Advances in Botanical Research, Volume 25

Advances in BOTANICAL RESEARCH incorporating Advances in Plant Pathology VOLUME 25 The Plant Vacuole Advances in B...

Author: J. A. Callow | Roger A. Leigh | Dale Sanders

18 downloads 1199 Views 26MB Size Report

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

DOWNLOAD PDF

Advances in

BOTANICAL RESEARCH incorporating Advances in Plant Pathology VOLUME 25

The Plant Vacuole

Advances in

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

School of Biological Sciences, University of Birmingham, Birmingham, U K

Editorial Board J. H. ANDREWS

H. G. DICKINSON M. KREIS R. M. LEECH R. A. LEIGH E. LORD D. J. READ I. C. TOMMERUP

University of Wisconsin-Madison, Madison, USA University of Oxford, Oxford, UK Universitt! de Paris-Sud, Orsay, France University of York, York, UK I A CR-Rothamsted, Harpenden, U K University of California, Riverside, U S A University of Shefield, Shefield, UK CSIRO, Perth, Australia

Advances in

BOTANICAL RESEARCH incorporating Advances in Plant Pathology

The Plant Vacuole edited by

R. A. Leigh

D. Sanders

and

Biochemistry and Physiology Department, IACR-Rothamsted, Harpenden, Herts, U K

The Plant Laboratory, Biology Department, University of York, York, UK

Series editor

J. A. CALLOW School of Biological Sciences, University of Birmingham, Birmingham, UK

VOLUME 25

1997

ACADEMIC PRESS San Diego

London

Boston New York

Sydney Tokyo Toronto

This book is printed on acid-free paper Copyright

0 1997 by

ACADEMIC PRESS

All rights reserved. N o part of this publication may be reproduced or transmitted in any form or by any means electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Academic Press, Inc. 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.apnet.com Academic Press Limited 24-28 Oval Road, London NWl 7DX, UK

http://www.hbuk.co.uk/ap/ A catalogue record for this book is available from the British Library ISBN 0-12-005925-8

Typeset by Keyset Composition, Colchester, Essex Printed in Great Britain by Hartnolls Limited, Bodmin, Cornwall

97 98 99 00 01 02 EB 9 8 7 6 5 4 3 2 1

We dedicate this volume to Professors Tom ap Rees, Harold Woolhouse and Horst Marschner who all sadly died during 1996 and who were each inspirational in leading their own fields of research in the plant sciences. Roger A . Leigh Dale Sanders

This Page Intentionally Left Blank

CONTENTS

CONTRIBUTORS TO VOLUME 25 .......................... CONTENTS OF VOLUMES 14-24

xvii

.............................

xix

SERIES PREFACE ..................................................

xxvii

PREFACE ..............................................................

xxix

The Biogenesis of Vacuoles: Insights from Microscopy F. MARTY I . Introduction

...........................................................................

1

I1. Sorting of Vacuolar Precursors .................................................

2

I11. The Autophagic Pathway .........................................................

A . Starvation-induced Cellular Autophagy in Plant Cells .............

7 13

IV . Endocytic Pathways and Vacuole Biogenesis ............................... A . Endocytic-like Plasma Membrane Resorption After Secretion .. B . Plasma Membrane Internalization in Water-stressed Cells .......

17 25 25

IV . Ontogeny of Vacuoles Specialized in Protein Storage ....................

27

VI . Conclusions ...........................................................................

32

.................................................................

33

References ............................................................................

33

Acknowledgements

Molecular Aspects of Vacuole Biogenesis D . C . BASSHAM and N . V . RAIKHEL I . Introduction

...........................................................................

43

I1. Targeting of Soluble Proteins to the Vacuole .............................. A . N-terminal Propeptides .....................................................

45 46

...

CONTENTS

Vlll

B . C-terminal Propeptides ..................................................... C . Internal Targeting Signals ...................... ....... D . Transport of Plant Proteins to the Yeast Vacuole .................. 111. Mechanism of Protein Transport to the Vacuole

.......................... ................

A . Multiple Mechanisms for Transport to the Vacuole B . Components of the Vacuolar Transport Machinery

48 49 50 50 51

IV . Transport of Membrane Proteins to the Tonoplast ...........

V . Perspectives .......................................................... Acknowledgements

.........................

....................................

References ............................................................................

56 56

The Vacuole: a Cost-Benefit Analysis J . A . RAVEN I . Introduction

...........................................................................

59

I1 . Demonstrated and Hypothesized Functions of Vacuoles. and Alternative Means of Performing these Functions ........................

62

Demonstrated and Hypothesized Costs of Producing and Maintaining Vacuoles. and of Costs of Alternative Means of Performing Vacuolar Functions .................................................

62

I11*

IV . A Case History: “Vacuolate” (Eu)bacteria

...................................

62

V . Another Case History: Vacuoles and Buoyancy .............................

75

VI . Costs and Benefits of Vacuolation: Simple Analyses and the Allocation of Costs Among Various Benefits ................................

78

VII . Evolutionary Aspects ................................................................

81

VIII . Conclusions .............................................................................

82

...................................................................

82

Acknowledgements References

..............................................................................

82

ix

CONTENTS

The Vacuole and Cell Senescence P . MATILE I . Introduction .......................................................................... A . Scnescence and Death ..................................................... B . Functions of Vacuoles in Cell Senescence ............................

87 87 88

.............. ....................... 89 I1. Leaf Senescence . A . Differential Se elles .................. 89 90 B . Vacuolar Hydrolases ......................................................... 92 C . Autophagic Activity of Vacuoles ..... 94 D . Autolysis ........................................................................ 96 E . Accumulation and Export of Solutes 97 F. Vacuoles and the Breakdown of Chlor ........................ 102 G . Secondary Compounds ............... 111. Senescence and Autolysis in Various Cell Phenotypes

IV .

Programmed Cell Death

V . Retrospect References

...................

..................................................

103 105

.............................................................................

1Oh

...........

107

............................................................

Protein Bodies: Storage Vacuoles in Seeds G . GALILI and E . M . HERMAN I . Introduction

...........................................................................

113

I1. Ontogeny of PSVs .................................................................

114

111. The Golgi Apparatus Mediates the Deposition of PSV Constituents in Dicotyledonous Plants .......................................

116

IV V.

Transport of Storage Proteins to Vacuoles in Monocotyledonous Plants ................................................................................... Developmental Regulation of the PSV Tonoplast

........................

VI . Enzyme Composition of PSVs ...........................

120 123 127

VII . Diversity of Vacuolar Storage Proteins and Enzymes .... A . Globulin Storage Proteins .................................................. B . Prolamin Storage Proteins .......... .... .......................

128 128 129

VIII . Assembly and Processing of Storage Proteins .............................. A . Assembly of the Storage Proteins Within the ER is Assisted by Molecular Chaperones ..................................................

129 130

CONTENTS

X

B . Proteolytic Processing of Prolegumins inside Vacuoles ............ 132 IX . Expression of Storage Protein Genes in Transgenic Plants ............. 133 Acknowledgements

.................................................................

............................................................................

References

135

135

Compartmentation of Secondary Metabolites and Xenobiotics in Plant Vacuoles M . WINK I . Introduction ........................................................................... A . Secondary Metabolites as Defence and Signal Compounds of Plants ......................................................................... B . Fate of Xenobiotics in Plants ............................................. C . Aims and Scope ...............................................................

I1. Vacuolar Storage of Secondary Compounds and Xenobiotics ......... A . Secondary Compounds ...................................................... B. Xenobiotics ..................................................................... C . Mechanisms Underlying Vacuolar Sequestration .................... 111. Conclusions

141 143 145 145 145 153 154

...........................................................................

159

.................................................................

160

............................................................................

160


141

Solute Composition of Vacuoles R . A . LEIGH I . Introduction

...........................................................................

I1. Variability of Vacuolar Solute Composition ................................. A . X Ray Microanalysis ......................................................... B . Ion-selective Microelectrodes .............................................. C. Single-cell Sampling and Analysis (SiCSA) ...........................

111. Regulation of Vacuolar Solute Pools

IV. A Model

172 175 177 180

..........................................

182

...............................................................................

187

...........................................................................

189

V . Conclusions References

171

............................................................................

189

xi

CONTENTS

The Vacuole and Carbohydrate Metabolism

.

C . J POLLOCK and A . KINGSTON-SMITH I . Introduction

...........................................................................

195

I1. Methodological Approaches ..................................................... A . Compartmental Analysis ................................................... B . Non-aqueous Fractionation and Stereological Analysis ........... C . Direct Sampling of Vacuolar Sap ........................................ D . Preparation of Isolated Vacuoles ........................................ E . Analysis of Transport Functions in Isolated Membrane Vesicles .......................................................... I11. Sucrose and its Component Hexoses .......................................... A . Are Sucrose and its Component Hexoses Found in Vacuoles? B . Are Sucrose. Glucose and Fructose Accumulated Actively in Vacuoles? ....................................................... C . Are Sucrose-metabolizing Enzymes Located in the Vacuole? IV.

Fructans

...

201 202 204 207

...............................................................

208

...........................................................................

210

.................................................................

211


198 200

...............................................

VI . Other Carbohydrates VII . Conclusions

198

.

................................................................................

V. Raffinose-series Oligosaccharides

196 197 197 197 198

............................................................................

211

Vacuolar Ion Channels of Higher Plants G . J . ALLEN and D . SANDERS I . Introduction .......................... A . Vacuoles as Ion Stores ...................................................... B . Electrochemical Potential Differences for Ions Across the Vacuolar Membrane ................................ ............ C. Polarity of Membrane Potential and Ionic Cur he Vacuolar Membrane ...... ................................ D . General Properties of Ion Some Definitions ..... E . Experimental Characterization of Ion Channels in Vacuoles .... I1*

Cation Channels ................................................................. A . SV Channels ............... ............ ........ B . FV Channels ................................................................... C . Vacuolar K+ (VK) Channels ..............................................

218 219 221 221 222 226 226 230 231

xii

CONTENTS

232 D . Other Inward-rectifying K+ Channels .................................. E . Hydrostatic and Osmotic Pressure (HOP)-activated Channels . . 233 233 F. Vacuolar Voltage-gated Ca2+ (VVCa'J Channels ......... G . Inositol 1.4.5.Trisphosphat e.gated Ca + Channels .................. 236 239 H . Ryanodine Receptor Homologues ............................ I11. Anion Channels ....... .......................................................... A . Malate (VMAL) Channels ........................................ B . Chloride (VCI) Channels ................................................... IV . Summary of Individual Channel Characteristics V . Integration of Vacuolar Channel Activity VI . Conclusions

............

243

....................................

243

...........................................................................

246

.................................................................

247

Acknowledgements

............................................................................

References

241 241 242

247

The Physiology. Biochemistry and Molecular Biology of the Plant Vacuolar ATPase U . LUTTGE and R . RATAJCZAK ................................................

I . Introduction

253

I1. Phylogeny .............................. I11. Ontogeny

.............................

IV . Properties

.....................

................... 255

................. 251 .........................

262

........................................

267

VI . Electron Microscopy ...............................................................

270

V.

Holoenzyme Subunit - Fine Structure

VII . Physiological Functions and Ecophysiological Responses VIII . Cell Physiological Regulation

IX . Conclusions and Outlook Acknowledgements References

.

.................................

281

.........

.................

276

284 ......................

285

.................................................

285

CONTENTS

...

XI11

The Molecular and Biochemical Basis of Pyrophosphate-Energized Proton Translocation at the Vacuolar Membrane R.-G. ZHEN. E . J . KIM and P . A . REA I . Introduction

.........................................................................

298

I1 .

Reaction Mechanism ............................................................... A . 1,1-Diphosphonates as Type-specific Inhibitors .... B . Cautionary Note Concerning In Vivo Studies ........................ C . Steady State Kinetics of Substrate Hydrolysis ........................ D . Oxygen Exchange Reactions ..............................................

299 299 300 302 304

111.

Molecular Identity and Sequence .............. A . Identification of the Catalytic Subunit . B . Molecular Cloning of cDNAs Encoding V-PPase C . Isoforms of the Substrate-binding Subunit ............................

307 307 308 310

IV .

Structure-function Relations ..................................................... A . One Polypeptide is Sufficient for Pump Function ................... B . AVP Does Not Functionally Complement Yeast V-ATPase .................................................... C . Homomulti .......................... ... ... D . Revised Topological Model ................................................ E . Identification of Substrate-protectable Maleimide-reactive Cysteine Residue ..... .................................................... F . Potential Coupling Sites ........... ......................

311 311

V . Future Research

321 324

...................................................................

329

.................................................................

331

...........................................................................

332


312 314 317

The Bioenergetics of Vacuolar H+ Pumps J . M . DAVIES I . Introduction

..........................................................................

...................... I1 . Determination of the Coupling Ratio A . Kinetic Estimates of the Coupling Ratio .............................. B . Thermodynamic Determination of the Coupling Ratio ............ C . Use of Patch Clamp Electrophysiology ....................

I11. Patch Clamp Studies of the V-PPase .......... A . The V-PPase and Potassium ............................................... B . Vectorial Activation by Potassium .......

340 340 340 341 342 343 343 344

xiv

CONTENTS

Modelling the V-PPase as a (K+/Hf) Symporter ................... Observed Reversal Voltage of the V-PPase .......................... Deduction of the V-PPase Coupling Ratio ............................ Validation and Future Directions ........................................

346 346 349 350

IV . Patch Clamp Studies of the V-ATPase ....................................... A . Isolation of the V-ATPase Pump Current ............................. B . Reversal Voltage of the V-ATPase and Determination of n .... C . Is the Non-integer Coupling Ratio an Artefact? .................... D . Mechanistic Implications of the Variable Coupling Ratio ........

352 352 354 356 357

C. D. E. F.

V . Physiological Consequences ...................................................... A . Acidification by the V-ATPase ........................................... B . V-PPase and K+ Accumulation ..........................................

358 358 359

VI . Conclusions ...........................................................................

359

.................................................................

360

............................................................................

360


Transport of Organic Molecules Across the Tonoplast E . M A R T I N O I A and R . RATAJCZAK I . Introduction

...........................................................................

I1. Carbohydrates .......................................... 111. Organic Acids

IV. Amino Acids

.........................

........................................................................

.........................................................................

V . Polyamines and Peptides

....... ...............................................

VI . Transport of Secondary Products of Plant Metabolism VII . Conclusion

367 372 382 386

.................. 388

..............................................................

390

.................................................................

390

............................................................................

390


366

Secondary Inorganic Ion Transport at the Tonoplast E. B L U M W A L D and A . GELLI I . Introduction

...........................................................................

401

xv

CONTENTS

11.

............................................... .............................................

402 402 404 406 407 408

I11. Anions .......................................................... A . Chloride ......................................................................... B . Nitrate ... ................ ............ C . Other Anions ..................................................................

408 408 409 411

...........................................................................

412

A . Sodium ........................................................................... B . Potassium ..... ........................

E . Heavy Metals

IV. Conclusions References

...........................................................................

413

Aquaporins and Water Transport Across the Tonoplast M . J . CHRISPEELS. M . J . DANIELS and A . WEIG I . Introduction

...........................................................................

I1. How are Water Channel Proteins Assayed? I11. The Discovery of Aquaporins in Plants

................................

...............

419 420 422

IV . Aquaporin TIP and Aquaporin PIP are Members of a Large Gene Family ..................................................................................

423

...........................................

423

VI . The Activity of a Seed-Specific Tonoplast Aquaporin is Regulated by Phosphorylation ... .................................

426

....................

427

.............................................

428

V. The Structure of the Aqueous Pore

VII . Developmental Regulation of Tonoplast Aquaporins VIII . Are Aquaporins Active in Plants?

IX . Do Aquaporins Play a Role in Water Transport in the Plant? Acknowledgements References

....... 428

.................................................................

430

........................................................

SUBJECT INDEX .................. AUTHOR INDEX ...............

430

....................

433 44 1


CONTRIBUTORS TO VOLUME 25

DR G. J . ALLEN, The Plant Laboratory, Biology Department, University of York, PO Box 373, York YO1 5YW, U K DR D. C . BASSHAM, MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, M I 48824-1312, U S A PROFESSOR E. BLUMWALD, Department of Botany, University of Toronto, 25 Willcocks Street, Toronto, Ontario M5S 3B2, C A N A D A PROFESSOR M. J. CHRISPEELS, Department of Biology 0116, University of California, San Diego, 9500 Gilman Drive, La Jolla, C A 92093-0116, USA DR M. J. DANIELS, Department of Biology 0116, University of California, Sun Diego, 9500 Cilman Drive, La Jolla, C A 92093-OI16, U S A DR J . M. DAVIES, Department of Plant Sciences, University of Cambridge, Downing Street, Cambridge CB2 -TEA, U K DR G. GALILI, Department of Plant Genetics, The Weizmann Institute of Science, Rehovot 76100, I S R A E L DR A. GELLI, Department of Botany, University of Toronto, 25 Willcocks Street, Ontario M5S 3B2, C A N A D A PROFESSOR E. M. HERMAN, Plant Molecular Biology Laboratory, United States Department of Agriculture, Agricultural Research Service, Beltsville, M D 20705, U S A DR E. J. KIM, Plant Science Institute, Department of Biology, University of Pennsylvania, Philadelphia, PA 19104-6018, U S A DR A. H. KINGSTON-SMITH, Institute of Grassland and Environmental Research, Plas Gogerddan, Aberystwyth, Dyfed SY23 3 E B , U K PROFESSOR R. A. LEIGH, Biochemistry and Physiology Department, IA CR-Rothamsted, Harpenden, Hertfordshire A LS 2JQ, U K PROFESSOR U . LUJTTGE, Technische Hochschule Darmstadt, lnstitut f u r Botanik, Schnittspahnstrasse 3-5, 0-64287 Dnrmstadt, G E R M A N Y DR E. MARTINOIA, Genetique Physiologique et Moltculaire, BBtiment de Botanique, 40, Avenue du Recteur Pineau, F-86022, Poitiers, F R A N C E PROFESSOR F. MARTY, Laboratoire de phytoBiologie Cellulaire, Universite' de Bourgogne, BP 138, 210 Dijon Cedex, F R A N C E PROFESSOR P. MATILE, Institut fur Pflanzenbiologie, Universitiit Zurich, Zollikerstrasse 107, CH-8008 Zurich, S W I T Z E R L A N D PROFESSOR C . J . POLLOCK, lnstitute of Grassland and Environmental Research, Plas Gogerddan, Aberystwyth, Dyfed SY23 SEB, U K

xviii

CONTRIBUTORS TO VOLUME 25

DR N. V. RAIKHEL, MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI 48824-1312, USA DR R. RATAJCZAK, Technische Hochschule Darmstadt, Institut fur Botanik, Schnittpahnstrasse 3-5, 0-64287, Darmstadt, G E R M A N Y PROFESSOR J . A . RAVEN, Department of Biological Sciences, University of Dundee, Dundee DDl 4HN, U K PROFESSOR P. A . REA, Plant Science Institute, Department of Biology, University of Pennsylvania, Philadelphia, P A 19104-6018, USA PROFESSOR D. SANDERS, The Plant Laboratory, Biology Department, University of York, PO Box 373, York YO1 5YW, UK DR A. WEIG, Department of Biology 0116, University of California, San Diego, 9500 Gilman Drive, La Jolla, C A 92093-0116, U S A PROFESSOR M. WINK, Institut fur Pharmazeutische Biologie, Universitat Heidelberg, I m Neuenheimer Feld 364, 0-69120 Heidelberg, G E R M A N Y DR R.-G. ZHEN, Plant Science Institute, Department of Biology, University of Pennsylvania, Philadelphia, P A 19104-6018, USA

xix

CONTENTS TO VOLUMES 14-24

Contents to Volume 14 Protein Targeting R. J. ELLIS and C. ROBINSON

Control of Isoprenoid Biosynthesis in Higher Plants J . C. GRAY

Dunaliella: A Green Alga Adapter to Salt M. GINZBURG

Contents to Volume 15 Perception of Gravity by Plants T. BJORKMAN

Crassulacean Acid Metabolism: a Re-Appraisal of Physiological Plasticity in Form and Function H. GRIFFITHS

Potassium Transport in Roots L. V. KOCHIAN and W. J. LUCAS

Sporogenesis in Conifers R. I. PENNELL

xx


Contents of Volume 16 Lipid Metabolism in Algae J. L. HARWOOD and A . L. JONES

The Alteration of Generations P. R. BELL The Formation and Interpretation of Plant Fossil Assemblages R. A. SPICER

Primary Productivity in the Shelf of North-West Europe

P. M. HOLLIGAN

Contents of Volume 17 Plant Evolution and Ecology During the Early Cainozoic Diversification M. E. COLLINSON Origin and Evolution of Angiosperm Flowers E. M. FRIIS and P. K. ENDRESS Bacterial Leaf Nodule Symbiosis I. M. MILLER

Fracture Properties of Plants J. F. V. VINCENT


xxi

Contents of Volume 18 Photosynthesis and Stomata1 Responses to Polluted Air, and the Use of Physiological and Bacterial Responses for Early Detection and Diagnostic Tools H . SAXE

Transport and Metabolism of Carbon and Nitrogen in Legume Nodules J. G. STREETER

Plants and Wind P. VAN GARDINGEN and J. GRACE

Fibre Optic Microprobes and Measurement of the Light Microenvironment within Plant Tissues T. C. VOGELMANN, G. MARTIN, G. CHEN and D. BUITRY

Contents of Volume 19 Oligosaccharins S. ALDINGTON and S. C. FRY

Are Plant Hormones Involved in Root to Shoot Communication? M. B. JACKSON

Second-Hand Chloroplasts: Evolution of Cryptomonad Algae G. 1. McFADDEN

The Gametophyte43porophyte Junction in Land Plants R. LIGRONE, J. G. DUCKETT and K. S. RENZAGLIA

xxii


Contents of Volume 20 Global Photosynthesis and Stomata1 Conductance: Modelling the Controls by Soil and Climate F. 1. WOODWARD and T. M. SMITH

In vivo NMR Studies of Higher Plants and Algae R. G. RATCLIFFE

Vegetative and Gametic Development in the Green Alga Chlamydomonas H. VAN DEN ENDE

Salicylic Acid and its Derivatives in Plants: Medianes, Metabolites and Messenger Molecules W. S. PIERPOINT

Contents of Volume 21 Defense Responses of Plants to Pathogens E. KOMBRINK and I. E. SOMSSICH

On the Nature and Genetic Basis for Resistance and Tolerance to Fungal Wilt Diseases of Plants C. H. BECKMAN and E. M. ROBERTS

Implication of Population Pressure on Agriculture and Ecosystems A. H. EHRLICH

Plant Virus Infection: Another Point of View G. A. DE ZOETEN

The Pathogens and Pests of Chestnuts S. L. ANAGNOSTAKIS


xxiii

Fungal Avirulence Genes and Plant Resistance Genes: Unraveling the Molecular Basis of Gene-for-Gene Interactions P. J. G. M. DE WIT

Phytoplasmas: Can Phylogeny Provide the Means to Understand Pathogenicity B. C. KIRKPATRICK and C. D. SMART

Use of Categorical Information and Correspondence Analysis in Plant Disease Epidemiology S. SAVARY, L. V. MADDEN, J. C. ZADOKS and H. W. KLEIN-GEBBINCK

Contents of Volume 22 Mutualism and Parasitism: Diversity in Function and Structure in the “Arbuscular” (VA) Mycorrhizal Symbiosis F. A. SMITH and S. E. SMITH

Calcium Ions as Intracellular Second Messengers in Higher Plants A. A. R. WEBB, M. R. McAINSH, J. E. TAYLOR and I. M. HETHERINGTON

The Effects of Ultraviolet-B Radiation on Plants: A Molecular Perspective B. R. JORDAN

Rapid, Long-Distance Signal Transmission in Higher Plants M. MALONE

Keeping in Touch: Responses of the Whole Plant to Deficits in Water and Nitrogen Supply A. J. S. McDONALD and W. J. DAVIES

xxiv


Contents of Volume 23 The Value of Indexing for Disease Control Strategies D. E. STEAD, D. L. EBBELS and A. W. PEMBERTON

Detecting Latent Bacterial Infections S. H. DE BOER, D. A. CUPPELS and R. GITAITIS

Sensitivity of Indexing Procedures for Viruses and Viroides H. HUTTINGA

Detecting Propagules of Plant Pathogenic Fungi S. A. MILLER

Assessing Plant-Nematode Infestations and Infections K. R. BARKER and E. L. DAVIS

Potential of Pathogen Detection Technology for Management of Diseases in Glasshouse Ornamental Crops I. G. DlNESEN and A. VAN ZAAYEN

Indexing Seeds for Pathogens J. LANGERAK, R. W. VAN DEN BULK and A. A. J. M. FRANKEN

A Role for Pathogen Indexing Procedures in Potato Certification S. H. DE BOER, S. A. SLACK, G. VAN DEN BOVENKAMP and I. MASTENBROEK

A Decision Modelling Approach for Quantifying Risk in Pathogen Indexing C. A. LEVESQUE and D. M. EAVES

Quality Control and Cost Effectiveness of Indexing Procedures C. SUTULAR


xxv

Contents of Volume 24 Contributions of Population Genetics to Plant Disease Epidemiology and Management M. G . MILGROOM and W. E . FRY

A Molecular View Through the Looking Glass: the Pyrenopezizu brassicae-Brussica Interaction A. M. ASHBY

The Balance and Interplay Between Asexual and Sexual Reproduction in Fungi M. CHAMBERLAIN and D. S. INGRAM

The Role of Leucine-Rich Repeat Proteins in Plant Defences D. A . JONES and J. D. G. JONES

Fungal Life-styles and Ecosystem Dynamics: Biological Aspects of Plant Pathogens, Plant Endophytes and Saprophytes R. J. RODRIGUEZ and R. S. REDMAN

Cellular Interactions between Plants and Biotrophic Fungal Parasites M. C. HEATH and D. SKALAMERA

Symbiology of Mouse-Ear Cress (Arabidopsis thuliuna) and Oomycetes E . B. HOLUB and J. L. BEYNON

Use of Monoclonal Antibodies to Detect, Quantify, and Visualize Fungi in Soils F. M. DEWEY, C. R. THORNTON and C . A. GILLIGAN

Function of Fungal Haustoria in Epiphytic and Endophytic Infections P. T. N. SPENCER-PHILLIPS

xxvi


Towards an Understanding of the Population Genetics of Plant-Colonizing Bacteria

B . HAUBOLD and P. B. RAINEY Asexual Sporulation in the Oomycetes A. R. HARDHAM and G . J. HYDE

Horizontal Gene Transfer in the Rhizosphere: a Curiosity or a Driving Force in Evolution? J. WOSTEMEYER, A. WOSTEMEYER and K. VOIGT

The Origins of Phytophthora Species Attacking Legumes in Australia J. A. G. IRWIN, A. R. CRAWFORD and A. DRENTH

SERIES PREFACE

Advances in Botanical Research is one of Academic Press’ longest standing serials, and has established an excellent reputation over more than 30 years. Advances in Plant Pathology, although somewhat younger, has also succeeded in attracting a highly respected name for itself over a period of more than a decade. The decision has now been made to bring the two serials together under the title of Advances in Botanical Research incorporating Advances in Plant Pathology. The resulting synergy of the merging of these two serials is intended to greatly benefit the plant science community by providing a more comprehensive resource under one “roof”. John Andrews and Inez Tommerup, the previous editors of Advances in Plant Pathology, are now on the editorial board of the new series. Our joint aim is to continue to include the very best articles, thereby maintaining the status of a high-impact factor review series.


PREFACE

The idea for this thematic volume came from the realization that it is now just over 20 years since Wagner and Siegelman (Science 190, 1298-1299) published a method for the isolation of large numbers of intact higher plant vacuoles by the disruption of protoplasts from petals of tulip and Hippeasfruin. This breakthrough allowed many long-standing theories about the solute and enzymatic composition of the vacuole to be tested and questions about the nature of transport systems in the vacuolar membrane to be addressed. Since that time our knowledge of the role of the plant vacuole has increased enormously as a battery of powerful techniques arising from developments in electrophysiology (e.g. patch clamping) and molecular biology have allowed the activities of this large organelle to be unravelled in ever more increasing detail. These methods have identified new functions not previously ascribed to the organelle, and the vacuole is now recognized as a truly multifunctional compartment with roles in solute storage, stress responses, intracellular signalling, intracellular digestion and plant defence. The selective transport properties of the vacuolar membrane are central to many of the functions of the vacuole and serve to retain toxic materials within the vacuolar sap while exchanging metabolically useful compounds with the cytosol in response to the needs of the cell. Appropriately, therefore, there has been much emphasis on understanding the nature of transporters in the vacuolar membrane, and these have led to the description of two H + pumps in this membrane, a number of H+-coupled solute transporters, a variety of ion channels and a multigene family of aquaporins (the TIPS). Intensive studies of the Ca2+ channels in the vacuolar membrane in relation to intracellular signalling have revealed the complex nature of the regulation of transport at the vacuolar membrane, which may provide a paradigm for how the activities of the vacuole are integrated with those of the rest of the cell. The molecular definition of membrane and soluble components associated with the vacuole has led to new opportunities to study the process of vacuole formation and, again, developments in cytological methods, such as confocal microscopy and immunolocalization techniques, have allowed a better understanding of how the vacuole is formed and maintained, and the pathways followed by different components as they move to their final destinations in the vacuolar membrane or lumen. The various chapters in this volume record all of these developments and

PREFACE

xxx

point out opportunities for further research. We hope that the various contributions will provide a useful compendium of the current state of understanding of vacuoles at a variety of levels and will encourage greater investigation of this unique organelle. Roger A. Leigh Dale Sanders

The Biogenesis of Vacuoles: Insights from Microscopy

F. MARTY

Laboratoire de phytoBiologie Cellulaire, Vniversitk de Bourgogne, Dijon, France

I. 11.

Introduction

................................................................................

Sorting of Vacuolar Precursors

......................................................

1 2

111.

The Autophagic Pathway .............................................................. A. Starvation-induced Cellular Autophagy in Plant Cells ................

IV.

Endocytic Pathways and Vacuole Biogenesis ................................... 17 A. Endocytic-like Plasma Membrane Resorption After Secretion ..... 25 B. Plasma Membrane Internalization in Water-stressed Cells ........... 25

V. VI.

Ontogeny of Vacuoles Specialized in Protein Storage

.......................

Conclusions ............................................................................... Acknowledgements ..................................................................... References ................................................................................

7 13

27 32 33 33

I. INTRODUCTION The vacuole of plant cells, like the vacuole of algae and fungi (including yeasts) is an acidic compartment which shares some of its basic properties with mammalian lysosomes. It is a multifunctional organelle, with specific properties which are central to the cellular strategies of development in plants. Vacuoles of plant cells were discovered with the early light microscope and, as meant by the etymology of the word, were subsequently defined as the cell space empty of cytoplasmic matter. As usual in experimental science, methodological and instrumental progress has delineated differently the Advances in Botanical Research Vol. 25 incorporating Advances in Plant Pathology ISBN 0-12-005Y25-8

Copyright 0 lW7 Academic Pre5s Limited All rights of reproduction in any form reserved

2

F. MARTY

present definitions of vacuoles, which largely depend on the tools and concepts used for their study. For example, electron microscopy made us aware that, in mature plant cells, the conspicuous central vacuole surrounded by the vacuolar membrane (tonoplast) is but one component from a series of specialized membranebound compartments and transitional elements referred to collectively as the secretory system of plant cells. This secretory system consists of the endoplasmic reticulum (ER), Golgi apparatus, and vacuole as well as transport vesicles connecting these compartments and the plasma membrane. This is a highly dynamic and intricate system through which proteins, lipids, and polysaccharides flow. A variety of inclusions (e.g. protein storage and endocytic vesicles) are provisionally recorded as “vacuole-like” components. Biochemistry and physiology of the vacuole have been made possible by the isolation of intact vacuoles. A specific, although yet incomplete, set of soluble and membrane proteins with unique properties has been specifically ascribed to the vacuole. Recent analyses of the gene products affected in vacuolar mutants have provided new insights into the intimate organization of the pathways leading to the vacuole in yeast. Molecular tools now allow important questions to be addressed regarding the transport, targeting and assembly of the vacuolar components in plants. Therefore, present definitions of the vacuole are still largely operational and derived from a combination of microscopy, biochemistry, genetics and molecular biology. In this review summarizing our current understanding of their ontogeny, vacuoles are provisionally defined as the intracellular compartments resulting from the end-point (terminal) differentiation of the secretory pathway in plant cells. They are ontogenetically linked with other components of the vacuolar system. Experimental evidence suggests that material within the vacuolar system in plants derives both from a direct intracellular biosynthetic pathway and a confluent endocytic pathway. Variations on this theme are documented by recent studies on vacuoles specialized in protein storage. The reader is referred to other chapters of this volume for detailed information and to previous reviews (Matile, 1975, 1978, 1987; Marty ef al., 1980; Boller and Wiemken, 1986; Chrispeels, 1991; Wink, 1993) for more complete summaries of earlier work.

11. SORTING OF VACUOLAR PRECURSORS All results from recent plant cell studies support the view that the basic mechanisms that specifically organize the endomembrane system of eukaryotes are highly conserved. In plant cells, as in animal cells and yeasts, forward (anterograde) transport through the vacuolar pathway begins at the

VACUOLE BIOGENESIS

3

endoplasmic reticulum. Protein-secreting cells contain abundant E R cisternae, interconnected by tubular elements at their edges (Marty, 1973; Parker and Hawes, 1982). Secretory proteins that are inserted into, or translocated across, the ER membrane contain sorting signals required for the targeting and the retention in most of the compartments along the secretory pathway (Chrispeels and Raikhel, 1992). For some membrane and soluble proteins, e.g. the prolamins of the ER-protein bodies (see below), the target organelle is the ER itself, and they are not transported further. All other proteins are carried to the Golgi apparatus via a still elusive vesiculo-tubular intermediate (or “transitional”) compartment (Hellgren et al., 1993). The Golgi stacks in plant cells, like their counterparts in animal cells, consist of three (cis. medial, and trans) discrete groups of cisternae which can be distinguished on the basis of distinct cytochemical reactivities (Staehelin and Moore, 1995). The post-Golgi compartment nearest to the Golgi stack consists of a polygonal tubular network at the trans side of the Golgi stack (Marty, 1973, 1978). It was originally called GERL, an acronym given to the region of smooth ER that is located at the trans aspect of the Golgi apparatus and that appears to produce Lysosomes (Novikoff, 1976). It is now referred to as the trans-Golgi network or TGN (Griffiths and Simons, 1986). Although this relatively simple model of Golgi organization is substantiated by less experimental support in plant than in animal cells, it influences most current views of Golgi functions in plants. Along the early biosynthetic pathway (Fig. 1), newly synthesized proteins undergo an elaborate series of covalent modifications which begin in the E R and are continued further in the Golgi apparatus and post-Golgi compartments (Harris and Watson, 1991). The Golgi apparatus is mainly involved in the synthesis of complex polysaccharides and the sequential modification (mostly glycosylation) of soluble and membrane secretory proteins (Staehelin et al.. 1991; Zhang and Staehelin, 1992; Satiat-Jeunemaitre and Hawes, 1993a; Fitchette-Laine et al., 1994). Newly processed vacuolar proteins thus transit through the early stages of the secretory pathway together with proteins that are destined to be exported in the extracellular medium or delivered to the plasma membrane. Vacuolar components are probably sorted and diverted out of this common secretory pool in the TGN and delivered to the vacuole via an intermediate (pre/provacuolar) compartment (Marty, 1978). The TGN appears to vary in size according to specific cell requirements. For instance, in actively vacuolating plant cells, the TGN is widely extended and can be readily characterized morphologically and cytochemically, whereas it is barely detectable in cells secreting the polysaccharides for the extracellular matrix. In vacuolating cells from the root meristem, the TGN consists of a twisted, smooth surface, polygonal meshwork of anastomosing

4

F. MARTY

Fig. 1 . Endoplasmic reticulum (ER) and Golgi apparatus (G) from the secretory pathway in suspension-cultured cells of sycamore (Acer pseudoplatanus L.). The cell wall (CW) and the vacuole (V) are terminal compartments of the pathway. The bar represents 0.5 pm. tubules extending from small disc-like cisternae facing t h e Golgi stacks at a distance greater than that between Golgi cisternae in the stacks. T h e TGN is a specialized organelle that is possibly common to several Golgi stacks in the cell (Marty, 1973). Tubules from the TGN, even when located far from the cisternal areas, possess clathrin-coated swellings with internal small

VACUOLE HIOGENESIS

5

vesicles of unknown origin. Their possible identity with the partially coated reticulum (PCR) described in some plant materials (Beevers, 1996; Tanchak ef al., 1984; Pesacreta and Lucas, 1985; Hillmer et a / . , 1988; Hippe et al., 1989; Fowke el a/.. 1991) remains to be cstablished. Smooth-surfaced vesicular carriers as well as coated vesicles of different compositions have been shown to operate along the discontinuous elements of the early biosynthetic pathway. However, the identification of clathrinand nonclathrin-coated vesicles has to be refined in order to obtain a hetter evaluation of the respective contributions of the different classes of vesicular carriers to the pathway. Although direct experimcntal evidence is still lacking, the TGN from plant cells, like its homologue in animal cells, is most likely an acidic compartment. Acidification of the interior of the organelle and the formation of a positive membrane potential inside can be achieved by two distinct electrogenic H + pumps: an H+-ATPase (Churchill ef al., 1983; Bennett et ul., 3984; Sze et al., 1992) and an H + pyrophosphatase (PPase) (Walker and Leigh, 1981: Rea and Poole, 1993; Sato et nl., 1994). I n this respect, it is noteworthy that a vacuolar-type proton pump was localizcd in the membranes of the Golgi cisternae (Ali and Akazawa, 1986; Hurlcy and Taiz, 1989). Proton gradients across the TGN membrane are essential for proper sorting of proteins in the post-Golgi vacuolar pathway, as indicated by the disruptive effects of pH perturbants on the sorting of vacuolar proteins at the TGN (Boss et al., 1984; Bednarek and Raikhel, 1992; Gomez and Chrispeels, 1993). Most distinctively, the TGN is specifically reactive for acid phosphatase (Marty, 1978). The reaction is probably produced by resident enzymes o r by phosphatases in transit to the vacuole. Other acid hydrolases, such as a thiolacetic acid esterase, were also detected by cytochcmical reaction in the TGN. The acid hydrolases, like other secretory proteins, that have moved through the Golgi complex condense in the lumen of the TGN before they are sorted into developing vesicles, which can be considered as the prime precursors of the vacuole. Kinetic characteristics and differential sensitivity to pharmacological agents suggest that several distinct sorting machineries could operate, and vacuolar membrane proteins such as TIP (for Tonoplast Intrinsic Protein; see Johnson et al., 1990) may be sorted by a mechanism at least partly different from those used for soluble vacuolar proteins (Gomez and Chrispeels, 1993; Matsuoka et al., 1995) but the routes are likely very similar. A major intrinsic protein similar to the aquaporin y-TIP and specifically present in the tonoplast of storage parcnchyrna cells o f beetroot has been immunolocalized, but at a low frequency, in the membranes of the ER, Golgi-derived vesicles and the TGN (Marty-Mazars ef a / ., 1995). Similarly, the V-ATPase which is associated mainly with thc vacuolar membrane in mature cells has been shown to be associated with the ER, GoIgi-derived vesicles, provacuoles and tonoplast i n vacuolating root cells (Herman et ul.,

6

F. MARTY

1994). The immunological detection of the peripheral regulatory subunit B of the V-ATPase by monoclonal antibodies suggests that it becomes associated with the membrane integral sector of the V-ATPase complex on the E R in the early stages of vacuolation. Most of the assembled V-ATPase complexes are destined for the tonoplast of mature vacuoles. A few are possibly diverted to other destinations such as the plasma membrane in rapidly differentiating cells (Herman et al., 1994), as is also suspected for other tonoplast proteins (Robinson et al., 1996a). Part of our understanding of the membrane traffic through the early steps of the biosynthetic pathway has come from the use of pharmacological agents, such as brefeldin A (BFA) and monensin, which disorganize the flow of membrane (Bednarek and Raikhel, 1992; Satiat-Jeunemaitre and Hawes, 1994). Immunocytochemical and biochemical studies have shown that the fungal metabolite BFA inhibits the early steps of both the exocytic pathway to the cell surface (Satiat-Jeunemaitre and Hawes, 1992, 1993b; Driouich et al., 1993; Henderson er af.,1994; Schindler et al., 1994) and the transport of soluble proteins to the vacuole (Holwerda et af., 1992; Gomez and Chrispeels, 1993). Concomitantly, it was shown to perturb reversibly the positioning and structure of the Golgi stacks (Satiat-Jeunemaitre and Hawes, 1993b). At low concentrations, BFA induces an increase in the number of trans-Golgi cisternae and trans-Golgi-derived vesicles. At higher concentrations it causes a vesiculation and dissociation of the Golgi stacks as well as the swelling of the E R cisternae (Driouich et al., 1993). To date, however, there is no evidence that BFA induces the reabsorption of the entire Golgi stack into the ER as it does in animal cells (Satiat-Jeunemaitre and Hawes, 1994). Morphological and immunocytochemical results suggest that the TGN is the main structural target of the sodium ionophore monensin (Boss et af., 1984; Shannon and Steer, 1984; Zhang et al., 1993; Satiat-Jeunemaitre et al., 1994). By altering the Na+ and the H+ gradients across biological membranes, monensin probably affects the acidification of the TGN and disrupts the sorting of vacuolar and exported secretory molecules at the exit site from the Golgi complex. It induces the osmotic swelling of the TGN, the production of large smooth vesicles and the accumulation of coated vesicles in the vicinity of the TGN and in association with the large smooth vesicles (Fig. 2). The biosynthesis and/or processing machinery of polysaccharides is clearly altered by the drug. In contrast, results reported so far on its effects on the secretion and processing of proteins are diverse and vary with the plant materials, drug concentrations and experimental methods (Satiat-Jeunemaitre et al., 1994; Zhang et al., 1996). In mesophyll protoplasts from transgenic tobacco plants transformed with genes encoding the soluble lectin phytohaemagglutinin (PHA) and the tonoplast intrinsic protein a-TIP, both BFA and monensin were shown to block the transport of PHA to the vacuole, but neither drug stopped the arrival of a-TIP in the tonoplast (Gomez and Chrispeels, 1993). As suggested by in situ

VACUOLE BlOGENESlS

7

Fig. 2. Modifications of thc rrrms-(iolgi network ( T G N ) and accumulation of coated vesicles (arrowheads) after inonensin treatment of suspension-cultured cells of sycamore ( A w r ~ ) . s ~ ~ i ~ ~ l ~ ) / ~ lL. ( ~)./ f lCi; i, i ~Golgi .s apparatus. The bar represents 0 . 5 pin.

labelling, the tonoplast and soluble proteins follow very similar, if not identical, intracellular pathways (Vitale rf ol., 19x4; Grcenwood and Chrispeels, 1985; Marty-Mazars et al.. 1995: Pueyo et (11.. 1995). Taken together with the results of transport kinetics studies (Gome, and Chrispeels, 1993), these findings suggest that soluble proteins such as PHA and membrane proteins such as a - T I P reach their vacuolar destinations at different speeds by using different molecular mechanisms on the same route.

111. THE AU'TOPHAGIC PATHWAY Starting from the premise that a comprclicnsive morphological characterizdtion o f an organelle can help in undcrstanding its function, the intermediate compartment which lies between the trans-Golgi sorting site and the vacuole has been carefully examined. This compartment is predominant in cells where new vacuoles a r e being formed. Its discovery comes largely from morphological and cytochemical studies of actively vacuolating cells in the root meristem using conventional as well as high-voltage electron microscopes (Marty. 1978). T h e largely undifferentiated cells from the root meristems contain neither

8

F. MAKTY

Fig. 3. Provacuoles (PV) act as an intermediate Compartment along the biogenetic pathway between the Golgi apparatus (G) and the vacuole (V). Transient figures suggestive of fusion events between provacuoles and multivesicular bodies (MVB) indicate that provacuoles might be the confluent site of endocytic and vacuolar pathways. Suspension-cultured cell of sycamore (Acer pseucioplafunus L.). The bar represents 0.5 p m . vacuole nor identifiable vacuolar precursors, but ER cisternae and Golgi stacks are present. In these cells, numerous vesicles are budding off the T G N at the exit site from the Golgi stacks. These vesicles, which make a novel intermediate compartment along the vacuolar pathway, have been collectively referred to as provacuoles because ontogenetically they stand as the immediate precursors of the vacuole (Fig. 3). These TGN-derived vesicles mediate the transport between the Golgi complex and the vacuole a n d , therefore, may represent a physiological prevacuolar compartment along t h e vacuolar biogenetic pathway. In situ localization of acid phosphatase activity and selective staining by electron-opaque tracers have shown that the T G N a n d the provacuoles react similarly and have structural connections. Nascent provacuoles apparently bud from nodes of the T G N meshwork. Their diameter (-0.100pm) is readily larger than the diameter (-0.015 p m ) of the tubules from the T G N . T h e availability of high-voltage microscopes has made it possible to appreciate the three-dimensional organization of this post-Golgi compartment (Fig. 4). Striking sequences of membrane tubulation a n d cytoplasmic

confinement have been obtained (Marty, 1978, 1980, 19x3). Vesicular provacuoles, still close to the TGN, grow into extensive tubules having roughly the same bore (0.1 p m ) as the diameter of the vesicles from which they derive. Numerous vesicles. resulting presumably from micro-invaginations of the membrane of the provacuoles themselves, fill the lumen of the tubular provacuoles. The exact identity of the constituents and trafficking dynamics through this intermediate pre/provacuolar compartment remain obscure. Isolation. characterization and immunolocalization o f its components have yet to be achieved. Likewise, the molecular mechanisms that control its formation and thc flux through i t remain unknown. The discovery of molecular markers for this provacuolar compartment may provide a genetic and biochemical mcans to examine these processes in detail. Identification of the components involved in the sorting of provacuoles requires the characterization of molecular markers and their in situ localization at high resolution. The provacuoles are good candidates to house, for instance, the newly discovered putative plant vacuolar 1995). sorting receptors (Kirsch et ol., 1994; Bassham ef d., Moreover, the intermediate provacuolar compartment might be a critical juncture in post-Golgi traffic. Provacuoles might be a site where the endocytic and vacuolar biogenetic pathways converge near the late-Golgi compartment. As reported below, endocytic tracers have been observed in a variety of vesicles and endosomes near the Golgi apparatus and the vacuole (Joachim and Robinson, 1984; Hubner et d., 1985; Fowke et ul., 1991; Oparka el 01.. 1991, 1993; Low et ul., 1993; Villanueva et al., 1993; Low and Chandra, 1994). On the basis of immunocytochemical studies it has been hypothesized recently that a minority of tonoplast proteins can follow an alternative exocytic-endocytic circuitous route to the vacuole (Robinson et af., 1996a). According to this suggestion, a few newly synthesized proteins destined to the tonoplast, including the two Hf-pumps (V-ATPase and H+-PPase) and the aquaporin, y-TIP, can escape the direct intracellular pathway to the vacuole and first move to the plasma membrane by exocytosis and are then targeted to the tonoplast in a way which resembles the endocytic pathway to lysosomes in animal cells (Kornfeld and Mellman. 1989). The extensive tubular provacuoles in actively vacuolating cells may represent an exaggerated version of a uhiquitous prevacuolar compartment similar to that described in mammalian cells and yeasts (Griffiths et uf.. 1988; Piper et al., 1995). The proliferation into this exaggerated form would occur either if membrane flow o u t of this compartment were slowed down, or if the membrane input from the Golgi apparatus and/or the endocytic tributary were increased. In a rather synchronous manner. the tubulur provacuoles produce clusters of digitate cxtcnsions (-0. I p m i n diameter) which form cage-like structures enclosing a portion of cytoplasm (Marty, 1978). Adjacent tubules within the cage fuse through transitory palmar joints (Fig. 5 ) . As fusion

F. MARTY

Fig. 4. Tubular provacuoles in actively vacuolating cells from the root meristem of Euphorbia charircius L. Provacuoles are selectively stained by the zinc iodideosmium reaction (see Marly, 1978). The specinlen ( 2 p m thick) was examined without counterstain at 2.5 MV with the Toulousc 3 MV electron microscope. The bar represents 1 p m .

VACUOLE BlOCiENESlS

11

Fig. 5. Sequential stages o f cytoplasmic confinement by provacuoles involved in cellular autophagy. 1, provacuoles forming a cage-like trap; 2, formation of an cnveloping cavity when the bars of the cage fuse; 3 , the enveloping system is complete and forms a functional autophagosome. Note that tuhular provacuoles (PV) are still present in the cytoplasm. G. Golgi stacks. Cells were from the root meristem of Eicphorhia characim L. Provacuoles were sclectively stained by the zinc iodideosmium reaction. ?'he specimen (2 prn thick) was examined without counterstain at 2.5 MV with the Toulouse 3 MV electron microscope. The bar represents 1 pm.

12

F. MARTY

progresses in a zipper fashion, a continuous cavity enwraps the portion of cytoplasm tightly. When observed by the conventional electron microscope, thin sections through these ball-shape structures are interpreted as autophagosomes at early stages of formation (see Fig. 2 in Marty, 1978). A narrow ringlike cavity bounded by inner and outer membranes encircles a piece of cytoplasm. Equatorial thin sections at earlier steps in the formation of cage-like structure show cross-sections of the tubular provacuoles. Cytochemical studies show that the TGN, provacuoles and autophagosomes share unique properties. They are acidic compartments and contain acid phosphatase as well as other lysosomal acid hydrolases which are thought to be in transit to the vacuole. Morphological and cytochemical observations have shown that the cytoplasm in the autophagosome is digested sometime after it has been totally closed off. Thc digestive enzymes are most likely released from the surrounding cavity as soon as the inner boundary membrane has lost its tightness. Indeed, it was hypothesized that the inner and outer membranes of the sequestering envelope at the early stages of its completion are equally impermeable to the acid hydrolases because they both derive from the junction of adjacent tubular provacuoles (Marty, 1978). The integrity of the outermost membrane is well sustained by anabolic reactions of the surrounding cytoplasm. In contrast, the innermost membrane which is topologically separated from the cytoplasm cannot benefit from the thermodynamic upkeep, and, therefore, becomes disorganized and leaky. The hydrolytic enzymes are thus released from the surrounding cavity into the interior of the autophagosome and digest their substrates. Upon completion of the digestive process, a typical vacuole is formed. Whereas ribosomes, ER and occasionally other sequestered organelles are rapidly degraded, the ghost of the inner membrane from the autophagic system is usually the last structure to disappear inside the vacuole. This is likely to be due to the slow rate of digestion of the phospholipids. The outer membrane remains impermeable to acid hydrolases, whose digestive activities are consequently confined within the forming vacuole. The outer membrane thus prevents cellular autolysis, and becomes the tonoplast. All the young vacuoles formed more or less simultaneously in the same cell fuse together and enlarge to give rise to a few large vacuoles. In elongating tissues, mRNAs specific for the aquaporin y-TIP accumulate at the time of vacuolation or shortly thereafter and then subside as the cells reach their full size (Ludevid et al., 1992). These results suggest that the tonoplast aquaporin is needed for the specific transport of water through the tonoplast which results in vacuole enlargement. Provacuoles are occasionally seen to merge directly with the vacuole. This process would account for tonoplast extension and vacuolar content accretion during cell enlargement. Although the provacuoles are much less extended i n the differentiated fully vacuolated cells, they likely act as a physiological intermediate compartment along the biogenetic pathway between the Golgi

VACUOLE BIOGENESIS

13

apparatus and the vacuole and might be the confluent site of the endocytic and vacuolar pathways (see Fig. 3). Likewise, the molecular machinery that controls the traffic through it will determine whether it is analogous to the prevacuolar “class E” compartment in yeast (Raymond et al.. 1992; Horazdovsky et al.. 1995; Piper rf af., 199.5) and to the prelysosomal compartment in mammalian cells (Griffiths ef al., 1988). Early fractionation studies on vacuoles (Leigh and Branton, 1976; Marty and Branton, 1980) have made possible the generation of antibodies specifically directed against the purified fractions. The tonoplast from the central vacuole of fully differentiated ce!ls is now being probed by polyclonal (Figs 6 and 7) and monoclonal (Fig. 8) antibodies raised against specific proteins (Hurley and Taiz, 1989; Herman et d., 1994; Dozolme et af.,1995; Marty-Mazars et al., 199.5).The antibodies are used to label the same proteins along their biosynthetic pathways. Early precursors and intermediate compartments including ER, Golgi-derived vesiclcs and provacuoles are labelled but at a frequency lower than that of the tonoplast. Patches of plasma membrane were occasionally decorated, suggesting that isoforms of tonoplast proteins might be present in the plasma membrane. Alternatively, a small fraction of tonoplast proteins might rcach its final destination after a circuitous transport to and from the cell surface (see above; Robinson et al., 1996a). These results illustrate the difficulty in understanding dynamic cellular events. It must be remembered that the location in which a molecule is found is a kinetic effect that may result from one slow step in a flowing process. Thus, vacuolar proteins located in the plasma membrane may be either missorted or in transit rather than having a physiological role at the cell surface. The formation of a central vacuole is restricted to a few cells in the meristems of a plant and these cells contain vacuoles at different stages of development. This is a serious drawback in the biochemical analysis of vacuolation when large populations o f cells at a same stage of vacuole formation are needed. Experimentally, evacuolated protoplasts which synchronously regenerate a new central vacuole have provided an alternative model system (Lorz et al., 1976; Griesbach and Sink, 1983; Burgess and Lawrence, 1985; Hortensteiner et nl., 1992). Kinetics studies conibining electron microscopy and biochemistry have shown that the vacuoles are regenerated after 20 h . During the reformation of vacuoles. soluble hydrolases as well as both proton pumps (V-ATPasc and PPase) and other tonoplast polypeprides are synthesized (Hiirtensteiner et al., 1992. 1994). A

STAKVAJION-INDUCED CEL.1.ULAR ACJTOPHAGY IN PLANT CELLS

Cellular autophagy has been shown to be induced by sucrose starvation in sycamore (Douce er al., 1995; Aubert rt (11.. 1996, and references herein). tobacco (Moriyasu and Ohsurni, 199h), rice (Chen rt al., 1994) and yeast

14

F. MART?

Fig. 6. lmmunolocalization o f a major intrinsic protein related t o TIPS on the tonoplast from the shoot meristeniatic cells ot' beetroot. (a) Immunofluorescence labelling of the tonoplast with a polyclonal monospecific antiserum directcd against :I tonoplast intrinsic protein of 27 kDa; (b) the same field observed under phase contrast optics. The bar represents 10 Fm. (From Marty-Mazars er a/. (IgYS), with permission from Wissenschaftliche Verlagsgesellschaft, Stuttgart.)

VACUOLE BIOGENESIS

15

Fig. 7. Electron microscopic cryosection prepared from the shoot meristem of beetroot and incubated with the same polyclonal antiserum as in Fig. 6 (immunogold electron microscopy localization of the tonoplast intrinsic protein related to TIPS). V,vacuole; N, nucleus. The bar represents I0 pm. (From Marty-Mazars et ul. (1995), with permission from Wissenschaftliche Verlagsgesellschaft . Stuttgart.)

16

F. MARTY

Fig. 8. Immunolocalization of ,,,bTIP106, a polypeptide of 106 kDa recognized by the monoclonal antibody Tem 106 in the tonoplast of cells from the shoot meristem of cauliflower (Brassica oleracea var. hotrytis). (a) lmmunofluorescence labelling; (b) same field observed using Nomarski optics. The bar represents 1 0 p m . (From Dozolme et af. (1995). with permission from The Company of Biologists Limited, Cambridge.)

VACUOLE BIOGENESIS

17

(Baba er ul., 1994). In sycamore suspension-cultured cells, autophagic vacuoles are formed in the cytoplasm as early as 14 h after the cells arc deprived of sucrose when all the endogenous reserves of carbohydrates (starch and sucrose) have been consumed (Aubert et al., 1996). Portions of cytoplasm are first sequestered in double membrane-bounded envelopes and then eventually digested (Figs 9 and 10). The small vacuoles thus newly made in the cytoplasm protrude into the central vacuole before they eventually become incorporated into it. Many small vacuoles thus accumulate in t h e central vacuole of starved cells. Their membrane remains visible for some time before being totally digested. Coordinated nuclear magnetic resonance studies on the same cells demonstrate that a massive breakdown of membrane polar lipids parallels the formation of autophagic vacuoles. In particular, phosphatidylcholines decrease whereas phosphorylcholines, which arc resistant to further degradation, steadily accumulated. These findings show that phosphorylcholine can be used, therefore, as a reliable biochemical marker of autophagy in sucrose-starved cells (Aubert et al., 1996). By replacing sucrose by glycerol or pyruvate it was also shown that the induction of cellular autophagy in these cells is controlled by the supply of mitochondria with respiratory substrate and not by the decrease in the concentrations of sucrose and hexose phosphates. Although they already contain a large central vacuole, cells are capable of reinitiating complete sequences of autophagy in their peripheral cytoplasm.

IV.

ENDOCYTIC PATHWAYS AND VACUOLE BIOGENESIS

Endocytosis is the process unique to eucaryotic cells whereby portions of the plasma membrane invaginate and pinch off to form membrane-bounded vesicles containing some of the ambient extracellular material as well as molecules adsorbed on the cell surface. Therefore, it generates intracellular vesicles and vacuole-like cavities with possible interactions with the central vacuole (Fig. 11). Endocytosis has been well documented in animal cells, where several pathways serving a variety o f functions have been described (Goldstein rt al., 1985; Rodman e1 al., 1990; Smythe and Warren, 1991; Robinson et of.. 1996b). The clathrin-dependent pathway involved in receptor-mediated endocytosis is constitutive, and it is the main endocytic route in many animal cells. Receptor-mediated endocytosis also occurs at caveolae, which are tiny membrane invaginations coated by a protein called caveolin. When both clathrin-coated pits and caveolae pinch off t o form closed vesicles at the plasma membrane some extraccllular tluid may be included. However, fluid phase endocytosis primarily involves uncoated membrane depressions much

18

F. MARTY

Fig. 9. Cellular autophagy in sycamore suspension-cultured cells deprived of sucrose. Autophagic vacuoles (A) at different stages of the digestive process are formed in the cytoplasm. V, vacuole. The bar represents 1 pm.

VACUOLE BIOGENESIS

I9

Fig. 10. Cellular autophagy in sycamore cells grown in sucrose-free culture medium. Small globular vacuoles (V*) resulting from a starvation-induced cellular autophagy are formed in the cytoplasm and subsequently taken into the central vacuole (V). N. nucleus. The bar represents 1 p m . (From Aubert Pt al. (1996), with permission from The Rockefeller University Press, New York.)

20

F. MAKTY

Fig. 11. Endocyiosis in a suspensiori-cultured cell of sycainore (Acer pseudoplrtrrrmrs L.). Intracellular vesicles rescmhling endosomes (E) suggestive of a fluid phase cndocytosis, multivesicuivr bodies (MVB). Golgi vesicles and coated vesicles are ohserved togcther with vacuoles ( V ) . 'The plasma membrane invaginates locally (arrowheads) and forms rnctnbrnnc-hounded vesicles. The bar represents 0.5 p m .

VACUOLE BlOGENESlS

21

larger than coated pits. In this way, even in the absence of a specific uptake of ligand, large quantities of plasma membrane can be returned to the cell interior. This internalization process is usually viewed as a regulatory mechanism essential for the adjustment of the cell surface to the cell volume during periods of intense secretory activity and exocytosis. The occurrence of endocytosis in plants has been debated on the basis of theoretical arguments (Cram, 1980; Raven, 1987; Saxton and Breidenbach, 1988; Robinson et ul., 1992). Early calculations led to concluding that endocytosis was not energetically feasible in permanently turgid plant cells. Thermodynamics of endocytosis makes it now clear that vesicle formation at the plasma membrane is possible in plant cells under low turgor pressure; it is now generally recognized that endocytosis does occur (Low and Chandra, 1994). In vacuolating root tip cells, morphological measurements shown that the vacuole volume fraction (vacuole volume/cell volume) increases considerably to make up 80% of the cell volume (Patel el uf., 1990). These observations demonstrate a requirement for the delivery of new additional membrane to match swelling rates. At steady state cell growth, a simple physical adjustment of membrane surface areas would require the insertion of equal amounts in the tonoplast and the plasma membrane. In elongating cells, where secrction of cell wall material is needed to maintain wall thickness as the cells enlarge, the surface area of plasma membrane that is added by exocytosis of Golgi-derived vesicles was estimated to be two to eight times greater than the surface area measured in fully elongated cells. This comparison suggests that plasma membrane recycling must occur in these cells (Phillips eruf.,1988; Samuels and Bisalputra, 1990). Stereological analysis of exocytic vesicles i n secretion-inhibited cells demonstrates that cells produce more plasma membrane than is needed for cell extension. Endocytosis is presumed to recycle all the excess membrane by internalizing an area equivalent to the whole cell plasma membrane every 10-200min (Emons and Traas, 1986; Steer and O'Driscoll, 1991; Satiat-Jeunemaitre ef ul. , 1996). It is inferred from these calculations that plasma membrane is recycled towards intracellular compartments and endocytosis most likely acts to remove excess plasma membrane added during exocytosis of secretory vesicles. Thc occurrence of endocytosis in plants is now substantiated by experimental evidence, although many aspects of the endocytic membrane traffic remain poorly understood. Conventional electron microscopy and subcellular fractionation studies provide evidence for plant cell structures morphologically and functionally homologous to components that were shown to perform endocytosis in animal cells (Low and Chandra. 1994; Beevers, 1996). Smooth membrane depressions and coated pits are common at the plasma membrane of actively growing cells. where they pinch off to form smooth and coated vesicles, respectively, with the same size and morphology as their 1985; Depta and Robinson, 1986; counterparts in animal cells (Mersey er d.,

22

F. MARTY

Fig. 12. Coated vesicles (CV) in a sycamore suspension-culturedcell. The clathrin coat is clearly visible on the cytoplasmic surface of the vesicle. ER, endoplasmic reticulum. The bar represents 0.25 pm.

Balusek et af., 1988; Coleman et af., 1991; Fowke et al., 1991; Robinson et ul., 1991). The coat is primarily made of clathrin, a complex protein with heavy and light chains assembled in triskelions to build a mixed hexagonaVpentagonal lattice (Fig. 12). There is now evidence for the existence of a P-adaptin (Holstein et al., 1994) and a-adaptin (D. G. Robinson, personal communication) in plant clathrin-coated vesicles, but other adaptor polypeptides as well as receptors remain to be identified. Both plant and animal clathrin cages have the same dimensions, although heavy and light chains of plant clathrin are about 10 000-15 000 Da larger than their animal homologues (Coleman et al., 1991; Demmer et al., 1993). Coated vesicles are also observed near the Golgi stacks and the partially coated reticulum from which they are probably separating (Emons and Traas, 1986; Coleman et al., 1987; Hillmer et al., 1988). These coated vesicles (60-70 nm) are smaller than those originating from the plasma membrane (100nm) and might have a different intracellular target such as vacuolar membranes (Harley and Beevers, 1989). Vesicles in the process of uncoating can be seen (Tanchak et af., 1988), and many coated as well as uncoated vesicles are distributed throughout the cytoplasm. The PCR (Fig. 13) has branching tubules which are presumably associated

VACUOLE BIOGENESIS

23

Fig. 13. Partially coated reticulum (PCR) in a sycamore suspension-cultured cell. Note the numerous coated regions (arrowheads) and t h e branching tubules interacting with double membrane-bounded vesicles (*). The bar represents 0.5 p m .

with the Golgi apparatus (Hillmer et al., 1988; Tanchak et al., 1988; Griffing, 1991). Occasionally, they have been found to be continuous with small vacuoles (Samuels and Bisalputra, 1990). However, its homology to the trans -Golgi network (TGN) or to the compartment of uncoupling for receptorligand (CURL; Rodman et al., 1990) remains controversial. Multivesicular bodies (see Fig. 1 1) contain numerous entrapped vesicles (50-100 nm) which are frequently observed as arising by invagination of the surrounding membrane (Tanchak and Fowke, 1987; Samuels and Bisalputra, 1990; Fowke et a l . , 1991). Distinctive plaques of clathrin-like coat and coated pits are present on the outer surface of the membrane. Multivesicular bodies are strikingly similar in plant and in animal cells, wherein their role in endocytosis is well characterized (Gruenberg and Howell, 1989; Kornfeld and Mellman, 1989). The functions of the endocytic pathway have been investigated in intact cells in planfa, in suspension-cultured cells, as well a5 in protoplasts (see review by Low and Chandra, 1994, and references within) by using macromolecular ligands (e.g. lectin-gold conjugates, silver-enhanced BSA-

24

F. MARTY

gold, cationized ferritin, fluorescein isothiocyanate (F1TC)-labelled elicitors, and FITC-labelled biotinylated proteins) or membrane-impermeant solutes (e.g. heavy metal salts such as those containing Pb2+ and La”). Cationized ferritin and lectin-gold conjugates were used as non-specific cell surface binding macromolecules whereas solutes like those of heavy metal salts and fluorescent dyes (e.g. lucifer yellow CH) labelled the fluid phase. Both markers are expected to delineate the pathway for fluid phase endocytosis as described in animal cells (Steinman et al., 1983). Alternatively, biological ligands such as FITC-labelled elicitors and FITC-labelled biotinylated proteins (Horn et al., 1989, 1990) were used to specify the receptor-mediated endocytosis as documented in animal cells (Smythe and Warren, 1991). Despite reservations about the extent and biological relevance of fluid phase endocytosis, experimental evidence from several laboratories is now supporting the existence of functional endocytic pathways in plant cells (see Low and Chandra, 1994, and references within). The different transport steps have been defined in various cells, usually by monitoring the loading (and enrichment) of a vesicle with a cargo marker. The movement of endocytosed markers has been putatively deduced and the overall kinetics estimated from the time course of intracellular distribution of the tracers using fluorescence and electron microscopy (see below). Non-specific markers havc been shown to bind to the plasma membrane of protoplasts within seconds after initial exposure (Fowke et al., 1991). The wall of intact cells acts as a physical barrier and, as a consequence, the binding of non-specific as well as specific ligands is delayed. In a few cases, ligands were shown first to bind to the cell surface and then to internalize in a temperature- and energy-dependent process (Horn et a[. , 1989. 1990; Fowke et al., 1991). Uptake usually involves coated pits, which quickly form coated vesicles free in the cytoplasm. Within 2 min the internalized markers are transferred to the tubular elements of partially coated reticulum (PCR), where they accumulate with time. The earliest Golgi labelling was shown to occur in peripheral vesicles only 4min after initial uptake (Joachim and Robinson, 1984; Hubner et al., 1985; Hillmer et d . , 1986; Samuels and Bisalputra, 1990; Fowke et a!., 1991). Markers were then observed in multivesicular bodies within G12 min, where they gradually accumulate, indicating that they could come either from the PCR and/or the Golgi apparatus. It has been suggested that the multivesicular bodies could fuse with the plasma membrane releasing back their contents in the intracellular medium (Fowke et al., 1991). However, a divergent route has been documented (Tanchak and Fowke, 1987; Record and Griffing, 1988). Indeed, internalized markers were shown to be translocated from the multivesicular bodies to the central vacuole or to peripheral (pro)vacuoles, suggesting that the multivesicular bodies ultimately fuse with vacuolar membranes. There is a general agreement on the loading of small vacuoles by the

VACUOLE BIOGENESIS

25

internalized molecules (Joachim and Kobinson, 1984: Griffing and Fowke, 1985; Hillmer el ul., 1986; Tanchak and Fowke, 1987; Record and Griffing, 1988; Owen et al., 1991). However, results differ concerning the central vacuoles. For instance, non-specific markers, such as cationized ferritin, and elicitors were found to be accumulated in the central vacuole of soya bean protoplasts and suspension culture cells. respectively (Tanchak and Fowke, 1987; Horn et al., 1989; Fowke et al., 1991) whereas the biotin-linked proteins were delivered primarily to the cytoplasm (Low et al., l993), and the central vacuole of bean leaf and carrot protoplasts could not be labelled with cationized ferritin or lucifer yellow (Joachim and Robinson, 1984; Hillmer rt al., 1989). Two distinct routes of internalization by clathrin-mediated endocytosis were therefore suggested to operate in plant cells: (1) plasma membrane to endosonial compartment (Golgi apparatus. PCR, niultivesicular bodies and/or provacuoles) and (2) plasma membrane to provacuoles and/or vacuoles (Low and Chandra, 1994). A

kNDOC’YTIC-LIKE PLASMA MEMBRANE RESORPTION AFTER S t C R E l ION

Novel intermediary structures showing plasma membrane internalization have been described in suspension culture cells of sycamore (Staehelin and Chapman, 1987). Dynamic membrane events are specifically associated with vesicle-mediated secretion and immediate plasma membrane recycling. The exocytic vesicle, after fusion with the plasma membrane and discharge of its content, is flattened and folds back. Concomitantly, the pore resulting from the membrane fusion is converted to a slit curling in a horseshoe shape. I t was suggested that membrane recycling after exocytosis o f secretory vesicles is made by a mixed mode that involves both internalization of membrane by endocytosis and of individual molecules by unknown mechanisms. Although coated pits are present at the plasma membrane, their number could not account for an exclusively endocytic recycling of the membranes. Instead, membranes directly derived from the exocytic vesicles follow unique intermediary configuration, resulting i n a retraction and reduction i n size, consistent with membrane recycling events t o t h e EK and/or Golgi apparatus (Staehelin and Chapman, 1987). B.

PLASMA MEMBRANE INTERNAL.I%ATION I N WATER-STRESSED CELLS

Kapid retrieval of plasma membrane to the cell interior, together with a fluid phase i n t e r n a 1iza t i on of extra ce I 1ul a r ( per i p I asm i c ) ma tc r i a I , occurs when protoplasts and intact plant cells function in the absence of positive turgor pressure (Stcponkus and Wicst. 1979; Oparka et al., 1990, 1991. 1993; Steponkus, 1991; Wartenberg rt ul., 1992). Such conditions can be met daily

26

F. MARTY

or seasonally when plants are challenged by a dearth of water. Under conditions of environmental stress such as drought, salt or freezing, cell dehydration occurs, and water loss from the vacuolar sap is among the most dramatic cellular responses to water deficit. An efflux of water from isolated protoplasts follows the decrease of the water potential of the suspending medium. Water efflux from the central vacuole results in a cell volume decrease. The dynamics of the plasma membrane of protoplasts during osmotic excursions has been thoroughly studied by using fluorescent probes, differential interference phase contrast optics and high-resolution video microscopy (Steponkus, 1991). The plasma membrane starts to flutter and invaginate, giving rise to numerous rather large intracellular vesicles (0.3-1 .0 pm in diameter). After several minutes, when osmotic equilibration is reached, the protoplasts recover their spherical shape. During the shrinking process, membrane material is deleted from the plasma membrane. Internalization of vesicles leads to a decrease in plasma membrane area and cell volume. The vesicles occur first in clusters under the plasma membrane and then may travel in the cytoplasmic strands. Upon return to an isotonic solution after osmotic shrinkage in a hypertonic medium, protoplasts usually burst before they are able to recover their initial isotonic surface area. Video microscopy of living protoplasts as well as high-resolution electron microscopy of their membranes suggest that the vesicles are made by internalization of plasma membrane patches, without preferential segregation of lipids or proteins. The vesicles thus formed during the osmotic shrinkage are not readily recycled back into the plasma membrane during subsequent expansion (Steponkus, 1991). Similar observations have been made in intact epidermal cells subjected to plasmolysis/deplasmoIysis excursions. The internalized vesicles can be loaded with fluid phase tracers. Several discrete pools of internalized vesicles have been successively labelled by using consecutive plasmolysis/deplasmolysis cycles (Oparka et al., 1990, 1993; Diekmann et a/. , 1993). Surprisingly, vesicles from the same pool do not fuse together nor do they fuse with the central vacuole, and vesicles from different pools do not fuse together either. None of the vesicles internalized as a consequence of plasmolysis were returned to the plasma membrane when the intact cells were subjected to deplasmolysis. Vesicles of this origin were observed over periods of several hours co-existing with the central vacuole in cells as well as in protoplasts which had recovered from osmotic shock. These observations suggest that the limited expansion of the plasma membrane in deplasmolyzing cells is not mediated by the osmotically induced vesicles. Vesicles which originate from the plasma membrane under hyperosmotic conditions are considerably larger (0.1-1 .0 pm) than coated vesicles and likely lack the molecular machinery operating along the endocytic pathway (Oparka et a / ., 1991). The high ratio of volume to surface area that these large vesicles offer

VACUOLE BIOGENESIS

27

may be particularly useful for diminishing greatly the cell volume or for sampling the extracytoplasmic environment with a minimal loss of plasma membrane. Their function still remains obscure. The vacuole system of the stomatal complex is highly dynamic during cell differentiation as well as in differentiated guard cells during stomatal movement (Palevitz et al., 1981). Using this system, internalization of osmotically induced vesicles could not be demonstrated when guard cells were challenged by osmotic shocks mimicking those that occur during stomatal movement (Diekman etal. 1993). More work is needed to appreciate the extent to which this inward endocytic-like pathway occurs in plant cells under more normal physiological conditions. Moreover, Lucifer Yellow C H and other anionic fluorescent probes which are used as markers of fluid phase endocytosis in animal cells are not suitable for similar studies in plant cells because they are likely to be transported through the plasma membrane and the tonoplast by probenecid-sensitive transporters whose activities cannot be totally blocked (Cole et al., 1991; Oparka et al., 1993). While vesicle-mediated internalizations of plasma membrane are clearly documented in plant cells, their routes need to be precisely mapped by using reliable tracers. Endocytic vesicles do belong to the vacuolar apparatus but their direct contribution in the construction of the central vacuole remains a subject of debate.

V.

ONTOGENY OF VACUOLES SPECIALIZED I N PROTEIN STORAGE

Recent studies on the assembly and transport of seed storage proteins in legumes and cereals have shown that these proteins can be sorted at diverse exit sites along the vacuolar pathway. As a consequence, proteins are stored in a variety of compartments which can be specific to the plant species, tissue, stage of cell differentiation or protein category (Fig. 14). The major storage proteins in legume seeds are the globulins and lectins (Shewry et al., 1995). They are co-translationally inserted in the E R and transported via the Golgi apparatus to the vacuoles, where they are finally deposited (Chrispeels, 1991; Dombrowski ef a / ., 1993; Schroeder et al., 1993; Herman, 1994). Cereal grains differ from legume seeds by accumulating prolamins, another category of storage proteins (Shotwell and Larkins, 1988; Lending and Larkins, 1989; Krishnan et al., 1991; Shewry et ul., 1995). Cereal prolamins, like legume globulins, are co-translationally loaded into the endoplasmic reticulum. However, in many cereals, including rice, maize and sorghum, the prolamins are assembled in protein aggregates which are retained in the ER (Khoo and Wolf, 1970; Larkins and Hurkman, 1978; Lending ef al., 1988;

28

Fig. 14. Protein storage vacuoles in parenchyma cells from the developing cotyledons of ii bean (Phn.wolits vulgaris L. ). Electron microscopic cryoscctions show that storage protcins are accumulated in tubular, ringlike as well as globular vacuoles. 'The bar represents 0.5 p m .

VACUOLE HIOGENESIS

29

Geli et al., 1994). These protein deposits, bound by the rough EK membrane, are classically referred to as “protein bodies”. Rice grains accumulate both prolamins and globulins. The latter, which are the major storage proteins, are transported via the Golgi apparatus to the vacuolar compartment, which is clearly distinct from the ER-derived, prolamin-containing protein bodies (Krishnan et al., 1986). By contrast to maize, rice and sorghum. the prolamins of other cereals, including wheat, barley and oat, are accumulated in vacuoles together with the globulins (Parker and Hawes, 1982; Bechtel and Barnett, 1986; Kim et al., 1988; Shotwell and Larkins. 1988; Lending et d., 1989; Levanony et al., 1992). Endosperm cells from barley grains routinely use various compartments of the secretory pathway to store prolamins (called hordeins in barley) as part of a programmed developmental process. The ontogeny of the protein stores varies with the age of the endosperm cell, which is correlated with the position of the cell in the tissue. In the oldest cells, which are located deep within the endosperm, most of the prolamins are stored in the E R as protein bodies, whereas in younger cells close to the superficial aleurone layer, the prolamins are all accumulated in vacuoles. Prolamins are deposited in both locations as well as in intermediate compartments along the vacuolar pathway in cells between these two developmental extremes (Cameron-Mills and von Wettstein, 1980; Rechinger m l . . 1993). Studies also indicate that the storage proteins that are usually stored in a specific compartment are capable of accumulating in another compartment of the pathway. For instance. the pea globulin (called vicilin), regularly stored in the vacuole, is accumulated in E R protein bodies when the KIIEL retention signal is added at its carboxy amino acid terminus (Wandelt el al., 1992). Similarly, storage proteins normally accumulated in the vacuole of the endosperm cells of‘ barley are deposited in E R protein bodies of the barley line Nevsky, which does not synthesize y-hordeins, a specific isoforrn of prolaniins (Rexach et ril., 1992). Conversely, a maize prolamin isoform (B-zein) which is naturally retained in E R protein bodies has been found addressed to different compartments, including the vacuole, along the secretory pathway in different organs of transgenic tobacco plants (Hoffman et al., 1987; Bagga et al., 1995). The different localizations of prolamins in cereals may result from different sorting mechanisms operating along the vacuolar pathway. For instance, the ER chaperone BiP has been immunolocalized at the luminal surface of the protein bodies where nascent storage proteins are added (Zhang and Boston, 1992; Okita and Rogers, 1996). Interestingly, BiP was occasionally detected with prolamins in the vacuole of wheat endosperm (Levanony et al.. 1992). The architectural components of the cell may serve as topological sorting devices. It has been elegantly shown by in sitir hyhridization in endosperm cells of rice that mRNAs for different species of storage proteins are segregated on morphologically distinct EK membranes (Li et al., 1993).

30

F. MARTY

Globulin mRNAs are preferentially located on cisternal E R whereas prolamin mRNAs are selectively present on protein body-forming domains at the junction where several cisternae converge. The segregation of transcripts on the E R may be critical for efficient sorting of different proteins in subdomains from the same compartment. The concentration of newly synthesized prolamins could specify where the protein bodies can form and, as a consequence, globulins would have free access to the Golgi apparatus and transGolgi pathways. The cytoskeleton plays a role in storage protein deposition. In maize endosperm cells actively forming protein bodies and starch, the protein bodies are juxtaposed with a reticulate array of microtubules and are enmeshed in a protein complex made of the elongation factor (EF)-la and actin (Clore et al., 1996). Physical aggregation of proteins, retention signals, and targeting sequences give positive molecular information for sorting decisions. But proteins may well escape these commands and accumulate by default in intermediate compartments along the secretory pathway. Immunocytochemical studies suggest that the transport of prolamins to the vacuole is mediated by the Golgi apparatus in wheat endosperm cells (Kim et al., 1988; Levanony et al., 1992; Rubin et al., 1992; Galili et al., 1993). However, wheat prolamins might be also translocated in the vacuoles by autophagy (Levanony et al., 1992). Prolamins would be first assembled into E R protein bodies before being surrounded by small vesicles of yet unknown origin, abundantly present in the cells at this stage. The vesicles (provacuoles?) apparently fuse with each other to form a vacuole containing the protein body inclusion. Membrane remnants are also often present with the protein inclusion inside the vacuole, suggesting an autophagic-like process. It is not known whether such autophagy shares common mechanisms with the cellular autophagy involved in the formation of the vacuoles in the same tissue. The origin of the protein storage vacuole in maturing legume seeds is not yet settled (Robinson and Hinz, 1996). Early descriptions suggested that the pre-existing vegetative vacuole in the young parenchyma cells of the cotyledons gradually filled up with storage proteins and then subdivided to generate numerous small storage vacuoles. In contrast to these earlier suggestions, recent morphological and cytochemical studies indicate that the original vegetative vacuoles are replaced by the protein storage vacuole in a process which strikingly resembles autophagy. Indeed, the vegetative vacuole of immature parenchyma cells becomes surrounded by a smooth cisternal, tubular membrane system that already contains deposits of storage proteins (Craig, 1986; Hoh et ul., 1995). Later on, the encircled vegetative vacuole is digested, and only membrane remnants can be seen in the developing storage vacuole. However, the origin of the novel protein storage vacuole is not clearly

VACUOLE RIOGENESIS

31

understood (Hoh et al., 1995). The two Hi--pumps, the V-ATPase and H+-PPase, as well as the seed-specific aquaporin a-TIP, but not y-TIP which is specific to the vegetative vacuole, were immunolocalized on the membrane of the protein storage vacuole. The storage proteins, vicilin and legumin, were immunodetected in the lumen as expected. The transport of vicilin and legumin to the vacuole in parenchyma cells of developing pea cotyledons is mediated by sniooth-surfaced vesicles but not by clathrin-coated vesicles as previously assumed (Hohl et al., 1996). Because immunocytochemical studies have indicated that the Golgi apparatus mediates the transport of the soluble storage proteins and of membrane proteins to the repository Compartment, the storage compartment is a post-Golgi compartment, but clearly distinct from the vegetative vacuole. In the cotyledons from germinating seedlings, when the protein storage is replaced by a vegetative vacuole, another type of developmentally regulated sequestration and disposal of organelles has been described. It involves the local invagination of the tonoplast and the subsequent engulfment of cytoplasmic fragments in the protein storage vacuole. Concurrently with the hydrolysis of the storage proteins by newly synthesized endoproteases and the disappearance of specific membrane proteins, the engulfed material and its surrounding membrane are degraded (Van der Wilden et al., 1980; Herman et al., 1981; Melroy and Herman, 1991; Herman 1994; Inoue et al., 1995). This mechanism occasionally reported in active vegetative cells (Marty, 1978) is enhanced in cells undergoing senescence (Wittenbach et al., 1982). In cells from the aleurone layer of non-germinated grains of barley, the protein storage vacuoles containing the aspartic proteinase are morphologically and biochemically distinct from the vacuoles that contain the cysteine protease aleurain. Upon activation of the aleurone cells by gibberellins, the vacuoles of the two classes merge in large vacuoles. making possible the digestion of the storage protein by aleurain (Holwerda et al., 1990). The protein storage vacuoles and the “aleurain vacuoles” could thus be considered as intermediate compartments in the ontogeny of the large terminal vacuoles (Okita and Rogers, 1996). Similarly, it has been recently shown by immunocytochemistry that pea and barley root tip cells from young seedlings contain two separate vacuole-related organelles defined by the tonoplast intrinsic protein a-TIP and TIP-Ma27, respectively (Paris et al., 1996). In barley root tip cells barley lectin is present in a-TIP-positive vacuoles, but absent from TIP-Ma27-positive vacuoles, while aleurain is distinctively contained in TIP-Ma27-positive vacuoles, but absent from a-TIP-positive vacuoles. These separate compartments are therefore functionally distinct: a-TIP defines a compartment where storage proteins are protected against degradative enzymes whereas TIP-Ma27 defines a separate acidic, lytic compartment. As cells differentiate and develop large vacuoles,

32

F. MARTY

these two compartments were described as merging and the marker membrane antigens a-TIP and TIP-Ma27 are localized in the same membrane (tonoplast) at least in certain regions of the vacuolar compartment (Paris et al., 1996). Further understanding of the biogenesis and dynamics of the two separate compartments is awaited. The ontogeny of the compartments specialized in protein storage is diverse, and not all stores are homologous, although all belong to the vacuolar apparatus of plant cells.

VI. CONCLUSIONS Recent structural insights into the vesicular traffic between compartments of the endomembrane system and the identification of specific markers make it possible to begin to draw a number of preliminary conclusions on the intracellular pathways leading to the vacuole. There is now evidence that the central vacuoles of plant cells represent the final destination for a significant fraction of all intracellular traffic, and in the vast majority of cells from ground tissues of vegetative organs, vacuoles are the terminal endomembrane compartments. They serve as a true milieu interieur for these cells. As supported by a number of studies during the past several years, vacuoles receive input from the biosynthetic and endocytic pathways and share a number of characteristics with the precursors involved in their formation. Many aspects of the pathways have been investigated in different systems by a variety of methodologies. The mapping of these confluent pathways by morphological and cytochemical methods provides the structural framework of our present concepts on vacuole biogenesis. For example, vacuoles are built in part by newly synthesized molecules which are transported through multiple compartments prior to reaching their target: transit through the E R , intermediate compartment, Golgi apparatus, TGN, and delivery to pre/provacuoles before retention in the vacuole. Autophagy appears as a key step in the formation of new vacuoles in unvacuolated differentiating cells as well as in already vacuolated starving cells. In addition, extracellular macromolecules - previously exocytosed by the same cell, or alien molecules - may gain entry into the cell by different types of endocytosis. The various endocytic vesicles may be integral parts of the vacuolar system but their direct contribution to the biogenesis of the central vacuole remains unsettled. Variations on this theme might explain the complexity of the vacuolar system in a variety of plant cells. Indeed, in a number a cell types, such as secretory cells or cells specialized in protein storage, vacuoles are diverse in origin, composition and function. Any one of the compartments (whether they are E R , intermediary, Golgi, post-Golgi, pre/provacuolar compartments or endocytic vesicles) along the biogenetic vacuolar pathways may create via

VACUOLE B I OCiF. N E S I S

33

“hypertrophy”, “over-loading” or “over-expression” a distinct inclusion, provisionally identified as “vacuole-like”. This reflects our present inability to define more precisely vacuoles using molecular criteria. In recent years, a number of proteins that reside in membranes of the secretory system have been identified. The cloning of genes encoding proteins from the vacuolar pathways and their putatively deduced structure have allowed a molecular description of the mechanisms involved in their assembly and transport to the vacuole. Much additional work is nceded to adjust the molecular findings to the structural framework. This will clarify the functions and formation of the organelles involved in the vacuole biogcnesis. High-resolution immunocytochemistry at the electron microscope as well as in s i t u hybridization will now allow a molecular mapping of the origins of vacuoles in plant cclls.

ACKNOWLEDGEMENTS The author is grateful to the past and present members of his laboratory and colleagues elsewhere for their contributions to the work discussed in this review. Work described from the author’s laboratory was supported by grants from the Ministere de I’Enseignement Superieur et de la Recherche. Mission Scientifique et Technique (DSPTS. EA 469), the Centre National de la Recherche Scientifique (CNRS, Departemcnt des Sciences dc la Vie), the Conseil Regional de Bourgognc and the Delegation Regionale la Recherche et a la Technologie.

REFERENCES Ali, M. S. and Akazawa, T. (1986). Association o f H+-translocating ATPase in Golgi membrane system from suspension-cultured cells of sycamore. Plunf Physiology 81, 222-227. Aubert, S . , Gout, E., Bligny, R . , Marty-Mazars, D.. Barrieu, F., Alabouvette, J . , Marty. F. and Douce, R . (lYY6). Ultrastructural and biochemical characterization of autophagy in higher plant ctAls subjected to carbon deprivation control by the supply of mitochondria with rcspiratory substrates. Journal o f Cell Biology 133. 1251-1263. Baba, M., ‘Takeshige, K., Baba, N . and Ohsumi. Y. (1994). Ultrastructural analysis of the autophagic process in yeast: detection of autophagosomes and their characterization. Journcrl of Cdl Biology 124. 003-913. Bagga, S . . Adams, H.. Kenip. J . D. and Scngupta-Ciopalan. C. (1995). Accumulation of 15-kilodaltons zein in novel protein bodieh in transgenic tobacco. Plcint P / I ~ S ~ O / 107, O ~ J13-23. > Balusck, K . , Depta. H. and Rokinson, D. G . (1988). Two polypeptides (30 and 38 k D a ) in plant coated vesicles with clathrin light chain properties. Protoplusmu 146. 147-176. Bassham. D. C . , Gal, S., da Silva Conceicao, A . and Kaikhel, N . (1995). An Arcrhiclop.si.s syntaxin homologue isolated by functional complementation o f a

34

F. MARTY

yeast pep 12 mutant. Proceedings of the National Academy of Sciences of the USA 92, 7262-7266. Bechtel, D. B. and Barnett, B. D. (1986). Development and evaluation of freezextch, freeze-fracture technique applied to developing wheat endosperm. American Journal of Botany 3, 863-873. Bednarek, S. Y. and Raikhel, N. V. (1992). Intracellular trafficking of secretory proteins. Plant Molecular Biology 20, 133-150. Beevers, L. (1996). Clathrin-coated vesicles in plants. International Review of Cytology 167, 1-35. Bennett, A. B . , O’Neill, S. D. and Spanswick, R. M. (1984). H+-ATPase activity from storage tissue of Beta vulgaris. Plant Physiology 74, 538-544. Boller, T. and Wiemken, A. (1986). Dynamics of vacuolar compartmentation. Annual Review of Plant Physiology 37, 137-164. Boss, W. F., MorrC, D. J. and Mollenhauer, H. H. (1984). Monensin-induced swelling of Golgi apparatus cisternae mediated by a proton gradient. European Journal of Cell Biology 34, 1-8. Burgess, J . and Lawrence, W. (1985). Studies of the recovery of tobacco mesophyll protoplasts from the evacuolation treatment. Protoplasma 126, 140-146. Cameron-Mills, V. and von Wettstein, D. (1980). Protein body formation in the developing barley endosperm. Carlsherg Research Communication 45, 577594. Chen, M. H., Liu, L. F., Chen. Y. R., Wu, H. K. and Yu, S. M. (1994). Expression of a-amylases, carbohydrate metabolism, and autophagy in cultured rice cells is coordinately regulated by sugar nutrient. Plant Journal 6 , 625-636. Chrispeels, M. J. (1991). Sorting of proteins in the secretory system. Annual Review of Plant Physiology and Plant Molecular Biology 42, 21-53. Chrispeels, M. J. and Raikhel, N. (1992). Short peptide domains target proteins to vacuoles. Cell 68, 613-616. Churchill, K . A , , Holaway, B. and Sze, H. (1983). Separation of two types of electrogenic H+-pumping ATPases from oat roots. Plant Physiology 73, 921-928. Clore, A. M., Dannenhoffer, J . M. and Larkins, B. A. (1996). EF-la is associated with a cytoskeletal network surrounding protein bodies in maize endosperm cells. Plant Cell 8, 2003-2014. Cole, L., Coleman, J. 0. D . , Kearns, A , , Morgan, G. and Hawes, C. (1991). The organic anion transport inhibitor, probenecid, inhibits the transport of lucifer yellow at the plasma membrane and the tonoplast in suspension-cultured plant cells. Journal of Cell Science 99, 545-555. Coleman, J. 0. D., Evans, D., Hawes, C., Horsley, D. and Cole, L. (1987). Structure and molecular organization of higher plant coated vesicles. Journal of Cell Science 88, 3545. Coleman, J . 0. D., Evans, D. E., Horsley, D. and Hawes, C . R. (1991). The molecular structure of plant clathrin and coated vesicles. In “Endocytosis, Exocytosis and Vesicle Traffic in Plants” (C. R. Hawes, J. 0. D. Coleman and D. E. Evans, eds), pp. 41-63. Society for Experimental Biology, Seminar Series 45. Craig, S. (1986). Fixation of a vacuole-associated network of channels in proteinstoring pea cotyledon cells. Protoplasma 135, 67-70. Cram, W. J . (1980). Pinocytosis in plants. New Phytologist 84, 1-17. Demmer, A , , Holstein, S. E. H., Hinz, G., Schauermann, G. and Robinson, D. G. (1993). Improved coated vesicle isolation allows better characterization of clathrin polypeptides. Journal of Experimental Botany 44,23-33.

VACUOLE RIOGENESIS

35

Depta, H . and Robinson, D. G . (1986). The isolation and enrichment of coated vesicles from suspension-cultured carrot cells. Proroplasma 130, 162-170. Diekman, W.. Hedrich. R . , Raschke, K . and Robinson, D. G. (1993). Osmocytosis and vacuolar fragmentation in guard cell protoplasts: their relevance to osmotically-induced volume changes in guard cells. Journal of Experirniental Botany 44, I 569-1 577. Dombrowski. J. E . , Schroeder. M. R . , Bcdnarek. S . Y . and Raikhel, N . V. (1993). Determination of the functional elements within the vacuolar targeting signal of barley lectin. Plant Cell 5, 587-596. Douce, R.. Aubert. S., Bligny, R.. Dornc, A.-J., Gout, E . and Marty. F. (1995). Formation of autophagic vacuoles and accumulation of deacylation products of membrane polar lipids during the coursc of sucrose starvation in higher plant cclls. In “Plant Lipid Metabolism” (J. C. Kader and P. Mazliak, eds), pp. 188-196. Kluwer. Dordrecht. Dozolme, P., Marty-Mazars, D . , Clemencet. M.-C. and Marty, F. (1995). Monoclonal antibody TeM 106 reacts with a tonoplast intrinsic protein of 106 kDa from Brassica oleracea L. Journal of Cell Science 108, 4, 1509-1517. Driouich, A . , Zhang. G . F. and Staehelin. 1.. A . (1993). Effect of brefeldin A on the structure of the Golgi apparatus and on the synthesis and secretion of proteins and polysaccharides in sycamore maple (Acer pseudoplatanus) suspension-cultured cells. Plant Physiology 101, 1363-1373. Emons, A. M. C . and Traas, J . A . (1986). Coated pits and coated vesicles on the plasma membrane of plant cells. European Journal of Cell Biology 41, 5744. Fitchette-Laine, A.-C., Gomord, V., Chekkali, A . and Faye, L. (1994). Distribution of xylosylation and fucosylation in the plant Golgi apparatus. Plant Journal 5 , 673482. Fowke, L. C . , Tanchak, M. A . and Galway, M. E. (1991). Ultrastructural cytology of the endocytotic pathway in plants. I n “Endocytosis, Exocytosis and Vesicle Traffic in Plants” (C. R. Hawes, J . 0. D. Coleman and D. E . Evans, eds), pp. 15-40, Society for Experimental Biology, Seminar Series 45. Galili, G., Altschuler, Y. and Levanony, H . (1993). Assembly and transport of seed storage proteins. Trends in Cell Biology 3. 437442. Geli, M. I . . Torrent, M. and Ludevid, D. (1994). Two structural domains mediate two sequential events in y-zein targeting: protein endoplasmic reticulum retention and protein body formation. Plant Cell 6, 1911-1922. Goldstein. J . L . , Brown, M. S., Anderson, R. Ci. W., Russel, D. W. and Schneider, W. J . (1985). Receptor-mediated endocytosis: concepts emerging from the LDL receptor system. Annual Review of Cell Biology I , 1-39, Gomez, L . and Chrispeels, M. J . (1993). Tcmoplast and soluble vacuolar proteins are targeted by different mechanisms. Plant Cell 54, 1 1 13-1124. Greenwood, J . S . and Chrispeels. M. J . (1985). Correct targeting of the bean storage protein phaseolin in the seeds of transformed tobacco. Plant Physio1og.y 79, 65-71, Griesbach. R. G . and Sink. K. C . (1983). Evacuolation of mesophyll protoplasts. Plant Science Letters 30, 297-301. Griffing, L. R . (1991). Comparisons of Golgi structure and dynamics in plant and animal cells. Journal of Electron Microscopy Techniques 17, 179-199. Griffing, L. R. and Fowke, L. C. (1985). Cytochemical localization of peroxidase in soybean suspension culture cells and protoplasts: intracellular vacuole differentiation and presence of peroxydase in coated vesicles and multivesicular bodies. Protoplasma 128, 22-30.

36

F. MARTY

Griffiths, G. and Simons, K. (1986). The trans Golgi network: sorting at the exit site of the Golgi complex. Science 234, 438-443. Griffiths, G., Holflack, B., Simons, K., Mellman, I. and Kornfeld, S. (1988). The mannose-6-phosphate receptor and the biogenesis of lysosomes. Cell 52, 329-341. Gruenberg, J. and Howell, K. E. (1989). Membrane traffic in endocytosis: insights from cell-free assays. Annual Review of Cell Biology 5, 453-481. Harley, S. M . and Beevers, L. (1989). Coated vesicles are involved in the transport of storage proteins during seed development in Pisum safivum L. Plunt Physiology 91, 674-678. Harris, N. and Watson, M . D. (1991). Vesicle transport to the vacuole and the central role of the Golgi apparatus. In “Endocytosis, Exocytosis and Vesicle Traffic in Plants” (C. R. Hawes, J . 0. D. Coleman and D. E. Evans, eds), pp. 143-164. Society for Experimental Biology, Seminar Series 45. Hellgren, L., Morrk, D. J., Sellden, G . and Sandelius, A. S. (1993). Isolation of a putative vesicular intermediate in cell-free transfer of membrane from transitional endoplasmic reticulum to the Golgi apparatus of etiolated seedlings of garden pea. Journal of Experimental Botuny 44, 197-20.5. Henderson, J., Satiat-Jeunemaitre, B., Napier, R . and Hawes, C. (1994). Brefeldin A-induced disassembly of the Golgi apparatus is followed by disruption of the endoplasmic reticulum in plant cells. Journal of Experimental Botany 45, 1347-1351. Herman, E. M. (1994). Multiple origins of intravacuolar protein accumulation of plant cells. Advances in Structural Biology 3, 243-283. Herman, E. M., Baumgartner, B. and Chrispeels, M . J . (1981). Uptake and apparent digestion of cytoplasmic organelles by protein bodies (protein storage vacuoles) in mung bean cotyledons. European Journul of Cell Biology 24, 226235. Herman, E. M . , Li, X . , Su, R. T., Larsen, P., Hsu, H. T. and Sze, H. (1994). Vacuolar-type H+-ATPases are associated with the endoplasmic reticulum and provacuoles of root tip cells. Plant Physiology 106, 1313-1324. Hillmer, S . , Depta, H. and Robinson, D. G. (1986). Confirmation of endocytosis in higher plant protoplasts using lectin-gold conjugates. European Journal of Cell Biology 41, 142-149. Hillmer, S., Freundt, H. and Robinson, D. G. (1988). The partially coated reticulum and its relationship to the Golgi apparatus in higher plant cells. European Journal of Cell Biology 47, 206-212. Hillmer, S., Quader, H., Robert-Nicoud, M . and Robinson, D. G. (1989). Lucifer yellow uptake in cells and protoplasts of Duucus carota visualized by laser scanning microscopy. Journal of Experimental Botuny 40, 4 17423. Hippe, S., During, K. and Kreuzaler, F. (1989). In situ localization of a foreign protein in transgenic plants by immunoelectron microscopy following high pressure freezing, freeze substitution and low temperature embedding. European Journal of Cell Biology 50, 230-234. Hoffman, L. M . , Donaldson, D. D., Bookland, R., Rashka, K. and Herman, E. M . (1987). Synthesis and protein-body deposition of maize 15-kd zein in transgenic tobacco seeds. EMBO Journal 6, 3213-3221. Hoh, B., Hinz, G . , Jeong, B.-K. and Robinson, D. G. (1995). Protein storage vacuoles form de novo during pea cotyledons development. Journal of Cell Science 108, 299-310. Hohl, I . , Robinson, D. G., Chrispeels, M. J. and Hinz, G. (1996). Transport of storage proteins to the vacuole is mediated by vesicles without a clathrin coat. Journal of Cell Science 109, 2539-2550.

VACUOLE BIOGENESIS

37

Holstein, S. E. H . , Drucker, M. and Robinson, D. G . (1904). Identification of a P-type adaptin in plant clathrin-coated vesicles. Journul of Cell Science 107, 945-953. Holwerda. B. C., Calvin, N . J., Baranski, T. J. and Rogers, J. C. (1990). l r z vitro processing of aleurain, a barley vacuolar thiol protease. Plant Cell 2, 1091110h. Holwerda. B. C., Padgett, H . S. and Rogers. J. C. (1992). Proaleurain vacuolar targeting is mediated by short contiguous peptide interactions. Plant Cell 4, 307-3 18. Horazdovsky, B. F., Dewald, D. B. and Emr, S . D. (1995). Protein transport to the yeast vacuole. Current Opinion in Cell Biology 7, 544-551. Horn, M. A . , Heinstein, P. F. and Low, P. S. (1989). Receptor-mediated endocytosis in plant cells. Plant Cell 1, 1003-1009. Horn, M . A . , Heinstein, P. F. and Low, P. S. (1090). Biotin-mediated delivery of exogenous macromolecules into soybean cells. Plant Physiology 93, 14921496. Hiirtensteiner, S., Martinoia, E . and Amrhein, N . (1992). Reappearance of hydrolytic activities and tonoplast proteins in the regenerated vacuoles of evacuolated protoplast. Plunta 187, 113-121. Hiirtensteiner, S.. Martinoia, E . and Amrhein, N. (1994). Factors affecting the reformation of vacuoles in evacuolated protoplasts and the expression of the two vacuolar proton pumps. Plantu 192, 395-403. Hubner, R . , Depta, H. and Robinson, D. G . (1985). Endocytosis in maize root cap cells. Evidence obtained using heavy metal salt solutions. Protoplusmu 129, 2 I 4-222. Hurley, D. and Taiz, L. (1989). Immunocytochemical localization of the vacuolar H+-ATPase in maize root tip cells. Plant Physiology 89, 391-395. Inoue, K . , Motozaki, A , , Takeuchi, Y . , Nishimura, M. and Hara-Nishimura, I . ( 1995). Molecular characterization of proteins in protein-body membrane that disappear most rapidly during transformation of protein bodies into vacuoles. Plant Journal 7, 235-243. Joachim, S. and Robinson, D . G. (1984). Endocytosis of cationic ferritin by bean leaf protoplasts. European Joiirriul of’ CPll Biology 34. 212-216. Johnson, K. D., Hiifte, H . and Chrispeels. M. J. (1990). An intrinsic tonoplast protein of protein storage vacuoles in seeds is structurally related to a bacterial solute transporter (GlpF). Plant Cell 2, 525-532. Khoo, V. and Wolf, M. J. (1970). Origins and development of proteins granules in maize cndosperm. American Journal of Botany 57, 1042-1050. Kim, W. T., Franceschi, V. R.. Krishnan. H . B. and Okita, 7’. W. (1988). Formation of wheat protein bodies: involvement o f the Golgi apparatus in gliadin transport. Plunlu 173, 173-182. Kirsh. T., Paris. N . . Butler, J. M.. Beevers. L. and Rogers, J. C. (1994). Purification and initial characterization of a potential plant vacuolar targeting receptor. Proceedings of the Nutionul Acudemv of Sciences of the USA 91, 3403-3407. Kornfeld, S. and Mellman, I. (1989). The biogenesis of lysosomes. Annual Revicw of’ C‘dI Biology 5, 483-525. Krishnan. H . B., Franceschi, V. R. and Okita, T. W. (1986). Inimunochemical studies on the role of the Golgi complex in protein body formation in rice seeds. Planto 169, 471480. Krishnan, H . B., White, J . A . and Pueppke, S . G. (1091). Immunocytochemical localization of wheat prolamins in the lumen of the rough endoplasmic reticulum. C’unudiun Journal of Botuny 69, 2574-2577.

38

F. MARTY

Larkins, B. A. and Hurkman, W. J . (1978). Synthesis and deposition of zein protein bodies of maize endosperm. Plant Physiology 62, 256-263. Leigh, R. and Branton, D. (1976). Isolation of vacuoles from root storage tissues of Beta vulgaris L. Plant Physiology 58, 656-662. Lending, C. R. and Larkins, B. A. (1989). Changes in the zein composition of protein bodies during maize endosperm development. Plant Cell 1, 1011-1023. Lending, C. R . , Kriz, A. L., Larkins, B. A. and Bracker, C. E. (1988). Structure of maize protein bodies and immunocytochemical localization of zeins. Protoplasma 143, 5142. Lending, C. R., Chestun, R. S . , Shaw, K . L. and Larkins, B. A. (1989). Immunolocalization of avenin and globulin storage proteins in developing endosperm of Avena saliva L. Planta 178, 315-324. Levanony, H., Rubin, R., Altschuler, Y . and Galili, G. (1992). Evidence for a novel route of wheat storage proteins to vacuoles. Journal of CeN Biology 119, 1117-1 128. Li, X . , Franceschi, V. R. and Okita, T. W. (1993). Segregation of storage protein mRNAs on the rough endoplasmic reticulum membranes of rice endosperm cells. Cell 72, 869-879. Lorz, H., Harms, C. T. and Potrykus, I. (1976). Isolation of “vacuoplasts” from protoplasts of higher plants-. Biocheme und Physiologie der P$anzen 169, 617-620. Low, P. S. and Chandra, S. (1994). Endocytosis in plants. Annual Review of Plant Physiology and Plant Molecular Biology 45, 609-631. Low, P. S . , Legendre, L., Heinstein, P. F. and Horn, M. A. (1993). Comparison of elicitor and vitamin receptor-mediated endocytosis in cultured soybean cells. Journal of Experimental Botuny 44, 269-274. Ludevid, D., Hofte, H., Himelblau, E. and Chrispeels. M. J . (1992). The expression pattern of the tonoplast intrinsic protein y-TIP in Arabidopsis thalianu is correlated with cell enlargement. Plant Physiology 100, 1633-1639. Marty, F. (1973). Dissemblance des faces golgiennes et activite des dictyosomes dans les cellules en cours de vacuolisation de la racine d’Euphorbiu churacias L. Comptes Rendus de I’Academie des Sciences (Paris) 277, 1749-1752. Marty, F. (1978). Cytochemical studies on GERL, provacuoles, and vacuoles in root meristematic cells of Euphorbia. Proceedings of the National Academy of Sciences of the U S A 75, 852-856. Marty, F. (1980). High voltage electron microscopy of membrane interactions in wheat. Journal of Histochemistry and Cytochemistry 28, 1129-1 132. Marty , F. (1983). Microscopie Clectronique a haute tension de I’appareil provacuolaire dans les cellules mkristematiques des racines de radis (Raphanus sativus L.). Annales des Sciences Naturelles, Botanique (Paris) 5, 245-260. Marty, F. and Branton, D. (1980). Analytical characterization of beetroot vacuole membrane. Journal of Cell Biology 87, 72-83. Marty, F., Branton, D. and Leigh, R. (1980). Plant vacuoles. In “The Biochemistry of Plants” (P. K. Stumpf and E. E. Conn, eds) Vol. 1, pp. 625-658. Academic Press, New York. Marty-Mazars, D . , Clemencet, M-C., Dozolme, P. and Marty, F. (1995). Antibodies to the tonoplast from the storage parenchyma cells of beetroot recognize a major intrinsic protein related to TIPS. European Journal of Cell Biology 66, 106-118. Matile, P. (1975). “Cell Biology Monographs”, Vol 1, p. 183. Springer-Verlag, Wien. Matile, P. (1978). Biochemistry and function of vacuoles. Annual Review of Plunt

VACUOLE RIOGENESIS

39

Physiology 29, 193-2 13. Matile, P. (1987). The sap of plant cells. New Phytologist 105, 1-26. Matsuoka, K . , Bassham, D . C., Raikhel, N . and Nakamura. K. (1995). Different sensitivity to wortmannin of two vacuolar sorting signals indicates the presence of distinct sorting machineries in tobacco cells. Journal of Cell Biology 6, 1307-1 3 18. Melroy, D. L. and Herman, E. M. (1991). TIP, an integral membrane protein of the protein-storage vacuoles o f the soybean cotyledon undergoes developmentally regulated membrane accumulation and removal. Planta 184, 113-122. Merscy, €3. G . , Griffing, L. R . , Rennie, P. J . and Fowke, L. C. (1985). The isolation of coated vesicles from protoplasts of soybean. Planta 163, 317-327. Moriyasu. Y. and Ohsumi, Y . (1996). Autophagy in tobacco suspension-cultured cells in response to sucrose starvation. Hunt Physiology 111, 1233-1241. Novikoff, A . B. (1976). The endoplasmic reticulum: a cytochemist’s view (a review). Proceedings of the National Academy of Sciences of the U S A 73, 2781-2787. Okita. T. W. and Rogers, C. J . (1996). Chnpartmentation of proteins in the endomembrane system of plant cells. Annual Review of Plant Physiology and Plant Molecular Biology 47, 327-350. Oparka, K. J., Prior, D. A . M . and Harris, N . (1990). Osmotic induction of fluid-phase endocytosis in onion epidermal cells. Planta 180, 555-561. Oparka, K. J . , Cole, L., Wright, K. M.. Hawes, C. R.. Evans, D. E. and Coleman, J. 0. D . (1991). Fluid-phase endocytosis and the subcellular distribution of fluorescent probes in plant cells. I n “Endocytosis, Exocytosis and Vesicle Traffic in Plants” (C. R . Hawes. J. 0 . D. Coleman and D. E. Evans, eds), pp. 81-102. Society for Experimental Biology. Seminar Series 45. Oparka, K. J . , Wright, K. M . , Murant, E . A . and Allan, E. J . (1993). Fluid-phase endocytosis: do plants need it’? Journal of Experimental Botany 44, 247-255. Owen, T. P.. Jr, Platt-Aloia, K. A . and Thomson, W. W. (1991). Ultrastructural localization of lucifer yellow and endocytosis in plant cells. Protoplasma 160, 115-120. Palevitz. B. A . , O’Kane. D. J., Kobres. R . E . and Raikhel, N. (1981). The vacuole system in stomata1 cells of Alliurn: vacuole movements and changes in morphology in differentiating cells as revealed by epifluorescence, video and electron microscopy. Protoplasma 109, 23-55. Paris. N., Stanley, C. M . , Jones, R. J. and Rogers, J . C. (1996). Plant cells contain two functionally distinct vacuolar compartments. Cell 85, 563-572. Parker, M. L. and Hawes, C. R. (1982). ‘The Golgi apparatus in devcloping endosperm of wheat (Triticum aestivum L . ) . Planta 154, 277-283. Patel, D. D., Barlow, P. W. and Lee, R . B. (1990). Development of vacuolar volume in the root tips of pea. Annuls of Botany 65, 159-169. Pesacreta. T. C. and Lucas, W. J. (1985). Presence of a partially-coated reticulum in angiosperms. Protoplasma 125, 173-184. Phillips, G . D., Preshaw, C. and Steer. M . W. (1988). Dictyosome vesicle production and plasma membrane turnover in auxin-stimulated outer epidermal cells of coleoptile segments from Avena strtivu (L.). Protoplusma 145, 59-65, Piper, R . C., Cooper, A . A . . Yang, H . and Stevens, T. H. (1995). VPS27 controls vacuolar and endocytic traffic through a prevacuolar compartment in Succharomyces cerevisiae. Journal of Cell Biology 131, 603417. Pueyo, .I. J.. Chrispeels, M. J. and Herman, E. M. (1995). Degradation of transport-competent destabilized phaseolin with a signal for retention in the endocytic reticulum occurs in the vacuole. Planta 196, 586-596. Raven, J . A . (1987). The role of vacuoles. New Phytologist 196, 357422.

40

F. MARTY

Raymond, C. K., Howald-Stevenson, I., Vater, C . A . and Stevens, T. H. (1992). Morphological classification of a yeast vacuolar protein sorting mutant: evidence for a prevacuolar compartment in class E vps mutants. Molecular Biology of the Cell 3 , 1389-1402. Rea, P. A. and Poole, R. J. (1993). Vacuolar H+-translocating pyrophosphatase. Annual Review of Plant Physiology and Plant Molecular Biology 44, 157180. Rechinger, K . B., Simpson, D. J . , Svendsen, I. and Cameron-Mills, V. (1993). A role for hordein in the transport and targeting of prolamin polypeptides to the vacuole of developing barley endosperm. Plant Journal 4, 841-853. Record, R. D. and Griffing, L. R . (1988). Convergence of the endocytic and lysosomal pathways in soybean protoplasts. Planta 176, 425-432. Rexach, M., d’Enfert, C., Wuestehube, L. and Schekman, R. (1992). Genes and proteins required for vesicular transport from the endoplasmic reticulum. Antonie Leeuwenhoek International Journal of General Microbiology 61, 87-92. Robinson, D. G. and Hinz, G . (1996). Multiple mechanisms of protein body formation in pea cotyledons. Plant Physiology and Biochemistry 34, 155-163. Robinson, D. G., Balusek, K., Depta, H., Hoh, B . and Holstein, S. E. H. (1991). Isolation and characterisation of plant coated vesicles. In “Endocytosis, Exocytosis and Vesicle Traffic in Plants’’ (C. R. Hawes, J. 0. D. Coleman and D. E. Evans, eds), pp. 65-79. Society for Experimental Biology, Seminar Series 45. Robinson, D. G., Hedrich, R., Herkt, B., Diekmann, W. and Robert-Nicoud, M . (1992). Endocytosis in plants: problems and perspectives. In “Endocytosis” (P. J. Courtoy, ed.), NATO AS1 Series, pp. 459-466. Springer-Verlag, Berlin. Robinson, D. G., Hashke, H.-P., Hinz, G . , Hoh, B., Maeshima, M . and Marty, F. (19964. Immunological detection of tonoplast polypeptides in the plasma membrane of pea cotyledons. Planta 198, 95-103. Robinson, M. S . , Watts, C. and Zerial, M. (1996b). Membrane dynamics in endocytosis. Cell 84, 13-21, Rodman, J. S . , Mercer, R. W. and Stahl, P. D. (1990). Endocytosis and transcytosis. Current Opinion in Cell Biology 2, 664-672. Rubin, R., Levanony, H. and Galili, G. (1992). Evidence for the presence of two different types of protein bodies in wheat endosperm. Plant Physiology 99, 7 18-724. Samuels, A. L. and Bisalputra, T. (1990). Endocytosis in elongating root cells of Lobelia erinus. Journal of Cell Science 97, 157-165. Satiat-Jeunemaitre, B. and Hawes, C. (1992). Reversible dissociation of the plant Golgi apparatus by brefeldin A. Biology of the Cell 74, 325-328. Satiat-Jeunemaitre, B. and Hawes, C. (1993a). Insights into the secretory pathway and vesicular transport in plant cells. Biology of the Cell 79, 7-15. Satiat-Jeunemaitre, B. and Hawes, C. (1993b). Redistribution of a Golgi glycoprotein in plant cells treated with brefeldin A. Journal of Cell Science 103, 11531166. Satiat-Jeunemaitre, B. and Hawes, C. (1994). G A m (A general agreement on traffic and transport) and brefeldin A in plant cells. Plant Cell 6 , 463-467. Satiat-Jeunemaitre, B., Fitchette-LainC, A.-C., Alabouvette, J., Marty-Mazars, D., Hawes, C., Faye, L. and Marty, F. (1994). Differential effects of monensin on the plant secretory pathway. Journal of Experimental Botany 45, 685-698. Satiat-Jeunemaitre, B., Steele, C. and Hawes, C. R. (1996). Maintenance of the exocytotic and endocytic apparatus involved in protein targeting in plant cells.

VACUOLE BIOGENESIS

41

Plcmt Physiology and Biochemistry 34, 183-195. Sato, M. H . , Kasahara, M.. Ishii, N., Homareda. H., Matsui, H. and Yoshida. M. (1994). Purified vacuolar inorganic pyrophosphatase consisting of a 75 kDa polypeptide can pump H + into reconstituted proteoliposomes. Journal of Riologicul Chemistry 269, 6725-673-8. Saxton, M. I . and Breidenbach, R. W. (1988). Receptor-mediated endocytosis in plants is energetically possible. Plant I’h.vsiology 86, 993-995. Schindler, T., Bergfeld. R., Holh, M. and Schopfer, P. (1994). Inhibition of Golgi-apparatus function by brefcldin A in maize coleoptiles and its consequences on auxin-mediated growth. cell-wall extensibility and secretion of cell-wall proteins. Planta 92, 404413. Schroeder. M. R., Dombrowski, J . E . , Bednarek. S . Y., Borkhsenious, 0. N. and Raikhel, N. V. (1993). Molccular basis of post-translational modifications and targeting of barley lectin to the vacuoles in barley and in transgenic tobacco plants. Journal of Experimental Bottrny 44. 315-319. Shannon. T. M. and Steer, M. W. (1984). ‘The root cap as a test system for the cvaluation of Golgi inhibitors. 11 Effects o f potential inhibitors on slime droplet formation and structure of the secretory system. Journal of Experimental Botany 35. 1708-1714. Shewry, P. R . , Napier, J . A . and Tatham. A . (1Y95). Seed storage proteins: structures and biosynthesis. Plant Cell 7, 945-956. Shotwell, M. A . and Larkins, B. A. (19x8). The biochemistry and molecular biology of seed storage proteins. In “The Biochemistry of Plants” (B. .I.Miflin, ed.), pp. 297-345. Academic Press. New York. Smythe. E . and Warren, G . (1991). The mechanism of receptor-mediated endocytosis. Eitropeuti Journal of Biochemistry 202, 689-699. Staehelin, L. A . and Chapman, R . L. (1987). Secretion and membrane recycling in plant cells: novel intermediary structures visualized in ultrarapidly frozen sycamore and carrot suspension-culture cells. I’lanta 171, 43-57. Staehelin, L. A . and Moore. I . (1995). The plant Golgi apparatus: structure, functional organization and trafficking mechanisms. Annual Review of Plant Physiology and Plurit Molecular Biology 46. 261-288. Staehelin. L. A , . Giddings, T. H.. Levy. S . , Lynch, M. A, Moore, P. J . and Swords, K . M. M. (1991). Organization of the secretory pathway of cell wall glycoproteins and complex polysaccharides in plant cells. In “Endocytosis, Exocytosis and Vesicle Traffic in Plants” (C. R. Hawes. J . 0. D. Coleman and D. E . Evans, eds), pp. 183-198. Society for Experimental Biology, Seminar Series 45. Steer, M. W. and O’Driscoll. D. (1991). Vesicle dynamics and membrane turnover in plant cells. In “Endocytosis, Exocytosis and Vesicle Traffic in Plants” (C. R . Hawes, J . 0. D. Coleman and D. E . Evans, eds). pp. 129-142. Society for Experimental Biology, Seminar Series 45. Steinman. R . M.. Mellman, I. S . , Muller. W. A . and Cohn. Z . A . (1983). Endocytosis and recycling of plasma-membrane. Joirrnal of Cell Biology 96, 1-27. Steponkus. P. L. (1991). Behaviour of the plasma membrane during osmotic excursions. In “Endocytosis, Exocytosis and Vesicle Traffic in Plants” (C. R . Hawes, J. 0. D. Coleman and D. E . Evans, eds), pp. 103-128. Society for Experimental Biology, Seminar Series 45. Steponkus, P. L. and Wiest, S. C. (1979). Freeze-thaw-induced lesions in the plasma membrane. In “Low Temperature Stress in Crop Plants: The Role of the Membrane” (J. M. Lyons. D. Graham and J . K. Raison. eds), pp. 231-254. Academic Press, New York.

42

F. MARTY

Sze, H., Ward, J . M. and Lai, S. (1992). Vacuolar, H+-ATPases from plants: structure, function and isoforms. Journal of Bioenergetics and Biomembranes 24, 371-381. Tanchak, M. A. and Fowke, L. C. (1987). The morphology of multivesicular bodies in soybean protoplasts and their role in endocytosis. Protoplasma 138, 173-182. Tanchak, M., Griffing, L. R., Mersey, B . G. and Fowke, L. C. (1984). Endocytosis of cationized ferritin by coated vesicles of soybean protoplasts. Plantu 162, 48 1-486. Tanchak, M. A., Rennie, P. J . and Fowke, L. C. (1988). Ultrastructure of the partially coated reticulum and dictyosomes during endocytosis by soybean protoplasts. Planta 175, 433-441. Van der Wilden, W., Herman, E. M. and Chrispeels, M. J . (1980). Protein bodies of mung bean cotyledons as autophagic organelles. Proceedings of the National Academy of Sciences of the USA 77, 428-432. Villanueva, M. A , , Taylor, J . , Sui, X. and Griffing, L. R. (1993). Endocytosis in plant protoplasts: visualization and quantitation of fluid-phase endocytosis using silver-enhanced bovine serum albumin-gold. Journal of Experimental Botany 44, 257-281. Vitale, A. , Ceriotti, A , , Bollini, R . and Chrispeels, M. J . (1984). Biosynthesis and processing of phytohemagglutinin in developing bean cotyledons. European Journal of Cell Biology 141, 97-104. Walker, R. R. and Leigh, R. A . (1981). Mg2+-dependent,cation-stimulated inorganic pyrophosphatase associated with vacuoles of red beet (Beta vulgaris L.). Planta 153, 15&155. Wandelt, C. I . , Khan, R. I . , Craig, S . , Schroeder, H. E., Spencer, D. and Higgins, T. J. V. (1992). Vicilin with carboxy-terminal KDEL is retained in the endoplasmic reticulum and accumulates to high levels in the leaves of transgenic plants. Plant Journal 2, 181-192. Wartenberg, M., Hamann, J . , Pratsch, I. and Donath, E. (1992). Osmotically induced fluid-phase uptake of fluorescent markers by protoplasts of Chenopodium album. Protoplasma 166, 61-66. Wink, M. (1993). The plant vacuole: a multifunctional compartment. Journal of Experimental Botany 44, 231-246. Wittenbach, V. A., Lin, W. and Herbert, R. R. (1982). Vacuolar localization of proteases and degradation of chloroplasts in mesophyll protoplasts from senescing primary wheat leaves. Plant Physiology 69, 98-102. Zhang, F. and Boston, R. S. (1992). Increases in binding protein (BiP) accompany changes in protein body morphology in three high-lysine mutants of maize. Protoplasma 171, 142-152. Zhang, G . F. and Staehelin, L. A. (1992). Functional compartmentation of the Golgi apparatus of plant cells. Plant Physiology 99, 1070-1083. Zhang, G. F., Driouich, A . and Staehelin, L. A. (1993). Effect of monensin on plant Golgi: re-examination of the monensin-induced changes in cisternal architecture and functional activities of the Golgi apparatus of sycamore suspension-cultured cells. Journal of Cell Science 104, 819-831. Zhang, G.-F., Driouich, A. and Staehelin, L. A. (1996). Monensin-induced redistribution of enzymes and products from Golgi stacks to swollen vesicles in plant cells. European Journal of Cell Biology 71, 332-340.

Molecular Aspects of Vacuole Biogenesis

D. C. BASSHAM and N . V. R A I K H E L

MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, M I 48824-1312, USA

I. 11.

111.

IV. V.

Introduction

..........................................................................

43

Targeting of Soluble Proteins to the Vacuole .............................. .. A . N-terminal Propeptides .................................... .. ........ .. ... .... B. C-terminal Propeptides . . . . . . . . . . .. . . .. .. ... .. .. . .. .. . . . .. . .. . . . . . . . . .. . . . . . . . C. Internal Targeting Signals . . . .. . . . . . . . .. . . .. ole . .. . . . . .. ... . . . . . . . . D. Transport of Plant Proteins t o the Yeast

45 46 48 49 50

A.

Mechanism of Protein Transport to the Vacuole ............................ Multiple Mechanisms for Transport to the Vacuole . .. .... . . .... .. . . . B. Components of the Vacuolar Transport Machinery . ..... .. . .. . . . . . ..

50 51 51

Transport of Membrane Proteins to the Tonoplast

54

Perspectives .......................... Acknowledgements .. .. . . . .. ... . . . . . . . . .. . . . . . .. . . . .. . . . .. .. .. . . . . . .. .. . . . . . . . . .. . .. . . References . ........................................................

55 56 56

1.

INTRODUCTION

T h e transport of many newly synthesized proteins to the vacuole occurs through the secretory pathway (Figs 1 and 2 ; Marty, this volume). Proteins containing an N-terminal signal peptide are initially translocated across the endoplasmic reticulum (ER) membrane into the lumen (or, in the case of membrane proteins, inserted into the E R membrane). Further transport to Copyiiplit 0 1497 Academic Prmr Limited All rights ot reproducrion in any iorm reserved

44

D. C. BASSHAM and N . V. RAIKHEL

Fig. I. Electron micrograph showing cells froin the middle portion of an Ambidopsis rlzaliuna root. Organelles o f thc secretory pathway are visible. Various vacuolar morphologies can be seen within different cell types of the root. V, vacuolc; small arrow, endoplasmic reticulum; large arrowhead, Golgi. Scalc bar represcnts 2 win.

