Advances in Botanical Research, Volume 19

Advances in

BOTANICAL RESEARCH VOLUME 19

Advances in

BOTANICAL RESEARCH Editor-in-Chief J. A. CALLOW

School of Biological Sciences, University of Birmingham, Birmingham, England

Editorial Board M. E. COLLINSON H. G. DICKINSON R. A. LEIGH D. J. READ

Kings College, London, England University of Oxford, Oxford, England Rothamsted Experimental Station, England University of Sheffield, Sheffield, England

Advances in

BOTANICAL RESEARCH Edited by

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

VOLUME 19

1993

ACADEMIC PRESS Harcourt Brace & Company, Publishers

London

San Diego New York Boston Sydney Tokyo Toronto

This book is printed on acid-free paper

ACADEMIC PRESS LIMITED 24/28 Oval Road, London NW17DX

United States Edition published by ACADEMIC PRESS INC. San Diego, CA 92101

Copyright 01993 by ACADEMIC PRESS LIMITED

All rights reserved No 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.

A CIP record for this book is available from the British Library ISBNCk12-005919-3

Typeset by Phoenix Photosetting, Chatham, Kent Printed in Great Britain by TJ Press, Padstow, Cornwall

CONTRIBUTORS TO VOLUME 19

S. ALDINGTON, Centrefor Plant Science, Universityof Edinburgh, Daniel Rutherford Building, The King's Buildings, Mayfield Road, Edinburgh EN9 3JH, Scotland, UK J. G. DUCKET", School of Biological Sciences, Queen Mary and Westfield College, University of London, Mile End Road, London El 4NS, UK S. C. FRY, Centre for Plant Science, University of Edinburgh, Daniel Rutherford Building, The King's Buildings, Mayfield Road, Edinburgh EH9 3JH, Scotland, UK M. B. JACKSON, Department of Agricultural Sciences, University of Bristol, AFRC Institute of Arable Crops Research, Long Ashton Research Station, Bristol BS18 9AF, UK R. LIGRONE, Dipartimento di Biologia Vegetale, Universita di Napoli, Via Forla 223, I-80139 Napoli, Italy G. I. McFADDEN, Plant Cell Biology Research Centre, School of Botany, University of Melbourne, Parkville VIC 3052, Australia K. S. RENZAGLIA, School of Biological Sciences, Box 23590A, East Tennessee State University, Johnson City, TN 37614, U S A

This Page Intentionally Left Blank

PREFACE

In this volume of Advances in Botanical Research we start with an article by Aldington and Fry on the “oligosaccharins”, the name given to that group of diverse oligosaccharides which exerts biological activity in plants, at low concentration. Oligosaccharins have been implicated in a wide range of physiological processes and are frequently described as being “hormonelike”. Some of these roles are by now fairly well established, most notably in defence response elicitation. Other roles are still speculative and controversial and this review attempts to explore some of these areas of controversy, to identify gaps in our knowledge of them and to provide pointers for future work. It brings together a wide range of topics, including methods for preparation and chemical characterization, the range of physiological effects, modes of action and transport properties. Intuitively, it seems fairly obvious that plants must possess strong controlling mechanisms to balance the growth of their various organs and a great deal of research does demonstrate that the growth and behaviour of shoots is coupled closely with that of roots, and that the internal controls are strongly influenced by environment. It has often been suggested that these environmental influences operate indirectly, by regulating the hormonal traffic between the two organs rather than through more direct influences following changes in water or mineral supply and the main thrust of Jackson’s article is to assess the evidence relating to this hypothesis. It would appear that no unequivocal conclusions can yet be reached because of limitations in the experiments that have sought to determine hormonal fluxes. The author identifies the need for more quantitative studies which take advantage of modern physicochemical and immunological methods and for computer-based modelling techniques which would enable a more comprehensive exploration of the hormone ‘economy’ of the whole plant. The general theory of endosymbiosis of photosynthetic prokaryotes as a basis for evolution of green algae and subsequently land plants is supported by a wealth of morphological, biochemical and molecular evidence. The origin of photosynthetic capacity in other groups of algae is less certain and the very diversity of algal chloroplasts has prompted speculation that they may have arisen from separate endosymbiotic events involving many different prokaryotes, or even the entrapment of photosynthetic [vii]

...

Vlll

PREFACE

eukaryotes. McFadden’s article reviews the morphological and molecular evidence relating to the origin of cryptomonad algae, reaching the conclusion that this group arose from an association between an unknown predatory phagotrophic flagellate and a chloroplast-containing eukaryote, probably a red alga, and thus involves no less than four different evolutionary lineages. The phenomenon of “alternation of generations” and its relevance to the study of phylogeny, taxonomy and functional biology of land plants was discussed extensively by Bell in volume 16 of this series. In the present volume, the article by Ligrone, Duckett and Renzaglia takes this analysis a step further by considering one aspect of this in greater detail. In all land plants there is an embryonic phase, of variable duration, during which the sporophyte generation is in direct physical contact with the gametophyte and the interface between these two generations, the so-calledplucentu, thus plays a critical role in integrating the two phases of the life-cycle. In their review, the authors present a detailed and comparative anatomical and ultrastructural analysis of this interface, including the first detailed and systematic study of many groups of land plants. As usual, I would like to thank the authors for their excellent contributions, for their patience with the editor and their efforts to make his task easier.

JA CALLOW

CONTENTS

. . . . . .

V

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

vii

CONTRIBUTORS TO VOLUME 19 . . . . . . PREFACE

,

,

. .

,

Oligosaccharins S. ALDINGTON and S. C . FRY I.

Introduction . . . . . . . . . . . . . . . . . . . . . . . A. Origin of the Oligosaccharin Concept . . . . . . . . . . B. Preparation of Oligosaccharins . . . . . . . . . . . . . C. Bioassays . . . . . . . . . . . . . . . . . . . . . . D. Purification and Chemical Characterization of Oligosaccharides . . . . . . . . . . . . . . . . . .

11. Physiology of Oligosaccharin Effects . . . . . . . . . . . . A. Fungal Oligo-p-glucans . . . . . . . . . . . . . . . B. Xyloglucan-derived Oligosaccharides as Growth Regulators C. Oligosaccharides of Pectins . . . . . . . . . . . . . . D. Oligo-P-xylansasPossibleOligosaccharins . . . . . . . E. Chito-oligosaccharidesandRelatedFragments . . . . . F. OligosaccharinsfromN-IinkedGlycoproteins? . . . . . G. Conclusions . . . . . . . . . . . . . . . . . . . .

.

2 2 3 5

6

. . . .

7 7 12 17 32 34 37 38

111. Mode of Action of Oligosaccharins . . . . . . . . . . . . . A. Evidence for Receptors . . . . . . . . . . . . . . . . B. Rapid Effects of Oligosaccharins . . . . . . . . . . . . C. DirectEffectsofOligosaccharidesonEnzymes . . . . . .

41 41 46 56

IV.

Natural Occurrence of Oligosaccharins . . . . . . . . . . . A. Natural Occurrence of Xyloglucan Oligosaccharides . . . B. NaturalOccurrenceofPecticOligosaccharides . . . . . C. Glycoprotein-derived Oligosaccharins . . . . . . . . . D. Conclusion . . . . . . . . . . . . . . . . . . . . .

t ixl

. .

.

. .

. . .

58 58 59 61 62

X

CONTENTS

V . Mechanism of Formation and Degradation of Oligosaccharins . . A . Xyloglucan Oligosaccharides . . . . . . . . . . . . . . B . Pectic Oligosaccharides . . . . . . . . . . . . . . . . C . The Role of Chitinases, p-Glucanases and Other Enzymes .

62 62 66 73

VI . Movement of Oligosaccharins within the Plant: True Hormones? . A . Possible Transport of Xyloglucan Oligosaccharides . . . . B . Non-transport of Wound Signals . . . . . . . . . . . . C . Transport of Elicitors . . . . . . . . . . . . . . . . .

74 75 75 76

VII . Concluding Remarks . . . . . . . . . . . . . . . . . . . .

77

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

77

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

77

Acknowledgements

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

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

104

I1. The Hormone Message Concept . . . . . . . . . . . . . . .

111.

A . Different Kinds of Hormonal Message . . . . . . . . . . B . Quantifying Hormonal Messages in Transpiration Stream . C . Assessing Developmental Impact of Hormonal Messages . .

106 106 107 111

Evidence for Regulation of Root : Shoot Ratio by Roots . . . . A . Nutrient Control Theory . . . . . . . . . . . . . . . B . Shortcomings of Nutrient Control Theory . . . . . . . . C. Conclusions . . . . . . . . . . . . . . . . . . . . .

112 112 113 116

IV . Examples of Hormone-like Action of Roots on Shoots . . . . . A . Early Research . . . . . . . . . . . . . . . . . . . B . Leaf Senescence . . . . . . . . . . . . . . . . . . . C . Shoot Extension. Photosynthesis and Flowering . . . . . D . Conclusions . . . . . . . . . . . . . . . . . . . . . V . Cytokinins . . . . . . . . . . . . . . . A . Introduction and Early Research . . . B . Development in Unstressed Plants . . C . Root Excision Studies . . . . . . . . D . Responses to Mineral Nutrient Shortage E . Effects of Other Stresses Applied to Roots F. Conclusions . . . . . . . . . . . . .

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

117 117 117 120 122

. . . .

123 123 125 128 131 133 138

VI . Gibberellins . . . . . . . . . . . . . . . . . . . . . . . A . Introduction . . . . . . . . . . . . . . . . . . . . B . Studies on Unstressed Plants . . . . . . . . . . . . . .

138 138 139

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

. . . . .

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

CONTENTS

xi

C . Effects of Root Excision and Environmental Stresses Applied . . . . . . . . . . . . . . . . . . . . . toRoots D . Conclusions . . . . . . . . . . . . . . . . . . . . .

142 143

VII . Ethylene . . . A . Introduction B . Flooding . C . Conclusions

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

VIII . Abscisic Acid . . . . . . . . . . . . . . A . Introduction . . . . . . . . . . . . . B . Water Deficiency and Stomata1Closure C . Water Deficiency and Leaf Expansion . D . Soil Flooding . . . . . . . . . . . . . E . Various Other Stresses . . . . . . . . F. Conclusions . . . . . . . . . . . . . IX . Final Remarks

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

144 144 145 149

. . . . . . . .

149 149 150 159 160 164 166

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

. . . . . . . . .

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

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

Acknowledgements References

. . . .

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

166 168

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

168

Second-hand Chloroplasts: Evolution of Cryptomonad Algae G . I . McFADDEN I . Introduction

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

I1. Overview of Cryptomonad Features 111.

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

The Nucleomorph . . . . . . . . . . . . . . . . A . Nucleus-like Organelle . . . . . . . . . . . B . DNA in the Nucleomorph . . . . . . . . . C . Eukaryotic Ribosomes around the Nucleomorph D . Origin of the Nucleornorph . . . . . . . . . E . Isolation of the Nucleomorph . . . . . . . .

IV . The Chloroplast . . . . . . . . . . . . . . A . Chloroplast Membranes . . . . . . . B . Storage Product . . . . . . . . . . C . Photosynthetic Pigments . . . . . . D . Chloroplast Genome . . . . . . . . V.

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

. . . . .

. . . . . .

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

190 192 192 192 195 196 200 203

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

208 208 208 208 210

Cryptomonads as Endosymbionts: Parasites of Cryptomonads and Endosyrnbionts of Cryptomonads . . . . . . . . . . . . .

213

xii

CONTENTS

VI . Second-hand Chloroplasts in Other Algae VII . Role of the Nucleomorph

. . . . . . . . . .

214

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

216

VIII . Summary . . . . . . . . . . . . . . . . . . . . . . . . . IX . Taxonomic Appendix Acknowledgements

218

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

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

219 220

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

220

The Gametophyte-Sporophyte Junction in Land Plants R . LIGRONE. J . G . DUCKEIT and K . S. RENZAGLIA I . Introduction

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

232

I1. Bryophytes . . . . . . . . . . . . . . . . . . . . . . . . A . Mosses (Bryopsida) . . . . . . . . . . . . . . . . B . Liverworts (Hepatopsida) . . . . . . . . . . . . . C . Anthocerotes (Anthocerotopsida) . . . . . . . . . .

. . . . . .

234 235 253 275

I11. The Taxonomic Significance of the Placenta in Bryophytes and Implications for Phylogeny . . . . . . . . . . . . . .

. .

283

IV . Pteridophytes . . . . . . . . . . . . . . . . . . . . V . Seed Plants

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

Acknowledgements

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

295 301 306

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

307

AUTHOR INDEX

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

319

SUBJECT INDEX

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

337

Oligosaccharins

SUZANNE ALDINGTON and STEPHEN C . FRY

Centre for Plant Science. University of Edinburgh. Daniel Rutherford Building. The King’s Buildings. Mayfield Road. Edinburgh EH9 3JH. Scotland. U K

I . Introduction . . . . . . . . . . . . . . . . . A . Origin of the Oligosaccharin Concept . . . B . Preparation of Oligosaccharins . . . . . . C . Bioassays . . . . . . . . . . . . . . . . D . Purification and Chemical Characterization of Oligosaccharides . . . . . . . . . . .

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

2 2 3 5

. . . . . . .

6

I1. Physiology of Oligosaccharin Effects . . . . . . . . . . . . . A . Fungal Oligo-P-glucans . . . . . . . . . . . . . . . .

7 7

. . . . . . . . . . .

12 17 32 34 37 38

. . . . . . .

B . Xyloglucan-derived Oligosaccharides as Growth Regulators . . . . . . . . . . . . . . . . C . Oligosaccharides of Pectins . . . . . . . . . D . Oligo-P-xylansasPossibleOligosaccharins . . E . Chito-oligosaccharides and Related Fragments F. Oligosaccharins from N-linked Glycoproteins? G . Conclusions . . . . . . . . . . . . . . . .

111.

. . . . . .

. . . . . .

. . . . . .

. . . . .

Mode of Action of Oligosaccharins . . . . . . . . . . . . . A . Evidence for Receptors . . . . . . . . . . . . . . . . B . Rapid Effects of Oligosaccharins . . . . . . . . . . . . C . Direct Effects of Oligosaccharides on Enzymes . . . . . .

41 41

46

56

IV . Natural Occurrence of Oligosaccharins . . . . . . . . . . . 58 A . Natural Occurrence of Xyloglucan Oligosaccharides . . . 58 B . Natural Occurrence of Pectic Oligosaccharides . . . . . . 59 C . Glycoprotein-derived Oligosaccharins . . . . . . . . . . 61 D . Conclusion . . . . . . . . . . . . . . . . . . . . . . 62 Advancesin Botanical Research Vol . 19 Copyright 01993 Academic Press Limited ISBN 0-12-005919-3

All rights of reproduction in any form rescrved

2

S. ALDINGTON AND S. C. FRY

V.

Mechanism of Formation and Degradation of Oligosaccharins . . A. Xyloglucan Oligosaccharides . . . . . . . . . . . . . . B. Pectic Oligosaccharides . . . . . . . . . . . . . . . . C. The Role of Chitinases, P-Glucanasesand Other Enzymes .

62 62 66 73

VI.

Movement of Oligosaccharins within the Plant: True Hormones? . . . . . . . . . . . . . . . . . . . . . . . A. PossibleTransport of Xyloglucan Oligosaccharides . . . . B. Non-transport of Wound Signals . . . . . . . . . . . . C. Transport of Elicitors . . . . . . . . . . . . . . . . .

75 75 76

Concluding Remarks . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . .

77 77 77

VII.

74

I . INTRODUCTION A.

ORIGIN OF THE OLIGOSACCHARIN CONCEPT

The idea that biologically active oligosaccharides (called oligosaccharins) exist is about 17 years old (Ayers et al., 1976a,b.c; Albersheim and Valent, 1978), and is currently the subject of much imaginative research and speculation. Many exciting claims, some substantiated, have been made as to the significance of oligosaccharins. On the other hand, the phrase “believe in oligosaccharins” is still common enough to raise doubts. It thus seems appropriate to assess the current status of the oligosaccharin concept, and to evaluate objectively the biological roles ascribed to oligosaccharins. Other recent reviews include those by Dixon and Lamb (1990), Aldington et al. (1991) and Ryan and Farmer (1991). Oligosaccharins are particular oligosaccharides which, at low concentrations, exert biological effects on plant tissue other than as carbon or energy sources (Albersheim et al., 1983). Thus, while all oligosaccharins are oligosaccharides, not all oligosaccharides are oligosaccharins. “Elicitors”, in contrast to oligosaccharins, are any substances that evoke defence-related responses (especially phytoalexin synthesis) in plants. An elicitor does not have to be an oligosaccharide. An oligosaccharin does not have to evoke a defence-related response. Some, but by no means all, elicitors are oligosaccharins. Most of the known oligosaccharins are derived from cell wall polysaccharides, although it seems rather arbitrary to make this a necessary part of the definition. We would also count as oligosaccharins any biologically active oligomers that, although rich in sugar residues, also contained some non-carbohydrate material, e.g. phenolic, peptide or acyl groups. We would not count polysaccharides (say, molecular weight > 5000) as oligosaccharins, although some biologically active polysaccharides may act by virtue of possessing a particular oligosaccharin domain within their

OLIGOS ACCHARINS

3

larger structure. More precise definitions of “oligosaccharin” are neither possible nor desirable at this rapidly advancing stage in the development of the subject. The oligosaccharin concept grew out of plant pathology. The first oligosaccharins to be recognized were of fungal origin and their biological effects on Angiosperm tissues appeared to be related to the activation of defence responses (Albersheim and Valent, 1978). Soon afterwards it was found that oligosaccharins obtained from higher plant sources (so-called endogenous elicitors) can also evoke similar defence responses (Hahn et al., 1981; Lee and West, 1981a,b). Only later were plant-derived oligosaccharins shown to exert effects that appeared to be unrelated to disease resistance so that a role could be proposed in the life of the healthy plant (Albersheim and Darvill, 1985).

B. PREPARATION OF OLIGOSACCHARINS

Like cytokinins, which were first demonstrated in samples of autoclaved DNA, oligosaccharins were also first prepared by artificial means. The vast majority of research still uses such artificial oligosaccharins, a fact that may detract slightly from the credibility of oligosaccharins as biologically relevant signalling molecules. The limited evidence for the natural occurrence of free oligosaccharins is discussed in Section IV. Artificial oligosaccharins are prepared by the partial degradation of polysaccharides or whole cell walls, usually by one of four methods. Method 1. Degradation is brought about by partial acid hydrolysis (Nothnagel et al., 1983; Yamazaki et al., 1983; Broekaert and Peumans, 1988). Depending on the severity of the conditions used (acid concentration, temperature and time), this treatment cleaves a certain proportion of the glycosidic bonds in polysaccharides. Unfortunately, some glycosidic bonds (especially apiose, arabinofuranose and fucose) are much more acid-labile than others. Therefore, during partial acid hydrolysis, many of the theoretically possible oligosaccharide structures are not isolated. Nevertheless, acid hydrolysis has the advantages of cheapness and reproducibility. Dilute trifluoroacetic acid is often used because this volatile acid can readily be removed in vacuo, after hydrolysis. Also, sodium trifluoroacetate does not appear to be more damaging to plant cells than NaCl, so traces of residual trifluoroacetate that may remain after evaporation and neutralization with NaOH would not be expected to have any gross effects on metabolism. Method 2. It has been proposed that oligosaccharins could be prepared by treating the cell walls with alkali. This has been less extensively used as its effects are more difficult to define. The principal effect of cold alkalis on cell walls is solubilization of polysaccharides (especially hemicelluloses). Some

4

S. ALDINGTON AND S. C. FRY

polysaccharides are partially degraded by cold alkali via a “peeling” of sugar units, one by one, from the reducing terminus of the chain (Kuhn et af., 1958; Whistler and BeMiller, 1958). The rate and extent to which this occurs depends on the nature of the polysaccharide. A second effect of alkalis (at least at higher temperatures) is to cause the cleavage, by any of several mechanisms, of a small proportion of the mid-chain glycosidic linkages. Thirdly, even very mild alkali treatment will hydrolyse ester-linked substituents, e.g. ferulate and methyl esters, that may be present on the polysaccharides. Method 3. Cell walls are fragmented by autoclaving or other heat treatments. These treatments will cause partial degradation of polysaccharides by several means including Hf-catalysed hydrolysis and OH-catalysed p-elimination (Barrett and Northcote, 1965), to yield fragments of a wide variety of sizes. Each of the first three methods is liable to yield fragments of material extraneous to the cell wall. Even the most highly purified cell wall preparations are likely to be contaminated with small amounts of other material, including membranes, nuclear proteins, RNA, DNA, and polyphenolics such as tannin bodies. Any of these polymers may yield oligomers upon treatment with acid, alkali or autoclaving. Some of these fragments possess biological activity: phenolic substances have diverse effects on plants (Isaiah, 1971; Corcoran et af., 1972; Danks et af., 1975; Blum and Dalton, 1985), and any contaminating DNA would yield cytokinins upon autoclaving (Miller et af., 1955). In addition, heating can cause carbohydrates to undergo chemical reactions producing substances such as maltol and isomalto1 (components of the aroma of freshly-baked bread) (Backe, 1910), and to react with proteins and amino acids to produce substances known to food scientists as “non-enzymic browning products” (Eble et af.,1983; Goodwin, 1983). Not always has sufficient consideration been given to the possibility that “oligosaccharin” activities associated with cell wall fragments produced by these methods may not be due to simple oligosaccharides. Method 4. The fourth and certainly the best method of preparing oligosaccharins is by partial enzymic degradation of polysaccharides or cell walls. We would assume that, if oligosaccharins are produced in vivo, it would be by enzymic degradation. Therefore, if the right enzyme(s) can be found, we have the ideal way to make “realistic” oligosaccharins. The enzymecatalysed reaction can be stopped at various stages, thus generating fragments of diverse sizes. After the digestion, the enzymes can be precipitated, removed chromatographically, or inactivated by boiling. Enzymes that have been used in this way include cellulase (York et af., 1984; McDougall and Fry, 1988), pectinase (Bishop et al., 1981; Branca et al. , 1988), pectate lyase (Davis et al., 1986a,c) and chitinase (Kurosaki et al., 1988). Research has until recently been hampered by unavailability of these enzymes in pure

OLIGOS ACCHARINS

5

form, but specialist suppliers are beginning to offer some of them so future prospects are good. Other methods for the preparation of oligosaccharins have been mooted, e.g. ultrasonication of polysaccharides, and the enzymic synthesis of oligosaccharides de novo via transglycosylation reactions (Bezukladnikov and Elyakova, 1988; Nilsson, 1988; Usui et al., 1990), but relatively little use has been made of these methods to date. The chemical synthesis of oligosaccha; et rides de novo is a rapidly developing field (Sharp et af., 1 9 8 4 ~Ossowski al., 1984; Nakahara and Ogawa, 1987,1990; Sakai et al., 1990; Torgov et al., 1990; Cheong et al., 1991) and provides a very powerful means of confirming the proposed structures of oligosaccharins as well as of exploring structureactivity relationships (for review, see Aldington et al., 1991). Oligosaccharides are among the very few natural products to be (1) hydrophilic enough not to partition from water into butanol (unlike many plant hormones, which are lipophilic weak acids) and (2) t o have molecular weights of about 600-3000 (larger than most intermediary metabolites; smaller than proteins, nucleic acids and polysaccharides). These features make oligosaccharides relatively easy to isolate. Separation methods commonly used include phase partitioning, gel-permeation chromatography to determine the size of the active molecules (Kobata et al., 1987), ion-exchange chromatography (Redgwell and Selvendran, 1986) or electrophoresis (Stoddart and Northcote, 1967) to determine charge, and highpressure liquid chromatography (HPLC) to effect final purification (Sharp er al., 1984a; McDougall and Fry, 1991a). C. BIOASSAYS

The definition of “oligosaccharin” demands some effect on plant tissue. Therefore, having prepared (or possibly isolated from natural sources) a mixture of oligosaccharides, the only way to demonstrate the presence of oligosaccharins is to perform a bioassay. Unfortunately, bioassays are notorious for their irreproducibility. There are many possible reasons for this lack of consistency: batches of plant material may vary genetically; one year’s harvest of seeds may differ phenotypically from the next; local conditions under which seedlings are grown may vary in subtle ways; different organs are used; scientists differ in the way they handle the plants; the physical stress inflicted on the plant by administration of the oligosaccharin may vary. Tissue cultures are particularly prone to change between one sub-culturing and the next and certainly change as they pass through the growth cycle. Oligosaccharin folklore is full of stories about differences between types of Petri dish, effects of volatile substances derived from particular plastics, auspicious corners of the greenhouse where the assay is always successful, and even of the benefits of “green fingers”!

6

S. ALDINGTON AND S. C. FRY

Ultimately, bioassays are the only way to detect biological activities, and the present chapter illustrates the widespread support that has emerged for the existence of biological activities of oligosaccharides despite the experimental difficulties. The most reproducible bioassays are those in which the measured effect is relatively close to the initial action of the oligosaccharin. In contrast, with morphogenetic effects such as flowering (Gollin et al., 1984), the ultimate effect (flower development) may be far removed in space and time from the initial molecular action of the oligosaccharin, and thus the chain of events is particularly susceptible to interruption. At the other extreme, direct effects of oligosaccharins on enzyme action, or on membrane functions, are rapid and depend on few or no intervening steps. However, studies limited to this level would miss some of the possible systemic effects of oligosaccharins (cf. Hammerschmidt and KuC, 1982; Wildon et al., 1989). Bioassays are also at the heart of the methodology required to purify an oligosaccharin. The methods of polysaccharide fragmentation outlined in Section 1.B generate mixtures of oligosaccharides, most of which may lack oligosaccharin activity. The initial crude preparation can be subjected to a series of separation methods, but at each step a bioassay is required to indicate which fractions contain the activity. D. PURIFICATION AND CHEMICAL CHARACTERIZATION OF OLIGOSACCHARIDES

It is not possible here to discuss this topic in detail (see Fry, 1988; Dey, 1990), but the current status will be briefly reviewed. Oligosaccharides are often initially fractionated. according to their native charge, by ionexchange chromatography, e.g. on a column of QAE- (quaternary aminoethyl-) Sephadex (Baydoun and Fry, 1985). The second criterion is often size: a sub-fraction of the oligosaccharides, such as the neutral ones, is further fractionated by gel-permeation chromatography (gel filtration), e.g. on Bio-Gel P-2 or Sephadex G-25 (Sharp el al., 1984b). These gels are often set up in moderately large columns, e.g. of 1litre capacity, to which about 50ml of sample containing about 1g of oligosaccharides can be applied. Often the lowest molecular weight fraction with biological activity is taken for further purification because this will be the easiest to characterize chemically and because it may be the essential core structure conferring biological activity on larger molecules. Further fractionation can be achieved with a variety of chromatographic methods, e.g. preparative paper chromatography (which will accommodate about 10 mg per sheet; for reviews, see Bailey and Pridham, 1962; Fry, 1988), affinity chromatography on immobilized lectins (Blake and Goldstein, 1980; Green and Baenzinger, 1989), or semi-preparative HPLC (which will typically accommodate about 100 kg per run; McDougall and Fry, 1990,1991a). A recent development in

OLIGOSACCH ARINS

7

(analytical rather than preparative) HPLC is the use of anion-exchange chromatography on a pellicular resin (Dionex “CarboPac”) with 0.1 M sodium hydroxide plus sodium acetate as eluent: under such alkaline conditions even “neutral” oligosaccharides acquire a negative charge, by ionization of some of the -OH groups, and can thus be separated by anion-exchange chromatography. Excellent resolution of some closely related oligosaccharides is possible, providing evidence for or against homogeneity of an oligosaccharide preparation (Hardy and Townsend, 1989; McDougall and Fry, 1991a). The major limitation of this method for preparative purposes is the need to remove the NaOH from the sample very quickly after chromatography so as to minimize alkaline degradation. Another promising advance in HPLC of oligosaccharides is the use of graphitized carbon columns (Koizumi et al., 1991). Once purified, the oligosaccharide can be structurally characterized. Features that can readily be determined include monosaccharide composition (by acid hydrolysis and chromatography), anomeric configuration (by susceptibility to specific glycosidases), pyranose/furanose ring form (by susceptibility to mild acid hydrolysis), the occurrence of certain repeating motifs, e.g. Xyl-a-(1+6)-Glc (by enzymic release; Kooiman, 1961), and the reducing terminus (by NaB3H4-reduction to the corresponding [3H]alditol; Hall and Patrick, 1989). Precise linkages between other sugar residues in the oligosaccharide, as well as the presence of non-carbohydrate moieties, can often be determined by the rapidly advancing techniques of ‘H and 13C nuclear magnetic resonance spectroscopy (NMR) and fast-atombombardment mass-spectrometry (FAB-MS), as well as by conventional methylation analysis. Despite the many recent advances in carbohydrate chemistry, it remains considerably more difficult to sequence an oligosaccharide than an oligopeptide. This is because oligosaccharide structures have more variables D- versus L-isomers, pyranose versus than oligopeptides-including furanose ring-forms, a-versus P-anomers, (1+2)-, (1+3)-, (1+4)-linkages, etc. , and the presence of numerous modifications, e.g. deoxy-sugars, sugar acids, amino-sugars and O-acetyl-sugars. The automatic oligosaccharide sequencer is a long way off! It is also more difficult to apply the techniques of molecular biology to oligosaccharides than to peptides because the former are several steps further removed from transcription than the latter.

11. PHYSIOLOGY OF OLIGOSACCHARIN EFFECTS A.

FUNGAL OLIGO-8-GLUCANS

Oligo-P-glucans were the first oligosaccharins to be recognized; it is therefore appropriate to trace the work on them first and in some detail. It is well

8

S. ALDINGTON AND S. C. FRY

established that plants, when challenged by microorganisms, can often resist becoming diseased by switching on any of a wide range of defence responses such as the accumulation of phytoalexins, lignin, silica, callose, extensin, peroxidase, chitinases and P-glucanases, and the activation of the hypersensitive response (EsquerrC-Tugaye and Mazau, 1974;Bell, 1981;Bird and Ride, 1981; Bailey, 1982; Hammerschmidt and KuE, 1982; Kratka and Kfidela, 1984; Mauch et al., 1988a,b). Studies in the mid-l970s, aimed at identifying the factor(s) by which plant cells can recognize the presence of foreign (fungal) cells, demonstrated that fungal cell wall components played a particularly important role. Specifically, it was shown that mixed-linkage P-( 1 4 ),( 1+6)-~-glucans, which are prominent components of the cell walls of many fungi but not of higher plants, were able to activate the synthesis of phytoalexins in uninfected plants (Ayers et al., 1976a,b,c; Ebel etal., 1976;for reviews, see Albersheim et al., 1981; West, 1981;Lamb et a f . , 1989). Commercial yeast extract (i.e. autolysate) was also found to contain ethanol-precipitable elicitor activity, which was due largely to the P-glucan component rather than to the more abundant a-mannan (Hahn and Albersheim, 1978). Phytoalexins are low molecular weight compounds with anti-microbial properties; they are virtually absent in healthy tissues but are synthesized and accumulated by the plant after exposure to microorganisms, at and fairly near the site of infection. Phytoalexins are chemically diverse-they include phenolics, terpenoids and polyacetylenes, the precise compound(s) formed depending on the plant species challenged (Grisebach and Ebel, 1978; Bailey and Mansfield, 1982). The production of phytoalexins by plants appears to be a widespread response which can aid in disease resistance. Elicitation of phytoalexins has, in a number of cases, been shown to depend on the synthesis of the rate-limiting enzyme in the biosynthetic pathway (Hahlbrock et al., 1981;Chappell et al., 1991). Plants may inhibit the growth of many microorganisms (both prokaryotes and eukaryotes) by accumulating high concentrations of phytoalexins, which have a very indiscriminate action. One of the differences between a successful and an unsuccessful infection may be that phytoalexins are not synthesized rapidly enough or in sufficient quantities (Darvill and Albersheim, 1984; KuE and Rush, 1985; Ebel, 1986). Detailed studies by Albersheim’s group, designed to determine the precise components of the fungal cell wall material responsible for the biological activity, established that quite small oligosaccharides of P-( 1 4 , (1+6)-~-glucan were effective (Albersheim and Valent, 1978). Partial acid hydrolysis of the fungal polysaccharide yielded a mixture of oligosaccharides; these were fractionated according to size by gelpermeation chromatography. The fractions obtained were bioassayed for their ability to elicit the synthesis of a flavonoid phytoalexin (glyceollin) in excised soybean cotyledons. After NaBb-reduction (in order to simplify

9

OLIGOSACCHARINS

the chromatography), it turned out that the smallest highly activated material was a heptasaccharide (strictly a hexasaccharidyl-alditol), which was purified to homogeneity by HPLC, and whose complete chemical structure was determined (Sharp et al., 1984a,b) (Table I). The structure of the active heptasaccharide was elegantly confirmed by chemical synthesis (Sharp et al., 1984c; Ossowski et al., 1984), and it was shown that seven other, closely related, heptasaccharides had much less activity (Sharp et al., 1984a). The active principal was a simple, pure oligosaccharide with no amino acid or other constituents. This discovery provided a very firm footing for the oligosaccharin concept, contradicting the view that cell wall carbohydrates were inert, purely structural and relatively uninteresting substances. Although much of this work was done with soybean cotyledons, it was established that P-( 1+3),( 1+6)-~-glucans also elicited the synthesis of different phytoalexins in a wide range of other plant species (Cline et al., 1978; Darvill and Albersheim, 1984). The oligosaccharin phenomenon therefore seemed to have a wide applicability. The active heptasaccharide was effective in very low dosesapproximately 0.lpmol per cotyledon (Sharp et al., 1984b). The doseresponse curve resembled a rectangular hyperbola, such as is frequently seen in enzyme kinetics; the concentration giving a half-maximal effect (equivalent to the K,) was about 10 nM (Cheong et al., 1991). A careful investigation of the structure-activity relationships of a wide range of related oligosaccharides (Table I) showed that, although several oligo-P-glucosides were able to elicit phytoalexin synthesis at high concentrations ((LM to mM), effectiveness in the 5-50 nM range depended on the presence of at least the following core hexasaccharide (Cheong et al., 1991): ~ - G l c -1+6)-~-Glc-( ( 1+6)-~-Gk-(1 - - + 6 ) - ~ - Gt) l~( (*I 3 3

t

1 D-GIc(*)

t

1 D-G~

where D-GIc is a P-D-glucopyranose residue. Larger oligosaccharides (e.g. compounds 1 and 2,Table I) were active if they possessed this motif within their structure.The two non-reducing terminal glucose residues marked (*) were essential for maximal activity (Sharp et al., 1984a), and activity was reduced if either of these residues was converted to glucosamine (compounds 6 and 7) or N-acetylglucosamine (compounds 8 and 9). Modification of the reducing terminus (t) had relatively little effect. Through this discriminating recognition system, the plant is presumably provided with a means of detecting small amounts of fungal P-(1+3),(1+6)-glucans in the presence of the much larger amounts of other P-glucans that are found in all higher plants.

TABLE I Concentrations of oligo-p-glucosides required to elicit phytoalexin synthesis in vivo and to compete with the specific membrane-binding of a highly elicitor-active, radiolabelled oligo-/3-glucoside in vitro

Compound structurea

Concentration (nM) required for half-maximal biological effect . ) b or membrane binding ( TI .)" ..

(ml.. 101

102

103

104

I

I

I

I

105 I

I

I

I

I

I

I

I

I

I

I

I

I

1

I

I

.......... I I

............. ......... .............:...... :.: ..: ..I.. :::::: :.. .:.I:.. ... ............ . . .I.... . .....~.;~.-...$<::.: ....... .:.:.. I

I

I

I

I

I

I

I

I

I

:::.:.:::......:............ ................. :........ ..,::...:::..:: ................. :.....;..::::. ..............

106

107

I

I

10

.............. ...... ..I..t:l l::.):

I I I I

11 I

I

I

I

I

I

I

I

I

I

I

I

......:I . I.....:.I ..I:...>1;...... .:, ................... ..... ..: ;...:]...I... 2.1.. .I::.t~.:t.:rl..-.:l:t:t:,~-.~ ...I..] .>%

12

I

I

2.1

13 14

Symbols used in structures: 0 , propyl group; 0 , ally1 group; 0, reducing terminal glucose unit; 0, glucose residue; +, P-(1+6)-linkage; t , P-(1+3)linkage; =),other glycosidic linkage; 0, glucitol unit; 6, glucosamine residue; 0 , N-acetylglucosamine residue. All the sugar residues are in the P-D-pyranose form. .) shows the approximate concentration of each The soybean cotyledon bioassay devised by Ayers et al. (1976a) was used; the upper bar oligo-P-glucoside required to elicit half-maximal accumulation of glyceollin. The “maximal” response was the amount of glyceollin elicited by saturating implies that the concentration indicated gave less than a half-maximal response. levels of compound 3. The symbol ‘The concentration of the same oligo-P-glucosides required to inhibit (competitively) 50% of the specific binding of a radioiodinated derivative of compound 3 was also tested (lower bar, CIII]...;see Section III.A.l). Data summarized from Cheong et al. (1991) and Cheong and Hahn (1991).

(m..

m+

12

S . ALDINGTON AND S. C. FRY

The above investigations have provided a clear indication of the chemical nature of the active principle from fungal cell walls. In addition, numerous other studies have probed the diverse biological effects of various carbohydrate-rich fragments derived from fungal cell walls. These preparations have often been less well defined chemically. Effects reported include: (1) the accumulation of lysozyme and chitinase (Bernasconi et al., 1986), (2) induction of an arabinosyltransferase (Bolwell, 1984, 1986) and a prolyl hydroxylase (Bolwell et al. ,1985b;Bolwell and Dixon, 1986)involved in glycoprotein synthesis, (3) evocation of quantitative and qualitative changes in the phenolic components of the plant cell wall (Bolwell et al. , 1985a) and in the levels of an enzyme (phenylalanine ammonia-lyase) partly responsible for these (Bolwell et al., 1 9 8 5 ~ ) ~ (4) the synthesis of diverse low molecular weight secondary metabolites in plant tissue cultures, including several commercially valuable alkaloids, e.g. berberine in cultured cells of Thalictrum rugosum (Funk et al., 1987; see also Constabel and Eilert, 1986), ( 5 ) a burst of respiration (Funk etal., 1987) and ethylene synthesis (Piatti et al., 1991) in cultured cells. Nigeran, a different mixed-linkage fungal polysaccharide [a-( 1-+3), (1+4)-~-glucan], is also capable of eliciting the synthesis of phenylalanine ammonia lyase, in Petunia cell suspension cultures, although at the relatively high concentration of 400 mg 1-’ (Hagendoorn et al. , 1990). Perhaps surprisingly, it seems that p-( 1 4 ),( 1+6)-~-glucans can block the induction by other fungal components of another defensive reactionthe hypersensitive response (Section II.C.4)-in potato protoplasts (Doke and Tomiyama, 1980a,b). It is unclear why the plant should respond “less defensively” to a combination of two fungal components than to one. For further discussion of possible synergism between oligosaccharins, see Section II.G.3.

B. XYLOGLUCAN-DERIVED OLIGOSACCHARIDES AS GROWTH REGULATORS

Xyloglucan was first described as a storage polysaccharide of certain seeds (Kooiman, 1961). Only later was it shown to be present in large amounts in the primary cell walls of higher plants, although here, as a structural component, is where it undoubtably plays its major role (Bauer et al., 1973). It is a hemicellulosic polysaccharide, i.e. it cannot readily be extracted from the cell wall in hot water or by chelating agents, but can be solubilized by cold concentrated alkali-although even NaOH at 6.0 M, the optimum

F = a-L-fucose G = P-D-glucose L = P-D-galactose X = a-D-xylose

fi , il = (1+2)-glycosidic linkages

t,4

+ +

= = =

A

Fig. 1.

(1+6)-glycosidic linkages (1+4)-glycosidic linkage (1+4)-glycosidic linkage susceptible to hydrolysis by cellulase.

General arrangement of major sugar residues in xyloglucan.

