Advances in Botanical Research, Volume 40

Advances in

BOTANICAL RESEARCH incorporating Advances in Plant Pathology VOLUME 40

Advances in

BOTANICAL RESEARCH incorporating Advances in Plant Pathology Editor-in-Chief J. A. CALLOW

School of Biosciences, The University of Birmingham, UK Editorial Board

A. R. HARDHAM J. S. HESLOP-HARRISON M. KREIS R. A. LEIGH E. LORD D. G. MANN P. R. SHEWRY D. SOLTIS

Australian National University, Canberra, Australia University of Leicester, UK Universite de Paris-Sud, Orsay, France University of Cambridge, Cambridge, UK University of California, Riverside, USA Royal Botanic Garden, Edinburgh, UK IACR-Long Ashton Research Station, UK University of Florida at Gainesville, USA

Advances in

BOTANICAL RESEARCH incorporating Advances in Plant Pathology

Series Editor

J.A. CALLOW School of Biosciences, University of Birmingham, Birmingham, UK

VOLUME 40

2003

Amsterdam Boston Heidelberg London New York Oxford Paris San Diego San Francisco Singapore Sydney Tokyo

ELSEVIER Ltd The Boulevard, Langford Lane Kidlington, Oxford OX5 1GB, UK  2003 Elsevier Ltd. All rights reserved. This work is protected under copyright by Elsevier, and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone: (+44) 1865 843830, fax: (+44) 1865 853333, e-mail: [email protected]. You may also complete your request on-line via the Elsevier homepage (http://www.elsevier.com), by selecting ‘Customer Support’ and then ‘Obtaining Permissions’. In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (+1) (978) 7508400, fax: (+1) (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (+44) 207 631 5555; fax: (+44) 207 631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of Elsevier is required for external resale or distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier’s Science & Technology Rights Department, at the phone, fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. Academic Press An imprint of Elsevier 84 Theobald’s Road London WC1X 8RR, UK http://www.academicpress.com

Academic Press An imprint of Elsevier 525 B Street, Suite 1900 San Diego, CA 92101-4495, USA http://www.academicpress.com

First edition 2003 Library of Congress Cataloging in Publication Data A catalog record from the Library of Congress has been applied for. British Library Cataloguing in Publication Data A catalogue record from the British Library has been applied for. ISBN:

0-12-005940-1

Typeset by Keyword Publishing Services, Barking, Essex Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

CONTENTS

CONTRIBUTORS TO VOLUME 40 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xi

CONTENTS OF VOLUMES 29–38 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

Starch Synthesis in Cereal Grains K. TOMLINSON AND K. DENYER I. II. III. IV. V. VI. VII.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Endosperm Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Supply of Carbohydrate to the Endosperm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starch Synthesis in the Endosperm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starch Accumulation During Grain Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starch Granule Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Structure of Granules, Amylose and Amylopectin . . . . . . . . . . . . . . . . . . . . . . . . A. Granule Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Amylopectin Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Amylose Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII. Enzymes Involved in Starch Biosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. ADP Glucose Pyrophosphorylase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. The Plastidial ADPG Transporter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Starch Synthase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Starch-Branching Enzyme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Isoamylase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Other Enzymes Involved in Starch Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX. Granule and Polymer Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Amylopectin Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Amylose Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Granule Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X. The Control of Starch Synthesis in Endosperms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. The Control of Metabolic Pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Coarse Regulation of Starch Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Fine Regulation of Starch Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Metabolic Control Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Starch Synthesis and Yield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2 3 4 6 9 12 15 15 17 19 20 20 23 25 31 35 36 37 37 37 38 40 40 41 43 45 46 47 47

CONTENTS

vi

The Hyperaccumulation of Metals by Plants M.R. MACNAIR I. II. III. IV. V.

VI.

VII.

VIII. IX.

The Phenomenon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Taxonomic Distribution of Hyperaccumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Variation Within Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Genetics of Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Mechanism of Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. From Soil to Plant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Into the Plant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Where Does it End Up? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Control of Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evolution of Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Increased Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Defence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Inadvertant Uptake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Drought Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Allelopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Ecological Consequences of Hyperaccumulation . . . . . . . . . . . . . . . . . . . . . . . . . . A. Effects on Other Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Effects on Herbivores and Higher Trophic Levels . . . . . . . . . . . . . . . . . . . . . . . . C. Co-evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Phytoremediation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

64 69 71 73 74 74 76 80 80 82 83 84 88 89 90 90 91 91 92 93 98 98

Plant Chromatin — Learning from Similarities and Differences J. BRZESKI, J. DYCZKOWSKI, S. KACZANOWSKI, P. ZIELENKIEWICZ AND A. JERZMANOWSKI I. II. III. IV. V.

VI. VII. VIII. IX. X.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chromatin Basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chromatin Structural Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modulation of Chromatin Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ATP-dependent Chromatin Remodelling Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . A. Arabidopsis Homologues of the Central ATPase and Other Core Subunits of the Prototype SWI/SNF-type Complexes . . . . . . . . . . . . . . . . . . . . B. CHD/Mi2 Subfamily in Arabidopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Core Histone Modifications by Acetylation and Methylation . . . . . . . . . . . . . . . . . . Plant Enzymes Responsible for Acetylation and Methylation of Core Histones Methylation of DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Problem of Linker Histones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

108 109 110 110 111 117 121 121 124 126 130 133 134 134

CONTENTS

vii

The Interface Between the Cell Cycle and Programmed Cell Death in Higher Plants: from Division unto Death D. FRANCIS I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. The Cell Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Programmed Cell Death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Cell Cycle Checkpoints and the Route to PCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Phosphoregulation of the Cell Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. The Checkpoints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Plant Cell Cycle Checkpoints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Do Plants Require Cell Cycle Checkpoints? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Instances of PCD in Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Aerenchyma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. PCD in Cell Cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Xylogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. PCD in the Endosperm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Oxidoreductive States and PCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Fungal Elicitors of PCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

144 145 147 148 149 152 157 158 160 161 161 164 167 168 169 170 171 172

The Importance of Extracellular Carbohydrate Production by Marine Epipelic Diatoms G.J.C. UNDERWOOD AND D.M. PATERSON I. Marine Benthic Diatoms and Biofilms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Motility and EPS Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Production of Colloidal Carbohydrate and EPS by Diatoms . . . . . . . . . . . . . . . . . . A. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Carbohydrate Production in Microphytobenthos in Culture . . . . . . . . . . . . . . C. Carbohydrate Production in Microphytobenthos in situ . . . . . . . . . . . . . . . . . . IV. Comparative Biochemistry of Diatom EPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Production Pathways of EPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Biochemical Composition of Diatom EPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Comparison of Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Natural Patterns of Colloidal Carbohydrate and EPS Distribution . . . . . . . . . . . . A. Horizontal Distribution of EPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Depth Distribution of EPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Seasonal Variation in EPS Concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Tidal Variation in Colloidal Carbohydrate Concentrations . . . . . . . . . . . . . . . E. Loss of Carbohydrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Impact of Environmental Stresses on Exudation in Field . . . . . . . . . . . . . . . . . VI. Sediment-stability, Flocculation and EPS Production . . . . . . . . . . . . . . . . . . . . . . . . . A. Measuring the Influence of EPS on Physical Dynamics . . . . . . . . . . . . . . . . . . B. The Problem of the Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

184 187 190 190 192 198 203 203 204 210 216 216 218 219 220 221 222 223 224 225

viii

CONTENTS

C. The Influence of EPS and Organisms on Sediment Dynamics . . . . . . . . . . . . D. The EPS Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Areas of Further Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

225 228 230 231 231

Fungal Pathogens of Insects: Cuticle Degrading Enzymes and Toxins A.K. CHARNLEY I. II. III. IV. V.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overview of the Infection Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Invasion of Host Cuticle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cuticle-degrading Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Proteolytic Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Chitinolytic Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Lipolytic and Esterolytic Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Regulation of Enzyme Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Production of Cuticle-degrading Enzymes by Other Entomopathogenic Fungi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Evidence for a Role for Cuticle-degrading Enzymes in Fungal Pathogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . H. Evolutionary Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. Role of Cuticle-degrading Enzymes in Virulence and Specificity . . . . . . . . . . VI. Toxins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Incidence of Insecticidal Toxins Amongst Entomopathogenic Fungi . . . . . . C. A Survey of Toxic Metabolites Produced by Entomopathogenic Fungi . . . D. Cyclicpeptide Toxins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Proteases and Other Enzymes as Toxins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Multiple Toxins – Synergy or Specificity? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. The Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

242 244 245 246 251 251 253 257 259 260 263 265 270 274 276 276 277 280 286 296 298 299 300 300

AUTHOR INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 SUBJECT INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345

CONTRIBUTORS TO VOLUME 40 J. BRZESKI Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warsaw University, Pawin´skiego 5A, 02-106 Warsaw, Poland A.K. CHARNLEY Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, UK K. DENYER John Innes Centre, Norwich Research Park, Colney, Norfolk NR4 7UH, UK J. DYCZKOWSKI Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warsaw University, Pawin´skiego 5A, 02-106 Warsaw, Poland D. FRANCIS School of Biosciences, Cardiff University, PO Box 915, Cardiff CF10 3TL, UK A. JERZMANOWSKI Institute of Biochemistry and Biophysics, Polish Academy of Sciences and Laboratory of Plant Molecular Biology, Warsaw University, Pawin´skiego 5A, 02-106 Warsaw, Poland S. KACZANOWSKI Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warsaw University, Pawin´skiego 5A, 02-106 Warsaw, Poland M.R. MACNAIR School of Biological Sciences, University of Exeter, Hatherly Laboratories, Prince of Wales Rd, Exeter, EX4 4PS, UK D.M. PATERSON Gatty Marine Laboratory, University of St. Andrews, Fife, KY16 9AJ, Scotland K. TOMLINSON John Innes Centre, Norwich Research Park, Colney, Norfolk NR4 7UH, UK G.J.C. UNDERWOOD Department of Biological Sciences, University of Essex, Essex, UK P. ZIELENKIEWICZ Institute of Biochemistry and Biophysics, Polish Academy of Sciences and Laboratory of Plant Molecular Biology, Warsaw University, Pawin´skiego 5A, 02-106 Warsaw, Poland

This Page Intentionally Left Blank

CONTENTS OF VOLUMES 29–38

Contents of Volume 29 The Calcicole–Calcifuge Problem Revisited J. A. LEE Ozone Impacts on Agriculture: an Issue of Global Concern M. R. ASHMORE and F. M. MARSHALL Signal Transduction Networks and the Integration of Responses to Environmental Stimuli G. I. JENKINS Mechanisms of Na+ Uptake by Plants A. AMTMANN and D. SANDERS The NaCl-induced Inhibition of Shoot Growth: The Case for Disturbed Nutrition with Special Consideration of Calcium Nutrition D. B. LAZOF and N. BERNSTEIN

Contents of Volume 30 Nitrate and Ammonium Nutrition of Plants: Physiological and Molecular Perspectives G. FORDE and D. T. CLARKSON Secondary Metabolites in Plant–Insect Interactions: Dynamic Systems of Induced and Adaptive Responses J. A. PICKETT, D. W. M. SMILEY and C. M. WOODCOCK

xii

CONTENTS OF VOLUMES 29–38

Biosynthesis and Metabolism of Caffeine and Related Purine Alkaloids in Plants H. ASHIHARA and A. CROZIER Arabinogalactan-Proteins in the Multiple Domains of the Plant Cell Surface M. D. SERPE and E. A. NOTHNAGEL Plant Disease Resistance: Progress in Basic Understanding and Practical Application N. T. KEEN

Contents of Volume 31 Trichome Diversity and Development E. WERKER Structure and Function of Secretory Cells A. FAHN Monoterpenoid Biosynthesis in Glandular Trichomes of Labiate Plants D. L. HALLAHAN Current and Potential Exploitation of Plant Glandular Trichome Productivity S. O. DUKE, C. CANEL, A. M. RIMANDO, M. R. TELLEZ, M. V. DUKE and R. N. PAUL Chemotaxonomy Based on Metabolites from Glandular Trichomes O. SPRING Anacardic Acids in Trichomes of Pelagonium: Biosynthesis, Molecular Biology and Ecological Effects D. J. SCHULTZ, J. I. MEDFORD, D. COX-FOSTER, R. A. GRAZZINI, R. CRAIG and R. O. MUMMA

CONTENTS OF VOLUMES 29–38

xiii

Specification of Epidermal Cell Morphology B. J. GLOVER and C. MARTIN Trichome Initiation in Arabidopsis A. R. WALKER and M. D. MARKS Trichome Differentiation and Morphogenesis in Arabidopsis M. HU¨LSKAMP and V. KIRIK Trichome Plasmodesmata: A Model System for Cell-to-Cell Movement F. WAIGMANN and P. ZAMBRYSKI

Contents of Volume 32 Plant Protein Kinases Plant Protein-Serine/Threonine Kinases: Classification into Subfamilies and Overview of Function D. G. HARDIE Bioinformatics: Using Phylogenetics and Databases to Investigate Plant Protein Phosphorylation E. R. INGHAM, T. P. HOLTSFORD and J. C. WALKER Protein Phosphatases: Structure, Regulation and Function S. LUAN Histidine Kinases and the Role of Two-component Systems in Plants G. E. SCHALLER Light and Protein Kinases J. C. WATSON Calcium-dependent Protein Kinases and their Relatives E. M. HRABAK

xiv

CONTENTS OF VOLUMES 29–38

Receptor-like Kinases in Plant Development K. U. TORII and S. E. CLARK A Receptor Kinase and the Self-incompatibility Response in Brassica J. M. COCK Plant Mitogen-activated Protein Kinase Signalling Pathways in the Limelight S. JOUANNIC, A.-S. LEPRINCE, A. HAMAL, A. PICAUD, M. KREIS and Y. HENRY Plant Phosphorylation and Dephosphorylation in Environmental Stress Responses in Plants K. ICHIMURA, T. MIZOGUCHI, R. YOSHIDA, T. YUASA and K. SHINOZAKI Protein Kinases in the Plant Defence Response G. SESSA and G. B. MARTIN SNF1-Related Protein Kinases (SnRKs) – Regulators at the Heart of the Control of Carbon Metabolism and Partitioning N. G. HALFORD, J.-P. BOULY and M. THOMAS Carbon and Nitrogen Metabolism and Reversible Protein Phosphorylation D. TOROSER and S. C. HUBER Protein Phosphorylation and Ion Transport: A Case Study in Guard Cells J. LI and S. M. ASSMANN

Contents of Volume 33 Foliar Endophytes and Their Interactions with Host Plants, with Specific Reference to the Gymnospermae W.-M. KRIEL, W. J. SWART and P. W. CROUS

CONTENTS OF VOLUMES 29–38

Plants in Search of Sunlight D. KOLLER The Mechanics of Root Anchorage A. R. ENNOS Molecular Genetics of Sulphate Assimilation M. J. HAWKESFORD and J. L. WRAY Pathogenicity, Host-specificity, and Population Biology of Tapesia spp., Causal Agents of Eyespot Disease of Cereals J. A. LUCAS, P. S. DYER and T. D. MURRAY

Contents of Volume 34 BIOTECHNOLOGY OF CEREALS Cereal Genomics K. J. EDWARDS and D. STEVENSON Exploiting Cereal Genetic Resources R. J. HENRY Transformation and Gene Expression P. BARCELO, S. RASCO-GAUNT, C. THORPE and P. A. LAZZERI Opportunities for the Manipulation of Development of Temperate Cereals J. R. LENTON Manipulating Cereal Endosperm Structure, Development and Composition to Improve End Use Properties P. R. SHEWRY and M. MORELL

xv

xvi

CONTENTS OF VOLUMES 29–38

Resistance to Abiotic Freezing Stress in Cereals M. A. DUNN, G. O’BRIEN, A. P. C. BROWN, S. VURAL and M. A. HUGHES Genetics and Genomics of the Rice Blast Fungus Magnaporthe grisea: Developing an Experimental Model for Understanding Fungal Diseases of Cereals N. J. TALBOT and A. J. FOSTER Impact of Biotechnology on the Production of Improved Cereal Varieties R. G. SOLOMON and R. APPELS Overview and Prospects P. R. SHEWRY, P. A. LAZZERI and K. J. EDWARDS

Contents of Volume 35 Recent Advances in the Cell Biology of Chlorophyll Catabolism H. THOMAS, H. OUGHAM and S. HO¨RTENSTEINER The Microspore: A Haploid Multipurpose Cell A. TOURAEV, M. PFOSSER and E. HEBERLE-BORS The Seed Oleosins: Structure Properties and Biological Role J. NAPIER, F. BEAUDOIN, A. TATHAM and P. SHEWRY Compartmentation of Proteins in the Protein Storage Vacuole: A Compound Organelle in Plant Cells L. JIANG and J. ROGERS Intraspecific Variation in Seaweeds: The Application of New Tools and Approaches C. MAGGS and R. WATTIER

CONTENTS OF VOLUMES 29–38

Glucosinolates and their Degradation Products R. F. MITHEN

Contents of Volume 36 Aphids: Non-persistent Transmission T. P. PIRONE and K. L. PERRY Persistent Transmission of Luteoviruses by Aphids B. REAVY and M. A. MAYO Fungi M. J. ADAMS Whitefly Transmission of Plant Viruses J. K. BROWN and H. CZOSNEK Beetles R. C. GERGERICH Thrips As Vectors of Tospoviruses D. E. ULLMAN, R. MEIDEROS, L. R. CAMPBELL, A. E. WHITFIELD, J. L. SHERWOOD and T. L. GERMAN Virus Transmission by Leafhoppers, Planthoppers and Treehoppers (Auchenorrhyncha, Homoptera) E. AMMAR and L. R. NAULT Nematodes S. A. MacFARLANE, R. NEILSON and D. J. F. BROWN Other Vectors R. T. PLUMB

xvii

xviii

CONTENTS OF VOLUMES 29–38

Contents of Volume 37 ANTHOCYANINS IN LEAVES Anthocyanins in Leaves and Other Vegetative Organs: an Introduction D. W. LEE and K. S. GOULD Le Rouge et le Noir: Are Anthocyanins Plant Melanins? G. S. TIMMINS, N. M. HOLBROOK and T. S. FEILD Anthocyanins in Leaves: History, Phylogeny and Development D. W. LEE The Final Steps in Anthocyanin Formation: a Story of Modification and Sequestration C. S. WINEFIELD Molecular Genetics and Control of Anthocyanin Expression B. WINKEL-SHIRLEY Differential Expression and Functional Significance of Anthocyanins in Relation to Phasic Development in Hedera helix L. W. P. HACKETT Do Anthocyanins Function as Osmoregulators in Leaf Tissues? L. CHALKER-SCOTT The Role of Anthocyanins for Photosynthesis of Alaskan Arctic Evergreens During Snowmelt S. F. OBERBAUER and G. STARR Anthocyanins in Autumn Leaf Senescence D. W. LEE A Unified Explanation for Anthocyanins in Leaves? K. S. GOULD, S. O. NEILL and T. C. VOGELMANN

CONTENTS OF VOLUMES 29–38

xix

Contents of Volume 38 An Epidemiological Framework For Disease Management C. A. GILLIGAN Golgi-independent Trafficking of Macromolecules to the Plant Vacuole D. C. BASSHAM Phosphoenolpyruvate Carboxykinase: Structure, Function and Regulation R. P. WALKER and Z.-H. CHEN Developmental Genetics of the Angiosperm Leaf C. A. KIDNER, M. C. P. TIMMERMANS, M. E. BYRNE and R. A. MARTIENSSEN A Model for the Evolution and Genesis of the Pseudotetraploid Arabidopsis thaliana Genome Y. HENRY, A. CHAMPION, I. GY, A. PICAUD, A. LECHARNY and M. KREIS

This Page Intentionally Left Blank

Starch Synthesis in Cereal Grains

KIM TOMLINSON AND KAY DENYER

John Innes Centre, Norwich Research Park, Colney, Norfolk NR4 7UH, UK

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Endosperm Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. The Supply of Carbohydrate to the Endosperm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Starch Synthesis in the Endosperm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Starch Accumulation During Grain Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Starch Granule Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. The Structure of Granules, Amylose and Amylopectin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Granule Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Amylopectin Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Amylose Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII. Enzymes Involved in Starch Biosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. ADP Glucose Pyrophosphorylase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. The Plastidial ADPG Transporter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Starch Synthase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Starch-Branching Enzyme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Isoamylase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Other Enzymes Involved in Starch Synthesis . . . . . . . . . . . . . . . . . . . . . . IX. Granule and Polymer Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Amylopectin Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Amylose Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Granule Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X. The Control of Starch Synthesis in Endosperms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Advances in Botanical Research Vol. 40 incorporating Advances in Plant Pathology ISBN 0-12-005940-1

2 3 4 6 9 12 15 15 17 19 20 20 23 25 31 35 36 37 37 37 38 40

Copyright  2003 Elsevier Ltd All rights of reproduction in any form reserved

2

K. TOMLINSON and K. DENYER

A. B. C. D. E.

The Control of Metabolic Pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Coarse Regulation of Starch Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fine Regulation of Starch Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Metabolic Control Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starch Synthesis and Yield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

40 41 43 45 46 47 47

ABSTRACT In this review, we consider how starch is made in the developing endosperms of the major cereal crops, wheat, maize and rice and how this process is controlled. We describe how carbohydrate is supplied to the endosperm, how the endosperm develops and the pattern of starch accumulation in the developing endosperm. The nature of starch in cereals in terms of granule morphology, the internal structure of granules and the structure of the glucan polymers which comprise starch is described. We review the current understanding of granule and polymer initiation and synthesis giving detailed information about the properties of the enzymes involved including the roles of the multiple isoforms of several of these enzymes. We concentrate on the nature and importance of the cereal-endosperm-specific pathway of starch synthesis involving the production of the substrate for starch synthesis, ADPG in the cytosol rather than the plastid. Finally, we give a summary of our current understanding of the control of biochemical pathways in general and of starch synthesis in endosperms in particular.

I. INTRODUCTION In this review, we consider how starch is made in developing cereal grains and how this process is controlled. Starch accounts for approximately threequarters of the dry weight of the mature grain and is a major determinant of yield (Watson, 1987). The major fate of carbon imported into the developing grain is incorporation into starch. We will concentrate on starch synthesis in the major cereal crops, maize, rice and wheat since these are the species for which we have the most information. We will assume that the reader is familiar with the anatomy of the cereal grain and the economic importance of cereal grains as a food, an animal feed and a source of industrial raw materials. We recommend an article in an earlier volume of this series by Shewry and Morell (2001) and others (cereals: Evers and Millar, 2002; maize: Watson and Ramstad, 1987; rice: Hoshikawa, 1993 and other articles in the Science of the Rice Plant, Vols 1 and 2) for more information on these topics. Cereals are cultivated grasses and their grains are single-seeded fruits. During grain development, some starch is synthesised in both the embryo and the pericarp (wheat: Chevalier and Lingle, 1983; Black et al., 1996; Evers et al., 1999 and maize: Watson, 1987). However, in wheat and possibly other cereals, starch accumulated in both these organs is later

STARCH SYNTHESIS IN CEREAL GRAINS

3

broken down so that by the time the grain is mature, they contain little or no starch. The endosperm is the site of synthesis of most of the starch that accumulates in the cereal grain. We will therefore concentrate on starch synthesis in the endosperm starting with a description of how the endosperm forms during development.

II. ENDOSPERM DEVELOPMENT The process of endosperm development, from fertilisation to the establishment of a multicellular endosperm, is probably essentially similar in all cereals and has been described in detail elsewhere (Olsen et al., 1999; Olsen, 2001; Evers and Miller, 2002). In brief, after fertilisation, repeated nuclear divisions give rise to an endosperm cell containing a large central vacuole surrounded by a layer of cytoplasm containing many nuclei (the coenocytic stage). Cell walls form between the nuclei and these nuclei divide further to form cells that eventually replace the central vacuole and fill the endosperm. Further enlargement of the endosperm is achieved by cell division as well as cell expansion. The spatial organisation of cell divisions within the endosperm, that generates a gradient of cell age within the endosperm, is basically similar between species. Cellularisation within the coenocytic endosperm begins at the embryo end (wheat: Chaudhury et al., 2001; rice: Hoshikawa, 1993) and proceeds towards the centre of the endosperm (barley: Bosnes et al., 1992; Olsen et al., 1999; and rice: Hoshikawa, 1993). At this early stage, the youngest cells are those in the centre of the endosperm. However, most of the later cell divisions occur in the outer endosperm so that, by the time cell division ceases, the youngest endosperm cells are those in the outer endosperm (Evers and Millar, 2002). Clonal analysis of endosperm was used to trace the development of cell-lineages within the maize endosperm (McClintock, 1978; Lopes and Larkins, 1993). This work also showed that the oldest cells were in the centre of the endosperm. In most cases, clones of cells originated in the centre of the endosperm and proliferated outwards. There are, however, some differences between species in the precise pattern of cell divisions. In rice, there is a layer of meristematic cells all around the outer edge of the developing endosperm from the earliest stages of endosperm development. Thus, most (86%) cell divisions occur at the periphery and older endosperm cells are pushed inwards (Hoshikawa, 1973). The formerly meristematic cells in rice endosperm eventually differentiate into the aleurone layer. In maize, cell divisions initially occur throughout the endosperm but rapidly become confined to

4

K. TOMLINSON and K. DENYER

Fig. 1. Diagram of cereal grains. (A) Transverse mid-section of a developing wheat grain showing three types of endosperm cells (adapted from Olsen et al., 1992). (B) Transverse section through the crown of a mature kernel of yellow dent maize showing the distribution of horny and floury endosperm (adapted from Wolf et al., 1952).

the peripheral cell layers (Watson, 1987). In wheat and barley, meristematic activity is confined to the outer endosperm cells in the dorsal area over the crease rather than being all around the edge of the endosperm as in rice. This activity gives rise to the prismatic cells (see Fig. 1a and Bosnes et al., 1992). Elsewhere in the endosperm of wheat and barley, random cell divisions occur giving rise to the irregular central cells in the cheeks of the endosperm (Olsen et al., 1999). There are also gradients of cell death within developing endosperms, the patterns of which vary between species. In maize endosperm, a wave of cell death moves through the kernel from the crown towards the base (Young et al., 1997; Young and Gallie, 2000). In contrast, in wheat there is no distinct wave of cell death, instead cell death occurs at random throughout the endosperm (Young and Gallie, 1999).

III. THE SUPPLY OF CARBOHYDRATE TO THE ENDOSPERM All of the macro- and micronutrients, including the carbon required for starch synthesis, must be imported into the endosperm from the maternal tissues. Exactly which metabolite is the major imported source of carbon for starch synthesis in the endosperm is not entirely clear and probably varies between species and with developmental age. It is generally assumed for all grains that transfer from the maternal tissues must be apoplastic. The maternal and filial tissues are genetically distinct and microscopic examination shows that for all but the very earliest stages of development, these tissues are clearly symplastically isolated. Thus, during grain-filling,

STARCH SYNTHESIS IN CEREAL GRAINS

5

the bulk of the carbon transferred from maternal to filial tissues must pass through the apoplast. In wheat grains, sucrose produced by photosynthesis in the leaves moves in the phloem along the vascular bundle in the crease, then across the tissues at the base of the crease and into the endosperm cavity and finally it moves radially into and through the endosperm (Ugalde and Jenner, 1990a). There is a two-fold gradient of sucrose concentration across the endosperm from the cavity to the periphery, decreasing in the outward direction (Ugalde and Jenner, 1990b). Sucrose does not appear to be cleaved as it moves from the phloem to the endosperm in wheat. This is shown by experiments in which 14 C-fructosyl-sucrose was supplied to developing endosperm. Very little randomisation of the label between the hexose-moieties of sucrose was observed in sucrose recovered from the endosperm suggesting that hydrolysis of sucrose is not a prerequisite for transport into the endosperm (Jenner, 1973; Jenner, 1974; Ugalde and Jenner, 1990b). That sucrose is probably the form in which carbohydrate is transported into the endosperm was also shown by experiments in which radioactive sugars were supplied to developing wheat endosperms via a needle into the endosperm cavity (Wang et al., 1993a). The rates of starch synthesis from sucrose in such experiments were comparable with those observed in vivo (Wang et al., 1993a). Thus, sucrose can be readily absorbed and metabolised by the cells of the endosperm. In developing maize kernels, some hydrolysis of sucrose by invertases bound to the cell walls of the endosperm is necessary for normal grain development. The maize mutant miniture-1, that lacks an isoform of cell wall-bound invertase, has impaired endosperm development and a drastically reduced final grain weight (Vilhar et al., 2002). The products of sucrose hydrolysis, in this case, are probably required primarily to maintain cell division in the early stages of endosperm development. To what extent they also provide substrates for starch synthesis is not clear. It has been suggested that in maize and sorghum, the specialised cells at the base of the endosperm (basal endosperm cells) may re-synthesise sucrose from imported hexose prior to the transport of sucrose and its consumption by cells in the rest of the endosperm (Griffith et al., 1987; Singh et al., 1991). However, there is also contrasting evidence that in maize, as in wheat, cleavage of sucrose to hexoses is not specifically needed for the uptake of carbohydrate from the apoplast into the cytosol of the endosperm cells. When an analog of sucrose, 10 -fluorosucrose that is resistant to hydrolysis by invertases, was supplied to the leaves of the husk around a developing maize cob, it passed from the maternal tissues into the endosperm at the same rate and to the same extent as sucrose (Schmalstig and Hitz, 1987).

6

K. TOMLINSON and K. DENYER

On arrival in the cytosol of the endosperm cells, sucrose destined to support starch synthesis is metabolised by sucrose synthase and alkaline invertase. The activity of sucrose synthase is high compared with that of invertase (maize: Doehlert and Felker, 1987; barley: Lingle and Chevalier, 1984; rice: Perez et al., 1975). However, there is evidence to suggest that the primary metabolic role of sucrose synthase is to provide the substrates for pathways other than starch synthesis such as the pathway of cellulose synthesis. A double mutant of maize lacking two of the three known sucrose synthase isoforms, SH1 and SUS1 has less than 1% of the normal sucrose synthase activity and yet accumulates 54% of the normal amount of starch (Carlson et al., 2002). This suggests that invertase and/or the third, minor sucrose synthase isoform encoded by Sus3 are responsible for providing the majority of the substrate for starch synthesis.

IV. STARCH SYNTHESIS IN THE ENDOSPERM Starch accumulation in the endosperm starts immediately after cellularisation of the endosperm is complete. It continues at the same, rapid and approximately linear rate during the phase of cell division and during maturation, after cell division has ceased (Bosnes et al., 1992). In many biochemical and molecular studies of starch synthesis in cereal endosperms, there is a tendency to assume that the cells of the endosperm are all doing exactly the same thing at the same time and to the same extent. In truth, the cells in the endosperm are very different from one another in terms of their metabolism. In Section II, we described how the development of the endosperm gives rise to cells with different morphologies and to gradients of cell age within the endosperm. It is likely that these differences are largely responsible for differences between cells in starch synthesis and other metabolic processes. The biggest difference between cells in the endosperm is between the aleurone cells and those in the rest of the endosperm. The aleurone cells contain less starch and more protein than the starchy endosperm cells (e.g. rice: Hoshikawa, 1993). In rice and wheat, there are gradients of starch content and granule size that reflect the gradients of age of the endosperm cells. The sub-aleurone cells in wheat contain less starch and smaller starch granules than the bulk of the starchy endosperm cells. This is possibly because starch synthesis starts first in the older, central endosperm and occurs later in the younger, peripheral endosperm. Thus, the concentration of starch is highest in the centre of the grain and the starch granules in the centre of the endosperm are larger (Bradbury et al., 1956; Briarty and

STARCH SYNTHESIS IN CEREAL GRAINS

7

Hughes, 1979; Ugalde and Jenner, 1990b; Evers and Miller, 2002). Starch synthesis also begins first in the central endosperm in maize. However, in addition to this minor gradient from the centre outwards to the periphery, in maize there is also a major gradient of granule formation from the crown to the base of the endosperm (Boyer et al., 1976; Boyer et al., 1977). Variation in starch content and composition between cells in the endosperm is most eloquently demonstrated by recent studies of partial waxy mutants of barley (Oscarsson et al., 1997; Andersson et al., 1999). In these mutants, the activity of the enzyme responsible for the synthesis of the amylose component of starch, GBSSI, is reduced by a lesion in the 50 noncoding region (Patron et al., 2002). This leads to a change in the temporal and spatial expression of GBSSI and thus, of amylose synthesis. Amylose accumulates in these mutants only in the cells in the outer endosperm and only relatively late in endosperm development (Fig. 2). This gradient of amylose synthesis appears to be peculiar to these particular waxy mutants. In wild-type barley endosperms, there is no apparent gradient in amylose content between the outer and inner cells (Andersson et al., 1999). However, the pattern of expression of the mutant form of the GBSSI gene presumably reflects some underlying gradient that is present in the wild-type grains but that does not normally affect amylose synthesis. Another example of heterogeneity in starch metabolism between endosperm cells is shown by variation in the amount of starch and a soluble form of starch called phytoglycogen in the endosperm of sugary mutants of cereals (maize: Boyer et al., 1977; rice: Nakamura et al., 1997

Fig. 2. Developing endosperm from the waxy barley mutant, Iyatoma Mochi. Whole grains were fixed in formaldehyde, embedded in wax, sectioned and stained with iodine solution. Sample was from a grain of 50–70 mg fresh weight and the section was taken from the outer endosperm, opposite the crease. The bar represents 50 mm.

8

K. TOMLINSON and K. DENYER

and barley: Burton et al., 2002a). In all of these mutants, the cells in the endosperm vary in their starch and phytoglycogen contents. For example in rice, starch in the sugary endosperms is restricted to the outer endosperm cells whereas phytoglycogen is much more abundant in the inner region (Nakamura et al., 1997). Presumably these differences reflect variations between cells in the enzymes responsible for starch and phytoglycogen synthesis but this is not yet understood. In addition to the cell-to-cell variations in starch content, there are also gradients of sucrose content (wheat: Ugalde and Jenner, 1990b), enzyme activities and protein content within the developing endosperm. In wheat endosperm in mid to late development, declines in the activities of some starch biosynthetic enzymes at the end of the period of starch accumulation occur first at the embryo end (our unpublished data) and in maize endosperm, the activities of enzymes associated with starch synthesis are higher in the upper and middle endosperm than in the lower endosperm (Doehlert, 1990). Unlike the content of starch that is lowest in the subaleurone (Ugalde and Jenner, 1990b), the protein content per cell varies very little across wheat endosperm (Evers, 1970) or is highest in the cells of the outer endosperm (Ugalde and Jenner, 1990c). Thus, the cells in the outer endosperm contain proportionally more protein than those in the inner endosperm. Differences in protein content (Watson, 1987) and in the proportions of different types of storage proteins (Geetha et al., 1991) between regions of maize endosperm have been noted. Areas of maize endosperm containing high protein levels are called ‘horny’ whereas low protein regions are called ‘floury’ (Wolf et al., 1952). In the endosperm of yellow dent maize, the horny endosperm is mainly at the back and sides of the kernel and the floury endosperm in the centre and in the crown (Fig. 1b). In barley, there are ‘steely’ and ‘mealy’ areas of endosperm that vary in texture and degradability upon germination. The differences between ‘steely’ and ‘mealy’ regions may at least in part, be due to variations in protein content as in the ‘horny’ and ‘floury’ regions of maize endosperm but the nature of the differences between these areas of endosperm is not completely understood (Palmer, 1989). The metabolic differences between endosperm cells have consequences for the study of starch and starch synthesis in cereal grains. Measurements of enzyme activities, levels of transcripts and metabolite contents are normally averaged over the whole of the endosperm. In many cases, this approximation will not adversely affect the interpretation of the results but it may have an impact in some instances. It is therefore, important to consider the heterogeneity of cell metabolism and starch within the endosperm. In the case of the waxy barley mutants measured above, on average, the amylose

STARCH SYNTHESIS IN CEREAL GRAINS

9

content of the starch was reduced. However, this averaging hides the fact that the starch contains some granules with near-normal amylose content (from the peripheral endosperm cells) and others with zero amylose (from the central endosperm).

V. STARCH ACCUMULATION DURING GRAIN DEVELOPMENT The pattern of starch accumulation in all cereal grains is essentially similar to that shown in Fig. 3 for barley. After an initial lag, during which cell division is taking place in the endosperm, there is a phase of rapid starch accumulation followed by a phase during which little if any, starch accumulates (e.g. maize; Salvador and Pearce, 1995). The accumulation of the amylose and amylopectin components of starch occurs at different rates during the grain-filling period (Fig. 3c). Amylose accumulation lags behind that of amylopectin and continues after amylopectin synthesis ceases. This is true for both the A-type and the B-type granules in barley. Thus, there is an increase in the ratio of amylose to amylopectin throughout development in starch as a whole and in both the A- and B-type granules. The reason for this may be that amylose synthesis occurs within the amylopectin matrix (see Section VIII.C.2.) and is therefore dependent upon the prior synthesis of this matrix. The lag is of the order of several days in barley (Fig. 3c) suggesting that, after its synthesis, the amylopectin matrix is not completely filled with amylose for some considerable time. The final starch content of the grain is determined by the rate of starch synthesis during the phase of rapid accumulation and also by the duration of this phase. These two factors can vary independently with changes in growth conditions and between cultivars. High temperatures during the starch-filling period for example, can decrease the duration of grain filling without having a large impact on the maximum rate of starch synthesis (Chowdhury and Wardlaw, 1978). However, effects on both rate and duration are more commonly seen (e.g. Sofield et al., 1977a; Wiegand and Cuellar, 1981). The rate of starch accumulation is thought to be very close to the rate of starch synthesis. In other words, it is believed that there is no significant turnover of starch during the grain-filling period. Measurement of the rate of starch synthesis in wheat endosperm in vitro using radioactively labelled substrates showed that this was very similar to the rate of starch accumulation in vivo (Wang et al., 1993a). This suggests that in vivo, there is no net movement of label into starch and then out of starch into other products. This does not mean that starch-degrading enzymes are not active

10

K. TOMLINSON and K. DENYER

Fig. 3. Starch accumulation in developing wild-type barley grains. Data from McDonald et al., (1991). (a) The accumulation of starch and its components. f ¼ starch; œ ¼ amylopectin; m ¼ % amylose; i ¼ amylose. (b) Estimates of the rate of starch accumulation for data shown in (a). g ¼ points used to fit line (1) which represents the period of rapid grain-filling. m ¼ points used to fit line (2) which represents the period of late grain-filling. m þ g þ f ¼ points used to fit generalised logistic curve. This curve accounts for 99.6% of the variation (Genstat.). (c) The rate of amylose (grey line) and amylopectin (black line) accumulation in the A- and B-type granules. Rate were derived from curve-fitting to contents (generalised logistic) as in part b.

STARCH SYNTHESIS IN CEREAL GRAINS

11

or important in the starch biosynthetic process. Specific enzymes, such as isoamylase (see Section VIII.E) have proven and important roles in the determination of starch structure. The products of these enzymes may be recycled into starch directly or they may be degraded further prior to re-incorporation into starch (Myers et al., 2000). Assuming that there is no starch turnover, the rate of starch synthesis during endosperm development can be estimated from measurements of the rate of accumulation of starch. In some species and conditions, there may be a phase of rapid, linear starch accumulation that changes abruptly to a complete cessation of starch synthesis (e.g. maize, Keeling, 1999; wheat, Sofield et al., 1977a). Most studies of wheat, however, suggest that the rate of starch synthesis is declining steadily for a significant part of the grainfilling period. Modelling of data such as that shown in Fig. 3a for barley suggests that at least two linear curves are required to describe the rate of starch synthesis accurately from the onset of grain filling to grain maturity (Fig. 3b). A better ‘fit’ to the rate of synthesis is obtained using a non-linear curve (Fig. 3b). The non-linear accumulation of starch in wheat is reflected in the pattern of dry weight accumulation (Chanda et al., 1999). A significant decline in the rate of dry weight accumulation during the grain-filling period has been noted in many studies (Sofield et al., 1977a; Simmons and Crookston, 1979; Loss et al., 1989; Jenner, 1991). The effect of this decline on yield can be estimated from the data in Sofield et al. (1977a). On average 12% (range 0 to 34%; data for different species and environmental conditions) of the dry weight of the grain is accumulated during the period of declining rate, after the linear phase of grain filling. If the linear rate were maintained in the late grain filling period, rather than declining, then assuming no change in duration of grain filling, the predicted increase in yield would average approximately 6%. The factors responsible for the declining rate are not known. There is a decline in the rate of starch synthesis as well as a reduction in water content. The possibility that the cessation of growth is primarily driven by water loss perhaps via control of the entry of water into the grain has been considered (Sofield et al., 1977b). However, the data suggest that the decline in the accumulation of dry matter starts before there is a measurable decrease in water content. Therefore, water loss is more likely to be a consequence of, rather than a cause of, the decline in the accumulation of dry matter and the onset of maturation. The primary cause of the reduction in the rate of starch accumulation is still to be discovered. Some starch degradation may occur at the end of grain development. This occurs particularly when the grain prematurely germinates. Even very limited damage to starch granules due to pre-harvest sprouting in wheat can

12

K. TOMLINSON and K. DENYER

have a serious effect on flour quality (Kruger, 1989). For this reason, measurements of -amylase activity – such as the Hagberg Falling Number – are major determinants of the value and potential end-uses of wheat grain. However, despite large and economically important changes in starch structure, pre-harvest sprouting, even when it is severe, has very little impact on starch content.

VI. STARCH GRANULE MORPHOLOGY Starch granule morphology is largely genetically controlled and varies greatly between cereal species. Whereas environmental conditions have only a small impact on granule morphology, genetic defects can cause radical alterations. For example, granules from different low-starch mutants of barley have very varied morphologies (Fig. 4). The nature of this genetic control is almost entirely unknown. It is likely that granule morphology is in large part determined by the structure of the amylopectin molecules from which the granules are built. For granules that are approximately spherical, there may be little or no additional genetic control of morphology beyond the determination of amylopectin structure. However, for granules with very complex shapes, such as the A-type granules in the Triticeae (see Evers, 1971 and Fig. 4), it is difficult to imagine that the morphology is solely a consequence of amylopectin structure. Careful observations of growing A-type granules in wheat suggest that their synthesis involves the deposition of starch on certain areas of the granule surface only (Evers, 1971). The mechanism that exists within the cell to achieve this spatial coordination of starch synthesis is unknown. A filamentous network of proteins – the plastoskeleton – has recently been found to exist inside plastids in addition to the actin and tubulin networks that exist in the cytosol (Reski, 2002). This is thought to maintain the shape of the plastids and is involved in the process of plastid division. Perhaps it could also provide a framework for the spatial organisation of enzymes involved in starch synthesis within plastids. Several systems of classification of granule morphology have been devised to describe starch granules in cereal endosperms and other plants. The least complex of these defines two types of granules: simple and compound granules (Fig. 5). Granules that form each in a separate plastid are called simple granules whereas compound granules are formed when many granules initiate per plastid. More complex classification systems, with three major classes, five sub-classes and 17 different types of granule morphology exist (Reichert, 1913), but will not be described further here.

STARCH SYNTHESIS IN CEREAL GRAINS

13

Fig. 4. Starch granule morphology in barley endosperm from wild-type (Bomi) and low-starch mutants (Notch 2, Risø13 and Risø527). Mature seeds were cracked to reveal the endosperm and observed using a Scanning Electron Microscope. The genotypes and scale are indicated. Notch2 accumulates small, irregularly shaped granules and is an isoamylase mutant (Burton et al., 2002b). The A-type granules Risø13 and several other low-starch barley mutants (data not shown) are irregularly shaped and frequently dimpled. The A-type granules of Risø527 are near-spherical in contrast to those in the wild-type which are discus-shaped. The nature of the biochemical lesions responsible for the low-starch phenotypes of Risø13 and Risø527 are not yet known. The bar represents a distance of 10 mm. Pictures were taken by M. Parker, Institute of Food Research, Norwich, UK.

Simple granules that form one per plastid are often regular in shape. However, in cells in which the starch content is very high such as those in maize endosperm, the starch granules and plastids can become so tightly packed that they become irregularly shaped and angular due to the pressure of neighbouring granules against one another. Compound granules such as those in rice and oats comprise several hundred separately initiated ‘sub-granules’ or granulae (Reichert, 1913). The granulae can be regular in shape if the starch content of the plastid is relatively low as in early endosperm development or in certain low-starch mutants such as barley Risø 17 (Burton et al., 2002a). Usually however, granules inside the same plastid become tightly appressed and irregular in shape and may partly fuse together. In this case, even when the starch is extracted, the granulae from one plastid may occasionally remain together as one compound granule.

14

K. TOMLINSON and K. DENYER

Fig. 5. Simple and compound starch granules. (a) A transmission electron micrograph showing a plastid from a developing pea embryo. The plastid contains a single, simple starch granule. Simple granules are also found in maize endosperm. (b) A light micrograph showing a cell from a developing oat endosperm. The cell contains severals plastids. Some of the plastids contain compound starch granules. Picture was taken by T. Verhoeven, John Innes Centre. (c) Diagrams representing plastids containing either simple or compound starch granules.

In some species, there is more than one type of starch granule in the cells of the endosperm. The most extreme example of such a bimodal distribution of granule size and shape is the A- and B-type starch granules found in the endosperm of members of the tribe Triticeae (such as wheat and barley). For a detailed description of the initiation and development of A- and B-type granules, see Shewry and Morell (2001) and references therein. Briefly, the A-type granules are disc-shaped and are initiated about 10 days earlier than the more spherical B-type granules. The two sorts of starch granule are thought to co-exist in the same plastid although they are confined to different regions within the plastid. In each plastid, one A-type granule forms in the central space and then later, several B-type granules form in stromules – tubular protrusions from the main body of the plastid (Parker, 1985; Langeveld et al., 2000). Since there are multiple granules initiating in each plastid, this type of granule morphology could be considered as a special sort of compound granule. In oat endosperm, as well as compound starch granules there are separate single granules that are roughly spherical (Reichert, 1913). It is not clear whether these spherical granules are initiated in separate plastids to the compound granule and are therefore simple granules or whether they initiate in the same plastids as the compound granules. Oats are closely related to the

STARCH SYNTHESIS IN CEREAL GRAINS

15

Fig. 6. The phylogeny of the grass family. Diagram based on information given by the Grass Phylogeny Working Group (2001). The relationships between the major cereal species only are shown. C ¼ species with compound (or A- and B-types) granules. S ¼ species with simple granules.

Triticeae and are similar to the Triticeae in having a bimodal distribution of granule types. Thus, it is tempting to speculate that the spherical granules in oats form in stromules as do the B-type granules in the Triticeae. However, we are not aware of any evidence that either supports or refutes this idea. The number of granules that form per plastid is genetically determined and different granule morphologies appear to be restricted to particular taxonomic groups within the grass family (Fig. 6). As mentioned above, the peculiar A- and B-type granule morphology is restricted to the Triticeae. In addition, the starch granules of the sub-family Panicoideae (including maize) are usually simple whereas most other grasses have compound granules (Tateoka, 1962). Based on the relative abundance of the compound-granule type of morphology within the grass family, Tateoka (1962) suggested that this was likely to be the ancestral state from which the simple-granule type was derived. As more information about the processes controlling granule initiation becomes available, we may eventually be in a position to understand at the molecular level how granule morphology – and more specifically granule number per plastid – evolved within the grass family.

VII. THE STRUCTURE OF GRANULES, AMYLOSE AND AMYLOPECTIN A. GRANULE STRUCTURE

Despite large variations between species in granule morphology, the internal organisation of starch granules is less variable. The structures represented in Fig. 7 are seen in starches from all plants, not just in cereals. For reviews of starch structure, see Kossmann and Lloyd (2000), Gallant et al. (1997) and references therein.

16

K. TOMLINSON and K. DENYER

α

α

Fig. 7. The internal structure of starch granules. (a) Hard and soft growth rings revealed by amylase digestion. A starch granule from a potato tuber was cracked open, etched with amylase and viewed at two magnifications. The granule is approximately 15–30 mm in diameter. Picture taken by E. Pilling, John Innes Centre. (b). Layers of large and small blocklets within the hard and soft regions of the growth rings, respectively. Each blocklet may comprise a superhelix of amylopectin (Gallant et al., 1997). Within the blocklet/supehelix are crystalline and amorphous layers with a repeat distance of 9 nm (Jenkins et al., 1993). (c) A fragment of an amylopectin molecule showing clusters of double helices. Within a cluster, adjacent glucan chains form double helices which pack into a crystalline array. Most of the branch points and the chains connecting the clusters are contained within the amorphous layers (Jane et al., 1997).

STARCH SYNTHESIS IN CEREAL GRAINS

17

Rings of hard (semi-crystalline, not readily digested by amylase) and soft (relatively disorganised, easily digested by amylase) material can be seen in starch granules when these are observed under the light microscope or when fractured granules are etched by treatment with -amylase and viewed by scanning electron microscopy (SEM) (Gallant et al., 1997). What is seen as rings are actually concentric, three-dimensional spherical shells. These shells, often referred to as growth rings, are 100 nm or more apart. It has been suggested that one growth ring forms per day although there is no direct evidence for this. High-magnification microscopy of various kinds (SEM, transmission electron microscopy [TEM] and atomic force microscopy [AFM]) has shown that growth rings appear to be composed of elongated structures known as blocklets that are stacked on top of one another (Gallant et al., 1997; Ridout et al., 2002). These occur in both the hard and soft layers of the growth ring, although the blocklets in the hard, crystalline layers are generally larger than those in the soft layers. The diameter of the blocklets varies between 20–500 nm depending on the source of the starch. Studies of partially degraded starch granules suggest that molecules of the major glucose polymer in starch (amylopectin, see below) are organised into ‘super-helices’ (Oostergetel and van Bruggen, 1993). These may relate to the blocklets observed by microscopy. Starch is an -glucan composed of two glucose-polymers, amylose and amylopectin. Both polymers are made up of D-glucose units linked through glycosidic bonds between carbons 1 and 4 of adjacent glucose monomers (Fig. 7). In cereal endosperms and other plant storage organs, the ratio of amylose to amylopectin is fixed within fairly narrow limits: between approximately 75% amylopectin : 25% amylose and 66% amylopectin : 33% amylose. However, mutants with very different ratios of amylose to amylopectin exist. For some of these mutants, the distinction between amylose and amylopectin becomes blurred. For example, abnormal amylopectin molecules with fewer than normal branches take on some of the physical characteristics of amylose molecules. Such starches are sometimes referred to, perhaps misleadingly, as ‘high-amylose’ starches. The structure of the amylose and amylopectin polymers and their organisation within starch granules will be discussed in the following two sections. B. AMYLOPECTIN STRUCTURE

Amylopectin is a large, multiply branched glucose polymer (Fig. 7). The great complexity and size of amylopectin prevent the determination of its precise structure (Thompson, 2000). Measurements of the lengths of the chains

18

K. TOMLINSON and K. DENYER

within amylopectin and the non-random distribution of the branch points joining the constituent chains together suggest that amylopectin consists of clusters of short chains (Hizukuri, 1986). The clusters contain short glucose chains (containing approximately 6–20 glucose units) and there are longer chains of glucose extending between and joining adjacent clusters. Within the starch granule, adjacent amylopectin molecules are thought to lie parallel to one another with their clusters of short chains in register. The chains within the clusters lie approximately along the radius of the starch granule with their non-reducing ends towards the granule surface. Pairs of short chains form double helices that pack together to form crystalline arrays. Between the parallel crystalline arrays (crystalline lamellae) are noncrystalline regions (amorphous lamellae) containing the branch points that join together the chains within the clusters. A strong periodicity within amylopectin crystals of 9 nm is seen in small angle X-ray scattering profiles (Jenkins et al., 1993). This repeat distance is remarkably constant (ranging from 8.5 to 9.4 nm; Jenkins et al., 1993) regardless of the species from which the starch was extracted and despite variations in the composition of starch from different mutants. This 9-nm repeat is thought to correspond to the radial distance between adjacent amylopectin arrays and therefore represents the average size of a cluster. Two types of amylopectin crystals called A-type and B-type have been found in plant starches. The chains within the amylopectin clusters in A-type crystals are more densely packed than in the B-types (Wang et al., 1998). Granules contain only A-type crystals (A-type starch, e.g., found in most normal cereal seeds), only B-type (B-type starch, e.g., found in seeds of high-amylose maize mutants) or both types of crystals (C-type starch, e.g., found in pea seeds). The type of crystal formed depends upon the lengths of the chains within the cluster and the exact positioning of the branch points (Jane et al., 1997). Starches with amylopectin with relatively short average chain-lengths, have A-type crystals. In these crystals, the branch points are found in both the crystalline and non-crystalline regions of the amylopectin clusters. B-type starches have amylopectin with longer average chain-lengths and most of the branch points are within the non-crystalline region of the clusters. The differences between the amylopectin molecules that form A-type and B-type crystals are very subtle. The average amylopectin chainlengths in A- and B-type starches can vary by as little as one glucose unit (Jane et al., 1997). However, the nature of the amylopectin crystals has a profound effect on properties of starch such as its susceptibility to hydrolysis by -amylase. A model that equates crystalline amylopectin to a side-chain liquid crystal reveals the importance of the cluster structure in determining whether or not

STARCH SYNTHESIS IN CEREAL GRAINS

19

amylopectin can crystallise (Waigh et al., 2000; Donald, 2001). In this model, the double-helices within clusters are likened to rigid units (mesogens) linked via flexible spacers to a flexible backbone (the region of high branch-density within the cluster and the inter-cluster chains, respectively). The model shows that to crystallise, the chains within the clusters of amylopectin must be long enough to form double helices and the ‘spacers’ must be sufficiently flexible to allow ordered packing. It is assumed that the existence of clusters within amylopectin explains why this molecule can pack to form semi-crystalline granules. Glycogen, like amylopectin, is a branched polymer of glucose but differs from amylopectin in that it does not contain clusters. Glycogen has chains that are on average, shorter than those in amylopectin and these are randomly, rather than periodically distributed within the molecule. Unlike amylopectin, glycogen does not crystallise. It can only form small, amorphous particles consisting, very often, of individual molecules. In order to form clusters, the amylopectin molecule must be synthesised with regions of high branch-point density separated by regions with relatively few branch points. Despite the progress being made in understanding the roles of the different enzymes involved in starch synthesis (see Section VIII), we still do not understand how this is achieved. Essentially, similar glucan synthases and branching enzymes are present in bacteria that make glycogen with no clusters and in plants that make amylopectin with clusters. It is generally assumed that some unique property of plant branching enzymes or starch synthases, or both, is responsible for the structure and formation of amylopectin clusters. However, we cannot rule out an important role for some as yet unknown protein or process within plants that is crucial to the synthesis of clusters.

C. AMYLOSE STRUCTURE

Amylose is a relatively infrequently branched molecule compared with amylopectin and is smaller in mass. It consists of between several hundred and a few thousand glucose residues, depending upon biological source. The location of amylose within the granule and in relation to amylopectin is not completely understood. Amylose molecules are distributed throughout the starch granule as single chains, not as double helices (Gidley, 1992) and are interspersed amongst the amylopectin molecules (Jane et al., 1992; Kasemsuwan and Jane, 1994). Amylose molecules in the centre of granules are larger but less concentrated than those at the periphery of the granule (Jane and Shen, 1993). Amylose is not crystalline in the granule even though

20

K. TOMLINSON and K. DENYER

purified amylose readily crystallises from solution (Lu et al., 1997). It is located within the granule in positions that allow ready access to the exterior of the granule. Most or all of the amylose can be leached away from the residual amylopectin component of the granule at temperatures just above that required for gelatinisation (Gidley, 1992). However, it is suggested that amylose exists within the semi-crystalline blocklets as well as in the amorphous regions between blocklets. Amylose appears to disrupt the packing of amylopectin double helices within the crystalline arrays (Jenkins and Donald, 1995). There are mutants of cereals and other plants that have no amylose in their starch granules. They are sometimes referred to as ‘waxy’ mutants after the waxy mutant of maize that lacks amylose in its seeds and has a waxy appearance. Despite the lack of amylose, waxy mutants have a nearnormal internal and external granule structure. This shows that amylose is not required for starch granule synthesis or structural integrity. Its role is therefore unclear. However, the fact that there is remarkably little variation in amylose content amongst wild-type plant storage organs suggests that there is some as yet unknown advantage to having approximately 25% of the mass of starch granule in the form of amylose.

VIII. ENZYMES INVOLVED IN STARCH BIOSYNTHESIS A. ADP GLUCOSE PYROPHOSPHORYLASE

The enzyme ADP glucose pyrophosphorylase (AGPase) converts glucose 1-phosphate and ATP to ADPG and pyrophosphate. The involvement of AGPase in starch synthesis is shown by mutants that have reduced AGPase activity. These are unable to synthesise a normal amount of starch. The maize mutants shrunken-2 and brittle-2 have less than 2% of the normal AGPase activity in the endosperm (Preiss et al., 1990) and consequently synthesise less than 25% of the normal amount of starch per grain (Tsai and Nelson, 1966; Singletary et al., 1997). Similarly, a mutant of barley, Risø 16 that has 31% of the normal AGPase activity synthesises less than 80% of the normal amount of starch (Johnson et al., 2003). In higher plants, AGPase is a multi-subunit enzyme consisting of two small and two large subunits (SSU and LSU, respectively) that are encoded by distinct genes (Preiss et al., 1991). There is evidence from random and site-directed mutagenesis of the subunits to suggest that they have distinct roles. In the absence of the LSU, the SSU has almost normal enzyme activity but is less sensitive to allosteric regulators than the heterotetrameric

STARCH SYNTHESIS IN CEREAL GRAINS

21

enzyme (Lin et al., 1988; Ballicora et al., 1995). In contrast, the LSU expressed in Escherichia coli in the absence of the SSU is unable to form an active enzyme (Ballicora et al., 1995). These results suggest that the SSU is the catalytic subunit and the LSU has primarily a regulatory role. However, this absolute division of labor between the subunits might be over-simplistic. Kavakli et al. (2001) provide evidence that the LSU may have a role in the binding of the substrate, glucose 1-phosphate. 1. The Location of AGPase One of the most interesting aspects of starch synthesis in cereals is that the location of AGPase in the cells of the endosperm is different from that in all other plant organs. Uniquely in cereal endosperms, AGPase occurs in both the plastids and the cytosol whereas in the other organs of cereals and in non-cereals, all the AGPase activity is plastidial. The cytosolic AGPase in cereals plays an important role in supplying ADPG for starch synthesis. Mutants that lack cytosolic AGPase activity but have normal plastidial AGPase activity have reduced starch contents (maize: Denyer et al., 1996a; barley: Johnson et al., 2003). This shows that the plastidial AGPase alone is not sufficient to support normal rates of starch synthesis. Evidence for the dual location of AGPase in cereal endosperms comes from plastid isolation experiments. In these experiments, the very delicate plastids are released from the endosperm cells by chopping the tissue with razor blades and then collected by centrifugation. Obviously, not all the plastids in the tissue are recovered in the pellet after centrifugation – some are not released by the chopping process and some break. However, those plastids that are recovered in the pellet are relatively free from cytosolic enzymes. By comparing the proportions of cytosolic enzymes and plastidial enzymes in the pellet and the supernatant with the proportions of AGPase in these fractions, the sub-cellular distribution of AGPase can be determined (provided that it is shown that there were no losses of enzyme activities during plastid isolation). If the AGPase is entirely plastidial it will show a similar distribution to the plastidial enzymes. This is observed in, for example, experiments with pea embryos (Denyer and Smith, 1988) and tomato fruits (Beckles et al., 2001a). However, in all cereal endosperms investigated in this way so far (barley: Thorbjornsen et al., 1996a; maize: Denyer et al., 1996a; rice: Sikka et al., 2001; wheat: Tetlow et al., 2003), the AGPase distribution was intermediate between those of the plastidial and cytosolic enzymes suggesting a dual location. These experiments can also give an estimate of the relative activity of the cytosolic and plastidial AGPase. This varies from species to species but in general, most of the

22

K. TOMLINSON and K. DENYER

activity in the endosperm is cytosolic (e.g., 95% of maize endosperm AGPase is cytosolic). Although the plastid isolation experiments discussed above provide good evidence for the dual location of AGPase in cereal endosperms, there is also some evidence against the cytosolic location of AGPase, at least in maize. Immunogold localisation studies (Miller and Chourey, 1995; Brangeon et al., 1997) did not reveal a cytosolic AGPase. The antibody labelled the interiors of starch granules and cell walls but not the cytosol. The reason for this discrepancy is not known. Immunogold-labelled pictures are very persuasive but they rely on the preservation of the antigenicity of the target protein and on the specificity of the antiserum. Thus, artifactual immunogold labelling is the most reasonable explanation for the discrepancy between these biochemical and immunogold studies. One of the theoretical consequences of synthesising ADPG in the cytosol as well as in the plastids is that the ratio of ADPG : UDPG is expected to be high because the reactions catalysed by AGPase and UDPglucose pyrophosphorylase will be coupled and close to equilibrium (Beckles et al., 2001b). Thus, a high ratio of ADPG : UDPG in a tissue may indicate that the tissue contains both a cytosolic and plastidial AGPase. Examination of a few organs where the AGPase location was known supported this idea. Screening the ADPG : UDPG ratio in a large number of plant organs, suggested that, in all the species of grasses examined, ADPG is synthesised in the cytosol of the endosperm cells as well as in the plastids whereas in other species and tissues, ADPG is synthesised exclusively in the plastids (Beckles et al., 2001b). 2. The Evolution of AGPase The cytosolic and plastidial forms of AGPase in the endosperm are encoded by different genes (Hannah et al., 2001, and references therein; Johnson et al., 2003). As only cereals possess a cytosolic form of AGPase, it is assumed that this enzyme evolved more recently than the plastidial AGPase that all plant species possess. The evolution of the cytosolic AGPase required the loss of an active transit peptide from both the SSU and the LSU. The genes encoding the cytosolic SSUs in cereals are very similar over most of their sequence to those encoding the plastidial subunits (Hannah et al., 2001; Johnson et al., 2003). Presumably, in the ancestor of the modern cereals, the gene encoding the cytosolic SSU arose by duplication and divergence of a gene encoding a plastidial SSU. This could have been a plastidial SSU expressed in the endosperm or elsewhere in the plant. It is likely that the cytosolic LSU evolved from a plastidial LSU in a similar manner. However, for most cereal species, there is insufficient information at present to

STARCH SYNTHESIS IN CEREAL GRAINS

23

distinguish clearly the genes encoding the cytosolic LSU from those encoding plastidial LSU. Until we can match LSU genes to the proteins they encode, it is not possible either to support or refute this hypothesis. The evolution of the cytosolic SSU in cereal endosperms has involved the acquisition of a first exon different from that encoding the plastidial SSU. The first exon of the plastidial SSU gene contains a sequence encoding a transit peptide whereas the first exon of the cytosolic SSU lacks such a sequence (Thorbjornsen et al., 1996b; Hannah et al., 2001; Sikka et al., 2001; Johnson et al., 2003). In maize, for example, the first exon of the plastidial SSU encodes the transit peptide and the N-terminus of the mature protein whereas, the first exon of the cytosolic SSU is shorter than that of the plastidial SSU and quite different in sequence. It is difficult to imagine that the unique first exon of the cytosolic SSU could have evolved from an ancestral plastidial gene. In contrast, the rest of the cytosolic and plastidial SSU genes are very similar in sequence. It is possible that the first exon of the cytosolic SSU represents a piece of DNA that was recruited from elsewhere in the genome. The first exons of the cytosolic SSUs are very divergent between species. The first exon of the cytosolic SSU from maize, for example, is strikingly different from that of barley, wheat and rice (Hannah et al., 2001). This raises the possibility that the cytosolic SSU in cereal endosperms may have evolved more than once during the evolution of our modern cereals. Perhaps the plastidial SSU duplicated and then, independently in the ancestors of maize and other cereals, a new first exon was acquired that did not encode a transit peptide. In barley and wheat, a single gene encodes two AGPase SSU transcripts by the alternative use of two first exons (Thorbjornsen et al., 1996b; Burton et al., 2002b). One of these transcripts encodes the cytosolic SSU of AGPase and the other potentially encodes a plastidial SSU although no protein corresponding to this transcript has been identified so far (Johnson et al., 2003). Instead, as in other cereal species, there is a second, separate gene in barley and wheat that encodes the plastidial AGPase SSU (Burton et al., 2002b; Johnson et al., 2003). The nature and origin of the gene encoding the cytosolic SSU and a second, alternative transcript is intriguing. It could possibly represent a cytosolic SSU gene that is not fully duplicated or diverged from the ancestral plastidial SSU gene. B. THE PLASTIDIAL ADPG TRANSPORTER

As discussed above, the production of cytosolic ADPG via the cytosolic AGPase evolved uniquely in the ancestors of the grasses. This pathway

24

K. TOMLINSON and K. DENYER

requires both a cytosolic AGPase and the means to import ADPG into plastids. The transporter protein responsible for the movement of ADPG from the cytosol into the plastids is believed to be an inner envelope protein that, in maize is encoded by the Brittle-1 gene (Cao et al., 1995; Sullivan and Kaneko, 1995). Investigation of the Brittle-1 mutant of maize clearly implicates a 34–44 kDa membrane protein as being necessary for ADPG transport (Shannon et al., 1998 and references therein). The Brittle-1 protein is capable of transporting ADPG from the cytosol into the plastid (Shannon et al., 1998) probably in exchange for AMP (Mo¨hlmann et al., 1997; Emes et al., 2003). Proteins homologous to Brittle-1 are present in the plastid envelopes of developing endosperm of wheat (Andon et al., 2002) and barley (our own unpublished results). However, genes encoding proteins similar in sequence to Brittle-1 are also found in non-cereals (e.g. Arabidopsis thaliana, accession number AL034567). Although the sub-cellular location of these in most non-cereal species is unknown, Bt-1–like proteins have been found in the chloroplast membranes of pea leaves (Peltier et al., 2000) and amyloplasts isolated from cultured cells of sycamore trees have been found to competitively transport ADPG, ATP, ADP and AMP (Pozueta-Romero et al., 1991). This raises the possibility that Bt-1-like proteins that are capable of transporting ADPG may be universally present in the plastid envelopes of plants. The function of these in tissues other than cereal endosperms is not clear. It was suggested that sucrose synthase could synthesise ADPG as well as UDPG in the cytosol and that this might represent an alternative source of ADPG for starch synthesis (BarojaFerna´ndez et al., 2001). However, there is evidence to suggest that in the presence of both UDP and ADP, sucrose synthase preferentially synthesises UDPG and not ADPG (for reviews see Pontis, 1977; ap Rees, 1995). Thus, it is not clear whether there is a cytosolic pool of ADPG in organs other than cereal endosperm. It is intriguing to speculate in what order the genes required to encode the cytosolic AGPase and the ADPG transporter evolved, when neither protein alone would theoretically confer any advantage to the plant. It is difficult to imagine that there would be a strong evolutionary pressure towards selection of a cytosolic AGPase, requiring two new genes, in the absence of an ADPG transporter. The evolution of the cytosolic AGPase makes more sense, however, if the ancestor of the grasses, perhaps like all plants, already possessed a protein in the plastidial envelope capable of transporting ADPG. Discovery of the role of the Brittle-1-like proteins in non-cereals may shed light on this evolutionary question.

STARCH SYNTHESIS IN CEREAL GRAINS

25

C. STARCH SYNTHASE

Starch synthase (SS) elongates glucan polymers by the sequential addition of single glucose moieties from ADPG. It is generally assumed that the enzyme adds these glucose moieties to the non-reducing end of the chain. However, it has been suggested recently that rather than the non-reducing ends, SS might add glucose units to the reducing ends (Mukerjea et al., 2002). In the proposed mechanism, an SS protein is covalently bound to the reducing end of the amylopectin chain. During the elongation process, the covalently bound SS is displaced by another SS protein that is carrying the next glucose unit to be added to the chain. However, elongation of the single reducing end of an amylopectin molecule rather than the many non-reducing ends of the constituent chains seems unlikely on theoretical grounds and is not supported by other experimental evidence. An isoform of starch synthase (SSII – see below) when acting on a chemically synthesised branched maltooligosaccharide, specifically elongates one of the two available non-reducing ends and not the reducing end (Damager et al., 2003). This suggests that for at least one isoform of SS, elongation occurs at the non-reducing ends of glucan chains. SS exists in all plants in multiple isoforms. For example, there are genes encoding five SS isoforms in the Arabidopsis thaliana genome (GBSSI, SSI, SSII, SSIII, SSIV). Sequence comparison suggests that these genes represent at least four different classes of starch synthase. SSIII and SSIV are similar in sequence and may represent a single class of isoform. All of these four classes of SS are expressed in the endosperms of both maize (Klo¨sgen et al., 1986; Gao et al., 1998; Harn et al., 1998; Knight et al., 1998) and wheat (Clark et al., 1991; Li et al., 1999a,b, 2000). The evidence available suggests that all four classes are probably present in all starch-synthesising cells (Smith, 1999) of plants although the relative contributions of the isoforms to the total starch synthase activity may vary between organs and species. All classes of SS have conserved functional domains towards the C-terminal end of the protein (Fig. 8) that are also found in bacterial glycogen synthase (GS). Thus, GS and SS probably share a common ancestral gene. The C-terminal domains common to all SS isoforms (Cao et al., 1999) include two putative ADPG binding sites (characterised by a ‘GGL’ motif) and a glucosyl transferase domain that is also found in other types of enzymes such as sucrose synthase and sucrose phosphate synthase (Denyer et al., 2001). The N-terminal regions of the different classes of SS isoform vary in length. For each class of SS, the amino acid sequence of the N-terminal region also varies between species. Removal of the N-terminal region does not inactivate the enzymes suggesting that it does not have a

26

K. TOMLINSON and K. DENYER

vital role in catalysis (e.g., maize SSI, Imparl-Radosevich et al., 1998). Its function is presently unknown but it may have a regulatory role possibly via interaction with other proteins or in determining substrate specificity (see below). The nature of the evolution of the four classes of SS from the ancestral GS-like gene is difficult to guess. Presumably the ancestral gene duplicated several times and these duplicated forms diverged to result in the multiple SS isoforms that we see in plants today. Multiple forms of SS homologous to at least some of the classes of SS seen in plants are also present in the green alga Chlamydomonas rienhardtii, suggesting that the divergence of the different classes of SS within starch-synthesising organisms was a very early evolutionary event (Wattebled et al., 2002). The early divergence of the different classes of SS is also supported by comparison of the structures of the genes encoding GBSSI, SSI, SSII and SSIII. There are differences between these genes in the positions of the introns and in the locations of conserved motifs within the C-terminal catalytic domain in wheat (Morell et al., 2001). Comparison of SSII genes in wheat and Arabidopsis thaliana suggests that this class of isoform was established at least 100 million years ago, prior to the separation of the monocotyledonous and dicotyledonous groups of plants. The exon sequences and the positioning of introns in the C-terminal region of the SSIII genes in these two species are conserved (Morell et al., 2001). Analysis of sequences encoding glycogen and starch synthases shows that SSIII is more like GS than are the other forms of SS (Li et al., 2000). This may indicate that SSIII evolved prior to the evolution of other forms of SS. In cereals, most of the classes of starch synthase are encoded by multiple genes that are expressed in different tissues. For example, in maize there are two SSII genes. SSIIa is expressed most strongly in the endosperm and SSIIb in the leaves (Harn et al., 1998). There are also multiple, tissue-specific genes encoding GBSSI. In wheat, GBSSI is expressed in the endosperm and in pollen grains and a second GBSSI gene is expressed in the pericarp but not in the endosperm or pollen (Vrinten and Nakamura, 2000). It is expected that these different, tissue-specific isoforms have broadly similar roles in starch synthesis but the subtle differences in their activities cannot be excluded. 1. The Roles of SSI, SSII and SSIII The conservation between species (cereals and non-cereals) of the various classes of isoform of SS at the amino acid level suggests that each has a unique and conserved role in starch synthesis. Studies of mutant and transgenic plants have demonstrated that GBSSI has a unique role in

STARCH SYNTHESIS IN CEREAL GRAINS

27

amylose synthesis that is conserved between species (for a review see Smith et al., 2001). Similarly, evidence is accumulating from studies of mutants and transgenic plants lacking particular isoforms of SS and also from studies of the kinetic properties of the isolated enzymes that SSI, SSII and SSIII make distinct contributions to the synthesis of amylopectin. The different classes of SS synthesise chains of different lengths. SSI from maize and rice preferentially synthesises very short amylopectin chains. This was shown by biochemical studies of maize SSI (expressed in E. coli and purified prior to analysis; Commuri and Keeling, 2001) and by recent studies of a mutant of rice lacking SSI due to the insertion of a retro-transposon into the gene encoding SSI (Nakamura, 2002). In contrast, SSII synthesises chains of intermediate length. Mutants of pea (Craig et al., 1998) and rice (Umemoto et al., 2002) and transgenic potatoes (Edwards et al., 1999) that lack SSII have reduced amounts of these chains of intermediate length (pea: dp10–20; rice: dp15–25; potatoes: dp13–25). SSIII is responsible for the synthesis of amylopectin chains as long or longer than those made by SSII. The dull1 mutant of maize that lacks SSIII activity (Gao et al., 1998) and transgenic potatoes that have reduced SSIII activity (Fulton et al., 2002), have decreased amounts of long chains (dp > 17) (Wang et al., 1993b; Fulton et al., 2002). The chain-length preferences of the isoforms of SS may in part be determined by the lengths of their N-terminal domains. Although the sequences of these domains are not conserved between species, their length is roughly conserved. The N-terminal domains of the SSIII class are the longest, those of SSII are intermediate in length and those of SSI are the shortest (Fig. 8). It may be no coincidence that the lengths of the N-terminal

Fig. 8. Isoforms of starch synthase from Arabidopsis. The bars represent the amino acid sequences of the isoforms of starch synthase from Arabidopsis. The length of the bar is proportional to the number of amino acid residues. Comparison of the amino acid sequences shows that they share a common domain (shown in black). The N-terminal regions (left) vary in size and sequence between isoforms and are shown in various shades of gray.

28

K. TOMLINSON and K. DENYER

domains vary in a similar manner to the chain lengths preferentially synthesised by the different isoforms of SS. Thus, the isoform with the shortest N-terminal extension, SSI makes the shortest amylopectin chains. We hypothesise that the N-terminal domain functions as a kind of ‘molecular ruler’, guiding the SS to substrate-chains of particular lengths. In the future, comparison of the products synthesised by full-length and N-terminally truncated forms of SS will help to determine whether this view of the role of the N-terminal domains is correct. 2. The Role of GBSSI Amylose is synthesised by a specific isoform of starch synthase, GBSSI. This is shown by mutants, such as the waxy mutants of cereals that lack amylose (see Section VII). All such mutants lack GBSSI activity (Smith et al., 2001). Unlike other isoforms of starch synthase that exist either partly or completely in the soluble phase of the plastid, GBSSI exists almost entirely within the granule. Investigation of the location of GBSSI within potato starch granules suggests that it is concentrated into discrete internal concentric rings (Han and Hamaker, 2002). Similar rings were also observed with maize starch whereas with wheat A-type starch, GBSSI was located mostly in the peripheral region of the granules. The rings of GBSSI are probably related to the growth ring structure of the granules, although it is not clear from these studies whether there is more GBSSI in the semicrystalline or in the amorphous layers. The discrete localisation of GBSSI within the granule suggests that amylose may also be synthesised, and may therefore accumulate, in discrete regions of the granule. However, as described in Section VII.C, the available evidence suggests that amylose molecules are interspersed amongst amylopectin molecules (Jane et al., 1992; Kasemsuwan and Jane, 1994). Analysis of granule permeability shows that granules are freely permeable to small molecules such as the substrate for amylose synthesis, ADPG. Therefore, amylose synthesis should be possible deep within the granule. This appears to occur in transgenic plants with reduced activities of GBSSI (Kuipers et al., 1994; Tatge et al., 1999). However, in wild-type plants, amylose synthesis probably occurs mostly just underneath the granule surface within a framework of the most recently deposited amylopectin. GBSSI is able to synthesise very long chains. These typify the amylose molecule but are also synthesised by GBSSI in amylopectin (Baba et al., 1987; Delrue et al., 1992; Denyer et al., 1996b; Fulton et al., 2002). Some of the unique properties of GBSSI which allow this isoform, but not others, to synthesise long chains have been discovered. The enzyme is able to add several glucose molecules to the growing amylose molecule before releasing

STARCH SYNTHESIS IN CEREAL GRAINS

29

the chain and re-binding to another. This processive elongation, as it is called, will favour the synthesis of long chains. Other isoforms of starch synthase that have been investigated (Denyer et al., 1999a) do not elongate processively. Instead they add a single glucose molecule per chain before dissociating from this product. Due to this distributive mode of action, and the abundance of possible substrate chains, in each successive round of polymerisation, a different chain will be elongated. Distributive elongation will therefore tend to result in the synthesis of short chains. The domains and motifs within the GBSSI protein that are responsible for its unique properties – such as the processive mode of action and tight binding to starch granules – are not yet known. 3. Competition between the Amylose and Amylopectin Pathways As described above, the starch-synthetic enzymes make their contributions largely to the synthesis of either amylose or amylopectin but there is also some overlap in the contributions of the enzymes to the synthesis of the two starch polymers. GBSSI is known to synthesise amylose but it is also responsible for the extra-long chain component of amylopectin (see above). Similarly, SBE is involved primarily in amylopectin synthesis but it is assumed that the same isoforms of SBE are also responsible for the branching of amylose (Takeda et al., 1990). For this reason, the synthesis of amylose influences amylopectin synthesis and vice versa. This interaction leads to competition between amylose and amylopectin synthesis. The nature of this competition and how it might occur will be described below. There is evidence that, in the wild-type, amylose is synthesised at the expense of amylopectin. This is provided by studies of amylose-free mutants such as the waxy mutants of cereals. Mutants that lack GBSSI, and therefore have no amylose, have instead more than the normal amount of amylopectin. In fact, the amylose is replaced by an almost equivalent amount of amylopectin such that the starch content of these mutants is near-normal. The starch granules of waxy mutants are also approximately the same size and shape as those in the wild-type (Jane et al., 1994; Yoo and Jane, 2002b). In the absence of amylose, the amylopectin molecules in amylose-free (waxy) mutants of wheat are larger in mass but they are more compact, as indicated by their gyration radius, than those in the wild-type (Yoo and Jane, 2002a). The increase in molecular weight of amylopectin entirely accounts for the increase in amylopectin content in amylose-free wheat (Yoo and Jane, 2002b). This shows that the number of amylopectin molecules in the granule is largely unaffected by the presence or absence of amylose.

30

K. TOMLINSON and K. DENYER

It is, therefore, unlikely that there is competition between the pathways of amylose and amylopectin synthesis at the level of polymer initiation. There is also evidence that amylose synthesis is limited by amylopectin synthesis. Mutations in amylopectin-synthesising enzymes that reduce amylopectin accumulation can, at least in some circumstances result in an increase in amylose content. The amylose content is increased, not just as a proportion of the starch, but in terms of absolute amounts per seed. For example, the ae mutant of maize that has a reduced activity of starchbranching enzyme has a lower than normal amylopectin content (47 mg per endosperm instead of the normal 92 mg; Singletary et al., 1997). The amylose content of this mutant is increased (40 mg per endosperm instead of 33 mg; Wang et al., 1993b). However, the effect on amylose synthesis of mutations in the amylopectin-synthesising enzymes are variable and difficult to interpret. First, some mutants with reduced amylopectin content do not have more amylose than normal. For example, the dull mutant of maize that lacks SSIII and has a lower than normal amylopectin content, has an increased proportion of amylose (31% amylose instead of 27%) but the absolute amount of amylose per grain is only slightly changed (32 mg per endosperm in the mutant and 33 mg in the wild-type; Wang et al., 1993b). Second, accurate measurements of the amounts of amylose and amylopectin are difficult to achieve for mutant starches with altered amylose and amylopectin structures. For example, the presence of intermediate material or abundant extra-long amylopectin chains in some mutant starches can interfere with amylose and amylopectin assays (discussed in Wang et al., 1993b). There are several possible mechanisms to explain the observed competition between amylose and amylopectin synthesis. First, amylose- and amylopectin-synthesising pathways may compete for the common substrate, ADPG. It is known that, in some circumstances, the availability of ADPG can influence the relative amounts of amylose and amylopectin that are synthesised. In mutants that have a reduced capacity to synthesise ADPG, the proportion of amylose in starch is lower than in wild-type plants (Clarke et al., 1999). This is thought to be due to the poor affinity of GBSSI for ADPG compared with the affinity of other isoforms of starch synthase. Thus, GBSSI is relatively more affected by a decline in ADPG concentration than are the amylopectin-synthesising enzymes. In a similar but opposite manner, if the ADPG concentration were to rise, as it might in a mutant with a lower rate of amylopectin synthesis, this could explain the increase in the rate of amylose synthesis that is observed. The increase in amylopectin synthesis observed in the absence of amylose is probably not caused by increasing ADPG concentration alone. For

STARCH SYNTHESIS IN CEREAL GRAINS

31

developing pea embryos, it was estimated that the ADPG concentration in wild-type plastids was near-saturating for the amylopectin-synthesising isoforms of starch synthase (Clarke et al., 1999). Therefore an increase in ADPG concentration in the amylose-free mutant would not be predicted to result in a substantial increase in the rate of amylopectin synthesis. There are very few estimates of ADPG concentrations in the plastids of any plants and these estimates are particularly difficult for cereal endosperms where ADPG exists in the cytosol as well as in the plastids. Thus, these ideas have not yet been tested in cereals. Second, the inhibitory effect of amylose synthesis on amylopectin synthesis could be due to disruption of the packing of the amylopectin chains by amylose or the extra-long chains of amylopectin that are synthesised by GBSSI. As described above, in the presence of amylose, the amylopectin molecules are smaller and more loosely packed. Thus, the products of GBSSI may physically interfere with amylopectin synthesis. Third, the long chains synthesised by GBSSI could directly inhibit the activity of the amylopectin-synthesising isoforms of starch synthase. Commuri and Keeling (2001) noticed that SSI binds very tightly to long chains of amylopectin that are ineffective as substrates. It is not known whether other amylopectin-synthesising isoforms of starch synthase also bind to amylose in a similarly unproductive manner. The competitive inhibition of SS by the long chains of amylopectin or amylose would reduce the activity of SS available for polymer elongation. This could explain the reduced rate of amylopectin synthesis seen in wild-type plants containing an active GBSSI compared with the rate in waxy mutants. Finally, interactions between GBSSI and amylopectin influence the ability of this enzyme to synthesise amylose (Denyer et al., 1999b). This suggests that amylose synthesis may be dependent on the prior synthesis of an amylopectin framework in which to house GBSSI. The dependence of amylose synthesis on the synthesis of an amylopectin framework may account for the upper limit (approximately 33%) on the amount of amylose synthesised in wild-type starch granules. There may be a limit on the available space within the framework for amylose synthesis. This might also explain why amylose synthesis lags behind the synthesis of amylopectin (see Fig. 3c).

D. STARCH-BRANCHING ENZYME

Starch-branching enzyme (SBE) cuts an (1,4) linkage between glucose units in a chain and then adds the severed chain via an (1,6) linkage to

32

K. TOMLINSON and K. DENYER

another chain. The nature of the substrate for SBE is not entirely clear. The substrate for SBE could be a single chain. In this case, the cutting and joining of chains would be spatially and temporally separate parts of the reaction mechanism. This two-step mechanism for SBE is implicit in many of the current models of amylopectin synthesis (e.g., Nakamura, 2002). However, there is some evidence that the substrate for SBE might be a double helix formed by a pair of adjacent amylopectin chains (Borovsky et al., 1979). In this case, SBE cuts one of the chains in the double helix and then transfers the cut end to the adjacent chain (Fig. 9). Multiple cutting and joining along a double helix would result in a cluster of branch points

Fig. 9. Modes of action of SBE on amylopectin as postulated by Borovsky et al. (1976) and Borovsky et al. (1979). The glucan chains are shown as lines, the nonreducing ends as arrow heads, the 1,6-linkages (branch-points) as dots and starchbranching enzyme as a circle labeled: SBE. (A) Two step mechanism. (i) The substrates are two single glucan chains. (ii) SBE cuts the donor chain and (iii) joins a section of this to the acceptor chain. (B) One step mechanism. (i) The substrate is a double-helix of glucan chains. (ii) Multiple attack of the helix by SBE (iii) results in the introduction of several branch points. Either chain can act as acceptor or donor. (iv) the product shown in (iii) depicted in exploded form.

STARCH SYNTHESIS IN CEREAL GRAINS

33

subtending short chains. Such a structure is reminiscent of the cluster of branch points that have been proposed to exist within amylopectin prior to elongation of the chains by starch synthase (e.g., Myers et al., 2000). This mechanism of action, being dependent on the prior formation of doublehelices within the substrate, is consistent with the low optimum temperature of the SBE reaction in vitro (Borovsky et al., 1975) and the requirement for substrate chains longer than the minimum required to form a stable double helix (Borovsky et al., 1976). Based on sequence comparisons, there are two types of SBE (SBEI or SBEB and SBEII or SBEA) in cereal endosperms and probably in all starchsynthesising cells (Burton et al., 1995; Morell et al., 1997). In cereals, SBEI and SBEII are expressed at different times during endosperm development. For example, in developing wheat endosperm, SBEI is preferentially expressed late in development whereas the SBEII isoform accounts for most of the SBE expression in young endosperms (Morell et al., 1997). In cereals, there are two forms of SBEII, SBEIIa and SBEIIb. Whilst sequences closely related to SBEIIb of cereals are also found in dicotyledonous species, the SBEIIa isoform may be specific to monocotyledonous species (Blauth et al., 2001). In maize, barley and rice, SBEIIb is endosperm-specific but SBEIIa is also expressed in other parts of the plant (Gao et al., 1997; Sun et al., 1998; Nakamura, 2002). In the endosperm of wheat and other members of the tribe Triticeae, there is an additional form of SBE (SBEIc) of unusually large molecular size (140–145 kDa compared with 90–100 kDa for SBE I and II; Peng et al., 2000). This protein appears to be more abundant on A-type starch granules than on B-type granules and this has lead to the suggestion that this form of SBE may in some way determine granule morphology in the Triticeae (Peng et al., 2000). Confirmation of this awaits the results of transformation experiments in which this isoform is eliminated. Biochemical studies of isolated isoforms of SBE show that these differ significantly in their properties, with the SBEI preferentially transferring longer chains than SBEII (Guan and Preiss, 1993; Takeda et al., 1993a). Maize endosperm SBEIIa and SBEIIb have similar but not identical properties (reviewed in Preiss et al., 1991). The differences between SBEI and SBEII with respect to substrate preference and products suggest that they may have different roles in the synthesis of amylopectin. Based on this hypothesis, it has been proposed that branching (Guan and Preiss, 1993) and the synthesis of the clusters within amylopectin (Nakamura, 2002) may occur in two phases. First, SBEI synthesises a lightly branched region of the amylopectin cluster that is destined to lie mainly within the amorphous lamella. Second, the products of SBEI serve as the substrates

34

K. TOMLINSON and K. DENYER

for the combined actions of starch synthases and SBEII. These synthesise a highly branched region of the amylopectin cluster that will become the crystalline lamella. Mutants that lack SBEIIb provide evidence in support of the above model. For example, the amylose-extender mutants of maize (Stinard et al., 1993) and rice (Nishi et al., 2001) that lack SBEIIb, accumulate amylopectin with fewer branches and fewer short chains (Takeda et al., 1993b; Nishi et al., 2001). However, mutants lacking either SBEI or SBEIIa do not in all cases provide evidence for the suggested roles of these isoforms. Whilst mutants of rice that lack SBEI accumulate amylopectin that has fewer intermediate and long chains (Nakamura, 2002), SBEI mutants of maize have amylopectin structures that are indistinguishable from wild-type starch (Blauth et al., 2002). Mutants of maize and rice that lack SBEIIa also have essentially normal amylopectin structure (Blauth et al., 2001). An additional problem with the use of mutants to deduce the roles of the isoforms of SBE is that frequently, there may be pleiotropic effects of the mutations on the activities of other starch-biosynthetic enzymes. For example, rice mutants lacking SBEIIb also have reduced (approximately 50%) activity of SS in the endosperm (Nishi et al., 2001). Therefore, the changes in amylopectin structure might not be attributable only to changes in the nature and amount of SBE. An alternative approach to discover the roles of the different isoforms of SBE is to express them individually or in various combinations in heterologous systems such as Escherichia coli (Guan et al., 1995) or Saccharomyces cerevisiae (Seo et al., 2002). Two important conclusions can be drawn from this work so far. First, expression of SBEI and SBEII in these heterologous systems does not result in the production of a semicrystalline polymer with a polymodal distribution of chain lengths like amylopectin but gives a soluble polymer with a uniform distribution of chain lengths more like glycogen. Therefore, the properties of SBE alone are not sufficient to determine the polymodal distribution of chain lengths characteristic of amylopectin. A similar conclusion was also reached from earlier work on SBE in Pisum sativum L. (Tomlinson et al., 1997). Second, SBEI alone did not support significant glucan synthesis but was effective in combination with both SBEIIa and SBEIIb (Seo et al., 2002). This result adds some weight to the two-phase model described above in that it suggests that SBEI and SBEII act synergistically. However, unlike the above model, Seo et al. (2002) suggested that their data could be interpreted to mean that SBEII activity was needed to generate the substrate for SBEI rather than the other way round. Very clear evidence of the synergistic interaction between SBEI and SBEII is also provided by

STARCH SYNTHESIS IN CEREAL GRAINS

35

studies of transgenic potato tubers (Schwall et al., 2000 and references therein). E. ISOAMYLASE

Isoamylase is a starch-debranching enzyme that cuts the (1,6) linkages within amylopectin. Studies of mutants of several species that lack isoamylase, have shown that this enzyme is required for normal starch synthesis. In cereals, isoamylase mutants exist in maize (sugary-1; James et al., 1995), rice (sugary-1; Kubo et al., 1999) and barley (Riso17, Notch-2; Burton et al., 2002a). In the absence of isoamylase, reduced amounts of starch, often in the form of more numerous, smaller granules are commonly observed, as described above (see Section IX.C). It seems likely that plants generally possess genes encoding three classes of isoamylase, isa-1, isa-2 and isa-3 (Hussain et al., 2003). In all of the cereal isoamylase mutants mentioned above, the isoform affected directly by the mutation is encoded by isa-1. We do not yet know the relative contributions of the three isoamylase genes to the total isoamylase activity in cereal endosperms. The isoamylase activity in the endosperms of the sugary mutants is reduced to undetectable levels (as judged by assays based on native gel electrophoresis). Taken at face value, this would suggest that isa-1 accounts for most or all of the isoamylase activity in the endosperm. However, there is evidence from work on the isoamylase isoforms of potato that the isoforms encoded by both isa-1 and isa-2 are required for isoamylase activity. If this is also true for cereals, then, in the absence of isa-1, isa-2 would be predicted to be inactive. Isoamylase plays a unique role in starch synthesis. Its role is different from that of limit dextrinase (or the pullulanase-like debranching enzyme), which is another type of starch-debranching enzyme present in starchsynthesising cells. Both these enzymes are able to carry out essentially the same debranching reaction with amylopectin as the substrate. However, mutants of maize that lack limit dextrinase have normal starch contents and do not accumulate phytoglycogen (Dinges et al., 2003). Also, limit dextrinase activity cannot substitute for the lack of isoamylase in isoamylase-deficient mutants (Zeeman et al., 1998; Burton et al., 2002a). In addition to a reduction in starch synthesis, a soluble form of starch called phytoglycogen is often, but not always synthesised in isoamylasedeficient cells. In fact, the relative amounts of phytoglycogen and starch that accumulate in these mutant cells varies enormously between species, between cells within the same endosperm and even between plastids within the same cell. In the rice sugary-1 mutant, EM-914, which has no detectable

36

K. TOMLINSON and K. DENYER

isoamylase protein, there is phytoglycogen but almost no starch (Nakamura et al., 1997; Kubo et al., 1999). In the maize mutant, su1-R4582::Mu1, a null mutant for isoamylase, the phytoglycogen content is more than four times greater than the starch content (Dinges et al., 2001). In contrast, in the null isoamylase mutants of barley, Riso17 and Notch-2, the phytoglycogen content is less than the starch content (Burton et al., 2002a). In these barley isoamylase mutants, most of the cells in the endosperm contain both starch and phytoglycogen although adjacent plastids can vary enormously in their glucan composition. However, in sugary-1 mutants of rice with reduced isoamylase activity, such as EM-41, phytoglycogen accumulates in the cells in the middle of the endosperm whilst cells at the edge of the endosperm contain some starch but little, if any, phytoglycogen (Kubo et al., 1999). These data suggest that, whilst the lack of isa-1 predisposes the cell to make phytoglycogen, it does not make phytoglycogen synthesis inevitable. The isa-1 mutants of cereals suggest that the lack of isoamylase may be necessary, but not on its own sufficient, to trigger phytoglycogen synthesis. The nature of the enzymatic conditions required in addition to the lack of isoamylase to stimulate phytoglycogen synthesis are unknown. Only one other mutant, the sugary-3 of maize, is known to accumulate a soluble glucan similar to phytoglycogen (Stinard, 1992; Pan et al., 1995). The su-3 mutant is affected in a gene other than that encoding isa-1, but its nature is not yet known. Elucidation of the nature of the enzyme(s) affected in su-3 may shed light on the conditions required for phytoglycogen synthesis. Various theories have been proposed to explain the role of isoamylase in starch and phytoglycogen synthesis. Summaries of these have been presented elsewhere (Smith, 1999; Kossmann and Lloyd, 2000). Briefly, isoamylase could take part in the determination of the structure of amylopectin by trimming away some of the branches made by starchbranching enzyme (Ball et al., 1996; Mouille et al., 1996; Myers et al., 2000; Nakamura, 2002). It could help to destroy soluble glucans that would otherwise accumulate at the expense of starch granules (Zeeman et al., 1998). It could be involved in the initiation of starch granules and glucan particles and influence the number of these that form per plastid (Burton et al., 2002a). These ideas are not mutually exclusive and further investigation is required to reveal completely the role of isoamylase in starch synthesis. F. OTHER ENZYMES INVOLVED IN STARCH SYNTHESIS

In addition to the enzymes described above, it is likely that other enzymes are involved in the synthesis of starch granules. The starch-synthesising

STARCH SYNTHESIS IN CEREAL GRAINS

37

plastid contains enzymes that were traditionally regarded as starchdegrading enzymes. These include amylase, disproportionating enzyme, starch phosphorylase and limit dextrinase (see reviews by Smith, 1999; Kossmann and Lloyd, 2000). The roles of these enzymes in starch synthesis, particularly in cereal endosperms, are largely unknown. Like isoamylase, they may be required to degrade soluble starch polymers that might otherwise accumulate at the expense of starch granules (Zeeman et al., 1998). Alternatively, or in addition, degradative enzymes may be directly involved in the synthesis of a crystallisation-competent amylopectin molecule. Further studies of the roles of starch-degrading enzymes in starch synthesis are underway and will help to resolve these issues.

IX. GRANULE AND POLYMER INITIATION A. AMYLOPECTIN INITIATION

Our understanding of the initiation or ‘priming’ of amylopectin molecules is very limited and lags far behind the understanding of similar processes for other macromolecules such as glycogen (Roach and Skurat, 1997) or cellulose (Peng et al., 2002). In the case of glycogen, which is a molecule very similar to amylopectin, a protein primer called glycogenin is required for initiation in animals and yeast (Smythe and Cohen, 1991). For reviews of its properties see Alonso et al. (1995) and Roach and Skurat (1997). Glycogenin is able to glucosylate itself creating a short chain of -1,4-linked glucose residues covalently linked to the glycogenin molecule via a tyrosine residue. The hunt for a similar self-glucosylating proteins in plants that might be responsible for the priming of amylopectin initially looked promising and various candidate protein primers were identified. These were called amylogenin (Singh et al., 1995), reversibly glucosylated polypeptide (RGP; Langeveld et al., 2002) or UDP-glucose:protein transglucosylase (UPTG; Ardila and Tandecarz, 1992). However, the available evidence suggests that these proteins are involved in the synthesis in the endoplasmic reticulum of polysaccharide components of the cell wall rather than in amylopectin synthesis in the plastids (Dhugga et al., 1997; Delgado et al., 1998; Langeveld et al., 2002). Thus, no amylopectin-priming protein has yet been identified. B. AMYLOSE INITIATION

In addition to the possible involvement of a priming protein as described above for amylopectin initiation, two other models of amylose priming

38

K. TOMLINSON and K. DENYER

have been proposed. First, amylose may be synthesised from short malto-oligosaccharide (MOS) precursors (Denyer et al., 1996b). These MOS may in turn be produced by the action of starch-degrading enzymes on amylopectin or on soluble glucans in the plastid stroma. The synthesis of amylose-like molecules from MOS primers has been observed in vitro (Baba et al., 1987; Denyer et al., 1996b, 1999a; Van de Wal et al., 1998; Zeeman et al., 2002) and evidence for their involvement in amylose synthesis in vivo comes from work on starch synthesis in Arabidopsis thaliana leaves (Zeeman et al., 2002). A mutant of Arabidopsis thaliana (dpe1, lacking disproportionation enzyme activity) that accumulates higher than normal amounts of MOS, also synthesises more amylose than the wild-type. Second, the primer for amylose might be amylopectin (Ball et al., 1998; Van de Wal et al., 1998). Elongation of chains of amylopectin by GBSSI, followed by cleavage of these chains is suggested to give rise to amylose. The observed elongation of amylopectin in vitro (Baba et al., 1987; Denyer et al., 1996b, 1999a) and in vivo by GBSSI (Delrue et al., 1992; Fulton et al., 2002) is consistent with this idea. Long amylopectin chains from the starch of the green alga, Chlamydomonas reinhardtii, were cleaved in vitro by an unknown mechanism to form amylose (Van de Wal et al., 1998). However, there is no evidence for the operation of this mechanism in vivo. The synthesis of amylose from an amylopectin precursor was not observed in Arabidopsis thaliana (Zeeman et al., 2002).

C. GRANULE INITIATION

Almost nothing is known about the nature and control of granule initiation. It is also not clear whether the initiation of starch granules and the initiation of amylose and amylopectin polymers are aspects of the same or different processes. If a starch granule consists of a single, extremely large amylopectin molecule, then initiation of granules and of amylopectin would obviously be the same process for both. However, this seems very unlikely from estimates of the size of amylopectin molecules (Yoo and Jane, 2002b). An amylopectin molecule is much smaller than a starch granule suggesting that there are many amylopectin molecules per granule. This means that the initiation of the starch granule could be very different from amylopectin initiation. The initiation of one or many starch granules in the plastid stroma could be a physical process determined by the structure of the first-formed -glucans and the environment within the plastid. In vitro, low molecular weight linear or lightly branched starch polymers will, under certain

STARCH SYNTHESIS IN CEREAL GRAINS

39

conditions, spontaneously form spherulites with radially oriented crystalline lamellae (Nordmark and Zeigler, 2002). These spherulites are similar to starch granules in the orientation of the lamellae and in that both spherulites formed in vitro and native granules have at their core, small holes of similar dimensions. It is possible therefore, that -glucans in the plastid stroma crystallise to form spherulites and that these are then enlarged by the addition of amylopectin to form granules. Variations between species in the structure of the -glucans formed in early granule synthesis and/or the environment within the plastid could influence the number and nature of the spherulites that form per plastid. Recently, we found evidence for the involvement of a starch biosynthetic enzyme, isoamylase in the control of the number of starch granules per plastid (Burton et al., 2002a). Isoamylase is a starch-debranching enzyme. Its properties and role in starch synthesis was discussed in Section VIII.E. Lack of, or reduction in, isoamylase activity during starch synthesis leads to an increase in the number of granules that are initiated per plastid. This effect of isoamylase on granule number per plastid seems to be universal in plants. Increases in the number of granules per plastid have been observed in isoamylase mutants of cereals (sugary maize; Boyer et al., 1977; sugary rice: Kubo et al., 1999; barley: Burton et al., 2002a) and in transgenic potatoes with reduced isoamylase activity (AM Smith and C Martin, personal communication). In barley, which normally synthesises A- and B-type granules, lack of isoamylase leads to the synthesis of compound starch granules (Burton et al., 2002a). In rice, which normally synthesises compound granules, the lack of isoamylase leads to the production of ‘super-compound’ granules with an increased number of component granulae (Kubo et al., 1999). The role of isoamylase in granule initiation is not understood. It may affect granule initiation by directly influencing the structure of the firstformed -glucans in the plastid or by influencing the number of these glucans via an effect on their initiation. Bacterial isoamylase is able to remove the -glucan chain from glycogenin in vitro, thus rendering it incapable of priming glycogen synthesis (Lomako et al., 1992). If similar priming proteins exist in plants, they might also be subject to degradation by isoamylase. Therefore, an increase in the amount of glycogenin-like primer and in the number of glucans initiated would be expected in isa-1 mutants with reduced isoamylase activity. The total number of glucans initiated is greater in isa-1 mutants due to the synthesis of phytoglycogen in addition to amylopectin. The increase in glucan initiation may in turn lead to an increase in granule initiation, although how this occurs is not yet clear.

40

K. TOMLINSON and K. DENYER

X. THE CONTROL OF STARCH SYNTHESIS IN ENDOSPERMS In this section, we will consider the control of starch synthesis within the grain. First, we will discuss the control of metabolic pathways in general. This is necessary to provide a framework in which to understand the available information on the control of the rate of starch synthesis in endosperms. Second, we will highlight some aspects of the control of starch synthesis in cereal endosperms. In considering these, we are aware that the way in which the rate of starch synthesis in cereal endosperms is controlled probably varies between species, at different times in development, between different cells in the endosperm, in different environments and at different times during the day. Therefore, although we hope to identify general principles, we do not expect to be able to identify one, all-purpose description of the control of starch synthesis applicable to all endosperms and conditions. A. THE CONTROL OF METABOLIC PATHWAYS

The two terms ‘control’ and ‘regulation’ are often used interchangeably but have also been assigned precise and different meanings (Hofmeyr and Cornish-Bowden, 1991). Control refers to the quantitative effect that a change in enzyme activity has on the flux through the pathway. Regulation is a description of the mechanisms involved in altering enzyme activity. Enzyme activities can be regulated in different ways (Stitt, 1999). Coarse regulation involves changes in the amount of the enzyme protein due to changes in the rates of transcription and/or translation, or due to protein turnover. Fine regulation involves changes in the activity of a pre-existing enzyme due to changes in its substrates or effectors, or due to posttranslational modifications such as phosphorylation or reduction. Although control of a pathway involves the regulation of the activity of one or more enzymes, regulation of enzyme activity does not necessarily lead to an alteration in the flux through the pathway. Enzymes may also be regulated in order to maintain the concentrations of metabolite pools or to coordinate their activity with that of other enzymes elsewhere in the pathway in the absence of a change in flux. This means that enzymes of key importance in controlling the flux through a pathway are not necessarily those with regulatory properties. It has also been demonstrated that enzymes with no regulatory properties and with activities greatly in excess of the flux, can under some circumstances, exert a significant degree of control over the flux (see Stitt, 1999, for examples).

STARCH SYNTHESIS IN CEREAL GRAINS

41

A general theory describing how pathways are controlled was proposed by Kacser and Burns (1973). The quantitative methods of analysis that have been developed based upon this theory have come to be called metabolic control analysis. For descriptions of the theory and how to apply it to the study of the regulation of plant metabolism, we recommend Kacser (1987), ap Rees and Hill (1994) and Stitt (1999). The quantitative contribution of an enzyme to the control of the flux through the pathway can be measured and is called its flux control coefficient. A flux control coefficient reflects an enzyme’s capacity for coarse control. Generally speaking, an enzyme that is subject to fine regulation will have a lower flux control coefficient than an unregulated enzyme (Kacser, 1987; ap Rees and Hill, 1994). This is because changes in the amount of the enzyme-protein will have less impact on the flux when they are buffered by changes in the extent of fine regulation of the enzyme’s activity. Understanding the control of metabolism in these terms is necessary for a rational approach to the genetic engineering of metabolic pathways. Regulated enzymes with low flux control coefficients are not good targets for manipulation if that manipulation consists of changing the amount of the enzyme without changing its nature. This is because the regulatory mechanisms – allosteric effectors or covalent modifications – will over-ride alterations in enzyme concentration. However, enzymes with high flux control coefficients are good targets for manipulation in this way, although experience has shown that it is rare for a single enzyme in a pathway to exert the majority of the control. This means that to increase the flux through a pathway it may be necessary to increase the activity of several enzymes with high flux control coefficients. Regulated enzymes with low flux control coefficients are good targets for genetic engineering if the introduced gene encodes an enzyme with different regulatory properties to the endogenous enzyme. For example, substituting an unregulated enzyme for a regulated one may increase the flux through the pathway. This is because the unregulated enzyme will provide a bypass for a step in the pathway that could otherwise be limiting.

B. COARSE REGULATION OF STARCH SYNTHESIS

The amount of starch in the mature grain depends upon when starch synthesis starts, when it finishes and on the rate of starch synthesis during the period of maximal starch accumulation. Coarse control may be important in some or all of these processes. The maximum catalytic activity of enzymes involved in starch synthesis increases and then declines during

42

K. TOMLINSON and K. DENYER

grain development in a manner that approximately corresponds to the changing rate of starch synthesis. For example, in rice, the developmental changes in the activity of soluble SS, SBE and sucrose synthase correlate well (correlation coefficients between 0.69 and 0.84) with the measured changes in the rate of starch synthesis during the grain-filling period (Yang et al., 2002). Our measurements of enzyme activities in wheat show a similar correlation for starch synthase (Fig. 10). In contrast, the developmental changes in the activity of AGPase did not correlate with changes in the rate of starch synthesis in rice (Yang et al., 2002) or wheat (Fig. 10). This shows that there is coordinated coarse control of many, but not all, of the enzymes in the pathway. The activities of enzymes in the pathway of starch synthesis vary enormously relative to one another and to the rate of starch synthesis. In rice (Yang et al., 2002), maize (Singletary et al., 1994) and wheat (Jenner 1991; Hawker and Jenner, 1993), there is only just sufficient soluble SS activity to account for the rate of starch synthesis. In contrast, AGPase activity is present in excess (AGPase activity is 4–15-fold greater than the rate of starch accumulation). It is clear from this that in order to achieve

Fig. 10. Comparison of the rate of starch synthesis with the activity of starch synthase and AGPase. The rate of starch synthesis was calculated from measurement of starch content (data not shown). The generalised logistic plot of this rate is shown in A and B (crossed line; note different scales). (A) The rate of starch accumulation and the activity of starch synthase decline during development are quantitatively similar. (B) Between 10–30 days after flowing, the activity of AGPase is increasing and is far higher than the rate of accumulation of starch.

STARCH SYNTHESIS IN CEREAL GRAINS

43

any substantial increase in the rate of starch synthesis in the endosperm by genetic manipulation, it will be necessary to increase the activity of soluble SS. There is evidence for some species but not for others that coarse control of SS activity may account for the decline in the rate of starch synthesis at the end of the grain-filling period. In wheat, soluble SS activity declines in parallel with the decline in the rate of starch accumulation whilst the activity of AGPase remains unaltered (Hawker and Jenner, 1993) or declines later (Fig. 10). In maize, the accumulation of starch stops abruptly and does not correlate with changes in the activities of any of the enzymes in the pathway (Keeling, 1999). Thus, in maize there is no evidence that coarse control leads to the cessation of starch synthesis and the mechanism of regulation at this stage in development remains unknown.

C. FINE REGULATION OF STARCH SYNTHESIS

From a theoretical point of view, control of starch synthesis in cereal grains by fine regulation might be expected to be less important than it is in leaves. In photosynthetic cells, there is a need for sensitive and rapid modulation of the rate of starch synthesis during the day in response to the availability of carbon and the rate of sucrose synthesis and export. However, the supply of substrates for starch synthesis is likely to vary less rapidly and unpredictably in cereal grains than it is in leaves. Control of starch synthesis in grains at the developmental level by coarse regulation may therefore be more important. Consistent with these theoretical expectations are differences between leaves and grains in the regulatory properties of AGPase. There is ample evidence to show that AGPase in leaves is activated by 3-phosphoglyceric acid (3PGA) and inhibited by inorganic phosphate (Pi) (Preiss and Sivak, 1998; Kleczkowski, 1999). The ratio of 3PGA/Pi in the chloroplast varies during the day in response to variations in photosynthetic metabolism. Variations in these metabolites regulate AGPase activity and thus, the rate of starch synthesis (Kleczkowski, 1999). There have been several investigations of the regulatory properties of AGPase in cereal endosperms. These studies suggest that, in some species, the regulatory properties of the cytosolic AGPase in cereal endosperms may be different from those of the AGPase in photosynthetic tissues (Emes et al., 2003). AGPase from wheat (Olive et al., 1989; Gomez-Casati and Iglesias, 2002) and barley (Kleczkowski et al., 1993a, b; Doan et al., 1999) endosperm is relatively insensitive to activation by 3PGA. It is inhibited by Pi but is less

44

K. TOMLINSON and K. DENYER

sensitive to this metabolite than a typical leaf AGPase. In contrast, AGPase from rice endosperm was found to be sensitive to both 3PGA and Pi (Sikka et al., 2001). AGPase in maize endosperm is also sensitive to activation by fructose 6-phosphate (F6P; Plaxton and Preiss, 1987). The significance of this is not known but it is speculated that F6P in the endosperm may be an indicator of sucrose availability. Thus, the sensitivity to F6P may be a recently acquired (in evolutionary terms) allosteric property that allows activation of AGPase when sucrose is available for starch synthesis. It has been suggested that although inhibition by Pi may be physiologically relevant in a non-photosynthetic, starch-storing organ such as a cereal endosperm, regulation by 3PGA may not (Salamone et al., 2002). Sensitivity to 3PGA in the AGPase of the cereal endosperm may be no more than ‘evolutionary carry-over’ – reflecting the fact that the endosperm AGPase evolved from the leaf AGPase and therefore retains some of its properties. In a recent study comparing the regulatory properties of the plastidial and whole-endosperm (mostly cytosolic) forms of AGPase from wheat grains, Tetlow et al. (2003) found that the leaf enzyme was the most sensitive to regulation by 3PGA and Pi, the endosperm-cytosolic enzyme was the least sensitive and the endosperm-plastid enzyme had properties that were intermediate between those of the other two. This ranking of sensitivities reflects the assumed order of evolution of these enzymes and therefore supports the hypothesis of Salamone et al. (2002). Recent work has shown that AGPase activity in potato tubers is sensitive to the redox state of the plastid as well as being regulated allosterically (Ballicora et al., 2000; Tiessen et al., 2002). In an oxidising environment, disulphide bridges form between the two SSU in the holoenzyme rendering the enzyme less active. The site of the cysteine residue involved in the disulphide bond has been identified as cys12. This residue is conserved in the plastidial SSU of cereal endosperm and in maize leaves but not in the cytosolic SSU of cereal endosperms (Ballicora et al., 2000). Thus, although the plastidial AGPase in cereal endosperm may be regulated by the redox state inside the plastid, it is possible that the cytosolic AGPase has lost this regulatory property. To our knowledge however, this has not been directly confirmed by biochemical analysis of the purified isoforms of AGPase from any cereal endosperm. In an attempt to increase the rate of starch synthesis in cereal grains by alleviating the regulation exerted by AGPase, site-specific mutagenesis with the transposable element dissociation was used to create a revertant form of AGPase in maize endosperm, one of which proved to be un-regulated (Giroux et al., 1996). The revertant, Rev6 with alterations in the gene encoding the cytosolic LSU of AGPase was found to be much less sensitive

STARCH SYNTHESIS IN CEREAL GRAINS

45

(14-fold) to Pi inhibition than the wild-type enzyme but to have normal levels of activation by 3PGA. The grain weight in Rev6 was increased by 11–18% relative to wild-type maize. However, the starch content as a percentage of the dry weight of the grain was not altered. This means that AGPase may exert control over the total dry weight accumulation of the grain rather than specifically over the rate of starch synthesis. More recently, the same modified gene encoding the LSU of AGPase in Rev6 maize was introduced into wheat (Smidansky et al., 2002). Preliminary data suggests that there may have been an increase in grain number per plant and wholeplant biomass but no change in the individual grain weight. To explain these results, the authors suggest that AGPase may be exerting control over some as yet unidentified physiological process in very early flower development rather than specifically controlling the pathway of starch synthesis. No other starch-synthetic enzymes, apart from AGPase, are known to have allosteric effectors. However, some of these enzymes may be regulated by phosphorylation. Members of the SSIII class of soluble SS have conserved motifs with similarity to the consensus motif for phosphoserine/ threonine-binding in 14-3-3 proteins (Sehnke et al., 2001). Biochemical characterisation of protease-treated starch granules from both Arabidopsis thaliana leaves and maize endosperm indicated that a 14-3-3 protein may be located within starch granules. 14-3-3 proteins are known to participate in phosphorylation-related regulatory pathways in plants. In Arabidopsis thaliana leaves, a reduction in one subgroup of 14-3-3 proteins led to an increase in the rate of starch synthesis and altered starch structure (Sehnke et al., 2001). An increase in the activity of starch synthase from spinach leaves by a calcium and calmodulin-specific protein kinase was also observed (Dreier et al., 1992). The role of phosphorylation of starch synthases (or of any of the other enzymes in the pathway) in the control of starch synthesis in cereal endosperms remains to be discovered.