MOLECULAR ASPECTS OF VACUOLE BIOGENESIS

45

Fig. 2. Electron micrograph showing orgmelles of the secretory pathway in an Arubidop.si,s rhnlianu leaf. V. vacuole; ER. endoplasmic reticulum; G, Golgi. Scale bar represents 0.2 p m .

the vacuole often occurs via t h e Golgi apparatus, although this is not always the case (see Marty, this volume; Galili and Herman. this volume). Proteins are carried along the secretory pathway in a series of small membrane-bound vesicles, and transport is thus mediated by a process of vesicle budding from one compartment and fusion with the next. In this review we will discuss the signals involved in the targeting of proteins to the plant vacuole and the mechanisms by which vesicular transport to this organelle occurs.

11. TARGETING OF SOLUBLE PROTEINS TO THE VACUOLE Solublc vacuolar proteins are sorted from secreted proteins at the trans-Golgi network and transported to the vacuole in membrane-bound vesicles.

46

D. C. BASSHAM and N. V. RAIKHEL

Secretion is the “default pathway” - proteins lacking a targeting signal (other than the signal sequence) are transported to the cell surface. Proteins destined for the vacuole thus require sorting information in order for them to reach their correct site of function (for reviews, see Bednarek and Raikhel, 1992; Nakamura and Matsuoka, 1993). Protein transport to the mammalian lysosome (considered to be an equivalent organelle to the plant vacuole in some respects) has been well studied. Most soluble lysosomal proteins are specifically modified by the phosphorylation of mannose residues of their carbohydrate side-chains. These residues are recognized by mannose 6-phosphate receptors which mediate the transport of the proteins to the lysosome (Kornfeld and Mellman, 1989). Many plant vacuolar proteins are also modified by glycosylation in the E R and Golgi apparatus. In this case, however, the glycans do not contain sorting information, as demonstrated by removal of the glycosylation sites of various vacuolar proteins, which has no effect on their vacuolar targeting (Voelker et a l . , 1989; Sonnewald et ul., 1990; Wilkins et ul., 1990). 1.1 yeast, vacuolar targeting signals on soluble proteins do not consist of carbohydrate modifications. Instead, vacuolar proteins contain an N-terminal propeptide immediately following the signal sequence which is able to direct them to the vacuole. This short stretch of amino acids is recognized by a protein-specific receptor (at least in the case of carboxypeptidase Y, a soluble vacuolar protease; Marcusson et ul., 1994) which is responsible for the deposition of the protein in the vacuole. For a range of plant vacuolar proteins, it has now been demonstrated that targeting is dependent upon short peptide sequences within the protein. These signals can occur in the form of a propeptide (either N-terminal (NTPP) or C-terminal (CTPP)) which is removed proteolytically during or after transport to the vacuole, or can form part of the mature protein (Fig. 3A; Chrispeels and Raikhel, 1992). A . N-TERMINAL PROPEPTIDES

Several plant vacuolar proteins have been identified as containing vacuolar sorting information in an NTPP. Sporamin is a major vacuolar protein of the tuberous root of Zpomoeu bututus (sweet potato). It contains a 21-aminoacid signal sequence followed by a 16-amino-acid propeptide (Fig. 3B) which is removed to form the mature protein (Matsuoka et al., 1990). The wild-type prosporamin protein is correctly targeted to the vacuole in transgenic Nicotianu tabacum (tobacco) plants and transformed tobacco cell culture, although it is processed to a different size to the native protein in sweet potato (Matsuoka et al., 1990). However, when a construct lacking the NTPP is expressed in a tobacco cell culture, the protein is secreted to the medium (Matsuoka and Nakamura, 1991). This shows that the prosporamin NTPP is required for targeting to the vacuole and thus contains sorting information.

A

Carboxy-terminal propeptides

Barley lectin Tobacco chitinase Tobacco p- 1 ,3-glucanase Tobacco AP24

VFAEAIAANSTLVAE GLLVDTM VSGGVWDSSVETNATASLVSEM QAHPNFPLEMPGSDEVAK

Bednarek et a / , 1990 Neuhaus et al., 1991 Melchers et a/. 1993 Melchers et a/ , 1993

.

Amino-terminal propeptides

Sweet potato sporamin Barley aleurain

HSRFNPIRLPTTHEPA SSSSFADSNPWVTDRAASTLE . . . .

Matsuoka and Nakamura, 1991 Holwerda et a/., 1992

Fig. -3. Plant vacuolar targeting signals. ( A ) Schematic diagram of the types of targeting signals within vacuolar proteins. SP. signal peptidc; INTERNAL, signal hithin the mature protein. ( B ) Sequences of known vacuolar targeting propcptides.

48


Site-directed mutagenesis of the propeptide was used to determine some of the essential features for vacuolar targeting. While single amino acid substitutions in many cases do not affect function of the propeptide, the substitution of Ile28 and Am26 with Gly leads to significant proportions of the protein being secreted, indicating that these two residues are critical for correct sorting (Nakamura et a f . , 1993). These residues lie within a short sequence motif which is conserved between the NTPPs of several proteins (Nakamura et al., 1993) and may thus be of functional importance. Another protein containing an NTPP is aleurain, a vacuolar thiol protease. The targeting determinants of the proprotein have been identified by incorporating regions of the aleurain cDNA into the equivalent domain of proendoproteinase B (proEP-B), a secreted thiol protease. The proaleurain NTPP is able to redirect proEP-B to the vacuole in transgenic tobacco, showing that it contains vacuolar sorting information (Holwerda et al., 1992). Smaller regions of the propeptide are able to redirect a portion of the proEP-B to the vacuole, suggesting that these regions function together for efficient targeting. This has been confirmed by deletion of the appropriate regions from proaleurain, which leads to partial secretion (Holwerda et a f . ,1992). B . C-TERMINAL PROPEPTIDES

Stretches of amino acids at the C terminus of vacuolar proteins can also function as targeting signals. Hordeum vulgare (barley) lectin (BL) is a homodimeric chitin-binding protein involved in plant defence (Raikhel et a f . , 1993). It contains a 15-amino-acid hydrophobic CTPP (Fig. 3B) which is glycosylated, although the glycan is not required for transport to the vacuole (Wilkins et a f . , 1990). When a mutant BL cDNA lacking the CTPP is expressed in transgenic tobacco plants, the protein is secreted to the extracellular space (Bednarek et a f . ,1990), indicating that the CTPPcontains targeting information. A construct encoding a fusion protein consisting of BL (containing the CTPP) fused to the C terminus of Cucumis sativus (cucumber) chitinase, a normally secreted protein, has been expressed in tobacco and the protein is found in the vacuole. When the CTPP alone is fused to the C terminus of cucumber chitinase, approximately 70% of the protein is found in the vacuole (Bednarek and Raikhel, 1991). The CTPP is therefore able to act independently as a vacuolar targeting signal in the absence of the rest of the BL protein, although at a lower efficiency. The CTPP is thus both necessary and sufficient for vacuolar targeting. The presentation of the CTPP to the sorting machinery by the protein being transported may be important in determining the efficiency of sorting to the vacuole. Tobacco contains several different isoforms of chitinase, also a defencerelated protein, some of which are vacuolar and some extracellular. The vacuolar forms of chitinase contain a C-terminal extension when compared with extracellular forms. Tobacco basic chitinase A, normally a vacuolar


49

isoform, lacking the C-terminal extension of seven amino acids has been found to be extracellular. Cucumber chitinase (an extracellular chitinase) containing the seven-amino-acid CTPP of tobacco basic chitinase A (Fig. 3B) is found in the vacuole of tobacco cells. This CTPP thus constitutes the vacuolar targeting signal of the tobacco chitinase (Neuhaus et al., 1991). However, there is no primary structure homology between the CTPPs of different proteins. Extensive site-directed mutagenesis has been performed on the CTPPs of both BL (Dombrowski et al., 1993) and tobacco chitinase A (Neuhaus et al., 1994). No consensus sequence is found in the targeting signals, and in both cases many amino acid substitutions within the sequence of the CTPP are tolerated without abolishing the targeting activity. In the case of BL, a series of deletions indicated that as few as three residues are sufficient for a functional CTPP, whereas deletions of single internal amino acids of the tobacco chitinase CTPP cause secretion. The addition of glycine residues to the C terminus of the BL CTPP disrupts sorting, as does the movement of the glycosylation site closer to the C terminus. The CTPP is thus thought to interact with components of the sorting machinery from its C terminus. It is likely that the sorting machinery recognizes some structural features of the BL and tobacco chitinase CTPPs rather than specific amino acid sequences. Other proteins containing CTPPs which have been studied include tobacco P-1,3-glucanase and AP24, both pathogenesis-related vacuolar proteins. Proteins expressed from mutant constructs lacking CTPPs are secreted in transgenic tobacco, indicating that the CTPPs function as vacuolar targeting signals (Melchers et al., 1993). C. INTERNAL TARGETING SIGNALS

Proteins such as phytohaemagglutinin (PHA) do not contain cleavable vacuolar targeting signals but contain targeting information within regions of the mature protein (von Schaewen and Chrispeels, 1993). Portions of PHA have been fused to invertase, which is normally secreted, and the invertase activity in the vacuoles and secreted was measured. This indicated that a 30-amino-acid stretch within the protein (residues 84-1 13) is sufficient to redirect invertase to the vacuole. This region is predicted to lie at the surface of the PHA molecule, by comparison with the crystal structure of homologous lectins, and thus is in an appropriate position to interact with components of the targeting machinery. The targeting of the Vicia fabu (field bean) seed vacuolar protein legumin has been studied by creating fusions of regions of this protein with chloramphenicol acetyltransferase (CAT) and expression in transgenic tobacco plants (Saalbach et a f . ,1991). In this case, the entire legumin a chain is required for efficient transport of CAT to the vacuole. Smaller portions

50


of the protein are able to redirect CAT to the vacuole at a lower efficiency. It was therefore concluded that legumin contains multiple targeting signals which act together for efficient vacuolar transport. Interestingly, it has been noted that in roots, certain proteins which are normally vacuolar are found in the cell wall. These include proteinase inhibitors (Narvaez-Vasquez et a f . , 1993) and PHA (Kjemtrup et al., 1995); the PHA in root cell walls has been shown to be identical in sequence to the vacuolar PHA. In root meristem cells, PHA is correctly targeted to the vacuole, whereas in elongating cells, PHA is found only in the cell wall, suggesting that alternative targeting of the protein occurs in different cell types. It has been suggested that the targeting signal of PHA may not be recognized by the elongating cells, leading to its secretion (Kjemtrup etal., 1995). D. TRANSPORT OF PLANT PROTEINS TO THE YEAST VACUOLE

The presence of peptide targeting signals on both plant and yeast vacuolar proteins has led to the suggestion that vacuolar targeting signals may be conserved between these organisms. When a series of deletions of PHA are fused to the secreted reporter protein invertase and expressed in yeast cells, a short region of PHA is found to be sufficient to redirect invertase to the yeast vacuole (Tague et a f . , 1990). However, this domain is not sufficient to confer vacuolar localization of the reporter protein expressed in transgenic plants, and another region of PHA has been identified as the true plant vacuolar targeting signal (von Schaewen and Chrispeels, 1993; see above). This difference in the signal responsible for transport to the vacuole in yeast compared with plant cells has been observed for several other proteins. Sweet potato sporamin and barley lectin are also localized to the vacuole in yeast, but their NTPP and CTPP, respectively (identified as the plant targeting signals), are not required for this localization (Matsuoka and Nakamura, 1992; Gal and Raikhel, 1994). In addition, two plant vacuolar lectins, seed lectin and DB58 from the legume Dofichos bifiorus, are secreted when expressed in yeast (Chao and Etzler, 1994). It appears, therefore, that plant vacuolar targeting signals are not recognized by the yeast sorting machinery but that some plant vacuolar proteins can be transported to the yeast vacuole by a pathway independent of these signals.

111. MECHANISM OF PROTEIN TRANSPORT TO THE

VACUOLE The signals for targeting to the plant vacuole have thus been characterized in some detail now, with extensive mutagenesis performed in some cases to determine the features of the signals which are essential for function. However, very little is known about the mechanism by which these targeting


51

signals are recognized and divert proteins from the route of secretion for deposition in the vacuole. In particular, it is not known whether proteins containing the various types of targeting signals are transported to the vacuole in the same vesicle or whether there are classes of vesicles responsible for the transport of different proteins. A . MULTIPLE MECHANISMS FOR TRANSPORT 1 0 THE VACUOLE

One important question which arises from the differences in vacuolar targeting signals and the possibility of more than one receptor is whether there are also multiple mechanisms for the transport of proteins to the vacuole. Co-expression of barley lectin and sporamin in transgenic tobacco plants followed by electron microscopic immunolocalization and pulse-chase analysis indicate that these two proteins are localized to the same vacuoles in leaves (Schroeder et ul., 1993). The proBL CTPP and prosporamin NTPP have been shown to be interchangeable in vacuolar targeting by exchange of the two signals (Matsuoka el af., 1995). The mechanism of transport of these two proteins has been investigated in transgenic tobacco cells using the fungal metabolite wortmannin, an inhibitor of phosphatidylinositol 3-kinase (PI 3-kinase) activity. A PI 3-kinase (Vps34 protein) has been demonstrated to be essential for vacuolar protein sorting in yeast (Stack et al., 1993). Both BL and sporamin containing the proBL CTPP are missorted to the cell surface in the presence of wortmannin, whereas proteins containing the prosporamin NTPP are correctly targeted to the vacuole under these conditions (Matsuoka ef ul., 1995). This indicates that CTPP- and NTPP-mediated transport occurs via different mechanisms, CTPP-mediated transport being sensitive to wortniannin and NTPP-mediated transport insensitive to wortmannin. The target of wortniannin i n tobacco cells has also been investigated: phospholipid synthesis as well as PI kinase activities were shown to be inhibited by this metabolite. A comparison of dose-dependencies of inhibition suggests that the synthesis of phospholipids may be involved in CTPP-dependent vacuolar transport (Matsuoka et af., 1995). B.

COMPONENTS OF THE VACUOLAR TRANSPORT MACHINERY

Both from analogy with other systems. and experiments showing that the vacuolar sorting process is saturable (see below). it is assumed that receptor proteins recognize vacuolar targeting signals and cause proteins containing these signals to be packaged into transport vesicles destined f o r the vacuole.

I. NTPP-binding protein The only candidate thus far identified for a vacuolar targeting signal receptor is a protein from pea which is able to specifically bind t o the proaleurain

52


NTPP (Kirsch et al., 1994). A proaleurain affinity column consisting of immobilized NTPP was used to isolate an approximately 80 kDa glycoprotein (BP-80) from a detergent extract of pea clathrin-coated vesicles, which are thought to transport proteins to the vacuole (Harley and Beevers, 1989). The C-terminal 5 kDa of this membrane protein is extravesicular, with the N terminus inside the vesicle. The intravesicular portion alone is still retained on the proaleurain column, indicating that it contains the proaleurain binding region. The binding properties of the 80 kDa binding protein have been studied by the binding of radiolabelled peptide corresponding to the proaleurain NTPP to solubilized vesicles, and bound peptide measured in a precipitation assay. The binding data are consistent with a single binding component in the extract, and unlabelled peptide is able to compete for binding, indicating a saturable binding site. The binding is pH-dependent, with a maximum at about pH 6, consistent with a role in vacuolar transport. The receptor is proposed to bind to its ligand in the Golgi apparatus and to dissociate again in the low pH of the vacuole. Photoaffinity cross-linking of the labelled NTPP peptide confirms that the peptide is bound to a protein of approximately 80 kDa, and that the same protein is probably functioning in both the affinity chromatography and binding experiments (Kirsch et al., 1994). The binding specificity of BP-80 has been examined using affinity columns containing peptides corresponding to vacuolar targeting signals from several proteins. In addition to binding the proaleurain NTPP, BP-80 is also retained on columns containing the prosporamin NTPP and the targeting sequence from Bertholletia excelsa (Brazil nut) 2s albumin, but not the probarley lectin CTPP (Kirsch et al., 1996). BP-80 may therefore mediate the targeting of a subset of vacuolar proteins, but is unlikely to be involved in the transport of all proteins to the vacuole. The existence of a receptor for the CTPP-containing vacuolar protein tobacco chitinase has been implied by overexpression studies in tobacco protoplasts. At low levels of expression, the chitinase is found exclusively in the vacuole; however, at high expression levels a proportion of the protein is secreted (Neuhaus et al., 1994). This indicates that sorting to the vacuole is a saturable process and therefore is likely to be receptor-mediated, although no candidates for a receptor have yet been isolated. In addition to attempts to isolate a receptor protein for vacuolar proteins, several proteins have been identified which may be components of the machinery responsible for vesicle formation, targeting or fusion between the TGN and the vacuole. 2. AtPEP12 - a plant syntaxin homologue Data from various organisms, including biochemical studies in mammalian cells and yeast genetics, indicate that some of the basic machinery of vesicle transport is conserved between diverse organisms and cell types. Proteins


53

implicated in vesicle docking and fusion at the presynaptic membrane of neuronal cells (Sollner el al., 1993) have been found to have homologues in yeast which function in vesicle trafficking through the secretory pathway (for a review, see Bennett and Scheller, 1994). These include the presvnaptic membrane proteins syntaxin and SNAP-25 (synaptosomal-associated protein of 25 kDa) which interact with the synaptic vesicle protein synaptobrevin, along with soluble factors, to form a docking complex and to allow vesicle fusion to occur. Different isoforms of syntaxin and synaptobrevin exist on distinct cell membrane types, implying that each vesicle fusion step of the secretory pathway may involve a particular isoform of each protein. These membrane proteins could therefore help to regulate the specificity of fusion by ensuring that a vesicle fuses with the correct membrane type (Bennett and Scheller, 1994). The yeast PEP12 gene encodes a syntaxin honiologue involved in vacuolar targeting (Jones, 1977; Rothman et al., 1989) which resides on the yeast vacuolar or prevacuolar membrane (Bennett and Scheller, 1994). The protein is therefore thought to act as a receptor for vesicles transporting proteins between the trans-Golgi network and the vacuole. A pep12 mutant lacks carboxypeptidase Y (CPY) activity, a vacuolar protease, as CPY is secreted in this mutant instead of being transported to the vacuole. CPY is normally activated by proteolysis in the vacuole, and thus the secreted CPY is inactive. This phenotype of the pep12 mutant has been utilized in a screen for an Arabidopsis thafianahomologue of PEP12. The mutant was transformed with an Arabidopsis cDNA library and colonies screened for the restoration of CPY activity. A cDNA (called A t P E P l 2 ) was isolated which complements the yeast mutant and was found to be homologous to the yeast PEP12 gene and other syntaxins (Bassham et ul., 1995). Northern blot analysis indicated that the mRNA is present in all tissues of Arahidopsis tested, but at a very low level in leaves compared with roots, stems and flowers. The significance of this is unclear, as leaves appear capable of correctly transporting constitutively expressed foreign proteins to the vacuole (N. V. Raikhel, unpublished results). The RNA is present in all cell types in both roots and leaves. shown by in s i m hybridization experiments, and cell-specific expression therefore cannot account for the low level in leaves. The AtPepl2 protein is thus a candidate for a component of the plant vacuolar transport machinery, but confirmation o f this awaits localization of the protein and functional studies. However, it does appear that, despite the differences between the targeting signals of yeast and plant vacuolar proteins, some components of the transport machinery may be conserved between these organisms. 3. PI 3-kinase activity As mentioned above, a PI 3-kinase (VPS34)is essential for vacuolar transport in yeast (Stack er al., 1993). although it is unclear whether this is the case

54

D. C . BASSHAM and N. V. RAIKHEL

in plants. This has led Welters et al. (1994) to clone a PI 3-kinase from Arabidopsis by polymerase chain reaction, based on conserved regions in several PI 3-kinases, including the yeast VPS34 gene. The Arabidopsis gene (called AtVPS34) shows homology to yeast VPS34 and to a mammalian PI 3-kinase, and overexpression in transgenic plants indicated that the gene encodes a functional PI 3-kinase. However, AtVPS34 is not able to complement a yeast vps34 mutant, although the C-terminal third containing the catalytic site can functionally replace this region of yeast VPS34 in a chimeric gene. The function of AtVPS34 in vivo is not yet known.

4. lnvolvement of GTP-binding proteins in vacuolar transport It has been shown for many vesicle transport events that small GTP-binding proteins are key regulators of transport activity and specificity (for a review, see Balch, 1990). Vesicles carrying vacuolar precursor proteins have been purified from pumpkin cotyledons and vesicle membranes isolated. GTPbinding proteins in the membranes were identified by ligand blot analysis using radiolabelled GTP. Two proteins (25 and 27 kDa) bind GTP in this assay and do not bind other NTPs (Shimada et al., 1994). Detergent is required to solubilize the proteins from the membranes, suggesting that they are integral membrane proteins. However, a role for these proteins in transport remains to be established. A class of large GTP-binding proteins has also been implicated in several membrane trafficking events. One member of this class is yeast Vpsl, which is necessary for correct sorting to the vacuole (Rothman et al., 1990; Vater et al., 1992). An Arubidopsis cDNA (aG68) has been isolated which shows high homology to VPSl in the N-terminal GTP-binding domain, although diverges at the C terminus (Dombrowski and Raikhel, 1995). The C-terminal portion of the protein was implicated in protein-protein interactions, which could indicate that the aG68 protein has a function different to that of Vpsl. In addition, aG68 cannot complement the yeast vpsl mutant. aG68 could therefore be involved in a different transport process in plants, although this has yet to be determined.

IV. TRANSPORT OF MEMBRANE PROTEINS TO THE TONOPLAST Despite the identification of targeting signals in a number of soluble vacuolar proteins, the sorting of vacuolar membrane proteins has been studied very little, and the signals and mechanisms by which targeting to the tonoplast is achieved are still unclear. a-TIP (tonoplast intrinsic protein) is a tonoplast protein with six membrane-spanning domains which travels through the secretory pathway to the vacuole. Fusion of a portion of a-TIP consisting of the sixth transmembrane domain and the cytoplasmic tail to the reporter


5s

protein phosphinotricine acetyltransfersse (PAT) is sufficient to redirect this reporter to the tonoplast. As an a-TIP deletion mutant lacking the cytoplasmic tail is still correctly targeted to the tonoplast, it appears that the sixth transmembrane domain is sufficient for tonoplast targeting of the reporter protein (Hofte and Chrispeels. 1992). However, it is not clear whether this domain contains a specific targeting signal or whether the vacuole is the default destination for membrane proteins which enter the secretory system in plant cells, requiring n o targeting signal. The rncchanisms of transport of a vacuolar membrane protein (a-TIP) and soluble protein (PHA) have been compared in transgenic tobacco protoplasts using the vesicle transport inhibitors monensin and brefeldin A (BFA; Gomez and Chrispeels, 1993). Both of these inhibitors prevent the transport of PHA to the vacuole. However, transport of a-TIP is not affected by either monensin or BFA, and some vesicle transport must therefore continue in t h e presence of these compounds. From these data it appears that PHA and a-TIP transport occurs by different mechanisms and that these two proteins are probably carried in different types of vesicles.

V.

PERSPECTIVES

The transport of soluble proteins through the plant secrctory pathway to the vacuole has been studied in some detail at the level of the targeting signals within the proteins. However, many major questions remain about the mechanisms by which the proteins are transported. Few components of the vesicle transport machinery have been isolated, and the similarities between these components and those of other organisms such as yeast are unclear. Some components (such as Pepl2) may be common to all vesicle transport systems, but others could be plant-specific. In particular. these are expected to include receptors for vacuolar proteins containing various types of targeting signals. It is not known whether all proteins are transported to the vacuole in the same type of vesicles or if different vesicle classes function in the transport of various proteins. In addition, the mechanism of transport of membrane proteins to the tonoplast is still unclear, and the det'ault destination of membrane proteins within the secrctory pathway is not known. Another aspect of vacuolar transport which is now being studied in plants is the differential regulation of transport processes and machineries between various tissues and cell types. There appears to be a difference in the expression level of some components o f the sorting machinery between tissues which may reflect functional variations i n divergent tissues and their constituent cell types. Mechanistic differences between cell types have also been implicated, as certain proteins which are vacuolar in leaves are found extracellularly in roots. The availability of marker proteins in the various organelles of the plant endomemhrane system will now allow many of thcse issues to bc addressed.

56


ACKNOWLEDGEMENTS We thank members of the Raikhel group for their helpful comments and discussions. Research was supported by grants from the Department of Energy No. DE-AC02-76ERO-1338 and the National Science Foundation NO. MCB-9507030 to N.V.R.

REFERENCES Balch, W. E. (1990). Small GTP-binding proteins in vesicular transport. Trends in Biochemical Sciences 15, 473-477. Bassham, D. C., Gal, S . , Conceiqio, A. D. S. and Raikhel, N. V. (1995). An Arabidopsis syntaxin homologue isolated by functional complementation of a yeast pepI2 mutant. Proceedings of the National Academy of Sciences of the USA 92, 7262-7266. Bednarek, S. Y. and Raikhel, N. V. (1991). The barley lectin carboxyl-terminal propeptide is a vacuolar protein sorting determinant in plants. Plant Cell 3, 11%-1206. Bednarek, S . Y. and Raikhel, N. V. (1992). Intracellular trafficking of secretory proteins. Plant Molecular Biology 20, 133-150. Bednarek, S . Y., Wilkins, T. A . , Dombrowski, J. E. and Raikhel, N . V. (1990). A carboxyl-terminal propeptide is necessary for proper sorting of barley lectin to vacuoles of tobacco. Plant Cell 2, 1145-1155. Bennett, M. K . and Scheller, R. H. (1994). A molecular description of synaptic vesicle membrane trafficking. Annual Review of Biochemistry 63, 63-100. Chao, Q. and Etzler, M. E. (1994). Incorrect targeting of plant vacuolar lectins in yeast. Journal of Biological Chemistry 269, 20 866-20 871. Chrispeels, M. J. and Raikhel, N . V. (1992). Short peptide domains target proteins to plant vacuoles. Cell 68, 613-616. Dombrowski, J. E. and Raikhel, N . V. (1995). Isolation of a cDNA encoding a novel GTP-binding protein of Arabidopsis thaliana. Plant Molecular Biology 28, 1121-1 126. Dombrowski, J. E., Schroeder, M. R., Bednarek, S. Y. and Raikhel, N. V. (1993). Determination of the functional elements within the vacuolar targeting signal of barley lectin. Plant Cell 5, 587-596. Gal, S. and Raikhel, N . V. (1994). A carboxy-terminal plant vacuolar targeting signal is not recognized by yeast. Plant Journal 6, 235-240. Gomez, L. and Chrispeels, M. J . (1993). Tonoplast and soluble vacuolar proteins are targeted by different mechanisms. Plant Cell 5, 1 1 13-1124. Harley, S. M. and Beevers, L. (1989). Coated vesicles are involved in the transport of storage proteins during seed development in Pisum sativum L. Plant Physiology 91, 674-678. Hbfte, H. and Chrispeels, M. J . (1992). Protein sorting to the vacuolar membrane. Plant Cell 4. 995-1004. Holwerda, B. C., Padgett, H. S. and Rogers, J . C. (1992). Proaleurain vacuolar targeting is mediated by short contiguous peptide interactions. Plant Cell 4, 307-3 18, Jones, E. W. (1977). Proteinase mutants of Saccharomyces cerevisiae. Genetics 85, 23-33. Kirsch, T.. Paris, N . , Butler, J. M., Beevers, L. and Rogers, J . C . (1994). Purification

MOI.ECULAR ASPECTS OF VACUOLE BIOGENESIS

57

and initial characterization of a potential plant vacuolar targeting receptor. Proceedirigs of the Natiotiul Acudemy of Sciences of t h U ~ S A 91, 3403-3407. Kirsch, T., Saalbach, G., Raikhel, N. V. and Beevcrs. L. (19%). Interaction of a potential vacuolar targeting receptor with amino- and carboxy-terminal targeting determinants. Plant Physiology 111, 469474. Kjemtrup. S., Borksenious, 0.. Raikhel, N. V. and Chrispeels, M. J. (1995). Targeting and release of phytohemagglutiiiin from the roots of bean seedlings. Plant Physiology 109, 603410. Kornfeld. S. and Mellman. I. (1089). The biogcncsis of lysosomes. Annitul Rrview o j Cell Biology 5. 483-525. Marcusson, E. G., Horazdovsky, B. F., Cereghino. 1. L.. Gharakhanian, E. and Emr, S. D. (1994). The sorting receptor for yeast vacuolar carboxypeptidase Y is encoded by the VPSIO gene. Cdl 77, 579-586. Matsuoka. K. and Nakamura, K . (1991). Propeptide of a precursor to a plant vacuolar protein required for vacuolar targeting. Proceedings o f ihe National Acacienzy of Sciences of the USA 88, 834-838. Matsuoka. K. and Nakamura. K. (1992). Transport of a sweet potato storage protein, sporamin. to the vacuole in yeast cells. Plurzt Cell Physiology 33, 453-462. Matsuoka, K., Matsumoto, S.. Hattori, T., Machida, S. and Nakamura, K. (1990). Vacuolar targeting and post-translational processing of the precursor to the sweet potato tuberous root storage protein in heterologous plant cells. Journal o f Biological Chemistry 265, 19 75C-19 757. Matsuoka, K., Bassham. D. C.. Raikhel, N. V. and Nakamura. K. (1995). Different sensitivity to wortmannin of two vacuolar sorting signals indicates the presence of distinct sorting machineries in tohacco cells. Journul of Cell Biology 130. 1307- 1318.

Melchers, L. S., Sela-Buurlage, M. B., Vloemans, S. A., Woloshuk, C . P . , Van Roekel, J . S. C . , Pen, J . . van den Elzen. P. J . M. and Cornelissen. B . 3. C . (1993). Extracellular targeting of the vacuolar tobacco proteins AP24, chitinase and p- 1,3-glucanase in transgenic plants. Plant Moleciilar Biology 21, 583593. Nakamura, K. and Matsuoka, K. (1993). Protein targeting to the vacuole in plant cells. Plunt Physiology 101, 1-5. Nakamura. K . , Matsuoka, K., Mukumoto, F. and Watanabe, N. (1993). Processing and transport to the vacuole of ;I precursor to sweet potato sporamin in transformed tobacco cell line BY-2. Journal of Experimental Botany 44(Supplement), 33 1-338. Narvaez-VBsquez, J . , Franceschi, V. R. and Ryan, C. A. (1903). Proteinase-inhibitor synthesis in tomato plants: evidence for extracellular deposition in roots through the secretory pathway. Pluntu 189, 257-266. Neuhaus, J.-M.. Sticher, L., Meins, F. and Bollcr, ‘r. (1991). A short C-terminal sequence is necessary and sufficient for the targeting of chitinases to the plant vacuole. Proceetlings of the Nutionul Academy of Sciences of the U S A 88, 10 362-10 366. Neuhaus. J.-M., Pietrzak, M. and Boller. -1. (1904). Mutation analysis of the C-terminal vacuolar targeting peptidc of tobacco chitinase: low specificity o f the sorting system. and gradual transition between intracellular retention and secretion into the extracellular space. Plant Joicriztzl 5, 15-54. Raikhel, N. V., Lee, 14.-I. and Broekaert. W. F. (1903). Structure and function of chitin-binding proteins. Annuul Review qf Plant Physiology und Plunt Molecnlur Biology 44, 5 9 1 4 1 5 . Rothman. J . H., Howald, 1. and Stevens, ‘I. H . (1Y89). Characterization of genes

58


required for protein sorting and vacuolar function in the yeast Saccharomyces cerevisiae. EMBO Journal 8, 2057-2065. Rothman, J . H . , Raymond, C. K., Gilbert, T., O’Hara, P. J. and Stevens, T. H. (1990). A putative GTP-binding protein homologous to interferon-inducible Mx proteins performs an essential function in yeast protein sorting. Cell 61, 1063-1074. Saalbach, G . , Jung, R., Kunze, G . , Saalbach, I., Adler, K. and Muntz, K. (1991). Different legumin protein domains act as vacuolar targeting signals. Plant Cell 3, 695-708. Schroeder, M . R . , Borkhsenious, 0. N., Matsuoka, K., Nakamura, K. and Raikhel, N. V. (1993). Colocalization of barley lectin and sporamin in vacuoles of transgenic tobacco plants. Plant Physiology 101, 4511158. Shimada, T., Nishimura, M. and Hara-Nishimura, I . (1994). Small GTP-binding proteins are associated with the vesicles that are targeted to vacuoles in developing pumpkin cotyledons. Plant Cell Physiology 35, 995-1001. Sollner, T., Whiteheart, S. W., Brunner, M., Erdjument-Bromage, H., Gerornanos, S., Tempst, P. and Rothman, J. E. (1993). SNAP receptors implicated in vesicle targeting and fusion. Nature 362, 318-324. Sonnewald, U., von Schaewen, A . and Willmitzer, L. (1990). Expression of mutant patatin protein in transgenic tobacco plants: role of glycans and intracellular location. Plant Cell 2 , 345-355. Stack, J . H., Herman, P. K., Schu, P. V. and Ernr, S. D. (1993). A membraneassociated complex containing the VpslS protein kinase and the Vps34 PI 3-kinase is essential for protein sorting to the yeast lysosome-like vacuole. EMBO Journal 12, 2195-2204. Tague, B. W., Dickenson, C. D. and Chrispeels, M. J . (1990). A short domain of the plant vacuolar protein phytohemagglutinin targets invertase to the yeast vacuole. Plant Cell 2 , 533-546. Vater, C . A . , Raymond, C. K . , Ekena, K., Howald-Stevenson, I . and Stevens, T. H. (1992). The VPSl protein, a homolog of dynamin required for vacuolar protein sorting in Saccharomyces cerevisiae. is a GTPase with two functionally separable domains. Journal of Cell Biology 119, 773-786. Voelker, T. A., Herman, E. M. and Chrispeels, M. J. (1989). In vitro mutated phytohemagglutinin genes expressed in tobacco seeds. Role of glycans in proteins targeting and stability. Plant Cell 1, 95-104. von Schaewen, A. and Chrispeels, M. J. (1993). Identification of vacuolar sorting information in phytohemagglutinin, an unprocessed vacuolar protein. Journal of Experimental Botany 44(Supplement), 339-342. Welters, P., Takegawa, K . , Emr, S. D. and Chrispeels, M . J. (1994). AtVPS34, a phosphatidylinositol 3-kinase of Arabidopsis thaliana, is an essential protein with homology to a calcium-dependent lipid binding domain. Proceedings of the National Academy of Sciences of the USA 91, 11 398-11 402. Wilkins, T. A , , Bednarek. S. Y . and Raikhel, N. V. (1990). Role of propeptide glycan in post-translational processing and transport of barley lectin to vacuoles in transgenic tobacco. Plant Cell 2, 301-313.

The Vacuole: a Cost-Benefit Analysis

J . A . RAVEN

Depurtment of Biological Sciences, University of Dundee, Dundee DDl 4HN, UK

I.

Introduction

........................................................................

50

11. Demonstrated and Hypothesized Functions of Vacuoles. and Alternative Means of Performing these Functions .........................................

62

Demonstrated and Hypothesized Costs o f Producing and Maintaining Vacuoles, and of Costs of Alternative Means of Performing Vacuolar Functions ..............................................................................

62

111.

IV. V. VI.

A Case History: ”Vacuolate” (Eu)bacteria

.................................

62

...........................

75

Another Case History: Vacuoles and Buoyancy

Costs and Benefits of Vacuolation: Simple Analyses and the Allocation of Costs Among Various Benefits ..............................................

78

VII.

........................................

81

VIII.

.............................................................. Conclusions . Acknowledgements ......................... ..... References .............................................................................

82 82 81

1.

INTRODUCTION

A largc number of functions have been demonstrated, or suggested, for the aqueous vacuoles of plants (including algae) and fungi (including the “pseudofungal” oomycetes), as well as vacuole-like structures (not “gas

60

J. A. RAVEN

vesicles”) in prokaryotes (e.g. Raven, 1987; Fassing el al., 1995). If we are to understand the distribution of vacuoles, then these functions of the vacuole which yield quantifiable benefits must be considered in the context of the costs of synthesizing and maintaining the vacuoles, including any consequential requirements of vacuolation such as increased resource allocation to cell walls. Such cost-benefit analyses are predicated on maximizing the inclusive fitness of the organisms by optimal acquisition and allocation of resources (Osmond et al., 1980). The costs and benefits must be expressed in the same resource units: the appropriate units in a given environment will be determined by the resource which is most limiting for growth (Liebig, 1840) as determined by sensitivity analysis. When light is limiting the growth rate then energy (or carbon) is the appropriate resource to concentrate on in cost-benefit analyses (Raven, 1984a,b, 1985). If water is limiting for a land plant then reduced carbon, whose production from atmospheric C 0 2 involves the expenditure of water in transpiration, may be an appropriate unit to use (Raven, 1985, 1987), remembering, of course, the water retained during growth in the vacuoles or their alternatives. Similar arguments apply to limitation by phosphorus or nitrogen (Raven, 1987). These considerations related to the role(s) of vacuolation in the acquisition of limiting resource, i.e. alleviating stress sensu Grime (1979). We must also consider biomass retention in the sense of the costs and benefits of vacuoles in reducing biophagy, including parasitism, i.e. overcoming biotic disturbance sensu Grime (1979). Considerations so far all relate to biomass production or retention. Inclusive fitness involves successful reproduction, so we must also consider biotic resources needed for pollination (more generally, fertilization) or for dispersal of propagules. Finally, the multiple proposed benefits, and known costs, of vacuolation mean that multiobjective optimization methods must eventually be used (Farnsworth and Niklas, 1995). Having considered what is meant by cost-benefit analyses, the end-product of such analyses over a wide phylogenetic range of organisms, metabolic types and of life-forms (sensu Raunkaier, 1934; Luther, 1949) should be some rationale for the extent of vacuolation (vacuolar volume per unit cytoplasmic volume) in different organisms. The range of vacuo1e:cytoplasm among phototrophs is from 0.05 (many microalgae; host invertebrates in various symbioses with microscopic photolithotrophs (or chemolithotrophs)) to 250 (giant-celled macroalgae; photosynthetic cells of CAM (crassulacean acid metabolism) plants) (Raven, 1987, 1993a, 1995a; Winter, Robinson and Heldt, 1993, 1994). Such a rationale requires consideration of alternatives to vacuolation in providing non-cytoplasmic fluid phases (coelom, haemocoel, coelenteron in various metazoa; apoplasmic water in plants) (Raven, 1987).

THE VACI!OLE: A COST-BENEFIT ANALYSIS

61

Space does not permit a detailed quantitative cost-benefit analysis of all vacuolar functions. Such detail is given for many functions by Raven (1OX7), and the conclusions are briefly summarized here as an introduction to the tabulated material (Tables I-IV) and to the two case histories (text, and Fig. 1) presented in this article. An inescapable effect of vacuolation of a cell of a given shape is to increase the surface area per unit volume of cytoplasm. This increases the rate of resource (gaseous or dissolved chemicals: photons) acquisition from resourcelimited environments on a unit cytoplasm basis. Although extra material and energy costs are incurred in building and maintaining a vacuolate cell as compared to a non-vacuolate cell with the same volume of cytoplasm, the benefits generally exceed the costs under resource-limited conditions. However, under resource-saturated conditions, vacuolation may mean a lower maximum specific growth rate due to resource diversion from catalytic apparatus to the additional wall as well as vacuolar material in the vacuolate cell. Alternatives to vacuolation in achieving additional resource acquisition rates per unit cytoplasm under resource-limited conditions include changes in cell shape (less closely approximating to a sphere) or the occurrence of apoplasmic volume in multicellular organisms. These alternatives do not exclude vacuolation, and themselves have material and energy costs similar to those of vacuolation which partly offset the benefits which accrue at low resource availability. Vacuolation also has other costs and other benefits. These further costs only apply in particular habits, e.g. slow turgorholume adjustment of large vacuolate cells in relation to the rate o f change of external osmolarity in estuaries, and a possible decrease in desiccation tolerance as a result o f vacuolation which mainly affects "lower" plants in desiccation-prone environments. Benefits in addition to these geometric effects related to resource acquisition in resource-deficient environments include the storage, manipulation and protection of already acquired resources, the accumulation o f end-products of (for example) acid-base regulation, and contributions t o outbreeding and dispersal, as well as further contributions to resource acquisition in CAM and in rhizospherc acidification attendant on accumulation of endogenously produced organic acids R S their salts. While some o f these benefits may be attained in other ways (e.g. storage of hcxose as polysaccharide in plastids or cytosol rather than mono- or disaccharidcs in thc vacuole) others cannot (e.g. die1 changes in low-M, organic acids in CAM). Overall these additional benefits outweigh the additional costs of vacuolation and may account for vacuolation in those plants (if such there be) which invariably grow naturally in resource-rich habitats and in which vacuolation might be expected to diminish fitness by decreasing the achieved specific growth rate.

62

J. A . RAVEN

11. DEMONSTRATED AND HYPOTHESIZED FUNCTIONS OF VACUOLES, AND ALTERNATIVE MEANS OF PERFORMING THESE FUNCTIONS Table I lists a number of vacuolar functions (benefits), and of alternative ways of performing these functions. From this mainly qualitative menu, two examples not considered by Raven (1987) are subsequently taken for analysis in the text.

111. DEMONSTRATED AND HYPOTHESIZED COSTS OF PRODUCING AND MAINTAINING VACUOLES, AND OF COSTS OF ALTERNATIVE MEANS OF PERFORMING VACUOLAR FUNCTIONS Table I1 lists a number of costs of producing and maintaining vacuoles. and of costs of alternative means of performing vacuolar functions.

IV. A CASE HISTORY: “VACUOLATE” (EU)BACTERIA Prokaryotes d o not have an endomembrane system, and thus cannot have a vacuole in the sense that eukaryotic cells do, i.e. a hypertropied lysosome, since a lysosome is a differentiated part of the endomembrane system. However, certain eubacteria do have aqueous P phases (sensu Mitchell, 1979) bounded by protein-lipid bilayer membranes which occupy a considerable fraction of the cell volume (Fassing et al., 1995; Jamasch, 1995). These prokaryotes are marine S*--oxidizing eubacteria of the genera Beggiatoa and Thioplaca and have filaments of vacuolate cells up to 100 pin or more in diameter. They obtain the energy needed for (chemolithotrophic) growth and maintenance by oxidizing S2- or So using 0 2 or N 0 3 - as an electron acceptor. Their habitat is ocean floor sediments in which the rain of reduced organic carbon from surface photolithotrophy exceeds the rate at which O2 can diffuse into the sediments. Microbial chemo-organotrophy oxidizes more moles of organic carbon than moles of 0 2 are available, so S042- is used as terminal electron acceptor instead of 0 2 . Sulfate is abundant in seawater: 25 rnol mP3 S042- (8 mol of electrons consumed upon complete reduction of 1 mol) relative to -0.25 mol m-3 O2 (4 rnol of electrons consumed upon complete reduction of 1 rnol), i.e. 200 times more electron acceptors per unit volume of seawater as S042- than as 02.Of course, S042- is an energetically much inferior electron acceptor than is 0 2 in terms of joules available per electron transferred from organic carbon to the acceptor. Beggiatoa and Thioplaca live near the chemocline separating the seawater containing O2 and NO3- from the anoxic sediment containing H2S.

TABLE I Vacitolar ficnctions and alferiiativc wrzy of performing these fiinction.r Function pel-foi-med by vacuole

References ~

~~

~~

~

( I ) Geometric ( a ) Increases the surface area of plasmalemma exposed to the cnvironnient per unit volume o f cytoplasm. Increases the potential for light absorption per unit cytoplasmic volume from a given radiation field. Increases the potential for solute and water influx across the plasmalemma by lipid solution passive flux with a given concentration difference, or mediated passive o r active transport at thc plasmalemma. Increases the volume o f substrate exploited (important if thc solute has a lo\\ diffusion cocfficicnt in the suhstratc. e.g. phosphate in oxidized soil)

Alteration of cell shape from near-spherical to less spherical (can be combined with vacuolation), yielding evaginations (root, shoot. thallus hairs) or invaginatiom (transfer cells). Altered morphology at the supracellular level. yielding intercellular gas spaces in plants. gills/trachea/lungs/boo~ lungs in metazoa (can be combined with vacuolation in plants). Presence o f internal apoplasmic aqueous spaces (xylem. leptome in embryophytes; intercellular spaces in algae: blood. haemocoel. coelom. coelenteron in metazoa). Animal spaccs can, like vacuoles, transmit hydrostatic pressure and be part o f a hydrostatic skeleton: plant apopl;isrnic spaces generally cannot. Photosynthetic (symbiotic) animals generally have an exoskeleton (corals. for arne ni fe ra . hiva lves : n o t certa I n h ydroids . or Cori I d i m )

Raven (1981. 1Y8la.b, 19x7. I9Y3a. lYY5c. 1996)

(13) \':icuoles x e iinportant in cells which undergo I-apid (seconds-minutes) changes in volume. e.g stomatal guard cells: pulvinar motor cells). Delegation of most of the volume change to the vacuole meam that concentrations of proteins. metaholitea or effectors in the cytoplasm are little altered.

N o viable alternative which does not cause change5 in protein. metabolites or effector concentrations in the cytoplasm which would disturb metabolism

Raven (1987)

However, estuarine algae often have cell walls with a low bulk elastic modulus which means that changes in external osmolarity as a function of time in the tidal cycle are reflected initially (and. to a large extent. over the time between high and low tide) in protoplast wdunie changes. and hence changes in protein, mctabolite or effector concentrations

Reed ct a / . (1980) Young et (11. (1987)

TABLE I (continued) Vacuolar functions and alternative ways of performing these functions Function performed by vacuole

Alternative means of performing the function

References

Rapid increase in volume of traps of Utricularia caused by pumping ions (hence water) out of the apoplasmic lumen of a trap. followed by "firing", yielding an inrush of water (plus. with luck. prey)

Sydenham and Findlay (1975)

Pressurization o f internal tissues leading to explosive ("squirting") seed dispersal ( Echalliiim, Echiriocystis. Itnyatiens. Oxalis), a result of vacuolar accumulation of solutes (sugars, glycosides) in highly vacuolate tissue (cytoplasmic pressure = vacuolar pressure)

An internal aqueous apoplasmic phase containing the seeds which is pressurized by solute accumulation and contained by the pericarp could cause explosive seed dispersal; no examples known. Proposal is analogous to '.jet propulsion" of cephalopods. opposite of firing of the Utriciclaria trap

Fahn (1974) Niklas (1992) Walker er al. (1995)

Contractile roots operate (in part) by collapse and death of highly vacuole cell layers between the cell layers

Some contractile roots lack this collapse and death of cell

Fahn (1974) Putz et a / . (1995)

layers; all contractile roots depend on cortical cell growth, which increases the radial extent of the cells while decreasing their axial extent

(2) Physical

(a) Radiation absorptiotilscattering Increased absorption of photosynthetically active radiation via smaller package effect in photosynthetic cells (see (1))

Changed cell shape (see (1))

Raven (1987)

Increased absorption of UV-B radiation caused by the presence of UV-B-absorbing solutes in vacuoles (cannot protect DNA unless it is suspended in vacuole on cytoplasmic strands. as i n Mougeotia or is always on shaded side in a vector radiation field. as in leaves with UV-B absorbers in the epidermal vacuoles)

Presence of UV-B-absorbing material in extracellular matrix (cyanobacteria, Phaeocystis). Presence of UV-B-reflecting cuticle. (External UV-B barriers can protect all cell components)

Raven (1987. 1991. 1995~) Raven and Johnston (1992) Marchant et al. (1991) Garcia-Pichel and Castenholz (1991) Edwards et al. (1996)

Increased photon absorption via an increased pathlength of photons in photosynthetic celliorgan if vacuoles have a large difference in refractive index relative to cell wallskytoplasm (not proven)

Increased pathlength for photons in photosynthetic organ due to the presence of an intercellular pas space with a much lower refractive index than cell wall/cytoplasm/vacuole. or intercellular CaCOi

Ramus (1978) Raven (1987. 1996) Vogelman (1993)

Production o f colour attractive to pollinatorsipropagule disperser5 caused hy the presence of pigments in the vacuole (anthocyanins. betalains)

Colour production by lipophilic pigments in plastids (does not produce identical spectral range to the vacuolar pigments). Colour production by thin-film interference in cell walls (does not reduce photosynthesis underlying fruit tissue as much as does pigment)

Raven ( 1987) Lee (1991)

Increased absorption of adaxially incident photosynthetically active radiation by back-scattering from anthocyanin-containing vacuoles o f abaxial epidermal cells of leaves

Back-scattering via thin-film interference in cell walls of ahaxial epidermal cells

Lee et al. (1979. 1990) Lee ( 1986)

Focusing o f radiation related to phototaxidphotosynthesis by the difference between the refractive index of wallicytoplasmivacuole and medium

Focusing in non-vacuolate cells of given cytoplasmic volume involves different cytoplasmic volume/focal length relation. Solid lens in some single-celled eukaryotes (Dinophyceae). and metazoa focuses light on photoreceptor. e.g. retina in some metazoa

Vogelman (1993)



References

(b) Change in overall celllorganisrn density Decrease in density to value lower than that of fluid medium, hence positive buoyancy. via exclusion of Ca’+, Mg’+, ( K + ) . S042- from, emphasis on NH4+, Ht in, vacuoles when cells are isomolar with/hyperosmolar to medium (only works in high-osmolarity environments such as seawater). Occurs in both planophytic (sensu Luther, 1949) organisms, permitting upward movement in water column and halophytic (sensu Luther, 1949). related to posture

Positive buoyancy in high-osmolarity environments achieved by H+/NH4+, C1- in apoplasmic phases of some metazoa. Positive buoyancy (in high-osmolarity environments) by the use of lipid rather than polysaccharide as the energy store

Denton (1974) Dromgoole (1982) Lambert and Lambert (1978) Luther (1949) Villareal (1992) Villareal and Carpenter ( 1994) Villareal et al. (1993)

Positive buoyancy in high- or low-osmolarity environments achieved by prokaryote gas vesicles

Walsby (1994) Dromgoole (1982, 1990)

Positive buoyancy in high- or low-osmolanty environments achieved by extracellular gas spaces in multicellular aquatic planohapto- or rhizophytic plants (Enterornorpha, Codiurn, Dumontia. many brown algae in Fucales, Laminariales, Durvillaeales, Scytosiphonales; aquatic vascular plants) and in some metazoa (teleost swim bladder)

Walsby (1972) Raven (1996)

Flagellar (dyneidtubulin) or muscular (myosinlactin) motility permits organisms which are denser than the aqueous medium upwards in media of high or low osmolarity

Raven and Richardson (1984)

Hypothetical positive buoyancy in air of hydrogenic or methanogenic organisms

Raven (unpublished science fiction) Gibor (unpublished science fiction)

Increase in density relative to a highor low-osmolarity medium due to high-density solution andor solid components of vacuole

Dense components of cytoplasm (stored polysaccharide, protein, polyphosphate) or cell wall (Si02, CaC03).

Walsby (1972) Bauman et al. (1978)

Large surface area per unit volume of wind-dispersed seeddfruits; gas-filled dead cells derived from vacuolate cells with high C:N, C:P ratios

Effective premortem retranslocation of N , P to living part of the plant, i.e. dead “wings” etc. could have originated from non-vacuolate (low C:N, C:P) cells

(c) Graviperception by high-density vacuolar components moving (towards gravitational atractor) within cell (BaS04 crystals in characean rhizoids)

Graviperception by amyloplasts in vascular plants, CaS04 (etc.) crystals in animal cytosol

Raven (1995~) Clifford et al. (1989) Kiss (1994) Severs et al. (1989)

Hypothetical graviperception by movement of gas vesicles of prokaryotes (away from gravitational atractor) within cells

Raven (unpublished science fiction)



References

(3)Chemical (a) Storage of water, and watersoluble compounds (storage = deposition with stochastic or deterministic likelihood of withdrawal and further use). Strict limits on the extent to which cytoplasm can be diluted (water storage) or to which the concentration ( 5 a few moIm-3) of inorganic phosphate, total Ca’+, NH4+ or NO3- can vary in cytoplasm; any additional storage of these resources (and of K + , Mg2+, etc.) can occur in the vacuole. Low energy costs of storage of NO3- in vacuoles relative to storage in a reduced form. Vacuolar storage of “CO2” for -12 h as organic acid in the vacuole is the only method of “COf” storage used in CAM plants. Vacuolar storage of free Ca2+ at a higher electrochemical potential than that of free Ca2+ in the cytosol, thus permitting (downhill) efflux across the tonoplast to act as a means of signalling, may not be consistent with accumulation of large amounts of oxalate (see (b) below)

Apoplasmic water is stored in some saccate intertidal algae during emersion, and in the wood of trees. Nutrient solutes are apparently less frequently stored apoplasmically. Nutrient storage in desiccation-tolerant seeds, vegetative organs/organisms as polymers (C, energy as polysaccharides, lipids; N , C, energy as protein; P as polyphosphate, phytic acid). Storage of fuel for thermogenesis in aroid spadices as starch (and lipid) rather than soluble sugars: relatively non-vacuolate nature of spadix cells related to need for large heat production (a function of cytoplasmic (mitochondrial) volume) per unit volume if temperature is to be significantly above ambient. Extracellular mucilage of the colonial marine alga Phaeocysfis serves inter alia as a store of organic C and energy. Metazoan nutrient storage generally involves insoluble material. No known method of storing “CO,” for -12 h except as organic acids stored in the vacuole

Raven (1982, 1984a, 1987) Zhen et al. (1991, 1992) Raven and Spicer (1996) Welbaum et al. (1993) Canny and Huang (1993) Trebacz et al. (1995) Lee and Ratcliffe (1983) Mimura (1995) Mimura et al. (1990) Rebeille et al. (1983) BermadingerStabentheiner and Stabentheiner (1995) Lancelot and Rousseau (1994) Sanders et al. (1992) Kinzel (1989) Leigh and Wyn Jones (1984)

(b) Accumulation of defence materials (nuclear, chemical, biological) and optical attractants, UV-B absorbers, photosynthetically active radiation in back-scatterers (accumulation = synthesis and sequestration without (generally) remetabolism). Defence compounds includes free radical scavengers (active against ionizing and UV-B radiation, chemicals) and anti-biophage agents. Also include solutes which offset the effects of (stable) material necessarily produced in metabolism, e.g. OH- from NO3assimilation neutralized by organic acid production with a salt of the organic anion accumulated in the vacuole (generally soluble, sometimes precipitated, e .g. Ca( COO)2). Vacuoles with total oxalate in excess of total Ca2+ may have free Ca2+ concentrations too low to allow downhill Ca2+ efflux to the cytosol to participate in signalling. Also in this category are salts moved to the shoot in emergent halophytes; one fate for these is vacuolar accumulation

Alternatives to vacuolar storage of solutes interacting with (non-ionizing) radiation were considered in 2(a) above. Free radical scavengers occur mainly in cytoplasm where free radicals are (metabolically) produced and there is a higher concentration of damageable components. OHneutralization can include apoplasmic C a C 0 3 precipitation (could not occur in vacuoles of “normal” pH) and (like H + disposal) direct or indirect efflux to the environment. Emergent halophytes can exclude un-needed salts from their roots, or excrete them via salt glands or (?) in abscised leaves

Raven (1987, 1995c) Osmond ei al. (1980) Kinzel (1989)

TABLE I (continued) Vacuolar functions and alternative ways of pegorming these functions Function performed by vacuole


References

(c) Non-enzymic catalysis by H + of reactions involving inorganic and organic compounds. Hydrolysis of sucrose in Citrus juice vesicles is initially enzymic, later (as vacuolar pH decreases to pH 2.5) the enzyme activities disappear and all hydrolysis can be accounted for by H + catalysis

Acid invertase in vacuoles with a higher pH than that of Citrus juice vesicles, or enzymes in other cell compartments (including acid invertase in low-pH cell walls)

Echeverria (1990)

Possible occurrence of H+-catalysed, H+-consuming conversion of HC03to C 0 2 in vacuoles of photosynthetic cells provided vacuole is large enough or has a low enough pH relative to rate of C 0 2 consumption in photosynthesis. Involves transport of exogenous HC03- across plasmalemma and tonoplast, with C 0 2 transport from vacuole to plastids (and some leakage to medium!)

Extracellular H+-catalysed conversion of HC03- to C 0 2 in acid zones on the surface of certain freshwater macrophytes, with C 0 2 entry and fixation. Extracellular catalysis of HC03- to C 0 2 conversion by carbonic anhydrase with COz entry and fixation. HC03- transport into cell with intracellular conversion to C 0 2 and then fixation. All of these mechanisms ‘‘leak’’ C 0 2

Raven (1997)

TABLE I1 Vacuolar costs, and the costs of performing vacuolar functions by other meuns Costs of vacuolation

(1) Synthesis of vacuoles and of concomitant additional quantities of other cell components. Costs of making the additional tonoplast and plasmalemma membrane, and of filling the vacuole with inorganic salts, is only -1% of the cost of synthesizing the rest of the cell for a vacuo1e:cytoplasm volume ratio of 10. Laplace’s principle requires that the turgor-resisting wall has not only a greater area, but also a greater thickness. in the vacuolate cell with the same volume of cytoplasm as a non-vacuolate cell. The additional wall cost may double the overall cost of cell synthesis per unit cytoplasm The diversion of resources to synthesis of vacuoles and the attendant additional plasmalemma and cell wall may reduce the maximum specific growth rate (p.,,,) relative to that of a non-vacuolate cell with the same cytoplasmic volume per cell; however, this has not thus far been demonstrated

Costs of alternatives to vacuolation Change of cell shape as an alternative (and addition) to vacuolation as a means of increasing surface area per unit cytoplasmic volume involves (as does vacuolation per se) a need for more wall synthesis. Polymer synthesis (for storage in cytoplasm) may cost more (per unit energy, N or P stored) than transport into the vacuole of the monomers of this polymer synthesis. However, the cytoplasmic storage structures (if any) may have lower synthetic costs (energy, C, N, P) than vacuoles and (by not increasing cell volume as much as vacuoles storing the same quantity of material) lower resource costs of synthesis of walls (where present)

References Raven (1987) Bouma et al. (1994)

TABLE I1 (continued) Vacuolar costs, and the costs of performing vacuolar functions by other means Costs of vacuolation

Costs of alternatives to vacuolation

References

(2) Maintenance of vacuoles and concomitant additional quantities of cell components. The larger area of tonoplast (hypertrophied lysosomes) and plasmalemma membranes increases solute leakage intolfrom the cytosol; solute homoiostasis aspects of maintenance cost four times as much in a vacuolate cell (vacuo1e:cytoplasm ratio of 10) as a non-vacuolate cell with the same volume of cytoplasm and a similar shape

Maintenance of alternatives to vacuolation (e.g. pigmented chromoplasts; stored polymers; accumulated CaC03) individually involves less energy expenditure in solute homoiostasis than does a vacuole. This is probably also true of the alternatives to vacuolation in toto, unless there are extreme changes in shape in increasing surface area per unit cytoplasmic volume as an alternative to vacuolation of an isodiametric cell increasing costs of solute homoiostasis

Raven (1987)

( 3 ) Response to rapid changes in external osmolarity. Giant-celled algae with vacuo1e:cytoplasm volume ratio, of 20 or more require very large transtonoplast and trans-plasmalemma ion fluxes in these cells with low surface area per unit volume to adjust turgor after rapid changes in external osmolarity

No direct analogues in multicellular algae (or coenocytic algae with interwoven, small-diameter hyphae, e.g. Codium) with a hollow (saccate) thallus; essentially no turgor in the internal aqueous phase in these algae with similar fresh/dry weight ratio (including water in sac) to giant algal cells

Raven (1987) Guggino and Gutknecht (1982)

Unknown costs of tolerance of bursting/resealing of certain giant-celled algae when exposed to rapid hyposmotic shocks

(4) High vacuo1e:cytoplasm volume ratios seem to be incompatible with desiccation tolerance. Further work is needed here. Certainly a (relative) absence of vacuolation does not necessarily confer desiccation tolerance

There seems to be n o constraint on desiccation tolerance resulting from the various alternatives to vacuolation. However, there is a height limit of about 2 m (which may be exceeded in the Vreeziaceae) for desiccation-tolerant vascular plants, perhaps related to problems of refilling long-distance transport conduits, including the apoplasmic xylem. Absence of vacuolation may restrict length of conduit initials, thus imposing more resistance in long-distance transport pathways (more frequent cross-walls in phloem, xylem tracheids, or remains of cross-walls in xylem vessel elements)

Raven (1987) Kinchin (1994) Cavalier-Smith (1978)

( 5 ) Movement by dynein/tubulin (flagella, cilia), myosinkactin (muscles. amoeboid movement) or other mechanochemical transducers at a given velocity has a higher energy cost in a vacuolate than a non-vacuolate organism with the same kolume o f cytoplasm (more mass (corrected for density relative to medium) to move; greater surface area for frictional interaction with the medium)

A similar volume of apoplasmic water per unit cytoplasm

Raven and Richardson ( 1984)

(6) Alleged problems with cell division in highly vacuolate cells. Meristematic cells usually non-vacuolate; but many algal cells can divide despite being high vacuolate (e.g. Cludophoru)

Many alternatives to vacuolation relate to differentiated tissues which would normally have further cell division

(blood, haemocoel, coelom, coelenteron) would have a similar energetic cost for movement to that of vacuoles

Fahn (1974)

74

J. A . RAVEN

Jannasch (1995) suggested that Beggiatoa does not execute gliding movements between the more oxidized and more reduced ends of the chemocline, and that it relies on natural vertical movements of the chemocline to alternatively supply oxidant (02, N 0 3 - ) and reductant (S2-). Oxidant (NO3-) can be stored in the vacuole during an oxidizing episode, and (possibly) S2- and So can be stored in a reducing episode. This permits the cells to have a better supply of (internal) oxidant during reducing episodes and vice versa. Thioplaca exhibits gliding motility, and it has been suggested to migrate vertically across the chemocline (Fassing et al., 1995; Huetel et al., 1996; Schulz et al., 1996). At the oxidized surface of the sediment the NO3concentration is -25 mmol mP3. Thioplaca accumulates NO3- 20 000-fold to 500 mol mP3 in its “vacuole”, and then migrates downwards to the S2--rich reduced zone where S2- is oxidized, using NO3- as an electron acceptor, to So, then S042-, and N 2 0 and N2, respectively. Whether the cells migrate upwards with “vacuoles” full of So or S2- is not clear from Fassing et al. (1995); the retention of S2- is unlikely except at unrealistically high intracellular pH values in view of the probable high permeability of H2S. At all events, the downward transport of NO3- in motile filaments can move oxidant down to the S2- much faster than can diffusion. Nitrate is a less energetically desirable oxidant than is 02,but is easier to transport since it can be accumulated in vacuoles whence its leakage is slow because of its low permeability coefficient in the lipid parts of membranes (Raven, 1984a). Oxygen could be transported using a haemoglobin- or myoglobin-like carrier, but this would be a “leaky” transporter into anoxic regions as would the corresponding transport of S2-/HS-/H2S associated with haemoglobin, as in the pogonophoron Riftia (Childress et al., 1991). McHatton et al. (1996) showed similar NO3- accumulation in (motile) Beggiatoa, and suggested that it behaved as did Thioplaca (cf. Jannasch, 1995). At the moment the energetics of NO3- accumulation cannot be accurately estimated from a mechanistic point of view, although an estimate of the minimum thermodynarnic energy cost of NO3- accumulation is possible. This estimate is some 26 kJ (mol NO3-)-’. Minimum energetic costs of gliding mobility are increased by vacuolation due to the larger surface area per unit cytoplasm. Furthermore, the costs of vacuolation must be added to find the total costs of vacuolation including the related transport (of cells and of molecules) phenomena. The benefits of the transport can, of course, be expressed in the same units (energy conversion rate per unit of cytoplasm volume); in this case the computation involves the energy conversion rate based on diffusion of NO3- (and 0,) into the sediment and that based on the “active transport” of NO3- down from the sediment surface to the S2--rich zone. Alas, the information from which the latter “benefit” can be quantified is not readily come by.

THE VACUOLE: A COST-BENEFIT ANALYSIS

75

V. ANOTHER CASE HISTORY: VACUOLES AND BUOYANCY Table I indicates that manipulation of the content of solutes in the “true” vacuole can regulate buoyancy. Positive buoyancy can only be achieved in this way in aqueous media of relatively high density and thus relatively high osmolarity. Positive or negative buoyancy can only be of use in controlling the position of planophytic (sensu Luther, 1949) aquatic organisms relative to the water surface if the vertical motion imparted by the buoyancy exceeds the vertical component of bulk water movement. Vacuole-related buoyancy can be achieved by manipulation of the ionic composition of the vacuole such that “heavy” ions (Ca2+, Mg2+, SO4*-, (K+)) are diminished while “light” ions (Na’, NH4+, H + ) are favoured. Granted the availability of the ions in the (seawater) medium, the production of a “buoyant” vacuolar sap need not cost any more than that of a “non-buoyant” sap. This is true at least in mechanistic terms with 1 mol of ATP (or its equivalent) used per mole of ions transported across a plasma membrane and 0.5 mol of ATP per mole of ions transported across the tonoplast: the highest ionic gradients required could be maintained by these ion-to-ATP stoichiometries (Raven, 1984a). However, although the manipulation of vacuolar ions can be considered “costless” granted the occurrence of the vacuole, the constraints on the ionic content of the vacuole related to buoyancy could restrict the use of the vacuole for storage of dense solutes (e.g. NO3- rather than NH4+). Furthermore, the vacuole-tocytoplasm ratio must be high for this mechanism to work, since the cytoplasm is denser than seawater; this argument applies a fortiori to the required ratio of vacuolar volume to that of dense, silicified walls of diatoms. The constraint on the nature of vacuolar storage materials is, to some extent, offset by the large volume of vacuole relative to cytoplasm needed to obtain buoyancy, although there are still constraints on the content of “dense” stored solutes. The high vacuole-to-cytoplasm ratio needed to obtain buoyancy would require a higher wall-to-cytoplasm ratio, which might offset in part the buoyancy effect. Data are available for the vacuolar fraction and wall fraction as a function of cell size in diatoms (Raven, 1987, 1995a). An alternative to favouring “light” ions in a “normal” vacuole in reducing density is the use of lipid (low density) rather than polysaccharide (high density) as the organic carbon and energy storage material. This option of lipid is, to varying extents, seen in diatoms in parallel with “light” ion accumulation in the vacuole. Another component which reduces overall cell density is the gas vesicle (Walsby, 1994). This mechanism is confined to certain prokaryotes, and has not been found so far in those with true (aqueous) “vacuoles” (Walsby, 1994; Fassing et al., 1995). The volume of gas vesicles needed to give a certain degree of buoyancy is, of course, much smaller than that of an aqueous vacuole since the gas has a density

76

J . A. RAVEN

that of the average solid plus fluid cell contents, although the need to have pressure-resistant (proteinaceous) walls for gas vesicles involves a very substantial energy cost (Walsby, 1994). However, the small fraction of cell volume occupied by gas vesicles means that a given volume of cytoplasm has its surface area increased less when gas vesicles cause a given reduction in density than is the case for aqueous vacuoles with ‘‘light’’ ions. This in turn means a more rapid upward motion for the gas-vesicle-containing cells, as shown by the application of hydrodynamic principles, for a given difference in density between the cell and its medium. A rather different means of adjusting the vertical position of cells in a non-turbulently mixed water body is that of flagellar motility (Raven, 1982; Raven and Richardson, 1984). Walsby (1994) has shown that the energetic and nitrogen cost of the construction of gas vesicles (in cyanobacteria) greatly exceeds that of the flagellar apparatus (of eukaryotic algae) and that inclusion of running (energy) costs of operating the flagellar apparatus does not offset this difference in energetic constructional costs for any generation time attainable with balanced growth (20 times the minimum generation time) (Raven, 1986; Geider et al., 1985). Furthermore, the flagellar mechanism offers a more immediate (seconds) regulation of the direction of vertical movement. Adding ballast (carbohydrate as polysaccharide) to gas-vesiculate cyanobacteria to an extent which reverses the sign of the density difference between cells and medium takes tens of minutes or hours of net photosynthesis, as does accumulation of K+ salts (Walsby, 1994). It is likely that similar temporal considerations apply to alteration of buoyancy in algae with ‘‘light’’ions in their vacuole, granted the surface area per unit volume in these diatoms and the area-based net ion fluxes across the tonoplast and plasmalemma (Raven, 1984a, 1988). The more temporally flexible flagellar mechanism is not available to the cyanobacteria (with the exception of one marine strain, Synechococciis: Waterbury et al., 1985), with some means of swimming not associated with “normal” structures of bacterial flagella) o r to the walled (silicified) vegetative cells of diatoms (flagella only occur on the wall-less male gametes of centric diatoms). The discussion thus far of buoyancy regulation/flagellar motility has concentrated on the costs of the various “cell-positioning” mechanisms. What of the benefits? Raven and Richardson (1984) have performed a cost-benefit analysis of vertical diel migration by dinoflagellates growing in (normal) die1 light-dark cycles in stratified water bodies with a greater supply of nutrients (e.g. inorganic nitrogen and phosphorus) and depth from chemoorganotrophy than at the surface dominated by photolithotrophy. The strategy here is upward migration around dawn and downward migration around dusk, thus optimizing the acquisition of the “co-limiting” resources light (only available near the surface in the photophase) and the inorganic nutrients nitrogen, phosphorus, iron, etc. (available over the whole diel cycle but scarce near the surface).


77

The behaviour of certain of the planktonic diatoms which can regulate their density via control of vacuolar composition fits a similar paradigm (Villareal, 1992; Villareal et al., 1993; Villareal and Carpenter, 1994). However, the periodicity of the cyclic vertical migration is more than 24 h ; Villareal and Carpenter (1994) suggest 7-12 days for Erhmodiscus rex (cell volume m3). The “limiting” inorganic nutrient in the habitats investigated by Villareal and co-workers (central Pacific gyre) is nitrogen, as shown by the cellular C:N:P ratio (Villareal and Carpenter, 1994). The diatoms have more N03- in their cell (vacuolar) sap whcn ascending than when desccnding; for Rhizosolenia sp. (cell volume 10-“’m3) the concentrations are 9.7 k 2.9 mol m-3 when rising and 2.0 k 2.3 mol m - 3 when descending (Villareal et al., 1993). Natural-abundance “N/I4N studies show that Rhizosolenia obtains much of its nitrogen from subnutricline NO3- with a higher lsN/I4N than surface-water combined nitrogen (Villareal et al., 1993). The large size of the vacuole relative to the cytoplasm means that relatively modest NO3- concentrations (up to 27 mol m-3 in E. rex) can contribute up to 54% of the total (inorganic plus organic) cellular nitrogen quota (Villareal and Carpenter, 1994). The carriage of nitrogen from the nutricline to the surface waters as NO3 by ascending cells of these large diatoms illustrates the potential conflict between the storage function of the vacuole and its role in decreasing overall cell density. Nitrate is a “heavy” ion, and the buoyancy function would be better served by NO3- reduction t o yield the “light” ion NH4+ at the nutricline prior to or during ascent (Fig. 1 ) . However, energetic considerations may militate against this energetically costly reduction in a low-light environment. This appears to explain why the benthic marine brown macroalga Laminaria accumulates the NO3- available in winter (light energy available at the depth limit of its occurrence) as NO3- in its vacuoles, with reduction and assimilation later in the year (Raven, 1987). It may also explain why the freshwater red macroalga Lemanea, whose main growth occurs in winter with limited light availability, uses exogenous NH4+ as its nitrogen source and eschews the more abundant (at least in agriculturally influenced streams) NO3- (Raven, 1987; MacFarlane and Raven, 1990). At all events the absolute concentration of “heavy” NO3- is only a small fraction of the total ionic content of these large-celled diatoms, so that variations in the content of other ions can contribute to the conversion o f an NO3- -rich “floater” into an NO3-- poor “sinker”, possibly with contributions from the fraction of organic carbon and energy stored as polysaccharide rather than lipid. In addition to these and similar large diatoms, the “vertical migrations with NO3- transport” paradigm may well also apply to the large-celled, vacuolate phycoma stage of the green (prasinophyte) alga Halosphaera and to the large-celled, vacuolate but non-flagellate (in the vegetative state) dinoflagellates such as Pyrocystis (Villareal and Carpenter, 1994). Lest these migrations of large-celled vacuolate planktonic algae seem

78 1.050

-

1.045 -

1.040

-

1.035 -

3

E 0)

F 3

.-fn

$ n

J . A. RAVEN

-.-

NH,CI

-v-

NaCl KCI NaNO,

-4-

KNO,

-0-

-A-

1.030 1.025

-

1.0201.0151.010 1.005 7 0

100

200

300

400

500

600

700

Concentration/mol m-3 Fig. 1. Density of solutions of potential vacuolar solutes of large-celled diatoms as a function of concentration. (From tabulated data in Washburn (1928).)

remote from the preoccupations of most vacuoleers, a few comments on their global significance are in order. Planktonic algae in the ocean transform significantly more nitrogen each year than d o terrestrial (including cultivated) plants (Raven et al., 1993). Of this combined nitrogen taken up by marine planktonic algae, some 20% is as NO3- recycled from the ocean depths (Raven et al., 1993). While much of this moves upwards as macroscopic upwellings and microscopic eddy diffusion, there could be a very important role in parts of the open ocean for the N03--transporting vertical migrations based on regulation of vacuolar ionic composition; at least 1% of the oceanic nitrogen assimilation (i.e. more than 6Tmol N per year) could be moved upward by this mechanism.

VI. COSTS AND BENEFITS OF VACUOLATION: SIMPLE ANALYSES AND THE ALLOCATION OF COSTS AMONG VARIOUS BENEFITS The geometrically unavoidable consequence of vacuolation without a change in cell shape is an increase in the surface area per unit volume of cytoplasm (see Table I), which can increase the rate of resource acquisition, at least

800


79

TABLE 111 Costs and benefits of vacuolation for aquatic unicells growing under resource-saturating conditions, for a non-vacuolate cell 0.5 or 5 p m in radius and a vacuolate cell 1.11 or 11.1 p m in radius with a vacuo1e:cytoplasm volume ratio of 10 and the same cytoplasmic volume as for the 0.5 or 5 p m radius cell, respectively (from Tables 4-7 of Raven, 1987) Resource being acquired and used

Benefit (additional resource acquisition) of vacuolation

Cost (additional resource cost of cell synthesis) of vacuolation

Energy

Nil

U p to twice as much energy needed for cell synthesis per unit of cytoplasm

Carbon

Nil (except in the unlikely case of limitation of C acquisition by area of plasmalemma to house transporters, in which case the rate may be fivefold higher)

Up to twice as much energy needed for cell synthesis per unit of cytoplasm

Nitrogen

Nil (except in the unlikely case of limitation of N acquisition by area of plasmalemma to house transporters, in which case rate may be fivefold higher)

U p to 40% more N needed for cell synthesis per unit of cytoplasm

Phosphorus

Nil (except in the unlikely case of limitation of P acquisition by area of plasmalemma to house transporters, in which case rate may be fivefold higher)

U p to 10% more P needed for cell synthesis per unit of cytoplasm

under resource-limiting conditions (Tables 111 and IV). The increase in the rate of resource acquisition is especially significant for photons, but can also apply to chemical resources (Table IV). These benefits are paralleled by the resource costs (see Table 11) of cell synthesis. Under resource-saturating conditions the resource costs of vacuolation outweigh the resource-acquisition benefits (see Table 111); the reverse can be the case under resourcelimiting conditions, most generally in the case of photon absorptiodenergy costs (see Table IV). This simple analysis deals in processes which all organisms carry out, i.e. resource acquisition and growth. However, Tables I and I1 show many more costs and benefits of vacuolation which generally (an exception is storage of resources) are not universal. Thus, the increased energy cost of motility of a given volume of cytoplasm at a given velocity as a result of vacuolation

TABLE IV Costs and benefits of vacuolation for aquatic unicells growing under resource-limiting conditions, for a non-vacuolate cell 0.5 or 5 p m in radius and a vacuolate cell 1.11 or 11.1 p m in radius with a vacuo1e:cytoplasm volume ratio of 10 and the same cytoplasmic volume as the 0.5 or 5 p m radius cell, respectively (from Fig. 2 and Table 2 of Raven, 1987)

Benefit of vacuolation for cell of radius Resource limiting growth

0.5 um (non-vacuolateY 1.i1pA (vacuo~ate)’

5 urn .~ (non-vacuolateM 11.1 pm (vacuolate)

Cost of vacuolation for cell of radius

0.5 um (non-vacuolateY . \ 1.11p m (vacuolate) ‘

5 urn .~ (non-vacuolateY 11.1 p m (vacuolate)

Energy (assuming 20mol of chromophore per 1 m3 of cytoplasmic volume)

Photon absorption 1.15 times that of non-vacuolate cell

Photon absorption 2.33 times that of non-vacuolate cell

Up to twice as much energy needed for cell synthesis per unit cytoplasm

Up to twice as much energy needed for cell synthesis per unit cytoplasm

Carbon

Zero for extracellular diffusion of solutes bearing C, N or P

Very small for extracellular diffusion of solutes bearing C, N or P

Up to twice as much C needed for cell synthesis per unit of cytop 1asm

Up to twice as much C needed for cell synthesis per unit of cytoplasm

Nitrogen

Up to five times as much C, N or P entry if influx is limited by plasmalemma area available for transporters

Up to five times as much C, N or P entry if influx is limited by plasmalemma area available for transporters

Up to 40% more N needed for cell synthesis per unit of cytoplasm

Up to 40% more N needed for cell synthesis per unit of cytoplasm

Up to 10% more P needed for cell synthesis per unit of cytoplasm

Up to 10% more P needed for cell synthesis per unit of cytoplasm

Phosphorus


81

(see Table 2) is an exception among vacuole organisms, as is the benefit of increased visual attraction as a result of vacuolar pigments (see Table I). Quantitative analysis of the allocation of costs (see Table 11) among numerous benefits (see Table I), i.e. multiobjective optimization (Farnsworth and Niklas, 1995) is potentially possible, but the multiplicative effects of the increasing number of assumptions which must be made as more costs and benefits are included in the analysis limit the extent to which such analyses can be used for the multitude of costs and benefits in Tables I and 11. Eventually, with more quantitative knowledge, such multiobjective optimization analyses will become more realistic, and will contribute to the ultimate goal of relating vacuolation to include fitness (Osmond et af., 1990).

VII. EVOLUTIONARY ASPECTS In view of the lack of multiobjective optimization analyses noted above, it may be considered premature to consider evolutionary aspects of vacuolar costs and benefits. However, I wish to make two general points, one involving environmental variability over geological time, and the other to nonvacuolate, invertebrate-based phototrophic symbioses. The environmental variable considered is atmospheric C 0 2 . The first terrestrial embryophytes (bryophytes, tracheophytes) some 420 million years ago were confronted with C 0 2 partial pressures 10-20 times the present value of 36Pa (Raven, 1995b). Carbon dioxide partial pressures higher than the present value have predominated over the intervening 420 million years, with values probably as low as the extant C 0 2 level in the Upper Carboniferous, and even lower values in the interglacial and, especially, glacial episodes in the last million or so years (Raven, 1995b). Today there are advantages in vacuolation in homoiohydric plants (i.e. those which have cuticle, stomata and intercellular gas spaces, and thus can remain hydrated for some time in the absence of an adequate water supply to the roots and/or with a very large evaporative demand from the atmosphere) in terms of minimizing the resistance to C 0 2 diffusion in the aqueous phase (Raven, 1993b, and references therein). These advantages would have been greater at the last glacial maximum (18 000 years ago) with C 0 2 partial pressure at 18-20 Pa, but much less in the high-C02 environments of the Early and Mid Palaeozoic and the Mesozoic and most of the Tertiary. However, the light-absorption advantages of vacuolation would still have been present (see Table 1). At least as far as photosynthetic structures are concerned there might have been a smaller selection pressure for large vacuoles in the more distant past (except for the Upper Carboniferous) but larger selection pressure in the more recent past. Quantitative assessment of these suggestions is difficult except in terms of overall cell size: vacuoles do not fossilize well. The phototrophic symbioses between invertebrates and microalgae are

82

J. A. RAVEN

non-vacuolate, yet they coexist with (vacuolate) macroalgae and higher plants (see Table I, and Raven, 1993a). Corals can achieve a large surface area per unit cytoplasmic volume by non-spherical shapes and the presence of a coelenteron; this is less readily achieved in the giant clams. These animalinvertebrate symbioses will form useful material for future evolutionary cost-benefit analyses of vacuolation.

VIII.

CONCLUSIONS

At the risk of being too Panglossian (Voltaire, 1759), it would be possible to conclude that the semiquantitative analyses currently possible show that the evolutionary costs of vacuoles are outweighed by their evolutionary benefits. However, much more quantitative analysis is needed to test the suggestion that we have currently identified all of the costs and benefits of vacuolation and their quantitative importance.

ACKNOWLEDGEMENTS Past and present colleagues have catalysed and refined my thoughts on the costs and benefits of vacuolation. Experimental work on the role of vacuoles in acid-base regulation of higher plants and of cell size in resource acquisition and storage by algae has been funded by the AFRC/BBSRC and NERC.

REFERENCES Bauman, F. G., Idenberg, H. D. and Gennaro, J., Jr (1978). The inverse relationship between nutrient N concentration and coccolith calcification in cultures of the cocclithophorid Hymenomonas sp. Journal of Protozoology 25, 253-256. Bermadinger-Stabentheiner, E. and Stabentheiner, A. (1995). Dynamics of thermogenesis and structure of epidermal tissues in inflorescences of Arum maculatum. New Phytologist 131, 41-50. Bouma, T. J., Visser, R. De, Janssen, J. M. U. A,, Kock, M. J. De, Leeawer, P. H. and Lambers, H. (1994). Respiratory energy requirements and rate of protein turnover in vivo determined by the use of an inhibitor of protein synthesis and a probe to assess its effect. Physiologia Plantarum 92, 585594. Canny, M. J. and Huang, C. X. (1993). What is in the intercellular spaces of roots? Evidence from the cryo-analytical scanning electron microscope. Physiologia Plantarum 87, 1-8. Cavalier-Smith, T. (1978). Nuclear volume control by nucleoskeletal DNA, selection for cell volume and cell growth rate, and the solution of the DNA C-value paradox. Journal of Cell Science 34, 247-278. Childress, J. J., Fisher, C. R., Faruzzi, J. A,, Kochevar, R. E., Sanders, N.K. and Alayse, A. M. (1991). Sulfide-driven autotrophic balance in the bacterial symbiont-containing hydrothermal vent tubeworm, Riftia pachyptila Jones. Biological Bulletin 180, 135-153.


83

Clifford, P. E., Douglas, S. and McCartney, G. W. (1989). Amyloplast sedimentation in shoot statocytes having a large, central vacuole: further interpretations from electron microscopy. Journal of Experimental Botany 40. 1341-1346. Denton, E. J . (1974). On the buoyancy and the lives of modern and fossil cephalopods. Proceedings of the Royal Society of London B 185, 273-299. Dromgoole, F. I. (1982). The buoyant properties of Codium. Botanica Marina 25, 391-397. Dromgoole, F. I . (1990). Gas-filled structures, buoyancy and support in marine macro-algae. Progress in Phycological Research 7 , 169-21 1. Echeverria, E. (1990). Developmental transmission from enzymatic to acid hydrolysis of sucrose in acid limes (Citrus aurantifofia). Plant Physiology 92, 168-171. Edwards, D., Abbott, G. D. and Raven, J . A. (1996). Cuticles in early land plants: a palaeoecological evaluation. In “Plant Cuticles” (G. Kierstens, ed.), pp. 1-31. Bios Scientific Publications, Oxford. Fahn, A. (1974). “Plant Anatomy” 2nd edn. Pergamon Press, Oxford. Farnsworth, K. D. and Niklas, K. J . (1995). Theories of optimization, form and function in branching architecture in plants. Functional Ecology 9, 355-363. Fassing, H., Gallardo, H., Canfield, D. E., Forster, S . , Glud, R. N., Gunderson, J . K., Kuver, J., Ramsing, N. B., Teske, A , , Thamdrup, B. and Ulloa, 0. (1995). Concentration and transport of nitrate by mat-forming sulphur bacterium Thioplaca. Nature 374, 713-715. Garcia-Pichel, F. and Castenholz, R. W. (1991). Characterization and biological implications of scytonemin, a cyanobacterial sheath pigment. Journal of Phycology 27, 395409. Geider, R. J., Osborne, B. A. and Raven, J . A. (1985). Light effects on growth and photosynthesis of Phoeodactylum tricornutum. Journal of Phycology 21, 609-619. Grime, J . P. (1979). “Plant Strategies and Vegetation Processes.” Wiley, Chichester. Guggino, S. and Gutknecht, J. (1982). Turgor regulation following acute osmotic shock. Journal of Membrane Biology 67, 155-164. Huetel, M., Forster, S . , Kloser, S. and Fossing, H. (1996). Vertical migration in the sediment-dwelling sulfur bacterium Thioplaca spp. in overcoming diffusion limitations. Applied and Environmental Microbiology 62, 1863-1872. Jannasch, H. W. (1995). Life at the sea floor. Nature 374, 676-677. Kinchin, I . M. (1994). “The Biology of Tardigrades.” Portland Press, London. Kinzel, H. (1989). Calcium in the vacuoles and cell walls of plant tissue. Flora 182, 99-1 25. Kiss, J . Z . (1994). The response to gravity is correlated with the number of statoliths in Chara rhizoids. Plant Physiology 105, 937-940. Lambert, C. C. and Lambert, G. (1978). Tunicate eggs utilize ammonium ions for flotation. Science 200, 64-65. Lancelot, C. and Rousseau, V. (1994). Ecology of Phaeocystis: the key role of colony forms. In “The Haptophyte Algae” (J. C. Green and B. S. C. Leadbeater, eds), pp. 229-245. Clarendon Press, Oxford. Lee, D. W. (1986). Unusual strategies of light absorption in rainforest herbs. In “On the Economy of Plant Form and Function” (T. J. Givnish, ed.), pp. 105-131. Cambridge University Press, Cambridge. Lee, D. W. (1991). Ultrastructural basis and function of iridescent blue colour of fruits of Elaeocarpus. Nature 349, 260-262. Lee, D. W . , Lowry, J. B. and Bone, B. C. (1979). Abaxial anthocyanin layer in leaves of tropical rain forest plants: enhancer of light capture in deep shade. Biotropica 11, 70-77.

84

J . A. RAVEN

Lee, D. W., Bone, R. A., Tarsis, S. C. and Storch, D. (1990). Correlates of leaf optical properties in tropical forest of sun and extreme shade plants. American Journal of Botany 7 7 , 370-380. Lee, R. B. and Ratcliffe, R. G. (1983). Phosphorus nutrition and the intracellular distribution of inorganic phosphate in pea root tips: a quantitative study using 31P nmr. Journal of Experimental Botany 34, 1222-1244. Leigh, R. A. and Wyn Jones, G. (1984). A hypothesis relating critical potassium concentrations for growth to the distribution and functions of this ion in the plant cell. New Phytologist 97, 1-14. Liebig, J. (1840). “Organic Chemistry in its Application to Agriculture and Physiology.” Taylor and Walton, London. Luther, H. (1949). Vorschlag zu einer Okologischen Grundeinteilung der Hydrophyten. Acta Botanica Fennica 44, 1-15. MacFarlane, J. J. and Raven, J. A. (1990). C, N and P nutrition of Lemanea mamillosa Kiitz. (Batrachospermales, Rhodophyta) in the Dighty Burn, Angus, UK. Plant, Cell and Environment 13, 1-13. Marchant, H. J . , Davidson, A. T. and Kelly, G. J . (1991). UV-B protecting compounds in the marine algae Phaeocystis pouchettii from Antarctica. Marine Biology 109, 391-395. McHatton, S. C., Barry, J. P., Jannasch, H. W. and Nelson, D. C. (1996). High nitrate concentrations in vacuolate, autotrophic Beggiatoa spp. Applied and Environmental Biology 62, 954-958. Mimura, T. (1995). Homeostasis and transport of inorganic phosphate in plants. Plant and Cell Physiology 36, 1-7. Mimura, T., Dietz, K.-J., Kaiser, W., Schramm, M. J., Kaiser, G. and Heber, V. (1990). Phosphate transport across biomembranes and cytosolic phosphate homeostasis in barley leaves. Planta 180, 139-146. Mitchell, P. (1979). Direct chemiosmotic ligard conduction mechanisms in protonmotive complexes. In “Membrane Bioenergetics” (C. P. Lee, G. Schotz and L. Ernster, eds), pp. 361-372. Addison-Wesley, Reading, Mass. Nicklas, K. J. (1992). “Plant Biomechanics. An Engineering Approach to Plant Form and Function.” University of Chicago Press, Chicago. Osmond, C. B., Bjorkman, 0. and Anderson, D. J. (1980). “Physiological Processes in Plant Ecology: Toward a Synthesis with Atriplex.” Springer-Verlag, Berlin. Putz, N., Hiining, G . and Froebe, H.A. (1995). Cost and advantage of soil channel formation by contractile roots in successful plant movement. Annals of Botany 75, 633-639. Ramus, J. (1978). Seaweed anatomy and photosynthetic performance: the ecological significance of light guides, heterogeneous absorption and multiple scatter. Journal of Phycology 14, 352-362. Raunkaier, C. (1934). “The Life Forms of Plants and Statistical Plant Geography.” Clarendon Press, Oxford. Raven, J. A. (1981). Nutritional strategies of submerged benthic plants: the acquisition of C, N and P by rhizophytes and haptophytes. New Phytologist 88, 1-30. Raven, J. A. (1982). The energetics of freshwater algae: energy requirements for biosynthesis and volume regulation. New Phytologist 92, 1-20. Raven, J. A. (1984a). “Energetics and Transport in Aquatic Plants.” Liss, New York. Raven, J. A. (1984b). A cost-benefit analysis of photon absorption by photosynthetic unicells. New Phytologist 98, 593-625.


85

Raven, J. A. (1985). Regulation of pH and generation of osmolarity in vacuolar land plants: costs and benefits in relation to efficiency of use of water, energy and nitrogen. New Phytologist 101, 25-77, Raven, J. A. (1986). Evolution of plant life forms. I n “On the Economy of Plant Form and Function” (T. J. Givnish, ed.), pp. 421-492. Cambridge University Press, Cambridge. Raven, J. A. (1987). The role of vacuoles. New Phytologisr 106, 357-422. Raven, J . A. (1988). Algae. In “Solute Transport in Plant Cells and Tissues” (D. A. Baker and J. L. Hall, eds), pp. 166-219. Longman Scientific and Technical. Harlow. Raven, J . A. (1991). Responses of aquatic photosynthetic organisms to increased solar UV-B. Journal of Photochemistry and Photobiology B : Biology 9, 239-244. Raven, J . A. (1993a). Energy and nutrient acquisition by autotrophic symbioses and their asymbiotic ancestors. Symbiosis 14, 33-60. Raven, J. A. (1993b). The evolution of vascular plants in relation to quantitative function of dead water-conducting cells and of stomata. Biological Reviews 68, 337-363. Raven, J. A. (199%). Scaling the seas. Plant, Cell and Environment 18, 10901loo. Raven, J. A. (1995b). The early evolution of land plants: aquatic ancestors and atmospheric interactions. Botanical Journal of Scotland 47, 151-175. Raven, J . A. (1995~).Costs and benefits of low intracellular osmolarity in cells of freshwater algae. Functional Ecology 9, 701-707. Raven, J. A. (1996). Into the voids: the distribution, function, development and maintenance of gas spaces in plants. Annals of Botany 78, 137-142. Raven, J. A. (1997). Inorganic carbon acquisition by marine autotrophs. Advances in Botanical Research (in press). Raven, J. A. and Johnston, A. M. (1992). Inorganic carbon supply to algae submerged in acid wetland pools: analysis using the natural abundance of stable isotopes. Botanical Journal of Scotland 46, 321-330. Raven, J. A. and Richardson, K. (1984). Dinophyte flagella: a cost-benefit analysis. New Phytologist 98, 259-276. Raven, J . A . and Spicer, R. A. (1996). Crassulacean acid metabolism. Biochemistry, ecophysiology and evolution, I n “Proceedings of the International Symposium on Crassulacean Acid Metabolism, Panama City, 1993” (K. Winter and J. A. C. Smith, eds), pp. 360-385. Springer-Verlag, Berlin. Raven, J . A., Wollenweber, B. and Handley, L. L. (1993). The quantitative role of ammonia/ammonium transport and metabolism by plants in the global nitrogen cycle. Physiologia Plantarum 89, 512-518. Rebeille, F., Bligny, R., Martin, J.-B. and Douce, R. (1983). Relationship between the cytoplasm and vacuole phosphate pool in Acer pseudoplatanus cells. Archives of Biochemistry and Biophysics 225, 143-148. Reed, R. H., Collins, J. C. and Russell, G. (1980). The effects of salinity upon cellular volume of the marine red alga Porphyrn purpurea (Roth.) C. Ag. Journal of Experimental Botany 31, 1521-1537. Sanders, D., Davies, J. M . , Rea, P. A , , Brosnan, J. M. and Johannes, E . (1992). Transport of H + , K + , and Ca2+ at the vacuolar membrane of plants. In “Plant Organelles” (A. K. Tobin, ed.) S.E.B. Seminar Series SO, pp. 169-188. Cambridge University Press, Cambridge. Schulz, H. N., Jorgensen, B. B., Fossing, H. A. and Ramsing, N. B. (1996). Community structure of filamentous, sheath-building sulfur bacteria, Thioplaca

86

J . A. RAVEN

spp., off the coast of Chile. Applied and Environmental Microbiology 62, 1855-1862. Sievers, A., Kruse, S., Kuo-Huang, L.-L. and Wendt, M. (1989). Statoliths and microfilaments in plant cells. Planta 179, 275-278. Sydenham, P. H. and Findlay, G. P. (1975). Transport of solutes and water by resetting bladders of Utricularia. Australian Journal of Plant Physiology 2, 335-35 1. Trebacz, K., Simonis, W. and Schonknecht, G. (1994). Cytoplasmic Ca2+, K+, C1and NO,- activities in the liverwort Conocephalum conicum L. at rest and during action potentials. Plant Physiology 106, 1073-1084. Villareal, T. A. (1992). Buoyancy properties of the giant diatom Ethmodiscus. Journal of Plankton Research 14, 459-463. Villareal, T. A. and Carpenter, E. J . (1994). Chemical composition and photosynthetic characteristics of Ethmodiscus rex (Bacillariophyceae): evidence for vertical migration. Journal of Phycology 30, 1-18. Villareal, T. A., Altabet, M. A. and Culver-Rymsza, K. (1993). Nitrogen transport by vertically migrating diatom mats of the North Pacific Ocean. Nature 363, 709-712. Vogelman, T. C. (1993). Plant tissue optics. Annual Review of Plant Physiology and Plant Molecular Biology 44, 231-235. Voltaire (1759). “Candide”. Walker, N. A., Patrick, J. W., Zhang, W.-H. and Fieuw, S. (1995). Efflux of photosynthate and acid from developing seed coats of Phaseolus vulgaris L.: a chemiosmotic analysis of pump-driven efflux. Journal of Experimental Botany 46, 539-549. Walsby, A. E. (1972). Gas-filled structures providing buoyancy in photosynthetic organisms. Society for Experimental Biology Symposia 26, 233-250. Walsby, A. E. (1994). Gas vesicles. Microbiological Reviews 58, 94-144. Washburn, E. W. (ed.) (1928). “International Critical Tables of Numerical Data, Physics, Chemistry and Technology”, Vol. 111. McGraw-Hill, New York. Waterbury, J. B., Willey, J. M., Franks, D. G., Valois, F. W. and Watson, S. W. (1985). A cyanobacterium capable of swimming motility. Science 230, 74-76. Welbaum, G. E., Meinzer, F. C., Grayson, R. L. and Thornham, K. T. (1993). Evidence for and consequences of a barrier to solute diffusion between the apoplast and vascular bundles in sugar cane stalk tissue. Australian Journal of Plant Physiology 19, 611-623. Winter, H., Robinson, D. G. and Heldt, H. W. (1993). Subcellular volumes and metabolite concentrations in barley leaves. Planta 191, 180-190. Winter, H., Robinson, D. G. and Heldt, H. W. (1994). Subcellular volumes and metabolite concentrations in spinach leaves. Planta 193, 530-535. Young, A. J., Collins, J. C. and Russell, G. (1987). Ecotypic variation in the osmotic responses of Enteromorpha intestinalis (L.) Link. Journal of Experimental Botany 38, 1309-1324. Zhen, G., Koyro, H.-W., Leigh, R. A., Tomos, A. D. and Miller, A. J. (1991). Compartmental nitrate concentrations in barlev root cells measured with nitrate-selective microelectrodes and single-cell‘ sap sampling. Planta 185, 356-361. Zhen, R. G., Smith, R. J. and Miller, A. J . (1992). A comparison of nitrate-selective micro-electrode made with different nitrate sensors and the measurement of intracellular nitrate activities in cells of excised barley roots. Journal of Experimental Botany 43, 131-138.

The Vacuole and Cell Senescence

P. MATILE

Institut fur wanzenbiologie, Universitat Zurich, Zollikerstrasse 107, CH-8008 Zurich, Switzerland

I.

11.

Introduction ............................................................................. A. Senescence and Death ........................................................ B. Functions of Vacuoles in Cell Senescence .............................. Leaf Senescence ............. A. Differential S B. Vacuolar Hydrolases .

E. F. G.

...................

.................................................... Accumulation and Export of Solutes ..................... Vacuoles and the Breakdown of Chlorophyll .......... Secondary Compounds ...............................

111.

Senescence and Autolysis in Various Cell Phenotypes

IV.

Programmed Cell Death

V.

Retrospect .......... References ..........

.........

.........................

A.

89 89 90 92 94 96 97 102 103

105

.......................... 106 .......... ..... 107

....... .......

I.

....................

87 87 88

INTRODUCTION SENESCENCE AND DEATH

Death is intimately related to life, and life is unimaginable without senescence, the path of development that ultimately leads to death. Senescence and death have important functions in plant development, and it is easily appreciated that they are programmed and regulated in a subtle Advances in Botanical Kescnrch Vol. 25 incorporating Advances in Plant Pathology ISBN o- I m s m - 8

Copynght Q 1997 Academic P r e s L.imiled All rights of reproduction in any form reserved

88

P. MATILE

manner. In plants the entire life cycle is associated with senescence and death of individual cells, tissues and whole organs. For example, organs of flowers such as petals or stamens that after pollination have lost their function, senesce and eventually wither. Such a vitally important function as longdistance transport of water in the xylem depends on the differentiation of meristematic cells into longitudinally connected rows of dead capillaries. Thus, senescence and death of some cell phenotypes start at the embryonic stage and continue as long as the plant is sprouting. Apart from xylem elements there are many instances of other cell phenotypes that have a function as dead cells, e.g. sclereids, or are extremely short lived, e.g. root hairs. The well-defined topographical and temporal pattern of senescence and death clearly demonstrates an underlying genetic programme. Senescence is distinct from other developmental processes only by its end-point, death. As plants cope with normally scarce availabilities of soil-borne nutrients, it is understandable that in most cases senescence is associated with the remobilization of cytoplasmic components and with the recycling of valuable nutrient elements to other parts of the plant. Senescence plays, therefore, an important role in the maintenance of growth and development under conditions of limited budgets of nutrients. Senescent leaves are typically sources of nitrogen, phosphorus, sulfur, potassium and magnesium. Nothing could better illustrate the significance as well as the efficiency of recycling of nitrogen than an observation of Mei and Thimann (1984) on the development of oat plants with only the seed protein as the nitrogen source: the wretched plants were able to complete the life cycle and ultimately produce a single seed. Leaves remain turgescent down to the end of the senescence process, indicating that the intactness of membranes and the compartmentation of solutes are retained. Indeed, the catabolic reactions and metabolism of breakdown products could not take place in an orderly fashion if senescent cells were not perfectly intact. Persistence of subcellular compartmentation is also demonstrated by the phenomenon that leaves do not turn brown until the very end of senescence. In other words, browning may be regarded as a mark of death: it is the result of enzymic oxidation of phenolics which, in the living cell, were sequestered in the vacuoles and, hence, spatially separated from phenol oxidase , a plastid-located enzyme. B. FUNCTIONS OF VACUOLES IN CELL SENESCENCE

Vacuoles represent a multifunctional compartment. Since senescence must be regarded as a normal developmental process taking place in viable cells, practically all the functions of vacuoles outlined in other chapters of this volume are relevant. Thus, ion homoeostasis in the cytoplasm is based on compartmentation in the cell sap, and transport across the tonoplast is as important for metabolic functions in senescent cells as it is during other stages

THE VACUOLE AND CELL SENESCENCE

89

of development. Leaf senescence is subject to hormonal regulation, and the potential role of vacuoles in the compartmentation of calcium ions, signal transduction and gene expression must, therefore, be considered as well. There is also little doubt that the temporary storage of intermediary metabolites such as organic acids and amino acids continues to be a function of vacuoles as cells are induced to senesce. Orderly metabolism also requires the sequestration in the cell sap of such potentially toxic secondary compounds as phenolics. It will be seen that in senescent mesophyll cells, water-soluble products of chlorophyll (Chl) breakdown are handled like secondary compounds and deposited in the vacuoles. The compartmentation in vacuoles of unspecific hydrolases such as proteases, RNAases and acid phosphatase may also be regarded as a kind of detoxification of the cytoplasm. It is quite tempting to consider a role of these enzymes in the degradation of cytoplasmic components as it occurs in senescent cells. A corresponding function of the vacuole (as part of the lytic compartment of plant cells) has indeed been proposed in the past (Matile, 1975). Yet, the mechanism and subcellular organization of protein breakdown in senescent cells have so far not been elucidated. It will be seen that such phenomena as the remobilization of protein in the chloroplasts and the accumulation of hydrolytic potential in the vacuoles are still poorly understood, if at all.

11. A

LEAF SENESCENCE

DIFFERENTIAL SENESCENC'E LN TISSUES AND ORGANELLES

Leaves have a limited lifespan. After developnient to maturity they begin to senesce in a matter of hours (e.g. petals of ephemeral flowers), days, weeks or, in the case of evergreen species, years. Regardless of the lifespan, senescence appears to be governed by a genetic programme aimed at the orderly breakdown of cell constituents and the recycling of valuable nutrients to other parts of the plant. Experimentally, foliar senescence is usually followed by measuring declines of Chl and protein. This procedure largely covers breakdown processes in the mesophyll. However, the various tissues of leaves senesce very differently. Whereas in the xylem senescence of the tracheary elements occurs during leaf expansion, the phloem part of vascular bundles, which has an important function in the export of breakdown products, does not senesce before the mobilization in the mesophyll is nearly completed. Senescence is also delayed in the epidermal tissue, particularly in the guard cells which retain the capacity of turgor-driven adjustment of stomata1 aperture and control of water loss. A pattern of differential senescence cannot only be observed with regard to tissues but also to the various organelles of mesophyll cells. As judged by the loss of protein, the chloroplasts represent the principal site

90

P. MATILE

of degradation. The abundant soluble protein ribulose-l,5-bisphosphate carboxylase/oxygenase (Rubisco) together with the apoproteins of the photosynthetic pigments contributes largely to the overall loss of protein in the senescent leaves. Other organelles such as mitochondria and peroxisomes are excluded from degradation. The functional intactness of mitochondria is important with regard to the requirements for metabolic energy. Protein synthesis, which in senescent leaves takes place at surprisingly high rates (Klerk et af., 1992), is just one of several energy-consuming parts of metabolism. It is also important that membranes, notably the plasma membrane and tonoplast, retain semipermeability. As senescence advances, individual mesophyll cells may collapse, but as long as the epidermal cells remain viable the leaves remain turgescent. Regarding the function of vacuoles it is certainly intriguing to consider a role of the various hydrolytic enzymes in digestive processes taking place in senescent cells. The chloroplasts of mesophyll cells are particularly interesting as they represent the major source of nitrogen available for remobilization and recycling. Chloroplasts have been claimed to be degraded sequentially in senescent wheat leaves (Camp et af., 1982), and ultrastructural observations have even suggested a movement of whole organelles into the central vacuole (Wittenbach et af., 1982). Such a mechanism of digestion of sequestered chloroplasts appeared to provide a logical explanation of organelle-selective breakdown. In the meantime it has been demonstrated, however, that the number of chloroplasts remains practically constant throughout the senescence period (Martinoia et af., 1983; Mae et af., 1984; Wardley et al., 1984; Grover et af., 1987). It is now generally accepted that chloroplasts remain intact and in the course of senescence develop into a distinct type of plastid designated gerontoplasts (see Matile, 1992). Hence, if vacuoles play a role in the differentiation of gerontoplasts it would be a matter of transfer of components across the threefold membrane barrier of the plastid envelope and the tonoplast. B. VACUOLAR HYDROLASES

The lysosomal character of vacuoles is documented by numerous results of compartmentation analysis with vacuoles isolated from protoplasts (Boller and Kende, 1979; see also Marty et al., 1980; Boller, 1982; Wagner, 1982; Boller and Wiemken, 1986). The catalogue of vacuolar hydrolases comprises endopeptidases, carboxypeptidase, RNAase, DNAase, acid phosphatase, phosphodiesterase and several glycosidases. Most of these enzymes have a pH optimum matching the usually acidic pH of cell saps. They are unspecific in the sense that they can be assessed with artificial substrates such as p-nitrophenyl- or umbelliferylglycnsides in the case of glycosidases. With regard to protein breakdown in senescent cells, special attention must be paid to the proteolytic enzymes. Thus, the vacuolar endopeptidases hydrolyse such


91

convenient substrates as haemoglobin or azocoll as well as relevant substrates such as Rubisco. Compartmentation analysis with protoplasts as a source of vacuoles is not overly accurate, so that locations of the predominant hydrolases in other subcellular compartments cannot be excluded. A major disadvantage of work with protoplasts is, however, the loss of the apoplastic space that is inevitably associated with the digestion of cell walls. Indeed, the apoplast has been identified as a second major part of the lytic compartment of plant cells (see Matile, 1975). In oat leaves, a substantial proportion of the total acid endopeptidase activity has been recovered in the apoplast wash (van der Valk and van Loon, 1988). Likewise, the cell walls of leaves of the French bean contain an azocoll-digesting proteinase (Van der Wilden et al., 1983). A twofold location in vacuoles and extracellular space has also been found for P-1,3-glucanase, an enzyme which, together with the vacuolar chitinase, may have an antimicrobial function (see Boller, 1993). The digestive potential of the two major extraplasmatic cell compartments, vacuole and apoplast, has been displayed repeatedly in terms of autodigestion of cytoplasmic constituents in homogenized tissues. The efficient degradation of proteins and nucleic acids taking place upon the disruption of the subcellular organization demonstrates quite clearly that in vivo a subtle control of exposure of endogenous substrates to the digestive enzymes in the lytic compartment would be required should, indeed. these enzymes have a function in the living cells. Digestion in senescent tissues caused by the release of lysosomal enzymes into the cytoplasm is rather unlikely because the vacuolar compartmentation appears to be unchanged as senescence is induced in the mesophyll cells (Table 1; Heck et af., 1981; Wittenbach et al., 1982). Genes that in maize leaves are expressed specifically during senescence include a vacuolar endopeptidase (homologous to rice oryzain) and also a gene homologous to a castor bean vacuolar processing enzyme (Smart et al., 1995). Thus, new vacuolar hydrolases appear not only to be synthesized in senescent leaves but also properly targeted and processed. Unfortunately, reports on the intracellular digestion of cytoplasmic constituents within the vacuoles are extremely scarce. With regard to protein breakdown the vacuolar proteolytic system may have been superseded upon the discovery of proteases that are clearly located outside the lytic compartment, as, for example, in the chloroplasts, or by the specific targeting of protein for degradation by the cytoplasmic ubiquitin system (see Huffaker, 1990). Correlations that have been reported to exist between enzyme activities in cell-free extracts and processes taking place in senescent leaves are hardly relevant. Nevertheless, it should be mentioned that activities of typical vacuolar hydrolases have been observed to increase considerably, in some cases even dramatically, when leaves began to senesce. In senescent leaves of cereals, protein mobilization appeared to be correlated with increasing

92

P. MATILE

TABLE I Compartmentation of various hydrolases in vacuoles isolated from barley mesophyll protoplasts. Comparison of data obtained from mature green and senescent primary leaves. Activities present in preparations of vacuoles were related to total activities in the lysed protoplasts taking a-rnannosidase activity as a vacuolar marker. Data from Martinoia (1982) YO of total activity in vacuoles

Hydrolase

Mature leaves

Senescent leaves

-__

a-Mannosidase Acid phosphatase Phosphodiesterase Acid protease a-Galactosidase /3-Galactosidase

100 75 55 87 75 53

100 65

79 91 58

37

endopeptidase activities (e.g. Martin and Thimann, 1972; Peterson and Huffaker, 1975; Thomas, 1978; Wittenbach, 1978; Cheng and Kao, 1984; see also Feller, 1986), particularly when senescence was studied in detached leaves. RNAases represent further examples of presumably vacuolar enzymes that tend to increase in activity in senescent leaf tissue of various species (see Matile, 1975). Some of these marked changes seem to occur only under unnatural conditions such as in detached leaves kept in permanent darkness. Thus, the increase of two major endopeptidases in senescent barley leaves occurs only in the detached leaves (Miller and Huffaker, 1985; see also Huffaker, 1990). Likewise, the expression in wheat leaves of three RNAase activities occurred only upon dark-induced senescence and was absent under natural conditions (Blank and McKeon, 1991). Since in many cases the exact localization of the hydrolase in question with regard to the cell phenotype is unknown, conclusions about functions in the senescence-associated degradation processes can hardly be drawn. The enzyme may be located in epidermal cells whilst the degradation observed may take place in the mesophyll. Hence, for the time being it can only be stated that in cells of senescent leaf tissue a digestive potential is maintained or even rendered more powerful as senescence is induced. And yet, the relevance of the lytic compartment for the degradation of proteins, nucleic acids and other cell constituents still remains to be elucidated. C. AUTOPHAGIC ACTIVITY OF VACUOLES

Autophagy has been documented largely in terms of electron micrographs showing structures within vacuoles that are more or less clearly of cytoplasmic origin. When static images of electron micrographs were translated into the


93

dynamism of living cells, tonoplasts were thought to engulf cytoplasmic material by invagination; upon the budding off of such invaginations. portions of cytoplasm would be transferred into the cell sap (see Matile, 1975). There are certainly possible pitfalls associated with this conception of autophagic activity as based on morphological evidence. Thus, the fixation of highly vacuolated and, particularly, of senescent cells may produce distortions of membranes, giving the impression of protrusions into the vacuolar space that may have nothing to do with transfer into the digestive compartment. Very few studies include the analysis of serial sections to demonstrate convincingly the budding off of tonoplast invaginations into the cell sap (e.g. Herman et af., 1981). Another disadvantage of morphological investigations must be seen in the lack of a biochemical correlate: in most cases nothing is known about the nature of autophagocytosed and eventually digested cell constituents. There is no doubt that autophagy is an important attendant symptom of differentiation of meristematic into functionally specialized cells. The autophagic nature of the vacuolation process has above all been shown in differentiating laticifers (Marty, 1978; Fineran, 1983). Similarly, disintegration of chloroplasts through formation of autophagic vacuoles has been observed in developing zygotes of Spirogyra (Ogawa, 1982). The documentation of autophagy is less convincing with regard to senescent leaf cells. As judged by the occurrence of deposits in the cell sap, intravacuolar vesicles, and (apparent) invaginations of tonoplasts, increasing autophagy has been observed during senescence of flower petals (Matile and Winkenbach, 1971; Smith ef af., 1992). Regarding the senescence of green leaves, attention must be drawn to the question of whether or not autophagic activity of the central vacuole plays a role in gerontoplast development. Electron microscopists have described in great detail the transition of chloroplasts into gerontoplasts (see references in Matile, 1992), but in adequately fixed specimens they have not observed a morphologically obvious transfer of plastidic material into vacuoles. An exception, though unrelated to senescence, is the observation of evaginations from the plastid envelope that apparently bud into the vacuole (Vaughn and Duke, 1981). However, there is increasing evidence that the chloroplasts themselves represent an autophagic compartment. Not only have several proteolytic enzymes been localized in chloroplasts but it has also been shown that unassembled proteins or prolamellar bodies in greening etioplasts are digested within the organelles (see Huffaker, 1990). The formation within isolated chloroplasts of distinct fragments of the Kubisco large subunit (Mitsuhashi et a[., 1992) further corroborates the view that chloroplasts are indeed capable of degrading endogenous protein. Although the problem of orderly mobilization of protein in senescent chloroplasts is by no means solved, a role played by vacuoles appears to be rather unlikely.

94

P. MATILE D. AUTOLYSIS

Electron microscopic observations suggest that the rupture of the tonoplast marks the end of the orderly senescence process (see Matile, 1975). As a consequence of abolished vacuolar compartmentation the hydrolases are released and digest cell constituents such as the nucleus, mitochondria and cytosol, which, in the viable senescent cell, had been excluded from degradation because they had important functions such as in senescencespecific protein synthesis, generation of metabolic energy, synthesis of amides for export via the phloem and general house-keeping. Such a role of vacuoles in cell autolysis may shed light onto the apparently unnecessary new synthesis of hydrolases in senescent leaves. In the ephemeral flowers of the morning glory, Zpornoea tricolor, the fading of the corolla is preceded by an enormous increase of RNAase activity (Matile and Winkenbach, 1971; Baumgartner et al., 1975) which is obviously not correlated at all with the decline of RNA (Fig. 1). Indeed, in the homogenized tissue the comparatively low RNAase activity in the flower buds is sufficient for the complete hydrolysis of RNA in a matter of minutes. DNAase activity, which also increases concomitant with RNAase, may have a function in the digestion of nuclear DNA beginning at the onset of flower fading, most probably reflecting the progressive autolysis of cells in the mesophyll (Matile and Winkenbach, 1971). In senescing wheat leaves the DNA content was found to remain almost constant during the first few days and thereafter began to decrease dramatically (Lamattina et al., 1985); again this phenomenon may point to progressive autolysis in the mesophyll at advanced stages of senescence. The lytic potential of vacuoles may be maintained and even intensified in senescent cells in order to achieve a rapid degradation of the remaining cell constituents when subcellular compartmentation collapses. This view is endorsed by recent studies of differentiation in vitro of isolated mesophyll cells of Zinnia elegans into tracheary elements (Fukuda, 1994, 1996). As required by the function of tracheids, the final event of differentiation is autolysis (Fig. 2). It is preceded by the expression of a gene with sequence homology to papaya proteinase I (Ye and Varner, 1993) and by the synthesis of potent nucleases (Thelen and Northcote, 1989). Undoubtedly, these are vacuolar enzymes which, upon the abolishment of subcellular compartmentation, may take charge of the hydrolytic cleaning of the autolysed tracheid. Leigh (1979) has doubted that vacuoles of mature plant cells play a role in the degradation of cytoplasmic components; he has interpreted the vacuolar hydrolases as remains of those that had a lysosomal function during early stages of cell differentiation. And yet, the presence of an apparently functionless digestive machinery belongs to the standard equipment of

95

THE VACUOLE AND CELL SENESCENCE Anthesis

12

00 I

I

00 I

I

Senescence

11 I

I

00 I

12 I

..... .... 1

00

12

00

12

00

I

.

~

I

~

L

12

Time by the clock

Fig. 1 . Senescence in the ephemeral flowers of Ipornoea tricolor is associated with marked increases of RNAase and DNAase activities. The loss of DNA is likely to indicate the autolysis of mesophyll cells and the hydrolysis of nuclear DNA, respectively. Acid-soluble phosphates produced upon the breakdown of nucleic acids are withdrawn from the corolla. (Redrawn from Matile and Winkenbach (1971).)

mature plant cells. When tobacco protoplasts were evacuolated artificially, not only were vacuoles subsequently regenerated but they were also newly supplied with the typical hydrolases (Hortensteiner et al., 1992). The digestive machinery may have a dual function. According to Boller (1986) it represents a constitutive weapon of plant cells for defence against invading microbes, and in the context of cell senescence it is likely to represent a tool for wholesale hydrolysis after cell death.

96

P. MATILE 0

Days

1

2

3

I

I

IT

4

House keeping events Oediflerentiation events

t

Differentiation

Secondary wall thickening

I Initiation

4

-

RNase

b

ssDNase

SH-Protease

Fig. 2. Sequential events taking place during transdifferentiation of single mesophyll cells of Zinnia eleguns into tracheary elements. (Simplified, with emphasis on the synthesis of hydrolytic enzymes preceding autolysis, from Fukuda (1994).)

E. ACCUMULATION AND EXPORT OF SOLUTES

So far only one consequence of tonoplast rupture has been considered: the hydrolases released will do away quickly with all digestible remains of the cytoplasm. Another consequence is the release of vacuolar solutes which, together with new hydrolytically produced solutes, should ultimately be loaded into sieve tubes for export to other parts of the plant. Minor veins are located in close proximity of mesophyll cells, and the release of solutes will soak the apoplastic continuum so that uptake by bundle sheet or companion cells can take place. Neighbouring mesophyll cells that are still viable may also take up solutes and store them in the vacuole until they autolyse likewise. Senescent mesophyll cells retain the capacity of accumulating solutes. In detached leaves, export via the phloem is interrupted, and amounts of amino acids increase in the vacuoles of viable mesophyll cells (Table 11); amounts of inorganic phosphate also increase probably at the expense of degraded nucleic acids (Martinoia, 1982). As senescence advances, the preparation of protoplasts for the purpose of compartmentation analysis becomes increasingly difficult. Reduced yields per unit of fresh leaf weight (Martinoia, 1982) may reflect increasing fragility of protoplasts but may also be due to increasing proportions of lysed cells. The hypothetical interplay of asynchronous autolysis in the mesophyll, accumulation of solutes by neighbouring viable cells and the efficient

97


TABLE I1 Contents and compartmentation in vacuoles of amino acids and inorganic phosphate in mesophyll protoplasts from mature and senescent hiirley primary leaves. Excised leaves were allowed to senesce in darkness. Data from Martinoia (1982) Concentration (pmol per lo6 protoplasts)

Amino acids p,

Yo of total in vacuoles

Mature

Senescent

Mature

Senescent

1.6 0.45

4.2 0.77

53 99

102

95

withdrawal of solutes to be exported from the leaf into the phloem are illustrated in Fig. 3. The efficiency of withdrawal of nutrients from the mesophyll may be very high. Nevertheless, in the fully senescent leaves the residual protein normally accounts for some 1 0 4 0 % of total protein in the mature leaves. This is probably a consequence of differential senescence in the various cell phenotypes; the epidermal cells may account for the bulk of the residual protein. F.

VACUOLES AND THE BREAKDOWN OF CHLOROPHYLL

The loss of Chl is generally employed as a convenient parameter for recording the progress of leaf senescence. This is justified because the disappearance of the green colour is positively correlated with other relevant parameters such as loss of protein. In contrast to protein degradation, which plays an important role in the recycling of nitrogen from senescent leaves to other parts of the plant, Chl is not catabolized for the benefit of the four nitrogen atoms per molecule to be reused. The structural analysis of several catabolites has yielded ample evidence that breakdown of Chl-porphyrin does not proceed beyond the stage of linear tetrapyrroles. Studies with a stay-green mutant of Festuca pratensis have shown, however, that Chl breakdown is a prerequisite to the remobilization of apoproteins of the chloroplast pigmentprotein complexes (Thomas, 1982; Thomas and Hilditch, 1987) which account for a substantial proportion of the total protein of chloroplasts. It appears that the apoproteins are out of reach of proteases as long as they are complexed with Chl. And yet, the mere disassembly of complexes for the benefit of protein degradation is not feasible because Chl is photodynamic, i.e. toxic, in nature. Breakdown of Chl-porphyrin and the photodynamic inactivation associated with it must be regarded as a prerequisite for apoprotein remobilization to take place in viable senescent mesophyll cells.

98

P. MATILE Mesophyll

Bundle Companion sheet cell

Sieve tube

Fig. 3. Hypothetical role of vacuoles and autolysis in mesophyll senescence. It is assumed that solutes remain accumulated in senescing cells until autolysis takes place. Solutes released from autolysing cells will diffuse into the apoplast and may be partially absorbed by neighbouring viable cells. Digestion of residual cytoplasmic components by the vacuolar hydrolases is another important consequence of the collapse of subcellular compartmentation. The resulting solutes will also be available for reabsorption and eventual loading into the sieve tubes. Transport processes not shown in the model concern a possible drain of the vascular bundle via the symplastic route.

The radiolabelling of Chl in the pyrrole units during greening of barley primary leaves has allowed the fate of the porphyrin moiety during subsequent senescence to be followed (Peisker et al., 1990). In the course of leaf yellowing, the label, of which initially about 75% was present in Chls

THE VACUOLE AND CELL. SENESCENCE

99

a and h of the mature leaves, progressively accumulated in the fraction of water-soluble compounds. A number of radiolabelled colourless catabolites have been identified in this fraction; upon isolation and purification one of them turned out to be a linear tetrapyrrolic derivative of Chl a (Krautler et al., 1991). In attached leaves senescing under natural conditions the “C present in breakdown products of Chl was quantitatively retained, i.e. there was no export to other parts of the plant. In the senescent mesophyll cells, the water-soluble catabolites were found to be accumulated in the vacuoles; within the accuracy of compartmentation analysis with protoplasts and isolated vacuoles, all of the “C stemming from Chl was localized in the vacuoles (Peisker, 1991). This result was obtained with detached leaves induced to yellow rapidly in permanent darkness and also with attached primary leaves that were allowed to seiiesce slowly under natural conditions. Moreover, the vacuolar location of distinct catabolites (Matile et al., 1988) such as Hv-FCC-2 (Duggelin et al., 1988) and Hv-NCC[ RP141 (Bortlik et al., 1990) has been demonstrated (for the terminology of Chi catabolites, see Ginsburg and Matile, 1993). It appears that the breakdown products of Chl-porph yrin, like secondary compounds. are sequestered in the vacuolar sap of senescing mesophyll cells. Upon death and shedding of leaves they may contribute to the diet of soil microorganisms. Work with another experimental system, cotyledons of oilseed rape, corroborated the notion of similarity between Chl catabolites and watersoluble secondary metabolites. In the senescent cotyledons three NCCs (non-fluorescent chlorophyll catabolites) were identified and found to represent the complement of the Chl originally present in the mature green leaves (Ginsburg and Matile, 1993). The major NCC, Bn-NCC-1, was identified as an oxidatively cleaved derivative of phaeophorbide a (Fig. 4); an interesting modification concerns the hydroxylation and esterification of the hydroxyl group with malonic acid, respectively, in the side-chain of pyrrole B (Muhlecker et al.. 1993). One of the minor NCCS, Bn-NCC-2, was identified as the glucosyl analogue of NCC-1, and Bn-NCC-3 turned out to represent the aglycon of NCC-2 (Fig. 4; Muhlecker and Krautler, 1996). Modifications such as hydroxylations and conjugations with malonic acid or glucose are, indeed, very common in the metabolism of secondary compounds. In the case of Chl breakdown they are most probably responsible for the amazing diversity of catabolites so far identified in different plant species in terms of structure and/or polarity upon reverse-phase highperformance liquid chromatography. Whereas the NCCs represent secondary or final products of breakdown, FCCs (fluorescent chlorophyll catabolites) most probably are early or even primary catabolites. In minute quantities they occur in the leaves when rates of Chl breakdown are high. In vitro they are produced by t h e action of an oxygenase which cleaves phaeophorbide a , with reduced ferredoxin acting

P. MATILE

100

i M

&OPhytyi

Chlorophyll-a

Bn-NCC-1 R = COCH,COO'K@ BnNCC-2 R = C,H,,O, (p-Glucose) Bn-NCC-3 R = H

Fig. 4. Structures of Chl a and of the three predominant catabolites accumulated in senescent cotyledons of oilseed rape, Bn-NCC-1, -2 and -3. Note that all three catabolites, which account for practically all the Chl broken down, are derivatives of Chl a; it seems that in senescent leaves Chl b is converted to Chl a prior to cleavage of the porphyrin macrocycle.

as a reductant (Schellenberg et al., 1993; Ginsburg et al., 1994; Hortensteiner et al., 1995). In intact barley gerontoplasts the major FCC produced in organello, Hv-FCC-2, is released into the medium in an ATP-dependent fashion, indicating a release into the cytosol in vivo (Matile et al., 1992a). In the senescent mesophyll cells, Hv-FCC-2 accumulates, together with several NCCs, in the vacuoles. Indeed, the tonoplast appears to be equipped with a carrier that recognizes Chl catabolites. The uptake of catabolites into vacuoles isolated from barley mesophyll protoplasts was found to be strictly ATP dependent, and had features of a primary active transport (Hinder et al., 1996). In a heterologous system the barley vacuoles transported the radiolabelled Bn-NCC-1 prepared from rape cotyledons. The two catabolites from barley, Hv-FCC-2 and Hv-NCC(RP14), inhibited the uptake of Bn-NCC-1, indicating that the carrier in the tonoplasts does not distinguish the specific structural features of catabolites. Rather surprisingly, the transport of Bn-NCC-1 was not inhibited by bilirubin, a catabolite of haem in mammals. Therefore, one or several of the structural peculiarities that distinguish Chl catabolites from degradation products of haem (e .g. the formyl group in pyrrole B,the isocyclic ring, or hydroxylations in side-chains

101


)

thgiL)

Inner-

(

Chlorophyllase . .