14

S . ALDINGTON AND S. C. FRY

concentration, may take many days for efficient extraction of xyloglucan at 25°C (Edelmann and Fry, 1992a). Xyloglucan from the primary cell wall is composed of the sugar residues D-glucose, D-xylose, D-galactose, L-fucose and L-arabinose, in decreasing order of abundance. It is readily distinguished from the other main xylosecontaining hemicelluloses, the xylans, by the fact that the xylose residues are a-linked rather than (3-linked. The arrangement of the major sugar residues of xyloglucan is shown in Fig. 1. All the sugar residues shown in Fig. 1 are in the pyranose ring form; a small amount of arabinofuranose may also be present. The galactose residues of xyloglucan are often acetylated (York et af.,1988). The structure and functions of xyloglucan have been reviewed (Hayashi, 1989; Fry, 1989a). Xyloglucan can be fragmented into a limited number of major oligosaccharides by exhaustive digestion with pure cellulase [EC 3.2.1.4; endo-p(1-+4)-~-glucanase]. This enzyme attacks the (3-( 1+4)-~-glucan (celluloselike) backbone of xyloglucan wherever there is a non-xylosylated glucose residue (marked * in Fig. 1; Bauer et af., 1973). Since these tend to occur every fourth residue along the backbone, the major oligosaccharides generated (XG7, XG8, XG9, XG9n, and XG10) are based on a G-+G-+G+G (cellotetraose) core, which may bear a variety of substituents (Valent et af.,1980; Kato and Matsuda, 1980; Matsushita et af.,1985). Other fragments such as XG5 may arise because the spacing of non-xylosylated glucose units is not completely regular and because of partial breakdown of some of the initially formed cellotetraose-based oligosaccharides by the action of contaminating enzymes present in commercial cellulase preparations. In addition, a few non-xylosylated glucose residues may carry arabinofuranose residues and therefore be protected from the action of cellulase; this leads to the production of an oligosaccharide of degree of polymerization (DP) 17 (Kiefer et al., 1990). York et af. (1984) were the first to show that one particular xyloglucanderived oligosaccharide, XG9, can regulate plant growth. XG9 can, at an optimal concentration of 1nM, partially block the promotion of growth ~ acid (2,4-D, an artificial auxin). caused by 1 p , 2,4-dichlorophenoxyacetic At higher concentrations, e.g. 100 nM, XG9 was much less effective; this was surprising because growth inhibitors are usually more effective at higher concentrations. These observations have been reproduced and extended in three other laboratories (McDougall and Fry, 1988; Emmerling and Seitz, 1990; Hoson and Masuda, 1991). It was confirmed that highly purified XG9 was active, showing that the activity resided in the structure of the oligosaccharide itself, rather than in a contaminant (McDougall and Fry, 1988, 1989a,b, 1991a). It was shown that the activity was critically dependent on the a-L-fucose residue present in XG9 since XG8 (McDougall and Fry, 1989a) and a mixture of fucose-free oligosaccharides from pine (Nealey et

-

15

OLIGOSACCHARINS

al., 1989) were inactive. The XYl

1

Glc+Glc-.

..

t XYl unit from the non-reducing terminus of XG9 was apparently irrelevant to the growth-inhibiting activity because XG5 [Fuc+Gal+Xyl+Glc+Glc] and the commercially-available 2’-fucosyl-lactose [Fuc+Gal+Glc] (Kuhn et al., 1958) were also active at low concentrations (McDougall and Fry, 1989b). On the other hand, the L-fucose was not sufficient for activity, as shown by the lack of effect of free L-fucose or methyl-a-L-fucopyranoside (McDougall and Fry, 1989b). No effect of XG9 on the growth induced by indoleacetic acid (IAA) was seen in several bioassays, e.g. using pea internodes, Azuki bean epicotyls, cucumber hypocotyls and oat coleoptiles (Hoson and Masuda, 1991). It was also found that the inhibitory effect of XG9 on the action of 2,4-D in pea stem segments could not be reversed by increasing the 2,4-D concentration, showing that the effect was uncompetitive (Hoson and Masuda, 1991). Since 1nM XG9 also blocked the stimulatory action of low pH on the elongation of pea stem segments (Lorences et al., 1990), it seems likely that XG9 interferes with some basic process common to the action of both H+ and 2,4-D on elongation, such as the generation of turgor or the production of a wall-loosening enzyme. There is an intriguing report, possibly related to the above, that 1nM XG9 will antagonize the beneficial effect of auxin on the regeneration of isolated carrot protoplasts; again, the activity was lost at 1 nM XG9 (Emmerling and Seitz, 1990). It will be of great interest to see whether any other auxin responses are antagonized by 1nM XG9. Priem et al. (1990) have reported that an L-fucose-containing oligosaccharide (see Section 1I.F) is able to regulate the growth of flax hypocotyls. It seems possible that this effect could operate via a mechanism similar to that of the L-fucose-containing XG9. Questions need to be answered about the importance of the (3-D-galactose residue to which the fucose is attached in XG9 but which is missing in Priem’s oligosaccharin. Tran Thanh Van and Mutaftschiev (1990) reported a stimulation of elongation in wheat coleoptiles by surprisingly low concentrations (- 1-10 p ~ of) unspecified heptaand nonasaccharides of xyloglucan obtained from Rubus culture filtrates. The surprising loss of ability of XG9, at higher concentrations, to block the response of pea stem segments to 2,4-D appeared to be due to a second, growth-prompting effect of this oligosaccharide. This second effect was not exhibited by XG5 or 2’-fucosyl-lactose, both of which lack the cellotetraose core (McDougall and Fry, 1989b). In agreement with this, it was found that XG9, added to stem segments in the absence of 2,4-D, was able to promote

-

-

16

S. ALDINGTON AND S. C. FRY

1.o

XG 7

Fig. 2. The effect of HPLC-purified xyloglucan-oligosaccharides on the elongation of pea stem segments. The difference in elongation between untreated and treated segments ( A L ) is plotted against concentration for each oligosaccharide. The letters a-e indicate the statistical significance of the apparent deviation of AL from 0: a,p<0.001; b, 0.001
elongation (Fig. 2). The optimum concentration for the growth-promoting effect in pea was about 1 p~ (McDougall and Fry, 1990). It thus appeared that low concentrations (- 1nM) of XG9 had one effect (growth inhibitory), whereas moderate concentrations (- 0.1-1 .O FM) had a second effect (growth promoting). The latter dominated the former. These two effects had different structural requirements, While the growth-inhibiting effect of nM XG9 was dependent on the L-fucose group and independent of the cellotetraose core, the growth-restoring activity of p~ xyloglucan-oligosaccharideswas independent of L-fucose but dependent on the cellotetraose core. Thus, 1 p XG7, ~ XG8, XG9 and XG9n all promoted growth in pea stem segments (Fig. 2) (McDougall and Fry, 1990).

OLIGOSACCHARINS

17

However, cellotetraose itself was not effective, so at least one of the a-D-xyloseunits must have been essential (Lorences et al., 1990). The mode of action of xyloglucan-derived oligosaccharides as growth promoters is discussed in Section 1II.C.1. C. OLIGOSACCHARIDES OF PECTINS

Pectic oligosaccharides were the first oligosaccharins of higher plant origin to be recognized. Pectins are major polysaccharides of the primary cell wall, especially in non-graminaceous seed plants, and also occur at a high concentration in the middle lamella. They are rich in a(1+4)-linked D-galacturonic acid (GalA) residues; other characteristic building blocks include D-galactose, L-arabinose and L-rhamnose (Deuel and Stutz, 1958; McNeil et al., 1984; O’Neill et al., 1990). Many of the galacturonic acid groups are methylesterified, especially in young tissue. Pectins can readily be fragmented by enzymic digestion. For example, pure pectinase [the term is used here synonymously with polygalacturonase (EC 3.2.1.19, also known as endopolygalacturonase, galacturonanase, etc.] partially hydrolyses the pectic polysaccharides of the primary cell wall to liberate small oligogalacturonides composed typically of between two and five a-(1+4)-~-galacturonic acid residues, and two larger fragments called rhamnogalacturonans I and I1 (RG-I and RG-11). The highly branched RG-I and RG-I1 presumably lack the unsubstituted (GalA)24 unit which is attacked by pectinase and are thus released from the cell wall as large, intact fragments. The basic constitution of the pectic fragments obtained from primary cell walls is summarized in Table 11. Di- to tetrasaccharides have also often been prepared from commercial Citrus “polygalacturonic acid”, a chemically de-esterified linear a-(1+4)-~-galacturonanwith few other sugar residues: o-GalA-tD-GalA-to-GalA~o-GalA-,o-GalA-,D-GalA-to-GalA-,D-GalA--,D-GalA-,

...

TABLE I1 Fragments obtained by pectinase digestion of the primary cell walls of dicotyledons Fragment Oligogalacturonides RG-I RG-I1

DP”

Major sugar residue(s)b

2-5

a-D-GalA Backbone: a-D-GalA, a-L-Rha; side-chains: a-L-Ara, P-D-Gal, Xyl a-D-GalA, P-L-Rha, a-D-Gal, ~-L-Fuc, 2-O-Me-~-Fuc,a-L-Arap, L-Araf, P-D-GalA, a-L-Rha, D-Api, P-D-GkA, KDO, DHA, L-aceric acid, 2-0-Me-Xyl

- 1000 - 3&60

“DP, degree of polymerization of fragments. ”Fordata, see Darvill el al. (1978). Spellman el al. (1983), Melton eral. (1986), Stevenson er al. (1988a,b), Thomas ef al. (1989) and O’Neill et al. (1990).

18

S. ALDINGTON AND S. C. FRY

Oligogalacturonides larger than the pentasaccharide can be generated from polygalacturonic acid and/or primary cell wall pectins (1) by partial acid hydrolysis (Nothnagel etal., 1983), (2) by hydrolysis with a fungal pectinase in the presence of a specific enzyme inhibitor to prevent the hydrolysis going to completion (see Section V.B.Z), or (3) by brief digestion with a pectin or pectate lyase. Reviews of pectic enzymes are given by Deuel and Stutz (1958), Rexova-Benkova and MarkoviE (1976) and Rombouts and Pilnik (1980). 1. Pectic oligosaccharides as growth regulators

Oligogalacturonides, produced by use of pectinase, have been shown to block the growth-promoting effect of auxin in pea stem segments (Branca et al., 1988). In the absence of oligogalacturonides, segments exhibited the usual biphasic response to auxin, the first peak in growth rate occurring after 30 min and the second after 120 min-probably corresponding to wall acidification and deposition of new wall material, respectively (Vanderhoef, 1980). Treatment of the segments with oligogalacturonides greatly reduced the initial elongation response and practically abolished the later response. Branca et al. (1988) interpreted this as meaning that the oligogalacturonides were mainly effective on the long-term phase of auxin-promoted growth. However, since the oligogalacturonides did alter the early response as well, it is quite possible that they acted primarily on the early phase, but that the effect had been magnified by the onset of the second phase. The most effective partial pectinase digests of polygalacturonic acid contained oligogalacturonides of degree of polymerization (DP) 10-17. At 75 mg 1-' (= 25-42 FM, depending on DP distribution), this mixture gave a 36% inhibition of IAA-induced elongation; higher concentrations did not lose their effectiveness (Filippini et al., 1992), in contrast to the situation with xyloglucan-derived oligosaccharides (see Section 1I.B). Thus, the necessary concentration for growth inhibition by oligogalacturonides was between four and five orders of magnitude higher than for xyloglucan oligosaccharides. The oligogalacturonide effect was reversible by higher auxin concentrations, and was apparently specific to the auxin response since the same oligogalacturonides had little effect on cytokinin- and gibberellin-induced growth (Branca et al., 1988). Oligogalacturonides of D P S 4 were inactive (Filippini et al., 1992), agreeing with the structural requirements for phytoalexin-elicitor activity (see Section II.C.3). Longterm responses to auxin in pea and tobacco stem explants were also antagonized by oligogalacturonides (Bellincampi et al., 1990).

-

-

2. Wound signal activity Considerable interest and some controversy has centred on the role of oligosaccharides as wound hormones (Bowles, 1990). Many plants can respond to localized injury by the systemic induction of specific proteins,

OLIGOSACCHARINS

19

particularly protease inhibitors, which may serve the useful purpose of preventing an invading microorganism or insect from efficiently digesting the plant’s proteins (for recent examples see Pefia-Cortes et al., 1988; Bowles, 1990; Keil et al., 1990; Ryan, 1990). A message appears to radiate from the site of injury in one leaf, causing other parts of the same leaf as well as distant undamaged leaves of the same plant to switch on this defence response within about 10h (Ryan, 1968; Ryan and Huisman, 1970). The induction of protease inhibitors was dependent on light and on RNA synthesis, and was partially inhibited by auxins. The induction was optimal at 36°C and failed to occur at G20”C (Green and Ryan, 1973). The hypothetical messenger (wound hormone) was termed “protease inhibitor inducing factor” (PIIF) (Green and Ryan, 1972,1973; Ryan, 1974); it was shown that the dose of PIIF required for half-maximal response in neighbouring leaves was exported from the injured leaf within 45-200min (Green and Ryan, 1972, 1973). Certain pectic substances were found to evoke protease inhibitor synthesis (Ryan, 1974; Bishop et al., 1981, 1984; Ryan et al., 1981; Bishop and Ryan, 1987), but since there is evidence that even quite small oligogalacturonides are unable to migrate out of injured leaves (Section VI.B), we refer to them as “wound signals” rather than as PIIF. Extracts of autoclaved tomato leaves (Ryan, 1974) and steamed tissues from many other plants and fungi (McFarland and Ryan, 1974) all induced protease inhibitor synthesis in (relatively) uninjured tomato leaves. Besides being heat-stable, the tomato wound signal was insoluble in organic solvents (Ryan, 1974). Its apparent molecular weight was 5000-10000 according to gel-permeation chromatography on Sephadex G-50 (Bishop et al., 1981). These data are compatible with the wound factor being a carbohydrate. Purified pectinase converted the leaf-juice wound signal to a mixture of oligogalacturonides (DP 3 2) which were also active; one experiment even appears to show wound signal activity in the monosaccharide fraction (Bishop et al., 1981). The activity of mono- to trisaccharides was elsewhere denied (Ryan et al., 1981), but the activity of the di- to hexasaccharides has now been reinstated (Bishop et al., 1984). Moloshok and Ryan (1989) claimed that the A4.’-unsaturated di- and trisaccharides, produced from polygalacturonic acid by the action of pectate lyase, are less active. Half~ or 152 p . ~ maximal protease induction was obtained with 8 p . disaccharide A4.’-unsaturated disaccharide. In the above work, excised “uninjured” tomato leaves and seedlings were supplied with oligogalacturonides through their cut petioles or hypocotyls. [A clean cut through the petiole or hypocotyl with a sharp razor blade does not constitute “injury” in this system since the cut does not, of itself, induce the lamina to synthesize protease inhibitors .] That oligogalacturonides may play a wound-signalling role in the intact plant is suggested by the fact that very small wounds, which would not normally evoke systemic protease

20

S. ALDINGTON AND S. C. FRY

inhibitor synthesis, can be made to do so by application of pectinase to the injured tissue (Walker-Simmons et al., 1984); presumably the exogenous enzyme generates endogenous pectic oligosaccharides which act as wound signals on the neighbouring uninjured tissue. Cell suspension cultures of tomato also synthesize protease inhibitors in response to pectic oligosaccharides (Walker-Simmons and Ryan, 1986). Wound-signal activity, capable of inducing the synthesis of protease inhibitors, has been demonstrated in several negatively charged substances of pectic origin, including (1) some commercial batches of pectin (Ryan et al., 1981), ( 2 ) pectic fragments of DP>25 solubilized by autoclaving plant tissues (Ryan, 1974; McFarland and Ryan, 1974; Bishop etal., 1981; Ryan et al. , 1981), (3) rhamnogalacturonan-I (RG-I) (DP 2000) generated from cultured Acer cell walls by pectinase digestion (Ryan et al., 1981), and (4) simple oligogalacturonides of DP 2 2 (Bishop et al., 1984; Ryan, 1988). No wound-signal activity was found in RG-I1 (Ryan et al., 1981). Chemical reduction of the galacturonic acid residues (converting them into galactose residues) abolished wound-signal activity in RG-I (Ryan, et al. , 1981), demonstrating the need for negatively charged moieties. However, it is clear that the structural requirements for wound-signal activity are considerably less exacting than those for many other biological effects of oligogalacturonides. Since it is difficult to see what precise features oligogalacturonides of D P 2 2 could have in common with RG-I but not with RG-11, it seems possible that the different pectic substances exert their wound-signalling effect by different means. In this connection, it may be significant that wound-signalling activity has been reported for another, completely unrelated group of oligosaccharides, derived from chitosan and therefore positively charged (Walker-Simmons et al., 1983, 1984; Walker-Simmons and Ryan, 1984; cf. Section II.E.2). In addition, a somewhat similar systemic response can be triggered by a localized heat stimulus (Wildon et al., 1989). It may thus be that protease inhibitor synthesis is designed as a generalized defence response that can be triggered by any of a wide range of molecular cues. Systemic signalling leading to lignification can also be induced by localized fungal infection (Hammerschmidt and KuC, 1982; KuC, 1982), though it is not clear whether this can also be mimicked by oligosaccharins.

3. Pectic oligosaccharides as elicitors of phytoalexin synthesis The synthesis of phytoalexins (see Section 1I.A) can be induced by a wide variety of elicitors including: (1) biotic elicitors of fungal origin, e.g. P-(1+3),(1+6)-~-glucan (Ayers et al., 1976c) and a-(1+3),( 1+4)-~-glucan (Hagendoorn et al., 1990) (Section II.A), glycoproteins (de Wit and Roseboom, 1980), lipids (Bostock et al., 1981, 1982) and pectinase (Stekoll and West, 1978; Lee and West, 1981a,b);

OLIGOSACCHARINS

21

(2) biotic elicitors of plant origin (so called endogenous elicitors) which will be discussed here; and (3) abiotic elicitors such as heavy metals and ultraviolet light (KuC and Rush, 1985). Ribonuclease A is also active as an elicitor, even after all its enzyme activity has been destroyed by heating, and may perhaps also be classified as an “abiotic” elicitor (Dixon et al., 1983b). Lyon and Albersheim (1982) showed that soybean stem tissues, when killed by a freezekhaw cycle, released a heat-labile factor (putative enzyme) that liberated a low molecular weight heat-stable elicitor (putative oligosaccharin) from purified plant cell walls. It was suggested that the abiotic elicitors (many of which, e.g. UV light and heavy metals, also kill plant cells) stimulate phytoalexin accumulation by causing some cell disruption, as in a freezekhaw cycle, thus releasing enzymes capable of liberating oligosaccharins from the walls of neighbouring cells (Darvill and Albersheim, 1984). The elucidation of the type of endogenous elicitor responsible for phytoalexin synthesis has been one of the most successful developments in the oligosaccharin story. Furthermore, some of the enzymes for phytoalexin synthesis are known and hence studies of their mRNA levels have shown how these can be affected by oligosaccharins (Section III.B.5). Stekoll and West (1978) and Lee and West (1981a,b) showed that a fungal pectinase could elicit the phytoalexin casbene in castor beans. Hahn e f al. (1981) showed that elicitor-active carbohydrate material from plant sources was rich in D-galacturonic acid, suggesting a pectic polysaccharide, with smaller amounts of L-rhamnose and D-xylose. D-Galacturonic acid was apparently an essential constituent of the elicitor-active fragments. Further characterization of the endogenous fragments showed that the D-galacturonic acid residues were a-(l-+4)-linked (Bruce and West, 1982, 1989; Nothnagel et al., 1983; Jin and West, 1984), again diagnostic of pectic material. Usually a given plant species will produce a given phytoalexin regardless of the elicitor used. A rare exception to this rule was provided by Marinelli et al. (1991), who showed that whereas a crude mixture of pectin-digesting enzymes from Aspergillus elicited 6-hydroxymellein in cultured carrot cells, a pure pectinase from Fusarium elicited 6-methoxymellein in the same cells. Bruce and West (1982) found that pectic fragments from the cell walls of castor bean were able to elicit casbene synthase, an enzyme in the biosynthetic pathway of the phytoalexin casbene. Analysis of the crude elicitor suggested that about 70% of the carbohydrate present was D-galacturonic acid or its methyl ester. Casbene synthase was also strongly elicited by fractions rich in D-galacturonic acid residues from other systems, e.g. the endogenous elicitor from soybean cell walls. A pectin-rich fraction of molecular weight 5000 with wound-signal activity isolated from autoclaved tomato leaves (see Section II.C.2) also elicited casbene synthesis in

-

22

S. ALDINGTON AND S. C. FRY

castor beans as well as a different phytoalexin, pisatin, in pea pods (WalkerSimmons et al., 1983). Nothnagel et al. (1983) found that the most active fragment (obtained by partial acid hydrolysis of soybean cell walls) was rich in dodeca-a-(1+4)-~-galacturonide. The optimum dose for elicitation was 1-10 g 1-' (of which 90 pl was applied per cotyledon), which is considerably higher than for the oligo-P-glucans (Section 1I.A). The authors were reluctant to conclude that this dodecasaccharide was itself the active principle, since the possibility could not be excluded that a minor contaminant was active at much lower concentrations. Jin and West (1984) further characterized some castor bean cell wall fragments which elicited casbene synthase activity (assayed in vitro using [3H]geranylgeranyl pyrophosphate as substrate) and showed that oligogalacturonides with d 8 residues were inactive, and that a tridecaa-(1+4)-~-galacturonidewas the most active of the larger oligomers tested. Activity of the fragments was reduced if they were fully methylesterified (thus abolishing their negative charge) and restored if the acidic groups were regenerated by mild acid hydrolysis. Activity was also decreased if the reducing terminus was borohydride-reduced (which also diminishes the negative charge because the product, L-galactonic acid, spontaneously lactonizes-see Aldington et al., 1991). Not all polyanions were active: inactive ones included chondroitin sulphate, hyaluronic acid, polyglutamic acid and heparin. The size of the oligomer, its polyanionic character and the specific molecular shape of oligogalacturonides are thus among the important structural features necessary for activity (West et al., 1984). Davis et al. (1984, 1986a,c) detected an elicitor of the phytoalexin pterocarpan from Erwinia carotovora culture filtrate. Elicitor activity co-purified with a pectate lyase. This enzyme solubilized oligogalacturonides from the soybean cell wall, citrus pectin and polygalacturonic acid. Of the oligomers released from polygalacturonic acid by this lyase the most active elicitor was a decasaccharide of a-(1+4)-~-galacturonic acid residues with a 4,5unsaturated non-reducing terminal residue (A4s5-GalA) (formed by the action of the lyase). The corresponding undecamer showed minor elicitor activity. When either oligogalacturonides solubilized from parsley cells by pectate lyase or else the purified enzyme from E. carotovora was added to cell suspension cultures of parsley, the phytoalexin, coumarin, accumulated and there was also an increase in P-(1+3)-~-glucanase activity. Again, oligogalacturonides were implicated as regulators of the plant defence response (Davis and Hahlbrock, 1987). Kurosaki et al. (1985a,b) provided evidence that oligogalacturonides might not be the whole answer. Kurosaki et al. (1985a) induced phytoalexin production by addition of partial pectinase digests of carrot cells to suspension-cultured carrot cells. The hydrolysates from 7-day-old cells had

OLIGOSACCHARINS

23

the most potent elicitor activity. At this stage, the pectic polysaccharides in the cells are highly esterified, whereas older cultures have mostly nonesterified uronic acids (Asazimu et a f . , 1984). This was interpreted to suggest, in contrast to the results of Nothnagel et al. (1983) and Jin and West (1984), that maximal elicitor activity was dependent on at least partial esterification. However, more direct evidence would be required to establish this. Kurosaki et al. (1985b), using either pectinase or trypsin, released elicitor-active fragments from cultured carrot cells. Fragments obtained by pectinase digestion were essentially oligogalacturonides; their activity was destroyed by further pectinase treatment. Those obtained by tryptic digestion included two types of elicitor: one lost activity after treatment with Pronase E (another protease) and may thus have contained an important oligopeptide moiety, and the other lost activity after treatment with pectinase and may thus have been pectic material held within the cell wall by the presence of structural protein. This suggested that phytoalexin synthesis was elicited by a heterogeneous mixture of fragments containing oligogalacturonide and peptide moieties; the possibility of synergism is developed in Section II.G.3. Further evidence has been presented for a role of oligopeptide material. Dixon et a f . (1989) used fractionated material obtained from autoclaved bean hypocotyls and material hydrolysed from bean cell walls to elicit phenylalanine ammonia lyase (PAL) and phytoalexin accumulation. The elicitor-active components from the hypocotyls were found in neutral and uronic acid-containing fractions; treatment with proteinase K increased the activity. Pectic fragments, especially the nonasaccharide, prepared by treatment of pure polygalacturonic acid with the commercial enzyme mixture “Pectolyase”, also had elicitor activity, but were less potent elicitors than those from the plant. This again suggested that maximal elicitor activity resides in a complex mixture of pectic and other components. In conclusion, the most effective endogenous elicitor-active oligosaccharins are oligogalacturonides of a certain size (DP between 9 and 13) which can be prepared from a variety of sources by different methods (Collmer and Keen, 1986). However, the situation may not be as straightforward as these results suggest; other components may well be influential. It also seems possible that the precise structural requirements for elicitor activity, and therefore also the effects of chemical modification on pectic fragments, may vary according to the responding plant species. 4. Possible evocation of the hypersensitive response by pectic (and other?) oligosaccharins The rapid death of host cells in close proximity to a site of infection is a widely observed phenomenon-the hypersensitive response-generally associated with active plant defence. This response, which is distinct from phytoalexin synthesis, is evoked by many non-pathogenic microorganisms

24

S. ALDINGTON AND S. C . FRY

and somehow prevents colonization by the microorganism (Maclean et al., 1974; Keen, 1982). It seems likely that the plant cells are induced to commit suicide in this way by recognizing some chemical component of the potential parasite; the hypersensitive response is not usually evoked by mechanical injury alone. Furthermore, the hypersensitive response is not simple death, but requires protein metabolism and can be inhibited by Blasticidin S (Furuichi et al., 1979) or cycloheximide (Masuta et al., 1991). Extracellular components recovered from culture filtrates of Phytophthora infestans triggered the hypersensitive response in potato (Keenan et al., 1985). Specific microbial components implicated in this recognition process include: (1) a combination of factors derived from cell walls of Phytophthoru infestans (Kurantz and Zacharius, 1981); (2) lipids of P. infestuns (Kurantz and Zacharius, 1981); (3) glycoproteins from Cladosporium fulvum (de Wit and Kodde, 1981) and Colletotrichum lindemuthianum (Tepper and Anderson, 1986); (4) oligopeptides coded for by avirulence genes of Cladosporiumfulvum (Schottens-Toma and de Wit, 1988). In addition, considerable evidence has indirectly suggested a role for host cell wall fragments in evoking the hypersensitive response. In particular, several lines of evidence suggest that pectic oligosaccharides, possibly liberated from the host’s cell wall by the action of microbial enzymes, are important. At least two of these pieces of evidence predate the oligosaccharin concept: (1) the production of polymethylgalacturonase activity by various races of Xanthomonas malevacearumappeared to correlate with their tendency to trigger hypersensitive necrosis in cotton plants (Hopper et al., 1975); (2) purified pectate lyase from Erwinia rubrifaciens killed a cell suspension culture of tobacco; enzyme concentrations as low as 10 mg 1-1 were sufficient (Gardner and Kado, 1976). Fushtey (1957) first suggested that enzymic degradation of pectic polysaccharides of the plant might result in the release of phytotoxic materials which in turn were responsible for cell death. Mussel1and Strand (1976) suggested that the mechanism of killing cells is probably characteristic of the role of pectic enzymes in many necrotrophic diseases. This mechanism is characterized by the rapid effects of enzymes, efflux of water and salts and high concentrations of pectic enzymes at the site of infection. However, neither these authors nor Basham and Bateman (1975a,b) could provide evidence for the participation of materials released from the host cell walls by the pectic enzymes. More recently, work has been conducted with the oligosaccharin concept as the working hypothesis, and more direct evidence has been presented

OLIGOSACCHARINS

25

which suggests an involvement of plant cell wall fragments. Yamazaki et al. (1983) prepared acid-solubilized wall fragments of sycamore cells. These fragments were added at 50-400 mg 1-’ to suspension-cultured sycamore cells and resulted in the death of a significant proportion of these cells. Fragments generated from starch and citrus pectin by similar acid treatments did not have this “killing” effect. The mixture of fragments produced by partial acid hydrolysis of sycamore cell walls also strongly inhibited [‘4C]leucine incorporation into protein, an indicator of decreased cell vitality, in cultured sycamore cells. The active fragments could be largely inactivated, and/or degraded to products small enough to pass through 1000-molecular weight cut-off dialysis tubing, by heating at 121°C in 2 M trifluoroacetic acid for 2 h-conditions which would hydrolyse glycosidic linkages-suggesting that the active fragments were oligosaccharides. It was suggested that the activity of the fragments was related to the phytotoxic effect of pectin-degrading enzymes (Basham and Bateman, 1975a; Bateman, 1976; Hislop et al., 1979; Marinelli et d., 1991) in that such enzymes would liberate toxic oligosaccharides from the cell wall. In subsequent work designed to gain information on the chemical nature of the active principle, Fry et al. (1983) acid-hydrolysed sycamore cell walls and fractionated the fragments by gel-permeation chromatography on BioGel P-4 into four peaks corresponding to oligosaccharides of D P approximately 1-3, 4-10, 10-25 and 2 2 5 . Fragments of surprisingly low apparent molecular weight had considerable killing activity; in particular, an active fragment co-chromatographed with a marker trisaccharide. This lowmolecular weight material was not a simple oligogalacturonide but did appear to originate from the sycamore cell wall pectins. Biological activity may have been due to a non-uronic acid substitutent of the pectic oligosaccharide. This suggested that pectic fragments mediated the killing action of pectic enzymes. Cervone et al. (1987a) purified a pectinase from a commercial preparation from Aspergillus niger. The homogeneous enzyme was able to elicit a necrotic response in cowpea pods (Vigna); the catalytic activity of the enzymes was found to be necessary for elicitor function. A mixture of oligo-a-( 1+4)-~-galacturonideswith D P 2 4 produced by partial pectinase digestion of polygalacturonic acid evoked a necrotic response in Vigna which was indistinguishable from that evoked by native pectinase. Oligosaccharides released from Vigna hypocotyl cell walls by partial digestion with pectinase also triggered the necrotic response. In these three cases the responses observed were very similar to necrosis incuded in Vigna pods by inoculation with A . niger mycelium. D-Galacturonic acid and its a-(l-4)linked disaccharide did not elicit a necrotic response. Non-pectic fragments may also contribute. Doares et al. (1989) and Bucheli et al. (1990) detected heat-labile (presumably enzymic) material, secreted by the rice blast pathogen Magnaporthe grisea, which could kill

-

26

S. ALDINGTON AND S. C. FRY

cultured maize cells. The putative enzymes were able to solubilize, from maize cell walls, heat-stable factors (putative oligosaccharins) which had a similar phytotoxic effect. Several putative enzymes were involved which synergisticallywere very effective at killing cells. However, no combination of three purified enzymes (pectin lyase, pectinmethylesterase and xylanase) from culture filtrates of the fungus was found which could kill plant cells or release toxic fragments from the plant cell wall (Bucheli et a f . ,1990). It has, nevertheless, been suggested that M . grisea secretes a xylanase and an arabinosidase which can act in concert to release feruloylated oligoarabinoxylans that act as killing factors (Kauffman et a f . , 1990). If feruloyl groups are important in the biological activity of an oligosaccharin, then great care is needed to avoid their alkali-catalysed hydrolysis. The possible role of feruloyl groups is further discussed in Section 1I.D. Recent work by the present authors has shown that several other plant and fungal polysaccharides can be converted, by partial hydrolysis with acid or “Driselase”, into fragments that inhibit the incorporation of [14C]leucine into protein by cultured tomato cells. Sycamore rhamnogalacturonan-11, without any further treatment, also had this effect when added to tomato cells. The structure of RG-I1 is exceedingly complex (Table II), a fact for which no explanation can currently be given in terms of cell wall architecture or extensibility; an informational or signalling role has thus always seemed a plausible reason for the complexity of RG-11. Hahne and Lorz (1988) have also presented evidence that (unidentified) fragments released from plant cell walls during the enzymic isolation of protoplasts can be toxic to the protoplasts. Pectic oligosaccharides can also suppress the hypersensitive response. Baker et af. (1986) found that tobacco leaves infiltrated with small amounts of the same pectate lyase as used by Atkinson et al. (1986) (see Section 1II.B.1) did not then respond hypersensitively when inoculated with incompatible bacteria (i.e. bacteria that, under normal circumstances, trigger an effective defence response and are therefore not successful pathogens). In enzyme-treated tissue, bacterial multiplication was reduced and symptoms caused by both compatible and incompatible bacteria were suppressed, although the bacteria were not killed. The effect of pectate lyase was localized to the infiltrated tissue (cf. KuL, 1982) and was effective against challenge inoculations for up to 4 days. Baker et al. (1985) had previously found that heat-stable enzymic reaction products also prevented the hypersensitive response when tobacco was infiltrated 3-24 h prior to bacterial inoculation. This suggests that pectic enzymes or the oligosaccharides which they generate may, if present prior to infection, help to “immunize” the plant. As with the oligosaccharins that elicit the synthesis of phytoalexins, it seems likely that different plant tissues have different structural requirements for fragments to activate the hypersensitive response.

OLIGOSACCHARINS

27

Morphogenesis-regulating activity of pectic oligosaccharides Soluble substances prepared both by pectinase-catalysed digestion and by KOH extraction of the walls of cultured sycamore cells appeared to modulate, at l-lOmgI-', the course of development in thin cell layers of tobacco (Tran Thanh Van et a l . , 1985). Tran Thanh Van and Mutaftschiev (1990) later claimed to detect effects of the pectic fragments at concentrations as low as 0.1 p,g I-'. The ratio of explants that rooted, produced floral initials, or established calli was shifted in response to a delicate balance of factors in the medium, including pH, hormone concentrations and the amount and type of cell wall components added. However, both the pectinase- and KOH-solubilized samples were predominantly of high molecular weight: the former had been exhaustively dialysed (presumably in standard 12 kDa cut-off dialysis tubing) and would therefore have been mainly rhamnogalacturonans (Table 11), and the latter were standard (deacylated but otherwise relatively intact) hemicellulose preparations (for a review of methods used for extraction of hemicelluloses, see Edelmann and Fry, 1992a). Therefore, the substances used cannot be described as "oligosaccharins". Eberhard et al. (1989) and Mohnen et al. (1990) re-examined this work by studying in depth the response of tobacco explants to oligo-a(1+4)-~-galacturonides.The presence of oligogalacturonides, of the right D P and in the right quantities, altered the concentrations of auxin and cytokinin required for the promotion of particular developmental pathways. A t 0.5 FM,the most effective oligogalacturonides were of D P 12-14; those of D P 6-11 were only effective at substantially higher concentrations. Filippini et al. (1992) have shown that oligogalacturonides induce several disparate responses that could all broadly be described as antagonistic to auxin action. Thus, an oligogalacturonide preparation of D P 10-17, but not a DP 1-4 preparation: 5.

-

(1) inhibited auxin-induced elongation in pea stem segments at 75-150mgl-' (cf. Section 1I.B); (2) inhibited auxin-induced rooting in tobacco leaf explants at 1-10mg1-'; (3) inhibited the auxin-induced production of membrane-associated auxin-binding proteins at 10-100 mg I-'; and (4) inhibited somatic embryogenesis in carrot cell cultures at 100 mgl-'. The embryogenesis may perhaps be regarded as auxin-induced because, although the medium in which the embryos develop is low in auxin, the competence to undergo embryogenesis is conferred by a brief pre-treatment in high-auxin medium (de Vries et al. , 1988). A general ability to block the action of auxin would mean that many of the numerous effects of auxins on morphogenesis might be expected t o be modulated by added oligogalacturonides. In this context, both the reported stimulation by oligogalacturonides of the enzymic oxidation of indoleacetic

28

S. ALDINGTON AND S. C. FRY

acid (Section III.C.2) and the inhibition by oligogalacturonides of the binding of indoleacetic acid to a membrane protein (Section III.A.2) deserve careful attention.

6. Oligosaccharin-induced changes in cell wall composition Changes are known to occur within the host’s cell wall in response to infection. This is not surprising because for many pathogens penetration of the wall is important for a successful infection; furthermore, many pathogens produce cell wall degrading enzymes (see Section V.B.2). Therefore, it would be expected that the host should have evolved certain defence mechanisms to reinforce the cell wall structure in response to infection. Changes such as lignification, accumulation of hydroxyproline-rich glycoproteins and synthesis of callose occur during infection, and can also be induced by chemical factors originating from host and potential pathogen. Here, then, is another possible role for oligosaccharins. Lignijication has been implicated as a defence response in many plants (Vance et al., 1980; Glazener, 1982; KuC, 1982), with lignin being rapidly deposited in the walls and cytoplasm of host cells during infection. Lignin is an aromatic polymer formed by the oxidative coupling of p-coumaryl, coniferyl and sinapyl alcohols, generally assumed to be catalysed by peroxidase (Sarkanen and Ludwig, 1971). It is hydrophobic, insoluble and extremely resistant to enzymic digestion. Incrustation with lignin can protect cell wall polysaccharides from enzymic digestion. Microbial penetration is impeded by lignification owing to the resistance of lignin to microbial enzymes and also to the toxicity of low molecular weight precursors of lignin. The fungal wall itself may become “lignified” by the action of fungal oxidases on the hydroxycinnamyl alcohols, further restricting fungal growth. Considerable evidence has shown that fungal wall fragments can induce lignification; chitin-derived oligomers are especially potent (Ride and Barber, 1987; Section 1I.E). Robertsen (1986) has also demonstrated that endogenous elicitors may be involved in the production of a lignin-like material in cucumber seedlings resistant to Cladosporium cucumerinum; lignin was not produced by susceptible plants on fungal penetration. Lignification was induced in vivo by oligogalacturonides from cucumber cell walls, by chitosan and by polygalacturonic acid. Active oligogalacturonides had a DP between 8 and 14; the most active fragments were the decamer and undecamer, which were effective down to a concentration of 0.5 p ~ . Suspension cultures of castor bean treated with pectic fragments have also been shown to accumulate lignin (Bruce and West, 1989); the most effective preparation was a mixture of oligogalacturonides with an average chain length of 7. Furthermore, substantial changes in peroxidase isozyme pattern occurred following treatment with these oligosaccharides. During infection, pectic fragments are probably released from the host’s

OLIGOSACCHARINS

29

cell wall by pectic enzymes produced by the pathogen. Enzymes produced in culture filtrates of C. cucumerinum grown on pectin were purified (Robertsen, 1987). One such enzyme, a pectinase (as defined earlier), induced lignification in cucumber hypocotyl segments and released active fragments from cucumber cell walls and polygalacturonic acid. However, since neither the enzymes nor the oligogalacturonides induced lignification specifically in resistant hypocotyl segments, other factors must determine specificity. Interestingly, Robertsen and Svalheim (1990) found that the lignin induced by oligogalacturonides in the cucumber hypocotyl is formed mainly from p-coumaryl alcohol, which usually constitutes only a minor component of dicotyledonous lignin (1-5%). They also showed that the induction of lignification was blocked by inhibitors of phenylalanine ammonia lyase or peroxidase (often regarded as the “first” and “last” enzymes in the lignin biosynthetic pathway) ; this suggests that the oligosaccharins cause an increased total flux of carbon skeletons into phenolic metabolism and not merely a diversion from hydroxycinnamates (e.g. ferulate) to lignin. Hammerschmidt et al. (1985) had previously observed an increase in p coumaryl rich lignin when cucurbits were inoculated with non-pathogens. They suggested that this type of lignin was advantageous in that its lack of methoxy groups could allow faster synthesis requiring less energy, and in that it may become more heavily cross-linked than the other lignin types. Hydroxyproline-rich glycoproteins (HRGPs) are another group of cell wall polymers which are difficult to degrade, and hence form an effective defence barrier. The best known HRGPs are the extensins, a family of highly basic glycoproteins with tetraarabinoside side-chains (Lamport, 1980) and cross-linked by isodityrosine bridges (Fry, 1982). The extensin of cultured cells was resistant to trypsin unless the tetraarabinoside side-chains were first removed by mild acid hydrolysis (Lamport, 1977). Another possible defensive role is to act as agglutinins of pathogens (Leach et al., 1982). Hydroxyproline-rich glycoproteins were first found to increase on infection of melon seedlings with Colletotrichum lagenarium (EsquerrC-TugayC and Mazau, 1974). The accumulation of HRGPs, which can be induced by oligosaccharins from both the host and the fungal pathogen, showed some race-cultivar specificity since it occurred in resistant cultivars earlier and to a greater extent than in susceptible cultivars; in incompatible interactions of bean and Colletotrichum lindemuthianum there was an early increase in HRGP mRNAs being detected 4 h after oligosaccharin treatment (Showalter etal., 1985). Induction of HRGP tends to be a slower but more prolonged response to oligosaccharins than phytoalexin elicitation. Hydroxyproline-rich glycoprotein mRNAs are expressed systemically as well as locally with mRNAs being detected in uninfected cells distant from the infection site (Showalter et al., 1985), and found in roots which are not directly affected by the disease (Rumeau et al., 1988).