D. METABOLIC CONTROL ANALYSIS

The flux control coefficients of some of the enzymes of starch synthesis in cereal endosperms have been estimated by measuring the quantitative effect of a reduction in enzyme activity on the rate of starch synthesis. In these experiments, it is necessary to change the activity of one enzyme only. This has been achieved in two ways. First, the activity of soluble SS in wheat endosperm was specifically reduced by thermal inactivation (Jenner et al., 1993; Keeling et al., 1993). Soluble SS is unusually sensitive to high temperatures. Irreversible inactivation of SS occurs after incubation of

46

K. TOMLINSON and K. DENYER

isolated grains at temperatures of 25–30 C or more. In the conditions used in these experiments, there were no significant effects on the activities of enzymes involved in starch synthesis other than soluble SS and no known effects on other aspects of endosperm development (Jenner et al., 1993; Keeling et al., 1993). After specifically inhibiting SS by heating, the activity of SS and the rate of starch synthesis in vitro from radioactively labelled glucose or sucrose were measured at 20 C. The decrease in starch synthase activity was almost exactly matched by a decrease in the rate of starch synthesis. From these experiments, the flux control coefficient for starch synthase was estimated to be very high (close to unity) which suggests that the rate of starch synthesis in wheat endosperm is controlled largely by the amount of this enzyme. Second, the flux control coefficients of AGPase, SBE and sucrose synthase were determined for developing maize kernels. Mutants with reduced activity of these enzymes were used (Singletary et al., 1997). The coefficients were estimated by comparison of the enzyme activity and rate of starch synthesis in wild-type maize with that in mutants containing two or three doses of the mutant allele. With two doses of the mutant allele, the perturbation of the activity of other enzymes in the pathway of starch synthesis was minimal. The control coefficients calculated from these data were low (0.04 for AGPase, 0.14 for SBE and 0.01 for sucrose synthase) indicating that the activity of none of these enzymes exerts a significant degree of control over the rate of starch synthesis during early to mid grain filling.

E. STARCH SYNTHESIS AND YIELD

Understanding what controls the starch content per grain is important to the understanding of the control of yield in cereal crops. However, the starch content of the grains is only one of many components determining yield. In fact, it is possible for a crop with many small grains of low starch content to out-yield a crop with a few large grains of high starch content. Yield concerns the amount of starch per plant or more generally, the amount of starch per area of field. To understand yield therefore, it is necessary to understand the control of photosynthesis in the source tissues as well as the control of starch synthesis in the sink and the mechanisms of communication and coordination between these two processes. For more information on yield in cereals, we recommend (Evans and Wardlaw, 1976; Jenner et al., 1991; Satorre and Slafer, 1999; Reynolds et al., 2002; Richards et al., 2002; Tollenaar and Lee, 2002).

STARCH SYNTHESIS IN CEREAL GRAINS

47

Starch synthesis in the grains will be influenced by the supply of the substrate, sucrose, by the source tissues. Environmental factors that adversely affect source strength such as poor light, pest damage, etc. could limit starch synthesis in the grain. The major ways in which source limitation affects yield is via decreases in the number of reproductive structures (tillers, ears or grains) rather than via effects on starch synthesis in the grain per se. In favourable environmental conditions, there is evidence for wheat that the supply of sucrose to the developing grain is not limiting for starch synthesis for most, if not all, of the grain-filling period. Measurements of sucrose content showed that even when the rate of starch synthesis in the wheat grain declines in late development, there is no decrease in the amount of sucrose available to the endosperm cells (Jenner and Rathjen, 1975). It has been suggested that substrate supply could be more limiting late in grain filling when the flag leaf senesces than it is earlier in development. Correlations between the shortening of the grain-filling period and enhanced senescence in the leaves have been noted (Chowdhury and Wardlaw, 1978, and references therein). However, it is not clear whether differences in the duration of photosynthesis in the leaf are the cause, or are caused by, differences in the duration of grain filling. Evidence both for and against source-limitation in late grain-filling is discussed by Evans and Wardlaw (1976).

ACKNOWLEDGEMENTS We thank Dr. Alison M. Smith, John Innes Institute, Norwich, UK and Dr. Peter L. Keeling, BASF Plant Science L.L.C., ExSeed Research, Ames, Iowa, USA for constructive criticism of the manuscript. The John Innes Centre is supported by a competitive strategic grant from the Biotechnology and Biological Sciences Research Council (BBSRC) UK. Kim Tomlinson thanks Syngenta for financial support.

REFERENCES Alonso, M. D., Lomako, J., Lomako, W. M. and Whelan, W. J. (1995). A new look at the biogenesis of glycogen. Faseb Journal 9, 1127–1136. Andersson, L., Fredriksson, H., Bergh, M. O., Andersson, R. and Aman, P. (1999). Characterisation of starch from inner and peripheral parts of normal and waxy barley kernels. Journal of Cereal Science 30, 165–171. Andon, N. L., Hollingworth, S., Koller, A., Greenland, A. J., Yates, J. R. and Haynes, P. A. (2002). Proteomic characterization of wheat amyloplasts using identification of proteins by tandem mass spectrometry. Proteomics 2, 1156–1168.

48

K. TOMLINSON and K. DENYER

ap Rees, T. and Hill, S. A. (1994). How should we study the regulation of plant metabolism? Plant, Cell and Environment 17, 587–599. ap Rees, T. (1995). Where do plants make ADP-Glucose? In ‘Sucrose metabolism, biochemistry, physiology and molecular biology’, (H. G. Pontis, G. L. Salerno and E. J. Echeverria, eds.), pp. 143–155. American Society of Plant Physiologists, USA. Ardila, F. and Tandecarz, J. (1992). Potato tuber UDP-glucose: protein transglucosylase catalyses its own glucosylation. Plant Physiology 99, 1342–1347. Baba, T., Yoshii, M. and Kainuma, K. (1987). Acceptor molecule of granular-bound starch synthase from sweet-potato roots. Starch/Sta¨rke 39, 52–56. Ball, S., Guan, H. P., James, M., Myers, A., Keeling, P., Mouille, G., Buleon, A., Colonna, P. and Preiss, J. (1996). From glycogen to amylopectin – a model for the biogenesis of the plant starch granule. Cell 86, 349–352. Ball, S., van de Wal, M. and Visser, R. (1998). Progress in understanding the biosynthesis of amylose. Trends in Plant Science 3, 462–467. Ballicora, M. A., Laughlin, M. J., Fu, Y. B., Okita, T. W., Barry, G. F. and Preiss, J. (1995). Adenosine 50 -diphosphate-glucose pyrophosphorylase from potatotuber – significance of the N-terminus of the small-subunit for catalytic properties and heat-stability. Plant Physiology 109, 245–251. Ballicora, M. A., Frueauf, J. B., Fu, Y. B., Schurmann, P. and Preiss, J. (2000). Activation of the potato tuber ADP-glucose pyrophosphorylase by thioredoxin. Journal of Biological Chemistry 275, 1315–1320. Baroja-Fernandez, E., Munoz, F. J., Akazawa, T. and Pozueta-Romero, J. (2001). Reappraisal of the currently prevailing model of starch biosynthesis in photosynthetic tissues: a proposal involving the cytosolic production of ADP-Glucose by sucrose synthase and occurrence of cyclic turnover of starch in the chloroplast. Plant and Cell Physiology 42, 1311–1320. Beckles, D. M., Craig, J. and Smith, A. M. (2001a). ADP-Glucose pyrophosphorylase is located in the plastid in developing tomato fruit. Plant Physiology 126, 261–266. Beckles, D. M., Smith, A. M. and ap Rees, T. (2001b). A cytosolic ADP-Glucose pyrophosphorylase is a feature of Graminaceous endosperms, but not of other starch-storing organs. Plant Physiology 125, 818–827. Black, M., Corbineau, F., Grzesik, M., Guy, P. and Come, D. (1996). Carbohydrate metabolism in the developing and maturing wheat embryo in relation to its desiccation tolerance. Journal of Experimental Botany 47, 161–169. Blauth, S. L., Yao, Y., Klucinec, J. D., Shannon, J. C., Thompson, D. B. and Guilitinan, M. J. (2001). Identification of Mutator insertional mutants of starch-branching enzyme IIa in corn. Plant Physiology 125, 1396–1405. Blauth, S. L., Kim, K. N., Klucinec, J., Shannon, J. C., Thompson, D. and Guiltinan, M. (2002). Identification of Mutator insertional mutants of starch-branching enzyme 1 (SBE1) in Zea mays L. Plant Molecular Biology 48, 287–297. Borovsky, D., Smith, E. E. and Whelan, W. J. (1976). On the mechanism of amylose branching by potato Q-enzyme. European Journal of Biochemistry 62, 307–312. Borovsky, D., Smith, E. E. and Whelan, W. J. (1975). Purification and properties of potato 1,4--D-glucan 6--(1,4--glucano)-transferase. Evidence against a dual catalytic function in amylose-branching enzyme. European Journal of Biochemistry 59, 615–625. Borovsky, D., Smith, E. E., Whelan, W. J., French, D. and Kikumoto, S. (1979). The mechanism of Q-enzyme action and its influence on the structure of amylopectin. Archives of Biochemistry and Biophysics 198, 627–631.

STARCH SYNTHESIS IN CEREAL GRAINS

49

Bosnes, M., Weideman, F. and Olsen, O.-A. (1992). Endosperm differentiation in barley wild-type and sex mutants. Plant Journal 2, 661–674. Boyer, C. D., Daniels, R. R. and Shannon, J. C. (1976). Abnormal starch granule formation in Zea mays L. endosperms possessing the amylose-extender mutation. Crop Science 16, 298–301. Boyer, C. D., Daniels, R. R. and Shannon, J. C. (1977). Starch granule (amyloplast) development in endosperm of several Zea mays L. genotypes affecting kernel polysaccharides. American Journal of Botany 64, 50–56. Bradbury, D., Cull, I. M. and MacMasters, M. M. (1956). Structure of the mature wheat kernel. III. Microscopic structure of the endosperm of hard red winter wheat. Cereal Chemistry 33, 361–373. Brangeon, J., Reyss, A. and Prioul, J. L. (1997). In situ detection of ADPglucose pyrophosphorylase expression during maize endosperm development. Plant Physiology and Biochemistry 35, 847–858. Briarty, L. G. and Hughes, C. E. (1979). The developing endosperm of wheat – a stereological analysis. Annals of Botany 44, 641–658. Burton, R. A., Bewley, J. D., Smith, A. M., Bhattacharyya, M. K., Tatge, H., Ring, S., Bull, V., Hamilton, W. D. O. and Martin, C. (1995). Starch-branching enzymes belonging to distinct enzyme families are differentially expressed during pea embryo development. Plant Journal 7, 3–15. Burton, R. A., Jenner, H., Carrangis, L., Fahy, B., Fincher, G. B., Hylton, C., Laurie, D. A., Parker, M., Waite, D., van Wegen, S., Verhoeven, T. and Denyer, K. (2002a). Starch granule initiation and growth are altered in barley mutants that lack isoamylase activity. Plant Journal 31, 97–112. Burton, R. A., Johnson, P. E., Beckles, D. M., Fincher, G. B., Jenner, H. L., Naldrett, M. J. and Denyer, K. (2002b). Characterization of the genes encoding the cytosolic and plastidial forms of ADP-glucose pyrophosphorylase in wheat endosperm. Plant Physiology 130, 1464–1475. Cao, H., Sullivan, T. D., Boyer, C. D. and Shannon, J. C. (1995). Bt1, a structural gene for the major 39–44 kDa amyloplast membrane polypeptides. Physiologia Plantarum 95, 176–186. Cao, H. P., Imparl-Radosevich, J., Guan, H. P., Keeling, P. L., James, M. G. and Myers, A. M. (1999). Identification of the soluble starch synthase activities of maize endosperm. Plant Physiology 120, 205–215. Carlson, S. J., Chourey, P. S., Helentjaris, T. and Datta, R. (2002). Gene expression studies on developing kernels of maize sucrose synthase (SuSy) mutants show evidence for a third SuSy gene. Plant Molecular Biology 49, 15–29. Chanda, S. V., Narmada, K. and Singh, Y. D. (1999). Dry matter accumulation and associated changes in biochemical parameters during wheat grain development. Journal of Agronomy and Crop Science-Zeitschrift Fur Acker Und Pflanzenbau 182, 153–159. Chaudhury, A. M., Koltunow, A., Payne, T., Luo, M., Tucker, M. R., Dennis, E. S. and Peacock, W. J. (2001). Control of early seed development. Annual Review of Cellular and Developmental Biology 17 IS, 677–700. Chevalier, P. and Lingle, S. E. (1983). Sugar metabolism in developing kernels of wheat and barley. Crop Science 23, 272–277. Chowdhury, S. I. and Wardlaw, I. F. (1978). The effect of temperature on kernel development in cereals. Australian Journal of Agricultural Research 29, 205–223. Clark, J. R., Robertson, M. and Ainsworth, C. C. (1991). Nucleotide sequence of a wheat (Triticum aestivum) cDNA clone encoding the waxy protein. Plant Molecular Biology 16, 1099–1101.

50

K. TOMLINSON and K. DENYER

Clarke, B. R., Denyer, K., Jenner, C. F. and Smith, A. M. (1999). The relationship between the rate of starch synthesis, the adenosine 50 -diphosphoglucose concentration and the amylose content of starch in developing pea embryos. Planta 209, 324–329. Commuri, P. D. and Keeling, P. L. (2001). Chain-length specificities of maize starch synthase l enzyme: studies of glucan affinity and catalytic properties. Plant Journal 25, 475–486. Craig, J., Lloyd, J. R., Tomlinson, K., Barber, L., Edwards, A., Wang, T. L., Martin, C., Hedley, C. L. and Smith, A. M. (1998). Mutations in the gene encoding starch synthase II profoundly alter amylopectin structure in pea embryos. Plant Cell 10, 413–426. Damager, I., Olsen, C. E., Blennow, A., Denyer, K., Moller, B. L. and Motawia, M. S. (2003). Chemical synthesis of methyl 60 --maltosyl--maltotrioside and its use for investigation of the action of starch synthase II. Carbohydrate Research 338, 189–197. Delgado, I. J., Wang, Z. H., de Rocher, A., Keegstra, K. and Raikhel, N. V. (1998). Cloning and characterization of AtRGP1 – a reversibly autoglycosylated Arabidopsis protein implicated in cell wall biosynthesis. Plant Physiology 116, 1339–1349. Delrue, B., Fontaine, R., Routier, F., Decq, A., Wieruszeski, J.-M., Van den Koornhuyse, N., Maddelein, M.-L., Fournet, B. and Ball, S. (1992). Waxy Chlamydomonas reinhardtii: Monocellular algal mutants defective in amylose biosynthesis and granule-bound starch synthase accumulate a structurally modified amylopectin. Journal of Bacteriology 174, 3612–3620. Denyer, K. and Smith, A. M. (1988). The capacity of plastids from developing pea cotyledons to synthesize acetyl coA. Planta 173, 172–182. Denyer, K., Dunlap, F., Thorbjornsen, T., Keeling, P. and Smith, A. M. (1996a). The major form of ADP-Glucose pyrophosphorylase in maize endosperm is extra-plastidial. Plant Physiology 112, 779–785. Denyer, K., Clarke, B., Hylton, C., Tatge, H. and Smith, A. M. (1996b). The elongation of amylose and amylopectin chains in isolated starch granules. Plant Journal 10, 1135–1143. Denyer, K., Waite, D., Motawia, S., Moller, B. L. and Smith, A. M. (1999a). Granule-bound starch synthase I in isolated starch granules elongates malto-oligosaccharides processively. Biochemical Journal 340, 183–191. Denyer, K., Waite, D., Edwards, A., Martin, C. and Smith, A. M. (1999b). Interaction with amylopectin influences the ability of granule-bound starch synthase I to elongate malto-oligosaccharides. Biochemical Journal 342, 647–653. Denyer, K., Johnson, P., Zeeman, S. and Smith, A. M. (2001). The control of amylose synthesis. Journal of Plant Physiology 158, 479–487. Dhugga, K. S., Tiwari, S. C. and Ray, P. M. (1997). A reversibly glycosylated polypeptide (RGP1) possibly involved in plant cell wall synthesis: purification, gene cloning and trans-Golgi localization. Proceedings of the National Academy of Sciences USA 94, 7679–7684. Dinges, J. R., Colleoni, C., Myers, A. M. and James, M. G. (2001). Molecular structure of three mutations at the maize sugary 1 locus and their allelespecific phenotypic effects. Plant Physiology 125, 1406–1418. Dinges, J. R., Colleoni, C., James, M. G. and Myers, A. M. (2003). Mutational analysis of the pullulanase-type debranching enzyme of maize indicates multiple functions in starch metabolism. Plant Cell 15, in press. Doan, D. N. P., Rudi, H. and Olsen, O.-A. (1999). The allosterically unregulated isoform of ADP-glucose pyrophosphorylase from barley endosperm is the

STARCH SYNTHESIS IN CEREAL GRAINS

51

most likely source of ADP-glucose incorporated into endosperm starch. Plant Physiology 121, 965–975. Doehlert, D. C. and Felker, F. C. (1987). Characterization and distribution of invertase activity in developing maize (Zea mays) kernels. Physiologia Plantarum 70, 51–57. Doehlert, D. C. (1990). Distribution of enzyme-activities within the developing maize (Zea-mays) kernel in relation to starch, oil and protein accumulation. Physiologia Plantarum 78, 560–567. Donald, A. M. (2001). Plasticization and self-assembly in the starch granule. Cereal Chemistry 78, 307–314. Dreier, W., Preusser, E. and Grundel, M. (1992). The regulation of the activity of soluble starch synthase in spinach leaves by a calcium-calmodulin dependent protein-kinase. Biochemie und Physiologie Der Pflanzen 188, 81–96. Edwards, A., Fulton, D. C., Hylton, C. M., Jobling, S. A., Gidley, M., Rossner, U., Martin, C. and Smith, A. M. (1999). A combined reduction in activity of starch synthases II and III of potato has novel effects on the starch of tubers. Plant Journal 17, 251–261. Emes, M. J., Bowsher, C. G., Hedley, C., Burrell, M. M., Scrase-Field, E. S. and Tetlow, I. J. (2003). Starch synthesis and carbon partitioning in developing endosperm. Journal of Experimental Botany 54, 569–575. Evans, L. T. and Wardlaw, I. F. (1976). Aspects of the comparative physiology of grain yield in cereals. Advances in Agronomy 28, 301–359. Evers, A. D. (1970). Development of the endosperm of wheat. Annals of Botany 34, 547–555. Evers, A. D., Blakeney, A. B. and O’Brien, L. (1999). Cereal structure and composition. Australian Journal of Agricultural Research 50, 629–650. Evers, T. (1971). Scanning electron microscopy of wheat starch. 3. Granule development in the endosperm. Starch-Starke 23, 157–162. Evers, T. and Millar, S. (2002). Cereal grain structure and development: Some implications for quality. Journal of Cereal Science 36, 261–284. Fulton, D. C., Edwards, A., Pilling, E., Robinson, H. L., Fahy, B., Seale, R., Kato, L., Donald, A. M., Geigenberger, P., Martin, C. and Smith, A. M. (2002). Role of granule-bound starch synthase in determination of amylopectin structure and starch granule morphology in potato. Journal of Biological Chemistry 277, 10834–10841. Gallant, D. J., Bouchet, B. and Baldwin, P. M. (1997). Microscopy of starch: evidence of a new level of granule organization. Carbohydrate Polymers 32, 177–191. Gao, M., Fisher, D. K., Kim, K. N., Shannon, J. C. and Guiltinan, M. J. (1997). Independent genetic control of maize starch-branching enzymes IIa and IIb (Isolation and characterization of Sbe2a cDNA). Plant Physiology 114, 69–78. Gao, M., Wanat, J., Stinard, P. S., James, M. G. and Myers, A. M. (1998). Characterization of dull1, a maize gene coding for a novel starch synthase. Plant Cell 10, 399–412. Geetha, K. B., Lending, C. R., Lopes, M. A., Wallace, J. C. and Larkins, B. A. (1991). Opaque-2 modifiers increase gamma-zein synthesis and alter its spatial distribution in maize endosperm. Plant Cell 3, 1207–1219. Gidley, M. J. (1992). Structural disorder in starch granules and its loss during gelatinisation. In ‘Gums and stabilisers for the food industry’, (G. O. Phillips, P. A. Williams and D. J. Wedlock, eds.), pp. 87–92. IRL Press, Oxford, UK.

52

K. TOMLINSON and K. DENYER

Giroux, M. J., Shaw, J., Barry, G., Cobb, B. G., Greene, T., Okita, T. and Hannah, L. C. (1996). A single gene mutation that increases maize seed weight. Proceedings of the National Academy of Sciences USA 93, 5824–5829. Gomez-Casati, D. F. and Iglesias, A. A. (2002). ADP-glucose pyrophosphorylase from wheat endosperm. Purification and characterization of an enzyme with novel regulatory properties. Planta 214, 428–434. Grass Phylogeny Working Group (2001). Phylogeny and subfamilial classification of the grasses (Poaceae). Annals of the Missouri Botanical Gardens 88, 373–457. Griffith, S. M., Jones, R. J. and Brenner, M. L. (1987). In vitro sugar transport in Zea mays L. kernels. 1. Characteristics of sugar absorption and metabolism by developing maize endosperm. Plant Physiology 84, 467–471. Guan, H. P. and Preiss, J. (1993). Differentiation of the properties of the branching isozymes from maize (zea-mays). Plant Physiology 102, 1269–1273. Guan, H. P., Kuriki, T., Sivak, M. and Preiss, J. (1995). Maize branching enzyme catalyzes synthesis of glycogen-like polysaccharide in glgb-deficient Escherichia-coli. Proceedings of the National Academy of Sciences USA 92, 964–967. Han, X. Z. and Hamaker, B. R. (2002). Location of starch granule-associated proteins revealed by confocal laser scanning microscopy. Journal of Cereal Science 35, 109–116. Hannah, L. C., Shaw, J. R., Giroux, M. J., Reyss, A., Prioul, J. L., Bae, J. M. and Lee, J. Y. (2001). Maize genes encoding the small subunit of ADP-glucose pyrophosphorylase. Plant Physiology 127, 173–183. Harn, C., Knight, M., Ramakrishnan, A., Guan, H. P., Keeling, P. L. and Wasserman, B. P. (1998). Isolation and characterization of the zSSIIa and zSSIIb starch synthase cDNA clones from maize endosperm. Plant Molecular Biology 37, 639–649. Hawker, J. S. and Jenner, C. F. (1993). High-temperature affects the activity of enzymes in the committed pathway of starch synthesis in developing wheat endosperm. Australian Journal of Plant Physiology 20, 197–209. Hizukuri, S. (1986). Polymodal distribution of the chain lengths of amylopectins and its significance. Carbohydrate Research 147, 342–347. Hofmeyr, J. H. S. and Cornishbowden, A. (1991). Quantitative assessment of regulation in metabolic systems. European Journal of Biochemistry 200, 223–236. Hoshikawa, K. (1973). Morphogenesis of endosperm tissue in rice. Japan Agricultural Research Quarterly 7, 153–159. Hoshikawa, K. (1993). Anthesis, fertilization and development of caryopsis. In ‘Science of the rice plant. Volume 1. Morphology’, (T. Matsuo and K. Hoshikawa, eds.), pp. 339–375. Food and Agriculture Policy Research Center, Tokyo, Japan. Hussain, H., Mant, A., Seale, R., Zeeman, S., Hinchliffe, E., Edwards, A., Hylton, C., Bornemann, S., Smith, A. M., Martin, C. and Bustos, R. (2003). Three isoforms of isoamylase contribute different catalytic properties for the debranching of potato glucans. Plant Cell 15, 133–149. Imparl-Radosevich, J. M., Li, P., Zhang, L., Mckean, A. L., Keeling, P. L. and Guan, H. P. (1998). Purification and characterization of maize starch synthase I and its truncated forms. Archives of Biochemistry and Biophysics 353, 64–72. James, M. G., Robertson, D. S. and Myers, A. M. (1995). Characterisation of the maize gene sugary1, a determinant of starch composition in kernels. Plant Cell 7, 417–429.

STARCH SYNTHESIS IN CEREAL GRAINS

53

Jane, J., Xu, A., Radosavljevic, M. A. and Seib, P. A. (1992). Location of amylose in normal starch granules: 1. Susceptibility of amylose and amylopectin to cross-linking reagents. Cereal Chemistry 69, 405–409. Jane, J. and Shen, J. J. (1993). Internal structure of potato starch granule revealed by chemical gelatinization. Carbohydrate Research 247, 279–290. Jane, J., Kasemsuwan, T., Leas, S., Ames, I. A., Zobel, H., Darien, I. L. and Robyt, J. F. (1994). Anthology of starch granule morphology by scanning electron microscopy. Starch/Sta¨rke 46, 121–129. Jane, J., Wong, K. S. and McPherson, A. E. (1997). Branch-structure difference in starches of A- and B-type x-ray patterns revealed by their Naegeli dextrins. Carbohydrate Research 300, 219–227. Jenkins, P. J., Cameron, R. E. and Donald, A. M. (1993). A universal feature in the structure of starch granules from different botanical sources. Starch/Sta¨rke 45, 417–421. Jenkins, P. J. and Donald, A. M. (1995). The influence of amylose on starch granule structure. International Journal of Macromolecules 17, 315–321. Jenner, C. F. (1973). The uptake of sucrose and its conversion to starch in detatched ears of wheat. Journal of Experimental Botany 24, 295–306. Jenner, C. F. (1974). An investigation of the association between the hydrolysis of sucrose and its absorption by grains of wheat. Australian Journal of Plant Physiology 1, 319–329. Jenner, C. F. and Rathjen, A. J. (1975). Factors regulating the accumulation of starch in ripening wheat grain. Australian Journal of Plant Physiology 2, 311–322. Jenner, C. F., Ugalde, T. D. and Aspinall, D. (1991). The physiology of starch and protein deposition in the endosperm of wheat. Australian Journal of Plant Physiology 18, 211–226. Jenner, C. F. (1991). Effects of exposure of wheat ears to high-temperature on drymatter accumulation and carbohydrate-metabolism in the grain of two cultivars. Australian Journal of Plant Physiology 18, 179–190. Jenner, C. F., Siwek, K. and Hawker, J. S. (1993). The synthesis of [C-14] starch from [C-14] sucrose in isolated wheat grains is dependent upon the activity of soluble starch synthase. Australian Journal of Plant Physiology 20, 329–335. Johnson, P. E., Patron, N. J., Bottrill, A. R., Dinges, J. R., Fahy, B. F., Parker, M. L., Waite, D. N. and Denyer, K. (2003). A low-starch barley mutant, Risø 16, lacking the cytosolic small subunit of ADP-glucose pyrophosphorylase, reveals the importance of the cytosolic isoform and the identity of the plastidial small subunit. Plant Physiology 131, 684–696. Kacser, H. and Burns, J. A. (1973). The control of flux. Symposia of the Society for Experimental Biology 28, 65–104. Kacser, H. (1987). Control of metabolism. In ‘‘The Biochemistry of Plants’’, (D. D. Davies, ed.), pp. 39–67. Academic Press, New York, USA. Kasmsuwan, T. and Jane, J. (1994). Location of amylose in normal starch granules. II. Location of phosphodiester cross-linking revealed by phosphorus-31 nuclear magnetic resonance. Cereal Chemistry 71, 282–287. Kavakli, I. H., Greene, T. W., Salamone, P. R., Choi, S. B. and Okita, T. W. (2001). Investigation of subunit function in ADP-glucose pyrophosphorylase. Biochemical and Biophysical Research Communications 281, 783–787. Keeling, P. L., Bacon, P. J. and Holt, D. C. (1993). Elevated temperature reduces starch deposition in wheat endosperm by reducing the activity of soluble starch synthase. Planta 191, 342–348.

54

K. TOMLINSON and K. DENYER

Keeling, P. L. (1999). From enzyme activity to flux control: a quest to understand starch deposition in developing cereal grains. In ‘Plant Carbohydrate Biochemistry’, (J. A. Byrant, M. M. Burrel and N. J. Kruger, eds.), pp. 91–102. BIOS Scientific Publishers Ltd., Oxford, UK. Kleczkowski, L. A., Villand, P., Luthi, E., Olsen, O.-A. and Preiss, J. (1993a). Insensitivity of barley endosperm adp-glucose pyrophosphorylase to 3-phosphoglycerate and orthophosphate regulation. Plant Physiology 101, 179–186. Kleczkowski, L. A., Villand, P., Preiss, J. and Olsen, O.-A. (1993b). Kinetic mechanism and regulation of ADP-glucose pyrophosphorylase from barley (Hordeum vulgare) leaves. Journal of Biological Chemistry 268, 6228–6233. Kleczkowski, L. A. (1999). A phosphoglycerate to inorganic phosphate ratio is the major factor in controlling starch levels in chloroplasts via ADP-glucose pyrophosphorylase regulation. FEBS letters 448, 153–156. Klo¨sgen, R. B., Gierl, A., Schwarz-Sommer, Z. and Saedler, H. (1986). Molecular analysis of the waxy locus of Zea mays. Molecular and General Genetics 203, 237–244. Knight, M. E., Harn, C., Lilley, C. E., Guan, H., Singletary, G. W., MuForster, C., Wasserman, B. P. and Keeling, P. L. (1998). Molecular cloning of starch synthase I from maize (W64) endosperm and expression in Escherichia coli. Plant Journal 14, 613–622. Kossmann, J. and Lloyd, J. (2000). Understanding and influencing starch biochemistry. Critical Reviews in Plant Sciences 19, 171–226. Kruger, J. E. (1989). Biochemistry of preharvest sprouting in cereals. In ‘Preharvest Field Sprouting in Cereals’, (N. F. Derera, ed.), pp. 61–84. CRC Press Inc., Boca Raton, Florida, USA. Kubo, A., Fujita, N., Harada, K., Matsuda, T., Satoh, H. and Nakamura, Y. (1999). The starch-debranching enzymes isoamylase and pullulanase are both involved in amylopectin biosynthesis in rice endosperm. Plant Physiology 121, 399–409. Kuipers, A. G. J., Jacobsen, E. and Visser, R. G. F. (1994). Formation and deposition of amylose in the potato-tuber starch granule are affected by the reduction of granule-bound starch synthase gene-expression. Plant Cell 6, 43–52. Langeveld, S. M. J., van Wijk, R., Stuurman, N., Kijne, J. W. and de Pater, B. S. (2000). B-type granule containing protrusions and interconnections between amyloplasts in developing wheat endosperm revealed by transmission electron microscopy and GFP expression. Journal of Experimental Botany 51, 1357–1361. Langeveld, S. M. J., Vennik, M., Kottenhagen, M., van Wijk, R., Buijk, A., Kijne, J. W. and de Pater, S. (2002). Glucosylation activity and complex formation of two classes of reversibly glycosylated polypeptides. Plant Physiology 129, 278–289. Li, Z., Rahman, S., Kosarhashemi, B., Mouille, G., Appels, R. and Morell, M. K. (1999a). Cloning and characterization of a gene encoding wheat starch synthase I. Theoretical and Applied Genetics 98, 1208–1216. Li, Z. Y., Chu, X. S., Mouille, G., Yan, L. L., Kosarhashemi, B., Hey, S., Napier, J., Shewry, P., Clarke, B., Appels, R., Morell, M. K. and Rahman, S. (1999b). The localization and expression of the class II starch synthases of wheat. Plant Physiology 120, 1147–1155. Li, Z. Y., Mouille, G., Kosar-Hashemi, B., Rahman, S., Clarke, B., Gale, K. R., Appels, R. and Morell, M. K. (2000). The structure and expression of the wheat starch synthase III gene. Motifs in the expressed gene define the

STARCH SYNTHESIS IN CEREAL GRAINS

55

lineage of the starch synthase III gene family. Plant Physiology 123, 613–624. Lin, T. P., Caspar, T., Somerville, C. R. and Preiss, J. (1988). A starch-deficient mutant of Arabidopsis thaliana with low ADP-Glucose pyrophosphorylase activity lacks one of the two subunits of the enzyme. Plant Physiology 88, 1175–1181. Lingle, S. E. and Chevalier, P. (1984). Movement and metabolism of sucrose in developing barley kernels. Crop Science 24, 315–319. Lomako, J., Lomako, W. M. and Whelan, W. J. (1992). The substrate-specificity of isoamylase and the preparation of apo-glycogenin. Carbohydrate Research 227, 331–338. Lopes, M. A. and Larkins, B. A. (1993). Endosperm origin, development and function. Plant Cell 5, 1383–1399. Loss, S. P., Kirby, E. J. M., Siddique, K. H. M. and Perry, M. W. (1989). Graingrowth and development of old and modern Australian wheats. Field Crops Research 21, 131–146. Lu, T. J., Jane, J. L. and Keeling, P. L. (1997). Temperature effect on retrogradation rate and crystalline structure of amylose. Carbohydrate Polymers 33, 19–26. McClintock, B. (1978). Development of the maize endosperm as revealed by clones. In ‘The Clonal Basis of Development’, (S. Subtelny. and I. Sussex, eds.), pp. 217–237. Academic Press, London, UK. Miller, M. E. and Chourey, P. S. (1995). Intracellular immunolocalization of adenosine 50 -diphosphoglucose pyrophosphorylase in developing endosperm cells of maize (Zea mays L). Planta 197, 522–527. Mohlmann, T., Tjaden, J., Henrichs, G., Quick, W. P., Hausler, R. and Neuhaus, H. E. (1997). ADP-glucose drives starch synthesis in isolated maize endosperm amyloplasts: characterization of starch synthesis and transport properties across the amyloplast envelope. Biochemical Journal 324, 503–509. Morell, M. K., Blennow, A., Koshar-Hashemi, B. and Samuel, M. S. (1997). Differential expression and properties of starch branching enzyme isoforms in developing wheat endosperm. Plant Physiology 113, 201–208. Morell, M. K., Li, Z., and Rahman, S. (2001). Starch biosynthesis in the small grained cereals: wheat and barley. In ‘Starch: Advances in Structure and Function.’, (T. L. Barsby, A. M. Donald and P. J. Frazier, eds.), pp. 129–137. The Royal Society of Chemistry, Cambridge, UK. Mouille, G., Maddelein, M. L., Libessart, N., Talaga, P., Decq, A., Delrue, B. and Ball, S. (1996). Preamylopectin processing: A mandatory step for starch biosynthesis in plants. Plant Cell 8, 1353–1366. Mukerjea, R., Yu, L. and Robyt, J. F. (2002). Starch biosynthesis: mechanism for the elongation of starch chains. Carbohydrate Research 337, 1015–1022. Myers, A. M., Morell, M. K., James, M. G. and Ball, S. G. (2000). Recent progress toward understanding biosynthesis of the amylopectin crystal. Plant Physiology 122, 989–997. Nakamura, Y., Kubo, A., Shimamune, T., Matsuda, T., Harada, K. and Satoh, H. (1997). Correlation between activities of starch debranching enzyme and alpha-polyglucan structure in endosperms of sugary-1 mutants of rice. Plant Journal 12, 143–153. 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. Nishi, A., Nakamura, Y., Tanaka, N. and Satoh, H. (2001). Biochemical and genetic analysis of the effects of amylose-extender mutation in rice endosperm. Plant Physiology 127, 459–472.

56

K. TOMLINSON and K. DENYER

Nordmark, T. S. and Ziegler, G. R. (2002). Structural features of non-granular spherulitic maize starch. Carbohydrate Research 337, 1467–1475. Olive, M., Ellis, R. J. and Schuh, W. W. (1989). Isolation and nucleotidesequences of cDNA clones encoding ADP-glucose pyrophosphorylase polypeptides from wheat leaf and endosperm. Plant Molecular Biology 12, 525–528. Olsen, O.-A., Potter, R. H. and Kalla, R. (1992). Histo-differentiation and molecular biology of developing cereal endosperm. Seed Science Review 2, 117–131. Olsen, O.-A., Linnestad, C., and Nichols, S. E. (1999). Developmental biology of the cereal endosperm. Trends in Plant Science 4, 253–257. Olsen, O.-A. (2001). Endosperm development: Cellularization and cell fate specification. Annual Review of Plant Physiology and Plant Molecular Biology 52, 233–268. Oostergetel, G. T. and van Bruggen, E. F. J. (1993). The crystalline domains in potato starch granules are arranged in a helical fashion. Carbohydrate Polymers 21, 7–12. Oscarsson, M., Parkkonen, T., Autio, K. and Aman, P. (1997). Composition and microstructure of waxy, normal and high-amylose barley samples. Journal of Cereal Science 26, 259–264. Palmer, G. H. (1989). ‘Cereal Science and Technology’. Aberdeen University Press, Aberdeen, UK. Pan, D., Stinard, P., Dietrich, J. and Nelson, O. E. (1995). A comparison of debranching enzyme activities in su1 and su3 endosperms. Maize Genetics Conference Abstracts 37, 5–6. Parker, M. L. (1985). The relationship between A-type and B-type starch granules in the developing endosperm of wheat. Journal of Cereal Science 3, 271–278. Patron, N. J., Smith, A. M., Fahy, B. F., Hylton, C. M., Naldrett, M. J., Rossnagel, B. G. and Denyer, K. (2002). The altered pattern of amylose accumulation in the endosperm of low-amylose barley cultivars is attributable to a single mutant allele of granule-bound starch synthase I with a deletion in the 50 -non-coding region. Plant Physiology 130, 190–198. Peltier, J. B., Friso, G., Kalume, D. E., Roepstorff, P., Nilsson, F., Adamska, I. and van Wijk, K. J. (2000). Proteomics of the chloroplast: Systematic identification and targeting analysis of lumenal and peripheral thylakoid proteins. Plant Cell 12, 319–341. Peng, L., Kawagoe, Y., Hogan, P. and Delmer, D. (2002). Sitosterol-beta-glucoside as primer for cellulose synthesis in plants. Science 295, 147–150. Peng, M. S., Gao, M., Baga, M., Hucl, P. and Chibbar, R. N. (2000). Starchbranching enzymes preferentially associated with A-type starch granules in wheat endosperm. Plant Physiology 124, 265–272. Perez, C. M., Perdon, A. A., Resurreccion, A. P., Villareal, R. M. and Juliano, B. O. (1975). Enzymes of carbohydrate metabolism in the developing rice grain. Plant Physiology 56, 579–583. Plaxton, W. C. and Preiss, J. (1987). Purification and properties of nonproteolytic degraded ADP-glucose pyrophosphorylase from maize endosperm. Plant Physiology 83, 105–112. Pontis, H. G. (1977). Riddle of sucrose. MTP International Reviews of Science and Plant Biochemistry 13, 79–117. Pozueta-Romero, J., Frehner, M., Viale, A. M. and Akazawa, T. (1991). Direct transport of ADPglucose by an adenylate translocator is linked to starch biosynthesis in amyloplasts. Proceedings of the National Academy of Sciences USA 88, 5769–5773.

STARCH SYNTHESIS IN CEREAL GRAINS

57

Preiss, J. and Sivak, M. (1998). Biochemistry, molecular biology and regulation of starch synthesis. Genetic Engineering 20, 177–223. Preiss, J., Danner, S., Summers, P. S., Morell, M., Barton, C. R., Yang, L. and Nieder, M. (1990). Molecular characterization of the brittle-2 gene effect on maize endosperm ADP-Glucose pyrophosphorylase subunits. Plant Physiology 92, 881–885. Preiss, J., Ball, K., Smith-White, B., Iglesias, A., Kakefuda, G. and Li, L. (1991). Starch biosynthesis and its regulation. Biochemical Society Transactions 19, 539–547. Reichert, E. T. (1913). ‘The Differentiation and Specificity of Starches in Relation to Genera, Species, etc. Stereochemistry Applied to Protoplasmic Processes and Products, and as a Strictly Scientific Basis for the Classification of Plants and Animals. Part I’, Carnegie Institution, Washington, USA Reski, R. (2002). Rings and networks: the amazing complexity of FtsZ in chloroplasts. Trends in Plant Science 7, 103–105. Reynolds, M. P., Trethrowan, R., Crossa, J., Vargas, M. and Sayre, K. D. (2002). Physiological factors associated with genotype by environment interaction in wheat. Field Crops Research 4049, 1–22. Richards, R. A., Rebetzke, G. J., Condon, A. G. and van Herwaarden, A. F. (2002). Crop physiology and metabolism. Breeding opportunities for increasing the efficiency of water use and crop yield in temperate cereals. Crop Science 42, 111–121. Ridout, M. J., Gunning, A. P., Parker, M. L., Wilson, R. H. and Morris, V. J. (2002). Using AFM to image the internal structure of starch granules. Carbohydrate Polymers 50, 123–132. Roach, P. J. and 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. Salamone, P. R., Kavakli, I. H., Slattery, C. J. and Okita, T. W. (2002). Directed molecular evolution of ADP-glucose pyrophosphorylase. Proceedings of the National Academy of Sciences USA 99, 1070–1075. Salvador, R. J. and Pearce, R. B. (1995). Proposed standard system of nomenclature for maize grain filling events and concepts. Maydica 40, 141–146. Satorre, E. H. and Slafer, G. A. (1999). ‘‘Wheat: Ecology and Physiology of Yield Determination’’, Food Products Press, Binghamton, New York, USA. Schmalstig, J. G. and Hitz, W. D. (1987). Transport and metabolism of a sucrose analog (10 -fluorosucrose) into Zea mays endosperm without invertase hydrolysis. Plant Physiology 85, 902–905. Schwall, G. P., Safford, R., Westcott, R. J., Jeffcoat, R., Tayal, A., Shi, Y. C., Gidley, M. J. and Jobling, S. A. (2000). Production of very-high-amylose potato starch by inhibition of SBE A and B. Nature Biotechnology 18, 551–554. Sehnke, P. C., Chung, H. J., Wu, K. and Ferl, R. J. (2001). Regulation of starch accumulation by granule-associated plant 14-3-3 proteins. Proceedings of the National Academy of Sciences USA 98, 765–770. Seo, B., Kim, S., Scott, M. P., Singletary, G. W., Wong, K., James, M. G. and Myers, A. M. (2002). Functional interactions between heterologously expressed starch-branching enzymes of maize and the glycogen synthases of brewer’s yeast. Plant Physiology 128, 1189–1199. Shannon, J. C., Pien, F. M., Cao, H. P. and Liu, K. C. (1998). Brittle-1, an adenylate translocator, facilitates transfer of extraplastidial synthesized ADPglucose into amyloplasts of maize endosperms. Plant Physiology 117, 1235–1252.

58

K. TOMLINSON and K. DENYER

Shewry, P. R. and Morell, M. (2001). Manipulating cereal endosperm structure, development and composition to improve end-use properties. Advances in Botanical Research Incorporating Advances in Plant Pathology 34, 165–236. Sikka, V. K., Choi, S. B., Kavakli, I. H., Sakulsingharoj, C., Gupta, S., Ito, H. and Okita, T. W. (2001). Subcellular compartmentation and allosteric regulation of the rice endosperm ADPglucose pyrophosphorylase. Plant Science 161, 461–468. Simmons, S. R. and Crookston, R. K. (1979). Rate and duration of growth of kernels formed at specific florets of spring wheat. Crop Science 19, 690–693. Singh, D. G., Lomako, J., Lomako, W. M., Whelan, W. J., Meyer, H. E., Serwe, M. and Metzger, J. W. (1995). -Glucosylarginine: a new glucose-protein bond in a self-glucosylating protein from sweet corn. FEBS letters 376, 61–64. Singh, R., Goyal, R. K., Bhullar, S. S. and Goyal, R. (1991). Factors, including enzymes, controlling the import and transformation of sucrose to starch in the developing sorghum caryopsis. Plant Physiology and Biochemistry 29, 177–183. Singletary, G. W., Banisadr, R. and Keeling, P. L. (1994). Heat stress during grain filling in maize: effects on carbohydrate storage and metabolism. Australian Journal of Plant Physiology 21, 829–841. Singletary, G. W., Banisadr, R. and Keeling, P. L. (1997). Influence of gene dosage on carbohydrate synthesis and enzymatic activities in endosperm of starchdeficient mutants of maize. Plant Physiology 113, 293–304. Smidansky, E. D., Clancy, M., Meyer, F. D., Lanning, S. P., Blake, N. K., Talbert, L. E. and Giroux, M. J. (2002). Enhanced ADP-glucose pyrophosphorylase activity in wheat endosperm increases seed yield. Proceedings of the National Academy of Sciences USA 99, 1724–1729. Smith, A. M. (1999). Making starch. Current Opinion in Plant Biology 2, 223–229. Smith, A. M., Zeeman, S. C. and Denyer, K. (2001). The synthesis of amylose. In ‘Starch: Advances in Structure and Function.’, (T. L. Barsby, A. M. Donald and P. J. Frazier, eds.), pp. 150–163. The Royal Society of Chemistry, Cambridge, UK. Smythe, C. and Cohen, P. (1991). The discovery of glycogenin and the priming mechanism for glycogen biogenesis. European Journal of Biochemistry 200, 625–631. Sofield, I., Wardlaw, I. F., Evans, L. T. and Zee, S. Y. (1977a). Nitrogen, phosphorus and water contents during grain development and maturation in wheat. Australian Journal of Plant Physiology 4, 799–810. Sofield, I., Evans, L. T., Cook, M. G. and Wardlaw, I. F. (1977b). Factors influencing the rate and duration of grain filling in wheat. Australian Journal of Plant Physiology 4, 785–797. Stinard, P. S. (1992). Sugary-3 (su3), a new endosperm mutant with sugary/shrunken phenotype. Maize Genetics Cooperation Newsletter 66, 4–5. Stinard, P. S., Robertson, D. S. and Schnable, P. S. (1993). Genetic isolation, cloning, and analysis of a mutator-induced, dominant antimorph of the maize amylose extender1 locus. Plant Cell 5, 1555–1566. Stitt, M. (1999). The first will be last and the last will be first: non-regulated enzymes call the tune? In ‘‘Plant Carbohydrate Biochemistry’’, (J. A. Byrant, M. M. Burrel and N. J. Kruger, eds.), BIOS Scientific Publishers Ltd., Oxford, UK. Sullivan, T. D. and Kaneko, Y. (1995). The maize brittle1 gene encodes amyloplast membrane polypeptides. Planta 196, 477–484. Sun, C., Sathish, P., Ahlandsberg, S. and Jansson, C. (1998). The two genes encoding starch-branching enzymes IIa and IIb are differentially expressed in barley. Plant Physiology 118, 37–49.

STARCH SYNTHESIS IN CEREAL GRAINS

59

Takeda, Y., Shitaozono, T. and Hizukuri, S. (1990). Structures of sub-fractions of corn amylose. Carbohydrate Research 199, 207–214. Takeda, Y., Guan, H. P. and Preiss, J. (1993a). Branching of amylose by the branching isoenzymes of maize endosperm. Carbohydrate Research 240, 253–263. Takeda, C., Takeda, Y. and Hizukuri, S. (1993b). Structure of the amylopectin fraction of amylomaize. Carbohydrate Research 246, 273–281. Tateoka, T. (1962). Starch grains of endosperm in grass systematics. Botanical Magazine – Tokyo 75, 377–383. Tatge, H., Marshall, J., Martin, C., Edwards, E. A. and Smith, A. M. (1999). Evidence that amylose synthesis occurs within the matrix of the starch granule in potato tubers. Plant, Cell and Environment 22, 543–550. Tetlow, I. J., Davies, E. J., Vardy, K. A., Bowsher, C. G., Burrel, M. M. and Emes, M. J. (2003). Subcellular localisation of ADPglucose pyrophosphorylase in developing wheat endosperm and analysis of the properties of a plastidial isoform. Journal of Experimental Botany 54, 1–11. Thompson, D. B. (2000). On the non-random nature of amylopectin branching. Carbohydrate Polymers 43, 223–239. Thorbjornsen, T., Villand, P., Denyer, K., Olsen, O.-A. and Smith, A. M. (1996a). Distinct isoforms of ADPglucose pyrophosphorylase occur inside and outside the amyloplasts in barley endosperm. Plant Journal 10, 243–250. Thorbjornsen, T., Villand, P., Kleczkowski, L. A. and Olsen, O.-A. (1996b). A single gene encodes two different transcripts for the ADP-glucose pyrophosphorylase small subunit from barley (Hordeum vulgare). Biochemical Journal 313, 133–138. Tiessen, A., Hendriks, J. H. M., Stitt, M., Branscheid, A., Gibon, Y., Farre, E. M. and Geigenberger, P. (2002). Starch synthesis in potato tubers is regulated by post-translational redox modification of ADP-glucose pyrophosphorylase: A novel regulatory mechanism linking starch synthesis to the sucrose supply. Plant Cell 14, 2191–2213. Tollenaar, M. and Lee, E. A. (2002). Yield potential, yield stability and stress tolerance in maize. Field Crops Research 4050, 1–9. Tomlinson, K. L., Lloyd, J. R. and Smith, A. M. (1997). Importance of isoforms of starch-branching enzyme in determining the structure of starch in pea leaves. Plant Journal 11, 31–43. Tsai, C.-Y. and Nelson, O. E. (1966). Starch-deficient maize mutant lacking adenosine diphosphate glucose pyrophosphorylase activity. Science 151, 341–343. Ugalde, T. D. and Jenner, C. F. (1990a). Route of substrate movement into wheat endosperm. 1. Carbohydrates. Australian Journal of Plant Physiology 17, 693–704. Ugalde, T. D. and Jenner, C. F. (1990b). Substrate gradients and regional patterns of dry-matter deposition within developing wheat endosperm. 1. Carbohydrates. Australian Journal of Plant Physiology 17, 377–394. Ugalde, T. D. and Jenner, C. F. (1990c). Substrate gradients and regional patterns of dry-matter deposition within developing wheat endosperm. 2. Amino-acids and protein. Australian Journal of Plant Physiology 17, 395–406. Umemoto, T., Yano, M., Satoh, H., Shomura, A. and Nakamura, Y. (2002). Mapping of a gene responsible for the difference in amylopectin structure between japonica-type and indica-type rice varieties. Theoretical and Applied Genetics 104, 1–8.

60

K. TOMLINSON and K. DENYER

Van de Wal, M., D’Hulst, C., Vincken, J.-P., Bule´on, A., Visser, R. and Ball, S. (1998). Amylose is synthesised in vitro by extension of and cleavage from amylopectin. Journal of Biological Chemistry 273, 22232–22240. Vilhar, B., Kladnik, A., Blejec, A., Chourey, P. S. and Dermastia, M. (2002). Cytometrical evidence that the loss of seed weight in the miniature1 seed mutant of maize is associated with reduced mitotic activity in the developing endosperm. Plant Physiology 129, 23–30. Vrinten, P. L. and Nakamura, T. (2000). Wheat granule-bound starch synthase I and II are encoded by separate genes that are expressed in different tissues. Plant Physiology 122, 255–263. Waigh, A. T., Kato, K. L., Donald, A. M., Clarke, C. J. and Riekel, C. (2000). Sidechain liquid-crystalline model for starch. Starch/Sta¨rke 52, 450–460. Wang, H. L., Patrick, J. W., Offler, C. E. and Wardlaw, I. F. (1993a). A novel experimental system for studies of photosynthate transfer in the developing wheat-grain. Journal of Experimental Botany 44, 1177–1184. Wang, Y. J., White, P., Pollak, L. and Jane, J. (1993b). Characterization of starch structures of 17 maize endosperm mutant genotypes with Oh43 inbred line background. Cereal Chemistry 70, 171–179. Wang, T. L., Bogracheva, T. Y. and Hedley, C. L. (1998). Starch: as simple as A, B, C? Journal of Experimental Botany 49, 481–502. Watson, S. A. (1987). In ‘Corn: Chemistry and Technology’, (S. A. Watson and P. E. Ramstad eds.), pp. 53–82. American Association of Cereal Chemists, Minnesota, USA. Watson, S. A. and Ramstad, P. E. (1987). ‘Corn: Chemistry and Technology’, American Association of Cereal Chemists, Minnesota, USA. Wattebled, F., Buleon, A., Bouchet, B., Ral, J. P., Lienard, L., Delvalle, D., Binderup, K., Dauvillee, D., Ball, S. and Hulst, D. (2002). Granule-bound starch synthase I. A major enzyme involved in the biogenesis of B-crystallites in starch granules. European Journal of Biochemistry 269, 3810–3820. Wiegand, C. L. and Cuellar, J. A. (1981). Duration of grain filling and kernel weight of wheat as affected by temperature. Crop Science 21, 95–101. Wolf, M. J., Buzan, C. L., MacMasters, M. M. and Rist, C. E. (1952). Structure of the mature corn kernel. I. Gross anatomy and structural relationships. Cereal Chemistry 29, 321–333. Yang, J., Zhang, J., Wang, Z., Zhu, Q. and Liu, L. (2002). Activities of enzymes involved in sucrose-to-starch metabolism in rice grains subjected to water stress during filling. Field Crops Research 4144, 1–13. Yoo, S. H. and Jane, J. (2002a). Molecular weights and gyration radii of amylopectins determined by high-performance size-exclusion chromatography equipped with multi-angle laser-light scattering and refractive index detectors. Carbohydrate Polymers 49, 307–314. Yoo, S. H. and Jane, J. (2002b). Structural and physical characteristics of waxy and other wheat starches. Carbohydrate Polymers 49, 297–305. Young, T. E., Gallie, D. R. and DeMason, D. A. (1997). Ethylene-mediated programmed cell death during maize endosperm development of wild-type and shrunken2 genotypes. Plant Physiology 115, 737–751. Young, T. E. and Gallie, D. R. (1999). Analysis of programmed cell death in wheat endosperm reveals differences in endosperm development between cereals. Plant Molecular Biology 39, 915–926. Young, T. E. and Gallie, D. R. (2000). Programmed cell death during endosperm development. Plant Molecular Biology 44, 283–301.

STARCH SYNTHESIS IN CEREAL GRAINS

61

Zeeman, S. C., Umemoto, T., Lue, W. L., Au-Yeung, P., Martin, C., Smith, A. M. and Chen, J. (1998). A mutant of Arabidopsis lacking a chloroplastic isoamylase accumulates both starch and phytoglycogen. Plant Cell 10, 1699–1711. Zeeman, S. C., Smith, S. M. and Smith, A. M. (2002). The priming of amylose synthesis in Arabidopsis leaves. Plant Physiology 128, 1069–1076.

This Page Intentionally Left Blank

The Hyperaccumulation of Metals by Plants

MARK R. MACNAIR

School of Biological Sciences, University of Exeter, Hatherly Laboratories, Prince of Wales Rd, Exeter, EX4 4PS

I. II. III. IV. V.

VI.

VII.

VIII. IX.

The Phenomenon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Taxonomic Distribution of Hyperaccumulation . . . . . . . . . . . . . . . . . . . . . . . Variation Within Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Genetics of Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Mechanism of Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. From Soil to Plant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Into the Plant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Where Does it End Up? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Control of Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evolution of Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Increased Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Defence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Inadvertant Uptake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Drought Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Allelopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Ecological Consequences of Hyperaccumulation . . . . . . . . . . . . . . . . . A. Effects on Other Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Effects on Herbivores and Higher Trophic Levels . . . . . . . . . . . . . . . . . C. Co-evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Phytoremediation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Advances in Botanical Research Vol. 40 incorporating Advances in Plant Pathology ISBN 0-12-005940-1

64 69 71 73 74 74 76 80 80 82 83 84 88 89 90 90 91 91 92 93 98 98

Copyright  2003 Elsevier Ltd All rights of reproduction in any form reserved

64

M. R. MACNAIR

ABSTRACT Hyperaccumulation is a fascinating phenomenon in which plants accumulate unusually large amounts of metals in their aerial parts. In this review the definition of the phenomenon is considered, and it is noted that there has probably been overreporting of it; certainly only the hyperaccumulation of nickel and zinc, and to a lesser extent cadmium and arsenic, have been established and studied experimentally. The character appears to have evolved independently on a number of occasions, but the number of evolutionary events is much less than the number of observed species, since some whole clades appear to show the phenomenon. Hyperaccumulation appears generally to be a species-level phenomenon, though there is within-species variation in degree and specificity of accumulation. The possible mechanisms of accumulation are reviewed. There is evidence that hyperaccumulators preferentially grow roots in areas of high metal concentration; have elevated levels of uptake into root cell symplasm; and have reduced root vacuolar transport. The potential reasons for the evolution of the character are explored. There is some evidence that the elevated level of metals may protect plants from herbivory and parasitism, but there are problems with this hypothesis, especially for the evolution of zinc hyperaccumulation. The ecological effects of hyperaccumulation are less well studied, but should be a priority for research if the phenomenon is utilised to remediate metal contaminated land, or to mine metals on a widespread scale.

I. THE PHENOMENON Metal hyperaccumulation is a fascinating phenomenon in which plants accumulate exceptional amounts of metal in their aerial parts. An extreme example is the New Caledonian tree, Sebertina acuminata, which when damaged oozes a bright blue latex containing over 11% nickel (25% by dry weight) (Jaffre´ et al., 1976). The phenomenon is exciting considerable interest at the moment, particularly from the perspectives of genetics and physiology (how do they do it?), evolution (why do they do it?) and the use of these plants as a practical technology to clean up metal contaminated land (phytoremediation). One of the first records of a plant containing unusually high quantities of metal was due to Baumann (1885) who discovered exceptionally high concentrations of zinc in the leaves of Thlaspi calaminare (¼ Thlaspi caerulescens). The term hyperaccumulation was coined by Brooks et al. (1977) in studies of the nickel content of plants growing on serpentine soils. Since then, Brooks and his colleagues, and others, have catalogued a large number of plants with high levels of metals in their aerial parts, and have defined critical values of metals that must be exceeded in plant specimens in order for them to be so categorised (Table I). The most recent consensus value of total number of hyperaccumulating species for each metal is also given in Table I (Baker et al., 2000). However, these figures must be regarded with some caution. The basic methodology used in many of these studies is to trawl through herbarium

THE HYPERACCUMULATION OF METALS BY PLANTS

65

TABLE I Numbers of species described as hyperaccumulators of various metals, together with the threshold required to be so defined Metal

Threshold (% leaf dry wt)

No of species

>0.01 >0.1 >0.1 >0.1 >1.0 >0.1 >1.0

1 28 37 14 5 317 11

Cadmium Cobalt Copper Lead Manganese Nickel Zinc Source: Baker et al. (1999).

specimens and to test field-collected leaves for metal contents. If a single specimen is found above the thresholds given in Table I, that species is deemed a hyperaccumulator, and is added to the catalogue (e.g., Reeves, 1992). This methodology is clearly open to a number of criticisms. First, herbarium specimens are liable to be more or less contaminated with soil. If the soil is highly contaminated, even tiny amounts attached to the outside of the leaves will lead to an erroneously high value for the internal concentration. It is impossible to be sure that all specimens are completely free of contamination, though Reeves (1992) suggests that contamination can be inferred if unusually high concentrations of elements such as Fe or Cr co-occur. Secondly, because of individual variation, the probability of detecting a hyperaccumulator species will depend on the number of specimens tested, as well as the mean level of metal in individuals of that species. Thus there is a great danger of an ascertainment bias, since workers may tend to preferentially test many specimens in species which have been well-collected, or are related to species that have previously been shown to be hyperaccumulators. Finally, of course, there is potential for great error caused by inaccurate or invalid taxonomy of the herbarium sheets. It is thus clear that the occurrence of the phenomenon has been overreported. Consider just two examples. a. Brooks (1998) listed 16 species that were said to hyperaccumulate zinc (Table II). Of these one (Silene cucubalus) does not exceed the 1% threshold; three (Arenaria patula, Haumaniastrum katangense and Viola calaminaria) are based on a very small number of positive ‘hits’, and must be considered doubtful accumulators (note that Paton and Brooks (1996) in a wide survey of H. katangense found no evidence of it being a zinc

66

M. R. MACNAIR

TABLE II List of species given by Brooks (1998) as zinc hyperaccumulators, together with maximum zinc content recorded Species

%Zn

Arenaria patula Cardaminopsis halleri Haumaniastrum katangense Noccaea eburneosa Silene cucubalus Thlaspi alpestre T. brachypetalum T. bulbosum T. caerulescens T. calaminare T. limosellifolium T. praecox T. rotundifolium T. stenopterum T. tatraense Viola calaminaria

1.31 1.36 1.98 1.05 0.47 2.50 1.00 1.05 2.73 3.96 1.10 2.10 2.10 1.60 2.70 1.00

Comment Doubtful record ¼ Arabidopsis halleri very doubtful record ¼ S. vulgaris; not exceeding threshold ¼ T. caerulescens

¼ T. caerulescens

¼ T. brachypetalum ssp. tatrense Doubtful record

accumulator). Apart from Cardaminopsis halleri (¼ Arabidopsis halleri), all the other species are members of the section of the Thlaspi genus that Meyer (1973) separated into the genus Noccaea and which Mummenhoff et al. (1997) showed was a genuine monophyletic clade. Of these species, several are synonyms. In fact, it is likely that there are only nine genuine zinc hyperaccumulators in this list, and since the Thlaspi species are closely related, it is possible that these only represent two independent evolutionary events leading to the development of this character. Note, however, the recent report of Long et al. (2002) that Sedum alfredi shows this character while three other Sedum species do not. b. Copper hyperaccumulation has been reported in a number of species from the Shaba copper belt in Zaire. Many of the reports are based on a single field-collected specimen, and the only published report on experimental work on these plants (Morrison et al., 1979) failed to replicate the phenomenon. We have also grown a number of these species in the laboratory from seeds collected from the wild under controlled conditions and in no case have we been able to find any evidence that they accumulate large amounts of this metal. They are all extremely tolerant of copper, as befits plants that grow in highly contaminated soil, but none behave substantially different to the highly copper-tolerant plant, Mimulus guttatus (which Macnair, 1981, showed was not an accumulator) or related ‘normal’ agricultural species (Table III). Of course, this limited sample does not prove that some of the untested

THE HYPERACCUMULATION OF METALS BY PLANTS

67

TABLE III Copper concentration of roots and shoots (g g dry wt1) of alleged copper hyperaccumulators compared with the copper-tolerant, non-accumulating species Mimulus guttatus and two cultivated ‘‘normal’’ species, Vigna radiata (mung beans) and Ipomoea batatas (sweet potato). A: Plants grown at 24 M Cu in hydroponic solution for 4 weeks. B: Plants grown for 4 weeks at 48 M Cu Status1

Root conc.

Shoot conc.

A. Vigna radiata Ipomoea batatas V. dolomitica I. alpina Mimulus guttatus

N N H H T

9045±435 5491±302 2107±304 3978±294 4557±282

370±39 277±37 52±7 54±18 117±16

B. V. dolomitica I. alpina Rendlia cupricola Aeollanthus biformifolius

H H H H

5495±619 8201±1311 7562±1850 19822±4828

131±23 164±37 160±17 128±28

Species

1

N: Normal plant; T: Copper-tolerant plant; H: Copper hyperaccumulator (Brooks, 1998).

species are not accumulators, nor that there are no accumulating genotypes of these species that we have not tested. In addition, it may be that accumulation will occur in the field whereas it will not occur in the laboratory. This could occur if, for instance, a mycorrhizal symbiont were required to manifest the phenomenon. Also, one of the tested species, Aeollanthus biformifolius, is a perennial, regenerating from corms. The author’s laboratory could never get the corms to sprout. It is conceivable that only shoots regenerating from corms have the levels of copper reported by Malaisse et al. (1978), as suggested by Morrison et al. (1979). However, the author remains sceptical that copper hyperaccumulation is a genuine phenomenon. By far the largest proportion of hyperaccumulators are reported to be accumulators of nickel. All known accumulators of this metal are found on serpentine (¼ultramaphic) soils, which tend to be naturally contaminated with nickel. Many are serpentine endemics, i.e. are only found on this substrate. There may be some over-reporting, but in general this phenomenon is well-established and widespread. In some floras, the distinction between accumulators and non-accumulators is clear (Fig. 1) while in others there is no obvious disjunction (Fig. 2). The original formulation of the phenomenon was based on disjunct distributions such as

68

M. R. MACNAIR

Fig. 1. Maximum values of Ni content (mg g1) in 173 species of Alyssum. The dotted line indicates the hyperaccumulator threshold. (Data from Brooks et al., 1979).

Fig. 2. Maximum values of Ni content (mg g1) in 116 species from serpentine soils in Cuba. The dotted line indicates the hyperaccumulator threshold. (data from Reeves et al., 1999).

in Fig. 1; it is questionable whether, if Cuban plants had been studied earlier than Mediterranean ones, hyperaccumulation would have been defined in the way it currently is, with the same arbitrary thresholds. An increasing number of species are being tested under standard laboratory conditions, and in all these genuine hyperaccumulators show a pattern of accumulation in which shoot concentrations exceed root concentrations, while related ‘normal’ species do not (e.g., zinc: Baker et al., 1994b; Zhao et al., 2000; Assunc¸a˜o et al., 2001; Long et al., 2002; Macnair, 2002; nickel: Kra¨mer et al., 1996; Assunc¸a˜o et al., 2001; arsenic: Zhao et al., 2002a; but not cadmium: Lombi et al., 2000). Baker (1981) first drew attention to the difference between accumulators and normal plants

THE HYPERACCUMULATION OF METALS BY PLANTS

69

Fig. 3. The relationship between soil metal concentration and plant aerial part concentration in normal plants (solid line) and accumulators (dotted line). If hyperaccumulation is defined by an arbitrary concentration, soil metal concentrations must exceed some threshold for plants to exhibit the phenomenon.

(Fig. 3) and noted the property that these species had a shoot : root ratio >1. For plants to exceed the arbitrary hyperaccumulator threshold they have to be growing in a medium with a sufficient metal, and be sufficiently tolerant to survive. These latter properties may be less interesting or unusual than the difference between the accumulator/non-accumulator phenotype (i.e. a shoot : root ratio>1), but clearly this cannot be determined on field-collected plants.

II. TAXONOMIC DISTRIBUTION OF HYPERACCUMULATION It is clear that there is some taxonomic grouping of hyperaccumulators. For instance, a substantial number of hyperaccumulators occur in the Brassicaceae. However, since many of these occur in the genera Alyssum and Thlaspi, it is possible that the number of evolutionary events leading to this character in this family may be small. In Alyssum, all accumulators occur in the Odontarrhena section of the genus (Brooks et al., 1979), while in Thlaspi, all known accumulators occur in the Noccaea, Thlaspiceras, and Raparia sections of the genus, which Mummenhoff et al. (1997) show is a monophyletic clade. It is possible that the evolution of hyperaccumulation occurred just once in this genus, at the base of this clade. We are currently testing this hypothesis: preliminary results (S.I. Taylor, unpublished) indicate that almost all species in this clade accumulate metals.

70

M. R. MACNAIR

This includes species not normally found on contaminated soils, and which have not hitherto been classified as hyperaccumulating. Apart from this clustering in the Brassicaceae, there is little evidence of a widespread taxonomic bias. However, the uncertainties about the definition of this character noted above, and the undoubted over-reporting of the phenomenon, may have led to a blurring of the pattern so that a taxonomic signal cannot be detected. Borhidi (2001) has reviewed the taxonomic distribution of nickel hyperaccumulation. It is overwhelmingly concentrated in the dicotyledons, amongst which he notes three broad groups. In the extratropical flora, 93 out of 110 species occur in the Brassicaceae, but the other species are distributed widely amongst plant families. In the tropics, two groups are apparent: the old palaeotropical flora distributed on serpentines widely across geographic regions, and the young isolated Cuban flora which contains a large number of hyperaccumulators (see Fig. 2). In the old Palaeotropical flora, there is little pattern, as found in the temperate regions. Amongst the Cuban flora, however, Borhidi (2001) has found some intriguing patterns associated with potential ecological/adaptive features. Firstly, families with high alkaloid contents do not appear to contain nickel hyperaccumulators, though they do have serpentine endemics. In the Rubiaceaea, which generally has a high alkaloid content, hyperaccumulation was only found in tribes within the family that have low alkaloid levels. There was no correlation, however, with other secondary compounds such as aromatic substances or volatile oils. Secondly, he found a negative correlation with families which are specialised for nutrient poor soils, even though these families are well represented on the poor soils found on serpentine. Finally, he found that climbing species (lianes and vines) are not hyperaccumulators. Another approach was taken by Broadley et al. (2001) who tested whether the pattern of accumulation of metals by ‘normal’ species was taxonomically correlated with the occurrence of hyperaccumulation. They searched the literature for comparative studies that had looked at the accumulation of heavy metals in the laboratory, and tested whether different plant orders had different patterns of accumulation of metals. They found that there was a substantial difference between orders, particularly for zinc and nickel, with about 45% of the total sum-of-squares being accounted for by this taxonomic level. This indicates that there are fundamental differences in plant mineral nutrition between plant orders, which suggests that these evolved a long time ago. However, there proved to be little correlation between this character and hyperaccumulation, since the orders with the highest mean accumulation are not always the ones with the largest number of hyperaccumulators. This suggests that hyperaccumulation has evolved as

THE HYPERACCUMULATION OF METALS BY PLANTS

71

a distinct adaptation unrelated to the normal pattern of mineral acquisition by the relatives of the species which show this character.

III. VARIATION WITHIN SPECIES There are clearly differences between closely related species in the character ‘hyperaccumulation’. Thus Arabidopsis halleri hyperaccumulates zinc, while A. petraea (its sister species, Koch et al., 1999) does not. Similarly the Californian serpentine endemic Streptanthus polygaloides is the only recorded hyperaccumulator in this genus, though other species are also part of the serpentine flora (Boyd and Moar, 1999). So one would expect there to be some species in which some populations hyperaccumulate, and others do not. There are very few, if any, examples, however. One possible example is Senecio coronatus, where plants growing on some serpentine soils hyperaccumulate nickel, while those on others do not (Boyd et al., 2002). However, the author is not aware of any test of this species under controlled conditions to determine whether this phenotypic difference is caused by their innate biology or the particular properties of the substrates on which they are found. Within-species variation has been most rigorously investigated in T. caerulescens, and, to a lesser extent, in A. halleri, though Boyd and Martens (1998) have also investigated this phenomenon in T. montanum and found nickel accumulation to be a constitutive trait. All populations of T. caerulescens tested to date have proved to hyperaccumulate zinc and/or nickel and frequently cadmium as well (Ingrouille and Smirnoff, 1986; Baker et al., 1994b; Lloyd-Thomas, 1995; Pollard and Baker, 1996; Meerts and van Isacker, 1997; Assunc¸a˜o et al., 2001). However, the level of accumulation of different metals under standard conditions differs, with some populations preferentially accumulating zinc, others nickel (Lloyd-Thomas, 1995; Assunc¸a˜o et al., 2001). The level of accumulation of cadmium is particularly variable, with one population (Ganges, Lombi et al., 2000) accumulating this metal to over 10,000 ppm (most populations are around the 100 ppm range). There is some indication that for zinc accumulation, at least, populations from calamine habitats accumulate less zinc than populations from less zinc-contaminated sites (Meerts and van Isacker, 1997), though Assunc¸a˜o et al. (2001) found one population from uncontaminated soil that had very low levels of metal uptake. However, it is clear that the variation between populations is large, and it is not yet known what environmental factors correlate with this between-population variation. In A. halleri, Bert et al. (2002a) studied 33 populations from both contaminated and

72

M. R. MACNAIR

uncontaminated sites. They looked at field-collected material, and found the variation between and within populations in zinc and cadmium levels to be enormous. Bert et al. (2000) looked at two populations under standard conditions and found that the population from an uncontaminated site accumulated more zinc than that from the calamine site. However, Macnair (2002) has studied 17 populations under standard conditions, and found that, though the variation between populations in zinc-accumulating ability was large, there was no relationship between accumulating ability and zinc contamination of the population (Fig. 4). Genetic variation within populations has been even less studied. Pollard and Baker (1996) investigated two populations of the self-fertilising T. caerulescens. They found evidence of heritable variation in one of them for zinc accumulation. Meerts and van Isacker (1997) investigated four populations of this species, two from metallicolous sites, and two from normal soils. Significant within-population variation for zinc accumulation was only found in the two populations from normal soils: they suggested that natural selection on metallicolous soils might have exhausted any genetic variation originally present. Pollard et al. (2000) studied the heritability of both zinc and nickel accumulation in five populations of T. caerulescens. They also found that the population from an uncontaminated site had the highest heritability for metal accumulation. Macnair (2002) has studied three populations of the out-crossing A. halleri. Significant heritability for zinc accumulation was found in all three populations, and a small but significant amount of variation in zinc content between plants in

Fig. 4. The relationship between total soil zinc concentration, and the mean zinc accumulation phenotype in 17 populations of Arabidopsis halleri. Zinc accumulation was measured under standard conditions, and the value plotted is the first principal component of accumulation in three separate external concentrations. The open circles are populations where the zinc soil concentration was estimated. (Redrawn from data in Macnair, 2002).

THE HYPERACCUMULATION OF METALS BY PLANTS

73

Fig. 5. The zinc content of 530 Arabidopsis halleri plants from a variety of populations grown in 10 mM Zn for 4 weeks. (Redrawn from data in Macnair, 2002.)

the field could be accounted for by their genotype. Intriguingly, heritability was greatest when plants were grown in low external zinc concentrations, suggesting that selection for increased accumulating ability will be easiest at such low concentrations. Figure 5 shows the distribution of leaf zincconcentrations after 5 weeks at low external zinc concentration – it is tempting to interpret this graph as showing a polymorphism for zinc accumulating ability. At higher external zinc concentrations, all plants hyperaccumulate zinc (Macnair, 2002).

IV. GENETICS OF ACCUMULATION Because of the lack of intraspecific variation in accumulation noted above, it has proved difficult to investigate the formal genetics of accumulation. To do this, one needs to be able to cross an accumulator with a nonaccumulator, and study the patterns in segregating families. This is not possible in the model hyperaccumulator, T. caerulescens, though we have been able to make crosses between A. halleri and its non-accumulating sister species, A. petraea. Macnair et al. (2000) showed that the F1 of this cross was zinc-tolerant, and had an accumulating phenotype similar in mean to A. halleri. Macnair et al. (1999) studied the F2, and showed that the character of hyperaccumulation was both highly heritable and highly variable in this family. No segregation into distinct classes was apparent, and it was speculated that the number of genes governing the character was probably not many, but equally probably not just one. Analysis of the relationship between zinc tolerance and zinc accumulation in this segregating F2 indicated that the two characters were genetically independent (Macnair et al., 1999), and thus have different genes governing

74

M. R. MACNAIR

them and different genetic bases. This independence is also implied by the patterns of accumulation and tolerance in T. caerulescens, where populations of very limited zinc tolerance, but high zinc accumulation, have been found (Ingrouille and Smirnoff, 1986; Meerts and van Isacker, 1997; Assunc¸a˜o et al., 2001). Bert et al. (2002b) have produced a segregating backcross to A. petraea from the A. halleri  A. petraea cross. They have investigated cadmium hyperaccumulation in this cross. They also find that there is no evidence for a single major gene for this character, and find only partial correlation between cadmium and zinc accumulation. This suggests that some, but not all, of the genes involved in zinc accumulation may also pleiotropically give cadmium accumulation.

V. THE MECHANISM OF ACCUMULATION In considering the mechanism of accumulation, we need to consider how the metal gets from the soil into the plant; how it gets into the root symplasm; how it gets from root to shoot; and where and in what form the metal is stored (Fig. 6).

A. FROM SOIL TO PLANT

There has been some discussion as to whether hyperaccumulators are accessing a different soil metal pool than normal plants. This could occur if the plant had the ability to search for (Fig. 6a) or mobilise a different source of metal (Fig. 6b,c), or if it possessed microbial partners (mycorrhizae, endophytes or rhizosphere partners) that were able to do this. There is evidence that the zinc hyperaccumulators T. caerulescens and A. halleri do actively search for zinc. Schwartz et al. (1999) and Whiting et al. (2000), using rhizoboxes which contained two different soils of differing zinc content, found that T. caerulescens would preferentially grow roots in soils contaminated with zinc or cadmium. Haines (2002) also found this to be true, but only in one population of T. caerulescens; a second population did not show the response. The author’s laboratory has investigated this aspect of plant behaviour in A. halleri using a hydroponic split-pot experiment. Plants were established with their roots split between two solutions, where the two solutions could have either the same or different concentrations of zinc. Figure 7 shows that A. halleri plants preferentially grow their roots in the higher concentration. A. thaliana (A. halleri’s sister species) did not show this behaviour. In toxic soils, a nontolerant species is likely to inhibit its growth in the most contaminated soil

THE HYPERACCUMULATION OF METALS BY PLANTS

75

Fig. 6. Potential ways in which a hyperaccumulator could differ from normal plants so as to increase accumulation of a metal ion (Mþ) from the soil. (A) Preferential root proliferation in areas of local high [Mþ]. (B) Increased release of metal from insoluble forms (M-X). (C) Increased release of metal adsorbed to soil particles. (D) Increased flux into the symplasm, decreased flux into the root vacuole, and/or increased flux from root symplasm into xylem. (E) Increased uptake from xylem into leaf symplasm and/or increased transport into leaf vacuoles. Size of arrows indicates magnitude of a flux; dotted lines indicate the hyperaccumulator state.

patches; if a hyperaccumulator actively grows its roots in such a patch, it is inevitable that its metal concentration will ultimately be higher. The effect of this behaviour on shoot metal contents has been investigated by Haines (2002). He grew T. caerulescens plants in pots which either had uncontaminated soil, contaminated soil, or a heterogeneous mixture of contaminated and uncontaminated soil. He ensured that the total zinc available to the plants was the same in the homogeneous and heterogeneous pots. He found that the plants grew best in the heterogeneous soil, and put 67% of their roots in the contaminated patches. At the end of the experiment, the total zinc content of the plants was the same, whether they had their roots in homogeneous or heterogeneous soil. There is some suggestion that roots of T. caerulescens can acidify the soil or growing medium, and release zinc from insoluble forms (Knight et al., 1997). Many other species can also do this, however, and it is not clear that

76

M. R. MACNAIR

Fig. 7. Relative rootgrowth of plants in a split pot experiment. Arabidopsis halleri seedlings were established with their roots distributed between two pots, with the right hand pot always having the same or higher Zn concentration than the left hand one. Seedlings were grown for 4 weeks and root biomass in each pot determined. Data is the difference between RHS and LHS expressed as a percentage of total root biomass ( SE). Data points are labelled as LHS/RHS in mM Zn.

hyperaccumulator species have a specifically greater ability to mobilise insoluble forms. Whiting et al. (2001a,c) investigated the ability of T. caerulescens to mobilise zinc from insoluble forms, such as ZnS. They used both accumulation by T. caerulescens (Whiting et al., 2001c) and a bioassay using non-accumulator species grown in the same pot (Whiting et al., 2001a), to assess zinc mobilisation. They found very little evidence that T. caerulescens was in fact mobilising these insoluble zinc fractions. Zhao et al. (2001) compared the root exudates of T. caerulescens with those of canola and wheat in their ability to mobilise zinc, and found no evidence for increased efficiency. Indeed, when grown in low iron or zinc, the wheat exudates were more efficient. Salt et al. (2000) investigated the root exudates of the hyperaccumulator T. goesingense and compared them with the nonaccumulator T. arvense. They found that histidine was a component of the exudates of both species, but concentrations were greater in T. arvense, and that the accumulator did not produce more histidine in response to the metal. They concluded that the release of acids such as histidine or citrate was not an important part of the hyperaccumulator response. Thus any effect of the plant on metal solubilisation is likely to be small.

B. INTO THE PLANT

Once the metal has reached the root, it has to be taken up by the plant (Fig. 6c) and translocated to the leaf cells (Fig. 6d). Theoretically, there are

THE HYPERACCUMULATION OF METALS BY PLANTS

77

two routes that an ion could use to reach the leaves: the apoplastic or symplastic. Since the Casparian strip separates the root xylem from the soil solution, it has been normally assumed that most metal must pass into a cell, and thus the symplastic route is the most important. This view has been challenged by White et al. (2002) who used the published fluxes of zinc across the root cell membrane (Lasat et al., 1996; Pence et al., 2000) to calculate the maximum rate of zinc accumulation in the shoots. They find that zinc is observed to accumulate faster in the shoots than is possible according to their calculations. They also found that the relative growth rate (RGR) and the shoot to root fresh weight ratio of the plants were critical: if plants grew too fast, or had too high a shoot : root ratio, it was impossible for zinc to be transported across the cell membrane fast enough to deliver sufficient zinc to the shoots to reach the observed shoot concentrations. This argument has been countered by Ernst et al. (2002) who argue that metal uptake by hyperaccumulators is too specific for the apoplastic route to play any major part in metal uptake, since an apoplastic pathway should not differentiate between ions. In addition, there would have to be substantial structural differences in Casparian strip or root architecture to explain the difference between accumulators and related non-accumulators; such differences have not been seen. This debate will doubtless spur experiments to specifically distinguish the two routes. Considering the symplastic route, divalent hydrophilic ions cannot pass by diffusion into the root (Marshner, 1995), and thus must be transported across the cell membrane into the symplasm. Active ion uptake tends to follow Michaelis–Menten kinetics, and Lasat et al. (1996) compared Zn uptake by T. caerulescens and T. arvense. They found that the transport mechanisms of both species had similar affinity for zinc, but that T. caerulescens had a Vmax 4 higher than the corresponding value for T. arvense. This suggests that both species use essentially the same transport molecules to take zinc up from the medium (since the affinity was the same) but that T. caerulescens has a greater number of them (since the Vmax is higher). Lasat et al. (1998) found that the xylem sap of T. caerulescens had a 5 higher Zn concentration than in T. arvense, while T. arvense had a higher root vacuolar zinc content and a lower rate of efflux from vacuole to cytoplasm. There was no difference between the species in cell wall binding of the metal, or of cytoplasmic concentrations. This suggests that, in addition to the difference in the number of transport molecules taking zinc from soil to root cell, there are other differences between the species, so that T. arvense preferentially stores the zinc in its vacuoles, while T. caerulescens moves the zinc into the xylem. In the leaf, the zinc has to be taken up by leaf mesophyll cells. Lasat et al. (1998) also looked at the kinetics of zinc uptake

78

M. R. MACNAIR

by leaf discs and protoplasts. Much smaller differences were found between T. caerulescens and T. arvense in the pattern of zinc uptake into these materials, and what differences there were, were only found at high external zinc concentrations when the non-tolerant T. arvense might be expected to be suffering some toxicity. This might suggest that the fundamental mechanism responsible for hyperaccumulation in T. caerulescens is a ‘push’ from the roots, rather than a ‘suck’ from the leaves. The last few years have seen an explosion of knowledge concerning metal transporters, and it is clear that there are a number of different classes of transporters with different properties (e.g. see reviews by Guerinot, 2000; Williams et al., 2000; Ma¨ser et al., 2001). So far, transporters related to the iron transporter IRT1 have been most extensively studied with respect to zinc and cadmium uptake by hyperaccumulators. In Arabidopsis thaliana, Guerinot and colleagues have isolated a number of zinc transporters (ZIP family, ‘ZRT, IRT-like Protein’) (Grotz et al., 1998; Guerinot, 2000). ZIP1, ZIP3 and ZIP4 are all induced in plants grown in low zinc concentrations, the first two being found in the roots, while ZIP4 is found in both roots and leaves (Guerinot, 2000). Homologues of these genes have been found in T. caerulescens (ZNT1: Pence et al., 2000; ZNT1 and ZNT2: Assunc¸a˜o et al., 2001). ZNT1 shows great similarity to ZIP4. Both ZNT1 and ZNT2 show greater expression in T. caerulescens than in T. arvense, particularly in root tissue. In both species, ZNT1 expression is down-regulated in plants with relatively high zinc levels (Pence et al., 2000). In T. arvense, zinc deficient plants have a greater ZNT1 expression (as shown by Northern blotting) and a 2-fold greater Vmax for Zn uptake than plants grown in adequate or high zinc. In contrast, in T. caerulescens, ZNT1 transcription is only reduced at high external zinc levels, where the Vmax is also reduced 4-fold. Note that even at this reduced level it is still comparable with that of the up-regulated T. arvense. The pattern of expression clearly agrees with the physiological differences, and thus it is tempting to suggest that these genes are involved in the hyperaccumulation phenotype. Another zinc-transporting gene isolated in A. thaliana is ZAT (van der Zaal et al., 1999). Transgenic plants with this gene overexpressed show greater zinc tolerance and an increased root zinc content. van der Zaal et al. (1999) suggested that the ZAT protein is involved in internal compartmentation of zinc. Assunc¸a˜o et al. (2001) have also isolated the T. caerulescens homologue of this gene (ZTP1). This gene shows greater expression in leaves than in roots, and the expression is correlated with zinc tolerance, in that a population of T. caerulescens from a calamine site had higher expression than populations from serpentine or normal soil.

THE HYPERACCUMULATION OF METALS BY PLANTS

79

The mechanism of zinc accumulation in T. caerulescens is thus becoming clearer, but much more work is needed before we can understand fully the roles of the genes so far identified, or to determine whether any of the other classes of metal transporters (e.g., CPx-type ATPases and Nramp) that are known to be involved in metal homeostasis in other organisms are involved in the tolerance or accumulation syndrome. In contrast with zinc, much less is known about the physiology or molecular mechanisms of accumulation of cadmium or nickel. Zhao et al. (2002b) have recently studied the Ganges population of T. caerulescens, which is an exceptional accumulator of cadmium. They compared the kinetics of Cd uptake by this population with that of Prayon, a population that accumulates Cd more than ‘normal’ plants, but less than Ganges. They found that while Cd uptake in both populations obeyed Michaelis–Menten kinetics, the Ganges population had a 4.5-fold greater rate of Cd uptake than Prayon. In addition, they found that Zn2þ, Mn2þ or Ca2þ appeared to compete with Cd2þ uptake in the Prayon plants, but not in Ganges. This would suggest that Ganges has evolved a cadmium transporter that is different to that in Prayon, rather than just more highly expressed. It is surprising that nickel accumulation has attracted less attention. Kra¨mer et al. (1997) compared rates of nickel uptake by the nickel hyperaccumulator T. goesingense with the non-accumulator T. arvense. They found no evidence of greater transport of nickel in the accumulator, and suggested that it was the greater tolerance of the former that led to the hyperaccumulation phenotype. Kra¨mer et al. (1996) studied the nickel hyperaccumulator Alyssum lesbiacum and compared it to two nonaccumulating species. They found that the concentration of histidine was greatly elevated in the xylem sap of A. lesbiacum following nickel treatment, and that the concentration of nickel in the sap was highly correlated with histidine concentration in all three species. Histidine has the highest association constant for complex formation with nickel of all the amino or organic acids (Kra¨mer et al., 1996) and so it is tempting to suggest that a greater synthesis of this amino acid plays a major role in producing nickel accumulation. This is supported by the observation that exogenously supplied histidine increased both the tolerance to, and xylem transport of nickel by the non-accumulator A. montanum. Persans et al. (2001) have cloned a metal transporter gene belonging to the CDF family (as is the ZAT gene of A. thaliana and the ZNT1 gene of T. caerulescens, see above) from the nickel hyperaccumulator T. goesingense. They find that this gene is highly expressed in this species, and that when expressed in yeast, the gene confers a number of metal tolerances. Interestingly, in T. goesingense, the transcript can be found in two alternative versions,

80

M. R. MACNAIR

spliced and unspliced. It is the spliced version which confers greatest nickel tolerance in yeast.

C. WHERE DOES IT END UP?

Once in the leaf, the metal must be taken up by leaf cells and is presumably stored in the vacuoles. Whilst some authors have suggested that the metal may be stored in insoluble forms (e.g., Va´zquez et al., 1994), it seems likely that these results are artefacts of sample preparation (Ku¨pper et al., 1999). Ku¨pper et al. (1999) extracted the contents of single cells of T. caerulescens, and concluded that the zinc was in a soluble form. This agreed with the results of Zhao et al. (1998) which showed that most of the zinc in T. caerulescens was water soluble. In A. halleri, the fact that all the zinc can be extracted from leaves by very gentle extraction techniques (Macnair and Smirnoff, 1999; Zhao et al., 2000) would also suggest that the zinc is in soluble forms. It is likely that the metal ion is balanced by a variety of organic ligands, particularly malate, citrate and, in A. halleri, an unidentified organic acid. There is evidence that the metal may not be uniformly distributed across the leaf. In T. caerulescens, both upper and lower epidermis have about a 4 greater concentration of zinc than mesophyll cells (Ku¨pper et al., 1999), though Frey et al. (2000) showed that the metal seems to be excluded from the stomatal cells. Ku¨pper et al. (2001) and Psaras et al. (2000) have also found nickel to be preferentially located in the epidermis of a variety of nickel hyperaccumulators, with again lower concentrations in the stomata. In contrast, in A. halleri, the zinc appears to be largely stored in mesophyll cells, though there was some localisation at the base of the trichomes (Ku¨pper et al., 2000). Note, however, that even in species which have higher concentrations in the epidermis, the majority of the metal will actually be contained in the mesophyll, since this makes up such a greater proportion of the total leaf.

D. CONTROL OF ACCUMULATION

Macnair (2002) showed that the final concentration of zinc in leaves and plants of A. halleri is highly variable. Whilst some of this variation could be ascribed to genetic variation between individuals, there are clearly other factors which determine the final leaf concentration of metal. Young leaves have less zinc than old (Ku¨pper et al., 1999; Macnair and Smirnoff, 1999), but even old leaves have very different metal concentrations. Zinc levels do

THE HYPERACCUMULATION OF METALS BY PLANTS

81

not continue to increase indefinitely: mean zinc concentrations plateau out after a time (Macnair, 2002). Important factors influencing final leaf concentrations could be shoot : root ratios and the activity of metal transporters (White et al., 2002). Transpiration rates and thus the flow of soil solution could also have an effect. It is clear that in hyperaccumulators both root growth (Whiting et al., 2000) and activity of transporters (Pence et al., 2000) is under genetic regulation in response to external zinc. What is unknown is how any of these potential factors is regulated in response to internal zinc status. Thus does the plant sense leaf zinc concentration in some way, and control root or shoot growth, or transcription of zinc transporters as a result? Or is all control influenced only by external zinc concentration? The potential complex interrelations between these factors are illustrated by the data of Haines (2002). He grew T. caerulescens plants in three different treatments: a control soil, and two soils containing the same total amount of zinc per pot. In one of these the zinc was homogeneously distributed, while in the other it was heterogeneously distributed. As noted above, in the heterogeneous treatment the plants preferentially grew roots in the high zinc patches. The shoot dry mass and total root dry mass were also greatest in the heterogeneous treatment. However, the total zinc translocated to the shoots was the same for both zinc treatments, with the result that the shoot concentrations were lower in the heterogeneous treatment. This was not true for other nutrients investigated (N, P and K) where the 50% greater shoot mass was accompanied by a substantial increase (between 30% and 80%) in total shoot mineral content. Haines (2002) argues that, because only 67% of the plants’ roots were in the zinc patches in the heterogeneous soil, then the fact that the plants in both zinc treatments had similar zinc contents suggests that there must be another mechanism leading to greater uptake in the heterogeneous treatment. However, if one assumes that zinc uptake depends on total root mass, then the greater root biomass in the heterogeneous treatment means that there was actually a greater biomass of roots in the zinc soils in this treatment, and thus one would have expected a greater total zinc content in the heterogeneous treatment. Therefore the fact that it was not higher suggests that the roots in the heterogeneous treatment were actually less efficient at scavenging zinc than in the homogeneous treatment. It is possible that, because the external zinc concentration in the heterogeneous treatment was higher in the zinc patches than in the homogeneous treatment, zinc transport was downregulated (Pence et al., 2000), leading to lower overall accumulation. Clearly, there needs to be more research on the factors regulating final leaf metal concentration.

82

M. R. MACNAIR

VI. EVOLUTION OF ACCUMULATION The taxonomic distribution of hyperaccumulation, whereby in many of the species exhibiting this phenomenon it is clearly a derived character, strongly suggests that this character is currently adaptive. However, the direct evidence for selection developing or maintaining this character is thin. Huitson (unpublished) used the F2 between A. halleri and A. petraea to generate variation in this character, and planted out ramets of 400 clones in the field at two contaminated sites in northern France. The survivorship of the plants was monitored for several months, and she characterised the clones in the laboratory for their zinc-accumulating ability under standard conditions. She found (Fig. 8) that the plants with the greatest accumulating ability had the highest survivorship. This result is consistent with the hypothesis that accumulation is selected for in the field, though it remains possible that the character being selected is one that is genetically linked to the accumulation phenotype. However, this tells us nothing about the reasons for the difference in survivorship. Boyd and Martens (1992), in a classic review, discuss a variety of hypotheses for the adaptive role of hyperaccumulation. They identify five principal hypotheses: increased tolerance; protection against predators or pathogens; inadvertant uptake; drought tolerance; allelopathy.

Fig. 8. Survivorship of clones of the F2 of a cross between Arabidopsis halleri and A. petraea in contaminated sites in France. Ramets of 400 F2s were planted out in 3 contaminated sites in Northern France in April, and survivorship determined in July when overall survivorship was about 30%. The ability of the clones to accumulate zinc was determined in the laboratory under standard conditions at 75 mM Zn. The clones were grouped according to their Zn content, and the average proportion of surviving ramets was determined (Huitson and Macnair, unpublished).

THE HYPERACCUMULATION OF METALS BY PLANTS

83

A. INCREASED TOLERANCE

Under this hypothesis, hyperaccumulation is an additional or alternative tolerance mechanism to those employed by tolerant but non-accumulating species growing in the same, toxic environment. The mechanisms of tolerance are still largely obscure (see Hall, 2002, for a recent review), but part of the overall mechanism must involve the sequestration of the metal in a part of the cell or plant where the toxic metal cannot interfere with the normal cellular processes. In normal plants the metal is largely sequestered in root vacuoles; it is possible that sequestering the metal in leaves or shoots is an alternative or better mechanism. It is certainly true that hyperaccumulators tend to be very tolerant of the metal they hyperaccumulate. It is for instance, almost impossible to kill either A. halleri or T. caerulescens with zinc. There is, however, little evidence from the distribution of plants in the field that hyperaccumulators are more tolerant than other species living in the same habitat. All hyperaccumulators grow in environments where their roots intermingle with those of more normal plants. Since ultimately tolerance has to be defined ecologically – i.e. a tolerant plant is one that can grow where others cannot – there is no evidence of intrinsically greater tolerance of hyperaccumulators than non-accumulators. In addition, as we have seen, in the zinc accumulators T. caerulescens and A. halleri there is genetic evidence that tolerance and accumulation are genetically independent characters. Note, however, that the genetic evidence of independence relates only to zinc accumulation (and possibly cadmium accumulation in A. halleri, Lombi et al., 2001a; Bert et al., 2002b). No similar experiments have been possible with nickel accumulators, and it remains possible that hyperaccumulation of this element represents an alternative tolerance mechanism. Thus, Kra¨mer et al. (1996) found that histidine was associated with the accumulation of nickel by Alyssum lesbiacum (see above). They also found that the addition of histidine to the growth medium of the non-tolerant non-accumulating conspecific A. montanum increased both the tolerance to, and the accumulation of, this element. Kra¨mer et al. (1997) compared the Ni hyperacccumulator T. goesingense with the ‘normal’ species T. arvense. They found that at low concentrations of Ni there was little difference between the species in the way they took up this metal, but T. arvense was much more sensitive to it. They suggested that the difference between the species was solely caused by the differences in tolerance. Boyd et al. (2000) compared the nickel tolerance of three species: a hyperaccumulator (Streptanthus polygaloides), a congeneric serpentine endemic species that was not an accumulator (S. breweri) and a species that neither grew naturally on

84

M. R. MACNAIR

serpentine nor accumulated (Brassica oleracea). They showed that the order of nickel tolerance was hyperaccumulator> endemic>‘normal’. These data are consistent with nickel accumulation being a mechanism of nickel tolerance, but clearly the study is limited since only one species was studied of each type. Final resolution of this will only come when we have a clearer understanding of the genetics and molecular biology of these characters. Another potential relationship between tolerance and accumulation is the possibility that hyperaccumulation is a mechanism for excreting the metal. Once a metal has been taken up by a plant there may be rather little opportunity for disposal of the metal. Transporting the metal to the leaves, and then shedding them, offers an opportunity to dispose of the burden. However, ‘normal’ plants accumulate metals in their roots. In trees and perennial plants, roots ‘turn over’ in the soil just as leaves do above ground (Marschner, 1995). Thus it is not obvious that ‘‘churning’’ the metal through leaves is more efficient than through roots.

B. DEFENCE

All plants are continually challenged by predators, pathogens and parasites, and most plants expend much energy in secondary metabolism producing compounds of greater or lesser toxicity to their enemies. It is conceivable that it would be less costly to a plant which was growing in a soil containing high concentrations of metals to take up the metal as an alternative toxic material to deter predators or pathogens. This hypothesis has excited considerable interest and experimental test. A number of experiments have compared plants containing metal with those without. The basic procedure is either to vary the growth medium to achieve variation in metal content (i.e. grow plants of one accumulating species either on metal contaminated medium or not) or to compare the accumulator grown in the presence of metal with a related non-accumulator. Some studies have investigated whether the plants differ in their toxicity to their predators, actual or potential; others have investigated whether predators’ feeding behaviour is altered when faced with a plant containing high metal levels. Boyd and co-workers (Boyd and Martens, 1994; Martens and Boyd, 1994; Boyd and Moar, 1999; Boyd et al., 2002) have shown that Ni hyperaccumulation makes these plants highly toxic to a number of generalist herbivores. Whether comparing hyperaccumulators grown on normal soil with those grown on nickel, or hyperaccumulators with non-accumulators, the plants with high nickel are always more toxic to the

THE HYPERACCUMULATION OF METALS BY PLANTS

85

predators than plants with more normal nickel levels. Thus, Boyd and Moar (1999) compared the survival and growth of the polyphagus noctuid moth Spodoptera exigua fed on three species of Strepanthus which had been grown on either a high nickel or low nickel soil. One of the species, S. polygaloides, is a nickel hyperaccumulator and the plants in the two treatments differed substantially in mean nickel content: (1480 ppm vs. 20 ppm). Survivorship after 10 days on the low nickel leaves was 52%, but on the high nickel leaves was only 4%. The other two species, S. breweri and S. tortuosus, were collected from serpentine soils, but are not hyperaccumulators, and their leaf nickel content differed much less between the two treatments (S. breweri: 40 vs. 9 ppm; S. tortuosus: 93 vs. 0.5 ppm). There were no differences in survivorship between the two leaf types for either species, but the larvae grown on the high nickel S. tortuosus did pupate slower and were smaller than those grown on control leaves. Boyd and Moar (1999) suggest that these sublethal effects may not just be due to nickel, and speculate that the treatment (i.e. plant growth medium) may have affected the plants in other ways than simply in the amount of nickel in their leaves. While the demonstration that the nickel in a leaf is toxic implies that a plant will achieve greater protection from herbivores, in practice the plant is still being damaged by a herbivore until the herbivore has ingested the lethal dose. If the plant is small, it might be severely damaged before the herbivore has been killed. It would be more effective as a protectant if herbivores were deterred from eating the leaves. In an elegant experiment, Pollard and Baker (1997) tested this aspect of the interaction. They grew two accessions of T. caerulescens in either a low or high zinc medium. The high-zinc plants accumulated 10 as much zinc in their leaves as the ones grown in the low zinc medium. The plants were presented to three herbivores: locusts (Schistocerca gregaria), slugs (Deroceras caruanae) and caterpillars (Pieris brassicae). All three herbivores showed a preference for the low-zinc plants, with the caterpillars rejecting the high-zinc leaves without even apparently tasting them. Boyd et al. (2002) have also shown that snails preferred lownickel Senecio coronatus when offered a choice between leaves containing either high or low nickel. This effect on the behaviour of herbivores clearly increases the effectiveness of the plants’ defence. A number of cases have been published where the metal did not provide protection. Martens and Boyd (2002) conducted a field experiment in which they grew plants of S. polygaloides with different nickel concentrations in the field. Plants were protected by systemic insecticide against insects or by exclosures that would prevent the access of large herbivores. There was no real evidence that the nickel afforded protection against either class of herbivore, and indeed all the plants suffered considerable damage from large

86

M. R. MACNAIR

herbivores. The metal also does not appear to offer protection against specialist feeders which avoid the metal containing leaves. Thus phloem feeding aphids do not appear to be affected by the metal in either Zn-accumulating A. halleri (Ernst et al., 1990; Macnair unpublished) or Ni-accumulating S. polygaloides (Boyd, 1998). We also find that common greenhouse pests such as mealybugs or thrips are equally damaging to A. halleri whatever the metal status of the plants is. Boyd et al. (1999) also found that hyperaccumulation did not protect S. polygaloides from attack by the parasitic plant Cuscuta californica. This plant attaches itself both to xylem and phloem, though presumably obtains most of its nutrients via the latter route. From the xylem, however, it did obtain elevated concentrations of nickel, even if they were not as high as in the parasitised S. polygaloides (Boyd et al., 1999). Turning to pathogens, elevated nickel was found to protect two hyperaccumulating species of Alyssum from attack by two species of Pythium, the cause of damping off (Ghaderian et al., 2000). Plants were grown in a range of metal concentrations, with or without the fungi, and survivorship after 12 days increased from 0% in media lacking nickel to around 50% of control values (A. serpyllifolium) or 75% (A. murale) in media containing nearly 1.7 mM nickel. In contrast, A. saxatile (a nonaccumulator) did not see any increase in survivorship. Note, however, that these very high medium nickel concentrations are unlikely to be met in the field, and will almost certainly be toxic to the non-accumulator. In S. polygaloides, Boyd et al. (1994) showed that high nickel levels protected this species from a number of pathogens. However, Davis et al. (2001) investigated the protection of nickel against Turnip Mosaic Virus in the same species. They found, paradoxically, that high leaf nickel levels increased the susceptibility of these plants to this pathogen. However, the basic protocol of the above studies can be criticised. The problem is that in all the examples given above, the difference in metal content is confounded either with species or with growth medium and thus the difference in predation or infection cannot be unequivocally be ascribed to the metal. Jhee et al. (1999) overcame this objection by using two populations of T. caerulescens that differed in their accumulating phenotype. They still found that the plants with less zinc were attacked more than those with higher zinc concentrations. We have been using the variation in accumulation in the F2 of the cross between A. halleri and A. petraea to test the predation hypothesis. In order to increase the variation, F2 plants were grown in one of 4 different growth media with increasing zinc content. Plants were scored for zinc content, and plants with a wide range of zinc contents were offered to snails as an example of a generalist predator.

THE HYPERACCUMULATION OF METALS BY PLANTS

87

Overall, there was a decrease in predation with increasing zinc content (Fig. 9), but in practice ANOVA showed that all significant effects are caused by the growth medium; within a growth medium there is no effect of zinc. Thus within a treatment, plants with high zinc were not protected more than those with less. The explanation of this effect is not clear, but it does raise doubts about the interpretation of the experiments where plants have been compared which have been grown in different media. Note that similar growth medium effects were postulated by Boyd and Moar (1999) in their work comparing non-accumulating Strepanthus species (see above). There are other reasons to doubt the generality of the predation hypothesis. In the case of zinc hyperaccumulation, it is not clear that the character evolved in a zinc-contaminated environment. For instance, A. halleri is found on both normal and polluted soils, and it would be most parsimonious to suggest that it evolved on normal soils, and that the

Fig. 9. The relationship between the amount of a plant eaten by a snail, and the zinc content of the plant. Individuals from an F2 cross between Arabidopsis halleri and A. petraea, together with A. halleri plants, were grown in one of four Zn concentrations. Plant zinc level was determined, and 6 F2s and 2 A. halleri plants were chosen with a range of internal Zn concentrations and grown with a single snail for 1 week. Plant fresh weight was determined before and afterwards; the change in mass is plotted. Positive values represent net growth over the week. Solid circles: A. halleri plants; open squares: F2 plants grown at 10 mM Zn; open triangles: F2 plants grown at 50 mM Zn; open diamonds: F2 plants grown at 100 mM Zn; open circles: F2 plants grown at 250 mM Zn (Huitson and Macnair, 2003).

88

M. R. MACNAIR

evolution of hyperaccumulation pre-adapted the species to invade polluted soils. It is not at all clear from the experiments discussed above that small differences in zinc content led to the sort of protection from predation that would lead to evolution of this character under this selective agent. In particular, predation by small generalist predators such as snails is most damaging to a plant when it is a small seedling; in practice once a plant is of some size, predation has to be substantial before the fitness of the plant is much reduced (and conversely, the selective advantage of being more protected is thereby less). But small seedlings do not have much zinc in them, particularly when grown in a relatively uncontaminated environment. It takes time for a seedling to build up to the hyperaccumulating levels that have been shown to be protective. Note that this argument does not hold in the case of nickel accumulation. All known nickel accumulators evolved on serpentine soils that are generally rich in nickel. A mutation giving accumulation of nickel would, in these species, have immediately afforded the high nickel concentrations that have been shown to protect adult plants.

C. INADVERTANT UPTAKE

This hypothesis suggests that hyperaccumulation is not an adaptation for an increased level of a specific metal per se. It is possible that the plant is seeking some other nutrient, and has evolved an extremely efficient mechanism for acquiring it; metals such as zinc or nickel are also acquired inadvertantly. This is an essentially non-adaptive hypothesis, and should be a ‘null hypothesis’ against which the adaptive hypotheses are tested. In support of the hypothesis we can note that many hyperaccumulators of one metal have also been shown to be able to accumulate other metals if tested under standard conditions: it is clear that the mechanism of hyperaccumulation is relatively non-specific in many species, and thus it would be dangerous to postulate an adaptive value that depended on the specific metal normally found in the plant in the field. It is also likely that in the model hyperaccumulator, T. caerulescens, accumulation is a character that evolved at the base of a large extant clade that includes many species in the Thlaspiceras, Noccaea and Raparia sections of the genus (Mummenhof et al., 1997). All species tested so far in this clade (S.I. Taylor, unpublished) can accumulate zinc under standard conditions, but do not show this character in the field. This observation is consistent with metal accumulation having evolved under low external metal conditions, where it does not lead to

THE HYPERACCUMULATION OF METALS BY PLANTS

89

elevated leaf concentrations, and hyperaccumulation was only subsequently manifested in species that colonised high metal substrates. What it is that hyperaccumulators may be seeking if not toxic metals is not yet clear. One possibility is iron. Iron is required by plants for healthy growth, and is frequently a limiting factor. Iron-deficient dicotyledons show a number of adaptations including the induction of a membrane-bound reductase, the release of reductants/chelators and the stimulation of a proton pump (Marschner, 1995). Iron is then taken up as FeII. Cohen et al. (1998) found that Fe deficiency in peas stimulated cadmium uptake by the plants. They suggested that this was due to the upregulation of the Fe-transporter IRT1, which is also able to transport other divalent cations such as Cd2þ and Zn2þ, and is related to the ZIP family of zinctransporters. In graminaceous species, iron deficiency leads to the stimulation of the release of phytosiderophores which form very stable complexes with FeIII which are then taken up by a specific transport system (Marschner, 1995). The release of siderophores can also increase the availability and uptake of zinc (Zhang et al., 1991). However, it should be noted that there is no strong evidence that zinc or nickel hyperaccumulators accumulate more iron than related non-accumulator species, nor that iron competes with zinc, nickel or cadmium uptake by hyperaccumulators.

D. DROUGHT TOLERANCE

All nickel hyperaccumulators occur on serpentine, and many are endemic to this substrate. Serpentine soils are known to support an unusual flora, and there has been much discussion of the edaphic factors responsible for this so-called ‘serpentine syndrome’. However, one factor that has frequently been postulated as a major contributor is drought. Serpentine weathers very quickly, and produces soils with poor water retaining properties. Many serpentine species show xeromorphic characteristics. Baker and Walker (1990) noted that the nickel hyperaccumulator Hybanthus floribundus was the only species in its arid habitat that did not show any specific xeromorphic adaptation, but it did have very high epidermal Ni concentrations. They suggested that these high nickel concentrations might reduce cuticular transpiration, though the mechanism by which this might occur is unclear. They more generally speculated that hyperaccumulated metals might act as an osmoticum under conditions of low water availability. Many plants, when faced with a decreasing water potential, synthesise various organic solutes that decrease the osmotic potential so that a plant can maintain turgor. These organic compounds include organic acids, sugars,

90

M. R. MACNAIR

sugar-alcohols amino acids and quaternary ammonium compounds. It is tempting to suggest that acquiring ions from the soil (particularly ones that are in relatively high concentration, such as nickel in serpentine soils) would be energetically less demanding than synthesising organic compounds. However, it is not obvious that the concentrations of metals are sufficient to be very osmotically active. Thus, 1000 ppm Ni (the hyperaccumulation threshold) represents about 2 mM in fresh leaves. This is small compared with the total concentration of ions in the cell. Typical values for proline, for instance, a very common osmoticum (Raymond and Smirnoff, 2002), are much greater. For instance, Voetberg and Sharp (1991) report that proline concentrations reached 100 mM in maize root tips at low water potential.

E. ALLELOPATHY

Boyd and Martens (1992) speculated that metal released from leaf litter into the soil around hyperaccumulators would inhibit the growth of metalsensitive species. This would reduce competition for the hyperaccumulators. However, it will not inhibit the growth of metal-tolerant species. In any metalliferous habitat, including serpentine, any species able to grow must be metal-tolerant. It is not clear that the increase in metal around hyperaccumulators will aid in reducing competition with other species, unless hyperaccumulators are more tolerant than other species, and thus able to survive in higher concentrations of metals. As we have seen, hyperaccumulation as a tolerance mechanism has been postulated, particularly for nickel, but it is difficult to see how this second-order effect could have been the process that led to the evolution of this character (i.e. that for which it is adaptive).

VII. THE ECOLOGICAL CONSEQUENCES OF HYPERACCUMULATION In considering the ecological effects of hyperaccumulation, we must only consider effects which arise solely because of hyperaccumulation. Thus many nickel hyperaccumulators are an important part of the flora of serpentine soils. However, they are normally only a minor component (with important exceptions such as the dominance of some Alyssum species in some Mediterranean sites), and many of their principal ecological effects will be indistinguishable from those of co-occurring, non-accumulator species.