1

riiaaiua a*

? Mg-Dechelatase

J

Fe3f

Oxygenase

Outer Envelope Membrane

Modification

t

I

~

Cytosol

ATP / I

t

Tonoplast

Chlorophyll

_'cn- FCC

@ Chlorophyllide

0

Phaeophorbide

Pn- NCCs

Fig. 5 . Model of Chl catabolism in senescent mesophyll cells as based on current knowledge. The inner membrane of the gerontoplast envelope has recently been identified as the site of at least two Chl catabolic enzymes, chlorophyllase and phaeophorbide-a oxygenase. The stepwise degradation of Chl molecules eventually yields a primary fluorescent catabolite which, in an energy-dependent fashion, is released from the senescent chloroplasts. The final catabolites are probably formed in the cytosol and transported across the tonoplast.

102

P. MATlLE

of pyrroles A and/or B) may be decisive for the recognition of linear tetrapyrroles by the carrier in the tonoplast. A current concept of Chl breakdown in senescent leaves featuring the final deposition of catabolites in the vacuoles is illustrated in Fig. 5. Of the enzymes engaged in the overall process, so far only the ring cleaving “oxygenase” has been found to be regulated in a senescence-specific fashion (Ginsburg ef al., 1994; Hortensteiner et al., 1995; Vicentini ef al., 1995). Chlorophyllase, dechelatase and also the catabolite carrier in the tonoplast appear to represent constitutive components of the amazingly complicated and energy-consuming catabolic machinery.

G . SECONDARY COMPOUNDS

Leaf senescence in forest and ornamental trees is highly esteemed for its aesthetic value. In many species the autumnal foliage not only has a golden appearance due to the partial retention of carotenoids during Chl degradation but turns orange or red due to the concomitant synthesis of anthocyanins. In Populus tremuloides the accumulation of such typical vacuolar phenolics as cyanidin-3 glucoside and cyanidin-3 galactoside has been observed to differ depending on the environmental conditions and also on the genotype (Chang et al., 1989). This is consistent with the general notion that over the years the autumn coloration of trees is variable, indicating that the senescenceassociated syntheses of secondary compounds are unlikely to have a vital function. A most intriguing kind of secondary metabolism occurs in the autumn leaves of Ginkgo biloba. In this species, Chl breakdown is associated not only with an outstandingly high retention of carotenoids but is also accompanied by the synthesis of a fluorescent compound, 6-hydroxykynurenic acid (Schennen and Holzl, 1986), a secondary metabolite of tryptophane which appears to have the effect of an optical brightener (Fig. 6; Matile ef al., 1992b). This compound is located in the vacuoles of scattered mesophyll cells which, in the course of senescence, apparently undergo a kind of transdifferentiation into idioblasts (Matile, 1994). In conjunction with the highly retained carotenoids it appears to cause the unique golden autumn foliage of Ginkgo trees. It is difficult to interpret the biological function of secondary compounds accumulated at the very end of leaf development. Since the phenomenon occurs only in a limited number of species, it is hardly relevant for the general metabolism of senescent leaves. The biosynthesis of secondary compounds may consume such negligible quantities of energy and organic materials that the underlying genetic programmes may have evolved in the absence of selection pressure. Indeed, these vacuolar secondary compounds may have no biological function except for a high aesthetic merit.


103

530

100

r

&

r CJ)

80 Emission

‘z %

c

60

Q)

c

.-C

Q)

40

300

8ur

500

600 nm

/oJ3qcom

?!

G

400

on

20

Oi-

Sept. 18

Senescence period (weeks)

Oct. 30

Fig 6. Breakdown of chlorophyll, retention of carotenoids and accumulation of 6-hydroxykynurenin i n autumn leave\ of Ginkgo hrluba. (From Matile (195)4).)

111.

SENESCENCE AND AUTOLYSIS IN VARIOUS CELL PHENOTYPES

The following examples of cell differentiation have been selected in order t o illustrate the widespread occurrence and importance of senescence and cell death in plant development. The most convincing and best known example is the differentiation of tracheids in the xylem. As already mentioned, the transdifferentiation of isolated Zinnia mesophyll cells into tracheary elements has offered a unique opportunity for the study of differentiation, including the final events of senescence and autolysis, from the viewpoint of biochemistry and gene expression (Fukuda, 1994). Earlier work on the maturation of tracheids described the morphological events (Srivastava and Singh, 1972; Wodzicki and Humphreys, 1973, 1974; Wodzicki and Brown, 1973) and in some cases also the subcellular distribution of acid phosphatase considered as a

104

P. MATILE

cytochemical marker of the lytic compartment (Gahan and Maple, 1966; Charvat and Esau, 1975; Cronshaw and Bentwood, 1977). These studies demonstrate, in one way or another, that subcellular organization, including intactness of the central vacuole, remains normal during the typical formation of secondary wall thickening. This is not surprising as this process requires an orderly metabolism and metabolic energy provided by intact mitochondria. Acid phosphatase has been observed to be strictly compartmented first in the endoplasmic reticulum and Golgi and later in the vacuole. Filamentous structures budding off into the vacuole near the end of tracheid development and eventually fragmenting into intravacuolar spherules (Wodzicki and Humphreys, 1973) may indicate autophagic activity taking place before autolysis is initiated upon the rupture of the tonoplast. Whether changes of the morphology of the nucleus that have been observed during differentiation (Lai and Srivastava, 1976) are due to so far unidentified components of the senescence programme remains to be seen. In any case, the finding in Zinnia that autolysis is preceded by the accumulation of presumably vacuolar hydrolases (Fukuda, 1994, 1996) is intriguing as it suggests that the final phase of development is part of a genetic programme which is ultimately aimed at the wholesale digestion of cytoplasmic structures in the lumen of tracheids. Another example of apparently programmed cell death and autolytic breakdown of cell constituents is represented by resin ducts developing in a lysogenic fashion. The internal cavity of this type of duct develops upon the disintegration of a group of cells that are surrounded by epithelial resin-producing cells. At later stages, autolysis of epithelial cells takes place so that in the end the cavity is filled with resin and the undigested remains of the disintegrated cells (Joel and Fahn, 1980). The digestive glands of the carnivorous genus Pinguicula possess a group of secretory head cells in which the digestive enzymes are synthesized and transferred both into the vacuoles and into the walls. According to Heslop-Harrison and Heslop-Harrison (1981) the head cells senesce during the final phase of maturation and eventually undergo autolysis resulting in the discharge of the digestive enzymes. Senescence with vacuolar implications and autolysis have also been observed in reproductive tissues such as in embryo suspensors of certain species (Nagl, 1976; Gartner and Nagl, 1980), in unpollinated ovaries of the pea (Carrasco and Carbonell, 1988) and in the tapetum of Lilium (Reznickova and Dickinson, 1982). A last example concerns the fungus Coprinus lagopus, which is conspicuous by the rapid senescence and autolysis of the mature fruiting body. Disintegration is associated with degradation of the hyphal cell walls by the action of chitinases. During maturation of the fruiting bodies, chitinase and other hydrolases are accumulated in the vacuoles, and are released into the cell walls upon the collapse of subcellular compartmentation (Iten and Matile,


105

1970). Autolytic processes associated with starvation or with the release of spores are quite common in the development of fungi and algae.

IV. PROGRAMMED CELL DEATH Cell death is a normal and vitally important event in plant ontogeny. The development of tracheary elements emphasizes the significance of autolysis for the function of this cell phenotype. Differential senescence in mesophyll and epidermal cells of leaves provides another example for tissue specificity of development ending with death. It is quite obvious that the temporally and spacially defined senescence and autolytic finale must be governed by a genetic programme. Such a death programme is currently being investigated intensively in animal systems. It is termed “apoptosis”, and comprises a sequence of events that ultimately result in the fragmentation of individual cells into dead “apoptotic bodies” which are eventually removed by phagocytes (Martin et al., 1994). The death programme in plant cells appears to be different, at least as far as the final events are concerned: autolysis seems to be typical for postmortem development of plant cells, in contrast to phagocytic removal of the remains of dead animal cells. Therefore, use of the term “programmed cell death” may be preferable to address development to death in plants, with the more fashionable term “apoptosis” reserved for the specific death programme in animal systems. However, plant and animal systems possibly have in common certain parts of the programme. In the sequence of the apoptosis programme the activation of an endonuclease that cleaves nuclear DNA into fragments of more or less defined length precedes other events that eventually lead to the final formation of apoptotic bodies. Very recently, fragmentation of nuclear DNA has been shown to occur also during differentiation of tracheary elements in higher plants (Mittler and Lam, 1995). This has been demonstrated by employing an immunofluorescence kit, ApopTag, which had been developed for the detection of fragmented DNA in mammalian cells. Whether or not nDNA fragmentation in developing tracheary elements is due to the same mechanism of endonuclease activation as in the apoptosis programme is not yet clear. There is a possibility that in plant cells, fragmentation of nDNA is merely a consequence of autolysis and is brought about by the action of endonucleases released from disrupted vacuoles. The decisive transition of individual cells from life to death is likely to be marked by the release of toxic compounds and hydrolases from ruptured vacuoles. Such a highly active hydrolase as acid phosphatase when released into the cytoplasm will cause an immediate arrest of intermediary metabolism which depends so much on phosphate esters. In plant cell development, death and autolytic digestion of residual cell constituents may, therefore, be

106

P. MATILE

associated with changes in membrane properties that eventually lead to leakage and disrupture. Following this line of thought, the genetic death programme may concern the expression of suicide proteins that are responsible for the damage of membranes generally and tonoplasts specifically. Indeed, membrane deteriorations have been widely observed to occur during senescence of leaf and other tissues. They have been assessed in terms of increasing phase transition temperatures and decreasing membrane fluidity, respectively, phospholipid degradation, lipid peroxidation and accumulation of products of peroxidative degradation of polyunsaturated fatty acids, changes in membrane proteins, and leakage of ions and small molecules (e.g. Suttle and Kende, 1980; Dhindsa et al., 1981; Adam et al., 1983; P a d s and Thompson, 1984; Fobel et al., 1987; Borochov et al., 1990; Paliyat and Droillard, 1992; Yamane et a f . , 1993). Since most of these and similar studies were done with whole organs or membrane preparations from whole organs, it is difficult to associate the changes observed with development in specific cell phenotypes or in distinct membranes. It is also difficult to distinguish events that precede and eventually cause the breakdown of subcellular compartmentation from those which are a consequence of chaotic conditions prevailing in autolysing cells. In the case of petal senescence in Tradescantia, new protein synthesis has been demonstrated to be required for ethylene-induced breakdown of phospholipids and leakiness of membranes (Suttle and Kende, 1980). In terms of gene expression, protcins that have an effect on phospholipid degradation in the tonoplast may be considered as candidates of the cell death programme in plants.

V.

RETROSPECT

It is fair to state that specific functions of vacuoles in cell senescence have not been identified so far. The accumulation of Chl catabolites in vacuoles of senescent mesophyll cells may be regarded as an exception. And yet. this function is not overly important for the understanding of the senescence process in the chloroplasts. It merely provides another example of the normal function of vacuoles as subcellular dustbins. Much more important may be the continued capacity of vacuoles of senescent cells to accumulate solutes and ions for the benefit of maintenance of homoeostasis in the cytoplasm as long as the cells are viable. An understanding of protein breakdown in senescent cells, particularly in developing gerontoplasts, must certainly be regarded as one of the most urgent problems. A role of vacuolar hydrolases in protein remobilization cannot be excluded but is not very likely. With regard to the lytic potential in the cell sap it is perhaps the very end of the senescence process, autolysis, that may provide an explanation of its function. There is little doubt that


107

t h e c u r r e n t enthusiasm f o r apoptosis will cncourage experimental w o r k on t h e genetics a n d biochemistry of cell d e a t h in plants. T h e Zinnia system may play an i m p o r t a n t role because it provides a u n i q u e o p p o r t u n i t y t o s t u d y programmed cell d e a t h in a more or lcss uniform p o p u l a t i o n of synchronously developing cells. A system of this kind with cells f r o m Arahidopsis instead of Zinrziu would unquestionably be an ideal experimental tool f o r exploring t h e p h e n o m e n o n of cell d e a t h in plants at a genetic and molecular level.

REFERENCES Adam, Z . , Borochov, A . , Mayak, S. and Halevy. A . H. (1983). Correlative changes in sucrose uptake, ATPase activity and membrane fluidity in carnation petals during senescence. Physiologia Pluntururri 58, 257-262. Raumgartncr. B., Kende, H . and Matile, Ph. (1975). RNase in Japanese morning glory. Purification and demonstration of de novo synthesis. Plant Physiology 55, 734-737. Blank, A . and McKeon, T. A . (1991). Exprcssion of three RNase activities during natural and dark-induced senescence in wheat leaves. Plant Physiology 97, 1409- I41 3. Boller, Th. (1986). Enzymatic equipment of plant vacuoles. PhysiologiP VPgCtule 20, 241-257. Boller, T h . (1982). Hydrolytic enzymes in plant disease resistance. In “Plant-Microbe Interactions” (T. Kosuge and E. W. Nester, eds), Vol. 2, pp. 385413. Macmillan, New York. Boller, Th. (1993). Antimicrobial functions of the plant hydrolases, chitinase and p-I ,3-glucanase. In “Mechanisms of Plant Defense Responses” (B. Fritig and M. Legrand, eds), pp. 391400. Kluwer, Dordrecht. Boller. T h . and Kende, H . (1979). Hydrolytic enzymes in the central vacuole of plant cells. Plunr Physiology 63, 1113-1132. Boller, T h . and Wiemken, A . (1986). Dynamics of vacuolar compartmentation. Annuul Review of Plant Physiology 37. 137-164. Uorochov. A . , Drori, A . , Tirosh, T., Borochov-Neori. H . and Mayak. S. (1990). Quantitative and qualitative changes in membrane proteins during petal senescence. Journal of Planr P/i.ysiology 136, 203-207. Rortlik. K.. Peisker, C. and Matile, Ph. (1990). A novel type of chlorophyll catabolite in senescent barley leaves. Jourriul o f ’ Plrrril Physiology 136, 161-164. Camp, P. J . . Huber, S . C . , Burke, J . J . and Moreland, D . E. (1982). Biochemical changes that occur during senescence of wheat leaves. Plunt Physiology 70, I64 1 - I 646. Carrasco, P. and Carbonell, J . (1988). Involvement of a neutral proteolytic activity in the senescence of unpollinated ovaries of Pisurri sutivum. Physiologin Pluntcirurn 72, 61(M16. Chang, K.-G.. Fcchner, G . H . and Schroeder. H . A . (1989). Anthocyanins in autumn leavcs o f quaking aspen in Colorado. Forest Scii7rice 35, 229-236. Charvat. I . and Esau, K. (1975). An ultrastructural study of acid phosphatase localization in Phaseoliis vulgaris xylem by use of an azo-dye method. Journal of‘ Cell Science 19, 543-569. Cheng, S . H . and Kao, C. H. (1984). The role of proteolytic enzymes in protein degradation during senescence of rice Icaves. Physiologiu Plantarurn 62, 231-237.

108

P. MATILE

Cronshaw, J. and Bentwood, B. J. (1977). Cytochemical localization of p-glycerophosphatase in the xylem of Nicotiana tabacum and Pisum sativum. Biology of the Cell 28, 65-74. Dhindsa, R. S., Plumb-Dhindsa, P. and Thorpe, T. A. (1981). Leaf senescence: correlated with increased levels of membrane permeability and lipid peroxidation, and decreased levels of superoxide dismutase and catalase. Journal of Experimental Botany 32, 93-101. Duggelin, T., Schellenberg, M., Bortlik, K. and Matile, Ph. (1988). Vacuolar location of lipofuscin- and proline-like compounds in senescent barley leaves. Journal of Plant Physiology 133, 492-497. Feller, U. (1986). Proteolytic enzymes in relation to leaf senescence. In “Plant Proteolytic Enzymes” (M. J. Dalling, ed.), Vol. 11, pp. 49-68. CRC Press, Boca Raton. Fineran, B. A. (1983). Differentiation of non-articulated laticifers in poinsettia (Euphorbia pulcherrima Willd.). Annals of Botany 52, 279-293. Fobel, M., Lynch, D. V. and Thompson, J. E. (1987). Membrane deterioration in senescing carnation flowers. Coordinated effects of phospholipid degradation and the action of membraneous lipoxygenase. Plant Physiology 85, 204-211. 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, 29S325. Gahan, P. B. and Maple, A. J. (1966). The behaviour of lysosome-like particles during cell differentiation. Journal of Experimental Botany 17, 151-155. Gartner, P.-J. and Nagl, W. (1980). Acid phosphatase activity in plastids (plastolysosomes) of senescing embryo-suspensor cells. Planta 149, 341-349. Ginsburg, S. and Matile, Ph. (1993). Identification of catabolites of chlorophyllporphyrin in senescent rape cotyledons. Plant Physiology 102, 521-527. Ginsburg, S . , Schellenberg, M. and Matile, Ph. (1994). Cleavage of chlorophyllporphyrin. Requirement for reduced ferredoxin and oxygen. Plant Physiology 105, 545-554. Grover, A., Sabat, S. C. and Mohanty, P. (1987). Does the loss of leaf chlorophyll during senescence of primary wheat leaf arise due to loss in chloroplast number or chlorophyll content? Biochemie und Physiologie der Pj7anzen 182, 481484. Heck, U., Martinoia, E. and Matile, Ph. (1981). Subcellular localization of acid proteinase in barley mesophyll protoplasts. Planta 151, 198-200. Herman, E. M., Baumgartner, B. and Chrispeels, M. J. (1981). Uptake and apparent digestion of cytoplasmic organelles by protein bodies (protein storage vacuoles) in mung bean cotyledons. European Journal of Cell Biology 24, 226-235. Heslop-Hamson, Y. and Heslop-Harrison, J. (1981). The digestive glands of Pinguicula: structure and cytochemistry. Annals of Botany 47, 293-319. Hinder, B . , Schellenberg, M., Rodoni, S., Ginsburg, S . , Vogt, E., Martinoia, E., Matile, P. and Hortensteiner, S. (1996). How plants dispose of chlorophyll catabolites: Directly energized uptake of tetrapyrrolic breakdown products into isolated vacuoles. Journal of Biological Chemistry 271, 27 233-27 236. Hortensteiner, S., Martinoia, E. and Amrhein, N. (1992). Reappearance of hydrolytic enzymes and tonoplast proteins in the regenerated vacuole of evacuolated protoplasts. Planta 187, 113-121. Hortensteiner, S., Vicentini, F. and Matile, Ph. (1995). Chlorophyll breakdown in senescent leaves: enzymic cleavage of phaeophorbide a in vitro. New Phytology 129, 237-246.


109

Huffaker, R. C. (1990). Proteolytic activity during senescence of plants. New Phytology 116, 199-231. Iten, W. and Matile, Ph. (1970). Role of chitinase and other lysosomal enzymes of Coprinus lagopus in the autolysis of fruiting bodies. Journal of General Microbiology 61, 301-309. Joel, D. M. and Fahn, A. (1980). Ultrastructure of resin ducts of Mangifera indica L. (Anacardiaceae). Differentiation and senescence of shoot ducts. Annals of Botany 46, 225-233. Klerk, H., Tophof, S. and Van Loon, L. C. (1992). Synthesis of proteins during the development of the first leaf of oat (Avena sativa). Physiologia Plantarum 85, 595-605. Krautler, B., Jaun, B., Bortlik, K., Schellenberg, M. and Matile, Ph. (1991). On the enigma of chlorophyll degradation: the constitution of a secoporphinoid catabolite. Helvetica Chimica Acta 30, 1315-1318. Lai, V. and Srivastava, L. M. (1976). Nuclear changes during the differentiation of xylem vessel elements. Cytobiologie 2, 220-243. Lamattina, L., Pont-Lezica, R. and Conde, R. D. (1985). Protein metabolism in senescing wheat leaves. Determination of synthesis and degradation rates and their effects on protein loss. Plant Physiology 77, 587-590. Leigh, R. A. (1979). Do plant vacuoles degrade cytoplasmic components? Trends in Biochemical Science 4, 37-38. Mae, T., Kai, N., Makino, A. and Ohira, K. (1984). Relation between ribulose bisphosphate carboxylase and chloroplast number in naturally senescing primary leaves of wheat. Plant Cell Physiology 25, 333-336. Martin, C. and Thimann, K. V. (1972). The role of protein synthesis in the senescence of leaves. I. The formation of protease. Plant Physiology 49, 64-71. Martin, S. J . , Green, D. R. and Cotter, T. G. (1994). Dicing with death: dissecting the components of the apoptosis machinery. Trends in Biochemical Science 19, 26-30. Martinoia, E. (1982). Abbau von Chloroplasten in alternden Gerstenmesophyllzellen. Ph.D. Thesis No. 7148, ETH Zurich. Martinoia, E., Heck, U., Dalling, M. J. and Matile, Ph. (1983). Changes in chloroplast number and chloroplast constitutents in senescing barley leaves. Biochemie und Physiologie der PfIanzen 178, 147-155. Marty, F. (1978). Cytochemical studies on GERL, provacuoles, and vacuoles in root meristematic cells of Euphorbia. Proceedings of the National Academy of Sciences of the USA 75, 852-856. Marty, F., Branton, D. and Leigh, R. A. (1980). Plant vacuoles. In “The Biochemistry of Plants” (N. E . Tolbert ed.), Vol. 1, pp. 625-658. Academic Press, London. Matile, Ph. (1975). “The Lytic Compartment of Plant Cells.” Cell Biology Monographs, Vol. 1. Springer Verlag, Berlin. Matile, Ph. (1992). Chloroplast senescence. In “Crop Phorosynthesis: Spatial and Temporal Determinants” (N. R. Baker and H. Thomas, eds), pp. 413-440. Elsevier, Amsterdam. Matile, Ph. (1994). Fluorescent idioblasts in autumn leaves of Ginkgo biloba. Botanica Helvetica 104, 87-92. Matile, Ph. and Winkenbach, F. (1971). Function of lysosomes and lysosomal enzymes in the senescing corolla of the morning glory (Ipomoea pupurea). Journal of Experimental Botany 22, 759-771. Matile, Ph., Ginsburg, S., Schellenberg, M. and Thomas, H. (1988). Catabolites of chlorophyll in senescing barley leaves are localized in the vacuoles of mesophyll

110

P. MATILE

cells. Proceedings of the National Academy of Sciences of the USA 85, 9529-9532. Matile, Ph., Schellenberg, M. and Peisker, C. (1992a). Production and release of a chlorophyll catabolite in isolated senescent chloroplasts. Planta 187, 230235. Matile, Ph., Flach, B. and Eller, B. M. (1992b). Spectral optical properties, pigments and optical brighteners in autumn leaves of Ginkgo biloba L. Botanica Acta 105, 13-17. Mei, H. S. and Thimann, K. V. (1984). The relation between nitrogen deficiency and leaf senescence. Plant Physiology 62, 157-161. Miller, B. L. and Huffaker, R. C. (1985). Differential induction of endoproteinases during senescence of attached and detached barley leaves. Plant Physiology 78, 442-446. Mitsuhashi, W., Crafts-Brandner, S. J . and Feller, U. (1992). Ribulose-1.5-bisphosphate carboxylase/oxygenase degradation in isolated pea chloroplasts incubated in the light or in the dark. Journal of Plant Physiology 139, 635-658. Mittler, R. and Lam, E. (1995). In situ detection of nDNA fragmentation during the differentiation of tracheary elements in higher plants. Plant Physiology 108, 489-493. Muhlecker, W., Krautler, B., Ginsburg, S. and Matile, Ph. (1993). Breakdown of chlorophyll: the constitution of a secoporphinoid chlorophyll catabolite from senescent rape leaves. Helvetica Chimica Acta 76, 29762980, Miihlecker, W. and Krautler, B. (1996). Breakdown of chlorophyll: constitution of nonfluorescing chlorophyll-catabolites from senescent cotyledons of the dicot rape. Plant Physiology and Biochemistry 34, 61-75. Nagl, W. (1976). Ultrastructural and developmental aspects of autolysis in embryosuspensors. Berichte der Deutschen Botanischen Gesellschaft 89, 301-31 1 . Ogawa, S. (1982). Disintegration of chloroplasts during zygote formation in Spirogyra verriculosa. Botanical Magazine of Tokyo 95, 249-260. Paliyath, G. and Droillard, M. J . (1992). The mechanism of membrane deterioration and disassembly during senescence. Plant Physiology and Biochemistry 30, 789-8 12. Pauls, K. P. and Thompson, J . E. (1984). Evidence for the accumulation of peroxidized lipids i n membranes of senescing cotyledons. Plant Physiology 75, 1152-1157. Peisker, C. (1991). Chlorophyllabbau in ver ilbenden Gerstenbliittern: Entwicklung und Anwendung einer Methode zur I C-Markierung von Chlorophyll. Ph.D. Thesis, University of Zurich. Peisker, C., Thomas, H., Keller, F. and Matile, Ph. (1990). Radiolabelling of chlorophyll for studies of catabolism. Journal of Plant Physiology 136, 544-549. Peterson, L. W. and Huffaker, R. C. (1975). Loss of ribulose 1,5-diphosphate carboxylase and increase in proteolytic activity during senescence of detached primary barley leaves. Plant Physiology 55, 1009-1015. Reznickova, S. A. and Dickinson, H. G. (1982). Ultrastructural aspects of storage lipid mobilization in the tapetum of Lilium hybrida var. enchantment. Planta 155, 400-408. Schellenberg, M., Matile, Ph. and Thomas, H . (1993). Production of a presumptive chlorophyll catabolite in vitro: requirement for reduced ferredoxin. PIanta 191, 4 17-420. Schennen, A. and Holzl, J . (1986). 6-Hydroxykynureninsaure, die erste N-haltige

9

THE VACUOLE A N D CELL SENESCENCE

111

Verbindung aus den RlCttern von Ginkgo hiloha. Planta Medica 52. 235236. Smart, C. M.,Hosken, S. E.. Thomas, H . , Greaves, J . A., Blair, B. C. and Schuch, W. (1995). The timing of maize leaf senescence and characterization of senescence-related cDNAs. Physiologia Plantarum 93. 673482. Smith, M. T., Saks. Y. and Van Staden. J . (1992). Ultrastructural changes in the petals of senescing flowers of Dianthtcs cmyophyllus L. Annals of Botany 69, 277-285. Srivastava. L. M. and Singh, A . P. (1972). Certain aspects of xylem differentiation in corn. Canadian Journal of’ Botany 50, 1795-1804. Suttle, J . C. and Kende, H . (1980). Ethylene action and loss of membrane integrity during petal senescence in Tradescantia. Plant Physiology 65, 1067-1072. Thelen, M. P. and Northcote, D. H. (1989). Identification and purification of a nuclease from Zinnia elegans L. : a potential molecular marker for xylogcnesis. PIantu 179, 181-195. Thomas. H . (1978). Enzymes of nitrogen metabolism in detached leaves of Lolium temulentum during senescence. Planta 142, IGl-169. Thomas, H. (1982). Leaf senescence in a non-yellowing mutant of Fesruca pratensis. 1. Chloroplast membrane polypeptides. Plunta 154, 212-218. Thomas, H . and Hilditch. P. (1987). Metabolism of thylakoid membrane proteins during foliar senescence. In “Plant Senescence: Its Biochemistry and Physiology” (W. W. ‘Thornson, E. A. Nothnagel and R . C. Huffaker. eds), pp. 114-122. The American Society of Plant Physiologists. Rockville. Van der Valk, H. C.P. M . and Van Loon, L. C. (1988). Subcellular localization of proteases in developing leaves of oats (Avena scrtiva L..). Plant Physiology 87, 536-542. Van der Wilden. W., Segers, J . H. L. and Chrispeels, M . J . (1983). Cell walls of Phuseo1u.s vulgaris leaves contain the azocoll-digesting proteinase. Plant Physio1og.y 73, 576.578. Vaughn. K . C. and Duke, S. 0. (1981). Evaginations from the plastid envelope, a method f o r transfer of substances from plastid to vacuole. Cytohios 32, 89-96, Vicentini, F . , Hdrtensteiner. S., Schcllenbcrg. M . , Thomas, H . and Matile, Ph. (1995). Chlorophyll breakdown in senescent leaves: Identification of the lesion in a stay-green genotype of Festucu pratensis. New Phyrologist 129, 247-252. Wagner. G . J . (1982). Compartmentation in plant cells: the role of the vacuole. Recent Adiunces in Phytochernisrry 16, 1 4 5 . Wardley, T. M..Bhalla, P. L. and Dalling, M. J . (1984). Changes in the number and composition of chloroplasts during senescence of mesophyll cells of attached and detached primary leaves of wheat (Triricum aestivum L.). Plant Physiology 7 5 , 42 1-424. Wittenbach, V. A . (1978). Breakdown of ribulose biphosphate carboxylase and change in proteolytic activity during dark-induced senescence of wheat seedlings. Plant Physiology 62, 604608. Wittenbach, V. A , , Lin, W. and Hebert. R. R . (1982). Vacuolar localization of proteases and degradation of chloroplasts in mesophyll protoplasts from senescing primary wheat leaves. Plunt Physiology 69, 98-102. Wodzicki, T. J . and Brown. C. L. (lY73). Organization and breakdown of the protoplast during maturation of pine tracheids. American Jourttal of Botany 60, 63 1-640. Wodzicki. T. J . and Humphreys, W. J . (1073). Maturing pine tracheids. Organization of intravacuolar cytoplasm. Journul of Cell Biology 56, 263-265.

112

P. MATILE

Wodzicki, T. J. and Humphreys, W. J. (1974). Cytodifferentiation of maturing pine tracheids: the final stage. Tissue and Cell 4, 525-528. Yarnane, K . , Abiru, S., Fujishige, N., Sakiyama, R. and Ogata, R. (1993). Export of soluble sugars and increase in membrane permeability of Gladiolus florets during senescence. Journal of the Japanese Society of Horticultural Science 62, 575-580. Ye, Z.-H. and Varner, J. E. (1993). Gene expression patterns associated with in vitro tracheary element formation from isolated mesophyll cells of Zinnia elegans. Plant Physiology 103, 805-813.

Protein Bodies: Storage Vacuoles in Seeds

G. GALILI

Department of Plant Genetics, The Weizmann Institute of Science, Rehovot 76100, Israel E. M.HERMAN

Plant Molecular Biology Laboratory, United States Department of Agriculture, Agricultural Research Service, Beltsville, MD 20705, USA

I.

Introduction

............. ....... ....................................... .............. ,

11. Ontogeny of PSVs .....................................................................

113 114

111. The Golgi Apparatus Mediates the Deposition of PSV Constituents in Dicotyledonous Plants ..... ....... .. ... .. ..., ... ........... ., ,.. ..... ............ .. 116 IV. V. v1.

Transport of Storage Proteins to Vacuoles in Monocotyledonous Plants ................................................................................... 120 Developmental Regulation of the PSV Tonoplast

.. ........ .... ...... . ....

Enzyme Composition of PSVs ..................................................

123 127

VII . Diversity of Vacuolar Storage Proteins and Enzymes .......... . .. .. . .... 128 A. Globulin Storage Proteins ................ ......... .. .................. ..... 128 B. Prolamin Storage Proteins ................................................. 129 VIII.

Assembly and Processing of Storage Proteins .............................. 129 A. Assembly of the Storage Proteins Within the ER is Assisted by Molecular Chaperones ...... ......... ................. ....... ... ............ 130 B. Proteolytic Processing of Prolegumins inside Vacuoles ............ 132

Advances in Botanical Research Vol. 25 incorporating Advances in Plant Pathology

ISBN 012-005925-8

Copyright @ 1997 Academic Press Limited All rights of reproduction in any form reserved

114

G . GALILl and E. M. HERMAN

IX. Expression of Storage Protein Genes in Transgenic Plants ............. 133 Acknowledgements ................................................................. 135 References ............................................................................ 135

I. INTRODUCTION Seeds of many plant species store reserve proteins in protein storage vacuoles (PSVs). The questions of the ontogeny of seed PSVs parallel the questions on the origin of the central vacuole in meristematic cells (for a review, see Marty, this volume). Differing interpretations on the direct involvement of the endoplasmic reticulum (ER) in vacuole ontogeny and the alternative mechanism of formation mediated by the Golgi apparatus have dominated the thinking of investigators studying the origins of the central vacuole as well as PSVs. Understanding the ontogeny of PSVs is further complicated by the substantial differences that have been demonstrated in the mechanisms of protein trafficking to the vacuole and PSV that are used in cereals and many of the dicotyledonous plants. This has shown that there are multiple mechanisms of protein sequestration in the PSVs in seeds, as in vegetative vacuoles (for a review and discussion, see Herman, 1994). In this review, we have detailed the process and developmental dynamics of PSV formation in plants. The primary purpose of PSV formation is to sequester nitrogen in the form of storage proteins for utilization during germination. Deposition of the storage proteins in protein bodies inside vacuoles has been shown to be a complex process involving some of the mechanisms of cotranslational glycosylation and glycan modification, precursor processing, oligomer formation and protein assembly and aggregation as well as more than one mechanism of protein trafficking. We have reviewed and contrasted the various processes that result in the deposition of storage proteins inside vacuoles in cereals and dicotyledonous seeds.

11. ONTOGENY OF PSVs Seed storage cells consist of two functional types: (i) cells that survive into germination, including both the developing embryo and some endosperm or aleurone cells of dicotyledonous seeds and the aleurone cells of cereals; and (ii) cells that do not continue into germination, such as starchy endosperm cells of cereal grains. Most studies on the ontogeny of PSVs have been performed on dicotyledonous embryonic cells. PSVs are transiently differentiated vacuoles that appear to originate from the subdivision of the central vacuole during seed maturation and resume the morphology of the central vacuole during seed germination. This process appears to occur coordinately with the respective deposition and mobilization of storage proteins. T h e ontogeny of the PSVs is still the subject of debate. In many dicotyledonous

PROTEIN BODIES: STORAGE VACUOLES I N SEEDS

11s

seeds, immature embryonic cells closely resemble vegetativc cells by possessing a central vacuole that is replaced during the later stages of seed maturation by numerous protein-tilled PSVs. How the transition from a single “empty” central vacuole to numerous protein-filled vacuoles occurs has been the subject of much research and continuing debate. Light and electron microscopy investigations have indicated that the central vacuole is subdivided coordinately with the deposition of storage proteins, producing numerous protein-filled vacuoles (Craig et a / . , 1979. 1980). The process of forming a protein-tilled vacuole appears to occur by several different mechanisms. For instance in pea seeds, electron microscopy and immunocytochemistry with anti-storage protein antibodies have shown that storage proteins are sequestered and packaged into protein aggregates on thc inner surface of the tonoplast that is budded off the vacuole as a mature protein-tilled PSV (Hoh ef (11.. 199s). The various seed storage proteins within the PSV matrix appear to stratify, resulting i n different domains enriched in each protein (Hinz et al., 1995). In other legumes as well as other dicotyledonous plants, the subdivision of the vacuole into incipient PSVs appears to precede the deposition of most of the storage proteins (Fig. l a ) . This results in numerous small partially tilled PSVs that are gradually filled with storage proteins, resulting in mature, protein-filled PSVs (Fig. 1 b). Because this later mechanism for formation of PSVs appears to occur over a protracted period of time, this allows for additional complexity and subdomains to be established within t h e protein matrix of the PSV. These subdomains can include not only amorphous deposits of storage proteins, but also a protein crystalloid that is assembled during the course of PSV maturation. These PSVs also often contain deposits of phytin (inositol polyphosphate) and other inclusions derived from autophagy of the cytosolic constituents. The interpretation that PSVs arise from subdivision of the embryonic vegetative vacuole is not universally held. Prominent among those with an alternative interpretation are Robinson and co-workers (Robinson et ul., 1995), who have suggested that PSVs might constitute an entirely new vacuole that is synthesized de n o w and replaces the pre-existing central vacuole. Among these data are micrographs that compellingly illustrate the presence of two distinctly different and coexisting vacuoles in storage parenchyma cells of maturing pea seeds. One of the vacuole types appears to be incipient PSVs containing dense aggregations of storage proteins that appear to be in the process of releasing mature protein-filled PSVs by budding. The othcr vacuole type retains the “empty” appearance of the pre-existing embryonic central vacuole (Robinson et of., 1995). The isolated membranes of the PSV and its vegetative vacuole precursor possess different intrinsic tonoplast polypeptides and densities in equilibrium sucrose gradients (Hoh et af. 1995). This has been interpreted to indicate that the vegetative vacuole and PSV are entirely different structures having independent origins, with the PSV

116

G . GALILI and E. M. HERMAN

replacing the vegetative vacuole as seed maturation progresses. Indirect support for the hypothesis of Robinson and associates has recently been obtained by confocal microscopy, using antibodies against several vacuolar markers, including a- and y-TIP (Paris et al., 1996). This study showed that a given vegetative plant cell may contain two functionally distinct vacuoles, which differ both in their tonoplast and inner resident proteins. How these two distinct models of vegetative vacuole transformation into the PSV and the PSV being formed independently and replacing the vegetative vacuole can be reconciled remains to be determined. It is possible that there is more than one general pattern of PSV ontogeny, and the process used to form the PSV might be developmentally or environmentally regulated and therefore appear to be quite different in the various seeds used as experimental material or the conditions of plant growth. The situation of PSV ontogeny and storage protein deposition is further complicated in monocotyledonous seeds such as wheat. During early and intermediate stages of wheat grain development, the storage proteins are first deposited into small, vegetative-like vacuoles, which then fuse to form large central vacuoles containing large protein body inclusions (Fig. 2). At later stages of seed maturation, no vacuoles can be visualized microscopically, and the protein bodies remain in the cytoplasm. Moreover, wheat, and perhaps other cereals too, exhibit a unique mechanism for the sequestering of storage proteins into the vacuole, which will be discussed in detail later on.

111. THE GOLGI APPARATUS MEDIATES THE DEPOSITION OF PSV CONSTITUENTS IN DICOTYLEDONOUS PLANTS The current models for PSV formation in dicotyledonous plants are based on the concept that protein deposition occurs as the consequence a Golgi-mediated secretory process that delivers precursor or pro-storage proteins to the incipient PSV. There is much structural and biochemical evidence to support a Golgi apparatus-mediated model for PSV ontogeny. Structural evidence showing the secretion of dense protein-containing vesicles from the Golgi (see Fig. 3 for example) that appear to contain storage proteins dates from the earliest period of the availability of ultrastructural

Fig. 1. The ontogeny of PSVs from the central vegetative vacuole in maturing soya bean cotyledon cells. (A) Electron micrograph showing a portion of an immature cotyledon cell in which the vegetative vacuole is in the process of subdividing coordinately with the deposition of storage proteins. (B) Electron micrograph showing a portion of a late maturation cotyledon cell in which the subdivision of the vacuole is complete and the individual PSVs contain dense deposits of storage proteins. CW, cell wall; ER, endoplasmic reticulum; G , Golgi; PB, protein body.


117

118

G. GALlLl and E. M. HERMAN

Fig. 2. Electron micrographs from developing wheat endosperm cells at 13 days after anthesis. (a) Small vacuoles containing single or several small protein bodies. (h) Large central vacuoles containing large protein bodies. Small protein bodies inside small vacuoles join the central vacuoles (arrowheads). CW, cell wall; S, starch; PB, protein body; V, vacuole; I, inclusion body. Bars: (a) 1.8. pm; (b) 5 pm.(Reproduced in part from Galili el al. (1993) and Levanony et al. (1992).)

techniques. In the 1980s with the advent of electron microscopic immunocytochemical techniques, many investigators (Craig and Goodchild, 1984a; Herman and Shannon, 1984a,b, 1985; Greenwood and Chrispeels, 1985) published micrographs showing a clear association of storage proteins, lectins and some enzymes with the ER, Golgi apparatus and differentiating PSV (see Figs 3 and 4 as examples). This has been interpreted to indicate that storage proteins and other PSV polypeptides are delivered to the PSV by the endomembrane secretory route, resulting in specific immunolabelling of each of the constituent compartments of the endomembrane system. Additional support for the functional role of the Golgi in packaging and processing storage proteins and lectins results from biochemical investigations of the processing of the cotranslationally attached N-glycans found o n many vacuolar glycoproteins (Chrispeels, 1983,1991; Faye et al., 1986,1989; Sturm et al., 1987). There is considerable evidence in animal cells that the processing o f the high-mannose glycan side-chains by trimming the mannosyl sugars and


Fig. 3.

119

Role of thz Golgi apparatus in storage protein transport to the vacuole.

( a ) Electron micrograph showing Golgi apparatus with associated secretion vesicles in a midrnaturation soya bean cotyledon cell. !ndividual Golgi vesicles appear t C J be

~e to the PSV\. ( h ) An electron progressivelv secreted t o carrv precursor ~ t o r u protcins micrograph showing the apparent fusion of ;I deiisc secretion vesicle t o thc I'SV in 21 midmaturation soya bean cotyledon cell. bIR. cndoplasmic reticulum; Ci. C i o l K i ; OB. oil body; PSV, protein storage vacuole.

120


attaching other sugars to form complex glycans occurs in the Golgi apparatus (Chrispeels, 1983, 1991). Several of the vacuole storage proteins and lectins are processed by the addition of sugars such as xylose, fucose and galactose to the N-glycans, resulting in the formation of complex glycans (Sturm et al., 1988). Electron microscopic immunocytochemistry of soya bean aleurone cells shows that the complex glycan proteins accumulated in the PSV initially acquire the xylose in the medial-trans-Golgi (Yaklich and Herman, 1995). The glycosyl-transferases that mediate the addition of the sugars to the N-glycans have been shown to be present in Golgi fractions purified from maturing seeds (Sturm et al., 1987; Moore et al., 1991). Additional evidence for the function of the Golgi in the trafficking of storage proteins has been obtained through the use of inhibitors that prevent correct trafficking through the Golgi. Monensin is an ionophore that affects the trans-Golgi, preventing secretion in animal cells by interfering with the proper assembly of Golgi secretion vesicles. Monensin treatment of maturing pea seeds causes storage protein trafficking to be redirected away from the vacuole to the cell surface, resulting in secretion of the vacuolar proteins into extracellular space (Craig and Goodchild, 1984b). The new membrane of the PSV is presumably added to the differentiating tonoplast as the conversion from embryonic central vacuole to PSV commences. This membrane is presumed to enclose the precursor storage proteins secreted from the Golgi apparatus. Few studies have examined the trafficking and functional role of the Golgi apparatus in PSV membrane ontogeny. Melroy and Herman (1991) localized a specific PSV tonoplast protein, TIP or seed-specific aquaporin (see Chrispeels et al., this volume) to the Golgi apparatus using electron microscopic immunocytochemistry , thereby providing direct evidence for the role of the Golgi.

IV. TRANSPORT OF STORAGE PROTEINS TO VACUOLES IN MONOCOTYLEDONOUS PLANTS Like dicotyledonous seeds, in cereal grains the Golgi apparatus participates in the vacuolar deposition of storage proteins and ontogeny of the tonoplast membrane. Electron microscopy studies (Bechtel et al., 1989; Levanony et a l . , 1992; H. Levanony and G. Galili, unpublished findings) also suggest that the E R contributes significantly to the ontogeny of the tonoplast. These studies have shown connections between E R membranes and electron-lucent vesicles (apparently provacuoles) in endosperm cells of maturing wheat grains. These vesicles, which form inside ER-enriched areas or around protein bodies in the cytoplasm, also contain vacuolar acid phosphatases. During grain maturation, the provacuoles fuse, forming large, vegetative-type central vacuoles that occupy most of the cell volume (Levanony et al., 1992; Galili et al., 1993). Deposition of the storage proteins in vacuoles parallels

PROTEIN BODIES: STORAGE VACUOLES IN SEEDS

121

Fig. 4. Immunogold localization of the acid hydrolase cu-galactosidase in a maturing soya bean cotyledon cell. The immunogold particles label the ER, Golgi apparatus and PSV matrix, providing structural evidence of the progression of vacuolar matrix proteins through the endomembrane system for deposition in the PSV. ER, endoplasmic reticulum; G, Golgi; PSV, protein storage vacuole; SV, secretion vesicle.

the maturation of these organelles. The storage proteins are first deposited as inclusions in the small vacuoles. The small vacuoles appear to fuse and concurrently the storage protein aggregates inside them also fuse to form large protein inclusions (Levanony et al., 1992; Galili et al., 1993). Although the Golgi apparatus is involved in the transport of wheat prolamins to vacuoles (Fig. 5, right), these storage proteins do not progress to the vacuoles solely by endomembrane trafficking. Some prolamins initially aggregate within the E R lumen and bud from the E R as protein bodies, which consist of a prolamin matrix as well as additional E R luminal molecular chaperones that apparently assist in proper folding and assembly of the prolamins (Levanony et af., 1992; Rubin et al., 1992; Galili et al., 1993, 1995a,b). These protein bodies then appear to be internalized into vacuoles by a process analogous to autophagy (Levanony et al., 1992; Galili et al., 1993). Formation of protein bodies within the E R of wheat endosperm cells is not unexpected, because similar prolamin storage proteins from maize and

122

G . GALII.1 and E. M. HERMAN

Fig. 5. Schematic representation of the two different routes of wheat storage proteins to vacuoles. Right: the paradigm route including vesicular transport from the ER via the Golgi apparatus to vacuoles. Left: protein bodies formed within the ER become surrounded by electron-lucent vesicles (provacuoles) and are internalized into vacuoles by a process analogous to autophagy. (Reproduced with permission from Galili er ul. (Ic)O3).)

sorghum also accumulate in protein bodies within this organelle (Larkins and Hurknian, 1978; Shotwell and Larkins, 1989). However, in contrast to the protein bodies in wheat, the protein bodies of maize endosperm are not transported to the vacuole at any stage of grain development; they remain as cytoplasmic proteins, and some of them may remain within the E R . The autophagic internalization of wheat storage proteins into the small vacuoles seems to be initiated by the formation and attachment of the provacuoles (present abundantly in the endosperm cells) around the entire surface of the ER-derived protein bodies (Fig. 5 , left). The provacuoles then fuse with each other around the protein bodies, forming small vacuoles containing protein body inclusions. In many cases, the protein bodies inside the vacuole are surrounded by one or two incomplete membranes, in addition to the vacuolar membrane (Levanony et a [ . , 1992; Galili et al., 1993). These membranes appear to contain the tonoplast marker enzyme pyrophosphatase, supporting the notion that the direct internalization of the protein bodies from the E R into vacuoles occurs by an autophagy process (H. Levanony and G. Galili, unpublished results). Interestingly, the vacuoles in wheat

PliOrFlN 130DItS S I O K A G E VACUOI ES I N SELD\

123

endosperm are autophagic in general, since they also sequester membranes and other cytoplasmic material (Rechtt.1 pf d..1Y8Y). How are the provacuoles formed? As was discussed previously, electron microscopy studies suggest t h e y may devclop, a t least in part. directly from the ER. One possibility is that newly synthesized tonoplast membrane proteins assemble into functional componcnts already on the ER membrane and thus pump ions into the ER lumen, reducing its pH. Thus, newly synthesized vacuolar hydrolases may he activated in the lumen of the E R and digest its contents, resulting in the formation of electron-lucent vesicles. or provacuoles. Indeed, these vesicles wcre shown to contain the vacuolar enzyme acid phosphatase (Bechtel r t 01.. 1980). Further elucidation of the mechanism of the direct transport of storage proteins from the ER to vacuoles awaits detailed analysis o f the protein composition in the membrane o f the electron lucent vesicles, and the incomplete membranes surrounding the protein bodies. Direct transport of proteins from the 11K to the vacuole appears not to be restricted to plant cells. In hyperstimulated thyroid hormone-secreting cells of rats, secretion of thyroid-stimulating hormone is retarded. The hormone first accumulates in intracisternal granules within the E R and then is transported dircctly from the EK to the lysosome (an analogous organelle to plant vacuoles) for its disposal (Noda and Farquhar, 1992). Electron microscopy analysis suggested that this transport occurs by a specific pathway in which E R vesicles (part rough/part smooth cistcrna), containing the thyroid-stimulating hormone. are converted into lysosome-like structures (Noda and Farquhar, 1992). The extent o f divergence or similarity o f this pathway to the one transporting whcnt storage proteins from the ER to vacuoles has still to be studied.

V. DEVELOPMENTAL REGULATION OF THE PSV TONOPLAST Although subdivision of the vacuole in seeds of dicotyledonous plants is developmentally coordinated with the deposition of storage proteins, it is not yet known what signals induce the alteration of the vacuole. Subdivision appears to occur with the onset of storiige protein accumulation but whether the storage proteins are the signal inducing the subdivision is not known. This hypothesis is as yet untestable, because there are no known seed mutants in which all storage protein synthesis is repressed but vacuoles still mature normally. However, the ability of vegetative cells t o accumulate storage proteins without the subdivision o f the vacuole (for ii review, see Herman, 1994) certainly suggests that storage protein accumulation may not be the primary signal for the subdivision of thc vacuole into PSVs. It is far more likely that the signal and the mechanism tor vacuolar subdivision occurs at

124


the level of the tonoplast, because the PSV tonoplast is differentiated and possesses a set of polypeptides specific for late seed maturation (Johnson et al., 1989; Inoue et al., 1995), as well as an enriched sterol content (Herman et al., 1984). The relationship between the acquisition of embryonic tonoplast proteins and the differentiation of the central vacuole into PSVs is unclear. The first PSV-specific tonoplast polypeptide to be discovered was a-TIP (tonoplast intrinsic protein) (Johnson et al., 1989). TIP is a member of a large family of water channel proteins, or aquaporins, that facilitate the passage of water through the tonoplast and plasma membrane (reviewed by Chrispeels et al., this volume). The seed-specific a-TIP likely facilitates the passage of water into and/or out of the PSV during seed desiccation and rehydration. Chrispeels and co-workers have shown that the function of a-TIP is regulated by phosphorylation (Johnson and Chrispeels, 1992; Maurel et af., 1995), indicating that water flow through the PSV tonoplast may be physiologically regulated at the correct stages of development. a-TIP is apparently found in embryonic seed cells of diverse plant species (as examples, see Johnson et af., 1989; Maeshima et af., 1994), supporting the contention that it is essential to facilitate the passage of water during desiccation and rehydration. a-TIP is accumulated in the tonoplast of embryo, cotyledon and aleurone/endosperm cells (Johnson et al., 1989). Seed-specific a-TIP accumulation is not directly correlated with storage protein accumulation. In the soya bean, a-TIP is accumulated in the PSV membranes of storage parenchyma cells that accumulate vacuolar storage proteins, as well as in the tonoplast of cotyledon vascular bundle cells that do not accumulate storage proteins (Melroy and Herman, 1991). a-TIP is densely accumulated in the PSV tonoplast during the late stages of seed maturation after much of the storage protein accumulation has been completed (Melroy and Herman, 1991). The late maturation accumulation of a-TIP indicates it cannot be an inducing protein for the differentiation of the vacuole into PSVs. The vegetative vacuole precursor of the PSV in pea embryonic cells contains the homologue of y-TIP, which has been shown by Chrispeels and colleagues to be widely distributed in the tonoplast of cells of various vegetative organs (see Chrispeels et al., this volume). The y-TIP disappears as differentiation of the vegetative vacuole into PSVs occurs. Hoh et af. (1995) have used sucrose gradients to examine the distribution of two different isoforms of the TIP. They showed that the vegetative vacuoles of pea seeds initially possess the vegetative-type ?-TIP, which is replaced by seed-PSVspecific a-TIP as the seed matures. In isopycnic centrifugation studies, the vegetative vacuole equilibrates at about 30% sucrose, while mature PSV tonoplast equilibrates at -40% sucrose. This may be the consequence of increasing amounts of sterols accumulated in PSVs (Herman et al., 1984). A simple interpretation that the density of the membrane increases with the


125

acquisition of embryo-specific tonoplast constituents is confused by the presence of a membrane fraction, equilibrating at -22% sucrose at an intermediate development stage, that contains a-TIP. How the 22% sucrose membrane fraction functions as the precursor to the 40% sucrose membrane fraction, if indeed it does, remains to be elucidated. However, Hoh et af. (1995) have interpreted these results to indicate that the PSV membrane does not originate from the vegetative vacuole. Whether the exchange of y-TIP for a-TIP as PSV differentiation occurs proves that the PSV is synthesized de novo, as suggested by Hoh and colleagues, or whether there is a gradual insertion and turnover of the new and old membrane constituents, gradually altering the composition of the incipient PSV, as suggested by Melroy and Herman (1991), remains to be determined. In addition to a-TIP, there are other membrane proteins that appear to be specific for the embryonic PSV. Inoue et af. (1995) have shown that pumpkin PSV tonoplasts possess major 23, 27, 28, 32 and 73 kDa polypeptides. The 23 and 28 kDa proteins appear to be homologues of a-TIP. The 27 and 32 kDa proteins are synthesized as a common precursor that is post-translationally cleaved after asparagine. This cleavage is a common processing event in many seed storage proteins and some vacuolar enzymes, but this is the first instance where a membrane protein has been shown to be processed in this manner. The vegetative vacuole of immature pea cotyledon cells appears to be deficient in the two major vacuolar proton pumps, H + pyrophosphatase and V ATPase. As PSV differentiation occurs, both of the proton pumps are accumulated in the PSV tonoplast coordinately with a-TIP. a-TIP is rapidly removed from the PSV membrane during germination. During this time the storage proteins are mobilized, and coordinately the PSVs fuse to reform the central vegetative vacuole. The process of vacuole reformation occurs in seeds whether they are terminally senescent during germination or whether the cotyledons regreen and persist for some time. The removal of a-TIP from the PSV membrane appears to occur by autophagy of the tonoplast into the PSV, removing the embryonic tonoplast by progressive internalization and degradation (Fig. 6; Melroy and Herman, 1991). The replacement of tonoplast with new membrane material possessing a different polypeptide composition results in the alteration of the PSV membrane by progressive dilution converting the embryonic membrane into a central vacuole membrane. The 27 and 32 kDa proteins are rapidly removed from the PSV during germination, and autophagic vesicles presented in micrographs by Inoue et af. (1995) appear to indicate that the removal of this protein from the PSV tonoplast occurs by the same mechanism as that proposed for destruction of a-TIP (Melroy and Herman, 1991). The dynamic changes that occur in reforming the central vacuole from the pre-existing PSVs indicate that new tonoplast proteins replace those removed by autophagy. Maeshima et af. (1994) investigated this process in pumpkin

126


Fig. 6. The autophagic sequestration of tonoplast into the PSV. Immunogold labelling of @-TIPin a PSV of B germinating soya bean cotyledon detects both the tonoplast and an autophagic vesicle that has been sequestered within thc PSV matrix. The progressive autophagic internalization and disposal of the seed-specific tonoplast into the PSV apparently results in the transformation of the PSVs into the central vegetative vacuole during seed germination. PSV. protein storage vacuole.

cotyledons. They found that PSVs are deficient in the two major proton pumps used to maintain the acidic environment of the vacuolar sap. As germination progresses and the PSVs fuse to reform the central vacuole, the H+-pyrophosphatase and H +-ATPase are accumulated in the tonoplast. Coordinately, the seed-specific u-TIP (28 kDa) is lost and replaced by a homologue of the vegetative or y-TIP (23 kDa). yT1P has been shown to be widely distributed in vegetative organs. Storage proteins are mobilized by the coordinated actions of a de nuvo synthesized thiol protease (Briaty et id.,1970; Baumgartner and Chrispeels, 1977; Baumgartner et ul., 1978; Rogers et al.. 1985; Mitsuashi and Minamikawa. 1989; Koehler and Ho, 1990; Watanabe et u / . , 1991; Tanaka et al., 1993) and carboxypeptidase that require an acidic environment. The accumulation of proton pumps in the PSVvacuole tonoplast, coordinately with the synthesis of the proteases, appears to be well designed to produce an acidic hydrolytic environment that will facilitate the rapid mobilization of storage protein reserves.

PROTEIN BODIES: STORAGE VACCJOLES IN SEEDS

VI.

127

ENZYME COMPOSITION OF PSVs

In addition to storage proteins, PSVs contain an array of hydrolytic enzymes during both seed maturation and germination. Maturing legume seed PSVs have been shown to possess activities of RNAase (Chappell ef al., 1980; Van der Wilden et a l . , 1980), glycosidase (Van der Wilden et al., 1980), phospholipase D (Herman and Chrispeels. 1980; Van der Wilden et al., 1980) and carboxypeptidase (Van der Wilden et a l . , 1980). PSVs also can contain abundant lectin proteins (Etzler, 1985). Lectins are characterized by their ability to bind to, but not modify, specific glycosyl residues. Whether lectins constitute another class of storage protein or are defence proteins to protect the seed is an active debate. Because PSVs accumulate storage proteins that have evolved to be substrates for proteases during germination, until recently it was widely assumed that PSVs do not possess significant levels of proteases during seed maturation. However, Kalinski et al. (1992) cloned and studied a seed-specific member of the thiol protease papain superfamily that is accumulated during soya bean seed maturation. Nielsen and co-workers (1995) identified additional proteolytic activities in developing soya bean seeds that they propose are sequestered in the vacuolc (Scott et al., 1992). Many different seeds, including the castor bean, soya bean and jackbean, accumulate a protease that processes precursor storage proteins, as well as some lectins and enzymes, on the C-terminal side of exposed asparagine residues, creating multipolypeptide chain proteins (Hara-Nishamura et al., 1995). The processing enzyme is developmentally regulated to be accuniulated within the PSV at the same stage as its substrates. This processing activity may serve to activate some PSV proteins, including the jackbean lectin concanvalin A (Bowles et ul., 1986), soya bean P34 (a cysteine protease; Kalinski et al., 1992) and a pumpkin tonoplast protein (Hara-Nishamura et al., 1995). During germination, additional enzyme activities are accumulated in the PSV. Most prominent among these are germination-specific papain superfamily thiol proteases that function to mobilize the storage proteins. Proteases of this family have been characterized in dicotyledonous and monocotyledonous seeds (Baumgartner and Chrispeels, 1977; Baumgartner et al., 1978; Rogers et a l . , 1985; Mitsuashi and Minamikawa, 1989; Koehler and Ho, 1990; Watanabe et al., 1991; Yamauchi el al., 1992; Tanaka et d., 1993). These enzymes are synthesized de n o w , accumulated in the PSV and, after activation by post-translational processing, rapidly mobilize the storage proteins, producing PSV with an increasingly -‘empty” appearance in light and electron microscopy. Simultaneously with the mobilization of the storage proteins, the individual PSVs fuse to reform a central vacuole. In electron microscopy observations, the fusing PSVs appear to be highly dynamic with membrane internalizations. These vacuoles also contain both tonoplast and cytoplasmic organelles that appear t o be in the process of degradation by

128

0. GALILI and E. M. HERMAN

the action of PSV hydrolytic enzymes. Apart from the proteases involved in storage protein mobilization, other PSV enzymes have received scant attention from investigators, although increases in activity of some of these enzymes, for instance RNAase, suggests that de novo synthesis supplements the pre-existing enzyme activity remaining from seed maturation.

VII. DIVERSITY OF VACUOLAR STORAGE PROTEINS AND ENZYMES Plant seeds consist of two major classes of proteins: (i) proteins that function during seed development (i.e. metabolic proteins, house-keeping proteins, defence proteins, etc.); and (ii) storage proteins that function as stores of nitrogen and energy during germination. In mature seeds, the storage proteins account for about 80-90% of total seed proteins and are broadly defined as proteins that are unique to seeds, are deposited in protein bodies, and are efficiently degraded and mobilized in the germinating embryo (Spencer and Higgins, 1979). Seed proteins are also classified based on their solubility into four major classes: albumins (water-soluble), globulins (salt-soluble), prolamins (soluble in alcohol-water mixtures) and glutelins (soluble in dilute acids or bases; in some plant species they are called glutenins) (Osborne, 1924). With this classification the major storage proteins in many dicotyledonous plants and some cereals (rice and oat) are globulins. In other cereals, such as wheat, barley and maize, the major storage proteins are prolamins and glutenins. Recent molecular studies have shown that some of the glutenin storage proteins, although insoluble in alcohol-water mixtures, are structurally similar to the prolamin storage proteins. The lack of solubility of these glutenins in alcohol-water mixtures is due to polymerization by intermolecular disulfide bonds, and they can be rendered alcohol-soluble upon addition of reducing agents. The characterization, classification and structure of the various vacuolar storage proteins have been described in a series of excellent reviews (Shewry et al., 1989; Shotwell and Larkins, 1989; Shewry and Tatham, 1990; Shewry, 1995) and will only briefly be described here. A . GLOBULIN STORAGE PROTEINS

The major globulin storage proteins fall into three major classes collectively called legumins, vicilins and lectins. The legumins are non-glycosylated proteins which assemble into hexamers having a sedimentation coefficient of 11-12s. Each legumin protein consists of two polypeptides, one acidic and one basic, that are linked by an intermolecular disulfide bond. The acidic and basic subunits are derived from the proteolytic cleavage of precursor prolegumins (Shotwell and Larkins, 1989; Vitale and Bollini, 1995). The vicilins are a specific group of proteins, which separate as trimers with a


129

sedimentation coefficient of -7s. These proteins are not cleaved proteolytically, but some of them are post-translationally glycosylated (see later). B.

PROLAMIN SI'ORAGE PROTEINS

The major prolamins stored in vacuoles are classified into three groups as sulfur (S)-poor, S-rich and high molecular weight (HMW) prolamins. All of these proteins are characterized by their content of a domain that is built of small amino acid sequence repeats, rich in glutamine and proline and poor in sulfur-containing amino acids (Shotwell and Larkins, 1989; Shewry, 1995). In the S-poor prolamins, this domain accounts for over 90% of the polypeptide. The S-rich prolamins contain an additional domain at their C terminus, which consists of non-repetitive amino acid sequences and contains between six and eight cysteine residues forming -3-4 intramolecular disulfide bonds. The HMW prolamins are three-domain polypeptides consisting of a central repetitive region that is flanked by N- and C-terminal non-repetitive sequences. In the case of these proteins, the non-repetitive domains contain most or all of the cysteine residues. The S-poor and some of the S-rich prolamins are monomeric, while some of the S-rich prolamins, as well as the HMW prolamins, appear as polymers linked by non-covalent and intcrmolecular disulfide bonds. These aggregated prolamins are also commonly designated as HMW and low molecular weight (LMW) glutenins (Shotwcll and Larkins, 1989; Shewry, 1995).

VIII.

ASSEMBLY AND PROCESSING OF STORAGE PROTEINS

All vacuolar proteins, including the storage proteins, cotranslationally enter into the rough ER. Insertion into the rough ER is achieved by means of an N-terminal signal peptide, which is present in all storage proteins and is removed upon insertion into the organelle. However, subsequent assembly and maturation processes appear to be specific for each storage protein typc. The 7s globulins assemble into trimers within the ER, and are deposited as trimers inside vacuoles. Some of these proteins, for instance bean phaseolin, are glycosylated within the ER, and their glycan residues are subsequently modified within the Golgi. The 11s globulins are translated as pro-proteins, containing the basic and acidic domains, and they first assemble into pro-globulin trimers within the ER. Upon entrance into the vacuole, the globulins follow three additional maturation steps: (i) specific cleavage to form the acidic and basic subunits; (ii) linkage of the acidic and basic domains by an intermolecular disulfide bond; and (iii) assembly of each two trimers into a single hexamer. The prolamins assemble by non-covalent interactions, and in the case of

130


the HMW and aggregated prolamins also by intermolecular disulfide bonds to form large insoluble complexes. However, the specificity of these interactions and the intracellular site where they occur are still unknown. As some of the prolamins assemble into protein bodies within the E R , it is conceivable that some of the interactions between them occur within the organelle. Nevertheless, the assembly of the gliadins must be a highly regulated process inasmuch as if all of the prolamins would aggregate into insoluble complexes within the ER, it is difficult to conceive how they could diffuse into vesicles that bud off the E R enroute to the Golgi. Thus, the assembly of at least some of the prolamins must be prevented until they have been exported from the E R to the Golgi. A.

ASSEMBLY OF T H E STORAGE PROTEINS WITHIN THE ER IS ASSISTED BY MOLECULAR CHAPERONES

It has long been thought that folding and assembly of secretory proteins within the ER occurs spontaneously. However, this hypothesis was reexamined in the last decade upon the discoveries that the E R contains a variety of enzymes and molecular chaperones that assist in the folding and assembly of secretory proteins (Rothman, 1989, 1994; Freedman et al., 1994). Several molecular chaperones have been identified within the E R (Gething and Sambrook, 1992; Vitale et al., 1993). Among these proteins, the HSP-70-related molecular chaperone called binding protein (BiP) has been characterized most extensively. BiP interacts transiently in an ATP-dependent manner with a large number of nascent soluble and membrane proteins that transverse the ER, preventing their aggregation as malfolded proteins. This binding also operates as a “quality control” system to tag malfolded proteins and mark them for degradation. Indeed, in many cases, malfolded secretory proteins form much more stable associations with BiP than their correctly folded counterparts (Gething and Sambrook, 1992; Vitale et af., 1993). Another intensely characterized ER-resident chaperone is protein disulfide isomerase (PDI), which catalyses the formation and isomerization of disulfide bonds in nascent secretory proteins (Freedman eta!., 1994). PDI is likely to play a major role in the folding and assembly of some storage proteins, e.g. cereal prolamins and 11s globulins, because they contain cysteine residues that form intra- and intermolecular disulfide bonds (Freedman et al., 1994). Nielsen and associates (Dickinson et al., 1987) have studied the assembly of soya bean 11s prolegumins, following translation in vitro with rabbit reticulocyte lysate. They initially found that the prolegumin can self-assemble into trimers correctly in the test tube (Dickinson et af., 1987). However, in subsequent experiments, Nielsen et al. (1995) found that upon increasing production of the proglobulin, trimer formation reached a maximum, while insoluble monomers started to accumulate. This suggestcd that the


131

reticulocyte lysate may contain specific saturatable component(s) that promotes trimer formation. Interestingly, in vitro synthesis of the same proglobulin in wheat germ lysate resulted in much less efficient assembly into trimers, and trimer formation was improved upon addition of reticulocyte lysate (Dickinson et al., 1987; Nielsen et al., 1995). This implies that the wheat germ lysate contains low levels of the trimer assembling factor(s). Although the nature of this factor is still not known, the fact that in vitro assembly was more efficient in the presence of ATP indicates that BiP is likely to be involved in this process. Nevertheless, BiP does not apparently operate alone in assisting trimer formation, as addition of purified BiP to the in vitro assembly reaction had no stimulatory effect (Nielsen et al., 1995). By immunoprecipitation of tunicamycin-treated bean seed extracts with anti-phaseolin (bean 7s globulin) or anti-BiP sera, Vitale and associates (D’Amico et al., 1992) could identify ATP-reversible association between BiP and the non-glycosylated storage protein. Yet, in this report, no such interactions were found in non-treated seeds, apparently due to the relatively long periods of pulse labelling. In a later study, Vitale et al. (1995a) showed that BiP also transiently interacts with monomeric, but not trimeric, wild-type phaseolin. These results, coupled with those of expression of wild-type and malfolded phaseolin in Xenopus oocytes and in transgenic tobacco plants (Ceriotti et al., 1991, 1995; Pedrazzini et al., 1994), suggest that phaseolin is subjected to a quality control within the E R , and only correctly folded and assembled protein leaves this compartment en route to vacuoles (Vitale et al., 1993; Vitale and Bollini, 1995). Recently, Gillikin et al. (1995) showed ATP-reversible interactions between BiP and P-conglycinin in developing soya bean seeds under natural conditions. The role of E R enzymes and molecular chaperones in folding and assembly of cereal prolamins is also not understood. Some evidence for the interactions between prolamins and BiP has been recently obtained in rice (Li et al., 1993), maize (Boston et al., 1995; Gillikin et al., 1995) and wheat (Galili et al., 1995b). Yet, it is still unknown whether BiP is just required to prevent incorrect folding of prolamins, or whether it is necessary to assist in their assembly into protein bodies by keeping the nascent storage proteins “assembly competent”. As many of the prolamins, particularly those of wheat, barley and rye, contain intra- and intermolecular disulfide bonds, their folding and assembly may also be assisted by redox-regulating enzymes, such as PDI. Indeed, PDI was shown to catalyse the formation of intramolecular disulfide bonds in wheat S-rich prolamins translocated in vitro into canine microsomes (Bullied and Freedman, 1988). Moreover, formation of the intramolecular bonds between the conserved cysteine residues of wheat S-rich prolamins was recently shown to play a major role in their assembly and deposition (Shimoni and Galili, 1996). Another redox-regulating enzyme that may control gliadin assembly is thioredoxin. This enzyme was shown to catalyse the in vitro reduction of S-rich prolamins under physiological

132


conditions (Kobrehel et af., 1992). The in vivo role of thioredoxin in wheat prolamin assembly is still questionable, as there is no direct proof for the presence of thioredoxin within the ER. However, indirect evidence supporting such a localization has recently been obtained by subcellular fractionation (Kobrehel et af., 1992). Nevertheless, the E R of developing grains of wheat, and possibly also of other cereals, contains a large family of low molecular weight cysteine-rich proteins called purothionines, which were shown to possess thioredoxin-like activities (Wada and Buchanan, 1981). It is thus possible that the rate and type of disulfide bond formation in prolamins may be affected by a balance between enzymes such as PDI, thioredoxin and thioredoxin-like proteins within the ER. Resolving this issue is challenge for future research. Beside its interactions with storage proteins (D’Amico et af., 1992; Pedrazzini et af., 1994; Gillikin er af., 1995), BiP may also have a role in the assembly of PSVs in dicotyledonous seeds. In the pea, where the PSVs bud from the vacuole as mature protein-filled organelles, Hinz et af. (1995) showed that the 2S, 7s and 11s proteins are distributed so that the 2s proteins are on the periphery of the PSV matrix and the 11s proteins occur within the interior core of the PSV matrix. The 7s proteins are distributed throughout the entire PSV matrix. Robinson et af. (1995) showed that BiP is localized at the interface between the peripheral matrix and inner core that separates the 2 s and 11s proteins. This is curious because BiP is widely assumed to be restricted to the E R lumen and nuclear envelope in all eukaryotes, and in the specialized circumstances of prolamin protein bodies that are directly derived from the ER. Robinson and associates (Robinson et af., 1995) interpret this result to indicate a direct E R origin for the pea PSV in accordance with the observations on wheat and rice seeds. However, the storage proteins of the pea have been shown to be localized in the Golgi apparatus (Craig and Goodchild, 1984a), indicating that these proteins are not directly transported from the E R to vacuole, bypassing the Golgi. How these observations may be reconciled remains to be determined, and it is possible that more than one mechanism of PSV ontogeny is operating simultaneously in the pea.

B.

PROTEOLYTIC PROCESSING OF PROLEGUMINS INSIDE VACUOLES

Many seed proteins have been shown to be proteolytically cleaved on the C-terminal side of asparagine residues. Important examples of this include the 2 s albumins (Altenbach et af., 1992; Hara-Nishimura et af., 1993), 11s globulins also termed legumins (Staswick el a!., 1984; Hayashi et af., 1988; Nielsen et af., 1995), a very few 7s storage proteins termed vicilins (Gatehouse et af., 1982), lectins, which represent at least three different and unrelated gene families, including the legume lectins and related a-amylase


133

inhibitor (Bowles et uf., 1986; Gatehouse et uf., 1987; Sanito et al., 1992), castor bean rich (Roberts et al., 1985) and the cereal lectins (Wilkins and Raikhel, 1989), as well as the maturing seed thiol protease of the soya bean (Kalinski et a f . , 1992). In addition, many other seed enzymes and proteins are likely to be shown to exhibit cleavage after asparagines. The enzyme that processes all of these proteins has been identified as an asparaginyl endopeptidase that is related to a putative cysteine protease of the parasite Shistosorna mansoni. This enzyme is not related to the widely distributed and conserved papain superfamily cysteine proteases.

IX. EXPRESSION OF STORAGE PROTEIN GENES IN TRANSGENlC PLANTS Several genes encoding legume globulins and lectins have been expressed in transgenic plants. When expression was driven by a storage protein promoter, all of the proteins were correctly processed and stably accumulated in seed PSVs. The stability of the storage proteins inside the transgenic seed PSVs also suggests that the proteins assembled correctly, achieving a specific structure that is resistant to degradation by PSV-resident proteases. Thus, the signals determining the assembly, transport, processing and accumulation of the storage proteins inside PSVs are apparently regulated by highly conserved signals and mechanisms. Although wild-type storage proteins are stable in PSVs, modification of the proteins can result in unstable products. Hoffman rt al. (1988) modified phaseolin by inserting a short amino acid sequence containing several methionine residues (HiMet phaseolin) in an attempt to increase the methionine content of the protein. Although it was expressed and synthesized at the same levels as the unmodified control, almost no protein accumulated. Pueyo et al. (1995) showed that the vacuole is the intracellular site where the HiMet phaseolin is degraded. This result suggests that modification of the phaseolin caused exposure of some protein domains to proteolytic degradation by PSV-resident proteases. Interestingly, supplementing the HiMet phaseolin with a C-terminal amino acid sequence, K/HDEL (a sequence that induces continuous retrieval of reticuloplasmins from the cis-Golgi to the ER lumen causing retention within the organelle), results in the accumulation of HiMet protein in the ER and Golgi. This shows that HiMet is stable until it progresses to the vacuole. Nevertheless, the HiMet could accumulate in immature vacuoles that have just begun to accumulate storage proteins. Perhaps the protease responsible for HiMet degradation is developmentally regulated and accumulated coordinately with the storage proteins. Another interesting set of results regarding the assembly and transport of storage proteins was obtained by expressing maize prolamins in seeds of

134

G. GALILI and E. M. HERMAN

transgenic tobacco and petunia plants. Maize prolamins (called zeins) accumulate inside the ER and are distinguished as a-, p-, y- and 6-zeins (Shotwell and Larkins, 1989). The a-zeins are mostly composed of small peptide repeats rich in glutamine and proline, while the other zeins have entirely different structures and are rich in methionine (p-zeins) and cysteine ( y - and 6-zeins). Although all zeins accumulate in the ER of maize endosperm, a 15 kDa p-zein, expressed in transgenic tobacco under control of the phaseolin promoter, accumulated stably in PSVs and not in the E R (Hoffman et af., 1987). In a subsequent study, the same 15 kDa p-zein was expressed under the 35s promoter, and it accumulated in both cytoplasmic protein bodies that were derived by fission from the ER, as well as in intravacuolar inclusions that likely originated by autophagy of the cytosolic protein bodies (Bagga et al., 1995). The 35s promoter differs from the phaseolin promoter by being expressed mainly at early stages of seed development. Thus, the intracellular site of accumulation of the 15 kDa p-zein may differ, depending on whether the protein is expressed at the early stages of seed maturation, i.e. in vegetative-type cells (35s promoter), or in embryonic cells formed at the later stages of seed maturation (phaseolin promoter). Whether zein accumulated at later stages of development accompanies the intrinsic storage proteins through the endomembrane system or whether it is accumulated in the vacuole by autophagy must be resolved to reconcile the differences in protein localization during early and late stages of seed development. The differences in zein localization and accumulation at early and late stages of seed development appear to indicate that there is either a fundamental difference between PSVs and vegetative vacuoles, or that cells of early and late stages of seed development differ in the microenvironment of the ER, which may affect the folding and aggregation of zeins inside the organelle. In other studies (Williamson et af., 1988; Ohtani et al., 1991), a maize a-zein expressed with a phaseolin promoter was unstable in developing seeds of transgenic petunia and tobacco plants. Although the intracellular site of disposal of this protein was not determined, it was probably degraded inside the PSVs. Storage proteins routing to PSVs apparently pass a “quality control” inside the ER. To study this, Vitale and associates (Pedrazzini et al., 1994) characterized the fate of a wild-type and deletion mutant of bean phaseolin in protoplasts and leaves of transgenic tobacco plants. Both the wild-type and malfolded phaseolin were unstable in leaf cells. However, while the turnover of the wild-type protein was inhibited by treatment with brefeldin A, that of the malfolded protein was not (Vitale et af., 1995b). As brefeldin A is known to block the transport of secretory proteins from the ER to the Golgi, these results suggest that the wild-type and mutant phaseolin follow different intracellular routes for their disposal. While the wild-type protein is apparently transported via the Golgi to vacuoles where it is degraded by


135

vacuolar proteases, the situation with the malfolded protein is not clear. Vitale et al. (1995b) suggested that the brefeldin A-insensitive turnover of the deletion mutant protein implies that it is degraded within the ER. However, since autophagy may not be inhibited by brefeldin A, it is impossible yet to eliminate the possibility that the mutant phaseolin is transported from the E R to vacuoles by autophagy, and that it is also degraded in this organelle. As this study was conducted with leaf cells, it would be intriguing to express the mutant phaseolin protein using a phaseolin promoter and test the stability of the malfolded protein and whether it is disposed within the ER or the PSVs of the seed embryonic cells.

ACKNOWLEDGEMENTS We thank Dr Brian A. Larkins for critical reading of this review and helpful comments. The work in the laboratories of both authors was supported by research grants from BARD: The United States-Israel Binational Agricultural Research and Development, Funds Nos. IS-1805-90 (GG) and US-233493 (EMH and GG). EMH was supported by a United States Department of Agriculture, Agricultural Research Service Fellowship to visit the Weizmann Institute during FY95. GG is an incumbent of the Bronfman Chair of Plant Sciences.

REFERENCES Altenbach, S. B., Pearson, K. W. and Sun, S. S. M. (1992). Nucleotide sequence of cDNAs encoding two members of the Brazil nut methionine-rich 2 s albumin gene family. Plant Physiology 98, 1520-1522. Bagga, S., A d a m , H., Kemp, J. D. and Sengupta-Gopalan, C. (1995). Accumulation of 15-kilodalton zein in novel protein bodies in transgenic tobacco. Plant Physiology 107, 13-23. Baumgartner, B. and Chrispeels, M. J. (1977). Purification and characterization of vicilin peptidohydrolase, the major endopeptidase in the cotyledons of mung bean seedlings. European Journal of Biochemistry 77, 223-233. Baumgartner, B., Tokuyasu, K. T. and Chrispeels, M. J. (1978). Localization of vicilin peptidohydrolase in the cotyledons of mung bean seedlings by immunofluorescence microscopy. Journal of Cell Biology 79, 10-19. Bechtel, D . B., Frend, A., Kaleikau, L. A., Wilson, J. D. and Shewry, P. R. (1989). Vacuole formation in wheat starchy endosperm. Food Microstructure 8, 183- 190. Boston, R. S., Gillikin, J. W. and Wrobel, R . L. (1995). Coordinate induction of three ER-luminal stress proteins in maize endosperm mutants. Journal of Cellular Biochemistry, Supplement 19A (Abstract A3-202), p 143. Bowles, D. J., Marcus, S. E., Pappin, D. J. C., Findlay, J. B. C., Eliopolus, E., Maycox, P. R. and Burgess, J. (1986). Posttranslational processing of concanavalin A precursors in jack bean cotyledons. Journal of Cell Biology 102, 1284-1 297. Briaty, L. G., Coult, D. A. and Boulter, D. (1970). Protein bodies of germinating

136


seeds of Vicia faba. Changes in fine structure and biochemistry. Journal of Experimental Botany 21, 513-524. Bulleid, N. J. and Freedman, R. B. (1988). Defective co-translational formation of disulphide bonds in protein disulphide-isomerase-deficient microsomes. Nature (London) 335, 649-651. Ceriotti, A., Pedrazzini, E., Fabbrini, M. S., Zoppe, M., Bollini, R. and Vitale, A. (1991). Expression of wild-type and mutated vacuolar storage protein phaseolin in Xenopus oocytes reveals relationships between assembly and intracellular transport. European Journal of Biochemistry 202, 959-968. Ceriotti, A., Pedrazzini, E., Bielli, A., Giovinazzo, G., Bollini, R. and Vitale, A. (1995). Assembly and intracellular transport of phaseolin, the major storage protein of Phuseolus-vulgaris L. Journal of Plant Physiology 145, 64-53. Chappell, J., Van der Wilden, W. and Chrispeels, M. J. (1980). The biosynthesis of ribonuclease and its accumulation in protein bodies in the cotyledons of mung beans. Developments in Biology 76, 115-125. Chrispeels, M. J. (1983). The Golgi apparatus mediates the transport of phytohemagglutinin to the protein body in bean cotyledons. Planta 158, 14G151. Chrispeels, M. J. (1991). Sorting of proteins in the secretory system. Annual Review of Plant Physiology and Plant Molecular Biology 42, 21-53. Craig, S. and Goodchild, D. J. (1984a). Periodic-acid treatment of sections permits on-grid localization of pea seed vicilin in ER and Golgi. Protoplusma 122, 35-44. Craig, S. and Goodchild, D. J. (1984b). Golgi-mediated vicilin accumulation in pea cotyledon cells is redirected by monensin and nigercin. Protoplasma 122, 91-97. Craig, S., Goodchild, D. J. and Hardham, A. R. (1979). Structural aspects of protein accumulation in developing pea cotyledons. I. Qualitative and quantitative changes in parenchyma cell vacuoles. Australian Journal of Plant Physiology 6, 81-98. Craig, S., Goodchild, D. J. and Miller, C. (1980). Structural aspects of protein accumulation in developing pea cotyledons. 11. Three-dimensional reconstitution of vacuoles and protein bodies from serial sections. Australian Journal of Plant Physiology 7, 329-337. D’Amico, L., Valsasina, B., Daminati, M. G., Fabbrini, M. S., Nitti, G., Bollini, R., Ceriotti, A. and Vitale, A. (1992). Bean homologs of the mammalian glucose-regulated proteins: induction by tunicamycin and interaction with newly synthesized seed storage proteins in the endoplasmic reticulum. Plant Journal 2, 44-55, Dickinson, C. D., Floener, L. A., Lilley, G. G. and Nielsen, N. C. (1987). Self-assembly of proglycinin and hybrid proglycinin synthesized in vitro from cDNA. Proceedings of the National Academy of Sciences of the USA 84, 5525-5529. Etzler, M. E. (1985). Plant lectins: molecular and biological aspects. Annual Review of Plant Physiology and Plant Molecular Biology 36, 209-234. Faye, L., Strum, A., Bollini, R., Vitale, A. and Chrispeels, M. J. (1986). The position of the oligosaccharide side-chains on phytohemagglutinin and their accessibility to glycosidases determines their subsequent processing in the Golgi. European Journal of Biochemistry 158, 741-745. Faye, L., Johnson, K. D., Strum, A. and Chrispeels, M. J. (1989). Structural biosynthesis and function of asparagine-linked glycans on plant proteins. Physiologia Plantarum 75, 309-314. Freedman, R. B., Hirst, T. R. and Tuite, M. F. (1994). Protein disulphide isomerase:

PROTEIN BODIES: STORAGE VAC‘IJOLES IN SEEDS

137

building bridges in protein folding. Trends in Biochemical Science 19, 331336. Galili. G.. Altschuler, Y. and Levanony, H. (1093). Assembly and transport of seed storage proteins. Trends in Cellular Biology 3 , 437-443. Galili. G . , Altschuler, Y., Levanony, H . , Giorini-Silfen, S . , Shimoni, Y. and Shani, N. (1995a). Assembly and transport of wheat storage proteins. Journal qf Plum Physiology 145, 6 2 M 3 1. Galili, G., Giorini-Silfen. S . , Shimoni, Y., Altschuler. Y., Levanony, H . , Shani, N., Karchi, H. and Galun. E. (199%). Assembly and intracellular transport of wheat storage proteins. Journal of C~llularBiochemistry, Supplement 19A419, 132. Gatehouse. J . A., Lycett, G. W., Croy, R . R. D. and Boulter, D . (1982). The post-translational proteolysis of the subunits of vicilin from pea (Pissum sativum I>.), Biochemical Journal 207, 629-632. Gatehouse, J. A , , Bown, A . D., Evans, I. M., Gatehouse, L. N . , Jobes, D., Preston, P. and Croy, R. R . D . (1987). Sequence of the seed lectin gene from pea (Pissumsativum L.). Nucleic Acids Research 15, 7642. Gething, M.-J. and Sambrook J . (1992). Protein folding in the cell. Nature (London) 355, 33-45. Gillikin, J . W., Fontes, E. P. B. and Boston. R . S . (1995). Protein-protein interactions within the endoplasmic reticulum. In “Methods in Cell Biology” (D. W. Galbraith, D. P. Bourque and H. J . Bohnert, eds), Vol. 50, pp. 309-323. Academic Press, San Diego. Greenwood, J. S. and Chrispeels, M . J . ( 1985). Immunocytochemical localization of phaseolin and phytohemagglutinin in the endoplasmic reticulum and Golgi complex of developing bean cotyledons. Planta 164, 295-302. Hara-Nishimura, I., Takeuchi, Y . , Inoue, K . and Nishimura, M. (1993). Vesicle transport and processing of the precursor to 2s albumin in pumpkin. Plant Journal 4, 793400. Hara-Nishimura, I . , Shimada, T., Hiraiwa, N . and Nishimura, M. (1995). Vacuolar processing enzyme responsible for maturation of seed proteins. Journal of Plant Physiology 145, 632440. Hayashi. M., Mori H., Nishimura, M., Akazawa, ‘I. and Hara-Nishimura, I. (1988). Nucleotide sequence of cloned cDNA coding for pumpkin 1IS-globulin p subunit. European Journal of Biochemistry 172, 627-032. Herman, E. M. (1994). Multiple origins of intravacuolar protein accumulation of plant cells. Advances in Structural Resenrch 3 , 243-283. Herman, E. M.and Chrispeels, M. J . (1980). Phospholipase 13 and phosphatidic acid phosphatase: acid hydrolases involved in phospholipid catabolism in mung bean cotyledons. Planf Physiology 66, 1001-1007. Herman, E. M. and Shannon, L. M. (1984a). Imniunocytochemical localization of concanavalin A in developing jack bean cotyledons. Plunta 161, 97-104. Herman, E. M. and Shannon, L. M. (1984b). The role of the Golgi apparatus in the deposition of the seed lectin of Bauhinia purpuren (Leguminoseae). Protoplasma 121. 163-170. Herman, E. M. and Shannon, L. M. (198S). Accumulation and subcellular localization of a-galactosidase in developing soya bean cotyledons. Plant Physiology 77, 886-890. Herman, E. M., Platt-Aloeia, K. A . , Thomson, W. W. and Shannon, L. M. (1984). Freeze fracture and filipin cytocheniical observations o f developing soybean cotyledon protein bodies and Golgi apparatus. European Journal of Cell Biology 35, 1-7.

138


Hinz, G., Hoh, B., Hohl, I. and Robinson, D. G. (1995). Stratification of storage proteins in the protein storage vacuole of developing cotyledons of Pisum sativum L. Journal of Plant Physiology 145, 437-442. Hoffman, L. M., Donaldson, D. D., Bookland, R., Rashka, K. and Herman, E. M. (1987). Synthesis and protein body deposition of maize 15 kD zein in transgenic tobacco seeds. EMBO Journal 6, 3213-3221. Hoffman, L. M., Donaldson, D. D. and Herman, E. M. (1988). A modified storage protein is synthesized, processed and degraded in the seeds of transgenic plants. Plant Molecular Biology 11, 717-729. Hoh, B., Hinz, G., Jeong, B.-K. and Robinson, D. G. (1995). Protein storage vacuoles form de novo during pea cotyledon development. Journal of Cell Science 108, 299-310. Inoue, K., Motozaki, A., Takeuchi, Y., Nishimura, M. and Hara-Nishimura, I. (1995). Molecular characterization of proteins in protein body membrane that disappear most rapidly during transformation of protein bodies into vacuoles. Plant Journal 7, 235-243. Johnson, K. D. and Chrispeels, M. J . (1992). Tonoplast-bound protein kinase phosphorylates tonoplast intrinsic protein. Plant Physiology 100, 1787-1795. Johnson, K. D., Herman, E. M. and Chrispeels, M. J. (1989). An abundant, highly-conserved tonoplast protein in seeds. Plant Physiology 91, 1006-1013. Kalinski, A. J., Melroy, D. L., Dwivedi, R. S. and Herman, E. M. (1992). A soybean vacuolar protein (P34) related to thiol proteases which is synthesized as a glycoprotein precursor during seed maturation. Journal of Biological Chemistry 267, 12 068-12 076. Kobrehel, K., Wong, J. H., Balogh, A., Kiss, F., Yee, B. C. and Buchanan, B. B. (1992). Specific reduction of wheat storage proteins by thioredoxin. Plant Physiology 99,919-924. Koehler, S. and Ho, D. (1990). Hormonal regulation, processing, and secretion of cysteine proteinase in barley aleurone. Plant Cell 2 , 769-783. Larkins, B. A. and Hurkman, W. J. (1978). Synthesis and deposition of zein in protein bodies of maize endosperm. Plant Physiology 62, 256-263. Levanony, H., Rubin, R., Altschuler, Y. and Galili, G. (1992). Evidence for a novel route of wheat storage proteins to vacuoles. Journal of Cell Biology 119, 1117-1 128. Li, X., Wu, Y., Zhang, D.-Z., Gillikin, J. W., Boston, R. S., Franceschi, V. R. and Okita, T. W. (1993). Rice prolamine protein body biogenesis: A BiP-mediated process. Science 262, 1054-1056. Maeshima, M., Hara-Nishimura, I., Takeuchi, Y. and Nishimura, M. (1994). Accumulation of vacuolar H+-pyrophosphatase and H+-ATPase during reformation of the central vacuole in germinating pumpkin seeds. Plant Physiology 106, 61-69. Maurel, C., Kado, R. T., Geurn, J. and Chrispeels, M. J . (1995). Phosphorylation regulates the water channel activity of the seed-specific aquaporin a-TIP. EMBO Journal 14, 3028-3035. Melroy, D. L. and Herman, E. M. (1991). TIP, an integral membrane protein of the soybean seed protein storage vacuole, undergoes developmentally regulated membrane insertion and removal. Planta 184, 113-122. Mitsuashi, W. and Minamikawa, T. (1989). Synthesis and posttranslational activation of sulfhydryl-endopeptidase in cotyledons of germinating Vigna mungo seeds. Plant Physiology 89, 274-279. Moore, P. J., Swords, K. M. M., Lynch, M. A. and Staehelin, L. A. (1991). Spatial organization of the assembly pathways of glycoproteins and complex polysac-


139

charides in the Golgi apparatus of plant cells. Journal of Cell Biology 112, 589-602. Nielsen, N. C., Jung, R., Nam, Y., Beaman, T. W., Oliveira, L. 0. and Bassuner, R. B. (1995). Synthesis and assembly of 11s globulins. Journal of Plant Physiology 145, 641-647. Noda, T. and Farquhar, M. G. (1992). A non-autophagic pathway for diversion of ER secretory proteins to lysosomes. Journal of Cell Biology 119, 85-97. Ohtani, T., Galili, G., Wallace, J. C., Thompson, G. A. and Larkins, B. A. (1991). Normal and lysine-containing zeins are unstable in transgenic tobacco seeds. Plant Molecular Biology 16. Osborne, T. B. (1924). The vegetable proteins. In “Monographs on Biochemistry”. Longrams Green, London. Paris, N., Stanley, C. M., Jones, R. L. and Rogers, J. C. (1996). Plant cells contain two functionally distinct vacuolar compartments. Cell 85, 563-572. Pedrazzini, E., Giovinazzo, G., Bollini, R., Ceriotti, A. and Vitale, A. (1994). Binding of BiP to an assembly-defective protein in plant cells. Plant Journal 5 , 103-110. Pueyo, J. U., Chrispeels, M. J. and Herman, E. M. (1995). Degradation of transport-competent destabilized phaseolin with a signal for retention in the endoplasmic reticulum occurs in the vacuole. Planta 196, 586-596. Roberts, L. M., Lamb, I. F., Pappin, D. J. C. and Lord, J. M. (1985). The primary sequence of Ricinus comrnunis agglutinin. Comparison with ricin. Journal of Biological Chemistry 260, 15 682-15 686. Robinson, D. G., Hoh, B., Hinz, G. and Jeong, B.-K. (1995). One vacuole or two vacuoles: do protein storage vacuoles arise de novo during pea cotyledon development? Journal of Plant Physiology 145, 654-664. Rogers, J. C., Dean, D. and Heck, G. R. (1985). Aleurin: a barley thiol protease closely related to mammalian chatepsin H. Proceedings of the National Academy of Sciences of the USA 82, 6512-6516. Rothman, J. E. (1989). Polypeptide chain binding proteins: catalysts of protein folding and related processes in cells. Cell 59, 591-601. Rothman, J. E. (1994). Mechanisms of intracellular protein transport. Nature (London) 312, 55-63. Rubin, R., Levanony, H. and Galili, G. (1992). Characterization of two types of protein bodies in developing wheat endosperm. Plant Physiology 99, 718724. Sanito, A , , Daminti, M. G., Vitale, A . and Bollini, R. (1992). The a-amylase inhibitor of bean seed: two step proteolytic maturation in the protein storage vacuoles of the developing cotyledon. Physiologia Plantarum 85, 425-432. Scott, M. P., Jung, R., Muntz, K. and Nielsen, N. C. (1992). A protease responsible for post-translational cleavage of a conserved Asn-Gly linkage in glycinin, the major seed storage protein of soybean. Proceedings of the National Academy of Sciences of the USA 89, 658-662. Shewry, P. R. (1995). Cereal seed storage proteins. In “Seed Development and Germination” (H. Kigel and G. Galili, eds), pp. 45-72. Marcel Dekker, New York. Shewry, P. R. and Tatham, A. S. (1990). The prolamin storage proteins of cereal seeds: structure and evolution. Biochemical Journal 261, 1-12. Shewry, P. R., Halford, N. G. and Tatham, A. S. (1989). The high molecular weight subunits of wheat, barley and rye: genetics, molecular biology, chemistry and role in wheat gluten structure and functionality. In “Oxford Survey of Plant Molecular and Cell Biology” (B. J . Miflin, ed.), Vol. 6, pp. 163-219. Oxford

140


University Press, New York. Shimoni, Y. and Galili, G. (1996). Intramolecular disulfide bonds between conserved cysteines in wheat gliadins control their deposition into protein bodies. Journal of Biological Chemistry 271, 18 869-18 874. Shotwell, M. A. and Larkins, B. A. (1989). The biochemistry and molecular biology of seed storage proteins. In “The Biochemistry of Plants” (A. Marcus, ed.), Vol. 15, pp. 297-345. Academic Press, San Diego, CA. Spencer, D. and Higgins, T. J. V. (1979). Molecular aspects of seed protein biosynthesis. Current Advances in Plant Science 34, 1-15. Staswick, P. E., Hermodson, M. A. and Nielsen, N . C. (1984). The amino acid sequence of the A2Bla subunit of glycinin. Journal of Biological Chemistry 259, 13 424-13 430. Sturm, A , , Johnson, K . D., Szumilo, T., Elbein, A. D. and Chrispeels, M. J. (1987). Subcellular localization of glycosidases and glycotransferases involved in the processing of N-linked oligosaccharides. Plant Physiology 85, 741-745. Sturni, A., Volker, T., Herman, E. M. and Chrispeels, M. J . (1988). Correct targeting, glycosylation and Golgi-processing of the bean vacuolar protein phytohemagglutinin in transgenic tobacco. Planta 175, 170-183. Tanaka, T., Minamikawa, T., Yamauchi, D. and Ogushi, Y. (1993). Expression of an endopeptidase (EP-C1) in Phaseolus vulguris plants. Plant Physiology 101, 421428. Van der Wilden, W., Herman, E. M. and Chrispeels, M. J. (1980). Protein bodies of inung bean cotyledons as autophagic organelles. Proceedings of the National Academy qf Sciences of the USA 77, 42&432. Vitale, A . and Bollini, R. (1995). Legume storage proteins. In “Seed Development and Germination” (H. Kigel and G. Galili, eds), pp. 73-102. Marcel Dekker, New York. Vitale, A., Ceriotti, A . and Denecke, J . (1993). The role of the endoplasmic reticulum in protein synthesis, modification and intracellular transport. Journal of Experimental Botany 44, 1417-1444. Vitale, A , , Bielli, A. and Ceriotti, A. (1995a). The binding-protein associates with monomeric phaseolin. Plant Physiology 107, 1411-1418. Vitale, A , , Pedrazzini, E., Bielli, A , , Giovinazzo, G., Faoro, F., Bollini, R. and Ceriotti, A . (199%). Quality control in the endoplasmic reticulum. Journal of Cellular Biochemistry, Supplement 19A-020, 132. Wada, K. and Buchanan, B. B. (1981). Purothionin: a seed protein with thioredoxin activity. FEBS Letters 124, 237-240. Watanabe, H., Abe, K., Emori, Y . , Hosoyama, H. and Aria, S. (1991). Molecular cloning and gibberelin-induced expression o f multiple cysteine proteases of rice seeds. Journal of Biological Chemistry 266, 16 897-16 902. Wilkins, T. A. and Raikhel, N . V. (1989). Expression of rice lectin is governed by two temporally and spatially regulated mRNAs in developing embryos. Planf Cell 1, 541-549. Williamson, J . O., Galili, G . , Shaw, B. A., Larkins, B. A. and Gelvin, S. B. (1988). The synthesis of a 19kD zein protein in transgenic petunia plants. Plant P h y ~ i o l ~ g88, y 1002-1007. Yaklich, B. and Herman, E. M. (1995). Protein storage vacuoles of soybean aleurone cells accumulate a unique glycoprotein as well as proteins thought to be embryo specific. Plant Science 107, 57-67. Yamauchi, D., Akasofu, H. and Minamikawa, T. (1992). Cysteine endopeptidase from Vigna mungo: gene structure and expression. Plant Cell Physiology 33, 789-797.

Compartmentation of Secondary Metabolites and Xenobiotics in Plant Vacuoles

M. WINK

lnstitut f u r Pharmazeictisdie Biologie, Universitat Heidelberg, Im Neuenheimer Feld 364, 0-69120 Heidelberg, Germany

141

Introduction ............................................................................. Secondary Metabolites as Defence and Signal Compounds of Plants . .. . . . .. .. ... .. .... . . . . . . . .. . . . .. . . . . B. Fate of Xenobiotics in Plants ................................................ C. Aims and Scope _ . .. . . . _ _ _ _ . . . . . . . . . _ _ .

141 143 145

11.

Vacuolar Storage of Secondary Compounds and Xenobiotics .. . . . .. . . . . . A , Secondary Compounds .. , .. .. , . .. . . . , . ., , , . , . . .. ... .. .. . . . .. . . . , .. .. . . B. Xenobiotics ............ ..................................................... C. Mechanisms Underlyin uolar Sequestration . . .. . . . . .

145 145 153 154

111.

Conclusions .............................................................................. Acknowledgements . .. .. . . . . . , . , , .. . . . . , .. , . . . . .. .. .. , , . ... .. . . . . .. . . .. . . . . . .. .. .. . . . .. References .. .. . . . .. . . . . . . . . . . . .. .. ... .. .. . . . .. . . .. . . . . . . .. . .. . . . . .. . .. . . . . . . . . . . .. . . . . . ..

160

I.

A.

I.

159 160

INTRODUCTION

A . SECONDARY METABOLITES AS DEFENCE AND SIGNAL COMPOUNIX OF PLANTS

Plants need to protect themselves against herbivores (mostly insects and grazing vertebrates) and microorganisms. Various defence strategies can be observed in plants (for reviews, see Levin, 1976; Swain, 1977; Wink, 1988, 1093a; Harborne. 1993; Rosenthal and Rerenbaum, 1991), which are not

142

M. WINK

independent and which may operate cooperatively and synergistically. The strategies include: Physical protection by thorns, spikes, trichomes, o r glandular or stinging hairs (“weapons”), by thick bark or cuticles of roots and stems or by robust seed coats (“armour”). The production and storage of defence chemicals (or allelochemicals), which are abundant and a typical trait of all plants. The following situations can be distinguished: (1) Plant surfaces are usually covered by a hydrophobic cuticle consisting of antibiotic and deterrenthepellent cuticular waxes which may contain other biologically active allelochemicals such as flavonoids. (2) Plants can synthesize inhibitory proteins (e.g. lectins, protease inhibitors, toxalbumins) or enzymes (e.g. chitinase, glucanases, hydrolases, nucleases) which could degrade microbial cell walls or other microbial constituents, o r peroxidase and phenolase, which could help to inactivate microbial toxins (“xenobiotics”, see below) produced by pathogens. (3) As a most important trait, plants can produce and store secondary metabolites (of which more than 50 000 compounds have been described so far) with deterrenthepellent or toxic properties against microorganisms, viruses and/or herbivores. These products are often stored at strategically important sites: epidermal tissues or in cells adjacent to an infection, or in plant parts that are especially important for reproduction and survival (flowers, fruits, seeds, bark or roots). Considering their synthesis, three situations can be distinguished: - allelochemicals are often produced constitutively; - some may be activated by wounding or infection (“preformed defence chemicals”), such as cyanogenic glycosides, glucosinolates, coumaryl glycosides, alliin, ranunculin, etc.); - in some instances, a de nova synthesis induced by elicitors (so-called phytoalexins), infection or herbivory can be observed. In addition to defence, secondary metabolites are employed by plants to attract pollinating insects or seed-dispersing animals, e .g. by coloured compounds such as betalains (within the Centrospermae), anthocyanins, carotinoids, flavonoids or fragrances, such as terpenes, amines and aldehydes. In this case we can consider the secondary metabolites as attracting signal substances. Also, physiological roles, such as ultraviolet protection (by flavonoids or coumarins), nitrogen transport or nitrogen storage (some alkaloids, non-protein amino acids, lectins and protease inhibitors), or photosynthetic pigments (carotinoids) may be additional features. Although the biological functions of most plant-derived secondary metabolites have not been studied experimentally in depth, it is now generally assumed that many compounds are important for the survival and fitness of

SECONDARY METABOLITES AND XENOBIOTICS

143

a plant (Fraenkel, 1959; Wink, 1988, 1993a) and that they are not useless waste products, as was suggested earlier in this century. Secondary metabolites are often not directed against a single organism, but generally against a variety of potential enemies, or they may combine the roles of both deterrents and attractants (e.g. anthocyanins and many essential oils can be attractants in flowers, but are also insecticidal and antimicrobial). Thus, many natural products have dual or even multiple functions. It might be argued that the defence hypothesis cannot be valid since most plants, even those with extremely poisonous metabolites (from the human point of view), are nevertheless attacked by pathogens and herbivores. However, we have to understand and accept that chemical defence is not an absolute process. Rather, it constitutes a general barrier which will be effective in most circumstances, i.e. most potential enemies are repelled or deterred. However, plants with chemical defences also represent an ecological niche for potential pathogens and herbivores. During evolution a few organisms have generally been successful in specializing towards that niche, i.e. with respect to a particular toxic plant in that they found a way to sequester the toxins or become immune to them. This is especially apparent in the largest class of animals, the insects (with several million species on our earth), which are often highly host plant-specific (Bernays and Graham, 1988; Bernays and Chapman, 1994). The number of these “specialists” is exceedingly small for a given plant species as compared to the number of potential enemies that are present in the ecosystem. These specialists are the exception to the general rule, similar to the situation of some viruses and bacteria which have overcome our powerful immune system. A prerequisite for both the defence and the signal functions is the sequestration of critical amounts of the respective active secondary metabolites, otherwise the necessary effect (which is usually dose-dependent) could not be achieved. Whereas lipophilic material (e.g. many terpenoids) is stored in resin ducts, oil containers, dead cells or trichomes (Wiermann, 1981), hydrophilic compounds (e.g. alkaloids, non-protein amino acids, organic acids and glycosides) are usually confined to the vacuole, which seems to be especially adapted to the bulk storage of allelochemicals (the term “defence compartment and signal compartment” was coined to emphasize this vacuolar function; Wink, 1993b). As mentioned before, the vacuoles employed for defence or signalling can be expected in cells that are positioned in strategically favourable positions, such as epidermal tissues, or in flowers or fruits, i.e. they show a high degree of cell- and tissue-specificity.

B. FATE OF XENOBIOTICS IN PLANTS

Xenobiotics can be characterized as compounds that are not indigenously present in a particular plant species. In nature, xenobiotics can be secondary

144

M. WINK

metabolites which are produced by other plants and released to the environment either via the rhizosphere or by leaching from aerial parts. These compounds often interfere with the germination and seedling growth of plants of the same or other species which compete for light, water o r nutrients. These interactions have been called “allelopathic” (for reviews, see Rice, 1984; Waller, 1987; Inderjit et al., 1995). When plants are infected by fungi or bacteria, the pathogens often produce secondary compounds (“phytotoxins”) to weaken the defence of the host plant. A more recent exposure of plants to xenobiotics is due to agrochemicals and industrial pollutants. Mechanisms have evolved in plants during evolution to cope with secondary metabolites produced either by competing plants or by pathogens. These mechanisms can also be used by plants when exposed to man-made xenobiotics (the so-called “green liver” concept: Sandermann et al., 1985; Dodge, 1989; Hathway, 1989). The handling of xenobiotics shows some similarities between animals and plants (for reviews, see Cole, 1994; Sandermann, 1994). As an initial reaction, lipophilic compounds are often oxidized, reduced or hydrolysed (Jacoby and Ziegler, 1990) in order to reveal or introduce a functional group which will enhance reactivity and polarity of the molecules (“phase 1” reaction in the pharmacological literature). In a second step these compounds are conjugated with more polar molecules, such as sugars, amino acids, acids or glutathione (“phase 2”). In “phase 3”, these conjugates are eliminated or sequestered in a safe place. Although obvious similarities exist, marked differences are also apparent (for reviews, see Baldwin, 1977; Menn, 1978) (Table I): Phase I , Whereas cytochrome-P-450 hydroxylases have broad and overlapping substrate specificities in animals (in order to cope with a wide variety of dietary secondary metabolites), plant cytochrome-P-450 enzymes with their multiple isoforms (Donaldson and Luster, 1991) are more selective and naturally function to catalyse specific reactions in the biosynthesis of secondary metabolites (Cole, 1994). Any modification of a xenobiotic would be more a side-reaction. However, Sandermann (1994) has argued that the specificity can also be due to special enzymes involved only in the metabolism of xenobiotics. In addition to cytochrome-P-450 hydroxylases, xenobiotics can be modified by demethylases and peroxidases, which are abundant in plants. Phase 2. Whereas conjugation reactions with sulfate, amino acids and glucuronic acid are common in animals, plants often add sugar moieties to xenobiotics with aid of UDP-O-glucosyltransferases and UDP-Nglucosyltransferases. The resulting glycosylderivatives are often further acylated with malonic acid by 0-or N-malonyltransferases. For plants, special importance is attributed to glutathione transferases, of which different isoforms with varying substrate specificities exist (Timmermann, 1989);

SECONDARY METABO1,ITES AND XENOBIOTICS

145

TABLE I Fate of xenobiotics in plants arid aniniuls Plants _ _ ~ _ _ _ _ _

Phase I (transformation) Phuse 2 (conjugation)

Phase 3 (excretion/ compartmentation)

An im a I s ~

Hydroxylation, reduction, hydrolysis

Hydroxylation, reduction, hydrolysis

Glucosylation + malonylation Glucuronic acid conjugates Conjugation with amino Gluthatione S-conjugation acids Glutathione S-conjugation Conjugation with sulfates Binding to lignin/polysaccharides Vacuolar sequestration

Excretion with urinelfileces

these gluthathione S-transferases conjugate xenobiotics with reduced glutathione. Phase 3. Whereas in animals conjugates of xenobiotics are eliminated from the body tissues via optimized excretory systems (involving a specific ATPase), other mechanisms are required in plants, which do not have an active excretory system: conjugates are either bound to insoluble constituents, e.g. lignin and polysaccharides (Langebartels rt al., 1986), o r they are sequestered in the vacuole. C. AIMS AND SCOPE

In this review the experimental evidence for the vacuolar sequestration of secondary metabolites and xenobiotics is summarized. Because the questions of how these compounds pass the tonoplast and how they are accumulated against a concentration gradient in the vacuole are discussed in other chapters of this volume, these topics will be treated more briefly in this chapter.

11. VACUOLAR STORAGE OF SECONDARY COMPOUNDS AND XENOBIOTICS A. SECONDARY COMPOUNDS

It had been observed by light microscopy as early as the last century that many of the coloured flower pigments (e.g. anthocyanins, flavonol glycosides) or the red betalaines of Beta roots are exclusively sequestered in the vacuole.

146

M. WINK

Modern microspectrometric investigations (Schnabl et al., 1986; Gonnet and Hieu, 1992) and direct determinations in isolated vacuoles (Table 11) have confirmed these early observations. The fate of the non-coloured metabolites remained almost unknown for many decades. Progress came when it was possible to prepare protoplasts from a wide number of plants in the 1970s (Ruesink, 1980) and to produce intact vacuoles without a significant efflux of vacuolar contents. Several protocols, which employ the lysis of protoplasts in slightly hypotonic media and centrifugation on sucrose or Ficoll gradients, often result in pure and intact vacuoles (for reviews, see Marty et al., 1980; Wagner, 1982; Leigh, 1983; Ryan and Walker-Simmons, 1983; Mader, 1984; Willenbrink, 1987). In order to separate the vacuoles from the surrounding medium, centrifugation of the vacuoles through a silicone oil layer was another important step forward (Wagner, 1982; Mende and Wink, 1987). As a general rule, hydrophilic and water-soluble secondary metabolites are sequestered in the vacuole (Table 11), although their site of synthesis is usually the cytoplasm. A few compounds are synthesized in the chloroplast, such as quinolizidine alkaloids (Wink and Hartmann, 1982) or the piperidine alkaloid coniine (Roberts, 1981). Protoberberine alkaloids, such as berberine, are synthesized in small vesicles which later fuse with the tonoplast to release their alkaloidal content into the vacuole (Sato et al,, 1990, 1993, 1994). Several glycosides become malonylated by 0-malonyltransferase in the cytoplasm. 0-malonylglucosides are very labile and thus difficult to isolate; it is likely that many more compounds exist in this form in vivo (Sandermann, 1994). The charged malonyl residue appears to act as a signal for transport into the vacuole (Matern et al., 1986; Mackenbrock et al., 1992), and malonylglycosides of isoflavones, isoflavanones and pterocarpans were exclusively located in vacuoles from Cicer arientinum cell suspension cultures (Mackenbrock et al., 1992). Sometimes quite high concentrations of secondary metabolites are reached in vacuoles. For example, in latex vacuoles (vesicles) of Chelidonium majus, concentrations of sanguinarine, chelidonine and berberine up to 5001000 mM were found (Hauser and Wink, 1990), and latex vacuoles of Papaver somniferum sequester up to 500mM morphine (Pham and Roberts, 1991). Since many of these compounds are toxic and dangerous to the plant producing them, Matile (1984) considers the vacuole mainly as a site for detoxification, and coined the term “toxic compartment”. This is a plausible description from the point of view of plant biochemistry. But as explained in the introduction, “defence and signal compartment” (Wink, 1993b) would better describe the functional role and corresponding ecological importance. The close integration of vacuolar sequestration into the defence strategy of a plant is exemplified in the following. In Sorghum, the cyanogenic glycosides are stored in epidermal vacuoles which, however, lack P-glucosidase, which is localized in chloroplasts of


147

TABLE I1 Vacuolar sequestration of defence and signal compounds Compounds

Phenolics Anthocyanins Bergenin Coumaroylglycosides (esculin) Flavonolglycosides

Gallic acid 7-Glucosylpleurostirnin Isoflavanone malonylglycosides Isoflavone malonylglycosides Kaempherol 3,7-O-glycoside Orientin C-glycosides Pterocarpan malonyl glycosides Quercetin 3-triglucoside 7-Rhamnosyl-6-hydroxyluteolin Shikimic acid Tricin 5-glucoside

Terpenoids Convallatoxin and other cardenolides Gen tiopicroside Oleanolic acid (3-0-glucoside) Oleanolic acid (3-0-glucuronide) Primary cardiac glycosides lanatoside A, C; purpureaglycoside A Saponines (avenacosides) Oligosaccharides Gentianose Gen tiobiose Stachyose Nitrogen-containing compounds (excluding alkaloids) Cyanogenic glycosides (linamarin) Glucosinolates Sinapoylglucosides

Alkaloids A j m alicine Atropine Nicotine

References Wiermann (1981), Hrazdina et al. (1982), Ishikura (1981) Taneyama (1992) Oba et al. (1981), Werner and Matile (1985) Wiermann (1981), Matern (1987), Hopp and Seitz (1987), Van Genderen and Van Hemert (1986), Hrazdina el al. ( 1982) Taneyama (1992) Harborne et al. (1993) Mackenbrock et al. (1992) Mackenbrock et al. (1992) Schnabl et al. (1986) Harborne et al. (1993) Mackenbrock et al. (1992) Weissenbock et al. (1986) Harborne er al. (1993) Hollhnder-Czytko and Amrhein (1983) Harborne et al. (1993) Loffelhardt et al. (1979) Keller (1986) Szakiel and Janiszowska (1993) Szakiel and Janiszowska (1993) Kreis and Reinhard (1987), Christmann et al. (1993) Urban et al. (1983) Keller and Wiemken (1982) Keller and Wiemken (1982) Keller and Matile (1985)

Saunders and Conn (1978) Liithy and Matile (1984), Wei et al (1981) Sharma and Strack (1985) Deus-Neumann and Zen k (1984), Renaudin (1989) Mende and Wink (1987) Saunders (1979); Renaudin and Guern ( 1987)

148

M. WINK

TABLE I1 (continued) Vacuolar sequestration of defence and signal compounds Compounds Secondary metabolites Berberine Betaine Betalaines Capsaicin Catharanthine Codeine Dopamine Lupanine Morphine Noscapine Papaverine Polyamines (S)-Reticuline Sanguinarine Scopolamine (S)-Scoulerine Senecionine N-oxide Serpentine Solanidine Thebaine Vindoline Defence proteins Ph ytohaemagglutinin Protease inhibitors Chitinase p-1,3-Glucanase

References Sato et al. (1993, 1994) Matoh et al. (1987) Leigh (1983) Fujiwake et al. (1980) Deus-Neumann and Zenk (1984) Pham and Roberts (1991), Roberts (1 987) Homeyer and Roberts (1984), Roberts (1987) Mende and Wink (1987) Pham and Roberts (1991), Roberts ( 1987) Pham and Roberts (1991) Pham and Roberts (1991), Roberts ( 1987) Pistocchi et al. (1988) Deus-Neumann and Zenk (1986) Matile et al. (1970) Deus-Neumann and Zenk (1984) Deus-Neumann and Zenk (1986) Ehmke et al. (1987, 1988) Deus-Neumann and Zenk (1984), Blom et al. (1991b) Han et al. (1989) Pham and Roberts (1991), Roberts (1987) Deus-Neumann and Zenk (1984), Brisson et al. (1992) Chrispeels (1991), Chrispeels and Raikhel (1992) Ryan and Walker-Simmons (1983) Mackenbrock et al. (1992) Mackenbrock et al. (1992)

adjacent cells (Saunders and Conn, 1978; Kojima et a f . , 1979; Wajant et a f . , 1995). Upon wounding, for example by a herbivore, the cellular integrity breaks down, and both vacuolar contents and P-glucosidase come into contact (Fig. 1). The cyanogenic glycosides are hydrolysed. Hydroxynitrile lyase, which releases HCN, is found in the cytoplasm of both epidermal and mesophyll cells. The HCN generated is a strong inhibitor of mitochondria1 respiration and thus a strong toxin. In Hevea brasifiensis leaves, the cyanogenic glycoside linamarin is also exclusively stored in the vacuole, and the linamarase in the apoplastic space (Gruhnert et al., 1994), indicating that


149

henzaldehyde

C3N

7''$ CW

I!-glucosl-

R

dase dhurrin

hydroxynitril-

plucote

lyase

HCN

Fig. 1. Compartmentation of the cyanogenic glycoside dhurrin in Sorghum leaves and its degrading enzymes. (After Kojima et al. (1979) and Wink (1993b).)

the spatial compartmentation of cyanogenic glycosides and their metabolizing enzymes have evolved several times during evolution. In root tissues of horseradish (Armoracia rusticana), the glucosinolates in addition to ascorbate (an activator of myrosinase) are also stored in vacuoles (Fig. 2), but the hydrolysing enzyme, the myrosinase, is localized in the cytoplasmic membranes and in cell walls (Matile, 1980, 1984; Luthy and Matile, 1984). Upon wounding, all components come into contact, and mustard oil is released, which is a strong animal deterrent, a membranedestabilizing agent and an antibiotic. The storage and activation of coumaroylglycosides follows a similar strategy

150

M. WINK

sinigrine

ally1isothiocyanate

Fig. 2. Compartmentation of glucosinolates and myrosinase in the roots of horseradish (Armoracia rusticana). (After Matile (1984) and Wink (1993b).)

(Oba et al., 1981; Alibert et al., 1985). In Melilotus alba, trans- and 2-cis-2-hydroxycinnamicacid is sequestered in the vacuoles of epidermal and mesophyll cells; because of the abundance of mesophyll cells, they contain 90% or more of the glucosides present in leaves. The corresponding P-glucosidase appeared to be localized in the extracytoplasmic space. When leaves of Melifotus are wounded, the compartmentation breaks down, and both glucosides and P-glucosidase come into contact. As a result, coumarins are generated, which function as active defence compounds. Quinolizidine alkaloids, which figure as the characteristic secondary metabolites of many legumes, especially within the Papilionoideae (Wink,


151

1993c; Kass and Wink, 1995), are produced by chloroplasts of leaf mesophyll cells (Wink and Hartmann, 1982). After synthesis they are exported to the vacuole, apparently by means of a tonoplast proton antiport carrier (Mende and Wink, 1987). Some of the alkaloids are transported via the phloem (Wink and Witte, 1984) to all other parts of a lupin plant, especially the stems and fruits. In stems and petioles, the alkaloids enter the epidermal cells and are taken up into the vacuole against a concentration gradient by a transport system (Wink and Mende, 1987) (Fig. 3). Concentrations of quinolizidine alkaloids in epidermal vacuoles can be as high as 200mM. Since these alkaloids serve as defence compounds against herbivores (e.g. insects) the epidermal localization can be interpreted as a means to place them at a strategically important site, where they can ward off intruders just when they start nibbling at a plant (Wink, 1985, 1987, 1988, 1992). Cardiac glycosides are produced in a number of unrelated plant families, such as the Scrophulariaceae, Apocynaceae, Ranunculaceae, Brassicaceae and Asclepiadaceae. In Digitalis lanata it could be shown that cardenolides are synthesized in green tissue. Following synthesis, the primary glycosides (such as lanatoside A and C) with a terminal glucose in the sugar side-chain (but not so the secondary glycosides) are accumulated against a concentration gradient in the vacuoles of the source tissue, or after phloem transport in those of the sink tissues (Holz et a f . , 1992; Christmann et a f . , 1993). Cardiac glycosides inhibit Na+,K+-ATPase, and are thus strong toxins in animals. Although cardiac glycosides are usually very effective in defence, a number of insects are known which have overcome this barrier and which use the dietary toxins for their own defence. A well-studied example is the monarch (Danaus pfexippus), which became insensitive to cardiac glycosides through a single point mutation in the ouabain-binding site of Na+,K+-ATPase (Holzinger et al., 1992; Holzinger and Wink, 1996). Peptides and proteins which are employed in defence are also often localized in the vacuole, and include protease inhibitors, lectins, enzymes or other toxalbumins. In the case of the toxic proteins, which are typical for many seeds, we can see a dual function: besides their role as defence compounds, they serve at the same time as nitrogen stores. Upon wounding or infection, the synthesis of these defence proteins (such as chitinase or protease inhibitors) can be enhanced, and the newly made proteins may end up in the vacuole or the extracellular space/cell wall, depending on their sorting signals (Chrispeels, 1991). Some storage proteins have very low abundances of essential amino acids, such as methionine in seeds of legumes or lysine in Gramineae. These deficiencies can also be interpreted as an antiherbivore strategy, because a herbivore will show symptoms of ill health and growth retardation when reared on such a diet. These examples clearly demonstrate the intricate but cooperative interplay between different compartments, tissues and their integration in the overall defence strategy of a plant.

152

M. WINK

H+

ATP

ADP

1y , "+

>TT

p+@ H+

CYTOPLASM

H+

1

I

VACUOLE

SECONDARY MEI’ABOLITES AND XENOBIOTICS B.

1.53

XENOBIO’TICS

As mentioned in the introduction, xenobiotics are metabolized in plants to form glycosides which may further be acylated by malonic acid; other conjugates are with glutathione and cysteine. As a general rule xenobiotics enhance their hydrophilic properties, i.e. water solubility, by these procedures, which take place in the cytoplasm. It has been shown for a number of xenobiotics, especially for some pesticides, that the corresponding conjugates are sequestered in the vacuole (Coupland, 1991). For example, the plant growth regulator (2,4dich1orophenoxy)acetic acid (2,4-D) diffuses into plant cells ( e . g in Phuseolus vulgaris), where it becomes hydroxylated in the 4-position. The hydroxylation occurs via the “NIH shift” mechanism and involves the displacement of the CI group (Thomas et a l . , 1964). 4-OH-2,s-D is then glucosylated and malonylated; 0-(ma1onyl)glucosyl 4-OH-2,4-D is stored in the vacuole (Schmitt and Sandermann, 1982; Sandermann, 1987). In the soya bean, 2,4-D forms amide conjugates with various amino acids instead (especially glutamyl and aspartyl derivatives) (Mumma and Davidonis, 1983). In addition, the precursor of ethylene is stored as 1-(ma1onylamino)cyclopropane-I-carboxylic acid in the vacuole (Bouzayen et al., 1989). The herbicide paraquat slowly accumulates in the vacuoles of root cells of maize seedlings, but is also translocated to the shoot (DiTomaso et al., 1993). Xenobiotics and secondary metabolites stored in the vacuole can be modified by some enzymes, for example peroxidases which are often also sequestered in the vacuole (for a review, see Wink, 1093b). It was shown that anthocyanins, anthocyanidins, alkaloids arid many other compounds were oxidized by vacuolar peroxidases (Calderon et al., 1992; Wink, 1994). I t should be recalled that peroxidases may have evolved for the detoxification of microbial toxins. In addition to vacuolar storage. some conjugated xenobiotics, such as the 0-(malony1)glucoside of pentachlorophenol and N-(malonyl)-3,4dichloroaniline. are released from the cells and stored in the apoplast (Winkler and Sandermann, 1989). I n plant cell cultures, an export of secondary metabolites and other compounds into the medium (“extracellular lytic and storage compartment”) can be regularly observed (Sandermann, 1994; Wink, 198.5, 1994). For example, in cell cultures of Lupinuspolyphyllus we even observed a malonyl derivative of lupanine (which was hitherto unknown from quinolizidine alkaloids) in the medium (Wink, 1994). Fig. 3. Compartmentation of quinolizidine alkaloids in stems of Lupinus polyphyllus. (a) Vacuolar sequestration in epidermal cells. Recall that the synthesis of these alkaloids takes place in the chloroplasts of leaf mesophyll cells (Wink, 1987, 1 9 9 3 ~ ) . (b) Schematic drawing of lupanine transport across the tonoplast (Mende and Wink, 1987). employing a lupanine proton antiport mechanism.