30

S . ALDINGTON AND S. C. FRY

7. Induction of ethylene synthesis by pectic oligosaccharides The increased production of ethylene is one of the earliest chemically detectable events in pathogen-infected plants (Ecker and Davis, 1987) and is usually followed by activation of other defence responses such as the synthesis of phytoalexins (Chappell et al., 1984), HRGPs (Roby et al., 1985) and chitinases (Boller et al., 1983; Roby et al., 1986). Both fungal and plant cell wall fragments have been implicated in the induction of ethylene synthesis. Initial evidence for the involvement of plant cell wall fragments came from the work of Anderson et al. (1982). They applied the cell wall-degrading enzyme mixture “Cellulysin” (which contains enzymes that hydrolyse pectins, hemicelluloses and cellulose) to tobacco leaf disks and found that ethylene production was induced rapidly, increasing 17-fold in 3 h. The acid (ACC), a precursor of synthesis of 1-aminocyclopropane-1-carboxylic ethylene, also increased considerably. Chemical inhibition of ACC synthase resulted in the inhibition of ethylene production and of ACC accumulation. These results suggested that either the cell wall-degrading enzymes or their products could stimulate ethylene biosynthesis. Much evidence now implicates pectic oligosaccharides. For example, Labavitch (1983) noted increases in ethylene production when partially digested preparations of citrus pectin were injected into oranges, thus suggesting that pectic fragments may be oligosaccharins that induce ethylene synthesis. However no further information on structure of active fragments was given. The endogenous components from melon leaf cell walls which triggered HRGP biosynthesis in melons (Roby et al., 1985) were also found to increase ethylene production. Although the active components were not characterized, they had been obtained by autoclaving the cell walls and would thus have included pectic fragments; they were more effective than fungal wall fragments. There may have been a causal relationship between HRGP and ethylene production as suggested by Roby et al. (1985). An inhibitor of ethylene synthesis also inhibited HRGP accumulation, and ethylene was found to increase the level of HRGP mRNA. The possible role of membrane depolarization in initiating both these effects is discussed later (Section III.B.l). Macerase (a pectinase-containing enzyme mixture) was used to digest isolated cell walls from cultured pear cells, producing pectic fragments (Tong et al., 1986). Both Macerase and the soluble products of Macerasedigested cell walls induced ethylene production in pear cells; an increase in ethylene production was observed within 1 h after treatment and peaked after 2 h. Production rates were higher in response to Macerase products than in response to the enzymes themselves. Macerase-digested pectin from pear fruit and tobacco leaf cell walls also increased the rate of ethylene production by pear cells (Campbell and Labavitch, 1991). Further evidence that the active fragments capable of inducing ethylene

OLIGOSACCHARINS

31

synthesis were of pectic origin came from Baldwin and Biggs (1988). Pectolyase (which contains fungal pectinase, endo-pectin lyase and xylanase) was injected into orange fruit. Ethylene production was stimulated and levels of ACC also increased. Tomato pectinase had a similar effect (Baldwin and Pressey, 1988). Ethylene production was found to be due to enzyme products rather than to the enzyme mixture itself, as partially acid hydrolysed pectic material also induced a dramatic increase in ethylene 24 h after injection into the fruit. Following gel-permeation chromatography, certain fractions (although size was not specified) were enriched in ethyleneinducing activity. Too much or too little hydrolysis of the substrate resulted in reduced ethylene-inducing activity of the products. This suggests that pectic oligosaccharides need to be of a specific size for maximum activity. These results imply that pectic fragments are involved in the induction of ethylene in host-pathogen interactions. Similarly, pectic fragments released from the cell wall during fruit ripening are also associated with ethylene production. However, it is surprising that an a-D-galacturonidase (exopolygalacturonase) also induces ethylene formation (Baldwin and Pressey, 1990), even though the product of this enzyme, the monosaccharide, does not. It has also been reported that yeast polysaccharides [prepared by the method of Hahn and Albersheim (1978) and thus containing P-glucans but apparently not freed of a-mannans] will induce ACC synthase activity in plant tissue cultures within 10min (Piatti et al., 1991). Thus, several oligosaccharins, as well as wounding (Boller et al., 1983), can induce the synthesis of ethylene and of various “defensive” proteins. Treatment of plant tissues with exogenous ethylene can also often induce the synthesis of “defensive” proteins. For example, Ecker and Davis (1987) reported an ethylene-induced increase in mRNA for phenylalanine ammonia lyase (PAL), chalcone synthase, HRGPs etc., and Vogeli e f al. (1988) found that mRNA for chitinase and P-glucanase was induced. When wounded bean leaf tissue was fed aminoethoxyvinyl-glycine (AVG, an inhibitor of ethylene synthesis), little ethylene was formed in response to wounding and the wound-induced increase in chitinase activity was abolished (Boller et al., 1983). It can therefore be suggested that ethylene acts as a second messenger for wounding and perhaps also for oligosaccharins. However, Chappell et al. (1984) found that exogenous ethylene did not induce PAL, although fungal elicitors induced both ethylene (within 1h) and, later, PAL. Pre-treatment of the tissue for 1h with AVG “completely” inhibited elicitor-induced ethylene biosynthesis, though PAL induction was only 50% inhibited. This suggests that PAL was induced by the elicitor independently of ethylene. Similarly, in cultured cells of Eschscholtzia, sanguinarine synthesis was elicited by yeast polysaccharides in the presence of 0.25 mM AVG, which “completely” blocked ethylene synthesis; it was also shown that the yeast polysaccharide was still able to elicit ethylene synthesis in the presence of saturating levels of exogenous ethylene (Piatti et

32

S. ALDINGTON AND S. C. FRY

al. ,1991). Mauch et al. (1984) found that high concentrations of indoleacetic acid stimulated ethylene synthesis; chitinase and p-glucanase were also induced and this enzyme induction could be prevented by AVG. Fungal elicitors resulted in additional ethylene production; AVG partially blocked this but had only marginal effects on enzyme induction. This again suggests that ethylene and elicitors are separate stimuli for enzyme induction. Paradies et al. (1980) also showed that phytoalexin synthesis was independent of ethylene induction. Nevertheless, Hughes and Dickerson (1991) found that induction of chitinase and p-glucanase by elicitors could be abolished by AVG (opposing the results of Mauch et al. , 1984); AVG had no effect on enzyme induction by exogenous ethylene. Also, Hughes and Dickerson (1989) found that, in a bean cultivar resistant to Colletotrichumlindemuthianum, PAL could not be induced by exogenously supplied ethylene in the absence of elicitors but only in the presence of a sub-optimal concentration of elicitor. Ethylene had to be introduced to the cultivar within 2 h of elicitor application. AVG partially inhibited this induction, but again only when added within 2 h of the elicitor, and the effect of AVG could be reversed by the addition of ACC or ethylene. A susceptible cultivar did not respond to exogenous ethylene. It was suggested that ethylene may function by activating the elicited response when elicitor levels are sub-optimal.

D . OLIGO-P-XYLANS AS POSSIBLE OLIGOSACCHARINS

The discussion of ethylene induction leads us on to the possible regulatory role of p-( 1+4)-~-xylanasesand the oligosaccharides which they can produce. Xylans (the term is here taken to include arabinoxylans, glucuronoxylans, glucuronoarabinoxylans, etc.) are present in the primary and secondary cell walls of all higher plants that have been examined. They make up a higher proportion of the primary walls of the Gramineae than of Dicotyledons. They have a linear backbone of p-(1+4)-linked D-xylose residues, some of which are substituted with a-L-arabinose and a-D-glucuronic acid residues. Many of the glucuronic acid residues are often in the form of the 4-O-methyl ether, some of the xylose and arabinose residues may be acetylated, and, at least in the Gramineae, a few of the arabinose residues carry ester-linked ferulic acid groups. An extracellular protein from Phytophthora parasitica, possibly a xylanase, induced stress metabolites in tobacco callus (Farmer and Helgeson, 1987). Dean et al. (1989) and Fuchs et al. (1989) showed that a particular xylanase (ethylene-inducing xylanase, EIX) present in “Cellulysin” and in certain plant pathogens is able to induce ethylene biosynthesis in plant tissues. The obvious suggestion from this observation is that oligo-p-xylans, generated by the action of EIX, are oligosaccharins. However, the ability to

OLIGOSACCHARINS

33

induce ethylene biosynthesis is not a trait common to all xylanases. Thus, it seems possible that a specific repeating unit of xylan-perhaps akin to the XG9 unit of xyloglucan (Section 1I.B) or the key hexasaccharide unit of the oligo-P-glucans (Section 1I.A)-may be an essential component of active oligosaccharins. Certain xylanases are indeed remarkably specific for the site at which they attack their substrate (Nishitani and Nevins, 1991). Dean et al. (1991) showed that the major product of the action of EIX on pure commercial oat and birch xylans was the tetrasaccharide, 6-(1+4)-~-xylotetraose;some larger oligo-P-xylans bearing arabinose and xylose side-chains were also formed. EIX did not solubilize any detectable carbohydrate from tobacco leaf cell walls, leading the authors to speculate that the enzyme itself, rather than any oligosaccharide products, was directly responsible for inducing ethylene synthesis. However, it seems possible that small amounts of oligosaccharides were solubilized but not detected (limits of detection were not stated); it is also possible that the EIX would digest wall-bound xylans more effectively in the presence of other enzymes, e.g. pectinases, which are normally present in infected tissue, and which might open up the structure of the cell wall to expose the xylans. Ishii and Saka (1991) obtained possible evidence for an oligosaccharin effect in a xylan-derived fragment. They showed that the feruloylated trisaccharide. Ferulate

I

5 ~-Arabinose-a-( 1+3)-~-Xylose-P-(1+4)-~-Xylose, or FAXX, released from bamboo shoot cell walls by enzymic hydrolysis, inhibited the auxin-stimulated bending of rice lamina joints; the corresponding non-feruloylated ologosaccharide (AXX) lacked this effect. As emphasized in Section I, phenolics (including ferulic acid) are well known to exert numerous effects on plant growth and development, and it has not been established that the effect of FAXX is not a simple consequence of its ferulate group, which would perhaps be mimicked by pure ferulic acid or methyl ferulate; however, if the AXX moiety of the FAXX molecule is required for activity, this will be an interesting novel oligosaccharin. A xylanase was also shown to evoke the synthesis of pathogenesis-related proteins in tobacco (Lotan and Fluhr, 1990); this was demonstrated to be independent of its ability to induce ethylene synthesis. Ishii (1988) had also shown that a xylanase purified from Trichoderma viride could kill suspension-cultured rice cells-perhaps by solubilizing xylan-based oligosaccharins. In contrast, no toxic effect was seen in maize cell cultures treated with a xylanase (alone) from Magnaporthe grisea (Bucheli et al., 1990; see Section II.C.4).

34

S. ALDINGTON AND S. C. FRY

The role of xylan-derived oligosaccharins would appear to be an exciting new field in which significant advances can be expected in the next few years. E. CHITO-OLIGOSACCHARIDES AND RELATED FRAGMENTS

Chitin, a component of the fungal cell wall, is essentially a neutral homopolymer of (3-( 1-+4)-linked N-acetyl D-glucosamine (GlcNAc) residues (Fig. 3); however, there may be stretches of non-N-acetylated residues within the overall structure. Chitosan, essentially a positively-charged homopolymer of (3-(1+4)-linked D-glucosamine (GlcN) residues (Fig. 3), is

H

NH

I R

H

NH

I R

H

NH

I

R

Fig. 3. The structure of a short segment of a chitin (R = acetyl) or chitosan chain (R = H).

formed from chitin as a post-synthetic modification by the action of an acetamidase, and tends to be found in the walls of the lower fungi such as the Zygomycetes (where their cationic nature is balanced by anionic polymers). Chitosans may retain a low proportion of N-acetylated residues (Wessels, 1986). Oligosaccharides of both chitin and chitosan have been demonstrated to have numerous physiological effects on suspension-cultured cells and plant tissues. 1. Oligosaccharides of chitin Perhaps the most widely known effect of chitin fragments is the induction of lignification. This response occurs in wheat leaves when inoculated with non-pathogens: rapid lignification ensues at the wound margins, suggesting protection against spread by the potential pathogen (Ride, 1975; Bird and Ride, 1981; Beardmore et al., 1983; Moerschbacher et al., 1986; Ride and Barber, 1987). There appears to be no racekultivar specificity in the induction of the defence response (Callow, 1984; Deverall and Deakin, 1987; Moerschbacher et al., 1989). The size of the chitin oligomers which had appreciable oligosaccharin activity ranged from the tetrasaccharide through to the hexasaccharide; the trisaccharide was slightly active. Digestion of the tetrasaccharide with chitinase, which catalyses the reaction

OLIGOSACCHARINS

GlcNAc-GlcNAc-GlcNAc-GlcNAc

35

+ H2O + GIcNAc-GIcNAc + GIcNAc-GIcNAc,

abolished its oligosaccharin activity, supporting the view that chitotetraose, despite its structural simplicity, was the active principle. None of the oligomers of chitosan tested (up to the tetrasaccharide) possessed significant ability to induce lignification (Ride and Barber, 1987; Barber et al., 1989). However, polymeric chitosan was active, and lignification was also found to be induced by a number of apparently unrelated materials, including a heat-stable component of a commercial (“Onozuka R-10”) cellulase preparation, wheat germ agglutinin (weakly), eicosapentaenoic acid, HgC12 and CdCI2 (Barber and Ride, 1988). Pachyman [a fungal P-(l--+3)-~-glucan] and the p - ( 1 4 ) , ( 1-+6)-~-glucan from Phytophthora were also active, but the hepta-P-glucoside (compound 3 of Table 11) was not, despite its strong activity as an elicitor of glyceollin in soybean (Section 1I.A). The size of active chitin-derived oligosaccharides seems to vary depending on the response studied. Another response to chitin fragments is the induction of chitinase in melon plants. Here, the most efficient oligosaccharins were found to range from the hexasaccharide to the nonasaccharide. Smaller (DP 1 4 ) and larger oligosaccharides (DP 1 S 2 0 ) had little or no effect. The chitin oligosaccharides, which were injected under the epidermis of a cotyledon, induced chitinase within 6 h , the effect being maximal between 12 and 24 h. Interestingly, the size of the oligomer appeared to govern the type of chitinase activity induced: the hexasaccharide and heptasaccharide induced exo-activity whilst the heptasaccharide through to the nonasaccharide induced endo-activity (Roby et al., 1987). Kurosaki et al. (1988) showed that chitinase-solubilized fragments of the walls of the fungus Chaetomium globosum, and of chitin, induced chitinase activity in carrot cells. Chitinase-inducing activity was distributed in several fractions following separation of oligosaccharides by gel-permeation chromatography. The same oligomers also induced the accumulation of phenolic acids in the cultured carrot cells; however, for this effect the most potent oligosaccharins were observed in high molecular weight fractions.

2. Oligosaccharides of chitosan Cationic oligosaccharides, derived from chitosan, have also been reported to possess several biological activities. For example, they induce chitinases and P-glucanases in pea pods; similar effects were seen in pods infected with both compatible and incompatible fungi (Mauch et al., 1988). Chitosan fragments can both induce plant defence responses and inhibit fungal growth (Hadwiger and Beckman, 1980; Hadwiger and Loschke, 1981; Walker-Simmons et al., 1983). They also induce callose synthesis in higher plant cells (Kauss et a f . , 1989), the DP and the degree of acetylation of the oligosaccharides governing activity. Kendra and Hadwiger (1984) demonstrated that the smallest chitosan oligomer with elicitor activity (for the

36

S. ALDINGTON AND S. C. FRY

formation of the phytoalexin pisatin) and able to inhibit fungal growth was the trisaccharide, although the heptasaccharide had maximum activity. Walker-Simmons et af. (1983) found that not only were chitosan fragments potent elicitors of pisatin, but they also were the most potent known inducers of proteinase inhibitor accumulation in excised tomato cotyledons. The smallest chitosan oligomer with this activity was the pentasaccharide (Ryan, 1987, 1988). Another phytoalexin found to be induced by chitosan fragments was glyceollin in soybean suspension cultures; other effects coinduced in this system were a decrease in the rate of cell growth and an increase in wall-bound phenolics and callose; the host cell walls became more resistant to degradation by the fungal enzyme mixture, Driselase (Kohle el af., 1984). Chitosan has been reported to have disruptive effects on membranes of suspension-cultured soybean cells and (perhaps hence) to activate callose synthesis. Callose is a polysaccharide containing a high proportion of p-( 1+3)-linked D-glucose residues. It is deposited in response to physical and chemical stress. The speed of its synthesis and its localized deposition may well indicate that it is an important defence response. When suspension cultures of soybean were treated with chitosan fragments, callose synthesis began 20 min later. Although excess Ca2+ blocked the response (Young et af., 1982), callose synthesis was dependent on the availability of moderate levels of Ca2+ (Kauss et af., 1983) and was only observed when the membrane was disrupted sufficiently to result in electrolyte leakage (Kohle et al., 1985). Chitosan fragments, especially those of high molecular weight (Young and Kauss, 1983), caused a rapid release of Ca2+ from cells, presumably leading to an alteration in the local Ca2+ concentration in the cell wall and at the plasma membrane. A disruptive effect of chitosan (10 mg 1-') on cultured cells of rice was also reported to be Ca2+ dependent (Masuta et af., 1991). The chitosan caused browning and death in the routine medium, which contained 3.1 mM Ca2+, but not in a medium containing 4.0 mM Ca2+. The effect of the chitosan, which may be a hypersensitive response (cf. Section II.C.4), was blocked by pre-treatment of the cells with cycloheximide, suggesting that the browning and death was an active process, dependent on protein synthesis. Although membrane effects are very obvious and rapid in suspension cultures, Kendra and Hadwiger (1987a) found that, in pea endocarp tissue inoculated with Fusarium sofani, a leakage of electrolytes similar to that induced by chitosan was not detectable until after 24 h, when there was a sharp increase in leakage in the compatible interaction, probably due to extensive fungal growth rather than to any specific effect of chitosan. The induction of a resistance response, apparently by chitosan, occurred within 6 h (as shown by the complete suppression of a non-pathogenic fungus), and the hypersensitive response became apparent several hours later. It seems possible, however, that localized chitosan effects at the initial site of fungal

37

OLIGOSACCHARINS

penetration might come into play much earlier but on an insufficient scale to cause detectable electrolyte leakage. In contrast, in suspension cultures all the cells in the population are simultaneously exposed to the same dose of oligosaccharins and their massed response is therefore much more readily detectable. SOZOH HOCH, ~

o

I

~

o

HO

OH NH

NH

NH

NH

R

COCH,

COCH,

COCH,

I

I

I

I

Fig.4. The structure of an oligosaccharin from Rhizobium meliloti that specifically induces lucerne root hairs to deform in preparation for nodulation. Most hydrogen atoms omitted for clarity. R- represents CH3-(CH2)s-CH=CH-(CH,),-CH=CH-(C=O)-. Adapted from Lerouge et al. (1990).

The symbiotic N2-fixingbacterium Rhizobium meliloti produces an extracellular oligosaccharin (Fig. 4), structurally related to chitotetraose, which at nM concentrations induces the root hairs of its host plant to deform in preparation for colonization by the bacteria (Lerouge et al., 1990). This oligosaccharin is unusual in being sulphated and in that one of the glucosamine residues is N-acylated with a hexadecadienoic acid. It is not yet known whether this oligosaccharin is produced de novo or by the breakdown of a polymer. However, considerable progress is being made in the genetic analysis of this system (Baev et al., 1991; Govers et al., 1991) and exciting advances can be expected.

-

F. OLIGOSACCHARINS FROM N-LINKED GLYCOPROTEINS?

As mentioned in Section II.B, Priem et al. (1990) showed that an oligosaccharide with the structure cx -D-M a n

1

J 6

~-D-M~~-~+~-P-D-M~~-~+~-P-D-G~CNAC-~+~2

3

t

t

1 P-D-XYl

1 CX-L-FUC

38

S. ALDINGTON AND S. C. FRY

was able, at 0.7-7.0 nM, to synergize with auxin, and, at 70 nM, to antagonize the action of auxin on elongation in flax hypocotyls. The structure of this oligosaccharide is similar to what would be obtained upon cleavage of an N-acetylglucosaminyl+L-asparagine bond in many N-linked glycoproteins by an appropriate endoglucosaminidase. In this connection, it is interesting that certain extracellular N-linked glycoproteins, which normally accumulate in the culture medium of carrot cells during somatic embryogenesis, appear to be essential for embryogenesis (de Vries et al., 1988). Thus, treatment of the cultures with tunicamycin, which inhibits the N-glycosylation of proteins, blocked somatic embryogenesis at a very early stage. The blockage could be relieved by addition of purified glycoproteins (especially a 38 kDa isoperoxidase) isolated from the media of untreated cultures (Cordewener et al., 1991). A plausible explanation of these results is that the glycoproteins are either themselves active as signalling molecules, or that they are partially digested apoplastically to liberate their carbohydrate moieties, which then act as oligosaccharins essential for embryogenesis. Other glycoproteins with “hormone-like’’ effects include a fungal glycoprotein that induced lignification in wheat (Kogel et al., 1988, 1991) and soybean lectin (“agglutinin”), which induced cell division in soybean cultures (Howard et al., 1977). The fungal glycoprotein, isolated from the walls of the rust Puccinia graminis, has a molecular weight of 67 000 and seems to depend for its biological effect on the carbohydrate moiety, which consists of 50% mannose, 47% galactose and traces of N-acetylglucosamine (Kogel et al., 1988).

G. CONCLUSIONS

1. Diversity of oligosaccharins Sections L A - F have shown that many diverse carbohydrates may exhibit oligosaccharin activity in one cell-type or another. Active oligosaccharins have been obtained from p-( 1+3),( 1+6)-~-glucans,xyloglucans, homogalacturonan-type pectins, rhamnogalacturonan-I, chitin, chitosan, probably xylans, and possibly N-linked glycoproteins. There are also reports of oligosaccharins from cellulose [p-( 1+4)-~-glucan] (Lorences el al., 1990) 1+3),( 1-+4)-~-glucan] (see Section 1I.A). No oligosacchaand nigeran [a-( rin activity seems to have been reported for fragments of some other cell wall polymers, including 6-( 1-3) ,( 1+4)-~-glucan (a major hemicellulose of the primary cell walls of grasses), callose [p-(1+3)-~-glucan], rhamnogalacturonan-11, arabinogalactans, P-mannans or extensins. Quite possibly, though, some of these polysaccharides and glycoproteins will be found in future experiments to yield active oligosaccharins in response to the right cleavage treatments.

OLIGOSACCHARINS

39

2. Successful host or successful pathogen? Many of the responses of plants to oligosaccharins appear to be defence mechanisms. Are they useful in helping the plant to resist disease? The speed with which the plant is able to respond to a signal is of paramount importance, enabling the defence responses to be activated rapidly. Hence, as plant cells can respond very quickly to oligosaccharins (see Section III.B), the early appearance of oligosaccharins as an alarm signal could be very advantageous to the plant. Usually, responses to oligosaccharins appear just as rapidly and as strongly in susceptible and resistant cultivars (Dixon, 1986). However, care must be taken in assessing specificity-non-specific elicitors were isolated from culture filtrates of Cludosporiumfulvum grown axenically ,but specific elicitors were isolated from apoplastic fluid of tomato leaves infected with compatible races of C. fulvum (de Wit and Spikman, 1982). Halverson and Stacey (1986) suggested that in incompatible interactions, the presence of both fungal and plant cell wall elicitors would provoke a more dramatic defence response. In compatible interactions, such synergism may not occur, resulting in lower levels of phytoalexin, which the pathogen may be able to tolerate or detoxify. Although synergistic effects have been reported (Section II.G.3), there is as yet no information as to whether they enhance defence responses specifically in incompatible interactions. Most fungal pathogens secrete pectic enzymes, and pectic oligosaccharins can probably be solubilized from the primary cell walls of all higher plants. If endogenous elicitors act as alarm signals to activate defence mechanisms, this would mean that all plants have the potential for a resistance response and its elicitation; and most infectious agents have the ability to trigger a resistance response. But susceptibility still occurs (KuC and Rush, 1985). What else could race-cultivar specificity depend on? Robertsen (1987) suggested several possibilities in the infection of cucumber with Cludosporium cucumerinum. As the fungus secretes a pectinase (i.e. an endohydrolase) which can release approximately the same amount of elicitor from the cell walls of both susceptible and resistant hypocotyls, race-cultivar specificity may depend on (1) different ratios of pectinase :galacturonidase and thus on the durability of the active (i.e. moderately large) oligogalacturonides in vivo [excess galacturonidase may destroy elicitor activity]; and (2) the host cell’s ability to lignify [the fungus may be able to inhibit lignification]. The ability of the plant to resist infection may be related to the number of different responses rapidly elicitated by a variety of oligosaccharins which may be present at the interface between the host and potential pathogen. Any one oligosaccharin can elicit more than one response. For example, a phytoalexin (coumarin) and P-glucanase are both induced by Phytophthoru P-glucan in parsley (Davis and Hahlbrock, 1987). In addition, different

40

S. ALDINGTON AND S. C. FRY

oligosaccharins may evoke different defence responses in the same tissue: for example, phytoalexin elicitation may require a different oligosaccharin from that which provokes the hypersensitive response (Albersheim et al., 1986). Therefore, in the incompatible infection, numerous defence responses could be occurring which together are much more deleterious to the invading microorganism.

3. Synergism between oligosaccharins Synergistic effects between different oligosaccharins may well be important in vivo. Darvill and Albersheim (1984) pointed out that, in the elicitation of phytoalexins, it would be advantageous to have two simultaneous signals for maximum sensitivity. This would help the plant to distinguish between purely physical damage and fungal invasion. The hepta-P-glucoside from Phytophthora megasperma cell walls (see Section 1I.A) is, weight-forweight, a much more effective elicitor of phytoalexin accumulation in soybean than the endogenous dodecagalacturonide (see Section II.C.3). However, the application of the two elicitors together produced a synergistic effect on the synthesis of phytoalexins and P-( 1+3)-~-glucanasesin cultured parsley cells (Davis et al., 1986b; Davis and Hahlbrock, 1987): 50-fold less oligo-P-glucan was required to elicit phytoalexin synthesis in the presence of oligogalacturonides than in their absence. Similarly, fatty acids isolated from the cell wall of P. infestans can sometimes elicit phytoalexin synthesis, but elicitor activity is greatly enhanced in the presence of an insoluble plant cell wall material (Kurantz and Zacharius, 1981) or an oligo-P-glucan from the fungal cell wall (Bostock et al., 1982); in this system the oligo-P-glucan alone showed very little activity. Phosphate starvation may also synergize with elicitor effects (Dunlop and Curtis, 1991). And in a further example of synergism, Tepper and Anderson (1990) found that oligosaccharides from citrus pectin and a galactoglucomannan (purified from the incompatible a race of Colletotrichum lindemuthianum), acted together to induce phenolic accumulation. How these synergistic effects are brought about is unknown. There may be two receptor sites, one for each elicitor, which on activation of both cause a cascade effect. Alternatively, one elicitor may disrupt the plasma membrane, allowing the other to penetrate the cell and exert its effects more rapidly. The matter is open to experimentation. One practical consequence of synergism is that it may complicate efforts to purify individual oligosaccharins, since biological activity would be lost with increasing purity. In some cases, the structural requirements for oligosaccharin activity are very stringent (e.g. the hepta-P-glucoside fungal elicitor, and perhaps the xyloglucan-derived XG9 auxin antagonist). At the other extreme, many different pectin-related materials, from the disaccharide (GalA2) to a polysaccharide composed of about 1000 sugar residues (rhamnogalacturonan-I), all possess wound-signal activity, inducing the synthesis of protease

OLIGOSACCHARINS

41

inhibitors in tomato. Intermediate degrees of stringency are seen with the oligosaccharins of pectin (acting other than as a wound signal), chitin and chitosan, and with the xyloglucan oligosaccharins (XG7 to XG9n) that promote cell expansion. These differences in structural requirements lead to speculations about modes of action of the various oligosaccharins. Thus, while a very high degree of stringency suggests interaction with a discerning receptor, such as a membrane protein, the lower stringencies might suggest a more physical effect-such as on the permeability properties of membranes or an effect on the action of an enzyme (Aldington et al., 1991). The following section therefore examines the evidence for various suggested modes of action of oligosaccharins.

111. MODE OF ACTION OF OLIGOSACCHARINS Very little hard evidence is available regarding the mechanism of action of oligosaccharins. Since oligosaccharins are thought of as being apoplastic signals, there has been considerable speculation on the possible role of receptors situated in the plasma membrane. The idea of receptors in the plasma membrane has always been very plausible by analogy with studies on animal cells where plasma membrane-located receptors have been characterized and purified for a number of oligopeptide hormones; the corresponding genes have been isolated and sequenced. Being able to do this with animal cells has proved valuable in elucidating the signal transduction pathways. Recently, progress has been made in the study of receptors for elicitors of fungal origin. Although no receptors for elicitors of plant origin have been clearly demonstrated, there is some circumstantial evidence that they may exist. Primary effects of some oligosaccharins are also becoming better understood following observations that they cause membrane disruption as indicated by “leaky” cells, membrane depolarization and lipid peroxidation. These disruptive effects, which occur very quickly, may help to explain how some oligosaccharins are able to induce very rapid changes in mRNA synthesis. Finally, some oligosaccharides act as substrates and/or effectors of particular enzymes, and the possible significance of this for the mode of action of oligosaccharins will be discussed. A.

EVIDENCE FOR RECEPTORS

1. Receptors for oligo-P-glucans As elicitors of fungal origin were the first type of oligosaccharin to be identified, it is perhaps fitting that binding sites for these elicitors have been the first to be detected. The observation that isolated protoplasts of higher

42

S. ALDINGTON AND S. C. FRY

plants can be agglutinated by treatment with fungal P-( 1+3),( 1 + 4 ) - ~ glucan was the first experimental evidence for the existence of membranelocalized binding sites for this carbohydrate (Peters et al., 1978). The earliest report of a binding site for radioactive oligo-P-glucans came from Yoshikawa et al. (1983) using a membrane fraction from soybean cotyledons. They showed that these membranes were able to bind mycolaminarin, a P-(1+3)-linked D-glucan with a small number of 6-linked branches, which has quite a low phytoalexin-elicitor activity when compared with other fungal P-glucans. Schmidt and Ebel (1987) detected, in plant membranes, high-affinity binding sites for a 3H-labelled oligo-p-glucan mixture isolated from Phytophthora megasperma cell walls, a much more effective elicitor of phytoalexin synthesis than mycolaminarin. Specific binding sites were found in membrane fractions from soybean roots, hypocotyls, cotyledons and cell cultures (Ebel and Grisebach, 1988). There was a strong correlation between the ability of non-radioactive ligands to compete for these binding sites and their elicitor activity. When Cosio et al. (1988) found that blocking the reducing terminus of partially chacterized elicitor-active oligo-P-glucans did not alter their binding affinity, these authors were able to synthesize an oligo-pglucosyl-tyramine conjugate suitable for radio-iodination to a very high specific radioactivity. This conjugate increased the sensitivity of binding assays whilst behaving similarly to the ~ligo-P-[~H]glucan.P-Glucanbinding activity in microsomal membranes of soybean was later also assayed by use of a similar radio-iodinated oligo-~-glucosyl-[2-(4-aminophenyl)ethylamine] conjugate. The hepta-P-glucoside (compound 3 of Table I) exhibited tighter binding and higher phytoalexin elicitor activity than mixed oligo-P-glucans released by acid hydrolysis from the mycelial cell wall (Cosio et al., 1990a). Cosio et al. (1990b) succeeded in solubilizing the binding site (as a large, protein-detergent complex) and suggested that binding of fungal oligo-P-glucans depended on a common structural determinant. Cheong and Hahn (1991) also labelled oligo-P-glucosyl-tyramine conjugates by '251-iodination and demonstrated binding to a soybean membrane preparation. Binding constants agreed well with elicitor activity (Table I). The Kdwas 0.66-0.82 nM and the specific binding capacity of the membranes was about 1.2pmol oligosaccharide per mg protein. The in vitro binding h 15min; the binding was reversible and the tlh for disoccurred with t ~ = sociation was 3 h. Binding was prevented by heating and by protease pre-treatment of the membranes. The elicitor-binding site exhibited a high degree of specificity with respect to the oligosaccharides that it binds, and four oligo-P-glucans, ranging in size from a hexasaccharide to a decasaccharide, which all induced phytoalexin accumulation similarly, were equally effective competitive inhibitors of the binding of the hepta-P-glucoside conjugate (Table I).

-

OLIGOSACCHARINS

43

The work in the laboratories of Ebel and Hahn has thus clearly implicated membrane-localized binding sites as receptors for oligo-P-glucan elicitors. However, no evidence has yet been presented, e.g. by showing that chemical or genetic interference with the binding abolished the responsiveness of cells to the elicitors, which would prove that the binding sites are true receptors.

2. Receptors for other oligosaccharins? Many, perhaps all, plants contain lectins-proteins and glycoproteins that recognize and bind non-covalently to specified carbohydrate groups in glycoproteins, oligo- and polysaccharides, glycolipids, etc. As a relic from the time when lectins (“phytohaemagglutinins”) were mainly studied for their effect on blood cells, the term “lectin” is arbitrarily reserved for those (g1yco)proteins with at least two carbohydrate-binding sites per molecule so that they can cross-link structures bearing appropriate carbohydrate domains, thus for instance causing the agglutination of red blood cells. Lectins could theoretically act as receptors for certain oligosaccharins, especially those derived from chitin (Barber et al., 1989) since there are many plant lectins that recognize N-acetyl-P-D-glucosamine residues. Indeed, legume root lectins may well be responsible for recognizing the bacterial oligosaccharin that induces nodulation (Lerouge et al., 1990) since transformation of white clover cultures with the gene for a pea lectin allowed a pea-specific Rhizobium to nodulate the clover roots (Diaz et al., 1989). a-L-Fucose-specific lectins are also known which bind xyloglucan (Hayashi and Maclachlan, 1984) and would probably bind the oligosaccharins XG9, XG5 and 2’-fucosyl-lactose. Such lectins are present in at least some plants, e.g. gorse (Ulex europaeus), and might even turn out on closer inspection to occur in small amounts in all plants. A lectin that binds oligogalacturonides is also known (Benhamou et al., 1988). However, there are no known lectins that bind two other classes of oligosaccharins-the oligo-P-glucans and the oligo-P-xylans. Conversely, several lectins are known that recognize a-D-mannose, a-D-galactose, P-D-galactose, and N-acetyl-a-D-galactosamine residues-none of which is known to be an essential part of a well authenticated oligosaccharin. Thus, a close functional link between the oligosaccharins and the classical lectins seems unlikely. This, of course, is not to say that carbohydrate-binding (g1yco)proteins other than classical lectins are unimportant in oligosaccharin action: witness the oligo-P-glucan binding site (Section 1II.A.1). A variation on this hypothesis-a binding site in a microbial cell surface that recognizes extracellular plant carbohydrates-is provided by the observation that the carbohydrate moieties of root slime provide binding sites for fungal zoospores (Hinche and Clarke, 1980; Longman and Callow, 1987). Thus, the perception of specific carbohydrates by cell surfaces is certainly not unprecedented.

44

S. ALDINGTON AND S. C. FRY

Oligogalacturonides are presumably too large and too polar to pass through plasma membranes unaided; some sort of interaction with the plasma membrane probably occurs. Horn et al. (1989) used fluorescence microscopy to follow the fate in suspension-cultured soybean cells of fluorescein-labelled dodeca-a-(1-+4)-galacturonide (GalAI2) and of a fluorescein-labelled elicitor from the fungal pathogen Verticillium dahliae (probably a glycoprotein of molecular weight > 30 000), which both elicit glyceollin synthesis. In both cases, the ligand quickly accumulated at the cell surface before becoming internalized into the vacuole or bound to the tonoplast. The GalA12-fluorescein complex required less time for internalization (2 h) than the larger fungal elicitor (4 h). Two fluorescein-labelled non-elicitor polypeptides-bovine serum albumin (molecular weight 68 000) and insulin (molecular weight 6000)-were not able to enter the cells even after an 8 h incubation, nor did they bind to the cell surface. A GalA12-[’251]iodotyro~ine conjugate was taken up whereas an ovomucoid[1251]iodotyrosineconjugate was not (Fig. 5 ) . The GalAI2derivatives were internalized intact (rather than after breakdown to the monosaccharide), by an energy-dependent, temperature-dependent process. Uptake of the GalA12-[1251]iodotyrosine conjugate was reduced about 10-fold by addition of a 10-fold molar excess of non-radioactive GalA12. It was suggested that uptake was receptor-mediated. Filippini et al., (1992) have shown that oligogalacturonides can interfere with the specific binding of indoleacetic acid (IAA) to isolated membranes. Thus, it seems possible that oligogalacturonides and IAA may have a common binding site in the plasma membrane. Chitosan oligosaccharins, which affect plant cell membranes rapidly in suspension cultures and more slowly in intact tissues, can apparently bind to the plant cell wall, membrane and nucleus. When [3H]chitosan was applied to the surface of pea endocarp tissue, the label was detected in the cytoplasm and nucleus within 15 min (Hadwiger et al., 1981; Kendra and Hadwiger, 1987b). On the other hand, 3H-labelled oligosaccharides of xyloglucan (including XG9) failed to enter cultured spinach cells to any appreciable extent even after an incubation period of 72 h (Baydoun and Fry, 1989; Smith and Fry, 1991). This lends support to the idea that the uptake of oligosaccharins, where it does occur, is mediated by specific binding sites in the plasma membrane and is not simply a consequence of “accidental” trapping during endocytosis. The mechanism by which 1nM XG9 inteferes with the growth-promoting action of 2,4-D and H + in pea stem segments is unknown. The possibility should be considered that it acts by interfering with the transport or binding of auxin, a plasma membrane phenomenon (Barbier-Brygoo et al., 1991). The specific structural requirements for the growth-inhibiting activity of XG9-especially the dependence on an a-L-fucose residue-suggest a specific receptor for XG9 and related fucose-containing oligosaccharides

-

-

45

OLIGOSACCHARINS

20

15 r

rl

X -1 J

w

x2

10

t : 2 0

2c

5 A c I

1

2

3

4

5

TIME (HWRS) Fig. 5. Time-course of internalization of 'Z51-labelled conjugates of oligogalacturonides and ovomucoid in cultured soybean cells. 0 , '251-labelledoligogalacturonides at 23°C; A , l r s I labelled oligogalacturonides at 4°C; 0,'2sI-labelled oligogalacturonides at 23°C in the presence of non-radioactive oligogalacturonides; 0, '2SI-labelledovomucoid at 23°C. From Horn er al. (1989).

(McDougall and Fry, l988,1989a,b). Since XG9 is not significantly taken up by cells, the putative receptor would have to be located on the plasma membrane or (less probably) in the cell wall. The very low concentration optimum (-nM) of XG9, some orders of magnitude lower than the K , values of typical enzymes, is also compatible with a receptor, similar to a hormone receptor. However, no work has yet been reported to show XG9-binding sites, either in membranes or in any other subcellular location. A binding site for a less well characterized lignification-inducing glycoprotein from Puccinia graminis (see Section 1I.F) has also recently been reported (Kogel et al., 1991). Specific binding sites of molecular weight 30000 were found in the plasma membrane of wheat and barley leaf cells.

46

S. ALDINGTON AND S. C. FRY

Finally, a novel approach to the demonstration of putative membranelocalized oligosaccharin-receptors is to demonstrate biological effects on isolated protoplasts of carbohydrates that have been conjugated to molecules or particles so large that they cannot possibly penetrate the plasma membrane. Thus, the finding of effects of silica-immobilized sugar residues (Lienart et al., 1991) is circumstantial evidence for receptors in the plasma membrane. B. RAPID EFFECTS OF OLIGOSACCHARINS

If, as suggested in Section III.A, there are oligosaccharin receptors in the plasma membrane, what do these receptors do after binding an oligosaccharin? Some very fast, almost instantaneous responses by plant cells have been reported to occur after the addition of oligosaccharins. These responses tend to be related to effects which the oligosaccharins have on the plant cell membrane, suggesting some sort of direct link between the two. 1. Membrane depolarization Some oligosaccharins can cause rapid membrane depolarization. Fungal elicitors prepared from the ethanol-soluble material of mycelia from two plant pathogens (Colletotrichum lagenarium and Phytophthora parasitica var. nicotianae) were applied to the roots of host plants-melon and tobacco, respectively (Pelissier et al., 1986; see Fig. 6b). Depolarization occurred within seconds in both systems; in melons, a new steady state was reached after lmin, which remained stable for at least a further 25min. Removal of elicitors caused repolarization. Elicitors from C. lagenarium also caused depolarization in maize roots; this effect therefore seems unrelated to pathogen specificity. After elicitors from P. parasitica had been fractionated by gel-permeation chromatography on Bio-Gel P-2, those of intermediate and high molecular weight induced a marked membrane depolarization and were very active inducers of ethylene biosynthesis. The elicitors caused depolarization by reducing the electrogenic component. The ethylene-inducing oligosaccharins derived from melon cell wall pectins (see Section II.C.7) also induced rapid membrane depolarization in roots of host and non-host plants (Esquerre-Tugaye et al., 1985). Mayer and Ziegler (1988) treated soybean tissue with a highly purified glucan elicitor preparation (molecular weight 4000-8000) from the cell walls of the fungus Phytophthora megasperma f.sp glycinea. At 1 mg l-', the elicitor induced depolarization within 2 min of contact with the plant cells. However, after a further lOmin, the membrane hyperpolarized. The concentration of the glucan which induced nearly maximal phytoalexin synthesis (0.1 mgl-') did not induce depolarization, but did cause the hyperpolarization. This effect was maximal at pH 6, which was also the

47

OLIGOSACCHARINS On A

-201

B

-185

1

NaCN

Low DP

C Off D High DP

-100

f

I

I

I

I

I

,

off

I

I

Time (5-min intervals)

(4

1

z2

f .parasitica E.

1C.

lagenarium E.

a

-15017

i

I \

rc

/Melon t-E

-loo/

f

-E

5 min

Fig. 6. Effects of oligosaccharins on plant cell membrane potential. The effectors were added at and then rinsed away at t . (a) Depolarization of tomato leaf mesophyll cells by NaCN (1 mM; curve A), small oligogalacturonides (mean DP = 4) (1 g I-'; curves B, C), and large oligogalacturonides (mean DP = 15) (1 gl-'; curve D). The units for the y-axis are indicated on the individual curves. The gaps (-I )-I were 10 min. From Thain et al. (1990). (b) Depolarization of melon and tobacco root hairs by elicitors (-0.1 g l - ' ; E.) from Colletotrichum lagenarium and Phytophthora parasitica, respectively. From PClissier et al. (1986).