THE HYPERACCUMULATION OF METALS BY PLANTS

91

A. EFFECTS ON OTHER PLANTS

If hyperaccumulators, by virtue of their more efficient metal uptake systems, were able to reduce the effective metal concentration in the soil solution, then it is possible that they could enhance the local environment for other, less tolerant, species. Thus, Whiting et al. (2001b) found that the uptake of zinc by T. caerulescens could ameliorate the toxicity of the soil to the nontolerant T. arvense. However, in so far as hyperaccumulators generally are found on metalliferous soils, and co-occurring species will have to be tolerant of these soils in order to have an intimate association with the hyperaccumulator, then this effect is likely to be relatively unimportant. Negative effects on co-occurring plants could occur if the hyperaccumulator mobilises and releases metal which is then able to have a toxic effect on its neighbours. These possible allelopathic effects have already been discussed (see above). This effect will only be very significant if the hyperaccumulator is able to access a pool of immobile metal that would otherwise be of little ecological effect. But as we have seen, there is little evidence that hyperaccumulators are able to actively mobilise metals from the soil. It is possible that the metal released by the litter from hyperaccumulator species will be in a more biologically active form than otherwise. This could have a particularly detrimental effect on soil microorganisms, and perhaps inhibit nutrient cycles mediated by these organisms. This could certainly affect co-occurring species (but of course it would also affect the hyperaccumulators). On many highly contaminated sites (e.g. old mines) the inhibitory effects of metals on the decomposer community has led to the build up of a significant peat layer below the metal-tolerant plants. Schlegel et al. (1992) found that the microbial community sampled from below nickel hyperaccumulator species was more nickel tolerant than the community sampled from unvegetated serpentine sites. This certainly supports the hypothesis that the hyperaccumulators are increasing the concentration of available metal in the litter layer. B. EFFECTS ON HERBIVORES AND HIGHER TROPHIC LEVELS

As discussed above, the high metal contents of the foliage of hyperaccumulators deters predators, and is toxic to most of those which have been tested. This could potentially have an effect on the population dynamics of generalist herbivores if hyperaccumulators are common in an environment. However, many metalliferous sites have a rather low-productivity flora and a soil that is potentially directly toxic to many herbivores, and these effects are likely to be much more important in determining herbivore population

92

M. R. MACNAIR

sizes than the toxic effects of the hyperaccumulator leaves. A potentially more interesting interaction could arise if hyperaccumulators are toxic to their pollinators, if they had elevated levels of metal in pollen or nectar. One could predict therefore, that flowers will contain less metal than the leaves; this prediction is borne out in A. halleri (Ku¨pper et al., 2000). By increasing the amount of bioavailable metal in the above-ground ecosystem, hyperaccumulators could have a detrimental effect on the functioning of the ecosystem. It is arguable, however, as to whether it is likely that the addition of a minority of hyperaccumulator species will increase the metal sufficiently to have noticeable effects. The point is that in a metalliferous environment there will already be a lot of metal moving through the ecosystem. For instance, the granitic intrusion of Dartmoor contains naturally elevated levels of arsenic. It is possible to detect the arsenic at all levels of the food chain, such that birds of prey in the South West of Britain have higher levels of arsenic in their tissues than the same species in South West Scotland (Erry et al., 1999, 2000). Would the presence of arsenic hyperaccumulators in the flora of the area make any significant overall difference? Most animal species have processes to excrete metals from their bodies; a small increase in the mean metal content of the plant trophic level will almost certainly simply lead to an overall increase in the rate of elimination of metals from animals. Note that these arguments might not apply if an ecosystem were engineered that had a very substantial proportion of hyperaccumulators over a large area. Such an event could happen if hyperaccumulators were deliberately planted on serpentine soil in order to mine the nickel (see below). Then the overall amount of bioavailable metal would be substantially increased, and would almost certainly have a local effect on the fauna. These effects would clearly need to be researched as part of a risk assessment of the technology.

C. CO-EVOLUTION

Notwithstanding the probable lack of an overall effect on an ecosystem from the inclusion of a proportion of hyperaccumulators, the presence of such species will afford an ecological opportunity, and could drive the co-evolution of species that can interact with the hyperaccumulators. The evolution of a specialist microbial community under hyperaccumulators has already been noted (see above). One can conceive of at least two potential co-evolutionary interactions. First, herbivores could evolve tolerance to the metal and so circumvent any defensive role for the metal. This has occurred

THE HYPERACCUMULATION OF METALS BY PLANTS

93

with all other plant defence compounds, so we can predict that this would occur with metal hyperaccumulators also. Second, and potentially more interestingly, the herbivore could take up the metal from the plant and use it as a defence in its turn. For instance, some insects are known to sequester toxic plant alkaloids which in their turn makes the insects highly toxic or distasteful to predators (e.g., Brower and Moffitt, 1974; Rothschild et al., 1975). Research with vertebrate predators has shown that these alkaloids have emetic properties, and birds that have eaten a single alkaloidcontaining specimen will reject further individuals of the same species (Brower, 1969). Thus the storage of these compounds protects the insects from attack. Could the same happen with metals? Boyd and Wall (2001) report a herbivorous hemipteran, Melanotrichus boydi, which was found on all populations of the nickel hyperaccumulator S. polygaloides and which had elevated internal levels of nickel (590–1020 mg Ni g1). This nickel was presumably acquired from its diet, and we can assume that this species has evolved tolerance to this metal. However, Boyd and Wall (2001) suggest that this nickel might also be protective to the insect. They tested four different invertebrate predators, and found that one of the four, a spider, had lower survivorship when fed exclusively on this species. This is suggestive that nickel could possibly be protective, but note that killing a predator after it has eaten you is not a very effective form of defence! The experiments are also rather artificial since the predators were given no choice of prey: in reality these generalist predators will be sampling many different prey species. A defensive role would be more convincing if it could be shown that predators learned to avoid this species after a limited exposure to them.

VIII. PHYTOREMEDIATION The legacy of human industry and intensive agriculture has left large areas of land more or less contaminated with various metals. At one extreme are old mines, where the tailings can frequently contain high (several %) quantities of metals. Many industries (e.g. gas works, tanneries etc.) routinely deposited waste materials on their sites, and the result can be sites highly contaminated with a range of organic and inorganic toxic substances. Less extreme contamination is very widespread, either by the more widespread deposition of wastes from industry or human activities (e.g. lead from car exhausts) or by the effects of intensive agriculture. For instance, many sewage sludges used as agricultural fertilisers contain significant amounts of metals, and many pesticides do or did contain

94

M. R. MACNAIR

metals (e.g. Bordeaux mixture, copper; many early pesticides were based on arsenic). The environmental effects of these metals are profound. The ecosystems are disrupted as the most sensitive species are killed, and many metalcontaminated sites support very depauperate faunas and floras (though note that Cornish mine sites paradoxically provide a haven for a number of species that have been reduced or eliminated in intensively farmed more benign sites, Spalding and Haes, 1995). More importantly from a regulatory perspective, these sites pose a potential hazard to human health. Metals such as cadmium, arsenic and lead are very toxic to humans, and regulators are rightly concerned that as sites change their use, the legacy of previous contamination should not cause adverse effects on health for future users. Thus most developed world countries are actively seeking to decontaminate such sites. Traditional technologies are costly and either involve the removal of the contaminated soil (which simply transfers the contamination from one site to another) or washing the soil in some way that removes the contamination in situ. Such technologies will tend to destroy the soil structure, however, so may only serve to remove the hazard to human health. It is therefore a very attractive notion to use plants to extract the contaminants from the soil, leaving it clean but undamaged. Note that this would also be potentially much cheaper! This possibility is often called phytoextraction. This is only one of a number of potential uses of plants to remediate contaminated land (phytoremediation) but is the only one which is relevant to hyperaccumulators. Other potential phytoremediation strategies have been reviewed and discussed elsewhere (e.g., Chaney et al., 1997; Raskin et al., 1997; Salt et al., 1998; Horne, 2000; Mench et al., 2000). Thus in phytoextraction, plants would be grown on the contaminated sites, the plants would remove the metals from the soils and translocate some of it to the aerial parts, and these can then be harvested, thus permanently removing a portion of the metal in the soil. Three issues arise immediately when considering the applicability of this technology: the concentration of metal in the harvestable material; the biomass of the crop; and the specificity of the plant for the metal. Clearly, the amount of metal removed will be the product of the biomass times the concentration. Hyperaccumulators offer a way of maximising the concentration, but most are of low biomass; it might seem attractive to use fast-growing, high biomass crops that take up lower concentrations, but it seems unlikely that the great difference in concentration in aerial parts could be offset by the difference in biomass (see Table IV). Clearly the best solution will be to use a high-biomass hyperaccumulator, but these seem unusual (though note that

95

THE HYPERACCUMULATION OF METALS BY PLANTS

TABLE IV Amount of metal contained in a crop (kg ha1) compared to the amount of metal contained in the top 20 cms of a soil contaminated with 500 g g1 metal, depending on the biomass of the crop, and the concentration of the metal in its aerial parts

Soil Normal Hyperaccumulator

1

Biomass (t ha1)

Concentration (mg g1)

Total metal (kg ha1)

– 10 20 5 5 10

500 100 200 1000 10000 5000

12001 1 4 5 50 50

Assumes soil has a density of 1.2.

the South African plant, Berheya coddii seems to be an exception: Robinson et al. (1997) found that this plant has both a high biomass production and a high nickel content). The specificity of most hyperaccumulators for particular metals may be more of a problem: since most sites are multiply contaminated, it would be most effective if a potential phytoextractor could extract more than one metal. As we have seen, however, many species will only extract one metal (mostly nickel) and the multiple capabilities of T. caerulescens (nickel, zinc and cadmium) or A. halleri (zinc and cadmium) are unusual, and many of the more toxic metals are probably not accumulated at all (e.g. chromium) or only by a very few species (e.g. lead and arsenic). Clearly modern biotechnology offers the possibility, once the genes for accumulation are identified, of putting together a number of different hyperaccumulator systems in a high biomass crop, and, ultimately, if we could understand and alter the specificity of these systems, of engineering accumulator systems for metals unknown in nature. These prospects are a very long time in the future! The potential to use these hyperaccumulators to clean up land has however been tested in a number of studies. In the first, McGrath and colleagues grew a number of species of hyperaccumulators and normal species, all Brassicaceae, on a series of plots at the Rothamsted Experimental Station, which had been contaminated to different degrees by the long-term application of sewage sludge (McGrath et al., 1993; Baker et al., 1994a). They found that the three populations of zinc hyperaccumulators (2 populations of T. caerulescens, and one of A. halleri) were much more effective at removing zinc from these soils than four species of nickel hyperaccumulator or the three normal species (Table V). These data illustrate

96

M. R. MACNAIR

TABLE V Extraction of zinc by a variety of Brassica species from contaminated soil at the Woburn Market Garden Experiment. The number of croppings required to reduce the soil zinc loading from 444 g g1 (the maximum at the site) to 300 g g1 (the European limit) is also given. (From Baker et al., 1994a.) Species Hyperaccumulators: Thlaspi caerulescens A T. caerulescens B Arabidopsis halleri Non-accumulators: Brassica napus Raphanus sativus

Metal in plant (kg ha1)

Number of croppings

30.1 27.6 10.3

13 14 37

0.5 0.2

832 2046

the points made above about the importance of hyperaccumulation vis a´ vis biomass (Table IV) and of specificity, since only the hyperaccumulators would be able to reduce the zinc loadings of these soils to acceptable levels in less than 50 croppings. Further experiments with T. caerulescens have been performed by Brown et al. (1994) and by Robinson et al. (1998) who have shown that this species is effective at reducing cadmium levels as well as zinc (though about 10 more slowly). The potential of this technology have also been illustrated by work with chelator-assisted phytoextraction (Salt et al., 1998). This technology does not use hyperaccumulators, but induces hyperaccumulation in normal plants by using chelators to mobilise metals. The chelator–metal complex is then taken up by plants and translocated to the shoots, killing the plants, but the plant can still be harvested. Blaylock (2000) describes experiments on two sites heavily contaminated with lead using this technology. The sites were sown with Brassica juncea and the chelator EDTA was applied via irrigation. After 6 weeks the aerial parts of the plants were harvested and removed for disposal. The soil was ploughed, and the process repeated two more times. After the three croppings, the lead burden of both sites was significantly reduced. At one of the sites, before treatment 68% of the site had a lead concentration exceeding 800 ppm. After treatment, all of the site had been reduced to below this level. Whether these technologies will ever have a widespread application must be in doubt, however, particularly in Europe. There are a number of problems. First, effective clean-up will only be possible if a soil is mildly contaminated with a single metal capable of being hyperaccumulated

THE HYPERACCUMULATION OF METALS BY PLANTS

97

(in practice, Zn, Ni or Cd). If the level of contamination is too high, the process will take too long, even with a high biomass hyperaccumulator (see Table V). Second, the author has found considerable scepticism about the technology in discussions with regulators from Britain’s Environment Agency. They are concerned about the overall philosophy of turning a low-level waste problem into a high-level waste problem (i.e. what do you do with cropped material containing high levels of toxic metal?). There are also issues relating to protecting the rest of the environment from the crop whilst the hyperaccumulator is being grown. How will rabbits or other mobile herbivores be prevented from feeding on the crop and moving the metal from the site? What will be the effects on the local insect population? Could these transfer the metal to higher trophic levels? Finally, of course, if a genetically engineered plant were to be developed, the release of GMs into the environment would be an issue. In the case of chelator-enhanced phytoextraction, the danger of increasing the mobility of toxic metals and of contaminating local ground water will surely limit its general applicability (Puschenreiter et al., 2001; Lombi et al., 2001b). In most cases, British regulators appear to prefer the use of the site to be related to the level of contamination, rather than reduce the contamination to enable a different use. The value in phytoremediation lies in it being an environment-friendly technology; the potential economic benefit being the possibility of increasing the value of land at a lower cost than alternative technologies. Another possible use of phytoextraction technology makes use of the intrinsic value of the crop itself. Whilst it will not be possible to ever de-contaminate a highly contaminated site by phytoextraction (Table IV) it would be possible, if a suitable crop could be established on a highly contaminated site, to produce a harvest containing an amount of metal that could be smelted to recover the metal. Table VI illustrates the potential value of such a crop for nickel, zinc and cadmium, assuming reasonable levels of these metals in the crop and biomass production. It is clear that zinc is of far too low value for this technology to be of any use. The metal with the greatest potential is nickel: there are large areas of serpentine soil that have high nickel levels and are of limited use for high-output agriculture and thus produce low incomes. The prospects for this technology have recently been reviewed by Brooks et al. (2001). They include the energy fixed by the crop as a potential income generator. They suggest that if a high biomass plant such as B. coddii were grown on a suitable soil with fertiliser amendments, it should be possible to harvest 22 t ha1 at about 0.5% Ni. This would produce a nickel yield of about 110 kg ha1, worth about $579 in November 2001. If the energy value of the biomass is included (estimated as $288) then the total value of the crop is around $867 ha1. However, how much of this would be retained by

98

M. R. MACNAIR

TABLE VI Value of the metal in a hyperaccumulating crop, assuming a biomass yield of 10 t ha1. Plant concentration is the likely maximum plant concentration achievable with existing plants; yield is the total metal in the crop; metal price is a typical price of refined metal Metal Cadmium Zinc Nickel

Plant conc (mg g1)

Yield (kg ha1)

Metal price ($ kg1)

Crop value ($ ha1)

500 10000 5000

5 100 50

60 0.8 7.3

300 80 365

the grower must be questioned; the processing and handling costs of such material will be high. Brooks et al. (2001) suggest that the grower might get 50% of the nickel and all the energy value, giving a crop value to the farmer of $577 ha1, which exceeds the typical value of a wheat crop at $500 ha1. However, there are many hurdles to be overcome if this value is to be realised, as Brooks et al. (2001) duly note.

IX. CONCLUSIONS Hyperaccumulation is thus a fascinating phenomenon where the questions still far exceed the answers. It is clearly adaptive, but we do not yet know to what, though a defensive function (particularly in nickel hyperaccumulation) is a strong possibility. Research into the mechanisms of accumulation are revealing a variety of transport proteins, the study of which may shed further light on the normal regulation of metal status in plants, but also might help to understand the structural bases for metal transport specificity. The ecological effects of the phenomenon are least understood, and research in this area has lagged behind the molecular and experimental aspects. The holy grail of a commercial exploitation of hyperaccumulation is used as a justification for much research in this area. This is likely to be elusive for the foreseeable future; in the author’s opinion the phenomenon is of sufficient intrinsic interest that its investigation is justified without any immediate commercial applications.

REFERENCES Assunc¸a˜o, A. G. L., Da Costa Martins, P., De Folter, S., Vooijs, R., Schat, H. and Aarts, M. G. M. (2001). Elevated expression of metal transporter genes in three accessions of the metal hyperaccumulator Thlaspi caerulescens. Plant Cell and Environment 24, 217–226.

THE HYPERACCUMULATION OF METALS BY PLANTS

99

Baker, A. J. M. (1981). Accumulators and excluders – strategies in the response of plants to heavy metals. Journal of Plant Nutrition 3, 643–654. Baker, A. J. M. and Walker, P.L. (1990). Ecophysiology of metal uptake by tolerant plants. In ‘Heavy Metal Tolerance in Plants: Evolutionary Aspects’ (A.J. Shaw, ed.), pp. 155–177. CRC Press, Boca Raton. Baker, A. J. M., McGrath, S. P., Sidoli, C. M. D. and Reeves, R. D. (1994a). The possibility of in situ heavy metal decontamination of polluted soils using crops of metal-accumulating plants. Resources, Conservation and Recycling 11, 41–49. Baker, A. J. M., Reeves, R. D. and Hajar, A. S. M. (1994b). Heavy-metal accumulation and tolerance in British populations of the metallophyte Thlaspi caerulescens J. & C. Presl. (Brassicaceae). New Phytologist 127, 61–68. Baker, A. J. M., McGrath, S. P., Reeves, R. D. and Smith, J. A. C. (2000). Metal hyperaccumulator plants: a review of the ecology and physiology of a biological resource for phytoremediation of metal-polluted soils. In ‘Phytoremediation of Contaminated Soil and Water’ (N. Terry and G.S. Ban˜uelos, eds.), pp. 85–108. Lewis Publishers, Boca Raton. Baumann, A. (1885). Das verhalten von zinksalen gegen Pflanzen und im boden. Landwirtsch.Vers.Stn. 31, 1–53. Bert, V., Macnair, M. R., de Languerie, P., Saumitou-Laprade, P. and Petit, D. (2000). Zinc tolerance and accumulation in metallicolous and nonmetallicolous populations of Arabidopsis halleri (Brassicaceae). New Phytologist 146, 225–233. Bert, V., Bonnin, I., Saumitou-Laprade, P., de Languerie, P. and Petit, D. (2002a). Do Arabidopsis halleri from nonmetallicolous populations accumulate zinc and cadmium more effectively than those from metallicolous populations? New Phytologist 155, 47–57. Bert, V., Meerts, P., Saumitou-Laprade, P., Salis, P., Gruber, W. and Verbruggen, N. (2002b). Genetic basis of Cd tolerance and hyperaccumulation in Arabidopsis halleri. Plant and Soil 249, 9–18. Blaylock, M. J. (2000). Field demonstrations of phytoremediation of leadcontaminated soils. In ‘Phytoremediation of Contaminated Soil and Water’ (N. Terry and G.S. Ban˜uelos, eds.), pp. 1–12. Lewis Publishers, Boca Raton. Borhidi, A. (2001). Phylogenetic trends in Ni-accumulating plants. South African Journal of Science 97, 544–547. Boyd, R. S. (1998). Hyperaccumulation as a plant defensive strategy. In ‘Plants that hyperaccumulate heavy metals’ (R.R. Brooks, ed.), pp. 181–202. CAB International, Wallingford. Boyd, R. S. and Martens, S. N. (1992). The raison d’eˆtre for metal hyperaccumulation by plants. In ‘The Vegetation of Ultramafic (Serpentine) Soils’ (A.J.M. Baker, J. Proctor and R.D. Reeves, eds.), pp. 279–289. Intercept, Andover. Boyd, R. S. and Martens, S. N. (1994). Nickel hyperaccumulated by Thlaspi montanum var montanum is acutely toxic to an insect herbivore. Oikos 70, 21–25. Boyd, R. S. and Martens, S. N. (1998). Nickel hyperaccumulation by Thlaspi montanum var. montanum (Brassicaceae): a constitutive trait. American Journal of Botany 85, 259–265. Boyd, R. S. and Moar, W. J. (1999). The defensive function of Ni in plants: response of the polyphagous herbivore Spodoptera exigua (Lepidoptera: Noctuidae) to hyperaccumulator and accumulator species of Streptanthus. Oecologia 118, 218–224.

100

M. R. MACNAIR

Boyd, R. S. and Wall, M. A. (2001). Responses of generalist predators fed high-Ni Melanotrichus boydi (Heteroptera: Miridae): elemental defense against the third trophic level. The American Midland Naturalist 146, 186–198. Boyd, R. S., Shaw, J. J. and Martens, S. N. (1994). Nickel hyperaccumulation defends Streptanthus polygaloides (Brassicaceae) against pathogens. American Journal of Botany 81, 294–300. Boyd, R. S., Martens, S. N. and Davis, M. A. (1999). The nickel hyperaccumulator Streptanthus polygaloides (Brassicaceae) is attacked by the parasitic plant Cuscuta californica (Cuscutaceae). Madron˜o 46, 92–99. Boyd, R. S., Wall, M. A. and Watkins, J. E. (2000). Correspondence between Ni tolerance and hyperaccumulation in Streptanthus (Brassicaceae). Madron˜o 47, 97–105. Boyd, R. S., Davis, M. A., Wall, M. A. and Balkwill, K. (2002). Nickel defends the South African hyperaccumulator Senecio coronatus (Asteraceae) against Helix aspersa (Mollusca: Pulmonidae). Chemoecology 12, 91–97. Broadley, M. R., Willey, N. J., Wilkins, J. C., Baker, A.J.M., Mead, A. and White, P.J. (2001). Phylogenetic variation in heavy metal accumulation in angiosperms. New Phytologist 152, 9–27. Brooks, R. R. (1998). Geobotany and hyperaccumulators. In ‘Plants that Hyperaccumulate Heavy Metals’ (R.R. Brooks, ed.), pp. 55–94. CAB International, Wallingford. Brooks, R. R., Lee, J., Reeves, R. D. and Jaffre´, T. (1977). Detection of metalliferous rocks by analysis of herbarium specimens of indicator plants. Journal of Geochemical Exploration 7, 49–77. Brooks, R. R., Morrison, R. S., Reeves, R. D., Dudley, T. R. and Akman, Y. (1979). Hyperaccumulation of nickel by Alyssum Linnaeus (Cruciferae). Proceedings of the Royal Society of London, Series B. 203, 387–403. Brooks, R. R., Robinson, B. H., Howes, A. W. and Chiarucci, A. (2001). An evaluation of Berkheya coddii Roessler and Alyssum bertolonii Desv. for phytoremediation and phytomining of nickel. South African Journal of Science 97, 558–560. Brower, L. P. (1969). Ecological Chemistry. Scientific American 220, 22–29. Brower, L. P. and Moffitt, C. M. (1974). Palatability dynamics of cardenolides in the monarch butterfly. Nature 249, 280–283. Brown, S. L., Chaney, R. L., Angle, J. S. and Baker, A. J. M. (1994). Phytoremediation potential of Thlapsi caerulescens and bladder campion for zinc- and cadmium-contaminated soils. Journal of Environmental Quality 23, 1151–1157. Chaney, R. L., Malik, M., Li, Y. M., Brown, S. L., Brewer, E. P., Angle, J. S., and Baker, A.J.M. (1997). Phytoremediation of soil metals. Current Opinion in Biotechnology 8, 279–284. Cohen, C. K., Fox, T. C., Garvin, D. F. and Kochian, L. V. (1998). The role of irondeficiency stress responses in stimulating heavy-metal transport in plants. Plant Physiology 116, 1063–1072. Davis, M. A., Murphy, J. F. and Boyd, R. S. (2001). Nickel increases susceptibility of a nickel hyperaccumulator to Turnip mosaic virus. Journal of Environmental Quality 30, 85–90. Ernst, W. H. O., Schat, H. and Verkleij, J. A. C. (1990). Evolutionary biology of metal resistance in Silene vulgaris. Evolutionary Trends in Plants 4, 45–50. Ernst, W. H. O., Assunc¸a˜o, A. G. L., Verkleij, J. A. C. and Schat, H. (2002). How important is apoplastic zinc xylem loading in Thlaspi caerulescens? New Phytologist 155, 4–5.

THE HYPERACCUMULATION OF METALS BY PLANTS

101

Erry, B. V., Macnair, M. R., Meharg, A. A., Shore, R. F. and Newton, I. (1999). Arsenic residues in predatory birds from an area of Britain with naturally and anthropogenically elevated arsenic levels. Environmental Pollution 106, 91–95. Erry, B. V., Macnair, M. R., Meharg, A. A. and Shore, R. F. (2000). Arsenic contamination in wood mice (Apodemus sylvaticus) and bank voles Clethrionomys glareolus) on abandoned mine sites in southwest Britain. Environmental Pollution 110, 179–187. Frey, B., Keller, C., Zierold, K. and Schulin, R. (2000). Distribution of Zn in functionally different leaf epidermal cells of the hyperaccumulator Thlaspi caerulescens. Plant Cell and Environment 23, 675–687. Ghaderian, Y. S. M., Lyon, A. J. E. and Baker, A. J. M. (2000). Seedling mortality of metal hyperaccumulator plants resulting from damping off by Pythium spp. New Phytologist 146, 219–224. Grotz, N., Fox, T., Connolly, E., Park, W., Guerinot, M. L. and Eide, D. (1998). Identification of a family of zinc transporter genes from Arabidopsis that respond to zinc deficiency. Proceedings of the National Academy of Sciences of the United States of America 95, 7220–7224. Guerinot, M. L. (2000). The ZIP family of metal transporters. Biochimica et Biophysica Acta 1465, 190–198. Haines, B. J. (2002). Zincophilic root foraging in Thlaspis caerulescens. New Phytologist 155, 363–372. Hall, J. L. (2002). Cellular mechanisms for heavy metal detoxification and tolerance. Journal of Experimental Botany 53, 1–11. Horne, A. J. (2000). Phytoremediation by constructed wetlands. In ‘Phytoremediation of Contaminated Soil and Water’ (N. Terry and G.S. Ban˜uelos, eds.), pp. 13–40. Lewis Publishers, Boca raton. Huitson, S. B. and Macnair, M. R. (2003) Does zinc protect the zinc hyperaccumulator Arabidopsis halleri from herbivory by snails? New Phytologist 159, 453–460. Ingrouille, M. J. and Smirnoff, N. (1986). Thlaspi caerulescens J. & C. Presl. (T. alpestre L.) in Britain. New Phytologist 102, 219–233. Jaffre´, T., Brooks, R. R., Lee, J. and Reeves, R. D. (1976). Sebertia acuminata: a hyperaccumulator of nickel from New Caledonia. Science 193, 579–580. Jhee, E. M., Dandridge, K. L., Christy, A. M. and Pollard, A. J. (1999). Selective herbivory on low-zinc phenotypes of the hyperaccumulator Thlaspi caerulescens (Brassicaceae). Chemoecology 9, 93–95. Knight, B., Zhao, F. J., McGrath, S. P. and Shen, Z. G. (1997). Zinc and cadmium uptake by the hyperaccumulator Thlaspi caerulescens in contaminated soils and its effects on the concentration and chemical speciation of metals in soil solution. Plant and Soil 197, 71–78. Koch, M., Bishop, J. and Mitchell-Olds, T. (1999). Molecular systematics and evolution of Arabidopsis and Arabis. Plant Biology 1, 529–537. Kra¨mer, U., Cotter-Howells, J. D., Charnock, J. M., Baker, A. J. M. and Smith, J.A.C. (1996). Free histidine as a metal chelator in plants that accumulate nickel. Nature 379, 635–638. Kra¨mer, U., Smith, R. D., Wenzel, W. W., Raskin, I. and Salt, D. E. (1997). The role of metal transport and tolerance in nickel hyperaccumulation by Thlaspi goesingense Ha´la´csy. Plant Physiology 115, 1641–1650. Ku¨pper, H., Zhao, F. J., and McGrath, S. P. (1999). Cellular compartmentation of zinc in leaves of the hyperaccumulater Thlaspi caerulescens. Plant Physiology 119, 305–311.

102

M. R. MACNAIR

Ku¨pper, H., Lombi, E., Zhao, F. J. and McGrath, S. P. (2000). Cellular compartmentation of cadmium and zinc in relation to other elements in the hyperaccumulator Arabisopsis halleri. Planta 212, 75–84. Ku¨pper, H., Lombi, E., Zhao, F. J., Weishammer, G. and McGrath, S. P. (2001). Cellular compartmentation of nickel in the hyperaccumulators Alyssum lesbiacum, Alyssum bertolonii and Thlaspi goesingense. Journal of Experimental Botany 52, 2291–2300. Lasat, M. M., Baker, A. J. M. and Kochian, L. V. (1996). Physiological characterization of root Zn2þ absorption and translocation to shoots in Zn hyperaccumulator and nonaccumulator species of Thlaspi. Plant Physiology 112, 1715–1722. Lasat, M. M., Baker, A. J. M. and Kochian, L. V. (1998). Altered Zn compartmentation in the root symplasm and stimulated Zn absorption into the leaf as mechanisms involved in Zn hyperaccumulation in Thlaspi caerulescens. Plant Physiology 118, 875–883. Lloyd-Thomas, D. (1995). Heavy metal hyperaccumulation by Thlaspi caerulescens J&C Presl. Dissertation, University of Sheffield. Lombi, E., Zhao, F. J., Dunham, S. J. and McGrath, S. P. (2000). Cadmium accumulation in populations of Thlaspi caerulescens and Thlapsi goesingense. New Phytologist 145, 11–20. Lombi, E., Zhao, F. J., McGrath, S. P., Young, S. and Sacchi, A. (2001a). Physiological evidence for a high affinity cadmium transporter highly expressed in a Thlaspi caerulescens ecotype. New Phytologist 149, 53–60. Lombi, E., Zhao, S. -L., Dunham, S. J. and McGrath, S. P. (2001b). Phytoextraction of heavy metal-contaminated sites: natural hyperaccumulation versus chemically enhanced phytoextraction. Journal of Environmental Quality 30, 1919–1926. Long, X. X., Yang, X. E., Ye, Z. Q., Ni, W. Z. and Shi, W. Y. (2002). Differences of uptake and accumulation of zinc in four species of Sedum. Acta Botanica Sinica 44, 152–157. Macnair, M. R. (1981). The uptake of copper by plants of Mimulus guttatus differing in genotype primarily at a single major copper tolerance locus. New Phytologist 88, 723–730. Macnair, M. R. (2002). Within and between population genetic variance for zinc accumulation in Arabidopsis halleri. New Phytologist 155, 59–66. Macnair, M.R. and Smirnoff, N. (1999). Use of zincon to study uptake and accumulation of zinc by zinc tolerant and hyperaccumulating plants. Communications in Soil Science and Plant Analysis 30, 1127–1136. Macnair, M. R., Bert, V., Huitson, S. B., Saumitou-Laprade, P. and Petit, D. (1999). Zinc tolerance and hyperaccumulation are genetically independent characters. Proceedings of the Royal Society of London, Series B. 266, 2175–2179. Macnair, M. R., Tilstone, G. H. and Smith, S. E. (2000). The genetics of tolerance and accumulation in higher plants. In ‘Phytoremediation of Contaminated Soil and Water’ (N. Terry and G.S. Ban˜uelos, eds.), pp. 235–250. Lewis Publishers. Boca Raton. Ma¨ser, P., Thomine, S., Schroeder, J. I., Ward, J. M., Hirschi, K., Sze, H., Talke, I. N., Amtmann, A., Maathuis, F. J. M., Sanders, D., Harper, J. F., Tchieu, J., Gribskov, M., Persans, M. W., Salt, D. E., Kim, S. A. and Guerinot, M. L. (2001). Phylogenetic relationships within cation transporter families of Arabidopsis. Plant Physiology 126, 1646–1667.

THE HYPERACCUMULATION OF METALS BY PLANTS

103

Malaisse, F., Gre´goire, J., Brooks, R. R., Morrison, R. S. and Reeves, R. D. (1978). Aeollanthus biformifolius De Wild: A hyperaccumulator of copper from Zaı¨ re. Science 199, 887–888. Marschner, H. (1995). ‘‘Mineral Nutrition of Higher Plants’’. Academic Press, London. Martens, S. N. and Boyd, R. S. (1994). The ecological significance of nickel hyperaccumulation – a plant-chemical defense. Oecologia 98, 379–384. Martens, S. N. and Boyd, R. S. (2002). The defensive role of Ni hyperaccumulation by plants: a field experiment. American Journal of Botany 89, 998–1003. McGrath, S. P., Sidoli, C. M. D., Baker, A. J. M. and Reeves, R. D. (1993). The potenial for the use of metal-accumulating plants for the in situ decontamination of metal-polluted soils. In ‘Integrated Soil and Sediment Research: a Basis for Proper Protection’ (H.J.P. Eijsackers and T. Hamers, eds.), pp. 673–676. Kluwer Academic Publishers. Meerts, P., and Van Isacker, N. (1997). Heavy metal Tolerance and accumulation in metallicolous and non-metallicolous populations of Thlaspi caerulescens from continental Europe. Plant Ecology 133, 221–231. Mench, M., Vangronsveld, J., Clijsters, H., Lepp, N. W. and Edwards, R. (2000). In situ metal mobilization and phytostabilization of contaminated soils. In ‘Phytoremediation of Contaminated Soil and Water’ (N. Terry and G.S. Ban˜uelos, eds.), pp. 323–358. Lewis publishers, Baco Raton. Meyer, F. K. (1973). Conspectus der ‘‘Thlaspi’’ Arten Europas, Africas und Vorderasiens. Feddes Repertorium 84, 449–470. Morrison, R. S., Brooks, R. R., Reeves, R. D. and Malaisse, F. (1979). Copper and cobalt uptake by metallophytes from Zaı¨ re. Plant and Soil 53, 535–539. Mummenhoff, K., Franzke, A. and Koch, M. (1997). Molecular data reveal convergence in fruit characters used in the classification of Thlaspi s. l. (Brassicaceae). Botanical Journal of the Linnean Society 125, 183–199. Paton, A. and Brooks, R. R. (1996). A re-evaluation of Haumaniastrun species as a geobotanical indicator of copper and cobalt. Journal of Geochemical Exploration 56, 37–45. Pence, N. S., Larsen, P. B., Ebbs, S. D., Letham, D. L. D., Lasat, M. M., Garvin, D. F., Eide, D. and Kochian, L. V. (2000). The molecular physiology of heavy metal transport in the Zn/Cd hyperaccumulator Thlaspi caerulescens. Proceedings of the National Academy of Sciences of the United States of America 97, 4956–4960. Persans, M. W., Nieman, K. and Salt, D. E. (2001). Functional activity and role of cation-efflux family members in Ni hyperaccumulation in Thlaspi goesingense. Proceedings of the National Academy of Sciences of the United States of America 98, 9995–10000. Pollard, A. J. and Baker, A. J. M. (1996). The quantitative genetics of zinc hyperaccumulation in Thlaspi caerulescens. New Phytologist 132, 113–118. Pollard, A. J. and Baker, A. J. M. (1997). Deterrence of herbivory by zinc hyperaccumulation in Thlaspi caerulescens (Brassicaceae). New Phytologist 135, 655–658. Pollard, A. J., Dandridge, K. L. and Jhee, E. M. (2000). Ecological genetics and the evolution of trace elements in plants. In ‘Phytoremediation of contaminated soil and water’ (N. Terry and G.S. Ban˜uelos, eds.), pp. 251–264. Lewis Publishers. Boca Raton. Psaras, G. K., Constantinidis, T., Cotsopoulos, B. and Manetas, Y. (2000). Relative abundance of nickel in the leaf epidermis of eight hyperaccumulators:

104

M. R. MACNAIR

evidence that the metal is excluded from both guard cells and trichomes. Annals of Botany 86, 73–78. Puschenreiter, M., Stoger, G., Lombi, E., Horak, O. and Wenzel, W. W. (2001). Phytoextratcion of heavy metal contaminated soils with Thlaspi goesingense and Amaranthus hybridus: Rhizosphere mainipulation using EDTA and ammonium sulfate. Journal of Plant Nutrition and Soil Science-Zeitschrift fur pflanzenernahrung und bodenkunde 164, 615–621. Raskin, I., Smith, R. D. and Salt, D. E. (1997). Phytoremediation of metals: using plants to remove pollutants from the environment. Current Opinion in Biotechnology 8, 221–226. Raymond, M. and Smirnoff, N. (2002). Proline metabolism and transport in maize seedlings at low water potential. Annals of Botany 89, 813–823. Reeves, R. D. (1992). The Hyperaccumulation of Nickel by Serpentine Plants. In ‘The Vegetation of Ultramaphic (Serpentine) Soils’ (A.J.M. Baker, J. Proctor and R.D. Reeves, eds.), pp. 253–277. Intercept, Andover. Reeves, R. D., Baker, A. J. M., Borhidi, A. and Berazan, R. (1999). Nickel hyperaccumulation in the serpentine flora of Cuba. Annals of Botany 83, 29–38. Robinson, B. H., Brooks, R. R., Howes, A. W., Kirkman, J. H. and Gregg, P. E. H. (1997). The potential of the high biomass nickel hyperaccumulator Berkheya coddii for phytoremediation and phytomining. Journal of Geochemical Exploration 60, 115–126. Robinson, B. H., Leblanc, M., Petit, D., Brooks, R. R., Kirkman, J. H. and Gregg, P. E. H. (1998) The potential of Thlaspi caerulescens for phytoremediation of contaminated soils. Plant and Soil 203, 47–56. Rothschild, M., von Euw, J., Reichstein, T., Smith, D. A. S. and Pierre, J. (1975). Cardenolide storage in Danaus chrysippus (L.) with additional notes on D. plexipus (L.). Proceedings of the Royal Society of London, Series B. 190, 1–31. Salt, D. E., Smith, R. D. and Raskin, I. (1998). Phytoremediation. Annual Review of Plant Physiology and Plant Molecular Biology 49, 643–668. Salt, D. E., Kato, N., Kra¨mer, U., Smith, R. D. and Raskin, I. (2000). The role of root exudates in nickel hyperaccumulation and tolerance in accumulator and nonaccumulator species of Thlaspi. In ‘Phytoremediation of contaminated soil and water’ (N. Terry and G.S. Ban˜uelos, eds.), pp. 189–200. Lewis Publishers, Boca Raton. Schlegel, H. G., Meyer, M., Schmidt, T., Stoppel, R. D. and Pickhardt, M. (1992). A community of nickel-resistant bacteria under nickel-hyperaccumulating plants. In ‘The Vegetation of Ultramaphic (Serpentine) Soils’ (A.J.M. Baker, J. Proctor and R.D. Reeves, eds.), pp. 305–317. Intercept, Andover. Schwartz, C., Morel, J. L., Saumier, S., Whiting, S. N. and Baker, A. J. M. (1999). Root development of the zinc-hyperaccumulator plant Thlaspi caerulescens as affected by metal origin, content and localization in soil. Plant and Soil 208, 103–115. Spalding, A. and Haes, E. C. M. (1995). Contaminated land – a resource for wildlife: a review and survey of insects on metalliferous mine sites in Cornwall. Land Contamination and Reclamation 3, 24–29. van der Zaal, B. J., Neuteboom, L. W., Pinas, J. E., Chardonnens, A. N., Schat, H., Verkleij, J. A. C. and Hooykaas, P. J. J. (1999). Overexpression of a novel Arabidopsis gene related to putative zinc-transporter genes from animals can lead to enhanced zinc resistance and accumulation. Plant Physiology 119, 1047–1055. Va´zquez, M. D., Poschenrieder, C., Barcelo´, J., Baker, A. J. M., Hatton, P. and Cope, G. H. (1994). Compartmentation of zinc in roots and leaves of the

THE HYPERACCUMULATION OF METALS BY PLANTS

105

zinc hyperaccumulator Thlaspi caerulescens J&C Presl. Botanica Acta 107, 243–250. Voetberg, G. S. and Sharp, R. E. (1991). Growth of the maize primary root at low water potentials. III. Role of increased proline deposition in osmotic adjustment. Plant Physiology 96, 1125–1130. White, P. J., Whiting, S. N., Baker, A. J. M. and Broadley, M. R. (2002). Does zinc move apoplastically to the xylem in roots of Thlaspi caerulescens? New Phytologist 153, 201–207. Whiting, S. N., Leake, J. R., McGrath, S. P. and Baker, A. J. M. (2000). Positive responses to Zn and Cd by roots of the Zn and Cd hyperaccumulator Thlaspi caerulescens. New Phytologist 145, 199–210. Whiting, S. N., Leake, J. R., McGrath, S. P. and Baker, A. J. M. (2001a). Assessment of Zn mobilization in the rhizosphere of Thlaspi caerulescens by bioassay with non-accumulator plants and soil extraction. Plant and Soil 237, 147–156. Whiting, S. N., Leake, J. R., McGrath, S. P. and Baker, A. J. M. (2001b). Hyperaccumulation of Zn by Thlaspi caerulescens can ameliorate Zn toxicity in the rhizosphere of co-cropped Thlaspi arvense. Environmental Science and Technology 35, 3237–3241. Whiting, S. N., Leake, J. R., McGrath, S. P. and Baker, A. J. M. (2001c). Zinc accumulation by Thlaspi caerulescens from soils with different Zn availability: a pot study. Plant and Soil 236, 11–18. Williams, L. E., Pittman, J. K. and Hall, J. L. (2000). Emerging mechanisms for heavy metal transport in plants. Biochimica et Biophysica Acta 1465, 104–126. Zhang, F., Ro¨mheld, V. and Marschner, H. (1991). Release of zinc mobilizing root exudates in different plant species as affected by zinc nutritional status. Journal of Plant Nutrition 14, 675–686. Zhao, F. J., Shen, Z. G. and McGrath, S. P. (1998). Solubility of zinc and interactions between zinc and phosphorus in the hyperaccumulator Thlaspi caerulescens. Plant Cell and Environment 21, 108–114. Zhao, F. J., Lombi, E., Breedon, T. and McGrath, S. P. (2000). Zinc hyperaccumulation and cellular distribution in Arabidopsis halleri. Plant Cell and Environment 23, 507–514. Zhao, F. J., Hamon, R. E. and McLaughlin, M. J. (2001). Root exudates of the hyperaccumulator Thlaspi caerulescens do not enhance metal mobilization. New Phytologist 151, 613–620. Zhao, F. J., Dunham, S. J. and McGrath, S. P. (2002a). Arsenic hyperaccumulation by different fern species. New Phytologist 156, 27–31. Zhao, F. J., Hamon, R. E., Lombi, E., McLaughlin, M. J. and McGrath, S. P. (2002b). Characteristics of cadmium uptake in two contrasting ecotypes of the hyperaccumulator Thlaspi caerulescens. Journal of Experimental Botany 53, 535–543.

This Page Intentionally Left Blank

Plant Chromatin — Learning from Similarities and Differences

JAN BRZESKI1, JERZY DYCZKOWSKI1, SZYMON KACZANOWSKI1, PIOTR ZIELENKIEWICZ1,2 AND ANDRZEJ JERZMANOWSKI1,2 1

Institute of Biochemistry and Biophysics, Polish Academy of Sciences and 2 Laboratory of Plant Molecular Biology, Warsaw University, Pawin´skiego 5A, 02-106 Warsaw, Poland

I. II. III. IV. V.

VI. VII. VIII. IX. X.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chromatin basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chromatin structural dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modulation of chromatin structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ATP-dependent chromatin remodelling mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Arabidopsis homologues of the central ATPase and other core subunits of the prototype SWI/SNF-type complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. CHD/Mi2 subfamily in Arabidopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Core histone modifications by acetylation and methylation . . . . . . . . . . . Plant enzymes responsible for acetylation and methylation of core histones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methylation of DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The problem of linker histones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Advances in Botanical Research Vol. 40 incorporating Advances in Plant Pathology ISBN 0-12-005940-1

108 109 110 110 111 117 121 121 124 126 130 133 134 134

Copyright  2003 Elsevier Ltd All rights of reproduction in any form reserved

108

J. BRZESKI ET AL.

ABSTRACT The availability of the complete sequence of Arabidopsis genomic DNA has allowed in-depth comparative analyses of plant proteins homologous to components of yeast and animal complexes involved in chromatin remodelling. These studies have uncovered an astonishing diversity of plant proteins that are potentially involved in ATP-dependent modulation of nucleosome structure and in post-translational modification of the core histone tails. Of the small fraction of these proteins that have been studied, all were shown to play important roles in plant development. These broad comparative analyses have provided insights into plant-specific features of evolutionarily conserved chromatin processes. An interesting example is the scarcity or absence from many plant DNA-dependent ATPases, histone acetyltransferases and histone methyltransferases of the protein–protein recognition modules (like bromo- and chromodomains), which are highly conserved in yeast and animals. The comparison in plants and animals of the phenotypic effects of disturbances in the native ratio of linker histone variants has shed some light on the possible functions of these abundant and still mysterious chromatin components. Of general importance for deciphering the biochemical modification code underlying the epigenetic organisation of chromatin was a recent discovery of a link between histone H3 lysine 9 methylation and DNA methylation in Arabidopsis. Although not yet as advanced as in yeast and animals, chromatin research in plants has moved rapidly to a stage at which major progress in understanding the chromatin-based regulation of plant development will soon be possible.

I. INTRODUCTION The availability of the entire genomic sequences of Arabidopsis and of several animals (Drosophila melanogaster, Caenorhabditis elegans, Homo sapiens) has permitted the comparison of developmental mechanisms of plants and animals. The results of this analysis indicate that in the two kingdoms the processes fundamental for development, i.e. pattern formation and cell–cell signalling, are controlled by non-homologous genes (Meyerowitz, 2002). Thus, these processes have evolved independently in plants and animals. In contrast, the regulatory mechanisms at the level of chromatin show a high degree of evolutionary conservation. Not only are the major structural proteins, the histones, highly conserved, but also many proteins involved in histone modification and nucleosome remodelling (Verbsky and Richards, 2001). This conservation extends to important control strategies: for example, in both animals and plants the Polycomb group (Pc-G) proteins are involved in the repression of homeotic-type genes during development (Meyerowitz, 2002). However, even a cursory examination of the differences between animal and plant cells suggests that chromatin-based regulatory mechanisms, although evolved from a common ancestral system, may not be exactly the same. For example, the unique ability of somatic plant cells to dedifferentiate, i.e. undergo

PLANT CHROMATIN

109

genetic reprogramming, requires mechanisms ensuring the plasticity of chromatin-level gene repression that may not occur in animals (Weigel and Ju¨rgens, 2002). While studies on chromatin in plants have not yet advanced to a stage comparable to that reached in yeast or animals, the available information about homologous proteins and processes in plants and other kingdoms allows some insight into plant-specific features of chromatin regulation. Several excellent reviews covering various aspects of plant chromatin have recently been published (Meyer, 2001; Fransz and de Jong, 2002; Li et al., 2002; Reyes et al., 2002). Therefore, we will focus on more recent data concerning the ATP-dependent mechanisms of chromatin remodelling and their possible links with histone and DNA modifications as well as on the functions of linker histones.

II. CHROMATIN BASICS Eukaryotic genomes are organised into chromatin, a structure that enables packaging of hundreds of millions of base pairs of DNA into a microscopicsize nucleus, while ensuring efficient and highly ordered use of the information contained in the DNA sequence. The nucleosome, the fundamental building block of chromatin, is composed of 145–147 base pairs of DNA wrapped around a protein octamer made of two molecules each of core histones H2A, H2B, H3 and H4. This nucleosome core particle is associated with a molecule of linker histone H1 that stabilises entry/exit of the DNA helix on the nucleosome surface and protects an additional  20 bp of DNA from nuclease digestion. Each core histone contains a carboxy-terminal helical domain with a highly conserved histone-fold that is critical for octamer assembly and interactions with DNA, and a flexible and basic N-terminal domain which protrudes outside the core particle and has sites for post-translational modifications (such as acetylation, methylation, phosphorylation and ubiquitination). Histone H1 does not have a histonefold. Its central globular domain (GH1) through which it associates with the core particle, belongs to the family of winged-helix type proteins (other typical members of this family are: Hepatocytic Nuclear Factor 3  – HNF3 , Human Regulatory factor X – RFX-1), and is flanked by flexible and highly basic N- and C-terminal tails that can bind and neutralise negatively charged DNA. Nucleosomes, spaced at the 160–210 bp intervals along chromosomal DNA, form the nucleosomal arrays. While the packaging of DNA into nucleosomes has been shown to impede the access of DNA-binding proteins and RNA polymerase in vitro, it is

110

J. BRZESKI ET AL.

probably the higher order structures, i.e. the architecture of the nucleosomal arrays, that regulate much of the biological function of chromatin in the cell (Horn et al., 2002).

III. CHROMATIN STRUCTURAL DYNAMICS Most of the current knowledge concerning chromatin structural transitions comes from in vitro solution studies using defined nucleosomal arrays composed of 12 nucleosomes (for a recent review see Hansen, 2002). An increase in the concentration of mono- or divalent cations causes loosely organised extended nucleosomal arrays to fold intramolecularly, first into moderately and then into maximally compact conformations. The latter state is reached at >60 mM Naþ or >0.3 mM Mg2þ and has the appearance of a rod-like compact fibre with a diameter of about 30 nm. The increase in divalent (but not monovalent) cation concentration can also induce intermolecular oligomerisation of nucleosomal arrays, a phenomenon reminiscent of the fibre–fibre interactions observed in native chromatin. Interestingly, this oligomerisation does not require extensive intramolecular folding and occurs with both maximally or moderately folded as well as fully extended arrays (Hansen, 2002). This suggests that in vitro and probably also in vivo, the chromatin fibre undergoes dynamic structural transitions, adopting many different folded and/or oligomeric structures. In vivo, the highest level of compaction is achieved within chromosome domains. To what extent are chromatin structural transitions determined by features of the nucleosome? The results of in vitro studies indicate that core histone tails are essential for chromatin compaction occurring at elevated salt as well as being determinants of the structural dynamics of the compacted structures. The tails are critical because they mediate nucleosome– nucleosome interactions that drive the formation of higher-order compact structures. The in vitro studies also demonstrated that linker histones, contrary to earlier thinking, are not required for the induction of folding of the chromatin fibre. However, they do stabilise both the folded and the oligomeric chromatin states. Under the same ionic conditions, the structures formed in the presence of linker histones are more regular and homogeneous than those formed in their absence (Hansen, 2002).

IV. MODULATION OF CHROMATIN STRUCTURE All processes that involve DNA as a substrate or template (e.g., repair, recombination, transcription, replication) can potentially be regulated by

PLANT CHROMATIN

111

modulation of the DNA accessibility. It is now generally accepted that this regulation occurs at two levels. Firstly, the primary regulators, either activators or repressors, bind to specific DNA motifs. These proteins are modular in nature and in addition to sequence-specific DNA-binding domains they possess functionally separate effector domains. The biological outcome of their association with DNA depends on the properties and biochemical activities of cofactors that are bound via the effector domains. These cofactors form the second level of regulation. The repressor– corepressor or activator–coactivator complexes initiate the assembly of much larger complexes which integrate diverse regulatory signals. These can be the ATP-dependent chromatin remodelers altering nucleosome stability and/or position on target DNA sequences, or enzymes which add or remove post-translational covalent modifications such as acetylation, methylation, phosphorylation and ubiquitination of the core histone tails, creating a biochemical code that marks chromatin regions. Chromatin DNA can also be modified by the methylation of cytosines. Stable changes in DNA accessibility can also be achieved by heterochromatinisation induced by specialised proteins. All of these mechanisms cooperate to ensure the efficient control of genetic activity.

V. ATP-DEPENDENT CHROMATIN REMODELLING MECHANISMS The structural dynamics of chromatin fibres are influenced by a battery of specialised enzymes that utilise the energy of ATP hydrolysis to alter the chromatin state. These enzymes are classified within the broad group of ‘ATP-dependent chromatin remodelling factors’. The term ‘chromatin remodelling’ has no precise definition and the process is not well understood, either at the nuclear level or in the context of native chromatin fibres. In in vitro biochemical assays, chromatin remodelling is reflected by the ability of protein complexes displaying an enzymatic activity to affect different aspects of nucleosomal structure. These include: induction of histone octamer movement along a DNA molecule, induction of regular nucleosome spacing, altering histone–DNA interactions and even complete disruption of histone–DNA contacts. Combined genetic and biochemical approaches have resulted in purification and characterisation of different types of chromatin remodelling enzymes from yeast, Drosophila and vertebrates. All enzymes characterised so far appear to function within multi-subunit complexes, with the number of subunits ranging from two to more than 10. The composition of subunits

112

J. BRZESKI ET AL.

associated with a given enzyme is generally conserved within the enzyme subfamily. The distinctive feature of all complexes is the occurrence of a central catalytic ATPase, belonging to the SWI2/SNF2 family which falls within a large superfamily of DEXD/H ATPases defined by characteristic motifs within conserved ‘DEXD/H’ and ‘HelicC’ domains. The DEXD/H superfamily is one of the largest groups of ancient proteins present in all living organisms. Many of its members are involved in various aspects of DNA and RNA metabolism. The distinctive feature of the SWI2/SNF2 family is the SNF2_N domain, a variant of the typical DEXD/H domain with a wellconserved C-terminal extension of approximately 100 aminoacids (http:// pfam.wustl.edu/cgi-bin/getdesc?name¼SNF2_N). SWI2/SNF2-like proteins have not been found in prokaryotes and appear to have evolved after the separation of the eukaryotic lineage. A more detailed characterisation of the SWI2/SNF2 family is provided by phylogenetic analysis (Eisen et al., 1995; Verbsky and Richards, 2001; this study). Since sequence homology is limited to the core ATPase domain, the different subfamilies may have evolved in a series of independent gene fusion events. The members of only four subfamilies have been shown to act as catalytic subunits in chromatin remodelling complexes. They are: SWI2/SNF2 (Cote et al., 1994; Cairns et al., 1996), ISWI (Tsukiyama and Wu, 1995; Ito et al., 1997; Varga-Weisz et al., 1997), CHD/Mi2 (Xue et al., 1998; Zhang et al., 1999) and INO80 (Shen et al., 2000). The ATPase subunits provide the basis for the currently accepted classification of chromatin remodelling complexes. The prototype SWI/SNF (SWItch/Sucrose-Non-Fermenting) complex is built around the founding member of the remodelling ATPase family, the SWI2/SNF2 (Cote et al., 1994). The unique feature of the SWI2/SNF2 subfamily is the occurrence of a conserved C-terminally located motif of roughly 110 amino–acids called the bromodomain, which is implicated in the recognition of the acetylated core histone tails. Yeast has two highly related complexes: SWI/SNF and RSC (Cote et al., 1994; Cairns et al., 1996), each of which contains the SWI2/SNF2-type ATPase plus ten or more subunits, several of which are identical or highly similar. The discovery of human homologues of the SWI2/SNF2 ATPase: hBrm and BRG1, allowed the purification and characterisation of human SWI/SNF-like complexes which were shown to contain eight or nine subunits. The composition of these complexes vary with the cell type (Wang et al., 1996). A similar complex, called BRM, has been purified from Drosophila and contains proteins homologous to those in yeast and human SWI/SNF-type complexes (Kal et al., 2000). The ISWI subfamily of complexes have been purified from a number of different sources. A distinctive feature of the ISWI-like ATPases is the

PLANT CHROMATIN

113

chromatin recognition module consisting of two SANT-like domains located in the C-terminal part of the enzyme. In Drosophila, which has one ISWI-type protein, three ISWI-containing remodelling complexes have been discovered so far. In the NURF (NUcleosome Remodelling Factor) complex, the ISWI ATPase is associated with a large 301 kDa protein, an inorganic pyrophosphatase, and the p55 protein containing a WD-repeat (Tsukiyama and Wu, 1995; Gdula et al., 1998; Martinez-Balbas et al., 1998; Xiao et al., 2001). All the subunits, including ISWI, interact directly with NURF 301 (Xiao et al., 2001), a protein which acts as the hub for the formation of the complex. In vitro, NURF is able to increase the rate of transcription by facilitating the access of transcription factors to promoter sequences (Mizuguchi et al., 1997). The CHRAC (CHRomatin Accessibility Complex) consists of ISWI, the 180 kDa Acf1 subunit and two small histone fold proteins (Varga-Weisz et al., 1997; Corona et al., 2000; Eberharter et al., 2001). The ACF complex is very similar to CHRAC; it contains both ISWI and Acf1 but lacks the two smaller subunits (Ito et al., 1997). In in vitro assays both complexes catalyse nucleosome movement in cis and hence can induce regular spacing of nucleosomes on DNA (Ito et al., 1997, 1999; Varga-Weisz et al., 1997; Langst et al., 1999). Chromatin remodelling complexes with a subunit composition similar to that of CHRAC and ACF have been discovered in many vertebrates and in yeast (Tsukiyama et al., 1999; Gushin et al., 2000; Poot et al., 2000). Both Acf1 and NURF 301 belong to a family of related proteins that all associate with homologues of ISWI (Ito et al., 1997; Eberharter et al., 2001). The distinguishing feature of the Acf1-related proteins is a combination of conserved domains including the bromodomain, PhD fingers, a DDT domain and BAZ- and WAC sequence motifs. Interestingly, related proteins predicted from Arabidopsis genome sequence analysis do not contain the bromodomain. The occurrence of two chromodomains is a distinctive feature of the CHD/Mi2 nucleosome remodelling ATPases. Mi2 has been shown to be a member of NuRD (Nucleosome Remodelling and Deacetylation) complexes in various species. As the name suggests, these complexes, in addition to other subunits, also contain the histone deacetylase HDAC1/HDAC2, which is associated with the Mi2 ATPase (Xue et al., 1998; Zhang et al., 1999). The subunit composition of NuRD led to speculation that its remodelling activity renders the histone tails accessible to deacetylation. Consistent with this idea, the ATPase activity has been shown to increase the rate of nucleosomal histone deacetylation (Ng et al., 1999; Feng and Zhang, 2001). As histone hypoacetylation is generally correlated with repression of transcription, it was speculated that NuRD complexes are

114

J. BRZESKI ET AL.

involved in establishing a repressive chromatin environment. NuRD can associate with the MBD2 protein to form the MeCP1 complex that preferentially interacts with methylated DNA (Feng and Zhang, 2001). Thus, the MBD2 may target the remodelling and histone deacetylase activities of NuRD to nucleosomes-containing methylated DNA. Individual members of the SWI2/SNF2, ISWI and CHD/Mi2 subfamilies have been analysed as isolated recombinant proteins outside the complex context and shown to have chromatin remodelling activity in vitro (Corona et al., 1999; Phelan et al., 1999, 2000; Brehm et al., 2000). However, additional subunits are required to reconstitute full enzymatic activity of the complexes. Recently, similar studies have been performed on members of two other subfamilies of SWI2/SNF2 group: the human ERCC6/CSB protein, involved in DNA repair, and the Arabidopsis DDM1 (Decrease in DNA Methylation) protein (At5g66750). The recombinant ERCC6/CSB was tested in a number of biochemical assays and shown to have chromatin remodelling activity (Citterio et al., 2000). We applied a similar strategy to demonstrate the in vitro remodelling activity of DDM1 (Brzeski and Jerzmanowski, 2003). Thus, both proteins define novel subfamilies of chromatin remodelling factors, although it is not known whether they form larger complexes in vivo. So far, plant chromatin remodelling enzymes have received little attention. While DDM1 is the only plant protein which has been shown to remodel chromatin in biochemical assays, 36 other genes encoding putative proteins of the SWI2/SNF2 family can be identified in the Arabidopsis genome. To see how these proteins are placed in a broader context of the SWI2/SNF2 family we have surveyed the proteomes of five fully sequenced eukaryotes: Saccharomyces cerevisiae, S. pombe, D. melanogaster, C. elegans and A. thaliana, supplemented with human and mouse sequences, for putative chromatin remodelling ATPases (Fig. 1). While the phylogenetic tree of the SWI2/SNF2 family confirms the existence of a number of well-defined subfamilies (see the discussion above), any conclusions concerning early evolutionary events and the exact relationships between subfamilies are uncertain due to a relatively weak bootstrap support of the deep nodes of the tree. However, despite the weakness of the statistical support and the differences between the minimum parsimony and the distance-based trees (Fig. 1), the overall pattern shows that the SWI2/SNF2, CHD/Mi2, ISWI and DDM1 subfamilies are closely related in evolutionary terms. The two other subfamilies of chromatin remodelling ATPases: INO80 and ERCC6/CSB are more distantly related. Interestingly, two different subfamilies, one containing the Drosophila

PLANT CHROMATIN

115

Fig. 1. Reconstruction of the phylogenetic tree of the SWI2/SNF2 family of ATPases with distance-based (A) and minimum parsimony (B) methods. To generate the data set the proteomes of A. thaliana, S. cerevisiae, S. pombe, C. elegans, D. melanogaster, M. musculus and H. sapiens were searched using PSI-BLAST. Six independent database searches were conducted using sequences characteristic for the six known families of the remodeling ATPases. The query sequences were trimmed to about 100 amino acid long signature motif of the SWI2/SNF2 family. This procedure resulted in 167 non-redundant sequences. The sequences were then aligned using ClustalX. Gap-containing columns were removed prior to the phylogenetic analysis. To reconstruct the evolutionary trees distance-based (panel A) and minimum parsimony (panel B) methods were used as implemented by the Neighbor and Protpars programmes of the Phylip package (Felsenstein, 1993). For clarity, only the Arabidopsis and ‘‘marker’’ sequences are included. The bootstrap values supporting defined families are shown. The shaded areas depict protein families with experimentally proven chromatin remodeling activity.

116

Fig. 1.

J. BRZESKI ET AL.

Continued

DOMINO and the other the yeast Fun30p protein, show a rather closer relationship to INO80. The DOMINO protein has been shown to act in developmental processes in the fruit fly (Ruhf et al., 2001). Even though these proteins have not been examined in chromatin remodelling assays, it is tempting to speculate that DOMINO and Fun30p define novel chromatin remodelers. Nothing is known about the chromatin remodelling activity of members of the more distant subfamilies described in Fig. 1. The genes encoding the SWI2/SNF2-like proteins are duplicated in an event that

PLANT CHROMATIN

117

appears to have occurred relatively recently. An interesting exception is the CHD/Mi2 subfamily which was divided early in evolution into three closely related branches: the two types of CHD/Mi2 ATPases and the KISMET group. KISMET-like proteins were not found in plants and fungi. Arabidopsis proteins are represented in all subfamilies described by the phylogenetic tree on Fig. 1, suggesting that the spectrum of chromatin remodelling factors is conserved between all major taxa with the only exception being the KISMET group of the CHD/Mi2-type subfamily, as mentioned above. A. ARABIDOPSIS HOMOLOGUES OF THE CENTRAL ATPASE AND OTHER CORE SUBUNITS OF THE PROTOTYPE SWI/SNF-TYPE COMPLEXES

All prototype SWI/SNF-type complexes studied so far contain a minimal structural and functional core composed of four evolutionarily conserved subunits; homologues of yeast proteins SWI2/SNF2 (the ATPase, major catalytic subunit), SNF5, SWI3 and SWP73 (Sawa et al., 2000; Sudarsanam and Winston, 2000). An assembly of minimum core subunits has been shown to remodel chromatin in vitro with an efficiency comparable to that of the whole complex (Sudarsanam and Winston, 2000). Both experimental analyses (Brzeski et al., 1999; Sarnowski et al., 2002) and database surveys (Verbsky and Richards, 2001) indicate that the existence of the prototype SWI/SNF complexes in plants is highly probable, although so far, no complete plant chromatin remodelling complex has been described. The results of homology searches, in agreement with the parsimonious phylogenetic tree described by Verbsky and Richards (2001) place four Arabidopsis proteins (At5g19310, At3g06019, At2g28290 and At2g46020) within the canonical SWI2/SNF2 subfamily. Which of these four putative ATPases are most likely candidates for the major catalytic subunit of a SWI/ SNF-type complex? Analysis of the available data from two-hybrid screens (Khavari et al., 1993; Treich et al., 1998) and co-immunoprecipitation studies (Murchardt et al., 1995) for yeast and animal core subunits indicates that the N-terminal fragment of SWI2/SNF2-type ATPases is responsible for binding the SWI3- and SNF5-type subunits. The most characteristic conserved motif of this fragment is domain 2 (Fig. 2). Upon searching the protein database with the sequence of domain 2, only four Arabidopsis proteins were found (Fig. 2). Three of them: At2g28290, At3g06010 and At5g19310 are members of a set of four proteins listed in an earlier survey aimed at identifying putative Arabidopsis SWI2/SNF2 ATPases (Verbsky and Richards, 2001). The fourth is a hypothetical protein annotated as

118

J. BRZESKI ET AL.

At2g46010. It consists of 943 amino acids resembling the N-terminal tail of the SWI2/SNF2-type ATPases. The adjacent ORF At2g46020 is the fourth protein of the set of Verbsky and Richards (2001). It has the SWI2/SNF2type ATPase domain but lacks the typical N-terminal tail. Because these two ORFs are on the same DNA strand and less than 200 bp apart, we suggest that they encode fragments of one protein, a SWI2/SNF2-type ATPase with a canonical N-terminal tail. We will refer to it temporarily as At2g46010/ 46020. However, the similarity of At2g46010/46020 domain 2 to SWI2/ SNF2 domain 2 is significantly lower than that of the domain 2 of two other putative Arabidopsis SWI2/SNF2 ATPases, At2g28290 and At3g06010 (compare the expectation values shown in Fig. 2). All SWI2/SNF2 ATPases which form the catalytic subunit of SWI/SNFtype complexes in yeast, Drosophila and humans contain a bromodomain. Of the four candidate ATPases of Arabidopsis, only At2g46010/46020 has a C-terminal region which resembles a bromodomain. To confirm this observation we performed a careful analysis of the genomic regions (20 kb and 2 ORFs up- and downstream) adjacent to the four identified

Fig. 2. Schematic organisation of the prototype yeast SWI2/SNF2 ATPase and its Arabidopsis homologues. Black – domain 2, light gray – region including seven motifs characteristic for all of the SWI2/SNF2-type ATPases, dark gray – bromodomain. Expectation values are shown under the domains.

PLANT CHROMATIN

119

SWI2/SNF2-type ATPase genes. None of these regions encodes a bromodomain motif, which rules out the possibility of an error in genome annotation. However, some other Arabidopsis proteins may contain a bromodomain characteristic of the SWI2/SNF2 proteins. SWI2/SNF2type bromodomains form a subgroup and have common amino acids in at least 19 positions. We obtained an alignment of 266 bromodomains from the PRODOM database (Corpet et al., 2000). The Hidden Markov Model (HMM) constructed from this alignment was used to search the Arabidopsis protein and translation databases (the programmes for construction of HMMs were from the GCG package). The resulting sequences were aligned to the model and compared manually to SWI2/ SNF2-type bromodomains for sequence similarities. None of the over 25 bromodomain-like sequences found in Arabidopsis has the SWI2/SNF2type bromodomain ‘signature’ amino acids in more than 7 of these 19 positions. A general sequence comparison with the bromodomain of At2g46010/46020 confirms that it has no significant similarity to the bromodomain of the yeast SWI2/SNF2 ATPase. A homologue of yeast SNF5 named BSH has been identified in Arabidopsis and was shown to complement the snf5 mutation in yeast (Brzeski et al., 1999). The human homologue of SNF5, INI1, binds to hBRM via aa 203-329 (Murchardt, 1995). Upon searching the protein database, 20 sequences similar to this region were found. 14 of them (excluding redundant hits and BSH) were aligned and used to create an HMM. Following a search of Arabidopsis protein and translation databases, only BSH, annotated as At3g17590, and encoded by a single copy gene, was found to possess significant similarity. Yeast SWI3 interacts with SWI2/SNF2 via a region which contains a SANT domain that is evolutionarily conserved in Drosophila and human SWI3 homologues (Crosby et al., 1999). This region was used to search the protein database. Sequences present in four Arabidopsis proteins: At1g21700, At2g33610, At4g34430 and At2g47620 were identified with an expectation value of below 1010. We conclude that these four proteins are true Arabidopsis homologues of SWI3. The sequences responsible for the interactions or function of SWP73-type proteins have yet to be identified. The Arabidopsis genome has two highly similar homologues of yeast SWP73: At3g01890 and At5g14170, which show 83.7% sequence identity to each other. Is it possible to predict the composition and properties of putative SWI/ SNF-type complexes in Arabidopsis, based on the results of the genomewide analysis discussed above? Two of the four putative Arabidopsis SWI2/ SNF2-type ATPases: (At5g19310 and At2g46010/46020) have a poorly

120

J. BRZESKI ET AL.

conserved domain 2, which may be important for binding of SWI3-type subunits. This could indicate that they bind other core subunits differently, perhaps utilizing a binding site closer to the N-terminus (Treich et al., 1998), or that they have different preferences for the SWI3-type subunits. Alternatively, these two proteins may perform functions not related to the SWI/SNF complex. The bromodomain is a distinguishing feature of the classical SWI2/SNF2-type ATPases (see the discussion above). Surprisingly, unlike the yeast and animal proteins, the putative plant prototype SWI2/ SNF2 ATPases either lack or have a degenerate form of the bromodomain motif (Fig. 2). Two of the putative ATPases (At3g06010 and At2g28290 or SYD) are more likely to interact with SWI3- and SNF5-type subunits. As revealed by sequence analysis, no other Arabidopsis proteins could provide a ‘SWI2/SNF2-type’ bromodomain for the putative complex. Because the bromodomain plays a role in the recognition of histone H3 and H4 tails when they are acetylated at specific lysines (Hudson et al., 2000), its absence or degeneration could have a profound effect on the mode of operation of SWI/SNF-type complexes in plants (see further discussion below). SNF5 protein is essential for co-ordinating the assembly and nucleosome remodelling activity of the SWI/SNF complex (Geng et al., 2001). The finding that only one SNF5-type gene (BSH) occurs in the Arabidopsis genome makes BSH an ideal marker of any putative Arabidopsis SWI/SNFtype complex. The presence of four SWI3 and four SWI2/SNF2 homologues in Arabidopsis is surprising, but perhaps reflects the general trend in plants of multiplication of genes encoding the SWI3- and SWI2/SNF2-type proteins. The functional SWI/SNF complex in yeast contains one molecule each of SWI2/SNF2, SNF5 and SWP73 and two molecules of SWI3. In Arabidopsis, in addition to a single SNF5-type protein, there are four potential SWI2/ SNF2-type ATPases, four SWI3- and two SWP73-type proteins. The presence of conserved binding sites in all four SWI3-type proteins and in at least two SNF2-type proteins, and the fact that at least three of the four SWI3-type proteins can interact and form heterodimers (Sarnowski et al., 2002) indicates that in Arabidopsis there is the potential for multiple SWI/SNF-type complexes with different combinations of core subunits, a situation reminiscent of that in mammals (Sif et al., 2001). It is possible that different Arabidopsis SWI/SNF type complexes are active in different physiological conditions and/or developmental phases (Wang et al., 1996). Recently, the SPLAYED (SYD) gene, encoding one of the four putative SWI2/SNF2 homologues (At2g28290) has been identified as a component in the temporal control of the switch from vegetative to reproductive development in Arabidopsis. SYD is also required for regulation of floral

PLANT CHROMATIN

121

homeotic gene expression, meristem maintenance during reproductive development and the correct morphogenesis of carpels and ovules (Wagner and Meyerowitz, 2002). Depletion of BSH (Brzeski et al., 1999) using an antisense strategy would be expected to inactivate all SWI/SNFtype complexes. Such anti-BSH plants show a pleiotropic phenotype involving reduction in apical dominance and infertility. This phenotype suggests the involvement of BSH in numerous physiological functions. The recent finding that one of the Arabidopsis SWI3 homologues (At2G33610) interacts with FCA, a protein regulating flowering time (Sarnowski et al., 2002) is in accordance with the hypothesis that Arabidopsis has a family of specialised SWI/SNF-type complexes. The importance of these complexes in development is emphasised by the embryo-lethal phenotype of plants homozygous for an insertional mutation in BSH (K. Olczak — personal communication).

B. CHD/Mi2 SUBFAMILY IN ARABIDOPSIS

PICKLE (At2g25170) is the only member of the CHD/Mi2 subfamily in Arabidopsis that has been analysed so far. Pickle mutants express embryonic traits even after germination. The mutant phenotype is strongly enhanced in the presence of gibberellin biosynthesis inhibitors. Interestingly, Ogas et al. (1999) have shown that a pickle genetic background results in postembryonic derepression of the gene LEC1, that is thought to serve as a critical activator of embryo development. These data suggest that PICKLE acts as part of a gibberellin-dependent switch that regulates the transition from embryonic to postembryonic development.

VI. CORE HISTONE MODIFICATIONS BY ACETYLATION AND METHYLATION The N-terminal unstructured tails of core histones are subject to posttranslational modifications, mostly acetylation, methylation and phosphorylation, which are targeted to specific amino acids. Because of the critical role of tail domains in chromatin fibre condensation, these modifications may exert a profound effect on chromatin structural dynamics. For example, histone acetylation interferes with interaction of nucleosomal arrays which is reflected by a requirement for increased concentration of Mg2þ to achieve 50% chromatin oligomerisation in vitro (Tse et al., 1998; Pollard et al., 1999). In contrast, methylation increases the tightness of the

122

J. BRZESKI ET AL.

association between core histone tails and DNA. Such alterations may affect the interactions of the histone tails and DNA with specific proteins needed to establish stable open or repressive chromatin states. Acetylation of core histone tails is mediated by histone acetyltransferases (HATs) and removed by histone deacetylases (HDACs). The first report suggesting a link between histone acetylation and regulation of transcription was published in 1964 (Allfrey et al., 1964). However, the breakthrough discoveries came much later with the demonstration that the Tetrahymena HAT enzyme is a homologue of the Gcn5 transcriptional co-activator from yeast (Brownell et al., 1996) and that a mammalian histone deacetylase is homologous to Rpd3, a yeast repressor (Taunton et al., 1996). Both Gcn5 and Rpd3 are members of multiprotein complexes involved in the control of many yeast genes. Since then numerous studies in yeast and animals using chromatin immunoprecipitation have confirmed the general link between transcriptional competence of genes and the specific pattern of core histones acetylation (Krebs et al., 1999; Parekh and Maniatis, 1999). In contrast, deacetylation of histones was shown to be associated with a transcriptionally repressed chromatin state (Suka et al., 2001). The targets of acetylation are lysines located in conserved positions of the N-terminal tails, mostly in histones H3 (positions 9, 14, 18, 23) and H4 (positions 5, 8, 12, 16) but to some extent also in histones H2A and H2B. The methylation of histones is catalysed by histone methyltransferases (HMTs). The first report on the core histone methylation at "-NH2 groups of lysines appeared in 1964 (Murray, 1964). Interestingly, methylation of lysines in the C-terminal unstructured tail of histone H1 was reported in a slime mould Physarum polycephalum (Jerzmanowski and Maleszewski, 1985). In contrast to acetylation, the methylation of core histones is rather stable. No global histone demethylase activity has been found which could counterbalance the activity of HMTs and help maintain dynamic equilibrium between methylated and non-methylated core histones. It is now known that methylation, like acetylation, occurs at lysines located at specific conserved sites of histones H3 (positions: 4, 9, 27 and 36) and H4 (position 20). In mammals, the lysines at specific sites of H3 and H4 N-termini are either acetylated or methylated, with the exception of lysine 9 of H3 which can be targeted for both modifications. Methylation can also occur at arginine residues in the core histone tails (Stallcup, 2001). The discovery that some of the genes encoding mammalian homologues of the Drosophila position-effect variegation (PEV) modifier SU(VAR)3-9 proteins have site-specific lysine HMT activity towards lysine 9 of histone H3 (H3K9) established a link between histone methylation and the propagation of heterochromatic chromatin domains (Lachner and Jenuwein, 2002). It is

PLANT CHROMATIN

123

now known that the catalytic activity of this specific methyl transferase is dependent on the presence of the 130-amino acid SET domain in association with an adjacent cysteine-rich domain (Jenuwein, 2001). In mammals, over 70 genes with the SET domain have been identified. Some encode proteins with HMT activity specific for lysines located at different conserved positions in the H3 and H4 N-terminal tails (Jenuwein, 2001). The effects of H3 methylation strongly depend on the position of the affected lysine. Whereas H3K9 methylation in mammals is associated with repressed chromatin, H3K4 methylation seems to mark active chromatin domains (Litt et al., 2001; Noma et al., 2001). Most DNA in the cells of highly differentiated multicellular organisms is located within hypoacetylated, transcriptionally inactive heterochromatin. Recent data show that modifications of core histones by acetylation which affect chromatin accessibility, occur both at a broad level, leading to partial decondensation of whole domains (Litt et al., 2001; Schu¨beler et al., 2001), and at a local level as targeted acetylation of histones in promoterassociated nucleosomes (Brown et al., 2000; Forsberg and Bresnick, 2001). While broad scale acetylation establishes the transcriptionally competent domains, it does not result in transcriptional activation of particular genes. The latter is achieved by recruiting HAT-containing co-activator complexes to specific promoters by transcriptional activators. However, such complexes (for example the large SAGA complex in yeast) have other subunits which can function as activators independently of the HAT subunit. Histone H3K9 methylation, which is mainly linked with transcriptionally inactive chromatin, is not restricted to constitutive heterochromatin domains. It is also found in euchromatic regions where it marks transcriptionally inactive genes. H3K9 methylation establishes an affinity for several repressive elements, like HP1 (Heterochromatin Protein 1) and the HDAC-containing NuRD complex (see Lachner and Junewein, 2002, for review). In Neurospora and Arabidopsis, H3K9 methylation has also been shown to be a signal for DNA methylation. In Arabidopsis the Kryptonite gene encodes a SET domain-containing protein with H3K9 methyltransferase activity. Kryptonite-mediated methylation of H3K9 recruits the chromodomain containing LHP1 protein (an Arabidopsis homologue of HP1) which in turn recruits the DNA methyltransferase (Lindroth et al., 2001; Jackson et al., 2002) (see discussion below). The increasing flow of data from different model organisms documents the importance of the interplay between histone H3 methylation occurring at different positions (like those at H3K9 and H3K4) as well as between methylation and acetylation at the same and neighbouring histone tails (Rice and Allis, 2001). The interplay between acetylation and H3

124

J. BRZESKI ET AL.

phosphorylation at Ser 10 may be of similar importance (Clayton et al., 2000). Combinations of the above and other modifications (like ubiquitination) could well provide an effective code for marking chromatin sites. Many chromatin modifiers possess domains that can recognise specific nucleosomal modifications, for example the bromodomain occurs both in yeast and animal SWI2/SNF2 ATPases and in Gcn5-type HATs. Thus, the histone code may be recognised by co-operating chromatin regulators including chromatin modifying complexes capable of ATP-dependent remodelling and/or containing HAT and HDAC activities, as well as by chromodomaincontaining proteins like HP1 or chromo-DNA methyltransferases.

VII. PLANT ENZYMES RESPONSIBLE FOR ACETYLATION AND METHYLATION OF CORE HISTONES In plants, histone H3 lysines are acetylated at positions 9, 14, 18 and 23 and those of histone H4 at positions 5, 8, 12, 16 and 20 (Waterborg, 1990, 1992). Also acetylated, albeit to a lower extent, are the N-terminal tails of histone H2A and H2B. A recent study by Chua et al. (2001) confirmed that in plants, as in yeast and animals, the histone acetylation status is linked to the transcriptional state. It was also demonstrated that H3 and H4 histones associated with the enhancer/promoter region of a transcriptionally activated pea plastocyanin gene are hyperacetylated. Functional studies of several plant HDACs have confirmed their importance in gene control. In particular, the antisense silencing of AtHD1, an Arabidopsis homologue of yeast Rpd3, which resulted in hyperacetylation of histone H4, was correlated with a broad spectrum of developmental abnormalities (Tian and Chen, 2001). Interestingly, while these plants showed ectopic expression of the normally methylated SUPERMAN (SUP) gene indicating its release from transcriptional silencing, they did not show a changed pattern of DNA methylation. Analysis of the effects of mutations in HDA6, a gene encoding another Arabidospsis homologue of Rpd3, indicated that the deacetylation caused by this enzyme plays a role in the silencing of transgenes (Murfett et al., 2001). An important hint as to the vital role of histone acetylation in plants came from the observation that the HC toxin produced by Cochliobolus carbonum, a fungal pathogen of maize, is a specific inhibitor of plant HDACs (Brosch et al., 1995). Taken together, the above results suggest that the role played by core histone acetylation/deacetylation in gene regulation in plants is in general similar to that found in fungi and animals. However, a comparison of the global distribution of acetylated H3 and H4 histones in chromosomes indicates that there may be some important differences

PLANT CHROMATIN

125

between plants and animals. Whereas H3 and H4 show similar acetylation patterns throughout mammalian chromosomes (Belyaev et al., 1996), the patterns for these two histones differ visibly in field bean chromosomes (Belyaev et al., 1998). The current classification distinguishes several different groups of HATs and HDACs (Khochbin et al., 2001; Roth et al., 2001). HATs are classified into groups A and B. The group B HATs catalyse the acetylation of cytoplasmic histone H4 at Lys 5 and Lys 12, an event preceding the assembly of the nucleosome. The HATs of group A are nuclear enzymes which can be further subdivided into GCN5/MYST-, CBP/p300-, TAFII 250- and the mammal-specific nuclear receptor co-activator family, by virtue of their differing substrate specificities. The HDACs are subdivided into RPD3/-, SIR2- (Silent Information Regulator 2) and the plant-specific HD2 family. Plant histone acetylases and deacetylases have been intensively studied and are relatively well characterised biochemically (see Lusser et al., 2001, for a recent review). A global picture of the diversity of plant HATs and HDACs was provided by a recent comparative genome analysis of A. thaliana, S. cerevisiae, S. pombe, C. elegans and D. melanogaster (Pandey et al., 2002). Plants, like animals and fungi, possess a single member of each of the three subfamilies of the GCN5/MYST family, which suggests that these HATs have an old and well-conserved function. However, the three kingdoms show considerable differences in the size of the other families of HATs and HDACs, as well as in the domain composition of their members. These differences indicate a substantial functional diversification of chromatin mechanisms linked to histone acetylation which occurred during plant evolution. One of the most significant examples again concerns the occurrence of the bromodomain. Arabidopsis has only one bromodomain in the conserved TAFII 250-type HAT, compared with two bromodomains occurring in a contiguous fashion in the corresponding animal HATs. Moreover, in contrast to animal p300/CBP HATs, all of which have a bromodomain, the Arabidopsis CBP-type HAT has no bromodomain at all. This is reminiscent of the lack of bromodomains in putative Arabidopsis SWI2/SNF2-type ATPases and homologues of Acf1 proteins (see above). In plants, unlike animals, not only Lys 9 of H3 but also Lys14, Lys18 and Lys 23 of H3 and Lys 20 of H4 can be targets for either acetylation or methylation (http://research.nhgri.nih.gov./histones/posttrans/shtml). The enzymes responsible for most of these diverse methylations are yet to be characterised. As shown by a recent genome-wide screen combined with RT-PCR analysis of cDNAs, Arabidopsis has at least 29 active genes encoding SET-domain proteins (Baumbusch et al., 2001). This represents

126

J. BRZESKI ET AL.

considerable diversity when compared to the number of SET-domain proteins in other organisms. The Arabidopsis SET genes fall into four evolutionarily conserved families named after their Drosophila prototype members: E(z), TRX, ASH1 and SU(VAR)3-9. Interestingly, the SU(VAR)3-9 family consisting of homologues of yeast and animal histone H3-specific methyltransferases, is the biggest with 15 members. One of them is Kryptonite, the first Arabidopsis protein shown to be a histone H3K9 methyltransferase (Jackson et al., 2002). None of the Arabidopsis SU(VAR)3-9 type proteins contain an N-terminal chromodomain present in the animal homologues and required for specific binding to methylated H3. It is unclear whether the novel YDG domain identified in Arabidopsis SU(VAR)3-9 homologues can substitute functionally for the chromodomain. A characteristic property of animal histone acetyltransferases and histone methyltransferases is the occurrence of protein-binding domains that recognise the products of their own enzyme activity (the bromodomain for the acetylated- and the chromodomain for the methylated histone N-termini). These recognition modules create the potential for processivity, i.e. for enhancing the effect of modifications by replicating its pattern in neighbouring nucleosomes. Such self-perpetuation of an epigenetic signal may be critical for its long-term effect and stability in chromatin. It is highly intriguing that in Arabidopsis the key enzymes involved in histone modification and in the ATP-dependent chromatin remodelling lack the protein recognition modules which could mediate processivity. Does this indicate that at least some of the interactions mediated by histone modifications are less stable, i.e. more easily reversible, in plants than in animals?

VIII. METHYLATION OF DNA Arabidopsis has three classes of DNA methylating enzymes. As the literature on this subject has recently been reviewed (Finnegan and Kovac, 2000) we will concentrate on some recent studies that have addressed the interdependence of DNA and histone H3 methylation. The MET1 methyltransferase is mainly responsible for the typical CpG methylation (Finnegan et al., 1996; Roenmus et al., 1996; Genger et al., 1999). CHROMOMETHYLASE 3 (CMT3) is required for maintenance of the methylation status at CpXpG sites (Lindroth et al., 2001), although it cooperates with DOMAINS REARRANGED METHYLASE 1 and 2 (DRM1 and DRM2) in maintaining CpXpG and assymetric methylation at some other loci (Cao and Jacobsen, 2002a). DRM1 and 2 also act as de-novo

PLANT CHROMATIN

127

DNA methylases (Cao and Jacobsen, 2002b). The first observation suggesting a relationship between histone H3 lysine 9 methylation and DNA methylation in Arabidopsis came from genetic studies on gene silencing at the SUPERMAN locus (Jackson et al., 2002). The epigenetic silencing of the SUP locus, associated with dense methylation at CpXpG and assymetric sites, is released in cmt3 mutants. A genetic screen for suppressors of SUP gene silencing revealed a loss-of-function allele of the KRYPTONITE (KYP) gene encoding the H3K9-specific methyltransferase (see above). Interestingly, the depletion of KYP function was associated with reduced DNA methylation. These results suggested that KYP acts upstream of CMT3 DNA methylase. CMT3 was demonstrated to interact in vitro with LHP1, an Arabidopsis homologue of HP1 protein, which specifically recognises histone H3 tails methylated at lysine 9 (see above). Therefore, it seemed likely that LHP1 might serve to recruit CMT3, which in turn locked-up the inactive, silent state at the SUP locus. This view was further supported by the results of chromatin immunoprecipitation studies on DNA methylation mutants. Gendrel et al. (2002) used the ddm1 mutant with severe decrease in cytosine methylation at both CpG and CpXpG symmetric sites, to analyse the pattern of histone H3K4 and K9 methylation at over 50 different loci in the heterochromatic knob of chromosome IV. In general, the depletion of DDM1 function resulted in a decrease in H3K9 methylation and an increase in H3K4 methylation at the analysed loci. However, the effect was apparently limited to a subset of loci and could not be unambiguously correlated with transcriptional derepression. Furthermore, the overall content of K9 and K4 methylated H3 was not affected, suggesting that the ddm1 mutation had relatively little effect on the H3 methylation level, in contrast to the observed 70% reduction in DNA methylation (Vongs et al., 1993; Kakutani et al., 1995). Nevertheless, these results suggested that DDM1 is involved not only in DNA methylation but also in the maintenance of the H3 methylation pattern. Another possible interpretation is that DDM1 acts primarily on DNA methylation and only indirectly on H3 methylation, presumably by a feedback loop regulation. In a recent study, Johnson et al. (2002) extended the above analysis using a collection of additional DNA and H3 methylation mutants that included ddm1, cmt3, met1, the double cmt3/met1 mutant and kyp. In agreement with previous data demonstrating the upstream role of H3K9 methylation over DNA methylation, they found that DNA methylase mutants did not affect H3K9 methylation level in the 180-bp CEN repeats. These results did not support the idea of a feedback loop between DNA methylation and H3K9 methylation. Interestingly, the depletion of H3K9 methylation in the

128

J. BRZESKI ET AL.

kyp line affected DNA methylation only at CpXpG sites, suggesting that the CpG methylation does not rely on H3K9 methylation or that Arabidopsis has a different H3K9 HMTase that is specifically linked to CpG DNA methylation. The latter hypothesis seems very unlikely. It raises the question of how the CMT3 and MET1 DNA methylases could discriminate between chromatin-containing histone H3 methylated at the lysine 9 by different HMTases. It appears possible that CMT3 and MET1 associate with two different HMTases to form complexes that sequentially methylate H3K9 and cytosines at CpXpG and CpG sites, respectively. However, the methylated CpG and CpXpG sites are often found on the same nucleosome. Surprisingly, the mutation in DDM1 caused a drop in H3K9 methylation in the 180-bp CEN repeats almost to the level seen in kyp mutants. This observation suggests a direct cooperation between DDM1 and KYP HMTase. The DNA methylation level was apparently reduced at both CpXpG and CpG sites. These data prompted the authors to speculate that DDM1 ‘lives a double life’. According to the proposed model, DDM1 would act in concert with KYP HMTase to establish H3K9 methylated nucleosomes that could be recognised by CMT3 via its interaction with LHP1. The second role of DDM1 would be to cooperate with MET1 to methylate DNA at CpG sites. The above model, which is based on studies of transcriptionally silent or not very active 180-bp CEN repeats, was challenged by the results of an analysis of Ta2 and Ta3 retrotransposons. These loci are normally heterochromatic and transcriptionally silent but can become active in the context of mutations affecting DNA methylation. In agreement with the previous results the authors observed a loss of H3K9 methylation at the Ta2 locus in ddm1 and kyp mutants, whereas no loss was observed in cmt3 and met1 mutants. Interestingly however, the loss of H3K9 methylation occurred in double cmt3/met1 mutants. Transcriptional profiling of the mutants demonstrated the reactivation of the Ta2 element in ddm1 and cmt3/met1 mutants, while no release of transcriptional silencing at the Ta2 locus was observed in the kyp mutant. These data led the authors to speculate that both CpXpG and CpG DNA methylation systems lock-up an inactive state that is only marked (in case of CpXpG) by H3K9 methylation. When DNA methylation was erased, transcription occurred allowing nucleosome exchange and hence the loss of K9-methylated histone H3. The lack of transcriptional activity in the kyp mutant (which presumably fails to methylate DNA only at CpXpG sites) could be attributed to DNA methylation persisting at CpG sites. However, in contrast to the proposed scenario, the authors detected transcriptional activity of the Ta2 element in the met1 mutant which did not show any loss of H3K9 methylation.

PLANT CHROMATIN

129

Another piece of this puzzle has recently been added by immunocytologic studies on DNA and histone H3 methylation patterns in the nuclei of met1 and ddm1 mutants (Soppe et al., 2002). In the nuclei of wild-type Arabidopsis the majority of methyl-cytosine and H3K9-met signals co-localise in the bright DAPI-stained chromocentre dots. The staining with DAPI of the met1 and ddm1 nuclei revealed much smaller and less numerous chromocentres, indicating that the heterochromatic regions have been ‘dissolved’. Consistent with this observation, in FISH experiments some of the pericentromeric repetitive sequences appeared dislocated from the chromocentres. In agreement with the previous studies, the depletion of DDM1 function resulted in a comparable reduction in the chromocentre staining with antibodies raised against H3K9-met and methyl-cytosine. However, an analogous experiment with the nuclei of the met1 mutant line brought a surprise: there was a strong reduction in both DNA and H3K9 methylation in chromocentres. These data provoked a conclusion that is in conflict with the previous models. The authors proposed that the CpG DNA methylation precedes the modification of histone H3 lysine 9, which in turn induces DNA methylation at the CpXpG sites and the spreading of heterochromatin to the pericentromeric sequences. Another interpretation might be that the loss of DNA methylation in met1 enables transcription at dissolved chromocentres that in turn dislocates the K9-methylated histone H3 and causes equalisation of the staining pattern. Unfortunately, the transcription profile of pericentromeric sequences has yet to be established. What predictions can be made from the two models and can they be verified by the existing data? If heritable CpG methylation had the primary function of inducing H3K9 methylation and CpXpG modification leading to the nucleation or spreading of heterochromatin, the depletion of MET1 CpG methylase should influence the overall nuclear level of H3K9 and CpXpG methylation. Indeed, the met1 mutants showed a reduced level of CpXpG methylation in some loci. However, quantification of the H3K9 level in the ddm1 line, even though it was not very rigorous, failed to show any dramatic effect on histone H3 modification (Gendrel et al., 2002). Thus, the complex relationship between cytosine methylation at different sites and H3K9 methylation is far from being clear. It seems likely that the complex network of dependencies involved, vary at different loci. Such a variability might be dependent on a local dominance of the particular modification. Perhaps, there are ‘reader molecules’ in the nuclei capable of sensing subtle variations in the modification patterns and translating them into specific expression profiles. However, the existing data provide no support for such a hypothesis.

130

J. BRZESKI ET AL.

Not much can be inferred from the available data on how the chromatin remodelling activity of DDM1 might be involved in generation of the heterochromatin environment. Two hypotheses can be sketched, both of which require further studies to be verified. The often cited model assumes that the remodelling activity of DDM1 unwinds chromatin fibres to permit the methylation of DNA and of H3K9 (Jeddeloh et al., 1999a, Finnegan and Kovac, 2000; Verbsky and Richards, 2002). According to this view, DDM1 activity might be linked to DNA replication and the early events in chromatin assembly. Also, a tight association of DNA and/or H3K9 methylases with DDM1 can be expected. The model explains the overall loss of DNA methylation observed in ddm1 mutants by the inability of DNA methylases to efficiently modify nucleosomal DNA. It also assumes that the release of transcriptional silencing leads in turn to exchange of the nucleosomal histones and the redistribution of the H3K9 methylation pattern. Alternatively, the loss of CpG methylation could interfere with a correct recruitment of KRYPTONITE and similar HMTases. An alternative model postulates that DDM1 recognises chromatin fibres that have already been modified and acts as a folding factor creating a condensed chromatin environment that is not accessible to active de-methylation. Although the DNA de-methylation enzymes have not yet been identified, active de-methylation has been observed in plants during responses to environmental cues (Steward et al., 2002 and references therein). This model predicts the role for DDM1 in maintaining heterochromatin in the somatic cells. In fact, DDM1 activity is required to silence viral genes delivered to the somatic cells (Morel et al., 2000). The two models are not necessarily mutually exclusive and both explain why the phenotypic aberrations in ddm1 lines accumulate over many generations whereas the phenotype of the met1 plants is already present in the first generation (Finnegan et al., 1996; Kakutani et al., 1996; Ronemus et al., 1996; Jeddeloh et al., 1999b).

IX. THE PROBLEM OF LINKER HISTONES Linker (H1) histone is the only structural element of the nucleosome that has not yet been convincingly proven to play a role in the chromatin biochemical code. On the contrary, there is considerable difficulty in reconciling the results of experiments on structural and biochemical properties of H1 with the results of work aimed at establishing its function in vivo. The data from experiments probing the position of linker histones in nucleosomes and in the compact forms of nucleosomal filaments, as well as

PLANT CHROMATIN

131

from studies on the distribution and properties of different H1 variants, have been invariably interpreted as indicative of a critical function in chromatin architecture and dynamics and consequently in the regulation of such fundamental processes as transcription and chromosome condensation. At the same time complete knockouts of genes encoding H1 in unicellular Protista and fungi demonstrated that these simple Eukaryotes survive and prosper without linker histones (Shen et al., 1995; Ramon et al., 2000). Similarly, the knockout and overexpression of genes encoding different H1 variants in the mouse has so far produced no evidence that the variability of H1 in this complex multicellular organism is indeed functionally important (Fan et al., 2001). The variability of linker histones is also a characteristic of flowering plants (Jerzmanowski et al., 2000). Phylogenetic analysis of plant linker histones based on the sequences of the conserved GH1 domain, shows that in addition to major somatic-type variants there is a distinct branch of the so-called ‘stress-inducible’ variants of H1. These minor variants represent an old isoform, that arose before the separation of mono- and dicotyledonous plants. Another striking feature of the phylogenetic tree of plant H1s is the branch representing the ‘hybrid’ proteins. This branch groups sequences derived from proteins in which a typical GH1 domain is fused to domains characteristic of proteins from outside the H1 family. The most typical representatives of such proteins are the HMG1/Y proteins. The reason why the function of linker histones in higher eukaryotes is still poorly understood lies partly in the unusually efficient compensatory mechanisms. Suppression of one or more types of variants leads to immediate up-regulation of the remaining types, so that the overall ratio of linker histones to nucleosomes is preserved. Since this mechanism operates equally well in animals and in plants (Fan et al., 2001; Prymakowska-Bosak et al., 1999), it probably plays an important and conserved function in maintaining the proper stoichiometry of linker histones in chromatin. Clearly, the significance of the variability of linker histones must be secondary to maintaining their overall physiological level. Two recent works may provide some explanation of this effect. Horn et al. (2002) using nucleosomal arrays reconstituted in vitro on DNA consisting of twelve 208 bp repeats (208-12 DNA) showed that the presence of linker histones inhibited the remodelling of nucleosomes by human SWI/SNF, Xenopus Mi-2 and Xenopus ACF (an ISWI-type) complexes i.e. the three major classes of ATP-dependent chromatin-remodelling complexes. However, the remodelling could be rescued by phosphorylation of H1. Karymov et al. (2001) used the similar arrays of nucleosomes reconstituted on 208-12 DNA to study the effect of DNA methylation on chromatin structure. These

132

J. BRZESKI ET AL.

authors demonstrated that an excess compaction observed for the arrays reconstituted on the methylated DNA was the result of the additive effect of the methylation and the presence of H1. Neither of these two factors could bring about this effect without the other. Thus, in the in vitro system used, the DNA methylation must work in conjunction with linker histone binding to cause increased chromatin fibre compaction. Taken together, the results of these two studies indicate that H1 can itself block the access of chromatin remodelers and that DNA methylation is a factor capable of enhancing this effect. Are there any hints from the functional studies on the role of H1 that could corroborate the conclusions based on in vitro assays? Given the lack of any clear phenotypic effects of changed proportions of linker histone variants in mice (Fan et al., 2001), it is particularly interesting that such effects were documented in invertebrates and in angiosperm plants. In C. elegans, a decrease in one of the eight variants of H1 in this invertebrate led to the activation of a normally repressed reporter transgene in germ line cells without affecting the expression of this reporter in somatic cells. Correlated with this release of silencing were defects in germline differentiation and in some cases even sterility. Most importantly, the observed defects in germline development resembled those caused by mutation of the mes genes involved in germline-specific chromatin repression in C. elegans (Jedrusik and Schulze, 2001). Some of the mes genes are homologues of the Drosophila Polycomb Group genes involved in gene repression during development. In tobacco, it was possible, using the antisense approach, to construct plants which had a completely reversed ratio of major somatic variants (H1A and H1B) to minor variants (H1C to H1E) (Prymakowska-Bosak et al., 1999; Przewloka et al., 2002). Although this change had little or no effect on the growth and development of the plant, it did lead to disturbances in male gametogenesis and eventually to a male sterility phenotype. A common feature of the C. elegans and tobacco phenotypes is the deregulation of the precise genetic systems responsible for germline development. When comparing the effects in C. elegans and tobacco it should be remembered that these two organisms differ substantially in at least one aspect of chromatin repression. C. elegans has no methylated DNA and repression by chromodomain-containing proteins (HP1, Polycomb) is induced by DNA methylation-independent events. In contrast, the heterochromatin assembly in plants (as in vertebrates) is linked to DNA methylation. As discussed earlier, the H3Lys9 methylation in Arabidopsis requires HP1 (LHP1) to mediate the binding of chromodomain-containing DNA methyltransferase (Lindroth et al., 2001). Interestingly, the recent demonstration that H1 can directly interact with chromodomain-containing HP1 and Polycomb proteins suggests that

PLANT CHROMATIN

133

linker histones can facilitate and/or mediate the functions of these proteins (Nielsen et al., 2001). In plants, unlike in vertebrates, the inherited DNA methylation pattern is not erased during the first cleavage stages of the zygote. Interestingly, the only developmental phase in plants in which the overall level of methylated cytosine in DNA substantially decreases is the postmeiotic maturation of the male gametes (Oakley et al., 1997). The association between the disturbances in the native profile of linker histone variants and the correct pattern of gene expression needed to control precise developmental events in both C. elegans and tobacco suggest that linker histones may be critical for maintaining the repressed chromatin conformation at potentially active genes (i.e. mostly in euchromatin as opposed to constitutive heterochromatin), independently of whether the repression is linked with DNA methylation or not. The eventual confirmation of the universal role of H1 in establishing repressed euchromatin states must await the construction of multicellular eukaryotes completely lacking linker histones.

X. PERSPECTIVES Chromatin, in addition to being a superb architectural solution to the problem of DNA packaging, provides a structural basis for an immensely complex network of interactions. This is hardly surprising given the fact that it integrates an enormous number of various input data which have to be translated into a common language and communicated to the information system of DNA. This notwithstanding, chromatin serves as a cellular memory, ensuring the transmittance of characteristic features of the cells from one generation to the next. The true complexity of the chromatin network has become apparent with the uncovering of the diversity of histone and DNA modifications. Studies of plant chromatin, may add a great deal to our understanding of the general logic of chromatin-based regulation. A major goal is to correlate the characteristic features of plant chromatin with the specificity of plant developmental strategies. A good example is the plant-specific deficiency of the protein–protein recognition domains in HATs, HMTs and some of the SNF2-type remodelling ATPases. How does this relate to the universality of the histone-code hypothesis and, at the functional level, to the plasticity of the cellular differentiation–dedifferentiation systems? It should be remembered that plants are a valuable model and a natural reference system for studying the biological effects of DNA methylation in mammals. None of the other widely studied eukaryotic model organisms (yeast,

134

J. BRZESKI ET AL.

Drosophila, C. elegans) uses DNA methylation. However, in order to be useful for broad comparative analysis, plant chromatin research must provide more detailed information on the molecular organisation and biochemical properties of the plant structural proteins, enzymes and multisubunit complexes involved in chromatin regulation. Of particular importance would be the development of a reliable in vitro system (nucleosomal arrays reconstituted with recombinant plant histones) suitable for studying the molecular mechanisms of chromatin remodelling by plantspecific components.

ACKNOWLEDGEMENTS We would like to thank John Gittins for critical reading and comments on the manuscript. This work was supported by a Howard Hughes Medical Institute grant No. 55000312 (for A.J.), Polish Committee for Scientific Research grants Nos. 6PO4A 00320 and PBZ-039/PO4/2001 and a Foundation for Polish Science grant No. 2/2000. The laboratory is supported by Centre of Excellence in Molecular Biotechnology.

REFERENCES Allfrey, V. G., Faulkner, R. and Mirsky, A. E. (1964). Acetylation and methylation of histones and thier possible role in the regulation of RNA synthesis. Proceedings of the National Academy of Sciences USA 51, 786–794. Baumbusch, L. O., Thorstensen, T., Krauss, V., Fischer, A., Naumann, K., Assalkhou, R., Schulz, I., Reuter, G. and Aalen, R. B. (2001). The Arabidopsis thaliana genome contains at least 29 active genes encoding SET domain proteins that can be assigned to four evolutionarily conserved classes. Nucleic Acids Research 29, 4319–4333. Belyaev, N. D., Houben, A., Baranczewski, P. and Schubert, I. (1998). The acetylation patterns of histones H3 and H4 along Vicia faba chromosomes are different. Chromosome Research 6, 59–63. Belyaev, N. D., Keohane, A. M. and Turner, B. M. (1996). Differential underacetylation of histones H2A, H3 and H4 on the inactive X chromosome in human female cells. Human Genetics 97, 573–578. Brehm, A., Langst, G., Kehle, J., Clapier, C. R., Imhof, A., Eberharter, A., Muller, J. and Becker, P.B. (2000). dMi-2 and ISWI chromatin remodelling factors have distinct nucleosome binding and mobilization properties. EMBO Journal 19, 4332–4341. Brosch, G., Ransom, R., Lechner, T., Walton, J. D. and Loidl, P. (1995). Inhibition of maize histone deacetylases by HC toxin, the host selective toxin of Cochliobolus carbonum. The Plant Cell 7, 1941–1950. Brown, C. E., Lechner, T., Howe, L. and Workman, J. L. (2000). The many HATs of transcription coactivators. Trends in Biochemical Sciences 25, 15–19. Brownell, J. E., Zhou, J., Ranalli, T., Kobayashi, R., Edmondson, D.G., Roth, S. Y. and Allis, C. D. (1996). Tetrahymena histone acetylotransferase A: a

PLANT CHROMATIN

135

homolog to yeast Gcn5p linking histone acetylation to gene activation. Cell 84, 843–851. Brzeski, J., Podstolski, W., Olczak, K. and Jerzmanowski, A. (1999). Identification and analysis of the Arabidopsis thaliana BSH gene, a member of the SNF5 gene family. Nucleic Acids Research 27, 2393–2399. Brzeski, J. and Jerzmanowski, A. (2003). Deficient in DNA Methylation 1 (DDM1) defines a novel family of chromatin remodeling factors. Journal of Biological Chemistry Published as JBC Papers in Press 10.1074/ jbc.M209260200 Cairns, B. R., Lorch, Y., Li, Y., Zhang, M., Lacomis, L., Erdjument-Bromage, H., Tempst, P., Du, J., Laurent, B. and Kornberg, R. D. (1996). RSC, an essential, abundant chromatin-remodeling complex. Cell 87, 1249–1260. Cao, X. and Jacobsen, S. E. (2002a). Locus-specific control of asymmetric and CpNpG methylation by the DRM and CMT3 methyltransferase genes. Proceedings of the National Academy of Sciences USA 99 Suppl 4, 16491–16498. Cao, X. and Jacobsen, S. E. (2002b). Role of the Arabidopsis DRM methyltransferases in de novo DNA methylation and gene silencing. Current Biology 12, 1138–1144. Chua, Y. L., Brown, A. P. and Gray, J. C. (2001). Targeted histone acetylation and altered nuclease accessibility over short regions of the pea plastocyanin gene. The Plant Cell 13, 599–612. Citterio, E., Van Den Boom, V., Schnitzler, G., Kanaar, R., Bonte, E., Kingston, R. E., Hoeijmakers, J. H. and Vermeulen, W. (2000). ATPdependent chromatin remodeling by the Cockayne syndrome B DNA repair-transcription-coupling factor. Molecular and Cellular Biology 20, 7643–7653. Clayton, A. L., Rose, S., Barratt, M. J. and Mahadevan, L. C. (2000) Phosphoacetylation of histone H3 on c-fos- and c-jun-associated nucleosomes upon gene activation. EMBO Journal 19, 3714–3726. Corona, D. F., Eberharter, A., Budde, A., Deuring, R., Ferrari, S., Varga-Weisz, P., Wilm, M., Tamkun, J. and Becker, P. B. (2000). Two histone fold proteins, CHRAC-14 and CHRAC-16, are developmentally regulated subunits of chromatin accessibility complex (CHRAC). EMBO Journal 19, 3049–3059. Corona, D. F., Langst, G., Clapier, C. R., Bonte, E. J., Ferrari, S., Tamkun, J. W. and Becker, P. B. (1999). ISWI is an ATP-dependent nucleosome remodeling factor. Molecular Cell 3, 239–245. Corpet, F., Servant, F., Gouzy, J. and Kahn, D. (2000). ProDom and ProDom-CG: tools for protein domain analysis and whole genome comparisons. Nucleic Acids Research 28, 267–269. Cote, J., Quinn, J., Workman, J. L. and Peterson, C. L. (1994). Stimulation of GAL4 derivative binding to nucleosomal DNA by the yeast SWI/SNF complex. Science 265, 53–60. Crosby, M. A., Miller, C., Alon, T., Watson, K. L., Verrijzer, C. P., Goldman-Levi, R. and Zak. N. B. (1999). The trithorax group gene moira encodes a brahma-associated putative chromatin-remodeling factor in Drosophila melanogaster. Molecular and Cellular Biology 19, 1159–1170. Eberharter, A., Ferrari, S., Langst, G., Straub, T., Imhof, A., Varga-Weisz, P., Wilm, M. and Becker, P. B. (2001). Acf1, the largest subunit of CHRAC, regulates ISWI-induced nucleosome remodelling. EMBO Journal 20, 3781–3788.

136

J. BRZESKI ET AL.

Eisen, J. A, Sweder, K. S. and Hanawalt, P. C. (1995). Evolution of the SNF2 family of proteins: subfamilies with distinct sequences and functions. Nucleic Acids Research 23, 2715–2723. Fan, Y., Sirotkin, A., Russell, R. G., Ayala, J. and Skoultchi, A. I. (2001). Individual somatic H1 subtypes are dispensable for mouse development even in mice lacking the H1(0) replacement subtype. Molecular and Cellular Biology 21, 7933–7943. Felsenstein, J. 1993. PHYLIP (Phylogeny Inference Package) version 3.5c. Distributed by the author. Department of Genetics, University of Washington, Seattle. Feng, Q. and Zhang, Y. (2001). The MeCP1 complex represses transcription through preferential binding, remodeling, and deacetylating methylated nucleosomes. Genes and Development 15, 827–832. Finnegan, E. J., Peacock, W. J. and Dennis, E. S. (1996). Reduced DNA methylation in Arabidopsis thaliana results in abnormal plant development. Proceedings of the National Academy of Sciences USA 93, 8449–8454. Finnegan, E. J. and Kovac, K. A. (2000). Plant DNA methyltransferases. Plant Molecular Biology 43, 189–201. Forsberg, E. C. and Bresnick, E. H. (2001). Histone acetylation beyond promoters: long-range acetylation patterns in the chromatin world. BioEssays 23, 820–830. Fransz, P. F. and de Jong, J. H. (2002). Chromatin dynamics in plants. Current Opinion in Plant Biology 5, 560–567. Gdula, D. A., Sandaltzopoulos, R., Tsukiyama, T., Ossipow, V. and Wu, C. (1998). Inorganic pyrophosphatase is a component of the Drosophila nucleosome remodeling factor complex. Genes and Development 12, 3206–3216. Gendrel, A. V., Lippman, Z., Yordan, C., Colot, V. and Martienssen, R. A. (2002). Dependence of heterochromatic histone H3 methylation patterns on the Arabidopsis gene DDM1. Science 297, 1871–1873. Geng, F., Cao, Y. and Laurent, B. C. (2001). Essential roles of Snf5p in Snf-Swi chromatin remodeling in vivo. Molecular and Cellular Biology 21, 4311–4320. Genger, R. K., Kovac, K. A., Dennis, E. S., Peacock, W. J. and Finnegan, E. J. (1999). Multiple DNA methyltransferase genes in Arabidopsis thaliana. Plant Molecular Biology 41, 269–278. Guschin, D., Geiman, T. M., Kikyo, N., Tremethick, D. J., Wolffe, A. P. and Wade, P. A. (2000). Multiple ISWI ATPase complexes from xenopus laevis. Functional conservation of an ACF/CHRAC homolog. Journal of Biological Chemistry 275, 35248–35255. Hansen, J. C. (2002). Conformational dynamics of the chromatin fiber in solution: determinants, mechanisms, and functions. Annual Review of Biophysics & Biomol. Structure 31, 361–392. Horn, P. J., Carruthers, L. M., Logie, C., Hill, D. A., Solomon, M. J., Wade, P. A., Imbalzano, A.N., Hansen, J.C. and Peterson, C.L. (2002). Phosphorylation of linker histones regulates ATP-dependent chromatin remodeling enzymes. Nature Structural Biology 9, 263–267. Hudson, B. P., Martinez-Yamout, M. A., Dyson, H. J. and Wright, P. E. (2000). Solution structure and acetyl-lysine binding activity of the GCN5 bromodomain. Journal of Molecular Biology 304, 355–370. Ito, T., Bulger, M., Pazin, M. J., Kobayashi, R. and Kadonaga, J. T. (1997). ACF, an ISWI-containing and ATP-utilizing chromatin assembly and remodeling factor. Cell 90, 145–155. Ito, T., Levenstein, M. E., Fyodorov, D. V., Kutach, A. K., Kobayashi, R. and Kadonaga, J.T. (1999). ACF consists of two subunits, Acf1 and ISWI, that

PLANT CHROMATIN

137

function cooperatively in the ATP-dependent catalysis of chromatin assembly. Genes and Development 13, 1529–1539. Jackson, J. P., Lindroth, A. M., Cao, X. and Jacobsen, S. E. (2002). Control of CpNpG DNA methylation by the KRYPTONITE histone H3 methyltransferase. Nature 416, 556–560. Jeddeloh, J. A., Stokes, T. L. and Richards, E. J. (1999a). Maintenance of genomic methylation requires a SWI2/SNF2-like protein. Nature Genetics 22, 94–97. Jeddeloh, J. A., Bender, J. and Richards, E. J. (1999b). The DNA methylation locus DDM1 is required for maintenance of gene silencing in Arabidopsis. Genes and Development 12, 1714–1725. Jedrusik, M. A. and Schulze, E. (2001). A single histone H1 isoform (H1.1) is essential for chromatin silencing and germline development in Caenorhabditis elegans. Development 128, 1069–1080. Jenuwein, T. (2001). Re-SET-ting heterochromatin by histone methyltransferases. Trends in Cell Biology 11, 266–273. Jerzmanowski, A. and Maleszewski, M. (1985). Phosphorylation and methylation of Physarum histone H1 during mitotic cycle. Biochemistry 24, 2360–2367. Jerzmanowski, A., Przewloka, M. and Grasser, K. D. Linker histones and HMG1 proteins of higher plants. (2000). Plant Biology 2, 586–597. Kakutani, T., Jeddeloh, J. A., Flowers, S. K., Munakata, K. and Richards. E.J. (1996). Developmental abnormalities and epimutations associated with DNA hypomethylation mutations. Proceedings of the National Academy of Sciences USA 93, 12406–12411. Kal, A. J., Mahmoudi, T., Zak, N. B. and Verrijzer, C. P. (2000). The Drosophila brahma complex is an essential coactivator for the trithorax group protein zeste. Genes and Development 14, 1058–1071. Khavari, P. A., Peterson, C. L., Tamkun, J. W. and Crabtree, G. R. (1993). BRG1 contains a conserved domain of the SWI2/SNF2 family necessary for normal mitotic growth and transcription. Nature 366, 170–174. Khochbin, S., Verdel, A., Lemercier, C. and Seigneurin-Berny, D. (2001). Functional significance of histone deacetylase diversity. Current Opinion in Genetics & Development 11, 162–166. Krebs, J. E., Kuo, M. H., Allis, C. D. and Peterson, C. L. (1999). Cell-cycle regulated histone acetylation required for expression of the yeast HO gene. Genes and Development 13, 1412–1421. Lachner, M., Jenuwein, T. (2002). The many faces of histone lysine methylation. Current Opinion in Cell Biology 14, 286–298. Langst, G., Bonte, E. J., Corona, D. F. and Becker, P. B. (1999). Nucleosome movement by CHRAC and ISWI without disruption or trans-displacement of the histone octamer. Cell 97, 843–852. Li, G., Hall, T. C. and Holmes-Davis, R. (2002). Plant chromatin: development and gene control. BioEssays 24, 234–243. Lindroth, A. M., Cao, X., Jackson, J. P., Zilberman, D., McCallum, C. M., Henikoff, S. and Jacobsen, S. E. (2001). Requirement of CHROMOMETHYLASE3 for maintenance of CpXpG methylation. Science 292, 2077–2080. Litt, M. D., Simpson, M., Gaszner, M., Allis, C. D. and Felsenfeld. G. (2001). Correlation between histone lysine methylation and developmental changes at the chicken beta-globin locus. Science 293, 2453–2455. Martinez-Balbas, M. A., Tsukiyama, T., Gdula, D. and Wu, C. (1998). Drosophila NURF-55, a WD repeat protein involved in histone metabolism. Proceedings of the National Academy of Sciences USA 95, 132–137. Meyer, P. (2001). Chromatin remodelling. Current Opinion in Plant Biology 4, 457–462.