154

M. WINK C. MECHANISMS UNDERLYING VACUOLAR SEQUESTRATION

I . Uptake across the tonoplast Biomembranes are semipermeable, and small and lipophilic molecules, such as 02,COz, N2 or benzene, can pass membranes rapidly by free diffusion. Polar metabolites, such as sugars, ions and charged molecules (e.g. amino and other acids), will diffuse freely at a very low rate. The tonoplast does not appear to differ it its semipermeability from other membranes, such as the plasmalemma. For example, lipophilic synthetic dye probes, such as methyl red, methylene blue, auramine 0 and acridine orange, pass both membranes readily by diffusion (Wink, 1990). Dyes that carry charged sulfate groups or are glycosides, such as procion blue, carminic acid, alizarin red and indigocarmine, cannot pass either membrane by free diffusion (Wink, 1990). For the charged or polar molecules, the tonoplast and the plasmalemma thus form penetration barriers. As discussed by Martinoia and Ratajczak and Blumwald and Gelli (this volume), the uptake of ions, sugars and amino acids is achieved with aid of special channels, carriers and pumps (Hediger, 1994), some of which have already been purified and reconstituted in artificial membranes, such as the carriers for glutamine o r arginine (Thume and Dietz, 1991) and malate (Martinoia et a f . , 1991). In a few cases the genes for corresponding plasma membrane carriers have been isolated and expressed (Sauer and Tanner, 1989; Riesmeier et a f . , 1992; Hsu et a f . , 1993; Tsay et a f . , 1993). Most of the defence and signal compounds and conjugated xenobiotics found in the vacuole (Table 11) are hydrophilic, polar (such as sugars and glycosides) or even charged molecules, such as non-protein amino acids, alkaloids, some glycosides or proteins. Since these compounds have not been synthesized in the vacuole, they have to pass the tonoplast first in order to accumulate in the vacuole. Considering the structural diversity of secondary metabolites, we cannot assume that the biochemical mechanisms which lead to their sequestration in vacuoles are identical. More lipophilic metabolites will pass the tonoplast by simple diffusion, as was shown for several weakly basic alkaloids, such as nicotine, ajmalicine, colchicine, vinblastine, ergotamine, sanguinarine, vindoline, quinine and cinchonamine (Kurkdjian, 1982; Renaudin and Guern, 1987; McCaskill et al., 1988; Renaudin, 1989; Hauser and Wink, 1990; Blom et a f . , 1991a). Also, ascorbate and dehydroascorbate appear to reach the vacuole of Hordeum wufgareprotoplasts without a carrier, although transport across the plasmamembrane was carrier mediated (Rautenkranz et al., 1994). Other alkaloids, which are charged species under cytosolic pH conditions or polar glycosides, appear to pass the tonoplast with aid of a carrier mechanism. Examples are the alkaloids (S)-scoulerine, (S)-reticuline, catharanthine (Deus-Neumann and Zenk, 1984, 1986), atropine, lupanine,


155

sparteine, 13-hydroxylupanine (Mende and Wink, 1987), senecionine-Noxide (Ehmke et al., 1987, 1988), dopamine (Homeyer and Roberts, 1984) and polyamines (Pistocchi et al., 1988). Examples for glycosides and other polar metabolites include primary cardiac glycosides (Kreis and Reinhard, 1987; Kreis and Holz, 1991), coumaroylglycosides (Alibert et al., 1985; Rataboul et al., 1985; Werner and Matile, 1985), acylated anthocyanins (Hopp and Seitz, 1987), flavonoids such as apigenin-7O-(p-D-malonylglucoside) (Matern et al., 1986; Matern, 1987), 1-(malonylamino)cyclopropane-1-carboxylicacid (Bouzayen et al., 1989) and glucosides and glucuronides of the triterpene oleanolic acid. In the case of latex vacuoles of Papaver sornniferurn, morphine uptake was ATP stimulated, but the authors suggest the presence of an alkaloid channel instead of a specific morphine transporter (Roberts et a f . , 1991). It is remarkable that for (S)or (R)-reticuline or (S)- and (R)-scoulerine, these carriers discriminated the naturally occurring (S)-configurated compounds and were thus stereoselective (Deus-Neumann and Zenk, 1986). Whether particular carriers exist for all these allelochemicals or whether they can hijack existing carriers for primary metabolites cannot be stated with certainty at present. The transport system for l-aminocyclopropane-lcarboxylic acid (ACC) can be classified as a neutral L-amino acid carrier with a high affinity for ACC and other non-polar aminoacids (Saftner, 1994), supporting this possibility. Oat aleurone protoplasts and epidermal cells of Allium cepa take up a number of fluorescent membrane probes and sequester them in the vacuole. The uptake of carboxyfluorescein, lucifer yellow, cascade blue hydrazide and sulforhodamine G into vacuoles can be inhibited by probenecid, indicating that the transport is carrier mediated (Oparka et al., 1991; Wright and Oparka, 1994). It is very likely that this carrier is responsible for the transport of indigenous compounds and that these xenobiotics have hijacked it. Since the above-mentioned membrane probes have physicochemical properties similar to some phloem-mobile xenobiotics, these results have obvious implications for the detoxification and compartmentation of xenobiotics in plants (Wright and Oparka, 1994). For other xenobiotics, evidence has been presented that derived glutathione derivatives (such as glutathione S-conjugates of N-ethylmaleimide and of metolachlor) cross the vacuolar membrane with the aid of a group-specific transporter that is widely distributed in plants (Martinoia et a f . , 1993; Li et al., 1995a,b, 1996) and which is remarkably similar to the glutathione S-conjugate export pumps of mammalian liver. 2. Vacuolar sequestration against a concentration gradient The concentrations of ions, sugars, acids, signal and defence chemicals is often remarkably high in vacuoles and orders of magnitude lower in the cytoplasm (Leigh et a f . , 1981; Boller and Wiemken, 1986). Thus, all these molecules have to be sequestered in the vacuole against a concentration

156

M. WINK

gradient. The question to be considered next concerns, therefore, the driving force for uphill accumulation in vacuoles. As discussed by Liittge and Ratajczak, and Davies in this volume, the major tonoplast protein is an H+-ATPase (Leigh and Walker, 1980; Sze, 1985; Bremberger et al., 3988), which in addition to a pyrophosphatase (Leigh and Walker, 1980; Rea and Sanders, 1987; Bremberger et al., 1988; Zehn et al., this volume) transports protons into the vacuole (Thom and Komor, 1984; Sze, 1985; Rea and Sanders, 1987; Hedrich et al., 1989; Taiz, 1992). Thus, the vacuolar hydrogen ion concentration is orders of magnitude higher than that of the cytoplasm. This gradient (proton motive force, Ph4F) can be utilized for secondary active transport systems (Reinhold and Kaplan, 1984; Blumwald, 1987; Hedrich and Schroder, 1989; Kurkdjian and Guern, 1989), such as substrate proton antiport mechanisms (Hager and Hermsdorf, 1981; Liittge et ul., 1981; Thom and Komor, 1984; Blumwald and Poole, 1985a,b; Briskin ef al., 1985; Blumwald, 1987; Blackford et al., 1990; Getz, 1991) as discussed by Martinoia and Ratajczak and Blumwald and Gelli in more detail in this volume. There is some experimental evidence that some secondary metabolites and conjugates of xenobiotics are also transported into the vacuole by an H + antiport mechanism, and examples include lupanine (Mende and Wink, 1987), (S)-reticuline, (S)-scoulerine (Deus-Neumann and Zenk, 1986), and l-(malony1amino)cyclopropane-l-carboxylicacid (Bouzayen et al., 1989). In all these cases, vacuolar uptake was dependent on Mg ATP and could be inhibited by reagents which dissipate proton gradients. In addition, lupanine transport was enhanced by K+ ,indicating either that the membrane potential could be additionally involved (Mende and Wink, 1987) or that K+ activates the H+-translocating pyrophosphatase as was shown for beet vacuoles (Leigh, 1983). Conversely, although the transport of glutathione S-conjugates of xenobiotics was MgATP-dependent, transport was not inhibited by compounds which disrupt secondary activated uptake processes, i.e. proton antiport mechanisms cannot be the driving force in this case. Instead, the carrier appears to be a specific ATPase with a pronounced sensitivity to vinblastine, vanadate and verapamil, similar to the export pumps in mammalian liver (Martinoia et al., 1993; Tommasini et al., 1993; Li et al., 199Sa). The yeast cadmium factor protein (YCF1) is a vacuolar glutathione S-conjugate pump and also shows substantial sequence homology to the human multidrug resistance-associated protein (MRPl), and might also be related to the plant glutathione S-conjugate transporter (Li et al., 1996; Tommasini et af., 1996). In plants, the transport of oxidized glutathione (GSSG) is also achieved by this pump and is competitively inhibited by the glutathione S-conjugate of the herbicide metolachlor (Tommasini et ai., 1993). Once in the vacuole the conjugates are degraded by a carboxypeptidase, suggesting that glutathione S-conjugates represent the


157

transport form but not the storage form of xenobiotics (Wolf et al., 1996). The activity of these transporters appears to be inducible by some xenobiotics, such as herbicide antidotes (so-called safener) (Gaillard e l al., 1994; Li et al., 199Sb). Recently, it has been shown that the protein encoded by the Bronze-2 gene in maize, which is responsible for the deposition of anthocyanins in the vacuole, is a glutathione S-transferase (Marrs et al., 1995). Thus, anthocyanins, e.g. cyanidin-3-glucoside, as well as xenobiotics are conjugated with glutathione. Many herbicide conjugates are metabolized to cysteine conjugates and then acylated with malonic acid. Since malonylated cyanidin-3-glucoside is the major maize anthocyanin, Marrs et al. (1995) postulate by analogy that anthocyanin is transported into the vacuole as the gluthatione S-conjugate with the aid o f the ATP-dependent glutathione S-conjugate “export” pump. Once in the vacuole, anthocyanins are further converted lo the malonyl derivative. However, Hopp and Seitz (1987) had shown that a carrier system exists for acylated anthocyanins at the tonoplast which would contradict the assumption of Marrs et al. (1995). Whether the glutathione conjugate pump also transports other secondary metabolites remains to be shown. It has been emphasized that an apparent uphill transport can also be achieved by certain “trapping” reactions (Matile, 1978, 1984; Boller and Wiemken, 1986). Normally, passive transport or diffusion will come to a standstill when equal concentrations are reached on both sides of the tonoplast. If we assume that the molecule which enters the vacuole is trapped or otherwise changed, then it is removed from the equilibrium, and the transport process can go on until all trapping molecules are exhausted and equal concentrations of “free” molecules are reached on both sides. The trapping reactions discussed (Table 111) include “ion traps” (which could be relevant for nitrogenous compounds, which become protonated in the vacuole (Nishimura, 1982; Boller and Wiemken, 1986; Guern et al., 1987; Renaudin, 1989), e.g. alkaloids do not pass membranes as protonated molecules except when highly lipophilic (Kurkdjian, 1982; Renaudin, 1989; Hauser and Wink, 1990)). The monoterpene indole alkaloids ajmalicine and vindoline appear to cross the tonoplast by simple diffusion (Renaudin and Guern. 1987; McCaskill et al., 1988; Renaudin, 1989; Blom et al., 199lb) and not by carrier-mediated transport, as reported earlier (Deus-Neumann and Zenk, 1984, 1986). In the vacuole. ajmalicine is effectively converted into the more polar serpentine by basic peroxidases, which cannot leak out of the vacuole (Blom et al., 1991b). It has been suggested that intravacuolar serpentine binds to the tonoplast (Pradier et a l . , 1988). Other trapping reactions include crystallization (often observed for calcium oxalate; Franceschi and Horner, 1980), conformational changes (e.g. apigenin 7-0-(6-0-malonylglycoside) (Matern et al., 1983, 1986; Matern, 1987) and cyanidin-3-O-sinapoylxylosylglycosylgalactoside (Hopp and Seitz, 1987), and isomerization (discussed for cis- and trans-coumarylglycosides;

158

M. WINK

TABLE I11 Mechanismsfor uptake and sequestration of defencetsignal compounds and xenobiotics in vacuoles ~~

Uptake mechanisms Membranelvesicle fusion: -Endoplasmic reticulum vesicles (e.g. prolamin protein bodies) Simple diffusion: -Lipophilic secondary metabolites Active/passive transport (channeldtransporters): -Ions, amino acids, organic acids, sugars, polar secondary metabolites; conjugated xenobiotics; glutathione S-conjugates Mechanisms for uphill accumulation Active transport: -ATPase-coupled “pumps” Secondary active transport: -Proton gradients generated by H+-ATPase, pyrophosphatase (proton motive force) used by proton antiport carriers Trapping mechanisms: -Protonation of alkaloids in acidic vacuoles -Binding of secondary compounds to tannin, other phenolics or polyphosphates -Complexation of alkaloids with chelidonic acid, meconic acid -Binding of secondary metabolites to proteins? -Crystallization (calcium oxalate) -Conjugation -Change of conformation or configuration

Rataboul et ul., 1985). Whereas the trans to cis isomerization is almost irreversible under natural conditions and would trap the coumarylglycosides permanently in the vacuole, the conformational shift observed with acylated flavonoids and anthocyanins is pH-dependent and thus reversible (Matern, 1987). Another trapping procedure could be the binding of molecules to complexing compounds such as tannins or polyphosphates (Matile, 1978). Examples for such complexing reagents are chelidonic and meconic acids. Isolated latex vacuoles of Chelidonium mujm take up sanguinarine and various other lipophilic alkaloids (Matile, 1978; Hauser and Wink, 1990). It could be shown that these vesicles contain about 660mM (range 200-1300 mM) chelidonic acid, which readily complexes alkaloids (Hauser and Wink, 1990). The reaction of these alkaloids with chelidonic acid obviously provides the buffer for apparent accumulation against a concentration gradient. We do not know, however, the mechanism for concentrating chelidonic acid in vesicles. In the case of latex vacuoles of Papaver somniferum, the concentration of meconic acid can be as high as 250mM,


159

and was considered as relevant for trapping of morphine, codeine, papaverine and other poppy alkaloids (Pham and Roberts, 1991). In cells of Coptis juponicu, the berberine content of the vacuoles was correlated with their malic acid accumulation; the resulting malate-berberine complex might trap berberine in the vacuole (Sato el ul., 1992). The accumulation of lupanine in epidermal vacuoles was also favoured by malate, suggesting a similar mechanism (Wink and Mende, 1987). Crystallization, which has been observed for the alkaloids berberine and sanguinarine and for calcium oxalate (Renaudin and Guern, 1990), is another, albeit rare, trapping reaction. Summarizing, it is apparent that the uptake of metabolites into the vacuole can be directly or secondarily by energy-dependent export “pumps” or proton antiports, or can be accomplished by various trapping or binding reactions. The trapping mechanisms are compatible with both transport and diffusion processes.

111. CONCLUSIONS A typical feature of plants and other sessile organisms is the production of secondary metabolites, which can be considered as chemical defence and signal compounds. They are mediators of plant-plant, plant-herbivore and plant-microorganism interactions, and thus important for the fitness of plants. Although chemical defence appears to be the major function, some coloured or scented metabolites play important additional roles in reproductive biology (attraction of pollinating or seed-dispersing animals). Defence and signal functions can only be achieved if local concentrations of secondary metabolites are high enough. Most of these compounds are synthesized in the cytosol or in plastids. Since many of the allelochemicals are also toxic for the producing plant, they need to be stored in a separate compartment, which is the vacuole in the case of hydrophilic compounds. The storage of vacuolar defence or signal compounds is often tissue-specific, e.g. many compounds are accumulated in a strategically favourable position, such as epidermal cells, which have to ward off small enemies in the first place. Here, the vacuoles function as “defence or signal compartments”. Lipophilic compounds may cross the tonoplast by simple diffusion, whereas hydrophilic ones, such as amino acids, organic acids, ions and many polar allelochemicals (alkaloids, glycosides), are taken up by carrier- o r channelmediated processes. The driving force for uphill transport can be proton gradients which are generated by tonoplast-associated H+ ATPases and H+ pyrophosphatases. In other cases, diverse trapping processes within the vacuole seem to be involved. In general, xenobiotics of plant, microbial or industrial origin are handled by plants in a similar way to secondary metabolites, although some marked differences also exist. Vacuolar sequestration is a most important trait for

160

M. WINK

both secondary metabolites and man-made pesticides; in the future, it will be important to discuss results from both fields in a closer context as reactions seen today obviously evolved under similar evolutionary constraints. Depending on the cell type in question, a vacuole can function in a variety of ways: besides turgor regulation the vacuole can serve as a lytic, storage, defence or signal compartment (Wink, 1993b). It seems likely that some of these functions are exclusive, and not all vacuoles within one cell or within the same tissue must have identical properties. To make the picture even more complex, all these functions must be considered as dynamic and not static, being regulated in space and time (Boller and Wiemken, 1986) and under environmental constraints.

ACKNOWLEDGEMENTS Work from our laboratory was supported by grants of the Deutsche Forschungsgemeinschaft. I would like to thank my co-workers, D r P. Mende, Dr M. T. Hauser and Dr R. Perrey for cooperation, and Mrs C. Theuring, U. Schade, M. Weyerer and U. Dostal for technical assistance.

REFERENCES Alibert, G., Boudet, A. M., Canut, H. and Rataboul, P. (1985). Protoplasts in studies of vacuolar storage compounds. In “Physiological Properties of Plant Protoplasts” (P. E. Pilet, ed.), pp. 105-115, Springer-Verlag, Heidelberg. Baldwin, B. C. (1977). Xenobiotic metabolism in plants. I n “Drug Metabolism - From Microbe to Man” (D. V. Parke and R. L. Smith, eds), pp. 191-217, Taylor and Francis, London. Bernays, E. A. and Chapman, R. F. (1994). “Host-Plant Selection by Phytophagous Insects”. Chapman and Hall, New York. Bernays, E. A. and Graham, M. (1988). On the evolution of host specificity in phytophagous arthropods. Ecology 69, 886-892. Blackford, S . , Rea, P. A. and Sanders, D. (1990). Voltage sensitivity of H+/Ca++ antiport in higher plant tonoplast suggests a role in vacuolar calcium accumulation. Journal of Biological Chemistry 265, 9617-9620. Blom, T. J. M.,Van Vliet, T. B., Schripsema, J., Val, J., Van h e n , F., Verpoorte, R. and Libbenga, K. R. (1991a). Uptake and accumulation of the alkaloids quinine and cinchonamine in cultured cells of Cinchona robusta and Catharanthus roseus. Journal of Planr Physiology 138, 436442. Blom, T. J. M., Sierra, M.,Van Vliet, T. B., Franke-van Dijk, M. E., De Koning, P., Van hen, F., Verpoorte, R. and Libbenga, K. R. (1991b). Uptake and accumulation of ajmalicine into isolated vacuoles of cultured cells of Cuthuranthus roseus (L.) G. Don. and its conversion into serpentine. Planta 183, 170-177. Blumwald, E. (1987). Tonoplast vesicles as a tool in the study of ion transport at the plant vacuole. Physiologia Plantarum 69, 731-734. Blumwald, E. and Poole, R.J.(1985a). Na+/H+-antiport in isolated tonoplast vesicles from storage tissue of Beta vulgaris. Plant Physiology 78, 163-167. Blumwald, E. and Poole, R.J. (1985b). Nitrate storage and retrieval in Beta vulgaris:


161

effect of nitrate and chloride on proton gradients in tonoplast vesicles. Proceedings of the National Academy qf Sciences of the USA 82, 3683-3687. Boller, T. and Wiemken, A. (1986). Dynamics of vacuolar compartmentation. Annual Review of Plant Physiology 37, 137-164. Bouzayen, M., Latche, A., Pech, J.-C. and Marigo, G. (1989). Carrier-mediated uptake of I-(ma1onylamino)-cyclopropane-1-carboxylic acid in vacuoles isolated from Catharanthus roseus cells. Plant Physiology 91, 1317-1322. Bremberger, C., Haschke, H.-P. and Liittge, U. (1988). Separation and purification of the tonoplast ATPase and pyrophosphatase from plants with constitutive and inducible crassulacean acid metabolism. Pluntu 175, 465-470. Briskin, D. P., Thornley, W. R. and Wyse, R. E. (1985). Membrane transport in isolated vesicles from sugarbeet taproot. 11. Evidence for a sucrose/K+-antiport. Plant Physiology 78, 871-875. Brisson, L., Charest, P. M., De Luca, V. and Ibrahim, R. K. (1992). Immunocytochemical localization of vindoline in mesophyll protoplasts of Catharanthus roseus. Phytochemistr-v 31. 465470. Calderon, A. A., Garcia-Florenciano, E.. Munoz, R. and Ros Barcelo, A. (1992). Gamay grapevine peroxidase: its role in vacuolar anthocyani(di)n degradation. Vifis 31, 139-147. Christmann, J., Kreis, W. and Reinhard, E. (1993). Cardenolide sinks and occurrence of cardenolides in the sieve tubes of Digitalis lanata. Botanica Acta 106, 419-427. Chrispeels, M. J. (1991). Sorting of proteins in the secretory system. Annual Review of Plant Physiology and Plant Molecular Biology 42, 21-53. Chrispeels, M. J . and Raikhel, N. V. (1992). Short peptide domains target protein to vacuoles. Cell 68, 613416. Cole, D. J . (1994). Detoxification and activation of agrochemicals in plants. Pesticide Science 42, 2OY-222. Coupland, D. (1991). The role of compartmentation of herbicides and their metabolites in resistance mechanisms. I n “Herbicide Resistant Weed Crops” (J. C. Caseley, G . W. Cussans and R. K. Atkin, eds), pp. 263-278. Butterworth-Heinemann, Oxford. Deus-Neumann, B. and Zenk, M. H. (1984). A highly selective alkaloid uptake system in vacuoles of higher plants. Planta 162, 250-260. Deus-Neumann, B. and Zenk, M. H. (1986). Accumulation of alkaloids in plant vacuoles does not involve an ion-trap mechanism. Plunta 166, 44-53. DiTomaso, J. M., Hart, J. J. and Kochian, L. V. (1993). Compartmentation analysis of paraquat fluxes in maize roots as a means of estimating the rate of vacuolar accumulation and translocation to shoots. Plant Physiology 102, 467-472. Dodge, A. D. (ed.) (1989). Herbicides and plant metabolism. “Society of Experimental Biology, Seminar Series 38”. Cambridge University Press, Cambridge. Donaldson, R. P. and Luster, D. G. (1991). Multiple forms of plant cytochromes p-450. Plant Physiology 96, 669-674. Ehmke, A , , von Borstel, K. and Hartmann, T. (1987). Specific uptake of the N-oxides of pyrrolizidien alkaloids by cells, protoplasts and vacuoles from Senecio cell cultures. In “Plant Vacuoles” (B. Marin, ed.), NATO AS1 Series, Vol. 134. pp. 301-304. Plenum Press, New York. Ehmke, A , , von Borstel, K. and Hartmann, T. (1988). Alkaloid-N-oxides as transport and vacuolar storage compounds of pyrrolizidine alkaloids in Senecio vulgaris. Planta 176, 83-90. Franceschi, V. R. and Homer, H. T. (1980). Calcium oxalate cristals in plants. Botunical Reviews 46, 361-427.

162

M. WINK

Fraenkel, G. (1959). The raison d’Ctre of secondary substances. Science 129, 1466-1470. Fujiwake, H., Suzuki, T. and Iwai, K. (1980). Intracellular localisation of capsaicin and its analogues in Capsicum fruit. Plant Cell Physiology 21, 1023-1030. Gaillard, C., Dufaud, A., Tommasini, R., Kreuz, K., Amrhein, N. and Martinoia, E. (1994). A herbicide antidote induces the activity of both the herbicide detoxifying enzyme for the detoxified herbicide. FEBS Letters 352, 219-221. Getz, H. P. (1991). Sucrose transport in tonoplast vesicles of red beet roots is linked to ATP hydrolysis. Planta 185, 261-268. Gonnet, J . F. and Hieu, H. (1992). In situ micro-spectrophotometric and microspectrocolorimetric investigation of vacuolar pigments in flowers of cultivars of carnation (Dianthus caryophyllus). Journal of Horticultural Science 67, 663676. Gruhnert, C., Biehl, B. and Selmar, D. (1994). Compartmentation of cyanogenic glucoside and their degrading enzymes. Planta 195, 36-42. Guern, J . Renaudin, J. P. and Brown, S. C. (1987). The compartmentation of secondary metabolites in plant cell cultures. In “Cell Culture and Somatic Cell Genetics” (F. Constabel and I. Vasil, eds), pp. 43-76. Academic Press, New York . Hager, A. and Hermsdorf, P. (1981). A H+/Ca++-antiporter in membranes of microsomal vesicles from Maize coleoptiles, a secondary energized Ca++ pump. Zeitschrift fur Naturforschung 36c, 1009-1012. Han, S. R., Campbell, W. F. and Salunkhe, D. K. (1989). Ultrastructural localization of solanidine in potato tubers. Journal of Food Biochemistry 13, 377-389. Harborne, J . B. (1993). “Introduction to Ecological Biochemistry”, 4th edn. Academic Press, New York. Harborne, J. B., Greenham, J., Williams, C. A., Eagles, J . and Markham, K. R. (1993). Ten isoprenylated and C-methylated flavonoids from the leaves of three Vellozia species. Phytochemistry 34, 219-226. Hathway, D.E. (1989). “Molecular Mechanisms of Herbicide Selectivity”. Oxford University Press, Oxford. Hauser, M.-T. and Wink, M. (1990). Uptake of alkaloids by latex vesicles and isolated mesophyll vacuoles of Chelidonium majus (Papaveraceae). Zeitschrift fur Naturforschung 45c, 949-957. Hediger, M.A. (1994). Structure, function and evolution of solute transporters in procaryotes and eucaryotes. Journal of Experimental Biology 196, 1 5 4 9 . Hedrich, R. and Schroder, J. (1989). Physiology of ion channels and ion pumps of higher plant cells. Annual Review of Plant Physiology and Plant Molecular Biology 40,539-569. Hedrich, R., Kurkdjian, Guern, J. and Flugge, U. I., (1989). Comparative studies on the electrical properties of the H + translocating ATPase and pyrophosphatase of the vacuoladlysosomal compartment. E M B O Journal 8, 28352841. Hollander-Czytko, H. and Amrhein, N. (1983). Subcellular compartmentation of shikimic acid and phenylalanine in buckwheat cell suspension cultures grown in the presence of shikimate pathway inhibitors. Plant Science Letters 29, 89-96. Holz, H., Kreis, W., Haug, B. and Reinhard, E. (1992). Storage of cardiac glycosides in vacuoles of Digitalis lanata mesophyll cells. Phytochemistry 31, 11671171. Holzinger, F. and Wink, M. (1996). Mediation of cardiac glycoside insensitivity in the monarch (Danaus plexippus): role of an amino acid substitution in the


163

ouabain binding site Na+.K+-ATPase. Journal of C‘hemical Ecology 22, 193 1- 1947. Holzinger, F., Frick, C. and Wink. M. (1992). Molecular basis for the insensitivity of the monarch (Danuus plexippus) to cardiac glycosides. FEBS Letters 314, 477-480. Homeycr. B. C. and Roberts, M. F. (1984). Dopatnine accumulation in Papaver somnifentm latex. Zeitschrift fiir Naturforschung 39c. 1034-1037. Hopp, W. and Seitz, H. U. (1987). The uptake of acylated anthocyanin into isolated vacuoles from a cell suspension culture of Daucus carotu. Plunta 170. 74-85. Hrazdina, G . , Marx. G. A. and Hoch, H. C. (1982). Distribution of secondary plant metabolites and their biosynthetic enzymes in pea (Pisum sativum) leaves. Anthocyanins and flavonol glycosides. Plunt Physiology 70, 745-748. Hsu, L.-C.. Chiou, T.-J., Chen, L. and Bush. D. R. (1993). Cloning a plant amino acid transporter by functional complementation of a yeast amino acid transporter mutant. Proceedings of the Nirtional Academy of Sciences of the USA 90, 7441-7445. Inderjit. Dakshini, K. M. M. and Einhellig, F. A. (19%). “Allelopathy, Organisms, Processes and Applications”. American Chemical Society Symposium Series 582. American Chemical Society, Washington. DC. Ishikura, N . (1981). Flavonoids in the petal cells of Atzagallis arvensis f . coerulea containing a blue crystalline anthocyanin. Zcitschrift f u r Pflanzenphysiologie 103, 469473. Jacoby, W. B. and Ziegler, D. M . (1990). The enzymes of detoxification. Journal of Biological Chemistry 265, 20 7 15-20 7 1 8 . Kiss, E. and Wink, M. (1995). Molecular phylogeny of the Papilionoideae (family Leguminosae): RhcL gene sequences versus chemical taxonomy. Botunica Acta 108, 149-162. Keller, F. (1986). Gentiopicroside is located in the vacuoles of root protoplasts of Gentiuna Iutea. Journal of Plant Physiology 122, 473-476. Keller, F. and Matile, P. (1985). The role o f the vacuole in storage and mobilization of stachyose in tubers of Stachys sioboldii. Journal of Plant Physiology 119, 369-380. Keller, F. and Wiemken, A. (1982). Differential compartmentation of sucrose and gentianose in the cytosol and vacuoles o f storage root protoplasts from Getitiana lutea. Plant MI Reports I , 274-277. Kojima, M., Poulton, J . E., Thayer, S. and Conn, E . E. (1979). Tissue distribution of dhurrin and of enzymes involved in its metabolism in leaves of Sorghum bicolor. Plant Physiology 63. 1022-11128. Kreis, W. and H d z , H. (1991). Zellularer Transport und Speicherung von Naturstoffen. Naturwissetischaftliche Rundschuu 44. 463470. Kreis, W. and Reinhard. E. (1987). Selective uptake and vacuolar storage of primary cardiac glycosides by suspension-cultured Digitalis lanata cells. Journal of Plant Physiology 128, 31 1-326. Kurkdjian, A. (1982). Absorption and accumulation of nicotine by Acer pseudoplatanus and Nicotiuna tubacum cells. Phvsiologiu Vegetalis 20, 73-83. Kurkdjian, A. and Guern, J. (1989). Intraccllular pH: measurements and importance in cell activity. Annual Review of Plimt Physiology and Plant Molecular Bio1og.y 40, 271-303. Langebartels, C., Harms, H.. Arjmand, M., Schiifer. W. and Sandermann, H. (1986). Cell wall components as binding sites for xenobiotics. I n “Cell Walls ’86”. (B. Vian, D. Reis and R. Goldberg, eds), pp. 158-163, LJniversitie Pierre and Marie Curie, Paris.

164

M.WINK

Leigh, R. A. (1983). Methods, progress and potential for use of isolated vacuoles of solute transport in higher plants. Physiologia Plantarum 57, 390-396. Leigh, R. A. and Walker, R. R. (1980). ATPase and acid phosphatase activities associated with vacuoles isolated from storage roots of red beet. Planta 150, 222-229. Leigh, R. A., Ahmad, N. and Wyn Jones, R. G. (1981). Assessment of glycinebetaine and proline compartmentation by analysis of isolated beet vacuoles. Planta 153, 34-41. Levin, D. A. (1976). The chemical defences of plants to pathogens and herbivores. Annual Review of Ecology and Systematics 7, 121-159. Li, Z.-S., Zhao, Y. and Rea, P. A. (1995a). Magnesium adenosine 5’-triphosphateenergized transport of glutathione-S-conjugates by plant vacuolar membrane vesicles. Plant Physiology 107, 1257-1268. Li, Z.-S., Zhen, R.-G. and Rea, P. A. (1995b). l-Chloro-2,4-dinitrobenzene-elicited increase in vacuolar glutathione-S-conjugate transport activity. Plant Physiology 109. 177-185. Li, Z.-S.,’Szczypka, M., Lu, Y.-P., Thiele, D. J. and Rea, P. A. (1996). The yeast cadmium factor protein (YCF1) is a vacuolar glutathione S-conjugate pump. Journal of Biological Chemistry 271, 6.5094517. Loffelhardt, W., Kopp, B. and Kubleka, W. (1979). Intracellular distribution of cardiac glycosides in leaves of Convallaria majalis. Phytochemistry 18, 12891291. Luthy, B. and Matile, P. (1984). The mustard oil bomb: rectified analysis of the subcellular organization of the myrosinase system. Biochemie und Physiologie der Pjlanzen 179, 5-12. Luttge, U., Smith, A., Marigo, G. and Osmond, C.B. (1981). Energetics of malate accumulation in the vacuoles of Kalanchoe tubiflora cells. FEBS Letters 126, 81-84. McCaskill, D. G., Martin, D. L. and Scott, A. I. (1988). Characterization of alkaloid uptake by Catharanthus roseus (L.) G . Don protoplasts. Plant Physiology 87, 402-408. Mackenbrock, U . , Vogelsang, R. and Barz, W. (1992). Isoflavone and pterocarpan malonylglucosides and /3-1,3-glucan- and chitin-hydrolases are vacuolar constituents in chickpea (Cicer arientinum L.) . Zeitschrift fur Naturforschurig 47c, 815-822. Mader, M. (1984). Die pflanzliche Vakuole aus biochemischer Sicht. Biologie in Unserer Zeit 14, 171-176. Marrs, K. A,, Alfenito, M. R., Lloyd, A. M. and Walbot, V. (1995). A glutathione S-transferase involved in vacuolor transfer encoded by the maize gene Bronze-2, Nature 375, 397-400. Martinoia, E., Vogt, E., Rentsch, D. and Amrhein, N. (1991). Functional reconstitution of the malate carrier of barley mesophyll vacuoles in liposonies. Biochimica et Biophysica Acta 1062, 271-278. Martinoia, E., Grill, E., Tommasini, R., Kreuz, K. and Amrhein, N. (1993). ATP-dependent glutathione-S conjugate “export” pump in the vacuolar membrane of plants. Nature 364, 247-249. Marty, F., Branton, D. and Leigh, R. A. (1980). Plant vacuoles. In “The Biochemistry of Plants: A Comprehensive Treatise” (N. E. Tolbert, ed.), Vol. 1, pp. 625-658. Macmillan Press, New York. Matern, U. (1987). Die Isomerenfalle fur Sekundarmetabolite, eine Alternative zum Ionenfallen-Modell. Biologie in Unserer Zeit 17, 148-152. Matern, U., Heller, W. and Himmelspach, K. (1983). Conformational changes


165

of apigenin7-0-(6-0-malonylglycoside), a vacuolar pigment from parsley, with solvent composition and proton concentration. European Journal o f Biochemistry 133, 439448. Matern, U., Reichenbach, C. and Heller, W., (1986). Efficient uptake of flavonoids into parsley (Petroselinum hortense) vacuoles requires acylated glycosides. Planta 167, 183-189. Matile, P. (1978). Biochemistry and function of vacuoles. Annual Review of Plant Physiology 29, 193-213. Matile, P. (1980). The “mustard oil bomb”: compartmentation of myrosinase systems. Biochemie und Physiologie der Pflanzen 175, 722-731. Matile, P. (1984). Das toxische Kompartiment der Pflanzenzelle. Naturwlssenschuften 71, 18-24. Matile, P., Jans, B. and Rickenbacher, R. (1970). Vacuoles of Chelidonium latex: lysosomal property and accumulation of alkaloids. Biochemie und Physiologie der Pfanzen 161, 447-458. Matoh, T., Watanabe, J. and Takahashi, E. (1987). Sodium, potassium, chloride, and betaine concentrations in isolated vacuoles from salt-grown Atriplex grnelini leaves. Plant Physiology 84, 173-177. Mende, P. and Wink, M. (1987). Uptake of the quinolizidine alkaloid lupanine by protoplasts and vacuoles of Lupinus polyphyllus cell suspension cultures. Journal of Plant Physiology 129, 229-242. Menn, J. J . (1978). Comparative aspects of pesticide metabolism in plants and animals. Environmental Health Perspectives, 113-124. Mumma, R. 0. and Davidonis, G. H. (1983). Plant tissue culture and pesticide metabolism. In “Progress in Pesticide Biochemistry and Toxicology” (D. Hutson and T. R. Roberts, eds), Vol. 3, pp. 255-278. J. Wiley, Chichester. Nishimura, M. (1982). pH in vacuoles isolated from castor bean endosperm. Plant Physiology 70, 742-744. Oba, K., Conn, E., Canut, H . and Boudet, A. M. (1981). Subcellular localization of 2-(/3~-glucosyloxy-)cinnamicacids and the related /3-glucosidase in leaves of Melilotus alba. Plant Physiology 68, 1359-1363. Oparka, K. J., Murant, E. A , , Wright, K . M., Prior, D. A. M. and Harris, N. (1091). The drug probenicid inhibits the vacuolar accumulation of fluorescent anions in onion epidermal cells. Journal o f CeN Science 99, 557-563. Pham, T. D. T. and Roberts, M. F. (1991). Quantitative characterization of the contents of Papaver somniferum latex vacuoles. Phytochemical Analysis 2, 68-73. Pistocchi, R., Keller, F., Bagni, N. and Matile, P. (1988). Transport and subcellular localisation of polyamines in carrot protoplasts and vacuoles. Plant Physiology 87, 514-518. Pradier, J. M . , Barbier-Brygoo, H., Ephitikhine, G. and Guern, J . (1988). Interaction of an alkaloid, serpentine, with tonoplast vacuoles from Catharanthus roseus G . Don. Compte Rendue Academie Sciences, Series 3 306, 283-289. Rataboul, P., Alibert, G., Boller, T. and Boudet, A. M. (1985). Intracellular transport and vacuolar accumulation of o-coumaric acid glucoside in Melilotus alba mesophyll cells protoplasts. Biochimica et Biophysica Acta 816, 25-36. Rautenkranz, A. A. F., Li, L., Machler, F., Martinoia, E. and Oertli, J. J. (1994). Transport of ascorbic and dehydroascorbic acids across protoplast and vacuole membranes isolated from barley (Hordeum vugare L. cv Gerbel) leaves. Plant Physiology 106, 187-193. Rea, P. and Sanders, D. (1987). Tonoplast energizations: two H+-pumps, one membrane. Physiologiu Plantarurn 71, 131-141.

166

M. WINK

Reinhold, L. and Kaplan, A. (1984). Membrane transport of sugars and amino acids. Annual Review of Plant Physiology 35, 45-83. Renaudin, J . P. (1989). Different mechanisms control the vacuolar compartmentation of ajmalicine in Catharanthus rosem cell cultures. Plant Physiology and Biochemistry (Paris) 27, 613-621. Renaudin, J . P. and Guern, J . (1987). Ajamlicine transport into vacuoles isolated from Catharanthus roseus. In “Plant Vacuoles” (B. Marin, ed.), NATO AS1 Series, Vol. 134, pp. 339-347. Plenum Press, New York. Renaudin, J. P. and Guern, J . (1990). Transport and vacuolar storage of secondary metabolites in plant cell cultures. In “ Secondary Products from Plant Tissue Culture” (B. C. Charwood and M . J . C. Rhodes, eds), Proceedings of the Phytochemical Society of Europe 30, pp. 59-78. Rice, E. L. (1984). “Allelopathy”, 2nd edn. Academic Press, Orlando. Riesmeier, J . W., Willmitzer, L. and Frommer, W. B. (1992). Isolation and characterisation of a sucrose carrier cDNA from spinach by functional expression in yeast. EMBO Journal 11, 4705-4713. Roberts, M. F. (1981). Enzymic synthesis of coniceine in Conium maculaturn chloroplasts and mitochondria. Plant Cell Reports 1, 10-13. Roberts, M. F. (1987). Papaver latex and alkaloid storage vacuoles. In “Plant Vacuoles” (B. Marin, ed.), NATO AS1 Series. Vol. 134, pp. 513-528. Plenum Press, New York. Roberts, M. F., Homeyer, B. C. and Pham, T. D. T. (1991). Further studies of sequestration of alkaloids in Papaver somniferum latex vacuoles. Zeitschrift fur Naturforschung 46c, 377-388. Rosenthal, G. A . and Berenbaum, M. R. (1991). “Herbivores. Their Interaction with Secondary Plant Metabolites”. Academic Press, San Diego. Ruesink, A. (1980). Protoplasts of plant cells. Methods of Enzymology 69, 69-84. Ryan, C. A. and Walker-Simmons, M. (1983). Plant vacuoles. Methods of Enzyrnology 96, 580-589. Saftner, R. A. (1994). Stereoselectivity and structural determinants in molecular recognition by the ACC transport system in isolated maize mesophyll vacuoles. Physiologia Plantarum 92, 543-554. Sandermann, H. (1987). Pestizid-Ruckstande in Nahrungspflanzen. Die Rolle des pflanzlichcn Metabolismus. Naturwissenschaften 74, 573-578. Sandermann, H. (1994). Higher plant metabolism of xenobiotics: the “green liver” concept. Pharmacogenetics 4, 225-241. Sandermann, H., Diesperger, H. and Scheel, D. (1985). Metabolism of xenobiotics by plant cell cultures. In “Plant Tissue Culture and its Biotechnological Applications” (W. Barz, E. Reinhard and M. H. Zenk, eds), pp. 178-196. Springer-Verlag, Heidelberg. Sato, H., Kobayashi, Y., Fukui, H. and Tabata, M. (1990). Specific differences in tolerance to exogenous bernerine among plant cell cultures. Plant Cell Reports 9, 133-136. Sato, H., Taguchi, G., Fukui, H. and Tabata, M. (1992). Role of malic acid in solubilizing excess berberine accumulating in vacuoles of Coptis japonica. Phytochemistry 31, 345 1-3454. Sato, H., Tanaka, S. and Tabata, M. (1993). Kinetics of alkaloid uptake by cultured cells of Coptis japonica. Phytochemistry 34, 697-701. Sato, H . . Tanaka, T., Tanaka, S. and Tabata, M. (1994). Binding of berberine to a membrane fraction from Coptis cells. Phytochemistry 36, 1363-1367. Sauer, N. and Tanner, W. (1989). The hexose carrier from Chlorella: cDNA cloning of a eukaryoitic Ht-contransporter. FEBS Letters 259, 4346.


167

Saunders, G. A. (1979). Investigations of vacuoles isolated from tobacco. I. Quantification of nicotine. Plant Physiology 64, 74-78. Saunders, G. A. and Conn, E. E. (1978). Presence of the cyanogenic glycoside dhurrin in isolated vacuoles from Sorghum. Plant Physiology 61, 154-157. Schmitt, R. and Sandermann, H. (1982). Specific localisation of P-D-glucoside conjugates of 2,4-D in soybean vacuoles. Zeitschrift fur Naturforschung 37c, 772-777. Schnabl, H., Weissenbock, G. and Scharf, H. (1986). In vivo microspectrophotometric characterization of flavonol glycosides in Vicia faba guard and epidermal cells. Journal of Experimental Botany 37, 61-72. Sharma, V. and Strack, D. (1985). Vacuolar localisation of 1-sinapoyl glucose: L-malate sinapoyltransferase in protoplasts from cotyledons of Raphanus sutivus. Planta 163, 563-568. Swain, T. (1977). Secondary compounds as protective agents. Annual Review of Plant Physiology 28, 479-501. Szakiel, A. and Janiszowska, W. (1993). The kinetics of transport of oleanolic acid monoglycosides into vacuoles isolated from Calendula ofFcinalis leaf protoplasts. Acta Biochimia Polonia 40, 136-138. Sze. H. (1985). H+-translocating ATPase: advances using membrane vesicles. Annual Review of Plant Physiology 36, 175-208. Taiz, L. (1992). The plant vacuole. Journal of Experimental Biology 172, 113-122. Taneyama. M. (1992). Studies on C-glycosides in higher plants. IV Intracellular distribution of bergenin and related compounds in Sarifraga stolonifera leaves. Botanical Magazine, Tokyo 105, 565-571. Thorn, M. and Komor. E. (1984). H+-sugar antiport as the mechanism of sugar uptake by sugar cane vacuoles. FEBS Letters 173, 1 4 . Thomas, E. W . , Loughman, B. C. and Powell, R. G. (1964). Metabolic fate of 2,4-D in the stem tissue of Phaseolus vulgaris. Nature 204, 884-885. Thume, M. and Dietz. K.-H. (1991). Reconstitution of the tonoplast amino-acid carrier into liposomes. Planta 185, 569-575. Timmermann, K . P. (1989). Molecular characterisation of corn glutathione-Stransferase isozymes involved in herbicide detoxification. Physiologia Plantarum 77, 465471. Tommasini, R., Martinoia, E., Grill, E . , Dietz, K. J. and Amrhein, N. (1993). Transport of oxidised glutathione into barley vacuoles: evidence for the involvement of the glutathione-S-conjugate ATPase. Zeitschrift fur Naturforscliung 48c, 868-871. Tommasini, R., Evers, R., Vogt, E., Mornet, C., Zaman, G. J. R., Schinkel, A. H., Borst, P. and Martinoia, E. (1996). The human multidrug resistanceassociated protein functionally complements the yeast cadmium resistance factor 1. Proceedings of the National Academy of Sciences of the USA 93, 67436748. Tsay, Y.-F., Schroeder, J., Feldmann, K . A . and Crawford, N. M. (1993). The herbicide sensitivity gene CHLl of Arabidopsis enocodes a nitrate-inducible nitrate transporter. Cell 72, 705-713. Urban, B.. Laudenbach, U. and Kesselmcier, J . (1983). Saponin distribution in the etiolated leaf tissue and subcellular localization of steroidal saponins in etiolated protoplasts of oat, Avena sativa. Protoplusma 118, 121-123. Van Genderen, H. H. and Van Hemert, J . (1986). Sites of flavone accumulation in some genotypes of Silene pratensis (Rafn.) Godron et Gren. (Caryophyllaceae). New Phytologist 104, 3S39. Wagner. G . J . (1982). Compartmentation in plant cells: the role of the vacuole. In

168

M. WINK

“Recent Advances in Phytochemistry” (L. Creasy and G. Hrazdina, eds), Vol. 16, pp. 1-45. Plenum Press, New York. Wajant, H., Riedel, D., Benz, S. and Mundry, K.-W. (1995). Immunocytological localization of hydroxynitrile lyases from Sorghum bicolor L. and Linum usitatissimum L. Plant Science 103, 145-154. Waller, G. R. (1987). “Allelochemicals: Role in Agriculture and Forestry”. American Chemical Society Symposium Series 330. American Chemical Society, Washington, DC. Wei, X., Roomans, G. M., Seveus, L. and Pihakaski, K. (1981). Localization of glucosinolates in roots of Sinapis alba using X-ray microanalysis. Scanning Electron Microscopy 2 , 481-488. Weissenbock, G., Hedrich, R. and Sachs, G. (1986). Secondary phenolic products in isolated guard cell, epidermal cell and mesophyll cell protoplasts from pea (Pisum sativum L.) leaves: distribution and determination. Protoplasma 134, 141-148. Werner, C. and Matile, P. (1985). Accumulation of coumaroylglucosides in vacuoles of barley mesophyll protoplasts. Journal of Plant Physiology 118, 237-249. Wiermann, R., (1981). Secondary plant products and cell and tissue differentiation. In “The Biochemistry of Plants”, Vol. 7, pp. 85-116. Academic Press, New York. Willenbrink, J. (1987). Die pflanzliche Vacuole als Speicher. Naturwissenschaften 74, 22-29. Wink, M. (1985). Composition of the spent culture medium. Time course of ethanol formation and the excretion of hydrolytic enzymes into the culture medium of suspension-cultured cells of Lupinus polyphyllus. Journal of Plant Physiology 121, 287-293. Wink, M. (1987). Quinolizidine alkaloids: Biochemistry, metabolism, and function in plants and cell suspension cultures. Planta Medica 53, 509-514. Wink, M. (1988). Plant breeding: Importance of plant secondary metabolites for protection against pathogens and herbivores. Theoretical Applied Genetics 7 5 , 225-233. Wink, M. (1990). Physiology of secondary product formation in plants. In “Secondary Products from Plant Tissue Culture” (B. C. Charwood and M. J. C. Rhodes, eds). Proceedings of the Phytochemical Society of Europe 30, pp. 23-41. Wink, M. (1992). The role of quinolizidine alkaloids in plant-insect interactions. In “Insect-Plant Interactions (E. A. Bernays, ed.), Vol. 4, pp. 131-166. CRC Press, Boca Raton. Wink, M. (1993a). Allelochemical properties or the raison d’etre of alkaloids. In “The Alkaloids.” (G. A. Cordell, ed.), Vol. 43, pp. 1-118. Academic Press, San Diego. Wink, M. (1993b). The plant vacuole: a multifunctional compartment. Journal of Experimental Botany 44(Supplement), 231-246. Wink, M. (1993~).Quinolizidine alkaloids. In “Methods in Plant Biochemistry” (P. G. Waterman, ed.), Vol. 8, pp. 197-239. Academic Press, London. Wink, M. (1994). The cell culture medium - a functional extracellular compartment of suspension cultured cells. Plant Cell, Tissue and Organ Culture 38, 307-319. Wink, M. and Hartmann, T. (1982). Localization of the enzymes of quinolizidine alkaloid biosynthesis in leaf chloroplasts of Lupinus polyphyllus. Plant Physiology 70, 74-77. Wink, M. and Mende, P. (1987). Uptake of lupanine by alkaloid-storing epidermal cells of Lupinus polyphyllus. Planta Medica 53, 465469. Wink, M. and Witte, L. (1984). Turnover and transport of quinolizidine alkaloids:


169

diurnal variation of lupanine in the phloem sap, leaves and fruits of Lupinus albus L. Planta 161, 519-524. Winkler, R. and Sandermann, H. (1989). Plant metabolism of chlorinated anilines: isolation and identification of N-glucosyl and N-malonyl conjugates. Pesticide Biochemistry and Physiology 33, 239-248. Wolf, A. E., Dietz, K.-J. and Schroeder, P. (1996). Degradation of glutathione S-conjugates by a carboxypeptidase in the plant vacuole. FEBS Letters 384, 31-34. Wright, K. M. and Oparka, K. J. (1994). Physicochemical properties alone do not predict the movement and compartmentation of fluorescent xenobiotics. Journal of Experimental Botany 45, 35-44.


Solute Composition of Vacuoles

R. A. LEIGH

Biochemistry and Physiology Department, IACR-Rothamsted, Harpenden, Hertfordshire A L5 2JQ, UK

I. 11.

Introduction

.....

.................

........

Variability of Vacuolar Solute Composition .. ..................... A . X Ray Microanalysis .......................... ..................... B. Ion-selective Microelectrodes ....................................... C. Single-Cell Sampling and Analysis (SiCSA

111.

Regulation of Vacuolar Solute Pools ..

IV.

A Model

References

............ 182

...............................................................................

I.

172 175

189

INTRODUCTION

The pre ence of a I rge, water-filled vacuole occupying the majority of the intracellular volume confers a number of benefits on plant cells (Raven, 1987, this volume), including the ability to accumulate a wide variety of solutes to relatively high concentrations. Without a vacuole, plant cells would be constrained to accumulating only those solutes that are compatible with the effective operation of metabolic processes in the cytoplasm, so restricting the intracellular concentrations of solutes which interfere with metabolism through toxicity or feedback inhibition. However, the presence of a vacuole overcomes thcse problems by effectively separating the solutes within it from mainstream metabolism, allowing them to accumulate to high concentrations without affecting the rest of the cell. The flexibility that this confers on the solute relations of plant cells has been utilized to good effect in a number Advaricch iii Boianic,il Rcqearch incorporiitirig Advance, 111 Plan1

ISBN 0-12-IXISY?S-X

Vol. 25 P;ithology

Copynphl 0 1997 Acildcmlc Prea, Limited All rights of reproduction in aiiy Iorm reserved

172

R. A. LEIGH

of different ways. For instance, in situations where the soil contains high concentrations of potentially toxic ions (e.g. Na+ and C1- in saline soils), they can, nonetheless, be taken up and used beneficially to lower vacuolar sap osmotic potential while metabolically compatible ions and other solutes are accumulated in the cytosol (Wyn Jones et af.,1977; Leigh and Wyn Jones, 1986). In other cases, the presence of the vacuole allows plants to store soluble products of metabolism (e.g. malate in plants with crassulacean acid metabolism or sucrose in sugar beet) without risk of them being further metabolized or of them allosterically inhibiting enzymes. In addition, it is now also clear that the vacuole has an important role in intracellular signalling by acting as a reservoir for the storage of free Ca2+, which is released in response to specific stimuli (e.g. Johannes et af., 1992; Allen and Sanders, this volume). In a similar way, the vacuole is important in cytosolic pH regulation by providing a large sink into which H + can be pumped (Raven, 1987). Finally, toxic secondary metabolites with important roles in plant defence are compartmentalized into the vacuole and released only when the cell is damaged (Wink, 1993, this volume; Bennett and Wallsgrove, 1994). Here I shall concern myself with the factors that influence the concentrations of osmotically significant solutes within plant vacuoles. This is an area which has been reviewed extensively in the past (e.g. Matile, 1978, 1987; Boller and Wiemken, 1986; Leigh and Wyn Jones, 1986; Wink, 1993), but many of the conclusions have been based on extrapolations from whole-tissue analysis or from the use of techniques such as vacuole isolation or nuclear magnetic resonance (NMR) which give information about the composition of populations of vacuoles rather than of individual organelles. Although much of our knowledge of the behaviour of solutes in vacuoles still rests on these approaches, and I shall draw on them extensively below, recent methodological developments have made it possible to measure quantitatively the concentration of individual solutes in individual vacuoles. These methods are revealing details about the relationships between vacuolar composition, whole-tissue composition, solute supply, and water relations parameters (particularly osmotic pressure and turgor) which will be important in understanding whether, and how, the composition of vacuoles is regulated. They are also indicating that there is considerable intercellular heterogeneity in vacuolar composition. Therefore, my main aim is to show how these techniques are confirming and/or changing views on vacuolar solute composition that have been gained from earlier studies.

11. VARIABILITY OF VACUOLAR SOLUTE COMPOSITION It is well established that plant tissues and cells can vary greatly in their solute composition depending upon the prevailing supply of solutes from the environment and the rate at which compounds are used or produced in

SOLUTE COMPOSITION OF VACUOLES

173

metabolism. I shall not attempt to summarize this large literature, which has been reviewed before by others (e.g. Matile, 1978, 1987; Boller and Wiemken, 1986; Leigh and Wyn Jones, 1986; Wink, 1993), but instead will restrict myself to a few examples that illustrate the type of information which can be gained. The flexibility that plant cells have over their solute composition as a result of having a vacuole is illustrated by the work of Mott and Steward (1972). They showed that both the total sap osmotic pressure and the concentrations of ions and sugars in carrot explants changed with growth conditions (Fig. 1). In explants grown initially in a culture medium in which the major solutes were 1 m M KCI and 1 m M NaCl, the composition of the cell sap was dominated by these salts (Fig. lA), but when these explants were transferred to a medium containing 60 mM sucrose the ion concentrations decreased, and after 7 days the solute composition was dominated by sugars (Fig. 1B). This change was accompanied by a rise in sap osmotic pressure due partly to the cells adjusting their turgor in response to the higher external osmotic pressure of the sucrose medium, but also to a non-iso-osmotic replacement of the salts by the sugars. It is also possible to use whole-tissue studies to demonstrate that the storage of compounds in the vacuole is a regulated process which is initiated only in response to particular conditions within the tissue and, once begun, proceeds in balance with other aspects of metabolism. Thus the accumulation of free nitrate in barley leaves (which is known to be the result of deposition of nitrate in the vacuole; Martinoia et al., 1981; Granstedt and Huffaker, 1982; Miller and Smith, 1996) is not started until the total tissue nitrogen concentration has reached a minimum threshold value above which nitrate accumulates linearly as a function of total tissue nitrogen content (Hommels er al., 1989; Zhen and Leigh, 1990). At low internal nitrogen concentrations, all nitrate taken up by the plants is assimilated and none is diverted into storage. However, once the nitrate supply exceeds metabolic needs, excess nitrate is stored in the vacuole. The amount stored is a constant proportion of the extra nitrogen accumulated, indicating that the balance between nitrate assimilation and storage is maintained over a wide range of excess nitrate supply. Similar relationships are also seen for other nutrient ions that are stored in the vacuole (for examples for Pi,see Lee and Ratcliffe, 1983; Kakie, 1969), indicating that the principle is of general utility in understanding processes regulating the storage of nutrient ions. Useful though such whole-tissue studies are, they rely on assumptions about the location of the solutes in the vacuole. Fortunately, more detailed studies that give information directly about the composition of vacuoles have generally confirmed that solutes thought to be vacuolar in location are found in this organelle (for reviews, see Leigh and Wyn Jones, 1986; Miller and Smith, 1996). Thus 3'P-NMR (Roberts, 1984; Lee and Ratcliffe, 1983; Ratcliffe, 1994) has been used to distinguish and quantify the vacuolar and

B

A Ions

82% Sugars 28% OP = 303 mosm/kg

OP = 440 mosm/kg

Fig. 1. Percentage contributions of different solutes to the total solute composition of carrot explants grown for ( A ) 7 days in a medium with 1 mM KC1 and 1 mM NaCl and then (B) transferred for a further 7 days to a medium without salts but containing 60 mM sucrose. OP, osmotic pressure. (Redrawn from the results of Mott and Steward (1972).)


175

cytosolic Pipools in plants and to show that decreases in total tissue Pi content in response to phosphorus deprivation are due entirely to changes i n the vacuolar pool (Lee and Ratcliffe, 1983). Similarly, isolated vacuoles have been used to study the solute composition of this organelle (see Leigh, 1983: Leigh and Wyn Jones, 1986) and have the advantage that, when isolated from individual cell types, they provide information about the behaviour of solutes in these different cells (e.g. Dietz et ul., 1992a,b; Leigh and Tomos, 1993).

However. both NMR and analysis of isolated vacuoles provide information that is an average from populations o f large numbers of vacuoles, and t h e underlying behaviour of solutes within individual vacuoles is not revealed. Thus, while such approaches have confirmed which solutes are in vacuoles and how they behave in response to different conditions, they are inevitably of limited use in reaching a detailed interpretation of what is happening in individual vacuoles. Without knowledge at this finer level it is not possible to say whether changes in vacuolar composition that are seen with averaging techniques result from the same behaviour i n all vacuoles or reciprocal changes occurring in different vacuolcs. Fortunately, studies using techniques such as X ray niicroanalysis (Echlin and Taylor, 1986; van Steveninck and van Steveninck, 1991: Stelzer and Lehmann. 1993), ion-selective microelectrodes (Miller, 1994), and single-cell sampling and analysis (SiCSA; Tomos et af., 1994) are providing information on the composition and behaviour of single vacuoles and thus giving a better basis on which to interpret larger-scale behaviour. A.

X RAY MICROANALYSIS

Although X ray microanalysis is not B fully quantitative technique and thus rarely provides accurate information about the absolute concentrations of solutes in vacuoles or other organelles (Echlin and Taylor, 1986; van Steveninck and van Steveninck, 1991), it can provide information about the relative abundance of different inorganic elements in vacuoles and thus whether different cell types differ significantly i n their composition. Leigh and Storey (1993) used the technique to measure the relative peak heights for potassium, sodium, calcium, phosphorus and chlorine in X ray microanalysis spectra of vacuoles of epidermal and mesophyll cells in leaves of barley plants grown under a variety of nutrient treatments. T h e y showed that the composition of the vacuoles in each cell type was quite distinct, with calcium and chlorine being detectable in epidermal cells but not in mesophyll cells, whereas the reverse is true for phosphorus (Table I). Williams et al. (1993) used the technique in a different mode, and made maps of the distribution of elements in barley leaves and found patterns similar to those observed by Leigh and Storey (1993) and, in addition, showed that magnesium and sulfur are preferentially accumulated in bundle sheath cells.

176

R. A. LEIGH

TABLE I Percentage of mesophyll and epidermal cells in barley leaves with detectable levels of various elements in their vacuoles. The elements were measured by X ray microanalysis, and results are summarized across 38 leaf samples from 13 different treatments. Data from Leigh and Storey (1993) o /'

Cell type

Mesophyll Adaxial epidermis Abaxial epidermis Adaxial plus abaxial

of cells with a detectable level of

Total number of cells measured

nutrient P

c1

Ca

K

Na

414 641 610 1251

87 0 0 0

5 57 56 56

0

64 53 59

75 60 51 56

51 38 31 35

Other X ray microanalysis studies of leaves from cereals (Hodson and Sangster, 1988; Boursier and Lauchli, 1989; Huang and van Steveninck, 1989) have reached the same general conclusions as the above studies, but results from other plants (e.g. Lupinus luteus (Treeby et al., 1987; Treeby and van Steveninck, 1988) and Atriplex spongiosa (Storey et al., 1983)), while confirming that cells differ in their vacuolar compositions, have found different patterns from those observed in barley. Thus, in L . luteus (Treeby and van Steveninck, 1988) phosphorus concentrations were higher in vacuoles of epidermal cells than in photosynthetic cells, the complete opposite to that seen in barley leaves. The physiological significance of the asymmetric distributions of nutrient elements between different cell types of barley has been discussed in detail elsewhere (Leigh and Storey, 1993; Leigh and Tomos, 1993; Williams et al., 1993). The separation of calcium and phosphorus may be of particular importance in preventing precipitation of phosphorus as calcium phosphate and thus for maintaining a store of readily available Pi for photosynthesis in mesophyll cells. The apparent exclusion of chlorine from the vacuoles of mesophyll cells may be important in salinity tolerance, and the finding by Huang and van Steveninck (1989) that the more salt-tolerant barley cultivar California Mariout was better at preventing chlorine accumulation in the vacuoles of mesophyll cells than the more salt sensitive cultivar Clipper seems to support this. Despite its high spatial resolution, X ray microanalysis has limitations that make it difficult to interpret the observations unequivocally. Thus, although calibration curves can be obtained showing that the X ray signal rises in proportion to the concentration of an element (e.g. Koyro and Stelzer, 1988; Huang and van Steveninck, 1989), there is some doubt that the results are truly accurate (van Steveninck and van Steveninck, 1991; Stelzer and


177

Lehmann, 1993). In addition, the lower limits of detection for many of the elements of interest are in the range 2 W 0 m M (Lazof and Lauchli, 1991), and thus there may still be osmotically and metabolically significant concentrations of elements in vacuoles which are giving no detectable X ray signal. Also, the signal obtained is derived from all chemical forms of an element, and thus assumptions have to be made about the chemical entities that are present. This is not a problem for elements such as potassium which are always present as ions, but for others such a calcium which can be present as the free ion, chelates or insoluble compounds, additional information is needed to interpret the results. Fortunately, development and deployment of multibarrelled ion-selective microelectrodes and of the SiCSA technique have provided ways of overcoming these limitations, and offer the possibility of measuring ions and other solutes in plant cells with high sensitivity and good spatial resolution.

B.

ION-SELECTIVE MICROELECTRODES

Ion-selective microelectrodes (Thomas, 1978; Ammann, 1986; Felle, 1993; Miller, 1994) have been widely used to measure ions in plant cells (e.g. see Penny and Bowling, 1974, 1975; Penny et a l . , 1976; Kelday and Bowling, 1980; MacRobbie and Lettau, 1980; Felle and Bertl, 1986; Miller and Sanders, 1987;Zhen et al., 1991;Walker et al., 1995, 1996). They are useful because they provide a direct determination of activity which is more thermodynamically relevant than concentration and, by necessity, the results are from individual cells. The value of this approach is exemplified by the work of Penny and Bowling (1974, 1975; Penny et al., 1976), who used different types of ion-selective electrodes to measure the changes in ion concentrations in different cells of the stomatal complex of Cornrnelina communis during opening and closing. The results (Table 11) show that there are large differences in pH and K+ and CI- concentrations in different cell types in the complex and that the directions of the ion gradients between different cells are reversed during opening and closing of the stomatal pore. Thus the cell type and its physiological state is a major determinant of the concentrations of ions within it. In subsequent work, Kelday and Bowling (1980) used the same approach to measure CI- concentrations in different cell types in C. communis roots. They found no evidence for radial gradients of C1- concentration across the root but showed that the concentration of C1- in the external solution determined the concentration of this ion in the cells (27-35 mM when grown in 12 mM C1-, 2.7-3.5 mM when in 0.3mM c1-) . Penny and colleagues assumed that the tip of the ion-selective electrode was in the vacuole and thus did not consider the possibility that some of the differences they measured might have been due to the electrode being

178

R. A. LEIGH

TABLE I1 Ion-selective microelectrode measurements of pH and of Ki, and Cl- concentrations in cells of open and closed stomata1 complexes of Commelina communis Ion concentration (mM) or pH"

c1-

K+

Cell type ~

Open

Closed

Open

Closed

Open

Closed

5.60

5.19 5.60 5.78 5.74

448 293 08 73

95 156 199 448

121 62 47 86

33 35 55 117

~.

Guard cell Inner lateral subsidiary Outer lateral subsidiary Epidermal

ND 5.56 5.11

'Data for pH from Penny and Bowling (1975); for K+ from Penny and Bowling (1974); for CI- from Penny et al. (1976). ND, not determined. inserted into different compartments. Fortunately, the acidic pH values they measured do suggest that the majority of impalements were into the vacuole (Table 11). However, other studies (Miller and Sanders, 1987; Zhen ef ul., 1991; Maathuis and Sanders, 1993) have shown that impalements with ion-selective microelectrodes can yield measurements that clearly divide into two populations, which could indicate either that different cells have distinctly different ion concentrations or that two different compartments are being impaled. Thus, Miller and Sanders (1987) using Ca2+-selective electrodes in cells of the giant alga Nifellopsis found that the values of intracellular free Ca2+ activities fell into two well-separated populations with values between 10-395nM and 0.3-1.56mM, respectively. Based on a knowledge of the cytosolic free Ca2+ concentrations in other cells, they assigned the smaller values to the cytosol and the larger ones to the vacuole. They then showed that illumination of the cells caused a decrease in cytosolic Ca2+ activity which they ascribed to Ca2+ uptake by the chloroplasts. In this work, prior knowledge of the free Ca2+ concentrations expected in the cytosol made it relatively easy to assign the values to different compartments, but Zhen et uf. (1991), who measured two populations of intracellular nitrate concentrations in barley root epidermal cells, had to use additional methods to determine how to assign the values they measured. After growth in 10 mM nitrate for 24 h, measurements with nitrate-selective microelectrodes yielded two populations of nitrate activities with mean values of 5.4 and 41.8mM, respectively. To determine whether one of these was vacuolar they used the SiCSA technique (see below) to sample sap from individual vacuoles, and then measured the nitrate concentration in it using a miniaturized enzymelinked fluorescence assay. The results showed that the population with the larger values was vacuolar in origin, and so the smaller values were assigned to the cytosol.

SOLUTE COMPOSITION OF: VACUOLES

179

The above studies confirm that it is possible to use ion-selective microelectrodes to measure compartmental ion concentrations/activities in different types of plant cells, and these values can be assigned to the cytosol and vacuole with confidence in some cases. However, in situations where the vacuole and cytosol may have similar concentrations of an ion this may be less easy, e.g. in K+-replete cclls where K+ concentrations in the vacuole and cytosol are expected to be similar (Leigh and Wyn Jones, 1984). Fortunately. the development of triple-barrelled microelectrodes by Walker et ul. (1995) has overcome this problem. These electrodes incorporate a pH-sensing barrel in addition to the membrane potential-measuring and the ion-selective barrels normally present in electrodes (see Miller, 1994). The pH-sensing barrel allows unequivocal assignment of the measurements to either the vacuole or cytosol, based on the difference in pH between these compartments (vacuolar pH 5.0-5.5, cytosolic pH 7-7.5; Kurkdjian and Guern, 1989). As well as allowing detailed measurements of the dynamics o f changes in ion activities in the two compartments, the triple-barrelled microelectrodes also provide measurements of the membrane potential and pH gradients across the plasma membrane and tonoplast allowing thermodynamic calculations to be made to predict the need for active or passive ion transport at each of these membranes and the feasibility of the coupling of active fluxes to H + gradients (for examples, see Felle et al., 1992; Miller and Smith, 1992, 1996; Maathuis and Sanders, 1993; Miller, 1994; Walker et al., 1996). Using triple-barrelled K+-selective microelectrodes, Walker et ul. ( 1996) measured the changes in vacuolar and cytosolic K + activities in barley root cells grown at different external K + concentrations. The aim was to test the veracity of the predictions of Leigh and Wyn Jones (1984) concerning the behaviour of vacuolar and cytosolic K-' concentrations in response to changes in tissue K+. The measurements (Fig. 2) confirmed the model and showed that vacuolar K + activity varied proportionately with tissue K + concentration whereas cytosolic K+ activity remained constant over a wide range of tissue K+ concentrations, decreasing only under extreme K+ deficiency. However, several additional observations were made by Walker et al. (1996), indicating the usefulness of being able to make accurate and unequivocal measurements in each compartment in different cell types and of being able simultancously to measure pH as well as K + activity. Firstly, there was a significant difference in the behaviour of cytosolic K+ activities in epidermal and cortical cells of the roots; K+ activity in epidermal cells declined more than that in cortical cells in response to K + deficiency. Secondly, there was no evidence that the K + activity in the vacuole reached a minimum value below which it did not decline. This value, set at 10-20mM by Leigh and Wyn Jones (1984), was invoked by them because it was needed to trigger the proposed decline in cytosolic K + . However, the microelectrode measurements indicate that vacuolar K+ activity declines linearly with tissue K + concentration and does

180

R. A. LEIGH

125

Q E

g loo

t

0

b

25 0

0

25

50

75

100

125

150

175

Tissue K ' (mM)

Fig. 2. Changes in cytosolic (solid symbols) and vacuolar (open symbols) K+ activities in barley root epidermal cells in response to changes in total root K+ concentration. Compartmental K+ activities were determined with triple-barrelled microelecrodes. (Redrawn from the data of Walker et al. (1996).)

not level off in the way predicted (Fig. 2). Thus, although the behaviour of cytosolic K+ concentration conformed to the predictions, the reason for its decline at low tissue K+ probably does not relate to the behaviour of the vacuolar K+ pool. Finally, the simultaneous pH measurements indicated that once the cytosolic K+ activity declined below about 70mM there was a parallel decrease in cytosolic pH, raising the possibility that responses of plants to Kf deficiency are caused by this acidification of the cytosol as well as by the decrease in K+ activity in this compartment. The use of triple-barrelled microelectrodes will allow the behaviour of a number of ions in the vacuole and cytosol to be characterized in great detail and the energetics of their transport at the plasma membrane and tonoplast to be investigated. However, only a small number of ions can be determined simultaneously (depending on how many ion-selective barrels are incorporated into the electrode), and the results cannot be easily related to accompanying changes in other solutes or the water relations parameters of the cells. However, this can be achieved with the SiCSA method. C. SINGLE-CELL SAMPLING AND ANALYSIS (SiCSA)

This SiCSA technique (Tomos et al., 1994) is an elaboration of the pressure probe technique developed originally to measure turgor in plant cells (Hiisken et al., 1978). The approach relies on the observation that when the


181

pressure probe is inserted into a cell, a sample of sap is forced into its tip. By suitable modification of the probe, the speed at which this sample is taken can be increased so that dilution by water flow into the cell is prevented and a representative sample of cell sap is obtained (Malone et al., 1989). The picolitre-sized sample is then be stored under oil, and subsamples taken for the determination of osmotic pressure and the concentrations of inorganic and organic solutes (for description of methods, see Tomos et af., 1994). Various considerations suggest that the sap samples are vacuolar in origin (Malone et al., 1991), and this has been confirmed for barley leaf epidermal cells by showing that malate dehydrogenase is undetectable in the samples (Fricke et a f . , 1994a). However, the same study showed that this enzyme was present in mesophyll cell extracts, indicating that samples from these cells had some cytoplasmic contamination. The advantage of the SiCSA technique is its ability to measure a number of parameters simultaneously. Thus with this single method it is possible to measure turgor, sap osmotic pressure, and the major solutes that contribute to these, all at the resolution of single cells. Hence both water and solute relations of individual cells can be determined and the composition of different cell types analysed in detail in space and in time (Fricke et al., 1994a,b,c, 1995, 1996; Pritchard et al., 1996). The SiCSA technique has been used to confirm that the composition of epidermal cells in barley leaves is distinctly different from that of mesophyll and bundle sheath cells, and has extended the results from X ray microanalysis (see Table I; see also Leigh and Storey, 1993; Williams et af., 1993) by showing that the epidermis also has lower sugar, amino acid and organic acid concentrations than the mesophyll (Fricke et af., 1994a; see also Dietz et al., 1994). The spatial resolution obtainable with SiCSA has been used to show that there are differences between cells within an epidermal layer and between cells in the upper and lower epidermis. The analysis of files of adjacent epidermal cells in the upper epidermis of barley showed that there are definite patterns of solute distribution (Fricke et al., 1995). Cells overlying veins have high concentrations of C1- and low nitrate concentrations, while the reverse is true for cells between the veins. Calcium concentration is highest in cells close to the stomates, and this may be a mechanism for lowering the apoplastic Ca2+ level around the guard cells in order not to interfere with Ca2+ signalling mechanisms involved in stomata1 closing (Fricke ef al., 1995; for a similar role for epidermal trichomes, see De Silva et al., 1996). The ion gradients between epidermal cells are generally maintained in response to environmental treatments such as changes in light intensity (Fricke ef af., 1995) or salinity (Fricke et al., 1996), but d o moderate with age (Fricke et af., 1994b, 1995, 1996) and presumably in response to the overall physiological state of the leaf as found by Penny and Bowling (1974, 1975) in their work on Commelina (see Table 11). Finally, measurements on the upper and lower epidermis in barley have shown significant

182

R. A . LEIGH

differences in the accumulation of C1- and N03- between the two layers, with nitrate preferentially accumulated in the upper epidermis and CI- in the lower (Fricke et al., 1994~).The sap osmotic pressure and the sum of the concentrations of the two anions were similar between the two layers, indicating that the differences in anion composition were due to iso-osmotic ion substitution. These asymmetric distributions of ions between upper and lower epidermis indicate that solutes are not merely passively following the transpiration stream to reach epidermal cells (Leigh and Tomos, 1993). A similar conclusion was reachcd by Atkinson (1991), who showed that the concentration of Ca2+ was similar in the upper and lower epidermal layers of Comrnelina communis despite a four-fold higher rate of transpiration from the lower surface. The implications of this for the pathways of transport of solutes from the xylem to the mesophyll and epidermis have been considered by Leigh and Tomos (1993). They concluded that it is unlikely that the results can be explained by a simple model in which cells are supplied by the transpiration stream which first passes the mesophyll and then moves to the sites of water evaporation in the epidermis. Instead, they proposed that in cereal leaves there is a separation of the supply streams for the two cell populations. The mestome/bundle sheath selectively absorbs the solutes destined for the mesophyll and then transfers them to the mesophyll via either the symplast or the apoplast while solutes destined for the epidermis move directly to these cells along a vein extension pathway (Canny, 1990a,b). As these examples indicate, the ability to determine the solute composition of cells in greater detail is demonstrating that changes in tissue composition cannot be interpreted in terms of the “average” cell. Instead, different solutes are located in different cell types and, as a consequence, the composition of each cell type will respond differently to alterations at the whole tissue level. Hence attempts to modify the solute composition of plants may require a more detailed knowledge of controls operating at the level of individual cells because assumptions based on whole tissue analyses may be in error because the true behaviour of solute pools located in different cell types is masked by this approach.

111. REGULATION OF VACUOLAR SOLUTE POOLS It is clear from data such as those shown in Fig. 2 that the concentration of a solute in the vacuole can vary widely depending on its supply, its metabolism, and the growth requirements of the plant. It now seems clear that there is no lower limit to the concentration to which a solute concentration may decrease in the vacuole, i.e. no threshold lower value (Leigh and Wyn Jones, 1984), and there is also evidence to suggest that there is no upper limit providing there is regulation of turgor. This is in contrast


183

to earlier indications that the maximum vacuolar concentrations of some individual solutes are regulated. Thus, measurements of the K + concentration in leaves of barley and other grasses indicated that the total tissue K + concentration does not normally exceed 200 mM (Asher and Ozanne, 1967; Ahmad and Wyn Jones, 1972; Leigh and Johnston, 1983a,b) even when the K + supply is supraoptimal (Barraclough and Leigh, 1993). Such observations strongly suggested some form of regulation, and measurements with SiCSA in cells of leaves grown under these conditions d o confirm that they contain 200mM K + (Fricke et al., 1996). However, when barley is grown with excessive K + in the nutrient solution (200mM KCI or KN03), the K + concentration in epidermal and mesophyll cells can increase to over 400 mM with a concomitant increase in CI- or N 0 3 - concentration and osmotic pressure (Fricke et a f . , 1996). Similarly, when barley plants are exposed to high concentrations of NaCl the concentrations of Na+ and CI- in the vacuoles of leaf epidermal cells increase with the size of the imposed stress (Fig. 3A; see also Fricke et al., 1996). The rise in Na+ and C1- concentrations in the cells is offset by a concomitant decline in the concentrations of Kf and NO3- but there is still an increase i n sap osmotic pressure which rises proportionately with the increase in external NaCI concentration (Fig. 3B). However, measurements of turgor (Fig. 3B; W. Fricke, personal communication) indicate that the rise in sap osmotic pressure is not matched by a concomitant increase in turgor, suggesting that this parameter is more closely controlled than the osmotic pressure or the concentrations of individual ions. A more extreme version of this response has also been reported in the extreme halophyte Suaeda maritimu. When grown in 400 mM NaCl the osmotic pressures of leaf sap rangcd from about 960 to 1600 mosm kg-' (approximately 2.4-4.0 MPa) depending on leaf age, compared with about 520-800 mosm kg-' (1.3-2.0 MPa) in non-salinized plants (Clipson et a f . , 1985). However, the turgor of leaf epidermal cells was never greater than 0.4 MPa and was lower in the salt-treated plants than in those grown without salt. Thus, as in barley, Suaeda does not translate the increased intracellular osmotic pressure due to salt uptake into an equivalent change in turgor. Thus in both barley and Suaeda there must be an adjustment of the extracellular water potential in parallel with the rise in sap osmotic pressure. In Suaeda this appears to be due to accumulation of NaCl outside t h e cells rather than a hydrostatic pressure resulting from the transpirational component of the cell wall water potential, because immersing the leaves to eliminate transpiration only had a small effect on turgor (Clipson et al., 1985). Adjustment of turgor in response to a large increase in intracellular solute concentration also occurs in storage roots of Beta vulgaris (which includes sugar beet, fodder beet and red beet), and analysis of the different plants indicates that there is some genetic control over vacuolar solute concentrations. Thus, the concentrations of sucrose in the storage root of sugar beet are four-fold or more higher than those in fodder beet (Watson and Selman,

184

R. A. LEIGH

6oo 500

f'

E 400

-

0 .c

L?

-0

300 200 100

0 0

25

50

75

100

125

150

External NaCl concentration(mM)

2.5

I

!-

0.5 I 0

I

I

I

I

I

25

50

75

100

125

I

150

External NeCl concentration(mM)

Fig. 3. The changes in (A) concentrations of K+ (o), NO3- (O), Na+, ( 0 ) and Cl- (B) and (B) of osmotic pressure (m) and turgor ( 0 ) in epidermal cells from barley leaves grown in different external concentrations of NaCI. (Drawn from the data of Fricke et al. (1996) and W. Fricke (personal communication).)

1938), reaching over 1 M in high-yielding varieties (Bell et al., 1996). The increase in sucrose is accompanied by an increase in sap osmotic pressure, which can reach a value of over 1200 mosm kg-I (3.0 MPa) in mature sugar beet storage roots (Bell and Leigh, 1996; Bell et al., 1996). It has been shown that all of the sucrose in red beet storage root is in the vacuole (Leigh et al., 1979; Pollock and Kingston-Smith, this volume), so these increases in sucrose concentration observed in storage roots directly reflect vacuolar behaviour. However, as in the case of leaves of NaC1-stressed barley or


185

Suaeda (see above), there is evidence that the rise in osmotic pressure in not translated completely into a change in turgor. Thus, Tomos et a f . (1992) reported that cells of sugar beet storage roots maintain turgor constant at about 0.7 MPa throughout development even though sap osmotic pressure increased from 0.7 to 2.0 MPa. Again there was little evidence that this was due to cell wall hydrostatic pressure caused by transpiration because turgor changed only slightly over the dayhight cycle (Tomos et af., 1992).Thus there must have been some concomitant change in extracellular solute concentrations to match the change in sap osmotic pressure. Leigh and Tomos (1983) suggested that the size of this adjustment must be related to the ability of the different subspecies to accumulate high concentrations of sucrose. Leaves of crassulacean acid metabolism (CAM) plants are another system where large changes in the concentrations of solutes occur in the vacuole. In this case they are the result of diurnal variations in malic acid concentration, which in Kalanchoe dazgrernontiana varies from about 40 mM in the light period to over 200mM at the end of the dark period (Liittge, 1987). This night-time increase in malate concentration is accompanied by an increase in sap osmotic pressure which is accounted for by the change in malate anion concentration as the accompanying change in H+ concentration is osmotically negligible (Smith and Luttge, 1985). However, the turgor in the leaf cells, whether estimated using a pressure bomb or measured directly with a pressure probe, is always lower than the leaf sap osmotic pressure (Steudle et al., 1980; Smith and Luttge, 1985; Murphy and Smith, 1994), although there is a diurnal variation in turgor that has the opposite periodicity to the accumulation of malate, i.e. is higher in the day when malate levels are low (Smith and Liittge, 1985). This periodicity of turgor is due to opening of stomates at night, resulting in an increase in xylem tension which changes the extracellular water potential and offsets the rise in turgor that would otherwise result from the increase in malate concentration (Smith and Liittge, 1985; Murphy and Smith, 1994). Whether there is also a contribution of apoplastic solutes to this regulation of turgor remains to be shown. Hence, CAM plants also moderate changes in turgor in response to increases in vacuolar osmotic pressure resulting from the diurnal cycling of malate into and out of the vacuole. These various examples indicate that individual vacuolar solutes can reach very high concentrations with little evidence of an upper limit. However, in all cases, there is evidence that the rise in sap osmotic pressure is offset by a change in extracellular water potential which limits the increase in turgor. This regulation or moderation of turgor may be the key response, and its evolution may have allowed plants to increase the solute storage capacity of the vacuole without the need to develop mechanically strong cell walls to withstand the large turgors that would otherwise develop. In all of the above examples, the cells have a physiological need to accumulate high concentrations of solutes within their vacuoles either for

186

R. A . LEIGH

storage or to cope with excessive concentrations in the environment. In these conditions, increasing total solute concentrations in the vacuole while regulating turgor through some other means may be the only response possible because the changes in vacuolar composition are large. However, during mobilization of stored solutes from the vacuole, the composition of the vacuole can change with less dramatic shifts in osmotic pressure (and therefore turgor) because loss of one solute from the vacuole is compensated for by the uptake of another. An example is the mobilization of nitrate from vacuoles of lettuce when light intensity is increased (Blom-Zandstra and Lampe, 1985) or other anions are supplied (Blom-Zandstra and Lampe, 1983). The decrease in nitrate is accompanied by concomitant increases in the concentrations of other anions and glucose, which together compensate electrically and osmotically for the decrease in nitrate concentration (BlomZandstra and Lampe, 1985). Mobilization of sugars from vacuoles is also accompanied by compensatory changes in other solutes. Thus, when discs of red beet storage tissue are washed in aerated solutions, a vacuolar acid invertase activity is induced which hydrolyses the sucrose, decreasing its concentration to negligible levels after 3 days (Bacon et al., 1965; Leigh et al., 1979). However, to maintain sap osmotic pressure, only a proportion of the hexoses produced by the invertase activity is mobilized from the vacuole (Perry et al., 1987). With 200 mM mannitol (thought to be similar to the extracellular water potential of red beet storage root tissue; Leigh and Tomos, 1983) in the external medium, the decrease in sucrose and increase in hexose are stoichiometric, leading to maintenance of the initial sap osmotic pressure and turgor. However, with external solutions hypo- or hyperosmotic to this value, the accumulation of hexose does not match the loss of sucrose, so that there is regulation of osmotic pressure and turgor back towards their values in the control (200 mM mannitol) treatment. Regulation is not precise, however, and both parameters stabilize at values significantly different from the control, which may indicate that in this tissue turgor is regulated to fall within a particular range of values rather than to a precise value (Perry el al., 1987). Complete mobilization of the hexoses can be induced by adding dilute mixtures of KCl and NaCl to the external medium. These accumulate in the tissue and the concentration of hexoses decreases. However, this change is accompanied by an increase in sap osmotic pressure and turgor, indicating that the value of these parameters is dependent on the nature of the solutes in the vacuole. A salt-dependent change in turgor also occurs in cereal roots. When grown in “low salt” conditions (i.e. in 0.5 mM CaS04 only) the roots contain high concentration of hexose (Table 111; Pitman et al., 1971), presumably because the seed supply of ions is insufficient to meet fully the osmotic needs of the cells, and so sugars are diverted into this role. When transferred to “high salt” conditions (a solution containing dilute salts), ions are absorbed and

187


TABLE I11 Concentrations of sugurs and salls, eslimuted sap osmotic yre.mire (OP) rind mcasureti turgor in “low-” and “high-” salt cereul roots Growth condition -

Measurementa

-

~

_

_

~ ~~-

- -

0 5 WIMCc1S0,j _

~

_

[K+1 (mM) “a’] (mM) [Sugars] (mM) Calculated OP (mosm kg-’)h Turgor (MPa)

_

- -

-

~

10 mM KCI

~

20

~

88

7

1

77 121 0.3

12 190 0 55

‘Data for solute concentrations are for barley roots and are taken from Pitman et al. (1968, 1971). Turgors are for wheat roots and are from Pritchard ef al. (1987). ’Assumes CI- as the counterion and osmotic coefficient of 1 for all solutes.

the sugar level declines (Table 111; Pitman et ul., 1968, 1971). Measurements of turgor in mature cells of wheat roots grown under similar conditions show that the transition from “low” to “high” salt status is accompanied by an increase in turgor (Table 111; Pritchard er ( i f . , 1987, 1989). All of these examples illustrate that there is no simple model that can describe how the vacuolar concentration of a single solute is regulated. Depending on the solute, species and environmental treatment, the concentrations of individual solutes can vary from very low concentrations to very high concentrations and sap osmotic pressure can be maintained constant o r can vary considerably. However, all of these ideas can be reconciled if it is assumed that the parameter which is maintained relatively constant is turgor and that the ability of cells to achieve this will determine both the upper limit to total solute concentration in the vacuole and the change that occurs during mobilization of stored solutes from thc vacuole.

1V. AMODEL It seems that plant cells can display a wide variety of behaviour with regard to their vacuolar composition. Thus they can ( I ) have either low or high sap osmotic pressure depending on the nature of the solutes in the vacuolc (e.g. low- and high-salt barley roots), (2) change the solute composition of the vacuole iso-osmotically (e.g. loss of nitrate in lettuce leaves) or ( 3 ) increase vacuolar solutes with a parallel increase in osmotic pressure but with more moderate changes in turgor (e.g. NaCl accumulation in barley and Suaedu or sucrose in sugar beet storage roots). So is there any underlying consistency

188

R. A. LEIGH

to these different behaviours or is each system and response unique? Here I wish to suggest a model that can be used to bring all of the above observations together. The elements of this model are:

1. The availability of solutes during the development of a cell determines the concentrations of individual solutes in the vacuole and therefore the sap osmotic pressure and turgor. There is a certain minimum turgor that must be generated by cells, and to achieve this the cell will divert sugars to the vacuole if inorganic salts are not available (as in low-salt barley roots). However, to minimize the amounts of carbon that are put into this role, the turgor that is generated under these conditions will be lower than that developed if salts are available (Table 111). If salts become available the sugars are mobilized as salts are accumulated, and sap osmotic pressure and turgor rise to a value typical of an “inorganic” vacuole. This then sets the maximum “non-stressed” turgor for the cell. If any inorganic ions are mobilized from the vacuole for metabolism (e.g. nitrate in lettuce leaves) these can be replaced by sugars and other carbon-containing compounds such as malate, but this appears to be done iso-osmotically, suggesting that once the turgor “set-point’’ has been determined by salt uptake it is not subsequently decreased to lower values. 2. When a cell is in a situation where it is required to adjust its water potential by accumulating solutes in the vacuole (e.g. NaCl stress) or when the cell is physiologically adapted to store high concentrations of solutes for short or long periods (e.g. CAM plants and sugar beet, respectively) then sap osmotic pressure may increase beyond the value set by the above considerations but the overall size of this rise may be partially offset by loss of other solutes (e.g. of K N 0 3 from vacuoles of barley epidermal cells accumulating NaCl; Fig. 3). To prevent the rise in osmotic pressure translating into large changes in turgor, accumulation of extracellular solutes or other mechanisms are used to decrease extracellular water potential. The extent of these extracellular adjustments is genetically determined and influences both salt tolerance (e.g. compare barley and Suaeda above) and the capacity for solute storage (e.g. sugar beet and red beet). If this suggested behaviour is correct it would imply that adaptation to vacuolar osmotic changes resulting from stresses such as salt tolerance or from the accumulation of high concentrations of storage solutes such as sucrose requires adaptation of water relations at two levels. Firstly, increases in vacuolar osmotic pressure must be matched by an equivalent change in the cytosolic osmotic pressure to maintain water potential equilibrium across the tonoplast. This will be done by the accumulation in the cytosol of compatible solutes such as glycinebetaine (Wyn Jones et al. , 1977). Secondly, turgor will also be adjusted by the mechanisms outlined above. Thus at least


189

three major compartments, the vacuole, cytosol and apoplast, participate in these changes, and without an integrated understanding of the adjustments in all three, the nature of the factors controlling the composition of the vacuole will not be fully understood.

V.

CONCLUSIONS

The ability to analyse solutes in individual vacuoles and to be able to relate their concentrations to changes in the cytosol or to other parameters such as sap osmotic pressure and turgor is providing new insights into how vacuolar composition is regulated. It is now clear that different cell types can behave independently in their solute relations and that conclusions based on averaging techniques will have to be modified to take account of this. The integrated information that is now available from the application of single-cell techniques is showing just how complex tissue solute relations are when broken down to the level of different cell types, and that regulation of turgor rather than solute concentrations per se may be an overriding feature determining the total solute concentrations in a vacuole. Understanding the mechanisms that provide and respond to this regulation is now a major challenge.

REFERENCES Ahmad, N. and Wyn Jones, R. G. (1972). Tissue distribution of glycinebetaine, proline and inorganic ions in barley at different times during the plant growth cycle. Journal of Plant Nutrition 5 , 195-205. Asher, C. J. and Ozanne, P. G. (1967). Growth and potassium content of plants in solution cultures maintained at constant potassium concentrations. Soil Science 103, 155-161. Ammann, D. (1986). “Ion-selective Microelectrodes, Principles, Design and Application”. Springer-Verlag, Berlin. Atkinson, C. J. (1991). The flux and distribution of xylem sap calcium to adaxial and abaxial epidermal tissue in relation to stomata1 behaviour. Journal of Experimental Botany 42, 987-993. Bacon, J . S. D., MacDonald, I. R. and Knight, A. H. (1965). The development of invertase activity in slices of the root of Beta vulgaris L. washed under asceptic conditions. Biochemical Journal 91, 175-182. Barraclough, P. B. and Leigh, R. A. (1993). Grass yield in relation to potassium supply and the concentration of cations in tissue water. Journal of Agricultural Science, Cambridge 121, 157-168. Bell, C. I . and Leigh, R. A. (1996). Differential effects of turgor on sucrose and potassium transport at the tonoplast and plasma membrane of sugar beet storage root tissue. Plant, Cell and Environment 19, 191-100. Bell, C. I., Milford, G. F. J. and Leigh, R. A . (1996). Assimilate partitioning in sugar beet. In “Photoassimilate Distribution in Plants and Crops: Source-Sink Relationships” (E. Zamski and A. A. Schaffer, eds), pp. 691-707. Marcel Dekker. New York.

190

R . A . LEIGH

Bennett, R. N. and Wallsgrove, R. M. (1994). Secondary metabolites in plant defence. New Phytologist 127. 617633. Blom-Zandstra, G. and Lampe, J. E. M. (1983). The effect of chloride and sulphate salts on the nitrate content in lettuce plants (Lactuca sativa L.). Journal of Plant Nutrition 6, 611428. Blom-Zandstra, M. and Lampe, J. E. M. (1985). The role of nitrate in osmoregulation of lettuce (Lactuca sativa L.) grown at different light intensities. Journal of Experimental Botany 36, 1043-1052. Boiler, T. and Wiemken, A. (1986). Dynamics of vacuolar compartmentation. Annual Review of Plant Physiology 37, 137-164. Boursier, P. and LBuchli, A. (1989). Mechanisms of chloride partitioning in the leaves of salt-stressed Sorghum bicolor L. Physiologia Plantarum 77, 537-544. Canny, M. J. (1990a). What becomes of the transpiration stream? New Phytologist 114, 341-368. Canny, M. J . (1990b). Rates of apoplastic diffusion in wheat leaves. New Phytologist 116, 263-268. Clipson, N . J. W., Tomos, A . D., Flowers, T. J . and Wyn Jones, R. G. (1985). Salt tolerance in the halophyte Suaeda maritima L. Dum. The maintenance of turgor pressure and water potential gradients in plants growing at different salinities. Planta 165, 392-396. De Silva, D. L. R., Hetherington, A. M. and Mansfield, T. A. (1996). Where does all the calcium go? Evidence of an important regulatory role for trichomes in two calcicoles. Plant, Cell and Environment 19, 880-886. Dietz, K.-J., Schramm, M., Betz, M . , Busch, H . , Diirr, C. and Martinoia, E. (1992a). Characterization of the epidermis from barley primary leaves. 1. Isolation of epidermal protoplasts. Plunta 187, 425430. Dietz, K.-J., Schramm, M., Lang, B., Lanzl-Schramm, A., Diirr, C. and Martinoia, E. (1992b). Characterization of the epidermis from barley primary leaves. 11. The role of the epidermis in ion compartmentation. Planta 187, 431-437. Dietz, K . J., Hollenbach, B. and Helwege, E. (1994). The epidermis of barley leaves is a dynamic storage compartment of carbohydrates, amino acids and nitrate. Plant Physiology 92, 31-36. Echlin, P. and Taylor, E. (1986). The preparation and X-ray microanalysis of bulk frozen hydrated plant tissue. Journal of Microscopy 161, 329-348. Felle, H. H. (1993). Ion-selective microelectrodes: their use and importance in modern plant cell biology. Botanica Acta 106, 5-12. Felle, H. H. and Bertl, A. (1986). The fabrication of H+-selective liquid-membrane microelectrodes for use in plant cells. Journal of Experimental Botany 37, 1416-1428. Felle, H. H., Tretyn, A. and Wagner, G. (1992). The role of the plasma membrane Ca'+-ATPase in Ca2+ homeostasis in Sinapis alba root hairs. Planta 188, 306-313. Fricke, W., Leigh, R. A. and Tomos, A. D. (1994a). Concentrations of inorganic and organic solutes in sap sampled from individual epidermal, mesophyll and bundle sheath cells of barley leaves. Planta 192, 316316. Fricke, W., Leigh, R. A. and Tomos, A. D. (1994b). Single-cell sap sampling and analysis of the solute contents of individual epidermal cells of barley leaves. Changes along the leaf blade during aging and NaCl stress. Planta 192, 3 17-323. Fricke, W., Pritchard, J., Leigh, R. A. and Tomos, A. D. (1994~).Cells in the upper and lower epidermis of barley (Hordeum vulgare L.) leaves exhibit distinct patterns of vacuolar solutes. Plant Physiology 104, 1201-1208.


191

Fricke, W . , Hinde, P. S., Leigh, R. A . and 'I'omos, A . D . (1995). Vacuolar solutes in the upper epidermis of barley leaves. Intracellular differences follow distinct patterns. Platita 196, 4 ( H 9 . Fricke, W . , Leigh, R . A . and Tonios. A. D. (1996). The intercellular distribution o f vacuolar solutes in the epidermis and mesophyll of barley leaves changes in response t o NaCI. Journal of Experitnetitul Wornny 47, 1413-1426. Granstedt, R . C. and Huffaker, R . C. (1982). Identification of the leaf vacuole as a major nitrate storage pool. Plarrt P h j ~ ~ i o l o g70, . ~ 41(&413. Hodson, M. J . and Sangster. A . G . (1988). Obscrvationson the distribution of mineral elements in the leaf of wheat (Triticum uestivum L.), with particular reference to silicon. Annuls qf' Botany 62, 463471. Homniels. C. H.. Tanczos. 0. G . and Kuiper. P. .I.C . (1989). Responses of internal nitrogen and phosphorus concentrations of two Taruxucum microspecies of contrasting mineral ecology: critical N and P concentrations of whole plants. Physiologia Planiurum 77, 555-561 . Huang, C. X . and van Steveninck. R . F. M . (19x9). Maintenance of low CIconcentrations in mesophyll cells of leaf blades of barley seedlings exposed to salt stress. Plant Physiologv 90, 144W1443. Hiisken. 11.. Zimmermann, U . and Steudle, E. (1978). Pressure probe technique for measuring water relations of cells in higher plants. Plunt Physiology 61, 158- 163. Johannes, E . . Brosnan, J. M. and Sanders, D. (1992). Calcium channels in the vacuolar membrane of plants: multiple pathways lor intracellular calcium mobilization. Philosophical Transoc~ionsof the Royal Society of London, Series B 338, 104-112. Kakie. T. (1969). Phosphorus fractions in tobacco plants a s affected by phosphate application. Soil Science and Plarrr Nutrition (Tokyo) 15, 81-85. Koyro. H.-W. and Stelzer, R . (1988). Ion concentrations in the cytoplasm and vacuoles of rhizoderniis cells from NaCI treated Sorghum. Spartinu, and Puccinelliu plants. Joiirnul of Plunr Physiology 133, 441-446. Kurkdjian, A . and Guern, J. (1989). lntracellular pH: measurement and importance in cell activity. Annuul Review oJ Plant Phvsiolohgy and Plant Molecular Biology 40, 271-303. Kelday, L. S. and Bowling, D. J. F. (1980). Profiles of chloride concentration and PD in the root of Commelinu comrnunis L. Journal of Experimental Botuny 31. 1347-1355. Lazof. D. and Lauchli, A. (1991). Complementary analysis of freeze-dried and frozen-hydrated plant tissue by electron-probe X-ray microanalysis: spectral resolution and analysis of calcium. Plurita 184, 327-333. Lee. K. B. and Ratcliffe, R . G . (1983). Phosphorus nutrition and the intracellular distribution o f inorganic phosphate in pea root tips: a quantitative study using "P-NMR. Journd 0.f Experimental Botritiy 34, 1222-1244. Leigh. R . A. (1983). Methods. progre rnd potential for thc use of isolated vacuoles in studies o f solute transport in higher plant cells. Plrysiologia Pluntarum 57, 390-396. Leigh. R. A . and Johnston, A . E. (1983a). Concentrations of potassium in the dry matter and tissue water of field-grown spring barley and their relationships to grain yield. Journal of Agricidtural Science, Cambridge 101, 675-685. Lcigh. R . A. and Johnston, A . E. (1YX3h). The effects of fertilizers and drought on the concentrations of potassium in the dry matter and tissue water of field-grown spring barley. Journd of Agriculrurai Lsrierzce, Cambridge 101, 741-748. Leigh, R . A . and Storey. R . (1993). Intercellular compartmentation of ions in barley

192

R . A. LEIGH

leaves in relation to potassium supply and salinity. Journal of Experimental Botany 44, 755-762. Leigh, R. A. and Tomos, A. D. (1983). An attempt to use isolated vacuoles to determine the distribution of sodium and potassium in cells of storage roots of red beet (Beta vulgaris L). Planta 159, 469-475. Leigh, R. A. and Tomos, A. D. (1993). Ion distribution in cereal leaves: pathways and mechanisms. Philosophical Transactions of the Royal Society, Series B 341, 75-86. Leigh, R. A. and Wyn Jones, R. G. (1984). A hypothesis relating critical potassium concentrations for growth to the distribution and functions of this ion in the plant cell. New Phytologist 97, 1-13. Leigh, R. A. and Wyn Jones, R. G. (1986). Cellular compartmentation in plant nutrition: the selective cytoplasm and the promiscuous vacuole. In “Advances in Plant Nutrition” (P. B. Tinker and A. Lauchli, eds), Vol. 2, pp. 249-279. Praeger Scientific, New York. Leigh, R. A., ap Rees, T., Fuller, W. A. and Banfield, J. (1979). The location of acid invertase activity and sucrose in the vacuoles of storage roots of beetroot (Beta vulgaris). Biochemical Journal 178, 539-547. Liittge, U. (1987). Carbon dioxide and water demand: Crassulacean acid metabolism (CAM), a versatile ecological adaptation exemplifying the need for integration in ecophysiological work. New Phytologist 106, 593-629. Maathuis, F. J. M. and Sanders, D. (1993). Energization of potassium uptake in Arabidopsis thaliana. Planta 191, 302-307. MacRobbie, E. A. C. and Lettau, J. (1980). Ion content and aperture in “isolated” guard cells of Commelina communis L. Journal of Membrane Biology 53, 199-205. Malone, M., Leigh, R. A. and Tomos, A. D. (1989). Extraction and analysis of sap from individual wheat leaf cells: The effect of sampling speed on the osmotic pressure of extracted sap. Plant Cell and Environment 12, 919-926. Malone, M., Leigh, R. A. and Tomos, A. D. (1991). Concentrations of vacuolar inorganic ions in individual cells of intact wheat leaf epidermis. Journal of Experimental Botany 42, 305-309. Martinoia, E., Heck, U. and Wiemken, A. (1981). Vacuoles as storage compartments for nitrate in barley leaves. Nature 289, 292-294. Matile, Ph. (1978). The biochemistry and function of vacuoles. Annual Review of Plant Physiology 29, 193-213. Matile, Ph. (1987). The sap of plant cells. New Phytologist 105, 1-26. Miller, A. J. (1994). Ion-selective microelectrodes. In “Plant Cell Biology - A Practical Approach” (N. J. Harris and K. Oparka, eds), pp. 283-296. IRL Press, Oxford. Miller, A. J. and Sanders, D. (1987). Depletion of cytosolic free calcium induced by photosynthesis. Nature 326, 397400. Miller, A. J. and Smith, S. J . (1992). The mechanism of nitrate transport across the tonoplast of barley root cells. Planta 187, 554-557. Miller, A. J. and Smith, S. J. (1996). Nitrate transport and compartmentation in cereal root cells. Journal of Experimental Botany 47, 843-854. Mott, R. L. and Steward, F. C. (1972). Solute accumulation in plant cells. I. Reciprocal relations between electrolytes and non-electrolytes. Annals of Botany 36, 621-639. Murphy, R. and Smith, J. A. C. (1994). A critical comparison of the pressure-probe and pressure-chamber techniques for estimating leaf-cell turgor pressure in Kalanchoe daigremontiana. Plant, Cell and Environment 17, 15-29.


193

Penny, M. G. and Bowling, D. J. F. (1974). A study of potassium gradients in the epidermis of intact leaves of Commelinii communis L. in relation to stomatal opening. Planta 119, 17-25. Penny, M. G. and Bowling, D. J. F. (1975). Direct determination of pH in the stomatal complex of Commelina. Planta 122, 209-212. Penny, M. G., Kelday, L. S. and Bowling, D. J. F. (1976). Active chloride transport in the leaf epidermis of Commelina communis in relation to stomatal activity. Planta 130, 291-294. Perry, C. A., Leigh, R. A., Tomos, A. D . , Wyse, R. E. and Hall, J. L. (1987). The regulation of turgor pressure during sucrose mobilisation and salt accumulation by excised storage root tissue of red beet. Planta 170, 353-361. Pitman, M. G., Courtice, A. C. and Lee, B. (1968). Comparison of potassium and sodium uptake by barley roots at high and low salt status. Australian Journal of Biological Sciences 21, 871-881. Pitman, M. G., Mowat, J. and Nair, H. (1971). Interactions between processes for the accumulation of salt and sugar in barley plants. Australian Journal of Biological Sciences 24, 619-631. Pritchard, J., Tomos, A. D. and Wyn Jones, R. G. (1987). Control of wheat root elongation growth. I. Effects of ions on growth rate, wall rheology and cell water relations. Journal of Experimental Botany 38, 948-959. Pritchard, J . , Williams, G., Wyn Jones, R. G. andTomos, A. D. (1989). Radial turgor pressure profiles in growing and mature zones of wheat roots - a modification of the pressure probe. Journal of Experimental Botany 40,567-571. Pritchard, J., Fricke, W. and Tomos, A. D. (1996). Turgor regulation during extension growth and osmotic stress of maize roots. An example of single-cell mapping. Plant and Soil (in press). Ratcliffe, R. G. (1994). In vivo NMR studies of higher plants and algae. Advances in Botanical Research 20, 43-123. Raven, J . A. (1987).The role of vacuoles. New Phytologist 106, 357422. Roberts, J. K. M. (1984). Study of plant metabolism in vivo using NMR spectroscopy. Annual Review of Plant Physiology 35, 375-377. Smith, J. A. C. and Luttge, U. (1985). Day-night changes in leaf water relations associated with the rhythm of crassulacean acid metabolism in Kalanchoe daigremontiana. Planta 163, 272-282. Stelzer, R. and Lehmann, H. (1993). Recent developments in electron microscopical techniques for studying ion localization in plant cells. In “Plant Nutrition - From Genetic Engineering to Field Practice” (N. .I. Barrow, ed.), pp. 3 5 4 5 . Kluwer Academic, Dordrecht. Steudle, E., Smith, J . A. C. and Liittge, U. (2980). Water relation parameters of individual mesophyll cells of the crassulacean acid metabolism plant Kalanchoe daigremontiana. Plant Physiology 66, 1155-1 163. Storey, R., Pitman, M. G., Stelzer, R. and Carter, C. (1983). X-ray microanalysis of cells and cell compartments in Atriplex spongiosa. I. Leaves. Journal of Experimental Botany 34, 778-794. Thomas, R. C. (1978). “Ion-sensitive Microelectrodes. How to Make and Use Them”. Academic Press, London. Tornos, A. D., Leigh, R. A , , Palta, J. A. and Williams, J . H. H . (1992). Sucrose and cell water relations. In “Carbon Partitioning Between and Within Organisms” (C. J. Pollock and J . F. Farrar, eds), pp. 71-89. Biosis Scientific, Oxford. Tomos, D., Hinde, P., Richardson, P., Pritchard, J. and Frickc, W. (1994). Microsampling and measurements of solutes in single plant cells. In “Plant Cell

194

R. A . LEIGH

Biology - A Practical Approach” (N. J . Harris and K. Oparka, eds), pp. 297-314. IRL Press, Oxford. Treeby, M. T. and van Steveninck, R. F. M. (1988). Effects of salinity and phosphatc on ion distribution in lupin leaflets. Physiologia Plantarum 73, 317-322. Treeby, M. T., van Steveninck, R. F. M. and D e Vries, H . M. (1987). Quantitative estimates of phosphorus concentrations within Lupinus luteus leaflets by means of electron probe microanalysis. Plant Physiology 85, 331-334. van Steveninck, R. F. M. and van Steveninck, M. E. (1991). Microanalysis. It1 “Electron Microscopy of Plant Cells” (J. L. Hall and C. R. Hawes, eds), pp. 415-455. Academic Press, London. Watson, D. J . and Selman, I. W. (1938). A comparative physiological study of sugar beet and mangold with respect to growth and sugar accumulation. 11. Changes in sugar content. Annals of Botany 2, 827-846. Walker, D. J . , Smith, S. J . and Miller, A. J. (1995). Simultaneous measurement of intracellular pH, K+ o r NO3- in barley root cells using triple-barrelled ion-selective microelectrodes. Plant Physiolology 108, 743-75 I . Walker, D. J., Leigh, R. A . and Miller, A . J . (1996). Potassium homeostasis in vacuolate plant cells. Proceedings of the National Academy of Sciences of the USA 93, 1051&10 514. Williams, M. L., Thomas, B . J . , Farrar. J . F. and Pollock, C. J . (1993). Visualizing the distribution of elements within barley leaves by energy dispersive X-ray image maps (EDX maps). New Phyfologist 125, 367-372. Wink, M. (1993). The plant vacuole: a multifunctional compartment. Journal of Experimental Botany 44(Supplement), 231-246. Wyn Jones, R. G., Storey, R.. Leigh, R. A . , Ahmad, N. and Pollard, A . (1977). A hypothesis on cytoplasmic osmoregulation. In “Regulation of Cell Membrane Activities in Plants” (E. MarrC and 0. Ciferri, eds). pp. 121-136. ElseviedNorth Holland Biomedical Press, Amsterdam. Zhen, R. G. and Leigh, R. A . (1090). Nitrate accumulation in relation to growth and tissue N concentration. Plant and Soil 124, 157-160. Zhen, R. G., Koyro, H-W., Leigh, R. A , , Tomos, A . D. and Miller, A . J. (1991). Compartmental nitrate concentrations in barley root cells measured with nitrate-selective microelectrodes and by single-cell sampling. Planta 185, 356361.

The Vacuole and Carbohydrate Metabolism

C. J. POLLOCK and A. H. KINGSTON-SMITH

Institute of Grassland and Environmental Research, Plas Gogerddan, Aberystwyth, Dyfed SY23 3EB, UK I. Introduction

............................................................................

11. Methodological Approaches

A. B. C. D. E.

.......................................................

C.

IV.

196

Compartmental Non-aqueous Fr Direct Sampling Preparation of I Analysis of Transport Functions in Isolated Membrane Vesicles ..

111. Sucrose and its Component Hexoses ........................................... A. Are Sucrose and its Component Hexoses Found in Vacuoles?

B.

195

..

198 198 200

Are Sucrose, Glucose and Fructose Accumulated Actively in Vacuoles? ....................................... Are Sucrose-metabolizing Enzymes Loca

Fructans

....

.........................................................................

204

................................................

207

................................................................

208

VII. Conclusions ........................................... Acknowledgements ....................................................... References ............................................

211

V. VI.

Raffinose-series Oligosaccharides Other Carbohydrates

This review is dedicated to the memory of Tom ap Rees, teacher, botanist and friend.

I. INTRODUCTION Sachs (1864) was the first researcher to propose a role for the vacuole in the storage of carbohydrates. By using ethanol to precipitate fructans, he Advances in Botanical Research Vol. 2.5 incorporating Advances in Plant Pathology ISBN 0-12-(l0.592.5-8

Copyrighr 0 1997 Academic Press Limited All rights nf reproducfion in any form reserved

196

C . J. POLLOCK and A. KINGSTON-SMITH

observed the resulting sphaerocrystals in the vacuoles of a range of asteraceous plants. Further consideration of the potential role of the vacuole in carbohydrate metabolism was based upon indirect evidence, or even the assumption that it must be involved because of the high concentrations of soluble sugars found in some plant organs and the very high percentage of cell volume commonly occupied by the vacuole (Pollock, 1984). The ability to prepare isolated vacuoles of adequate purity and intactness has finally permitted a direct examination of these observations and inferences, and the purpose of this review is to summarize the main features of such studies. In addition, however, we wish to address critically the evidence that the vacuole may be the site of carbohydrate accumulation. Plant metabolism is compartmentalized to a high degree, with metabolite and enzyme contents which can differ markedly between compartments. The vacuole is capable of accumulating a wide range of inorganic and organic molecules and of sequestering them against a concentration gradient (Matile, 1987), but there is conflicting evidence over the extent to which such processes are involved in vacuolar metabolism of primary carbohydrates (i.e. those synthesized directly from the products of primary metabolism in both source and sink tissue). We wish, therefore, to consider, for each of the major classes of carbohydrate whose metabolism may involve a vacuolar step, the involvement of the vacuole in their metabolism (i.e. enzymatic synthesis or degradation), their storage (i.e. their presence within intact vacuoles) and their accumulation (sequestration against a concentration gradient). Where appropriate, we will also consider the mechanisms by which these metabolic processes are regulated and the interrelationships with primary metabolism. Sucrose is the pivotal metabolite in terms of soluble carbohydrate metabolism in plants (Pollock and Cairns, 1991; Huber et al., 1992; Pollock and Farrar, 1996). Its synthesis in leaves, the dynamics of its export and its subsequent metabolism in sink tissues controls the rate and distribution of a wide range of catabolic and biosynthetic reaction pathways. Sucrose itself is a vacuolar metabolite (Matile, 1987) and is the biosynthetic precursor of a range of other soluble sugars (Kandler and Hopf, 1984), some of which may be found in vacuoles. We will, therefore consider the interrelationships between sucrose metabolism, in both cytoplasm and vacuole, and the metabolism of other vacuolar carbohydrates.

11.

METHODOLOGICAL APPROACHES

Over the past 20 years, significant advances have been made in the study of metabolic compartmentation in higher plants. In the main, these have been based upon the extraction, purification and analysis of discrete organelles, and there is no doubt that the major impetus for studies on vacuolar carbon metabolism has resulted from the successful preparation of intact, functional

THE VACUOLE A N D CARBOHYDRATE METABOLISM

197

vacuoles. However, other techniques have also been used to estimate metabolite and enzyme distribution between different cellular compartments. A . COMPARTMENTAL ANALYSIS

Developed for the study of ion fluxes and subsequently used by Moorby and Jarman ( 1 975) for the measurement of carbon movement, this technique utilizes the efflux kinetics of soluble, labelled components from tissue samples into an aqueous suspension medium in order to estimate the relative lability of the different metabolic pools within the tissue. This technique has been applied successfully (Farrar and Farrar, 1986) to the study of primary carbon metabolism in leaves. When combined with measurements of flux from intact leaves and analysis of the chemical constitution of the labelled pools (Farrar and Farrar, 1985), these studies have distinguished multiple pools of sucrose and fructans in leaves and have led to the proposal that the less labile of these represents the vacuolar pool. Estimates of residence half-times can be made at various stages throughout the die1 cycle, permitting analysis of the interactions between photosynthesis and primary carbon partitioning (Farrar and Farrar, 1986). The advantage of this approach is that high-level organization within the tissues is not disrupted during the experiments, although the assignation of localization to the distinct kinetic pools is, of necessity, indirect. The technique also requires that uniform labelling of the metabolites can be achieved prior to the start of the efflux measurements. B . NON-AQUEOUS FRACTIONATION A N D STEREOLOGICAL ANALYSIS

This approach has been used by Heldt and co-workers (Winter et al., 1993; Winter et al., 1994) to calculate metabolite concentrations in different cellular compartments. Metabolite contents are measured in whole tissues using non-aqueous fractionation of lyophilized, homogenized material (Gerhardt and Heldt, 1984). This technique determines distribution within different compartments by reference to marker enzyme activities following nonaqueous gradient centrifugation. To convert contents to concentrations. the volumes of the different compartments are calculated following measurement of their relative areas in cross-section, allowing for shrinkage during fixation (Winter el al., 1993). This technique has been applied to leaves of spinach and barley, and has produced concentration estimates for a range of metabolites which are discussed in more detail below (see Section 111). C. DIRECT SAMPLING OF VACUOLAR SAP

Tomos et al. (1992a) have discussed the use of methods to measure soluble carbohydrate contents in samples of vacuolar sap withdrawn directly from

198

C. J. POLLOCK and A. KINGSTON-SMITH

individual cell vacuoles using the pressure probe apparatus (Husken et al., 1978). The assay technique involves prevention of evaporation by storage under water-saturated hexane, accurate subsampling using calibrated picolitre constriction pipettes and the use of fluorimetric measurement of reduced pyridine nucleotides linked to enzyme assays of free and combined hexoses (Tomos et al., 1992b). The methods have been applied with success to epidermal cells, and current studies are extending this approach to mesophyll cells. D. PREPARATION OF ISOLATED VACUOLES

Both mechanically and enzymically prepared vacuoles have been used to study carbohydrate metabolism (e.g. Leigh et al., 1979; Wagner et al., 1983; Darwen and John, 1989; Carpita et al., 1991). The length of incubation in hydrolytic enzymes and the fact that these preparations may have activities against some or all of the carbohydrates which are found naturally in the tissue under test (Winters et al., 1992) makes mechanical isolation the preferred option. Unfortunately, yields are low and hence the method is not suitable for leaves. Estimates of enzyme distribution are less affected by the time taken to prepare the organelles than are measurements of metabolites, and there is no doubt that the data obtained from such isolation studies have led to many of the current hypotheses concerning the compartmentation of carbohydrate metabolism in plants. E. ANALYSIS OF TRANSPORT FUNCTIONS IN ISOLATED MEMBRANE VESICLES

Preparations of isolated vacuoles can be used to prepare purified tonoplast membranes and thence membrane vesicles which are capable of measurable rates of metabolite transport (e.g. Briskin et al., 1985; Greutert and Keller, 1993). Transport processes can be driven by imposed gradients of pH or membrane potential and the uptake of labelled substrates can be measured (Bush, 1993). By varying the methods used to energize the vesicles, information can be obtained on the link between metabolite uptake and the direction of proton movement. Such approaches have been used to study sugar movement into vacuoles, but not all tissues may be used as a source of active vesicles (Bush, 1993).

111. SUCROSE AND ITS COMPONENT HEXOSES The synthesis of sucrose in higher plant cells occurs in the cytoplasm (Huber et al., 1992). Sucrose-phosphate synthase (EC 2.4.1.14; Fig. 1) is believed to be the major enzyme of sucrose synthesis (ap Rees, 1984). Free glucose

THE VACUOLE AND CARBOHYDRATE METABOLISM

199

Synthesis: G 6-P + UTP ?2UDPG + PPi

UDPG pyrophosphorylase

UDPG + F 6-P Q sucrose 6-P + UDP

Sucrose phosphate synthase

sucrose 6-P+sucrose

+ Pi

Sucrose phosphate phosphatase

Breakdown: sucrose + H,O+

glucose + fructose

sucrose + UDP

UDPG + fructose

lnvertase Sucrose synthase

Transformation: sucrose + sucrose-+ 1-kestose [G-F-F]

+ fructose

sucrose + galactinol+ raffinose + myo-inositol

Sucrose:sucrose fructosyl transferase Raffinose synthase

Fig. 1. The principal reactions of sucrose metabolism in higher plants. G 6-P, glucose 6-phosphate; F 6-P, fructose 6-phosphate; UDPG, UDPglucose.

and fructose are thought to be products of sucrose cleavage or polymer hydrolysis, rather than being formed from hexose phosphates (ap Rees et al., 1981). Sucrose degradation can occur via the action of invertases or sucrose synthase, and the enzymology of higher-plant sucrose metabolism is summarized in Fig. 1. Sucrose, glucose and fructose are of widespread occurrence in higher plants and may reach high concentrations in certain tissues. This has led to the general assumption that the vacuole is involved in the storage of sucrose although the direct evidence for this assumption is restricted to a relatively few systems. At least two of these (sugar cane internodes and beet taproots) are derived from improved crop plants and thus may be atypical. Although the agricultural importance of sucrose is derived from its storage function (Pollock et al., 1995), its major physiological role is as a phloemmobile sugar; the principal form in which carbon is transported between organs in higher plants (ap Rees, 1984). Sucrose is known to occur in multiple pools in leaves, representing, it is believed, synthetic, storage and transport pools (Farrar and Farrar, 1985, 1986). In leaves, these pools are all fairly labile, since, under steady state growth conditions, the net rate of sucrose synthesis over a 24 h period roughly matches its rate of export (Farrar and Farrar, 1985). Kinetic analysis of sucrose metabolism in sucrose-storing sinks is less straightforward, since turnover is slower and concentration changes are more gradual than in leaves, but there is some evidence for multiple pools

200

C. J . POLLOCK and A. KINGSTON-SMITH

(Pollock and Farrar, 1996). The critical questions are the extent to which vacuoles are involved in any of these pools, and whether or not there are specific enzymic or transport functions which are localized in these organelles.

A . ARE SUCROSE AND ITS COMPONENT HEXOSES FOUND IN VACUOLES?

There is now good evidence for the occurrence of sucrose in vacuoles. Mechanical isolation and analysis of vacuoles from red beet taproots showed a very strong positive correlation between the presence of the vacuolar marker betanin and the sucrose content of various fractions containing isolated vacuoles (Leigh et al., 1979). These experiments were done quickly enough to minimize potential losses of small molecules, and did not involve the use of crude preparations of hydrolytic enzymes to prepare protoplasts. Vacuolar preparations from enzymically generated protoplasts of a range of species also suggest that sucrose is found in vacuoles, although the proportion of the total sucrose pool may be quite low (Thom et al., 1982; Wagner et al., 1983; Frehner et al., 1984; Keller, 1988). Similarly, there are a number of reports of studies on isolated vacuoles suggesting that glucose and fructose are also found in these organelles (Keller, 1995, and references therein). These reports have been confirmed and extended by the work of Tomos and collaborators, who have used direct sampling of vacuolar sap from intact wheat leaves to demonstrate the preponderance of glucose in epidermal cell vacuoles, contrasting with the bulk tissue abundance of sucrose and fructans (Tomos et al., 1992a). Confirmation of the significance of tissue localization of carbohydrate metabolism in leaves raises a problem in the interpretation of organelle localization data from isolated leaf vacuoles. Although epidermal tissue may be removed prior to leaves being treated with hydrolytic enzymes to prepare protoplasts, the tissue remaining is far from homogeneous. Jellings and Leech (1982) estimated that mesophyll cells make up only 55% of the cell population in cereal leaves, and Williams et al. (1989) have shown that there are differences in the partitioning pattern for chloroplast starch within the population of photosynthetic parenchymatous cells in barley leaves. One is forced to conclude, therefore, that sugar measurements from pooled vacuoles isolated from heterogeneous tissues may be an average from a range of cells with differing metabolite status. We would propose, therefore, that conclusions based on such evidence about the presence or absence of specific accumulation processes should be treated with caution. Such reservations are less significant in the case of tuberous sinks, where tissue uniformity is much higher. Single-cell sampling offers an excellent opportunity to assess the relative importance of tissue level and cell level compartmentation in determining allocation patterns within leaves.