48

S. ALDINGTON AND S. C. FRY

optimum pH for induction of PAL. It was suggested that hyperpolarization played a role in the induction of phytoalexin synthesis but that depolarization did not. Results reported by Tomiyama et af. (1983) suggested a much slower effect of living Phytophthora infestans on the membrane potential of infected potato cells. Infection by compatible races did not affect the membrane potential within 24 h; infection with incompatible races resulted in depolarization one to several hours after fungal penetration of the cell. It is understandable that exogenous purified oligosaccharins should have quicker and more widespread effects than fungal infection. More recently, Thain et al. (1990) studied the effects of oligogalacturonides on the membrane potential of tomato leaf cells. The membrane potential (about - 200 mV) that is normally present in these cells is thought to be maintained by the operation of an electrogenic H+-pump; it can be approximately halved by application of 1mM cyanide (Fig. 6a). At 181-' (- 1.4 m ~ )small , oligogalacturonides (DP 1-7; mean 4) caused a large and rapid depolarization (by -50 mV within -0.5 min), which was reversed within 5-10 min when the oligosaccharins were removed. Similarly, larger oligogalacturonides (DP 10-20; mean 15) at 1 gl-' ( - 0 . 4 m ~ ) caused a depolarization which was initially rapid (- 25 mV within 1min) and then followed by a slower continued depolarization (by a further -25 mV within 10min); this effect was also reversed within 10 min after removal of the oligosaccharins (Fig. 6a). The difference between the two oligosaccharin preparations may have been related to the different defence responses evoked by them: both the preparations induced protease inhibitor synthesis; phytoalexin synthesis was elicited by the larger fragments only (see Sections II.C.2 and 3). How pectic oligosaccharins evoke the transcription of protease inhibitor genes is unknown; indeed, as discussed in Section VI.C, it is unclear whether they do evoke it at all, or whether their action is limited to initiating the dispatch of the mysterious second messenger, the true PIIF. The effect of oligogalacturonides (and of injury) on protease inhibitor induction in tomato can be reversibly blocked by pre-treatment of the plant with acetylsalicylic acid and related benzoic acid derivatives (Doherty et af., 1988), although in cultured rice cells salicylic acid induced a protease inhibitor (Masuta et af., 1991). However, in view of the major uncertainty about the target of the oligogalacturonides, there is little that can yet be concluded from this about the mode of action of wound signals. So what are the consequences of oligosaccharin-induced depolarization? Is electrolyte leakage enhanced? Is membrane permeability affected allowing internalization of oligosaccharins? Addition of polycations (e.g. chitosan) to cell cultures rapidly resulted in electrolyte leakage owing to an increase in membrane permeability (Young et al., 1982; see Section II.E), although polyanions (e.g. polygalacturonic acid and poly-L-aspartic acid)

-

-

-

OLIGOSACCHARINS

49

that induced the release of Ca2+from whole cells did not affect membrane permeability (Young and Kauss, 1983). The major electrolyte leaked by cultured cells on addition of chitosan was K + , with 5 1 5 % of cellular K+ exiting within the first 30 min; by the time this efflux had diminished, callose synthesis had reached a constant rate. Ca2+ may possibly enter the cell by the membrane depolarization occurring owing to the K+ efflux but there was no correlation between the degree or time course of electrolyte leakage and callose formation (Kauss, 1987). Atkinson et al. (1985) provided evidence that a K+ efflux/H+ influx is an important preliminary to the hypersensitive response of tobacco to Pseudomonas syringae. Net K+ efflux began 1.0-1.5 h after inoculation of tobacco cells with bacteria and it reached a maximum in 2.5-3.0 h, dropping within 5 h. Purified pectate lyase from Erwinia chrysanthemi induced a similar K + efflux/H+ influx in suspension-cultured tobacco cells (Atkinson et al., 1986). This may point to a mechanism by which pectic enzymes (or perhaps their oligosaccharide products) induce the hypersensitive response. Peever and Higgins (1989) also reported a rapid enhancement of electrolyte leakage in tomato-Cladosporium fulvum interactions. Specific elicitors isolated from apoplastic fluid induced electrolyte leakage only in cultivars resistant to the race of C.f u f v u mused. A glycoprotein non-specific elicitor induced electrolyte leakage in both resistant and susceptible cultivars. Dow and Callow (1979) had previously shown that culture filtrate of three races of C.fulvum elicited rapid electrolyte leakage from isolated tomato leaf mesophyll cells, the active elicitors in this case being glycopeptides. Uptake of [''C]leucine across the plasma membrane in cultured sycamore cells was found to be inhibited by the same acid-solubilized plant cell wall fragments that caused cell death (Fry et al., 1983; cf. Section II.C.4). This again indicates an interference by oligosaccharins with the activities of membranes.

2. Oxidative metabolism The generation of the superoxide anion radical ( 0 2 - - ) appears to act as a trigger for several defence responses when inflammatory (animal) cells are exposed to microbial challenge (Badwey and Karnovsky, 1980). One effect is to produce more reactive oxygen species such as hydroxy radicals (OH.) (Halliwell, 1978; Halliwell and Gutteridge, 1984), which can attack many biological molecules, especially polyunsaturated fatty acids, which are thereby peroxidated. Hence the lipids in membranes are particularly vulnerable to free radicals of oxygen: RH+OH* * R*+H20 Re + 0 ROO.

2

+ R'H

+

(1)

ROO- (peroxide radical)

(2)

+ ROOH (peroxide)

(3)

-+ R'.

50

S. ALDINGTON AND S. C. FRY

where R are R’ are fatty acids. This chain reaction can be terminated at eqn (2) by reduction of the peroxide radical with glutathione (G-SH) and glutathione peroxidase: 2 ROO.

+ 2 G-SH + 2 ROOH + G-S-S-G

(4)

Epperlein et al. (1986) attempted to assess the involvement of reactive oxygen species in phytoalexin (glyceollin) accumulation in soybeans responding to an abiotic elicitor, AgNO3. The cotyledons were treated with acetaldehyde), excessive a 02---generating system (xanthine oxidase being prevented with superoxide dismutase which accumulation of 02-* catalyses the dismutation of 0 2 - * to H202

+

2 02-.

+ 2H+ + H202 + 0 2

(5)

supplied in this way did not mimic AgN03 in eliciting glyceollin was not directly involved in phytoalexin synthesis, suggesting that 02-* elicitation. However, OH. appeared to have a role: scavengers of OH. such as mannitol, dimethylsulphoxide, benzoate and methionine inhibited the AgN03-induced accumulation of glyceollin. This suggested a possible role for lipid peroxidation in phytoalexin accumulation. As this type of reaction was shown to occur in plant cells with an abiotic elicitor, the possibility of biotic elicitors initiating events by free radical generation has been investigated. Reactive oxygen species have now been detected in host-pathogen interactions and as a rapid response to elicitors. For example, suspensioncultured soybean cells exhibited a greatly enhanced luminol-dependent chemiluminescence 20-30 min after the addition of P-glucan elicitor from Phytophthora megasperma; the response increased for at least a further hour. The increase in chemiluminescence was inhibited by exogenous superoxide dismutase and catalase, and also by peroxidase inhibitors, suggesting . H 2 0 2were involved in generating the luminescence and that both 0 2 -and that peroxidase played a role in the formation of these active oxygen species (Lindner et al., 1988). The rapid production of H202 by soybean cells, and its use by endogenous cell wall peroxidases following the addition of elicitors (crude autoclaved cell walls/membranes; Apostol et al., 1987) from Verticillium dahliae and oligogalacturonides, caused vigorous peroxidatic activity within 5 min. A variety of water-soluble exogenous test compounds, including IAA and certain fluorescent dyes, were rapidly destroyed by oxidation (Apostol et al., 1989) (Fig. 7). IAA has been shown to inhibit the induction of chitinase (a pathogenesis-related enzyme) in Nzcotiana tubacum; the enzyme could not be detected in the upper leaves where the IAA concentration was relatively high, but represented 1 4 % of total soluble protein in the lower leaves and roots where there was less IAA (Shinshi et al., 1987). Hence, the destruction of IAA as a consequence of oligosaccharin treatment may be related to 02-*

OLIGOSACCHARINS

51

NADH t Hi

Fig. 7. A model of the arrangement of components involved in the oxidative burst evoked by elicitors. According to this model, any component which interferes with receptor-reductase coupling, or which consumes electrons [e.g. Fe(CN),'-] or reduces H2O2 [e.g. Fe(CN),"], or which blocks the oxidase (e.g. KCN) would be expected to inhibit both the bleaching and transmembrane signalling functions of the oxidative burst. From Apostol et al. (1989).

evocation of certain defence responses. The same response might also account for the ability of oligogalacturonides to antagonize many of the effects of IAA (see Section II.C.l);if so, we would predict that oligogalacturonides should not affect the response of plants to "non-oxidizable" auxins such as 2,4-D. The generation of 0 2 - - may be involved in the hypersensitive response (Doke, 1983a,b). Potato tuber disks inoculated with spores of an incompatible race of Phytophthora infestans, which evoked the hypersensitive response, increased in ability to reduce extracellular cytochrome c 1-4 h after inoculation (Doke, 1983a); the increase was also evoked by fungal wall components (Doke, 1983b). [Spores, which germinated on the disks, started to penetrate cells 1h after inoculation.] Infiltration of the disks with superoxide dismutase delayed both the hypersensitive response and phytoalexin accumulation in the incompatible interaction. The conclusion was that the reducing activity was dependent on 0 2 - * cyt c (Fe")

+ 0 2 - * + cyt c (Fe2+) + 0 2

(6) Another hypersensitive-like response, the browning and death of cultured rice cells induced by chitosan, was blocked by catalase, suggesting a

52

S. ALDINGTON AND S. C. FRY

role for H202, although exogenous H 2 0 2itself did not induce browning and death (Masuta et al., 1991). The response was also blocked by vitamins C and E (free-radical scavengers) but not by superoxide dismutase. Lipid peroxidation, which can cause dramatic alterations to membrane permeability, has been reported in several plant tissues on exposure to elicitors. It appears to be a characteristic response to mechanical damage and microbial infection (Chai and Doke, 1987);it has been reported to occur during the hypersensitive response and to be initiated by 0 2 - * (Adam et al., 1989). Keppler and Novacky (1987) found an increase in lipoxygenase activity in cucumber cotyledons infected with incompatible bacteria. Lipoxygenase (EC 1.13.11.12) catalyses the reaction: ,

..--CH=CH-CH,-CH=CH--.

.. + 0

2

+ ...-CH-CH=CH-CHSH-.

.. (7)

I

0-OH

The target group (. ..-CH=CH-CH2-CH=CH-. ..) occurs in several common fatty acids, e.g. linoleic, linolenic, arachidonic (= eicosatetraenoic) and eicosapentaenoic acids, some of which are themselves elicitors (Bostock et al., 1981) and precursors of traumatin and jasmonic acid. One way in which lipoxygenase may be beneficial during infection is to scavenge harmful free fatty acids, so any increase in lipoxygenase activity may be a response to free radical-initiated membrane damage. Increased levels of lipoxygenase activity were induced in the system studied by Peever and Higgins (1989), in which lipid peroxidation was also induced. Preparations from Colletotrichum findernuthianurn (including a galactoglucomannan) elicited phytoalexins and stimulated the accumulation of products indicative of lipid peroxidation in bean tissues 6 hours after treatment. These responses were also stimulated by generators of activated oxygen species, although whether they were stimulated more quickly than by the polysaccharide was not stated (Rogers et al., 1988). The accumulated evidence suggests that elicitors of defence responses, especially elicitors of fungal origin, do play some role in generating activated oxygen species, and that the presence of such species can have rapid effects on plant cell membranes. Hence, membrane damage could be an early signal to the plant of a potential pathogen, and this could be beneficial in allowing the release of autolytic enzymes that generate oligosaccharins from the plant’s own cell wall (Lyon and Albersheim, 1982).

3. Protein phosphorylation Another interesting rapid oligosaccharin effect rather different from those described above was reported by Grab et al. (1989). Following the addition of a P-glucan elicitor from Phytophthora megasperma, there were rapid changes in phosphate turnover in several phosphoproteins. The effect of the

OLIGOSACCHARINS

53

elicitor on protein phosphorylation was tested after in vivo labelling with [32P]orthophosphate. Decreases and increases in the labeling of several phosphoproteins occurred about 5 min after elicitor application. One particular polypeptide (molecular weight 69000) showed a decrease in 32Pin response to elicitor treatment but was strongly phosphorylated in vitro in the presence of a low molecular weight factor ( M , = 1000) from soybean cell cultures. This “effector” was partially characterized and found to be negatively charged at pH 7.3; its activity was reduced after treatment with alkaline phosphatase and with crude preparations of pectinase and cellulase. Farmer et al. (1989, 1991) have shown that plasma membranes isolated from tomato and potato are promoted in their ability to incorporate label from [ Y - ~ ~ P I A into T P proteins in vitro by the addition of 0.1-1.0mM oligogalacturonides of D P 20. Specific protein phosphorylation was also reported to occur by Dietrich et al. (1990) in parsley cells elicited by a fungal P-glucan. A 45 kDa protein, found in microsomal and cytoplasmic fractions, was phosphorylated as early as 1min after treatment. A 26 kDa nuclear protein was also phosphorylated rapidly. Changes in protein phosphorylation correlated with the biological response of the cells and appeared to depend on the presence of Ca2+ in the medium. Phytoalexin accumulation was reduced in Ca2+-deprivedcells. It was hypothesized that signal transduction may involve the activation of a Ca2+-dependentprotein kinase.

-

4. Second messengers D o levels of putative second messengers, such as intracellular Ca2+,change during oligosaccharin treatment, as predicted by the work of Dietrich et al. (1990)? Kurosaki et al. (1987b) found that addition of verapamil (a Ca2+channel blocker) to carrot cell cultures, 30 min (but not 60 min) after fungal elicitors, inhibited subsequent phytoalexin accumulation, suggesting that intracellular Ca2+ concentration was important in elicitor action. They also found that intracellular levels of cAMP increased on addition of an elicitor. This suggested that Ca2+ and cAMP may participate as second messengers in the regulation of phytoalexin production. Hahn and Grisebach (1983), however, denied any role of CAMP. Direct measurements, using plant cells transformed with the gene for a Ca2+-sensitive fluorescent protein, aequorin, have demonstrated a rapid elevation of intracellular Ca2+concentration by P-glucan elicitors (Knight et al., 1991). Recent work in the same laboratory (Messiaen et al., 1992) has shown that A4~s-oligogalacturonideslarge enough to form an oligosaccharide-Ca2+ “egg-box’’ conformation (see Jarvis, 1984)-and to elicit phytoalexins-also caused substantial increases in intracellular free Ca2+, especially near the plasma membrane. This effect was blocked by verapamil. It thus appears plausible that some oligosaccharins act primarily by interacting with the plasma membrane, secondarily inducing a change in

54

S. ALDINGTON AND S. C. FRY

intracellular free Ca2+. Certain protein kinases are Ca2+-dependent,so the tertiary effect could be on protein phosphorylation. Ca2+ also regulates the phosphatidylinositol phosphate signalling system in animals, and so it is interesting to note that Kurosaki et al. (1987a) have demonstrated a turnover of phosphatidylinositol during the elicitation of phytoalexins in carrot cell cultures. 5. R N A synthesis Following infection by a potential pathogen, the transcription of defencerelated genes is activated as part of a massive switch in the pattern of host gene expression (Cramer et al., 1985). Ryder et a f . (1986) treated bean cell cultures (Phaseolus vulgaris) with elicitors released by heating walls of the fungus Colletotrichum lindemuthianum. This inhibited the synthesis of several polypeptides whilst stimulating the synthesis of at least 60 others. Stimulation was usually observed only 1h after elicitor treatment. The pronounced change in the pattern of translation was due to a switch in the pattern of mRNA synthesis. Marked changes were also observed in bean hypocotyls infected with an incompatible race of C. lindemuthianum. PAL and chalcone synthase (CHS) mRNA accumulated within 35 h of inoculation; this may actually represent a rapid response as it takes 3 0 4 0 h following inoculation for the spores to germinate and come into contact with the first host cell. PAL is the first enzyme on the pathway from primary metabolites to phenolic phytoalexins and lignin; CHS is the first enzyme on the branch leading to flavonoid synthesis. Somssich et al. (1989) isolated numerous cDNA clones corresponding to genes which were rapidly activated in parsley cells by fungal elicitors. Transcription of 18 different genes was rapidly and transiently activated. The genes encoding PAL and CHS were activated within 2-3min of elicitor treatment (K. Edwards etal., 1985; Lawton and Lamb, 1987). Ryder et al. (1986) reported that the elicitor stimulation of cinnamyl alcohol dehydrogenase (CAD) mRNA in suspension-cultured bean cells was more rapid than that observed for PAL and CHS. CAD is the enzyme that diverts metabolic flux from general phenolics towards lignin synthesis. The sequence for mRNA accumulation was (1) CAD, (2) PAL and CHS, (3) HRGP. The increase in HRGP mRNA occurred much more slowly, usually requiring a lag period of at least 1h after elicitation (Showalter et al., 1985; Templeton and Lamb, 1988). Transcripts encoding chitinase also accumulated very rapidly and with kinetics similar to those of C A D mRNA (Hedrick et al., 1988). Somssich et al. (1986), using the elicitor from Phytophthora megasperma f. sp. glycinea which induces phytoalexin accumulation in parsley, reported a rapid increase in mRNAs for certain pathogenesis-related (PR) proteins. There was a four-fold increase in the transcription rate of the PR1 gene within 5 min and a three-fold increase for the PR2 gene within 20min (Fig. 8).

OLIGOSACCHARINS

200

55

-I

Time after addition of elicitor (min)

Fig. 8. Rapid effects of an elicitor (P-glucan from the fungus Phytophthoru megasperma) on transcription of the genes for two pathogenesis-related proteins (PR1 and PR2), pcoumarate: CoA ligase (4CL), and phenylalanine ammonia lyase (PAL), relative to that of an elicitor-insensitive gene (LF14). in cultured parsley cells. Nuclei were isolated at the times indicated and allowed to incorporate [IY,-~’P]NTP into “run-off” RNA, which was then hybridized to the relevant cDNA and assayed for 32P.From Somssich et ul. (1986).

Templeton and Lamb (1988) have described the early transcriptional events of several enzymes in bean-Colletotrichum lindemuthianum interactions. The rapidity of these responses suggested that there were very few steps between oligosaccharin recognition and activation of defence-related transcription. The genes were activated much more quickly in incompatible interactions than in compatible interactions. Furthermore, the transcription of these genes was also induced some distance from the infection site. The induction of protease inhibitors in tomato leaves by pectic fragments also appears to depend on transcription, but the protease inhibitor mRNAs do not accumulate very rapidly (Nelson et al., 1981). The mechanism by which gene transcription is modulated by oligosaccharins is unknown. In the case of oligosaccharides of chitosan, which can enter the plant cell and accumulate in the nucleus, it could be proposed that they interact directly with the chromatin. Indeed, it was shown that, not surprisingly for a polycation, chitosan will bind in vitro to DNA molecules and hence change their physical properties (Hadwiger et al., 1981). It was also shown that chitosan inhibited the incorporation of [’Hluridine into mRNA (Kendra and Hadwiger, 1987b). I t was proposed that defence

56

S. ALDINGTON AND S. C. FRY

responses initiated by the host were due to structural changes in the host’s nuclear material, with the hypersensitive response and host membrane deterioration being consequences of these responses. However, it seems improbable that other, non-cationic, oligosaccharins act in this way, and even oligosaccharins of chitin would presumably have to act by a different mechanism. It seems clear that plant cells exhibit numerous diverse rapid responses to oligosaccharins. There would appear to be little real evidence on which to decide objectively which, if any, of these are required for the ultimate physiological effects observed (Section 11), and in what cause-and-effect order they may operate.

C. DIRECT EFFECTS OF OLIGOSACCHARIDES ON ENZYMES

1. X y Zoglucan endotransgly cosylase Although XG9 at 1 nM inhibits the growth-promoting effect of 2,4-D in pea stem segments, at concentrations 3 100nM it does not (York et al., 1984; McDougall and Fry, 1988). This suggests that, at the higher concentrations, a second (growth-restoring) effect of XG9 comes into play, nullifying the growth-inhibiting effect (see Section 1I.B). The fact that XG5 and 2‘fucosyl-lactose do not become less effective inhibitors of auxin-promoted growth at 3 l 0 O n ~suggests that they lack this second function of XG9. Direct tests of this hypothesis led to the discovery of the ability of XG7, XG8, XG9 and XG9n to promote elongation in the absence of auxin (McDougall and Fry, 1990). What is the mode of action of XG9 as a growth restorer/prornoter? An important clue was the finding by FarkaS and Maclachlan (1988) that oligosaccharides derived from xyloglucan can apparently stimulate the activity of “cellulase” in a cell-free system. The effect was only observed when xyloglucan was used as substrate (not carboxymethylcellulose), only when the “cellulase” was obtained from plants (not from fungi), and only in viscometric (not reductiometric) assays of substrate degradation. Since xyloglucan oligosaccharides are an end-product of the action of cellulase, it would be unusual if they did indeed stimulate cellulase activity-a case of positive feedback. In fact, the enzyme responding to the oligosaccharides now appears likely to have been xyloglucan endotransglycosylase (XET; see Section 1I.B) (Fry et al., 1992) rather than cellulase. The oligosaccharides XG7, XG8, XG9 and XG9n (which all have four glucose units in the backbone) are substrates for XET, whereas XG5 and 2’-fucosyl-lactose (which have two and one glucose units, respectively) are not. The stimulation of xyloglucan degradation (decrease in viscosity) by the larger oligosaccharides would be due to their ability to compete with other xyloglucan molecules as glycosyl acceptors in the XET reaction. Endotransglycosyl-

57

OLIGOSACCHARINS

ation between twopolysaccharide molecules (0000... and o. ...) results in no change in the mean molecular weight of the reactants since for every bond broken a new bond is re-formed:

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

00000000000000000000000000000000000000000000000000000000000000000000

+

+

0000000000000000000000000000000000000000

......

However, when this principle is applied to endotransglycosylation between .) equimolar amounts of xyloglucan (say, molecular weight 200 000; and XG9 (molecular weight 1500; a), the products should have a mean molecular weight of 100 000

-

-

00000000000000000000000000000000000000000000000000000000000000000000

+. + 0000000000000000000000000000.

+

0000000000000000000000000000000000000000

and, since XG9 contributes only negligibly to viscosity, the viscosity of the reaction mixture would be strongly reduced by the occurrence of this reaction. Thus, there is a correlation between the ability of xyloglucan oligosaccharides to promote (and/or restore) growth and their effectiveness as substrates for XET. Could this be a causal relationship? The action of X E T in vivo, in the absence of xyloglucan oligosaccharides, may promote the extensibility of the cell wall by cleaving load-bearing, intermicrofibrillar xyloglucan chains-thus allowing incremental growth (Fry, 1989b)-and then re-joining the cut ends to different acceptor xyloglucan chains, thereby restoring the original strength of the cell wall. When FM XGOs are experimentally added into the system, however, the XET can frequently transfer the cut xyloglucan chains on to these short “stubs”, using them, instead of the neighbouring structural polysaccharides, as acceptors; the strength of the cell wall would thus fail to be restored, and cell extension would continue for longer than usual.

2. Other enzymes Several polyvalent anions, e.g. pyrophosphate, oxalate, oxomalonate and citrate, dramatically promote the oxidation of IAA by plant peroxidases in vitro (Pressey, 1990). The optimum concentration of these anions was 100 FM. Pressey (1991) has also shown that a similar effect is exerted on peroxidase by oligogalacturonides that have been enzymically modified by oxidation of the reducing terminal galacturonate unit to galactarate. Galac, oligomers of the form taric acid itself was active above about 300 p ~ but

-

[D-Galacturonic acid-a-( 1-+4)],-Galactarate

58

S. ALDINGTON AND S. C. FRY

were active at considerably lower concentrations. For example, the penta. (1991) also saccharide appeared to be effective at about 50 p ~Pressey states that several plant tissues were shown to contain a uronate oxidase that converted normal oligogalacturonides (the products of pectinase digestion) to these oxidized derivatives. Promotion of IAA oxidation in vivo by an enzymic oxidation product could be a further factor explaining the fact that oligogalacturonides antagonize several of the effects of IAA on plants (see also Sections III.A.2 and B.2). A plant pectinmethylesterase has also been reported to be inhibited by oligogalacturonides of D P 3 8 (Termote et al., 1977). It is interesting to consider whether this effect may be related to growth regulation.

IV. NATURAL OCCURRENCE OF OLIGOSACCHARINS If a biological signalling role is to be ascribed to oligosaccharins, it is essential to demonstrate their natural occurrence at the right time, place and concentration to carry out the biological functions envisaged for them. Clearly the appropriate polysaccharides and glycoproteins (from which oligosaccharins may arise) occur in vivo, but surprisingly few studies have addressed the question of whether the oligosaccharins themselves occur. This is particularly true for the oligo-P-(1+3),(1-+6)-glucans, which, despite the detailed information available about the action of artificial oligosaccharins (Section II.A), do not appear to have been demonstrated to occur naturally at all. Although the mere existence of biological responses to artificial oligosaccharins, some extremely potent and specific, may suggest that they do play a natural role, this kind of evidence is not proof. Other reasons for developing techniques for studying the natural occurrence of oligosaccharins are: (1) so that changes in their endogenous levels within plant tissues can be monitored, e.g. in responses to environmental and hormonal stimuli or to infection; and (2) so that potential natural sources can be explored for possible novel oligosaccharins that might never be generated by the in vitro methods currently used (Section 1.B). A . NATURAL OCCURRENCE OF XYLOGLUCAN OLIGOSACCHARIDES

Acetylated and non-acetylated forms of XG9 and related oligosaccharides of xyloglucan (DP 5-10) have been detected in vivo in the spent medium of suspension-cultured spinach cells (Fry, 1986). The culture medium can be regarded as an extension of the apoplast of these cells, and the discovery of XG9 in the spent medium indicates that it might also occur extraprotoplasmically in intact plant organs. The concentration of XG9 plus its acetates was estimated at 0.4 p ~ . The method used to enable detection of this low concentration of XG9

-

OLIGOSACCHARINS

59

involved the prior in vivo feeding to suspension-cultured spinach cells of high activity ~ - [ ~ H ] f u c oand s e ~-[~H]arabinose, which are incorporated into the L-fucose and D-xylose residues, respectively, of xyloglucan. The culture filtrates were then collected and fractionated on the basis of molecular weight by gel-permeation chromatography on Bio-Gel P-2. This simple method is the most sensitive means currently available for detecting trace levels of specific carbohydrates, permitting the detection of picogramme quantities. The nonasaccharide-enriched, 3H-labelled material was further fractionated by two-dimensional paper chromatography, and a spot was detected that appeared to be an acetylated derivative of XG9. After mild alkaline hydrolysis to remove the acetyl group(s), evidence that the material was indeed [3H]XG9 and not another oligosaccharide of similar size and RF was obtained by use of acid- and enzyme-catalysed hydrolysis, paper electrophoresis in the presence of borate or molybdate (Weigel, 1963), and related techniques (Fry, 1986; for a review of the methods, see Fry, 1988). No evidence has yet been obtained that spinach cell cultures are responsive to xyloglucan oligosaccharins. Thus, the biological significance l ~in spinach culture filtrates is difficult to assess. of the finding of 0.4 l ~XG9 This concentration is supraoptimal for the inhibition of 2,4-D-stimulated growth in pea stem segments, and closer to that required for growth promotion in the absence of auxin (Section 1I.B).

B. NATURAL OCCURRENCE OF PECTIC OLIGOSACCHARIDES

1. Wound signals The natural occurrence of pectic fragments, possibly acting as wound signals in mechanically injured plant tissue, has been considered. Ryan (1974) found that when fresh tomato leaves were ground in a mortar and pestle at 2°C and then squeezed through a hand-operated garlic press, the juice possessed wound-signal activity. This is evidence for the natural occurrence of wound signal in wounded tissue, although, unfortunately, this particular material was not studied further. The detailed characterization of a tomato wound signal was conducted on extracts of autoclaved leaves (Ryan, 1974, and subsequent papers). Autoclaving of plant homogenates would certainly generate soluble pectic fragments from wall-bound pectins; this would occur both by acid hydrolysis (even at pH4-6; see Smidsrod et al., 1966) and, especially in the presence of certain ions such as would be found in crude tomato leaf juice, by p-elimination degradation (Barrett and Northcote, 1965; Aspinall and Cottrell, 1970). The wound signal found in autoclaved preparations may therefore not have been the same as that found in the cold homogenate. The finding of high concentrations of soluble pectic substances in autoclaved tissue is therefore not evidence for their natural occurrence. If the wound signal solubilized at 2°C also turns out to be pectic, this will

60

S. ALDINGTON AND S. C. FRY

support the hypothesis that such substances play a natural role in the activation of protease inhibitor synthesis upon injury, but it will not be sufficient evidence. It poses the difficult question of whether or not uninjured leaves contain the same amounts of soluble pectic fragments. No-one has yet reported the presence of soluble pectic oligosaccharides in uninjured tissues, although this might not be as difficult as it sounds; for example apoplastic fluid could be collected by the vacuum-infiltration/ centrifugation method (de Wit and Spikman, 1982), or the tissue could be homogenized under conditions carefully chosen to prevent both the chemical and the enzymic degradation of pectins. It is difficult to estimate what increase in concentration of endogenous pectic fragments would be needed in order to provide credible evidence for their natural involvement in the wound-signalling system. This is because there is no precise estimate of the quantity of exogenous pectic fragments necessary for the induction of protease inhibitor synthesis. Exogenous pectic fragments ( M , 5000-10 000) have generally been administered through the transpiration stream of excised leaves, and would thus have had to pass along the petiole before arriving in the lamina. The minimum dose that it is necessary to supply to the petiole to obtain a response in the lamina is -0.1 g I-' for 30 min (Ryan et al., 1981); the transpiration rate under the conditions used in earlier work was p1 min-' (Ryan, 1974). These figures suggest that the minimum dose for induction of protease inhibitor is of the order of 1-10 pg of pectic material per leaf. However, it is unclear whether the oligosaccharides need to travel as far as the lamina; the proportion of the pectic material that completes the journey into the lamina has not been adequately investigated. Preliminary studies using ['4C]pectins and oligo-[14C]galacturonideshave indicated that a substantial proportion becomes immobilized at certain points within the xylem of the petiole (E. A.-H. Baydoun and S. C. Fry, unpublished); the formation of pectic plugs in the xylem has also been reported in other circumstances, e.g. as a cause of premature wilting in cut rose flowers (Burdett, 1970). It has long been known that the media of plant cell suspension cultures accumulate soluble polysaccharides, some of which are pectic, during cell growth (Aspinall etal., 1969; Fry, 1980). Some of these, e.g. from tomato, tobacco and lucerne cultures, have wound-signal activity (Walker-Simmons and Ryan, 1986). There has been no evidence for an increase in their concentration in response to injury (e.g. caused by stirring instead of shaking; Brett, 1978), nor does there appear to be any evidence for pectic oligosaccharides (cf. evidence for xyloglucan oligosaccharides-Section 1V.A). Levels of the polysaccharides did tend to increase when the sucrose supply was exhausted, so their accumulation may be a response to starvation (Walker-Simmons and Ryan, 1986).

-

OLIGOSACCHARINS

61

2. Other pectin-related oligosaccharins Hargreaves and Bailey (1978) demonstrated the presence of elicitor-active substances (which may possibly have included pectic fragments) in damaged Phaseolus hypocotyls. Benhamou et al. (1991) used a polygalacturonic acid-binding lectin, complexed to gold, to study the distribution of pectin during the infection of bean tissues with Colletotrichum lindemuthianum. At the same time a PGIP (Section V.B.2) which binds to C. lindemuthianum pectinase, was also tagged with gold and used to localize the site of enzyme accumulation in infected host tissues. The level of pectinase activity increased several-fold when the fungus developed in host plant tissues. The enzyme was found to diffuse freely in the host cell wall and apparently caused degradation of pectins of the primary cell walls and middle lamella. Pectic material appeared to become solubilized and to accumulate at specific sites such as adjacent to the intercellular spaces and in the coagulated cytoplasm of infected host cells. This may reflect the natural production of oligosaccharins of the oligogalacturonide class, which might contribute to the activation of defence responses (Section 1I.C). However, in common with many cytological approaches, the method used does not give unequivocal information about the chemical nature (especially DP) of the material localised, and thus it cannot be definitely concluded that biologically active oligogalacturonides were generated. Details of the methodology are provided by Benhamou etal. (1988), and a similar application of the techniques to the case of the tomato-Fusarium oxysporum interaction has been described (Benhamou et al., 1990). Another difficulty to be overcome is the measurement of the concentration of pectic oligosaccharins in vivo: as much as 0.1-1.0 mM is required for many of the reported biological effects of pectic oligosaccharides and it is far from certain that such high concentrations are generated in vivo.

C. GLYCOPROTEIN-DERIVED OLIGOSACCHARINS

The lignification-inducing glycoprotein of Puccinia walls (see Section 11.F) has been reported to be present in the apoplastic fluid of rust-infected wheat leaves (Kogel et al., 1988), suggesting natural occurrence. Culture filtrates from Silene cell suspension cultures were found to contain a growthregulating heptasaccharide, possibly related to the carbohydrate moieties of N-linked glycoproteins (Priem etal., 1990; Section 1I.F). This, together with the embryogenesis-regulating glycoproteins which also accumulate (de Vries et al., 1988) and the growth-regulating XG9 (Fry, 1986), is another example of the value of the spent media of cell cultures as a source from which to seek novel oligosaccharins. Culture media also accumulate

62

S. ALDINGTON AND S. C. FRY

arabinogalactan proteins-a group of polymers currently considered likely to play a role in the control of morphogenesis (Stacey et al., 1990). A systematic survey of the oligosaccharides that accumulate in culture media, to see how many of them are oligosaccharins, would be valuable. D. CONCLUSION

The study of the natural occurrence of oligosaccharins is in its infancy. This is surprising because the techniques required are simple enough (in vivo feeding of radioactive sugars followed by paper chromatography of the oligomer fraction is a good start!) and the question of natural occurrence is pivotal to the biological relevance of the whole oligosaccharin concept.

V. MECHANISM OF FORMATION AND DEGRADATION OF OLIGOSACCHARINS A. XYLOGLUCAN OLIGOSACCHARIDES

1. Synthesis of xyloglucan oligosaccharides The machinery theoretically required for the production of oligosaccharides from xyloglucan is present in vivo (Fry et al., 1990). It has been shown that xyloglucan is degraded in the stems of many plants during auxin-promoted elongation. This degradation can be shown by pulse-chase experiments in which [14C]sugars are briefly fed to the tissue to generate a cohort of radioactive xyloglucan molecules, which can be extracted and analysed. Some of the initially-formed [ ''C]xyloglucan subsequently vanishes, presumably by degradation to products which are lost upon dialysis (Labavitch and Ray, 1974; Gilkes and Hall, 1977). An apparent solubilization of wall-bound xyloglucan, with little hydrolysis, to yield xyloglucan freely soluble in the apoplast, may also occur (Terry et al., 1981). In addition, the average molecular weight of xyloglucan molecules that remain wall bound may decrease in response to treatment with auxin or H + (Nishitani and Masuda, 1982). Although these studies of xyloglucan turnover were carried out with a view to explaining mechanical wall-loosening during plant growth, the same xyloglucan-degrading machinery could also be responsible for the production of oligosaccharins. The proportion of total wall-bound labelled xyloglucan degraded in vivo is small, at least in rapidly growing Rosa cell cultures (Edelmann and Fry, 1992b), but xyloglucan is so abundant in the cell wall (-10% w/v) that even a small percentage conversion to oligosaccharides would easily lead to the nM and perhaps also p~ concentrations required for oligosaccharin activity (Section 1I.B). All these changes in xyloglucan could theoretically be catalysed by cellu-

63

OLIGOSACCHARINS

lase. Pea stems contain small amounts of cellulase constitutively, and synthesize much greater quantities in response to auxins, especially 2,4-D, at concentrations high enough to promote lateral swelling rather than elongation of stems (Verma et af., 1975). The preferred cell wall substrate of pea cellulase is xyloglucan; cellulases cleave the P-glucan backbone of xyloglucan at specific sites (Section 1I.B). Thus, at least some plant cellulases, as well as fungal cellulases, can cleave xyloglucan in vitro to its constituent oligosaccharide units, mainly XG7 and XG9 (Bauer et af., 1973; Hayashi et al., 1984). Other plant cellulases, e.g. one from tobacco callus, appear to require more than one non-xylosylated glucose residue at the site of attack (Truelsen and Wyndaele, 1992) and will therefore only hydrolyse xyloglucans that have a low xylose :glucose ratio (e.g. that of tobacco; Kato and Noguchi, 1976). To release XG9, XG7 etc. from xyloglucan, cellulase would have either (1) to attack the first non-xyosylated glucose residue away from the reducing terminus of the polysaccharide ( * in the diagram) or (2) to attack two contiguous non-xylosylated glucose residues of the P-glucan backbone (e.g. T, 7 in the diagram) [in the following diagram, the polysaccharide chain has been grossly shortened and galactose and fucose omitted for clarity]: x

x

x

x

x

x

x

x

x

x

x

x

x

1

1

1

1

1

1

1

1

1

1

1

1

1

x i

x .

x

1

1

G-G-G~G-G-G-G~G~G-G-G-GIG-G-GCi-GfG-G-G-G-G-ti-G-(i-G-G-G-G~G~G-G

T

T

T

T

X

X

X

X

T

X

1

T

T

X

X

X

Although some endo-acting glycanases have been shown to carry out multiple attacks on a single chain, thus converting the polysaccharide directly to di- and trisaccharides (English et af., 1972; Bezukladnikov and Elyakova, 1990), this has not so far been demonstrated with a plant cellulase. Nevertheless, even stochastic cleavage of xyloglucan by cellulase would release some XG9, and this might be adequate for an oligosaccharin that is active at concentrations as low as 1 nM. Thus it is reasonable to propose that cellulase catalyses the post-synthetic degradation of xyloglucan to generate oligosaccharins such as XG9. Alternatively, however, it could be suggested that XG9 is synthesized within the protoplast as a pre-formed nonasaccharide, and secreted as such. The mere detection of XG9 in spent media of cell cultures (Section 1V.A) does not distinguish between these two possibilities. Kinetic in vivo labelling studies have recently indicated that the XG9 arises by polysaccharide breakdown. ~-['H]Fucose was fed to suspensioncultured spinach cells and the kinetics of labelling of XG9 and xyloglucan were contrasted (McDougall and Fry, 1991b). After a short lag period (- 0.5 h; presumably the time required for newly synthesized polysaccharide molecules to transit the Golgi vesicular system), [3H]xyloglucan started to appear extracelularly. Its accumulation in the culture medium

64

S. ALDINGTON AND S. C. FRY

continued at a constant rate for at least 7 h. Extracellular [3H]XG9 and [3H]XG9-acetate also appeared in the culture medium after a short lag, but accumulated at a constantly accelerating rate for several hours (indicated by a linear relationship between concentration of [3H]XG9 and ?-Fig. 9).

0

10

20 30 40 Square of incubation time

50

60

(h2)

Fig. 9. Kinetics of accumulation of [3H]XG9 and of soluble [3H]xyloglucanin the culture Values are cpm per 4-ml culture. The medium of spinach cells after addition of ~-[~H]fucose. abscissa shows t2: the fact that the XG9 curve plotted in this way is linear indicates that the accumulation of [3H]XG9exhibits a constant acceleration.

This indicates that the free [3H]XG9 was not synthesized directly from NDP-[3H]sugars and rapidly secreted via the Golgi vesicular system, but that it arose from a gradually labelled pool of precursor material. The steady acceleration in the rate of formation of [3H]XG9would most likely be due to an enzyme (e.g. cellulase) acting at a constant rate on a large, relatively stable, pool of substrate (apoplastic xyloglucan) that was itself increasing in specific radioactivity at an almost constant rate. This mode of origin of XG9 is that proposed in the classic oligosaccharin hypothesis-i.e. that soluble signalling molecules arise by partial enzymic cleavage of wall polysaccharides (Fig. 10). Although it may be reasonable to suggest that cellulase is the enzyme that liberates XG9 from xyloglucan, a new enzyme has recently been found that could also play this role. This is xyloglucan endotransglycosylase (XET) (Smith and Fry, 1991; Fry e t a l . , 1992; Section III.C.l), which cuts (t) the backbone of a xyloglucan molecule, releasing one portion (containing the original reducing terminus) and possibly forming a reactive polysaccharideenzyme (E) intermediate:

65

OLIGOSACCHARINS

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

l

l

l

l

l

l

l

l

l

l

i

l

l

l

l

l

G-G-G-G-C-G~G-G-G-G-G-GIC-G-G-G-G-G-G-~~-G-G-G-G-C~-G-G~ T r T T r r T T X

X

X

X

X

X

X

X

i€ +

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

1

1

1

1

1

1

J

l

l

l

l

l

l

l

l

l

G-Ci-G-G-G-G-G-G-tG-G-G-G-G-G-G-G-G-G-G

~G-G-G-G-G-G-CwG-G-G-G-€

I

t X

t X

X

r

t

T

X

X

X

T X

T

X

from which the polysaccharide can be transferred on to the non-reducing terminus of a different xyloglucan molecule (bb b b b b b b b b b b b b ), x

x

1

1

x 1

x 1

x 1

x J

.