138

J. BRZESKI ET AL.

Meyerowitz, E. M. (2002). Plants compared to animals: the broadest comparative study of development. Science 22, 1482–1485. Mizuguchi, G., Tsukiyama, T., Wisniewski, J. and Wu, C. (1997). Role of nucleosome remodeling factor NURF in transcriptional activation of chromatin. Molecular Cell 1, 141–150. Morel, J. B., Mourrain, P., Beclin, C. and Vaucheret, H. (2000). DNA methylation and chromatin structure affect transcriptional and post-transcriptional transgene silencing in Arabidopsis. Current Biology 10, 1591–1594. Murchardt, C., Sardet, C., Bourachot, B., Onufryk, C. and Yaniv, M. (1995). A human protein with homology to Saccharomyces cerevisiae SNF5 interacts with the potential helicase hbrm. Nucleic Acids Research 23, 1127–1132. Murfett, J., Wang, X. J., Hagen, G. and Guilfoyle, T. J. (2001). Identification of Arabidopsis histone deacetylase HDA6 mutants that affect transgene expression. The Plant Cell 13, 1047–1061. Murray, K. (1964). The occurrence of "-N-methyl lysine in histones. Biochemistry 3, 10–15. Ng, H. H., Zhang, Y., Hendrich, B., Johnson, C. A., Turner, B. M., ErdjumentBromage, H., Tempst, P., Reinberg, D. and Bird, A. (1999). MBD2 is a transcriptional repressor belonging to the MeCP1 histone deacetylase complex. Nature Genetics 23, 58–61. Nielsen, A. L., Oulad-Abdelghani, M., Ortiz, J. A., Remboutsika, E., Chambon, P. and Losson, R. (2001) Heterochromatin formation in mammalian cells: interaction between histones and HP1 proteins. Molecular Cell 7, 729–739. Noma, K., Allis, C. D. and Grewal, S. I. (2001). Transitions in distinct histone H3 methylation patterns at the heterochromatin domain boundaries. Science 293, 1150–1155. Oakley, E. J., Podesta, A. and Jost, J-P. (1997). Developmental changes in DNA methylation of the two tobacco pollen nuclei during maturation. Proceedings of the National Academy of Sciences USA 94, 11721–11725. Ogas, J., Kaufmann, S., Henderson, J. and Somerville, C. (1999). PICKLE is a CHD3 chromatin-remodeling factor that regulates the transition from embryonic to vegetative development in Arabidopsis. Proceedings of the National Academy of Sciences USA 96, 13839–13844. Pandey, R., Muller, A., Napoli, C. A., Selinger, D. A., Pikaard, C. S., Richards, E. J., Bender. J., Mount, D. W. and Jorgensen, R. A. (2002). Analysis of histone acetyltransferase and histone deacetylase families of Arabidopsis thaliana suggests functional diversification of chromatin modification among multicellular eukaryotes. Nucleic Acids Research 30, 5036–5055. Parekh, B. S. and Maniatis. T. (1999). Virus infection leads to localized hyperacetylation of histones H3 and H4 at the IFN-beta promoter. Molecular Cell 3, 125–129. Phelan, M. L., Schnitzler, G. R. and Kingston, R. E. (2000). Octamer transfer and creation of stably remodeled nucleosomes by human SWI-SNF and its isolated ATPases. Molecular and Cellular Biology 20, 6380–6389. Phelan, M. L., Sif, S., Narlikar, G. J. and Kingston, R. E. (1999). Reconstitution of a core chromatin remodeling complex from SWI/SNF subunits. Molecular Cell 3, 247–253. Pollard, K. J., Samuels, M. L., Crowley, K. A., Hansen, J. C. and Peterson, C. L. (1999). Functional interaction between GCN5 and polyamines: a new role for core histone acetylation. EMBO Journal 18, 5622–5633. Poot, R. A., Dellaire, G., Hulsmann, B. B., Grimaldi, M. A., Corona, D. F., Becker, P. B., Bickmore, W. A. and Varga-Weisz, P. D. (2000). HuCHRAC,

PLANT CHROMATIN

139

a human ISWI chromatin remodelling complex contains hACF1 and two novel histone-fold proteins. EMBO Journal 19, 3377–3387. Prymakowska-Bosak, M., Przewloka, M. R., Slusarczyk, J., Kuras, M., Lichota, J., Kilianczyk, B. and Jerzmanowski, A. (1999). Linker histones play a role in male meiosis and the development of pollen grains in tobacco. The Plant Cell 11, 2317–2329. Przewloka, M. R., Wierzbicki, A. T., Slusarczyk, J., Kuras, M., Grasser, K. D., Stemmer, C. and Jerzmanowski, A. (2002). The ‘drought-inducible’ histone H1s of tobacco play no role in male sterility linked to alterations in H1 variants. Planta 215, 371–379. Ramon, A., Muro-Pastor, M. I., Scazzocchio, C. and Gonzalez, R. (2000). Deletion of a unique gene encoding a typical histone H1 has no apparent phenotype in Aspergillus nidulans. Molecular Microbiology 35, 223–233. Reyes, J. C., Hennig, L. and Gruissem, W. (2002). Chromatin-remodeling and memory factors. New regulators of plant development. Plant Physiology 130, 1090–1101. Rice, J. C. and Allis, C. D. (2001). Histone methylation versus histone acetylation: new insights into epigenetic regulation. Current Opinion in Cell Biology 13, 263–273. Ronemus, M. J., Galbiati, M., Ticknor, C., Chen, J. and Dellaporta, S. L. (1996). Demethylation-induced developmental pleiotropy in Arabidopsis. Science 273, 654–657. Roth, S. Y., Denu, J. M., Allis, C. D. (2001). Histone acetyltransferases. Annual Review of Biochemistry 70, 81–120. Ruhf, M. L., Braun, A., Papoulas, O., Tamkun, J. W., Randsholt, N. and Meister, M. (2001). The domino gene of Drosophila encodes novel members of the SWI2/SNF2 family of DNA-dependent ATPases, which contribute to the silencing of homeotic genes. Development 128, 1429–1441. Sarnowski, T. J., Swiezewski, S., Pawlikowska, K., Kaczanowski, S. and Jerzmanowski, A. (2002). AtSWI3B, an Arabidopsis homolog of SWI3, a core subunit of yeast Swi/Snf chromatin remodeling complex, interacts with FCA, a regulator of flowering time. Nucleic Acids Research 30, 3412–3421. Sawa., H., Kouike, H. and Okano, H. (2000). Components of the SWI/SNF complex are required for asymmetric cell division in C. elegans. Molecular Cell 6, 617–624. Schu¨beler, D., Groudine, M. and Bender, M. A. (2001). The murine beta-globin locus control region regulates the rate of transcription but not the hyperacetylation of histones at the active genes. Science 293, 2453–2455. Shen, X., Mizuguchi, G., Hamiche, A. and Wu, C. (2000). A chromatin remodelling complex involved in transcription and DNA processing. Nature 406, 541–544. Shen, X., Yu, l., Weir, J. W. and Gorovsky, M. A. (1995). Linker histones are not essential and affect chromatin condensation in vivo. Cell 82, 47–56. Sif, S., Saurin, A. J, Imbalzano, A. N. and Kingston, R. E. (2001). Purification and characterization of mSin3A-containing Brg1 and hBrm chromatin remodeling complexes. Genes and Development 15, 603–618. Soppe, W. J., Jasencakova, Z., Houben, A., Kakutani, T., Meister, A., Huang, M. S., Jacobsen, S.E., Schubert, I., Fransz, P.F. (2002). DNA methylation controls histone H3 lysine 9 methylation and heterochromatin assembly in Arabidopsis. EMBO Journal 21, 6549–6559. Stallcup, M. R. (2001). Role of protein methylation in chromatin remodeling and transcriptional regulation. Oncogene 20, 3014–3020.

140

J. BRZESKI ET AL.

Steward, N., Ito, M., Yamaguchi, Y., Koizumi, N. and Sano, H. (2002). Periodic DNA methylation in maize nucleosomes and demethylation by environmental stress. Journal of Biological Chemistry 277, 37741–37446. Sudarsanam. P. and Winston, F. (2000). The Swi/Snf family of nucleosomeremodeling complexes and transcriptional control. Trends in Genetics 16, 345–351. Suka, N., Suka, Y., Carmen, A. A., Wu, J. and Grunstein. M. (2001). Highly specific antibodies determine histone acetylation site usage in yeast heterochromatin and euchromatin. Molecular Cell 8, 473–479. Taunton, J., Hassig, C. A. and Schreiber, S. L. (1996). A mammalian histone deacetylase related to the yeast transcriptional regulator Rpd3p. Science 272, 408–411. Tian, L. and Chen, Z. J. (2001). Blocking histone deacetylation in Arabidopsis induces pleiotropic effects on plant gene regulation and development. Proceedings of the National Academy of Sciences USA 98, 200–205. Treich, I., Ho, L. and Carlson, M. (1998). Direct interaction between Rsc6 and Rsc8/ Swh3, two proteins that are conserved in SWI/SNF-related complexes. Nucleic Acids Research 26, 3739–3745. Tse, C., Sera, T., Wolffe, A. P. and Hansen, J. C. (1998). Disruption of higher-order folding by core histones acetylation dramatically enhances transcription of nucleosomal arrays by RNA polymerase III. Molecular and Cellular Biology 18, 4629–4638. Tsukiyama, T. and Wu, C. (1995). Purification and properties of an ATP-dependent nucleosome remodeling factor. Cell 83, 1011–1020. Tsukiyama, T., Palmer, J., Landel, C. C., Shiloach, J. and Wu, C. (1999). Characterization of the imitation switch subfamily of ATP-dependent chromatin-remodeling factors in Saccharomyces cerevisiae. Genes and Development 13, 686–697. Varga-Weisz, P. D., Wilm, M., Bonte, E., Dumas, K., Mann, M. and Becker, P. B. (1997). Chromatin-remodelling factor CHRAC contains the ATPases ISWI and topoisomerase II. Nature 388, 598–602. Verbsky, M. L. and Richards, E. J. (2001). Chromatin remodeling in plants. Current Opinion in Plant Biology 4, 494–500. Vongs, A., Kakutani, T., Martienssen, R. A. and Richards, E. J. (1993). Arabidopsis thaliana DNA methylation mutants. Science 260, 1926–1928. Wagner, D. and Meyerowitz, E. M. (2002). SPLAYED, a novel SWI/SNF ATPase homolog, controls reproductive development in Arabidopsis. Current Biology 12, 85–94. Wang, W., Cote, J., Xue, Y., Zhou, S., Khavari, P. A., Biggar, S. R., Muchardt, C., Kalpana, G. V., Goff, S. P., Yaniv, M., Workman, J. L. and Crabtree, G. R. (1996). Purification and biochemical heterogenity of mammalian SWISNF complex. EMBO Journal 15, 5370–5382. Waterborg, J. H. (1990). Sequence analysis of acetylation and methylation in two histone H3 variants of alfalfa. Journal of Biological Chemistry 265, 17157–17161. Waterborg, J. H. (1992). Identification of five sites of acetylation in alfalafa histone H4. Biochemistry 31, 6211–6219. Weigel, D. and Ju¨rgens, G. (2002). Stem cells that make stems. Nature 415, 751–754. Xiao, H., Sandaltzopoulos, R., Wang, H. M., Hamiche, A., Ranallo, R., Lee, K. M., Fu, D. and Wu, C. (2001). Dual functions of largest NURF subunit NURF301 in nucleosome sliding and transcription factor interactions. Molecular Cell 8, 531–543.

PLANT CHROMATIN

141

Xue, Y., Wong, J., Moreno, G. T., Young, M. K., Cote, J. and Wang, W. (1998). NURD, a novel complex with both ATP-dependent chromatin-remodeling and histone deacetylase activities. Molecular Cell 2, 851–861. Zhang, Y., Ng, H. H., Erdjument-Bromage, H., Tempst, P., Bird, A. and Reinberg, D. (1999). Analysis of the NuRD subunits reveals a histone deacetylase core complex and a connection with DNA methylation. Genes and Development 13, 1924–1935.

This Page Intentionally Left Blank

The Interface Between the Cell Cycle and Programmed Cell Death in Higher Plants: from Division unto Death

DENNIS FRANCIS

School of Biosciences, Cardiff University, PO Box 915, Cardiff CF10 3TL, UK

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. The Cell Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Programmed Cell Death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Cell Cycle Checkpoints and the Route to PCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Phosphoregulation of the Cell Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. The Checkpoints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Plant Cell Cycle Checkpoints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Do Plants Require Cell Cycle Checkpoints? . . . . . . . . . . . . . . . . . . . . . . . III. Instances of PCD in Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Aerenchyma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. PCD in Cell Cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Xylogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. PCD in the Endosperm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Oxidoreductive States and PCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Fungal Elicitors of PCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Advances in Botanical Research Vol. 40 incorporating Advances in Plant Pathology ISBN 0-12-005940-1

144 145 147 148 149 152 157 158 160 161 161 164 167 168 169 170 171 172

Copyright  2003 Elsevier Ltd All rights of reproduction in any form reserved

144

D. FRANCIS

ABSTRACT In this article, interfaces between the cell cycle and programmed cell death (PCD) have been reviewed both in planta and in cells in culture. In animals, the most obvious interface between these two processes are the cell cycle checkpoints, surveillance systems that check the integrity of DNA and block the cell cycle whilst damage is being repaired or replication is normalised. Catastrophic DNA damage results in cell cycle exit into PCD. The extent to which cell cycle checkpoints may be functioning in the plant cell cycle has been explored, and a model of a plant cell cycle checkpoint is presented based on evidence available from animal systems and based on the presence, in Arabidopsis, of key components of these checkpoints: ATM, a sensor of DNA damage, WEE1, a negative regulator of cell division and 14-3-3 proteins; the latter protect phosphorylated proteins in cell cycle checkpoints. Both environmental and chemically induced PCD (e.g. UVB and ethylene) have been reviewed and evidence of cell cycle specific exit into PCD is presented. Instances of PCD in the plant have also been examined (e.g. aerenchyma, endosperm and xylogenesis). In each case, the extent to which a cell cycle checkpoint could be operating has been examined. Finally, oxidoreductive states of cells in tissues and the effect of fungal elicitors of PCD has been surveyed again in relation to cell cycle – PCD interfaces. A somewhat weakly coloured picture emerges for this interface in plants compared with animals. However, sufficient evidence exists to suggest that cross-talk between the cell cycle and PCD is an important component of life unto death of the plant cell.

I. INTRODUCTION Arguably, the cell cycle and programmed cell death (PCD) have been the hottest topics in cell biology for the past 10 years (e.g. Nurse, 2001). This is particularly true in mammalian cell biology but there has also have been considerable advances in our understanding of both the plant cell cycle and plant cell death (Inze, 2000; Lam et al., 2000). Clearly in plants there are several instances where PCD is irrevocably the end-product of a differentiation pathway (e.g. xylogenesis, see Fukuda, 2000) whilst there are fewer instances where it functions to rid a tissue or an organ of unwanted cells. This seems to be the converse of animal PCD where it serves as a clearing house to remove mutagenised cells from germ layers (Raff, 1992). That is not to say that there aren’t various instances of developmentally regulated PCD in mammals; finger and toe formation from webbed digits during embryology immediately spring to mind (Jacobson et al., 1997). To encapsulate all discoveries about the cell cycle, and PCD in one review would be overly ambitious. Instead, the intention in this review is to consider the extent to which these processes ‘talk’ to each other and to highlight interfaces where pivotal genes link together. Note that necrosis, another form of cell death, is probably not regulated but is a function of massive nutrient loss together with a general winding down of all living

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

145

processes. Whilst it is clearly an important component of the plant life cycle, it is not central to the type of cross-talk sought here. In animals, key interfaces between the cell cycle and PCD are the cell cycle checkpoints, molecular networks that prevent cell division when DNA is damaged or DNA replication is perturbed. Genes that constitute checkpoints in fission yeast and animals will be reviewed and against this backdrop, a model of plant cell cycle checkpoints will be proposed. Euphemistically, the review may be considered as ‘cell division unto programmed cell death’. Other than Drosophila genes that are often named topically (e.g. the PCD genes GRIM (Chen et al., 1996) and REAPER (White et al., 1996)), the conventions of Arabidopsis genetics, GENE (first three letters italicised) and PROTEIN (first three letters in normal font) will be adhered to. However, confusingly, fission yeast (Schizosaccharomyces pombe) cell cycle genes encompass a unique nomenclature whereby genes are referred to in lower case italics (e.g. cdc2) and the protein in normal font but beginning with a capital letter (e.g. Cdc2). This convention is necessary because the budding yeast gene, for example, CDC2, is not the homologue to fission yeast cdc2.

A. THE CELL CYCLE

During a cell cycle, a competent proliferative cell replicates its chromosomes and gives rise to offspring that have identical sets of chromatids. Chromosome replication occurs during S-phase (DNA synthetic phase) and chromosome partitioning during M phase (mitosis). In animals, early embryonic events are rapid; chromosomes replicate and partition in the so-called true cell cycle comprising M and S (Fig. 1) (Cross et al., 1989). Indeed in Drosophila replication and partitioning can occur in cell cycles that last a mere 10 min (Edgar and O’Farrell, 1989, 1990). Only later in development is there the interpolation of presynthetic interphase (G1) and premitotic interphase, (G2) with the inevitable slowing down of the cell cycle (Fig. 1). For example in Drosophila early embryogenesis comprises 13 synchronous nuclear divisions (solely M and S). The 14th division occurs with the concomitant interpolation of G1 and G2 and the zygotic transcription of regulatory genes of the cell cycle (Edgar and O’Farrell, 1989, 1990). One such gene that falls neatly into this scheme is STRING, a Drosophila homologue of the fission yeast cell division cycle (cdc) gene, cdc25. This gene is transcribed in late interphase (G2) and therefore, it is not transcribed until the 14th division; prior to this, G2 does not exist and all STRING mRNA is of maternal origin (Edgar and O’Farell, 1990). However

146

D. FRANCIS

Fig. 1. The embryonic, or true cell cycle, comprising Mitotic (M) and DNA synthetic (S) phases, which converts to the post-embryonic cell cycle through the additions of presynthetic interphase (G1) and premitotic interphase (G2) (based on Cross et al., 1989).

from the 14th division on, the expression of zygotic STRING begins but is under a spatio-temporal control whereby patterning genes cue the expression of cell cycle genes (O’Farrell, 2001). Hence the classical M, G1, S, G2 cell cycle follows on from the true cell cycle in animals and perhaps this also occurs in plants. Unfortunately nothing is known about cell cycles during early plant embryogenesis and one can only postulate about comparable free running M-S cell cycles that subsequently lengthen due to the interpolation of G1 and G2. During its passage through the cell cycle, the cell is subject to intricate surveillance mechanisms that assess its competence for the various hurdles en-route. They are more commonly referred to as checkpoints. Hartwell and Weinert (1989) defined a checkpoint empirically: when step B is dependent on the completion of step A, that dependence is governed by a checkpoint unless a loss-of-function mutation exists that relieves the dependence. For example, in budding yeast, the transition from G2 to mitosis is dependent on intact DNA but this is abolished by deletion of the RAD9 gene (radiation damage). Clearly, the RAD9 protein is a functional part of the checkpoint that monitors DNA integrity (Paulovich et al., 1997). In general terms, when cells are stressed checkpoints must be satisfied in order that cell cycle transitions occur. However, as noted by O’Connell et al. (1997), proteins

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

147

that make up these checkpoints are non-essential and only alter cell cycle transitions under stress conditions. There are two major checkpoints of the cell cycle operating in S-phase (S/M), and at G2/M. In budding yeast, G2 phase is lacking and some of the machinery of mitosis is synthesised before S-phase has ended (e.g. spindle formation). Hence in the budding yeast system, the DNA damage checkpoint operates by blocking the metaphase– anaphase transition and is often referred to as the spindle checkpoint (Yamamoto et al., 1996). In human cells, DNA damage during mitosis delays metaphase/anaphase in a spindle-assembly checkpoint (Mikhailov et al., 2002). So the cell cycle is a mechanism for partitioning and replication of chromosomes with an intricate surveillance mechanism lying in wait to determine whether or not these major processes should occur; more of this later.

B. PROGRAMMED CELL DEATH

Programmed cell death has been defined in animal systems as a type of cell death, which is part of an organism’s life cycle, is initiated by specific physiological signals, and requires de novo transcription (Ellis et al., 1991). Another term, apoptosis, often used interchangeably with PCD, was originally used to define particular features of cells undergoing PCD. Ultrastructural and biochemical changes include chromatin condensation, membrane blebbing, and DNA laddering (Kerr et al., 1972). However, whether all PCD has apoptotic features is questionable because some animal cells undergo PCD physiologically but without any of the characteristic apoptotic features (Schwartz et al., 1993). In this review, the terms will be used interchangeably unless there is specific information that defines particular types of dying plant cells. Changes characteristic of PCD in plants include: progressive and sometimes rapid degeneration of organelles, proteolysis, protoplast autolysis and DNA laddering that ultimately leads to nuclear degradation. Whether such features can be described as diagnostic of PCD in plants is questionable because whilst they may typically occur in some tissue systems exhibiting PCD they may not occur in others. For example, and as noted by Jones and Dangl (1996), whilst DNA laddering has been observed during the PCD of tracheary elements (Mittler et al., 1997), it is not always an observable feature of plant cells undergoing PCD (Buckner et al., 2000; Herbert et al., 2001). One biochemical marker of PCD in plants is cysteine protease; a peak of activity occurs at the onset of cytoplasmic loss in TEs undergoing PCD (Fukuda, 2000) and such activity increases dramatically during leaf senescence (Smart, 1994). However, once

148

D. FRANCIS

again, the extent to which the activity of a particular enzyme can be used as a universal marker of PCD continues to be hotly debated. There is an ever-growing number of factors linked with PCD including inositol triphosphate, ceramides, Ca2þ fluxes, reactive oxygen species (ROS) and suppressors (e.g. BC12) or activators (e.g. Bax) of PCD (Stewart, 1994; Steller, 1995). In particular, hydrogen peroxide is a key signalling molecule during the so-called hypersensitive reaction (HR) following pathogen invasion (Neill et al., 2002). Hence the route to PCD includes classical transcellular calcium movement and phosphoregulation of protein kinases acting in series to deliver signals that activate transcription factors. Most of these regulators of PCD will be discussed in this review although not exhaustively but more where they offer clues about the checkpoint interface between the cell cycle and cell death. In the nematode, Caenorhabditis elegans, PCD is regulated by ced-3, the homologue of the mammalian protease gene, ICE (e.g. Raff, 1992). However, this should be regarded as the very tip of an enormous body of literature on animal PCD that will not be elaborated upon here. To the author’s knowledge, homologues of the ced-type genes do not exist in the Arabidopsis genome leading Lam et al. (2000) to conclude that plants probably enact PCD in a remarkably different way to animal cells.

II. CELL CYCLE CHECKPOINTS AND THE ROUTE TO PCD Most of what we know about cell cycle checkpoints stems from studies with yeasts and animals. Rather than a tangential walk among known genes and their function in yeasts and animals per se, the aim will be to consider the extent to which knowledge from these systems can be applied to plant systems. For vertebrate cells stressed by DNA damaging agents, their successful progression to mitosis depends on compliance with DNA replication (S/M) and DNA damage checkpoints (G2/M) (see Rhind and Russell, 1998, 2000). In both checkpoints, a gamut of genes forms an intricate surveillance mechanism that checks the fidelity of replicated DNA. The latter is proof read for the presence of either single or double-strand breaks. DNAdependent protein kinases are in the forefront of this proof reading procedure and, in turn, provide the cue for the up-regulation of DNAdependent repair enzymes. Repairing DNA takes time and whilst this occurs, G2 lengthens and the cell is delayed from entering mitosis (Rhind and Russell, 1998, 2000). If the damage is catastrophic, or, is liable to

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

149

propagate harmful errors, then another set of machinery orchestrated by p53, is induced to steer the doomed cell into PCD (Kastan et al., 1991; Kuerbitz et al., 1992; Lane, 1992). The mechanism for this in vertebrates is by phosphoregulation of cell cycle proteins. However, in order to understand the nature of these checkpoints, some background knowledge is necessary about phosphoregulation of cell cycle proteins during the normal transition from G2 to mitosis. A. PHOSPHOREGULATION OF THE CELL CYCLE

The central focus of entry into mitosis is cdc2, a gene first discovered in fission yeast that is absolutely essential for both the G2/M and G1/S transitions (Nurse and Bissett, 1981). It encodes a protein kinase that phosphorylates several substrates in order that the cell can proceed into mitosis (Simanis and Nurse, 1986; Nurse, 1990). In plants, there is a plethora of cdc2 homologues mostly referred to as cyclin-dependent kinase (CDK) genes (Francis and Inze, 2001). During the plant cell cycle different CDKs are expressed at different times. G1/S and S-phase are characterised by the expression of CDKA (Joubes et al., 2000). The G2/M transition is characterised by the expression of CDKA, CDKB and CDKF in series (Fig. 2), the latter being expressed in mitosis itself (Joubes et al., 2000; Me´sza´ros et al., 2000). CDKs that function in late G2 are most likely subject

Fig. 2. Simplified model of major transitions of the plant cell cycle: G1/S, dependent on a cytokinin/sucrose-mediated up-regulation of D-type cyclins. CDKA binds to CycD and phosphorylates ( ) the retinoblastoma (Rb) protein which in an unphosphorylated form, represses the transcription factor, E2F. Hyperphosphorylated Rb detaches from E2F enabling the latter to facilitate the transcription of genes necessary for the G1/S transition. G2/M, dependent on the expression of CDKA and CDKB in series and CDKF in mitosis. A cytokinin signal leads to the dephosphorylation of CDKB enabling the latter to drive cells into mitosis (model based on John, 1996; Murray et al., 1998; Me´sza´ros et al., 2000).

150

D. FRANCIS

to regulation by cytokinins acting on a CDC25-like phosphatase (Zhang et al., 1996). Detailed attempts have been made to clarify the numerous plant CDKs. For example, the Arabidopsis CDK expressed in late G2 is known as Arath; CDKB1;1 (e.g. Joubes et al., 2000). The classification of these genes is becoming complex because there seem to be more variants of CDC2 in plants than in animals and it is not clear what they are all doing. The abovementioned gene will be simply referred to as AtCDKB and so on but where a generic explanation seems more valid it will be referred to as CDC2. As mentioned above (Fig. 2), and unlike animal systems, at least three CDKs are expressed in series in late G2/M of the plant cell cycle. Using Medicago sativa (alfalfa) as a model culture system, Dudits and his colleagues have isolated three cdc2 orthologues and their encoded protein kinases exhibit different patterns of kinase activity during the cell cycle. Cdc2MS A and B peak in kinase activity in S-phase and G2, respectively. Cdc2MsD peaks in late G2 while Cdc2MsF peaks in mitosis. To the author’s knowledge, the latter is a unique expression pattern for a cyclindependent kinase in the plant cell cycle and perhaps owes its uniqueness to the presence of plant specific substrates such as the mitotic machinery in the plant cell. In support of this, immunolocalisation studies show that Cdc2MsF is localised in the preprophase band, the prophase/metaphase spindle and the phragmoplast (Me´sza´ros et al., 2000). In Arabidopsis, CDKA;1 is also expressed at G1/S-phase. It sits among a cascade of reactions (Fig. 2) involving a cytokinin/sucrose-mediated induction of plant D-type cyclins, the retinoblastoma(Rb) protein and the transcription factor, E2F (more of this interaction later) (see Murray et al., 2000). Another example of regulation by plant growth factors, is the cyclindependent kinase inhibitors (CKIs in animals and ICKs in plants), proteins that mask the entire ATP-binding domain of Cdc2 kinase. In Arabidopsis, ICK1 and 2 are up-regulated by abscisic acid, blocking CDK activity and preventing cell division (Wang et al., 1997; Zhou et al., 2002). Dependency of plant cell cycle proteins on plant growth regulator signalling (see Francis and Sorrell, 2001) is worth emphasising in relation to interfaces between the cell cycle and PCD in plants. There may well be some surprising and unusual proteins involved in plant cell cycle checkpoints compared with yeast and animal ones because they have evolved differently and are uniquely responsive to plant growth regulator signalling. Perhaps not surprisingly, G2/M, a major transition point of the cell cycle, involves a number of negative controls that stop CDC2 kinase exhibiting catalytic activity until conditions are ideal for mitosis. In vertebrates, at least three protein kinases negatively regulate CDC2 through the addition of a

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

151

phosphate group(s) at amino acid residues that border an ATP-binding domain of the CDC2 kinase (Fig. 3). WEE1 and MIK1 are nuclear-located and phosphorylate tyrosine15 of CDC2; they show functional redundancy in fission yeast (Lundgren et al., 1991). MYT1, a cytoplasmically located kinase related to WEE1, is a dual inhibitory kinase phosphorylating threonine14 and tyrosine15 (Mueller et al., 1995). In plants, a homologue to WEE1 has been cloned in maize (ZmWEE1, Sun et al., 1999), Arabidopsis (AtWEE1, Sorrell et al., 2002) and partially in tobacco (NtWEE1, I. Siciliano, H.J. Rogers and D. Francis, unpublished data) but, currently, there are no plant homologues to MIK1 or MYT1. Much of the control of the cell cycle is exerted through phosphoregulation (see John, 1996). To become activated, CDC2 kinase (AtCDKA or AtCDKB) must bind to an essential, albeit non-catalytic regulatory partner, cyclinB (Joubes et al., 2000). Once again, there are many plant cyclins and space does not permit detailed consideration of them (see Me´sza´ros et al., 2000). Cyclin binding, which alters the conformation of the CDC2 kinase is an essential prerequisite for the so-called T-loop to flip across the catalytic face of the protein thereby exposing an ATP-binding domain (Fig. 4) (Morgan, 1997). At the same time, a cyclin-dependent activating kinase (CAK kinase) catalyses the phosphorylation of threonine160/167 (Krek and Nigg, 1991). A CAK kinase homologue has been isolated in Arabidopsis (Umeda et al., 1998). However, O’Farrell (2001) proposed that this is not the final all-or-nothing signal that activates CDC2 because cells can accumulate cyclins in excess of the requirements for mitosis (see O’Farrell, 2001). In animals and fission yeast, a dual phosphatase, encoded by CDC25 removes the inhibitory phosphate groups from threonine14 and tyrosine15 bordering the aforementioned ATP-binding domain of CDC2 (Russell and Nurse, 1987; O’Farrell, 2001). Hence, CDC25 emerges as a positive regulator of the G2/M transition but as yet, a plant homologue has not been identified.

Fig. 3. Phosphoregulation of CDC2cyclin. Negative regulation is provided by WEE1/MIK1 that phosphorylate tyrosine15 of CDC2 and MYT1, a threonine14-tyrosine15 dual kinase. In competition for the same sites is the positive regulator, CDC25 phosphatase. CAK kinase phosphorylates CDC2 at residue threonine 161/167 facilitating binding with CYCLINB.

152

D. FRANCIS

Fig. 4. Activation of a CDK–cyclin complex. Cyclin binding to the CDK, characterised and facilitated by phosphorylation of tyrosine160/167, causes the so-called T loop of the CDK to flip across the molecule thereby exposing an ATPbinding site and enabling Cdc25 phosphatase to remove an inhibitory phosphate on tyrosine15. This is the final all-or-nothing signal that activates the CDK to drive the cell into mitosis (based on O’Farrell, 2001; Alberts et al., 2002). The cartoon is not a representation of protein conformation B. THE CHECKPOINTS

As mentioned earlier, the entry of a cell into mitosis is dependent upon faithful nuclear DNA replication. The fidelity of this replicated DNA is checked in the DNA replication and DNA damage checkpoints, which exercise control over the phosphoregulating complex that acts on Cdc2 (Hartwell and Weinert, 1989; Rhind and Russell, 2000). This is often referred to as a dependence of mitosis on DNA replication (S/M). In vertebrates, the sensor genes for DNA damage or for perturbed replication are, ATR/ATM and in fission yeast, rad3. ATR (ataxia-telangiectasia and rad3 related) is most highly expressed in testes; it encodes a protein kinase in meiotic cells (Wright et al., 1998). ATR is a member of the phosphatidylinositol 3-kinase-like family (Smith and Jackson, 1999), is nuclear-localised and is associated with unsynapsed regions of paired chromosomes (Keegan et al., 1996; Smith et al., 1999; Suzuki et al., 1999). AT is a rare inherited human disorder affecting chromosomal stability, sensitivity to ionising

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

153

radiation and the immune system (Savitsky et al., 1995). ATM kinase activity is characteristic of AT but cells from AT patients in which the ATM gene is mutated have reduced survival and are unable to activate cell cycle checkpoints following exposure to ionising radiation (Meyn, 1999). The ATM protein is the sensor for ionising radiation and the initiator of the DNA replication and damage checkpoints (Fig. 5A). An ATM homologue has been cloned in Arabidopsis (Garcia et al., 2000). ATR and ATM phosphorylate and activate downstream checkpoint kinases (Durocher and Jackson, 2001). They also form part of a complex of DNA repair enzymes at the site of either single, or double strand-breaks in nuclear DNA (Caspari and Carr, 2002). The exact composition of the checkpoints shows some interspecific variation (Fig. 5B). For example, fission yeast Rad3, a homologue to ATM, is the sensor when cells are challenged with hydroxyurea (DNA replication checkpoint) or UV radiation (DNA damage checkpoint) (Bentley et al., 1996). Another kinase, Cds1 (checking DNA synthesis), is the effector of the DNA replication checkpoint acting downstream of the Rad3 kinase (Al-Khodairy, et al., 1994). Cds1 phosphorylates Cdc25 phosphatase thereby inactivating it and at the same time, phosphorylates Mik1 thereby activating it (Rhind and Russell, 1998, 2000). In the DNA damage checkpoint, the kinase Chk1 (checkpoint kinase) is the effector phosphorylating Cdc25 and Mik1 (Baber-Furnari et al., 2000). However, an equally good case has been made for Wee1 kinase being the major inhibitory kinase that is targeted by Chk1 (O’Connell et al., 1997; Raleigh and O’Connell, 2000; Lee et al., 2001). In humans, the DNA replication checkpoint sensor is probably ATR whilst hCHK1 is the checkpoint effector (Matsuoka et al., 1998). Through phosphorylation, hCHK1 inactivates CDC25 (Sanchez et al., 1997). The human DNA damage checkpoint is more complicated because ATM is the sensor for gamma irradiation and through phosphorylation promotes the activity of hCDS1 that in turn phosphorylates p53 on serine20, at least in vitro (Shieh et al., 2000). ATR is the sensor for UV-induced DNA damage and, via CHK1, suppresses CDC25C (note in humans there are three different variants of CDC25: A, B and C). In mouse ES cells, knockouts (chk, ) were generated and when treated with UV or gamma irradiation, they escaped the DNA damage checkpoint (Hirao et al., 2000). ATR/ATM do not act alone but are part of a protein complex that senses DNA damage. In fission yeast, Rad3, the ATM homologue, binds with: Rad1, Rad9, Rad17 and Hus1 (hydroxyurea sensitive) together with Rad 26 (Durocher and Jackson, 2001). Rad17 is involved in processive DNA replication through binding to PCNA (proliferating cell nuclear antigen)

154

D. FRANCIS

Fig. 5. Cell cycle checkpoints. (A) higher eukaryote model in which ATM/ATR are sensors of perturbation of DNA replication or DNA damage inflicted by hydroxyurea (HU), ionising radiation (IR) or ultraviolet light (UV). ATM/ATR phosphorylate CHK1, the effector checkpoint kinase, that in turn phosphorylates both WEE1 (increased, or stabilised activity (")) and CDC25 (loss of activity (#)).

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

155

while Rad1, Rad9 and Hus1 form a protein complex capable of moving along the DNA molecule as a sliding clamp (Durocher and Jackson, 2001). The homologue of Rad26 in humans is known as ATRIP (ATR interacting protein) (Cortez et al., 2001). ATR/ATRIP can sense double strand breaks in DNA, although seemingly in the absence of the other proteins, a signal is not transduced to the effector CHK proteins. A consensus view is that ATR/ ATRIP are recruited to the chromatin in response to DNA damage through independent mechanisms (Cortez et al., 2001). Although ATR/ATM exists in plants, we know very little about other members of this complex, nor has a plant homologue to ATRIP been identified. Despite the differences in fine detail, the explanation of the function of the checkpoints is generic in that the activating phosphatase for Cdc2 kinase (Cdc25) is suppressed whereas the inhibitory kinase for Cdc2 (Wee1 or Mik1) is activated. The end result is that the cells are held in G2 until DNA replication recommences or until DNA repair is complete. Another noncatalytic protein, a 14-3-3/Rad24 protein is also an important part of this network (Figs. 5 and 6). 14-3-3 proteins are characterised by an ability to bind to phosphorylated serine, tyrosine residues of other proteins, including CDC25 and WEE1, thereby preventing access to these sites by other phosphatases (e.g. Blasina et al., 1997; Lee et al., 2001; Hutchins et al., 2002). Details about the mechanisms of CDC25 inactivation and WEE1/MIK1 activation are emerging constantly so that there is a stream of views and counterviews on how they operate. In vertebrates, most data indicate that once phosphorylated by CDS1/CHK1, CDC25 moves from the nucleus to the cytoplasm (Chan et al., 1999). The phosphatase is then restrained in the cytoplasm through binding to a 14-3-3-protein (in this case the  variant). Upon completion of the DNA repair, the 14-3-3 detaches from CDC25 enabling the latter to be dephosphorylated, it returns to the nucleus, and then activates CDC2 kinase so that the cell can divide (Chan et al., 1999). In fission yeast, things are different. For example, although phosphorylation of Cdc25 occurs at the DNA damage checkpoint, nuclear exclusion of

In each case the phosphorylated site is protected by a 14-3-3 protein although not necessarily the same one in each case. The overall effect is that WEE1 phosphorylates CDC2-CYLINB (decreased activity (#)) so that cells are held in G2 until DNA replication is restored (replication checkpoint) or repair is complete (damage checkpoint). (B) the equivalent fission yeast model, in which Rad3, a homologue to ATM/ATR, is the sensor, Cds1 is the effector of the HU-induced DNA replication checkpoint whilst Chk1 is the effector of the UV-induced DNA damage checkpoint. Thereafter the cartoon is essentially the same as A although in this system there is functional redundancy between Mik1 and Wee1 and hypothesised binding of Wee1/Mik1 with a 14-3-3 protein (based on Rhind and Russell, 2000).

156

D. FRANCIS

Fig. 6. Putative cell cycle checkpoints in plants. The model is based on Fig. 5A. Straight lines represent known effects whereas dotted lines are hypothetical links. Note that plant CHK1-like and CDC25-like proteins have yet to be fully identified.

this phosphatase was found to be unnecessary (Lopez-Girona et al., 2001). Less contentiously, 14-3-3 proteins are key components of checkpoints in lower and higher eukaryotes. Although the exact molecular mechanisms are unclear, data and models support the idea of a 14-3-3 protein binding to the phosphorylated form of the Cdc25 phosphatase although fewer details exist about binding between Wee1 and a 14-3-3 protein. In Xenopus laevis, CDC25C is either activated or deactivated through phosphorylation. In an auto activation feedback loop, CDC2-cyclin B and PLX1 (polo-like kinase in Xenopus) phosphorylates Cdc25 on multiple serine/threonine residues (Hutchins et al., 2002). The 14-3-3 protein prevents other protein phosphatases from gaining access to the phosphorylated residue. Indeed, in Xenopus, 14-3-3 binding masks a nuclear localisation signal on Cdc25C (Hutchins et al., 2002). A 14-3-3 also binds to the phosphorylated XWEE1. However, the effect is the exact converse; phosphorylation of XWEE1 at serine549 activates/stabilises WEE1 kinase activity which in turn negatively regulates XCDC2 (Lee et al., 2001).

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

157

C. PLANT CELL CYCLE CHECKPOINTS

As may have been noted from the above account, some of the genes of the DNA replication and damage checkpoints have been cloned in Arabidopsis: ATM, rad3 (ATR), a putative checkpoint 14-3-3 protein, and one of the targets, WEE1. What then can we say about the efficacy of a DNA damage checkpoint in plants? Not surprisingly, much less work has been done in plants although an S/M checkpoint has been partially characterised in zygotes of Fucus spiralis (Corellou et al., 2000). Addition of aphidicolin inhibited DNA replication and cell division. Here, the DNA replication checkpoint prevented chromatin condensation through to cytokinesis and retained CDK-like proteins in a phosphorylated state (Corellou et al., 2000). However, we know little of the genetic makeup of plant cell cycle checkpoints. Moreover, we lack homologues to cdc25 and the checkpoint kinases. A passing comment by Rhind and Russell (2000) suggested a Chk1 homologue in plants but it was not revealed in the Arabidopsis genome sequencing programme. A tangential clue about the existence of this latter kinase in plants comes from the discovery of the COP9 signalosome complex in Arabidopsis and its homologous complex in insects and humans (Wei and Deng, 1999). In Arabidopsis, this complex is important in repressing phytochrome-mediated responses in darkness and hence is integral to the etiolated response in plants (Wei and Deng, 1999). Indeed, it was mutants of Arabidopsis that responded in the dark as if they were in the light that led to the detection of the COP9 complex. The complex comprises eight protein sub-components (Wei and Deng, 1999). In vertebrates, the consensus view is that the complex is capable of forming the protein lid of the 26S proteosome, a sort of proteinaceous dustbin for ubiquitinated proteins destined for proteolytic destruction (Wei and Deng, 1999). In fission yeast, each of the eight proteins has been characterised and sub-unit 1 seems to act during the S-phase of the cell cycle although more work is required to verify its role (Mundt et al., 1999). Hence, this is the only subunit of the COP9 signalosome that shows a cell cycle function and in the author’s view, a Chk1 homologue has yet to be unequivocally identified in plants. So, what is the molecular composition of DNA replication and damage checkpoint in plants? As mentioned above, there are plant homologues to some but by no means all of the key players in the DNA replication and damage checkpoint pathways. Neither a p53-like gene nor one capable of encoding a full-length CDC25 phosphatase has been found in the Arabidopsis genome. Since WEE1 homologues exist in plants, logically one might think that there is a competing phosphatase(s), capable of catalysing the removal of phosphates from plant CDC2. Recently, we

158

D. FRANCIS

isolated an Arabidopsis gene that encodes a 362 amino acid protein with 28% sequence homology with TWINE, a Drosophila homologue to fission yeast cdc25 (D.A. Sorrell, D. Francis, H.J. Rogers and J.R. Dickinson, unpublished data; Alphey et al., 1992). The predicted protein lacks the regulatory domain of a CDC25 although it exhibits phosphatase activity when expressed in E coli (J. Joubes, L. DeVelder and D. Inze unpublished data). However, much more work is required to discover whether it truly functions as a plant CDC25. In the model proposed in Figure 6, WEE1 features strongly. Indeed, it may have a stronger role in plant cell cycle checkpoints than in fission yeast, given that there is neither a Mik1 nor a Myt 1 in plants. Moreover, in fission yeast and Xenopus, in response to DNA damage, Wee1 is phosphorylated by a Chk1 resulting in the maintenance of Y15 phosphorylation of CDC2 and hence G2 delay (O’Connell et al., 1997; Raleigh and O’Connell, 2000). As mentioned earlier, 14-3-3 proteins have an important role in the checkpoint by protecting specific phosphorylated residues of WEE1 and CDC25 from attack by native phosphatases. In Arabidopsis we found that in two hybrid screens, a specific 14-3-3 (GF14!), was detected when fission yeast Cdc25 was used as a bait to an Arabidopsis cDNA library, and was able to complement both the DNA damage and DNA replication checkpoints in fission yeast (D.A. Sorrell, A.M. Marchbank. J.R. Dickinson, D. Francis, N.G. Halford, C.S. Grierson, unpublished data). GF14! is one of five plant 14-3-3s able to restore UV resistance to the fission yeast mutant rad24 (Kuromori and Yamamoto, 2000). However, in our study only GF14! was able to both bind to Cdc25 and to exhibit a cell cycle checkpoint function in fission yeast. Since, in fission yeast, Wee1 is also a target of the Chk1 kinase, the plant WEE1 may be an integral target of the plant cell cycle checkpoints. Indeed, as mentioned above, in the absence of Mik1, Myt1 and Cdc25, it is the only potential target that we know anything about in the plant cell cycle checkpoint. Hence, I hypothesise that in both the DNA damage and replication checkpoints plant WEE1 is phosphorylated by a plant-like Chk1 kinase (or at least a functionally equivalent protein) and that the phosphorylated residue(s) is/are protected by a plant 14-3-3. Whether it could be GF14! is currently under scrutiny in the author’s laboratory. This hypothesis has been proven in various animal systems (Rowley et al., 1992; O’Connell et al., 1997; Michael and Newport, 1998; Raleigh and O’Connell, 2000). D. DO PLANTS REQUIRE CELL CYCLE CHECKPOINTS?

Plants are continually bombarded with environmental stresses, not least harmful UVB radiation. As noted by Britt (1995), because of their sessile

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

159

habit, plants have evolved efficient mechanisms for the management of environmentally induced DNA damage. Hence, proliferative cells must be well-equipped to deal with DNA-related damage or perturbation. However, this is not to overlook that meristems are probably well protected from the harmful effects of UVB radiation or IR. For example, the shoot apical meristem is normally covered by several leaf primordia and young leaves. Higher plant leaves synthesise flavonoids at the cuticular surface; they protect the inner tissues from harmful radiation (Rozema et al., 1997). However, proliferative cells in emerging leaves may be far less protected. In any event, cell cycle checkpoints, which are remarkably well conserved from yeasts through to humans, must be an important part of the plants armoury against environmental damage. This is particularly pertinent in relation to the classic response of plants exposed to sub-lethal UVB radiation where typically they exhibit stunted aerial growth, a loss of apical dominance and an outgrowth of lateral shoots from previously dormant axillary buds (Britt, 1999). Axillary buds break dormancy through a cell cycle activation. For example, in Helianthus tuberosus, cells in vegetative lateral buds are activated from G1 to S-phase, and a semi-synchronous population of cells enters mitosis (Tepfer et al., 1981); this is the first important step in the outgrowth of the axillary bud. Hence, in plants exposed to UVB, cell cycle checkpoints must be operating to allow for DNA repair prior to the lateral outgrowth response. Clearly, whole plant, cellular and molecular responses of plants to UV irradiation is a well-documented field of plant research and the topic has been reviewed extensively elsewhere (Rozema et al., 1997; Jansen et al., 1998; Britt, 1999). What follows is a very brief survey of UV-induced DNA repair and how this may relate to cell cycle checkpoints. UV radiation induces oxidative damage and cross links into DNA but the most common lesions are pyrimidine dimers and so-called pyrimidine[6-4]-pyrimidinone dimers. The two main types of repair involve either direct reversal, or excision of the damaged segment. The toxic and mutagenic effects of UV radiation can be negated by subsequent exposure to light in a bandwidth spanning the UVA through-to-the blue spectrum (360–420 nm). Under these conditions, photolyases are activated that reverse UV-induced pyrimidine dimers to monomers. An Arabidopsis homologue to microbial photolyase has been identified (Ahmad, 1997; Jiang et al., 1997) and, a gene encoding a 6-4 photolyase, UVR3, has been cloned. In addition, plants are capable of repairing UV-induced dimers in darkness in a repair pathway that runs parallel with photolyases. As noted by Britt (1999), plants probably share with other organisms

160

D. FRANCIS

the ability to undergo nucleotide excision repair. In effect, the damaged strand is nicked at its 50 and 30 ends and the undamaged DNA is used as a template to restore the original sequence. There are a number of homologues to budding yeast RAD genes involved in DNA damage recognition, and encoding 50 and 30 endonucleases: RAD1 (S. pombe homologue is rad10), RAD2, RAD4, RAD10, and RAD14. Moreover an atm mutant of Arabidopsis is sensitive to ionising radiation and in this mutant, transcriptional induction is impaired for the following DNA repair-like genes: AtRAD51, AtPARP-1, AtGR1 and AtLIG4 (Garcia et al., 2003). This work provides important molecular evidence of links between the repair of DNA and DNA damage detected via the ATM sensor. Regarding the influence of UVB on the whole plant, conclusions tend to be more tentative given the differences that exist concerning light sources and light quality deployed in lab- compared with field-based studies. As noted by Jansen et al. (1998), there are fine lines to be drawn between UVB-induced damage, repair and acclimation. Moreover, recent research has identified multiple targets for UVB although much more needs to be learnt about their environmental relevance. However, at the molecular level, mutants of Arabidopsis deficient in UVR genes show increased UV sensitivity and reduced growth responses. For example, double mutants deficient in excision repair (uvr1) and one of two different types of photolyase (uvr2 and uvr3) exhibited reduced root growth when challenged with UV radiation. In particular, those that were deficient in photo repair were also deficient in photoactivation of root growth (Jiang et al., 1997). As noted by these authors, simple conclusions regarding particular environmental DNA damaging agents are restricted given the complexity of the effects that each lesion may have on transcription, DNA replication and cell cycle control. Nevertheless, the genetic approach is revealing the relative contribution of differing forms of UV-induced damage that promises to deliver very interesting insights into UV-induced genotoxicity.

III. INSTANCES OF PCD IN PLANTS Thus far, a case has been presented for the existence of cell cycle checkpoints in plants and a model has indicated putative players in the checkpoint team. What now follows is a glimpse at various instances of PCD both during plant development and in cell cultures. In each example, a question will be raised about whether a cycle interface exists or indeed whether one is necessary.

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

161

A. AERENCHYMA

Roots with internal holes are either infected or are exhibiting aerenchyma, intercellular spaces that facilitate gas diffusion in roots experiencing waterlogged conditions (Armstrong, 1979). This is not a random affair but occurs by a tightly regulated pattern of PCD (Armstrong and Armstrong, 1994). In maize roots, aerenchyma is induced by ethylene (Drew et al., 1979; Jackson et al., 1985; Arunika et al., 2001). The familiar PCD-related chain of reactions: cytosolic destruction, apoptosis, chromatin condensation and nuclear blebbing, are all hallmarks of aerenchyma formation (see Brailsford et al., 1993). A recent study in maize confirmed that cortical tissue is targeted for aerenchyma formation. The first detectable events were plasma membrane invagination together with the accumulation of numerous vesicles immediately beneath the plasma membrane (Arunika et al., 2001). By the time that cytoplasmic changes were observed, chromatin condensation became apparent. These authors drew attention to other studies in both plants and animals in which cytoplasmic changes that led to PCD were observed before chromatin condensation (Jacobsen et al., 1994; Earnshaw, 1995; Mittler et al., 1997). Aerenchyma formation in maize is a form of PCD that resembles both apoptosis and cytoplasmic cell death (Arunika et al., 2001). Ethylene induces aerenchyma formation in non-cycling cortical cells. Whether as a prerequisite for aerenchyma formation, exit into PCD is cell cycle specific is unknown. Presumably, since these cortical cells are essentially non-proliferating, cell cycle checkpoints would be redundant in this system? B. PCD IN CELL CULTURES

Cell cultures, which are free of developmental constraints, have become useful systems for studying PCD, greatly assisted by the availability of easily synchronised cell populations so that cellular responses to putative agents of PCD can be tested in a relatively homogeneous group of cells. Perhaps the best for this purpose is the tobacco TBY-2 cell line that can be easily synchronised using aphidicolin, an inhibitor of the replicative form of DNA polymerase (Nagata et al., 1992). More recently, an Arabidopsis cell line has yielded equally good levels of synchrony following aphidicolin treatment (Mengers and Murray, 2002). At one point it was questioned whether PCD could occur in cell cultures given the relevance of its occurrence in relation to plant developmental processes. However, in Arabidopsis cell cultures, genes expressed in senescing organs also accumulate in log phase of cell cultures (Callard

162

D. FRANCIS

et al., 1996). Notably, features of PCD, inter-nucleosomal cleavage and chromatin compaction, were observed in these Arabidopsis cultures at times corresponding to a loss of cell viability (Callard et al., 1996). As noted by McCabe and Leaver (2000), cells have intrinsic cell suicide programmes with constitutive expression of PCD machinery (Weil et al., 1996). PCD invariably involves a spike in cytosolic calcium. In soybean and tobacco cultures such an increase induces PCD as does oxidative bursts. Hydrogen peroxide added to Arabidopsis cultures can also induce PCD and it can be suppressed by the addition of catalase (Neill et al., 2002). Fungal elicitors of the HR can also induce PCD when added to cell cultures. For example, the addition of xylanase, an HR inducer from Trichoderma virida can induce various hallmark features of PCD when added to cell suspensions (Yano et al., 1998, 1999). The bacterial elicitor, harpin, can also induce PCD in culture (Desikan et al., 1998). In each of these instances, whether the induction of PCD was dependent on cells being positioned at a particular cell cycle phase (e.g. G1/S or G2/M) was not explored. 1. Ethylene induces cell cycle specific PCD The extent to which exit into PCD might be cell cycle specific was investigated using the tobacco TBY-2 cell line. We used a known inducer of PCD, ethylene, in these experiments. Cells at the S/G2 transition were subjected to ethylene treatment (Fig. 7). Viable cells exhibited a G2 phase

Fig. 7. Cell cycle-specific exit into PCD in the tobacco TBY-2 cell line. Ethylene given to cells at S/G2 results in an extension of the G2 phase, a peak of cells exhibiting nuclear DNA fragmentation in late G2 but enables the majority of the cycling cells to divide and cycle as normal (based on Herbert et al., 2001).

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

163

that was one hour longer than in wild-type, and in late G2 there was a marked increase in the level of mortality and DNA shearing, in the form of exposed 50 ends of DNA (as revealed by the TUNEL assay in Herbert et al., 2001). To induce detectable levels of PCD in the TBY-cell line it was necessary to provide a high exogenous concentration of ethylene to the cells (17,000 ml l1). Inevitably, this prompts the question: how physiologically meaningful is such a high dose of ethylene? The peak of ethylene-induced mortality in late G2 was suppressed by the simultaneous addition of silver ions to the medium at a concentration that was identical to that used in in vivo studies (Drew et al., 1981). A dose-dependent curve also revealed that concentrations less than 17,700 ml l1 were ineffective in inducing mortality in the TBY-2 cell line. Moreover, the maximum level of mortality was approximately 30%. Cells that survived this treatment (the majority), exhibited a rate of cell division that was no different from wild-type. Hence, we would have expected catastrophic levels of mortality if our ethylene treatment was a massive perturbation. The data supported a model of cell cycle specific exit into PCD. For viable cells it was hypothesised that the one hour delay of entry of cells into mitosis was because of the operation of a DNA-damage type checkpoint control (Herbert et al., 2001). It is further hypothesised that a plant ATM/ATR-mediated network, of the type projected in Fig. 3 operates in response to the stress imposed by ethylene. It would follow that catastrophic ethylene-induced DNA damage would lead the cell into PCD. In support of this hypothesis was the observation of maximal nuclear DNA fragmentation at the G2/M transition of these synchronised cells (Herbert et al., 2001). More recently, the simultaneous addition of ethylene and the ethylene receptor blocker, MCP at S/G2, totally suppressed ethylene-induced increases in mortality at the G2/M boundary. Indeed, the mortality levels in this treatment were no different from control levels throughout a synchronised cell cycle (C.B. Orchard, R.J. Herbert, H.J. Rogers and D. Francis, unpublished data). These data suggest strongly that ethylene was inducing cell cycle-specific PCD in the TBY-2 cell line. Moreover, we have now transformed the TBY-2 cells with the dominant-negative etr1-1 gene from Arabidopsis. This transgenic line exhibits high levels of mortality throughout the cell cycle (between 10–20%). However, the addition of exogenous ethylene did not affect this underlying level of mortality. Hence, through both a chemical and genetic approach we have been able to demonstrate that ethylene must be binding to its receptor protein in order to generate a transcellular signal that leads to cell cycle-specific cell death. Presumably, where the DNA damage is less extensive a plant cell cycle

164

D. FRANCIS

checkpoint is induced that facilitates the recovery of many cells from the damage inflicted by exogenous ethylene. 2. Exit into PCD from G1/S Much has been made here, and in earlier sections, about the G2/M transition in relation to exit from the cell cycle into PCD but what about other cell cycle transitions? The retinoblastoma (Rb) protein, is an important regulator of the G1/S transition in plants and animals. It represses the activity of a transcription factor, E2F that is essential for the G1/S-phase transition (see Section II.A). In mammals, hyperphosphorylation of Rb by Cdk4-cyclinD and Cdk2-cyclinE releases E2F into an active state to drive cells into S-phase (see Murray et al., 1998). Also, in mammals, Rb has been detected as a substrate for an apoptotic protease (Kasten and Giordano, 1998). This is consistent with the idea that apoptosis could be linked with the cell cycle machinery through the phosphorylated status of Rb either through signalling to a G1 cell to enter the cell cycle or through signalling to exit into PCD. Although the nature of the signalling pathways remains rudimentary, a common denominator would seem to be a sensing mechanism that monitors the fidelity of DNA prior to replication. There is strong evidence of exit into PCD at the G1/S transition (Tanaka et al., 2000).

IV. XYLOGENESIS The initiation of a tracheary element (TE) through to its maturation is, perhaps, the perfect example of PCD representing a transition from a functional live to a functional dead cell. The regulated autolysis of the protoplasts, the lignification of the cell wall and the hollowing of the entire cell are all geared to make it a conduit for the passage of water and nutrients about the plant. The pioneer work of H. Fukuda and A. Komamine will not be plagiarised here; the reader is encouraged to explore their reviews and papers (e.g. Fukuda, 1997, 2000; Fukuda and Komamine, 1980a,b). The question posed is what do we know about cell cycle exit in this transition from cambial initial to TE? The question is reminiscent of a swathe of older cell cycle literature that centred on the necessity of cells to go through a cell cycle before differentiating as TEs (e.g. Shininger, 1979; Fukuda and Komamine, 1981a). Various cellular systems have been used to study differentiation of TEs. Komamine devised a cell culture system generated from Zinnia elegans for this purpose. Freshly isolated cells from the Zinnia mesophyll are, in effect, non-cycling parenchyma cells that in culture would be expected to re-enter

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

165

the cell cycle. Early studies of Zinnia cultures suggested that the cell cycle was uncoupled from TE formation, because well over half of the cells differentiated into TEs in the absence of an intervening mitosis (Fukuda and Komamine, 1980b). DNA synthesis occurred in only a subpopulation of the TEs (Fukuda and Komamine, 1981a) so that mesophyll cells were able to differentiate as TEs either with or without cell cycle progression (Fukuda and Komamine, 1981b, reviewed by Fukuda, 1992). However, when the cells were treated with inhibitors of DNA replication, TE differentiation was perturbed (Fukuda and Komamine, 1981b), suggesting that S-phase progression was important for the differentiation process. Since previous studies indicated that cell cycle progression was unnecessary for TE formation, Komamine and his colleagues suggested a requirement for DNA repair synthesis in the absence of the S phase. In support of this idea, inhibitors of repair activity blocked the differentiation of TEs (Sugiyama and Komamine, 1987). Induction of repair-like enzymes is associated with ATR/ATRIP activity in animals (see Section II.C), and, hence, may be linked with checkpoint controls. However, for TEs, the extent of DNA damage would clearly escape checkpoint control so that PCD is the end point. However, how this could be achieved is unknown. A recent elegant study returned to the issue of TE differentiation in the Zinnia system (Mourelatou, 2001). As mentioned above (Fig. 8A), this is a model system in which mesophyll cells are cultured for 48 h and then an auxin and a cytokinin are added (Fukuda and Komamine, 1980a). Over the next 48 h the cells go through a number of well-documented stages before becoming functional TEs: I, a dedifferentiation stage comprising rearrangements of actin, changes in gene expression together with DNA repair-type synthesis; II, an increase in transcription rates, increase in microtubules and an elaboration of the endoplasmic reticulum and III, secondary wall thickening (Fukuda, 2000). The work of M. Moraletou, M. McCann and J. Doonan (unpublished data) consisted of an analysis of cell cycle genes during the differentiation of the Zinnia mesophyll cells using the Fukuda and Komamine (1980a) culture system. If at the start of the ‘second’ 48 h, aphidicolin was added for 24 h, cells were blocked in early S-phase but they went on to become functional TEs (Fig. 8B). However, if roscovitine was added 24 h following the start of the second 48 h stage (i.e. 48 þ 24 in Fig. 8C), cells went through three cell cycles but became blocked in early S-phase of the fourth cell cycle. Importantly, in this treatment, the cells failed to differentiate (Fig. 8C). Roscovitine targets the transition from G1 to S-phase but does not entirely block the cell cycle. If roscovitine was added at the start of the second 48 h stage but was removed 24 h later, cells would form TEs as normal (Fig. 8D). These data are consistent in showing that

166

D. FRANCIS

Fig. 8. The Zinnia elegans system of tracheary element (TE) formation. (A) Addition of auxin and cytokinin, 48 h following the start of the culture of mesophyll cells results in them forming TEs 48 h later (48 þ 48 h) even if the plant growth regulators are removed after 24 h (48 þ 24). (B) Addition of aphidicolin (apd) at the beginning of the second 48 h for 24 h (48 þ 24) does not inhibit TE formation. (C ) Addition of roscovitine (rvt) 24 h following the start of the second 48 h (48 þ 24), inhibits TE formation whereas (D), addition of rvt for 24 h at the beginning of the second 48 h does not inhibit TE formation (based on Morelatou (2001)).

events occurring during the early S-phase are important during the differentiation process. The suggestion is that by impeding CDK activity during early S-phase, the programme of events leading to PCD is blocked. Hence, there is a tangible link between the cell cycle and PCD although whether this link involves cell cycle checkpoints is currently unknown. In dedifferentiating tobacco mesophyll cells re-entering the cell cycle (Zhao et al., 2001), a late G1 phase, referred to by these authors as the ‘supra G1’ requires the presence of auxin and cytokinin. This supra G1 just before S phase entry, could be a special feature of cells re-entering the cell cycle and would correspond to ‘START’ in yeast (reviewed by Forsburg and Nurse, 1991). The question raised by Mourelatou (2001) was whether this ‘early’ S phase represented a unique window that enhances TE formation by maintaining cells at their most ‘competent’ phase for differentiation. Thus, the G1/S transition maybe necessary for differentiation although how this is different in a differentiating cell compared to a cycling one is unknown. However, Zinnia cells arrested

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

167

in early S phase by aphidicolin exhibit enhanced levels of TE formation (Mouraletou, 2001).

V. PCD IN THE ENDOSPERM In monocots endosperm is a long-lived nutritive tissue upon which the embryo depends prior to the acquisition of photosynthetic competence. In wheat it develops following the double fertilisation of the ovule by two male gametes. One fuses with two polar nuclei that then sparks a series of synchronous nuclear divisions that within 5 days post-anthesis generates a nuclear number of 512-1024 triploid nuclei. Cellularisation follows after which there is a limited mitotic phase, final cell number increasing to over 100,000 (Gao et al., 1992). During this period, PCD occurs in cells close to the embryo, as a cavity is created that is essentially a photosynthetic conduit between endosperm and embryo (Young and Gallie, 1999). During the late cellular phase a decline in cell number can occur (e.g., Gao et al., 1992). Maize endosperm goes through the same phases but the timing differs (Young and Gallie, 2000). Cellularisation begins 3–5 days after pollination (DAP) and between days 5–12, a combination of cell division and cell expansion leads to an enlarged endosperm. In this system, nuclear DNA synthesis continues beyond the period of cell divisions. The end result is that cells become endopolyploid so that between 16–18 DAP, cells have enlarged and contain many times the amount of DNA in diploid cells. From 12–15 DAP, marks the transition from the cellular to maturation phases characterised by starch deposition. During the late maturation phase, maize endosperm undergoes marked PCD and this follows from the wave of endoreduplication as characterised by the rise and fall of endosperm nuclear DNA amount over this period (Young and Gallie, 2000). Endoreduplication cycles are characterised by the eventual appearance of thickened bands of chromosomes known as polytene chromosomes. Endoreduplicated DNA may provide a ‘store’ of replicated DNA to encode proteins in large amounts required for the mobilisation of nutrients. PCD can be delayed by treatment of the endosperm with inhibitors of ethylene action/synthesis and abscisic acid inhibits aleurone cell death, a treatment that can also be simulated by okadaic acid, a protein phosphatase inhibitor (Kuo et al., 1996; Wang et al., 1996). Clearly during endoreduplication cycles, the normal dependence of mitosis on nuclear DNA replication has been uncoupled as cells bypass mitosis. Moreover, the appearance of long banded polytene chromosomes indicates that the normal contraction cycle of chromosomes that normally

168

D. FRANCIS

peaks in mitosis is lost. This uncoupling can also be considered in terms of checkpoints. The rapid ‘re-replication’ of the genome is occurring in the absence of normal cell division. In effect, the DNA damage checkpoint is probably no longer working and the rapidity of the DNA replication mechanism must place a strain on the fidelity of the nascent DNA. Hence it seems reasonable to suggest that a lack of fidelity of the DNA replication mechanisms would normally be promoting the DNA damage checkpoint. But as already intimated, this checkpoint may have ceased to function in these endoreduplicating nuclei. Hence, if the frequency of errors in replication is increased, the DNA is becoming increasingly damaged, which in the absence of a DNA damage checkpoint may tip the cells into PCD. The occurrence of endopolyploidy and subsequent PCD in developing endosperm was also highlighted by Young and Gallie (2000).

VI. OXIDOREDUCTIVE STATES AND PCD The balance between life and death, in terms of cell proliferation or PCD, is affected strongly by the oxidoreductive state of the cells. A strong case exists for a role for hydrogen peroxide and nitric oxide in signalling pathways that steer a cell to PCD. H2O2 together with superoxide and hydroxyl radicals are collectively known as reactive oxygen species (ROS). To counter this oxidative damage are antioxidant enzymes such as catalase (Vranova et al., 2002). Oxidative stress results when H2O2 is produced more rapidly than it is metabolised. In Arabidopsis cell cultures, H2O2 can induce PCD together with the expression of genes required for the defence against pathogens (Desikan et al., 1998, 2000) but it probably also affects DNA integrity (Neill et al., 2002). Clearly, much is known about how active oxygen species (AOS) such as superoxide radical (O.2) and H2O2 can alter the redox state which, in turn, can affect many cellular proteins (Vranova et al., 2002). For example, in relation to pathogen attack, at low concentrations, AOS can induce defence genes, although high concentrations induce cell death. In transgenic plants over expressing H2O2-generating enzymes, cell death occurred spontaneously and could be easily induced by stress (Chamnongpol et al., 1998; Kazan et al., 1998). O2, but not H2O2 could initiate a cell death phenotype in the lsd1 mutant of Arabidopsis (Jabs et al., 1996). Lsd1 plants grown in long days, formed necrotic lesions profusely and cell death occurred unchecked. Exogenous ethylene increases O2-dependent cell death, whereas perturbing ethylene perception blocks O2 accumulation and the spread of lesions (Overmyer et al., 2000). PCD is a typical cellular fate

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

169

following these changes in redox potential. Redox status can also influence the cell cycle although different systems react differently to molecules that are directly influenced by the redox status. For example, in Lycium barbarum reduced glutathione (GSH) treatment resulted in an increase in the proportion of cells undergoing mitosis, whereas removal of GSH had the opposite effect (Cui et al., 1999). Conversely, ascorbate and dehydroascorbate could regulate cell division in the tobacco TBY-2 cell line. A peak in ascorbate levels but not GSH coincided with a peak in the mitotic index during an exponential growth phase of this cell line (De Pinto et al., 1999). Clearly, much published literature exists about the oxidoreductive status of proliferating cells and those undergoing PCD. More data are now required to bring this information together.

A. FUNGAL ELICITORS OF PCD

Fungal pathogens can induce the so-called hypersensitive response (HR) during which an oxidative burst coincides with the induction of cell death at the site of pathogen attack. The HR response encompasses both cell death and defence gene expression. The whole response is dependent on the presence of the plant pathogen or elicitors produced by it. The localised sacrifice of plant cells within the vicinity of the pathogen limits the spread of the invading fungus. The sequence of events for HR depends on calcium influxes and signalling. For example, Ca2þ signalling precedes membrane depolarisation and electrolyte leakage prior to the HR caused by Phytophthora infestans (Doke, 1983; Doke et al., 1998). However, there is strong cross-talk between cell death during HR and during normal development. For example, AtSAG12, a marker for senescence (BuchananWollaston, 1997), is also expressed in cells surrounding Tobacco Mosaic Virus- and bacterial-induced HR lesions in transgenic tobacco (Pontier et al., 1999). Resistance to fungal invasion depends on the expression of R genes. The characteristic features of their encoded products are leucine-rich repeats (LRRs) or serine-threonine kinase domains consistent with signalling systems. Surprisingly, there is no relationship between the category of R gene and the type of organism against which it confers resistance. However, several R gene products show similarity to the nematode CED-4 and the mammalian homologue, APAF-1 suggesting a link between R genes of HR and the regulation of PCD. Only putative homologues of animal apoptosis suppressor genes Bcl-2 (Dion et al., 1997) and defender against death (DAD) (Dong et al., 1998) have been detected in plants. Although Heath (1998) concluded cautiously about links between

170

D. FRANCIS

plant DAD-like genes and PCD (Heath, 1998) DAD-1 is generally regarded as a phytosuppressor of PCD; rice DAD-1 can suppress PCD in animals (Tanaka et al., 1997). In an elegant review of apoptosis in plant diseases, Gilchrist (1998) forged a link between cell cycle exit from G2 to nuclear DNA fragmentation. Positive TUNEL (Terminal Desoxyribosyl-Transferase mediated duTP Nick End Labelling) reactions and DNA ladders are hallmark features of fungal or fungal elicitor-induced HR (Ryerson and Heath, 1996). Clearly, DNA fragmentation is also the end-point of pathways of caspase activity, and Caþ signalling and endonuclease activities. Moreover oxidative stress can cause cell cycle arrest (Reichfeld et al., 1999). Clearly oxygen stress creates huge redox changes within the cell that can lead to PCD. It would follow that one fall-out from oxygen stress is DNA fidelity that either sustains viability or commits the cell to suicide. However, currently, we lack the molecular links between DNA damage detectors in plants and the road to PCD.

VII. CONCLUSIONS The aim of this review has been to search for interfaces between the cell cycle and programmed cell death in plants. The striking interfaces in yeasts and animals are the cell cycle checkpoints. A simple conclusion is that we know hardly anything about the molecular landscape of plant cell cycle checkpoints compared with a relatively vast amount known about checkpoints in yeast and animal systems. Clearly, when human cell cycle checkpoints are faulty, cells divide out of control with alarming consequences in the form of cancerous tumours. Often, cancers have mutated versions of key checkpoint proteins, notably p53 (Lane, 1992). Mostly, plants do not develop cancers (Doonan and Hunt, 1996) or if they do, the tumour is not life threatening (e.g. crown gall). However, a plant’s sessile habit dictates that it must adapt to its environment and it must also be able to adapt to changing environmental conditions that increasingly contain more and more DNA-damaging pollutants. As noted by Britt (1999), plants do have efficient mechanisms to cope with DNA damage. Hence, plants probably also possess very efficient cell cycle checkpoints. Given the strong conservation of key cell cycle genes such as CDC2, and the DNA damage sensor, ATM, it is reasonable to conclude that plant cell cycle checkpoints share some components with animal ones. Nevertheless, this is not to conclude that plant cell cycle checkpoints mirror the animal ones. Indeed the absence of plant homologues to p53 and checkpoint kinases must reflect

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

171

evolutionary divergence. The unique regulation of key cell cycle transitions by plant growth regulators, suggests that analysis of plant cell cycle checkpoints will reveal some unusual plant proteins that are able to perform checkpoint-like functions. In this review some instances of PCD have been illustrated with the aim of seeking interfaces between PCD and the cell cycle. Aerenchyma forms as terminally differentiated cortical cells in the root breakdown. Effectively, the cell cycle is complete prior to the collapse of these cells. Hence, cell cycle checkpoints are probably irrelevant to the formation of this tissue. However, PCD in cell cultures has been shown to be cell cycle-specific leading to the hypothesis that cell cycle checkpoints are overridden in catastrophically damaged cells. Xylogenesis seems to be linked to early S-phase of the cell cycle. Here, exit into PCD may be related to expression of genes such as those encoding Rb and CDKs that are specifically expressed at the G1/S transition. In the endosperm, PCD seems to be linked with endoreduplication. Here, the author’s hypothesis is that the continual replication of nuclear DNA in the absence of mitosis effectively means that the cells have escaped checkpoint control. Moreover, endoreduplication probably places a strain on the fidelity of the DNA replication mechanism so that eventually the cell flips into PCD. Clearly there are other instances of PCD in plants that have not been covered in this review (e.g. reproductive tissues) mainly because of the timely recent reviews of these systems elsewhere (see Wu and Cheung, 2000). In these instances, links between the cell cycle and PCD may also be operative. Finally, some consideration has been given to the oxidoreductive state of cells and tissues that undergo PCD. Exploration of the links between cell cycle control and fungal elicitors of PCD is worthy of more research. Finally, the pendulum of this review has swung largely to cell cycle controls and cell cycle checkpoints as opposed to PCD itself. However, as pointed out at several points in this paper, PCD is expertly covered elsewhere whilst cross-talk between the cell cycle and PCD has been the main theme here. With the molecular techniques now available it should now be possible to fill the gaps in the plant cell cycle checkpoints and test the extent to which they truly interface with the processes of PCD. Hopefully ‘from division unto death’ will be a short-lived euphemism.

ACKNOWLEDGEMENTS I thank Dr. Maria Morelatou for allowing me access to unpublished data from her PhD thesis. Other unpublished work mentioned in this review stems from BBSRC funding (ref. 72/P10942).

172

D. FRANCIS

REFERENCES Ahmad, M. (1997). An enzyme similar to animal type II photolyases mediates photoreactivation in Arabidopsis. Plant Cell 9, 199–207. Alberts, B., Johnson, A., Lewis, J., Raff, M., Roberts, K. and Walter, P. (2002). Molecular Biology of the Cell. 4th edition. Garland Science, New York. Al-Khodairy, F., Fotou, E., Sheldrick, K. S., Griffiths, D. J., Lehman, A. R. and Carr, A. M. (1994). Identification and characterisation of new elements involved in checkpoint and feedback controls in fission yeast. Molecular Biology of the Cell 5, 147–160. Alphey, L., Jimenez, J., White-Cooper, H., Dawson, I., Nurse, P. and Glover, D. (1992). Twine, a cdc25 homologue that functions in the male and female germline of Drosophila. Cell 69, 977–988. Armstrong, J. and Armstrong, W. (1994). Chlorophyll development in mature lysigenous and schizogenous root aerenchymas provides evidence of continuing cell viability. New Phytologist 126, 493–497. Armstrong, W. (1979). Aeration in higher plants. Advances in Botanical Research 7, 225–332. Arunika, H. L. A. N., Gunawardena, A. N., Pearce, D. M., Jackson, M. B., Hawes, C. R. and Evans, D. E. (2001). Characterisation of programmed cell death during aerenchyma formation induced by ethylene or hypoxia in roots of maize (Zea mays L.). Planta 212, 205–214. Baber-Furnari, B. A., Rhind, N., Boddy, M., Shanahan, P., Lopez-Girona, A. and Russell, P. (2000). Regulation of mitotic inhibitor mik1 helps to enforce the DNA damage checkpoint. Molecular Biology of the Cell 11, 1–11. Bentley, N. J., Holtzman, D. A., Flaggs, G., Keegan, K. S., DeMaggio, A., Ford, J. C., Hoekstra, M. and Carr, A. M. (1996). The Schizosaccharomyces pombe rad 3 checkpoint gene. EMBO Journal 15, 6641–6651. Blasina, A., Paegle, E. S. and McGowan, C. H. (1997). The role of inhibitory phosphorylation of CDC2 following DNA replication and radiationinduced damage in human cells. Molecular Biology of the Cell 8, 1013–1023. Boddy, M., Furnari, B., Mondesert, O., Russell, P. (1998). Replication checkpoint enforced by kinases Cds1 and Chk1. Science 280, 909–912. Brailsford, R. W., Voesenek, L. A. J. C., Blom, C. W. P. M., Smith, A. R., Hall, M. A. and Jackson, M. B. (1993). Enhanced ethylene production by primary roots of Zea mays L. in response to sub-ambient partial pressures of oxygen. Plant Cell & Environment 16, 1071–1074. Britt, A. B. (1995). Repair of DNA damage induced by ultraviolet radiation. Plant Physiology 108, 891–896. Britt, A. B. (1999). Molecular genetics of DNA repair in higher plants. Trends in Plant Science 4, 20–25. Britt, A. B., Chen J. J., Wykoff, D. and Mitchell D. A. (1993). UV-sensitive mutant of Arabidopsis defective in the repair of pyrimidine-pyrimidinone (6-4) dimers. Science 261, 1571–1574. Buchanan-Wollaston, V. (1997). The molecular biology of leaf senescence. Journal of Experimental Botany 48, 181–199. Buckner, B., Johal, G. S. and Janick-Buckner, D. (2000). Cell death in maize. Physiologia Plantarum 108, 231–239. Callard, D., Axelos, M. and Mazzolini, L. (1996). Novel molecular markers for the phases of the growth cycle of Arabidopsis thaliana cell suspension cultures are expressed during organ senescence. Plant Physiology 112, 705–715.

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

173

Caspari, T. and Carr, A. M. (2002). Checkpoints: How to flag up double-strand breaks. Current Biology 12, R105–R107 Chamnongpol, S., Wilekens, H., Moder, W., Langebartels, C., Sandermann, H., Jr Va Montague, M., Inze, D. and Van Camp, W. (1998). Defense activation and enhanced pathogen tolerance induced by H2O2 in transgenic plants. Proceedings of the National Academy of Sciences, U.S.A. 95, 5818–5823. Chan, T. A., Hermeking, H., Lengauer, C., Kinzler, K. W. and Vogelstein, B. (1999). 14-3-3 is required to prevent mitotic catastrophe after DNA damage. Nature 401, 616–620. Chen, P., Nordstrom, W., Gish, B. and Abrams, J. M. (1996). Grim, a novel cell death gene in Drosophila. Genes and Development 10, 1773–1782. Correlou, F., Bisgrove, S. R., Kropf, D. L., Meijer, L., Kloareg, B. and Bouget, F. -Y. (2000). A S/M DNA replication checkpoint prevents nuclear and cytoplasmic events of cell division including centrosomal axis alignment and inhibits activation of cyclin-dependent kinase-like proteins in fucoid eggs. Development 127, 1651–1660. Cortez, D., Guntuku, S., Qin, J. and Elledge, S. J. (2001) ATR and ATRIP: partners in checkpoint signalling. Science 294, 1713–1716. Cross, F., Roberts, J. and Weintraub, H. (1989). Simple and complex cell cycles. Annual Review of Cell Biology 5, 341–395. Cui, K., Xing, G., Liu, X., Xing, G. and Wang, Y. (1999). Effect of hydrogen peroxide on somatic embryogenesis of Lycium barbarum L. Plant Science 146, 9–16. DePinto, M. C., Francis, D. and De Gara, L. (1999). The redox state of the ascorbate-dehydroascorbate pair as a specific sensor of cell division in tobacco BY-2 cells. Protoplasma 209, 90–97. Desikan, R., Neill, S. J. and Hancock, J. T. (2000). Hydrogen peroxide-induced gene expression in Arabidopsis thaliana. Free Radical Biology and Medicine 28, 773–778. Desikan, R., Reynolds, A., Hancock, J. T. and Neill, S. J. (1998) Harpin and hydrogen peroxide both initiate programmed cell death but have differential effects on gene expresion in Arabidopsis suspension cultures. Biochemical Journal 330, 115–120. Dion, M., Chamberland, H., St-Michel, C., Plante, M., Darveau, A., Lafontaine, D. G. and Brisson, L. F. (1997). Detection of a homologue of bcl-2 in plant cells. Biochemistry and Cellular Biology 75, 457–461. Doke, N. (1983). Involvement of superoxide anion generation in the hypersensitive response of potato tuber tissue to infection with an incompatible race of Phytophthera infestans and to the hypha wall components. Physiological Plant Pathology 23, 345–357. Doke, N., Sanchez, L. M., Yoshioka, H., Kawakita, K., Miura, Y. and Park, H. -J. (1998). Hypersensitive response in potato tubers. In Molecular Genetics of Host-Specific Toxins in Plant Diseases (Kohmoto, J. and Yoder, O.C. eds). pp. 331–341. Kluwer Academic Publishers, Dordrecht, Netherlands. Dong, Y. -H., Zhan, X. -C., Kvarnheden, A., Atkinson, R. G., Morris, B. A. and Gardner, R. C. (1998). Expression of a cDNA from apple encoding a homologue of DAD1, an inhibitor of programmed cell death. Plant Science, 139, 165–174. Doonan, J. and Hunt, T. (1996). Why don’t plants get cancer? Nature 380, 481–482. Drew, M. C., Jackson, M. B. and Giffard, S. (1979). Ethylene promoted adventitious rooting and development of cortical air spaces in Zea mays L. Planta 147, 83–88.

174

D. FRANCIS

Drew, M. C., Jackson, M. B., Giffard, S. C. and Campbell, R. (1981). Inhibition by silver ions of gas space (aerenchyma) formation in adventitious roots of Zea mays L. subjected to exogenous ethylene or to oxygen deficiency. Planta 153, 217–224. Durocher, D. and Jackson, S. P. (2001). DNA-PK, ATM and ATR sensors of DNA damage: variations on a theme? Current Opinion in Cell Biology 13, 225–231. Earnshaw, W. C. (1995). Apoptosis – lessons from in vitro systems. Trends in Cell Biology 5, 217–220. Edgar, B. A. and O’Farrell, P. H. (1990). The three postblastoderm cell cycles of Drosophila embryogenesis are regulated in G2 by string. Cell 62, 469–480. Edgar, B. A. and O’Farrell, P. H. (1989). Genetic control of cell division patterns in the Drosophila embryo. Cell 57, 177–187. Ellis, R. E., Yuan, J. and Horwitz, H. R. (1991). Mechanisms and functions of cell death. Annual Review of Cell Biology 7, 663–698. Ford, J. C., Alkhodairy, F., Fotou, E., Sheldrick, K. S., Griffiths, D. J. F. and Carr, A. M. (1994). 14-3-3-protein homologs required for the DNA-damage checkpoint in fission yeast. Science 265, 533–535. Forsburg, S. L. and Nurse, P. (1991). Cell cycle regulation in the yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe. Annual Review of Cell Biology 7, 227–256. Francis, D. and Inze, D. (2001). The plant cell cycle. In The Plant Cell Cycle and its Interfaces (D. Francis, ed.). Sheffield Academic Press, Sheffield, U.K. pp. 1–18. Francis, D. and Sorrell, D. A. (2001). The interface between the cell cycle and plant growth regulators: a mini review. Plant Growth Regulation 33, 1–12. Fukuda, H. (1989a). Cytodifferentiation in isolated single cells. Botanical Magazine, Tokyo 102, 491–501. Fukuda, H. (1989b). Regulation of tubulin degradation in isolated Zinnia mesophyll cells in culture. Plant & Cell Physiology 30, 243–252. Fukuda, H. (1992). Tracheary element formation as a model system of cell differentiation. International Review of Cytology 136, 289–332. Fukuda, H. (1994). Redifferentiation of single mesophyll cells into tracheary elements. International Journal of Plant Science 155, 262–271. Fukuda, H. (1996). Xylogenesis: initiation, progression, and cell death. Annual Review of Plant Physiology and Plant Molecular Biology 47, 299–325. Fukuda, H. (1997). Tracheary element differentiation. Plant Cell 9, 1147–1156. Fukuda, H. (2000). Programmed cell death of tracheary elements as a paradigm in plants. Plant Molecular Biology 44, 245–253. Fukuda, H. and Komamine, A. (1980a). Establishment of a new experimental system for the study of tracheary element differentiation from single cells isolated from the mesophyll of Zinnia elegans. Plant Physiology 65, 57–60. Fukuda, H. and Komamine, A. (1980b). Direct evidence for cytodifferentiation to tracheary elements without intervening mitosis in a culture of single cells isolated from the mesophyll of Zinnia elegans. Plant Physiology 65, 61–64. Fukuda, H. and Komamine, A. (1981a). Relationship between tracheary element differentiation and the cell cycle in single cells isolated from the mesophyll of Zinnia elegans. Physiologia Plantarum 52, 423–430. Fukuda, H. and Komamine, A. (1981b). Relationship between tracheary element differentiation and DNA synthesis in single cells isolated from the mesophyll of Zinnia elegans – Analysis by inhibitors of DNA synthesis. Plant & Cell Physiology 22, 41–49.

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

175

Gao, X., Francis, D., Ormord, J. C. and Bennett, M. D. (1992). Changes in cell number and cell division activity during endosperm development in allohexaploid wheat, Triticum aestivum L. Journal of Experimental Botany 43, 1603–1609. Garcia, V., Bruchet, H., Camescasse, D., Granier, F., Bouchez, D. and Tissier, A. (2003). AtATM is essential for meiosis and the somatic response to DNA damage in plants. Plant Cell 15, 119–132. Garcia, V., Salanoubat, M., Choisne, N. and Tissier, A. (2000). An ATM homologue from Arabidopsis thaliana: complete genomic organisation and expression analysis. Nucleic Acids Research 28, 1692–1699. Gilchrist, D. G. (1998). Programmed cell death in plant disease: The purpose and promise of cellular suicide. Annual Review of Phytopathology 36, 393–414. Gould, K. L. and Nurse, P. (1989). Tyrosine phosphorylation of the fission yeast cdc2þ protein kinase regulates entry into mitosis. Nature 342, 39–45. Hall-Jackson, C. A., Cross, D. A. E., Morrice, N. and Smythe, C. (1999). ATR is a caffeine-sensitive DNA-activated protein kinase with a substrate specificity distinct from DNA-PK. Oncogene 18, 6707–6713. Hartwell, L. H. and Weinert, T. A. (1989) Checkpoints: controls that ensure the order of cell cycle events. Science 246, 629–634. Heath, M. C. (1998). Apoptosis, programmed cell death and the hypersensitive response. European Journal of Plant Pathology 104, 117–124. Herbert, R. J., Vilhar, B., Orchard, C. B., Rogers, H. J., Davies, M. S. and Francis, D. (2001). Ethylene induces cell death at particular phases of the cell cycle in the tobacco TBY-2 cell line. Journal of Experimental Botany 52, 1615–1623. Hirao, A., Kong, Y. Y., Matsuoka, S., Wakeham, A., Ruland, J., Yoshida, H., Liu, D., Elledge, S. J. and Mak, T. W. (2000). DNA damage-induced activation by the checkpoint kinase, Chk2. Science 287, 1824–1827. Hutchins, R. A., Dikovskaya, D. and Clarke, P. R. (2002). Dephosphorylation of the inhibitory phosphorylation site S287 in Xenopus Cdc25 by protein phosphatase-2A is inhibited by 14-3-3 binding. FEBS Letters 528, 267–271. Inze, D. (ed.) (2000). The plant cell cycle. Plant Molecular Biology 43 special edition, pp. 244, Kluwer Academic Publishers, Dordrecht, The Netherlands. Jabs, T., Dietrich, R. A. and Dangl, J. L. (1996). Initiation of runaway cell death in an Arabidopsis mutant by extracellular superoxide. Science 273, 1853–1856. Jackson, M. B., Fenning, T. M., Drew, M. C. and Saker, L. R. (1985). Stimulation of ethylene production and gas space (aerenchyma) formation in adventitious roots of Zea mays L. by small particle pressures of oxygen. Planta 165, 486–492. Jacobson, M. D., Burne, I. F. and Raff, M. C. (1994). Programmed cell-death and bcl-2 protection in the absence of the nucleus. EMBO Journal 13, 1899–1910. Jacobson, M. D., Weil, M. and Raff, M. C. (1997). Programmed cell death in animal development. Cell 13, 1899–1910. Jansen, M. A. K., Gaba, V. and Greenberg, B. M. (1998). Higher plants and UV-B radiation: balancing damage, repair and acclimation. Trends in Plant Science 3, 131–135. Jiang, C. -Z., Yee, J., Mitchell, D. L. and Britt, A. B. (1997). Photorepair mutants of Arabidopsis. Proceedings of the National Academy of Sciences USA 94, 7441–7445. Jin, P., Gu, Y. and Morgan, D. O. (1996). Role of inhibitory CDC2 phosphorylation in radiation-induced G2 arrest in human cells. Journal of Cell Biology 134, 963–970.

176

D. FRANCIS

John, P. C. L. (1996). The plant cell cycle: conserved and unique features in mitotic control. In, Progress in Cell Cycle Research (L. Meijer, S. Guidet, L. Vogel, eds.). pp. 59–72, Plenum Press, New York, USA. Joubes, J., Chevalier, C., Dudits, D., Heberle-Bors, E., Inze, D., Umeda, M. and Renaudin, J. -P. (2000). CDK-related protein kinases in plants. Plant Molecular Biology 43, 607–620. Jones, A. M. and Dangl, J. L. (1996). Logjam at the styx, porgrammed cell death in plants. Trends in Plant Science 1, 114–119. Kastan, M. B., Onyekwere, O., Sidransnsky, D., Vogelstein, B. and Graig, R. W. (1991). Participation of p53 protein in the cellular response to DNA damage. Cancer Research 51, 6304–6311. Kasten, M. M. and Giordano, A. (1998). pRB and the Cdks in apoptosis and the cell cycle. Cell Death and Differentiation 5, 132–140. Kazan, K., Murray, F. R., Goulter, K. C., Lewellyn, D. J. and Manners, J. M. (1998). Induction of cell death in transgenic plants expressing a fungal glucose oxidase. Molecular Plant-Microbe Interactions 11, 555–562. Keegan, K. S., Holtzman, D. A., Plug, A. W., Christenson, E. R., Brainerd, E. E., Flaggs, G., Bentley, N. J., Taylor, E. M., Meyn, M. S. and Moss, S. B. (1996). The Atr and Atm protein kinases associate with different sites along meotically pairing chromosomes. Genes and Development 10, 2423–2437. Kerr, J. F. R., Wyllieb, A. H. and Currie, A. R. (1972). Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. British Journal of Cancer Research 26, 239–257. Krek, W. and Nigg, E. A. (1991). Differential phosphorylation of vertebrate p34cdc2 at the G1/S and G2/M transitions of the cell cycle: identification of major phosphorylation sites. EMBO Journal 10, 305–316. Kuerbitz, S. J., Plunkett, B. S., Walsh, W. V. and Kastan, M. B. (1992). Wild-type p53 is a cell cycle checkpoint determinant following radiation. Proceedings of the National Academy of Sciences, U.S.A. 92, 8876–8880. Kuo, A. L., Cappelluti, S., CervantesCervantes M., Rodriguez, M. and Bush, D. S. (1996). Okadaic acid, a protein phosphates inhibitor, blocks calcium changes, gene expression, and cell death induced by gibberellin in wheat aleurone cells. Plant Cell 8, 259–269. Kuromori, T. and Yamamoto, M. (2000). Members of the Arabidopsis 14-3-3 gene family trans-complement two types of defects in fission yeast. Plant Science 158, 155–161. Lam, E., Fukuda, H. and Green berg, J. (eds.) (2000). Programmed cell death in higher plants. Plant Molecular Biology 44 special edition, 200 pp. Kluwer Academic Publishers, Dordrecht, The Netherlands. Lane, D. P. (1992). p53, Guardian of the genome. Nature 358, 15–16. Lee, J., Kumagai, A. and Dunphy, W. G. (2001). Positive regulation of Wee1 by Chk1 and 14-3-3 proteins. Molecular Biology of the Cell 12, 551–563. Lindsay, H. D., Griffiths, D. J., Edwards, R. J., Christensen, P. U., Murray, J. M., Osman, F., Walworth, N. and Carr, A. M. (1998). S-phase specific activation of Cds1 kinase defines a subpathway of the checkpoint response in Schizosaccharomyces pombe. Genes and Development 12, 382–395 Lopez-Girona, A., Furnari, B., Modesert, O. and Russell, P. (1999). Nuclear localisation of Cdc25 is regulated by DNA damage and a 14-3-3 protein. Nature 397, 172–175 Lopez-Girona, A., Kanoh, J. and Russell P. (2001). Nuclear exclusion of Cdc25 is not required for the DNA damage checkpoint in fission yeast. Current Biology 11, 50–54.

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

177

Lundgren, K., Walworth, N., Booher, R., Dembski, M., Kirschner, M. and Beach, D. (1991). mik1 and wee1 cooperate in the inhibitory tyrosine phosphorylation of cdc2. Cell 54, 1111–1122. Matsuoka, S., Hunag, M. and Elledge, S. J. (1998). Linkage of ATM to cell cycle regulation by the Chk2 kinase. Science 282, 1893–1897. McCabe, P. F. and Leaver, C. J. (2000). Programmed cell death in cell cultures. Plant Molecular Biology 44, 359–368. McGowan, C. H. and Russell, P. (1995). Cell cycle regulation of human WEE1. EMBO J 14, 2166–2175. Mengers, M. and Murray, J. A. H. (2002). Synchronous Arabidopsis suspension cultures for analysis of cell cycle gene activity. Plant Journal 30, 203–212. Me´sza´ros, T., Misklczi, P., Ayaydin, F., Pettko´-Szandtner, A., Peres, A., Magyar, Z., Horva´th, G. B., Bako´, L., Fehe´r, A. and Dudits, D. (2000). Multiple cyclindependent kinase complexes and phosphatases control G2/M progression in alfalfa cells. Plant Molecular Biology 43, 595–605. Meyn, M. S. (1999). Ataxia telangiectasia cancer and the pathobiology of the ATM gene. Clinical Genetics 55, 289–304. Michael, W. M. and Newport, J. (1998). Coupling of mitosis to the completion of S phase through Cdc34-mediated degradation of Wee1. Science 282, 1886–1889. Mikhailov, A., Cole, R. W. and Rieder, L. (2002). DNA damage during mitosis in human cells delays the metaphase/anaphase transition via the spindleassembly checkpoint. Current Biology 12, 1797–1806. Mittler, R., Simon, L. and Lam, E. (1997). Pathogen-induced programmed cell death in tobacco. Journal of Cell Science 110, 1333–1344. Morgan, D. O. (1997). Cyclin-dependent kinases: engines, clocks and microprocessors. Annual Review of Cell and Developmental Biology 13, 261–291. Mourelatou, M. (2001). Proliferation versus differentiation in the Zinnia elegans model system. Ph.D. Thesis. University of East Anglia, U.K. Mueller, P. R., Coleman, T. R., Kumagai, A. and Dunphy, W. G. (1995). Cloning and functional-characterization of myt1, a membrane-associated inhibitory kinase that phosphorylates cdc2 on both thr14 and tyr15. Science 270, 86–90. Mundt, K. E., Porte, J., Muray, J. M., Brikos, C., Christensen, P. U., Caspari, T., Hagan, I. M., Millar, J. B. A., Simanis, V., Hofman, K. and Carr, A. M. (1999). The COP9/signalosome complex is conserved in fission yeast and has a role in S phase. Current Biology 9, 1427–1430. Murray, J. A. H., Doonan, J., Riou-Khamlichi, C., Meijer, M. and Okenfull, E. A. (2000). G1 cyclins and the regulation of the G1/S transition. In ‘The Plant Cell Cycle and its Interfaces’ (D. Francis, ed.). pp. 19–42. Sheffield Academic Press, Sheffield, U.K. Murray, J. A. H., Freeman, D., Greenwood, J., Huntley, R., Makkerh, J., RiouKhamlichi, C., Sorrell, D. A., Cockcroft, C., Carmichael, J. P., Soni, R. and Shah, Z. H. (1998). Plant D cyclins and retinoblastoma protein homologues. In Plant Cell Division. pp. 99–127 (D. Francis, D. Dudits and D. Inze, eds.). Portland Press Ltd., London. Nagata, T. N., Nemoto, Y. and Hasezewa, S. (1992). Tobacco BY-2 cell line as the ‘HeLa’ cell line in the cell biology of higher plants. International Review of Cytology 132, 1–30. Neill, S. J., Desikan, R., Clare, A., Hurst, R. D. and Hancock, J. T. (2002). Hydrogen peroxide and nitric oxide as signalling molecules in plants. Journal of Experimental Botany 53, 1237–1247.

178

D. FRANCIS

Nurse, P. (1990). Universal control mechanisms regulating onset of mitosis. Nature 344, 503–508. Nurse, P. (2001). Cyclin-dependent kinases and cell cycle control. Nobel Lecture, December 9th, 2001. http://www.nobel.se/medicine/laureates/2001/nurselecture.html Nurse, P. and Bisset, Y. (1981). Gene required on G1 for the commitment to the cell cycle in G2 for control in mitosis in fission yeast. Nature 292, 558–560. O’Connell, M. J., Raleigh, J. M., Verkade, H. M. and Nurse, P. (1997). Chk1 is a wee1 kinase in the G2 DNA damage checkpoint inhibiting cdc2 by Y15 phosphorylation. EMBO Journal 16, 545–554. O’Farrell, P. H. (2001) Triggering the all-or-nothing switch into mitosis. Trends in Cell Biology 11, 512–519. Overmyer, K., Tuominen, H., Kettunen, R., Betz, C., Langebartels, C., Sanderann, H. and Kangasja¨rvi, J. (2000). Ozone-sensitive Arabidopsis rcd1 mutant reveals opposite roles for ethylene and jasmonate signalling pathways in regulating superoxide-dependent cell death. Plant Cell 12, 1849–1862. Paulovich, A. G., Toczyski, D. P. and Hartwell, L. H. (1997). When checkpoints fail. Cell 88, 315–321. Peng, C. Y., Graves, P. R., Thoma, R. S., Wu, Z., Shaw, A. S. and Piwcnica-Worms, H. (1997). Mitotic and G2 checkpoint control: regulation of 14-3-3 protein binding by phosphorylation of Cdc25 on serine-216. Science 216, 1501–1505. Pontier, D., Gan, S., Amasino, R. M., Roby, D. and Lam, E. (1999). Markers of hypersensitive response and senescence show distinct patterns of expression. Plant Molecular Biology 39, 1243–1255. Raff, M. C. (1992). Social controls on cell-survival and cell death. Nature 356, 397–400. Raleigh, J. M. and O’Connell, M. J. (2000). The G(2) DNA damage checkpoint targets both Wee1 and Cdc25. Journal of Cell Science 113, 1727–1736. Reichfeld, J. P., Vernoux, T., Lardon, F., Van Montague, M. and Inze, D. (1999). Specific checkpoints regulate plant cell cycle progression in response to oxidative stress. Plant Journal 17, 647–656. Rhind, N. and Russell, P. (1998). Mitotic DNA damage and replication checkpoints in yeast. Current Biology in Cell Biology 10, 749–758. Rhind, N. and Russell, P. (2000). Chk1 and Cds1: linchpins of the DNA damage and replication checkpoint pathways. Journal of Cell Science 113, 3896–3899. Rothblum-Oviatt, C. J., Ryan, C. E. and Piwcnica-Worms, H. (2001). 14-3-3 binding regulates catalytic activity of human Wee1 kinase. Cell Growth and Differentiation 12, 581–589. Rowley, R., Hudson, J. and Young, P. G. (1992). The Wee1 protein is required for radiation-induced mitotic delay. Nature 356, 353–355. Rozema, J., van de Staaij, J., Bjo¨rn, L. O. and Caldwell, M. (1997). UV-B as an environmental factor in plant life: stress and regulation. Trends in Ecology and Evolution 1, 22–28. Russell, P. and Nurse, P. (1986). cdc25þ functions as an inducer in the mitotic control of fission yeast. Cell 45, 145–153. Russell, P. and Nurse, P. (1987). Negative regulation of mitosis by wee1þ, a gene encoding a protein kinase homolog. Cell 49, 559–567. Ryerson, D. E. and Heath, M. C. (1996). Cleavage of nuclear DNA into oligonucleosomal fragments during cell death induced by fungal infection or by abiotic treatments. Plant Cell 8, 393–402.

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

179

Sanchez, Y., Wong, C., Thoma, R. S., Wu, R. Q., Piwnica Worms, H., Elledge, S. J. (1999). Conservation of the Chk1 checkpoint pathway in mammals: Linkage of DNA damage to Cdk regulation through Cdc 25. Science 277, 1497–1501. Savitsky, K., Bar-Shira, A., Gilad, S., Rotman, G., Ziv, Y., Vanagaite, L., Tagle, D. A., Smith, S. and Uziel, et al., (1995) A single ataxia telangiectasia gene with a product similar to PI-3 kinase. Science 268, 1749–1753. Schwartz, L. M., Smith, S. W., Jones, M. E. E. and Osburne, B. A. (1993). Do all programme cell deaths occur via apoptosis? Proceedings of the National Academy of Sciences, USA 90, 980–984. Shieh, S. Y., Ahn, J., Tamai, K., Taya, Y. and Prives, C. (2000). The human homologs of checkpoint kinases Chk1 and Cds1 (Chk2) phosphorylate p53 at multiple DNA damage-inducible sites. Genes & Development 14, 289–300. Shininger, T. L. (1979). The control of vascular development. Annual Review of Plant Physiology 30, 313–337. Simanis, V. and Nurse, P. (1986). The cell control gene cdc2þ of fission yeast encodes a protein kinase potentially regulated by phosphorylation. Cell 45, 261–268. Smart, C. M. (1994). Gene expression during leaf senescence. New Phytologist 126, 419–448. Smith, G. C. M. and Jackson, S. P. (1999). The DNA-dependent protein kinase. Genes and Development 13, 916–934. Sorrell, D. A., Marchbank, A., McMahon, K., Dickinson, J. R., Rogers, H. J. and Francis, D. (2002). A WEE1 homologue from Arabidopsis thaliana. Planta 215, 518–522. Stals, H., Casteels, P., Van Montague, M. and Inze, D. (2000). Regulation of cyclindependent kinases in Arabidopsis thaliana. Plant Molecular Biology 43, 583–593. Steller, H. (1995) Mechanisms and genes of cellular suicide. Science 267, 1445–1449. Stewart, B. W. (1994). Mechanisms of apoptosis: integration of genetic, biochemical and cellular interactions. Journal of the National Cancer Institute 86, 1286–1296. Sugiyama, M. and Komamine, A. (1987). Relationship between DNA synthesis and cytodifferentiation to tracheary elements. Oxford Surveys of Plant Molecular and Cell Biology 4, 343–346. Sun, Y., Dilkes, B. P., Zhang, C., Dante, R. A., Carneiro, N. P., Lowe, K. S., Jug, R., Gordon-Kamm, W. J. and Larkins, B. A. (1999). Characterization of maize (Zea mays L.) Wee1 and its activity in developing endosperm. Proceedings of the National Academy of Sciences, USA 96, 4180–4185. Suzuki, K., Kodama, S. and Watnabe, M. (1999). Recruitment of ATM protein to double stranded DNA irradiated with ionizing radiation. Journal of Biological Chemistry 274, 25571–25575. Tanaka, H., Arakawa, H., Yamaguchi, T., Shiraishi, K., Fukuda, S., Matsui, K., Takei, Y. and Nakamura, Y. (2000). A ribonucleotide reductase gene involved in a p53-dependent cell-cycle checkpoint for DNA damage. Nature 404, 42–49. Tanaka, Y., Makishima, T., Sasabe, M., Ichinose, Y., Shiraishi, T., Nishimoto, T. and Yamada, T. (1997). dad-1, a putative programmed cell death suppressor gene in rice. Plant and cell Physiology 38, 379–383. Tepfer, S. S., Nougarede, A. and Rondet, P. (1981). Seasonal studies of vegative buds of Helianthus tuberosus: concentration of nuclei hel in phase G1 during winder dormancy. Canadian Journal of Botany 59, 1918–1927.

180

D. FRANCIS

Umeda, M., Bhalerao, R. P., Schell, J., Uchimaya, H. and Koncz, C. (1998). A distinct cyclin-dependent kinase-activating kinase of Arabidopsis thaliana. Proceedings of the National Academy of Sciences U.S.A. 95, 5021–5026. Vranova, E., Inze, D. and Van Breusegem, F. (2002). Signal transduction during oxidative stress. Journal of Experimental Botany 53, 1227–1236. Vranova, E., Langebartels, C., Van Montague, M., Inze, D. and Van Camp, W. (2000). Oxidative stress, heat shock and drought differentially affect expression of a tobacco protein phosphatase 2C. Journal of Experimental Botany 51, 1763–1764. Wang, H., Fowke, L. C. and Crosby, W.L. (1997). A plant cyclin-dependent kinase inhibitor gene. Nature 386, 451. Wang, M., Oppedijk, B. J., Lu, X., van Duijn, B. and Schilperoort, R. A. (1996). Apoptosis in barley aluerone during germination and its inhibition by abscisic acid. Plant Molecular Biology 32, 1125–1134. Wei, N. and Deng, X-W. (1999). Making sense of the COP9 signalosome. Trends in Genetics 15, 98–103. Weil, M., Jacobsen, M. D., Coles, H. R. S., Davies, T. J., Gardner, R. L., Raff, K. D. and Raff, M. C. (1996). Constitutive expression of the machinery for programmed cell death. Journal of Cell Biology 133, 1053–1059. White, K., Tahaoglu, E. and Steller, H. (1996). Cell killing by the Drosophila gene reaper. Science 271, 805–807. Wright, J. A., Keegan, K. S., Herendeen, D. R., Bentley, N. J., Carr, A. M., Hoekstra, M. F. and Concannon, P. (1998). Protein kinase mutants of human ATR increase sensitivity to UV and ionizing radiation and abrogate cell cycle checkpoint control. Proceedings of the National Academy of Sciences, U.S.A., 95, 7445–7450. Wu, H. and Cheung, A. Y. (2000). Programmed cell death in plant reproduction. Plant Molecular Biology 44, 267–281. Wu, Z., Piwicninca-Worms, H. and Elledge, S. J. (1997). Conservation of the Chk1 checkpoint pathway in mammals: linkage of DNA damage to Cdk regulation through Cdc25. Science 277, 1497–1501. Yamamoto, A., Guacci, V. and Koshland, D. (1996). Pds1p, an inhibitor of anaphase in budding yeast, plays a critical role in the APC and checkpoint pathway(s). Journal of Cell Biology 133, 99–110. Yano, A., Suzuki, K. and Shinshi, H. (1999). A signalling pathway, independent of the oxidative burst, that leads to hypersensitive cell death in cultured tobacco cells includes a serine protease. Plant Journal 18, 105–109. Yano, A., Suzuki, K., Uchimiya, H. and Shinshi, H. (1998). Induction of hypersensitive cell death by a fungal protein in cultures of tobacco cells. Molecular Plant-Microbe Interactions 11, 115–123. Young, T. E. and Gallie, D. R. (1999). Analysis of programmed cell death in wheat endosperm reveals differences in endosperm development between cereals. Plant Molecular Biology 39, 915–926. Young, T. E. and Gallie, D. R. (2000). Programmed cell death during endosperm development. Plant Molecular Biology 44, 283–301. Zeng, Y., Forbes, K. C., Wu, Z., Moreno, S., Piwnica-Worms, H. and Enoch, T. (1998). Replication checkpoint requires phosphorylation of the phosphatase Cdc25 by Cds1 or Chk1. Nature 395: 507–510. Zhang, K., Letham, D. S. and John, P. C. L. (1996). Cytokinin controls the cell cycle at mitosis by stimulating the tyrosine dephosphorylation and activation of p34cdc2-like histone kinase. Planta 200, 2–12.

THE INTERFACE BETWEEN THE CELL CYCLE AND PCD

181

Zhao, J., Morozova, N., Williams, L., Libs, L., Avivi, Y. and Grafi, G. (2001). Two phases of chromatin decondensation during dedifferentiation of plant cells: Distinction between competence for cell–fate switch and a commitment for S phase. Journal of Biological Chemistry 276, 22772–22786 Zhou, Y., Fowke, L. C. and Wang, H. (2002). Plant CDK inhibitor studies of interactions with cell cycle regulators in the yeast two-hybrid system and functional comparisons in transgenic Arabidopsis plants. Plant Cell Reporter 20, 967–975.

