0 2 2 6 1 4 6

3

4

1 7

15 0 0 0

1 17 0

No. of ESTs

N N

N N N N N N N

N

N

N Y

N N N N

N N Y

Signal peptide

1 1

1 1 1 1 1 1 1

1

1

1 1

1 1 0 1

0 0 0

No. of transmembrane regions

555 536

533 533 534 534 534 534 536

534

536

528 541

526 525 446 532

474 537 539

Protein length (a.a.)

Table 6.1 Family GT8 genes identified in seven sequenced higher plant genomes. All encoded proteins contain the Glyco_transf_8 (PF01501) Pfam domain

Glycosyltransferases of the GT8 Family ■

171

XP_002465655 ABF94604 CAO15050 NP_197051 NP_186753 XP_002324094 XP_002305374 NA NA NA NA XP_002438104 BAD37314 NA NP_001048110 XP_002454284 NA XP_002282423 XP_002332802 XP_002329246

estExt_fgenesh1_pg.C_ chr_14427 LOC_Os03g11330.1 GSVIVP00021044001 AT5G15470.1 (AtGAUT14)

AT3G01040.1 (AtGAUT13)

eugene3.00170460 eugene3.00041059 Glyma08g26480.1 Glyma18g49960.1 Glyma13g05950.1 Glyma19g03460.1 fgenesh1_pg.C_ chr_10000882 LOC_Os06g12280.1 Bradi1g45210.1

LOC_Os02g51130.1 e_gw1.4.16023.1 Bradi3g59370.1

GSVIVP00038116001 eugene3.03090007 gw1.123.42.1

rice (os) Brachypodium (bd) rice (os) sorghum (sb) Brachypodium (bd) grape (vv) poplar (pt) poplar (pt)

rice (os) grape (vv) Arabidopsis (at) Arabidopsis (at) poplar (pt) poplar (pt) soybean (gm) soybean (gm) soybean (gm) soybean (gm) sorghum (sb)

NA

Bradi1g70290.1

Brachypodium (bd) sorghum (sb)

GenBankID

GeneID

GAUT GAUT GAUT

GAUT GAUT GAUT

GAUT GAUT

GAUT GAUT GAUT GAUT GAUT GAUT GAUT

GAUT

GAUT GAUT GAUT

GAUT

GAUT

GT8 clade

GAUT-b GAUT-b GAUT-b

GAUT-b GAUT-b GAUT-b

GAUT-b GAUT-b

GAUT-a GAUT-a GAUT-a GAUT-a GAUT-a GAUT-a GAUT-b

GAUT-a

GAUT-a GAUT-a GAUT-a

GAUT-a

GAUT-a

GT8 subclade

1 2 2

15 0 0

32 0

0 0 1 1 2 1 0

13

1 3 13

3

2

No. of ESTs

N N N

Y Y Y

Y Y

N N N N N N Y

N

N N N

N

N

Signal peptide

1 1 0

1 1 1

1 1

1 1 1 1 1 1 1

1

1 1 1

1

1

No. of transmembrane regions

543 565 504

494 493 508

505 501

529 529 539 540 535 535 505

534

578 533 533

489

564

Protein length (a.a.)

■

Species

Table 6.1 Continued

172 Plant Polysaccharides

XP_002273962 XP_002453869 NP_001046899 NA XP_002447180

GSVIVP00010370001 Sb04g020140 LOC_Os02g29530.1 Bradi3g43810.1

estExt_fgenesh1_pg.C_ chr_62314 Bradi5g23250.1

Arabidopsis (at) soybean (gm) soybean (gm) Brachypodium (bd) sorghum (sb)

poplar (pt) NP_565485 NA NA NA XP_002465134

Glyma13g37650.1 Glyma12g32820.1 Bradi1g60010.1

estExt_fgenesh1_pm.C_ chr_10735

XP_002319923

NP_001054014 XP_002279062 XP_002329358

LOC_Os04g54360.1 GSVIVP00026021001 estExt_ Genewise1Plus.C_1200026 fgenesh4_pm.C_LG_ XIII000435 AT2G20810.1 (AtGAUT10)

NA

NA NA XP_002310951 NA ACU20809 XP_002301803 NP_189150

Glyma19g05060.1 Glyma13g06990.1 eugene3.00080075 Glyma18g33210.1 Glyma08g46210.1 eugene3.00002521 AT3G25140.1 (AtGAUT8)

Brachypodium (bd) rice (os) grape (vv) poplar (pt)

NP_566170

AT3G02350.1 (AtGAUT9)

Arabidopsis (at) soybean (gm) soybean (gm) poplar (pt) soybean (gm) soybean (gm) poplar (pt) Arabidopsis (at) grape (vv) sorghum (sb) rice (os) Brachypodium (bd) sorghum (sb)

GenBankID

GeneID

Species

GAUT-c

GAUT-c GAUT-c GAUT-c

GAUT-c

GAUT-c

GAUT-c GAUT-c GAUT-c

9

14 6 2

35

0

9 0 0

0

5

GAUT-c GAUT-c

15 7 15 0

4 4 0 42 43 5 64

39

No. of ESTs

GAUT-b GAUT-b GAUT-b GAUT-b

GAUT-b GAUT-b GAUT-b GAUT-b GAUT-b GAUT-b GAUT-b

GAUT-b

GT8 subclade

N

N N N

N

N

N N N

N

Y

N Y Y Y

Y N Y N N Y N

N

Signal peptide

1

1 1 1

1

1

1 1 1

1

1

0 1 1 1

1 1 1 0 1 1 1

1

No. of transmembrane regions

544

534 534 540

537

535

557 533 535

566

556

137 535 534 524

553 553 555 509 557 555 560

562

Protein length (a.a.)

■

GAUT

GAUT GAUT GAUT

GAUT

GAUT

GAUT GAUT GAUT

GAUT

GAUT

GAUT GAUT GAUT GAUT

GAUT GAUT GAUT GAUT GAUT GAUT GAUT

GAUT

GT8 clade

Glycosyltransferases of the GT8 Family

173

XP_002330135 XP_002318580 NA NA NA NA NA XP_002328269 NP_195540 CAO47521 XP_002439065 NA ABB47337 BAD61814 NP_182171 NP_191672 XP_002272447

eugene3.01290051 grail3.0138001201 Glyma05g07410.1 Glyma17g08910.1 Glyma06g22730.1 Glyma04g31770.1 Glyma05g09200.1 eugene3.00660198 AT4G38270.1 (AtGAUT3)

GSVIVP00036268001 estExt_Genewise1.C_ chr_108333 Bradi1g29780.1

LOC_Os10g21890.1 LOC_Os06g51160.1 AT2G46480.1 (AtGAUT2)

AT3G61130.1 (AtGAUT1)

GSVIVP00026306001

Brachypodium (bd) rice (os) rice (os) Arabidopsis (at) Arabidopsis (at) grape (vv)

NP_001050360 CAO62433 CAO62431 NP_564057

LOC_Os03g30000.1 GSVIVP00000008001 GSVIVP00000005001 AT1G18580.1 (AtGAUT11)

rice (os) grape (vv) grape (vv) Arabidopsis (at) poplar (pt) poplar (pt) soybean (gm) soybean (gm) soybean (gm) soybean (gm) soybean (gm) poplar (pt) Arabidopsis (at) grape (vv) sorghum (sb)

GenBankID

GeneID

GAUT

GAUT

GAUT GAUT GAUT

GAUT

GAUT GAUT

GAUT GAUT GAUT GAUT GAUT GAUT GAUT GAUT GAUT

GAUT GAUT GAUT GAUT

GT8 clade

GAUT-d

GAUT-d

GAUT-d GAUT-d GAUT-d

GAUT-d

GAUT-d GAUT-d

GAUT-c GAUT-c GAUT-c GAUT-c GAUT-c GAUT-c GAUT-d GAUT-d GAUT-d

GAUT-c GAUT-c GAUT-c GAUT-c

GT8 subclade

14

41

17 2 1

0

3 1

0 1 0 5 5 4 9 0 10

6 4 2 19

No. of ESTs

N

N

N Y N

N

N N

N N N N N N N Y Y

N N N N

Signal peptide

0

1

1 0 0

0

0 0

1 0 0 1 1 1 0 0 0

1 1 0 1

No. of transmembrane regions

228

674

687 602 529

1060

618 513

532 490 474 537 535 535 585 656 681

542 499 211 538

Protein length (a.a.)

■

Species

Table 6.1 Continued

174 Plant Polysaccharides

rice (os) Brachypodium (bd) rice (os) sorghum (sb) Brachypodium (bd) soybean (gm) soybean (gm)

Brachypodium (bd) rice (os) sorghum (sb)

sorghum (sb) Brachypodium (bd) rice (os) rice (os) sorghum (sb)

BAD10126 NA NP_001061555 XP_002444200 NA NA NA

LOC_Os08g23780.1 Sb07g014890 Bradi3g20550.1

Glyma17g00790.1 Glyma07g40020.1

NP_001063487 XP_002454735

NA

NP_001063757 NA XP_002460421

XP_002462717 NA

XP_002466644

XP_002302550 NA NA NA NA EEC81302 NA

XP_002320745

GenBankID

LOC_Os09g30280.1 fgenesh1_pg.C_ chr_4003251 LOC_Os08g38740.1 Bradi3g39270.1

LOC_Os09g36190.1 LOC_Os09g36180.1 estExt_Genewise1Plus.C_ chr_27961 Bradi4g33280.1

estExt_Genewise1Plus.C_ chr_12326 Sb02g030820 Bradi4g36050.1

estExt_Genewise1Plus.C_ LG_XIV0504 eugene3.00021408 Glyma09g40260.1 Glyma18g45750.1 Glyma03g02250.1 Glyma07g08910.1 LOC_Os06g49810.1 Bradi1g12180.1

poplar (pt)

poplar (pt) soybean (gm) soybean (gm) soybean (gm) soybean (gm) rice (os) Brachypodium (bd) sorghum (sb)

GeneID

Species

GAUT-e GAUT-e

GAUT-e GAUT-e GAUT-e

GAUT-e GAUT-e

GAUT-e GAUT-e

GAUT-e

GAUT-d GAUT-d GAUT-e

GAUT-d GAUT-d

GAUT-d

GAUT-d GAUT-d GAUT-d GAUT-d GAUT-d GAUT-d GAUT-d

GAUT-d

GT8 subclade

4 20

51 3 1

2 0

8 0

0

23 3 0

2 0

3

0 13 12 9 14 23 1

0

No. of ESTs

N N

Y Y Y

Y Y

Y N

Y

N N Y

N N

Y

N Y N Y N Y Y

N

Signal peptide

0 0

1 1 1

1 1

1 0

1

1 1 1

1 1

1

1 1 0 2 0 0 0

1

No. of transmembrane regions

399 399

644 649 661

727 704

708 371

697

698 1081 705

684 692

589

645 665 607 845 613 589 590

688

Protein length (a.a.)

■

GAUT GAUT

GAUT GAUT GAUT

GAUT GAUT

GAUT GAUT

GAUT

GAUT GAUT GAUT

GAUT GAUT

GAUT

GAUT GAUT GAUT GAUT GAUT GAUT GAUT

GAUT

GT8 clade

Glycosyltransferases of the GT8 Family

175

XP_002329942 XP_002308736 XP_002451032 NA ABA94533 NP_563771 NP_850150 XP_002282102 XP_002307628 NA NA NA NA NA CAO45341 NP_565893 XP_002323701

eugene3.01370031 fgenesh4_pg.C_LG_ VI000014 e_gw1.5.15262.1 Bradi4g14910.1

LOC_Os11g37980.1 AT1G06780.1 (AtGAUT6)

AT2G30575.1 (AtGAUT5)

GSVIVP00023858001 eugene3.00051260 fgenesh4_pg.C_LG_ II000411 Glyma19g34420.1 Glyma03g31590.1 Glyma02g15990.1 Glyma10g03770.1 GSVIVP00033610001 AT2G38650.1 (AtGAUT7)

e_gw1.XVI.562.1

soybean (gm) soybean (gm) soybean (gm) soybean (gm) grape (vv) Arabidopsis (at) poplar (pt)

sorghum (sb) Brachypodium (bd) rice (os) Arabidopsis (at) Arabidopsis (at) grape (vv) poplar (pt) poplar (pt)

NA NA CAO23686 NP_568688

Glyma15g12900.1 Glyma09g01980.1 GSVIVP00011843001 AT5G47780.1 (AtGAUT4)

soybean (gm) soybean (gm) grape (vv) Arabidopsis (at) poplar (pt) poplar (pt)

GenBankID

GeneID

GAUT

GAUT GAUT GAUT GAUT GAUT GAUT

GAUT GAUT GAUT

GAUT

GAUT GAUT

GAUT GAUT

GAUT GAUT

GAUT GAUT GAUT GAUT

GT8 clade

GAUT-g

GAUT-f GAUT-f GAUT-f GAUT-f GAUT-g GAUT-g

GAUT-f GAUT-f GAUT-f

GAUT-f

GAUT-f GAUT-f

GAUT-f GAUT-f

GAUT-e GAUT-e

GAUT-e GAUT-e GAUT-e GAUT-e

GT8 subclade

0

21 17 6 7 3 27

6 3 0

5

1 15

0 0

4 0

7 14 22 46

No. of ESTs

N

Y Y N Y N N

Y N N

Y

N N

N Y

Y Y

Y Y Y Y

Signal peptide

1

1 1 1 1 0 1

1 1 0

1

1 1

0 1

1 1

1 1 1 1

No. of transmembrane regions

621

626 626 576 586 452 620

588 606 365

611

549 590

434 537

666 649

658 658 613 617

Protein length (a.a.)

■

Species

Table 6.1 Continued

176 Plant Polysaccharides

Brachypodium (bd) rice (os)

Brachypodium (bd) Brachypodium (bd) rice (os) Brachypodium (bd) sorghum (sb) sorghum (sb)

rice (os) Brachypodium (bd) rice (os) sorghum (sb) rice (os) rice (os) sorghum (sb) NA NA BAD45664 NA XP_002436791 XP_002452580 NA NP_001048067

Bradi4g44070.1

LOC_Os06g13760.1 Bradi1g44470.1

Sb10g008830 estExt_fgenesh1_pg.C_ chr_42380 Bradi3g59760.1

LOC_Os02g50600.1

AAT01328 XP_002441280 EEC68769 NP_001065619 NA

NP_001060656 NA

XP_002461225

NA NA NA XP_002465323 NP_001050007 NA

XP_002326255

GenBankID

LOC_Os05g40720.1 Sb09g023760 LOC_Os12g02910.1 LOC_Os11g03160.1 fgenesh1_pm.C_ chr_8000106 Bradi3g61120.1

estExt_Genewise1.C_ chr_211179 LOC_Os07g48370.1 Bradi1g17570.1

estExt_ Genewise1Plus.C_281089 Glyma09g40610.1 Glyma18g45230.1 Glyma16g09420.1 Sb01g036430 LOC_Os03g21250.1 Bradi1g63520.1

poplar (pt)

soybean (gm) soybean (gm) soybean (gm) sorghum (sb) rice (os) Brachypodium (bd) sorghum (sb)

GeneID

Species

GATL-a

GATL-a

GATL-a GATL-a

GATL-a GATL-a

GAUT-g

GAUT-g

GAUT-g GAUT-g GAUT-g GAUT-g GAUT-g

GAUT-g GAUT-g

GAUT-g

GAUT-g GAUT-g GAUT-g GAUT-g GAUT-g GAUT-g

GAUT-g

GT8 subclade

8

2

0 15

2 0

0

0

2 0 2 3 1

7 0

4

4 3 0 2 2 0

0

No. of ESTs

Y

Y

Y Y

Y Y

Y

N

Y Y N N N

Y Y

Y

N N N Y Y Y

N

Signal peptide

1

1

1 1

1 0

1

0

2 1 1 1 0

374

380

368 427

372 358

633

565

668 639 660 643 581

626 622

628

1 1 1

563 658 245 629 632 626

591

Protein length (a.a.)

0 1 0 1 1 1

1

No. of transmembrane regions

■

GATL

GATL

GATL GATL

GATL GATL

GAUT

GAUT

GAUT GAUT GAUT GAUT GAUT

GAUT GAUT

GAUT

GAUT GAUT GAUT GAUT GAUT GAUT

GAUT

GT8 clade

Glycosyltransferases of the GT8 Family

177

NP_564983 NP_173827 NP_001049861 NA XP_002468004

AT1G24170.1 (AtGATL8)

LOC_Os03g18890.1 Bradi1g64830.1

estExt_Genewise1.C_ chr_19676 GSVIVP00028063001 Glyma10g01960.1 Glyma02g01880.1 Glyma19g40180.1 Glyma03g37560.1 eugene3.00021805 grail3.0061016301

grape (vv) soybean (gm) soybean (gm) soybean (gm) soybean (gm) poplar (pt) poplar (pt)

Arabidopsis (at) Arabidopsis (at) rice (os) Brachypodium (bd) sorghum (sb) CAO40071 NA NA NA NA XP_002302739 XP_002320324

XP_002311789

ACU19924 NA CAO40973 NA NA XP_002311853

Glyma19g01910.1 Glyma13g04780.1 GSVIVP00028923001 Glyma02g03090.1 Glyma01g04460.1 fgenesh4_pm.C_LG_ VIII000888 estExt_Genewise1Plus.C_ LG_VIII2617 AT1G70090.1 (AtGATL9)

poplar (pt)

NP_189474

AT3G28340.1 (AtGATL10)

Arabidopsis (at) soybean (gm) soybean (gm) grape (vv) soybean (gm) soybean (gm) poplar (pt)

GenBankID

GeneID

GATL GATL GATL GATL GATL GATL GATL

GATL

GATL GATL

GATL

GATL

GATL

GATL GATL GATL GATL GATL GATL

GATL

GT8 clade

GATL-b GATL-b GATL-b GATL-b GATL-b GATL-b GATL-b

GATL-b

GATL-b GATL-b

GATL-a

GATL-a

GATL-a

GATL-a GATL-a GATL-a GATL-a GATL-a GATL-a

GATL-a

GT8 subclade

6 45 30 46 37 4 2

3

24 0

21

36

0

4 7 9 3 5 0

10

No. of ESTs

Y Y Y Y Y Y Y

Y

Y Y

Y

Y

Y

Y Y Y Y Y Y

Y

Signal peptide

1 0 0 1 1 1 1

0

0 0

1

340 360 358 347 347 368 369

372

369 368

394

391

379

1 1

382 382 381 379 379 384

366

Protein length (a.a.)

1 1 0 1 1 0

1

No. of transmembrane regions

■

Species

Table 6.1 Continued

178 Plant Polysaccharides

NP_171772 NA XP_002466674

NP_190645 NA NA CAO22912 NP_564077 ABK94510 XP_002302469

AT1G02720.1 (AtGATL5)

Bradi1g12560.1

estExt_Genewise1.C_ chr_12425 LOC_Os03g47530.1 LOC_Os10g31650.1 GSVIVP00014689001 estExt_fgenesh4_pm.C_ LG_VII0375 fgenesh4_pg.C_ scaffold_57000218 AT3G50760.1 (AtGATL2)

Glyma01g38520.1 Glyma02g06640.1 GSVIVP00025008001 AT1G19300.1 (AtGATL1)

eugene3.00400186 estExt_Genewise1Plus.C_ LG_II1880 Glyma06g03770.1 Glyma04g03690.1 Glyma17g36650.1 Glyma14g08430.1 LOC_Os04g44850.1 e_gw1.6.14698.1

soybean (gm) soybean (gm) soybean (gm) soybean (gm) rice (os) sorghum (sb)

Arabidopsis (at) soybean (gm) soybean (gm) grape (vv) Arabidopsis (at) poplar (pt) poplar (pt)

poplar (pt)

NA NA NA NA NP_001053393 XP_002446838

XP_002327710

ABF98189 AAP54070 CAO61226 XP_002310780

NP_192122

AT4G02130.1 (AtGATL6)

rice (os) rice (os) grape (vv) poplar (pt)

NP_191825

AT3G62660.1 (AtGATL7)

Arabidopsis (at) Arabidopsis (at) Arabidopsis (at) Brachypodium (bd) sorghum (sb)

GenBankID

GeneID

Species

GATL GATL GATL GATL GATL GATL

GATL GATL

GATL GATL GATL GATL

GATL

GATL

GATL GATL GATL GATL

GATL

GATL

GATL

GATL

GATL

GT8 clade

GATL-c GATL-c GATL-c GATL-c GATL-d GATL-d

GATL-c GATL-c

GATL-c GATL-c GATL-c GATL-c

GATL-c

GATL-c

GATL-c GATL-c GATL-c GATL-c

GATL-c

GATL-c

GATL-b

GATL-b

GATL-b

GT8 subclade

4 4 2 2 10 0

4 1

5 1 2 28

4

0

24 0 3 0

1

0

12

23

58

No. of ESTs

Y N Y Y Y Y

Y Y

Y N Y Y

Y

Y

Y N N Y

N

Y

Y

Y

Y

Signal peptide

1 0 1 1 0 0

1 0

1 0 1 0

1

1

1 0 0 0

0

1

0

0

0

No. of transmembrane regions

367 319 353 362 342 343

361 352

352 333 340 352

342

353

361 265 186 353

292

358

362

347

362

Protein length (a.a.)

Glycosyltransferases of the GT8 Family ■

179

GATL GATL GATL GATL GATL GATL GATL PGSIP-A PGSIP-A

XP_002281658 XP_002314884 XP_002312381 NA NA NP_001060465 XP_002461136 XP_002465236 NP_001050151 NA XP_002277334 XP_002271296 XP_002311936 XP_002315420 NP_187277 NA NA XP_002280832 NP_175891

GSVIVP00030290001 grail3.0006037101 grail3.0010045901 Glyma02g11100.1 Glyma01g22480.1 LOC_Os07g45260.1 fgenesh1_pm.C_ chr_2000674 e_gw1.1.15754.1 LOC_Os03g24510.1 Bradi1g61830.1

GSVIVP00019872001 GSVIVP00023650001 grail3.0090012501 grail3.0106011001 AT3G06260.1 (AtGATL4)

Glyma07g38430.1 Glyma17g02330.1 GSVIVP00022838001 AT1G54940.1 (AtPGSIP4)

sorghum (sb) rice (os) Brachypodium (bd) grape (vv) grape (vv) poplar (pt) poplar (pt) Arabidopsis (at) soybean (gm) soybean (gm) grape (vv) Arabidopsis (at)

NP_563925

AT1G13250.1 (AtGATL3)

GATL GATL GATL

GATL GATL GATL GATL GATL GATL GATL

GATL

GATL

NA

Bradi5g16690.1

Brachypodium (bd) Arabidopsis (at) grape (vv) poplar (pt) poplar (pt) soybean (gm) soybean (gm) rice (os) sorghum (sb)

GenBankID

GeneID

GT8 clade

1 1 0 0

2 1 0 0 2

GATL-e GATL-e GATL-e GATL-e GATL-e GATL-e GATL-e

0 7 0

6 0 0 12 6 10 0

14

0

No. of ESTs

GATL-e GATL-e GATL-e

GATL-d GATL-d GATL-d GATL-d GATL-d GATL-e GATL-e

GATL-d

GATL-d

GT8 subclade

Y Y Y N

Y Y Y Y Y

N Y Y

Y Y Y Y Y Y Y

Y

Y

Signal peptide

1 1 1 1

0 0 0 0 1

0 1 1

0 0 0 1 0 1 0

1

0

No. of transmembrane regions

351 347 546 558

352 356 348 349 352

310 348 352

345 343 347 343 339 378 335

346

343

Protein length (a.a.)

■

Species

Table 6.1 Continued

180 Plant Polysaccharides

rice (os) sorghum (sb) Brachypodium (bd) rice (os)

grape (vv) Arabidopsis (at) soybean (gm) soybean (gm) soybean (gm) sorghum (sb) Brachypodium (bd) Brachypodium (bd) sorghum (sb)

XP_002282762 NP_195059 NA NA NA XP_002458785 NA NA XP_002453984 BAD15458 XP_002441128 NA NP_001044991

Glyma04g04080.1 Glyma14g09070.1 Glyma17g36100.1 e_gw1.3.18262.1 Bradi5g27680.1

Bradi3g45800.1

fgenesh1_pg.C_ chr_4001767 LOC_Os02g35020.1 Sb09g020930 Bradi2g24740.1

LOC_Os01g65780.1

AAK92624 XP_002468380 XP_002326938

LOC_Os03g08600.1 Sb01g044930 estExt_Genewise1_ v1.C_400718 GSVIVP00014811001 AT4G33330.1 (AtPGSIP3)

XP_002319544 NA NA NA NA

XP_002328440

fgenesh4_pg.C_ scaffold_70000240 e_gw1.XIII.1015.1 Glyma10g29570.1 Glyma19g42380.1 Glyma03g39820.1 Bradi1g72350.1

poplar (pt) soybean (gm) soybean (gm) soybean (gm) Brachypodium (bd) rice (os) sorghum (sb) poplar (pt)

NP_172373

AT1G08990.1 (AtPGSIP5)

Arabidopsis (at) poplar (pt)

GenBankID

GeneID

Species

GT8 subclade

88

6 18 0

0

0

4 0 1 0 0

1 4

22 3 0

0 2 0 0 1

0

1

No. of ESTs

N

N N N

N

Y

Y N N N N

N Y

Y Y N

N N N Y N

N

N

Signal peptide

1

636

655 632 634

645

0 0 1 1

612

588 598 593 630 581

590 597

615 606 361

426 541 517 434 608

431

567

Protein length (a.a.)

0

1 1 1 1 1

1 1

1 1 0

0 0 0 0 1

0

1

No. of transmembrane regions

■

PGSIP-A

PGSIP-A PGSIP-A PGSIP-A

PGSIP-A

PGSIP-A

PGSIP-A PGSIP-A PGSIP-A PGSIP-A PGSIP-A

PGSIP-A PGSIP-A

PGSIP-A PGSIP-A PGSIP-A

PGSIP-A PGSIP-A PGSIP-A PGSIP-A PGSIP-A

PGSIP-A

PGSIP-A

GT8 clade

Glycosyltransferases of the GT8 Family

181

XP_002456736 CAO60879 XP_002307480 NP_177838 NA NA NA XP_002275240 NP_566615 XP_002307022 XP_002310513 NA NA NA NA NA NP_001047475 NA CAE05448 XP_002446789

estExt_Genewise1.C_ chr_310243 GSVIVP00014288001 e_gw1.V.184.1 AT1G77130.1 (AtPGSIP2)

Glyma02g40480.1 Glyma14g28370.1 Glyma0214s00200.1 GSVIVP00034844001 AT3G18660.1 (AtPGSIP1)

eugene3.00050013 eugene3.00070393 Glyma06g15690.1 Glyma04g39240.1 Glyma05g32370.1 Glyma05g04770.1 Glyma08g15640.1 LOC_Os02g41520.1 Bradi3g49200.1

LOC_Os04g43700.1 estExt_fgenesh1_pg.C_ chr_61566

grape (vv) poplar (pt) Arabidopsis (at) soybean (gm) soybean (gm) soybean (gm) grape (vv) Arabidopsis (at) poplar (pt) poplar (pt) soybean (gm) soybean (gm) soybean (gm) soybean (gm) soybean (gm) rice (os) Brachypodium (bd) rice (os) sorghum (sb)

NA

Bradi2g56810.1

Brachypodium (bd) sorghum (sb)

GenBankID

GeneID

PGSIP-B PGSIP-B

PGSIP-A PGSIP-A PGSIP-A PGSIP-A PGSIP-A PGSIP-A PGSIP-A PGSIP-B PGSIP-B

PGSIP-A PGSIP-A PGSIP-A PGSIP-A PGSIP-A

PGSIP-A PGSIP-A PGSIP-A

PGSIP-A

PGSIP-A

GT8 clade

GT8 subclade

2 14

2 0 1 0 5 0 0 7 4

9 1 1 1 10

6 0 32

12

3

No. of ESTs

Y Y

N N N N N N N Y Y

N N N N N

N N N

N

N

Signal peptide

4 6

0 0 0 1 1 1 0 5 7

1 1 1 1 0

0 1 1

1

1

No. of transmembrane regions

480 537

545 631 537 627 641 628 483 548 1189

645 542 591 636 656

652 630 619

634

633

Protein length (a.a.)

■

Species

Table 6.1 Continued

182 Plant Polysaccharides

NA NP_197349 XP_002266145 XP_002319122 NA NA NA NA XP_002269578 NA NA NP_565817 NP_193393 NP_001053509 XP_002448294 NA NP_001065345 XP_002467515 NA XP_002280616 XP_002262651 NP_176250 NP_176248 XP_002265947 XP_002311774

Bradi5g15780.1

AT5G18480.1 (AtPGSIP6)

GSVIVP00002209001 eugene3.00130489 Glyma03g40980.1 Glyma19g43630.1 Glyma20g37000.1 Glyma10g30700.1 GSVIVP00009599001 Glyma11g03550.1 Glyma05g04630.1 AT2G35710.1 (AtPGSIP7)

AT4G16600.1 (AtPGSIP8)

LOC_Os04g46750.1 Sb06g024740 Bradi5g17880.1

LOC_Os10g40640.1 Sb01g029400 Bradi3g33080.1

GSVIVP00033193001 GSVIVP00019669001 AT1G60470.1 (AtGolS4)

AT1G60450.1 (AtGolS7)

GSVIVP00002727001 grail3.0009037801

Brachypodium (bd) Arabidopsis (at) grape (vv) poplar (pt) soybean (gm) soybean (gm) soybean (gm) soybean (gm) grape (vv) soybean (gm) soybean (gm) Arabidopsis (at) Arabidopsis (at) rice (os) sorghum (sb) Brachypodium (bd) rice (os) sorghum (sb) Brachypodium (bd) grape (vv) grape (vv) Arabidopsis (at) Arabidopsis (at) grape (vv) poplar (pt)

GenBankID

GeneID

Species

GT8 subclade

153 82

4

25 27 10

23 3 0

12 12 0

0

14 3 8 10 14 1 24 1 4 45

43

0

No. of ESTs

N N

N

N N N

Y Y Y

N Y Y

Y

Y Y N Y Y Y Y N Y Y

Y

Y

Signal peptide

0 0

0

0 0 0

6 6 5

5 6 6

5

5 5 5 6 5 5 5 5 6 6

5

6

No. of transmembrane regions

336 339

333

318 317 335

493 487 488

429 476 467

495

546 547 485 553 542 537 473 432 478 498

538

545

Protein length (a.a.)

■

GolS GolS

GolS

GolS GolS GolS

PGSIP-C PGSIP-C PGSIP-C

PGSIP-C PGSIP-C PGSIP-C

PGSIP-C

PGSIP-B PGSIP-B PGSIP-B PGSIP-B PGSIP-B PGSIP-B PGSIP-C PGSIP-C PGSIP-C PGSIP-C

PGSIP-B

PGSIP-B

GT8 clade

Glycosyltransferases of the GT8 Family

183

NP_001060697 XP_002461242 NA XP_002467954 NP_001049939 NA XP_002314975

NP_850902 NP_567741 NP_197768 XP_002330017 XP_002319473

Sb01g037090 LOC_Os03g20120.1 Bradi1g64120.1

fgenesh4_pm.C_LG_ X000590 fgenesh4_pm.C_LG_ VIII000417 AT5G30500.1(AtGolS8)

AT4G26250.1(AtGolS6)

AT5G23790.1(AtGolS5)

gw1.131.196.1 estExt_fgenesh4_pm.C_ LG_XIII0025 e_gw1.XIII.549.1 GSVIVP00022562001

poplar (pt) grape (vv)

Arabidopsis (at) Arabidopsis (at) Arabidopsis (at) poplar (pt) poplar (pt)

poplar (pt)

rice (os) sorghum (sb) Brachypodium (bd) sorghum (sb) rice (os) Brachypodium (bd) poplar (pt)

XP_002319472 CAO17390

XP_002312306

XP_002314613

estExt_fgenesh4_pg.C_LG_ X0618 LOC_Os07g48830.1 e_gw1.2.20933.1 Bradi1g17200.1

poplar (pt)

GenBankID

GeneID

GolS GolS

GolS GolS

GolS

GolS

GolS

GolS

GolS

GolS GolS GolS

GolS GolS GolS

GolS

GT8 clade

GT8 subclade

1 0

1 6

0

0

0

2

4

14 55 0

38 5 1

5

No. of ESTs

N N

N N

N

N

N

N

N

Y N N

N N N

N

Signal peptide

0 0

0 0

0

0

0

0

0

0 0 0

0 0 0

0

No. of transmembrane regions

335 332

306 338

334

337

329

326

326

350 342 338

329 322 345

339

Protein length (a.a.)

■

Species

Table 6.1 Continued

184 Plant Polysaccharides

XP_002320958 XP_002301531 NA NA XP_002279114 XP_002279157 NP_176053 NP_172406 NA AAM96867 NA NA

eugene3.00140617 estExt_fgenesh4_pm.C_ LG_II0906 Glyma19g40680.1 Glyma03g38080.1 GSVIVP00028165001 GSVIVP00028167001 AT1G56600.1 (AtGolS2)

AT1G09350.1 (AtGolS3)

Glyma20g22700.1 Glyma10g28610.1 Glyma03g38910.1 Glyma19g41550.1

soybean (gm) soybean (gm) grape (vv) grape (vv) Arabidopsis (at) Arabidopsis (at) soybean (gm) soybean (gm) soybean (gm) soybean (gm)

XP_002281304 XP_002281369 XP_002281261 CAO17387 NP_182240

GSVIVP00022565001 GSVIVP00022570001 GSVIVP00022561001 GSVIVP00022559001 AT2G47180.1 (AtGolS1)

grape (vv) grape (vv) grape (vv) grape (vv) Arabidopsis (at) poplar (pt) poplar (pt)

GenBankID

GeneID

Species

GolS GolS GolS GolS

GolS

GolS GolS GolS GolS GolS

GolS GolS

GolS GolS GolS GolS GolS

GT8 clade

GT8 subclade

11 21 5 11

18

5 7 19 9 9

2 1

8 13 6 2 28

No. of ESTs

N N N N

N

N N N N N

N N

Y Y Y Y N

Signal peptide

0 0 0 0

0

0 0 0 0 0

0 0

0 0 0 0 0

No. of transmembrane regions

325 329 332 331

335

336 340 342 340 336

337 338

325 324 325 162 345

Protein length (a.a.)

Glycosyltransferases of the GT8 Family ■

185

186

■

Plant Polysaccharides

GATL PGSIP-A 100

1 100

100

GAUT

63 100

Arabidopsis

100

Poplar

65

Grape Soybean

100

Rice Sorghum Brachypodium

GolS 100

PGSIP-B PGSIP-C

Figure 6.1 The phylogeny of higher plant GT8 proteins. The multiple sequence alignment of 319 full-length plant GT8 protein sequences (Table 6.1) was used to perform a maximum likelihood (ML) phylogeny reconstruction, to which bootstrap analyses were applied 100 times. The statistical bootstrap values are shown beside the branches to show the confidence levels with regard to the clustering of relevant proteins into each group. The colour scheme used for the branches is as follows: Arabidopsis, dark blue; poplar, green; grape, aqua; soybean, yellow; rice, red; sorghum, brown; Brachypodium, purple.

The full-length GT8 protein sequences identified from the screen of fully sequenced higher plant genomes were aligned using MAFFT v6.603 (Katoh et al. 2005) and the resulting alignment was used to perform maximum likelihood phylogeny reconstruction using PhyML v2.4.4 (Guindon & Gascuel 2003). The phylogeny of plant GT8 proteins shown in Fig. 6.1 suggests that there are six monophyletic plant GT8 clades that are distributed into two widely divergent groups having only distant evolutionary relatedness. One group contains the GAUT and GATL proteins, some of which have been linked functionally to cell wall polysaccharide biosynthesis. The other group contains the PGSIP and GolS proteins, some of which have been linked to starch and raffinose or stachyose oligosaccharide synthesis, respectively. Detailed phylogenetic studies of GT8 domains from all sequenced organisms suggests that the two broad groups of plant GT8 proteins each arose from distinct ancestral bacterial genes (Yin et al. 2010). A more detailed discussion of each of these groups of proteins follows later in this chapter. Each of the monophyletic clades identified in the phylogenetic reconstruction of the GT8 protein sequences was subjected to further individual analysis to discern subclade structures within each clade (Figs 6.2–6.7). The full-length protein sequences for each clade were realigned using MAFFT, and maximum likelihood phylogenies were built using PhyML as previously. The results of these studies demonstrate that most of the clades can

0.1 100 100

94

100 100

100 71 67

100

100

100

60 100 30 100 100

74 100 99

65 100 91 54

66

100

100 100

38 100 86 100

100 74

100

100 52

100

100

19

100 52 31 100 100

21 89 70 43

100

49

100

92

100

63 98 100 47 46 100 100

100

100

100 99

81

58 100

77

100 100 100

33 100 27

54 100

58

100

77

100

98 70 25

99 45

100 100 100 28 100 100 64

88

100 100

82

100 99

91

91 80 100

100 100 42 95 95

100

59 69 100

65

100

100

96 100 63

100

100 41

67

100 100 100 89 65 100

40

100

100 100

100

100

100

exon

intron 0 1 2: intron phase

50 99 56

97

100

100

100 50 97

0 0 0 sb/Sb03g033400 10 0 0 0 os/LOC Os01g52710.1 1 0 0 0 0 bd/Bradi2g48710.1 1 0 0 gm/Glyma08g42280.1 0 0 0 0 2 gm/Glyma14g03110.1 0 0 0 gm/Glyma02g45720.1 0 0 0 pt/estExt Genewise1Plus.C LG XIV2539 0 0 0 0 2 vv/GSVIVP00018928001 0 0 0 0 2 AtGAUT15/AT3G58790.1 0 0 02 AtGAUT12/AT5G54690.1 0 0 0 02 pt/estExt fgenesh4 pm.C LG XIII0357 0 0 0 02 pt/eugene3.00111083 0 0 0 0 2 vv/GSVIVP00027429001 1 0 0 0 0 gm/Glyma12g34280.1 0 0 0 02 gm/Glyma13g36280.1 0 0 0 02 gm/Glyma06g41630.1 1 0 0 0 0 gm/Glyma12g16550.1 0 0 0 02 sb/estExt Genewise1Plus.C chr 85936 0 0 0 02 os/LOC Os12g38930.1 1 0 0 0 0 bd/Bradi4g03670.1 1 0 0 0 bd/Bradi1g70290.1 1 0 0 0 0 sb/estExt fgenesh1 pg.C chr 14427 0 0 0 02 os/LOC Os03g11330.1 0 0 0 vv/GSVIVP00021044001 0 0 0 02 AtGAUT14/AT5G15470.1 1 0 0 0 0 AtGAUT13/AT3G01040.1 0 0 0 02 pt/eugene3.00170460 1 0 0 0 0 pt/eugene3.00041059 0 0 0 02 gm/Glyma08g26480.1 0 0 0 02 gm/Glyma18g49960.1 0 0 0 02 gm/Glyma13g05950.1 1 0 0 0 0 gm/Glyma19g03460.1 0 sb/fgenesh1 pg.C chr 10000882 0 os/LOC Os06g12280.1 0 bd/Bradi1g45210.1 0 os/LOC Os02g51130.1 0 sb/e gw1.4.16023.1 0 bd/Bradi3g59370.1 0 0 vv/GSVIVP00038116001 0 0 pt/eugene3.03090007 0 pt/gw1.123.42.1 0 0 AtGAUT9/AT3G02350.1 0 0 gm/Glyma19g05060.1 0 0 gm/Glyma13g06990.1 0 0 pt/eugene3.00080075 0 gm/Glyma18g33210.1 0 0 gm/Glyma08g46210.1 0 0 pt/eugene3.00002521 0 0 AtGAUT8/AT3G25140.1 0 vv/GSVIVP00010370001 0 0 sb/Sb04g020140 0 0 os/LOC Os02g29530.1 0 0 bd/Bradi3g43810.1 0 sb/estExt fgenesh1 pg.C chr 62314 0 bd/Bradi5g23250.1 0 os/LOC Os04g54360.1 0 vv/GSVIVP00026021001 0 pt/estExt Genewise1Plus.C 1200026 0 pt/fgenesh4 pm.C LG XIII000435 0 AtGAUT10/AT2G20810.1 0 gm/Glyma13g37650.1 0 gm/Glyma12g32820.1 0 bd/Bradi1g60010.1 0 sb/estExt fgenesh1 pm.C chr 10735 0 os/LOC Os03g30000.1 2 0 vv/GSVIVP00000008001 0 vv/GSVIVP00000005001 0 AtGAUT11/AT1G18580.1 0 pt/eugene3.01290051 pt/grail3.0138001201 gm/Glyma05g07410.1 0 gm/Glyma17g08910.1 0 gm/Glyma06g22730.1 0 gm/Glyma04g31770.1 0 1 0 0 gm/Glyma05g09200.1 0 1 0 0 00 pt/eugene3.00660198 0 200 0 0 2 0 AtGAUT3/AT4G38270.1 0 1 01 vv/GSVIVP00036268001 0 1 sb/estExt Genewise1.C chr 108333 1 0 1 0 1 0 0 0 2 bd/Bradi1g29780.1 0 20 0 0 2 0 os/LOC Os10g21890.1 0 1 0 0 os/LOC Os06g51160.1 0 2 0 1 0 AtGAUT2/AT2G46480.1 1 0 0 0 10 2 0 0 AtGAUT1/AT3G61130.1 0 vv/GSVIVP00026306001 0 0 1 0 20 0 00 pt/estExt Genewise1Plus.C LG XIV0504 0 0 1 0 20 2 2 pt/eugene3.00021408 1 0 0 0 10 2 0 0 gm/Glyma09g40260.1 2 0 1 00 01 gm/Glyma18g45750.1 10 00 1 0 2 0 0 10 2 0 gm/Glyma03g02250.1 11 2 0 1 1 gm/Glyma07g08910.1 1 010 2 0 0 os/LOC Os06g49810.1 0 0 1 0 2 02 bd/Bradi1g12180.1 0 0 1 0 2 02 sb/estExt Genewise1Plus.C chr 12326 0 0 1 02 0 02 sb/Sb02g030820 0 0 1 02 0 02 bd/Bradi4g36050.1 0 0 1 0 2 0 02 os/LOC Os09g36190.1 1 0 1 1 0 0 0 os/LOC Os09g36180.1 2 1 0 2 0 0 sb/estExt Genewise1Plus.C chr 27961 11 0 2 0 0 bd/Bradi4g33280.1 11 0 2 0 0 os/LOC Os09g30280.1 0 0 sb/fgenesh1 pg.C chr 4003251 1 00 2 0 2 0 0 os/LOC Os08g38740.1 00 0 2 0 0 bd/Bradi3g39270.1 1 00 1 0 2 0 0 os/LOC Os08g23780.1 100 1 0 2 0 0 sb/Sb07g014890 0 0 1 0 2 002 bd/Bradi3g20550.1 0 0 gm/Glyma17g00790.1 0 0 gm/Glyma07g40020.1 10 0 1 0 2 0 0 gm/Glyma15g12900.1 10 0 1 0 2 0 0 gm/Glyma09g01980.1 1 00 1 0 2 0 0 vv/GSVIVP00011843001 100 1 0 2 0 0 AtGAUT4/AT5G47780.1 10 0 1 0 2 0 0 pt/eugene3.01370031 0 0 1 0 2 00 2 pt/fgenesh4 pg.C LG VI000014 0 0 1 sb/e gw1.5.15262.1 1 00 2 0 0 bd/Bradi4g14910.1 0 0 1 0 02 os/LOC Os11g37980.1 1 0 0 10 2 0 0 AtGAUT6/AT1G06780.1 0 0 1 0 2 00 2 AtGAUT5/AT2G30575.1 10 0 1 0 2 0 vv/GSVIVP00023858001 0 0 1 0 2 00 2 pt/eugene3.00051260 22 2 02 0 pt/fgenesh4 pg.C LG II000411 0 0 1 0 2 00 2 gm/Glyma19g34420.1 0 0 1 0 2 00 2 gm/Glyma03g31590.1 10 0 1 2 0 0 gm/Glyma02g15990.1 0 0 1 2 00 2 gm/Glyma10g03770.1 0 0 1 vv/GSVIVP00033610001 0 0 1 02 0 002 AtGAUT7/AT2G38650.1 0 0 1 2 pt/e gw1.XVI.562.1 1 2 0 0 pt/estExt Genewise1Plus.C 281089 22 2 02 2 0 0 gm/Glyma09g40610.1 2 2 0 1 1 2 gm/Glyma18g45230.1 1 gm/Glyma16g09420.1 1 00 1 0 2 0 0 sb/Sb01g036430 0 0 1 0 2 0 02 os/LOC Os03g21250.1 1 0 0 1 0 2 0 0 bd/Bradi1g63520.1 1 0 0 1 0 2 0 0 sb/estExt Genewise1.C chr 211179 1 00 1 0 2 0 0 os/LOC Os07g48370.1 0 0 1 0 2 0 02 bd/Bradi1g17570.1 0 0 1 0 2 0 0 0 02 os/LOC Os05g40720.1 0 0 1 0 20 0 02 sb/Sb09g023760 0 0 1 0 2 0 0 02 os/LOC Os12g02910.1 0 0 1 0 2 0 02 os/LOC Os11g03160.1 0 10 2 0 0 sb/fgenesh1 pm.C chr 8000106 0 0 1 0 2 bd/Bradi3g61120.1 1 0 0 1 0 2 0 0 bd/Bradi4g44070.1 5'

0kb

1kb

2kb

3kb

4kb

5kb

0

0

0

2

GAUT-a 0

0 2

GAUT-b

GAUT-c

0

GAUT-d 0

0

1

1 1 0 0

00

2

GAUT-e

GAUT-f

0

1

0

2

GAUT-g

6kb

3'

7kb

8kb

17kb

Figure 6.2 The phylogeny of the GAUT clade and structures of the GAUT genes. This ML phylogeny was built based on the multiple sequence alignment of full length GAUT protein sequences identified from fully sequenced higher plant genomes. The bootstrap values are shown to indicate the confidence level of each grouping. Subclades are named in the ‘GAUT-alphabet’ format according to the phylogenetic grouping and gene structure conservation. The colour scheme for the branches is the same as that used in Fig. 6.1. The gene structure was plotted using the GSDS server (Guo et al. 2007) with intron– exon coordinate information acquired from the downloaded annotation files of the seven genomes. The intron–exon structure for each gene is shown on the right. The intron phase indicates the position of the intron within a codon. If it is not located within a codon the phase is 0. If it is located within a codon and after the first base of the codon the phase is 1. If it is located within a codon and after the second base of the codon, the phase is 2.

188

■

Plant Polysaccharides

be further classified into distinct well-supported subfamilies. The only exceptions were the PGSIP-A and GolS clades, where the sequences available from the fully sequenced plant genomes do not provide sufficient resolution to clearly discern the subclade structure of the clade. All of the subclades in the cell wall biosynthesis-related clades (GAUT and GATL) correspond to different monophyletic clusters with significant bootstrap support values greater than or equal to 70%. Each of the GAUT and GATL subclades contain sequences from each of the seven plant genomes, indicating that the proteins in each subclade are orthologous to each other. This result is not surprising, considering that all plants need to synthesize walls and hence their polysaccharide constituents. The subclade structures of the clades not related to cell wall synthesis (PGSIP and GolS) are more complex. Not all subclades in the PGSIP and GolS groups contained representative proteins from all seven plant genomes examined, and some subclades lacked either monocot or dicot representatives. These results suggest that monocots and dicots diverged in their needs for, and uses of, enzymes involved in starch and raffinose biosynthesis, the annotated functions of the PGSIP and GolS families, respectively.

6.3 GT8 clades related to plant cell wall polysaccharide synthesis Representatives of each of two clades (GAUT and GATL) in family GT8 were initially implicated in plant cell wall synthesis based on altered wall phenotypes exhibited by plants carrying mutations in these genes (Bouton et al. 2002; Lao et al. 2003; Shao et al. 2004). Subsequently, AtGAUT1 was demonstrated to functionally encode a homogalacturonan galacturonosyltransferase (Sterling et al. 2006) involved in pectic polysaccharide synthesis. Additional mutational and biochemical studies on members of the GAUT and GATL clades further supported the hypothesis that members of these two clades are involved in plant cell wall polysaccharide biosynthesis (Caffall et al. 2009; Y. Kong, G. Zhou, Y. Yin, Y. Zhu, S. Pattathil, M.G. Hahn, submitted for publication). Bioinformatic analyses of the entire GT8 family also identified amino acid sequence motifs that are unique to the GAUT and GATL clades among the family GT8 clades (Sterling et al. 2006; Yin et al. 2010), further substantiating the functional separation of the GAUT and GATL from other clades within the GT8 family. 6.3.1 Galacturonosyltransferase (GAUT) clade The GAUT clade consists of genes encoding proven and putative glycosyltransferases involved in plant cell wall polysaccharide biosynthesis. The GAUT proteins were first identified as being involved in cell wall polysaccharide synthesis in Arabidopsis through studies of a plant carrying a

Glycosyltransferases of the GT8 Family

■

189

mutation in the QUASIMODO1/GAUT8 gene (Bouton et al. 2002). Biochemical proof of a role of this family of proteins in cell wall synthesis was independently obtained by enzyme purification, cloning and functional expression studies that led to the identification of Arabidopsis GAUT1 as a pectin biosynthetic homogalacturonan galacturonosyltransferase (Sterling et al. 2006). The proteins encoded by almost all GAUT genes are type II membrane proteins carrying a single transmembrane domain. 6.3.1.1 Phylogeny of GAUT clade Initial phylogenetic studies of the GAUT family carried out in Arabidopsis thaliana identified three clades of GAUT proteins: clade A (AtGAUTs 1–7), clade B (AtGAUTs 8–11) and clade C (AtGAUTs 12–15) (Sterling et al. 2006). Subsequently, a search of the sequenced genomes of rice (Oryza sativa) and poplar (Populus trichocarpa) for GAUT homologs identified 21 poplar and 22 rice GAUTs and led to a refinement of the GAUT phylogeny into 7 subclades (Caffall et al. 2009). Inclusion of additional GAUTs from more recently sequenced plant genomes (e.g. grape, soybean, sorghum and Brachypodium) has substantiated the existence of seven distinct GAUT subclades, each with strong statistical support (bootstrap values >90%) (Fig. 6.2). For the purposes of this review, we have designated the seven GAUT subclades (a–g) without implying specific evolutionary relatedness between the subclades, since the evolutionary relationships between the GAUT subclades is unclear because of the relatively weak statistical support for these nodes. However, there is very strong statistical support for the relatedness of genes within a subclade (see Fig. 6.2). In addition, the structures of the GAUT genes are more conserved within subclades than across the subclades. For example, among the seven GAUT subclades, the GAUT-b and -c subclades have one or two introns, many fewer than the other GAUT subclades (Fig. 6.2). All of the major GAUT subclades contain both monocot and dicot proteins (Fig. 6.2). This is, at first glance, somewhat surprising given the functional association of the GAUTs with pectic polysaccharide biosynthesis (see below) and the fact that pectin constitutes only about 10% of the walls of monocots (Mohnen et al. 2008). This broad distribution of GAUT proteins in all GAUT clades must then reflect a key role for these glycosyltransferases in all higher plants. The relative distributions of monocot genes vs. dicot genes in each GAUT subclade are not uniform (Fig. 6.2). Thus, the GAUT-g clade has almost twice as many monocot representatives as dicot representatives. Furthermore, not all GAUTs have orthologues in all plants. For example, no monocot genes orthologous to AtGAUT12 could be identified, which is consistent with functional studies (see below). Likewise there are no obvious orthologues to AtGAUT2 in any plant included in the current analysis, suggesting that this Arabidopsis gene may be a pseudogene. Whether or not the individual GAUT subclades identified through phylogenetic analyses reflect the function(s) of their members with respect to synthesis of specific plant cell

190

■

Plant Polysaccharides

wall polysaccharides remains unclear and awaits more detailed functional studies on representative GAUT genes from each subclade. 6.3.1.2 Function of GAUT proteins All of the GAUT genes, with the possible exception of GAUT2 which may be a pseudogene, are expressed in all tissues examined to date, consistent with a function for GAUTs in wall biosynthesis. Some GAUTs, however, show higher levels of expression in some tissues. For example, GAUT8 and GAUT12 are highly expressed in vascular tissues in Arabidopsis stem (Orfila et al. 2005; Peña et al. 2007; Persson et al. 2007; Caffall et al. 2009). Functional studies of GAUT proteins have relied largely on molecular genetic and biochemical studies. The latter involved partial purification of GAUT1 from solubilized membrane fractions of Arabidopsis suspensioncultured cells. Proteomic analysis of the trypsin-digested solubilized protein fraction with the greatest amount of GalAT activity revealed the presence of only two putative glycosyltransferases; AtGAUT1 and AtGAUT7. Heterologous expression of AtGAUT1 in human embryonic kidney cells yielded α1,4-GalAT activity in the presence of homogalacturonan oligosaccharide acceptors (Sterling et al. 2006). Furthermore, anti-GAUT1 polyclonal antibodies immunoabsorbed GalAT activity from the partially purified Arabidopsis GalAT enzyme preparations. These data provide explicit biochemical evidence that AtGAUT1 is a pectin homogalacturonan (HG) α1,4galacturonosyltransferase (HG:GalAT) (Sterling et al. 2006) which transfers GalA from UDP-GalA on to the non-reducing end of homogalacturonan acceptors. This is the only GAUT for which enzyme activity has been confirmed in vitro. Interestingly, although no GalAT activity was detected in heterologously expressed AtGAUT7, AtGAUT7 co-purifies and coimmunoprecipates with AtGAUT1 and with HG:GalAT activity (M. A. Atmodjo, Y. Sakuragi, X. Zhu, A. Burrell, S.S. Mohanty, J.A. Atwood III, R. Orlando, H.V. Scheller and D. Mohnen, submitted for publication), suggesting that AtGAUT7 and AtGAUT1 are components of a GalAT protein complex involved in homogalacturonan biosynthesis. Analyses of Arabidopsis plants carrying mutations in individual GAUT genes have provided evidence for the involvement of GAUT proteins in the synthesis of several polysaccharides present in higher plant cell walls. For example, gaut8 (qua1-1) mutants show defects in cell adhesion and have reduced wall GalA and Xyl content, and a qua1-1 mutant stem microsomal membrane protein preparation shows reduced GalAT and xylan synthase activity compared to WT (Orfila et al. 2005; Brown et al. 2007), suggesting that that GAUT8 is involved in pectin and/or xylan synthesis. Biochemical and immunohistochemical analyses of gaut12 (irx8) mutants, which have a collapsed xylem phenotype (Brown et al. 2005), demonstrated that these mutants have somewhat reduced levels of GalA in their walls (Persson et al. 2007) and reduced stem secondary wall xylan (Persson et al. 2007; Peña et al. 2007). Chemical studies of gaut12 walls revealed the absence of a Xyl-

Glycosyltransferases of the GT8 Family

■

191

and GalA-containing pentasaccharide at the ends of xylan chains (Peña et al. 2007). The results of these biochemical analyses of gaut12 mutants have lead to the proposal that GAUT12 may be an α1,4-GalAT that participates in the synthesis of the so-called xylan primer/cap (Peña et al. 2007) or of a specific subfraction of pectic homogalacturonan (Persson et al. 2007). Interestingly, monocots lack any apparent GAUT12 homolog, suggesting that GAUT12 may have a specialized function in the synthesis of glucuronoxylan in dicots. Lastly, a function for GAUT11 in Arabidopsis seed mucilage and wall production has been proposed recently, based on associated changes in seed mucilage production and expansion in Arabidopsis gaut11 mutants, suggesting a role for this protein in rhamnogalacturonan synthesis (Caffall et al. 2009). For GAUT8, GAUT11, and GAUT12, however, no data exist, as yet, that would permit conclusions about the biochemical functions of these proteins. An extensive analysis of the glycosyl residue compositions of cell walls isolated from 26 homozygous T-DNA mutants of 13 of the 15 Arabidopsis genes has provided additional information about GAUT protein function (Caffall et al. 2009). These analyses demonstrated that gaut6, 8, 9, 10, 11, 12, 13 and 14 mutants have significant and reproducible changes in the levels of galacturonic acid, xylose, rhamnose, galactose and/or arabinose compared to walls of wild type Arabidopsis grown under the same growth conditions (Caffall et al. 2009). These results strongly implicate these GAUTs as being involved in the synthesis of at least six different biosynthetic linkages in pectins and/or xylans, based on the different patterns of changes in wall composition observed in the distinct gaut mutant walls. An initial hypothesis for GAUT function, based on the demonstrated homogalacturonan galacturonosyltransferase activity for GAUT1, is that the GAUTs function as GalATs involved in pectin synthesis. The phenotypes of several gaut mutants are compatible with this hypothesis. The walls of gaut6, 9, 10 and 11 mutants have significant reductions in GalA, suggesting a role in pectin, and possibly homogalacturonan synthesis, for the proteins encoded by these genes. Indeed, more detailed studies of GAUT6, including preliminary heterologous expression data, provide compelling evidence that GAUT6 is a putative pectin biosynthetic HG:GalAT (Caffall 2008; Caffall et al. 2009). Cell walls of the gaut 13 and 14 mutants have increased GalA and Gal content and reduced Xyl and Rha content compared to wildtype plants. Such changes could reflect a role for these genes in rhamnogalacturonan I (RG-I) synthesis, although such a function would require detailed enzymological studies of the expressed proteins. The inability to recover homozygous gaut1 or gaut4 mutants, and the high expression level of these genes in Arabidopsis, suggests a critical function for these proteins in plants. GAUT4 has 83% amino acid similarity to GAUT1, and a reasonable hypothesis is that GAUT4 is also a GalAT involved in pectin, and possibly, homogalacturonan synthesis. However, proof of this requires critical studies of the enzyme activity of the expressed protein. In summary, the available results are consistent with a

192

■

Plant Polysaccharides

role for the entire GAUT family as GalATs involved in pectin, and possibly xylan, synthesis. 6.3.2 Galacturonosyl transferase-like (GATL) clade A family of genes closely related to the GAUT family was identified on the basis of protein alignments (Lao et al. 2003; Sterling et al. 2006), and named galacturonosyl transferase-like (GATL) genes based on the conservation of specific amino acid motifs between the GATL and GAUT proteins (Sterling et al. 2006). GATL genes and their corresponding proteins are distinguishable from GAUTs based on three characteristics: (1) GATL genes and proteins are smaller than GAUTs; (2) GATL proteins lack an identifiable trans-membrane domain that is found in almost all GAUTs; and (3) GATL genes for the most part lack introns (Fig. 6.3), unlike the GAUT genes, which typically have extensive intron/exon structures (Fig. 6.2). Nonetheless, the conservation of

56

0.1

60

100

100 100

70 81

100 61

75

100 97 99 83 100 100 49

100 47

100 21

100

10

97 28 36 71 100 49 82

99 98 7

19 40

83

95

22 37 97 100

33

76

88 96 100

84

84 100 51

100

42

75

100 100 95 85 92

100 22 95

exon intron 0 1 2: intron phase

100 40 58 100

os/LOC Os06g13760.1 bd/Bradi1g44470.1 sb/Sb10g008830 sb/estExt fgenesh1 pg.C chr 42380 bd/Bradi3g59760.1 os/LOC Os02g50600.1 AtGATL10/AT3G28340.1 gm/Glyma19g01910.1 gm/Glyma13g04780.1 vv/GSVIVP00028923001 gm/Glyma02g03090.1 gm/Glyma01g04460.1 pt/fgenesh4 pm.C LG VIII000888 pt/estExt Genewise1Plus.C LG VIII2617 AtGATL9/AT1G70090.1 AtGATL8/AT1G24170.1 os/LOC Os03g18890.1 bd/Bradi1g64830.1 sb/estExt Genewise1.C chr 19676 vv/GSVIVP00028063001 gm/Glyma10g01960.1 gm/Glyma02g01880.1 gm/Glyma19g40180.1 gm/Glyma03g37560.1 pt/eugene3.00021805 pt/grail3.0061016301 AtGATL7/AT3G62660.1 AtGATL6/AT4G02130.1 AtGATL5/AT1G02720.1 bd/Bradi1g12560.1 sb/estExt Genewise1.C chr 12425 os/LOC Os03g47530.1 os/LOC Os10g31650.1 vv/GSVIVP00014689001 pt/estExt fgenesh4 pm.C LG VII0375 pt/fgenesh4 pg.C scaffold 57000218 AtGATL2/AT3G50760.1 gm/Glyma01g38520.1 gm/Glyma02g06640.1 vv/GSVIVP00025008001 AtGATL1/AT1G19300.1 pt/eugene3.00400186 pt/estExt Genewise1Plus.C LG II1880 gm/Glyma06g03770.1 gm/Glyma04g03690.1 gm/Glyma17g36650.1 gm/Glyma14g08430.1 os/LOC Os04g44850.1 sb/e gw1.6.14698.1 bd/Bradi5g16690.1 AtGATL3/AT1G13250.1 vv/GSVIVP00030290001 pt/grail3.0006037101 pt/grail3.0010045901 gm/Glyma02g11100.1 gm/Glyma01g22480.1 os/LOC Os07g45260.1 sb/fgenesh1 pm.C chr 2000674 sb/e gw1.1.15754.1 os/LOC Os03g24510.1 bd/Bradi1g61830.1 vv/GSVIVP00019872001 vv/GSVIVP00023650001 pt/grail3.0090012501 pt/grail3.0106011001 AtGATL4/AT3G06260.1 gm/Glyma07g38430.1

gm/Glyma17g02330.1

2

GATL-a

1

1

1

GATL-b 0 2

GATL-c

2

GATL-d 0

GATL-e

5'

0bp

3'

500bp

1000bp

1500bp

2000bp

Figure 6.3 The phylogeny of the GATL clade and gene structures of the GATL genes. See legend of Fig. 6.2 for further details.

Glycosyltransferases of the GT8 Family

■

193

specific amino acid motifs in the C-terminal domains of both GAUTs and GATLs suggest possible functional commonalities, particularly with respect to their roles in cell wall biosynthesis (Sterling et al. 2006; Yin et al. 2010). 6.3.2.1 Phylogeny of GATL clade Phylogenetic analysis of full-length GATL protein sequences based on genes from the fully sequenced plant genomes, using PhyML, a maximum likelihood algorithm (Guindon & Gascuel 2003), yields a tightly clustered tree with short branch lengths (Fig. 6.3). The latter suggests that the GATLs have diverged less than have the GAUTs, whose tree overall has significantly longer branch lengths (Fig. 6.2). The GATLs cluster into five distinct clades (a–e) with reasonable statistical support for the nodes defining these clades. Further support for this GATL tree topology is provided by the fact that each GATL clade subdivides into a monocot clade and a dicot clade, a frequent observation in gene family phylogeny comparisons between monocots and dicots. The latter also implies that the GATLs have evolved separately in the two major higher plant lineages. Thus, it may prove problematic to infer GATL function across the monocot/dicot divide. Among the seven fully sequenced plant genomes examined in this overview of family GT8, Arabidopsis and soybean have the largest diversity of GATLs. AtGATL10 and Glyma19g01910 have no easily recognizable counterparts in the other five plants. The subdivision of the GATLs into five subclades tempts one to infer five distinct functions for GATL genes in higher plants. However, very few experimental studies have been done to identify GATL function (see below), so firm conclusions on this point are premature. There is some pairing of GATL genes in the tree: for example, AtGATL5 with AtGATL6 and AtGATL7, AtGATL8 with AtGATL9, and the pairing of all GATLs from poplar (arising from the recent whole genome duplication event in the evolution of the poplar genome; see Tuskan et al. 2006). While such gene/protein pairings suggest functional redundancies within each pair, this must be proved experimentally. The little experimental evidence that exists to date, for AtGATL5 and AtGATL6 (see below) and PdGATL1.1 and PdGATL1.2 (Kong et al. 2009; Lee et al. 2009), suggests that these pairs of genes are not fully redundant functionally and that some functional specialization of these gene/protein pairs has occurred since duplication and divergence. 6.3.2.2 Function of GATL proteins Currently, little is known about the function(s) of most of the GATL proteins. What functional evidence exists is largely derived from protein localization, gene expression and mutational studies. Analysis of the domain structure of the GATLs using sequence analysis tools (as predicted by TMHMM v2.0; Krogh et al. 2001) reveals that about half of the GATL proteins from the fully sequenced plant genomes are

194

■

Plant Polysaccharides

predicted to contain a single transmembrane domain (Table 6.1). However, close examination of the GATL sequences reveals that these predicted transmembrane domains are located at or only a few amino acids away from the N-terminus of the protein and are thus, most likely, signal peptides. Those GATL proteins not predicted to contain a transmembrane domain are predicted to have a signal peptide (Table 6.1). Thus, all GATL proteins appear to be directed to the endomembrane system. No other consensus membraneanchoring sequences (e.g. for the attachment of lipid anchors) were identified in the GATL amino acid sequences. Thus, most GATLs are predicted to be soluble proteins. The predicted domain structure of the GATL proteins raises the possibility that these proteins are not involved in cell wall synthesis at all, but rather play roles as glycosylating agents in other cellular processes (e.g. glycosylation of hormones and other metabolites), perhaps in subcellular compartments other than the Golgi where most matrix cell wall polysaccharides are thought to be synthesized (Mohnen et al. 2008). Thus, one key element in the discovery of GATL function is the establishment of the intracellular location of GATL proteins. The only experimental studies on GATL localization published thus far have been carried out on GATLs from Arabidopsis and poplar. AtGATL1 (PARVUS) has been reported to be localized to the endoplasmic reticulum (ER), on the basis of heterologous expression studies carried out in carrot protoplasts (Lee et al. 2007). AtGATL1, PdGATL1.1 and PdGATL1.2 were localized to both ER and Golgi on the basis of transient expression studies in Nicotiana benthemiana leaves (Kong et al. 2009; Y. Kong, G. Zhou, Y. Yin, Y. Xu, S. Pattathil and M.G. Hahn, submitted for publication). Other as yet unpublished studies in Arabidopsis have demonstrated that several other AtGATL proteins (AtGATL2, AtGATL5, AtGATL6, and AtGATL9) are localized in the Golgi, and in some cases, in the ER (Y. Kong, G. Zhou, Y. Yin, Y. Xu, S. Pattathil and M.G. Hahn, submitted for publication). Thus, the available evidence indicates that GATL proteins are in the correct subcellular compartment to participate in cell wall or glycoprotein synthesis. Given the apparent absence of transmembrane or other membrane-anchoring domains in these proteins, another mechanism must be responsible for retaining the GATL proteins in the endomembrane system. Otherwise, the GATL proteins would be secreted via the default secretory pathway that targets proteins entering the endomembrane system to the apoplast, unless they are otherwise directed by signal sequences (Rojo & Denecke 2008). It is tempting to speculate that GATLs are retained in the endomembrane system because they participate with bona fide membrane-anchored proteins in biosynthetic complexes that are responsible for cell wall polysaccharide biosynthesis. However, most such enzymes that have been studied show Golgi localization patterns (Mohnen et al. 2008), and the ER localization of at least some GATLs is difficult to reconcile with such a model. Thus, the mechanism by which GATLs are retained within the endomembrane system remains to be experimentally determined.

Glycosyltransferases of the GT8 Family

■

195

Studies in Arabidopsis have demonstrated that most GATL genes are expressed throughout the plant, though with higher levels of expression for individual GATL genes in particular tissues or cell types (Kong et al. 2009; Y. Kong, G. Zhou, Y. Yin, Y. Xu, S. Pattathil and M.G. Hahn, submitted for publication). Transcriptomic analyses in poplar and Arabidopsis have documented co-expression of at least some of the GATLs with other genes known to be involved in cell wall synthesis. For example, GATL1 has been found to be co-regulated with other cell wall genes in xylem (Ko et al. 2006; Leplé et al. 2007; Minic et al. 2009), as have GATL6 and GATL9 (Kubo et al. 2005). These results suggest an involvement of GATL1 in the synthesis of polysaccharides important for secondary wall formation, such as xylan (Brown et al. 2007; Lee et al. 2007; Kong et al. 2009). GATL5 has been shown to be co-expressed with other genes involved in synthesis of seed mucilage (Y. Kong, G. Zhou, M. Atmodjo, A.M.A. Abdeen, T. Western, D. Mohnen and M.G. Hahn, in preparation). Other studies have placed GATL genes in context with other cell wall biosynthetic genes whose expression is altered by responses to nutrient depletion (GATL8, GATL9; Vlieghe et al. 2003) or environmental stress (GATL4; Tsabary et al. 2003) and developmental programmes (GATL7; Mandaokar et al. 2003). Whether or not such correlations observed in transcriptomic analyses allow conclusions about functional associations of GATLs with other cell wall synthetic or modifying enzymes awaits experimental verification. In any case, these transcriptomic studies imply a role for GATLs in polysaccharide biosynthesis. Mutational studies have provided some experimental support for the hypothesis that GATLs are involved in the synthesis of several cell wall polysaccharides, including xylan (GATL1) and pectins (GATL5 and GATL6). Mutations in GATL1 (parvus, glz1) were first isolated as plants with reduced stature (Lao et al. 2003; Shao et al. 2004). Early glycosyl composition analyses of leaves from the parvus mutant showed decreased xylose content and modest increases in rhamnosyl and galactosyl residues, particularly in young leaves (Lao et al. 2003). These results, indicating an involvement of GATL1 in xylan and possibly pectin synthesis, have been substantiated, at least in part, by recent studies of xylan synthesis in parvus mutants (Brown et al. 2007; Lee et al. 2007; Kong et al. 2009; Lee et al. 2009). These studies focused on secondary wall formation in stem and root xylem tissues and examined changes in xylan structure in the mutant plants. The thickness of the secondary walls in xylem tissues are dramatically reduced in parvus. Furthermore, a tetrasaccharide present at the non-reducing terminus of xylan chains in wildtype plants (Peña et al. 2007) is absent in the mutant and there is a significant decrease in xylan content of mutant walls (Brown et al. 2007; Lee et al. 2007). However, the phenotype of parvus/glz1 mutants is complex, with many character traits (e.g. stature, organ size, pollen production and viability) affected, in addition to the effects on secondary wall formation (Lao et al. 2003; Shao et al. 2004). Not all of these mutant characters can easily be explained solely on the basis of defects in xylan synthesis and additional studies will be necessary to exclude a role for

196

■

Plant Polysaccharides

GATL1 in pectin biosynthesis as had been suggested in earlier studies (Lao et al. 2003). A systematic analysis of insertional mutations in other GATL genes has resulted in the association of at least some GATLs with the synthesis of pectic polysaccharides in Arabidopsis. Mutations in AtGATL5 result in a significant decrease in the synthesis of mucilage in Arabidopsis seeds. The effect of the gatl5 mutation is enhanced by mutation of a second, closely related gene (Fig. 6.3), AtGATL6, resulting in the almost complete absence of seed mucilage. The single gatl6 mutant on its own shows little if any effect on mucilage biosynthesis. Arabidopsis seed mucilage is largely composed of unbranched rhamnogalacturonan I (Deng et al. 2009). Further detailed characterization of the gatl5 mutant provides strong evidence that GATL5 participates in the synthesis of pectic polysaccharides (Y. Kong, G. Zhou, M. Atmodjo, A.M.A. Abdeen, T. Western, D. Mohnen and M.G. Hahn, in preparation). In addition to its expression in developing seeds, GATL5 is expressed throughout the plant, particularly in the vasculature of root and stem. These results suggest that GATL5 participates in pectin synthesis in other cellular contexts besides mucilage synthesis. An insertional mutation in AtGATL9 leads to defects in cell adhesion (Y. Kong, G. Zhou and M.G. Hahn, unpublished results), a phenotype that has been observed in plants carrying mutations in genes thought to be associated with pectic polysaccharide biosynthesis (Bouton et al. 2002) (see previous section on GAUT genes). A 20% reduction in GalA content was detected in cell walls of gatl9 mutants, and over-expression studies also suggest that GATL9 participates in pectin biosynthesis. However, further detailed studies are necessary to demonstrate definitively that GATL9 is a pectin biosynthetic enzyme. Arabidopsis plants carrying single mutations in other GATL genes do not display easily identifiable morphological or cell wall abnormalities (Y. Kong and M.G. Hahn, unpublished results), suggesting that some redundancies exist with regard to GATL function at least in this plant.

6.4 GT8 clades not related to cell wall synthesis The proteins in the remaining GT8 clades have not been tied experimentally to cell wall synthesis. Three clades (PGSIP clades) include proteins that were originally identified (Qi et al. 2005; Chatterjee et al. 2005) on the basis of their sequence similarity to fungal and mammalian proteins, called glycogenins, which are self-glycosylating proteins that serve as initiators of glycogen biosynthesis in those organisms (Roach & Skurat 1997). The other clade (GolS clade) contains proteins that have been associated with the synthesis of raffinose family oligosaccharides that appear to have roles in plant stress responses. These non-cell-wall-related GT8 clades include proven and putative α-d-glycosyltransferases that use protein or inositol acceptors and

Glycosyltransferases of the GT8 Family

■

197

glucose or galactose nucleotide sugars as donors. All of these clades have characteristic features and sequence motifs that distinguish them from each other and from the GT8 families related to cell wall synthesis, discussed previously. 6.4.1 Plant glycogenin-like (PGSIP) clades All living organisms store chemical energy in some way in order to survive when their living conditions change. The energy-storing molecules are synthesized at times when nutrition is easily available, and are used when sustenance is scarce or in times of stress. The energy storage carbohydrate in most organisms other than plants is glycogen. The biosynthesis of glycogen, an α1,4-linked glucan with α1,6 branch points at approximately every ten glucosyl residues, is a three-step process that includes initiation, elongation and branching (Alonso et al. 1995a). The first step in the biosynthesis of glycogen, initiation, is carried out by glycogenin, a self-glucosylating protein that serves as an initiator of glycogen synthesis, and is believed to play a key regulatory role in glycogen synthesis (Roach & Skurat 1997; Lomako et al. 2004). The initiation step involves transfer of a glucosyl residue from UDP-Glc to a single tyrosyl residue on glycogenin forming a Glc-1-O-tyrosyl linkage on glycogenin (Smythe et al. 1988). This step is one of several glycogenin self-glucosylation reactions, which are repeated about ten times to yield glycogenin with a covalently bound oligoglucan oligosaccharide. It has been suggested that glycogenin exists functionally as a dimer in which one subunit glycosylates the other (Alonso et al. 1995b). Starch, which is used as a storage polymer in plants and is synthesized in the chloroplast, is structurally similar to glycogen, in that it also consists of an α1,4-linked glucan with α1,6 branch points. Unlike glycogen, however, starch is made of two basic polymers, amylopectin and amylose. The branched amylopectin has approximately 1 branch point per 30 glucose units while amylose is less branched (Hoover 2001). Some plant starches, such as waxy maize starch, are composed only of amylopectin (AngellierCoussy et al. 2009). Starch is semicrystalline, with granular sizes ranging from ∼2 μm in rice up to 85 μm in other plant species. Another difference between glycogenin and amylopectin, aside from the frequency of branching, is that the activated sugar for glycogenin initiation is UDP-Glc, while ADP-Glc is used for starch synthesis (Ball et al. 1998). Although glycogen and amylopectin have very similar chemical characteristics, glycogen is fully soluble and non-crystalline (Calder 1991), while amylopectin is more crystalline (Ball et al. 1996). Plant proteins with sequence similarity to mammalian glycogenin were first identified in Arabidopsis (Chatterjee et al. 2005) and rice (Qi et al. 2005), and in two thermo-acidophilic red algae (Barbier et al. 2005). These proteins were named plant glycogenin-like starch initiation proteins (PGSIP) (Chatterjee et al. 2005), which we also refer to in this review as plant

198

■

Plant Polysaccharides

glycogenin-like proteins, and it has been suggested that they may be involved in the initiation of starch amylopectin biosynthesis in a manner similar to glycogenin in glycogen synthesis. Some of the genes encoding PGSIPs have been shown to be upregulated by auxin (Goda et al. 2004), wounding (Guan & Nothnagel 2004), and by environmental stress (Qi et al. 2005) or disease (Schaff et al. 2007). 6.4.1.1 Phylogeny of PGSIP clades Phylogenetic analysis using PhyML, a Maximum Likelihood algorithm (Guindon & Gascuel 2003), distributed the glycogenin-like proteins from the seven higher plant genomes into three distinct phylogenetic clades whose evolutionary relationships are not completely clear due to the relatively weak statistical support for some of the nodes in this part of the family GT8 tree (see Fig. 6.1). The PGSIP-B and PGSIP-C clades likely derive from a common ancestor. However, the evolutionary relationship of these two clades with the PGSIP-A clade is not clear. All PGSIP proteins contain the following consensus MEME motif (Bailey & Elkan 1994): YSK [ FL ] RLWQLTDYD [ KR ][ VI ][ VI ] F [ IL ] DADL [ LI ][ VI ] L [RK ] NIDFLFA [ CM ][PG][ QE] In addition, each PGSIP clade is characterized by one or more MEME motifs that are unique to the proteins in a given clade. The functional significance, if any, of the distribution of the PGSIP proteins into three clades cannot be ascertained at present, as no plant PGSIP protein has yet been functionally characterized (see below). The PGSIP-A clade contains both monocot and dicot PGSIP genes that are distributed among several subclades (Fig. 6.4). However, the orthologous relationships across the monocot/dicot divide in this clade are not clear for all subclades. For example, there are two monocot subclades most closely related to two dicot subclades containing AtPGSIP1 (At3g18660) and AtPGSIP2 (At1g77130), but the two monocot clades branch earlier than the two dicot clades. The three monocot genomes included in this analysis do not appear to contain genes orthologous to AtPGSIP4 or AtPGSIP5, suggesting that these gene products play distinctive roles in dicots. The PGSIP-B and PGSIP-C clades each consist of single monophyletic groups of genes split among monocots and dicots (Fig. 6.5, Fig. 6.6). A distinguishing characteristic of the PGSIP-B clade is that the genes in this clade have, on average, twice as many exons as do the genes in either the PGSIP-A or PGSIP-C clades. The exon/intron gene structure is also highly conserved within both the PGSIP-B and PGSIP-C clades (Fig. 6.5, Fig. 6.6), but less well conserved within the PGSIP-A clade (Fig. 6.4).

100

67 59

100

82

100

100 100

100 100 100 100

100

99 92

32

82

100 72

55

61

100

97 64

100 89 77

100

100

100

99

100

100

100

100

100

exon

intron

0 1 2: intron phase

5'

vv/GSVIVP00022838001 AtPGSIP4/AT1G54940.1 AtPGSIP5/AT1G08990.1 pt/fgenesh4 pg.C scaffold 70000240 pt/e gw1.XIII.1015.1 gm/Glyma10g29570.1 gm/Glyma19g42380.1 gm/Glyma03g39820.1 bd/Bradi1g72350.1 os/LOC Os03g08600.1 sb/Sb01g044930 pt/estExt Genewise1 v1.C 400718 vv/GSVIVP00014811001 AtPGSIP3/AT4G33330.1 gm/Glyma04g04080.1 gm/Glyma14g09070.1 gm/Glyma17g36100.1 sb/e gw1.3.18262.1 bd/Bradi5g27680.1 bd/Bradi3g45800.1 sb/fgenesh1 pg.C chr 4001767 os/LOC Os02g35020.1 sb/Sb09g020930 bd/Bradi2g24740.1 os/LOC Os01g65780.1 bd/Bradi2g56810.1 sb/estExt Genewise1.C chr 310243 vv/GSVIVP00014288001 pt/e gw1.V.184.1 AtPGSIP2/AT1G77130.1 gm/Glyma02g40480.1 gm/Glyma14g28370.1 gm/Glyma0214s00200.1 vv/GSVIVP00034844001 AtPGSIP1/AT3G18660.1 pt/eugene3.00050013 pt/eugene3.00070393 gm/Glyma06g15690.1 gm/Glyma04g39240.1 gm/Glyma05g32370.1 gm/Glyma05g04770.1 gm/Glyma08g15640.1 2

2

1

2

2

0kb

1kb

20

2

1

1

1

1

1

1

1

1

3kb

00112

2

2kb

1

0 1

2 2

2

1 1

1

2

2

1

1 1 1

2

2

1

1 0

2

0

0

2 2

2

2

2

2 2

2

1 2

0

1

2 2

2

11

1

11

0

0

2

2

2

2

0

1

2 1

0

2

2 1

0

2

1

2

1

2

0

01

2

1

0

2 220

1

1

1

1

0

4kb

0

0

5kb

0

6kb

2

7kb

8kb

01 2

2

9kb

3' 17kb

■

Figure 6.4 The phylogeny of the PGSIP-A clade and gene structures of PGSIP-A genes. See legend of Fig. 6.2 for further details.

100

77

98

0.1

Glycosyltransferases of the GT8 Family

199

200

■

Plant Polysaccharides

0.1

os/LOC Os02g41520.1 bd/Bradi3g49200.1 os/LOC Os04g43700.1 sb/estExt fgenesh1 pg.C chr 61566 bd/Bradi5g15780.1 AtPGSIP6/AT5G18480.1 vv/GSVIVP00002209001 pt/eugene3.00130489 gm/Glyma03g40980.1 gm/Glyma19g43630.1 gm/Glyma20g37000.1 gm/Glyma10g30700.1

97 100 80 65

37 92

100 100 100

exon

intron

0 1 2: intron phase

0

11 0

0

000

12

11 0

0 111

000

02 00

0 11

0 000

0

1 1

12 2

1 20

0

0

12 12

0 000 12

12 0000 20 1

20

0 1 1

1

2 0 000

1

00 0

2

2 0

0 0 00 12 22

2 0 000

1

2 0 000

1

2 0 0010

0

22 0 2 20

22200

5' 3' 0kb 1kb 2kb 3kb 4kb 5kb 6kb 7kb 8kb 9kb 10kb11kb12kb13kb14kb

Figure 6.5 The phylogeny of the PGSIP-B clade and gene structures of the PGSIP-B genes. See legend of Fig. 6.2 for further details.

0.1

100 91 100 100 52 100 100 51

exon

vv/GSVIVP00009599001 gm/Glyma11g03550.1 gm/Glyma05g04630.1 AtPGSIP7/AT2G35710.1 AtPGSIP8/AT4G16600.1 os/LOC Os04g46750.1 sb/Sb06g024740 bd/Bradi5g17880.1 os/LOC Os10g40640.1 sb/Sb01g029400 bd/Bradi3g33080.1 intron

2 2 2

0 1 2: intron phase 5'0kb

0 12 0

12 0

12

12

1

2

12 20 1 20 1 2 0 1 2 0 12 12 0 1 2 0 12 1 1 1

1kb

2kb

3kb

4kb

5kb

6kb

7kb

3'

Figure 6.6 The phylogeny of the PGSIP-C clade and gene structures of the PGSIP-C genes. See legend of Fig. 6.2 for further details.

6.4.1.2 Function of PGSIP proteins The function(s) of plant glycogenin-like proteins remains to be established. What little information exists originates from mutational, bioinformatic and transcriptomic analyses. There are published reports of plant selfglucosylating proteins and polypetides (Ardila & Tandecarz 1992; Singh et al. 1995; Langeveld et al. 2002), but none have been shown to be involved in the initiation of starch biosynthesis. The initial discovery and naming of the PGSIP protein clades was made based on a search of the Arabidopsis genome for genes encoding proteins with sequence similarity to mammalian and yeast glycogenin (Chatterjee et al. 2005). Six of these proteins (AtPGSIP1– 6) have amino acid motifs consistent with a possible function of these proteins as glycosyltransferases. However, they lack the conserved Tyr residue that is the site of self-glycosylation in glycogenin. In yeast, glycogenin is glycosylated at multiple Tyr sites (Mu et al. 1996) and the Arabidopsis PGSIPs do contain alternative Tyr residues that could be used as the sites for self-glycosylation. More critically, only two Arabidopsis glycogeninlike proteins, AtPGSIP1 and AtPGSIP3, are predicted (see http:// www.cbs.dtu.dk/services/ChloroP/; Emanuelsson et al. 1999) to contain a transit peptide to target the polypeptide to the chloroplast (Chatterjee et al.

Glycosyltransferases of the GT8 Family

■

201

2005 and unpublished results), the organelle in which starch biosynthesis occurs. Indeed, plants carrying a knock-out mutation in PGSIP1 are starchdeficient in their leaves (Chatterjee et al. 2005). These results suggest that PGSIP1 plays a role in starch biosynthesis, at least in Arabidopsis. However, no enzymatic activity has been demonstrated for this protein as yet. The absence of a clear transit peptide for chloroplast targeting in the other PGSIP proteins raises questions about the possible role(s) of these proteins in starch biosynthesis. To date, no experimental verification of the subcellular localization of any of the PGSIPs has been reported. AtPGSIP7 and AtPGSIP8 lack conserved motifs involved in binding of the donor ligand, UDP-Glc, and also lack a conserved His residue (in the HxxGxxKPW motif at position 277) important for Mn2+ binding (Chatterjee et al. 2005). Thus, it is doubtful that the proteins in the PGSIP-C clade are involved in starch biosynthesis, and detailed studies will be required to define the function(s) of the proteins in this clade. Transcriptomic analyses have revealed that PGSIPs are up-regulated in response to hormones (Goda et al. 2004; Qi et al. 2005), wounding (Guan & Nothnagel 2004), and environmental stress (Qi et al. 2005) or disease (Schaff et al. 2007). However, these analyses by themselves are not enlightening as to the function(s) of the up-regulated PGSIPs. Interestingly, bioinfomatic identification of Arabidopsis genes co-expressed with cellulose synthases (CESA) identified, among others, AtPGSIP1 as being co-expressed with secondary cell wall CESA 4, 7, and 8 (Persson et al. 2005). These results suggest a possible role for at least this PGSIP in secondary cell wall biosynthesis in Arabidopsis. These transcriptomic and bioinformatic analyses will need to be followed up with detailed molecular genetic and biochemical studies to define the functions of PGSIPs in plants. 6.4.2 Galactinol synthase (GolS) clade Plants are constantly exposed to changes in their environment including nutrient depletion, temperature extremes, and drought. To survive such challenges plants have developed means to store energy, maintain cellular integrity and preserve cellular metabolism. One such survival mechanism is through the formation of carbohydrates that can be transported and/or stored, and that can serve as protectant molecules (Peterbauer & Richter 2001; Browse & Lange 2004). Raffinose family oligosaccharides (RFOs) are examples of such carbohydrates. RFOs are water-soluble non-reducing carbohydrate molecules. There is a substantial body of literature on RFO accumulation and function in plants (Peterbauer & Richter 2001; Browse & Lange 2004), so only a few highlights and examples will be presented here. All plants, at some point, synthesize some RFOs, but many neither transport nor accumulate large quantities in their tissues and/or organs. For example, ryegrass has been shown to accumulate small amounts of raffinose and loliose under normal, non-stress,

202

■

Plant Polysaccharides

conditions (Amiard et al. 2003). In contrast, other plants store RFOs in large concentrations, sometime 25–80% of their dry weight, in tubers (French 1954; Keller & Matile 1985) and in photosynthesizing leaves where they are localized in mesophyll cells (Senser & Kandler 1967; Bachmann et al. 1994). RFOs also accumulate in seeds (Korytnyk & Metzler 1962, Handley et al. 1983a; Handley et al. 1983b; Blackman et al. 1992; Haritatos et al. 2000; Peterbauer et al. 2001), as well as in vegetative tissues of many plants (French 1954; Zimmermann 1957; Handley et al. 1983a), and serve as the predominant transport carbohydrate in cucurbits (Handley et al. 1983a; Keller & Pharr 1996). RFOs are believed to prolong seed survival and storage by protecting the embryo during the desiccation that occurs during seed maturation (Peterbauer et al. 2001). Plants can also accumulate RFOs in response to diverse abiotic stresses (Nishizawa et al. 2008) such as temperature extremes or drought. Under freezing conditions, for example, ice forms first in the extracellular compartments of plants that are exposed to low temperatures, causing a reduction of water potential and consequent loss of water from cells by osmosis (Browse & Lange 2004). Similar processes take place under drought conditions due to desiccation and dehydration. RFOs, under these conditions, are thought to act as osmolytes to maintain cellular integrity and function (Nishizawa et al. 2008). RFOs are synthesized by a transfer of a galactosyl residue from galactinol to the glucosyl residue of sucrose, resulting in an α1,6 linkage. The first step in the synthesis of any RFO is the synthesis of galactinol that is catalysed by galactinol synthases (GolS), which are members of CAZy family GT8. The only known function for galactinol, to date, is that of a galactosyl donor in RFO biosynthesis (Saravitz et al. 1987; Peterbauer et al. 2002). Raffinose (GalSuc), stachyose (diGal-Suc), verbacose (triGal-Suc), etc. are formed as a result of the action of several galactosyltransferases (e.g. raffinose synthase, stachyose synthase, etc.) (Peterbauer et al. 2001). 6.4.2.1 Phylogeny of GolS clade Phylogenetic analysis using PhyML, a maximum likelihood algorithm (Guindon & Gascuel 2003), distributed the galactinol synthases from the seven higher plant genomes into a single, strongly supported monophyletic clade (see Fig. 6.1). The evolutionary relationship of the GolS clade to the other non-cell-wall synthetic GT8 clades remains unclear due to the relatively weak statistical support for the nodes in this part of the family GT8 tree (see Fig. 6.1). The subclade structure within the GolS clade is unclear, as most of the nodes between subgroups of GolS proteins have very weak statistical support (Fig. 6.7). Such weak support for subclades within the GolS clade was also reported in earlier phylogenetic analyses of GolS sequences (Volk et al. 2003; Zhao et al. 2004). Nonetheless, it is clear that monocots have a smaller number and diversity of GolS genes compared with dicots, and that

100

100

100

100

100

100

65

100

89

87

93

100

exon

intron

0 1 2: intron phase

5' 0bp

vv/GSVIVP00033193001 vv/GSVIVP00019669001 AtGolS4/AT1G60470.1 AtGolS7/AT1G60450.1 vv/GSVIVP00002727001 pt/grail3.0009037801 pt/estExt fgenesh4 pg.C LG X0618 os/LOC Os07g48830.1 sb/e gw1.2.20933.1 bd/Bradi1g17200.1 sb/Sb01g037090 os/LOC Os03g20120.1 bd/Bradi1g64120.1 pt/fgenesh4 pm.C LG X000590 pt/fgenesh4 pm.C LG VIII000417 AtGolS8/AT5G30500.1 AtGolS6/AT4G26250.1 AtGolS5/AT5G23790.1 pt/gw1.131.196.1 pt/estExt fgenesh4 pm.C LG XIII0025 pt/e gw1.XIII.549.1 vv/GSVIVP00022562001 vv/GSVIVP00022565001 vv/GSVIVP00022570001 vv/GSVIVP00022561001 vv/GSVIVP00022559001 AtGolS1/AT2G47180.1 pt/eugene3.00140617 pt/estExt fgenesh4 pm.C LG II0906 gm/Glyma19g40680.1 gm/Glyma03g38080.1 vv/GSVIVP00028165001 vv/GSVIVP00028167001 AtGolS2/AT1G56600.1 AtGolS3/AT1G09350.1 gm/Glyma20g22700.1 gm/Glyma10g28610.1 gm/Glyma03g38910.1 gm/Glyma19g41550.1 1 0 0 0 0 0 0

0 0

0

0

0 0 0

0 0

0

0 0 0 0 0 0 0

0

0 0

0

500bp

0 0 0

0 0

0 0 0

0 0

0

0

0

0

0

0 0

0 0

0

0

0 0 0

0 0

0

0 0

0

0

0

0

0

0

1

0

0

0

0 0

0

2 0

0 0

0

2

0

0

0

1000bp

0

0 0

0

0

0

0

1500bp

0

0

0

0 0

0 0 0 0 0 0 0 0 0 0

0

0

0

0 0

0 2000bp

0

0 2500bp

0

3' 3000bp

■

Figure 6.7 The phylogeny of the GolS clade and gene structures of the GolS genes. See legend of Fig. 6.2 for further details.

37

18

48

98

100

59 36 9

75

70

91

98

100

100

100

100

59

98

88

69

28

12

74

26

0.1

Glycosyltransferases of the GT8 Family

203

204

■

Plant Polysaccharides

the monocot galactinol synthases cluster together in one subclade that is most closely related to a dicot cluster containing AtGolS4 and AtGolS7 (Fig. 6.7). Thus, it appears that dicots have developed or retained a greater diversity of uses for RFOs than have monocots. 6.4.2.2 Function of GolS proteins Galactinol (O-α-d-galactopyranosyl-[1→1]-l-myo-inositol) is synthesized by galactinol synthase (EC 2.4.1.123) which transfers a galactosyl residue from UDP-Gal on to myo-inositol in a reversible reaction yielding galactinol and UDP. This is a regulatory step in RFO biosynthesis (Robbins & Pharr 1987; Saravitz et al. 1987; Hitz et al. 2002). Galactinol synthase was first isolated from pea seeds (Frydman & Neufeld 1963), and later from Cucumis sativum (Pharr et al. 1981), and cotyledons of kidney bean (Liu et al. 1995). GolS activity has been associated with RFO biosynthesis in cucumber fruit (Handley et al. 1983a) and maturing soybean seeds (Saravitz et al. 1987). GolS enzymatic activity is readily assayable, as both donor and acceptor are known and available. Furthermore, GolS proteins do not have any transmembrane domains (Table 6.1) and are thus expected to be soluble proteins, greatly simplifying functional assays of heterologously expressed GolS proteins. Indeed, a number of GolS genes have been shown to encode functional galactinol synthases (Zhao et al. 2004). There is an extensive literature on the regulation and expression of GolS genes in plants, particularly with respect to environmental stresses such as desiccation (including both drought and seed maturation), salinity, and low temperatures (Obendorf 1997; Peterbauer & Richter 2001). At least in dicots, it appears that different GolS genes have specialized functions. For example, two GolS genes in Arabidopsis (AtGolS1 and AtGolS2) (Panikulangara et al. 2004) and one in maize (ZmGolS2) (Zhao et al. 2004) are highly expressed in mature and dry seeds, where they most likely acting as desiccation stress response factors. Other galactinol synthase genes are expressed at low levels in seeds and are induced by diverse stresses, including temperature stresses (Taji et al. 2002; Fowler & Thomashow 2002; Amiard et al. 2003; Downie et al. 2003; Panikulangara et al. 2004; Zhao et al. 2004). AtGolS1 and AtGolS2 are also up-regulated in leaves in response to drought and salinity (Nishizawa et al. 2008), by a combination of high light and heat stress (Panikulangara et al. 2004; Nishizawa et al. 2006), or by treatment with peroxide (Nishizawa et al. 2008). AtGolS3 (Nishizawa et al. 2008), and GolS genes in Ajuga reptans (Bachmann et al. 1994), tomato (Downie et al. 2003) and alfalfa (Cunningham et al. 2003) are induced during cold stress. GolS transcript levels decrease rapidly when plants are returned to normal growth conditions (Cunningham et al. 2003). Molecular genetic studies confirmed a role for GolS genes and RFOs in plant responses to environmental and pathogen stress. Over-expression of AtGOLS2 resulted in the improved drought tolerance of transgenic Arabidopsis plants (Panikulangara et al. 2004). These plants also showed

Glycosyltransferases of the GT8 Family

■

205

increased levels of galactinol and raffinose in leaves (Panikulangara et al. 2004). A knock-out mutation in the AtGolS1 gene in Arabidopsis results in impaired galactinol flux and raffinose synthesis (Panikulangara et al. 2004). Galactinol synthase is also shown to protect plants against pathogen infections. A cucumber galactinol synthase CsGolS1 gene was isolated from root tissue colonized by Pseudomonas chlororaphis (Kim et al. 2008). Over-expression of CsGolS1 in transgenic tobacco plants resulted in a constitutive resistance against pathogen infection and stimulated the accumulation of defencerelated genes (Kim et al. 2008). The CsGolS1 over-expressing transgenic plants also showed an increased tolerance to drought and high salt stress conditions (Kim et al. 2008).

6.5

Conclusions

The CAZy family GT8 genes present in higher plants are divided into two distantly related sets of proteins. One set, which includes the GAUTs and GATLs, has been strongly implicated as having roles in the synthesis of pectins and xylans, two major groups of polysaccharides present in plant cell walls. The second set, which includes PGSIPs and GolSs, appear not to be directly involved in plant cell wall synthesis. The PGSIPs have been suggested to play a role in priming starch biosynthesis, while the GolSs are key enzymes in the synthesis of the raffinose family of oligosaccharides that play important roles in environmental stress responses in plants. The large number and diversity of GT8 proteins in higher plants highlights the importance of these proteins for the biology of plants. However, much work remains to be done to define the biochemical functions of GT8 proteins, particularly those in the GAUT, GATL, and PGSIP clades.

Acknowledgements This work was supported by the US Department of Energy (BioEnergy Science Center and grant DESG0296ER20220). The BioEnergy Science Center is a U.S. Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science. Research on GAUTs and GATLs is also supported in part by grants from the US National Science Foundation (MCB-0646109) and the USDA (NRI-CREES-2006-35318-17301).