+

G - r G - r G - G - G - G - G - t G - r G - t G - G - t G - t ~ ~ ~ ~ ~ ~ E~ ~ ~ ~ ~ ~ ~ ~

r

X

r

X

T

X

thus effecting transglycosylation, rather than being transferred on to H 2 0 (as in hydrolysis) XET was shown to be a different enzyme from cellulase (Fry etal., 1992). If XET attacks xyloglucan close to the reducing terminus,

-

:.\

[3H]fucose

&*

xyloglucan

*;:*

. oligosaccharides (

.-t ,

Fig. 10. Two models to show how ["H]XG9 could theoretically accumulate in the culture medium (cf. Fig. 9). In hypothesis (a), 3H from GDP-[3H]fucose is incorporated into xyloglucans, some of which are later broken down to oligosaccharides. In hypothesis (b), some 3H is incorporated directly into XG9, which is then secreted ready-made. The data in Fig. 9 support hypothesis (a).

66

S. ALDINGTON AND S. C. FRY

it can release X G 9 (just as it can use X G 9 as an acceptor substrate), although, as with cellulase, such sub-terminal attacks do not appear to be significantly preferred over other potential cleavage sites in the substrate xyloglucan molecule (P. R. Hetherington and S. C . Fry, unpublished). There is, however, some evidence suggestive of multiple attacks on a single xyloglucan molecule (Smith and Fry, 1991) and this might suggest a special role for XET in generating X G 7 , X G 9 etc. in vivo. 2. Degradation of xyloglucan oligosaccharides If X G 9 is a natural signalling molecule, it might be expected to have a relatively short half-life within plant tissue: when it had evoked the appropriate response, the signal would be removed from the system. However, X G 9 seems to be remarkably stable in the presence of cultured plant cells. When IfucosyZ-’H]XG9, [xyZosyl-’H]XG9 or [reducing terminal-’H]XG9 was fed to actively-growing spinach cell cultures, only a neglible proportion of the ’H was taken up o r bound by the cells (Baydoun and Fry, 1989; Smith and Fry, 1991). A small amount of the [’H]XG9 was broken down, but much free [‘H]XG9 remained in the culture medium after 7 2 h. Thus, at least this particular plant tissue does not readily degrade this oligosaccharin. The principal means of removing X G 9 in the spinach system appeared to be by a form of sequestration rather than hydrolysis. By 7 2 h, in the above experiments, an appreciable amount (15-70%) of the total ’H had become sequestered by covalent binding to a soluble, extracellular polymer. Subsequent work has shown that this polymer was xyloglucan, and that the “sequestration” was catalysed by XET (Section V . A . l ) ,for which [’H]XG9 is capable of acting as an acceptor substrate (Smith and Fry, 1991). Plants do contain enzymes theoretically capable of achieving the complete hydrolysis of X G 9 to monosaccharides (Koyama et al., 1983; Edwards et al., 1985, 1988; Fanutti et aZ., 1991; FarkaS et al., 1991). The a-L-fucosidase described by FarkaS et al. (1991)is able to hydrolyse the fucose residues from X G 9 but not those of xyloglucan itself. It is unclear why this battery of enzymes, if present in cultured cells, fails to bring about any appreciable degradation of exogenous [‘H]XG9. B. PECTIC OLIGOSACCHARIDES

1. Formation of pectic oligosaccharides as wound signals In their high molecular weight, wall-bound condition, pectins are unlikely to act as signalling molecules since they are present constitutively. It is only when they are solubilized that they are likely to be effective “alarm calls”. It should be emphasized that, although artificial pectic oligosaccharides have numerous well-documented biological effects, there is to date no clear evidence for the natural occurrence of biologically active pectic oligosac-

OLIGOSACCHARINS

67

charides (Section 1V.B). However, we may reasonably ask what could solubilize them at times of stress. There seem to be two plausible hypotheses: (1) Chelating agents, e.g. citrate and oxalate, are released from the cytosol or vacuole of injured cells and, by breaking Ca2+ bridges in the cell wall, could solubilize some pectins. Chelating agents at room temperature will solubilize small amounts of pectin from most primary cell walls and larger amounts from certain specialized cell types, e.g. the parenchyma of ripe fruit (Jarvis, 1982, 1984). However, pectins are constantly being secreted in soluble form during wall growth (Hanke and Northcote, 1974; Boffey and Northcote, 1975), so the outside of the plasma membrane will normally be coated with newly-secreted, soluble, “periplasmic” pectic molecules; therefore, mere solubilization of wall pectins may be insufficient as a stress signal. Partial degradation to oligosaccharins is probably required. (2) Cellular injury could induce the plant to synthesize pectin-degrading enzymes, release them from a compartmentalized condition, or activate them. Certain plant tissues, especially ripening fruits and abscission zones, have indeed been reported to contain pectinases (Rexova-Benkova and MarkoviE, 1976; Rombouts and Pilnik, 1980), but as yet there is no definite evidence for their action on the primary cell walls of other tissues in vivo, let alone for an enhancement of their concentration or effectiveness upon injury. Addressing the possibility of in vivo conversion of pectic polysaccharides to oligosaccharides, Bishop et al. (1981) attempted to demonstrate the shortening of [3H]pectic fragments (initial mean DP 30-60) when fed, via the transpiration stream of the hypocotyl, into expanded tomato cotyledons. Water-soluble material subsequently extracted from the cotyledons (boiled for 5 min) was analysed by gel-permeation chromatography on Sephadex G-25: the majority of the recovered 3H now appeared to co-elute with di- or trigalacturonides (Ve/Vo=1.9). However, the authors do not state what percentage of the [3H]pectin supplied to the hypocotyl actually reached the cotyledons and thus became available for analysis: if this was a small proportion, it could already have been enriched in the smallest components of the labelled pectic preparation-for example owing to an inability of larger pectic substances to migrate up the xylem-and the results would therefore be misleading. The production of pectic oligosaccharides by contact between aphid saliva and plant cell walls has been demonstrated (Campbell, 1986). The accumulation of pectic fragments in honeydew is due to the presence of pectinase in the saliva (Ma et al., 1990), and may be part of the wound-signalling system that culminates in the synthesis of protease inhibitors (Section II.C.2). It would not, however, be part of the response to simple mechanical injury,

68

S. ALDINGTON AND S. C. FRY

where no saliva is usually present. There seems to have been no convincing evidence for the partial hydrolysis of pectins to oligosaccharides during the mechanical wounding of vegetative plant tissues. 2. Formation of pectic oligosaccharides as elicitors In several cases where pectic oligosaccharins have been found to induce biological responses, initial observations were made with pectic enzymes rather than with pectic oligosaccharides. For instance, (1) pectinase from Rhizopus stolonifer elicited phytoalexins in castor bean (Lee and West, 1981a; Bruce and West, 1982); (2) endo-pectate lyase from Erwinia carotovora induced phytoalexins in soybean (Davis et al., 1984); (3) purified pectinase from Aspergillus niger was implicated in hypersensitivity (Cervone et al., 1987a). In each case, the biological effect was induced by the purified enzyme. Only later was activity found to be associated with oligosaccharides generated by the (in vitro) action of these enzymes. The hypothesis is reasonable but unproven that the enzymes act in vivo to generate sufficient of the right oligosaccharides to induce a biological effect. Pathogenic fungi and bacteria are known to produce a variety of enzymes that degrade the plant cell wall (Cooper, 1976). Pectic enzymes are usually the first to be synthesised when fungi are grown axenically with isolated cell walls of Dicotyledons as the carbon source, and the same may also be true in planta (Cooper, 1984; Cervone et al., 1986; Collmer and Keen, 1986). This is understandable since the first layer of the plant cell met by a pathogen, the middle lamella, is particularly rich in pectins. In particular, maceration of tissue in rot diseases is believed to be due to pectic enzymes. Mutants deficient in pectic enzymes tend to be impaired in their pathogenicity (Mussel1 and Strand, 1976). However, it is difficult to correlate the production of pectic enzymes in axenic cultures with that in planta. Enzyme production in cultures depends on numerous factors such as the isolates used, the composition of the culture medium and the age of the culture. Enzyme activity in planta may differ from that seen in cultures owing to the added effects of plant products (e.g. PGIP-see below). Culture filtrates of two races of Cladosporium fulvum have been shown to contain enzymes active on plant cell walls (S. Aldington and S. C. Fry, unpublished), although this fungus does not penetrate the host cell wall and there is thus no obvious requirement for such enzymes in planta. Wijesundera et al. (1989) compared the enzyme activities produced by Colletotrichum lindemuthianum in axenic culture and in infected bean hypocotyls. The culture filtrate contained a-arabinofuranosidase, a- and P-galactopyranosidase, protease, pectinase and two forms of pectin lyase. The first four of these were also recovered from infected hypocotyls, but there was no extractable

OLIGOSACCHARINS

69

pectinase activity and only one form of pectin lyase. Extraction of pectic enzymes from infected tissues may also be hampered by binding to the cell walls (Skare et al., 1975). Furthermore, the amounts of enzymes produced in planta may be very low and require the use of highly sensitive assays for their detection. Bashan et al. (1985) detected pectinase activity in leaves of susceptible and resistant tomato cultivars within 48 h of infection with Pseudornonas syringae, suggesting that this enzyme may be involved in the primary stages of disease development (see also Bateman and Basham, 1976). Pectic enzymes are also secreted by mutualistic mycorrhizal fungi (Garcia-Romera et al., 1991); it will be interesting to discover how these fungi avoid eliciting the plant’s defence responses. Other enzymes produced later appear to relate to the progressive degradation of the cell wall. As more wall components become accessible, so the pathogen produced corresponding degradative enzymes. Page1 and Heitefuss (1990) found that pectinase, pectate lyase, cellulase, protease and xylanase appeared sequentially 10, 14, 16, 19 and 22h, respectively, after inoculation of potato with Erwinia. Pectinase peaked at 12-14 h whereas pectate lyase peaked at 22 h. In contrast, pathogens of the Gramineae produce xylanase as a key enzyme early during infection (Cooper, 1984; see Section V.C.3), agreeing with the predominance of xylans in grass cell walls. Pectinase (EC 3.2.1.15) catalyses the mid-chain hydrolysis of a-(1-+4)-~-galacturonic acid residues within those parts of a pectin molecule that consist of several contiguous non-esterified residues. Although they are endo-hydrolases, some pectinases have a marked tendency to catalyse multiple attacks close together on a single polysaccharide chain and thus to release large amounts of di- and trisaccharide once one mid-chain glycosidic bond has been hydrolysed (English et al., 1972). Other pectic enzymes may also play a role in pathogenesis. For example, pectin lyase is produced by Aspergillus niger (Albersheim and Killias, 1962) and pectate lyase by Erwinia carotovora (Davis et al., 1984); these both catalyse an elimination reaction resulting in an unsaturated bond between C4 and C5 at the new non-reducing end (A4,5-galacturonicacid). A specific rhamnogalacturonase has also been reported (Schols etal., 1990). Pectin methyl esterases may also be induced (Miller and Macmillan, 1971) whose action can facilitate the action of pectinase. Many pectic enzymes have been found to be inducible by pectins and sometimes by D-galacturonic acid and its derivatives (Cooper, 1976). The release of oligogalacturonides may therefore be simultaneously advantageous to the pathogen (for continuing induction of these enzymes and thus wall degradation) and disadvantageous to it, amplifying the oligosaccharin signal to the host cell. Certain proteins which were isolated from plant cell walls were found to be capable of inhibiting pectinase activity secreted by phytopathogens (Albersheim and Anderson, 1971; Albersheim and Valent, 1974). These

70

S. ALDINGTON AND S. C. FRY

pectinase-inhibiting proteins (PGIPs) have since been found in the cell walls of a variety of plants (see Cervone et a f . , 1987b). PGIPs appear to inhibit only pectinases of fungal origin; plant and bacterial pectinases were not affected although a plant a-D-galacturonidase (EC 3.2.1.67) was inhibited (Cervone et a f . , 1989b, 1990). PGIPs may retard the pectinase-catalysed hydrolysis of polygalacturonic acid (Cervone et a f . , 1987b) leading to the accumulation of oligomers with DP > 4, which can evoke necrosis (Cervone et a f . , 1987a) and induce PAL (De Lorenzo et a f . , 1987). Cervone et a f . (1989a) found that oligogalacturonides (DP > 10) capable of eliciting phytoalexin synthesis were produced after incubation of polygalacturonic acid with pectinase for 1 min. After 15 min, the active fragments had been depolymerized to inactive oligosaccharides (DP < 6). With the same substrate and enzyme concentrations, but in the presence of PGIP, the level of elicitor-active oligogalacturonides continued to increase for 24 h (Fig. 11); only by 48h did oligosaccharides with D P < 6 become predominant. Hence, by increasing the length of time that the active oligogalacturonides are present at the host-fungal interface, PGIPs may convert fungal pectinase from an elicitor destroyer to an elicitor enhancer. The prolonged accumulation of elicitor-active oligosaccharides would also have been assured by the simple expedient of using less pectinase (Cervone et a f . , 1989a). The above picture of the situation may thus be over-simplified. If a localized site of fungal colonization were to emit fungal pectinases, a concentration gradient of enzyme would be created with the lowest concentrations towards the periphery. At some point in this gradient, it might be expected that the pectinase concentration would be just right to hydrolyse wall pectins to moderate sized oligosaccharides without unduly rapidly degrading these further to elicitor-inactive fragments. Might PGIP be more sophisticated in its effects than so far envisaged? It would be advantageous to the plant for the PGIP to permit pectinase to act efficiently on pectic polysaccharides (say, DP> 30), but to inhibit the action of the enzyme on elicitor-active oligogalacturonides (DP == 8-14). If, on the other hand, PGIP efficiently inhibited all action of fungal pectinase, including the hydrolysis of pectic pofysaccharides, few soluble elicitor molecules would be generated in the first place. One way in which PGIP might achieve this goal would be to abolish the propensity of fungal pectinases to catalyse multiple attacks on a single polygalacturonic acid chain (see English et a f . , 1972), i.e. to cause the enzyme to detach from the substrate poly- or oligosaccharide molecule after each hydrolytic event and search for a new substrate molecule rather than simply moving two or three residues along the same chain as normally occurs (English et a f . ,1972). Apparently no work has yet critically tested this possible action of PGIP. De Lorenzo et a f . (1990) showed that the PGIPs from four different cultivars of Phaseolus vufgaris inhibited equally the pectinases from three different races of Coffetotrichumfindemuthianum. The abilities of these

OLIGOSACCH ARINS

1.6

71

Untreatedpolygalacturonic acid

24 h digestion with pectinase and PGIP

o.!31

0.61

Fractions (1 ml)

Fig. 11. Effect of a pectinase inhibitor (PGIP) from Phuseolus on the hydrolysis of polygalacturonic acid (PGA) by Aspergillus pectinase. (a) Untreated PGA; (b) PGA treated for 15 min with pectinase; (c) PGA treated for 24 h with the same concentration of pectinase but in the presence of PGIP (5mol PGIP per mol pectinase). The products were analysed by anionexchange chromatography on “Mono Q” (Pharmacia) with a 0 . 2 -1 . 0 ~NH4HC03gradient and assayed for uronic acid residues by the rn-hydroxybiphenyl method. The elution of authentic mono-, tri- and decagalacturonic acid is indicated by the arrows labelled 1, 3 and 10. From Cervone er ul. (1989b).

fungal enzymes to generate oligogalacturonides capable of eliciting phytoalexins in soybean were equally affected by PGIPs from different cultivars of P . vulgaris whether PGIPs from compatible or incompatible hosts were used. These results suggest that pectinase and PGIP are not involved in race-cultivar specificity although, as the authors point out, they used an in

72

S. ALDINGTON AND S. C. FRY

vitro system, generating active oligogalacturonides from polygalacturonic acid rather than from the more complicated pectins of the primary cell wall. No clear evidence exists showing the presence of both pectic enzymes and pectic oligosaccharins in vivo during the course of infection. Nevertheless, circumstantial evidence suggests that pectic fragments may well arise through the action of pectic enzymes produced by potential pathogens during infection. 3. Formation of pectic oligosaccharides in ripening fruit Various pectic enzyme preparations induce ethylene biosynthesis and pectic oligosaccharides have similar effects (Section II.C.7). Many fruits synthesize pectinase during ripening at about the same time as they produce large amounts of ethylene (Fischer and Bennett, 1991). This has been seen for example in the fruits of avocado (Awad and Young, 1979), peach (Hinton and Pressey, 1974), and tomato (Babbitt et al., 1973), though not in the receptacles (“fruits”) of strawberry (Abeles and Tadeka, 1990) or apples (Abeles and Biles, 1991). Despite the exceptions, it is attractive to suggest that the initial production of small amounts of pectinase early in the ripening process leads to the release of pectic oligosaccharins which evoke ethylene synthesis, which induces the climacteric production of larger amounts of pectinases and other fruit-softening enzymes-an amplification system. This hypothesis is supported by the fact that soluble products of the in vitro autolysis of isolated tomato fruit cell walls can evoke ethylene synthesis in unripe fruit (Brecht and Huber, 1988). 4. Degradation of pectic oligosaccharides in plant tissue The fate of pectic oligosaccharins in vivo has been remarkably little studied. Horn et al. (1989) showed that oligogalacturonides can be taken up by plant cells and recovered intact 40 min later; however, their longevity in the cell is unknown. A priori it would seem advantageous for the plant to have a disposal system for any biological signal so that the response could be switched off when no longer beneficial. In the case of wound signalling, oligogalacturonides would need to be degraded to the monosaccharide for complete inactivation (Bishop et al., 1984), whereas in the case of phytoalexin elicitation, ethylene induction, inhibition of IAA-induced growth etc., degradation to D P < 8 would be adequate. Degradation to D P < 8 could be catalysed by a pectinase, although whether this actually occurs in vegetative tissue in vivo has apparently not been critically studied. Pectinases do not usually yield much galacturonic acid (Rombouts and Pilnik, 1980), so degradation to the monosaccharide would probably require the action of an a-D-galacturonidase (EC 3.2.1.67; “exopolygalacturonase”). Such enzymes occur in plants (Konno and Tsukumi, 1991), but their action on oligogalacturonides in vivo has not been demonstrated. The demonstration of the existence of an enzyme in certain plant tissues does not

73

OLIGOSACCHARINS

establish that it catalyses any particular reaction in vivo, still less that it acts in all plant tissues: witness the stability of XG9 in living spinach cell cultures despite the existence of plant a-D-xylosidases and a-L-fucosidases (Section V.A.2).

C. THE ROLE OF CHITINASES, P-GLUCANASES AND OTHER ENZYMES

1. Enzymes acting on chitin and chitosan Many plant tissues contain chitinases, both constitutively and in increased amounts after infection [they are therefore described as “pathogenesisrelated” (PR) proteins] or after hormonal treatments (Boller, 1985). Chitinases may be directly toxic to fungal cells (Pegg and Vessey, 1973; Schlumbaum et al., 1986), especially in conjunction with p-glucanases (Mauch et al., 1988b). The action of chitinases on the walls of invading fungal hyphae, especially at the growing hyphal tips where the chitin is most exposed, could also generate oligosaccharins of chitin which would elicit defence responses: whether this occurs in vivo has not been established, although the elicitor activity associated with chito-oligosaccharides would suggest that such an action would be very beneficial (Section 1I.E). Ride (1992) has shown that when [3H]chitinis applied to wheat leaves, 2% of the 3H is solubilized to oligosaccharides, including the oligosaccharin, chitotetraose. Prolonged action of plant chitinase could also degrade chitooligosaccharides to the disaccharide, which is too small to be an oligosaccharin (Barber et al., 1989). Ride (1992) has shown, however, that [3H]chitotetraose is relatively stable when applied to wheat leaf tissue. A chitin acetamidase has been detected in cucumber leaves infected by Colletotrichum lagenarium (Siegrist and Kauss, 1990). This enzyme might de-N-acetylate the chitin, making it resistant to chitinase and so suppressing the formation of oligosaccharins. In addition, chitin oligomers might have oligosaccharin activity destroyed (in N-acetyl-dependent bioassaysSection 1I.E.1) or conferred (in bioassays dependent on the free amino group-Section II.E.2) by the action of acetamidase.

-

2. P-D-Giucanases Fungi growing in artificial media often slough p-( 1 4 ),( 1+6)-~-glucans into the culture medium, and some of these polysaccharides are elicitor active (Ayers et al., 1976a,b,c). In addition (and perhaps required for elicitor activity of the polysaccharides), it seems likely that endo-acting p-D-glucanases, which are present in many plant tissues, attack the soluble and wall-bound p-( 1+3) ,( 1+6)-glucans of invading fungi and thereby liberate soluble oligosaccharides such as those with elicitor activity (Section

14

S. ALDINGTON AND S . C. FRY

1I.A). Indeed, Yoshikawa et al. (1981) and Keen and Yoshikawa (1983) have demonstrated that soybean endo+-( 1+3)-glucanases will solubilize elicitor-active P-glucan fragments from hyphal walls of Phytophthora megasperma. The continued action of P-glucanases might (depending on the branching patterns of the substrate) progressively reduce the size of any oligosaccharides initially formed until they were too small to be oligosaccharins. This process would be enhanced by the action of P-glucosidases (i.e. exoglucanases), as has been demonstrated in vitro (Cline and Albersheim, 1981a,b). However, the fate of oligo-P-glucans in vivo has not been investigated experimentally. P-Glucanases could also have a direct damaging effect on fungal hyphae, e.g. by attacking their walls (Mauch et al., 1988a,b); they can also act as antiviral proteins (Edelbaum et al., 1991).

3. Other polysaccharidases Novel endogenous elicitors might arise through the action of the variety of enzymes known to be produced during fungal infection of plant tissues (Section V.B.2). Tomato leaves infected with the fungus Cladosporium fulvum accumulated in their apoplastic fluid a number of enzyme activities which, in vitro, can partially hydrolyse plant cell walls to liberate soluble oligosaccharides. These enzymes can be assayed by a very sensitive method in which the substrate consists of uniformly 14C-labelledwalls isolated from plant cells cultured in the presence of ~-[U-’~C]glucose as sole carbon source (Aldington and Fry, 1992). This assay will detect enzymes that attack any wall polymers (even hitherto unknown polymers) and that yield soluble fragments as products. It was found that tomato leaves infected with an incompatible race of the fungus accumulated within 3-5 days an enzyme activity (probably an endoarabinanase) that solubilized arabinose-rich oligosaccharides from plant cell walls. Leaves infected with a compatible race accumulated this activity later in the infection process (Aldington and Fry, 1992). It will be interesting to determine whether the arabinose-rich oligosaccharides solubilized by this enzyme in vitro possess any oligosaccharin activity in vivo-perhaps activating a defence response.

VI. MOVEMENT OF OLIGOSACCHARINS WITHIN THE PLANT: TRUE HORMONES? Oligosaccharins have a wide range of biological effects, some of which are exerted by exceedingly low concentrations, and they have been associated with systemic responses. They have for this reason frequently been described as “hormone-like,’. A hormone is a substance that acts at a site distant from its site of production, and this implies intercellular transport within the plant. Does this occur?

OLIGOSACCHARINS

7s

A . POSSIBLE TRANSPORT OF XYLOGLUCAN OLIGOSACCHARIDES

It has been speculated that XG9, formed in response t o the auxin that is present in the main shoot, is transported down the main stem to the axillary buds, where it is responsible for preventing their growth, i.e. for imposing apical dominance (Albersheim and Darvill, 1985). This hypothesis would ascribe a classical hormonal role to XG9. Exogenously supplied nonasaccharide must reach its site of action in the pea stem segment for its effect to be noted. The target cells for the action of XG9 are unknown but it seems reasonable to suggest that they might be those of the tissue most responsive to auxin-stimulated elongation, i.e. the epidermis. If it is further assumed that the relatively large, hydrophilic, XG9 molecule cannot readily penetrate the cuticle and is excluded from the symplast by the plasma membrane (Baydoun and Fry, 1989), and yet can inhibit the growth of 6-10mm pea stem segments (York et af., 1984; McDougall and Fry, 1988; Lorences et al., 1990), it can be suggested that XG9 is capable of moving distances of at least a few millimetres through the apoplast of a stem. However, direct studies of the movement of exogenous XG9 have not been reported.

B. NON-TRANSPORT OF WOUND SIGNALS

The discovery that pectic oligosaccharides possess wound-signal activity, evoking the biosynthesis of protease inhibitors, prompted Bishop et af. (1981) to conclude that these oligosaccharides are the protease inhibitor inducing factor (PIIF), previously defined as the long-distance wound hormone (Green and Ryan, 1973). However, while it is plausible that endogenous pectic oligosaccharides might increase in concentration at injury sites, and well-established that exogenous pectic oligosaccharides can evoke protease inhibitor synthesis in excised leaves, the missing link in the argument is whether pectic oligosaccharides can be transported from injured to uninjured leaves. Such transport would require the oligosaccharides to move both basipetally (out of the injured leaf, presumably in the phloem) and acropetally (into the uninjured leaf, possibly in the xylem). Even the smallest pectic oligosaccharides are too large to traverse such distances in the time required purely by diffusion. To provide an estimate of the velocity at which PIIF moves within the tomato plant, Green and Ryan (1972, 1973) inflicted a standardized injury on the lamina of one leaf and then, after an additional time t (0-8 h), they excised the injured leaf close to the base of its petiole, thus preventing further export of PIIF. The response (accumulation of protease inhibitor in a neighbouring uninjured leaf) at 48 h was taken at a measure of the amount of PIIF that had moved the few centimetres from the injury site on the

76

S. ALDINGTON AND S. C. FRY

lamina past the excision point in the petiole within time t. Half-maximal response was observed when t was -45min. It was concluded that PIIF covered this basipetal leg of its journey at 3-5 cm h-’. However, this is a minimum estimate: the 45 min delay may have been due to the time taken by the injured lamina to produce enough PIIF to evoke a response, each individual PIIF molecule being instantly exported at a velocity much greater than 3-5 cm h-I. To investigate whether pectic oligo- and polysaccharides can indeed move from leaf t o leaf, as would be required if they were PIIF, three different preparations of soluble, relatively low molecular weight [U-14C]rhamnogalacturonans were applied to small injury sites on the leaves of potted tomato seedlings and the re-distribution of the radioactivity was monitored by autoradiography (Baydoun and Fry, 1985). Most of the 14Cremained at the injury site. A small amount of it moved acropetally towards the tips and margins of the treated leaf, in a pattern characteristic of xylem transport (Canny, 1990). No detectable I4C (corresponding to S 1ng of pectic material) moved basipetally. The lack of export of applied pectic fragments was confirmed with [3H]oligogalacturonides of DP 6 9 and 10-14 (Baydoun and Fry, 1985). The treated lamina tissue had been injured so that the 14C-and 3H-labelled pectic substances were certainly brought into contact with cell surfaces. Wounds are clearly valid application sites for testing the transport of a putative wound hormone (even if not of other hormones!). The acropetal movement of some of the labelled pectic material via the xylem shows that the applied pectic substances came into contact with vascular tissue. [‘4C]Sucrose applied in the same way was efficiently translocated away from the injury site, both acropetally and basipetally, so that phloem loading must have been possible under the experimental conditions used. We conclude that pectic oligosaccharides cannot move basipetally in tomato plants, and that, if they are involved in the induction of protease inhibitor biosynthesis in response to injury, they act locally in the immediate vicinity of the wound rather than as long-distance wound hormones. It is for this reason that we use the term “wound signal” rather than “PIIF” to describe pectic oligosaccharins; PIIF is the long-distance hormone by which injured leaves communicate their plight to neighbouring uninjured leaves and its identity remains a mystery.

C. TRANSPORT OF ELICITORS?

Since the hypersensitive response tends to be highly localised, any oligosaccharins involved in its evocation seem, apriori, likely to remain close to the site of infection and thus not to be hormones in the normal sense of the word. However, not all defence responses are as localized as the hypersensitive response. For example, phytoalexin synthesis often occurs over a

OLIGOSACCHARINS

77

large proportion of the area of an infected leaf (Holliday et al., 1981; Bailey and Mansfield, 1982) or of locally elicited tissue (Dixon et al., 1983b), and localized infection can sometimes trigger systemic responses, e.g. the accumulation of peroxidase (KuC, 1982), HRGP-mRNA (Showalter et al., 1985) and chitinase (Roby et al., 1988): in these cases, more mobile signals would be implicated. It is possible that some oligosaccharins may be mobile enough to serve this role (though probably not the oligogalacturonides-see Section V1.B); alternatively, oligosaccharides may prompt signals which can be rapidly transmitted to more distant parts of the plant. Ethylene is arguably one such signal; Ecker and Davis (1987) suggested that locally produced ethylene is suited to activate defence-response genes both nearby and at a distance.

VII.

CONCLUDING REMARKS

Oligosaccharins have been implicated in a wide range of botanical processes. Some of their proposed roles, especially as elicitors of phytoalexin synthesis, are well established, even if there are still conspicuous gaps in our knowledge of the oligosaccharins responsible-such as their mode of action, whether they occur naturally and (if so) how they are generated, transported and degraded. Some other roles are still much more speculative and even the existence of the biological response in artificial bioassays requires definitive confirmation or negation. However, despite all the gaps, the oligosaccharin concept is clearly here to stay, and, because of the gaps, the coming decade will no doubt be an exciting period in the history of the oligosaccharin concept.

ACKNOWLEDGEMENTS The authors are very grateful to the EEC for the award of a “BRIDGE” contract enabling them to work in this area.

REFERENCES Abeles, F. B . and Biles, C. L. (1991). Cellulase activity in developing apple fruits. Scientia Horticultura 47, 77-87. Abeles, F. B. and Takeda, F. (1990). Cellulase activity and ethylene in ripening strawberry and apples. Scientia Horticultura 42, 269-275. Adam, A., Farkas,T., Somlyai, G . , Hevesi, M . and Kiraly, Z . (1989). Consequence of 0 2 . - generation during a bacterially induced hypersensitive reaction in tobacco; deterioration of membrane lipids. Physiological and Molecular Plant Pathology 34, 13-26. Albersheim, P. and Anderson, A. J. (1971). Proteins from plant cell walls inhibit

78

S. ALDINGTON AND S. C. FRY

polygalacturonase secreted by plant pathogens. Proceedings of the National Academy of Science, USA 68, 1815-1819. Albersheim, P. and Darvill, A. G. (1985). Oligosaccharins. ScientiJc American 253, 44-50. Albersheim, P. and Killias, U. (1962). Studies relating to the purification and properties of pectin transeliminase. Archives of Biochemistry and Biophysics 97,107-115. Albersheim, P. and Valent, B. S. (1974). Host-pathogen interactions. VII. Plant pathogens secrete proteins which inhibit enzymes of the host capable of attacking the pathogen. Plant Physiology 53, 684-687. Albersheim, P. and Valent, B. S. (1978). Host-pathogen interactions in plants: plants when exposed to oligosaccharides of fungal origin defend themselves by accumulating antibiotics. Journal of Cell Biology 78, 627443. Albersheim, P., Darvill, A. G . , McNeil, M., Valent, B. S., Hahn, M. G., Lyon, G., Sharp, J. K. and Desjardins, A. E . (1981) Structure and function of complex carbohydrates active in regulating plant-microbe interactions. Pure and Applied Chemistry 53, 79-88. Albersheim, P., Darvill, A. G., McNeil, M., Valent, B. S . , Sharp, J. K . , Nothnagel, E. A., Davis, K. R. and Yamazaki, N. (1983). Oligosaccharins: naturallyoccurring carbohydrates with biological regulatory functions. In “Structure and Function of Plant Genomes” (0.Ciferri and L. Dure, eds), pp. 293-312. Plenum, New York. NATO AS1 on the Structure and Function of the Plant Genome, Lake Garda, Italy. Albersheim, P., Darvill, A. G . , Sharp, J. K., Davis, K. R. and Doares, S. H. (1986). Studies on the role of carbohydrates in host-microbe interactions. In “Recognition in Microbe-Plant Symbiotic and Pathogenic Interactions” (B. Lugtenberg, ed.), pp. 297-309. Springer-Verlag, Berlin. NATO AS1 Series, H4. Aldington, S. and Fry, S. C. (1992). Plant cell wall-lysing enzymes in the apoplast of Fulvia fulva-infected tomato leaves. Canadian Journal of Botany (in press). Aldington, S., McDougall, G. J. and Fry, S. C. (1991). Structure-activity relationships of biologically active oligosaccharides. Plant, Cell and Environment 14, 625-636. Anderson, J. D., Mattoo, A. K. and Lieberman, M. (1982). Induction of ethylene biosynthesis in tobacco leaf discs by cell wall digesting enzymes. Biochemical and Biophysical Research Communications 107, 588-596. Apostol, I . , Low, P. S., Heinstein, P. F., Stipanovic, R.D. and Altman, D. W. (1987). Inhibition of elicitor-induced phytoalexin formation in cotton and soybean cells by citrate. Plant Physiology 84, 12761280. Apostol, I., Heinstein, P. F. and Low, P. S. (1989). Rapid stimulation of an oxidative burst during elicitation of cultured plant cells. Plant Physiology 90, 109-116. Asamizu, T., Nakayama, N. and Nishi, A. (1984). Pectic polysaccharides in carrot cells growing in suspension culture. Planta 160, 469473. Aspinall, G. O., Molloy, J . A. and Craig, J. W. T. (1969). Extracellular polysaccharides from suspension-cultured sycamore cells. Canadian Journal of Biochemistry 47, 1063-1070. Aspinall, G. 0. and Cottrell, I . W. (1970). Lemon-peel pectin. 11. Isolation of homogenous pectins and examination of some associated polysaccharides. Canadian Journal of Chemistry 48, 1283-1289. Atkinson, M. M., Huang, J.-S. and Knopp, J. A. (1985). The hypersensitive reaction of tobacco to Pseudomonas syringae pv. pisi: activation of a plasmalemma K+/H+ exchange mechanism. Plant Physiology 79, 843-847. ~

OLIGOSACCHARINS

79

Atkinson, M. M., Baker, C. J. and Collmer, A . (1986). Transient activation of plasmalemma K+ efflux and H+ influx in tobacco by a pectate lyase isozyme from Erwinia chrysanthemi. Plant Physiology 82, 142-146. Awad, M. and Young, R. E. (1979). Postharvest variation in cellulase, polygalacturonase, and pectinmethylesterase in avocado (Persea americana Mill, cv. Fuerte) fruits in relation to respiration and ethylene production. Plant Physiology 64, 306308. Ayers, A. R., Ebel, J., Finelli, F., Berger, N. and Albersheim, P. (1976a). Hostpathogen interactions. IX. Quantitative assays of elicitor activity and characterization of the elicitor present in the extracellular medium of cultures of Phytophthora megasperma var. sojae. Plant Physiology 57, 751-759. Ayers, A. R., Ebel, J., Valent, B. S. and Albersheim, P. (1976b). Host-pathogen interactions. X. Fractionation and biological activity of an elicitor isolated from the mycelial walls of Phytophthora megasperma var. sojae. Plant Physiology 57,760-765. Ayers, A. R., Valent, B. S., Ebel, J. and Albersheim, P. (1976~).Host-pathogen interactions. XI. Composition and structure of wall-released elicitor fractions. Plant Physiology 57,766774. Babbitt, J. K., Powers, M. J. and Patterson, M. E. (1973). Effects of growthregulators on cellulase, polygalacturonase, respiration, color, and texture of ripening tomatoes. Journal of the American Society for Horticultural Science 98,77-81. Backe, M. A. (1910). Recherches sur I’isomaltol. Comptes Rendus de I’Acadkmiedes Sciences Paris Skrie III 151, 78-80. Badwey, J. A. and Karnovsky, M. L. (1980). Active oxygen species and the functions of phagocytic leukocytes. Annual Review of Biochemistry 49, 695-726. Baev, N., Endre, G., Petrovics, G., Banfalvi, Z. and Kondorosi, A. (1991). Six nodulation genes of nod box locus 4 in Rhizobium meliloti are involved in nodulation signal production: nodM codes for D-glucosamine synthetase. Molecular and General Genetics 228, 113-124. Bailey, J. A. (1982). Mechanisms of phytoalexin accumulation. In “Phytoalexins” (J. A. Bailey and J. W. Mansfield, eds), pp. 289-318. Halsted, Wiley, New York. Bailey, J. A. and Mansfield, J. W. (1982). “Phytoalexins”. Halsted, Wiley, New York. Bailey, R. W. and Pridham, J. B. (1962). The separation and identification of oligosaccharides. Chromatographical Reviews 4, 114-136. Baker, C. J., Atkinson, M. M. and Collmer, A. (1985) Effect of cell wall fragments released by pectate lyase on the hypersensitive response in tobacco. Phytopathology 75, 1372-1373 (Abstr). Baker, C. J., Atkinson, M. M., Roy, M. A. and Collmer, A. (1986). Inhibition of the hypersensitive response in tobacco by pectate lyase. Physiological and Molecular Plant Pathology 29,217-225. Baldwin, E. A. and Biggs, R. H. (1988). Cell-wall lysing enzymes and products of cell wall digestion elicit ethylene in citrus. Physiologia Plantarum 73, 5 8 4 4 . Baldwin, E. A. and Pressey, R. (1988). Tomato polygalacturonase elicits ethylene production in tomato fruit. Journal of the American Society for Horticultural Science 113,92-95. Baldwin, E. A. and Pressey, R. (1990). Exopolygalacturonase elicits ethylene production in tomato. HortScience 251, 779-780. Barber, M. S. and Ride, J. P. (1988). A quantitative assay for induced lignification in

80

S. ALDINGTON AND S. C. FRY

wounded wheat leaves and its use to survey potential elicitors of the response. Physiological and Molecular Plant Pathology 32, 185-197. Barber, M. S., Bertram, R. E. and Ride, J. P. (1989). Chitin oligosaccharides elicit lignification in wounded wheat leaves. Physiological and Molecular Plant Pathology 34, 3-12. Barbier-Brygoo, H., Ephritikhine, G., Klambt, D., Maurel, C., Palme, K., Schell, J. and Guern, J. (1991). Perception of the auxin signal at the plasma membrane of tobacco mesophyll protoplasts. The Plant Journal 1, 83-93. Barrett, A. J. and Northcote, D. H. (1965). Apple fruit pectic substances. Biochemical Journal 94,617-627. Basham, H. G. and Bateman, D. F. (1975a). Relationship of cell death in plant tissue treated with homogeneous endopectate lyase to cell wall degradation. Physiological Plant Pathology 5,249-262. Basham, H. G. and Bateman, D. F. (1975b). Killing of plant cells bypecticenzymes: the lack of direct injurious interaction between pectic enzymes or their soluble reaction products and plant cells. Phytopathology 65, 141-153. Bashan, Y., Okon, Y. and Henis, Y. (1985). Detection of cutinases and pectic enzymes during infection of tomato by Pseudomonas syringae pv. tomato. Phytopathology 75, 94&945. Bateman, D. F. (1976) Plant cell wall hydrolysis by pathogens. In “Biochemical Aspects of Plant-Parasite Relationships” (J. Friend and D. R. Threlfall, eds), pp. 79-103. Academic Press, New York. Bateman, D. F. and Basham, H. G. (1976). Degradation of plant cell walls and membranes by microbial enzymes. In “Physiological Plant Pathology”, Encyclopedia of Plant Physiology NS” (R. Heitefuss and P. H. Williams, eds), Vol. 4, pp. 316355. Springer-Verlag, Berlin. Bauer, W. D., Talmadge, K. W., Keegstra, K. and Albersheim, P. (1973). The structure of plant cell walls. 11. The hemicellulose of the walls of suspensioncultured sycamore cells. Plant Physiology 51, 174-187. Baydoun, E. A.-H. and Fry, S. C. (1985). The immobility of pectic substances in injured tomato leaves and its bearing on the identity of the wound hormone. Planta 165,269-276. Baydoun, E. A.-H. and Fry, S. C. (1989). In vivo degradation and extracellular polymer-binding of xyloglucan nonasaccharide, a natural anti-auxin. Journal of Plant Physiology 134,453459. Beardmore, M., Ride J. P.and Granger, J. W. (1983). Cellular lignification as a factor in the hypersensitive resistance of wheat to stem rust. Physiological Plant Pathology 22,209-220. Bell, A. A. (1981). Biochemical mechanisms of disease resistance. Annual Review of Plant Physiology 32,2141. Bellincampi, D., Branca, C., De Lorenzo, G., Desiderio, A., Ricci, A., Salvi, G. and Cervone, F. (1990). a-D-Ohgogalacturonides interfere with the auxin activity in pea and tobacco explants. Giornale Botanica Italian0 124, 72-73. Benhamou, N., Gilboa-Garber, N., Trudel, J. and Asselin, A. (1988). A new lectin-gold complex for ultrastructural localisation of galacturonic acids. Journal of Histochemistry and Cytochemistry 36,1403-1411. Benhamou, N., Chamberland, H. and Pauze, F. J. (1990). Implication of pectic components in cell surface interactions between tomato root cells and Fusarium oxysporum f. sp. radicis-lycopersici. A cytochemical study by means of a lectin with polygalacturonic acid binding specificity. Plant Physiology 92, 995-1003. Benhamou, N., Lafitte, G., Barthe, J.P. and Esquerrk-TugayC, M.-T. (1991). First localization of fungal endopolygalacturonase in Colletotrichum-

OLIGOS ACCHARINS

81

infected bean tissues, and correlation with pectin breakdown. Plant Physiology 96S, 55. Bernasconi, P., Jolles, P. and Pilet, P.-E. (1986). Increase of lysozyme and chitinase in Rubus calli caused by infection and some polymers. Plant Science 44,79-83. Bezukladnikov, P. W. and Elyakova, L. A . (1988). Preparation of 14C-labelled (1-+3)-~-~-gluco-oligosaccharidesby transglycosylation. Carbohydrate Research 184, 26&270. Bezukladnikov, P. W. and Elyakova, L. A. (1990). Transglycosylation and multiple attack of endo-( 1+3)-P-~-glucanaseL-IV from Spisula sachalinensis: a new approach to the evaluation of the degree of multiple attack on polysaccharides. Carbohydrate Research 203, 119-127. Bird, P. M. and Ride, J. P. (1981). The resistance of wheat to Septoria nodorum: fungal development in relation to host lignification. Physiological Plant Pathology 19, 289-299. Bishop, P. D. and Ryan, C. A. (1987). Plant cell wall polysaccharides that activate natural plant defenses. Methods in Enzymology 138, 715-724. Bishop, P. D., Makus, D. J., Pearce, G. and Ryan, C. A . (1981). Proteinase inhibitor-inducing factor activity in tomato leaves resides in oligosaccharides enzymically released from cell walls. Proceedings of the National Academy of Sciences, USA 78,3536-3540. Bishop, P. D., Pearce, G., Bryant, J. E. and Ryan, C. A. (1984). Isolation and characterization of the proteinase inhibitor-inducing factor from tomato leaves. Identity and activity of poly- and oligogalacturonide fragments. Journal of Biological Chemistry 259, 13 172-13 177. Blake, D. A. and Goldstein, I. J. (1980). Resolution of nucleotide sugars and oligosaccharides by lectin affinity chromatography. Analytical Biochemistry 102, 103-109. Blum, U. and Dalton, B. R. (1985). Effects of ferulic acid an allelopathic compound on leaf expansion of cucumber seedlings grown in nutrient culture. Journal of Chemical Ecology 11,279-301. Boffey, S. A. and Northcote, D. H. (1975). Pectin synthesis during the regeneration of plasmolysed tobacco leaf cells. Biochemical Journal 150,433-440. Boller, T. (1985). Induction of hydrolases as a defence reaction against pathogens. In “Cellular and Molecular Biology of Plant Stress” (J. L. Key and T. Kosuge, eds), pp. 247-262. A. R. Liss, New York. Boller,T., Gehri, A . , Mauch, F. and Vogeli, U. (1983). Chitinase in bean leaves: induction by ethylene, purification, properties, and possible function. Planta 157, 22-31. Bolwell, G. P. (1984). Differential patterns of arabinosylation by membranes of suspension-cultured cells of Phaseolus vulgaris (French bean) after subculture or elicitation. Biochemical Journal 222, 427435. Bolwell, G. P. (1986). Microsomal arabinosylation of polysaccharide and elicitorinduced carbohydrate-binding glycoprotein in French bean. Phytochemistry 25, 1807-1813. Bolwell, G. P. and Dixon, R. A. (1986). Membrane-bound hydroxylases in elicitortreated bean cells. Rapid induction of the synthesis of prolyl hydroxylase and a putative cytochrome P-450. European Journal of Biochemistry 159, 163-169. Bolwell, G. P., Robbins, M. P. and Dixon, R. A. (1985a). Elicitor-induced prolyl hydroxylase from French bean (Phaseolus vulgaris). Localization, purification and properties. Biochemical Journal 229, 693-699. Bolwell, G. P., Robbins, M. P. and Dixon, R. A. (1985b). Metabolic changes in elicitor-treated bean cells. Enzymic responses associated with rapid changes in cell wall components. European Journal of Biochemistry 148,571-578.