This Page Intentionally Left Blank

The Importance of Extracellular Carbohydrate Production by Marine Epipelic Diatoms

GRAHAM J. C. UNDERWOOD1 AND DAVID M. PATERSON2 1

Department of Biological Sciences, University of Essex, Essex, UK 2 Gatty Marine Laboratory, University of St Andrews, Fife, Scotland, UK

I II III

IV

V

VI

VII

Marine Benthic Diatoms and Biofilms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Motility and EPS Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Production of Colloidal Carbohydrate and EPS by Diatoms . . . . . . . . . . A. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Carbohydrate Production in Microphytobenthos in Culture . . . . . . . C. Carbohydrate Production in Microphytobenthos in situ . . . . . . . . . . . . Comparative Biochemistry of Diatom EPS . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Production Pathways of EPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Biochemical Composition of Diatom EPS . . . . . . . . . . . . . . . . . . . . . . . . . C. Comparison of Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Natural Patterns of Colloidal Carbohydrate and EPS Distribution . . . . A. Horizontal Distribution of EPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Depth Distribution of EPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Seasonal Variation in EPS Concentrations . . . . . . . . . . . . . . . . . . . . . . . . D. Tidal Variation in Colloidal Carbohydrate Concentrations . . . . . . . . . E. Loss of Carbohydrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Impact of Environmental Stresses on Exudation in Field . . . . . . . . . . Sediment-stability, Flocculation and EPS Production . . . . . . . . . . . . . . . . . A. Measuring the Influence of EPS on Physical Dynamics . . . . . . . . . . . . B. The Problem of the Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. The Influence of EPS and Organisms on Sediment Dynamics . . . . . . D. The EPS Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Areas of Further Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Advances in Botanical Research Vol. 40 incorporating Advances in Plant Pathology ISBN 0-12-005940-1

184 187 190 190 192 198 203 203 204 210 216 216 218 219 220 221 222 223 224 225 225 228 230 231 231

Copyright  2003 Elsevier Ltd All rights of reproduction in any form reserved

184

G.J.C. UNDERWOOD AND D.M. PATERSON

ABSTRACT Soft-sediment habitats in intertidal and shallow subtidal marine ecosystems frequently support extensive populations of benthic microalgae (microphytobenthos). These algal assemblages are dominated by species of motile benthic diatoms and form biofilms, a matrix of cells, sediments and extracellular polymeric substances (EPS), that create a complex microhabitat and act to stabilise sediments. Diatom EPS consists of a relatively undefined complex mixture of proteins, proteoglycans and carbohydrates. This complexity causes problems in extracting and analysing EPS and in the intercomparison of studies. This chapter reviews our current knowledge on the production rates, patterns and composition of benthic diatom EPS in culture and field studies. Production patterns are dynamic, changing with cell growth phase, photosynthesis and irradiance, nutrient conditions, and are also linked to endogenous cell rhythms. Meta-analysis of published monosaccharide composition data identified at least four major types of EPS produced by benthic diatoms, with varying patterns of production and composition. It is clear that more detailed research on the structural and physical properties of EPS are needed to understand its role in the environment. The natural occurrence of EPS is closely linked to diatom biomass, a pattern consistent over both macro (km) and micro (m) scales. EPS is lost from sediments by various routes; solubilisation and removal by overlying water, bacterial degradation and consumption by deposit-feeding invertebrates. Work is needed to quantify these pathways and clarify the importance of EPS in coastal carbon cycles. Diatom EPS is a widely cited mechanism for increasing sediment stability and stabilisation by biofilms is well described. However, data are not consistent and developments in our knowledge of the structure and function of EPS are needed to explain how EPS binds and interacts within the sediment-biofilm matrix and affects the rheology of sediment.

I

MARINE BENTHIC DIATOMS AND BIOFILMS

The diatoms (Bacillariophyta) are the most diverse group of algae containing as many as 200,000 extant species (Mann and Droop, 1996; Mann, 1999). These unicellular algae are the major primary producers in many aquatic habitats and it has been estimated that diatoms are responsible for approximately 40% of the total global primary production in marine systems (Falciatore and Bowler, 2002; Medlin, 2002). The diatoms belong to the division Heterokontophyta and are divided into two main taxonomic groups based on the shape of the cell; the centric diatoms (radially symmetrical) and the pennate diatoms (bilaterally symmetrical) (Round et al., 1990; Van Den Hoek et al., 1995). The basic cell structure in nearly all cases can be described as a silica pill box, with 2 halves (valves), bound together with girdle bands (Fig. 1). Centric diatoms are predominantly planktonic, while pennate diatoms can be planktonic, found attached to surfaces, or growing on and moving through sediments (Van den Hoek et al., 1995).

EXTRACELLULAR CARBOHYDRATE PRODUCTION

185

Fig. 1 A: Features of the silica frustule of a stylised Naviculoid epipelic diatom, showing the upper epivalve and lower hypovalve, the raphe slit (raphe) and the bands of silica (copulae) which join the valves. B: Valve (upper) and girdle (side) view of a naviculoid diatom.

Many of the pennate diatoms possess the ability to move. The mechanism of movement in diatoms is unique for microbial cells and relies on the extrusion of mucilage through a slit in the surface of the silica frustule (cell wall). This slit, known as the raphe, may be present on only a single valve of the frustule (monoraphid) or on both valves (biraphid). The ability to move allows diatoms to inhabit depositional ecosystems dominated by sandy and muddy sediments. In shallow systems (estuaries, subtidal areas, tropical mangrove and coral reef flat environments), wherever light can penetrate with sufficient radiance to support photosynthesis, benthic microalgae, particularly diatoms, are important constituents of the biological community (MacIntyre et al., 1996). In estuarine systems, benthic microalgae, also known as ‘microphytobenthos’ (MPB), can contribute up to 50% of the total autotrophic production and up to 33% of total estuarine carbon inputs in some systems (Underwood and Kromkamp, 1999; Cahoon, 1999). While numerous algal groups are present within the microphytobenthos, the most abundant group of benthic microalgae in fine (cohesive) sediments are epipelic (living in mud) diatoms (Admiraal, 1984; Underwood and Kromkamp, 1999; Paterson and Hagerthey, 2001). Populations of epipelic diatoms are diverse (at least 200 spp.) and are characterised by motile, biraphid species. The abundance of cells can be sufficient to colour the sediment surface and produce thick biofilms (Fig. 2). Microphytobenthic biofilms can have high rates of photosynthesis and a significant proportion of their photo-assimilated carbon is released into the environment as extracellular carbohydrates. This material collects in the surficial sediments (Taylor and Paterson, 1998) where it can be broken

186

G.J.C. UNDERWOOD AND D.M. PATERSON

Fig. 2 A. Natural assemblage of epipelic diatoms visualised by light microscopy Scale bar ¼ 40 mm B. Cell of Nitzschia epithemioides showing the presence of mucilage (m), stained with acidic alcian blue, surrounding the cell and present in the culture media. Scale bar ¼ 20 mm. C. Biofilm dominated by Cylindrotheca closterium on mudflats in the Severn Estuary, U.K. Note the cohesive nature of the film. Scalpel ¼ 15 cm. D. Extensive diatom biofilms in the Colne estuary, Essex, U.K. showing the extensive area covered in favourable conditions (with Dr G.J.C Underwood and Dr M. Consalvey).

down by bacteria, washed away and/or buried with the sediment bed (Underwood and Smith, 1998b; Yallop et al., 2000). Some of this carbohydrate material consists of low molecular weight exudates, but other components are polymeric in nature, forming mucilage. High concentrations of mucilage can be present in biofilms, where a matrix of cells, sediment and mucilage forms a coherent structure on the sediment surface (Yallop et al., 1994). The biofilm itself is comprised of a complex mixture of polymeric compounds produced by the microorganisms that form the system, including bacteria, algae and fungi. The generic name given to these exudates are extracellular polymeric substances (EPS), but most research has been carried out on the dominant carbohydrate component of this polymeric material (Underwood et al., 1995). These biofilms can act as a protective ‘skin’ within the surface layers of the sediment, which significantly alters the erosion and deposition characteristics of the system (Paterson and Black, 1999; Paterson, 2001).

EXTRACELLULAR CARBOHYDRATE PRODUCTION

187

It is known that diatom mucilages are rich in polysaccharides and proteoglycans and are secreted through the channels and pores in the diatom frustule. In some benthic species, these compounds produce structures (tubes, pads and stalks) that are used for attachment to surfaces (Fig. 3) and perhaps help to create favourable micro-environments in which the cells live (Daniel et al., 1987; Decho 1990; Hoagland et al., 1993). Such structures often contribute to biofouling problems (Hoagland et al., 1993; Lewis et al., 2002). Epipelic diatoms do not produce permanent structures but secrete large quantities of extracellular mucilages that are involved in motility (Hoagland et al., 1993; Lind et al., 1997; Wetherbee et al., 1998). However, the total amount of EPS produced by benthic diatoms is far in excess of that required for movement (Edgar and Pickett-Heaps, 1984), but it is hypothesised that this additional EPS plays an important and multifunctional role in biofilm ecology (Decho, 2000). EPS and other carbohydrate-rich exudates in the aquatic environment are ecologically significant since they can be utilized by bacteria, meio- and macro-fauna as a carbon source (Decho, 1990, 2000; van Duyl et al., 1999, 2000; Middleburg et al., 2000), and exopolymers can increase the stability of sediments (Paterson and Black, 1999; Widdows et al., 2000). This latter process is especially important in intertidal situations where the mudflat habitat is often sensitive to erosion. A wide variety of other properties have been attributed to EPS; such as protection from desiccation, as a polyelectrolyte to bind organic and inorganic nutrients, as a matrix for exoenzymes (e.g. phosphatases), as a carbon reserve (heterotrophy) (Decho 1990; Staats et al., 2000a, b). Thus extracellular carbohydrate production by diatoms is a significant route by which benthic diatom primary production enters the trophic web and influences the physical environment. In this review, we will describe the existing data on the rates and patterns of production of extracellular carbohydrates by benthic marine diatoms, the chemical composition of this material, and the evidence for its important role in influencing sediment properties.

II

MOTILITY AND EPS PRODUCTION

Motility is an essential adaptation for photosynthetic organisms in depositional environments, allowing cells to migrate into the very narrow illuminated (photic) zone near the surface (Kelly et al., 2001; Paterson and Hagerthey, 2001) after periods of sediment mixing or deposition. Epipelic diatom biofilms show endogenous rhythms of migration with cells migrating to and from the surface in response to environmental cues (Seroˆdio et al.,

188

G.J.C. UNDERWOOD AND D.M. PATERSON

Fig. 3 Scanning electron microscopy (SEM) images of diatom cells and structures. a: Low-temperature SEM (LTSEM) of a natural surface biofilm from the Eden estuary, Scotland (Scale bar ¼ 80 mm). b. LTSEM showing a vertical fracture of the sediment thought the surface biofilm region (Scale bar ¼ 40 mm). c–f. The surface of selected diatoms after the removal of the organic coating that normally covers the cells. The ornamentation of the silica valves is clear showing c: punctae (bar marker ¼ 10 mm, d, fine striae (bar marker ¼ 5 mm), e, fine punctae (bar marker ¼ 5 mm), f: robust striae on Tryblionella sp., with a band of EPS associated with the laterally displaced raphe slit. (bar marker ¼ 10 mm). g; LTSEM of mucilage strand associated with the raphe of a Gyrosigma sp. Note that the surface patterning is occluded by an organic coating. (Bar marker ¼ 5 mm). h. LTSEM of pair of epiphytic diatoms (Gomphonema sp.) attached by a mucilage stalk to their host plant. Note that although underlying pattern of striae can be seen on the surface, the pores and channels are occluded by an organic coating. (Bar marker ¼ 30 mm). i. LTSEM of a tube forming diatom. These diatoms live within a mucilaginous tube generated by the cells themselves. The tube appears to be branched and some diatoms can be seen adhering to the outside of the tube (Bar marker ¼ 30 mm).

EXTRACELLULAR CARBOHYDRATE PRODUCTION

189

1997; Smith and Underwood, 1998; Underwood and Kromkamp, 1999; Consalvey, 2002). Two different theories explaining diatom motility have been suggested in the literature over the past 40 years. One is the hydration theory (Jarosch, 1962; Gordon and Drum, 1970; Gordon, 1987), in which energy produced by hydration of adhesive mucilages provides the motive force to move the diatom cell. The second, and now more widely accepted theory, was put forward by Edgar and Pickett-Heaps (1984). This model was based on the work of Edgar and Zavortink (1983) showing actin bundles in the cell underlying the raphe region of the frustule, and electron microscopy showing the presence of charged polysaccharide mucilage in the raphe of cells exhibiting motility (Edgar and Pickett-Heaps, 1982). In this model mucilage is secreted through the raphe, hydrates and swells, and adheres to the substratum. Cell movement is then generated as the mucilage associated with trans-membrane complexes in the plasma membrane is moved along the line of the raphe by intracellular microfilament bundles. This theory has received substantial support in recent years, with the observation (by electron microscopy) of mucilage in the raphe canal (Lind et al., 1997; Wang et al., 2000). Paterson (Paterson 1986; Paterson et al., 1986) first used low-temperature electron microscopy to visualise natural diatom assemblages and show mucilage being extruded through the diatom raphe (Fig. 3), while Wang et al. (2000) using a combination of high resolution electron microscopic and immuno-cytochemistry, produced images of EPS ‘trails’ secreted by motile Achnanthes longipes and found similarities in the composition of polymers involved in motility and EPS incorporated into more permanent attachment structures. Using Atomic Force Microscopy (AFM), Higgins et al. (2002) showed that the biraphid diatom, Craspedostauros australis (previously identified as Stauroneis decipiens), produced a soft mucilage which covered the cell from pore openings in the cell frustule. Strands of a separate, highly-adhesive mucilage, were secreted from the raphe. Trails of similar material, appearing as a continuous, swollen ridge of material with diffuse mucilage on either side, have been measured using AFM behind diatoms of the genus Pinnularia moving over glass slides (Higgins et al., 2000). Application of specific actin- or myosin-inhibitor drugs can reversibly inhibit diatom movement, with cell motility returning within 5 s of drugs being removed (Poulsen et al., 1999). Recent work using antibodies indicates a key role for proteoglycans and glycoproteins as the functional components in diatom motility. Proteoglycan epitopes have been identified by labeling with monoclonal antibodies (Lind et al., 1997; Wustman et al., 1998). The importance of these proteoglycans in gliding is shown by the inhibition of adhesion and motility in the marine biraphid

190

G.J.C. UNDERWOOD AND D.M. PATERSON

diatom Stauroneis decipiens when monoclonal antibodies were added (Lind et al., 1997). Proteoglycans are present in extremely low amounts within diatom EPS which also contains substantial amounts of polysaccharides, uronic acids and sulphated sugars. The importance of these different components in the functional properties of EPS is a topic of debate (Wetherbee et al., 1998).

III PRODUCTION OF COLLOIDAL CARBOHYDRATE AND EPS BY DIATOMS A. TERMINOLOGY

Research into the role of diatom extracellular mucilages in ecology is complicated by a highly variable and inconsistent range of terminology. This is due, in part, to the different scientific disciplines that embrace the study of microbial EPS, in particular polysaccharide biochemists, medical researchers, algal physiologists, ecologists and sedimentologists. Each discipline has introduced or adapted terms from other studies, not necessarily using the same degree of rigour, and this had led to confusion as to the nature and form of the EPS that a particular study is investigating (e.g. Laspidou and Rittmann, 2002). This problem is compounded by the variable nature of diatom mucilage itself. These exudates do not fall neatly into different ‘types’ but range along a continuum from rigid stalks and pads, through mucilages in various stages of hydration (gels, slimes) to colloidal and dissolved carbohydrate components soluble in aqueous media. The tertiary state of the polymer is influenced by the chemical composition of the EPS as well as the interaction between molecules of EPS, the environment and surrounding polymers (Decho, 1990, 1994). Extraction of EPS from cultures and natural sediments is highly dependent on the protocol used, as well as physiological status of the cells and the biochemical properties of the carbohydrates. Thus differences between the fractions of EPS obtained by different extraction techniques are the combined influence of EPS composition and structure, environmental conditions and operational methodology. The challenge for algal EPS research is to define protocols for extracting fractions along this continuum of EPS types that are (a) consistent and (b) reflect the biological utility or relevance of the molecules. The physical structures produced by diatoms have been called stalks, pads and tubes, trails and extra-cellular matrix (ECM) (Hoagland et al., 1993; Wetherbee et al., 1998; Higgins et al., 2000; Wang et al., 2000). These are physically complex structures, with a broad range of biochemical constituents (Hoagland et al., 1993). Various techniques to harvest and

EXTRACELLULAR CARBOHYDRATE PRODUCTION

191

analyse tubes and stalks have been developed, including sequential extractions using various solvents, to remove different components (Hoagland et al., 1993; Wustman et al., 1997). Such approaches have not been applied to the amorphous mucilages produced by motile benthic diatoms. Many of the carbohydrate components in stalks are cross-linked, methylated or sulphated and different parts of stalks can have varying composition (Daniel et al., 1980; Wustman et al., 1997). This has been clearly shown using FITC-lectin staining for different sugars in the stalk-forming diatoms Achnanthes longipes and Cymbella cistula (Wustman et al., 1997). These structures appear to self-assemble after secretion of the polysaccharides from the cell (Wustmann et al., 1997; Wang et al., 2000). Stalk-forming diatoms are rarely found in epipelic habitats, though epipsammic and epiphytic taxa do produce adhesive pads to fix themselves to sand grains (Fig. 3). However, after periods of low tidal energy, tubeforming diatoms can be found among the sediment matrix, often belonging to the genera Berkeleya, Haslea, Parlibellus and Encyonema. These species live within a mucilage tube system (Fig. 3), but their carbohydrate dynamics have not been studied. A comprehensive review of stalk-forming diatoms is given by Hoagland et al. (1993). Studies of sediment-inhabiting diatoms generally use the term extracellular polymeric substance (EPS) to describe the mucilage materials (including glycoproteins) produced within biofilms. Even here though, there is confusion in the terminology. An attempt was made to standardise methods and terminology (Underwood et al., 1995). In this paper, an extraction protocol was described, in which the carbohydrate material that remained in the supernatant after extraction of a sample of cells or sediment as colloidal carbohydrate (Underwood et al., 1995; Smith and Underwood, 1998, 2000; Sutherland et al., 1998a; Taylor and Paterson, 1998; Staats et al., 1999, 2000). The material remaining in the pellet was termed bulk carbohydrate. The colloidal carbohydrate material can be further separated into low molecular weight (LMW) mono- and polysaccharides, and larger polymeric molecules (exopolymers or EPS) by precipitation in 70% ethanol (Decho 1990; Underwood et al., 1995). The quantity of colloidal carbohydrate obtained depends on the extraction method (fresh versus freeze-dried sediments) and generally increases with increasing temperature, incubation period and astringency of the extraction media (saline, NaOH, EDTA, acid) (Underwood et al., 1995; de Brouwer et al., 2002). Depending on the nature of the original sample, especially natural mud samples, it is important to realise that the precipitate derived is not pure EPS but may contain large quantities of impurities and salt and this must be considered. The relative ease of extraction of different components had led to additional

192

G.J.C. UNDERWOOD AND D.M. PATERSON

terms such as ‘loose’ (to describe material present in culture media after centrifugation of cells), bound, non-attached, and attached EPS (Underwood et al., 1995; Staats et al., 1999; de Brouwer and Stal (2002) and references therein). The relationship between these different fractions is unclear. More recently, molecular filters have been used to size-fractionate material from various aqueous extractions. This allows different molecular size thresholds to be defined (de Brouwer and Stal, 2001). A similar approach is to precipitate EPS from aqueous extracts using an alcohol series (Boulcott, 2001). These approaches result in a range of EPS fractions being obtained. All of these fractions have potentially different provenances that may relate to the physiology of the diatom and may have different properties in the environment. As more investigations of the dynamic nature of the biofilm-EPS matrix are carried out, it is becoming apparent that the methods we described in Underwood et al. (1995) are not ideal. The standard approach for a number of years has been to freeze-dry sediment before extraction of colloidal and EPS components. This has been extensively used in field studies. One of the major advantages of this technique is increased gravimetric accuracy compared to working with wet sediments. However, it is likely that freeze-drying may change both carbohydrate yield and polymer properties (Underwood et al., 1995; Perkins, pers. comm.). There is now more data to show that, as with bacteria (Kazy et al., 2002), numerous different types of EPS are produced by benthic diatoms (see Section IV). These issues are highly relevant, as biologically distinct EPS may become merged within extractions. In this review, we use the term ‘colloidal’ to define carbohydrate material present in either culture media or extracted from cells using water or saline extracts. The polymeric component of this material (cEPS) is the carbohydrate fraction precipitated out of colloidal extracts by using ethanol. Where other author’s extractions are compatible with our definition of these terms, we use them as such, otherwise they are prefaced by the authors terminology (bound, soluble, molecular size fraction) to indicate that different fractions are being considered.

B. CARBOHYDRATE PRODUCTION IN MICROPHYTOBENTHOS IN CULTURE

Detailed studies on the production of extracellular carbohydrates have been carried out on nine species of benthic marine diatom and in various estuarine and coastal habitats (Table I). Of these species, Cylindrotheca closterium is most commonly studied, partly due to the ease with which this species grows in culture. Some general features are common to all these studies.

TABLE I Hourly (h) and daily (d) rates of extracellular carbohydrate production (glucose equivalents) by cultures (C) of benthic marine diatoms and in undisturbed sediments supporting biofilms (S) either in situ or in laboratory mesocosms

Amphora exigua A. exigua Cylindrotheca closterium C. closterium C. closterium C. closterium C. closterium C. closterium

C C C

logarithmic logarithmic transition

colloidal EPS EPS

C C C C C

sol EPS bound EPS bound EPS EPS EPS

C. C. C. C.

closterium closterium closterium closterium

C C C C

stationary logarithmic stationary stationary log/station transition logarithmic logarithmic

C. closterium

C

C. closterium

C

Navicula perminuta C N. perminuta C N. perminuta C

increase light and dark 6.4 light period increase light period increase 10 h illum. period 6 h illum. period

20 7

de Brouwer and Stal, 2002 de Brouwer and Stal, 2002 de Brouwer and Stal, 2002 Staats et al., 2000 Staats et al., 2000

3,000 500

Boulcott, 2001 Boulcott, 2001 de Brouwer et al., 2002 Underwood and Smith, 1998a

4.7

31.5

Smith and Underwood, 2000

8.68

54.1

Smith and Underwood, 2000

2.1 1.3

Boulcott, 2001 Boulcott, 2001 Underwood and Smith, 1998a

max rates, 14:10 L:D

3.91

EPS

max rates, 14:10 L:D

colloidal

max rates, 14:10 L:D

max rates, 14:10 L:D

Boulcott, 2001 Boulcott, 2001 Staats et al., 1999

8.3 2.6 1.6

colloidal EPS EPS EPS

colloidal EPS EPS

14.4 8.7 0.93

2.59

163

108

193

log/station transition log/station transition log/station transition logarithmic logarithmic log/station transition

mg glucose ng glucose pg glucose Reference equiv. mg equiv. mg equiv. Chl a1 d1 Chl a1 h1 cell1 d1

EXTRACELLULAR CARBOHYDRATE PRODUCTION

Organism/field site C/S Growth phase Colloidal/EPS Extraction notes

(continued )

194

TABLE I Continued. C/S Growth phase colloidal/EPS extraction notes

N. perminuta

C

N. perminuta

C

Navicula salinarum Nitzschia frustulum Nz. frustulum Nz. frustulum Nitzschia sigma Nitzschia sigma Nitzschia sigma Nitzschia sp. Nitzschia sp. Nitzschia sp. Stauroneis amphioxys Stauroneis amphioxys Surirella ovata Surirella ovata Surirella ovata Tagus estuary, PT

C C C C C C C C C C C C C C C S

log/station transition log/station transition transition stationary transition transition stationary transition transition stationary logarithmic stationary transitional stationary stationary transition transition

mg glucose ng glucose pg glucose Reference equiv g equiv. mg equiv. Chl a1 d1 Chl a1 h1 cell1 d1

EPS

max rates, 14:10 L:D

3.7

21.6

Smith and Underwood, 2000

colloidal

max rates, 14:10 L:D

15.91

128.4

Smith and Underwood, 2000

10 max rates, 14:10 L:D max rates, 14:10 L:D max rates, 14:10 L:D max rates, 14:10 L:D max rates, 14:10 L:D max rates, 14:10 L:D Increase in L & D Increase in L Increase in L

1.92 5.1 16.51 5.09 3 10.1 3.4

Staats et al., 1999 Underwood and Smith, 1998a Smith and Underwood, 2000 Smith and Underwood, 2000 Underwood and Smith, 1998a Smith and Underwood, 2000 Smith and Underwood, 2000 de Brouwer and Stal, 2002 de Brouwer and Stal, 2002 de Brouwer and Stal, 2002 McConville et al., 1999 McConville et al., 1999 Underwood and Smith, 1998a Smith and Underwood, 2000 Smith and Underwood, 2000 Perkins et al., 2001

EPS EPS EPS colloidal EPS EPS colloidal sol EPS bound EPS bound EPS mucilage EPS loose EPS EPS EPS colloidal colloidal

80 21.2 61.3 212 6.3 42.2 1,800 200 2.43 4.35

max rates, 14:10 L:D max rates, 14:10 L:D max rates, 14:10 L:D

1.6 1.6 24.5

67 9.48 65.8 5,700

G.J.C. UNDERWOOD AND D.M. PATERSON

Organism/field site

S S S S S S S S S S S S S S S S S S S S S S S

top 200 mm 200–400 mm 800–1000 mm

L, D, light and dark periods respectively.

colloidal colloidal colloidal colloidal EDTA EDTA EPS EPS EPS/EDTA EPS EPS EPS EPS EPS EPS EPS low MW low MW low MW low MW low MW low MW low MW

diel emersion diel emersion diel emersion diel emersion diel emersion diel emersion diel emersion diel emersion coll diel emersion. diel emersion diel emersion diel emersion shade 910 mmol m2 s1 1600 mmol m2 s1 EDTA extraction diel emersion diel emersion EDTA extraction diel emersion diel emersion

19,900 7,600 2,200 10,360 382–418 80 483 95 220–850 164–472 13,200 18,200 960 580 1,572 0 490–850 472–800 1,300 240 3,265 30,450 40,000

de Brouwer and Stal, 2001 de Brouwer and Stal, 2001 de Brouwer and Stal, 2001 Yallop et al., 2000 van Duyl et al., 2000 de Brouwer and Stal, 2001 Underwood and Smith, 1998a Daborn et al., 1993 van Duyl et al., 1999 van Duyl et al., 2000 Underwood, 2002 Underwood, 2002 Perkins et al., 2001 Perkins et al., 2001 Staats et al., 2000 Staats et al., 2000 van Duyl et al., 1999 van Duyl et al., 2000 Perkins et al., 2001 Staats et al., 2000 Staats et al., 2000 Underwood, 2002 Underwood, 2002

EXTRACELLULAR CARBOHYDRATE PRODUCTION

Westerschelde, NL Westerschelde, NL Westerschelde, NL Severn estuary, UK Ems Dollard, NL Westerschelde, NL Alresford Creek, UK Bay of Fundy, CA Ems Dollard, NL Ems Dollard, NL Fiji muddy shore, FJ Fiji sandy shore, FJ Tagus estuary, PT Tagus estuary, PT Westerschelde, NL Westerschelde, NL Ems Dollard. NL Ems Dollard, NL Tagus estuary, PT Westerschelde, NL Westerschelde, NL Fiji muddy shore, FJ Fiji sandy shore, FJ

195

196

G.J.C. UNDERWOOD AND D.M. PATERSON

For all taxa so far examined, when grown in batch culture, the concentrations of extracellular carbohydrates in the medium increase significantly as the cells enter the transition from exponential growth into stationary phase (Sutherland et al., 1998a; Underwood and Smith 1998a; Staats et al., 1999, 2000b; Smith and Underwood, 2000). Smith and Underwood (2000) showed that similar growth dynamics occurred with duplicate clonal lines of five taxa of benthic diatoms, with peak production rates of EPS ranging between 1.64–5.13 mg glucose equivalents mg chl a1 d1. Similar daily rates of EPS production for benthic diatoms have been reported in other studies (Table I). Such values reflect net accumulation over a 24 h period, usually including a period of darkness. De Brouwer and Stal (2002) measured hourly rates of production of between 200–3000 ng glucose equivalents mg Chl a1 h1 of bound-EPS (removed by a 1-h incubation in water at 30 C) during illuminated periods in cultures of C. closterium and Nitzschia spp. These rates are much higher than those for cEPS present in the culture media (67–212 ng glucose equiv. mg Chl a1 h1) measured by Smith and Underwood (2000). Data normalised to cell density show a slightly different pattern. Cell-normalised production rates for colloidal carbohydrate range from 2.1–65.8 pg glucose equivalents cell1 day1 (median 42 pg glucose equiv. cell1 day1) and for EPS, from 0.93–31.5 pg cell1 day1 (median 8.7 pg glucose equiv. cell1 day1, Table I). In a study by Boulcott (2001), three diatom taxa showed significantly higher rates of carbohydrate production per cell when in logarithmic growth than when in stationary phase. The rates of production measured by Boulcott (2001) for EPS (ranging from 0.05–14.4 pg glucose equiv. cell1 day–1) are slightly lower than the maximum rates (between 6.31–34.16 pg glucose equiv. cell1 day–1) reported for five benthic diatom taxa by Smith and Underwood (2000) and closer to the rate of 1.6 pg gluc. equiv. cell1 day1 of ‘soluble EPS’ fraction produced by C. closterium measured by de Brouwer et al. (2002). Care has to be taken when comparing patterns of data for extracellular carbohydrate production, as it has been shown that patterns can vary depending on whether absolute value or biomass-normalised (cells, Chl a) data are used (Wolfstein and Stal, 2002). There is now increasing data to show that the production patterns of different EPS fractions vary with the physiological status of the diatom cells. de Brouwer and Stal (2002) measured six times the rate of cell-normalised production of bound-EPS (i.e. more closely associated with the cell) by C. closterium in logarithmic compared to stationary phase growth, while ‘soluble EPS’ was produced at a constant rate per cell throughout the growth curve (Table I). McConville et al. (1999) found that Stauroneis amphioxys cultures initially showed little EPS production, followed by

EXTRACELLULAR CARBOHYDRATE PRODUCTION

197

increased amounts of EPS secreted as a mucilage sheet during the transition from logarithmic to stationary phase. Once cells were in stationary phase, the production of mucilage EPS decreased and high rates of soluble EPS were released into the media as a result of dissolution of the mucilage sheet and continued production by the cells. The proportion of EPS in the secreted colloidal fraction also changes during growth curves. During logarithmic growth, about 20% of the colloidal exudates are polymeric (cEPS), but this can rise up to 50% for cells in stationary phase (Smith and Underwood, 2000). Cells maintained in stationary phase for longer periods of time (>10 d) tend to show a decrease in the rates of EPS being produced (McConville et al., 1999; Smith and Underwood, 2000; de Brouwer et al., 2002). McConville et al. (1999) observed that the mucilage layer surrounding the diatom cells disintegrated in older cultures, releasing material into the culture media. Studies on the composition of the EPS show that there is also a change in sugar composition though the growth curves (see Section IV). These data all indicate that benthic diatom cells produce higher concentrations and greater proportions of EPS in their exudates as they become nutrient-limited and enter stationary phase, and the composition of the EPS changes. Significant differences in the rates of EPS production between species under defined laboratory conditions have been measured. Nitzschia sigma, Navicula perminuta and Cylindrotheca closterium had significantly higher rates of EPS production than Nitzschia frustulum and Surirella ovata (Smith and Underwood, 2000). de Brouwer and Stal (2002) reported rates of EPS production by C. closterium to be approximately double those of an unidentified Nitzschia species under the same culture conditions. This raises the interesting question whether biofilm species composition in natural conditions will affect the production of EPS. This has not been investigated in detail, but there is some evidence to suggest that species composition does affect EPS dynamics in natural biofilms. The shift to greater colloidal carbohydrate and EPS production in cultures entering stationary phase appears related to nutrient limitation. Cultures of benthic diatoms show increased colloidal carbohydrate production when they are nutrient-limited (Underwood and Smith, 1998; Staats et al., 1999, 2000b; Smith and Underwood, 2000), and similar results have been seen in subtidal, nutrient-poor sands (Wulff et al., 2000), muddy sediments (Sutherland et al., 1998a; Staats et al., 2000b), tropical beaches and coral reef flats (Underwood, 2002). Alcoverro et al. (2000) found that EPS accounted for <5% of exudates during exponential and transition phase growth, but was approximately 65% of exudates in stationary phase. However, imposition of either N or P limitation increased EPS production

198

G.J.C. UNDERWOOD AND D.M. PATERSON

to between 50–70% of carbon in any phase of growth. Labelling experiments using 14C-bicarbonate on benthic diatom cultures have shown that between 30–70% of photoassimilates can be present in the colloidal fraction within a 1 h incubation, with cEPS containing between 5–30% of the photosynthetically labelled carbon (Table II). This overflow of excess carbon from the cell is a proposed mechanism by which the cells can maintain electron flow through their photosystems and thus protect them from photodamage under high light conditions. Staats et al. (2000b) found  increased EPS secretion when cells were grown on NHþ 4 compared to NO3 as a nitrogen source. This suggests that when not using energy to reduce nitrate, cells may channel excess energy into carbon production and then secrete the material. Staats et al. (2000b) also found that P-limitation increased EPS production. Similar processes occur within phytoplankton populations (Fogg and Thake, 1987). An important proviso must be made here concerning culture studies. Once benthic biraphid diatoms are maintained in culture, the cells lose their natural pattern of cell migration and sometimes even the ability to move. This is circumstantial evidence to suggest that the nature of EPS production may also change on placing diatoms in culture. The evidence that polymer produced through the raphe for locomotion is different in structure from that produced through pores in the frustules suggest that this may well be the case (Higgins et al., 2002). Therefore, while similarities between the results of laboratory and field studies exist, extrapolation of the results of culture studies to the natural field situation must be carried out with caution.

C. CARBOHYDRATE PRODUCTION IN MICROPHYTOBENTHOS IN SITU

In field studies, hourly rates of production can be obtained by measuring the net accumulation (normalised to Chl a) of EPS and other carbohydrate fractions over diel tidal emersion periods. Such an approach can both underestimate EPS production, because loss processes such as microbial mineralisation, and adsorption to sediment particles that would reduce extracted yield, will take place, and could also overestimate diatom EPS production, due to contamination from other sediment EPS sources. This means that wide margins of error may surround these values. From field studies, EPS production rates range from 0 to 18.2 mg glucose equiv. mg Chl a1 h1 (Table I). The highest values in this range are from a tropical intertidal shore where high light stress conditions applied (Underwood, 2002).

EXTRACELLULAR CARBOHYDRATE PRODUCTION

199

Temperate, intertidal mudflats have a median EPS production rate of approximately 400 ng glucose equiv. mg Chl a1 h1 (Table I). This falls somewhat above the range for loose-EPS production and at the bottom of the range of bound-EPS produced in culture (Table I). The majority of the field data are based on extractions of cEPS from freeze-dried sediments, which increases the yield (Section IIIA). In situ measurements of photosynthesis and EPS production using 14C incorporation allow both the rate of primary production (mg C m2 h1) and the allocation of photoassimilates into different extracellular fractions to be determined. Using these techniques colloidal carbohydrate production rates between 380 and 5200 mg C m2 h1 and EPS production rates of between 20–1000 mg C m2 h1 have been measured (Table II). Such values will be strongly influenced by both diatom biomass and light intensity. The percent allocation of 14C label to colloidal (1.7–73%) and cEPS (0.05–18%) measured in natural biofilms is very similar to that measured in cultures (Table II). This clearly indicates that substantial amounts of carbohydrate-rich exudates are produced by natural biofilms under conditions favourable for photosynthesis. There are conflicting data on the relationship between the rates of photosynthesis and the rates of colloidal carbohydrate and EPS production. Smith and Underwood (1998) observed that the rate of colloidal carbohydrate production was significantly correlated with the rate of primary production. Similar patterns have since been recorded in other studies (Staats et al., 2000a; Perkins et al., 2001). Perkins et al. (2001), working in the Tagus Estuary at irradiances of 1300–1600 mmol m2 s1, found that within 1 h, up to 37% of photoassimilated carbon was present in the colloidal fraction. This is higher (approx. 10%) than values measured at lower irradiances (80–620 mmol m2 s1) in the Humber Estuary by Smith and Underwood (1998). Higher irradiances have been thought to lead to greater exudate production, by the overflow hypothesis proposed by Staats et al. (2000a). Field data support this hypothesis. Wolfstein and Stal (2002) showed a clear linear relationship between the rate of production of intracellular and extracellular carbohydrates and irradiance. Extremely high rates of colloidal production have also been measured on tropical sand and mudflats, with light levels in excess of 2200 mmol m2 s1 (Underwood, 2002). The current literature indicates that between 1.7 to 73% of photoassimilates can be released as exudates into the colloidal carbohydrate pool of natural biofilms, with a median value between 30–40% (Table II). Staats et al. (2000a) used DCMU to block photosynthesis in cultures of C. closterium and in natural sediments and found no increases in any extracellular carbohydrate fractions. This led these authors to conclude that oxygenic photosynthesis is a prerequisite for extracellular carbohydrate

200

Organism/field site

Colloidal/EPS Extraction notes

Reference % Into Irradiance EPS EPS/colloidal production mg C m2 h1 during 1 h

Amphora coeffeaeformis Cylindrotheca closterium Navicula perminuta Nitzschia hybridaeformis Nitzschia longissima Surirella ovata Amphora coeffeaeformis Cylindrotheca closterium Navicula perminuta Nitzschia hybridaeformis Nitzschia longissima Surirella ovata Alresford Creek, UK Humber estuary, UK Humber estuary, UK Humber estuary, UK mixed microphytobenthos

EPS EPS EPS EPS EPS EPS colloidal colloidal colloidal colloidal colloidal colloidal EPS EPS EPS EPS EPS

200 30 30 200 200 30 200 30 30 200 200 30 200 1390 1480 80–620 200

lab exp in situ in situ in situ

20–175 50 70 30–150

30.1 10 9 18.6 26.7 5 70.3 62 60 45 50 31 0.05–0.8 0.43 0.31 0.2–1.0 16.1

Goto et al., 1999 Smith and Underwood, Smith and Underwood, Goto et al., 1999 Goto et al., 1999 Smith and Underwood, Goto et al., 1999 Smith and Underwood, Smith and Underwood, Goto et al., 1999 Goto et al., 1999 Smith and Underwood, Smith and Underwood, Underwood and Smith, Underwood and Smith, Smith and Underwood, Goto et al., 1999

2000 2000 2000 2000 2000 2000 1988 1988b 1988b 1988

G.J.C. UNDERWOOD AND D.M. PATERSON

TABLE II Rates of production (in terms of C) and percentage incorporation of photoassimilates over a 1 h period (14C incorporation) into various extracellular carbohydrate fractions by cultures of marine benthic diatoms (C) and in natural sediment biofilms (S)

a

mg C mg Chl a h1.

EPS EPS EPS EPS colloidal colloidal colloidal colloidal colloidal colloidal colloidal colloidal attached EPS

midday 1600 pm normal 1300 pm shade 460 photosynthetron 1303 lab exp 200 in situ 80–620 200 estimation 13 C estimation midday 1600 pm normal 1300 pm shade 700 photosynthetron 1303

400 300 1000 400–1100a 380–1700 400–5200

3800 1500 1100 35–80a

4 4.4 18 1.7–8.0 7.3–10.9 38.9 73 39 33 37 32

Perkins et al., 2001 Perkins et al., 2001 Perkins et al., 2001 Wolfstein and Stal, 2002 Smith and Underwood, 1988 Smith and Underwood, 1988 Goto et al., 1999 de Brouwer and Stal, 2001 Middleburg et al., 2000 Perkins et al., 2001 Perkins et al., 2001 Perkins et al., 2001 Wolfstein and Stal, 2002

EXTRACELLULAR CARBOHYDRATE PRODUCTION

Tagus estuary, PT Tagus estuary, PT Tagus estuary, PT Westerschelde, NL Alresford Creek, UK Humber estuary, UK mixed microphytobenthos Westerschelde, NL Westerschelde, NL Tagus estuary, PT Tagus estuary, PT Tagus estuary, PT Westerschelde, NL

201

202

G.J.C. UNDERWOOD AND D.M. PATERSON

production (Staats et al., 2000a; de Brouwer and Stal, 2001). Such a conclusion is at variance with other data that show that EPS production in benthic diatoms can occur in darkness. Smith and Underwood (1998, 2000) showed that when placed in darkness, biofilms and cultures increased both the rate of production of EPS (increased concentrations) as well as significant increase in the proportion of cEPS present in the colloidal fraction. By conducting pulse-chase experiments with 14C, it was shown that labelled EPS was produced by cells for up to 4 h after darkness was imposed. When the amount of 14C label produced over the time course was calculated, up to 16.4% of carbon fixed during the illuminated period ended up in the EPS fraction after 5 h darkness (Smith and Underwood, 1998). The consequence of darkness was to significantly increase the proportion of cEPS (Smith and Underwood, 2000). Secretion of low molecular weight exudates ceased in darkness, suggesting that this fraction is immediately dependent on photosynthesis. Recent culture work (de Brouwer et al., 2002) has shown that EPS loosely associated with cells (present in the media) was continuously produced in both light and darkness, but that bound-EPS was only produced in the light. This bound fraction decreased during periods of darkness, possibly due to solubilisation (similar to that reported by McConville et al., 1999) or to heterotrophic utilisation by the diatoms. Other field measurements have also recorded increases in EPS in darkened sediments (van Duyl et al., 1999, 2000). van Duyl et al. (1999) found that both high and low molecular weight carbohydrate fractions increased at the same rate in the top 2 mm of intertidal sediments during tidal emersion during light and dark periods. High molecular weight concentrations increased by 55% (19.4 mg glucose equivalents cm3 wet sediment h1), while low molecular weight carbohydrate concentrations showed a 95% increase (42.6 g glucose equivalents cm3 wet sediment h1). The ‘dark production’ of extracellular carbohydrates by diatoms appears to be mediated by light history. Perkins et al. (2001) found that shading biofilms as a pre-treatment resulted in increased EPS production against ambient controls, with the proportion of photoassimilates (14C) allocated to EPS increasing from 4 to 18% with shading pretreatment. Such data are clearly at variance with other studies (Wolfstein and Stal, 2002) and the reasons for this have yet to be determined. Why do diatoms produce EPS in the dark? There is extensive data showing that epipelic diatoms can move and vertically migrate under dark conditions (Hay et al., 1993; Seroˆdio et al., 1997; Perkins et al., 2001; Underwood, 2002) Evidence from AFM and lectin labelling studies (op cit.) clearly shows that carbohydrate-rich EPS is secreted during motility (Section II). So reallocation of carbon to EPS will be necessary for

EXTRACELLULAR CARBOHYDRATE PRODUCTION

203

movement in darkness. Whether such movement is sufficient to explain the increases in EPS concentrations measured in natural biofilms remains to be determined.

IV

COMPARATIVE BIOCHEMISTRY OF DIATOM EPS A. PRODUCTION PATHWAYS OF EPS

The common storage compounds of freshwater and marine diatoms are -1,3-linked glucans (e.g. leucosin, chrysolaminarin) (Myklestad and Haug, 1972; Myklestad, 1988) and these glucan reserves are utilized when cells are not illuminated, with rapid catabolism of glucan by exo-(-1,3)-D-glucanase (Myklestad and Haug, 1972; Varum and Myklestad, 1984; Myklestad, 1988). Although little literature exists on benthic diatom species, Wustman et al. (1998) demonstrated a chrysolaminarin from A. longipes, and recent sugar-composition work (Boulcott, 2001) has found that the storage compound of C. closterium is a glucan. Studies using 14C as a tracer in cultures of a number of species, and in sediments, have shown accumulation of photosynthetically fixed 14C within intracellular glucan fractions, and a delay of approximately 3 h before substantial amounts of 14C were present in the extracellular EPS (Smith and Underwood, 1998, 2000; Boulcott, 2001). Diatoms placed in darkness continued to secrete EPS, proportional to glucan utilization, but with little low molecular weight extracellular carbohydrate being produced (Smith and Underwood, 1998, 2000; Perkins et al., 2001). Under longer-term dark incubations, cell glucan reserves lasted between 3–4 days, after which EPS production also ceased (Smith and Underwood, 2000). Interestingly, this is a similar time period over which natural biofilms will maintain their rhythms of vertical migration when placed in darkness (Seroˆdio et al., 1997). These studies suggest that a glucan may be either the precursor of EPS, or that it acts as the photoassimilate carbon store providing energy for EPS synthesis during periods of darkness. Further support for this hypothesis has come from studies showing a significant correlation between glucan catabolism and EPS production in the dark in a number of benthic species (Smith and Underwood, 1998, 2000). The results from various inhibitor experiments (Boulcott, 2001) indicate at least two possible routes by which photosynthetically assimilated carbon can appear in EPS. Intracellular glucans, synthesised by an enzyme pathway with a slow turnover rate are produced in the light. This glucan is metabolized, in both light and darkness, by an enzyme system (glucanase) that has a high turnover rate and can

204

G.J.C. UNDERWOOD AND D.M. PATERSON

be blocked by the addition of cycloheximide, which inhibits de novo protein synthesis. This glucan-EPS pathway can be blocked using PNGP (P-nitrophenyl -D-glucopyranoside) or cycloheximide, resulting in a reduction in EPS production. Yet EPS synthesis is not completely inhibited by disruption of the glucan degradation pathway. There is also a compositional mismatch between glucan (consisting of >90% glucose) and a number of the EPS types reported containing a number of other sugars and uronic acids, though some extracellular fractions are glucose rich (Section IV). This indicates that while glucose from glucan may be directly incorporated into EPS, other sugars also need to be synthesised for the production of some of the EPS types produced by benthic diatoms. The role of glucan may be as an energy and carbon source for those pathways, particularly in darkness. A number of studies have shown that diatom EPS is synthesized in the Golgi body and transported in vesicles to the cell membrane. These vesicles fuse with the cell membrane and the EPS is extruded through the raphe, apical pore field, or other fissures in the silica frustule (Cooksey and Cooksey, 1986; McConville et al., 1999; Wang et al., 2000). It has been discovered by using monoclonal antibody immuno-cytochemistry, that the pathway from intracellular to fully assembled extracellular stalks in A. longipes is a complex process. Several different EPS polymers are initially co-localized in intracellular vesicles near the nucleus. These polymers are then sorted, segregated and deposited at specific locations on the plasma membrane. These EPS pass through the diatotepum (a multilayered organic structure associated with the silica frustule) and are extruded from pores or slits in the frustule, where the polymers self-assemble to form the stalk structure (Wustman et al., 1997, 1998; Wang et al., 2000). No such detailed studies have been carried out on EPS produced by motile diatoms.

B. BIOCHEMICAL COMPOSITION OF DIATOM EPS

The biochemical composition of the EPS of approximately 30 genera of diatoms has been examined (for review see Hoagland et al., 1993). The majority of these studies have been carried out on planktonic diatoms, with only a few benthic species investigated. Many of the investigations that have involved benthic species have concentrated on attached forms due to their role in biofouling (Hoagland et al., 1993). This means there is only a limited subset of data concerning marine benthic motile taxa (Table III). Detailed studies on Stauroneis decipiens (biraphid, motile) and Achnanthes longipes (monoraphid, with a motile phase preceding stalk production) indicate that the predominant extracellular polymers are

EXTRACELLULAR CARBOHYDRATE PRODUCTION

205

complex proteoglycans that may be packaged into biocomposite structures (Wetherbee et al., 1998; Wustman et al., 1998). The polysaccharide portions of these proteoglycans are highly substituted with sulphate and incorporate galactosyl, glucuronosyl and fucosyl residues. Other studies on Cylindrotheca closterium, Navicula salinarum and Staurophora (Stauroneis) amphioxys have also measured substantial proportions of acidic polysaccharides, protein, sulphated groups, methylated sugars and uronic acids in EPS (Staats et al., 1999; McConville et al., 1999). Staats et al. (1999) found that the composition of non-attached EPS was different from EPS closely associated with the cells of C. closterium, although there was less variation in these EPS fractions with N. salinarum. Carbohydrate was about 80% of the weight of EPS in N. salinarum and non-attached EPS from C. closterium. However, C. closterium attached-EPS was only 60% carbohydrate, with 20% uronic acids and 10% sulphate. The monosaccharide composition was also different between EPS fractions. Staats et al. (1999) found that non-attached EPS of C. closterium and N. salinarum had high proportions of xylose, rhamnose and galactose, while attached EPS was over 80% (mol%) glucose. De Brouwer and Stal (2002) also found similar differences in monosaccharide composition between soluble and bound EPS fractions in cultures of C. closterium and Nitzschia sp. The bound EPS fraction was more glucose rich and the soluble fraction had a more balanced distribution of monosaccharide components. They also showed that the glucose component of the bound fraction of these two species increased during light periods (by an extra 10–15%), whereas there was no significant change on the sugar spectrum of soluble EPS. Related temporal studies have been done in situ by de Brouwer and Stal (2001) investigating the monosaccharide composition of different size classes of EPS (separated using molecular filters into >100, 100–50, 50–10, <10 kDa size classes) at the beginning and end of emersion periods on a tidal flat. They found greatest increases in a >100 kDa and <10 kDa fractions over the course of a tidal emersion period. Glucose substantially increased after a period of emersion in all 4 EPS size classes. In the uppermost 200 mm layer of sediment, glucose was 85% of the monosaccharide pool by the end of the emersion period. Fractionation of isolated EPS though an alcohol precipitation series revealed that Cylindrotheca closterium produced two distinct types of EPS during a growth curve (Boulcott, 2001). During logarithmic growth, cells exhibited a high cell-normalised rate of production of a ‘complex EPS’ containing seven different sugar components that precipitated in 20% ethanol and an additional EPS (containing mannose, galactose, glucose and uronic acid) with precipitated at 80% ethanol during stationary phase

206 TABLE III Monosaccharide composition (%) of extracellular carbohydrate fractions (colloidal, EPS) from culture (C) and field (F) studies of benthic diatoms and marine biofilms, grouped by similarity according to cluster analysis. FUC ¼ fucose, ARA ¼ arabinose, RHA ¼ rhamnose, GAL ¼ galactose, GLC ¼ glucose, MAN ¼ mannose, XYL ¼ xylose, URON ¼ uronic acids, UNKN ¼ unknown sugars C/F extraction

Phase L/D cluster FUC ARA RHA GAL GLC MAN XYL URON UNKN Reference

Chaetoceros decipiens Coscinodiscus nobilis Marennes st1 Stauroneis amphioxys Stauroneis amphioxys Stauroneis amphioxys Stauroneis amphioxys Stauroneis amphioxys Stauroneis amphioxys Chaetoceros curvisetus Chaetoceros debilis Westeschelde, NL Humber A Humber B1 Humber B2 Humber C1 Humber C2 Humber D1 Humber D2 Humber A Humber B1 Humber B2

C C F C C C C C C C C F F F F F F F F F F F

COLL A EPS A COLL B1 EPS B1 EPS B1 EPS B1 EPS B1 EPS B1 EPS B1 COLL B1 COLL B1 COLL em B2 COLL B2 COLL B2 COLL B2 COLL B2 COLL B2 COLL B2 COLL B2 COLL B2 COLL B2 COLL B2

Colloidal colloidal EDTA log soluble log mucilage trans soluble trans mucilage stn soluble stn mucilage Coll Coll coll > 10KD Coll Coll Coll Coll Coll Coll Coll EDTA EDTA EDTA

32 34 15 7 7 10 11 12 10 14 30 4 10 8 10 10 6 9 5 3 9 13

0 0 6 2 1 0 0 0 0 0 0 12 6 6 7 8 6 10 6 5 9 7

34 15 12 5 9 12 11 11 11 13 17 8 24 24 20 21 23 20 21 18 14 12

17 0 30 21 25 30 28 26 28 30 29 22 18 20 17 18 19 19 21 22 24 24

5 16 8 13 7 2 2 2 2 5 5 25 12 11 16 13 11 12 12 14 14 14

7 19 22 9 9 7 10 8 12 12 10 20 21 19 22 22 24 24 24 23 24 23

5 6

8 9

0 9 7 25 17 17 16 20 16 0 0 8 8 12 8 8 12 7 13 15 6 7

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Haug and Myklestad, 1976 Percival et al., 1980 de Brouwer et al., in press McConville et al., 1999 McConville et al., 1999 McConville et al., 1999 McConville et al., 1999 McConville et al., 1999 McConville et al., 1999 Haug and Myklestad, 1976 Haug and Myklestad, 1976 de Brouwer and Stal, 2001 de Brouwer et al., 2003 de Brouwer et al., 2003 de Brouwer et al., 2003 de Brouwer et al., 2003 de Brouwer et al., 2003 de Brouwer et al., 2003 de Brouwer et al., 2003 de Brouwer et al., 2003 de Brouwer et al., 2003 de Brouwer et al., 2003

G.J.C. UNDERWOOD AND D.M. PATERSON

Species

F F F F F F F F F F F F F F F F

C. closterium C. closterium C. closterium Cylindrotheca fusiformis Navicula sp. Seawater mucilage C. closterium C. closterium Nitzschia sp Nitzschia sp

C C C C C F C C C C

EDTA EDTA EDTA EDTA Coll Coll Coll EDTA EDTA Coll Coll Coll EDTA EDTA EDTA EDTA > 10KD coll Lg20a coll S20a coll log coll capsular coll Sol Sol Sol Sol

COLL COLL COLL COLL COLL COLL COLL COLL COLL COLL COLL COLL COLL COLL COLL COLL em

B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 C1

13 13 8 9 14 12 11 11 12 12 8 4 12 9 7 4

7 8 8 7 6 6 6 6 6 6 6 3 9 8 6 10

13 14 11 12 17 14 14 10 10 19 19 17 17 16 12 12

25 28 25 26 21 21 20 25 24 23 27 25 26 26 21 18

12 13 11 14 7 13 12 13 13 13 15 8 14 13 13 20

23 15 20 23 21 22 23 21 21 17 12 22 12 12 20 15

EPS EPS EPS COLL EPS EPS EPS EPS EPS EPS

C1 C1 C1 C1 C1 C1 C2 C2 C2 C2

4 3 4 0 0 0 0 0 0 0

0 0 0 16 2 14 0 0 0 0

6 5 6 16 11 9 9 9 12 14

13 12 13 17 18 36 29 28 21 25

28 35 28 20 29 16 7 9 8 11

7 7 7 14 15 7 18 17 15 18

Db Lb Db Lb

8 7 8 18 18 19

7 9 17 9 15 12 14 15 14 10 14 22 10 15 21 35

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

de de de de de de de de de de de de de de de de

Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer Brouwer

et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 et al., 2003 and Stal, 2001

33 31 33 0 7 0 36 37 44 33

0 0 0 0 0 0 0 0 0 0

Boulcott, 2001 Boulcott, 2001 Boulcott, 2001 de Angelis et al., 1993 Stal et al., 1994 de Angelis et al., 1993 de Brouwer and Stal, 2002 de Brouwer and Stal, 2002 de Brouwer and Stal, 2002 de Brouwer and Stal, 2002

EXTRACELLULAR CARBOHYDRATE PRODUCTION

Humber C1 Humber C2 Humber D1 Humber D2 Marennes st1 Marennes st3 Marennes st4 Marennes st3 Marennes st4 Dollard st1 Dollard st2 Dollard st3 Dollard st1 Dollard st2 Dollard st3 Westeschelde, NL

(continued )

207

208

TABLE III Continued. C/F extraction

phase L/D cluster FUC ARA RHA GAL GLC MAN XYL URON UNKN Reference

Westeschelde, NL

F

COLL im

NL NL NL

C C C C C C C F C F F h F

Westeschelde, NL

F

Seawater mucilage C. closterium N salinarum Eden, UK N subinflata

F C C F C

C. closterium C .closterium C. closterium C. closterium C. closterium C. closterium C. closterium Westeschelde, C. closterium Westeschelde, Westeschelde, Westeschelde,

NL

EDTA > 10KD coll L40a,b coll L60a,b coll L80a,b coll D40a,b coll D60a,b coll S60a coll stn coll < 10KD bound coll < 10KD coll > 10KD EDTA < 10KD EDTA < 10KD coll non-att non-att bulk

EPS EPS EPS EPS EPS EPS EPS COLL EPS COLL COLL COLL

C2

5

10

12

20

25

18

C2 C2 C2 C2 C2 C2 C2 em D1 D D2 im D2 im D2 em D2

5 4 0 5 0 0 1 0 0 0 3 0

0 0 0 0 0 0 0 2 0 1 5 12

7 7 0 7 0 0 1 1 9 2 5 15

17 16 15 16 16 17 15 40 14 30 12 30

12 9 8 16 11 16 22 30 55 55 45 35

7 9 18 10 22 17 14 4 11 1 12 0

COLL im

D2

0

4

7

17

55

0

COLL EPS EPS total ?

D2 D3 D3 D3 E

0 0 0 10 0

10 0 0 0 0

9 15 6 0 0

20 12 19 15 2

48 23 42 37 94

5 4 14 6 2

8 7 0 8 0 0 2

9 46 20 14 1

30

0

de Brouwer and Stal, 2001

45 47 58 37 50 49 44 18 11 4 5 0

0 0 0 0 0 1 0 0 0 0 0 0

Boulcott, 2001 Boulcott, 2001 Boulcott, 2001 Boulcott, 2001 Boulcott, 2001 Boulcott, 2001 Boulcott, 2001 de Brouwer and de Brouwer and de Brouwer and de Brouwer and de Brouwer and

8

0

de Brouwer and Stal, 2001

0 0 0 0 0

0 0 0 0 0

de Angelis et al., 1993 Staats et al., 1999 Staats et al., 1999 Taylor et al., 1999 Bhosle et al., 1995

Stal, Stal, Stal, Stal, Stal,

2001 2002 2001 2001 2001

G.J.C. UNDERWOOD AND D.M. PATERSON

Species

C C C C C C C C F F F F C C C C

attach attach bound bound bound

EPS EPS EPS L EPS D EPS L EPS 0.1N H2SO4 glucan 0.1M H2SO4 glucan top 200 m EPS Coll coll Coll EPS hot water EPS Coll COLL Coll COLL Coll COLL Coll COLL

E E E E E E E E E E E E F F F F

0 0 0 0 0 0 1 0 1 2 1 2 0 16 20 0

3 1 0 0 0 0 0 0 2 0 0 1 0 8 0 0

1 9 5 6 4 9 1 0 3 0 1 2 24 20 33 70

2 4 8 7 5 5 0 1 3 5 3 5 8 17 8 0

83 85 72 70 82 81 92 91 85 82 83 78 0 0 0 0

8 1 6 8 5 4 0 2 3 2 4 6 34 7 10 7

4 0

0 0 0 0 0 0 0 0

3

0 0 8 9 5 0 5 5 2 0

0 7 9 0

0 0 0 0

34 25 20 23

1 1 0

0

Lg20 ¼ log phase EPS precipitated in 20% ethanol, S60 ¼ stationary phase EPS precipitated in 60% ethanol etc. L ¼ material produced in light, D carbohydrate produced in darkness. em, im ¼ at the end of an emersion or immersion phase respectively. a

b

Staats et al., 1999 Staats et al., 1999 de Brouwer and Stal, 2002 de Brouwer and Stal, 2002 de Brouwer and Stal, 2002 Stal et al., 1994 Boulcott, 2001 Boulcott, 2001 de Brouwer and Stal, 2001 Taylor et al., 1999 Gretz and Underwood, unpub. Gretz and Underwood, unpub. Allan et al., 1972 Allan et al., 1972 Allan et al., 1972 Allan et al., 1972

EXTRACELLULAR CARBOHYDRATE PRODUCTION

C. closterium N salinarum C. closterium Nitzschia sp Nitzschia sp Navicula sp. C. closterium C. closterium Westerschelde, NL Eden,UK Alresford creek, UK Alresford creek, UK Nz frustulum Nz angularis N incerta Asterionella socialis

209

210

G.J.C. UNDERWOOD AND D.M. PATERSON

(Boulcott, 2001). Incubating cultures in darkness did not result in any significant changes in the monosaccharide distribution of the EPS among these fractions, both of which were produced in darkness. These data on monosaccharide composition may explain differences in EPS composition measured using PyMS (Smith and Underwood, 2000). The EPS compositional data for C. closterium from Boulcott (2001) are in close agreement with the findings of de Brouwer and Stal (2002), except that Boulcott (2001) identified two distinct EPS types within the colloidal extract which may be equivalent to the ‘soluble EPS’, produced in both light and darkness found by de Brouwer and Stal (2002). In contrast to these studies showing diversity of monosaccharide composition and changes due to growth phase and light–dark conditions, McConville et al. (1999) found no significant differences in the extracellular soluble and mucilage EPS produced by Staurophora amphioxys over a growth curve. While the patterns of production of these fractions changed, the monosaccharide composition remained relatively constant and low in glucose (2–13%).

C. COMPARISON OF STUDIES

One of the problems in comparing different studies that have investigated diatom EPS composition is the range of techniques used. As discussed above, this makes decisions concerning the comparability of fractions very difficult. Some authors have measured monosaccharide composition, while more detailed investigations have included methylation analysis, and analysis of proteins and uronic acids. To investigate whether unifying patterns emerge from these studies of EPS composition, a meta-analysis was carried out. This analysis is based on data from 13 studies of extracellular carbohydrate production in cultures of benthic diatoms and studies of monosaccharide composition of colloidal and EPS extracts from intertidal sediments supporting rich diatom biofilms (Table III). Only information on monosaccharide and uronic acid content was sufficiently comprehensive to be used in this analysis. The different extraction methods used in these studies are indicated in Table III. Certain problems had to be overcome before the data were in suitable form for analysis. Some studies separated mannose from xylose while others did not. In the meta-analysis, mannose and xylose fractions were combined. Ribose was only reported in three studies (Allen et al., 1972; de Angelis et al., 1993; Taylor et al., 1999). It is not clear whether ribose was absent or just not measured in other studies, but ribose data had to be removed. Some workers report small amounts of

EXTRACELLULAR CARBOHYDRATE PRODUCTION

211

unknown sugars within their samples. We have assumed that where no unknown sugar was reported there was none. Finally, various workers have reported uronic acids as present in the EPS samples. As uronic acids appear to be an important component of EPS we have left these data in the analysis. Cluster analysis identified six groupings within the data (separated at the 60% similarity level), clustered into three broader divisions (Fig. 4, Table III). There were significant differences between these six groups in the proportions of monosaccharides (Table IV). Principal components 1 and 2 explained 45% of the variation in the data set, with the groups identified by cluster analysis mapping onto the distribution of data within the PCA (Fig. 4). Principal component 1 was positively related to the proportions of the monosaccharides, rhamnose, fucose, mannose/xylose, and negatively related to glucose and uronic acids. Principal component 2 mainly reflected a gradient in the abundance of arabinose and mannose/xylose and unknown sugars. Samples in clusters D1 þ D2 and F were the farthest apart on this

Fig. 4 Scatter plot of Principal Components 1 and 2 derived from analysis of monosaccharide composition of diatom mucilage and carbohydrate extracts from sediments. Samples are coded according to the different clusters obtained from cluster analysis (insert, and Table 3). Vectors showing the significant correlations of different monosaccharides with PC1 and PC2 are indicated. Key to clusters: 5 ¼ A, g ¼ B1, œ ¼ B2, m ¼ C1, 4 ¼ C2,  ¼ D1, f ¼ D2, * ¼ D3, ¼ E, ^ ¼ F. Carbohydrates coded as Table 4.

Sample cluster type

ANOVA

A 2

B1 10

B2 26

C1 7

C2 12

D2 7

D3 3

E 12

F 4

FUC

33.0 > all

9.3 >C,D,E

2.5

1.7

0.4

3.3

0.6

0

6.7 >A,C2

0.8

4.8

0

0.5

RHA

9

6.5

6.9

7

2.7

18.2

19.6

23.3 >A,E,F

15.3 E

3.8

p < 0.001

GLC

10.5

27.7 >A,C, D3,E,F 5.4

36.8 >all exc A 8.3

p < 0.001

GAL

24.5 >C,D, E,F 8.5

6.9 >all exc.C1 16.2 >C2, D,E,F 22.6 >A, E,F

9 >C, D,E 2.0

p < 0.001

ARA

13.1
13

24.5

12.8

0

p < 0.001

13

12.2

9.5

14.5

p < 0.001

4.5

12.5

15.3 > D2,E 42 >all

3.9

URONICS

6.6

0

3.4

0

p < 0.001

UNKNOWN

0

0

20.4 > C,D,E 11.5 > D3,E,F 0

34 >A,B, C2,E 8

83.1 >all

MAN/XYL

46.1 >all exc E 4.7

0

0

1.3

25.5 >all

p < 0.001

22 >all exc C2 0

0.1

p < 0.001

G.J.C. UNDERWOOD AND D.M. PATERSON

Sugar type/ cluster size

11.3

212

TABLE IV Mean % sugar composition of nine different clusters of carbohydrate fractions extracted from cultures of diatoms and field samples. Clusters were derived from cluster analysis and were separated at a > 60% similarity for major groups and a > 70% similarity for subgroups. There were significant differences in relative abundance of sugar between clusters (ANOVA), and differences between sample clusters (post hoc Tukey test) are indicated (>/< mean value significantly greater/less than cluster types listed, exc ¼ excluding). Abbreviation for sugars as in Table III.

EXTRACELLULAR CARBOHYDRATE PRODUCTION

213

plot, indicating greatest difference, and samples in cluster B2 were tightly grouped together. Clusters A, B and C were grouped together at a 55% similarity. Group A consisted of extracellular carbohydrates from two planktonic species of diatoms, Coscinodiscus nobilis and Chaetoceros decipiens, characterised by having significantly higher amounts of fucose and rhamnose than the other clusters (Table IV and Fig. 4). Cluster B grouped colloidal carbohydrates from cultures of planktonic Chaetoceros species, mucilage produced by the benthic diatom Staurophora amphioxys (McConville et al., 1999) and material extracted from natural sediments. The subgroup B1 included colloidal and EPS material from cell cultures and a sample of EDTAextracted colloidal carbohydrate from mudflats in Marennes-Ole´ron Bay, France. These extracts had significantly greater proportions of fucose and galactose than other clusters (Table IV). Subgroup B2 included 26 samples of field-extracted colloidal carbohydrates, either extracted in saline or using EDTA. All these samples were tightly clustered in the PCA plot (Fig. 4). These samples had significantly higher proportions of fucose, arabinose, rhamnose, galactose, mannose/xylose than most of the C, D, E and F clusters (Table IV). There was little similarity between the composition of EPS samples from cell cultures and from the field. In contrast to cluster B, data from both diatom cultures and field-derived samples were grouped together in Cluster C. Subgroup C1 contained the > 10 kD size fraction of EDTA-extracted colloidal carbohydrate material from sediment at the end of a tidal emersion period (de Brouwer and Stal, 2001), precipitated EPS from Adriatic seawater supporting blooms of Cylindrotheca fusiformis, colloidal carbohydrate extracted from cultures of C. fusiformis (de Angelis et al., 1993), and EPS from cultures of C. closterium produced in both log and stationary phase of growth (Boulcott, 2001). Samples grouped in subgroup C1 had moderate proportions of glucose and galactose, and significantly higher mean abundance of uronic acids (Table II), an exception being the Adriatic – C. fusiformis samples which had no uronic acid. Subgroup C2 consisted of EPS from a number of studies of C. closterium and a Nitzschia species (Boulcott, 2001; de Brouwer and Stal, 2002) as well as a >10 kD EDTA-extracted carbohydrate fraction extracted from Westerschelde sediments at beginning of a tidal emersion period. These fractions were rich in uronic acid (Table III). Generally fucose and arabinose were low in all C-cluster samples, which separated them from clusters A and B (Table IV). Clusters D and E were more closely related to each other that the rest of the data set. Cluster D also contained both sediment samples and diatomculture material. Generally, the sediment data consisted of polymer <10kD,

214

G.J.C. UNDERWOOD AND D.M. PATERSON

extracted using either saline or EDTA (Table IV). Subgroup D2 contained mainly colloidal carbohydrate fractions from sediments, with ‘bound’ EPS from C. closterium, while subgroup D3 contained non-attached EPS from C. closterium as well as bound sediment material. Cluster D samples were low in fucose, arabinose and rhamnose (compared to clusters A and B), low in mannose/xylose and uronic acids (Clusters C1 and C2) and rich in galactose and glucose (Table IV). Cluster E contained either attached, capsular or bound EPS from diatom cultures, a colloidal carbohydrate extract from the Eden estuary, the top biofilm slice from de Brouwer and Stal (2001), and EPS from a dense biofilm in Arlesford creek, Essex (top 2 mm). This cluster was characterised by very high glucose abundance (mean 83.1%). Also included in this cluster were intracellular glucan samples extracted from C. closterium (Boulcott, 2001). The final cluster (F) contained culture data from Allen et al. (1972). These samples had significantly higher amounts of unidentified carbohydrate and rhamnose than the other clusters (Fig. 4). Because of the difficulties of comparing studies, it has only been possible to carry out this comparison at the level of monosaccharide composition. The relative abundance of different monosaccharide groups does not convey any information about the structural or biochemical properties of the compounds concerned. This information can only be derived from modern analytical techniques such as nuclear magnetic resonance (NMR) analysis (Nicolaus et al., 2002).. Such studies are rare with benthic marine taxa or natural biofilms. Nevertheless, it is apparent that there are some major differences in monosaccharide composition that should provide a basis for comparison and also to promote further investigation. In the meta-analysis, there was no consistent grouping of results by laboratory, suggesting that patterns were not attributable to laboratory practise. In general, two major families of EPS types appear to exist (A–C and D–E). Material from cultures and field samples was represented in each family, as described above. The diversity of monosaccharides in the first family was greater, while in the second family, glucose was a more significant component. These data, in conjunction with the information presented in Sections IIIB and C and IVA and B can be used to produce a conceptual view of benthic diatom EPS production (Fig. 5). Living diatom cells appear to have a covering of mucilage and other organic material over the silica frustule (Fig. 3) and AFM suggests that this coating mucilage is probably different from the EPS produced for movement (Higgins et al., 2000, 2002). No detailed compositional analysis has been done on this EPS. The degree to which coating-mucilage merges into the next fraction is uncertain. The glucose-rich bound or attached EPS shows a high degree of

EXTRACELLULAR CARBOHYDRATE PRODUCTION

215

Fig. 5 Conceptual model of the range of different benthic diatom EPS types and features of their production patterns. EPS types refer to the Cluster Analysis in Table III.

light-dependent production and can be removed from cells using various extraction techniques. This bound EPS is very different in composition from the more labile cEPS and other colloidal carbohydrates present in cultures and sediments. At least seven major types of these cEPS can be identified (Table III). It is in this pool that the most diversity in production patterns and sugar composition (and probably in tertiary structure) has been found. The production of these EPS depends on environmental conditions and the physiological status of the cells (Staats et al., 1999; Smith and Underwood, 2000; Boulcott, 2001). These differences in sugar composition are important as they could result in different physical properties and potential interactions within the environment (Zhou et al., 1998; Decho, 2000). The final major pool consists of low molecular weight exudates. These show high light-dependent patterns of production. These compounds have been little studied, though they are likely to be the most biologically reactive in terms of transfer of carbon into other trophic groups. The boundaries and flux of material between these different groups of extracellular carbohydrate exudates are unresolved, and more detailed biochemical analysis, linked with ecological and physiological investigations, are required to investigate the properties and interactions of these molecules. This is not an easy task, given the fine scale of the micro-environment that epipelic biofilms inhabit and the need to obtain sufficient material for biochemical characterisation (Taylor, 1998).

216

G.J.C. UNDERWOOD AND D.M. PATERSON

V NATURAL PATTERNS OF COLLOIDAL CARBOHYDRATE AND EPS DISTRIBUTION Colloidal carbohydrate and EPS concentrations tend to be maximal when epipelic diatoms are found at high biomass. Therefore, conditions that support the colonisation and growth of epipelic biofilms lead to elevated levels of EPS. Light, nutrients, and the nature of the sediment are all critical, as are the activity of bacteria and the influence of grazing infauna. The colloidal carbohydrate content of intertidal sediments can be high, up to 5 mg g1 in thick epipelic biofilms (Underwood and Paterson, 1993). In intertidal sediments, the concentrations of colloidal carbohydrates are closely related to the microphytobenthic biomass. Significant correlations between colloidal carbohydrate content and Chl a content were observed in some early studies (Madsen et al., 1993; Underwood and Paterson, 1993; Underwood et al., 1995). The factors that control the occurrence of colloidal carbohydrate at the sediment surface can be expressed conceptually in terms of production and loss factors. The nature of the transformation of EPS between the various bound and colloidal forms is not well investigated, therefore it may be best to consider that the transient pool of EPS in sediments exists as an equilibrium between complementary states (Fig. 5), recognising that any extraction procedure will alter and influence the nature and the properties of the EPS extracted. The nature of the extraction technique makes comparison between studies difficult but several authors have reported on the distribution of variation fraction of EPS within sediment, and the water column.

A. HORIZONTAL DISTRIBUTION OF EPS

Underwood and Smith (1998a) investigated Chl a: colloidal carbohydrate distributions for a range of European mudflats and published a model describing the relationship. This model was: Log ðcoll: Carbohydrate content þ 1Þ ¼ 1:40 þ 1:02  ðlog Chl a content þ 1Þ ð1Þ The relationship had an r2 of 64.6% and was validated against a set of independent data from other European estuaries. Subsequently, many

EXTRACELLULAR CARBOHYDRATE PRODUCTION

217

studies have found this model to be robust for other estuaries (Taylor, 1998; van Duyl et al., 1999; Blanchard et al., 2000; Widdows et al., 2000). In validating their model, Underwood and Smith (1998a) found that this relationship only fitted for sediments dominated by fine clay particles. This has been confirmed in other studies, such as across a wide horizontal transect encompassing silts and sandy sites in the Humber estuary (Blanchard et al., 2000). The species composition of the biofilm was also important, and the Underwood and Smith (1998a) model was found to only apply where epipelic diatoms were greater than 50% of the microphytobenthic assemblage. Similar patterns were found in developing saltmarsh (Underwood, 1997). When samples were grouped according to biofilm species composition, diatom-dominated and mixed taxa assemblages (with > 50% RA diatoms) showed significant colloidal carbohydrate : Chl a relationships, while other assemblages dominated by green algae or cyanobacteria did not show this pattern (Underwood, 1997). The species of microphytobenthos present within a sediment biofilm is partly dependent on sediment particle size (Underwood and Kromkamp, 1999; Paterson and Hagerthey, 2000) and therefore, the effects of sediment type and assemblage composition are not independent of each other. A study of tropical intertidal and subtidal habitats found significant differences in species composition between sites, associated with differing colloidal carbohydrate : Chl a relationships in subtidal seagrass-sediments and coral sands and on intertidal silty sediments (Underwood, 2002). Though large data sets may reveal significant relationships, individual sites can show different patterns. Thornton et al. (2002) found that pooled data from six sites in the Colne estuary showed a significant relationship between colloidal: Chl a, but at the individual site level this only occurred at two sites. Thornton et al. (2002) found that epipelic diatom assemblage composition did not explain differences in data between sites, with species composition appearing to have little effect on the sediment colloidal carbohydrate: Chl a relationship. In contrast, Underwood et al. (1998) found that G. limosum and Nitzschia sigma were correlated with high concentrations of carbohydrates in biofilms at Colne point saltmarsh. The general pattern of EPS distribution over the intertidal is for increasing levels from the tidal edge toward the shore (Underwood and Paterson, 1993; Taylor, 1998). However, it has also been shown that within this coarse pattern, EPS concentrations vary depending on the nature and surface topography of the intertidal region. For example, on the Skeffling flats of the Humber estuary, the mid-intertidal zone is dominated by ridges and runnels. The ridges were found to have a higher concentration of colloidal carbohydrate than adjacent runnels and this was attributed

218

G.J.C. UNDERWOOD AND D.M. PATERSON

Fig. 6 Conceptual diagram of the build-up and decline of colloidal carbohydrate content in the upper 1 mm of sediment during a tidal exposure period, under reasonable conditions for biofilm development (light, temperature, nutrients, disturbance). At the beginning of the emersion period (Emersion 1), the distribution of colloidal carbohydrate in the sediment is relatively uniform with depth. As the transient biofilm migrates to the sediment surface and begins to photosynthesise, a slight elevation in colloidal carbohydrate can be found (1 h–2 h). As photosynthesis continues this surface elevation increases and begins to be noted at greater depth (4 h). Following the next period of tidal immersion the surface elevation of colloidal content has been lost (based on the data of Taylor, 1998).

to carbohydrate dissolution within the saturated troughs (Paterson et al., 2000; Blanchard et al., 2000). However, this evidence is based on relatively few samples and often within a fairly restricted spatial scale. The limited evidence available suggests that EPS will have a patchy and variable distribution depending on the local sources and sinks. We know very little of EPS patch dynamics (Fig. 6) or the combined result of production, loss and turnover rates that drive this variability and this would be a topic worthy of further attention. B. DEPTH DISTRIBUTION OF EPS

Microbial activity in sediments and soils is usually elevated around the interface between the water/air and the sediment bed. The most extreme expression of this accumulation of activity and biomass at the surface is found where fine sediments are closely packed and a biofilm or microbial mat is formed at the surface (Boudreau and Jorgensen, 2001; Paterson, 2001). Since the microbes are concentrated at the surface, this region becomes an area of rapid metabolism and cycling of metabolic products. EPS distribution is also constrained by microbial activity and patterns of distribution largely reflect the biomass at the surface. The spatial

EXTRACELLULAR CARBOHYDRATE PRODUCTION

219

constraints of microbial activity in sediments were only fully understood after the development of suitable techniques to measure related products on micrometer scale (Revsbech, 1994). From these studies it might be inferred that EPS concentration should reflect the depth of photosynthetic activity, but this is more difficult to assess. Results from the literature are varied and this probably reflects both the analytical technique and operational fraction examined as well as the depth resolution of the measurements. De Brouwer et al. (2002) found little variation in EPS fractions with depth using a relatively coarse analysis. However, fine scale analysis on the level of the microbial assemblages has been problematic. This has been resolved by a new methodology which combines the freeze sampling of sediments in situ with subsequent fine sectioning and analysis (Paterson, 2000). This technique, known as the Cryolander system (Wiltshire et al., 1997), allowed extremely fine scale resolution of materials in natural sediments. Taylor and Paterson (1998) were the first to employ the system to determine the depth profile of EPS with a depth resolution of 200 mm (e.g. Fig. 6). This work supported the studies of Underwood and Smith (1998a) in that the colloidal carbohydrates correlated well with the biomass of microphytobenthos and decreased rapidly with depth whereas the remaining carbohydrate material was found to be constant or to increase with depth (Taylor and Paterson, 1998). De Brouwer (2002) also noted the variation in behaviour of different EPS fraction with respect to depth distribution. This is not surprising given that different molecules and, even tertiary states, of the same polymer will have different solubility and relative binding capacities (Liu and Fang, 2002). Surface elevation of colloidal carbohydrate content was only found where transient diatom biofilms were present at the sediment surface and this was also found by Paterson et al. (2000). This surface elevation decreased with depth declining to background levels within 2 mm (2000 mm) of the sediment surface (Fig. 6). On sediments without a biofilm there was no increase in colloidal EPS from background levels. Similar results have been noted by Staats et al. (2001).

C. SEASONAL VARIATION IN EPS CONCENTRATIONS

Frostick and McCave (1979) first noted that variation in sediment dynamics resulted from seasonal changes in biological influences. This suggestion is supported by seasonal fluctuations in EPS, particularly the colloidal carbohydrate considered to be important in terms of sediment dynamics. In the Netherlands, recent work has confirmed the seasonal variation in EPS

220

G.J.C. UNDERWOOD AND D.M. PATERSON

fractions related to the increase in microphytobenthos during summer months and reduction in levels when diatom biomass was low in the winter (Staats et al., 2001; de Deckere et al., 2002). In general, these variations are more extreme moving up the intertidal as diatom biomass increases.

D. TIDAL VARIATION IN COLLOIDAL CARBOHYDRATE CONCENTRATIONS

A number of authors have noted an increase in colloidal carbohydrate concentration on the surface of intertidal sediments during the tidal exposure period (Underwood and Smith, 1998; van Duyl et al., 1999, 2000). These effects are usually most extreme where there is a visible surface biofilm (Fig. 2). The reported patterns vary. An increase in EPS in the early part of the tidal exposure period has often been measured, after which levels become more stable until near the end of the exposure when there may be a further elevation in concentration (Taylor, 1998; de Brouwer and Stal, 2001). Other studies have noted linear increases in carbohydrate (Underwood and Smith, 1998; Perkins et al., 2001) or little change in content or concentration (de Brouwer et al., 2000). The difference in results is usually related to time of year and/or development of the biofilm but may also reflect variation in the techniques used. Dewatering and compaction of the sediment can lead to unwarranted conclusions about the content and concentration of material in sediments if care is not taken. This is clearly demonstrated by Flemming and Delafontaine (2000) and this paper should be compulsory reading for anyone interested in studying sediment properties and the concentration or content of material on a microscale. The study of Perkins et al. (2003), although related specifically to Chl a, is also highly relevant, clearly demonstrating errors in interpretation that can occur when working on a fine scale. Ambient light climate can also influence the production and build-up of carbohydrate fractions on the sediment surface. In a detailed study, Perkins et al. (2001) demonstrated that shaded natural biofilms produced more EPS than ambient (unshaded) controls. Shaded biofilm showed a linear increase in EPS normalised to Chl a with time during the exposure period. The authors suggested that low light may stimulate movement and hence locomotive EPS production. This may also be the reason for increased levels of colloidal carbohydrate sometime observed towards the end of the emersion period as the diatom begin to migrate back into the sediment (Smith and Underwood, 1998; Taylor, 1998).

EXTRACELLULAR CARBOHYDRATE PRODUCTION

221

E. LOSS OF CARBOHYDRATES

Concentrations of colloidal carbohydrates in intertidal sediments are closely linked to algal biomass and rhythms in microalgal photosynthesis (Underwood and Paterson, 1993; Underwood et al., 1995; Smith and Underwood, 1998; Staats et al., 2000) and there is evidence that biofilmderived LMW carbohydrates are more labile (Smith and Underwood, 1998; Underwood and Smith, 1998b; Taylor et al., 1999) than the polymeric (EPS) component (Sundh et al., 1992; Karner and Rassoulzadegan, 1995; Sinsabaugh and Findlay, 1995). Polymeric molecules in the aquatic environment are cleaved by a range of hydrolytic, extracellular endo- and exo-acting enzymes (of microbial origin). Extracellular enzyme activities have been shown to be highly variable, altering on diel time scales (Karner and Raz, 1995), and inducible by organic matter composition and concentration (Pusch et al., 1999; van Duyl et al., 2000). Extracellular enzyme activity may also be enhanced by the polysaccharide matrix of a biofilm, which acts to maintain enzymes in close proximity to bacterial cells and can buffer changes in organic substrate concentrations (Freeman and Locke, 1995). Thus, extracellular carbohydrates are hypothesised to be a major component of the input into the sediment DOC cycle, but the processes involved in the fractionation and transformation of such biofilm-derived carbohydrates through the pools are not known although some pulse-chase experiments have begun to address this issue (Middleburg et al., 2000; Yallop et al., 2000). Periodic tidal exposure is associated with rapid changes in physical and chemical properties of the surface sediments (Christie et al., 2000), coupled with changes throughout the tidal emersion period in the production rates and composition of carbohydrate inputs from MPB biofilms (Smith and Underwood, 1998; Taylor, 1999). The exact fate of EPS as the tide returns is unknown but a number of authors have recorded the wash off of material and microphytobenthos into the water column (Bailie and Walsh, 1980; Wiltshire et al., 1998). This material may provide a rich source of organic material to pelagic bacteria in both freshwater (Tien et al., 2002) and marine systems or may become absorbed onto the surface of suspended particles (Eisma, 1998). Even in this case, the EPS continue to have a significant role in the dynamics of sediments since the concentration of EPS on the surface of particles influences the flocculation of the material and hence the deposition and retention of particles by the sediment bed (Dyer, 1986; Eisma, 1998). After the tidal inundation period, the concentration of EPS on the surface of the sediment is lowered, often returning to levels close to initial values.

222

G.J.C. UNDERWOOD AND D.M. PATERSON

However, under condition of high light and without sediment disturbance, there is general increase in the levels of colloidal carbohydrate at the beginning of each subsequent exposure period leading to a gradual temporal increase in EPS. This was shown by Paterson et al. (1990) in a laboratory system where diatom biofilms were maintained in tidal tanks for a period of 7 days. However, the authors pointed out that this accumulation does not usually continue. In nature, the biofilm may be disturbed by bioturbatory activity or by hydrodynamic stress (tides, waves, storms). However, if left undisturbed, then EPS may accumulate to such a level that the mucilaginous matrix becomes a semi-permanent feature of the sediment surface. When this occurs, a microbial mat develops on the surface of the sediment. This process produces the so-called ‘blister mats’ or ‘bubble mats’ described in the literature (Yallop et al., 1994). Here, the polymeric material has sufficient physical integrity to trap oxygen bubbles which give the mat its blistered appearance. Measurements of colloidal carbohydrate from such system have given some of the highest field records of EPS concentrations (Yallop et al., 1994). The hypothesis put forward by these authors was that as the polymer matrix increases in concentration and undergoes cycles of desiccation, it hardens and become less prone to re-hydration. Under these circumstances it has been noted that cell migration within the biofilm ceases and the mat becomes a permanent feature of the surface of the sediment. This was also suggested by Admiraal (1984) as a factor inhibiting diatom migration. F. IMPACT OF ENVIRONMENTAL STRESSES ON EXUDATION IN FIELD

Some authors have suggested that diatoms may respond to environmental stresses, including the presence of xenobiotic compounds, through variation in the nature and/or quantity of EPS produced. Decho (1990) suggested that EPS in the surrounding microenvironment may shield the diatom from toxins. There is some circumstantial evidence for this. Diatoms exposed to heavy metals often appear to alter their growth, forming closely adpressed cushion-like colonies (Miles, 1994). The mechanism proposed for this formation is the interaction between the heavy metal and the diatom polymers. EPS polymers form many cross linkages between adjacent strands, many of these linkages are weak and based on hydrogen bonding. Metal molecules, however, can form the basis of much stronger bonding between adjacent molecular chains, leading to the formation of a much more coherent gel structure (Decho, 1994). The nature of the bond depends partly on the metal, alkali earth metals form ‘bi-dendate’ ionic bridges

223

EXTRACELLULAR CARBOHYDRATE PRODUCTION

TABLE V The influence of metal concentration on the relative locomotive abilities of Naviculoid diatoms. A mixed culture of naviculoid cells were exposed to metal concentration (Pb, Cu, Cd and Al) ranging from 0.01 to 50 g ml1 for an exposure period 1 or 24 hours and the velocity of diatom locomotion recorded. A control value for the assemblage of 23 m s1 was recorded Metal Concentration (mg ml1) 0.01

0.1

1

10

50

Diatom velocity (mm s1 þ S.D.) Control ¼ 23.19 (10.25) Pb 1 h Pb 24 h Cu 1 h Cu 24 h Cd 1 h Cd 24 h Al 1 h Al 24 h

14.12 16.01 21.08 8.80 22.83 19.29 23.35 22.50

(4.6) (6.7) (5.6) (2.3) (6.1) (7.8) (8.3) (4.5)

19.18 21.23 21.11 0 20.05 20.97 22.55 17.85

(5.3) (8.5) (7.4) (4.2) (6.7) (5.5) (2.6)

20.04 16.95 20.31 0 19.47 6.74 23.83 19.06

(3.7) (5.8) (4.4) (5.5) (4.0) (2.3) (2.6)

19.61 0 18.43 0 21.36 0 21.09 0

(5.2) (6.0) (7.2) (3.6)

20.37(1.9) 0 19.68 (3.4) 0 21.14 (6.2) 0 22.34 (5.0) 0

Key: Pb ¼ lead, Cu ¼ Copper, Cd ¼ Cadmium, Al ¼ Aluminium.

and these metals can be replaced by transition metals which form more stable ‘multi-dentate bridges’ (Decho, 1990). The result is that growing colonies of diatoms become associated in a tight mucilaginous matrix. A by-product of this may be the local depletion of metal ion concentration around the diatom cells. Other effects of heavy metals on diatoms have been shown that also relate to the nature of EPS secretion. Studies by Barranguet et al. (2002) have noted the toxicity of heavy metals to diatoms in terms of photo-physiological performance, but it has also been shown that the rate of movement of epipelic forms varies with the concentration of the metal. In one study, the diatoms showed a consistent decrease in locomotive ability as the concentration of metal in the overlying medium increased (Table V). This effect may be due to reduced production and secretion of locomotive EPS by cadmium toxicity or because of changes in EPS conformation and viscosity caused by the availability metal ions or most probably a mixture of both factors. This question requires further study.

VI SEDIMENT-STABILITY, FLOCCULATION AND EPS PRODUCTION The analysis of the role and function of EPS in sediments is not a purely academic exercise. A major problem in current coastal research is the

224

G.J.C. UNDERWOOD AND D.M. PATERSON

requirement to model the transport pathways of particulate material through aquatic systems (Burt et al., 1997; Black et al., 2002). This work impinges on aspect of coastal management, maintenance of navigable waterways and the transport, retention and magnification of particleassociated pollutants. Sediments are often categorised in terms of their size and related behaviours. Particles greater than 63 mm in size are considered to act independently in response to hydrodynamic forcing. These particles are the non-cohesive sands and gravels. Particles smaller than 63 m may show inter-particle attraction, through a mechanism involving uneven distribution of surface charge, resulting in hydrogen bonding between adjacent surfaces. Thus, particles do not behave independently and show cohesive behaviour; these effects become more extreme as the particles become smaller, truly cohesive behaviour being manifest with particles below 20 mm in maximum dimension. While the modelling of non-cohesive sediment transport is welldeveloped (Soulsby, 1997), cohesive sediments provide a much more complex and intractable problem (Burt et al., 1997; Whitehouse et al., 2000). The nature of the inter-particle attraction varies with numerous external factors such as the electrolytic capacity of the pore waters, the packing of the material and the physiochemical environment. In addition, it is now clear that the organisms that inhabit sediments are not simply neutral exploiters of the habitat but, whether actively or passively, influences the structure and mechanical properties of the system introducing a biological element to an already complex problem (Black et al., 2002). The biological influence is highly variable but can be extremely significant in terms of the stability of particular system. The most pervasive effect cited in the literature is related to the production, release and biomechanical activity of EPS which ramifies through any sediment inhabited by microbial life. Other mechanisms of mechanical stabilisation can be more effective than EPS (Paterson and Black, 1999), but the occurrence of organic secretions in surficial sediment is ubiquitous.

A. MEASURING THE INFLUENCE OF EPS ON PHYSICAL DYNAMICS

Two major problems arise when scientists seek to understand the mechanisms of biogenic stabilisation of cohesive sediments. Firstly, it would be useful to have a reliable means of determining the stability of the sediment. Secondly, the quantity and quality of the EPS present must be determined. The first of these problems is beyond the scope of this review, but readers are referred to a number of recent papers and reviews on this

EXTRACELLULAR CARBOHYDRATE PRODUCTION

225

topic (Black and Paterson, 1997; Tolhurst et al., 2000a; Tolhurst et al., 2000b). However, it is clear that no consensus on methodology exists and that this area of research is still problematic. The second area is no less problematic. The analysis of EPS from sediments has largely been based on a few simple assay systems. The most common of these is the Dubois assay (Dubois et al., 1956), which is based on the acid hydrolysis of the mixed sample of carbohydrate into monomeric units which are then quantified by a colour reaction against a glucose standard curve (Underwood et al., 1995). The problem that arises here is that we understand very little of the mechanistic action of EPS polymers and their behaviour in sediments. The Dubois assay is based on an operational separation of the EPS found within the sediments. The logic is to separate easily extractable material (the so-called colloidal phase) from the more refractive material. This is done by varying the severity of the extraction procedure. The various fractions collected can then be correlated with the response of the sediment to erosive stress. This work has met with mixed results and there may be some fundamental reasons for this as outlined below. B. THE PROBLEM OF THE CONTROL

As outlined, EPS is ubiquitous in natural sediments. Therefore the absolute influence of EPS on sediment stability is extremely hard to determine and the majority of studies to date produce data of an ordinal or, at best, interval nature. The solution would be to produce abiotic control sediment. With sandy sediment this is possible by cleaning the sediment to hydrolyse away organic material. A reasonable control system can then be produced. This is not possible with cohesive sediments. The reactive nature of clay minerals means that any treatment of natural sediment to remove EPS has a fundamental effect on the sediment bed and can also influences the status of the clay mineral particles. An alternative is to employ a pure clay mineral such as kaolinite as a control. This has been used in laboratory experiments but does not replicate the behaviour of natural sediment which is a much more complex mix of grains with varying sizes and mineralogy. We therefore cannot yet produce an adequate control that would provide a clearer understanding of the pervasiveness and influence of EPS on natural systems. C. THE INFLUENCE OF EPS AND ORGANISMS ON SEDIMENT DYNAMICS

The evidence that organisms and, in particular diatoms, influence the physical dynamics of depositional systems has been accumulating for many

226

G.J.C. UNDERWOOD AND D.M. PATERSON

years. The earliest references to mucilage and slime date back to the nineteeth century when Paracelcus (1813) considered that the mucilage that developed on submerged rock was actually part of the process of rock formation. Later, Huxley (1868) noted the mucilaginous material found on the ocean bed and incorrectly considered it to be organic and alive. He named the ‘organism’ Bathybius haecklii and noted its role as an agent of biostabilisation. It is perhaps surprising that we are still discussing the exact nature and influence of mucilage in the environment after such a long span of time, but it seems that after these early studies the science of sedimentology and biology developed separately. It is only in the last 15–20 years that biological and sedimentological workers recognised the need for an interdisciplinary approach to this study. This work began after the field studies of Scoffin (1968, 1970) and the laboratory studies of Holland et al. (1974). Holland and co-workers determined that diatoms grown in culture on sediment influenced the erosional behaviour of the substratum. They also noted that different diatom species varied in terms of their stabilising potential and considered that this was due to the nature of the polymer produced. The importance of diatom stabilisation in nature was then supported by the work of Coles (1979) who brought forward the idea that diatom biofilms stabilised muddy sediments as part of a successional process leading to the development of saltmarsh systems from tidal flats. This work was perhaps not given the prominence it deserved. However, during the 1980s a number of papers were produced that began to formalise and expand our knowledge of biological interaction with sediment dynamics. This series of papers considered not only mechanical nature of the effects of organism on sediment and interaction with boundary flow conditions (Nowell et al., 1981, 1989), but also the development of the necessary theoretical background and hydrodynamic experience to replicate natural conditions (Nowell and Jumars, 1987). Thus, as biologists were beginning to consider the effects of organism so were some of the more innovative engineers. An early study by Parchure (1984) linked the supply of nutrients (and hence enhanced biological activity) to a reduction in the erosional potential of a clay mineral bed, maintained in a laboratory flume system. The effects were considered to be from an increase in microbial activity in the sediments. This work was highly innovative for its time and was a precursor to many future laboratory flume studies. The complexity of the natural system also led to the development of varied field flumes: devices to determine the natural stability of material in situ (Hawley, 1991; Childers and Day, 1988; Maa et al., 1993). It is interesting to note the drive to create such systems often came from biologists wishing to understand the ecology of natural systems rather than from sedimentologists trying to determine the natural erosive behaviour of sediments. However, attention from both

EXTRACELLULAR CARBOHYDRATE PRODUCTION

227

sedimentologists and biologist led to the development of several field-based flume systems which began to provide valuable information on the behaviour of natural sediments. A critical contribution in terms of defining the effects of organisms on sediments came from the work of Manzenrieder (1983) who employed a field flume to measure the erosion potential of natural sediments. He produced a mathematical measure of the effects of microbial assemblages, which he described as a biological stabilisation coefficient. Sb ¼

Verit ðbiolÞ Verit ðsterileÞ

ð2Þ

Under this system, the biological effect on erosion was always positive or neutral and varied from 1 (no effect) to the maximum (value of 8.7) that Manzenrieder reported. This system could not be followed for all sediment since the Verit sterile required a control and this is only possible for non-cohesive sandy sediments. Since the 1980s is has been accepted that biological activity influences sediment transport and that the major influence seems to be through the presence of EPS in sediments. The focus then turned to the search for proxy measurement that would serve as a predictive function for sediment transport. A number of possibilities were investigated with two major contenders emerging. The logic was that the stabilising effect of microbial assemblages would be related to their biomass. A convenient traditional proxy measure of biomass for photosynthetic organism is their Chl a content. The literature contains a number of studies where the distribution of Chl a has been used as a predictive parameter for sediment stability (Reithmuller et al., 1998), but these relationships may be highly site dependent because if the relationship rests on a correlation between Chl a and EPS this will only be valid under certain conditions. The site dependency of such relationships has been pointed out by Defew et al. (2002). An alternative approach to demonstrate the importance of biology and microbial assemblages in the field was to inhibit their activity. One of the first studies to try and remove microbiological activity from sediment and assess the influence on bed dynamics was carried out by de Boer (1981). His poisoning of a large area of sediment resulted in mass sediment loss from the foreshore. The mechanism may well have been through the loss of EPS production in the system and reduce sediment cohesion. Other factors that reduce microbial biomass may also act to influence sediment stability through the reduction in the production of EPS. The importance of grazing in this context was described by Daborn et al. (1993) reported a relationship between the abundance of the grazing amphipod, Corophium volutator,

228

G.J.C. UNDERWOOD AND D.M. PATERSON

diatom biomass and the stability of a shore in the Bay of Fundy (Nova Scotia). Similar result were found by Gerdol and Hughes (1994) inhibiting the amphipod Corophium volutator. This cascade relationship where the removal of grazers enhances biofilm development and influences sediment stability has now been supported by studies using a variety of methods. Underwood and Paterson (1993) used formalin to remove infaunal activity and noted increased sediment deposition and enhanced stability as diatom biomass, and EPS concentrations, increased. De Decker et al. (2001) also noted an increase in sediment stability after sediment were treated with an insecticide and this treatment also changed the nature of the erosion from the surface of the bed leading to lower erosion rates where infauna were removed.

D. THE EPS PROBLEM

As explained above, the nature of EPS in sediment is not well understood. It is known that the EPS secreted from diatoms is largely carbohydrate and several authors have reported a relationship between the content of some operational fraction of EPS and sediment stability for both intertidal and subtidal systems (Amos et al., 1998; Sutherland et al., 1998a, b; Yallop et al., 2000). The operational fraction with the most influence seems to be the colloidal extract from sediment sample. The problem of determining and predicting cohesive sediment transport is therefore inseparable from an understanding of the biology of the system and the dynamics of EPS (Fig. 7). It is not always the case that sediment stability is biologically mediated but when this occurs it can be a first order factor controlling the behaviour of the sediment. It is not surprising that the search for a predictive biological proxy for sediment stability has met with mixed results (Reithmuller et al., 1998; Defew et al., 2003). The results of these studies have been very varied with reasonable correlation produced under certain circumstances while providing no useful relationship in others (Houwing, 1999). The nature of the ‘binding capacity’ of EPS is still poorly known. De Brouwer et al. (2002) made the first attempts to examine the rheological behaviour of extracted EPS in sediment but found little effect. However, it appears that the in situ production of EPS is more effective than extracted material (Dade et al., 1990). To date, we have very little knowledge on how extraction procedures may alter the molecular structure and efficacy of polymer binding and this work should be repeated using a variety of extraction methods. Other approaches are now also being attempted to measure the relative binding capacity of molecules of EPS by biochemical

EXTRACELLULAR CARBOHYDRATE PRODUCTION

229

Fig. 7 Conceptual diagram of the production and loss terms of colloidal carbohydrate, within the surface of intertidal sediment. It should be noted that chemical transformations work in both directions. Bulk carbohydrate material can be broken-down by dissolution and extracellular exo- and endo-enzyme activity into smaller units while dehydration may lead to the aggregation of smaller units into large more recalcitrant molecules.

techniques (Lui and Fang, 2002) and by atomic force microscopy (Higgins et al., 2002). More direct approaches include the examination of the erosion process at a microscale using laser holography (Black et al., 2001; Sun et al., 2002). It seems certain that this combination of approaches into the binding capacity of selected EPS under different conditions will greatly enhance our knowledge of the effects of EPS in situ. However, there still remains the problem of EPS extraction from sediments. The separation of EPS into various operational fractions has been explained. The physical properties and structure of EPS will also be radically affected by the extraction and isolation procedures. The evidence for this hypothesis is still sparse but supported by the experiments of Dade et al. (1990) in which EPS extracted from bacteria has less effect on sediment stability than the natural population in situ (Dade et al., 1990). This may be part of the problem encountered by de Brouwer et al. (2002) in his study of rheological effects.

230

G.J.C. UNDERWOOD AND D.M. PATERSON

It may be some time before we are certain of the conformational arrangement and binding capacity of EPS in its natural milieu.

VII

AREAS OF FURTHER WORK

Despite their importance in aquatic systems, we still have an incomplete understanding of how the composition and dynamics of extracellular carbohydrates by benthic diatoms changes in response to environmental conditions. Benthic diatoms can be subject to nutrient limitation, photoinhibitory irradiances (including UVB), and salinity changes (Underwood et al., 1998; Underwood and Kromkamp, 1999), all interposed over their endogenous rhythms of migration. A major challenge is to fully understand the dynamics and variability of EPS production in natural systems (Fig. 7). The major areas for future research can be summarised as follows: . . . .

Understanding the biochemical nature of the EPS produced. Understanding the factors that control EPS production in situ. Determining the importance of EPS as a vector for carbon flow. Modelling the effect EPS on the sediment erosion, transport and deposition cycle.

The limited data currently available shows that there are significant differences in the properties of EPS produced by diatoms under different conditions. This suggests that the chemical and physical properties of EPS produced by natural biofilms in the field may vary significantly with environmental conditions and the physiological status of the diatom cells. Differences in sugar composition, size and structural form, as well as environment conditions will influence the degree to which polymers bind to sediments, respond to dehydration and are degraded by bacteria (Fig. 7). The mechanisms by which diatoms change the types of EPS being produced are not understood. Are changes in composition just a response to high light stress and nutrient limitation or are the cells responding to environmental variations at a more subtle level? Developments in the field of cell–environment signalling and genomic and proteomic approaches may be useful here. These challenges will require considerable effort and the development of new methodologies and approaches. As long ago as 1989, Alberts stated ‘With the current methods it takes longer to determine the structure of half a dozen linked sugars than to determine the nucleotide sequence of a DNA molecule containing many thousands of nucleotides’. While work on DNA has been revolutionised since this time and many new techniques are available to

EXTRACELLULAR CARBOHYDRATE PRODUCTION

231

examine EPS structure, their use for environmental samples is still limited and our knowledge of this is only increasing slowly.

ACKNOWLEDGEMENTS Preparation of this article was partly supported by NERC grants GT4/96/ 96/MAS and NER/A/S/2001/00536 and NFS grant IBN-0110875 to GJCU and NERC grant NER/A/S/2000/00513, EU Awards HIMOM (EVK3 CT2001-00052) and TIDE (EVK3-CT-2001-00064) to DMP.

REFERENCES Admiraal, W. (1984). The ecology of estuarine sediment-inhabiting diatoms. Biopress 3, 269–322. Alberts, B. (1989). Small molecules, energy and biosynthesis. In ‘Molecular Biology of the Cell.’ (Alberts, B., ed) p 43. Garland Publishing House, New York. Alcoverro, T., Conte, E. and Mazzella, L. (2000). Production of mucilage by the Adriatic epipelic diatom Cylindrotheca closterium (Bacillariophyceae) under nutrient limitation. Journal of Phycology 36, 1087–1095. Allan, G. G., Lewin, J. and Johnson, P. G. (1972). Marine polymers. IV. Diatom polysaccharides. Botanica Marina 15, 102–108. Amos, C. L., Brylinski, M., Sutherland, T. F., O’Brian, D., Lee, S. and Cramp, A. (1998). The stability of mudflats in the Humber estuary. In ‘Sedimentary Processes in the Intertidal Zone’ (K. Black, A. Cramp and D. M. Paterson, eds.). Geological Society of London. Barranguet, C., Plans, M., van der Grinten E., Sinke J. J. and Admiraal, W. (2002). Development of photosynthetic biofilms affected by dissolved and sorbed copper in a eutrophic river. Environmental Toxicology and Chemistry 21(9), 1955–1965. Bhosle, N. B., Sawant, S. S., Garg, A. and Wagh, A. B. (1995). Isolation and partial chemical analysis of exopolysaccharides from the marine fouling diatom Navicula subinflata. Botanica Marina 38, 103–110. Black, K. S., Sun, H., Craig, G., Paterson, D. M., Watson, J. and Tolhurst, T. (2001). Incipient erosion of biostabilized sediments examined using particle-field optical holography. Environmental Science and Technology, 35, 2275–2281. Black, K. S., Tolhurst, T. J., Hagerthey, S. E., and Paterson, D. M. (2002). Working with Natural Cohesive Sediments. Journal of Hydraulic Engineering, 128, 1–7. Blanchard, G. F., Paterson, D. M., Stal, L. J., Richard, P., Galois, R., Huet, V., Kelly, J., Honeywill, C., de Brouwer, J., Dyer, K., Christie, M. and Seguines, M. (2000). The effect of geomorphological structures on potential biostabilisation by microphytobenthos on intertidal mudflats. Continental Shelf Research 20, 1243–1256. Boudreau, B. P. and Jorgensen B. B. (2001). The benthic boundary layer: Transport processes and biogeochemistry. pp. 404, Oxford University Press, Oxford. Boulcott, M. (2001). The dynamics of carbohydrate production, storage and excretion in epipelic diatoms. Ph.D. thesis, University of Essex.

232

G.J.C. UNDERWOOD AND D.M. PATERSON

Burt, N., Parker, R. and Watts, J. (eds.). (1997). ‘Cohesive Sediments’. pp. 458. Wiley and Sons, UK. Cahoon, L. B. (1999). The role of benthic microalgae in neritic ecosystems. Oceanography and Marine Biology: an Annual Review 37, 47–86. Childers, D. L. and Day, J. W. (1988). A flow-through flume technique for quantifying nutrient and materials fluxes in microtidal estuaries. Estuarine and Coastal Shelf Science 16, 483–494. Christie, M. C., Dyer, K.R., Blanchard, G., Cramp, A., Mitchener, H. J. and Paterson D. M. (2000). Temporal and spatial distributions of moisture and organic contents across a macro-tidal mudflat. Continential Shelf Research 20, 1219–1242. Coles, S. M. (1979). Benthic microalgal populations on intertidal sediments and their role as precursors to salt marsh development. In ‘Ecological Processes in Coastal Environments: The First European Symposium of the British Ecological Society’ (R. C. Jeffries and A. J. Davies, eds.), pp. 25–42. Blackwell Scientific, Oxford. Consalvey, M. (2002). The structure and function of microphytobenthic biofilm. Ph.D. Thesis, University of St Andrews. Cooksey, K. and Cooksey, B. (1986). Adhesion of fouling diatoms to surfaces: Some biochemistry. In ‘Algal Biofouling. Studies in Environmental Science’. (L. V. Evans and K. D. Hoagland, eds.), p. 28. Elsevier, Amsterdam. Daborn, G. R., Amos, C. L., Berlinsky, M., Christian, H., Drapeau, G., Faas, R. W., Grant, J., Long, B., Paterson, D. M., Perillo, G. M. E. and Piccolo, M. C. (1993). An ecological ‘cascade’ effect: Migratory birds affect stability of intertidal sediments. Limnology and Oceanography 38(1), 225–231. Dade, B. W., Davis, J. D., Nichols, P. D., Nowell, A. R. M., Thistle, D., Trexler, M. B. and White, D. C. (1990). Effects of bacterial exopolymer adhesion on the entrainment of sand. Journal of Geomicrobiology 8, 1–16. Daniel, G. F., Chamberlain, H. L. and Jones, E. B. G. (1980). Ultrastructure observations on the marine fouling diatom Amphora. Helgola¨nder Meeresuntersuchungen 34, 123–149. Daniel, G. F., Chamberlain, A. H. L. and Jones, E. B. G. (1987). Cytochemical and electron microscopical observations on the adhesive materials of marine fouling diatoms. British Phycology 22, 101–118. de Angelis, F., Barbarulo, M. V., Bruno, M., Volterra, L. and Nicoletti, R. (1993). Chemical composition and biological origin of ‘dirty sea’ mucilages. Phytochemistry 34, 393–395. de Boer, P. L. (1981). Mechanical effects of micro-organisms on intertidal bedform migration. Sedimentology 28, 129–132. de Brouwer, J. F. C. and Stal, L. J. (2001). Short-term dynamics in microphytobenthos distribution and associated extracellular carbohydrates in surface sediments of an intertidal mudflat. Marine Ecology Progress Series 218, 33–44. de Brouwer, J. F. C. and Stal, L. J. (2002). Daily fluctuations of expolymers in cultures of the benthic diatoms Cylindrotheca closterium and Nitzschia sp. (Bacillariophyceae). Journal of Phycology 38, 464–472. de Brouwer, J. F. C., Bjelic, S., de Deckere, E. M. G. T. and Stal, L. J. (2000). Interplay between biology and sedimentology in a mudflat (Biezelingse Ham, Westeschelde, The Netherlands). Continental Shelf Research 20, 1159–1177. de Brouwer, J. F. C., Ruddy, G. K., Jones, T. E. R. and Stal, L. J. (2002) Sorption of EPS to sediment particles and the effect on the rheology of sediment slurries. Biogeochemistry, 61, 57–71.

EXTRACELLULAR CARBOHYDRATE PRODUCTION

233

de Brouwer, J. F. C., Wolfstein, K. and Stal, L. J. (2002). Physical characterisation and diel dynamics of different fractions of extracellular polysaccharides in an axenic culture of a benthic diatom. European Journal of Phycology 37, 37–44. de Brouwer, J. F. C., de Deckere, E. M. G. T. and Stal, L. J. (2003). Distribution of extracellular carbohydrates in three mudflats in west-Europe. Estuarine and Coastal and Shelf Science 56, 313–324. de Deckere, E. M. G. T., Kornman, B. A., Staats, N., Termaat, G. R., de Winder, B., Stal, L. J. and Heip, C. H. R. (2002). The seasonal dynamics of benthic (micro) organisms and extracellular carbohydrates in an intertidal mudflat and their effect on the concentration of suspended matter. In ‘Fine sediment dynamics in the Marine Environment’ (J. C. Winterwerp and C. Kranenberg, eds.). Proceedings of INTERCOH-2000, Elsevier. Decho, A. W. (1990). Microbial exopolymer secretions in ocean environments: their role(s) in food webs and marine processes. Oceanography Marine Biology Annual Review 28, 73–153. Decho, A. W. (1994). Molecular-scale events influencing the macroscale cohesiveness of exopolymers. In ‘Biostabilization of Sediments’ Oldenburg: Bibliotheks. und Informationssystem der Universita¨t Oldenburg (BIS)–Verlag’ (W. E. Krumbein, D. M. Paterson and L. J. Stal, eds.). pp. 135–148. Decho, A. W. (2000). Microbial biofilms in intertidal systems: an overview. Continental Shelf Research 20, 1257–1274. Defew, E. C., Tolhurst, T. J. and Paterson, D. M. (2002). Site-specific features influence sediment stability of intertidal mudflats. Hydrology and Earth Systems Science 6, 971–982. Defew, E. C., Tolhurst, T. J., Paterson, D. M. and Hagerthey, S. E. (2003). Can the stability of intertidal sediments be predicted from proxy parameters? An in situ investigation. In ‘Coastal zone topics 5: The estuaries and coasts of north-east Scotland.’ (Raffaelli, D., Solan, M., Paterson, D. M., Buck, A. L. and Pomfret, J. R., eds.) Estuarine Coastal Sciences Association. 61–70 Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A. and Smith, F. (1956). Colorimetric method for determination of sugars and related substances. Analytical chemistry 28, 350–356. Dyer, K. R. (1986). Estuarine and Coastal Sediment Dynamics. Wiley Interscience, 342pp, Chichester, UK. Eisma, D. (1998). ‘Intertidal Deposits’. CRC Press, Boca Raton. Edgar, L. A. and Pickett-Heaps, J. D. (1982). Ultrastructural localisation of polysaccharides in the motile diatom Navicula cuspidata. Protoplasma 113, 10–22. Edgar, L. A. and Pickett-Heaps, J. D. (1984). Diatom locomotion. Progress in Phycology Research 3, 47–88. Edgar, L. A. and Zavortink, M. (1983). The mechanism of diatom locomotion. II. Identification of actin. Proceedings of the Royal Society London B2 18, 345–348. Falciatore, A. and Bowler, C. (2002). Revealing the molecular secrets of marine diatoms. Annual Review of Plant Biology 53, 109–130. Flemming, B. W. and Delafontaine, M. T. (2000). Mass physical properties of muddy intertidal sediments: some applications, misapplications and non-application. Continental Shelf Research 20(10–11) 1179–1197. Fogg, G. E. and Thake, B. (1987). ‘Algal cultures and phytoplankton ecology’. University of Wisconsin Press, London.