References Alonso, M.D., Lomako, J., Lomako, W.M., Whelan, W.J. (1995a) A new look at the biogenesis of glycogen. FASEB Journal, 9, 1126–1137.

206

■

Plant Polysaccharides

Alonso, M.D., Lomako, J., Lomako, W.M., Whelan, W.J. (1995b) Catalytic activities of glycogenin additional to autocatalytic self-glucosylation. Journal of Biological Chemistry, 270, 15315–15319. Amiard, V., Morvan-Bertrand, A., Billard J-P., Huault, C., Keller, F., Prud’homme M-P. (2003) Fructans, but not the sucrosyl-galactosides, raffinose and loliose, are affected by drought stress in perennial ryegrass. Plant Physiology, 132, 2218–2229. Angellier-Coussy, H., Putaux J-L., Molina-Boisseau, S., Dufresne, A., Bertoft, E., Perez, S. (2009) The molecular structure of waxy maize starch nanocrystals. Carbohydrate Research, 344, 1558–1566. Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature, 408, 796–815. Ardila, F.J., Tandecarz, J.S. (1992) Potato tuber UDP-glucose:protein transglucosylase catalyzes its own glucosylation. Plant Physiology, 99, 1342–1347. Bachmann, M., Matile, P., Keller, F. (1994) Metabolism of the raffinose family oligosaccharides in leaves of Ajuga reptans L. – Cold acclimation, translocation, and sink to source transition: Discovery of chain elongation enzyme. Plant Physiology, 105, 1335–1345. Bailey, T.L., Elkan, C. (1994) Fitting a mixture model by expectation maximization to discover motifs in biopolymers. In: Proceedings of the Second International Conference on Intelligent Systems for Molecular Biology (eds. R. Altman, D.L. Brutlag, P.D. Karp, R.H. Lathrop, D.B. Searls), pp 28–36. AAAI Press, Menlo Park, CA. Ball, S., Guan H-P., James, M., et al. (1996) From glycogen to amylopectin: A model for the biogenesis of the plant starch granule. Cell, 86, 349–352. Ball, S.G., van de Wal, M.H.B.J., Visser, R.G.F. (1998) Progress in understanding the biosynthesis of amylose. Trends in Plant Science, 3, 462–467. Barbier, G., Oesterhelt, C., Larson, M.D., Halgren, R.G., Wilkerson, C., Garavito, R.M., Benning, C., Weber, A.P.M. (2005) Comparative genomics of two closely related unicellular thermo-acidophilic red algae, Galdieria sulphuraria and Cyanidioschyzon merolae, reveals the molecular basis of the metabolic flexibility of Galdieria sulphuraria and significant differences in carbohydrate metabolism of both algae. Plant Physiology, 137, 460–474. Bendtsen, J.D., Nielsen, H., von Heijne, G., Brunak, S. (2004) Improved prediction of signal peptides: SignalP 3.0. Journal of Molecular Biology, 340, 783–795. Blackman, S.A., Obendorf, R.L., Leopold, A.C. (1992) Maturation proteins and sugars in desiccation tolerance of developing soybean seeds. Plant Physiology, 100, 225–230. Bouton, S., Leboeuf, E., Mouille, G., et al. (2002) QUASIMODO1 encodes a putative membrane-bound glycosyltransferase required for normal pectin synthesis and cell adhesion in Arabidopsis. Plant Cell, 14, 2577–2590. Brown, D.M., Goubet, F., Vicky, W.W., Goodacre, R., Stephens, E., Dupree, P., Turner, S.R. (2007) Comparison of five xylan synthesis mutants reveals new insight into the mechanisms of xylan synthesis. Plant Journal, 52, 1154–1168. Brown, D.M., Zeef, L.A.H., Ellis, J., Goodacre, R., Turner, S.R. (2005) Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics. Plant Cell, 17, 2281–2295. Browse, J., Lange, B.M. (2004) Counting the cost of a cold-blooded life: Metabolomics of cold acclimation. Proceedings of the National Academy of Sciences of the U S A, 101, 14996–14997.

Glycosyltransferases of the GT8 Family

■

207

Caffall, K.H. (2008) Expression and characterization of galacturonosyltransferase-6 (GAUT6) of the galacturonosyltransferase-1 (GAUT1)-related gene family of Arabidopsis thaliana. PhD thesis, University of Georgia, Athens, GA. Caffall, K.H., Pattathil, S., Phillips, S.E., Hahn, M.G., Mohnen, D. (2009) Arabidopsis thaliana T-DNA mutants implicate GAUT genes in the biosynthesis of pectin and xylan in cell walls and seed testa. Molecular Plant, 2, 1000–1014. Calder, P.C. (1991) Glycogen structure and biogenesis. International Journal of Biochemistry, 23, 1335–1352. Cantarel, B.L., Coutinho, P.M., Rancurel, C., Bernard, T., Lombard, V., Henrissat, B. (2009) The Carbohydrate-Active EnZymes database (CAZy): an expert resource for glycogenomics. Nucleic Acids Research, 37, D233–D238. Chatterjee, M., Berbezy, P., Vyas, D., Coates, S., Barsby, T. (2005) Reduced expression of a protein homologous to glycogenin leads to reduction of starch content in Arabidopsis leaves. Plant Science, 168, 501–509. Cunningham, S.M., Nadeau, P., Castonguay, Y., Laberge, S., Volenec, J.J. (2003) Raffinose and stachyose accumulation, galactinol synthase expression, and winter injury of contrasting alfalfa germplasms. Crop Science, 43, 562–570. Deng, C., O’Neill, M.A., Hahn, M.G., York, W.S. (2009) Improved procedures for the selective chemical fragmentation of rhamnogalacturonans. Carbohydrate Research, 344, 1852–1857. Dong, Q., Lawrence, C.J., Schlueter, S.D., et al. (2005) Comparative plant genomics resources at PlantGDB. Plant Physiology, 139, 610–618. Downie, B., Gurusinghe, S., Dahal, P., et al. (2003) Expression of a GALACTINOL SYNTHASE gene in tomato seeds is up-regulated before maturation desiccation and again after imbibition whenever radicle protrusion is prevented. Plant Physiology, 131, 1347–1359. Emanuelsson, O., Nielsen, H., von Heijne, G. (1999) ChloroP, a neural network-based method for predicting chloroplast transit peptides and their cleavage sites. Protein Science, 8, 978–984. Fowler, S., Thomashow, M.F. (2002) Arabidopsis transcriptome profiling indicates that multiple regulatory pathways are activated during cold acclimation in addition to the CBF cold response pathway. Plant Cell, 14, 1675–1690. French, D. (1954) The raffinose family of oligosaccharides. Advances in Carbohydrate Chemistry, 9, 149–184. Frydman, R.B., Neufeld, E.F. (1963) Synthesis of galactosylinositol by extracts from peas. Biochemical and Biophysical Research Communications, 12, 121–125. Gibbons, B.J., Roach, P.J., Hurley, T.D. (2002) Crystal structure of the autocatalytic initiator of glycogen biosynthesis, glycogenin. Journal of Molecular Biology, 319, 463–477. Goda, H., Sawa, S., Asami, T., Fujioka, S., Shimada, Y., Yoshida, S. (2004) Comprehensive comparison of auxin-regulated and brassinosteroid-regulated genes in Arabidopsis. Plant Physiology, 134, 1555–1573. Goff, S.A., Ricke, D., Lan T-H., et al. (2002) A draft sequence of the rice genome (Oryza sativa L. ssp japonica). Science, 296, 92–100. Guan, Y., Nothnagel, E.A. (2004) Binding of arabinogalactan proteins by Yariv phenylglycoside triggers wound-like responses in Arabidopsis cell cultures. Plant Physiology, 135, 1346–1366. Guindon, S., Gascuel, O. (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology, 52, 696–704.

208

■

Plant Polysaccharides

Guo, A.Y., Zhu, Q.H., Chen, X., Luo, J.C. (2007) GSDS: a gene structure display server. Yi Chuan, 29, 1023–1026. Handley, L.W., Pharr, D.M., McFeeters, R.F. (1983a) Carbohydrate changes during maturation of cucumber fruit. Implications for sugar metabolism and transport. Plant Physiology, 72, 498–502. Handley, L.W., Pharr, D.M., McFeeters, R.F. (1983b) Relationship between galactinol synthase activity and sugar composition of leaves and seeds of several crop species. Journal of the American Society for Horticultural Science, 108, 600–605. Haritatos, E., Medville, R., Turgeon, R. (2000) Minor vein structure and sugar transport in Arabidopsis thaliana. Planta, 211, 105–111. Hitz, W.D., Carlson, T.J., Kerr, P.S., Sebastian, S.A. (2002) Biochemical and molecular characterization of a mutation that confers a decreased raffinosaccharide and phytic acid phenotype on soybean seeds. Plant Physiology, 128, 650–660. Holst, O. (2007) The structures of core regions from enterobacterial lipopolysaccharides – an update. FEMS Microbiology Letters, 271, 3–11. Hoover, R. (2001) Composition, molecular structure, and physicochemical properties of tuber and root starches: a review. Carbohydrate Polymers, 45, 253–267. Hurley, T.D., Stout, S., Miner, E., Zhou, J., Roach, P.J. (2005) Requirements for catalysis in mammalian glycogenin. Journal of Biological Chemistry, 280, 23892–23899. Jaillon, O., Aury J-M., Noel, B., et al. (2007) The grapevine genome sequence suggests ancestral hexaploidization in major angiosperm phyla. Nature, 449, 463–465. Katoh, K., Kuma, K., Toh, H., Miyata, T. (2005) MAFFT version 5: improvement in accuracy of multiple sequence alignment. Nucleic Acids Research, 33, 511–518. Keller, F., Matile, P. (1985) The role of the vacuole in storage and mobilization of stachyose in tubers of Stachys sieboldii. Journal of Plant Physiology, 119, 369–380. Keller, F., Pharr, D.M. (1996) Metabolism of carbohydrates in sinks and sources: galactosyl-sucrose oligosaccharides. In: Photoassimilate Distribution in Plants and Crops: Source-Sink Relationships (eds E. Zamsui, A.A. Schaffer), pp 157–183. Marcel Dekker, New York. Kim, M.S., Cho, S.M., Kang, E.Y., et al. (2008) Galactinol is a signaling component of the induced systemic resistance caused by Pseudomonas chlororaphis O6 root colonization. Molecular Plant Microbe Interaction, 21, 1643–1653. Ko J-H., Beers, E.P., Han K-H. (2006) Global comparative transcriptome analysis identifies gene network regulating secondary xylem development in Arabidopsis thaliana. Molecular Genetics and Genomics, 276, 517–531. Kong, Y., Zhou, G., Avci, U., et al. (2009) Two poplar glycosyltransferase genes, PdGATL1.1 and PdGATL1.2, are functional orthologs to PARVUS/AtGATL1 in Arabidopsis. Molecular Plant, 2, 1040–1050. Korytnyk, W., Metzler, E. (1962) Formation of raffinose and stachyose in lima beans (Phaseolus lunatus). Nature, 195, 616–617. Krogh, A., Larsson, B., von Heijne, G., Sonnhammer, E.L.L. (2001) Predicting transmembrane protein topology with a hidden Markov model: Application to complete genomes. Journal of Molecular Biology, 305, 567–580. Kubo, M., Udagawa, M., Nishikubo, N., et al. (2005) Transcription switches for protoxylem and metaxylem vessel formation. Genes and Development, 19, 1855–1860. Langeveld, S.M.J., Vennik, M., Kottenhagen, M., et al. (2002) Glucosylation activity and complex formation of two classes of reversibly glycosylated polypeptides. Plant Physiology, 129, 278–289.

Glycosyltransferases of the GT8 Family

■

209

Lao, N.T., Long, D., Kiang, S., et al. (2003) Mutation of a family 8 glycosyltransferase gene alters cell wall carbohydrate composition and causes a humidity-sensitive semi-sterile dwarf phenotype in Arabidopsis. Plant Molecular Biology, 53, 687–701. Lee, C., Zhong, R., Richardson, E.A., Himmelsbach, D.S., McPhail, B.T., Ye, Z.-H. (2007) The PARVUS gene is expressed in cells undergoing secondary wall thickening and is essential for glucuronoxylan biosynthesis. Plant and Cell Physiology, 48, 1659–1672. Lee, C., Teng, Q., Huang, W., Zhong, R., Ye, Z.-H. (2009) The poplar GT8E and GT8F glycosyltransferases are functional orthologs of Arabidopsis PARVUS involved in glucuronoxylan biosynthesis. Plant and Cell Physiology, 50, 1982–1987. Leplé, J.C., Dauwe, R., Morreel, K., et al. (2007) Downregulation of cinnamoylcoenzyme A reductase in poplar: Multiple-level phenotyping reveals effects on cell wall polymer metabolism and structure. Plant Cell, 19, 3669–3691. Liu, J.J., Odegard, W., de Lumen, B.O. (1995) Galactinol synthase from kidney bean cotyledon and zucchini leaf – purification and N-terminal sequences. Plant Physiology, 109, 505–511. Lomako, J., Lomako, W.M., Whelan, W.J. (2004) Glycogenin: the primer for mammalian and yeast glycogen synthesis. Biochimica Biophysica Acta, 1673, 45–55. Mandaokar, A., Kumar, V.D., Amway, M., Browse, J. (2003) Microarray and differential display identify genes involved in jasmonate-dependent anther development. Plant Molecular Biology, 52, 775–786. Minic, Z., Jamet, E., San-Clemente, H., et al. (2009) Transcriptomic analysis of Arabidopsis developing stems: a close-up on cell wall genes. BMC Plant Biology, 9, 6–24. Mohnen, D., Bar-Peled, M., Somerville, C. (2008) Cell wall polysaccharide synthesis. In: Biomass Recalcitrance – Deconstructing the Plant Cell Wall for Bioenergy (ed. D.E. Himmel), pp 94–187. Blackwell Publishing Ltd., Oxford. Mu, J., Cheng, C., Roach, P.J. (1996) Initiation of glycogen synthesis in yeast – requirement of multiple tyrosine residues for function of the self-glucosylating Glg proteins in vivo. Journal of Biological Chemistry, 271, 26554–26560. Müller, E., Hinckley, A., Rothfield, L. (1972) Studies of phospholipid-requiring bacterial enzymes. III. Purification and properties of uridine diphosphate glucose:lipopolysaccharide glucosyltransferase I. Journal of Biological Chemistry, 247, 2614–2622. Nishizawa, A., Yabuta, Y., Shigeoka, S. (2008) Galactinol and raffinose constitute a novel function to protect plants from oxidative damage. Plant Physiology, 147, 1251–1263. Nishizawa, A., Yabuta, Y., Yoshida, E., Maruta, T., Yoshimura, K., Shigeoka, S. (2006) Arabidopsis heat shock transcription factor A2 as a key regulator in response to several types of environmental stress. Plant Journal, 48, 535–547. Obendorf, R.L. (1997) Oligosaccharides and galactosyl cyclitols in seed desiccation tolerance. Seed Science Research, 7, 63–74. Orfila, C., Sørensen, S.O., Harholt, J., Geshi, N., Crombie, H., Truong H-N., Reid, J.S.G., Knox, J.P., Scheller, H.V. (2005) QUASIMODO1 is expressed in vascular tissue of Arabidopsis thaliana inflorescence stems, and affects homogalacturonan and xylan biosynthesis. Planta, 222, 613–622. Panikulangara, T.J., Eggers-Schumacher, G., Wunderlich, M., Stransky, H., Schöffl, F. (2004) Galactinol synthase1. A novel heat shock factor target gene responsible for

210

■

Plant Polysaccharides

heat-induced synthesis of raffinose family oligosaccharides in Arabidopsis. Plant Physiology, 136, 3148–3158. Paterson, A.H., Bowers, J.E., Bruggmann, R., et al. (2009) The Sorghum bicolor genome and the diversification of grasses. Nature, 457, 551–556. Peña, M.J., Zhong, R., Zhou G-K., et al. (2007) Arabidopsis irregular xylem8 and irregular xylem9: Implications for the complexity of glucuronoxylan biosynthesis. Plant Cell, 19, 549–563. Persson, K., Ly, H.D., Dieckelmann, M., Wakarchuk, W.W., Withers, S.G., Strynadka, N.C.J. (2001) Crystal structure of the retaining galactosyltransferase LgtC from Neisseria meningitidis in complex with donor and acceptor sugar analogs. Nature Structural Biology, 8, 166–175. Persson, S., Caffall, K.H., Freshour, G., Hilley, M.T., Bauer, S., Poindexter, P., Hahn, M.G., Mohnen, D., Somerville, C. (2007) The Arabidopsis irregular xylem8 mutant is deficient in glucuronoxylan and homogalacturonan, which are essential for secondary cell wall integrity. Plant Cell, 19, 237–255. Persson, S., Wei, H.R., Milne, J., Page, G.P., Somerville, C.R. (2005) Identification of genes required for cellulose synthesis by regression analysis of public microarray data sets. Proceedings of the National Academy of Sciences of the U S A, 102, 8633–8638. Peterbauer, T., Lahuta, L.B., Blöchl, A., et al. (2001) Analysis of the raffinose family oligosaccharide pathway in pea seeds with contrasting carbohydrate composition. Plant Physiology, 127, 1764–1772. Peterbauer, T., Mucha, J., Mach, L., Richter, A. (2002) Chain elongation of raffinose in pea seeds. Isolation, charcterization, and molecular cloning of a multifunctional enzyme catalyzing the synthesis of stachyose and verbascose. Journal of Biological Chemistry, 277, 194–200. Peterbauer, T., Richter, A. (2001) Biochemistry and physiology of raffinose family oligosaccharides and galactosyl cyclitols in seeds. Seed Science Research, 11, 185–197. Pharr, D.M., Sox, H.N., Locy, R.D., Huber, S.C. (1981) Partial characterization of the galactinol forming enzyme from leaves of Cucumis sativus L. Plant Science Letters, 23, 25–33. Qi, Y., Kawano, N., Yamauchi, Y., Ling, J., Li, D., Tanaka, K. (2005) Identification and cloning of a submergence-induced gene OsGGT (glycogenin glucosyltransferase) from rice (Oryza sativa L.) by suppression subtractive hybridization. Planta, 221, 437–445. Raetz, C.R.H., Whitfield, C. (2002) Lipopolysaccharide endotoxins. Annual Review of Biochemistry, 71, 635–700. Roach, P.J., Skurat, A.V. (1997) Self-glucosylating initiator proteins and their role in glycogen biosynthesis. Progress in Nucleic Acid Research and Molecular Biology, 57, 289–316. Robbins, N.S., Pharr, D.M. (1987) Regulation of photosynthetic carbon metabolism in cucumber by light intensity and photosynthetic period. Plant Physiology, 85, 592–597. Rojo, E., Denecke, J. (2008) What is moving in the secretory pathway of plants? Plant Physiology, 147, 1493–1503. Saravitz, D.M., Pharr, D.M., Carter, T.E., Jr. (1987) Galactinol synthase activity and soluble sugars in developing seeds of four soybean genotypes. Plant Physiology, 83, 185–189.

Glycosyltransferases of the GT8 Family

■

211

Schaff, J.E., Nielsen, D.M., Smith, C.P., Scholl, E.H., Bird, D.M. (2007) Comprehensive transcriptome profiling in tomato reveals a role for glycosyltransferase in Mimediated nematode resistance. Plant Physiology, 144, 1079–1092. Senser, M., Kandler, O. (1967) Occurence and distribution of galactinol in leaves of higher plants. Phytochemistry, 6, 1533–1540. Shao, M., Zheng, H., Hu, Y., et al. (2004) The GAOLAOZHUANGREN1 gene encodes a putative glycosyltransferase that is critical for normal development and carbohydrate metabolism. Plant and Cell Physiology, 45, 1453–1460. Singh, D.G., Lomako, J., Lomako, W.M., et al. (1995) β-Glucosylarginine: A new glucose-protein bond in a self-glucosylating protein from sweet corn. FEBS Letters, 376, 61–64. Smythe, C., Caudwell, F.B., Ferguson, M., Cohen, P. (1988) Isolation and structural analysis of a peptide containing the novel tyrosyl-glucose linkage in glycogenin. EMBO Journal, 7, 2681–2686. Sterling, J.D., Atmodjo, M.A., Inwood, S.E., et al. (2006) Functional identification of an Arabidopsis pectin biosynthetic homogalacturonan galacturonosyltransferase. Proceedings of the National Academy of Sciences of the U S A, 103, 5236–5241. Sterling, J.D., Quigley, H.F., Orellana, A., Mohnen, D. (2001) The catalytic site of the pectin biosynthetic enzyme α-1,4-galacturonosyltransferase is located in the lumen of the Golgi. Plant Physiology, 127, 360–371. Taji, T., Ohsumi, C., Iuchi, S., et al. (2002) Important roles of drought- and coldinducible genes for galactinol synthase in stress tolerance in Arabidopsis thaliana. Plant Journal, 29, 417–426. Tsabary, G., Shani, Z., Roiz, L., Levy, I., Riov, J., Shoseyov, O. (2003) Abnormal ‘wrinkled’ cell walls and retarded development of transgenic Arabidopsis thaliana plants expressing endo-1,4-beta-glucanase (cel1) antisense. Plant Molecular Biology, 51, 213–224. Tuskan, G.A., DiFazio, S., Jansson, S., et al. (2006) The genome of black cottonwood, Populus trichocarpa (Torr. and Gray). Science, 313, 1596–1604. Vlieghe, K., Vuylsteke, M., Florquin, K., et al. (2003) Microarray analysis of E2Fa-DPaoverexpressing plants uncovers a cross-talking genetic network between DNA replication and nitrogen assimilation. Journal of Cell Science, 116, 4249–4259. Volk, G.M., Haritatos, E.E., Turgeon, R. (2003) Galactinol synthase gene expression in melon. Journal of the American Society for Horticultural Science, 128, 8–15. Yin, Y., Chen, H., Hahn, M.G., Mohnen, D., Xu, Y. (2010) Evolution and function of the plant cell wall synthesis-related glycosyltransferase family 8. Plant Physiology, 153, 1729–1746. Yu, J., Hu, S.N., Wang, J., et al. (2002) A draft sequence of the rice genome (Oryza sativa L. ssp. indica). Science, 296, 79–92. Zhao T-Y., Thacker, R., Corum, J.W.I.I.I., et al. (2004) Expression of the maize GALACTINOL SYNTHASE gene family: (I) Expression of two different genes during seed development and germination. Physiologia Plantarum, 121, 634–646. Zimmermann, M.H. (1957) Translocation of organic substances in trees. I. The nature of the sugars in the sieve tube exudate of trees. Plant Physiology, 32, 288–291.

Annual Plant Reviews (2011) 41, 213–234 doi: 10.1002/9781444391015.ch7

Chapter 7

GENES AND ENZYMES OF THE GT31 FAMILY: TOWARDS UNRAVELLING THE FUNCTION(s) OF THE PLANT GLYCOSYLTRANSFERASE FAMILY MEMBERS Jack Egelund1, Miriam Ellis2, Monika Doblin2, Yongmei Qu2 and Antony Bacic2 1

University of Copenhagen, Department of Molecular Biology, Building 4-2-15, Ole Maaløes Vej 5, 2200 Copenhagen, Denmark 2 Plant Cell Biology Research Centre, School of Botany, The University of Melbourne, Vic 3010, Australia Manuscript received August 2008

Abstract: The galactosyltransferases (GalTs) have been extensively studied in mammals where they are involved in the synthesis of both N- and O-glycans on glycoproteins. In contrast, only a few studies have been published characterizing plant GalTs even though plants assemble many complex carbohydrates and glycoconjugates not found in other eukaryotes or bacteria, such as pectins, galactomannans, xyloglucans, arabinogalactan-proteins (AGPs), proline-rich proteins and extensins. Many enzymes characterized within family GT31 are mammalian and include the fringe proteins (GlcNAc-β-(1,3)-Fuc), the chondroitin synthases (GlcUA-β-(1,3)-GalNAc), and the β-(1,3)-GalTs. We attempted to categorize the putative β-(1,3)-GalTs and, where possible, we predicted their putative substrate specificity based on secondary structure and motifs shared with known β-(1,3)GalTs. Ninety-four plant sequences are assigned to CAZy family GT31, including 33 from Arabidopsis thaliana (At), and 39 from Oryza sativa (Os), but only one plant enzyme has as yet been biochemically characterized, At-GalT1, which is involved Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

213

214

■

Plant Polysaccharides

in the production of the Lewisa structure of N-glycans (Gal β-(1,3)-GlcNAc). Phylogenetic analysis identified 11 distinct clades, of which 4 are plant-specific. Clade 1 proteins contain the plant-specific DUF604 domain. Clade 7 is defined by a galectin-containing domain and both clades 7 and 10 contain GalT domains. The possible substrate specificity of these enzymes is predicted. Clade 11 proteins contain no obvious domains so no function can be assigned. Enzymes in clades 7 and 10 are speculated to be involved in the synthesis of both proteoglycans, such as AGPs, and N-glycans. Keywords: AGPs; CAZy family GT31; DUF604 galactosyltransferase; galectin; N-glycan; Lewisa structure

domain;

β-(1,3)-

7.1 Introduction Campbell et al. (1997) classified glycosyltransferases (GTs) into different families, currently 92 in CAZy (CAZy web server at http://www.cazy.org/ GlycosylTransferases.html), on the basis of the identity of the donor sugar, the relative acceptor/product stereochemistry, and sequence homologies (Amado et al. 1999; see also Coutinho and Henrissat 2011, Chapter 4). Plant researchers have attempted to first purify and then clone GTs that are involved in the synthesis of cell wall polysaccharides. However, few enzymes have been successfully purified and cloned using this approach because many GTs occur at low levels among a whole set of related enzymes and it is very difficult to retain their activity after solubilization (Edwards et al. 1999; Perrin et al. 1999). In contrast, bioinformatic approaches have allowed many putative GTs to be identified (e.g. Perrin et al. 2001). Thus it is possible to search the large amount of available genomic and expressed sequence tags (ESTs) data to identify GT candidate genes from different species based on their overall sequence homology, conserved motifs and structural features such as hydropathy profile and hydrophobic cluster analysis (HCA). Bioinformatic analyses are also useful for predicting the amino acids that confer catalytic activity and substrate specificity also forming the basis for future biochemical studies. Galactosyltransferases (GalTs) have been extensively studied in mammals. The mammalian GalTs are all type II membrane proteins and catalyse the addition of Gal to acceptors in various types of glycoconjugates through β-(1,4), β-(1,3), α-(1,3) and α-(1,4) linkages. In bacteria, at least 15 GalTs have been characterized and unlike their eukaryotic counterparts, they do not share the same topology: they can be either membrane-bound or soluble (Hennet 2002). In plants, membrane-bound GalTs are involved in the biosynthesis of pectins, galactomannans, xyloglucans and AGPs (McNab et al. 1968; Panayotatos and Villemez 1973; Mascara and Fincher 1982; Schibeci et al. 1984; Goubet et al. 1993; Goubet and Morvan 1994; Edwards et al. 1999; Geshi et al. 2000; Peugnet et al. 2001) as well as plastid galactolipids (Awai et al. 2001).

Genes and Enzymes of the GT31 Family

■

215

In addition to the classical N-glycans on glycoproteins, plants assemble many complex carbohydrates and glycoconjugates not found in other eukaryotes or bacteria (Oxley et al. 2004) including arabinogalactan-proteins (AGPs), proline-rich proteins, extensins and solanaceous lectins – the so called hydroxyproline-rich glycoproteins (HRGPs) (Showalter 1993; Nothnagel 1997; Johnson et al. 2003). The AGPs are generally regarded as the most complex of the HRGPs and are widely distributed in plant secretions and cell walls and at cell surfaces where they are attached to the plasma membrane via a glycosylphosphatidylinositol (GPI) anchor (Clarke et al. 1979; Schultz et al. 1998). The AGPs that have been characterized have been shown to contain less than 10% protein and more than 90% carbohydrate (w/w), which is typically rich in galactose (Gal) and arabinose (Ara). The polysaccharide chains in AGPs are β-(1,3)-galactans (Bacic et al. 1987; Tan et al. 2007) with β-(1,6)-galactopyranose (Galp) side chains terminated mostly with arabinofuranose (Araf) residues (Fincher et al. 1983). Some also contain glucuronic acids (GlcA) and short oligosaccharides of l-Araf (Qi et al. 1991; Goodrum et al. 2000; Kieliszewski 2001; Li and Kong 2004). Based on our current knowledge it is presumed that for the assembly of the galactan chains several GalT enzymes are involved and that these enzymes work in coordination to regulate the density, length and sequences of carbohydrate chains on AGPs. Identification and characterization of novel GalTs initiating and extending the polysaccharides of AGPs will therefore be of great scientific interest for expanding the current understanding of the biosynthesis of AGPs and the role of their glycan moieties in biological function. In this chapter we briefly describe the GT31 family members whose enzyme activity has been biochemically defined. On the basis of identified conserved regions in these proteins we have adopted a bioinformatic approach to identify and systematically characterize the putative GalTs responsible for synthesizing the β-(1,3)-Gal linkage from CAZy GT31. Furthermore, where possible, we have attempted to predict the possible substrate specificity of these GalTs based on secondary structure and conserved motifs shared with known β-(1,3)-GalTs (Qu et al. 2008). We would stress that these predictions, based upon bioinformatic studies, need to be experimentally validated.

7.2 Identification and characterization of the first β-(1,3)-GalTs 7.2.1 The first β-(1,3)-GT family members of non-plant origin Sasaki and co-workers (Japanese patent JP1994181759-A/1) identified the first β-(1,3)-GalT gene by expression cloning using mRNA from a melanoma cell line WM 26–4 to direct Lewisa and sialyl-Lewisa expression in Burkitt lymphoma Namalwa KJM-1 cells. Using β-(1,3)-GalT protein sequence to

216

■

Plant Polysaccharides

search human and mouse EST and genomic databases resulted in the identification and cloning of more than 10 homologous genes (Amado et al. 1999). In recent years, enzymatic activities using different sugar donors (UDP-Gal, UDP-GlcNAc, and UDP-GalNAc) and different acceptors (GlcNAc-, Gal-, GalNAc-, and Man-based) have been characterized, and all form β-(1,3)linkages (Amado et al. 1998; Hennet et al. 1998; Kolbinger et al. 1998; Isshiki et al. 1999; Zhou et al. 1999; Okajima et al. 2000; Bai et al. 2001; Shiraishi et al. 2001; Togayachi et al. 2001; Iwai et al. 2002; Müller et al. 2002; Schwientek et al. 2002). This has now led to the identification of a large homologous family of β-(1,3)-GTs, which include eight β-(1,3)-GalTs, six related β(1,3)-N-acetylglucosaminyltransferases (β-(1,3)-GlcNAcTs), two β-(1,3)-Nacetylgalactosaminyltransferases (β-(1,3)-GalNAcTs) (Amado et al. 1999; Hiruma et al. 2004), and Fringe, a β-(1,3)-GlcNAcT which transfers GlcNAc to fucosyl-Ser/Thr (Moloney et al. 2000). All these β-(1,3)-GTs belong to CAZY family GT31 (Table 7.1 and Fig. 7.1). 7.2.2 The first β-(1,3)-GalTs of plant origin Recently the first GT of plant origin in CAZy family GT31 was identified. At1g26810 (GALT1, Fig. 7.1, clade 7) possesses β-(1,3)-GalT activity involved in the formation of the Lewisa [Fuc-(1,4)-α-(Gal-(1,3)-β)-GlcNAc-R] structure on N-glycans in Arabidopsis thaliana (Strasser et al. 2007). Together with the work from Qu and co-workers (Qu et al. 2008), who adopted a bioinformatic approach to identify and systematically characterize the putative GalTs responsible for synthesizing the β-(1,3)-Gal linkage in Arabidopsis thaliana, these studies form the basis for future experiments to biochemically define the enzymatic activity of many of the plant-specific GTs in CAZy GT31. Their putative functions are discussed in Section 7.5.

7.3 Grouping of accessions based on their phylogenetic relationship CAZy family GT31 currently contains 521 entries, representing no less than 106 different organisms (http://www.cazy.org/GT31_all.html), that have been assigned the PFAM (PF01762) domain specific of GalTs, although in the vast majority of the accessions it is difficult to discern, as will be discussed below. The phylogenetic tree shown in Fig. 7.1 represents the evolutionary relationships between accessions of known activities (Table 7.1) and all the plant accessions; all other accessions have been left out on the illustration in order to generate an easily comprehensible overview of the family. As illustrated in Fig. 7.1, the accessions group into 11 clades (Table 7.1):

• Clade 1, accessions of plant origin harbouring a domain of unknown function (DUF604) identified by Qu et al. (2008), which will be elaborated upon in Section 7.5.3 • Clade 2, β-(1,3)-GlcTs involved in synthesis of O-fucosylglycan of TSR1.

At1g33250 At5g12460 At4g15240 At2g37730 At3g11420 At1g05280

1000

At4g00300 At1g01570 At4g11350 At4g23490 At5g41460 At1g07850

Os10g0516600 Os03g0124100 Os03g0269900 Os10g0534700 Os02g0681100 Os04g0578800

217

■

Genes and Enzymes of the GT31 Family

Clade 1 Os-Os08g0137300 Tp-BAE71200 Os-OSJNBa0071M09 Vv-CAN65210 DUF 604 domain Os-H0404F02_17 Mt-ABE82879 containing proteins Mt-ABE81721 Mt-ABN08498 Plant Specific Clade Mm-β3GTL β 1,3 Glucosyl-Clade 2 Hs-β3GTL transferases

1000 500

Dm-Fringe Mm-Radical-Fringe Hs-Manic-Fringe Mm-Lunatic-Fringe Mm-Manic-Fringe

1000

Chondroitin sulfate synthases

Hs-CSGlcA-T Hs-CSS1 Hs-CSS2 Hs-CSS3

708

709 999 1000

Clade 6

Sp-PVG3 1000 1000

1000

Os02g0577300 Os01g0328900 Os06g0229200

Clade 7

At3g06440 At1g26810-GalT1 Vv-CAN60055 Os03g0692500 Os03g0803600 Os03g0803900 Os07g0195200

At4g21060 At1g27120 AT1g74800 At5g62620

Accessions containing a GalT and a Galectin specific domain

Os08g0130900 Vv-CAN60867 Os-OJ1715_A07 Vv-CAN69092 Mt-ABE93312 Vv-CAN79615 Sb-AAL73538 Plant Specific Clade

1000 745

598 993 680

856

Hs-β3GnT5 Mm-β3GnT5 Hs-β3GalNAcT1 Hs-β3GalT1 Mm-β3GalNAcT1 Mm-β3GalT-I Gg-β3GalT1

Hs-β3GalT2 Hs-β3GalT5 Mm-β3GalT2 Mm-β3GalT5

Dm-Brainiac

1000

Hs-β3GalT4 Mm-β3GalT4 Rn-β3GalT4 Hs-β3GalNAcT2 Mm-β3GalNAcT2

1000 902

Hs-β3GalT6 Ce-Sqv2 Mm-β3GalT6 Ol-CAJ84717

1000

541 1000 1000

At4g32120 At5g53340 At2g25300

999

942 1000

Os060176200 Sl-CAD30015 Os08g0386700 Vv-CAN78789 Os-CAD44836 Vv-CAN66996

At1g11730 At1g05170 At2g32430 At1g32930 At4g26940

1000

1000

Clade 8 β 1,3 Galactosyl, GlcNAc, and GnT transferases

1000

Hs-β3GnT2 Hs-β3GnT3 Hs-β3GnT6 Hs-β3GnT8 790 Mm-β3GnT2 Hs-β3GnT4 Mm-β3GnT7

848 339

Clade 4

Clade 5 Core 1 β 1,3 Galactosyltransferases

Hs-C1Rn-C1β3GalT1β3GalT1 Ce-C1Mm-C1β3GalT1β3GalT1

1000

Clade 3

Fringe proteins

At3g14960 At1g53290 At2g26100 At5g57500 Os060192400 Os-OJ1081_G10.8 Os09g433000 Osg0432900-

At1g33430 At1g22015 At1g77810 Os01g0877400 Os08g0116900

Clade 10 Accessions containing a GalT specific domain

Os05g0427200 Os03g0577500 Os02g0566800 Nt-CAA06925 Os09g0452900 Vv-CAN63417 Os02g0164300 Os6g0679500

Os05g0199500 Vv-CAN63091 Os06g0156900 Vv-CAN65751 Hv-ABL11234

Os05g0552200 Os02g0785000 Os09g0433500 Os-OJ1316_E06 Os-OJ1715_H01

Clade 9

Vv-CAN80704 Zm-AAV64218

Plant Specific Clade Clade 11 Accessions containing no known domains Plant Specific Clade

Figure 7.1 Phylogenetic overview of CAZy family GT31. A standard sequence alignment from Clustal X version 1.8 (Thompson et al. 1997) was used to generate an unrooted tree, bootstrapping using 1000 replicates. To ensure an easy overview only major branch points are shown. Species names are abbreviated as follows: At, Arabidopsis thaliana; Ce, Caenorhabditis elegans; Dm, Drosophila melangaster; Gg, Gorilla gorilla; Hs, Homo sapiens; Hv, Hordeum vulgare; Mt, Medicago truncatula; Mm, Mus musculus; Nt, Nicotiana tabacum; Os, Oryza sativa; Rn, Rattus norvegicus; Sp, Schizosaccharomyces pombe; Sl, Solanum lycopersicum; Sb, Sorghum bicolour; Tp, Trifolium pratense; Vv, Vitis vinifera; Zm, Zea mays.

218

■

Plant Polysaccharides

Table 7.1 Glycosyltransferases of known function currently classified in CAZy family GT31 (http://www.cazy.org/fam/GT31.html)

Enzyme

Protein Clade name

β−(1,3)GlcT

Clade 2

Fringe proteins

Clade 3

Chondroitin sulfate synthases

Core β-(1,3)-GalT

Clade 4

Clade 5

Clade 6

β-(1,3)-GalT

Hs-β3GTL Mm-β3GTL Dm fringe Hs-manic fringe Mm-manic fringe Mm-radical fringe Mm-lunatic fringe Hs-CSS1

Hs-CSS2 Hs-CSS3 Hs-CSGlcA-T HsC1β3GalT1 MmC1β3GalT1 RnC1β3GalT1 CeC1β3GalT1 Sp-PVG3

Clade 7

At-GalT1

Clade 8

Hs-β3GalT1

Mmβ3GalT1 Hs-β3GalT2

Linkage

Synthesized glycans

Glc-β(1,3)-Fucα-1

O-fucosylglycan of TSR1

GlcNAc-β(1,3)-Fuc

O-fucosylglycan of EGF repeats

Accession

Reference

AB101481

1,2

AB253762 AAA64525

2,3 4,5,6

AAC51358 AAB71669 AAB71670 AAB71668 GlcUA-β(1,3)GalNAc

Gal-β(1,3)GalNAc

Gal-β(1,3)-Gal Gal-β(1,3)GlcNAc Gal-β(1,3)GlcNAc

Gal-β(1,3)GlcNAc

Chondroitin sulfate proteoglycans

AB023207

7

AB086063 AB086062 AB037823 AF155582

8 9 10 11

AF157962

11

AF157963

11,12

AF269063

11,13

Gal-β-(1,3) epitope of N-glycans Lea structure of N-glycans

CAB58972.1

14

At1g26810

15

Type 1 chain core structures of glycolipids and glycoproteins

E07739

16

AF029790

17

Y15060

16

Gal-β-(1,3)GalNAc-α-1, Ser/Thr Core 1 of O-glycan

Type 1 chain core structures of glycolipids and glycoproteins

Genes and Enzymes of the GT31 Family

■

219

Table 7.1 Continued Enzyme

Protein Clade name Mmβ3GalT2 Hs-β3GalT4 Mmβ3GalT4 Rn-β3GalT4 Hs-β3GalT5

Clade 9

β-(1,3)GlcNAcT

Clade 9 Clade 8

Clade 8

Mmβ3GalT5 Hs-β3GalT6 Mmβ3GalT6 Ce-Sqv2 Dm Brainiac

Hs-β3GnT2 Mmβ3GnT2 Hs-β3GnT3 Hs-β3GnT4 Hs-β3GnT5 Mmβ3GnT5 Hs-β3GnT6 Mmβ3GnT7 Hs-β3GnT8

β-(1,3)GalNAcT

Clade 8

Hsβ3GalNAcT1 Mmβ3GalNAcT1

Linkage

Gal-β(1,3)GalNAc

Gal-β(1,3)GalNAc

Gal-β(1,3)-Galb Gal-β(1,3)-Gal GlcNAc-β(1,3)-Man GlcNAc-β(1,3)-Gal GlcNAc-β(1,3)-Gal GlcNAc-β(1,3)-Gal GlcNAc-β(1,3)-Gal GlcNAc-β(1,3)-Gal GlcNAc-β(1,3)-Gal GlcNAc-β(1,3)GalNAc GlcNAc-β(1,3)-SO3Gal GlcNAc-β(1,3)-Gal GalNAc-β(1,3)-Gal

Synthesized glycans

GD1b/GM1/GA1 gangliosides

Gb5 globoside (SSEA-3)/ CA19-9 epitope of sLea

GAG linkage

GAG linkage

Accession

Reference

AF029791

17

Y15061

16

AF082504

18

AB003478 AB020337

19 20

AF254738

21

AY050570

22

AY050569

22

AY241927

23

Man-β-(1,4)-Glc AAF45918 core structure of arthroseries glycolipids Polylactosamine AB049584

24,25

Polylactosamine

AF092050

27

Core 1

AB049585

26

Polylactosamine

AB049586

26

Lc3Cer

AB045278

28

AY029203

29

26,27

Core 3 structure of O-glycans Keratan sulfate

AB073740

30

AF502429

31,32

Polylactosamine

ABI75895

33

Globoside

Y15062

16,17,34

AF029792

220

■

Plant Polysaccharides

Table 7.1 Continued Enzyme

Protein Clade name Clade 9

Hsβ3GalNAcT2

Linkage GalNAc-β(1,3)GlcNAc

Mmβ3GalNAcT2

Synthesized glycans GalNAc-β-(1,3)GlcNac-β-1-R

Accession

Reference

BC029564

35

AB116655

GAG, glycosaminoglycan; GalNAcT, N-acetylgalactosylaminosyltransferase; GalT, galactosyltransferase; GlcNAcT, N-acetylglucosaminosyltransferase; GlcT, glucosyltransferase; EGF, epidermal growth factor like; Lc3Cer, lactotriaosylceramide; sLea, sialyl Lewisa; SSEA, stage-specific embryonic antigen; TSR1, thrombospondin type 1 repeat. For species abbreviations see Fig. 7.1. References: 1, Kozma et al. (2006); 2, Sato et al. (2006); 3, Heinonen et al. (2006); 4, Moloney et al. (2000); 5, Brückner et al. (2000); 6, Rampal et al. (2005); 7, Kitagawa et al. (2001); 8, Yada et al. (2003a); 9, Yada et al. (2003b); 10, Gotoh et al. (2002); 11, Ju et al. (2002a) 12, Ju et al. (2002b); 13, Ju et al. (2006); 14, Andreishcheva et al. (2004); 15, Strasser et al. (2007); 16, Amado et al. (1998); 17, Hennet et al. (1998); 18, Daniotti et al. (1999); 19, Miyazaki et al. (1997); 20, Isshiki et al. (1999); 21, Zhou et al. (2000); 22, Bai et al. (2001); 23, Hwang et al. (2003); 24, Müller et al. (2002); 25, Schwientek et al. (2002); 26, Shiraishi et al. (2001); 27, Zhou et al. (1999); 28, Togayachi et al. (2001); 29, Henion et al. (2001); 30, Iwai et al. (2002); 31, Kataoka and Huh (2002); 32, Seko and Yamashita (2004); 33, Ishida et al. (2005); 34, Okajima et al. (2000); 35, Hiruma et al. (2004).

• Clade 3, β-(1,3)-GlcNAcTs involved in synthesis of O-fucosylglycan EGF repeats

• Clade 4, β-(1,3)-GlcATs involved in synthesis of chondroitin sulphate proteoglycans

• Clade 5, Core 1 β-(1,3)-GalTs involved in synthesis of Gal-β-(1,3)-GalNAcα-1, the Ser/Thr core 1 of O-glycans

• Clade 6, β-(1,3)-GalTs involved in synthesis of Gal-β-(1,3) epitopes of N-glycans

• Clade 7, galactoside-binding lectin (galectin) plant-specific accessions

• • • •

including the recently identified β-(1,3)-GalT from Arabidopsis thaliana involved in synthesis of the Lewisa structure of N-glycans (Strasser et al. 2007). The significance of the presence of both the galectin and GalT domain (manually assigned; Qu et al. 2008) which is found in all members of this clade will be elaborated upon in Sections 7.5.1 and 7.5.2. Clade 8, containing β-(1,3)-GalT/GlcNAcT/GnTs involved in the synthesis of various glycoprotein, proteoglycan and glycolipid structures (for details see Table 7.1) Clade 9, β-(1,3)-GalNAcT involved in the synthesis GalNAcT-β-(1,3)GlcNAc-β-1-R Clade 10, plant-specific accessions containing a GalT domain (manually assigned; Qu et al. 2008) which will be elaborated upon in Section 7.5.1 Clade 11, plant-specific accessions containing no known features/motifs, besides the PFAM (PF01762) domain. The relevance for further dissecting their specific function(s) is discussed in Section 7.5.4.

Genes and Enzymes of the GT31 Family

(A) TM N

ST

I

■

221

Catalytic domain II III IV V C

(B) TM N

ST

Galectin domain

I

II

III

IV

V C

HxNxR VxN

FxxG/A RxxxRXT/SW

DxD

GxxYxxS E/DDV GxW/C

Figure 7.2 Conserved motifs in the Arabidopsis thaliana GTs. Schematic domain structure of the two classes of putative plant β-(1,3)-GalT (illustration modified from Qu et al. 2008). A. Classical β-(1,3)-GTs. B, Galectin-containing domain β-(1,3)-GTs. Five conserved domains (I, II, III, IV and V) as identified by Qu et al. (2008). N-, N-terminus; C-, C-terminus; TM, transmembrane region; ST, stem region.

Interestingly clades 1, 7, 10 and 11 seem to be plant-specific (see Fig. 7.2 for a detailed overview of the four plant-specific clades). These four clades will be analysed further in Section 7.5.

7.4

Conserved motifs and implications for catalysis

7.4.1 Motif of the catalytic domain belongs to the GT-A fold superfamily Although the CAZy GTs that have been crystallized exhibit insignificant or at best very low sequence similarity, at the 3D structure level they adopt one of two folds: the GT-A fold or the GT-B fold (Coutinho et al. 2003, Wimmerova et al. 2003), presumably reflecting the evolutionary origin from a few precursor sequences. The GT-A fold was first described from an enzyme isolated from Bacillus subtillis (SpsA) and GT-B was first described from T4 phage (β-GlcT) (reviewed by Coutinho et al. 2003). Since the GT-B superfamily is not represented in CAZy GT31 they are not discussed further in this chapter. The GT-A fold has been described as having two tightly associated β/α/β domains. The existence of highly identical regions among different GTs has led to the identification of small peptide stretches or motifs that are highly conserved among all GTs. The DXD motif (often seen as e.g. TDD, DVD, EDD, DDD; Wiggins and Munro 1998; Shibayama et al. 1998) is highly conserved among GTs of the GT-A superfamily (Charnock et al. 2001). Based on crystal structure data (Breton et al. 2001), the DXD motif, located in a hydrophobic pocket, has been reported to interact mainly with the phosphate groups of the nucleotide donor through the coordination of a metal cation.

222

■

Plant Polysaccharides

The presence of a DXD motif alone, however, is not in itself sufficient to classify a protein as a GT, as evidenced for example by a search performed by Coutinho et al. (2003) in which it was revealed that 51% of all entries in SwissProt (Version 40) contained the DXD motif; within the CAZy database only 71% contain the DXD motif as well as the α-(1,4)-GalT (LgtC) from Neisseria meningitidis, which contains more than one DXD motif (reviewed by Gastinel 2001). The accessions in CAZy family GT31 all share the catalytic domain (DXD) of the GT-A superfamily. The slightly modified version of the DXD motif found in some accessions (e.g. IDN, DDN and DDN) may suggest the use of a donor substrate different from the other accessions. The catalytic function of the DXD motifs in the CAZy family GT31 accessions will, however, require experimental verification. 7.4.2 Sequence analysis and functional assignment of conserved motifs Sequence alignment (data not shown; Thompson et al. 1997) revealed several regions and amino acid residues that are highly conserved within the β-(1,3)GTs, although their substrate specificity and biological functions are different (Table 7.1). The relative locations of the domain structures and conserved motifs shared with the plant accessions in CAZy family GT31 are summarized in Fig. 7.2. The β-(1,3)-GTs share several conserved motifs and hydrophobic regions. Notably, none of these conserved regions, except the DXD motif, occurs in β-(1,4)- and α-(1,3)-GalT sequences. Towards the N-terminus is a conserved hydrophobic region, the trans-membrane domain (TMD), followed by a galectin-binding lectin domain (in some of the accessions). This is followed by the five major conserved motifs (Fig. 7.2):

• • • • •

I, RxxxRxT/SW II, FxxG/A III, the DXD motif IV, GxxYxxS V, E/DDV plus GxW/C

Towards the C-terminus, a variety of mutagenesis studies and mutants have proved that the three β-(1,3)-GT motifs (RxxxRxT/SW, DXD, E/DDV and GxW/C) are critical for catalysis (Munro & Freeman 2000; Griffitts et al. 2001; Hellberg, et al. 2002; Malissard et al. 2002). They are involved in sugar donor binding, acceptor binding and catalysis. Although some sequences, such as fucosyl-based β-(1,3)-GlcNAcTs (Fringe; Table 7.1 and Fig. 7.1, clade 3) GalNAc-based β-(1,3)-GalTs (Core 1 β-(1,3)-GalTs; Table 7.1 and Fig. 7.1, clade 5)- are highly divergent depending on their acceptors, most of these motifs are still recognizable in their sequences (data not shown; Ju et al. 2002a; Ju et al. 2002b; Correia et al. 2003). More interestingly, although

Genes and Enzymes of the GT31 Family

■

223

β-(1,3)-GlcATs also share weak sequence similarity with β-(1,3)-GTs (Kitagawa et al. 2001; Yada et al. 2003a; Yada et al. 2003b), the three critical motifs of β-(1,3)-GTs are identifiable in their sequences. A similar EDY/V motif also exists in bacterial β-(1,3)-GalTs and some members of the GT2 family (data not shown; Watanabe et al. 2002; Keenleyside et al. 2001), and it has therefore been speculated that the E residing in this motif might be a catalytic residue (Keenleyside et al. 2001) as in β-(1,3)-GlcAT. These results suggested that the maintenance of the β-(1,3)-linkage, instead of the donor substrate, has dictated the conservation of domains within these proteins (Hennet 2002). For a detailed discussion of these conserved motifs see Qu et al. (2008). 7.4.3 Implication of different substrate specificity for β-(1,3)-GTs Acceptor substrate specificity studies of β-(1,3)-GTs revealed significant differences in biological functions. The current knowledge of substrate specificity of the β-(1,3)-GT gene family is summarized in Table 7.1. The enzymes that transfer Gal to different sugar acceptors are clustered into different clades and are found in clades 5, 6, 7, 8 and 9 (Table 7.1), whereas GTs within the same clade may or may not share similar acceptor specificity (Table 7.1). Interestingly, β-(1,3)-GalTs which use Gal-based acceptors are more closely related to β-(1,3)-GlcNAcTs than other β-(1,3)-GalTs which use GlcNAcbased acceptors. This indicates that the recognition of the acceptor is the main driving force in the evolution of these proteins, instead of the donor substrate (Breton et al. 1998).

7.5

Domains conserved within the plant-specific clades

The vast majority of the plant-specific proteins all possess a hydrophobic membrane spanning domain close to the N-terminus indicative of typical type II membrane proteins, followed by a stem region that separates the N-terminal domain from the C-terminal catalytic region – a typical feature of Golgi localized GTs (Fig. 7.2; reviewed by Gibeaut 2000; Breton et al. 2001; Gastinel 2001; Keegstra & Raikhel 2001; Hennet 2002; Breton et al. 2006). In general, these GTs are composed of one or more functional domains, besides the PFAM (PF01762) domain, as will be discussed below. The identification of such domain(s) can therefore provide insights into the specific function of the protein. Using this knowledge we analysed the accessions from the four plant-specific clades (clades 1, 7, 10 and 11; Fig. 7.3) using the Protein Family (PFAM) database, which is a large collection of protein families (June 2008, 9318 families), each represented by multiple sequence alignments and hidden Markov models (Finn et al. 2006; http://pfam.sanger.ac.uk/).

224

■

A

Plant Polysaccharides

C

Clade 1 1000 1000

425 641 892 998

989 1000

732 1000 990 1000

529 960 998 1000 790 857 1000

509

1000

At1g33250 Os10g0516600 At5g12460 Vv-CAN65210 Os08g0137300 At4g15240 At1g05280 At2g37730 At3g11420 Os-OSJNBa0071M09 Os03g0269900 Os10g0534700 Os03g0124100 Os02g0681100 1000 Os04g0578800 Os-H0404F02_17 At4g00300 At1g01570 Mt-ABN08498 At4g11350 1000 At4g23490 At5g41460 Tp-BAE71200 At1g07850 Mt-ABE82879 1000 Mt-ABE81721

Clade 10 Os060176200 Vv-CAN78789 At2g25300 At4g32120 Vv-CAN66996 Sl-CAD30015

1000 999 1000 999

982

613 995

At5g53340

1000

1000

925

1000

1000 980 965 545 523

869

1000

481 1000

928 369

879 998 602 742

B

Clade 7

1000

999 914

1000 1000

774 1000 609 976

994

870 813 1000

974 923 1000

856 1000 1000 1000 962

997

Os02g0577300 At1g26810-GalT1 Os06g0229200 At3g06440 Os01g0328900 Vv-CAN60055 Os03g0692500 At1g27120 AT1g74800 At5g62620 Vv-CAN69092 Vv-CAN79615 Os03g0803900 Os07g0195200 Os03g0803600 Os-OJ1715_A07 Sb-AAL73538 Os08g0130900 At4g21060 Vv-CAN60867 Mt-ABE93312

970

1000 976 1000

D

Os08g0386700 Os-CAD44836 Os08g0116900 At1g11730 Os02g0164300 Os6g0679500 Os03g0577500 At4g26940 Vv-CAN63417 At1g05170 At2g32430 At1g32930 At1g33430 Os01g0877400 Os05g0427200 Os02g0566800 Os09g0452900 At1g22015 AT1G77810 Nt-CAA06925 Os05g0199500 At2g26100 Vv-CAN63091 Vv-CAN65751 At3g14960 At1g53290 Os06g0156900 Hv-ABL11234

Clade 11 1000 1000 1000

945 999 495 975 320 1000

Zm-AAV64218 Os060192400 At5g57500 Vv-CAN80704 Os-OJ1081_G10.8 Os09g433000 Osg0432900 Os05g0552200 Os02g0785000 Os09g0433500 Os-OJ1316_E06 Os-OJ1715_H01

Figure 7.3 Evolutionary relationship of the four plant-specific subclades (see Fig. 7.1) shown in detail. (A) Clade 1: proteins containing DUF 604 domain. (B) Clade 7: proteins containing galectin domain. (C) Clade 10. (D) Clade 11. Species names are abbreviated as follows: At, Arabidopsis thaliana; Hv, Hordeum vulgare; Mt, Medicago truncatula; Nt, Nicotiana tabacum; Os, Oryza sativa; Sl, Solanum lycopersicum; Tp, Trifolium pratense; Vv, Vitis vinifera; Zm, Zea mays.

As a result we found that three main domains could be identified: a galactosyltransferase (GalT) domain, a galactoside-binding lectin (galectin) domain and a domain of unknown function (DUF604) (for details see Qu et al. 2008). The importance of these domains is discussed in the following section.

Genes and Enzymes of the GT31 Family

■

225

7.5.1 Galactosyltransferase domain-containing clades 7 and 10 All accessions in the CAZy GT31 family have been assigned the PFAM GalT domain (http://pfam.sanger.ac.uk/) which includes the galactosyltransferases UDP-galactose:2-acetamido-2-deoxy-d-glucose-3βgalactosyltransferase and UDP-Gal:β-GlcNAc β-(1,3)-galactosyltransferase. Particular GalTs transfer Gal from UDP-Gal on to GlcNAc terminal residues in the synthesis of the lacto-series oligosaccharides types 1 and 2 (Hennet et al. 1998; Kolbinger et al. 1998). However, based upon E-values, the presence of the PFAM domain is weak for many of the accessions as illustrated by the recent attempts that manually identified only a total of 20 genes (Qu et al. 2008) in clades 7 and 10 in Arabidopsis thaliana as putative β-(1,3)-GalTs possibly involved in AGP biosynthesis (Fig. 7.1). These findings are further corroborated by the Interpro database (http://www.ebi.ac.uk/interpro/) which showed that the domain is widely distributed throughout all eukaryotes, as illustrated on the taxonomy chart (Fig. 7.4A), but is only present in 20 of the 33 Arabidopsis thaliana accessions of CAZy GT31 (Table 7.2) and, interestingly, this domain is only present in clades 7 and 10 of the plantspecific clades. 7.5.2 Galactoside-binding lectin (galectin) domain specific to clade 7 The galactoside-binding lectin (galectin) domain is defined by a conserved carbohydrate recognition domain (CRD). Galectins exclusively bind βgalactosides, like lactose, although they display a wide range of substrate specificities due to the structural differences in the CRD (Barondes et al. 1994; Dodd & Drickamer 2001). In mammals the basic ligand recognized by these galectins is either N-acetyl-lactosamine (Gal-β-(1,4)-GlcNAc), GalNAc, or Gal-β-(1,3)-GlcNAc, found on the ends of Asn N-linked and Ser/Thr Olinked oligosaccharides on various glycoproteins (Oxley et al. 2004) which are known to be developmentally regulated and may be involved in differentiation, cellular regulation and tissue construction, but their precise function is currently unknown. The galectin domain-containing GTs are mainly found in mammals, but not exclusively as seen on the taxonomy chart (Fig. 7.4B). Interestingly, the galectin domain (HxNxR, VxN, Fig. 7.2), can be identified in all the plantspecific accessions in clade 7 (Table 7.2). The plant-specific accessions have one CRD inserted between the stem region and catalytic domain (Fig. 7.2), which indicates that these accessions may be responsible for the addition of the first Gal residue to the Hyp residue of the backbone of members of the HRGP superfamily and act in a similar manner to the polypeptide N-acetylgalactosaminyltransferase in Homo sapiens (Hagen et al. 1999; Tenno et al. 2002). These galectin-domain-containing GTs are a large family of GalNAcTs that add GlcNAc to mammalian mucins and other protein backbones initiating O-glycosylation on either the nascent polypeptide or on a glycopeptide

A 6 23 32 247 18 47 162 24 31 539

B 6 36 68 353 8 41 230 34 51 416

C

62

Saccharomyces cerevisiae Fungi Caenorhabditis elegans Nematoda Metazoa Fruit fly Arthropoda Chordata Mouse Human Eukaryota

Unclassified Virus Archaea Bacteria Cyanobacteria Synechocystis PCC 6803 Oryza sativa (rice) Arabidopsis thaliana Green plants Plastid group Other eukaryotes

Saccharomyces cerevisiae Fungi Caenorhabditis elegans Nematoda Metazoa Fruit fly Arthropoda Chordata Mouse Human Eukaryota

Unclassified Virus Archaea Bacteria Cyanobacteria Synechocystis PCC 6803 Oryza sativa (rice) Arabidopsis thaliana Green plants Plastid group Other eukaryotes

Saccharomyces cerevisiae Fungi Caenorhabditis elegans Nematoda Metazoa Fruit fly Arthropoda Chordata Mouse Human Eukaryota

Unclassified Virus Archaea Bacteria Cyanobacteria Synechocystis PCC 6803 Oryza sativa (rice) Arabidopsis thaliana Green plants Plastid group Other eukaryotes

1

126 44 199 285 1

1 2

37 12 57 57

38 17 62 62

Figure 7.4 Distribution of conserved domains illustrated on taxonomy charts. (A) Members containing GalT domain (Interpro entry IPR002659). (B) Members containing galectin-binding lectin domain (Interpro entry IPR001079). (C) Members containing domain of unknown function DUF604 (Interpro entry IPR006740). The taxonomy charts provide an overview of the taxonomic range of the conserved domains associated with each Interpro entry and the number of sequences associated with each family. The circular display has the taxonomy-tree root as its centre. Selected model organisms populate the outer most circle. Nodes of the taxonomy-tree are placed on the inner circles. Radial lines lead to the description for each node (Mulder et al. 2007; http://www.ebi.ac.uk/interpro). Figures modified from Interpro (http://www.ebi.ac.uk/interpro).

Genes and Enzymes of the GT31 Family

■

227

Table 7.2 The plant-specific clades of CAZy family GT31 and their proposed functions

Subclade

TMD

GalT domaina

DUF domain

Galectin domain

1

Yesb

ND

Yesb

ND

7

Yesb

Yesb

ND

Yesb

10

Yesb

Yesb

ND

ND

11

Yesb

ND

ND

ND

Proposed function β-(1,3)-GalTs; substrate unknown β-GalTs involved in the synthesis of AGPsc β-(1,3)-GalTs involved in the synthesis of N- and O-glycans, and β-(1,3)GlcNAcTs; substrate unknown β-(1,3)-GalTs; substrate unknown

Subclades according to Fig. 7.1. Transmembrane domain (TMD), GalT specific domain (manually assigned; Qu et al. 2008) or domain of unknown function (DUF604 domain, manually assigned; Qu et al. 2008), and proposed catalytic activity and acceptor from Qu et al. (2008). ND, not detected. a The PFAM domain (PF01762) specific to GalT has been assigned to all accessions in the CAZy family GT31. b Represents the majority of the accessions. c This subclade also contains GALT1 (At1g26810), the β-(1,3)-GalT involved in synthesis of the Lewisa epitope on plant N-glycans (Strasser et al. 2007).

acceptor (Hassan et al. 2000). The catalytic units are known to have distinct acceptor substrate specificity based primarily on the peptide sequence, whereas the lectin (galectin) domain, in some isoforms, can modulate the substrate specificity of the catalytic domain to allow glycosylation of additional peptide acceptor sites flanked by the GalNAc residues. The precise mechanism of this lectin-induced activation of glycopeptide activity is the subject of speculation but the in vitro data suggest that the in vivo functions of some of the lectin-containing GlcNAcT isoforms is to act after the initiation of glycosylation by other isoforms to complete glycosylation of other Ser/Thr residues (Hassan et al. 2000). Strasser et al. (2007) recently demonstrated that a member of this clade (At1g26810; GALT1, Fig. 7.1)- adds β(1,3)-Gal to N-glycans to form the Lewisa epitope on plant glycoproteins. In contrast, Qu et al. (2008) were able to demonstrate that At1g77810, also a member of clade 7, possessed β-(1,3)-GalT activity as demonstrated by its capacity to add a Gal residue to Gal-β-(1,3)-GalOMe. Thus clade 7 members may have multiple and distinct functions in plant N- and O-glycan biosynthesis (see Table 7.2). 7.5.3 Domain of unknown function (DUF604) in clade 1 DUF604 is a conserved region found in several uncharacterized plant proteins, 17 from Arabidopsis thaliana and 38 from Oryza sativa, as determined by Interpro (illustrated on the taxonomy chart in Fig. 7.4C). Thus, a manual

228

■

Plant Polysaccharides

search with the accessions of CAZy GT31 identified the presence of the DUF604 in all accessions in clade 1, of which 11 belong to Arabidopsis thaliana (Table 7.1; Fig. 7.1). Although, the presence of the DUF604 domain in this clade does not currently help with the assignment of function to these novel GT members, it does indicate a strict plant-specific function. It is worth noting that if one searches TAIR (http://arabidopsis.org/ servlets/search) with the keyword ‘Fringe’ then 12 Arabidopsis genes are annotated as ‘Fringe-related’. This assignment most likely arises due to the automated detection of the PF02434 domain (http://pfcm.sanger.ac.uk/ search/sequence) in At1g33250, a clade 1 member (see Fig. 7.1), based upon rather weak E-values. This is also reflected by the poor E-values for the other accessions (approximately 0.013) and we therefore believe that their relationship to Fringe is circumspect. This highlights the dangers of using automated searches with no cut-off filter settings and why verification using manual control of these searches is always recommended. 7.5.4 No identifiable domains in accessions of clade 11 The remaining plant proteins which constitute accessions in clade 11 do not display any obvious domains. They do, however, contain a TMD region and when analysed manually the presence of a DXD motif, typical of GTs of the GT-A superfamily, can be found, indicating that these sequences are likely to be GTs. Because the information that can be extracted using bioinformatics is limited, ascribing a function to this clade is problematic.

7.6 Conclusions The structures of GTs have evolved to accommodate their specific nonreducing terminal sugar adaptively according to their reaction type, inverting or retaining (Negishi et al. 2003). The structural conservation of the various subgroups of β-(1,3)-GalTs may indicate functional conservation within each subfamily and could provide clues for further proving their activities and determining substrate specificity. In plants, most β-(1,3)-GalTs are expected to be Gal-based, since Gal residues are mainly attached to Gals in Gal-containing molecules. Strasser et al., (2007) identified the first β-(1,3)-GalT of plant origin as being involved in the assembly of the Lewisa epitope (see Sections 7.2.2 and 7.5.2). The major β-(1,3)-Gal-containing molecules in plants are the arabino-(3,6)-galactan (AG) chains on arabinogalactan-proteins (AGPs) (Schultz et al. 1998). It is therefore tempting to speculate that the plantspecific β-(1,3)-GalT-containing clades 7 and 10 are primarily involved in the assembly of the glycan moiety of this important class of plant proteoglycans (Table 7.2). We are now in a position to begin detailed biochemical and molecular studies to define their precise substrate specificities.

Genes and Enzymes of the GT31 Family

■

229

Acknowledgements This work was supported by a grant from the Australian Research Council (Discovery Projects DP0343454, DP0663374). Yongmei Qu acknowledges the support of a University of Melbourne International Research Scholarship and a University of Melbourne Research Scholarship. Jack Egelund was supported by the Danish Agency for Science Technology and Innovation (274-09-0082), the Carlsberg Foundation and the Villum Kann Rasmussen Research Center ‘Pro-Active Plants’. We acknowledge the excellent administrative support of Ms Joanne Noble.

References Amado, M., Almeida, R., Carneiro, F., et al. (1998) A family of human β3galactosyltransferases – Characterization of four members of a UDP-galactose: β-N-acetyl-glucosamine/β-N-acetyl-galactosamine β-1,3-galactosyltransferase family. Journal of Biological Chemistry, 273(21), 12770–12778. Amado, M., Almeida, R., Schwientek, T., Clausen, H. (1999) Identification and characterization of large galactosyltransferase gene families: galactosyltransferases for all functions. Biochimica et Biophysica Acta – General Subjects, 1473, 35–53. Andreishcheva, E.N., Kunkel, J.P., Gemmill, T.R., Trimble, R.B. (2004) Five genes involved in biosynthesis of the pyruvylated Galβ1,3-epitope in Schizosaccharomyces pombe N-linked glycans. Journal of Biological Chemistry, 279, 35644–35655. Awai, K., Maréchal, E., Block, M.A., et al., (2001) Two types of MGDG synthase genes, found widely in both 16:3 and 18:3 plants, differentially mediate galactolipid syntheses in photosynthetic and nonphotosynthetic tissues in Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the U S A, 98, 10960–10965. Bacic, A., Churms, S.C., Stephen, A.M., Cohen, P.B., Fincher, G.B. (1987) Finestructure of the arabinogalactan protein from Lolium multiflorum. Carbohydrate Research, 162, 85–93. Bai, X.M., Zhou, D.P., Brown, J.R., Crawford, B.E., Hennet, T., Esko, J.D. (2001) Biosynthesis of the linkage region of glycosaminoglycans – cloning and activity of galactosyltransferase II, the sixth member of the β1,3-galactosyltransferase family (β3GalT6). Journal of Biological Chemistry, 276, 48189–48195. Barondes, S.H., Castronovo, V., Cooper, D.N.W., et al. (1994) Galectins – a family of animal β-galactoside-binding lectins. Cell, 76, 597–598. Breton, C., Bettler, E., Joziasse, D.H., Geremia, R.A., Imberty, A. (1998) Sequencefunction relationship of prokaryotic and eukaryotic galactosyltransferases. Journal of Biochemistry, 123, 1000–1009. Breton, C., Mucha, J., Jeanneau, C. (2001) Structural and functional features of glycosyltransferases. Biochimie, 83, 713–718. Breton, C., Šnajdrová, L., Jeanneau, C., Koca, J., Imberty, A. (2006) Structures and mechanisms of glycosyltransferases. Glycobiology, 16, 29–37. Brückner, K., Perez, L., Clausen, H., Cohen, S. (2000) Glycosyltransferase activity of Fringe modulates Notch-Delta interactions. Nature, 406, 411–415. Campbell, J.A., Davies, G.J., Bulone, V., Henrissat, B. (1997) A classification of nucleotide-diphospho-sugar glycosyltransferases based on amino acid sequence similarities. Biochemical Journal, 326, 929–939.

230

■

Plant Polysaccharides

Charnock, S.J., Henrissat, B., Davies, G.J. (2001) Three-dimensional structures of UDP-sugar glycosyltransferases illuminate the biosynthesis of plant polysaccharides. Plant Physiology, 125, 527–531. Clarke, A.E., Anderson, R.L., Stone, B.A. (1979) Form and function of arabinogalactanproteins. Phytochemistry, 18, 521–540. Correia, T., Papayannopoulos, V., Panin, V., et al. (2003) Molecular genetic analysis of the glycosyltransferase Fringe in Drosophila. Proceedings of the National Academy of Sciences of the U S A, 100, 6404–6409. Coutinho, P.M., Henrissat, B. (2011) Annotating carbohydrate-active enzymes in plant genomes: present challenges. In: Annual Plant Reviews, Vol. 41, Plant polysaccharides (ed. P. Ulvskov), Ch. 4, pp. 93–108, Blackwell Publishing Ltd., Oxford. Coutinho, P.M., Deleury, E., Davies, G.J., Henrissat, H. (2003) An evolving hierarchical family classification for glycosyltransferases. Journal of Molecular Biology, 328, 307–317. Daniotti, J.L., Martina, J.A., Zurita, A.R., Maccioni, H.J. (1999) Mouse beta 1,3-galactosyltransferase (GA1/GM1/GD1b synthase): protein characterization, tissue expression, and developmental regulation in neural retina. Journal of Neuroscience Research, 58, 318–327. Dodd, R.B., Drickamer, K. (2001) Lectin-like proteins in model organisms: implications for evolution of carbohydrate-binding activity. Glycobiology, 11, 71–79. Edwards, M.E., Dickson, C.A., Chengappa, S., et al. (1999) Molecular characterisation of a membrane-bound galactosyltransferase of plant cell wall matrix polysaccharide biosynthesis. Plant Journal, 19, 691–697. Fincher, G.B., Stone, B.A., Clarke, A.E. (1983) Arabinogalactan-proteins – structure, biosynthesis, and function. Annual Review of Plant Physiology, 34, 47–70. Finn, R.D., Mistry, J., Schuster-Böckler, B., et al., (2006) Pfam: clans, web tools and services. Nucleic Acids Research, 34, 247–251. Gastinel, L.N. (2001) Galactosyltransferases: A structural overview of their function and reaction mechanisms. Trends in Glycoscience and Glycotechnology, 13, 131–145. Geshi, N., Jorgensen, B., Scheller, H.V., Ulvskov, P. (2000) In vitro biosynthesis of 1,4,-β-galactan attached to rhamnogalacturonan I. Planta, 210, 622–629. Gibeaut, D.M. (2000) Nucleotide sugars and glycosyltransferases for synthesis of cell wall matrix polysaccharides. Plant Physiology and Biochemistry, 38, 69–80. Goodrum, L.J., Patel, A., Leykam, J.F., Kieliszewski, M.J. (2000) Gum arabic glycoprotein contains glycomodules of both extensin and arabinogalactan-glycoproteins. Phytochemistry, 54, 99–106. Gotoh, M., Yada, T., Sato, T., et al. (2002) Molecular cloning and characterization of a novel chondroitin sulfate glucuronyltransferase that transfers glucuronic acid to N-acetylgalactosamine. Journal of Biological Chemistry, 277, 38179–38188. Goubet, F., Morvan, C. (1994) Synthesis of cell-wall galactans from flax (Linum usitatissimum L) suspension-cultured cells. Plant and Cell Physiology, 35, 719–727. Goubet, F., Martini, F., Thoiron, B., Morvan, C. (1993) Incorporation of D-[U-C-14] glucose and 14CO2 into the cell-wall polymers of flax hypocotyl, in the course of the fiber differentiation. Plant and Cell Physiology, 34, 841–848. Griffitts, J.S., Whitacre, J.L., Stevens, D.E., Aroian, R.V. (2001) Bt toxin resistance from loss of a putative carbohydrate-modifying enzyme. Science, 293, 860–864. Hagen, F.K., Hazes, B., Raffo, R., deSa, D., Tabak, L.A. (1999) Structurefunction analysis of the UDP-N-acetyl-D-galactosamine: polypeptide N-acet ylgalactosaminyltransferase. Journal of Biological Chemistry, 274, 6797–6803.

Genes and Enzymes of the GT31 Family

■

231

Hassan, H., Reis, C.A., Bennett, E.P., et al. (2000) The lectin domain of UDP-N-acetylD-galactosamine:polypeptide N-acetylgalactosaminyltransferase-T4 directs its glycopeptide specificities. Journal of Biological Chemistry, 275, 38197–38205. Heinonen, T.Y.K., Pelto-Huikko, M., Pasternack, L., Makl, M., Kainulainen, H. (2006) Murine ortholog of the novel glycosyltransferase, B3GTL: Primary structure, characterization of the gene and transcripts, and expression in tissues. DNA and Cell Biology, 25, 465–474. Hellberg, A., Poole, J., Olsson, M.L. (2002) Molecular basis of the globosidedeficient P-k blood group phenotype – Identification of four inactivating mutations in the UDP-N-acetylgalactosamine: Globotriaosylceramide 3-beta-Nacetylgalactosaminyltransferase gene. Journal of Biological Chemistry, 277(33), 29455–29459. Henion, T.R., Zhou, D., Wolfer, D.P., Jungalwala, F.B., Hennet, T. (2001) Cloning of a mouse β1,3-N-acetylglucosaminetransferase GlcNAc(β1,3)Gal(β1,4) Glc–ceramide synthase gene encoding the key regulator of lacto–series glycolipid biosynthesis. Journal of Biological Chemistry, 276, 30261–30269. Hennet, T. (2002) The galactosyltransferase family. Cellular and Molecular Life Sciences, 59, 1081–1095. Hennet, T., Dinter, A., Kuhnert, P., Mattu, T.S., Rudd, P.M., Berger, E.G., (1998) Genomic cloning and expression of three murine UDP-galactose: β-Nacetylglucosamine β-1,3-galactosyltransferase genes. Journal of Biological Chemistry, 273, 58–65. Hiruma, T., Togayachi, A., Okamura, K., et al. (2004) A novel human β1,3-N-acetylg alactosaminyltransferase that synthesizes a unique carbohydrate structure, GalNAc β1–3GlcNAc. Journal of Biological Chemistry, 279, 14087–14095. Hwang, H.Y., Olson, S.K., Brown, J.R., Esko, J.D., Horvitz, H.R. (2003) The Caenorhabditis elegans genes sqv-2 and sqv-6, which are required for vulval morphogenesis, encode glycosaminoglycan galactosyltransferase II and xylosyltransferase. Journal of Biological Chemistry, 278, 11735–11738. Ishida, H., Togayachi, A., Sakai, T., et al. (2005) A novel β1,3-Nacetylglucosaminyltransferase (β3Gn-T8), which synthesizes poly-N-acetyllactosamine, is dramatically upregulated in colon cancer. FEBS Letters, 579, 71–78. Isshiki, S., Togayachi, A., Kudo, T., et al. (1999) Cloning, expression, and characterization of a novel UDP-galactose: β-N-Acetylglucosamine β1,3-galactosyltransferase (β3Gal-T5) responsible for synthesis of type 1 chain in colorectal and pancreatic epithelia and tumor cells derived therefrom. Journal of Biological Chemistry, 274, 12499–12507. Iwai, H., Masaoka, N., Ishii, T., Satoh, S. (2002) A pectin glucuronosyltransferase gene is essential for intercellular attachment in the plant meristem. Proceedings of the National Academy of Sciences of the U S A, 99, 16319–16324. Johnson, K.L., Jones, B.J., Bacic, A., Schultz, C.J. (2003) The fasciclin-like arabinogalactan proteins of Arabidopsis. A multigene family of putative cell adhesion molecules. Plant Physiology, 133, 1911–1925. Ju, T., Brewer, K., D’Souza, A., Cummings, R.D., Canfield, W.M. (2002a) Cloning and expression of human core 1 β1,3-galactosyltransferase. Journal of Biological Chemistry, 277, 178–186. Ju, T., Cummings, R.D., Canfield, W.M. (2002b) Purification, characterization, and subunit structure of rat core 1 β1,3-galactosyltransferase. Journal of Biological Chemistry, 277, 169–177.

232

■

Plant Polysaccharides

Ju, T., Zheng, Q., Cummings, R.D. (2006) Identification of core 1 O-glycan T-synthase from Caenorhabditis elegans. Glycobiology, 16, 947–958. Kataoka, K., Huh, N.H. (2002) A novel β1,3-N-acetylglucosaminyltransferase involved in invasion of cancer cells as assayed in vitro. Biochemical and Biophysical Research Communications, 294, 843–848. Keegstra, K., Raikhel, N. (2001) Plant glycosyltransferases. Current Opinion in Plant Biology, 4: 219–224. Keenleyside, W.J., Clarke, A.J., Whitfield, C. (2001) Identification of residues involved in catalytic activity of the inverting glycosyl transferase WbbE from Salmonella enterica serovar borreze. Journal of Bacteriology, 183, 77–85. Kieliszewski, M.J. (2001) The latest hype on Hyp-O-glycosylation codes. Phytochemistry, 57, 319–323. Kitagawa, H., Uyama, T., Sugahara, K. (2001) Molecular cloning and expression of a human chondroitin synthase, Journal of Biological Chemistry, 276, 38721–38726. Kolbinger, F., Streiff, M.B., Katopodis, A.G. (1998) Cloning of a human UDPgalactose:2-acetamido-2-deoxy-d-glucancose 3-beta-galactosyltransferase catalyzing the formation of type 1 chains. Journal of Biological Chemistry, 273, 433–440. Kozma, K., Keusch, J.J., Hegemann, B., et al. (2006) Identification and characterization of a β1,3-glucosyltransferase that synthesizes the Glc-β-1,3-Fuc disaccharide on thrombospondin Type 1 repeats. Journal of Biological Chemistry, 281, 36742–36751. Li, A.X., Kong, F.Z. (2004) Syntheses of arabinogalactans consisting of beta-(1–6)linked d-galactopyranosyl backbone and alpha-(1–3)-linked L-arabinofuranosyl side chains. Carbohydrate Research, 339, 1847–1856. Malissard, M., Dinter, A., Berger, E.G., Hennet, T. (2002) Functional assignment of motifs conserved in beta 1,3-glycosyltransferases – A mutagenesis study of murine UDP-galactose:beta-N-acetylglucosamine beta 1,3-galactosyltransferase-I. European Journal of Biochemistry, 269, 233–239. Mascara, T., Fincher, G.B. (1982) Biosynthesis of arabinogalactan-protein in Lolium multiflorum (Ryegrass) endosperm cells – In vitro incorporation of galactosyl residues from UDP-galactose into polymeric products. Australian Journal of Plant Physiology, 9, 31–45. McNab, J.M., Villemez, C.L., Albersheim, P. (1968) Biosynthesis of galactan by a particulate enzyme preparation from Phaseolus aureus seedlings. Biochemical Journal, 106, 355–360. Miyazaki, H., Fukumoto, S., Okada, M., Hasegawa, T., Furukawa, K., Furukawa, K. (1997) Expression cloning of rat cDNA encoding UDP-galactose:GD2 β1,3galactosyltransferase that determines the expression of GD1b/GM1/GA1. Journal of Biological Chemistry, 272, 24794–24799. Moloney, D.J., Panin, V.M., Johnston, S.H., et al. (2000) Fringe is a glycosyltransferase that modifies Notch. Nature, 406, 369–375. Mulder, N.J., Apweiler, R., Attwood, T.K., et al. (2007) New developments in the InterPro database. Nucleic Acids Research, 35, 224–228. Müller, R., Altmann, F., Zhou, D., Hennet, T. (2002)The Drosophila melanogaster brainiac protein is a glycolipid specific β1,3 N-acetylglucosaminyltransferase. Journal of Biological Chemistry, 277, 32417–32420. Munro, S., Freeman, M. (2000) The Notch signalling regulator Fringe acts in the Golgi apparatus and requires the glycosyltransferase signature motif DxD. Current Biology, 10, 813–820.

Genes and Enzymes of the GT31 Family

■

233

Negishi, M., Dong, J., Darden, T.A., Pedersen, L.G., Pedersen, L.C. (2003) Glucosaminylglycan biosynthesis: what we can learn from the X-ray crystal structures of glycosyltransferases GlcAT1 and EXTL2. Biochemical and Biophysical Research Communications, 303, 393–398. Nothnagel, E.A. (1997) Proteoglycans and related components in plant cells. International Review of Cytology, 174, 195–291. Okajima, T., Chen, H.H., Ito, H., et al. (2000) Molecular cloning and expression of mouse GD1α/GT1aα/GQ1bα synthase (ST6GalNAc VI) gene. Journal of Biological Chemistry, 275, 6717–6723. Oxley, D., Currie, G.C., Bacic, A. (2004) Analysis of carbohydrate from glycoproteins. In: Purifying Proteins for Proteomics: A Laboratory Manual, (ed R.J. Simpson), ch 25, pp. 579–636, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, USA. Panayotatos, N., Villemez, C.L. (1973) Formation of a β-(1–4)-D-galactan chain catalyzed by a Phaseolus aureus enzyme. Biochemical Journal, 133(2):263–271. Perrin, R., Wilkerson, C., Keegstra, K. (2001) Golgi enzymes that synthesize plant cell wall polysaccharides: finding and evaluating candidates in the genomic era. Plant Molecular Biology, 47, 115–130. Perrin, R.M., DeRocher, A.E., Bar-Peled, M., et al. (1999) Xyloglucan fucosyltransferase, an enzyme involved in plant cell wall biosynthesis. Science, 284, 1976–1979. Peugnet, I., Goubet, F., Bruyant-Vannier, M.P., et al. (2001) Solubilization of rhamnogalacturonan I galactosyltransferases from membranes of a flax cell suspension. Planta, 213, 435–445. Qi, W., Fong, C., Lamport, D.T.A. (1991) Gum-arabic glycoprotein is a twisted hairy rope – a new model based on O-galactosylhydroxyproline as the polysaccharide attachment site. Plant Physiology, 96, 848–855. Qu, Y., Egelund, J., Gilson, P.R., et al. (2008) Identification of a novel group of putative Arabidopsis thaliana β-(1,3)-galactosyltransferases. Plant Molecular Biology, 68, 43–59. Rampal, R., Li, A.S.Y., Moloney, D.J., et al. (2005) Lunatic fringe, manic fringe, and radical fringe recognize similar specificity determinants in O-fucosylated epidermal growth factor-like repeats. Journal of Biological Chemistry, 280, 42454–42463. Sato, T., Sato, M., Kiyohara, K., et al., (2006) Molecular cloning and characterization of a novel human β1,3-glucosyltransferase, which is localized at the endoplasmic reticulum and glucosylates O-linked fucosylglycan on thrombospondin type 1 repeat domain. Glycobiology, 16, 1194–1206. Schibeci, A., Pnjak, A., Fincher, G.B. (1984) Biosynthesis of arabinogalactan-protein in Lolium multiflorum (Italian ryegrass) endosperm cells – subcellular distribution of galactosyltransferases. Biochemical Journal, 218, 633–636. Schultz, C., Gilson, P., Oxley, D., Youl, J., Bacic, A. (1998) GPI-anchors on arabinogalactan-proteins: implications for signalling in plants. Trends in Plant Science, 3, 426–431. Schwientek, T., Keck, B., Levery, S.B., et al. (2002) The Drosophila gene brainiac encodes a glycosyltransferase putatively involved in glycosphingolipid synthesis. Journal of Biological Chemistry, 277, 32421–32429. Seko, A., Yamashita, K. (2004) β1,3-N-acetylglucosaminyltransferase-7 (3Gn-T7) acts efficiently on keratan sulfate-related glycans. FEBS Letters, 556, 216–220.

234

■

Plant Polysaccharides

Shibayama, K., Ohsuka, S., Tanaka, T., Arakawa, Y., Ohta, M. (1998) Conserved structural regions involved in the catalytic mechanism of Escherichia coli K-12 WaaO (RfaI). Journal of Bacteriology, 180, 5313–5318. Shiraishi, N., Natsume, A., Togayachi, A., et al. (2001) Identification and characterization of three novel β1,3-N-acetylglucosaminyltransferases structurally related to the β1,3-galactosyltransferase family. Journal of Biological Chemistry, 276, 3498–3507. Showalter, A.M. (1993) Structure and function of plant-cell wall proteins. Plant Cell, 5, 9–23. Strasser, R., Bondili, J.S., Vavra, U., et al. (2007) A unique 1,3-Galactosyltransferase is indispensable for the biosynthesis of N-glycans containing Lewis a structures in Arabidopsis thaliana. Plant Cell, 19, 2278–2292. Tan, L., Xu, J., Lamport, D., et al. (2007) The AGP Hyp-arabinogalactan backbone is a folded polysaccharide of reverse-turn hairpins generated by β-(1–6)-linked repeats of β-(1–3) trigalactosyl units. Copenhagen 2007 XIth Cell Wall Meeting. Abstract. Physiologia Plantarum, 130, 68. Tenno, M., Toba, S., Kezdy, F.J., Elhammer, A.P., Kurosaka, A. (2002) Identification of two cysteine residues involved in the binding of UDP-GalNAc to UDP-GalNAc: polypeptide N-acetylgalactosaminyltransferase 1(GalNAc-T1). European Journal of Biochemistry, 269, 4308–4316. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G. (1997) The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research, 24, 4876–4882. Togayachi, A., Akashima, T., Ookubo, R., et al. (2001) Molecular cloning and characterization of UDP-GlcNAc:lactosylceramide β1,3-N-acetylglucosaminyltransferase (beta 3Gn-T5), an essential enzyme for the expression of HNK-1 and Lewis X epitopes on glycolipids. Journal of Biological Chemistry, 276, 22032–22040. Watanabe, M., Miyake, K., Yanae, K., et al. (2002) Molecular characterization of a novel beta 1,3-galactosyltransferase for capsular polysaccharide synthesis by Streptococcus agalactiae type Ib. Journal of Biochemistry, 131, 183–191. Wiggins, C.A.R., Munro, S. (1998) Activity of the yeast MNN1 alpha-1,3mannosyltransferase requires a motif conserved in many other families of glycosyltransferases. Proceedings of the National Academy of Sciences of the U S A, 95, 7945–7950. Wimmerova, M., Engelsen, S.B., Bettler, E., Breton, C., Imberty, A. (2003) Combining fold recognition and exploratory data analysis for searching for glycosyltransferases in the genome of Mycobacterium tuberculosis. Biochimie., 85, 691–700. Yada, T., Gotoh, M., Sato, T., et al. (2003a) Chondroitin sulfate synthase-2: Molecular cloning and characterization of a novel human glycosyltransferase homologous to chondroitin sulfate glucuronosyltransferase, which has dual enzymatic activities. Journal of Biological Chemistry, 278, 30235–30247. Yada, T., Sato, T., Kaseyama, H., et al. (2003b) Chondroitin sulfate synthase-3: Molecular cloning and characterization. Journal of Biological Chemistry, 278, 39711–39725. Zhou, D., Dinter, A., Gallego, R.G., et al. (1999) A β-1,3-N-acetylglucosaminyltransferase with poly-N-acetyllactosamine synthase activity is structurally related to β-1,3galactosyltransferases. Proceedings of the National Academy of Sciences of the U S A, 96, 406–411. Zhou, D., Henion, T.R., Jungalwala, F.B., Berger, E.G., Hennet, T. (2000) The β1,3galactosyltransferase β3GalT-V is a stage-specific embryonic antigen-3 (SSEA-3) synthase. Journal of Biological Chemistry, 275, 22631–22634.

Annual Plant Reviews (2011) 41, 235–250 doi: 10.1002/9781444391015.ch8

Chapter 8

GLYCOSYLTRANSFERASES OF THE GT34 AND GT37 FAMILIES Kenneth Keegstra and David Cavalier MSU-DOE Plant Research Laboratory, Michigan State University, Plant Biology Building, East Lansing, MI 48824, USA Manuscript received February 2009

Abstract: CAZy family GT37 contains 10 genes in Arabidopsis and a similar or higher number in other plant species. One of the Arabidopsis genes encodes a fucosyltransferase that is involved in xyloglucan biosynthesis. The biological role of the other genes in this family is not known, but it has been postulated that they all encode fucosyltransferases that are responsible for the addition of fucose to various cell wall polymers. If this hypothesis is correct, all the genes in this family encode proteins that perform the same biochemical function, i.e. transfer of fucose, but their biological roles are different. In contrast, the genes in CAZy family GT34 are known to encode proteins with two different biochemical functions. Some of the genes encode a galactosyltransferase that is involved in galactomannan biosynthesis. Other genes from this family encode xylosyltransferases that are involved in xyloglucan biosynthesis. Despite utilizing different donor molecules, both types of enzymes in this family utilize similar acceptors, creating an α-1,6linked branch on the β-1,4-linked backbone. Keywords: arabinogalactan protein; fucosyltransferase; galactomannan; galactosyltransferase; xyloglucan; xylosyltransferase

8.1

Introduction

The biosynthesis of plant cell wall matrix polysaccharides occurs in the Golgi apparatus and involves the participation of glycan synthases that produce polysaccharide backbones and glycosyltransferases that add the side chains (Perrin et al. 2001). Glycosyltransferases move a sugar from an Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

235

236

■

Plant Polysaccharides

activated donor molecule, usually a nucleotide diphosphate sugar, to a specific acceptor molecule. The specificity of these enzymes is such that in most cases the product is no longer an acceptor for the transfer of additional sugar residues. The plant glycosyltransferases characterized to date are type II integral membrane proteins with a single transmembrane domain near the N-terminus and a large globular domain that extends into the Golgi lumen and contains the active site of the enzyme (Keegstra & Raikhel 2001, Lerouxel et al. 2006). Although genes encoding many putative glycosyltransferases have been identified as a result of the massive amounts of DNA sequence information that are now available, relatively few of the enzymes have been characterized in sufficient detail that we know their biochemical function or their biological role in cell wall biosynthesis. In very few cases have the donor and acceptor substrate specificity of these enzymes been characterized experimentally. In many cases, the location of the protein within the cells has not been determined nor has their topology in the membrane been established. Consequently workers in the plant cell wall field have a bolus of candidate genes, but a great deal of biochemistry and cell biology needs to be done to identify the biochemical and biological functions of these candidates. The first two Golgi enzymes involved in plant cell wall polysaccharide biosynthesis to be identified and characterized at the molecular level were a fucosyltransferase from CAZy family GT37 (http://www.cazy.org/) that is involved in xyloglucan biosynthesis (Perrin et al. 1999) and a galactosyltransferase from CAZy family GT34 (http://www.cazy.org/) that is involved in galactomannan biosynthesis (Edwards et al. 1999). This chapter provides a brief review of this early work and provides a status report on the current understanding of other plant enzymes that are members of these two families of glycosyltransferases.

8.2 Family GT37 enzymes 8.2.1 Xyloglucan fucosyltransferase (FUT1) Xyloglucans are complex cell wall polysaccharides that are composed of a β-1,4-linked glucan backbone that is substituted with α-1,6-linked xylosyl residues (see Fig. 8.1) in highly recurrent patterns. Based upon the xylosyl residue substitution patterns, most xyloglucans can be classified as being either ‘XXXG-type’ or ‘XXGG-type’ xyloglucans (Vincken et al. 1997; Hoffman et al. 2005). The xylosyl residues are further substituted with other sugars depending upon the plant species. In many dicot plants, the side chain adjacent to the unsubstituted glucosyl residue of the XXXG-repeat may contain a terminal fucosyl residues attached to a galactosyl residue (see Fig. 8.1). Using seed storage xyloglucan, which lacks fucosyl residues, as an acceptor substrate, Maclachlan and his colleagues developed an assay to measure xyloglucan fucosyltransferase activity (Camirand et al. 1987, Farkas

Glycosyltransferases of the GT34 and GT37 Families

■

237

a-L-Fucp 1 ↑

2 b-D-Galp 1

b-D-Galp 1

↑

α-D-Xylp 1 ↑

2 a-D-Xylp 1

2 a-D-Xylp 1 ↑

↑ ↑

a-D-Xylp 1 ↑

6 6 6 6 →4)-b-D-Glcp-(1→4)-b-D-Glcp-(1→4)-b-D-Glcp- (1→4)-b-D-Glcp-(1→4)-b-D-Glcp-(1→4)-b-D-Glcp-(1→4)-b-D-Glcp-(1→4)-b-D-Glcp-(1 6 6 ↑ ↑ 1 1 a-D-Xylp a-D-Xylp 2 ↑ 1 b-D-Galp

Figure 8.1 Typical structure of xyloglucan polysaccharide from dicot plants. For structural details of xyloglucan polymers from individual plants see the information in Hoffman et al. (2005) and Vincken et al. (1997) and references in these papers.

& Maclachlan 1988). The availability of a reliable assay with the proper acceptor substrates is an important step that allows biochemical investigations of the glycosyltransferase. Indeed, the activity of many cell wall biosynthetic enzymes still cannot be measured because suitable acceptor substrates are lacking. Maclachlan and his colleagues went on to characterize this enzyme activity and demonstrate that it was present in the Golgi fraction of plant cell extracts (Brummell et al. 1990). Wulff et al. (2000) later demonstrated that the active site of this enzyme is located in the lumen of the Golgi apparatus. Perrin et al. (1999) used the assay developed by Maclachlan’s group to purify the fucosyltransferase activity from pea microsomes (originally called FT1, but now referred to as FUT1). Although the highly purified preparation still contained two proteins, amino acid sequence information was obtained from both of them. Sequence information from one of the two proteins lead to the identification of an Arabidopsis EST and a full- length cDNA clone corresponding to that EST was shown to produce a protein with xyloglucan fucosyltransferase activity when expressed in mammalian COS cells, thereby confirming that the cDNA encoded the proper enzyme. Faik et al. (2000) extended this work by isolating a pea cDNA clone encoding the purified pea protein and showing that the predicted protein had 62% sequence identity at the amino acid level to the predicted Arabidopsis protein (see further discussion on sequence comparisons below). In addition, Faik et al. (2000) performed a detailed characterization of the highly purified pea enzyme with respect to its enzymatic properties and its specificity for both donor and acceptor substrates. They demonstrated that the purified enzyme was specific for xyloglucan as an acceptor and added fucose only to a single galactosyl residue in tamarind xyloglucan to produce the oligosaccharide XLFG when the polymeric product was digested with

238

■

Plant Polysaccharides

endo-glucanase (see Fry et al. (1993)for an explanation of the nomenclature used in describing these oligosaccharides). Vanzin et al. (2002) examined the in vivo role of the Arabidopsis protein and demonstrated that it is responsible for fucosylation of xyloglucan. They demonstrated that mur2 mutant plants contain a missense mutation in the gene encoding the xyloglucan fucosyltransferase protein. Extracts from the mutant plants lack detectable xyloglucan fucosyltransferase activity and the xyloglucan from the walls of the mutant plants lacks detectable fucose. Despite these biochemical defects, the mur2 plants grow and develop normally, with the only morphological defect being collapsed trichome papillae (Vanzin et al. 2002). Perrin et al. (2003) extended the in vivo work by investigating the impact of a knock-out mutation and of overexpression of the gene encoding xyloglucan fucosyltransferase in Arabidopsis. Consistent with the results of Vanzin et al. (2002), Perrin et al. (2003) found that xyloglucan isolated from roots and leaves of the knock-out mutant lacked fucose. Interestingly, they observed that overexpression of the gene encoding xyloglucan fucosyltransferase resulted in elevated levels of enzyme activity but did not increase the levels of fucose in the xyloglucan isolated from the walls of these plants. The reasons for this discrepancy remain to be determined. Finally, Perrin et al. (2003) used RT-PCR and promoter GUS fusion constructs to examine the expression patterns and found that the gene encoding xyloglucan fucosyltransferase was expressed at highest levels in rapidly growing tissues, which is consistent with its role in cell wall biosynthesis. Once the sequence of the Arabidopsis genome became available, Sarria et al. (2001) identified a series of related genes in Arabidopsis. They demonstrated that Arabidopsis contains nine additional genes related to the one encoding the xyloglucan fucosyltransferase (called AtFUT1); the additional genes were labelled AtFUT2 through AtFUT10 (see Fig. 8.2). The genetic

Figure 8.2 Phylogenetic analysis of select members of CAZy family GT37. Protein sequences from select taxa found in CAZy family GT37, including A. thaliana (At), M. truncatula (Mt), O. sativa (Os), P. sativum (Ps), and P. tremula × P. alba (PtxPa), were used to construct a phylogenetic tree. The rice sequences were recovered from the database described by Cao et al. (2008), http://ricephylogenomics.ucdavis.edu/cellwalls/gt/). Fulllength protein sequences were aligned as described by Liepman et al. (2007). Phylograms were constructed from the aligned sequences using the neighbour-joining method (Saitou & Nei 1987). The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates with values greater than 50% displayed) are shown next to the branches (Felsenstein, 1985). The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Poisson correction method (Zuckerkandl & Pauling 1965) and are in units of the number of amino acid substitutions per site. Phyogenetic analyses were conducted in MEGA4 (version 4.0.2; Tamura et al. 2007). FUT, fucosyltransferase.