82

S. ALDINGTON AND S. C. FRY

Bolwell, G. P., Bell, J. N., Cramer, C. L., Schuch, W., Lamb, C. J. and Dixon, R. A . (1985~).L-Phenyalanine ammonia-lyase from Phaseolus vulgaris. Characterisation and differential induction of multiple forms from elicitor-treated cell suspension cultures. European Journal of Biochemistry 149,411419, Bostock, R. M., KuC, J. and Laine, R. A. (1981) Eicosapentaenoic and arachidonic acids from Phytophthora infestans elicit fungitoxic sesquiterpenes in the potato. Science 212, 67-69. Bostock, R. M., Laine, R. A. and KuC, J. A. (1982). Factors affecting the elicitation of sesquiterpenoid phytoalexin accumulation by eicosapentaenoic and arachidonic acids in potato. Plant Physiology 70, 1417-1424. Bowles, D. J. (1990). Defense-related proteins in higher plants. Annual Reviews in Biochemistry 59, 873-907. Bowles, D. (1991). Signalling in plants: the wound response of plants. Current Biology 1, 165-167. Branca, C. A., De Lorenzo, G. and Cervone, F. (1988). Competitive inhibition of the auxin-induced elongation by a-D-ohgogalacturonides in pea stem segments. Physiologia Plantarum 72,499-504. Brecht, J. K. and Huber, D. J. (1988). Products released from enzymically active cell wall stimulate ethylene production and ripening in preclimacteric tomato (Lycopersicon esculentum Mill.) fruit. Plant Physiology 88,1037-1041. Brett, C. T. (1978). Synthesis of P-(l-+3)-glucan from extracellular uridine diphosphate glucose as a wound response in suspension-cultured soybean cells. Plant Physiology 62,377-382. Broekaert, W. F. and Peumans, W. J. (1988). Pectic polysaccharides elicit chitinase accumulation in tobacco. Physiologia Plantarum 74, 74&744. Bruce, R. J. and West, C. A. (1982). Elicitation of casbene synthetase activity in castor bean: the role of pectic fragments of the cell wall in elicitation by a fungal endo-polygalacturonase. Plant Physiology 69, 1181-1 188. Bruce, R. J. and West, C. A.(1989). Elicitation of lignin biosynthesis and isoperoxidase activity by pectic fragments in suspension-cultures of castor bean. Plant Physiology 91, 889-897. Bucheli, P.,Doares, S. H., Albersheim, P. and Darvill, A. (1990). Host-pathogen interactions XXXVI. Partial purification and characterization of heat-labile molecules secreted bv the rice blast pathogen that solubilize plant cell wall fragments that kill plint cells. Physioligical i n d Molecular Plan; Pathology 36, 159- 173. Burdett, A . N. (1970). The cause of bent neck in cut roses. Journal of the American Society for Horticultural Science 95, 427431. Callow, J . A . (1984). Cellular and molecular recognition between higher plants and fungal pathogens. In “Encyclopedia of Plant Physiology NS” (H. F. Linskens and J. Heslop-Harrison, eds), Vol. 17, pp. 212-237. Springer-Verlag, Berlin. Campbell, A. D. and Labavitch, J. M. (1991). Induction and regulation of ethylene biosynthesis by pectic oligomers in cultured pear cells. Plant Physiology 97, 699-705. Campbell, B. C. (1986). Host-plant oligosaccharins in the honeydew of Schizaphis grarninum (Rondani) (Insecta, Aphididae). Experientia 42, 451452. Canny, M.J. (1990). What becomes of the transpiration stream? New Phytologist 114, 341-368. Cervone, F., De Lorenzo, G., Salvi, G. and Camardella, L. (1986). Molecular evolution of fungal polygalacturonase. In “Biology and Molecular Biology of Plant-Pathogen Interactions”. NATO AS1 Series Vol. H1, pp. 385-392. Ed. J. A. Bailey. Springer-Verlag, Berlin. Cervone, F., De Lorenzo, G., Degrh, L. and Salvi, G. (1987a). Elicitation of

OLIGOSACCHARINS

83

necrosis in Vigna unguiculata Walp. by homogeneous Aspergillus niger endopolygalacturonase and by a-D-galacturonate oligomers. Plant Physiology 85, 626-630. Cervone, F., De Lorenzo, G., Degra, L., Salvi, G. and Bergami, M. (1987b). Purification and characterization of a polygalacturonase-inhibiting protein from Phaseolus vulgaris L. Plant Physiology 85, 631-637. Cervone, F., De Lorenzo, G., D’Ovidio, R., Hahn, M. G., Ito, Y . , Darvill, A. G. and Albersheim, P. (1989a). Phytotoxic effects and phytoalexin-elicitor activity of microbial pectic enzymes. In “Phytotoxins and Plant Pathogenesis” (A. Ballio, R. D. Durbin and A. Graniti, eds), pp. 473477. Springer-Verlag, Berlin. NATA AS1 Series. Cervone, F., Hahn, M . G., De Lorenzo, G., Darvill, A. G . and Albersheim, P. (1989b). Host-pathogen interactions. XXXIII. A plant protein converts a fungal pathogenesis factor into an elicitor of plant defence responses. Plant Physiology 90,542-548. Cervone, F., De Lorenzo, G., Pressey, R., Darvill, A. G. and Albersheim, P. (1990). Can Phaseolus PGIP inhibit pectic enzymes from microbes and plants? Phytochemistry 29, 447-449. Chai, H. B. and Doke, N. (1987). Superoxide anion generation: a response of potato leaves to infection with Phytophthora infestans. Phytopathology 77, 645-649. Chappell, J., Hahlbrock, K. and Boller, T. (1984). Rapid induction of ethylene biosynthesis in cultured parsley cells by fungal elicitor and its relationship to the induction of phenylalanine ammonia-lyase. Planta 161, 475480. Chappell, J., Vonlanken, C. and Vogeli, U. (1991). Elicitor-inducible 3-hydroxy-3methylglutaryl coenzyme A reductase activity is required for sesquiterpene accumulation in tobacco cell suspension cultures. Plant Physiology 97, 693698. Cheong, J.-J. and Hahn, M. G. (1991). A specific, high-affinity binding site for the hepta-P-glucoside elicitor exists in soybean membranes. Plant Cell 3,137-147. Cheong, J.-J., Birberg, W., Fugedi, P., Pilotti, A., Garegg, P. J., Hong, N . , Ogawa, T. and Hahn, M. G. (1991). Structure-activity relationships of oligo-Pglucoside elicitors of phytoalexin accumulation in soybean. Plant Cell 3, 127-136. Cline, K. and Albersheim, P. (1981a). Host-pathogen interactions. XVI. Purification and characterization of a P-glucosyl hydrolase/transferase present in the walls of soybean cells. Plant Physiology 68,207-220. Cline, K. and Albersheim, P. (1981b). Host-pathogen interactions. XVII. Hydrolysis of biologically-active fungal glucans by enzymes isolated from soybean cells. Plant Physiology 68, 221-228. Cline, K . , Wade, M. and Albersheim, P. (1978). Host-pathogen interactions. XV. Fungal glucans which elicit phytoalexin accumulation in soybean also elicit the accumulation of phytoalexins in other plants. Plant Physiology 62, 918-921. Collmer, A. and Keen, N. T. (1986). The role of pectic enzymes in plant pathogenesis. Annual Reviews in Phytopathology 24, 383409. Constabel, F. and Eilert, U . (1986). Elicitation of product accumulation. ZAPTC Newsletter, no. 50, 2-8. Cooper, R. M. (1976). Regulation of synthesis of cell wall-degrading enzymes of plant pathogens. In “Cell Wall Biochemistry Related to Specificity in Host Plant-Pathogen Interactions” (B. Solheim and J . Raa, eds), pp. 163-206. Universitetsforlaget, Oslo. Cooper, R. M . (1984). The role of cell wall-degrading enzymes in infection and damage. In “Plant Diseases: Infection Damage and Loss” (R. K. S. Wood and G. J. Jellis, eds), pp. 13-27. Blackwell, Oxford.

84

S. ALDINGTON AND S. C. FRY

Corcoran, M. R., Geissman, T. A. and Phinney, B. 0. (1972). Tannins as gibberellin antagonists. Plant Physiology 49, 323-330. Cordewener, J., Booij, H., Van der Zandt, H., Van Engelen, F., Van Kammen, A. and De Vries, S. (1991). Tunicamycin-inhibitedcarrot somatic embryogenesis can be restored by secreted cationic peroxidase isoenzymes. Planta 184, 478-486. Cosio, E. G., Popperl, H., Schmidt, W. E. and Ebel, J. (1988). Highaffinity binding of fungal 9-glucan fragments to soybean (Glycine m a L.) microsomal fractions and protoplasts. European Journal of Biochemistry 175,309-315. Cosio, E. G., Frey, T., Verduyn, R., Van Boom, J. and Ebel, J. (1990a). Highaffinity binding of a synthetic heptaglucoside and fungal glucan phytoalexin elicitors to soybean membranes. FEBS Letters 271,223-226. Cosio, E. G., Frey, T. and Ebel, J. (1990b). Solubilization of soybean membrane binding sites for fungal P-glucans that elicit phytoalexin accumulation. FEBS Letters 264,235-238. Cramer, C. L., Ryder, T. B., Bell, J. N. and Lamb, C. J. (1985). Rapid switching of plant gene expression induced by fungal elicitor. Science 227,1240-1243. Danks, M. L., Fletcher, J. S. and Rice, E. L. (1975). Effects ofphenolic inhibitors on growth and metabolism of gl~cose-UL-'~Cin Paul's Scarlet Rose cellsuspension cultures. American Journal of Botany 62, 311-317. Darvill, A. G., McNeil, M. and Albersheim, P. (1978). Structure of plant cell walls VIII. A new pectic polysaccharide. Plant Physiology 62,41&422. Darvill, A. G. and Albersheim, P. (1984). Phytoalexins and their elicitors. A defense against microbial infection in plants. Annual Reviews in Plant Physiology 35, 243-275. Davis, K. R., Lyon G. D., Darvill, A. G. and Albersheim, P. (1984). Host-pathogen interactions XXV. Endopolygalacturonic acid lyase from Erwinia carotovora elicits phytoalexin accumulation by releasing plant cell wall fragments. Plant Physiology 74, 52-60. Davis, K. R. and Hahlbrock, K. (1987). Induction of defense responses in cultured parsley cells by plant cell wall fragments. Plant Physiology 85, 12861290. Davis, K. R., Darvill, A. G. and Albersheim, P. (1986a). Host-pathogen interactions. XXX. Characterization of elicitors of phytoalexin accumulation in soybean released from soybean cell walls by endopolygalacturonic acid lyase. Zeitschrift fur Naturforschung 41c, 39-48. Davis, K. R., Darvill, A. G. and Albersheim, P. (1986b). Host-pathogen interactions XXXI. Several biotic and abiotic elicitors act synergistically in the induction of phytoalexin accumulation in soybean. Plant Molecular Biology 6, 23-32. Davis, K. R., Darvill, A. G., Albersheim, P. and Dell, A. (1986~).Host-pathogen interactions. XXIX. Oligogalacturonides released from sodium polypectate by endopolygalacturonic acid lyase are elicitors of phytoalexins in soybean. Plant Physiology 80,568-577. De Lorenzo, G., Ranucci, A., Bellincampi, D., Salvi, G. and Cervone, F. (1987). Elicitation of phenylalanine ammonia lyase in Daucus carota by oligogalacturonides released from sodium polypectate by homogeneous polygalacturonase. Plant Science 51, 147-150. De Lorenzo, G., Ito, Y., D'Ovidio, R., Cervone, F., Albersheim, P. and Darvill, A. G. (1990). Host-pathogen interactions. XXXVII. Abilities of the polygalacturonase-inhibiting proteins from four cultivars of Phaseolus vulgaris to inhibit the endopolygalacturonases from three races of Colletotrichum lindemuthianum. Physiological and Molecular Plant Pathology 36,421-435. de Vries, S. C., Booij, H., Janssens, R., Vogels, R., Saris, L., Loschiavo, F., Terzi,

OLIGOSACCHARINS

85

M. and Van Kammen, A. (1988). Carrot somatic embryogenesis depends on the phytohormone-controlled presence of correctly glycosylated extracellular proteins. Genes and Development 2,462-476. de Wit, P. J. G. M. and Roseboom, P. H. M. (1980). Isolation, partial characterization and specificity of glycoprotein elicitors from culture filtrates, mycelium and cell walls of Cladosporiumfulvum (syn. Fulvia fulva). Physiological Plant Pathology 16,391-408. de Wit, P. J. G. M. and Kodde, E. (1981). Further characterization and cultivarspecificity of glycoprotein elicitors from culture filtrates and cell walls of Cladosporium fulvum (syn. Fulvia fulva). Physiological Plant Pathology 18, 297-3 14. de Wit, P. J. G. M. and Spikman, G. (1982). Evidence for the occurrence of race and cultivar-specificelicitors of necrosis in intercellular fluids of compatible interactions of Cladosporium fulvum and tomato. Physiological Plant Pathology 21,l-11. Dean, J. F. D., Gamble, H. R. and Anderson, J. D. (1989). The ethylene biosynthesis-inducing xylanase: its induction in Trichoderma viride and certain plant pathogens. Phytopathology 79, 1071-1078. Dean, J. F. D., Gross, K. C. and Anderson, J. D. (1991). Ethylene biosynthesisinducing xylanase. 111. Product characterization. Plant Physiology 96, 571576. Deuel, H. and Stutz, E. (1958). Pectic substances and pectic enxymes. Advances in Enzymology 20, 341-382. Deverall, B. J. and Deakin, A. L. (1987). Genetic tests of the basis of wheat cultivar selectivity in symptom elicitation by preparations from rust pathogens. Physiological and Molecular Plant Pathology 30, 225-232. Dey, P. M. (ed.) (1990). “Methods in Plant Biochemistry”, Vol. 2, Carbohydrates. Academic Press, London. Diaz, C. L., Melchers, L. S., Hooykaas, P. J. J., Lugtenberg, B. J. J. and Kijne, J. W. (1989). Root lectin as a determinant of host-plant specificity in the Rhizobium-legume symbiosis. Nature 338, 579-581. Dietrich, A., Mayer, J. E. and Hahlbrock, K. (1990). Fungal elicitor triggers rapid, transient, and specific protein phosphorylation in parsley cell suspension cultures. Journal of Biologial Chemistry 265, 63604368. Dixon, R. A. (1986). The phytoalexin response: elicitation, signallingand control of host gene expression. Biological Reviews 61, 239-291. Dixon, R. A. and Lamb, C. J. (1990). Molecular communication in interactions between plants and microbial pathogens. Annual Review of Plant Physiology and Plant Molecular Biology 41,339-367. Dixon, R. A., Dey, P. M. and Lamb, C. J. (1983a). Phytoalexins: enzymology and molecular biology. Advances in Enzymology and Related Areas of Molecular Biology 53, 1-136. Dixon, R. A., Dey, P. M., Lawton, M. A. and Lamb, C. J. (1983b). Phytoalexin induction in French bean. Intercellular transmission of elicitation in cell suspension cultures and hypocotyl sections of Phaseolus vulgaris. Plant Physiology 71, 251-256. Dixon, R. A., Jennings, A. C., Davies, L. A., Gerrish, C. and Murphy, D. L. (1989). Elicitor active components from French bean hypocotyls. Physiological and Molecular Plant Pathology 34, 99-1 15. Doares, S. H., Bucheli, P., Albersheim, P. and Darvill, A. G . (1989). Hostpathogen interactions. XXXIV. A heat-labile activity secreted by a fungal phytopathogen releases fragments of plant cell walls that kill plant cells. Molecular Plant-Microbe Interactions 2, 346-353.

86

S. ALDINGTON AND S. C. FRY

Doherty, H. M., Selvendran, R. R. and Bowles, D. J. (1988). The wound response of tomato plants can be inhibited by aspirin and related hydroxybenzoic acids. Physiological and Molecular Plant Pathology 33, 377-384. Doke, N. (1983a). Involvement of superoxide anion generation in the hypersensitive response of potato tuber tissues to infection with an incompatible race of Phytophthora infestans and to the hyphal wall components. Physiological Plant Pathology 23,345-357. Doke, N. (1983b). Generation of superoxide anion by potato tuber protoplasts during the hypersensitive response to hyphal wall components of Phytophthora infestans and specific inhibition of the reaction by suppressors of hypersensitivity. Physiological Plant Pathology 23, 359-367. Doke, N. and Tomiyama, K. (1980a). Effect of hyphal wall components from Phytophthora infestans on protoplasts of potato tuber tissues. Physiological Plant Pathology 16,169-176. Doke, N. and Tomiyama, K. (1980b). Suppression of the hypersensitive response of potato tuber protoplasts to hyphal wall components by water soluble glucans isolated from Phytophthora infestans. Physiological Plant Pathology 16, 177186. Dow, J. M. and Callow, J. A. (1979). Leakage of electrolytes from isolated leaf mesophyll cells of tomato induced by glycopeptides from culture filtrates of Fulvia fulva (Cooke) Ciferri (syn. Cladosporium fulvum). Physiological Plant Pathology 15, 27-34. Dunlop, D. S. and Curtis, W. R. (1991). Synergistic response of plant hairy-root cultures to phosphate limitation and fungal elicitation. Biotechnology Progress 7,434-438. Ebel, J., Ayers, A. R. and Albersheim, P. (1976). Host-pathogen interactions. XII. Response of suspension-cultured soybean cells to the elicitor isolated from Phytophthora megasperma var. sojae, a fungal pathogen of soybeans. Plant Physiology 57,775-779. Ebel, J. (1986). Phytoalexin synthesis: the biochemical analysis of the induction process. Annual Reviews in Phytopathology 24, 235-264. Ebel, J. and Grisebach, H. (1988). Defence strategies of soybean against the fungus Phytophthora megasperma f. sp. glycinea: a molecular analysis. Trends in Biochemical Science 13, 23-27. Eberhard, S., Doubrava, N., Marfa, V., Mohnen, D., Southwick, A., Darvill, A. and Albersheim, P. (1989). Pectic cell wall fragments regulate tobacco thincell-layer explant morphogenesis. Plant Cell 1, 747-755. Eble, A. S., Thorpe, S. R. and Baynes, J. W. (1983). Nonenzymatic glucosylation and glucose-dependent cross-linking of protein. Journal of Biological Chemistry 258,9406-9412. Ecker, J. R. and Davis, R. W. (1987). Plant defense genes are regulated by ethylene. Proceedings of the National Academy of Sciences, USA 84,5202-5206. Edelbaum, O . , Sher, N., Rubinstein, M., Novick, D., Tal, N., Moyer, M., Ward, E., Ryals, J. and Sela, I. (1991). Two antiviral proteins, gp35 and gp22, correspond to P-1,3-glucanase and an isoform of PR-5. Plant Molecular Biology 17, 171-173. Edelmann, H. G. and Fry, S. C. (1992a). Factors that affect the extraction of xyloglucan from the primary cell walls of suspension-cultured rose cells. Carbohydrate Research 228,423-431. Edelmann, H. G. and Fry, S. C. (1992b). Kinetics of integration of xyloglucan into the walls of suspension-cultured rose cells. Journal of Experimental Botany 43, 463-470. Edwards, K., Cramer, C. L., Bolwell, G. P., Dixon, R. A., Schuch, W. and Lamb,

OLIGOSACCHARINS

87

C. J. (1985). Rapid transient induction of phenylalanine ammonia-lyase mRNA in elicitor-treated bean cells. Proceedings of the National Academy of Sciences, USA 82,67314735, Edwards, M., Dea, I. C. M., Bulpin, P. V. and Reid, J. S. G. (1985). Xyloglucan (amyloid) mobilisation in the cotyledons of Tropaeolum majus L. seeds following germination. Planta 163, 133-140. Edwards, M., Bowman, Y. J. L., Dea, I. C. M. and Reid, J. S. G. (1988). A P-D-galactosidase from nasturtium (Tropaeohm majus L.) cotyledons. Purification, properties, and demonstration that xyloglucan is the natural substrate. Journal of Biological Chemistry 263, 43334337. Emmerling, M. and Seitz, H. U. (1990). Influence of a specific xyloglucannonasaccharide derived from cell walls of suspension-cultured cells of Daucus carota L. on regenerating carrot protoplasts. Planta 182, 174-180. English, P. D., Jurale, J. B. and Albersheim, P. (1971). Host-pathogen interactions. 11. Parameters affecting polysaccharide-degrading enzyme secretion by Colletotrichum lindemuthianum grown in culture. Plant Physiology 47, 1 4 . English, P. D., Maglothin, A., Keegstra, K. and Albersheim, P. (1972). A cell wall-degrading endopolygalacturonase secreted by Colletotrichum lindemuthianum. Plant Physiology 49,293-297. Epperlein, M. M., Noronha-Dutra, A. A. and Strange, R. N. (1986). Involvement of the hydroxyl radical in the abiotic elicitation of phytoalexins in legumes. Physiological and Molecular Plant Pathology 28, 67-77. Esquerre-TugayC, M.-T. and Mazau, D. (1974). Effect of a fungal disease on extensin, the plant cell wall glycoprotein. Journal of Experimental Botany 25, 509-5 13. EsquerrC-TugayC, M.-T., Mazau, D., PClissier, B., Roby, D., Rumeau, D. and Toppan, A. (1985). Induction by elicitors and ethylene of proteins associated to the defence of plants. In “Cellular and Molecular Biology of Plant Stress” (J. Key and T. Kosuge, eds), pp. 459-473. Alan R. Liss, New York. Fanutti, C., Gidley, M. J. and Reid, J. S. G. (1991). A xyloglucan-oligosaccharidespecific a-D-xylosidase or exo-oligoxyloglucan-a-xylohydrolasefrom germinated nasturtium (Tropaeolum majus L.) seeds. Purification, properties and its interaction with a xyloglucan-specificendo-(1+4)-P-~-glucanase and other hydrolases during storage-xyloglucan mobilisation. Planta 184, 137-147. FarkaS, V. and Maclachlan, G. (1988). Stimulation of pea 1,4-P-glucanase activity by oligosaccharides derived from xyloglucan. Carbohydrate Research 184, 213220. FarkaS, V., Hanna, R. and Maclachlan, G. (1991). Xyloglucan oligosaccharide a-L-fucosidase activity from growing pea stems and germinating nasturtium seeds. Phytochemistry 30,3203-3207. Farmer, E. E. and Helgeson, J. P. (1987). An extracellular protein from Phytophthora parasitica var. nicotianae is associated with stress metabolite accumulation in tobacco callus. Plant Physiology 85,733-740. Farmer, E. E., Pearce, G. and Ryan, C. A. (1989). In vitro phosphorylation of plant plasma membrane proteins in response to the proteinase inhibitor inducing factor. Proceedings of the National Academy of Sciences, USA 86,1539-1542. Farmer, E. E., Moloshok, T. D., Saxton, M. J. and Ryan, C. A. (1991). Oligosaccharide signalingin plants. Specificityof oligouronide-enhanced plasma membrane protein phosphorylation. Journal of Biological Chemistry 266, 3140-3145. Filippini, F., Lo Schiavo, F., Terzi, M., Branca, C., Bellincampi, D . , Desiderio, A., Salvi, G., De Lorenzo, G. and Cervone, F. (1992). Oligogalacturonides interact with auxin-binding sites and affect auxin-regulated development of cells and tissue explants. Plant Journal (in press).

88

S. ALDINGTON AND S. C. FRY

Fischer, R. L. and Bennett, A. B. (1991). Role of cell wall hydrolases in fruit ripening. Annual Review of Plant Physiology and Plant Molecular Biology 42, 675703. Fry, S. C. (1980). Gibberellin-controlled pectinic acid and protein secretion in growing cells. Phytochemistry 19, 735740. Fry, S. C. (1982). Isodityrosine, a new cross-linking amino acid from plant cell-wall glycoprotein. Biochemical Journal 204, 4 4 W 5 5 . Fry, S. C. (1986). In-vivo formation of xyloglucan nonasaccharide: a possible biologically active cell-wall fragment. PIanta 169, 443-453. Fry, S. C. (1988). “The Growing Plant Cell Wall: Chemical and Metabolic Analysis”. Longman, London. Fry, S. C. (1989a). The structure and functions of xyloglucan. Journal of Experimental Botany 40, 1-11. Fry, S. C. (1989b). Cellulases, hemicelluloses and auxin-stimulated growth: a possible relationship. Physiologia Plantarum 75, 532-536. Fry, S. C., Darvill, A. G. and Albersheim, P. (1983). Amino acid transport and protein synthesis: possible primary targets of biologically active cell wall fragments. In “Interactions between Nitrogen and Growth Regulators in Plant Development”, Monograph 9 (M. B. Jackson, ed.), pp. 33-44. British Plant Growth Regulator Group, Wantage, Oxfordshire. Fry, S. C., McDougall, G.J., Lorences, E. P., Biggs, K. J. and Smith, R. C. (1990). Oligosaccharins from xyloglucan and cellulose: Modulators of the action of auxin and Hf on plant growth. In “Hormone Perception and Signal Transduction in Animals and Plants”, pp. 285-298. Company of Biologists Ltd., Cambridge (Society for Experimental Biology). Fry, S. C., Smith, R. C., Renwick, K. F., Martin, D. J., Hodge, S. K. andMatthews, K. J. (1992). Xyloglucan endotransglycosylase, a new wall-loosening enzyme activity from plants. Biochemical Journal 282, 821-828. Fuchs, Y . , Saxena, A., Gamble, H. R. and Anderson, J. D. (1989). Ethylene biosynthesis-inducing protein from Cellulysin is an endoxylanase. Plant Physiology 89, 138-143. Funk, C., Gugler, K. and Brodelius, P. (1987). Increased secondary product formation in plant cell suspension cultures after treatment with a yeast carbohydrate preparation (elicitor). Phytochemistry 26,401-405. Furuichi, N., Tomiyama, K., Doke, N. and Nozue, M. (1979). Inhibition of further development of hypersensitive reactivity to Phytophthora infestans by Blasticidin S in cut tissues of potato tuber at various stages of aging process. Annals of the Phytopathological Society of Japan 45, 215-220. Fushtey, S. G. (1957). Studies in the physiology of parasitism XXIV. Further experiments on the killing of plant cells by fungal and bacterial extracts. Annals of Botany 21,273-286. Garcia-Romera, I., Garcia-Garrido, J. M., Martinez-Molina, E. and Ocampo, J. A. (1991). Production of pectolytic enzymes in lettuce root colonized by Glomus mosseae. Soil Biology and Biochemistry 23, 597-601. Gardner, J. M. and Kado, C. I. (1976). Polygalacturonic acid trans-eliminase in the osmotic shock fluid of Erwinia rubrifaciens: characterisation of the purified enzyme and its effect on plant cells. Journal of Bacteriology 127, 451-460. Gilkes, N. R. and Hall, M. A. (1977). The hormonal control of cell wall turnover in Pisum sativum L. New Phytologist 78, 1-15. Glazener, J. A. (1982). Accumulation of phenolic compounds in cells and formation of lignin-like polymers in cell walls of young tomato fruits after inoculation with Botrytis cinerea. Physiological Plant Pathology 20, 11-25. Gollin, D. J., Darvill, A. G . and Albersheim, P. (1984). Plant cell wall fragments

OLIGOSACCHARINS

89

inhibit flowering and promote vegetative growth in Lemna gibba G3. Biology of the Cell 51,275-280. Goodwin, J. C. (1983). Isolation of 3-O-a-~-gluco- and 3-O-P-~-galactopyranosyloxy-2-fury1 methyl ketones from non-enzymic browning of maltose and lactose with secondary amino acids. Carbohydrate Research 115,281-287. Govers, F., Harmsen, H., Heidstra, R., Michielsen, P., Prins, M., Van Kammen, A. and Bisseling, T. (1991). Characterization of the pea ENOD12B gene and expression analyses of the two ENOD12 genes in nodule, stem and flower tissue. Molecular and General Genetics 228, 160-166. Grab, D., Feger, M. and Ebel, J. (1989). An endogenous factor from soybean (Glycine max L.) cell cultures acivates phosphorylation of a protein which is dephosphorylated in vivo in elicitor-challenged cells. Planta 179, 340-348. Green, E. D. and Baenziger, J. U. (1989). Characterization of oligosaccharides by lectin affinity high-performance liquid chromatography. Trends in Biochemical Science 14, 168-172. Green, T. R. and Ryan, C. A. (1972). Wound-induced proteinase inhibitor in plant leaves: a possible defense mechanism against insects. Science 175, 776-777. Green, T. R. and Ryan, C. A. (1973). Wound-induced proteinase inhibitor in tomato leaves: some effects of light and temperature on the wound response. Plant Physiology 51, 19-21. Grisebach, H. and Ebel, J. (1978). Phytoalexins, chemical defence substances of higher plants. Angewandte Chemie-International Edition in English 17, 6 3 5 4 7 . Hadwiger, L. A. and Beckman, J. M. (1980). Chitosan as a component of peaFusarium solani interactions. Plant Physiology 66, 205-21 l. Hadwiger, L. A. and Loschke, D. C. (1981). Molecular communication in hostparasite interactions: hexosamine polymers (chitosan) as regulator compounds in race-specific and other interactions. Phytopathology 71, 756762. Hadwiger, L. A., Beckman, J. M. and Adams, M. J. (1981). Localization of fungal components in the pea-Fusarium interaction detected immunochemically with anti-chitosan and anti-fungal cell wall antisera. Plant Physiology 67, 170-175. Hagendoorn, M. J. M . , Traas, T. P., Boon, J. J. and Van Der Plas, L. H. W. (1990). Orthovanadate induced lignin production, in batch and continuous cultures of Petunia hybrida. Journal of Plant Physiology 137, 72-80. Hahlbrock, K . , Lamb, C. J., Purwin, C . ,Ebel, J . , Fautz, E. and Schafer, E. (1981). Rapid response of suspension-cultured parsley cells to the elicitor from Phytophthora megasperma var. sojae. Induction of the enzymes of general phenylpropanoid metabolism. Plant Physiology 67, 768-773. Hahn, M. G. and Albersheim, P. (1978). Host-pathogen interactions. XIV. Isolation and partial characterization of an elicitor from yeast extract. Plant Physiology 62, 107-111. Hahn, M. G. and Grisebach, H. (1983). Cyclic AMP is not involved as a second messenger in the response of soybean to infection by Phytophthora megasperma f. sp. glycinea. Zeitschrift fur Naturforschung 38c, 578-582. Hahn, M. G., Darvill, A. G. and Albersheim, P. (1981). Host-pathogen interactions. XIX. The endogenous elicitor, a fragment of a plant cell wall polysaccharide that elicits phytoalexin accumulation in soybeans. Plant Physiology 68, 1161-1169. Hahne, G. and Lorz, H. (1988). Release of phytotoxic factors from plant cell walls during protoplast isolation. Journal of Plant Physiology 132, 345-350. Hall, N. A. and Patrick, A. D. (1989). A high-performance liquid chromatography method for the analysis of picomole amounts of oligosaccharides. Analytical Biochemistry 178, 378-384.

90

S . ALDINGTON AND S. C. FRY

Halliwell, B. (1978). Superoxide-dependent formation of hydroxyl radicals in the presence of iron chelates. Is it a mechanism for hydroxyl radical production in biochemical systems? FEBS Letters 92, 321-326. Halliwell, B. and Gutteridge, J . M. C. (1984). Oxygen toxicity, oxygen radicals, transition metals and disease. Biochemical Journal 219, 1-14. Halverson, L. J. and Stacey, G . (1986). Signal exchange in plant-microbe interactions. Microbiological Reviews 50, 193-225. Hammerschmidt, R. and KuC, J . (1982). Lignification as a mechanism for induced systemic resistance in cucumber. Physiological Plant Pathology 20, 61-71. Hammerschmidt, R., Lamport, D. T. A. and Muldoon, E. P. (1984). Cell wall hydroxyproline enhancement and lignin deposition as an early event in the resistance of cucumber to Cladosporium cucumerinum. Physiological Plant Pathology 24, 43-47. Hammerschmidt, R., Bonnen, A. M., Bergstrom, G. C. and Baker, K. K . (1985). Association of epidermal lignification with nonhost resistance of cucurbits to fungi. Canadian Journal of Botany 63, 2393-2398. Hanke, D. E. and Northcote, D. H. (1974). Cell wall formation by soybean callus protoplasts. Journal of Cell Science 14, 29-50. Hardy, M. R. and Townsend, R. R. (1989). Separation of fucosylated oligosaccharides using high-pH anion-exchange chromatography with pulsedamperometric detection. Carbohydrate Research 188, 1-7. Hargreaves, J. A. and Bailey, J. A. (1978). Phytoalexin production by hypocotyls of Phaseolus vulgaris in response to constitutive metabolites released by damaged bean cells. Physiological Plant Pathology 13, 89-100. Hayashi, T. and Maclachlan, G. (1984). Pea xyloglucan and cellulose. I . Macromolecular organization. Plant Physiology 75, 5 9 M 0 4 . Hayashi, T., Wong, Y . 4 . and Maclachlan, G. (1984). Pea xyloglucan and cellulose. 11. Hydrolysis by pea endo-l,4-P-glucanases. Plant Physiofogy 75,605610. Hayashi, T. (1989). Xyloglucans in the primary cell wall. Annual Review of Plant Physiology and Plant Molecular Biology 40, 139-168. Hedrick, S. A., Bell, J. N., Boller, T. and Lamb, C. J . (1988). Chitinase cDNA cloning and mRNA induction by fungal elicitor, wounding and infection. Plant Physiology 86, 182-186. Hinch, J . M. and Clarke, A. E. (1990). Adhesion offungal zoospores to root surfaces is mediated by carbohydrate determinants of the root slime. Physiological Plant Pathology 16, 303-307. Hinton, D. R. and Pressey, R. (1974). Cellulase activity in peaches during ripening. Journal of Food Science 39,783-785. Hislop, E. C., Keon, J. P. R. and Fielding, A. H. (1979). Effectsof pectin lyasefrom Monilinia fructigena on viability, ultrastructure and localization of acid phosphatase of cultured apple cells. Physiological Plant Pathology 14, 371-381. Holliday, M. J . , Keen, N. T. and Long, M. (1981). Cell death patterns and accumulation of fluorescent material in the hypersensitive response of soybean leaves to Pseudomonas syringae pv. glycinea. Physiological Plant Pathology 18, 279-287. Hopper, D. G., Venere, R. J . , Brinkerhoff, L. A. and Gholson, R. K. (1975). Necrosis induction in cotton. Phytopathology 65, 206-213. Horn, M. A., Heinstein, P. F. and Low, P. S. (1989). Receptor-mediated endocytosis in plant cells. Plant Cell 1, 1003-1009. Hoson, T. and Masuda, Y. (1991). Effect of xyloglucan nonasaccharide on cell elongation induced by 2,4-dichlorophenoxyacetic acid and indole-3-acetic acid. Plant and Cell Physiology 32, 777-782.