234

G.J.C. UNDERWOOD AND D.M. PATERSON

Freeman, C. and Locke, M. A. (1995). The biofilm polysaccharide matrix: A buffer against changing organic substrate supply? Limnology and Oceanography 40, 273–278. Frostick, L. E. and McCave, I. N. (1979). Seasonal shifts of sediment within an estuary mediated by algal growth. Estuarine and Coastal and Marine Science 9, 569–576. Gerdol, V. and Hughes, R. G. (1994). Effect of corophium volutator on the abundance of benthic diatoms, bacteria and sediment stability in 2 estuaries in south-eastern England. Marine Ecology-Progress Series 114, 109–115. Gordon, R. (1987). A retaliatory role for algal projectiles, with implications for the mechanochemistry of diatom gliding motility. Journal Theoretical Biology 126, 419–436. Gordon, R. and Drum, R. W. (1970). A capillarity mechanism for diatom gliding locomotion. Proceedings National Academy of Science USA 67, 338–344. Goto, N., Kawamura, T., Mitamura, O. and Terai, H. (1999). Importance of extracellular organic carbon production in the total primary production by tidal-flat diatoms in comparison to phytoplankton. Marine Ecology Progress Series 190, 289–295. Haug A. and Myklestad, S. (1976). Polysaccharides of marine diatoms with special reference to Chaetoceros species. Marine Biology 34, 217–222. Hawley, N. (1991). Preliminary observations of sediment erosion from a bottom resting flume. Journal Great Lakes Research 17, 361–367. Holland, A. F., Zingmark, R. G. and Dean, J. M. (1974). Quantitative evidence concerning the stabilization of sediments by marine benthic diatoms. Marine Biology 27, 191–196. Hay, S. I., Maitland, T. C. and Paterson, D. M. (1993). The speed of diatom migration through natural and artificial substrata. Diatom Research 8, 371–384. Higgins, M. J., Crawford S. A., Mulvaney, P. and Wetherbee, R. (2000). The topography of soft, adhesive diatom ‘trails’ as observed by Atomic Force Microscopy. Biofouling 16, 133–139. Higgins, M. J., Crawford, S. A., Mulvaney, P. and Wetherbee, R. (2002). Characterization of the adhesive mucilages secreted by live diatom cells using atomic force microscopy. Protist 153, 25–28. Hoagland, K. D., Rosowski, J. R., Gretz, M. R. and Roemer, S.C. (1993). Diatom extracellular polymeric substances: function, fine structure, chemistry and physiology. Journal Phycology 29, 537–566. Houwing, E. J. (1999). Determination of the critical erosion threshold of cohesive sediments on intertidal mudflats along the Dutch Wadden Sea coast. Estuarine and Coastal Shelf Science 49, 545–555. Huxley, T. H. (1868). On some organisms living at great depths in the North Atlantic Ocean. Quarterly Journal of Microscopy, VIII, 203–212. Jarosch, R. (1962). Gliding. In ‘Physiology and Biochemistry of Algae’ (R. A. Lewin ed.), pp. 573–581. New York, Academic Press. Karner, M. and Rassoulzadegan, F. (1995). Extracellular enzyme activity: indications for high short-term variability in a coastal marine ecosystem. Microbial Ecology 30, 143–156. Kazy. S. K., Sar, P., Singh, S. P., Sen, A. K. and D’Souza, S. F. (2002). Extracellular polysaccharides of a copper-sensitive and a copper-resistant Pseudomonas aeruginosa strain: synthesis, chemical nature and copper binding. World Journal of Microbiology and Biotechnology 18(6), 583–588.

EXTRACELLULAR CARBOHYDRATE PRODUCTION

235

Kelly, J., Honeywill, C. and Paterson, D. M. (2001). Microscale analysis of chlorophyll-a in cohesive, intertidal sediments: the implications of microphytobenthos distribution. Journal of Marine Biology and Ecology 81, 151–162. Laspidou, C. S. and Rittmann, B. E. (2002). A unified theory for extracellular polymeric substances, soluble microbial products, and active and inert biomass. Water Research 36(11), 2711–2720. Lewis, R. J., Johnson, L. M. and Hoagland. K. D. (2002). Effects of cell density, temperature, and light intensity on growth and stalk production in the biofouling diatom Achnanthes longipes (Bacillariophyceae). Journal of Phycology 38(6), 1125–1131. Lind, J. L., Heimann, K., Miller, E. A., van Vliet, C., Hoogenraad, N. J. and Wetherbee R. (1997). Substratum adhesion and gliding in a diatom are mediated by extracellular proteoglycans. Planta 203, 213–221. Liu, H. and Fang, H. H. P. (2002). Characterization of electrostatic binding sites of extracellular polymers by linear programming analysis of titration data. Biotechnology and Bioengineering 80(7), 806–811. Maa, J. P.-Y., Wright, L. D., Lee, C. H. and Shannon, T. W. (1993). VIMS Sea carousel – a field instrument for studying sediment transport. Marine Geology 115(3–4), 271–287. MacIntyre, H. L., Geider R. J. and Miller, D. C. (1996). Microphytobenthos: The ecological role of the ‘secret garden’ of unvegetated, shallow-water marine habitats. I. Distribution, abundance and primary production. Estuaries 19, 86–201. Madsen, K. N., Nilsson, P. and Su¨ndback, K. (1993). The influence of benthic microalgae on the stability of a subtidal shallow water sediment. Journal Experimental Marine Biology Ecology 170, 159–177. Mann, D. G. (1999). The species concept in diatoms. Phycologia 38(6), 437–495. Mann, D. G. and Droop, S. J. M. (1996). Biodiversity, biogeography and conservation of diatoms. Hydrobiologia 336(1–3), 19–32. Manzenrieder, H. (1983). Retardation of initial erosion under biological effects in sandy tidal flats. Leichtweiss, Inst. Tech. University Braunschweig. pp. 469–479. McConville, M. J., Wetherbee, R. and Bacic, A. (1999). Subcellular location and composition of the wall and secreted extracellular sulphated polysaccharides/proteoglycans of the diatom Stauroneis amphioxys Gregory. Protoplasma 206, 188–200. Medlin, L. K. (2002). Why silica or better yet why not silica ? Speculations as to why the diatoms utilise silica as their cell wall material. Diatom Research 17, 453–459. Middleburg, J. J., Barranguet, C., Boschker, H. T. S., Herman, P. M. J., Moens, T. and Heip, C. H. R. (2000). The fate of intertidal microphytobenthos carbon: an in situ 13C-labelling study. Limnology Oceanography 45, 1224–1234. Miles, A. (1994). Heavy metals in estuarine systems. Ph.D Thesis, University of Bristol. Myklestad, S. (1988). Production, chemical structure, metabolism and biological function of the (1–3)-linked-B-D-glucans in diatoms. Biological Oceanography 6, 313–326. Myklestad, S. and Haug, A. (1972). Production of carbohydrates by the marine diatom Chaetoceros affinis var. Willei (Gran) Hustedt I. Effect of the concentration of nutrients in the culture medium. Journal of Experimental Marine Biology and Ecology 9, 125–136. Nicolaus, B., Lama, L., Panico, A., Moriello, V. S., Romano, I. and Gambacorta, A. (2002). Production and characterization of exopolysaccharides excreted

236

G.J.C. UNDERWOOD AND D.M. PATERSON

by thermophilic bacteria from shallow, marine hydrothermal vents of Flegrean Ares (Italy). Systematic and Applied Microbiology 25(3), 319–325 OCT 2002. Nowell, A. R. M. and Jumars, P. A. (1987). Flumes: theoretical and experimental considerations for simulation of benthic environments. Oceanography Marine Biology Annual Review 25, 91–112. Nowell, A. R., Jumars, P. A. and Eckman, J. E. (1981). Effects of biological activity on the entrainment of marine sediments. Marine Geology. 42, 133–153. Nowell, A. R. M., Jumars, P. A., Self, R. F. L. and Southard, J. B. (1989). The effects of sediment transport and deposition on infauna: Results obtained in a specially designed flume. In ‘Ecology of Marine Deposit Feeders’ (G. Lopez, G. Taghon and J. Levington eds.), Vol. 31, pp. 245–266. Lecture Notes on Coastal and Estuarine Studies. Nowell, A. R. M., McCave, I. N. and Hollister, C. D. (1985). Contributions of HEBBLE to understanding marine sedimentation. Marine Geology 66, 397–409. Paracelcus (1813). cited in Krumbein, W. E., Paterson, D. M. & Stal, L. J. (1994). Biostabilization of sediments. Carl von Ossietzky Universita¨t Oldenburg, Germany. Parchure, T. M. (1984). Erosional behavior of deposited cohesive sediments. Ph. D. Thesis, University of Florida. Paterson, D. M. (1986). The migratory behaviour of diatom assemblages in a laboratory tidal micro-ecosystem examined by low-temperature scanning electron microscopy. Diatom Research 1, 227–239. Paterson, D. M. (2001). The fine-structure and properties of the sediment interface. In ‘The Benthic Boundary Layer: Transport, Processes and Biogeochemistry’ (B. B. Boudreau and B. B. Jorgensen, eds.), pp. 127–143 Oxford University Press, Oxford. Paterson, D. M. and Black, K. S. (1999). Water flow, sediment dynamics and benthic biology. Advances in Ecological Research 29, 155–193. Paterson, D. M. and Hagerthey, S. E. (2001). Microphytobenthos in contrasting coastal ecosystems: Biology and dynamics. In ‘Ecological Comparisons of Sedimentary shores’ (K. Reise, ed.), pp. 151, 105–125. Ecological studies. Paterson, D. M., Crawford R. M. and Little, C. (1986). The structure of benthic diatom assemblages: A preliminary account of the use and evaluation of low-temperature scanning electron microscopy. Journal Experimental Marine Biology and Ecology 96, 279–289. Paterson, D. M., Tolhurst, T. J., Kelly, J., Honeywill, C., de Deckere, E. M. G. T., Huet, V., Shayler, S. A., Black, K. S., de Brouwer, J. F. C. and Davidson, I. (2000). Variations in sediment stability and biogeochemical parameters across Skeffling mudflat, Humber Estaury. Continental Shelf 20(10–11), 1373–1396. Percival, E., Rahman, M. A. and Weigel, H. (1980). Chemistry of the polysaccharides of the diatom Coscinodiscus nobilis. Phytochemistry 19, 809–811. Perkins, R. G., Honeywill, C., Consalvey, M., Austin, H. A., Tolhurst, T. J. and Paterson, D. M. (2003). Changes in microphytobenthic chlorophyll a and EPS resulting from sediment compaction due to de-watering: opposing patterns in concentration and content. Continental Shelf Research 23, 575–586. Perkins, R. G., Underwood, G. J. C., Brotas, V., Snow, G. C., Jesus, B. and Ribeiro, L. (2001). Responses of microphytobenthos to light: primary production

EXTRACELLULAR CARBOHYDRATE PRODUCTION

237

and carbohydrate allocation over an emersion period. Marine Ecology Progress Series 223, 101–112. Poulsen, N. C., Spector, I., Spurck T. P., Schultz T. F. and Wetherbee R. (1999). Diatom gliding is the result of an actin-myosin motility system. Cell Motility and the Cytoskeleton 44, 23–33. Pusch, M., Fiebig, D., Brettar, I., Eisenmann, H., Ellis B. K., Kaplan, L. A., Lock, M. A., Naegeli, M. W. and Traunspurger, W. (1999). The role of microorganisms in the ecological connectivity of running waters. Freshwater Biology 40, 453–495. Reithmuller, R., Hakvoort, J. H. M., Heinke, M., Heymann, K., Khul, H. and Witte. G. (1998). Relating erosion shear stress to tidal flat surface colour. In ‘Sedimentary processes in the intertidal zone’ (K. S. Black, D. M. Paterson and A. Cramp, eds.) Vol. 139, pp. 1–10. Revsbech, N. P. (1994). Analysis of microbial mats by use of electrochemical microsensors: recent advances. In ‘Microbial Mats, NATO ASI Series’ (L. J. Stal and P. Caumette, eds.), 35, 135–147. Round, F. E., Crawford, R. M. and Mann, D. G. (1990). The Diatoms, biology and morphology of the genera. Cambridge University Press, Cambridge. Scoffin, T. P. (1968). An underwater flume. Journal Sedimentary Petrology 38, 44–247. Scoffin, T. P. (1970). The trapping and binding of subtidal carbonate sediments by marine vegetation in Bimini Lagoon, Bahamas. Journal Sediment Petrology 40, 249–273. Seroˆdio, J., da Silva, J. M. and Catarino, F. (1997). Non-destructive tracing of migratory rhythms of intertidal benthic microalgae using in vivo chlorophyll a fluorescence. Journal Phycology 33, 542–553. Sinsabaugh, R. L. and Findlay, S. (1995). Microbial production, enzyme activity and carbon turnover in surface sediments of the Hudson River estuary. Microbiology Ecology 30, 127–141. Smith, D. J. and Underwood, G. J. C. (1998). Exopolymer production by intertidal epipelic diatoms. Limnology Oceanography 43, 1578–1591. Smith, D. J. and Underwood, G. J. C. (2000). The production of extracellular carbohydrate exopolymers (EPS) by estuarine benthic diatoms: the effects of growth phase and light and dark treatment. Journal of Phycology 36, 321–333. Staats, N., de Winder, B., Stal, L. J. and Mur, L. R. (1999). Isolation and characterisation of extracellular polysacchardies from the epipelic diatoms Cylindrotheca closterium and Navicula salinarum. European Journal of Phycology 34, 161–169. Staats, N., Stal, L. J., de Winder, B. and Mur, L. R. (2000a). Oxygenic photosynthesis as driving process in exopolysaccharide production of benthic diatoms. Marine Ecology Progress Series 193, 261–269. Staats, N., Stal, L. J. and Mur, L. R. (2000b). Exopolysaccharide production by the epipelic diatom Cylindrotheca closterium: effects of nutrient conditions. Journal of Experimental Marine Biology and Ecology 249, 3–27. Staats, N., de Deckere, E. M. G. T., de Winder, B. and Stal, L. J. (2001). Spatial patterns of benthic diatoms. carbohydrates and mud on a tidal flat in the Ems-Dollard estuary. Hydrobiologia 448, 107–115. Stal, L. J., Villbrandt, M. and de Winder, B. (1994). EPS in benthic phototrophic microorganisms. In ‘Biostabilization of Sediments’ (W. E. Krumbein, D. M. Paterson and L. J. Stal eds.), pp. 377–384. Bibliotheks und Informationssystem der Universita¨t Oldenburg (BIS)-Verlag, Oldenburg.

238

G.J.C. UNDERWOOD AND D.M. PATERSON

Sun, H., Dong, H., Paterson, D. M., Perkins, R., Player, M. A. and Watson, J. (2002). In-line digital video holography for the study of erosion processes in sediments. Measurement Science and Technology. 13, 7–12. Sundh, J. (1992). Biochemical composition of dissolved organic carbon derived from phytoplankton and used by heterotrophic bacteria. Applied Environmental Microbiology 58, 2938–2947. Sutherland, T. F, Grant, J. and Amos, C. L. (1998a). The effect of carbohydrate production by the diatom Nitzschia curvilineata on the erodibility of sediment Limnology and Oceanography, 43(1), 65–72. Sutherland,T. F, Amos, C. L. and Grant, J. (1998b). The effect of buoyant biofilms on the erodibility of sublittoral sediments of a temperate microtidal estuary. Limnology and Oceanography 43(2), 225–235. Taylor, I. S. (1998). Fine-scale distribution of carbohydrates on intertidal sediments in relation to diatom biomass and sediment properties. Ph.D Thesis, University of St Andrews, pp. 211. Taylor, I. and Paterson, D. M. (1998). Microspatial variation in carbohydrate concentrations with depth in the upper millimetres of intertidal cohesive sediments. Estuarine and Coastal Shelf Science 46, 359–370. Taylor, I., Paterson, D.M., and Mehlert, A. (1999). The quantitative variability and monosaccharide composition of sediment carbohydrates associated with intertidal diatom assemblages. Biogeochemistry 45(3), 303–327. Thornton, D. C. O., Dong, L. F., Underwood, G. J. C. and Nedwell, D. B. (2002). Factors affecting microphytobenthic biomass, species composition and production in the Colne estuary (UK) Aquatic Microbial Ecology 27, 285–300. Tien, C. J., Krivtsov, V., Levado, E., Sigee, D. C. and White, K. N. (2002). Occurrence of cell-associated mucilage and soluble extracellular polysaccharides in Rostherne Mere and their possible significance. Hydrobiologia, 485(1–3), 245–252. Tolhurst, T. J., Riethmu¨ller, R. and Paterson, D. M. (2000a). In situ versus laboratory analysis of sediment stability from intertidal mudflats. Continental Shelf Research 20(10–11), 1317–1334. Tolhurst, T. J., Black, K. S., Paterson, D. M., Mitchener, H., Termaat, R. and Shayler, S. A. (2000b). Comparison of four in situ sediment erosion devices. Continental Shelf Research 20(10–11), 1397–1418. Underwood, G. J. C. (1997). Microalgal colonisation in a saltmarsh restoration scheme. Estuarine and Coastal Shelf Science 44, 471–481. Underwood, G. J. C. (2002). Adaptations of tropical marine microphytobenthic assemblages along a gradient of light and nutrient availability in Suva Lagoon, Fiji. European Journal of Phycology 37, 449–462. Underwood, G. J. C. and Kromkamp, J. (1999). Primary production by phytoplankton and microphytobenthos in estuaries. Advances in Ecological Research 29, 93–153. Underwood, G. J. C. and Smith, D. J. (1998a). Predicting epipelic diatom exopolymer concentrations in intertidal sediments from sediment chl. a Microbial Ecology 35, 116–125. Underwood, G. J. C. and Smith, D. J. (1998b). In situ measurements of exopolymer production by intertidal epipelic diatom-dominated biofilms in the Humber estuary. In ‘Sedimentary Processes in the Intertidal Zone’. (K. S. Black, D. M. Paterson and A. Cramp eds.). Geological Society, London. Special Publications 139, 125–134.

EXTRACELLULAR CARBOHYDRATE PRODUCTION

239

Underwood, G. J. C., Paterson, D. M. and Parkes, R. J. (1995). The measurement of microbial carbohydrate exopolymers from intertidal sediments. Limnology and Oceanography 40, 1243–1253. Underwood, G. J. C., Phillips, J. and Saunders, K. (1998). Distribution of estuarine benthic diatom species along salinity and nutrient gradients. European Journal of Phycology 33, 173–183. Van den Hoek, C., Mann, D. G. and Jahns, H. M. (1995). Algae: An introduction to phycology. Cambridge, pp. 623. van Duyl, F. C., de Winder, B., Kop, A. J. and Wollenzien, U. (1999). Tidal coupling between carbohydrate concentrations and bacterial activities in diatom-inhabited intertidal mudflats. Marine Ecology Progress Series 191, 19–32. van Duyl, F. C., de Winder, B., Kop A. J. and Wollenzien, U. (2000). Consequences of diatom mat erosion for carbohydrate concentrations and hetertrophic bacterial activities in intertidal sediments of the Ems-Dollard estuary. Continential Shelf Research 20, 1335–1349. Varum, K. M. and Myklestad S. (1984). Effects of light, salinity and nutrient limitation on the production of -1, 3-D-Glucan and Exo-D-Glucanase activity in Skeletonema costatum (Grev.) Cleve. Journal of Experimental Marine Biology and Ecology 83, 13–25. Wang, Y., Chen, Y., Lavin, C. and Gretz, M. R. (2000). Extracellular matrix assembly in diatoms (Bacillariophyceae). IV. Ultrastructure of Achnanthes longipes and Cymbella cistula as revealed by high-pressure freezing/freeze substitution and cryo-field emission scanning electron microscopy. Journal of Phycology 36, 367–378. Wetherbee, R., Lind, J. L., Burke, J. and Quatrano, R. S. (1998). The first kiss: establishments and control of initial adhesion by raphid diatoms. Journal of Phycology 34, 9–15. Widdows, J., Brown, S., Brinsley, M. D., Salkeld, P. N. and Elliott, M. (2000). Temporal changes in intertidal sediment erodability: influence of biological and climatic factors. Continental Shelf Research 20, 1275–1289. Wiltshire, K. H., Blackburn, J. and Paterson, D. M. (1997). The cryo-lander: a new method for in situ sampling of unconsolidated sediments minimising the distortion of sediment fabric. Journal of Sediment Research 67, 980–984. Wiltshire, K. H., Tolhurst, T., Paterson, D. M., Davidson, I. and Gust, G. (1998). Pigment fingerprints as markers of erosion and changes in cohesive sediment surface properties in simulated and natural erosion events. In ‘Sedimentary Processes in the Intertidal Zone’ (K. S. Black, D. M. Paterson and A. Cramp eds.), Vol. 139, pp. 99–114. Geological. Society, London, Special publication. Wolfstein, K. and Stal, L. J. (2002). Production of extracellular polymeric substances (EPS) by benthic diatoms: effect of irradiance and temperature. Marine Ecology Progress Series 236, 13–22. Wulff, S., Wa¨ngberg, A˚., Sundba¨ck, K., Nilsson, C. and Underwood, G. J. C. (2000). Effects of UVB radiation on a marine microbenthic community growing on a sand-substratum under different nutrient conditions. Limnology Oceanography 45, 1144–1152. Wustman, B. A., Gretz, M. R. and Hoagland, K. D. (1997). Extracellular matrix assembly in diatoms (Bacillariophyceae) I. A model of adhesives based on chemical characterisation and localisation of polysaccharides from the marine diatom Achnanthes longipes and other diatoms. Plant Physiology 113, 1059–1069.

240

G.J.C. UNDERWOOD AND D.M. PATERSON

Wustman, B. A., Lind, J., Wetherbee, R. and Gretz, M. R. (1998). Extracellular matrix assembly in diatoms (Bacillariophyceae) III. Organization of fucoglucuronogalactans within the adhesive stalks of Achnanthes longipes. Plant Physiology 116, 1431–1441. Yallop, M. L., de Winder, B., Paterson, D. M. and Stal, L. J. (1994). Comparative structure, primary production and biogenic stabilisation of cohesive and non-cohesive sediments inhabited by microphytobenthos. Estuarine and Coastal and Shelf Science 39, 565–582. Yallop, M. L., Paterson, D. M. and Wellsbury, P. (2000). Interrelationships between rates of microbial production, exopolymer production, microbial biomass, and sediment stability in biofilms of intertidal sediments. Microbial Ecology. 39, 116–127. Yallop, M. L., Paterson, D. M. and Wellsbury, P. (2000). Inter-relationships between rates of microbial production, exopolymer production, microbial biomass and sediment stability in biofilms of intertidal sediments. Microbial Ecology 39, 116–127. Zhou, J., Mopper, K. and Passow, U. (1998). The role of surface-active carbohydrates in the formation of transparent exopolymer particles by bubble adsorption of seawater. Limnology and Oceanography 43, 1860–1871.

Fungal Pathogens of Insects: Cuticle Degrading Enzymes and Toxins

A. KEITH CHARNLEY

Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, UK

I. II. III. IV. V.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overview of the Infection Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Invasion of Host Cuticle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cuticle-degrading Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Proteolytic Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Chitinolytic Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Lipolytic and Esterolytic Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Regulation of Enzyme Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Production of Cuticle-degrading Enzymes by Other Entomopathogenic Fungi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Evidence for a Role for Cuticle-degrading Enzymes in Fungal Pathogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . H. Evolutionary Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. Role of Cuticle-degrading Enzymes in Virulence and Specificity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Toxins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Incidence of Insecticidal Toxins Amongst Entomopathogenic Fungi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. A Survey of Toxic Metabolites Produced by Entomopathogenic Fungi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Cyclicpeptide Toxins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Proteases and Other Enzymes as Toxins . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Multiple Toxins – Synergy or Specificity? . . . . . . . . . . . . . . . . . . . . . . . . .

Advances in Botanical Research Vol. 40 incorporating Advances in Plant Pathology ISBN 0-12-005940-1

242 244 245 246 251 251 253 257 259 260 263 265 270 274 276 276 277 280 286 296 298

Copyright  2003 Elsevier Ltd All rights of reproduction in any form reserved

242

A. K. CHARNLEY

VII. The Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300

ABSTRACT Pathogenic fungi are important natural regulators of insect populations. However, many attempts to harness their potential for pest control have met with comparatively minor commercial success. Studies on mechanisms of pathogenesis have yet to contribute to the development of more efficient mycoinsecticides, but new insights into the pathogenic process are laying the groundwork. Significant progress has been made in particular in understanding enzymes involved with the penetration of host cuticle and the role of insecticidal toxins. Insect cuticle comprises up to 70% protein and it is not surprising that extracellular fungal proteases appear to be particularly important in the penetration process. Subtilisins, chymotrypsins, trypsins and metalloproteases, usually with multiple isoforms of each, provide an impressive backed-up arsenal. Pathogenic fungi produce a wide variety of toxic metabolites, which vary from low molecular weight products of secondary metabolism to complex cyclic peptides and proteolytic enzymes. Comparatively few compounds have been found in diseased insects, in quantities sufficient to account for symptoms of mycosis. An exception, a family of cyclic peptides called the destruxins, are dealt with in detail. The potential for synergy between toxins is explored also.

I. INTRODUCTION Insect fungal pathogens hold a special place in the study of microbial pathogenesis. Agostino Bassi’s monograph in 1835 established for the first time that a microorganism (the fungus later to be called Beauveria bassiana in Bassi’s honour) could cause an infectious disease in an animal (the silkworm, Bombyx mori) (see Fig. 1). Prophetically Bassi suggested further that microbes could be used to control pest insects (Bassi, 1835). Natural epizootics of insect fungal diseases are comparatively common, and their impact on insect populations further demonstrates the potential of microbial pest control (Carruthers and Soper, 1987). This fact was recognised in the latter part of the 19th century and culminated in the seminal attempts by Metchnikoff and Paliokov to use the Deuteromycotina fungal pathogen Metarhizium anisopliae for insect control (Gillespie, 1988). Despite these and other early successes, synthetic chemical pesticides have been the mainstay of insect pest control for the last 50 years. However, the advent of insecticide resistance and concern over the environmental impact of agricultural inputs focus attention on biologically based forms of pest control. Mycoinsecticides have a toe-hold in a biological crop protection market (see e.g. Fig. 1c) dominated by the toxins from Bacillus thuringiensis (Bt) and crop plants transformed with Bt delta endotoxin genes (Charnley, 1997).

FUNGAL PATHOGENS OF INSECTS

243

Fig. 1. (a) Agostino Bassi*, (b) Frontise piece to Bassi’s 1835 paper in Italian*, (c) A formulation of Beauveria bassiana developed by Myotech and now produced by Emerald BioAgriculture Corporation for control of whitefly, aphids and thrips on vegetables and ornamentals (with permission from S. Jaronski), (d). Frontise piece to Bassi’s 1835 paper in English*, (e) Beauveria bassiana sporulating on cadavers of house flies (with permission from D. Steinkraus) *taken from Bassi (1835) (with permission from the American Phytopathological Society)

244

A. K. CHARNLEY

Studies on the mechanisms of fungal pathogenesis in insects have yet to contribute to the development of more efficient commercial mycoinsecticides, but new insights into the pathogenic process are laying the groundwork. Comparative genomics and post-genomic studies are providing exciting new insights into the evolution of virulence, host adaptation and gene function in pathogenic bacteria (e.g. Waterfield et al., 2002). A similar revolution in the study of fungal pathogens is on the horizon with the publication or pending publications of the genome sequences of the plant pathogen Magnaporthe grisea and the human pathogens Candida albicans and Cryptococcus neoformans (e.g. Tunlid and Talbot, 2002). A further 14 fungal pathogens are the subject of either full sequencing or expressed sequence tag projects including two insect pathogens, Metarhizium anisopliae and Conidiobolus coronatus (Tunlid and Talbot, 2002; Freimoser et al., 2003a,b). In the last 20 years the study of mechanisms of fungal pathogenesis in insects has progressed particularly with regard to the study of the enzymes involved with invasion of host cuticle and toxins. This contribution reviews current knowledge in these areas. Previous significant reviews include: (Roberts, 1980; Charnley and St. Leger, 1991; Khachatourians, 1991, 1996; St. Leger, 1993, 1995; Clarkson and Charnley, 1996; Vey et al., 2001; Anke and Sterner, 2002; Soledade et al., 2002).

II. TAXONOMY Relationships between fungi and insects may be mutualistic, through commensal to obligately pathogenic. The term entomogenous is often used to encompass all types of association between insects and fungi, with disease-causing fungi being referred to as entomopathogenic. A further distinction can be made between fungi which are aggressively pathogenic like Metarhizium anisopliae and opportunists like the wound pathogen Mucor haemalis (McCoy et al., 1988; Samson et al., 1988; Tanada and Kaya, 1993). Entomopathogenic fungi are found in most taxonomic groupings in the fungal kingdom, apart from the higher Basidiomycetes. The primitive water fungi, Mastigomycotina, have representatives with complex life cycles e.g. Coelomomyces psorophorae a mosquito pathogen with an obligate copepod secondary host. Among the Ascomycotina, Cordyceps spp. have fruiting structures or perithecia which can dwarf the cadavers of their insect victims. Entomophthorales are widespread members of the Zygomycotina. Mummified aphids stricken by fungi of this group are familiar features of cereal crops in temperate regions. The most widespread insect pathogenic

FUNGAL PATHOGENS OF INSECTS

245

fungi are found in the Hyphomycetous Deuteromycotina of the genera Metarhizium and Beauveria. The fungi described above are all destructively pathogenic. Many species of Entomophthorales are host-specific pathogens. Beauveria bassiana and Metarhizium anisopliae, facultative pathogens, have broad host ranges though considerable specificity occurs among isolates. Laboulbeniomycetes (Ascomycotina) on the other hand are biotrophic and specific to a single host or even a specific location on the host. They remain largely external gaining nutrition via a penetrant haustoria while apparently causing little harm. Most Trichomycetes (Zygomycotina) have a commensal existence in the guts of their Dipteran hosts, though some are pathogenic. Fast growing opportunists like Mucor haemalis may invade through wounds and it is important to note in the context of this review that many fungi are pathogenic if their spores are injected. This experimental strategy by-passes the exoskeleton and highlights the importance of the cuticle as a barrier to microbial pathogens.

III. OVERVIEW OF THE INFECTION PROCESS Unique among entomopathogenic microorganisms, fungi do not have to be ingested and can invade their hosts directly through the exoskeleton or cuticle. Therefore they can infect non-feeding stages such as eggs and pupae. The site of invasion is often between the mouthparts, at intersegmental folds or through spiracles, where locally high humidity promotes germination and the cuticle is non-sclerotised and more easily penetrated (Charnley, 1989; Hajek and St. Leger, 1994). M. anisopliae and B. bassiana have hydrophobic spores that appear to bind to insect cuticle by non-specific interactions though failure to adhere to particular insect species may help to define isolate host range. Zoospores of Lagenidium giganteum are host selective. Cuticle-degrading enzymes are present on the surface of conidia of M. anisopliae and therefore there is the potential for the fungus to modify the cuticle surface to aid attachment. Host and fungal lectins have been implicated also in the process of attachment. Germination in vitro of nutrient-dependent spores of M. anisopliae and B. bassiana is consequent upon a non-specific accessible source of carbon and nitrogen though in vivo isolate specificity may depend on response to qualitative and quantitative differences in available nutrients on host cuticle. More selective pathogens appear to have more specific requirements. Ability to withstand antifungal compounds in the cuticle such as short chain fatty acids is a prerequisite for successful invasion (see Boucias and Pendland, 1991). The importance of signal exchange between

246

A. K. CHARNLEY

host and pathogen is becoming increasingly clear and is first seen in the cues which cause the fungus to stop horizontal growth on the surface of the cuticle and initiate penetration. Differentiation of the germ tube to produce the holdfast structure, or appressorium, is most completely understood for M. anisopliae. Isolate 2575 (formerly ME1) requires low concentrations of a complex carbon and nitrogen source and a hard surface. Metarhizium isolates which have come from Homoptera, form appressoria in media (high concentrations of simple sugars) which are repressive for isolates from Coleoptera. This is probably an adaptation to parasitism as the cuticle of plant-sucking bugs (Homoptera) is contaminated with sugars from their copious liquid excreta (St. Leger et al., 1992b). Once the fungus breaks through the cuticle and underlying epidermis then it may grow profusely in the haemolymph, in which case death is probably the result of starvation or physiological/biochemical disruption brought about by the fungus. Alternatively insecticidal secondary metabolites may contribute to the demise of the insect and in this case extensive growth of the fungus may only occur on the cadaver of the host (Roberts, 1980; Gillespie and Claydon, 1989). For many fungi the reality is probably somewhere between these two extremes. Few studies have looked at the effect of fungal infection on host physiology/behaviour. This is unfortunate because sublethal or prelethal effects of mycosis may be just as useful as the death of the host from the point of view of crop protection. Detrimental effects of mycosis on food consumption, egg laying and flight behaviour have been recorded (Nnakumusana, 1985; Seyoum et al., 1995). The life cycle is completed when the fungus sporulates on the cadaver of the host. Under the right conditions, particularly high RH, the fungus will break out through the body wall of the insect producing aerial spores (Fig. 2). This may allow horizontal or vertical transmission of the disease within the insect population. Resting spores produced within the dead insect will enable the fungus to survive for long periods under adverse conditions (Samson et al., 1988).

IV. INVASION OF HOST CUTICLE The many light and electron microscope studies on cuticular penetration by entomopathogenic fungi have led to the conclusion that both physical force and enzymic degradation are involved (see Fig. 3 for an overview of the structure of insect cuticle). Ingression of cuticle around a fungal penetration peg suggests mechanical penetration, but this does not necessarily preclude a role for enzymes also (Fig. 4). Physical penetration is prominent in host invasion by some entomophthoralean pathogens where characteristic

FUNGAL PATHOGENS OF INSECTS

247

Fig. 2. (a) Stromata bearing perithecia of a Cordyceps spp. on a fly cadaver (with permission from H. Evans), (b) Erynia neoaphidis on Macrosiphum euphorbiae, note white halo of spores on the leaf around dead aphids (centre-left), (c) Metarhizium anisopliae sf. acridum sporulating on a cadaver of the desert locust, Schistocerca gregaria, (d) symptoms of mycosis caused by Metarhizium anisopliae sf. anisopliae 2575 on 5th instar caterpillars of the tobacco hornworm Manduca sexta, note the black melanic pigment, a host defensive response. (e). Akanthomyces aculeatus, a Deuteromycotina on a leaf mimicking moth from Papua New Guinea (with permission from C. Prior), (f). Verticillium lecanii sporulating on a dead aphid.

248

A. K. CHARNLEY

Fig. 3. Structure and composition of insect cuticle, (a) a transmission electron micrograph of cuticle from a newly moulted 5th instar caterpillar of the tobacco hornworm moth Manduca sexta, note the folds of the epicuticle and the procuticle laid down in lamellae (appears striated) (from Hassan and Charnley, 1987) with permission from Elsevier); (b) a diagram showing the main components of the cuticle and associated epidermis.

FUNGAL PATHOGENS OF INSECTS

249

Fig. 4. (a) Shows a scanning electron micrograph of Metarhizium anisopliae on the surface of the cuticle of a newly moulted 5th instar Manduca sexta caterpillar, note the folded epicuticle (‘‘pimples’’), appressoria form preferentially on the flat surface at the base of the seta; (b) a diagram depicting the invasion of cuticle by the fungus, note the surface features and the displacement of the lamellae around the hyphal bodies in the cuticle, suggesting a mechanical component; (c) cartoon (with permission from S. Fairhurst) emphasising the physical side of the penetration process; (d) transmission electron micrograph of Metarhizium anisopliae in cuticle of 5th instar Manduca sexta caterpillars. Note the lamellae (striations) of the procuticle, which are disturbed around the hyphal bodies, indicating mechanical ingress, and also the partial clearing of the lamellar structure around the fungus suggestive of partial enzymic hydrolysis, probably the action of proteases (from Hassan and Charnley (1987) with permission from Elsevier); (e) A cartoon (with permission from S. Fairhurst) to emphasise the role of the cuticle-degrading enzymes in the penetration process.

250

A. K. CHARNLEY

irradiate and tetraradiate fissures appear in the epicuticle (Brobyn and Wilding, 1983; Butt, 1987). Initial fungal growth within the outer layers of the procuticle often occurs laterally. These subepicuticular expansions can cause fractures, which favour penetration (Brey et al., 1986). Vertical penetrant hyphae may appear thin and constricted within the outer layers of the procuticle (exocuticle) presumably due to resistance to mechanical penetration engendered by the sclerotisation of the proteins, whereas the growing tip swells in the more pliant unsclerotised inner layers (endocuticle) (Robinson, 1966). Indentation or displacement of lamellae by lateral penetrant hyphae in both exo- and endocuticles is another clear sign of a mechanical component to the penetration process (see Charnley, 1984, and Fig. 4). Disappearance of the wax layer beneath appressoria of Metarhizium anisopliae on wireworm (Elaterid) cuticle indicates enzyme activity (Zacharuk, 1970b) as does the presence of circular holes around germ tubes of Beauveria bassiana at the point of entry into larvae of Heliothis zea (Pekrul and Grula, 1979). Early studies (reviewed by Charnley, 1984) showed changes in the histochemistry of insect cuticle around penetrant fungal hyphae consistent with enzymic hydrolysis. Brey et al. (1986) reported wide zones of complete histolysis in cuticle beneath penetrant hyphae of Conidiobolus obscurus in Acyrthosiphon pisum. In a transmission electron microscope study Askary et al. (1999) also found significant degradation of cuticle around hyphae of Verticillium lecanii in the aphid Macrosiphum euphorbiae. However, fine structural studies of many other insect–fungus interactions suggest that significant enzymolysis is not the norm. Absence of mechanical damage or displacement of lamellae suggests the action of enzymes, but without obvious zones of histolysis it must be assumed that fungal enzymes are usually limited to the vicinity of fungal structures in the initial stages of infection (Charnley, 1984, and Fig. 4). However, Goettel et al. (1989) and Hassan and Charnley (1989) noted clearing of the lamellar pattern but not complete histolysis around hyphae of Metarhizium anisopliae in the cuticle of Manduca sexta (Fig. 4). A consideration of the chemistry and physical properties of the cuticle a priori provides a way of assessing the likely impact of hydrolases. The tanned proteins of the cement layer may resist the proteases secreted profusely by the appressoria. The outer epicuticle is resistant to chemical and enzymic degradation (Hepburn, 1985), but in most insects it is probably fragile and thus susceptible to mechanical force (see St. Leger, 1991). In contrast, the inner epicuticle may yield to a combination of endoprotease and lipoprotein lipases. The procuticle constitutes the bulk of the cuticle and must provide a significant barrier to the invading fungus. This layer comprises chitin fibrils embedded in a protein matrix, together with lipids and quinones (Neville, 1984) (Fig. 3). The mechanical properties

FUNGAL PATHOGENS OF INSECTS

251

of different cuticles depend on the proportions of the two main constituents, the nature and extent of hydration of the proteins and the degree of sclerotisation also called tanning (because of homology with the curing of leather) viz. cross-linking of the proteins by quinones (Hillerton, 1984). In soft-bodied juvenile endopterygote insects like Lepidopteran caterpillars, exocuticle is essentially restricted to the head, while in the majority of adult insects and juvenile exopterygote insects, non-sclerotised cuticle is present only at joints and arthrodial membranes. Pliant cuticles have a higher chitin (46%) content than rigid cuticles (as little as 17%) (Hillerton, 1984). Clearly chitinases and proteases would be particularly useful for the fungus in facilitating passage through and gaining nutrient from the procuticle. Passage of the fungus across the procuticle may be more or less direct or involve a degree of lateral proliferation between cuticular lamellae before or during vertical penetration (Charnley, 1984). Thickness of the procuticle correlates with disease resistance in that young larvae (with thin cuticles) are more susceptible than old larvae (with thick cuticles). The degree of cuticle sclerotisation appears also to have a strong influence on penetrability. Although there are reports that sclerotised cuticle can be traversed, in the main it seems that insects with heavily sclerotised body segments are invaded via arthrodial membranes or spiracles (see Charnley, 1984; St. Leger, 1991). There is good reason for this behaviour. The resistance to compressive force of sclerotised cuticle (exocuticle see Fig. 3) and enzymic hydrolysis (from endogenous moulting fluid enzymes (Hepburn, 1985) and purified fungal enzymes (St. Leger et al., 1986d) suggests that it presents a stronger barrier than pliant, non-sclerotised endo and mesocuticle. While a considerable amount of work has been done on cuticle-degrading enzymes, hydrolysis of cuticle polymers could be facilitated by weak organic acids. Oxalic acid is produced both on the surface of some mycosed insects and in the haemolymph of infected insects and can hydrolyse cuticle proteins in vitro (Bidochka and Khachatourians, 1991). However, alkalisation of the cuticle during infection, which promotes protease production and activity (St. Leger et al., 1999), militates against a direct role in pathogenicity. Furthermore, hyperproductive oxalic acid mutants showed no difference in pathogenicity from wild-type (Bidochka and Khachatourians, 1993a).

V. CUTICLE-DEGRADING ENZYMES A. OVERVIEW

Cuticle-degrading enzymes have been studied in detail in Metarhizium anisopliae (see Table I for a compendium of Metarhizium anisopliae protease

252

TABLE I Proteases produced by Metarhizium anisopliae Enzyme Subtilisin

Trypsin

Mr (kDa)

Substrate

Specificity X ¼ amino acid branched at second carbon atom, phenylalanine preferred

Pr1A

8–9

10.2

30.2

Ala-Ala-X

Pr1B

8–9

31.5 18.5

Phe-Leu-X

Mp

7

9 8.3 7.3

Pr2A Pr2B

9 9 7

4.4 4.9 4.4–5.4

30 27 31–105

8

4 and 4.3

74

Aminopeptidase M(multiple isoforms) Post proline dipeptidyl peptidase IV (2 isoforms) Post alanine dipeptidyl peptidase (3 isozymes) Serine carboxypeptidase Zinc carboxypeptidase

pI

5.8 6.8 MeCPA

Based on St. Leger and Bidochka (1996).

7.5

9.8

Similar to Pr1A Broad, large hydrophobic residues preferred Val-Leu-Arg>Lys Val-Leu-Arg>>Lys Broad, alanine preferred Y-Pro-X

Inhibited by 1,10-phenylanthroline Inhibited by DFP

Lys-Ala-X 30 35

Broad, phenylalanine preferred

Specifically complements Pr1 Preference for branched aliphatic and aromatic COOH-terminal aa, complements Pr1

A. K. CHARNLEY

Chymotrypsin Metalloprotease

pH optimum

FUNGAL PATHOGENS OF INSECTS

253

activities). While it is useful to build up a body of knowledge on a single organism, generalisations cannot be made because of the huge diversity in biology and biochemistry between groups of entomopathogenic fungi. Initial studies showed that when M. anisopliae is grown on comminuted locust cuticle in liquid medium, a range of extracellular cuticle-degrading enzymes are produced corresponding to the major components of insect cuticles, viz. protein, chitin, and lipid (St. Leger et al., 1986c). Enzymes appeared sequentially. Esterase and proteolytic enzymes (endoprotease, aminopeptidase, and carboxypeptidase) were produced first (24 h) followed by Nacetylglucosaminidase (NAGase). Chitinase and lipase were produced 3–5 days later. The order of appearance of the enzymes is supported by the sequence of cuticle constituents solubilised into the culture medium, where a rapid increase in amino sugars followed early release of amino acids. Since chitinase is an inducible enzyme (Smith and Grula, 1983; St. Leger et al., 1986e) (see later), and cuticular chitin is masked by protein (St. Leger et al., 1986d), the late appearance of chitinase is presumably a result of induction as chitin eventually becomes available after degradation of encasing cuticle proteins. The late detection of lipase appears to be due to the fact that the enzyme is largely cell bound in young cultures (St. Leger, Charnley, and Cooper, unpublished). By testing purified enzymes against locust cuticle in vitro, St. Leger et al. (1986d) showed that pretreatment or combined treatment with endoprotease (Prl; see later) was necessary for high chitinase activity. Samsinakova et al. (1971) and Smith et al. (1981) also concluded that cuticular chitin is shielded by protein from studies using semipure commercial enzyme preparations against cuticles from Galleria mellonella larvae and Heliothis zea larvae, respectively. When locust exuviae (non-digested remains of old cuticle shed at ecdysis; exocuticle only) were used as substrate for purified pathogen enzymes instead of cuticle from larval sclerites (exo- and endocuticle), comparatively little hydrolysis occurred (St. Leger et al., 1986d). Similarly, while unsclerotised cuticle from fledgling locusts was rapidly degraded by fungal proteases, the cross-linking of cuticle proteins with glutaraldehyde (as a model for sclerotisation) substantially reduced their susceptibility to proteolytic attack (St. Leger, Charnley, and Cooper, unpublished). B. PROTEOLYTIC ENZYMES

1. Endoproteases The most prominent endoprotease in cuticle cultures of M. anisopliae 2575, termed Pr1, is an alkaline, serine enzyme with an essential histidine residue in the active site (St. Leger et al., 1987a). Prl possesses a broad primary

254

A. K. CHARNLEY

specificity for amino acids with a hydrophobic side group at the second carbon atom (e.g., phenylalanine, methionine, and alanine) but also possesses a secondary specificity for extended hydrophobic peptide chains with the active site recognising at least five subsite residues. This comparative non-specificity accounts for it being a good general protease with activity against a range of proteins (casein, elastin, bovine serum albumin, collagen) and insect cuticle (St. Leger et al., 1987a). Essential binding of Prl to negatively charged cuticle groups is dictated by its basic nature (St. Leger et al., 1986b). Only following adsorption does the active site come into contact with susceptible peptide bonds; solubilised peptides are further degraded until a chain length of about 5 is obtained (St. Leger et al., 1986d). The Pr1 cDNA has been cloned, revealing that Pr1 is synthesised as a large precursor containing a signal peptide, a propeptide and a mature 29 kDa protein. Originally Pr1 was termed a ‘chymoelastase’ in line with its substrate specificity, however, the predicted amino acid sequence places Pr1 in the subtilisin subclass of serine proteases, which are the predominant class of microbial extracellular proteases. St. Leger et al. (1994b) resolved the Pr1 activity of M. anisopliae 2575 produced on cockroach cuticle into four isoforms. Three of the purified isoforms (pI 10.2, 9.8, 9.3) had similar primary specificities viz. phenylalanine at P1 was more reactive than leucine. With regard to secondary subsite specificity, pI 10.2 isoform differed from the others in preferring alanine over bulky hydrophobic groups at S1 or S2. They appeared to be equally effective in degrading proteins from insect cuticle. Multiple isoforms of metalloprotease were also found. They were inhibited by 1,10phenanthroline and phosphoramidon, a specific inhibitor of thermolysinlike metalloproteases. The metalloproteases were not produced to the same degree as the subtilisins although they had a similar amino acid specificity. A further Metarhizium subtilisin gene Pr1B has been cloned and sequenced (Joshi et al., 1997). Application of contour-clamped homogeneous electric field electrophoresis (CHEF) separated the M. anisopliae genome into seven chromosomes. Hybridisation of Pr1B occurred strongly to one chromosome and weakly to two others, which, along side evidence from Southern analysis, suggested single copies of both the Pr1B gene and the 1st Pr1 gene cloned (Pr1A) and the existence of a third subtilisin Pr1C. It is not yet clear whether the four Pr1 isoforms so far purified are products of the three ORFs. Bidochka and Melzer (2000) showed that Pr1 genes from M. anisopliae make up a multigene family and suggested that they may be derived from a common ancestral subtilisin. The RFLP pattern for Pr1A is more complex than that for Pr1B or Pr1C. Thus Pr1A could be the original gene which duplicated. However, the situation is further complicated by

FUNGAL PATHOGENS OF INSECTS

255

the revelation from an EST project that M. anisopliae genome has at least 11 subtilisin genes (Pr1 A-K) (Freimoser et al., 2003a). Bagga et al. (2003) have divided the genes into three subfamilies. Further phylogenetic analysis comparing subtilisins from Metarhizium with those from fungi and other organisms confirms the authenticity of these subdivisions and suggests that gene duplication occurred before speciation of major fungal lineages. Recently Screen and St. Leger (2000) identified a chymotrypsin (CHY1) by express sequence tag analysis of cuticle cultures. CHY1 belongs to the S2 group of chymotrypsins. It is most closely related to S. griseus protease C. Both have only 15% identity to mammalian S1 chymotrypsin and have Ala190 and Thr213 in the S1 specificity pocket. These residues define a specificity for hydrophobic amino acids, so that like Pr1, CHY1 is most effective against Suc (Ala)2 Pro Phe. Substituting Leu or Meth for the Phe reduced catalytic efficiency, replacing Phe with Ala or Val left only trace activity. The latter two substitutions produce elastase substrates which are effectively hydrolysed by Pr1a. Interestingly plant and insect Ascomycotina and actinomycete bacteria are the only known microbial sources of S1 trypsins. So M. anisopliae may have acquired CHY1 by lateral gene transfer from an actinomycete. Pr2, a third class of serine protease produced by M. anisopliae (2575) occurs as four isozymes (ca. pl 4–4.9) with little activity against insect cuticle, insoluble insect cuticular proteins or elastin but high activity against casein and basic solubilised cuticular proteins. It has a primary specificity for arginine and lysine residues comparable to that of bovine trypsin and is sensitive also to trypsin inhibitors (e.g. leupeptin, tosyl-lysine-chloroketone, soybean trypsin inhibitor). Maximum activity was against Val Leu Arg AFC. The catalytic efficiency of Pr2 can be influenced by subsite residues at a distance from the cleaved site (St. Leger et al., 1987a,b, 1996a). N termini of the two main isoforms (pI 4.4 and 4.9) resemble those of other trypsins. Cole et al. (1993) purified a pI 4.6 enzyme with trypsin-like specificity, which they termed a cysteine protease on the basis of its susceptibility to inhibition by sulphydryl reagents. St. Leger et al. (1996a) have suggested, however, that the studied by Cole et al. (1993) is in fact equivalent to their pI 4.9 isoform (the same isolate was used in both studies) that may have free sulphydryl groups close to the active site. Why M. anisopliae should have such a complex protease arsenal with subtilisin, chymotrypsin, metalloproteases and trypsins, in most cases with multiple isoforms, is not clear. Comparative analysis of the 10 Pr1 genes that code for exocellular enzymes, expressed on cuticle by M. anisopliae 2575, showed differences in the S4, S3 and S2 subsites consistent with variation in catalytic efficiency and secondary specificity noted in biochemical analyses

256

A. K. CHARNLEY

(Bagga et al., 2003). Other differences in sequence will affect environmental stability and adsorption as well as substrate activity between isoforms. Pr1 D-J all have small non-polar or hydrophilic residues at 189 (Gly, Thr, His, Asp) which will preclude an interaction with certain types of proteases inhibitor e.g. chymotrypsin inhibitor 2. Though Pr1A is the predominant form produced during growth on cuticle, ESTs for Pr1A are 10 times more abundant than Pr1J, the next most abundant transcripts, the minor isoforms could play a key role in pathogenesis. Multiple Pr1 isoforms with subtle differences in specificity may act synergistically in the degradation of host cuticular protein, enable the pathogen to operate in changing environmental conditions and counter host defensive enzyme inhibitors. A spectrum of proteases may help a facultative pathogen exploit a wider range of substrates. They may be expressed at different stages of infection and have slightly different functions cf. proteases produced by Leishmania (Mottram et al., 1997). The mycopathogenic fungus Trichoderma harzianum produces many chitinases and expression is dependent on the host infected (Haran et al., 1996). Alternatively duplicate enzyme systems may help a facultative pathogen survive saprophytically in different ecological niches. The use of inhibitors with Metarhizium culture filtrates showed that Pr1-like activity accounts for 92% of the insoluble cuticle-degrading ability present. The greater efficiency of Pr1 is a reflection of the higher alanine and low basic amino acid content of many insect cuticular proteins which would favour digestion by a subtilisin (St. Leger et al., 1996a). However, the low activity of Pr2 enzymes against cuticle does not preclude a role in fungal penetration. Trypsins have activity against arginine and lysine residues, which are found in particular on the surface of globular proteins. Thus Pr2 may complement Pr1 by opening up proteins for further hydrolysis and providing peptides for nutrition. Pr2 could additionally be involved in cellular control mechanisms, catalysing specific proteolytic inactivation and activation processes (St. Leger et al., 1987c). In this context it is interesting that inhibition of M. anisopliae (2575) Pr2 with tosyl-lysine-chloroketone selectivity repressed formation by germlings of infection structures, implying a role for Pr2 in control of differentiation (St. Leger, unpublished). A gene encoding one isoform of the trypsin-like enzyme Pr2 from M. anisopliae has been cloned; the predicted amino acid sequence is very similar to the trypsin of Fusarium oxysporum and clearly distinguishes Pr2 from the subtilisin subclass of serine proteases (Smithson et al., 1995). 2. Exoproteases The action of Pr1 on cuticle releases peptides with a mean residue length of 5 amino acids. Further degradation by peptidases will be required to

FUNGAL PATHOGENS OF INSECTS

257

provide fungal nutrition. Two classes of aminopeptidase were isolated from cuticle-grown cultures of M. anisopliae and classified as an aminopeptidase M of broad specificity and a post-proline dipeptidyl aminopeptidase IV (St. Leger et al., 1995a). The aminopeptidase M (pH optimum 7–8, 33 kDa) exists as six isozymes (pl 5–6) with optimal activity for alanine residues and side activities versus other apolar and hydrophobic amino acids. The enzyme is sensitive to typical inhibitors of metalloenzymes (e.g., EDTA, 1,10-phenanthroline). The dipeptidyl aminopeptidase (pH optimum 8, 74 kDa) exists as two isozymes (pl ca. 4.6) and removes X-prolyl groups from polypeptides (X ¼ an apolar amino acid). The enzyme is inhibited by DFP (but not PMSF), indicating that it is a serine hydrolase. Without this enzyme degradation of proteinase-derived cuticle peptides by aminopeptidase would terminate at proline residues because of the low specificity of Pr1 for this residue. Neither peptidase alone hydrolysed intact insect cuticle. However, when combined with Pr1 they effected enhanced release of amino acids (St. Leger et al., 1993a). St. Leger et al. (1994a) purified a serine carboxypeptidase produced during growth on cockroach cuticle. The enzyme (Mr 30 kDa, pI 9.97, pH optimum 6.8) had a broad primary specificity towards amino acids with hydrophobic side groups in a series of N-terminally blocked dipeptides, with phenylalanine being most rapidly hydrolyzed. It complements Pr1, both showing a preference for cleavage of bonds C-terminal to aromatic residues. Both are produced during C and N deprivation and operate synergistically to provide nutritional amino acids. Joshi and St. Leger (1999) isolated a cDNA clone for a zinc carboxypeptidase MeCPA by reverse transcriptase differential display PCR from a cockroach cuticle culture of M. anisopliae. It is of interest because it is a metallocarboxypeptidase, other fungal CPAs being serine enzymes including the one identified by St. Leger et al. (1994a) from Metarhizium (see above). MeCPA has a preference for branched aliphatic and aromatic COOHterminal amino acids. This requirement is met by the peptides released by the action of Pr1 which cleaves COOH-terminal to aromatic amino acids like Phe. Thus, Pr1 and MeCPA complement each other. These enzymes are also regulated similarly viz produced under C and N derepressed conditions, but optimally in the presence of insect cuticle (see later).

C. CHITINOLYTIC ENZYMES

St. Leger et al. (1991a) purified endochitinase from culture filtrates of M. anisopliae grown on 1% ground chitin. The purified enzyme failed to

258

A. K. CHARNLEY

hydrolyze arylglycosides or chitobiose (N-acetylglucosamine dimer), showed only trace activity against chitotriose (trimer), but rapidly degraded chitotetraose (tetramer). Colloidal chitosan (deacetylated form of chitin) and crystalline chitin were less amenable to degradation than colloidal chitin, but activity against them was still substantial. The chitinase had many similarities to those produced by other microorganisms (e.g., Stirling et al., 1979). These properties include a pH optimum of 5.3, a molecular weight of 33 kDa, and the lack of any requirement for a cofactor. Hydrolysis of crystalline chitin produced only one low-molecular-weight reaction product within 24 h, viz. N-acetylglucosamine (NAG). The absence of intermediary oligomers among chitin breakdown products probably means that NAG is released directly from insoluble chitin. Either the chitinase has an exo-acting component or alternatively the reaction proceeds by a singlechain processive mechanism as described for some other endo-acting polysaccharidases (Cooper et al., 1978). This involves the random cleaving of bonds followed by release of monomers or dimers from exposed ends so that a single macromolecule is completely degraded before a new one is attacked. Such a mechanism, especially if it involved simultaneous digestion of several parallel chains, could result in the rapid degradation of chitin fibrils and in addition produce monomers for nutrition and induction for further enzyme synthesis. An IEF study has shown large numbers of chitinase isoforms (10) produced in culture filtrates from M. anisopliae (St. Leger et al., 1993b). The big difference in molecular weights of the enzymes suggests that they are products of different genes rather than post-translational modifications, e.g. glycosylation (St. Leger et al., 1996b). The isoforms produced on cockroach cuticle by M. anisopliae sf. anisopliae 2575 and M. anisopliae sf. acridum 324 may be broadly classified by pI as basic or acidic. The latter are dominant. Two isoforms 43.5 and 45 kDa, pI 4.8 have been purified. In this case similarities between N-termini suggest that they may not be separate gene productions but rather result from post-translational modifications, particularly glycosylation. Pinto et al. (1997) found a 30-kDa endochitinase in a different isolate of M. anisopliae with similar Mr, pH and temperature to one reported previously by St. Leger et al. (1991a). However, a 60-kDa endochitinase by Kang et al. (1998, 1999) differed from the St. Leger et al. (1991a, 1996b) isoforms not only in Mr but also protein N-terminus and ORF nucleotide sequence. It could be one of the eight uncharacterised chitinase isoforms identified by St. Leger et al. (1993b). In both the latter cases the enzymes were similar to the St. Leger et al. isoforms in having both endo and exoactivity. Bogo et al. (1998) isolated a cDNA for a chitinase gene

FUNGAL PATHOGENS OF INSECTS

259

encoding an endochitinase of 58 kDa. The ORF encoded a 423 amino acid protein with a predicted final product of 43 kDa. This compared well with the 45-kDa chitinase purified biochemically by St. Leger et al. (1996b). N-Acetylglucosaminidase activity has been partially purified from culture filtrates of M. anisopliae grown on 1% ground chitin. The enzyme had substantial activity against p-nitrophenol acetylglucosamine, as well as chitobiose, chitotriose, and chitotetraose, the major product in each case being NAG, showing that the enzyme is a true NAGase rather than a chitobiase (St. Leger et al., 1991a). The enzyme had little activity against colloidal or crystalline chitins. Its size, 110-120 kDa, is within the range for similar enzymes from other sources (e.g., Reyes and Byrde, 1973).

D. LIPOLYTIC AND ESTEROLYTIC ENZYMES

There has been little work done on the enzymes that are active against lipids and esters. This is perhaps not an important omission in the dissection of the role of enzymes in cuticle penetration, because substrates for these enzymes are not important components of epi- or procuticle. Hydrocarbons that are major constituents of the wax layer are used by germinating fungi (Lecuona et al., 1991). However, exocellular enzymes are not involved. In general, esterases may be differentiated from lipases because shortchain fatty acids (C2–C4) are preferentially hydrolysed by the former and long-chain esters > C8) by the latter (Shnitka, 1974). Esterase activity produced by M. anisopliae in young cultures (3 days) was greatest against short- and intermediate-length p-nitrophenol esters with only trace activity occurring above C10 (St. Leger, Charnley, and Cooper, unpublished), suggesting that lipase is not produced extracellularly by young mycelia. Activity against C14 rose in older cultures (7–14 days). Late arrival of extracellular lipase in vitro was confirmed using the ‘true’ lipase substrate olive oil (St. Leger et al., 1986a). The major esterase peak eluted from a Sephadex G100 column with a profile very similar to that obtained for endoprotease Pr1. As this protease also degrades p-nitrophenol esters, it is probably a major contributor to total esterase activity. Flat-bed IEF, however, revealed 25 distinct esterases (isozymes) from culture filtrates of M. anisopliae grown on locust cuticle. On the basis of their reactions with naphthyl esters, the isozymes appeared to have different substrate specificities. However, all the bands were inhibited by PMSF, indicating that they are serine carboxyesterases (esterase B) and not arylesterases (esterase A) that are inhibited by N-ethyl-maleimide. Esterases catalyse many enzymatic reactions though preferentially hydrolysing aliphatic or

260

A. K. CHARNLEY

aromatic esters and amides (Shnitka, 1974; Heymann, 1980). The considerable heterogeneity of esterases could account for their collective lack of specificity. Multiple enzyme strategies are believed to play an important role in the ability of an organism to adapt to different environments (Moon, 1975; Somero, 1975), presumably including that provided by an insect host.

E. REGULATION OF ENZYME PRODUCTION

Production of the right enzymes, in sufficient quantity, in an appropriate sequence, at the right place and time must be critical for successful parasitism. Pr1 is controlled by multiple regulatory circuits. The Pr1a promoter region has been sequenced (up to 1 kb upstream of the ATG). Binding sites similar to those of the ‘nitrogen regulator’ (AREA) and ‘carbon regulator’ (CREA) of A. nidulans have been identified. In addition genes coding for AREA and CREA-like regulatory proteins have been cloned and sequenced from M. anisopliae and have been shown to function when transformed into A. nidulans mutants (Screen et al., 1997, 1998). Pr1 is the major protein product by appressoria on artificial surfaces or on host cuticle and thus there is an element of developmental regulation. Pr1 mRNA was not present in spores but Pr1 is the primary translation product in germlings forming appressoria. Radiolabelled Pr1 was secreted into the medium just 6.5–7.2 min after the addition of a pulse of [35S] methionine. Rapid processing of the original large primary translation product was not affected by the Pr2 inhibitor TLCK, which is not consistent with a role in protein processing for this enzyme suggested earlier (St. Leger et al., 1989a). In culture both Pr1 and Pr2 occurs with carbon and nitrogen starvation (St. Leger et al., 1988c). In minimal medium (salts and no nutrients), the soluble protein BSA repressed production of Pr1, while it allowed enhanced synthesis of Pr2. Generally Pr2 is less tightly regulated than Pr1. Both during production in culture or during appressorium formation Pr1 is repressed by readily utilised metabolites (e.g., glucose or alanine) (St. Leger et al., 1988c, 1989a). The second messenger system that mediates the effects of starvation and regulates transcription of the Pr1 gene has not yet been established. Evidence militates against the involvement of cAMP (St. Leger et al., 1988c, 1989b). A role for calcium is suggested by the fact that secretion but not synthesis of Pr1 in M. anisopliae is inhibited by agents which depress calcium-dependent protein phosphorylation (St. Leger et al., 1989b). However, antagonists of

FUNGAL PATHOGENS OF INSECTS

261

calmodulin (a sensor and effector of many Ca2þ-dependent messages) that are potent inhibitors of protein synthesis and phosphorylation during germination do not affect Pr1 synthesis nor secretion by mycelia (St. Leger et al., 1989b). Extracellular levels of both Pr1 and Pr2 were enhanced in cultures supplied with insect cuticle or other insoluble polymers (e.g., cellulose) that were insufficient to produce catabolite repression (CR). Under these conditions regulation of Pr1 gene expression was again exerted at the level of transcription (St. Leger et al., 1995b). By comparison with the situation on rich media up to 32 new proteins appeared on cuticle or chitin (some 19 also were less abundant); prominent among them was Pr1 though there were trypsin and chymotrypsin homologues. Pr1 is specifically induced by a peptide component of insect cuticle, but not by other soluble or insoluble proteinaceous substrates (Paterson et al., 1994a,b). The feeding of elastin or collagen to derepressed established mycelia (starved for carbon and nitrogen) did not enhance Pr1 production significantly and, as found previously, soluble proteins were repressive. The carbohydrate polymers cellulose and xylan gave derepressed basal levels only. Peptides in the range 150–2000 Da, released from insect cuticle by either Pr1 or Pr2 induced Pr1 production to a similar level to that obtained with untreated cuticle. Deproteinised and lipid-extracted cuticles supported little Pr1 production. Pr2, by comparison, is induced non-specifically by protein (Paterson et al., 1993). Thus, rapid protease synthesis is only possible in host tissues where the concentration of readily metabolisable compounds is low. This is the case with insect cuticles as the components are largely insoluble until released by cuticle-degrading enzymes (St. Leger et al., 1986d). However, repression could operate if ever the release from cuticle of degradation products exceeded fungal requirements. Thus, the pathogenic process involving infection-related morphogenesis and enzyme production occurs only when it is necessary for the pathogen to establish a nutritional relationship with the host. Less is known about the regulation of exopeptidases, but as for the endoproteases production occurs under C and N derepression, optimal production on insoluble polymers (particularly protein and host cuticle) and repression by low molecular weight compounds e.g. the zinc carboxypeptidase MeCPA (Joshi and St. Leger, 1999). Chitinase synthesis is regulated in M. anisopliae (St. Leger et al., 1986e) by products of chitin degradation, through an inducer–repressor mechanism. High chitinase activity was found only in cultures supplied with chitin, but not with other polymers such as pectin, xylan, and

262

A. K. CHARNLEY

cellulose. Slow feeding of cultures with sugars or alanine in a carbondeficient medium demonstrated that the most effective inducer of chitinase was N-acetylglucosamine (NAG). Glucosamine also allowed production, possibly an adaptation to the fact that chitin from natural sources (including insect cuticle) appears to be partially deacetylated (Hackman and Goldberg, 1974). It is unlikely that chitobiose could function as a major inducer of chitinase in M. anisopliae as e.g. cellobiose does for cellulose (Cooper and Wood, 1975) because NAGase would degrade it to NAG and the major product of chitinase activity is NAG. Interestingly, NAGase was produced constitutively and was little affected by catabolite repression. Co-ordinated expression of all endochitinase isoforms does not occur since a mutant in which all other isoforms were produced at low levels secreted large amounts of the major 48-kDa form (St. Leger et al., 1996b). St. Leger et al. (1998) showed that proteases and chitinases were only synthesised at the pH at which they function effectively, irrespective of whether the medium contained an inductive cuticle substrate, though for most enzymes cuticle increased activity 3-fold over that at the optimum pH. Aminopeptidases were produced at pH 7 (optimum enzyme activity ¼ pH 7), metalloproteases produced at pH 6–8 (optimum ¼ pH 7), trypsins and subtilisins produced at pH 8 (optimum ¼ pH 8). Northern analysis of RNA corresponding to 7 cDNA sequences encoding proteases and chitinases confirmed that ambient pH played a major role in regulating gene expression of secreted proteins. During fungal infection the pH of the host insect cuticle rose from 6.3–7.7. Alkalinisation of the cuticle, possibly by the fungus itself, is a physiological signal that triggers the production of pathogenicity factors. Interestingly chitinase was produced at pH 5 and 8 (optimum ¼ pH 5). pH 8 is the optimum for Pr1 which is needed for chitinase activity. Consistent with the changes observed in cuticle pH during mycosis St. Leger et al. (1999) observed that M. anisopliae 2575 neutralises environmental pH during growth on solid media containing yeast extract by the production of ammonia or organic acids (oxalic, succinate, acetic). Acid non-producing mutants had reduced ability to grow at pH 8 suggesting that acid production is linked to the ability to grow at higher pH. Hyperproductive acid mutants had reduced protease activity because of the acidification of the medium. Ammonia is produced by deaminases that are regulated by induction and catabolite repression. While this work establishes the possibility that ammonia may be a virulence factor for pathogenic fungi it also complicates the interpretation of the role of organic acid production by these fungi.

FUNGAL PATHOGENS OF INSECTS

263

F. PRODUCTION OF CUTICLE-DEGRADING ENZYMES BY OTHER ENTOMOPATHOGENIC FUNGI

Pr1-like enzymes have been resolved in culture filtrates from B. bassiana, V. lecanii, N. rileyi, and A. aleyrodis (St. Leger et al., 1987b; El-Sayed et al., 1993; Chrzanowska et al., 2001). Trypsin, Pr2-like, endoproteases are also produced by these fungi, though they are specific for a Phe Val Arg group and demonstrate less sensitivity to trypsin inhibitors (St. Leger et al., 1987b; Gupta et al., 1993b, 1994). Endo- and exo-acting proteolytic enzymes have been identified in cultures of other entomopathogenic fungi viz. the collagenases produced by Entomophthora coronata (Zygomycete) (Hurion et al., 1977) and Lagenidium giganteum (Oomycete) (Hurion et al., 1979), and chymotrypsins produced by Erynia spp. (Entomophthorales; Zygomycete) (Samuels et al., 1990). Pr1-like genes have been cloned and sequenced from two isolates of B. bassiana also (Joshi et al., 1995; Kim et al., 1999). Both genes had strong homology with each other and with Pr1a from M. anisopliae and protease K from T. album. Genes that are significantly similar to Pr1 are present in the entomopathogens A. flavus and V. lecanii (St. Leger et al., 1992a; Leal et al., 1997). Pr1 from B. bassiana is repressible by glucose and N-acetylglucosamine and enhanced by protein, though unlike the enzyme from M. anisopliae no specific induction by cuticle protein has been established (Bidochka and Khachatourians, 1987, 1988a,b). Catabolite repression probably accounts for the low expression of Pr1 from M. anisopliae in host haemolymph during mycosis. However, this level of control may not occur in all species or all isolates. Substantial amounts of extracellular proteases were present in haemolymph of Bombyx mori infected with B. bassiana, estimated by ELISA using a polyclonal antibody. Enzyme activity in the haemolymph was low possibly due to protease inhibitors (Shimizu et al., 1993a). Bye and Charnley (unpublished) showed significant differences in regulation of Pr1-like enzymes between isoforms of the same isolate and between isolates of V. lecanii. Of particular note is an isoform from KV71 with a pI of 8.6 that is induced specifically by NAG. For isolate KV42 all isoforms were not repressed by low molecular weight C or N individually though they were in combination. Consistent with the reduced effect of catabolite repression in vitro substantial subtilisin activity was found in aphids infected with this isolate during the early stages of mycosis (Fig. 5). Pr1 from five isolates of V. lecanii was produced to equal extents on locust and (host) aphid cuticles, suggesting no host-specific induction.

264

A. K. CHARNLEY

Fig. 5. Sections through whole adult peach potato aphids, Myzus persicae, infected with Verticillium lecanii (2d post inoculation). They have been immunostained with the Vectastain ABC system using rabbit antibodies raised against Pr1 protease from either isolate KV71 or KV 42. (a) Control uninfected aphid counter stained with toluidine blue and acid fuchsin, (b) Aphid infected with isolate KV71, gold staining (g) shows location of antibody bound to protease, only a small area towards the bottom of the section, there is little protease in these insects outside of the cuticle because of catabolite repression, (c) Aphid infected with isolate KV42, gold staining (g) shows location of antibody bound to protease. It is all around the periphery of the insect in the cuticle and within subepidermal fatbody. This slide has been counterstained to show the other tissues, (d) Similar to (c) but without the counter stain so the immunostain only is visible.

FUNGAL PATHOGENS OF INSECTS

265

Endo and exochitinases have been found consistently in culture supernatants of other entomopathogenic fungi, including Nomuraea rileyi and B. bassiana (Coudron et al., 1984; Elsayed et al., 1989; Bidochka et al., 1993; St. Leger et al., 1996b). Exochitinase from B. bassiana was induced by low concentrations but not repressed by high concentrations of the chitin monomer N-acetylglucosamine (Bidochka and Khachatourians, 1993b). Repression was exerted selectively by other low molecular weight compounds. Some amino acids were repressive while others were not. It appeared that the carbon skeleton of amino acids which repressed exochitinase synthesis were catabolised later in the TCA cycle i.e. -ketoglutarate, succinyl-CoA and fumarate. Thus regulation could be where the amino acids enter the TCA cycle (Khachatourians, 1991). In one of the few published papers on a lipase from an entomopathogenic fungus Hegedus and Khachatourians (1988) showed that lipase from B. bassiana was induced by olive oil but only during early stationary phase (5d post-germination) and they suggested that in vivo the enzyme may be important in colonisation of the haemolymph rather than the cuticle. This makes intuitive sense as most insects have high concentrations of diacylglycerol in the haemolymph.

G. EVIDENCE FOR A ROLE FOR CUTICLE-DEGRADING ENZYMES IN FUNGAL PATHOGENESIS

St. Leger et al. (1987c) extracted Pr1- and Pr2-like enzymes and an aminopeptidase from wings of C. vomitoria and abdominal cuticle of fifthinstar larvae of M. sexta, about 16 h after inoculation with M. anisopliae. Endoprotease activity was separated into two compounds which closely resembled Pr1 and Pr2 in pI, substrate specificity, and inhibitor spectrum. Purified extracts of infected blowfly wings tested by Ouchterlony gel diffusion against specific antiserum to Pr1 gave a single precipitin line identical to that given by the pure enzyme, confirming the presence of Prl during infection. The translucent wings of C. vomitoria have also been used to locate histochemically proteolytic enzymes during penetration (St. Leger et al., 1987c). Substrates and inhibitors specific for Prl and Pr2 established the production of these enzymes on appressoria, which developed 10–24 h after inoculation. Aminopeptidase differed from endoprotease in that it was not present on immature appressoria, and the activity extended into the mucilage surrounding mature appressoria and appressorial plates. Pulse labelling with [35S] methionine in conjunction with Western blot analysis employing Pr1 antibody showed that Pr1 forms the

266

A. K. CHARNLEY

majority of the protein being synthesised by appressoria in vitro and in vivo (St. Leger et al., 1989a). An immunogold technique has been used to locate Pr1 in cuticle during penetration of larvae of M. sexta (Goettel et al., 1989). Protease was present on and close to infection structures and penetrant hyphae. The label was found more diffusely in the cuticle during later stages of pathogenesis. The two main Pr2 isoforms (pI 4.4 and 4.9) from M. anisopliae were ultrastructurally located to cell walls of the appressoria and other prepenetrant structures of the fungus in contact with the cuticle (St. Leger et al., 1996a). Within the cuticle, label was present on penetration peg and hyphal bodies. In contrast to Pr1 the enzymes were restricted to the vicinity of the fungus, even during later stages of infection. A matrix-like or fibrous material was present around the elements of the fungus which could be mucilage. It extended beyond the deposition of the enzyme. The zinc carboxypeptidase MeCPA has a specificity that complements Pr1 and the two enzymes could work co-operatively in releasing amino acids for nutrition from cuticular proteins. Consistent with this Joshi and St. Leger (1999) used immunogold to locate MeCPA to infection structures during cuticle penetration. Pr1 appears to be a pathogenicity determinant by virtue of its considerable ability to degrade cuticle (St. Leger et al., 1986d) and its production at high levels by the pathogen in situ during infection (St. Leger et al., 1987c). Simultaneous application of turkey egg white inhibitor (TEI) and conidia significantly delayed mortality of Manduca larvae compared with larvae inoculated with conidia, supporting the importance of Pr1 in penetration (St. Leger et al., 1988b). The inhibitor also reduced melanisation of cuticle (a host response to infection) and invasion of the haemolymph as well as maintaining the host’s growth rate. TEI or antibodies raised against Pr1 delayed penetration of the cuticle but did not affect spore viability or prevent growth and formation of appressoria on the cuticle surface. This suggests that inhibition of Pr1 reduced infection by limiting fungal penetration of the insect cuticle. In vitro studies using TEI showed that accumulation of protein degradation products from the cuticle, including ammonia, was dependent on active Pr1. This confirms its major part in solubilising cuticle proteins and making them available for nutrition. It is interesting to note that Pr1 is resistant to serpins (a key class of protease inhibitor present in host cuticle and haemolymph) (St. Leger and Bidochka, 1996) and melanin produced in situ (St. Leger et al., 1988a). Attributes which are consistent with an enzyme adapted for a role in insect parasitism. Gillespie et al. (1998) failed to find a correlation between subtilisin production and pathogenicity for locusts among isolates of M. anisopliae.

FUNGAL PATHOGENS OF INSECTS

267

A complication in this and other studies of this kind is that the different genetic backgrounds of the isolates tested may make them polymorphic for other characteristics that play a part in pathogenesis. Bidochka and Khachatourians (1990) found that a UV-induced protease-deficient mutant of B. bassiana displayed reduced virulence, but pleiotropic effects cannot be ruled out. Analysis of a Pr1A null mutant, produced by transformationmediated gene disruption, did not provide unambiguous demonstration of a key role for Pr1 in the disease process. In those cases where there were no additional heterologous integrations mutants still retained near-normal pathogenicity under certain bioassay conditions (St. Leger, 1995). The onset of cuticle invasion coincided with secretion of high levels of Pr1b and a metalloprotease suggesting that these other cuticle-degrading proteases, that have broadly similar specificities, might partially substitute for Pr1a (St. Leger et al., 1994b). This apparent redundancy in proteases helps to explain the previous observation that simultaneous application of a Pr1 inhibitor and Metarhizium conidia to the insect surface did not prevent infection, but did cause a significant delay in mortality (St. Leger et al., 1988b). Targeted gene disruption has been employed extensively in investigations of the part played by cell-wall degrading enzymes in fungal pathogenesis of plants. Here too null mutants e.g. of protease, cutinase and xylanase genes have failed often to have significant effects on pathogenicity (see review by Hamer and Holden, 1997), a pattern repeated with human and animal protease genes, even when knock-outs of several genes are achieved in the same isolate. These results are disappointing given the wealth of biochemical evidence that in many cases has implicated the corresponding proteins in pathogenesis. A variety of explanations have been offered: the bioassay was not sensitive enough to reveal the effect of the mutation, there was biochemical compensation by products of related genes (isoforms), there was functional redundancy (other enzymes can carry out the same task as appears to be the case with Metarhizium) and new group of similar enzymes produced in vivo. An alternative strategy to reverse genetics to investigate the role of a gene in pathogenicity is to determine the phenotype of mutants expressing multiple copies of the target gene. Such an experiment has been done with the subtilisin protease Pr1a from M. anisopliae (St. Leger et al., 1996c). Mutants containing multiple copies of the ORF behind a heterologous constitutive promoter produced protease in vitro on cuticle supplemented with 1% N-acetylglucosamine. The latter adjuvant is repressive in wild-type cultures. Large amounts of Pr1a were produced in the haemolymph during mycosis whereas only trace amounts were present in wild-type infections.

268

A. K. CHARNLEY

The enzyme caused hydrolysis of proteins, activation of endogenous trypsins involved in the regulation of prophenoloxidase and extensive melanisation; effects mimicked by the injection of pure Pr1. Survival time for infected insects dropped from 120 to 93 h. Accelerated death was due to the toxic action of the melanin as mutants had reduced fungal growth in the haemolymph. Reduced LT50 was not matched by reduced LC50 which would be expected if the increased protease had improved efficiency of cuticle penetration. This suggests that there are other constraints on cuticle penetration and that Pr1 protease was not limiting. Chitin constitutes 17–50% of the dry weight of insect cuticle; more pliant cuticles have a higher chitin content than stiff cuticles (Hillerton, 1984). In the main, chitin fibrils are laid down parallel to the cuticular surface and as such present a potential barrier to penetration by entomopathogenic fungi. Thus it is interesting that St. Leger et al. (1987c) failed to find evidence of the production of chitinase during the first critical 40 h after inoculation of C. vomitoria wings or abdominal cuticle of M. sexta larvae with M. anisopliae. NAGase activity was extracted from infected cuticle, but this enzyme has only trace activity against polymeric chitin. The apparent absence of chitinase from infected cuticle could be due to inadequate extraction or inhibitors in the cuticle (chitinases in vitro binds tightly to locust chitin in a nonionic manner (St. Leger et al., 1986b). Nevertheless, failure to detect the products of chitin hydrolysis in infected cuticle indicates that the activity of chitinase, if present, is negligible compared to that of protease. The slow appearance of chitinase in vivo (St. Leger et al., 1987c) is consistent with in vitro results (St. Leger et al., 1986c) and is probably due to the fact that chitinase is an inducible enzyme (St. Leger et al., 1986e) and cuticular chitin is masked by protein (St. Leger et al., 1986d). It seems likely that chitinase functions largely to provide nutrients during the saprophytic phase of fungal growth in cuticle of moribund insect hosts. St. Leger et al. (1996b) raised polyclonal antibodies against the two purified chitinases from M. anisopliae and used them for ultrastructural location of the enzymes. No label was found on pre-penetrant structures or around the fungus during penetration of the cuticle up to 36 h postinoculation, at a time when high levels of Pr1 and Pr2 are present. Label appeared and increased in intensity 36–60 h post-inoculation, both on the hyphal wall and, more often, extending into the surrounding cuticle. This is consistent with biochemical evidence that prior hydrolysis of protein by proteases expose chitin fibrils which results in the production and activity of chitinases. The importance of chitin as a mechanical barrier to penetration and as a stabiliser of the cuticular protein matrix in the absence of fungal chitinase, is

FUNGAL PATHOGENS OF INSECTS

269

evident from studies using acylurea insecticides that specifically inhibit chitin synthesis in insects. Diflubenzuron (as Dimilin) and teflubenzuron worked synergistically with Metarhizium to kill Manduca sexta (Hassan and Charnley, 1983) and Schistocerca gregaria (Joshi et al., 1992) respectively. Ultrastructural observations demonstrated that fungal penetration through Dimilin-treated cuticle was dramatically enhanced (Hassan and Charnley, 1983). Postecdysial Dimilin-treated cuticle (without chitin) was almost completely destroyed in contrast to pre-ecdysial cuticle (laid down prior to insecticide treatment) where hydrolysis was apparently selective (presumably protein only) and restricted to the vicinity of the fungal hyphae. Consistent with these ultrastructural observations, pharate fifth-instar Manduca cuticle, produced during treatment with Dimilin and thus completely disrupted by the insecticide (Hassan and Charnley, 1989), was considerably more susceptible to Pr1 than control cuticle (St. Leger, Charnley, and Cooper, unpublished). However, when multiple copies of chitinase gene from a M. anisopliae sf. acridum isolate 324 under a constitutive promoter from Aspergillus nidulans were transformed into M. anisopliae sf. anisopliae isolate 2575, the mutant did not show altered virulence to M. sexta (Screen et al., 2001). Since the mutant 2575 isolate produced large amounts of the isolate 324 chitinase in non-inducing and inducing (chitin containing) media, neither wild-type levels of chitinase nor its mode of regulation are limiting for cuticle penetration. It is interesting that overexpression of the heterologous gene caused early production of the endogenous chitinase, probably due to the production of soluble inducers. Askary et al. (1999) used the chitin-specific lectin wheat germ agglutinin to determine the effect of V. lecanii on chitin within the cuticle of the aphid Macrosiphum euphorbiae. The lectin was tagged with gold in a TEM study. Reduced labelling around penetrant hyphae was consistent with hydrolysis of chitin by the fungus. However, these effects on chitin were only observed when the fungus was already established in the aphid tissues suggesting that as in M. anisopliae so in V. lecanii chitinase is of secondary importance to the pathogenic process. St. Leger et al. (1987c) detected esterase on pregerminating and germinating conidia and appressoria of M. anisopliae on wings of C. vomitoria with naphthyl acetate and naphthyl propionate as substrates. It is very difficult to differentiate histochemically between esterase and lipase. Although many workers have identified lipase produced by entomopathogenic fungi solely on the basis of activity against Tweens (e.g., Michel, 1981), such substrates are degraded by non-specific esterases (Pearse, 1972). Most microbial lipases are serine enzymes (Brockerhoff and Jensen, 1974)

270

A. K. CHARNLEY

and as such would not be distinguished from the non-specific esterases produced by M. anisopliae. The different sites of activity against Tween 80 (localised on appressoria) and naphthyl-AS-nonanoate (localised on conidia) suggest, however, that at least for M. anisopliae the enzymedegrading Tween is distinguishable from at least one medium-chain-length non-specific esterase (St. Leger et al., 1986e). Doubts may be cast as to the involvement of lipases in penetration because St. Leger et al. (1987c) failed to extract true lipase (with activity against olive oil) from cuticles of M. sexta and C. vomitoria infected with M. anisopliae. Failure to detect may reflect numerous factors such as binding of lipase to fungal cell walls and host cuticles. However, notwithstanding this the possible biological role of lipase in pathogenesis is not as obvious as many have assumed. Ultrastructural studies which have demonstrated the early histolysis of the ‘wax’ layer on cuticles underneath infection structures might implicate lipase/esterase activity (e.g. Zacharuk, 1970a). However, in most insects ‘extractable’ cuticular lipids are composed mainly of a complex mixture of alkanes and alkenes with triacylglycerols and wax esters (potential substrates for lipases and esterases) being comparatively minor components (Hepburn, 1985). Lipids could only have a major role if the ‘bound’ surface lipids and ‘non-extractable’ lipid fraction contain esters. Unfortunately, however, very little is known of the chemistry of bound cuticular lipids or their esters (Blomquist, 1984). Wigglesworth (1970) suggested tentatively that the outer epicuticle (Fig. 3) is a multiple polyester cross-linked by ester bonds. If this is correct, then an esterase with characteristics somewhat similar to cutinase would be required for its hydrolysis. However, no such enzyme is produced by M. anisopliae or B. bassiana (St. Leger, Charnley, and Cooper, unpublished). Lipoprotein lipases (active against olive oil emulsion activated with human blood plasma) are secreted. This enzyme perhaps in conjunction with Pr1 would aid penetration of the inner epicuticle (polymerised lipoprotein) (Fig. 3).

H. EVOLUTIONARY CONSIDERATIONS

It has been suggested that fungi emerged onto land as endophytes of plants (Lewis, 1987). Entomopathogenic fungi could have evolved to attack herbivorous insects either from endophytes or from plant pathogens because they would associate in a common niche. Such host shifts would presumably involve adaptations of extracellular hydrolases to allow degradation of insect cuticle (St. Leger and Bidochka, 1996). Perhaps it is no coincidence that some common underlying mechanisms of fungal pathogenesis occur

FUNGAL PATHOGENS OF INSECTS

271

against insects and plants (Charnley, 1984). However, ability to infect insects must have evolved independently several times since entomopathogens are found in several divergent fungal groups. Thus while many phytoand entomopathogenic fungi are found in the same taxonomic grouping, the Entomophthorales have no plant-pathogenic relatives. St. Leger and Bidochka (1996) suggested alternatively that insect-pathogenic fungi evolved from alimentary canal commensals or fungi that employ insects for transmission of spores, while Evans (1988) suggested a switch to parasitisim by saprophytes living off insect cadavers. Entomophthorales fall within the Phylum Zygomycota and as such are towards the bottom of the fungal evolutionary tree, only preceeded by the Chytridiomycota (Berbee and Taylor, 2001). Among the Entomophthorales, Conidiobolus coronatus is one of the least specialised and least evolved; many of the others are host-adapted and difficult to growth in vitro on standard media (Evans, 1989). C. coronatus is probably essentially an opportunistic pathogen making the most of weakened hosts of many species (Papierok, 1986). In a recent EST analysis of genes expressed by C. coronatus during growth on insect cuticle as with M. anisopliae (Freimoser et al., 2003b) found chitinases, multiple subtilisins (one with homology to a gene from Thermus aquatica others with homology to C. carbonum genes), trypsin (with homology to a C. carbonum gene), metalloproteases (homology to a gene from Aspergillus fumigatus) and aspartyl protease (similar to pepsinogen from A. niger). In contrast to M. anisopliae subtilisins, those from C. coronatus and several other entomophthoralean fungi tested showed enhanced expression and activity rather than repression following addition of soluble protein to an inducing insoluble protein source. It would be interesting to know if this attribute allows production in host haemolymph during mycosis. A pathogen that relies primarily on biomass rather than toxins to overcome its host (see later) would benefit enormously from being able to hydrolyse host proteins early on rather than just accessing free amino acids (which are in high concentration in insect haemolymph). Some of the more specialised of the Entomophthorales viz. Erynia rhizospora, E. dipterogena and E. neoaphidis have a different kind of serine protease to the subtilisins found in the Hyphomycetes, that combine chymotrypsin and trypsin-like activity (Samuels et al., 1990). Unlike the subtilisins, these entomophthoralean enzymes have poor locust-cuticle degrading ability. This may be because their enzymes are highly adapted to host cuticle or that like the insect’s own moulting fluid cuticle degrading proteases they inherently have low cuticle-degrading ability. Controlled degradation of the old cuticle at the moult may be best served by slow

272

A. K. CHARNLEY

hydrolysis. Similarly the Entomophthoralean pathogens have an extended parasitic phase which may be promoted by a less aggressive degradation of host cuticle than that adopted by the toxin-assisted, subtilisin-facilitated necrotrophic Hyphomycetes (Samuels et al., 1991, 1993a,b; Samuels and Paterson, 1995). Facultative pathogens and saprophytes, which have to contend with a greater variety of sources of nutrients, often are polymorphic for their depolymerases and overall have greater genetic variation than obligate, specialised pathogens. Saprophytes such as Neurospora crassa, N. nidulans and Aspergillus spp. produce the broadest spectrum of exocellular depolymerases. This versatility should help such fungi tackle a broad diet, including the remains of dead insects and plants (St. Leger and Screen, 2000). Broad-spectrum proteases (subtilisins) are produced by saprophytes as well as insect pathogens. Thus, these enzymes are not per se an adaptation to insect pathogenicity (Gunkel and Gassen, 1989; St. Leger and Screen, 2000). The large number of Pr1 isoforms in M. anisopliae compared to subtilisins in saprophytes and plant-pathogenic fungi suggests that the proliferation of Pr1 genes in Metarhizium took place after the split between the plant and insect-infecting Pyrenomycetes (Bagga et al., 2003). St. Leger and Screen (2000) compared proteases produced by an opportunistic human pathogen, A. fumigatus, an insect pathogen, M. anisopliae and the plant pathogen Haematonectria haematococca. All three fungi produced subtilisins on mucin and lung polymers but smaller amounts on plant cell walls. However, insect cuticle was only stimulatory to M. anisopliae. Indeed only Metarhizium produced a full complement of isoforms on cuticle. The production of a similar complement of enzymes by all three species grown on host-related polymers suggests that some of the underlying mechanisms of fungal pathogenesis may be similar in insects, plants and animals. But there is some development of host-specific regulation of these factors. Consistent with this Bidochka et al. (1999) showed that V. lecanii 973, produced three key subtilisin isoforms only during growth on insect cuticle. Thus adaptation from dead to live host may be accompanied by altered properties and/or regulation of key hydrolases, allowing expression under conditions in which similar enzyme genes in nonpathogens are not transcribed (St. Leger et al., 1991b). It is interesting that such diverse fungi as M. anisopliae, V. lecanii, Tolypocladium niveum, Paecilomyces farinosus and Beauveria bassiana produce a subtilisin-like protease under nutrient limiting conditions (on chitin, cellulose, cuticle), which is repressed when low molecular weight compound such as N-acetylglucosamine is added. Regulation is at the level of transcription and is an adjustment to altered conditions and not a response to stress.

FUNGAL PATHOGENS OF INSECTS

273

Fungal pathogens of other organisms also produce subtilisins. VCP1 produced by the nematode pathogen Verticillium chlamydosporium has a similar pI (ca. 10), Mr (ca. 33 kDa) to Pr1 from M. anisopliae and is immunologically related. The two enzymes differ in inhibitor profile, substrate specificity and N-terminal sequence (Segers et al., 1999). A Pr1like enzyme has been cloned from the mycoparasitic fungus Trichoderma harzianum. Intriguingly, this protease, PrB1, which has the same substrate specificity as Pr1 against synthetic peptides, is also induced in a host-related manner, in this case by fungal cell walls or chitin (Geremia et al., 1993). In contrast to subtilisins, trypsins appear to have a specific role in pathogenesis (St. Leger et al., 1997). They are not produced by saprophytes N. crassa, N. nidulans and Aspergillus spp. They are the major proteases produced by plant-pathogenic Verticillium spp. (St. Leger and Screen, 2000) and at least some other plant-pathogenic fungi, e.g. Cochliobolus carbonum (Murphy and Walton, 1996). In contrast in M. anisopliae and V. lecanii subtilisins are produced in greater quantity than trypsins. The trypsins produced by the plant pathogens and the insect pathogens have different specificities. Whereas the insect pathogenic B. bassiana, V. lecanii, M. anisopliae have trypsins that are particularly active against Bz Phe Val Arg, the plant pathogens both from Verticillium spp and other genera have broad spectrum trypsins e.g. Cochliobolus carbonum (Murphy and Walton, 1996; St. Leger and Roberts, 1997; St. Leger and Screen, 2000). Expressed sequence tag libraries (EST) from cuticle-grown cultures of M. anisopliae sf. anisopliae isolate 2575 and the grasshopper-specific pathogen M. anisopliae sf. acridum isolate ARSEF 324 confirmed the wealth of cuticle-degrading enzymes revealed by earlier biochemical studies on this species (Freimoser et al., 2003a). In particular proteases comprised 36% and 20% of the ARSEF 2575 and ARSEF 324 libraries respectively, with transcripts of 11 subtilisin genes in the former and 3 in the latter. In comparison with the richness of proteases genes in these entomopathogens there is a relative paucity of such genes in the genomes of S. cerevisiae, N. crassa and A. nidulans and EST libraries on other Ascomycetes; no chymotrypsins, often no trypsins, and only 2 or 3 subtilisin genes. Thus while subtilisins per se are not an adaptation to entomopathogenicity, the production of a variety of such enzymes may be critical to success. Indeed paralogs of all 11 of the Pr1 genes expressed on cuticle by the generalist M. anisopliae sf. anisopliae isolate 2575 were found in sf. anisopliae isolate ARSEF820 (ex beetle) and sf. acridum isolate 324 (grasshopper specific) (Bagga et al., 2003). Insect cuticle comprises up to several hundred different proteins and a cocktail of enzymes with slightly different specificities may accelerate host invasion. Furthermore gene duplication and divergent

274

A. K. CHARNLEY

evolution may facilitate subtle and individual changes in regulation that promote production of each isoform at the right time, in the right place, in the right quantity and in the right combination. I. ROLE OF CUTICLE-DEGRADING ENZYMES IN VIRULENCE AND SPECIFICITY

The possibility that virulence among isolates may be correlated (at least in part) with cuticle-degrading enzyme activity has stimulated several studies, with conflicting results (see Charnley, 1984). El Sayed et al. (1989) showed a relationship between chitinase activity (exo and endoenzymes) and virulence among isolates of B. bassiana. The greatest difference occurred during germination, though the most rapid increase in chitinases occurred at the onset of blastospore formation. However, the numbers of isolates involved was very small. Gupta et al. (1994) found high virulence among five isolates of B. bassiana for Galleria mellonella was associated with high activity of chymoelastase, chymotrypsin and chitinases. Comparisons between isolates for pathogenicity and production of enzymes, however, may only reveal the great variability within a species for numerous factors, many of which may influence but be unrelated to cuticle-degrading enzyme activity. Induction of mutants within a common genetic background is an alternative approach which has also been exploited. Dasilva et al. (1989) found M. anisopliae mutants selected for high amylase and lipase that showed greater virulence for Triatoma infestans, whereas hyperproductive protease mutants were either similar or less virulent than wild-type. Hypoproductive protease mutants from B. bassiana showed reduced virulence towards Melanoplus sanguinipes (Bidochka and Khachatourians, 1990). Comparable studies by others do not all show similar association. Interpretation of the results is, however, complicated as pleiotropic effects cannot be ruled out; in no case was a point mutation established. Paris and Ferron (1979) found a link (which is ambiguous; see earlier) between virulence and lipase in mutants of B. brogniartii. However, mutants deficient in other respects were also avirulent, suggesting that, as one might expect, pathogenicity was a function of many attributes. Similarly, Pekrul and Grula (1979) found that low pathogenicity of B. bassiana mutants against H. zea larvae, was not simply a consequence of the lack of a suitable enzyme cocktail. All mutants possessed varying levels of the three major enzyme activities regardless of pathogenicity and some mutants contained very high levels of certain activities and yet were poor pathogens. Another approach to unravelling the role of cuticle-degrading enzymes in host specificity and virulence is to look at the properties of the enzymes

FUNGAL PATHOGENS OF INSECTS

275

concerned. As is clear from this review, at present only endoproteases for Deuteromycete entomopathogens have been studied in sufficient detail to draw any significant conclusions. Cuticle-degrading endoproteases, with similar modes of action, are produced in quantity by all species studied (St. Leger et al., 1987b), and it seems unlikely that they contribute to host specificity or virulence, though the occurrence of two or three types of extracellular protease in five genera of entomopathogen implies an indispensable function for these enzymes (St. Leger et al., 1987b). Leal et al. (1997) using nested PCR, have demonstrated significant variation in Pr1 gene sequence among 54 isolates from seven fungal species (primarily M. anisopliae but including M. flavoviride). Furthermore Pr1-like enzymes from isolates of M. anisopliae and other entomopathogenic Deuteromycetes differ in amino acid sequence since they vary in their response to a polyclonal antibody raised against Pr1 from M. anisopliae (2575) (St. Leger et al., 1987b). Even a few strategic amino acid substitutions can affect cuticle binding and hence activity of Pr1 from M. anisopliae (St. Leger et al., 1986a, 1992a). Thus it is possible that virulence between isolates can be influenced by variations in Pr1 sequence. Furthermore protein composition of insect cuticle varies between insect species. Isolate specificity could in part depend on the impact of Pr1 enzymes on host cuticular proteins. Thus it is interesting that purified basic isoforms of Pr1 from aphid isolates of V. lecanii degraded aphid cuticle significantly more effectively than non-host locust cuticle (Bye and Charnley, unpublished). Cuticles of Hyalophora cecropia (Cox and Willis, 1985) and Locusta migratoria (Hojrup et al., 1986) have a non-uniform distribution of charge, with negatively charged and positively charged proteins predominating in flexible cuticle (e.g. arthrodial membranes) and rigid body-wall cuticle, respectively. Thus, it is possible that regions of the cuticle may be favourable or unfavourable to binding (and thus degradation) by individual enzymes, with consequences for the parts of the insect body which can be invaded by enzymatic action. This may influence speed of penetration and thus virulence. Gillespie et al. (1998) looked at the ability of partially pure cocktails of Pr1 isoforms from cuticle cultures of 19 locust-pathogenic isolates of Metarhizium spp. to degrade insect cuticle. The cuticle types used were Manduca sexta pupal cuticle, adult desert locust (Schistocerca gregaria) wing cuticle, abdominal cuticle and pharate adult locust abdominal cuticle. These cuticle types differ in their protein composition and degree of sclerotisation. Pr1 hydrolysed the locust cuticle to different degrees. The data suggest a hierarchy for the susceptibility to hydrolysis in the order pharate adult abdominal>adult abdominal>wing. There was no significant

276

A. K. CHARNLEY

correlation between the ability of the enzymes to degrade any of the cuticle types and median lethal time of the isolates against the desert locust. Bidochka and Khachatourians (1994) also found that proteases from M. anisopliae and B. bassiana had differential action against acid and basic proteins from different parts of grasshopper cuticle. The possibility that particular protease isoforms may play a part in host specificity is suggested by a recent paper by Wang et al. (2002). They found three spontaneous mutants of M. anisopliae that did not produce Pr1A or Pr1B; mutants were identified by nested PCR and confirmed by Southern analysis. These mutants had wild-type virulence to Galleria mellonella but 20% reduction in pathogenicity for Tenebrio molitor. The possible importance of regulation not only in determining entompathogenicity per se but also species specificity is suggested by the observation that an EST library of the grasshopper-pathogen M. anisopliae sf acridum isolate ARSEF 324 grown on non-host cuticle (cockroach) contained transcripts of only a few of the trypsin and subtilisin genes identified in the genome whereas the EST library of the generalist isolate ARSEF 2575 was fully representative of the genome (Freimoser et al., 2003a). The virulence of facultative pathogens like M. anisopliae is the result of many factors that operate in concert to overcome the host. Identifying putative virulence determinants is potentially even more problematic for an opportunistic fungus like A. fumigatus in which there has probably been no selection for specific virulence genes (St. Leger and Screen, 2000).

VI. TOXINS A. OVERVIEW

The definition of what constitutes a ‘toxin’ depends on the scientific discipline. A bacteriologist working on mammalian pathogens would use the term to describe high molecular weight proteins that have detrimental effects on their hosts (Roberts, 1980). Plant pathologists have restricted the term ‘toxin’ largely to low molecular weight compounds that are bioactive in low concentrations and are often products of so-called ‘secondary metabolism’ (Graniti, 1972). In this review no restrictions will be made on what can or cannot be called a toxin. Though, reflecting the state of knowledge, the focus will be primarily on small molecules, the potential effects of fungal proteases in the haemolymph during mycosis will also be considered thus bringing together both aspects of this review.

FUNGAL PATHOGENS OF INSECTS

277

Compounds that have no apparent role in the primary metabolic processes of living things have been termed secondary metabolites; a name that reflects early views of their significance as a means of disposing of waste metabolic intermediates. ‘Secondary metabolism’ is particularly well developed in higher plants and fungi (Vining, 1990). Amongst the latter the most prolific producers occur in groups with adaptability to changing environments e.g. Basidiomycetes and Ascomycetes. Thus, it has been suggested that secondary metabolism is an area where biochemical evolution is taking place. In this view occasionally substances are manufactured that are of benefit to the producer. The problem with this approach is that it implies that some of the complex molecules synthesised by fungi may have no functional significance. This is hard to equate with the metabolic costs involved in their production. The range of biological activities of diverse and often complex molecules has ensured a large interest and commensurate literature. B. INCIDENCE OF INSECTICIDAL TOXINS AMONGST ENTOMOPATHOGENIC FUNGI

A key problem for a pathologist is how to make sense of the bewildering diversity of toxic metabolites produced in vitro by entomopathogenic fungi in the context of the disease process. In part this is a technical problem viz. how to detect small quantities of chemicals in small insects. Additionally as 60 years of research on the mode of action of synthetic chemical insecticides have shown, detailed analysis of the in vitro effects of biocides may be hard to reconcile with symptoms expressed in vivo. A priori it seems reasonable that insect pathogenic fungi might produce metabolites that would serve one or more of the following functions: toxic to the host and help to cause death; immunosuppressive to aid the fungus overcome host defence; antibiotic to suppress competition from other pathogens or the saprophytic flora on the cadaver post mortem; toxic to mycophagous organisms and thus provide a defence outside of the host. Early commentators concluded from a consideration of histopathology that entomopathogens among the lower fungi (e.g. of the genera Coelomomyces, Lagenidium, Entomophthora) overcome susceptible hosts by utilising available nutrients in the haemolymph, followed by colonisation of parts or all of the living host (Evans, 1989). They may need to be fast growing when they are pathogens of rapidly developing insects or those which have short-lived adult stages e.g. aphids. Sporulation occurs very soon after host death – the saprophytic phase is minimal or completely absent. Locusts and grasshoppers infected with Entomophthora grylli climb

278

A. K. CHARNLEY

to the tops of plants prior to death. This so-called ‘summit disease’ may promote spore dispersal. It is certainly not in the best interests of the host which expose themselves to predation by this behaviour. While it has been suggested that fungal metabolites may control this behaviour, summit disease occurs in a number of insect species infected with viruses or fungi suggesting possibly a preprogrammed insect behaviour that is triggered nonspecifically by microbial pathogens. Consistent with these observations on mycosis, no toxic secondary metabolites have been described in cultures from Entomophaga, Erynia, Lagenidium or Harposporium species (Anke and Sterner, 2002). Initial reports of a high molecular weight toxin produced by Entomophaga aulicae (¼Entomophthora egressa) (Dunphy and Nolan, 1982), have not been confirmed (Tyrrell, 1990). A 32-kDa cell lytic factor has been semi-purifed from larvae of the spruce budworm Choristoneura fumiferana mycosed with E. aulicae (Milne et al., 1994). However claims that this protein is primarily responsible for symptoms observed in moribund larvae and host death are premature since it has not been established that the protein comes from the fungus. Some of the most primitive of the Entomophthorales like Conidiobolus spp. may produce toxic secondary metabolites in vitro (Roberts, 1980) e.g. 4, 40 hydroxymethyl azoxybenzene carboxylic acid (Fig. 6b) from C. thromboides (Claydon, 1978) and a 30-kDa insecticidal protein (Bogus and Scheller, 2002) from C. coronatus. Furthermore symptoms of wax moth larvae infected with C. coronatus were consistent with the effects of a toxin (Bogus and Szczepanik, 2000). Thus it is interesting that (Freimoser et al., 2003b) in their EST study of gene expression in C. coronatus found few genes for toxic enzymes (e.g. no phospholipases), insecticidal secondary metabolite or antibiotic synthesising enzymes. Instead the high proportion of ESTs encoding ribosomal proteins and enzymes of intermediary metabolism suggest a facility for rapid growth. Of course, as these authors state this does not preclude the existence of a novel protein toxin. In contrast to most of the Entomophthorales, the Deuteromycete and Ascomycete entomopathogens tend to have wider host ranges and kill their hosts rapidly. For many of these fungi circumstantial evidence is consistent with the involvement of fungus-derived biocides in pathogenesis: sparse growth of the fungus in the haemolymph of the host prior to death; changes in host behaviour such as reduced activity, paralysis, reduced feeding; pathogenic changes in the ultrastructure of host tissues in advance of penetrating hyphae; compromised immune system; reduced microbiota on the cadaver. Rapid killing consequent upon the activity of toxins, resulting in a low fungal biomass at the time of host death may reduce

FUNGAL PATHOGENS OF INSECTS

279

Fig. 6. (a) Cytochalasin (from Metarhizium anisopliae) a perhydroindole with a macrocyclic ring, (b) 4, hydroxymethyl azoxybenzene 4 carboxylic acid (from Conidiobolus thromboides ( ¼ Entomophora virulenta), (c) Beauvericin (from Beauveria bassiana, B. brogniartii, V. lecanii). A cyclic hexadepsipeptide comprising three residues of D--hydroxyvaleric acid alternating with three molecules of N-methyl-L-phenylalanine, (d) Swainsonine (from Metarhizium anisopliae), an indolizidine alkaloid, (e) Enniatin (from Fusarium spp.). Cyclohexadepsipeptides comprising three molecules of d-2-hydroxyisovaleric acid alternating with L-amino acids or N-methyl-L-amino acids. Enniatin A has 3 N-methylvaline residues (R1, R2, R3). R4 and R5 ¼ CH2.

the ability of the fungus to compete with the saprophytic microbiota unless antibiotic secondary metabolites are also produced. In common with other members of the Entomophthorales, C. coronatus uses rapid growth to overcome its host and exploit the cadaver before competitors overrun it.

280

A. K. CHARNLEY

What follows is a review of the most important of the toxic secondary compounds that have been identified in cultures of entomopathogenic fungi and an assessment of their actual or potential role in the development of disease. Evidence for the involvement of secondary metabolites in pathogenesis is limited to just a handful of compounds. Since some of these fungi are facultatively pathogenic with an alternative saprophytic existence in soil, the adaptive significance of these compounds in some cases may be to deter fungivorous insects or other Arthropods. Furthermore horizontal transfer of genes may result in the acquisition by entomopathogens of a variety of useful defensive compounds originally evolved by saprophytes to prevent mammalian and insect mycophagy. A further interesting ecological aspect of tritrophic relationships is seen in the periodic population decline of spruce budworm caused by the insecticidal cyclic peptide enniatin A/A1 (Fig. 6e), produced by the phylloplane fungus Fusarium avanaceum (Strongman et al., 1987). There are other examples of plants gaining protection from insect and mammalian grazers through toxins produced by endophytic fungi (Clay, 1988). If, has been suggested, that fungi emerged onto land as endophytes of plants (Lewis, 1987) and entomopathogenic fungi evolved to attack herbivorous insects from endophytes because they associate in a common niche, then they would have a ready-made arsenal of insecticidal toxins. Interestingly B. bassiana exists endophytically in certain genotypes of maize (Vakili, 1990). Furthermore translocation within plants of conidia initially applied externally can provide season-long control of the European cornborer, Ostrinia nubilalis (Bing and Lewis, 1991). Though dealing with microbial competitors on or off the host or its cadaver must be a key issue for an insect pathogenic fungus, there are comparatively few reports of antibiotic activity. This of course probably reflects a failure to look. C. A SURVEY OF TOXIC METABOLITES PRODUCED BY ENTOMOPATHOGENIC FUNGI

A huge variety of toxic metabolites are produced by entomopathogenic fungi (see Table II which is not exhaustive, and Figs. 6 and 7). Of particular interest among non-protein metabolites are the viridoxins from ethanolic extract of mycelia of M. anisopliae var. flavoviride, which are closely related to Colletochin and Colletotrichin produced by the plant pathogen Colletotrichum nicotianae (Gupta et al., 1993a). The viridoxins are insecticidal against the Colorado beetle Leptinotarsa decemlineata. Oosporein, a red-pigmented dibenzoquinone antimicrobial pigment,

FUNGAL PATHOGENS OF INSECTS

281

is produced by isolates of B. bassiana and B. brogniartii and some other nonentomopathogenic soil fungi (Eyal et al., 1994; Strasser et al., 2000). Oosporein is antibiotic against gram-positive bacteria and to a lesser extent gram-negative also (Brewer et al., 1984). Oosporein appears to be produced in cuticle but not haemolymph during mycosis of Manduca sexta by B. bassiana so it could help to suppress the saprophytic surface microbiota of the host, though there was no significant correlation between isolate virulence and incidence of oosporein in cadavers, which militates against a significant role in pathogenesis (Foley, Reynolds and Charnley, unpublished). Oosporein disrupts membranes (Jeffs and Khachatourians, 1997) which may account for inhibitory effects on Naþ/Kþ and Ca2þ ATPases. Interestingly oosporein was the only secondary metabolite produced in quantity by isolates of B. brogniartii (Strasser et al., 2000). Oxalic acid is a pathogenicity determinant of some plant pathogens (Godoy et al., 1990) and this organic acid is also produced by Beauveria bassiana (Kodaira, 1961). Twelve isolates of B. bassiana all produced significant quantities of oxalic acid in Czapek-Dox liquid medium (mean of 400  46 mg l1) (Foley, Charnley and Reynolds, unpublished). Although oxalic acid (108  9.5 mg ml1) was present in the haemolymph of Manduca caterpillars infected with a virulent isolate of B. bassiana 48 h after inoculation, significant amounts were also present in uninfected controls (36  2.2 mg ml1). Oxalic acid could help facilitate invasion of the cuticle. Bidochka and Khachatourians (1991) showed that oxalic acid can solubilise cuticular protein and suggested that it may synergise with cuticle-degrading enzymes. However, subsequently they established that hyperproductive oxalic acid mutants were no different in pathogenicity for grasshoppers than wild-type (Bidochka and Khachatourians, 1993a). Isolates of Aspergillus flavus and Aspergillus parasiticus may be saprophytic, plant-pathogenic or insect-pathogenic. Isolates of all three nutritional strategies produce aflatoxins. The toxic effects of these compounds on insects include delayed development, diminutive pupal and adult size, reduced fecundity and sterility (Wright et al., 1982). Fifteen pathogenic strains of A. flavus of silkworm (Bombyx mori) produced aflatoxins in vitro and aflatoxins have been extracted from infected larvae, though it is not clear whether in sufficient quantities to be the cause of death (Murakoshi et al., 1977). Aflatoxin production in vivo has been demonstrated also by Drummond and Pinnock (1990). Although 14.2 mg g1 of aflatoxin wet weight of insect was extracted from infected mealybugs 7 d after inoculation, not all pathogenic isolates produced aflatoxins and a highly virulent isolate lost the ability to produce aflatoxin without any reduction in pathogenicity. Entomopathogenic strains of Aspergillus spp.