Glycosyltransferases of the GT34 and GT37 Families AtFUT5 At2g15370

99 98

AtFUT10 At2g15350

99

AtFUT4 At2g15390

87

AtFUT6 At1g14080

100

AtFUT7 At1g14070 AtFUT8 At1g14100 PsFUT1 AtFUT2 At2g03210 AtFUT1 At2g03220

63

64

PtxPaFUT1 AtFUT3 At1g74420 PtxPaFUT4

100 67

PtxPa FUT6 PtxPa FUT3 MtFUT1

96

75

PtxPaFUT5 PtxPaFUT2

98

PtxPaFUT7

100

Os02g52640 Os03g50800 Os06g10950

99 71

Os06g10960 Os06g10930

76

Os02g52630

67

Os10g03650

99

Os06g10910

99

Os06g10970 100

Os02g17534 Os02g17600 Os02g52610.1

54

100 Os02g52610.2

Os02g52590 Os02g52560

97

Os06g10980

86

Os02g25630

96 99

Os0437650 Os04g37640 Os09g28460 Os06g10920

74 100

0.1

Os08g24750

■

239

240

■

Plant Polysaccharides

evidence provided by Vanzin et al. (2002) and Perrin et al. (2003) provide compelling evidence that AtFUT1 is the only gene involved in the fucosylation of xyloglucan, at least in Arabidopsis. As noted above, the FUT1 genes from Arabidopsis and pea are not closely related (only 62% identity at the amino acid level, and see Fig. 8.2). Interestingly, recent work with pea has lead to some observations that could be explained if pea has more than one gene encoding a xyloglucan fucosyltransferase (Wen et al. 2008). More complete EST or genomic sequence information from pea will be needed to evaluate this possibility. In summary, xyloglucan fucosyltransferase genes have been identified in many species and the enzymes from pea and Arabidopsis have been characterized. These enzymes are highly specific, adding a fucosyl residue to a single galactosyl residue within the repeating unit of xyloglucan (see Fig. 8.1) (Perrin et al. 1999; Faik et al. 2000). The pea enzyme is located in the Golgi and the active site faces the Golgi lumen (Wulff et al. 2000). The evidence is compelling that Arabidopsis has a single gene encoding this activity and disruption of this gene has minimal impact on plant morphology and physiology (Vanzin et al. 2002; Perrin et al. 2003). It is possible that this conclusion is unique to Arabidopsis and that other species may have additional genes encoding this activity and that disruption of their functions may have biological consequences (Wen et al. 2008). 8.2.2 Other FUT genes If one accepts the evidence that AtFUT1 is the only gene encoding a xyloglucan fucosyltransferase, this raises the question of what is the biochemical function and the biological role of each of the other nine Arabidopsis FUT genes. Similar questions need to be answered for the many homologues present in other plants. While there is no published information to answer these questions, some speculation and some unpublished results are available that may shed some light on the roles of these other homologues. The first point to make is that many of these genes are annotated in the various databases as xyloglucan fucosyltransferases. This annotation is not based on any biochemical or biological information and is almost certainly incorrect. In the speculation presented below, the hypothesis is made that all of the other genes in family GT37 encode proteins with fucosyltransferase activity. While this is a reasonable hypothesis, experimental evidence is needed to support it. But if one accepts this hypothesis, then the next issue is to identify the acceptor molecules. One can prepare a list of candidate acceptors by using structural information to identify the cell wall components that are known to contain fucose. These include rhamnogalacturonan I and rhamnogalacturonan II (For a review see Mohnen (2008)), and arabinogalactan proteins (AGPs) (Tsumuraya et al. 1984; Misawa et al. 1996; van Hengel & Roberts 2002). While the carbohydrate portions of N-linked glycoproteins

Glycosyltransferases of the GT34 and GT37 Families

■

241

are known to contain fucose, the enzymes studied to date that add these fucosyl residues belong to CAZy family GT10 (Wilson et al. 2001; Strasser et al. 2004). At present there is no information on which, if any, of the genes shown in Fig. 8.2 encode enzymes that add fucosyl residues to rhamnogalacturonan I or rhamnogalacturonan II. However, unpublished work from our group provides evidence that FUT4 and possibly FUT6 add fucosyl residues to AGPs (T. Wagner, N. Raikhel & K. Keegstra, manuscript in preparation). While the biological role of fucosyl residues on AGPs is unknown, given the postulated role of AGPs in intercellular recognition and communication, it is attractive to consider the possibility that fucose plays an important role in the biological functions of these molecules (van Hengel & Roberts 2003). Certainly, fucose is known to play an important role in intercellular recognition events in many organisms including humans. If fucosyl residues play similar important roles in plants, and if CAZy family GT37 are all fucosyltransferases, then this would be an important category of enzymes to investigate in detail. Along these lines, it is interesting to note that in the rice genome, CAZy family GT37 has expanded significantly compared to Arabidopsis and contains 22 genes (Fig. 8.2) (Cao et al. 2008; see http://ricephylogenomics. ucdavis.edu/cellwalls/gt/ for details). Although most of the rice genes are annotated as encoding a xyloglucan fucosyltransferase, this seems very unlikely, given the low levels of xyloglucan in rice and the lack of fucose in rice xyloglucan; see Hoffman et al. (2005) and references therein. Indeed, based on the absence of fucosylated xyloglucan in rice and the lack of a rice gene with high sequence similarity to AtFUT1, Yokoyama & Nishitani (2004) predict that rice does not have a xyloglucan fucosyltranferase. Thus, the expansion of family GT37 genes in rice may reflect an importance of fucosyl residues in other cell wall molecules, possibly in AGPs. The significant expansion of family GT37 genes when comparing rice to Arabidopsis raises interesting and important questions about the biological roles of these genes and the proteins that they encode.

8.3

Family GT34 enzymes

8.3.1 Galactomannan galactosyltransferase J. S. Grant Reid, Mary Edwards and their colleagues performed a series of elegant, detailed studies on the biosynthesis of galactomannan polysaccharides in the developing seeds of several legume species (Meier & Reid 1982; Edwards et al. 1989; Edwards et al. 1992). These polysaccharides have a β1,4-linked mannan backbone that is substituted in a variable pattern by side chains consisting of α-1,6-linked galactosyl residues (Fig. 8.3) (Meier & Reid 1982, Edwards et al. 1992). Synthesis of the galactomannans requires a

a-D-Galp 1

Plant Polysaccharides a-D-Galp 1 ↑

■

a-D-Galp 1 ↑

242

↑

6 6 6 →4)-b-D-Manp-(1→4)-b-D-Manp-(1→4)-b-D-Manp- (1→4)-b-D-Manp-(1→4)-b-D-Manp-(1→4)-b-D-Manp-(1→4)-b-D-Manp-(1→4)-b-D-Manp-(1 6 ↑ 1 a-D-Galp

Figure 8.3 Typical structure of galactomannans from the endosperm walls of various seeds. See the review by Meier & Reid (1982) for details.

mannan synthase and a galactosyltransferase (Edwards et al. 1989). Recent work by several groups has established that the mannan synthase enzymes are encoded by cellulose synthase-like proteins from the A family (Dhugga et al. 2004; Liepman et al. 2005; Liepman et al. 2007). With respect to the galactosyltransferase responsible for adding the side chains to the mannan backbone, biochemical methods were used to solubilize and then purify the enzyme activity from fenugreek (Trigonella foenumgraecum L) (Edwards et al. 1999). Like the work with the xyloglucan fucosyltransferase summarized above, the purification of the galactosyltransferase required an assay to follow purification of the enzyme. The assay used by Edwards et al. (1999) employed UDP-Gal as the donor substrate and β-1,4-linked mannan oligosaccharides that must be at least five mannosyl residues in length as the acceptor; oligosaccharides with eight mannosyl residues were the optimum acceptors. Peptides were isolated from the purified protein and information derived from sequencing these peptides was used to synthesize oligonucleotide primers that were used to isolate a cDNA clone encoding the protein. Expression of the cDNA clone in Pichia pastoris confirmed that the cDNA encoded a protein with the proper enzyme activity. Computer analysis of the deduced amino acid sequence demonstrated that the full-length protein was predicted to have a single transmembrane domain near the N-terminus. Subsequent analysis of this and related sequences demonstrated that they are classified in CAZy family GT34 (Campbell et al. 1997; Coutinho et al. 2003). Edwards et al. (2002) performed a detailed analysis of the substrate specificity of the fenugreek galactosyltransferase as well as a detailed characterization of the products. They concluded that the solubilized enzyme from fenugreek adds a single α-1→6 linked galactosyl residue to the oligosaccharide acceptor. The location of the galactosyl residue was dependent upon the length of the acceptor molecule, but it was not added at random locations. Based on their analysis of the products, they proposed that the solubilized enzyme binds to six adjacent mannosyl residues and transfers a galactosyl residue specifically to the third residue from the non-reducing end of the oligosaccharide (Edwards et al. 2002). Reid et al. (2003) demonstrated that the fenugreek gene is involved in galactomannan biosynthesis in vivo. They expressed the fenugreek galactosyltransferase gene in tobacco plants under the control of the 35S promoter

Glycosyltransferases of the GT34 and GT37 Families

■

243

and demonstrated that the galactomannan that normally accumulates in the endosperm cell walls with a very low level of galactosyl substitution had a much higher level of galactosyl substitution in the transgenic plants. Edwards et al. (2004) provided further evidence that a related gene is involved in galactomannan biosynthesis in vivo. They use overexpression and antisense methods to generate transgenic lines of Lotus japonicus that had elevated or reduced levels of its endogenous galactomannan galactosyltransferase activity. The mannans produced by the resulting transgenic plants had elevated or reduced levels of galactosyl residues, consistent with the conclusion that this enzyme is required for the addition of the side chain residues of the endosperm galactomannans. The galactosyltransferase genes involved in galactomannan biosynthesis have now been identified from several different species, including coffee, which stores galactomannan in its seeds (Pré et al. 2008). The enzymes from the legume seeds have been characterized in detail by J. S. Grant Reid and his colleagues (Edwards et al. 1992; Edwards et al. 1999; Edwards et al. 2002). In those plant species where genomic sequence information is available, family GT34 is relatively small in size; seven genes in Arabidopsis (see Fig. 8.4) and ten genes in rice (Cao et al. 2008; http://ricephylogenomics. ucdavis.edu/cellwalls/gt/). But as described in more detail below, this family contains at least two different enzyme activities. 8.3.2 Xyloglucan xylosyltransferase Faik et al. (2002) developed an in vitro assay for the xyloglucan xylosyltransferase activity that adds the xylosyl residues to the xyloglucan backbone (see Fig. 8.1). They observed that the xylosyltransferase activity from pea microsomes had similar properties to the galactosyltransferase activity reported by Edwards et al. (1999). Specifically, they noted that transfer of xylose from UDP-xylose required the presence of an acceptor consisting of a β-1,4-linked glucan oligosaccharide that was at least five glucosyl residues long (Faik et al. 2002). Based on this observation, they hypothesized that some of the Arabidopsis genes from CAZy family GT34 might be involved in xyloglucan biosynthesis (see Fig. 8.4). Indeed, one gene, At3g62720 (called AtXXT1 in Fig. 8.4) produced xylosyltransferase activity when it was expressed in Pichia pastoris (Faik et al. 2002). Preliminary characterization of the products provided evidence consistent with the hypothesis that this enzyme was involved in xyloglucan biosynthesis. Cavalier and Keegstra (2006) extended these observations and performed a detailed characterization of the enzymes produced by two Arabidopsis genes, At3g62720 and Atg4g02500 (called AtXXT1 and AtXXT2, respectively, in Fig. 8.4). Each gene could be expressed successfully in several heterologous systems, but the enzymes produced in insect cells using the baculovirus system were used for most for most of their studies. The specificity of each enzyme for acceptor and donor substrates was examined;

244

■

Plant Polysaccharides

AtXXT1 At3g62720

100

AtXXT2 At4g02599

100

Os03g18820.1 100 Os03g18820.2 97

100 Os02g32750.1 96

Os02g32750.2 Os12g05380.1

92

AtGT3 At5g07720

100 100 100

AtGT4 At1g18690 AtXXT5 At1g74380 Os03g19310 Os03g19330

100

Os02g49140

99

Os11g34390 100

96

AtGT6 At4g37690 AtGT7 At2g22900 TfGMGT

100 100

LjGMGT Os07g23494.1

0.2

Figure 8.4 Phylogenetic analysis of select members of CAZy family GT34. Protein sequences from select taxa found in CAZy family GT34, including A. thaliana (At), L. japonicus (Lj), O. sativa (Os), and T. foenum-graecum (Tf), were used to construct a phylogenetic tree. Full-length protein sequences were aligned as described by Liepman et al. (2007). Phylograms were constructed from the aligned sequences using the neighbourjoining method (Saitou & Nei, 1987). The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates with values greater than 50% displayed) are shown next to the branches (Felsenstein, 1985). The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Poisson correction method (Zuckerkandl & Pauling, 1965) and are in units of the number of amino acid substitutions per site. Phyogenetic analyses were conducted in MEGA4 (version 4.0.2; Tamura et al., 2007). GMGT; galactomannan galactostltransferase; GT; putative glycosyltransferase; XXT, xyloglucan xylosyltransferase.

each transferred xylose from UDP-xylose to cellooligosaccharides, with cellohexaose being the preferred acceptor. If the ratio of donor to acceptor was kept high and the reactions were allowed to proceed for longer times, each enzyme was capable of adding multiple xylosyl residues to a single acceptor molecule. The first xylose residue is added to the fourth glucosyl residue

Glycosyltransferases of the GT34 and GT37 Families

■

245

from the reducing end of cellohexaose while the second xylose is added to the third glucosyl residue from the reducing end of the oligosaccharide. Like the earlier work by Faik et al. (2002), the observations of Cavalier and Keegstra (2006) are consistent with the hypothesis that these two enzymes are involved in xyloglucan biosynthesis. More recently, Cavalier et al. (2008) provided genetic evidence that AtXXT1 and AtXXT2 are involved in xyloglucan biosynthesis in vivo; Zabotina et al. (2008) provided genetic evidence that AtXXT5 is involved in xyloglucan biosynthesis. In each case these authors used reverse genetics in Arabidopsis to explore the biological functions of these genes. Single mutants of AtXXT1 and AtXXT2 had no morphological defects and the levels of xyloglucan were reduced only slightly (Cavalier et al. 2008). This is not surprising given the biochemical evidence that the two genes encoded proteins with very similar functions (Cavalier & Keegstra 2006). However, in double mutant plants lacking both AtXXT1 and AtXXT2, xyloglucan could not be detected using a variety of biochemical and cell biological methods. Surprisingly, the plants grew and developed relatively normally. The major defects were slightly slower rates of development, slightly smaller stature of mature plants and a lack of normal root hairs. One major conclusion of these observations is that AtXXT1 and AtXXT2 play a critical role in the biosynthesis of xyloglucan. Secondly, and more importantly, these observations raise important fundamental questions about how the double mutant plants can survive as well as they do without detectable xyloglucan. A single mutant lacking AtXXT5 was morphologically normal, but had biochemical defects in the levels and substitution patterns of xyloglucan (Zabotina et al. 2008). Based on these observations the authors concluded that this protein is a xylosyltransferase involved in adding some of the xylosyl residues to xyloglucan. Given the regular pattern of xylose substitution in the xyloglucan polymer (see Fig. 8.1) one important question that remains to be answered is which of these three enzymes is involved in adding which of the xylosyl residues present in xyloglucan. Another important question is whether the xylosyl transferases form a complex with the glycan synthase that forms the backbone of xyloglucan. The later enzyme is thought to be encoded by a CslC gene (Cocuron et al. 2007). Cocuron et al. (2007) provided evidence that the activity of the the CslC protein is impacted by the presence of the XXT1 protein, leading these authors to speculate that the two proteins form a complex in the Golgi membrane. This attractive hypothesis is depicted in Fig. 8.1 in the review by Lerouxel et al. (2006). The hypothesis of a functional complex involving xyloglucan xylosyltransferase and glucan synthase is consistent with earlier biochemical observations that both UDP-glucose and UDP-xylose are needed to allow in vitro synthesis of xyloglucan (Ray 1980); the biochemical evidence in favour of such complexes was reviewed by Hayashi (1989). It will also be interesting to determine whether similar complexes might be involved in the synthesis of other polysaccharides, for example between the galactosyltransferase and mannan synthase that are responsible for galactomannan biosynthesis.

246

■

Plant Polysaccharides

While two of the seven Arabidopsis genes in CAZy family GT34 have been shown to encode enzymes with xyloglucan xylosyltransferase activity (Faik et al. 2002; Cavalier & Keegstra 2006), the biochemical activity of the other five proteins remains unknown. Reverse genetic studies provide evidence that XXT5 is involved in xyloglucan biosynthesis in vivo, although in vitro evidence that it encodes a xylosyltransferase are still lacking (Zabotina et al. 2008). Based on the clustering of the sequences within this family (see Fig. 8.4), it is reasonable to postulate that the other two genes in the top portion of the tree (labelled GT3 and GT4 in Fig. 8.4) also encode xylosyltransferases, possibly involved in xyloglucan biosynthesis in some way. On the other hand, it is also reasonable to postulate that the two genes in the bottom of the tree (labelled GT6 and GT7 in Fig. 8.4) encode galactosyltransferases involved in galactomannan biosynthesis. This hypothesis is based on the similarity of these sequences to the genes from fenugreek and lotus that are known to encode galactomannan galactosyltransferases (see Fig. 8.4). Work is under way in our lab to test this prediction experimentally.

8.4 Concluding comments The genes in CAZy families GT34 and GT37 provide an interesting contrast to each other. The current evidence is that the genes in family GT37 all encode fucosyltransferase enzymes. That is, they all transfer l-fucose from GDP-fucose to a variety of different acceptor molecules. In the case of FUT1 from pea and Arabidopsis, the enzymes transfer fucose to a specific galactosyl residue on xyloglucan. While the acceptor specificity of the other members of this family has not been determined, preliminary unpublished evidence supports the conclusion that some members of this family transfer fucose to arabinogalactan proteins. The sugar to which the fucose was attached has not been determined for the Arabidopsis enzyme. However, fucosylated arabinogalactan proteins from radish, a close relative of Arabidopsis, contain fucose α-1,2-linked to arabinosyl residues (Tsumuraya et al. 1984). So, it is possible that the Arabidopsis enzyme transfers the L-fucose to L-arabinose. Regardless of the details, the important conclusion is that all of the family GT37 enzymes appear to transfer the same sugar to a variety of different acceptor molecules. In contrast, the family GT34 genes encode enzymes that transfer different sugars, galactose or xylose, from different acceptor molecules, UDP-galactose or UDP-xylose, to very similar acceptor molecules. The galactosyltransferase identified as the original member of family GT34 transfers galactose from UDP-galactose to an oligosaccharide consisting of β-1,4-linked mannosyl residues in vitro (Edwards et al. 1999). It is assumed that in vivo the enzyme catalyses the transfer of galactose to a growing mannan chain. However, the xylosyltransferases identified more recently (Faik et al. 2002, Cavalier & Keegstra 2006) transfer xylose from UDP-xylose to β-1,4-linked glucan

Glycosyltransferases of the GT34 and GT37 Families

■

247

oligosaccharides in vitro and presumably to a growing glucan chain in vivo. Thus, the members of this family recognize different donor molecules, but recognize very similar acceptor molecules to form the same α-1,6 linkages.

References Brummell, D., Camirand, A., Maclachlan, G. (1990) Differential distribution of xyloglucan glycosyl transferases in pea Golgi dictyosomes and secretory vesicles. Journal of Cell Science, 96, 705–710. Camirand, A., Brummell, D., Maclachlan, G. (1987) Fucosylation of xyloglucan: Localization of the transferase in dictyosomes of pea stem cells. Plant Physiology, 84, 753–756. Campbell, J., Davies, G., Bulone, V., Henrissat, B. (1997) A classification of nucleotidediphospho-sugar gylcosyltransferases based on amino acid sequence similarities. Biochemical Journal, 326, 929–939. Cao, P., Bartley, L., Jung, K., Ronald, P. (2008) Construction of a rice glycosyltransferase phylogenomic database and identification of rice-diverged glycosyltransferases. Molecular Plant, 1, 858–877. Cavalier, D.M., Keegstra, K. (2006) Two xyloglucan xylosyltransferases catalyze the addition of multiple xylosyl residues to cellohexaose. Journal of Biological Chemistry, 281, 34197–34207. Cavalier, D.M., Lerouxel, O., Neumetzler, L., et al. (2008) Disrupting two Arabidopsis thaliana xylosyltransferase genes results in plants deficient in xyloglucan, a major primary cell wall component. Plant Cell, 20, 1519–1537. Cocuron, J.C., Lerouxel, O., Drakakaki, G., et al. (2007) A gene from the cellulose synthase-like C family encodes a β-1, 4 glucan synthase. Proceedings of the National Academy of Sciences of the U S A, 104, 8550–8555. Coutinho, P., Deleury, E., Darvill, A., Henrissat, B. (2003) An evolving hierarchical family classification for glycosyltransferases. Journal of Molecular Biology, 328, 307–317. Dhugga, K., Barreiro, R., Whitten, B., et al. (2004) Guar seed β-mannan synthase is a member of the cellulose synthase super gene family. Science, 303, 363–366. Edwards, M., Bulpin, P.V., Dea, I.C.M., Reid, J.S.G. (1989) Biosynthesis of legumeseed galactomannans in vitro. Planta, 178, 41–51. Edwards, M., Scott, C., Gidley, M.J., Reid, J.S.G. (1992) Control of mannose/galactose ratio during galactomannan formation in developing legume seeds. Planta, 187, 67–74. Edwards, M.E., Choo, T.S., Dickson, C.A., Scott, C., Gidley, M.J., Reid, J.S. (2004) The seeds of Lotus japonicus lines transformed with sense, antisense, and sense/ antisense galactomannan galactosyltransferase constructs have structurally altered galactomannans in their endosperm cell walls. Plant Physiology, 134, 1153–1162. Edwards, M.E., Dickson, C.A., Chengappa, S., Sidebottom, C., Gidley, M.J., Reid, J.S.G. (1999) Molecular characterisation of a membrane-bound galactosyltransferase of plant cell wall matrix polysaccharide biosynthesis. Plant Journal, 19, 691–697. Edwards, M.E., Marshall, E., Gidley, M.J., Reid, J.S. (2002) Transfer specificity of detergent-solubilized fenugreek galactomannan galactosyltransferase. Plant Physiology, 129, 1391–1397.

248

■

Plant Polysaccharides

Faik, A., Bar-Peled, M., Derocher, A., et al. (2000) Biochemical characterization and molecular cloning of an α-1, 2-fucosyltransferase that catalyzes the last step of cell wall xyloglucan biosynthesis in pea. Journal of Biological Chemistry, 275, 15082–15089. Faik, A., Price, N., Raikhel, N.V., Keegstra, K. (2002) An Arabidopsis gene encoding an α-xylosyltransferase involved in xyloglucan biosynthesis. Proceedings of the National Academy of Sciences of the U S A, 99, 7797–7802. Farkas, V., Maclachlan, G. (1988) Fucosylation of exogenous xyloglucans by pea microsomal membranes. Archives of Biochemistry and Biophysics, 264, 48–53. Felsenstein, J. (1985) Confidence limits on phylogenies: An approach using the bootstrap. Evolution, 39. Fry, S.C., York, W.S., Albersheim, P., et al. (1993) An unambiguous nomenclature for xyloglucan-derived oligosaccharides. Physiologia Plantarum, 89, 1–3. Hayashi, T. (1989) Xyloglucans in the primary cell wall. Annual Reviews in Plant Physiology and Plant Molecular Biology, 40, 139–168. Hoffman, M., Jia, Z., Pena, M., et al. (2005) Structural analysis of xyloglucans in the primary cell walls of plants in the subclass Asteridae. Carbohydrate Research, 340, 1826–1840. Keegstra, K., Raikhel, N.V. (2001) Plant glycosyltransferases. Current Opinion in Plant Biology, 4, 219–224. Lerouxel, O., Cavalier, D.M., Liepman, A.H., Keegstra, K. (2006) Biosynthesis of plant cell wall polysaccharides – a complex process. Current Opinion in Plant Biology, 9, 621–630. Liepman, A., Wilkerson, C., Keegstra, K. (2005) Expression of cellulose synthase-like (Csl) genes in insect cells reveals that CslA family members encode mannan synthases. Proceedings of the National Academy of Sciences of the U S A, 102, 2221–2226. Liepman, A.H., Nairn, C., Willats, W., Sorensen, I., Roberts, A., Keegstra, K. (2007) Functional genomic analysis supports conservation of function among cellulose synthase-like A gene family members and suggests diverse roles of mannans in plants. Plant Physiology, 143, 1881–1893. Meier, H., Reid, J.S.G. (1982) Reserve polysaccharides other than starch in higher plants. In: Encyclopedia of Plant Physiology (eds F.A. Loewus, W. Tanner). SpringerVerlag, Berlin. Misawa, H., Tsumuraya, Y., Kaneko, Y., Hashimoto, Y. (1996) α-L-fucosyltransferases from radish primary roots. Plant Physiology, 110, 665–673. Mohnen, D. (2008) Pectin structure and biosynthesis. Current Opinion in Plant Biology, 11, 266–277. Perrin, R., Derocher, A., Bar-Peled, M., et al. (1999) Xyloglucan fucosyltransferase, an enzyme involved in plant cell wall biosynthesis. Science, 284, 1976–1979. Perrin, R., Jia, Z., Wagner, T.A., et al. (2003) Analysis of xyloglucan fucosylation in Arabidopsis. Plant Physiology, 132, 768–778. Perrin, R., Wilkerson, C.G., Keegstra, K. (2001) Golgi enzymes that synthesize plant cell wall polysaccharides: Finding and evaluating candidates in the genomic era. Plant Molecular Biology, 47, 115–130. Pré, M., Caillet, V., Sobilo, J., Mccarthy, J. (2008) Characterization and expression analysis of genes directing galactomannan synthesis in coffee. Annals of Botany (London), 102, 207–220. Ray, P.M. (1980) Cooperative action of β-glucan synthetase and UDP-xylose xylosyl transferase of Golgi membranes in the synthesis of xyloglucan-like polysaccharide. Biochimica et Biophysica Acta, 629, 431–444.

Glycosyltransferases of the GT34 and GT37 Families

■

249

Reid, J.S.G., Edwards, M.E., Dickson, C.A., Scott, C., Gidley, M.J. (2003) Tobacco transgenic lines that express fenugreek galactomannan galactosyltransferase constitutively have structurally altered galactomannans in their seed endosperm cell walls. Plant Physiology, 131, 1487–1495. Saitou, N., Nei, M. (1987) The neighbor-joining method: A new method for reconstructing phylogenetic trees. Molecular Biology and Evolution, 4, 406–425. Sarria, R., Wagner, T., O’Neill, M., et al. (2001) Characterization of a family of Arabidopsis genes related to xyloglucan Fucosyltransferase1. Plant Physiology, 127, 1595–1606. Strasser, R., Altman, F., Mach, L., Steinkellner, H. (2004) Generation of Arabidopsis thaliana plants with complex N-glycans lacking β-1,2-linked xylose and core α-1,3linked fucose. FEBS Letters, 561, 132–136. Tamura, K., Dudley, J., Nei, M., Kumar, S. (2007) MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution, 24, 1596–1599. Tsumuraya, Y., Hashimoto, Y., Yamamoto, S., Shibuya, N. (1984) Structure of L-arabino-D-galactan-containing glycoproteins from radish leaves. Carbohydrate Research, 134, 215–228. Van Hengel, A., Roberts, K. (2002) Fucosylated arabinogalactan-proteins are required for full root cell elongation in Arabidopsis. Plant Journal, 32, 105–113. Van Hengel, A.J., Roberts, K. (2003) AtAGP30, an arabinogalactan-protein in the cell walls of the primary root, plays a role in root regeneration and seed germination. Plant Journal, 36, 256–270. Vanzin, G., Madson, M., Carpita, N., Raikhel, N.V., Keegstra, K., Reiter, W. (2002) The mur2 mutant of Arabidopsis thaliana lacks fucosylated xyloglucan because of a lesion in fucosyltransferase AtFUT1. Proceedings of the National Academy of Sciences, 99, 3340–3345. Vincken, J.P., York, W.S., Beldman, G., Voragen, A.G. (1997) Two general branching patterns of xyloglucan, XXXG and XXGG. Plant Physiology, 114, 9–13. Wen, F., Celoy, R.M., Nguyen, T., et al. (2008) Inducible expression of Pisum sativum xyloglucan fucosyltransferase in the pea root cap meristem, and effects of antisense mRNA expression on root cap cell wall structural integrity. Plant Cell Reports, 27, 1125–1135. Wilson, I.B.H., Rendic, D., Freilinger, A., et al. (2001) Cloning and expression of cDNAs encoding α-1,3-fucosyltransferase homologues from Arabidopsis thaliana. Biochimica et Biophysica Acta, 1527, 88–96. Wulff, C., Norambuena, L., Orellana, A. (2000) GDP-fucose uptake into the Golgi apparatus during xyloglucan biosynthesis requires the activity of a transporterlike protein other than the UDP-glucose transporter. Plant Physiology, 122, 867–877. Yokoyama, R., Nishitani, K. (2004) Genomic basis for cell-wall diversity in plants. A comparative approach to gene families in rice and Arabidopsis. Plant and Cell Physiology, 45, 1111–1121. Zabotina, O.A., Van De Ven, W.T., Freshour, G., et al. (2008) The Arabidopsis XXT5 gene encodes a putative α-(1,6)-xylosyltransferase that is involved in xyloglucan biosynthesis. Plant Journal, 56, 101–115. Zuckerkandl, E., Pauling, L. (1965) Evolutionary divergence and convergence in proteins. In: Evolving Genes and Proteins (eds V. Bryson, H.J. Vogel). Academic Press, New York.

Annual Plant Reviews (2011) 41, 251–264 doi: 10.1002/9781444391015.ch9

Chapter 9

GLYCOSYLTRANSFERASES OF THE GT43 FAMILY Nadine Anders and Paul Dupree Department of Biochemistry, University of Cambridge, Building O, Downing Site, Cambridge CB2 1QW, UK Manuscript received March 2009

Abstract: Members of the CAZy family GT43 are present in the plant and animal kingdoms. Two members of the Arabidopsis GT43 family, IRX9 and IRX14, are suggested to act as β-1,4-xylosyltransferases, transferring UDP-Xyl to xylosyloligomers, a process required for glucoronoxylan (GX) synthesis. Absence of IRX9 or IRX14 leads to a decrease in xylan chain length and reduced xylose content of cell walls. The mutant plants exhibit collapsed xylem vessels and, in the case of irx9, dwarfed plant growth. However, the enzymatic activity of the purified proteins remains to be demonstrated, as does the function of most plant GT43 family members. Interestingly, all animal GT43 proteins studied so far catalyse a different enzymatic reaction. They act as β-1,3-glucuronosyltransferases transferring GlcA to a terminal Gal residue of diverse carbohydrate epitopes and are involved in the synthesis of proteoglycans, glycoproteins and/or glycolipids. Thus, GT43 proteins likely diversified in evolution in order to facilitate related, but not identical, biochemical reactions involved in different biological processes, adjusting the properties of the plant cell wall and the animal extracellular matrix to the specific requirements in the two kingdoms. Keywords: glucuronosyltransferase; irregular xylem phenotype; xylan synthesis; xylosyltransferase

9.1 Introduction The GT43 family of glycosyltransferases are inverting enzymes harbouring a GT-A fold. Members of this CAZy family are present in the plant and animal kingdoms, whereas no related proteins can be found in primitive Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

251

252

■

Plant Polysaccharides

Clade b Po_EEE9159 Po_EEE93481 At5g67230 At4g36890_IRX14

CE42859 CE42087 CE42926 CE42859 CE42872 DmGlcAT-BSII DmGlcAT-BSI

Os06g47340.1

HsGlcAT-I HsGlcAT-P HsGlcAT-S CE01694_SQV8

SgGT43B1-1

Phypa_31864 Phypa_129721

DmGlcAT-I

Phypa_32848 Os04g55670.1 Os01g48440.1 Os05g48600.1 At1g27600 Po_EEE90375 SgGT43A1-1 Phypa_173857 Phypa_57016 Po_EEE93073

Os01g06450.1 Os05g03174.1 At2g37090_IRX9 PoGT43B PoGT43A

Os04g01280.1 Os10g13810.1

Clade c Plants

Os07g49370.1 Os03g17850.1

Clade a Po_EEF03372

Animals

Figure 9.1 Phylogenetic analysis of plant and animal GT43 proteins. Sequences were retrieved from diverse databases. Oryza sativa (green), Gramene; Physcomitrella patens (blue), Cosmoss; Caenerorhabtitis elegans (dark grey), wormbase; Drosophila melanogaster (light grey), Arabidopsis thaliana (orange), Selaginella moellendorffii (lilac), JGI; Populus trichocarpa (red) and Homo sapiens (black, bold), NCBI. Sequence alignments of the catalytic regions (ClustalW) were used to generate a neighbour-joining tree. Bootstrap analysis was performed to resample 500 trees with a support threshold of 50%. Note that plant GT43s, classified into three different clades (a–c), are much more diverse than animal GT43s.

eukaryotes like fungi, protozoa or algae (Fondeur-Gelinotte et al. 2006). Phylogenetic analysis divides plant, invertebrate and vertebrate GT43s into multiple clades (Fig. 9.1). Interestingly, multiplication and diversification into clades has occurred independently in plants, invertebrates and vertebrates (Fondeur-Gelinotte et al. 2006). However, separation into three clades in plants appears to be a more recent event. In most genomes investigated to date the number of GT43 family members is small with about three or four representatives. In monocots, however, the family is increased in number, e.g. up to ten GT43s in rice and sorghum (Fig. 9.1, data not shown). Moreover, GT43 genes are more highly expressed in cereals than in dicots, suggesting an advantage in boosting GT43 activity or in diversifying its function.

Glycosyltransferases of the GT43 Family

■

253

A O

HO HO

+

HO

OH

OH

HO

O

O HO

O

OH

UDP-Xyl

B HO HO

COOH

Xyl O OH

+

OH

O

Complex oligosaccharide

O

UDP-GlcA

OH

HO O OH

HO O

OH

O

OH

OH

O

Gal2 COOH

OH

OH

HO O

O

n> 1

UDP

HO HO

Complex oligosaccharide

Xyl O

O HO

O

HO

OH

Xyl

HO O

HO

O

O

n> 1

UDP

HO HO

OH

HO O

O OH

HO O

Gal1

O OH

HO O

OH

O OH

OH

O

Ser

Core protein

Ser

Core protein

O

Xyl

HO O

OH

O

O

Figure 9.2 Enzyme reactions of GT43 family members. Note that the carboxyl group (red) at C6 of UDP-GlcA is the only difference between donor substrates in (A) and (B). (A) Putative enzyme reaction of β-1,4-xylosyltransferases in Arabidopsis. Donor substrate, UDP-Xyl; acceptor substrate, Xyl-β-1,4-Xyl-β-1,4-Xyl; product: Xyl-β-1,4-Xyl-β-1,4-Xyl-β1,4-Xyl. (B) Enzyme reaction of the human UDP-GlcA β-1,3-glucuronyltransferase GlcAT-I. Donor substrate, UDP-GlcA; acceptor substrate, Gal-β-1,3-Gal-β-1,4-Xyl; product, GlcA-β-1,3-Gal-β-1,3-Gal-β-1,4-Xyl.

GT43 proteins seem to catalyse a different reaction in plants and animals. Even though the biochemical activity of purified plant GT43 proteins remains to be demonstrated, at least two GT43 members of Arabidopsis thaliana, IRX9 and IRX14, are suggested to function as β-1,4-xylosyltransferases (EC 2.4.2.-, Fig. 9.2A). IRX9 and IRX14 are required for the elongation of the xylan backbone of glucoronoxylan (GX), a highly abundant hemicellulosic component of the plant secondary cell wall. Reduced length of glucuronoxylan in the mutants leads to collapse of xylem cells and severe growth defects, underlining the important role of xylan in plant cell wall integrity. In contrast to plant GT43s, extensive biochemical and structural analysis on animal GT43s revealed their function as β-1,3-glucuronyltransferases (EC 2.4.1.135, Fig. 9.2B). Animal GT43 proteins are involved in the synthesis of carbohydrate linkage regions of proteoglycans, glycoproteins and/or glycolipids that are important components of the extracellular matrix or

254

■

Plant Polysaccharides 308

Consensus Identity

clade a:

clade b:

Plants

clade c:

Animals

At1g27600 Po_EEE90375 Po_EEE93073 Po_EEF03372 Os01g48440.1 Os04g01280.1 Os05g48600.1 Os10g13810.1 Phypa_173857 Phypa_57016 SgGT43A1-1 At4g36890_IRX14 At5g67230 Po_EEE91159 Po_EEE93481 Os04g55670.1 Os06g47340.1 Phypa_129721 Phypa_31864 Phypa_32848 SgGT43B1-1 At2g37090_IRX9 PoGT43A PoGT43B Os01g06450.1 Os03g17850.1 Os05g03174.1 Os07g49370.1 CE42087 CE42859 CE42872 CE42898 CE42926 CE01694_SQV8 DmGlcAT-BSII DmGlcAT-I DmGlcAT-BSI HsGlcAT-I HsGlcAT-P HsGlcAT-S

Figure 9.3 Amino acid alignment of a selected region of plant and animal GT43s (ClustalW). The region encompasses His308 of GlcAT-I (arrow) that is required for the recognition of the carboxy group at C6 of the UDP-GlcA donor substrate. Interestingly, this residue is not highly conserved in plant GT43s and completely absent in the representatives of clade b. Note the differential conservation of amino acid residues in animals in comparison to conserved residues in the different clades of plant GT43s.

the outer leaflet of plasma membranes, respectively. Since the β-1,3glucuronyltransferases operate at branching points of diverse glycan biosynthesis, interference with their function affects diverse biological processes related to cell–cell and cell–matrix interaction. The sugar UDP-GlcA used for the transfer reaction is conserved as donor substrate in all studied animal GT43 proteins, whereas the proteins differ in their specificity towards the carbohydrate acceptor. Structural and functional analysis of human GT43s revealed the molecular determinants for donor and acceptor specificity. However, the residues required for specificity are not highly conserved in plants, supporting the idea of a divergent biological and biochemical function of GT43 proteins in plants and animals (Fig. 9.3, data not shown).

Glycosyltransferases of the GT43 Family

■

255

9.2 GT43 glycosyltransferases in plants – putative β-1,4-xylosyltransferases Our knowledge on the enzymatic function of GT43 proteins in plants is limited, as the activity of purified GT43s has not been shown so far, and is moreover limited to analysis in dicots. This is surprising, as the number of GT43s is massively enlarged in the genomes of grasses and their expression is relatively higher, suggesting their involvement in grass-specific biosynthesis of cell–wall components (Mitchell et al. 2007). Therefore, understanding GT43 function might be of interest in regard to human nutrition and generation of sustainable bioenergy. The best-characterized plant GT43 proteins are IRX9 and IRX14 from Arabidopsis thaliana. Arabidopsis encodes four GT43 family members: IRX9 (At2g37090), IRX14 (At4g36890), At1g27600 and At5g67230. Phylogenetic analysis revealed that plant GT43s can be divided into three phylogenetic clades (clades a–c) with an amino acid identity between the Arabidopsis representatives of each clade less than 30% (Fig. 9.1). IRX14 and At5g67230 belong to the same clade (b) with an overall amino acid identity of 62.2%. Separation of IRX9 and At1g27600 into clades a and c appears to be a more recent event. Representatives of all three clades are present in both monocot and dicot higher plants, whereas the nonvascular moss Physcomitrella patens and the vascular spikemoss Selaginella moellendorffii lack a possible IRX9 ortholog in clade c. IRX9 was identified as a gene involved in cell wall synthesis due to its high expression in cells of secondary cell wall production and co-expression with cellulose synthase genes (Brown et al. 2005; Persson et al. 2005). The irx9 mutant plants are dwarfed and of poor fertility, displaying reduced secondary wall thickness and more characteristically collapsed xylem cells (Brown et al. 2005; Peña et al. 2007). Extensive analysis of cell wall sugar composition revealed a decrease of xylose and a concomitant approximately 70% reduction of glucuronoxylan (GX) content in irx9 mutants (Brown et al. 2005; Bauer et al. 2006; Brown et al. 2007; Peña et al. 2007). Even though the irx14 mutant plants show no obvious growth phenotype, the mutant displays a irregular xylem phenotype, too, and the content of xylose and GX is reduced in irx14, although to a lesser extent than in irx9 (Brown et al. 2007). Fractionation of cell-wall polysaccharides, quantitative PACE analysis and mass spectrometry revealed a shift of GX GlcA-side chain substitution towards its methylated form in both mutants while retaining the wildtype proportion of xylose residues substituted by either MeGlcA or GlcA-side chains (Brown et al. 2007). The characteristic complex oligosaccharide at the reducing end of GX Xyl-β-1,4-Xyl-α-1,3-Rha-α-1,2-GalA-1,4-Xyl is unaltered in the mutants. However, the molecular weight of GX decreases in irx9, strongly suggesting a defect in xylan chain elongation (Brown et al. 2007; Peña et al. 2007). In fact, analysis of the enzymatic activity of microsomal fractions from irx9 and irx14 mutant plants showed a decrease in xylosyltransferase activity in transferring Xyl from UDP-Xyl on to xylooligomers

256

■

Plant Polysaccharides

(Brown et al. 2007; Lee et al. 2007; Peña et al. 2007). Thus, these data strongly imply that IRX9 and IRX14 function as UDP-Xyl:xylan β-1,4xylosyltransferases. Alternatively, it is possible that IRX9 and IRX14 may be indirectly required for β-1,4-xylosyltransferase activity, not exhibiting this enzymatic activity on their own. The genome of poplar (Populus trichocarpa) encodes seven GT43 genes (Fig. 9.1). Two of them, PoGT43A (EEE72007) and PoGT43B (EEF05217), belong to clade c and it was shown for PoGT43B and their close homologues in hybrid aspen PttGT43A and PttGT43B that they are specifically expressed in secondary walls (Aspeborg et al. 2005; Zhou et al. 2007). Like IRX9-GFP, which localizes to the Golgi apparatus, the subcellular compartment of hemicellulose synthesis, PoGT43B-YFP localizes to the Golgi apparatus in Arabidopsis (Peña et al. 2007; Zhou et al. 2007). Moreover, PoGT43B is able to functionally replace IRX9 function in irx9 mutants and to restore xylan biosynthesis activity of isolated irx9 microsomal fractions, suggesting the conservation of the function of clade c GT43 proteins in flowering plants (Lee et al. 2007; Zhou et al. 2007).

9.3 GT43 glycosyltransferases in animals – β-1,3-glucuronosyltransferases All animal GT43 family members investigated so far exhibit β-1,3glucuronosyltransferase activity. They transfer glucuronic acid (GlcA) to non-reducing ends of specific carbohydrate epitopes of proteoglycans, glycoproteins and/or glycolipids, forming a β-1,3-linkage. The mammalian genomes of human (Homo sapiens), rat (Rattus norvegicus) and mouse (Mus musculus) each encode three GT43 genes. The in vitro function of the human representatives GlcAT-I, GlcAT-P and GlcAT-S has been extensively studied and the function appears to be conserved in the respective homologues of other vertebrate species. GlcAT-I is involved in the biosynthesis of the glycosaminoglycan (GAG)– protein linkage region of proteoglycans. The enzyme transfers a glucuronic acid moiety from UDP-GlcA to the terminal Gal residue (Gal2) of the Galβ-1,3-Gal-β-1,4-Xyl carbohydrate that is covalently bound to serine residues in the core proteins of proteoglycans with heparin/heparan sulfate and chondroitin/dermatan sulfate modifications (Kitagawa et al. 1998; Tone et al. 1999). GlcAT-I plays a key role in the process of proteoglycan assembly since the completion of the linkage region is essential for the conversion of a core protein into a functional proteoglycan. Variation of GlcAT-I function by antisense RNA analysis or its overexpression in cells or tissues producing high amounts of proteoglycans, like chondrocytes or cartilage explants, affect the proteoglycan content accordingly (Venkatesan et al. 2004). This might be of interest in the context of therapeutic treatment of human disorders accompanied by loss of PG, such as osteoarthritis, one of the most prevalent chronic human disorders (Venkatesan et al. 2004).

Glycosyltransferases of the GT43 Family

■

257

Mammalian GlcAT-P and GlcAT-S (in rat GlcAT-D) also transfer GlcA to a terminal Gal-moiety, however their acceptor substrates differ from GlcATI. Both GlcAT-S and GlcAT-P transfer GlcA to Gal-β-1,4-GlcNAc; GlcAT-S furthermore transfers GlcA to Gal-β-1,3-GlcNAc (Shimoda et al. 1999; Tone et al. 1999; Oka et al. 2000; Kakuda et al. 2004; Kakuda et al. 2005 20 20 19 19). Both GlcAT-S and GlcAT-P are required for the synthesis of the sulphated HNK-1 carbohydrate epitope (HSO3–3-GlcA-β-1,3-Gal-β-1,4-GlcNAc), which is present on glycoproteins and glycolipids. HNK-1 is characteristic for diverse cell adhesion molecules most abundant in the nervous system (Kakuda et al. 2004) and functionally associated with neurite outgrowth (Martini et al. 1992), cell adhesion (Künemund et al. 1988) and synaptic plasticity (Yamamoto et al. 2002). The mammalian orthologues of GlcAT-P and GlcAT-S are differentially expressed in the mouse and rat brain (Kizuka et al. 2006; Inoue et al. 2007). Interestingly, their overexpression in rat neural crest cells affects cell migration (Nagase et al. 2003). GlcAT-P knock-out mice lack the HNK-1 epitope in the central nervous system almost completely and show defects in higher functions of the brain, like hippocampal longterm potentiation and spatial memory formation (Yamamoto et al. 2002). Moreover, reduced transcription of human GlcAT-P may be involved in schizophrenia and related psychoses in humans (Jeffries et al. 2003). Diversification of vertebrate and invertebrate GT43s in three phylogenetic clades occurred independently in evolution (Fondeur-Gelinotte et al. 2006). However, the function of GT43 family members of vertebrate and invertebrate species seems to be highly conserved. The genome of Drosophila melanogaster encodes three GT43s: DmGlcAT-I, DmGlcAT-BSI, and DmGlcAT-BSII. Like human GlcAT-I, the DmGlcAT-I is strictly specific to the acceptor substrate Gal-β-1,3-Gal-β-1,4-Xyl, whereas DmGlcAT-BSI and DmGlcAT-BSII (where BS stands for ‘broad specificity’) act on a wide variety of substrates with non-reducing terminal β-1,3- and β-1,4-linked Gal residues (Kim et al. 2003). Strikingly, no GT43 mutant phenotype has been reported in Drosophila, although the regulatory function of proteoglycans in invertebrate development is well documented. However, a mutant phenotype for one of the seven GT43 members in C. elegans has been described. Like its mammalian counterparts, SQV-8 (squashed vulva-8) acts as β-1,3-glucuronosyltransferase, preferentially using Gal-β-1,3-Gal-β as carbohydrate acceptor (Bulik et al. 2000). The sqv-8 mutant shows a dramatic reduction in chondroitin-modified proteoglycans and exhibits defects in epithelium morphogenesis, namely the invagination process of the vulva (Bulik et al. 2000).

9.4

Structural characteristics of GT43 proteins

The known activities of GT43 family of glycosyltransferases indicate they are inverting enzymes, as they utilize UDP-α-D-GlcA to form a β-1,3-linkage through an in-line displacement mechanism (Pedersen et al. 2002; Lairson et al. 2008). They possess a GT-A fold with their typical overall structure

258

■

Plant Polysaccharides

containing a short cytosolic tail, a transmembrane helix conferring type II membrane protein topology, a stem region and a globular catalytic domain. In general, the catalytic domain can be divided into two subdomains with the active side in a cleft that extends across the two subdomains. The N-terminal domain contains the majority of residues associated with donor substrate binding, whereas the C-terminal subdomain contains the acceptor binding sites. Like most GT-A fold proteins, GT43 proteins posses a characteristic DXD motif (in human GlcAT-I: Asp194, Asp195 and Asp196). This motif is involved in hydrogen bonding to the UDP portion of the donor substrate as well as in interacting directly with an essential metal ion (Mn2+) required as cofactor for catalysis (Pedersen et al. 2002). However, the aspartate residues are not invariant and can slightly differ in their relative positions. For example, mutational and functional analysis of the DXD motif of GlcAT-I (DDD, see above) showed the importance of Asp195 rather than Asp194 (Gulberti et al. 2003). The crystal structure of all three human GT43 proteins has been determined and suggests the formation of functional homodimers with C-terminal residues of one monomer extending into the active site of the other (Pedersen et al. 2000; Kakuda et al. 2004; Shiba et al. 2006). Disulfide linkages seem to be involved in dimerization, as mutating a conserved Cys33 residue in GlcAT-I abolished its dimer formation and moreover led to a decrease in catalytic efficiency of the protein (Ouzzine et al. 2000). The crystal structure of the catalytic domain of GlcAT-I in the presence of donor and acceptor substrates as well as the catalytic Mn2+ ion indicated specific conserved amino acid residues involved in substrate binding (Pedersen et al. 2000). The significance of these residues for enzymatic activity and/or binding efficiency has been structurally and biochemically analysed. In addition to the residues of the DXD motif, residues Asp113 and Tyr84 of GlcAT-I interact with the uridine base and were experimentally confirmed to be required for enzymatic activity (Pedersen et al. 2000; Fondeur-Gelinotte et al. 2006). Interaction with the GlcA moiety of the donor substrate (UDPGlcA) is predominantly mediated by His308, Arg116 and Arg156, whose presence is essential for function (Pedersen et al. 2000; Ouzzine et al. 2002; Fondeur-Gelinotte et al. 2006). A network of interactions with the terminal Gal2 moiety of the carbohydrate Gal-β-1,3-Gal-β-1,4-Xyl, involving Glu227, Arg247, Asp252 and Glu281, is essential for acceptor substrate binding and catalytic activity. Especially, Glu281 is positioned to function as a catalytic base by protonating the incoming 3-hydroxyl group at the C-3 position of the acceptor substrate (Pedersen et al. 2000; Pedersen et al. 2002). Interestingly, however, mutational analysis combined with molecular modelling provided evidence that the interaction of Trp243 with the Gal1 residue is a critical determinant in acceptor specificity of GlcAT-I, although other residues are described to interact with the Gal1 moiety as well (Pedersen et al. 2000; Kakuda et al. 2004; Gulberti et al. 2005; Fondeur-Gelinotte et al. 2007). The mechanisms by which the GlcAT-I, GlcAT-P and GlcAT-S recognize the donor substrate are almost the same (Shiba et al. 2006). However, some

Glycosyltransferases of the GT43 Family

■

259

differences exist in the structural models for acceptor substrate recognition, which is expected given the differences in substrate specificity determined by biochemical analysis (Kakuda et al. 2004; Shiba et al. 2006). Replacing the residues relevant for acceptor recognition in GlcAT-S by those of GlcAT-P (W234F, A309V) leads to the loss of β-1,3-linkage acceptor recognition, the characteristic feature of GlcAT-S (Shiba et al. 2006). Strikingly, the residues described to be important for donor and acceptor recognition in animal GT43s are not highly conserved in plants. This finding is consistent with the data suggesting different donor and acceptor substrates for GT43 proteins in plants. Most interestingly, GlcAT-I seems to require a carboxyl group on C6 as a key feature for its sugar donor selectivity (Ouzzine et al. 2002). His308 seems to confer this selectivity, as replacing His308 with arginine alters the donor specificity of GlcAT-I. In contrast to the wildtype GlcAT-I the mutated protein efficiently transfers glucose or mannose from UDP-Glc or UDP-Man, respectively (Ouzzine et al. 2002). The plant GT43 homologues are thought to transfer xylose from UDP-Xyl on to the acceptor substrate. UDP-Xyl is structurally identical to UDP-GlcA except that it lacks the characteristic carboxyl group at C6 (Fig. 9.2). Consistent with this, the His308 of GlcAT-I is not conserved in plant GT43s and completely absent in representatives of clade b (Fig. 9.3). Future research will have to elucidate whether residues characteristic for plant GT43 are involved and important for a plant-specific recognition of donor and acceptor substrates.

9.5

Concluding remarks

The enzymatic function of GT43 proteins in plants and animals seems to have diverged early in evolution. Plant GT43 proteins might function as β1,4-xylosyltransferases, whereas the animal GT43 proteins so far characterized function as β-1,3-glucuronosyltransferases. Interestingly, IRX10 (At1g27440) and its closest homologue IRX10-L (At5g61840), two members of the CAZy family GT47, appear to function as β-1,4-xylosyltransferases, too. The irx10/irx10-L double mutant shows similar macro-, microscopical and molecular phenotypes to irx9 and irx14 (Brown et al. 2009; Wu et al. 2009). In addition, IRX9, IRX10 and IRX14 are highly expressed in stem tissue (AtGeneExpress), the location of secondary cell wall synthesis. It is striking that abolition of IRX9, IRX10/IRX10-L and IRX14 function revealed similar defects in Arabidopsis, despite their very different amino acid sequences. Human GlcAT-I, -P and -S, which are more than 42% identical, vary in their substrate specificity. Hence, diversification of the plant GT43s into phylogenetic clades may be accompanied by acquisition of specific function(s). Analysis of the enzymatic activity of purified plant GT43 proteins is therefore crucial to the understanding of the specific function of GT43 members, which might act together in xylan synthesis but utilize diverse donor or acceptor substrates. Alternatively, they might act

260

■

Plant Polysaccharides

in different protein complexes facilitating the synthesis of different types of xylan. Apart from the GT43 family members and the IRX10-members of the GT47 family, two other proteins, AtCSLD5 and QUASIMODO, have been shown to influence xylansynthase activity of microsomal membranes (Orfila et al. 2005; Bernal et al. 2007). However, analysis of the mutants shows pleiotropic effects, which rather suggest an indirect function on xylan synthesis. Nevertheless, these data may indicate that the chapter of xylan biosynthesis is not closed with the analysis of the GT43 and GT47 proteins. Three articles were published on the function of plant GT43 proteins and their functional redundancy in the family after submission of this review (Keppler and Showalter 2010; Lee et al. 2010; Wu et al. 2010).

References Aspeborg, H., Schrader, J., Coutinho, P.M., et al. (2005) Carbohydrate-active enzymes involved in the secondary cell wall biogenesis in hybrid aspen. Plant Physiology, 137, 983–997. Bauer, S., Vasu, P., Persson, S., Mort, A.J., Somerville, C.R. (2006) Development and application of a suite of polysaccharide-degrading enzymes for analyzing plant cell walls. Proceedings of the National Academy of Sciences of the U S A, 103, 11417– 11422. Bernal, A.J., Jensen, J.K., Harholt, J., et al. (2007) Disruption of ATCSLD5 results in reduced growth, reduced xylan and homogalacturonan synthase activity and altered xylan occurrence in Arabidopsis. Plant Journal, 52(5), 791–802. Brown, D.M., Goubet, F., Wong, V.W., Goodacre, R., Stephens, E., Dupree, P., Turner, S.R. (2007) Comparison of five xylan synthesis mutants reveals new insight into the mechanisms of xylan synthesis. Plant Journal, 52, 1154–1168. Brown, D.M., Zeef, L.A., Ellis, J., Goodacre, R., Turner, S.R. (2005) Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics. Plant Cell, 17, 2281–2295. Brown, D.M., Zhang, Z., Stephens, E., Dupree, P., Turner, S.R. (2009) Characterization of IRX10 and IRX10-like reveals an essential role in glucuronoxylan biosynthesis in Arabidopsis. Plant Journal, 57, 732–746. Bulik, D.A., Wei, G., Toyoda, H., et al. (2000) sqv-3, -7, and -8, a set of genes affecting morphogenesis in Caenorhabditis elegans, encode enzymes required for glycosaminoglycan biosynthesis. Proceedings of the National Academy of Sciences of the U S A, 97, 10838–10843. Fondeur-Gelinotte, M., Lattard, V., Gulberti, S., et al. (2007) Molecular basis for acceptor substrate specificity of the human beta1,3-glucuronosyltransferases GlcAT-I and GlcAT-P involved in glycosaminoglycan and HNK-1 carbohydrate epitope biosynthesis, respectively. Glycobiology, 17, 857–867. Fondeur-Gelinotte, M., Lattard, V., Oriol, R., et al. (2006) Phylogenetic and mutational analyses reveal key residues for UDP-glucuronic acid binding and activity of beta1,3-glucuronosyltransferase I (GlcAT-I). Protein Science, 15, 1667–1678. Gulberti, S., Fournel-Gigleux, S., Mulliert, G., et al. (2003) The functional glycosyltransferase signature sequence of the human beta 1,3-glucuronosyltransferase is a XDD motif. Journal of Biological Chemistry, 278, 32219–32226.

Glycosyltransferases of the GT43 Family

■

261

Gulberti, S., Lattard, V., Fondeur, M., et al. (2005) Phosphorylation and sulfation of oligosaccharide substrates critically influence the activity of human beta1,4galactosyltransferase 7 (GalT-I) and beta1,3-glucuronosyltransferase I (GlcAT-I) involved in the biosynthesis of the glycosaminoglycan-protein linkage region of proteoglycans. Journal of Biological Chemistry, 280, 1417–1425. Inoue, M., Kato, K., Matsuhashi, H., Kizuka, Y., Kawasaki, T., Oka, S. (2007) Distributions of glucuronyltransferases, GlcAT-P and GlcAT-S, and their target substrate, the HNK-1 carbohydrate epitope in the adult mouse brain with or without a targeted deletion of the GlcAT-P gene. Brain Research, 1179, 1–15. Jeffries, A.R., Mungall, A.J., Dawson, E., et al. (2003) beta-1,3-Glucuronyltransferase-1 gene implicated as a candidate for a schizophrenia-like psychosis through molecular analysis of a balanced translocation. Molecular Psychiatry, 8, 654–663. Kakuda, S., Sato, Y., Tonoyama, Y., Oka, S., Kawasaki, T. (2005) Different acceptor specificities of two glucuronyltransferases involved in the biosynthesis of HNK-1 carbohydrate. Glycobiology, 15, 203–210. Kakuda, S., Shiba, T., Ishiguro, M., et al. (2004) Structural basis for acceptor substrate recognition of a human glucuronyltransferase, GlcAT-P, an enzyme critical in the biosynthesis of the carbohydrate epitope HNK-1. Journal of Biological Chemistry, 279, 22693–22703. Keppler, B.D., Showalter, A.M. (2010) IRX14 and IRX14-LIKE, two glycosyl transferases involved in glucuronoxylan biosynthesis and drought tolerance in arabidopsis. Mol Plant, 2010, Jun 30. Kim, B.T., Tsuchida, K., Lincecum, J., Kitagawa, H., Bernfield, M., Sugahara, K. (2003) Identification and characterization of three Drosophila melanogaster glucuronyltransferases responsible for the synthesis of the conserved glycosaminoglycanprotein linkage region of proteoglycans. Two novel homologs exhibit broad specificity toward oligosaccharides from proteoglycans, glycoproteins, and glycosphingolipids. Journal of Biological Chemistry, 278, 9116–9124. Kitagawa, H., Tone, Y., Tamura, J., et al. (1998) Molecular cloning and expression of glucuronyltransferase I involved in the biosynthesis of the glycosaminoglycanprotein linkage region of proteoglycans. Journal of Biological Chemistry, 273, 6615–6618. Kizuka, Y., Matsui, T., Takematsu, H., Kozutsumi, Y., Kawasaki, T., Oka, S. (2006) Physical and functional association of glucuronyltransferases and sulfotransferase involved in HNK-1 biosynthesis. Journal of Biological Chemistry, 281, 13644–13651. Künemund, V., Jungalwala, F.B., Fischer, G., Chou, D.K., Keilhauer, G., Schachner, M. (1988) The L2/HNK-1 carbohydrate of neural cell adhesion molecules is involved in cell interactions. Journal of Cell Biology, 106(1), 213–223. Lairson, L.L., Henrissat, B., Davies, G.J., Withers, S.G. (2008) Glycosyltransferases: structures, functions, and mechanisms. Annual Review of Biochemistry, 77, 521–555. Lee, C., O’Neill, M.A., Tsumuraya, Y., Darvill, A.G., Ye, Z.H. (2007) The irregular xylem9 mutant is deficient in xylan xylosyltransferase activity. Plant Cell Physiology, 48, 1624–1634. Lee, C., Teng, Q., Huang, W., Zhong, R., Ye, Z.H. (2010) The Arabidopsis family GT43 glycosyltransferases form two functionally nonredundant groups essential for the elongation of glucuronoxylan backbone. Plant Physiol, 153(2), 526–541. Martini, R., Xin, Y., Schmitz, B., Schachner, M. (1992) The L2/HNK-1 carbohydrate epitope is involved in the preferential outgrowth of motor neurons on ventral roots and motor nerves. European Journal of Neuroscience, 4(7), 628–639.

262

■

Plant Polysaccharides

Mitchell, R.A., Dupree, P., Shewry, P.R. (2007) A novel bioinformatics approach identifies candidate genes for the synthesis and feruloylation of arabinoxylan. Plant Physiology, 144, 43–53. Nagase, T., Sanai, Y., Nakamura, S., Asato, H., Harii, K., Osumi, N. (2003) Roles of HNK-1 carbohydrate epitope and its synthetic glucuronyltransferase genes on migration of rat neural crest cells. Journal of Anatomy, 203, 77–88. Oka, S., Terayama, K., Imiya, K., et al. (2000) The N-glycan acceptor specificity of a glucuronyltransferase, GlcAT-P, associated with biosynthesis of the HNK-1 epitope. Glycoconjate Journal, 17, 877–885. Orfila, C., Sørensen, S.O., Harholt, J., Geshi, N., Crombie, H., Truong, H.N., Reid, J.S., Knox, J.P., Scheller, H.V. (2005) QUASIMODO1 is expressed in vascular tissue of Arabidopsis thaliana inflorescence stems, and affects homogalacturonan and xylan biosynthesis. Planta, 222(4), 613–622. Ouzzine, M., Gulberti, S., Levoin, N., Netter, P., Magdalou, J., Fournel-Gigleux, S. (2002) The donor substrate specificity of the human beta 1,3-glucuronosyltransferase I toward UDP-glucuronic acid is determined by two crucial histidine and arginine residues. Journal of Biological Chemistry, 277, 25439–25445. Ouzzine, M., Gulberti, S., Netter, P., Magdalou, J., Fournel-Gigleux, S. (2000) Structure/function of the human Ga1beta1,3-glucuronosyltransferase. Dimerization and functional activity are mediated by two crucial cysteine residues. Journal of Biological Chemistry, 275, 28254–28260. Pedersen, L.C., Darden, T.A., Negishi, M. (2002) Crystal structure of beta 1,3-glucuronyltransferase I in complex with active donor substrate UDP-GlcUA. Journal of Biological Chemistry, 277, 21869–21873. Pedersen, L.C., Tsuchida, K., Kitagawa, H., Sugahara, K., Darden, T.A., Negishi, M. (2000) Heparan/chondroitin sulfate biosynthesis. Structure and mechanism of human glucuronyltransferase I. Journal of Biological Chemistry, 275, 34580–34585. Peña, M.J., Zhong, R., Zhou, G.K., et al. (2007) Arabidopsis irregular xylem8 and irregular xylem9: implications for the complexity of glucuronoxylan biosynthesis. Plant Cell, 19, 549–563. Persson, S., Wei, H., Milne, J., Page, G.P., Somerville, C.R. (2005) Identification of genes required for cellulose synthesis by regression analysis of public microarray data sets. Proceedings of the National Academy of Sciences of the U S A, 102, 8633– 8638. Shiba, T., Kakuda, S., Ishiguro, M., et al. (2006) Crystal structure of GlcAT-S, a human glucuronyltransferase, involved in the biosynthesis of the HNK-1 carbohydrate epitope. Proteins, 65, 499–508. Shimoda, Y., Tajima, Y., Nagase, T., Harii, K., Osumi, N., Sanai, Y. (1999) Cloning and expression of a novel galactoside beta1, 3-glucuronyltransferase involved in the biosynthesis of HNK-1 epitope. Journal of Biological Chemistry, 274, 17115–17122. Tone, Y., Kitagawa, H., Imiya, K., Oka, S., Kawasaki, T., Sugahara, K. (1999) Characterization of recombinant human glucuronyltransferase I involved in the biosynthesis of the glycosaminoglycan-protein linkage region of proteoglycans. FEBS Letters, 459, 415–420. Venkatesan, N., Barre, L., Benani, A., et al. (2004) Stimulation of proteoglycan synthesis by glucuronosyltransferase-I gene delivery: a strategy to promote cartilage repair. Proceedings of the National Academy of Sciences of the U S A, 101, 18087–18092.

Glycosyltransferases of the GT43 Family

■

263

Wu, A.M., Rihouey, C., Seveno, M., et al. (2009) The Arabidopsis IRX10 and IRX10LIKE glycosyltransferases are critical for glucuronoxylan biosynthesis during secondary cell wall formation. Plant Journal, 57, 718–731. Wu, A.M., Hörnblad, E., Voxeur, A., Gerber, L., Rihouey, C., Lerouge, P., Marchant, A. (2010) Analysis of the Arabidopsis IRX9/IRX9-L and IRX14/IRX14-L pairs of glycosyltransferase genes reveals critical contributions to biosynthesis of the hemicellulose glucuronoxylan. Plant Physiol, 153(2), 542–554. Yamamoto, S., Oka, S., Inoue, M., et al. (2002) Mice deficient in nervous systemspecific carbohydrate epitope HNK-1 exhibit impaired synaptic plasticity and spatial learning. Journal of Biological Chemistry, 277, 27227–27231. Zhou, G.K., Zhong, R., Himmelsbach, D.S., McPhail, B.T., Ye, Z.H. (2007) Molecular characterization of PoGT8D and PoGT43B, two secondary wall-associated glycosyltransferases in poplar. Plant Cell Physiology, 48, 689–699.

Annual Plant Reviews (2011) 41, 265–284 doi: 10.1002/9781444391015.ch10

Chapter 10

GLYCOSYLTRANSFERASES OF THE GT47 FAMILY Naomi Geshi, Jesper Harholt, Yumiko Sakuragi, Jacob Krüger Jensen1 and Henrik Vibe Scheller2 Department of Plant Biology and Biotechnology, Faculty of Life Sciences, University of Copenhagen Thorvaldsensvej 40, 1871 Frederiksberg, Denmark 1 Present address: MSU-DOE Plant Research Laboratory, Michigan State University, Plant Biology Building, East Lansing, MI 48824, USA 2 Present address: Joint BioEnergy Institute, Lawrence Berkeley National Laboratory, 5885 Hollis St, Emeryville, CA 94608, USA Manuscript received December 2009

Abstract: CAZy glycosyltransferase family 47 (GT47) consists of enzymes that are all predicted to be inverting glycosyltransferases. The family includes enzymes with diverse donor and acceptor specificities, making it difficult to predict the function of family members that have not been biochemically characterized. The GT47 family is dominated by plant proteins, many or all of which are likely involved in cell wall biosynthesis and located in the Golgi vesicles. Based on the phylogeny, the plant sequences can be classified into six groups (A–E). Most of them remain to have their biochemical function determined, but a few members are relatively well characterized: MUR3 and XGD1 have been shown biochemically to function as xyloglucan β-1,2-galactosyltransferase and xylogalacturonan β-1,3-xylosyltransferase, respectively. Wall phenotype has been demonstrated in mutants of arad1, gut1/irx10 and fra8/irx7, but biochemical function of the corresponding proteins remains to be elucidated. The mutant phenotypes suggest that ARAD1 functions as a pectin arabinosyltransferase while GUT1/IRX10L and FRA8/IRX7 are involved in xylan biosynthesis. Keywords: ARAD1; FRA8/IRX7; MUR3; pectin; XGD1; xyloglucan

(glucurono)arabinoxylan;

Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

GUT1/IRX10L;

Edited by P. Ulvskov

265

266

■

Plant Polysaccharides

10.1 Introduction The CAZy glycosyltransferase family 47 (GT47) contains members from plant and animal species. No GT47 members have been found in prokaryotes and a single sequence from the fungus Cryptococcus neoformans is the only one that is not from a plant or animal. The genomes of the beststudied animal species, such as Homo, Danio, Drosophila and Caenorhabditis, encode two to four members in GT47. These animal proteins are known as exostosins, which are heparan synthases and close homologues to heparan synthases. Heparan synthases are composed of two domains: an inverting β-1,4-GlcA transferase domain and a retaining α-1,4-GlcNAc transferase domain. The inverting domain belongs to GT47 while the retaining domain belongs to family GT64 and the heparan synthases are therefore members of both GT47 and GT64. The plant homologues in GT47 have only a single domain. Plants also have members of GT64 but no plant protein has been found with a dual domain structure as in the animal heparan synthases. GT47 is dominated by plant sequences; there are more than ten times as many members from plant species than from animal species. This high prevalence of plant sequences has made GT47 an obvious place to look for proteins involved in cell wall biosynthesis (Scheller et al. 2007). Although most of the GT47 members in plants still have an unknown function it seems likely that most if not all the proteins are involved in cell wall biosynthesis. As will be clear from the detailed treatment below, the plant members of GT47 are highly diverse with respect to the substrates they use and the linkages they form. GT47 members are involved in biosynthesis of pectin and hemicellulose, they can form several different linkages, and they use both neutral and acidic substrates.

10.2 Phylogenetic analysis of CAZy GT47 As of January 2008 the CAZy family GT47 contains 158 plant sequences from over 20 plant species. The majority of the sequences originate from Arabidopsis thaliana (39) and Oryza sativa (japonica cultivar) (35) as these organisms are fully sequenced. In addition to the sequences found in CAZy, two rice sequences were added. The two rice sequences are orthologues to At3g57630 and At1g21480, two Arabidopsis outliers not grouped into clades. The phylogenetic relationship between all these sequences is illustrated in Fig. 10.1. The tree was constructed after manual trimming of the protein sequences and exclusions of fragments that did not represent full-length genes, resulting in a total of 109 sequences. The resulting phylogenetic tree is made up of six separate groups, A–F, of which C and D are further divided into two and three subgroups, respectively. The nomenclature for the groups originates from Li et al. (2004).

Glycosyltransferases of the GT47 family

■

267

D

C3 Musa balbisian Mus a acumin a ata Vitis Medicago At5 g19 670 At5 g3 70 00

Rice

F

At 5g 22 94 0 Ri Trit ce icu At2g28 ma 110/FR Populus ven A8/IRX 7 sis

Populus 5 x rice Vitis

o Medicag 45 67 g1 At4 Rice 40 go 80 ica g3 ed At4 M

E

C1

40 At1g274 X10L R 1/I UT /G 40 18 g6 At5

D1

Ri

ce Ri ce

At3 g5 763 0 At1g2 1480 Vitis Rice Rice

0 10 82 25 g116 5g At5 At 0 9 7 2 At4g3

1 AD /AR 00 51 g3 At2

is m cu ice o Cu Rdicag 10 0 e MAt5g2532026 g Rice At5 Medicago At5g1 1 130 VitisVitis A Ol t5g33 im 29 ara 0/X bid GD1 op sis

Rice

C2

180 g42 At3 Medicago

At5 g44 930 /

ARA D2 Medicag o Rice At1 g74 680 Rice Rice At3g03650 s lu pu Po 400 Rice ice R 10 g45 4 At3 67 1g At e Ric0 89 16 5g At

At 1g 34 27 0 Rice

B

C

At5 g03 80 At 0 3g 07 62 0

D2

Medicago

Rice

Ric e

Rice e Ric e Ric

At R 4g ice 22 58 0 e 0 0 Ric 22 75 2 2 g6 g3 40 At5 icago At2 327 g 0 Med At2 3199 R3 g At2 370/MU 20 At2g 40 RicAet2g290

0 125 g4 50 At5 g634 0 0 1 47 99 At 68 13 1g 4g At is At Vit

A Figure 10.1 A phylogenetic tree of GT47 including Arabidopsis, rice and Medicago as the major species. Several other species are included where apparent full length sequences are available. The clade nomenclature is from Li et al. (2004).

At present we can say only very little about what the evolutionary relationships between these proteins may reflect in terms of the function of the proteins in different plant species. We expect to find functional conservation within the different groups where some of the genes may either be redundant or involved in related functions. One such case could be the eight genes in group B, where it has been proposed that all these are αarabinofuranosyltransferases involved in pectin biosynthesis (Scheller et al. 2007, and Section 10.5). Another example is group A where at least two proteins are involved in xyloglucan galactosylation (see Section 10.3).

268

■

Plant Polysaccharides

However, since there are no cases so far where the function of more than one GT in a group or subgroup has been elucidated we cannot predict the degree of conservation within and between groups in biochemical function of the GTs in GT47. With the exceptions of subgroup C3, species with type I walls and species with type II walls are both represented in all the GT47 groups and subgroups. This observation may reflect the fact that while there are differences in the ratio between the various cell wall components in type I and type II walls, the basic cell wall carbohydrate components are still the same. Hence, we might expect differences in expression patterns between grasses such as rice and a typical dicot such as Arabidopsis rather than differences in the inventory of GTs. This logic has been used recently to identify candidate transferases for arabinoxylan synthesis in rice (Mitchell et al. 2007). The single-copy Arabidopsis gene (At4g38040) in subgroup C1 is interesting as both of the other two major species represented in the tree, Medicago and Oryza, along with Vitis, are represented in that subgroup. Furthermore, in silico expression analysis shows that At4g38040 is ubiquitously and highly expressed. With the lack of close homologues, redundancy seems unlikely and we may predict that analysis of e.g. a knock-out mutant would be indicative of the function of the GT. There are two single-copy clades in GT47. These two clades contain At1g21480 (clade E) and At3g57630 (clade F) with rice orthologues. That these clades do have some function, or at least had an ancestral function, is indicated by the fact that there are rice and Vitis fragment orthologues in both clades. Furthermore, both clades have Physcomitrella orthologues, showing that the diversion of these two clades happened before the division of mosses and higher plants (see later for discussion on Physcomitrella). Clade E actually points towards some ancestral gene which was present before the division of the primordial plant/alga gene into alga- and plantspecific genes. If the protein sequence of At3g57630 is blasted using BlastP, then the highest scoring sequences are the rice and Vitis orthologues, which is not a surprise. But the next genes in line are actually Chlamydomonas sequences. The algae sequence found by this simple blast, indicates that an ancestral alga had a glycosidic linkage catalysed by this protein. Its could be hypothesized that this glycosidic linkage is found in the cell wall. The cell wall of Chamydomonas consists of glycoprotein resembling extensins from plant cell (Hicks et al. 2001). So there is a definite link between clade E in higher plants and Chlamydomonas GT47 sequences linking Chlorophyta and streptophyta cell walls. With the full genome sequence available for Physcomitrella patens it is now possible to investigate the phylogenetic relationship of more distantly related species, rather than only the relationship between angiosperms as discussed above. The cell wall composition of P. patens has recently been investigated in some detail and there are three polymers that seem not to be present at

Glycosyltransferases of the GT47 family

■

269

all, or present in very low amounts in P. patens as compared to angiosperms: xylogalacturonan, fucosylated xyloglucan and mixed-linkage β-glucan (Møller et al. 2007). A comparison of the genomes could lead to information on the function of the GTs. Figure 10.2 shows a tree for GT47 compromising Arabidopsis and Physcomitrella. Compared to Fig. 10.1, the tree in Fig. 10.2 shows the expected deeper branching when two distantly related species are analysed. This is seen for example in group B, where the Arabidopsis and Physcomitrella proteins group into separate subgroups. However, the branching of group B does not indicate that that there is a fundamental difference between Arabidopsis and Physcomitrella in the inventory of sequences. From the cell wall analysis of Physcomitrella we also known that arabinan is present (Møller et al. 2007) and we know that member(s) of group B are involved in arabinan biosynthesis (Harholt et al. 2006; see Section 10.5 below). Interestingly, there are no close orthologues to At2g35100, ARAD1, which is the only member of subgroup B in which knock-out plants have a phenotype (Harholt et al. 2006; J. Harholt, J.K. Jensen, C. Søgaard, H.V. Scheller, unpublished results). Group A and D seem to have evolved as group B with orthologous genes. Group C is interesting, as this group contain only a few Physcomitrella genes. Subgroup C1 is again confirmed as a singlecopy group in Arabidopsis with orthologues in Physcomitrella, supporting the notion above that the GT has a conserved and important function. Like groups A, B and D, subgroup C3 has several orthologues in both Arabidopsis and Physcomitrella. In contrast, subgroup C2 differs in having only one member in Physcomitrella compared to the seven homologues in Arabidopsis. This could indicate that the biological function within the group is specific to higher plants and that the genes in higher plants have evolved towards differentiation in function or spatial expression. Subgroup C2 contains XGD1 which is a xylogalacturonan xylosyltransferase (Jensen et al. 2008) and since Physcomitrella is devoid of xylogalacturonan there is a consistency in the poor representation of subgroup C2 in the moss. With a cut-off of 1e-25 used for the blast, some Physcomitrella-specific clades are observed.

10.3

Group A

The MUR3 gene (At2g20370) was the first of only two GTs in GT47 to have its biochemical activity demonstrated (Madson et al. 2003).The mur3 mutants (missense mutants; mur3–1 (S470L) and mur3–2 (A290V)) were identified in an Arabidopsis mutant collection that was generated by chemical mutagenesis and screened for altered monosaccharide composition in the cell wall (Reiter et al. 1997). Both mur3–1 and mur3–2 mutants were found to be deficient in fucose. The mutants showed altered trichome papillae, and reduced wall strength in hypocotyls (Ryden et al. 2003; Peña et al. 2004), but otherwise a rather normal growth phenotype. More recently, additional allelic mutants

a Phyp

At 1g 21 48 0

a yp Ph

F

At4g38040

Phypa

E

At5g25310 At3g07620 At5g03800 XGD1 At Ph yp At A3g4218 a 5g t5g 0 11 2 13 026 0 0

C

a yp Ph

At 3g 57 63 0

10 16 5 5g1 90 4 7 t 7 16 A 32 4g At4g t A At5g25820 A At5 t5g370 00 g19 670 0 94 22 5g t A t2g28110 A Phypa Phypa At1g 2744 0

D

40 18 g6 At5

At 3g At3 036 5 g At1 4540 0 g74 0 680

0 741 g6 &2 1 t A AD1 0 AR 27 0 34 89 1g t 16 A g 5 At

Ph

yp

B

a

Phypa

Phypa

2g 29 04 P At1g 6847 hypa 0 0

At

a 50 yp 12 Ph 4 50 g 4 3 5 At t1g6 A pa Phy

At5g62220 ypa Ph

Phypa

0 3275 0 0 74 99 32 g31 g 2 At At2 At2g

Mur3

At 4g 22 58 0

A Figure 10.2 The phylogenetic relation betwen Arabidopsis and Physcomitrella patens in GT47. Note that none of the clades contain solely Arabidopsis or Physcomitrella patens sequences. A deeper branching than in Fig. 10.1 can be observed corresponding to the longer evolutionary distance between Arabidopsis and Physcomitrella patens.

Glycosyltransferases of the GT47 family

■

271

were characterized by another group (Tamura et al. 2005), and in contrast to the mur3 mutants the growth was severely affected in these new mutants. The details are described below, but in short, nonsense and T-DNA knockout mutants are severely dwarfed (Tamura et al. 2005). Cloning of the corresponding gene revealed that the MUR3 gene encodes a protein with type II membrane topology and with a sequence similarity to the β-GlcA transferase domain of animal exostosins (Madson et al. 2003). MUR3 transcripts are present in all major plant organs, and the protein is targeted to the Golgi apparatus. Recombinant protein expressed in Pichia pastoris showed biochemical activity of the protein as a β-1,2-Gal transferase transferring galactose onto the third xylose residue in the xyloglucan XXXG core structure. This galactose is often substituted with fucose and this explains the fucosedeficient phenotype of mur3–1 and mur3–2. The primary deficiency of the mur3–1and mur3–2 mutants in galactose is masked by the many other galactose-containing polysaccharides and glycoproteins. Apparently, MUR3 does not transfer galactose onto the second xylose residue, indicating the presence of at least one other β-1,2-Gal transferase for complete synthesis of the xyloglucan structure present in Arabidopsis. MUR3 homologues are widely present in many flowering plants, including important crop species such as alfalfa, tomato and barley. In Arabidopsis, ten homologous genes are present in the genome (classified AtGT11–20 in Li et al. 2004). Some of these may work as other xyloglucan-specific Gal transferases. AtGT18 is a specially good candidate since T-DNA insertion lines of AtGT18 resulted in reduced galactose substitution of the second xylose in the xyloglucan core structure, i.e. a different site than that shown in mur3 (Li et al. 2004; W.-D. Reiter, personal communication). However, biochemical activity data has not yet confirmed the specific involvement of GT18 in the xyloglucan Gal transferase reaction. Li et al. (2004) suggested another possibility for MUR3 homologues: that some may be inverting larabinosyltransferases, considering the structural similarity of α-l-Ara and β-d-Gal. In fact, in solanaceous species l-Ara largely replaces d-Gal in the xyloglucan structure (Jia et al. 2005), and therefore their MUR3 homologues may encode xyloglucan-specific Ara transferase. As mentioned above, new mutant alleles with a much stronger effect on phenotype were recently characterized (Tamura et al. 2005). Tamura and coworkers observed that the katamari1 (kam1) mutant, an allelic mutant of mur3, forms large aggregates of endomembranes, which is not observed in mur3–1 and mur3–2. The kam1–1 mutant was created by chemical mutagenesis similarly to mur3–1 and mur3–2, but the kam1–1 mutation introduced a stop codon after the N-terminal transmembrane domain (Gln-62 to STOP) which resulted in deficiency of the MUR3 protein, while in mur3–1 and mur3–2 single amino acid residues are altered but the protein is still produced. Tamura et al. (2005) additionally investigated two T-DNA insertional mutants in which the MUR3 protein is not produced, and showed that endomembranes also form aggregates in these mutants. In contrast to the

272

■

Plant Polysaccharides

mur3 mutants, kam1–1 and the T-DNA knock-out mutants have a strong dwarf phenotype. In addition it was found that actin could be coimmunoprecipitated with MUR3 protein using MUR3 antibody. These results suggest that MUR3 has a role for organizing the endomembrane system in addition to the role as xyloglucan Gal transferase. Tamura et al. (2005) suggested MUR3 as a bifunctional protein, in which one domain interacts with actin and another domain acts as a xyloglucan Gal transferase. However, a mechanism explaining the interaction between MUR3 protein and actin filament is not clear since MUR3 is a type II membrane protein and most of the protein is present in the luminal side of Golgi. If there is any interaction between MUR3 and actin, it must be via the short N-terminal sequence before the transmembrane domain of MUR3, where none of the sequence motifs for interaction with actin appear to be present. Thus, the interaction may be indirect, and more detailed study of the interaction mechanism remains to be carried out.

10.4 Group D Several members of group D have been shown to be involved in cell wall biosynthesis. NPGUT1 was identified as a gene in which T-DNA insertion causes loss of tight intracellular attachment in callus of Nicotiana plumbaginifolia (Iwai et al. 2002). The deduced NpGUT1 amino acid sequence had a significant homology to animal β-glucuronosyltransferase (GlcAT) domain. Surprisingly, NpGUT1 does not encode a clear transmembrane domain, unlike most of the Golgi-localized cell wall GTs, and localization of NpGUT1 protein in the secretory pathway has not been demonstrated. NpGUT1 expression is specific for the meristems in N. plumbaginifolia. Analysis of cell walls in the npgut1 mutant revealed that its pectic rhamnogalcturonan II (RG-II) had no detectable level of glucuronic acid and contained only half the amount of galactose compared with the wild type (Iwai et al. 2002). Most of the RG-II in wild type is present as a dimer via a covalent linkage of borate diester, but RG-II in the npgut1 mutant was mostly present as a monomer. Complementation of the mutant with the wild type NpGUT1 recovered RG-II dimer formation, implicating that the disaccharide Gal-GlcA in the terminal RG-II side chain is important for the RGII cross linking, and the cross linking is important for cell-to-cell adhesion. Iwai et al. (2002) postulated that NpGUT1 is an RG-II specific β-GlcA transferase, but there is a discrepancy in this hypothesis. Pectin RG-II is a very complex polysaccharide containing 12 different sugars, but the structure is well conserved in all land plants examined so far (Pérez et al. 2003). Therefore, we would expect that the genes encoding the enzymes involved in RGII biosynthesis are also well conserved in different plant species. However, in case of NpGUT1, none of the closest homologues from Arabidopsis and poplar seem to be involved in RG-II biosynthesis, but instead in glucuronoxylan biosynthesis (see below). From the expression profile, it is clear that

Glycosyltransferases of the GT47 family

■

273

NpGUT1 has a role in primary wall biosynthesis, but alteration of RG-II might be a secondary effect of the mutation. Perhaps the NpGUT1 gene product has a specific function in Nicotiana species that is not conserved in other plants. The closest homologues of NpGUT1 are Arabidopsis AtGUT1/AtIRX10L (At5g61840), followed by AtGUT2/AtIRX10 (At1g27440) and poplar PttGT47A. Arabidopsis FRAGILE FIBER8/IRX7 (FRA8/IRX7) and its poplar orthologue PttGT47C are also in group D, but these are in a different branch in the phylogenetic tree (see Fig. 10.1). The nomenclature used here follows the original designation by Iwai et al. (2002) of At5g61840 as AtGUT1, which is also followed in NCBI. Other workers have used AtGUT1 for At1g27440 and AtGUT2 for At5g61840 (e.g. Zhong et al. 2005; Ubeda-Tomas et al. 2007) and this designation is used in CAZy. However, as described above, the biochemical function suggested for AtGUT1 and AtGUT2 is clearly different from NpGUT1, thus At5g61840 and At1g27440 should be named differently from GUT according to their function. In contrast to NpGUT1, the homologues in Arabidopsis and poplar are predicted at least by some programs to be type II membrane proteins (see Section 10.7). Expression of NpGUT1 is specific for meristems, whereas PttGT47A and AtGUT2 are highly expressed in wood-forming tissues in poplar and in secondary thickened hypocotyls in Arabidopsis, respectively (Ubeda-Tomas et al. 2007). These are the tissues where secondary wall formation is active. AtGUT1/AtIRX10L is also expressed in secondary thickened hypocotyls, but also in other tissues in Arabidopsis. In addition, the AtGUT2/ AtIRX10 gene is coexpressed with CESA4, CESA7 and CESA8 (Brown et al. 2005, Persson et al. 2005), which are involved in cellulose synthesis during secondary wall formation (Taylor et al. 1999; Taylor et al. 2000), and a T-DNA insertion line for atgut2/atirx10 shows collapsed vessel phenotype (irx10 in Brown et al. 2005). Single mutants of atgut1/atirx10L and atgut2/atirx10 have a relatively normal growth phenotype, but a double knock-out is severely dwarfed, indicating a redundant function encoded by the two GTs (Marchant et al. 2007). The phenotype is less severe in the double mutant with heterozygosity at one locus compared with homozygous/homozygous, indicating that the function depends on gene dosage. Furthermore, the double mutant of atgut1/atirx10 heterozygote/atgut2/atirx10 homozygote is more severely affected than the atgut1/atirx10L homozygote/atgut2/atirx10 heterozygote, indicating the essential role of AtGUT2/AtIRX10L function in glucuronoxylan biosynthesis (Wu et al. 2008; A. Marchant, personal communication). When these mutants are investigated with LM10 antibody, which labels xylans with no substitutions or low substitutions (McCartney et al. 2005), the level of LM10 epitope also follows the gene dosage; the epitope is less abundant in the intermediate mutant and lacks in the double knock-out. The dwarfed phenotype in the double knock-out can be complemented both by AtGUT1/AtIRX10L and AtGUT2/AtIRX10 genes driven by the AtGUT2/ AtIRX10L and AtGUT1/AtIRX10 promoters, respectively, whereas NpGUT1 cannot. RG-II dimerization or structure is normal in atgut1/atirx10L and

274

■

Plant Polysaccharides

atgut2/atirx10 single mutants as well as in different combination of double mutants. In contrast, the level of glucuronic acid in xylan is reduced by 90– 95% and 5% in atgut1/atirx10L and atgut2/atirx10, respectively (Séveno et al. 2009; P. Lerouge, personal communication). These lines of evidence suggest distinct functions of NpGUT1 and AtGUT1/AtIRX10L/AtGUT2/AtIRX10. FRAGILE FIBER8 (FRA8) in Arabidopsis and its homologous gene in poplar (PttGT47C) are also members in subgroup D. The fra8 mutant was found in a screen for reduced fibre wall strength in Arabidopsis and the mutated gene was cloned (Zhong et al. 2005). PttGT47C has been confirmed as a FRA8 orthologue since PttGT47C can complement the fra8 mutant (Zhou et al. 2006). FRA8 as well as PttGT47C have a type II membrane protein topology, and FRA8 targets the Golgi apparatus. Differently from npgut1 and similarly to atgut1 and atgut2, FRA8 mutation does not affect the level of glucuronic acid in pectin and the level of RG-II dimerization, indicating a different role of FRA8 than in pectin biosynthesis. Mutation in FRA8 caused a decrease in the number of glucuronoxylan chains and glucuronic acid residues in overall glucuronoxylan. Glucuronoxylan is the most predominant hemicellulose present in secondary walls in dicots and is composed of a linear backbone of 1,4 linked β-d-Xyl residues, of which approximately 5% are substituted with α-d-GlcA or 4-O-methyl α-d-GlcA (Darvill et al. 1980; Zablackis et al. 1995). The reducing end of the glucuronoxylan from Arabidopsis, birch and spruce has been shown to contain the oligosaccharide of (1→4)-β-d-Xylp-(1→4)- β-d-Xylp-(1→3)-α-l-Rhap-(1→2)α-d-GalpA-(1→4)-d-Xylp (Shimizu et al. 1976; Andersson et al. 1983; Peña et al. 2007). This specific oligosaccharide was depleted in the fra8 mutant. In Arabidopsis, IRREGULAR XYLEM8 (IRX8) and IRREGULAR XYLEM9 (IRX9) as well as FRA8 are suggested to be involved in the biosynthesis of glucuronoxylan in secondary cell walls (Brown et al. 2005; Persson et al. 2005; Zhong et al. 2005; Bauer et al. 2006; Peña et al. 2007; Persson et al. 2007). The specific oligosaccharide described above is also decreased in irx8, but is more abundant in irx9 (Peña et al. 2007). Investigation of biosynthetic enzyme activity in the mutants showed that the irx9 mutation results in a substantial reduction in Xyl transferase activity but has no effect on GlcA transferase activity, whereas neither XylT nor GlcAT activity was affected by fra8 and irx8 mutations (Lee et al. 2007). These results are not a direct demonstration of the enzymatic activity of FRA8, but the results indicate that FRA8 protein may be involved in the biosynthesis of the complex oligosaccharide in the reducing end of xylan in Arabidopsis. As described above, reduction of α-d-GlcA and 4-O-methyl α-d-GlcA in xylan is commonly observed in atgut1/atirx10L, atgut2/atirx10 as well as in the fra8 mutant. However this may be an indirect effect based on an assumption that all the GTs in family 47 are inverting enzymes. Xylan-specific α-GlcA transferase would be retaining and therefore neither structurally similar to a β-GlcA transferase, nor likely to be found in GT47. If FRA8 is involved in making the specific hetero-oligosaccharides found in the reduc-

Glycosyltransferases of the GT47 family

■

275

ing end of xylan, we can speculate that it might encode either the β-1,3xylosyltransferase or the α-1,2-rhamnosyltransferase, since these enzymes would be inverting enzymes as expected in GT47.

10.5

Group B

Group B of GT47 consists of eight proteins in Arabidopsis, one of which has been characterized in detail. At2g35100 has been designated ARABINAN DEFICIENT 1 (ARAD1) and has been shown to be involved in pectic arabinan biosynthesis (Harholt et al. 2006). The arad1–1 mutant was identified in a reverse screen of Arabidopsis knock-out mutants in GT47 based on whole cell wall monosaccharide composition. Knock-out mutants of the gene (arad1–1 and arad1–2) contain only 25% and 50% of wild type arabinose in leaf and stem, respectively. Many polymers in the cell wall contain arabinose but immunohistochemistry identified arabinan to be the affected polymer in arad1 mutants whereas other polymers were unaffected. In purified RG-I, the reduction in arabinose content was reduced by 70% as compared to the wild type. Pectic arabinan contains arabinose in several different linkages but linkage analysis of the purified RG-I showed a general decrease in the different kind of linkages of arabinose indicating that the arad1 mutants contain shorter arabinan side chains in the RG-I. Based on this general decrease, ARAD1 was proposed to be an α-1,5-arabinosyltransferase involved in elongation of the pectic arabinan side chains (Harholt et al. 2006). Transformation of arad1 with CaMV35S:ARAD1 led to complementation of the arabinan deficient phenotype, but the highly elevated levels of ARAD1 protein in the plants did not lead to increased arabinan content (Harholt et al. 2006). This could indicate either a very tight regulation of enzyme activity or that another component is limiting, perhaps a subunit of a protein complex that includes the GT. Another possibility is that nucleotide sugar substrates are limiting. Despite the dramatic and specific change in the arabinan side chains of RGI in the arad1 mutants, the plants have no visible phenotype. However, careful analysis of the mutants has demonstrated the importance of arabinans for the mechanical properties of the plants. Stress/strain experiments on arad1 hypocotyls show that these are less stiff and can withstand less force than the wild type before breakage (J. Harholt, P. Ryden, P. Ulvskov & H.V. Scheller, unpublished). A link between RG-I arabinan and physical properties of the wall has previously been suggested based on the effect of expressing fungal arabinanase in potato tubers (Ulvskov et al. 2005) and of treating stomatas with arabinanase (Jones et al. 2005). Analysis of the arad1 mutants has also demonstrated the importance of arabinan for interactions with pathogens. Cell wall polysaccharides are thought to play important roles in protection against invading pathogens. However, only limited examples are known about the exact function(s) of

276

■

Plant Polysaccharides

individual polysaccharides and little, if not nothing, has been known about the cell wall polymers synthesized by the GT47 family proteins in host defence. The arad1 mutants were subjected to an infection assay with the necrotic fungal pathogen Botrytis cinera and were observed to show increased susceptibility to the pathogen infection (Y. Sakuragi, M. Nafisi, J. Harholt, H.V. Scheller, unpublished). This response of increased susceptibility is unique in light of that the previously studied cell wall mutants generally showed increased resistance to various pathogens (Ellis et al. 2002; Vogel et al. 2002; Nishimura et al. 2003; Zhu et al. 2003; Gasper et al. 2004; Vogel et al. 2004; Hernández-Blanco et al. 2007). This somewhat counter-intuitive observation relates to the fact that the plant responds to altered wall structure by mobilizing pathogen defences. However, in the case of arad1 mutants this was apparently not the case. Detailed characterization of the molecular basis of increased susceptibility is currently underway. Expression analysis of the genes in group B show that they are generally all ubiquitously expressed and the expression levels are low to moderate. Investigations of monosaccharide composition of leaf cell walls of T-DNA knock-out mutants representing six out of the remaining seven members have so far not revealed any new mutant phenotypes in pectic arabinan or any other cell wall polymer (J. Harholt & H.V. Scheller, unpublished). One of the group B GTs in Arabidopsis, At3g03650, has also been identified in a mutant screen for embryonic developmental arrest, but no further characterization was reported and ESTs are also available from vegetative tissue indicating that At3g03650 is not only expressed in embryonic tissue (Pagnussat et al. 2005). No further characterization was reported by Pagnussat et al. and complementation or other controls were not included. A T-DNA insertion mutant in At3g03650 analysed in our lab did not show any visual phenotype or change in monosaccharide composition of the cell walls (J. Harholt & H.V. Scheller, unpublished data). The closest homologue to ARAD1 is At5g44930 and a T-DNA mutant of the corresponding gene has been investigated in detail, but no differences from the wild type have so far been observed. Purified pectic fractions of the cell wall of the At5g44930 T-DNA mutant show the same monosaccharide composition as for the wild type. At5g44930 is, however, not redundant to ARAD1. The double knockout mutant of the two genes does not show a more severe phenotype than arad1. Furthermore, in a cross-complementation experiment using the 35S promoter to express At5g44930 there was no complementation of the arad1 mutant (J. Harholt, J.K. Jensen, C. Søgaard, & H.V. Scheller, unpublished). Hence, at present the functions of the remaining seven Arabidopsis members of group B remain unknown. Considerable efforts have been made to establish the catalytic activity of ARAD1. Initial attempts to express the protein in Pichia pastoris were unsuccessful but as for XGD1 significant amounts of recombinant ARAD1 have been obtained by transient expression in N. benthamiana. Konishi et al. (2006) have developed a sensitive assay for detecting arabinan α-(1,5)-

Glycosyltransferases of the GT47 family

■

277

arabinofuranosyltransferase but substantial efforts to establish UDP-Araf arabinosyltransferase activity in solubilized microsome preparations containing recombinant ARAD1 have so far been inconclusive (H.V. Scheller, T. Konishi, N. Geshi, S.O. Sørensen, J.K. Jensen, & T. Ishii, unpublished). This would suggest either that ARAD1 needs additional protein partners to be active or that ARAD1 cannot use linear arabino-oligosaccharides as acceptor. One possibility is that ARAD1 is responsible for attaching the first, or one of the first, arabinose residues to the RG-I backbone whereas another enzyme is responsible for the arabinan chain elongation. Since biochemical activity has not been established, it can not excluded that ARAD1 is not an Ara transferase and that the phenotype of arabinan deficiency is a secondary effect. However, the lack of strong pleiotropic effects in the arad1 mutants supports the specific involvement of ARAD1 in arabinan biosynthesis.