THE VACUOLE AND CARBOHYDRATE METAHOI.ISM

201

B. ARE SUCROSE. GLUCOSE AND FRUCTOSE ACCUMULATED ACTIVELY IN VACUOLES’

The evidence to support active accumulation in vacuolcs of either source or sink organs is not wholly convincing. The crux of the problem is the relative volume of cytoplasm and vacuole. The cytoplasm can comprise as little as 5 % of the total cell volume, so a “passive” equalization of concentrations between the two compartments would still lead to the vacuole being significantly the largest storage pool. Measurements of accumulation based upon differential distribution of solutes between protoplasts and isolated vacuoles (Matile, 1987) will identify such processes where differences are large, but cannot rule out activc accumulation just because differences are small. This problem has led many workers to concentrate on the transport properties of isolated vacuoles or tonoplast vesicles, on the assumption that the presence of an active transport mechanism at the tonoplast indicates that a particular tissue has at least the capacity to accumulate the substrate in viva (Lucas and Madore, 1988; Bush, 1993). Considering specifically uptake into intact vacuoles o r tonoplast vesicles, rather than into intact protoplasts or plasma membrane vesicles, two distinct transport mechanisms have been proposed. Proton-sucrose antiports have been described and characterized for a number of sink tissues including sugar beet tap roots (Briskin et uf., 1985), red beet (Willenbrink and Doll, 1979; Getz, 1987, 1991) and Japanese artichoke (Keller, 1992; Greutert and Keller, 1993). Similar systems for the movement of glucose have been described from pea leaves (Guy et al., 1979), sugar cane (Thom and Komor, 1984) and maize (Rausch et af., 1987). However, the evidence for a proton-sucrose antiport in sugar cane is less clear-cut. Getz et al. (1991) reported small increases in sucrose transport activity into tonoplast vesicles prepared from cane stalks when measured in the presence of ATP or an imposed pH gradient (both consistent with the presence of a sucrose-proton antiport). By contrast, other investigations on suspension cells from the same tissue (Preisser and Komor, 1991; Preisser et af., 1992) have produced evidence for the existence of a facilitated diffusion system which would not lead to elevated sucrose concentrations within the vacuole. A similar process has been suggested for sucrose movement into barley leaf vacuoles (Kaiser and Heber, 1984), with transport of I4C sucrose showing no dependence upon ATP or PP,, high K , values and competitive inhibition by other sugars. A rapid vacuole isolation technique pioneered by these workers (Kaiser et al., 1982) indicated that sucrose was partitioned rapidly into the vacuole with similar kinetics to its rate of synthesis, and led them to propose a facilitated diffusion mechanism. Suggestions that the vacuoles used in these studies might have been damaged and hence lack effective H+ transporting activity were disproved following the demonstration of ATP-dependent malate and CI- transport (Martinoia et al., 1985; Kaiser

202

C. J . POLLOCK and A. KINGSTON-SMITH

TABLE I Calculations of metabolite concentrations in the subcellular compartments of spinach and barley leaves ~~

~

Concentration (mM) in: Species

Metabolite

Stroma

Cytosol

Vacuole

Spinach

F1,6BP UDPglucose Sucrose Glutamine Nitrate

0.55 0.07 1 s) voltage activation times. The question of Ca2+ permeation was addressed directly by Ward and Schoeder (1994), who derived a permeability ratio Pc,:PK = 3:l based on measurement of reversal potentials. The notion of Ca2+ permeation was supported in similar subsequent studies (Allen and Sanders, 1995; Schulz-Lessdorf and Hedrich, 1995), although the issue of anion permeation remains contentious. Thus, disparity between measured values of E,,, and those calculated on the basis of a channel permeable only to Ca2+ and K+ has been ascribed to anion permeation (Schulz-Lessdorf and Hedrich, 1995). Alternatively this disparity can be attributed to Mg2+ permeation, which then obviates the requirement to invoke anion permeation (Allen and Sanders, 1996). The estimates of Pca:PKvary with the specific proportions of the two ions present on either side of the membrane (Allen and Sanders, 1996), and this behaviour is characteristic of multi-ion pores where movement of ions through the channel is non-independent. This then undermines the basic approach for calculation of ionic permeability ratios by measurement of reversal potentials since the assumption of independent ion movement is violated (Hille, 1992). Thus, quantitative considerations of permeation properties are not possible for the SV channel using reversal potential methods. Nevertheless, the clear shifts in Ere, with Eta, taken together with the measurements of SV channel-mediated currents in simple CaC12 solutions, are strongly indicative of Ca2+ permeation (Ward and Schroeder, 1994;

VACUOLAR ION CHANNELS

229

Schulz-Lessdorf and Hedrich, 1995; Allen and Sanders, 1996). Furthermore, SV channel-mediated whole-vacuole and single-channel currents decrease with a rise in free Ca2+ (Ward and Schroeder, 1994; Allen and Sanders, 1995, 1996), which suggests that Ca2+ permeation and selectivity are facilitated through high-affinity binding of the ion within an otherwise non-selective pore, as is the case for animal plasma membrane Ca2+ channels. Several characteristics of SV channels, including widely variable ionic permeability ratios in mixtures of CaZ+ and K+ and the existence of negative apparent permeability ratios calculated for some ionic conditions (Allen and Sanders, 1996), have been predicted by an eight-state model for catalysis of ion permeation in which K + and Ca2+ compete for a common binding site (Gradmann, 1996). Noting the inappropriate permeability ratios calculated when independent electrodiffusion is assumed, this class of model might turn out to provide a better description of Ca2+ and K + permeation through SV channels. However, whatever the mechanism of its permeation, Ca2+ will move via SV channels info the cytosol, driven by the very large electrochemical potential difference for Ca2+, and despite the outward voltage rectification of the SV channel. These Ca2+ activation and Ca2+ permeability properties of SV channels suggest that the channels could participate in Ca2+-induced Ca2+ release from the vacuole. In many respects (Ca2+ permeability, activation by Ca2+ and high pH) SV channels bear remarkable similarities to the YVCCl channel of yeast (Bert1 et al., 1992b). However, the voltage-dependence of the yeast channel favours openings at cytosol-negative potentials, and in this respect the two channel types differ. 3. Pharmacology A wide range of compounds has been reported to inhibit currents through SV channels, including tetraethylammonium (TEA), 9-aminoacridine, quinacrine and quinine (Weiser and Bentrup, 1993). Surprisingly, the acetylcholine receptor antagonist (+)-tubocurarine and the K+ channel blocker charybdotoxin are also effective (Weiser and Bentrup, 1990, 1991, 1993). A number of inhibitors of anion transport, including 4,4’-diisothiocyanatostilbene-2,2’-disulfonicacid (DIDS), 4-acetamido-4‘isothiocyanatostilbene-2,2’-disulfonicacid (SITS) and Zn2+ also inhibit SV channels (Hedrich and Kurkdjian, 1988), and this observation might have reinforced the early notion that the channel is significantly permeable to anions. However, pharmacological profiles are poor guides to selectivity, as indicated, for example, by the general reactivity of DIDS with the &-amino groups of lysine residues. Function The gating and permeability properties of SV channels indicate that they could participate in Ca2+-induced K + and Ca’+ release from the vacuole, 4.

230

G . J . ALLEN and D . SANDERS

or even in K+ uptake into the lumen, depending on the prevailing electrochemical potential difference for K+ . The way in which the gating properties of SV channels, as well as those relating to phosphorylation state (Allen and Sanders, 1995), might be integrated into coordinated responses of vacuolar channels during signalling is discussed in section V of this chapter.

B. FV CHANNELS

A channel that activates instantaneously in response to voltage changes was first reported in beet vacuoles and named the fast vacuolar (FV) channel, to designate its rapid kinetics and to distinguish it from the SV channel (Hedrich and Neher, 1987). Currents characteristic of FV channels have also been recorded from broad bean guard cell vacuoles (Allen and Sanders, 1996) and barley mesophyll vacuoles (Schonknecht et a f . , 1996). 1 . Gating FV channels are voltage-dependent. Currents carried by FV channels are outwardly rectifying: whole-vacuole recordings show larger currents at positive potentials in symmetrical conditions (Allen and Sanders, 1996; see also Fig. 1A), whereas in the physiological range of negative membrane potentials (0 to -40mV) there is a limited conductance. Thereafter, FV channels activate progressively with negative-going voltage. These voltagedependent gating characteristics give whole vacuole FV channel currents a characteristic shape (see Fig. 1A). Intriguingly, FV channels show an opposite Ca2+ dependence to that exhibited by SV channels, with inhibition of activity at [Ca2'], >300 nM in beet and >100nM in guard cells (Hedrich and Neher, 1987; Allen and Sanders, 1996). In addition, vacuolar Ca2+ inhibits FV-mediated currents, especially in the inward direction (J. M. Ward and J. I. Schroeder, personal communication; Schonknecht et al., 1996), and the currents are very sensitive to cytosolic pH, with an optimum at pH 7.3 (G. J. Allen, unpublished observations).

2. Permeation and selectivity Studies of whole vacuole currents indicate that FV channels are cationselective (Allen and Sanders, 1996; Schonknecht et al., 1996), and selectivity for K+ over C1- is confirmed by single-channel studies (Hedrich and Neher, 1987). Single channels have a conductance of 30pS in symmetric 200mM KCl (Hedrich and Neher, 1987). Nothing is known regarding the pharmacology of FV channels.


23 1

3. Functions While a definite demonstration of the function of FV channels is lacking, some proposals have been made on the basis of the known properties. One possibility is that the channel facilitates K+ release from guard cell vacuoles during stomata1 closure, since currents are fivefold greater in the presence of a tenfold cytosol-directed Kf gradient in the presence of 10 nM [Ca*+], (Allen and Sanders, 1996). An alternative possibility is that FV channel function relates to provision of a shunt conductance for the vacuolar electrogenic H+ pumps. This proposal arose from observations on beet vacuoles that instantaneous currents measured in whole-vacuole mode, and possibly FV channelmediated, are activated by both ATP and increased cytosolic pH (Davies and Sanders, 1995). In normal conditions, therefore, this channel could provide a pathway for return current flow, facilitating luminal acidification in the absence of an opposing membrane potential. Conversely, in metabolically restricted conditions (for example, anoxia), with cytosolic pH low and H + ATPase activity decreased through a drop in ATP level, the channel-mediated K+ leak from the vacuole would be reduced. To date, however, the ATP dependence of FV channels has not been studied at the single-channel level.

C. VACUOLAR K' (VK) CHANNELS

VK channels comprise a second, but discrete, type of instantaneously activating K+ channel. They have been described in most detail in vacuoles of broad bean guard cells (Ward and Schroeder, 1994; Allen and Sanders, 1996). 1. Gating One feature which distinguishes VK from FV channels is the voltage independence of VK channels. VK channels also show an opposing dependence on [Ca2+Ic:they are fully activated at S p M Ca2+ (Ward and Schroeder, 1994), although detailed titration studies have demonstrated that activation begins at concentrations above 100 nM (Allen and Sanders, 1996). Finally, VK channels are activated at low cytosolic pH (Ward and Schroeder, 1994): activity reaches a maximum at pH (7.5, and declines progressively over the pH range 7.0-8.0.

2. Perrrieation and selectivity Single-channel studies in symmetric 100 mM KCI have revealed that VK channels have a single-channel conductance of 70 pS (Ward and Schroeder, 1994). They are very highly selective for K f , again in contrast to FV channels. Thus, although VK channels will conduct Rb' and NH4+ to some extent,

232

G. .I.ALLEN and D . SANDERS

neither Cs+, Na+ nor Li+ permeate detectably (Ward and Schroeder, 1994). Nothing is known regarding the pharmacology of VK channels.

3. Function VK channels have so far been described only in guard cells, where they are thought to have a role in facilitating release of vacuolar K+ which is required for stomatal closure. The [Ca2+], dependence of VK channels is in accord with this proposal, since guard cell [Ca2+], normally rises during stomatal closure (Gilroy et al., 1990). However, the observation that cytosolic pH rises slowly in response to the closing stimulus ABA (Irving et al., 1992; Blatt and Armstrong, 1993), coupled with the substantial decrease in VK channel activity above pH 7.5, suggests that other pathways are available for sustained K+ release during closure. One such pathway might be through the SV channel (Ward and Schroeder, 1994). An additional role for VK channels has also been proposed by Ward and Schroeder (1994). This relates to the capacity of VK channels, on opening, to activate voltage-gated ion channels through membrane depolarization towards E K . The proposal is considered in the context of integration of ion channel activities towards the end of this review.

Distribution Although VK channels have not been firmly identified in cell types other than guard cells, there is some evidence for their presence in at least one other cell type where they are likely to co-reside with FV-like channels. In the freshwater alga Eremosphaera viridis, there are two components to the instantaneous current (Linz and Kohler, 1994). The first component is linear with voltage and is carried by a channel with a unitary conductance of 75 pS (in symmetric 100mM KCl). In some respects, this channel resembles the VK channel, although its activity is decreased by a rise in [Ca2'],. The second component is active principally at positive potentials (i.e. strongly outwardly rectifying) and is carried by a channel with a unitary conductance of 35 pS. These features are similar to those of FV channels, as is the decrease of channel activity at low cytosolic pH. It appears that evolutionary prototypes of VK and FV channels might occur in this alga. 4.

D. OTHER INWARD-RECTIFYING K+ CHANNELS

Other channels that are K+-selective and active at negative potentials (i.e. opposite to that of SV channels) have been described in vacuoles from a number of species, including beet (Pantoja et al., 1992c), tobacco (Ping et a f . , 1992b) and Vigna unguicufata (runner bean: Maathuis and Prins, 1991). These channels exhibit a slow time-dependence in the whole-vacuole mode, = 20-5O:l). Although they might be and are markedly K+-selective (PK:PCI similar to FV channels with respect to their voltage dependence, they are


233

distinguished by their dependence on [Ca2'], (Maathuis and Prins, 1991). The pharmacological and functional properties of these channels are not known. E.

HYDROSTATIC AND OSMOTIC PRESSURE (HOP)-ACTIVATED CHANNELS

A class of ion channel which is activated by pressure in the lumen of beet vacuoles, or in the presence of an osmotic gradient, has been described by Alexandre and Lassalles (1991). The channel is inward rectifying and slightly selective for Kf over C1-. Although the function of the channel is not known, it seems possible that it could be geared to regulate vacuolar volume during osmotic stress and hence - more importantly - cytosolic volume. F. VACUOLAR VOLTAGE-GATED Ca'

'

(VVCa) CHANNELS

In addition to SV channels, which are Ca"-permeant and depolarization ativated, Ca2+ channels which are activated on membrane hyperpolarization have been reported in vacuoles of beet (Johannes et al., 1992a; Gelli and Blumwald, 1993) and guard cells (Allen and Sanders, 1994a). 1. Gating The channels activate at negative potentials, and have been studied both at the single-channel (Johannes et a l . , 1992a; Allen and Sanders, 1994a) and whole-vacuole (Gelli and Blumwald, 1993) levels. Currents through VVCa channels show some time dependence in whole-vacuole recordings, but single-channel currents are activated instantaneously. The channels are strongly rectifying: they open only rarely at positive potentials. Typically, activation is over the voltage range (-20 to -50 mV) normally associated with the steady-state vacuolar membrane potential. Activity of VVCa channels is either insensitive to the prevailing [Ca2'], (Johannes etal., 1992a; Allen and Sanders, 1994a), or inhibited when [Ca2'], exceeds values of 1 p M (Gelli and Blumwald, 1993). However, all studies report activation as luminal Ca'+ increasos over the millimolar range: the half-maximal activation constant in beet at 0 niV is 1.4 mM Ca2+ (Johannes and Sanders, 199Sa). The dcpendenco of activation on Ca2+ concentration predicts that binding of two Ca'+ ions is required to open the channel (Johannes and Sanders, 199Sa). The effect of luminal Ca2+ is to shift the threshold for voltage activation to less negative potentials, thereby leading to an increase in open state probability over the physiological range of membrane potentials (Johannes et d.,1992a). Analysis of the voltage dependence of the response leads to the conclusion that the binding sites for Ca" gating are located 30% through the electric field of the membrane from the luminal side (Johannes and Sanders, 199Sa). Activation by luminal Ca2+

234

G. J . ALLEN and D. SANDERS

is curious, because the large size of the vacuolar Ca2+ pool implies that rapid changes in channel gating compatible with fine control of activity cannot be attained through changes in luminal Ca2+.Thus it might be that the activation of the channel by luminal Ca2+ serves a more general purpose, such as restriction on the accumulation of excess Ca2+ in the vacuole. Luminal pH also plays a significant role in the control of channel activity (Johannes et al., 1992b; Allen and Sanders, 1994a). At non-physiological pH around neutrality, channel activity is high - to such an extent that the retention of such activity in vivo would outstrip the ability of vacuolar Ca2+ sequestration mechanisms to sustain a Ca2+ gradient across the vacuolar membrane (Bush, 1993). However, activity progressively decreases with luminal pH, until at physiological values around p H 5.5, channel openings are very infrequent indeed (Allen and Sanders, 1994a). The physiological implications are significant. Modest alkalinization of the vacuole has been observed during stomata1 opening (Bowling and Edwards, 1984), and this might play a role in the short-term control of channel activity in stimulusresponse coupling. Thus, luminal H+ might serve not only to prevent uncontrolled leakage of Ca2+ through VVCa channels in normal conditions but also, unlike luminal Ca2+, to regulate Ca2+ release in short-term responses to environmental stimuli.

2. Permeation and selectivity With 5-20 mM Ca2+ as a charge carrier on the luminal side, single-channel studies have revealed unitary conductances of 6 and 12 pS (beet: Gelli and Blumwald, 1993; Johannes et ad., 1992a) and 14 and 27 pS (guard cells: Allen and Sanders, 1994a). Intriguingly, in beet, spontaneous and reversible transitions between the 12 pS conductance state and one at 4 pS have been reported, and this is accompanied be a marked decrease in open probability (Johannes and Sanders, 1995b). Ionic selectivity for Ca2+ over K + , determined by measurement of reversal potentials in bi-ionic conditions, has been reported as 20:l (beet: Johannes et al., 1992a; Gelli and Blumwald, 1993) and 6:l (guard cells: Allen and Sanders, 1994a). Other alkali earth cations are also transported well by these channels, especially Ba2+, which, as with many animal Ca2+ channels, yields greater currents than those produced by Ca2+ (Gelli and Blumwald, 1993; Allen and Sanders, 1994a; Johannes and Sanders, 1995a). However, detailed permeation studies (Johannes and Sanders, 1995a) have revealed that, like SV channels, VVCa channels behave as multi-ion pores. A bona fide quantitative assessment of permeability ratios is therefore not possible (for the same reasons as given for SV channels). Nevertheless, VVCa channels bear many of the hallmark qualities of Ca2+ channels, including larger currents carried by those ions (K+, Mg2+) which exhibit low apparent permeability ratios. The explanation for this apparent paradox is that ion-binding site(s) within the channel which are required to endow the


235

channel with ionic selectivity are also instrumental in slowing the passage of the ion (Hille, 1992). Those ions binding most strongly are therefore those for which the channel exhibits selectivity in complex solutions, but in simple solutions with only one permeant ion present, the unitary currents appear smaller than for those ions binding less strongly. The corollary is that large unitary currents carried through VVCa channels at high luminal K + concentrations can be titrated with modest additions of Ca2+. The affinity and location of the putative ion-binding sites endowing selectivity can then be investigated by monitoring the Ca2+ concentration and voltage dependence of the inhibited current. Such studies have resulted in the conclusion that the half-maximal inhibition constant at 0 mV is 300 pM Ca2+ (Johannes and Sanders, 1995a). This inhibition constant is only rather weakly voltage dependent, and quantitative analysis suggests a membrane location 9% through the electrical field from the luminal side. Some circumstantial evidence has been obtained for a second ion-binding site nearer to the cytosolic side (Johannes and Sanders, 1995a).

3. Pharmacology VVCa channels are potently inhibited by lanthanides, in particular La'+ (Gelli and Blumwald, 1993) and Gd3+ (Johannes et ul., 1992a; Allen and Sanders, 1994a). Blockade is associated with a drop in open state probability rather than in unitary current. In addition, verapamil and nifedipine - which are normally thought of as plasma membrane Ca2+ channel antagonists are effective in decreasing activity (Gelli and Blumwald, 1993; Allen and Sanders, 1994a). 4. Function VVCa channels are ideally poised to play a role in intracellular Ca2+ mobilization, and hence in intracellular signalling. They catalyse a relatively specific influx of Ca2+ into the cytosol, and respond over a range of membrane potentials believed to pertain in vivo. There are hints of involvement in signal transduction networks involving elevation of vacuolar pH, since this is a critical determinant of activity. The switch from the low-activity, low-conductance state to the high-activity, high-conductance state (Johannes and Sanders, 199%) might also confer essential physiological properties, although until it is known what triggers these state transitions, implications for function must remain speculative. 5 . Deyolarization-activated Ca2+ channels Calcium-selective channels activating on depolarization have been described in vacuoles from beet (Pantoja et a f . , 1992b), Arubidopsis (Ping et al., 1992a) and tobacco (Ping et al., 1992b). These channels are outwardly rectifying and hence could in principle carry Ca2+ from cytosol to vacuole. One proposed

236

G. J . ALLEN and D . SANDERS

function for these channels is that they clear cytosolic Ca2+ loads, for example after vacuolar release of Ca2+ in cell signalling (Pantoja et a f . , 1992b). However, this function is not possible, since the release channels would not be capable of generating an outward Ca2+ gradient from the cytosol to the lumen. Furthermore, as shown in Table I, the electrochemical potential for Ca2+ is strongly directed into the cytosol. It seems very likely that these depolarization-activated Ca2+ channels are actually manifestations of the SV channel, as suggested by Ward and Schroeder (1994). G.

INOSITOL 1,4,5-TRISPHOSPHATE-GATED Ca2+ CHANNELS

Additional pathways exist for mobilization of vacuolar Ca2+. These are gated by ligands thought to have roles in intracellular signalling. The best characterized are those gated by inositol 1,4,S-trisphosphate (Imp3). An increasing body of evidence from microinjection and metabolic studies suggests that InsP3 has a role in controlling a number of processes in plant cells, including stomata1 closure (Gilroy et al., 1991), osmoregulation (Einspahr et af., 1988; Srivastava et al., 1989; Cho et af., 1993), modulation of turgor in the motor cells of leaf pulvini by red light (Kim et al., 1996), and pollen tube growth (Franklin-Tong et al., 1996). According to models developed from work on animal cells, InsP3 is produced by G proteinmediated hydrolysis of the plasma membrane lipid phosphatidylinositol 4,s-bisphosphate on perception of an appropriate agonist (Berridge, 1993). Binding of InsP3 to its receptor then results in Ca2+ mobilization from the endoplasmic reticulum lumen. Both the receptor and Ca2+ release properties are conferred by a homo-tetrameric array on the endoplasmic reticulum. Many of the details of the pathway for InsP3 production in plants remain to be determined, and it cannot be assumed that the animal cell model will apply in detail to plants, although evidence for stimulus-induced inositol lipid turnover is beginning to emerge in plants ( D r ~ b a k 1992). , The presence of an InsP3-gated channel on the vacuolar membrane of plants points to a role for this organelle as an InsP3-mobilizable store of Ca2+,at least in some cases. In this context, it should be noted that other intracellular locations for InsP3-gated channels in plants are not excluded. Indeed, although membrane fractionation studies on carrot suspension cultures failed to detect InsP3sensitive Ca2+ release at any membrane other than the vacuole (Canut et a f . , 1993), recent work on cauliflower florets has yielded a firm indication that InsP3 will also mobilize Ca2+ from a membrane fraction with much greater buoyant density than that of the vacuolar membrane - perhaps the plasma membrane (Muir, 1996).

I . ZnsI', binding and specificity Mobilization of Ca2+ from plant membrane vesicles was first demonstrated by Drobak and Ferguson (1985) using Ca2+-loaded zucchini hypocotyl

VACUOLAR I O N CHANNELS

237

microsomes. Subsequent studies using vacuole-enriched microsomes from the roots of oats or beet (Schumaker and Sze, 1987; Brosnan and Sanders, 1990) or intact vacuoles from Acer cell cultures (Ranjeva et al., 1988) established that the vacuole comprised an InsP3-mobilizable Ca2+ store, and that the dose dependence on InsP3 is similar to that for animal systems, with half-maximal effects at 200-600 nM. Binding sites specific for InsP3 have been solubilized from red beet microsomes, with binding co-purifying with vacuolar membranes (Brosnan and Sanders, 1993). The Kd for InsP3 binding is 120nM, which is in good agreement with the dose dependence found for Ca2+ release once allowance is made for the fact that ATP - which competes with InsP3 for binding - is also present in the Ca2+ release media. Other inositol phosphates (Imp2, InsP4) are ineffective in Ca2+release and in competing for InsP3 binding sites when applied at concentrations < 1 pM (Schumaker and Sze, 1987; Brosnan and Sanders, 1990, 1993). InsP3-binding sites have also been identified in Chenopodium, although specific binding is apparent only at elevated levels of Ca’+ in the millimolar concentration range (Scanlon et al., 1996). The physiological relevance of these sites might therefore be questioned, although it is possible that the membrane isolation protocol results in modification of binding properties. The Imp3-binding site density has been estimated from the equilibrium binding studies on beet to be < 1 pmol mg-’ (Brosnan and Sanders, 1993). Despite this low abundance, Biswas et ul. (1995) were able to use heparin affinity chromatography to purify to apparent homogeneity a protein of M, = 110 000 which could be reconstituted to yield InsP3-gated Ca2+ release activity. The disparity in M , between this protein and the mammalian receptor (250 000 for the monomer) is surprising, given that many of the Ca2+ release and InsP3-binding properties are conserved between animals and plants, and that the InsP3-binding and Ca” channel domains of the mammalian receptor are at the N and C termini, respectively (Taylor and Marshall, 1992). 2. Gating of InsP.3-dependent currents Whole vacuole currents are elicited by InsP3 with a half-saturation constant of 220nM (Alexandre et a / . , 1990), which is within the range observed in vesicle studies. Currents are inwardly rectifying over physiological negative membrane potentials, and this is reflected in the behaviour of Imp3-gated single channels for which open-state probability increases continuously over the range -20 to -90mV. Although several reports have commented on a failure to replicate the observation of InsP3-gated currents in vacuoles (Johannes et al., 1992a; Ping et a / . , 1992a; Gelli and Blumwald, 1993), it appears that, in beet at least, substantial hyperosmotic shock is required prior to vacuole isolation in order that measurable currents are obtained (Allen and Sanders, 1994b). Indeed, the magnitude of the InsP3-induced current

238

G . J. ALLEN and D. SANDERS

increases as an exponential function of the osmotic pressure of the pretreatment solution. It is not known whether this sensitization of vacuoles to InsP3 relates to increased receptor expression, to post-translational events, or even, perhaps, to protection of InsP3 sensitivity during vacuole isolation. Currents elicited by InsP3 are readily reversible on InsP3 washout, as anticipated for a bonafide ligand-gated channel (Alexandre et al., 1990; Allen and Sanders, 1994b). There is no evidence that whole-vacuole currents are sensitive to the prevailing [Ca2+], (Allen and Sanders, 1994b), and, in this respect, gating of plant InsP3 channels appears to differ from that of many animal counterparts, where activation by [Ca”], comprises part of a positive-feedback mechanism for Ca2+-induced Ca2+ release (Taylor and Marshall, 1992). The effects of luminal Ca2+ remain to be investigated. 3. Permeation and selectivity Whether assayed at the single-channel or the whole-vacuole level, InsP3gated currents appear to be very highly selective for Ca2+ over K + , with reported values in excess of 100:l (Alexandre et al., 1990; Allen and Sanders, 1994b). Single-channel conductance has been reported as 30 pS with 5 mM Ca2+ on the luminal side (Alexandre et al., 1990). Subsequent investigations yielded ill-defined and variable estimates of single-channel conductance, largely as a result of extremely rapid gating kinetics (Allen and Sanders, 1994b). 4. Pharmacology Antagonists of InsP3-gated Ca2+ release from vacuoles of plants are similar to those identified in animals. The most potent inhibitor described to date is low molecular mass heparin (M,= 5000), which inhibits Ca2+ release by competition with InsP3 for binding with a Ki = 34 nM in beet (Brosnan and Sanders, 1990, 1993; Johannes et al., 1992b). Heparin with a higher M , is considerably less efficacious (Johannes et al., 1992b). Electrophysiological studies suggest that heparin might also be inhibitory to an unidentified Ca2+-sensitive current in the same membrane, thereby possibly limiting its use as a selective inhibitor of InsP3-gated Ca2+ currents (Alexandre and Lassalles, 1992). A second, less potent inhibitor is 8-(N,N-diethylamino)octyl 3,4,5trimethoxybenzoate (TMB-8), which inhibits in the micromolar range (Schumaker and Sze, 1987; Ranjeva et al., 1988; Johannes et al., 1992a). By contrast, ruthenium red and ryanodine, which interact with mammalian endomembrane Ca2+ channels, are ineffective (Muir et al., 1997).

Function While the very high selectivity and the nature of the activating ligand leave little doubt that InsP3-gated channels function to release Ca2+ during signal

5.


239

transduction, the spectrum of physiological stimuli to which the vacuolar receptor responds has yet to be elucidated. In the pollen tube, for example, where InsP3 might play a crucial role in the control of growth, available evidence indicates a paucity of vacuolar material in the tip and subtip regions which are responsive to InsP3 (Franklin-Tong et al., 1996). Conversely, the prominence of the vacuole in beet storage root, coupled with the upregulation of vacuolar InsP3-gated channels during osmotic stress (Allen and Sanders, 1994b), points to a role for the vacuolar channels in turgor regulation (which is well established in beet: Cram, 1980). The electrophysiological studies on vacuolar InsP3-gated Ca2+ channels give rise to the interesting possibility that these channels could function in stimulus-response coupling without changes in InsP3 levels in one of two ways. First, a general increase in activity might be achieved (through increased expression or post-translational modification of the receptor), as for the response of beet to osmotic stress. This would place InsP3-gated Ca2+ release away from the early signalling events. Second, as originally pointed out by Alexandre et al. (1990), the strong inward rectification of the channel enables large changes in activity to be generated by membrane hyperpolarization alone. In either case, InsP3 would need to be present, but only at constant, basal concentration, to permit activation. H . RYANODINE RECEPTOR HOMOLOGUES

Ryanodine receptors were originally identified in the sarcoplasmic reticulum as the voltage-gated ion channels responsible fbr Ca2+ release during contraction of skeletal muscle. However, in many other animal cell types, coordinate action of ryanodine receptors with InsP3 receptors is thought to give rise to complex patterns of Ca2+ signalling (Berridge, 1993). In such instances, ryanodine receptors are thought to reside in the endoplasmic reticulum. At least one ryanodine receptor isoform (RYR2) is activated - either directly or indirectly -by the NAD metabolite cyclic ADP-ribose (cADPR), and cell types in which cADPR is thought to play a role in Ca2+signalling include non-skeletal muscle, brain and sea urchin eggs (Galione, 1994). The possibility of a similar role for cADPR in plants has yet to be examined critically. Nevertheless, plant vacuoles possess ryanodine receptor-like channels which are activated by cADPR. +

1. Ligand and voltage gating

Release of Ca2+ by cADPR can be demonstrated from vacuolar-enriched membrane vesicles of beet using a radiometric approach, or from intact vacuoles using whole-vacuole patch clamp (Allen et al., 1995). Using either methodology, the cADPR concentrations eliciting a half-maximal response are in the range 2 W O n M (Allen et al., 1995; Muir and Sanders, 1996). The non-cyclic isomer, ADPR, is ineffective at 100 nM.

240

G. J. ALLEN and D. SANDERS

Although the dose-response relationship for cADPR is voltage insensitive, cADPR-elicited whole-vacuole currents are markedly voltage sensitive (Allen et al., 1995). The response to voltage is similar to that of InsP3-gated channels in the same membrane: instantaneous, and with strong inward rectification over physiological membrane potentials. This behaviour is also reflected in the properties of single cADPR-gated channels. Rectification is induced by the presence of Ca2+ on the luminal side of the membrane (G. J. Allen, unpublished data). As in the case of InsP3-gated channels in animal cells, [Ca2+], is a key activator of animal ryanodine receptors, which also participate in Ca2+-induced Ca2+ release. However, there is no evidence for Ca2+ activation of cADPR-elicited currents in beet vacuoles, and at concentrations greater than 1 p M , the response in Vicia guard cell vacuoles is actually inhibited (G. J . Allen, unpublished data). 2. Permeation and selectivity Although electrical activity is induced by cADPR in outside-out membrane patches, channels appear very fast in their gating characteristics, and it is not possible to resolve bona fide single-channel events. The best estimates of selectivity have therefore come from reversal potential measurements of cADPR-elicited whole-vacuole currents, where the permeability ratio Pc,:PK- 15:l has been determined for beet (Allen et al., 1995). This implies that the channel is significantly less selective for Ca2+ than plant InsP3-gated channels. 3. Pharmacology It is with respect to agonists and antagonists that cADPR-gated Ca2+ release at the vacuole exhibits most functional similarity to the ryanodine receptors of animals. Ryanodine itself - a plant alkaloid - can have agonistic or antagonistic effects in animals, depending on concentration. Ryanodine alone, applied to vacuole-enriched microsomes, elicited release of Ca2+ with a half-saturation constant of 40 nM (Muir and Sanders, 1996). Significantly, pretreatment with ryanodine releases just that fraction of the intravesicular Ca2+ pool which is cADPR sensitive, such that subsequent treatment with cADPR is without effect on Ca2+ release. Caffeine, also an agonist of ryanodine receptors in animals, likewise induces Ca2+ release while rendering the vesicles insensitive to subsequent addition of cADPR. Conversely, cADPR-elicited Ca2+ release is blocked by the ryanodine receptor antagonists ruthenium red and procaine (Allen et al., 1995; Muir and Sanders, 1996). 4. Function Calcium channels activated by cADPR co-reside in the same vacuoles as InsP3-gated channels, and yet the two channel types can clearly be distinguished on the basis of pharmacological profile, ionic selectivity and


24 1

the discrete identities of the activating ligand (Allen et al., 1995). While it is anticipated that cADPR-elicited release of Ca2+ from the vacuole will play a role in signal transduction, there is as yet no information on which pathways are involved.

111. ANION CHANNELS By contrast with cation channels, far less is known about the behaviour of anion channels. The principal anionic constituents of higher plant vacuoles are normally malate and/or C1-, and consequently most studies have addressed the issue of permeation by these ions.

A . MALATE (VMAL) CHANNELS

A consistent picture is now beginning to emerge regarding the mechanism of channel-mediated export of malate from the cytosol into the vacuole. Some of these studies have been performed on the CAM plants Graptopetalum paraguayense (Iwasaki et al., 1992) and Kalanchoe daigremontiana (Cheffings et al., 1997), which comprise favourable material because the vacuole undergoes diurnal cycling of vacuolar malate filling.

1 . Gating Malate-permeable channels have been identified from whole-vacuole and single-channel patch clamp studies both in CAM plants (see above) and in sugar beet (Pantoja et al., 1992c) and Arabidopsis thaliana (Cerana et al., 1995). The currents are very strongly inward rectifying, corresponding to anion uptake into the vacuolar lumen over the physiological range of negative membrane potentials. Activation occurs only at potentials negative of the reversal potential for the divalent species (maI2-: Iwasaki et al., 1992; Cerana et al., 1995), suggesting that malate efflux though the channel is prevented even in the event of a favourable membrane potential. Time constants for activation of the current are remarkably slow: in Kalanchoe, for example, Cheffings et al. (1997) report two exponential components with time constants of 0.8 and 5 . 3 s , together with an instantaneous one. The mechanism by which rectification is achieved is unclear at present. One possibility is that luminal C1- blocks malate re-entry to the cytosol (i.e. outward current), and some evidence for this is provided by the observation that luminal C1- inhibits VMAL channels by decreasing the open state probability (Plant et al., 1994). The single-channel conductance is not affected by luminal CI-. Cytosolic Ca2+ and ATP are without effect on VMAL channels (Iwasaki et al., 1992; Cerana et al., 1995; Cheffings et al., 1997).

242


2. Permeation and selectivity Thermodynamically, a membrane potential of -10 to -40 mV is competent in driving cytosolic export of malate into the vacuole (even to the high vacuolar levels occurring in CAM plants) providing the permeant species is maI2- and, in the case of CAM plants, protonation to the monoanionic form (Hmal-) occurs on the luminal side. Selectivity studies have confirmed that the reversal potentials of VMAL-mediated currents are in accord with permeation of maI2- (Pantoja el al., 1992c; Cerana et al., 1995). Singlechannel recordings with 10 mM malate on the cytosolic side yield a unitary conductance of 120 pS in Graptopetalum. Reversal potential measurements in Arabidopsis indicate that a number of other organic divalent anions also permeate with equal efficacy to malate, including succinate and fumarate, while oxaloacetate is somewhat less permeant (Cerana et al., 1995). In sugar beet, permeation of malate channels by NO3-, acetate and even H2P042- has been described (Plant et al., 1994). 3. Function It seems clear that VMAL channels are ideally suited for vacuolar uptake of malate by CAM and other plants, but not for malate mobilization into the cytosol. The identity of the latter pathway remains to be established. In addition, there is the distinct possibility that VMAL channels in some species are responsible for the vacuolar accumulation of a number of inorganic ions, although there is still substantial uncertainty on this point: Cerana et al. (1995) point out that activation by cytosolic malate of an inorganic anion conductance in Arabidopsis might arise either as a result of permeation though VMAL channels, or through separate inorganic ion channels. B. CHLORIDE (VCI) CHANNELS

There is some evidence that, based on their different inhibitor profiles, uptake of malate and of Cl- into vacuoles proceeds by different pathways (Martinoia et al., 1990). However, reports of Cl--permeable channels in higher plant vacuoles have been scant. Ping et al. (1992b) recorded the activity of single C1- channels in tobacco vacuoles. The single-channel conductance is 11OpS in 100mM C1-, and the channel opens at negative potentials, thereby putatively carrying C1- from the cytosol to the vacuole. Klughammer et al. (1992) have detected a number of single-channel conductances for C1-, NO3- and S042- in planar lipid bilayers into which membrane vesicles from barley have been fused. However, one problem with the bilayer approach is that even small amounts of contaminating membranes can contribute to the observations of single channels, so that unless copurification studies are performed with marker enzymes, localization cannot be assured. In addition, there is no possibility with vacuolar membranes to relate channel orientation in the bilayer to that in the native


243

membrane, so the physiological meaning of any observed rectification cannot be ascertained.

IV. SUMMARY OF INDIVIDUAL CHANNEL CHARACTERISTICS The properties of vacuolar ion channels, as described in detail in the preceding account, are summarized in 'Table 11. This table comprises a generalized account, and is not intended to describe in detail those interesting variations in channel behaviour which relate to cell or tissue type and which are noted in the text. Rather, a general catalogue is intended, and this will doubtless require revision as new channel types are discovered and new functions ascertained.

V.

INTEGRATION OF VACUOLAR CHANNEL ACTIVITY

Characterization of the behaviour of individual channel types with respect to ion permeation and gating properties is a prerequisite to appreciation of their respective physiological roles. Nevertheless, a full understanding of channel function can only be attained when the activity of a given channel type is viewed against the background of other transport processes at the same membrane. These other transport processes might influence a number of factors which in the preceding discussion have been shown to impact on the activities of various channels, including membrane voltage, pH,, [Ca"], and, as a result of these potentially localized changes, the activities of phosphatases and kinases. Some intriguing interactions have been proposed in the context of vacuolar Ca2+ mobilization (Ward and Schroeder, 1994), and while these proposals remain largely hypothetical, they do form the basis for further experimental investigation. The proposals centre around the role of the SV channel as a vehicle for Ca2+-induced Ca2+ release (CICR). The insensitivity of the InsP3-gated channel and ryanodine receptor homologues to [Ca2'], in plants has already been noted as a principal point of divergence between ligandgated endomembrane Ca2+ release channels in plant and animal systems. Ward and Schroeder (1994) propose that in guard cells Ca2+ release through these or VVCa channels could trigger activation of Ca2+-permeable SV channels in two ways. First, direct activation of SV channels could arise through elevation of [Ca2+],. Second, activation of [ CaZf],-sensitive VK channels would depolarize the vacuolar membrane towards E K , with the result that SV channels might then enter their range of voltage activation. Essentially, amplification of the primary Ca'+ signal would be achieved even though the channels responsible for initial CaZf release are insensitive to [Ca2'lc. Localized changes in [Ca2'], could play a special role, not only in

TABLE TI Summary of the properties of vacuolar ion channels

Channel

Permeant ion(s)

Gating kineticdvoltagedependence

Unitary conductance

Other regulators

Inhibitors

Function

sv

Ca2+, K +

[Ca2+lCactivated, time-dependent, positive voltages

Calmodulin, Zn2+, DIDS, SITS, 50-250 pS (100 mM KCI) phosphorylation, TEA, PHC 9-aminoacridine, Decreases in quinacrine , increasing Ca2+ quinine, +turbocuraine, charybdotoxin

Fv

K+ (Cl-)

Positive and negative voltages, instantaneous.

30 pS (200 KC1)

PHC

-

K+ release, “shunt” for ATPase

VK

K+

[Ca2+lCactivated, instantaneous, voltage-independent Pressure activated, osmotically activated, voltage independent

70 pS (100 KCI)

PHC

-

K+ release

20pS (200 KCI)

-

-

Volume regulation

HOP

K+ (Cl-)

Ca2+ K+ release flux,

WCa

Ca2+

Negative voltages, instantaneous (single channel), time dependent (whole vacuole)

IP3 gated

Ca2+

Negative voltages, instantaneous

cADPR gated

Ca2+

Negative voltages, instantaneous

VMAL

ma12-

c1-

VCl

PHV [Ca”],

La3+, Gd3+, verapamil, nifedipine

Ca2+ release

-

Low-M, heparin, TMB-8

Ca2+ release

-

[Ca2+Ic (guard cells)

Ruthenium red, ryanodine, procaine

Ca’+ release

Voltages negative of Emal,time dependent

120 p s (10 mM malate)

[Cl-lv

-

Malate into vacuole

Negative voltages, instantaneous.

110pS (100mM KCl)

-

-

C1- into vacuole

pH,, cytosolic pH; pH,, vacuolar pH.

6-27 pS (5-20 mM Ca2+)

30 pS ( 5 mM Ca2+)

246


the activation of SV channels, but also in the inactivation of FV channels (Alexandre and Lassalles, 1991). This latter aspect of control could be important, since FV channels tend to clamp the membrane potential to E K . Down-regulation of FV channel activity in conditions of elevated [Ca2'], could then enable the membrane potential to move from the restricted range imposed by the K + gradient. Positive feedback in Ca2+ signalling at the level of SV channels must inevitably be subject to negative-feedback regulation, since mobilization of anything other than a small proportion of the sizeable vacuolar Ca2' pool into the cytosol would be potentially lethal. The inhibition of SV channel activity by the Ca2+-dependent protein phosphatase calcineurin (Allen and Sanders, 1995) might provide one mechanism for control and downregulation of CICR. However, if this is indeed the case, then the action of phosphatase must be delayed with respect to initial SV channel activation by elevation of [Ca2'],. Such delay could occur if activation of phosphatase were to occur at higher [Ca2'], than the activation of CICR. The contrasting dependence of VK. and FV channels on both pH, and [Ca2'], has already been noted. These diverse responses might give clues to the ways in which Ca2+-dependent and -independent signalling events can converge on the same response (Allen and Sanders, 1996). Thus, stomatal closure could be achieved by vacuolar K+ release either through FV channels in a Ca2+-independent event controlled by an increase in pH, or through VK channels in a Ca2+-dependent event without an increase in pH,. These findings could go some way to reconciling seemingly disparate reports in the literature regarding the centrality or otherwise of cytosolic Ca2+ signalling in stomatal closure (MacRobbie, 1997).

VI. CONCLUSIONS This review has highlighted many areas of ignorance in our understanding of vacuolar channels, especially with respect to function, but even in relation to such basic properties as ion permeation (Gradmann, 1996). However, these uncertainties must be viewed against a background in which just over a decade ago, vacuolar ion channels were essentially uncharacterized in higher plants. Progress during that decade has been astonishing with respect to identification and characterization of the properties of vacuolar ion channels at the electrophysiological level. Despite this, and the progress in our molecular understanding of water channels at the same membrane (see Chrispeels et af., this volume), not one vacuolar ion channel has, to date, been cloned. The challenge for the next decade will be to place the pioneering electrophysiological work with vacuolar ion channels on a firm physiological footing with the range of molecular, optical and biochemical techniques which have yet to be applied to this experimental system.


247

ACKNOWLEDGEMENTS We thank Susan Brudenell for valuable and accurate help with manuscript preparation and the Biotechnology and Biological Sciences Research Council for continuing support to this laboratory. This review is dedicated to the memory of LGDS - scholar and teacher.

REFERENCES Alexandre, J . and Lassalles, J. P. (1991). Hydrostatic and osmotic pressure activated channel in plant vacuole. Biophysical Journal 60. 13261336. Alexandre, J . and Lassalles, J . P. (1992). Intracellular Ca2+ release by InsP3 in plants and effect of buffers on Ca'+ diffusion. Philosophical Transactions of the Royal Society of London, Series B 338, 53-61. Alexandre, J., Lassalles. J. P. and Kado, R. T. (1990). Opening of Ca2+ channels in isolated red beet root vacuole membrane by inositol 1,4,5-trisphosphate. Nature 343, 567-570. Allen, G. J. and Sanders, D. (1994a). Two voltage-gated calcium release channels coreside in the vacuolar membrane of guard cells. Plant Cell 6, 685494. Allen, G. J. and Sanders, D. (1994b). Osmotic stress enhances the competence of Beta vulgaris vacuoles to respond to inositol 1,4,5 trisphosphate. Plant Journal 6, 687-695. Allen, G. J. and Sanders, D. (1995). Calcineurin, a type 2B protein phosphatase, modulates the Ca2+-permeable slow vacuolar ion channel of stomata1 guard cells. Plant Cell 7, 1473-1483. Allen, G. J. and Sanders, D. (1996). Control of ionic currents in guard cell vacuoles by cytosolic and luminal calcium. Plant Journal 10, 1055-1067 Allen, G. J . , Muir, S. R. and Sanders, D. (1995). Release of Ca2+ from individual plant vacuoles by both InsP, and cyclic ADP-ribose. Science 268, 735-737. Amodeo, G., Escobar, A. and Zeiger, E . (1994). A cationic channel in the guard ccll tonoplast of Allium c q a . Plant Physiology 105, 999-1006, Bates, G . W., Goldsmith, M. H. M. and Goldsmith, T. H . (1982). Separation of tonoplast and plasma membrane potential and resistance in cells of oat coleoptiles. Journal of Membrane Biology 66, 15-23. Berridge, M . J . (1993). Inositol trisphosphate and calcium signalling. Nature 361, 197-205. Bertl, A , , Blumwald, E., Coronado, R . , ef al. (1992a). Electrical measurements on endomembranes. Science 258. 873-874. Bertl, A , , Gradmann, D. and Slayman, C. L. (1992b). Calcium- and voltagedependent ion channels in Saccharomyces cerevisiae. Philosophical Transactions of the Royal Society of London. Series B 338, 63-72. Bethke, P. C . and Jones, R . L. (1994). Ca'+-calmodulin modulates ion channel activity in storage protein vacuoles of barley aleurone cells. Plant Cell 6, 277-285. Bethmann. B.. Thaler. M., Simonis, W. and Schiinknecht, G. (1995). Electrochemical potential gradients of H + , K', Ca2+, and C1- across the tonoplast of the green alga Eremosphaera viridis. Plant Physiology 109, 1317-1326. Bieleski, R . L. (1968). Effect of phosphorus deficiency on levels of phosphorus compounds in Spirodelu. Planr Physiology 43, 1309-1316.

248


Biswas, S., Dalal, B., Sen, M. and Biswas, B. B. (1995). Receptor for myo-inositol trisphosphate from the microsomal fraction of Vigna radiata. Biochemical Journal 306, 631-636. Blatt, M. R. and Armstrong, F. (1993). K+ channels of stomatal guard cells: abscisic-acid-evoked control of the outward rectifier is mediated by cytoplasmic pH. Planta 191, 330-341. Bowling, D. J. F. and Edwards, A. (1984). pH gradients in the stomatal complex of Tradescantia virginiana. Journal of Experimental Botany 35, 1641-1645. Brosnan, J. M. and Sanders, D. (1990). Inositol trisphosphate-mediated calcium release in beet microsomes is inhibited by heparin. FEBS Letters 260, 70-72. Brosnan, J. M. and Sanders, D. (1993). Identification and characterization of high-affinity binding sites for inositol trisphosphate in red beet. Plant Cell 5 , 931-940. Bush, D. S. (1993). Regulation of cytosolic calcium in plants. Plant Physiology 103, 7-13. Bush, D. S . (1995). Calcium regulation in plant cells and its role in signalling. Annual Review of Plant Physiology and Plant Molecular Biology 46, 95-122. Canut, H., Carrasco, A., Rossignol, M. and Ranjeva, R. (1993). Is vacuole the richest store of IP3-mobilizable calcium in plant cells? Plant Science 90, 135-143. Cerana, R., Giromini, L. and Colombo, R. (1995). Malate-regulated channels permeable to anions in vacuoles of Arabidopsis thaliana. Australian Journal of Plant Physiology 22, 115-121. Cheffings, C. M., Pantoja, O., Ashcroft, F. M. and Smith, J. A. C. (1997). Malate transport and vacuolar ion channels in CAM plants. Journal of Experimental Botany (in press). Cho, M. H., Shears, S. B. and Boss, W. F. (1993). Changes in phosphatidylinositol metabolism in response to hyperosmotic stress in Daucus carota L. cells grown in suspension culture. Plant Physiology 103, 637-647. Colombo, R., Cerana, R., Lado, P. and Peres, A. (1988). Voltage-dependent channels permeable to K+ and Na+ in the membrane of Acer pseudoplatanus vacuoles. Journal of Membrane Biology 103, 227-236. Colombo, R., Cerana, R., Camattari, G. and Giromini, L. (1996). Effects of divalent cations and heavy metals on ion channels of Arabidopsis thaliana. Journal of Experimental Botany 47(Supplement 67). Coyaud, L., Kurkdjian, A., Kado, R. and Hedrich, R. (1987). Ion channels and ATP-driven pumps involved in ion transport across the tonoplast of sugarbeet vacuoles. Biochimica et Biophysica Acta 902, 263-268. Cram, W. J. (1976). Negative feedback regulation of transport in cells. The maintenance of turgor, volume and nutrient supply. In “Encyclopedia of Plant Physiology, New Series” (U. Liittge and M. G. Pitman, eds), Vol. 2A, pp. 284-316. Springer-Verlag, Berlin. Cram, W. J. (1980). Chloride accumulation as a homeostatic system: negative feedback signals for concentration and turgor maintenance differ in a glycophyte and a halophyte. Australian Journal of Plant Physiology 7, 237-250. Cram, W. J. (1983a). Characteristics of sulfate transport across plasmalemma and tonoplast of carrot root cells. Plant Physiology 72, 204-211. Cram, W. J. (1983b). Sulphate accumulation is regulated at the tonoplast. Plant Science Letters 31, 329-338. Davies, J. M. and Sanders, D. (1995). ATP, pH and Mg2+ modulate a cation current in Beta vulgaris vacuoles: a possible shunt conductance for the vacuolar H+-ATPase. Journal of Membrane Biology 145, 75-86.


249

Drobak, B. K. (1992). The plant phosphoinositide system. Biochemical Journal 288, 697-712. Drobak, B. K. and Ferguson, I. B. (1985). Release of Ca2+ from plant hypocotyl microsomes by inositol 1,4,5-trisphosphate. Biochemical and Biophysical Research Communications 130, 1241-1246. Einspahr, K. J., Peeler, T. C. and Thompson, G. A. Jnr (1988). Rapid changes in polyphosphoinositide metabolism associated with the response of Dunaliella salina to hypoosmotic shock. Journal of Biological Chemistry 263, 57755779. Felle, H. (1988). Cytoplasmic free calcium in Riccia fluitans and Zea mays: interactions of Ca2+ and pH? Planta 176, 248-255. Felle, H. (1993). Ion selective microelectrodes - their use and importance in modern plant-cell biology. Botonica Acta 106, 5-12. Findlay, G. P. and Hope, A. B. (1964). Ionic relations of cells of Chara australis. VII. The separate electrical characteristics of the plasmalemma and tonoplast. Australian Journal of Biological Science 17, 62-83. Flowers, T. J., Troke, P. F. and Yeo, A. R. (1977) The mechanism of salt tolerance in halophytes. Annual Review of Plant Physiology 28, 89-121. Franklin-Tong, V. E., Drobak, B. K., Allan, A. C., Watkins, P. A. C. andTrewavas, A. J. (1996). Growth of pollen tubes of Papaver rhoeas is regulated by a slow-moving calcium wave propogated by inositol 1,4,5-trisphosphate. Plant Cell 8, 1305-1321. Galione, A. (1994). Cyclic ADP-ribose, the ADP-ribosyl cyclase pathway and calcium signalling. Molecular and Cellular Endocrinology 98, 125-131. Gambale, F., Cantu, A. M., Carpaneto, A. and Keller, B. U. (1993). Fast and slow activation of voltage-dependent ion channels in radish vacuoles. Biophysical Journal 65, 1837-1843. Gelli, A. and Blumwald, E. (1993). Calcium retrieval from vacuolar pools. Plant Physiology 102, 1139-1146. Gilroy, S., Fricker, M. D., Read, N. D. and Trewavas, A. J. (1991). Role of calcium in signal transduction of Commelina guard cells. Plant Cell 3, 333-344. Gilroy, S . , Read, N. D. and Trewavas, A. J. (1990). Elevation of cytoplasmic calcium by caged calcium or caged inositol trisphosphate initiates stomata1 closure. Nature 346, 769-771. Gradmann, D. (1996). Selectivity of ion channels: competitive catalysis versus independent electrodiffusion. Journal of Experimental Botany 47, 1733-1736. Hamill, 0. P., Marty, A., Neher, E., Sakmann, B. and Sigworth, F. J. (1981). Improved patch clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pfliigers Archiv 391, 85-100. Hedrich, R. and Neher, E. (1987). Cytoplasmic calcium regulates voltage dependent ion channels in plant vacuoles. Nature 329, 833-836. Hedrich, R. and Kurkdjian, A. (1988). Characterization of an anion-permeable channel from sugar beet vacuoles: effect of inhibitors. E M B O Journal 7 , 3661-3666. Hedrich, R., Fliigge, U. I. and Fernandez, J . M. (1986). Patch-clamp studies of ion transport in isolated plant vacuoles. FEBS Letters 204, 228-232. Hedrich, R., Barbier-Brygoo, H., Felle, H., Fliigge, U. I., Liittge, U., Maathuis, F. J. M., Marx, S., Prins, H. B. A., Raschke, K., Schnabl, H., Schroeder, J. I., Struve, I., Taiz, L. and Ziegler, P. (1988). General mechanisms for solute transport across the tonoplast of plant vacuoles: a patch-clamp survey of ion channels and proton pumps. Botanica Acta 101, 7-13. Hille, B. (1992). “Ionic Channels of Excitable Membranes”, 2nd edn, pp. 374-385.

250

G. J. ALLEN and D. SANDERS

Sinauer, Sunderland, MA. Irving, H. R., Gehring. C. A. and Parish, R. W. (1992). Changes in cytosolic pH and calcium of guard cells precede stomatal movements. Proceedings of the National Academy of Sciences of the U S A 89, 1790-1794. Iwasaki, I . , Arata, H., Kijima, H. and Nishimura, M. (1992). Two types of channels involved in the malate ion transport across the tonoplast of a crassulacean acid metabolism plant. Plant Physiology 98, 1494-1497. Johannes, E. and Sanders, D. (1995a). Lumenal calcium modulates unitary conductance and gating of an endomembrane calcium release channel. Journal of Membrane Biology 146, 21 1-224. Johannes, E. and Sanders, D. (1995b). The voltage-gated Ca2+ release channel in the vacuolar membrane of sugar beet resides in two activity states. FEBS Letters 365, 1 4 . Johannes, E., Brosnan, J. M. and Sanders, D. (1992a). Parallel pathways for intracellular Ca2+ release from the vacuole of higher plants. Plant Journal 2, 97- 102. Johannes, E., Brosnan, J. M. and Sanders, D. (1992b). Calcium channels in the vacuolar membrane of plants: multiple pathways for intracellular Ca2+ mobilization. Philosophical Transactions of the Royal Society of London, Series B 338, 105-1 12. Kim, H. Y . , Cote, G. G. and Crain, R. C. (1996). Inositol 1,4,5-trisphosphate may mediate closure of K+ channels by light and darkness in Sarnanea sarnan motor cells. Planta 198, 279-287. Klughammer, B., Benz, R., Betz, M., Thume, M. and Dietz, K. J. (1992). Reconstitution of vacuolar ion channels into planar lipid bilayers. Biochirnica et Biophysicu Acta 1104, 308-316. Lado, P., Colombo, R. and Cerana, R. (1989). K+ channels in the tonoplast of Acer pseudoplatanus cells. I n “Plant Membrane Transport: The Current Position” (J. Dainty, M. I . DeMichelis, E. Marre and F. Rasi-Caldogno, eds), pp. 179-184. Elsevier, Amsterdam. Leigh, R. A. and Branton, D. (1976). Isolation of vacuoles from root storage tissue of Beta vulgaris L. Plant Physiology 58, 45-62. Linz, K. W. and Kohler, K. (1994). Vacuolar ion currents in the primitive green alga Erernosphaeru viridis: the electrical properties are suggestive of both the Characeae and higher plants. Protoplasma 179, 3445. Maathuis, F. J. M. and Prins, H. B. A. (1990). Patch clamp studies on root cell vacuoles of a salt-tolerant and a salt-sensitive Planlago species. Plant Physiology 92, 23-28. Maathuis, F. J. M. and Prins, H. B. A. (1991). Outward current conducting ion channels in tonoplasts of Vigna unguiculata. Journal of Plant Physiology 139, 6349. MacRobbie, E. A. C. (1995). ABA-induced ion efflux in stomatal guard cells: multiple actions of ABA inside and outside the cell. Plant Journal 7, 565-576. MacRobbie, E. A. C. (1997). Signalling in guard cells and regulation of ion channel activity. Journal of Experimental Botany (in press). MacRobbie, E. A. C. and Lettau, J. (1980). Ion content and aperture in “isolated” guard cells of Commelina communis L. Journal of Membrane Biology 53, 199-205. Martinoia, E., Vogt, E. and Amrhein, N. (1990). Transport of malate and chloride into barley mesophyll vacuoles. ljifferent carriers are involved. FEBS Letters 261, 109-111.


25 1

Miller, A . J. and Sanders, D. (1987). Depletion of cytosolic free calcium induced by photosynthesis. Nature 326, 397400. Muir, S . R . (1996). Ligand-gated calcium channels in higher plant membranes. D.Phil. Thesis, University of York. Muir, S . R. and Sanders, D. (1996). Pharmacology of Ca2+ release from red beet microsomes suggests the presence of ryanodine receptor homologs in higher plants. FEBS Letters 395, 3 9 4 2 . Muir, S. R . , Bewell, M. A., Sanders, D. and Allen, G . J . (1997). Ligand-gated Ca2+ channels and Ca2+ signalling in higher plants. Journal of Experimental Botany (in press). Pantoja, 0.. Dainty, J . and Blumwald, E. (1989). Ion channels in vacuoles from halophytes and glycophytes. FEBS Letters 255, 92-96. Pantoja, O., Dainty, J. and Blumwald, E. (1992a). Cytoplasmic chloride regulates cation channels in the vacuolar membrane of plant cells. Journal of Membrane Biology 125, 219-229. Pantoja, O., Gelli, A. and Blumwald, E. (1992b). Voltage-dependent calcium channels in plant vacuoles. Science 255, 1567-1570. . of vacuolar nialate Pantoia, O., Gelli, A. and Blumwald. E. ( 1 9 9 2 ~ )Characterization and Kf channels under physiological conditions. Plant Physiology 100. 1137-1 141. Ping, Z., Yabe, I . and Muto, S. (1992a). Voltage-dependent Ca2+ channels in the plasma membrane and the vacuolar membrane of Arubidopsis thafiuna. Biochimica et Riophysica Acru 1112, 287-290. Ping, Z., Yabe, I . and Muto, S . (1992b). Identification of K f , CI-, and Ca2+ channels in the vacuolar membrane of tobacco cell suspension cultures. Protoplasmu 171, 7-18. Plant, P. J., Gelli, A . and Blumwald, E. (1994). Vacuolar chloride regulation of an anion-selective tonoplast channel. Journal of Membrane Biology 140, 1-12. Ranjeva, R . , Carrasco, A. and Boudet. A . M. (1988). lnositol trisphosphate stimulates the release of calcium from intact vacuoles from Acer cells. FEBS Letters 230, 137-141. Raschke, K . and Hedrich, R . (1989). Patch clamp measurements on isolated guard cell protoplasts and vacuoles. Methods in Enzymology 174, 3 12-330. Reid, N. D., Shacklock, P. S . , Knight, M . R . and Trewavas, A. J. (1993). Imaging calcium dynamics in living plant cells and tissues. Cell Biology International 17, 111-125. Reifarth, F. W . , Weiser, T. and Bentrup, F. W. (1994). Voltage- and Ca*+dependence of the K + channel in vacuolar membrane of Chenopodium rubrum. L. suspension cells. Biochimica et Biophysica Acta 1192, 79-87. Sanders, D., Muir, S. R. and Allen, G . J . (1995). Ligand- and voltage-gated calcium release channels at the vacuolar membrane. Biochemical Society Transactions 23, 85&861. Scanlon, C. H . , Martinec, J . , Machackova, I.. Rolph, C. E. and Lumsden. P. J . (1996). Identification and preliminary characterisation of a calcium-dependent high-affinity binding site for inositol trisphosphate from Chenopodium rubrum. Plant Physiology 110, 867-874. Schonknecht, G., Pottosin, I . and Tikhonova, L. (1996). Charactcrization of fast vacuolar currents. Journal of Experimental Botany 47( Supplement 69). Schulz-Lessdorf, B. and Hedrich, R. (1995). Protons and calcium modulate SV-type channels in the vacuolar-lysosomal compartment - channel interaction with calmodulin inhibitors. PIunta 197, 6 5 5 4 7 1 . Schumaker, K. S. and Sze. H . (1987). Inositol 1,4.5-trisphosphate releases Ca2+ from

252


vacuolar membrane vesicles of oat roots. Journal of Biological Chemistry 262, 3944-3946. Smith, J. A. C. and Bryce, J. H. (1992). Metabolite compartmentation and transport in CAM plants. In “Plant Organelles” (A. K. Tobin, ed.), pp. 141-167. Cambridge University Press, Cambridge. Spanswick, R. M. and Williams, S. E. (1964). Electrical potentials and Na, K, and CI concentrations in the vacuole and cytoplasm of Nitella translucens. Journal of Experimental Botany 15, 193-200. Srivastava, A., Pines, M. and Jacoby, B. (1989). Enhanced potassium uptake and phosphatidylinositol-phosphate turnover by hypertonic mannitol shock. Physiologia Plantarum 11, 320-325. Taylor, C. W. and Marshall, I. C. B. (1992). Calcium and inositol 1,4,5-trisphosphate receptors: a complex relationship. Trends in Biochemical Science 11, 403-407. Trebacz, K., Simonis, W. and Schonknecht, G. (1994). Cytoplasmic Ca2+, K’, CI-, and NO3- activities in liverwort Conocephalum conicum L. at rest and during action potentials. Plant Physiology 106, 1073-1084. Walker, D. J., Smith, S. J. and Miller, A. J. (1995). Simultaneous measurement of intracellular pH and K+ or NO3- in barley root cells using triple-barreled ion-selective microelectrodes. Plant Physiology 108, 743-751. Walker, D. J., Leigh, R. A. and Miller, A. J . (1996). Potassium homeostasis in vacuolate plant cells. Proceedings of the National Academy of Sciences of the USA 93, 10510-10514. Ward, J. M. and Schroeder, J. I. (1994). Calcium-activated K+ channels and calcium-induced calcium release by slow vacuolar channels in guard cell vacuoles implicated in the control of stomata1 closure. Plant Cell 6, 669-4583. Ward, J. M., Pei, Z.-M. and Schroeder, J. I. (1995). Roles of ion channels in initiation of signal transduction in higher plants. Plant Cell I, 833-844. Weiser, T. and Bentrup, F. W. (1990). (+)-Tubocurarine is a potent inhibitor of cation channels in the vacuolar membrane of Chenopodium rubrum L. FEBS Letters 211, 220-222. Weiser, T. and Bentrup, F. W. (1991). Charybdotoxin blocks cation-channels in the vacuolar membrane of suspension cells of Chenopodium rubrum L. Biochimica et Biophysica Acta 1066, 109-110. Weiser, T. and Bentrup, F. W. (1993). Pharmacology of the SV channel in the vacuolar membrane of Chenopodium rubrum suspension cells. Journal of Membrane Biology 136, 43-54. Weiser, T., Blum, W. and Bentrup, F. W. (1991). Calmodulin regulates the Ca2+-dependent slow-vacuolar ion channel in the tonoplast of Chenopodium rubrum suspension cells. Planta 185, 440-442. Zhen, R.-G. and Leigh, R. A. (1990). Nitrate accumulation by wheat (Triticum aestivum) in relation to growth and tissue N concentrations. Plant and Soil 124, 157-163. Zhen, R.-G., Koyro, H.-W., Leigh, R. A., Tomos, A. D. and Miller, A. J. (1991). Compartmental nitrate concentrations in barley root cells measured with nitrate-selective microelectrodes and by single-cell sap sampling. Planta 185, 356-361. Zhen, R.-G., Smith, S. J. and Miller, A. J. (1992). A comparison of nitrate-selective microelectrodes made with different sensors and the measurement of intracelMar nitrate activities in cells of excised barley roots. Journal of Experimental Botany 43, 131-138.

The Physiology, Biochemistry and Molecular Biology of the Plant Vacuolar ATPase

U. LUITGE and R. RATAJCZAK

Technische Hochschule Darrnstadt, Institut fur Botanik, Schnittspahnstrasse 3-5, 0-64287 Darrnstadt, Germany

I.

Introduction

. . . . . . . . . . . . . . . .. .. . . . . . . . . . . . . . . . .. .. .. . . . . .

11.

Phylogeny

. ... .. . .. ... ...... .... . .....

111.

Ontogeny

.. .. .. .. . . .

IV.

Properties

V.

Holoenzyme Subunit

VI.

Electron Microscopy

VII.

VIII.

IX.

-

253 255

.........................................

257

............................................................

262

Fine Structure

. ._.. _..._...... .. ..... .... .. .. .. .. ... . . 267

. ... .. .. .. . . .. . . . ... . . .. ..... . .. ... .. .... .. . . . .. . . . .. .... . . . . . . 270

Physiological Functions and Ecophysiological Responses Cell Physiological Regulation

.. .. ... . ..... .. .. .. ..

.. ... .. .. . . . ... 276 28 1

284 Conclusions and Outlook . . .. ...... .. . .. .. Acknowledgements ........................................................ ..... 285 285 .................... References . .. ... .. .. . .. . .. . ..... ... .. .. . .. .

I.

INTRODUCTION

Primary active transport in plant cells, i.e. the transmembrane movement of solute particles by direct consumption of energy available from dissociation of covalent phosphoric acid-ester bonds, is limited to a few simple cations, i.e. H + and Ca2+ and possibly Na+ and C1-, with the H+- and Ca2+transporting ATPases at the plasma membrane, endomembranes and the Advances in Botanical Research Vol 25 incorporating Advanccs in Plant Pathology ISBN 0-12-lx).5Y25-X

Copyright 0 1YY7 Academic Press Limited All riphts of reproduction In any form reserved

254

U . LUTTGE and R. RATAJCZAK

tonoplast, the H+-transporting pyrophosphatase (V-PPase) at the tonoplast (see Zhen er al. and Davies, this volume), and putative Na+- and CI--transporting ATPases at the plasma membrane of halophytes and marine algae (for references to some of this general knowledge see, for example, Luttge (1993); Ca2+-ATPase at the tonoplast - Diaz de Leon and Wyn Jones (1985), Evans et al. (1991), Pfeiffer and Hager (1993); Na+-ATPase at the plasma membrane - Shono et af. (1995)). Although this simple picture may soon be more or less severely complicated by the discovery of ABC* transporters at the tonoplast, which use ATP hydrolysis energy directly to transport organic solutes (Martinoia et al., 1993; Li ef af., 1995; Salt and Rauser, 1995; Martinoia and Ratajczak, this volume), it sets the scene for discussion of the vacuolar H+-pumping ATPase, the V-ATPase. H+-ATPases in general are located in different membrane systems. They are important for cellular metabolism in different ways. From the functional point of view ATPases can be divided into two groups, i.e. enzymes generating ATP by using the energy of a proton gradient across a membrane (the (M)F-type H+-ATPases of the inner mitochondria1 membrane (Futai et al., 1989), the CF-type H+-ATPase of the thylakoid membrane (Graber et al., 1990), and the (B)F-type Hf-ATPase of eubacteria (Futai et af., 1989)) and enzymes generating a proton gradient across a membrane by using the energy of ATP hydrolysis (the E1E2- or P-type H+-ATPase of the plasma membrane (Serrano, 1990) and the V-type H+-ATPase of the tonoplast). While P-ATPases are homodimers of two 100 kDa polypeptides, both the F-ATPases and V-ATPases are protein complexes consisting of a variety of subunits (see Section V). Moreover, when visualized by electron miroscopy both the F-ATPase and the V-ATPase exhibit a characteristic ‘head-andstalk’ structure (see Section VI). This membrane peripheral ‘head-and-stalk’ domain is denominated F1 or V1 for the F-ATPase and the V-ATPase, respectively, while the membrane integral domains of the enzyme are called F, and, in analogy, V,. Thus, structurally the ATP-hydrolysing V-ATPase is more related to the F-ATPase, which is synthesizing ATP under physiological conditions, rather than to the ATP-hydrolysing P-ATPase. While operating in concerted action and within interconnected regulation networks together with the other ion pumps, the V-ATPase in itself is a key element in cell physiology. Since the advent of tonoplast membrane purification about 20 years ago (Buser-Suter et af.,1982; Marin, 1985) much work has been performed on this enzyme, and it has been reviewed repeatedly (Marin, 1985; Sze, 1985; Sze et af.,1992a,b). Nevertheless, it still provides us with many fascinating questions regarding its phylogeny, ontogeny and molecular fine structure, as reviewed here. This is particularly due to the very complex multisubunit structure of the V-ATPase holoenzyme.

*ATP-binding cassette transport protein.

PLANT VACUOLAR ATPase

255

Moreover, variable cytological localization, flexible and versatile changes of both its activity and molecular fine structure in response to environmental factors and cell physiological control parameters have been revealed recently. This opens the outlook on a new field, where the V-ATPase with its plastic responses is considered as a steering element in whole-cell and whole-plant behaviour .

11.

PHYLOGENY

Due to the structural similarities of F-ATPases and V-ATPases at the level of the multisubunit holoenzymes and to the ubiquity and antiquity of these types of ATPases attention has focused on the evolutionary relationships among both types of ATPases. The knowledge of DNA or protein sequences of F- and V-ATPase subunits allowed sequence alignment and determination of the degree of sequence homology and, thus, opened the possibility to construct phylogenetic trees, which has been intensively reviewed in recent years (Nelson and Taiz, 1989; Gogarten ef nl., 1992; Gogarten and Taiz, 1992; Kibak et al., 1992; Nelson, 1992). Most sequence data are available for three of the V-ATPase subunits, i.e. the catalytic A subunit and the regulatory B-subunit of V , and the proteolipid subunit c of the proton conducting V,, domain (Table I; for V-ATPase subunits, see also Sections IV and V). Thus, phylogenetic investigations were performed by comparison of the sequences of subunits A , B, and c of the V-ATPase with subunits p (catalytic), (Y (regulatory)* and c (proteolipid) of the F-ATPase or by comparison of subunit A , B, and c sequences from different organisms to address to the following two questions: 1. Where do the V-ATPases originate from? 2. What are the evolutionary relationships between V-ATPases of different plant species? Comparison of sequences of subunits A and B with subunits p and a confirmed that V-ATPases and F-ATPases are genetically related but independent groups of proton pumps. Subunits A and /3, respectively, are highly conserved, with about 60% identity in different organisms. Conversely, the overall identity of subunits A and p among each other is only 25%, i.e. relatively low (Nelson and Taiz, 1980). Nevertheless, a number of highly conserved regions have been dctccted, including the catalytic site (Zimniak rt al., 1988). Moreover, the honiology of subunits A and B and

*Note that the F-ATPase nomenclature starts with the regulatory subunit (a)followed by the catalytic subunit (p). in contrast to the V-ATPase nomenclature, where A is the catalytic and B the regulatory subunit, so that the functional correspondence is A-p and B-a, respectively.

256

U. LUTTGE and R. RATAJCZAK

TABLE I Species for which V-ATPase subunit sequences have been published, and the number of isoforms of the respective subunits. Asterisks indicae sequence data derived from polymerase chain reaction cloning of partial cDNAs. Letters in parentheses denote references (see footnote) Subunit

Organism

A (catalytic)

Avena sativa* (c) Chenopodium sp.* (c) Clematis sp.* ( c ) Daucus carota (m) Gossypium hirsutum ( I ) Hydrastis sp.* (c) Lycopersicon esculentum * (c) Mesembryanthemum crystalhum* ( h ) Nicotiana tabucum* (c) Zea mays* (n) Arabidopsis thaliana ( i ) Gossypium hirsutum (k) Hordeum vulgare (a) Nicotiana tabacum (e) Mesembryanthemum crystallinurn* ( h ) Hordeum vulgare (b) Arabidopsis thaliana (j) Avena sativa ( f ) Beta vulgaris* (g) Clusia minor* ( h ) Daucus carota* ( h ) Gossypium hirsutum ( d ) Mesembryanthemum crystallinurn* ( h ) Zea mays* (n)

B (regulatory)

E (31 kDa) c (proteolipid)

Number of isoforms

a, Berkelman et al. (1994); b, Dietz et al. (1995); c, Gogarten et al. (1992); d, Hasenfratz et al. (1995); e, Hortensteiner et al. (1994); f , Lai et al. (1991); g, Liiw, R. and Rausch, T., unpublished; h, Liiw et al. (1996); i, Manolson et al. (1988); j , Perera et al. (1995); k , Wan and Wilkins (1994); 1, Wilkins (1993); m, Zimniak et al. (1988); n, Viereck et al. (1996).

/3 and

a, respectively, indicates that an ancient gene duplication gave rise to the catalytic and non-catalytic subunits of V-ATPases and F-ATPases. The observation that subunits /3 and a of the archaebacterial plasma membrane H+-ATPase exhibit a higher degree of homology to subunits A and B of the eukaryotic V-ATPase than to the p and a subunits of the eukaryotic F-ATPase led to the conclusion that the V-ATPases and the F-ATPases evolved from the same enzyme which was present in the common ancestor (Gogarten and Taiz, 1992). This is supported by the fact that the archaebacterial H+-ATPase resembles the eukaryotic V-ATPase more than the


257

eukaryotic F-ATPase in terms of its insensitivity to azide and its sensitivity to nitrate (see Section IV). However, this conclusion might be premature since horizontal gene transfer between different organisms would also be sufficient to explain structural similarities of ATPase subunits (Hilario and Gogarten, 1993). Since only very few full-length sequences of V-ATPase subunits are known (Table I) the polymerase chain reaction (PCR) recently became a useful tool to deduce evolutionary relationships from V-ATPase sequence comparisons. Figure 1 shows dendograms obtained by sequence analysis and comparison of partial cDNAs of V-ATPase subunits A and c using PCR in a number of higher plant species. In both cases monocotyledons are clearly separated from dicotyledons, and in the dicotyledonous species, families and subclasses are also grouped together as expected from the currently accepted angiosperm taxonomy. This shows the power of the method. However, more sequence data are needed for more far-reaching evolutionary analyses. As shown in Table I in many cases several V-ATPase A , B and c subunits in a given plant are encoded by more than one gene. Different types of V-ATPase consisting of different subunit isoforms might be located in different cellular membrane systems (see Section 111). Additionally or alternatively, different isoforms could be expressed during the development of the plant and in response to certain environmental conditions. Recent reports indicate a differential expression of V-ATPase subunits in a tissue-specific manner (Hasenfratz et a l . , 1995; Low et al., 1996) which could also involve differential expression of isoforms, as it was demonstrated for subunit c of Gossypium hirsutum (Hasenfratz et al., 1995). In Citrus limon var. Schaub Rough Lemon different V-ATPase isoforms seem to be present in epicotyls and fruits (Miiller el a l . , 1996).

111. ONTOGENY With its combination of integral and peripheral membrane subunits the structure of the V-ATPase makes its assembly and intracellular transport pathway particularly interesting (Kane and Stevens, 1992). Indeed, the V-ATPase which is quite well characterized at the physiological, biochemical and molecular biological level could serve as a model system for the investigation of multisubunit protein ontogeny. However, up to now only a small amount of information has been obtained about the assembly, intracellular transport and turnover of the plant V-ATPase. For about 10 years it has been known that the V-ATPase is not exclusively located at the tonoplast, although the enzyme usually is used as a vacuolar marker. Based either on co-migration of V-ATPase with marker enzymes during purification of membrane vesicles or on immunoelectron microscopy using specific antibodies cross-reacting with V-ATPase subunits, several authors reported

Phylogenetic relationships of V-ATPase subunits Subunit A

A

I

r-

-1

1

I

Species Hydrastis spec.

Family

I

Sumass

Ranunculaceae

Ranunculidae

Chenopodium spec.

Chenopodiaceae

Ca 0phfidae

Daucus carofa L.

Apiaceae

Rosidae

Solanaceae

Lamiidae

Avena safiva L.

Poaceae

Liliidae

vulgaris L.

Chenopodiaceae

Mesembryanthemum crystallinurn L.

Aizoaceae

Daucus carota L

Apiaceae

Clma minor L

Clusiaceae

Clematis spec.

Lympersicon esculenfurn L. Nicotiana fabacum L.

Subunit c

1

A vena sabva L Zea mays L (isoform 1)

I

Poaceae

I

;&&ae Rosidae

class Dicotyledonae

Monocotyledonae

Dicotyledonae

Dilleniidae

Liliidae

Monocoty. ledonae

Fig. 1. Phylogenetic relationships of V-ATPases: dendograms deduced from partial cDNA sequences of genes encoding subunit A and subunit c of the V-ATPase in different plant species from different families. (Data of subunit A and subunit c are from Gogarten et al. (1992) and Liittge et al. (1995b), respectively. Taxonomy is according to Ehrendorfer (1991).)


259

the existence of V-ATPases in other cellular membranes. Plant V-ATPases were found in membranes of the endoplasmic reticulum (ER) (Churchill et al., 1983; Hager and Biber, 1984) and the Golgi (Chanson and Taiz, 1985; Ali and Akazawa, 1986). Later V-ATPases were detected in purified clathrin-coated vesicles (Depta et al.. 1991; Oberbeck et al., 1994) and in the plasma membrane of Vigna radiata hypocotyl (Mito et al., 1988; Kimura et al., 1988), Zea mays coleoptile cells (Hurley and Taiz, 1989), Ricinus cornrnunis cotyledons (Williams et al., 1990) and Pisurn sativurn cotyledons (Robinson et al., 1996). In Gossypium hirsuturn ovules the subcellular distribution of the V-ATPase is quite dynamic (C.-Y. Wan, M.-P. Hasenfratz, D. Gunasekera and T. A . Wilkins, unpublished data). While a minor fraction of membrane-associated V1 subunits A and B is present at the tonoplast in preanthesis ovules, during the period of rapid trichome expansion V-ATPase activity is almost exclusively associated with the tonoplast. A priori the presence of V-ATPases in different membrane systems is not surprising because the vacuole is an integral part of the endomembrane system including the E R , the Golgi network, clathrin-coated vesicles, secretory vesicles, plasma membrane, the nuclear envelope and transition vesicles (Harris, 1986; Marty, this volume; Bassham and Raikhel, this volume). Due to the function of the endomembrane system including membrane and organelle biogenesis with membrane vesicle exchange, storage of solutes within the organelles and transport of substances destined for extracellular secretion intracellular membrane flow more or less closely connects the various types of membrane with each other. However, one has to ask the question if the V-ATPases which are present in different membranes are functional proton pumps energizing secondary transport processes in the respective organelles or if they are precursors of the mature V-ATPase holoenzyme occurring during assembly. The current model for V-ATPase assembly in yeast (Kane and Stevens, 1992) suggests independent formation of the V,, and V1 domains. V, is assumed to be assembled in the E R , while there is evidence that subunits of V, are synthesized and assembled in t h e cytosol. It is still unclear in which cellular compartment of yeast V1 and V, are coupled to obtain a functional V-ATPase. Two recent reports indicate that in the plant cell the mechanism of V-ATPase assembly might be different. Using monoclonal antibodies directed against peripheral V-ATPase subunits, Herman et al. (1994) detected V, subunits in addition to V, subunits in the ER, Golgi-derived membrane vesicles and provacuoles of Avena saliva root tip cells. By application of different techniques of membrane vesicle purification, Oberbeck et a f . (1994) found V-ATPases with a complete set of peripheral and membrane integral subunits in highly purified fractions deriving from endoplasmic reticulum and Golgi of corn (Zea mays) root cells. These findings support a model in which the V-ATPase holoenzyme is completely assembled in the E R and becomes part of the vacuole directly from the ER or is sorted via the Golgi to

260


endomembrane systems (Fig. 2). However, further work on the assembly of the V-ATPase holoenzyme is needed since both the studies on Avena sativa and Zea mays could not provide a conclusive model. In animal cells, functional V-ATPase was found in different endomembrane systems of the endocytotic pathway, including clathrin-coated vesicles (Forgac, 1992) and endosomes (Forgac, 1989; Mellmann et al., 1986) where lumenal acidification enhances ligand-receptor binding and, therefore, is essential for receptor-mediated endocytosis. It has also been found in the plasma membrane of renal epithelial cells (Gluck et al., 1994), insect midgut epithelial cells (Dow, 1994) and the osteoclast (Baron et a f . , 1994). However, for the plant cell the question still remains open if the V-ATPases found in cellular membranes other than the tonoplast are active in ATP hydrolysis and proton transport. In contrast to previous studies on the V-ATPase of Golgi vesicle fractions (Chanson and Taiz, 1985; Ali and Akazawa, 1986), Oberbeck et al. (1994) could not detect ATP hydrolysis o r proton transport activity of V-ATPases in highly purified ER or Golgi vesicles, while the V-ATPase of tonoplast vesicles isolated from the same tissue was active. The authors mentioned that the V-ATPase in ER or Golgi might be inactive and needs activation by post-translational modification of enzyme structure which could occur at the tonoplast. However, for the same reason, i.e. structural differences of the V-ATPases of the tonoplast, the ER or Golgi (or the existence of different isoforms of the V-ATPase occurring in different membranes; see Section IV), different properties of the enzyme, e.g. in terms of pH dependency or ion sensitivity, could also be invoked. Since Oberbeck et at. (1994) performed assays of enzyme activity under one given condition for all membrane vesicle populations it cannot completely be excluded that the ER and Golgi V-ATPases do not function as ATP-driven proton pumps under the conditions used but might in others. Different properties of V-ATPases located in different cellular membranes have already been detected. For example, Golgi-associated V-ATPases seem to be less sensitive to inhibition by nitrate than the tonoplast V-ATPase (Chanson and Taiz, 1985; Ali and Akazawa, 1986), and Vincente and Vale (1994) detected a V-ATPase in ER membrane vesicles from corn roots which according to inhibitor sensitivities is distinct from the V-ATPase of the tonoplast. An alternative promising technique to investigate the assembly of the V-ATPase at the tonoplast is to follow the reformation of vacuoles in evacuolated protoplasts (Burgess and Lawrence, 1985). This method allows the ontogeny of the vacuole to be studied in a synchronized system. Electron microscopy performed during vacuole reformation in evacuolated tobacco protoplasts confirmed the old notion that the ER is involved in vacuole formation (Hortensteiner et al., 1992). In subsequent studies (Hortensteiner et al., 1994; Hoffmann and Hampp, 1994) it was shown that both the V-ATPase and the V-PPase disappeared in cells after evacuolation and reappeared during reformation of the vacuole. However, at present it seems


26 1

Fig. 2 . Ontogenetic relationships of V-ATPases: distribution and activity of the plant V-ATPases in different cellular compartments and putative trafficking of V-ATPases via membrane flow (arrows) within the cell. Question marks indicate that the activity of the V-ATPase has not been unequivocally demonstrated for the respective compartment. ccv, clathrin-coated vesicles; pm, plasma membrane; psv, protein storage vacuole; pv, provacuole. References are given in brackets: a , Ali and Akazawa (1986); b, Chanson and Taiz (198s); c, Churchill et al. (1983); d , Depta et al. (1991); e, Hager and Biber (1984); f, Herman er al. (1994); g, Hurley and Taiz (1989); h, Oberbeck et al. (1994); i , Maeshima et al. (1994); j, Vincente and Vale (1994); k , Mito et al. (1988); I , Kimura er al. (1988); rn, Williams et al. (1990); n, Robinson et al. (1996).

262


to be impossible to isolate sufficient amounts of microsomal fractions from evacuolated protoplasts during different stages of reformation of the vacuole to separate different membrane vesicle populations, which is necessary for the investigation of assembly and trafficking of V-ATPases using biochemical methods. Alternatively, immunoelectron microscopy of cells might be helpful for the investigation of V-ATPase trafficking. An in vivo event of vacuole reformation occurs during germination of pumpkin (Cucurbita sp.) seeds (Nishimura and Beevers, 1979). In the late stage of seed maturation, protein storage vacuoles are generated by fragmentation of the central vacuole (Hara-Nishimura et al., 1987). During germination the proteins stored in the protein storage vacuoles are hydrolysed, and protein storage vacuoles are fused, reforming a single central vacuole. Maeshima et al. (1994) demonstrated that both the V-ATPase and the V-PPase are present in the membrane of protein storage vacuoles and function as proton pumps acidifying the interior of the vesicles, which is necessary to stimulate the activity of acid hydrolases. During germination the amount of proton pumps in the protein storage vacuoles increased, indicating a de novo synthesis of V-ATPase and V-PPase and incorporation into the storage vacuole membrane prior to the formation of the central vacuole. However, nothing is known about the protein trafficking to the protein storage vacuoles.

IV. PROPERTIES Initial evidence for the existence of a V-ATPase at the plant tonoplast came from studies of the vacuole-like Hevea lutoids (D’Auzac, 1975, 1977). ATP hydrolysis activity of the enzyme was maximal at p H 7.5-8.0 and was stimulated by C1- . Subsequent investigations showed that stimulation of the V-ATPase by anions is a common feature of all V-ATPases studied so far (Table 11). The activity of the V-ATPase can be distinguished from the activity of other ATP-hydrolyzing proton pumps by its sensitivity to different inhibitors (Sze, 1985; Table 11). The V-ATPase has long been considered to be insensitive to vanadate (but see below) in distinction to the P-ATPase, where vanadate prevents formation of the phosphorylated intermediate necessary in the functional cycle (Quail, 1979). The V-ATPase is sensitive to nitrate (which also inhibits the activity of the F-ATPase) but is insensitive to azide and oligomycin*, which are F-ATPase inhibitors. Thus, for about one decade V- ATPase activity was characterized as azide-resistant, nitrate-sensitive ATP hydrolysis and ATP-dependent H+-transport activity. *Note that o for oligomycin has been used to denominate the membrane integral part of F-ATPases as F , , where oligomycin is binding.


263

More recently it has been discovered, however, that some plant V-ATPases are in fact vanadate-sensitive to some extent, and the H+-transporting ATPase of tonoplasts of Acer pseudopfatanus cells was shown to be a V-ATPase having a phosphorylated intermediate and operating with a plasma membrane-type ATPase catalytic mechanism (Magnin et a f . , 1995). Another example is the V-ATPase of Citrus limon fruit tonoplasts, which has been demonstrated to be vanadate-sensitive (Muller et a f . , 1996). Thus, it is important to have a rather specific inhibitor of the V-ATPase, i.e. the macrolide antibiotic bafilomycin A, as described by Bowman et a f . (1988). This antibiotic inhibits the activity of V-ATPases from different sources at the very low concentration of lop9 M. Bafilomycin A l also inhibits M), while F-ATPases are P-ATPases, at much higher concentrations ( completely unaffected (Drose et al., 1993). The mechanism of inhibition of the plant V-ATPase by bafilomycin Al is unknown. Recently, members of another group of macrolide antibiotics, concanamycins, were found to be even more effective inhibitors than bafilomycin A l (Drose et a f . ,1993). Total inhibition of V-ATPase by concanamycin is obtained at concentrations around M. It was demonstrated that bafilomycin A, not only inhibits the V-ATPase of plant cells but also the enzyme of animal cells (Zhang et a f . , 1994) and yeast (Crider et a f . , 1994) by blocking the membrane integral protontranslocating V , domain. These authors suggested denoting the membrane integral domain of the V-ATPase VB instead of Vat. In fact, in view of the dubious specificity of various inhibitors - now even including vanadate (see above) - and the localization of the “V”-type ATPases not only in vacuolar membranes but also in plasma membranes, e.g. in plants in Vigna radiata hypocotyl (Kimura et a f . , 1988; Mito et al., 1988), Zea mays coleoptile cells (Hurley and Taiz, 1989), Ricinus communis (Williams et a f . , 1990) and Pisum sativum (Robinson et a f . , 1996) cotyledons, and in animals in renal epithelial cells (Gluck et al., 1994), insect midgut epithelial cells (Dow, 1994) and osteoclasts (Baron et a f . , 1994), Bafilomycin-sensitivity at the moment appears to be the only differential criterion to distinguish this type of H+-ATPase from other Hf-ATPases. Thus, “vacuolar” appears to be a misnomer, and it would be a step further not only to call the proteolipid “Vs” but the whole enzyme “B-ATPase”. However, since the “V”-ATPase of lemon fruit tonoplasts exhibits very low bafilomycin sensitivity (Muller et a f . , 1996), even the expression “B-ATPase” would not be sufficient to denominate all H+-ATPases of the vacuolar type. Beside the diagnostic inhibitors mentioned above, V-ATPase activity is also inhibited by a variety of protein-modifying agents which covalently bind to amino acid residues of different V-ATPase subunits (Table 11). By use

‘V‘, actually is a misnomer from the analogy to F, where o refers to oligomycin binding

TABLE I1 Properties of the plant V-ATPase Stimulators c1maIate2Inhibitors Diagnostic inhibitors N03c10,Bafilomycin A l Concanamycin Protein-modifying reagents DCCD DES

DIDS FITC Nbd-C1 NEM PCMBS Anti-calmodulin drugs w5 w7 Calmidazolium Calcium channel antagonists Verpamil Diltiazem

C1-, Br- > HC03- (w) Maximal stimulation at 50 mM (d,p,q) Half-maximal stimulation at 1mM (w,x) Maximal stimulation at 25-50 mM (p,q) Inhibition of ATP hydrolysis Inhibition of H+ transport Z50 = 2.5-10

mM (i,n,w,y) Z50 = 5 mM (w) 150 = 1-26 nM (c,f,y,z) Complete inhibition at 1 nM (s) 150 = 3-20

p M (i,j,k,n,m,w,x)

Z50 = 4-20

p M (a,n,s,w)

I50 =

mM (Y)

Z50 = 0.1 nM

(f)

(s) (s) Z50 = 25 FM (s) I50 = 10 p M I50 = 20 p M

Z50= 0.33 mM (v) 150

= 0.5-10 p M (j,I,n,w,b')

150 = 0.4-12

/.LM(s,w)

I50 = 2-30

p M (f,n,s,w) 150 = < 20 p M ( f ) 150 = 7.5 p M (m) Z50 = 71 nM (m) 150 =

26 nM (m)

150 = 398 nM

(m) Z50 = 500 nM (m)

Others Oryzalin Quercitin Be ticoline [(w-32P]3-o-(4-benzoy1) benzoyladenosine 5I-triphosphate Km (Mg A - w Turnover rate Stoichiometry

Zyj = 10 /LM (u)

Z, I,,

= 13 p M = 50 nM

(w) (a')

150 =

2 PM (w)

1x1= 1 1 PM (k) 0.19-0.81mM (a,p,q,r,w,y,a'); 0.77mM and 2 p M (y) 50ATP s-l (8) 2 H+/ATP (b,h); 1.8-3.3 H+/ATP (e)

ZS0, concentration of inhibitor required to give 50% inhibition; DCCD, N,N'-dicyclohexylcarbodiimide;DES, diethylstilbestrol; DIDS. 4,4'-dithioisocyano-2,2'-stilbene disulfonic acid; FITC, fluorescein 5'-isothiocyanate; Nbd-C1, 7-chloro-4-nitrobenzo-2-oxa-1.3diazole: NEM, N-ethylmaleimide; PCMBS, p-chloromercuriphenyl suifonic acid. References are given in parentheses: a, Banuls et al. (1993);b, Bennett and Spanswick (1984);c, Bowman er al. (1988);d , D'Auzac (1975);e, Davies et al. (1994):f. Hoffmann-Thoma and Willenbrink (1993);g, Liittge et al. (1995a); h, Liittge et al. (1981);i, Mandala and Taiz (1985);j. Mandala and Taiz (1986); k, Manolson et al. (1985);1, Matsuura-Endo et al. (1990); m, Pfeiffer (1995); n, Randall and Sze (1986); 0 , Rea et al. (1987); p, Struve and Liittge (1987); q, Struve and Luttge (1988); r, Struve et al. (1985);s, Takeshige et al. (1988); t, Tazawa et al. (1995); u , Tu ef al. (1995);v, Tzeng et al. (1992);w, Wang and Sze (1985);x, Ward and Sze (1992b);y , Warren er al. (1992);z, White (1994); a', Willmer et al. (1995); b', Yamanishi and Kasamo (1992).

266


of these protein-modifying reagents it was possible to reveal information about the function of different subunits (see Section V). Binding of NEM (see Table I1 for inhibitor abbreviations) (Wang and Sze, 1985; Randall and Sze, 1986), FITC (Tzeng et al., 1992) or Nbd-CI (Mandala and Taiz, 1986; Yamanishi and Kasamo, 1992) to the nucleotide-binding subunit A inactivates the V-ATPase, indicating that subunit A is the catalytic subunit of the V-ATPase. Nbd-CI seems to bind at or near the catalytic site since the V-ATPase can be protected from inhibition by addition of its substrate MgATP (Mandala and Taiz, 1986). The role of subunit A as the catalytic subunit is supported by the fact that its amino acid chain has a nucleotidebinding motif which is also present in the catalytic F-ATPase @-subunit containing a cysteine and a tyrosine residue which might be the binding site of Nbd-C1. From spectroscopic data, Yamanishi and Kasamo (1992) suggested that a cysteine residue is involved in Nbd-CI binding. Binding of the ATP analogue (a-32P]3-o-(4-benzoyl)benzoyladenosine5’-triphosphate to subunit B indicates that subunit B also contains a nucleotide-binding site (Manolson et al., 1985). This is supported by the presence of a putative nucleotide-binding sequence in subunit B (Manolson et al., 1988). In analogy to the F-ATPase a subunit, the V-ATPase B subunit is considered to have a regulatory function. DCCD inhibits the V-ATPase activity by covalent modification of the proteolipid subunit c (Manolson et al., 1985; Mandala and Taiz, 1986; Randall and Sze, 1986; Rea et al., 1987) which builds up the proton translocating V, domain of the V-ATPase. Interestingly, activity is inhibited by binding of one DCCD molecule per V-ATPase holoenzyme. Since V, comprises multiple copies of subunit c (Rea et a f . , 1987; Kaestner et al., 1988) this indicates a tight cooperation of c subunits during proton translocation. Several other substances inhibit the activity of V-ATPases (see Table 11). Among these inhibitors, anticalmodulin drugs and calcium channel antagonists are of special interest. Pfeiffer (1995) investigated the effect of a number of these compounds on V-ATPase activity. The direct inhibition of V-ATPase indicates that V-ATPase itself might play a role in the regulation of Ca*+-dependent metabolic responses and, thus, might function as a component of signal transduction chains (see Section VIII). The apparent K , values of the V-ATPase for MgATP range between 0.2 and 0.8 mM (see Table 11). However, investigation of reconstituted, partially purified V-ATPase of Kalanchoe daigremontiana (Warren et al. , 1992) revealed two distinct K , values of 0.77mM and 2 p M , respectively, indicating either the existence of two or more catalytic centres or cooperativity between nucleotide-binding sites. An estimation of the turnover number of the V-ATPase of Mesembryanthemum crystallinum was possible by relating the rate of ATP hydrolysis to the amount of V-ATPase of tonoplast vesicles which was determined immunologically (Ratajczak et al., 1994). Assuming a molecular mass of the V-ATPase holoenzyme of 584 kDa


267

(or 9.7 x 10-I6mg, which can be calculated by addition of the apparent molecular masses of the different subunits assuming three copies of subunit A, three copies of subunit B, six copies of subunit c and single copies of other subunits per V-ATPase holoenzyme (Arai et al., 1988)), from the specific - 1 h-1 (Ratajczak e t a l . , ATP hydrolysis activity of 300 pmol ATP mgV.AwPase 1994) a turnover rate of 50 ATP s-’ was estimated (Luttge et al., 1995a). Values of the turnover number which can be obtained from the ATP hydrolysis activity of purified V-ATPase are in the same order of magnitude as the value calculated from the immunological quantification of the V-ATPase, e.g. a turnover number of 30 ATP s - l is given by the activity of 180 pmol ATP mgprotein-lh-’ of the purified V-ATPase of Kalanchoe daigremontiana (Warren et a l . , 1992). Investigations on plants performing crassulacean acid metabolism (CAM) for thermodynamic reasons indicate a 2 H+/ATP stoichiometry of the V-ATPase (Luttge et al., 1981; Smith et al., 1982). Subsequently, a 2 H+/ATP stoichiometry was demonstrated for the V-ATPase of Beta vulgaris (Bennett and Spanswick, 1984). An interesting observation was published recently by Davies et al. (1994), who used patch clamp techniques and found a variable coupling ratio of ATP hydrolysis and proton transport ranging between 1.75 and 3.28 in dependence on cytoplasmic and vacuolar pH (see also Davies, this volume). A variable ATP-proton stoichiometry might be one factor of regulation of V-ATPase activity in vivo (see Section VIII). From differential effects of inhibitors on ATP hydrolysis and H+-transport activities, Brauer et al. (1993) suggested an indirect mechanism of coupling of H+-transport to ATP hydrolysis. In favour for such a model is the effect of oryzalin on V-ATPase activity (Tu et al., 1995). Oryzalin did not significantly inhibit ATP hydrolysis but decreased ATP-dependent H+-transport without increasing proton leakage of the tonoplast, which is the effect of common ionophores. However, further work is needed to understand the molecular coupling mechanism of ATP hydrolysis and H+-transport.

V.

HOLOENZYME SUBUNIT - FINE STRUCTURE

Initial investigations of the subunit composition of higher plant V-ATPase revealed only three polypeptides which were present in fractions of purified V-ATPase as “major components” (Mandala and Taiz, 1985, 1986). The apparent molecular masses of these polypeptides were about 70 (subunit A), 60 (subunit B) and 16 kDa (subunit c). Later it turned out that the subunit composition of the V-ATPase is much more complex, as could be shown by purification experiments and immunological investigations using different plant materials (see Fig. 3). Meanwhile up to 10 different subunits have been reported for the V-ATPase of Avena sativa, Hordeum vulgare and Pyrus communis. The “major components” (subunits A , B and c) were found in

268

U . LU7TGE and R. RATAJCZAK

ton oplas t membrane 16-20kDa iC)

'\

I

12 kDa

[ b,f,j I

13 kDa [ b,f,j I Fig. 3. Subunit composition and fine structure of the plant V-ATPase holoenzyme complex. The diagram shows V-ATPase subunits found in 10 different plant materials. Different subunits are denominated by their apparent molecular mass and, where possible, by characters in parentheses, following the nomenclature of Moriyama and Nelson (1989) and Ratajczak et al. (1994). The location of subunits in the membrane peripheral V I domain and in the membrane integral V, domain was demonstrated by dissociation experiments for Avena sativa (Lai ef al., 1988; Ward and Sze, 1992a), Beta vulgaris (Rea et al., 1987), and Mesembryanthemum crystallinum (Berndt, 1993; Luttge et al., 1995a). For V-ATPase subunits from all other materials allocation of subunits to V I and V, was made according to similarities of apparent molecular mass. Subunits which are present in all materials investigated so far are hatched (with the exception that subunit C was not found by Mandala and Taiz (1985, 1986) in Zea mays roots). For all other subunits characters in square brackets indicate the plant


269

all V-ATPases studied so far and, thus, seem to be subunits essential for V-ATPase function. In addition, a polypeptide exhibiting a molecular mass ranging from 37 to 52kDa (subunit C ) in preparations from different materials is present in all V-ATPases with the exception of the maize V-ATPase described by Mandala and Taiz (1985, 1986). Additional subunits of the V-ATPase reported in the literature were only detected in certain materials. Differences in subunit composition between V-ATPases of various plants suggest that subunit composition is either tissue-specific, environmentally or developmentally regulated (see Section VII). The assignment of subunits to the membrane peripheral V1 domain and the membrane integral V, domain was achieved by dissociation experiments. When incubated with a chaotropic reagent like potassium iodide or with 0.1 mM ethylenediamine tetraacetic acid (EDTA), V, dissociates from V, and the two domains of the V-ATPase can be separated. Chaotropic treatment of V-ATPases of Avena sativa (Lai et al., 1988; Ward and Sze, 1992a), Beta vulgaris (Rea et al., 1987) and Mesembryanthemum crystallinum (Berndt, 1993; Luttge et al., 1995a) led to the allocation of subunits to V1 and V, which is depicted in Fig. 3 with up to eight V1 subunits (63-72 kDa (A), 52-60 kDa (B), 37-52 kDa (C), 42-44 kDa, 30-42 kDa (D), 31-32 kDa (Di), 27-32kDa (E) and 27-28kDa (E,)) and up to five V, subunits (95-115 kDa, 30-32 kDa, 16-20 kDa (c), 13 kDa and 12 kDa). By size exclusion chromatography a total molecular mass of the VATPase holoenzyme complex of 400 kDa (Zea mays), 510 kDa (Kalanchoe daigremontiana) and 650 kDa (Avena sativa) was determined (for references, see Fig. 3). A different experimental approach, i.e. radiation inactivation of ATP hydrolysis activity and proton transport activity, revealed a molecular mass of 446 and 394 kDa, respectively, of the V-ATPase of Beta vulgaris (Sarafian et al., 1992). These molecular masses are all within the same order of magnitude as would be expected from the addition of the apparent molecular masses of the different subunits assuming the subunit stoichiometry suggested by Arai et al. (1988) (see above).

Fig. 3 . continued. material in which they have been found and the respective references: a, Acer pseudoplatanus (Magnin et al., 1995); b, Avena sativa root (Randall and Sze, 1987; Kaestner et al., 1988; Ward and Sze, 1992a,b); c, Arachis hypogea seedling (Sen and Sharma, 1994); d, Beta vulgaris roots (Manolson et al., 1985; Rea et al., 1987; Parry el al., 1989); e, Citrus sinensis leaf (Batiuls et a l . , 1993); f, Hordeum vulgare root (DuPont and Morrissey, 1992); g, Kalanchoe daigremontiana leaf (Bremberger et al., 1988; Behre et al., 1992; Warren ef al., 1992); h, Mesembryanthemum crystallinum (C3 photosynthesis) leaf (Bremberger et al., 1988; Berndt, 1993; Ratajczak et al., 1994), i, M . crystallinum (crassulacean acid metabolism) leaf (Bremberger et a l . , 1988; Berndt, 1993; Ratajczak et a l . , 1994), j, Pyrus communis fruit (Hosaka et al., 1994).

270


VI. ELECTRON MICROSCOPY The V-ATPase can be visualized in the electron microscope by preparation of negatively stained membrane vesicles or of replicas of freeze fracture surfaces of vesicle membranes. The former presents top and side views of the V1 domain on the face and the rim of the vesicles, respectively (Fig. 4A), while the latter shows the intramembrane particles of the V, domain (Fig. 4B). The earliest reports about negatively stained membranes showing a head and stalk structure for the V-ATPase like that of the mitochondria1 and chloroplastic coupling factors (MF,Fl- and CF,F1-ATPases, respectively) are for bovine brain chromaffin granules (Kanaseki and Kadota, 1969; Schmidt and Winkler, 1982; Moriyama et al., 1991), proton-secreting cells of rat kidney, and toad and turtle urinary bladders (Brown et al., 1987), vacuoles of the fungus Neurospora crassa (Bowman et af., 1989) and acidosomes of Dictyostelium discoideum (Nolta et al., 1991). For higher plants the head and stalk structure of the V1 domain was depicted electron microscopically in tonoplast vesicles of Daucus carom (Lee Taiz and Taiz, 1991), Mesembryanthemum crystallinum (Klink and Liittge, 1991), Gfycine max ( M o d et al., 1991) and Beta vulgaris (Getz and Klein, 1995). In comparative studies, the V,VI-ATPase was clearly distinguished structurally from the MF,F1-ATPase, as it is larger and has a different shape with a typical cleft at the apex and also was found to occur in clusters (Lee Taiz and Taiz, 1991; Dschida and Bowman, 1992; Getz and Klein, 1995). Clusters were never seen in M . crystallinum. Dimensions of the V1 domain of the V-ATPase of N . crassa, B. vulgaris and D . carota are shown in Fig. 5. First, it must be noted that the ATPase head is supported by a rather slim stalk (see also Fig. 6) because some illustrations of the V-ATPase in the literature present rather thick stalk regions. Second, the basal arm-like structures of the stalk seen by Dschida and Bowman (1992), Getz and Klein (1995) and Lee Taiz and Taiz (1991) in these materials but not by Klink and Liittge (1991) and E. Berndt, I . Emig, G. Konig, C. Rezmer, U. Liittge, and R. Ratajczak (unpublished data) in M . crystallinum (Fig. 6) are intriguing. It is not clear if they represent genuine stalk polypeptides or membrane-associated polypeptides connected with the stalk. In the annual facultative halophyte M . crystallinum, where salinity induces a shift from C3 photosynthesis to CAM (see Section VII), the V1 domain of the ATPase changes sizes when plants are NaCl treated. Both the head and the stalk become somewhat wider and the length of stalk plus head above the membrane surface gets longer (Fig. 6). The increased widths of the stalk and head diameter are readily explained by the two additional stalk polypeptides Di and Ei appearing under salinity stress (see also Fig. 3 and Section VII) and by the upper end of the stalk protruding between the A


27 1

Fig. 4. (A) V-ATPase molecules seen in negatively stained tonoplast vesicles o f Mesemhrycznrhetnum crystallinurn with their heads in top view on the vesicle face and their head and stalk structures in side view at the vesicle edges and ( B ) their membrane integral proteolipid particles on freeze fracture replicas. The bar represents 100 nm. (From Luttge et a / . (199Sa).)

272

U . LUTTGE and R. RATAJCZAK -1

11.5 I -I

(9.41 9.0 1

I (13.6) 12.9

A

B

Fig. 5. Models and dimensions (nm) of the V1 domain of V-ATPases of (A) Neurospora crassa (Dschida and Bowman, 1992), (B) Beta vulgaris (Getz and Klein, 1995) and Daucus carota (numbers in parentheses in (B) from Lee Taiz and Taiz (1991)), obtained from negatively stained tonoplast vesicles.

and B subunits into the head domain. That the latter may occur is not implausible in view of the argument that the larger size of the V,V1-ATPase as compared to the MF,F1-ATPase cannot be exclusively explained by the larger subunits of the former but that in addition “the vacuolar ATPase may also have more open space between some of these components” (Dschida and Bowman, 1992). Up to very recently the head of the V-ATPase in the entire literature was always depicted as a hexamer of three A plus three B subunits although actually no evidence was ever presented for this assumption. Marginal support was inherent in biochemical analyses of Arai et al. (1988) which gave a subunit A:B ratio of 2.93:3.22 which by rounding up allows an A3B3 stoichiometry. However, the real reason for assuming an A3B3 stoichiometry is pure analogy to the hexameric structures of the MF,FIATPase and especially the CF,FI-ATPase. In the latter, the hexameric structure was elegantly depicted by improving the signal-to-noise ratio of electron micrographs using sophisticated alignment strategies of electron micrographs of the CF,F1-ATPase heads (Graber et al., 1990; Boekema and Bottcher, 1992). The study of two-dimensional crystals of the CF,FI-ATPase even allowed detection of conformational changes during the functioning of the CF,F,-ATPase (Boekema and Bottcher, 1992). Due to the low abundance

Fig. 6. Models and dimensions of the head and stalk (V, domain) and the membrane integral proteolipid (V, domain) of the V-ATPase of Mesembryanthemum crystallinum under control conditions and salinity stress, respectively. as obtained from measurements of intramembrane particles on electronmicrographs of freeze fracture replicas (Rockel et al., 1994) and digitized side views of V-ATPase (image in the centre) of negatively stained tonoplast vesicles (Emig, 1995).

274

U . LU'ITGE and R . RATAJCZAK

of tonoplast membrane as compared to the thylakoids in plant cells and, hence, of V,V1-ATPase as compared to CF,FI-ATPase, not enough material has yet been prepared of the former to obtain two-dimensional crystals. Nevertheless, using rotational analysis (Markham et a f . , 1963) and with experiments of signal-to-noise improvement by alignment of electron micrographs of single heads, similar approaches were applied to the V-ATPase of M . crystaffinurn (Kramer et a f . , 1995). It turned out that the head structure is not always a hexamer but is sometimes a pentamer. In rotational image analyses of about 1400 ATPase heads, an average 26.5% showed a hexameric and 29.8% a pentameric structure. The rest remained fuzzy, probably due to insufficient primary image contrast. Occurrence of a tetrameric structure was excluded. With a simultaneous occurrence of pentameric and hexameric V-ATPase heads, decimal numbers for the ratios of A and B subunits are of course possible. If an A subunit is missing from the pentameric head (Kramer et a f . , 1995) the A:B ratio obtained from the rotational image analyses is 0.83, i.e. close to the value of 0.91 obtained from the biochemical analyses of Arai et a f . (1988). It is assumed that the hexameric head structure is the catalytically active configuration. The pentameric head structure may be a relatively stable intermediate product of degradation and/or assembly of the holoenzyme. Clearly, as shown by assembly studies (see Section 111) and subunit turnover during stress adaptations and physiological control of enzyme activity (see Section VII) the V,V1-ATPase is less stable than the CF,F1- and MF,F,ATPases. This may also be due to the open space between subunits (Dschida and Bowman, 1992). Several indirect arguments, mainly enzyme abundance estimates, suggest that the membrane integral particles (IMPs) seen on freeze-fracture replicas of tonoplast vesicles (Fig. 4B) belong to the V, domain of the V-ATPase (Klink et a f . , 1990; Klink and Luttge, 1992). Direct evidence comes from quantitative comparative studies of native tonoplast membranes and reconstituted proteoliposomes. For K . daigremontiana, Fig. 7A shows that relative abundance of different size classes of IMPs in native tonoplast vesicles and in reconstituted proteoliposomes obtained by using total tonoplast protein is similar with a major peak at a diameter of 9.1 nm and a shoulder at 7.3 nm. Proteoliposomes prepared with purified V-ATPase only showed the peak of 9.1 nm, while a peak of 6.5-7.3nm was obtained with proteoliposomes made with purified tonoplast PPase. Thus, even these two membrane enzymes can be distinguished among the IMPs with diameters of 9.1 nm for the ATPase and -7 nm for the PPase (Mariaux et a f . , 1994). Quantitative and statistical analysis of IMPs also provides a readily applicable approach for studying physiological and ecological adaptations of the V, domain. During the maturation of leaves in M. crystaflinum the tonoplast PPase and the putative water channel VM23, which also may appear as IMPs, disappear, and, thus, the IMPs in mature leaves are solely

B l

'

I

'

I

r

50 40

30

20 10

-

n

so 50 8C 40 m rn

c 30 3

m a, .> c, m 0

20 10

0 50 40

30 20 10

0 0

5

10

15

0

5

10

15

IMP diameter (nm) Fig. 7. Relative abundance of different size classes of intramembrane particles in various preparations. (A) Kalanchoe daigrernontianu. Upper panel: native tonoplast vesicles (solid line) and proteoliposomes obtained by reconstitution with total solubilized tonoplast protein (dotted line). Centre panel: proteoliposomes obtained with two different fractions of purified tonoplast V-PPase. Lower panel: proteoliposomes obtained with purified V-ATPase (Mariaux et al., 1994). (B) Ecophysiological adaptations. Upper panel: Mesernbryanthernurn crystallinuni under control conditions (dotted line) and under salinity stress with CAM expression (solid line) (Rockel et a / . , 1994). Centre panel: Kalunchoe blossfeldiana cv. Tom Thumb under long days (dotted line) and under short days with enhanced CAM expression (solid line). Lower panel: Hordeurn vulgare under control conditions (dotted line) and with 150mM NaCl in the root medium (solid line) (Mariaux, 1994).

276


due to the V, domain of the V-ATPase. Under salinity stress, and with the expression of CAM, the IMPs become significantly larger (Fig. 7B and Fig. 6). This is accompanied by increases in the mRNA and protein levels of subunit c building up the membrane integral proteolipid V, domain (see Section VII). It is concluded that this means that there is a higher stoichiometry of c subunits per holoenzyme under stress than in the controls (Rockel et al., 1994). Similarly in Kalunchoe blossfeldiana cv. Tom Thumb, IMPs with larger diameters dominate when CAM expression is enhanced due to short day treatment, and this is accompanied by increased subunit c to subunit C or D stoichiometries (see Section VII), while no changes occur in barley under salt stress (Fig. 7B; Mariaux, 1994). Activities of the V-ATPase and densities of holoenzyme molecules per cell or vacuole and per unit tonoplast area have been obtained for various materials using different methods (Table 111). It is noteworthy that, notwithstanding the diversity in materials and methods used, all calculated densities of V-ATPase holoenzyme molecules per unit of tonoplast area are in the same order of magnitude, i.e. 1 X I d to 3 X lo3 pm-2. The higher densities are clearly obtained in leaves performing CAM (with 2.7 x lo3 pm-2 on average in the obligate CAM plant, K. daigremontiana, and in M . crystallinum in the CAM state) as compared to C3 leaves (with 1.2 X lo3 pmP2 on average in the obligate C3 plant, C. rubrum, and M . crystallinum in the C 3 state).

VII. PHYSIOLOGICAL FUNCTIONS AND ECOPHYSIOLOGICAL RESPONSES By pumping protons into the vacuole, the V-ATPase has two major types of functions: it removes H + ions from the cytosol and thus, it is part of a cytoplasmic pH-stat mechanism; it energizes the tonoplast by establishing an electrochemical proton gradient, Aj&+, across this membrane (see Davies, this volume). Both of these functions are basic physiological requirements in plant cell biology. Thus, prima facie the V-ATPase may be considered as a housekeeping enzyme. With respect to cytoplasmic pH-stat the V-ATPase is one element in a network of interrelated and connected components like the second H+-pump of the tonoplast, i.e. the tonoplast V-PPase (see Zhen et al., this volume), the H+-pumps of other endomembranes (see above: Section III), the plasma membrane H+-ATPase, H+/solute symport and antiport mechanisms and metabolic reactions affecting cytoplasmic acid/base equilibria. By energizing the tonoplast membrane through primary active transport of H + ions the V-ATPase drives a wealth of secondary active uniport, symport and antiport mechanisms (see various other chapters in this

TABLE I11 A T P hydrolysis activity and density of V-ATPase molecules per cell or vacuole and per unit tonoplast area in different plant materials as obtained by different methods Per vacuole or cell Plant material

Activity (nmol s-l)

Kalanchoe daigremontiana Immunonegative staining of tonoplast vesicles (a) 2.8 x lop6 Intramembrane particles of tonoplast vesicles (b) 4.2 X Entire vacuoles (c) 4.0 x 10-6 Mesembryanthemum crystallinum: controls Immunonegative staining of tonoplast vesicles (a) 2.6 x lop6 Entire vacuoles (d) 1.6 x Mesembryanthemum crystallinum: NaCl treated Immunonegative staining of tonoplast vesicles (a) 11.7 x lop6 Intramembrane particles of tonoplast vesicles (e) 14.0 X Entire vacuoles (d) 11.1 x 10-6 Chenopodium rubrum Cell suspension cultures, entire vacuoles - patch clamp studies (f)

Density (number)

Per unit tonoplast area Activity Density (mol mP2s - ' ) (number pm-')

79.7 x 106 121.7 X lo6 115.9 x lo6

77 118 112

2215 3380 3220

114.3 x lo6 72.0 x lo6

35 22

1540 970

189.5 x lo6 225.6 x lo6 179.1 x lo6

146 174 138

2360 2810 2230

-

1170

3.3 x 106

For tonoplast vesicles of K. daigremontiana and M. crystallinum ATPase densities were measured directly by relating immunogold labelling and intramembrane particles to unit tonoplast area; ATP hydrolysis and the ratio of ATPase-holoenzyme protein to total tonoplast protein were determined and a molecular mass of 584 000 g mol-' of the ATPase holoenzyme (Sze et al., 1992a) was assumed for calculations. For entire isolated vacuoles of K. daigremontiana and M . crystallinurn, measured ATP hydrolysis activity was related directly to unit tonoplast area, and it was assumed for calculations that the ratio of V-ATPase molecule/ATP hydrolysis was the same as in the experiments with vesicles. Morphometric analyses of K. daigremontiana and M . crystallinum leaf tissues were used to convert tonoplast area to a vacuole or cell basis. For Chenopodium rubrum estimates are based on analyses of protein per vacuole and compared with the electric current through the V-ATPase measured with the patch clamp technique. References are given in parentheses: a, Ratajczak et al. (1995); b, Mariaux (1994); c, Smith et al. (1984a); d, Struve et al. (1985); e , Rockel et al. (1994); f, Weiser and Bentrup (1994).

278


volume). In this way - possibly and partially together with the V-PPase it is largely responsible for functioning of the vacuole as a repository for waste solutes and a reservoir for solutes having roles in metabolism. With this importance of the vacuole, however, the V-ATPase also has essential functions in ecological adaptations. This evidently pertains to mineral nutrition with varying availability of nutrients in the environment. Unfortunately under such conditions the cytosolic nutrient ion levels effective at the tonoplast surface, i.e. in the microenvironment of the V-ATPase, mostly are basically unknown. This, so far, has hampered evaluation of the involvement of the V-ATPase in ecophysiological reactions based on its known in vitro responses to anions and cations (see Section VIII). Much more evidence is available for the participation of the V-ATPase in responses to stress due to chilling and salinity, where the V-ATPase itself undergoes changes of its molecular fine structure. In tonoplast vesicles from the hypocotyls of chilling-sensitive mung bean seedlings, H+-transport declined markedly when plants were kept below 10°C while there was no response even at 0°C in chilling-insensitive pea seedlings. This effect in the mung bean was not due to increased permeabilities of the vesicle membrane to H + or accompanying anions and cations but to a marked inhibition of the catalytic activity of both the V-ATPase and V-PPase (Yoshida and Matsuura-Endo, 1991). These changes were only observed after chilling in vivo and not in vitro. In other materials, cold inactivation of the V-ATPase does occur in vitro, but then it is not reversible (Moriyama and Nelson, 1989). This implies involvement of cytosolic interactions putatively in dissociatiodassembly equilibria of some of the nine detected subunits of the mung bean V-ATPase (Yoshida, 1991; Ward et al., 1992). This speculation originally was based on the observation that recovery was energy-dependent but did not require de novo protein synthesis (Yoshida, 1991). Somewhat later it was supported by the observation that the reduction of the V-ATPase activity observed after a 3-day exposure of mung bean hypocotyls to 0°C was associated with the decrease of the levels of several V-ATPase subunits in the tonoplast vesicle fraction but that the membraneintegral proteolipid was not affected (Matsuura-Endo et al., 1992). Hence, a reversible partial degradation of the V-ATPase holoenzyme by selective release of peripheral subunits in vivo during chilling is an intriguing possibility. There also appears to be acclimation as observed in tomato cell suspension cultures, where the temperature optimum of the V-ATPase broadened and shifted to a lower temperature range in tonoplast membranes from cells maintained at 9°C as compared to 28°C (DuPont and Mudd, 1985). In suspension cells of mung bean, the V-ATPase was very cold-sensitive at the early stage of exponential growth and less sensitive at the late state (Yoshida et al., 1993). Thus, although not confirmed by investigations at the whole-plant level yet, cold sensitivity of the V-ATPase could depend on the developmental status of the cell.


279

Much work has been devoted to following responses of the V-ATPase to salinity stress. Some changes were observed using crude microsome preparations from glycophytes and halophytes, but purified tonoplast fractions from salt-sensitive and salt-tolerant plants did not reveal substantial differences (reviewed by Liittge, 1993; Rausch et al., 1996). Strong evidence for specific salinity responses of the V-ATPase came from work with cultured tobacco cells. The kinetics of the enzyme change from hyperbolic to sigmoidal when cells are challenged with 428mM NaCl (Reuveni, 1992) and the specific H+-transport activity increases fourfold, which overcompensates for a reduction in the amount of the enzyme in the tonoplast (Reuveni et al., 1990). The amount of transcript (i.e. mRNA) of the 70 kDa subunit (subunit A) present was increased by short-term NaCl treatment in both salt-adapted and unadapted cultured cells of tobacco (Narasimhan et al., 1991) as well as intact plants of tomato (Binzel, 1995). In the latter, the increase of subunit A mRNA was only observed in expanded leaves and not in roots or young leaves and it was transient, returning to control levels within seven days after the onset of NaCl stress. Conversely, the transcript level of the plasma membrane H+-ATPase was increased in both roots and expanded leaves. This may reflect the concerted action of the H+-ATPases in response to salinity stress with salt accumulation via the roots and sequestration in vacuoles of expanded leaves until a steady-state vacuolar filling is reached (Binzel, 1995). In leaves of Citrus sinensis, subunit A shows strongly increased turnover when plants are under salinity stress in the root medium. A 35 kDa fragment of subunit A is formed which still exhibits ATP hydrolysis activity but can be separated from the holoenzyme by size exclusion chromatography (Baiiuls et al., 1995). Conversely, in barley under salinity stress in the root medium there was only a small increase in the amount of V-ATPase per unit of membrane area (Ratajczak et al., 1995) without any apparent changes in subunit stoichiometries (Mariaux, 1994). In the annual facultative halophyte Mesembryanthemum crystallinum the reaction to salinity is accompanied by a switch from C3 photosynthesis to CAM during which the V-ATPase also shows conspicuous responses. A standard experiment is shown in Fig. 8. CAM-induction is indicated by increasing dayhight changes of malate levels (Amal) after the onset of the salt treatment. When the salt is washed out again from the root medium, Amal remains high for a few days and then declines to the level attained in control plants of the same age which develop a certain level of CAM by ageing. The amount of V-ATPase holoenzyme protein in the tonoplast as determined immunologically shows a similar pattern although with two important differences: (i) it starts to increase immediately with the onset of salinity stress while Amal lags behind; (ii) it declines immediately when salt is removed from the root medium while again Amal follows later (Ratajczak et al., 1994). These quantitative changes of V-ATPase holoenzyme protein are accompanied by a remarkable array of other quantitative

280


150

..r E

E -

I

I

90

-

60

d

30

'

T

'A I

120

0 300 .-c a, + g 250

1

T

Q

200 \

a,

g

100

2

50

a

/

150

I

I

I

I

I I

0

I

I

I

I g o

i-T

I

I

I

I

2

4

6

8

I

I

I

I

I

10 12 14 16 18 20

time (days) Fig. 8. Time course of salinity-induced responses of M . crystallinurn: (A) day-night changes of malate levels (Amal), (B) amount of V-ATPase protein related to total tonoplast (TP) protein. The downward-pointing arrow indicates time of addition of NaCl (400 mM) to the root medium of salt-treated plants (solid symbols); the upward-pointing arrow indicates time of washing out of the salt from the root medium of some of these plants (*,A).Open symbols, controls; solid squares, plants treated with 400mM NaCl during the whole experiment. (From Ratajczak et al. (1994) .)


28 1

and qualitative responses of the enzyme in both its V, and its V1 domain: 0

0

0

0

0

0

The V-ATPase density per unit tonoplast surface-area increases (Ratajczak et al., 1995; see also Table 111 and Section VI). The amounts of mRNA and protein of subunit c increase (Rockel et al., 1994; Low et al., 1996; Tsiantis et al., 1996), the intramembraneous particles built up of subunit c increase in size (Rockel et al., 1994). Two new polypeptides (D, and El, see also Section VI) appear (Bremberger and Luttge, 1992a,b; Ratajczak et al., 1994) which are firmly attached to the holoenzyme as shown by immunoprecipitation with the holoenzyme when an anti-A serum is used (Ratajczak et al., 1994). This is associated with an increased stability of the ATPase complex against detergents (Ratajczak, 1994) and chaotropic reagents (Berndt, 1993). Cross-linking studies (E. Berndt, I. Emig, G. Konig, C. Rezmer, U. Luttge and R. Ratajczak, unpublished data) and the increased thickness of the stalk (see Fig. 6) suggest that the new polypeptides D, and El are part of the stalk region of the V1 domain. Comparison of the N-terminal amino acid sequence of the D, polypeptide (Bremberger and Luttge, 1992b) with the sequence of the V-ATPase subunit B derived from a partial cDNA clone indicated that D, originates from subunit B by proteolytic removal of a protein fragment of about 20 kDa from its N-terminus (Zhigang et al., 1996). As shown by phenolic extraction of leaf tissue the polypeptide D, was found to occur in vivo, and is not a proteolytic artifact due to the time-consuming procedure of tonoplast vesicle preparation (Zhigang et al., 1996). Thus, proteolytic processing of subunit B may have functional implication. Notwithstanding the possible enhancement of subunit B turnover it has not been shown, so far, that the pentameric head structure is more abundant in the CAM performing leaves of salt-treated M . crystallinurn than in the controls (Kramer et al., 1995).