282

TABLE II Toxins produced by entomopathogenic fungi Chemistry

Name of compound

Fungus

Effect

Reference

Indolizidine alkaloid Perhydroindole with a macrocyclic ring A dibenzoquinone

swainsonine cytochalasins

M. anisopliae ditto

 mannosidase inhibitor Inhibit cell movement

Tamerler et al. (1998) Aldridge and Turner (1969)

oosporein

B. bassiana

Vining et al. (1962)

Diterpene derivatives of polysubstituted g-pyrones Organic acid Hydoxamic acid 3-(6,8-dimethyl-E, E, E-deca-2, 4,6,-trienoyl)-1,4-dihydroxy-5(p-hydroxyphenyl)-2(1H)-pyridone 3-(4,6-dimethyl-E, E-octa-2, 4-dienoyl)-1, 4-dihydroxy-5-(p-hydroxyphenyl)2(1H)-pyridone 5-n-butylpyridine-2-carboxylic acid 5-hydroxy-2-(hydroxymethyl)4H-pyran-4-one

viridoxins

M. flavoviride

Red pigment, antibacterial (gram positive>gram negative) Insecticidal

oxalic acid tolypocin bassianin

B. bassiana T. geodes B. bassiana

Hydrolyses protein Siderophore Yellow pigment

Kodaira (1961) Jegorov et al. (1993) Wat et al. (1977)

tenellin

B. bassiana

Yellow pigment

ditto

fusaric acid kojic acid

Fusarium solani Aspergillus flavus

Claydon et al. (1977) Beard and Walton (1969); Dowd (1999)

Polyketide

aflatoxin B1

ditto

6-(1-prophenyl)-5,-6-dihydro-5hydroxypyran-2-one)

phomalactone

Hirsutella thompsonii

Insecticidal Insecticidal, inhibits insect phenoloxidase Insecticidal, interferes with growth and development Insecticidal, antifungal (inhibits germination of B. bassiana) Insecticidal by injection

Entomophthora virulenta ( syn Conidiobulus thromboides Drechseler)

Murakoshi et al. (1977)

Krasnoff and Gupta (1994)

Claydon (1978)

A. K. CHARNLEY

40 -hydroxymethylazoxybenene-4carboxylic acid

Gupta et al. (1993a)

triterpenoid (3S,6R)-4-methyl-6-(1-methylethyl)3-phenylmethyl-1, 4-perhydrooxazine-2,5-dione Pyridine-2,6-dicarboxylic acid Two C25 keto acids Diterpene derivatives of polysubstituted - g - pyrones Linear peptide

M. anisopliae B. bassiana

Antibiotic Platelet aggregation inhibitor

Turner and Aldridge (1983) Kagamizono et al. (1995)

dipicolinic acid viridoxins

V. lecanii ditto M. anisopliae var flavoviride

Insecticidal ditto Insecticidal

Claydon and Grove (1982) ditto Gupta et al. (1993a)

efrapeptins

Tolypocladium niveum

Krasnoff and Gupta (1991)

H. thompsonii

Protein

Mr 14 kDa, hirsutellin Mr 10 kDa

Insecticidal, antimicrobial, inhibitor of ATPase Ribosomal inhibitor

B. bassiana

Immunosuppressive

Protein

TF2

Beauveria sulfurescens

Protein

Mr 100–129 kDa

Beauveria sulfurescens

Protein Cyclic peptide

hydrophobin cyclosporin

Cyclic peptide

beauvericin

M. anisopliae Tolypocladium inflatum, Tolypocladium spp, B. bassiana, B. brogniartii, V. lecanii B. bassiana, P. fumoso-roseus, Fusarium spp.

Cytotoxic to insect but not mammalian cell lines Insecticidal, causes cuticle melanisation Insecticidal Immunosuppressive

Cyclic peptide

destruxin

M. anisopliae

Cyclic peptide

beauverolides

Cyclic peptide Cyclic peptide

bassianolides enniatins

B. bassiana, B. brogniartii, P. fumoso-roseus, Isaria sp., B. bassiana, V. lecanii, Fusarium spp., Alternaria kikuchiana

Protein

Insecticidal, antibacterial (gram positive), antifungal, ionophore Immunosuppressive

Mazet and Vey (1995); Liu et al. (1996) Mazet et al. (1994) Mollier et al. (1994a) Mollier et al. (1994b) St. Leger et al. (1992c) Weiser and Matha (1988a); Jegorov et al. (1990)

FUNGAL PATHOGENS OF INSECTS

helvolic acid bassiatin

Hamil et al. (1969)

Insecticidal Insecticidal

Suzuki et al. (1977) Visconti et al. (1992)

283

Immunosuppressive

Kodaira (1961); Vilcinskas et al. (1997c) Grove (1980)

284

Fig. 7.

A. K. CHARNLEY

Structures of key members of the destruxin series

produce a variety of secondary metabolites including cyclopiazonic acid, aflatrem (Richard and Gallagher, 1979) and Kojic acid (Beard and Walton, 1969). The last named is insecticidal and can act synergistically with aflatoxin B-1 (Dowd, 1988).

FUNGAL PATHOGENS OF INSECTS

285

Cytochalasins are a family of perhydro-indoles with a macrocyclic ring (Table II, Fig. 6a). They were first isolated from M. anisopliae (Aldridge and Turner, 1969) though other fungi are now known to produce them. An injected dose of up to 150 mg g1 has no detrimental effects on Galleria mellonella larvae (Roberts, 1980). However, cytochalasins are inhibitors of filamentous actin formation and in vitro reduce phagocytosis and spreading of insect haemocytes (Vilcinskas et al., 1997b,c). Thus by interfering with the host cellular immune response these compounds could still have a role in mycosis. A variety of proteinaceous toxins have been found. Cyclic peptides will be considered separately in the next section. The linear molecules will be surveyed here. Efrapeptins are a group of linear peptides, comprising 15 amino acids. They are produced only by Tolypocladium spp. (Krasnoff and Gupta, 1991) and its possible teleomorph Cordyceps subsessilis (Hodge et al., 1996). There is intra and interspecific variation in efrapeptin production (Krasnoff and Gupta, 1992). Efrapeptins have limited antifungal and antibacterial activity as well as being insecticidal by injection and/or or by contact (Matha et al., 1988; Krasnoff et al., 1991). Antifeedant and growth inhibitory properties have been noted in treated insects also (Bandani et al., 2000). They have been extracted from the haemolymph and whole bodies of G. mellonella infected with T. niveum but not in concentrations sufficient to cause host death. Furthermore, though limited growth of T. niveum in non-natural hosts such as G. mellonella prior to death is restricted, consistent with a toxin (Bandani et al., 2000), hyphae ramify through mycosed mosquitoes and Chaoboridae (natural hosts of Tolypocladium spp.) and death has been attributed to starvation (Turlington et al., 1990). Weiser and Matha (1988b) suggested that efrapeptins probably evolved as broad scale antibiotics rather than as components of the pathogenic process. Efrapeptins are potent inhibitors of ATPases from bacteria, and chloroplasts and mitochondria from a variety of organisms (Krasnoff et al., 1991; Fricaud et al., 1992). They also blocked cell surface expression of viral glycoproteins in a cell line infected with Newcastle disease virus and vesicular stomatitis virus but not the uptake of ricin and diptheria toxin from which Muroi et al. (1996) concluded that efrapeptins block exocytic but not endocytic trafficking of proteins. Interestingly though, efrapeptins are inhibitors of F-ATPase (Cross and Kohlbrenner, 1978), a key enzyme in ATP regeneration from ADP. The mite pathogen Hirsutella thompsonii produces Hirsutellin A, a 15-kDa, pI 10.5, non-glycosylated, thermostable protein (Liu et al., 1995) that is toxic by contact to its host, the citrus rust mite Phyllocoptruta oleivora (Omoto and McCoy, 1998). Hirsutellin is also insecticidal by

286

A. K. CHARNLEY

injection or by mouth to a range of insects, including aphids, fruit flies, mosquitoes (Mazet and Vey, 1995) and Lepidoptera which it kills slowly (within 8d) (Liu et al., 1995). It is cytolytic to insect cells, but no other eukaryotic or prokaryotic cells tested (Liu et al., 1996). It inhibits protein synthesis by specific cleavage of rRNA and as such is similar to the wellcharacterised ribosomal-inhibiting proteins (RIPs),  sarcin, mitogelin and restrictocin (Liu et al., 1996; Boucias et al., 1998). However, the hirsutellin gene is unique. The cloned cDNA does not contain the RNAase motif of fungal RIPs but does have other features in common with these proteins e.g. a series of consensus phosphorylation and myristoylation sites (Boucias et al., 1998). Furthermore it encodes a single polypeptide rather than the two observed in some RIPs (Liu et al., 1996). Hirsutellin A was isolated from 162 mite-associated isolates of H. thompsonii. Most isolates (100/162) contained the HtA gene. However, there was a lack of correlation between either the presence of the HtA gene or the amount of hirsutellin protein (semiquantified using a monoclonal antibody) and mortality induced by culture filtrates or pathogenicity of isolates for Galleria mellonella. Thus HtA is not required for survival or pathogenicity and strains are likely to produce other toxins not yet characterised (Maimala et al., 2002). Cell-free haemolymph of larvae of the lepidopteran S. exigua in the later stages of infection with B. bassiana contains a >10 kDa protein that is highly toxic by injection to healthy larvae. The protein caused a permanent reduction in the numbers of spreading plasmatocytes though haemocyte viability and phagocytcic competence (determined in vitro) was not affected; symptoms also seen in haemocytes from mycosed insects. Insects appeared normal but died later in development (Mazet et al., 1994).

D. CYCLICPEPTIDE TOXINS

1. Overview Depsipeptides are a very varied group of compounds that comprise hydroxy and amino acids joined by amide and ester linkages to form a cyclic structure (Visconti et al., 1992). The amino acids include methylated and other non-protein forms. The same compounds can be produced by a number of species (see Table III). Each compound often comprises a family of similar molecules which differ in one or a few residues. In addition there are a number of related depsipeptide families e.g. beauverolides (produced by B. bassiana, B. brogniartii, P. fumoso-roseus, Isaria sp.), bassianolides (B. bassiana, V. lecanii), beauvericins (B. bassiana, P. fumoso-roseus, Fusarium spp.) beauvaricins (B. bassiana), isarolids (¼ beauverolides), isariins (Isaria

FUNGAL PATHOGENS OF INSECTS

287

TABLE III Insecticidal secondary metabolites common to both entomopathogenic* and non-entomopathogenic fungi Compound

Chemistry

Species

Reference

Beauvericin

Cyclic peptide

Beauveria bassiana* Paecilomyces fumoso-roseus* Paecilomyces tenuipes* Fusarium proliferatum F. semitectum F. subglutians Metarhizium anisopliae* Aschersonia sp* Ophiosphaerella herpotricha Alternaria brassicae Fusarium oxysporum* F. acuminatum F. avenaceum F. lacteritium*

Hamil et al. (1969)

Destruxins

Enniatins

Cyclic peptides

Cyclic peptides

Peeters et al. (1983) Nilanonta et al. (2002) Logrieco et al. (1997) Gupta et al. (1991) Logrieco et al. (1998) Kodaira (1961) Krasnoff et al. (1996) Venkatsubbaiah et al. (1994) Buchwaldt and Jensen (1991) Madry et al. (1983) Visconti et al. (1992) Strongman et al. (1987) Tsantrizos et al. (1993)

fwelina, Isaria cretacea) and enniatins (Fusarium spp., Alternaria kikuchiana) (Anke and Sterner, 2002). Beauvericin (see Fig. 6c) is a hexadepsipeptide, comprising a cyclic repeating sequence of three molecules of phenylalanine alternating with three molecules of hydroisovaleric acid. It is closely related to the enniatins (Fig. 6e) (Visconti et al., 1992). It was first identified from B. bassiana (Grove and Pople, 1980; Gupta et al., 1995) though it has recently been identified also in some plant pathogens e.g. Fusarium spp. (Gupta et al., 1991; Logrieco et al., 1998). Not all isolates of B. bassiana produce beauvericin (Frappier et al., 1975). Several analogues of beauvericin have been described (Gupta et al., 1995). Beauvericin is cationophoric and will increase permeability of membranes to Naþ and Kþ ions (Steinrauf, 1985) and uncouple oxidative phosphorylation in isolated mitochondria. Apoptosis-like cell death was initiated in a murine cell line by beauvericin as a consequence of an increase in the cytoplasmic calcium concentration (Ojcius et al., 1991). Beauvericin is bacteriocidal (Ovchinnikov et al., 1971) and toxic to some insects (Gupta et al., 1995) but not all (Champlin and Grula, 1979).

288

A. K. CHARNLEY

Fusarium lateritium, a pathogen of the scale insect, Hemiberlesia rapax, produces enniatins (Tsantrizos et al., 1993). These are less active against the mosquito Aedes aegypti, than beauvericin but are more insecticidal against the blowfly Calliphora vicina (Grove and Pople, 1980). Plant-pathogenic Fusarium spp. also produce beauvericin and enniatin and both toxins have been implicated in disease (Hohn, 1997). Beauverolide L was isolated by Jegorov et al. (1990) from mycelia of B. brogniartii. It is a tetradepsipeptide with a Mr of 516. Fifteen members of the family have been characterised. Bassianolide, a cyclo-octadepsipeptide that comprises four molecules each of L-N-methyl leucine and D--hydroxyisovaleric acid, was extracted from mycelia of B. bassiana (Suzuki et al., 1977). It is lethal to B. mori by ingestion. An injected dose caused atonic symptoms in B. mori, similar symptoms are seen in larvae infected with the fungus (Murakoshi et al., 1978). Cyclosporins are cyclic undecapeptides with immunosuppressive and antiinflammatory activity in humans as well as having antiparasitic and antifungal properties (Wartburg and Traber, 1988). They were originally isolated from Trichoderma polysporum (Ruegger et al., 1976) and many kinds of soil fungi, but more recently from Tolypocladium cylindrosporium, a pathogen of mosquito larvae. B. nivea, B. bassiana, B. brogniartii and Verticillium sp are among many entomopathogenic fungi that have subsequently been found to produce cyclosporins, though they were absent from isolates of Hirsutella sp, Metarhizium sp, Paecilomyces farinosus and P. fumoso-roseus (Jegorov et al., 1990). Cyclosporins A, B and C have insecticidal properties (Weiser and Matha, 1988a). Swollen mitochondria and vacuolated rough endoplasmic reticulum occur in toxin-treated cells of mosquito larvae and in Malpighian tubules incubated in vitro (Dumas et al., 1996b). However, injection or feeding of cyclosporin to G. mellonella and Spodoptera exigua had no effect, probably because of rapid removal from the haemolymph. Haemolymph proteins, particularly lipophorin, bind cyclosporin and remove it to the fatbody (Vilcinskas et al., 1997a). 2. Destruxins: A Case History The entomopathogenic fungus Metarhizium anisopliae produces a family of cyclic peptide toxins known as the destruxins (DTX) (see Fig. 7). To date some 30 variants have been identified in cultures of this fungus; all comprise five amino acids and an hydroxyacid (Kodaira, 1961; Suzuki et al., 1970; Suzuki and Tamura, 1972; Pais et al., 1981; Gupta et al., 1989; Wahlman and Davidson, 1993; Chen et al., 1995; Yeh et al., 1996; Jegorov et al., 1998). DTXs are produced during active fungal growth (Roberts, 1966a; Samuels

FUNGAL PATHOGENS OF INSECTS

289

et al., 1988a; Amiri-Besheli et al., 2000) and DTX A, E and B tend to predominate in culture. Destruxins have been described also from the entomopathogenic fungus Aschersonia sp. (Krasnoff et al., 1996). It is interesting that DTXs (including two not found in M. anisopliae cultures) are produced by three unrelated plant pathogenic fungi, Alternaria brassicae (Bains and Tewari, 1987), Trichothecium roseum (Springer et al., 1984) and Ophiosphaerella herpotricha (Venkatsubbaiah et al., 1994). Natural analogues of destruxins have also been found e.g. roseotoxin (Engstrom et al., 1975) and bursephalocids (Kawazu et al., 1993). Injection of DTX into lepidopteran larvae and adult Diptera causes immediate, tetanic muscular paralysis, followed by flaccidity; insects recover from low doses while high doses are lethal (Kodaira, 1961; Roberts, 1966b; Samuels et al., 1988a,b,c). However, DTX has a wide range of other effects. DTXs inhibit Malpighian tubule fluid secretion in Schistocerca gregaria (James et al., 1993) and ecdysteroid secretion by the prothoracic glands of Manduca sexta (Sloman and Reynolds, 1993) while stimulating the heart beat of M. sexta (Samuels, 1998). Cytopathological effects have been observed in cultured cells (Quiot et al., 1985; Odier et al., 1992) and in cells of the midgut and Malpighian tubules treated in vitro or in vivo with low doses of DTX (Vey and Quiot, 1989a; Dumas et al., 1996b). Symptoms include vacuolisation of the cytoplasm caused by dilation of the endoplasmic reticulum, appearance of vesicles on microvilli and aggregation of chromatin in nuclei (Quiot et al., 1985; Dumas et al., 1996a). DTX also inhibits viral replication in culture cells (Quiot et al., 1980; Yeh et al., 1996) via an effect on RNA and DNA synthesis. Several studies have shown immunomodulatory effects of DTXs including inhibition of nodule formation around injected spores of Aspergillus niger (Vey, 1985; Huxham et al., 1989; Vilcinskas et al., 1997c). DTXs can be toxic by injection, ingestion (Amiri et al., 1999) and/or topical application (Poprawski et al., 1994), depending on variant and host, though it is not clear how penetration of the cuticle is achieved. Lepidoptera and adult Diptera are particularly susceptible; on the whole other insects less so. The injected 24 h-median lethal dose for DTX A and B on Bombyx mori caterpillars was 0.015–0.030 mg g1 (Kodaira, 1961) but Galleria mellonella was 10–30 times less susceptible (Roberts, 1966b). DTX A, B, C are repellant and antifeedant for a number of insects including Plutella xylostella and Phaedon cochleariae (Robert and Riba, 1989; Amiri et al., 1999). DTX may, however, have a direct effect on growth also (Brousseau et al., 1996; Amiri et al., 1999). DTXs exhibit toxicity towards vertebrates. The injected intraperitoneal toxicity against mice is 1–1.35 mg kg1 for DTX A and 13.2–16.9 mg kg1 for DTX B (Kodaira, 1961) though effects

290

A. K. CHARNLEY

on fish and amphibians maybe less marked (Debeaupuis and Lafont, 1985; Genthner and Middaugh, 1992; Genthner et al., 1998). The first critical experiments on the effects of DTXs on lepidopteran muscle indicated that paralysis was due to calcium-dependent depolarisation of the muscle membrane (Samuels et al., 1988c; Bradfisch and Harmer, 1990). Both studies established that DTX specifically and reversibly gated calcium channels in the plasma membrane. Destruxin-induced degranulation of crayfish (Pacifastacus leniuseulus) haemocytes was also calcium dependent, being abolished in calcium-free conditions and by cadmium ions (Cerenius et al., 1990). Dumas et al. (1996a) found that DTXs induce calcium influx and phosphorylation of intracellular proteins within lepidopteran cell lines. The effects on intracellular calcium appeared secondary and may be brought about by the binding of DTX to an unidentified receptor (Dumas et al., 1994). Ultrastructural changes in Malpighian tubules and midgut epithelial cells in vitro caused by DTE E were Ca2þ dependent viz. the effects were absent in calcium-free media and in the presence of CdCl2, a calcium channel blocker (Dumas et al., 1996a). Although experiments by Samuels et al. (1988c) suggested that DTX did not exhibit ionophoric activity, structurally related molecules, the enniatins and beauvericin, move ions across membranes (Steinrauf, 1985) and Hinaje et al. (2002) demonstrated that DTX A can move calcium across liposomal membrane barriers. A direct involvement for secondary messengers appears to be ruled out for the inhibition of ecdysone secretion by lepidopteran prothoracic glands (Sloman and Reynolds, 1993) and fluid secretion by locust Malpighian tubules (James et al., 1993) because in both cases DTX inhibits responses to treatments elevating both cAMP and calcium. It is interesting that DTX B has been shown to be a specific, dose-dependent and reversible inhibitor of vacuolar-type ATPase from the yeast Saccharomyces cerevisiae, (Muroi et al., 1994). This enzyme maintains acidic homeostasis in membranebound organelles in eukaryotic cells. Acidification of intracellular compartments, a pivotal event in many aspects of cell physiology, was also found to be blocked by DTX B. However, while DTX inhibited V-ATPase from mung bean and yeast (Muroi et al., 1994) it had little effect on V-ATPase from barley and critically V-ATPase of brush border membrane vesicles from midgut columnar cells of G. mellonella (Bandani et al., 2001). In comparative studies DTX E often proves to be the most toxic of the DTX family probably because of its reactive epoxide group (Dumas et al., 1994). A number of studies have synthesised analogues of DTX to investigate structure–activity relationship. D-Lac-destruxin E, with lactic acid in place of the hydroxy acid in DTX E, has lower activity than the

FUNGAL PATHOGENS OF INSECTS

291

parent molecule. Synthetic analogues in which the ester bond is replaced by an amide bond are not active, showing the importance of the depsipeptide bond. Substituting the hydroxy acid with an alcohol or carboxy acid on side chain R resulted in a relatively inactive compund. Opening up the epoxide sidechain in DTX E to form a diol (which occurs in vivo during detoxification) reduces activity. None of the analogues were better than the parent (Cavelier et al., 1996, 1997, 1998). Also in a sequence of hemisynthetic analogues, none was as good as the parent (Dumas et al., 1994). The naturally occurring polar DTX D and DTX Ed are less active than DTX A or DTX E. Thus the epoxy group or double bond confer a high toxicity; a radical with free COOH is weaker. The ability of insects to detoxify DTX may have a bearing on host specificity of isolates. Removal of DTX from the haemolymph occurs much more quickly in vivo than in vitro suggesting that detoxification occurs primarily outside the haemolymph (Cherton et al., 1991). The half-life of DTX in G. mellonella is 1 h. This corresponds to the timescale of recovery from paralysis from a comparable dose (Jegorov et al., 1992). Injected 3H destruxin was found in cuticle, haemolymph, midgut and hindgut; with high concentration in the middle two (Jegorov et al., 1992). This is consistent with ultrastructural observations that cytotoxicity is greatest in midgut and haemocytes (Vey and Quiot, 1989b). DTX E is detoxified by hydrolysis and conjugation with glutathione in the migratory locust, Locusta migratoria (Loutelier et al., 1994); the fat body in particular is responsible. Opening up the epoxide sidechain in DTX E forms a diol which is then excreted through the Malpighian tubules (Cherton et al., 1991). Detoxification of DTX E occurs by similar mechanisms in G. mellonella (Hubert et al., 1999) and E-diol has little activity upon injection in this insect (Dumas et al., 1994). Detoxification of DTX A occurs by linearising the molecule (Lange et al., 1991, 1992). Interference with the cellular immune system seems the most likely impact of low doses of DTX during mycosis. DTXs reduced zymosan-induced nodule formation and activation of phenoloxidase in Periplaneta americana and Schistocerca gregaria (Huxham et al., 1989). Injected blastospores of M. anisopliae cause morphological alterations to plasmatocytes (a class of insect blood cell involved with phagocytic uptake of microorganisms) of G. mellonella. The cells from injected insects, when presented with particles that normally evoke phagocytosis, remained rounded, did not attach to glass surfaces, had surface blebs and failed to form actin filaments and filipodia. In vitro DTX A and DTX E had similar effects on the morphology and behaviour of plasmatocytes (Vilcinskas et al., 1997b). Additional features of DTX-treated cells viz swollen nuclei with clumped chromatin and blebbing are typical of cells undergoing programmed cell death (apoptosis). Thus the

292

A. K. CHARNLEY

toxin may be inducing apoptosis in haemocytes via the stimulation of key intracellular proteins. Consistent with this DTX E and concanamycin, another V-ATPase inhibitor, induced apoptosis in human tumor cells overexpressing epidermal growth factor receptor and stimulated by epidermal growth factor (Yoshimoto and Imoto, 2002). Cytochalasin D, also produced by M. anisopliae in vitro (see earlier), like DTX, inhibited attachment. However, in contrast to DTX, spreading was not affected and haemocytes had a somewhat different appearance. Thus whereas the effects of DTX on haemocytes mirrored that seen in larvae infected with M. anisopliae, the same was not true of cytochalasin. Suggesting that the latter has at best a minor role in pathogenesis, consistent with this view, cytochalasins have not been found yet in mycosed insects (Vilcinskas et al., 1997b). Hydrophobic DTXs are enriched on cell walls of growing fungi (Matha et al., 1992). If such DTXs are co-located with the equally hydrophobic cytochalasins on the surface of fungi they could work synergistically against the haemocytes (Vilcinskas et al., 1997b). This strategy could also work for other cyclic peptides such as beauverolides or cyclosporins that are probably too hydrophobic to be found free in quantity in the haemolymph. The retention of secondary metabolites to the fungal cell-wall would not only place the toxins in a prime spot to target immunocompetent haemocytes, but also minimise the amount needed for activity. Injecting silica beads coated with either beauverolide or cyclosporin into Galleria mellonella provoked nodule formation, synthesis of antimicrobial peptides, as well as attachment and spreading of plasmatocytes; but no phagocytosis. Direct injection caused a smaller effect, probably because the toxins were taken up by haemolymph proteins (lipophorins) and removed from circulation (in the fat body). Conversely in vitro the toxins inhibit phagocytosis of silica beads, yeast cell and blastospores of B. bassiana in a dose-dependent manner but did not affect haemocyte attachment or spreading though cytoskeleton of haemocytes was impeded (Vilcinskas et al., 1999). However, the effects of beauverolide and cyclosporin on the immune response are not seen in vivo during mycosis by B. bassiana, arguing against a role for these compounds in pathogenesis. In summary, of all the cyclic peptide toxins produced by entomopathogenic fungi the best evidence exists for an involvement of DTXs in pathogenesis of M. anisopliae. However, there is considerable variation in the quantity of DTX produced in vitro by isolates of M. anisopliae (Pais et al., 1981; Samuels et al., 1988a; Loutelier et al., 1994). Some isolates produce little or no DTX particularly those from M. anisopliae var. majus and M. anisopliae var. acridum (Kaijiang and Roberts, 1986;

FUNGAL PATHOGENS OF INSECTS

293

Kershaw et al., 1999). Production of destruxins is not dramatically affected by the composition of the culture medium (Roberts, 1966a; Jegorov et al., 1989). This suggests that DTX is produced constitutively and thus it is likely that in vitro studies are a true reflection of isolate capability and thus DTX production is not a prerequisite for pathogenicity. Identifying fungal metabolites in mycosed insects is the first step to establishing a role for the chemicals in pathogenesis. DTX B and desmethyl destruxin B were identified in B. mori infected with M. anisopliae by Suzuki et al. (1971) and quantified by the amount of -alanine present (assuming all the -alanine comes from DTX, though cuticle may also contain -alanine). The amount present in dead insects was sufficient to have caused mortality. Vey et al. (1986) also extracted biologically significant quantities of DTX from the haemolymph of Oryctes rhinoceros and B. mori using HPLC. Trace amounts of DTX A and B were found in G. mellonella infected with M. anisopliae var. anisopliae and var. majus but not var. flavoviride isolates (Amiri-Besheli et al., 2000). Samuels et al. (1988b) extracted DTX A from M. sexta infected with the virulent isolate 2575 but not from insects infected with two pathogenic but less virulent isolates. Onset of paralysis in diseased insects, similar to that seen in DTX-injected caterpillars, has suggested a direct link between DTXs and death from mycosis in some cases (Roberts, 1966b; Suzuki et al., 1971; Vey et al., 1986; Samuels et al., 1988b). Paralysis could interfere with haemolymph circulation, gaseous exchange and other vital functions. Given the variety of tissues susceptible to DTX the lethal lesion may not involve the muscles. Failure of an isolate to produce in vivo a sufficient quantity of DTX to induce muscular paralysis, therefore, does not preclude an involvement for DTX in pathogenesis. Samuels et al. (1988b) showed that among insects from 5 orders only Lepidoptera and adult Diptera were knocked down by low doses of DTX. This may reflect the sensitivity of the muscles from these different types of insects though other factors such as detoxification may also be important. If DTX has a role in pathogenesis of M. anisopliae then in most cases tissues other than muscles must be targeted. A positive correlation between virulence and destruxin production among hyper- or hypoproductive mutants would constitute good evidence for the involvement of DTX in mycosis. Indeed Al-Aidroos and Roberts (1978) found a spontaneous mutant of M. anisopliae, significantly more virulent for mosquito larvae than wild-type, that had enhanced in vitro destruxin production. However, the mutation probably involved several loci since the mutant also had more dense sporulation and more rapid in vitro germination than the wild-type. Amiri-Besheli et al. (2000) found that most virulent isolates produced large quantities of DTX but some low toxin producers were also pathogen.

294

A. K. CHARNLEY

Kershaw et al. (1999) also studied the relationship between DTX production and isolate virulence using three species of insect and isolates of Metarhizium. A significant negative correlation was found between the titre of DTX produced in vitro by isolates of M. anisopliae var. anisopliae pathogenic for the vine weevil Otiorhynchus sulcatus and median lethal time, suggesting a role for the toxin in isolate virulence. The same was true for isolates active against M. sexta. Growth form of the fungus in the haemolymph was clearly associated with virulence towards M. sexta. The more virulent isolates tended to grow more profusely as blastospores than hyphal fragments; the reverse was true for the weaker pathogens. The larger size of the hyphal fragments compared to blastospores suggested that there was greater growth of less virulent isolates in the haemolymph of insects prior to death. The appearance of infected insects was consistent with this interpretation. Five of the seven most virulent isolates all displayed flaccid paralysis prior to death. This symptom is also caused by injection of DTX (Samuels et al., 1988b,c). A key exception was isolate 703. This is highly virulent for M. sexta yet does not produce DTX in vitro, grows largely as hyphal fragments in the haemolymph of infected insects and does not cause host paralysis. A recent study, using the fungal-specific ergosterol as a measure of fungal biomass, revealed a much greater growth of 703 than 2575 (a virulent, high DTX producer) in mycosed Manduca prior to death (Graystone and Charnley, unpublished). Thus there are are two possible virulence strategies among isolates of M. anisopliae var. anisopliae pathogenic for Manduca viz the ‘toxin strategy’ and the ‘growth strategy’. For desert locusts, Schistocerca gregaria, a strong positive correlation was found only between in vitro toxin production and % mortality of individuals where sporulation did not occur on the cadaver. To account for this Kershaw et al. (1999) suggested that if DTX kills locusts before the fungus has established itself then the pathogen may not compete effectively with the saprophytic flora and as a result fails to sporulate. They concluded that, in the pathogenesis of M. anisopliae var. anisopliae for all three insects, there was a relationship between the titre of DTX production by isolates in vitro and killing power. The fact that little or no DTX is produced by certain specialist isolates of M. anisopliae e.g. M. anisopliae var. majus, pathogenic for scarabaeid beetles (Kaijiang and Roberts, 1986), M. anisopliae var. album, hemipteran specific (Amiri-Besheli et al., 2000), M. anisopliae var. acridum, pathogens of locusts and grasshoppers (Kershaw et al., 1999) is also consistent for a role for DTX in isolate virulence. Definitive evidence for the involvement of destruxin in virulence would come from the use of transformation-mediated gene disruption to produce a single lesion mutant of the ‘destruxin gene’. In common with other

FUNGAL PATHOGENS OF INSECTS

295

cyclic peptide antibiotics produced by fungi e.g. beauvericin, enniatins and cyclosporins, DTXs are probably synthesised non-ribosomally by a multifunctional peptide synthetase (for review, see Kleinkauf and von Dohren, 1990). These enzymes possess a multidomain structure and employ the thiotemplate mechanism to activate, modify and link together by amide or ester bonds the constituent amino acids (Stachelhaus and Marahiel, 1995). The domains (one for each amino acid residue) act as independent enzymes. The order of the domains defines the sequence of the eventual peptide. Type I domains handle hydroxy and amino acids, type II domains are bigger and have the additional function of methylating the residue they handle. The advantage of this in comparison to ribosomal synthesis is a 6-fold lower consumption of ATP per peptide bond, though substrate fidelity is sacrificed, as evidenced by the variety of molecules produced in the series. The destruxin synthetase gene has not yet been cloned, though recent work in our lab has resulted in the cloning and sequencing of a cyclic peptide synthetase (of unknown function) from M. anisopliae ( isolate 2575) and has provided evidence for the existence of others (Bailey et al., 1996; Bailey et al., unpublished). In an express sequence tag study of M. anisopliae isolates grown on insect cuticle, (Freimoser et al., 2003a) found several transcripts encoding enzymes (cyclic peptide synthetases, reductases) involved in the synthesis of toxic metabolites in the library of M. anisopliae sf. anisopliae 2575 and the absence of counterparts in the library of M. anisopliae sf. acridum. The former as has been noted above kills its hosts quickly apparently through the action of toxins, while the specific grasshopper/locust pathogen M. anisopliae sf. acridum invades all tissues of the host and the insect dies when it is filled with fungal biomass (Inglis et al., 2001). This provides further support of the ‘toxin’ and ‘growth’ strategies employed by isolates of Metarhizium spp. suggested by Kershaw et al. (1999) (see above). A number of fungal genes involved in the production of cyclic peptide toxins have been cloned and sequenced. They tend to be clustered, often being positioned with regulatory elements and autoresistance genes less than 2 kb from each other. Most fungal genes for other biosynthetic pathways follow the normal eukaryotic model of dispersion throughout the genome. Walton (2000) has argued that the grouping of genes in this way is an aid to retention during horizontal transmission; they are more likely to persist in an active form if they are moved together. Horizontal movement of fungal genes between isolates and species through hyphal anastamosis is important for the retention of the characteristic in the light of the relative inefficiency of vertical transmission due to the instability of fungal genomes. The alternative is the clustering of genes to aid co-regulation. However,

296

A. K. CHARNLEY

secondary metabolite genes are controlled by trans-acting transcription factors which can control the expression of dispersed or clustered genes. The horizontal movement hypothesis accounts for the presence of the toxin in some isolates of a species and not others. Some species have an arsenal of toxins e.g. B. bassiana has beauverolides, bassianolides, beauvericins and cyclosporins. The gene clusters coding for these attributes may have been acquired by horizontal transfer. This would account for the occurrence of the same molecule in a number of different fungal species e.g. destruxin in the insect pathogens M. anisopliae and Aschersonia sp. and in three unrelated plant pathogenic fungi, Alternaria brassicae (Bains and Tewari, 1987) , Trichothecium roseum (Springer et al., 1984) and Ophiosphaerella herpotricha (Venkatsubbaiah et al., 1994). What we don’t know yet is whether the cocktail of toxins acquired by a particular species and/or isolate influences host range.

E. PROTEASES AND OTHER ENZYMES AS TOXINS

Although subtilisins and metalloproteases from M. anisopliae and other entomopathogenic Deuteromycotina are particularly active against insect cuticle their broad specificity could make them effective weapons during later stages of mycosis. Kucera (1980, 1981) was the first to demonstrate the insecticidal effects of M. anisopliae proteases (partially purified serine and cystiene enzymes) upon injection into wax moth larvae. More recently several authors (Vilcinskas et al., 1997b,c; Vilcinskas and Wedde, 1997; Griesch and Vilcinskas, 1998; Griesch et al., 2000) have shown that in vitro proteases from B. bassiana and M. anisopliae impaired attachment, spreading, cytoskeleton formation and inhibited phagocytosis of plasmatocytes from G. mellonella. Metalloproteases were more effective than serine enzymes (trypsin and chymotrypsin). Similar symptoms were seen also in vivo during mycosis (Vilcinskas et al., 1997b) The role of proteases in the toxic activities of pathogens and predators is well established outside of the fungi e.g. proteases produced by Pseudomonas aeruginosa play an important part in pathogenesis (Lysenko and Kucera, 1971) and snake venoms comprise proteases and other enzymes (Zeller, 1977). However, the problem with seeking a role for fungal proteases in pathogenesis outside of cuticle invasion is that there have been few reports of significant activity in haemolymph and other tissues during the pathogenic phase. A number of studies have established the importance of catabolite repression in the regulation of fungal proteases (St. Leger et al., 1988c; Paterson et al., 1994b). Thus the low

FUNGAL PATHOGENS OF INSECTS

297

Pr1 activity observed in haemolymph during later stages of infection of M. sexta with M. anisopliae (St. Leger et al., 1996c) could be due to the repressive effects of amino acids and other low molecular weight metabolites. That this may not be true for all isolates of all species of entomopathogenic fungi is shown by the fact that Shimizu et al. (1993a,b) detected elastase-type (‘subtilisin’) B. bassiana protease serologically in the haemolymph of mycosed B. mori larvae. However, activity was low, they suggested, because of host protease inhibitors. We have found recently that regulation of Pr1 enzymes can vary between isolates and isoforms within isolates of V. lecanii. Thus while Pr1s from KV42 are not subject to C or N repression in vitro, KV71 are. As a consequence in vivo during pathogenesis of the peach potato aphid Myzus persicae, significant Pr1 activity appeared early in aphids infected with KV42, but not KV71 (Bye and Charnley, unpublished). Furthermore immunochemical staining of whole body sections of infected aphids using specific antibodies showed the widespread presence of Pr1 in cuticle and peripheral fatbody of insects in the early stages (2 days) of mycosis with KV42, but little in aphids infected with KV71 (see Fig. 5) (Bye and Charnley, unpublished). The subtilisins of C. coronatus are less subject to catabolite repression (Freimoser et al., 2003b). Thus they could be produced in the haemolymph during mycosis. Consistent with this the blackening of the haemolymph in mycosed insects infected with this fungus is reminiscent of the activation of host prophenoloxidase by Metarhizium Pr1 subtilisin experimentally and in vivo by hyperproductive, constituent mutants (St. Leger et al., 1996c). The presence of an array of protease inhibitors in insect haemolymph with activity against pathogenic enzymes implies an essential function and thus the potential importance of fungal enzymes in disease development. M. sexta has 15 different plasma serpins (Kanost, 1999). One of these has 12 alternatively spliced versions of exon 9, each encoding a different reactive site with a different proteinase targeting. Interestingly Vilcinskas and Wedde (1997) showed a protease inhibitor in the haemolymph of G. mellonella induced by the injection of B. bassiana spores. This inhibitor inhibited pathogen proteases and germination in vitro. Fungal invasion through the cuticle following topical application (a ‘natural infection’) did not elicit production of the inhibitor. Thus fungal proteases may have a part to play in pathogenicity other than during host invasion through the cuticle and destruction of the cadaver post mortem. The effects of a hyperproductive Pr1 mutant under the direction of a constitutive promoter suggest reasons why fungi have not generally evolved to release highly active proteases during the haemolymph growth phase. St. Leger et al. (1996c) found the extracellular enzyme produce

298

A. K. CHARNLEY

during mycosis caused in M. sexta by these mutants caused widespread activation of the phenoloxidase cascade and the production of large amounts of melanin. This contributed to early death of the insect and a large significant reduction in spore production by the fungus on the cadaver. Phospholipases are important cytotoxins produced by insect and mammalian bacterial pathogens (Lysenko, 1985). While M. anisopliae and other entomopathogenic Deuteromycotina are known to secrete phospholipases in vitro and transcripts of the corresponding genes have been found in EST libraries of isolates grown on cuticle (Freimoser et al., 2003a), the importance of these enzymes in mycosis is yet to be established. Undoubtedly the increasing application of molecular techniques to entomopathogenic fungi will reveal other novel specific toxins, some of which may even be host-specific as has been found with phytopathogenic fungi. Recent research has shown that Nomuraea rileyi secretes a specific enzyme that oxidises the hydroxyl group at position C22 of haemolymph ecdysteroids and prevents moulting in larvae of the Japanese silkmoth Bombyx mori (Kiuchi et al., 2003). If this mirrors what happens in mycosis then posssibly N. rileyi in common with insect nuclearpolyhedrosis viruses has evolved a way to interfere with moulting by which it keeps its host feeding and thus in an optimum nutritional state. Interestingly DTX from M. anisopliae inhibits production of ecdysone by isolated prothoracic glands from M. sexta in vitro (Sloman and Reynolds, 1993). This could account for instances where development is disrupted in caterpillars infected with M. anisopliae (Sloman and Reynolds, 1993).

F. MULTIPLE TOXINS – SYNERGY OR SPECIFICITY?

There appear to have been few experimental demonstrations of synergy between toxic insecticidal secondary metabolites produced by entomopathogenic fungi. In one of the few Dowd (1988) showed enhanced toxicity of combined application of two co-occuring fungal metabolites, aflatoxin B- and kojic acid produced by A. flavus, to two species of lepidopteran caterpillar. Hydrophobic DTXs are enriched on cell walls of growing fungi (Matha et al., 1992). Also, as stated earlier, if such DTXs are co-located with the equally hydrophobic cytochalasins on the surface of fungi they could work synergistically against the haemocytes (Vilcinskas et al., 1997b). Yoder and Turgeon (2001) have suggested that pathogens carry more genes involved in the production of secondary metabolites, particularly non-ribosomal multifunctional peptide synthetases (NRPSs) and polyketide

FUNGAL PATHOGENS OF INSECTS

299

synthetases, than saprophytes. Their survey of the literature is enriched for phytopathogens. But it is clear from the present review that Deuteromycete entomopathogens in common with their plant pathogenic cousins produce a range of cyclic peptide toxins that derive from NRPSs. As has been noted several times in this review useful analogies can be made between fungal pathogenesis in insects and plants. Good evidence exists for the role of NRPSs as virulence factors in phytopathogens (Herrmann et al., 1996; Walton, 1996; Johnson et al., 2000). The abundance of NRPS genes in these plant pathogens (Yoder and Turgeon, 2001), appears to be matched by the situation in M. anisopliae. We have cloned and sequenced an NRPS from M. anisopliae (PES1) comprising a 23733 bp ORF encoding a protein of 5158 amino acids (Bailey, Reynolds, Charnley and Clarkson, unpublished). Analysis shows that this polypeptide contains only four domains with no evidence for amino acid methylation motifs. The inferred amino acid sequence shows similarity to those of other peptide synthetases. The conserved domain region contains sequences similar to those of the six core motifs and this domain shows most similarity (36% identity) to the second domain of the gene encoding the HC tetrapeptide toxin from the plant pathogenic fungus Cochliobolus carbonum. DTX contains six residues and we predict that the gene encoding DTX synthetase will contain six domains and include methylation motifs associated with two of the domains. The sequenced gene is therefore unlikely to encode DTX synthetase. Southern analysis of a wide range of isolates of M. anisopliae demonstrated that all contained DNA that hybridised strongly to a probe based on PES1. No clear difference in pattern or intensity of hybridisation was seen between DTX producers and non-producers. This observation adds weight to the contention that the sequenced gene encodes a peptide synthetase other than DTX synthetase. No Northern analysis with this clone was carried out. However, an RT-PCR demonstrated that this gene is expressed in M. anisopliae (2575). Through a PCR strategy using primers based on those described by Nikolskaya et al. (1995) we found sequences consistent with at least 17 peptide synthetase domains including the four in PES1, some of which include methylation motifs. Thus, M. anisopliae may have multiple NRPS genes that synthesise several families of cyclic peptide.

VII. THE FUTURE The 21st century is an exciting time to be involved in research on microbial pathogenesis of eukaryotes. The availability of the sequences of whole

300

A. K. CHARNLEY

genomes from an increasing number of organisms is giving realistic expectation of the resolution of long-standing far-reaching questions on the nature of pathogenesis. Where do pathogens acquire pathogenicity genes from? There is evidence from human pathogenic bacteria for horizontal transfer of chromosomal genes (‘pathogenicity islands’) and plasmids. A similar phenomenon may occur in phytopathogens (Rosewich and Kistler, 2000). Do pathogens have unique pathogenicity genes or are ‘housekeeping’ genes co-opted (Yoder and Turgeon, 2001)? Evidence for phytopathogenic fungi suggests both situations prevail. It is still too early to expect answers to such searching questions for entomopathogenic fungi, though there are some tantalising pieces of evidence. CHY1 a chymotrypsin produced by M. anisopliae belongs to the S2 group of chymotrypsins. It is most closely related to protease C from the actinomycete Streptomyces griseus. Both have only 15% identity to mammalian S1 chymotrypsin. Plant- and insect-pathogenic Ascomycotina and actinomycete bacteria are the only known microbial sources of S1 trypsins. Furthermore, S1 and S2 homologues appear to be absent from the ascomycete yeasts S. cerevisiae and the saprophytic filamentous ascomycetes N. crassa and A. nidulans. So M. anisopliae may have acquired CHY1 by lateral gene transfer from an actinomycete. Such a process would be facilitated by the sharing, in soil, of a common habitat (Screen and St. Leger, 2000). This study was an outcome of the production of expressed sequence tag libraries (EST) from cuticle-grown cultures of two isolates of M. anisopliae (Freimoser et al., 2003a). Several other ESTs were found with closer homology to bacterial than fungal genes, again consistent with gene transfer.

ACKNOWLEDGEMENTS I would like to thank R. J. St. Leger and A. Vey for helpful comments on a draft of the manuscript, R. J. St. Leger for providing preprints of unpublished papers, S. Fairhurst for drawing the super cartoons and D. Steinkraus, H. Evans and C. Prior for generously allowing me use of some of their photographs. I am grateful also to the BBSRC and the EC for funding.

REFERENCES Al-Aidroos, K. and Roberts, D. W. (1978). Mutants of Metarhizium anisopliae with increased virulence toward mosquito larvae. Canadian Journal of Genetics and Cytology 20, 211–220. Aldridge, D. C. and Turner, W. B. (1969). Structures of Cytochalasin C and Cytochalasin D from Metarhizium anisopliae. Journal of the Chemical Society Section C Organic Chemistry, 923–928.

FUNGAL PATHOGENS OF INSECTS

301

Amiri, B., Ibrahim, L. and Butt, T. M. (1999). Antifeedant properties of destruxins and their potential use with the entomogenous fungus Metarhizium anisopliae for improved control of crucifer pests. Biocontrol Science and Technology 9, 487–498. Amiri-Besheli, B., Khambay, B., Cameron, S., Deadman, M. L. and Butt, T. M. (2000). Inter- and intra-specific variation in destruxin production by insect pathogenic Metarhizium spp., and its significance to pathogenesis. Mycological Research 104, 447–452. Anke, H. and Sterner, O. (2002). Insecticidal and nematicidal metabolites from fungi. In ‘The Mycota volume X – Industrial applications’ (H.D. Osiewacz, ed.), pp. 109–127, Springer, Heidelberg. Askary, H., Benhamou, N. and Brodeur, J. (1999). Ultrastructural and cytochemical characterization of aphid invasion by the hyphomycete Verticillium lecanii. Journal of Invertebrate Pathology 74, 1–13. Bagga, S., Screen, S. E. and St. Leger, R. J. (2003). Reconstructing the diversification of subtilisins in the pathogenic fungus Metarhizium anisopliae. Gene (In press). Bailey, A. M., Kershaw, M. J., Hunt, B. A., Paterson, I. C., Charnley, A. K., Reynolds, S. E. and Clarkson, J. M. (1996). Cloning and sequence analysis of an intron-containing domain from a peptide synthetaseencoding gene of the entomopathogenic fungus Metarhizium anisopliae. Gene 173, 195–197. Bains, P. S. and Tewari, J. P. (1987). Purification, chemical characterisation and host-specificity of the toxin produced by Alternaria brassicae. Physiological Molecular Plant Pathology 30, 259–271. Bandani, A. R., Amiri, B., Butt, T. M. and Gordon-Weeks, R. (2001). Effects of efrapeptin and destruxin, metabolites of entomogenous fungi, on the hydrolytic activity of a vacuolar type ATPase identified on the brush border membrane vesicles of Galleria mellonella midgut and on plant membrane bound hydrolytic enzymes. Biochimica Et Biophysica Acta–Biomembranes 1510, 367–377. Bandani, A. R., Khambay, B. P. S., Faull, J. L., Newton, R., Deadman, M. and Butt, T. M. (2000). Production of efrapeptins by Tolypocladium species and evaluation of their insecticidal and antimicrobial properties. Mycological Research 104, 537–544. Bassi, A. (1835). Del mal del segno. Reproduced in ‘Phytopathological Classics’ (G. C. Ainsworth and P. J. Yarrow, eds.), American Phytopathological Society. Beard, R. L. and Walton, G. S. (1969). Kojic acid as an insecticidal mycotoxin. Journal of Invertebrate Pathology 14, 53–59. Berbee, M. L. and Taylor, J. W. (2001). Fungal molecular evolution: gene trees and geological time. In ‘The Mycota Vol 7A – Systematics and Evolution’ (D. J. McLaughlin, E. G. McLaughlin and P. A. Lemke, eds.), pp. 229–245, Springer, Berlin. Bidochka, M. J. and Khachatourians, G. G. (1987). Purification and properties of an extracellular protease produced by the entomopathogenic fungus Beauveria bassiana. Applied and Environmental Microbiology 53, 1679–1684. Bidochka, M. J. and Khachatourians, G. G. (1988a). N-acetyl-D-glucosaminemediated regulation of extracellular protease in the entomopathogenic fungus Beauveria bassiana. Applied and Environmental Microbiology 54, 2699–2704.

302

A. K. CHARNLEY

Bidochka, M. J. and Khachatourians, G. G. (1988b). Regulation of extracellular protease in the entomopathogenic fungus Beauveria bassiana. Experimental Mycology 12, 161–168. Bidochka, M. J. and Khachatourians, G. G. (1990). Identification of Beauveria bassiana extracellular protease as a virulence factor in pathogenicity toward the migratory grasshopper, Melanoplus sanguinipes. Journal of Invertebrate Pathology 56, 362–370. Bidochka, M. J. and Khachatourians, G. G. (1991). The implication of metabolic acids produced by Beauveria bassiana in pathogenesis of the migratory grasshopper, Melanoplus sanguinipes. Journal of Invertebrate Pathology 58, 106–117. Bidochka, M. J. and Khachatourians, G. G. (1993a). Oxalic-acid hyperproduction in Beauveria bassiana mutants is related to a utilizable carbon source but not to virulence. Journal of Invertebrate Pathology 62, 53–57. Bidochka, M. J. and Khachatourians, G. G. (1993b). Regulation of extracellular N-acetyl-D-glucosaminidase production in the entomopathogenic fungus Beauveria bassiana. Canadian Journal of Microbiology 39, 6–12. Bidochka, M. J. and Khachatourians, G. G. (1994). Protein hydrolysis in grasshopper cuticles by entomopathogenic fungal extracellular proteases. Journal of Invertebrate Pathology 63, 7–13. Bidochka, M. J. and Melzer, M. J. (2000). Genetic polymorphisms in three subtilisinlike protease isoforms (Pr1A, Pr1B, and Pr1C) from Metarhizium strains. Canadian Journal of Microbiology 46, 1138–1144. Bidochka, M. J., Tong, K. I. and Khachatourians, G. G. (1993). Partial purification and characterization of two extracellular N-acetyl-D-glucosaminidases produced by the entomopathogenic fungus, Beauveria bassiana. Canadian Journal of Microbiology 39, 40–45. Bidochka, M. J., Burke, S. and Ng, L. (1999). Extracellular hydrolytic enzymes in the fungal genus Verticillium: adaptations for pathogenesis. Canadian Journal of Microbiology 45, 856–864. Bing, L. A. and Lewis, L. C. (1991). Suppression of Ostrinia nubilalis (Hubner) (Lepidoptera, Pyralidae) by Endophytic Beauveria bassiana (Balsamo) Vuillemin. Environmental Entomology 20, 1207–1211. Blomquist, G. J. (1984). Cuticular lipids of insects. In Conference Report (D. W. Roberts and J. R. Aist, eds.), pp. 54-60: Rockerfeller Foundation. Bogo, M. R., Rota, C. A., Pinto, H., Ocampos, M., Correa, C. T., Vainstein, M. H. and Schrank, A. (1998). A chitinase encoding gene (chit1 gene) from the entomopathogen Metarhizium anisopliae: Isolation and characterization of genomic and full-length cDNA. Current Microbiology 37, 221–225. Bogus, M. I. and Scheller, K. (2002). Extraction of an insecticidal protein fraction from the parasitic fungus Conidiobolus coronatus (Entomophthorales). Acta Parasitologica 47, 66–72. Bogus, M. I. and Szczepanik, M. (2000). Histopathology of Conidiobolus coronatus (Entomophthorales) infection in Galleria mellonella (Lepidoptera) larvae. Acta Parasitologica 45, 48–54. Boucias, D. G. and Pendland, J. C. (1991). Attachment of mycopathogens to cuticle: the initial event of mycosis in arthropod hosts. In ‘The Fungal Spore and Disease Initiation in Plants and Animals.’ (G. T. Cole and H. C. Hoch, eds.), pp. 101–128. Plenum Press, New York. Boucias, D. G., Farmerie, W. G. and Pendland, J. C. (1998). Cloning and sequencing of cDNA of the insecticidal toxin Hirsutellin A. Journal of Invertebrate Pathology 72, 258–261.

FUNGAL PATHOGENS OF INSECTS

303

Bradfisch, G. A. and Harmer, S. L. (1990). Omega-Conotoxin Gvia and Nifedipine inhibit the depolarizing action of the fungal metabolite, destruxin-B on muscle from the tobacco budworm (Heliothis virescens). Toxicon 28, 1249–1254. Brewer, D., Jen, W. C., Jones, G. A. and Taylor, A. (1984). The antibacterial activity of some naturally occuring 2,5-dihydroxy-1,4-benzoquinones. Canadian Journal of Microbiology 30, 1068–1072. Brey, P. T., Latge, J. P. and Prevost, M. C. (1986). Integumental penetration of the Pea Aphid, Acyrthosiphon pisum, by Conidiobolus obscurus (Entomophthoraceae). Journal of Invertebrate Pathology 48, 34–41. Brobyn, P. J. and Wilding, N. (1983). Invasive and developmental processes of Entomophthora muscae infecting houseflies. Transactions of the British Mycological Society 81, 72–74. Brockerhoff, H. and Jensen, R. G. (1974). ‘Lipolytic Enzymes’. Academic Press, New York. Brousseau, C., Charpentier, G. and Belloncik, S. (1996). Susceptibility of spruce budworm, Choristoneura fumiferana Clemens, to destruxins, cyclodepsipeptidic mycotoxins of Metarhizium anisopliae. Journal of Invertebrate Pathology 68, 180–182. Buchwaldt, L. and Jensen, J. S. (1991). Hplc purification of destruxins produced by Alternaria brassicae in culture and leaves of Brassica napus. Phytochemistry 30, 2311–2316. Butt, T. M. (1987). A fluorescence microscopy method for the rapid localization of fungal spores and penetration sites on insect cuticle. Journal of Invertebrate Pathology 50, 72–74. Carruthers, R. I. and Soper, R. S. (1987). Fungal diseases. In ‘Epizootology of Insects Diseases.’ (J. R. Fuxa and Y. Tanada, eds.), pp. 357–416. Wiley and Sons, New York. Cavelier, F., Jacquier, R., Mercadier, J. L. and Verducci, J. (1996). Destruxin analogues: Depsi peptidic bond replacement by amide bond. Tetrahedron 52, 6173–6186. Cavelier, F., Jacquier, R., Mercadier, J. L., Verducci, J., Traris, M. and Vey, A. (1997). Destruxin analogs: variations of the alpha-hydroxy acid side chain. Journal of Peptide Research 50, 94–101. Cavelier, F., Verducci, J., Andre, F., Haraux, F., Sigalat, C., Traris, M. and Vey, A. (1998). Natural cyclopeptides as leads for novel pesticides: Tentoxin and destruxin. Pesticide Science 52, 81–89. Cerenius, L., Thornquist, P.-O., Vey, A., Johansson, M.W. and Soderhall, K. (1990). The effect of the fungal toxin destruxin E on isolated crayfish haemocytes. J. Insect. Physiol. 36, 785–789. Champlin, F. R. and Grula, E. A. (1979). Noninvolvement of Beauvericin in the entomopathogenicity of Beauveria bassiana. Journal of Invertebrate Pathology 37, 1122–1125. Charnley, A. K. (1984). Physiological aspects of destructive pathogenesis in insects by fungi: a speculative review. In ‘Invertebrate–Microbial Interactions’ (J. M. Anderson, A. D. M. Rayner and D. W. H. Walton, eds.), pp. 229–276. Cambridge University Press, Cambridge. Charnley, A. K. (1989). Mechanisms of fungal pathogenesis in insects. In ‘The Biotechnology of Fungi for Improving Plant Growth’ (J. M. Whipps and R. D. Lumsden, eds.), pp. 85–127. Cambridge University Press, Cambridge. Charnley, A. K. (1997). Entomopathogenic fungi and their role in pest control. In ‘The Mycota Vol IV, Environmental and Microbial Relationships’

304

A. K. CHARNLEY

(D. Wicklow and M. Soderstrom, eds.), pp. 185–201. Springer-Verlag, Heidelberg. Charnley, A. K. and St. Leger, R. J. (1991). The role of cuticle-degrading enzymes in fungal pathogenesis of insects. In ‘The Fungal Spore and Disease Initiation in Plants and Animals’ (G. T. Cole and H. C. Hoch, eds.), pp. 129–156, Plenum press, London. Chen, H. C., Yeh, S. F., Ong, G. T., Wu, S. H., Sun, C. M. and Chou, C. K. (1995). The Novel Desmethyldestruxin B–2 from Metarhizium anisopliae, that suppresses hepatitis-B virus surface-antigen production in human hepatoma-cells. Journal of Natural Products-Lloydia 58, 527–531. Cherton, J. C., Lange, C., Mulheim, C., Pais, M., Cassier, P. and Vey, A. (1991). Direct in vitro and in vivo monitoring of destruxins metabolism in insects using internal surface reversed-phase high- performance liquid-chromatography. 1. Behavior of E-destruxin in locusts. Journal of ChromatographyBiomedical Applications 566, 511–524. Chrzanowska, J., Banas, J. and Kolaczkowska, M. (2001). Purification and characterization of Beauveria bassiana proteinases. Acta Biotechnologica 21, 73–81. Clarkson, J. M. and Charnley, A. K. (1996). New insights into the mechanisms of fungal pathogenesis in insects. Trends in Microbiology 4, 197–203. Clay, K. (1988). The effect of fungi on the interaction between host plants and their herbivores. Canadian Journal of Plant Pathology 9, 380–388. Claydon, N. (1978). Insecticidal secondary metabolites from entomogenous fungi: Entomophthora virulenta. Journal of Invertebrate Pathology 32, 319–324. Claydon, N. and Grove, J. F. (1982). Insecticidal secondary metabolic products from the entomogenous fungus Verticillium lecanii. Journal of Invertebrate Pathology 40, 413–418. Claydon, N., Grove, J. F. and Pople, M. (1977). Insecticidal secondary metabolic products from the entomogenous fungus Fusarium solani. Journal of Invertebrate Pathology 30, 216–223. Cole, S. C. J., Charnley, A. K. and Cooper, R. M. (1993). Purification and partial characterization of a novel trypsin-like cysteine protease from Metarhizium anisopliae. FEMS Microbiology Letters 113, 189–196. Cooper, R. M. and Wood, R. K. S. (1975). Regulation of synthesis of cell-wall degrading enzymes by Verticillium lecanii and Fusarium oxysporum f. sp. lycopersici. Physiological Plant Pathology 5, 135–156. Cooper, R. M., Rankin, B. and Wood, R. K. S. (1978). Cell-wall degrading enzymes of vascular wilt fungi II. Properties and modes of action of polysaccharidases of Verticillium albo-atrum and Fusarium oxysporum f. sp. lycopersici. Physiological Plant Pathology 13, 101–134. Coudron, T. A., Kroha, M. J. and Ignoffo, C. M. (1984). Levels of chitinolytic activity during development of 3 entomopathogenic fungi. Comparative Biochemistry and Physiology B 79, 339–348. Cox, D. L. and Willis, J. H. (1985). The cuticular proteins of Hyalophora cecropia from different anatomical regions and metamorphic stages. Insect Biochemistry 15, 349–362. Cross, R. L. and Kohlbrenner, W. E. (1978). The mode of inhibition of oxidative phosphorylation by efrapeptin A-23871 evidence for an alternating site mechanism for ATP synthesis. Journal of Biological Chemistry 253, 4865–4873. Dasilva, J. C., Desiqueira, J. P. and Marcondes, C. B. (1989). Selection of Metarhizium anisopliae for extracellular enzyme-production and virulence toward Triatoma infestans. Revista Brasileira De Genetica 12, 1–7.

FUNGAL PATHOGENS OF INSECTS

305

Debeaupuis, J. P. and Lafont, P. (1985). Effet de quelques mycotoxines et substances genotoxiques de synthese sur le developpement de l’embryon et de la Brachydanio rerio. Comptes Rendus De L Academie Des Sciences Serie 3 Sciences De La Vie-Life Sciences 300, 167–170. Dowd, P. F. (1988). Synergism of aflatoxin B-1 toxicity with the co-occuring fungal metabolite kojic acid to two caterpillars. Entomologia Experimentalis et Applicata 47, 69–72. Dowd, P. F. (1999). Relative inhibition of insect phenoloxidase by cyclic fungal metabolites from insect and plant pathogens. Natural Toxins 7, 337–341. Drummond, J. and Pinnock, D. E. (1990). Aflatoxin production by entomopathogenic isolates of Aspergillus parasiticus and Aspergillus flavus. Journal of Invertebrate Pathology 55, 332–336. Dumas, C., Robert, P., Pais, M., Vey, A. and Quiot, J. M. (1994). Insecticidal and cytotoxic effects of natural and hemisynthetic destruxins. Comparative Biochemistry and Physiology C-Pharmacology Toxicology & Endocrinology 108, 195–203. Dumas, C., Matha, V., Quiot, J. M. and Vey, A. (1996a). Effects of destruxins, cyclic depsipeptide mycotoxins, on calcium balance and phosphorylation of intracellular proteins in lepidopteran cell lines. Comparative Biochemistry and Physiology C-Pharmacology Toxicology & Endocrinology 114, 213–219. Dumas, C., Ravallec, M., Matha, V. and Vey, A. (1996b). Comparative study of the cytological aspects of the mode of action of destruxins and other peptidic fungal metabolites on target epithelial cells. Journal of Invertebrate Pathology 67, 137–146. Dunphy, G. B. and Nolan, R. A. (1982). Mycotoxin production by the protoplast stage of Entomophthora egressa. Journal of Invertebrate Pathology 39, 261–263. Elsayed, G. N., Coudron, T. A., Ignoffo, C. M. and Riba, G. (1989). Chitinolytic activity and virulence associated with native and mutant isolates of an entomopathogenic fungus, Nomuraea rileyi. Journal of Invertebrate Pathology 54, 394–403. El-Sayed, G. N., Ignoffo, C. M., Leathers, T. D. and Gupta, S. C. (1993). Cuticular and non-cuticular substrate influence on expression of cuticle-degrading enzymes from conidia of an entomopathogenic fungus Nomuraea rileyi. Mycopathologia 122, 79–87. Engstrom, G., Delance, J., Richard, A. L. and Baetz, J. (1975). Purification and characterisation of roseotoxin B, a toxic cyclodepsipeptide from Trichothecium roseum. Journal of Agriculture and Food Chemistry 23, 244–253. Evans, H. C. (1988). Coevolution of entomogenous fungi and their insect hosts. In ‘Coevolution of fungi with plants and animals’ (K. A. Pirozynski and D. L. Hawksworth, eds.). pp. 149–171. Academic Press, London. Evans, H. C. (1989). Mycopathogens of insects of epigeal and aerial habitats. In ‘Insect Fungus Interactions’, (N. Wilding, N. M. Collins, P. M. Hammond and J. F. Weber, eds.), pp. 205–238. Academic Press, London. Eyal, J., Mabud, A., Fischbein, K. L., Walter, J. F., Osborne, L. S. and Landa, Z. (1994). Assessment of Beauveria bassiana Nov. EO-1 strain, which produces a red pigment for microbial control. Applied Biochemistry and Biotechnology 44, 65–80.

306

A. K. CHARNLEY

Frappier, F., Ferron, P. and Pais, M. (1975). Chemistry of entomopathogenic fungi: beauvellide, a new cyclodepsipeptide isolated from Beauveria tenella Phytochemistry 14, 2703–2705. Freimoser, F. M., Screen, S., Bagga, S., Hu, G. and St. Leger, R. J. (2003a). Expressed sequence tag (EST) analysis of two subspecies of Metarhizium anisopliae reveals a plethora of secreted proteins with potential activity in insect hosts. Microbiology 149, 1–9. Freimoser, F. M., Screen, S. E., Hu, G. and St. Leger, R. J. (2003b). EST analysis of genes expressed by the zygomycete pathogen Conidiobolus coronatus during growth on insect cuticle. Microbiology 149, 1893–1900. Fricaud, A. C., Walters, A. J., Whitehouse, D. G. and Moore, A. L. (1992). The roles of adenylate kinase and the adenylate carrier in the regulation of plant mitochondrial respiratory activity. Biochimica et Biophysica Acta 1099, 253–261. Genthner, F. J. and Middaugh, D. P. (1992). Effects of Beauveria bassiana on Embryos of the Inland Silverside Fish (Menidia beryllina). Applied and Environmental Microbiology 58, 2840–2845. Genthner, F. J., Chancy, C. A., Couch, J. A., Foss, S. S., Middaugh, D. P., George, S. E., Warren, M. A. and Bantle, J. A. (1998). Toxicity and pathogenicity testing of the insect pest control fungus Metarhizium anisopliae. Archives of Environmental Contamination and Toxicology 35, 317–324. Geremia, R. A., Goldman, G. H., Jacobs, D., Ardiles, W., Vila, S. B., Vanmontagu, M. and Herreraestrella, A. (1993). Molecular characterization of the proteinase-encoding gene, Prb1, related to mycoparasitism by Trichoderma harzianum. Molecular Microbiology 8, 603–613. Gillespie, A. T. (1988). Use of fungi to control pests of agricultural importance. In ‘Fungi in Biological Control Systems.’ (M. N. Burge, ed.), pp. 37–60. Manchester University Press, Manchester. Gillespie, A. T. and Claydon, N. (1989). The use of entomogenous fungi for pest control and the role of toxins in pathogenesis. Pesticide Science 27, 203–215. Gillespie, J. P., Bateman, R. and Charnley, A. K. (1998). Role of cuticledegrading proteases in the virulence of Metarhizium spp. for the desert locust, Schistocerca gregaria. Journal of Invertebrate Pathology 71, 128–137. Godoy, G., Steadman, J. R., Dickman, M. B. and Dam, R. (1990). Use of mutants to demonstrate the role of oxalic acid in pathogenicity of Sclerotinia sclerotiorum. Physiological and Molecular Plant Pathology 37, 179–191. Goettel, M. S., St. Leger, R. J., Rizzo, N. W., Staples, R. C. and Roberts, D. W. (1989). Ultrastructural-localization of a cuticle-degrading protease produced by the entomopathogenic fungus Metarhizium anisopliae During Penetration of Host (Manduca sexta) cuticle. Journal of General Microbiology 135, 2233–2239. Graniti, A. (1972). The evolution of the toxin concept in plant pathology. In ‘Phytotoxins in Plant Diseases’ (R. K. S. Woods, A. Ballio and A. Graniti, eds.), pp. 1–18. Academic Press, New York. Griesch, J. and Vilcinskas, A. (1998). Proteases released by entomopathogenic fungi impair phagocytic activity, attachment and spreading of plasmatocytes isolated from haemolymph of the greater wax moth Galleria mellonella. Biocontrol Science and Technology 8, 517–531. Griesch, J., Wedde, M. and Vilcinskas, A. (2000). Recognition and regulation of metalloproteinase activity in the haemolymph of Galleria mellonella: a new

FUNGAL PATHOGENS OF INSECTS

307

pathway mediating induction of humoral immune responses. Insect Biochemistry and Molecular Biology 30, 461–472. Grove, J. F. (1980). Cyclodepsipeptides from Beauveria bassiana. 3. The isolation of Beauverolides Ba, Ca, Ja, and Ka. Journal of the Chemical Society-Perkin Transactions 1, 2878–2880. Grove, J. F. and Pople, M. (1980). The insecticidal activity of beauvericin and the enniatin complex. Mycopathologia 70, 103–105. Gunkel, F. A. and Gassen, H. G. (1989). Proteinase K from Tritirachium album Limber. Characterisation of the chromosomal gene and expression of the cDNA in Escherichia coli. European Journal of Biochemistry 179, 185–194. Gupta, S., Roberts, D. W. and Renwick, J. A. A. (1989). Preparative isolation of destruxins from Metarhizium anisopliae by high-performance liquidchromatography. Journal of Liquid Chromatography 12, 383–395. Gupta, S., Krasnoff, S. B., Underwood, N. L., Renwick, J. A. A. and Roberts, D. W. (1991). Isolation of Beauvericin as an insect toxin from Fusarium semitectum and Fusarium moniliforme var subglutinans. Mycopathologia 115, 185–189. Gupta, S., Krasnoff, S. B., Renwick, J. A. A., Roberts, D. W., Steiner, J. R. and Clardy, J. (1993a). Viridoxin-a and Viridoxin-B – Novel toxins from the fungus Metarhizium flavoviride. Journal of Organic Chemistry 58, 1062–1067. Gupta, S. C., Leathers, T. D., Elsayed, G. N. and Ignoffo, C. M. (1993b). Purification and characterization of trypsin from an Entomopathogen, Nomuraea rileyi Nrrl-13755. Current Microbiology 27, 103–107. Gupta, S. C., Leathers, T. D., El-Sayed, G. N. and Ignoffo, C. M. (1994). Relationships among enzyme activities and virulence parameters in Beauveria bassiana infections of Galleria mellonella and Trichoplusia ni. Journal of Invertebrate Pathology 64, 13–17. Gupta, S., Montllor, C. and Hwang, Y. S. (1995). Isolation of Novel Beauvericin Analogs from the Fungus Beauveria bassiana. Journal of Natural ProductsLloydia 58, 733–738. Hackman, R. H. and Goldberg, M. (1974). Light scattering and infra-red spectrophotometric studies of chitin and chitin derivatives. Carbohydrate Research 38, 35–45. Hajek, A. E. and St. Leger, R. J. (1994). Interactions between fungal pathogens and insect hosts. Annual Review of Entomology 39, 293–322. Hamer, J. E. and Holden, D. W. (1997). Linking approaches in the study of fungal pathogenesis: a commentary. Fungal Genetics and Biology 21, 11–16. Hamil, R. L., Higgins, C. E., Boaz, H. E. and Gorman, M. (1969). The structure of beauvericin, a new depsipeptide antibiotic toxic to Artemia salina. Tetrahedron Letters 49, 4255–4258. Haran, S., Schickler, H., Oppenheim, A. and Chet, I. (1996). Differential expression of Trichoderma harzianum chitinases during mycoparasitism. Phytopathology 86, 980–985. Hassan, A. E. M. and Charnley, A. K. (1983). Combined effects of diflubenzuron and the entomopathogenic fungus Metarhizium anisopliae on the tobacco hornworm Manduca sexta. In 10th International Congress on Plant Protection, pp. 790. Brighton: BCPC. Hassan, A. E. M. and Charnley, A. K. (1987). The effect of dimilin on the ultrastructure of the integument of Manduca sexta. Journal of Insect Physiology 33, 669 et seq.

308

A. K. CHARNLEY

Hassan, A. E. M. and Charnley, A. K. (1989). Ultrastructural-study of the penetration by Metarhizium anisopliae through dimilin-affected cuticle of Manduca sexta. Journal of Invertebrate Pathology 54, 117–124. Hegedus, D. D. and Khachatourians, G. G. (1988). Production of an extracellular lipase by Beauveria bassiana. Biotechnology Letters 10, 637–642. Hepburn, H. R. (1985). Structure of the integument. In ‘Comprehensive Insect Physiology, Biochemistry and Pharmacology’ (G. A. Kerkut and L. I. Gilbert, eds.). Pergammon Press, Oxford. Herrmann, M., Zocher, R. and Haeses, A. (1996). The effect of disruption of the enniatin synthetase gene on the virulence of Fusarium avenaceum. Molecular Plant Microbe Interactions 9, 226–232. Heymann, E. (1980). Carboxyesterases and amidases. In ‘Enzymatic Basis of Detoxification’ (W. B. Jakoby, ed.), pp. 291–323. Academic Press, New York. Hillerton, J. E. (1984). Cuticle: mechanical properties. In ‘Biology of the Integument .I. Invertebrates’ (J. Bereiter-Hahn, A. G. Matolsty and K. S. Richards, ed.), pp. 583–625. Springer-Verlag, Berlin. Hinaje, M., Ford, M., Banting, L., Arkle, S. and Khambay, B. (2002). An investigation of the ionophoric characteristics of destruxin A. Archives of Biochemistry and Biophysics 405, 73–77. Hodge, H. T., Krasnoff, S. B. and Humber, R. A. (1996). Tolypocladium inflatum is the anamorph of Cordyceps subsessilis. Mycologia 88, 715–719. Hohn, T. M. (1997). Fungal phytotoxins: biosynthesis and activity. In ‘Plant Relationships’ (G. C. Caroll and P. Tudzynski, eds.), pp. 129–144. Springer, New York. Hojrup, P., Andersen, S. O. and Roepstorff, P. (1986). Isolation, Characterization, and N-terminal sequence studies of cuticular proteins from the migratory locust, Locusta migratoria. European Journal of Biochemistry 154, 153–159. Hubert, M., Cherton, J. C., Vey, A., Lange, C. (1999). Metabolites of A and E destruxins in Galleria mellonella larvae using negative-ion fast-atom bombardment mass spectrometry. Rapid communications in mass spectrometry 13, 179–184. Hurion, N., Fromentin, H. and Keil, B. (1977). Proteolytic enzymes of Entomophthora coronata characterization of a collagenase. Comparative Biochemistry and Physiology b- biochemistry 56B, 259–264. Hurion, N., Fromentin, H. and Keil, B. (1979). Specificity of a collagenolytic enzyme from the fungus Entomophthora coronata: comparison with the bacterial collagenase from Achromobacter iophagus. Archives of Biochemistry and Biophysics 192, 438–445. Huxham, I. M., Lackie, A. M. and McCorkindale, N. J. (1989). Inhibitory effects of cyclodepsipeptides, destruxins, from the fungus Metarhizium anisopliae, on cellular-immunity in insects. Journal of Insect Physiology 35, 97–105. Inglis, G. D., Goettel, M. S., Butt, T. M. and Strasser, H. (2001). Use of hyphomycetous fungi for managing insect pests. In ‘Fungi as Biocontrol Agents’ (T. M. Butt, C. W. Jackson and N. Magan, eds.), pp. 23–69. CAB International, Wallingford. James, P. J., Kershaw, M. J., Reynolds, S. E. and Charnley, A. K. (1993). Inhibition of desert locust (Schistocerca gregaria) malpighian tubule fluid secretion by destruxins, cyclic peptide toxins from the insect pathogenic fungus Metarhizium anisopliae. Journal of Insect Physiology 39, 797–804. Jeffs, L. B. and Khachatourians, G. G. (1997). Toxic properties of Beauveria pigments on erythrocyte membranes. Toxicon 35, 1351–1356.

FUNGAL PATHOGENS OF INSECTS

309

Jegorov, A., Matha, V. and Hradec, H. (1992). Detoxification of Destruxins in Galleria mellonella L Larvae. Comparative Biochemistry and Physiology C-Pharmacology Toxicology & Endocrinology 103, 227–229. Jegorov, A., Sedmera, P., Havlicek, V. and Mat’ha, V. (1998). Destruxin Ed(1) a cyclopeptide from the fungus Metarhizium anisopliae. Phytochemistry 49, 1815–1817. Jegorov, A., Matha, V., Sedmera, P. and Roberts, D. W. (1989). Destruxins from Metarhizium anisopliae. In Proceedings of the International Conference on Biopesticides, Theory and Practice (Ed. by A. Jegorov and V. Matha), pp. 64–70. Jegorov, A., Matha, V. and Weiser, J. (1990). Production of Cyclosporins by Entomopathogenic Fungi. Microbios Letters 45, 65–70. Jegorov, A., Matha, V., Husak, M., Kratochvil, B., Stuchlik, J., Sedmera, P. and Havlicek, V. (1993). Iron uptake system of some members of the genus Tolypocladium: crystal structure of the ligand and its iron (III) complex. Journal of Chemical Society Dalton Transactions 1, 1287–1293. Johnson, R. D., Johnson, L., Itoh, Y., Kodama, M., Otani, H. and Kohmoto, K. (2000). Cloning and characterization of a cyclic peptide synthetase gene from Alternaria alternaria apple pathotype whose product is involved in AM-toxin synthesis and pathogenicity. Molecular Plant Microbe Interactions 13, 742–753. Joshi, L., Charnley, A. K., Arnold, G., Brain, P. and Bateman, R. (1992). Synergism between entomopathogenic fungi, Metarhizium spp., and the benzoylphenyl urea insecticide, teflubenzuron, against the desert locust, Schistocerca gregaria. Brighton Crop Protection Conference: Pests and Diseases, Vols. 1, 2 and 3, 369–374. Joshi, L. and St. Leger, R. J. (1999). Cloning, expression, and substrate specificity of MeCPA, a zinc carboxypeptidase that is secreted into infected tissues by the fungal entomopathogen Metarhizium anisopliae. Journal of Biological Chemistry 274, 9803–9811. Joshi, L., St. Leger, R. J. and Bidochka, M. J. (1995). Cloning of a cuticle-degrading protease from the entomopathogenic fungus, Beauveria bassiana. FEMS Microbiology Letters 125, 211–217. Joshi, L., St. Leger, R. J. and Roberts, D. W. (1997). Isolation of a cDNA encoding a novel subtilisin-like protease (Pr1B) from the entomopathogenic fungus, Metarhizium anisopliae using differential display–RT-PCR. Gene 197, 1–8. Kagamizono, T., Nishino, E., Matsumoto, K., Kawashima, A., Kishimoto, M., Sakai, N., He, B. M., Chen, Z. X., Adachi, T., Morimoto, S. and Hanada, K. (1995). Bassiatin, a new platelet aggregation inhibitor produced by Beauveria bassiana K-717. Journal of Antibiotics 48, 1407–1412. Kaijiang, L. and Roberts, D. W. (1986). The Production of destruxins by the entomogenous fungus Metarhizium anisopliae var major. Journal of Invertebrate Pathology 47, 120–122. Kang, S. C., Park, S. and Lee, D. G. (1998). Isolation and characterization of a chitinase cDNA from the entomopathogenic fungus, Metarhizium anisopliae. FEMS Microbiology Letters 165, 267–271. Kang, S. C., Park, S. and Lee, D. G. (1999). Purification and characterization of a novel chitinase from the entomopathogenic fungus, Metarhizium anisopliae. Journal of Invertebrate Pathology 73, 276–281. Kanost, M. R. (1999). Serine proteinase inhibitors in arthropod immunity. Developmental and Comparative Immunology 23, 291–301.

310

A. K. CHARNLEY

Kawazu, K., Murakami, T., Ono, Y., Kanzaki, H., Kobayashi, A., Mikawa, T. and Yoshikawa, N. (1993). Isolation and characterization of 2 novel nematocidal depsipeptides from an imperfect fungus, Strain D1084. Bioscience Biotechnology and Biochemistry 57, 98–101. Kershaw, M. J., Moorhouse, E. R., Bateman, R., Reynolds, S. E. and Charnley, A. K. (1999). The role of destruxins in the pathogenicity of Metarhizium anisopliae for three species of insect. Journal of Invertebrate Pathology 74, 213–223. Khachatourians, G. G. (1991). Physiology and genetics of entomopathogenic fungi. In ‘Humans, Animals and Insects’ (D. K. Arora, L. Ajello and K. G. Mukerji, eds.), pp. 613–661. Marcel Dekker, New York. Khachatourians, G. G. (1996). Biochemistry and molecular biology of entomopathogenic fungi. In ‘Human and Animal Relationships’ (R. J. Howard and J. S. Miller, eds.), pp. 331–363. Springer-Verlag, Heidelberg. Kim, H. K., Hoe, H. S., Suh, D. S., Kang, S. C., Hwang, C. and Kwon, S. T. (1999). Gene structure and expression of the gene from Beauveria bassiana encoding bassiasin I, an insect cuticle-degrading serine protease. Biotechnology Letters 21, 777–783. Kiuchi, M., Yasui, H., Hayasaka, S. and Kamimura, M. (2003). Entomogenous fungus Nomuraea rileyi inhibits host insect moulting by C22 oxidising inactivation of haemolymph ecdysteroids. Archives of Insect Biochemistry and Physiology 52, 35–44. Kleinkauf, H. and von Dohren, H. (1990). Non-ribosomal biosynthesis of peptide antibiotics. European Journal of Biochemistry 192, 1–15. Kodaira, Y. (1961). Biochemical studies on the muscardine fungi in the silkworms, Bombyx mori L. Journal of Faculty Textile Technology Sinshu University, 5, 1–68. Krasnoff, S. B. and Gupta, S. B. (1991). Identification and biosynthesis of efrapeptins in the fungus Tolypocladium geodes Gams (Deuteromycotina: Hyphomycetes). Journal of Chemical Ecology 17, 1953–1960. Krasnoff, S. B. and Gupta, S. (1992). Efrapeptin production by Tolypocladium fungi (Deuteromycotina, Hyphomycetes) – intraspecific and interspecific variation. Journal of Chemical Ecology 18, 1727–1741. Krasnoff, S. B. and Gupta, S. (1994). Identification of the antibiotic phomalactone from the entomopathogenic fungus Hirsutella thompsonii var synnematosa. Journal of Chemical Ecology 20, 293–302. Krasnoff, S. B., Gupta, S., St. Leger, R. J., Renwick, J. A. A. and Roberts, D. W. (1991). Antifungal and insecticidal properties of the efrapeptins – metabolites of the fungus Tolypocladium niveum. Journal of Invertebrate Pathology 58, 180–188. Krasnoff, S. B., Gibson, D. M., Belofsky, G. N., Gloer, K. B. and Gloer, J. B. (1996). New destruxins from the entomopathogenic fungus Aschersonia sp. Journal of Natural Products 59, 485–489. Kucera, M. (1980). Proteases from the fungus Metarhizium anisopliae toxic for Galleria mellonella larvae. Journal of Invertebrate Pathology 35, 304–310. Kucera, M. (1981). The production of toxic protease by the entomopathogenic fungus Metarhizium anisopliae in submerged culture. Journal of Invertebrate Pathology 38, 33–38. Lange, C., Loutelier, C., Cherton, J. C., Cassier, P., Vey, A. and Pais, M. (1992). Desorption of ions from locust tissues. 2. Metabolites of E-destruxin using negative-ion fast–atom-bombardment mass-spectrometry. Rapid Communications in Mass Spectrometry 6, 28–31.

FUNGAL PATHOGENS OF INSECTS

311

Lange, C., Mulheim, C., Cherton, J. C., Cassier, P., Vey, A. and Pais, M. (1991). Desorption of ions from locust tissues. 1. Behavior of E- destruxin using positive-ion fast-atom-bombardment mass-spectrometry. Rapid Communications in Mass Spectrometry 5, 169–174. Leal, S. C. M., Bertioli, D. J., Butt, T. M., Carder, J. H., Burrows, P. R. and Peberdy, J. F. (1997). Amplification and restriction endonuclease digestion of the Pr1 gene for the detection and characterization of Metarhizium strains. Mycological Research 101, 257–265. Lecuona, R., Riba, G., Cassier, P. and Clement, J. L. (1991). Alterations of insect epicuticular hydrocarbons during infection with Beauveria bassiana or Beauveria brongniartii. Journal of Invertebrate Pathology 58, 10–18. Lewis, D. H. (1987). Evolutionary aspects of mutualistic associations between fungi and photosynthetic organisms. In ‘Evolutionary Biology of the Fungi’ (A. D. M. Rayner, C. M. Brasier and D. Moore, eds.), pp. 161–178. Cambridge University Press, Cambridge. Liu, W. Z., Boucias, D. G. and McCoy, C. W. (1995). Extraction and characterization of the insecticidal toxin Hirsutellin A produced by Hirsutella thompsonii var thompsonii. Experimental Mycology 19, 254–262. Liu, J. C., Boucias, D. G., Pendland, J. C., Liu, W. Z. and Maruniak, J. (1996). The mode of action of hirsutellin A on eukaryotic cells. Journal of Invertebrate Pathology 67, 224–228. Logrieco, K., Ritieni, A., Moretti, A., Randazzo, G. and Bottalico, A. (1997). Beauvericin and fusaproliferin: New emerging Fusarium toxins. Cereal Research Communications 25, 407–413. Logrieco, A., Moretti, A., Castella, G., Kostecki, M., Golinski, P., Ritieni, A. and Chelkowski, J. (1998). Beauvericin production by Fusarium species. Applied and Environmental Microbiology 64, 3084–3088. Loutelier, C., Lange, C., Cassier, P., Vey, A. and Cherton, J. C. (1994). Nonextractive metabolism study of E-destruxin and a-destruxin in the locust, Locusta migratoria L. 3. Direct high-performance liquid-chromatographic analysis and parallel fast-atom- bombardment mass-spectrometric monitoring. Journal of Chromatography B-Biomedical Applications 656, 281–292. Lysenko, O. (1985). Non-sporeforming bacteria pathogenic to insects: incidence and mechanisms. Annual Review of Microbiology 39, 673–695. Lysenko, O. and Kucera, M. (1971). Microorganisms as sources of new insecticidal chemicals – toxins. In ‘Microbial Control of Insects and Mites’ (H. D. Burges and N. W. Hussey, eds.), pp. 205–227. Academic Press, London. Madry, N., Zocker, R. and Kleinkauf, H. (1983). Enniatin production by Fusarium oxysporum in chemically defined media. European Journal of Applied Microbiology and Biotechnology 17, 103–105. Maimala, S., Tartar, A., Boucias, D. and Chandrapatya, A. (2002). Detection of the toxin Hirsutellin A from Hirsutella thompsonii. Journal of Invertebrate Pathology. Matha, V., Weiser, J. and Olejnicek, J. (1988). The effect of tolypin in Tolypocladium niveum crude extract against mosquito and blackfly larvae in the laboratory. Folia Parasitologia 35, 381–383. Mazet, I., Hung, S. Y. and Boucias, D. G. (1994). Detection of toxic metabolites in the haemolymph of Beauveria bassiana infected Spodoptera exigua larvae. Experientia 50, 142–147.

312

A. K. CHARNLEY

Mazet, I. and Vey, A. (1995). Hirsutellin-a, a Toxic Protein Produced in vitro by Hirsutella thompsonii. Microbiology 141, 1343–1348. McCoy, C. W., Samson, R. A. and Boucias, D. G. (1988). Entomogenous fungi. In ‘Vol V Microbial Insecticides, Part A Entomogenous Protozoa and Fungi’ (C. M. Ignoffo, ed.), pp. 151–236. CRC Press, Boca Ratan. Michel, B. 1981. Recherches experimentales sur la penetration des chapignons pathogens chez les insectes., pp. 170: University of Montpellier. Milne, R., Wright, T., Welton, M., Budau, C., Gringorten, L. and Tyrrell, D. (1994). Identification and partial purification of a cell-lytic factor from Entomophaga aulicae. Journal of Invertebrate Pathology 64, 253–259. Mollier, P., Lagnel, J., Fournet, B., Aioun, A. and Riba, G. (1994a). A glycoprotein highly toxic for Galleria mellonella larvae secreted by the entomopathogenic fungus Beauveria sulfurescens. Journal of Invertebrate Pathology 64, 200–207. Mollier, P., Lagnel, J., Quiot, J. M., Aioun, A. and Riba, G. (1994b). Cytotoxic activity in culture filtrates from the entomopathogenic fungus Beauveria sulfurescens. Journal of Invertebrate Pathology 64, 208–213. Moon, T. W. (1975). Temperature adaptation, isozymic function and the maintenance of heterogeneity. In ‘Isozymes II. Physiological Function’ (A. P. Marbet, ed.). Academic Press, New York. Mottram, J. C., Frame, M. J., Brooks, D. R., Tetley, L., Hutchison, J. E., Souza, A. E. and Coombs, G. H. (1997). The multiple cpb cysteine proteinase genes of Leishmania mexicana encode isoenzymes that differ in their stage regulation and substrate preferences. Journal of Biological Chemistry 272, 14285–14293. Murakoshi, S., Ichinoe, M., Kumata, H. and Kurata, H. (1977). Productivity of aflatoxins and some biological effects in Aspergillus flavus link isolated from cadavers of the silkworm Bombyx mori L. Applied Entomology and Zoology 12, 255–259. Murakoshi, S., Ichinoe, M., Suzuki, A., Kanaoka, M., Isogai, A. and Tamura, S. (1978). Presence of toxic substance in fungus fruiting bodies of the entomopathogenic fungi, Beauveria bassiana and Verticillium lecanii. Applied Entomology and Zoology 13, 97–102. Muroi, M., Shiragami, N. and Takatsuki, A. (1994). Destruxin B, a specific and readily reversible inhibitor of vacuolar type Hþ translocating ATPase. Biochemical and Biophysical Research Communications 205, 1358–1365. Muroi, M., Kaneko, N., Suzuki, K., Nishio, T., Oku, T., Sato, T. and Takatsuki, A. (1996). Efrapeptins block exocytic but not endocytic trafficking of proteins. Biochemical and Biophysical Research Communications 227, 800–809. Murphy, J. M. and Walton, J. D. (1996). Three extracellular proteases from Cochliobolus carbonum: Cloning and targeted disruption of ALP1. Molecular Plant-Microbe Interactions 9, 290–297. Neville, A. C. (1984). Cuticle: organisation. In ‘Biology of the Integument, I. Invertebrates’ (J. Bereiter-Hahn, A. G. Matolsty and K. S. Richards, eds.). Springer-Verlag, Berlin. Nikolskaya, A. N., Panaccione, D. G. and Walton, J. D. (1995). Identification of peptide synthetase-encoding genes from filamentous fungi producing hostselective phytotoxins or analogs. Gene 165, 207–211. Nilanonta, C., Isaka, M., Kittakoop, P., Trakulnaleamsai, S., Tanticharoen, M. and Thebtaranonth, Y. (2002). Precursor-directed biosynthesis of beauvericin analogs by the insect pathogenic fungus Paecilomyces tenuipes BCC 1614. Tetrahedron 58, 3355–3360.

FUNGAL PATHOGENS OF INSECTS

313

Nnakumusana, E. S. (1985). Laboratory infection of mosquito larvae by entomopathogenic fungi with particular reference to Aspergillus parasiticus and its effects on fecundity and longevity of mosquitoes exposed to conidial infections in larval stages. Current Science 54, 1221–1228. Odier, F., Vey, A. and Bureau, J. P. (1992). In vitro effect of fungal cyclodepsipeptides on leukemic-cells – study of destruxin-A, destruxin-B and destruxin-E. Biology of the Cell 74, 267–271. Ojcius, D. M., Zychlinsky, A., Zheng, L. M. and Young, J. D.-E. (1991). Ionophoreinduced apoptosis: role of DNA fragmentation and calcium fluxes. Experimental Cell Research 197, 43–49. Omoto, C. and McCoy, C. W. (1998). Toxicity of purified fungal toxin hirsutellin A to the citrus rust mite Phyllocoptruta oleivora. Journal of Invertebrate Pathology 72, 319–322. Ovchinnikov, Y. A., Ivanov, V. T. and Mikhaleva, I. I. (1971). The synthesis and some properties of beauvericin. Tetrahedron Letters 2, 159–162. Pais, M., Das, B. C. and Ferron, P. (1981). Depsipeptides from Metarhizium anisopliae. Phytochemistry 20, 715–723. Papierok, B. (1986). Some noteworthy biological problems within the genus Conidiobolus. In ‘Fundamental and Applied Aspects of Invertebrate Pathology’ (R. A. Samson, J. M. Vlak and D. Peter, eds.), pp. 197–200. Foundation of the Fourth International Colloquium of Invertebrate Pathology, Wageningen. Paris, S. and Ferron, P. (1979). Study of the virulence of some mutants of Beauveria brogniartii. Journal of Invertebrate Pathology 34, 71–77. Paterson, I. C., Charnley, A. K., Cooper, R. M. and Clarkson, J. M. (1993). Regulation of production of a trypsin-like protease by the insect pathogenic fungus Metarhizium anisopliae. FEMS Microbiology Letters 109, 323–327. Paterson, I. C., Charnley, A. K., Cooper, R. M. and Clarkson, J. M. (1994a). Partial characterization of specific inducers of a cuticle-degrading protease from the insect pathogenic fungus Metarhizium anisopliae. Microbiology 140, 3153–3159. Paterson, I. C., Charnley, A. K., Cooper, R. M. and Clarkson, J. M. (1994b). Specific induction of a cuticle-degrading protease of the insect pathogenic fungus Metarhizium anisopliae. Microbiology 140, 185–189. Pearse, A. G. E. (1972). ‘Histochemistry, Theoretical and Applied’. Churchill Livingstone, Edinburgh. Peeters, H., Zocher, R., Madry, N. and Kleinkauf, H. (1983). Incorporation of radioactive precursors into beauvericin produced by Paecilomyces fumosoroseus. Phytochemistry 22, 1719–1720. Pekrul, S. and Grula, E. A. (1979). Mode of infection of the earworm (Heliothis zea) by Beauveria bassiana as revealled by scanning electron microscopy. Journal of Invertebrate Pathology 34, 238–247. Pinto, A. D. S., Barreto, C. C., Schrank, A., Ulhoa, C. J. and Vainstein, M. H. (1997). Purification and characterization of an extracellular chitinase from the entomopathogen Metarhizium anisopliae. Canadian Journal of Microbiology 43, 322–327. Poprawski, T. J., Robert, P. H. and Maniania, N. K. (1994). Contact activity of the mycotoxin destruxin E to Empoasca vitis. Journal of Applied Entomology 117, 135–143. Quiot, J. M., Vey, A. and Vago, C. (1985). Effects of mycotoxins on invertebrate cells In vitro. Advances in Cell Research 4, 203–212. Quiot, J. M., Vey, A., Vago, C. and Pais, M. (1980). Anti-viral activity of a mycotoxin – study of the toxin of the hyphomycete Metarhizium anisopliae

314

A. K. CHARNLEY

(Metsch) Sorok in lepidoptera cell-culture. Comptes Rendus Hebdomadaires Des Seances De L Academie Des Sciences Serie D 291, 763–766. Reyes, F. and Byrde, R. J. W. (1973). Partial purification and properties of Nacetylglucosaminidase from the fungus Sclerotina fructigena. Biochemical Journal 131, 381–388. Richard, J. L. and Gallagher, R. T. (1979). Multiple toxin production by an isolate of Aspergillus flavus. Mycopathologia 67, 161–163. Robert, P. and Riba, G. (1989). Toxic and repulsive effects of spray, per os and systemic applications of destruxin-E to aphids. Mycopathologia 108, 179–183. Roberts, D. W. (1966a). Toxins from the entomogenous fungus Metarhizium anisopliae 1. Production in submerged and surface cultures, and in inorganic and organic nitrogen media. Journal of Invertebrate Pathology 8, 212–221. Roberts, D. W. (1966b). Toxins from the entomogenous fungus Metarhizium anisopliae II. symptoms and detection in moribund hosts. Journal of Invertebrate Pathology 8, 222–227. Roberts, D. W. (1980). Toxins of entomopathogenic fungi. In ‘Microbial Control of Insects, Mites and Plant Diseases’ (H. D. Burges, ed.), pp. 441–463. Academic Press, New York. Robinson, R. K. (1966). Studies on penetration of insect integument by fungi. Pest Articles and News Summaries 12, 131–142. Rosewich, U. L. and Kistler, H. C. (2000). Role of horizontal gene transfer in the evolution of fungi. Annual Review of Phytopathology 38, 325–363. Ruegger, A., Kuhn, M., Lichti, H., Loosli, H. R., Huguenin, R., Quiqierez, C. and Wartburg, V. A. (1976). Cyclosporin A, ein immunosuppressiv wirksamer Peptidmetabolit aus Trichoderma polysporum. Helvetica Chimica Acta 59, 1075–1092. Samsinakova, A., Misikova, S. and Leopold, J. (1971). Action of enzymatic system of Beauveria bassiana on the cuticle of the greater wax moth larvae Galleria mellonella. Journal of Invertebrate Pathology 18, 322–330. Samson, R. A., Evans, H. C. and Latge, J. P. (1988). ‘Atlas of Entomopathogenic Fungi.’ Springer-Verlag, New York. Samuels, R. I. (1998). A sensitive bioassay for destruxins, cyclodepsipeptides from the culture filtrates of the entomopathogenic fungus Metarhizium anisopliae (Metsch.) Sorok. Anais da Sociedade Entomologica do Brasil 27, 229–235. Samuels, R. I., Charnley, A. K. and Reynolds, S. E. (1988a). Application of reversed-phase hplc in separation and detection of the cyclodepsipeptide toxins produced by the entomopathogenic fungus Metarhizium anisopliae. Journal of Chromatographic Science 26, 15–19. Samuels, R. I., Charnley, A. K. and Reynolds, S. E. (1988b). The role of destruxins in the pathogenicity of 3 strains of Metarhizium anisopliae for the tobacco hornworm Manduca sexta. Mycopathologia 104, 51–58. Samuels, R. I., Reynolds, S. E. and Charnley, A. K. (1988c). Calcium-channel activation of insect muscle by destruxins, insecticidal compounds produced by the entomopathogenic fungus Metarhizium anisopliae. Comparative Biochemistry and Physiology C-Pharmacology Toxicology & Endocrinology 90, 403–412. Samuels, R. I., Charnley, A. K. and St. Leger, R. J. (1990). The partial characterization of endoproteases and exoproteases from 3 species of entomopathogenic entomophthorales and 2 species of deuteromycetes. Mycopathologia 110, 145–152.

FUNGAL PATHOGENS OF INSECTS

315

Samuels, R. I., Charnley, A. K. and Reynolds, S. E. (1991). A comparison of insect cuticle degrading proteases from insect moulting fluid and from cultures of entomopathogenic fungi. In Proceedings of the Vth International Colloquium on Invertebrate Pathology and Microbial Control, pp. 313. Adelaide, Australia. Samuels, R. I., Charnley, A. K. and Reynolds, S. E. (1993a). An aminopeptidase from the molting fluid of the tobacco hornworm, Manduca sexta. Insect Biochemistry and Molecular Biology 23, 615–620. Samuels, R. I., Charnley, A. K. and Reynolds, S. E. (1993b). A cuticle-degrading proteinase from the molting fluid of the tobacco hornworm, Manduca sexta. Insect Biochemistry and Molecular Biology 23, 607–614. Samuels, R. I. and Paterson, I. C. (1995). Cuticle degrading proteases from insect molting fluid and culture filtrates of entomopathogenic fungi. Comparative Biochemistry and Physiology B-Biochemistry & Molecular Biology 110, 661–669. Screen, S. E. and St. Leger, R. J. (2000). Cloning, expression, and substrate specificity of a fungal chymotrypsin – Evidence for lateral gene transfer from an actinomycete bacterium. Journal of Biological Chemistry 275, 6689–6694. Screen, S., Bailey, A., Charnley, K., Cooper, R. and Clarkson, J. (1997). Carbon regulation of the cuticle-degrading enzyme PR1 from Metarhizium anisopliae may involve a trans-acting DNA binding protein CRR1, a functional equivalent of the Aspergillus nidulans CREA protein. Current Genetics 31, 511–518. Screen, S., Bailey, A., Charnley, K., Cooper, R. and Clarkson, J. (1998). Isolation of a nitrogen response regulator gene (nrr1) from Metarhizium anisopliae. Gene 221, 17–24. Screen, S. E., Hu, G. and St. Leger, R. J. (2001). Transformants of Metarhizium anisopliae sf anisopliae overexpressing chitinase from Metarhizium anisopliae sf acridum show early induction of native chitinase but are not altered in pathogenicity to Manduca sexta. Journal of Invertebrate Pathology 78. Segers, R., Butt, T. M., Carder, J. H., Keen, J. N., Kerry, B. R. and Peberdy, J. F. (1999). The subtilisins of fungal pathogens of insects, nematodes and plants: distribution and variation. Mycological Research 103, 395–402. Seyoum, E., Moore, D. and Charnley, A. K. (1995). Reduction in flight activity and food consumption by the desert locust, Schistocerca gregaria, after infection with Metarhizium flavoviride. Zeitschrift fur Angewandte Entomologie 118, 310–315. Shimizu, S., Tsuchitani, Y. and Matsumoto, T. (1993a). Production of an extracellular protease by Beauveria bassiana in the hemolymph of the silkworm, Bombyx mori. Letters in Applied Microbiology 16, 291–294. Shimizu, S., Tsuchitani, Y. and Matsumoto, T. (1993b). Serology and substratespecificity of extracellular proteases from 4 Species of entomopathogenic hyphomycetes. Journal of Invertebrate Pathology 61, 192–195. Shnitka, T. K. (1974). Esterases-nonspecific esterases. In ‘Electron Microscopy of Enzymes’ (M. A. Hayat, ed.), pp. 1–53. Van Nostrand-Reinhold. Princeton, N.J. Sloman, I. S. and Reynolds, S. E. (1993). Inhibition of ecdysteroid secretion from Manduca prothoracic glands in vitro by destruxins – cyclic depsipeptide toxins from the insect pathogenic fungus Metarhizium anisopliae. Insect Biochemistry and Molecular Biology 23, 43–46. Smith, R. J. and Grula, E. A. (1983). Chitinase is an inducible enzyme in Beauveria bassiana. Journal of Invertebrate Pathology 42, 319–326.

316

A. K. CHARNLEY

Smith, R. J., Pekrul, S. and Grula, E. A. (1981). Requirements for sequential enzymatic activities for penetration of the integument of the cornworm Helithis zea. Journal of Invertebrate Pathology 38, 335–344. Smithson, S. L., Paterson, I. C., Bailey, A. M., Screen, S. E., Hunt, B. A., Cobb, B. D., Cooper, R. M., Charnley, A. K. and Clarkson, J. M. (1995). Cloning and characterization of a gene encoding a cuticle- degrading protease from the insect pathogenic fungus Metarhizium anisopliae. Gene 166, 161–165. Soledade, M., Pedras, C., Zahira, L. I. and Ward, D. E. (2002). The destruxins: synthesis, biosynthesis, biotransformation, and biological activity. Phytochemistry 59, 579–596. Somero, I. (1975). The role of isoenzymes in adaptation to varying temperatures. In ‘Isozymes II. Physiological Function’ (A. P. Marbet, ed.). Academic Press, New York. Springer, J. P., Cole, R. J., Dorner, J. W., Cox, R. H., Richard, J. L., Barnes, C. L. and van der Helm, D. (1984). Structure and conformation of roseiotoxin. Journal of American Chemical Society 106, 2388–2392. St. Leger, R. J. (1991). Integument as a barrier to microbial infections. In ‘Physiology of the Insect Epidermis’ (K. Binnington and A. Retnakaran, eds.), pp. 284–306. St. Leger, R. J. (1993). Biology and mechanisms of insect-cuticle invasion by Deuteromycete fungal pathogens. In ‘Parasites and Pathogens of Insects’ (N. E. Beckage, S. N. Thompson and B. A. Federici, eds.), pp. 211–229. Academic Press. St. Leger, R. J. (1995). The role of cuticle-degrading proteases in fungal pathogenesis of insects. Canadian Journal of Botany 73, S1119–S1125. St. Leger, R. J. and Bidochka, M. J. (1996). Insect-fungus interactions. In ‘Invertebrate Immunology’ (K. Soderhall, Vasta, G. and Iwanaga, S, eds.), pp. 441–478. SOS Publications, Fairhaven, NJ. St. Leger, R. J. and Roberts, D. W. (1997). Engineering improved mycoinsecticides. Trends in Biotechnology 15, 83–85. St. Leger, R. J. and Screen, S. E. (2000). In vitro utilization of mucin, lung polymers, plant cell walls and insect cuticle by Aspergillus fumigatus, Metarhizium anisopliae and Haematonectria haematococca. Mycological Research 104, 463–471. St. Leger, R. J., Charnley, A. K. and Cooper, R. M. (1986a). Cuticle-degrading enzymes of entomopathogenic fungi – mechanisms of interaction between pathogen enzymes and insect cuticle. Journal of Invertebrate Pathology 47, 295–302. St. Leger, R. J., Charnley, A. K. and Cooper, R. M. (1986b). Cuticle-degrading enzymes of entomopathogenic fungi mechanisms of interaction between pathogen enzymes and insect cuticle. Journal of Invertebrate Pathology 47, 295–302. St. Leger, R. J., Charnley, A. K. and Cooper, R. M. (1986c). Cuticle-degrading enzymes of entomopathogenic fungi: synthesis in culture on cuticle. Journal of Invertebrate Pathology 48, 85–95. St. Leger, R. J., Cooper, R. M. and Charnley, A. K. (1986d). Cuticle-degrading enzymes of entomopathogenic fungi cuticle degradation in vitro by enzymes from entomopathogens. Journal of Invertebrate Pathology 47, 167–177. St. Leger, R. J., Cooper, R. M. and Charnley, A. K. (1986e). Cuticle-degrading enzymes of entomopathogenic fungi regulation of production of Chitinolytic Enzymes. Journal of General Microbiology 132, 1509–1518.

FUNGAL PATHOGENS OF INSECTS

317

St. Leger, R. J., Charnley, A. K. and Cooper, R. M. (1987a). Characterization of cuticle-degrading proteases produced by the entomopathogen Metarhizium anisopliae. Archives of Biochemistry and Biophysics 253, 221–232. St. Leger, R. J., Cooper, R. M. and Charnley, A. K. (1987b). Distribution of chymoelastases and trypsin-like enzymes in five species of entomopathogenic deuteromycetes. Archives of Biochemistry and Biophysics 258, 123–131. St. Leger, R. J., Cooper, R. M. and Charnley, A. K. (1987c). Production of cuticledegrading enzymes by the entomopathogen Metarhizium anisopliae during infection of cuticles from Calliphora vomitoria and Manduca sexta. Journal of General Microbiology 133, 1371–1382. St. Leger, R. J., Charnley, A. K. and Cooper, R. M. (1988a). Production of polyphenol pigments and phenoloxidase by the entomopathogen Metarhizium anisopliae. Journal of Invertebrate Pathology 52, 215–220. St. Leger, R. J., Durrands, P. K., Charnley, A. K. and Cooper, R. M. (1988b). Role of extracellular chymoelastase in the virulence of Metarhizium anisopliae for Manduca sexta. Journal of Invertebrate Pathology 52, 285–293. St. Leger, R. J., Durrands, P. K., Cooper, R. M. and Charnley, A. K. (1988c). Regulation of production of proteolytic enzymes by the entomopathogenic fungus Metarhizium anisopliae. Archives of Microbiology 150, 413–416. St. Leger, R. J., Butt, T. M., Staples, R. C. and Roberts, D. W. (1989a). Synthesis of proteins including a cuticle-degrading protease during differentiation of the entomopathogenic fungus Metarhizium anisopliae. Experimental Mycology 13, 253–262. St. Leger, R. J., Roberts, D. W. and Staples, R. C. (1989b). Calcium-mediated calmodulin-mediated protein synthesis and protein phosphorylation during germination growth and protease production by Metarhizium anisopliae. Journal of General Microbiology 135, 2141–2154. St. Leger, R. J., Cooper, R. M. and Charnley, A. K. (1991a). Characterization of chitinase and chitobiase produced by the entomopathogenic fungus Metarhizium anisopliae. Journal of Invertebrate Pathology 58, 415–426. St. Leger, R. J., Staples, R. C. and Roberts, D. W. (1991b). Changes in translatable messenger RNA species associated with nutrient deprivation and protease synthesis in Metarhizium anisopliae. Journal of General Microbiology 137, 807–816. St. Leger, R. J., Frank, D. C., Roberts, D. W. and Staples, R. C. (1992a). Molecular cloning and regulatory analysis of the cuticle-degrading protease structural gene from the entomopathogenic fungus Metarhizium anisopliae. European Journal of Biochemistry 204, 991–1001. St. Leger, R. J., May, B., Allee, L. L., Frank, D. C., Staples, R. C. and Roberts, D. W. (1992b). Genetic differences in allozymes and in formation of infection structures among isolates of the entomopathogenic fungus Metarhizium anisopliae. Journal of Invertebrate Pathology 60, 89–101. St. Leger, R. J., Staples, R. C. and Roberts, D. W. (1992c). Cloning and regulatory analysis of starvation-stress gene, Ssga, encoding a hydrophobin-like protein from the entomopathogenic fungus, Metarhizium anisopliae. Gene 120, 119–124. St. Leger, R. J., Cooper, R. M. and Charnley, A. K. (1993a). Analysis of aminopeptidase and dipeptidylpeptidase Iv from the entomopathogenic fungus Metarhizium anisopliae. Journal of General Microbiology 139, 237–243. St. Leger, R. J., Staples, R. C. and Roberts, D. W. (1993b). Entomopathogenic isolates of Metarhizium anisopliae, Beauveria bassiana, and Aspergillus

318

A. K. CHARNLEY

flavus produce multiple extracellular chitinase isozymes. Journal of Invertebrate Pathology 61, 81–84. St. Leger, R. J., Bidochka, M. J. and Roberts, D. W. (1994a). Characterization of a novel carboxypeptidase produced by the entomopathogenic fungus Metarhizium anisopliae. Archives of Biochemistry and Biophysics 314, 392–398. St. Leger, R. J., Bidochka, M. J. and Roberts, D. W. (1994b). Isoforms of the cuticledegrading PR1 protease and production of a metalloprotease by Metarhizium anisopliae. Archives of Biochemistry and Biophysics 313, 1–7. St. Leger, R. J., Joshi, L., Bidochka, M. J. and Roberts, D. W. (1995a). Multiple aminopeptidases produced by Metarhizium anisopliae. Journal of Invertebrate Pathology 65, 313–314. St. Leger, R. J., Joshi, L., Bidochka, M. J. and Roberts, D. W. (1995b). Proteinsynthesis in Metarhizium anisopliae growing on host cuticle. Mycological Research 99, 1034–1040. St. Leger, R. J., Joshi, L., Bidochka, M. J., Rizzo, N. W. and Roberts, D. W. (1996a). Biochemical characterization and ultrastructural localization of two extracellular trypsins produced by Metarhizium anisopliae in infected insect cuticles. Applied and Environmental Microbiology 62, 1257–1264. St. Leger, R. J., Joshi, L., Bidochka, M. J., Rizzo, N. W. and Roberts, D. W. (1996b). Characterization and ultrastructural localization of chitinases from Metarhizium anisopliae, Metarhizium flavoviride, and Beauveria bassiana during fungal invasion of host (Manduca sexta) cuticle. Applied and Environmental Microbiology 62, 907–912. St. Leger, R. J., Joshi, L., Bidochka, M. J. and Roberts, D. W. (1996c). Construction of an improved mycoinsecticide overexpressing a toxic protease. Proceedings of the National Academy of Sciences of the United States of America 93, 6349–6354. St. Leger, R. J., Joshi, L. and Roberts, D. W. (1997). Adaptation of proteases and carbohydrases of saprophytic, phytopathogenic and entomopathogenic fungi to the requirements of their ecological niches. Microbiology (UK) 143, 1983–1992. St. Leger, R. J., Joshi, L. and Roberts, D. (1998). Ambient pH is a major determinant in the expression of cuticle-degrading enzymes and hydrophobin by Metarhizium anisopliae. Applied and Environmental Microbiology 64, 709–713. St. Leger, R. J., Nelson, J. O. and Screen, S. E. (1999). The entomopathogenic fungus Metarhizium anisopliae alters ambient pH, allowing extracellular protease production and activity. Microbiology-UK 145, 2691–2699. Stachelhaus, T. and Marahiel, M. A. (1995). Modular structure of genes encoding multifunctional peptide synthetases required for non-ribosomal peptide synthesis. FEMS Microbiology Letters 125, 3–14. Steinrauf, L. K. (1985). Beauvericin and the other enniatins. Metal Ions in Biological Systems 19, 139–171. Stirling, J. L., Cook, E. A. and Pope, A. M. S. (1979). Chitin and its degradation. In ‘Fungal walls and hyphal growth’ (J. A. Burnett and C. Trinci, eds.), pp. 169–188, Cambridge University Press, Cambridge. Strasser, H., Abendstein, D., Stuppner, H. and Butt, T. M. (2000). Monitoring the distribution of secondary metabolites produced by the entomogenous fungus Beauveria brongniartii with particular reference to oosporein. Mycological Research 104, 1227–1233.