10.6

Group C

XGD1 (At5g33290) belongs to group C and encodes a pectic xylogalacturonan xylosyltransferase (Jensen et al. 2008). The XGD1 protein is a type II membrane protein and is targeted to the Golgi apparatus. XGD1 is most highly expressed in leaves, where pectic xylogalacturonan (XGA) is abundant (Zandleven et al. 2007). T-DNA insertional mutation in XGD1 resulted in a substantial reduction of xylose in the isolated pectin fraction compared with the wild type. The xgd1 mutants lack detectable level of XGA cleavable by an XGA-specific hydrolase, while the same treatment generated substantial amount of fragments from wild type plants. XGD1 expressed in N. benthamiana had the activity of transferring xylose onto oligogalacturonides. When endogenous acceptors present in wild type plants were used in the enzyme assays with UDP-Xyl, the synthesized products were cleavable by the specific XGA hydrolase. The LM8 antibody was raised against a pea XGA which was highly substituted with xylose residues (Willats et al. 2004). This antibody recognizes only a few tissues in Arabidopsis, namely root tips and the septum of siliques, whereas the abundant XGA in leaves was not recognized (Jensen et al. 2008). Surprisingly, the LM8 epitope in siliques and root tips was unaffected in the walls of xgd1 mutant plants. This indicates the presence of at least two different types of XGA with different patterns of xylose substitution in Arabidopsis and requiring different xylosyltransferases for their synthesis. In addition to XGD1, group C contains six other Arabidopsis GTs in the same branch as XGD1. These are obvious candidates for synthesizing different types of XGA, but so far no information on mutant phenotypes or biochemical activity is available for any of these six genes and proteins.

278

■

Plant Polysaccharides

10.7 Subcellular localization and protein–protein interactions Subcellular localization of cell wall biosynthetic proteins is a very important aspect of cell wall biosynthesis. Generally, wall biosynthesis is believed to take place in the Golgi vesicles except for cellulose and callose, which are made on the plasma membrane. However, the subcellular localization has not been rigorously characterized and only sporadic information is available with individual cell-wall biosynthetic enzymes. Thus, we cannot exclude that some biosynthetic reactions – e.g. formation of primers – take place in the endoplasmic reticulum. Furthermore, we do not know much about the subcompartmentalization of the Golgi apparatus, and we cannot exclude that some biosynthetic reactions take place in different types of post-Golgi vesicles. For the Arabidopsis members of GT47 we have compiled predictions made by publicly available bioinformatic tools, published experimental data, and our own preliminary results below. Bioinformatics-based prediction of protein subcellular localization depends on the software to be used. ARAMEMNON is a web-based prediction database that compiles results obtained from more than a dozen programs including TmHMM and SignalP (Schwacke et al. 2003). A brief summary of results available through this database and SignalP is presented in Table 10.1. The GT47 proteins that have been studied in some detail include AtGUT1/ AtIRX10L, AtGUT2/AtTRX10, FRA8/IRX7, ARAD1, MUR3 and XGD1, which have all been described above. As for AtGUT1/AtIRX10L and AtGUT2/AtIRX10, there is a confusion of nomenclature (see Section 10.4), but we use AtGUT1/AtIRX10L (At5g61840) and AtGUT2/AtIRX10 (At1g27440) throughout in this section. For all these six proteins there is positive evidence that they are located in the Golgi, in most cases based on bioimaging using fluorescent fusion proteins or by proteomic LOPIT analysis (Dunkley et al. 2006). Secondary structure prediction programs agree that FRA8 and MUR3 are typical type II membrane proteins with a signal anchor. However, for the other proteins the results are more ambiguous. Some programs predict that these proteins are soluble or that they have a hydrophobic signal peptide that is cleaved off, while other programs agree that all the proteins are type II membrane proteins. AtGUT1/AtIRX10L and AtGUT2/AtIRX10 have weakly defined hydrophobic regions in the N-termini. Both proteins have been located in the Golgi vesicles by proteomic analysis (Dunkley et al. 2006) and we have shown that an AtGUT1/AtIRX10L-YFP fusion protein that was transiently expressed in N. benthamiana showed a punctate signal pattern that is characteristic of Golgi localization (Y. Sakuragi & H.V. Scheller, unpublished). However, bioinformatics analysis indicates that these transmembrane domains serve as signal peptides and that processed proteins are secreted (Table 10.1). This

Glycosyltransferases of the GT47 family

■

279

Table 10.1 Subcellular localization of GT47 proteins. Subcellular localizations of GTs are indicated based on existent experimental data, where available, and on analysis by publicly available bioinformatics programs

GUT1 GUT2 MUR3 FRA8 ARAD1

XGD1

Experimental data

Predictions by Aramemnon

Subcellular localization

TMsa

N-terb

Subcellular targeting

1 (0, 2)

Cyto

SP

1 (0, 2)

Cyto

SP

1 (2, 3)

Cyto

SP = Mite

1 (2,3)

Cyto

SP > Mit

Yes (21aa) Yes (21aa) Yes (55aa) No

1 (2, 3)

Cyto

SP > Mit

1 (2)

Cyto

SP

Golgi (LOPIT, bioimaging) Golgi (LOPIT) Golgi(SDG, bioimaging) Golgi (bioimaging) Golgi (bioimaging, LOPIT) Golgi (bioimaging)

SignalPHMM

SignalPNN

Yes (21aa) Yes (22aa) No

Yes (22aa) Yes (22aa) No

No

No

No

Yes (33aa)

Yes (28aa)

Yes (28aa)

No

Yes (33aa)

PrediSi

Cyto, cytosolic; Mit, mitochondria; N-ter, N-terminal domain; SP, secreted protein; TM, transmembrane domain. a Predicted TMs with confidence values less than 0.5 are shown in parentheses. b No GPI anchor site predicted by the bi-Pi program. Predictions for signal peptides are shown in the three rightmost columns. ‘Yes’ indicates that signal peptides are predicted by the respective program with predicted cite of signal cleavage in parentheses (where ‘aa’ indicates the position at which the amino acid residue where the predicted cleavage occurs), while ‘No’ indicates absence of predicted signal cleavage.

apparent discrepancy between the experimental and bioinformatics data may represent the false assignment of signal cleavage sites by the programs in these proteins (see Table 10.1). Alternatively, and more intriguingly, it is plausible that AtGUT1/AtIRX10L interacts with a yet unknown Golgitargeting protein and is tethered in this subcellular compartment via protein complex formation. For ARAD1 and XGD1 the predictions agree that both proteins have a transmembrane helix in the N-terminus, while ambiguity exists about whether this sequence is cleaved off or not (Table 10.1). YFP-fusions of both proteins have been located in the Golgi (Jensen et al. 2008; J.K. Jensen, Y. Sakuragi & H.V. Scheller, unpublished). Furthermore, immunoblot analysis has demonstrated that the epitope tags of His and myc fused to the N-terminus of ARAD1 are retained, providing evidence that the transmembrane helix of ARAD1 is not cleaved off (J.K. Jensen, J. Harholt & H.V. Scheller, unpublished).

280

■

Plant Polysaccharides

10.8 Conclusion GT47 is a large family of GTs in plants. Six Arabidopsis proteins from three different groups are all clearly involved in cell wall biogenesis and so far no GT47 protein has been shown to have a function not related to the cell wall. As for other GTs involved in cell wall biogenesis, it has been difficult to determine their exact role. Many insertional mutants have been investigated but in most cases the phenotypic changes are too subtle to be readily detected. We believe that it will turn out to be a more useful approach to determine the biochemical activity of all the proteins using robust methods for heterologous production and analysis of activity.

References Andersson, S.-I., Samuelson, O., Ishihara, M., Shimizu, K. (1983) Structure of the reducing end-groups in spruce xylan. Carbohydrate Research, 111, 283–288. Bauer, S., Vasu, P., Persson, S., Mort, A.J., Somerville, C.R. (2006) Development and application of a suite of polysaccharide-degrading enzymes for analyzing plant cell walls. Proceedings of the National Academy of Sciences of the U S A, 103, 11417–11422. Brown, D.M., Zeef, L.A.H., Ellis, J., et al. (2005) Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics. Plant Cell, 17, 2281–2295. Darvill, J.E., McNeil, M., Darvill, A.G., Albersheim, P. (1980) The structure of plant cell walls. XI. Glucuronoarabinoxylan, a second hemicellulose in the primary cell wall of suspension-cultured sycamore cells. Plant Physiology, 66, 1135–1139. Dunkley, T.P., Hester, S., Shadforth, I.P., et al. (2006) Mapping the Arabidopsis organelle proteome. Proceedings of the National Academy of Sciences of the U S A, 103, 6518–6523. Ellis, C., Karafyllidis, K., Wasternack, C., Turner, J.G. (2002) The Arabidopsis mutant cev1 links cell wal signaling to jasmonate and ethylene responses. Plant Cell, 14, 1557–1566. Gasper, Y.M., Nam, J., Schultz, C.J., et al. (2004) Characterization of the Arabidopsis lysine-rich arabinogalactan-protein AtAGp17 mutant (rat1) that results in a decreased efficiency of Agrobacterium transformation. Plant Physiology, 135, 2161–2171. Harholt, J., Jensen, J.K., Sørensen, S.O., Orfila, C., Pauly, M., Scheller, H.V. (2006) ARABINAN DEFICIENT 1 is a putative arabinosyltransferase involved in biosynthesis of pectic arabinan in Arabidopsis. Plant Physiology, 140, 49–58. Hernández-Blanco, C., Feng, D.X., Hu, J., et al. (2007) Impairment of cellulose synthases required for Arabidopsis secondary cell wall formation enhances disease resistance. Plant Cell, 10, 890–903. Hicks, G.R., Hironaka, C.M., Dauvillee, D., et al. (2001) When simpler is better. Unicellular green algae for discovering new genes and functions in carbohydrate metabolism. Plant Physiology, 127, 1334–1338.

Glycosyltransferases of the GT47 family

■

281

Iwai, H., Masaoka, N., Ishii, T., Satoh, S. (2002) A pectin glucuronosyltransferase gene is essential for intercellular attachment in the plant meristem. Proceedings of the National Academy of Sciences of the U S A, 99, 16319–16324. Jensen, J.K., Sørensen, S.O., Harholt, J., et al. (2008) Identification of a xylogalacturonan xylosyltransferase involved in pectin biosynthesis in Arabidopsis. Plant Cell, 20, 1289–1302. Jia, Z., Cash, M., Darvill, A.G., York, W.S. (2005) NMR characterization of endogenously O-acetylated oligosaccharides isolated from tomato (Lycopersicon esculentum) xyloglucan. Carbohydrate Research, 340, 1818–1825. Jones, L., Milne, J.L., Ashford, D., McCann, M.C., McQueen-Mason, S.J. (2005) A conserved functional role of pectic polymers in stomatal guard cells from a range of plant species. Planta, 221, 255–264. Konishi, T., Ono, H., Ohnishi-Kameyama, M., Kaneko, S., Ishii, T. (2006) Identification of a mung bean arabinofuranosyltransferase that transfers arabinofuranosyl residues onto (1, 5)-linked alpha-L-arabino-oligosaccharides. Plant Physiology, 141, 1098–1105. Lee, C.H., O’Neill, M.A., Tsumuraya, Y., et al. (2007) The irregular xylem9 mutant is deficient in xylan xylosyltransferase activity. Plant Cell Physiology, 48, 1624–1634. Li, X., Cordero, I., Caplan, J., Mølhøj, M., Reiter W-D. (2004) Molecular analysis of 10 coding regions from Arabidopsis that are homologous to the MUR3 xyloglucan galactosyltransferase. Plant Physiology, 134, 940–950. Madson, M., Dunand, C., Li, X., et al. (2003) The MUR3 gene of Arabidopsis encodes a xyloglucan galactosyltransferase that is evolutionarily related to animal exostosins. Plant Cell, 15, 1662–1670. Marchant, A., Wu A-M., Hörnblad, E., et al. (2007) Functional characterization of the Arabidopsis GT47 GUT1 and GUT2 glycosyltransferases. Physiologia Plantarum, 130, Suppl. XIth Cell Wall Meeting Abstract 6. McCartney, L., Marcus, S.E., Knox, J.P. (2005) Monoclonal antibodies to plant cell wall xylans and arabinoxylans. Journal of Histochemistry and Cytochemistry, 53, 543–546. Mitchell, R.A.C., Dupree, P., Shewry, P.R. (2007) A novel bioinformatics approach identifies chadidate genese for the synthesis and feryloylation of arabinoxylan. Plant Physiology, 144, 43–53. Møller, I., Sørensen, I., Bernal, A.J., et al. (2007) High-throughput mapping of cellwall polymers within and between plants using novel microarrays. Plant Journal, 50, 1118–1128. Nishimura, M.T., Stein, M., Hou, B.H., Vogel, J.P., Edwards, H., Somerville, S.C. (2003) Loss of a callose synthase results in salicylic acid-dependent disease resistance. Science, 301, 969–972. Pagnussat, G.C., Yu, H.J., Ngo, Q.A., et al. (2005) Genetic and molecular identification of genes required for female gametophyte development and function in Arabidopsis. Development, 132, 603–614. Peña, M.J., Ryden, P., Madson, M., Smith, A.C., Carpita, N.C. (2004) The galactose residues of xyloglucan are essential to maintain mechanical strength of the primary cell walls in Arabidopsis during growth. Plant Physiology, 134, 443–451. Peña, M.J., Zhong, R., Zhou G-K., et al. (2007) Arabidopsis irregular xylem8 and irregular xylem9: Implications for the complexity of glucuronoxylan biosynthesis. Plant Cell, 19, 549–563.

282

■

Plant Polysaccharides

Pérez, S., Rodríguez-Carvajal, M.A., Doco, T. (2003) A complex plant cell wall polysaccharide: rhamnogalacturonan II. A structure in quest of a function. Biochimie, 85, 109–121. Persson, S., Wei, H.R., Milne, J., et al. (2005) Identification of genes required for cellulose synthesis by regression analysis of public microarray data sets. Proceedings of the National Academy of Sciences of the U S A, 102, 8633–8638. Persson, S, Caffall, K.H., Freshour G., et al. (2007) The Arabidopsis irregular xylem8 mutant is deficient in glucuronoxylan and homogalacturonan, which are essential for secondary cell wall integrity. Plant Cell, 19, 237–255. Reiter W-D., Chapple, C., Somerville, C.R. (1997) Mutant of Arabidopsis thaliana with altered cell wall polysaccharide composition. Plant Journal, 12, 335–345. Ryden, P., Sugimoto-Shirasu, K., Smith, A.C., Findlay, K., Reiter W-D., McCann, M.C. (2003) Tensile properties of Arabidopsis cell walls depend on both a xyloglucan cross-linked microfibrillar network and rhamnogalacturonan II-borate complexes. Plant Physiology, 132, 1033–1040. Scheller, H.V., Jensen, J.K., Sørensen, S.O., Harholt, J., Geshi, N. (2007) Biosynthesis of pectin. Physiologia Plantarum, 129, 283–295. Schwacke, R., Schneider, A., Van der Graaff, E., et al. (2003) ARAMEMNON, a novel database for Arabidopsis integral membrane proteins. Plant Phyiology, 131, 16–26. Séveno, M., Voxeur, A., Rihouey, C., et al. (2009) Structural characterisation of the pectic polysaccharide rhamnogalacturonan II using an acidic fingerprinting methodology. Planta, 230, 947–957. Shimizu, K., Ishihara, M., Ishihara, T. (1976) Hemicellulases of brown rotting fungus, Tyromyces palustris. II. The oligosaccharides from the hydrolysate of a hardwood xylan by the intracellular xylanase. Mokuzai Gakkaishi, 22, 618–625. Tamura, K., Shimada, T., Kondo, M., et al. (2005) KATAMARI1/MURUS3 is a novel Golgi membrane protein that is required for endomembrane organization in Arabidopsis. Plant Cell, 17, 1764–1776. Taylor, N.G., Scheible, W.R., Cutler, S., et al. (1999) The irregular xylem3 locus of arabidopsis encodes a cellulose synthase required for secondary cell wall synthesis. Plant Cell, 11, 769-779. Taylor, N.G., Laurie, S., Turner, S.R. (2000) Multiple cellulose synthase catalytic subunits are required for cellulose synthesis in Arabidopsis. Plant Cell, 12, 2529-2539. Ubeda-Tomas, S., Edvardsson, E., Eland, C., et al. (2007) Genomic-assisted identification of genes involved in secondary growth in Arabidopsis utilizing transcript profiling of poplar wood-forming tissues. Physiologia Plantarum, 129, 415–428. Ulvskov, P., Wium, H., Bruce, D., Jørgensen, B., Qvist, K.B., Skjøt, M., Hepworth, D., Borkhardt, B., Sørensen, S.O. (2005) Biophysical consequences of remodeling the neutral side chains of rhamnogalacturonan I in tubers of transgenic potatoes. Planta, 220, 609–620. Vogel, J.P., Raab, T.K., Schiff, C., Somerville, S.C. (2002) PMR6, a pectate lyase-like gene required for powdery mildew susceptibility in Arabidopsis. Plant Cell, 14, 2095–2106. Vogel, J.P., Raab, T.K., Somerville, C.R., Somerville, S.C. (2004) Mutations in PMR5 result in powdery mildew resistance and altered cell wall composition. Plant Journal, 40, 968–978. Willats, W.G.T., McCartney, L., Steele-King, C.G., et al. (2004) A xylogalacturonan epitope is specifically associated with plant cell detachment. Planta, 218, 673–681.

Glycosyltransferases of the GT47 family

■

283

Wu, A-M., Rihouey, C., Seveno, M., et al. (2008) The Arabidopsis IRX10 and IRX10LIKE glycosyltransferases are critical for glucuronoxylan biosynthesis during secondary cell wall formation. Plant Journal, 57, 718–731. Zablackis, E., Huang, J., Müller, B., Darvill, A.G., Albersheim, P. (1995) Chracterization of the cell-wall polysaccharides of Arabidopsis thaliana leaves. Plant Physiology, 107, 1129–1138. Zandleven, J., Sorensen, S.O., Harholt, J., et al. (2007) Xylogalacturonan exists in cell walls from various tissues of Arabidopsis thaliana. Phytochemistry, 68, 1219-1226. Zhong, R., Pena, M.J., Zhou G-K., Narin, C.J., et al. (2005) Arabidopsis Fragile Fiber8, which encodes a putative glucuronyltransferase, is essential for normal secondary wall synthesis. Plant Cell, 17, 3390–3408. Zhou, G.K., Zhong, R.Q., Richardson, E.A., et al. (2006) The poplar glycosyltransferase GT47C is functionally conserved with Arabidopsis Fragile fiber8. Plant and Cell Physiology, 47, 1229–1240. Zhu, Y., Nam, J., Carpita, N.C., Matthysse, A.G., Gelvin, S.B. (2003) Agrobacteriummediated root transformation is inhibited by mutation of an Arabidopsis cellulose synthase-like gene. Plant Physiology, 133, 1000–1010.

Annual Plant Reviews (2011) 41, 285–304 doi: 10.1002/9781444391015.ch11

Chapter 11

THE PLANT GLYCOSYLTRANSFERASE FAMILY GT64: IN SEARCH OF A FUNCTION Ellinor Edvardsson1, Sunil Kumar Singh2, Min-Soo Yun1,3, Agata Mansfeld4, Marie-Theres Hauser4 and Alan Marchant1,5 1 Department of Forest Genetics and Plant Physiology, Swedish University of Agricultural Sciences (SLU), SE-901 83 Umeå, Sweden 2 Department of Plant Cell and Molecular Biology, Indian Institute of Advanced Research (HAR), Koba, Gandhinagar-382007, India 3 Current address: Plant Genetic Engineering Research Unit, National Institute of Agrobiological Sciences, 2-1-2 Kannondai, Tsukuba, Ibaraki 305-8602, Japan 4 Institute of Applied Genetics and Cell Biology, BOKU – University of Natural Resources and Applied Life Sciences, Muthgasse 18 A-1190 Vienna, Austria 5 School of Biological Sciences, Life Sciences Building 85, Highfield Campus, University of Southampton, SO17 1BJ, UK

Manuscript received January 2009

Abstract: The glycosyltransferase 64 (GT64) family includes members from a diverse range of species including human, Xenopus and Drosophila as well as plant species including Arabidopsis, the moss Physcomitrella and poplar. The majority of the animal GT64 proteins are bimodular, consisting of an N-terminal GT47 domain linked to a GT64 domain. The animal GT64 domain has been found to have GlcNAc transferase activity and functions during heparan sulphate biosynthesis. However, the biochemical activity of the plant GT64 proteins has yet to be established. Arabidopsis has three GT64 members which have been named ECTOPICALLY PARTING CELLS1 (EPC1), EPC-LIKE1 (EPC-L1) and EPC-LIKE2 (EPC-L2). The EPC1 and EPC-L1 proteins are comprised of a single GT64 domain whereas the EPC-L2 protein has an additional N-terminal domain which does not show homology to any other protein other than EPC-L2 homologues from other plants. Mutation of the Arabidopsis GT64 EPC1 results in a plant with severely Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

285

286

■

Plant Polysaccharides

reduced stature, demonstrating the importance of its function for normal plant development. Neutral sugar analysis of the epc1 cell wall shows a reduction in galactose but there is currently no enzymatic activity demonstrated to explain this alteration. In contrast to EPC1, EPC-L2 is specifically expressed in reproductive tissues and mutations in the gene result in defects in pollen development, reduced seed formation and embryo abortions at the late globular stage. Keywords: Arabidopsis; development; EPC1; glycosyltransferase 64

11.1 Introduction The biosynthesis of complex polysaccharide structures is a key process in living organisms. Central to this are the glycosyltransferases (GTs), a large and diverse group of enzymes which typically catalyse transfer of a sugar moiety from a nucleoside diphosphate sugar donor to an acceptor to form a glycosidic bond. To date more than 415 GTs have been identified in the Arabidopsis genome (Scheible & Pauly 2004) and it is likely that there are more to be discovered. Glycosyltransferases function in the synthesis of various cell wall polysaccharides (see Scheible & Pauly 2004 for a review), glycoproteins and glycolipids (Jarvis et al. 2000; Jorasch et al. 2000) as well as in the transfer of sugars to small molecules such as flavonoids (Hirotani et al. 2000), hormones such as indole-3-acetic acid and brassinosteroids (Szerszen et al. 1994; Poppenberger et al. 2005) phytotoxins and xenobiotics (for review see Bowles et al. 2005). The biological activity of only a small proportion of the GTs have been determined thus far, and establishing function often presents a major challenge because of the many and varied reactions that they can potentially catalyse.

11.2 GT64 family members are found in a diverse range of species The GT64 family is a relatively new addition to the CAZY classification of glycosyltransferases and is characterized by the possession of a putative α-1,4-N-acetylglucosaminyltransferase domain. Examples of GT64 family members are found in a wide range of species including, but not limited to human, Xenopus, Caenorhabditis and Drosophila (http://www.cazy.org/ fam/acc_GT.html). Within the plant kingdom proteins belonging to the GT64 family have been identified in the basic single-celled alga Ostreococcus lucimarinus, grape (Vitis vinifera), poplar (http://genome.jgi-psf.org/ Poptr1_1/Poptr1_1.home.html, Edvardsson & Marchant unpublished results), rice, the moss Physcomitrella patens and Arabidopsis thaliana (Singh et al. 2005; Bown et al. 2007). The high degree of sequence conservation between the plant members suggests that examples from the GT64 family will be found in most if not all plant species.

The Plant Glycosyltransferase Family GT64: in Search of a Function

■

287

Originally the GT64 family was included in the GT47 classification, stemming from the fact that most of the animal GT64 proteins are bimodular, possessing an N-terminal GT47 family domain and a C-terminal GT64 domain. In contrast, no examples have been described to date of a bimodular plant protein possessing both GT47 and GT64 domains. Currently the bestcharacterized GT64 family members are those from human and Drosophila systems. The human EXOSTOSIN or EXT family is made up of five members (EXT1, EXT2 and EXT-LIKE 1, 2 and 3) that have been shown to function at different stages during the synthesis of heparan sulfate (Lind et al. 1998; McCormick et al. 2000; Kim et al. 2001). The name exostosin refers to the finding that mutations of either EXT1 or EXT2 results in a disorder known as hereditary multiple exostoses (HME) which is characterized by formation of benign outgrowths on the ends of long bones (Solomon 1963). Four of the human EXTs are bimodular, consisting of an N-terminal GT47 domain which for EXT1 and EXT2 has been shown to catalyse addition of β-1,4-GlcA residues and a C-terminal GT64 domain which adds α-1,4-GlcNAc residues (Kitagawa et al. 1999). The EXT1 and EXT2 proteins have been shown to form a hetero-oligomeric complex in the Golgi apparatus which creates a more active enzyme (McCormick et al. 2000; Senay et al. 2000). In contrast, both the human and murine EXT-L2 proteins contain just the GT64 domain and exhibit GalNAc/GlcNAc transferase activity. EXT-L2 is believed to function in the initiation of heparan sulfate biosynthesis (Kitagawa et al. 1999). Screens for mutations affecting morphogen signalling pathways in Drosophila have identified three genes named tout-velu (Ttv), sister of tout velu (Sotv) and brother of tout velu (Botv) which are homologues of the human EXT1, EXT2 and EXTL3. Mutation of Ttv, Sotv or Botv has been shown to impair HS synthesis which affected Hedgehog (Hg), Wingless (Wg) and βDecapentaplegic (Dpp) signalling (Bornemann et al. 2004; Han et al. 2004; Takei et al. 2004).

11.3

The Arabidopsis GT64 family

11.3.1 Structure of the GT64 genes and proteins The Arabidopsis GT64 family is small, made up of just three members encoded by the At3g55830, At1g80290 and At5g04500 genes. Analysis of a mutant in the At3g55830 gene led to it being named ECTOPICALLY PARTING CELLS 1 (EPC1) (Singh et al. 2005) and here it is proposed to name At1g80290 as EPC-LIKE1 (EPC-L1) and At5g04500 as EPC-LIKE2 (EPC-L2). Despite being designated as members of the same family, the EPC1, EPC-L1 and EPC-L2 proteins show only 27% identity and 42% similarity to each other. However, sequence comparisons between GT64 proteins from Arabidopsis, poplar, rice and human show that there are highly conserved homologues

288

■

Plant Polysaccharides

PtEPC1-A

PtEPC1-B

AtEPC1

Os06g49150 EXTL2(human) EXTL3(human) EXT2(human)

PtEPCL1 AtEPC-L1

EXTL1(human) EXT1(human)

PtEPCL2

Os03g48010 Os05g46260 AtEPC-L2

Figure 11.1 Phylogenetic tree of GT64 family members from Arabidopsis, poplar, rice and human. Protein sequences from Arabidopsis (EPC1, EPCL1 and EPCL2), poplar (PttEPC1.1- estExt_fgenesh4_pg.C_LG_X1712; PttEPC1.2-eugene3.00080589, PttEPCL2estExt_fgenesh4_pg.C_LG_X2099 and PttEPCL1; gw1.147.192.1), rice (Os03g0684500, Os05g0540000 and P0018H04.8) and human (EXT1, EXT2, EXTL1, EXTL2 and EXTL3) were aligned using ClustalX (1.8) and the phylogenetic tree created using Phylodraw from the resulting .dnd file.

for EPC1, EPC-L1 and EPC-L2 in other plant species. In contrast, the five human EXT proteins are more divergent from the plant GT64s and cluster together on a separate branch of the tree (Fig. 11.1). The EPC1 gene contains 5 exons encoding a predicted protein of 334 amino acids with a DXD motif between residues 166 and 168 which is typical of the GT-A superfamily (Franco & Rigden 2003). The DXD motif, which is conserved across the GT64 family, is believed to be important for interaction with a Mn2+ cofactor that stabilizes binding of the diphosphate moiety of the UDP-sugar substrate (Ünligil & Rini 2000). The EPC1 protein has a single predicted transmembrane span between amino acids 28 and 49 (http:// phobius.sbc.su.se/; Krogh et al. 2001) and has been reported to localize to small highly mobile, punctate cytoplasmic structures resembling the Golgi apparatus (Bown et al. 2007). This is consistent with a putative role at the site of cell wall polysaccharide synthesis, but to date the experimental evidence to support this possible functional role is lacking. The EPC-L1 gene encodes a predicted protein of 329 amino acids, similar in length to EPC1 and which, like EPC1, has a DXD motif between amino acids 139 and 141. However, in contrast to EPC1, the EPC-L1 gene lacks introns and the EPC-L1 protein does not have any predicted transmembrane

The Plant Glycosyltransferase Family GT64: in Search of a Function

■

289

domains (http://phobius.sbc.su.se). This is also the case for the closest poplar EPC-L1 homologue encoded by the gene gw1.147.192.1. However, the EPC-L1 protein has a predicted signal peptide (amino acids 1–22) which is absent in both EPC1 and EPC-L2. The EPC-L2 protein is predicted to be more than double the size of EPC1 and EPC-L1, having 764 amino acids, and is encoded by the At5g04500 gene which has 4 exons. The final exon encodes the C-terminal 563 amino acids and so resembles the structure of the At1g80290 (EPC-L1) gene in having no introns within the part of the gene encoding the GT64 domain. The C-terminal 320 amino acids of EPC-L2 shows homology to the other GT64 family members but the N-terminal domain is unique in the Arabidopsis genome, although it is significant to note that poplar has a highly conserved EPC-L2 homologue which contains both the GT64 domain and the N-terminal domain. The bimodular arrangement of EPC-L2 is further highlighted by the prediction of three transmembrane regions between residues 40–62, 394–421 and 478–499 which may place the two globular domains (residues 63–393 and residues 500–764) on the same face of the membrane. This putative bimodular arrangement of EPC-L2 is interesting given the structure of four of the human EXTs which comprise both GT47 and GT64 domains. However, the N-terminal domain of EPC-L2 exhibits no apparent homology to any other known or predicted glycosyltransferases and so the possible function of this domain has yet to be determined. Future experiments will be necessary to determine the function of the two domains and whether this function is dependent on the physical link between them. The relatively low sequence identity between the Arabidopsis GT64 members raises the question of whether they are derived from a common ancestor. Comparison of the GT64 family members from Arabidopsis and poplar provides evidence that the function of the proteins may be distinct and that the function is conserved between the two species. This is apparent from the high degree of sequence identity between the poplar EPC1 homologues and the Arabidopsis EPC1 which is much higher that the identity between the three Arabidopsis GT64s. The same is true for the poplar EPC-L1 and EPC-L2 homologues, which are highly conserved with the Arabidopsis EPC-L1 and EPC-L2 respectively although the poplar EPC-L2 does have an apparent additional 63-amino-acid insertion within the N-terminal half of the protein. The gene structure is also highly conserved between the two species. The Arabidopsis and poplar EPC1 genes display the same exon/ intron organization, whereas the EPC-L1 genes in Arabidopsis and Poplar both lack introns. The structure of the Arabidopsis and poplar EPC-L2 genes is less well conserved, although in both genes the entire GT64 domain is encoded by the final exon. This apparent gene structure conservation provides evidence that evolution of the plant GT64 family took place prior to the divergence of poplar and Arabidopsis which is estimated to have occurred around 110 million years ago (Wikström et al. 2001).

290

■

Plant Polysaccharides

11.3.2 Function of Arabidopsis GT64s proteins 11.3.2.1 At3g55830 EPC1 Thus far, of the three GT64 genes in Arabidopsis only At3g55830 (EPC1) has been studied in any detail, yet its function remains somewhat enigmatic. Originally the Arabidopsis At3g55830 gene was identified on the basis of homology to a poplar gene which was up-regulated during the early stages of wood formation prior to the initiation of secondary cell wall formation (Hertzberg et al. 2001; Singh et al. 2005). T-DNA insertion mutants in the At3g55830 gene resulted in a severe phenotype characterized by reduced leaf expansion, aberrant vascular development, infrequent formation of a stunted inflorescence structure which only rarely produced seed, and a reduction in root elongation (Singh et al. 2005). The severity of the epc1 mutant phenotype demonstrates that this gene has an important role during development and indicates that EPC-L1 and EPC-L2 are not functionally redundant with EPC1. Indeed it was found that expression of EPC-L1 under control of the EPC1 promoter was unable to rescue the epc1 mutant phenotype, whereas pEPC1::EPC1 expression resulted in a complete rescue (Singh et al. 2005; Singh & Marchant, unpublished results). Several lines of evidence demonstrated a reduction in cell-to-cell adhesion in the epc1 mutant within hypocotyl and leaf tissues. Coupled with the reduced cell-to-cell adhesion in the cortical parenchyma of the hypocotyl there was a dramatic increase in the amount of secondary growth within the stele tissues, although it is still not clear whether the cell adhesion defect directly contributes to this phenomenon. A subsequent independent study also isolated the At3g55830 gene from a screen for stature mutants in a T-DNA tagged Arabidopsis population in the WS ecotype. Despite the fact that this new epc1–2 allele displayed similar but slightly less severe growth phenotypes compared to epc1 (Singh et al. 2005), the authors did not detect evidence of cell separation in either epc1 or epc1–2 (Bown et al. 2007). The reasons for this difference in phenotype are unclear but it may reflect the effect of subtle differences in growth conditions, although this remains to be fully tested. The epc1–2 allele was shown to be hypersensitive to the hormone abscisic acid (ABA) while the response to the other hormones giberellic acid (GA), auxin, ethylene and epibrassinolide was normal. Interestingly, while the hypocotyl growth of epc1–2 in response to GA was normal, higher levels were required to stimulate germination compared to the wild type. GA is known to act antagonistically to ABA in seed germination, supporting the conclusion that epc1–2 is hypersensitive to ABA. Analysis of the cell wall composition of the epc1 mutants has not provided a clear indication as to the function of the protein. Analysis of epc1 cell wall material showed an increase in glucose which was argued to reflect the abnormal callose deposition in the mutant tissues that is likely to be a wound-type response. There was also a small reduction in galactose though

The Plant Glycosyltransferase Family GT64: in Search of a Function

■

291

this was not significant compared to the wild-type (Singh et al. 2005). Further analysis of the epc1–2 allele found that β-(1–4)-galactan was reduced by around 50%, although this was not supported by immunolocalization studies using the LM5 antibody which recognizes β-(1–4)-galactan tetramers. Thus although there are indications of alterations in cell wall components, the evidence is far from providing solid clues as to the function of the enzyme. In a further study a hypomorphic allele of EPC1 was identified in a genetic screen for modulators of the cellulose deficient mutant pom pom1 (pom1) and therefore was named ENHANCER OF POM1 (EPO) (Hauser et al. 1995; Mouille et al. 2003; Mansfeld 2005). Work carried out on the epo/epc1–3 allele has provided further insight into a possible role for EPC1 in cell wall biosynthesis. The mutation in epc1/epo results in an amino acid substitution of a conserved Gly288 to Cys (Fig. 11.2A, B) and leads to a conformational change in the loop above the putative substrate binding pocket. Analysis of the α-1,4-N-acetylhexosaminyltranferase catalytic domain of the human EXTL2 protein has presented the possibility that the protein forms a homooctameric complex (based on data from the Protein Quaternary Structure

A

1

B

epc1-3/epo GGT TGT Gly288Cys ATG

TAA

115

2188

2498 bp

G288C

HsEXTL2 :T-QNCDDIAMNFIIAKHIGK-TSGIFVKPVNMDNLEKE------TNSGYSGMWHRAEHALQRSYCINKLVNIYDSMPLR MmEXTL2like :T-QNCDDIAMNFLVTRHTGK-PSGIFVKPINMVNLEKE------TN-GYSGMWHRAEHFLQRSYCINKLVNIYDGMPLK :T-QNCDDIAMNFIVSKHTGK-SSGIFVKPINIINLEKE------TN-GYSGMWHRAEHFLQRSYCLNKLVNIFSGMPLK RnEXTL2 XlNP_001085503:T-QNCDDITMNFMVANHLGK-ASGVLVKPTDMRNLEKE------AGSGYTGMWHRAEHLLQRSYCLNKLAEIYGTMPLK GgXP_422308 :T-QNCDDIAMNFLVAKHIGK-PSGVFVKPVDLRNLEKD------TNSGYSGMWHRAEHLLQRSYCVNKLVNVYDGMPLK DrNP_001070029:T-QNCDDIAMNFIVARQLSKRPSGVFVKPVHMSNLEKD------ASSGFVGMWHRPEHMLQRSYCLNKLAQIYGHMPLR BtNP_001069692:T-QNCDDIAMNFIIAKHTGK-TSGVFVKPVNIANLEKE------TTGGYSGMWHRAEHFLQRSYCINKLVNIYDSMPLK :I-QNCDDIAMNFIIAKHTGK-TSGVFVKPVNMGNLEKE------SNGGYPGMWHRAEHFLQRSYCINKLVNIYNSMPLK CfXP_537051 At3g55830 :KNRNCEDIAMSFLIANATNA--PAIWVKG-------------KIYEIGSTGISSIGGHTEKRTHCVNRFVAEFGKMPLV VvESTmerge :KNRNCEDIAMSFLVANVTGS--PPIWVKG-------------KIFEIGSTGISSLGGHTEKRSQCVNRFAMEYGRMPLV LsESTmerge :RNRNCEDIAMSFLVANATGA--PPIWAKG-------------KIYEIGSTGISSLGGHSDKRTECVNRFVSEFGKMPLV PtESTmerge :KNRNCEDIAMSFLVANATGA--PPIWVKG-------------KIFEIGSTGISXLGGHGERRTXCVNXFAAEFGRMPLV GmESTmerge :KNRNCEDIAMSFLVANATGA--PPIWVKG-------------KIFEIGSTGISSLGGHSERRTECVNRFAAVYGRMPLV OsiEAZ02288 :ENRNCEDIAMSFLVANVTGS--PPIWVQGGHTEVPFRDVTAGRIFEIGSSGISSLKGHDLQRSKCLNTFSAMYGHMPLV OsjBAD53800 :ENRNCEDIAMSFLVANVTGS--PPIWVQGGHTEVPFRDVTAGRIFEIGSSGISSLKGHDLQRSKCLNTFSAMYGHMPLV

Figure 11.2 Localization and effects of the epc1–3/epo mutation. (A) Schematic representation of the EPC1 (At3g55830) gene showing the position and identity of the missense mutation in the epc1–3/epo allele. (B) Multiple alignment of the C-terminal conserved 70 amino acids. The amino acid substitution in epc1–3/epo affects a glycine residue that is conserved between the human (Hs, Homo sapiens), mouse (Mm, Mus musculus), rat (Rn, Rattus norvegicus), frog (Xl, Xenopus laevis), chicken (Gg, Gallus gallus), zebra fish (Dr, Danio rerio) and cow (Bt, Bos taurus) Exostosin-like 2 (EXTL2) protein and the closest plant homologues in grape (Vv, Vitis vinifera), rice (Osi, Oryza sativa indica, Osj, Orzyza sativa japonica), Populus (Pt, Populus tremula), soybean (Gm, Glycine max) and lettuce (Ls, Lactuca serriola). The alignment was generated with ClustalX (1.8) and decorated in GeneDoc (black, 100% identity, grey 80% identity and light grey 60% identity).

292

■

Plant Polysaccharides

EPC1

epc1-3/epo

*

*

*

*

*

MmEXTL2 dimer MmEXTL2 octamer

Figure 11.3 3D modelling prediction of EPC1 and epc1–3/epo mutant proteins based on previous models of the catalytic domain of the mouse α-1,4-Nacteylhexosaminyltranferase (MmEXTL2). The EPC1 and epc1–3/epo proteins were modelled with Geno3D (Combet et al. 2002) based on the catalytic domain of the mouse α-1,4-N-acteylhexosaminyltranferase (MmEXTL2) with UDP and acceptor GlcUAβ1– 3Galβ1-O-naphthalenelmethanol (PDB accessions 1ON8 1ON6 and 1OMX). The Gly288 to cystine amino acid substitution in epc1–3/epo changes the loop structure (star) above the substrate pocket that faces the pore in the octamere. The protein backbones are shown the rainbow colouring mode where the N-terminus is blue and the C-terminus red and the sequences between follow a spectral rainbow. Donor and acceptor substrates are in red and lilac spheres. Residues between 277 and 281 are disordered but become ordered upon acceptor binding. The EXTL2 catalytic domain is predicted to dimerize and for the dimers to form octamers. Figures were created using FirstGlance in Jmol.

server http://www.ebi.ac.uk/thornton-srv/databases/cgi-bin/pdbsum/ GetPage.pl?pdbcode=1on6) If EPC1 folds in a similar way to the human EXTL2 then the affected loop in the epo/epc1–3 allele would face the pore in the octamer (Fig. 11.3). Thus it is possible that the Gly288 to Cys mutation may directly affect the catalytic activity of the mutant protein. POM1 encodes the chitinase-like protein AtCTL1 and is synonymous to ECTOPIC DEPOSITION OF LIGNIN IN PITH 1 (elp1, Zhong et al. 2000; Zhong et al. 2002; Rogers et al. 2005), ECTOPIC ROOT HAIR2 (erh2) (Schneider et al. 1997), SHORT HYPOCOTYL (shy-631 and shy-840, Reed et al. 1998) and was found in a screen for increased organ regeneration (Cary et al. 2001) and reduced abiotic stress tolerance (HOT2; Hong et al. 2003; Kwon et al. 2007). The identification of pom1 alleles in different genetic screens highlights the pleiotropic nature of the mutant phenotypes. However, most of the

The Plant Glycosyltransferase Family GT64: in Search of a Function

■

293

phenotypes such as aberrant cell expansion can be explained by a defect in cellulose synthesis (Mouille et al. 2003; Table 11.1). Interestingly, the phenotypes of the majority of the cellulose biosynthesis mutants are enhanced by sucrose but this enhancement is lower in epo/epc1–3 mutants (Table 11.1). Reduced cellulose in primary and secondary cell walls is known to induce wound and stress responses and this is consistent with the observed ectopic deposition of callose in the epc1 mutant (Singh et al. 2005). While the epc1–3/epo single mutants did not show a visible phenotype, the double mutants with pom1–9 exhibit synergistically reduced growth, lower cellulose content (Table 11.1) and enhanced ectopic deposition of lignin in stems. The genetic interaction of epc1–3/epo is not dependent on a pom1 null allele since the same synergistic defects were seen in both missense and null alleles of pom1 (Hauser, unpublished data), indicating that EPC1 is either a positive regulator of POM1 activity and/or they act in a common pathway. Further insight into a possible function of EPC1 during cellulose formation has been obtained from studies looking at the genetic interaction between epo/epc1–3 and different mutant alleles of the endo-1–4-β-d-glucanase KORRIGAN1 (KOR1). It has been postulated that KOR1 functions in chain termination during cellulose biosynthesis or in the degradation of glucan chains which are not incorporated into cellulose microfibrils (Mølhøj et al. 2001). Mutants in KOR1 have been identified in several independent genetic screens for root and hypocotyl morphogenesis genes (lit-1, kor1–1, rsw2–1, rsw2–3, Hauser et al. 1995; Nicol et al. 1998; Lane et al. 2001, Sato et al. 2001), as well as in screens for defects in primary (kor1–2, Zuo et al. 2000) and secondary cell wall formation (irx2–1, irx2–2, Szyjanowicz et al. 2004). Double mutants between the lions-tail (lit-1) allele of KOR1 (G344R mutation) and epc1–3/epo exhibit an enhanced phenotype although this is not observed with other alleles including kor1–1 (reduced KOR1 expression level) or rsw2–3 (S183N mutation) (Table 11.1). This indicates that EPC1 may act on the same pathway as POM1 and/or KOR1. Both KOR1-GFP and EPC1-GFP (Bown et al. 2007) are targeted to an intracellular compartment consistent with a Golgi localization. Further preliminary data indicate that a POM1-GFP fusion protein localizes to mobile intracellular compartments which may also represent the Golgi apparatus. These results suggest that EPC1 might traffic in the same compartments as KOR1 and POM1, consistent with a possible interacting function. However, a definitive function for EPC1 based on this data remains elusive and awaits further work. It has been found that chitinases can cleave the GlcNAc residues present in plant AGPs and that this activity can promote somatic embryogenesis (van Hengel et al. 1998) Thus POM1 either together with or separately from KOR1 and EPC1 could potentially act on GlcNAc-containing glycoproteins, AGP or glycolipids. AGPs have been proposed to tag the attachment sites for cellulose microfibrils and regulate their deposition (Gibeaut & Carpita 1991). Furthermore, AGP might be a source of molecular signals controlling the cell wall integrity (see later for further discussion about AGPs). Despite

10.3 10.9 7.8 3.3 9.4 4.0 10.7 1.8 1.0 5.6 4.6

± ± ± ± ± ± ± ± ± ± ±

1.2 1.0 0.6 1.2a 0.8 1.4a 2.0 0.3 0a 1.5 0.7

12.2 10.8 3.3 2.5 6.4 7.8 18.2 3.3 2.0 5.5 5.7

± ± ± ± ± ± ± ± ± ± ± 0.9 1.0 0.3 0.12a 1.4 1.5 2.2 0.4 0.4a 1.1 0.7

13.6 12.8 4.8 2.6 8.7 10.5 15.8 8.4 2.2 13.8 14.0

± ± ± ± ± ± ± ± ± ± ± 1.6 1.9 0.3 0.5a 1.3 0.8 1.8 1.4 0.3a 4.1 4.1

16.1 16.2 6.6 2.8 4.4 5.2 18.0 11.4 5.8 13.8 13.6

± ± ± ± ± ± ± ± ± ± ± 2.8 0.8 1.4 0.5a 0.6 0.9 1.5 1.5 0.5a 3.6 4.6

4.5% sucrose

0% sucrose

4.5% sucrose

0% sucrose 28.6 26.2 23.93 12.8 20.6 20.6 6.1 15.9 3.2 32.8 37.8

± ± ± ± ± ± ± ± ± ± ±

Plant height (cm) 2.1 3.1 .6 3.6a 9.8 9.8 1.5 1.6 0.8a 4.3 4.5

1.8 ± 1.8 ± 1.0 ± 1.4 ± 2.6 ± n.d. 1.7 ± 1.1 ± n.d. 1.2 ± n.d. 0.3

4.8 ± 0.6 2.8 ± 1.0 1.9 1.2 ± 0.7 0.4 0.2

0.6 0.5 0.4 0.3 0.5

3.2 3.6 2.0 2.0 1.4

± ± ± ± ±

4.5% sucrose(seedling)

0.6 0.4 0.1 0.2 0.3

0% sucrose(seedling)

Cellulose (mg/g FW)

n.d., not determined. a Student’s t-test between single cellulose deficient and their double mutants with epc1–3/epo p < 0.05.

Ws epc1–3/epo pom1–9 epo pom1–9 kor1–1 epo kor1–1 Col lit-1 epo lit-1 rsw2–3 epo rsw2–3

Genotype

Hypocotyl length (mm)

Root length (mm)

± ± ± ± ±

6.2 4.9 5.7 5.6a 3.8 31.3 ± 5.3 26.0 ± 7.6 n.d. 34.4 ± 6.5

27.2 24.8 22.2 14.7 11.7

Stems

■

Table 11.1 Summary of the phenotypes of epc1–3/epo, cellulose deficient and double mutant combinations. The effects of sucrose on root growth and elongation of etiolated hypocotyls were quantified 7 days after germination on MS medium with and without sucrose. The plant height was measured at the end of the vegetation period. Cellulose was quantified from 17-day-old in vitro cultured seedlings grown on either 0% or 4% sucrose and from stems of 4-week-old soil-grown plants using a modified version of the protocol of Updegraff (1969)

294 Plant Polysaccharides

The Plant Glycosyltransferase Family GT64: in Search of a Function

■

295

these hypotheses, detailed chemical analyses of cell wall components including AGPs, glycoproteins, glycolipids of epc1 mutants and overexpressors as well as biochemical and cell biological studies will be needed to finally reveal the function of EPC1 and its relation to POM1/AtCTL1 and KOR1. 11.3.2.2 At1g80290 EPC-L1 Currently virtually nothing is known about the possible function of EPC-L1. The gene is found to be expressed throughout the plant using a promoter:GUS reporter (Singh & Marchant, unpublished data) as well as from publicly available Genevestigator microarray data (https://www.genevestigator. ethz.ch/at/). In order to try to elucidate the function of EPC-L1, two independent homozygous Ds insertion lines (GT12821 and GT12784) were isolated from the CSHL population. In stark contrast to the epc1 phenotype, neither of the epc-L1 homozygous insertion lines displayed any visible phenotype. However, despite being homozygous for the insertion, quantitative RT-PCRs reveal that both lines retain more than 20% of the wild type EPC-L1 message level, making it possible that sufficient functional protein is still made accounting for the lack of a phenotype. (Singh & Marchant, unpublished results). In an alternative approach, RNAi lines in which the EPC-L1 gene is specifically down-regulated have been generated using resources from the AGRIKOLA project but this also has not revealed any visible phenotype. Despite this it is evident that the EPC-L1 protein is highly conserved throughout the plant kingdom, indicating that it is functionally important. 11.3.2.3 At5g04500 EPC-L2 The Arabidopsis EPC-L2 protein differs from EPC1 and EPC-L1 in having an additional domain at its N-terminus. The possible function of this domain is elusive as it does not show significant homology to any other protein in the Arabidopsis proteome. It is interesting to note that close homologues of EPC-L2 are found in both poplar and rice, demonstrating that the function of the protein is likely to be conserved in multiple species. Based on its expression pattern from the Genevestigator database as well as promoter::GUS analysis (Yun & Marchant, unpublished results), the function of EPC-L2 appears to be limited to reproductive tissues including anthers, pollen and developing embryos. The poplar EPC-L2 homologue is likely to have a similar expression pattern to its Arabidopsis counterpart, as the cDNA has been isolated from flower buds (http://www.populus. db.umu.se/result.php?id=117036). Consistent with the Arabidopsis EPC-L2 expression pattern, analysis of three independent T-DNA insertion alleles of EPC-L2 all display defects in pollen development and embryogenesis (Yun & Marchant, unpublished results). Compared to the wild type, homozygous epc-L2 mutants show a 50% reduction in the number of seeds that form and of these around one-fifth are small, shrivelled and nonviable. Although the homozygous epc-L2 mutant has obvious defects in the development of a subset of seed, the normal-appearing seed that are

296

■

Plant Polysaccharides

formed show no difference in germination compared to the wild type and produce wild type-appearing plants. Closer examination of developing seed shows that in a subset, embryo development is arrested at late globular/ heart stage and this is coupled with an apparent inhibition of normal endosperm formation (Yun, Singh, & Marchant, unpublished results). Arrest of embryogenesis at heart stage has previously been observed in a number of Arabidopsis mutants including fis1/mea, fis2 and fie/fis3 in which the endosperm fails to cellularize following fertilization (Ohad et al. 1996; Chaudhury et al. 1997; Grossniklaus et al. 1998; Ohad et al. 1999). In addition to the defect in seed development, epc-L2 pollen germination efficiency and pollen tube elongation are both reduced compared to the wild-type (Yun & Marchant, unpublished results). This may provide at least a partial explanation for the reduced seed formation by epc-L2. Thus while EPC-L2 is not absolutely essential for fertilization and seed development, it does play a significant role in reproductive processes. The limited expression pattern of EPC-L2 has hampered efforts to identify the biochemical defect in the mutant tissues owing to the difficulty in obtaining sufficient material for analysis.

11.4 Possible activities of the plant GT64 enzymes The only solid evidence for a specific activity for the GT64 enzymes comes from the studies into the human and Drosophila EXT family of proteins which have been shown to function during the synthesis of heparan sulphate (HS). HS is found in the extracellular space as proteoglycans (HSPGs) and is abundant on the surfaces of cells. Synthesis of HS is a complex multistep process which is initiated by formation of a tetrasaccharide comprising glucuronic acid-galactose-galactose-xylose (GlcAβ1–3Galβ1–3Galβ1– 4Xylβ1), which is O-linked to a serine residue in the core protein (Sugahara & Kitagawa 2002). A GlcNAc residue is then added to the reducing end by GlcNAc transferase I (GlcNAc-TI) and elongation then proceeds via the action of GlcNAc-TII and GlcA-TII sequentially adding β1,4-GlcA and α1,4-GlcNAc residues (Lidholt & Lindahl 1992). Studies in Drosophila have shown that the EXT proteins Ttv, Sotv and Botv which function in the synthesis of HSPGs play an important role in the function and distribution of signalling molecules such as Wingless (Wg), Hedgehog (Hh) and (Dpp) during wing development (Bornemann et al. 2004; Han et al. 2004; Perrimon et al. 2004; Takei et al. 2004). Plants are not known to make heparin-related polysaccharides although the pectic rhamnogalacturonan I (RG-I) backbone, like HS, is a heteropolymer comprising a charged (GalUA) and a non-charged residue (Rha). The RG-I backbone is decorated with side chains of 1,4-β-d-galactose or 1,5-α-larabinose residues linked to the rhamnose residues. PACE analysis of the epc1–2 allele found a 50% reduction in the level of β(1,4)-galactan whereas the level of 1,5-α-l-arabinose and the homogalacturonan were unaltered in comparison to the wild type (Bown et al. 2007). The earlier study of Singh

The Plant Glycosyltransferase Family GT64: in Search of a Function

■

297

and coworkers (2005) observed a small but non-significant reduction in galactose in cell wall preparations from seedlings. The plant GT64 enzymes are predicted to have a retaining mechanism whereas the formation of the RG-I 1,4-β-d-galactose side chain would require an inverting mechanism making it unlikely that EPC1 functions directly in its synthesis. Studies into the distribution of the β(1,4)-galactan epitope in Arabidopsis root tissues have found that it is associated with the transition zone between cell division and cell elongation (McCartney et al. 2003). It has been speculated that the reduced β(1,4)-galactan in the epc1–2 allele may be a consequence of a reduced cell elongation caused by increased sensitivity to ABA. Alternatively ABA may act via a mechanism which reduces the β(1,4)-galactan level thereby reducing cell elongation (Bown et al. 2007). Despite these plausible theories, the authors did not detect an alteration in the level of the β(1,4)galactan epitope recognized by the LM5 monoclonal antibody. 11.4.1 Could EPC1 function in AGP synthesis? A further possibility that cannot be discounted is that EPC1 functions either directly or indirectly in the synthesis of arabinogalactan proteins (AGPs), a class of plant proteoglycans which are located at the cell surface and which are also found in gums and secretions (Seifert & Roberts 2007). The AGPs are a highly diverse group of glycoproteins which contain common features such as a core protein which is O-glycosylated with a complex carbohydrate primarily composed of arabinose and galactan. AGPs have been implicated in a large number of developmental processes including cell division, embryo patterning, xylogenesis and pollen tube guidance and selfincompatability (Seifert & Roberts 2007). Given the apparent alteration in ABA sensitivity in the epc1–2 alleles it is interesting to note that Arabidopsis mutants in the AGP30 gene result in a reduced response to ABA in root tissues as well as in ABA-induced seed dormancy (van Hengel & Roberts 2003). The analytical tools currently available to study AGP distribution are limited to β-glucosyl Yariv reagent (Yariv et al. 1962) and antibodies such as JIM13 and CCRC-M7 directed against AGPs. The drawback with these approaches is that they are non-discriminatory and so it is difficult to discern alterations in a specific AGP. Although no difference was detected between wild type and epc1–2 in immunolocalization experiments using JIM13 and CCRC-M7 to detect AGPs, it is possible that a specific AGP structure may be altered in epc1–2 (Bown et al. 2007). It is known that AGPs are abundant in the extracellular matrix of the stylar transmitting tract of a number of plant species (Cheung & Wu 1999) and this has led to the proposal that they may play important roles in pollen tube germination and elongation (Fincher et al. 1983; Cheung et al. 1995). This is supported by the finding that an AGP found in the stylar transmitting tract of Nicotiana alata promotes in vitro pollen tube growth (Wu et al. 2000). The Arabidopsis AtAGP18 gene which encodes a classical AGP is required to

298

■

Plant Polysaccharides

initiate female gametogenesis also highlights the emerging importance of AGPs during reproduction (Acosta-Garcia & Vielle-Calzada 2004). Thus, the defect in pollen tube growth and embryo development observed in the epcL2 mutant could be due to a defect in synthesis of one or more AGPs. 11.4.2 Could EPC1 play a role in FLA biosynthesis? Although two independent studies into EPC1 have yielded some conflicting results, a possible speculative function for the protein arises from the data. Proteins containing fasciclin domains have been shown to function in cell adhesion in a range of organisms including plants, animals, yeast and fungi (Huber & Sumper 1994; Kawamoto et al. 1998; Shi et al. 2003). In plants the fasciclin-like arabinogalactan proteins (FLAs) are a subclass of the arabinogalactan proteins (AGPs) which contain a fasciclin domain. Arabidopsis has at least 21 FLAs, although there may be more as their identification is hampered by the lack of a clear consensus sequence. It is thought that some of the FLAs may function in cell adhesion (Johnson et al. 2003) which is of particular interest given the observed defects in epc1 cell adhesion (Singh et al. 2005). It has been found that at least three of the FLAs (FLA1, 3, 8) are down-regulated in response to ABA (Johnson et al. 2003). Thus in the epc1–2 mutant which shows hypersensitivity to ABA it is possible that one or more of the FLA genes may be down-regulated. If any of these down-regulated FLAs are involved in cell adhesion then it could result in a cell separation phenotype. Differences in growth conditions between the studies by Singh et al. (2005) and Bown et al. (2007) may have induced different stress responses and hence different levels of ABA. Indeed, ABA has been referred to as the stress hormone since it is released in response to stimuli such as drought, salinity and cold. It is therefore possible to speculate that the Singh study may have observed ectopic cell separation as a consequence of higher ABA levels resulting in reduced FLAs. This hypothesis remains to be tested. 11.4.3 EPC1 as a negative regulator of ABA signalling A further possibility has been suggested based on the increased sensitivity to ABA exhibited by the epc1–2 mutant. It is speculated that EPC1 may glycosylate a plasma membrane protein which is involved in the negative regulation of ABA signalling. There is currently no direct evidence to support this, though it is known that at least one pathway which negatively regulates ABA signalling involves a plasma membrane associated ROP10 GTPase (Zheng et al. 2002; Bown et al. 2007). Work carried out to date on epc1, while demonstrating the importance of the protein for normal development, has not provided any firm evidence as to the function of the protein. A yeast-2-hybrid screen using the soluble domain of EPC1 has been carried out to try to identify interacting proteins,

The Plant Glycosyltransferase Family GT64: in Search of a Function

■

299

as it was hypothesized that this approach may yield some clues as to the function of the protein if association with another known protein was detected. However, this approach failed to identify any interacting proteins. There are several possible reasons for this, but removing EPC1 from its normal membrane location may affect its ability to interact with other protein partners. It does not rule out the possibility that interacting proteins do exist. Additionally, attempts to express EPC1 in the yeast Pichia pastoris have proved unsuccessful to date, preventing any in vitro tests of putative function (Yun & Marchant, unpublished results). Alternative expression systems such as baculovirus or Drosophila cells (Liepman et al. 2005) may have to be tried to overcome this stumbling block.

11.5

Concluding remarks

Both EPC1 and EPC-L2 clearly play important but distinct developmental roles in Arabidopsis and the high degree of sequence conservation indicates that this function is likely to be conserved in other plant species. Despite the clear phenotypes exhibited by both epc1 and epc-L2 there is no strong biochemical or functional activity data to help determine the biochemical activities of the proteins. Until this challenging block is overcome, the GT64 family in plants will remain in search of a function.

References Acosta-Garcia, G., Vielle-Calzada, J.P. (2004) A classical arabinogalactan protein is essential for the initiation of female gametogenesis in Arabidopsis. Plant Cell, 16, 2614–2628. Bornemann, D.J., Duncan, J.E., Staatz, W., Selleck, S., Warrior, R. (2004) Abrogation of heparin sulfate synthesis in Drosophila disrupts the Wingless, Hedgehog and Decapentaplegic signaling pathways. Development, 131, 1927–1938. Bowles, D., Isayenkova, J., Lim, E.K., Poppenberger, B. (2005) Glycosyltransferases: managers of small molecules. Current Opinion in Plant Biology, 8, 254–263. Bown, L., Kusaba, S., Goubet, F., et al. (2007) The ectopically parting cells 1–2 (epc1–2) mutant exhibits an exaggerated response to abscisic acid. Journal of Experimental Botany, 58, 1813–1823. Cary, A., Uttamchandani, S.J., Smets, R., Van Onckelen, H.A., Howell, S.H. (2001) Arabidopsis mutants with increased organ regeneration in tissue culture are more competent to respond to hormonal signals. Planta, 213, 700–707. Chaudhury, A.M., Ming, L., Miller, C., Craig, S., Dennis, E.S., Peacock, W.J. (1997) Fertilization-independent seed development in Arabidopsis thaliana. Proceedings National Academy of Sciences of the U S A, 94, 4223–4228. Cheung, A.Y., Wu, H-M., (1999) Arabinogalactan proteins in plant sexual reproduction. Protoplasma, 208, 87–98. Cheung, A.Y., Wang, H., Wu, H. (1995) A floral transmitting tissue-specific glycoprotein attracts pollen tubes and stimulates their growth. Cell, 82, 383–393.

300

■

Plant Polysaccharides

Combet, C., Jambon, M., Deléage, G., Geourjon, C. (2002) Geno3D: automatic comparative molecular modelling of protein. Bioinformatics, 18, 213–214. Fincher, G.B., Stone, B.A., Clarke, A.E. (1983) Arabinogalactan-proteins: structure, biosynthesis and function. Annual Review of Plant Physiology, 34, 7–70. Franco, O.L., Rigden, D.J. (2003) Fold recognition analysis of glycosyltransferase families: further members of structural superfamilies. Glycobiology, 13, 707–712. Gibeaut, M.C., Carpita, N.C. (1991) Tracing cell-wall biogenesis in intact cells and plants – selective turnover and alteration of soluble and cell-wall polysaccharides in grasses. Plant Physiology, 97, 551–561. Grossniklaus, U., Vielle-Calzada, J-P., Hoeppner, M.A., Gagliano, W.B. (1998) Maternal control of embryogenesis by MEDEA, a polycomb group gene in Arabidopsis. Science, 280, 446–450. Han, C., Belenkaya, T.Y., Khodoun, M., Tauchi, M., Lin, X. (2004) Distinct and collaborative roles of Droposphila EXT family proteins in morphogen signalling and gradient formation. Development, 131, 1563–1575. Hauser, M.T., Morikami, A., Benfey, P.N. (1995) Conditional root expansion mutants of Arabidopsis. Development, 121, 1237–1252. Hertzberg, M., Aspeborg, H., Schrader, J., et al. (2001) A transcriptional roadmap to wood formation. Proceedings National Academy of Sciences USA, 98, 14732–14737. Hirotani, M., Kuroda, R., Suzuki, H., Yoshikawa, T. (2000) Cloning and expression of UDP-glucose:flavonoid 7-O-glucosyltransferase from hairy root cultures of Scutellaria baicalensis. Planta, 210, 1006–1013. Hong, S.W., Lee, U., Vierling, E. (2003) Arabidopsis hot mutants define multiple functions required for acclimation to high temperatures. Plant Physiology, 132, 757–767. Huber, O., Sumper, M. (1994) Algal-CAMs: isoforms of a cell adhesion molecule in embryos of the alga Volvox with homology to Drosophila facicline I. EMBO Journal, 13, 4212–4222. Jarvis, P., Dörmann, P., Peto, C.A., Lutes, J., Benning, C., Chory, J. (2000) Galactolipid deficiency and abnormal chloroplast development in the Arabidopsis MGD synthase 1 mutant. Proceedings of the National Academy of Sciences of the U S A, 97, 8175–8179. Johnson, K.L., Jones, B.J., Bacic, A., Schultz, C.J. (2003) The fasciclin-like arabinogalactan proteins of Arabidopsis. A multigene family of putative cell adhesion molecules. Plant Physiology, 133, 1911–1925. Jorasch, P., Warnecke, D.C., Lindner, B., Zähringer, U., Heinz, E. (2000) Novel processive and nonprocessive glycosyltransferases from Staphylococcus aureus and Arabidopsis thaliana synthesize glycoglycerolipids, glycophospholipids, glycosphingolipids and glycosterols. European Journal Biochemistry, 267, 3770–3783. Kawamoto, T., Noshiro, M., Shen, M., et al. (1998) Structural and phylogenetic analyses of RGD-CAP/βig-h3, a fasciclin-like adhesion protein expressed in chick chondrocytes. Biochimica Biophysica Acta, 1395, 288–292. Kim, B-T., Kitagawa, H., Tamura, J-I., et al. (2001) Human tumor suppressor EXT gene family members EXTL1 and EXTL3 encode α1,4-N-acetylglucosaminyltransferases that likely are involved in heparin sulfate/ heparin biosynthesis. Proceedings of the National Academy of Sciences of the U S A, 98, 7176–7181. Kitagawa, H., Shimakawa, H., Sugahara, K. (1999) The tumor suppressor EXTlike gene EXTL2 encodes an alpha 1,4-N-acetylhexosaminyltransferase that transfers N-acetyl-galactosamine and N-acetylglucosamine to the common

The Plant Glycosyltransferase Family GT64: in Search of a Function

■

301

glycosaminoglycan-protein linkage region – the key enzyme for the chain initiation of heparin sulfate. Journal of Biological Chemistry, 274, 13933–13937. Krogh, A., Larsson, B., von Heijne, G., Sonnhammer, E.L.L. (2001) Predicting transmembrane protein topology with a hidden Markov model: Application to complete genomes. Journal of Molecular Biology, 305, 567–580. Kwon, Y., Kim, S.H., Jung, M.S., et al. (2007) Arabidopsis hot2 encodes an endochitinase-like protein that is essential for tolerance to heat, salt and drought stresses. Plant Journal, 49, 184–193. Lane, D.R., Wiedemeier, A., Peng, L., et al. (2001), Temperature-sensitive alleles of RSW2 link the KORRIGAN endo-1,4-β-glucanase to cellulose synthesis and cytokinesis in Arabidopsis. Plant Physiology, 126, 278–288. Lidholt, K., Lindahl, U., (1992) Biosynthesis of heparin. The d-glucancuronosyl- and N-acetyl-d-glucancosaminyltransferase reactions and their relation to polymer modification. Biochemical Journal, 287, 21–29. Liepman, A.H., Wilkerson, C.G., Keegstra, K. (2005) Expression of cellulose synthaselike (Csl) genes in insect cells reveals that CslA family members encode mannan synthases. Proceedings of the National Academy of Sciences of the U S A, 102, 2221–2226. Lind, T., Tufaro, F., McCormick, C., Lindahl, U., Lidholt, K. (1998) The putative tumor supressors EXT1 and EXT2 are glycosyltransferases required for the biosynthesis of heparin sulfate. Journal of Biological Chemistry, 273, 26265– 26268. Mansfeld, A. (2005) Molecular and genetic characterzation of ‘ENHANCER OF POM1’, a new component of root morphogenesis in Arabidopsis thaliana. PhD thesis, Universität für Bodenkultur, Vienna. McCormick, C., Duncan, G., Goutsos, K.T., Tufaro, F. (2000) The putative tumor suppressors EXT1 and EXT2 form a stable complex that accumulates in the Golgi apparatus and catalyzes the synthesis of heparan sulfate. Proceedings of the National Academy of Sciences of the U S A, 97, 668–673. McCartney, L., Steele-King, C.G., Jordan, E., Knox, P. (2003) Cell wall pectic (1–4)-βd-galactan marks the acceleration of cell elongation in the Arabidopsis seedling root meristem. Plant Journal, 33, 447–454. Mouille, G., Robin, S., Lecomte, M., Pagant, S., Höfte, H. (2003)Classi cation and identi cation of Arabidopsis cell wall mutants using Fourier Transform InfraRed (FTIR) micro-spectroscopy. Plant Journal, 35, 393–404. Mølhøj, M., Ulvskov, P., Dal Degan, F. (2001) Characterization of a functional soluble form of a Brassica napus membrane-anchored endo-1,4-β-glucanase heterologously expressed in Pichia pastoris. Plant Physiology, 127, 674–684. Nicol, F., His, I., Jauneau, A., Vernhettes, S., Canut, H., Hofte, H. (1998) A plasma membrane-bound putative endo-1,4-beta d-glucanase is required for normal wall assembly and cell elongation in Arabidopsis. EMBO Journal, 17, 5563–5576. Ohad, N., Margossian, L., Hsu, Y.C., Williams, C., Repeth, P., Fischer, R.L. (1996) A mutation that allows endosperm development without fertilization. Proceedings of the National Academy of Sciences of the U S A, 93, 5319–5324. Ohad, N., Yadegari, R., Margossian, L., et al. (1999) Mutations in FIE, a WD polycomb group gene, allow endosperm development without fertilization. Plant Cell, 11, 407–415. Perrimon, N., Hacker, U., Sanson, B., Tabata, T. (2004) Wingless, hedgehog and heparin sulfate proteoglycans. Development, 131, 2509–2513.

302

■

Plant Polysaccharides

Poppenberger, B., Fujioka, S., Soeno, K., et al. (2005) The UGT73C5 of Arabidopsis thaliana glucosylates brassinosteroids. Proceedings of the National Academy of Sciences of the U S A, 102, 15253–15258. Reed, J.W., Elumalai, R.P., Chory, J. (1998), Suppressors of an Arabidopsis thaliana phyB mutation identify genes that control light signaling and hypocotyl elongation. Genetics, 148, 1295–1310. Rogers, L.A., Dubos, C., Surman, C., et al. (2005) Comparison of lignin deposition in three ectopic lignification mutants. New Phytologist, 168, 123–140. Sato, S., Kato, T., Kakegawa, K., Ishii, T. et al. (2001) Role of the putative membranebound Endo-1,4-β-glucanase KORRIGAN in cell elongation and cellulose synthesis in Arabidopsis thaliana. Plant Cell Physiology, 42, 251–263. Scheible, W-R., Pauly, M. (2004) Glycosyltransferases and cell wall biosynthesis: novel players and insights. Current Opinion Plant Biology, 7, 285–295. Seifert, G.J., Roberts, K. (2007) The biology of arabinogalactan proteins. Annual Review of Plant Biology, 58, 137–161. Schneider, K., Wells, B., Dolan, L., Roberts, K. (1997) Structural and genetic analysis of epidermal cell differentiation in Arabidopsis primary roots. Development, 124, 1789–1798. Senay, C., Lind, T., Muguruma, K., et al. (2000) The EXT1/EXT2 tumor suppressors: catalytic activities and role in heparan sulfate biosynthesis. EMBO Reports, 1, 282–286. Shi, H., Kim, Y.S., Guo, Y., Stevenson, B., Zhu, J-K. (2003) The Arabidopsis SOS5 locus encodes a putative cell surface adhesion protein and is required for normal cell expansion. Plant Cell, 15, 19–32. Singh, S.K., Eland, C., Harholt, J., Scheller, H.V., Marchant, A. (2005) Cell adhesion in Arabidopsis thaliana is mediated by ECTOPICALLY PARTING CELLS 1 – a glycosyltransferase (GT64) related to the animal exostosins. Plant Journal, 43, 384–397. Solomon, L. (1963) Hereditary multiple exostoses. Journal of Bone and Joint Surgery, 45B, 292–304. Sugahara, K., Kitagawa, H. (2002) Heparin and heparan sulfate biosynthesis. IUBMB Life, 54, 163–175. Szerszen, J.B., Szczyglowski, K., Bandurski, R.S. (1994) iaglu, a gene from Zea mays involved in conjugation of growth hormone indole-3-acetic acid. Science, 265, 1699–1701. Szyjanowicz, P.M., McKinnon, I., Taylor, N.G., Gardiner, J., Jarvis, M.C., Turner, S.R. (2004) The irregular xylem 2 mutant is an allele of korrigan that affects the secondary cell wall of Arabidopsis thaliana. Plant Journal, 37, 730–740. Takei, Y., Ozawa, Y., Sato, M., Watanabe, A., Tabata, T. (2004) Three Drosophila EXT genes shape morphogen gradients through synthesis of heparin sulfate proteoglycans. Development, 131, 73–82. Updegraff, D.M. (1969) Semimicro determination of cellulose in biological materials. Analytical Biochemistry, 32, 420–424. Ünligil, U.M., Rini, J.M. (2000) Glycosyltransferase structure and mechanism. Current Opinion Structural Biology, 10, 510–517. van Hengel, A.J., Guzzo, F., van Kammen, A., de Vries, S.C. (1998) Expression pattern of the carrot EP3 endochitinase genes in suspension cultures and in developing seeds. Plant Physiology, 117, 43–53.

The Plant Glycosyltransferase Family GT64: in Search of a Function

■

303

Van Hengel, A.J., Roberts, K., (2003) AtAGP, 30, an arabinogalactan-protein in the cell walls of the primary root, plays a role in root regeneration and seed germination. Plant Journal, 36, 256–270. Wikström, N., Savolainen, V., Chase, M.W. (2001) Evolution of the angiosperms: calibrating the family tree. Proceedings of the Royal Society London Series B, 268, 2211–2220. Wu, H-M., Wong, E., Ogdahl, J., Cheung, A.Y. (2000) A pollen tube growth-promoting arabinogalactan protein from Nicotiana alata is similar to the tobacco TTS protein. Plant Journal, 22, 165–176. Yariv, J, Rapport, M.M., Graf, L. (1962) The interaction of glycosides and saccharides with antibody to the corresponding phenylazo glycosides. Biochemical Journal, 85, 383–388. Zheng, Z-L., Nafisi, M., Tam, A., et al. (2002) Plasma membrane associated ROP10 small GTPase is a specific negative regulator of abscisic acid responses in Arabidopsis. Plant Cell, 14, 2787–2797. Zhong, R., Kays, S.J., Schroeder, B.P., Ye, Z.H. (2002) Mutation of a chitinase-like gene causes ectopic deposition of lignin, aberrant cell shapes, and overproduction of ethylene. Plant Cell, 14, 165–179. Zhong, R., Ripperger, A., Ye, Z.H. (2000) Ectopic deposition of lignin in the pith of stems of two Arabidopsis mutants. Plant Physiology, 123, 59–70. Zuo, J., Niu, Q-W., Nishizawa, N., Wu, Y., Kost, B., Chua, N-H. (2000) KORRIGAN, an Arabidopsis endo-1,4-β-glucanase, localizes to the cell plate by polarized targeting and is essential for cytokinesis. Plant Cell, 12,1137–1152.

Annual Plant Reviews (2011) 41, 305–320 doi: 10.1002/9781444391015.ch12

Chapter 12

GLYCOSYLTRANSFERASES OF THE GT77 FAMILY Bent Larsen Petersen, Kirsten Faber and Peter Ulvskov Department of Plant Biology and Biotechnology, Faculty of Life Sciences, University of Copenhagen Thorvaldsensvej 40, 1871 Frederiksberg, Denmark Manuscript received April 2009

Abstract: GT77 does not include entries from animals, prokaryotes or fungi, but it spans the entire green plant lineage from the simplest marine microalgae to angiosperms. Only two biochemical activities have been demonstrated in the family: the xylosyltransferases involved in rhamnogalacturonan-II synthesis and a cytoplasmic galactosyltransferase from the single slime mould member of the family. With so little biochemical information available, we will instead analyse and use phylogenetic and taxonomic information to characterize the family and occasionally develop working hypotheses regarding function. Some evolutionary contours of the GT77 family emerge, which are interesting in themselves, and lead us to propose which organisms may turn out to be useful models for functional dissection of family GT77 and possibly other families of glycosyltransferases. Keywords: β-arabinosides; extensin; rhamnogalacturonan-II

12.1

Introduction

GT77 is interesting and unique in many ways: it is a plant-only family but for a single slime mould member, and as we will argue, it is likely to become a big family. The gene family is as old as a plant-only family can be. GT77 is not exclusively a cell wall polysaccharide GT family, however. The latest addition to the family consists of several genes from the marine microalgae Ostreococcus tauri (Derelle et al. 2006) and Ostreococcus lucimarinus (Palenik et al. 2007). These Prasinophytes are interesting: tiny eukaryotes, 1–3 μm in

Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

305

306

■

Plant Polysaccharides

diameter, with one plastid, one Golgi apparatus, a mitochondrion but no vacuole. Their origin dates back to the root of the green plant lineage some 1500 million years ago. They are described as wall-less and, as expected, have no genes in plant cell wall signature families like GT47 and GT34 although a single sequence fragment is classified to family GT8. They have quite a few entries in family GT2, but only one of them, CAL57811, has a predicted topology, which is somewhat similar to the glycan synthases of the CesA superfamily. Of all the cell wall related families, GT77 has the highest number of Ostreococcus entries. Which glycans, if any, cover their plasma membrane in the absence of a true wall has not yet been investigated, so not too much can be deduced from the distribution of entries across CAZy families.

12.2 The oldest cell wall We have proposed that two unclassified Chlamydomonas genes belong to family GT77 (Egelund et al. 2007). Although Chlamydomonas is not wallless, its wall is not of the higher plant type with complex polysaccharides as the major components. Rather, its primary wall is mostly made up of glycoproteins (Roberts et al. 1972). The inclusion of Chlamydomonas and Ostreococcus in family GT77 prompted us to ask whether we could identify GT77 candidate genes from ancient organisms with cell walls. The first endosymbiontic event leading to the green plant lineage occurred very early in eukaryote evolution, no later than 1558 million years ago, and the split between red and green algae took place shortly after that (Yoon et al. 2004). Galdieria sulphuraria is a red, thermophilic and acidophilic alga of the Cyanidales order that has been sequenced (Barbier et al. 2005). Even though the wall has not been analysed in detail, it is known that it contains cellulose albeit in relatively small amounts; galactose is a predominant monosaccharide in the wall and the protein content is high (Dr Andreas Weber, personal communication). Genes encoding class III peroxidases have been identified in Galdieria (Oesterhelt et al. 2008) that might be involved in crosslinking of aromatic residues, e.g. present in cell wall glycoproteins. Weber and colleagues have compared the genome of Galdieria to that of a closely related but wall-less member of Cyanidales, Cyanidioschyzon merolae, and observed that the former has genes putatively encoding fucosyland galactosyltransferases that the wall-less family member lacks (Barbier et al. 2005). The Galdieria genome was compared to a database representing all GTs in CAZy. Galdieria sequences for which a hit to GT77 was better than a hit to other GT families and better than non-GT hits to Arabidopsis were analysed by the secondary structure prediction server, Phyre (http:// www.sbg.bio.ic.ac.uk/phyre/; Bennett-Lovsey et al. 2007). Only sequences

Glycosyltransferases of the GT77 Family

■

307

that were predicted to be glycosyltransferases were considered. The proteomes of Selaginella moellendorffii (http://genome.jgi-psf.org/Selmo1/ Selmo1.home.html) and Physcomitrella patens (http://genome.jgi-psf.org/ Phypa1_1/Phypa1_1.home.html) were treated similarly, as was the proteome of Cyanidioschyzon merolae, but the publicly available data (http:// merolae.biol.s.u-tokyo.ac.jp) was at the time of writing not entirely up to date compared to the status of the sequencing effort (Nozaki et al. 2007). Chlamydomonas reinhardtii has also been fully sequenced (http://genome.jgipsf.org/Chlre3/Chlre3.home.html), and a draft proteome prepared.

12.3

Pfam and fold prediction

These suggested additions are rather conservative for the following reason. Most of the GT77-hits that did not pass the requirement that the Phyre prediction server should annotate them as GTs were often annotated as ‘sevenbladed beta-propeller WD40 repeat-like’. The Dictyostelium gene for which biochemical proof of function as an α-galactosyltransferase has been published (Ercan et al. 2006) also receives this annotation. The WD40 family is involved in a wide variety of functions including glycosyl transfer, so one cannot infer that these sequences do not encode glycosyltransferases (personal communication, Dr Lawrence Kelley of the Phyre team). Recognition via a conserved domain and a Pfam HMM-logo would facilitate classification of GT-candidates to family GT77, but the HMM-logo of the nucleotide-diphospho-sugar transferase family, PF03407, did not embrace members of GT77, and neither did PF04488 which carries the annotation ‘glycosyltransferase sugar-binding region containing DXD motif ’. A DXD motif can be identified unambiguously in many but not all GT77s. An enquiry with the Pfam team (Finn et al. 2008) initially did not lead to a revision of PfamA families but to the definition of a new PfamB family PB000588. The motif, if complete, spans from after the signal peptide or membrane anchor and additionally about 300 amino acids into the sequence, thus covering the DXD motif (PB000588 begins in position 1 in the cytoplasmic Dictyostelium galactosyltransferase). PB000588 occurs as more than one shorter fragment in some of the algal species, but is not missing in any of the suggested additions to family GT77 (Table 12.1). PB000588 was a preliminary Pfam construct (vers. 23) which does not exist anymore, so the Pfam prediction engine can no longer assign new sequences to GT77. However, PB000588 disappeared as PF03407 was expanded to include GT77. Family GT77 is rather divergent for a family that comprises only eukaryotes; the minimum identity is 9.1% (gene fragments not considered), and this limit is not pushed any lower with the suggested additions; the slime mould gene is still a clear outlier.

308

■

Plant Polysaccharides

Table 12.1 Key to Fig. 12.1. The numbers in the second column refer to the labels in the phylogenetic tree. The identifier is a locus, if available, an accession number, or if not available, information derived from a fasta header. For Physcomitrella, Selaginella and Chlamydomonas the format is species identifier_protein ID_remainder of the JGI fasta header. A gene name or additional accession number may appear in parenthesis. The presence of PfamB-domains is indicated with asterisks so that *** corresponds to e-values less than 10−100; ** to e-values between 10−100 and 10−25, and * to the range 10−5–10−25. Protein topology is indicated as number of transmembrane helices as predicted by Phobius with S meaning that a signal peptide is predicted rather than membrane anchor. All membrane anchors are N-terminal for proteins with one predicted transmembrane helix

Species Dictyostelium discoideum Arabidopsis thaliana Arabidopsis thaliana Arabidopsis thaliana Oryza sativa Vitis vinifera Physcomitrella patens Physcomitrella patens Selaginella moellendorffi Chlamydomonas reinhardtii Ostreococcus tauri Ostreococcus tauri Chlamydomonas reinhardtii Arabidopsis thaliana Arabidopsis thaliana Arabidopsis thaliana Linum usitatissimum Arabidopsis thaliana Oryza sativa

Number in Fig. 12.1

Identifier

PfamB 588

1

DdDQ340632 (AgtA)

***

0

2

At1g19360 (RRA3)

***

1

3

At1g75110 (RRA2)

***

1

4

At1g75120 (RRA1)

***

1

5 6

*** ***

1 0

***

0

***

1

***

1

**

S

**

S

**

S

**

1

14

Os07g0587100 VITISV_034847 (CAN76537.1) Phypa_222973_estExt_ Genewise1.C_2460030 Phypa_151990_e_ gw1.351.9.1 Selmo_103279_e_ gw1.27.291.1 (SmRRA1) Chlamy_155494_acegs_ kg_5000061 (XP_001699928.1) Ot06g02170 (CAL53476.1) Ot04g02460 (CAL51644.1) Chlamy_137973_estExt_ gwp_1W.C_260040 (XP_001695508.1) At4g01770 (RGXT1)

***

1

15

At4g01750 (RGXT2)

***

2

16

At4g01220

***

1

17

DQ124303 (AAZ94713)

***

2

18

At1g56550 (RGXT3)

***

2

19

Os05g0386900

***

2

7 8 9 10

11 12 13

PfamB 13934

TM

■

309

PfamB 13934

TM

Glycosyltransferases of the GT77 Family Table 12.1 Continued Species Physcimitrella patens Selaginella moellendorffi

Number in Fig. 12.1 20 21

Ostreococcus tauri Chlamydomonas reinhardtii

22

Chlamydomonas reinhardtii

24

Selaginella moellendorffi

25

Physcomitrella patens Medicago truncatula

26

Arabidopsis thaliana Vitis vinifera

28

Oryza sativa Ostreococcus tauri Ostreococcus lucimarinus Ostreococcus lucimarinus Ostreococcus tauri Ostreococcus lucimarinus Ostreococcus tauri Ostreococcus lucimarinus Ostreococcus tauri Chlamydomonas reinhardtii

30 31

Chlamydomonas reinhardtii

23

27

29

32 33 34 35 36 37 38 39

40

Identifier Phypa_117225_e_ gw1.19.80.1 Selmo_87970_e_ gw1.9.211.1 (SmRGXT1) Ot01g01450 (CAL50056.1) Chlamy_56951_estExt_ GenewiseH_1.C_170024 (EDP03806, XP_001692787.1) Chlamy_114491_e_ gwW.10.240.1 (XP_001690605.1) Selmo_127789_e_ gw1.87.147.1 (SmGT77C1) Phypa_123701_e_ gw1.46.33.1 MtrDRAFT_ AC174363g20v1 (ABP03837.1) At2g35610 (xeg113) VITISV_040773 (CAN79199.1) Os03g0586300 Ot03g04870 (CAL53138.1) OSTLU_34304 (ABO95359.1) OSTLU_43023 (ABO99483.1) Ot17g00950 (CAL57822.1) OSTLU_18496 (ABP00281.1) Ot11g01770 (CAL55782.1) OSTLU_88612 (ABO98770.1) Ot02g07030 (CAL52578.1) Chlamy_195312_estExt_ fgenesh2_ pg.C_3510001 (XP_001697686.1) Chlamy_153079_ Chlre2_kg_53000011 (XP_001700329.1)

PfamB 588 ***

0

***

2

*

*

0

**

**

1

**

**

0

***

***

0

***

**

1

***

***

1

***

***

1

***

**

2

*** **

*** **

1 0

**

**

S

**

**

S

*

*

0

*

*

0

**

*

0

**

*

0

*

*

0

*

*

0

*

**

0

310

■

Plant Polysaccharides

Table 12.1 Continued Species

Number in Fig. 12.1

Identifier

Chlamydomonas reinhardtii

41

Chlamydomonas reinhardtii

42

Chlamydomonas reinhardtii

43

Cyanidioschyzon merolae Galderia sulphuraia Oryza sativa

44

Chlamy_153077_ Chlre2_kg_53000009 (XP_001700358.1) Chlamy_179139_2_ pg.C_63000007 (XP_001701267.1) Chlamy_145976_ Chlre2_ kg.scaffold_16000067 (XP_001692532) gnl_CMER_CMC104C

45 46

Vitis vinifera

47

Arabidopsis thaliana Physcomitrella patens Selaginella moellendorffi

48

Physcomitrella patens Galderia sulphuraia Oryza sativa Oryza sativa Oryza sativa Oryza sativa Oryza sativa Oryza sativa Oryza sativa Oryza sativa Arabidopsis thaliana Arabidopsis thaliana Arabidopsis thaliana Arabidopsis thaliana Arabidopsis thaliana Oryza sativa Oryza sativa

PfamB 588

PfamB 13934

TM

*

*

0

*

**

1

*

*

S

**

1

stig_4;Gs07340.1

**

0

P0022D06 (AAU03099.1) VITISV_008966 (CAN65937.1) At1g70630

***

1/S

***

0

***

0

**

0

**

0

**

0

52

Phypa_117686_e_ gw1.21.44.1 Selmo_76356_e_ gw1.1.1371.1 (SmGT77D1) Phypa_218033_estExt_ Genewise1.C_1510037 stig_11;Gs17840.1

*

S

53 54 55 56 57 58 59 60 61

Os03g0849800 Os07g0294800 Os08g0530100 Os03g0849900 Os01g0921000 Os01g0920700 Os01g0920500 Os01g0921100 At5g44820

*** *** *** *** *** *** ** *** ***

1 S S 1 1 S 1 S S

62

At4g19970

***

2

63

At4g15970

***

1

64

At2g02060

**

1

65

At1g14590

***

1

66 67

Os04g0585400 Os02g0686300

*** ***

1 1

49 50

51

■

311

PfamB 13934

TM

Glycosyltransferases of the GT77 Family Table 12.1 Continued Number in Fig. 12.1

Identifier

PfamB 588

Oryza sativa Arabidopsis thaliana Arabidopsis thaliana Arabidopsis thaliana Arabidopsis thaliana Vitis vinifera

68 69

Os03g0129500 At5g40900

*** ***

0 0

70

At1g28710

***

1

71

At1g28700

***

S

72

At1g28695

***

1

73

***

0

Oryza sativa

74

VITISV_031574 (CAN64421.1) Os03g0731800

***

1

Species

12.4

Establishing GT77

Previously, the uniqueness and complexity of the plant cell wall have prompted the cell wall community to speculate to what degree the corresponding biosynthetic machinery was classified within the CAZy database system, where the classification scheme is primarily based on the simple homology relationships to enzymes of other kingdoms with known function. This was investigated using more complex approaches, such as incorporating 2D and 3D structure information in the homology searches and applying this to the Arabidopsis proteome. The search, which was constrained to candidate GTs adopting a typical type II protein structure, resulted in a pool of 27 putative GTs, which were not classified within the CAZy system (Egelund et al. 2004). Two of these accessions, At4g01770 and At4g01750, now designated RGXT1 and RGXT2, respectively, were shown to be xylosyltransferases implicated in the synthesis of pectic rhamnogalacturonan-II (RG-II) (Egelund et al. 2006), and this functional characterization of RGXT1 and RGXT2 gave rise to the formation of family GT77. The type II topology was, and still is (Fasmer Hansen et al. 2009), used for the identification of putative Golgi localized glycosyltransferases, but not all entries are predicted to adopt this topology; see Table 12.1. It should be borne in mind, though, that these predictions are known to be rather far from perfect. Family GT77 contains the known sequences α-1,3-xylosyltransferase (EC 2.4.2.39) and α-1,3-galactosyltransferase (EC 2.4.1.37). Arabinosyltransferase (EC 2.4.2.-), also listed in the header, is still a working hypothesis. The predicted mechanism of transfer is retaining and the sequences adopt the GT-A (SpSA) fold. The two Populus trichocarpa entries are not included in the present analysis. The reason is that Geisler-Lee et al. (2006) reported 16 Populus sequences belonging to GT77. This was based on the first set of gene

312

■

Plant Polysaccharides

models of the poplar genome. Consolidating this information with the version 1.1 of the poplar genome and with the CAZy database falls outside the scope of this review. 12.4.1 The Dictyostelium discoideum (1,3)-α-D-galactosyltranferase GT78/AgtA encodes a soluble UDP-galactose:fucoside α-1,3-galactosyltransferase that modifies the cytoplasmic glycoprotein Skp1 (Ercan et al. 2006). Skp1 is an adaptor-like protein in E3SCF-ubiquitin ligases and other multiprotein complexes of the cytoplasm and nucleus. In Dictyostelium, Skp1 is modified by an unusual pentasaccharide containing the Gal-α1,3-Fuc linkage, which has not been reported previously on glycoconjugates in either prokaryotes or eukaryotes (Ketcham et al. 2004). Skp1 is modified in stepwise fashion just like O-glycosylation pathways in the Golgi apparatus (Young 2004). However, in this case the enzymes all appear to reside in the cytoplasm, in contrast to sequential distribution along subcompartments of the Golgi. 12.4.2 Clade B – rhamnogalacturonan-II biosynthesis Clade B contains the second of the two known activities of GT77, namely the rhamnogalacturonan xylosyltransferase-1 and -2 (RGXT1 and -2) enzymes, which are proposed to function in the biosynthesis of pectic RG-II where they catalyse the transfer of d-xylose to the internal fucose of the RG-II A-chain, yielding an α(1,3)linkage between the xylose and fucose moiety (Egelund et al. 2006). This particular linkage is only known in the complex polysaccharide RGII (see Fig. 3.3). RGXT1 and RGXT2 encode polypeptides of 361 and 367 amino acids (41 175 and 41 799 Dal), respectively, and are 90% identical at the amino acid level (see branching in Fig. 12.1). The two genes are located closely together on chromosome 4, separated by only 6.2 kb in the Arabidopsis genome. Many homologous genes are situated in clusters created by recent, local duplications (Coutinho et al. 2003). At1g56550 and At4g01220 are highly identical to RGXT1 and RGXT2 (68–75% at the amino acid level) and share the same overall type II protein structure containing a N-terminal TMD region and a DxD motif embedded in hydrophobic surroundings situated in the large C-terminal domain holding the catalytic activity (Egelund et al. 2004). Recently, At1g56550 was expressed in Pichia pastoris and shown to be yet another (1,3)-α-dxylosyltransferase having similar acceptor specificities to those reported for the RGXT1 and -2 enzymes (Egelund et al. 2008). If RGXT3, as suggested, is paralogous to RXGT1 and RGXT2, the potential presence of yet another compensating gene product may further rationalize the earlier reported somewhat feeble phenotype of the T-DNA single mutants rgxt1 and rgxt2 (Egelund et al. 2006).

■

Glycosyltransferases of the GT77 Family

313

Dicotyledon embryophytes Grasses Lycophytes Bryophytes Chlorophyte algae

Cla de D

Marine microalgae 52

50

51

Rhodophyte algae

49 48

57 56 54 60 62

59

58

47

53 55

61

46

64 63 45

65 66 67

44

68 69

73 74

43

42

3 4

A de Cla

6 7

5

41 39 40 38

8

37 36

10 9

Clade B

35

4

33

13

93

32 31

30

25

Pfa

12

34

26 27 28

24

11

13

14 20 15 19 16 18 21 17

mB

2

E de Cla

70 71 72

29

eC

22

d Cla 23

1

Figure 12.1 Unrooted phylogeny based on GT77 protein sequences including genes from Chlamydomonas, Galdieria, Cyanidioschyzon, Physcomitrella and Selaginella not formally in CAZy at present. The tree is a neighbour-joining tree prepared using Muscle for alignment (Edgar 2004) and the Phylip programs protdist, neighbor and drawtree (http:// evolution.genetics.washington.edu/phylip.html). Clade stability and hence assignment was based on comparison to a consensus tree prepared from 600 trees using Phylip’s seqboot.

314

■

Plant Polysaccharides

Besides the Arabidopsis accessions, clade B contains a Linum usitatissimum accession (AAZ94713), which has been annotated as the flax (1,3)-α-d-XylT ortholog of RGXT (Paynel et al. 2006) as well as a rice sequence (Os07g0587100). Expressed sequence tags (ESTs), putatively orthologous to members of clade B, have so far only been found in angiosperms (data not shown). RG-II occurs in all vascular plants and its structure is almost invariant in the species so far analysed, which include lycopods, extant representatives of early vascular plants. Our addition of a Selaginella sequence to this clade is thus not surprising. Bryophyte cell walls, on the other hand, contain some of the rare monosaccharides characteristic of RG-II, but have so far not been demonstrated to synthesize the entire polysaccharide. They may instead produce oligosaccharides that are homologous to RG-II (see Chapter 1). One way of gauging how closely the moss sequence is related to the vascular plant sequences is to look not only at the protein sequence, but also at the gene structure. Our underlying assumption is that similarities in the positions, but not the lengths, of introns correlate with evolutionary distance. The splice patterns shown in Fig. 12.2 for clade B clearly tie the vascular plant sequences together, while the Physcomitrella sequence now appears more distant than the amino acid sequence suggests. We thus do not infer the presence of real RG-II in mosses from the existence of a clade B member. The recent characterization of RGXT3 leads us to propose that all clade B members catalyse related reactions. This raises a question about redundancy, however, with Arabidopsis having up to four genes where only one putative orthologue has so far been identified in rice. None of the algae for

Arabidopsis At4g01770 (RGXT1)

Arabidopsis At4g01750 (RGXT2)

Arabidopsis At4g01220

Arabidopsis At1g56550 (RGXT3)

Rice Os05g0386900

Selaginella SmRGXT1

Physcomitrella e_gw1.19.80.1, 117225

Figure 12.2

Gene structures in clade B with exons shown in grey and introns in white.