VIII. CELL PHYSIOLOGICAL REGULATION These clear ecophysiological reactions of the V-ATPase immediately raise the question of the nature of possible signal transduction chains between the pertinent environmental factors and V-ATPase responses. Much work has been performed to check elements of cell physiological regulation networks known from animals and plants. However, it is still difficult to obtain an overview, because the experimental observations are isolated, sometimes contradictory, and often only apply to the situation in vitro. Nevertheless, Table 1V presents an attempt to give a brief keyword-type summary of the work, which may be commented on briefly as follows:

282

0

0


The role of the V-ATPase in cytoplasmic pH-stat mechanisms has already been mentioned above (see Section VII); feedback and a regulation of the V-ATPase itself by cytosolic p H may be important, as pH can also be envisaged as a secondary messenger in signal transductions (Felle, 1989; Kurkdjian and Guern, 1989). Substrate is evidently important for function; it is observed that cells switch to V-PPase operation when ATP is limiting, e.g. under hypoxia (Brauer et al., 1992; Carystinos et a[., 1995; Darley et al., 1995). It has already been mentioned that it is still hard to relate anion and cation sensitivity to nutritional responses. Initially cations where found to have little effect, although divalent cations are known to be necessary as part of the substrate, where Mn2+ and Zn2+ can partially replace Mg2+ in the metal-ATP2- complex. More recently it is suggested that Ca2+ and possibly Zn2+ participate as secondary messengers in regulatory networks. The stimulation by C1- and the inhibition by NO3- are distinctive features of the V-ATPase which also has been called “anion ATPase”. Nitrate possibly acts as a chaotropic solute determining ATPase subunit dissociation. Inorganic phosphate is a non-competitive inhibitor and may be involved in metabolic regulation of the V-ATPase. Regulations by redox activity including blue light effects, phosphorylation, phytohormones and light possibly involving phytochrome effecting extension growth, and via modification of membrane lipids remain intriguing possibilities.

Two more specific observations regard V-ATPase regulation at the transcriptional and posttranslational level, respectively: (i) In M . crystallinurn, transcript levels of V-ATPase subunits, particularly of subunit c in mature leaves, appear to be under diurnal control. In young leaves there is a slight day-night rhythm of subunit c mRNA. Under salt stress its amplitude is increased, and subunit A and B mRNA also show small day-night changes. In mature leaves the subunit c mRNA rhythm is more pronounced, and it is markedly increased in amplitude by salinity stress while there are no rhythms and salinity responses in A subunit mRNA whereas B subunit mRNA also oscillates (Rockel, 1996). In roots there is no diurnal rhythmicity, but salinity somewhat increases transcript levels of all three subunits (Low et al., 1996) while the rhythm is also fully expressed in photoautotrophic suspension cell cultures of M . crystallinurn (Rockel, 1996). This shows tissue-specific variations in V-ATPase transcription activity as also observed in Gossypiurn hirsutum (Hasenfratz et al., 1995). By contrast to the latter authors, who concluded from their mRNA analyses that the three major subunits (A, B and c) are coordinately regulated, the work of Low et al. (1996) with M . crystallinurn showed tqat transcription of different subunits occurred independently. This underlines the prominent role of subunit c in adaptation of M . crystallinurn to salinity and the expression of CAM in mature

TABLE IV Cell physiological responses of the V-ATPase in higher plants ~

Response

Experimental observations

pH sensitivity

pH optimum of V-ATPase in vitro 8.G8.5 (7.0-7.5) Increase of pH optimum from pH 7.4 to 8.4 due to salinity stress in tonoplast vesicles but not with solubilized V-ATPase from M . crystallinurn ATP >> GTP > NTP Little or no effect ATP-divalent cation complexes as substrate: Mg 2 Mn >> Ca, Co Ca2+/calmodulin stimulates activity Zn2+ inhibits, ZnATP2- can also serve as substrate C1- stimulated, NO3- and P, inhibited SH group effective reagents inhibit V-ATPase activity Blue light mediates V-ATPase inhibition by singlet oxygen and H 2 0 2which oxidize essential SH groups SH groups are essential in the catalytic A subunit and in the regulatory B subunit of the V-ATPase head The lipid microenvironment is important; phospholipids in the liquid crystalline phase are necessary for activity Tonoplast proteins are phosphorylated - perhaps only in vivo Tonoplast protein phosphorylation inhibits V-ATPase activity Plant lysophosphatidylcholine and the chemically similar animal lipid platelet-activating factor stimulate V-ATPase activity via protein-kinase stimulation; among a number of tonoplast proteins the V-ATPase B subunit is phosphorylated ABA may or may not stimulate and cytokinin may inhibit V-ATPase activity w,x.y,z,a‘,b’ Light inhibiting extension growth and vacuole enlargement downregulates subunit A and c mRNA C’