FUNGAL PATHOGENS OF INSECTS

319

Strongman, D. B., Strunz, G. M., Giguere, P., Yu, C.-M. and Calhoun, L. (1987). Enniatins from Fusarium avenaceum isolated from balsam fir foliage and their toxicity to spruce budworm larvae, Choristoneura fumiferana. Journal of Chemical Ecology 14, 753–764. Suzuki, A. and Tamura, S. (1972). Isolation and structure of protodestruxin from Metarrhizium anisopliae. Agricultural and Biological Chemistry 36, 896–898. Suzuki, A., Taguchi, H. and Tamura, S. (1970). Isolation and structure elucidation of three new insecticidal cyclodepsipeptides, destruxins C and D and desmethyldestruxin B produced by Metarhizium anisopliae. Agricultural and Biological Chemistry 34, 813–816. Suzuki, A., Kawakami, K. and Tamura, S. (1971). Detection of destruxins in silkworm larvae infected with Metarhizium anisopliae. Agricultural and Biological Chemistry 35, 1641–1643. Suzuki, A., Kanaoka, M., Isogai, A., Murakoshi, S., Ichinoe, M. and Tamura, S. (1977). Bassianolide, a new insecticidal cyclodepsipeptide from Beauveria bassiana and Verticillium lecanii. Tetrahedron Letters 25, 2167–2170. Tanada, Y. and Kaya, H. K. (1993). ‘Insect Pathology’. Academic Press, San Diego. Tsantrizos, Y.S., Xu, X.J., Sauriol, F. and Hynes, R.C. (1993) Novel quinazolinones and enniatins from Fusarium lateritium. Canadian Journal of Chemistry 71, 1362–1367. Tunlid, A. and Talbot, N. J. (2002). Genomics of parasitic and symbiotic fungi. Current Opinion in Microbiology 5, 513–519. Turlington, L. W., Woodward, D. L. and Colwell, A. E. (1990). Pathogenesis of Tolypocladium cylindrosporium (Deuteromycotina: Hyphomycetes) in Chaoborus astictopus (Diptera: Chaoboridae). Journal of Invertebrate Pathology 55, 126–129. Turner, W. B. and Aldridge, D. C. (1983). Triterpenoids and steroids. In ‘Fungal Metabolites 2nd Edition (W. B. Turner and D. C. Aldridge, eds.), pp. 225–336. Academic Press, London. Tyrrell, D. (1990). Pathogenesis of Entomophaga aulicae. 1. Disease symptoms and effect of infection on weight-gain of infected Choristoneura fumiferana and Malacosoma disstria Larvae. Journal of Invertebrate Pathology 56, 150–156. Vakili, N. G. (1990). Biocontrol of stalk rot in corn. In Proceedings of the 44th annual corn and sorgum research conference, pp. 87–105. Venkatsubbaiah, P., Tisserat, N. A. and Chilton, W. S. (1994). Metabolites of Ophiosphaerella herpotricha. Mycopathologia 128, 155–159. Vey, A. (1985). Efficiency of the encapsulation reaction in insects – effect on the structure and the viability of the encapsulated fungus. Developmental and Comparative Immunology 9, 174–174. Vey, A. and Quiot, J. M. (1989a). Cytotoxic effect of destruxins cyclodepsipeptide toxins produced by the entomopathogenic fungus Metarhizium anisopliae investigated in-vitro and in the insect host. Canadian Journal of Microbiology 35, 1000–1008. Vey, A. and Quiot, J. M. (1989b). In vitro and insect-host cyto-toxicity of destruxins, cyclodepsipeptide mycotoxins produced by the entomopathogen, Metarhizum anisopliae. Canadian Journal of Microbiology 35, 1000–1008. Vey, A., Quiot, J. M. and Pais, M. (1986). Toxemia of a fungal origin in invertebrates and its cytotoxic consequences – studies on the infection caused by Metarhizium anisopliae (Hyphomycete, Moniliales) in lepidoptera and coleoptera. Comptes Rendus Des Seances De La Societe De Biologie Et De Ses Filiales 180, 105–112. Vilcinskas, A. and Wedde, M. (1997). Inhibition of Beauveria bassiana proteases and fungal development by inducible protease inhibitors in the

320

A. K. CHARNLEY

haemolymph of Galleria mellonella larvae. Biocontrol Science and Technology 7, 591–601. Vilcinskas, A., Kopacek, P., Jegorov, A., Vey, A. and Matha, V. (1997a). Detection of lipophorin as the major cyclosporin-binding protein in the hemolymph of the greater wax moth Galleria mellonella. Comparative Biochemistry and Physiology C-Pharmacology Toxicology & Endocrinology 117, 41–45. Vilcinskas, A., Matha, V. and Gotz, P. (1997b). Effects of the entomopathogenic fungus Metarhizium anisopliae and its secondary metabolites on morphology and cytoskeleton of plasmatocytes isolated from the greater wax moth, Galleria mellonella. Journal of Insect Physiology 43, 1149–1159. Vilcinskas, A., Matha, V. and Gotz, P. (1997c). Inhibition of phagocytic activity of plasmatocytes isolated from Galleria mellonella by entomogenous fungi and their secondary metabolites. Journal of Insect Physiology 43, 475–483. Vilcinskas, A., Jegorov, A., Landa, Z., Gotz, P. and Matha, V. (1999). Effects of beauverolide L and cyclosporin A on humoral and cellular immune response of the greater wax moth, Galleria mellonella. Comparative Biochemistry and Physiology C-Pharmacology Toxicology & Endocrinology 122, 83–92. Vining, L. C. (1990). Functions of secondary metabolites. Annual Review of Microbiology 44, 395–427. Vining, L. C., Kelleher, W. J. and Schwarting, A. E. (1962). Osporein production by a strain of Beauveria bassiana originally identified as Amanita muscaria. Canadian Journal of Microbiology 8, 931–933. Visconti, A., Blais, L. A., ApSimon, J. W., Greenhalgh, R. and Miller, J. D. (1992). Production of enniatins by Fusarium acuminatum and Fusarium compactum in liquid culture: isolation and characterization of three new enniatins, B2, B3, B4. Journal of Agriculture and Food Chemistry 40, 1076–1082. Wahlman, M. and Davidson, B. S. (1993). New destruxins from the entomopathogenic fungus Metarhizium anisopliae. Journal of Natural Products 56, 643–647. Walton, J. D. (1996). Host-selective toxins: agents of compatibility. Plant Cell 8, 1723–1733. Walton, J. D. (2000). Horizontal gene transfer and the evolution of secondary metabolite gene clusters in fungi: an hypothesis. Fungal Genetics and Biology 30, 167–171. Wang, C. S., Typas, M. A. and Butt, T. M. (2002). Detection and characterisation of pr1 virulent gene deficiencies in the insect pathogenic fungus Metarhizium anisopliae. FEMS Microbiology Letters 213, 251–255. Wartburg, V. A. and Traber, R. (1988). Cyclosporins, fungal metabolites with immunosuppressive activities. In ‘Progress in Medicinal Chemistry’ (G. P. Ellis and G. B. West, eds.), pp. p133. Elsevier Science Publishers. Wat, C.-K., McInnis, A. G., Smith, D. G., Wright, J. L. C. and Vining, L. C. (1977). The yellow pigments of Beauveria bassiana species. Structures of tenellin and bassianin. Canadian Journal of Chemistry 55, 4090–4098. Waterfield, N. R., Daborn, P. J. and ffrench-Constant, R. H. (2002). Genomic islands in Photorhabdus. Trends in Microbiology 10, 541–545. Weiser, J. and Matha, V. (1988a). The insecticidal activity of cyclosporines on mosquito larvae. Journal of Invertebrate Pathology 51, 143–150. Weiser, J. and Matha, V. (1988b). Tolypin, a new insecticidal metabolite of fungi of the genus Tolypocladium. Journal of Invertebrate Pathology 51, 94–96. Wigglesworth, V. B. (1970). Structural lipids in the insect cuticle and the function of of oenocytes. Tissue & Cell 2, 155–179.

FUNGAL PATHOGENS OF INSECTS

321

Wright, V. F., Vesonder, R. F. and Ciegler, A. (1982). In ‘Microbial and Viral Pesticides’ (E. Kurstak, ed.). Marcel Dekker. Yeh, S. F., Pan, W., Ong, G. T., Chiou, A. J., Chuang, C. C., Chiou, S. H. and Wu, S. H. (1996). Study of structure-activity correlation in destruxins, a class of cyclodepsipeptides possessing suppressive effect on the generation of hepatitis B virus surface antigen in human hepatoma cells. Biochemical and Biophysical Research Communications 229, 65–72. Yoder, O. C. and Turgeon, B. G. (2001). Fungal genomics and pathogenicity. Current Opinion in Plant Biology 4, 315–321. Yoshimoto, Y. and Imoto, M. (2002). Induction of EGF-dependent apoptosis by vacuolar-type Hþ-ATPase inhibitors in A431 cells overexpressing the EGF receptor. Experimental Cell Research 279, 118–127. Zacharuk, R. Y. (1970a). Fine structure of the fungus Metarrhizium anisopliae infecting 3 species of larval elateridae coleoptera Part 2 conidial germ tubes and appressoria. Journal of Invertebrate Pathology 15, 81–91. Zacharuk, R. Y. (1970b). Fine structure of the fungus Metarrhizium anisopliae infecting 3 species of larval elateridae coleoptera Part 3 penetration of the host integument. Journal of Invertebrate Pathology 15, 372–396. Zeller, E. A. (1977). Snake venom action: are enzymes involved in? Experientia 33, 143–150.

This Page Intentionally Left Blank

AUTHOR INDEX

Numbers in bold refer to pages on which full references are listed

A Aalen, R. B., 134 Aarts, M. G. M., 98 Abendstein, D., 318 Abrams, J. M., 173 Adachi, T., 309 Adamska, I., 56 Admiraal, W., 185, 222, 231 Ahlandsberg, S., 58 Ahmad, M., 159, 172 Ahn, J., 179 Ainsworth, C. C., 49 Aioun, A., 312 Akazawa, T., 48, 56 Akman, Y., 100 Al-Aidroos, K., 300 Al-Khodairy, F., 153, 172, 174 Alberts, B., 152, 172, 231 Alcoverro, T., 197, 231 Aldridge, D. C., 282, 283, 285, 300, 319 Allan, G. G., 209, 210, 214, 231 Allee, L. L., 317 Allfrey, V. G., 122, 134 Allis, C. D., 123, 134, 137, 138, 139 Alon, T., 135 Alonso, M. D., 37, 47 Alphey, L., 158, 172 Aman, P., 47, 56 Amasino, R. M., 178 Ames, I. A., 53 Amiri, B., 289, 301 Amiri-Besheli, B., 288, 293, 294, 301

Advances in Botanical Research Vol. 40 incorporating Advances in Plant Pathology ISBN 0-12-005940-1

Amos, C. L., 228, 231, 232, 238 Amtmann, A., 102 Andersen, S. O., 308 Andersson, L., 7, 47 Andersson, R., 47 Andon, N. L., 24, 47 Andre, F., 303 Angle, J. S., 100 Anke, H., 244, 278, 287, 301 ap Rees, T., 24, 41, 48 Appels, R., 54 ApSimon, J. W., 320 Arakawa, H., 179 Ardila, F., 37, 48 Ardiles, W., 306 Arkle, S., 308 Armstrong, J., 161, 172 Armstrong, W., 161, 172 Arnold, G., 309 Arunika, H. L. A. N., 161, 172 Askary, H., 250, 269, 301 Aspinall, D., 53 Assalkhou, R., 134 Assunc¸a˜o, A. G. L., 68, 71, 74, 78, 98, 100 Atkinson, R. G., 173 Au-Yeung, P., 61 Austin, H. A., 236 Autio, K., 56 Avivi, Y., 181 Axelos, M., 172 Ayala, J., 136 Ayaydin, F., 177 Copyright  2003 Elsevier Ltd All rights of reproduction in any form reserved

324

AUTHOR INDEX

B Baba, T., 28, 38, 48 Baber-Furnari, B. A., 153, 172 Bacic, A., 235 Bacon, P. J., 53 Bae, J. M., 52 Baetz, J., 305 Baga, M., 56 Bagga, S., 255, 256, 272, 273, 301, 306 Bailey, A. M., 295, 299, 301, 315, 316 Bains, P. S., 289, 296, 301 Bako´, L., 177 Baker, A. J. M., 64, 65, 68, 71, 72, 85, 89, 95, 96, 99, 100, 101, 102, 103, 104, 105 Baldwin, P. M., 51 Balkwill, K., 100 Ball, K., 57 Ball, S. G., 55 Ball, S., 36, 38, 48, 50, 55, 60 Ballicora, M. A., 21, 44, 48 Banas, J., 304 Bandani, A. R., 285, 290, 301 Banisadr, R., 58 Banting, L., 308 Bantle, J. A., 306 Bar-Shira, A., 179 Baranczewski, P., 134 Barbarulo, M. V., 232 Barber, L., 50 Barcelo´, J., 104 Barnes, C. L., 316 Baroja-Fernandez, E., 24, 48 Barranguet, C., 223, 231, 235 Barratt, M. J., 135 Barreto, C. C., 313 Barry, G. F., 48 Barry, G., 52 Barton, C. R., 57 Bassi, A., 242, 243, 301 Bateman, R., 306, 309, 310 Baumann, A., 64, 99 Baumbusch, L. O., 125, 134 Beach, D., 177 Beard, R. L., 282, 284, 301 Becker, P. B., 134, 135, 137, 138, 140 Beckles, D. M., 21, 22, 48, 49 Beclin, C., 138 Belloncik, S., 303 Belofsky, G. N., 310 Belyaev, N. D., 125, 134 Bender, J., 137, 138 Bender, M. A., 139 Benhamou, N., 301 Bennett, M. D., 175 Bentley, N. J., 153, 172, 176, 180 Berazan, R., 104 Berbee, M. L., 271, 301 Bergh, M. O., 47

Berlinsky, M., 232 Bert, V., 71, 72, 74, 83, 99, 102 Bertioli, D. J., 311 Betz, C., 178 Bewley, J. D., 49 Bhalerao, R. P., 180 Bhattacharyya, M. K., 49 Bhosle, N. B., 208, 231 Bhullar, S. S., 58 Bickmore, W. A., 138 Bidochka, M. J., 251, 252, 254, 263, 265, 266, 267, 270, 271, 272, 274, 276, 281, 301, 302, 309, 316, 318 Biggar, S. R., 140 Binderup, K., 60 Bing, L. A., 280, 302 Bird, A., 138, 141 Bisgrove, S. R., 173 Bishop, J., 101 Bisset, Y., 178 Bjo¨rn, L. O., 178 Bjelic, S., 232 Black, K. S., 186, 187, 224, 225, 229, 231, 236, 238 Black, M., 2, 48 Blackburn, J., 239 Blais, L. A., 320 Blake, N. K., 58 Blakeney, A. B., 51 Blanchard, G. F., 217, 218, 231, 232 Blasina, A., 155, 172 Blauth, S. L., 33, 34, 48 Blaylock, M. J., 96, 99 Blejec, A., 60 Blennow, A., 50, 55 Blom, C. W. P. M., 172 Blomquist, G. J., 270, 302 Boaz, H. E., 307 Boddy, M., 172 Bogo, M. R., 258, 302 Bogracheva, T. Y., 60 Bogus, M. I., 278, 302 Bonnin, I., 99 Bonte, E. J., 135, 137 Bonte, E., 135, 140 Booher, R., 177 Borhidi, A., 70, 99, 104 Bornemann, S., 52 Borovsky, D., 32, 33, 48 Boschker, H. T. S., 235 Bosnes, M., 3, 4, 6, 49 Bottalico, A., 311 Bottrill, A. R., 53 Bouchet, B., 51, 60 Bouchez, D., 175 Boucias, D. G., 245, 286, 302, 311, 312 Boudreau, B. P., 218, 231 Bouget, F. -Y., 173

AUTHOR INDEX Boulcott, M., 192, 193, 196, 203, 205, 207, 208, 209, 210, 213, 214, 215, 231 Bourachot, B., 138 Bowler, C., 184, 233 Bowsher, C. G., 51, 59 Boyd, R. S., 71, 82, 83, 84, 85, 86, 87, 90, 93, 99, 100, 103 Boyer, C. D., 7, 39, 49 Bradbury, D., 6, 49 Bradfisch, G. A., 290, 303 Brailsford, R. W., 161, 172 Brain, P., 309 Brainerd, E. E., 176 Brangeon, J., 22, 49 Branscheid, A., 59 Braun, A., 139 Breedon, T., 105 Brehm, A., 114, 134 Brenner, M. L., 52 Bresnick, E. H., 123, 136 Brettar, I., 237 Brewer, D., 281, 303 Brewer, E. P., 100 Brey, P. T., 250, 303 Briarty, L. G., 6, 49 Brikos, C., 177 Brinsley, M. D., 239 Brisson, L. F., 173 Britt, A. B., 158, 159, 170, 172, 175 Broadley, M. R., 70, 100, 105 Brobyn, P. J., 250, 303 Brockerhoff, H., 269, 303 Brodeur, J., 301 Brooks, D. R., 312 Brooks, R. R., 64, 65, 66, 67, 68, 69, 97, 98, 100, 101, 103, 104 Brosch, G., 124, 134 Brotas, V., 236 Brousseau, C., 289, 303 Brower, L. P., 93, 100 Brown, A. P., 135 Brown, C. E., 123, 134 Brown, S. L., 96, 100 Brown, S., 239 Brownell, J. E., 122, 134 Bruchet, H., 175 Bruno, M., 232 Brylinski, M., 231 Brzeski, J., 114, 117, 119, 121, 135 Buchanan-Wollaston, V., 172 Buchwaldt, L., 287, 303 Buckner, B., 147, 172 Budau, C., 312 Budde, A., 135 Buijk, A., 54 Buleon, A., 48, 60 Bulger, M., 136 Bull, V., 49 Bureau, J. P., 313

325

Burke, J., 239 Burke, S., 302 Burne, I. F., 175 Burns, J. A., 41, 53 Burrel, M. M., 59 Burrell, M. M., 51 Burrows, P. R., 311 Burt, N., 224, 232 Burton, R. A., 8, 13, 23, 33, 35, 36, 39, 49 Bush, D. S., 176 Bustos, R., 52 Butt, T. M., 250, 301, 303, 308, 311, 315, 317, 318, 320 Buzan, C. L., 60 Byrde, R. J. W., 259, 314

C Cahoon, L. B., 185, 232 Cairns, B. R., 112, 135 Caldwell, M., 178 Calhoun, L., 318 Callard, D., 161, 162, 172 Cameron, R. E., 53 Cameron, S., 301 Camescasse, D., 175 Campbell, R., 174 Cao, H., 24, 25, 49, 57 Cao, X., 126, 127, 135, 137 Cao, Y., 136 Cappelluti, S., 176 Carder, J. H., 311, 315 Carlson, M., 140 Carlson, S. J., 6, 49 Carmen, A. A., 140 Carmichael, J. P., 177 Carneiro, N. P., 179 Carr, A. M., 153, 172, 173, 174, 176, 177, 180 Carrangis, L., 49 Carruthers, L. M., 136 Carruthers, R. I., 242, 303 Caspar, T., 55 Caspari, T., 153, 173, 177 Cassier, P., 304, 310, 311 Casteels, P., 179 Castella, G., 311 Catarino, F., 237 Cavelier, F., 291, 303 Cervantes, M., 176 Chamberlain, A. H. L., 232 Chamberland, H., 173 Chambon, P., 138 Chamnongpol, S., 168, 173 Champlin, F. R., 287, 303 Chan, T. A., 155, 173 Chancy, C. A., 306 Chanda, S. V., 11, 49 Chandrapatya, A., 311

326

AUTHOR INDEX

Chaney, R. L., 94, 100 Chardonnens, A. N., 104 Charnley, A. K., 242, 244, 245, 248, 249, 250, 251, 253, 269, 271, 274, 275, 301, 303, 304, 306, 307, 308, 309, 310, 313, 314, 315, 316, 317 Charnock, J. M., 101 Charpentier, G., 303 Chaudhury, A. M., 3, 49 Chelkowski, J., 311 Chen, H. C., 288, 304 Chen, J. J., 172 Chen, J., 61, 139 Chen, P., 145, 173 Chen, Y., 239 Chen, Z. J., 124, 140 Cherton, J. C., 291, 304, 310, 311 Chet, I., 307 Cheung, A. Y., 171, 180 Chevalier, C., 176 Chevalier, P., 2, 6, 49, 55 Chiarucci, A., 100 Chibbar, R. N., 56 Childers, D. L., 226, 232 Chilton, W. S., 319 Chiou, A. J., 320 Chiou, S. H., 320 Choi, S. B., 53, 58 Choisne, N., 175 Chou, C. K., 304 Chourey, P. S., 22, 49, 55, 60 Chowdhury, S. I., 9, 47, 49 Christensen, P. U., 176, 177 Christenson, E. R., 176 Christian, H., 232 Christie, M. C., 221, 232 Christy, A. M., 101 Chrzanowska, J., 263, 304 Chu, X. S., 54 Chua, Y. L., 124, 135 Chuang, C. C., 320 Chung, H. J., 57 Ciegler, A., 320 Citterio, E., 114, 135 Clancy, M., 58 Clapier, C. R., 134, 135 Clardy, J., 307 Clare, A., 177 Clark, J. R., 25, 49 Clarke, B. R., 30, 31, 50, 54 Clarke, B., 50, 54 Clarke, C. J., 60 Clarke, P. R., 175 Clarkson, J. M., 244, 301, 304, 313, 315, 316 Clay, K., 280, 304 Claydon, N., 246, 278, 282, 283, 304, 306 Clayton, A. L., 124, 135 Clement, J. L., 311

Clijsters, H., 103 Cobb, B. D., 316 Cobb, B. G., 52 Cockcroft, C., 177 Cohen, C. K., 89, 100 Cohen, P., 37, 58 Cole, R. J., 316 Cole, R. W., 177 Cole, S. C. J., 255, 304 Coleman, T. R., 177 Coles, H. R. S., 180 Coles, S. M., 226, 232 Colleoni, C., 50 Colonna, P., 48 Colot, V., 136 Colwell, A. E., 319 Come, D., 48 Commuri, P. D., 27, 31, 50 Concannon, P., 180 Condon, A. G., 57 Connolly, E., 101 Consalvey, M., 186, 189, 232, 236 Constantinidis, T., 103 Conte, E., 231 Cook, M. G., 58 Cooksey, B., 204, 232 Cooksey, K., 204, 232 Coombs, G. H., 312 Cooper, R. M., 258, 262, 304, 313, 315, 316, 317 Cope, G. H., 104 Corbineau, F., 48 Cornishbowden, A., 40, 52 Corona, D. F., 113, 114, 135, 137, 138 Corpet, F., 119, 135 Correa, C. T., 302 Correlou, F., 157, 173 Cortez, D., 155, 173 Cote, J., 112, 135, 140, 141 Cotsopoulos, B., 103 Cotter-Howells, J. D., 101 Couch, J. A., 306 Coudron, T. A., 265, 304, 305 Cox, D. L., 275, 304 Cox, R. H., 316 Crabtree, G. R., 137, 140 Craig, G., 231 Craig, J., 27, 48, 50 Cramp, A., 231, 232 Crawford, R. M., 236, 237 Crawford, S. A., 234 Crookston, R. K., 11, 58 Crosby, M. A., 119, 135 Crosby, W.L., 180 Cross, D. A. E., 175 Cross, F., 145, 146, 173 Cross, R. L., 285, 304 Crossa, J., 57 Crowley, K. A., 138

AUTHOR INDEX Cuellar, J. A., 9, 60 Cui, K., 169, 173 Cull, I. M., 49 Currie, A. R., 176

D D’Hulst, C., 60 D’Souza, S. F., 234 Da Costa Martins, P., 98 Da Silva, J. M., 237 Daborn, G. R., 195, 227, 232 Daborn, P. J., 320 Dade, B. W., 228, 229, 232 Dam, R., 306 Damager, I., 25, 50 Dandridge, K. L., 101, 103 Dangl, J. L., 147, 175, 176 Daniel, G. F., 187, 191, 232 Daniels, R. R., 49 Danner, S., 57 Dante, R. A., 179 Darien, I. L., 53 Darveau, A., 173 Das, B. C., 313 Dasilva, J. C., 274, 304 Datta, R., 49 Dauvillee, D., 60 Davidson, B. S., 288, 320 Davidson, I., 236, 239 Davies, E. J., 59 Davies, M. S., 175 Davies, T. J., 180 Davis, J. D., 232 Davis, M. A., 86, 100 Dawson, I., 172 Day, J. W., 232 De Angelis, F., 210, 213, 232 De Boer, P. L., 227, 232 De Brouwer, J. F. C., 191, 192, 193, 194, 195, 196, 197, 201, 202, 205, 206, 207, 208, 209, 210, 213, 214, 219, 220, 228, 229, 231, 232, 233, 236 De Deckere, E. M. G. T., 220, 228, 232, 233, 236, 237 De Folter, S., 98 De Gara, L., 173 De Jong, J. H., 109, 136 De Languerie, P., 99 De Pater, B. S., 54 De Pater, S., 54 De Rocher, A., 50 De Winder, B., 233, 237, 239, 240 Deadman, M. L., 301 Dean, J. M., 234 Debeaupuis, J. P., 290, 305 Decho, A. W., 187, 190, 191, 215, 222, 223, 233

327

Decq, A., 50, 55 Defew, E. C., 227, 228, 233 Delafontaine, M. T., 220, 233 Delance, J., 305 Delgado, I. J., 37, 50 Dellaire, G., 138 Dellaporta, S. L., 139 Delmer, D., 56 Delrue, B., 28, 38, 50, 55 Delvalle, D., 60 DeMaggio, A., 172 DeMason, D. A., 60 Dembski, M., 177 Deng, X-W., 157, 180 Dennis, E. S., 49, 136 Denu, J. M., 139 Denyer, K., 21, 25, 28, 29, 31, 38, 49, 50, 53, 56, 58, 59 DePinto, M. C., 169, 173 Dermastia, M., 60 Desikan, R., 162, 168, 173, 177 Desiqueira, J. P., 304 Deuring, R., 135 Dhugga, K. S., 37, 50 Dickinson, J. R., 179 Dickman, M. B., 306 Dietrich, J., 56 Dietrich, R. A., 175 Dikovskaya, D., 175 Dilkes, B. P., 179 Dinges, J. R., 35, 36, 50, 53 Dion, M., 169, 173 Doan, D. N. P., 43, 50 Doehlert, D. C., 6, 8, 51 Doke, N., 169, 173 Donald, A. M., 19, 20, 51, 53, 60 Dong, H., 238 Dong, L. F., 238 Dong, Y. -H., 169, 173 Doonan, J., 170, 173, 177 Dorner, J. W., 316 Dowd, P. F., 282, 284, 298, 305 Drapeau, G., 232 Dreier, W., 45, 51 Drew, M. C., 161, 163, 173, 174, 175 Droop, S. J. M., 184, 235 Drum, R. W., 189, 234 Drummond, J., 281, 305 Du, J., 135 Dubois, M., 225, 233 Dudits, D., 176, 177 Dudley, T. R., 100 Dumas, C., 288, 289, 290, 291, 305 Dumas, K., 140 Dunham, S. J., 102, 105 Dunlap, F., 50 Dunphy, G. B., 278, 305 Dunphy, W. G., 176, 177 Durocher, D., 153, 155, 174

328

AUTHOR INDEX

Durrands, P. K., 317 Dyer, K. R., 221, 231, 232, 233 Dyson, H. J., 136 E Earnshaw, W. C., 161, 174 Ebbs, S. D., 103 Eberharter, A., 113, 134, 135 Eckman, J. E., 236 Edgar, B. A., 145, 174 Edgar, L. A., 187, 189, 233 Edmondson, D.G., 134 Edwards, A., 27, 50, 51, 52 Edwards, R. J., 176 Edwards, R., 103 Eide, D., 101, 103 Eisen, J. A, 112, 136 Eisenmann, H., 237 Eisma, D., 221, 233 El-Sayed, G. N., 263, 265, 274, 305, 307 Elledge, S. J., 173, 175, 177, 180 Elliott, M., 239 Ellis, B. K., 237 Ellis, R. E., 147, 174 Ellis, R. J., 56 Emes, M. J., 24, 43, 51, 59 Engstrom, G., 289, 305 Enoch, T., 180 Erdjument-Bromage, H., 135, 138, 141 Ernst, W. H. O., 77, 86, 100 Erry, B. V., 92, 101 Evans, D. E., 172 Evans, H. C., 247, 271, 277, 305, 314 Evans, L. T., 46, 47, 51, 58 Evers, A. D., 2, 8, 51 Evers, T., 2, 3, 6, 12, 51 Eyal, J., 281, 305 F Faas, R. W., 232 Fahy, B. F., 53, 56 Fahy, B., 49, 51 Falciatore, A., 184, 233 Fan, Y., 131, 132, 136 Fang, H. H. P., 219, 229, 235 Farmerie, W. G., 302 Farre, E. M., 59 Faulkner, R., 134 Faull, J. L., 301 Fehe´r, A., 177 Felker, F. C., 6, 51 Felsenfeld, G., 137 Felsenstein, 115, 136 Feng, Q., 113, 114, 136 Fenning, T. M., 175 Ferl, R. J., 57 Ferrari, S., 135 Ferron, P., 274, 306, 313

ffrench-Constant, R. H., 320 Fiebig, D., 237 Fincher, G. B., 49 Findlay, S., 221, 237 Finnegan, E. J., 126, 130, 136 Fischbein, K. L., 305 Fischer, A., 134 Fisher, D. K., 51 Flaggs, G., 172, 176 Flemming, B. W., 220, 233 Flowers, S. K., 137 Fogg, G. E., 198, 233 Fontaine, R., 50 Forbes, K. C., 180 Ford, J. C., 172, 174 Ford, M., 308 Forsberg, E. C., 123, 136 Forsburg, S. L., 166, 174 Foss, S. S., 306 Fotou, E., 172, 174 Fournet, B., 50, 312 Fowke, L. C., 180, 181 Fox, T. C., 100 Fox, T., 101 Frame, M. J., 312 Francis, D., 149, 150, 173, 174, 175, 179 Frank, D. C., 317 Fransz, P. F., 109, 136, 139 Franzke, A., 103 Frappier, F., 287, 306 Fredriksson, H., 47 Freeman, C., 221, 234 Freeman, D., 177 Frehner, M., 56 Freimoser, F. M., 244, 255, 271, 273, 276, 278, 295, 297, 298, 300, 306 French, D., 48 Frey, B., 80, 101 Fricaud, A. C., 285, 306 Friso, G., 56 Fromentin, H., 308 Frostick, L. E., 219, 234 Frueauf, J. B., 48 Fu, D., 140 Fu, Y. B., 48 Fujita, N., 54 Fukuda, H., 144, 147, 164, 165, 174, 176 Fulton, D. C., 27, 28, 38, 51 Furnari, B., 172, 176 Fyodorov, D. V., 136

G Gaba, V., 175 Galbiati, M., 139 Gale, K. R., 54 Gallagher, R. T., 284, 314 Gallant, D. J., 15, 16, 17, 51 Gallie, D. R., 4, 60, 167, 168, 180

AUTHOR INDEX Galois, R., 231 Gambacorta, A., 235 Gan, S., 178 Gao, M., 25, 27, 33, 51, 56 Gao, X., 167, 175 Garcia, V., 153, 160, 175 Gardner, R. C., 173 Gardner, R. L., 180 Garg, A., 231 Garvin, D. F., 100, 103 Gassen, H. G., 272, 307 Gaszner, M., 137 Gdula, D. A., 113, 136, 137 Geetha, K. B., 8, 51 Geider, R. J., 235 Geigenberger, P., 51, 59 Geiman, T. M., 136 Gendrel, A. V., 127, 129, 136 Geng, F., 120, 136 Genger, R. K., 126, 136 Genthner, F. J., 290, 306 George, S. E., 306 Gerdol, V., 228, 234 Geremia, R. A., 273, 306 Ghaderian, Y. S. M., 86, 101 Gibon, Y., 59 Gibson, D. M., 310 Gidley, M. J., 19, 20, 51, 57 Gierl, A., 54 Giffard, S. C., 173, 174 Giguere, P., 318 Gilad, S., 179 Gilchrist, D. G., 170, 175 Gilles, K. A., 233 Gillespie, A. T., 242, 246, 306 Gillespie, J. P., 266, 275, 306 Giordano, A., 164, 176 Giroux, M. J., 44, 52, 58 Gish, B., 173 Gloer, J. B., 310 Gloer, K. B., 310 Glover, D., 172 Godoy, G., 281, 306 Goettel, M. S., 250, 266, 306, 308 Goff, S. P., 140 Goldberg, M., 262, 307 Goldman, G. H., 306 Goldman-Levi, R., 135 Golinski, P., 311 Gomez-Casati, D. F., 43, 52 Gonzalez, R., 139 Gordon, R., 189, 234 Gordon-Kamm, W. J., 179 Gordon-Weeks, R., 301 Gorman, M., 307 Gorovsky, M. A., 139 Goto, N., 200, 201, 234 Gotz, P., 319, 320 Gould, K. L., 175

329

Goulter, K. C., 176 Gouzy, J., 135 Goyal, R. K., 58 Gre´goire, J., 103 Grafi, G., 181 Graig, R. W., 176 Granier, F., 175 Graniti, A., 276, 306 Grant, J., 232, 238 Grasser, K. D., 137, 139 Graves, P. R., 178 Gray, J. C., 135 Greenberg, B. M., 175, 176 Greene, T. W., 52, 53 Greenhalgh, R., 320 Greenland, A. J., 47 Greenwood, J., 177 Gregg, P. E. H., 104 Gretz, M. R., 234, 239, 240 Grewal, S. I., 138 Gribskov, M., 102 Griesch, J., 296, 306 Griffith, S. M., 5, 52 Griffiths, D. J. F., 174 Griffiths, D. J., 172, 176 Grimaldi, M. A., 138 Gringorten, L., 312 Grotz, N., 78, 101 Groudine, M., 139 Grove, J. F., 283, 287, 288, 304, 307 Gruber, W., 99 Gruissem, W., 139 Grula, E. A., 250, 253, 274, 287, 303, 313, 315, 316 Grundel, M., 51 Grunstein, M., 140 Grzesik, M., 48 Gu, Y., 175 Guacci, V., 180 Guan, H. P., 33, 34, 48, 49, 52, 54, 59 Guerinot, M. L., 78, 101, 102 Guilfoyle, T. J., 138 Guiltinan, M. J., 48, 51 Gunawardena, A. N., 172 Gunkel, F. A., 272, 307 Gunning, A. P., 57 Guntuku, S., 173 Gupta, S., 58, 263, 274, 280, 282, 283, 285, 287, 288, 307, 310 Guschin, D., 113, 136 Gust, G., 239 Guy, P., 48

H Hackman, R. H., 262, 307 Haes, E. C. M., 94, 104 Haeses, A., 308 Hagan, I. M., 177

330

AUTHOR INDEX

Hagen, G., 138 Hagerthey, S. E., 185, 187, 231, 233, 236 Haines, B. J., 74, 75, 81, 101 Hajar, A. S. M., 99 Hajek, A. E., 245, 307 Hakvoort, J. H. M., 237 Hall, J. L., 83, 101, 105 Hall, M. A., 172 Hall, T. C., 137 Hall-Jackson, C. A., 175 Hamaker, B. R., 28, 52 Hamer, J. E., 267, 307 Hamiche, A., 139, 140 Hamil, R. L., 283, 287, 307 Hamilton, J. K., 233 Hamilton, W. D. O., 49 Hamon, R. E., 105 Han, X. Z., 28, 52 Hanada, K., 309 Hanawalt, P. C., 136 Hancock, J. T., 173, 177 Hannah, L. C., 22, 23, 52 Hansen, J. C., 110, 136, 138, 140 Harada, K., 54, 55 Haran, S., 256, 307 Haraux, F., 303 Harmer, S. L., 290, 303 Harn, C., 25, 26, 52, 54 Harper, J. F., 102 Hartwell, L. H., 146, 152, 175, 178 Hasezewa, S., 177 Hassan, A. E. M., 248, 249, 250, 269, 307, 308 Hassig, C. A., 140 Hatton, P., 104 Haug, A., 203, 206, 234, 235 Hausler, R., 55 Havlicek, V., 309 Hawes, C. R., 172 Hawker, J. S., 42, 43, 52, 53 Hawley, N., 226, 234 Hay, S. I., 202, 234 Hayasaka, S., 310 Haynes, P. A., 47 He, B. M., 309 Heath, M. C., 169, 170, 175, 178 Heberle-Bors, E., 176 Hedley, C. L., 50, 51, 60 Hegedus, D. D., 265, 308 Heimann, K., 235 Heinke, M., 237 Heip, C. H. R., 233, 235 Helentjaris, T., 49 Henderson, J., 138 Hendrich, B., 138 Hendriks, J. H. M., 59 Henikoff, S., 137 Hennig, L., 139 Henrichs, G., 55

Hepburn, H. R., 250, 251, 270, 308 Herbert, R. J., 147, 162, 163, 175 Herendeen, D. R., 180 Herman, P. M. J., 235 Hermeking, H., 173 Herreraestrella, A., 306 Herrmann, M., 299, 308 Hey, S., 54 Heymann, E., 260, 308 Heymann, K., 237 Higgins, C. E., 307 Higgins, M. J., 189, 190, 198, 214, 229, 234 Hill, D. A., 136 Hill, S. A., 41, 48 Hillerton, J. E., 251, 268, 308 Hinaje, M., 308 Hinchliffe, E., 52 Hirao, A., 153, 175 Hirschi, K., 102 Hitz, W. D., 5, 57 Hizukuri, S., 18, 52, 59 Ho, L., 140 Hoagland, K. D., 187, 190, 191, 204, 232, 234, 235, 239 Hodge, H. T., 285, 308 Hoe, H. S., 310 Hoeijmakers, J. H., 135 Hoekstra, M. F., 180 Hoekstra, M., 172 Hofman, K., 177 Hofmeyr, J. H. S., 40, 52 Hogan, P., 56 Hohn, T. M., 288, 308 Hojrup, P., 275, 308 Holden, D. W., 267, 307 Holland, A. F., 226, 234 Hollingworth, S., 47 Hollister, C. D., 236 Holmes-Davis, R., 137 Holt, D. C., 53 Holtzman, D. A., 172, 176 Honeywill, C., 231, 235, 236 Hoogenraad, N. J., 235 Hooykaas, P. J. J., 104 Horak, O., 104 Horn, P. J., 110, 131, 136 Horne, A. J., 94, 101 Horva´th, G. B., 177 Horwitz, H. R., 174 Hoshikawa, K., 2, 3, 6, 52 Houben, A., 134, 139 Houwing, E. J., 228, 234 Howe, L., 134 Howes, A. W., 100, 104 Hradec, H., 308 Hu, G., 306, 315 Huang, M. S., 139 Hulert, M., 291, 308 Hucl, P., 56

AUTHOR INDEX Hudson, B. P., 120, 136 Hudson, J., 178 Huet, V., 231, 236 Hughes, C. E., 7, 49 Hughes, R. G., 228, 234 Huguenin, R., 314 Huitson, S. B., 101, 102 Hulsmann, B. B., 138 Hulst, D., 60 Humber, R. A., 308 Hunag, M., 177 Hung, S. Y., 311 Hunt, B. A., 301, 316 Hunt, T., 170, 173 Huntley, R., 177 Hurion, N., 263, 308 Hurst, R. D., 177 Husak, M., 309 Hussain, H., 35, 52 Hutchins, R. A., 155, 156, 175 Hutchison, J. E., 312 Huxham, I. M., 289, 291, 308 Huxley, T. H., 226, 234 Hwang, C., 310 Hwang, Y. S., 307 Hylton, C. M., 51, 56 Hylton, C., 49, 50, 52 I Ibrahim, L., 301 Ichinoe, M., 312, 319 Ichinose, Y., 179 Iglesias, A. A., 43, 52, 57 Ignoffo, C. M., 304, 305, 307 Imbalzano, A. N., 136, 139 Imhof, A., 134, 135 Imoto, M., 292, 321 Imparl-Radosevich, J. M., 26, 49, 52 Inglis, G. D., 295, 308 Ingrouille, M. J., 71, 74, 101 Inze, D., 144, 149, 173, 174, 175, 176, 178, 179, 180 Isaka, M., 312 Isogai, A., 312, 319 Ito, H., 58 Ito, M., 140 Ito, T., 112, 113, 136 Itoh, Y., 309 Ivanov, V. T., 313 J Ju¨rgens, G., 109, 140 Jabs, T., 168, 175 Jackson, J. P., 123, 126, 127, 137 Jackson, M. B., 161, 172, 173, 174, 175 Jackson, S. P., 152, 153, 155, 174, 179 Jacobs, D., 306 Jacobsen, E., 54

331

Jacobsen, M. D., 161, 180 Jacobsen, S. E., 126, 127, 135, 137, 139 Jacobson, M. D., 144, 175 Jacquier, R., 303 Jaffre´, T., 64, 100, 101 Jahns, H. M., 239 James, M. G., 35, 48, 49, 50, 51, 52, 55, 57 James, P. J., 289, 290, 308 Jane, J., 16, 18, 19, 28, 29, 38, 53, 55, 60 Janick-Buckner, D., 172 Jansen, M. A. K., 159, 160, 175 Jansson, C., 58 Jarosch, R., 189, 234 Jasencakova, Z., 139 Jeddeloh, J. A., 130, 137 Jedrusik, M. A., 132, 137 Jeffcoat, R., 57 Jeffs, L. B., 281, 308 Jegorov, A., 282, 283, 288, 291, 293, 309, 320 Jen, W. C., 303 Jenkins, P. J., 16, 18, 20, 53 Jenner, C. F., 5, 7, 8, 11, 42, 43, 45, 46, 47, 49, 50, 52, 53, 59 Jenner, H. L., 49 Jensen, J. S., 287, 303 Jensen, R. G., 269, 303 Jenuwein, T., 122, 123, 137 Jerzmanowski, A., 114, 122, 131, 135, 137, 139 Jesus, B., 236 Jhee, E. M., 86, 101, 103 Jiang, C. -Z., 159, 160, 175 Jimenez, J., 172 Jin, P., 175 Jobling, S. A., 51, 57 Johal, G. S., 172 John, P. C. L., 149, 151, 176, 180 Johnson, A., 172 Johnson, C. A., 138 Johnson, L. M., 235 Johnson, P. E., 20, 21, 22, 23, 49, 53 Johnson, P. G., 231 Johnson, P., 50 Johnson, R. D., 299, 309 Jones, A. M., 147, 176 Jones, E. B. G., 232 Jones, G. A., 303 Jones, M. E. E., 179 Jones, R. J., 52 Jones, T. E. R., 232 Jorgensen, B. B., 218, 231 Jorgensen, R. A., 138 Joshi, L., 254, 257, 261, 263, 266, 269, 309, 318 Jost, J-P., 138 Joubes, J., 149, 150, 151, 176 Jug, R., 179 Juliano, B. O., 56 Jumars, P. A., 226, 236

332

AUTHOR INDEX

K Kacser, H., 41, 53 Kaczanowski, S., 139 Kadonaga, J. T., 136 Kagamizono, T., 283, 309 Kahn, D., 135 Kaijiang, L., 292, 294, 309 Kainuma, K., 48 Kakefuda, G., 57 Kakutani, T., 127, 130, 137, 139, 140 Kal, A. J., 112, 137 Kalla, R., 56 Kalpana, G. V., 140 Kalume, D. E., 56 Kamimura, M., 310 Kanaar, R., 135 Kanaoka, M., 312, 319 Kaneko, N., 312 Kaneko, Y., 24, 58 Kang, S. C., 258, 309, 310 Kangasja¨rvi, J., 178 Kanoh, J., 176 Kanost, M. R., 297, 309 Kanzaki, H., 309 Kaplan, L. A., 237 Karner, M., 221, 234 Kasemsuwan, T., 19, 28, 53 Kastan, M. B., 149, 176 Kasten, M. M., 164, 176 Kato, K. L., 60 Kato, L., 51 Kato, N., 104 Kaufmann, S., 138 Kavakli, I. H., 21, 53, 57, 58 Kawagoe, Y., 56 Kawakami, K., 319 Kawakita, K., 173 Kawamura, T., 234 Kawashima, A., 309 Kawazu, K., 289, 310 Kaya, H. K., 244, 319 Kazan, K., 168, 176 Kazy, S. K., 192, 234 Keegan, K. S., 152, 172, 176, 180 Keegstra, K., 50 Keeling, P. L., 11, 27, 31, 43, 45, 46, 48, 49, 50, 52, 53, 54, 55, 58 Keen, J. N., 315 Kehle, J., 134 Keil, B., 308 Kelleher, W. J., 320 Keller, C., 101 Kelly, J., 187, 231, 235, 236 Keohane, A. M., 134 Kerr, J. F. R., 147, 176 Kerry, B. R., 315 Kershaw, M. J., 292, 294, 295, 301, 308, 310 Kettunen, R., 178

Khachatourians, G. G., 244, 251, 263, 265, 267, 274, 275, 281, 301, 302, 308, 310 Khambay, B. P. S., 301 Khambay, B., 301, 308 Khavari, P. A., 117, 137, 140 Khochbin, S., 125, 137 Khul, H., 237 Kijne, J. W., 54 Kikumoto, S., 48 Kikyo, N., 136 Kilianczyk, B., 139 Kim, H. K., 263, 310 Kim, K. N., 48, 51 Kim, S. A., 102 Kim, S., 57 Kingston, R. E., 135, 138, 139 Kinzler, K. W., 173 Kirby, E. J. M., 55 Kirkham, J. H., 104 Kirkman, J. H., 104 Kirschner, M., 177 Kishimoto, M., 309 Kistler, H. C., 300, 314 Kittakoop, P., 312 Kiuchi, M., 298, 310 Klo¨sgen, R. B., 25, 54 Kladnik, A., 60 Kleczkowski, L. A., 43, 54, 59 Kleinkauf, H., 295, 310, 311, 313 Kloareg, B., 173 Klucinec, J. D., 48 Knight, B., 75, 101 Knight, M. E., 25, 52, 54 Kobayashi, A., 309 Kobayashi, R., 134, 136 Koch, M., 71, 101, 103 Kochian, L. V., 100, 102, 103 Kodaira, Y., 281, 282, 283, 287, 288, 289, 310 Kodama, M., 309 Kodama, S., 179 Kohlbrenner, W. E., 285, 304 Kohmoto, K., 309 Koizumi, N., 140 Kolaczkowska, M., 304 Koller, A., 47 Koltunow, A., 49 Komamine, A., 164, 165, 174, 179 Koncz, C., 180 Kong, Y. Y., 175 Kop, A. J., 239 Kopacek, P., 319 Kornberg, R. D., 135 Kornman, B. A., 233 Kosar-Hashemi, B., 54, 55 Koshland, D., 180 Kossmann, J., 15, 36, 37, 54 Kostecki, M., 311

AUTHOR INDEX Kottenhagen, M., 54 Kouike, H., 139 Kovac, K. A., 126, 130, 136 Kra¨mer, U., 68, 79, 83, 101, 104 Krasnoff, S. B., 282, 283, 285, 287, 289, 307, 308, 310 Kratochvil, B., 309 Krauss, V., 134 Krebs, J. E., 122, 137 Krek, W., 151, 176 Krivtsov, V., 238 Kroha, M. J., 304 Kromkamp, J., 185, 189, 217, 230, 238 Kropf, D. L., 173 Kruger, J. E., 12, 54 Kubo, A., 35, 36, 39, 54, 55 Kucera, M., 296, 310, 311 Kuerbitz, S. J., 149, 176 Kuhn, M., 314 Kuipers, A. G. J., 28, 54 Kumagai, A., 176, 177 Kumata, H., 312 Kuo, A. L., 167, 137, 176 Kuo, M. H., 137 Ku¨pper, H., 80, 92, 101, 102 Kuras, M., 139 Kurata, H., 312 Kuriki, T., 52 Kuromori, T., 158, 176 Kutach, A. K., 136 Kvarnheden, A., 173 Kwon, S. T., 310

L Lachner, M., 122, 123, 137 Lackie, A. M., 308 Lacomis, L., 135 Lafont, P., 290, 305 Lafontaine, D. G., 173 Lagnel, J., 312 Lam, E., 144, 148, 176, 177, 178 Lama, L., 235 Landa, Z., 305, 320 Landel, C. C., 140 Lane, D. P., 149, 170, 176 Lange, C., 291, 304, 310, 311 Langebartels, C., 173, 178, 180 Langeveld, S. M. J., 14, 37, 54 Langst, G., 113, 134, 135, 137 Lanning, S. P., 58 Lardon, F., 178 Larkins, B. A., 3, 51, 55, 179 Larsen, P. B., 103 Lasat, M. M., 77, 102, 103 Laspidou, C. S., 190, 235 Latge, J. P., 303, 314 Laughlin, M. J., 48 Laurent, B. C., 135, 136

333

Laurie, D. A., 49 Lavin, C., 239 Leake, J. R., 105 Leal, S. C. M., 263, 275, 311 Leas, S., 53 Leathers, T. D., 305, 307 Leaver, C. J., 162, 177 Leblanc, M., 104 Lechner, T., 134 Lecuona, R., 259, 311 Lee, C. H., 235 Lee, D. G., 309 Lee, E. A., 46, 52, 59 Lee, J., 100, 101, 153, 155, 156, 176 Lee, K. M., 140 Lee, S., 231 Lehman, A. R., 172 Lemercier, C., 137 Lending, C. R., 51 Lengauer, C., 173 Leopold, J., 314 Lepp, N. W., 103 Letham, D. L. D., 103 Letham, D. S., 180 Levado, E., 238 Levenstein, M. E., 136 Lewellyn, D. J., 176 Lewin, J., 231 Lewis, D. H., 270, 280, 311 Lewis, J., 172 Lewis, L. C., 280, 302 Lewis, R. J., 187, 235 Li, G., 109, 137 Li, L., 57 Li, P., 52 Li, Y. M., 100 Li, Y., 135 Li, Z. Y., 25, 26, 54 Libessart, N., 55 Libs, L., 181 Lichota, J., 139 Lichti, H., 314 Lienard, L., 60 Lilley, C. E., 54 Lin, T. P., 21, 55 Lind, J. L., 187, 189, 190, 235, 239, 240 Lindroth, A. M., 123, 126, 132, 137 Lindsay, H. D., 176 Lingle, S. E., 2, 6, 49, 55 Linker histones, , 137 Linnestad, C., 56 Lippman, Z., 136 Litt, M. D., 123, 137 Little, C., 236 Liu, D., 175 Liu, H., 219, 235 Liu, J. C., 283, 286, 311 Liu, K. C., 57 Liu, L., 60

334

AUTHOR INDEX

Liu, W. Z., 283, 285, 286, 311 Liu, X., 173 Lloyd, J. R., 50, 59 Lloyd, J., 15, 36, 37, 50, 54, 59 Lloyd-Thomas, D., 71, 102 Lock, M. A., 237 Locke, M. A., 221, 234 Logie, C., 136 Logrieco, A., 287, 311 Logrieco, K., 287, 311 Loidl, P., 134 Lomako, J., 39, 47, 55, 58 Lomako, W. M., 39, 47, 55, 58 Lombi, E., 68, 71, 83, 97, 102, 104, 105 Long, B., 232 Long, X. X., 66, 68, 102 Loosli, H. R., 314 Lopes, M. A., 3, 51, 55 Lopez-Girona, A., 156, 172, 176 Lorch, Y., 135 Loss, S. P., 11, 55 Losson, R., 138 Loutelier, C., 291, 292, 310, 311 Lowe, K. S., 179 Lu, T. J., 20, 55 Lu, X., 180 Lue, W. L., 61 Lundgren, K., 151, 177 Luo, M., 49 Luthi, E., 54 Lyon, A. J. E., 101 Lysenko, O., 296, 298, 311

M Ma¨ser, P., 78, 102 Me´sza´ros, T., 149, 150, 151, 177 Maa, J. P.-Y., 226, 235 Maathuis, F. J. M., 102 Mabud, A., 305 MacIntyre, H. L., 185, 235 MacMasters, M. M., 49, 60 Macnair, M. R., 66, 68, 72, 73, 80, 81, 99, 101, 102 Macnair, M.R., 73, 80, 102 Maddelein, M. L., 55 Maddelein, M.-L., 50 Madry, N., 287, 311, 313 Madsen, K. N., 216, 235 Magyar, Z., 177 Mahadevan, L. C., 135 Mahmoudi, T., 137 Maimala, S., 286, 311 Maitland, T. C., 234 Mak, T. W., 175 Makishima, T., 179 Makkerh, J., 177 Malaisse, F., 67, 103 Maleszewski, M., 122, 137

Malik, M., 100 Manetas, Y., 103 Maniania, N. K., 313 Maniatis, T., 122, 138 Mann, D. G., 184, 235, 237, 239 Mann, M., 140 Manners, J. M., 176 Mant, A., 52 Manzenrieder, H., 227, 235 Marahiel, M. A., 295, 318 Marchbank, A., 179 Marcondes, C. B., 304 Marschner, H., 77, 84, 89, 103, 105 Marshall, J., 59 Martens, S. N., 71, 82, 84, 85, 90, 99, 100, 103 Martienssen, R. A., 136, 140 Martin, C., 49, 50, 51, 52, 59, 61 Martinez-Balbas, M. A., 113, 137 Martinez-Yamout, M. A., 136 Maruniak, J., 311 Matha, V., 283, 285, 288, 292, 298, 305, 308, 309, 311, 319, 320 Matsuda, T., 54, 55 Matsui, K., 179 Matsumoto, K., 309 Matsumoto, T., 315 Matsuoka, S., 153, 175, 177 May, B., 317 Mazet, I., 283, 286, 311 Mazzella, L., 231 Mazzolini, L., 172 McCabe, P. F., 162, 177 McCallum, C. M., 137 McCave, I. N., 219, 234, 236 McClintock, B., 3, 55 McConville, M. J., 194, 196, 197, 202, 204, 205, 206, 210, 213, 235 McCorkindale, N. J., 308 McCoy, C. W., 244, 285, 311, 312, 313 McGowan, C. H., 172, 177 McGrath, S. P., 95, 99, 101, 102, 103, 105 McInnis, A. G., 320 Mckean, A. L., 52 McLaughlin, M. J., 105 McMahon, K., 179 McPherson, A. E., 53 Mead, A., 100 Medlin, L. K., 184, 235 Meerts, P., 71, 72, 74, 99, 103 Meharg, A. A., 101 Mehlert, A., 238 Meijer, L., 173 Meijer, M., 177 Meister, A., 139 Meister, M., 139 Melzer, M. J., 254, 302 Mench, M., 94, 103 Mengers, M., 161, 177

AUTHOR INDEX Mercadier, J. L., 303 Metzger, J. W., 58 Meyer, F. D., 58 Meyer, F. K., 66, 103 Meyer, H. E., 58 Meyer, M., 104 Meyer, P., 109, 137 Meyerowitz, E. M., 108, 121, 138, 140 Meyn, M. S., 153, 176, 177 Michael, W. M., 158, 177 Michel, B., 269, 312 Middaugh, D. P., 290, 306 Middleburg, J. J., 187, 201, 221, 235 Mikawa, T., 309 Mikhailov, A., 147, 177 Mikhaleva, I. I., 313 Miles, A., 222, 235 Millar, J. B. A., 177 Millar, S., 2, 3, 6, 51 Miller, C., 135 Miller, D. C., 235 Miller, E. A., 235 Miller, J. D., 320 Miller, M. E., 22, 55 Milne, R., 278, 312 Mirsky, A. E., 134 Misikova, S., 314 Misklczi, P., 177 Mitamura, O., 234 Mitchell, D. A., 172 Mitchell, D. L., 175 Mitchell-Olds, T., 101 Mitchener, H. J., 232 Mitchener, H., 238 Mittler, R., 147, 161, 177 Miura, Y., 173 Mizuguchi, G., 113, 138, 139 Moar, W. J., 71, 84, 85, 87, 99 Moder, W., 173 Modesert, O., 176 Moens, T., 235 Moffitt, C. M., 93, 100 Mohlmann, T., 24, 55 Moller, B. L., 50 Mollier, P., 283, 312 Mondesert, O., 172 Montllor, C., 307 Moon, T. W., 260, 312 Moore, A. L., 306 Moore, D., 315 Moorhouse, E. R., 310 Mopper, K., 240 Morel, J. B., 130, 138 Morel, J. L., 104 Morell, M. K., 2, 14, 26, 33, 54, 55 Morell, M., 2, 14, 26, 57, 58 Moreno, G. T., 141 Moreno, S., 180 Moretti, A., 311

335

Morgan, D. O., 151, 175, 177 Moriello, V. S., 235 Morimoto, S., 309 Morozova, N., 181 Morrice, N., 175 Morris, B. A., 173 Morris, V. J., 57 Morrison, R. S., 66, 67, 100, 103 Moss, S. B., 176 Motawia, M.S., 50 Motawia, S., 50 Mottram, J. C., 256, 312 Mouille, G., 36, 48, 54, 55 Mount, D. W., 138 Mourelatou, M., 165, 166, 167, 177 Mourrain, P., 138 Muchardt, C., 140 Mueller, P. R., 151, 177 MuForster, C., 54 Mukerjea, R., 25, 55 Mulheim, C., 304, 310 Muller, A., 138 Muller, J., 134 Mulvaney, P., 234 Mummenhoff, K., 66, 69, 88, 103 Munakata, K., 137 Mundt, K. E., 157, 177 Munoz, F. J., 48 Mur, L. R., 237 Murakami, T., 309 Murakoshi, S., 281, 282, 288, 312, 319 Muray, J. M., 177 Murchardt, C., 117, 119, 138 Murfett, J., 124, 138 Muro-Pastor, M. I., 139 Muroi, M., 285, 290, 312 Murphy, J. F., 100 Murphy, J. M., 273, 312 Murray, F. R., 176 Murray, J. A. H., 149, 150, 161, 164, 177 Murray, J. M., 176 Murray, K., 122, 138 Myers, A. M., 11, 33, 36, 49, 50, 51, 52, 55, 57 Myers, A., 48 Myklestad, S., 203, 206, 234, 235, 239

N Naegeli, M. W., 237 Nagata, T. N., 161, 177 Nakamura, T., 26, 60 Nakamura, Y., 7, 8, 27, 32, 33, 34, 36, 54, 55, 59, 179 Naldrett, M. J., 49, 56 Napier, J., 54 Napoli, C. A., 138 Narlikar, G. J., 138 Narmada, K., 49

336

AUTHOR INDEX

Naumann, K., 134 Nedwell, D. B., 238 Neill, S. J., 148, 162, 168, 173, 177 Nelson, J. O., 318 Nelson, O. E., 56, 59 Nemoto, Y., 177 Neuhaus, H. E., 55 Neuteboom, L. W., 104 Neville, A. C., 250, 312 Newport, J., 158, 177 Newton, I., 101 Newton, R., 301 Ng, H. H., 113, 138, 141 Ng, L., 302 Ni, W. Z., 102 Nichols, P. D., 232 Nichols, S. E., 56 Nicolaus, B., 214, 235 Nicoletti, R., 232 Nieder, M., 57 Nielsen, A. L., 133, 138 Nieman, K., 103 Nigg, E. A., 151, 176 Nikolskaya, A. N., 299, 312 Nilanonta, C., 287, 312 Nilsson, C., 239 Nilsson, F., 56 Nilsson, P., 235 Nishi, A., 34, 55 Nishimoto, T., 179 Nishino, E., 309 Nishio, T., 312 Nnakumusana, E. S., 246, 312 Nolan, R. A., 278, 305 Noma, K., 123, 138 Nordmark, T. S., 39, 56 Nordstrom, W., 173 Nougarede, A., 179 Nowell, A. R. M., 226, 232, 236 Nurse, P., 144, 149, 151, 166, 172, 174, 175, 178, 179

O O’Brian, D., 231 O’Brien, L., 51 O’Connell, M. J., 146, 153, 158, 178 O’Farrell, P. H., 145, 146, 151, 152, 174, 178 Oakley, E. J., 133, 138 Ocampos, M., 302 Odier, F., 289, 313 Offler, C. E., 60 Ogas, J., 121, 138 Ojcius, D. M., 287, 313 Okano, H., 139 Okenfull, E. A., 177 Okita, T. W., 48, 53, 57, 58 Okita, T., 52 Oku, T., 312

Olczak, K., 135 Olejnicek, J., 311 Olive, M., 43, 56 Olsen, C. E., 50 Olsen, O.-A., 3, 4, 49, 50, 54, 56, 59 Omoto, C., 285, 313 Ong, G. T., 304, 320 Ono, Y., 309 Onufryk, C., 138 Onyekwere, O., 176 Oostergetel, G. T., 17, 56 Oppedijk, B. J., 180 Oppenheim, A., 307 Orchard, C. B., 175 Ormord, J. C., 175 Ortiz, J. A., 138 Osborne, L. S., 305 Osburne, B. A., 179 Oscarsson, M., 7, 56 Osman, F., 176 Ossipow, V., 136 Otani, H., 309 Oulad-Abdelghani, M., 138 Ovchinnikov, Y. A., 287, 313 Overmyer, K., 168, 178

P Paegle, E. S., 172 Pais, M., 288, 292, 304, 305, 306, 310, 313, 319 Palmer, G. H., 8, 56 Palmer, J., 140 Pan, D., 36, 56 Pan, W., 320 Panaccione, D. G., 312 Pandey, R., 125, 138 Panico, A., 235 Papierok, B., 271, 313 Papoulas, O., 139 Paracelcus, , 226, 236 Parchure, T. M., 226, 236 Parekh, B. S., 122, 138 Paris, S., 274, 313 Park, H. -J., 173 Park, S., 309 Park, W., 101 Parker, M. L., 14, 53, 56, 57 Parker, M., 49 Parker, R., 232 Parkes, R. J., 239 Parkkonen, T., 56 Passow, U., 240 Paterson, D. M., 185, 186, 187, 189, 191, 216, 217, 218, 219, 221, 224, 225, 228, 231, 232, 233, 234, 235, 236, 238, 239, 240 Paterson, I. C., 261, 272, 296, 301, 313, 315, 316

AUTHOR INDEX Paton, A., 65, 103 Patrick, J. W., 60 Patron, N. J., 7, 53, 56 Paulovich, A. G., 146, 178 Pawlikowska, K., 139 Payne, T., 49 Pazin, M. J., 136 Peacock, W. J., 49, 136 Pearce, D. M., 172 Pearce, R. B., 9, 57 Pearse, A. G. E., 269, 313 Peberdy, J. F., 311, 315 Peeters, H., 287, 313 Pekrul, S., 250, 274, 313, 316 Peltier, J. B., 24, 56 Pence, N. S., 77, 78, 81, 103 Pendland, J. C., 245, 302, 311 Peng, C. Y., 178 Peng, L., 37, 56 Peng, M. S., 33, 56 Percival, E., 206, 236 Perdon, A. A., 56 Peres, A., 177 Perez, C. M., 6, 56 Perillo, G. M. E., 232 Perkins, R. G., 194, 195, 199, 201, 202, 203, 220, 236 Perkins, R., 238 Perry, M. W., 55 Persans, M. W., 79, 102, 103 Peterson, C. L., 135, 136, 137, 138 Petit, D., 99, 102, 104 Pettko´-Szandtner, A., 177 Phelan, M. L., 114, 138 Phillips, J., 239 Piccolo, M. C., 232 Pickett-Heaps, J. D., 187, 189, 233 Pickhardt, M., 104 Pien, F. M., 57 Pierre, J., 104 Pikaard, C. S., 138 Pilling, E., 51 Pinas, J. E., 104 Pinnock, D. E., 281, 305 Pinto, A. D. S., 258, 313 Pinto, H., 302 Pittman, J. K., 105 Piwnica-Worms, H., 178, 180 Plans, M., 231 Plante, M., 173 Plaxton, W. C., 44, 56 Player, M. A., 238 Plug, A. W., 176 Plunkett, B. S., 176 Podesta, A., 138 Podstolski, W., 135 Pollak, L., 60 Pollard, A. J., 71, 72, 85, 101, 103 Pollard, K. J., 121, 138

337

Pontier, D., 169, 178 Pontis, H. G., 24, 56 Poot, R. A., 113, 138 Pople, M., 287, 288, 304, 307 Poprawski, T. J., 289, 313 Porte, J., 177 Poschenrieder, C., 104 Potter, R. H., 56 Poulsen, N. C., 189, 237 Pozueta-Romero, J., 24, 48, 56 Preiss, J., 20, 33, 43, 44, 48, 52, 54, 55, 56, 57, 59 Preusser, E., 51 Prevost, M. C., 303 Prioul, J. L., 49, 52 Prives, C., 179 Prymakowska-Bosak, M., 131, 132, 139 Przewloka, M. R., 132, 137, 139 Psaras, G. K., 80, 103 Pusch, M., 221, 237 Puschenreiter, M., 97, 104 Q Qin, J., 173 Quatrano, R. S., 239 Quick, W. P., 55 Quinn, J., 135 Quiot, J. M., 289, 291, 305, 312, 313, 319 Quiqierez, C., 314 R Ro¨mheld, V., 105 Radosavljevic, M. A., 53 Raff, K. D., 180 Raff, M. C., 144, 148, 175, 178, 180 Raff, M., 172 Rahman, M. A., 236 Rahman, S., 54, 55 Raikhel, N. V., 50 Ral, J. P., 60 Raleigh, J. M., 153, 158, 178 Ramakrishnan, A., 52 Ramon, A., 131, 139 Ramstad, P. E., 2, 60 Ranalli, T., 134 Ranallo, R., 140 Randazzo, G., 311 Randsholt, N., 139 Rankin, B., 304 Ransom, R., 134 Raskin, I., 94, 101, 104 Rassoulzadegan, F., 221, 234 Rathjen, A. J., 47, 53 Ravallec, M., 305 Ray, P. M., 50 Raymond, M., 90, 104 Rebers, P. A., 233 Rebetzke, G. J., 57

338

AUTHOR INDEX

Reeves, R. D., 65, 68, 99, 100, 101, 103, 104 Reichert, E. T., 12, 13, 14, 57 Reichfeld, J. P., 170, 178 Reichstein, T., 104 Reinberg, D., 138 Reithmuller, R., 227, 228, 237 Remboutsika, E., 138 Renaudin, J. -P., 176 Renwick, J. A. A., 307, 310 Reski, R., 12, 57 Resurreccion, A. P., 56 Reuter, G., 134 Revsbech, N. P., 219, 237 Reyes, F., 259, 314 Reyes, J. C., 109, 139 Reynolds, A., 173 Reynolds, M. P., 46, 57 Reynolds, S. E., 289, 290, 298, 301, 308, 310, 314, 315 Reyss, A., 49, 52 Rhind, N., 148, 152, 153, 155, 157, 172, 178 Riba, G., 289, 305, 311, 312, 314 Ribeiro, L., 236 Rice, J. C., 123, 139 Richard, A. L., 305 Richard, J. L., 284, 314, 316 Richard, P., 231 Richards, E. J., 108, 112, 117, 118, 130, 137, 138, 140 Richards, R. A., 46, 57 Richman, R., 179 Ridout, M. J., 17, 57 Rieder, L., 177 Riekel, C., 60 Riethmu¨ller, R., 238 Ring, S., 49 Riou-Khamlichi, C., 177 Rist, C. E., 60 Ritieni, A., 311 Rittmann, B. E., 190, 235 Rizzo, N. W., 306, 318 Roach, P. J., 37, 57 Robert, P. H., 313 Robert, P., 289, 305, 314 Roberts, D. W., 244, 246, 273, 276, 278, 285, 288, 289, 292, 293, 294, 300, 306, 307, 309, 310, 314, 316, 317, 318 Roberts, J., 173 Roberts, K., 172 Robertson, D. S., 52, 58 Robertson, M., 49 Robinson, B. H., 95, 96, 100, 104 Robinson, H. L., 51 Robinson, R. K., 250, 314 Roby, D., 178 Robyt, J. F., 53, 55 Rodriguez, M., 176 Roemer, S.C., 234 Roepstorff, P., 56, 308

Rogers, H. J., 175, 179 Romano, I., 235 Rondet, P., 179 Ronemus, M. J., 126, 130, 139 Rose, S., 135 Rosewich, U. L., 300, 314 Rosowski, J. R., 234 Rossnagel, B. G., 56 Rossner, U., 51 Rota, C. A., 302 Roth, S. Y., 125, 134, 139 Rothblum-Oviatt, C. J., 178 Rothschild, M., 93, 104 Rotman, G., 179 Round, F. E., 184, 237 Routier, F., 50 Rowley, R., 158, 178 Rozema, J., 159, 178 Ruddy, G. K., 232 Rudi, H., 50 Ruegger, A., 288, 314 Ruhf, M. L., 116, 139 Ruland, J., 175 Russell, P., 148, 151, 152, 153, 155, 157, 172, 176, 177, 178, 179 Russell, R. G., 136 Ryan, C. E., 178 Ryerson, D. E., 170, 178

S Su¨ndback, K., 235 Sacchi, A., 102 Saedler, H., 54 Safford, R., 57 Sakai, N., 309 Saker, L. R., 175 Sakulsingharoj, C., 58 Salamone, P. R., 44, 53, 57 Salanoubat, M., 175 Salis, P., 99 Salkeld, P. N., 239 Salt, D. E., 76, 94, 96, 101, 102, 103, 104 Salvador, R. J., 9, 57 Samsinakova, A., 253, 314 Samson, R. A., 244, 246, 312, 314 Samuel, M. S., 55 Samuels, M. L., 138 Samuels, R. I., 263, 271, 272, 288, 289, 290, 292, 293, 294, 314, 315 Sanchez, L. M., 173 Sanchez, Y., 153, 179 Sandaltzopoulos, R., 136, 140 Sandermann, H., 173, 178 Sanders, D., 102 Sano, H., 140 Sar, P., 234 Sardet, C., 138 Sarnowski, T. J., 117, 120, 121, 139

AUTHOR INDEX Sasabe, M., 179 Sathish, P., 58 Sato, T., 312 Satoh, H., 54, 55, 59 Satorre, E. H., 46, 57 Saumier, S., 104 Saumitou-Laprade, P., 99, 102 Saunders, K., 239 Saurin, A. J, 139 Savitsky, K., 153, 179 Sawa., H., 139 Sawant, S. S., 231 Sayre, K. D., 57 Scazzocchio, C., 139 Schu¨beler, D., 123, 139 Schat, H., 98, 100, 104 Schell, J., 180 Scheller, K., 278, 302 Schickler, H., 307 Schilperoort, R. A., 180 Schlegel, H. G., 91, 104 Schmalstig, J. G., 5, 57 Schmidt, T., 104 Schnable, P. S., 58 Schnitzler, G. R., 138 Schnitzler, G., 135 Schrank, A., 302, 313 Schreiber, S. L., 140 Schroeder, J. I., 102 Schubert, I., 134, 139 Schuh, W. W., 56 Schulin, R., 101 Schultz, T. F., 237 Schulz, I., 134 Schulze, E., 132, 137 Schurmann, P., 48 Schwall, G. P., 35, 57 Schwarting, A. E., 320 Schwartz, C., 74, 104 Schwartz, L. M., 147, 179 Schwarz-Sommer, Z., 54 Scoffin, T. P., 226, 237 Scott, M. P., 57 Scrase-Field, E. S., 51 Screen, S. E., 255, 269, 272, 273, 276, 300, 301, 306, 315, 316, 318 Screen, S., 260, 306, 315 Seale, R., 51, 52 Sedmera, P., 309 Segers, R., 273, 315 Seguines, M., 231 Sehnke, P. C., 45, 57 Seib, P. A., 53 Seigneurin-Berny, D., 137 Self, R. F. L., 236 Selinger, D. A., 138 Sen, A. K., 234 Seo, B., 34, 57 Seroˆdio, J., 187, 202, 203, 237

339

Sera, T., 140 Servant, F., 135 Serwe, M., 58 Seyoum, E., 246, 315 Shah, Z. H., 177 Shanahan, P., 172 Shannon, J. C., 24, 48, 49, 51, 57 Shannon, T. W., 235 Sharp, R. E., 90, 105 Shaw, A. S., 178 Shaw, J. J., 100 Shaw, J. R., 52 Shayler, S. A., 236, 238 Sheldrick, K. S., 172, 174 Shen, J. J., 19, 53 Shen, X., 112, 131, 139 Shen, Z. G., 101, 105 Shewry, P. R., 2, 14, 58 Shewry, P., 54 Shi, W. Y., 102 Shi, Y. C., 57 Shieh, S. Y., 153, 179 Shiloach, J., 140 Shimamune, T., 55 Shimizu, S., 263, 297, 315 Shininger, T. L., 164, 179 Shinshi, H., 180 Shiragami, N., 312 Shiraishi, K., 179 Shiraishi, T., 179 Shitaozono, T., 59 Shnitka, T. K., 259, 260, 315 Shomura, A., 59 Shore, R. F., 101 Siddique, K. H. M., 55 Sidoli, C. M. D., 99, 103 Sidransnsky, D., 176 Sif, S., 120, 138, 139 Sigalat, C., 303 Sigee, D. C., 238 Sikka, V. K., 21, 23, 44, 58 Simanis, V., 149, 177, 179 Simmons, S. R., 11, 58 Simon, L., 177 Simpson, M., 137 Singh, D. G., 37, 58 Singh, R., 5, 58 Singh, S. P., 234 Singh, Y. D., 49 Singletary, G. W., 30, 42, 46, 54, 57, 58 Sinke, J. J., 231 Sinsabaugh, R. L., 221, 237 Sirotkin, A., 136 Sivak, M., 43, 52, 57 Siwek, K., 53 Skoultchi, A. I., 136 Skurat, A. V., 37, 57 Slafer, G. A., 46, 57 Slattery, C. J., 57

340

AUTHOR INDEX

Sloman, I. S., 289, 290, 298, 315 Slusarczyk, J., 139 Smart, C. M., 147, 179 Smidansky, E. D., 45, 58 Smirnoff, N., 71, 74, 80, 90, 101, 102, 104 Smith, A. M., 21, 25, 27, 28, 36, 37, 48, 49, 50, 51, 52, 56, 58, 59, 61 Smith, A. R., 172 Smith, D. A. S., 104 Smith, D. G., 320 Smith, D. J., 186, 187, 191, 193, 194, 195, 196, 197, 199, 200, 202, 203, 210, 215, 216, 217, 219, 220, 221, 237, 238 Smith, E. E., 48 Smith, F., 233 Smith, G. C. M., 152, 179 Smith, J. A. C., 99, 101 Smith, R. D., 101, 104 Smith, R. J., 253, 315, 316 Smith, S. E., 102 Smith, S. M., 61 Smith, S. W., 179 Smith-White, B., 57 Smithson, S. L., 256, 316 Smythe, C., 37, 58, 175 Snow, G. C., 236 Sofield, I., 9, 11, 58 Soledade, M., 244, 316 Solomon, M. J., 136 Somero, I., 260, 316 Somerville, C. R., 55 Somerville, C., 138 Soni, R., 177 Soper, R. S., 242, 303 Soppe, W. J., 129, 139 Sorrell, D. A., 150, 151, 174, 177, 179 Southard, J. B., 236 Souza, A. E., 312 Spalding, A., 94, 104 Spector, I., 237 Springer, J. P., 289, 296, 316 Spurck, T. P., 237 St-Michel, C., 173 St. Leger, R. J., 244, 245, 246, 250, 251, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 265, 266, 267, 268, 269, 270, 271, 272, 273, 275, 276, 283, 296, 297, 300, 301, 304, 306, 307, 309, 310, 314, 315, 316, 317, 318 Staats, N., 187, 191, 192, 193, 194, 195, 196, 197, 198, 199, 202, 205, 208, 209, 215, 219, 220, 221, 233, 237 Stachelhaus, T., 295, 318 Stal, L. J., 192, 193, 194, 195, 196, 197, 199, 201, 202, 205, 206, 207, 208, 209, 210, 213, 214, 220, 231, 232, 233, 237, 239, 240

Stallcup, M. R., 122, 139 Stals, H., 179 Staples, R. C., 306, 317 Steadman, J. R., 306 Steiner, J. R., 307 Steinrauf, L. K., 287, 290, 318 Steller, H., 148, 179, 180 Stemmer, C., 139 Sterner, O., 244, 278, 287, 301 Steward, N., 130, 140 Stewart, B. W., 148, 179 Stinard, P. S., 34, 36, 51, 58 Stinard, P., 56 Stirling, J. L., 258, 318 Stitt, M., 40, 41, 58, 59 Stoger, G., 104 Stokes, T. L., 137 Stoppel, R. D., 104 Strasser, H., 281, 308, 318 Straub, T., 135 Strongman, D. B., 280, 287, 318 Strunz, G. M., 318 Stuchlik, J., 309 Stuppner, H., 318 Stuurman, N., 54 Sudarsanam, P., 117, 140 Sugiyama, M., 165, 179 Suh, D. S., 310 Suka, N., 122, 140 Suka, Y., 122, 140 Sullivan, T. D., 24, 49, 58 Summers, P. S., 57 Sun, C. M., 304 Sun, C., 33, 58 Sun, H., 229, 231, 238 Sun, Y., 151, 179 Sundba¨ck, K., 239 Sundh, J., 221, 238 Sutherland, T. F., 191, 196, 197, 228, 231, 238 Suzuki, A., 283, 288, 293, 312, 318, 319 Suzuki, K., 152, 179, 180, 312 Sweder, K. S., 136 Swiezewski, S., 139 Szczepanik, M., 278, 302 Sze, H., 102

T Tagle, D. A., 179 Taguchi, H., 318 Tahaoglu, E., 180 Takatsuki, A., 312 Takeda, C., 34, 59 Takeda, Y., 29, 33, 34, 59 Takei, Y., 179 Talaga, P., 55 Talbert, L. E., 58 Talbot, N. J., 244, 319

AUTHOR INDEX Talke, I. N., 102 Tamai, K., 179 Tamkun, J. W., 135, 137, 139 Tamkun, J., 135 Tamura, S., 288, 312, 318, 319 Tanada, Y., 244, 319 Tanaka, H., 164, 179 Tanaka, N., 55 Tanaka, Y., 170, 179 Tandecarz, J., 37, 48 Tanticharoen, M., 312 Tartar, A., 311 Tateoka, T., 15, 59 Tatge, H., 28, 49, 50, 59 Taunton, J., 122, 140 Taya, Y., 179 Tayal, A., 57 Taylor, A., 303 Taylor, E. M., 176 Taylor, I. S., 215, 217, 218, 220, 221, 238 Taylor, I., 185, 191, 208, 209, 210, 219, 221, 238 Taylor, J. W., 271, 301 Tchieu, J., 102 Tempst, P., 135, 138, 141 Tepfer, S. S., 159, 179 Terai, H., 234 Termaat, G. R., 233 Termaat, R., 238 Tetley, L., 312 Tetlow, I. J., 21, 44, 51, 59 Tewari, J. P., 289, 296, 301 Thake, B., 198, 233 Thebtaranonth, Y., 312 Thistle, D., 232 Thoma, R. S., 178, 179 Thomine, S., 102 Thompson, D. B., 17, 48, 59 Thorbjornsen, T., 21, 23, 50, 59 Thornton, D. C. O., 217, 238 Thorstensen, T., 134 Tian, L., 124, 140 Ticknor, C., 139 Tien, C. J., 221, 238 Tiessen, A., 44, 59 Tilstone, G. H., 102 Tisserat, N. A., 319 Tissier, A., 175 Tiwari, S. C., 50 Tjaden, J., 55 Toczyski, D. P., 178 Tolhurst, T. J., 225, 231, 233, 236, 238, 239 Tollenaar, M., 46, 59 Tomlinson, K. L., 34, 59 Tomlinson, K., 50 Tong, K. I., 302 Traber, R., 288, 320 Trakulnaleamsai, S., 312 Traris, M., 303

341

Traunspurger, W., 237 Treich, I., 117, 120, 140 Tremethick, D. J., 136 Trethrowan, R., 57 Trexler, M. B., 232 Tsai, C.-Y., 59 Tse, C., 121, 140 Tsuchitani, Y., 315 Tsukiyama, T., 112, 113, 136, 137, 138, 140 Tucker, M. R., 49 Tunlid, A., 244, 319 Tuominen, H., 178 Turgeon, B. G., 298, 299, 300, 320 Turlington, L. W., 285, 319 Turner, B. M., 134, 138 Turner, W. B., 282, 283, 285, 300, 319 Typas, M. A., 320 Tyrrell, D., 278, 312, 319 U Uchimaya, H., 180 Uchimiya, H., 180 Ugalde, T. D., 5, 6, 8, 53, 59 Ulhoa, C. J., 313 Umeda, M., 151, 176, 180 Umemoto, T., 27, 59, 61 Underwood, G. J. C., 185, 186, 187, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 202, 203, 210, 215, 216, 217, 219, 220, 221, 225, 230, 236, 237, 238, 239 Underwood, N. L., 307 Uziel, 179 V Va´zquez, M. D., 80, 104 Vago, C., 313 Vainstein, M. H., 302, 313 Vakili, N. G., 280, 319 Van Breusegem, F., 180 van Bruggen, E. F. J., 17, 56 Van Camp, W., 173, 180 van de Staaij, J., 178 Van de Wal, M., 38, 48, 60 Van Den Boom, V., 135 Van den Hoek, C., 184, 239 Van den Koornhuyse, N., 50 van der Grinten, E., 231 van der Helm, D., 316 van der Zaal, B. J., 78, 104 van Duijn, B., 180 van Duyl, F. C., 187, 195, 202, 217, 220, 221, 239 van Herwaarden, A. F., 57 Van Isacker, N., 71, 72, 74, 103 Van Montague, M., 173, 178, 179, 180 van Vliet, C., 235 van Wegen, S., 49

342

AUTHOR INDEX

van Wijk, K. J., 56 van Wijk, R., 54 Vanagaite, L., 179 Vangronsveld, J., 103 Vanmontagu, M., 306 Vardy, K. A., 59 Varga-Weisz, P. D., 112, 113, 135, 138, 140 Vargas, M., 57 Varum, K. M., 203, 239 Vaucheret, H., 138 Venkatsubbaiah, P., 287, 289, 296, 319 Vennik, M., 54 Verbruggen, N., 99 Verbsky, M. L., 108, 112, 117, 118, 140 Verdel, A., 137 Verducci, J., 303 Verhoeven, T., 49 Verkade, H. M., 178 Verkleij, J. A. C., 100, 104 Vermeulen, W., 135 Vernoux, T., 178 Verrijzer, C. P., 135, 137 Vesonder, R. F., 320 Vey, A., 283, 286, 289, 291, 293, 303, 304, 305, 310, 311, 313, 319 Viale, A. M., 56 Vila, S. B., 306 Vilcinskas, A., 283, 285, 288, 289, 291, 292, 296, 297, 298, 306, 319, 320 Vilhar, B., 5, 60, 175 Villand, P., 54, 59 Villareal, R. M., 56 Villbrandt, M., 237 Vincken, J.-P., 60 Vining, L. C., 277, 282, 320 Visconti, A., 283, 286, 287, 320 Visser, R. G. F., 54 Visser, R., 48, 60 Voesenek, L. A. J. C., 172 Voetberg, G. S., 90, 105 Vogelstein, B., 173, 176 Volterra, L., 232 von Dohren, H., 295, 310 von Euw, J., 104 Vongs, A., 127, 140 Vooijs, R., 98 Vranova, E., 168, 180 Vrinten, P. L., 26, 60

W Wa¨ngberg, ., 239 Wade, P. A., 136 Wagh, A. B., 231 Wagner, D., 121, 140 Wahlman, M., 288, 320 Waigh, A. T., 19, 60 Waite, D. N., 53 Waite, D., 49, 50

Wakeham, A., 175 Walker, P.L., 89, 99 Wall, M. A., 93, 100 Wallace, J. C., 51 Walsh, W. V., 176 Walter, J. F., 305 Walter, P., 172 Walters, A. J., 306 Walton, G. S., 282, 284, 301 Walton, J. D., 134, 273, 295, 299, 312, 320 Walworth, N., 176, 177 Wanat, J., 51 Wang, C. S., 276, 320 Wang, H. L., 5, 9, 60 Wang, H. M., 140 Wang, H., 150, 180 Wang, M., 167, 180 Wang, T. L., 18, 50, 60 Wang, W., 112, 120, 140, 141 Wang, X. J., 138 Wang, Y. J., 27, 30, 60 Wang, Y., 173, 189, 190, 191, 204, 239 Wang, Z. H., 18, 50 Wang, Z., 60 Ward, J. M., 102 Wardlaw, I. F., 9, 46, 47, 49, 51, 58, 60 Warren, M. A., 306 Wartburg, V. A., 288, 314, 320 Wasserman, B. P., 52, 54 Wat, C.-K., 282, 320 Waterborg, J. H., 124, 140 Waterfield, N. R., 244, 320 Watkins, J. E., 100 Watnabe, M., 179 Watson, J., 231, 238 Watson, K. L., 135 Watson, S. A., 2, 4, 8, 60 Wattebled, F., 26, 60 Watts, J., 232 Wedde, M., 296, 297, 306, 319 Wei, N., 157, 180 Weideman, F., 49 Weigel, D., 109, 140 Weigel, H., 236 Weil, M., 162, 175, 180 Weinert, T. A., 146, 152, 175 Weintraub, H., 173 Weir, J. W., 139 Weiser, J., 283, 285, 288, 309, 311, 320 Weishammer, G., 102 Wellsbury, P., 240 Welton, M., 312 Wenzel, W. W., 101, 104 Westcott, R. J., 57 Wetherbee, R., 187, 190, 205, 234, 235, 237, 239, 240 Whelan, W. J., 47, 48, 55, 58 White, D. C., 232 White, K. N., 238

AUTHOR INDEX White, K., 145, 180 White, P. J., 77, 81, 105 White, P., 60 White, P.J., 100 White-Cooper, H., 172 Whitehouse, D. G., 306 Whiting, S. N., 74, 76, 81, 91, 104, 105 Widdows, J., 187, 217, 239 Wiegand, C. L., 9, 60 Wieruszeski, J.-M., 50 Wierzbicki, A. T., 139 Wigglesworth, V. B., 270, 320 Wilding, N., 250, 303 Wilekens, H., 173 Wilkins, J. C., 100 Willey, N. J., 100 Williams, L. E., 78, 105 Williams, L., 181 Willis, J. H., 275, 304 Wilm, M., 135, 140 Wilson, R. H., 57 Wiltshire, K. H., 219, 221, 239 Winston, F., 117, 140 Wisniewski, J., 138 Witte, G., 237 Wolf, M. J., 4, 8, 60 Wolffe, A. P., 136, 140 Wolfstein, K., 196, 199, 201, 202, 233, 239 Wollenzien, U., 239 Wong, C., 179 Wong, J., 141 Wong, K. S., 53 Wong, K., 57 Wood, R. K. S., 262, 304 Woodward, D. L., 319 Workman, J. L., 134, 135, 140 Wright, J. A., 152, 180 Wright, J. L. C., 320 Wright, L. D., 235 Wright, P. E., 136 Wright, T., 312 Wright, V. F., 281, 320 Wu, C., 112, 113, 136, 137, 138, 139, 140 Wu, H., 171, 180 Wu, J., 140 Wu, K., 57 Wu, S. H., 304, 320 Wu, Z., 178, 180 Wulff, S., 197, 239 Wustman, B. A., 189, 191, 203, 204, 205, 239, 240 Wykoff, D., 172 Wyllieb, A. H., 176 X Xiao, H., 113, 140 Xing, G., 173 Xu, A., 53 Xue, Y., 112, 113, 140, 141

343

Y Yallop, M. L., 186, 195, 221, 222, 228, 240 Yamada, T., 179 Yamaguchi, T., 179 Yamaguchi, Y., 140 Yamamoto, A., 147, 180 Yamamoto, M., 158, 176 Yan, L. L., 54 Yang, J., 42, 60 Yang, L., 57 Yang, X. E., 102 Yaniv, M., 138, 140 Yano, A., 162, 180 Yano, M., 59 Yao, Y., 48 Yasui, H., 310 Yates, J. R., 47 Ye, Z. Q., 102 Yee, J., 175 Yeh, S. F., 288, 289, 304, 320 Yoder, O. C., 298, 299, 300, 320 Yoo, S. H., 29, 38, 60 Yordan, C., 136 Yoshida, H., 175 Yoshii, M., 48 Yoshikawa, N., 309 Yoshimoto, Y., 292, 320 Yoshioka, H., 173 Young, J. D.-E., 313 Young, M. K., 141 Young, P. G., 178 Young, S., 102 Young, T. E., 4, 60, 167, 168, 180 Yu, C.-M., 318 Yu, I., 139 Yu, L., 55 Yuan, J., 174

Z Zacharuk, R. Y., 250, 270, 321 Zak, N. B., 135, 137 Zavortink, M., 189, 233 Zee, S. Y., 58 Zeeman, S. C., 35, 36, 37, 38, 58, 61 Zeeman, S., 50, 52 Zeller, E. A., 296, 321 Zeng, Y., 180 Zhan, X. -C., 173 Zhang, C., 179 Zhang, F., 89, 105 Zhang, J., 60 Zhang, K., 150, 180 Zhang, L., 52 Zhang, M., 135 Zhang, Y., 112, 113, 114, 136, 138, 141 Zhao, F. J., 68, 76, 79, 80, 101, 102, 105 Zhao, J., 166, 181 Zhao, S. -L., 102

344 Zheng, L. M., 313 Zhou, J., 134, 215, 240 Zhou, S., 140 Zhou, Y., 150, 181 Zhu, Q., 60 Ziegler, G. R., 56 Zierold, K., 101

AUTHOR INDEX Zilberman, D., 137 Zingmark, R. G., 234 Ziv, Y., 179 Zobel, H., 53 Zocher, R., 308, 313 Zocker, R., 311 Zychlinsky, A., 313

SUBJECT INDEX

14-3-3 proteins 155–6, 158

A ACF complex 113, 131 Achnanthes longipes 189, 191, 203, 204 active oxygen species (AOS) 168 Acyrthosiphon pisum 250 ADPG, transportation 23–4 Aedes aegypti 288 Aeollanthus biformifolius 67 aerenchyma 161, 171 aflatoxins 281, 282, 284, 298 AGPase 20–3, 42 evolution of 22–3 flux control coefficient of 46 location of 21–2 Alternaria brassicae 287, 289, 296 Alternaria kikuchiana 283, 287 Alyssum 68, 69 Alyssum lesbiacum 79, 83 Alyssum montanum 83 Alyssum murale 86 Alyssum saxatile 86 Alyssum serpyllifolium 86 ammonia 262 -amylase 12, 18 amylopectin A-type crystals 18 accumulation 9 B-type crystals 18 structure 16, 17–19 amylose accumulation 9 initiation 37–8 structure 19–20 aphidicolin 161, 165, 166, 167

Advances in Botanical Research Vol. 40 incorporating Advances in Plant Pathology ISBN 0-12-005940-1

apoptosis 147, 161, 164 appressoria 246, 265–6 Arabidopsis 14-3-3 proteins in 158 atm mutant of 160 CAK kinase homologue in 151 cell cultures 161–2 COP9 complex in 157 genome sequence 108, 113, 114 HATs 125 homologues of prototype SWI/SNF-type complexes 117–21 SNF5-type 119, 120 SWI2/SNF2-type 117–19, 120 SWI3-type 119, 120, 121 SWP73-type 119, 120 lsd1 mutant of 168 methylation in 123, 127, 132 SET-domain proteins 125–6 Arabidopsis halleri cadmium accumulation in 71–2, 83 crossed with A. petraea 73–4, 82, 86–7 metal levels in flowers and leaves 92 multiple metal extraction by 95 predation protection in 86, 87–8 zinc accumulation in 65–6, 71, 72, 73, 83 zinc extraction by 95–6 zinc search by 74 zinc storage in leaves 80 zinc tolerance in 83 Arabidopsis petraea 71, 73–4, 82 Arabidopsis thaliana 25, 26, 27, 38, 45 genome analysis of 114–16, 125 zinc search by 74 zinc transporters in 78 arabinose 206–9, 211, 212, 213, 214 Aschersonia aleyrodis 263

Copyright  2003 Elsevier Ltd All rights of reproduction in any form reserved

346

SUBJECT INDEX

Ascomycetes 273, 277, 278 Aspergillus flavus 263, 281, 282, 298 Aspergillus fumigatus 271, 272, 276 Aspergillus nidulans 260, 269, 273, 300 Aspergillus niger 271, 289 Aspergillus parasiticus 281 ATM 153, 170 ATR 152–3, 154, 155, 165 ATRIP 155, 165

B Bacillus thuringiensis (Bt) 242 Baeria nivea 288 barley A-type granules in 14 B-type granules in 14 cytosolic SSU from 23 mutants Notch-2 13, 35, 36 Ris13 13 Ris16 20 Ris17 13, 35, 36 Ris527 13 proteins in plastid envelopes of 24 starch accumulation in 9, 10 bassianolides 283, 286, 288, 296 Bathybius haecklii 226 Beauveria bassiana 242, 243, 245, 250, 280, 292, 297 cuticle-degrading enzyme production by 263, 265, 267, 270, 272, 273, 276 toxic metabolite production by 279, 280–1, 282, 283, 286–8, 296 Beauveria brogniartii 274, 279, 280, 281, 283, 286, 288 beauvericin 279, 283, 287, 288, 290, 294, 296 beauverolides 283, 286, 288, 292, 296 benthic microalgae see microphytobenthos (MPB) Berheya coddii 95, 97 Berkerlya 191 biofilms 185–6, 188, 216, 220, 221 shaded 220 species composition of 217 transient 219 biofouling 204 biological stabilisation coefficient 227 blister mats 222 Bombyx mori 281, 288, 298 effects of DTX on 289, 293 infected by Beauveria bassiana 242, 263, 297 Brassica juncea 96 Brassica oleracea 84 Brassicaceae 69, 70, 95 BRG1 complex 112 Brittle-1 protein 24 BRM complex 112

bromodomains 112, 113, 118–19, 125 BSH protein 119, 120, 121 budding yeast 160

C Caenorhabditis elegans 108, 114–16, 125, 132, 133, 134 PCD in 148 CAK kinase 151 calamine soils 71, 72 Calliphora vicina 288 Calliphora vomitoria 265, 268, 269, 270 Candida albicans 244 carbohydrate bulk 191 colloidal 191, 192 diagram of production and loss of 229 low molecular weight (LMW) 191, 221 patterns of distribution 216–23 production in microphytobenthos see microphytobenthos (MPB) see also EPS cdc2 gene 149 CDC2 kinase 150–1 CDC25 153–6, 157, 158 CDK genes 149–50 ced-3 gene 148 cell cycle 145–7 DNA synthetic phase (S) 145, 146 mitotic phase (M) 145, 146 phosphoregulation of 149–52 premitotic interphase (G2) 145, 146 presynthetic interphase (G1) 145, 146 cell cycle checkpoints 148, 152–6 DNA damage (G2/M) 148, 152, 168 DNA replication (S/M) 148, 152 human 170 plant 157–8, 170–1 need for 158–60, 161 cEPS 192, 196–7, 198, 199, 202, 215 Chaetoceros decipiens 206, 213 CHD/Mi2 subfamily 112, 113–14, 117 in Arabidopsis 121 chitinolytic enzymes 257–9 regulation of 261–2 role in fungal pathenogenesis 268–9 Chlorophyll a 216, 217, 220, 227 Chlamydomonas reinhardtii 38 Choristoneura fumiferana 278 CHRAC complex 113 chromatin ATP-dependent remodelling mechanisms 111–21 basics 109–10 modulation of structure 110–11 perspectives 133–4 structural transitions 110 see also core histones; linker histones

SUBJECT INDEX CHY1 255, 300 CMT3 protein 126, 127 Cochliobolus carbonum 124, 271, 273, 279, 297 Colletotrichum nicotianae 280 conidia 266, 267, 269, 270 Conidiobolus obscurus 250, 278, 279 Conidiobulus coronatus 244, 271 COP9 complex 157 Cordyceps subsessilis 285 core histones 109, 110 acetylation 112, 121–2, 123, 124–5, 126 methylation 121, 122–3 and DNA methylation 126–30 phosphorylation 121, 124 Corophium volutator 227–8 Coscinodiscus nobilis 206, 213 Craspedostauros australis 189 Cryolander system 219 Cryptococcus neoformans 244 Cuscuta californica 86 cuticle, insect invasion of 246, 249, 250–1 structure of 248 cuticle-degrading enzymes 251–76 chitinolytic enzymes 257–9 esterolytic enzymes 259–60 evidence for role in fungal pathogenesis 265–70 evolutionary considerations 270–4 lipolytic enzymes 259 overview 252–3 production by Metarhizium anisopliae 251–62 production by other entomopathogenic fungi 263–5 proteolytic enzymes 253–7 regulation of production 260–2 role in virulence and specificity 274–6 cyclic peptide toxins 286–96 cyclosporins 283, 288, 292, 294, 296 Cylindrotheca closterium 186 carbohydrate production in cultures of 192, 193, 196, 197, 199, 200 EPS compositional data for 205, 207, 208, 209, 210, 213 storage compound of 203 Cylindrotheca fusiformis 207, 213 Cymbella cistula 191 cytochalasins 279, 282, 285, 292

D DAD genes 169–70 DDM1 protein 114, 127, 128, 129, 130 Deroceras caruanae 85 destruxins (DTX) 283, 284, 287, 288–96, 298 Deuteromycetes 275, 278, 298–9 DEXD/H superfamily 112

347

diatoms 184 centric 184 epipelic 185 assemblage composition 217 motility of 188–90 pennate 184, 185 planktonic 184, 204 Dimilin 269 DNA, methylation of 126–30, 131–3 DNA ladders 170 DOMINO protein 116 Drosophila melanogaster 108, 114–16, 125 Dubois assay 225

E E2F 164 efrapeptins 283, 285 Encyonema 191 endoproteases 253–6 regulation of 260–1 endoreduplication 167, 171 endosperm carbohydrate supply to 4–6 development 3–4 PCD in 167–8 starch synthesis in see starch synthesis in endosperms enniatins 279, 283, 287, 288, 290, 294 Entomophaga aulicae 278 Entomophthora coronata 263 Entomophthora grylli 277 Entomophthorales 271–2, 278, 279 EPS 186, 191 binding capacity of 228–30 biochemical composition 197, 204–10 comparison of studies 210–15 bound 192, 196, 214–15 colloidal see cEPS conceptual model of 214–15 distribution patterns depth distribution 218–19 horizontal distribution 216–18 loss of carbohydrates 221–2 metal concentration effects 222–3 seasonal variation 219–20 tidal variation 220 dynamics of 228–9 future research areas 230–1 influence on sediment dynamics 224–8 production 189 in culture 192–8 in situ 198–203 photosynthesis and 199, 202 production pathways 203–4 properties 187 ‘soluble’ 196–7, 210 terminology 191–2 ERCC6/CSB subfamily 114

348

SUBJECT INDEX

Erynia dipterogena 271 Erynia neoaphidis 271 Erynia rhizospora 271 Escherichia coli 21, 34, 158 esterolytic enzymes 259–60 role in fungal pathenogenesis 269–70 ethylene 162–4 etr1–1 gene 163 Euphorbiae 250 exoproteases 256–7 regulation of 261 extra-cellular matrix (ECM) 190 extracellular polymeric substances see EPS F FCA protein 121 fission yeast 145, 153, 155–6, 157, 158 flux control coefficients 41, 45, 46 fructose 6-phospate (F6P) 44 fucose 206–9, 211, 212, 213, 214 Fucus spiralis 157 Fun30p protein 116 fungal pathogens of insects future research 299–300 infection process overview 245–6 organic acid production by 262 taxonomy 244–5 see also cuticle, insect; cuticle-degrading enzymes; toxins, from entomopathogenic fungi fungal pathogens of plants 169, 267, 270–1, 272, 280, 299 Fusarium avanaceum 280 Fusarium lateritium 288 Fusarium oxysporum 256, 287 G galactose 205, 206–9, 212, 213, 214 Galleria mellonella 253, 286, 288, 292 effects of DTX on 289, 290, 291 infected by Beauveria bassiana 274, 296, 297 infected by Metarhizium anisopliae 276, 285, 296 infected by Tolypocladium niveum 285 GBSSI 7, 25, 26, 28–9, 30–1, 38 GF14omega 158 glucans, in EPS production 203–4 glucose 205, 206–9, 211, 212, 213, 214 glycogen 19, 37 glycogen synthase (GS) 25, 26 glycogenin 37, 39 glycoproteins 189, 191, 285 Gnaphalium limosum 217 Gomphonema sp. 188 grass family phylogeny 15 grazing 227–8 GRIM 145 Gyrosigma sp. 188

H Haematonectria haematococca 272 Haslea 191 hBrm complex 112, 119 HDA6 gene 124 HDAC1/HDAC2 113 Helianthus tuberosus 159 Heliothis zea 250, 253, 274 Hemiberlesia rapax 288 herbivores 85–6, 92–3 Hirsutella thompsonii 282, 283, 285, 286 hirsutellin 283, 285–6 histidine 79, 83 histone acetyltransferases (HATs) 122, 123, 124, 125, 126, 133 classification of 125 histone deacetylases (HDACs) 122, 123, 124, 125 classification of 125 histone methyltransferases (HMTs) 122–3, 126, 133 histones core see core histones linker see linker histones HMG1/Y proteins 131 Homo sapiens 108, 114–16 HP1 protein 123, 124, 132 Hyalophora cecropia 275 hydrogen peroxide 168 hyperaccumulation arsenic 92 cadmium 71, 72, 74 accumulation mechanism 79 copper 66–7 description 64–9 ecological consequences of 90–3 co-evolution 92–3 effects on herbivores 91–2 effects on higher trophic levels 92 effects on other plants 91 evolution of accumulation 82–90 allelopathy 90 defence 84–8 drought tolerance 89–90 inadvertant uptake 88–9 increased tolerance 83–4 genetics of accumulation 73–4 land clean-up potential see phytoextraction mechanism of accumulation 74–81 control of accumulation 80–1 from soil to plant 74–6 into plant 76–80 storage in leaves 80 nickel 67–8, 70, 71–2, 83–4 accumulation mechanism 79–80 and cuticular transpiration 89 and predation protection 85, 86, 88 taxonomic distribution 69–71

SUBJECT INDEX term coined 64 thresholds for 64–5 variation within species 71–3 zinc 65–6, 70, 71–3, 82 accumulation mechanism 77–9 and predation protection 86 and zinc tolerance 73–4 hypersensitive response (HR) 169–70 Hyphomycetes 272 I INI1 protein 119 initiation 37 INO80 subfamily 112, 114, 116 inorganic phosphate (Pi) 43–4, 45 iron deficiency 89 Isarea cretacea 287 Isarea fwelina 286–7 isoamylase 11, 35–6, 39 ISWI subfamily 112–13, 114 K KISMET group 117 kojic acid 282, 298 Kryptonite protein 123, 126, 130 L Lagenidium giganteum 245, 263 lead contamination 96 LEC1 gene 121 Leptinotarsa decemlineata 280 LHP1 protein 123, 127, 132 linker histones 109, 110, 130–3 methylation 132 phosphorylation 131 lipolytic enzymes 259 role in fungal pathenogenesis 269–70 Locusta migratoria 275, 291 Lycium barbarum 169 M Macrosiphum euphorbiae 269 Magnaporthe grisea 244 maize aerenchyma in 161 AGPase in 44 B. bassiana in 280 coarse regulation in 43 cytosolic SSU from 23 endosperm 167 GBSSI in 28 mutants 34 ae 30 amylose-extender 34 Brittle-1 24 brittle-2 20 dull 30 high-amylose 18

349

miniture-1 5 Rev6 44–5 shrunken-2 20 sugary 39 sugary-1 35, 36 sugary-3 36 waxy 20 simple granules in 14, 15 starch synthase in 25 WEE1 homologue in 151 malto-oligosaccharide (MOS) precursors 38 Manduca sexta 248, 275, 297 effects of DTX on 289, 293, 294, 298 infected by Beauveria bassiana 281 infected by Metarhizium anisopliae 247, 249, 250, 265, 266, 268, 269, 270, 293, 294 mannose 205, 206–9, 210, 211, 212, 214 MBD2 protein 114 MCP 163 MeCP1 complex 114 MeCPA 257, 266 Medicago sativa 150 Melanoplus sanguinipes 274 Melanotrichus boydi 93 MET1 protein 126, 128, 129 metabolic pathways control of 40–1 metabolic control analysis 41, 45–6 metal transporters 78–9 Metarhizium anisopliae 242, 244, 292 CHY1 in 300 cuticle-degrading enzyme production by 251–62 DTX production by 288, 289, 291–4, 295, 296, 298 infection process by 245, 246, 247, 249 NPRS genes in 299 proteases produced by 252, 296 toxic metabolite production by 282, 283, 285 Metarhizium flavoviride 275, 280, 282, 293 Mi2 ATPase 113, 131 microphytobenthos (MPB) 185 biomass of 216, 219, 220 carbohydrate production in 192–203 in culture 192–8 in situ 198–203 MIK1 151, 155 Mimulus guttatus 66–7 monosaccharides, in EPS 205–10, 211–14 mucilages 186, 187, 189, 226 variable nature of 190 Mucor haemalis 244, 245 Mus musculus 114–16 mycoinsecticides 242, 244 MYT1 151 Myzus persicae 264, 297

350

SUBJECT INDEX

N NAG 258, 259, 262, 263 Nassella nidulans 272, 273 Navicula perminuta 193, 194, 197, 200 Navicula salinarum 194, 205, 208, 209 necrosis 144 nematode see Caenorhabditis elegans Neurospora crassa 272, 273, 300 Nitzschia epithemioides 186 Nitzschia frustulum 194, 197, 209 Nitzschia sigma 194, 197, 217 Nitzschia sp. 194, 196, 205, 207, 209 Nomuraea rileyi 263, 298 NPRSs 298–9 nucleosome 109 NuRD complex 113–14, 123 NURF complex 113 O oats compound granules in 13, 14 simple granules in 14–15 oosporein 280–1, 282 Ophiosphaerella herpotricha 287, 289, 296 Oryctes rhinoceros 293 Ostrinia nubilalis 280 Otiorhynchus sulcatus 294 oxalic acid 262, 281, 282 P p53 protein 170 pads 190, 191 Paecilomyces farinosus 272, 288 Paecilomyces fumoso-roseus 283, 286, 287, 288 Parlibellus 191 pathogenicity islands 300 pathogens, defence against 86–8 PCD 147–8, 171 in endosperm 167–8 fungal elicitors of 169–70 instances in plants 160–4 oxidoreductive state effects 168–9 PCNA 153 peas 18, 24, 31, 89 peptidases 256–7 Periplaneta americana 291 Phaedon cochleariae 289 3-phosphoglyceric acid (3PGA) 43–4, 45 phospholipases 298 phosphorylation 40, 45, 286 photodamage protection 198 photosynthesis 199, 202, 219, 221 Phyllocoptruta oleivora 285 Physarum polycephalum 122 phytoextraction 94–8 chelator-assisted 96, 97

commercial application 97–8 drawbacks 96–7 phytoglycogen 35–6 phytopathogens 298–9, 300 Phytophthora infestans 169 phytoplankton 198 phytoremediation 93–4 see also phytoextraction PICKLE 121 Pieris brassicae 85 Pinnularia 189 Pisum sativum L. 34 Plutella xylostella 289 pollutants, particle-associated 224 Polycomb group (Pc-G) proteins 108, 132 polysaccharides, in EPS 190, 205 potatoes GBSSI in 28 isoamylase activity in 39 starch-branching enzyme in 34–5 Pr1 253–6, 257, 273, 275, 296–7 regulation of 260–1, 297 role in fungal pathogenesis 265–70 Pr2 255, 256 regulation of 260, 261 role in fungal pathogenesis 265, 268 PrB1 273 programmed cell death see PCD proline 90 proteases 253–7 regulation of 260–1 as toxins 296–8 proteoglycans 189–90, 205 Pseudomonas aeruginosa 296 Pyrenomycetes 272 Pythium 86

R R genes 169 RAD genes 146, 160 Rad3 153, 155 raphe 185, 188, 189, 198, 204 reactive oxygen species (ROS) 168 REAPER 145 retinoblastoma (Rb) protein 164 rhamnose 205, 206–9, 211, 212, 213, 214 ribose 210 rice compound granules in 13 cytosolic SSU from 23 DAD-1 170 mutants 34 amylose-extender 34 sugary 39 sugary-1 35, 36 roscovitine 165, 166 RSC complex 112

SUBJECT INDEX S Saccharomyces cerevisiae 34, 114–16, 125, 273, 290, 300 saprophytes 271, 272, 273, 298 SBE see starch-branching enzyme Schistocerca gregaria 85, 247, 269, 275, 289, 291, 294 Schizosaccharomyces pombe 34, 114–16, 125, 160 sclerotisation 251, 253 secondary metabolism 276–7, 280 sediments cohesive 224 influence of EPS and organisms on dynamics of 224–8 non-cohesive 224 stability of 224–5 Sedum alfredi 66 Senecio coronatus 71, 85 serpentine soils 67, 68, 70, 71, 88, 89–90 silica frustule 185, 204 snails 86–7 SNF5 protein 117, 119, 120 soybean 162 spherulites 39 spinach leaves 45 Spodoptera exigua 85, 286, 288 stalks 188, 190, 191, 204 starch accumulation 9–12 starch degradation 11–12 starch granules initiation 38–9 morphology 12–15 compound granules 12, 13 simple granules 12, 13 structure 16–17 starch synthase (SS) 19, 25–31, 42 flux control coefficient of 46 GBSSI 7, 25, 26, 28–9, 30–1, 38 SSI 25, 26, 27, 28 SSII 25, 26, 27 SSIII 25, 26, 27, 30, 45 SSIV 25 starch synthesis in endosperms 6–9 control of 40–7 coarse regulation 41–3 fine regulation 43–5 metabolic control analysis 45–6 and yield 46–7 starch-branching enzyme (SBE) 29, 31–5, 42 flux control coefficient of 46 SBEI 33–4 SBEII 33–4 Stauroneis (staurophora) amphioxys 194, 196–7, 205, 206, 210, 213 Stauroneis decipiens 189, 190, 204 Streptanthus brewerii 83, 85 Streptanthus polygaloides 71, 83, 85, 86, 93

351

Streptanthus tortuosus 85 Streptomyces griseus 255, 300 STRING 145–6 sucrose synthase 25, 46 summit disease 278 SUP gene 124, 127 Surirella ovata 194, 197, 200 SWI/SNF complex 112, 131 SWI2/SNF2 subfamily 112, 114, 115–20 phylogenetic tree of 115–16 SWI3 protein 117, 119, 120, 121 SWP73 protein 117, 119, 120 SYD gene 120–1

T tanning 251 Tenebrio molitor 276 Thermus aquatica 271 Thlaspi 66, 69 Thlaspi arvense 76, 91 nickel uptake by 79, 84 zinc transporters in 78 zinc uptake by 77–8 Thlaspi caerulescens 64, 73 cadmium accumulation in 71 cadmium uptake by 79 multiple metal extraction by 95 nickel accumulation in 71, 72 predation protection in 85, 86 zinc accumulation in 66, 71, 72, 88 zinc extraction by 95–6 zinc mobilisation by 75–6 zinc search by 74 zinc storage in leaves 80 zinc tolerance in 74, 83 zinc transporters in 78 zinc uptake by 77–8, 81 and soil toxicity 91 Thlaspi goesingense 76 nickel uptake by 79, 83 Thlaspi montanum 71 tobacco 132, 133, 151, 162, 166, 169 TBY-2 cell line 161, 162–3, 169 Tolypocladium cylindrosporium 288 Tolypocladium niveum 272, 283, 285 toxins, from entomopathogenic fungi 276–99 cyclicpeptide 286–96 incidence 277–80 multiple 298–9 overview 276–7 proteases and other enzymes as 296–8 survey of 280–6 tracheary element (TE) formation 164–7 trails 190 Triatoma infestans 274 Trichoderma album 263

352

SUBJECT INDEX

Trichoderma harzianum 256, 273 Trichoderma polysporum 288 Trichoderma virida 162 Trichothecium roseum 289, 296 Tryblionella sp. 188 tubes 188, 190, 191 TUNEL reactions 170 turkey egg white inhibitor (TEI) 266 TWINE 158 U uronic acids, in EPS 190, 205, 206–9, 210–11, 212, 213–14 UV radiation 158–60 V Verticillium chlamydosporium 273 Verticillium lecanii 247, 250, 269 cuticle-degrading enzyme production by 263, 264, 272, 273, 275, 297 toxic metabolite production by 279, 283, 286 viridoxins 280, 282, 283 W waxy mutants 20, 28, 29

WEE1 151, 155, 156, 158 wheat A-type granules in 14 B-type granules in 14 coarse regulation in 43 cytosolic SSU from 23 endosperm 167 GBSSI in 25, 28 proteins in plastid envelopes of 24 starch accumulation in 9, 11 starch degradation in 12 starch synthase in 25, 26 sucrose supply to 47 X Xenopus 132, 156, 158 xylogenesis 164–7 xylose 205, 206–9, 210, 211, 212, 214 Y yeast 79–80 yield, cereal crop 11, 46–7 Z zinc transporters 78, 89 Zinnia elegans 164–7