Glycosyltransferases of the GT77 Family

■

315

which sequence information is available have members in clade B. Charophyacean algae are much closer to higher plants, but it would still be very surprising if they turn out to have genes belonging to clade B. The development of RG-II is believed to be closely tied to the colonization of dry land. 12.4.3 Clade A – the mixed algal and higher plant clade Clade A can be viewed as either a narrow clade comprised solely of higher plant genes, or a broader clade that also includes algal genes. It was a stable clade with only the Chlamydomonas genes added and it remains wellformed after some of the Ostreococcus genes have made their way into the clade. Physcomitrella and Selaginella have two members each in clade A, positioned between the angiosperms and the algae. The catalytic activity of the clade A members is unknown. The Arabidopsis reduced arabinose mutant phenotype in mutants of the RRA1 and RRA2 genes in a fraction that is not easily released from the wall have led us to propose a role in protein glycosylation, notably extensin biosynthesis (Egelund et al. 2007). RRA1 and RRA2 encode polypeptides of 402 and 428 amino acids, respectively. They are predicted to be type II membrane proteins and are 17–23% identical to the accessions in clade B. Similarities are most pronounced in the vicinity of the DXD motif. Corresponding ESTs have been detected in various monocot and dicot plants, but the two genes are too similar to allow assignment of the individual EST sequences to either gene. ESTs thus provide very little information about RRA1 and RRA2 expression profiles. Homologous sequences from organisms outside the plant kingdom, other than the Dictyostelium discoideum (1,3)-α-d-GalT, have not been found. Chlamydomonas is a source of inspiration, with its wall made up of extensin-like glycoproteins. Ostreococcus tauri has a gene product, CAL51877 (annotated as a disintegrin; pfam00200), with similarity to Chlamydomonas cell wall protein GP2 (and putative prolyl 4-hydroxylase homologues to enable protein arabinosylation). It is hard to imagine, however, that Ostreococcus would be described as wall-less if its extracellular matrix comprised extensin-like structures. The Ostreococcus genes in clade A therefore raise more questions than answers until the alga and its GTs are better characterized. Physcomitrella, on the other hand, has a cell wall constituent that reacts with LM1, a monoclonal antibody that recognizes extensin (Møller et al. 2007). The arabinose residues in extensin (Fig. 12.3) are arabinosylfuranoses, which suggests involvement of the mutase discovered recently by Konishi et al. (2007) and previously known as reversibly glycosylateable polypeptide (Dhugga et al. 1991). These proteins also possess autoglycosyltransferase activity and belong to family GT75, a retaining family when they autoglycosylate. Their mutase activity is expected not to change anomeric configuration.

316

■

Plant Polysaccharides

HO O HO OHO HO O HO HO

Figure 12.3 13.

O OHO

O

NH O

O

OH

O

HO HO

Hyp-ara4, the most common short arabinan in extensin. See also Chapter

Arabidopsis At1g75120 (RRA-1)

Arabidopsis At1g75110 (RRA-2)

Arabidopsis At1g19360

Oryza Os07g0587100

Selaginella SmRRA1

Physcomitrella e_gw1.351.9.1, 151990

Chlamydomonas NW_001843705

Figure 12.4 Structures of a subset of genes of clade A; with exons shown in grey and introns in white.

The RGXTs and the Dictyostelium galactosyltransferase form α-1,3 linkages so β-1,3 is the expected linkage from a retaining GT77 arabinosyltransferase. No β-1,3-linked arabinosyl residues are found in extensin, or anywhere else in the plant cell wall that we are aware of. Most of the residues in the short arabinans of extensin are β-1,2-linked, however. It is thus tempting to propose that the members of clade A are β-arabinosyltransferases, but it should be borne in mind that the reduced arabinose phenotype may be an indirect effect rather than the consequence of an arabinosyltransferase knock-out. Chlamydomonas is the major source of inspiration with regard to proposing extensin arabinosylation and it is thus relevant again to look for relatedness in gene structure within the clade. Figure 12.4 shows a subset of the clade. The vascular plants are again closely related, and more distantly related to the moss gene. The Chlamydomonas genes are intronless and may

Glycosyltransferases of the GT77 Family

■

317

be proposed as ancestral homologues to clade A, rather than bona fide members of it. However, strong evidence for the involvement of GT77 in extensin arabinoside biosynthesis has recently been presented for the Arabidopsis member of clade C; see below. 12.4.4 Clade C and PfamB 13934 The demarcation of clade C in Fig. 12.1 is the natural expansion of the clade proposed by Egelund et al. (2007) with Vitis vinifera, Medicago truncatula, Selaginella moellendorffii and Physcomitrella patens. All members are single genes. Mutants of the single higher plant accessions are therefore expected a priori to have a higher chance of conferring phenotypes. Interestingly, using an enzyme challenge screen (see Chapter 2), a T-DNA knock-out mutant of At2g35610 was found in a screen (Pauly et al. 2007; Gille et al. 2009) to have a structural change in xyloglucan as well as overall wall xylose levels. These changes were, however, hypothesized to be indirect effect of a lesion in extensin, since arabinose leves are reduced in fractions enriched in extensin . The two neighbouring Chlamydomonas genes again point to extensin arabinosylation, but Gille et al. also present rather direct evidence. Linkage analysis of a Ba(OH)2-extract of At2g35610 knock-out mutant seedlings shows a 38% reduction of 2-linked arabinose and 80% reduction of 3-linked arabinose. The simplest explanation would be that At2g35610 encodes the arabinosyltransferase that elongates Hyp-Ara2 to Hyp-Ara3. The fourth arabinosyl residue is linked α-1,3, so a pronounced reduction in 3-linked arabinose will be observed if its arabinosyltransferase (inverting, hence not a GT77) has a strong preference for Hyp-Ara3. The sequences, with one exception, in this part of the phylogenetic tree feature another PfamB motif, PB013934. The single rice sequence in clade C sensu strictu appears to be an exception. The PfamB 13934 reappears in Os03g0586300, however, if the longer gene model, ABF97349.1, is taken to be the correct model. PfamB 13934 is not annotated with a catalytic function so presently it merely suggests that clade C sensu lato is a meaningful concept. 12.4.5 Clade D Like clade C, clade D contains what might appear to be orthologues: one gene of each of the higher plant species plus Selaginella and possibly Physcomitrella. Mutants of theses single accessions may therefore be expected to confer phenotypes. 12.4.6 Clade E – the youngest clade This clade has only angiosperm members and may have expanded during speciation of the flowering plant families. There are no current hypotheses for the function of any of these genes, but Mitchell et al. (2007) conducted an ingenious comparative expression survey of some of the genes in grasses

318

■

Plant Polysaccharides

versus dicots and found that this group in particular appeared to be more highly expressed in grasses. The rice sequences 53 and 56 contributed significantly to this result, using 61, 62, 64 and 65 as Arabidopsis genes to compare with. Their method compared normalized frequencies of estsequences not only from rice and Arabidopsis but also from other cereals and dicots; see supplementary Table 1 of Mitchell et al. (2007) . This may mean that grasses have recruited these genes for the synthesis of carbohydrates that exist only in grasses, or are more prevalent there. We shall not attempt to propose which polysaccharide epitopes this may be on this information alone.

12.5 Discussion Family GT77 is enigmatic, particularly if it turns out that the structure predictions by Phyre are correct, meaning that the family comprises GTs that feature the GT-A fold and other GTs with a different secondary structure. This may result in a split of the family. The Dictyostelium α-galactosyltransferase is involved in the glycosylation of Skp1, a cytoplasmic protein involved in polyubiquination (Ketcham et al. 2004). The glycosylation appears to occur entirely in the cytoplasm and is initiated by a soluble prolyl 4-hydroxylase followed by transfer of an Nacetylglucosaminyl residue. The latter is catalysed by GnT51, which belong to family GT60 (West 2003), a family of GTs from prokaryotes and lower eukaryotes, including Ostreococcus, but with no higher plant members. It is thus unlikely that the cytoplasmic glycosylation machinery exists in higher plants, although a diverse family of Skp1 genes does (Kong et al. 2001). The slime moulds diverged from the eukaryote lineage after the plants, but before the fungi. There is no a priori reason why cytoplasmic glycosylation should have been lost from the entire plant lineage. It may well exist in lower plants whose GTs have made or will make their way into GT77 (and GT60). We have looked at other evolutionarily old algae as we are of the opinion that these algae with their smaller genomes and far fewer GTs compared to higher plants will aid functional assignment of GTs in the family. So will comparative studies of Charophytes, the group of algae closest to higher plants, for which a number of genome sequencing projects are well under way. Not nearly enough is known about algal polysaccharide structures to fully benefit from the genomic information, however, and that applies not least to the wall-less species.

Acknowledgments Dr Andreas Weber of the University of Düsseldorf kindly provided the draft G. sulphuraria proteome. Support for this work is acknowledged from two grants from the Danish Agency for Science Technology and Innovation;

Glycosyltransferases of the GT77 Family

■

319

grant no. 2101-07-0071: ‘Glyco-engineering Plant Cells for Controlled Human O-Glycosylation: Novel Host Cells for Recombinant Production’ from DSF and grant 274-09-0314: ‘Evolution-guided short-cuts to gene discovery for improved biofuel crops’ from FTP; and by a grant from the Villum Kann Rasmussen Foundation to the ProActive Plant Centre, Denmark.

References Barbier, G., Oesterhelt, C., Larson, M.D., et al. (2005) Comparative genomics of two closely related unicellular thermo-acidophilic red algae, Galdieria sulphuraria and Cyanidioschyzon merolae, reveals the molecular basis of the metabolic flexibility of Galdieria sulphuraria and significant differences in carbohydrate metabolism of both algae. Plant Physiology, 137(2), 460–474. Bennett-Lovsey, R.M., Herbert, A.D., Sternberg, M.J., Kelley, L.A. (2007) Exploring the extremes of sequence/structure space with ensemble fold recognition in the program Phyre. Proteins, 70(3), 611–625. Coutinho, P.M., Stam, M., Blanc, E., Henrissat, B. (2003) Why are there so many carbohydrate-active enzyme-related genes in plants? Trends in Plant Science, 8, 563–565. Derelle, E., Ferraz, C., Rombauts, S., et al. (2006) Genome analysis of the smallest free-living eukaryote Ostreococcus tauri unveils many unique features. Proceedings of the National Academy of Sciences of the U S A, 103(31), 11647–11652. Dhugga, K.S., Ulvskov, P., Gallagher, S.R., Ray, P.M. (1991) Plant polypeptides reversibly glycosylated by UDP-glucose – possible components of golgi betaglucan synthase in pea cells. Journal of Biological Chemistry, 266(32), 21977–21984. Edgar, R.C. (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research, 32(5), 1792–1797. Egelund, J., Skjodt, M., Geshi, N., Ulvskov, P., Petersen, B.L (2004) A complementary bioinformatic approach to identify potential plant cell wall glycosyltransferase encoding genes. Plant Physiology, 136, 2609–2620. Egelund, J., Petersen, B.L., Motawia, J.S., et al. (2006) Biosynthesis of pectic rhamnogalacturonan, I.I.: Molecular cloning, characterization of Golgi-localized α(1→3)-xylosyltransferase encoded by the RGXT1 and RGXT2 genes of Arabidopsis thaliana. Plant Cell, 18, 2593–2607. Egelund, J., Obel, N., Ulvskov, P., Geshi, N., Pauly, M., Bacic A., Petersen, B.L. (2007) Molecular characterization of two Arabidopsis thaliana glycosyltransferase mutants, rra-1 and -2, which have a reduced content of arabinose in a polymer tightly associated with the cellulose residue. Plant Molecular Biology, 64, 439–451. Egelund, J., Damager, I., Faber, K., Olsen, C.E., Ulvskov, P., Petersen, B.L. (2008) Functional characterisation of a putative rhamnogalacturonan-II specific xylosyltransferase. FEBS Letters, 582, 3217–3222. Ercan, A., Panico, M., Sutton-Smith, M., et al. (2006) Molecular characterization of a novel UDP-galactose:fucoside α-3-galactosyltransferase that modifies Skp1 in the cytoplasm of Dictyostelium. Journal of Biological Chemistry, 281, 12713–12721. Fasmer Hansen, S., Bettler, E., Wimmerov, M., Imberty, A., Lerouxel, O., Breton, B. (2009) Combination of several bioinformatic approaches for the identification of new putative glycosyltransferases in Arabidopsis. Journal of Proteome Research, 8, 743–753.

320

■

Plant Polysaccharides

Finn, R.D., Tate, J., Mistry, J., et al. (2008) The Pfam protein families database. Nucleic Acids Research, 36, D281–D288. Geisler-Lee, J., Geisler, M., Coutinho, P.M., et al. (2006) Poplar carbohydrate-active enzymes. Gene identification and expression analyses. Plant Physiology, 140, 946–962. Gille, S., Hânsel, U., Ziemann, M., Pauly, M. (2009) Identification of plant cell wall mutants by means of a forward chemical genetic approach using hydrolases. Proceedings of the National Academy of Sciences of the U S A, 106, 14699–14704. Ketcham, C., Wang, F., Fisher, S.Z., et al. (2004) Specificity of a soluble UDP-galactose: fucoside alpha1,3-galactosyltransferase that modifies the cytoplasmic glycoprotein Skp1 in Dictyostelium. Journal of Biological Chemistry, 279(28), 29050–29059. Kong, H., Leebens-Mack, J., Ni, W., dePamphilis, C.W., Ma, H. (2001) Highly heterogenous rates of evolution in the SKP1 gene family in plants and animals: functional and evolutionary implications. Molecular Biology and Evolution, 21(1), 117–128. Konishi, T., Takeda, T., Miyazaki, Y., et al. (2007) A plant mutase that interconverts UDP-arabinofuranose and UDP-arabinopyranose. Glycobiology, 17(3), 345–354. Mitchell, R.A.C., Dupree, P., Shewry, P.R. (2007) A novel bioinformatics approach identifies candidate genes for the synthesis and feruloylation of arabinoxylan. Plant Physiology, 144, 43–53. Møller, I., Sørensen, I., Bernal, A.J., et al. (2007) High-throughput mapping of cellwall polymers within and between plants using novel microarrays. Plant Journal, 50(6), 1118–1128. Nozaki, H., Takano, H., Misumi, O., et al. (2007) A 100%-complete sequence reveals unusually simple genomic features in the hot-spring red alga Cyanidioschyzon merolae. BMC Biology, 5, 28 doi:10.1186/1741-7007-5-28. Oesterhelt C, Vogelbein S, Shrestha, R.P., Stanke M, Weber, A.P.M. (2008) The genome of the thermoacidophilic red microalga Galdieria sulphuraria encodes a small family of secreted class III peroxidases that might be involved in cell wall modification. Planta, 227(2), 353–362. Palenik, B., Grimwood, J., Aerts, A., et al. (2007) The tiny eukaryote Ostreococcus provides genomic insights into the paradox of plankton speciation. Proceedings of the National Academy of Sciences of the U S A, 104(18), 7705–7710. Pauly, M., Gille, S., Haensel, U., Ziemann, M. (2007) Identification and characterization of novel A. thaliana Xyloglucan Mutants using a xyloglucanase screen. Physiologia Plantarum, 130(4), abstract 20. Paynel, F., Chevalier, V., Bruyant, P., et al. (2006) Direct submission of AAZ94713 to NCBI. Roberts, K., Hills, G.J., Gurney-Smith, M. (1972) Structure, composition and morphogenesis of cell wall of Chlamydomonas reinhardi. I ultrastructure and preliminary chemical analysis. Journal of Ultrastructure Research, 40(5–6), 599–613. West, C.M. (2003) Evolutionary and functional implications of the complex glycosylation of Skp1, a cytoplasmic/nuclear glycoprotein associated with polyubiquitination. Cellular and Molecular Life Sciences, 60(2), 229–240. Yoon, H.S., Hackett, J.D., Ciniglia C., Pinto, G., Bhattacharya, D. (2004) A molecular timeline for the origin of photosynthetic eukaryotes. Molecular Biology and Evolution, 21(5), 809–818. Young, W.W. Jr. (2004) Organization of Golgi glycosyltransferases in membranes: Complexity via complexes. Journal of Membrane Biology, 198, 1–13.

Annual Plant Reviews (2011) 41, 321–342 doi: 10.1002/9781444391015.ch13

Chapter 13

HYDROXYPROLINE-RICH GLYCOPROTEINS: FORM AND FUNCTION Marcia J. Kieliszewski1, Derek T.A. Lamport2, Li Tan1 and Maura C. Cannon3 1

Department of Chemistry and Biochemistry , Ohio University, Athens, OH 45701, USA Biology and Environmental Science, University of Sussex, Falmer, Brighton, BN1 9RH, UK 3 Department of Biochemistry and Molecular Biology, 913 Lederle Graduate Research Tower, University of Massachusetts, Amherst, MA 01003, USA 2

Manuscript received August 2008

Abstract: The extracellular matrix is a morphogenetic substrate based on scaffold proteins comprising hydroxyproline-rich glycoproteins (HRGPs): collagens in animals and extensins in plants. Repetitive peptide motifs and glycomodules define the extensin superfamily, richly diverse in both structure and function. There are two major groups: highly basic classical extensins form insoluble covalent networks, while the classical arabinogalactan proteins (AGPs) are acidic and easily solubilized. In both extensins and AGPs the primary sequence encodes extensive post-translational modifications: addition of a GPI anchor to AGPs, crosslinking of extensins via isodityrosine (Idt) and diisodityrosine (di-Idt), and hydroxylation of proline residues where subsequent O-Hyp glycosylation defines the glycomodule. Non-contiguous Hyp directs addition of small, compact, beadlike arabinogalactan polysaccharides that coat the plasma membrane and hypothetically also act as a pectic plasticizer. However, in the basic extensins blocks of contiguous Hyp direct addition of small oligoarabinosides. The resulting hydrophilic glycomodules alternate with short hydrophobic Idt motifs to form self-assembling amphiphiles that react with acidic pectin. Thus an extensin pectate coacervate may be a template for the formation of the cell plate. Evolution of the cell plate and the acquisition of turgor required for land plant evolution were thus crucially dependent on HRGPs. Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

321

322

■

Plant Polysaccharides

Keywords: arabinogalactan proteins; diisodityrosine; extensins; hydroxyprolinerich glycoproteins; isodityrosine; O-glycosylation

13.1 Introduction 13.1.1 Background Intriguing similarities between the extracellular matrices of plants and animals include structural glycoproteins as essential scaffolding components. Some of these glycoproteins, such as the collagens of animals and hydroxproline-rich glycoproteins (HRGPs) of plants, share crucial functions: they increase tensile strength (Shirsat et al. 1996) and contribute to tissue integrity. There the similarities diverge; self-assembling HRGPs, notably extensins, template the new crosswall at cytokinesis (Cannon et al. 2008) whereas animals form a cleavage furrow. Other components such as the arabinogalactan-proteins (AGPs), richly endowed with acidic polysaccharides, resemble animal proteoglycans, although the amino acid compositions, glycosylation details and biological roles differ. 13.1.2 Definitions In an earlier review (Kieliszewski & Lamport 1994) we proposed that HRGPs, although highly diverse, actually represent a single family that comprises a phylogenetic continuum ranging from basic, minimally glycosylated PRPs (proline-rich proteins) and nodulins, to the acidic, highly glycosylated AGPs. Only recently have we come to appreciate that this definition depends on quite well-defined peptide modules and glycomodules with which nature builds the classically defined HRGPs. Thus in the broadest sense HRGP family members also include a wide range of proteins of varying functions that contain one or more HRGP (glyco)modules. Historically, HRGP family members were named largely based on the repetitive peptide motifs and their post-translational modifications: blocks of arabinosylated contiguous Hyp, particularly the Ser-Hyp4 motif (Lamport 1973) in the extensins – structural glycoproteins putatively covalently crosslinked through peptide motifs. Thus Ser-Hyp4 glycomotifs and ValTyr-Lys and Tyr-X-Tyr crosslinking motifs defined extensins, whereas related Pro-Hyp-Val-Tyr-Lys motifs and variants thereof defined the PRPs. AGPs were defined by the presence of arabinogalactans (Aspinall et al. 1969; Jermyn & Yeow 1975), later identified as Hyp-arabinogalactans (Hyp-AGs) (Lamport 1977; Pope 1977) and identified by their serendipitous reactivity with the β-Yariv reagent (Yariv et al. 1962; Yariv et al. 1967; Jermyn & Yeow 1975). HRGP chimeras possess both classical HRGP motifs and non-HRGP motifs (Johnson et al. 2003b); solanaceous lectins were the first identified with repetitive arabinosylated Ser-Hyp4 motifs fused to chitin-binding domains (Ashford et al. 1982; Kieliszewski et al. 1994; Van Damme et al.

Hydroxyproline-rich Glycoproteins: Form and Function

■

323

2004). More recent work identifies a plethora of GPI-anchored cell surface proteins that are chimeric HRGPs (Borner et al. 2003; Johnson et al. 2003a). On the other hand HRGP hybrids are classical HRGPs with mixed motifs, e.g. gum arabic glycoprotein (Qi et al. 1991; Goodrum et al. 2000; Johnson et al. 2003b) which contains extensin Ser-Hyp3–4 modules mixed with X-HypX-Hyp-X-Hyp (AGP motifs). Classical HRGPs represent extremes of a glycosylation and periodicity continuum: extensins, primarily defined by Ser-Hyp4 glycomodules and covalent crosslinking peptide motifs; classical AGPs contain AG-glycomodules, arabinosylated X-Hyp-Hyp glycomodules and GPI-anchors; finally the classical PRPs, which are defined by repetitive, ProHyp-Val-Tyr-Lys motifs (Keller 1993; Wilson et al. 1994; Fowler et al. 1999; Kauss et al. 2003) and variants, are often arabinosylated. Ultimately biological function will define particular HRGP types most likely consistent with our current definitions, given that structure and function are so closely related. HRGP nomenclature is evolving (Johnson et al. 2003b), particularly in light of new data gained by genome sequencing projects.

13.2

Post-translational modifications

Arguably the most abundant glycoproteins in nature, HRGPs are also the most extensively post-translationally modified. Notable modifications are the hydroxylation of specific proline residues by direct fixation of molecular oxygen (Lamport 1963; Lamport 1965); glycosylation (Lamport 1967; Pope 1977) of some but not all serine and hydroxyproline residues (Lamport et al. 1973; Cho & Chrispeels 1976; Allen et al. 1978); signal peptide processing; covalent intra- and intermolecular crosslinking of peptide motifs; and, for some HRGPs, C-terminal GPI anchor addition. Simple sequence-based rules direct post-translational modifications and these are becoming more sharply formulated (Xu et al. 2008). 13.2.1 Rules for proline hydroxylation Predicting proline hydroxylation is essential to algorithms that identify O-Hyp glycosylation sites. Early work suggested that hydroxylation was largely dependent on a PPII substrate conformation (Sadava & Chrispeels 1971). However, subsequent HRGP sequence data were more consistent with specific peptide motifs recognized by multiple prolyl hydroxylases (Kieliszewski & Lamport 1994) and confirmed by the identification of multiple prolyl hydroxylases in Arabidopsis (Hieta & Myllyharju 2002; Tiainen et al. 2005). More recently, a proposed code based on hydroxylation of a single Pro residue in vacuolar sporamin correctly identifies many arabinogalactosylation sites in AGPs (Shimizu et al. 2005). However, this algorithm underestimates the hydroxylation of PRPs, extensins, dipeptidyl Hyp in AGPs, and hydroxylation of the AGP analogues [VP]n → [VO]n and [TP]n → [TO]n (Tan et al. 2003) and the common AGP sequences TTP and TAP.

324

■

Plant Polysaccharides

Table 13.1 Observed hydroxylation of dipeptidyl proline sequences in angiosperm HRGPs

Not hydroxylated Asn-Pro Asp-Pro Cys-Pro Gln-Pro Leu-Pro Ile-Pro

Lys-Pro Met-Pro Phe-Pro Tyr-Pro Trp-Pro

Species-specific hydroxylation

Consistently hydroxylated

Inconsistently hydroxylated

Glu-Pro Gly-Pro His-Pro

Ala-Pro Ser-Pro Pro-Pro

Thr-Pro Val-Pro

Amino acid sequences from characterized HRGPs and Hyp-containing proteins and peptides (Table 13.1) show that Pro generally remains nonhydroxylated when preceded by Lys, Tyr, Leu, Ile, His, Cys, Gln, Glu, Met and Phe, presumably because bulky sidechains sterically hinder the hydroxylases. However, recognition of some sequences by prolyl hydroxylase is undoubtedly species specific. For example, His-Pro is not hydroxylated in tomato extensins (Smith et al. 1986) whereas it is in a maize HRGP (Kieliszewski et al. 1990). Likewise, Gly-Pro is not hydroxylated in Acacia (Goodrum et al. 2000) or tobacco (Shpak et al. 1999), but is in Arabidopsis (Schultz et al. 2004). Although bulky residues preceding Pro often restrict hydroxylation, smaller residues such as Thr, Val and especially Pro, Ser and Ala allow it, the exception being Gly (Shpak et al. 1999; Shpak et al. 2001; Shimizu et al. 2005), hence the successful identification of AGPs based on a biased amino acid composition (Schultz et al. 2002) and the extensive hydroxylation of poly-Pro blocks in extensins. 13.2.2 O-Hyp-glycosylation codes – the Hyp contiguity hypothesis Glycans attached to HRGP polypeptides occur as small, linear arabinooligosaccharides linked to Hyp (Lamport 1967) by the unusual β-linked anomeric configuration (Akiyama et al. 1980). Much larger, branched arabinogalactan polysaccharides (AGs) are linked through galactosyl-O-Hyp (Fincher et al. 1974; Lamport 1977; Pope 1977). The Hyp glycosylation profiles of naturally occurring extensins, AGPs, PRPs, HRGP hybrids and chimeras first suggested the existence of a glycosylation code. Thus, the Hyp contiguity hypothesis predicts O-Hyp glycosylation of proteins targeted for secretion based on their peptide sequence, blocks of contiguous Hyp directing oligoarabinoside addition with glycosylation site occupancy increasing with the number of contiguous Hyp residues (Kieliszewski et al. 1992; Kieliszewski & Lamport 1994; Shpak et al. 2001; Zhao et al. 2002; Tan et al.

Hydroxyproline-rich Glycoproteins: Form and Function

■

325

2003). The hypothesis further predicts that non-contiguous Hyp residues are sites of arabinogalactan polysaccharide attachment. We refer to glycosylated structural units as ‘glycomodules’, more specifically as arabinoglycomodules (Ara-module) and arabinogalactan glycomodules (AG-module). These glycomodules define most of the hydrophilic interactive surface and are thus major determinants of HRGP function. Thus, with the exception of the lightly glycosylated PRPs, they influence aspects of wall self-assembly, network and barrier formation, differentiation, and signalling. HRGPs share similar features. Classical AGPs in particular contain both contiguous Hyp residues that are arabinosylated and non-contiguous Hyp residues that are arabinogalactosylated within the same molecule. With such extensive glycosylation and Yariv reactivity it is possible to prepare mixtures consisting exclusively of classical AGPs (Lamport 1970) although rarely possible to purify a single AGP for biochemical analysis (Qi et al. 1991; Goodrum et al. 2000). However, expression of AGPs as green fluorescent protein (GFP) fusion glycoproteins removes that limitation (Shpak et al. 1999). Initially, we designed simple neoHRGPs consisting of single glycomodule repeats with a hydrophobic GFP tag. This greatly facilitated subsequent chromatographic purification of AGPs and unambiguous tests of the Hyp contiguity hypothesis in cell cultures. Many tissues of intact plants (Estevez et al. 2006) are Hyp-poor and thus likely do not express a full complement of HRGPs. Cultures of tobacco (Zhao et al. 2002) and Arabidopsis (Xu et al. 2008) are relatively Hyp-rich and therefore express a range of corresponding posttranslational glycosyltransferases. This enabled us to elucidate the sequence determinants of Hyp glycosylation and the first complete structures of Hyparabinogalactan polysaccharides (Tan et al. 2004; Xu et al. 2007) (Fig. 13.1). Early work involving HRGPs expressed through synthetic genes posed the question: To what extent does the Hyp contiguity hypothesis predict O-Hyp glycosylation sites in known glycoproteins? Thus tomato LeAGP-1 expressed as a GFP fusion in tobacco (Zhao et al. 2002) and subsequently in Arabidopsis cell cultures (Xu et al. 2008) yielded Hyp-glycosylation profiles entirely consistent with the predictions. Recently we examined a protein secreted by Arabidopsis cell cultures (Fig. 13.2). This putative phytocyanin (At4g31840) is a chimeric AGP, judging from six clustered non-contiguous Pro residues and the amino acid composition surrounding these Pro residues. The N-terminal region is probably folded as the oxidizing environment of the endoplasmic reticulum/Golgi promotes disulfide bond formation of the two CySH residues and the C-terminus encodes a GPI anchor. Based on the Pro hydroxylation/Hyp glycosylation predictions and the likelihood that Pro residues in folded regions of the polypeptide are not hydroxylated, we predicted six Hyp residues with a maximum of six arabinogalactan polysaccharides and no arabinosylated Hyp. Biochemical characterization of purified At4g31840 corroborated those predictions and its N-terminal sequence, NEVTVGGKSGDWKIPPSSSF, was

1 3 1 3

1 6

Ara 1 3

1 5

1 3 1 4

1 3

3

Rha

1 5

1 6

1 6

1

1 6

1 4 1 6

1 3

1 3

1 6

1 3

GlcUA

Gal

1 3

1

Hyp

6

1 4

Figure 13.1 Suggested structure of the 22-residue arabinogalactan AG-1 isolated from a human interferon-AGP fusion protein described in Xu et al. (2007). Characterized Nicotiana type II arabinogalactan polysaccharides possess repetitive subunits. Each complete 15-residue subunit consists of a main chain of β-1,3-galactose trisaccharide with two sidechains linked to the first and second Gal residue of the trigalactosyl block, numbering from the reducing end of the block. The sidechains, when fully elaborated, contain β-D-galactose, three α-L-arabinose residues, β-D-glucuronic acid, and α-Lrhamnose. A β-1,6-link connects consecutive galactose trisaccharide mainchain blocks with respective sidechains; however, the second subunit of AG-1 was incomplete and contained only seven sugar residues: 5 Gal residues and 2 Ara residues. From Xu, J., Tan, L., Goodrum, K.J., Kieliszewksi, M.J. (2007) High yields and extended serum half-life of human interferon α2b expressed in tobacco cells as Arabinogalactanprotein fusion. Biotechnology and Bioengineering, 97(5), 997–1008. With permission from John Wiley & Sons.

1

MASSSLLVTI

FLCISVFFFS

SVNA↓NEVTVG

GKSGDWKIPP

61

IVFKYEAGKD SVLQVTREAY

EKCNTTSPKA

SYTDGNTKVK

121

QKLRLVVITP RNSAFSPGPS PSEFDGPAVA

SSSFSFNEWA

QKARFKVGDF

LDQAGPVYFV SGTEGHCQKG

PTS↑GAAKLAG GFSVVFGLVL

GLWAFFF

Figure 13.2 The protein sequence of At4g31840 deduced from the gene. Mature At4g31840 is putatively a 129-residue early nodulin-like phytocyanin. Emboldened Pro residues at the C-terminus are predicted sites of hydroxylation/glycosylation. The symbol (↓) denotes the signal sequence cleavage site and (↑) indicates the predicted addition site for a GPI anchor.

Hydroxyproline-rich Glycoproteins: Form and Function

■

327

Table 13.2 Amino acid composition of the phytocyanin-AGP chimera encoded by At4g31840 Amino acid

Mol%

D E O S G H R T A P Y V M C I L F K W

9.7 9.3 4.5 12.5 9.8 1.1 2.7 6.3 8.0 4.5 2.9 7.3 0.4 n.d. 3.1 4.8 5.5 7.6 n.d.

n.d., not determined.

consistent with the signal peptide cleavage site predicted by Signal P (Bendtsen et al. 2004). Amino acid analysis of the mature GPI-anchored protein (Table 13.2) confirmed that At4g31840 contained six Hyp residues. Quantitative Hyp glycoside profiles indicated 92% of the total Hyp in At4g31840 was arabinogalactosylated, corresponding to 5.5 Hyp residues, with 8% (0.48 Hyp residues) remaining non-glycosylated. 13.2.3 Glycosylation enhances secretion of designer glycoproteins Application of the Hyp contiguity hypothesis allowed the design of glycoprotein-AG fusions that were glycosylated as predicted. For example, human interferon-(Ser-Hyp)10and20 and other therapeutic proteins expressed as AG fusions were reliably glycosylated (Xu et al. 2007). Desirably, they also had a greatly prolonged serum half-life. Such fusion glycoproteins were also secreted with greatly enhanced yields compared to similarly targeted nonglycosylated proteins (Xu et al. 2008). This is consistent with recent evidence of two secretory pathways, one involving bulk export of polysaccharides (Leucci et al. 2007). Thus the Golgi may act as a sorting station where glycosylation fast-tracks products for export.

328

■

Plant Polysaccharides

13.2.4 Glycosylation anomalies Although the Hyp contiguity hypothesis makes robust predictions, there are possible exceptions; it also identifies experimental anomalies that might otherwise remain undetected. For example, it was earlier noted that carrot DcAGP-1 lacked Hyp, according to amino acid analyses, but contained arabinogalactan polysaccharides – a suprising instance of a Hyp-poor AGP (Baldwin et al. 1993). However, subsequent identification of the DcAGP-1 gene (Baldwin et al. 2001) allowed its predicted amino acid composition to be compared to that determined experimentally for Hyp-poor DcAGP-1 (Table 13.3); it is clear that the biochemically characterized protein is not the same protein that is encoded by the DcAGP-1 gene. The genomic sequence of DcAGP-1 indicates a protein containing several potential AG glycosylation sites: one SP motif, four TP and twelve AP motifs which are all candidates for hydroxylation and arabinogalactosylation according to the Hyp contiguity hypothesis. Thus it is more likely that DcAGP-1, like a typical AGP, is arabinogalactosylated through Hyp residues rather than through Thr or Ser as first claimed.

Table 13.3 DcAGP1 isolated protein and genomic compositions compared. Major discrepancies are shown in bold type Isolated DcAGP1a Amino acid Asx Glx Ser Thr Gly Ala Arg Pro Val Met Ile Leu Phe Cys Lys His Tyr Hyp

DcAGP1 DNAb

Mol% 10.7 9.2 7.4 6.1 11.3 12.2 4.4 5.0 6.8 0.0 4.6 8.7 3.7 0.0 5.4 1.9 2.8 n.d.

3.8 4.6 4.1 9.6 6.0 10.1 1.4 17.4 9.6 0.5 1.4 6.4 2.8 2.8 11.0 6.0 2.8 n.a.

Asx, Asn or Asp; Glx, Gln or Glu; n.a., not applicable; n.d., none detected. a Baldwin et al. (1993). b Baldwin et al. (2001).

Hydroxyproline-rich Glycoproteins: Form and Function

■

329

Non-contiguous Hyp residues in clustered SP motifs of PRPs including some Arabidopsis extensins seem to be exceptions to general arabinogalactosylation. We speculate that flanking sequences containing basic or aromatic residues may suppress arabinogalactan polysaccharide addition. The Hyp glycosylation hypothesis raises many other questions: Does tissuespecific or cell-specific Hyp glycosylation occur? How much variation occurs between species? Is the Hyp glycosylation code similar in lower orders? This includes the non-flowering plants, mosses, liverworts, advanced algae like the ‘proto-liverwort’ Coleochaete (Buglass et al. 2008) right down to unicellular flagellated algae like Chlamydomonas (Miller et al. 1972; Roberts 1974; Hills et al. 1975; Goodenough et al. 1985; Goodenough & Heuser 1989). 13.2.5 Crosslinking code involving tyrosine motifs Although AGPs lack tyrosine for the most part and do not form covalently crosslinked networks, extensins are generally Tyr-rich, enabling them to behave as self-assembling amphiphiles and form networks further stabilized by covalent crosslinks (Brady et al. 1996; Brady et al. 1998; Cannon et al. 2008) putatively mediated by extensin peroxidase (Schnabelrauch et al. 1996). These networks involve short motifs, typically Tyr-Xaa-Tyr where the diphenyl ether isodityrosine (Idt) forms a very short intramolecular crosslink (Fry 1982; Epstein & Lamport 1984;). Idt may then react with a single Tyr residue on an adjacent polypeptide to create the trityrosine (pulcherosine) intermolecular crosslink (Brady et al. 1998; Cannon et al. 2008), or with another Idt to form the Tyr tetramer diisodityrosine (di-Idt) (Brady et al. 1996; Held et al. 2004; Cannon et al. 2008).

13.3

Molecular function, biological role

13.3.1 Arabinogalactan proteins (AGPs) Classical AGPs are hydrophiles when freely soluble, but AGPs occur initially as amphiphiles bound to the outer leaflet of the plasma membrane by a hydrophobic GPI C-terminal tail. This anchors AGPs to the periplasmic surface of the plasma membrane (Oxley & Bacic 1999; Sherrier et al. 1999; Svetek et al. 1999) possibly in lipid rafts via self-association of the C24 lignoceric fatty acid component. The abundance and periclinal orientation of periplasmic AGPs implies a surface coverage sufficient to act as a periplasmic cushion (Lamport et al. 2006) and barrier protecting the delicate plasma membrane. Phospolipases cleave the GPI anchor to yield soluble periplasmic AGPs that are then readily released by cell disruption. AGPs cannot diffuse freely through the wall because they are too large; even small Hyp-arabinogalactan polysaccharides, when folded, behave as globular ellipsoids (Tan et al. 2007) that approximate the porosity of pectin, <5 nm

330

■

Plant Polysaccharides

diameter (Carpita et al. 1979; Baron-Epel et al. 1988; Titel et al. 1997; Lamport et al. 2006). How then do soluble periplasmic AGPs pass through the cell wall and into the growth medium? The wall grows largely though layer-by-layer addition of large macromolecules like pectin and cellulose. Thus, initially soluble periplasmic AGPs gradually become embedded and move through a continuously expanding wall by ‘plug-flow’ extrusion (Lamport et al. 2006). Hypothetically, AGP monomers disrupt intermolecular pectic association and thus modulate wall rheology by acting as pectic plasticizers (Lamport 2001; Lamport et al. 2006). This is another possible control mechanism regulating cell extension and is supported by the fact that the Yariv reagent (β-d-glycosyl)3 Yariv phenylglycoside, which crosslinks AGPs, inhibits cell expansion (Roy et al. 1998). Rapidly extending pollen tube tips may represent the extreme scenario of pectin-rich cell walls well endowed with AGPs and concomitant high flow rheology. 13.3.2 Extensins as self-assembling amphiphiles Although some consider ‘cessation of growth’ as the major role of extensin (Cleland & Karlsnes 1967; Sadava & Chrispeels 1973; Cleland 2001), this pervasive view ignores other evidence such as: the lack of ‘correlation between HRGP synthesis and cessation of wall elongation’ and suggestions that extensins are not only ‘important at an early stage of wall assembly’ (Ye & Varner 1991) but also ‘for the correct assembly of other wall polymers’ including the prescient observation that ‘cytokinesis [may] require specific HRGPs’ (Cooper et al. 1994). More recent work supports these views of a dynamic role for extensins: For example, cell-cycle regulated expression of extensin in Catharanthus coincides with and thus contributes to the formation of the cell plate (Ito et al. 1998), while in ferns expression of a specific extensin coincides with the first mitosis of germinating spores (Uchida et al. 1998). All these data confirm the observation that ‘protein containing hydroxyproline is an integral part of the cell wall of actively growing cells’ (Lamport & Northcote 1960). The recent demonstration that RSH extensin (AtEXT3) is essential for cytokinesis (Hall & Cannon 2002; Cannon et al. 2008) goes further; the rsh mutant shows definitively how extensin plays a dynamic role in the initiation of growth: AtEXT3 has a propensity for self-assembly as dendritic structures (Cannon et al. 2008) (Fig. 13.3) and is direct evidence that highly periodic extensins (Smith et al. 1986) are self-assembling amphiphiles. This key concept (Rapaport 2006) led to the extensin HRGP self-assembly paradigm (Cannon et al. 2008) that depends crucially on the alternation of hydrophilic glycomodules with hydrophobic polypeptides that induce likewith-like self-recognition by homophilic alignment of hydrophilic (SerHyp4) glycomodules and hydrophobic peptide crosslinking modules (Brady et al. 1996; Cannon et al. 2008).

Hydroxyproline-rich Glycoproteins: Form and Function

A

■

331

C 4 1.5 3 1.0 2

500 nm

1

0.5

0

0.0 nm

–1 nm

B Counts

20 15

Gaussian fit: 140 +/– 20 nm

10 5 0 100 200 300 400 Segment length (nm)

500

Figure 13.3 The RSH extensin network imaged by atomic force microscopy (AFM). (A) RSH (9 μg/mL water) was deposited on a highly ordered pyrolytic graphite surface. Molecular height ranged from 0.6 nm to 5 nm, estimated using the nm ruler (right). The white bars correspond to the length of RSH (127 nm) estimated by assuming a strict polyproline II helical secondary structure (3 residues/turn; pitch = 9.4 angstroms) for the 404 amino acid protein. The segments of the dendritic structure correspond to a few RSH monomers (parallel to the 127 nm white bars in the circled areas) aligned to produce segments slightly longer than the RSH molecule in keeping with a staggered alignment. (B) Analysis of all segments in (A) yielded the histogram in (B). (C) Imaging of (YK)20 (10 μg/mL water), an extensin analog produced through synthetic genes. (YK)20 contains 20 tandem repeats of the arabinosylated extensin palindromic sequence SO4SOSO4YYYK, also in a polyproline II helix. (Scale bar in (A) and (C), 500 nm.) From M.C. Cannon, K. Terneus, Q. Hall, et al. (2008) Self-assembly of the plant cell wall requires an extensin scaffold. Proceedings of the National Academy of Sciences of the U S A, 105(6), 2226–2231. With permission from the National Academy of Sciences.

The dendritic structures observed via AFM (Fig. 13.3) suggest that orderly assembly of extensin monomers involves a zipper-like endwise association consistent with preferential crosslinking at the ends of Idt-rich oligomers as observed via TEM (Stafstrom & Staehelin 1986). Presumably the phase-rich environment (the interfacing of hydrophobic lipid and hydrophilic glycoproteins and polysaccharides) of the cell plate would optimize further supramolecular self-assembly, difficult to replicate experimentally as it requires isolation of the cell plate (Yasuhara 2005). However, if oligomeric extensin reacts with pectin (Smith et al. 1984) by a simple acid–base reaction then it most likely templates further orderly deposition of a mesoporous (∼5 nm diameter) coacervate of extensin pectate in

332

■

Plant Polysaccharides

Table 13.4 Examples of extensin-like HRGPs and HRGP chimeras predicted from the rice genome

Motifs in P-rich domain: Accession: Os01g0110200 Os01g0180000 Os01g0309100 Os01g0356900 Os01g0594300 Os01g0899700 Os01g02160 Os02g0138000 Os04g34170 Os05g0516400 Os05g0552600 Os06g49100 Os07g0105300 Os10g0480500 Os11g43640 Os12g0542200

Idt motifs

Ara modules X-Hyp3–5 (X = S,T,A,V)

Ara modules XP2 (X = S,T,A,V)

AG modules XP (X = S,T,A,V)

P-rich domain, total residues

Total residues

0 0 1 0 1 4 1 0 0 0 0 0 0 0 0 0

9 11 3 8 16 7 1 3 2 9 19 4 5 0 18 2

4 0 0 1 0 11 2 18 14 15 1 8 0 3 32 1

23 5 2 3 7 7 4 21 44 32 10 9 4 16 41 21

214 141 107 90 154 225 109 309 329 391 194 174 87 469 505 344

240 520 192 503 551 412 159 694 360 827 510 538 161 514 946 760

the nascent cell wall (Cannon et al. 2008). ‘Pectin is not just jelly!’ (Roberts 1990) and therefore it needs to be organized. 13.3.3 Role of multiple extensins Expressed extensin sequences of Arabidopsis (Cannon et al. 2008; SI in Table 13.4) and (putative) HRGP sequences of a wide range of other species show that they have motifs that generally correspond to the extensin network precursors P1, P2 and P3 first isolated as monomeric glycoproteins (P1 and P2) highly expressed in tomato cultures or as peptide fragments (P3) (Smith et al. 1984; Smith et al. 1986). Like Arabidopsis, tomato extensins fall into two major groups based on their Idt content:

• Class I: Idt-poor typified by tomato P1, which has the major repetitive motif Ser-Hyp4-Thr-Hyp-Val-Tyr-Lys

• Class II: Idt-rich, with two subclasses: 䊊

䊊

IIa: Idt-rich typified by tomato P2, whose major motif is Ser-Hyp4-ValTyr-Lys-Tyr-Lys. IIb: Idt-rich with the palindromic motif (underlined) Ser-Hyp4-SerHyp-Ser-Hyp4-Tyr-Tyr-Tyr-Lys. Typified by tomato P3 (Showalter et al. 1991; Zhou et al. 1992), this motif is highly conserved from ferns to Poaceae, together with equally conserved extensin Hyp-arabinosides whose unusual β-1,2-linked anomeric configuration may be designed to

Hydroxyproline-rich Glycoproteins: Form and Function

■

333

optimize specific H-bonding between the complementary surfaces of hydrophilic Ser-Hyp4 glycomodules. In tomato cell cultures an extensin scaffold is built of just three major components: P1-, P2- and P3-type extensins. We base this on the observation that these three extensins are insolubilized in the wall (Smith et al. 1986). Ionic desorption rapidly elutes P1 and P2 extensin monomers (>50% in 10 s) from cultured cells (Smith et al. 1984) but not the P3 type. This is possibly due to an anticlinal orientation of P1 and P2 (easily elutable) and a periclinal orientation of P3 (not elutable), whose orientation coupled with rigidity contributed by the exceptionally Hyp-rich palindromic motif would prevent elution through wall pores. This implies transmural insertion of some extensins (‘intussusception’) while bulk deposition of the wall, including deposition of other extensins, follows the rules of layer-by-layer selfassembly (‘apposition’) (Salditt & Schubert 2002; Ariga et al. 2006). The model in Fig. 13.4 of the extensin network within the cell wall is based on the RSH data (Fig. 13.3), P3 elution data, the striking differences between the Idt content of extensins, and the electron microscopy images of carrot EXT 1 showing that 60% of extensin crosslinks were at the end of molecules (Stafstrom & Staehelin 1986). C

M

P

5 nm

Extensins: anticlinal

periclinal

~80 nm

Figure 13.4 Hypothetical extensin matrix. Idt-poor extensins with exclusively terminal Idt may bridge Idt-rich extensins. Hypothetically this would enable anticlinally oriented Idt-poor (transmural) extensins and periclinally oriented Idt-rich extensins to form a threedimensional extensin scaffold and thus mechanically couple wall layers. Scales are approximations only. C, cytoplasm; M, plasma membrane; P, periplasmic AGPs represented as ‘beads on a string’ with ovals representing globular AGs. The cell wall matrix is represented by very small ovals with larger voids that correspond to the pectic porosity. Cellulose and xyloglucan are not shown.

334

■

Plant Polysaccharides

Although extensin apparently plays a pivotal role in cell wall assembly and structure (Cannon et al. 2008), many questions remain unanswered: Why are there such large extensin families of closely homologous glycoproteins? Are they ‘tailor-made’ for different cell types or do they represent gene redundancy? Extensins are undoubtedly multifunctional, reminiscent of multiple collagens tailored to the animal tissue. Their influence may be direct as in extensin networks per se or indirect by affecting cell wall assembly and consequently its structure. Extensins enhance the tensile strength of mechanically stressed tissues in stems (Shirsat et al. 1996; Shirsat et al. 2003) and roothairs (Bernhardt & Tierney 2000; Baumberger et al. 2003), and provide barriers to pathogens (Esquerre-Tugaye & Lamport 1979; Wei & Shirsat 2006). Hypothetically extensins could contribute to the regulation of cell extension (Petersen et al. 1999; Li & Cosgrove 2001; Grobe et al. 2002) by proteolysis of the Idt-poor transmural extensins that may tie wall layers together. This is not unreasonable, given that there are at least ten proteases in the Arabidopsis wall proteome (Boudart et al. 2005) and putative species specific proteolytic cut sites represented by translated motifs such as Val-Lys-Pro-Tyr-His-Pro of tomato P1 extensin. Furthermore, a cysteine endopeptidase active during apoptosis specifically cleaves the Lys-Lys of the glycosylated sequences SOOOKKPYYP and SOOOHKKPYYP in tobacco P1 extensin, suggesting that this protease dismantles the extensin scaffolds of senescing cells (Helm et al. 2008). This implies that similar proteases could remodel extensin scaffolds to facilitate cell expansion (Fig. 13.4).

13.4 Evolution 13.4.1 Conserved motifs The Idt-rich P3 type motif, SPPPPSPSPPPPYYYK, shows astonishing evolutionary conservation since the Devonian over ∼400 million years ago – the maidenhair fern, Adiantum capillus-veneris, (AcExt-1) 16-residue consensus motif (Uchida et al. 1998) SPPPPSPSPPPPYIYK differs from tomato P3 by one residue only, I → Y. In many of the other angiosperms, including potato, Populus trichocarpa, the orchid Bromheadia, Medicago, petunia and bean, the palindromic motif is identical with that of tomato or closely related, as with soybean (Accession AAB53156). In the Poaceae, however, annotated genomic sequences show that although inferred Ser-Hyp4 motifs are common, Idt motifs are much less evident. Why are grasses so different? Compared with other herbs the grasses represent a radical departure in growth habit, having largely escaped the tyranny of turgor and dependence on water for tissue support; this is reflected in their cell wall structure where pectin is largely replaced by glucuronoxylans, increased crosslinked ferulate and silica support. Extensins too reflect this change; available sequence data show extensin-like proteins with similarities to those of

Hydroxyproline-rich Glycoproteins: Form and Function

■

335

Arabidopsis although divergent in structure - rich in TP motifs but Idt-poor (see Table 13.4). 13.4.2 Evolution of the cell plate – nascent cell wall An essential role for extensin at cytokinesis has profound implications: The evolutionary origin of the cell plate itself may be predicated on the appearance of self-assembling amphiphiles potentially covalently crosslinked via di-Idt. This raises questions of phylogeny that cannot be adequately dealt with here. Nevertheless, one can trace the extensin HRGP family back to single-celled protists like Chlamydomonas with ‘animal-like’ characteristics absent from higher plants: a contractile vacuole regulates osmotic homeostasis; a centriole organizes the mitotic spindle and cytokinesis involves a cleavage furrow. Chlamydomonas lacks a cell plate and predates the appearance of cellulose as its fragile cell wall consists entirely of HRGPs (Roberts 1974; Goodenough et al. 1986). These remarkable glycoproteins are unique to the green plant line of evolution with detectable conservation of polypeptide sequence and Hyp glycosylation particularly the Hyp arabinoside anomeric configuration (Akiyama et al. 1980; Bollig et al. 2007). The Chlamydomonas genome encodes many ‘extensin-like’ proteins with a preponderance of SerPro motifs typical of AGPs (Shpak et al. 1999). One extensin, contains four Idt motifs whereas four others contain only two or three (Table 13.5). However, even this may suffice to peroxidatically crosslink the thin ‘W2’ inner wall of Chlamydomonas via tyrosine derivatives (Waffenschmidt et al. 1993) even though the suggested W2 component VSP-1 (UniProt Q07373) contains no Idt motifs. Evolution of the cell plate predestines turgor-driven growth – the sine qua non of landplants with their priapic growth habit. Plants evolved from chlorophyte algae like Chlamydomonas coincident with a transition from osmotically fragile cells to tough walls and hydraulic growth. The cell plate of a landplant originates from Golgi vesicles. Thus, evolution of the cell plate Table 13.5 Idt motifs in extensin-like proteins of Chlamydomonas

Accession

Motifs Idt [YXY]

YY

Y

SP3–7

SP2

SP

Total residues

AAK60544 AAK83527 AAT02521 L29029 AY450930 Q07373

2 3 3 2 4 0

1 2 3 1 3 2

36 39 45 13 60 15

0 0 25 1 12 1

1 0 4 1 17 0

21 36 11 70 152 31

955 991 1146 473 3409 300

336

■

Plant Polysaccharides

can be traced back to basal chlorophytes, notably, extensin-like scaffold proteins of Chlamydomonas and cellulosic scales exported from the Golgi to the cell surface of Prasinophytes (Brown & Romanovicz 1976).

13.5 Epilogue Fifty years ago in D.H. Northcote’s laboratory the primary cell wall was isolated (Lamport & Northcote 1960); this led to the discovery of its unique structural protein rich in hydroxyproline. The ramifications are diverse and profound. It questioned the classical textbook view of the primary cell wall as ‘cellulose embedded in an amorphous matrix’ and led to more dynamic models of an organelle with a unique proteome. Equally significant it begins to explain the origin of a self-assembling ‘morphogenetic fabric’ that underpins the evolution of landplants. The next 50 years must integrate the frayed edges of disparate fields – polymer chemistry, materials science and biology of the cell wall – into a seamless conceptual fabric.

Acknowledgments This work was supported by grants from the National Science Foundation (Award Numbers 0622428 and 0622394) and the Ohio University Bimimetic Nanoscience and Nanotechnology Program.

References Akiyama, Y., Mori, M., Kato, K. (1980) 13C-NMR analysis of hydroxyproline arabinosides from Nicotiana tabacum. Agricultural and Biological Chemistry, 44, 2487–2489. Allen, A.K., Desai, N.N., Neuberger, A., Creeth, J.M. (1978) Properties of potato lectin and the nature of its glycoprotein linkages. Biochemical Journal, 171, 665–674. Ariga, K., Nakanishi, T., Michinobu, T. (2006) Immobilization of biomaterials to nano-assembled films (self-assembled monolayers, Langmuir-Blodgett films, and layer-by-layer assemblies) and their related functions 267. Journal of Nanoscience and Nanotechnology, 6(8), 2278–2301. Ashford, D., Desai, N.N., Allen, A.K., Neuberger, A., O’Neill, M.A., Selvendran, R.R. (1982) Structural studies of the carbohydrate moieties of lectins from potato (Solanum tuberosum) tubers and thornapple (Datura stramonium) seeds. Biochemical Journal, 201, 199–208. Aspinall, G.O., Malloy, J.A., Craig, J.W.T. (1969) Extracellular polysaccharides from suspension-cultured sycamore cells. Canadian Journal of Biochemistry, 47, 1063–1070. Baldwin, T.C., McCann, M.C., Roberts, K. (1993) A novel hydroxyproline-deficient arabinogalactan protein secreted by suspension-cultured cells of Daucus carota. Plant Physiology, 103, 115–123.

Hydroxyproline-rich Glycoproteins: Form and Function

■

337

Baldwin, T.C., Domingo, C., Schindler, T., Seetharaman, G., Stacey, N., Roberts, K. (2001) DcAGP1, a secreted arabinogalactan protein, is related to a family of basic proline-rich proteins. Plant Molecular Biology, 45(4), 421–435. Baron-Epel, O., Gharyal, P.K., Schindler, M. (1988) Pectins as mediators of wall porosity in soybean cells. Planta, 175, 389–395. Baumberger, N., Steiner, M., Ryser, U., Keller, B., Ringli, C. (2003) Synergistic interaction of the two paralogous Arabidopsis genes LRX1 and LRX2 in cell wall formation during root hair development. Plant Journal, 35(1), 71–81. Bendtsen, J.D., Nielsen, H., von Heijne, G., Brunak, S. (2004) Improved prediction of signal peptides: SignalP 3.0. Journal of Molecular Biology, 340, 783–795. Bernhardt, C., Tierney, M.L. (2000) Expression of AtPRP3, a proline-rich structural cell wall protein from Arabidopsis, is regulated by cell-type-specific developmental pathways involved in root hair formation. Plant Physiology, 122(3), 705–714. Bollig, K., Lamshoeft, M., Schweimer, K., Marner, F.-J., Budzikiewiczc, H., Waffenschmidt, S. (2007) Structural analysis of linear hydroxyproline-bound O-glycans of Chlamydomonas reinhardtii – conservation of the inner core in Chlamydomonas and land plants. Carbohydrate Research, 342, 2557–2566. Borner, G.H.H., Lilley, K.S., Stevens, T.J., Dupree, P. (2003) Identification of glycosylphosphatidylinositol-anchored proteins in Arabidopsis. A proteomic and genomic analysis. Plant Physiology, 132(2), 568–577. Boudart, G., Jamet, E., Rossignol, M., et al. (2005) Cell wall proteins in apoplastic fluids of Arabidopsis thaliana rosettes: Identification by mass spectrometry and bioinformatics. Proteomics, 5(1), 212–221. Brady, J.D., Sadler, I.H., Fry, S.C. (1996) Di-isodityrosine, a novel tetrameric derivative of tyrosine in plant cell wall proteins: a new potential cross-link. Biochemical Journal, 315, 323–327. Brady, J.D., Sadler, I.H., Fry, S.C. (1998) Pulcherosine, an oxidatively coupled trimer of tyrosine in plant cell walls: its role in cross-link formation. Phytochemistry, 47, 349–353. Brown, R.M., Jr., Romanovicz, D.K. (1976) Biogenesis and structure of Golgi-derived cellulosic scales in Pleurochrysis. I. Role of the endomembrane system in scale assembly and exocytosis. Appied Polymer Symposium, 28, 537–585. Buglass, S., Lamport, D.T.A., Xu, J., Tan, Li., Kieliszewksi, M.J. (2008) Origin of the land plants: is Coleochaete their closest living relative? The writing is on the wall. 11th Cell Wall Meeting, Copenhagen, Abstract 17. Cannon, M.C., Terneus, K.A., Hall, Q., et al. (2008) Self-assembly of the plant cell wall requires an extensin scaffold. Proceedings of the National Academy of Sciences of the U S A, 105(6), 2226–2231. Carpita, N., Sabularse, D., Montezinos, D., Delmer, D.P. (1979) Determination of the pore size of cell walls of living plant cells. Science, 205, 1144–1147. Cho, Y.-P., Chrispeels, M.J. (1976) Serine-O-galactosyl linkages in glycopeptides from carrot cell walls. Phytochemistry, 15, 165–169. Cleland, R.E. (2001) Unlocking the mysteries of leaf primordia formation. Proceedings of the National Academy of Sciences of the U S A, 98(20), 10981–10982. Cleland, R.E., Karlsnes, A. (1967) A possible role for hydroxyproline-containing proteins in the cessation of cell elongation. Plant Physiology, 42, 669–671. Cooper, J.B., Heuser, J.E., Varner, J.E. (1994) 3,4-Dehydroproline inhibits cell wall assembly and cell division in tobacco protoplasts. Plant Physiology, 104, 747–752.

338

■

Plant Polysaccharides

Epstein, L., Lamport, D.T.A. (1984) An intramolecular linkage involving isodityrosine in extensin. Phytochemistry, 23, 1241–1246. Esquerre-Tugaye, M.T., Lamport, D.T.A. (1979) Cell surfaces in plant-microorganism interactions. I. A structural investigation of cell wall hydroxyproline-rich glycoproteins which accumulate in fungus-infected plants. Plant Physiology, 64, 314–319. Estevez, J.M., Kieliszewski, M.J., Khitrov, N., Somerville, C. (2006) Characterization of synthetic hydroxyproline-rich proteoglycans with arabinogalactan protein and extensin motifs in Arabidopsis. Plant Physiology, 142(2), 458–470. Fincher, G.B., Sawyer, W.H., Stone, B.A. (1974) Chemical and physical properties of an arabinogalactan-peptide from wheat endosperm. Biochemical Journal, 139, 535–545. Fowler, T.J., Bernhardt, C., Tierney, M.L. (1999) Characterization and expression of four proline-rich cell wall protein genes in Arabidopsis encoding two distinct subsets of multiple domain proteins. Plant Physiology, 121(4), 1081–1091. Fry, S.C. (1982) Isodityrosine, a new cross-linking amino acid from plant cell-wall glyoprotein. Biochemical Journal, 204, 449–455. Goodenough, U.W., Heuser, J.E. (1989) Molecular organization of cell-wall crystals from Chlamydomonas reinhardtii and Volvox carteri. Journal of Cell Science, 90, 717–733. Goodenough, U.W., Adair, W.S., Collin-Osdoby, P., Heuser, J.E. (1985) Structure of the Chlamydomonas agglutinin and related flagellar surface proteins in vitro and in situ. Journal of Cell Biology, 101, 924–941. Goodenough, U.W., Gebhart, B., Mecham, R.P., Heuser, J.E. (1986) Crystals of the Chlamydomonas reinhardtii cell wall: polymerization, depolymerization and purification of glycoprotein monomers. Journal of Cell Biology, 103, 405–417. Goodrum, L.J., Patel, A., Leykam, J.F., Kieliszewski, M.J. (2000) Gum arabic glycoprotein contains glycomodules of both extensin and arabinogalactan-glycoproteins. Phytochemistry, 54(1), 99–106. Grobe, K., Poppelmann, M., Becker, W.-M., Petersen, A. (2002) Properties of group I allergens from grass pollen and their relation to cathepsin B, a member of the C1 family of cysteine proteinases. European Journal of Biochemistry, 269, 2083–2092. Hall, Q., Cannon, M.C. (2002) The cell wall hydroxyproline-rich glycoprotein RSH is essential for normal embryo development in Arabidopsis. Plant Cell, 14, 1161–1172. Held, M.A., Tan, L., Kamyab, A., Hare, M., Shpak, E., Kieliszewksi, M.J. (2004) Diisodityrosine is the intermolecular cross-link of isodityrosine-rich extensin analogs cross-linked in vitro. Journal of Biological Chemistry, 279, 55474–55482. Helm, M., Schmid, M., Hierl, G., et al. (2008) KDEL-tailed cysteine endopeptidases involved in programmed cell death,intercalation of new cells and dismantling of extensin scaffolds. American Journal of Botany, 95, 1049-1062. Hieta, R., Myllyharju, J. (2002) Cloning and characterization of a low molecular weight prolyl 4-hydroxylase from Arabidopsis thaliana. Effective hydroxylation of proline-rich, collagen-like, and hypoxia-inducible transcription factor α-like peptides. Journal of Biological Chemistry, 277(26), 23965–23971. Hills, G.J., Phillips, J.M., Gay, M.R., Roberts, K. (1975) Self-assembly of a plant cell wall in vitro. Journal of Molecular Biology, 96, 431–444. Ito, M., Kodama, H., Komamine, A., Watanabe, A. (1998) Expression of extensin genes is dependent on the stage of the cell cycle and cell proliferation in suspensioncultured Catharanthus roseus cells. Plant Molecular Biology, 36(3), 343–351.

Hydroxyproline-rich Glycoproteins: Form and Function

■

339

Jermyn, M.A., Yeow, Y.M. (1975) A class of lectins present in the tissues of seed plants. Australian Journal of Plant Physiology, 2, 501–531. Johnson, K.L., Jones, B.J., Bacic, A., Schultz, C.J. (2003a) The fasciclin-like arabinogalactan proteins of Arabidopsis. A multigene family of putative cell adhesion molecules. Plant Physiology, 133(4), 1911–1925. Johnson, K.L., Jones, B.J., Schultz, C.J., Bacic, A. (2003b) Non-enzymic cell wall (glyco)proteins. In: The Plant Cell Wall, Annual Plant Reviews Vol. 8 (ed. J.K. Rose), pp. 111–154. Blackwell Publishing Ltd., Oxford. Kauss, H., Seehaus, K., Franke, R., Gilbert, S., Dietrich, R.A., Kroger, N. (2003) Silica deposition by a strongly cationic proline-rich protein from systemically resistant cucumber plants. Plant Journal, 33(1), 87–95. Keller, B. (1993) Structural cell wall proteins. Plant Physiology, 101, 1127–1130. Kieliszewski, M.J., Lamport, D.T.A. (1994) Extensin: repetitive motifs, functional sites, posttranslational codes and phylogeny. Plant Journal, 5, 157–172. Kieliszewski, M.J., Kamyab, A., Leykam, J.F., Lamport, D.T.A. (1992) A histidine-rich extensin from Zea mays is an arabinogalactan protein. Plant Physiology, 99, 538–547. Kieliszewski, M.J., Leykam, J.F., Lamport, D.T.A. (1990) Structure of the threoninerich extensin from Zea mays. Plant Physiology, 92, 316–326. Kieliszewski, M.J., Showalter, A.M., Leykam, J.F. (1994) Potato lectin: a modular protein sharing sequence similarities with the extensin family, the hevein lectin family, and snake venom disintegrins (platelet aggregation inhibitors). Plant Journal, 5, 849–861. Lamport, D.T.A. (1963) Oxygen fixation into hydroxyproline of plant cell wall protein. Journal of Biological Chemistry, 238, 1438–1440. Lamport, D.T.A. (1965) The protein component of primary cell walls. Advances in Botanical Research, 2, 151–218. Lamport, D.T.A. (1967) Hydroxyproline-O-glycosidic linkage of the plant cell wall glycoprotein extensin. Nature, 216, 1322–1324. Lamport, D.T.A. (1970) Cell wall metabolism. Annual Review ofPlant Physiology, 21, 235–270. Lamport, D.T.A. (1973) The glycopeptide linkages of extensin, O-D-galactosyl serine and O-L-arabinosyl hydroxyproline. In: Biogenesis of Plant Cell Wall Polysaccharides, pp. 149–164. Academic Press, New York. Lamport, D.T.A. (1977) Structure, biosynthesis and significance of cell wall glycoproteins. In: Recent Advances in Phytochemistry, 11 (eds F.A. Loewus and V.C. Runeckles), pp. 79–115. Plenum Press, New York. Lamport, D.T.A. (2001) Life behind cell walls: Paradigm lost, paradigm regained. Cellular and Molecular Life Sciences, 58, 1363–1385. Lamport, D.T.A., Northcote, D.H. (1960) Hydroxyproline in primary cell walls of higher plants. Nature, 188, 665–666. Lamport, D.T.A., Katona, L., Roerig, S. (1973) Galactosyl serine in extensin. Biochemical Journal, 133, 125–131. Lamport, D.T.A., Kieliszewksi, M.J., Showalter, A.M. (2006) Salt-stress upregulates periplasmic arabinogalactan-proteins: using salt-stress to analyse AGP function. New Phytologist, 169(3), 479–492. Leucci, M.R., Di Sansebastiano, G.P., Gigante, M., Dalessandro, G., Piro, G. (2007) Secretion marker proteins and cell-wall polysaccharides move through different secretory pathways. Planta, 225(4), 1001–1017.

340

■

Plant Polysaccharides

Li, L.C., Cosgrove, D.J. (2001) Grass group I pollen allergens (?-expansins) lack proteinase activity and do not cause wall loosening via proteolysis. European Journal of Biochemistry, 268(15), 4217–4226. Miller, D.H., Lamport, D.T.A., Miller, M. (1972) Hydroxyproline heterooligosaccharides in Chlamydomonas. Science, 176, 918–920. Oxley, D., Bacic, A. (1999) Structure of the glycosylphosphatidylinositol anchor of an arabinogalactan protein from Pyrus communis suspension-cultured cells. Proceedings of the National Academy of Sciences of the U S A, 96(25), 14246– 14251. Petersen, A., Grobe, K., Schramm, G., et al. (1999) Structure and function can determine important features in allergenicity: investigations on the group I allergens of the grasses. International Symposium on Food Allergens, 1(3), 95–101. Pope, D.G. (1977) Relationships between hydroxyproline-containing proteins secreted into the cell wall and medium by suspension-cultured Acer pseudoplatanus cells. Plant Physiology, 59, 894–900. Qi, W., Fong, C., Lamport, D.T.A. (1991) Gum arabic glycoprotein is a twisted hairy rope: a new model based on O-galactosylhydroxyproline as the polysaccharide attachment site. Plant Physiology, 96, 848–855. Rapaport, H. (2006) Ordered peptide assemblies at interfaces218. Supramolecular Chemistry, 18(5), 445–454. Roberts, K. (1974) Crystalline glycoprotein cell walls of algae: Their structure, composition and assembly. Philosophical Transactions of the Royal Society of London Series B, 268, 129–146. Roberts, K. (1990) Structures at the plant cell surface. Current Opinion in Cell Biology, 2, 920–928. Roy, S., Jauh, G.Y., Hepler, P.K., Lord, E.M. (1998) Effects of Yariv phenylglycoside on cell wall assembly in the lily pollen tube. Planta, 204(4), 450–458. Sadava, D., Chrispeels, M.J. (1971) Hydroxyproline biosynthesis in plant cells: peptidyl proline hydroxylase from carrot disks. Biochimica et Biophysica Acta, 227, 278–287. Sadava, D., Chrispeels, M.J. (1973) Hydroxyproline-rich cell wall protein (extensin): role in the cessation of elongation in excised pea epicotyls. Developmental Biology, 30, 49–55. Salditt, T., Schubert, U.S. (2002) Layer-by-layer self-assembly of supramolecular and biomolecular films. Reviews in Molecular Biotechnology, 90(1), 55–70. Schnabelrauch, L.S., Kieliszewski, M.J., Upham, B.L., Alizedeh, H., Lamport, D.T.A. (1996) Isolation of pI 4.6 extensin peroxidase from tomato cell suspension cultures and identification of Val-Tyr-Lys as putative intermolecular cross-link site. Plant Journal, 9, 477–489. Schultz, C.J., Rumsewicz, M.P., Johnson, K.L., Jones, B.J., Gaspar, Y.M., Bacic, A. (2002) Using genomic resources to guide research directions. The Arabinogalactan protein gene family as a test case. Plant Physiology, 129(4), 1448–1463. Schultz, C.J., Ferguson, K.L., Lahnstein, J., Bacic, A. (2004) Post-translational modifications of arabinogalactan-peptides of Arabidopsis thaliana: ER and GPI-anchor signal cleavage sites and hydroxylation of proline. Journal of Biological Chemistry, 279, 1–48. Sherrier, D.J., Prime, T.A., Dupree, P. (1999) Glycosylphosphatidylinositol-anchored cell-surface proteins from Arabidopsis. Electrophoresis, 20(10), 2027–2035.

Hydroxyproline-rich Glycoproteins: Form and Function

■

341

Shimizu, M., Igasaki, T., Yamada, M., et al. (2005) Experimental determination of proline hydroxylation and hydroxyproline arabinogalactosylation motifs in secretory proteins. Plant Journal, 42, 877–889. Shirsat, A.H., Bell, A., Spence, J., Harris, J.N. (1996) The Brassica napus extA extensin gene is expressed in regions of the plant subject to tensile stresses. Planta, 199, 618–624. Shirsat, A.H., Thomson, H.E.C., Elliott, K.A. (2003) The Brassica napus extA extensin gene negative regulatory region controls expression in response to mechanical stresses. Plant, 26(10), 1647–1655. Showalter, A.M., Zhou, J., Rumeau, D., Worst, S.G., Varner, J.E. (1991) Tomato extensin and extensin-like cDNAs: structure and expression in response to wounding. Plant Molecular Biology, 16, 547–565. Shpak, E., Leykam, J.F., Kieliszewski, M.J. (1999) Synthetic genes for glycoprotein design and the elucidation of hydroxyproline-O-glycosylation codes. Proceedings of the National Academy of Sciences of the U S A, 96(26), 14736–14741. Shpak, E., Barbar, E., Leykam, J.F., Kieliszewski, M.J. (2001) Contiguous hydroxyproline residues direct hydroxyproline arabinosylation in Nicotiana tabacum. Journal of Biological Chemistry, 276(14), 11272–11278. Smith, J.J., Muldoon, E.P., Lamport, D.T.A. (1984) Isolation of extensin precursors by direct elution of intact tomato cell suspension cultures. Phytochemistry, 23, 1233–1239. Smith, J.J., Muldoon, E.P., Willard, J.J., Lamport, D.T.A. (1986) Tomato extensin precursors P1 and P2 are highly periodic structures. Phytochemistry, 25, 1021–1030. Stafstrom, J.P., Staehelin, L.A. (1986) Cross-linking patterns in salt-extractable extensin from carrot cell walls. Plant Physiology, 81, 234–241. Svetek, J., Yadav, M.P., Nothnagel, E.A. (1999) Presence of a glycosylphosphatidylinositol lipid anchor on rose arabinogalactan proteins. Journal of Biological Chemistry, 274(21), 14724–14733. Tan, L., Leykam, J.F., Kieliszewski, M.J. (2003) Glycosylation motifs that direct arabinogalactan addition to arabinogalactan-proteins. Plant Physiology, 132(3), 1362–1369. Tan, L., Qiu, F., Lamport, D.T.A., Kieliszewski, M.J. (2004) Structure of a hydroxyproline (Hyp)-arabinogalactan polysaccharide from repetitive Ala-Hyp expressed in transgenic Nicotiana tabacum. Journal of Biological Chemistry, 279(13), 13156–13165. Tan, L., Xu, J., Lamport, D.T.A., Feng, Q., Cottrell, C., Jin, Q., Kieliszewksi, M.J. (2007) The AGP Hyp-arabinogalactan backbone is a folded polysaccharide of reverseturn hairpins generated by β-(1–6) linked repeats of β-(1–3) trigalactosyl units. 11th Cell Wall Meeting, Copenhagen, Abstract 68. Tiainen, P., Myllyharju, J., Koivunen, P. (2005) Characterization of a second arabidopsis thaliana prolyl 4-hydroxylase with distinct substrate specificity. Journal of Biological Chemistry, 280, 1142–1148. Titel, C., Woelecke, H., Afifi, I., Ehwald, R. (1997) Dynamics of limiting cell wall porosity in plant suspension cultures. Planta, 203(3), 320–326. Uchida, K., Muramatsu, T., Jamet, E., Furuya, Y. (1998) Control of expression of a gene encoding an extensin by phytochrome and a blue light receptor in spores of Adiantum capillus-veneris. Plant Journal, 15, 813–819.

342

■

Plant Polysaccharides

Van Damme, E.J.M., Barre, A., Rouge, P., Peumans, W.J. (2004) Potato lectin: an updated model of a unique chimeric plant protein. Plant Journal, 37(1), 34–45. Waffenschmidt, S., Woessner, J.P., Beer, K., Goodenough, U.W. (1993) Isodityrosine cross-linking mediates insolubilization of cell walls in Chlamydomonas. Plant Cell, 5, 809–820. Wei, G., Shirsat, A.H. (2006) Extensin over-expression in Arabidopsis limits pathogen invasiveness. Molec.Plant Pathol, 7, 579–592. Wilson, R.C., Long, F., Maruoka, E.M., Cooper, J.B. (1994) A new proline-rich early nodulin from Medicago truncatula is highly expressed in nodule meristematic cells. Plant Cell, 6, 1265–1275. Xu, J., Tan, L., Goodrum, K.J., Kieliszewksi, M.J. (2007) High yields and extended serum half-life of human interferon α2b expressed in tobacco cells as Arabinogalactan-protein fusion. Biotechnology and Bioengineering, 97(5), 997–1008. Xu, J., Tan, L., Lamport, D.T.A., Showalter, A.M., Kieliszewksi, M.J. (2008) The O-Hyp glycosylation code in tobacco and Arabidopsis and a proposed role of Hyp-glycans in secretion. Phytochemistry, 69(8), 1631–1640. Yariv, J., Rapport, M.M., Graf, L. (1962) The interaction of glycosides and saccharides with antibody to the corresponding phenylazo glycosides. Biochemical Journal, 85, 383–388. Yariv, J., Lis, H., Katchalski, E. (1967) Precipitation of arabic acid and some seed polysaccharides by glycosylphenylazo dyes. Biochemical Journal, 105, 1C–2C. Yasuhara, H. (2005) Caffeine inhibits callose deposition in the cell plate and the depolymerization of microtubules in the central region of the phragmoplast. Plant Cell Physiology, 46, 1083–1092. Ye, Z.-H., Varner, J.E. (1991) Tissue-specific expression of cell wall proteins in developing soybean tissues. Plant Cell, 3, 23–37. Zhao, Z.D., Tan, L., Showalter, A.M., Lamport, D.T.A., Kieliszewski, M.J. (2002) Tomato LeAGP-1 arabinogalactan-protein purified from transgenic tobacco corroborates the Hyp contiguity hypothesis. Plant Journal, 31(4), 431–444. Zhou, J., Rumeau, D., Showalter, A.M. (1992) Isolation and characterization of two wound-regulated tomato extensin genes. Plant Molecular Biology, 20, 5–17.

Annual Plant Reviews (2011) 41, 343–366 doi: 10.1002/9781444391015.ch14

Chapter 14

PLANT CELL WALL BIOLOGY: POLYSACCHARIDES IN ARCHITECTURAL AND DEVELOPMENTAL CONTEXTS Maureen C. McCann1 and J. Paul Knox2 1

Biological Sciences, Purdue University, West Lafayette, IN 47907, USA Centre for Plant Sciences, Biological Sciences, University of Leeds, Leeds LS2 9JT, UK

2

Manuscript received September 2009

Abstract: Land plant cell walls contain in the region of 11 major polysaccharides that are grouped as cellulose, hemicelluloses and pectins. Methodologies involving the use of molecular probes and advanced spectroscopies are revealing considerable diversity and complexity as to how these polysaccharides are configured within cell wall structures. Our knowledge of the integration of selected sets of polysaccharides into diverse and multifunctional primary and secondary cell walls is currently increasing although many challenges remain. In addition to understanding the structure and function of polysaccharides in the architectures of individual cell walls and in varied developmental contexts there is also a taxonomic dimension to cell wall polysaccharides that is likely to reveal insights into functions. Challenges for future studies include the understanding of aspects of possible polysaccharide redundancy during cell wall assembly and function, and of how specific polysaccharides are integrated with the mechanical attributes of cell walls and the placing of specific polysaccharide dynamics within physiological frameworks of regulation. Keywords: architecture; carbohydrate-binding modules; cell wall biology; microstructure; monoclonal antibodies; plant cell adhesion; spectroscopy.

Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

343

344

■

Plant Polysaccharides

14.1 Introduction Cell walls are central to many cell processes that are fundamental to plant growth and survival. Cell walls are also significant materials in the contexts of food, paper, polymers, textiles and fuel. Many of the sets of plant polysaccharides discussed elsewhere in this book are quantitatively the most important macromolecular components of cell walls. Our understanding of how these sets of polysaccharides are assembled to provide functionally diverse, mechanically flexible but robust sets of cell walls is the basis of this chapter. Cell wall polysaccharides are currently separated into cellulose, hemicelluloses and pectic polymers – groupings largely derived from both structural features and solubilization properties. The latter two groups contain polysaccharides of considerable structural diversity and complexity and, in addition, contain polymers that are subject to structural modulation within cell walls. It is well understood that primary and secondary cell walls largely dictate the morphology of plant cells and have considerable impacts upon the mechanical and functional properties of individual cells and organs (Evert 2006; Schopfer 2006). However, the integration of this cell-based understanding of how cell walls contribute to plant form and biology is not well integrated with the molecular understanding of the assembly of cell wall architectures and how these act and are adapted to accommodate or influence, for example, meristem development or organ flexibility and robustness. Although we have general inventories of the major structures found in cell wall polysaccharides, our knowledge of how they are assembled into patterns of diverse subcellular functional architectures, and of the dynamics of their modifications and molecular interactions during growth and development, is often limited. The biochemical complexity of cell wall polysaccharides has been widely explored and uncovered. The cell biological complexities of these polymers still require extensive exploration.

14.2 Plant cell wall biology basics Extendable primary cell walls, generating and resisting internal hydrostatic turgor pressures, enable multicellularity through the maintenance of cell-tocell attachments in addition to regulating cell and organ growth. Secondary cell walls provide tough internal strengthening to prevent the collapse of specific internal cell types such as xylem vessel elements. In structural terms, both primary and secondary cell walls are fibrous composites with several polysaccharides in common. Cellulose microfibrils form the fibrous component of both cell wall types and these are assembled in association with cellulose synthesis at plasma membranes (see Chapter 3). Tough cellulose microfibrils, stabilized by extensive internal hydrogen bonding, are present within cell wall layers,

Plant Cell Wall Biology

■

345

forming the basis of resistance to compressive and tensile forces. Cellulose microfibrils constrain turgor-pressure driven cell expansion to directions orthogonal to net microfibril orientations. The current view would be that it is the interlinking, intervening or interweaving polymers forming the cell wall matrices between the cellulose microfibrils that further determine, extend and direct the overall properties of cell walls (Cosgrove 2005; Thompson 2005). In primary cell walls pectin and hemicellulose matrix polymers mediate the regulated slippage apart of microfibrils when under tension, i.e. they mediate cell growth. A less hydrated packing of hemicellulose and often polyphenolic lignins can result in tough hydrophobic secondary cell wall materials that resist external forces. Primary cell walls will be the main focus here – as they are present in all plant cells that have a wall and they are central to cell and meristem development and plant growth – but facets of secondary cell wall structures are important in, for example, areas such as water transport, cell and organ biomechanics and cell wall exploitation. It is well established that the hemicelluloses attach and link cellulose microfibrils and this has led to these polymers being grouped together as ‘crosslinking glycans’. The major classes of hemicelluloses proposed to have microfibril binding and crosslinking properties are xyloglucans, xylans, mannans and mixed-linkage glucans. The specific distribution of structural variants of these polymers is taxonomically significant, with the first three classes widely distributed across the land plants (see Chapter 1). The mixedlinkage glucans are currently proposed to be restricted, within the land plants, to species of the Poales and also to the horsetails (Fry et al. 2008a; Sørensen et al. 2008). The major pectic polysaccharides are homogalacturonans (HG), the multidomain highly heterogeneous rhamnogalacturonan-I (RG-I) set of polymers (incorporating arabinans, galactans and arabinogalactans), the extreme heteropolymer rhamnogalacturonan-II (RG-II) and xylogalacturonans (Willats et al. 2001a; Mohnen 2008). Although heterogeneous, the major sets of pectic polysaccharides appear to occur widely across the land plants. It is becoming increasingly apparent that molecular compositions and configurations of cell wall polysaccharides differ between species and amongst organs, tissues and cell types of a single species. Indeed, the cell wall of a single cell has zones of differing composition implying varied functional architectures within the wall. The presence of varied patterns of cell wall polysaccharide occurrence surrounding cells and across tissues indicates the existence of precise targeting mechanisms for the delivery of polymers (or polymer-modifying enzymes) to distinct sites at plasma membranes and hence cell walls. In any consideration of polysaccharides in relation to cell wall biology, the major questions beyond polysaccharide synthesis and delivery to cell walls concern how polysaccharides are assembled into cell wall materials at cell surfaces and how the polymers interact within cell walls to make heterogeneous functional composites that are in turn

346

■

Plant Polysaccharides

assembled into multi-cell-wall systems ramifying throughout organs. In other words, how do the diverse configurations of polysaccharides that make up the cell walls in various cell types contribute to multiscale properties and diverse cell and organ functions?

14.3 Analytical tools to study cell wall microstructures and the diversity of cell wall architectures 14.3.1 Molecular probes The major tools and technologies used to assess and document the cell biological contexts of polysaccharides and cell wall diversity are molecular probes such as antibodies and microspectroscopies. Polysaccharide-directed antibodies can be used in conjunction with light and electron microscopies to detect polysaccharides in situ within cell wall materials but are also useful to study isolated cell wall polymers. Although specific antisera can be developed for polysaccharides, there is a trend to develop monoclonal antibodies that can be fully defined in terms of binding specificity and their extended availability allows use in a wide range of cell wall systems and research environments. Monoclonal antibodies are particularly useful for the detection and dissection of specific polysaccharide structural features in the context of complex polymers whether this complexity arises from the diversity of monosaccharides, polymer branching or non-glycan substitutions such as acetylation. Another important advantage of monoclonal antibodies is that the same probe can be used by all researchers so that a consensus concerning the occurrence and/or developmental dynamics of a particular structural feature or epitope of a polysaccharide can be established. Sets of cell wall-directed monoclonal antibodies are being developed and made available from various laboratories, e.g. Biosupplies, Australia (http:// www.biosupplies.com.au); CarboSource Services, Complex Carbohydrate Research Center, University of Georgia, USA (http://www.ccrc.uga.edu/ ∼carbosource/CSS_home.html); PlantProbes, University of Leeds, UK (http://www.plantprobes.net); and probes are collated at the WallBioNet: Plant Cell Wall Biosynthesis Research Network (http://glycomics.ccrc.uga. edu/wall/index.html). Most currently used monoclonal antibodies have been developed from cell-based rodent hybridoma procedures which appear to be the most effective route to monoclonal antibody generation for polysaccharides. However, the stimulation of an appropriate immune response and the isolation of an antibody to a target polysaccharide is not trivial and these issues have been discussed in more detail elsewhere (Willats & Knox 2003; Rao et al. 2008). Ideally, a polysaccharide-directed monoclonal antibody is developed from immunization with a defined oligo(poly)saccharide–protein conjugate and its precise binding specificity determined by oligosaccharide (hapten) inhibi-

Plant Cell Wall Biology

■

347

tion studies. This approach can be somewhat restricted by the limited availability of relevant oligosaccharides that are required for both of these aspects of antibody generation. Immunization of animals with complex mixtures of cell wall molecules can also be fruitful especially if combined with the rapid characterization of the binding profiles of the isolated antibodies using glycan microarray technologies (Moller et al. 2008). The difficulties and challenges confronting the generation of probes for cell wall polysaccharides are being overcome to some extent by the exploitation of the non-catalytic carbohydrate-binding module (CBM) domains of cell wall polysaccharide hydrolases (Boraston et al. 2004; Shoseyov et al. 2006). CBMs are a highly diverse set of protein modules which display wide binding specificities that include cellulose, xylans, mannans and pectins. CBM structures and specificities are collated at the Carbohydrate-Active Enzymes (CAZy) website (http://www.cazy.org). Recombinant His-tagged forms of CBMs are effective polysaccharide binding proteins and are directly amenable to antibody style assays and in situ labelling procedures (McCartney et al. 2004). CBMs have many advantages over antibodies, including the reduced requirement for high salt buffers and their smaller size. A further aspect of CBMs is that their sequences and structures are readily obtained which leads to the potential for engineering of both directly fluorescent protein tagged CBM molecules and also CBM specificities. A xylan binding member of CAZy family 4 CBM from Rhodothermus marinus has been engineered using a selection procedure to produce xyloglucan-directed CBMs (Gunnarsson et al. 2006). The transgenic expression of microbial cell wall polysaccharide-binding CBMs in plants can also be used to explore developmental mechanisms (Obembe et al. 2007). An emerging feature of CBM recognition of cell walls is the considerable variation and complexity observed in CBM families in binding to sets of cell wall polysaccharides both in vitro and in situ (Blake et al. 2006; McCartney et al. 2006; Hervé et al. 2009). In summary, these observations appear to indicate that these polysaccharide binding probes can discriminate between the contexts of polysaccharides when they are components of cell walls. Such observations indicate the considerable complexity and subtle diversity of cell wall structures, and CBMs will be valuable probes to explore and reveal this further. An important and challenging aspect of antibody/CBM binding patterns to cell walls is their interpretation. Does a restricted pattern of occurrence of a particular epitope indicate the absence of the polysaccharide, an altered polysaccharide structure or the masking of an epitope by other polymers? Treatment of cell walls in a section with a pectic homogalacturonandegrading enzyme can increase recognition of cellulose-directed CBMs (Blake et al. 2006). Moreover, recent work has demonstrated that pectic homogalacturonan can actively mask both xyloglucan (Marcus et al. 2008) and xylan (Hervé et al. 2009) detection in primary cell walls Examples of this unmasking by pectic HG degradation are shown in Fig. 14.1. The

348

■

Plant Polysaccharides

JIM5

e

PL/JIM5

LM15

e

cp

x e

cp

CBM15

PL/LM15

e

e

cp

cp

cp

x

x

x e

cp

PL/CBM15

e

cp

PL/LM11

Figure 14.1 Enzymatic treatment of sections can unmask polysaccharide epitopes for immunofluorescence detection. In transverse sections of tobacco stem sections treatment with pectate lyase (PL) results in the loss of the JIM5 pectic HG epitope, increased detection of the LM15 xyloglucan epitope particularly in epidermal (e) and cortical parenchyma (cp) cell walls (upper panels). A similar treatment exposes xylan polymers to detection by CBM15 and LM11 in primary cell wall regions associated with cell junction in the cortical parenchyma (lower panels). Micrographs with dashed edging indicate sections that were treated with PL prior to immunochemistry. x = xylem. Bars = 100 μm. Further information in Marcus et al. (2008) and Hervé et al. (2009).

possibilities of epitope masking, potentially indicating distinct and intimate polymer associations, will be an important matter for future studies of cell wall immunocytochemistry and cell wall biology. The full range of polysaccharide recognition capabilities of assembled panels of monoclonal antibodies and CBMs has recently been used in conjunction with glycan microarray methodologies to assess epitope/ligand

Plant Cell Wall Biology

■

349

occurrence in extracts of plant materials (Moller et al. 2007). This methodology, known as comprehensive microarray profiling (CoMPP), is a highthroughput, small-scale method for the assessment of the occurrence of specific structural features of cell wall polymers in organs or parts of organs and can be readily integrated with in situ analyses (Liepman et al. 2007; Moller et al. 2007; Singh et al. 2009). The refinement and extension of such approaches will be essential to allow the effective integration of knowledge of cell wall glycomes with other systematically collected data sets such as transcriptomes to support whole-system studies of specific plant processes. 14.3.2 Advances in microscopies Direct visualization of cell walls by fast-freeze, deep-etch, rotary-shadowed replicas has provided images that show native cell wall architectures, although pectins had first to be extracted (McCann et al. 1990). However, making replicas is technically difficult and other imaging technologies provide an easier route to imaging at least the microfibrils within cell walls. Field-emission scanning electron microscopy (FESEM) now provides the necessary resolution in scanning rather than transmission EM (Sugimoto et al. 2000). High-resolution atomic force microscopy (AFM) has the versatility to make measurements in air or under fluid. In a study of native maize primary cell walls, a range of microfibril morphologies and their varied interactions with matrix components has been observed using this technique (Ding & Himmel 2006) and clear AFM images of single cellulose chains can be obtained (Yokota et al. 2007). AFM can enable the study of the interactions and binding of individual molecules and thus has the potential to provide precise information to populate detailed models of cell wall architectures (Gunning et al. 2004; Gunning et al. 2009). Tomography is a technique that reconstructs the interior structure of an object by image collection associated with tilting the specimen grid in an electron beam. Dual-axis electron tomography has been used to investigate cellulose microfibrils in transmission EM, to a resolution of about 2 nm, resolving individual microfibrils and their 3D organization in the S2 layer of the walls of wood fibres (Xu et al. 2007). Such approaches will be important to determine the micromechanical properties of polysaccharides and the functional significance of the interactions and associations between polymer classes. High-resolution images of CBM binding (Ding et al. 2006) have been achieved by using quantum dots incorporated into a double His-tagged recombinant CBMs for direct imaging of individual cellulose microfibrils within secondary cell walls. Water-soluble and highly luminescent quantum dots of CdSe/ZnSe bind five histidines at the zinc surface while retaining desirable electronic properties for imaging at the EM level. Highly luminescent quantum dots bound to double His-tags in recombinant proteins has allowed the binding of single CBMs to cellulose microfibrils to be imaged at close to molecular resolution. Dual-axis tomography tilts the specimen

350

■

Plant Polysaccharides

around two orthogonal axes, and then combines the reconstructions of the two single-axis tilt series to reduce artefacts and improve resolution. 14.3.3 Spectroscopic and spectrometric technologies Raman and mid-infrared microscopies provide complementary techniques for chemical imaging of plant cell walls. Raman scattering depends on changes in the polarizability of functional groups due to molecular vibration, while infrared absorption depends on changes in intrinsic dipole moments. Infrared chemical imaging provides cellular resolution of cell wall compositions in tissue sections (Barron et al. 2005) using a step-scanning spectrometer equipped with a camera and a focal plane array detector. The spatial resolution is limited, but improved resolution is obtained with Raman imaging although sample fluorescence can be a problem. Laser-based confocal Raman microscopy has been used to generate intensity maps of lignin and cellulose distribution in cell wall layers of black spruce wood, resolving signals from cell corners, the middle lamella, and different S layers within the secondary wall (Agarwal 2006). The Raman effect is inherently weak. Wood tissues with abundant secondary cell walls have sufficient density to give enough Raman scattering to obtain a spectrum but this is not true of primary cell walls. However, recent developments in surface-enhanced Raman spectroscopy, with application of small silver or gold colloids to samples, or to the tip of a probe in contact with the sample, provides the opportunity for local signal amplification and much improved Raman signal detection (Ryder 2005). Combinations of visible and Raman imaging with AFM in a single instrument have also been advantageous because of not having to displace the sample (Osterberg et al. 2006).