Substrate specificity Cation sensitivity

Anion sensitivity Redox regulation/SH-groups

Regulation by lipid environment Regulation by phosphorylation

Regulation by phytohormones Regulation by light

References

~~~~~~

References are given in parentheses: a, Sze (1985); b, Struve and Luttge (1987); c, Struve and Luttge (1988); d , Smith et al. (1984b); e , Jochem et al. (1984); f, Pfeiffer (1995); g, Kastrup (1994); h, Bennett and Spanswick (1983); i, Griffith et al. (1986); j , Takeshige et al. (1992); k. Hager and Biber (1984); 1, Wang and Sze (1985); m, Hoffmann-Thoma and Willenbrink (1993); n, Krauss et al. (1987); o. Hager and Lanz (1989); p. Yamanishi and Kasamo (1992); q, Yamanishi and Kasamo (1993); r. Zocchi (1985); s, Garbarino et al. (1991); t, Scherer et al. (1988); u, Scherer and Stoffel (1987); v, Martiny-Baron et al. (1992); w, Kasai et al. (1993a); x, Kasai et al. (1993b); y, Kasai et al. (1994); z, Narasimhan et al. (1991); a’, Binzel and Dunlap (1995); b’, Tsiantis et al. (1996); c’, Viereck et al. (1996).

284

U . LUTTGE and R . RATAJCZAK

leaves (see also Sections VI and VII). The highest subunit c transcription levels are observed at the end of the light period and in the early dark period when leaf water relations are bound to be most strained due to daytime transpiration. This mRNA rhythm is not accompanied by a rhythm in subunit c protein abundance, and although it is immediately stimulated after application of salinity stress to the root medium, protein abundance only increases gradually over several days (Luttge et al., 1995b). It may be speculated that the initial dayhight rhythm of subunit c mRNA is part of a signalling system governing the C3 photosynthesis to CAM switch in drought and salt stressed M. crysfallinurn (Winter and Gademann, 1991). Another observation which is noteworthy in relation to V-ATPase rhythmicity is that V-ATPase activity changes during the cell cycle in tomato cell suspension cultures (DuPont and de Gracia-Zabala, 1985). (ii) It has been observed in species of Kalanchoe that the tonoplast V-ATPase is kinetically regulated by the tonoplast V-PPase. If the ATPase is energized by addition of MgATP2- to a tonoplast vesicle preparation a few minutes after energization of the V-PPase by MgPP:-, its initial H+-transport rate is markedly enhanced. This does not happen when both substrates are added simultaneously or when other nucleotides are used, and no reciprocal effect is seen, i.e. when MgPPt- is added after MgATP2-. Potassium ions are not involved since valinomycin does not alter the effect (Marquardt-Jarczyk and Liittge, 1990a,b). The nature of the interaction between the two enzymes is not known. In Kalanchoe blossfeldiana cv. Tom Thumb, tonoplast PPase protein levels are much higher under short-day conditions, which enhance CAM in the leaves, than under long-day conditions (Mariaux, 1994). Thus, the PPase-ATPase interrelationship may be a special feature of CAM in Kalanchoe. Conversely, in mature CAMperforming leaves of M. crysfallinurn the PPase disappears (see Section VI) and, hence, has nothing to do with CAM in these plants.

IX. CONCLUSIONS AND OUTLOOK Molecular characterization of V-ATPase is already highly advanced. Physiological functions and ecophysiological reactions are well perceived. With respect to the latter, it is apparent that the V-ATPase is more than a housekeeping enzyme, although it is not strictly induced by stress and therefore differs from stress enzymes. This behaviour opens new perspectives of viewing involvement of particular enzymes, such as V-ATPase, in ecophysiological responses, and a new term for enzymes with such properties and functions is needed. We have suggested the term “eco-enzyme”, and defined it as follows: “-

an enzyme which is directly involved in ecophysiological adaptations at the molecular level (in contrast to housekeeping enzyme),

PLANT VACUOLAR ATPase “-

“-

285

an enzyme which is post-translationally modified in response to stress (by phosphorylatioddephosphorylation,oxidatiodreduction, proteolysis, change in subunit stoichiometry), an enzyme which shows a moderate change in gene expression at the transcript and/or protein level in response to stress (in contrast to strongly induced stress enzymes)”

(Liittge et a l . , 199%). In context with the signal transduction chains that must be operative, this clearly implies cell physiological regulation as discussed in Section VIII (Table IV), which is still so poorly understood in vivo. It will be one of the major tasks of V-ATPase research in the future to develop understanding of the regulation of assembly and dissociation of the very complex multisubunit V-ATPase holoenzyme under various conditions as well as further perspectives of the functioning of its cell physiological responses in signal transduction networks.

ACKNOWLEDGEMENTS We thank Mrs Doris Schafer and Dr Ekkehart Berndt (Darmstadt, Germany) for providing the artwork.

REFERENCES Ali, M. S. and Akazawa, T. (1986). Association of H+-translocating ATPase in Golgi membrane system from suspension cultured cells of sycamore. Plant Physiology 81, 222-227. Arai, H., Terres, G., Pink, S. and Forgac, M . (1988). Topography and subunit composition of the coated vesicle proton pump. Journal of Biological Chemistry 263, 87968802. Bands, J . , Ratajczak, R. and Liittge, U . (1993). Characterization of a protontranslocating ATPase in a tonoplast-vesicle fraction from Citrus. Journal of Plant Physiology 142, 319-324. Bands, J . , Ratajczak, R . and Liittge, U. (1995). NaCI-stress enhances proteolytic turnover of the tonoplast H+-ATPase of Citrus sinensis: appearance of a 35 kDa polypeptide still exhibiting ATP-hydrolysis activity? Plant Cell and Environment 18, 1341-1344. Baron, R., Bartkiewicz, M . , David, P. and Hernando-Sobrino, N. (1994). Acidification and bone resorption: the role and characteristics of V-ATPases in the osteoclast. In “Organellar Proton-ATPases” (N. Nelson, ed.), pp. 49-73. Springer-Verlag, New York. Behre, B., Ratajczak, R. and Liittge, U. (1992). Selective reconstitution of the tonoplast H+-ATPase of the crassulacean-acid metabolism plant Kalanchoe daigrernontiana. Boranica Acta 105, 260-265. Bennett, A. B. and Spanswick, R. M. (1983). Optical measurements of ApH and A+ in corn root membrane vesicles. Kinetic analysis of CI- effects on a protontranslocating ATPase. Journal of Membrane Biology 7 1 , 95-107.

286


Bennett, A. B. and Spanswick, R. M. (1984). H+-ATPase activity from storage tissue of Beta vulgaris. 11. H+/ATP stoichiometry of an anion-sensitive H+-ATPase. Plant Physiology 74, 545-548. Berkelman, T., Houtchens, K. A . and DuPont, F. M. (1994). Two cDNA clones encoding isoforms of the B subunit of the vacuolar ATPase from barley roots. Plant Physiology 104, 287-288. Berndt, E. (1993). Strukturuntersuchungen der Tonoplasten H+-ATPase von Mesernbryanthemurn crystallinurn L. mit Hilfe von Dissoziationsexperimenten. Diplomarbeit, Technische Hochschule Darmstadt, Germany. Binzel, M. L. (1995). NaC1-induced accumulation of tonoplast and plasma membrane H+-ATPase message in tomato. Physiologia Plantarum 94, 722-728. Binzel, M. L. and Dunlap, J. R. (1995). Abscisic acid does not mediate NaC1-induced accumulation of 70-kDa subunit tonoplast Hf-ATPase message in tomato. Planta 179, 563-568. Boekema, E. J. and Bottcher, B. (1992). The structure of ATP synthase from chloroplasts. Conformational changes of CF1 studied by electron microscopy. Biochimica et Biophysica Acta 1098, 131-143. Bowman, B. J . , Dschida, W. J., Harris, T. and Bowman, E. J. (1989). The vacuolar ATPase of Neurospora crassa contains an F1-like structure. Journal of Biological Chemistry 264, 15 606-15 612. Bowman, E. J., Siebers, A. and Altendorf, K . (1988). Bafilomycins: a class of inhibitors of membrane ATPases from microorganisms, animal cells, and plant cells. Proceedings of the National Academy of Sciences USA 85, 7972-7976. Brauer, D., Conner, De N. and Tu, S.-I. (1992). Effects of pH on proton transport by vacuolar pumps from maize roots. Physiologia Plantarum 86, 63-70. Brauer, D., Tu, S.-I., Hsu, A. F. and Patterson, D. (1993). Evidence for an indirect coupling mechanism for the nitrate-sensitive proton pump from corn root tonoplast membranes. Physiologia Plantarum 89, 588-591. Bremberger, C. and Luttge, U. (1992a). Dynamics of tonoplast proton pumps and other tonoplast proteins of Mesembryanthemurn crystallinum during the induction of crassulacean acid metabolism. Planta 188, 575-580. Brernberger, C. and Luttge, U. (19921.3). Tonoplast ATPase of Mesernbryanthemum crystallinum. Partial amino acid sequence of subunits induced during the transition from C3-photosynthesis to crassulacean acid metabolism. Comptes Rendus A c a d h i e des Sciences Paris 315(SCrie III), 119-125. Bremberger, C., Haschke, H.-P. and Luttge, U. (1988). Separation and purification of the tonoplast ATPase and pyrophosphatase from CAM plants with constitutive and inducible CAM. Planta 175, 465-470. Brown, D., Gluck, S. and Hartwig, J. (1987). Structure of the novel membrane coating material in proton secreting epithelial cells and identification as an H+-ATPase. Journal of Cell Biology 105, 1637-1648. Burgess, J. and Lawrence, W. (1985). Studies of the recovery of tobacco mesophyll protoplasts from an evacuolation treatment. Protoplasma 126, 140-146. Buser-Suter, C., Wiemken, A. and Matile, P. (1982). A malic acid permease in isolated vacuoles of a crassulacean acid metabolism plant. Plant Physiology 69, 45H59. Carystinos, G. D., MacDonald, H. R., Monroy, A. F., Dhindsa, R. S. and Poole, R. J. (1995). Vacuolar H+-translocating pyrophosphatase is induced by anoxia or chilling seedlings of rice. Plant Physiology 108, 641-649. Chanson, A. and Taiz, L. (1985). Evidence for an ATP-dependent proton pump on the Golgi of corn coleoptiles. Plant Physiology 73, 921-928. Churchill, K. A., Holoway, B. and Sze, H. (1983). Separation of two types of


287

electrogenic H+-transporting ATPases from oat roots. Plant Physiology 73, 921-928. Crider, B. P., Xie, X. S. and Stone, D. K. (1994). Bafilomycin inhibits proton flow through the H+-channel of vacuolar proton pumps. Journal of Biological Chemistry 269, 17 379-17 381. Darley, C. P., Davies, J. M. and Sanders, D. (1995). Chill-induced changes in the activity and abundance of the vacuolar proton-pumping pyrophosphatase from mung bean hypocotyls. Plant Physiology 109, 659-665. D’Auzac, J. (1975). Caracterisation d’une ATPase membranaire en presence d’une phosphatase acide dans les lutoides du latex d’Hevea brasiliensis. Phytochemlstry 14, 671-675. D’Auzac, J. (1977). ATPase membranaire de vacuoles lysosomales, les lutoides du latex d’Hievea brasiliensis. Phytochemistry 16, 1881-1885. Davies. J. M., Hunt, I. and Sanders, D. (1994). Vacuolar H+-pumping ATPase variable transport coupling ratio controlled by pH. Proceedings of the National Academy of Sciences USA 91, 854743551, Depta, H . , Holstein, S. E. H., Robinson, D. G., Lutzelschwab, M . and Michalke, W. (1991). Membrane markers in highly purified clathrin-coated vesicles from Cucurbita hypocotyls. Planta 183. 434-442. Diaz de Leon, J . L. and Wyn Jones, R. G . (1985). Ca2+-ATPasesand their occurrence in vacuoles of higher plants. In “Biochemistry and Function of Vacuolar Adenosine-Triphosphatase in Fungi and Plants” (B. P. Marin, ed.), pp. 57-66, Springer-Verlag, Berlin. Dietz, K.-J., Rudloff, S . , Ageorges, A., Eckerskorn, C., Fischer, K. and Arbinger, B. (1995). Subunit E of the vacuolar H+-ATPase of Hordeum vulgare L.: cDNA cloning, expression and immunological analysis. Plant Journal 8, 521-529. Dow, J . A. T. (1994). V-ATPases in insects. In “Organellar Proton-ATPases” (N. Nelson, ed.), pp. 75-102. Springer-Verlag, New York. Drose, S., Bindseil, K. U . , Bowman, E. J . , Siebers, A , , Zeeck, A. and Altendorf, K . (1993). Inhibitory effects of modified Bafilomycins and Concanamycins on P- and V-type adenosinetriphosphatases. Biochemistry 32, 3902-3906. Dschida, W. J . and Bowman, B. J. (1992). Structure of the vacuolar ATPase from Neurospora crassa as determined by electron microscopy. Journal of Biological Chemistry 267, 18 783-18 789. DuPont, F. M. and de Gracia-Zabala, M. (1985). Preparation of membrane vesicles enriched in ATP-dependent proton transport from suspension cultures of tomato cells. Plant Physiology 77, 69-73. DuPont, F. M. and Morrissey, P. J. (1992). Subunit composition and Ca-ATPase activity of the vacuolar ATPase from barley roots. Archives of Biochemistry and Biophysics 294, 341-346. DuPont, F. M. and Mudd, J. B. (1985). Acclimation to low temperature by microsomal membranes from tomato cell cultures. Plant Physiology 77, 74-78. Ehrendorfer, F. (1991). Evolution und Systematik. In “Strasburger Lehrbuch der Botanik” (P. Sitte, H. Ziegler, F. Ehrendorfer, A. Bresinsky, eds), 33rd edn, pp. 473-835. Fischer, Stuttgart. Emig, I. (1995). Elektronenmikroskopische Untersuchungen der Seitenansicht der Tonoplasten-H+-ATPase bei der C,-CAM intermediaren Pflanze Mesembryanthemum crystullinum L. Diplomarbeit, Technische Hochschule Darmstadt, Germany. Evans, D. E., Briars, S.-A. and Williams, L. E. (1991). Active calcium transport by plant cell membranes. Journal of Experimental Botany 42, 285-303.

288


Felle, H. (1989). pH as a second messenger in plants. I n “Second Messengers in Plant Growth and Development” (F. Boss and D. J. Morre, eds), pp. 145-166. Alan R. Liss Inc., New York. Forgac, M. (1989). Structure and function of vacuolar class of ATP-driven proton pumps. Physiological Reviews 69, 765-796. Forgac, M. (1992). Structure, function and regulation of the coated vesicle V-ATPase. Journal of Experimental Biology 172, 155-169. Futai, M., Nuomi, T. and Maeda, M. (1989). ATP synthase (H+-ATPase): results by combined biochemical and molecular biological approaches. Annual Reviews of Biochemistry 58, 111-136. Garbarino, J . E., Hurkman, W. J . , Tanaka, C. K. and DuPont, F. M. (1991). In vitro and in vivo phosphorylation of polypeptides in plasma membrane and tonoplast-enriched fractions from barley roots. Plant Physiology 95, 12191228. Getz, H. P. and Klein, M. (1995). The vacuolar ATPase of red beet storage tissue: electron microscopic demonstration of the “head-and-stalk” structure. Botanica Acta 108, 14-23. Gluck, S. L., Nelson, R. D., Lee, B. S. M . , Holliday, L. S. and Iyori, M. (1994). Properties of kidney plasma membrane vacuolar H+-ATPases: proton pumps responsible for bicarbonate transport, urinary acidification, and acid-base homeostasis. In “Organellar Proton-ATPases” (N. Nelson, ed.), pp. 163-221. Springer-Verlag, New York. Gogarten, J. P. and Taiz, L. (1992). Evolution of proton pumping ATPases - rooting the tree of life. Photosynthesis Research 33, 137-146. Gogarten, J . P., Starke, T., Kibak, H., Fishmann, J. and Taiz, L. (1992). Evolution and isoforms of V-ATPase subunits. Journal of Experimental Biology 172, 137-147. Graber, P., Bottcher, B. and Boekema, E. J . (1990). The structure of the ATP-synthase from chloroplasts. I n “Bioelectrochemistry 111” (G. Milazzo and M. Blank, eds), pp. 247-276. Plenum Press, New York. Griffith, C. J . , Rea, P. A., Blumwald, E. and Poole, R. J . (1986). Mechanism of stimulation and inhibition of tonoplast H+-ATPase of Beta vulgaris by chloride and nitrate. Plant Physiology 81, 120-125. Hager, A. and Biber, W. (1984). Functional and regulatory properties of H+ pumps at the tonoplast and plasma membranes of Zea mays coleoptiles. Zeitschrift fiir Naturforschung 39c, 927-937. Hager, A. and Lanz, C. (1989). Essential sulfhydryl groups in the catalytic center of the tonoplast H+-ATPase from coleoptiles of Zea mays L. as demonstrated by the biotin-streptavidin-peroxidase system. Planta 180, 116-122. Hara-Nishimura, I . , Hayashi, M., Nishimura, M. and Akazawa, T. (1987). Biogenesis of protein bodies by budding from vacuoles in developing pumpkin cotyledons. Protoplasma 136, 49-55. Harris, N. (1986). Organization of the endomembrane system. Annual Review of Plant Physiology 37, 73-92. Hasenfratz, M., Tsou, C.-L. and Wilkins, T. A. (1995). Expression of two related vacuolar H+-ATPase 16-kilodalton proteolipid genes is differentially regulated in a tissue specific manner. Plant Physiology 108, 1395-1404. Herman, E. M., Li, X . , Su, R. T., Larsen, P., Hsu, H.-t. and Sze, H. (1994). Vacuolar-type H+-ATPases are associated with the endoplasmic reticulum and pro-vacuoles of root tip cells. Plant Physiology 106, 1313-1324. Hilario, E . and Gogarten, J . P. (1993). Horizontal transfer of ATPase genes - the tree of life becomes a net of life. Biosystems 31, 111-119.


289

Hoffmann, E. M. and Hampp, R. (1994). Studies of vacuole regeneration in evacuolated tobacco mesophyll protoplasts. Protein analysis by one- and two-dimensional microgel-electrophoresis. Physiologia Plantarum 92, 563570. Hoffmann-Thoma, G. and Willenbrink, J . (1993). Differential sensitivities towards inhibitors and SH-compounds of H+-ATPase and Hf -pyrophosphatase associated with intact vacuoles from sweet sorghum (Sorghum bicolor [L.] Moench.) stem parenchyma. Journal of Plant Physiology 141, 681-689. Hortensteiner, S . , Martinoia, E. and Amrhein, N . (1992). Reappearance of hydrolytic activities and tonoplast proteins in the regenerated vacuole of evacuolated protoplasts. Planta 187, 113-121. Hortensteiner, S., Martinoia, E. and Amrhein, N . (1994). Factors affecting the re-formation of vacuoles in evacuolated protoplasts and the expression of the two vacuolar proton pumps. Planta 192, 395-401. Hosaka, M., Kanayama, Y., Shiratake. K. and Yamaki, S. (1994). Tonoplast H+-ATPase of mature pear fruit. Phytochernistry 36, 565-567. Hurley, D. and Taiz, L. (1989). Immunological localization of the vacuolar H+-ATPase in maize root tip cells. Plant Physiology 89, 391-395. Jochem, P., Rona, J.-P., Smith, J . A. C . and Luttge, U. (1984). Anion-sensitive ATPase activity and proton transport in isolated vacuoles of species of the CAM genus Kalanchoe. Physiologia Plantarum 62, 41W15. Kaestner, K. H., Randall, S. K. and Sze, H. (1988). N,N’-Dicyclohexylcarbodiimidebinding proteolipid of the vacuolar H+-ATPase from barley roots. Journal of Biological Chemistry 263, 1388-1393. Kanaseki, T. and Kadota, K. (1969). The “vesicle in a basket”. A morphological study of the coated vesicle isolated from the nerve endings of the guinea pig brain, with special reference to the mechanism of membrane movements. Journal of Cell Biology 42, 202-220. Kane, P. M. and Stevens, T. H. (1992). Subunit composition, biosynthesis and assembly of the yeast vacuolar proton-translocating ATPase. Journal of Bioenergetics and Biomembranes 24, 383-393. Kasai, M., Sasaki, M., Tanakamaru. S., Yamamoto, Y. and Matsumoto, H. (1993a). Possible involvement of abscisic acid in increases in activities of two vacuolar H+-pumps in barley roots under aluminium stress. Plant Cell Physiology 34, 1335-1338. Kasai, M., Yamamoto, Y., Maeshima, M. and Matsumoto, H. (1993b). Effects of in vivo treatment with abscisic acid and or cytokinin on activities of vacuolar H+-pumps of tonoplast-enriched membrane vesicles prepared from barley roots. Plant Cell Physiology 34, 1107-11 15. Kasai, M., Yamamoto, Y. and Matsumoto, H. (1994). In vivo treatment of barley roots with vanadate increases vacuolar H+-translocatingATPase activity of the tonoplast-enriched membrane vesicles and the level of endogenous ABA. Plant Cell Physiology 35, 291-295. Kastrup, V. (1994). Regulation der H+-ATPasen durch divalente Metall-Ionen. Dissertation, Technische Hochschule Darmstadt, Germany. Kibak, H., Taiz, L., Starke, T., Bernasconi, P. and Gogarten, J. P. (1992). Evolution of structure and function of V-ATPases. Journal of Bioenergetics and Biomembranes 24, 415424. Kimura, T., Maeshima, M. and Asahi, T. (1988). Immunological studies on ATPases in mung bean hypocotyl plasma membrane: proposal of the presence of two molecular species of ATPase. Plant Cell Physiology 29, 883-888. Klink, R. and Luttge, U . (1991). Electronmicroscopic demonstration of a “head and

290


stalk" structure of the leaf vacuolar ATPase in Mesembryanthemum crystallinum. Botanica Acta 104, 122-131. Klink, R. and Liittge, U. (1992). Quantification of visible structural changes of the ATPase in the leaf tonoplast of Mesembryanthemum crystallinum by freezefracture replicas prepared during the C3-photosynthesis to CAM transition. Botanica Acta 105, 414-420. Klink, R., Haschke, H.-P., Kramer, D. and Liittge, U. (1990). Membrane particles, proteins and ATPase activity of tonoplast vesicles of Mesembryanthemum crystallinum in the C3- and CAM-state. Botanica Acta 103, 24-31. Gamer, D., Mangold, B., Hille, A , , Emig, I., Hess, A., Ratajczak, R. and Liittge, U. (1995). The head structure of a higher plant V-type H+-ATPase is not always a hexamer but also a pentamer. Journal of Experimental Botany 46, 16331636. Krauss, W., Schiebel, G., Eberl, D. and Hager, A. (1987). Blue light induced, reversible inactivation of the tonoplast-type H+-ATPase from corn coleoptiles in the presence of flavins. Phytochemistry and Phytobiology 45, 837-844. Kurkdjian, A. and Guern, J. (1989). Intracellular pH: measurement and importance in cell activity. Annual Review of Plant Physiology and Plant Molecular Biology 40, 271-303. Lai, S., Randall, S. K. and Sze, H. (1988). Peripheral and integral subunits of the tonoplast H+-ATPase from oat roots. Journal of Biological Chemistry 263, 16 731-16 737. Lai, S., Watson, J. C., Hansen, J. N. and Sze, H. (1991). Molecular cloning and sequencing of cDNAs encoding the proteolipid subunit of the vacuolar H+-ATPase from a higher plant. Journal of Biological Chemistry 266, 16 078-16 084. Lee Taiz, S. and Taiz, L. (1991). Ultrastructural comparison of the vacuolar and mitochondria1 H+-ATPases of Daucus carota. Botanica Acta 104, 117-121. Li, Z.-S., Zhao, Y. and Rea, P. A. (1995). Magnesium adenosine 5'-triphosphateenergized transport of glutathione-S-conjugates by plant vacuolar membrane vesicles. Plant Physiology 107, 1257-1268. Low, R., Rockel, B., Kirsch, M., Ratajczak, R., Hortensteiner, S., Martinoia, E., Liittge, U. and Rausch, T. (1996). Early salt stress effects on the differential expression of vacuolar H+-ATPase genes in roots and leaves of Mesembryanthemum crystallinum. Plant Physiology 110, 259-265. Liittge, U. (1993). Plant cell membranes and salinity: structural, biochemical and biophysical changes. Revista Brasileira de Fisiologia Vegetal 5, 217-224. Liittge, U., Smith, J. A. C., Marigo, G. and Osmond, C. B. (1981). Energetics of malate accumulation in the vacuoles of Kalanchoe tubiflora cells. FEBS Letters 126, 81-84. Liittge, U., Fischer-Schliebs, E., Ratajczak, R., Kramer, D., Berndt, E . and Kluge, M. (1995a). Functioning of the tonoplast in vacuolar C-storage and remobilization in crassulacean acid metabolism. Journal of Experimental Botany 46, 1377-1 388. Liittge, U., Ratajczak, R., Rausch, T. and Rockel, B. (1995b). Stress responses of tonoplast proteins: an example for molecular ecophysiology and the search for eco-enzymes. Acta Botanica Neerlandica 44,343-362. Maeshima, M., Hara-Nishimura, I., Tackeuchi, Y. and Nishimura, M. (1994). Accumulation of vacuolar H+-pyrophosphatase and H+-ATPase during reformation of the central vacuole in germinating pumpkin seeds. Plant Physiology 106, 61-69. Magnin, T., Fraichard, A., Trossat, C. and Pugin, A. (1995). The tonoplast


29 1

H+-ATPase of Acer pseudoplatanus is a vacuolar-type ATPase that operates with a phosphoenzyme intermediate. Plunt Physiol0g.v 109, 285-292. Mandala, S . and Taiz, L. (1985). Partial purification of a tonoplast ATPase from corn colcoptiles. Plant Physiology 78, 327-333. Mandala, S. and Taiz, L. (1986). Characterization of the subunit structure of the maizc tonoplast ATPase. Journal of Biological Chemistry 261, 12 850-12 855. Manolson, M. F., Rea, P. A . and Poole. R. J. (1985). Identification of 3-0-(4benzoy1)benzoyladenosine 5’-triphosphate and N,N’-dicyclohexylcarbodiimidebinding subunits of a higher plant H+-translocating tonoplast ATPase. Journal of Biological Chemistry 260, 12 273-12 279. Manolson, M. F., Quellette, B. F. F., Filion, M. and Poole, R . J . (1988). cDNA sequence and homologies of the “57 kDa” nucleotide-binding subunit of the vacuolar ATPase from Arahidopsis. Journal o f Biologicul Chemistry 263, 17 987-17 994. Mariaux, J.-B. (1994). Structural investigation of the tonoplast H+-ATPase of plants performing C3-photosynthesis or CAM and in relation to salt stress in barley (Hordeum vulgare L.). Dissertation, Technische Hochschule Darmstadt, Germany. Mariaux, J.-B., Becker, A , , Kemna, I . , Ratajczak, R . , Fischer-Schliebs, E., Kramer, D., Luttge. U. and Marigo, G. (1994). Visualization by freeze-fracture electron microscopy of intramembraneous particles corresponding to the tonoplast H+-pyrophosphatase and H+-ATPase of Kalanchoe daigremontiana Hamet et Perrier de la Bithie. Botanica Acta 107, 321-327. Marin, B. P. (1985). “Biochemistry and Function o f Vacuolar Adenosine-Triphosphatase in Fungi and Plants.” Springer-Verlag, Berlin. Markham, R . , Frey, S. and Hill, S . J. (1963). Methods for the enhancement of image detail and accentuation of structure in electron microscopy. Virology 20, 88-102. Marquardt-Jarczyk. G. and Luttge, U . (19Y0a). PP,ase-activated ATP-dependent H t transport at the tonoplast of mesophyll cells of the CAM plant Kalanchoe daigremontiana. Botanica Acta 103, 203-21 3. Marquardt-Jarczyk, G. and Luttge, U. ( 199Ob). Different valinomycin effects on tonoplast ATPase and PPiase of the CAM plant Kalanchoe daigremontiana measured by formation of H + gradients and membrane potentials. Comptes Rendus A c n d h i e des Sciences Paris 310 (Serie HI), 311-316. Martinoia, E., Grill. E., Tommasini, R.. Kreuz, K . and Amrhein, N. (1993). ATP-dependent glutathione S-conjugate ‘export’ pump in the vacuolar membrane of plants. Nature 364, 247-249. Martiny-Baron, G., Manolson, M. F., Poole. R . J . , Hecker, D. and Scherer, G. F. E. (1992). Proton transport and phosphorylation of tonoplast polypeptides from zucchini are stimulated by the phospholipid platelet-activating factor. Plant Physiology 99, 1635-1641. Matsuura-Endo, C., Maeshima, M. and Yoshida, S. (1090). Subunit composition of the vacuolar membrane H+-ATPase from mung bean. European Journal of Biochemistry 187, 745-751. Matsuura-Endo, C.. Maeshima, M. and Yoshida, S. (1992). Mechanism of the decline in vacuolar H+-ATPase activity in mung bean hypocotyls during chilling. Plant Physiology 100, 718-722. Mellmann, I., Fuchs. R . and Helenius, A . (1Y8h). Acidification of the endocytic and exocytic pathways. Annual Reviews of Biochemistry 55, 663-700. Mito, N., Kimura, T. and Asahi, T. (1988). Partial purification and characterization of an ATPase in mung bean hypocotyl plasma membrane: suggestion for a new

292


type of higher plant plasma membrane ATPase. Plant Cell Physiology 29, 875-882. Moriyama, Y. and Nelson, N. (1989). Cold inactivation of vacuolar proton-ATPases. Journal of Biological Chemistry 264, 3577-3582. Moriyama, Y., Yamamoto, A., Tashiro, Y. and Futai, M. (1991). Chromaffin granule H+-ATPase has F,-like structure. FEBS Letters 291, 92-96. Morre, D. J., Liedtke, C., Brightman, A. 0. and Scherer, G. F. E . (1991). Head and stalk structures of soybean vacuolar membranes. Planta 184, 343-349. Miiller, M. L., Irkens-Kiesecker, U., Rubinstein, B. and Taiz, L. (1996). On the mechanism of hyperacidification in lemon. Journal of Biological Chemistry 271, 1916-1924. Narasimhan, M. L., Binzel, M. L., Perez-Prat, E., Chen, Z., Nelson, D. E., Singh, N. K., Bressan, R. B. and Hasegawa, P. M. (1991). NaCl regulation of tonoplast ATPase 70-kilo-dalton subunit mRNA in tobacco cells. Plant Physiology 97, 562-568. Nelson, N. (1992). Evolution of organellar proton-ATPases. Biochimica et Biophysica Acta 1100, 109-124. Nelson, N. and Taiz, L. (1989). The evolution of H+-ATPases. Trends in Biochemical Sciences 14, 113-116. Nishimura, M. and Beevers, H. (1979). Hydrolysis of protein in vacuoles isolated from higher plant tissue. Nature 277, 412-413. Nolta, K. V., Padh, H. and Steck, T. L. (1991). Acidosomes from Dictyostelium initial biochemical characterization. Journal of Biological Chemistry 266, 18 31818 323. Oberbeck, K., Drucker, M. and Robinson, D. G. (1994). V-ATPase and pyrophosphatase in endomembranes of maize roots. Journal of Experimental Botany 45, 235-244. Parry, R. V., Turner, J. C. and Rea, P. A. (1989). High purity preparations of higher plant vacuolar H+-ATPase reveals additional subunits. Journal of Biological Chemistry 264, 20 025-20 032. Perera, I., Li, X. and Sze, H. (1995). Several genes encode nearly identical 16 kDa proteolipids of the vacuolar H+-ATPase from Arabidopsis thaliana. Plant Molecular Biology 29, 227-244. Pfeiffer, W. (1995). Effects of W-7, W-5, Verpamil and Diltiazem on vacuolar proton transport. Comparison of vacuolar H+-ATPase and H+-PPase from roots of Zea mays. Physiologia Plantarum 94, 284-290. Pfeiffer, W. and Hager, A. (1993). A Ca2+-ATPase and a Mg2+/H+-antiporter are present on the tonoplast membrane from roots of Zea mays. Planta 191, 377-385. Quail, P. H. (1979). Plant cell fractionation. Annual Reviews of Plant Physiology 30, 425-484. Randall, S. K. and Sze, H. (1986). Properties of the partially purified tonoplast H+-pumping ATPase from oat roots. Journal of Biological Chemistry 261, 1364-1371. Randall, S. K. and Sze, H. (1987). Probing the catalytic subunit of the tonoplast H+-ATPase from oat roots. Journal of Biological Chemistry 262, 7135-7141. Ratajczak, R. (1994). The non-ionic detergent Brij 58 conserves the structure of the tonoplast H+-ATPase of Mesembryanthemum crystallinum during solubilization and partial purification. Botanica Acta 107, 201-209. Ratajczak, R., Richter, J. and Liittge, U. (1994). Adaptation of the tonoplast V-type H+-ATPase of Mesembryanthemum crystallinum to salt stress, C3-CAM transition and plant age. Plant, Cell and Environment 17, 1101-1112.


293

Ratajczak, R . , Hille, A . , Mariaux, J.-B. and Luttge, U. (1995). Quantitative stress responses of the VoV1-ATPase of higher plants detected by immuno-electron microscopy. Botanica Acta 108, 505-5 13. Rausch, T . , Kirsch, M., Low, R., Lehr, A , , Viereck, R . and Zhigang, A . (1996). Salt stress responses of higher plants: the role of proton pumps and Na+/H+antiporters. Journal of Plant Physiology 148, 425-433. Rea, P. A , , Griffith, C. J. and Sanders, D. (1987). Irreversible inhibition of H+-ATPase o f higher plant tonoplast by chaotropic anions: evidence for peripheral location of nucleotide binding subunits. Biochimica et Biophysicu Acta 904, 1-12. Reuveni, M. (1992). Utilization of metabolic energy under saline conditions: changes in properties of ATP dependent enzymes in plant cells grown under saline conditions. Biologia Plantarum 34, 181-191. Reuveni, M., Bennett, A . B., Bressan, R . A . and Hasegawa, P. M. (1990). Enhanced H+-transport capacity and ATP hydrolysis activity of the tonoplast H+-ATPase after NaCl adaptation. Plant Physiology 94, 524-530. Robinson, D. G., Haschke, H.-P., Hinz, G . , Hoh. B., Maeshima, M. and Marty, F. (1996). Immunological detection of tonoplast polypeptides in the plasma membrane of pea cotyledons. Planra 198, 95-103. Rockel, B . (1996). Genexpression der Untereinheiten A , B und c der TonoplastenH+-ATPase von Mesembryanthernurn crystallinum L. im diurnalen Rhythmus und unter Salzstress. PhD thesis, Technische Hochschule Darmstadt. Germany. Rockel, B . , Ratajczak, R.. Becker, A . and Luttge, U . (1994). Changed densities and diameters of intra-membrane tonoplast particles of Mesembryanthemum crystallinum in correlation with NaC1-induced CAM. Journal o f Plant Physiology 143, 318-324. Salt, D. E. and Rauser, W. E. (1995). MgATP-dependent transport of phytochelatins across the tonoplast of oat roots. Plant Physiology 107, 1293-1301. Sarafian, V., Potier, M . and Poole, R . J . (1992). Radiation-inactivation analysis of vacuolar H+-ATPase and H+-pyrophosphatase from Beta vulgaris L. Biochemical Journal 283, 493-497. Scherer, G . F. E. and Stoffel, B. (1987). A plant lipid and platelet activating factor stimulate ATP-dependent H+-transport in isolated plant membrane vesicles. Planra 172. 127-130. Scherer, G . F. E.. Martiny-Baron, G . and Stoffel, B. (1988). A new set of regulatory molecules in plants: a plant phospholipid similar to platelet-activating factor stimulates protein kinase and proton-translocating ATPase in membrane vesicles. Planta 175, 241-253. Schmidt, W. and Winkler, H. (1982). Adrenal chromaffin granules. Evidence for an ultrastructural equivalent of the proton-pumping ATPase. European Journal of Cellular Biology 27, 9 6 1 0 4 . Sen, S. and Sharma. V. (1994). Tonoplast ATPase from peanut seedlings. Phytochemistry 36, 569-572. Serrano, R. (1990). Plasma membrane ATPase. I n “The Plant Plasma Membrane” (C. Larsson and I . M. Mdler, eds). pp. 127-153. Springer-Verlag, Berlin. Shono, M.. Wada, M. and Fujii, T. (1995). Partial purification of a Na+-ATPase from the plasma membrane of the marine alga Heterostigma akashiwo. Plant Phy.yiology 108, 1615-1621. Smith, J . A . C., Marigo, G . , Luttge, U . and Ball, E. (1982). Adenine-nucleotide levels during crassulacean acid metabolism and the energetics of malate accumulation in Kalanchoe tubiflora. Plant Science Letters 26, 13-21.

294

U. L U T G E and R. RATAJCZAK

Smith, J. A. C., Uribe, E. G., Ball, E. and Luttge, U. (1984a). ATPase activity associated with isolated vacuoles of the crassulacean acid metabolism plant Kalanchoe daigremontiana. Planta 162, 299-304. Smith, J. A. C., Uribe, E. G., Ball, E., Heuer, S. and Liittge, U. (1984b). Characterization of the vacuolar ATPase activity of the crassulacean-acidmetabolism plant Kalanchoe daigremontiana. European Journal of Biochemistry 141, 415-420. Struve, I . and Liittge, U. (1987). Characteristics of MgATP2--dependent electrogenic proton transport in tonoplast vesicles of the facultative crassulacean-acidmetabolism plant Mesembryanthemum crystallinum L. Planta 170, 111-120. Struve, I. and Liittge, U. (1988). Biochemical and immunological properties of solubilized tonoplast ATPase of the facultative CAM plant Mesembryanthemum crystallinum in the C3- and CAM state. Botanica Acta 101, 39-44. Struve, I., Weber, A , , Liittge, U., Ball, E. and Smith, J. A. C. (1985). Increased vacuolar ATPase activity correlated with CAM induction in Mesembryanthemum crystallinum and Kalanchoe blossfeldiana cv. Tom Thumb. Journal of Plant Physiology 117, 451-468. Sze, H. (1985). H+-translocating ATPases: advances using membrane vesicles. Annual Reviews of Plant Physiology 36, 175-208. Sze, H., Ward, J . M. and Lai, S. (1992a). Vacuolar H+-translocating ATPases from plants: structure, function and isoforms. Journal of Bioenergetics and Biomembranes 24, 371-381. Sze, H., Ward, J . M., Lai, S. and Perera, I. (1992b). Vacuolar-type H+-ATPases in plant endomembranes: subunit organization and multigene families. Journal of Experimental Biology 172, 123-135. Takeshige, K., Tazawa, M., Hager, A. (1988). Characterization of the H+translocating adenosine triphosphatase and pyrophosphatase of vacuolar membranes isolated by means of a perfusion technique from Chara corallina. Plant Physiology 86, 1168-1173. Takeshige, K., Mitsumori, F., Tazawa, M. and Mimura, T. (1992). Role of cytoplasmic inorganic phosphate in light-induced activation of H+-pumps in the plasma membrane and tonoplast of Chara corallina. PIanta 186, 466-472. Tazawa, M., Okazaki, Y . , Moriyama, Y. and Iwasaki, N. (1995). Concanamycin 4-B: a potent inhibitor of vacuolar pH regulation in Chara cells. Botanica Acta 108, 67-73. Tsiantis, M. S . , Bartholomew, D. M. and Smith, J . A. C. (1996). Salt regulation of transcript levels of a leaf vacuolar H+-ATPase in the halophyte Mesembryanthemum crystallinum. Plant Journal 9, 729-736. Tu, S. I., Patterson, D., Brauer, D. and Hsu, A . F. (1995). Inhibition of corn root membrane ATPase activities by oryzaline. Plant Physiology and Biochemistry 33, 141-148. Tzeng, C. M., Hsu, L. H. and Pan, R . L. (1992). Inhibition of tonoplast ATPase from etiolated mung bean seedlings by fluoresceine 5'4sothiocyanate. Biochemical Journal 285, 737-743. Viereck, R . , Kirsch, M., Low, R. and Rausch, T. (1996). Down-regulation of plant V-type H+-ATPase genes after light-induced inhibition of growth. FEBS Letters 384, 285-288. Vincente, J . A. F. and Vale, M. G. P. (1994). Proton transport by a fraction of endoplasmic reticulum and Golgi membranes of corn roots - comparison with the plasma membrane and tonoplast H+-pumps. Plant Science 96, 55-68. Wan, C. Y. and Wilkins, T. A. (1994). Isolation of multiple cDNAs encoding the vacuolar H+-ATPase subunit B from developing cotton (Gossypium hirsutum)


295

ovules. Plant Physiology 106, 393-394. Wang, Y. and Sze. H . (1985). Similarities and differences between the tonoplast-type and the mitochondria1 H+-ATPases of oat roots. Journal of Biological Chemistry 260, 10 434-10 443. Ward, J . M. and Sze. H. (1992a). Subunit composition and organization of the vacuolar H+-ATPase from oat roots. Plant Physiology 99, 170-179. Ward, J . M. and Sze, H . (1992b). Proton transport activity of the purified vacuolar H+-ATPase from oats. Direct stimulation by CI-. Plunf Physiology 99, 925-93 1 . Ward, J. M., Reinders, A . , Hsu, H.-T. and Sze, H . (1992). Dissociation and reassembly of the vacuolar H+-ATPase complex from oat roots. Plant Physiology 99, 161-169. Warren, M., Smith, J . A . C. and Apps, D. K. (1992). Rapid purification and reconstitution of a plant vacuolar ATPase using Triton X-114 fractionation: subunit composition and substrate kinetics of the H+-ATPase from the tonoplast of Kalanchoe daigremontiana. Biochimica et Biophysica Acto 1106, 117-125. Weiser, T. and Bentrup, F.-W. (1994). The chaotropic anions thiocyanate and nitrate inhibit the electric current through the tonoplast ATPase of isolated vacuoles from suspension cells of Clienopociium rubrum. Physiologia Plantarum 91, 17-22. White, P. J . (1904). Bafilomycin A , is a non-competitive inhibitor of the tonoplast H+-ATPase of maize coleoptiles. Journul of Experimental Botany 45, 13971402. Wilkins, T. A . (1993). Vacuolar H+-ATPase 69-kilodalton catalytic subunit cDNA from developing cotton (Gossypium hirsutum) ovules. Plant Physiology 102, 67Y-680. Williams, L. E., Nelson, S. J . and Hall, J . L. (1990). Characterization of solute transport in plasma membrane vesicles isolated from cotyledons of Ricinus communis L. Planta 182, 532-539. Willmer, C. M . , Grammatikopoulos, G . , Lasceve, G . and Vavasseur, A . (1995). Characterization of the vacuolar-type H+-ATPase from guard cell protoplasts of Commelina. Journal of Experimental Botany 46, 383-389. Winter, K. and Gademann, R. (1991). Daily changes in COz- and water vapour exchange, chlorophyll fluorescence, and leaf water relations in the halophyte Mesernbryanthemurn crystallinirm during the induction of crassulacean acid metabolism in response to high NaCI-salinity. Plant Physiology 95, 76g-776. Yamanishi, H. and Kasamo, K . (1992). Binding o f 7-chloro-4-nitrobenzo-2oxa-l,3diazole to an essential cysteine residue(s) in the tonoplast H+-ATPase from mung bean (Vigna radiata L.) hypocotyls. Plant Physiology 99, 652-658. Yamanishi, H. and Kasamo, K. (1993). Modulation of the activity of purified tonoplast H+-ATPase from mung bean (Vigna radiata L.) hypocotyls by various lipids. Plant Cell Physiology 34. 41 1-41C). Yoshida, S . (19Y 1). Chilling-induced inactivation and its recovery of tonoplast H+-ATPase in mung bean cell suspension cultures. Plant Physiology 95, 456-460. Yoshida. S . and Matsuura-Endo. C. (199 1). Comparison of temperature dependency of tonoplast proton translocation between plants sensitive and insensitive to chilling. Plunt Physiology 95, 504-508. Yoshida, S . , Hattanda, Y. and Suyama, 1'. (1993). Variations in chilling sensitivity of suspension-cultured cells of mung bean (Vigna radiata [L.] Wilczek) during the growth cycle. Plant Cell Physiology 34, 673-679.

296


Zhang, J . M., Feng, Y. and Forgac, M. (1994). Proton conduction and bafilomycin binding by the V-0 domain of the coated vesicle V-ATPase. Journal of Biological Chemistry 269, 23 518-23 523. Zhigang, A., Low, R., Rausch, T., Liittge, U. and Ratajczak, R. (1996). The 32 kDa tonoplast polypeptide Di associated with the V-type H+-ATPase of Mesembryanthemum crystallinurn L. in the CAM state: a proteolytically processed subunit B? FEBS Letters 389, 314-318. Zimniak, L., Dittrich, P., Gogarten, J. P., Kibak, H. and Taiz, L. (1988). The cDNA sequence of the 69 kDa-subunit of the carrot ATPase. Journal of Biological Chembtry 263, 9102-9112. Zocchi, G. (1985). Phosphorylation-dephosphorylationof membrane proteins controls the microsomal Hf-ATPase activity of corn roots. Plant Science 40, 153-159.

The Molecular and Biochemical Basis of Pyrophosphate-Energized Proton Translocation at the Vacuolar Membrane

R.-G. ZHEN, E. J. KIM and P. A. REA

Plant Science Institute, Department of Biology, University of Pennsylvania, Philadelphia, PA 19104-6018, USA

.............................................................................

298

11.

Reaction Mechanism ................................................................. A. 1,l-Diphosphonates as Type-specific Inhibitors ........................ B. Cautionary Note Concerning In Vivo Studies .......................... C. Steady State Kinetics of Substrate Hydrolysis ......................... D. Oxygen Exchange Reactions ....................

299 299 300 302

111.

Molecular Identity and Sequence .... A. Identification of the Catalytic Subunit ................................... B. Molecular Cloning of cDNAs Encoding V-PPase ..................... C. Isoforms of the Substrate-binding Subunit ..............................

1.

IV.

Introduction

Structure-function Relations ...................................... A. One Polypeptide is Sufficient for Pump Function .... B . AVP Does Not Functionally Complement Yeast V-ATPase

..........................................................

D. E. F. V.

307 308 310

312

Revised Topological Model ............................................ Identification of Substrate-protectable Maleimide-reactive Cysteine Residue ...................................... Potential Coupling Sites .........................

Future Research ....................................................................... Acknowledgements ......... ..................................... References .......................................................

Advances in Botanical Research Vol 25 incorporating Advances in Plant Pathology ISBN 0-12-oOSY25-8

329 331 332

Copyright 0 lW7 Academic Press Limited All rights of reproduction in any form reserved

298

R . 4 . ZHEN et al.

I. INTRODUCTION Inorganic pyrophosphatases (PPases) (EC 3.6.1.1) are ubiquitous enzymes that catalyse the hydrolysis of pyrophosphate (diphosphate, PPi) to orthophosphate (Pi). In most organisms, the bulk of PPase activity is soluble and considered to maximize the energy yield from the pyrophosphorylytic cleavage of nucleoside triphosphates, associated with many biosynthetic reactions, by dissipatively hydrolysing PPi. Soluble PPases have been almost exclusively investigated in the context of driving biosynthetic reactions that would otherwise be thermodynamically unfavourable (Voet and Voet, 1994). It is now clear, however, that some organisms contain membrane-associated PPases which, rather than simply dissipating the energy liberated upon PPj hydrolysis as heat, conserve some of the phosphoanhydride bond energy as transmembrane ion gradients (Baltscheffsky and Baltscheffsky, 1993; Rea and Poole, 1993). Examples are the reversible H+-translocating PPase (H+-PPi synthase) of chromatophores from the purple, non-sulfur bacterium Rhodospiriffum rubrum (Nyren et af., 1991), the membrane-linked PPases of animal and yeast mitochondria (Lundin et af., 1991) and the vacuolar H+-PPase (V-PPase) of plants (Rea et af., 1992c; Rea and Poole, 1993). The capacity of the mitochondria1 PPase for ion transport has yet to be demonstrated directly, but the enzymes from R . rubrum and plants are PPi-energized pumps active in the primary translocation of H+ (Baltscheffsky and Baltscheffsky, 1993) and H+ and/or K + , respectively (Davies e t a f . , 1992; Rea and Poole, 1993). One of these pumps, the V-PPase, will be considered here. Several lines of inquiry implicate PPi as an important energy source for plant vacuolar function in particular, and plant metabolism in general. First, the V-PPase is an abundant and universal component of plant vacuolar membranes, capable of generating a transtonoplast H+-electrochemical potential difference (A&+) of similar or greater magnitude than the vacuolar H+-ATPase (V-ATPase) (EC 3.6.1.3) on the same membrane (Pope and Leigh, 1987; Johannes and Felle, 1990; Maeshima and Yoshida, 1989; Rea et al., 1992a). Second, the vacuolar membrane is known to participate directly in a diversity of physiological processes, many of which depend on a transtonoplast A&+. Included amongst these are cytosolic pH stasis, compartmentation of regulatory Ca2+, sequestration of otherwise toxic substances such as secondary products, heavy metals and Na+, turgor regulation and nutrient storage and retrieval. Third, the V-PPase is probably the sole enzyme responsible for the direct hydrolysis of cytosolic PPi to orthophosphate (Weiner et al., 1987). When account is taken of the volume occupied by the plant vacuole - 40-90% of the total intracellular volume of a typical mature cell - it is apparent that the cytosol is often a thin interfacial zone delimited and regulated by the plasma membrane and tonoplast. Given that PPi is an energy source for a number of other metabolic reactions in

T H E VACUOLAR H+-PPase

299

plant cells, specifically cytosolic sucrose mobilization via sucrose synthase (Huber and Akazawa, 1986; Black et al., 1987) and glycolysis via PP,:fructose-&phosphate l-phosphotransferase (Black et al., 1987; Dennis and Greyson, 1987), the V-PPase has the potential for contributing to the regulation of carbon metabolism. Although significant progress in understanding the V-PPase at several levels has been accomplished during the last few years, the treatment here will be restricted to the identification of type-specific inhibitors of the enzyme, kinetic analyses of its substrate hydrolytic cycle and advances in the development of procedures for studying its structure-function relations. Particular emphasis will be placed on protein chemical and mutational analyses of the enzyme. The application of heterologous expression and site-directed mutagenesis studies of the functional pump will be discussed in the context of the minimum unit competent in PP,-dependent H + translocation, a revised topological model of the substrate-binding subunit* and the participation of specific amino acid residues in maleimide-mediated enzyme inactivation and PP,-energized transmembrane H + transport. Readers interested in evolutionary considerations or investigations of the physiological functions of the pump are referred to Rea and Poole (1993) and Davies (this volume).

11. A.

REACTION MECHANISM

1,l-DIPHOSPHONATES AS TYPE-SPECIFIC INHIBITORS

Most of the H + phosphohydrolases associated with plant membranes can be distinguished from each other by their inhibitor sensitivities. The F-type H+-ATPases of energy-coupling membranes, the P-type H+-ATPase of plasma membrane and the V-type Hf-ATPase of endomembranes are selectively inhibited by azide, orthovanadate and the bafilomycins, respectively. However, strict criteria for the identification of V-PPase activity in uncharacterized membrane fractions. other than through the measurement of azide-, orthovanadate- and bafilomycin-insensitive, K+-stimulated PPidependent Hf translocation (Rea and Turner, 1990), have been lacking. This deficiency has now been remedied by recognition of 1,l-diphosphonates containing a heteroatom (NH2 or OH) on the bridge carbon as exquisitely *The substrate-binding subunit of the V-PPasc migrates at M , = 66 000 (64 5 W 7 2 000) o n sodium dodecyl sulfate gels but the probable mass, computed from its deduced amino acid sequence. is 79-81 kDa. Thus. when referring to this polypeptide, thc terms “Mr 66 000 subunit” and “80 kDa subunit” are used interchangeably. Moreover, although it is now known that this subunit is probably the sole polypeptide species comprising the enzyme, it is referred to as the “substrate-binding subunit” because substratc-protectablc covalent modification was the first function t o be unequivocally assigned to it.

300

R.-G. ZHEN et al.

potent competitive inhibitors of the V-PPase (Baykov et al., 1993b; Zhen el al., 1994a). When screened at PPi concentrations corresponding to the K , of the enzyme, the apparent inhibition constants (KiaPP values) for PPi analogues follow the sequence dichloromethylenediphosphonate >> methylenediphosphonate > imidodiphosphate > hydroxymethylenediphosphonate > aminomethylenediphosphonate (AMDP) (Table I). Moreover, when measured at a PPi concentration (300 pM) corresponding to 10 times the K, - the usual concentration employed to assay the enzyme in vitro (e.g. Rea and Turner, 1990) and the concentration prevailing in the cytosol of plant cells (Weiner et al., 1987; Takeshige and Tazawa, 1989) - the most efficacious inhibitor, AMDP, is both potent and type-specific (Table I). It is 6-38-fold more effective than methylenediphosphonate and imidodiphosphate, the two PPi analogues most commonly used for investigations of the V-PPase (e.g. Chanson and Pilet, 1987), and is completely ineffective against other plant ion translocases (V-ATPase and P-type H+-ATPase) and soluble PPi hydrolases (soluble PPase, alkaline phosphatase and non-specific monophosphoesterase). Two factors are responsible for the potency and selectivity of AMDP: (a) the 6-12-fold greater intrinsic sensitivity of the V-PPase to AMDP by comparison with other PPases; and (b) The K , of the V-PPase for PPi. With the exception of the bacterial H+-PPi synthase (which is probably a V-PPase homologue; see Section III.B), all characterized PPases have K , values that are at least an order of magnitude smaller than that of the V-PPase (Zhen et af., 1994a). Since, for a competitive inhibitor, v = (V,,,[S])/(K, (1 + [I]/Ki}+ [S]), it follows that when K , = 1pM (e.g. in the case of a soluble PPase) and Ki = 2 pM (e.g. AMDP), 20 pM inhibitor (I) will inhibit activity ( v ) by only 3% when [PPi] = 300 pM. By contrast, if K, = 30 pM (e.g. in the case of the V-PPase), a similar concentration of I will cause approximately 50% inhibition under identical conditions. AMDP is unique among the compounds screened because it contains a positively charged amino group at pH 8.0. Its activity towards the V-PPase may denote a carboxylate group close to the substrate-binding site which stabilizes the enzyme-inhibitor complex by binding the protonated amino group of AMDP (Baykov et al., 1993b).

B. CAUTIONARY NOTE CONCERNING IN V N O STUDIES

The results of these in vitro studies demonstrate the utility of AMDP as a diagnostic probe for the preliminary identification of V-PPase activity, but extreme caution is warranted before application of this or similar compounds to in vivo studies. The suitability of 1,l-diphosphonates and other PPi analogues as V-PPase-specific inhibitors in vivo will be critically dependent

TABLE I Apparent inhibition constants ( K r P P values, p M ) for a series of pyrophosphate analogues with the structure 03P-R-PO3. Values in parentheses are percentage inhibitions exerted by 20 p M of the analogue in standard V-PPase assay medium containing 300 p M PP, and 1.3 m M M$+. V-PPase, H+-PP, synthase and soluble PPase activities were assayed using vacuolar membrane vesicles purified from Vigna radiata, chomatophores purified from R. rubrum and enzyme purified f r o m Saccharomyces cerevisiae, respectively. K,app

Compound

R

Methylenediphosphonate -CH2Aminomethylenediphosphonate XH(NH2)Hydroxymethylenediphosphona te -CH(OH)Ethane-1-hydroxy-1,l-diphosphonate -C(CH3)0H-C(Cbt Dichlorome thylenediphosphonate Imidodiphosphate -N HPyrophosphateb -0-

(YOinhibition)

V-PPase

Hf-PPi synthase

Soluble PPasea

68.0 (25) 1.8 (88) 5.7 6.5 >500 12.0 (67) 30.0

41.0 (53) 1.2 (91) 1.9 25.0 >loo0

1100 (0)

aValues from Smirnova et al. (1988). bApparent Michaelis constant for total PPi in the presence of 1 mM Mg2+.

30.0 (74) 36.0

20 (-3) 50

>3000 >3000 15.0 (-4) 0.9

302

R.-G. ZHEN

ei

al.

on the distribution of PP, and target PP, hydrolases between compartments within the cell. Neuhaus and Stitt (1991), for instance, describe the administration of imidodiphosphate to detached leaves via the transpiration stream and show that this is accompanied by the inhibition of sucrose synthesis concomitant with the depletion of cellular UDP-glucose and accumulation of PP,, hexose phosphates and fructose 1,6-bisphosphate. O n the basis of these findings and the observation that soluble PPase activity is almost exclusively located in the chloroplast stroma of photosynthetic tissues (Weiner et al., 1987), the V-PPase has been deduced to be the principal enzyme responsible for the disposal of the PP, generated by UDP-glucose pyrophosphorylase. By inference, the inhibition of sucrose synthesis and the elevation of cytosolic PP, seen during the administration of imidodiphosphate has been attributed to selective inhibition of the V-PPase and retardation of the UDP-glucose pyrophosphorylase reaction through end-product (PP,) accumulation (Quick et al., 1989; Neuhaus and Stitt, 1991). However, the validity of this conclusion is contingent on the relative contributions made by the cytosol, itself, and the chloroplast stroma to total cellular PP, turnover and the PP, concentrations prevailing in these compartments. Specifically, if stromal PP, concentrations are less than 1 pM, as indicated by the results of intracellular perfusion studies of Cham coralfinu (Takeshige and Tazawa, 1989), and the bulk of soluble PPase activity is restricted to the chloroplast stroma (Weiner et al., 1987), imidodiphosphate would be expected to not only inhibit the V-PPase but also the stromal PPase (Zhen et al., 1994a).

C. STEADY STATE KINETICS OF SUBSTRATE HYDROLYSIS

It is established that the V-PPase specifically acts on PP, and has a strict requirement for Mg2+ for activity (Walker and Leigh, 1981; Rea and Poole, 1993) but the mode of action of Mg2+ as activator and its state of association with PP, has been difficult to define. Thus, magnesium pyrophosphate (MgPP,) (Wang et al., 1986; Johannes and Felle, 1989; White et al., 1990), both MgPP, and dimagnesium pyrophosphate (Mg2PP,) (Walker and Leigh, 1981) and MgzPP,, alone (Leigh et al., 1992) have been concluded to be the active substrate species, while activation by Mg2+ has been attributed to cooperative (White et al., 1990) or non-cooperative binding of this divalent cation (Leigh et al., 1992). A major complication when attempting to determine the ligand interactions of any PPase, whether soluble or membrane bound, are the multiple ion complexes that PP, is capable of forming in aqueous media. In a reaction medium containing PP,, Mg2+ and K + , the complexes and ions present include free PP,, free Mg2+, free K', MgPP,, Mg2PP, and KPP,, as well as their intermediate association states and protonated forms. Since the

THE VACUOLAR H+-PPase

303

Fig. 1. Minimal reaction scheme for steady state kinetics of substrate hydrolysis by V-PPase of V. rudiuta. Dimagnesium pyrophosphate (Mg2PP,)is assumed to be the active substrate species. The complex E(Mg2PP,)is shown in braces to signify that it is not on the main K’dependent pathway. K M i and K M 2 are dissociation constants (0.26 f 0.02 and 0.1 k 0.01 mM, respectively), K , is the Michaelis constant (4.9 f 0.3 p M ) and V is the maximal velocity. E, enzyme; Mg, magnesium ion; PP,, pyrophosphate. This reaction scheme is approximated by a rate equation of the form: v = v/(1 + KM2/[Mg2+]+ K,(1 + K~,/[Mg”]/[Mg2pP,l)}.

concentrations of these ions and their complexes vary interdependently and any one of these species may function as substrate or cofactor, analyses of the substrate kinetics of PPases necessitate precise estimates of the pertinent stability constants under conditions identical to those employed for the kinetic measurements. For this reason, and the lack of binding data for the K-PP; system at 25°C and 0.1 M ionic strength (standard conditions for the assay of V-PPase activity), the appropriate parameters have been estimated potentiometrically (Baykov et af., l993a). When the relevant stability constants are enumerated and ionic strength is constrained at 0.1 M, the Mg’+-dependence of PP, hydrolysis is adequately approximated by a minimal reaction scheme containing three enzyme species: E, EMg and EMg(Mg2PPj) (Fig. 1). Since Mg2PPjis deduced to be the active substrate species and activation requires the binding of an additional Mg2+, three Mg’+ ions are considered to participate directly in catalysis. Positively cooperative binding of Mg2+ (White et d., 1990), competitive inhibition by free PP; and/or non-competitive inhibition by high Mg2PPj concentrations (Walker and Leigh, 1981; Johannes and Felle, 1989; Leigh et al., 1992) need not be invoked to explain the steady state kinetics of the enzyme from I/. rudiata. Formation of a ternary complex rather than a ping-pong mechanism involving formation of a phosphoryl intermediate is depicted in Fig. 1 because all attempts to autophosphorylate the substrate-binding subunit by the provision of [32P]PPj have failed (C. J . Britten and P. A . Rea, unpublished findings). The three-state model does not automatically preclude all other schemes.

304

R.-G. ZHEN et al.

For instance, formation of E(Mg2PPi) from the combination of EMg and EMgPPi cannot be excluded. Also, although the rate data are consistent with Mg2PPi being the active substrate species, EMg(Mg2PPi) may arise from the ordered addition of MgPPi and Mg2+, such that MgPPi is characterized by a sufficiently high Michaelis constant that the EMg(Mg2PPi) species is not stoichiometrically significant. Ascription of the active substrate species to Mg2PPj, rather than MgPPi, is primarily based on the comparative simplicity of the resulting scheme. Notable, however, is the basic operational equivalence between this scheme and that proposed by Leigh et al. (1992) in that catalysis entails the direct (non-cooperative) participation of three Mg2+ ions per PPi hydrolytic cycle. Distinct binding sites for K + and Mg2+ are implicated (Baykov et al., 1993a). The steady state kinetics of PPi hydrolysis and the kinetics of inactivation of the V-PPase by N-ethylmaleimide (NEM, see Section 1II.A) indicate that Mg2+ and K+ exert their stimulatory effects independently. Unlike K+, which determines maximal velocity (Fig. l), the primary effect of Mg2+ is to diminish K,, and this is independent of the nature of the monovalent cation. The dissociation constant for K+ binding, estimated from steady state substrate hydrolysis measurements, is not affected by the concentration of free Mg2+. Potassium and Mg2+ have opposite effects on the kinetics of inactivation by NEM: K+ promotes inactivation and Mg2+ retards inactivation, and neither cation interferes with the action of the other. The exact number of Mg2+-binding sites on the V-PPase is unclear. Whereas measurements of the effect of free Mg2+ on substrate hydrolysis are explicable in terms of a low-affinity binding site (KM1= 0.25-0.46 mM), estimates of the dissociation constant for Mg2+ binding, deduced from NEM inactivation experiments, reveal a high-affinity site (KM= 23-31 pM). Analogously, Gordon-Weeks el a!. (1996) have recently shown that protection from inhibition by the water-soluble carboxyl-selective reagent l-ethyl-3-(3-dimethylaminopropyl)carbodiimide(EDC), exhibits an Mg2+ concentration dependence consistent with a K, for free Mg2+ of 20-23 pM. Further studies are required to determine the mechanistic significance of this finding, but it may be that the V-PPase, like yeast soluble PPase (Cooperman, 1982), contains two distinct Mg2+ binding sites in addition to the Mg2PPi-binding site. In the case of the V-PPase, these are a high-affinity, NEWEDC-modifiable site and a low-affinity, NEWEDC-unreactive site.

D. OXYGEN EXCHANGE REACTIONS

Steady state measurements of substrate hydrolysis give insight into the stoichiometry of reactants and products but not their pathways of interconversion. However, by providing information on the covalent bond

305


formation step of phosphoryl transfer, oxygen exchange measurements can partially alleviate this shortfall (Hackney et al., 1980; Stempel and Boyer, 1986). In all cases where the rate of oxygen exchange has been compared with the rate at which substrate enters the cycle, exchange has been found to result from a reaction of the type R-P03 + H180H RH + H180P03. Hence, mass spectral analysis of the incorporation of l80into medium Pi provides an estimate of the rate of covalent bond formation-cleavage at the active site and measurements of the distribution of the Pi species containing "0 atoms enables enumeration of the partition coefficient (P,), a measure of the probability of bound Pi undergoing the reaction leading to the exclusion of a water molecule versus simple release back into the medium. Mass spectral analyses of reaction media initially containing highly enriched [ '80]P; and unlabelled water demonstrate medium P,-HOH exchange by vacuolar membrane vesicles purified from V . radiata. The data are consistent with a single reaction pathway in which exchange results from reversal of enzyme-bound PP; synthesis (Baykov et al., 1994). Four features of the exchange reaction demonstrate that the V-PPase is responsible. (i) Medium P,-HOH exchange exhibits a second-order dependence on Pi, indicating that the binding of two Pi molecules to the enzyme is required. This is consistent with the participation of PPj as an intermediate in the exchange reaction. (ii) The pronounced Mg2+ concentration dependence of the exchange reaction agrees with what is known of the substrate hydrolytic cycle. Since the hydrolysis of PP; requires formation of an MgZPP; complex and produces MgP; and Pi as products (Baykov et ~ l . 1993a), , it is reasonable that binding of both MgP; and free Pi will be required for the reaction resulting in medium P,-HOH exchange. (iii) The order of stimulation of oxygen exchange by monovalent cations (K+= Rb+ >Na+ > none) is the same as for V-PPase-mediated PP, hydrolysis and PPj-dependent H + translocation. (iv) Oxygen exchange is diminished by AMDP but not by F-type H+-ATPase, P-type H+-ATPase or soluble PPase inhibitors. By comparison with the rates of medium Pi-HOH exchange, the value of P, is relatively independent of monovalent cation, Pio r Mg2+ concentrations. The predicted distributions approximate the measured ones in all cases, and the average sum of the deviations for the five Pi species (P'xOoo,P"Ol, P"02, PI8O3, PI8O4, Pl8OS) is only 1.15%, indicating a uniform reaction scheme with a single catalytic pathway. Of the 13 models examined, Scheme 1 yields the best fit to the Pi and Mg2+ concentration-dependence data. Enzyme (E) is assumed to be active only as its Mg2+ complex (EM) (Leigh et ~ l . 1992; , Baykov et al., 1993a) and medium P,-HOH exchange is interpreted to depend on the binding of both Pi and MgP; to EM such that the reaction has a second-order dependence on Pi. In the specific case of the V-PPase from V . radiata the rate equation takes the form:

*

306

R . G . ZHEN et al. K,

E

K1

EM

F= EMP

K2

kex

EM(MP2)

It

EM(MP)2 in which [El,, [MI and [MP] are the concentrations of total enzyme, free Mg2+ and MgPi, respectively, k,, (= 16.5 +_ 1.4 nmol mg min-I) is the catalytic constant and K, (= 0.26mM), K,‘ (= 31 k 9 ) and K , (= 0.1 k 0.05 mM) are dissociation constants. Scheme 1 v,, = k,,[E],/(l

+ [MIK,’ + K,/[MP] + K1K2/[MPI2+ K1K2K,/[MPI2[M])

In principle, medium P,-HOH exchange could proceed through one of two pathways: by the formation of a phosphoryl-enzyme intermediate (E-P) or by the reversible generation of non-covalently bound PP, (E-P-P). Of these two alternatives, the latter is more probable. Not only does the V-PPase appear not to undergo autophosphorylation (see Section 1I.C) but medium P,-HOH exchange has a specific requirement for the binding of two P, molecules. This would be difficult to reconcile with E-P formation, which demands participation of only one P, molecule. The results of these studies are significant in two respects: (i) they are consistent with the reversible formation of V-PPase-bound PP, from P, without release of the bound PP, to the medium; and (ii) they imply, in accord with the results of steady state kinetic measurements, basic functional similarities between the V-PPase and soluble PPases. The finding that the P, values of the V-PPase for medium P,-HOH exchange (0.13-0.18) approximate the values for the soluble PPases from Escherichia coli and Sacchuromyces cerevisiae (Hackney, 1980; Springs et uf.,1981) indicates that the PP,-P, interconversion mechanisms of the two classes of enzyme are alike. In both cases, the rate of medium P,-HOH exchange exhibits a sharp dependence on Mg2+and P,, does not necessitate the formation of a phosphoryl-enzyme intermediate and results in a relatively invariant value for P,. The value of P, is an index of the number of P,-PP, interconversion cycles at the active site of the enzyme before the release of P, back to the bulk medium. As P, tends to zero, the rate of oxygen release approaches the rate of PP, formation so that each P, which contributes to the formation of E.P-P returns to the medium pool after having acquired only one oxygen atom from water. By contrast, as P, approaches its theoretical limit of 1, a medium P, upon reformation of enzyme-bound P, will have acquired nearly four oxygen atoms from water before returning to the medium. The equivalence of their P, values implies that the V-PPase and soluble PPases catalyse the formation of enzyme-bound PP, from P, with similar efficiencies; neither class of enzyme appears to be inherently more reversible than the other.


307

111. MOLECULAR IDENTlTY AND SEQUENCE Protein chemical identification of the substrate-binding subunit and the isolation and characterization of cDNAs encoding this polypeptide have been crucial for understanding the structure of the V-PPase and its relationship to other primary ion translocases. A.

IDENTIFICATION OF THE CATALYTIC SUBUNIT

Through the combined application of independent purification protocols, ligand-modified covalent labelling, direct sequencing and immunological techniques it is clear that the major M, 64 500-72 000 polypeptide identified in numerous V-PPase preparations corresponds to the same moiety, the substrate-binding subunit (e.g. Rea et ul., 1992a). The M, 64 500 polypeptide purified from Beta vulgaris co-migrates with the major polypeptide of the enzyme from V. rudiata, purified by an independent procedure, when the samples are subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under identical conditions, and the two polypeptides show indistinguishable patterns of ligand-modified labelling by [ I4C]NEM. The M, 64 500 and 67 000 polypeptides isolated from B. vulgaris by independent procedures (Britten rt al., 1989; Sarafian and Poole, 1989) co-migrate when electrophoresed under the same conditions and yield peptide fragments with amino acid sequences identical to each other (Rea et al.. l992a) and to the deduced translation products of cDNA clones isolated from several sources (see Section 111.B). Immunoblots of membranes prepared from Arabidopsis, 3. vulguris, V . radiata and Zea mays, probed with antibody affinity-purified against the M , 66 000 polypeptide of V. radiata, yield a single immunoreactive band migrating at M, 64 600-66 800 in all four preparations (Rea et al., 1992a). The pseudo (eukaryotic) “soluble PPase-like” subunit size of M, 37 0OWS 000 reported for the V-PPase from 2. mays and Acer pseudoplatunus (Chanson and Pilet, 1989; Fraichard et al., 1991) is an artifact arising from the combined effects of proteolysis and aggregation during sample preparation for SDS-PAGE (Rea et a f . , 1992a). The kinetics of labelling of the M, 64 500-72 000 subunit by [I4C]NEM indicate direct involvement in substrate binding (Britten et al., 1989; Rea et al., 1992a). The V-PPase activity of vacuolar membrane vesicles is subject to ligand-modified irreversible inhibition by NEM. Inhibition is pseudo-first order and quantitative protection is conferred by Mg2+ + PPi, whereas free PPj (PPi minus divalent cation) increases the potency of NEM. Treatment of vacuolar membrane vesicles with [ ‘“CINEM after pretreatment with [ ‘*C]NEM in the presence of Mg2+ + PPj - to block non-protectable, non-essential NEM-reactive groups and protect protectable groups - generates a single differentially “C-labelled polypeptide of M, 64 500 in B. vulgaris

308

R.-G. ZHEN et al.

(Britten et al., 1989; Rea et a f . , 1992a) and M , 66 000 in V. radiata (Rea et a f . , 1992a). Labelling is abolished by Mg2+ PPi and potentiated by free PPi. Because purified preparations of the V-PPase are primarily constituted of the M , 64 500-72 000 polypeptide and the same polypeptide is susceptible to (Mg2+ PPi)-protectable, free PPi-potentiated covalent modification by [I4C]NEM in vacuolar membrane vesicles, this polypeptide is deduced to contain the NEM-reactive, substrate-binding site, whose alkylation by NEM is responsible for irreversible inhibition of the enzyme. This conclusion is fully substantiated by peptide mapping and directed mutagenesis (see Section 1V.E) and the results of kinetic analyses which show that protection from NEM approximates Michaelis-Menten kinetics with respect to Mg2PPi concentration, yielding a K , value similar to that estimated from steady state substrate hydrolysis measurements (-2 pM: Gordon-Weeks et al., 1996).

+

+

B. MOLECULAR CLONING OF cDNAs ENCODING V-PPase

The V-PPase was originally cloned from Arabidopsis (Sarafian et al., 1992a). Partial cDNAs were selected by screening expression libraries with antibodies raised against the substrate-binding subunit purified from V. radiata (Maeshima and Yoshida, 1989; Rea et al., 1992a) and full-length clones were isolated by subsequent hybridization screens using the partial clones as probes (Sarafian et at., 1992a). Immunological and direct sequence data confirm the identity of the clones. The antibody used for the immunoscreens was affinity purified against the M , 66000 subunit of V. radiata. The same subunit is known to participate in substrate binding (see Section 1II.A). The deduced sequences of the polypeptide encoded by the cDNA clones show extensive identities to this subunit and the corresponding subunit from B. vulgaris (Sarafian et a f . , 1992a). Having established criteria and generated reagents for the identification of genes encoding the V-PPase, cDNAs encoding the substrate-binding subunit have been isolated from Hordeurn vulgare (Tanaka et al., 1993), B. vulgaris (Kim et a f . ,1994a) Nicotiana tubacum (Lerchl et al., 1995) and Oryza sativa (Sakakibara et a f . , 1996). In all cases, the clones encode extremely hydrophobic 760-775 amino acid (79-81 kDa) polypeptides showing greater than 85% sequence similarity (Fig. 2). A fundamental conclusion to come from the molecular cloning of V-PPase is recognition of its novelty. BLAST searches of the deduced amino acid sequences of all cloned V-PPases against GenBank release 90 (Altschul et a f . , 1990) failed to disclose any systematic sequence similarities between this pump and any other sequenced ion translocase. Likewise, phylogenic links between the V-PPase and soluble and mitochondria1 PPases seem very unlikely. All known soluble PPases have subunit sizes different from the V-PPase - 20 kDa for the enzyme from prokaryotes (Lahti et al., 1988) and

309


84.9

86.6

88.7

85.8

89.0

87 I

100.0

85.5

84.8

87.7

87.0

88.1

100.0

BVPl

87.7

87.0

89.7

87.4

100.0

BVP2

91.4

85.6

85.2

100.0

HVP

86.3

85.7

100.0

NVP

87.8

100.0

OVPl

100.0

OVP2

AVP

Fig. 2. Percentage amino acid sequence similarities among deduced sequences of polypeptides encoded by V-PPase cDNAs. The sequences shown are from A. thaliana (AVP; Sarafian et al., 1992a), B . vulgaris (putative isoforms BVPl and BVP2; Kim et a l . , 1994a). H . vulgare (HVP; Tanaka et a l . , 1993), N . tahacum (J. Lerchl, direct submission to GenBank; accession number X77915) and 0. sativa (isoforms OVPl and OVP2; Sakakibara et a l . , 1996). Similarity was computed according to Altschul et al. (1990).

32 kDa for the enzyme from eukaryotes (e.g. Kolakowski et al., 1988) - and none of the known sequences for soluble PPases align with the deduced sequence of the V-PPase (Rea et al., 1992~).The M , 32 000 catalytic subunit of the membrane-associated mitochondria1 PPase from S. cerevisiae is 49% identical to the soluble PPase from the same source (Lundin et al., 1991) and shows no sequence identities with the V-PPase (Cooperman et al., 1992; Rea et al., 1992c), and the corresponding enzyme from rat liver mitochondria does not cross-react with antibody raised against the V-PPase from V. radiata (Maeshima, 1991). The data available invoke monophylogenic origins, albeit remote, for the contemporary mitochondria1 membrane PPase and soluble PPase (Cooperman et al., 1992; Lundin et al., 1991). The V-PPase, on the other hand, appears to trace its origins to another line of descent which it may share with the reversible H+-PP, synthase of phototrophic bacteria (Rea et al., 1992~). The presence of an H+-PP, synthase o n the energy-coupling membranes of R. rubrurn has been known for some time (Baltscheffsky and Baltscheffsky, 1993), but only comparatively recently has it been shown that this translocase is an integral membrane protein with an apparent M , of between 56 000 (Nyren et al., 1991) and 70 000 (A. A. Baykov, personal communication). Two features of the M , 56000-70000 polypeptide are notable: (a) it immunoreacts with antibody raised against the substrate-binding subunit of the V-PPase purified from V. radiata (Nore et al., 1991) and KLH-conjugated synthetic peptides designed on the basis of the sequence of AVP (R.-G. Zhen and P. A . Rea, unpublished findings); and (b) unlike the M , 32 000 catalytic

310

R.-G. ZHEN

el af.

subunit of the mitochondria1 PPase, it alone is capable of mediating both PPi hydrolysis and H+ translocation (Nyren et al., 1991). In view of the extreme sensitivity of the H+-PPase from R . rubrum to inhibition by the type-specific inhibitor aminomethylenediphosphonate (Zhen et ul., 1994a; see Section 1I.A) and the strict equivalence of its PPi hydrolytic cycle with that of the V-PPase (Baykov et al., 1996), there is a strong possibility that these two pumps are evolutionarily related. C . ISOFORMS OF THE SUBSTRATE-BINDING SUBUNIT

A consequence of the isolation of cDNAs encoding the V-PPase from sources other than Arabidopsis is recognition of the existence of multiple genes encoding the substrate-binding subunit in some organisms. Analyses of multiple cDNA isolates from B. vulgaris have revealed two classes, designated BVPl and BVP2 (Kim et al., 1994a). Hybridization screens of N . tabucum have yielded three different classes of cDNA, typified by TVPS, 7VP9 and W P 3 1 (Lerchl et al., 1995). Polymerization chain reaction and hybridization screens of 0. sativa have resulted in the identification of two cDNA clones, OVPl and OVP2 (Sakakibara et al., 1996). Direct internal microsequencing of the substrate-binding subunit purified from V . radiata has disclosed two very similar but non-identical polypeptide species (Zhen et al., 1994b). In B. vulgaris, N . tabucum and 0. sativa the multiple clones appear to encode isoforms of the substrate-binding subunit. BVPZ and BVP2, for instance, specify closely related polypeptides with computed M , values of 80 550 and 80 000, respectively, exhibiting 88% identity with each other and 89% identity with AVP (Fig. 2). Their coding regions are 70% identical but their corresponding 5' and 3' non-coding regions are only 28 and 53% identical. Northern analyses of poly(A)+ RNA isolated from a range of tissue types and probed with RNAs transcribed from the 3' non-coding sequences of BVPl and BVP2 verify that both genes are expressed in the intact plant. Genomic Southern analyses indicate that two genes encode the V-PPase in this organism (Kim er al., 1994a). Similar studies of N . tabucum (Lerchl et al., 1995) and 0 . sativa (Sakakibara et al., 1996) support the same basic conclusion except that in the latter system the evidence for isoforms is categorical. OVPl and OVP2 are neither alleles nor the products of alternate splicing: RFLP mapping localizes OVPl and OVP2 to different loci on the same chromosome (chromosome 6 ) (Sakakibara et al., 1996). Virtually nothing is known of the significance of isoforms in any one of these systems. In N . tabucum a decrease in steady state transcript levels is found for all V-PPase isoforms, proceeding from sink to source leaves, but the decrease in TVP9-specific transcript is some 4-9-fold more pronounced than for W P . 5 and TVP31 (Lerchl et a l . , 1995). WP.5-specific transcripts are found at similar levels in midribs, stems and roots whereas TvP9- and


31 1

WP3Z-specific transcripts are much more abundant in roots (Lerchl et al., 1995). In 0. sativa, OVP2 is expressed at moderately higher levels in calli than roots and shoots, compared to OVPZ (Sakakibara et al., 1996). In B . vulgaris, on the other hand, analyses of 7-day-old dark- and light-grown seedlings, leaves, roots and storage roots of 1- and 7-month-old plants fail to reveal consistent and significant differences between the steady state levels of expression of BVPZ and BVP2 (Kim et al., 1994a). Evidently, the application of higher-resolution techniques, such as in situ hybridizations (Coen et al., 1990) or P-glucuronidase reporter analyses (Jefferson et al., 1987), in combination with exposure of the tissues concerned to a wider range of growth conditions, is needed to address the question of whether multiple genes for the V-PPase provide isoforms with expression characteristics peculiar to specific cell types, developmental stages and/or environmental factors, or generate enzyme molecules with modified catalytic and/or regulatory characteristics.

IV. STRUCTURE-FUNCTION RELATIONS A . ONE POLYPEPTIDE IS SUFFICIENT FOR PUMP FUNCTION

It was because of the inherent difficulty of reaching firm conclusions concerning the sufficiency of the substrate-binding subunit for V-PPase function by traditional biochemical methods that the independent strategy was adopted of heterologously expressing cDNAs for this subunit in S . cerevisiae and testing for PPi-dependent Hf translocation (Kim et a [ . , 1994b). Britten et al. (1992) describe a procedure for in vitro reconstitution of the transport function of the V-PPase purified from V. radiata to produce proteoliposomes capable of high rates of K+-activated, PPi-dependent H+ translocation, but SDS-PAGE of the final preparation reveals the coinsertion of two polypeptides of M , 20 000 and 21 000 in addition to the M , 66 000 subunit. For a different reason the findings of Sato et al. (1994) are also not conclusive. Although these authors succeeded in reconstituting higher-purity preparations of the substrate-binding subunit, the rates and extents of PP,-dependent Hf translocation observed are 5-10-fold lower than those achieved by the cruder preparations of Britten et al. (1992). Thus, in neither case could the need for additional subunits for full transport competence be discounted. By contrast, heterologous expression of the cDNA encoding this subunit demonstrates that it is the sole polypeptide species of the transportcompetent pump. Expression of the single cDNA species AVP for the substrate-binding subunit from Arabidopsis in yeast yields membranes capable of V-PPase-mediated H + translocation. Construction of the yeast-E. coli shuttle vector pYES2-AVP by insertion of the entire open-reading frame of AVP into the polylinker located between

312

R.-G. ZHEN et al.

the GA Ll promoter and CYCl termination sequences generates plasmid which, when transformed into S. cerevisiae, specifies galactose-inducible V-PPase activity in the vacuolar membrane-enriched fraction (Kim et ul., 1994b). The integrity of the heterologously expressed V-PPase and its non-identity with the endogenous soluble PPase of yeast, as indicated by the finding that it, alone, is activated by K+, inhibited by AMDP, active in transmembrane H+ pumping and subject to induction by galactose, unequivocally demonstrates that the pYES2-AVP encoded substrate-binding subunit by itself is sufficient for the assembly of the transport-competent pump. Direct participation of other polypeptides such as the M, 20 000 and 21 000 species detected in the in vitro reconstituted preparations of Britten et al. (1992), or a requirement for accessory polypeptides in order to achieve higher rates of PPi-dependent H+ translocation in the preparations described by Sat0 et al. (1994), need not be invoked. Heterologous expression of AVP in yeast also clarifies the discrepancy between the mass of the substrate-binding subunit computed from its deduced amino acid sequence and its estimated size. The open-reading frame of AVP encodes a 770 amino acid polypeptide with a computed mass of 80800 (Sarafian et al., 1992a), but the corresponding immunoreactive polypeptide of galactose-grown pYES2-AVP transformants has an M , of 66 800 on SDS gels. Although the possibility of post-translational modification of the polypeptide encoded by AVP cannot be eliminated completely, two factors make this seem improbable. First, the endogenous V-PPase of Arubidopsis is also constituted of a single polypeptide with an apparent M, of 66 800 (Rea et al., 1992a). If post-translational modification were operative it would have to be by identical mechanisms in both yeast and plant cells. Second, direct sequence acquired from the M , 64 500 and 66 000 substrate-binding subunits of B. vulguris and V. radiutu, respectively, include both the N terminus and sequences within 24 amino acid residues of the C terminus (Rea et al., 1992a; Maeshima and Yoshida, 1989; Sarafian et al., 1992a). While C-terminal proteolysis would be compatible with the direct sequence data, it cannot account for a mass shift of greater than 3 kDa. It is probable, in accord with earlier speculations (Sarafian et ul., 1992a), that the substrate-binding subunit of the V-PPase does, indeed, have a molecular weight of 79-81 kDa but migrates with an anomalously low M , on SDS gels. The binding of non-saturating amounts of SDS by membrane proteins is a common phenomenon and frequently accompanied by changes in apparent M, as a result of irregularities in the shape of the SDS-protein complex, exposure of charged amino acid residues or both (Maddy, 1976). B. AVP DOES NOT FUNCTIONALLY COMPLEMENT YEAST V-ATPase MUTANTS

One reason for choosing S . cerevisiae for heterologous expression of the V-PPase is the availability of strains mutated or disrupted for the V-ATPase

THE VACUOLAR H +-PPase

313

(Nelson and Nelson, 1990; Manolson et al., 1992; Yamashiro et al., 1990). Mutations in or disruptions of the structural genes encoding the V-ATPase make yeast cells hypersensitive to extracellular Ca2+ (Ohya et al., 1991) and neutral buffered media (Nelson and Nelson, 1990; Yamashiro et a f . , 1990; Ohya et al., 1991). V-ATPase mutants, unlike isogenic wild type strains, show diminished growth at pH values in excess of 6.5 units (Nelson and Nelson, 1990) and loss of viability at high extracellular Ca2+ concentrations (Ohya et a f . , 1991). It was therefore speculated that the conditional lethality of the V-ATPase mutant phenotype and the absence of endogenous V-PPase from S. cerevisiae might afford a means of phenotypically probing heterologous V-PPase expression in this system. Specifically, if heterologously expressed functional V-PPase can complement the loss of function (transmembrane H+ translocation) associated with the abolition of endogenous V-ATPase activity, it may be feasible to rescue the mutant phenotype. To date, however, all attempts to suppress the V-ATPase mutant phenotype and restore growth at high extracellular Ca2+ concentrations or neutral pH values through heterologous expression of the V-PPase have been unsuccessful (E. J. Kim and P. A. Rea, unpublished findings). At least two intrinsic factors, the Ca2+sensitivity of the V-PPase and the low cytosolic PP, content of yeast cells, may abrogate rescue by the V-PPase. The V-PPase is subject to pronounced inhibition by [Ca2+Ifree(Rea et a f . , 1992b), as is the heterologously expressed enzyme (Kim et al., 1994a). While the in vitro ko value for inhibition of the V-PPase by Ca2+ (2.4-3.4 pM; Reaetal., 1992b) would imply only limited inhibition in vivo if cytosolic [Ca2+Ifreeincreases from 0.15 to 0.90 p M upon transfer of S. cerevisiae V-ATPase mutants from Ca2+-deficient to Ca2+-replete media (Ohya et a f . , 1991), the validity of this conclusion will depend on the actual sensitivity of the V-PPase in vivo. Data relevant to the sensitivity of PPases in vivo are scant, but in the one system where the inhibitory potency of Ca2+ ha5 been estimated in situ (the soluble PPase of rat liver mitochondria1 matrix) the operational k0.5 is 0.3 p M (Davidson and Halestrap, 1989) -an order of magnitude lower than the value (2.6pM) obtained in vitro. If the k o 5 of the V-PPase were similarly overestimated, impaired endogenous V-ATPase function would undoubtedly diminish heterologous V-PPase function in yeast. The availability of cytosolic substrate may similarly limit V-PPase function in S. cerevisiae. The intracellular distribution of PP, in yeast differs markedly from that of plant cells (see Section 1I.B). While the total cellular PP, content of yeast cells may be high, ranging from 3 to 53 mM, depending on position in the cell cycle, at least 70% of the measured PP, is restricted to the vacuole (Urech et a[., 1978). Moreover, essentially all of the soluble PPase activity of S. cerevisiae is cytosolically localized. When account is taken of this and the fact that the soluble PPase has a K , for total PP, of 1 p M (Smirnova et al., 1988), large turnover number ( 2 1 2 ~ ~Springs '; et al., 1981) and represents approximately 1% of total cellular protein (Cooperman el al.,

314

R.-G. ZHEN et al.

1973), it is probable that the standing concentration of PPi in the cytosol of yeast cells is in the low micromolar range, well below the levels required by the V-PPase (Leigh et al., 1992; Baykov et al., 1993a). C. HOMOMULTIMERIC STRUCTURE

The sufficiency of the substrate-binding subunit for core catalysis and regulation does not preclude subunit-subunit interactions and oligomerization. Indeed, most estimates of the native size of the V-PPase indicate a homomultimer. What is not clear, however, is the exact degree of oligomerization and the conditions under which it applies. The apparent size of the V-PPase depends on how it is measured and which catalytic function, PPi hydrolysis or PP,-dependent Hf translocation, is assayed. Estimates of the structural size of the V-PPase, determined by gel filtration (Rea and Poole, 1986; Wang et a l . , 1986; Sato et a l . , 1991) and SDS-PAGE after cross-linking with dimethyl suberimidate (Maeshima, 1991) fall in the range 135-158 kDa. The apparent functional size of the enzyme, determined by radiation inactivation, on the other hand, is critically dependent on whether PPi hydrolysis or PPi-dependent H translocation is measured postirradiation. Estimates of the target size of the V-PPase for PPi hydrolysis yield values ranging from between 88 and 96 kDa (Fig. 3A) +

Fig. 3. Kinetics of inhibition of V-PPase-mediated PP, hydrolysis and H + translocation by (A) 3-(N-maleimidylpropionyl)biocytin (MPB) and (B) y irradiation. The results in part (A) and the considerations in Section 1V.C are consistent with the scheme shown in which the V-PPase is a functional monomer during PP, hydrolysis but a functional dimer during PPl-dependent H+ translocation. Covalent modification of only one of the two subunits is sufficient for the total abolition of H+ pumping whereas both subunits must be modified for the total abolition of PP, hydrolysis. Providing that the kinetics of modification of Cys634 (see Section 1V.E) are first order such that A = A,e-k"[MPBI, where A is the activity at time t , A, is the initial activity and k' is a pseudo first order rate constant, a plot of In(A/A,) against [MPB] will yield a straight line of slope -2k't for H+ pumping and -k't for PP, hydrolysis. The experimental1 determined rates of inactivation of H+ pumping and PP, hydrolysis are 1.50 x 1O'M-' min-' and 0.67 x lo3 M-' min-' , respectively (A). The kinetics of radiation inactivation indicate target sizes of 119.3 k 12.0 and 635.8 -t 94.1 kDa, respectively, when PP, hydrolysis and PP,-dependent H translocation are assayed postirradiation (B). For the results shown in part (A), vacuolar membrane vesicles purified from V . radiuta were reacted with MPB for 10min at 0°C. Reaction was terminated by the addition of dithiothreitol, and aliquots of the reaction mixture were assayed for V-PPase activity. For the results in part (B), membranes were purged with N2 and irradiated in solid C 0 2 (-78°C) with y rays from a 6oCosource for various lengths of time. Molecular size was estimated from the expression log(molecu1ar is the dose required to decrease enzyme size) = 5.89 - log D37,r- 0.0028t, where activity to 37% of the initial value and t is the temperature ("C) at which the samples were irradiated. +

A

B lysis (107-131 ma)

-

h ..-a

10.0

h .* .C

I

0

'

I

'

I

20

'

I

'

I

'

I

'

40

[MPBI (PM)

1

60

I

I

' 80

I

-

1.0' ' ' ' 0

'

I

5

'

I

'

'

1

10

Dose (MRad)

'

'

'

'

L 15

316

R.-G. ZHEN et al.

(Sarafian et a f . , 1992b; Pugliarello et al., 1991; R.-G. Zhen and P. A. Rea, unpublished findings) to between 139 kDa (Sato et a f . , 1991) and 160 kDa (Chanson and Pilet, 1989). However, when activity is measured as H+ translocation after sample irradiation, values of between 307 and 350 kDa (Sarafian et al., 1992b) and as high as 730 kDa (R.-G. Zhen and P. A. Rea, unpublished findings) (Fig. 3B) are obtained. These results are reconcilable only if it is assumed that the minimum estimates of the target size of the hydrolytic functionality (88-96 kDa) are the ones least subject to the artifactual, secondary, non-nihilatory effects of high-energy, ionizing radiation. If this assumption is valid, these results are in remarkably close agreement with the size of the substrate-binding subunit deduced from its cDNA sequence (79-81 kDa) (see Sections 1II.B and 1V.A). By extension, the double molecular size estimated by gel filtration and chemical cross-linking may mean that the V-PPase is a functional monomer and structural dimer during substrate hydrolysis. The physical meaning of the 3 4 f o l d (Sarafian et al., 1992b) or 5-6-fold (R.-G. Zhen and P. A. Rea, unpublished findings) greater radiation inactivation size for H+ transport versus PPi hydrolysis remains to be resolved, but several explanations may apply: (a) the transport activity of the V-PPase is more prone to secondary damage by y irradiation than substrate hydrolysis; or (b) the functional unit for PPrdependent H+ translocation is a homomultimer of several (3-6) identical 79-81 kDa subunits. The first explanation requires the secondary effects of radiation to be relatively specific to the V-PPase, since the radiation inactivation sizes for transport and substrate hydrolysis by the V-ATPase in the same preparation coincide (Sarafian et a f . , 1992b). The second explanation, on the other hand, would imply tight functional interdependence between subunits for transport but not substrate hydrolysis. The target molecular size of the substrate-binding subunit, estimated by 1251-linked Western analysis, is 64 kDa (Sarafian et a f . , 1992b), suggesting that irradiation annihilates only one subunit per ionizing hit. Consequently, if the V-PPase is a homomultimer during transport, its large inactivation size results from functional cooperation between subunits after irradiation rather than energy transfer between subunits during irradiation. There is therefore a possibility that autonomous subunits can catalyse PP; hydrolysis whereas H+ transport and/or transduction of the free energy of PPi hydrolysis to transmembrane H+ pumping demands functional contiguity between subunits. The differential sensitivity of PP; hydrolysis and PPi-dependent H+ translocation to irreversible inhibition by the membrane-impermeant sulfhydryl reagent 3-(N-maleimidylproprionyl)biocytin(MPB) (see Section 1V.E) is also consistent with alternate functional oligomeric states. Although MPB, like NEM, inhibits the V-PPase with single-site (pseudo-first order) kinetics, and the group whose alkylation is responsible for the abolition of activity is a single cysteine residue (Cys634, see Section IV.E), PPi-dependent


317

H + translocation is twice as sensitive to inhibition by MPB as PP, hydrolysis (Fig. 3A). In view of the MPB insensitivity of C634S or C634A mutant enzymes, whether PPi hydrolysis or PP,-dependent Hf translocation is assayed, and the fact that both hydrolytic and pumping activity are inhibited with pseudo-first order kinetics, the two-fold greater rate for the inhibition of pumping (k’ = 1.5 X lo3 M-’ min-’) versus substrate hydrolysis (k’ = 0.67 x lo3 M-l min-l) is indicative of a requirement for covalent modification of only half of the Cys634 residues for the total abolition of PPi-dependent H + translocation (Fig. 3). These results are consistent with the notion that the minimum unit competent in PP,-dependent H + translocation is a functional dimer of the hydrolytic moiety (Fig. 3). D . REVISED TOPOLOGlCAL MODEL

Now that the sufficiency of AVP for the elaboration of active V-PPase in S. cerevisiae has been established, the onus is not to identify new subunits but instead to determine the organization of the M , 66000 subunit itself. Pending crystallographic data, a rigorous working model for the number of transmembrane spans and the disposition of the intervening hydrophilic loops is a minimum requirement for rational structure-function analyses of the V-PPase. It is for this reason and because of emergent doubts over its precision that the original 13-span model of the substrate-binding subunit (Sarafian et al., 1992a) has been re-evaluated. The revised topological model of the V-PPase depicted in Fig. 4 was derived from the sequences of AVP (Sarafian et af., 1992a), B V Pl, BVP2 (Kim et al., 1994a) and HVP (Tanaka et al., 1993) using the TopPred I1 and MEMSAT (membrane structure and topology) programs of Claros and von Heijne (Karolinskaya Institute, Sweden) and Jones et al. (1994), respectively. TopPred I1 is a public domain Macintosh software package for predicting the topology of both prokaryotic and eukaryotic membrane proteins through the combined application of hydrophobicity analyses and the “positive-inside” and “charge-difference’’ rules (Sipos and von Heijne, 1993). MEMSAT, an IBM-PC-compatible program, is based on expectation maximization. From the distributions of amino acids compiled from membrane proteins, or portions thereof, of defined topology, the loglikelihood ratios (s,)for domain classes are calculated for each of the 20 amino acids according to the expression s, = In(q,/p,), where p, is the relative frequency of amino acid i in all sequences in the dataset and q, is the relative frequency of occurrence of the same amino acid in a particular domain. The s, values, or propensities, are then used to relate a sequence with a given topology. Examination of the structure of the M , 66000 subunit of the V-PPase by TopPred 11 consisted of three main stages. (i) The construction of hydrophobicity profiles using a trapezoid sliding window (von Heijne, 1992).

Fig. 4. Revised topological model of V-PPase from Arabidopsis showing positions of conserved cysteine and acidic amino acid residues subjected to mutagenesis. The model is based on predictions made using the TopPred I1 and MEMSAT algorithms of Claros and von Heijne (Karolinskaya Institute. Sweden) and Jones et al. (1994), respectively. Boxed sequences are transmembrane spans predicted by TopPred 11; hatched sequences are transmembrane spans predicted by MEMSAT. Conserved cysteine residues common to the deduced sequences from Arabidopsis (Sarafian et al., 1992a), B. vulgaris (isoforms 1 and 2; Kim et al., 1994b) and H . vulgare (Tanaka et a l . , 1993) are shown in white in black rectangles; conserved aspartate and glutamate residues are shown in white in black circles.


319

Depending o n the height and width of the hydrophobicity maxima and the preset “upper cut-off” and “lower cut-off” values for the computed hydrophobicity indices, spans were categorized as either “certain” or “putative”. (ii) Enumeration of the difference in representation of positively charged amino acid residues between the two sides of the membrane and tests of the adherence of any given model to the “positive-inside” rule - the bias in favour of arginine and lysine residues in hydrophilic loops with a cytosolic disposition in most polytopic membrane proteins (von Heijne, 1986). (iii) Application of the “charge difference” rule (Hartmann et uf., 1989) wherein the net charge difference between the 15 N-terminal and the 15 C-terminal residues flanking the most N-terminal transmembrane span is computed. Transmembrane orientation is correlated with the disposition of charged residues in the immediate vicinity of the first membrane span. The segment C-terminal to the first span is generally positively charged with respect to the N-terminal flanking regions in membrane proteins possessing a luminally oriented N terminus (Hartmann et ul., 1989). Deployment of the MEMSAT program, on the other hand, entailed analysis of segments of the sequence of the V-PPase in terms of their likelihood of being located within a particular topological element. Based on statistical analysis of the distribution of amino acids in membrane proteins, the MEMSAT program ranks amino acids according to their propensities for being associated with each of five types of topological element. These are two classes of hydrophilic loop, designated cytoplasmic (outside) loop (L,) and luminal (or inside) loop (LJ, and three classes of transmembrane helix domain, designated helix inside (Hi). helix middle (H,,,) and helix outside (Ho). Both programs predicted two main departures from the original 13-span model: three transmembrane spans in addition to the 13 predicted previously and an identical (luminal), rather than opposed (cytosolic N terminus; luminal C terminus), orientation for the N and C termini (Fig. 4). Significant is the basic equivalence between the predictions deriving from TopPred I1 and MEMSAT. Whereas TopPred I1 constrains the length of transmembrane spans at a specific value - 21 amino acids in this study - and is based on the assumption that all spans are perpendicular to the phospholipid bilayer, MEMSAT selects the best fit within a user-defined range of minimum and maximum lengths (17-25 amino acids in this study), thereby diminishing bias in favour of any one angle of intersection. Nevertheless, the margins of 11 of the 16 spans predicted by the two programs differed by no more than four amino acid residues, and the average lengths of the spans (21.5 by MEMSAT versus a fixed value of 21 for TopPred 11) were virtually identical. Of the spans predicted by MEMSAT, only two, spans IX (408-425) and XV (671-687), were shorter than the 20 amino acids required to traverse the entire bilayer. However, in both cases the counterpart helices predicted by TopPred I1 included all 17 of these residues. Accord-

320

R.-G. ZHEN et al.

ingly, when the MEMSAT settings were altered to increase the minimum span length from 17 to 19 residues, the overall topology of the V-PPase was unchanged; spans IX and XV were simply lengthened. The three transmembrane spans (V (230-252), VI (292-316), X (452-476)) not identified in the original 13-span model were overlooked as a result of the proximity of adjacent maxima in the hydrophilicity profiles (span X) and the use of window sizes so broad as to obscure hydrophobic segments adjacent to regions of extreme hydrophilicity (spans V and VI). TopPred 11, by contrast, applies a narrower sliding window (11 amino acid residues) and so permits better resolution of neighbouring transmembrane spans. While MEMSAT ranked the 16-span model highest, it should be appreciated that other models with high scores were also obtained. For instance, two models that were ranked just below the 16-span model contained 14 and 15 spans. Interestingly, these two models were nearly identical topologically to the 16-span model, in that the orientations of their component spans were conserved. In the 14-span model, the two lowest-scoring transmembrane spans (V and VI) were excluded, so preserving the orientation of the N and C termini and the remaining C-terminal spans. In the 15-span model, the last transmembrane span (XVI, 743-761), alone, was excluded, so transferring the C terminus from the luminal to the cytosolic face of the membrane while preserving the orientation of all of the other spans. A luminal orientation for the N terminus was deduced from the “charge difference” rule (Hartmann et al., 1989). Examination of the N-terminal residues immediately adjacent to the first transmembrane span (positions 14-34) of AVP revealed no positively charged residues and two negatively charged residues (Glu9, Glu13), giving a net charge of -2. The corresponding regions of BVPl, BVP2 and HVP had the same charge, attributable to Glu13 in all three sequences and Glu9 in HVP and Asp9 in BVPl and BVP2. Of the 15 amino acid residues on the C-terminal side of the first span of AVP, two (Arg36, Lys38) were positively charged and one (Asp42) was negatively charged, giving a net charge of + l . BVPl and BVP2 had three positive charges and one negative charge (net charge +2), and HVP six positive charges and no negative charges (net charge +6), in this region. The net charge difference of at least 3 (-2 for the N-terminal 15 residues before the first span versus +1 for the C-terminal 15 residues after the span for AVP, 4 for BVPl and BVP2,8 for HVP) in all four cases is consistent with a luminal orientation for the N terminus of the V-PPase. Hence, if the 16-span, or at least an even-span, model is a good approximation of the topology of the M, 66000 subunit, a luminal orientation must also be predicted for the C terminus. Inspection of Fig. 4 shows that the cytosolically disposed loops of the 16-span model for AVP contain a significantly greater number of arginine and lysine residues than the luminally oriented loops (83 versus 17%, respectively) in accord with the “inside-positive’’ rule (von Heijne, 1986).

T H E VACUOLAR H’-PPase

321

A similar pattern was found for the other three V-PPase sequences analysed. Likewise, the majority of the residues found in the hydrophilic loops o f the 16-span model are cytosolically oriented (79.7%, 45.6% of total) in good agreement with the expectation that the overall distribution of hydrophilic loops would favour the side of the membrane responsible for catalysis and ligand binding. By comparison, the 12-span model of the V-PPase proposed by Tanaka et af. (1993), on the basis of the deduced amino acid sequence of HVP, is unlikely on two counts: (a) it predicts that the majority (-58%) of the amino acid residues in putative hydrophilic loops are luminally oriented; and (ii) it places Cys634, which is known to have a cytosolic orientation (see Section 1V.E; Zhen et af., 1994b), on the luminal face of the membrane*.

E.

IDENTIFICATION O F SUBSTRATE-PKOTECTABLE MALEIMIDE-REACTIVE CYSTEINE RESIDUE

As an initial step towards the provision of protein chemical structural data for the V-PPase and as a direct test of some of the conclusions drawn from modeling its topology (see Section IV. D), the cysteine residue whose alkylation by maleimides is responsible for the inhibition of activity has been localized. By exploiting the protection conferred by substrate against inhibition by maleimides, the low permeability of the vacuolar membrane to MPB and the high level of sequence conservation amongst the V-PPases from disparate sources (see Section III.B), it has been possible to define the membrane orientation of the hydrophilic loop in which the maleimidereactive cysteine is located by peptide mapping and site-directed mutagenesis (Zhen et a / . , 1994b; Kim et al., 1995). NEM irreversibly inhibits the V-PPase in a substrate [(Mg” + PP,)] protectable manner with pseudo-first order kinetics (see Section 1II.A) (Britten er af., 1989; Rea et af., 1992a). Moreover, since NEM and the membrane-impermeant maleimide, MPB (see Section IV.C), inhibit the V-PPase with similar kinetics and compete for a common binding site on the M , 66 000 substrate-binding subunit, a single residue located in a cytosolically *Knight cf a / . (19%) have recently generated Arahidopsis plants expressing V-PPaseapoaequorin fusions. Assuming that the fusion of apoaequorin with the C terminus does not itself change the topology of the V-PPase, these studies are consistent with a cytosolic disposition for this portion of the M , 66000 polypeptide. It is therefore instructive to note that of the two models ranked just below the 16-span model, the 15-span model best accommodates these results and those dcrivcd from the mapping of Cys634. The last putative transmembrane span (XVI), alone. is excluded from the 15-span model, so transferring the C terminus from the lumen to the cytosol while leaving the rest of the structure unchanged. If this reasoning is correct, putative span XVI in Fig. 4 is probably not a span, in which case the last 200 residues of the V-PPase appear to have been topologically defined. A structure possessing an approximately 80-residue C-terminal extension is indicated.

322

R.-G. ZHEN et al.

disposed extramembranous domain is inferred to undergo covalent modification in both cases (Zhen er af., 1994b). Peptide mapping of this residue through selective labelling of the V-PPase with 14C-labelledNEM, purification of the M , 66000 subunit and its digestion with V8 protease generates a single ''C-labelled band (V814K) migrating at M , 14 000 on Tris-tricine gels that aligns with the C-terminal segment of the substrate-binding subunit. Because V814K encompasses only one cysteine residue at position 634 that is conserved amongst the V-PPases from Arabidopsis (Sarafian et af., 1992a), B. vufgaris (Kim et af., 1994a), H . vulgare (Tanaka et al., 1993), N . tabacum (Lerchl et af., 1995) and 0. saliva (Sakakibara et al., 1996), this residue, located in putative hydrophilic loop XI11 (Fig. 4), is concluded to contain the cytosolically oriented sulfhydryl group whose alkylation by maleimides is responsible for inactivation of the enzyme (Zhen et af., 1994b). Though capable of generating useful structural information, the results of protein chemical investigations of this type suffer from two limitations. (i) They do not provide any indication of the direct participation of Cys634 in catalysis and/or substrate binding. The fact that reaction of an enzyme with a group-specific reagent causes irreversible inhibition does not automatically imply that the functional group concerned is in the active site. Covalent modification of a non-essential residue could, for instance, result in a conformational change that inactivates the enzyme. By the same token, protection against inhibition by substrate does not mean that the susceptible group is in the substrate-binding site. It is equally likely that conformational changes accompanying substrate binding result in the occlusion of reactive sites that are otherwise remote from the active site. (ii) Technical constraints prohibited direct identification of the ['4C]NEM-modified residue (Zhen et a f . , 1994b). There is a possibility, albeit small, that the 14C-labelled residue is not a cysteine residue or that non-sequenceable N-terminally blocked peptides other than V 8 1 4 ~ate responsible for the signal seen on fluorograms. Thus, in order to assess the full significance of these findings, it is important to test the necessity of specific amino acid residues for maleimide action, catalysis and/or substrate binding by site-directed mutagenesis. In the case of the susccptibility of the V-PPase to inhibition by sulfhydryl reagents, analyses of the mutated enzyme after its heterologous expression in S. cerevisiae not only affords a means of independently identifying the cysteine residue, or residues, required for inactivation by maleimides but also offers the possibility of determining if one or more of these residues is involved in catalysis. Alignment of the deduced amino acid sequences of the V-PPase from Arabidopsis, B. vufgaris and H . vulgare discloses a total of nine conserved cysteine residues corresponding to positions 19, 78, 128, 136, 308, 343, 411, 444 and 634 in the sequence from Arabidopsis (Fig. 4). To determine which, if any, of these residues is the site of action for inhibition of the enzyme by NEM, each conserved cysteine residue was singly mutated to a serine residue


323

at positions 19, 78, 128, 136, 308, 343, 411 and 444 and a serine or alanine residue at position 634. Cys-, Ser substitutions were employed to conserve the electronegativity of the side-chain. The Cys+ Ala substitution at position 634 was introduced to assess the effects of eliminating this electronegative centre while minimizing the structural deformation that might otherwise arise from the introduction of a bulkier substituent. By testing the susceptibility of the mutated forms of the V-PPase to (Mg2+ + PPi)-protectable, free PPi-potentiated inhibition by NEM, the effects of mutagenesis on catalytic activity, NEM inhibitability and substrate binding could be examined in parallel (Kim et al., 1995). All of the substitutions, except for C634S and C634A, exert little o r no effect on the activity of the V-PPase or its sensitivity to inhibition by NEM. Cys-,Ser substitutions at positions 19, 78, 308, 343 and 411 yield enzyme exhibiting a pattern of (Mg2+ + PPJ-protectable, free PPi-potentiated inhibition by NEM similar to wild type V-PPase. Likewise, although C128S, C136S and C444S mutants are slightly less sensitive to inhibition by NEM in the presence of free PPi than wild-type enzyme, they are nevertheless quantitatively protected by Mg2+ + PPi. Furthermore, the specific activity of the V-PPase approximates the wild-type in all cases. In direct contrast, substitution of Cys634 with a serine or alanine residue yields heterologously expressed enzyme that is active in both PP, hydrolysis and PP,-dependent H+ translocation but insensitive to NEM irrespective of whether Mg2+ + PP, or free PP, is included in the NEM reaction medium. Together with the results from the protein chemical investigations, these studies unequivocally identify the sulfhydryl group of Cys634 as the reactive moiety. Two inferences therefore follow: (a) a cytosolic orientation must be assigned to the putative hydrophilic loop (XIII) encompassing Cys634 (Fig. 4), so defining the topology of this portion of the structure; and (b) earlier speculations concerning the location of the NEM-reactive sites, or sites, are refuted. It has been noted that Cys128 and Cys136, located in putative hydrophilic loop 11, are flanked by sequences possessing a spacing and alternation of acidic and basic residues equivalent to those regions of the soluble PPase from S. cerevisiae known, from site-directed mutagenesis and X-ray crystallography, to be critical for catalysis (Rea ef al., 1992c; Cooperman et al., 1992). Hence, while the V-PPase and soluble PPases appear to be remote evolutionarily (see Section IILB), it has been speculated that they may share convergent motifs related to the need for both classes of enzyme to interact with the same substrates, inhibitors and activators. Further, in view of the proximity of Cys128 and Cys136 to these motifs, the sensitivity of the V-PPase to covalent modification and inhibition by maleimides has been attributed to alkylation of one or both of these cysteine residues (Rea et al., 1992b). The direct participation of loop I1 residues other than cysteine in substrate binding and/or turnover has not yet been addressed, but the finding that C128S and C136S mutants are not only active in catalysis

324

R.-G. ZHEN et al.

but also retain sensitivity to (Mg2+ + PPJ-protectable, free PPi-potentiated inhibition by NEM clearly contradicts the proposed involvement of these cysteines in inhibition by maleimides. The revised topological model of the V-PPase (Fig. 4) and the sequence data acquired for the V-PPase from sources other than Arabidopsis reinforce this conclusion. Although loop I1 was originally assigned a cytosolic orientation on the basis of hydropathy and hydrophobic moment analyses (Sarafian et a f . , 1992b), application of the TopPred I1 and MEMSAT algorithms indicates a luminal orientation (Fig. 4). This is inconsistent with the direct interaction of this loop with cytosolic ligands. Accordingly, the E(X)4DXK(X)4D motif at position 119 in AVP is not conserved in BVPl and BVP2 (Kim et a f . , 1994b). AVP has the sequence EGFSTDNKPCTYQ _ _ but the corresponding sequences of BVPl and BVP2 are EGFSTSSQECTYD and EGTSTESQPCTYS, respectively. Of the four charged residues tenhtively&igneda role in catalysis in this region of AVP (Glu119, Asp124, Lys126, Aspl31) only one (Glu119) is conserved between BVPl and BVP2. An insight that could not have been gained other than through mutagenesis is the retention of catalytic activity by the maleimide-insensitive C634S and C634A mutants. This situation is reminiscent of the A subunit of yeast V-ATPase (Taiz et a f . , 1994) and E . cofi lac permease (Kaback, 1992). In the case of the V-ATPase, C261S mutants acquire insensitivity to NEM while retaining wild-type hydrolytic activity (Taiz et a f . , 1994). In the case of lac permease, only Cys154 appears to be important for transport, but even this residue is not essential (Kaback, 1992). When Cys154 is replaced with a valine and each of the other cysteine residues is replaced by serine, about 30% of the initial rate of transport and about 60% of the steady state level of accumulation of lactose is achieved by the “C-less” permease versus wild type, although the former is rendered insensitive to NEM (van Iwaarden et a f . , 1991). By analogy to these transporters and on the basis of the dispensability of all of the conserved cysteine residues of the V-PPase, the inhibitory action of maleimides on wild-type enzyme is attributed to structural deformation imposed by the introduction of a bulky substituted alkyl group on Cys634. Thus, while the substrate protectability of enzyme inhibition and covalent modification of Cys634 by maleimides indicate that this residue is close to or conformationally coupled with the substrate-binding site, direct participation of this or any other cysteine residue in the catalytic cycle of the V-PPase is excluded. F. POTENTIAL COUPLING SITES

Current mutational studies of the V-PPase are concerned with the involvement of acidic residues in substrate turnover and/or Hf translocation. Two factors prompted these studies. The first is the need to understand better the

THE VACUOLAR Hf-PPase

325

Fig. 5. Comparison of the deduced amino acid sequence of AVP (Sarafian et al., 1092b) with the amino acid sequences of the c peptides of F-ATPases. The F-ATPase subunits shown are R. rubrum subunit c (Falk and Walker, 1988), P. sativum chloroplast subunit I11 (Huttly et a l . , 1990) and P. sativum mitochondrial subunit 9 (Morikama and Nakamura, 1987). The C-terminal sequences of the c subunits and residues 229 through 245 of AVP are aligned. Identities and conservative substitutions are indicated by white boxes and shaded regions, respectively. The DCCD-reactive glutamine residues of the F-ATPase subunits are shown in bold. All of the V-PPase sequences listed in Fig. 2 contain the consensus sequence LFE(A/S)ITGYGLGGSSMALF.

identity and location of acidic residues with the potential for undergoing cycles of protonation and deprotonation within the plane of the membrane. On the basis of investigations of other Hf pumps, it might be anticipated that acidic residues associated with transmembrane spans directly participate in H+ uptake, translocation and release by the V-PPase. The second factor was the observations of Nyren et al. (1993). These authors noted that the sequences encompassed by positions 227 through 245 of the V-PPase from Arabidopsis bear a resemblance to the C-terminal regions of the c subunits of F-ATPases. The C-terminal sequence flanking Glu229 in AVP is 71, 65 and 67% similar (35, 47 and 39% identical) to R. rubrum c subunit (positions 58-74), Pisum sativum chloroplast subunit 111 (positions 61-77) and P. sativum mitochondrial subunit 9 (positions 55-72) (Fig. 5 ) . Since the c peptide is the most highly conserved subunit of the H+-conductive F, sector of F-ATPases and NJ”’dicyclohexy1carbodiimide (DCCD) binds to the only acidic residue (glutamate or aspartate) located in the middle of the second of the two transmembrane a helices of this polypeptide to abolish H + translocation, Nyren et al. (1993) propose that the sequence flanking Glu229 of the V-PPase plays a similar role. Specifically, given the sensitivity of the V-PPase to inhibition by DCCD (Chanson and Pilet, 1987; Maeshima and Yoshida, 1989), they suggest that Glu229 is the residue whose covalent modification by this hydrophobic carbodiimide is responsible for the inhibition of V-PPase activity. This proposal has been investigated directly by singly substituting most of the conserved aspartate and glutamate residues located near or within transmembrane spans (Fig. 4) and testing the capacity of wild-type and

326

R.-G. ZHEN et al.

mutant V-PPase for PPi hydrolysis, H+ translocation and inhibition by DCCD (Zhen and Rea, 1995). In all cases these residues were substituted by their corresponding amides. In some cases, when the dicarboxylic acid-., amide substitutions were found to have a pronounced effect on V-PPase function, mutants containing structurally conservative Asp + Glu or Glu- Asp substitutions at these positions were also generated. A total of eight positions (Glu119, Glu229, Glu305, Glu427, Asp504, Asp573, Glu667, Glu751) were mutated (Table I1 and Fig. 4). The mutants generated fall into three main classes: (I) those that exhibit rates of PPi hydrolysis and PPL-dependentH+ translocation similar to the wild type; (11) those that are selectively impaired in H+ translocation; and (111) those that are grossly impaired in both PPi hydrolysis and H+ translocation (Table 11). Glu+Gln substitutions at positions 119, 229, 667 and 751 and an Asp+ Asn substitution at position 573 generate enzyme with at least 15% of wild-type PPi hydrolytic and PPi-dependent H+ pumping activity. In all class I mutants, PPi hydrolytic activity and the rate and extent of PPtdependent H+ translocation are diminished proportionately (see Table 11). By contrast, the one class I1 mutant obtained, E427Q, retains approximately 50%, of wild-type PPi hydrolytic activity but is more than 90% impaired in PPi-dependent H+ translocation: coupling efficiency is only about 10% of the wild type. The remaining dicarboxylate mutants, E305Q and D504N, fell into class 111. They are completely deficient in PP,dependent H+ translocation and catalyse PPi hydrolysis at rates less than 10% of the wild type. In none of the mutants could an increase in the background or (Mg2+ + PPi)-dependent H+ conductance of the membrane or a gross impairment of insertion of the V-PPase into the membrane explain the results. Vacuolar membrane-enriched vesicles harbouring any one of the V-PPase mutants achieve similar rates and extents of V-ATPase-mediated H+ translocation irrespective of whether PPi is included in the V-ATPase assay medium. All of the V-PPase mutants show similar levels of M, 66 000 polypeptide associated with the vacuolar membrane-enriched fraction. Of all the mutants generated, only two - E305Q and D504N - exhibited a marked decrease in susceptibility to inhibition by DCCD. While the dicarboxylate- amide substitutions at positions 119, 229, 427, 573, 667 and 751 have little or no effect on DCCD inhibitability versus the wild type, the corresponding substitutions at positions 305 and 504 diminish DCCD inhibitability by more than two-fold. Significantly, E305D and D504E mutants are as sensitive to inhibition by DCCD as the wild-type enzyme. Three main conclusions stem from these studies. First, contrary to the speculations of Nyren et al. (1993), Glu229 is not essential for either PPi hydrolysis or H+ translocation. E229Q mutated V-PPase shows some impairment of PPi hydrolysis and H translocation but, since both processes are diminished in parallel, there is only a small diminution of coupling ratio. Moreover, in direct opposition to the imputed role of Glu229 in inhibition +

T A B L E I1 Effect of single substitution of aspartate or glutamate residues on heterologously expressed Arabidopsis V-PPase activity. Vacuolar membrane-enriched vesicles were prepared from S. cerevisiae BJ5459 pYES2-AVP transformants in which the codons specifying the amino acids indicated had been mutated in the AVP insert. Substitutions involving no change in net charge are underlined. PPase activity was measured as described (Kim et al., 1995). PP,-dependent H' translocation was assayed fluorimetrically (Rea and Turner, 1990) using acridine orange (2.5 p M ) as the transmembrane p H difference indicator in assay media containing membrane vesicles (200 pg), 1 mM Tris-PP,, l0OrnM KCI, 0.4 M glycerol, 1 mM Tris-EGTA and 5 m M Tris-HCl (pH 8.0). Reaction was initiated by the addition of 1.3 mM M g S 0 4 , and the decrease in fluorescence (AF) was monitored against time. The initial rate of H' translocation and steady state p H gradient were estimated as AP/o mg-' min-' and AF% mg-', respectively. The coupling ratio was calculated as (AP% min-')l(pmol PP, hydrolysed min-I).

H+ translocation Class

I

I1 111

PPi hydrolysis

Coupling ratio

Substitution

Initial rate (AF% mg-' min-I)

Steady state (AF% mg-')

(pmol mg-' min-')

((AF% ) (pmol PP, hydrolyzed)-')

Wild type

163.8

208.8

1.04

157.5

E119Q E229Q

125.3 22.8

185.0 41.3

E667Q E75 1 Q

58.1 108.0

1.33 0.23 1.50 0.91 0.49 1.15

94.2 99.1 186.0 277.5 118.6 93.9

E427Q E427D

221.4 -

E305Q D504E

ND, not detectable; NA, not applicable.

253.8

215.0 129.4 152.5

0.53

235.0

0.98

18.5 225.9

ND

ND

0.05

ND ND ND

ND

NA NA NA NA

9.8

22.2

ND ND

0.27 0.10 o.01

328

R.-G. ZHEN er al.

by DCCD, E229Q substituted V-PPase is no less sensitive to inhibition by DCCD than wild-type or E229D-mutated enzyme. An acidic residue at this position is not required for inhibition by this agent, suggesting that the alignments between this portion of the V-PPase and the c peptides of F-ATPases (Fig. 5 ) are coincidental. Similarly, the aspartate residue at position 573 and the glutamate residues at positions 119, 667 and 751 do not appear to be critical for PP, hydrolysis, PP,-dependent Hf translocation or inhibition by DCCD. Mutation of these residues to their corresponding amides exerts little or no effect on PP, hydrolytic activity, H + pumping, coupling ratio or DCCD inhibitability. Second, and in contrast to the amino acid residues at positions 119, 229, 573, 667 and 751, Glu305 and Asp504 appear to be critical for catalysis. Both E30SQ and D504N exhibit less than 10% wild-type PPi hydrolytic activity, no detectable PPz-dependent H+ translocation, and the activity remaining in these mutants is markedly less sensitive to inhibition by DCCD than wild-type enzyme. These characteristics, together with the finding that E305D and D504E mutants show a recovery of DCCD inhibitability, are consistent with the idea that an acidic residue at these positions is required for inhibition of the V-PPase by DCCD. However, the finding that structurally conservative Asp+ Glu or Glu+ Asp substitutions cause no recovery of catalytic activity in the case of GluS04 and only a moderate increase in PP, hydrolytic activity alone in the case of Asp305, suggests that the steric constraints for catalysis are more stringent than those for DCCD binding. Third, E427Q mutated enzyme behaves as if decoupled. Such mutants, though still active in PP, hydrolysis, mediate H+-translocation at less than 6% of the wild-type rate to yield an 8-9-fold diminished coupling ratio. While it may be premature to conclude that these results demonstrate a direct role for Glu427 in H+ transfer, since its substitution by glutamine might cause a structural change that indirectly effects enzyme function, the large recovery of wild-type H + pumping versus the modest increase in PP, hydrolytic activity by a Glu+ Asp substitution does nonetheless imply an important role for an acidic residue at this position for H+ translocation per se. Changing the E427Q substitution to an E427D substitution causes a greater than 20-fold increase in the capacity for H + translocation whereas the capacity for PP, hydrolysis increases by only two-fold (Table 11). If Glu427 does indeed directly participate in H transfer, two corollaries follow. (i) Since E427Q and E427D mutants are similarly sensitive to DCCD, and as sensitive as the wild type, inhibition of the V-PPase by DCCD does not have a direct bearing on H + translocation. By analogy to the results of structural studies of F- and V-ATPases, involvement of the M, 66000 polypeptide of the V-PPase in transmembrane H+ conduction has been inferred from the susceptibility of the enzyme to inhibition and covalent modification by DCCD (Chanson and Pilet, 1987; Maeshima and Yoshida, 1989). From the results described here and the known reaction specificity +


329

of DCCD, such a conclusion is not warranted. Reaction of DCCD with carboxyl groups can have several consequences depending on the accessibility of water. If accessible to water, the dicyclohexyl-0-isourea formed upon reaction of a carboxyl group with DCCD is rapidly converted to free dicyclohexylisourea. As a result, the protein is neither modified nor irreversibly inhibited. Only in the absence of water, or other nucleophiles, is the acyl group of the activated intermediate shifted to one of the nitrogen atoms, so causing the stable incorporation of dicyclohexylisourea. In short, DCCD does not exclusively inhibit H + translocation. Inhibition by this compound simply indicates that catalytic activity is directly or indirectly dependent on carboxyl functions sequestered from bulk-phase water. (ii) According to the revised topological model shown in Fig. 4, Glu427 has a cytosolic disposition. If this residue is involved in H + translocation, it seems likely that it forms part of an input channel responsible for the entry of H+ at the cytoplasmic face of the membrane. It is conceivable that the E427Q mutation acts to neutralize the y-carboxyl group that would otherwise be present at this position and thereby block protonation of the other H+-carrying residues of the pump.

V.

FUTURE RESEARCH

Many of the molecular and biochemical tools necessary for dissecting the structure-function characteristics of the V-PPase have been assembled. Indeed, through the application of techniques of the types detailed in this chapter a number of important insights into the mechanistic basis of V-PPase function have been gained. The suitability of the 1,l-diphosphonate, AMDP, as a type-specific inhibitor of both the V-PPase and its putative homologue, the phototrophic bacterial Hf-PPi synthase, has been demonstrated. The substrate-active Mg2+-PPi complex and the existence of both high and low affinity Mg*+-binding sites have been determined. The sequence of the substrate-binding subunit, its pronounced conservation amongst species and its novelty have been elucidated. Direct experimental verification of the sufficiency of the one subunit encoded by AVP, and by implication its homologues from other plants, for all of the known core catalytic and regulatory activities of the pump has been provided. The membrane orientation of the maleimide-reactive cysteine residue whose alkylation is responsible for pump inactivation has been mapped. The identities of acidic amino acid residues located near or within transmembrane spans that may be involved in H+ coupling and inhibition of the enzyme by DCCD, respectively, have been delineated. Priorities for future research of the V-PPase are its crystallization, the design of screens for random mutated, heterologously expressed enzyme, determination of the sequence of the H+-PPi synthase from R. rubrum and

330

R.-G. ZHEN et al.

detailed biochemical and physiological analyses of plants engineered for enhanced or diminished V-PPase expression. X-ray diffracting crystals will eventually be required for solving the three-dimensional structure of the V-PPase, and the provision of coordinates for modelling the results of mutagenesis. Topological models based on sequence data and the results of protein chemical and mutagenesis experiments provide indications of the gross organization of membrane proteins but they lack the critical third dimension. Moreover, the two-dimensional structures deduced from the application of computer algorithms to sequence data alone can be grossly misleading if hydrophobic stretches of sequence outside the plane of the membrane are associated with the cores of globular, potentially ligandactive, domains (Branden and Tooze, 1991). Similarly, while site-directed mutagenesis is a powerful tool for excluding the participation of specific amino acid residues in catalysis, it is limited by being directed. In lieu of, or pending, three-dimensional structural information, the sustained investigation of alternative selection procedures, organisms or mutant backgrounds for the development of workable screens for the heterologous expression of functional and non-functional enzyme is required. Such screens would provide a means of examining the structural basis of function in a blind, unprejudicial manner, by random mutagenesis. Molecular cloning of the bacterial H+-PPi synthase would undoubtedly facilitate these studies. When account is taken of the ready reversibility of the phototrophic bacterial H+-PPi synthase - its probable Hf:PP, stoichiometry of 2-3 (Baltscheffsky and Baltscheffsky, 1993) - it is conceivable that evolution of the plant V-PPase entailed mutation of one or two H + translocation sites on the bacterial H+-PPi synthase to K+-binding sites. Since the H+-PPi synthase, unlike its plant counterpart, is not K+ activated (Baltscheffsky and Baltscheffsky, 1993), such a scheme is not only capable of explaining the transition from a freely reversible to essentially irreversible H + pump but also the transition from a K+-insensitive to K+-activated pump. Sequence comparisons between the bacterial and plant enzymes may prove instrumental in the localization of sequences associated with homologuespecific functions ( K + activation/translocation in the case of the V-PPase), as well as a source of sequences for independent tests of the identity of residues involved in such shared functions as Mg2+ binding and Mg2PPi hydrolysis. A key area for future physiological investigations will be the study of transgenic plants that either express antisense transcripts (van der Krol et af., 1988) or ectopically overexpress genes encoding the pump (Lagramini et al., 1990). The feasibility and validity of this approach is exemplified by studies of photoassimilate partitioning and metabolite levels in transgenic plants expressing E. cofi soluble PPase (PPA) in the cytosol (Jelitto et al., 1992; Sonnewald, 1992). These studies not only demonstrate that viable plants with diminished cellular PPi levels are sustainable but also that PPi


331

diminution exerts measurable effects on the sucrose synthase pathway (Huber and Akazawa, 1986; Black et al., 1987). There are, however, two significant departures of the work of Jelitto et a f . (1992) and Sonnewald (1992) from direct manipulation of the V-PPase itself. First, kinetic restriction of the V-PPase consequent on ectopic expression of E . cofi P P A is unlikely in the plants characterized by Jelitto et al. (1992) and Sonnewald (1992). Steady state PP; levels were decreased by only 2-3-fold, from a value of 200-300 to 70-150 pM.Since in vitro the V-PPase has a Km(totalpp,) of 15-30 pM (Rea and Poole, 1993), its activity is unlikely to be rate limited in such transformants. By implication, the meaning of the findings of Ellebracht et al. (1994) is ambiguous. These authors note that in transgenic tobacco in which cytosolic PPi levels have been lowered, and they ussume the V-PPase inactivated by ectopic expression of E . coli P P A , the acidification of mesophyll vacuoles which ensues upon the onset of photosynthesis is unaffected. While they conclude that the V-PPase is dispensable for vacuolar acidification, the validity of this conclusion is critically dependent on whether the depression of cytosolic PP, levels is sufficient to stall the pump. Second, the studies of Jelitto et af. (1992), Sonnewald (1992) and Ellebracht et a f . (1994) do not address the central question of which enzyme or enzymes modulate PPi levels in vivo. E . coli P P A is heterologous and, in any case, the endogenous soluble PPase of plant cells is probably sequestered from the cytosol (Weiner et a f . , 1987). Consequently, the studies of P P A , though elegant, do not provide information on the identity of the elements responsible for regulating cytosolic PP, levels in vivo. It is for this reason that we and others are examining other strategies, principally the generation of plants showing diminished or enhanced V-PPase gene expression, for directly evaluating the contribution of this pump to PPi homeostasis and vacuolar acidification in the intact plant.

ACKNOWLEDGEMENTS Many of the studies reported here were supported by grant DE-FGO291ER20055 from the Department of Energy, grant MCB93-05281 from the National Science Foundation and a grant from the University Research Foundation, University of Pennsylvania awarded to P.A.R. Some of this work was also performed under the auspices of a DOE/NSF/USDA Triagency Plant Training Grant (DE-FG02-94ER20162) awarded to the Plant Science Institute. The origin and development of our studies of the V-PPase would not have been possible without the support, stimulation, criticism and/or collaboration of our colleagues Alex Baykov, Natalia Bakuleva, Chris Britten, Darren Chapman, Barry Cooperman, Catherine Darley, Julia Davies, Chris Griffith, Konrad Howitz, Yoncheol Kim, Roger Leigh,

332

R.-G. ZHEN et al.

Ze-Sheng Li, Yuping Lu, Morrie Manolson, Ron Poole, Andy Pope, Dale Sanders, Vahe Sarafian, Heven Sze and Janice Turner, and others of whose input we may not even be aware.

REFERENCES Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D. J. (1990). Basic local alignment search tool. Journal of Molecular Biology 215, 403410. Baltscheffsky, M. and Baltscheffsky, H. (1993). Inorganic pyrophosphate and inorganic pyrophosphatases. In “Molecular Mechanisms in Bioenergetics” (L. Ernster, ed.), pp. 331-348. Elsevier, Amsterdam. Baykov, A. A., Bakuleva, N. P. and Rea, P. A. (1993a). Steady-state kinetics of substrate hydrolysis by vacuolar H+-pyrophosphatase: a simple three-state model. European Journal of Biochemistry 217, 755-762. Baykov, A. A., Dubnova, E. B . , Bakuleva, N . P., Evtushenko, 0. A., Zhen, R.-G. and Rea, P. A. (1993b). Differential sensitivity of membrane-associated pyrophosphatases to inhibition by diphosphonates and fluoride delineates two classes of enzyme. FEBS Letters 327, 199-202. Baykov, A. A , , Kasho, V. N., Bakuleva, N. P. and Rea, P. A. (1994). Oxygenexchange reactions catalyzed by vacuolar H+-translocating pyrophosphatase. FEBS Letters 350, 323-327. Baykov, A. A , , Sergina, N. V., Evtushenko, 0. A. and Dubnova, E. B. (1996). Kinetic characterization of the hydrolytic activity of the H+-pyrophosphatase of Rhodospirillum rubrum in membrane-bound and isolated states. European Journal of Biochemistry 236, 121-127. Black, C. C., Mustardy, L., Sung, S. S., Kormanik, P. P., Xu, D.-P. and Paz, N. (1987). Regulation and roles of alternative pathways of hexose metabolism in plants. Physiologia Plantarum 69, 387-394. Branden, C. and Tooze, J. (1991). “Introduction to Protein Structure.” Garland, New York . Britten, C. J . , Turner, J. C. and Rea, P. A. (1989). Identification and purification of substrate-binding subunit of higher plant H+-translocating inorganic pyrophosphatase. FEBS Letters 256, 200-206. Britten, C. J . , Zhen, R.-G., Kim, E. J . and Rea, P. A. (1992). Reconstitution of transport function of vacuolar H+-translocating inorganic pyrophosphatase. Journal of Biological Chemistry 267, 21 850-21 855. Chanson, A. and Pilet, P. E. (1987). Localization in sucrose gradients of the pyrophosphate-dependent proton transport of maize root membranes. Plant Physiology 84, 1431-1436. Chanson, A. and Pilet, P. E. (1989) Target molecular size and sodium dodecyl sulfate polyacrylamide gel electrophoresis analysis of the ATP-dependent and pyrophosphate-dependent proton pumps from maize root tonoplast. Plant Physiology 90, 934-938. Coen, E. S., Romero, J . M., Doyle, S., Elliot, R., Murphy, G. and Carpenter, R. (1990). Floricaula: a homeotic gene required for flower development in Antirrhinum majus. Cell 63, 1311-1322. Cooperman, B. S. (1982). The mechanism of action of yeast inorganic pyrophosphatase. Methods in Enzymology 87, 526548. Cooperman, B. S . , Chiu, N. Y., Bruckmann, R. H., Bunick, G. J . and McKenna, G. P. (1973). Yeast inorganic pyrophosphatase. I . New methods of purification, assay, and crystallization. Biochemistry 12, 1665-1669.


333

Cooperman, B. S., Baykov, A. A. and Lahti, R. (1992). Evolutionary conservation of the active site of the soluble inorganic pyrophosphatase. Trends in Biochemicaf Sciences 17, 262-266. Davidson, A. M. and Halestrap, A. P. (1989). Inhibition of mitochondria1 matrix inorganic pyrophosphatase by physiological [Ca’+] and its role in the hormonal regulation of mitochondria1 matrix volume. Biochemical Journal 258, 817821. Davies, J . M . , Poole, R. J., Rea, P. A. and Sanders, D. (1992). Potassium transport into plant vacuoles energized directly by a proton-pumping inorganic pyrophosphatase. Proceedings of the National Academy of Sciences of the USA 89, 11 701-1 1 705. Dennis, D. T. and Greyson, M. F. (1987). Fructose 6-phosphate metabolism in plants. Physiologia Plantarum 69, 395404. Ellebracht, A. , Heber, U. and Sonnewald, U. (1994). Light-stimulated proton transport into the vacuoles of leaf mesophyll cells does not require energization by the tonoplast pyrophosphatase. Planta 193, 203-207. Falk, G. and Walker, J. E. (1988). DNA sequence of a gene cluster coding for subunits of the F, membrane sector of ATP synthase in Rhodospirillum rubrum: support for molecular evolution of the F, and F,, sectors. Biochemical Journal 254, 109- 122. Fraichard, A., Magnin, T. and Pugin, A. (1991). Properties and purification of a proton translocating pyrophosphatase in tonoplast of Acer pseudoplatanus. Plant Physiology 96S, 159. Gordon-Weeks, R., Steele, S. H. and Leigh, R. A. (1996). The role of magnesium, pyrophosphate and their complexes as substrates and activators of the vacuolar H+-purnping inorganic pyrophosphatase: studies using ligand protection from covalent inhibitors. Plant Physiology 111, 195-202. Hackney, D. D. (1980). Theoretical analysis of distribution of [I8O]Pjspecies during exchange with water. Application to exchanges catalyzed by yeast inorganic pyrophosphatase. Journal of Biological Chemistry 255, 5320-5328. Hackney, D. D., Stempel, K. E. and Boyer, P. D. (1980). Oxygen-18 probes of enzymic reactions of phosphate compounds. Methody in Enzymology 64, 6&83. Hartmann, E., Rapoport, T. A. and Lodish. H. F. (1989). Predicting the orientation of eukaryotic membrane proteins. Proceedings of the National Academy of Sciences of the USA 89, 5786-5790. Huber, S. C . and Akazawa, T. (1986). A novel sucrose synthase pathway for sucrose degradation in cultured sycamore cells. Plant Physiology 81, 1008-1013. Huttly, A . K . , Plant, A. L., Phillips, A. L., Auffret, A. D. and Gray, J . C. (1990). Nucleotide sequence and transcripts of the pea chloroplast gene encoding CF, subunit 111 of ATP synthase. Gene 90, 227-233. Jefferson, R. A., Kavanagh, T. A. and Bevan, M. W. (1987). GUS fusions: P-glucuronidase as a sensitive versatile gene fusion marker in higher plants. E M B O Journal 6, 3901-3907. Jelitto, T., Sonnewald, U., Willmitzer, L., Hajirezeai, M . and Stitt. M. (1992). Inorganic pyrophosphatase content and metabolites in potato and tobacco plants expressing E . coli pyrophosphatase in their cytosol. Planta 188, 238-244. Johannes, E. and Felle, H. (1989). The role of Mg’+ in proton transport by the tonoplast prophosphatase in vacuoles. Physiologia Plantarum 77, 326-33 1. Johannes, E. and Felle, H. (1990). Proton gradient across the tonoplast of Riccia @tans as a result of the joint action of two electroenzymes. Plant Physiology 93. 412417.

334

R.-G. ZHEN ef al.

Jones, D. T., Taylor, W. R. and Thornton, J. M. (1994). A model recognition approach to the prediction of alpha-helical membrane protein structure and topology. Biochemistry 33, 3038-3049. Kaback, H. R. (1992). In and out and up and down with Lac permease. International Reviews of Cytology 137A, 97-125. Kim, Y., Kim, E. J. and Rea, P. A. (1994a). Isolation and characterization of cDNAs encoding the vacuolar H+-pyrophosphatase of Beta vulgaris. Plant Physiology 106, 375-382. Kim, E. J., Zhen, R.-G. and Rea, P. A. (1994b). Heterologous expression of plant vacuolar pyrophosphatase in yeast demonstrates sufficiency of the substratebinding subunit for proton transport. Proceedings of the National Academy of Sciences of the USA 91, 6128-6132. Kim, E. J., Zhen, R.-G. and Rea, P. A . (1995). Site-directed mutagenesis of vacuolar H+-pyrophosphatase. Necessity of Cys634for inhibition by maleimides but not catalysis. Journal of Biological Chemistry 270, 2630-2635. Knight, H., Trewavas, A. J and Knight, M. R. (1996). Cold calcium signaling in Arabidopsis involves two cellular pools and a change in calcium signature after acclimation. Plant Cell 8, 489-503. Kolakowski, L. F., Schloesser, M. and Cooperman, B . S. (1988). Cloning, molecular characterization, and chromosome localization of the inorganic pyrophosphatase (PPA) gene from S. cerevisiae. Nucleic Acid Research 16, 441-452. Lagrimini, L. M., Bradford, S. and Rothstein, S. (1990). Peroxidase-induced wilting in transgenic tobacco plants. The Plant Cell 2, 7-18. Lahti, R., Pitkaranta, T., Valve, E., Ilta, I., Kukko-Kalske, E. and Heinonen, J. (1988). Cloning and characterization of the gene encoding inorganic pyrophosphatase of Escherichia coli K-12. Journal of Bacteriology 170, 5901-5907. Leigh, R. A., Pope, A. J., Jennings, I. R. and Sanders, D. (1992). Kinetics of the vacuolar H+-pyrophosphatase: the roles of magnesium, pyrophosphate, and their complexes as substrates, activators, and inhibitors. Plant Physiology 100, 1698-1705. Lerchl, J., Konig, S. and Sonnewald, U. (1995). Molecular cloning and expression analysis of isoforms encoding tonoplast-bound proton-translocating inorganic pyrophosphatase in tobacco. Plant Molecular Biology 29, 833-840. Lundin, M., Baltscheffsky, H. and Ronne, H. (1991). Yeast PPA2 gene encodes a mitochondrial inorganic pyrophosphatase that is essential for mitochondrial function. Journal of Biological Chemistry 266, 12 168-12 172. Maddy, A. H. (1976). A critical evaluation of the analysis of membrane proteins by polyacrylamide gel electrophoresis in the presence of dodecyl sulphate. Journal of Theoretical Biology 62, 315-326. Maeshima, M. (1990). Oligomeric structure of H+-translocating inorganic pyrophosphatase of plant vacuoles. Biochemical and Biophysical Research Communications 168, 1157-1162. Maeshima, M. (1991). H+-translocating inorganic pyrophosphatase of plant vacuoles: inhibition by Ca2+,stabilization by Mg2+, and immunological comparison with other inorganic pyrophosphatases. European Journal of Biochemistry 196, 11-17. Maeshima, M. and Yoshida, S. (1989). Purification and properties of vacuolar membrane proton-translocating inorganic pyrophosphatase from mung bean. Journal of Biological Chemistry 264, 20 068-20 073. Manolson, M. F., Proteau, D., Preston, R. A., Stenbitt, A,, Roberts, B. T., Hoyt, M. A., Preuss, D., Mulholland, J . , Botstein, D. and Jones, E. W. (1992). The VPHl gene encodes a 95-kD integral membrane polypeptide required for in


335

vivo assembly and activity of the yeast vacuolar H+-ATPase. Journal of Biological Chemistry 267, 14 294-14 303. Morikama, A. and Nakamura, K. (1987). The pea mitchondrial ATPase subunit 9 gene is located upstream of the ATPase a-subunit gene. Nucleic Acids Research 15, 4692. Nelson, H. and Nelson, N. (1990). Disruption of genes encoding subunits of yeast vacuolar H + ATPase causes conditional lethality. Proceedings of the National Academy of Sciences of the USA 87, 3503-3507. Neuhaus, H. E. and Stitt, M. (1991). Inhibition of photosynthetic sucrose synthesis by imidodiphosphate, an analog of inorganic pyrophosphate. Plant Science 76, 49-55. Nore, B. F., Sakainore, Y., Maeshima, M., Baltscheffsky, M. and Nyren, P. (1991). Immunological cross-reactivity between proton-pumping inorganic pyrophosphatases of widely phylogenically separated species. Biochemical and Biophysical Research Communications 181, 962-867. Nyren, P., Nore, B. F. and Strid, A. (1991). Proton pumping N,N'-dicyclohexylcarbodiimide-sensitive inorganic pyrophosphate synthase from Rhodospirillum rubrum: purification, characterization. and reconstitution. Biochemistry 30, 2883-2887. Nyren, P., Sakainore, Y. and Strid, A. (1993). Amino acid sequence similarities between the vacuolar proton-pumping inorganic pyrophosphatase and the c-subunit of F,,F,-ATPases. Plant and Cell Physiology 34, 375-378. Ohya, Y., Umemoto, N., Tanida, I., Ohta, A., Iida, H. and Anraku, Y. (1991). Calcium-sensitive mutants of Saccharornyces cerevisiae showing a Pet-phenotype are ascribable to defects of vacuolar membrane H+-ATPase activity. Journal of Biological Chemistry 266, 13 971-13 977. Pope, A. J. and Leigh, R. A. (1987). Some characteristics of anion transport at the tonoplast of oat roots, determined from the effects of anions on pyrophosphatedependent proton transport. Planta 172, 91-100. Pugliarello, M. C., Rasi-Caldogno, F., De Michelis, M. I. and Olivari, C. (1991). The tonoplast H+-pyrophosphatase of radish seedlings: biochemical characteristics. Physiologia Plantarum 83, 339-345. Quick, W. P., Neuhaus, H. E., Feil, R. and Stitt, M. (1989). Fluoride leads to an increase of inorganic pyrophosphate and an inhibition of photosynthetic sucrose synthesis in spinach leaves. Biochimica et Biophysica Acra 973, 263-271. Rea, P. A. and Poole, R. J. (1986). Chromatographic resolution of H+-translocating pyrophosphatase from H+-translocating ATPase of higher plant tonoplast. Plant Physiology 81, 126-129. Rea, P. A. and Poole, R. J. (1993). Vacuolar H+-translocating pyrophosphatase. Annual Review of Plant Physiology and Plant Molecular Biology 44, 157180. Rea, P. A. and Turner, J. C. (1990). Tonoplast adenosine triphosphatase and inorganic pyrophosphatase. Methods in Plant Biochemistry 3, 385-405. Rea, P. A., Britten, C. J. and Sarafian, V. (1992a). Common identity of substratebinding subunit of vacuolar H+-translocating inorganic pyrophosphatase of higher plant cells. Plant Physiology 100, 723-732. Rea, P. A , , Britten, C. J., Jennings, I. R., Calvert, C., Leigh, R. A. and Sanders, D. (1992b). Regulation of vacuolar H+-pyrophosphataseby calcium. A reaction kinetic analysis. Plant Physiology 100, 1706-1715. Rea, P. A., Kim, Y., Sarafian, V., Poole, R. J., Davies, J. M. and Sanders, D. (1992~).Vacuolar H+-translocating pyrophosphatase: a new category of ion translocase. Trends in Biochemical Sciences 17, 348-353.

336

R.-G. ZHEN et al.

Sakakibara, Y . , Kobayashi, H. and Kasamo, K. (1996). Isolation and characterization of cDNAs encoding vacuolar H+-pyrophosphatase isoforms from rice (Oryza sativa L.). Plant Molecular Biology 31, 1029-1038. Sarafian, V. and Poole, R. J. (1989). Purification of an H+-translocating inorganic pyrophosphatase from vacuole membranes of red beet. Plant Physiology 91, 34-38. Sarafian, V., Kim, Y., Poole, R. J. and Rea, P. A. (1992a). Molecular cloning and sequence of cDNA encoding the pyrophosphate-energized vacuolar membrane proton pump of Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the USA 89, 1775-1779. Sarafian, V . , Potier, M. and Poole, R. J. (1992b). Radiation-inactivation analysis of vacuolar H+-ATPase and H+-pyrophosphatase from Beta vulgaris L. Functional sizes for substrate hydrolysis and for H+ transport. Biochemical Journal 283, 493-397. Sato, M. H., Maeshima, M., Ohsumi, Y. and Yoshida, M. (1991). Dimeric structure of H+-translocating pyrophosphatase from pumpkin vacuolar membranes. FEBS Letters 290, 177-180. Sato, M. H., Kasahara, M., Ishii, N., Homareda, H. Matsui, H. and Yoshida, M. (1994). Purified vacuolar inorganic pyrophosphatase consisting of a 75-kDa polypeptide can pump H+ into reconstituted proteoliposomes. Journal of Biological Chemistry 269, 672545728. Sipos, L. and von Heijne, G. (1993). Predicting the topology of eukaryotic membrane proteins. European Journal of Biochemistry 213, 1333-1340. Smirnova, I . N., Kudryavtseva, N. A., Komissarenko, S. J., Tarusova, N. B. and Baykov, A. A . (1988). Diphosphonates are potent inhibitors of mammalian inorganic pyrophosphatase. Archives of Biochemistry and Biophysics 267, 280-284. Sonnewald, U. (1992). Expression of E. coli inorganic pyrophosphatase in transgenic plants alters photoassimilate partitioning. Plant Journal 2, 571-581. Springs, B . , Welsh, K.M. and Cooperman. B. S. (1981). Thermodynamics, kinetics, and mechanism in yeast inorganic pyrophosphatase catalysis of inorganic pyrophosphate: inorganic phosphate equilibration. Biochemistry 20, 63846391. Stempel, K. E. and Boyer, P. D. (1986). Refinements in oxygen-18 methodology for the study of phosphorylation mechanisms. Methods in Enzymology 126, 618-639. Taiz, L., Nelson, H., Maggert, K., Morgan, L., Yatabe, B., Taiz, S. L., Rubinstein, B. and Nelson, N. (1994). Functional-analysis of conserved cysteine residues in the catalytic subunit of the yeast vacuolar H+-ATPase. Biochimica et Biophysica Acta 1194, 329-334. Takeshige, K. and Tazawa, M. (1989). Determination of the inorganic pyrophosphate level and its subcellular localization in Chara corallina. Journal of Biological Chemistry 264, 3262-3266. Tanaka, Y., Chiba, K., Maeda, M. and Maeshima, M. (1993). Molecular cloning of cDNA for vacuolar membrane proton-translocating inorganic pyrophosphatase in Hordeum vulgare. Biochemical and Biophysical Research Communications 190, 962-967. Urech, K., Durr, M., Boller, T., Wiemken, A. and Schwencke, J. (1978). Localization of polyphosphate in vacuoles of Saccharomyces cerevisiae. Archives of Microbiology 116, 275-278. van der Krol, A. R., Mol, J. N. M. and Stuitje, A. R. (1988). Antisense genes in plants: an overview. Gene 72, 45-50.


337

van Iwaarden, P., Pastore, J. C., Konings, W. N . and Kaback, H. R. (1991). Construction of a functional lactose permease devoid of cysteine residues. Biochemistry 30, 9.595-9600. Voet, D. and Voet, J . G. (1994). “Biochemistry.” Wiley, New York. von Heijne, G. (1986). The distribution of positively charged residues in bacterial inner membrane proteins correlates with the transmembrane topology. EMBO Journal 5, 3021-3027. von Heijne, G. (1992). Membrane protein structure prediction. Hydrophobicity analysis and the positive-inside rule. Journal of Molecular Biology 225, 487494. Walker, R. R. and Leigh, R. A. (1981). Mg’+-dependent, cation-stimulated inorganic pyrophosphatase associated with vacuoles of red beet (Beta vulgaris L.). Pfanta 153, 1.5@155. Wang, Y. Z., Leigh, R . A., Kaestner, K. H. and Sze, H. (1986). Electrogenic H+-pumping pyrophosphatase in tonoplast vesicles of oat roots. Plant Physiology 81, 497-502. Weiner, H., Stitt, M. and Heldt, H. W. (1987). Subcellular compartmentation of pyrophosphatase and alkaline pyrophosphatase in leaves. Biochimica ef Biophysica Acta 893, 13-21.4. White, P. J., Marshall, J. and Smith, J . A. C. (1990). Substrate kinetics of the tonoplast H+-translocating inorganic pyrophosphatase and its activation by free Mg2+. Plant Physiology 93, 1063-1070. Yamashiro, C. T., Kane, P. M., Wolczyk, D. F.. Preston, R. A. and Stevens, T. H . (1990). Role of vacuolar acidification in protein sorting and zymogen activation: a genetic analysis of the yeast vacuolar proton-translocating ATPase. Molecular and Cellular Biology 10, 3737-3740. Zhen, R. G. and Rea, P. A . (1995). Directed mutagenesis of vacuolar H + PPase discloses potential coupling and DCCD-binding sites. Plant Physiology lWS, 21. Zhen, R.-G., Baykov, A. A., Bakuleva, N. P. and Rea, P. A. (1994a). Aminomethylene diphosphonate as a potent type-specific inhibitor of both plant and phototrophic bacterial H+-pyrophosphatases. Plant Physiology 104, 153-159. Zhen. R.-G., Kim, E. J . and Rea, P. A. (1994b). Localization of cytosolically oriented maleimide-reactive domain of vacuolar H+-pyrophosphatase. Journal of Biological Chemistry 269, 23 342-23 3.50.


The Bioenergetics of Vacuolar H+ Pumps

J . M . DAVIES

Department of Plant Sciences. University of Cambridge. Downing Street. Cambridge CB2 3EA. UK

............................................................................

340

I1 .

Determination of the Coupling Ratio .................................... A . Kinetic Estimates of the Coupling Ratio ................................ B . Thermodynamic Determination of the Coupling Ratio ......... C . Use of Patch Clamp Electrophysiology ..................................

340 340 341 342

111.

Patch Clamp Studies of the V-PPase ............................................ A . The V-PPase and Potassium ................................................ B . Vectorial Activation by Potassium ........................................ C . Modelling the V-PPase as a ( K + M f ) Symporter ..................... D . Observed Reversal Voltage of the V-PPase .... ............ E . Deduction of the V-PPase Couplin F. Validation and Future Directions

343 343 344 346 346 349 350

1V.

Patch Clamp Studies of the V-ATPase ... .............................. A . Isolation of the V-ATPase Pump Current .............................. B . Reversal Voltage of the V-ATPase and Determination of n ...... C . Is the Non-integer Coupling Ratio an Artefact? ...................... D . Mechanistic Implications of the Variable Coupling Ratio ..........

352 352 354 356 357

I.

V.

VI .

Introduction

358 Physiological Consequences ........................................................ A . Acidification by the V-ATPase ............................. ....... 358 B . V-PPase and K+Accumulation ............................................. 359 Conclusions ............................................................................. Acknowledgements ................................................................... References ..............................................................................

Advances in Botanical Research Vol 25 incorporating Advances 111 Plant Pathology ISBN 0-12-005925-8

359 360 360

Copyright 0 1997 Academic Press Limited All rights o f reproduction in any form reserved

340

J. M. DAVIES

I. INTRODUCTION The plant vacuolar membrane contains two H+-translocating pumps; the V-ATPase (EC 3.6.1.3;Liittge and Ratajczak, this volume) and the inorganic pyrophosphatase or V-PPase (EC 3.6.1.1; Zhen et al., this volume). In vivo both are thought to catalyse electrogenic transport of H + from the cytosol to the vacuolar lumen, driven by the free energy available from ATP and PPi hydrolysis, respectively. Their action results in luminal acidification and the establishment of a H + electrochemical potential gradient (APH+; Harold, 1986) across the vacuolar membrane. The energy of that gradient can be harnessed to drive secondary transport systems, coupled to either or both of the gradient’s components (i.e. the membrane voltage, A 9 and the pH gradient, ApH). Vacuolar pH is around 5.5 (e.g. Fox and Ratcliffe, 1990) but is often far more profoundly acidic, especially in fruits (pH 2.7-3.0) and leaves that accumulate oxalic acid (Smith and Raven, 1976). The long-standing question of why two H+ pumps with an ostensibly functional similarity co-reside in one membrane may be answered in part by examining the competence of each pump to acidify the vacuolar lumen. The key thermodynamic determinant of the capacity of a pump for luminal acidification is its operational transport coupling ratio (Lauger, 1991). This is effectively the number of H + ions (n)translocated per ATP (or PPi) molecule hydrolysed. In the literature, the term “stoichiometry” is often used interchangeably with coupling ratio but the two are not necessarily synonymous. The stoichiometry of a pump is equal to the number of bindingtransport sites for the translocated ion (Lauger, 1991), and so will differ from the coupling ratio if the catalytic reaction can proceed with some of those sites unoccupied. For the purposes of assessing the in vivo action of the vacuolar H + pumps, it is the coupling ratio which is of prime importance. In this chapter we shall examine the methods used for coupling ratio determination, work through an experimental study for both the V-PPase and V-ATPase and assess the coupling ratio data by applying them to putative in vivo examples of vacuolar energization.

11. DETERMINATION OF THE COUPLING RATIO A . KINETIC ESTIMATES OF THE COUPLING RATIO

A kinetic estimate of n comes simply from the ratio of observed initial rate of H + flux in a test system to the observed in vitro rate of ATP hydrolysis. The “test system” may be intact, isolated vacuoles, vacuolar vesicles or the purified enzyme reconstituted into liposomes. Methods for determining rates of H+flux include simple pH measurements of the reaction medium, nuclear

THE BIOENERGETICS OF VACUOLAR H+ PUMPS

341

magnetic resonance and ApH formation assessed using quenching of intravesicular H+-sensitive fluorescent dyes. The consensus from kinetic estimates is that for the V-ATPase, n has a value of around 2 (e.g. Bennett and Spanswick, 1984; Schmidt and Briskin, 1993) while that of the V-PPase is around 1 (e.g. Johannes and Felle, 1990; Schmidt and Briskin, 1993). Schmidt and Briskin (1993) have estimated that for these values of n , the maximum ApH that could be generated under putative in vivo conditions would be 5.0-5.4 by the V-ATPase and 3.6-5.8 by the V-PPase. However, such estimates of n are essentially indirect and subject to several sources of experimental error. Unquantified H + leakage (i.e. passive H + movement down AjiH+) will result in an underestimate of n. Likewise, uncoupling of the enzyme, with ATP hydrolysis proceeding in the absence of H + translocation, will also give an underestimate. Moreover, assays are usually only conducted at their optimal pH; for example, V-ATPase hydrolysis is measured at pH 6 whereas ApH formation is assayed at pH 8, yet such results are combined to deduce n. Overall, kinetic studies provide good "ball park" estimates of n , but a more accurate determination can be achieved using a thermodynamic approach which exploits the electrogenicity of the pumps.

B. THERMODYNAMIC DETERMINATION OF THE COUPLING RATIO

The theory underlying the thermodynamic determination of n has been described thoroughly by Rea and Sanders (1987). Briefly, we may describe the action of, for example, the V-ATPase thus;

n[H'Ic + ATP n[H'lV + ADP + Pi (1) where n is the coupling ratio of H+ translocated per ATP hydrolysed, the square brackets denote activities and the subscripts c and v refer to the cytosolic and vacuolar compartments, respectively. The free energy relationship of this reaction is given by AG = nF A 9 - RT In ([ADP][Pi][H+]vn/KATP [ATP][H+],")

(2)

where A 9 is the membrane voltage, KATp is the equilibrium constant for ATP hydrolysis and R , T and F have their usual meanings. The value of KATP can be calculated for given values of [H+],, [ATP], [ADP] and [PJ, using the dissociation constants of the reactants (Davies et al., 1993). For the special case of the pump at equilibrium (when AG is equal to zero), Equation (2) can be rearranged: A9

= (R T/nF) In

([ ADPI[ Pi][H+],"/KATdATP][ H+],")

(3)

The significance of Equation (3) is two-fold. Firstly, at equilibrium the pump

342

J . M. DAVIES

will pass no net current; it is poised between its forward (hydrolytic) mode (ATP hydrolysis coupled to H+ translocation into the vacuolar lumen), and its reverse (synthetic) mode (ATP synthesis coupled to H+ translocation into the cytosol). The membrane voltage at which the pump is at equilibrium is therefore termed its reversal voltage (Ere,), for the pump can reverse its poise around this point. Secondly, the Ere, of a pump is a measurable parameter; direct determination of Ere, is possible using patch clamp electrophysiology . Observed Ere, values may be substituted into the equation

Ere, = (RT/nF) In ([ADP][Pi][H+]VR/KATP[ATP][H+]cn) (4) to deduce the value of n for the experimentally imposed transmembrane H+ gradient and activities of ligands used to generate the pump current. How is this achieved? C. USE OF PATCH CLAMP ELECTROPHYSIOLOGY

To measure the Ere, of a pump requires the “whole vacuole” patch clamp configuration (Hedrich et al., 1988); the turnover of an individual pump is too low to resolve and so the activity of the entire population in the membrane must be measured. After establishment of a high electrical resistance seal between the patch microelectrode and the membrane of an isolated, intact vacuole, voltage pulses and suction are used to break the membrane held in the tip of the electrode. The defined contents of the electrode then diffuse into the lumen and place the membrane in electrical series with the electrode. As the isolated vacuole is bathed in a defined medium (analogous to the cytosol), the experimenter now has control of the ionic conditions at both faces of the vacuolar membrane. As the configuration also allows a set A 9 to be imposed, control of both components of APH+ can be achieved. To determine Ere,, the current-voltage (I-V) relationship of the pump at apparent steady state must be measured. Firstly, the I-V of the unenergized membrane is constructed by clamping the membrane over a set voltage range and measuring the current which is passed at each discrete voltage. If no H+ leak is observed, the pump is activated by superfusing the vacuoles with set concentrations of (for the V-ATPase) ATP, ADP and Pi (added simultaneously to permit the pump to run in both forward and reverse modes). A second I-V can then be constructed during continued superfusion with ligands (their activities can therefore be considered to be static). The difference between the energized and unenergized I-V relationships is taken to be that generated by the pump. On the difference curve, the voltage at which the current intersects the voltage axis is the voltage at which the pump is conducting no net current and that intercept is therefore the Ere, for those conditions (for example, see Fig. 5).


343

Patch clamp can therefore provide a direct and high-resolution assay of the Ere, of the current generated by the pump, which in turn can be used to deduce a value for n . It can also be used to examine other suspected properties of translocation by the electroenzyme - for example, the possibility that the V-PPase may translocate K+ in addition to H + . A note on polarity convention. The polarity convention for endomembrane voltage changed in 1992 (Bert1 et a f . , 1992); vacuolar A’P is now defined relative to the lumen (i.e. Pc- q,).Thus, vacuolar A* values reported prior to 1992 are positive (e.g. + 20 mV; Spanswick and Williams, 1964) but the polarity is now expressed as a negative value. With respect to the depiction of pump Z-V relationships, positive charge (e.g. H+) passing from the cytosol to the vacuolar lumen is termed “outward” (as the lumen is considered electrically equivalent to the extracellular space) and is represented in the upper half of the Z-V graph, above the voltage axis. Positive charge passing from the lumen to the cytosol is termed “inward” and is plotted below the voltage axis. The new polarity convention has been adopted here throughout.

111. PATCH CLAMP STUDIES OF THE V-PPase A. THE V-PPase AND POTASSIUM

A key characteristic of the V-PPase is its absolute dependence on K+ for both hydrolytic activity and H + pumping (e.g. Wang et al., 1986; White et al., 1990). Such dependency may not be simply the manifestation of enzyme regulation by K+ , but rather could be indicative of K+ translocation by the V-PPase. Certainly there is a requirement for active (i.e. energy-consuming) K + transport into the vacuolar lumen. Cytosolic K+ ([K+],) is thought to be maintained homeostatically around 80-100 mM (Maathuis and Sanders, 1994; Leigh and Wyn Jones, 1984) whereas vacuolar K+ ([K+],) varies, dependent on external K+ supply and cell type (Leigh and Wyn Jones, 1984; Malone et al., 1991). Under K+-replete conditions, [K’Iv is typically around 200mM and can be as high as 500mM in open stomata1 guard cells (MacRobbie, 1988). For argument’s sake if we take [K+], as 100mM and [K’Iv as 200mM, and as the in vivo tonoplast membrane potential is -20 to -50mV (Bates et al., 1982; Spanswick and Williams, 1964), then K+ transport into the lumen would be against an electrochemical potential gradient of 4-7 kJ. mol-l. Establishment and maintenance of such a K+ gradient could be facilitated by a (K+/H+) antiporter driven by A/ZH+. A highly specific K+/H+ vacuolar antiporter (with a putative 1:l coupling) has been described from membrane vesicles of Brassica napus hypocotyls (Cooper et al., 1991). However, as its in vitro activity is severely inhibited

344

J . M. DAVIES

by K+ concentrations over 25 mM, which would effectively render it inactive in vivo, its ability to facilitate vacuolar K+ accumulation appears compromised. Could the V-PPase drive luminal K + accumulation by acting as a (K+/H+) symporter, thus translocating both ions into the lumen? The first, simple test of this hypothesis is to examine the vectorial activation of pump activity by K+ . B. VECTORIAL ACTIVATION BY POTASSIUM

If the V-PPase were to act as a (K+/H+) symporter, it would be expected that the K+-binding site which stimulates the hydrolytic pump mode (i.e. PPj hydrolysis coupled to H+ translocation from cytosol to lumen) would be present at the cytosolic membrane face. Conversely, the synthetic pump mode (Pi synthesis coupled to H+ extrusion from the lumen) would be activated by K+ binding at the luminal face. These predictions have been tested using whole-vacuole patch clamp recordings of V-PPase ion translocation in isolated individual vacuoles from B. vulgaris root storage tissue (Davies et al., 1991,1992) and Chenopodium rubrum cell suspension cultures (Obermeyer et al., 1995). In Beta, with choline (which does not stimulate activity; Wang et al., 1986) substituting for K+ at both luminal and cytosolic faces, 100 pM cytosolic PPi failed to activate V-PPase activity of a vacuole (Fig. 1A). However, when K+ was then introduced selectively at the cytosolic face of the same vacuole, the same PPj concentration induced an outwardly directed current consistent with the entry of positive charge into the lumen (i.e. V-PPase operating in the hydrolytic mode; Fig. 1A). The mean PPi-dependent current (3.0 mA. m-') approximated that obtained with K+ present at both membrane faces (3.2mA. m-2). Such activity could not be induced when K+ was present only in the lumen, with choline at the cytosolic face (Davies et al., 1991). These results have been confirmed and extended in the more advanced study with Chenopodiurn. Obermeyer et al. (1995) have also reported a monovalent cation selectivity series for the PPz-dependent current, which suggests that Rb+ can adequately substitute for K+. The reciprocal experiments in which Pi at the cytosolic face should drive the reversal of the V-PPase indicate that for both Beta and Chenopodium, Pi can only induce an inwardly directed current (positive charge flowing from the lumen to the cytosol) when K + is present at the luminal face (Fig. 1B; Davies et al., 1992; Obermeyer et al., 1995). This current is indifferent to [K'],. If we refute such a current as being generated by the V-PPase, then we must invoke the existence of a [K+],-dependent Pi translocator. At present, however, the weight of evidence supports a freely reversible V-PPase, with a pattern of vectorial K+ activation entirely consistent with the action of a (K+/H+) symporter. This simplistic approach fails to distinguish between kinetic activation by K+ and K+ translocation. To make

345


A 50 ch'

+

PPi

0 50 ch'

50 K'

+

PPi

50 ch'

0.5 pA

L 10s 2Pi B

25

1

Fig. 1. Vectorial activation of the V-PPase by K + . (A) Current traces from a single Beta vacuole containing 50 mM choline (ch+) chloride, pH 5.5. With choline in the bathing medium, addition of 0.1 mM PP, failed to elicit a pump current. Using the same vacuole, when K + replaced choline at the cytoplasmic face, addition of PP, generated a current consistent with the translocation of positive charge into the vacuolar lumen. (B) Left panel: whole-membrane I-V relationships recorded from a single Beta vacuole containing 100 mM KCI and bathed in 100 mM choline chloride ( 0 , control; 0 , plus 10mM Pi at the cytoplasmic face). Right panel: I-V difference relationship derived from the whole vacuole recordings; the current represents PP,-dependent expulsion of positive charge from the lumen. (Data taken from Davies et al. (1991, 1992), with kind permission of the authors, and redrawn to the new polarity convention. )

346

J. M. DAVIES

that distinction requires an analysis of the effects of K+ on the thermodynamics of pump action. C. MODELLING THE V-PPase AS A (K+/H+) SYMPORTER

It has been argued in Section I1.B that the membrane voltage at which a pump reaction is at equilibrium is a function of the prevailing gradient of the translocated ion. Equation (1) can be rewritten for the V-PPase and expanded to describe concomitant and unidirectional K+/H+ translocation: n[H'Ic

+ m[K+], + PPi

n[H'],

+ m[K'], + 2Pj

(5) where n is now the coupling ratio for H+:PPi and m that for K+:PPi, with all other terms having the same meaning as given in Section 1I.B. The reversal voltage (Ere,) of such a V-PPase symporter then becomes: Ere, = RT/(n + m)F In ([Pi]*[H']," [K+lVmIKppi [H+]," [K+],") (6) where Kppi is the equilibrium constant for PPi hydrolysis, which can be calculated for given ionic conditions (Davies et al., 1993). Ere, is therefore a function of both ionic gradients (which can be manipulated experimentally); its value can only be affected by [K+] if the latter is translocated rather than being merely stimulatory. The absolute value of Ere, and its change in response to altered ionic gradients can be used to ascertain values of n and m for the pump. Using Equation (6) we can predict for example that for a set pH gradient and [K+],, increasing [K'Ic should result in a negative shift of the observed E,,,. D. OBSERVED REVERSAL VOLTAGE OF THE V-PPase

To date, V-PPase Ere, values have only been determined from Beta vacuoles (Davies et al., 1992). In a series of experiments, the I-V relationships of the V-PPase were resolved (using the approach described in Section 1I.C) with the pump activated by fixed concentrations of PPi (0.1 mM) and Pi (10 mM) Fig. 2. Effect of [K+], on the I-V relationships of the V-PPase. (A) Control ( 0 ) whole membrane I-V from a single Beta vacuole containing 30 mM KCl (pH 6.23) and bathed with 30 mM KCI (pH 8.30); ( 0 ) I-V in the presence of 0.1 mM PPi and 10 mM Pi at the cytoplasmic face. (B) Whole membrane I-V relationships from the same vacuole as (A), with 100 mM KCI (pH 8.16) at the cytoplasmic face (0,control; 0 , plus (PPi/Pi)). (C) I-V difference relationships derived from (A) and (B) for the (PP/Pi)-dependent current at [K+], 30 and 100mM. The observed E,,, values are the zero-current intercepts. As [K'], i-xreases, E,,, shifts negatively. (Data taken from Davies et al. (1992), with kind permission of the authors, and redrawn to the new polarity.)


A

PPi+Pi

control -50

-40

1

5

-40

C

2o

-20

PPi+Pi 0

I 1

i

347

348

J. M. DAVIES

added simultaneously to the cytoplasmic face. The V-PPase could therefore be driven in both hydrolytic and synthetic modes. Starting with a fixed trans-tonoplast K+ or H + gradient, the V-PPase I-V was recorded then determined again after the simple expedient of altering the ionic gradient by changing the bathing solution. An example of such an experiment is shown in Fig. 2, where (with a fixed p H gradient) [K'Ic was raised from 30 to 100 mM. The V-PPase I-V difference relationships clearly show the negative shift in Ere, predicted by Equation (6) for an increase in [K+],. Overall, absolute values of Ere, and shifts in Ere, were dependent on the magnitude of the K+ gradient (implicit in Equation (6)) rather than absolute values of [K+] on each side of the membrane. These effects were independent of the balancing anion used (Cl- o r gluconate-). The analogous experiments in which the [K'] gradient was fixed but the trans-tonoplast H+ gradient was altered also showed that ApH had a direct impact on E,,,, consistent with the predictions of Equation (6). That the observed Ere, values of the V-PPase current obey the predictions of Equation (6) (with Ere, responding to both H + and K+ gradients) argues strongly for the V-PPase as a symporter of both ions. Moreover, a number of alternative explanations for the effects of H + and K+ can reasonably be rejected at this stage. For example, if it were the case that, as Ere, responds to [K+], the V-PPase acts as a "pure" K+ pump, then previously observed PPi-dependent luminal H + accumulation (e.g. Johannes and Felle, 1990) could be accounted for by invoking an indirect H + transport system. The V-PPase would generate a trans-tonoplast K+ gradient which could drive a K+/H+ exchange (activated by PPi) to mediate luminal acidification. This hypothetical system may effectively be discounted as Ere, responds to [H+Ic as it does to a change in [K+],. Similarly, the same argument can be used against a V-PPase H + pump coupled to a PPi-stimulated K+/H+ exchange to facilitate K+ accumulation; the counterion of an exchanger should have no effect on the Ere, of the primary pump. It should also be noted that the values of [K'], used in the patch clamp study were sufficiently high to inhibit the (K+/H+) antiporter reported by Cooper et al. (1991). The possibility that PP{Pi activates a population of K+ channels must also be considered - such activation would compromise the validity of the observed E,,, values. The I-V subtraction procedure precludes the response of K+ channels to the experimental changes in [K+JCand [H'Ic but the resultant I-V difference relationship (the (PPi/Pi)-dependent current) may well have a component attributable to PPi/Pi activation of the macroscopic K+ channel current. However, I-V difference relationships obtained from Beta by either solely PPi or Pi activation with K+ symmetrical across the membrane fail to exhibit a reversal voltage. The current shows a low conductance, running effectively parallel to the voltage axis. This implies that such I-V relationships represent only pump activity. Were either ligand to activate a population of K + channels then one would anticipate their

349

THE BIOENERGETICS OF VACUOLAR H + PUMPS

interference to manifest in the difference I-V as the channel current reverses around the equilibrium voltage for K+ (EK) to oppose the polarity of the pump current. This has not been observed for PPi or Pi activation, which suggests that vacuolar K+ channels are not gated by either ligand. The Ere, values and shifts in Ere, (AE,,,) of the putative V-PPase pump current do not correspond with the predicted EK of the imposed K+ gradients. For example, in the case where K+ is symmetrical across the membrane at 100 mM (EK = 0 mV), the mean measured E,,, of the (PPl/Pi)-activated current is -7mV; when [K’Ic is changed to 30mM (EK= + 31 mV) the mean Ere, is only +I1 mV (Davies et a l . , 1992). That the full Nernstian shift is not observed when the K+ gradient is changed is not sufficient reason to discount the existence of (PPi/P,)-activated channels, but the results of the vectorial K+ activation experiments render that possibility unlikely. A more rigorous rebuttal would be provided by the further identification of the “true” V-PPase pump current by judicious use of inhibitors, as has been put into effect for identification of V-ATPase activity (see Section 1V.B). With this possible source of error in mind, the use of the V-PPase Ere, values to deduce the K+:H+ coupling ratio of the pump will be examined.

E. DEDUCTION OF THE V-PPase COUPLING RATIO

Substitution of the observed Ere, and AE,,, values from the Beta experiments into Equation (6) (with K , , values specific to each experimental condition) enabled values of n and m to be tested for goodness of fit (Davies er al., 1992). Neither Ere,nor AE,,, values were consistent with the V-PPase acting solely as a K+ pump or, more importantly as a “pure” H + pump. Moreover, substitution of integer values of n and m into Equation (6) failed to generate Ere, values in agreement with those obtained experimentally. However, with an overall ( n + m) integer coupling ratio of 3, non-integer values for n and m could predict both observed Ere, and AE,,, values adequately. Best fits to data were obtained with a coupling ratio of 1.3H+:1.7K+:PPi. An example of observed versus predicted Ere, and AE,,, for this coupling ratio is shown in Fig. 3. Without testing a greater range of K + and H + shifts, little can be said on the possible catalytic mechanism of a V-PPase symporter. However, the proposed non-integer coupling ratio values might be explained by translocation of a constant number (three) of positive charges per catalytic cycle. Of the three bindinghranslocation sites, at least one must translocate K or H+ , and competition between K+ and H + for the third site would lead to the phenomenological non-integer ratios. Some support for this explanation comes from the finding that for Beta vacuolar vesicles there is a linear relationship between the K , for K+ activation of hydrolysis and H+ concentration (Davies et al., 1994b). +

350

J. M. DAVIES

Erov

A

25

B

AErov A B C

> E

I

>

!!

4

0.

L

0

A: Observed value B: 1.3H+:1.7 K+ C: 1H'

m

r W

-50

.

Fig. 3. Comparison of (A) observed absolute Ere, values and AE,,, with (B) calculated values for a 1.3 H+:1.7 K+ V-PPase symporter and (C) a 1 H+:PPi V-PPase H+ pump. Data are mean values taken from experiments performed using Beta vacuoles: [K+], 100 mM, pH, 6.25; initial [K'Ic 100 mM, final [K+], 30 mM; initial pH, 8.22, final pH, 8.17. (Values taken from Davies et al. (1992), with kind permission of the authors.)

F. VALIDATION AND FUTURE DIRECTIONS

Biochemical validation studies have thus far refuted K+ translocation by the V-PPase (Sato et al., 1994; Ros et al., 1995), but may have been thwarted by the inherent leakiness of vesicles to K+. Sat0 et al. (1994) reconstituted the purified V-PPase of pumpkin hypocotyls into proteoliposomes, and failed to detect PPrdependent 42K+uptake; the time-courses and saturation values of intravesicular K+ (although not presented) were described as being identical in the presence or absence of PPi. This is curious given that, if the V-PPase were only an H + pump, H + pumping would generate a A* which would increasingly oppose passive K+ uptake, resulting in an altered time-course. In the study of Ros et ul. (1995), K+/H+ fluxes and A* in vacuolar vesicles from Vitis viniferu were measured using fluorescent probes. Again, PPi did not stimulate intravesicular K + accumulation (with K+ starting symmetrical at 10mM across the membrane). However, the A T generated was over 200mV, which, in addition to being an order of magnitude higher than that thought to obtain in vivo, would drive Kf out

THE BIOENERGETICS OF VACUOLAR Hf PUMPS

351

of the lumen passively, perhaps masking translocation into the lumen by the V-PPase. When A!€' was clamped at zero by FCCP, PP; failed to activate K+ accumulation; that the further addition of valinomycin (as a K+ ionophore) had no effect on intravesicular K+ suggests that the system was already at equilibrium, perhaps through the activity of native Kf channels. When all the available data point to the action of an enzyme as an Hf pump, we conclude by simple induction that it is an H + pump. When data (if sound) refute that conclusion, it is n o longer logical to make that induction - an alternative mode of action for that enzyme must be considered for it is no longer an H + pump. Here we have the crux of the problem for the V-PPase: are the patch clamp experiments technically sound and can they be validated? There are a number of pressing problems which must be solved before the case for the V-PPase as an (Hf/K+) pump could reasonably be accepted. Firstly, in any future patch clamp validation of the original study, the V-PPase pump current must be distinguished from any possible (PPi/Pi)activated K+ current to enable accurate determination of Ere, of the enzyme. This could be achieved by inhibiting specifically V-PPase activity, allowing resolution of an inhibitor-sensitive Z-V relationship, which should represent that of the enzyme itself. Pyrophosphate analogues such as the 1,ldiphosphonates are now known to be potent and specific V-PPase inhibitors (Zhen et al., 1994). Of these, aminomethylenediphosphonate may be appropriate for such studies, with an apparent Ki of 1 . 8 p M (Zhen et al., 1994). Secondly, it would be preferable if a single isoform system were studied by patch clamp; since the original study by Davies et al. (1992) it has been found that the Beta genome contains at least two genes encoding V-PPase (Kim et al., 1994a). The expression patterns of such possible isoforms have yet to be determined, and the catalytic properties of the gene products are unknown. Clearly, the possibility that the patch clamp measurements were from a dual-isoform V-PPase population with each isoform having potentially different translocation characteristics cannot yet be dismissed. However, this problem may be circumvented by expression of Arabidopsis V-PPase (known to be encoded by a single gene; Sarafian el al., 1992) in a suitable system. Kim et al. (1994b) made a notable advance by their expression of Arabidopsis cDNA (AVP) encoding the substrate-binding subunit of the V-PPase in the vacuolar membrane of Saccharomyces cerevisiae. As the heterologously expressed subunit exhibited the same hydrolysis, translocation and Kf activation characteristics of the native enzyme (demonstrating that this single polypeptide is a transport-competent pump) and as the yeast membrane appears to lack V-PPase activity, this can be considered a valid system for patch clamp studies. Moreover, it lends itself to the expression of site-directed AVP mutants, thus allowing the identity of putative K+ bindinghranslocation sites to be explored. Further validation of K+ translocation must come from biochemical

352

J. M. DAVIES

studies. Here an apparent impasse has already been reached for there has been no successful demonstration of PP,dependent K+ translocation by the reconstituted V-PPase.

IV. PATCH CLAMP STUDIES OF THE V-ATPase A.

ISOLATION OF THE V-ATPase PUMP CURRENT

Measurement of V-ATPase activity using the whole-vacuole patch clamp mode is complicated by the co-residence of the V-PPase. To apply the thermodynamic approach outlined in Section 1I.B and use the observed Ere, to deduce the H+:ATP coupling ratio II (from Equation (4)) requires a demonstration of free reversibility of the pump. Yet it has been shown by the work on the V-PPase that addition of Pi to the cytoplasmic membrane face elicits V-PPase activity. Thus, (ADP/Pj) activation of the V-ATPase synthetic mode (H+ translocation from the lumen to the cytosol) would occur concurrently with Pt-dependent V-PPase activity if K + were present in the vacuolar lumen. Such joint activation would not only give false measurements of V-ATPase synthetic activity but also distort Ere, values for that pump. The simplest solution to this problem has been to remove K+ completely from the recording system and replace it with choline - the V-PPase is thus inactivated. However, isolation of the V-ATPase I-V relationship may be further confounded by nucleotide gating of channels. There are sufficient precedents from animal channel studies for potent ATP and ADP regulation, and it is increasingly recognized that ATP may gate plant channels (e.g. Spalding and Goldsmith, 1993; Davies and Sanders, 1995). To obviate this potential problem, the I-V relationship of the Beta V-ATPase has been isolated using bafilomycin Al as a potent and specific inhibitor of V-type ATPases (Bowman et al., 1988). This approach simply adds another stage to the protocol for determining the pump I-V relationship. After recording the control whole-membrane I-V, the membrane is energized by set concentrations of ATP/ADP/Pi, and a second I-V is taken. In the continued presence of the nucleotide/Pi ligands, bafilomycin is added at a saturating concentration (600nM: Davies et al., 1994a) to inhibit the V-ATPase, and a final I-V is recorded. The V-ATPase I-V difference relationship is derived

Fig. 4. I-V relationships of the hydrolytic (pH, 8.0, pH, 5.5) and synthetic (pH, 7.6, pH, 4.8) modes of Beta V-ATPase. (A) Whole-membrane I-V relationships from two separate vacuoles with pump activation by 5mM ATP ( 0 ) or 5mM ADP and 10mM Pi (A). Solid symbols: additional presence of 600nM bafilomycin. (B) I-V difference relationships from (A) for the bafilomycin-sensitive (i.e. V-ATPase) current. (Reproduced from Davies et al. (1994a), with kind permission of the National Academy of Sciences, USA.)


353

A

-1

B

100 (Y

I

E

8

-E.

Hydrolytic mode

I

r'/ _ _ _ _ _ _ _ _ _ _ _ - - - _- -_- _- -_-_ _ _ _ - - - - - - I

I

-100

1

100

...... ......

........ ........

......

..... ........... ..............

4

Synthetic mode

-100

-

V,mV

354

J . M. DAVIES

by subtraction of the “bafilomycin-inhibited” Z-V from the “energized” Z-V. An example of this procedure is shown in Fig. 4, where the hydrolytic and synthetic V-ATPase modes have been isolated by their response to bafilomycin (Davies et al., 1994a). This also demonstrates that the pump is freely reversible in the choline-recording solutions and shows a slight voltage sensitivity.

B. REVERSAL VOLTAGE OF THE V-ATPase AND DETERMINATION OF n

Use of bafilomycin to isolate the V-ATPase pump current of a single vacuole prevents subsequent recordings of activity after exchange of bathing solution. In practice, it is not possible to wash out the inhibitor, and so only absolute Ere, values may be determined, not shifts. Despite this limitation it has been possible to explore the effects of the trans-membrane pH gradient on the Ere, of Beta V-ATPase by considering mean absolute values (bafilomycin inhibition of V-ATPase hydrolytic activity is pH-insensitive; Davies et al., 1994a). Figure 5 shows the whole-membrane Z-V and derived bafilomycinsensitive I-V relationships obtained from a single vacuole, with pH, set at 7.6 and pH, at 4.8 (ligands (in mM), 5 ATP, 5 ADP, 10 Pi). These pH values are close to those thought to obtain in the majority of plant cells in vivo. The observed Ere, in this example yields an n value (from Equation (4)) of close to 3, which is at odds with the coupling ratio of 2 frequently reported from vesicle studies (see Section 1I.A). However, it is in close agreement with the finding of n = 3 from turtle bladder V-ATPase, determined by short-circuit analysis of pump current at pH, 7.4 (Dixon and Al-Awqati, 1980). Moreover, mean Ere, values from Beta vary with the imposed pH gradient; the results of their substitution into Equation (4) to determine n are presented in Fig. 6. As with the V-PPase, we are faced with the phenomenon of non-integer values of n but, more strikingly, n is variable. There is a marked decrease of n as the cytosolic H+ concentration decreases at a given pH,; when pH, is constant, n decreases as the luminal H’ concentration increases. Thus it appears that n is governed by a mass action effect of the translocated ion. The physiological implications of this observation are discussed in Section V.A.