OLIGOSACCHARINS

91

Howard, J., Shannon, L., Oki, L. and Murashige, T. (1977). Soybean agglutinin. A mitogen for soybean callus cells. Experimental Cell Research 107,448450. Hughes, R. K. and Dickerson, A. G. (1989). The effect of ethylene on the induction of phenylalanine ammonia-lyase (PAL) by a fungal elicitor in Phaseolus vulgaris. Physiological and Molecular Plant Pathology 34, 361-378. Hughes, R. K. and Dickerson, A. G. (1991). Modulation of elicitor-induced chitinase and p-1,3-glucanase activity by hormones in Phaseolus vulgaris. Plant and Cell Physiology 32,853-861. Isaiah, A. (1971). Influence comparee de 1’(+)-acide abscissique, de la coumarine, de I’acide p-coumarique et de certains de leurs dirives sur la croissance de la premiere feuille de ble. Planta 96, 175-182. Ishii, S. (1988). Factors influencing protoplast viability of suspension-cultured rice cells during isolation process. Plant Physiology 88, 2C29. Ishii, T. and Saka, H. (1992). Inhibition of auxin-stimulated elongation of cells in rice lamina joints by a feruloylated arabinoxylan trisaccharide. Plant and Cell Physiology 33, 321-324. Jarvis, M. C. (1982). The proportion of calcium-bound pectin in plant cell walls. Planta 154, 344-346. Jarvis, M. C . (1984). Structure and properties of pectin gels in plant cell walls. Plant, Cell and Environment 7,153-164. Jin, D. F. and West, C. A. (1984). Characteristics of galacturonic acid oligomers as elicitors of casbene synthetase activity in castor bean seedlings. Plant Physiology 74, 989-992. Kato, Y. and Matsuda, K. (1980). Structure of oligosaccharides obtained by hydrolysis of mung bean xyloglucan with Trichoderma viride cellulase. Agricultural and Biological Chemistry 44,1759-1766. Kato, K. and Noguchi, M. (1976). Sugar composition of cell wall polysaccharides of suspension-cultured tobacco cells. Agricultural and Biological Chemistry 40, 1923-1928. Kato, Y . , Matsushita, J., Kubodera, T. and Matsuda, K. (1985). A novel enzyme producing isoprimeverose from oligoxyloglucans of Aspergillus oryzae. Journal of Biochemistry 97, 801-810. Kauffman, S . , Doares, S. H., Albersheim, P. and Darvill, A. G. (1990). Involvement of pathogen-secreted enzymes and plant cell wall-derived xylan fragments in plant cell death in the rice blast system. In “Plant-Microbe Interactions”, 5th International Symposium on Molecular Genetics, Interlaken, p. 32 [Abstract]. Kauss, H. (1987). Some aspects of calcium-dependent regulation in plant metabolism. Annual Reviews in Plant Physiology 38, 47-72. Kauss, H., Kohle, H. and Jeblick, W. (1983). Proteolytic activation and stimulation by Ca2+ of glucan synthase from soybean cells. FEBS Letters 158,8688. Kauss, H., Jeblick, W. and Domard, A. (1989). The degrees of polymerization and N-acetylation of chitosan determine its ability to elicit callose formation in suspension cells and protoplasts of Catharanthus roseus. Planta 178,385-392. Keen, N. T. (1982). Specific recognition in gene-for-gene host-parasite systems. Advances in Plant Pathology 1, 35-82. Keen, N. T. and Yoshikawa, M. (1983). P-1,3-Endoglucanase from soybean releases elicitor-active carbohydrates from fungus cell walls. Plant Physiology 71, 460-465. Keenan, P., Bryan, I. B . and Friend, J. (1985). The elicitation of the hypersensitive response of potato tuber tissue by a component of the culture filtrate of Phytophthora infestans. Physiological Plant Pathology 26, 343-355. Keil, M., Sanchez-Serrano, J., Schell, J. and Willmitzer, L. (1990). Localization of

92

S. ALDINGTON AND S. C. FRY

elements important for the wound-inducibleexpression of a chimeric potato proteinase inhibitor 11-CAT gene in transgenic tobacco plants. Plant Cell 2, 61-70. Kendra, D . F. and Hadwiger, L. A. (1984). Characterisation of the smallest chitosan oligomer that is maximally anti-fungal to Fusarium solani and elicits pisatin formation in Pisum sativum. Experimental Mycology 8, 276281. Kendra, D. F. and Hadwiger, L. A. (1987a). Cell death and membrane leakage not associated with the induction of disease resistance in peas by chitosan or Fusarium solani f. sp. phaseoli. Phytopathology 77, 100-106. Kendra, D. F. and Hadwiger, L. A. (1987b). Calcium and calmodulin may not regulate the disease resistance and uisatin formation of Pisum sativum to chytosan or Fusarium solani. Physiolo'gical and Molecular Plant Pathology 31, 337-348. Keppler, L. D. and Novacky, A. (1987). The initiation of membrane lipid peroxidation during bacteria-induced hypersensitive reaction. Physiological and Molecular Plant Pathology 30, 233-245. Kiefer, L. L., York, W. S., Albersheim, P. and Darvill, A. G. (1990). Structural characterization of an arabinose-containing heptadecasaccharide enzymatically isolated from sycamore extracellular xyloglucan. Carbohydrate Research 197, 139-158. Knight, M. R., Campbell, A. K., Smith, S. M. andTrewavas, A. J. (1991). Transgenic plant aequorin reports the effects of touch and cold-shock and elicitors on cytoplasmic calcium. Nature 352, 524-526. Kobata, A., Yamashita, K. and Takasaki, S. (1987). BioGel P-4 column chromatography of oligosaccharides: effective size of oligosaccharides expressed in glucose units. Methods in Enzymology 138, 84-94. Kogel, G., Beissman, B., Reisener, H. J. and Kogel, K. H. (1988) A single glycoprotein from Puccinia graminis f. sp. tritici cell walls elicits the hypersensitive lignification response in wheat. Physiological and Molecular Plant Pathology 33, 173-185. Kogel, G., Beissman, B., Reisener, H. J. and Kogel, K. (1991). Specific binding of a hypersensitive lignification elicitor from Puccinia graminis f. sp. tritici to the plasma membrane from wheat (Triticum aestivum L.). Planta 183, 164-169. Kohle, H., Young, D. H. and Kauss, H. (1984) Physiological changes in suspensioncultured soybean cells elicited by treatment with chitosan. Plant Science Letters 33, 221-230. Kohle, H., Jeblick, W . , Poten, F., Blaschek, W. and Kauss, H. (1985). Chitosanelicited callose synthesis in soybean cells as a Cazf-dependent process. Plant Physiology 77, 544-551. Koizumi, K., Okada, Y. and Fukuda, M. (1991). High-performance liquid chromatography of mono- and oligo-saccharides on a graphitized carbon column. Carbohydrate Research 215, 67-80. Konno, H. and Tsukumi, H. (1991). An exo-polygalacturonase from rice shoots. Phytochemistry 30, 2115-2118. Kooiman, P. (1961). The constitution of Tamarindus-amyloid. Recueli des Travaux Chimiques des Pays-Bas et de la Belgique 80,849-865. Koyama, T., Hayashi, T., Kato, Y. and Matsuda, K. (1983). Degradation of xyloglucan by wall-bound enzymes from soybean tissue. 11. Degradation of the fragment heptasaccharide from xyloglucan and the characteristic action pattern of the a-D-xylosidase in the enzyme system. Plant and Cell Physiology 24, 155-162. KrBtkB, J. and KGdela, V. (1984) Effect of alfalfa wilts on the hydroxyproline content in cell wall. Phytopathologische Zeitschrift 110, 127-133. KuC, J. (1982). Induced immunity to plant diseases. BioScience 32, 854-860.

OLIGOSACCHARINS

93

KuC, J. and Rush, J. S. (1985) Phytoalexins. Archives of Biochemistry and Biophysics 236, 455-472. Kuhn, R., Baer, H. H. and Gauhe, A. (1958). 2-a-~-Fucopyranosyl-~-Galaktose und 2-a-~-Fucopyranosyl-~-Talose. Zur Einwirkung von Alkali auf Oligosaccharide. Chemische Berichte 611,242-249. Kurantz, M. J. and Zacharius, R. M. (1981). Hypersensitive response in potato tuber: elicitation by combination of non-eliciting components from Phytophthora infestans. Physiological Plant Pathology 18, 67-77. Kurosaki, F., Tsurusawa, Y. and Nishi, A. (1985a). Phytoalexin production in cultured carrot cells treated with pectinolytic enzymes. Phytochemistry 24, 147%1480. Kurosaki, F., Tsurusawa, Y. and Nishi, A. (1985b). Partial purification and characterization of elicitors for 6-methoxymellein production in cultured carrot cells. Physiological Plant Pathology 27,209-217. Kurosaki, F., Tsurusawa, Y. and Nishi, A. (1987a). Breakdown of phosphatidylinositol during the elicitation of phytoalexin production in cultured carrot cells. Plant Physiology 85,601-604. Kurosaki, F., Tsurusawa, Y. and Nishi, A. (1987b). The elicitation of phytoalexins by Ca” and cyclic AMP in carrot cells. Phytochemistry 26, 1919-1923. Kurosaki, F., Tashiro, N. and Nishi, A. (1988). Role of chitinase and chitin oligosaccharides in lignification response of cultured carrot cells treated with mycelial cell walls. Plant and Cell Physiology 29, 527-531. Labavitch, J. (1983). Stimulationoffruit ethylene production. PlantPhysiology72S7158. Labavitch, J. and Ray, P. M. (1974). Relationship between promotion of xyloglucan metabolism and induction of elongation by indoleacetic acid. Plant Physiology 54,499-502. Lamb, C. J., Lawton, M. A., Dron, M. and Dixon, R. A. (1989). Signals and transduction mechanisms for activation of plant defenses against microbial attack. Cell 56,215-224. Lamport, D. T. A. (1977). Structure, biosynthesis and significance of cell wall glycoproteins. In “Recent Advances in Phytochemistry”, Vol. 11,The Structure, Biosynthesis and Degradation of Wood (F. A. Loewus and V. C. Runeckles, eds), pp. 79-115. Plenum, New York. Lamport, D. T. A. (1980). Structure and function of plant glycoproteins. In “The Biochemistry of Plants: A Comprehensive Treatise” (J. Preiss, ed.), Vol. 3, pp. 501-541. Academic Press, New York. Lawton, M. A. and Lamb, C. J. (1987). Transcriptional activation of plant defense genes by fungal elicitor, wounding and infection. Molecular and Cellular Biology 7,335-341. Leach, J. E., Cantrell, M. A. and Sequeira, L. (1982). Hydroxyproline-rich bacterial agglutinin from potato. Extraction, purification, and characterization. Plant Physiology 70, 1353-1358. Lee, S.-C. and West, C. A. (1981a). Polygalacturonase from Rhizopus stolonifer, an elicitor of casbene synthetase activity in castor bean (Ricinus cornmunis L.) seedlings. Plant Physiology 67, 633-639. Lee, S.-C. and West, C. A. (1981b). Properties of Rhizopus stolonifer polygalacturonase, an elicitor of casbene synthetase activity in castor bean (Ricinus communis L.) seedlings. Plant Physiology 67, 640-645. Lerouge, P., Roche, P., Faucher, C., Maillet, F., Truchet, G., PromC, J. C. and DCnariC, J. (1990). Symbiotic host-specificity of Rhizobium meliloti is determined by a sulphated and acylated glucosamine oligosaccharide signal. Nature 344,781-784. LiCnart, Y., Gautier, C. and Driguez, H. (1991). Silica-bound N-acetylglucosaminyl

94

S. ALDINGTON AND S. C. FRY

residues elicit laminarinase activity in Rubus protoplasts. Plant Science 77, 41-45.

Lindner, W. A., Hoffman, C. and Grisebach, H. (1988). Rapid elicitor-induced chemiluminescence in soybean cell suspension cultures. Phytochemistry 27, 2501-2503.

Longman, D. and Callow, J. A. (1987). Specific saccharide residues are involved in the recognition of plant root surfaces by zoospsores of Pythium aphanidermatum. Physiological and Molecular Plant Pathology 30, 139-150. Lorences, E. P., McDougall, G. J. and Fry, S. C. (1990). Xyloglucan- and cellooligosaccharides Antagonists of the growth-promoting effect of H+. Physiologia Plantarum 80, 109-113. Lotan, T. and Fluhr, R. (1990). Xylanase, a novel elicitor of pathogenesis-related proteins in tobacco, uses a non-ethylene pathway for induction. Plant Physiology 93, 811-817. Lyon, G. D. and Albersheim, P. (1982). Host-pathogen interactions. XXI. Extraction of a heat-labile elicitor of phytoalexin accumulation from frozen soybean stems. Plant Physiology 70,406-409. Ma, R., Reese, J. C., Black, W. C. and Bramel-Cox, P. (1990). Detection of pectinesterase and polygalacturonase from salivary secretions of living greenbugs, Schizaphisgraminum (Homoptera: Aphididae). Journal of Insect Physiology 36, 507-512. Maclean, D. J., Sargent, J. A., Tommerup, I. C. and Ingram, D. S. (1974). Hypersensitivity as the primary event in resistance to fungal parasites. Nature 249,186-187.

Maness, N. 0. and Mort, A. J. (1989). Separation and quantitation of galacturonic acid oligomers from 3 to 25 residues in length by anion-exchange high performance liquid chromatography. Analytical Biochemistry 178, 248-254. Marinelli, F., Di Gregorio, S. and Ronchi, V. N. (1991). Phytoalexin production and cell death in elicited carrot cell suspension cultures. Plant Science 77,261-266. Masuta, C., Van Den Bulcke, M., Bauw, G., Van Montague, M. and Caplan, A . B. (1991). Differential effects of elicitors on the viability of rice suspension cultures. Plant Physiology 97, 619-629. Matsushita, J., Kato, Y. and Matsuda, K. (1985). Further structural study of an endo-cellulase-derived mung bean xyloglucan oligosaccharide, decasaccharide. Agricultural and Biological Chemistry 49, 1533-1534. Mauch, F., Hadwiger, L. A. and Boller, T. (1984). Ethylene: symptom, not signal for the induction of chitinase and P-1,3-glucanase in pea pods by pathogens and elicitors. Plant Physiology 76, 607-611. Mauch, F., Hadwiger, L. A. and Boller, T. (1988a). Antifungal hydrolases in pea tissue. I. Purification and characterization of two chitinases and two P-1,3glucanases differentially regulated during development and in response to fungal infection. Plant Physiology 87, 325-333. Mauch, F., Mauch-Mani, B. and Boller, T. (1988b). Antifungal hydrolases in pea tissue. 11. Inhibition of fungal growth by combinations of chitinase and p-1,3glucanase. Plant Physiology 88,936-942. Mayer, M. G. and Ziegler, E. (1988). An elicitor from Phytophthora megasperma f. sp. glycinea influences the membrane potential of soybean cotyledonary cells. Physiological and Molecular Plant Pathology 33, 3 9 7 4 7 . McDougall, G. J. and Fry, S. C. (1988). Inhibition of auxin-stimulated growth of pea stem segments by a specific nonasaccharide of xyloglucan. Planta 175, 412416.

McDougall, G. J. and Fry, S. C. (1989a). Structure-activity relationships for xyloglucan oligosaccharides with antiauxin activity. Plant Physiology 89,883-887.

OLIGOSACCHARINS

95

McDougall, G. J. and Fry, S. C. (1989b). Anti-auxin activity of xyloglucan oligosaccharides: the role of groups other than the terminal a-r>-fucoseresidue. Journal of Experimental Botany 40, 233-238. McDougall, G. J. and Fry, S. C. (1990). Xyloglucan oligosaccharides promote growth and activate cellulase: evidence for a role of cellulase in cell expansion. Plant Physiology 93, 1042-1048. McDougall, G. J. and Fry, S. C. (1991a). Purification and analysis of growthregulating xyloglucan-derived oligosaccharides by high-pressure liquid chromatography. Carbohydrate Research 219, 123-132. McDougall, G. M. and Fry, S. C. (1991b). Xyloglucan nonasaccharide, a naturallyoccurring oligosaccharin, arises in vivo by polysaccharide breakdown. Journal of Plant Physiology 137, 332-336. McFarland, D. and Ryan, C. A. (1974). Proteinase inhibitor-inducing factor in plant leaves. A phylogenetic survey. Plant Physiology 54, 706708. McNeil, M., Darvill, A. G., Fry, S. C. and Albersheim, P. (1984). Structure and function of the primary cell walls of plants. Annual Reviews in Biochemistry 53. 625-663. Melton, L. D., McNeil, M., Darvill, A. G., Albersheim, P. and Dell, A. (1986). Structural characterization of oligosaccharides isolated from the pectic polysaccharide rhamnogalacturonan. 11. Carbohydrate Research 146, 279-305. Messiaen, J., Read, N. D., Van Cutsem, P. and Trewavas, A. J. (1992). Cell wall oligosaccharides increase cytosolic free calcium in carrot protoplasts. Journal of Cell Science (submitted). Miller, C. O., Skoog, F., Von Saltza, M. H. and Strong, F. M. (1955). Kinetin, a cell division factor from deoxyribonucleic acid. Journal of the American Chemical Society 77, 1329. Miller, L. and Macmillan, J. D. (1971). Purification and pattern of action of pectinesterase from Fusarium oxysporuni f. sp. vasinfectum. Biochemistry 10,57@576. Moerschbacher, B. M., Kogel, K. H., NOH, U . and Reisener, H.-J. (1986). An elicitor of the hypersensitive lignification response in wheat leaves isolated from the rust fungus Puccinia graminis f. sp. tritici. I. Partial purification and characterization. Zeitschrift fiir Naturforschung 41c, 830-838. Moerschbacher, B. M., Flott, B. E., NOH, U. and Reisener, H.-J. (1989). On the specificity of an elicitor preparation from stem rust which induces lignification in wheat leaves. Plant Physiology and Biochemistry 27, 305-314. Mohnen, D., Eberhard, S., MarfB, V., Doubrava, N . , Toubart, P., Gollin, D. J., Gruber, T. A . , Nuri, W., Albersheim, P. and Darvill, A . G. (1990). The control of root, vegatative shoot and flower morphogenesis in tobacco thin cell-layer explants (TCLs). Development 108, 191-201. Moloshok, T. and Ryan, C. A. (1989). Di- and trigalacturonic acid and A4.’-di- and A4~s-trigalacturonicacids: inducers of proteinase inhibitor genes in plants. Methods in Enzymology 179,566570. Mussell, H. and Strand, L. L. (1976). Pectic enzymes: involvement in pathogenesis and possible relevance to tolerance and specificity. In “Cell Wall Biochemistry Related to Specificity in Host Plant-Pathogen Interactions” (B. Solheim and J. Raa, eds), pp. 31-61. Universitetsforlaget, Oslo. Nakahara, Y. and Ogawa, T. (1987). Stereocontrolled, total synthesis of a - ~ - G a l A [(1+4)-a-~-GalA]8-(1+4)-P-~-GalA-l+OPr, a synthetic model for phytoalexin elicitor-active oligogalacturonic acids. carbohydrate Research 167, clk7. Nakahara, Y. and Ogawa, T. (1990). Synthesis of (1+4)-linked galacturonic acid trisaccharides, a proposed plant wound-hormone and a stereoisomer. Carbohydrate Research 200, 363-375.

96

S. ALDINGTON AND S. C. FRY

Nealey, L. T., Johnson, M. A . and Thompson N. S. (1989). Some properties of a xyloglucan isolated from suspension-cultured Lobolly pine (Pinus taeda) cells. Cellulose Chemistry and Technology 23, 671-682. Nelson, C. E., Walker-Simmons, M. and Ryan, C. A. (1981). Regulationof proteinase inhibitor synthesis in tomato leaves. In vitro synthesis of inhibitors I and I1 with mRNA from excised leaves induced with PIIF (proteinase inhibitor inducing factor). Plant Physiology 67, 841-844. Nilsson, K. G. I. (1988). Enzymatic synthesis of oligosaccharides. Trends in Biotechnology 6, 256-264. Nishitani, K . and Masuda, Y. (1982). Acid pH-induced structural changes in cell wall xyloglucans in Vigna angularis epicotyl segments. Plant Science Letters 28, 87-94. Nishitani, K. and Nevins, D. J. (1991). Glucuronoxylan xylanohydrolase: a unique xylanase with the requirement for appendant glucuronosyl units. Journal of Biological Chemistry 266, 6539-6543. Nothnagel, E. A., McNeil, M., Albersheim, P. and Dell, A. (1983). Host-pathogen interactions. XXII. A galacturonic acid oligosaccharide from plant cell walls elicits phytoalexins. Plant Physiology 71, 916926. O’Neill, M., Albersheim, P. and Darvill, A. G. (1990). The pecticpolysaccharides of primary cell walls. In “Methods in Plant Biochemistry” (P. M. Dey, ed.), Vol. 2, pp. 415-441. Academic Press, London. Ossowski, P., Pilotte, Garegg, P. J. and Lindberg, G. (1984). Synthesis of a glucoheptaose and a glucooctaose that elicit phytoalexin accumulation in soybean. Journal of Biological Chemistry 259, 11 337-11 340. Pagel, W. and Heitefuss, R. (1990). Enzyme activities in soft rot pathogenesis of potato tubers: effects of calcium, pH and degree of pectin esterification on the activities of polygalacturonase and pectate lyase. Physiological and Molecular Plant Pathology 37, %25. Paradies, I., Konze, J. R., Elstner, E. F. and Paxton, J. (1980). Ethylene: indicator but not inducer of phytoalexin synthesis in soybean. Plant Physiology 66, 1106-1109. Peever, T. L. and Higgins, V. J. (1989). Electrolyte leakage, lipoxygenase, and lipid peroxidation induced in tomato leaf tissue by specific and nonspecific elicitors from Cladosporium fulvum. Plant Physiology 90, 867-875. Pegg, G. F. and Vessey, J.C. (1973). Chitinase activity in Lycopersicon esculentum and its relationship to the in vivo lysis of Verticillium alboatrum mycelium. Physiological Plant Pathology 3, 207-222. PClissier, B . , Thibaud, J. B., Grignon, C. and Esquerre-Tugaye, M.-T. (1986). Cell surfaces in plant-microorganism interactions. VII. Elicitor preparations from two fungal pathogens depolarise plant membranes. Plant Science 46, 103-109. Pefia-CortCs, H., Sanchez-Serrano, J . , Rocha-Sosa, M. and Willmitzer, L. (1988). Systemic induction of proteinase-inhibitor-I1 gene expression in potato plants by wounding. Planta 174, 84-89. Pefia-CortCs, H., Sanchez-Serrano, J. J . , Mertens, R., Willmitzer, L. and Prat, S. (1989). Abscisic acid is involved in the wound-induced expression of the proteinase inhibitor I1 gene in potato and tomato. Proceedings of the National Academy of Sciences, USA 86,9851-9855. Peters, B. M., Cribbs, D . H. and Stelzig, D. A. (1978). Agglutination of plant protoplasts by fungal cell wall glucans. Science 201,364-365. Piatti, T., Boller, T. and Brodelius, P. E. (1991). Induction of ethylene biosynthesis is correlated with but not required for induction of alkaloid accumulation in elicitor-treated Eschscholtzia cells. Phytochemistry 30, 2151-2154.

w.,

OLIGOSACCHARINS

97

Pressey, R. (1990). Anions activate the oxidation of indoleacetic acid by peroxidases from tomato and other sources. Plant Physiology 93, 798-804. Pressey , R. (1991). Oxidized oligogalacturonides activate the oxidation of indoleacetic acid by peroxidase. Plant Physiology 96, 1167-1 170. Priem, B., Morvan, H. Hafez, A. M. A. and Morvan, C. (1990). Influence d’un xylomannoside d’origine vkgetale sur I’elongation de I’hypocotyle de lin. Comptes Rendus de I‘Academie des Sciences Paris SCrie I11 311, 411416. Redgwell, R. J. and Selvendran, R. R. (1986). Structural features of cell-wall polysaccharides of onion Allium cepa. Carbohydrate Research 157, 183-199. Rexova-Benkova, L. and MarkoviE, 0. (1976). Pectic enzymes. Advances in Carbohydrate Chemistry and Biochemistry 33, 323-385. Rickauer, M., Fournier, J. and Esquerre-Tugaye, M. T. (1989). Induction of proteinase inhibitors in tobacco cell suspension culture by elicitors of Phytophthora parasitica var. nicotianae. Plant Physiology 90, 1065-1070. Ride, J. P. (1975). Lignification in wounded wheat leaves in response to fungi and its possible role in resistance. Physiological Plant Pathology 5, 125-134. Ride, J. P. (1992). Recognition signals and initiation of host responses controlling basic incompatibility between fungi and plants. In “Perspectives in Plant Cell Recognition”, SEB Seminar Series 48, pp. 213-237. Ed. 3 . A. Callow and J. R. Green, Cambridge University Press. Ride, J. P. and Barber, M. S. (1987). The effects of various treatments on induced lignification and the resistance of wheat to fungi. Physiological and Molecular Plant Pathology 31,349-360. Robertsen, B. (1986). Elicitors of the production of lignin-like compounds in cucumber hypocotyls. Physiological and Molecular Plant Pathology 28, 137-148. Robertsen, B. (1987). Endo-polygalacturonase from Cladosporium cucumerinum elicits lignification in cucumber hypocotyls. Physiological and Molecular Plant Pathology 31, 361-374. Robertsen, B. and Svalheim, 0. (1990). The nature of lignin-like compounds in cucumber hypocotyls induced by a-l,4-linked oligogalacturonides. Physiologia Plantarum 79, 512-518. Roby, D., Toppan, A. and Esquerrk-Tugaye, M.-T. (1985). Cell surfaces in plantmicroorganism interactions. V. Elicitors of fungal and plant origin trigger the synthesis of ethylene and of cell wall hydroxyproline-rich glycoprotein in plants. Plant Physiology 77, 70C704. Roby, D., Toppan, A. and Esquerre-Tugaye, M.-T. (1986). Cell surfaces in plantmicroorganism interactions. VI. Elicitors of ethylene from Colletotrichum lagenarium trigger chitinase activity in melon plants. Plant Physiology 81, 228-233. Roby, D., Gadelle, A. andToppan, A. (1987). Chitin oligosaccharides aselicitors of chitinase activity in melon plants. Biochemical and Biophysical Research Communications 143, 885-892. Roby, D., Toppan, A. and Esquerre-Tugaye, M.-T. (1988). Systemic induction of chitinase activity and resistance in melon plants upon fungal infection or elicitor treatment. Physiological and Molecular Plant Pathology 33,409-41 7. Rogers, K. R., Albert, F. and Anderson, A. J . (1988). Lipid peroxidation is a consequence of elicitor activity. Plant Physiology 86, 547-553. Rombouts, F. M. and Pilnik, W. (1980). Pectic enzymes. In “Economic Microbiology” (A. H. Rose, ed.), Vol. 5, pp. 227-282. Academic Press, London. Rumeau, D., Mazau, D., Panabikres, F., Delseny, M. and Esquerre-Tugaye M.-T. (1988). Accumulation of hydroxyproline-rich glycoprotein mRNAs in infected or ethylene treated melon plants. Physiological and Molecular Plnnt Pathology 33, 419428.

98

S. ALDINGTON AND S. C. FRY

Ryan, C. A. (1968). Synthesis of chymotrypsin inhibitor I protein in potato leaflets induced by detachment. Plant Physiology 43, 1859-1865. Ryan, C. A. (1974). Assay and biochemical properties of the proteinase inhibitorinducing factor, a wound hormone. Plant Physiology 54,328-332. Ryan, C. A . (1987). Oligosaccharide signalling in plants. Annual Review of Cell Biology 3, 295-317. Ryan, C. A . (1988). Oligosaccharides as recognition signals for the expression of defensive genes in plants. Biochemistry 27, 8879-8883. Ryan, C. A. (1990). Protease inhibitors in plants: genes for improving defenses against insects and pathogens. Annual Review in Phytopathology 28, 425-449. Ryan, C. A. and Farmer, E . E. (1991). Oligosaccharide signals in plants: a current assessment. Annual Review of Plant Physiology and Plant Molecular Biology 42,651-674. Ryan C. A. and Huisman, W. (1970). The regulation of synthesis and storage of chymotrypsin inhibitor I in leaves of potato and tomato plants. Plant Physiology 45, 484489. Ryan, C. A., Bishop, P., Pearce, G., Darvill, A . G . , McNeil, M. and Albersheim, P. (1981). A sycamore cell wall polysaccharide and a chemically related tomato leaf polysaccharide possess similar proteinase inhibitor-inducing activities. Plant Physiology 68, 616-618. Ryder, T. B., Bell, J. N . , Cramer, C. L., Dildine, S. L., Grand, C . , Hedrick, S. A . , Lawton, M. A. and Lamb, C. J. (1986). Organization, structure and activation of plant defence genes. In “Biology and Molecular Biology of Plant-Pathogen Interactions” (J. A. Bailey, ed.), pp. 207-219. Springer-Verlag, Berlin. NATO AS1 Series H1. Sakai, K., Nakahara, Y. and Ogawa, T. (1990). Total synthesis of nonasaccharide repeating unit of plant cell wall xyloglucan: an endogenous hormone which regulates cell growth. Tetrahedron Letters 31, 3035-3038. Sarkanen, K. V. and Ludwig, C. H. (eds) (1971). “Lignins: Occurrence, Formation, Structure and Reactions”. Wiley, New York. Schlumbaum, A., Mauch, F., Vogeli, U . and Boller, T. (1986). Plant chitinases are potent inhibitors of fungal growth. Nature 324, 365-367. Schmidt, W. E. and Ebel, J . (1987). Specific binding of a fungal glucan phytoalexin elicitor to membrane fractions from soybean Glycine max. Proceedings of the National Academy of Sciences, U S A 84, 41174121. Schols, H. A . , Geraeds, C. C. J. M . , Searle-Van Leeuwen, M. F., Kormelink, F. J. M. and Voragen, A. G. J. (1990). Rhamnogalacturonase: a novel enzyme that degrades the hairy regions of pectins. Carbohydrate Research 206, 105-115. Schottens-Toma, I . M. J . and de Wit, P. J. G. M. (1988). Purification and primary structure of a necrosis-inducing peptide from the apoplastic fluids of tomato infected with Cladosporium fulvum (syn. Fulvia fulva). Physiological and Molecular Plant Pathology 33,59-67. Sharp, J. K., McNeil, M. and Albersheim, P. (1984a). The primary structures ofone elicitor-active and seven elicitor-inactive hexa(P-D-ghcopyranosyl)-D-ghcitols isolated from the mycelial walls of Phytophthora megasperma f. sp. glycinea. Journal of Biological Chemistry 259, 11 321-11 336. Sharp, J. K., Valent, B. and Albersheim, P. (1984b). Purification and partial characterization of a P-glucan fragment that elicits phytoalexin accumulation in soybean. Journal of Biological Chemistry 259, 11 312-11 320. Sharp, J. K., Albersheim, P., Ossowski, P., Pilotti, A., Garegg, P. and Lindberg, B. (1984~).Comparison of the structures and elicitor activities of a synthetic and a

OLIGOSACCHARINS

99

mycelial-wall-derived hexa(P-D-ghcopyranosyl)-D-glucitol. Journal of Biological Chemistry 259, 11 341-11 345. Shinshi, H. Mohnen, D. and Meins, F. (1987). Regulation of a plant pathogenesisrelated enzyme. Inhibition of chitinase and chitinase mRNA accumulation in cultured tobacco tissues by auxin and cytokinin. Proceedings of the National Academy of Sciences, USA 84, 89-93. Showalter, A. M., Bell, J. N., Cramer, C. L., Bailey, J. A., Varner, J. E. and Lamb, C. J. (1985). Accumulation of hydoxyproline-rich glycoprotein mRNAs in response to fungal elicitor and infection. Proceedings of the National Academy of Sciences, USA 82, 6551-6555. Siegrist, J. and Kauss, H. (1990). Chitin deacetylase in cucumber leaves infected by Colletotrichum lagenarium. Physiological and Molecular Plant Pathology 36, 267-275. Skare, N., Paus, F. and Raa, J. (1975). Production of pectinase and cellulase by Cladosporium cucumerinum with dissolved carbohydrates and isolated cell walls of cucumber as carbon sources. Physiologia Plantarum 33, 229-233. Smidsrod, O., Haug, A. and Larson, B. (1966). The influence of pH on the rate of hydrolysis of acidic polysaccharides. Acta Chemica Scandinavica 20, 1 0 2 6 1034. Smith, R. C. and Fry, S. C. (1991). Endotransglycosylation of xyloglucans in plant cell suspension cultures. Biochemical Journal 279, 529-535. Somssich, I. E., Schmelzer, E., Bollmann, J. and Hahlbrock, K. (1986). Rapid activation by fungal elicitor of genes encoding “pathogenesis-related” proteins in cultured parsley cells. Proceedings of the National Academy of Sciences, USA 83, 2427-2430. Somssich, I. E., Bollmann, J., Hahlbrock, K., Kombrink, E. and Schulz, W. (1989). Differential early activation of defense-related genes in elicitor-treated parsley cells. Plant Molecular Biology 12, 227-234. Spellman, M. W., McNeil, M . , Darvill, A. G. and Albersheim, P. (1983). Characterization of a structurally complex heptasaccharide isolated from the pectic polysaccharide rhamnogalacturonan 11. Carbohydrate Research 122, 131-153. Stacey, N. J., Roberts, K. and Knox, J. P. (1990). Patterns of expression of the JIM4 arabinogalactan protein epitope in cell cultures and during somatic embryogenesis in Daucus carota L. Planta 180, 285-292. Stekoll, M. and West, C. A. (1978). Purification and properties of an elicitor of castor bean phytoalexin from culture filtrates of the fungus Rhizopus stolonifer. Plant Physiology 61, 3 W 5 . Stevenson, T. T., Darvill, A. G. and Albersheim, P. (1988a). 3-Deoxy-~-lyxo-2heptulosaric acid, a component of the plant cell-wall polysaccharide rhamnogalacturonan-11. Carbohydrate Research 179, 26F288. Stevenson, T. T., Darvill, A. G. and Albersheim, P. (1988b). Structural features of the plant cell-wall polysaccharide rhamnogalacturonan-11. Carbohydrate Research 182, 207-226. Stoddart, R. W. and Northcote, D. H. (1967). Separation and measurement of microgram amounts of radioactive polysaccharides in metabolic experiments. Biochemical Journal 105, 61-63. Templeton, M. D. and Lamb, C. J. (1988). Elicitors and defence gene activation. Plant, Cell and Environment 11,395-401. Tepper, C. S. and Anderson, A. J. (1986). Two cultivars of bean display a differential response to extracellular components from Colletotrichum lindemuthianum. Physiological and Molecular Plant Pathology 29, 411420. Tepper, C. S. and Anderson, A. J. (1990). Interactions between pectic fragments

100

S. ALDINGTON A N D S. C. FRY

and extracellular components from the fungal pathogen Colletotrichum lindemuthianum. Physiological and Molecular Plant Pathology 36, 147-158. Termote, F., Rombouts, F. M. and Pilnik, W. (1977). Stabilization of cloud in pectinesterase active orange juice by pectic acid hydrolysates. Journalof Food Biochemistry 1, 15-34. Terry, M. E . , Jones, R. L. and Bonner, B. A. (1981). Soluble cell wall polysaccharides released from pea stems by centrifugation. I. Effect of auxin. Plant Physiology 68,531-537. Thain, J . F.,Doherty, H. M., Bowles, D. J. and Wildon, D . C. (1990).Oligosaccharides that induce proteinase inhibitor activity in tomato plants cause depolrisation of tomato leaf cells. Plant, Cell and Environment 13,560-574. Thomas, J. R., Darvill, A . G. and Albersheim, P. (1989). Isolation and structural characterization of the pectic polysaccharide rhamnogalacturonan I1 from walls of suspension-cultured rice cells. Carbohydrate Research 185,261-277. Tomiyama, K., Okamoto, H. and Katou, K. (1983). Effect of infection by Phytophthora infestans on the membrane potential of potato cells. Physiological Plant Pathology 22, 233-243. Tong, C. B., Labavitch, J. M. and Yang, S. F. (1986).The induction of ethylene production from pear cell culture by cell wall fragments. Plant Physiology 81, 929-930. Torgov, V. I., Nechaev, 0.A., Usov, A. I. and Shibaev, V. N. (1990). CAHTCJ TpacaXapHnHoro A neHTaCaXaPAAWOr0 4parMeHToB paCTUTeJIbHor0 KCAJIO~JIH)KaHa-IlOTeHUElanbH61X AHYA6HTOPOB ayKCAHa. [The synthesis Of the trisaccharide and pentasaccharide fragments of the plant xyloglucan, potential inhibitors of auxin.] Bfioopramyecmrr X W M H R 16,854-857. Tran Thanh Van, K. and Mutaftschiev, S. (1990).Signals influencing cell elongation, cell enlargment, cell division and morphogenesis. I n “Proceedings of the VII International Congress on Plant Tissue Cell Cultures”. (H. J. J. Nijkamp, L. H. W. van der Plas and J. van Aartrijk, eds), pp. 514519. Kluwer, Dordrecht. Tran Thanh Van, K., Toubart, P., Cousson, A., Darvill, A . G., Gollin, D. J . , Chelf, P. and Albersheim, P. (1985). Manipulation of the morphogenetic pathways of tobacco explants by oligosaccharins. Nature 314,615-617. Truelsen, T . A . and Wyndaele, R. (1992).Cellulase in tobaccocallus: regulation and purification. Journal of Plant Physiology (in press). Usui, T., Matsui, H. and Isobe, K. (1990). Enzymic synthesis of useful chitooligosaccharides utilizing transglycosylation by chitinolytic enzymes in a buffer containing ammonium sulfate. Carbohydrate Research 203, 65-77. Valent, B. S., Darvill, A. G., McNeil, M . , Robertsen, B. K. and Albersheim, P. (1980). A general and sensitive chemical method for sequencing the glycosyl residues of complex carbohydrates. Carbohydrate Research 79,165-192. Vance, C . P., Kirk, T. K. and Sherwood, R. T. (1980).Lignification a s a mechanism of disease resistance. Annual Reviews in Phytopathology 18,259-288. Vanderhoef, L.N. (1980). Auxin-regulated cell enlargement: is their action at the level of gene expression? I n “Genome Organization and Expression in Plants” (C. J. Leaver, ed.), pp. 159-173. Plenum, New York. Verma, D. P. S., Maclachlan, G. A . , Byme, H., and Ewings, D. (1975).Regulation and in vitro translation of messenger ribonucleic acid for cellulase from auxintreated pea epicotyls. Journal of Biological Chemistry 250, 101%1026. Vogeli, U . , Meins, F. and Boller, T. (1988).Co-ordinated regulation of chitinase and P-1,3-glucanase in bean leaves. Planta 174,364-372. Walker-Simmons, M., Hadwiger, L. and Ryan, C. A. (1983). Chitosans and pectic polysaccharides both induce the accumulation of the antifungal phytoalexin

OLIGOSACCHARINS

101

pisatin in pea pods and antinutrient proteinase inhibitors in tomato leaves. Biochemical and Biophysical Research Communications 110, 194-199. Walker-Simmons, M., Jin, D., West, C. A., Hadwiger, L. and Ryan, C. A. (1984). Comparison of proteinase inhibitor-inducing activities and phytoalexin elicitor activities of a pure fungal endopolygalacturonase, pectic fragments and chitosans. Plant Physiology 76, 833-836. Walker-Simmons, M. and Ryan, C. A. (1984). Proteinase inhibitor synthesis in tomato leaves. Induction by chitosan oligomers and chemically modified chitosan and chitin. Plant Physiology 76, 787-790. Walker-Simmons, M. and Ryan, C. A. (1986). Proteinase inhibitor I accumulation in tomato suspension cultures. Plant Physiology 80, 68-71. Weigel, H. (1963). Paper electrophoresis of carbohydrates. Advances in Carbohydrate Chemistry 18, 61-96. Wessels, J. G. H. (1986). Cell wall synthesis in apical hyphal growth. International Review of Cytology 104, 37-79. West, C. A. (1981). Fungal elicitors of the phytoalexin response in higher plants. Naturwissenschaften 68, 447457. West, C. A., Moesta, P., Jin, D . F., Lois, A. F. and Wickham, K. A. (1984). The role of pectic fragments of the plant cell wall in the response to biological stress. In “Cellular and Molecular Biology of Plant Stress” (J. J. Key and T. Kosuge, eds), pp. 335-350. Alan R. Liss, New York. Whistler, R. L. and BeMiller, J. N. (1958). Alkaline degradation of polysaccharides. Advances in Carbohydrate Chemistry 13, 289-329. Wijesundera, R. L. C., Bailey, J. A., Byrde, R. J. W. and Fielding, A. H. (1989). Cell wall degrading enzymes of Colletotrichum lindemuthianum: their role in the development of bean anthracnose. Physiological and Molecular Plant Pathology 34, 403-414. Wildon, D. C., Doherty, H. M., Eagles, G., Bowles, D. J. andThain, J. F. (1989). Systemic responses arising from localized heat stimuli in tomato plants. Annals of Botany 64,691-695. Yamazaki, N . , Fry, S. C., Darvill, A. G. and Albersheim, P. (1983). Host-pathogen interactions. XXIV. Fragments isolated from suspension-cultured sycamore cell walls inhibit the ability of the cells to incorporate [‘4C]leucine into proteins. Plant Physiology 72, 864-869. York, W. S., Darvill, A. G . and Albersheim, P. (1984). Inhibition of 2,4dichlorophenoxyacetic acid-stimulated elongation of pea stem segments by a xyloglucan oligosaccharide. Plant Physiology 75, 295-297. York, W,S., Oates, J. E., Van Halbeek, H., Albersheim, P., Tiller, P. R. and Dell, A. (1988). Location of the 0-acetyl substituents on a nonasaccharide repeating unit of sycamore extracellular xyloglucan. Carbohydrate Research 173, 113-132. Yoshikawa, M., Matama, M. and Masago, H. (1981). Release of a soluble phytoalexin elicitor from mycelial walls of Phytophthora megasperma var. sojae by soybean tissues. Plant Physiology 67, 1032-1035. Yoshikawa, M., Keen, N. T. and Wang, M.-C. (1983). A receptor on soybean membranes for a fungal elicitor of phytoalexin accumulation. Plant Physiology 73,497-506. Young, D. H. and Kauss, H. (1983). Release of calcium from suspension-cultured Glycine max cells by chitosan, other polycations, and polyamines in relation to effects on membrane permeability. Plant Physiology 73, 698-702. Young, D. H., Kohle, H. and Kauss, H. (1982). Effect of chitosan on membrane permeability of suspension-cultured Glycine max and Phaseolus vulgaris cells. Plant Physiology 70, 1449-1454.

This Page Intentionally Left Blank

Are Plant Hormones Involved in Root to Shoot Communication?

M . B . JACKSON Department of Agricultural Sciences. University of Bristol. AFRC Institute of Arable Crops Research. Long Ashton Research Station. Bristol BS18 9AF. UK

I.

Introduction

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

104

I1. The Hormone Message Concept . . . . . . . . . . . . . . . A . Different Kinds of Hormonal Message . . . . . . . . . . B . Quantifying Hormonal Messages in Transpiration Stream . C . Assessing Developmental Impact of Hormonal Messages . . . . I11. Evidence for Regulation of Root :Shoot Ratio by Roots A . Nutrient Control Theory . . . . . . . . . . . . . B . Shortcomings of Nutrient Control Theory . . . . . . C . Conclusions . . . . . . . . . . . . . . . . . . . .

.

. . . IV . Examples of Hormone-like Action of Roots on Shoots A . Early Research . . . . . . . . . . . . . . . . . B . Leaf Senescence . . . . . . . . . . . . . . . . . C . Shoot Extension. Photosynthesis and Flowering . . . D . Conclusions . . . . . . . . . . . . . . . . . . . .

. . . . . . .

V . Cytokinins . . . . . . . . . . . . . . . A . Introduction and Early Research . . . B . Development in Unstressed Plants . . C . Root Excision Studies . . . . . . . . D . Responses to Mineral Nutrient Shortage E . Effects of Other Stresses Applied to Roots F . Conclusions . . . . . . . . . . . . .