14.4 Cell wall architectures: primary cell walls Primary cell walls have the ability to withstand tensile forces and thus they generate and resist turgor pressure. Shared outer regions of primary cell walls known as middle lamellae form the interface between neighbouring cells and these are the cell wall domains directly responsible for cell adhesion, tissue cohesion and multicellular organ robustness. A wide range of structurally diverse polysaccharides is present in primary cell walls, allowing a range of potential interactions and crosslinking possibilities. Primary cell wall loosening has been the major focus of the study of cell expansion. Primary cell walls are extendable in highly controlled ways and therefore mechanisms must exist that allow discrete loosening of the cell wall matrix, permitting microfibril separation. At the same time sets of newly synthesized cellulose, hemicellulose and pectic polymers are integrated into expanding regions of cell walls. In an anisotropically extending cell only certain cell wall regions, e.g. longitudinal cell walls in an elongating cell,

Plant Cell Wall Biology

■

351

will be undergoing active extension and assembly processes. It is of interest that few instances of cell-based studies of polysaccharide heterogeneity have detected molecular distinctions among the major polysaccharides that relate directly to the differential wall extension around an individual cell. Hemicelluloses, or crosslinking glycans, are a class of polysaccharides that can hydrogen-bond to cellulose microfibrils: they may coat microfibrils but are also long enough to span between microfibrils and link them together to form a network. The two major crosslinking glycans of primary cell walls of flowering plants are xyloglucans and (glucuronoarabino)xylans. The precise details of how hemicelluloses link through hydrogen bonding to the surface of cellulose microfibrils and variations between hemicellulose polysaccharide classes are still largely open questions. Microscopy involving pectin removal and study of remaining cellulose and hemicellulose polymers has led to widely accepted ideas of coextensive networks in which the interlinked load-bearing cellulose–hemicellulose (crosslinking glycan) networks are embedded in a network of pectic polymers (McCann et al. 1990). This has been a useful, but perhaps simplistic view and has to be considered in the light of recent evidence indicting that xyloglucans can be covalently attached to pectin (Popper & Fry 2008) and that arabinan and galactan side chains (associated with pectic polymer RG-I) can attach to cellulose microfibrils in a similar manner to hemicelluloses (Zykwinska et al. 2008). Xyloglucans and xylans display a range of structural variants and these have taxonomic significance (Carpita 1996; O’Neill & York 2003; Peña et al. 2008). These factors, which are largely backbone substitutions by single residues or short oligosaccharide chains, but also acetylation, will alter properties, solubility and processing by enzymes. Major questions that require clarification for the hemicellulose groups of polymers are the functional significance of structural variants and the extent of multifunctionality and redundancy among this group of polymers. Recent work exploring the microstructure of tobacco stem primary cell walls has indicated that both xyloglucan and xylan epitopes can be detected in patterns of differing occurrences within primary cell wall types that may reflect roles other than cell expansion (Marcus et al. 2008; Hervé et al. 2009). It is clearly possible that hemicellulose polymers have roles beyond the crosslinking of cellulose microfibrils and the regulation of cell expansion. Genetic manipulation of xyloglucan synthesis and the isolation of Arabidopsis plants with little detectable xyloglucan appears to have limited impacts on morphology and growth in standard experimental conditions (Cavalier et al. 2008). This could indicate robust compensatory mechanisms in plant cell wall assembly (Ryden et al. 2003; Manfield et al. 2004) and possibly the interchangeability in hemicellulose polymer classes alluded to above. The mannan set of polymers appears to be widespread in primary cell walls but with varying relative abundances and structures. There are intriguing hints that mannans may play important developmental roles (Handford et al. 2003; BenˇováKákošová et al. 2006; Liepman et al. 2007). Key questions here are the reasons

352

■

Plant Polysaccharides

for such a range of hemicellulose polymers and the function roles of their modulation within cell walls. To what extent can one polysaccharide class, say xylans, do the job of another, say xyloglucans? 14.4.1 Modulation of cellulose–hemicellulose networks and cell enlargement Sets of wall-modifying enzymes can influence cellulose–hemicellulose networks. To date these have been largely studied in relation to cell growth but are also important for other processes such as cell wall dissolution at fruit ripening. Xyloglucan endo-transglycosylases (XET) carry out the transglycosylation of xyloglucan where one chain of xyloglucan is cleaved and reattached to the non-reducing terminus of another chain (see Chapter 2). Given such a mechanism, microfibrils could undergo a transient slippage but overall tensile strength of the interlocking xyloglucan matrix would not diminish. XETs may also function in the realignment of xyloglucan chains in different wall strata during growth, and in assembly of the wall as newly synthesized xyloglucans are incorporated. A barley XET has recently been identified to link xyloglucans to mixed-linkage glucans or cellulosic substrates, raising the possibility that XETs might link different polysaccharides in vivo (Hrmova et al. 2007). Enzyme activity favouring mixed-linkage glucan grafting to xyloglucans has also been identified in Equisetum species (Fry et al. 2008b). These novel transglycosylation activities have the potential to generate novel heteropolymeric networks that impact upon extendable cell wall architectures. In this context, a tomato endo-mannanase has been renamed mannan transglycosylase/hydrolase because it can carry out a transglycosylation reaction in the presence of mannan-derived oligosaccharides (Schröder et al. 2004; Schröder et al. 2009). Expansins catalyse wall extension in vitro without any detectable hydrolytic or transglycolytic events. Expansins act to reduce hydrogen bonding by an unknown mechanism, although this has been proposed to be the breakage of hydrogen bonds between cellulose and the load-bearing crosslinking glycans (Cosgrove 2005). Such an activity would disrupt the tethering of cellulose by hemicelluloses in a range of walls. Local transient induction of expansin expression on the flank of developing primordia leads to the induction of ectopic lamina tissue and modulation of leaf shape (Pien et al. 2001). Such observations demonstrate that cell walls and organ biophysical properties are intimately linked into developmental events. In growing organs the deposition of new cell wall material occurs uniformly along the expanding sides of cells. However, tip growth, the growth and deposition of new wall material strictly at the tip of a cell, occurs in some unadherent cell wall regions, notably root hairs, moss protonemata and pollen tubes. These cells can grow at prodigious rates (in excess of 200 nm/s for pollen tubes). The site of expansion in tip-growing cells is focused to the dome-shaped apex. Expansins are located specifically to the region of the trichoblast (hair-forming) cell wall that bulges out to initiate

Plant Cell Wall Biology

■

353

the root hair (Baluska et al. 2000). The localized wall loosening in the bulge by expansin may activate plasma membrane stretch-activated Ca2+ channels, transducing the biomechanical signal into a signal that can be interpreted by the cytoskeleton. A maize pollen allergen belonging to the expansin B family binds strongly to xylans, causing swelling of the maize cell wall, perhaps by disrupting arabinoxylan–cellulose interactions (Yennawar et al. 2006). The crystal structure of EXP B1 shows an unstructured glycosylated N-terminal extension and a two-domain folded structure with a highly conserved open surface spanning the two domains. The surface has many aromatic and polar residues suitable for binding a branched polysaccharide of about ten residues in length. The authors present an attractive hypothesis that expansin uses the strain energy stored in a taut cellulose-binding glycan to dissociate the glycan from the cellulose surface, using a 10° shift in angle between domains to cause a one-residue dislocation of the polysaccharide along the binding surface. Other factors that can influence cell wall loosening and extensibility include yieldin proteins that have been shown to directly affect the yield threshold of cell walls. Yieldins are homologous to acidic class III endochitinases and concanavalin B, and promote cell wall extension under acidic conditions (Okamoto-Nakazato 2002). Work also continues to determine the roles played by reactive oxygen species in primary cell wall extension processes and to identify the key polymers that are targeted by these factors (Schopfer & Liszkay 2006). 14.4.2 Pectins: modular multifunctional polymers of the primary cell wall matrix Pectins are galacturonic acid-rich polysaccharides – a definition that does not convey their extensive structural complexity nor their multifunctional contributions to plant growth processes. It appears that HG, RG-I, RG-II and xylogalacturonan pectic domains form a 3D crosslinked hydrated network coextensive with the cellulose–hemicellulose network and contribute to the regulation of cell wall extension, porosity, hydration, pH and mechanics. These factors in turn can regulate the access or activities of cell wall modifying enzymes to either the pectin network or the cellulose–hemicellulose network. Pectin cell biology has been covered extensively elsewhere (Willats et al. 2001a; Verhertbruggen & Knox 2007). HG is quantitatively the major component of pectin. The methylesterification of HG determines to a large extent the gelation characteristics of pectin: both the amount and the distribution of methyl groups on the HG backbone are important for interactions in the primary cell wall and also cell wall porosity. HG is secreted in a highly methyl-esterified form, and pectin methylesterase (PME) enzymes located in the cell wall cleave methyl groups to initiate binding and crosslinking by Ca2+ and other ions. Pectic polysaccharides appear to be essential for cell wall expansion and more highly methyl-esterified HG components potentiate this with crosslinked

354

■

Plant Polysaccharides

de-esterified HG contributing to wall tightening processes (Derbyshire et al. 2007). Pectic polymer contributions to cell wall assembly and mechanics and other processes are often hints and remain uncertain. The demonstration that pectic network components can also be attached to the cellulose– hemicellulose network by means of both covalent and non-covalent links leads to questions of load-bearing roles for pectic networks in overall primary cell wall architectures and integration with hemicellulose roles in these processes. This leads to notions of distinct populations of polymers in different aspects of architecture or functions – an idea that can also be extended to the hemicellulose polymers. Two molecules of RG-II, an elaborated HG domain, can complex with boron, forming a borate-diol ester, and can crosslink two pectin chains. The widespread occurrence of RG-II in the plant kingdom and its structural conservation indicate a distinct role in primary cell wall integrity for this constituent of pectin (O’Neill et al. 2004). Pectins are hybrid molecules and pectic HG can be covalently attached to RG-I domains within the matrix, although how these supramolecular networks are assembled and organized is still uncertain (Coenen et al. 2007). Recent work with the qua2 mutant indicates that HG and RG-I domains can have opposite effects on the hydrodynamic properties of pectins, and reduced relative levels of HG increase pectic polymer flexibility (Ralet et al. 2008). Within RG-I domains, neutral polymers (complex sets of arabinans, galactans and arabinogalactans) are attached at one end to the RG backbone, but extend into, and are highly mobile within, the wall space. Studies indicating that these side chains have the capacity to bind to cellulose microfibrils in the manner of hemicelluloses (Zykwinska et al. 2008) show that these domains clearly have the potential to have contrasting influences on overall wall properties. That the varied side chains can be trimmed by hydrolases provides a mechanism to influence these properties, although the roles of RG-I processing in terms of local wall properties and development are not well understood (Chávez-Montes et al. 2008; Verhertbruggen et al. 2009). In addition to RG-I domains attaching to microfibrils, biochemical evidence is also accumulating that pectins are covalently attached to the xyloglucans (Popper & Fry 2008). The complex domain nature of pectins, the diverse ways in which pectic polymers can be interlinked and the range of possibilities for interaction with other wall polymers indicate considerable capacity to impact on cell wall properties that is not yet fully understood. Although primary cell walls are largely polysaccharide composites, smaller amounts of glycoproteins appear to be important within wall structural frameworks. These proteins are encoded by large gene families and are developmentally regulated, with relative amounts varying among tissues and species (Johnson et al. 2003). Extensins are one of the best-studied sets of hydroxyproline-rich glycoproteins. Deposition and crosslinking of extensins is thought to help increase the mechanical strength of the cell wall and the oxidative crosslinking of an extensin has been shown to be important in forming a scaffold for cell plate assembly in arabidopsis (Cannon

Plant Cell Wall Biology

■

355

et al. 2008). Arabinogalactan-proteins (AGPs), although not quantitatively important polymers, carry complex branched heteroglycans that appear to be hypervariable in structural terms; this seems to be intimately tied in with cell development, but mechanistic understanding is scant. This group of polymers, located in the plasma membrane and cell wall, is implicated in several plant cell biological processes and AGP glycan domains also seem to be important factors in a range of chimeric proteins, presumably imparting distinct properties to these proteins in the cell wall environment (Seifert & Roberts, 2007).

14.5

In vitro polysaccharide composites

Primary cell walls are adapted to withstand tensile stresses whereas secondary walls are constructed to withstand compression (Jarvis & McCann 2000). To understand how stresses are resisted, we need to know how walls deform under load and how specific constituents of cell walls contribute to properties. Linking molecular knowledge of cell wall structures with an understanding of cell wall mechanical effects and properties and integrating this into models for multicellular structures in order to predict tissue-level and organ-level effects is currently one of the most challenging areas of cell wall polysaccharide biology, if not plant biology as a whole. Synthetic composites of cellulose with xylogucans or other polysaccharides have been made in vitro as an approach to understanding cell wall assembly and properties and to allow the biomechanical properties conferred by individual components to be measured (Chanliaud et al. 2004). For example, cellulosic pellicles of Acetobacter xylinus spontaneously form composites with other polysaccharides present in the culture medium that can be sampled for biomechanical tests. Enzymic treatment of cellulose/xyloglucan composites with a xyloglucan endo-hydrolase increased stiffness of the pellicle, while treatment with an XET enhanced viscoelasticity (Chanliaud et al. 2004). Xyloglucans with reduced galactose produced pellicles with low stress and breaking strain compared with cellulose alone, likely as an effect of xyloglucan aggregation introducing weak spots. The mechanical behaviour of the composite is consistent with observations of the reduced tensile strength of Arabidopsis mutant seedlings compromised in xyloglucan galactosylation (Ryden et al. 2003; Peña et al. 2004). Incorporation of pectins in such in vitro synthetic systems has also provided insight into their possible roles in cell wall assembly and properties (Chanliaud et al. 2002).

14.6

Cell wall diversity

An aspect of plant cell walls that is to some extent overlooked when models of cell wall architecture are considered is that of their structural heterogeneity and diversity. The molecular diversity of cell walls can be considered at

356

■

Plant Polysaccharides

three levels. The first of these is cell wall microstructure, i.e. how the component polysaccharides vary in structure and occurrence at the subcellular level within a single wall. The second is how the sets of primary and secondary cell walls, as components of diverse cell types and tissues, differ within an organ or plant. For example, do all parenchyma systems within an organ have the same configuration of cell wall polysaccharides? The third level is the variation in these factors between species, e.g. how similar are the cell walls in the epidermal or pith tissues of different species? All these levels of consideration are at an early stage of understanding and there are significant gaps in our knowledge, but some broad aspects of cell wall diversity are well established (Harris 2005; Knox 2008). The panels of cell wall-directed probes and spectroscopies discussed above can enable high-resolution imaging of cell wall diversity and indicate significant diversity at all three levels. Wall zones that to date have been detected to be molecularly distinct in terms of the structural features of polysaccharides present (epitopes detected), include inner and outer cell wall layers, middle lamellae, plasmodesmata, pit fields, primary and secondary cell wall thickenings, the cell wall lining, and materials filling intercellular spaces. These documented molecular cell wall domains largely involve the pectic sets of polymers. Pectic HG is clearly locally remodelled, initially by pectin methyl-esterases. At a cell level this is currently most clear for cell walls lining intercellular spaces where the LM7 HG epitope is widely found (Willats et al. 2001b) and as shown in Fig. 14.2. This observation highlights that cell-to-cell links are extremely poorly understood but are in fact important features of organ biomechanics and integrity. Pectic HG is widely implicated in cell adhesion maintenance and its loss and this abundant pectic domain, at a wider level, appears to be a focus for complex regulatory networks that can impact on primary cell wall attachments, mechanics, physiology and defence. The orchestration of HG structures across the varied landscape of primary cell walls will influence cell extensibility, porosity and mechanical properties as well as cell adhesion. Key factors here are how sets of enzymes (e.g. pectin methyl-esterases, polygalacturonases, pectate lyases – all with large gene families) can act on HG polymers to modulate HG structure and hence wall strength, wall porosity and middle lamellae crosslinking, all in a local spatially regulated manner. No direct adhesive role has yet been shown for the LM7 epitope and it may be involved in maintaining a microenvironment to allow other aspects of cell wall biochemistry to function in adhesive or separation events. It is likely that other aspects of cell wall microstructure in these regions are yet to be visualized. A xyloglucan epitope has been specifically detected at intercellular spaces of pith parenchyma cell walls and whether this is due to restricted occurrence or selective masking needs to be determined (Marcus et al. 2008). Another pectic epitope, a specific structure of xylogalacturonan that is not carried by xylogalacturonans of parenchyma systems, is of interest in that it relates to another cell separation process; in this case cell detachment from plant bodies, as seen in inner seed coat of legumes and root caps (Willats

Plant Cell Wall Biology

A

B i

i

■

357

C l

*

l i

Intact tomato fruit parenchyma

LM7 HG

TS hemp sclerenchyma fibre

LM11 xylan

TS hemp sclerenchyma fibre

LM16 RGI

Figure 14.2 Three examples of cell wall microheterogeneity. (A) Immunofluorescence detection of the LM7 HG epitope at the surface of an intact cell isolated from tomato fruit parenchyma. The epitope is present at cell wall regions that lined former adhesion planes (*) and intercellular space (i). (B) Immunofluorescence detection of the LM11 xylan epitope in outer cell wall (primary/S1) layers of a transverse section of hemp stem sclerenchyma fibre. Double headed bars indicate cell wall (primary + secondary) thicknesses between middle lamellae and cell lumen (l). (C) Equivalent cell to B showing the LM16 RG-I epitope in regions of inner cell wall layers. Arrowheads indicate immunofluorescence in each case. Bars = 10 μm. Image of tomato cell provided by J. J. Ordaz-Ortiz.

et al. 2004). This epitope and associated detached cell properties has also recently been detected at very specific locations in Arabidopsis flowers and siliques – locations identified from CoMPP analyses (Moller et al. 2007). A major gap in our understanding of cell-to-cell attachments is how middle lamellae polymers are integrated with primary cell wall polymers – there must be attachments across all cell wall layers between two adjacent cells. There is evidence that RG-II affects wall thickness by forming borate diesterlinked dimers and is important for growth and cell attachments (O’Neill et al. 2004). In addition, arabinans are implicated in cell adhesion events even though they can be undetectable in middle lamellae (Orfila et al. 2001). Detailed studies of cell wall microstructures, polysaccharide crosslinks and masking in the contexts of cell adhesion and its loss should provide insight into these topics. A major cell wall zone to be associated with a widely occurring modulation of RG-I structures appears to cell wall regions at pit fields and particularly the absence of the LM5 galactan epitope (Orfila & Knox 2000). The RG-I domain has been widely assessed in terms of cell biology by LM5 galactan and LM6 arabinan monoclonal antibodies (Willats et al. 2001a). The RG-I sets of polymers have been implicated in the control of mechanical properties and discussed elsewhere (McCartney et al. 2000; Willats et al. 2001a, Ulvskov et al. 2005) and there is a compelling case that RG-I-associated arabinan is required for guard cell wall flexibility in a range of species (Jones et al. 2005).

358

■

Plant Polysaccharides

Considerable work has studied developmental aspects of pectins in relation to cell development processes and a complex picture is emerging with extensive developmental regulation of pectic HG, RGI and xylogalacturonan sets of polymers. The demonstration that a localized de-methyl-esterification of HG is important for floral primordial growth in Arabidopsis (Peaucelle et al. 2008) is a clear indication for the importance of pectic HG and its methyl-esterification status. The Arabidopsis root is a useful system and amenable to direct immunolabelling of cell wall polysaccharides at intact root surfaces. An intriguing observation is the transient developmental occurrence of the LM5 1,4-galactan epitope in ground and epidermal tissues of the Arabidopsis seedling root that marks the onset of the acceleration of cell elongation in this organ (McCartney et al. 2003). The extent of this occurrence is modulated by factors that influence root growth. The patterns of detection of the LM5 galactan epitope at both subcellular (absent from pit fields) and tissue levels (transient occurrence in epidermal/cortical walls) indicate multiscale impacts on wall and cell properties. The CCRCM1 fucosylated epitope of xyloglucan has been similarly detected in relation to cell development in the arabidopsis root apex (Freshour et al. 2003). The identification of masking of xyloglucan epitopes (Marcus et al. 2008) may require both these cell development related patterns to be re-examined or reinterpreted, but whether a polysaccharide epitope is masked or detectable still indicates differences in cell wall structure. Arabinans are an intriguing but poorly understood set of cell wall polymers. They are often associated with pectic RG-I, but can also occur as nonconjugate polymers and as structural motifs in AGP glycans, and appear to be subject to complex processing in cell walls (Jones et al. 2005; Lee et al. 2005; Chavez-Montes et al. 2008; Verhertbruggen et al. 2009). Arabinans are abundantly associated with meristems (Skjøt et al. 2002; Verhertbruggen & Knox 2007) and degradative intervention can lead to serious disruption of meristem function but the mechanistic basis of this is far from clear (Skjøt et al. 2002; Ulvskov et al. 2005). Genetic dissection of arabinan synthesis is progressing (Harholt et al. 2006) but disruption of putative arabinan synthase genes belonging to CAZy family GT47 (Chapter 10) has been observed to have little impact on growth. Novel antibody tools that can provide insight into arabinan processing indicate that that this appears to be intimately associated with diverged cell function as well as showing intriguing differences between species (Verhertbruggen et al. 2009). A wider study of arabinan structures in the context of taxonomic contexts could be rewarding.

14.7 Cell wall architectures: secondary cell walls Secondary cell walls are currently viewed as having simpler composite structures with reduced levels of matrix polysaccharides and a less open and less hydrated structure in comparison with primary cell walls. Secondary

Plant Cell Wall Biology

■

359

cell walls generally have little or no pectic polysaccharides, contain xylans and or mannans as the main hemicellulose and display varied orientations of cellulose microfibrils in the three distinct layers that are often seen in secondary cell walls (Harris 2005). A distinguishing feature of some cells that generate secondary cell walls is the incorporation of lignins, complex networks of aromatic phenylpropanoids, that encrust all cell surface polysaccharide layers including middle lamellae. The composition of poplar secondary cell walls has been analysed and contrasted with primary cell wall composition: secondary cell walls are relatively enriched in cellulose and xylans, lignin, some structural proteins and minor amounts of pectins (Mellerowicz et al. 2001). The detailed cell biological analysis of polysaccharide occurrence has not been extensively documented for a wide range of secondary cell walls in diverse cell types such as xylem vessel elements, fibres and sclereids. The observations that have been made indicate diversity in composition. For example, the secondary cell walls of phloem-associated sclerenchyma fibres and xylem vessel elements in the stem of flax are clearly different in that only the phloem fibres appear to have abundant pectic galactan (Gorshkova & Morvan 2006) and the sclerenchyma fibres are distinct from those of industrial hemp where the LM5 epitope could not be detected (Blake et al. 2008). In the fibre cells that develop to control the coiling of tendrils in redvine (Brunnichia ovata), the LM5 and RG-I epitopes are detected only in the primary cell wall and inner layer of the secondary cell walls (Meloche et al. 2007; Bowling & Vaughn 2009). The LM5 and LM6 RG-I epitopes are present in cotton fibres and appears to be associated with peripheral primary cell walls that are important in early development (Singh et al. 2009). In flax and hemp xylan epitopes are detectable in phloem fibre cell peripheries in primary cell wall/S1 layers (Blake et al. 2008) whereas the fibre cells of coiling tendrils display xylan epitopes throughout secondary cell wall layers (Bowling & Vaughn 2009) which is usually seen for xylem cells. Mannan epitopes are abundant in the fibres of Vicia faba pistils and xylem (Chen et al. 2006) and xylem in gymnosperms (Harris 2005) as well at the periphery of hemp phloem fibres (Blake et al. 2008). Complex patterns of pectin and AGP epitope occurrence can be detected in relation to secondary cell wall structures (Blake et al. 2008; Bowling & Vaughn 2008) as shown in Fig. 14.2. Major questions to be answered here are how the molecular differences between, say, phloem fibres of different species, relate to fibre development and/or fibre properties. As cell wall biological methods develop, a more comprehensive view of the common and specific microstructural features of primary and secondary cell walls (e.g. Bowling & Vaughn 2008) and their diversity and functions will emerge. There may not always be a clear division between a cell with a secondary cell wall and a thickened primary cell wall. For example, the non-lignified secondary cell walls of certain seeds can be enriched in galactoglucomannans, pectic galactan or xyloglucan as storage polysaccharides

360

■

Plant Polysaccharides

(Harris 2005; Harris & Smith 2006). The thickened cell walls of the macrosclereid cell layer in pea testa contain abundant pectic HG (McCartney & Knox 2002). The uneven thickening of primary cell walls seen in collenchyma cells has not been extensively studied in terms of molecular components. In collenchyma of the tomato petiole, the LM5 galactan epitope has been observed to be restricted to inner cell walls (Willats et al. 2001a) and variations in cellulose structure in the wall thickenings revealed by CBM binding to celery petiole collenchyma cell walls (Blake et al. 2006). A survey of the thickenings of ingrowths seen in epidermal transfer cells of Vicia faba cotyledons has indicated that they are primary cell wall like in composition (Vaughn et al. 2007). Progress is being made in dissecting the molecular basis and control of a plant cell’s switch from primary to secondary cell wall formation (Zhong et al. 2006).

14.8 Prospects for plant cell wall biology We are on the verge of a rapid expansion in our understanding of the cell biology of the complex sets of polysaccharides that comprise plant cell walls. As we develop tools to map cell wall microheterogeneity we can perceive their considerable molecular complexity and diversity. Current major gaps in our knowledge of cell walls in relation to growth and development include systematic knowledge of the precise configurations and interactions of polysaccharides within the diverse cell walls of a growing organ, the functions of individual polysaccharides within a specific wall composites and also how the structural variants of the polysaccharide classes variedly influence cell wall architectures and properties. Inherent in all these studies will be the elucidation of regulatory and controlling mechanisms supporting the assembly of functionally specific wall architectures. At the level of organs, major gaps in knowledge include how cell and tissue wall architectures are integrated into the generation of organ mechanical properties and the nature of the involvement of cell walls in responses to diverse environmental and mechanical impacts.

Acknowledgements JPK acknowledges funding support from the United Kingdom’s Biotechnology and Biological Sciences Research Council and Department of Trade and Industry and the European Union research framework programmes.

References Agarwal, U.P. (2006) Raman imaging to investigate ultrastructure and composition of plant cell walls: Distribution of lignin and cellulose in black spruce wood (Picea mariana). Planta, 224, 1141–1153.

Plant Cell Wall Biology

■

361

Baluska, F., Salaj, J., Mathur, J., et al. (2000) Root hair formation: F-actin-dependent tip growth is initiated by local assembly of profiling-supported F-actin meshworks accumulated within expansin-enriched bulges. Developmental Biology, 227, 618–632. Barron, C., Parker, M.L., Mills, E.N.C., Rouau, X., Wilson, R.H. (2005) FTIR imaging of wheat endosperm cell walls in situ reveals compositional and architectural heterogeneity related to grain hardness. Planta, 220, 667–677. Benˇová-Kákošová, A., Digonnet, C., Goubet, F., et al. (2006) Galactoglucomannans increase cell population density and alter the protoxylem/metaxylem tracheary element ration in xylogenic cultures of Zinnia. Plant Physiology, 142, 696– 709. Blake, A.W., Marcus, S.E., Copeland, J.E., Blackburn, R.S., Knox, J.P. (2008) In-situ analysis of cell wall polymers associated with phloem fibre cells in stems of hemp, Cannabis sativa L. Planta, 228, 1–13. Blake, A.W., McCartney, L., Flint, J.E., et al. (2006) Understanding the biological rationale for the diversity of cellulose-directed carbohydrate-binding modules in prokaryotic enzymes. Journal of Biological Chemistry, 281, 29321–29329. Bowling, A.J., Vaughn, K.C. (2008) Immunocytochemical charcterization of tension wood: gelatinous fibers contain more than just cellulose. American Journal of Botany, 95, 655–663. Bowling, A.J., Vaughn, K.C. (2009) Gelatinous fibers are widespread in coiling tendrils and vines. American Journal of Botany, 96, 719–727. Boraston, A.B., Bolam, D.N., Gilbert, H.J., Davies, G.J. (2004) Carbohydrate-binding modules: fine tuning polysaccharide recognition. Biochemical Journal, 382, 769–781. Cannon, M.C., Ternus, K., Hall, Q., et al. (2008) Self-assembly of the plant cell wall requires an extensin scaffold. Proceedings of the National Academy of Sciences of the U S A, 105, 2226–2231. Carpita, N.C. (1996) Structure and biogenesis of the cell walls of grasses. Annual Review of Plant Physiology and Plant Molecular Biology, 47, 445–476. Cavalier, D.M., Lerouxel, O., Neumetzler, L., et al. (2008) Disrupting two Arabidopsis thaliana xylosyltransferase genes results in plants deficient in xyloglucan, a major primary cell wall component. Plant Cell, 20, 1519–1537. Chanliaud, E., Burrows, K.M., Jeronimidis, G., Gidley, M.J. (2002) Mechanical properties of primary plant cell wall analogues. Planta, 215, 989–996. Chanliaud, E., De Silva, J., Strongitharm, B., Jeronimidis, G., Gidley, M.J. (2004) Mechanical effects of plant cell wall enzymes on cellulose/xyloglucan composites. Plant Journal, 38, 27–37. Chávez Montes, R.A., Ranocha, P., Martinez, Y., et al. (2008) Cell wall modifications in arabidopsis plants with altered α-l-arabinofuranosidase activity. Plant Physiology, 147, 63–77. Chen, W., Stoddard, F.L., Baldwin, T.C. (2006) Developmental regulation of mannan, arabinogalactan-protein, and pectic epitopes in pistils of Vicia faba (faba bean). International Journal of Plant Science, 167, 919–932. Coenen, G.J., Bakx, E.J., Verhoef, R.P., Schols, H.A., Voragen, A.G.J. (2007) Identification of the connecting linkage between homo- or xylogalacturonan and rhamnogalacturonan type I. Carbohydrate Polymers, 70, 224–235. Cosgrove, D.J. (2005) Growth of the plant cell wall. Nature Reviews in Molecular and Cell Biology, 6, 850–861.

362

■

Plant Polysaccharides

Derbyshire, P., McCann, M.C., Roberts, K. (2007) Restricted cell elongation in Arabidopsis hypocotyls is associated with a reduced average pectin esterification level. BMC Plant Biology, 7, 31–42. Ding, S.-Y., Himmel, M.E. (2006) The maize primary cell wall microfibril: A new model derived from direct visualization. Journal of Agriculture and Food Chemistry, 54, 597–606. Ding, S.-Y., Xu, Q., Ali, M.K., Baker, J.O., et al. (2006) Versatile derivatives of carbohydrate-binding modules for imaging complex carbohydrates approaching the molecular level of resolution. Biotechniques, 41, 435–443. Evert, R.F. (2006) Esau’s Plant Anatomy, 3rd edition. John Wiley, New Jersey. Freshour, G., Bonin, P., Reiter, W.-D., Albersheim, P., Darvill, A.G., Hahn, M.G. (2003) Distribution of fucose-containing xyloglucans in cell walls of the mur1 mutant of Arabidopsis. Plant Physiology, 131,1602–1612. Fry, S.C., Nesselrode, B.H.W.A., Miller, J.G., Mewburn, B.R. (2008a) Mixed-linkage (1→3,1→4)-β-d-glucan is a major hemicellulose of Equisetum (horsetail) cell walls. New Phytologist, 55, 240–252. Fry, S.C., Mohler, K.E., Nesselrode, B.H.W.A., Franková, L. (2008b) Mixed-linkage glucan: xyloglucan endotransglucosylase, a novel wall-remodelling enzyme from Equisetum (horsetails) and charophytic algae. Plant Journal, 55, 240–252. Gorshkova, T., Morvan, C. (2006) Secondary cell-wall assembly in flax phloem fibres: role of galactans. Planta, 223, 149–158. Gunnarsson, L.C., Zhou, Q., Montanier, C., Karlsson, E.N., Brumer, H. 3rd., Ohlin, M. (2006) Engineered xyloglucan specificity in a carbohydrate-binding module. Glycobiology, 16, 1171–1180. Gunning, A.P., Bongaerts, R.J., Morris, V.J. (2009) Recognition of galactan components of pectin by galectin-3. FASEB Journal, 23, 415–424. Gunning, A.P., Kirby, A.R., Mackie, A.R., Kroon, P., Williamson, G., Morris, V.J. (2004) Watching molecular processes with the atomic force microscope: dynamics of polymer adsorption and desorption at the single molecule level. Journal of Microscopy, 216, 52–56. Harris, P.J. (2005) Diversity in plant cell walls. In: Plant Diversity and Evolution: Genotypic and Phenotypic Variation in Higher Plants (ed. R.J. Henry), pp 201–227. CAB International Publishing, Wallingford. Harris, P.J., Smith, B.G. (2006) Plant cell walls and cell-wall polysaccharides: structures, properties and uses in food products. International Journal of Food Science and Technology, 41, 129–143. Handford, M.G., Baldwin, T.C., Goubet, F., Prime, T.A., Miles, J., Yu, X., Dupree, P. (2003) Localisation and characterisation of cell wall mannan polysaccharides in Arabidopsis thaliana. Planta, 218, 27–36. Harholt, J., Jensen, J.K., Sørensen, S.O., Orfila, C., Pauly, M., Scheller, H.V. (2006) ARABINAN DEFICIENT 1 is a putative arabinosyltransferase involved in biosynthesis of pectic arabinan in Arabidopsis. Plant Physiology, 140, 49–58. Hervé, C., Rogowski, A., Gilbert, H.J., Knox, J.P. (2009) Enzymatic treatments reveal differential capacities for xylan recognition and degradation in primary and secondary plant cell walls. Plant Journal, 58, 413–422. Hrmova, M., Farkas, V., Lahnstein, J., Fincher, G.B. (2007) A Barley xyloglucan xyloglucosyl transferase covalently links xyloglucan, cellulosic substrates, and (1,3;1,4)-beta-d-glucans. Journal of Biological Chemistry, 282, 12951–12962.

Plant Cell Wall Biology

■

363

Jones, L., Milne, J.L., Ashford, D., McCann, M.C., McQueen-Mason, S.J. (2005) A conserved functional role of pectic polymers in stomatal guard cells from a range of plant species. Planta, 221, 255–264. Jarvis, M.C., McCann, M.C. (2000) Macromolecular biophysics of the plant cell wall: concepts and methodology. Plant Physiology and Biochemistry, 38, 1–13. Johnson, K.L., Jones, B.J., Schultz, C., Bacic, A. (2003) Non-enzymic cell wall (glyco) proteins. In: The Plant Cell Wall (ed. J.K.C. Rose), pp 111–154. Blackwell Publishing Ltd., Oxford. Knox, J.P. (2008) Revealing the structural and functional diversity of plant cell walls. Current Opinion in Plant Biology, 11, 308–313. Lee, K.J.D., Sakata, Y., Mau, S.-L., Pettolino, F., et al. (2005) Arabinogalactan-proteins are required for apical cell extension in the moss Physcomitrella patens. Plant Cell, 17, 3051–3065. Liepman, A.H., Nairn, C.J., Willats, W.G.T., Søresen, I., Roberts, A.W., Keegstra, K. (2007) Functional genomic analysis supports conservations of function among cellulose synthase-like A gene family members and suggests diverse roles of mannans in plants. Plant Physiology, 143, 1881–1893. Manfield, I.W., Orfila, C., McCartney, L., et al. (2004) Novel cell wall architecture of isoxaben-habituated Arabidopsis suspension-cultured cells: global transcript profiling and cellular analysis. Plant Journal, 40, 260–275. Marcus, S.E., Verhertbruggen, Y., Hervé, C., et al. (2008) Pectic homogalacturonan masks abundant sets of xyloglucan epitopes in plant cell walls. BMC Plant Biology, 8, 60. McCann, M.C., Wells, B., Roberts, K. (1990) Direct visualization of cross-links in the primary plant cell wall. Journal of Cell Science, 96, 323–334. McCartney, L., Blake, A.W., Flint, J., Bolam, D.N., Boraston, A.B., Gilbert, H.J., Knox, J.P. (2006) Differential recognition of plant cell walls by microbial xylan-specific carbohydrate-binding modules. Proceedings of the National Academy of Sciences of the U S A, 103, 4765–4770. McCartney, L., Gilbert, H.J., Bolam, D.N., Boraston, A.B., Knox, J.P. (2004) Glycoside hydrolase carbohydrate-binding modules as molecular probes for the analysis of plant cell wall polymers. Analytical Biochemistry, 326, 49–54. McCartney, L., Knox, J.P. (2002) Regulation of pectic polysaccharide domains in relation to cell development and cell properties in the pea testa. Journal of Experimental Botany, 53, 707–713. McCartney, L., Ormerod, A.P., Gidley, M.J., Knox, P. (2000) Temporal and spatial regulation of pectic (1→4)-β-d-galactan in cell walls of developing pea cotyledons: implications for mechanical properties. Plant Journal, 22, 105–113. McCartney, L., Steele-King, C.G., Jordan, E., Knox, J.P. (2003) Cell wall pectic (1→4)-βd-galactan marks the acceleration of cell elongation in the Arabidopsis seedling root meristem. Plant Journal, 33, 447–454. Mellerowicz, E.J., Baucher, M., Sundberg, B., Boerjan, W. (2001)Unravelling cell wall formation in the woody dicot stem. Plant Molecular Biology, 47, 239–274. Meloche, C.G., Knox, J.P., Vaughn, K.C. (2007) A cortical band of gelatinous fibers causes the coiling of redvine tendrils: a model based upon cytochemical and immunocytochemical studies. Planta, 225, 485–498. Mohnen, D. (2008) Pectin structure and biosynthesis. Current Opinion in Plant Biology, 11, 266–277.

364

■

Plant Polysaccharides

Moller, I., Sørensen, I., Bernal, A.J., et al. (2007) High-throughput mapping of cell wall polymers within and between plants using novel microarrays. Plant Journal, 50, 1118–1128. Moller, I., Marcus, S.E., Haeger, A., et al. (2008) High-throughput screening of monoclonal antibodies against plant cell wall glycans by hierarchial clustering of their carbohydrate microarray binding profiles. Glycoconjugate Journal, 25, 37–48. Obembe, O.O., Jacobsen, E., Timmers, J., et al. (2007) Promiscuous, non-catalytic, tandem carbohydrate-binding modules modulate cell wall structure and development of transgenic tobacco plants. Journal of Plant Research, 120, 605–617. Okamoto-Nakazato, A. (2002) A brief note on the study of yieldin, a wall-bound protein that regulates the yield threshold of the cell wall. Journal of Plant Research, 115, 309–313. O’Neill, M.A., York, W.S. (2003) The composition and structure of plant primary cell walls. In: The Plant Cell Wall (ed. J.K.C. Rose), pp. 1–54. Blackwell Publishing Ltd., Oxford. O’Neill, M.A., Ishii, T., Albersheim, P., Darvill, A.G. (2004) Rhamnogalacturonan-II: structure and function of a borate cross-linked cell wall pectic polysaccharide. Annual Review of Plant Biology, 55, 109–139. Orfila, C., Knox, J.P. (2000) Spatial regulation of pectic polysaccharides in relation to pit fields in cell walls of tomato fruit pericarp. Plant Physiology, 122, 775–781. Orfila, C., Seymour, G.B., Willats, W.G.T., et al. (2001) Altered middle lamella homogalact-uronan and disrupted deposition of (1→5)-α-L-arabinan in the pericarp of Cnr, a ripening mutant of tomato. Plant Physiology, 126, 210–221. Osterberg, M., Schmidt, U., Jaaskelainen, A.-S. (2006) Combining confocal Raman spectroscopy and atomic force microscopy to study wood extractives on cellulose surfaces. Colloids and Surfaces A Physiochemical Engineering Aspects, 291, 197–201. Peaucelle, A., Louvet, R., Johansen, J.N., et al. (2008) Arabidopsis phyllotaxis is controlled by the methyl-esterification status of cell wall pectins. Current Biology, 18, 1943–1948. Peña, M.J., Ryden, P., Madson, M., Smith, A.C., Carpita, N.C. (2004) The galactose residues of xyloglucan are essential to maintain mechanical strength of the primary cell walls in Arabidopsis during growth. Plant Physiology, 134, 443–451. Peña, M.J., Darvill, A.G., Eberhard, S., York W.S., O’Neill, M.A. (2008) Moss and liverwort xyloglucans contain galacturonic acid and are structurally distinct from the xyloglucans synthesized by hornworts and vascular plants. Glycobiology, 18, 891–904. Pien, S., Wyrzykowska, J., McQueen-Mason, S., Smart, C., Fleming, A. (2001) Local expression of expansin induces the entire process of leaf development and modifies leaf shape. Proceedings of the National Academy of Sciences of the U S A, 98, 11812–11817. Popper, Z.A., Fry, S.C. (2008) Xyloglucan-pectin linkages are formed intraprotoplasmically, contribute to wall assembly, and remain stable in the cell wall. Planta, 227, 781–794. Ralet, M.C., Crépeau, M.J., Lefèbvre, J., Mouille, G., Höfte, H., Thibault, J.F. (2008) Reduced number of homogalacturonan domains in pectins of an Arabidopsis mutant enhances the flexibility of the polymer. Biomacromolecules, 9, 1454–1460. Rao, Y., Buskas, T., Albert, A., O’Neill, M.A., Hahn, M.G., Boons, G.J. (2008) Synthesis and immunological properties of a tetrasaccharide portion of the B side chain of rhamnogalacturonan II (RG-II). Chembiochem, 9, 381–388.

Plant Cell Wall Biology

■

365

Ryden, P., Sugimoto-Shirasu, K., Smith, A.C., Findlay, K., Reiter, W.-D., McCann, M.C. (2003) Tensile properties of Arabidopsis cell walls depend on both a xyloglucan cross-linked microfibrillar network and rhamnogalacturonan II-borate complexes. Plant Physiology, 132, 1033–1040. Ryder, A.G. (2005) Surface-enhanced Raman scattering for narcotic detection and applications to chemical biology. Current Opinion in Chemical Biology, 9, 489–493. Schopfer, P. (2006) Biomechanics of plant growth. American Journal of Botany, 93, 1415–1425. Schopfer, P., Liszkay, A. (2006) Plasma membrane-generated reactive oxygen intermediates and their role in cell growth of plants. Biofactors, 28, 73–81. Schröder, R., Wegrzyn, T.F., Bolitho, K.M., Redgwell, R.J. (2004) Mannan transglycosylase: a novel enzyme activity in cell walls of higher plants. Planta, 219, 590–600. Schröder, R., Atkinson, R.G., Redgwell, R.J. (2009) Re-interpreting the role of endoβ-mannanases as mannan endotransglycosylase/hydrolases in the plant cell wall. Annals of Botany, 104, 197–204. Seifert, G.J., Roberts, K. (2007) The biology of arabinogalactan proteins. Annual Review of Plant Biology, 58, 137–161. Shoseyov, O., Shani, Z., Levy, I. (2006) Carbohydrate binding modules: biochemical properties and novel applications. Microbiology and Molecular Biology Reviews, 70, 283–295. Skjøt, M., Pauly, M., Bush, M.S., Borkhardt, B., McCann, M.C., Ulvskov, P. (2002) Direct interference with rhamnogalacturonan I biosynthesis in Golgi vesicles. Plant Physiology, 129, 95–102. Singh, B., Avci, U., Eichler Inwood, S.E., Grimson, M.J., et al. (2009) A specialized outer layer of the primary cell wall joins elongating cotton fibers into tissue-like bundles. Plant Physiology, 150, 684–699. Sørensen, I., Pettolino, F.A., Wilson, S.M., et al. (2008) Mixed linkage (1→3),(1→4)-βd-glucan is not unique to the Poales and is an abundant component of Equisetum arvense cell walls. Plant Journal, 54, 510–521. Sugimoto, K., Williamson, R.E., Wasteneys, G.O. (2000) New techniques enable comparative analysis of microtubule orientation, wall texture, and growth rate in intact roots of Arabidopsis. Plant Physiology, 124, 1493–1506. Thompson, D.S. (2005) How do cell walls regulate plant growth? Journal of Experimental Botany, 56, 2275–2285. Ulvskov, P., Wium, H., Bruce, D., et al. (2005) Biophysical consequences of remodeling the neutral side chains of rhamnogalacturonan I in tubers of transgenic potatoes. Planta, 220, 609–620. Vaughn, K.C., Talbot, M.J., Offler, C.E., McCurdy, D.W. (2007) Wall ingrowths in epidermal transfer cells of Vicia faba cotyledons are modified primary walls marked by localized accumulations of arabinogalactan proteins. Plant Cell Physiology, 48, 159–168. Verhertbruggen, Y., Knox, J.P. (2007) Pectic polysaccharides and expanding cell walls. In: Plant Cell Monographs 5, The Expanding Cell (eds J-P. Verbelen, K. Vissenberg), pp 139–158. Springer-Verlag, Berlin. Verhertbruggen, Y., Marcus, S.E., Haeger, A., et al. (2009) Developmental complexity of arabinan polysaccharides and their processing in plant cell walls. Plant Journal, 59, 413–425.

366

■

Plant Polysaccharides

Willats, W.G.T., Knox, J.P. (2003) Molecules in context: probes for cell wall analysis. In: The Plant Cell Wall (ed. J.K.C. Rose), pp 92–110. Blackwell Publishing Ltd., Oxford. Willats, W.G.T., McCartney, L., Steele-King, C.G., et al. (2004) A xylogalacturonan epitope is specifically associated with plant cell detachment. Planta, 218, 673–681. Willats, W.G.T., McCartney, L., Mackie, W., Knox, J.P. (2001a) Pectin: cell biology and prospects for functional analysis. Plant Molecular Biology, 47, 9–27. Willats, W.G.T., Orfila, C., Limberg, G., et al. (2001b) Modulation of the degree and pattern of methyl-esterification of pectic homogalacturonan in plant cell walls: implications for pectin methyl esterase action, matrix properties and cell adhesion. Journal of Biological Chemistry, 276, 19404–19413. Xu, P., Donaldson, L.A., Gergely, Z.R., Staehelin, L.A. (2007) Dual-axis electron tomography: A new approach for investigating the spatial organization of wood cellulose microfibrils. Wood Science and Technology, 41, 101–111. Yennawar, N.H., Li, A., Dudzinski, D.M., Tabuchi, A., Cosgrove, D.J. (2006) Crystal structure and activities of EXPB1 (zea m 1), a β-expansin and group-1 pollen allergen from maize. Proceedings of the National Academy of Sciencesof the U S A, 103, 14664–14671. Yokota, S., Ueno, T., Kitaoka, T., Wariishi, H. (2007) Molecular imaging of single cellulose chains aligned on a highly oriented pyrolytic graphite surface. Carbohydrate Research, 342, 2593–2598. Zhong, R., Demura, T., Ye, Z.H. (2006) SND1, a NAC domain transcription factor, is a key regulator of secondary wall synthesis in fibers of Arabidopsis. Plant Cell, 18, 3158–3170. Zykwinska, A., Thibault, J.-F., Ralet, M.-C. (2008) Organization of pectic arabinan and galactan side chains in association with cellulose microfibrils in primary cell walls and related models envisaged. Journal of Experimental Botany, 58, 1795–1802.

Annual Plant Reviews (2011) 41, 367–388 doi: 10.1002/9781444391015.ch15

Chapter 15

ENZYMATIC MODIFICATION OF PLANT CELL WALL POLYSACCHARIDES Jens Øbro1, Takahisa Hayashi2 and Jørn Dalgaard Mikkelsen3 1

Department of Biology, University of Copenhagen, Ole Maaløes vej 5, 2200 Copenhagen N, Denmark; present address: Novozymes A/S, Krogshøjvej 36, 2880 Bagsværd, Denmark 2 Research Institute for Sustainable Humanosphere, Kyoto University, Uji, Kyoto 611-0011, Japan 3 Department of Chemical and Biochemical Engineering, Technical University of Denmark, Søltofts Plads, building 229, 2800 Kgs. Lyngby, Denmark Manuscript received April 2008

Abstract: Plant cell walls are intricate structures with remarkable properties, widely used in almost every aspect of our life. Cell walls consist largely of complex polysaccharides and there is often a need for chemical and biochemical processing before industrial use. There is an increasing demand for sustainable processes that replace chemical treatments with white biotechnology. Plants can contribute significantly to this sustainable process by producing plant or microbial enzymes in planta that are necessary for plant cell wall modification or total degradation. This will give rise to superior food fibres, hydrocolloids, paper, textile, animal feeds or biofuels. Classical microbial-based fermentation systems could in the future face serious competition from plant-based expression systems for enzyme production. Plant expressed enzymes can either be targeted to specific cellular compartments for accumulation and storage, or targeted to the cell wall for immediate modification of the polysaccharides. Furthermore, thermoactivated enzymes can successfully be employed to avoid undue degradation of the cell wall and detrimental phenotypes resulting from this. A further use of enzyme expression in planta is the possibility of elucidating the biological role of specific polymers, such as the involvement of xyloglucan in tree bending, or to decrease the level of pectin in flax to simplify the retting step in the textile refinery process. Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

367

368

■

Plant Polysaccharides

Keywords: fungal polysaccharide hydrolases and lyases; heterologous expression; plant cell wall remodelling.

15.1 Introduction The walls that surround plant cells usually represent a major portion of the plant body and are of key importance in plant growth and development, and in resistance to pathogens. A vast number of industrial products have been generated from plant cell wall polymers. They include textiles, papers, bioethanol, pharmaceuticals and food ingredients such as dietary fibres, prebiotic oligomers and hydrocolloids (Farrokhi et al. 2006; Somerville 2006; Willats et al. 2006; von Wettstein 2007). To achieve the desired functionality of these industrial products the plant polymers have to be subjected to a range of chemical or biochemical processes. In the last decade, advances in biotechnology have provided new and improved methods to generate plant cell wall derived products with better properties (Giddings et al. 2000; Rishi et al. 2001; Farrokhi et al. 2006; Minic & Jouanin 2006). In addition, plant polymers can be modified in vivo by careful design of the glycosyl transferases, as with the introduction of β-glucan biosynthesis in Arabidopsis (Burton et al. 2006) or by expressing accessory proteins or enzymes responsible for the in vivo modification. The accessory proteins can either be enzymes degrading a specific polymer, such as xyloglucanase or cellulase (Park et al. 2004; Ransom et al. 2007), or noncatalytic proteins such as expansins and carbohydrate binding modules (Rochange et al. 2001; Obembe et al. 2007). Alternatively, plant and microbial enzymes can be produced in planta and deposited intra- or extracellularly, the so-called post-harvest modification. During post-harvest modification the deposited enzymes will be released or activated by heat and the substrates will be made available by maceration or other cell destructive processes. The enzymatic processes will subsequently transform the plant cell wall polymer into products with the desired functional property (Dai et al. 2000a; Dai et al. 2000b; von Wettstein et al. 2000; von Wettstein et al. 2003; Dai et al. 2005; von Wettstein et al. 2007). Designing the biosynthesis of the plant cell wall polymers by expression of transgenic enzymes may give rise to dwarf, detrimental or lethal phenotypes, but there are also examples of gene expression where the transgenic plants have demonstrated proof of concept by in vivo modifications, such as the elimination of galactan and arabinan side chains in potato pectin (Sørensen et al. 2000; Skjøt et al. 2002). Subcellular deposition of thermostable enzymes with low activity at ambient crop growth conditions is interesting for malting and brewing, the production of barley-based feed products for poultry (von Wettstein 2007), and improving plants for biofuel production (Dai et al. 2005; Oraby et al. 2007). The synthesis and deposition of thermostable enzymes in planta for post-harvest modifications may have a very high

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

369

potential in the future manufacturing of paper, textiles, feed, food and food ingredients. In feed for broiler chicks, it is estimated that with the present expression levels of an engineered, thermostable Bacillus macerans 1,3–1,4-βglucanase in transgenic barley plants, it is possible to feed 40 million broilers with 280 000 tonnes of non-transgenic barley containing only 56 tonnes of transgenic barley. The expression of the 1,3–1,4-β-glucanase is so high in the transgenic barley grains that only 25 acres (10 ha) of farmland planted with transgenic barley can satisfy the need (von Wettstein 2007). Plant-based enzyme production should be considered as an interesting alternative to the classical microbial-based fermentation systems. The commercial exploitation of this plant-based enzyme system seems to be viable, but further investigation with large-scale production is required before a final conclusion can be made. In the present review we have chosen to separate the new biotechnological solutions into two sections: (1) In vivo modification of plant polymers, and (2) post-harvest modification with enzymes expressed and deposited in plants.

15.2

In vivo modifications

Plant cell wall polymers are synthesized by numerous enzymes. A large number of genes have in recent years been identified by functional genomics and molecular genetics, and are currently being examined to evaluate their potential functions in cell wall biosynthesis. A number of reviews comprehensively describe this research, including Farrokhi et al. (2006), Minic and Jouanin (2006), Somerville (2006), Willats et al. (2006) and von Wettstein (2007). Modifications of plant polysaccharides in vivo are possible either by interfering directly with the enzymes involved in biosynthesis or by modifying or partly degrading specific polymers by the expression of hydrolytic enzymes or accessory proteins. Modification of plant polysaccharides is also possible by controlling the source/sink strength between plant tissues with the expression of invertase. This leads to resource allocation within the transgenic plant and a change in the direction of biosynthesis that can eventually lead to cell wall modifications as described by Sonnewald et al. (1997), Heyer et al. (2004) and Canam et al. (2006). Down-regulation of endogenous glycosyl transferases by knock-out mutations or by gene silencing in the biosynthesis of cellulose has been described and is reviewed by Somerville (2006). The concept of introducing old polymers into new plants, thus creating ‘gain-of-function’ mutants by heterologous expression of foreign glycosyl transferases, has been described by Burton et al. (2006). Such fundamental experiments are of great importance in evaluating the biological function of glycosyl transferase, and may in the future give rise to ‘tailor-made’ polysaccharides by more specific control of the synthesis and modification of the

370

■

Plant Polysaccharides

cell wall. This would be very valuable for many industrial and pharmaceutical products. The biological function of a specific cell wall polysaccharide can be evaluated by partly removing or modifying the polymer by a hydrolytic enzyme expressed in the plant. Research involving in vivo polysaccharide modifications by expressing active hydrolytic enzymes has been reported for pectin (Boudart et al. 2003), hemicellulose (Harpster et al. 2002; Park et al. 2004; Cho et al. 2006; Baba et al. unpublished), and cellulose (Park et al. 2003; Shani et al. 2004) (see Table 15.1. for a comprehensive list). Modifications of pectin have included the backbone and methylation pattern of homogalacturonan, the rhamnogalacturonan I backbone, and the galactan/arabinan side chains of rhamnogalacturonan I. One of the first attempts at in vivo modification by transgenic expression of active enzyme was published in 1990, where tomato endo-polygalacturonase (endoPG) was expressed in tobacco (Osteryoung et al. 1990). Although active endoPG enzyme accumulated in the apoplast of the transgenic plants no in vivo homogalacturonan degradation occurred (Osteryoung et al. 1990). In contrast, in vivo homogalacturonan degradation did occur when an endogenous apple fruit endoPG was constitutively overexpressed in apple trees. EndoPG overexpression resulted in a less rigid cell wall and reduced cell adhesion ultimately leading to a range of morphological changes (Atkinson et al. 2002). When Boudart et al. (2003) expressed a fungal endoPG transiently in tobacco plants, the enzyme reduced the galacturonic acid content of the cell wall by 30% and resulted in chlorotic and necrotic lesions on the leaves. A dwarfed phenotype was observed when another fungal endoPG was transiently expressed in Arabidopsis and tobacco plants (Capodicasa et al. 2004). This finding contradicts the phenotypes with reduced cell adhesion, and chlorotic and necrotic lesions described by Boudart et al. (2003) and Atkinson et al. (2002). Phenotypical changes are frequently observed when modifying homogalacturonan by endoPG, but there are also reports of homogalacturonan modification by endoPG expression that did not reveal any visual phenotypes. When a fungal endoPG was expressed in flax no phenotype was reported even though a 50% reduction in homogalacturonan occurred (Musialak et al. 2008). In addition to backbone modifications of pectin, homogalacturonan has also been subjected to modification, for example by transgenic expression of pectin methyl esterase or pectin methyl esterase inhibitors. Pilling et al. (2000) expressed a Petunia inflata pectin methyl esterase in potato plants. Although no chemical changes in homogalacturonan structure and methylation were observed, cell walls from the transgenic potato plants showed differences in ion-binding properties compared with wildtype cell walls. In addition, some developmental changes in growth rate of the younger transgenic plants and reduced tuber yield at maturity were observed (Pilling et al. 2000). Hasunuma et al. (2003) described constitutive expression of a fungal pectin methyl esterase in tobacco (Nicotiana tabacum) cell suspension

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

371

Table 15.1 In vivo and post-harvest modifications of plant cell wall polysaccharides through expression of hydrolytic enzymes Modification

Polymer

Enzyme

Localization

Organism

Reference

In vivo

Pectin

EndoPG (plant)

Apoplast

Tobacco

In vivo

Pectin

Apoplast

Potato

In vivo

Pectin

Apoplast

Potato

In vivo

Pectin

Pectin methylesterase (plant) Galactanase (fungal) Endopg (plant)

Osteryoung et al. 1990 Pilling et al. 2000

Apoplast

Apple

In vivo

Pectin

Golgi

Potato

In vivo

Pectin

Arabinanase (fungal) RG lyase (fungal)

Apoplast

Potato

In vivo

Pectin

Endopg (fungal)

Apoplast

Tobacco

In vivo

Pectin

Apoplast

Tobacco cells

In vivo

Pectin

Pectin methylesterase (fungal) Endopg (fungal)

Apoplast

In vivo

Pectin

Apoplast

Capodicasa et al. 2004 Hasunuma et al. 2004

In vivo

Pectin

Apoplast

Potato

In vivo

Pectin

Apoplast

Potato

In vivo

Pectin

Golgi Apoplast

Potato

Borkhardt et al. 2005 Martín et al. 2005 Ulvskov et al. 2005

In vivo

Pectin

Pectin methylesterase (fungal) Arabinanase (fungal) Galactosidase (plant) Arabinanase/ galactanase (fungal) EndoPG (fungal)

Tobacco/ Arabidopsis Tobacco

Apoplast

In vivo

Pectin

Apoplast

Tobacco/ Arabidopsis Flax

In vivo

Hemicellulose

Apoplast

Tomato

In vivo

Hemicellulose

Apoplast

Poplar

In vivo

Hemicellulose

Apoplast

Arabidopsis

In vivo

Hemicellulose

Apoplast

Arabidopsis

Shin et al. 2006

In vivo

Hemicellulose

Apoplast

Arabidopsis

Liu et al. 2007

In vivo

Cellulose

Apoplast

Arabidopsis

Park et al. 2003

EndoPG (fungal) RGase (fungal) Glucanase (plant) Xyloglucanase (plant) Xyloglucan endotransglucosidase/ hydrolase (plant) Xyloglucan endotransglucosidase/ hydrolase (plant) Xyloglucan endotransglucosidase/ hydrolase (plant) Glucanase (plant)

Sørensen et al. 2000 Atkinson et al. 2002 Skjøt et al. 2002 Oomen et al. 2002 Boudart et al. 2003 Hasunuma et al. 2003

Ferrari et al. 2008 Musilak et al. 2008 Harpster et al. 2002 Park et al. 2004 Cho et al. 2006

372

■

Plant Polysaccharides

Table 15.1 Continued Modification

Polymer

Enzyme

Localization

Organism

Reference

In vivo

Cellulose

Apoplast

Poplar

Post-harvest

Pectin

ER

Tobacco

Post-harvest

Hemicellulose

Apoplast

Barley

Post-harvest

Hemicellulose

Glucanase (plant) Arabinanase (fungal) β-Glucanase (bacterial) Xylanase (fungal)

Oil-bodies

Brassica napus

Post-harvest

Hemicellulose

Vacuole

Barley

Post-harvest

Hemicellulose

Apoplast

Barley

Post-harvest

Hemicellulose

Vacuole

Barley

Post-harvest

Hemicellulose

β-Glucanase (bacterial) β-Glucanase (fungal) Glucanase (bacterial) Xylanase (fungal)

Cytoplasm

Barley

Post-harvest

Hemicellulose

Cytosol

Rice

Post-harvest

Cellulose

Cytosol

Tobacco cells

Post-harvest

Cellulose

Cytosol

Post-harvest

Cellulose

Tobacco leaf and callus Tobacco

Post-harvest

Cellulose

Post-harvest

Cellulose

Post-harvest

Cellulose

Xylanase (bacterial) Glucanase (bacterial) Glucanase (fungal) Glucanase (bacterial) Glucanase (bacterial) Glucanase (bacterial) Glucanase (bacterial)

Shani et al. 2004 Skjøt et al. 2001 Jensen et al. 1996 Liu et al. 1997 Jensen et al. 1998 Nuutila et al. 1999 Horvath et al. 2000 Patel et al. 2000 Kimura et al. 2003 Kawazu et al. 1996 Dai et al. 1999 Kawazu et al. 1999 Ziegelhoffer et al. 1999 Dai et al. 2000a Dai et al. 2000b

Post-harvest

Cellulose

Post-harvest

Cellulose

Post-harvest

Cellulose

Post-harvest

Cellulose

Post-harvest

Cellulose

Post-harvest

Cellulose

Post-harvest

Cellulose

Post-harvest

Cellulose

Post-harvest

Cellulose

Post-harvest

Cellulose

Glucanase (bacterial) Glucanase (bacterial) Glucanase (bacterial) Glucanase (fungal) Glucanase (bacterial) Glucanase (bacterial) Glucanase (bacterial) Glucanase (bacterial) Glucanase (bacterial) Glucanase (bacterial)

Cytosol Cytosol Chloroplast ER Vacuole Apoplast Chloroplast Apoplast

Tobacco Alfalfa Tobacco Potato

Arabidopsis

Cytosol Apoplast Chloroplast Chloroplast

Tobacco

Cytoplasm

Barley

ER Vacuole Apoplast Chloroplast Apoplast

Tobacco

Maize

Cytosol

Duckweed

Apoplast

Rice

Apoplast

Maize

Chloroplast

Tobacco

Tobacco

Ziegler et al. 2000 Ziegelhoffer et al. 2001 Jin et al. 2003 Xue et al. 2003 Dai et al. 2005 Biswas et al. 2006 Sun et al. 2007 Oraby et al. 2007 Ransom et al. 2007 Yu et al. 2007

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

373

culture. This resulted in methanol overproduction since hydrolysis of the methyl esters in homogalacturonan releases methanol and acidic pectins (Hasunuma et al. 2003). When the same construct was transformed into tobacco plants the resulting transgenic plants had short internodes and small leaves, and displayed a dwarfed phenotype (Hasunuma et al. 2004). This phenotype could either be explained as a direct consequence of decreased methylation of homogalacturonan or by an increased susceptibility of the homogalacturonan to endogenous endoPG. Expression of pectin methyl esterase inhibitors in Arabidopsis decreased the activity of the endogenous pectin methyl esterase resulting in a higher degree of homogalacturonan esterification. In addition to the increase in esterification, expression of pectin methyl esterase inhibitors also led to improved resistance to Botrytis cinerea infections (Lionetti et al. 2007). A change in homogalacturonan content or degree of methylation does not always give rise to a change in phenotypes. A reason for this may be a variable stress response induced by the released degradation products of homogalacturonan (oligogalacturonide response) as suggested by Côte and Hahn (1994) and Ridley et al. (2001). The effects of in vivo degradation of the rhamnogalacturonan I backbone were investigated by expression of a fungal rhamnogalacturonan lyase in potato tubers. This resulted in transgenic tubers with reduced content of rhamnogalacturonan I and tubers with minor morphological alterations (Oomen et al. 2002) (Fig. 15.1). In contrast, no visual phenotypic changes were observed when a rhamnogalacturonanase were expressed in flax even though both potato and flax had approximately the same reduction in the rhamnogalacturonan I content (Oomen et al. 2002; Musialak et al. 2008). Alterations of the galactan side chains of rhamnogalacturonan I have been investigated in potato tubers by expression of galactanase and galactosidase.

Figure 15.1 Potato tubers expressing a rhamnogalacturonan lyase (RG lyase) have clear morphological changes compared with the unmodified wild type tubers. (Courtesy of K. Cankar and R. Visser, Wageningen UR Plant Breeding)

374

■

Plant Polysaccharides

This resulted in tubers with reduced galactosyl content but did not otherwise produce visual phenotypes in the plants (Sørensen et al. 2000; Martín et al. 2005). However, the tubers expressing galactanase had changes in their physical properties and were found to be more brittle (Ulvskov et al. 2005). Similar changes in physical properties of potato tubers were observed when a fungal arabinanase was targeted to the place of pectin biosynthesis, the Golgi vesicles (Mohnen 1999; Skjøt et al. 2002; Ulvskov et al. 2005). The Golgi targeting approach was investigated since expression of the fungal arabinanase to the apoplast of potatoes had a detrimental effect on the plants (Skjøt et al. 2002; Borkhardt et al. 2005). The authors suggested that the severe phenotype could be induced as a stress response to the release of arabinosyl oligomers similar to the well-described oligogalacturonide-induced response (Côte & Hahn 1994; Ridley et al. 2001). In contrast, when a similar arabinanase was expressed constitutively and deposited in the apoplast of Arabidopsis no detrimental phenotype was observed although a significant decrease in arabinan was observed (Øbro et al. unpublished). Another explanation of the severe phenotypic changes observed during arabinanase expression in potatoes could be that the plant recognizes the arabinanase enzyme itself in an as yet uncharacterized manner. This type of recognition has previously been suggested for fungal xylanase since a specific binding site for the xylanase can be found on tobacco membranes and protoplasts (Sharon et al. 1993; Hanania & Avni 1997) and a point-mutated fungal xylanase with severely reduced activity still elicits the same defense responses as un-modified xylanase (Enkerli et al. 1999). One example of an industrial use for in vivo modified pectin is in the fibre extraction process. Flax fibres have been used for textiles (linen) for at least 5000 years, and are thus one of the oldest fibres used by humans. Flax fibres are extracted from the stems of mature plants by an extraction process called retting. This involves enzymatic degradation of the pectin in the middle lamella to release the fibres. To improve retting of flax fibres, Musialak et al. (2008) described the production of two types of transgenic flax plant lines expressing either endo-polygalacturonase or rhamnogalacturonanase, both from Aspergillus aculeatus. Both lines had a reduction in their pectin content of approximately 50–60% and displayed improved retting efficiency (Musialak et al. 2008). The approach of improving retting efficiency through self-processing plants by expressing cell wall degrading enzymes could most likely also applied to other important fibre crops such as jute and hemp. In vivo modification of xyloglucan has also been examined by expression of xyloglucan-specific endo-1,4-β-glucanase (XEG) and xyloglucan endotransglucosydases (XET) to evaluate the function of this plant cell wall polymer. The hemicellulose xyloglycan is a key polysaccharide involved in the crosslinking of cellulose microfibrils in the cell wall, and consequently degradation and reconnection of xyloglucans could induce modifications of cell wall properties. The separation of microfibrils during cell elongation has been thought to require endogenous enzymes that cleave xyloglucan or

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

375

loosen its binding to the cellulose microfibrils (Hayashi 1989). This hypothesis was substantiated by overexpressing a xyloglucanase in transgenic poplar trees where cleavage of xyloglucan crosslinks, caused an increase in cellulose content and density in the cell walls of the transgenic plants (Park et al. 2004). While xyloglucanase (XEG) catalyses a cleavage of xyloglucan 1,4-β-glucan backbone (Matsumoto et al. 1997), the reconnection between xyloglucans can be catalysed by xyloglucan endo-transglucosylase (XET) (Rose et al. 2002). The very important function of xyloglucan has been recently demonstrated by Baba et al. (unpublished). A range of hydrolytic enzymes, including endogenous xyloglucanase, was overexpressed in transgenic poplars trees with knock-out of specific polysaccharides (Baba et al., unpublished). When placed horizontally, the basal regions of transgenic poplars stems expressing xyloglucanase could not bend upwards because of the low strain in the tension side of the xylem (Baba et al. unpublished). When a XET was expressed in Arabidopsis the molecular size of xyloglucan was increased but, as expected, it was decreased in the knock-out lines of the same gene (Liu et al. 2007). Furthermore, the knock-out line displayed a dwarfed phenotype but the XET expression resulted in an increased plant growth. The authors also found a decrease in the relative amount of cellulose with an increased amount of total sugars in the cell walls of the knock-out line (Liu et al. 2007). XET from Brassica campestris have also been expressed in Arabidopsis where oversized phenotypes with larger rosette leaves and stems were observed (Shin et al. 2006). In contrast, when a XET from hot pepper was expressed in Arabidopsis a different phenotype with leaves having increased numbers of small cells together with improved tolerance to severe water deficit and high salinity compared with that of the wild type was observed (Cho et al. 2006). In addition to the transgenic approaches, modification of plant cell wall polymers may also be drastically altered by specific mutations or by treatment with specific herbicides or other chemicals. Examples of these are the mutants rsw1 and Kor that are involved in cellulose biosynthesis and the qua2 gene involved in pectin formation. Mutation of the qua2 gene reduced homogalacturonan levels by approximately 50% but did not change the rhamnogalacturonan content of the pectin polymer (Mouille et al. 2007). A high ratio of rhamnogalacturonan can give rise to pectin with very high flexibility and possibly an interesting functional property. The rsw1 and Kor mutations give rise to accumulation of 1,4-β-glucan which may be of industrial interest, particularly with respect to food products such as fibres, hydrocolloids and prebiotics. It is known that the knock-out mutant of cellulose synthase and/or the inhibition of cellulose synthesis can result in a decrease in crystalline cellulose content and an increase in soluble 1,4-β-glucan in the walls. Arioli et al. (1998) observed that the rsw1 mutation disassembled the cellulose synthase complex, reduced the cellulose accumulation but increased the non-crystalline form of 1,4-β-glucan. In addition non-crystalline 1,4-βglucan was also observed in the Arabidopsis rsw2 mutants (allelic to the Kor

376

■

Plant Polysaccharides

gene) that code for a membrane-bound cellulase in plants (Lane et al. 2001; Mølhøj et al. 2001; Peng et al. 2002). Peng et al. (2001) further characterized the non-crystalline 1,4-β-glucan from cotton fibres. The 1,4-β-glucan could be solubilized without any detergent or alkali, it contained 1,4-β-linked glucose residues, and could be stained with Calcofluor White. In french bean cells habituated to grow in media containing a herbicide (DCB) large amounts of soluble 1,4-β-glucan was produced (Encina et al. 2002) and this was also the case for Arabidopsis cells habituated to grow in isoxaben (Manfield et al. 2004) Although the formation of 1,4-β-glucan in plants is observed in several plant systems, the mechanism of biosynthesis and the potential industrial use still remain to be elucidated. A number of experiments have also been carried out with an interesting added benefit on biomass yield. When cellulase (Park et al. 2003; Shani et al. 2004) and xyloglucanase (Park et al. 2004) were expressed in Arabidopsis and poplar an increased growth rate and biomass production was observed (Park et al. 2003; Park et al. 2004; Shani et al. 2004). An increase in growth rate and dry weight accumulation has also been observed when a bacterial carbohydrate binding module (CBM) was expressed in potato plants (SafraDassa et al. 2006). In contrast, the opposite effect with reduced growth was observed when a different CBM was expressed in tobacco plants (Obembe et al. 2007). Further experiments are needed before any firm conclusions are drawn with respect to the effects of expression of CBM in planta. It has also been suggested that overexpression of expansins, a cell wall loosening nonenzymatic protein, would soften the cell wall and lead to increased growth of the plants. Softening of the cell wall was indeed observed when an expansin was overexpressed in tomato fruits (Brummell et al. 1999). Likewise, larger cells and leaves were also found when an expansin was overexpressed in Arabidopsis (Cho & Cosgrove 2000). In contrast, when transgenic tomato plants expressed a cucumber expansin the transgenic plants were stunted and had shorter cortical and epidermal cells (Rochange & McQueen-Mason 2000; Rochange et al. 2001). On the basis of these findings it is difficult to predict how CBM and expansin expression will influence plant growth and development. Still, if the controversial findings of increased biomass accumulation were reproducible in other plant systems, it could be exploited commercially leading to an increase in biomass yield.

15.3 Post-harvest modifications In contrast to in vivo modification where the plant cell wall polymers are changed at the site and time of synthesis, post-harvest modification is defined as a process, where enzymes are produced in planta and deposited in appropriate compartments of the plant. In the post-harvest system the enzymes are first activated when the plant tissue is disintegrated during the refinery process. For heat-inducible enzymes activation may also require

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

377

higher temperature. While in vivo modifications may be used for generation of phenocopies for a mutant screen or as techniques to elucidate a suggested biological role of a specific cell wall component, the post-harvest modifications may fundamentally be better designed for industrial applications. One of the most interesting approaches in post-harvest modifications is the expression of enzymes that are inactive at normal plant growth temperatures, thus preventing the enzymes from degrading the cell wall prior to harvest (Montalvo-Rodriguez et al. 2000) (see Table 15.1). This concept was demonstrated when a truncated Acidothermus cellulolyticus endo-glucanase was expressed and deposited in the apoplastic space of Arabidopsis leaves. Expression of this enzyme had no detrimental effects on the transgenic plants although a very high protein level of the transgenic endo-glucanase was achieved (Ziegler et al. 2000). The same Acidothermus cellulolyticus endoglucanase has also been expressed as both a full-length and a truncated version in tobacco, where the enzymes were deposited at three sites: the cytosol, the chloroplasts and the apoplast (Ziegelhoffer et al. 2001). When the truncated endo-glucanase version was expressed and deposited in the apoplast the level of enzyme was increased more than 500-fold compared with the level present in the cytosol, when the full-length endo-glucanase gene was used. This difference in expression levels was probably due to proteases in the cytosol since the mRNA levels were similar (Ziegelhoffer et al. 2001). Chloroplastic accumulation has also been attempted in tobacco plants both with full-length versions (Dai et al. 2000a; Jin et al. 2003) and with truncated versions (Jin et al. 2003) of the Acidothermus cellulolyticus endoglucanase. Active enzyme could be recovered from both versions (Dai et al. 2000a; Jin et al. 2003). Dai et al. (2005) has later confirmed that the endoglucanase was indeed degraded by proteases in tobacco chloroplasts. Finally, duckweed (Lemna minor) and rice (Oryza sativa) have also been used to express the Acidothermus cellulolyticus endo-glucanase (Oraby et al. 2007; Sun et al. 2007). When the endo-glucanase was expressed in duckweed and deposited in the cytosol a relatively low level of enzyme was observed, as expected (Sun et al. 2007). In contrast, expression of the catalytic domain of endo-glucanase in the apoplastic space in rice had no apparent deleterious effects on plant growth and development although the enzyme accounted for up to 5% of the total soluble protein from leaves (Oraby et al. 2007). In addition to the endo-glucanase examples, expression of a number of xylanases has also been examined. A xylanase from the fungi Neocallimastix patriciarum was expressed and targeted to the cytoplasm of barley grain endosperm (Patel et al. 2000), while Kimura et al. (2003) demonstrated the possibility of expressing the catalytical domain of a bacterial xylanase in the cytosol of rice plants. Although active xylanase accumulated in the cytosol, the transgenic plants displayed no phenotypic changes (Kimura et al. 2003). That the xylanase in the transgenic rice plant indeed made a difference to the digestibility of the plant material was further examined by placing transgenic plant material, packed in nylon mesh, in the rumen of a goat. This

378

■

Plant Polysaccharides

indicated increased digestibility of the transgenic xylanase rice material compared to unmodified rice material (Ohmiya et al. 2003). In contrast to the rice expression, no viable transformants could be recovered when a Bacillus subtilis xylanase was constitutively expressed in wheat. When the Bacillus subtilis xylanase expression was restricted to the grain endosperm, almost normal fertility was observed. The development of the grain was, however, severely affected and it could not be used for animal studies (Bach et al., unpublished). A dual targeting approach has been carried out with a Trichoderma reesei xylanase by Hyunjong et al. (2006), where the enzyme was deposited in both the chloroplasts and peroxisomes (Hyunjong et al. 2006). This gave rise to a 200% increase in xylanase accumulation compared with xylanase targeted only to a single cellular compartment. The dual targeting was accomplished by fusing two signal peptides on to the same xylanase. At the N-terminal a transit peptide ensured chloroplast targeting, while a three-amino-acid tag (Ser-Lys-Leu) at the C-terminal facilitated peroxisome localization of the xylanase (Hyunjong et al. 2006). Finally, subcellular localization of hydrolytic enzymes has also been demonstrated in the endoplasmatic reticulum, vacuole and oil bodies (Liu et al. 1997; Dai et al. 2000b; Skjøt et al. 2001; Dai et al. 2005). In the case of targeting enzymes to oil bodies, an oleosin-coding region was fused to a xylanase gene originating from Neocallimastix patriciarum and the construct introduced in Brassica napus (Canola). With this approach active enzyme could be obtained from oil bodies of the transgenic Canola seeds (Liu et al., 1997). When an arabinanase was targeted to the endoplasmatic reticulum of tobacco plants the C-terminal KDEL-tag (Lys-Asp-Glu-Leu) ensured endoplasmatic reticulum retention (Skjøt et al. 2001). In this case there is no information about the transgenic protein concentration, but when Dai et al. (2005) targeted an endo-glucanase to endoplasmatic reticulum using a similar KDEL tag it resulted in low accumulation of the endo-glucanase (Dai et al. 2005). These examples illustrate the importance of subcellular localization of the enzymes in specific compartments of the plant. Both the cytosol and chloroplast compartments have a high level of proteolytic activity, which renders these sites less favoured for deposition of heterologous enzymes. In contrast, the apoplastic targeting of thermostable enzymes, or enzymes that in general are inactive at ambient growth temperatures, seem to be acceptable. The principle of post-harvest modification has already been applied in several industrial processes, such as malting, brewing, feed and bioethanol (Oraby et al. 2007; von Wettstein 2007). In the malting and brewing process, 1,3–1,4-β-glucans is a major concern when it comes to the brewing of beer, as β-glucan can have an adverse effect on the fermentation as well as increasing the viscosity of the wort and thus preventing filtration (Keegstra & Walton 2006). 1,3–1,4-β-glucans constitute on average 10% of the barley cell wall and up to 80% of the endosperm cell wall (Skendi et al. 2003; von Wettstein 2007). Purified β-glucanase is therefore often added during malting

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

379

(von Wettstein 2007). Barley grains actually contain endogenous β-glucanase, but this enzyme is not heat-stable and is irreversibly inactivated during the malting process; heat-stable β-glucanase is consequently used (von Wettstein 2007). The heat-stable β-glucanase is currently produced by fermentation but could just as well be expressed heterologously by transgenic barley plants. This would cut the production costs of purified β-glucanase for brewing. This was first carried out by Aspegren et al. (1995) where an active β-glucanase from Trichoderma reesei was expressed by transgenic suspension cultured barley cells. In the following year, Jensen et al. (1996) described the breeding of transgenic barley plants expressing a thermostabile hybrid βglucanase targeted to the endosperm of the grains (Jensen et al. 1996; Jensen et al. 1998). In a later study the same hybrid β-glucanase as described by Jensen et al. (1996) was optimized by codon design and deposited in the storage vacuoles of barley grains. This ensured a higher yield of recombinant enzyme (Horvath et al. 2000). These transgenic barley lines expressing βglucanase could easily be used for brewing, but consumer aversion to transgenic plants prevents their commercial implementation (von Wettstein 2007). Enzymes expressed in plants have also been used to improve the quality of feed. Approximately 15% of barley grains are used for brewing and the remaining 85% are used for animal feed (von Wettstein 2007). Barley grains have, however, a low nutritional value for some animals because of their high content of 1,3–1,4-β-glucans and arabinoxylans (Patel et al. 2000). These polymers are a problem particularly for monogastric animals like poultry and pigs, as these animals cannot degrade them. Feeding animals with barley reduces the digestibility and nutrient uptake of the feed as well as increasing the occurrence of indigestion, resulting in lower growth rates and reduced animal welfare (von Wettstein et al. 2000). Purified microbial βglucanase and xylanase is therefore often added to the feed (Bedford 1995; Patel et al. 2000; Malathi & Devegowda 2001). As with the enzymes used for malting, these enzymes are also produced by fermentation but could just as well be produced in planta. Also as with the enzymes expressed in planta that could be used for malting, the same thermostability requirements must be met since feed used for poultry is normally heat-sterilized to prevent salmonella infections (Matlho et al. 1997). The usefulness of transgenic barley for broiler feed was first reported by von Wettstein et al. (2000) when broiler chickens were fed a diet consisting of non-transgenic barley supplemented with transgenic barley expressing a thermotolerant β-glucanase. Broilers supplemented with transgenic barley had reduced indigestion and grew faster than broilers fed on non-transgenic barley (von Wettstein et al. 2000; von Wettstein et al. 2003; von Wettstein et al. 2007). It is estimated that with the present expression levels of the thermostable Bacillus macerans 1,3–1,4-βglucanase in transgenic barley plants, it would be possible to feed 40 million broilers with 280 000 tonnes of non-transgenic barley containing only 56 tonnes of transgenic barley. The production of these 56 tonnes of

380

■

Plant Polysaccharides

transgenic barley containing the 1,3–1,4-β-glucanase only require 25 acres (10 ha)of farmland to satisfy the need. The commercial exploitation of this plant-based enzyme system seems to be viable, but further investigations with large-scale production is required before a final conclusion can be made. Other cell wall degrading enzymes have also been expressed in barley grains. They include a modified Neocallimastix patriciarum xylanase, a hybrid Neocallimastix patriciarum/Piromyces cellulase, and a Trichoderma reesei 1,4-βglucanase (Nuutila et al. 1999; Patel et al. 2000; Xue et al. 2003). These transgenic barley lines could probably also be used for feed, but to our knowledge they have not been tested as animal feed. With the world’s reserves of fossil fuels rapidly diminishing there is an enormous interest in finding renewable sources of energy that could eventually replace them, such as biofuels. Ethanol produced from lignocellulosic plant material, also known as second-generation biofuels, could be a supplement to fossil fuels (Gray et al. 2006; Lin & Tanaka 2006; Ragauskas et al. 2006). However, there are many obstacles that must be overcome before this approach is economically feasible (Hahn-Hägerdal et al. 2006). One of the major obstacles is that the plant cell wall is very resistant to enzymatic degradation and thus difficult to degrade into fermentable sugars (Jørgensen et al. 2007). To circumvent this ‘biomass recalcitrance’ it might be feasible to manufacture transgenic self-processing plants that degrade their own cell wall in a controlled fashion by expression of hydrolytic enzymes. To avoid affecting the plant severely while the plant is growing it is possible to express and store inactive enzyme and to release the enzyme after harvest either by physical maceration of the plant tissue (Dai et al. 2005) and/or by thermal activation of the enzymes, i.e. a post-harvest approach (MontalvoRodriguez et al. 2000; Oraby et al. 2007). Although creating transgenic self-processing plants for biofuel production certainly has an appeal, there are limitations and drawbacks to this approach. The general concept of current technologies used for biofuel production involves harsh pretreatment steps of the plant material before enzymatic hydrolysis. These pretreatment steps increase the digestibility of plant material but would unfortunately also inactivate most enzymes present in the plants (Lin & Tanaka 2006; Jørgensen et al. 2007). Another possibility is to use plants as production organisms for proteins instead of producing the proteins by microbial fermentations.

15.4 Perspectives Although the plant cell wall is a very complex mixture of polymers, stateof-art-techniques such as system biology, bioinformatics, functional genomics and glycomics have increased our knowledge base to a very high level. A new application of plant science would be to generate crop plants with a

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

381

range of superior properties. These properties could include both input and output traits, like drought, cold and pest resistance, plants with increased nutrition value, plants expressing self-destructive thermostable enzymes, and plants with higher levels of plant cell wall polymers. The ideal plants would contain a cell wall comprising of only one homopolymer with high functional properties which could be easily extracted and processed to superior industrial products. By careful tampering with the structure of the cell wall polysaccharides through bioengineering it is possible to create plants with new and improved qualities that are beneficial as well as being economically attractive. We have presented examples of feasible industrial applications for fibre extraction, malting and animal feed. These applications can all be obtained through transgenic modifications of the plant cell wall polysaccharides, and we have highlighted technological ways to accomplish them. In addition to industrial applications, modifications of the plant cell wall polysaccharides can also facilitate the elucidation of the polysaccharides’ biological function in the wall. To achieve in situ modifications of plant polysaccharides it is possible to up- or down-regulate endogenous enzymes or to introduce genes into the plants by genetic engineering. If the goal of the modifications is to create plants with new, and possibly improved, polysaccharides, while the plants are alive the enzymes should be active at ambient growth temperatures and targeted either to the cell wall or to the place of polysaccharide synthesis. But if phenotypical changes and detrimental effects to the plants is to be avoided it is feasible to use enzymes with high temperature requirements or limit the enzymes’ access to their substrate by targeting the enzymes to cellular compartments not a part of, or involved in synthesis of, the cell wall. In this way the enzymes will not function until the plants are processed after harvest. Production of enzymes in plant is an attractive alternative to microbialbased enzyme production in fermenters. Although the microbial-based system is more flexible and can generate new enzymes at a higher pace, the plant-based system may compete in biofuel and other refinery process when the final cocktail of enzymes necessary for cell wall degradation and modification has been defined.

Acknowledgments The authors thank Wageningen UR Plant Breeding, especially K. Cankar and R. Visser, for providing pictures of transgenic potato tubers. Part of this work was supported by JSPS KAKENHI (No. 19208016) and the JSPS Global COE Program (E-04): In Search of Sustainable Humanosphere in Asia and Africa.

382

■

Plant Polysaccharides

References Arioli, T., Peng, L., Betzner, A.S., et al. (1998) Molecular analysis of cellulose biosynthesis in Arabidopsis. Science, 279, 711–720. Aspegren, K., Mannonen, L., Ritala, A., et al. (1995) Secretion of a heat-stable fungal beta-glucanase from transgenic, suspension-cultured barley cells, Molecular Breeding, 1, 91–99. Atkinson, R.G., Schröder, R., Hallett, I.C., Cohen, D., MacRae, E.A. (2002) Overexpression of polygalacturonase in transgenic apple trees leads to a range of novel phenotypes involving changes in cell adhesion. Plant Physiology, 129, 122–133. Bedford, M.R. (1995) Mechanism of action and potential environmental benefits from the use of feed enzymes. Animal Feed Science and Technology, 53, 145–155. Biswas, G.C.G., Ransom, C., Sticklen, M. (2006) Expression of biologically active Acidothermus cellulolyticus endoglucanase in transgenic maize plants. Plant Science, 171, 617–623. Borkhardt, B., Skjøt, M., Mikkelsen, R., Jørgensen, B., Ulvskov, P. (2005) Expression of a fungal endo-α-1,5-L-arabinanase during stolon differentiation in potato inhibits tuber formation and results in accumulation of starch and tuber-specific transcripts in the stem. Plant Science, 169, 872–881. Boudart, G., Charpentier, M., Lafitte, C., et al. (2003) Elicitor activity of a fungal endopolygalacturonase in tobacco requires a functional catalytic site and cell wall localization. Plant Physiology, 131, 93–101. Brummell, D.A., Harpster, M.H., Civello, P.M., Palys, J.M., Bennett, A.B., Dunsmuir, P. (1999) Modification of expansin protein abundance in tomato fruit alters softening and cell wall polymer metabolism during ripening. Plant Cell, 11, 2203–2216. Burton, R.A., Wilson, S.M., Hrmova, M., et al. (2006) Cellulose synthase-like CslF genes mediate the synthesis of cell wall (1,3;1,4)-beta-d-glucans. Science, 311, 1940–1942. Canam, T., Park, J.Y., Yu, K.Y., Campbell, M.M., Ellis, D.D., Mansfield, S.D. (2006) Varied growth, biomass and cellulose content in tobacco expressing yeast-derived invertases. Planta, 224,1315–1327. Capodicasa, C., Vairo, D., Zabotina, O., et al. (2004) Targeted modification of homogalacturonan by transgenic expression of a fungal polygalacturonase alters plant growth. Plant Physiology, 135, 1294–1304. Cho, H.-T., Cosgrove, D.J. (2000) Altered expression of expansin modulates leaf growth and pedicel abscission in Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the U S A, 97, 9783–9788. Cho, S.K., Kim, J.E., Park, J.A., Eom, T.J., Kim, W.T. (2006) Constitutive expression of abiotic stress-inducible hot pepper CaXTH3, which encodes a xyloglucan endotransglucosylase/hydrolase homolog, improves drought and salt tolerance in transgenic Arabidopsis plants. FEBS Letters, 580, 3136–3144. Côte, F., Hahn, M.G. (1994) Oligosaccharins: structures and signal transduction. Plant Molecular Biology, 26, 1379–1411. Dai, Z., Hooker, B.S., Anderson, D.B., Thomas, S.R. (2000a) Expression of Acidothermus cellulolyticus endoglucanase E1 in transgenic tobacco: biochemical characteristics and physiological effects. Transgenic Research, 9, 43–54. Dai, Z., Hooker, B.S., Anderson, D.B., Thomas, S.R. (2000b) Improved plant-based production of E1 endoglucanase using potato: expression optimization and tissue targeting. Molecular Breeding, 6, 277–285.

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

383

Dai, Z.Y., Hooker, B.S., Quesenberry, R.D., Gao, J.W. (1999) Expression of Trichoderma reesei exo-cellobiohydrolase I transgenic tobacco leaves and calli. Applied Biochemistry and Biotechnology, 77–79, 689–699. Dai, Z., Hooker, B.S., Quesenberry, R.D., Thomas, S.R. (2005) Optimization of Acidothermus cellulolyticus endoglucanase (E1) production in transgenic tobacco plants by transcriptional, post-transcription and post-translational modification. Transgenic Research, 14, 627–643. Enkerli, J., Felix, G., Boller, T. (1999) The enzymatic activity of fungal xylanase is not necessary for its elicitor activity. Plant Physiology, 121, 391–397. Encina, A., Sevillano, J.M., Acebes, J.L., Alvarez, J. (2002) Cell wall modifications of bean (Phaseolus vulgaris) cell suspensions during habituation and dehabituation to dichlobenil. Physiologia Plantarum, 114, 182–191. Farrokhi, N., Burton, R.A., Brownfield, L., et al. (2006) Plant cell wall biosynthesis: genetic, biochemical and functional genomics approaches to the identification of key genes. Plant Biotechnology Journal, 4, 145–167. Ferrari, S., Galletti, R., Pontiggia, D., et al. (2008) Transgenic expression of a fungal endo-polygalacturonase increases plant resistance to pathogens and reduces auxin sensitivity. Plant Physiology, 146, 669–681. Giddings, G., Allison, G., Brooks, D., Carter, A. (2000) Transgenic plants as factories for biopharmaceuticals. Nature Biotechnology, 18, 1151–1155. Gray, K.A., Zhao, L., Emptage, M. (2006) Bioethanol. Current Opinion in Chemical Biology, 10, 141–146. Hahn-Hägerdal, B., Galbe, M., Gorwa-Grauslund, M.F., Lidén, G., Zacchi, G. (2006) Bio-ethanol – the fuel of tomorrow from the residues of today. Trends in Biotechnology, 24, 549–556. Hanania, U., Avni, A. (1997) High-affinity binding site for ethylene-inducing xylanase elicitor on Nicotiana tabacum membranes. Plant Journal, 12, 113– 120. Harpster, M.H., Dawson, D.M., Nevins, D.J., Dunsmuir, P., Brummell, D.A. (2002) Constitutive overexpression of a ripening-related pepper endo-1,4-beta-glucanase in transgenic tomato fruit does not increase xyloglucan depolymerization or fruit softening. Plant Molecular Biology, 50, 357–369. Hasunuma, T., Fukusaki, E., Kobayashi, A. (2004) Expression of fungal pectin methylesterase in transgenic tobacco leads to alteration in cell wall metabolism and a dwarf phenotype. Journal of Biotechnology, 111, 241–251. Hasunuma, T., Fukusaki, E., Kobayashi, A. (2003) Methanol production is enhanced by expression of an Aspergillus niger pectin methylesterase in tobacco cells. Journal of Biotechnology, 106, 45–52. Hayashi, T. (1989) Xyloglucans in the primary cell wall. Annual Review of Plant Physiology and Plant Molecular Biology, 40, 139–168. Heyer, A.G., Raap, M., Schroeer, B., Marty, B., Willmitzer, L. (2004) Cell wall invertase expression at the apical meristem alters floral, architectural, and reproductive traits in Arabidopsis thaliana. Plant Journal, 39, 161–169. Horvath, H., Huang, J., Wong, O., et al. (2000) The production of recombinant proteins in transgenic barley grains. Proceedings of the National Academy of Sciences of the U S A, 97,1914–1929. Hyunjong, B., Lee, D.S., Hwang, I. (2006) Dual targeting of xylanase to chloroplasts and peroxisomes as a means to increase protein accumulation in plant cells. Journal of Experimental Botany, 57, 161–169.

384

■

Plant Polysaccharides

Jensen, L.G., Olsen, O., Kops, O., Wolf, N., Thomsen, K.K., von Wettstein, D. (1996) Transgenic barley expressing a protein-engineered, thermostable (1,3–1,4)-betaglucanase during germination. Proceedings of the National Academy of Sciences of the U S A, 93, 3487–3491. Jensen, L.G., Politz, O., Olsen, O., Thomsen, K.K., von Wettstein, D. (1998) Inheritance of a codon-optimized transgene expressing heat stable (1,3–1,4)-β-glucanase in scutellum and aleurone of germinating barley. Hereditas, 129, 215–225. Jin, R., Richter, S., Zhong, R., Lamppa, G.K. (2003) Expression and import of an active cellulase from a thermophilic bacterium into the chloroplast both in vitro and in vivo. Plant molecular biology, 51, 493–507. Jørgensen, H., Kristensen, J.B., Feldby, C. (2007) Enzymatic conversion of lignocellulose into fermentable sugars: Challenges and opportunities. Biofuels Bioproducts and Biorefining, 1, 119–134. Kawazu, T., Ohta, T., Ito, K., et al. (1996) Expression of a Ruminococcus albus cellulase gene in tobacco suspension cells. Journal of Fermentation and Bioengineering, 82, 205–209. Kawazu, T., Sun, J.L., Shibata, M., Kimura, T., Sakka, K., Ohmiya, K. (1999) Expression of a bacterial endoglucanase gene in tobacco increases digestibility of its cell wall fibers. Journal of Bioscience and Bioengineering, 88, 421–425. Keegstra, K., Walton, J. (2006) Beta-glucans – brewer ’s bane, dietician’s delight. Science, 311, 1872–1873. Kimura, T., Mizutani, T., Tanaka, T., Koyama, T., Sakka, K., Ohmiya, K. (2003) Molecular breeding of transgenic rice expressing a xylanase domain of the xynA gene from Clostridium thermocellum. Applied Microbiology and Biotechnology, 62, 374–379. Lane, D.R., Wiedemeier, A., Peng, L., et al. (2001) Temperature-sensitive alleles of RSW2 link the KORRIGAN endo-1,4-β-glucanase to cellulose synthesis and cytokinesis in Arabidopsis. Plant Physiology, 126, 278–288. Lin, Y., Tanaka, S. (2006) Ethanol fermentation from biomass resources: current state and prospects. Applied Microbiology and Biotechnology, 69, 627–642. Lionetti, V., Raiola, A., Camardella, L., et al. (2007) Overexpression of pectin methylesterase inhibitors in Arabidopsis restricts fungal infection by Botrytis cinerea. Plant Physiology, 143, 1871–1880. Liu, J.-H., Selinger, L.B., Cheng, K.-J., Beauchemin, K.A., Moloney, M.M. (1997) Plant seed oil-bodies as an immobilization matrix for a recombinant xylanase from the rumen fungus Neocallimastix patriciarum. Molecular Breeding, 3, 463–470. Liu, Y.B., Lu, S.M., Zhang, J.F., Liu, S., Lu, Y.T. (2007) A xyloglucan endotransglucosylase/hydrolase involves in growth of primary root and alters the deposition of cellulose in Arabidopsis. Planta, 226, 1547–1560. Malathi, V., Devegowda, G. (2001) In vitro evaluation of nonstarch polysaccharide digestibility of feed ingredients by enzymes. Poultry Science, 80, 302–305. Manfield, I.W., Orfila, C., McCartney, L., et al. (2004) Novel cell wall architecture of isoxaben-habituated Arabidopsis suspension-cultured cells: global transcript profiling and cellular analysis. Plant Journal, 40, 260–275. Martín, I., Dopico, B., Muñoz, F.J., et al. (2005) In vivo expression of a Cicer arietinum beta-galactosidase in potato tubers leads to a reduction of the galactan side-chains in cell wall pectin. Plant Cell Physiology, 46,1613–1622. Matlho, G., Himathongkham, S., Riemann, H., Kass, P. (1997) Destruction of Salmonella enteritidis in poultry feed by combination of heat and propionic acid. Avian Diseases, 41, 58–61.

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

385

Matsumoto, T., Takeda, T., Sakai, F., Hayashi, T. (1997) A xyloglucan-specific endo1,4-β-glucanase isolated from auxin-treated pea stems. Plant Physiology, 114, 661–667. Minic, Z., Jouanin, L. (2006) Plant glycoside hydrolases involved in cell wall polysaccharide degradation. Plant Physiology and Biochemistry, 44, 435–449. Mohnen, D. (1999) Biosynthesis of pectins and galactomannans. Comprehensive Natural Products Chemistry, (ed. B.M. Pinto). Elsevier. Montalvo-Rodriguez, R., Haseltine, C., Huess-LaRossa, K., et al. (2000) Autohydrolysis of plant polysaccharides using transgenic hyperthermophilic enzymes. Biotechnology and Bioengineering, 70, 151–159. Mouille, G., Ralet, M.C., Cavelier, C., et al. (2007) Homogalacturonan synthesis in Arabidopsis thaliana requires a Golgi-localized protein with a putative methyltransferase domain. Plant Journal, 50, 605–614. Musialak, M., Wróbel-Kwiatkowska, M., Kulma, A., Starzycka, E., Szopa, J. (2008) Improving retting of fibre through genetic modification of flax to express pectinases. Transgenic Research, 17, 133–147. Mølhøj, M., Ulvskov, P., Dal Degan, F. (2001) Characterization of a functional soluble form of a Brassica napus membrane-anchored endo-1,4-β-d-glucanase heterologously expressed in Pichia pastoris. Plant Physiology, 127 (2), 674–684. Nuutila, A.M., Ritala, A., Skadsen, R.W., Mannonen, L., Kauppinen, V. (1999) Expression of fungal thermotolerant endo-1,4-beta-glucanase in transgenic barley seeds during germination. Plant Molecular Biology, 41, 777–783. Obembe, O.O., Jacobsen, E., Timmers, J., et al. (2007) Promiscuous, non-catalytic, tandem carbohydrate-binding modules modulate the cell-wall structure and development of transgenic tobacco (Nicotiana tabacum) plants. Journal of Plant Research, 120, 605–617. Ohmiya, K., Sakka, K., Kimura, T., Morimoto, K. (2003) Application of microbial genes to recalcitrant biomass utilization and environmental conservation. Journal of Bioscience and Bioengineering, 95, 549–561. Oomen, R.J., Doeswijk-Voragen, C.H., Bush, M.S., et al. (2002) In muro fragmentation of the rhamnogalacturonan I backbone in potato (Solanum tuberosum L.) results in a reduction and altered location of the galactan and arabinan side-chains and abnormal periderm development. Plant Journal, 30, 403–413. Oraby, H., Venkatesh, B., Dale, B., et al. (2007) Enhanced conversion of plant biomass into glucose using transgenic rice-produced endoglucanase for cellulosic ethanol. Transgenic Research, 16, 739–749. Osteryoung, K.W., Toenjes, K., Hall, B., Winkler, V., Bennett, A.B. (1990) Analysis of tomato polygalacturonase expression in transgenic tobacco. Plant Cell, 2, 1239–1248. Park, Y.W., Baba, K., Furuta, Y., et al. (2004) Enhancement of growth and cellulose accumulation by overexpression of xyloglucanase in poplar. FEBS Letters, 564, 183–187. Park, Y.W., Tominaga, R., Sugiyama, J., et al. (2003) Enhancement of growth by expression of poplar cellulase in Arabidopsis thaliana. Plant Journal, 33, 1099–1106. Patel, M., Johnson, J.S., Brettell, R.I.S., Jacobsen, J., Xue, G.-P. (2000) Transgenic barley expressing a fungal xylanase gene in the endosperm of the developing grains. Molecular Breeding, 6, 113–123. Peng, L., Kawagoe, Y., Hogan, P., Delmer, D. (2002) Sitosterol-β-glucoside as primer for cellulose synthesis in plants. Science, 295, 147–150.