Fig. 5. Determination of Beta V-ATPase E,,,. (A) Whole-membrane Z-V relationships from a single vacuole, with the V-ATPase activated by (in mM) 5/5/10 ATP/ADP/Pi, respectively ( o ) , then inhibited by 600nM bafilomycin (a). (B) Z-V difference relationship from (A) for the bafilomycin-sensitive current. The E,,, corresponds to a coupling ratio of close to 3 for these conditions. (Reproduced from Davies et al. (1994a), with kind permission of the National Academy of Sciences, USA.)

355


200 -

B

N

0

E

a

-E,

-

E,."

c

c

r c

- 0 -

j~..-O;=e=O 4 -

I

I

I

I

_ - - ----I

c c

60

-60 ----_____---

n=3

V,mV

(n=2, Ere"= -73 mV)

-200 -

J . M. DAVIES

356

7

7.5

8 PHC

Fig. 6. Summary of Beta V-ATPase deduced coupling ratios as a function of pH,; data are mean values (SEM less than 0.12) obtained at pH, 4.32 ( 0 ) and 4.8 (a). (Data reworked from Davies et al. (1994a), with kind permission of the National Academy of Sciences, USA.)

C. IS THE NON-INTEGER COUPLING RATIO AN ARTEFACT?

The 16 kDa V-ATPase subunit (subunit c; the putative H+ channel of the enzyme) is known to be encoded by at least four different genes in oat (Lai et al., 1991). Moreover, recent studies by Kramer et al. (1995) on the V-ATPase of Mesembryanthemum have shown that the V1 catalytic head structure can exist as a hexamer (three copies each of the catalytic A subunit and B regulatory subunit; see also Luttge and Ratajczak, this volume) or as a pentamer (missing a copy of either an A or a B subunit). The possibility remains therefore that in Beta the absolute Ere,values have been determined from a heterogenous population of isoforms, with each isoform having a different but integer n yielding an overall non-integer value. This question could perhaps be resolved experimentally by studying a system known to have a single isoform. Yeast seems ideal for this purpose in that its genome contains only single copies of the genes encoding the c subunit and the B subunit of the catalytic sector (Nelson and Nelson, 1989; Nelson et al., 1989). It is critical that further work on the V-ATPase should aim to resolve this issue.


357

D. MECHANISTIC IMPLICATIONS OF THE VARIABLE COUPLING RATIO

If we accept the premise that the variable and non-integer values of n from Beta resulted from the action of a single enzyme isoform, then we can make some initial observations on the possible mode of action of the enzyme. As the concentration of cytosolic H + decreases, fewer ions are available for transport and so n decreases. But this facile observation that pH, will determine the extent of occupancy of the binding sites incorporates an implicit assumption that the enzyme will be catalytically competent even when some sites are empty. The enzyme must therefore execute a “slip” reaction, where ATP is hydrolysed without the translocation of the maximum number of H + that could be bound (Lauger, 1991). The concept of slip is one that has received relatively scant attention and suffers at present from a lack of formalized definition and treatment. What is clear, however, is that any slip reaction must be strictly regulated to maximize efficiency. For the Beta V-ATPase data, assessment of the slip reaction is effectively impeded by our lack of knowledge on just how many H + binding sites there are (i.e. the stoichiometry is unknown). Given that the reaction mechanism is also an unknown, it would be unsafe to conclude that the stoichiometry is simply the next highest integer to the maximum observed non-integer value of the coupling ratio. For example, coupling ratios of around 10 have been recorded for bacterial F-ATPases (which are structurally related to the V-ATPases); observed values differ greatly with experimental conditions, with seemingly no upper limit for n (e.g. van Walvaren et al., 1986). The data from Beta would put the stoichiometry at 4, as the maximum observed n was 3.28 (Fig. 6). With so many unknowns there is no recourse but to extract as much information out of the available data as possible. The Ere, of the V-ATPase current alone yields little insight on reaction mechanism, but the I-V relationship as a whole holds information on such vital parameters as the voltage sensitivity of charge translocation. To subject these data to reaction kinetic modelling affords an opportunity to deduce likely characteristics of the translocation process (the reader is referred to Hansen et al. (1981) for further information on the theory and practice of this type of analysis). Initial attempts to model the V-ATPase I-V relationships (Davies et al., 1996) suggest that an enzyme binding three H + ions could support the observed n values. This is provided that in addition to a charge translocation step which allows passage of all three H+ ions, there is a parallel slip translocation which would permit only one H+ ion to be transported per ATP molecule hydrolysed. This model can adequately describe the I-V relationships obtaincd when luminal pH is constant but pH, varies; it does not hold for the converse. The effect of pH, on n awaits more detailed analysis, but it seems likely that it would be exerted through the dissociation constants of the luminal face binding sites. Overall, at present the variable coupling ratio

358

J . M. DAVIES

of the V-ATPase presents an interesting phenomenon but one that awaits a mechanistic context!

V.

PHYSIOLOGICAL CONSEQUENCES

Although there are experimental pitfalls and weaknesses in both the kinetic and thermodynamic determination of the pump coupling ratio, it is useful to assess the consequences of the results through their predicted impact on vacuolar energization. However, construction of putative in vivo situations is limited and rendered crude by lack of information on, for example, the range of membrane voltage and the electrical coupling of the various transport systems in the vacuolar membrane. A.

ACIDIFICATION BY THE V-ATPase

The proposed overall coupling ratio of 3 for the V-PPase places a strict thermodynamic limit on its ability to acidify the vacuolar lumen. It is therefore rational to estimate the maximal acidification which could be achieved by the V-ATPase and assess how such values of pH, would affect the poise of the V-PPase. By setting AG to zero in Equation (2), the maximum pH, (i.e. the [H'], which would bring the V-ATPase to equilibrium if all other parameters remain constant) can be deduced. Equilibrium constants for ATP and PPi hydrolysis (KATP and Kpp,, respectively) under putative in vivo cytosolic conditions have been estimated here for 20°C according to Davies et al. (1993), using the following parameters (in mM): free cytosolic Mg2', 0.4 (Yazaki et al., 1988); free Ca2+, cytosolic, 0.0002; [PPJ, [ATP], [ADP] 0.25, 2.3 and 0.31, respectively (Weiner et af., 1987); [Pi], 5 (Rebeille et al., 1984); [K+],, 100 (Leigh and Wyn Jones, 1984). At pH 7.3. KATP is 6.48 x lo5 M and Kpp, is 5.97 X lo3 M. At this cytosolic pH, the coupling ratio of the V-ATPase would be approximately 3; the maximum pH, attainable would be 4.65 at A q = -20 mV, but only 5.16 at A q = -50mV. Control of A* appears critical to acidification. Both calculated values of pH, sufficiently encompass observed typical vacuolar PH. More acidic luminal pH could only be achieved by the V-ATPase were its coupling ratio to decrease. The minimum n observed using patch clamp was 1.75, obtained at pH, 8.0 (Davies et al., 1994a). Although there is no experimental evidence to link high cytosolic pH with profoundly acidic vacuolar pH, if pH, 8.0 is substituted into Equation (2) (with the necessary correction to KATP) then a maximum pH, of 2.88 could be attained at A q = -20 mV and 3.39 at -50 mV.


3.59

B . V-PPase A N D K t ACCUMULATION

If a typical pH, of 5.5 is considered, then at A* = -20mV the maximum [K'lV that could be attained through the action of the V-PPase with a coupling ratio of 1.3 H+:1.7 Kf would be approximately 670mM. This is sufficient for even guard cells. However, if the extreme cases are taken and the poise of the V-PPase assessed at the maximal pHv values that could theoretjcally be generated by the V-ATPase, then at pH, 7.3, AW -20 mV and pH, 4.65 the maximum [K'], would only be 150mM. At A* -50mV and pH, 5.16, [K'], would decrease further to approximately 82mM. As the E,,, of the V-PPase would be -41 mV, the enzyme would be reversed and would translocate K + into the cytosol. If the K + gradient were not to be so dissipated, the V-PPase would have to be under strict kinetic control (e.g. by Ca2+; Rea er al., 1992). This exercise highlights a crucial weakness in the argument for the V-PPase as the agent of vacuolar K + accumulation - its operational range as a symportcr would be severely restricted by pH, and AW. There is simply not enough known on the temporal dynamics of vacuolar pH and AW to assess how deleterious to the maintenance of the K + gradient those parameters would be. Potassium accumulation mediated by the V-PPase would be critical to the osmotic relations of guard cells and developing tissues. From the calculations above it appears that the V-PPase could be competent for the pronounced K+ accumulation of guard cells. Demonstration of V-PPase activity in that system is currently lacking, but preliminary studies using patch clamp and immunofluorescence microscopy indicate that there is a functional V-PPase in the vacuoles of Vicia and Triticum guard cells ( C . P. Darley, D. Sanders and J . M. Davies, unpublished data). Moreover, using antibodies raised against the V-PPase and V-ATPase, immunofluorescence microscopy of Triticum leaves suggest that there is a reciprocal accumulation of the pumps during development (C. P. Darley, J . L. Marrison, R. M. Leach, P. A. Rea, D. Sanders and J. M. Davies, unpublished data). In developing tissue, the V-PPase is the predominant pump whereas at maturity it is the V-ATPase. Thus, in provacuoles of young tissue, the V-PPase may facilitate K+ accumulation to drive (indirectly) cell expansion.

VI. CONCLUSIONS The study of the bioenergetics of vacuolar pumps assists not only in the understanding of their transport mechanisms but also of their physiological function. Bioenergetic studies are still in their infancy and must be continued to be developed on both the reductionist and holistic level.

360

J. M. DAVIES

ACKNOWLEDGEMENTS I would like to thank all those working on plant vacuolar H+ pumps, especially the laboratories at the University of Pennsylvania and IACR Rothamsted. This work was supported by the award of a Royal Society University Research Fellowship and is dedicated to Dorothy Davies.

REFERENCES Bates, G. W., Goldsmith, M. H. M. and Goldsmith, T. H. (1982). Separation of tonoplast and plasma membrane potentials and resistance in cells of oat coleoptiles. Journal of Membrane Biology 66, 15-23. Bennett, A. B. and Spanswick, R. M. (1984). H+-ATPase activity from storage tissue of Beta vulgaris. Plant Physiology 14, 545-548. Bertl, A., Blumwald, E., Coronado, R., Eisenberg, R., Findlay, G . , Gradmann, D., Hille, B., Kohle, K., Kolb, H.-A., MacRobbie, E., Meissner, G., Miller, C., Neher, E., Palade, P., Pantoja, O., Sanders, D., Schroeder, J., Slayman, C. L., Spanswick, R., Walker, A. and Williams, A. (1992). Electrical measurements on endomembranes. Science 258, 873-874. Bowman, E. J., Siebers, A. and Altendorf, K. (1988). Bafilomycins - a class of inhibitors of membrane ATPases from microorganisms, animal cells and plant cells. Proceedings of the National Academy of Sciences of the USA 85, 7972-7976. Cooper, S., Lerner, H. R. and Reinhold, L. (1991). Evidence for a highly specific K+/H+ antiporter in membrane vesicles from oil-seed rape hypocotyls. Plant Physiology 97, 1212-1220. Davies, J. M. and Sanders, D. (1995). ATP, pH and Mg2+ modulate a cation current in Beta vulgaris vacuoles: a possible shunt conductance for the vacuolar H+-ATPase. Journal of Membrane Biology 145, 75-86. Davies, J. M., Rea, P. A. and Sanders, D. (1991). Vacuolar proton-pumping pyrophosphatase in Beta vulgaris shows vectorial activation by potassium. FEBS Letters 218, 66-68. Davies, J. M., Poole, R. J., Rea, P. A. and Sanders, D. (1992). Potassium transport into plant vacuoles energized directly by a proton-pumping inorganic pyrophosphatase. Proceedings of the National Academy of Sciences of the USA 89, 11701-11 705. Davies, J. M., Poole, R. J. and Sanders, D. (1993). The computed free energy change of hydrolysis of inorganic pyrophosphate and ATP: apparent significance for inorganic-pyrophosphate-driven reactions of intermediary metabolism. Biochimica et Biophysica Acta 1141, 29-36. Davies, J. M., Hunt, I. and Sanders, D. (1994a). Vacuolar H+-pumping ATPase variable transport coupling ratio controlled by pH. Proceedings of the National Academy of Sciences of the USA 91, 8547-8551. Davies, J. M., Hunt, I. and Sanders, D. (1994b). Ion translocation stoichiometries of two endomembrane H+-pumps studied by patch clamp. In “Molecular and Cellular Mechanisms of H+ transport” (B. H. Hirst, ed.), NATO AS1 Series, Vol. H89, pp. 205-212. Springer-Verlag, Berlin. Davies, J. M., Sanders, D. and Gradmann, D. (1996). Reaction kinetics of the vacuolar H+-pumping ATPase in Beta vulgaris. Journal of Membrane Biology 150, 231-241.


361

Dixon, T. E. and Al-Awqati, Q. (1980). H+/ATP stoichiometry of proton pump of turtle urinary bladder. Journal of Biological Chemistry 255, 3237-3280. Fox, G. G. and Ratcliffe, R. G. (1990). P-31 NMR observations on the effects of the external pH on the intracellular pH values in plant cell cultures. Plant Physiology 93, 512-521. Hansen, U. P., Gradmann, D., Sanders, D. and Slayman, C. L. (1981). Inteml4.08rpretatio of current-voltage relationships for “active” ion transport systems: I. Steadystate reaction-kinetic analysis of class I mechanisms. Journal of Membrane Biology 63, 165-190. Harold, F. M. (1986). “The Vital Force: A Study of Bioenergetics.” Freeman, New York. Hedrich, R., Barbier-Brygoo, H., Felle, H., Flugge, U. I., Luttge, U., Maathuis, F. J. M., Marx, S., Prins, H. B. A , , Raschke, K., Schnabl, H. and Schroder, J. I. (1988). General mechanisms for solute transport across the tonoplast of plant vacuoles: a patch clamp survey of ion channels and proton pumps. Botanica Acta 101, 7-13. Johannes, E. and Felle, H. (1990). Proton gradient across the tonoplast of Riccia fluitans as a result of the joint action of two electroenzymes. Plant Physiology 93, 412417. Kim, Y., Kim, E. J. and Rea, P. A. (1994a). Isolation and characterization of cDNAs encoding the vacuolar H+-pyrophosphatase of Beta vulgaris. Plant Physiology 106. 375-382. Kim, E. J . , Zhen, R. G. and Rea, P. A. (1994b). Heterologous expression of plant vacuolar pyrophosphatase in yeast demonstrates sufficiency of substrate-binding subunit for H + transport. Proceedings of the National Academy of Sciences of the USA 91, 6128-6132. Kramer, D., Mangold, B., Hille, A., Emig, I., Hess, A., Ratajczak, R. and Luttge, U . (1995). The head structure of a higher plant V-type H+-ATPase is not always a hexamer but also a pentamer. Journal of Experimental Botany 46, 16331636. Lai, S., Watson, J. C., Hansen, J . N. and Sze, H. (1991). Molecular cloning and sequencing of cDNAs encoding the proteolipid subunit of the vacuolar H+-ATPase from a higher plant. Journal of Biological Chemistry 266, 16 078-16 084. Lauger, P. (1991). “Electrogenic Ion Pumps.” Sinauer, Sunderland, MA. Leigh, R. A. and Wyn Jones, R. G. (1984). A hypothesis relating critical potassium concentrations for growth to the distribution and function of this ion in the plant cell. New Phytologist 97, 1-14. Maathuis, F. J. M. and Sanders, D. (1994). Mechanism of high-affinity potassium uptake in roots of Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the USA 91, 9272-9276. MacRobbie, E. A. C. (1988). Stomata1 guard cells. In “Solute Transport in Plant Cells and Tissues” (D. A. Baker and J. L. Hall, eds), pp. 453497. Wiley, New York. Malone, M., Leigh, R. A. and Tomos, A. D. (1991). Concentrations of vacuolar inorganic ions in individual cells of intact wheat leaf epidermis. Journal of Experimental Botany 42, 305-309. Nelson, H. and Nelson, N. (1989). The progenitor of ATP synthases was closely related to the current vacuolar H+-ATPase. FEBS Letters 247, 147-153. Nelson, H . , Mandiyan, S. and Nelson, N. (1989). A conserved gene encoding the 57 kDa subunit of the yeast vacuolar H+-ATPase. Journal of Biological Chemistry 264, 1775-1778.

362

J. M. DAVIES

Obermeyer, G., Sommer, A. and Bentrup, F.-W. (1995). Potassium and voltage dependence of the vacuolar pyrophosphatase. Abstract P41, loth International Workshop on Plant Membrane Biology, Regensburg, FRG. Rea, P. A. and Sanders, D. (1987). Tonoplast energization: two H+ pumps, one membrane. Physiologia Plantarum 71, 131-141. Rea, P. A , , Britten, C. J . , Jennings, I. R., Calvert, C. M., Skiera, L. A., Leigh, R. A. and Sanders, D. (1992). Regulation of vacuolar H+-pyrophosphatase by free calcium. Plant Physiology 100, 1706-1715. Rebeille, F., Bligny, R. and Douce, R. (1984). Is the cytosolic Pi concentration a limiting factor for plant cell respiration? Plant Physiology 74, 355-359. Ros, R., Romieu, C., Gibrat, R. and Grignon, C. (1995). The plant inorganic pyrophosphatase does not transport K+ in vacuolar membrane vesicles multilabeled with fluorescent probes for H + , K + and membrane potential. Journal of Biological Chemistry 270, 1-7. Sarafian, V . , Kim, Y., Poole, R. J . and Rea, P. A. (1992). Molecular cloning and sequence of cDNA encoding the pyrophosphate-energized vacuolar membrane proton pump of Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the U S A 89, 1775-1779. Sato, M. H., Kasahara, M., Ishii, N., Homareda, H., Matsui, H. and Yoshida, M. (1994). Purified vacuolar inorganic pyrophosphatase consisting of a 75-kDa polypeptide can pump H+ into reconstituted proteoliposomes. Journal of Biological Chemistry 269, 672543728. Schmidt, A. L. and Briskin, D. P. (1993). Energy transduction in tonoplast vesicles from red beet (Beta vulgaris L.). Storage tissue: H’hbstrate stoichiometries for the H+-ATPase and H+-PPase. Archives of Biochemistry and Biophysics 301, 165-173. Smith, F. A . and Raven, J . A. (1976). H+ transport and regulation of cell pH. In “Encyclopedia of Plant Physiology” (U. Liittge and M. A. Pitman, eds), Vol. 2A, pp. 317-346. Springer-Verlag, Berlin. Spalding, E. P. and Goldsmith, M. H. M. (1993). Activation of K+ channels in the plasma membrane of Arabidopsis by ATP produced photosynthetically. Plant Cell 5, 471-484. Spanswick, R. M. and Williams, E. J. (1964). Electrical potentials and Nai. K+ and CI- concentrations in the vacuole and cytoplasm of Nitella translucens. Journal of Experimental Botany 15, 193-200. van Walvaren, H. S . , Haak, N. P., Krab, K. and Kraayenhof, R. (1986). Evidence for a high proton translocation stoichiometry of the H+-ATPase complex in well coupled proteoliposomes reconstituted from a thermophilic cyanobacterium. FEBS Letters 208, 138-142. Wang, Y. Z . , Leigh, R. A , , Kaestner, K. H. and Sze, H. (1986). Electrogenic H+-pumping pyrophosphatase in tonoplast vesicles of oat roots. Plant Physiology 81, 497-502. Weiner, H., Stitt, M. and Heldt, H. W. (1987). Subcellular compartmentation of pyrophosphate and alkaline pyrophosphatase in leaves. Biochimicu et Biophysica Acta 893, 19-21. White, P. J., Marshall, J . and Smith, J . A . C. (1990). Substrate kinetics of the tonoplast H+-translocating inorganic pyrophosphatase and its activation by free Mg2+. Plant Physiology 93, 1063-1070. Yazaki, Y . , Asukagawa, N . , Ishikawa, Y., Ohta, E. and Sakata, M. (1988). Estimation of cytoplasmic free Mg2+ levels and phosphorylation potentials in mung bean root tips by in vivo P-31 NMR spectroscopy. Plant Cell Physiology 29, 919-924.

THE BIOENERGETICS OF VACUOLAR

363

H+ PUMPS

Zhen, R.-G., Baykov, A . A . , Bakuleva, N . P. and Rea, P. A. (1994). Aminomethylenediphosphonate: a potent type-specific inhibitor of both plant and phototfophic bacterial H+-pyrophosphatases. Plant Physiology 104, 153-159. ~


Transport of Organic Molecules Across the Tonoplast

E. MARTINOIA

Gknktique Physiologique et Molkculaire, BBtiment de Botanique, 40, Avenue du Recteur Pineau F-$6022, Poitiers, France R. RATAJCZAK

Technische Hochschule Darmstadt, Institut fur Botanik, Schmittspahnstrasse 3-5, 0-64287 Darmstadt, Germany

I. Introduction

............................................................................

366

11. Carbohydrates

.........................................................................

367

111. Organic Acids

.........................................................................

372

IV. Amino Acids

..........................................................................

382

V. Polyamines and Peptides

.................

.........................

VI. Transport of Secondary Products of Plant Metabolism

...................

386 388

VII. Conclusion ....................... ......... ............. 390 Acknowledgements ..................................... ............. 390 390 References ........... .......................................................

Advance, in Botanical Research Vol. 25

incorporating Advances in Plan1 Pathology ISBN 0-12-005Y25-8

Copyright

0 1997 Academic Press Limited

All rights of reproduction in any form reserved

366

E. MARTINOIA and R . RATAJCZAK

I. INTRODUCTION The vacuole is the largest compartment of a mature plant cell and may occupy more than 80% of the total cell volume. In such mature cells, the cytosol is visible only as a thin layer which is separated from the cell wall by the plasmalemma and from the vacuolar sap (cell sap) by the vacuolar membrane (tonoplast). The constituents of the cell sap are mainly inorganic salts and water. The storage of inorganic ions within the vacuole enables the plant to reach a large size and a large surface area by accumulating salts from the environment which osmotically drive further water uptake, thereby spending only a minimum of energy for the energy-consuming synthesis of metabolites (Matile, 1978, 1987; Boller and Wiemken, 1986). From the classical point of view the major role of the vacuole is that of a storage compartment for potentially toxic compounds such as phenolics and alkaloids (Matile, 1984). Detoxification of the cytosol by the vacuole is, however, not restricted to the substances of plant metabolism but also to the storage of other potentially toxic compounds, such as microbial toxins and abiotic substances as agrochemicals which have been taken up by the plant (Martinoia et al., 1993). Finally, compartmentation experiments showed that besides inorganic salts and potentially toxic compounds, products of the primary metabolism such as sugars, amino acids, and organic acids are stored within the vacuole. These products, together with several inorganic ions such as nitrate and potassium, can be used for the growth of the plant and are therefore stored within the vacuole only temporarily. Compartmentation of most solutes between cytosol and vacuole is therefore a dynamic process and is reflected by variable vacuolar contents and ratios between cytosol and vacuole of a solute under different metabolic or growing conditions (Martinoia, 1992). In photosynthesizing protoplasts from barley leaves, the carbon fluxes to the vacuole can be studied using a very fast vacuole isolation procedure (Kaiser et al., 1982). The bulk of newly fixed carbon is converted into sucrose. Compartmentation analysis showed that, after a short lag phase, sucrose is transferred to the vacuole at rates comparable to those of its synthesis. Transport of organic acids such as malate or citrate is even faster while only a minor fraction of amino acids are transferred to the vacuole within the time span tested. In contrast, sugar phosphates are not taken up by vacuoles at all. In a similar experiment, Boller and Alibert (1983) showed that newly synthesized sucrose appears only slowly in the vacuole of Melilotus alba, while the uptake of organic acids was comparable to that described for barley. These observations suggest that the cytosolic metabolism is closely linked to the activity of vacuolar transporters and that our knowledge on vacuolar transport systems is an alternative and complementary way to learn more about regulatory mechanisms involved in cytosolic metabolism. In this chapter we will focus mainly on the compartmentation and transport of the products of primary metabolism. We will discuss only briefly the

TRANSPORT OF ORGANIC MOLECULES ACROSS THE TONOPLAST

367

transport of products of the secondary metabolism and xenobiotics, which is dealt with in more detail in the chaptcr by Wink in this volume.

11.

CARBOHYDRATES

Starch and sucrose are the main storage products of primary metabolism. Both are transitorily accumulated in leaves during the photosynthetic period, when the loading capacity of the phloem is limiting, and starch is degraded and sucrose exported during the night (Geiger, 1975; Giaquinta, 1983). In storage tissues, these carbohydrates are accumulated during a vegetative period and utilized as an energy source for growth in the subsequent period. In leaves as well as in storage tissues, sucrose has been partially localized in the vacuole. Compartmentation studies showed that, depending on the plant species and the metabolic conditions of a plant, the vacuolar proportion of sucrose may vary. Values ranging from 20 (Koster and Lynch, 1992) to 100% of the total cellular sucrose content (Leigh et al., 1979) have been reported for vacuoles of leaves and storage tissues. A constant vacuolar proportion of sucrose (75-8570) has been found in barley leaves, independent of the total content (G. Kaiser and E. Martinoia, unpublished findings). Similar results have been reported for spinach by Gerhardt et al. (1987) using the non-aqueous fractionation method. In a later report by the same group, the vacuolar sucrose content was lower; however, even in this case no large fluctuations in the vacuolar proportion were observed. For storage tissues, it was observed that most of the sucrose is localized in the vacuole in plants accumulating sucrose as storage carbohydrate (Leigh et al., 1979; Saftner et a [ . , 1983; Keller, 1988). However, in most plants accumulating alternative water-soluble carbohydrates, such as the tubers of Stachys siebofdii (Keller and Matile, 1985) or Gentiuna luteu (Keller and Wiemken, 1982), sucrose concentrations are reported to be much lower in the vacuole than in the cytosol. This effect is not likely to be due to a low sucrose permeability of the tonoplast in these plants, since a sucrose transport system has also been identified in Stachys (Keller, 1992). A glycoside tranferase activity which transfers one moiety of sucrose onto another carbohydrate thus decreasing the vacuolar sucrose content may be considered. However, such an activity is only known for plants synthesizing fructans, where a fructose moiety of sucrose is transferred to another sucrose by the enzyme sucrose-sucrose fructosyltransferase. In these plants, the sucrose concentration found in the vacuole is often not as low as in plants accumulating alternative carbohydrates (Wagner et ul., 1983; Frehner et a / . , 1984). In some cases, an explanation for a relatively low vacuolar sucrose content may be that sucrose metabolism in the cytosol is fast compared to the vacuolar transport system. Another possibility may be that the alternative carbohydrates are also a substrate for the sucrose transporter and therefore inhibit the transfer of sucrose to the

368

E. MARTINOIA and R. RATAJCZAK

vacuole. Furthermore, when performing compartmentation studies it should also be verified that metabolites are not lost during the isolation procedure. Transport experiments revealed that the mechanism of sucrose transport depends on the plant species. Kaiser and Heber (1984) showed that in isolated barley vacuoles sucrose crosses the tonoplast by facilitated diffusion. The transporter has a low affinity for sucrose ( K , 20-30mM) and is not inhibited by hexoses. Sugars with a fructosyl moiety were inhibitory and the uptake of sucrose could be completely inhibited by the sulfhydryl reagent p-chloromercuribenzene sulfonate (pCMBS). These results were confirmed by Martinoia et al. (1987); however, the affinity of the transport system was even lower ( K , 50 mM). In a careful compartmentation study, Preissner et a f . (1991) demonstrated that sugarcane suspension cells, which accumulate sucrose at comparable concentrations as the plant tissue does, surprisingly do not accumulate sucrose in their vacuole. Transport studies showed indeed that the sucrose carrier of sugarcane exhibits properties very similar to that of barley and was also partly inhibited by 0.5 mM pCMBS. Recently, it has been demonstrated that in tomato fruit sucrose crosses the tonoplast by facilitated diffusion (Preissner and Komor, 1991). However, in this case no pCMBS inhibition was observed. In some storage tissues an energized sucrose uptake mechanism has been demonstrated. The first indication that sucrose uptake is driven indirectly by the vacuolar H+-ATPase and by a sucrose/proton antiporter was shown by Doll et al. (1979) using red beet vacuoles. The affinity of the carrier is similar to that of barley vacuoles ( K , approximately 20 mM). Raffinose was a strong competitive inhibitor, whereas hexoses were only slightly inhibitory (Willenbrink and Doll, 1979). Very similar results were published by Getz (1991) using the same material. However, he observed two different saturable components, one with a K , of 1.7mM, the second with a K , of 31 mM. The K , values did not change in the presence of ATP. Using tonoplast vesicles from sugarbeet, Briskin et a f . (1985) showed that sucrose uptake ( K , 12.1 mM) was enhanced at least 10-fold after establishment of a pH gradient upon the addition of MgATP. An established proton gradient could be dissipated by sucrose but not by glucose or mannitol. Addition of carbonyl cyanide rn-chlorophenylhydrazone (CCCP), a protonophore, resulted in the release of sucrose from the vesicles, indicating that the enhancement by ATP of sucrose uptake and the maintenance of a concentration gradient is mediated indirectly through the pH gradient. Similar H+-dependent sucrose uptake was demonstrated for red beet and Stachys (Getz, 1991; Greutert and Keller, 1993). Attempts to calculate the stoichiometry of the sucrose/H+ antiporter by determining p H changes in the incubation medium suggest that one molecule of sucrose is exchanged for one proton (Getz and Klein, 1995). The fact that stimulation of sucrose uptake is much higher in vesicles than in intact vacuoles is most probably due to a pre-existing ApH in the intact


369

organelle. Since a stimulation of nearly 100% or even more could be observed for sucrose uptake in isolated vacuoles it may even be questioned whether this activation is only due to an enhanced ApH or whether the sucrose transport activity is also regulated by the membrane potential. To our knowledge no studies have been performed to solve this question. A first attempt to identify the vacuolar sucrose transporter was made by Getz et al. (1993) using monoclonal antibodies raised against a highly purified tonoplast fraction from red beet. Clones inhibiting sucrose uptake reacted with a 55-65 kDa and a 110-130kDa fraction. However, inhibition was not restricted to the vacuolar fraction since the same clones were also able to inhibit sucrose uptake in red beet root protoplasts. A transport system similar to that described for sucrose has been described for the uptake of stachyose in Japanese artichoke (Stachys sieboldii) tubers. As for most of the sucrose uptake systems described, the uptake system for stachyose has a low affinity for its substrate (about 50mM). It is driven by a proton gradient and could not be inhibited by pCMBS (Keller, 1992; Greutert and Keller, 1993). The substrate specificity is very similar to those of the sucrose carriers. These data indicate that in Stachys the same translocator is possibly responsible for the transport of sucrose and stachyose. However, not all sucrose transporters are able to recognize stachyose since barley does not take up this tetrasaccharide (F. Keller and E. Martinoia, unpublished findings). Similar results have recently been published by Niland and Schmitz (1995). They further confirmed the necessity of the fructose moiety for the recognition of the carbohydrate by the vacuolar transporter since the trisaccharide manninotriose, which corresponds to a stachyose lacking the terminal fructose moiety, did not inhibit stachyose uptake. An alternative pathway for sucrose accumulation via a UDPglucosedependent group translocator has been proposed by Brown and Coombe (1982). Experiments performed with sugarcane (Thom and Maretzki, 1985) and red beet (Thom et af., 1986; Getz, 1987; Voss and Weidner, 1988) seemingly supported this idea. However, the substance which accumulated in the vacuolar fraction and which comigrated with sucrose on chromatograms was later identified as laminaribiose, a (1-3)-/3 diglucoside synthesized by contaminating plasmalemma vesicles (Preissner and Komor, 1988; Maretzki and Thom, 1988). In many organs, as for example in grape berries, hexoses are accumulated to a much higher degree than sucrose (Moskowitz and Hrazdina, 1981). Vacuolar hexoses may derive from glucose transported into the vacuole from the cytosol or from vacuolar sucrose cleaved by vacuolar invertase. First results indicating that glucose is taken up by an energized system were presented by Guy et a f . (1979) in pea leaves. ATP stimulated the uptake by about 50%, while a protonophore almost completely inhibited uptake in the absence of ATP and only slightly in the presence of ATP. These results suggest a proton-glucose antiport at the tonoplast of pea leaves. However,

370


more information is needed to draw a clear-cut conclusion. More conclusive evidence for the presence of a proton-glucose antiport is given by Rausch et al. (1987). Using a fraction from maize coleoptiles enriched in tonoplast vesicles they observed a MgATP-dependent, highly specific uptake of 3-0-methylglucose ( K , 110mM). Both the pumping activity as well as the uptake of 3-0-methylglucose were inhibited in the presence of antibodies generated against the vacuolar H+-ATPase, indicating that glucose uptake is in fact driven by the vacuolar H+-ATPase. Thom and Komor (1984) showed that vacuolar uptake of 3-0-methylglucose in sugarcane is coupled to a release of protons from the vacuoles and to a decrease in the vacuolar membrane potential. In a later contribution, Preissner and Komor (1991), using highly purified vacuoles from sugarcane suspension cells which accumulate hexoses, could not observe an ATP-dependent effect. They assumed that the previously found energized glucose uptake was a result of serial and simultaneous function of plasmalemma and tonoplast transport systems in vacuoplasts (vacuoles surrounded by the plasmalemma). Data presented so far suggest that glucose uptake in some plants may be an energized process. However, to our knowledge no strict glucose-proton antiport as for sucrose has been shown to occur. As stated above, accumulation of hexoses within the vacuole, an observation which has often been reported (e.g. Moskowitz and Hrazdina, 1981; Keller and Matile, 1989; Preissner et al., 1991), is not a proof for energized hexose uptake since acid invertases are localized in the vacuolar sap (Boller and Wiemken, 1986). Fructan synthesis occurs within the vacuole (Wagner et al., 1983; Frehner et al. , 1984) where, in a first step, a fructose moiety of one sucrose molecule is transferred to another one by the enzyme sucrose-sucrose l-fructosyltransferase (Pollock and Cairns, 1991). This reaction yields a trisaccharide, kestose. Indirect proof for the origin of kestose in the cell sap may be obtained by the observation that it is not taken up by isolated vacuoles (U. Heck and E. Martinoia, unpublished findings). In fructan synthesis, one molecule of glucose is liberated per trisaccharide. The osmotic strength is therefore not altered, and it has to be postulated that glucose is exported from the vacuole and remetabolized in the cytosol to allow the plant to maintain a low water potential (Pollock and Cairns, 1991; Pollock and Kingston-Smith, this volume). Glucose can indeed be exchanged by facilitated diffusion between the cytosol and the cell sap (Martinoia et al., 1987). The permease is specific for D-glucose, has a K , of about 5 m M , a low VmaX,is not inhibited by pCMBS, and is distinct from the sucrose carrier as shown by inhibition experiments. Using the efflux analysis technique, Daie and Wilusz (1987) concluded that, in celery phloem segments, transport of glucose across the tonoplast also occurs by facilitated diffusion. A further plant where glucose uptake is apparently not energized is the tomato (Milner et a l . , 1995). The uptake system exhibits a very low affinity (120mM) and is, in contrast to glucose transport in barley, inhibited by pCMBS. Sucrose


371

and glucose are apparently transported by different carriers since sucrose transport is not inhibited by this sulfhydryl reagent. Fructose transport into the vacuole has been investigated in only a few cases and not in detail (Martinoia et al., 1987; Milner et al., 1995). It may be assumed that fructose is transported with a low affinity by the sucrose as well as the glucose carrier. However, more detailed investigations are required to clarify this point. Young coleoptiles are strong sink organs. and may be specialized to accumulate large amounts of metabolites. Young leaves also depend on imported metabolites. It may, therefore, be an interesting task to follow the properties of the vacuolar carbohydrate transport systems in plants throughout the development of certain phenotypes of cells and to investigate whether facilitated diffusion or energized uptake systems correspond to a plant or tissue species or whether developmental changes can be observed. Polyols are widely distributed in plants. They can be divided into alditols and cyclitols. Alditols are formed by reduction of sugars and the carbon atoms of these compounds are arranged in a straight chain. The most prominent alditols are sorbitol, present in most woody rosaceous plants, and mannitol, which is abundant in celery. The major group of cyclitols occurring in plants are derived from inositols and are formed by methylation of one of the six inositol hydroxyl groups (Popp and Smirnoff, 1995). These methylated polyols may play an important role in salt, cold, and drought tolerance. Tobacco expressing a bacterial mannitol-I-phosphate dehydrogenase gene has been shown to confer salt tolerance to these plants (Tarczynski et al., 1993). In Rosaceae, sorbitol is synthesized in the leaves, transported in the phloem and oxidized in the fruits, and therefore present only in low concentrations in this organ. In apple cotyledons, sorbitol accounted for about 50% of the carbohydrates; about 45% was localized extracellularly. From the cellular sorbitol about 50% could be localized within the vacuole (Yamaki, 1982). In apple fruit, sorbitol concentration was much lower (about 10% of the soluble carbohydrates). In this case, the cellular sorbitol appeared to be totally localized within the vacuole (Yamaki, 1984). Indeed, this author showed that sorbitol transport across the tonoplast was stimulated by MgATP in this tissue (Yamaki, 1987). However, it is not yet known whether this permease is specific for sorbitol since no inhibition studies were performed. Some of the sorbitol oxidase activity appears to be located in the vacuole and it was therefore postulated that oxidation of sorbitol to glucose may occur in the cytosol as well as in the vacuole. However, these investigations give only some preliminary indications, and the problem of sorbitol transport at the different levels should be analysed in more detail. A different situation is found in celery petioles. Mannitol is present in the vacuole at lower levels than glucose and fructose, and appears to be equally

372


distributed between the extravacuolar space and the vacuole (Keller and Matile, 1989). Indeed, transport experiments suggest that mannitol crosses the tonoplast slowly by facilitated diffusion (F. Keller and E. Martinoia, unpublished findings). In many plants, alditols and cyclitols are synthesized in high amounts in response to water stress imposed by drought, heat, cold o r salt (Popp and Smirnoff, 1995), and serve as compatible solutes. In this case the role of the polyols is not primarily that of a temporary storage compound but that of a protectant against environmental stress. Indeed, pinitol, a cyclitol accumulated during salt stress in Mesembryanthemum crystallinum, was localized in chloroplasts and cytosol (Paul and Cockburn, 1989). Carbohydrates are stored in the vacuole temporarily and are released according to the metabolic requirements in the cytosol. In those cases in which sugars cross the tonoplast by facilitated diffusion, the direction of the flux is simply determined by the respective concentration gradients. Fructan is degraded by fructan exohydrolase yielding free fructose. No clear-cut results are available on fructose release although in barley leaves fructose may cross the tonoplast by the sucrose- or glucose-specific permeases (Martinoia et ul., 1987). In red beet tap root, the vacuolar invertase increases during sprouting (Leigh et al., 1979), suggesting that withdrawal of the accumulated sucrose occurs by hydrolysis and efflux of the resulting monosaccharides from the cell sap to the cytosol. Transport systems for hexoses have, however, not yet been investigated in beet. In contrast to the sucrose and glucose transporter of the plasmamembrane, the respective carriers of the tonoplast have not been identified. The results of Getz et al. (1993), who showed that a monoclonal antibody could inhibit both the vacuolar and the plasma membrane sucrose transport, and of E. Martinoia, R. Lomoine and S. Delrot (unpublished results), who could inhibit transport of sucrose across the tonoplast by an antibody raised against the putative sucrose transporter of the plasmamembrane, indicate, however, that these transporters may share some structural similarities.

111.

ORGANIC ACIDS

The first experimental evidence for compartmentation of organic acids in plant cells came from the investigations of MacLennan et al. (1963). By feeding [14C]acetate to a variety of plant tissues, these authors demonstrated most elegantly that the bulk of various organic acids, i.e. aconitic acid, aspartic acid, citric acid, glutamic acid, isocitric acid, malic acid and succinic acid, must be located in the vacuole. A comparative study of several higher plant species (Nierhaus and Kinzel, 1970) revealed a great number of organic acids which are present in different amounts and combinations in leaf press saps and which at least are partially stored


373

in the vacuole. Some of the most abundant organic acids are malic and citric (ubiquitously distributed), aconitic (Gramineae, some Ranunculaceae, Pteridophytae, mosses), fumaric (some Boraginaceae), glyceric (some Liliaceae, Taraxacum), malonic (many Papilionaceae), tartaric (Pefargonium, some Papilionaceae and Polygonaceae, grapes), quinic and shikimik (Ginkgo, Sambucus nigra, Bryophytae, Pteridophytae), chlorogenic (e .g. Ranuncufus repens, Rosa canina, Ginkgo bifoba),ascorbic (e.g. Armoracia) and oxalic (Centrospermae, Violoceae, Labiatae, some Polygonaceae) acids, while several others are present in lower amounts. Since the vacuolar lumen can constitute more than 80% of the total cellular volume, it is obvious that the largest portion of the osmotically active solutes must be located in the vacuole (Leigh, 1983). Thus, considering the high malate content often observed in plants, most of the malate has been postulated to be localized within the vacuole. This assumption was demonstrated to be true by measurements of the malate content of isolated vacuoles of Kalanchoe sp. (Buser and Matile, 1977) and subsequently of isolated vacuoles from other plants (Gerhardt and Heldt, 1984; Schnabl and Kottmeier, 1984). The same assumption should hold true for other organic acids which are present in high amounts in certain species, e.g. citrate (SO mmol (kg fresh weight)-') or isocitrate (175 (mmol (kg fresh weight)-') in Kafanchoe cafycinum (Vickery, 1952). In the vacuole-like lutoids of Hevea brasifiensis, citrate concentrations higher than SO mM have been detected (Marin et a f . , 1982), while the concentration in the cytosol was found to range between 5 and 6 m M (D'Auzac ef a l . , 1982). The highest cell sap citrate concentration reported was 200mM in leaf tissue of the crassulacean acid metabolism (CAM) plants Clusia minor and Clusia rosea (Franc0 et al., 1992). In an elegant study applying "C nuclear magnetic resonance (NMR) spectroscopy to living Acer pseudoplatanus L. suspension cells, Gout et a f . (1993) could demonstrate that newly synthesized malic acid was accumulated in the cytoplasm; after reaching a cytoplasmic threshold concentration, malic acid was transported into the vacuole. In the same material, newly synthesized citric acid accumulated steadily in the vacuole. The first indications of intracellular transport of citrate in CAM were published by Olivares et al. (1993), who demonstrated by pulse-chase experiments that during the dark period in Clusia minor after malate formation by phosphoenolpyruvate carboxylase, citrate is synthesized from malate or oxaloacetate in the mitochondria followed by citrate export from the mitochondria. In plant tissues with high 1-aminocyclopropane-l-carboxylic acid (ACC) production this precursor of ethylene biosynthesis may be malonylated and subsequently stored in the vacuole (Tophof ef a f . , 1989; see also Wink, this volume). Monocarboxylic acids are also present in high amounts in distinct materials. Horseradish (Armoracia) roots accumulate high amounts of ascorbic acid, which plays a specific role as a cofactor and activator of the cytosolic enzyme myrosinase. In this plant, ascorbate is

374


located almost exclusively in the vacuole (Grob and Matile, 1980). The content of ascorbic acid in isolated barley protoplasts was much higher under high light intensities (Rautenkranz et al., 1994). Interestingly, the relative abundance of ascorbic acid in the vacuole increased from 15% under low-light conditions to 26% under high-light conditions. Treatment of buckwheat with an inhibitor of the phenylalanine ammonium lyase resulted in the production of high amounts of shikimic acid (Hollander-Czytko and Amrhein, 1983). Localization studies showed that this acid was localized almost completely in the vacuole. With a few exceptions, up to now transport processes across the vacuolar membrane have only been studied intensively for malate and citrate, mainly for two reasons: (i) malate and citrate are the most abundant organic anions in the plant cell, occurring in all plant species, and (ii) both compounds are of outstanding interest since they play several important roles in cellular metabolism (Martinoia and Rentsch, 1994). Both malate and citrate are intermediates of the tricarboxylic acid (TCA) and the glyoxylate cycle. The fact that the reactions catalysed by fumarase, malate dehydrogenase and aconitase are strongly in favour of the formation of malate and citrate, respectively, might explain the high abundance of these organic anions in the cell sap. Furthermore, malate can be synthesized by the carboxylation of phosphoenolpyruvate (PEP) and subsequent reduction of oxaloacetate. Malic acid synthesis and degradation play a role in the cytosolic “pH-stat’’ mechanism (Smith and Raven, 1979; Allan and Raven, 1987). In addition, malate functions as a shuttle of reduction equivalents from mitochondria and chloroplasts to the cytosol or peroxisomes (Douce and Neuburger, 1989; Heineke et al., 1991). In plants performing C4 photosynthesis, malate is involved in the transfer of C 0 2 and reduction equivalents from mesophyll to bundle sheath chloroplasts (Edwards and Huber, 1981). Roots of several plants excrete malic acid and citric acid into the soil to increase phosphate availability (Hoffland et al., 1992; Ohwaki and Hisata, 1992). Examples of metabolic processes which require malate transport across the tonoplast are (i) osmotic control, e.g. in guard cells where malic acid synthesis and degradation is involved in stomatal opening and closure (Raschke, 1979; Raschke et al., 1988), (ii) the neutralization of OH- ions, which are produced after reduction of N 0 3 - and S042- by the protons of malic acid and subsequent accumulation of cations together with malate in the vacuole, and (iii) storage of CO2 in plants performing CAM (Kluge and Ting, 1979; Ting, 1985). All of the three processes mentioned above show a rhythmic behaviour of the malate content of the cell sap. The stomatal guard cells of most plants use K2malate as a vacuolar osmoticum to increase stomatal aperture (Raschke, 1979). Malic acid is synthesized, H + extruded and K + taken up during opening. For stomatal closure, both the accumulation of K2ma!ate and H+/K+ exchange have to be reversed. While stomatal opening and

TRANSPORT OF ORGANIC MOLECULES ACROSS T H E TONOPLAST

375

closure is a rapid process, the concentration of malate in the cell sap shows a diurnal rhythmic behaviour in the processes involved in both the reduction of nitrate or sulfate and CAM. The OH- ions produced by the processes of sulfate or nitrate reduction are neutralized by protons of malic acid and malate accumulates together with cations taken up in the vacuoles of photosynthetically active tissue during the day. In the subsequent dark phase, malate is metabolized, thus leading to a dayhight rhythm of malate concentrations, with high malate values during the day and low malate values during the night. Amplitudes of day/night changes of the malate level of CD plants can reach 45 mmol (kg fresh weight)-' day-' in Moricandia arvensis (Winter et a f . , 1982) or 30mMday-' in the vacuoles of Spinacia oferacea (Gerhardt and Heldt, 1984). The rhythm of malate levels in photosynthetically active cells of CAM plants is inverted (for reviews, see Kluge and Ting, 1979; Ting, 1985). As a strategy to save water, plants performing CAM close their stomata during the day when the water vapour pressure difference between the leaf and the atmosphere is large. C 0 2 is taken up during the night and fixed to phosphoenolpyruvate via phosphoenolpyruvate carboxylase (PEPC). The oxaloacetate formed is reduced to malic acid, which serves as a storage form for CO? and reduction equivalents and is accumulated in the vacuole. During the subsequent light phase malic acid is released from the vacuole, decarboxylated, and the C 0 2 produced enters the Calvin cycle. This CAM rhythm leads to very high concentrations of malic acid in the cell sap during the night and low malate concentrations during the day, with dayhight concentration differences which may reach 320 mM in the bromeliad Aechmea nudicauis (Smith et al., 1986). Although it has been known for a long time that citrate shows a similar rhythmicity to malate in CAM plants (Wolf, 1960), this phenomenon has attracted only little attention (Luttge, 1988). Guthrie (1934) was the first to observe a five-fold increase of citric acid levels in Kalanchoe cal.ycinum during the night, accounting for 25% of the change in total acidity. Using the same species, Wolf (1939) determined the contribution of citric acid to be up to 82% under certain conditions . From thermodynamic considerations, Luttge et al. (1981) proposed an MgATP-dependent malic acid uptake into the vacuole. However, when Buser-Suter et al. (1982) presented the first evidence for a permease involved in malate fluxes in isolated vacuoles of the CAM plant Kafanchoe daigremontiana, they did not observe a MgATP-dependency of malic acid uptake. On the other hand, a few years later MgATP-dependent malic acid uptake into isolated intact mesophyll vacuoles o f Hordeum vufgare was observed (Martinoia et al., 1985). The authors could show that MgATP-stimulated uptake occurred against a concentration gradient and was not the result of enhanced malate exchange. Similar results of MgATP-dependent malate uptake were obtained later by other authors using either preparations from Kafanchoe daigremontiana (Nishida and Tominaga, 1987; Marquardt-Jarczyk

TABLE I Properties of the vacuolar malate transporter of various plant species and preparations (modified after Ratajczak et al., 1994)

Species/mode of photosynthesis

vmax

V,,, recalculateda Other organic anions which (nmol malate are transported by the malate min-’ (mp uptake system or inhibit protein)- ) malate uptake

Preparation

Measurement

Energization

K, (mM malate)

measured

Hordeum vulgarel C3l

vacuoles

[ “CIMalate uptake

H+-ATPase

2.5

800 nmol malate/(h x lo6 vacuoles)

-1300.0

Carharanthus roseusl C3?

vacuoles

[ “C]Malate uptake

H+-ATPase

4.5

-0.5

Kalanchoe ahigremontianal CAM3 Kalanchoe daigremontianal CAM4 Kalanchoe daigremonriand CAM’

vacuoles

[ I4C]Malate uptake

1.0

small vacuoles

[‘4C]Malate uptake

No energizationcatalysed diffusion H+-ATPase

2.5

tonoplast vesicles

H+ transporl

H+-ATPase

14.0

0.33 nmol malate/(h x 106 vacuoles) 150 nmol malate/(h x lo6 vacuoles) 25 nmol malate/(h x lo6 vacuoles) Not determined

~_______

Kalanchoe daigremontianal

Kalanchoe ahigremontianal CAM?

-250.0

-41.0 Not determined

Citrate > D-malate > oxaloacetate > tartronate > malonate > succinate D-Malate > oxaloacetate > fumarate, succinate, quinate > malonate. citrate D-Malate, tartrate > citrate > succinate, fumarate, malonate, isocitrate Not determined Fumarate > methylfumarate

> tartrate > oxaloacetate > D-malate > 2-oxoglutarate > o-tartrate > succinate > 4.0

>2.36 nmol (mg protein)-’ min-’

>2.36

2-met hylsuccinate Succinate

uptake

No energizationcatalysed diffusion

tonoplast vesicles

H+ transport

H+-ATPase

19.0

Not determined

Not determined

Not determined

tonoplast vesicles

[14C]Malate uptake

H+-ATPase

3.0

31.2 nmol malate (mg protein-’ min-’

31.2

Not determined

tonoplast vesicles

[ 14C]Malate uptake

K+ gradient

2.7

85.5 nmol malate (mg protein)-’ min-’

85.5

Citrate

tonoplast vesicles

[ I4C]Malate

CAM^ Kalanchoe ahigremonrianal CAM7 Kalanchoe daigremonrianal CAM7

~

References: 1, Martinoia et al. (1985); 2 , Marigo et al. (1988); 3 , Buser-Suter et al. (1982); 4, Nishida and Tominaga (1987); 5, White and Smith (1989); 6, Bettey and Smith (1993); 7, Ratajczak et al. (1994). aRecalculated where required, assuming that 10 pg of protein corresponds to lo6 vacuoles, as determined for Hordeum vulgare by Martinoia et al. (1991b).


377

and Liittge, 1990; White and Smith, 1989; Bettey and Smith, 1993; Ratajczak et al., 1994) or from cultured cells of Carharanthus roseus (Marigo et al., 1988). Further evidence for the involvement of the vacuolar H+-ATPase in malate uptake came from inhibitor studies. MgATP-dependent malate uptake could be inhibited by N,N’-dicyclohexylcarbodiimide (DCCD), diethyl stilbestrol (DBS) and nitrate, which are all known to be inhibitors of the vacuolar H+-ATPase (Martinoia et al., 1985; White and Smith, 1989). MgATP in the presence of ethylene diamine tetraacetic acid (EDTA) did not stimulate malate uptake (Martinoia et nl., 1985). Moreover, the highly specific vacuolar H +-ATPase inhibitor bafilomycin A, also inhibited the ATPstimulated malate uptake (Martinoia et al., 1993). Malate uptake was also stimulated several-fold by activation of the vacuolar Hf -pyrophosphatase (Bettey and Smith, 1993). Energization of native tonoplast vesicles from Kalanchoe daigremontiana by an artificial potassium gradient establishing only an inside-positive electrical membrane potential (A*) showed that A* was sufficient as the sole driving force for malate uptake. A proton gradient across the tonoplast is therefore not required to drive malate uptake into the vacuole (Ratajczak er al., 1994). Interestingly, the kinetic parameters for malate transport in the references mentioned above show considerable variation (Table I). The highest values for the apparent K , for malate of 19.0mM (Ratajczak et al., 1994) and 14.0 mM (White and Smith, 1989) have been obtained by measuring the effect of malate on H + transport activity after energization of malate uptake by MgATP. The other values, which were all obtained by direct determination of [‘4C]malate uptake, range from 1.0 to 4.5 mM malate independently of the mode of energization (i.e. no energization, energization via the vacuolar H+-ATPase, or energization via an artificial K+ gradient) or the use of either intact vacuoles or tonoplast vesicles. Taking into account that the cytosolic phosphoenolpyruvate carboxylase which is involved in malate synthesis is feedback inhibited by malate (Kluge and Osmond, 1972; Buchanan-Bollig and Kluge, 1981) with K , values ranging from 0.004 to 0.9mM (Greenway et al., 1978; Winter, 1980, 1982), high amounts of malate would accumulate in the cytosol, preventing malate synthesis if the vacuolar malate transporter had a low affinity for malate, as would be the case for K , values of 14.0 or 19.0mM. This indicates that indirect measurements of malate uptake by following the effect of malate on H + transport activity leads probably to a gross overestimation of K,, while the lower K,, values are more likely to reflect the physiological requirement. The similarity of the K , values obtained by direct measurement of [ “C]malate uptake using tonoplast vesicles or vacuoles from different sources indicates that the substrate affinity of the malate transporter is almost identical in plants performing C3 photosynthesis or CAM, In contrast, maximal rates of malate uptake seem to depend on the preparation used for uptake experiments (Table I). Interestingly the V,,, values obtained with small vacuoles (Nishida and

378


Tominaga, 1987) and tonoplast vesicles (Ratajczak et al., 1994) of Kalanchoe daigremontiana are of the same order of magnitude. Malate transport at maximal rates obtained for these materials would be about three times the rate required for a nocturnal in vivo malate accumulation of 200mM in Kalanchoe daigremontiana (Ratajczak et al., 1994), allowing the malate transporter to operate below substrate saturation. It has been shown in several studies that the malate uptake system is not specific for the natural enantiomer L-malate (Table I). Buser-Suter et a f . (1982) found that in vacuoles of Kalanchoe daigremontiana other dicarboxylic acids were inhibitory. In barley mesophyll vacuoles citrate, D-malate, phenylsuccinate, cis-aconitate, fumarate, oxaloacetate, tartronate, malonate, succinate and benzenetricarboxylate behaved as competitive inhibitors (Martinoia et a f . , 1985; Rentsch and Martinoia, 1991). Similar results have been obtained for suspension cells of Catharanthus roseus (Marigo et a f . , 1988) and Kalanchoe daigremontiana (Nishida and Tominaga, 1987; White and Smith, 1989; Ratajczak et al., 1994) where fumarate (White and Smith, 1989) and succinate (Bettey and Smith, 1993) exhibited an even higher affinity for the uptake system than malate. Contradictory results have been published for the uptake of N-malonyl-ACC into the vacuole. While Tophof et al. (1989) suggested from inhibition experiments that this compound is transported by the same carrier as malate, Bouzayen et al. (1989) proposed a separate carrier. Divalent malate, which is the predominant form in the cytosol, seems to be transported across the tonoplast, as suggested from the calculation of K , values obtained when uptake experiments were performed at different external pH (Rentsch and Martinoia, 1991). Marin et al. (1982) suggested MgATP-dependent citrate uptake to occur via a H2citrate-/H+ antiport in lutoids of Hevea brasiliensis. On the other hand, citrate uptake into tonoplast vesicles from tomato fruits was not specific for citrate and showed similarities to the malate uptake system, with the exception that citrate uptake was not stimulated by MgATP (Oleski et al., 1987). In this study, two saturable components of uptake could be distinguished, possibly due to the transport of &rate'- and Hcitrate2- via the same carrier with different affinities. In contrast, citrate uptake into barley mesophyll vacuoles was stimulated by MgATP, and showed only one saturable component with an apparent K , of 0.2mM (Rentsch and Martinoia, 1991). At physiological pH, &rate3- was suggested to be the transported form for both tomato and barley. Various di- or tricarboxylic acids inhibited malate and citrate uptake to similar extents, thus indicating that both carboxylic acids penetrate the tonoplast via the same carrier. If this assumption holds true, it would be of special interest to investigate the regulation of organic acid transport in CAM plants which accumulate both malic and citric acid during the nipht (see above). Reduction of nitrate in the cytosol which is accompanied by vacuolar nitrate efflux leads to an accumulation of malate in the vacuole (Marigo et


379

a / ., 1985). This observation was interpreted in terms of an antiport exchange of these two anions. However, experimental data suggest that the exchange is not mediated by a coupled antiport but by two distinct carriers (BlomZandstra et al., 1990; G.Kaiser and E. Martinoia, unpublished findings). Experiments with several protein-modifying reagents gave hints for the presence of several amino acid residues of carboxylate transporters which might be essential for the recognition o f di- and tricarboxylates. (i) Transport is efficiently inhibited by the histidine modifying reagent diethyl pyrocarbonate (Rentsch and Martinoia, 1991; Dietz et al., 1092). (ii) Uptake was inhibited by 4,4’-diisothiocyano-2,2’-stilbene disulfonic acid (DIDS) and pyridoxal phosphate (Bouyssou et al., 1990; Martinoia et al., 1990; Bettey and Smith, 1993). Since inhibition of both these protein-modifying reagents could be prevented by di- or tricarboxylates, a lysine and a histidine moiety may be involved in substrate recognition. (iii) An essential role of sulfhydryl groups was suggested since non-energized dicarboxylate uptake in Kalarzchoe daigrernontiana vacuoles was sensitive to the sulfhydryl modifying reagents N-ethylmaleimide (NEM) and pCMBS (Bettey and Smith, 1993). The identification of a channel which is permeable to monovalent cations and to anions such as chloride and malate (Hedrich et al., 1986;Coyaud et al., 1987, Pantoja et ul., 1992) gave rise to the speculation that all of these substances are transported across the tonoplast via the same channel. However, impermeable competitive inhibitors of malate uptake such as phenylsuccinic acid or 1,2,3-benzenetricarboxyIicacid did not inhibit chloride uptake (White and Smith, 1989; Martinoia et a / . , 1990). Furthermore, in contrast to malate uptake, chloride uptake was not inhibited by pyridoxal phosphate (Martinoia et al., 1990).Thus, the inhibition of malate uptake by chloride and vice versa observed in isolated vacuoles might not be due to a common binding site of both substances, but rather to competition for the energy source (Martinoia et al., 1990). On the other hand, patch clamp experiments with vacuoles have revealed a channel which opens only at potentials more negative than -30 mV (cytosol-negative) exhibiting properties similar to those of the uptake system described with tracer flux analyses (Iwasaki ef ul., 1992; Pantoja et d., 1992), and thus probably are identical to the malate carrier. In a recent study, Cerana et a / . (1995) described a channel mediating malate’- influx into the vacuole of Arabidopsis thaliana opening at potentials more negative than -60 mV. This channel also mediates uptake of succinate, fumarate, and oxaloacetate. Since the activation potential is dependent on the external (cytosolic) malate’concentration, thus indicating a malate sensor mechanism, this channel might play a role in accumulation and storage of malate in the vacuole. So far, no carboxylate transporter of the tonoplast has been identified, although the search for these transporters is important for the understanding of cellular malate compartmentation. Combination of hydroxyapatite chromatography, which previously has successfully been used for the

380


purification of di- and tricarboxylate carriers of the inner mitochondria1 membrane (Kaplan and Pedersen, 1985; Bisaccia et a f . , 1988; Bolli et al., 1989; Indivieri et a f . ,1989), and affinity chromatography using S-amino-l,2,3benzenetricarboxylic acid as a ligand, led to a partial purification of the malate carrier of barley mesophyll vacuoles (Martinoia et al., 1991b). Purification factors of 20-30 and approximately 30 000-50 000 were obtained after hydroxyapatite and affinity chromatography, respectively. After reconstitution of the partially purified malate transporter in artificial asolectin liposomes, characteristics of malate transport were similar to those of isolated mesophyll vacuoles. Partial purification of the vacuolar malate transporter of Kafanchoe daigremontiana was also performed by hydroxyapatite chromatography (Ratajczak et a f . ,1994). Taking into account the activity of proteoliposomes containing the reconstituted partially purified carrier, purification factors were 44 and 2000 compared to native tonoplast vesicles and reconstituted total tonoplast protein, respectively. Up to now neither for barley nor for Kalanchoe daigremontiana has identification of the malate transporter been possible, since on sodium dodecyl sulfatepolyacrylamide gels, several bands in the molecular mass range between 25 and 7OkDa were visible. Another strategy for the identification of the malate carrier is photoaffinity labelling. Rentsch et a f . (199.5) synthesized N‘(2-hydroxy-(-azido))-diazo-N-3,.5-benzenedicarboxylicacid and 5-azido-isophthalic acid as photoaffinity labels. Both compounds inhibited competitively and effectively malate and citrate uptake of barley vacuoles and Hevea brasifiensis lutoid vesicles. Binding of the photoaffinity label could be inhibited by several organic acids with a similar pattern as observed for citrate transport inhibition. The photoaffinity label specifically labelled a 27 kDa polypepide in both species which was purified to homogeneity from Hevea. The amino acid sequence deduced from the cDNA revealed a hydrophilic protein showing no similarities to other proteins. It has to be demonstrated whether this protein might be a regulatory component of the malate transport system. Recently, Lahjouji et a f . (1996) reported the labelling of a 37 kDa polypeptide of tonoplast vesicles from suspension cultures of Cutharanthus rosem by a photolysable malate analogue N(4-azidosalicyly1)aspartic acid. Polyclonal antibodies against the 37 kDa polypeptide strongly inhibited malate uptake both in tonoplast vesicles and in isolated vacuoles, suggesting the involvement of the 37 kDa polypeptide in vacuolar malate transport. While transport of malate into the vacuole is a secondary active process driven by the vacuolar proton pumps, efflux of malic acid does not require metabolic energy in most cases. Two mechanisms have been postulated for malic acid efflux in the literature: (i) a “lipid diffusion mechanism”, i.e. by movement of the undissociated malic acid through the lipid domains of the tonoplast (Liittge and Smith, 1984) and (ii) efflux via a vacuolar channel. One possible candidate for a malate efflux channel at the tonoplast is the slow-activating vacuolar-type (SV) channel opening at


381

cytosol-positive membrane potentials (Hedrich and Schroder, 1989). Iwasaki et al. (1992) described a channel in vacuoles of the CAM plant Craptopetalum paraguayense corresponding to the SV-type channel opening at high cytoplasmic Ca2+ concentrations, which could be one of the factors necessary for malate efflux from the vacuole (Williamson and Ashley, 1982). In guard cell vacuoles the SV-type channel shows a consistently higher conductivity compared to vacuoles from other materials, indicating that this channel may be involved in malate release during stomata1 closure (Hedrich et al., 1988). However, the work of Amodeo et al. (1994) and Ward and Schroeder (1994) demonstrates conclusively that the SV-type channel is essentially impermeable to anions and, therefore, it is highly unlikely that it would be involved in malate release. The ability to accumulate malic acid and to maintain steep concentration gradients between the vacuole and the cytoplasm as it is the case for photosynthetically active cells of CAM plants (Liittge and Smith, 1984) not only requires energization of the tonoplast but also depends on the biochemical and biophysical properties of t h e tonoplast itself. Supposing that a high degree of order of the tonoplast facilitates malic acid accumulation, measurements of tonoplast fluidity in membrane vesicle preparations from plants grown under certain environmental conditions and exhibiting different modes of photosynthesis might help to understand the regulation of malic acid efflux from the vacuole. Kliemchen et al. (1993) could demonstrate by application of electron paramagnetic resonance spectroscopy (EPR) that growth of Kalanchoe daigremontiana at elevated temperatures as well as the shift from C3 photosynthesis to CAM in Mesemhryanthemum crystallinum led to a rigidification of the tonoplast. In Kalanchoe daigremontiana this increase in the degree of order of the tonoplast corresponded to a decrease in the rate of malic acid release from the tonoplast during the light phase at high temperature (Kliemchen et al., 1993). The difference in the degree of order of the tonoplast isolated from Kalunchoe daigremontiuna grown under 35"C/25"C and 25"C/17"C day/night temperatures, respectively, was due to tonoplast proteins and not to tonoplast lipids. While the degree of order was almost identical in native tonoplast vesicles and liposomes consisting of tonoplast lipids after extraction of proteins in preparations from plants grown under low temperatures, membrane rigidity was significantly lower in protein-free liposomes compared to native tonoplast vesicles from plants grown under elevated temperatures (Schomburg and Kluge, 1994). Protein analysis revealed that the gross polypeptide composition of t h e tonoplast did not change under growth at elevated temperatures with the exception of the appearance of a 35 kDa polypeptide (M. Behzadipour, R . Ratajczak and M. Kluge, unpublished findings). This polypeptide might influence tonoplast rigidity. Another factor that might modulate tonoplast rigidity is the availability of free Ca2+ in the vacuolar lumen. Schomburg (1994) measured the concentrations of citrate, isocitrate, Ca2+ and Mg2+ and

382


dayhight changes of pH and malate levels in the cell sap from Kalanchoe daigremontiana leaves, and calculated the concentration of free Ca2+ using the known equilibrium constants for Ca2+ chelation. It turned out that there was a dayhight oscillation of free Ca2+, with the highest values attained at the end of the dark phase when malate accumulation is maximal. For CAM plants which accumulate malic acid and consequently acidify their vacuole very strongly, such a change in membrane rigidity and therefore in unspecific permeability may play an important role in the release of malate from the vacuole. However, for other plants, whose vacuolar pH is usually between 5 and 6, malate is mainly present in the Hmal- or the maI2- form. In this case diffusion across the lipid bilayer appears to be unlikely, and malate efflux more likely to occur via a channel. Such a channel could also play a role in CAM plants when large amounts of malic acid have already been exported from the vacuole, leading to a rise in vacuolar pH. This has already been implied by the assumption of Luttge and Smith (1984) that H2mal leakage out of the vacuoles may be only one of the factors responsible for change from net malic acid accumulation at the end of the dark period to net malic efflux from the vacuole in the early light period. Although it cannot be answered definitely at the moment if malate efflux from the vacuole occurs via the lipid phase of the tonoplast o r via a vacuolar channel, changes in the degree of order of the tonoplast might influence both possible transport mechanisms, and thus should be taken into account for the interpretation of the regulation of the mechanisms of malate efflux from the vacuole.

IV. AMINO ACIDS Amino acid transport across the tonoplast has been extensively studied in yeast and Neurospora crassa (Boller et al., 1975; Sato et at., 1985a,b; Zerez et al., 1986; Paek and Weiss, 1989), and for yeast it has been shown that at least seven different transport systems mediate the exchange between the cytoplasm and the vacuole. In these fungi the vacuole plays an important role in the temporary storage of high amounts of amino acids. In higher plants, compartmentation studies showed that a considerable amount of amino acids is located within the vacuole as well (Wagner, 1979; Heck et al., 1981; Martinoia et al., 1981; Alibert et al., 1982). The vacuolar proportion of amino acids depends on the total content of free amino acids, which can be manipulated depending on the nutritional condition and the metabolic state of a plant or cell (Table 11; Alibert et al., 1982). Higher vacuolar proportions of amino acids can be found where total contents of free amino acids are increased (Holliinder-Czytko and Amrhein, 1983). Calculation of the extravacuolar concentrations do not suggest that the concentrations of free amino acids in the cytosol remain absolutely constant; however, fluctuations appear to be much less than those found for the vacuolar


383

TABLE I I Cornpartmenrution of amino acids under different conditions Amino acid content ( n m o ~lo7) / Treatment Plants grown in H,O Plants grown in the presence of 20mM KNO, Plants scnesced in the dark for 4 days

Protoplasts

Vacuoles

% in vacuole

23.2 14.6

19.7 5.9

85 52

41.5

30.4

95

Data from Heck et al. (1981). Martinoia et ul. (1981) and E. Martinoia (unpublished findings).

concentrations. On the other hand, a simple calculation of vacuolarextravacuolar amino acids does not reflect the reality in the cell, since at least for green tissue it is known that chloroplasts contain high concentrations of amino acids and that different amino acids are accumulated preferentially in the chloroplast or in the cytosol. It may therefore be argued that amino acids which show a similar distribution between the vacuole and the extravacuolar space are amino acids preferentially localized in the cytosol, while those which are found only in trace amounts in the vacuole may be localized in other compartments. Indeed, in experiments with Cham cells, where the three compartments, chloroplasts-cytosol-vacuole, can be isolated from the same cell, it has been shown that the ratio of the different amino acids between the cytosol and the vacuole is quite constant, indicating that at least in this organism and under normal growth conditions there is no preferential uptake of a specific amino acid (Mimura et al., 1990). In barley leaves senescing in the dark, amino acids are accumulated as a result of protein breakdown (Table 11). The composition of free amino acids does not correspond, however, to the overall composition of protein-bound amino acids. For example, asparagine is by far the predominant amino acid (60-70% of the total free amino acid content (E. Martinoia, unpublished findings)). This suggests that either the proteins are not degraded within the vacuole or that an exchange of amino acids between the cell sap and the cytosol occurs even under conditions where amino acids are accumulated in the vacuole. Direct evidence for such a continuous exchange of amino acids between the vacuole and the cytosol have been reported for the alga Chum aitstralis. Sakano and Tazawa ( 1984) loaded several ''C-labelled amino acids to perfused Cham cells, and observed that these amino acids were exported from the vacuole and readily metabolized to other amino acids which, in their turn, accumulated in the vacuole. In several recent studies, interest was focused on the mechanism of amino acid transfer across the tonoplast. Phenylalanine uptake was found to be

384


stimulated by MgPPi and MgATP but not by other nucleotides (Homeyer and Schultz, 1988; Homeyer et al., 1989). This stimulation could be reversed by nitrate, an inhibitor of the vacuolar H+-ATPase, or by uncouplers, indicating that the stimulation was due to the tonoplast-bound proton pumps. The nature of the stimulation is, however, still unknown. Both valinomycin and nigericin inhibited the ATP-stimulated phenylalanine uptake. However, since the vacuolar K + concentration was not determined, the nature of the energization cannot be deduced unequivocally. Furthermore, a vacuolar accumulation of phenylalanine against its concentration gradient has not been shown. Therefore, it cannot be excluded that the transport rate of the phenylalanine uptake system is modulated by the A* and that the apparent increase in the uptake rate by MgATP does not reflect an energized uptake system but rather a modulation by the A*. The K , for phenylalanine was about 1 mM and inhibition was observed in the presence of other aromatic L-amino acids but not of their D enantiomers. An ATP-dependent but not activated amino acid transport system has been observed for most amino acids in isolated barley vacuoles (Dietz et af.,1990, 1994; Martinoia et af., 1991a; Gorlach and Willms-Hoff, 1992). In the absence of ATP, uptake of most amino acids into the vacuole is very slow. Hydrophobic and small amino acids cross the tonoplast faster than hydrophilic ones. Addition of ATP stimulates to different extents the uptake of the different amino acids. However, ATP appears to act directly as an effector of the carrier by modulating its capacity and not as a source of energy, since stimulation could be observed as response to Mg-free ATP and to the non-hydrolysable ATP analogues AMPPNP (adenylyl imidodiphosphate) and AMPPCP (adenylyl (P,ymethy1ene)diphosphate). However, these analogues have to be present in higher concentrations than ATP to obtain the same effect. The stimulation is specific for ATP and its analogues since other nucleotides failed to stimulate the transport of amino acids across the tonoplast. Interestingly, uptake of some amino acids, for example arginine, glycine or aspartate, is stimulated only by the addition of free ATP, whereas others, such as glutamine, leucine, alanine or phenylalanine, cross the tonoplast in the presence of free ATP as well as of MgATP. Amino acid uptake as a function of free ATP concentration exhibits a sigmoidal saturation curve. Below a concentration of 1 mM the effect of ATP is negligible. Maximal stimulation of uptake was reached between 3 and 5 mM free ATP. In contrast, the MgATP-dependent stimulation of amino acid transport exhibits Michaelis-Menten kinetics with a K , of about 2.5 mM. Addition of sulfhydryl reagents such as pCMBS or NEM stimulates the transporter to a similar or even higher extent than addition of ATP. Both the ATPand the MgATP-stimulated uptake of amino acids can be inhibited by hydrophobic amino acids. Phenylalanine, leucine and methionine inhibit the ATP-stimulated uptake almost completely at a concentration of 4 mM,


385

inhibition by alanine was 45% at this concentration, while u-phenylalanine, phenylethylamine and phenylpropanecarboxylic acid are not inhibitory. Half-maximal inhibition by phenylalanine occurred at 1 mM. Thume and Dietz (1991) showed that this transport system is present also in other plants such as Valerianella and Tulipa. Efflux of amino acids from isolated vacuoles is also stimulated by ATP and inhibited by hydrophobic amino acids, and is very similar to the amino acid uptake system (Dietz et al., 1989). This result suggests that amino acid transport across the tonoplast is mediated in both directions by the same permease. Interestingly, efflux of K f , NO3- and CIis also stimulated by free ATP, and it is tempting to speculate that the two systems are identical (Dietz et al., 1994). A first attempt to identify the ATP-dependent amino acid transporter was made by Thume and Dietz (1991). They succeeded in reconstituting and partially purifying this carrier. Using size exclusion chromatography they estimated the molecular mass of the functional amino acid carrier to be 210 kDa. Free ATP concentrations are reported to be only 5-10% of the total ATP concentration (Yazaki et al.. 1988), and it is therefore unlikely that under normal metabolic conditions the ATP-dependent amino acid transporter could be functional. However, it cannot be excluded that under specific conditions the concentration of free ATP may rise. Transport rates across this permease are very high and would change completely the constituents in the cytosol. It is therefore possible that the general function of the ATP/MgATP-dependent transporter in vivo is to allow, in a nearly closed state, a steady flux between the vacuole and the cytosol which may be a control mechanism for the maintenance of the cytosolic homeostasis. An additional possibility could be that a cytosolic factor which is lost during the vacuole isolation procedure might interact with the transporter and play an additional regulatory role. Furthermore, the total concentration of lipophilic amino acids are in the range of 1 mM and higher (Dietz et al., 1990), and based on their preferential cytoplasmic compartmentation it can be concluded that the ATP-dependent amino acid transporter is in the inhibited state under normal conditions, even for the substrates where the transport is stimulated by MgATP. It is tempting to speculate that under conditions where the content in hydrophobic amino acids is depleted, an efflux of amino acids from the vacuole occurs. However, under normal conditions the amino acid concentration is usually lower in the vacuole than in the cytosol. In this case an activation of the transporter would result in an influx of amino acids into the vacuole. The release of amino acids stored in the vacuole has therefore to be energy-dependent in most cases. Indeed, in the giant alga Charu australis, export of alanine and glutamine from the vacuole is accompanied by a rise in vacuolar pH, indicating the existence of a proton/amino acid symport (Amino and Tazawa, 1989) for the extrusion of amino acids. In higher plants no evidence has been presented up to now for an energized

386

E. MARTlNOlA and R. RATAJCZAK

efflux mechanism, and even in the Chara system it has to be verified that protons are not released by the ATP-dependent amino transporter. An alternative role for this ATP-dependent amino acid transporter has been proposed by Davies and Sanders (1995). Using the patch clamp technique they observed currents which corresponded to the fluxes observed in tracer experiments and showed that the potassium current was much higher than the current observed after addition of arginine. They argued that this channel may not only act as a K+ mobilization pathway but also as a counterion (shunt) conductance, allowing the two vacuolar proton pumps to acidify the vacuolar lumen. For positively charged amino acids an alternative vacuolar transport system has been demonstrated (Martinoia et a f . , 1991a) which does not depend on ATP. Arginine uptake in the absence of ATP is a saturable process ( K , 0.3-0.4 mM), and can be competitively inhibited by lysine but not by neutral or acidic amino acids (Martinoia et af., 1991a). Basic amino acids have been shown to accumulate in the vacuole-like lutoids of Hevea brasifiensis (D’Auzac et a f . , 1982; Marin et af., 1982), suggesting an energy-dependent uptake of these amino acids into the vacuoles of at least some plants. ACC is the precursor for ethylene biosynthesis, and it was discussed whether vacuolar localization of this amino acid might play a role in the modulation of ethylene synthesis. Indeed, 6040% of ACC was found to be localized within the vacuole (Tophof el al., 1989; Saftner and Martin, 1993). As for the proteogenic amino acids, transport was strongly stimulated, but not energized, by ATP since NO3-, which inhibits the vacuolar H+-ATPase, did not cause an inhibition of ACC uptake and since ATP could be replaced by the non-hydrolysable ATP analogue AMPPNP (Saftner and Martin, 1993). However, in this case only MgATP was able to stimulate the transport but not free ATP. The transport system exhibits a high affinity for ACC (20 pM), and is inhibited by valinomycin and protonophores, indicating that either the A T , ApH or both modulate the transport activity of ACC. In this case, too, hydrophobic L-amino acids inhibit the MgATP-dependent uptake of ACC (Saftner, 1994). However, large differences in inhibition could be observed when the four stereoisomers of l-amino-2-ethylcyclopropane-lcarboxylic acid were tested, indicating that recognition by the carrier is not only based on a hydrophobic L-amino acid.

V. POLYAMINES AND PEPTIDES Polyamines are ubiquitous polycationic metabolites. They are reported to have several important €unctions in both plants and animals (Galston and Sawhney, 1990) such as regulation of the cell cycle and cell differentiation.


387

In plants, putrescine can accumulate to high concentrations under a variety of stress conditions, and it was discussed that such a stress-mediated increase may be an adaptive advantage to plants. A portion of the polyamines are localized within the vacuole (Pistocchi et al., 1988; DiTomaso et al., 1992). Transport studies with carrot vacuoles showed a biphasic concentration dependence of uptake of the polyamine spermidine. Below 1 mM a saturable component ( K , 62 pM) was observed, while at higher concentrations up to 100 mM uptake rates increased proportionally to the concentration of spermidine in the medium (Pistocchi et af., 1988). Energization of polyamine uptake was not analysed in this study. A vacuolar channel which could correspond to the non-saturable uptake system observed by Pistocchi et al. (1988) was described by Colombo et al. (1992). The channel is activated by cytosol-positive potentials and is outward rectifying. The channel described by Colombo et al. (1992) shows striking similarities to the vacuolar SV channel (Hedrich et al., 1986; Hedrich and Schroeder, 1989), and it is tempting to speculate that it is indeed the same channel. Uptake of putrescine and the divalent cation-herbicide paraquat into barley vacuoles occurred by an ATP-modulated transporter similar to that described for amino acids and peptides. The transport system had only a low affinity for the polyamine and the herbicide (7.5 and 5.4 respectively) (Mornet et al., 1997). Like animal lysosomes, vacuoles contain a wide variety of acidic hydrolases. Most of the proteolytic activity (endo- and carboxypeptidases) observed in a plant cell are restricted to the vacuole (Boller and Wiemken, 1986). The role of these proteases is still unknown. It has been suggested that along with other hydrolytic activities they may contribute to the defence against herbivores and pathogens and to the general turnover of proteins (Boller, 1986; Boller and Wiemken, 1986). Isolated vacuoles can degrade some of the proteins they contain (Canut et al., 1985; Moriyasu and Tazawa, 1988), and misfolded proteins containing the endoplasmic reticulum retention motif HDEL (Pueyo et al., 1995). However, the question of whether the vacuole is involved in the degradation of cytosolic and plastid proteins is still open. It has been shown that chloroplasts degrade their own proteins by means of internal proteases (for a review, see Vierstra, 1993) and it would be therefore possible that oligopeptides deriving from partly digested proteins would be transported to the vacuole for final degradation. Di- and tripeptides are indeed transported into the vacuole. However, the transport system involved appears to be the same as described for amino acids (Jamai’ et al., 1995), and the general characteristics make it unlikely that this transport system is involved in the final degradation of proteins. Recent experiments using an alternative dipeptide with a very high specific activity confirmed the published results (A. Jamai and E . Martinoia, unpublished findings). Ubiquitin has been detected in plant vacuoles (Beers et al., 1992), suggesting that transport systems recognizing this compound may direct polypeptides into the vacuole.

388


VI. TRANSPORT OF SECONDARY PRODUCTS OF PLANT METABOLISM Plants synthesize an enormous number of secondary metabolites. Many of these have been shown to be exclusively localized in the vacuole. Two excellent reviews have been published on the role of the vacuole in plant secondary metabolism (Matile, 1984, 1987), and the article by Wink in this volume is devoted to the compartmentation and transport of these substances. We will therefore limit our contribution to a brief discussion of the different possibilities how such products, which are often potentially toxic, can be detoxified efficiently by the plant cell. In young shoots of Sorghum tricolor, dhurrin, a cyanogenic glycoside, makes up as much as 30% of the tissue dry weight, and is exclusively localized in the vacuole of epidermal cells (Saunders and Conn, 1978). This fact implies that the plant is not only able to transfer this substance into the vacuole but that mechanisms to accumulate such a substance very efficiently have evolved. A secondary product/H+ antiport mechanism may function in some cases. However, maximal accumulation is determined by the ApH between the cytosol and the vacuole, which normally is not much higher than 2. An antiport mechanism where more than one proton is exchanged for the secondary product may result in a higher concentration gradient between the cytosol and the vacuole. Conformational changes of the secondary product within the vacuole may be a further possibility to trap secondary substances. Indeed, apigenin-7-(6-0-malonyl)glucoside,a vacuolar pigment of parsley, is believed to be trapped inside the vacuole as a result of conformational changes at acidic pH (Matern et a f . , 1983). The trans isomer of o-coumaric acid is glycosylated in the cytosol. It is readily taken up by isolated vacuoles in which it is converted non-enzymically to the cis form, which does not cross the tonoplast (Rataboul et af., 1985). Enzymic conversion of 1sinapoylglucose within the vacuole has been proposed by Sharma and Strack (1985), who localized 1-sinapoylglucosex-malate sinapoyltransferase in the cell sap. Uptake of secondary products is often coupled to the pH gradient (Rataboul et al., 1985; Hopp and Seitz, 1987), suggesting that an effective retrieval is important even when trapping mechanisms exist. However, plants are not only able to accumulate their own products within the vacuole but also foreign biotic or abiotic glucosides or glutathione conjugates. It has been shown that coumarylglucosides such as esculin and scopolin are taken up most probably via an H+-antiport mechanism in barley mesophyll vacuoles, although the substances are not produced by barley (Werner and Matile, 1985). The same vacuoles are also able to transport glucosides of herbicides (Gaillard et al., 1994). An alternative mechanism for efficient vacuolar accumulation has been shown for xenobiotics which have been conjugated to glutathione by the action of glutathione Stransferases (Farrago et af., 1994). In this case, transport is directly energized

01 3 6

1 sucrose

12 16 20

25mm

I

fructose

T

Fig. 1. Transport of organic molecules across the tonoplast (centre: question marks indicate an unclear transport mechanism) and examples of in vivo uptake kinetics of some organic molecules into barley vacuoles isolated from photosynthesizing protoplasts (periphery: closed symbols indicate respective compound detected in protoplasts, open symbols indicate respective compound detected in the vacuolar fraction. Data from Kaiser el al., 1982).

390


by MgATP as in animals (Martinoia et al., 1993), and there is evidence that this ATPase belongs to the family of ABC transporters (R. Tommasini and E. Martinoia, unpublished findings). Calculations using known cytosolic concentrations for the metabolites involved show that such a direct energization may create a concentration gradient of 1-10' if a stoichiometry of 1:l for ATP hydrolysis and glutathione conjugate transported into the vacuole is assumed. Recently, it has been postulated that in maize the final step of anthocyanin biosynthesis consists in the transfer of a glutathione moiety to anthocyanin, and that this conjugate is then transferred into the vacuole (Marrs et al., 1995). All these observations suggest that plants contain both specific transport mechanisms responsible for the detoxification of their native secondary substances and general transporters which recognize different related substances and which are able to protect a plant if exposed to toxins or abiotic poisons. Further work is necessary to elucidate the number of transporters involved and their respective specificity.

VII.

CONCLUSION

A large number of transporters involved in the transfer of organic compounds from the cytosol into the vacuole have been described from the kinetic point of view (Fig. 1). In contrast to the vacuolar proton pumps, despite encouraging results at first, these transporters have not been identified, mainly due to their very low abundance. Further work will therefore be needed to identify these transporters and to understand their role in the maintenance of cytosolic homeostasis.

ACKNOWLEDGEMENTS We would like to thank Dr Felix Keller (University of Zurich), Professor Dr Ulrich Liittge, Professor D r Manfred Kluge, Dr Elke Fischer-Schliebs and Dr Sabine Steiger (Technische Hochschule Darmstadt) for critical reading of the manuscript and Mrs Doris Schafer (Technische Hochschule Darmstadt) for providing the artwork.

REFERENCES Alibert, G., Carrasco, A . and Boudet, A . M. (1982). Changes in biochemical composition of vacuoles isolated from Acer pseudoplatanus L. during cell culture. Biochimica et Biophysicu Acta 721, 22-29. Allan, S. and Raven, J . A . (1987). Intracellular pH regulation in Ricinm comrnunis grown with ammonium or nitrate as N source: the role of long distance transport. Journal of Experimental Botany 38, 580-596.


391

Amino, S . and Tazawa, M. (1989). Dependence of tonoplast transport of amino acids on vacuolar pH in Chara cells. Proceedings of the Japanese Academy, Series B 65, 34-36. Amodeo, G., Escobar, A. and Zeiger, E. (1994). A cationic channel in the guard cell tonoplast of Allium cepa. Plant Physiology 105, 999-1006. Beers, E. P., Moreno, T. N. and Callis, J. A. (1992). Subcellular localization of ubiquitin and ubiquinated proteins in Arabidopsis thaliana. Journal of Biological Chemistry 261, 15 432-15 439. Bettey, M. and Smith, J. A. C. (1993). Dicarboxylate transport at the vacuolar membrane of the CAM plant Kalanchoe dainrernontiana: sensitivitv to proteinmodifying and sulphydiyl reagents. Biochimica et Biophysica .Act; 1152, 270-279. Bisaccia, F., Indivieri, C. and Palmieri, F. (1988). Purification and reconstitution of two anion carriers from rat liver mitochondria: the dicarboxylate and the 2-oxoglutarate carrier. Biochimica et Biophysica Acta 933, 229-240. Blom-Zanstra, M., Koot, H. T. M., Hattum, J . and Borstlap, A. C. (1990). Interactions of uptake of malate and nitrate into isolated vacuoles from lettuce leaves. Planta 183, 10-16. Boller, T. (1986). Roles of proteolytic enzymes in interactions of plants with other organisms. In “Plant proteolytic enzymes” (M. J. Dalling, ed.), Vol. 1, pp. 67-96. CRC Press, New York. Boller, T. and Alibert, G. (1983). Photosynthesis in protoplasts from Meliliotw alba: distribution of products between vacuole and cytosol. Zeitschrijt fur Pflanzenphysiologie 110, 231-238. Boller, T. and Wiemken, A. (1986). Dynamics of vacuolar compartmentation. Annual Review of Plant Physiology 37, 137-164. Boller, T.. Diirr, M. and Wiemken, A. (1975). Characterization of a specific transport system for arginine in isolated yeast vacuoles European Journal of Biochemistry 54, 81-91. Bolli, R.,Naecz, K. A. and Azzi, A. (1989). Monocarboxylate and a-ketoglutarate carriers from bovine heart mitochondria. Journal of Biological Chemistry 264, 18 024-18 030. Bouyssou, H., Canut, H. and Marigo, G. (1990). A reversible carrier mediates the transport of malate at the tonoplast of Catharanthus roseus cells. FEBS Letters 93, 73-76. Bouzayen, M., Latche, A , , Alibert, G. and Pech, J. C. (1989). Carrier mediated uptake of 1-(malony1amino)cyclopropane-1-carboxylicacid in vacuoles isolated from Catharanthus roseus cells. Plant Physiology 91, 1317-1322. Briskin, D. P., Thornley, W. R. and Wyse, R. E. (1985). Membrane transport in isolated vesicles from sugarbeet taproot. Evidence for a sucrose/H+ antiport. Plant Physiology 78, 871-875. Brown, S. C. and Coombe, B. G. (1982). Sugar transport by an enzyme complex at the tonoplast of grape pericarp cells. Naturwissenschaften 69, 43-44. Buchanan-Bollig, I. C. and Kluge, M. (1981). Crassulacean acid metabolism (CAM) in Kalanchoe daigrernontiana: temperature response of phosphoenolpyruvate (PEP)-carboxylase in relation to allosteric effectors. Planta 152, 181-188. Buser, C. and Matile, P. (1977). Malic acid in vacuoles isolated from Bryophyllum leaf cells. Zeitschrijt fur Pflanzenphysiologie 82, 462-466. Buser-Suter, C., Wiemken, A. and Matile, P. (1982). A malic acid permease in isolated vacuoles of a crassulacean acid metabolism plant. Plant Physiology 69, 456459. Canut. H.. Alibert, G.. Boudet, A . M. (1985). Hydrolysis of intracellular proteins

392


in vacuoles isolated from Acer pseudoplatanus L. cells. Plant Physiology 79, 1090-1093. Cerana, R., Giromini, L. and Colombo, R. (1995). Malate-regulated channels permeable to anions in vacuoles of Arubidopsis thaliana. Australian Journal of Plant Physiology 22, 115-121. Colombo, R., Cerana, R. and Bagni, N. (1992). Evidence for polyamine channels in protoplasts and vacuoles of Arabidopsis thaliana cells. Biochemical and Biophysical Research Communications 182, 1187-1 192. Coyaud, L., Kurkdjian, A , , Kado, R. and Hedrich, R. (1987). Ion channels and ATP-driven pumps involved in ion transport across the tonoplast of sugarbeet vacuoles. Biochimica et Biophysica Acta 902, 263-268. Daie, J. and Wilusz, J. E. (1987). Facilitated transport of glucose in isolated phloem segments of celery. Plant Physiology 84, 711-715. D’Auzac, J., Cretin, H., Marin, B. and Lioret, C. (1982). A plant vacuolar system: the lutoids from Hevea brasiliensis latex. Physiologie Vkgktale 20, 31 1-331. Davies, J. M. and Sanders, D. (1995). ATP, pH and Mg2+ modulate a cation current in Beta vulgaris. Journal of Membrane Biology 145, 75-86. Dietz, K.-J., Martinoia, E. and Heber, U. (1989). Mobilisation of vacuolar amino acids in cells as affected by ATP and the level of cytosolic amino acids: ATP regulates but appears not to energize vacuolar amino acid release. Biochimica et Biophysica Acta 984, 57-62. Dietz, K.-J., Jager, R., Kaiser, G . and Martinoia, E. (1990). Amino acid transport across the tonoplast of vacuoles isolated from barley mesophyll vacuoles. Plant Physiology 92, 123-129. Dietz, K.-J., Canut, H. and Marigo, G. (1992). Identification of an essential histidine residue at the active site of the tonoplast malate carrier in Cutharanthus roseus cells. Journal of Membrane Biology 129, 137-143. Dietz, K.-J., Klughammer, B., Lang. B. and Thume, M. (1994 . Solute transport across the tonoplast of barley mesophyll vacuoles: Mg determines the specificity, and ATP and lipophilic amino acids the activity of the amino acid carrier. Journal of Membrane Biology 137, 151-158. DiTomaso, J., Hart, J. J. and Kochian, L. V. (1992). Transport kinetics and metabolism of exogenously applied putrescine in roots of intact maize seedlings. Plant Physiology 98, 611-620. Doll, S., Rodier, F. and Willenbrink, J. (1979). Accumulation of sucrose in vacuoles isolated from red beet tissue. Planta 144, 407411. Douce, R. and Neuburger, M. (1989). The uniqueness of plant mitochondria. Annual Review of Plant Physiology 40,371414. Edwards, G. E. and Huber, S. C. (1981). The C4 pathway. In “The Biochemistry of Plants: A Comprehensive Treatise” (M. D. Hatch and N. K. Boardman, eds), Vol. 8, pp. 237-281. Academic Press, New York. Farrago, S., Brunold, C. and Kreuz, K. (1994). Herbicide safeners and glutathione metabolism. Physiologia Plantarum 91, 537-542. Franco, A , , Ball, E. and Luttge, U. (1992). Differential effects of drought and light levels on accumulation of citric and malic acids during CAM in Clusia. Plant, Cell and Environment 15, 821-829. Frehner, M., Keller, F. and Wiemken, A. (1984). Fructan metabolism in Helianthus tuberosus: compartmentation in protoplasts and vacuoles isolated from tubers. Journal of Plant Physiology 116, 197-208. Gaillard, C., Dufaud, A., Tommasini, R., Kreuz, K., Amrhein, N. and Martinoia, E. (1994). A herbicide antidote (safener) induces the activity of both the herbicide detoxifying enzyme and of a vacuolar transporter for the detoxified

2

+


393

herbicide. FEBS Letters 352, 219-221. Galston, A. W. and Sawhney, R. K. (1990). Polyamines in plant physiology. Plant Physiology 94, 406-410. Geiger, D. R. (1975). Phloem loading. I n “Encyclopedia of Plant Physiology” (H. M. Zimmermann and J. A. Milburn, eds), Vol. 1, pp. 395-431. Springer-Verlag, Berlin. Gerhardt, R. and Heldt, H. W. (1984). Measurement of subcellular metabolite levels in leaves by fractionation of freeze-stopped material in nonaqueous media. Plant Physiology 75, 542-547. Gerhardt, R . , Stitt, M. and Heldt, H. W. (1987). Subcellular metabolite levels in spinach leaves. Plant Physiology 83, 399-407. Getz, H. P. (1987). Accumulation of sucrose in vacuoles released from isolated beet root protoplasts by both direct sucrose uptake and UDP-glucose-dependent group translocation. Plant Physiology and Biochemistry 25, 573-580. Getz, H. P. (1991). Sucrose transport in tonoplast vesicles of red beet roots is linked to ATP hydrolysis. Plantu 185, 261-268. Getz, H. P. and Klein, M. (1995). Characteristics of sucrose transport and sucrose-induced H+-transport on the tonoplast of red beet (Beta vulgaris L.) storage tissue. Plant Physiology 107, 459-467. Getz, H. P., Grosclaude, J., Kurkdjian, A . , Lelievre, F., Maretzki, A. and Guern, J. (1993). Immunological evidence for existence of a carrier protein for sucrose transport in tonoplast vesicles from red beet root storage tissue. Plant Physiology 102, 751-760. Giaquinta, R. T. (1983). Phloem loading of sucrose. Annual Review of Plant Physiology 34, 347-387. Gorlach. J. and Willms-Hoff, I . (1992). Glycine uptake into barley mesophyll vacuoles is regulated but not energized by ATP. Plant Physiology 99, 134-139. Gout, E., Bligny, R., Pascal, N. and Douce, R. (1993). I3C Nuclear magnetic resonance studies of malate and citrate synthesis and compartmentation in higher plant cells. Journal of Biological Chemistry 268, 3986-3992. Greenway, H., Winter, K. and Liittge, U . (1978). Phosphoenolpyruvate carboxylase during development of crassulacean acid metabolism and during a diurnal cycle in Mesembryanthemum crystallinum. Journal of Experimental Botany 29, 547-559. Greutert, H. and Keller, F. (1993). Further evidence for stachyose and sucrose H+/antiporters on the tonoplast of Jerusalem artichoke (Stachys sieboldii) tubers. Plant Physiology 102, 1317-1322. Grob, K. and Matile, P. (1980). Compartmentation of ascorbic acid in vacuoles of horseradish root cells. Note on vacuolar peroxidase. Zeitschrift f u r Pflanzenphysiologie 98, 235-243. Guthrie, J. D. (1934). Diurnal change in the citric acid content of the leaves of Bryophyllum calycinum. American Journal of Botany 21, 706. Guy, M., Reinhold, L. and Michaeli, D. (1979). Direct evidence for a sugar transport mechanism in isolated vacuoles. Plant Physiology 64,6144. Heck, U., Martinoia. E. and Matile, P. (1981). Subcellular localization of acid proteinase in barley mesophyll protoplasts. Planta 151, 198-200. Hedrich, R. and Schroeder, J . I . (1989). The physiology of ion channels and electrogenic pumps in higher plants. Annual Review of Plant Physiology 40, 539-569. Hedrich, R., Fliigge, U. I. and Fernandez, J. M. (1986). Patch-clamp studies of ion transport in isolated plant vacuoles. FEBS Letters 204, 228-232. Hedrich, R . , Barbier-Brygoo, H., Felle, H., Flugge, U . I . , Liittge, U., Maathius,

394


F. J. M., Mark, S . , Prins, H. B. A , , Raschke, K. M., Schnabl, H., Schroeder, J. M., Struve, I . , Taiz, L. and Ziegler, P. (1988). General mechanisms for solute transport across the tonoplast of plant vacuoles: a patch-clamp survey of ion channels and proton pumps. Botanica Acta 101, 7-13. Heineke, D., Riens, B., Grosse, H., Hoferichter, P., Flugge, U . I. and Heldt, H. W. (1991). Redox transfer across the inner chloroplast envelope. Plant Physiology 95, 1131-1137. Hoffland, E., Van de Boogard, R., Nelemans, J. and Findenegg, G. (1992). Biosynthesis and root exudation of citric and malic acids in phosphate-starved rape plants. New Phytologist 122, 675480. Hollander-Czytko, H . and Amrhein, N. (1983). Subcellular compartmentation of shikimic acid and phenylalanine in buckwheat cell suspension cultures grown in the presence of shikimate pathway inhibitors. Plant Science Letters 29, 89-96. Homeyer, U . and Schultz, G. (1988). Transport of phenylalanine into vacuoles isolated from barley mesophyll protoplasts. Planta 176, 378-382. Homeyer, U . , Litek, K., Huchzemeyer, B. and Schultz, G. (1989). Uptake of phenylalanine into barley vacuoles is driven by both tonoplast adenosine triphosphatase and pyrophosphatase. Plant Physiology 89, 1388-1393. Hopp, W. and Seitz, H. U. (1987). The uptake of acylated anthocyanin into isolated vacuoles from a cell suspension culture of Daucus carota. Planta 170, 74-85. Indivieri, C., Capobianco, L., Kramer, R. and Palmieri, F. (1989). Kinetics of the reconstituted dicarboxylate carrier from rat liver mitochondria. Biochimica et Biophysica Acta 977, 187-193. Iwasaki, I., Arata, H., Kijiama, H . and Nishimura, M. (1992). Two types of ion channels involved in the malate ion transport across the tonoplast of a crassulacean acid metabolism plant. Plant Physiology 98, 1494-1497. Jamai’, A., Gaillard, C., Delrot, S. and Martinoia, E. (1995). Dipeptide transport in barley mesophyll vacuoles. Planta 196, 430-433. Kaiser, G. and Heber, U . (1984). Sucrose transport into vacuoles isolated from barley mesophyll protoplasts. Planta 161, 562-568. Kaiser, G., Martinoia, E. and Wiemken, A. (1982). Rapid appearance of photosynthetic products in the vacuoles isolated from barley mesophyll protoplasts by a new fast method. Zeitschrift fur PJlanzenphysiologie 107, 103-1 13. Kaplan, R . S. and Pedersen, P. L. (1985). Isolation and reconstitution of the n-butylmalonate-sensitive dicarboxylate transporter from rat liver mitochondria. Journal of Biological Chemistry 260, 10 293-10 298. Keller, F. (1988). A large-scale isolation of vacuoles from protoplasts of mature carrot taproots. Journal Plant Physiology 132, 199-203. Keller, F. (1992). Transport of stachyose and sucrose by vacuoles of Japanese artichoke (Stachys sieboldii) tubers. Plant Physiology 98, 4 4 2 4 5 . Keller, F. and Matile, P. (1985). The role of the vacuole in storage and mobilization of stachyose in tubers of Stachys sieboldii. Journal of Plant Physiology 119, 369-380. Keller, F. and Matile, P. (1989). Storage of sugars and mannitol in petioles of celery leaves. New Phytologist 113, 291-299. Keller, F. and Wiemken, A. (1982). Differential compartmentation of sucrose and gentianose in the cytosol and vacuoles of storage root protoplasts from Gentiana lutea. Plant Cell Reports 1, 274-277. Kliemchen, A , , Schomburg., M., Galla, H. J., Liittge, U. and Kluge, M. (1993). Phenotypic changes in the fluidity of the tonoplast in response to temperature and salinity stress. Planta 189, 403-409.


395

Kluge, M. and Osmond, C. B. (1972). Studies on phosphoenolpyruvate carboxylase and other enzymes of crassulacean acid metabolism of Bryophyllum tubiflorum and Sedum praealtum. Zeitschrift fur Pflanzenphysiologie 66, 97-105. Kluge, M. and Ting, I . P. (1979). “Crassulacean Acid Metabolism. Analysis of an Ecophysiological Adaptation”. Springer-Verlag, Berlin. Koster, K . L. and Lynch, D. V. (1992). Solute accumulation and compartmentation during the cold acclimation of puma rye. Plant Physiology 98, 108-113. Lahjouji, K . , Carrasco, A., Bouyssou, H . , Cazaux, L., Marigo, G. and Canut, H. (1996). Identification with a photoaffinity reagent of a tonoplast protein involved in vacuolar malate transport of Catharanthus roseus. Plant Journal 9, 799-808. Leigh. R. A. (1983). Methods, progress and potential for the use of isolated vacuoles in studies of solute transport in higher plant cells. Physiologia Plantarum 57, 390-396. Leigh, R. A., ap Rees, T., Fuller, W. A. and Banfield, J. (1979). The location of acid invertase activity and sucrose in the vacuoles of storage roots of beetroot. Biochemical Journal 178, 539-545. Luttge, U. (1988). Day-night changes of citric-acid levels in crassulacean acid metabolism: phenomenon and ecophysiological significance. Plant Cell and Environment 11, 445451. Luttge, U . and Smith, J. A. C. (1984). Mechanism of passive malic-acid efflux from vacuoles of the CAM plant Kalanchoe daigremontiana. Journal of Membrane Bi010gy 81, 149-158. Liittge, U . , Smith, J. A . C., Marigo, G. and Osmond, C. B. (1981). Energetics of malate accumulation in the vacuoles of Kalanchoe tubiflora cells. FEBS Letters 126, 81-84. MacLennan, D. H., Beevers, H. and Harley, J. L. (1963). ‘Compartmentation’ of acids in plant tissues. Biochemical Journal 89, 316327. Maretzki, A. and Thom, M. (1988). High performance liquid chromatography-based reevaluation of disaccarides produced upon incubation of sugarcane vacuoles with UDP-glucose. Plant Physiology 88, 266269. Marigo, G., Bouyssou, H. and Belkoura. M. (1985). Vacuolar efflux of malate and its influence on nitrate accumulation in Catharanthus roseus cells. Plant Science 39, 97-103. Marigo, G., Bouyssou, H. and Laborie, D. (1988). Evidence for a malate transport into vacuoles isolated from Catharanthus roseus cells. Botanica Acta 101, 187-191. Marin, B., CrCtin, H. and D’Auzac, J. (1982). Energization of solute transport and accumulation at the tonoplast in Hevea latex. Physiologie Vkgktale 20, 333346. Marquardt-Jarczyk, G. and Luttge, U. (1990). Anion transport at the tonoplast of mesophyll cells of the CAM plant Kalanchoe daigremontiana. Journal of Plant Phy.yiology 136, 129-136. Marrs, K. A , , Alfenito, M. R., Lloyd, A. M. and Walbot, V. (1995). A glutathion S-transferase involved in vacuolar transfer encoded by the maize gene Bronze-2. Nature 375, 397-400. Martinoia, E. (1992). Transport processes in vacuoles of higher plants. Botunica Acta 105, 232-245. Martinoia. E. and Rentsch, D. (1994). Malate compartmentation - responses to a complex metabolism. Annual Review of Plant Physiology and Plant Molecular Biology 45, 447-467. Martinoia E., Heck, U . and Wiemken, A. (1981). Vacuoles as storage compartments

396


of nitrate in barley leaves. Nature 289, 292-294. Martinoia, E., Fliigge, U. I., Kaiser, G., Heber, U. and Heldt, H. W. (1985). Energy dependent uptake of malate into vacuoles isolated from barley mesophyll protoplasts. Biochimica et Biophysica Acta 806, 311-319. Martinoia, E., Kaiser, G., Schramm, M. J. and Heber, U. (1987). Sugar transport across the plasmalemma and the tonoplast of barley mesophyll protoplasts: evidence for different transport systems. Journal of Plant Physiology 131, 467-478. Martinoia, E., Vogt, E. and Amrhein, N. (1990). Transport of malate and chloride into barley mesophyll vacuoles. Different carriers are involved. FEBS Letters 261, 109-1 11. Martinoia, E . , Thume, M., Vogt, E . , Rentsch, D. and Dietz, K. J. (1991a). Transport of arginine and aspartic acid into isolated barley mesophyll vacuoles. Plant Physiology 97, 644-650. Martinoia, E., Vogt, E., Rentsch, D. and Amrhein, N. (1991b). Functional reconstitution of the malate carrier of barley mesophyll vacuoles in liposomes. Biochirnica et Biophysica Acta 1062, 271-278. Martinoia, E . , Grill, E., Tommasini, R., Kreuz, K. and Amrhein, N. (1993). ATP-dependent glutathione S-conjugate ‘export’ pump in the vacuolar membrane of plants. Nature 364, 247-249. Matern, U., Heller, W. and Himmelspach, K. (1983). Conformational changes of apigenin 7-0-(6-O-malonylglucoside), a vacuolar pigment from parsley, with solvent composition and proton concentration. European Journal of Biochemistry 133, 439448. Matile, P. (1978). Biochemistry and function of vacuoles. Annual Review of Plant Physiology 29, 193-213. Matile, P. (1984). Das toxische Kompartiment der Pflanzenzelle. Naturwissenschaften 71, 18-24. Matile, P. (1987). The sap of plant cells. New Phytologist 105, 1-26. Milner, 1. D., Ho, L. C. and Hall, J . L. (1995). Properties of proton and sugar transport at the tonoplast of tomato (Lycopersicum esculentum) fruit. Physiologia Plantarurn 94, 399-410. Mimura, T., Sakano, K. and Tazawa, M. (1990). Changes in the subcellular distribution of free amino acids in relation to light conditions in cells of Chara corallina. Botanica Acta 103, 42-47. Moriyasu, Y. and Tazawa, M. (1988). Degradation of proteins artificially introduced into vacuoles of Churu austrulis. Plant Physiology 88, 1092-1096. Mornet, C., Mondory, C., Gaillard, C. and Martinoia, E. (1997). Transport of paraquat and polyamines across a vacuolar membrane of barley vacuoles. Plant Physiology and Biochemistry 35 (in press). Moskowitz, A. H. and Hrazdina, G. (1981). Vacuolar contents of fruit subepidermal cells from Vitis species. Plant Physiology 68, 68-92. Nierhaus, D. and Kinzel, H. (1970). Vergleichende Untersuchungen iiber die organischen Sauren in Blattern hoherer Pflanzen. Zeitschrift fur Pfianzenphyiologie 64, 17-123. Niland, S. and Schmitz, K . (1995). Sugar transport into storage tubers of Stachys sieboldii Miq. : evidence for symplastic unloading and stachyose uptake into storage vacuoles by an H+-antiport mechanism. Botanica Acta 108, 24-33. Nishida, K. and Tominaga, 0. (1987). Energy-dependent uptake of malate into vacuoles isolated from CAM plant Kalanchoe daigrernontiana. Journal of Plant Physiology 127, 385-393. Ohwaki. Y. and Hirata, H. (1992). Differences in carboxylic acid exudation among


397

P-starved leguminous crops in relation to carboxylic acid contents in plant tissues and phospholipid level in roots. Soil Science- and Plant Nutrition 38, 235243. Oleski, N.. Mahdavi, P. and Bennett, A . B. (1987). Transport properties of the tomato fruit tonoplast. 11. Citrate transport. Plant Physiology 84, 997-1000. Olivares, E . , Faist, K.. Kluge, M. and Luttge, U. (1993). 14C02pulse-chase labelling in Clusia minor L. Journal of Experimental Botany 44, 1527-1533. Paek, Y. and Weiss, R. L. (1989). Identification of an arginine carrier in the vacuolar membrane of Neurospora crassa. Journal of Biological Chemistry 264, 72857290. Pantoja, 0. Gelli, A . and Blumwald, E. (1992). Characterization of vacuolar malate and Kf channels under physiological conditions. Plant Physiology 100, 1137-1141. Paul, M. J . and Cockburn, W. (1989). Pinitol. a compatible solute in Mesemhryanthemum crystallinurn L.? Journal of Experimental Botany 40, 1093-1098. Pistocchi, R . , Keller, F., Bagni, N . and Matile, P. (1988). Transport and subcellular localization of polyamines in carrot protoplasts and vacuoles. Plant Physiology 87, 514-518. Pollock, C. J . and Cairns, A . J . (1991). Fructan metabolism in grasses and cereals. Annual Review of Plant Physiology and Plant Molecular Biology 42, 77-101. Popp. M. and Smirnoff. N . (1995). Polyol accumulation and metabolism during water deficit. In “Environment and Plant Metabolism” (N. Smirnoff, ed.), pp. 199-215. Bios. Oxford. Preissner, J . and Komor, E . (1988). Analysis of the reaction products from incubation o f sugarcane vacuoles with uridine diphosphate-glucose: no evidence for a group translocator. Plant Physiology 88, 259-265. Preissner, J . and Komor, E . (1991). Sucrose uptake into vacuoles of sugarcane suspension cells. Planta 186, 109-1 14. Preissner. J., Spriigel, H. and Komor, E. (1991). Solute distribution between vacuole and cytosol of sugarcane suspension cells: sucrose is not accumulated in the vacuole. Planta 186, 203-21 1. Pueyo, J . J . , Chrispeels, M. J . and Herman, E. M. (1995). Degradation of transport-competent destabilized phaseolin with a signal for retention in the endoplasmic reticulum occurs in the vacuole. Planta 196, 586496. Raschke, K. (1979). Movements of stomata. I n “Encyclopedia of Plant Physiology” (W. Haupt and E. Feinleib, eds), Vol. 7, pp. 57-100. Springer-Verlag. Berlin. Raschke, K., Hedrich, R . , Reckmann, U. and Schroeder, J . (1988). Exploring biophysical and biochemical components of the osmotic motor that drives stomata1 movement. Botanica Acta 101, 283-294. Rataboul, P., Alibert, G., Boller, T. and Boudet, A . M. (1985). Intracellular transport and vacuolar accumulation of o-coumaric acid glucoside in Melilotus alba protoplasts. Biochimica et Biophysica Acta 816, 25-36. Ratajczak, R . , Kemna, I. and Liittge, U. (1994). Characteristics, partial purification and reconstitution of the vacuolar malate transporter of the C A M plant Kalanchot? daigremontiana Hamet et Perrier de la Bithie. Planfa 195, 226236. Rausch, T., Butcher, D. N. and Taiz, L. (1987). Active glucose transport and proton pumping in tonoplast membranes of Zea mays L. coleoptiles are inhibited by anti-H+-ATPase antibodies. Plant Physiology 85, 996-999. Rautenkranz, A. A . F., Li, L., Machler, F., Martinoia, E. and Oertli, J. (1994). Transport of ascorbic and dehydroascorbic acids across protoplast and vacuole

398


membranes isolated from barley (Hordeum vulgare L., cv. Gerbel) leaves. Plant Physiology 106, 187-193. Rentsch, D. and Martinoia, E. (1991). Citrate transport into barley mesophyll vacuoles - comparison with malate uptake activity. Planta 184, 532-537. Rentsch, D., Gorlach, J., Vogt, E., Amrhein, N. and Martinoia, E. (1995). Identification and sequencing of a citrate-binding protein of the vacuolar-like lutoid membrane of Hevea brasiliensis. Journal of Biological Chemistry 270, 30 525-30 531. Saftner, R.A. (1994). Stereoselectivity and structural determinants in molecular recognition by the ACC transport system in isolated maize mesophyll vacuoles. Physiologia Plantarum 92, 543-554. Saftner, R. A , , Daie, J. and Wyse, R. E. (1983). Sucrose uptake and compartmentation in sugar beet taproot tissue. Plant Physiology 72, 1-6. Saftner, R. A. and Martin, M. N. (1993). Transport of l-aminocyclopropane-lcarboxylic acid into isolated maize mesophyll vacuoles. Physiologia Plantarum 87, 535-543. Sakano, K. and Tazawa, M. (1984). Intracellular distribution of free amino acids between the vacuolar and extravacuolar compartments in internodal cells of Chara australis. Plant Cell Physiology 25, 1477-1486. Sakano, K. and Tazawa, M. (1985). Metabolic conversion of amino acids loaded in the vacuole of Chara australis internodal cells. Plant Physiology 78, 673677. Sato, T., Oshumi, Y. and Anraku, Y. (1985a). Substrate specificities of active transport systems for amino acids in vacuolar-membrane vesicles of Saccaromyces cerevisiae. Journal of Biological Chemistry 259, 11 505-1 1 508. Sato, T., Oshumi, Y. and Anraku, Y. (1985b). An arginine/histidine exchange transport system in vacuolar membrane vesicles of Saccaromyces cerevisiae. Journal of Biological Chemistry 259, 11 509-1 1 5 11. Saunders, J. A. and Conn, E. E. (1978). Presence of the cyanogenic glucoside dhurrin in isolated vacuoles from Sorghum. Plant Physiology 61, 154-157. Schnabl, H. and Kottmeier, C. (1984). Determinations of malate levels during the swelling of vacuoles isolated from guard-cell protoplasts. Planta 161, 27-31. Schomburg, M. (1994). Untersuchung iiber das thermotrope Phasenverhalten des Tonoplasten bei der CAM-Pflanze Kalanchoe daigremontiana. Ph.D. Thesis, Technische Hochschule Darmstadt. Schomburg, M. and Kluge, M. (1994). Phenotypic adaptation to elevated temperatures of tonoplast fluidity in the CAM plant Kalanchoe daigremontiana is caused by membrane proteins. Botanica Acta 107, 328-332. Sharma, V. and Strack, D. (1985). Vacuolar localization of 1-sinapoy1glucose:Lmalate sinapoyltransferase in protoplasts from cotyledons of Raphanus sativus. Planta 163, 563-568. Smith, F. A. and Raven, J. A. (1979). Intracellular pH and its regulation. Annual Review of Plant Physiology 30, 289-311. Smith, J. A. C., Griffith, H., Liittge, U . , Crook, C. E., Griffiths, N. M. and Stimmel, K.-H. (1986). Comparative ecophysiology of CAM and C , bromeliads. IV. Plant water relations. Plant Cell and Environment 9, 395410. Tarczynski, M. C., Jensen, R. G. and Bohnert, H. J. (1993). Stress protection of transgenic tobacco by production of the osmolyte mannitol. Science 259, 508-510.

Thom, M . and Komor, E. (1984). H+-sugar antiports the mechanism of sugar uptake by sugarcane vacuoles. FEBS Letters 173, 1 4 . Thom, M. and Maretzki, E. (1985). Group translocation as a mechanism for sucrose


399

transfer into vacuoles from sugarcane cells. Proceedings of the National Acadademy of Sciences of the USA 82, 4697-4701. Thorn, M., Leigh, R. and Maretzki, A . (1986). Evidence for the involvement of a UDP-glucose dependent group translocator in sucrose uptake into vacuoles of storage roots of red beet. Planta 167, 410-413. Thume, M. and Dietz, K. J. (1991). Reconstitution of the tonoplast amino-acid carrier into liposomes. Planta 185, 569-575. Ting, I. P. (1985). Crassulacean acid metabolism. Annual Review of Plant Physiology 99, 1118-1123. Tophof, S . , Martinoia, E . , Kaiser, G., Hartung, W. and Amrhein, N. (1989). Compartmentation and transport of 1-aminocyclopropane-I-carboxylic acid and N-malonyl-I-aminocyclopropane-1 -carboxylic acid in barley and wheat mesophyll cells and protoplasts. Physiologia Plantarum 75, 333-339. Vickery, H. B. (1952). The behaviour of isocitric acid in excised leaves of Bryophyllum calycinum during culture in alternating light and darkness. Plant Physiology 27, 9-17. Vierstra, R. D. (1993). Protein degradation in plants. Annual Review of Plant Physiology and Plant Molecular Biology 44, 385-410. Voss, M. and Weidner, M. (1988). Uridine 5’-diphospho-~-glucose-dependent vectorial sucrose synthesis in tonoplast vesicles of the storage hypocotyl of red beet (Beta vulgaris L. ssp. conditiva). Planta 173, 96-103. Wagner, G. J. (1979). Content and vacuole/extravacuole distribution of neutral sugars, free amino acids and anthocyanins. Plant Physiology 64, 88-93. Wagner, W., Keller, F. and Wiemken, A . (1983). Fructan metabolism in cereals: induction in leaves and compartmentation in protoplasts and vacuoles. Zeitschrift fur Pflanzenphysiologie 112, 359-372. Ward, J . M. and Schroeder, J . I. (1994). Calcium-activated K + channels and calcium-induced calcium release by slow vacuolar ion channels in guard cell vacuoles implicated in the control of stomata1 closure. Plant Cell 6,669-683. Werner, C. and Matile, P. (1985). Accumulation of coumarylglucosides in vacuoles of barley mesophyll protoplasts. Journal of Plant Physiology 118, 237-249. White, P. and Smith, J . A . C. (1989). Proton and anion transport at the tonoplast in crassulacean-acid-metabolism plants: specificity of the malate-influx system of Kalanchoe daigremontiana. Planta 179, 265-274. Willenbrink, J. and Doll, S. (1979). Characteristics of the sucrose uptake system of vacuoles from red beet tissue, kinetics and specificity of the sucrose uptake system. Planta 147, 15’3-162. Williamson, R . E. and Ashley, C. C. (1982). Free Ca2+ and cytoplasmic streaming in the alga Charu. Nature 296, 647-651. Winter, K. (1980). Dayhight changes in the sensitivity of phosphoenolpyruvate carboxylase to malate during crassulacean acid metabolism. Plant Physiology 65, 792-796. Winter, K. (1982). Properties of phosphoenolpyruvate carboxylase in rapidly prepared, desalted leaf extracts of the crassulacean acid metabolism plant Mesembryanthemurn crystallinum L. Planta 154, 298-308. Winter, K., Usuda, H., Tsuzuki, M . , Schmitt. M . , Edwards, G. E., Thomas, R. J. and Evert, R. F. (1982). Influence of nitrate and ammonia on photosynthetic characteristics and leaf anatomy of Moricandia arvensis. Planf Physiology 70, 616625. Wolf, J . (1939). Beitrage zur Kenntnis des Saurestoffwechsels sukkulenter Crassulaceen; IV. Beobachtungen zu Gehaltsschwankungen von Apfel- und Zitronensaure. Planta 29. 314-324.

400


Wolf, J . (1960). Der diurnale Saurerhythmus. In “Handbuch der Pflanzenphysiologie” (W. Ruhland, ed.), Vol. XII/2, pp. 809-889. Springer-Verlag, Berlin. Yamaki, S. (1982). Distribution of sorbitol, neutral sugars, free amino acids, malic acid and some hydrolytic enzymes in vacuoles of apple cotyledons. Plant Cell Physiology 23, 881-889. Yamaki, S. (1984). Isolation of vacuoles from immature apple fruit flesh and compartmentation of sugars, organic acids, phenolic compounds and amino acids. Plant Cell Physiology 25, 151-166. Yamaki, S. (1987). ATP-promoted sorbitol transport into vacuoles isolated from apple fruit. Plant Cell Physiology 28, 557-564. Yazaki, Y., Asukagawa, N., Ishikawa, Y., Ohta, E. and Sakata, M. (1988). Estimation of cytoplasmic free Mg2+ levels and phosphorylation potentials in mung bean root tips by in vivo 31P NMR spectroscopy. Plant Cell Physiology 29, 919-924. Zerez, C. R., Weiss, R. L., Franklin, C. and Bowman, B. J . (1986). The properties of arginine transport in vacuolar membrane vesicles of Neurospora crassa. Journal of Biological Chemistry 261, 8877-8882.

Secondary Inorganic Ion Transport at the Tonoplast

E. BLUMWALD and A . GELLI

Department of Botany, University of Toronto, 25 Willcocks Street, Toronto, Ontario M5S 3B2, Canuda

I.

Introduction .............................................................................

11.

Cations ....... ....................................................... A. Sodium ....................................................... C. D. E.

111.

IV.

40 1

402 402 404 Calcium ............................. ............ 406 407 Magnesium ....................................................................... Heavy Metals .. .................................... 408

A. B.

Anions ............. .................................................... Chloride .............................................. Nitrate ................................................................... C. Other Anions ......................... .........................

408 408 409 41 1

Conclusions ............................................................................ References .............................................................................

412 413

I. INTRODUCTION The plant vacuole has major roles in p H and ionic regulation of the cytosol, turgor regulation of the cell, the storage and retrieval of inorganic and organic nutrients, and detoxification of the cytosol, and it contains storage proteins and many lysosomal-type hydrolases (Matile, 1987). The main primary active transport process at the vacuolar membrane, the tonoplast, is the active accumulation of H+ by the action of two H+-translocating enzymes, the H+-ATPase and the pyrophosphatase (H+-PPase) (Rea and Sanders, 1987). The H+-pumps generate a proton-motive force, comprising gradients of both p H (ApH) and membrane potential ( A q ) across the tonoplast. The potential energy of the proton-motive force serves as the driving force for several Advances In Botanical Research Vol 25 incorporating Advances in Plant Pathology ISBN ~I-12-(IO5925-d

Copyright 0 l Y Y 7 Academic Press Limited All rights of reproduction in any form reserved

402


secondary transport processes. A secondary transport system involving a single ion can be driven by the membrane potential if the ion carries a net charge (uniport). A transport system will be driven by ApH if the flux of the ion is coupled to the flux of protons moving down their electrochemical gradient. In vivo, the “downhill” direction of protons is outward of the vacuole. Proton-coupled transport can occur in the same direction as the proton flux (Hf symport) or in the opposite direction (H+ antiport) (Blumwald, 1987). Studies characterizing the different transport mechanisms at the tonoplast, the methodology employed, the specificity and possible regulation of the transporters will be discussed. This review focuses on secondary ion transport mechanisms at the plant vacuole.

11. CATIONS A . SODIUM

Plant cells typically maintain a high K+/Na+ ratio in the cytoplasm. The transport of ions across the tonoplast is thought to play an important role in maintaining low cytosolic sodium concentrations. While salt-sensitive plants depend mainly on exclusion of sodium at the plasma membrane, salt-tolerant species accumulate large amounts of sodium in the vacuoles. Evidence consistent with the operation of a vacuolar Na+/H+ antiport has been obtained in different plant species using several techniques. Short-term in vivo 23Na nuclear magnetic resonance (NMR) spectroscopy studies of barley roots exposed to salt revealed evidence that supported the involvement of an Na+/H+ antiport in the transport of Na+ into the vacuole (Fan et a f . ,1989). Firstly, Na+ influx into root cells was consistently accompanied by vacuolar alkalinization and, secondly, changes in the accumulation of Na+ paralleled changes in the alkalinization of the vacuoles. Using the same procedure, Guern et a f . (1989) demonstrated the operation of an Na+/H+ antiport in intact vacuoles from Catharanthus roseus cells grown in a medium enriched in inorganic phosphate. The movement of H+ out of the vacuole was driven by an initial vacuolar gradient of 1.5 p H units, and Na+ accumulation into the vacuole was achieved against a four- to five-fold Na+ concentration gradient. This accumulation was both a saturable process with respect to external Na+ concentrations and a selective process, as K+ was less effective than Na+ in inducing the vacuolar alkalinization (Guern et al., 1989). Experiments based on the intravacuolar ApH-driven accumulation of fluorescent weak bases have also indicated the operation of an Na+/H+ antiport in vacuolar Na+ accumulation. In the salt-tolerant Beta vulgaris, an Na+/H+ exchange mechanism was demonstrated in intact vacuoles isolated from red beet storage roots (Niemietz and Willenbrink, 1985) and sugar beet cell suspensions (Blumwald and Poole, 1987). In both cases, the ATP-dependent acidification of the vacuoles,

SECONDARY INORGANIC ION TRANSPORT

403

measured by the fluorescence quenching of acridine orange, was reversed upon addition of external Na+. In similar studies, using isolated tonoplast vesicles, Na+/H+ antiport activity was reported in membranes isolated form storage tissue of red beet (Blumwald and Poole, 1985a) and sugar beet cell suspensions (Blumwald and Poole, 1987), from roots of the salt-tolerant Plantago maritima (Staal ef al., 1991), from salt-grown barley roots (Garbarino and Dupont, 1988) from tonoplast vesicles of the halophyte Atriplex gmelini (Matoh et al., 1989) and from sunflower roots (Ballesteros et al., 1997). In addition to Na+dependent H + efflux, a few studies have used H+-dependent 22Na+ influx measurements to demonstrate Na+/H+ antiport activity across the tonoplast (Barkla et al., 1990; Rea et a l . , 1990). 22Na+ uptake in tonoplast vesicles isolated from red beet storage tissue (Rea et ul., 1990) or sugar beet cell suspensions (Barkla et al., 1990) was dependent on the pH gradient across the tonoplast, and the influx was negligible in the absence of a pH gradient across the tonoplast vesicles. In some species the vacuolar Na+/H+ antiport appeared to be constitutive, while in others, such as Beta vulgaris, although constitutive the antiport was only activated by high NaCl concentrations. Increasing concentrations of NaCl in the growth medium of sugar beet cell suspensions did not change the K , but doubled the V,,, of the antiport (Blumwald and Poole, 1987). An increase in V,,, for the antiport with no change in apparent K , suggested the addition of more antiport molecules to the tonoplast in response to NaCl in the growth medium. A similar increase in tonoplast Na+/H+ antiport activity has been also reported after the exposure of Mesembryanthemum crysfallinurn to high NaCl concentrations (Barkla ef al., 1995). An inducible Na+/H+ antiport was also demonstrated in tonoplast from barley roots grown in the presence of NaCl (Garbarino and Dupont, 1988). The induction of the Na+/H+ antiport activity was very fast and appeared to be due to the activation of an existing protein rather than to de novo synthesis, since induction was observed in the presence of protein synthesis inhibitors (Garbarino and Dupont, 1989). Salt-treated plants lost their Na+/H+ antiport activity when transferred to a solution lacking Na+ with a similar time course to that seen during the induction by salt (Garbarino and Dupont, 1989). In Plantago species the vacuolar Na+/H+ antiport is only present in the salt-tolerant Plantago maritima, but not in the more salt-sensitive Plantago media (Staal ef al., 1991). The absence of Na+/H' antiport activity in the tonoplast of Plantago media may be related to a general property observed in salt-sensitive plants (Barkla et al., 1994). Moreover, in salt-tolerant species the accumulation of Na+ in the vacuoles via the vacuolar Na+/H+ antiport is accompanied by an outward rectification of the vacuolar cation channels that prevent a significant loss of the Na+ accumulated by the operation of the antiport (Pantoja et al., 1989; Maathuis and Prins, 1990). Blumwald and Poole (1985a) demonstrated that the Na+/H+ antiport of reed beet storage tissue was competitively inhibited by amiloride, a diuretic

404

E. BLUMWALD and A. GELLI

drug shown to be an inhibitor of Na+/H+ exchange systems in animals by binding at or near the Na+-binding site (Kinsella and Aronson, 1981). The demonstration that the 5-amino substituted amiloride analogues showed a greater inhibition of the vacuolar Na+/H+ antiport activity (Blumwald et a l . , 1987) allowed the use of high-specificity labelled amiloride analogues for the identification of the membrane proteins associated with the Na+/H+ antiport in plants. Barkla et a f . (1990) characterized the covalent binding of a radiolabelled amiloride analogue [3H]MIA (N5-methyl-I@-isobutylamiloride) to tonoplast proteins following photolysis. The correlation observed between the high-affinity MIA dissociation constant and the constants for inhibition of ApH-dependent Na+ uptake and Na+-dependent H + fluxes suggested that the high-affinity component represented a class of proteins that were likely associated with the vacuolar Na+/Hf antiport. Photolabelling of vacuolar membranes in the presence of MIA and amiloride suggested the association of three polypeptides of apparent molecular masses 170, 38 and 35 kDa with the vacuolar Na+/H+ antiport (Barkla et a f . , 1990). Antibodies raised against the 170 kDa polypeptide inhibited the Na+/H+ antiport in a competitive manner, suggesting that the 170 kDa polypeptide was associated with the antiport and that the antibody was binding to epitopes on the protein that were essential for its activity (Barkla and Blumwald, 1991). Higher amounts of this polypeptide were detected in tonoplast from sugar beets (Barkla and Blumwald, 1991) exposed to increasing NaCl concentrations and in microsomal membranes from leaves of salt-treated Mesembryanthemum crystallinum (Barkla et a f . , 1995). Further evidence for a role of the tonoplast Naf/Hf antiport-associated polypeptide in the salt tolerance of plants was suggested from the cross-reactivity of the antibody to membrane proteins from plant species differing in their ability to tolerate high levels of sodium (Barkla e t a f . , 1994). Antibodies raised against the 170 kDa polypeptide were used to screen tonoplast and microsomal membranes from salt-sensitive (pea and tomato) and salt-tolerant (sugar beets, barley and Mesembryanthemum crystallinum) plants. The antibodies only showed recognition of a 170 kDa polypeptide in membrane fractions of the salt-tolerant plants. These results were in agreement with those reported by Mennen et al. (1990), where the operation of a vacuolar Na+/H+ antiport was found not to be a ubiquitous feature of all higher plant cells. B . POTASSIUM

Potassium is a major nutrient for plants and is involved in a wide number of physiological processes that include the regulation of enzyme activity, charge balancing and osmotic regulation. Although the transport of other cations (e.g. Na+ and Ca2+) has been extensively investigated, information on the mechanisms for active proton transport at the plant vacuole is still scarce. Cytosolic K + concentrations have been reported in the range of


405

70-170 mM, while vacuolar K + can accumulate to concentrations of 200 mM (Luttge and Higinbotham, 1979). Given the cytosol-negative electrical potential difference at the tonoplast, an active K + translocation mechanism into the vacuole has to be considered. Evidence for the operation of a K+/H+ antiport was found in tonoplast-enriched fractions from zucchini (Scherer and Martiny-Baron, 1985) and Brassica nupus hypocotyls (Cooper et al., 1991). In zucchini a greater specificity of the antiport for K + was only observed at pH 8.0. In Brussica napus, the K+/H+ antiport was highly specific for K+ and was inhibited by high K + concentrations. The operation of a vacuolar transporter mediating the transport of K + from the vacuole to the cytosol has been proposed by Glass and co-workers. A substantial decrease in vacuolar Kf content was observed when barley roots were exposed to very low K + concentrations, although cytosolic K' concentrations remained constant (Memon et al., 1985). Based on energetic considerations, they proposed the operation of a K+/H+ symport mechanism (Fernando et al., 1992), similar to the system proposed for high-affinity K + absorption in Neurospora (Rodriguez-Navarro et ul., 1986). It should be noted that the calculations by Fernando et ul. (1992) were made assuming vacuolar membrane potentials of - 15 mV. At more negative vacuolar membrane potentials (< -30 mV) one would expect the passive movement of K+ from the vacuole into the cytosol through the vacuolar cation channel (Hedrich and Neher, 1987). An alternative pathway for active K + transport into the vacuole was suggested to be mediated by the vacuolar pyrophosphatase (Hf-PPase) (Davies et al., 1992). Using patch clamp techniques, the authors measured membrane currents in intact vacuoles from red beet (Beta vulgaris) storage tissue. A significant orthophosphate-dependent outward current, mediated by the enzyme in reverse mode, was evoked only when potassium was present at the vacuolar face of the tonoplast. These results led the authors to suggest that potassium was translocated by the H+-PPase. This hypothesis has been challenged by work carried out with purified and reconstituted PPase (Sato et al., 1994) and with vacuolar membrane vesicles (Ros er al., 1995). Sat0 and co-workers purified the vacuolar PPase from pumpkin, reconstituted the enzyme in proteoliposomes and showed that the transport of 42K+ transport into the liposomes was independent of the activation of H + translocation by the PPasc (Sato et al., 1994). Ros et al. (1995) used vacuolar membrane vesicles from Vitis vinifera suspension cells multilabelled with fluorescent probes for K + and H + to correlate proton pumping and K + transport into right-side-out, tightly sealed vesicles. Although both the H+-PPase hydrolytic activity and the pyrophosphate (PPJ-dependent H + transport into the vesicles was highly stimulated by the presence of K + , no incorporation of K + into the vesicles could be observed. Based on these observations, Ros et al. (1995) concluded that K + transport is not a general property of the vacuolar pyrophosphatase.

406

E. BLUMWALD and A. GELLI C. CALCIUM

Fluctuations in cytosolic free Ca2+ concentrations play an important role in the regulation of cell activities. In plants, the vacuole is the main storage compartment of intracellular Ca2+, and the release of vacuolar Ca2+ induces an elevation of free cytosolic Ca2+ concentration that plays an essential role in the stimulus/response coupling in signal transduction pathways in plant cells (Bush, 1993). Cytosolic Ca2+ is maintained at 200 nM or less (Trewavas and Gilroy, 1991) by the action of high-affinity membrane-bound transporters localized at the tonoplast and plasma membranes. The transport of Ca2+ into the vacuoles is achieved by a Ca2+/H+antiport driven by the electrochemical potential difference of H+ generated by the tonoplast H + pumps (Blumwald, 1987), and by the operation of an ATP-driven Ca2+-ATPase (Chanson, 1993). Ca2+/H+ antiport activities have been demonstrated indirectly using the tonoplast ATP or PPi-dependent proton pumps to generate an H+ electrochemical potential difference in tonoplast vesicles from oat roots (Schumaker and Sze, 1985), carrot cells (Bush and Sze, 1986), maize roots (Chanson, 1991) and vacuoles from Cutharunthus m e w cells (Guern et al., 1989). Direct studies using artificially imposed ApH gradients across tonoplast vesicles have shown the operation of a Ca2+/H+ antiport in red beet storage tissue (Blumwald and Poole, 1986; Blackford et uf., 1990) and oat roots (Schumaker and Sze, 1986). In all the in vitro studies the affinity of the antiport for Ca2+ was comparable, with apparent K , values ranging from 14 to 60pM. These K, values reported for the Ca2+/H+antiport are at least one order of magnitude higher than those reported for the Ca2+-ATPases (Evans et al., 1991). The apparent K , for external (cytosolic) Ca2+ was shown to be affected by the internal (vacuolar) pH in the range of pH 6.0-6.5, indicating a possible transmembrane effect of vacuolar proton binding on the affinity for cytosolic calcium (Blumwald and Poole, 1986). The effect of internal pH on the apparent K , for calcium was much steeper than a single-proton dissociation curve, and indicated a cooperative effect with a Hill coefficient of four, suggesting that four proton-dissociating groups were involved (Blumwald and Poole, 1986). The affinity of the Ca2+/H+ antiport for Ca2+ was also increased by calmodulin, resulting in a 100-fold decrease in the apparent K , for Ca2+ (Andreev et al., 1990). Based on the apparent electrogenicity of the Ca2+/H+ exchange, stoichiometries of 1H+/1Ca2+(Schumaker and Sze, 1986), 2H+/1Ca2+ (Blumwald and Poole, 1986) and 3H+/1Ca2+ (Blackford et al., 1990) have been reported. The number of H+ ions exchanged per Ca2+ ion transported by the antiport is relevant to whether or not the antiport, by running in reverse, could also serve for the influx of Ca2+ from the vacuole to the cytosol. Assuming that the accumulation of Ca2+ into the vacuole is solely due to the antiport activity, at least two H+ ions will be required to account for the Ca2+ gradients established across the vacuolar membrane. The stimulation of the


407

Ca2+/H+ antiport activity by positive vacuolar membrane potentials suggested a stoichiometry of 3H+/1Ca2+ (Blackford et a f . , 1990). With such a stoichiometry the antiport could not normally run in reverse to allow the influx of vacuolar Ca2+ into the cytosol. However, this assumption may not be valid because tonoplast-associated Ca2+-ATPases also contribute to the accumulation of Ca2+ into the vacuole (Bush, 1995). Schumaker and Sze (1990) reported the solubilization of the oat root vacuolar Ca'+/H+ antiport and its reconstitution in artificial liposomes. The antiport was solubilized with octylglucoside in the presence of phospholipids and glycerol, and the soluble proteins were reconstituted into liposomes by detergent dilution. The reconstituted antiport displayed similar properties to those found in the native tonoplast membranes (affinity of calcium, sensitivity to inhibitors, etc.), indicating that the Ca2+/Hf antiport was reconstituted in an active form (Schumaker and Sze, 1990). Recently, two Arabidopsis genes ( C A X I and CAX2) that suppress a mutant of Saccharornyces cerevisiae defective in vacuolar Ca2+ accumulation have been isolated (Hirschi et a f . , 1996). Both genes encode polypeptides possessing structural similarities to microbial Hf/Ca2+ antiports, and in vitro measurements demonstrated that expression of C A X l and CAX2 resulted in high- and low-efficiency ApH-dependent Ca2+ uptake (Hirschi et a f . , 1996).

D. MAGNESIUM

Magnesium ions play an important role in many cellular functions such as protein synthesis, glycolysis and oxidative phosphorylation, and act as a cofactor of numerous enzymic reactions. Mg2+ ions are abundant in the cytosol of plants, and its concentration can reach values of 1 mM or higher (Yazaki et a f . , 1988). An Mg2+/H+ antiport has been described in lutoids (a specialized vacuo-lysosomal compartment) from Hevea brusifiensis (Amalou et a f . , 1992). The exchange was electroneutral, supporting a stoichiometry of 1Mg2+/2H+, and it was inhibited by amiloride and imipramine. The affinity of the antiport for Mg2+ ( K , = 2.5 mM) was comparable to the affinity of the Mg2+/2Na+antiport reported in animal cells (Gunther et a f . , 1990). An Mg2+/H+ exchange was also shown in tonoplast vesicles isolated from maize roots (Pfeiffer and Hager, 1993). The divalent cation/H+ antiport activity was higher for Ca+ than for Mg+ into the vacuole, and the authors suggested that the antiport was physiologically involved in the transport of Mg2+ into the vacuole, but not Ca2+ (Pfeiffer and Hager, 1993). Amalou et a f . (1994) solubilized the Mg2+/Hf antiport from lutoid tonoplast and reconstituted the antiport in artificial liposomes. The proteoliposomes displayed Mg2+/H+ exchange, with a cation selectivity sequence of Zn'+ > Cd2+ > Mg2+, and the divalent cations Ca2+, Ba2+ and

408


Mn2+ were incapable of dissipating a preset pH gradient in the reconstituted proteoliposomes. Since the native membranes displayed both Ca2+/H+ and Mg2+/H+ exchange, the absence of Ca2+/H+ antiport activity in the proteoliposomes would suggest that the Ca2+/Hf antiport was eliminated during the solubilization/reconstitution step. Because of the different roles of Mg2+ and Ca2+ in cytosolic processes, these results would suggest that, similar to what has been observed in animal cells (Flatman, 1991), two different antiports mediate the transport of Ca2+ and Mg2+ into the lutoids (Amalou et uf., 1994). E. HEAVY METALS

Vacuoles have been suggested to be a major compartment for the accumulation of heavy metals (Krotz et al., 1989), and the active transport of these metals into the vacuole could play an important role in the detoxification of the cytosol. Although the transport of Mn2+, Cd2+ and Zn2+ into the vacuole was suggested in studies on the vacuolar Ca2+/H+ antiport from red beets (Blumwald and Poole, 1986) and the lutoid Mg2+/H+ antiport from Hevea brusifiensis (Amalou et al., 1994), only one study has directly demonstrated Cd2+ active transport in plant vacuoles (Salt and Wagner, 1993). The accumulation of Cd2+ into tonoplast vesicles isolated from oat roots was shown to be driven by a ApH generated by either the vacuolar H+-ATPase or by an artificial ApH generated using the ionophore nigericin to exchange K + for H+ (Salt and Wagner, 1993). The apparent K , (5.5 pM) for Cd2+/H+ antiport activity was comparable with the estimated range of physiological cytosolic Cd2+ concentrations in plants growing in Cd2+-contaminatedsoils (Krotz et al., 1989), suggesting that this antiport activity could play a role in the vacuolar accumulation of Cd2+ in oat roots (Salt and Wagner, 1993).

111. ANIONS A . CHLORIDE

The requirements of plants for C1- has been associated with the role this halogen plays in osmoregulation, charge balance in the protoplast and the stimulation of oxygen evolution during photosynthesis (Flowers, 1988). Some evidence indicates that the essential functions of C1- can be assumed by some other anions. For example, NO3- and Br- can substitute for CI- during the opening of stomata (Willmer, 1983), and during photosystem I1 activity NO3and Br- can also stimulate oxygen evolution (Critchley, 1985). However, in some cases these activities are only partially restored when C1- is substituted by a different anion, suggesting some more specific function for C1-


409

(Critchley, 1985; Flowers, 1988). In salt-tolerant plants CI- ions are compartmentalized in the vacuole in concentrations up to 300 mM (Matile, 1988) and play a significant role in osmoregulation. Fluctuations in cytoplasmic and vacuolar chloride concentrations have been shown to regulate the transport of other anions into the vacuole (Pantoja et af., 1992; Plant ef al., 1994). The dissipation of a vacuolar positive membrane potential (generated by the activation of the vacuolar H+-ATPase or the H+-PPase) by anions revealed the existence of a uniport that allows CI- to accumulate in the vacuole in response to the membrane potential generated by the H f pump (Bennett and Spanswick, 1983; Churchill and Sze, 1984; Blumwald and Poole, 1985b; Kastner and Sze, 1987; Pope and Leigh, 1987). Dissipation of the membrane potential by chloride was saturable with a K , of 2.3 mM (Kastner and Sze, 1987; Pope and Leigh, 1987). Chloride uptake was competitively inhibited by nitrate, suggesting that both nitrate and chloride may be transported by the same porter (Martinoia et uf., 1987). Tonoplast channels for passive ion movement of varying degrees of specificity are ubiquitous in plant cells, and several distinct types of channels involved in the transport of chloride have been identified (Hedrich and Neher, 1987; Pantoja et al., 1992; Tyerman, 1992). The operation of these channels could provide a uniport mechanism for the transport of chloride (and other anions) into the vacuole. Since the operation of the channel could also dissipate anion gradients established by active transport, one would expect the channel activity to be tightly controlled by factors such as membrane potential, Ca2+ and other ions. For example, intravacuolar chloride concentrations were shown to regulate the vacuolar anion channel activity; high vacuolar chloride concentrations favoured the transport of nitrate and phosphate into the vacuole (Plant et al., 1994), and the influx of anions into the vacuole was coupled to chloride efflux into the cytosol (P. J. Plant and E. Blumwald, unpublished results).

B. NITRATE

Nitrate, a major plant nutrient, is accumulated and stored in the vacuole (Martinoia et af., 1981), from where it can be retrieved under appropriate conditions, according to various metabolic demands (Granstedt and Huffaker, 1982). The transport of nitrate across the tonoplast can be driven by the cytosol-negative electrical membrane potential difference. Studies using isolated tonoplast vesicles from oats (Churchill and Sze, 1984) and red beet storage tissue (Blumwald and Poole, 1985b) showed nitrate transport driven by cytosol-negative membrane potentials generated by the activation of the tonoplast Hf-ATPase. Similar results were obtained with oat tonoplast vesicles when t h e cytosol-negative membrane potentials were generated

410


through the activation of the H+-PPase (Kastner and Sze, 1987; Pope and Leigh, 1987). Using 36C103-as an radioactive nitrate analogue, Chodera and Briskin (1990) showed that C103- uptake was driven by a cytosol-negative positive membrane potential, but not by an cytosol-alkaline pH gradient. The possibility that nitrate transport into the vacuole is mediated by an H+-coupled mechanism has been suggested (Blumwald and Poole, 1985b; Schumaker and Sze, 1987). Blumwald and Poole (1985b) reported that low nitrate concentrations induced an increase in vacuolar ATPase activity (an effect that was abolished by gramicidin), indicating a dissipation of the pH gradient and the electrical potential difference at vesicle transport sites distinct from the tonoplast H+-ATPase. Based on these results they proposed a model in which NO3- enters the vacuole by a uniport mechanism where it can be retrieved from the vacuole by the operation of an N03-/H+ symport. Schumaker and Sze (1987) observed that N03- dissipated an artificially imposed pH gradient in tonoplast vesicles. The dissipation of the pH gradient by NO3- was concentration dependent and was inhibited by the anion transport blocker 4,4’-siisothiocyano-2,2’-stilbenedisulfonate (DIDS). Based on these results they proposed the operation of an N03-/H+ antiport mechanism. The operation of active vacuolar nitrate transport has been challenged by results based on compartmental analysis of unidirectional nitrate fluxes in intact tissues. Devienne et al. (1994) examined long-term effects of NO3- concentrations on NO3- unidirectional fluxes and NO3distribution in intact wheat roots. Using compartmental analysis on 15N03fluxes they concluded that N03- concentrations derived from their analysis were close to those expected as a result of passive diffusion across the tonoplast, suggesting the absence of active vacuolar nitrate transport. Similar results were obtained in root segments of Allium cepa (Macklon et al., 1990). However, evidence in support of active vacuolar nitrate transport has been provided with ion selective electrodes used to measure proton gradients, electrical potential differences and nitrate gradients across vacuoles from barley root cells (Zhen e t a l . , 1991; Miller and Smith, 1992). Vacuolar nitrate activities were 8-20 times higher than the cytosolic nitrate activities, and the measured tonoplast electrical potential differences were no more negative than - 12 mV. These values would indicate that the observed vacuolar nitrate accumulation could not be accounted for by a passive mechanism driven by the cytosolic-negative vacuolar membrane potential. Calculations based on cytoplasmic and vacuolar pH values, tonoplast electrical potential differences and nitrate concentrations indicated that there was sufficient free energy in the p H gradient to drive nitrate accumulation into the vacuole. Based on these observations the authors suggested the operation of an N03-/H+ antiport with a 1:l stoichiometry (Miller and Smith, 1992). It is likely that the influx and efflux of nitrate from the vacuole are under regulatory control in the intact cell and that the activities of the uniport and H+-coupled nitrate cotransporters are regulated by metabolism.


411

For example, Aslam et a f . (1976) have given evidence that transfer of nitrate from a storage pool to a metabolic pool in etiolated barley leaves was regulated by light. Marigo et a f . (1985) indicated that the retrieval of nitrate may be coupled with nitrate reduction decreasing the cytosolic nitrate pool, thereby increasing the nitrate gradient across the tonoplast and favouring nitrate export. Recently, Plant et al. (1994) showed the regulation of nitrate transport into the vacuole by intravacuolar chloride. Vacuolar concentrations higher than 50 mM increased the inward currents of nitrate (and phosphate), and the authors suggested a mechanism by which chloride induced conformational changes in the anion channel protein favouring the uptake of nitrate into the vacuole (Plant etaf., 1994). All the evidence mentioned above would indicate that influx and efflux of nitrate from the vacuole are under metabolic control and that it is plausible that both uniport and H+-coupled transport are invoked at different developmental stages and that these transport mechanisms could be regulated by factors such as growth, sinkhource transitions and environmental factors.

C. OTHER ANIONS

Transport mechanisms allowing the influx and efflux of phosphate and sulfate in vacuoles have not been investigated in great detail. Active transport of sulfate into the vacuole was suggested by Kaiser et a f . (1989). Barley seedlings grown hydroponically in the presence of high concentrations of potassium sulfate showed accumulation of sulfate in the vacuoles of their mesophyll cells. The vacuolar sulfate accumulation was ATP-dependent and was inhibited by p-chloromercuribenzene sulfonate (pCMBS), an agent known to inhibit vectorial proton translocation into the vacuoles (Kaiser et a f . , 1989). Recently, three sulfate transporter cDNAs were isolated by complementation of a yeast mutant with a cDNA library derived from the tropical forage legume Styfosanrhes harnata (Smith et a f . , 1995). Two of these cDNAs, SHSTI and SHSTZ, encode high-affinity H+-sulfate cotransporters that mediate the uptake of sulfate by plant roots from low concentrations of sulfate in the soil solution. The third, SHST3, represents a different subtype encoding a lower-affinity H+-sulfate transporter which may be involved in the internal transport of sulfate between cellular or subcellular compartments within the plant (Smith et a f . , 1995). The accumulation of phosphate into vacuoles was characterized using 3'P NMR techniques (Rebeille et al., 1983; Mathieu et al., 1989). In Acer pseudoplatanus cells, phosphate starvation led to phosphate efflux from the vacuole, while vacuolar phosphate accumulation was observed in cells grown in excess of phosphate in the growth media (Rebeille et a f . , 1983). In Catharanthus roseus cells, the vacuoles accumulated phosphate to concentrations of up to 160mM (Mathieu er a f . , 1989). Although the uptake of

412


phosphate into isolated vacuoles was shown to be stimulated by ATP (Mimura et a f . , 1990), no experimental data supporting an active phosphate transport mechanism in plant vacuoles are available yet. Moreover, patch clamp experiments have shown that phosphate ions can passively permeate the tonoplast through the anion-selective channel (Plant et al., 1994).

IV. CONCLUSIONS Further progress in the understanding of vacuolar ion transport systems require the molecular identification and characterization of the transport proteins. The biochemical identification of these proteins is extremely difficult because cotransporters do not possess an enzymatic mechanism associated with their activity, and therefore the detection of the transporter activity during purification can only be achieved by reconstitution in artificial liposomes. These transporters are not abundant; for example, photolabelling of tonoplast proteins with a [3H]amiloride analogue estimated the number of Na+/H+ antiport units to be 10000 per vacuole (Barkla et al., 1990). Hedrich et af. (1986) estimated the presence of 1000 channels per vacuole. Because of their low abundance, the proteins have to be isolated from a large amount of tissue and concentrated through methods that could result in the inactivation of the protein in question. Thus, although reconstitution is possible, the reconstituted proteoliposome may contain the protein in an inactivated state, making its detection impossible. Molecular biological approaches have been successfully used to identify and clone several genes encoding for secondary transport systems. The selection of Arabidopsis mutants that were able to grow in the presence of chlorate allowed the identification of C H L l , encoding for an electrogenic nitrate transporter (Tsay et af., 1993). Yeast mutants have proven to be an important tool for the identification of genes from homologous and heterologous sources by complementation. Schachtman and Schroeder (1994) reported the identification and characterization of a cDNA for HKT1, a high-affinity K+ transporter. Expression of H K T l in Xenopus oocytes suggested that the HKTl K+ transporter uses a K+-H+ co-uptake mechanism similar to that reported for Neurospora (Rodriguez-Navarro et a f . , 1986). Further work by the same group (Rubio et a f . , 1995) demonstrated that HKTl mediated high-affinity K+-Na+ cotransport in oocytes and yeast. HKTl mutants conferred increased Na+ tolerance to yeast (lower inhibition of K+ uptake at high Na+ concentrations as well as reduced Na+ uptake), and the possible role of HKTl in Na+ uptake by plants was suggested (Rubio et af.,1995). Other examples of genes isolated by yeast complementation are a high-affinity NH4+ transporter (Ninnemann et al., 1994), a family of sulfate transporters (Smith et af., 1995) and high and low Hf/Ca2+ affinity antiports (Hirschi et a f . , 1996). Although the use of mutants has been an important


413

tool in the identification of plant genes coding for inorganic ion transporters, only a few of these genes encode for endomembranal transporters: SHST3 encoding for a sulfate transporter (Smith et al., 1995) and C A X l and CAX2 encoding for high- and low-affinity H+/Ca2+antiporters (Hirschi et a l . , 1996). The molecular identification of vacuolar cotransporters ultimately depends on the identification of appropriate phenotypes, which may be difficult for intracellular membrane transport mutants.

REFERENCES Amalou, Z . , Gibrat, R., Brugidou, C . , Trouslot, P. and d’Auzac, J . (1992). Evidence of an amiloride-inhibited Mg2+/2H+antiporter in lutoid (vacuolar) vesicles from latex o f Hevea brasiliensis. Plant Physiology 100, 255-260. Amalou, Z . . Gibrat, R., Trouslot, P. and d’Auzac, J. (1994). Solubilization and reconstitution of the Mg2+/2H+ antiporter of the lutoid tonoplast form Hevea brasiliensis latex. Plant Physiology 106, 79-85. Andreev, I . M., Koren’kov, V. and Molotokovsky, Y. G. (1990). Calmodulin stimulation of a Ca2+/nHt antiport across the vacuolar membrane of sugar beet taproot. Journal of Plant Physiology 136, .%7. Aslam, M., Oaks, A. and Huffaker, R. C. (1976). Effect of light and glucose on the induction of nitrate reductase and on the distribution of nitrate in etiolated barley leaves. Plant Physiology 58, 588-591. Ballesteros, E., Blumwald, E . , Donaire, J . P. and Belver, A. (1997). Na+/H+ antiport activity in tonoplast vesicles isolated from Sunflower roots induced by NaCl stress. Physiologia Plantarum 99: 328-334. Barkla, B. J . and Blumwald, E. (1991). Identification of a 170-kDa protein associated with the vacuolar Na+/H+ antiport of Beta vulgaris. Proceedings of the National Academy of Sciences of the U S A 88, 11 177-11 181. Barkla, B. J., Charuk, J. H. M., Cragoe, E . J.. Jr and Blumwald, E. (1990). Photolabeling of tonoplast from sugar beet cell suspensions by [3H]5-(N-methylN-isobuty1)amiloride. and inhibitor of the vacuolar Na+/H+ antiport. Plant Physiology 93, 924-930. Barkla, B. J., Apse, M. P., Manolson, M . F. and Blumwald, E. (1994). The plant vacuolar Na+/H+ antiport. In “Membrane Transport in Plants and Fungi: Molecular Mechanisms and Control” (M. R. Blatt, R. A. Leigh and D. Sanders, eds), pp. 141-153. The Company of Biologists, Cambridge. Barkla, B. J . , Zingarelli, L., Blumwald, E. and Smith, J . A. C. (1995). Tonoplast Na+/H+ antiport activity and its energization by the vacuolar H+-ATPase in the halophytic plant Mesembryanthemum crystallinurn L. Plant Physiology 109, 549-556. Bennett, A. B. and Spanswick, R. M. (1983). Optical measurements of ApH and AV in corn root membrane vesicles: kinetic analysis of CI- effects on a protontranslocating ATPase. Journal of Membrane Biology 71, 95-107. Blackford, S . , Rea, P. A. and Sanders, D. (1990). Voltage sensitivity of H+/Ca*+ antiport in higher plant tonoplast suggests a role in vacuolar calcium accumulation. Journal of Biological Chemistry 265, 9617-9620. Blumwald, E. (1987). Tonoplast vesicles as a tool in the study of ion transport at the plant vacuole. Physiologia Plantarurn 69, 731-734. Blumwald, E. and Poole, R. J . (1985a). Na+/H+ antiport in isolated tonoplast vesicles from storage tissue of Beta vulgaris. Plant Physiology 78, 163-167.

414

E. BLUMWALD and A. GELLI

Blumwald, E. and Poole, R. J . (1985b). Nitrate storage and retrieval in Beta vulgaris: effects of nitrate and chloride on proton gradients in tonoplast vesicles. Proceedings of the National Academy of Sciences o the USA 82, 3683-3687. Blumwald, E. and Poole, R, J. (1986). Kinetics of Ca /H+ antiport in isolated tonoplast vesicles from storage tissue of Beta vulgaris L. Plant Physiology 80, 727-73 1. Blumwald, E. and Poole, R. J. (1987). Salt tolerance in suspension cultures of sugar beet: induction of Na+/H+ antiport activity at the tonoplast by growth in salt. Plant Physiology 83, 884-887. Blumwald, E., Cragoe, E. J . , Jr and Poole, R. J . (1987). Inhibition of Na+/H+ antiport activity in sugar beet tonoplast by analogs of amiloride. Plant Physiology 85, 30-33. Bush, D. S. (1993). Regulation of cytosolic calcium in plants. Plant Physiology 103, 7-13. Bush, D. S. (1995). Ca2+ regulation in plant cells and its role in signalling. Annual Review of Plant Physiology and Plant Molecular Biology 46, 95-122. Bush, D. R. and Sze, H. (1986). Ca2+ transport in tonoplast and endoplasmic reticulum vesicles isolated from cultured carrot cells. Plant Physiology 80, 549-555. Chanson, A. (1991). A Ca2+/H+ antiport system driven by the tonoplast pyrophosphate-dependent proton pump from maize roots. Journal of Plant Physiology 137, 471476. Chanson, A. (1993). Active transport of proton and calcium in higher plant cells. Plant Physiological Biochemistry 31, 943-955. Chodera, A. J. and Briskin, D. P. (1990). Chlorate transport in isolated tonoplast vesicles from red beet (Beta vulgaris L . ) storage tissue. Plant Science 67, 151-160. Churchill, K. A. and Sze, H. (1984). Anion-sensitive H + pumping ATPase of oat roots: direct effects of C1-, NO3- and disulfonic stilbene. Plant Physiology 76, 490-497. Cooper, S . , Lerner, H. R. and Reinhold, L. (1991). Evidence for a highly specific K+/H+ antiporter in membrane vesicles from oil-seed rape hypocotyls. Plant Physiology 97, 12 112-12 120. Critchley, C. (1985). The role of chloride in Photosystem 11. Biochimica et Biophysica Acta 811, 33-46. Davies, J . M., Poole, R. J . , Rea, P. A. and Sanders, D. (1992). Potassium transport into plant vacuoles energized directly by a proton-pumping inorganic pyrophosphatase. Proceedings of the National Academy of Sciences of the USA 89, 11 701-11 705. Devienne, F., Mary, B. and Lamaze, T. (1994). Nitrate transport in intact wheat roots. 11. Long-term effects of NO3- concentration in the nutrient solution on N 0 3 - unidirectional fluxes and distribution within the tissues. Journal of Experimental Botany 45, 677-684. Evans, D. E., Briars, S. A . and Williams, L. E. (1991). Active calcium transport by plant cell membranes. Journal of Experimental Botany 236, 285-303. Fan, T. W. M., Higashi, R. M., Norlyn, J. and Epstein, E. (1989). In vivo 23Na and 3'P-NMR measurement of a tonoplast Na+/H+ exchange process and its characteristics in two barley cultivars. Proceedings of the National Academy of Sciences of the USA 86, 231-236. Fernando, M., Mehroke, J. and Glass, A. D. M. (1992). De novo synthesis of plasma membrane and tonoplast polypeptides of barley roots during short-term K+ deprivation. Plant Physiology 100, 1269-1276.

L


415

Flatman, P. W. (1991). Mechanisms of magnesium transport. Annual Review of Physiology 53, 259-271. Flowers, T. J. (1988). Chloride as a nutrient and as an osmoticum. In “Advances in Plant Nutrition” (P. B. Tinker and A. Lauchli, eds), Vol. 3, pp. 55-78. Praeger, New York. Garbarino, J. and Dupont, F. M. (1988). NaCl induces a Na+/H+ antiport in tonoplast vesicles from barley roots. Plant Physiology 86, 231-236. Garbarino, J . and Dupont, F. M. (1989). Rapid induction of Na+/H+ exchange activity in barley root tonoplast. Plant Physiology 89, 1-4. Granstedt, R. C. and Huffaker, R. C. (1982). Identification of the leaf vacuole as a major nitrate storage pool. Plant Physiology 70, 41W13. Guern, J . , Mathieu, Y . , Kurkdjian, A . , Manigault, P., Manigault, J . , Gillet, B., Beloeil, J . C. and Lallemand, J. Y. (1989). Regulation of vacuolar pH of plant cells. 11. A 3’P-NMR study of the modifications of vacuolar pH in isolated vacuoles induced by proton pumping and cation/H+ exchanges. Plant Physiology 89, 27-36. Gunther, T., Vorman, J. and Hollriegl, V. (1990). Characterization of Na+-dependent Mg2+ efflux from Mg2+-loaded rat erythrocytes. Biochimica et Biophysica Acta 898, 455-461. Hedrich, R. and Neher, E. (1987). Cytoplasmic calcium regulates voltage dependent ion channels in plant vacuoles. Nature 329, 833-836. Hedrich, R., Flugge, U . I. and Fernandez, J . M . (1986). Patch-clamp studies of ion transport in isolated plant vacuoles. FEBS Letters 204, 228-232. Hirschi, K. D., Zhen, R. G., Rea, P. A. and Fink, G. R. (1996). CAXl, and H+/Ca2+ antiporter from Arabidopsu, Proceedings of the National Academy of Sciences of the USA 93, 8782-8786. Kaiser, G., Martinoia, E., Schroppel-Meier, G. and Heber, U . (1989). Active transport of sulfate into the vacuole of plant cells provides halotolerance and can detoxify SO2. Journal of Plant Physiology 133, 756-763. Kastner, K. H. and Sze, H. (1987). Potential-dependent anion transport in tonoplast vesicles from oat roots. Plarzt Physiology 83, 483489. Kinsella, J . L. and Aronson, P. S. (1981). Amiloride inhibition of the Na+/H+ exchanger in renal microvillus membrane vesicles. American Journal of Physiology 241, 374-379. Krotz, R. M., Evangelou, B . P. and Wagner, G. J . (1989). Relationships between cadmium, Zinc, Cd-peptide, and organic acid in tobacco suspension cells. Plant Physiology 91, 180-187. Luttge, U . and Higinbotham, N. (1979). “Transport in Plants.” Springer-Verlag, New York. Maathuis, F. J . M. and Prins, H. B. A. (1990). Patch-clamp studies on root cell vacuoles of a salt tolerant and a salt sensitive Plantago species. Plant Physiology 92, 23-28. Macklon, A. E. S . , Ron, M. M. and Sim, A . (1990). Cortical cell fluxes of ammonium and nitrate in excised root segments of Allium cepa L.: studies using “N. Journal of Experimental Botany 41, 359-370. Marigo, G., Bouyssou, H. and Belkoura, M. (1985). Vacuolar efflux of malate and its influence on nitrate accumulation in Catharanthus roseus cells. Plant Science 39, 97-103. Martinoia, E., Heck, U . and Wiemken, A. (1981). Vacuoles as a storage compartment for nitrate in barley leaves. Nature 289, 292-293. Martinoia, E., Schramm, M. J . , Flugge, U . I. and Kaiser, G . (1987). Intracellular distribution of organic and inorganic anions in mesophyll cells: transport

416


mechanisms in the tonoplast. In “Plant Vacuoles” (B. Marin, ed.), NATO AS1 Series, pp. 407-416. Plenum Press, New York. Mathieu, Y., Guern, J., Kurkdjian, A , , Manigault, P., Manigault, J., Zielinska, T., Gillet, B., Belowil, J. C. and Lallemand, J . Y. (1989). Regulation of vacuolar pH of plant cells. I. Isolation and properties of vacuoles suitable for 31P-NMR studies. Plant Physiology 89, 19-26. Matile, P. (1987). The sap of plant cells. New Phytologist 105, 1-26. Matile, P. (1988). Biochemistry and function of vacuoles. Annual Review of Plant Physiology 29, 193-213. Matoh, T., Ishikawa, T. and Takahashi, E. (1989). Collapse of ATP-induced pH gradient by sodium ions in microsomal membrane vesicles prepared from Atriplex gmelini leaves. Plant Physiology 89, 180-183. Memon, A. R., Saccomani, M. and Glass, A. D. M. (1985). Efficiency of K+ utilization by barley varieties: the role of subcellular compartmentation. Journal of Experimental Botany 36, 1860-1876. Mennen, H., Jacoby, B. and Marschner, H. (1990). Is sodium proton antiport ubiquitous in plant cells? Journal of Plant Physiology 137, 180-183. Miller, A. J . and Smith, S. J. (1992). The mechanism of nitrate transport across the tonoplast of barley root cells. Planta 187, 554-557. Mimura, T., Dietz, K. J . , Kaiser, W., Schramm, M. J. and Heber, U. (1990). Phosphate transport across biomembranes and cytosolic phosphate homeostasis in barley leaves. Planta 180, 139-146. Niemietz, C. and Willenbrink, J. (1985). The function of tonoplast ATPase in intact vacuoles of red beet is governed by direct and indirect ion effects. Planta 166, 545-549. Ninnemann, O., Jauniax, J. C. and Frommer, W. B. (1994). Identification of a high affinity NH4+ transporter from plants. EMBO Journal 13, 3464-3471. Pantoja, O., Dainty, J. and Blumwald, E. (1989). Ion channels in vacuoles from halophytes and glycophytes. FEBS Letters 255, 92-96. Pantoja, O., Dainty, J. and Blumwald, E. (1992). Cytoplasmic chloride regulates cation channels in the vacuolar membrane of plant cells. Journal of Membrane Biology 125, 219-229. Pfeiffer, W. and Hager, A. (1993). A Ca2+-ATPAse and a Mg2+/Hf-antiporter are present on tonoplast membranes from roots of Zea mays L. Planta 191, 377-385. Plant, P. J., Gelli, A. and Blumwald, E. (1994). Vacuolar chloride regulation of anion-selective tonoplast channels. Journal of Membrane Biology 140, 1-12. Pope, A. J. and Leigh, R. A. (1987). Some characteristics of anion transport at the tonoplast of oat roots, determined from the effects of anions on pyrophosphatedependent proton transport. Planta 172, 91-100. Rea, P. A. and Sanders, D. (1987). Tonoplast energyzation: two H + pumps, one membrane. Physiologia Plantarum 71, 131-141. Rea, P. A., Sanders, D. and Leigh, R. A. (1990). Wash-activated Na+/H+exchange in tonoplast vesicles isolated from Beta storage root disks. In “Proceedings of the 7th International Workshop on Plant Membrane Transport” (M. S. Bailey, N. A. Walker and J. R. Smith, eds), pp. 277-291. University of Sydney, Sydney. Rebeille, F., Bligny, R., Martin, J. P. and Douce, R. (1983). Relationship between the cytoplasm and the vacuole phosphate pool in Acer pseudoplatanus cells. Archives of Biochemistry and Biophysics 225, 143-148. Rodriguez-Navarro, A., Blatt, M. R. and Slayman, C. L. (1986). A potassiumproton symport in Neurospora crassa. Journal of General Physiology 87, 649-674.

SECONDARY I NORG AN 1C ION TRANSPORT

417

Ros, R . , Romieu, C., Gibrat, R. and Grignon. C. (1995). The plant inorganic pyrophosphatase does not transport K+ in vacuole membrane vesicles multilabeled with fluorescent probes for H + , K+ and membrane potential. Journal of Biologicul Chemistry 270, 43684374. Rubio. F., Gassrnann, W. and Schroeder, J . 1. (1995). Sodium-driven potassium uptake by the plant potassium transporter H K T l and mutations conferring salt tolerance. Science 270, 1660-1663. Salt, D. E . and Wagner, G . J . (1993). Cadmium transport across tonoplast of vesicles from oat roots. Journal of Biologicul Chemistry 268, 12 297-12 302. Sato, M. H., Kasahara. M., Ishii, N., Homareda, H . , Matsui, H. and Yoshida, M. (1994). Purified vacuolar inorganic pyrophosphatase consisting of a 7.5 kDa polypeptide can pump H + into reconstituted proteoliposomes. Journa/ of Biologicul Chemistry 269, 67254728. Schachtman, D. P. and Schroeder, J . I. (1994). Structure and transport mechanism of a high-affinity potassium uptake transporter from higher plants. Nature 370, 655-658. Scherer, G . F. E. and Martiny-Baron, G. (1985). K+/H+ exchange transport in plant membrane vesicles is evidence for K + transport. Plant Science 41, 161-168. Schumaker, K . S. and Sze, H . (1985). A Ca’+/Hf antiport system driven by the proton electrochemical gradient of a tonoplast H+-ATPase from oat roots. Plant Physiology 79, 1 1 11-1 I 17. Schumaker, K. S. and Sze, H. (1986). Calcium transport into the vacuole of oat roots. Characterization of a Ca’+/H+ exchange activity. Journal of Biologicul Chemistry 161, 12 172-12 178. Schumaker, K. S. and Sze, H . (1987). Decrease of pH gradients in tonoplast vesicles by NO3- and C1-: evidence for H’-coupled anion transport. Plant Physiology 83, 490-496. Schumaker, K. S. and Sze, H. (1990). Solubilization and reconstitution of the oat root vacuolar H+/Ca*+ exchanger. Plant Physiology 92, 34(&345. Smith, F. W., Eating, P. M., Hawkesford, M. J. and Clarkson, D. T. (1995). Plant members of a family of sulfate transporters reveal functional subtypes. Proceedings of the National Academy of Sciences of the USA 92, 9373-9377. Staal, M.. Maathuis, F. J. M.. Elzenga, T. M.. Overbeek, J . H . M. and Prins, H. B. A . (1991). Na+/H+ antiport activity in tonoplast vesicles from roots o f the salt-tolerant Plantago maritima and the salt-sensitive Plantago media. Physiologia Plantarum 82, 179-184. Trewavas, A . and Gilroy, S. (1991). Signal transduction in plant cells. Trends in Genetics 7, 356361. Tsay, Y . , Schroeder, J. L., Feldmann. K. A . and Crawford, N. M . (1993). The herbicide sensitive gene C HLl of Arahidopsis enodes a nitrate-inducible nitrate transporter. Cell 72, 705-713. Tyerman, S. D. (1992). Anion channels in plants. Annual Review of Plant Physiology and Plant Molecular Biology 43, 351-373. Willmer, C. M. (1983). “Stomata.” Longman, London. Yazaki, Y . , Asukagawa, N . , Ishikaway. Y . , Eiji, 0. and Sakata, M . (1988). Estimation of cytosolic free Mg2+ levels and phosphorylation potentials in mung bean root tips by in vivo ”P-NMR spectroscopy. Plant Cell Physiology 29, 9 19-924. Zhen, R. G . , Koyro, H. W., Leigh, R. A.. Tornos, A . D. and Miller, A . J. (1991). Compartmental nitrate concentrations in barley root cells measured with nitrate-selective microelectrodes and by single-cell sap sampling. Planta 185, 356361.


Aquaporins and Water Transport Across the Tonoplast

M. J . CHRISPEELS, M. J . DANIELS and A. WEIG

Department qf Biology, 01 16, Uniwrsity of California, San Diego, 9500 Gilrnun Drive, La Jolla, C A 92093-0116 USA

I. 11.

Introduction

..........................................................................

How are Water Channel Proteins Assaycd? ................................

41Y 420

111.

The Discovery o l Aquaporins in I'lants

......................................

422

I v.

Aquaporin TIP and Aquaporin PIP arc Members of a Large Gene Family ..................................................................................

423

...........................................

423

VI. The Activity of a Seed-Specific Tonoplast Aquaporin is Regulated by Phosphorylation ...................................................

426

Developmental Kegulation of Tonoplast Aquaporins ....................

427

.............................................

428

V.

VII.

VIII.

The Structure o f the Aqueous Pore

Are Aquaporins Active in Plants'?

;I Role in Water Transport in the Plant'? ....... 428 ........................... Acknowledgements ............ References .......................................................................

IX. Do Aquaporins Play

I. INTRODUCTION T h e membranes o f living cells a r e differentially permeable to water and solutes: water passes more easily through a membrane than solutes, whose passage depends o n the presence of specific channels and carriers. But how Copyriplit 0 1YY7 Acadcnitc Press 1-imited All right5 ot ieproduction in a n y h r n i reserved

420

M J CHKISPEELS et

ti/

does water move through the membrane? It was assumed for a long time that osmotic water movement occurs strictly by diffusion of water molecules across the lipid bilayer. When it was noticed that certain membranes permitted a water flux that was greater than expected, the existence of water-filled pores was proposed as an explanation. For many membranes, t h e osmotic water permeability coefficient (Pf) is greater than the diffusional water permeability coefficient (Pd),suggesting the presence of pores o r water channels that ;illow for the mass movement of water. For example, results from measurements of water flux through the giant single-celled marine alga Vdonia indicate that water transport may be by diffusion alone (Gutknccht, 1967), but in Nifellcr, a freshwater alga with giant cells, similar measurements suggest the presence of pores estimated to be 3 A in diameter (Fensom and Wanless, 1967). By comparison, water molecules have an apparent diameter of 1.sA. The nature of thcse membrane pores has long been a mystery, and it was postulated that water could pass either through proteinaceous pores (water channel protcins) or through locally disorganized regions of the phospholipid bilayer. Considerably more work was done on water movement through the membranes o f specific animal cells such as erythrocytes or epithelial cells, than those of plant cells, and these studies have given the most insight to the niechanism involvcd. Unlike diffusional water permeability, water movement across erythrocyte membranes is reversibly inhibited by mercuric chloride which implied that ;I protein was responsible. The functional unit of the water channel in kidney tubules and red blood cell membranes is 30 kDa. as determined by radiation inactivation. Such observations point to the existence o f proteinaceous pores of water transport. I t was found that messenger K N A from cclls with high water permeability, when expressed in Xenopiu oocytcs, caused an increase in the water permeability of the oocyte. The molecular identity of the water channels remained elusive until the recent identification of the protein AQP1 (CHIP) in the erythrocyte plasma membrane (Preston ef af., 1902) and the protein y-TIP in the Arcrhidopsis thulinrza tonoplast (Maurel et al., 1993). These water channel proteins are refcrred to iis aquaporins.

11.

HOW ARE WATER CHANNEL PROTEINS ASSAYED?

Demonstration of a y ua pori ti -111ed i a t ed water per mea bi li t y was fi rs t accomplished hy the injection of Xenopus oocytes with aquaporin cKNA followed by measurements of the osmotic permeability of the oocytes. Normal oocytes are minimally permeable to water and swell slowly when cxposed to hypo-osmolar buffers. When the bathing medium is diluted tive-fold, it takes 4.5-60 min for the volume of oocytes to increase roughly I.5-fold, and for the oocyte membrane to rupture as a result. Oocytes injected

42 I

AOUAPORINS AND WAl’tK TRANSPORT I .60

-

1 .so -

1.40

-

g-TIP

-.-o---- GlpF ----&----

n.n

1.0

2.0

3.0

4.0

u n i 11 j c c t c d

5.0

6.0

Time (min) Fig. I . Swelling o f oocytes injected with three different MIP proteins: ./-TIP (g-TIP), tr-’TIP (a-TIP)and GIpF.

with cRNA for the erythrocyte protein AQPl or the tonoplast protein ?-TIP 2-4 days prior to a shift to a hypo-osmotic medium. swell 1G20 times more rapidly, and burst after 3-4 niin (Preston c’t a l . , 1992; Maurel el al., 1993) (Fig. 1 ) . The mere presence of a protein of the MIP (major intrinsic protein) family in the oocyte plasma membrane is n o t sufficient to increase the water permeability of the oocyte. Expression i n Xerzopus oocytes of the Escherichiu coli protein GIpF. which permits the facilitated transport of glycerol into the bacterial cells, greatly enhances their capacity to take up glycerol, but does not change the water permeability of the oocytes (Maurel et nl., 1994). The swelling rate can be used to calculate the osmotic permeability (P,) o f the membrane, and P+ increases from about 0.002 to 0.015 cm s-’ when plant or animal aquaporins are present. Membrane water transport by AQPl or ?-TIP was not accompanied by a measurable increase in conductance, indicating that the protein does not ge of ions. Even when the outside p t l is 4.0 and the H + gradient is very strong (1000-fold difference) there is no current, suggesting that H.lOf does not go through the channels. Both the small pore size and the presence of charged residues within the channel may restrict the passage of ions. This is true whether the oocytes are in an iso-osmotic o r a h y po -osmot i c me d i u m . The question arises as to why a tonoplast protein should be directed to

422

M. J. CHRISPEELS

el a1

the plasma membrane of the oocytes. The synthesis of new proteins can be ascertained with radioactive precursors, but the presence of y-TIP in the plasma membrane is inferred from the swelling assay, and there is no direct proof for its presence there. However, in animal cells the plasma membrane is the default destination for membrane proteins that enter the secretory system, and it is therefore a logical assumption that in oocytes a tonoplast protein will end up in the plasma membrane. Proof that a membrane protein has arrived at the plasma membrane of the oocyte can be obtained by immunofluorescence microscopy if antibodies to the protein are available, or by fractionating the oocyte membranes. New methods are now being developed to assay the activity of aquaporins by expressing them in yeast o r in E. coli. Spheroplasts prepared from these unicellular organisms shrink when exposed to a hypertonic medium. The rate of shrinking can be measured by determining the change in light scattering with a spectrophotometer equipped with a stopflow device. The rate of shrinking is much faster when functional aquaporins are present in the limiting membranes of the cells. This method has the advantage that the preparation of cRNA is eliminated and that gene constructs can be tested directly. Furthermore, with E. coli, there is no requirement for transport through the endomembrane system, only for incorporation into the limiting membrane.

111. THE DISCOVERY OF AQUAPORINS IN PLANTS In the late 1980s and early 1990s, a number of laboratories identified plant cDNAs that encode proteins with six putative membrane spanning domains: NOD26, a protein of the peribacteroid membrane in soya bean nodules (Fortin et al., 1987; Sandal and Marcker, 1988), TobRB7, a root-specific cDNA of tobacco (Yamamoto et al., 1990), clone 7a, a turgor-responsive cDNA of the pea (Guerrero et al., 1990), TIP, an abundant protein in the protein body membranes of bean cotyledons (Johnson et al., 1990) and RD28, a desiccation-induced cDNA of Arubidopsis (Yamaguchi-Shinozaki et a l . , 1992). Analysis of the amino acid sequences of these proteins showed them to be related to the previously described MIP of the bovine lens (Gorin eta/., 1984) and the E. coli solute transporter GlpF (Muramatsu and Mizuno, 1989). Many genes were cloned, but the function of the gene products remained elusive until the demonstration that injection of cRNAs of AQPl and y T I P into Xenopus oocytes causes the oocytes to swell in a hypo-osmotic medium. y-TIP, the first plant aquaporin to have demonstrated water transport activity, is a tonoplast protein (Hofte et al., 1992), but subsequent experiments showed that aquaporins called PIPS are also found in the plasma membranes of plant cells (Daniels et al., 1994; Kaldenhoff et al., 1994; Kammerloher et al., 1994). These findings have profound implications for

AQUAPORINS AND WATER 'I'RANSPOR?'

423

understanding water flow through living tissues of plants. Water flow is thought to be either apoplastic or transcellular, with the transcellular route accounting fcx the bulk o f the flow. The presence of water channel proteins i n the plasma membrane and the tonoplast would allow the plant to regulate most of its water flow either by changing the abundance of the aquaporins or their activity

JV. AQUAPORIN TIP AND AQUAPORIN PIP ARE MEMBERS OF A LARGE GENE FAMILY Research in our laboratory (Hiifte et d., 1992) and that of T. Schiffner (Kammerloher et al., 1994) showed that Aruhidopsis thulianu contains multiple PIP and TIP cDNAs. A recent search of the dbEST database indicates that there are at least 22 different expressed PIP and TIP genes in A . fhaliunu (Fig. 2). This surprisingly large number suggests considerable functional redundancy and/or cell type or tissue-specific isoforms of TIP and PIP. An cvolutionary tree showing the relatedness of the sequences indicates that they fall into two groups: the alrcacly known TIPS are in one group and the already known PIPs in the other. On this basis, we presume that the 1 1 sequences in the top half of Fig. 2 are PIPs, whereas the bottom half represent 'TIPS. Whether all these proteins have water channel activity still needs to be documented. So far this has been shown only for three TIPS and six PIPs. The development of easier methods to assay water channel activity will help indicate whether they are all water channels. In addition, i t remains to be determined if the facilitated transport is specific for water. For example, AQP3, a newly identified mammalian aquaporin, is permeable not only to water but also to glycerol and urea (Ishibashi et al., 1994). For only one plant protein, NOD26, has a different transport function been claimed. NOD26, when incorporated into lipid bilayers, forms ion channels (Weaver et al., 1994). Early work on MIP using the same method also led to the conclusion that it is an ion channel (Zampighi et a/., 1085), but subsequent work showed it to be a water channel (Mulders et al., 1995). The possibility therefore remains that NOD26 is a water channel or a multifunctional channel.

V.

THE STRUCTURE OF T H E AQUEOUS PORE

Where the oligoineric structure of aquaporins has been examined, the 25-30 kDa MIP proteins are found to form tetramcrs (Verbavatz el id., 1903). The low activation energy (EA

Advances in Botanical Research, Volume 25

Advances in Botanical Research, Vol. 25

Advances in Botanical Research, Volume 30

Advances in Botanical Research, Volume 41

Advances in Botanical Research, Volume 24

Advances in Botanical Research, Volume 30

Advances in Botanical Research, Volume 55

Advances in Botanical Research, Volume 42

Advances in Botanical Research, Volume 33

Advances in Botanical Research, Volume 9

Advances in Botanical Research, Volume 39