386

■

Plant Polysaccharides

Peng, L., Xiang, F., Roberts, E., et al. (2001) The experimental herbicide CGA 325’615 inhibits synthesis of crystalline cellulose and causes accumulation of noncrystalline β-1,4-glucan associated with CesA protein. Plant Physiology, 126, 981–992. Pilling, J., Willmitzer, L., Fisahn, J. (2000) Expression of a Petunia inflata pectin methyl esterase in Solanum tuberosum L. enhances stem elongation and modifies cation distribution. Planta, 210, 391–399. Ragauskas, A.J., Williams, C.K., Davison, B.H., et al. (2006) The path forward for biofuels and biomaterials. Science, 311, 484–489. Ransom, C., Balan, V., Biswas, G., Dale, B., Crockett, E., Sticklen, M. (2007) Heterologous Acidothermus cellulolyticus 1,4-beta-endoglucanase E1 produced within the corn biomass converts corn stover into glucose. Applied Biochemistry and Biotechnology, 137, 207–219. Ridley, B.L., O’Neill, M.A., Mohnen, D. (2001) Pectins: structure, biosynthesis, and oligogalacturonide-related signaling. Phytochemistry, 57, 929–967. Rishi, A.S., Nelson, N.D., Goyal, A. (2001) Molecular farming in plants: A current perspective. Journal of Plant Biochemistry and Biotechnology, 10, 1–12. Rochange, S.F., McQueen-Mason, S. (2000) Expression of a heterologous expansin in transgenic tomato plants. Planta, 211, 583–586. Rochange, S.F., Wenzel, C.l., McQueen-Mason, S. (2001) Impaired growth in trangenic plants over-expressing an expansin isoform. Plant Molecular Biology, 46, 581–589. Rose, J.K.C., Braam, J., Fry, S.C., Nishitani, K. (2002) The XTH family of enzymes involved in xyloglucan endotransglucosylation and endohydrolysis: current perspectives and a new unifying nomenclature. Plant Cell Physiology, 43, 1421–1435. Safra-Dassa, L., Shani, Z., Danin, A., Roiz, L., Shoseyov, O., Wolf, S. (2006) Growth modulation of transgenic potato plants by heterologous expression of bacterial carbohydrate-binding module. Molecular Breeding, 17, 355–364. Shani, Z., Dekel, M., Tsabary, G., Goren, R., Shoseyov, O. (2004) Growth enhancement of transgenic poplar plants by overexpression of Arabidopsis thaliana endo1,4–β-glucanase (cel1). Molecular Breeding, 14, 321–330. Sharon, A., Fuchs, Y., Anderson, J.D. (1993) The Elicitation of Ethylene Biosynthesis by a Trichoderma Xylanase Is Not Related to the Cell Wall Degradation Activity of the Enzyme. Plant Physiology, 102, 1325–1329. Shin, Y.K., Yum, H., Kim, E.S., et al. (2006) BcXTH, a Brassica campestris homologue of Arabidopsis XTH9, is associated with cell expansion. Planta, 224, 32–41. Skendi, A., Biliaderis, C.G., Lazaridou, A., Izydorczyk, M.S. (2003) Structure and rheological properties of water soluble β-glucans from oat cultivars of Avena sativa and Avena bysantina. Journal of Cereal Science, 38, 15–31. Skjøt, M., Kauppinen, S., Kofod, L.V., et al. (2001) Functional cloning of an endoarabinanase from Aspergillus aculeatus and its heterologous expression in A. or oryzae and tobacco. Molecular Genetics and Genomics, 265, 913–921. Skjøt, M., Pauly, M., Bush, M.S., Borkhardt, B., McCann, M.C., Ulvskov, P. (2002) Direct interference with rhamnogalacturonan I biosynthesis in Golgi vesicles. Plant Physiology, 129, 95–102. Somerville, C. (2006) Cellulose synthesis in higher plants. Annual Reviews of Plant Biology, 22, 53–78.

Enzymatic Modification of Plant Cell Wall Polysaccharides

■

387

Sonnewald, U., Hajirezaei, M.R., Kossmann, J., Heyer, A., Trethewey, R.N., Willmitzer, L. (1997) Increased potato tuber size resulting from apoplastic expression of a yeast invertase. Nature Biotechnology, 15, 794–797. Sun, Y., Cheng, J.J., Himmel, M.E., et al. (2007) Expression and characterization of Acidothermus cellulolyticus E1 endoglucanase in transgenic duckweed Lemna minor 8627. Bioresource Technology, 98, 2866–2872. Sørensen, S.O., Pauly, M., Bush, M., et al. (2000) Pectin engineering: modification of potato pectin by in vivo expression of an endo-1,4-beta-D-galactanase. Proceedings of the National Academy of Sciences of the U S A, 97, 7639–7644. Ulvskov, P., Wium, H., Bruce, D., et al. (2005) Biophysical consequences of remodeling the neutral side chains of rhamnogalacturonan I in tubers of transgenic potatoes. Planta, 220, 609–620. von Wettstein, D. (2007) From analysis of mutants to genetic engineering. Annual Review of Plant Biology, 58, 1–19. von Wettstein, D., Mikhaylenko, G., Froseth, J.A., Kannangara, C.G. (2000) Improved barley broiler feed with transgenic malt containing heat-stable (1,3–1,4)-betaglucanase. Proceedings of the National Academy of Sciences of the U S A, 97, 13512–13517. von Wettstein, D., Warner, J., Kannangara, C.G. (2003) Supplements of transgenic malt or grain containing (1,3–1,4)-beta-glucanase increase the nutritive value of barley-based broiler diets to that of maize. British Poultry Science, 44, 438–449. Willats, W.G.T., Knox, P., Mikkelsen, J.D. (2006) Pectin: new insights into an old polymer are starting to gel. Trends in Food Science and Technology, 17, 97–104. Xue, G.P., Patel, M., Johnson, J.S., Smyth, D.J., Vickers, C.E. (2003) Selectable markerfree transgenic barley producing a high level of cellulase (1,4-beta-glucanase) in developing grains. Plant Cell Reproduction, 21, 1088–1094. Yu, L.X., Gray, B.N., Rutzke, C.J., Walker, L.P., Wilson, D.B., Hanson, M.R. (2007) Expression of thermostable microbial cellulases in the chloroplasts of nicotine-free tobacco. Journal of Biotechnology, 131, 362–369. Ziegelhoffer, T., Raasch, J.A., Austin-Phillips, S. (2001) Dramatic effects of truncation and sub-cellular targeting on the accumulation of recombinant microbial cellulase in tobacco. Molecular Breeding, 8, 147–158. Ziegelhoffer, T., Will, J., Austin-Phillips, S. (1999) Expression of bacterial cellulase genes in transgenic alfalfa (Medicago sativa L.), potato (Solanum tuberosum L.) and tobacco (Nicotiana tabacum L.). Molecular Breeding, 5, 309–318. Ziegler, M.T., Thomas, S.R., Danna, K.J. (2000) Accumulation of a thermostablendo1,4-β-d-glucanase in the apoplast of Arabidopsis thaliana leaves. Molecular Breeding, 6, 37–46.

Annual Plant Reviews (2011) 41, 389–408 doi: 10.1002/9781444391015.ch16

Chapter 16

PRODUCTION OF HETEROLOGOUS STORAGE POLYSACCHARIDES IN POTATO PLANTS Xing-Feng Huang1, Jean-Paul Vincken1,2, Richard G.F. Visser1 and Luisa M. Trindade1 1 Graduate School Experimental Plant Sciences, Wageningen UR – Plant Breeding, Wageningen University, P.O. Box 386, 6700 AJ Wageningen, The Netherlands 2 Present address: Laboratory of Food Chemistry, Wageningen University, P.O. Box 8129, 6700 EV Wageningen, The Netherlands

Manuscript received April 2008

Abstract: Starch is the most important storage polysaccharide in higher plants. This polysaccharide is used in many industrial applications as it is abundant, renewable and biodegradable and it can be modified into a wide range of products used in food, animal feed, pharmaceuticals and industry. With the understanding of the starch biosynthesis pathway and the isolation of many of the genes involved in this process, the potential for producing novel starches with improved functionality through genetic modification of plants has become more clear. Several different strategies have been used to achieve the rational design of renewable polymers to meet specific requirements. One of these strategies is the regulation of starch biosynthesis and granule assembly in higher plants. A second strategy involves the expression of heterologous genes encoding biosynthetic or modifying enzymes from other organisms to produce novel products with novel functional properties. In this chapter, these recent developments in starch modification in potato plants, such as alteration of starch composition and production of novel storage polysaccharides, are reviewed. Keywords: alternan; dextran; fructan; mutan; starch modification; storage polysaccharides; transgenic potato Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

389

390

■

Plant Polysaccharides

16.1 Introduction Carbohydrates are the most abundant biological compounds on Earth. More than 100 billion tonnes of CO2 and H2O are converted into carbohydrates and other plant compounds by green plants during the process of photosynthesis every year (Nelson & Cox 2004). Most carbohydrates found in nature occur as polysaccharides containing from twenty to hundreds or thousands of monosaccharide units. Different kinds of polysaccharides differ in their recurring monosaccharide units, chain length, the types of bonds linking the units, and the degree of branching. Carbohydrates have different biological functions such as energy storage, like starch and glycogen, or structural components, (such as cellulose, hemicellulose, pectin and chitin; they) play important roles in the lubrication of skeletal joints and participate in recognition and adhesion between cells. Because of this functional diversity, polysaccharides are exploited in both food and non-food industries (Ramesh & Tharanathan 2003).

16.2 Starch: native and modified starch, consequences for its properties Starch is the most abundant storage carbohydrate in higher plants. It can be found in different plant organs such as cereal grains, roots, tubers, fruits and leaves and when it is stored in amyloplasts it is in a granular form. Starch is a versatile and highly useful carbon source because it is abundant, cheap and naturally renewable. Furthermore, it can be used both in food and as raw material for many industrial processes. Starch synthesized in the chloroplasts of green leaves as transitory starch during the day can provide carbon for non-photosynthetic metabolism at night. Starch produced in amyloplasts of tuberous tissues can act as a carbon store – storage starch. Starch granules (ranging in size from 1 μm to 100 μm, depending on the source) consist of an amorphous and a semicrystalline region, commonly known as growth rings (Myers et al. 2000), and have a complex structure with a hierarchical order composed of amylose and amylopectin. Amylose is an essentially linear molecule of α-1,4 linked glucose residues. It is the main component of the amorphous layers of the granules and typically constitutes about 20–25% of the granule mass. In cereals the amylose content can go up to 30%. Amylopectin is located in the semicrystalline region of the granules having the same α-1,4 linked glucan backbone as amylose but with 5–6% of α-1,6 bonds at the branch points, and accounts for the rest of starch granules. Amylopectin (molecular weight of 107–8 Da) is a far larger molecule than amylose (molecular weight of 5 × 105 Da). Amylopectin can form double helices, which are the basis of the semicrystalline structure of the starch granules.

Production of Heterologous Storage Polysaccharides in Potato Plants

■

391

16.2.1 Starch biosynthesis in storage organs The synthesis of starch granules takes place in the chloroplasts/amyloplasts of higher plants and involves three major steps: (i1) the transport of Glc-6-P into the plastids, (ii2) the synthesis of ADP-glucose from Glc-1-P and (iii3) the synthesis of starch from ADP-Glc, where many starch enzymes act together (Fig. 16.1). Sucrose transported by the phloem through the whole plant is the source of carbon for starch biosynthesis. It is converted into hexose-phosphate sugars and transported into the amyloplast to supply Glc-6-P (Kammerer et al. 1998). Sucrose can enter the cell via the apoplast, either by sucrose transporters or hexose transporters after hydrolysis, or via plasmodesmata (symplast). Once inside the cell, sucrose can be transported into the vacuole and/or hydrolysed by acid and alkaline invertases to glucose and fructose. Alternatively, sucrose synthase (SuSy) can convert sucrose into UDP-glucose and fructose in the cytosol. The hexoses formed in the vacuole can be transported into the cytosol and phosphorylated. The first committed step in the biosynthesis of both transient starch in chloroplasts and storage starch in amyloplasts is the synthesis of ADP-glucose from Glc-1-P and ATP, which is catalysed by ADP-glucose pyrophosphorylase (AGPase; EC 2.7.7.27). Starch synthase enzymes (SS; EC 2.4.1.21) transfer glucose from ADP-Glc to the non-reducing end of an α-1,4 glucan to synthesize the insoluble glucan polymers amylose and amylopectin. Based on sequence similarities in Arabidopsis, five isoforms of starch synthases have been predicted in plants: GBSS, SSI, SSII, SSIII and SSIV. The granule-bound starch synthase (GBSS), including GBSSI (in storage tissues) and GBSSII (in leaves and other non-storage tissues), catalyses the formation of amylose. The starch branching enzymes (SBEs; EC 2.4.1.18) creates α-1,6 linkages by cleaving α-1,4 linkages to generate the branched structure of the amylopectin molecule. Subsequently, amylopectin is crystallized into starch granules by concerted efforts of starch debranching enzymes (DBE; EC 2.4.1.41) and disproportionating enzyme (D-enzyme; EC 2.4.1.25) (Ball et al. 1998). Native starch has a limited number of uses because it does not have most of the desired characteristics demanded by industry. But it can be modified by physical treatment, chemical reagents or enzymes to suit various applications and to meet the specific needs of end users. Recently, novel starches with improved functionalities have been produced in planta through genetic modification techniques. By inhibiting native genes that are involved in starch biosynthesis, starches with changed amylose/amylopectin ratio and altered degree of branching of amylopectin have been achieved in plants. 16.2.2 Alteration of starch composition through the modeling of plant genes 16.2.2.1 Amylose-free starch The first genetically modified amylose-free (waxy) starch in potato was generated by silencing of the GBSS gene antisense RNA (Visser et al. 1991).

Hexokinase

Cytosol

Glucose 1P Glucose 6P

PGM

ADP-GPP

SS DBE

Glucose 6P

PGM

Glucose 1P

ADP-GPP

P

Plastid

Amylopectin

SP

Figure 16.1 The starch biosynthetic pathway. SuSy, sucrose synthase; OPPP, oxidative pentose phosphate pathway; Glucose 1P, glucose1-phosphate; UDP-GPP, UDP-glucose pyrophosphorylase; Glucose 6P, glucose-6-phosphate; PGM, phosphoglucomutase; ADP-GPP, ADPglucose pyrophosphorylase; GBSS, granule bound starch synthase; SS, starch synthases; SBE, starch branching enzymes; P, phosphate monoesters; SP, starch phosphorylases; DBE, debranching enzymes; SP, starch phosphorylases.

OPPP

Starch GBSS

Apoplast

Phloem

DBE SP Amylases Glucoronidase

Glucose

ADP-Glucose

GBSS SBE

Amylose

Symplast

ADP-Glucose

UDP-Glucose

SuSy

Sucrose

Sucrose

UDP-GPP

Invertase

Glucose + Fructose

Glucose 6P

Glycolysis

Vacuole

Hexoses

Invertase

Hexoses

Sucrose

■

Sucrose

Apoplast

Phloem

392 Plant Polysaccharides

Production of Heterologous Storage Polysaccharides in Potato Plants

■

393

Besides these transformed lines, an induced mutant was also obtained (Hovenkamp-Hermelink et al. 1987). Both the GM lines and the natural GBSS mutant genotype have improved paste clarity and stability (Visser et al. 1997). The naturally occurring mutant has been explored commercially for different applications by the Dutch company AVEBE under the brand name ELIANE™. A different amylose-free starch with short-chain amylopectin showing an extremely high freeze–thaw stability was created by simultaneous knock-down of three starch synthase genes (GBSS, SSII and SSIII) (Jobling et al. 2002). 16.2.2.2 High-amylose starch High-amylose starches are of great interest for the starch industry; for example, the development of starch-based films for their unique functional properties (van Soest & Borger 1997; Rindlav-Westling et al. 1998). One of the strategies used for the generation of starch with an increased level of amylose was the down-regulation of genes involved in the synthesis of amylopectin (Jobling et al. 1999). Inhibition of SBE A and B in potato lines has resulted in starch granules with more than 60% amylose (Schwall et al. 2000). One other strategy used for the increase of amylose content in starch granules was the tagging of antibodies to enzymes involved in amylopectin synthesis. A single-domain antibody directed against SBEII in potato was successfully used to produce starches with more than 50% amylose (Jobling et al. 2003). 16.2.3 Modification of starch properties by expression of bacterial proteins in plants Glycogen is a highly branched glucan molecule, like amylopectin, but it has a shorter average length of side chains and it is water-soluble (Myers et al. 2000). The formation of α-(1→6) branch points is catalysed by branching enzymes (BE)and (SBEs) in plants, and by glycogen branching enzymes in E. coli (Nakamura 2002). In order to increase the branching degree of potato tuber starch, several genes encoding branching enzymes involved in glycogen biosynthesis in E. coli were expressed in amylose-containing or/ and amylose-free potato (Stark et al. 1992; Shewmaker et al. 1994; Kortstee et al. 1996). A mutant E. coli ADP-Glc pyrophosphorylase (glgC) has been expressed in potato tuber amyloplasts resulting in increased starch accumulation in the tubers (Stark et al. 1992). Expression of E. coli glycogen synthase (glgA) in potato tuber amyloplasts leads to a number of changes in the starch composition: decreased specific gravity, a reduction in the percentage of starch, a decline in amylose/ amylopectin ratio and a reduced phosphorus content. The starches obtained from these transgenic potato lines have lower viscosity, reduced enthalpy and gelatinization properties when compared with control starches (Shewmaker et al. 1994).

394

■

Plant Polysaccharides

The expression of the E. coli branching enzyme (glgB) in the amylose-free potato mutant resulted in more branched starch structure and up to 25% increase in the branching degree (DE) of amylopectin. (Kortstee et al. 1996). Except for a slightly higher amylopectin percentage and a small difference in granule surface morphology, the expression of glgB in amylose-containing background did not show any significant changes in starch composition and properties (Krohn et al. 1994). In order to synthesize starch acetate in planta, an amyloplast-targeted E. coli maltose acetyltransferase gene (MAT) was expressed in tubers of wild type (Kardal) and mutant amylose-free (amf) potato plants. Acetyl groups were found in the starch granules of transgenic plants, although at a low level (Nazarian Firouzabadi et al. 2007a).

16.3 Production of novel storage polysaccharides in plants In the past decades the tremendous progress in molecular biology has offered valuable information on how to improve crop plants. Modifications of native polymers in planta can yield crops with added nutritional, environmental or commercial value. One example is the production of a freeze– thaw-stable potato starch exhibiting novel physicochemical properties, thereby increasing the number of industrial applications (Jobling et al. 2002). Production of novel polymers in plants by genetic modification is also a great opportunity to obtain plants with unique properties that cannot be generated by conventional breeding (Kok-Jacon et al. 2003). The relatively low costs and the productions of large amounts make plants as bioreactors an interesting option. Plants have been successfully used as bioreactors to produce compounds such as enzymes, antigens and polysaccharides in recent decades. In this review, we focus on recent developments in the use of transgenic potato plants to produce heterologous polysaccharides. 16.3.1 Glucans synthesized in transgenic plants Glucans are polysaccharides consisting of d-glucose monomers and linked by glycosidic bonds. Cellulose (β-1,4-glucan) and starch (α-1,4-and α-1,6glucan) in plants, glycogen (α-1,4-and α-1,6-glucan) in animals and bacteria and dextran (α-1,6-glucan) in bacteria are all glucans with different type of glucosidic linkages, degree and type of branching, length of the glucan chains, sizes and conformation. The industrial applications of glucans are not restricted to the food industry, whre thye are used as stabilizing, emulsifying, sweetening and gelling agents; they can also be used in the non-food industries as viscosifying and water-binding agents (Sutherland 1972; Welman & Maddox 2003). Glucansucrases (E.C. 2.4.1.5) catalyse the polymerization of glucose residues into a large variety of glucans with different sizes, structures and

Production of Heterologous Storage Polysaccharides in Potato Plants

■

395

linkage types (Sidebotham 1974; Monchois et al. 1999). With the exception of amylosucrase (E.C. 2.4.1.4), glucansucrases are extracellular enzymes mainly produced by soil bacteria, Streptococcus species from the oral flora and lactic bacteria Lactococci (Sidebotham 1974; Mooser 1992). Glucansucrases are classified based on the reaction catalysed and their substrate specificity: dextransucrase (DSR) (E.C. 2.4.1.5), alternansucrase (ASR) (E.C. 2.4.1.140), mutansucrase and reuteransucrase (GTF A) (E.C. 2.4.1.-) (van Hijum et al. 2006). Most of the glucansucrases have been included in the GH70 family, except amylosucrase which belongs to the GH13 family (Henrissat & Davies 1997). 16.3.1.1 Alternan Alternan is a unique polymer which is produced by three Leuconostoc mesenteroides strains: NRRL B-1355, NRRL B-1498 and NRRL B-1501 (Jeanes et al. 1954; Seymour et al. 1979; Seymour & Knapp 1980; Cote & Robyt 1982). It consists of alternating α-(1→3)/α-(1→6)-linked glucosyl residues, with approximately 7–11% branching (Seymour et al. 1977). Because of its special structure, alternan is highly soluble, has a low viscosity and is resistant to microbial and mammalian enzymes (Cote 1992; Cote et al. 1997). The alternansucrase (ASR) gene from Leuconostoc mesenteroides NRRL was expressed in potato, resulting in 0.3 and 1.2 mg g−1 fresh weight of alternan in potato tubers (Kok-Jacon et al. 2007). Expression of ASR did not interfere either with plant growth or development or tuber and starch yield, but it changed the morphology of the starch granules (Fig. 16.2). Even though the starch synthesizing genes AGPase and GBSSI were down-regulated, the physicochemical properties of the transgenic starches remain unaltered.

Figure 16.2 Scanning electron microscopy (SEM) analysis of starch granules from untransformed potato plants of the Kardal variety (A, D), transformed with the alternansucrase encoding gene asr (B, E), and with the dextransucrase encoding gene DsrS (C, F).

396

■

Plant Polysaccharides

16.3.1.2 Dextran Dextran is a glucan polymer synthesized by the dextransucrase DSRS (EC 2.4.1.5) of Lactobacillus, Leuconostoc and Streptococcus bacteria in the presence of sucrose (Simpson et al. 1995; Monchois et al. 1996; Kralj et al. 2004). Dextran is used in several industrial applications including chromatographic media and soil conditioner (Heinze et al. 2006), and in biomedical applications due to its biocompatibility properties, slow biodegradability and feasibility of incorporation of molecules into the matrices formed by dextrans (Ko et al. 1991; Mehvar 2000; Sinha & Kumria 2001). Dextran hydrogels and their chemical modifications have been evaluated as carriers for controlled release of drugs to targeted organs by slow dextranase hydrolysis (Hennink & van Nostrum 2002). Dextrans also contribute to human health since they are resistant to mammalian digestive enzymes in the small intestine but are readily fermented in the large intestine. Immobilized dextransucrase with soluble dextranase has been used for synthesis of prebiotic oligosaccharides (Kubik et al. 2004). The most widely studied dextran which contains 95% α-(1→6) linkages and 5% α-(1→3) branch linkages is produced by Leuconostoc mesenteroides NRRL B-512F strain (Groenwall & Ingelman 1948). The dextransucrase (DsrS) gene from L. mesenteroides was expressed in an amylose-containing potato cultivar (cv. Kardal) and the amylose-free (amf) potato mutant. The dextran was only detected in tuber juice and no dextran was detected inside the starch granule. Dextran concentration appeared two times higher in the Kardal (about 1.7 mg/g FW) than in the amf transformants. Starch granule morphology was affected by the dextran accumulation and the granules of the transformants exhibited round, protruding structures (Fig. 16.2). The accumulation of dextran did not interfere with the physicochemical properties of the transgenic starches or the starch content (Kok-Jacon et al. 2005a). 16.3.1.3 Mutan Mutan, a water-insoluble glucan containing a high proportion (up to 90%) α-(1→3) glucosidic linkages and associated with α-(1→6) linkages, was initially identified in S. mutans OMZ176 in 1970 (Guggenheim 1970) and is mainly produced by various Streptococci (Hamada & Slade 1980). Mutansucrase (E.C. 2.4.1.5 – GTFI) synthesizes mutan in the presence of sucrose and can be found in L. mesenteroides NRRL B-523, B-1149 and several Streptococcus strains (Sidebotham 1974; Mooser 1992). Mutan polymers are involved in the adhesion of oral flora microorganisms on the tooth surface (Hamada & Slade 1980) and account for about 70% of the carbohydrates present in dental plaque (Loesche 1986). This glucan polymer could potentially be used as functional food (pre-biotics) (Tuohy et al. 2003), but so far no specific applications of mutan polymers have been developed. A full-length (GtfI) and a truncated mutansucrase gene referred to as GtfICAT (GtfI without glucan-binding domain) isolated from Streptococcus

Production of Heterologous Storage Polysaccharides in Potato Plants

■

397

Figure 16.3 Scanning electron microscopy (SEM) analysis of the starch granules from untransformed potato plants of the Kardal variety (A, D), transformed with the full-length GtfI (B, E) and a truncated mutansucrase gene GtfICAT (C, F).

downei were expressed in an amylose-containing potato cultivar (c.v. Kardal) to produce mutan polymers (Kok-Jacon et al. 2005b). GtfICAT expression induced more severe development alterations, such as tuber phenotype and starch granule morphology, than that of GtfI (Fig. 16.3). Expression of the GtfICAT gene resulted in the adhesion of mutan material to starch granules. Expression of the GtfICAT interfered with the starch biosynthetic pathway as AGPase and GBSSI genes were down-regulated and starch content was decreased in the GtfICAT transformants. The rheological properties of the GtfICAT starches showed a higher retrogradation during cooling of the starch paste. Mutan can adhere to starch granules when a truncated mutansucrase (GtfICAT) is expressed in potato tuber, but it was not incorporated in the starch granule (Kok-Jacon et al. 2005b). In order to facilitate the incorporation of mutan polymers in the starch granule, GtfICAT was fused to the N- or C-terminus of a starch-binding domain (SBD) and introduced in potato plants (cv. Kardal and amf) (Nazarian Firouzabadi et al. 2007b). It was found that mutan was present inside starch granules, demonstrating that granuletargeting of GTFICAT was successful. Interestingly, the granules of these transformants had porous and spongy surfaces (Fig. 16.3), a higher melting temperature, and a more pronounced retrogradation behaviour, compared to those from controls. Except for the T onset (temperature at which starch granules start to gelatinize), these alterations were less pronounced than those observed in transformants expressing GtfICAT gene without appended SBD. In vitro production of mutan by incubating starch granules from transformants with an excess of sucrose was not evidenced. The results show

398

■

Plant Polysaccharides

that expression of granule-bound and ‘soluble’ GTFICAT can affect starch biosynthesis differently, and that the appended SBD inhibits the activity of GTFICAT in the engineered fusion protein (Nazarian Firouzabadi et al. 2007b). 16.3.1.4 Other novel carbohydrates Recently, some other carbohydrates with potential interest for the food industry, such as palatinose, cyclodextrins and trehalose, were synthesized in transgenic potato tubers. Palatinose (isomaltulose, 6-O-α-d-glucopyranosyld-fructose) is a structural isomer of sucrose with very similar physicochemical properties but is neither metabolizable nor cariogenic (Börnke et al. 2001). Expression of a bacterial sucrose isomerase (palI) which catalyses the conversion of sucrose into isomaltulose in transgenic potato tubers led to a nearly quantitative conversion of sucrose into palatinose (Börnke et al. 2002). Cyclodextrins (CDs) are cyclic oligosaccharides containing six (α), seven (β), or eight (γ) glucose residues. A cyclodextrin glycosyltransferases (CGT) gene from Klebsiella was transferred into potato to produce CDs. CDs were accumulated in tubers of transgenic potatoes (Oakes et al. 1991). To produce trehalose (α-O-glucopyranosyl-[1-l]-α-d-glucopyranoside), the otsA and otsB genes from kcherichia coli, which encode trehalose-6-phosphate synthase and trehalose-6-phosphate phosphatase respectively, were transferred into tobacco (Nicotiana tabacum) and potato (Solanum tuberosum). Up to 0.41 and 4.04 mg g−1 fresh weight of trehalose was produced in transgenic tobacco leaves and transgenic potato microtubers, respectively (Goddijn et al. 1997). 16.3.2 Fructans synthesized heterologously in transgenic potato plants Fructan is a polymer of fructose molecules synthesized from sucrose by various microbial and plant species. In plants, fructans are reserve carbohydrates (about 15% of all flowering plant species store fructans) and are likely to be involved in the adaptation of plant tissues to cold and arid environments (Hendry & Wallace 1993; Pilon-Smits et al. 1995). Microbial fructans have protective functions and occasionally energetic functions. Among the fructan-storing plants are many of significant economically important crops, such as cereals (such as e.g. barley, wheat, and oat), vegetables (such as e.g. chicory, onion, and lettuce), ornamentals (such as e.g. dahlia and tulip), and also forage grasses (such as e.g. Lolium and Festuca) (Hendry & Wallace 1993). Fructans of different origin can differ in the degree of polymerization (DP), the presence of branches, the type of linkage and the position of the glucose residue. In contrast to starch, which is synthesized in the amyloplast, fructans are synthesized and stored in the vacuole. Although sucrose is used as a precursor for both fructans and starch, in fructans the sugars are not phosphorylated.

Production of Heterologous Storage Polysaccharides in Potato Plants

■

399

Depending on the source, fructans can contain from 2 to more than 105 fructose units linked by β-(2→1) (inulin-type) or β-(2→6) (levan-type) glycosidic bonds. Plant fructans range from about ten up to a few hundred fructose units, while bacterial fructans present as part of extracellular polysaccharides are much larger, with a DP ranging from 104 to 106. Fructans with degrees of polymerization (DP) from 2 to 10 are commonly known as fructooligosaccharides (FOS) (Waterhouse & Chatterton 1993). Fructans can be found in a broad range of bacteria, including plant pathogens and the bacteria present in oral and gut floras of animals and humans (such as Bacillus, Streptococcus, Pseudomonas, Erwinia, and Actinomyces) (Hendry & Wallace 1993). In general, bacteria produce fructan molecules consisting mainly of β-(2→6)-linked fructosyl residues, occasionally containing β-(2→1)-linked branches (Dedonder 1966). Unlike bacterial fructans, which have seemingly uniform structure, plant fructans have much more structural diversity (Pollock & Cairns 1991). One of the simplest fructans, which is present in plants belonging to the Asterales (e.g. chicory), is linear inulin, which consists of β-(2→1)-linked fructose residues. Fructans of the levan type (also called phleins in plants) are found in grasses (Poaceae). The basic structure of this fructan type is a linear β-(2→6)linked fructose polymer as found, for example, in big bluegrass (Poa secunda) ([Chatterton & Harrison 1997; Wei et al. 2002]. ). Mixed fructan types, called gramminans, can also be present and consist of β-(2→6)-linked fructose residues with β-(2→1) branches (Sims et al. 1992; Livingston et al. 1993].). 16.3.2.1 Enzymes of fructan biosynthesis Plant fructans are synthesized from sucrose by the action of two or more different fructosyltransferases. Two enzymes are involved in the synthesis of inulin, the simplest form of fructan (Edelman & Jefford 1968). The first enzyme, sucrose:sucrose 1-fructosyltransferase (1-SST, EC 2.4.1.99), initiates fructan synthesis by catalysing the transfer of a fructosyl residue from sucrose to another sucrose molecule, resulting in the formation of the trisaccharide, 1-kestose. The second enzyme, fructan:fructan 1-fructosyltransferase (1-FFT, EC 2.4.1.100), transfers fructosyl residues from a fructan molecule to another fructan molecule or to sucrose. The action of 1-SST and 1-FFT results in the formation of a mixture of fructan molecules with different chain lengths. The 1-SST activity is present in all fructan-producing plants and the encoding gene has been isolated from different species (van der Meer et al. 1998; Vijn et al. 1998; Lüscher et al. 2000; Chalmers et al. 2003). In general, 1-SST and 1-FFT are both necessary and sufficient for the synthesis of inulintype fructans. Other enzymes are needed for the synthesis of more complex fructans, such as fructan:fructan 6G-fructosyltransferase (6G-FFT) which can produce neokestose by linking 1-kestose to sucrose via β-(2→6) linkage (Vijn et al. 1997), and the sucrose:fructan 6-fructosyltransferase (6-SFT, EC 2.4.1.10) which uses 1-kestose to synthesize the branched bifurcose (1- and 6-kestotetraose) (Duchateau et al. 1995). The biosynthesis of the complex

400

■

Plant Polysaccharides

mixed fructans and graminans has not been well studied. The combined action of 1-SST, 1FFT, 6-SFT and 6G-FFT could theoretically produce graminan (Ritsema & Smeekens 2003). In bacteria, fructans are synthesized extracellularly by bacterial fructosyltransferase (FTF) enzymes directly from sucrose. Most of the known bacterial FTFs are levansucrases (Lev; EC 2.4.1.10) which synthesize fructan polymers composed of β-(2→6)-linked fructose units (levan) (Gross et al. 1992; van Hijum et al. 2004). Although most of the bacteria produce levans, there are some FTF enzymes such as inulosucrases (Inu; EC 2.4.1.9), able to generate β-(2→1)-linked fructan polymers (inulin) (Baird et al. 1973). Bacterial inulin production is rare and has been reported only for Streptococcus mutans (Ebisu et al. 1975), Lactobacillus reuteri 121 (Olivares-Illana et al. 2002) and Leuconostoc citreum (van Hijum et al. 2002). In addition to fructosyltransferase activity, bacterial levansucrases can transfer fructosyl units to a variety of acceptor substrates, such as water (sucrose hydrolysis) and to other sugars such as sucrose (kestose synthesis), fructan (fructan polymerization), glucose (sucrose synthesis) and fructose (bifructose synthesis) (Cote & Ahlgren 1993; Banguela & Hernández 2006). 16.3.2.2 Industrial applications of fructans Fructans such as FOS and inulin have applications in the food and non-food industries (Fuchs 1991). Fructans have a low calorific value and dietary fibrelike properties and are considered the typical representative of prebiotics. Small fructans with DPs of 3–6 are sweet tasting and therefore constitute natural low-calorie sweeteners. High-DP fructans are now being used in alimentary products where they can replace fat. Because of its lower calorific value, neutral taste and excellent physical characteristics, inulin is extensively used as a food ingredient, as a fat replacer and dietary fibre. It is used in yogurts, spreads, and ice cream as a low-calorie replacement for fat (Ritsema & Smeekens 2003). Levan can be used as an emulsifier or encapsulating agent in a wide range of products, including biodegradable plastics, cosmetics, glues, textile coatings and detergents. Levan can substitute for dextran as a blood plasma volume extender, and it has been found to have antitumor and immunomodulatory activities in mice (Calazans et al. 2000].). 16.3.2.3 Fructans in transgenic plants The health-promoting properties of fructans increase their interest for industrial applications. The first fructosyltransferase was purified, cloned and characterized in 1995 (Sprenger et al. 1995). Since the mid-1990s, some bacterial and plant fructosyltransferase genes have been transferred into different plant species. The synthesis of the transgenic fructan in plants which normally do not accumulate fructan have resulted in better insights into fructan biosynthesis in plants and offered the opportunity for experimental investigation of carbon partitioning (Gerrits et al. 2001). Studying the responses of these transgenic plants to environmental stresses such as drought and cold

Production of Heterologous Storage Polysaccharides in Potato Plants

■

401

may lead to greater insight into the physiological role of fructans in stress resistance (Pilon-Smits et al. 1995; Pilon-Smits et al. 1999). Furthermore, introduction of fructosyltransferases in agronomically important crops may improve the commercial availability of fructan and may facilitate the production of a range of structurally different fructan molecules (Ebskamp et al. 1994). The fructosyltransferase gene from Bacillus spp., SacB, has not only been expressed in fructan-accumulating crops such as chicory, but also in plants which do not normally accumulate fructans such as tobacco, potato and sugar beet. The expression of bacterial fructosyltransferase genes in transgenic plants resulted in the accumulation of high molecular weight fructans. Bacterial fructan accumulation in potato plants and tubers dramatically affected the growth and carbohydrate partitioning, and the sink organs (roots, tubers) were reduced in weight. The starch content in the transformed tubers was inversely correlated with the fructan level, whereas the sucrose, glucose, fructose and protein levels increased substantially, in parallel with fructan concentration. In the transformed plants the fructans were located along the cell rim instead of in the vacuole (PilonSmits et al. 1996). Van der Meer and coworkers have fused the SacB gene from Bacillus subtilis to the vacuolar targeting sequence of the yeast carboxypeptidase Y (cpy) gene, under the control of the constitutive CaMV 35S promoter, and introduced it into potato. High molecular mass (>5 × 106 Da) fructans were accumulated in potato transformants both in the leaves (30% of dry weight) and microtubers (7% of dry weight) (van der Meer et al. 1994). The levansucrase from Erwinia amylovora was expressed in tubers of transgenic potato plants, resulting in the accumulation of fructans in the apoplast and vacuole up to 12–19% of the tuber dry weight. The molecular weight and structure of the fructan produced in transgenic plants was identical to levan isolated from E. amylovora (Rober et al. 1996). Levansucrase from Bacillus subtilis was targeted to the plastids of tobacco (Nicotiana tabacum) and potato (Solanum tuberosum) plants. Introduction of this enzyme leads to high-level fructan accumulation in chloroplasts and amyloplasts, respectively. Moreover, the starch structure in amyloplasts was altered (Gerrits et al. 2001). 1-SST and 1-FFT from Jerusalem artichoke were expressed in maize and potato. Potato tubers expressing both 1-SST and 1-FFT can accumulate more inulin (2.6 mg inulin/g tuber) than the transformants expressing only the 1-SST alone (1.8 mg inulin/g tuber). Inulin production in potato was relatively stable throughout tuber development and little evidence of degradation was observed (Stoop et al. 2007). A number of development aberrations, such as stunting, leaf bleaching and reduction of the number and weight of tubers and starch content in transgenic potato, were associated with the expression of bacterial fructosyltransferase genes in plants (van der Meer et al. 1994; Rober et al. 1996).

402

■

Plant Polysaccharides

16.4 Final remarks During photosynthesis, plants are able to fix atmospheric carbon dioxide and incorporate it into carbohydrates. The photosynthate produced in the source tissues travels through the plant in the form of sucrose and it can either be used by the plant in the primary (and secondary) metabolism or it can be stored in sink organs. In higher plants, the major reserve polysaccharide is starch. During the last 15 years the scientific community has been occupied with the increasing industrial request for carbohydrates with novel properties. Several different strategies have been used to study the starch biosynthesis pathway in higher plants and to produce novel products with novel functional properties. One of these strategies was to modulate the starch biosynthetic pathway by knocking out, knocking down or overexpressing endogenous genes involved in this pathway. A second strategy involved the expression of heterologous genes encoding biosynthetic or modifying enzymes from other organisms, including bacterial, fungal and animal enzymes. Although the activity of several of these enzymes was initially identified in carbohydrates other than starch, including glycogen, other glucans or different fructans, most of these enzymes have show an effect on different aspects of starch such as composition, granule morphology, etc. From an industrial and environmental point of view, starch is a very interesting resource as it is renewable and biodegradable and can be modified to a diverse range of products. With our understanding of the starch biosynthesis pathway becoming more and more clear, the potential for producing novel starches through genetic modification of plants may be fully realized. The engineering of starch biosynthesis may lead to modification of starch granule shape, amylose to amylopectin ratio, amylopectin chain length, phosphorylation and lipid content. Starches with modified properties will be used more efficiently and for a wider range of applications in both food and non-food industries. The concept of the biobased economy, where plants are seen as a bio-factories for various components, has increasingly gained interest in recent years. An interesting concept would be the synthesis of novel and multiple reserve carbohydrates in plants, with interesting properties for different industrial applications. Besides the improvement of their properties, one big challenge in carbohydrate research will be the maximization of storage sugars. In most of the cases described in this chapter, the synthesis of multiple reserve carbohydrates in the amyloplast did not result in an increase in the total amount of reserve carbohydrates in the plant. The tagging of distinct polysaccharides to different cell compartments will address the question of the limited storage of carbohydrates in the amyloplast and has great potential for increased accumulation of storage sugars in plants.

Production of Heterologous Storage Polysaccharides in Potato Plants

■

403

References Banguela, A., Hernández, L. (2006) Fructans: from natural sources to transgenic plants. Biotecnologia Aplicada, 23, 202–210. Baird, J.K., Longyear, V.M.C., Ellwood, D.C. (1973) Water insoluble and soluble glucans produced by extracellular glycosyltransferases from Streptococcus mutans. Microbios, 8, 143–150. Ball, S., van de Wal, M.H.B.J., Visser, R.G.F. (1998) Progress in understanding the biosynthesis of amylose. Trends in Plant Science, 3, 462–467. Börnke, F., Hajirezaei, M., Sonnewald, U. (2001) Cloning and characterization of the gene cluster for palatinose metabolism from the phytopathogenic bacterium Erwinia rhapontici. Journal of Bacteriology, 183, 2425–2430. Börnke, F., Hajirezaei, M., Sonnewald, U. (2002) Potato tubers as bioreactors for palatinose production. Journal of Biotechnology, 96, 119–124. Calazans, G.M.T., Lima R.C., de Franca F.P., Lopes C.E. (2000) Molecular weight and antitumour activity of Zymomonas mobilis levans. International Journal of Biological Macromolecules, 27, 245–247. Chalmers, J., Johnson, X., Lidgett, A., Spangenberg, G. (2003) Isolation and characterisation of a sucrose:sucrose 1-fructosyl transferase gene from perennial ryegrass (Lolium perenne). Journal of Plant Physiology, 160,1385–1391. Chatterton, N.J., Harrison P.A. (1997) Fructan oligomers in Poa ampla. New Phytologist, 136, 3–10. Cote, G., Robyt, J.F. (1982) Isolation and partial characterization of an extracellular glucansucrase from Leuconostoc mesenteroides NRRL B-1355 that synthesizes an alterning (1–6), (1–3)-α-d-glucan. Carbohydrate Research, 101, 57–74. Cote, G.L. (1992) Low-viscosity α-d-glucan fractions derived from sucrose which are resistant to enzymatic digestion. Carbohydrate Polymers, 19, 249–252. Cote, G.L., Ahlgren, J. (1993) Metabolism in microorganisms, Part I. Levan and levansucrase. In: Science and Technology of Fructans (eds M. Suzuki, N.J. Chatterton), pp 141–168. CRC Press, Boca Raton, FL. Cote, G.L., Leathers, T.D., Ahlgren, J.A., Wyckoff, H.A., Hayman, G.T., Biely, P. (1997) Alternan and highly branched limit dextrans: Low-viscosity polysaccharides as potential new food ingredients. In: Chemistry of Novel Foods (eds A.M. Spanier, M. Tamura, H. Okai, O. Mills), pp. 95–110. Allured Publishing, Carol Stream, IL. Dedonder, R. (1966) Levansucrase from Bacillus subtilis. Methods in Enzymology, 8, 500–505. Duchateau, N., Bortlik, K., Simmen, U., Wiemken, A., Bancal, P. (1995) Sucrose:fructan 6-fructosyltransferase, a key enzyme for diverting carbon from sucrose to fructan in barley leaves. Plant Physiology, 107, 1249–1255. Ebisu, S., Kato, K., Kotani, S., Misaki, A. (1975) Structural differences in fructans elaborated by Streptococcus mutans and Strep. salivarius. Journal of Biochemistry, 78, 879–887. Ebskamp, M.J.M., van der Meer, I.M., Spronk, B.A., Weisbeek, P.J., Smeekens, S.C.M. (1994) Accumulation of fructose polymers in transgenic tobacco. Nature Biotechnology, 12, 272–275. Edelman, J., Jefford, T.G. (1968) The mechanism of fructosan metabolism in plants as exemplified in Helianthus tuberosis. New Phytologist, 67, 517–531. Fuchs, A. (1991) Current and potential food and non-food applications of fructans. Biochemical Society Transactions, 19, 555–560.

404

■

Plant Polysaccharides

Gerrits, N., Turk, S.C.H.J., van Dun, K.P.M., et al. (2001) Sucrose metabolism in plastids. Plant Physiology, 125, 926–934. Goddijn, O.J.M., Verwoerd, T.C., Voogd, E., Krutwagen, R.W.H.H., De Graaf, P.T.H.M., Poels, J., et al. (1997) Inhibition of trehalase activity enhances trehalose accumulation in transgenic plants. Plant Physiology, 113, 181–190. Groenwall, A.J., Ingelman, B.G.A. (1948) Manufacture of infusion and injection fluids. US Patent 2, 437, 518. Gross, M., Geier, G., Rudolph, K., Geider, K. (1992) Levan and levansucrase synthesized by the fireblight pathogen Erwinia amylovora. Physiological and Molecular Plant Pathology, 40, 371–381. Guggenheim, B. (1970) Enzymatic hydrolysis and structure of water-insoluble glucan produced by glucosyltransferases from a strain of streptococcus mutans. Helvetica Odontologica Acta, 14(Suppl 5), 89. Hamada, S., Slade, H.D. (1980) Biology, immunology, and cariogenicity of Streptococcus mutans. Microbiological Reviews, 44, 331–384. Heinze, T., Liebert, T., Heublein, B., Hornig, S. (2006) Functional polymers based on dextran. Advances in Polymer Science, 205, 199–291. Hendry, G.A.F., Wallace, R.K. (1993) The origin, distribution, and evolutionary significance of fructans. In: Science and Technology of Fructans (eds M. Suzuki, N.J. Chatterton), pp 119–139, CRC Press, Boca Raton, FL. Hennink, W.E., van Nostrum, C.F. (2002) Novel cross linking methods to design hydrogels. Advanced Drug Delivery Reviews, 54, 13–36. Henrissat, B., Davies, G. (1997) Structural and sequence based classification of glycoside hydrolases. Current Opinions in Structural Biology, 7, 637–644. Hovenkamp-Hermelink, J.H.M., Jacobsen, E., Ponstein, A.S., et al. (1987) Isolation of an amylose-free starch mutant of the potato (Solanum tuberosum L.). Theoretical and Applied Genetics, 75(1), 217–221. Jeanes, A., Haynes, W.C., Williams, C.A., et al. (1954) Characterization and classification of dextrans from ninety six strains of bacteria. Journal of the American Chemical Society, 76, 5041–5052. Jobling, S.A., Schwall, G.P., Westcott, R.J., et al. (1999) A minor form of starch branching enzyme in potato (Solanum tuberosum L.) tubers has a major effect on starch structure: cloning and characterisation of multiple forms of SBE A. Plant Journal, 18, 163–171. Jobling, S.A., Westcott, R.J., Tayal, A., Jeffcoat, R., Schwall, G.P. (2002) Production of a freeze-thaw-stable potato starch by antisense inhibition of three starch synthase genes. Nature Biotechnology, 20, 295–299. Jobling, S.A., Jarman, C., Teh, M.M., Holmberg, N., Blake, C., Verhoeyen, M.E. (2003) Immunomodulation of enzyme function in plants by single-domain antibody fragments. Nature Biotechnology, 21, 77–80. Kammerer, B., Fischer, K., Hilpert, B., et al. (1998) Molecular characterization of a carbon transporter in plastids from heterotrophic tissues: the glucose 6-phosphate/ phosphate antiporter. Plant Cell, 10, 105–118. Ko, W., Wong, C., Yeung, H., Yung, M., Shaw, P., Tam, S. (1991) Increasing the plasma half-life of trichosanthin by coupling to dextran. Biochemical Pharmacology, 42, 1721–1728. Kok-Jacon, G.A., Ji, Q., Vincken, J-P., Visser, R.G.F. (2003) Towards a more versatile alpha-glucan biosynthesis in plants. Journal of Plant Physiology, 160, 765–777.

Production of Heterologous Storage Polysaccharides in Potato Plants

■

405

Kok-Jacon, G.A., Vincken, J-P., Suurs, L.C.J.M., Wang, D., Liu, S., Visser, R.G.F. (2005a) Production of dextran in transgenic potato plants. Transgenic Research, 14, 385–395. Kok-Jacon, G.A., Vincken J-P., Suurs, L.C.J.M., Visser, R.G.F. (2005b) Mutan produced in potato amyloplasts adheres to starch granules. Plant Biotechnology Journal, 3, 341–351. Kok-Jacon, G.A., Vincken J-P., Suurs, L.C.J.M., Wang, D., Liu, S., Visser, R.G.F. (2007) Expression of alternansucrase in potato plants. Biotechnology Letters, 29, 1135–1142. Kortstee, A.J., Vermeesch, A.M.S., De Vries, B.J., Jacobsen, E., Visser, R.G.F. (1996) Expression of Escherichia coli branching enzyme in tubers of amylose-free transgenic potato leads to an increased branching degree of the amylopectin. Plant Journal, 10, 83–90. Kralj, S.G.H., Van Geel-Schutten, M.M.G. Dondorff, S. Kirsanovs, M.J.E., Van Der Maarel C., Dijkhuizen, L. (2004) Glucan synthesis in the genus Lactobacillus: isolation and characterization of glucansucrase genes, enzymes and glucan products from six different strains. Microbiology, 150, 3681–3690. Krohn, B.M., Stark, D.M., Barry, G.F., Preiss, J., Kishore, G.M. (1994) Modification of starch structure in transgenic potato. Plant Physiology, 105(suppl), 37. Kubik, C., Sikora, B., Bielecki, S. (2004) Immobilization of dextransucrase and its use with soluble dextranase for glucooligosaccharides synthesis. Enzyme and Microbial Technology, 34, 555–560. Livingston, D.P.I., Chatterton N.J., Harrison P.A. (1993) Structure and quantity of fructan oligomers in oat (Avena spp.). New Phytologist, 123, 725–734. Loesche, W.J. (1986) Role of Streptococcus mutans in human dental decay. Microbiological Reviews, 50, 353–380. Lüscher, M., Hochstrasser, U., Vogel, G., et al. (2000) Cloning and functional analysis of sucrose:sucrose 1-fructosyltransferase from tall fescue. Plant Physiology, 124, 1217–1227. Mehvar, R. (2000) Dextrans for targeted and sustained delivery of therapeutic and imaging agents. Journal of Controlled Release, 69, 1–25. Monchois, V., Willemot, R.M., Remaud-Simeon, M., Croux, C., Monsan, P.F. (1996) Cloning and sequencing of a gene coding for a novel dextransucrase from Leuconostoc mesenteroides NRRL B-1299 synthesizing only α (1–6) and α (1–3) linkages. Gene, 182, 23–32. Monchois, V., Willemot, R.M., Monsan, P. (1999) Glucansucrases: Mechanism of action and structure–function relationships. FEMS Microbiology Reviews, 23, 131–151. Mooser, G. (1992) Glycosidases and glycosyltransferases. Enzymes, 20, 187–221. Myers, A.M., Morell, M.K., James, M.G., Ball, S.G. (2000) Recent progress towards understanding biosynthesis of the amylopectin crystal. Plant Physiology, 122, 989–997. Nakamura, Y. (2002) Towards a better understanding of the metabolic system for amylopectin biosynthesis in plants: Rice endosperm as a model tissue. Plant and Cell Physiology, 43, 718–725. Nazarian Firouzabadi, F., Vincken, J-P., Ji, Q., Suurs, L.C.J.M., Visser, R.G.F. (2007a) Expression of an engineered granule-bound E. coli maltose acetyl transferase in wild-type and amf potato plants. Plant Biotechnology Journal, 5, 134–145.

406

■

Plant Polysaccharides

Nazarian Firouzabadi, F., Kok-Jacon, G.A., Vincken, J-P., Ji, Q., Suurs, L.C.J.M., Visser, R.G.F. (2007b) Fusion proteins comprising the catalytic domain of mutansucrase and a starch-binding domain can alter the morphology of amylose-free potato starch granules during biosynthesis. Transgenic Research, 16, 645–656. Nelson, D.L., Cox, M.M. (2004) Carbohydrates and glycobiology. Chapter 7 in: Lehninger ’s Principles of Biochemistry (ed. W.H. Freeman), 4th edition, pp 238–272. New York. Oakes, J.V., Shewmaker, C.K., Stalker, D.M. (1991) Production of cyclodextrins, a novel carbohydrate, in the tubers of transgenic potato plants. Nature Biotechnology, 9, 982–986. Olivares-Illana, V., Wacher-Rodarte, C., Le Borgne, S., Lopez-Munguıa, A. (2002) Characterization of a cell-associated inulosucrase from a novel source: a Leuconostoc citreum strain isolated from Pozol, a fermented corn beverage from Mayan origin. Journal of Industrial Microbiology and Biotechnology, 28, 112–117. Pilon-Smits, E.A.H., Ebskamp, M.J.M., Paul, M.J., Jeuken, M.J.W., Weisbeek, P.J., Smeekens, S.C.M. (1995) Improved performance of transgenic fructan-accumulating tobacco under drought stress. Plant Physiology, 107, 125–130. Pilon-Smits, E.A.H., Ebskamp, M.J.M., Jeuken, M.J.W., et al. (1996) Microbial fructan production in transgenic potato plants and tubers. Industrial Crops and Products, 5, 35–46. Pilon-Smits, E.A.H., Terry, N., Sears, T., van Dun, K. (1999) Enhanced drought resistance in fructanproducing sugar beet. Plant Physiology and Biochemistry, 37, 313–317. Pollock, C.J., Cairns, A.J. (1991) Fructan metabolism in grasses and cereals. Annual Review of Plant Physiology and Plant Molecular Biology, 42, 77–101. Ramesh, H.P., Tharanathan, R.N. (2003) Carbohydrates – the renewable raw materials of high biotechnological value. Critical Reviews in Biotechnology, 23, 149–173. Rindlav-Westling, A., Stading, M., Hermansson, A.-M., Gatenholm, P. (1998) Structure, mechanical and barrier properties of amylose and amylopectin films. Carbohydrate Polymers, 36, 217–224. Ritsema, T., Smeekens, S. (2003) Fructans: beneficial for plants and humans. Current Opinion in Plant Biology, 6, 223–230. Rober, M., Geider, K., Muller-Roeber B., Willmitzer, L. (1996) Synthesis of fructans in tubers of transgenic starch-deficient potato plants does not result in an increased allocation of carbohydrates. Planta, 199, 528–536. Schwall, G.P., Safford, R., Westcott, R.J., Jeffcoat, R., Tayal, A., Shi, Y.C., et al. (2000) Production of very-high-amylose potato starch by inhibition of SBE A and B. Nature Biotechnology, 18, 551–554. Seymour, F.R., Slodki, M.E., Plattner, R.D., Jeanes, A. (1977) Six unusual dextrans: ethylation structural analysis by combined GLCMS of per-O-acetyl aldononitriles. Carbohydrate Research, 53, 153–166. Seymour, F.R., Knapp, R.D., Bishop, S.H., Jeanes, A. (1979) Structural analysis of Leuconostoc dextrans containing 3-O-α-Dglucosylated α-d-glucancosyl residues in both linear-chain and branch point positions, or only in branch-point positions, by methylation and by 13C-N.M.R. spectroscopy. Carbohydrate Research, 74, 41–62. Seymour, F.R., Knapp, R.D. (1980) Unusual dextrans. 13. Structural analysis of dextrans from strains of Leuconostoc and related genera, that contain 3-O-a-dglucancosylated a-d-glucancopyranosyl residues at the branch points, or in consecutive linear position. Carbohydrate Research, 81, 105–129.

Production of Heterologous Storage Polysaccharides in Potato Plants

■

407

Shewmaker, C.K., Boyer, C.D., Wiesenborn, D.P., et al. (1994) Expression of Escherichia coli glycogen synthase in the tubers of transgenic potatoes (Solanum tuberosum) results in a highly branched starch. Plant Physiology, 104, 1159–1166. Sidebotham, R. (1974) Dextrans. Advances in Carbohydrate Chemistry and Biochemistry, 30, 371–444. Simpson, C.L., Cheetham, N.W.H., Giffard, P.M., Jacques, N.A. (1995) Four glucosyltransferases, GtfJ, GtfK, GtfL and GtfM, from Streptococcus salivarius ATCC 25975. Microbiology, 141, 1451–1460. Sims I.M., Pollock C.J., Horgan R. (1992) Structural analysis of oligomeric fructans from excised leaves of Lolium temulentum. Phytochemistry, 31, 2989–2992. Sinha, V.R., Kumria, R. (2001) Colonic drug delivery: prodrug approach. Pharmaceutical Research, 18, 57–564. Sprenger, N., Bortlik, K., Brandt, A., Boller, T., Wiemken, A. (1995) Purification, cloning, and functional expression of sucrose:fructan 6-fructosyltransferase, a key enzyme of fructan synthesis in barley. Proceedings of the National Academy of Sciences of the U S A, 92, 11652–11656. Stark, D.M., Timmerman, K.P., Barry, G.F., Preiss, J., Kishore, G.M. (1992) Regulation of the amount of starch in plant tissues by ADP glucose pyrophosphorylase. Science, 258, 287–292. Stoop, J.M., Van Arker, J., Hakkert, J.C., Tyree, C., Caimi, P.G., Koops, A.J. (2007) Developmental modulation of inulin accumulation in storage organs of transgenic maize and transgenic potato. Plant Science, 173, 172–181. Sutherland, I.W. (1972) Bacterial exopolycaccharides. Advances in Microbial Physiology, 8, 143–212. Tuohy, K.M., Probert, H.M., Semejkal, C.W., Gibson, G.R. (2003) Using probiotics and prebiotics to improve gut health. Drug Discovery Today, 8, 692–700. van der Meer, I.M., Ebskamp, M.J.M., Visser, R.G.F., Weisbeek, P.J., Smeekens, S.C.M. (1994) Fructan as a new carbohydrate sink in transgenic potato plants. Plant Cell, 6, 561–570. van der Meer, I.M., Koops, A.J., Hakkert, J.C., Van Tunen, A.J. (1998) Cloning of the fructan biosynthesis pathway of Jerusalem artichoke. Plant Journal, 15, 489–500. van Hijum, S.A.F.T., van Geel-Schutten, G.H., Rahaoui, H., van der Maarel, M.J., Dijkhuizen, L. (2002) Characterization of a novel fructosyltransferase from Lactobacillus reuteri that synthesizes highmolecular- weight inulin and inulin oligosaccharides. Applied and Environmental Microbiology, 68, 4390–4398. van Hijum, S.A.F.T., Szalowska, E., van der Maarel, M.J., Dijkhuizen, L. (2004) Biochemical and molecular characterization of a levansucrase from Lactobacillus reuteri. Microbiology, 150, 621–630. van Hijum, S.A.F.T., Kralj, S., Ozimek, L.K., Dijkhuizen, L., Van Geel-Schutten, I.G.H. (2006) Structure-Function relationships of glucansucrase and fructansucrase enzymes from lactic acid bacteria. Microbiology and Molecular Biology Reviews, 70, 157–176. van Soest, J.J.G., Borger, D.B. (1997) Structure and properties of compression-molded thermoplastic starch materials from normal and high-amylose maize starches, Journal of Applied Polymer Science, 4, 631–644. Vijn, I., van Dijken, A., Sprenger, N., et al. (1997) Fructan of the inulin neoseries is synthesized in transgenic chicory plants (Cichorium intybus L.) harbouring onion (Allium cepa L.) fructan:fructan 6-G-fructosyltransferase. Plant Journal, 11, 387–398.

408

■

Plant Polysaccharides

Vijn, I., Van Dijken, A., Lüscher, M., et al. (1998) Cloning of sucrose:sucrose 1-fructosyltransferases from onion and synthesis of structurally defined fructan molecules from sucrose. Plant Physiology, 117, 1507–1513. Visser, R.G.F., Somhorst, I., Kuipers, G.J., Ruys, N.J., Feenstra, W.J., Jacobsen, E. (1991) Inhibition of expression of the gene for granule bound starch synthase in potato by antisense constructs. Molecular and General Genetics, 225, 289–296. Visser, R.G.F., Suurs, L.C.J.M., Bruinenberg, P.M., Jacobsen, E. (1997) Comparison between amylose-free and amylose containing potato starches. Starch/Staerke, 49, 438–443. Waterhouse, A.L., Chatterton, N.J. (1993) Glossary of fructan terms. In: Science and Technology of Fructans (eds M. Suzuki, N.J. Chatterton), pp 1–7. CRC Press, Boca Raton, FL. Wei, J.Z., Chatterton, N.J., Harrison, P.A., Wang, R.R., Larson, S.R. (2002) Characterization of fructan biosynthesis in big bluegrass (Poa secunda). Journal of Plant Physiology, 159, 705–715. Welman, A.D., Maddox, I.S. (2003) Exopolysaccharides from lactic acid bacteria: perspectives and challenges. Trends in Biotechnology, 21, 269–274.

Annual Plant Reviews (2011) 41, 409–424 doi: 10.1002/9781444391015.ch17

Chapter 17

GLYCAN ENGINEERING IN TRANSGENIC PLANTS Muriel Bardor1, José A. Cremata2 and Patrice Lerouge1 1

UPRES-EA 4358, IFRMP 23, Université de Rouen, 76821 Mont Saint Aignan Cédex, France 2 Department of Carbohydrate Chemistry, CIGB, Havana, Cuba Manuscript received October 2008

Abstract: N-glycosylation is a major post-translational modification step in the biosynthesis of proteins in eukaryotes. This process consists of two main steps, the early N-glycan processing occurring in the endoplasmic reticulum and its maturation in the Golgi apparatus. The early N-glycan processing steps are conserved among eukaryotes and are involved in the quality control of proteins. In contrast, the maturation steps in the Golgi apparatus give rise to a large variety of organism-specific complex structures. This divergence in the resulting Nglycan structures is a key issue when the plant is used as a cell factory for the production of human therapeutic proteins. In this chapter, we describe the main aspects of the N-glycan biosynthesis in plants as well as strategies that have been developed to engineer this N-glycosylation pathway in order to allow the production in transgenic plants of pharmaceutical proteins that carry human-like glycans. Keywords: immunogenicity; N-glycosylation; plant engineering; recombinant proteins; sialylation

17.1 Introduction Plants provide a low-cost and contamination-safe factory for the production of recombinant proteins. Plant systems also benefit from easy scale-up and established practices for efficient harvesting, transporting, storing and processing. Many therapeutic proteins, such as growth factors, antibodies, cytokines, enzymes and antigens produced for vaccination have been Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

409

410

■

Plant Polysaccharides

manufactured in leaves, tubers or seeds. Plant cells have the machinery to perform most of the post-translational modifications, including Nglycosylation, that are required by recombinant proteins to achieve adequate bioactivity and pharmacokinetics. Many studies have demonstrated that mammalian glycoproteins are efficiently glycosylated when expressed in plants (Fischer et al. 2004; Ma et al. 2005). However, the processing of plant N-glycans in the Golgi apparatus displays major differences from that of mammalian cells (Lerouge et al. 1998; Gomord et al. 2004). This inability to process authentic human N-glycan structures imposes a major limitation on the use of plants as expression system for therapeutic products. A number of strategies have emerged to engineer plant N-glycans into humancompatible molecules, mainly by knock-out strategies to remove immunogenic non-human glycoepitopes or knock-in methods to introduce the missing glycan sequences on N-glycans that are required for full biological function of the plant-derived therapeutic protein.

17.2 N-glycosylation: a major post-translational modification of secreted proteins N-glycosylation is a major post-translational modification step in the synthesis of proteins in eukaryotes. N-glycan processing in the secretory pathway is essential for proteins intended to be secreted or integrated into membranes. N-glycosylation starts when the protein is translated and translocated from the ribosome into the lumen of the endoplasmic reticulum (ER). In this processing, a dolicholphosphate oligosaccharide precursor (Glc3Man9GlcNAc2-PP-dolichol) is initially assembled at the cytoplasmic face and finished in the luminal face of the ER membrane (Burda & Aebi 1999). This precursor is used by the oligosaccharyltransferase (OST) multisubunit complex that catalyses its transfer on to the asparagine residues of the consensus sequences Asn-X-Ser/Thr, when X is other than Pro, of a target protein (Burda & Aebi 1999). The precursor is then deglucosylated/ reglucosylated to ensure the quality control of the neosynthesized protein through the interaction with ER-resident chaperones calreticulin and calnexin. These ER events are crucial for proper folding and oligomerization of secreted proteins (Helenius & Aebi 2001); they are highly conserved in eukaryotes investigated so far and involved a limited set of high-mannosetype N-glycans. In contrast, evolutionary adaptation of N-glycan processing in the Golgi apparatus has given rise to a large variety of organism-specific complex structures. Mannosidases located in this compartment first degrade the oligosaccharide precursor into oligomannose-type N-glycans ranging from Man9GlcNAc2 to Man5GlcNAc2. N-acetylglucosaminyltransferase I (GnT I) then transfers a first GlcNAc residue on the α(1,3)-mannose arm of Man5GlcNAc2 and opens the door to the synthesis of multiple structurally different complex-type N-glycans. Then, the action of α-mannosidase II and

Glycan Engineering in Transgenic Plants

■

411

Manα1-2Manα1 6 Manα α1 3 6 Manα1-2Manα1 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3

Precursor

Glcα1-2Glcα1-3Glcα1-3Manα1- 2Manα1-2Manα1 Manα1-2Manα1

Glucosidase I and II 6 Manα1 3 6 Manα1-2 Manα1 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3

Man9GlcNAc2

Manα1- 2Manα1-2Manα1

Manα1

Mannosidases

6 Manα1 3 6 Manα1 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 Manα1

Man5GlcNAc2

Manα1

GnT I

6 Manα1 3 6 Manα1 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 GlcNAcβ1-2Manα1

α-mannosidase II GnT II GlcNAcβ1-2Manα1 6 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 GlcNAcβ1-2Manα1

β(1,2)-XylT α(1,3)-FucT

GlcNAc2Man3GlcNAc2

β(1,4)-GalT α(1,6)-FucT Neu5AcT

GlcNAcβ1-2Manα1 6 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 2 3 GlcNAcβ1-2Manα1

Fucα1 Neu5Ac-Galβ1-4 β GlcNAcβ1-2Manα1 β

Xylβ1 6 6 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3

Fucα1

β(1,3)-GalT α(1,4)-FucT

Fucα1

Neu5Ac-Galβ1-4 GlcNAcβ1-2Manα1 4 Galβ1-3GlcNAcβ1-2Manα1

6 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 2 3 Galβ1-3GlcNAcβ1-2Manα1 4 Xylβ1

Multi-sialylated, multi-antenary mammalian N-glycans

Fucα1

Fucα1

Plant N-glycans

Lewisa

Figure 17.1 N-glycan processing in mammals and plants. GnT I and GnT II: Nacetylglucosaminyltransferase I and II. β(1,2)-XylT, β(1,2)-xylosyltransferase; α(1,3/4/6)FucT, α(1,3/4/6)fucosyltransferase; β(1,3/4)-GalT, β(1,3/4)-galactosyltransferase; Neu5AcT; N-acetylneuraminic acid transferase.

412

■

Plant Polysaccharides

GnT II allow the synthesis of the core GlcNAc2Man3GlcNAc2, the last common glycan intermediate in most eukaryotes (Fig. 17.1). The complex oligosaccharides arise from the transfer in the Golgi apparatus of monosaccharide residues on to the core GlcNAc2Man3GlcNAc2 under the action of organism-specific glycosyltransferases. As a consequence, mature proteins leaving the secretory pathway harbour multiple organism-specific complex N-glycans allowing the protein to acquire a set of glycan-mediated biological functions. As an illustration, N-glycans in mammals are maturated into polyantennary, polysialylated structures harbouring an α1,6-linked fucose residue on the proximal N-acetylglucosamine of the core (core-α(1,6)-fucose). In plants, the core is substituted by a xylose residue β(1,2)-linked to the mannose β(1,4)-linked to the inner chitobiose disaccharide (core-xylose) and by an α(1,3)-linked fucose residue (core-α(1,3)-fucose)) due to the action of the xylosyltransferase and α(1,3)-fucosyltransferase in the Golgi apparatus (Fitchette-Lainé et al. 1994). The substrate specificity for these glycosyltransferases requires at least the presence of one terminal GlcNAc already linked to the trimannosyl core. These antennas are elongated by Galβ(1–3)GlcNAc disaccharides that are frequently α(1,4)-fucosylated at the GlcNAc residue, forming the Lewisa epitope (Lerouge et al. 1998). α(1,4)-fucosylation is the last event in N-glycan maturation of most extracellular plant glycoproteins because the α(1,4)-fucosyltransferase (α(1,4)-FucT) requires Galβ(1–3) GlcNAc disaccharides as substrate (Fitchette-Lainé et al. 1997) (Fig. 17.1). 17.2.1 N-Glycosylation of plant-derived pharmaceuticals: antibodies as a glycoprotein model The expression of plant-derived antibodies, such as plant-derived IgGs, has received considerable attention since they exhibit a fast-growing interest in human therapy. For this reason, the study of the N-glycosylation of therapeutic proteins and its engineering has been mainly carried out on these molecules. In mammals, antibodies are usually N-glycosylated by biantennary structures on a conserved site (Asn297) located in the Fc domain. These complex N-glycans bear both a core-α(1,6)-fucose and non-reducing terminal β(1,4)-galactose residues and are poorly sialylated (Fig. 17.2) (Rudd et al. 1991; Raju et al. 2000). Cabanes-Macheteau et al. (1999) demonstrated that when expressed in plants, antibodies are also transported through the entire Golgi apparatus where the N-glycans are processed into a heterogeneous mixture of complex structures carrying both core-xylose and core-α(1,3)fucose residues (Fig. 17.2). Similar results were obtained for antibodies produced in various other plant systems (Zeitlin et al. 1998; Bardor et al. 2003a; Ramirez et al. 2003; Ko et al. 2003; Sriraman et al. 2004). Consistent with the well-established principle of cell- and tissue-specific glycosylation, studies on these so-called ‘plantibodies’ expressed in various plant systems have indicated that the N-glycan heterogeneity on these plant-derived pharmaceuticals is dependent on the plant expression system itself. Interestingly, the ratio between complex and high-mannose type N-glycans is significantly

Glycan Engineering in Transgenic Plants

IgG from hybridoma (Galβ1-4)GlcNAcβ1- 2Manα α1

413

Plant-derived IgG Fucα1

6 6 Manβ1- 4GlcNAcβ1- 4GlcNAc 3 (Galβ1-4)GlcNAcβ1- 2Manα1

■

(GlcNAcβ1)-2 Manα1 6 Manβ1- 4GlcNAcβ1- 4GlcNAc 3 2 3 (GlcNAcβ1)-2Manα1 Xylβ1

Fucα1

Figure 17.2 Chemical structures of glycans N-linked to IgG isolated from hybridoma cells or produced in transgenic plants.

decreased in older base leaves as compared to upper leaves, which reflects the influence of plant developmental stage in the N-glycosylation pattern of plantibodies (Elbers et al. 2001). Thus, the N-glycan population that decorates monoclonal antibodies expressed in plants is different from that produced in mice or CHO cells cultured in bioreactors, in which the most abundant species correspond mainly to the agalacto GlcNAc2Man3(Fuc) GlcNAc2 (Cabrera et al. 2004). Heavy and light chains of plant-derived antibodies were demonstrated to be correctly assembled into a functional complex (Hiatt et al. 1989; Bardor et al. 2003a). Taken together, these data demonstrated that the expression of sophisticated mammalian glycoproteins in plants, such as IgGs, results in the proper folding, oligomerization and glycosylation of the recombinant molecules. The structural differences observed between mammals and plant-derived molecules are restricted to N-linked glycan decorations and reflect the organism specificity of Golgi glycosyltransferases involved in late N-glycan maturation (Fig. 17.2). 17.2.2 Differences in N-glycan structure may compromise in vivo use of plant-derived therapeutic proteins The difference between the repertoires of plant and mammalian Golgi glycosidases and glycosyltransferases gives rise to organism-specific complex N-glycans that may strictly limit the use of plant-derived proteins as therapeutics. The core-xylose and core-α(1,3)-fucose epitopes (Fig. 17.1) are absent in mammalian cells. Immunization of goats (Kurosaka et al. 1991), rabbits

414

■

Plant Polysaccharides

(Faye et al. 1993) and rats (McManus et al. 1988) with plant glycoproteins was demonstrated to elicit the production of core-xylose and/or core-α(1,3)fucose-specific antibodies. Furthermore, it was demonstrated that such glycoepitopes are also able to elicit an immune response in humans (Bardor et al. 2003b). These epitopes can be also recognized by allergen-reactive IgE (Aalberse et al. 1981; Faye & Chrispeels 1988; Garcia-Casado et al. 1996; Wilson & Altmann 1998; van Ree et al. 2000). As stated above, it is well documented that the precursor oligosaccharide attached to the protein ensures the quality control of the neosynthesized protein through the interaction with ER-resident chaperones. In the case of IgG, N-glycans located on the Fc part of the molecule are required for optimal binding of antibodies to all classes of receptor and for their effector functions, including C1q binding from the complement cascade. As a consequence, the immune responses directed against plant-specific glycans cannot be overcome by using non-glycosylated plant-derived proteins. Deglycosylated antibodies lack effector functions and their half-life and serum stability are dramatically reduced (Nose & Wigzell 1983). The absence of mammalian-type glycan sequences on plant-derived therapeutic proteins may also be deleterious for in vivo efficiency of the recombinant molecule. For instance, altered glycosylation on the conserved N-glycosylation site of IgG heavy chain might alter effector functions by provoking distortion of protein conformation on the Cγ2 domain because a large number of contacts between Fc polypeptide and the core N-glycan exists. Presence of truncated oligosaccharides, such as N-glycans missing terminal β(1,4)-galactose residues, on antibodies led to a reduce affinity for the Fcγ receptors and C1q of the complement cascade and a deficiency in antibody-dependent cell-mediated cytotoxicity (ADCC) (Mimura et al. 2000; Jefferis 2005). Additionally, the lack of sialic acid synthesis in plants is also critical. Most therapeutic proteins carry terminal sialic acids on their N-glycans. The main exception is IgG, in which carbohydrate moieties are sequestered within the two CH2 peptide domains and thus are not fully matured into sialylated N-glycans within the Golgi apparatus. This terminal sialylation is required for numerous biological functions, the first one being the control of the halflife of the protein in the circulatory system. In the absence of terminal sialic acids, glycoproteins are detected by hepatic asialoglycoprotein receptors and cleared from the serum, rendering these proteins biologically shortlived and ineffective (Kelm & Schauer 1997). As a consequence, nonsialylated plant-made proteins might be rapidly eliminated from the bloodstream when injected into humans. As an illustration, tobacco-derived erythropoietin was demonstrated to be biologically active in vitro but not functional in vivo because of its removal from the circulation before reaching erythropoietic tissues (Matsumoto et al. 1995). To overcome potential drawbacks due to N-glycan structures, in planta remodelling strategies have recently emerged that enable the production of

Glycan Engineering in Transgenic Plants

■

415

plant-derived molecules with human-compatible carbohydrate profiles. These include retaining the recombinant glycoprotein in the ER and knockout or knock-in strategies to remove non-human epitopes or to introduce human missing sequences. This is discussed in detail in the following section and illustrated by data regarding the N-glycan engineering of plant-derived antibodies.

17.3

Strategies for glycan engineering in transgenic plants

17.3.1 Retention in the ER Plant-derived antibodies are able to efficiently bind to their target antigens but carry N-linked glycans having non-mammalian core-xylose and coreα(1,3)-fucose residues (Zeitlin et al. 1998; Cabanes-Macheteau et al. 1999; Bardor et al. 2003a; Ko et al. 2003; Ramirez et al. 2003; Sriraman et al. 2004) (Fig. 17.2). For immunotherapy requiring large quantities of repeatedly administered antibodies, the immunogenicity of these epitopes is of concern. Since these modifications occur in the late Golgi compartment, retention of the antibody in the ER (Fig. 17.1) is one of the strategies used so far. IgG heavy chains and/or light chains were fused to a C-terminal KDEL sequence to allow the retrieval between ER and the Golgi apparatus. The analysis of the N-glycosylation demonstrated that the resulting plant-derived antibodies mainly contain oligomannose-type N-glycans ranging from Man9GlcNAc2 to Man6GlcNAc2, thus indicating that the IgG is efficiently retained in the ER and, as a consequence, is largely not processed by Golgi transferases (Ko et al. 2003; Sriraman et al. 2004; Tekoah et al. 2004; Triguero et al. 2005; Petrucelli et al. 2006). This strategy appears to be an alternative way to produce antibodies devoid of plant immunogenic glycoepitopes. However, in vivo clearance of antibodies with oligomannose-type N-glycans is increased by binding to the macrophage mannose receptor in liver (Wright & Morrison 1998; Ko et al. 2003). 17.3.2 In planta reconstruction of human-like N-glycans by knock-in strategies The first attempts to rebuild mammalian-like N-glycans in transgenic plants by a knock-in strategy, by expressing human β(1,4)-galactosyltransferase in tobacco cells, were reported by Palacpac et al. (1999). β(1,4)-galactose is absent on the terminal GlcNAc of the plant core N-glycans and, as mentioned above, this carbohydrate motif is important for some protein bioactivity. For instance, the presence of a Galβ(1,4)-GlcNAc antennae on glycan N-linked to IgG is required to reach its optimal effector function (Mimura et al. 2000). Furthermore, if the final goal of the in planta remodelling strategies is the synthesis of mammalian-like sialylated lactosamine sequences, the presence of terminal β(1,4)-galactose is a prerequisite for the further

416

■

Plant Polysaccharides

transfer of sialic acids by expression of a sialyltransferase (Fig. 17.1). Transformation of plants with a human β(1,4)-galactosyltransferase allow the plant to acquire unusual N-glycans with terminal β(1,4)-Gal residues (Bakker et al. 2001). These results indicate that the human enzyme is fully functional and localizes correctly in the Golgi apparatus where it participates with endogenous transferases in the maturation of glycans N-linked to secreted proteins. Furthermore, the crossing of a tobacco plant expressing human β(1,4)-galactosyltransferase with a plant expressing the heavy and light chains of a mouse antibody resulted in the expression of a plantderived antibody that exhibits about 30% of N-glycans carrying β(1,4)-Gal extensions as observed in mammals (Bakker et al. 2001) (Fig. 17.3). Plant β(1,2)-xylosyltransferase and α(1,3)-fucosyltransferases involved in the synthesis of core-xylose and core-α(1,3)-fucose epitopes require at least one GlcNAc residue linked to Man3GlcNAc2 as substrate (Johnson & Chrispeels 1987; Zeng et al. 1997; Leiter et al. 1999) (Fig. 17.1, Fig. 17.3). Any substitution of the terminal GlcNAc units inhibits both transferases. Manα1 6 Manα α1 3 6 Manα1 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 Manα1

Manα1

GnT I

6 Manα1 3 6 Manα1 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3

(Manα1)

GlcNAcMan5GlcNAc2

GlcNAcβ1-2Manα1 GlcNAcβ1 2Manα1

XylT/β(1 4) GalT XylT/β(1,4)-GalT

α−mannosidase II Manα1

Galβ1-4GlcNAcβ1-2Manα1

6 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 GlcNAcβ1-2Manα1

6 Manα11 M 3 6 (Manα1) Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3

GlcNAcMan3GlcNAc2 GnT II

GlcNAcβ1-2Manα1 6 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3

GlcNAc2Man3GlcNAc2

GlcNAcβ1-2Manα1

β(1,2)-XylT α(1,3)-FucT

GlcNAcβ1-2Manα1

Galβ1-4GlcNAcβ1-2Manα1 6 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 2 3

β(1,4)-GalT Galβ1 4GlcNAcβ1 2Manα1 Galβ1-4GlcNAcβ1-2Manα1

6 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 2 3

Xylβ1

Fucα1

GlcNAcβ1-2Manα1 Xylβ1

Fucα1

GnT III

GlcNAcβ1-2Manα1 6 GlcNAcβ1-4 Manβ1- 4GlcNAcβ1- 4GlcNAcβ1-Asn 3 2 3 GlcNAcβ1-2Manα1 Fucα1 yβ Xylβ1

Figure 17.3 In planta N-glycan engineering by expression of mammalian glycosyltransferases. See Fig. 17.1 for abbreviations.

Glycan Engineering in Transgenic Plants

■

417

Considering this specificity, the competition between xylosylation/ fucosylation and β(1,4)-galactosylation in plants expressing a human β(1,4)galactosyltransferase was questionable. Indeed, a transfer of galactose on the GlcNAc2Man3GlcNAc2 glycan intermediate would hamper any further addition of immunogenic core-xylose and core-α(1,3)-fucose residues and would result in both the introduction of the missing β(1,4)-galactose and inhibition of deleterious core-xylosylation and core-α(1,3)-fucosylation. Mainly hybrid N-glycans having both human and plant glycoepitopes were identified on proteins from plants expressing a β(1,4)-galactosyltransferase, indicating that competition between endogenous glycosyltransferases and the recombinant enzyme had not occurred (Fig. 17.3). Numerous studies have established that the cytoplasmic tail, the transmembrane domain and the stem region (CTS) play a decisive role in the sub-Golgi distribution of glycosyltransferases. Thus, the subcellular localization of a transferase might be modulated by fusion of its catalytic domain with CTS from another transferase. This strategy was carried out by Bakker et al. (2006) who expressed in plants a human β(1,4)-galactosyltransferase fused to the CTS of the endogenous xylosyltransferase. Mainly glycans Galβ(1,4)GlcNAcMan5GlcNAc2 and Galβ(1,4)GlcNAcMan3GlcNAc2 were identified on proteins and IgG from such transformed plants (Fig. 17.3). This demonstrates that using this strategy, the hybrid transferase was relocated upstream in the Golgi apparatus with endogenous α-mannosidase II and GnT I. In this sub-Golgi compartment, the hybrid transferase was able to transfer terminal β(1,4)-galactose on available GlcNAcMan5GlcNAc2 or GlcNAcMan3GlcNAc2 thus inhibiting any further transfer of core-xylose and core-fucose (Fig. 17.3). Human N-acetylglucosaminyltransferase III (GnT III) has also been expressed in plants in order to engineer endogenous N-glycans in planta (Rouwendal et al. 2007). This transferase is able to introduce a β(1,4)-GlcNAc residue on the β-mannose of the core mammalian N-glycans (bisecting GlcNAc). The analyses of transformed plants revealed that most of the complex-type N-glycans in the plants expressing GnT-III were bisected (Fig. 17.3). Moreover, the authors have shown that the most of the N-glycans of an antibody produced in a plant expressing GnT III are also bisected. However, as observed for the expression of the human β(1,4)galactosyltransferase, expression of the GnT III did not induce any decrease of core-xylosylation and fucosylation. 17.3.3 Removal of immunogenic plant epitopes by knock-out strategies The use of mutants for the production of therapeutic protein devoid of immunogenic plant glycoepitopes has been investigated in Arabidopsis. Iduronidase has been successfully produced in the cgl mutant that is mutated in a gene encoding the Golgi GnT I and thus accumulates Man5GlcNAc2

418

■

Plant Polysaccharides

oligosaccharides on its proteins (Downing et al. 2006) (Fig. 17.1). Another Arabidopis mutant, mur1, is altered in a gene encoding a GDP-Man-4,6dehydratase involved in l-Fuc synthesis and thus N-linked glycans from this mutant do not carry the immunogenic α(1,3)-fucose (Rayon et al. 1999). More recently, an Arabidopis mutant inactivated in both β(1,2)xylosyltransferase and α(1,3)-fucosyltransferases has been isolated by Strasser et al. (2004). The same group then demonstrated that expression of an antibody in this knock-out line resulted in the production of a plantderived IgG that carries a mammalian-like GlcNAc2Man3GlcNAc2 oligosaccharide (Schahs et al. 2007). In addition to the use of mutants altered in the N-glycan pathway, the inactivation of the transfer of immunogenic epitopes in plants has also been successfully achieved by a RNA interference (RNAi) strategy in the aquatic plant Lemna minor (Cox et al. 2006). Coexpression of heavy and light chains with a RNAi transcript designed to silence endogenous β(1,2)xylosyltransferase and α(1,3)-fucosyltransferase resulted in the production of an antibody carrying exclusively a mammalian-like GlcNAc2Man3GlcNAc2 N-glycan (Cox et al. 2006). In recent years, particular attention has been paid to the moss Physcomitrella patens. This plant allows the expression and secretion of complex proteins into a simple aqueous medium, which facilitates downstream processing. Furthermore, P. patens is the only organism in the plant kingdom in which gene targeting, through homologous recombination, occurs as efficiently as in yeast. As in higher plants, this moss introduces on its proteins N-linked glycans carrying imunogenic core-xylose and core-α(1,3)-fucose (Viëtor et al. 2003). However, core-fucosylation and core-xylosylation has been inactivated by homologous recombination (Koprivova et al. 2004). A knock-in strategy has also been used in P. patens. Targeted insertion of the human β(1,4)-galactosyltransferase in both the endogenous β(1,2)-xylosyltransferase and α(1,3)-fucosyltransferase lead to the addition of terminal β(1,4)-galactose to N-glycans in addition to the removal of core xylosylation and fucosylation (Huether et al. 2005). 17.3.4 In planta sialylation of recombinant proteins Plants synthesize and activate 3-deoxy-d-manno-octulosonic acid (Kdo), an α-ketoacid constituent of rhamnogalacturonan-II (RG-II) and structurally related to sialic acid (York et al. 1985; Angatta & Varki 2002). However, for a long time it was believed that plants cannot synthesize sialic acids (Lerouge et al. 1998). In comparison with other expression systems for therapeutic proteins, this is a major disadvantage of plants because, with the exception of IgG, most human serum proteins require sialic acid on terminal positions of their N-glycans. In this context, the recent report by Shah et al. (2003) of the presence of sialylated glycoproteins in Arabidopsis cells received considerable attention. Then, two publications questioned the validity of these data and definitively concluded that N-acetyl- and N-glycolylneuraminic

Glycan Engineering in Transgenic Plants

■

419

acids (Neu5Ac and Neu5Gc), the two main mammalian sialic acids (Séveno et al. 2004), were absent plant cells or detected in amounts so small that they probably originated from contaminants (Zeleny et al. 2006). A search for genes encoding enzymes of the sialic acid pathway only revealed orthologues for sialyltransferase-like genes (Séveno et al. 2004; Harduin-Lepers et al. 2005). Although theses sialyltransferase-like gene products were found in the Golgi apparatus by a proteomic approach (Dunkley et al. 2006), their function has not been discovered to date. In contrast, a search in plant genomes has failed to identify genes potentially encoding GlcNAc or UDPGlcNAc epimerase required for ManNAc synthesis, as well as genes encoding Neu5Ac or CMP-Neu5Ac synthetases (Séveno et al. 2004). Furthermore, biochemical detection of d-mannosamine, the precursor monosaccharide of Neu5Ac, also failed (Paccalet et al. 2007). Taken together, these data allow us to conclude that sialic acids are absent in plants, thus suggesting that genetic manipulation will be required for the in planta synthesis of sialylated proteins by expressing enzymes able to synthesize CMP-Neu5Ac, its Golgi transporter and the appropriate sialyltransferases. The expression of a α-2,6 sialyltransferase in plant by Wee et al. (1998) was the first experiment on the way to the rebuilding of a sialic acid pathway in plants. The authors showed that the transferase was correctly targeted to the Golgi but does not exhibit any capacity for sialic transfer on secreted glycoproteins because of the absence of activated CMP-Neu5Ac in the plant Golgi apparatus. More recently, Misaki et al. (2006) demonstrated that plants do not exhibit any endogenous CMP-sialic acid synthetase activity or CMPsialic acid transport within the Golgi apparatus. They then expressed human CMP-sialic acid synthetase and CMP-sialic acid transporter in tobacco BY2 cells and showed that both proteins are functional in planta. The synthesis of Neu5Ac in plants, another critical step, was solved by the expression of bacterial Neu5Ac-synthesizing enzymes, Neu5Ac lyase from Escherichia coli and Neu5Ac synthase (neuB2) from Campylobacter jejuni (Paccalet et al. 2007). Taken together, these data demonstrate that the introduction into plants of the whole sialylation machinery is feasible and might be only limited by the capacity to steadily transform plants with multiple gene sequences and control the simultaneous expression of all these genes.

17.4

Conclusions

To meet the industrial need, future development of plant-derived recombinant glycoproteins as therapeutics is related to the success of glycoengineering strategies that can provide plant-derived proteins with appropriate glycan profiles lacking immunogenic plant glycoepitopes and harbouring mammalian-like glycan extensions required for their in vivo function. Recent success in both knock-in and knock-out strategies have demonstrated that having plants expressing therapeutic proteins carrying human-like Nglycans is not just a dream. Furthermore, despite the fact that through the

420

■

Plant Polysaccharides

modified glycosylation machinery numerous plant-expressed proteins have acquired unusual N-glycans with terminal β(1,4)-galactose residues or bisecting β(1,4)-GlcNAc residues, or lacking core-xylose and core-fucose, no obvious changes in the physiology of the transgenic plants were observed (Bakker et al. 2006; Rouwendal et al. 2007). This suggests that plants can tolerate, to a large extent, the alteration of their N-glycan processing. In addition, engineering of the N-glycosylation in P. patens did not induce any detectable phenotypic change in standard growth conditions (Huether et al. 2005). The introduction of a full sialic acid pathway in plants will likely be the main goal of future research concerning N-glycan engineering in plants, since producing recombinant sialylated glycoproteins is a major objective for plant biotechnology companies. Recent results in this area have demonstrated that the introduction into plants of the whole sialylation machinery is feasible and might be only limited by the capacity to steadily transform plants with multiple gene sequences and to control their expression. Will plant expression systems tolerate this in-depth alteration of their endogenous N-glycan pathway? For physicochemical reasons, the introduction of terminal acidic monomers such as sialic acids on endogenous proteins might have drastic impacts on plant development. Another problem directly or indirectly connected to sialylation is that, in some cases, it is not enough to have sialylated N-glycans; polyantennary-type oligosaccharides are needed, as is the case for EPO, for example. The assembly of polyantennary N-glycans requires a large number of glycosyltransferases that are absent in the plant glycosylation machinery. Thus, obtaining polyantennary, polysialylated Nglycans is the toughest job ever faced in plant remodelling glycosylation. The data reported in this chapter refer to the engineering of glycans linked to proteins in the context of the production of therapeutic proteins in transgenic plants. However, some of these strategies may help related projects in plant glycobiology such as the in planta engineering of polysaccharides. For instance, the remodelling of glycan sequences based on the relocalization up- or downstream of endogenous or exogenous glycosyltransferases within Golgi subcompartments may be of interest. This will first require identification of the full glycosyltransferase repertoire involved in cell wall biosynthesis and their subcellular localization.

Acknowledgements This research was supported in France by the CNRS and the University of Rouen. We also thank Medicago Inc., Québec, Canada for our fruitful collaboration on the expression of therapeutic glycoproteins in alfalfa plants and Sébastien Calbo for the critical reading of this article. J.A.C. is grateful for the support and collaboration between Biomedical Research and AgroBiotech Directions from the Center for Genetic Engineering and Biotechnology, Havana, Cuba.

Glycan Engineering in Transgenic Plants

■

421

References Aalberse, R.C., Koshte, V., Clemens, J.G. (1981) Immunoglobulin E antibodies that crossreact with vegetable foods, pollen, and Hymenoptera venom. Journal of Allergy and Clinical Immunology, 68, 356–364. Angatta, T., Varki, A. (2002) Chemical diversity in the sialic acids and related a-keto acids: an evolutionary perspertive. Chemical Review, 102, 439–469. Bakker, H., Bardor, M., Molthoff, J.W., et al. (2001) Galactose-extended glycans of antibodies produced by transgenic plants. Proceedings of the National Academy of Sciences of the U S A, 98, 2899–2904. Bakker, H., Rouwendal, G.J., Karnoup, A.S., et al. (2006) An antibody produced in tobacco expressing a hybrid beta-1,4-galactosyltransferase is essentially devoid of plant carbohydrate epitopes. Proceedings of the National Academy of Sciences of the U S A, 103, 7577–7582. Bardor, M., Loutelier-Bourhis, C., Paccalet, T., et al. (2003a) Monoclonal C5–1 antibody produced in alfalfa plants exhibits a N-glycosylation that is homogenous and suitable for glyco-engineering into human-compatible structures. Plant Biotechnology Journal, 1, 451–462. Bardor, M., Faveeuw, C., Fitchette, A.C., et al. (2003b) Immunoreactivity in mammals of two typical plant glyco-epitopes, core α(1,3)-fucose and core xylose. Glycobiology, 13, 427–434. Burda, P., Aebi, M. (1999) The dolichol pathway of N-linked glycosylation. Biochimica et Biophysica Acta, 1426, 239–257. Cabanes-Macheteau, M., Fitchette-Laine, A.C., Loutelier-Bourhis, C., et al. (1999) N-glycosylation of a mouse IgG expressed in transgenic tobacco plants. Glycobiology, 9, 365–372. Cabrera, G., Cremata, J.A., Valdés, R., et al. (2004) Influence of culture conditions on the N-glycosylation of a monoclonal antibody specific for recombinant hepatitis B surface antigen. Biotechnology and Applied Biochemistry, 40, 1–10. Cox, K.M., Sterling, J.D., Regan, J.T., et al. (2006) Glycan optimization of a human monoclonal antibody in the aquatic plant Lemna minor. Nature Biotechnology, 24, 1591–1597. Downing, W.L., Galpin, J.D., Clemens, S., et al. (2006) Synthesis of enzymatically active human alpha-L-iduronidase in Arabidopsis cgl (complex glycan-deficient) seeds. Plant Biotechnology Journal, 4, 169–181. Dunkley, T.P.J., Hester, S., Shadforth, I.P., et al. (2006) Mapping the Arabidopsis organelle proteome. Proceedings of the National Academy of Sciences of the U S A, 103, 6518–6523. Elbers, I.J.W., Stoopen, G.M., Bakker, H., et al. (2001) Influence of growth conditions and developmental stage on N-glycan heterogeneity of transgenic immunoglobulin G and endogenous proteins in tobacco leaves. Plant Physiology, 126, 1314–1322. Faye, L., Chrispeels, M.J. (1988) Common antigenic determinants in the glycoproteins of plants, molluscs and insects. Glycoconjugate Journal, 5, 245–256. Faye, L., Gomord, V., Fitchette-Lainé, A.-C., Chrispeels, M.J. (1993) Affinity purification of antibodies specific for Asn-linked glycans containing alpha 1-3 fucose or beta 1-2 xylose. Analytical Biochemistry, 209, 104–108. Fischer, R., Stoger, E., Schillberg, S., Christou, P., Twyman, R.M. (2004) Plant-based production of biopharmaceuticals. Current Opinion in Plant Biology, 7, 152–158.

422

■

Plant Polysaccharides

Fitchette-Lainé, A.C., Gomord, V., Chekkafi, A., Faye, L. (1994) Distribution of xylosylation and fucosylation in the plant Golgi apparatus. Plant Journal, 5, 673–682. Fitchette-Lainé, A.C., Gomord, V., Cabanes, M., et al. (1997) N-glycans harbouring the Lewis a epitope are expressed at the surface of plant cells. Plant Journal, 12, 1411–1417. Garcia-Casado, G., Sanchez-Monge, R., Chrispeels, M.J., Armentia, A., Salcedo, G., Gomez, L. (1996) Role of complex asparagine-linked glycans in the allergenicity of plant glycoproteins. Glycobiology, 6, 471–477. Gomord, V., Sourrouille, C., Fitchette, A.C., et al. (2004) Production and glycosylation of plant-made pharmaceuticals: the antibodies as a challenge. Plant Biotechnology Journal, 2, 83–100. Harduin-Lepers, A., Mollicone, R., Delannoy, P., Oriol, R. (2005) The animal sialyltransferases and sialyltransferase-related genes: A phylogenetic approach. Glycobiology, 15, 805–817. Helenius, A., Aebi, M. (2001) Intracellular functions of N-linked glycans. Science, 291, 2364–2369. Hiatt, A.C., Cafferkey, R., Bowdish, K. (1989) Production of antibodies in transgenic plants. Nature, 342, 76–78. Huether, C.M., Lienhart, O., Baur, A., et al. (2005) Glyco-engineering of moss lacking plant-specific sugar residues. Plant Biology, 7, 292–299. Jefferis, R. (2005) Glycosylation of recombinant antibody therapeutics. Biotechnology Progress, 21, 11–16. Johnson, K.D., Chrispeels, M.J. (1987) Substrate specificities of N-acetylglucosaminyl, fucosyl-, and xylosyltransferases that modify glycoproteins in the Golgi apparatus of bean cotyledons. Plant Physiology, 84, 1301–1308. Kelm, S., Schauer, R. (1997) Sialic acids in molecular and cellular interactions. International Review of Cytologyl, 175, 137–240. Ko, K., Tekoah, Y., Rudd, P.M., et al. (2003) Function and glycosylation of plantderived antiviral monoclonal antibody. Proceedings of the National Academy of Sciences of the U S A, 100, 8013–8018. Koprivova, A., Stemmer, C., Altmann, F., et al. (2004) Targeted knockouts of Physcomitrella lacking plant-specific immunogenic N-glycans. Plant Biotechnology Journal, 2, 517–523. Kurosaka, A., Yano, A., Itoh, N., Kuroda, Y., Nakagawa, T., Kawasaki, T. (1991) The structure of a neural specific carbohydrate epitope of horseradish peroxidase recognized by anti-horseradish peroxidase antiserum. The Journal of Biological Chemistry, 266, 4168–4172. Leiter, H., Mucha, J., Staudacher, E., Grimm, R., Glössl, J., Altmann, F. (1999) Purification, cDNA cloning, and expression of GDP-L-Fuc: Asn-linked GlcNAc α1,3-fucosyltransferase from mung beans. Journal of Biological Chemistry, 274, 21830–21839. Lerouge, P., Cabanes-Macheteau, M., Rayon, C., Fitchette-Laine, A.C., Gomord, V., Faye, L. (1998) N-glycoprotein biosynthesis in plants: recent developments and future trends. Plant Molecular Biology, 38, 31–48. Ma, J.K.C., Chikwamba, R., Sparrow, P., Fischer, R., Mahoney, R., Twyman, R.M. (2005) Plant-derived pharmaceuticals – the road forward. Trends in Plant Science, 10, 580–585.

Glycan Engineering in Transgenic Plants

■

423

Matsumoto, S., Ikura, K., Ueda, M., Sasaki, R. (1995) Characterization of a human glycoprotein (erythropoietin) produced in cultured tobacco cells. Plant Molecular Biology, 27, 1163–1172. McManus, M.T., Mc Keating, J., Secher, D.S., et al. (1988) Identification of a monoclonal antibody to abscission tissue that recognises xylose/fucose-containing Nlinked oligosaccharides from higher plants. Planta, 175, 506–512. Mimura, Y., Church, S., Ghirlando, R., et al. (2000) The influence of glycosylation on the thermal stability and effector function expression of human IgG1-Fc: properties of a series of truncated glycoforms. Molecular Immunology, 7, 697–706. Misaki, R., Fujiyama, K., Seki, T. (2006) Expression of human CMP-N-acetyneuraminic acid synthetase and CMP-sialic acid transporter in tobacco suspension-cultured cell. Biochemical and Biophyical Research Communications, 339, 1184–1189. Nose, M., Wigzell, H. (1983) Biological significance of carbohydrate chains on monoclonal antibodies. Proceedings of the National Academy of Sciences of the U S A, 80, 6632–6636. Paccalet, T., Bardor, M., Rihouey, C., et al. (2007) Engineering of a sialic acid synthesis pathway in transgenic plants by expression of bacterial Neu5Ac-synthesizing enzymes. Plant Biotechnology Journal, 5, 12–25. Palacpac, N.Q., Yoshida, S., Sakai, H., et al. (1999) Stable expression of human beta 1,4-galactosyltransferase in plant cells modifies N-linked glycosylation patterns. Proceedings of the National Academy of Sciences of the U S A, 96, 4692–4697. Petrucelli, S., Otegui, M.S., Lareu, F., et al. (2006) A KDEL-tagged monoclonal antibody is efficiently retained in the endoplasmic reticulum in leaves, but is both partially secreted and sorted to protein storage vacuoles in seeds. Plant Biotechnology Journal, 4, 511–527. Raju, T.S., Briggs, J.B., Borge, S.M., Jones, A.J.S. (2000) Species-specific variation in glycosylation of IgG : evidence for the species-specific sialylation and branchspecific galactosylation and importance for engineering recombinant glycoprotein therapeutics. Glycobiology, 10, 477–486. Ramirez, N., Rodriguez, M., Ayala, M., et al. (2003) Expression and characterization of an anti-(hepatitis B surface antigen) glycosylated mouse antibody in transgenic tobacco (Nicotiana tabacum) plants and its use in the immunopurification of its target antigen. Biotechnology and Applied Biochemistry, 38, 223–230. Rayon, C., Cabanes-Macheteau, M., Loutelier-Bourhis, C., et al. (1999) Characterization of N-glycans from Arabidopsis thaliana. Application to a fucose-deficient mutant. Plant Physiology, 119, 725–733. Rouwendal, G.J.A., Wuhrer, M., Florack, D.E.A., et al. (2007) Efficient introduction of a bisecting GlcNAc residue in tobacco N-glycans by expression of the gene encoding human N-acetylglucosaminyltransferase III. Glycobiology, 17, 334–344. Rudd, P.M., Leatherbarrow, R.J., Rademacher, T.W., Dwek, R.A. (1991) Diversification of the IgG molecule by oligosaccharides. Molecular Immunology, 28, 1369–1378. Schahs, M., Strasser, R., Stadlmann, J., Kunert, R., Rademacher, T., Steinkellner, H. (2007) Production of a monoclonal antibody in plants with a humanized Nglycosylation pattern. Plant Biotechnology Journal, 5, 657–663. Séveno, M., Bardor, M., Paccalet, T., Gomord, V., Lerouge, P., Faye, L. (2004) Glycoprotein sialylation in plants? Nature Biotechnology, 22, 5–6. Shah, M.M., Fuliyama, K., Flynn, C.R., Joshi, L. (2003) Sialylated endogenous glycoconjugates in plant cells. Nature Biotechnology, 21, 1470–1471.

424

■

Plant Polysaccharides

Sriraman, R., Bardor, M., Sack, M., et al. (2004) Recombinant anti-hCG antibodies retained in the endoplasmic reticulum of transformed plants lack core-xylose and core-α(13)-fucose residues. Plant Biotechnology Journal, 2, 279–287. Strasser, R., Altmann, F., Mach, L., Glössl, J., Steinkellner, H. (2004) Generation of Arabidopsis thaliana plants with complex N-glycans lacking β1,2-linked xylose and core α1,3-linked fucose. FEBS Letters, 561, 132–136. Tekoah, Y., Ko, K., Koprowski, H., et al. (2004) Controlled glycosylation of therapeutic antibodies in plants. Archives of Biochemistry and Biophysics, 426, 266–278. Triguero, A., Cabrera, G., Cremata, J., Yuen, C.T., Wheeler, J., Ramirez, N.I. (2005) Plant-derived mouse IgG monoclonal antibody fused to KDEL endoplasmic reticulum-retention signal is N-glycosylated homogeneously throughout the plant with mostly high-mannose-type N-glycans. Plant Biotechnology Journal, 3, 449–457. Van Ree, R., Cabanes-Macheteau, M., Akkerdaas, J., et al. (2000) Beta(1,2)-xylose and alpha(1,3)-fucose residues have a strong contribution in IgE binding to plant glycoallergens. Journal of Biological Chemistry, 275, 11451–11458. Viëtor, R., Loutelier-Bourhis, C., Fitchette, A.C., et al. (2003) Protein N-glycosylation is similar in the moss Physcomitrella patens and in higher plants. Planta, 203, 269–275. Wee, E.Q., Sherrier, D.J., Prime, T.A., Dupree, P. (1998) Targeting of active sialyltransferase to the plant Golgi apparatus. Plant Cell, 10, 1759–1768. Wilson, I.B., Altmann, F. (1998) Structural analysis of N-glycans from allergenic grass, ragweed and tree pollens: core alpha1,3-linked fucose and xylose present in all pollens examined. Glycoconjugate Journal, 15, 1055–1070. Wright, A., Morrison, S.L. (1998) Effect of C2-associated carbohydrate structure on Ig effector function: studies with chimeric mouse-human IgG1 antibodies in glycosylation mutants of Chinese hamster ovary cells. Journal of Immunology, 160, 393–402. York, W.S., Darvill, A.G., McNeil, M., Albersheim, P. (1985) 3-desoxy-D-manno-2octulosonic acid (KDO) is a component of rhamnogalacturonan II, a pectic polysaccharide in the primary cell walls of plants. Carbohydrate Research, 138, 109–126. Zeitlin, L., Olmsted, S.S., Moench, T.C., et al. (1998) A humanized monoclonal antibody produced in transgenic plants for immunoprotection of the vagina against genital herpes. Nature Biotechnology, 16, 1361–1364. Zeleny, R., Kolarich, D., Strasser, R., Altmann, F. (2006) Sialic acid concentrations in plants are in the range of inadvertent contamination. Planta, 5, 1–6. Zeng, Y., Bannon, G., Thomas Hayden, V., Rice, K., Drake, R., Elbein, A. (1997) Purification and specificity of β1,2-xylosyltransferase, an enzyme that contributes to the allergenicity of some plant proteins. Journal of Biological Chemistry, 272, 31340–31347.

Annual Plant Reviews (2011) 41, 425–450 doi: 10.1002/9781444391015.ch18

Chapter 18

POLYSACCHARIDE NANOBIOTECHNOLOGY: A CASE STUDY OF DENTAL IMPLANT COATING Marco Morra1, Clara Cassinelli1, Giovanna Cascardo1, Hanna Kokkonen2, Juha Tuukkanen2, Claudio Della Volpe3, Stefano Siboni3, Giordano Segatta3, Marco Brugnara3 and Giacomo Ceccone4 1

Nobil Bio Ricerche, Via Valcastellana 26, 14037 Portacomaro (AT), Italy Department of Anatomy and Cell Biology, University of Oulu, PO Box 5000, 90014 Oulu, Finland 3 Department of Materials Engineering and Industrial Technologies, University of Trento, Trento, Italy 4 Institute for Health and Consumer Protection, European Commission Joint Research Center, Ispra, Italy 2

Manuscript received October 2009

Abstract: Surface modification of biomaterials and medical devices by polysaccharides is presently a topic of great interest, both at fundamental and applied levels. Plant polysaccharides constitute an untapped resource in this regard. The very self-assembly properties of the polymers that the plant cell exploits during cell wall biogenesis may be exploited in assembly of contiguous coatings of medical devices and implants. Cell wall polysaccharides not only display advantageous physical properties but some are also known to be biologically active in a mammalian context. Bioactivities include anti-inflammatory and immunomodulatory effects that are hypothesized to suppress rejection and promote implant integration. These activities are in most cases associated with polysaccharides of the pectin family. The present study thus focuses on engineering biocompatible Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

425

426

■

Plant Polysaccharides

surfaces using rhamnogalacturonan-I. The target application is titanium dental implants and the study reports extensively on the methods used for surface characterization, both biological methods that probe biocompatability and nanotechnology methods used for the physical characterization of the modified surfaces. Keywords: dental implants; modified hairy regions; pectins; surface analysis; surface modification; titanium; wettability

18.1 Introduction: titanium dental implants and surface modifications The term ‘osseointegration’ is used to define the growth of bone in the close proximity of materials implanted in bony tissues. ‘Fibrointegration’, on the other hand, is the embedding of implanted materials in fibrous, soft tissue. A fibrous capsule that walls off the intruder from the body is often the ultimate fate of implanted materials. Careful experimental protocols and the use of biocompatible materials like titanium, as defined in the pioneering work of Branenmark (Branenmark et al. 1977), have made it possible to define the criteria for reliable and reproducible osteointegration of implanted bone-contacting devices. Osseointegrated titanium screws (Fig. 18.1), able to withstand masticatory loads and qualified by the criterion of less than 100 Å between bone and titanium (Ratner 2001), are nowadays at the basis of the burgeoning field of implant dentistry. Titanium implant devices are being widely used for a variety of indications, and most of the various techniques in use are evidence-based and predictable. Interfacial interactions at the

Figure 18.1 Titanium dental implant. Typical diameter ranges from 3 to 6 mm, length from 8 to 16 mm.

Polysaccharide Nanobiotechnology

■

427

bone–implant interface are recognized as the key to osteointegration and an enormous literature on titanium surfaces and interfaces exists (Davies 1991; Davies 2000, Brunette et al. 2001). There is a wide range of approaches to the surface modification of titanium to further improve clinical results and extend the spectrum of indications. Despite its significant success as a biomaterial, titanium and its surface are still being actively investigated: for instance, the need exists to address difficult clinical settings, such as an intended implant site compromised because of poor bone quality. The latter definition, from a clinical point of view, encompasses low bone density, in the case of highly cancellous bone (bone with a latticed or porous structure), or low vascularity, in the case of primarily cortical bone, or insufficient quantity of bone (in terms of the width of the alveolar ridge). Increased life expectancy poses new challenges for surface treatment of implant devices able to locally improve bone density or to accelerate initial healing times during integration even in old, or, in general, pathological bone. Briefly, many open issues still exist, despite the large literature and many applications mentioned earlier. Surface engineering approaches are actively investigated to promote cell adhesion and differentiation on titanium implant surfaces, in order to promote quick peri-implant bone formation. Traditionally, approaches to surface modification of titanium have been based on the control of surface topography (Brunette et al. 2001), on ceramic coatings and, more recently, on physicochemical (Rupp et al. 2006) or inorganic approaches (Cooper et al. 2006). At present, a significant research effort is aimed at the biochemical modification of titanium surfaces (BMTiS) (Puleo & Nanci 1999; Morra 2006; Morra 2007), whose goal is to use surface-linked or surface-delivered biomolecules and biomolecular signalling for the purpose of inducing specific cell and tissue responses at the host tissue– implant interface. BMTiS by proteins, peptides and polysaccharides has been reported in the literature (Morra 2006). A targeted research project was recently undertaken to investigate the role of nanometre-scale layers of enzymatically tailored pectin hairy regions, covalently linked to titanium surfaces, in peri-implant bone formation. The expertise of the multidisciplinary team involved includes plant biology, purification and manufacturing of tailored polysaccharides, surface modification and characterization of biomaterials, in vitro and in vivo biocompatibility assessment, and the manufacture of medical devices. The rationale for such a study is based on several factors:

• Like every implant procedure, bone formation around dental implants involves a healing process. Plant carbohydrates are a large reservoir of pharmacologically active molecules and since the beginning of human history we have exploited them unknowingly in traditional herbs and medicine (Paulsen 2000). A role for several plant polysaccharides in wound healing has been definitely proven, through interaction with the complement system and several stages of the inflammatory phase. Pectin hairy regions have often been suggested as the source of bioactivity.

428

■

Plant Polysaccharides

• Cell–cell interactions rely on displayed carbohydrate moieties and on sugar epitope-specific binding proteins; for example, members of the galectin and selectin gene families. Specific sugar epitopes are at the basis of a universal ‘sugar language’ that, in the spirit of BMTiS, should be understood and applied to direct the interfacial behaviour of medical devices. • Pectin hairy regions are nicely built polysaccharides that can be further finely tuned by enzymatic treatment (modified hairy regions, MHR) (Schols & Voragen 1996). The physicochemical properties of MHRs, in terms of chemical nature, number and length of side chains, for example, can affect interfacial properties and relevant interfacial cell behavior, such as cell adhesion, spreading and activity (Morra et al. 2004) • Surface linking of nanolayers of polysaccharides can control surface hydrophilicity, as reflected in wetting behaviour (Morra & Cassinelli 1999). The combination of the chemical flexibility of plant polysaccharides and MHRs described above with the requirements for proper surface hydrophilicity, as related to existing theories, is an important asset in the surface modification of medical devices. This chapter describes some outcomes of our research project on the surface modification of titanium dental implants by MHRs. Because of space limitations, it is obviously not possible to give a detailed account; rather, the aim of the chapter is to give an overview of a promising new field, where different disciplines merge and define a fascinating area of study, rich in implications. Section 18.2 discusses the evaluation of candidate MHRs in an in vitro test involving osteogenic cells and chemicophysical characterization by wettability measurements. The surface modification of titanium and titanium implant devices is then described, followed by some reflections and conclusions.

18.2 Rationale for the surface modification of titanium dental implants by nanolayers of MHRs 18.2.1 Background Pectic polysaccharides, present in most plant tissues, are an interesting area of investigation. Pectin structural elements depend on the plant material and the type of tissue (Schols & Voragen 1996), as widely reported in other chapters of this book. A simple procedure to isolate rhamnogalacturonan-I (RG-I)-rich fractions on a semi-large scale is to use commercial pectinase preparations to liquefy the tissue of fruit and vegetables (Schols et al. 1994). The enzymes degrade the homogalacturonan part (smooth regions) of the pectin present, leaving the so-called hairy regions (HR).

Polysaccharide Nanobiotechnology

■

429

The various structural elements of pectin could be of great potential interest in the surface modification of medical devices since they are polymers with negatively charged segments, and their hydrophilicity/hydrophobicity ratio can be affected by controlling the length of the side chains. Furthermore, the methyl esterification present in native pectins (partly neutralizing the charges present) can easily be removed by alkali or enzymes in order to vary the total charge on the molecule. Also, acetyl substitution of the RG-I segments will make the rather hydrophilic molecule slightly more hydrophobic. The different pectin/RG-I structures suggest the possibility of oriented immobilization of a RG backbone bearing side chains of controllable length and nature (Morra et al. 2004), affecting interfacial hydration and reportedly showing bioactive properties (Paulsen 2000). A number of structurally different enzymatically engineered MHRs were prepared and evaluated as molecular candidates for bone implant nanocoatings. The sugar content of MHRs described in the rest of this chapter is reported in Table 18.1. Considerations on how to link them to solid surfaces are reported in the following section. 18.2.2 Modification of solid surfaces by MHRs: general principles The plant polysaccharides that were investigated in the study are basically polyanionic, owing to the presence of the polygalacturonic acid backbone.

Table 18.1 Sugar composition (mol %) of the some of the modified hairy regions (MHRs) investigated in this work. Data were obtained by Rene Verhoef and Henk Schols, Laboratory of Food Chemistry, Wageningen University, Wageningen, The Netherlands

Sugar

Apple MHRα

Carrot MHRC

Potato MHRP

Apple MHRA

Apple MHRB

Fuc Rha Ara Gal Glc Man Xyl GalA GlcA OMe OAc Rha:GalA Total sugar (w/w%)

1 15 25 15 1 nd 12 31 nd 37 45 0.49 86

2 22 8 32 2 nd 2 30 2 16 46 0.75 82

1 10 14 28 4 nd nd 42 nd 57 59 0.24 70

0 9 51 10 0 nd 8 22 nd 42 60 0.41 86

0 11 11 20 3 nd 18 37 nd 34 11 0.30 78

nd, not determined; OAc, moles Ac per 100 moles GalA (max 200%); OMe, moles MeOH per 100 moles GalA (max 100%).

430

■

Plant Polysaccharides

They are water soluble, and immobilization to surfaces of biomaterials in general and titanium surface in particular requires the establishment of strong interactions. Examples involving both ionic and covalent coupling are reported in the literature. Polyanionic smooth regions of pectins can be linked to polycationic surfaces by polyelectrolyte formation. Marudova et al. (2005) investigated the deposition of alternating layers of pectin and polycationic chitosan at a solid surface using surface plasmon resonance. Binding to the surface was irreversible over the time scales examined. The thickness of the deposited layer was dependent on the biopolymer concentration. Sequential deposition resulted in the formation of multilayers with an essentially linear growth rate (Krzeminski et al. 2006). The multilayers were relatively hydrated structures with estimates of solids content in the range 10–32% w/w. The more highly esterified pectins had a tendency to form more hydrated structures, which showed a strong deswelling when poly-llysine (PLL) was added to a freshly deposited pectin layer. Covalent coupling of biomolecules to solid substrates is the strategy of choice for implant devices that require long-term stability of the surface layer. Covalent coupling requires suitable chemical groups on biomaterials surfaces, a step often called ‘surface functionalization’, as moieties suitable for coupling, the carbohydrate structure offers hydroxyl groups, which can be further manipulated to aldehyde through periodate oxidation, and carboxyl groups. As far as the substrate material is concerned, amino groups are by far the most frequently used groups to anchor polysaccharides, either by coupling to the polysaccharide carboxyl groups or by cyanoborohydride coupling to aldehyde groups. Surface functionalization of the substrate by amino groups can be achieved through several different techniques: silane coupling agents, polyethyleneimine (PEI) adsorption and deposition from plasma are the most frequently used (Hoffman & Hubbell 2004). Plasma (glow discharge) techniques are especially useful because they allow to deposit ultra-thin (nanoscale), adherent, conformal coatings (Yasuda 1985). By proper tuning of treatment conditions, these films can be endowed with excellent stability in an aqueous environment. In plasma treatment, gases or vapours are introduced into the vacuum chamber of the plasma reactor and current passes between the electrodes. Electrons accelerated by the electrical field create and sustain the plasma. The choice of the proper gas or vapour depends on the specific application of the treatment. Argon, oxygen or air are often used for surface cleaning, to remove organic contaminants. Amino-containing molecules are used to introduce amino groups that can be used for coupling reactions. In plasma discharges the precursor monomer is fragmented, and the molecular fragments recombine and are deposited as a thin film on the substrate surface. Allylamine is possibly the most common monomer, and its use for biochemical modification of titanium implants is described by Puleo et al. (2002). In our experiments allylamine deposition was used to functionalize a range of different substrates, from polystyrene Petri dishes for in vitro studies

Polysaccharide Nanobiotechnology

■

431

of cell behaviour to actual titanium implants (Morra et al. 2004). MHRs were covalently coupled to the functionalized substrates through carbodiimide chemistry. Relevant results, in terms of surface chemistry, are described in section 18.3, but it is of interest here to present some atomic force microscopy (AFM) result to provide insights into the surface hierarchy obtained by the process. In particular, the thickness of the surface-linked MHR layer in aqueous solution was measured by ‘scratching mode’ AFM on mica coated by covalently linked MHRs. Mica is the substrate of choice in AFM, because it can provide atomically flat planes upon cleaving, which allow the vertical resolution of AFM to be fully exploited. Briefly, freshly cleaved mica was functionalized by allylamine and different MHRs were covalently coupled by overnight reaction in an unbuffered aqueous solution containing watersoluble carbodiimide. Samples were washed and observed by AFM in 0.001 M NaCl solution. A 4 × 4 μm image was acquired, increasing the contact force until scratching and removal of the MHR surface layer was obtained. The height profile along a section passing through the scratched area was then measured and the thickness of the surface layer was evaluated. Figure 18.2 shows a typical result. In particular, the image shows a portion of the scratched area at the boundary with a non-scratched one. Outside the scratched area the mica surface appears coated by a homogeneous and dense layer, which is actually the covalently linked MHR layer. The figure shows an example of the measurements performed: a section through the scratched area was taken and the height profile measured along the white line shown in the figure. Depending on the MHR nature (molecular weight, length of side chains) the measured layer thickness ranged from 6 to 11 nm, consistent with a monolayer in ‘side on’ configuration, as expected from the nature of the interfacial coupling reaction (multipoint binding of polygalacturonan carboxyl groups). 18.2.3 Experiments on osteogenic cell adhesion to MHR-coated surfaces Bone is a dynamic tissue, which is continuously remodelled by two bone cell types: bone-resorbing osteoclasts and bone-forming osteoblasts. The operational areas of the coupled functions of these bone cells are called bone remodelling units, in which osteoblasts produce new bone after osteoclastic resorption of the old bone tissue. Osteoclasts differentiate from bone marrow haematopoietic stem cells by fusion of mononuclear progenitors into large multinuclear bone-resorbing cells. Osteoclasts can be detected as multinuclear tartrate-resistant acid phosphatase (TRACP)-positive cells. Bone resorption occurs via tight attachment of osteoclasts to the target bone area, in which both organic (mainly collagen) and inorganic (hydroxyapatite) bone matrix constituents are dissolved. Osteoblasts differentiate from mesenchymal stem cells. Mature osteoblasts are capable of producing collagenrich osteoid matrix and mineralizing it. Osteoblasts are adherent and

432

■

Plant Polysaccharides A

0

2

4

6

nm 8

10

12

14

16

B

0

100

200

300

400 500 Plane, nm

600

700

800

900

Figure 18.2 Result of ‘scratching mode AFM’ used to evaluate the thickness of the surface linked MHR layer. (A) AFM image at the boundary between scratched and nonscratched areas. (B) Section measurements along the white line of Fig. 18.2A, showing that the thickness of the surface MHR layer is about 10 nm.

anchorage-dependent cells, which have to attach to certain extracellular matrix (ECM) proteins, such as fibronectin, to function and differentiate properly. Kokkonen et al. (2007) studied how bone cells attach, proliferate, and differentiate on modified pectin coatings. Both continuous MC3T3-E1 and

Polysaccharide Nanobiotechnology

■

433

primary bone cells were cultured on Petri dishes surface-modified according to the principles described in the previous section. MHRA and MHRB (Table 18.1) were chosen for surface modification. The results show that the covalently linked layers of plant origin, have a significant effect on the behaviour of mammalian cells. MC3T3-E1 cells grew more numerously on MHRB than on MHRA. Clear and numerous focal adhesions were detected on MHRB, but none on MHRA. In agreement with previous observations, both the density and the spreading of primary bone cells were clearly diminished on MHRA compared with MHRB. The total number of TRACP+ cells (osteoclasts) on MHRB exceeded that of control tissue culture polystyrene (TCPS), whereas it did not significantly differ between MHR-B and cells cultured on control bone fragments. The amount of TRACP+ cells was clearly diminished on MHRA compared with MHRB, TCPS, and bone. TRACP+ osteoclasts with one or two nuclei and with three or more nuclei (multinuclear) were counted separately. On MHRA, the amount of cells with one or two nuclei was clearly lower than those on MHRB, TCPS, and bone. The number of cells with one or two nuclei did not differ significantly between control substrata. Interestingly, MHRB contained more TRACP+ cells with one or two nuclei than TCPS or bone. However, the amount of multinuclear osteoclasts was significantly greater on bone than on TCPS or either of the polysaccharides, between which the number of multinuclear osteoclasts did not differ. Calculation of the ratios of multinuclear TRACP+ cells of all TRACP+ cells showed significantly different results between both MHRs and bone; on bone the proportion of multinuclear TRACP+ cells of all the TRACP+ cells proved to be greater than that on polysaccharides. In contrast, such a difference was not detected between the MHRs. The activity of the osteoclasts on MHRB and bone was also verified by detecting the actin rings. On MHRB, approximately one-third of the osteoclasts exhibited an actin ring, whereas on bone all of the osteoclasts contained an actin ring. In contrast, actin rings were not detected on MHRA. Total cell numbers representing the whole mixed-cell populations were counted from the MHR samples. On MHRB the total cell number proved to be greater than that on MHRA. This observation can be considered consistent with MC3T3-E1 data. These results clearly show that bone cells are sensitive to modifications of pectic coatings. The cells preferred rhamnogalacturonans with shorter side chains in all parameters studied. This preference can be seen as both an increased number and also a more dispersed morphology of the cells on MHRB coating in comparison to that on MHRA coating, in agreement with results obtained using different cell lines. Moreover, results prove that bone cells can tolerate pectin-coated surfaces, and that the behaviour of bone cells depends at least on the length of the hairy regions of the rhamnogalacturonan molecules. This opens up possibilities of either stimulating or inhibiting local bone growth. Bussy et al. (2008) reported another interesting study involving bone cells on MHR coatings. Two different MHRs (MHRα and MHRB) from apple

434

■

Plant Polysaccharides

pectin, differing in relative amounts and lengths of neutral side chains, were covalently linked to Thermanox. In contrast to the study by Kokkonen et al. mentioned above, MHRα contains less arabinose than MHRA. MHRα therefore is suggested to have shorter arabinan side chains than MHRA, although still much longer than the chains in MHRB. In the study by Bussy and coworkers, bone tissue samples were obtained from the tibias of 14-dayold chick embryos. They were cut into 2-mm3 explants and cultivated in semisolid medium for 14 days in contact with Thermanox slides coated with MHRα or MHRB. Controls were uncoated Thermanox or aminated Thermanox. Depending on the surface properties of the substratum, cells could escape from each bone explant, spread, migrate, and proliferate, then develop a cell layer around the explant. Results show that cells migrated similarly on MHRB and the aminated control. Migration was increased on the Thermanox control, and cells did not migrate on MHRα. The ability of cells to differentiate was assessed on Thermanox and on MHRB; in both conditions, cell layers showed alkaline phosphatase (ALP) positive cells. Taken as a whole, the results obtained from chick embryo bone explant cultures 14 days post-seeding evidenced the existence of bone cell survival, migration, and differentiation on MHRB. In contrast, cells seeded on MHRα were unable to survive and form a cell layer around explants. Further studies performed on different osteogenic cells, including human mesenchymal cell from bone marrow, suggest that osteoblast differentiation, as indicated by ALP activity, is promoted on MHRB-coated surfaces. Thus, the short-haired MHRB is a potential good candidate as an interfacial layer in bone-contacting applications. 18.2.4 Wettability of MHR-coated surfaces The wettability of a solid surface is a function of the chemical nature of its outermost molecular layer. Interfacial interactions as dictated by surface chemistry and expressed by surface energetic parameters control a number of events highly relevant for implant devices. In particular, protein adsorption and the ensuing cell adhesion to adsorbed proteins have widely been discussed in terms of wetting theories (Vogler 2001). Recent studies on dental implant materials have revived previous suggestions that the hydrophilicity of implant surfaces may be especially important during initial conditioning by proteins and during initial cell adhesion (Rupp et al. 2006); and that implanted medical devices might benefit from hydrophilic surfaces and reduced interfacial foreign body response (Anderson & Jones 2007) eliminating the presence of potentially harmful cells and therefore promoting bony ingrowth and bone formation. Hydrophilic, covalently linked MHR layers able to support bone cell adhesion and differentiation can thus play an important role in the surface modification of titanium dental implants. To try to establish a relationship between the chemical nature and the wetting properties of surface-linked MHR layers a systematic study was

Polysaccharide Nanobiotechnology

■

435

undertaken to evaluate of the wettability of samples obtained by the covalent linking of MHRs to polystyrene culture dishes. As a first step, a roughness analysis of the samples was performed to assess the possible impact of roughness on the wettability results; in fact the presence of roughness is commonly considered one of the main causes of hysteresis, i.e. of the difference between advancing and receding contact angles, apart from chemical heterogeneity. The roughness has been checked using a Dektak3 device and using a scan length of 1–2 mm; five scans were performed on each sample and the mean value and its standard deviation considered. The parameter evaluated for each scan was the quantity Ra, which corresponds to the absolute mean value of the deviations of the profile with respect to the fitting plane; this quantity and its treatment are defined elsewhere (ANSI 1985). The following results were obtained from experiments: as far as roughness is concerned, data show a Ra value of 0.020 ± 0.007 μm for the substrate PS. Treated samples (both after amination and after MHR coupling) are rougher (typical Ra about 0.060 μm) but the difference between them is not significant. This shows that roughness is low and not able to modify the wettability, which therefore depends only on the chemistry of the surfaces. The wettability analysis was performed using two different liquids: Milli Q-water with an initial resistivity of 18.6 MΩm and phosphate buffered saline (PBS), the same as is used in cell culture experiments. The surface tension of the water, checked by the Wilhelmy technique at room temperature, was 71.8 ± 0.4 mJ/m2 and that of PBS was 72.6 ± 0.9 mJ/m2 at the same temperature. Two techniques were used: the Wilhelmy technique and the sessile method. In both cases it was possible to obtain three or four different values of the contact angle. In fact, apart from the advancing and receding angles it is possible to obtain a static ‘as placed’ angle in the sessile case and a further equilibrium angle in both methods. The technique used to obtain this equilibrium or ‘global equilibrium’ (Marmur 2006) is the application of a mechanical vibration of the meniscus by a loudspeaker and a wave generator, as fully described in the literature; this angle is known as the vibration induced equilibrium contact angle (VIECA; Della Volpe et al. 2001). It may be useful to recall that on a flat surface, such as this, the advancing and receding angles indicate respectively the areas of the surface with the lowest and highest energy. The static angle is generally close to the advancing angle, but not closely repeatable, and the equilibrium angle is a good estimate of the energy level of the surface; the difference between advancing and receding angles, i.e. the hysteresis, is a measure of the chemical heterogeneity. The details of the procedures and devices used are described elsewhere (Della Volpe et al. 2002; Della Volpe et al. 2006) and all the calculation methods are fully described in Brugnara et al. (2006). The measurements were performed at room temperature (in the range 20–25 °C).

436

■

Plant Polysaccharides

Table 18.2 Water contact angles of PS and modified PS at room temperature. The last column shows results obtained using phosphate buffered saline as wetting liquid. See text for further explanation

Material PS untreated (PS, thin 0.6 mm) PS untreated (PS, thick, 1.2 mm) TCPS, commercial PS RF freshly oxidized, 48 h TCPS, self-produced oven-stabilized PS, recently aminated (NH2) PS aminated, old (NH2) Apple MHRB Apple MHRA Potato MHRP Apple MHRα Carrot MHRC

θadvancing (Tilt) 96.5 87.5 70.4 48.5 51.3

± ± ± ± ±

1.6 2.2 2.1 7.0 4.0

65.5 47.7 31.4 32.1 28.1 21.6

± ± ± ± ± ±

1.6 2.2 3.6 3.8 1.4 4.4

θequilibrium (VIECA) 87.9 76.5 50.3 14.8 26.2 44 33.6 19.6 16.1 14.9 11.9 11.7

± ± ± ± ± ± ± ± ± ± ± ±

0.6 1.9 3.2 2.1 4.2 4.0 3.3 1.9 1.5 1.5 1.7 1.1

θreceding (Tilt)

θequilibrium (VIECA)

81.0 73.1 13.3 7.7 12.8

± ± ± ± ±

1.7 2.2 5.0 5.0 3.1

27.3 9.7 5.9 8.8 5.1 2.8

± ± ± ± ± ±

2.6 9.0 3.8 4.6 5.0 1.2

85.9 74.3 50.0 20.9 24.3 39.7

± ± ± ± ± ±

1.0 1.3 2.3 0.7 3.9 2.4

12.9 10.0 12.5 15.1 19.3

± ± ± ± ±

0.7 1.2 1.2 1.7 1.6

The results of the contact angle measurements are shown in Table 18.2. The values are listed as decreasing equilibrium angles. The polystyrene samples show different values, even in the same lot, depending on the chemistry of the surfaces. It is well known in surface characterization that even a small amount of contamination affects contact angle values, and it is definitely difficult to find ‘real-world’ samples that show exactly the same surface chemistry when accidental contamination is taken into account. In the present case, polystyrene generally appears hydrophobic or strongly hydrophobic, as expected, but the initial samples of each lot experience a different thickness and thermal treatment. The oxidation of polystyrene by the application of RF oxygen plasma decreases the contact angles, as well known from the literature: plasma treatment of polystyrene is commonly adopted to make it suitable for cell culturing (yielding TCPS). This effect depends strongly on the details of the treatment, however, and thus commercial TCPS and the self-oxidized polystyrene are different. Moreover, as is again well known from the literature the contact angles tend to return to higher values with the exposure to air or to a hydrophobic environment, so that the time elapsed after the treatment may also be significant. The introduction of amino functions and most of all, MHR surface-linking, strongly decreases contact angles, inducing the formation of a stable hydrophilic surface. It is in fact possible to distinguish among different MHRs because of their different wettability. In some cases the equilibrium angles are very close, while the advancing and receding angles are different, indicating different distribution of chemical functions or a different degree of heterogeneity of the treated surface. All of them are, in general, hydrophilic, but MHRB is

Polysaccharide Nanobiotechnology

■

437

Figure 18.3 Relationship between equilibrium contact angle and fibroblast adhesion. Data points referring to MHRB and MHRα-modified surfaces are labelled in the graph; for other data points refer to Table 18.2.

less so than MHRα. Strong acid–base interactions between water molecules and the long, hydrophilic MHRα side chains could account for the observed difference. It is of interest to correlate the wettability results shown in Table 18.2 with the cell adhesion results. In Fig. 18.3 this relationship is shown for the adhesion of fibroblast cells, as reported by Nagel (2008). The graph shows the typical ‘Baier window’; that is, a maximum of cell adhesion at intermediate contact angles (Baier 1975). All the MHR-modified surfaces are on the extreme left of the graph and show different levels of resistance to cell adhesion, owing to their hydrophilicity. Surfaces in the central portion of the window show a good level of cell adhesion; only on the extreme right the hydrophobic surface of PS Bacteria (a bacteriological grade polystyrene from commercial Petri dishes; Greiner Bio-One, Courtaboeuf, France), surfacemodified by the manufacturer to enhance hydrophobicity and used in some laboratories for direct bacterial cell culturing, is again cell-resistant. It is worthy of note that not only the adhesion but also the metabolism of the fibroblasts is correspondingly modified. Similar behaviour is detectable for osteoblasts, as shown in Fig. 18.4. (M. Morra, personal communication) Once again all the data on the left of the graph correspond to the position of MHR derivatives and the rest of the data to different types of TCPS surfaces; once

438

■

Plant Polysaccharides

Figure 18.4 Relationship between equilirium contact angle and osteoblast adhesion. Data points referring to MHRB and MHRα-modified surfaces are labelled in the graph; for other data points refer to Table 18.2.

again there is an optimum of adhesion in the mean interval (the so-called Baier window) and it is quite evident that it is possible to finely tune the surface properties of the pectins used and thus the cell adhesion from the incompatibility of the MHRα derivative to the cell-compatible MHRB.

18.3 Surface modification of titanium dental implants by MHRs In the previous sections the potential of MHRs as interfacial layers on bonecontacting devices has been substantiated by cell studies and wettability measurements. Moving a step further towards the goal of MHR surfacemodified titanium dental implants, the process and materials previously outlined have been applied to titanium disks and fixtures, and the relevant surface chemistry has been evaluated by surface-sensitive techniques. 18.3.1 Surface modification of titanium by MHR: surface chemical analysis Of the several surface analysis techniques, X-ray photoelectron spectroscopy (XPS) is the most widely exploited for chemical characterization, and there

Polysaccharide Nanobiotechnology

■

439

is a huge literature on XPS analysis of titanium dental implants (Sawase et al. 1998, Morra et al. 2003, Serro & Saramago 2003). XPS measurements of MHR-modifed titanium have been performed with an AXIS ULTRA Spectrometer (KRATOS Analytical, UK) described in detail elsewhere (Bretagnol et al. 2006). Briefly, samples were mounted on the stainless steel sample holder using a UHV compatible (silicon-free) copper double-sided adhesive tape The base pressure of the instrument was better than 8 × 10−10 Torr and the operating pressure better than 3 × 10−9 Torr. The samples were irradiated with monochromatic AlKα X-rays (hν = 1486.6 eV). The area of analysis was 400 × 700 μm2 and take-off angles (TOA) of 90°, 45 ° and 20 ° with respect to the sample surface. Surface charging was compensate by means of a filament (I = 1.9A, 3.6V) inserted in a magnetic lens system and all spectra were corrected by setting the C1s hydrocarbon component to 285.00 eV. Instrument calibration was performed using a clean, pure gold/copper sample and a pure silver sample (99.99%). Measured values for electron binding energies (BE) were 84 ± 0.02 eV, and 932 ± 0.04 eV (Briggs & Seah 1992). For each sample, a survey spectrum (0–1150 eV), from which the surface chemical compositions (at%) were determined, was recorded at a pass energy of 160 eV. In addition one set of high-resolution spectra (PE = 20 eV) was also recorded. Total acquisition time was kept below 20 min to avoid any possible X-ray-induced damage. There was no modification of the C1s peak shape under X-ray irradiation, indicating constant charge stabilization with time and the absence of sample degradation during analysis. The data were processed using the Vision 2.1 software (Kratos Analytical, UK) and CasaXPS v2.3.10 (Surface Spectra, UK). Sample compositions were obtained from the survey spectra after linear background subtraction and using the relative sensitivity factors (RSF) included in the software derived from Scofield cross-sections. This method is estimated to give an accuracy of 10% in the measurement of elemental compositions. C1s envelopes were fitted with a Gaussian–Lorentzian function (G/L = 30) and variable full width half maximum. Curve fitting of C1s peaks was carried out using the same initial parameters and inter-peak constrains to reduce scattering. Surface compositions obtained from XPS data collected at 45 ° TOA of different samples are presented in Table 18.3, and the C1s core level spectra after different surface treatment are reported in Fig. 18.5. Compositions of the parent MHRs obtained from aqueous solutions cast on silicon wafer are also reported for comparison. The MHR parent materials are composed essentially of carbon and oxygen, as expected, with a small amount of nitrogen also present in both materials. The O/C ratio is about 0.52 for MHRα and 0.54 for MHRB. The presence of nitrogen is not expected, but two possible sources could be responsible for this contamination: residue from the enzymatic treatment or surface contamination during preparation of the samples. In fact, nitrogen contamination is usually found in polymeric surfaces prepared in a standard laboratory

440

■

Plant Polysaccharides

Table 18.3 Surface chemical compositions of different titanium surfaces at 45 °TOA (standard deviation) Elemental concentration (at%)

Sample

Ti

O

C

N

Na

Others (Si, K, S, Cl)

MHRα MHRB Ti Ti+Am Ti+Am+MHRα Ti+Am+MHRB Ti sterilized Ti+Am sterilized Ti+Am+MHRα sterilized Ti+Am+MHRβ sterilized

– – 17.5 (1.2) 6.0 (0.45) 1.01 (0.13) 1.59 (0.1) 16.4 (0.7) 8.53 (0.4) 2.6 (0.3)

33.6 (0.27) 34.0 (0.41) 45.98 (1.7) 28.1 (0.8) 19.0 (0.5) 18.6 (0.5) 41.8 (0.8) 26.0 (0.9) 23.66 (1.4)

64.9 (0.45) 63.2 (0.17) 32.9 (2.5) 57.7 (0.5) 74.3 (2.0) 75.5 (0.7) 40.0 (1.4) 56.7 (1.7) 68.47 (1.6)

1.1 (0.13) 2.2 (0.20) 1.42 (1) 7.6 (0.16) 4.23 (1.2) 4.0 (0.2) 0.9 (0.1) 7.6 (0.14) 4.8 (0.06)

0.4 (0.03) 0.6 (0.22) 2.4 (0.9) – – – 0.8 (0.01) 0.29 (0.01) 0.3 (0.12)

<1 <1 <1 <1 <1 <1 <0.1 <1 <1

1.76 (0.5)

22.53 (0.33) 69.5 (0.7)

5.29 (0.01) –

<1

Am, aminated.

environment; indeed, a similar amount of nitrogen was found on alginates samples prepared under the same conditions. Significant amount of carbon was detected on the titanium surface, as widely reported in the relevant literature (Sawase et al. 1998; Morra et al. 2003; Serro & Saramago 2003). High-energy titanium surfaces are prone to adsorb ubiquitous hydrocarbons from the atmosphere. These would go undetected by conventional analytical techniques, but they are readily captured by surface-sensitive techniques like XPS. Analysis of the titanium Ti2p C1s and O1s core level spectra gives more information on the surface chemistry of the bare titanium substrate. In particular, the titanium 2p peak shows a single doublet at about 458.6 eV (Ti2p3/2) and 464.3 eV (Ti2p1/2) that is typical of a titanium oxide (native titanium is readily oxidized to titanium dioxide on contact with the atmosphere; the surface oxide layer is 3–6 nm thick) (Wagner et al. 1979; Crist 1999). This is also supported by the analysis of the O1s peak, which shows a major peak at 530.3 eV that is attributable to the titanium dioxide while the smaller component at about 532 eV is attributable to surface contamination. This is also supported by the analysis of the C1s core level spectrum (Fig. 18.5c); besides the CC/CH peak at 285 eV there are other two minor components related to COX (286.5 eV) and C=O moieties (Beamson & Briggs 1992). The surface composition changes drastically after coating with allylamine, as illustrated in Table 18.3. The titanium, carbon and oxygen signals decrease, while the nitrogen content increases to about 8 at%, as usually detected in allylamine films deposited by plasma (Lejeune et al. 2006; Siow et al. 2006). However, it should be noticed that the titanium signal is still present in a significant amount (up to about 6 at%).

D

CPS

10 292

20

30

40

50

60

70

80

C=O

COX

C-N

C-C

288 286 284 Binding Energy (eV)

COOX

290

CC/CH

COX

288 286 284 Binding Energy (eV)

Pos. %Area 285.0463 55.24 286.0463 18.66 286.6503 14.32 288.0226 9.03 289.2463 2.74

290

%Area 12.95 61.89 19.00 6.16

C=O

Pos. 285.0046 286.5546 287.9046 289.3046

COOX

Name C-C C-N COX C=O COOX

x 102

292

10

20

30

40

50

Name CC/CH COX C=O COOX

x 102

282

282

B

E

292

10

20

30

40

50

60

Pos. 285.0121 285.9521 286.5121 288.1686 289.1121

CC/CH

CN

COX

288 286 284 Binding Energy (eV)

COOX

290

CC/CH

COX

%Area 43.05 3.92 38.96 10.04 4.03

COO/C=O

Name CC/CH CN COX COO/C=O COOX

x 102

%Area 27.90 50.08 18.58 3.45

288 286 284 Binding Energy (eV)

C=O

Pos. 284.9943 286.5443 288.0300 289.2943

290

Name CC/CH COX C=O COOX

x 102

292

5

10

15

20

25

30

35

40

45

50

55

282

282

C

F

290

CN

288 286 284 Binding Energy (eV)

COOX

COO/C=O

COX

CC/CH

Pos. %Area 285.0279 45.07 285.9679 6.00 286.5279 35.47 288.2100 10.20 289.1279 3.26

288 286 284 Binding Energy (eV)

Name CC/CH CN COX COO/C=O COOX

x 102

292

10

20

30

40

50

60

290

C=O/COOX COX

CC/CH

%Area Pos. Name 284.9996 81.99 CC/CH 286.5588 9.30 COX C=O/COOX 288.7781 8.70

x 102

292

10

20

30

40

50

60

282

282

■

Figure 18.5 C1s core level spectra: (a) MHRα, (b) MHRB, (c) bare cpTi, (d) cpTi after allylamine deposition, (e) aminated cpTi after MHRα coupling, (f) aminated cpTi after MHR-B coupling.

CPS

CPS CPS

CPS CPS

A

Polysaccharide Nanobiotechnology

441

442

■

Plant Polysaccharides

Taking into account that at 45 ° TOA, the sampling depth is in the order of 5–7 nm, these results suggest that the allylamine film is not covering the underlying titanium dioxide substrate uniformly; however, this is somehow expected because of the high roughness of the titanium substrate (>10 μm) with respect to the allylamine film thickness (∼10–20 nm). In fact analyses performed at both 20 ° and 90 ° TOA (corresponding to an analysis depth of about 3–4 nm or 12 nm, respectively) support this idea. Moreover, the amount of oxygen normally incorporated in plasma-deposited allylamine films is below 15%, whereas in the present case it is definitely much higher (Table 18.3). However, assuming a stoichiometric composition of the titanium oxide (TiO2), and subtracting the corresponding oxygen from the total, the remaining oxygen concentration is about 16 at%, very close to that expected in a continuous plasma-deposited allylamine film. Analysis of the C1s core level spectrum shows (Fig. 18.5d), besides components related to CC/CH and C-O/CON moieties, a new component at 285.9 eV that is attributed to the C-N amine bonding. The N1s high-resolution spectrum presents a single peak at about 400 eV that is also typical of C-N type bonding. Furthermore, the O1s core level spectrum further supports these results, indicating that the component related to titanium dioxide is decreasing, while that due to the polymeric moieties (∼533 eV) is strongly increasing. The amount of amine present on the surface after plasma deposition has been determined to be about 5% by chemical derivatization with trifluoromethyl benzaldeyde (TFBA) following a published procedure (Choukourov et al. 2003). Further changes in the surface compositions are observed after the covalent immobilization of MHRs. In particular, the titanium signal shows a further decrease to about 1 at% and also the nitrogen content is reduced by a factor of 2. This indicates the successful coupling of both MHRs. This is also confirmed by the analysis of the C1s (Fig. 18.5e,f) and O1s core level spectra. As can be seen, the C1s peaks show an increase of the component related to the COX moieties, while the C-N component is dramatically reduced. Moreover, the O1s peaks (not shown) show a further increase of the component at higher binding energy with a correspondent reduction of the component related to the titanium substrate. On the other hand, it should be noticed that the carbon content increases dramatically while the oxygen concentration shows a decrease, with an O/C ratio of about 0.25, much lower that that of the parent MHRs (Table 18.3). However, this can be explained by the fact that the MHR layer thicknesses are definitely below the XPS analysis depth, with a perceptible contribution from the allylamine films underneath. The sterilization of the sample by ethylene oxide does not induce major changes in composition, as can be seen from Table 18.3. In particular, a slight increase of the titanium signal and a decrease of the carbon content can be noticed. This may indicate that the sterilization process is probably removing some of the surface contamination (mainly hydrocarbons) and a minor amount of MHR that was probably only physisorbed on the surface.

Polysaccharide Nanobiotechnology

■

443

However, the nitrogen content does not change dramatically and the C1s and O1s core level spectra also indicate that the MHRs are present and quite stable on the aminated titanium surfaces. 18.3.2 Surface modification of titanium by MHRs: wettability measurements on machined and rough surfaces In the previous section it was shown that MHRs can be covalently linked to titanium surfaces, yielding nanoscale overlayers that present the MHR chemistry at the device surface. Among the interfacial properties that are affected, it is expected that wettability will be controlled by the MHR overlayer, following general structure–property relationships described in section 18.2.4. A particularly noteworthy point here is to investigate the relationship between surface chemistry and surface topography. Presentday titanium dental implants do not have a ‘flat’ (usually called ‘machined’) surface. Rather, the surface topography is quite rough, and it is obtained by a number of approaches. Acid-etched surfaces have been investigated in depth, and it has definitely been established that the details of roughness affect cell behaviour, promoting interfacial osteogenesis (Boyan et al. 2003). A typical example of an acid-etched surface obtained by the so-called double acid etching (DAE) process is shown in Fig. 18.6. The image shows an osteoblast-like cell growing in culture on the rough surface. Note that the distance between peaks is much lower that the cell- size, a key point in directing cell behaviour by surface topography (Zhao et al. 2007). According to general principles, the wettability of acid-etched titanium surfaces will depend on the interplay between surface chemistry and surface

Figure18.6 Osteoblast-like cell cultured on acid-etched titanium. Note that the distance between peaks is much lower than cell length, as discussed in the text.

444

■

Plant Polysaccharides

topography. As previously shown, titanium surfaces adsorb ubiquitous (and low surface energy) hydrocarbons from the atmosphere and tend to become hydrophobic over time (e.g. when stored on the shelf before being sold). Surface roughness, as described in basic textbooks, in general magnifies the effect of surface chemistry: water cannot penetrate surface cracks and crevices of hydrophobic surfaces, but it immediately fills hydrophilic pores (wicking). Rough titanium implant surfaces are generally very hydrophobic, water (or, in a clinical setting, blood) cannot easily displace air from the pores, yielding poor wetting and poor interaction, as described earlier (Rupp et al. 2006). It is expected that the hydrophilic MHR could act as a permanently wettable surface layer: its surface energy is much lower than that of ‘native’ titanium, i.e. it is not high enough to promote hydrocarbon adsorption. Yet it is much higher than that of typical hydrophobic hydrocarbons, so it provides improved wettability, and it is readily hydrated upon contact with water or aqueous solutions. Water contact angle measurements were performed, using the drop method previously described, on machined and DAE titanium disks, both uncoated and MHRB coated. The results confirm the expected trend. On machined surfaces, the detected contact angle is not significantly different: 32 ± 3 ° untreated, 27 ± 4 ° MHRB coated. Note that experiments were performed on carefully cleaned titanium discs, about 1 week after cleaning. The untreated titanium disc surface was still relatively clean, as shown by the low value of its equilibrium angle. This value will increase upon storage, in a process that could take months before reaching a plateau value. In contrast, no modification should occur on the MHRB-coated surface. DAE titanium disks give a clear perception of the interplay between chemistry and topography. This time the equilibrium angles do are different, as can be appreciated in Fig. 18.7. Detected values are 44 ± 3 ° uncoated and 34 ± 3 ° MHRB coated, with a statistically significant difference. Note that this difference is expected to be highly magnified over time: the advancing contact angle of the uncoated surface, which at the time of this experiment was about 73 ° (45 ° for the MHRB coated), is expected to exceed 100 ° when hydrocarbon adsorption reaches a surface coverage high enough to make the surface cracks and crevices unavailable to water, as reported in the literature (Rupp et al. 2006).

18.4 Reflections and conclusions A multidisciplinary effort made it possible to place engineered polysaccharides of plant origin in the hands of surface-modifiers of medical devices. Support from mammalian cell biologists and surface characterization suggested possible rules and links between MHRs molecular structure and cell behaviour. MHR-coated titanium dental implants were prepared as a proof of concept in the use of engineered plant polysaccharides in the surface

Polysaccharide Nanobiotechnology

■

445

A

B

Figure 18.7 Examples of the wettability of (A) acid-etched titanium; (B) MHRB-coated acid-etched titanium. The wetting liquid is ultrapure water. As discussed in the text, the detected angles are affected both by surface topography and surface chemical composition. The observed wetting angle is close to the equilibrium one (VIECA), as discussed in the text.

modification of medical devices. Even if it is too early to draw conclusions, and if the ultimate goal of finding and defining MHR molecular structures that specifically trigger bone regeneration mechanisms is still to be reached, some reflections are mandatory. The studies mentioned above definitely show that MHR surface layers, a few nanometres thick, covalently linked to solid surfaces, do affect cell behaviour;. Although this may seem obvious (every modification of surface chemistry affects cell behaviour in vitro), the possibility of precise enzymatic tailoring, or of in planta modification, of

446

■

Plant Polysaccharides

pectins and pectin hairy regions, opens the way to an almost unlimited reservoir of complex polysaccharides structures. Many of these, as reported in previous highly valuable work (Paulsen 2000), show bioactive properties and the ability to interact with healing mechanisms. Rather than describing conclusions, the previous sentences suggest the opening of a promising and far-reaching field. It is worth mentioning here that a preliminary trial was performed to assess the effect of MHR coating on titanium implant devices in vivo. Four-week implants of titanium screws in the rabbit tibia and femur, uncoated and coated with MHRB, MHRα and whole high-methoxy pectin were performed following standard protocols. The results show that MHRB coating yields statistically significant more interfacial bone growth as compared to MHRα and to whole pectin, in agreement with in vitro results. The take-home message for further thought is that surface monolayers of MHRs direct interfacial bone regeneration behaviour in vivo, i.e. healing mechanisms ‘sense’ the interfacial MHR layer. Although the mean value was higher, however, MHRB-coated titanium did not perform significantly better than uncoated titanium, when evaluated from a statistical point of view. A number of considerations can be suggested: for instance the test was performed using machined titanium implants, and not rough ones. The latter, because of both the enhanced surface area and the discussed interplay between chemistry and topography, could magnify the effect of the coating. Further studies using rough titanium surfaces are planned, but it is clear that the most fundamental issue is the ongoing definition, at the molecular level, of the interactions between MHRs and the cascade of healing mechanisms. Targeted and finely tuned modifications could then fully exploit the potential of engineered bioactive plant polysaccharides in the surface modification of medical devices.

Acknowledgments This work is the result of the cooperative effort of the whole PectiCoat consortium (www.pecticoat.org). The financial support of the EC through the EU STREP Project # 517036 is acknowledged.

References Anderson, J.M., Jones, J.A. (2007) Phenotypic dichotomies in the foreign body reaction. Biomaterials, 28, 5114–5120. ANSI (1985) Surface texture. ANSI/ASME B 46. American National Standards Institute/American Society of Mechanical Engineers, New York, NY. Baier, R.E. (ed.) (1975) Applied Chemistry at Protein Interfaces. Advanced Chemistry Series 145, American Chemical Society, Washington DC.

Polysaccharide Nanobiotechnology

■

447

Beamson, G., Briggs, D. (eds) (1992) The XPS of Polymers Database, pp. 246–250. Surface Spectra, IM Publications, Chichester. Boyan, B.D., Lossdörfer, S., Wang, L., Zhao, G., Lohmann, C.H. (2003) Osteoblasts generate an osteogenic microenvironment when grown on surfaces with rough microtopographies. European Cell and Materials, 24, 22–27. Branenmark, P.I., Hansson, B.O., Adell, R., et al. (1977) Osseointegrated implants in the treatment of the edentulous jaw. Experience from a ten year period. Scandinavian Journal of Plastic and Reconstructive Surgery, 11(Suppl. 16), 1–132. Bretagnol, F., Valsesia, A., Ceccone, G., et al. (2006), Surface functionalization and patterning techniques to design interafces for biomedical and biosensor applications. Plasma Process and Polymers, 3(6/7), 443–455. Briggs, D., Seah, M.P. (1992) Practical Surface Analysis, Vol. I. John Wiley & Sons, New York. Brugnara, M., Della Volpe, C., Maniglio, D., Siboni, S., Negri, M., Gaeti, N. (2006) Wettability of porous materials. I: the use of Wilhelmy experiments: the cases of stones, wood and nonwoven fabric. In: Contact Angle, Wettability and Adhesion, Vol. 4 (ed. K.L. Mittal), pp. 115–142. VSP, New York. Brunette, D.M., Tengvall, P., Textor, M., Thomsen, P. (eds) (2001) Titanium in Medicine. Springer, Berlin. Bussy, C., Verhoef, R., Haeger, A., Morra, M., Duval, J.L., Vigneron, P. (2008) Modulating in vitro bone cell and macrophage behavior by immobilized enzymatically tailored pectins. Journal of Biomedical Materials Research A, 86(3), 597–606. Choukourov, A., Biederman., H., Slavinska, D., et al. (2003) The influence of pulse parameters on film composition during pulsed plasma polymerization of diamocyclohexane. Surface and Coating Technology, 174–175, 863–866. Crist, B.V. (1999) Handbook of Monochromatic X-Ray Spectra, Vol. 1, p. 484; Vol. 2, p. 385. XPS International, Mountain View, CA. Cooper, L.F., Zhou, Y., Takebe, J., Guo, J., Abron, A., Holmen, A., Ellingsen, J.E. (2006), Fluoride modification effects on osteoblast behavior and bone formation at TiO2 grit-blasted c.p. titanium endosseous implants. Biomaterials, 27, 926–936. Davies, J.E. (ed.) (1991) The Bone-Biomaterial Interface. University of Toronto Press, Toronto. Davies, J.E. (ed.) (2000) Bone Engineering. Em squared, Toronto. Della Volpe, C., Brugnara, M., Maniglio, D., Siboni, S., Wangdu, T. (2006) About the possibility of experimentally measuring an equilibrium contact angle and its theoretical and practical consequences. In: Contact Angle, Wettability and Adhesion, Vol. 4 (ed. K.L. Mittal), pp. 79–100. VSP, New York. Della Volpe, C., Maniglio, D., Morra, M., Siboni, S. (2002) The determination of a ‘stable-equilibrium’ contact angle on heterogeneous and rough surfaces. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 206, 46–67. Della Volpe, C., Maniglio, D., Siboni, S., Morra, M. (2001) An experimental procedure to obtain the equilibrium contact angle from the Wilhelmy method. Oil and Gas Science and Technology – Revue de l’IFP, 56, (1), 9–22. Hoffman, A.S., Hubbell, J.A. (2004) Surface immobilized biomolecules. In: Biomaterials Science, an Introduction to Materials in Medicine, 2nd edition (eds B.D. Ratner, A.S. Hoffman, F.J. Schoen, J.E. Lemons), pp. 225–236. Elsevier, Amsterdam. Kokkonen, H.E., Ilvesaro, J.M., Morra, M., Schols, H.A., Tuukkanen, J. (2007) Effect of modified pectin molecules on the growth of bone cells. Biomacromolecules, 8, 509–515.

448

■

Plant Polysaccharides

Krzeminski, A., Marudova, M., Moffat, J., et al. (2006) Deposition of pectin/poly-Llysine multilayers with pectins of varying degrees of esterification. Biomacromolecules, 7, 498–506. Lejeune, M., Bretagnol, F., Ceccone, G. (2006) Microstructural evolution of allylamine polymerized plasma films. Surface and Coatings Technology, 200, 5902–5907. Marmur, A. (2006) Soft contact: measurement and interpretation of contact angles. Soft Matter, 2, 12–17. Marudova, M., Lang, S., Brownsey, G.J., Ring, S.G. (2005) Pectin-chitosan multilayer formation. Carbohydrate Research, 340, 2144–2149. Morra, M., Cassinelli, C. (1999) Non-fouling properties of polysaccharides coated surfaces. Journal of Biomaterials Science, Polymer Edition, 10, 1107–1124. Morra, M., Cassinelli, C., Cascardo, G., et al. (2003) Surface engineering of titanium by collagen immobilization. Surface characterisation and in vitro and in vivo studies. Biomaterials, 24, 4639–4654. Morra, M., Cassinelli, C., Cascardo, G., et al. (2004) Effects on interfacial properties and cell adhesion of surface modification by pectic hairy regions. Biomacromolecules, 5, 2094–2104. Morra, M. (2006) Biochemical modification of titanium surfaces: peptides and ECM proteins. European Cells and Materials, 12, 1–15. Morra, M. (2007) Biomolecular modification of implant surfaces. Expert Reviews of Medical Devices, 4, 361–372. Nagel, M.D. (2008) Enzymatically-tailored pectins differentially influence the morphology, adhesion, cell cycle progression and survival of fibroblasts. Biochimica et Biophysica Acta, 1780(7–8), 995–1003. Paulsen, B.S. (2000) Bioactive Carbohydrate Polymers. Kluver Academic Publishers, Dordrecht. Puleo, D.A., Kissling, R.A., Sheu, M.S. (2002) A technique to immobilize bioactive proteins, including bone morphogenetic protein-4 (BMP-4), on titanium alloy. Biomaterials, 23, 2079–2087. Puleo, D.A., Nanci, A. (1999) Understanding and controlling the bone-implant interface, Biomaterials, 20, 2311–2321. Ratner, B.D. (2001) Replacing and renewing: synthetic materials, biomimetics and tissue engineering in implant dentistry. Journal of Dental Education, 65, 1340–1347. Rupp, F., Scheideler, L., Olshanska, N., de Wild, M., Wieland, M., Geis-Gerstorfer, J. (2006), Enhancing surface free energy and hydrophilicity through chemical modification of microstructured titanium implant surfaces. Journal of Biomedical Materials Research, 76A, 323–334. Sawase, T., Hai, K., Yoshida, K., Baba, K., Hatada, R., Atsuda, M. (1998) Spectroscopic studies of three osseointegrated implants. Journal of Dentistry, 24(6), 119–124. Schols, H.A., Voragen, A.G., Colquhoun, I.J. (1994) Isolation and characterization of rhamnogalacturonan oligomers, liberated during degradation of pectic hairy regions by rhamnogalacturonase. Carbohydrate Research, 256, 97–111. Schols, H.A., Voragen, A.G.J. (1996) In: Progress in Biotechnology 14: Pectins and Pectinases, Proceedings of an International Symposium, Wageningen (NL), 1995, (eds J. Visser, A.G.J. Voragen), pp. 3–19. Elsevier, Amsterdam. Serro, A.P., Saramago, B. (2003) Influence of sterilization on the mineralization of titanium implants induced by incubation in various biological model fluids. Biomaterials, 24, 4749–4760.

Polysaccharide Nanobiotechnology

■

449

Siow, K.S., Britcher, L., Kumar, S., Griesser, H.J. (2006) Plasma methods for the generation of chemically reactive surface immobilization and cell colonization: a review. Plasma Processes and Polymers, 3(6/7), 392–418. Vogler, E.A. (2001) How water wets biomaterials surfaces. In: Water in Biomaterials Surface Science (ed. M. Morra), pp. 269–290, John Wiley & Sons, Chichester. Wagner, C.D., Riggs, W.M., Davis L.E. (eds) (1979) Handbook of Photoelectron Spectroscopy, p. 69. Perkin-Elmer Corporation, Eden Prairie, MN. Yasuda, H. (1985) Plasma Polymerization. Academic Press, New York. Zhao, G., Raines, A.L., Wieland, M., Schwartz, Z., Boyan, B.D. (2007) Requirement for both micron- and submicron scale structure for synergistic responses of osteoblasts to substrate surface energy and topography. Biomaterials, 28, 2821–2829.

Annual Plant Reviews (2011) 41, 451–464 doi: 10.1002/9781444391015.index

INDEX

1,2-cis glycoside, 72 1,2-trans glycoside, 73 ABA see absisic acid absisic acid, 290, 297–298 AceA see aceric acid Acer, 14–15 aceric acid, 82–85 acetylated, 17–18, 33, 87 acetylglucosaminyltransferase, 411 acetyltransferase, 394 acid-etched, 443, 445 Acidothermus, 377 activator, 73, 124 activator cellobiose and laminaribiose, 143 ADP-Glc see ADP-glucose ADP-glucose, 197, 391, 393 AFM see atomic force microscopy AGP see arabinogalactan protein alga, 268, 286, 306, 315 alginate, 440 Alismatales, 21 Alkyl (aryl) thioglycoside, 73 alpha-glucan, 404

amylase, 32 amylopectin, 197, 198, 390–394, 402 amylose, 5, 197, 390–393, 402 angiosperms, 24, 268–269, 314–315, 334 Api, 6, 82 apiofuranose, 82–83 apiogalacturonan, 6, 21 apoptosis, 334 apple, 370, 433 Ara see arabinose Arabidopsis gene/mutant ARAD1 269, 275–276 AtFUT1, 239 AtFUT10, 238–239 AtFUT2, 238–239 AtFUT3, 239 AtFUT4, 239 AtFUT5, 239 AtFUT6, 239 AtFUT7, 239 AtFUT8, 239 AtGUT1, 272–274, 278 AtPGSIP1–7, 198, 200–201 AtXXT1, 243–245 EPC, 286–291, 293, 295, 296–298 epc1-3/epo, 291 epc1–2, 290–291, 296–298 epc1–3/epo, 291, 293, 295

Annual Plant Reviews Volume 41: Plant Polysaccharides, Biosynthesis and Bioengineering © 2011 Blackwell Publishing Ltd. ISBN: 978-1-405-18172-3

Edited by P. Ulvskov

451

452

■

INDEX

epc1/epo, 291 epcl1, 288 fra8, 274 FRA8, 274 glz1, 195 irx14, 255 IRX8, 274 irx9, 255 IRX9, 255 Kor, 375 mur1, 28, 418 mur2, 238 mur3, 269 Mur3, 270 mur3, 271 mur3, 271 MUR3, 278 MUR3, 269, 271–272 parvus, 195 POM1, 291 qua1-1, 190 qua1, 190 qua2, 354, 375 rgxt1, 312 RGXT1, 312 rgxt2, 312 RGXT2, 312 RGXT3, 312, 314 rsw1, 123, 375 XGD1, 269, 276–277 xgd1, 277 arabinan, 54, 67, 103, 269, 275–277, 316, 345, 351, 354, 357–358, 368, 374, 434 see also Table 5.2, 6.1, 7.1, 11.1, 12,1 and individual gene families arabino-oligosaccharides, 277 arabinanase, 275, 374, 378 arabinofuranosyltransferase, 267, 277 arabinogalactan protein, 53, 214–215, 225, 228, 240–241, 293, 295, 297–298, 322–323, 325–326, 329, 330, 333, 355, 358–359

AG-glycomodules, 323 AGP30, 297 AGPase, 397 arabinoglycomodules, 325 AtAGP18, 297 fasciclin, 298 LeAGP-1, 325 arabinogalactan, 18, 53, 56, 67, 86, 235, 240, 246, 297–298, 322, 325–326, 329 arabinogalactosylation, 323, 325, 329 arabinose, 12, 15–16, 18–21, 88, 108, 191, 215, 246, 271, 275, 277, 297, 315–317, 323, 325–326, 434 Araf, 32, 68, 87, 215 Arap, 12, 72, 88 arabinoside, 317 arabinosyltransferase, 275, 277, 316–317 arabinoxylan, 103, 268, 379 arabinoxylan–cellulose, 353 ARAD1, 269, 275–277 ARAMEMNON, 278 AtFUT see fucosyltransferase AtGUT1-2, 273–274, 278 atomic force microscopy, 349–350, 431–432 AtPGSIP1-7, 198, 200–201 AtXXT1, 243–245 AtXXT2, 243–245 AtXXT5, 243–245 autoglycosyltransferase, 315 autoradiography, 17, 34–35 auxin, 198, 290 barley, 368 biocompatibility, 396, 426, 427 bioethanol see biofuel biofuel, 151, 368, 380–381 post-harvest modification, 378 self-processing plants, 380

INDEX

biomass see biofuel bone implant, 287, 426–427, 431, 434, 438 Brachypodium, 173, 175, 177, 179, 181, 183 GT8, 170 Brassica heterologous expression in, 378 XET, 375 brassinosteroids, 286 broiler see chicken bryophyte, 14, 22 Caenorhabditis, 111, 132, 226, 252, 266, 286 calcium, 7, 29, 147, 353 callose, 10, 48, 119, 137, 141, 147, 150–152, 157, 168, 180–182, 185, 190–191, 194, 195, 197–200, 203, 316, 328, 331, 381 see also glucan synthase-like callose synthase HvGSL1, 133, 144 OsGsL2, 145 carbohydrate binding module CBM15, 348 xyloglucan-directed, 347 CAZy, 94, 214, 222, 241, 266, 286 cellularization in cereal grain, 119 cellulose, 19, 22, 118, 124–125, 142, 150, 368, 376, 380 celA, 121, 125 celABC, 125 celB, 125 CelB, 125 celC, 125 celD, 125 CelD, 125 celDE, 125 celE, 125 see also endo-glucanase

■

453

cellulose, 16, 20, 117, 124, 142, 150, 351, 352–354 definition, 2 radiolabeled, 17 cellulose synthase, 132, 150 AtCesA7, 149 CesA, 122, 124–127, 133, 142, 148, 150, 201, 306 CESA, 201 CesA2, 150 CESA4, 273 CESA7, 273 CESA8, 273 HvCesA1, 122 HvCesA2, 122, 133 HvCesA4, 122 HvCesA6, 122 HvCesA7, 122 HvCesA8, 122 Cellulose synthase-like Csl, 125, 127–128 CslA, 125, 126–127 CslAs, 125, 126 CslB, 126, 127 CslC, 126, 127, 245 CslD, 126, 127 CslE, 126, 127 CslF, 126, 128–129 Cslf, 129 CslF, 130, 133, 142, 150 CslFgene, 127 CslG, 126, 127 CslH, 126–128, 130 CslJ, 126–127 AtCslA15, 126 AtCslB5, 126 AtCslD, 127 AtCslD2, 126 AtCslD3, 126 AtCslD4, 126 AtCslD5, 126 HvCslF, 133 HvCslF9, 133 OsCslA4, 126

454

■

INDEX

OsCslA9, 126 OsCslC2, 126 OsCslC9, 126 OsCslD2, 126 OsCslD3, 126 OsCslF, 128–130 OsCslF1, 126, 129 OsCslF2, 126, 129 OsCslF3, 126, 129 OsCslF4, 129 OsCslF7, 126 OsCslF8, 129 OsCslF9, 129 OsCslH1, 126 OsCslH2, 126 charophyte, 21, 318 chicken, 291, 379 chicory, 398, 401 chitin, 141 chitinase, 292–293 chito-oligomer, 132 chitobiose, 412 chitosaccharide, 141 chitosan, 430 Chlamydomonas, 268, 306–307, 313, 315–317, 329 chlorophytes, 313, 336 chondrocytes, 256 chondroitin, 256 Coleochaete, 22, 329 collagen, 322, 334, 431 CoMPP see glycoarray conifer, 18, 21, 24 convergent block synthesis, 74, 76 core protein, 256 corn, 122, 132, 152, 217, 224 cross-link, 3, 12, 25, 26, 27, 28–29, 36, 272, 306, 322–323, 329–330, 333–334, 345, 350–354, 356–357, 374–375 cucumber, 202, 204, 376 Cyanidales, 306 Cyanidioschyzon, 306–307, 313

d-lyxo-5-hexosulopyranuronic, 26 dental, 396, 426–428, 434, 438–439, 443–444 deoxyhexose, 5, 29 dextran, 394, 396, 400 dextransucrase, 396 dha, 5, 82 diatoms, 119 diisodityrosine, 322, 329 dinoflagellate, 113, 116 direct glycosylation, 75 dityrosine see diisotyrosine DmGlcAT-BSI, 257 DmGlcAT-BSII, 257 DmGlcAT-I, 257 Driselase, 7, 12, 14–17, 21–24, 28, 32–33 DUF604, 214, 226, 228 DVD see DXD DXD, 138, 221, 222, 228, 258, 288, 307, 312, 315 ectopically parting cells see EPC electron-withdrawing group, 74, 75 ELISA, 55 embryophyte, 115, 121, 313 endo-glucanase, 13, 46, 238, 293, 374, 377–378 XEG, 13–15, 19, 21, 24, 33–34 XTH, 26 endo-mannanase, 15, 352 endo-polygalacturonase, 7, 67, 98, 356, 370, 373–374 xylogalacturonan-specific, 377 endo-transglucosylase, 19, 24, 375 XET, 26, 352, 355, 375 endo-xylanase, 15 endochitinase, 353 EPC-L1, 287–290, 295 EPC-L2, 285–290, 295, 296 EPC1, 285–291, 293, 295, 297, 298

INDEX

EPC1 GFP, 293 epc1-3/epo, 291 epc1–2, 290–291, 296–298 epc1–3/epo, 291, 293, 295 epc1/epo, 291 epcl1, 288 epcL2, 298 epibrassinolide, 290 epo/epc1–3, 291, 293 Erwinia, 399, 401 ethylene, 290 eudicots, 21 euglenids, 116 euglenoid, 119 euphyllophytes, 23 eusporangiate, 24 exostoses, 287 exostosin, 266, 271, 287 EXT1, 287 EXT2, 287–288 exostosin-like, 291 EXT-L2, 287 EXT-LIKE, 287 expansin, 352, 353, 376 extensibility, 24, 353, 356 extensin, 6, 315–317, 323, 329–330, 333–334, 354 extensin-like, 315, 334 fenugreek, 242, 246 fern, 24, 330, 334 ferulic acid, 19, 21, 24, 27, 32, 334 Fer-Ara-Xyl, 11 Fer-Ara, 11, 19, 32 feruloyl-oligosaccharides, 19 fibroblast, 437 fibrointegration, 426 fibronectin, 432 FLA see arabinogalactan protein flax, 314, 359, 370, 373, 374 FRA8, 274, 278 fringe, 216, 228

■

455

fructan, 398–402 accumulating, 401 producing, 399 storing, 398 fructose, 391–401 fructosyltransferase, 399–401 FTIR, 51–52, 57 Fuc see fucose fucose, 4, 12–13, 15–16, 18, 20, 29, 31, 47, 88, 222, 235–238, 240–241, 246, 269, 271, 306, 312, 358, 411–413, 417–418 Fuc-Xyl, 118 fucosyltransferase, 235–238, 240, 241–242, 243–246, 411 AtFUT1, 239–241 AtFUT10, 238–239 AtFUT2, 238, 239 AtFUT3, 239 AtFUT4, 239 AtFUT5, 239 AtFUT6, 239 AtFUT7, 239 AtFUT8, 239 PsFUT1, 239 PtxPaFUT1, 239 PtxPaFUT2, 239 PtxPaFUT4, 239 PtxPaFUT5, 239 PtxPaFUT7, 239 FUT1, 236, 240, 246 FUT3, 239 FUT4, 241 FUT6, 239, 241 FUT see fucosyltransferase Gal see galactose GalA see galacturonic acid galactan, 53–54, 67, 215, 297, 345, 351, 354, 357–359, 368, 370, 373

456

■

INDEX

galactanase, 373–374 galactinol, 168–169, 202, 204 galactinolsynthase AtGolS1, 204 galacto-oligosaccharide, 77 galactodecaoside, 78 galactododecaoside, 76 galactoglucomannan, 15, 117, 359 galactokinase, 33 galactolipid, 214 galactomannan, 15, 103, 214, 236, 241–246 galactose, 4 enantiomers, 33 enzyme resistant, 68 feruloyl and 4-coumaroyl ester groups, 21 galactinol, 168–169, 202, 204 galactomannan, 15, 103, 214, 235–236, 241–246 in LPS, 169 in mannan, 15–16 in skp1, 312 in xylan, 19 l and d-Gal, 19, 29 l-Gal and 2-O-Me-l-Gal in RG-II, 28 raffinose, 169 structure, 4 xyloglucan, 14, 271 galactosyltransferase, 202, 214–216, 224–226, 235, 242–243, 245–246, 306–307, 312, 316, 318, 417 AGP, 225 AgtA, 312 galactinol, 169 GalAT see galacturonosyltransferase LPS, 169 raffinose, 169 Skp1, 312, 318 galacturonic acid, 4, 6, 7, 12, 74, 87–88 galacturonic acid methyl ester, 7, 28

galacturonosyltransferase, 167, 168–169, 173, 175, 177, 186, 188–191, 193, 196 AtGAUT1, 187–188, 190 AtGAUT10/AT2G208101, 187 AtGAUT11/AT1G185801, 187 AtGAUT12, 187 AtGAUT2, 187 AtGAUT2/AT2G464801, 187 AtGAUT7, 190 AtGAUT7/AT2G386501, 187 galacturonosyltransferase-like, 167, 177, 179, 186, 188, 193–195, 196 AtGATL, 194 AtGATL1, 194 AtGATL1/AT1G193001, 193 AtGATL10, 193 AtGATL10/AT3G283401, 193 AtGATL2/AT3G507601, 193 AtGATL3/AT1G132501, 193 AtGATL4/AT3G062601, 193 AtGATL5, 193–194, 196 AtGATL5/AT1G027201, 193 AtGATL6, 193–194, 196 AtGATL6/AT4G021301, 193 AtGATL7, 193 AtGATL7/AT3G626601, 193 AtGATL8, 193 AtGATL8/AT1G241701, 193 AtGATL9, 193, 196 PdGATL1.1 and PdGATL1.2, 193, 194 Galdieria, 306, 313 galectin, 214, 221–225, 428 GalT see galactosyltransferase GATL see galacturonosyltransferase-like GAUT see galacturonosyltransferase GDP-fucose, 246 GDP-glucose, 126 GDP-Man-dehydratase, 418

INDEX

GDP-mannose, 126 GlcA see glucuronic acid GlcAT see glucuronic acid transferase GlcNAc see N-acetyl-glucosamine GlcNAc-TII, 296 glucan see mixed-linkage glucan, hemicellulose, xyloglucan, cellulose or callose glucan synthase GSL, 144, 146, 147, 148, 151 GSL1, 147 glucan synthase-like AtGSL1, 145 AtGsl1, 146 AtGSL10, 145 AtGsl10, 146 AtGSL11, 145 AtGsl11, 146 AtGSL12, 145 AtGsl12, 146 AtGSL2, 145 AtGsl2, 146 AtGSL2, 146 AtGSL3, 146 AtGSL4, 146 AtGSL5, 146 AtGSL6, 146 AtGSL7, 146 AtGSL8, 146 AtGSL9, 146 AtGSL10, 146 AtGSL11, 146 AtGSL12, 146 OsGSL1, 145 AtGSL3, 145 AtGsl3, 146 AtGSL4, 145 AtGsl4, 146 AtGSL5, 144, 145 AtGsl5, 146 AtGSL6, 145

■

457

AtGsl6, 146 AtGSL6, 147 AtGSL7, 145 AtGsl7, 146 AtGSL8, 145 AtGsl8, 146 AtGSL9, 145 AtGsl9, 146 definition, 143 glucanase, 378, 379 see also endo-glucanase glucansucrase, 394–395 glucitol, 14 glucomannan, 15, 17 Gluconacetobacter, 121, 137, 141, 142 glucosamine, 27 glucose, 4, 13, 14, 15, 20, 21 Glc-1-P, 391 radiolabeled, 17, 29, 31 glucosidase, 103 glucosylceramide, 132 glucosyltransferase, 169 glucuronic acid, 4, 12, 15, 16, 19, 88, 272, 274, 296 animal GT43, 253 charophyte, 21 deoxy, 4 electrophoresis, 35 exostosin, 271 FRA8, 274 glucuronomannan, 16, 32 GT2, 131, 139 GT31, 215 GT43, 253 GT64, 287 MeGlcA, 4, 15, 19 moss, 22 poales, 19 radiolabeled, 310 RGII, 12 xylan, 15, 32 glucuronidase, 16 glucuronoarabinoxylans, 15 glucuronomannan, 15, 16, 32

458

■

INDEX

glucuronosyltransferase, 253, 256–257, 259 GlcAT-I, 254, 256–259 GlcAT-P, 256–259 GlcAT-S, 256–259 GT47, 272 HsGlcAT-I, 252, 254 HsGlcAT-P, 254 HsGlcAT-P, 254 HsGlcAT-S, 254 DmGlcAT-BSI, 252 HsGlcAT-S, 254 xylan-specific, 274 glucuronoxylan, 191, 253, 272–274 glucuronyltransferase see glucuronosyltransferase glyco-array CoMPP, 45, 52–54, 357 glycogenin-like starch initiation protein, 168, 186, 188, 198, 200–201 PGSIP-A, 188, 198, 200 PGsiP-B, 183 PGSIP-B, 198, 200 PGsiP-B, 200 PGSIP-C, 198, 200 PGSIP-C, 201 PGSIP1, 201 glycosyl bromide, 73 donor, 72–76 fluoride, 73, 76 trichloroacetimidate, 73 glycosyl acceptor, 72, 74, 82 glycosylation direct, 75 reaction, 75 glycosylhydrolase families GH1, 103 GH13, 104 GH61, 103 glycosyltransferase families GT1, 103 GT2, 102, 109–111, 113, 115, 117, 119, 121–122, 125–127, 131–134, 137–141, 143, 146–151, 223, 306

GT3, 246 GT4, 102, 246 GT5, 102 GT6, 246 GT7, 246 GT8, 168–170, 173, 175, 177, 179, 181, 183, 185–186, 188–189, 191, 193, 195–198, 200–202, 204, 306 GT10, 102, 241 GT13, 102 GT14, 102 GT16, 102 GT18, 271 GT22, 102 GT24, 102 GT31, 102, 214–215 GT32, 102 GT33, 102 GT34, 234–237, 241–246, 306 GT37, 102, 235–238, 240–241, 243, 245–246 GT41, 102 GT43, 103, 252–259 GT47, 103, 259, 266–271, 273–276, 278, 287, 289, 306, 358 GT48, 109–111, 113, 115, 117, 119, 121, 127, 130–131, 133,137, 139, 141–149, 151 GT50, 102 GT57, 102 GT58, 102 GT59, 102 GT60, 318 GT61, 102 GT64, 266, 286–291, 293, 295–297 GT65, 102 GT66, 102 GT68, 102 GT75, 315 GT76, 102 GT77, 306, 307, 312–313, 315–318 glycosyltransferase folds GT-A, 134, 169, 221–222, 228, 257–258, 288, 318 GT-B, 221 Gnetophyte, 24

INDEX

GnT see Nacetylglucosaminyltransferase GolS see galactinol synthase GPI-anchor, 102, 323 GSL see glucan synthase like guar, 125 gum arabic, 323 GUT1-2, 273, 274, 278 hemicellulose, 3, 5, 12–20, 46, 253, 256, 266, 274, 344–345, 350–352, 354, 359, 370, 374, 390 definition, 3 hemp, 357, 359 heparan, 140, 266, 287, 296 heparin, 256, 296 heparosan, 131, 139–140 hexuronyltransferase, 93 hormone, 290, 298 indole-3-acetic acid, 286 hornwort, 22–23 horsetail, 24, 345 HPRG see hydroxyproline rich glycoprotein hyaluronan, 110, 131, 132, 133, 138–142 hydroxyproline, 225, 322, 323, 325, 326, 329, 330 Hyp-Ara2, 317 Hyp-Ara3, 317 hyp-Ara4, 316 Hyp-arabinogalactan, 329 Hyp-arabinogalactans, 322 Hyp-glycosylation, 325 hydroxyproline rich glycoprotein (HRGP), 215, 322–323, 325, 330 immunotherapy, 415 implant, 426–428, 430–431, 434, 438–439, 443–444 inflammatory, 427 inulin, 399–401

inulosucrase, 400 invertase, 369, 391 iregular xylem IRX10-L, 259 IRX10, 259 irx10/irx10-L, 259 IRX10/IRX10-L, 259 IRX14, 253, 255–256 irx14, 255 irx14, 259 IRX14, 259 irx2–2, 293 irx8, 274 IRX9-GFP, 256 IRX9, 253, 255–256 irx9, 255 irx9, 256 IRX9, 259 irx9, 259, 274 IRX9, 256 IRX see irregular xylem isodityrosine see diisodityrosine JIM13, 53 JIM5, 55 JIM7, 55 kam1–1, 271–272 KDEL, 415 Kdo, 4 KdoGalA, 12 ketohexoses, 33 ketoses, 4 KOR see korrigan korrigan KOR1-GFP, 293 kor1–1, 293, 295 KOR1, 293, 295 Lactobacillus, 400 Lactuca, 291 Lamiales, 21 laminaribiose, 24, 143

■

459

460

■

INDEX

laminarinase, 11 late stage oxidation, 75, 87 leaving groups, 72, 76 manganese, 138 lectin, 215, 222, 224–226, 322 galactoside-binding, 225 Leuconostoc, 396, 400 levan, 397–401 levansucrase, 400, 401 lichen, 116 lichenin, 117, 120 liverwort, 22–23, 329 LLGG, 21 Lolium, 137, 142, 398 Lotus, 243, 246 LSGG, 21 LTGG, 21 lycophytes see lycopodiophyte lycopodiophyte, 23, 313–314 magnesium, 134 MALDI-TOF, 47, 48, 55 manganese, 134, 138, 201, 258, 288 mannanase, 11 mannose, 4, 15–16, 108, 410–411, 413, 415, 417 radiolabeled, 31, 310 mannosidase, 410, 411, 417 Medicago, 224, 266–268, 317, 334 methyl-esterase, 7, 353, 356 methyl-esterification, 16, 68, 78, 80, 353, 358 methyl-fucose, 4, 12, 23, 29 methyl-xylose, 5, 12, 23 microalga, 313 microtubule, 56, 118, 124, 150 mixed-linkage glucan, 19 food and health, 151 MLG-xyloglucan, 26 occurrence, 19, 116 structure, 120 synthesis, 127–130 MLG see mixed-linkage glucan

monoclonal antibody CCRC-M7, 297 CCRCM1, 358 JIM13, 53 JIM5, 55 JIM7, 55 LM1, 315 LM10, 273 LM11, 348, 357 LM15, 348 LM16, 357 LM5, 53, 291, 297, 357–359 LM6, 53, 357, 359 LM7, 356–357 LM8, 277 moss, 22–23, 54, 116, 126, 268–269, 286, 314, 316, 329, 352, 418 mucilage, 15, 119, 191, 195–196 mucins, 225 mutan, 396–397, 400 mutansucrase, 396–397 mutase, 315 myo-inositol, 169, 204 N-acetyl-glucosamine, 141, 168, 216, 225, 293, 296, 410, 412, 413, 417 N-acetyl-lactosamine, 225 N-acetylgalactosaminyltransferase, 225, 287, 296 N-acetylglucosamine, 412 N-acetylglucosaminyltransferase, 93, 286, 291, 410 GnT-III, 415 GnT, 411–412, 417 GnT51, 318 N-acetylneuraminic acid, 411 N-glycan, 71, 102, 214–216, 225, 240, 410–415, 417–418 neu5acT, 411 N-glycolyl-neuraminic acids, 418 N-glycosylation, 102, 410, 412–415 core-fucose, 415 core-xylose, 412, 413–415, 417, 418

INDEX

n-Pentenyl glycoside, 73 Nasturtium, 127 non-participating group, 73 nucleotide sugar, 131, 288, 307 nucleotide-binding, 134, 137 O-acetylated glucomannan, 17 methylglucuronoxylan, 17 mixed-linked beta-glucan, 32 xyloglucan, 28 O-galacturonoyl–sugar, 27 O-glycan-related, 102 oak, 18 oat, 117 oligogalacturonide, 49, 68, 75–76, 78, 80, 277, 374 oligoglucantransferase, 125 orthogonal protecting groups, 78 osmoprotectant, 169 osteoblast, 431, 434, 437–438, 443 osteoclast, 431, 433 osteogenesis, 443 Ostreococcus, 286, 306, 315, 318 oxalate, 6, 22 PACE of oligosaccharides, 49 palatinose, 398 pea, 204, 237, 240, 243, 246, 277 pectin, 3, 5–7, 12, 16–18, 20–21, 22, 25, 53, 55, 71, 74, 334, 344–345, 347–351, 358, 370, 373–375, 390 AGPs and plascticisers, 330 AGPs and porosity, 329 ARAD, 275 arad1, 275 ARAD1, 277 CAZy, 102 coating, 426–428, 430–434, 438 COMPP, 52 definition, 2 electrophoresis, 49 enzymatic cleavage, 33, 46 ferulic acid in, 27

■

461

GT47, 266, 267 GT8, 168, 188–191, 193, 195–196 in secondary wall, 359 in wall model, 333, 353 LM5 and LM6, 359 LM7, 356 metabolism, 98 RGXT, 312 sources of oligosaccharides, 66 supramolecular interactions, 354 synthesis of metylated, 78–79 XGD1, 276, 277 pectinase, 428 pectinmethylesterase, 27 methylesterase-like, 27 phloem, 11, 119, 359, 391 Physcomitrella, 22, 54, 126, 252, 255, 268–270, 286, 307, 313–317, 416 Poa, 399 Poaceae, 116, 126–127, 334, 399 Poales, 11, 18–21, 24, 27, 116, 345 polyacrylamide, 45, 47, 49 polygalacturonase see endo-polygalacturonase POM1, 291, 293, 295 potato, 374, 396 prebiotic, 368, 375, 396, 400 prediction server TMHMM, 170, 193, 278 ARAMEMNON, 278 proline, 124, 215, 315, 318, 323 proline-rich proteins, 322–323, 325, 329 PRP see proline-rich proteins Psilotum, 24 pyrophosphorylase, 391–393 qua1-1, 190 QXXRW, 137–138 radish, 246 raffinose, 169, 186, 188, 196, 201, 202 regioselectivity, 74 reiterative synthesis, 78

462

■

INDEX

RG-II xylosyltransferase, 316 rgxt1, 312 RGXT1, 312 rgxt2, 312 RGXT2, 312 RGXT3, 312, 314 RGXT see RG-II xylosyltransferase rhamnogalacturonan, 6, 46, 103, 191, 196, 240–241, 370, 373, 375, 433 RG-I, 7, 18, 20, 25–26, 32, 33, 66–69, 86–88, 98, 275, 277, 296–297, 345, 353–354, 357–359, 426, 428 RG-II, 5, 7, 12, 20, 23–25, 28, 33, 66–67, 69, 71, 81–85, 98, 272–274, 312, 314–315, 345, 353–354, 357, 418 rhamnogalacturonan lyase, 68 rhamnogalacturonase, 68, 371, 373–374 rhamnopyranose, 68, 82, 87 rhamnose, 4, 6, 12, 17–18, 21, 23–24, 31, 33, 67–68, 84, 88, 191, 195, 296, 326 MeRha, 4, 24 rhamnosyltransferase, 275 Rhodothermus, 347 RRA1, 315, 316 RRA2, 315 rsh, 330 RSH, 330, 333 rsw1, 375 rsw2, 375 rsw2-1, 293 rsw2-3, 293, 295 Rubus, 116, 120 SBD see starch binding domain SBE see starch branding enzymes Ser-Hyp3-4, 323 Ser-Hyp4, 322–323, 333 sialic acid, 215, 412, 414–416, 418–419

sialyltransferase, 416, 419 signalling, 287, 296, 298, 325, 427 siliques, 277, 357 software tools see also prediction servers MEGA4, 238, 244 soyabean, 9 Sphagnum, 22, 26 spinach, 21 Spirodela, 6, 21 sporopollenin, 3 SSGG XG-sequence, 21 SSGGG XG-sequence, 21 stachyose, 169, 186, 202 starch, 16, 32, 102, 152, 168, 186, 188, 197–198, 200–201, 390–394, 396–398, 401–402 starch binding domain SBD, 397, 396 starch branching enzymes, 392, 393 starch synthase granule-bound, 391, 397 SSI, 391 SSII, 391, 393 SSIII, 391 SSIV, 391 stereoselectivity, 74, 76 sugar-beet, 7 SXGG XG-sequence, 21 SXGGG XG-sequence, 21 sycamore, 6 titanium, 426–428, 430–431, 434, 438–439, 442–445 tobacco, 127, 242, 325–326, 334, 348, 351, 370, 373–378, 398, 401, 414–415, 416, 419 tomato, 324 tragacanth, 12 transacylation, 27 transglycosylase, 352

INDEX

UDP magnesium, 134, 138 UDP-Araf, 277 UDP-Gal, 169, 204, 225, 242, 246 UDP-GalA, 190 UDP-galactose2acetamido-2-deoxy-dglucose3galactosyltransferase, 225 UDP-galactosefucoside, 312 UDP-GalGlcNAc, 225 UDP-Glc, 110, 120–121, 125, 131, 133–134, 137, 143, 147, 169, 197, 201, 245, 259, 391 UDP-Glc binding of, 124, 134 UDP-Glc radiolabeled, 145 UDP-GlcA, 139, 254, 256, 259 UDP-Glclipid, 125 UDP-GlcNAc, 141, 216 UDP-GlcpA, 140 UDP-GlcpNAc, 140–141 UDP-Man, 259 UDP-Xyl, 243–246, 256 unmasking of epitopes, 347 verbacose, 202 XEG see xyloglucanase XGA see xylogalacturonan xgd1, 277 XGD1, 278 XGD1, 269, 276–277 XGGG XG-sequence, 19 XLFG XG-sequence, 14, 21, 24, 47, 237 XLFG + Ac XG-sequence, 47 XLGGG XG-sequence, 21 XLLG XG-sequence, 15, 21, 24

■

463

XLXG XG-sequence, 14, 15 XPS, 439, 442 XSGG XG-sequence, 21 XSGGG XG-sequence, 21 XTH, 26 XXFG XG-sequence, 14, 21–24 XXGG XG-sequence, 19, 21, 236 XXGGG XG-sequence, 19, 21 XXGGGG XG-sequence, 19 XXLG XG-sequence, 14–15, 21, 24 XXSG XG-sequence, 21 XXT, 244–246 XXXG XG-sequence, 14–15, 21, 24, 47, 271 XXXG-repeat XG-sequence, 236 XXXG-type XG-sequence, 21, 236 XXXGol XG-sequence, 24 xylan, 10, 13, 15, 17–18, 32, 46, 53, 190–191, 193, 195, 253, 255–256, 259, 273–274, 345, 347–348, 352–353, 357, 359 acetylation, 15, 351 cleaving Araf, 19, 32 PACE, 49 structure, 15 taxonomy, 351 xylobiose from, 33 xylanase, 11, 374, 377–380 xylem, 6, 16, 18, 118, 190, 195, 253, 255, 344, 348, 359, 375 xylem8, 274 xylem9, 274 xylobiose, 11, 15, 17, 22, 33, 34

464

■

INDEX

xylogalacturonan, 10, 12, 25, 46, 48, 67, 69, 269, 277, 345, 353, 356, 358 xylogalacturonan hydrolase, 11, 277 xyloglucan, 3, 13–14, 22, 26, 32–33, 46, 48, 102, 117–119, 127, 214, 269, 271–272, 347, 348, 351–352, 354–355, 358 CAZy, 108 composites with cellulose, 355 COMPP, 53 definition, 13 fucosylation, 236 galactosylation, 267 in gymnosperms, 24 in Poales, 19 in primary wall, 6 in secondary wall, 18 in vivo modification, 374, 375 in wall models, 345 intercellular spaces, 356 MALDI-TOF, 47 OLIMP, 48 secondary walls, 359 structure, 237 tamarind, 15 taxonomical significance, 21 xylosylation, 245 xyloglucan oligomers LLGG, 21 LSGG, 21 LTGG, 21 SSGG, 21 SSGGG, 21 SXGG, 21 SXGGG, 21 XGGG, 19 XLFG, 14, 21, 24, 47, 237 XLFG + Ac, 47 XLGGG, 21 XLLG, 15, 21, 24 XLXG, 14, 15 XSGG, 21 XSGGG, 21 XXFG, 14, 21, 22, 24

XXGG, 19, 21, 236 XXGGG, 19, 21 XXGGGG, 19 XXLG, 14, 15, 21, 24 XXSG, 21 XXXG-repeat, 236 XXXG-type’, 21, 236 XXXG, 14–15, 21, 24, 47, 271 XXXGol, 26 xyloglucan-specific, 13, 46, 271 xyloglucan–RG-I, 26 xyloglucanase, 13–15, 19, 21, 24, 33–34, 368, 374–375 xyloglycan endo-transglycosylase AtEXT3, 330 xylooligomers, 255 xylose, 115, 190, 195, 237, 243–246, 251, 255, 259, 271, 274, 277, 310, 312, 411–413, 417 charophytic algae, 21 dissection, 32 grass xylan, 19 grass xyloglucan, 19 in glucuronomannan, 16 in reducing end oligosaccharide of xylan, 19 in xylan, 15 in xyloglucan, 13–14 linkage notation, 25 MALDI-TOF, 47 phellem, 18 radiolabeling, 31 reduced in mutant, 191 xylose replaced by acetyl group, 21 xylosidase, 14 xylosylation, 19, 21, 417–418 xylosyltransferase, 243–246, 253, 255, 259, 269, 274–275, 277, 312, 412, 417 Yariv reagent, 297, 322, 325, 330 Zea, 132, 324 zinc, 150, 349