. . . . .

. . . . . .

. . . . . .

. . . . . . . . . .

. . . . . . .

. . . .

. .

. .

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

VI . Gibberellins . . . . . . . . . . . . . . . . . . . . . . . A . Introduction . . . . . . . . . . . . . . . . . . . . Advances in Botanical Research Vol . 19 ISBN Ck12-005919-3

. . . . .

106 106 107 111 112 112 113 116 117 117 117 120 122 123 123 125 128 131 133 138 138 138

Copyright 01993Academic Press Limited All rightsof reproductionin any form reserved

104

M. B. JACKSON

Studies on Unstressed Plants . . . . . . . . . . . . . . Effects of Root Excision and Environmental Stresses Applied to Roots . . . . . . . . . . . . . . . . . . . . . D. Conclusions . . . . . . . . . . . . . . . . . . . . .

139

Ethylene . . . A . Introduction B. Flooding . C. Conclusions

144 144 145 149

B. C.

VII.

VIII.

IX.

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

Abscisic Acid . . . . . . . . . . . . . A. Introduction . . . . . . . . . . . . B. Water Deficiency and Stomata1 Closure C. Water Deficiency and Leaf Expansion D. Soil Flooding . . . . . . . . . . . . E. Various Other Stresses . . . . . . . F. Conclusions . . . . . . . . . . . .

. . . .

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

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

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

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

Final Remarks . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References

142 143

149 149 150 159 160 164 166 166 168 168

I. INTRODUCTION In plants the growth and behaviour of the shoots are coupled closely to growth and behaviour of the roots. This is reflected in root : shoot dry weight ratios, which are predictable at various stages of development (Davidson, 1969) implying the existence of a controlling mechanism balancing the growth of above- and below-ground parts. This impression of regulation is reinforced by evidence that particular root :shoot ratios are genetically linked (Monyo and Whittington, 1970; McMichael and Quisenberry, 1991) and that they return towards earlier values after physically constricting the roots (Richards, 198l), droughting or defoliation (Blaikie and Mason, 1990), or after excision of parts of the shoot (Buttrose, 1966) or large portions of the root (Biddington and Dearman, 1984; Fig. 1). Superimposed on this in-built regulation of the root : shoot ratio is a marked, and sometimes dominating, influence of environmental conditions around the roots, for example, increases in the ratio caused by salinity (Kuiper et al., 1990), or by restricting external nitrogen or phosphorus concentrations to levels that inhibit shoots more than roots (Hunt, 1975). Stressful soil conditions are also known to bring about many rapid morphological changes to shoots such as leaf epinastic curvature, foliar senescence, adventitious rooting and stomata1 closure, in addition to a loss of growth in dry matter. The way in which roots sense shortcomings in the soil and transmit a response to the shoot, where growth patterns are changed in consequence, is of obvious botanical interest and important in the context of agricultural crop performance.

PLANT HORMONES AND COMMUNICATION

105

t No roots removed o 76% roots removed at start 0

I

I

0

I

1

7

I

I

14

Time (days)

Fig. 1. Effect of removing 76% by weight of the root system of 2-week-old barley plants (Hordeum vulgare L. cv. Midas) on subsequent changes in root: shoot dry weight ratio. By the end of 14 days the initially small root:shoot ratios of the root-pruned plants once more approached those of unpruned plants. Similar effects were seen at both warm (16°C) and cool (8°C) root temperatures. From M. B. Jackson and J . V. Lake (unpublished results).

This review considers evidence that roots regulate the root :shoot ratio and other aspects of shoot development in stressed and unstressed plants by influencing the passage of plant hormones [auxin, gibberellins (GAS), cytokinins, ethylene, abscisic acid (ABA)] or their precursors between roots and shoots. A broad review is useful at this time because of a resurgence of experimental work that explores hormone involvement in root-shoot relationships (Jackson, 1985a; Kuiper and Kuiper, 1988; Davies and Jeffcoat, 1990) as an alternative to regulation directly through changes in water or mineral supply. The review may also help assess, in full bibliographic context, the remarks of Trewavas (1986) condemning the idea that hormone traffic between roots and shoots can reliably carry morphogenetic information. Trewavas suggested that any regulatory information thus carried would be destroyed by “noise” generated by environmental irregularities.

106

M. B. JACKSON

Thus, on the one hand we have the view that environmental change may transmit morphogenetically active hormonal-like messages from roots to shoots, and on the other, the view that such environmental changes degrade information any messages might otherwise convey. Much of the present chapter assesses the evidence for the generation of hormonal messages by roots experiencing environmental change.

11. THE HORMONE MESSAGE CONCEPT A . DIFFERENT KINDS OF HORMONAL MESSAGE

In principle, roots can influence hormone levels in the shoot either by exporting one or more hormones or precursors (e.g. cytokinins or the ethylene precursor 1-aminocyclopropane-1-carboxylicacid), or by acting as sinks for phloem-mobile hormones produced in shoot tissues (e.g. ABA). On this basis, roots may generate or modify several kinds of hormonal message (Cannell and Jackson, 1981; Jackson, 1981). Firstly, they could increase their output of hormones (positive message) or decrease their output (negative message), or they may become less active sinks for hormones made by shoot tissues, thereby causing an accumulation in source tissue (accumulative message). It is also possible for roots to become more active sinks for hormones from the shoots (debit message). An example of the latter may be found in the stimulation of root growth resulting from infection with the parasitic weed, S t r i p hermonthica (Parker, 1984). Research into the hormonal basis of root-shoot relationships has largely ignored most of these different kinds of operational message, and usually incorporates only one of them-sometimes twc-in the design o r interpretation of experiments. Because the contents of the xylem and phloem are not entirely isolated from one another (Canny and McCully, 1988), there is a strong possibility of hormonal transfer between the two. In some plants, certain zones may exist where such exchange is facilitated by an especially close juxtaposition of xylem and phloem (McCully, 1990). Thus, hormonal messages may be recycled between root and shoot, as shown experimentally by Hoad (1978) for ABA in droughted plants, and discussed recently by Wolf e f al. (1990) and Hartung and Slovik (1991). In addition, xylem sap may become more alkaline (Hartung et al., 1990) and phloem less alkaline (Baier and Hartung, 1991) in stressed plants. Since several hormones, particularly ABA, are weak acids they become increasingly ionized as the pH rises, thus maintaining the inward diffusion gradient for undissociated molecules that are membrane permeable (Hartung et al., 1988). Xylem sap, if alkalinized by stress, may therefore act as a trap for hormones such as ABA in phloem or produced by cells surrounding the xylem along its entire length. These

PLANT HORMONES AND COMMUNICATION

107

findings warn us that change in the amount of a hormone in xylem sap does not necessarily mean that it originated in the roots.

B. QUANTIFYING HORMONAL MESSAGES IN TRANSPIRATION STREAM

Many experiments assess the concentration of hormones in the transpiration stream of intact plants. Unfortunately, it is extremely difficult to sample the transpiration stream directly, since under-pressures driving transpiration are destroyed if xylem elements are punctured for sampling. Thus, estimates are usually made indirectly, using measurements made on sap that flows from root systems under osmotic pressure, or arbitrarily chosen external pressures, after the shoot is removed (de-topped plants). The flow of osmotically-driven sap is rarely more than 10% of daytime transpiration. Surprisingly, much experimentation has ignored the effect this large reduction in sap flow rate will inevitably have on solute concentration, through a lessening of dilution. The effect of a varying flow rate of sap through the xylem of root systems on hormone concentration can be readily demonstrated using a pressure vessel (Fig. 2a). In unstressed tomato plants, greater pressures give faster sap flow which, over the range approximating to transpiration rates of whole plants, dilute endogenous ABA in a linear fashion. It follows that concentrations of hormones in slowly flowing sap from de-topped plants cannot be equated with those of the intact plant and will be overestimates of the true concentration present in the transpiration stream. Furthermore, relative differences in concentration between sap from de-topped roots previously exposed to various treatments are likely to be distorted by the differences in sap flow rates these treatments must often cause through altered hydraulic resistance and root metabolism. Authors sometimes calculate the likely delivery rate of a hormone from roots to shoots of whole plants by multiplying sap flow rate from de-topped roots by the hormone concentration (e.g. Beever and Woolhouse, 1973; Bradford and Yang, 1980a; Heindle et af., 1982; Carlson et al., 1987; Neuman et af.,1990; Meinzer et af.,1991). In relating this value to the whole plant they are, axiomatically, assuming that the faster sap flow of the intact plant would dilute hormone strictly in proportion to the rate of transpiration, with no effect on total delivery. This seems to be a reasonable assumption, and for flooded and well-drained tomato plants delivery rates remain approximately the same over a range of flow rates similar to those of whole plant transpiration (Fig. 2b). In support of this, Radin et af. (1982) found delivery of exogenous ABA into detached Gossypium hirsutum leaves to be similar at different leaf conductances (i.e. different rates of water flow). However, there are at least two reports of delivery increasing with faster sap flow (Wagner and Michael, 1971; Meinzer et al., 1991), suggesting that delivery rates are best worked out using sap flows

M. B. JACKSON

108

16C m -

-E

-z

12c

I

c 0 .c c

!

8

-* -

2

c

8C

2

4c

Control

C

Sap flow (mm3SS’) Fig. 2 . Effect of increasing flow of sap through detached root systems of tomato plants grown in well-aerated soil or in soil flooded for 12 h on (a) concentration of abscisic acid in xylem sap and (b) rate of delivery of abscisic acid from cut stumps. Arrows show transpiration rates of comparable intact plants. Means of four replicates. From M. A. Else, W. J . Davies and M. B. Jackson (unpublished results).

approaching those of intact plants. When calculating delivery rates, some workers make the mistake of assuming a no-dilution effect, and consider hormone delivery from roots to be increased in proportion to whole plant transpiration flow (e.g. King, 1976; Smith and Dale, 1988). This can lead to multiplying the concentration measured in slow-flowing sap by the faster flows of whole plant transpiration in calculating hormone delivery. This misunderstanding will inevitably lead to gross over-estimations of hormone delivery from roots to shoots (e.g. Skene, 1967). To avoid making this mistake, it is imperative to multiply concentration by the flow rate of the sap used for hormone measurement, and not by some other flow rate. Clearly, there is unnecessary confusion and contradiction in the literature concerning the interpretation of measurements of hormones in xylem sap. Authors’ claims, therefore, always require careful scrutiny. A related concern originates in the well-known observation that in experimentally or environmentally perturbed plants, transpiration is often slower than normal because of stomata1 closure. Thus, even if an increase in concentration in the transpiration stream is correctly estimated in the

PLANT HORMONES AND COMMUNICATION

109

stressed plants, it could merely indicate less dilution of the same amount of hormone, with no real change in message or hormone delivery to the shoots. Consequently, there is a risk that claims of increased hormone transport from roots to shoots based on concentration measurements may be untenable if all that is really happening is decreased dilution caused by stomata1 closure. Thus, it is of paramount importance that increases in delivery rates are demonstrated before claims of changes in a hormonal message can be accepted with confidence. Recording increases or decreases in concentration are, by themselves not sufficient evidence of changes in the passage of hormones into the xylem sap from roots, or elsewhere. The potentially confounding effects that different sap flow rates can have on estimates of hormone concentration in xylem sap of intact plants and delivery rates from roots to shoots may be addressed in several ways. One possibility is to measure hormones in sap samples obtained after pressurizing detached roots to generate sap flows close to those of comparable whole transpiring plants. Although this approach is highly desirable and has been endorsed recently (Incoll and Jewer, 1987; Meinzer et al., 1991), no results obtained in this way have been published for hormones, except in flooded and well-drained tomato plants (Fig. 2). An alternative approach has been to pressurize the roots to a level equivalent to the leaf water potentials of intact plants. However, results obtained this way have assumed similar leaf water potentials in control and stressed plants (e.g. Neuman et al. , 1990; Smit et al., 1990). This assumption will often be incorrect. A more straightforward method is to take the first few microlitres of sap issuing from the surface of a cut stump, since this may represent true transpiration fluid issuing from the roots. The method appears to be satisfactory in maize (Zhang and Davies, 1990a) but in tomatoes, the first 5&100 ~1 of sap are heavily contaminated by extra ABA generated as a result of applying a collecting tube over the cut stump (Else e f al., 1991). Several authors have expressly avoided these first droplets of sap because they suspected that they would be contaminated (Henson and Wareing, 1976; Heindle et al., 1982; Cahill et al., 1986). Clearly there are obvious risks inherent in this simplest of methods. Alternative techniques include expelling small volumes of sap from cut leaf veins by over-pressurizing the roots of whole plants with a specially designed pressure bomb (Passioura and Munns, 1984; Munns and King, 1988; Schurr and Gollan, 1990). This seems to be the best way of measuring what is arriving at a particular leaf (Munns, 1990; Schurr and Gollan, 1990). However, it is less helpful for assessing what is initially delivered to the shoot from the root system since there will be losses or gains on the way up to leaf, and also a differential partitioning of the delivered hormone between leaves in proportion to their individual rates of transpiration. Another possible approach is to extract the contents of the xylem vessels of excised stem segments by applying increased or reduced pressures or centrifuging in the hope of obtaining a sample with hormone content

110

M. B. JACKSON

similar to that of the transpiration stream of the source plant (Loveys, 1984a,b; Tromp and Ovaa, 1990; Trejo and Davies, 1991). However, some components of this sap have been shown to differ from those of the transpiration stream (Ferguson, 1980). This may be because the small volumes involved are highly susceptible to either enrichment or depletion in the time between excision and extraction by the biochemical activities of the relatively large mass of surrounding living cells (Nonhebel et al., 1985; Griggs et al., 1988), by partitioning between pools of different acidities, or by metabolism within cell walls (Loveys and Robinson, 1987). Crushed cells at the ends of such stem segments may also introduce artifacts (Zhang and Davies, 1990a). Yet another method for trying to obtain an authentic sample of the transpiration stream is to place individual leaves in a Scholander-type chamber and pressurize until sap issues from the xylem (Hartung et al., 1990; Tardieu et al., 1992~).Some of the expressed fluid will undoubtedly be captured transpiration stream. However, a variable proportion of this fluid will have been expressed from cell interiors rather than just from xylem vessels. This is because xylem water will inevitably have been withdrawn into leaf cells when underpressures were released on cutting the leaf from the intact plant. This movement of sap into cells, and then out again under the pressure of the Scholander bomb, may therefore result in a misleading hormone enrichment since sap re-emerging from the cells may have become contaminated with hormones and other solutes. Also, the volume of sap expressed for analysis (50-100 p1 minimum) may exceed the volume of leaf xylem (Hartung et al., 1988), at least when small leaves are used. In these circumstances, the sap sample will always contain material expelled from leaf cells. In all work with sap flows from de-topped root systems, it is highly desirable that hormone measurements are made as quickly as possible after wounding artifacts are over, that is, before root biochemistry is altered by a shortage of assimilates from the shoots. Many workers use sap delivered over long periods (3 days-Salama and Wareing, 1979; 24 h-Sattelmacher and Marshner, 1978) during which time root biochemistry will have changed considerably with inevitable consequences for hormone production. Modern assay procedures for hormones are highly sensitive and preclude the need for such long collection times. Possibly the most useful estimates of hormone levels in the transpiration stream would comprise the following elements: (1) a concentration in sap flowing at rates of whole plant transpiration at the time of sampling; (2) a delivery rate calculated by multiplying the sap flow rate and hormone concentration derived in (1) above; (3) a “specific” delivery rate that takes into account the delivery rate calculated as in (2), together with the size of the root system supplying the hormone message (e.g. Bradford and Yang, 1980a; Coleman etal., 1990) and the weight or area of shoot tissue into which it will be delivered. This would generate expressions such as mol of hormone

PLANT HORMONES A N D COMMUNICATION

111

delivered per g root per m2 of leaf per min. There a r e , as yet, no published data of this kind, even though such data have the great advantage of allowing meaningful and direct comparisons t o be made between plants of different sizes and root :shoot ratios, growing under different experimental treatments and transpiring a t different rates. C. ASSESSING DEVELOPMENTAL IMPACT OF HORMONAL MESSAGES

Demonstrating a change in hormone delivery in sap, or in tissue concentration is in itself insufficient evidence t o establish any physiological significance. T h e several different kinds of additional information needed t o d o this convincingly are summarized in Table I and based on previously TABLE I Criteria for implicating a hormone in the regulation of a naturally occurring develoumental uhenomenon Principal criteria (1) Correlation Joint occurrence between the timing of developmental changes and alterations to the endogenous hormone concentration. The ideal is precedence by the hormonal change, commensurate with its speed of action. ( 2 ) Duplication Reproduction of the phenomenon by re-creating quantitatively, changes in the internal concentrations of hormone measured when the process occurred naturally. ( 3 ) Deletion and re-instatement (a) Prevention or inhibition of the phenomenon by removing or decreasing the internal hormone titre by chemical means or by molecular genetics. (b) Reversing the effect of (a) by demonstrably re-instating quantitatively the original internal hormone levels. (c) Inhibiting the phenomenon by interfering with the action of the hormone, preferably using a non-toxic, competitive inhibitor, or by molecular genetics or mutation. (4) Chemical specificity Evidence in (1)-(3) should not apply to other substances found in plants, other than precursors. (5) Relevance to higher levels of organization Developmental process studied should occur beyond the confines of the laboratory and relate to performance in environments to which organs, whole plants or populations are naturally subjected to. ( 6 ) Relevance to lower levels of organization Association between the action of exogenously supplied hormone and the naturally occurring phenomenon is retained at cellular, subcellular and biochemical levels. Ideally criteria (1)-(4) should apply to each aspect examined. (7) Generality Extent to which the proposed hormonal controls apply to other taxonomic groups with similar developmental traits should be established. From Jackson (1987).

112

M. B. JACKSON

published rules of evidence (Jackson, 1987) that are modifications to those of Jacobs (1959) and Koch (1876). Unfortunately, it is extremely rare to find any experimental system where all these requirements have been met. The rules in Table I form a useful basis for judging the thoroughness of a piece of research and of the claims made for hormonal intervention in root-shoot relationships. Important among the requirements are tests of physiological activity using the most appropriate hormone dose. This could be one that matches quantitatively the concentrations in xylem sap close to the target tissues of intact plants. Preferably, it would be the amount of hormone delivered from the root by transpiration to known amounts of target tissue in the shoot. This exemplary approach has been rarely adopted (e.g. Loveys, 1984b). Obviously, such tests cannot be made meaningfully without first obtaining realistic estimates of hormone levels in the transpiration stream. As discussed above, achieving this is not straightforward, concentrations in sap issuing from de-topped root systems at rates well below those of transpiration being especially misleading. Unfortunately, several workers have used concentrations in sap obtained in this way as a guide to choosing the concentrations of hormones for testing physiological potency (e .g. Smith and Dale, 1988; Nooden et al., 1990a,b).

111. EVIDENCE FOR REGULATION OF RO0T:SHOOT RATIO BY ROOTS As discussed in Section I, the size relationship between root and shoot is controlled by a combination of intrinsic and external factors. The questions here are whether roots themselves regulate the root:shoot ratio, and the extent to which any effect they have is based on factors other than nutrient or water supply (i.e. hormonal factors). A. NUTRIENT CONTROL THEORY

The division of labour between roots and shoots, arising from their very different environmental resources and absorbing properties, suggests that each might control the size of the other by limiting the supply of one or more constituents to the dependent part (i.e. negative control). Clearly, roots could not expand in size beyond the quantity of carbon assimilates available from the shoot, while it is inconceivable that shoot growth could proceed entirely unchecked by a finite supply of water or minerals available from the root. The root:shoot ratio of plants acclimatized to a given environment may thus be little more than the size ratio at which supply and demand of basic constituents from the environment are in equilibrium as roots and shoots continue to grow. The root :shoot ratio may, thus, be underpinned by a balance between the functional efficiency of roots and shoots along the

PLANT HORMONES AND COMMUNICATION

113

lines expressed arithmetically by Davidson (1969) as: root mass x rateabsorptionx k = shoot mass x ratephotosynthesis, where k is a constant representing the amount of mineral nutrient or water used per unit amount of carbon increment. Thornley (1972) has extended this approach by considering that shoot and root growth each separately depend on the concentration of carbon and minerals (mainly nirogen), which in turn is determined by how much of each is received from the supplying part. The amounts root and shoot receive presumably reflects the relative competitive abilities of root and shoot and their individual absorption capacity, e.g. for nitrogen (McDonald et ul., 1986). Wilson (1988) concluded that the Thornley model largely accounted for decreases in the root :shoot ratio seen in many plants as they grow larger, and for increases in the ratio in the face of deficits in water and major inorganic nutrients. An example of the latter effect is seen in the experiments of Hunt (1975), where the functional efficiency of roots for absorbing and supplying nitrogen was depressed by diluting the nitrogen concentration in the rooting medium. In the context of an overall decrease in plant growth, nitrogen shortage slowed root growth less than shoot growth, thereby increasing the root : shoot ratio. This is to be expected if the amount of available nitrogen falls away steeply with distance from the site of uptake, thus starving the shoot more than the root. The power of mineral nutrition to affect the root:shoot ratio is highlighted by a report that the ratio of root:shoot of individual but joined tillers of Curex can be made to vary 10-fold simply by contriving a different level of soil fertility around each tiller (Harper, 1977). The sophisticated regulation of mineral availability in solution cultures in the experiments of Ingestad and Lund (1986) also illustrates the controlling influence that nutrient supply can have on the absolute and relative growth rates of plant parts. Thus, in reviewing the problem, Wilson (1988) had little difficulty in rejecting hormonal controls in favour of mineral supply as the principal means by which roots influence the root :shoot dry weight ratio.

B. SHORTCOMINGS OF NUTRIENT CONTROL THEORY

There are problems with the idea that mineral supply from roots sensitively maintains a size of shoot that is in a predictable proportion to the size of the root. The notion necessarily rests on the principle that the mineral supply rate per unit weight of roots remains reasonably constant. Only if this holds true can mineral supply be a reliable indicator of root size, and thus a sensitive regulator of the root : shoot ratio. However, under conditions where mineral ions are readily available to roots, as in a complete nutrient solution, specific supply rates of nitrogen phosphorus and potassium (i.e. mg transported to the shoot per g dry weight of root) are greatly decreased if a large proportion of the roots (75-80%) is removed (Table 11). Thus, delivery

114

M. B. JACKSON

TABLE I1 Effect of pruning 7 5 4 0 % of roots from 7-day-old plants of barley (Hordeum vulgare), maize (Zea mays) or oilseed rape (Brassicd napus) on the specific supply rate of mineral nutrients to the shoot over the following 7 days. The final concentration of the nutrients in the shoot are also Riven

Parameter

Nitrogen

Phosphorous

Potassium

Control Pruned Control Pruned Control Pruned Specific supply rate (mgg-' root day-') Barley Maize Oilseed rape

18.3 18.7 53.0

13.0 8.5 13.0

2.8 3.8 4.4

1.5 0.9 0.7

19.6 25.7 37.2

17.7 16.7 9.9

62.4 47.4 65.1

55.7 38.8 63.3

8.7 10.2 5.9

6.5 7.4 5.7

68.0 61.7 48.6

68.2 55.1 50.7

Concentration in shoot (mg g-' dry wt)

Barley Maize

Oilseed rape

Means of 8 plants. 75% of roots were pruned from maize, 77% from barley and 80% from oilseed rape (dry weight basis). From J . V. Lake, K. Brown and M. B. Jackson (unpublished results).

rates of minerals per unit amount of root are not a reliable basis by which shoots may gain some measure of root size, even when external nutrient concentrations remain unchanged. Under such circumstances the fall in specific uptake rates by roots remaining and re-growing after substantial root pruning suggests that this parameter is governed more by a disproportionate decrease in shoot growth and, thus, shoot mineral demand rather then being a property of the roots themselves (Table 11). Such a feedback mechanism is well established (Pitman, 1972). This reasoning leads to the conclusion that the disproportionate decrease in shoot growth must have some cause other than mineral shortage. The dominance of the shoot in determining mineral absorption is illustrated by the relative consistency of mineral nutrient concentration in the shoot despite root pruning (Table 11), even though the mineral supply per unit weight of root is depressed greatly in the three species examined. A further difficulty is that surprisingly large reductions in root size, achieved by pruning, can be sustained with little change in shoot growth. For example, Tschaplinski and Blake (1985) reduced the weight of the root system of alder (Alnus glutinosa) seedlings from 1.73 g to 0.70 g by constricting them for 96 days using perforated polypropylene tubes submerged in aerated nutrient solution. This had almost no effect on growth in shoot weight, other aspects of shoot development or shoot-water relationships. Similarly, Das Gupta (1972) removed half the roots in the vertical plane from 28-day-old seedlings of sugar beet (Beta vulgaris) without reducing

PLANT HORMONES A N D COMMUNICATION

115

significantly the relative growth rate of the shoot over the subsequent 21 days. Buttrose and Mullins (1968) pruned the roots of grape vine (Vitis vinifera) to approximately 75% of the weight of unpruned control plants every week for 8 weeks without any statistically significant effect on shoot growth, while Humphries (1958) removed up to half the roots from barley or rye with no effect on shoot growth. There are other reports where shoot growth is restrained by pruning 50% or less of the roots (Humphries, 1959; Buttrose and Mullins, 1968; Carmi, 1986) or by severely constricting roots with small containers (Krizek et a f . , 1985; Ruff ef a f . , 1987) or cages suspended in nutrient solution (Richards and Rowe, 1977; Richards, 1981). However, recent careful re-examination of the latter approaches suggests that such effects may have been mediated by gross interference with water relations (Hameed et a f . , 1987; Tschaplinski and Blake, 1985) or root metabolism through restricting oxygen availability (Peterson et a f . , 1991), thus complicating interpretation. The few published examples of reduced growth in shoot dry weight from more modest root pruning (Humphries, 1959; Buttrose and Mullins, 1968; Carmi, 1986) are unlikely to be the result of water deficits since hydraulic conductance is increased rather than decreased by removing roots (Milligan and Dale, 1988). Thus, in certain situations or species, root size can be more sensitively measured than in others, but generally, size perception by the shoot is poor. We may conclude that while mineral supply is an important mechanism by which roots limit shoot growth, it is not necessarily a sensitive indicator of root size and thus an unlikely basis for the sensitive regulation of the root :shoot ratio. Indeed, it may be that shoots exercise a stronger control over the root:shoot ratio than the roots. This is suggested by grafting experiments between dwarf and normal sized tomato cultivars (van Staden ef al., 1987), by the influence of the shoot on specific uptake rates by roots (Pitman, 1972), and by reports that considerable amounts of shoot growth can sometimes occur in the absence of roots (Killingbeck, 1990). Presumably, in the short term, growing parts of the shoot can obtain minerals from non-growing shoot tissues, thereby displaying a limited degree of independence from the roots. It now remains to ask if there is any evidence that roots can influence growth of shoot dry mass in ways other than through the supply of minerals per se. Of course, failure to match demand for water by evaporation from the foliage will be one rather crude mechanism. Clues that other more subtle mechanisms exists include the simple observation by Vanden Driessche and Wareing (1966) that tree seedlings growing in a range of nutrient solution concentrations exhibited very different relative growth rates, despite having closely similar internal mineral nutrient concentrations. Thus, species differences in responses to a wide range of mineral nutrient concentrations cannot be attributed to differing requirements for minerals themselves as part of the biochemical and structural requirements for growth. Similar

116

M. B. JACKSON

conclusions were drawn by Kuiper and Staal(l987) from work with different species of Plantago, which were found to have very different responsiveness to changes in mineral supply. Thus, factors are at work that influence what Ingestad has termed “nutrient productivity”, i.e. the amount of growth sustained by a given availability of minerals. In his experiments, “nutrient productivity” is shown as the slope of relative growth rate against the internal nutrient concentration expressed as a percentage of the optimum concentration for growth (Ingestad and Lund, 1986). This expression is roughly analogous to the constant in Davidson’s equation (Section 1II.A; Davidson, 1969) expressing the size relationship between root and shoot. Application of hormone analogues such as benzyladenine has been found to increase the amount of growth in shoot dry mass per unit of absorbed mineral nutrient or of shoot mineral content (Richards and Rowe, 1977; Richards, 1978), suggesting that endogenous hormones could influence “nutrient productivity”. Furthermore, there are reports of growth responses in shoots to mineral deprivation that may be too fast or too slow to be explained easily in terms of direct mineral shortage in growing shoot cells (Kuiper and Staal, 1987).

C . CONCLUSIONS

Overall, it appears that root activities may have less to do with regulating the root :shoot dry weight ratio than the shoot. This may explain the apparent insensitivity of shoots to quite large reductions in the size of root systems well-supplied with minerals and water. This has serious consequences for the commonly held notion that reductions in growth in shoot dry mass caused by stress in the root environment are a result of decreased output of growth promoting hormones from the roots (negative messages). Shoot growth (in dry weight at least) is clearly often insensitive to a loss in putative output of hormonal growth promoters (e.g. cytokinins) equivalent to that emanating from at least 25% of the root system. Roots can of course influence shoot through mineral supply, especially where amounts in the rooting medium are already limiting overall plant growth. However, there is little direct support for believing that roots sensitively regulate the root :shoot ratio with minerals or with hormones. Some indirect evidence suggests that hormones could influence the extent to which growth in dry mass by shoots responds to a given mineral input from the roots. But, this is far from indicating a mechanism for regulating closely the root :shoot ratio. There remains a large body of evidence that particular aspects of shoot behaviour (e.g. stem extension, senescence, flowering, stomata1 apertures and leaf expansion) can be affected by roots in ways not readily explained in terms of water or mineral supply. Evidence for this is assessed in the subsequent sections.

PLANT HORMONES AND COMMUNICATION

117

IV. EXAMPLES OF HORMONE-LIKE ACTION OF ROOTS ON SHOOTS A. EARLY RESEARCH

In the “Silliman Lectures” to Yale University in 1937, Chibnall (1939) was “. . . tempted to suggest that some influence of the root system, possibly hormonic, is responsible for the regulation of protein levels in leaves”, although it was not until 1954 that his experimental results supporting such an idea were published (Chibnall, 1954). By that time other evidence supporting this hypothesis had appeared. For example, Went (1938a) reported an elongation-promoting effect on dark-grown pea shoots by roots extending into demineralized water. When shoots were grafted onto root systems in the dark, they re-commenced extension only when cell continuity across the graft union was established (Went, 1938b). He assumed that the role of the roots was to supply a graft-transmissible growth factor (“caulocaline”) rather than to drain the shoot of growth inhibitor. Working with light-grown tomato plants, Went (1943) demonstrated that an adventitious root system growing in nutrient-free peat and comprising approximately 10% of the mass of the main root system below, was capable of doubling the rate of stem extension. Similar tests were done with plants with vertically divided root systems. When one of the two “half” root systems was grown in well-aerated nutrient-free peat, gravel or pumice instead of nutrient solution, leaf senescence was retarded and stem extension promoted. Similar but simpler experiments have also been published by D e Ropp (1946), Galston (1948), Jordan and Skoog (1971) and Sabanek et al. (1985). Detached leaves of some species such as Phaseolus vulgaris or Nicotiana rustica can be induced to root in water. Mothes and Engelbrecht (1956) noted a coincidence between the arrest of protein decline in the lamina and the emergence of roots from the petiole. Humphries (1963) and Humphries and Thorne (1964) measured very slow rates of photosynthesis until just before the new roots emerged, when rates increased coincidentally with a halt in the decline of leaf chlorophyll. Along similar lines, Parthier (1964) showed that the ability of isolated leaves to incorporate l4C-rnethionine into protein declined steadily until the precise time that new adventitious roots first appeared when, from this point on, ‘‘C-incorporation rates increased markedly. The presence of roots clearly has influences on shoot behaviour that cannot easily be explained in terms of mineral nutrition. B. LEAF SENESCENCE

When a large proportion of a root system is removed so that the root :shoot ratio is suddenly very much reduced, shoot growth may not be inhibited

118

M. B. JACKSON

proportionately to the reduction in the supply of minerals from the root (Humphries, 1959). Similarly, when Topa and Mcleod (1986) withdrew oxygen for 15 days from roots of flooding tolerant Pinus serotinu and P. tuedu, mineral uptake into the shoots was depressed but with no associated reduction in growth of shoot dry weight. It is likely that in both these cases minerals used to sustain growth by the younger parts were provided by senescing leaves. In support of this interpretation, Darrall and Wareing (1981) showed that shoots of birch growing in nitrogen-deficient sand continued to grow with smaller than usual whole-shoot mineral concentrations. This appeared possible because younger growing parts obtained nitrogen from older senescing leaves. Of course, leaf senescence is atypical of more vigorous, healthy plants, even though their shoot growth may still be restrained by nutrient supply. The reason for the lack of senescence in the older leaves may be connected with the presence of a sizeable and healthy root system per se and less with the continued supply of minerals, for which mature leaves have in any case little or no requirement. The principle is made clear by the work of Ingestad (1982), which showed that shoot growth can be constrained by slow mineral supply over long periods without symptoms of leaf senescence, provided that the external supply is increased regularly to stay in proportion to the growth rate. Thus, total shoot size can be constrained by mineral supply in the absence of leaf senescence, but in association with the presence of an enlarged root system (Fig. 3). How do these roots prevent leaf senescence in the face of less than optimal mineral nutrition? The possibilities are that a full complement of healthy roots drains the shoot of some senescence-promoting influence, or alternatively, that a complete and healthy root system supplies one or more senescenceretarding metabolites, possibly hormones. Thus, one explanation for leaf senescence when roots are stressed is that the export of some senescence-promoter from the shoot system is slowed sufficiently enough for it to accumulate in leaves to damaging levels. It is well known that removing metabolic sinks increases sucrose in photosynthesizing leaves (Neales and Incoll, 1968; Farrar and Farrar, 1985). This in turn might inhibit photosynthesis (Herold, 1980) and damage chloroplasts through an excessive build-up of starch. However, Lawrence and Strangeways (1985) have shown the injurious effects of exogenously supplied sucrose to be artifactual and that endogenously generated increases in the sugar are innocuous. Similarly, Humphries (1963) followed the accumulation of assimilate in detached, rooted leaves over several weeks. Net assimilation rate (dry weight increase cmP2 week-’) remained unchanged despite the steady accumulation of dry matter in the lamina, resulting from the absence of growing sinks in the “shoot”. Weights of up to 7.4 mg cmP2were found, which is much greater than those in intact plants. However, leaf yellowing was seen only towards the end of the experiment, indicating how tolerant non-growing leaves can be of accumulated photosynthate and presumably

119

PLANT HORMONES AND COMMUNICATION

Leaves Stem X Root

0

E

60

.-cn

s??

0 K cn ._

5 40

H

. I -

-0m c

0

c

;Iz

20

0

I

I

I

1

5

10

15

Sub-optirnum (RNYO)

I

I

2 0 .E 5 1 c

0"

1

z2

.c

(1

0

Fig. 3. Dry weights of roots, shoots and stems of seedlings of birch grown in nutrient solutions replenished by frequent additions of mineral nutrients, with nitrogen given at different rates. After Ingestad and Lund (1979).

other metabolites. The latter may well have included the hormone abscisic acid, which accumulates in detached leaves (Jackson and Hall, 1987) or de-rooted seedlings (Smith and Dale, 1988), but not when they are senescent (Osborne et al., 1972). Another way to test if accumulations of substances in the leaf prompts senescence is to kill the phloem connection between root and leaf by girdling. This allows the inward movement of water and solutes from the root in the xylem while preventing outflow of materials. Kulaeva (1962) showed convincingly that girdling using steam does not simulate the senescence-promoting effects of de-rooting in Nicotiana. She also showed that roots are able to exert a senescence-retarding effect when connected to the leaves by xylem alone (see also the girdling experiments of Carmi and Koller, 1979). It seems unlikely that in plants with environmentally damaged or partially removed root systems, leaf senescence is neither explicable by mineral shortage nor by the accumulation of some senescence promoter. Changes in the output of senescence regulators from the roots seems more probable.

120

M. B. JACKSON C. SHOOT EXTENSION, PHOTOSYNTHESIS AND FLOWERING

Wareing and Nasr (1961) and Smith and Wareing (1964) examined the possibility that roots exercise control over shoot extension by means other than inorganic nutrition or by draining the shoot or growth inhibitors. They exploited the potential of shoot cuttings of willow (Safix virninafis),rooted at their basal end, to sprout axillary shoots on top of a loop formed by bending the stem (zone 1 in Fig. 4). It proved possible to induce axillary shoots to sprout in a second location along the stem if a second set of adventitious roots was induced to form further up by embedding a short length of stem in damp vermiculite. For example, axillary shoots at the tip of the stem (zone 2, Fig. 4) grew only if an additional set of adventitious roots were induced at the bottom of the loop (position B, Fig. 4). The effect was not explicable in

Total Iongth of sideshoots (cm)

Zone2

I

/ /

zone 1 1

zone 2

Roots at A only

1240

140

Roots at A & B (- minerals)

060

400

Roots at A & B (+ minerals)

000

I

660

Fig.4. Effect of inducing a second adventitious root system on cuttings of willow (Saliw viminalis) trained in a circle. With only basal roots present (position A), axillary shoots emerged mainly from zone 1. When a second set of roots was induced to grow out into a mineral-free medium, axillary buds were also induced to grow in zone 2. Based on results of Smith and Wareing (1964).

PLANT HORMONES AND COMMUNICATION

121

terms of enhanced mineral nutrient supply since roots were highly active even when grown in deionized water. That the roots acted as a sink for a phloem-mobile growth inhibitor was also discounted because the roots remained effective even when the stem between the roots and potentially responsive lateral buds was bark-girdled. In contrast, removing the roots, as they formed, prevented axillary bud outgrowth. Smith and Wareing (1964) concluded, quite reasonably, that roots release into the xylem some hormonal influence that is transported by transpiration or root pressure flow to sites of action upstream of the roots. Similar evidence for hormone-like effect of roots on shoots is to be found in the sequential rise and fall in net carbon dioxide fixation associated with the outgrowth of whorls of nodal roots of Sorghum saccharum (Jesko et al., 1971; Jesko, 1972) and Zea mays (Jesko and Vizarova, 1980). The effect occurs prior to the emergence of roots through the covering of leaf bases, and thus precedes their development into major sinks, or as organs of mineral or water uptake. Jesko and colleagues concluded that the young roots donate a stimulus for photosynthesis. Surprisingly, the results show that leaf expansion slows during these brief surges in photosynthesis. Roots can also influence flowering in ways that strongly suggest a hormone-like influence. Chailakhian (1961) showed that removing roots from Rudbeckia tricolour, a long-day plant, prevents flowering in response to long days. Roots also influence flowering in Scrofularia arguta,a long-day plant, or in the short-day species Perilla ocimoides and Chenopodium polyspermum. In each case, isolated vegetative buds, cultured in vitro, flower quickly even in non-inductive daylengths, provided the explants remain free of roots (Miginiac and Sotta, 1985, and references cited therein). In Silene, the need for long days can be obviated by first exposing only the roots to high temperatures (Wellensiek, 1985). In related work, with Cichorium intybus, a vernalizing treatment is needed before long days can induce flowering (Joseph, 1984, cited Miginiac and Sotta, 1985). From grafting experiments (Fig. 5) it is clear that it is the roots which respond to the chilling not the potentially flowering bud. O n the other hand, non-vernalized roots can inhibit flowering of a vernalized bud to which it is grafted. These examples point to a supression, by cold or by heat, of flowering inhibitor production in roots. The rooting behaviour of stem cuttings of blackcurrant (Ribes nigrum) also points to a flowering inhibitor produced by roots. Schwabe and Al-Doori (1973) found that cuttings cut down to less than 20 nodes long before being rooted will not flower in normally inductive short days. The effect was related to the distance between the root system and the shoot apex where flowering occurs. On longer shoots, with more than 20 nodes, flowering occurred in the upper buds in short days. However, this could be prevented by inducing a second set of adventitious rooting closer to the shoot tip. Other examples of hormone-like influences of roots on flowering are given by Bernier and Kinet (1986).

122

M. B. JACKSON 18°C Photoperiod:16h

f

Yo of vering bumS A

B

0

0

85

92

0

0

I oa

00

- -

Crossedgrafts

Autografts

c 1 e-

- -

Fig. 5. Flower initiation in apical buds of Cichoriurn inlybus cultured in vifro. To form a flower in a 16 h photoperiod the original explant comprising root ( V ) and shoot ( A ) requires 8 weeks vernalization at 3°C. Grafting combinations show that it is roots that sense chilling and transmit the response to buds across a graft union. Redrawn from Miginiac and Sotta (1985).

D. CONCLUSIONS

Even if the known plant hormones had remained undiscovered, the examples and arguments presented would still constitute a strong case for believing that hormone-like influence are involved and that mineral and water supply or assimilate accumulation are not the sole means by which roots influence shoots. The mere presence of growing roots would seem to exert a promoting influence on shoot extension and photosynthesis, and to stimulate or prevent flowering, although these phenomena are less thoroughly researched than that of the senescence-retarding effect roots have on leaves. The latter effect is difficult to explain exclusively in terms of mineral nutrient supply. Girdling experiments suggest that roots exert their effect by changes in the export of one or more senescence-retarding hormones (positive or negative messages) rather than modulating the withdrawal of a senescence promoter produced by the shoot (accumulative and debit messages). However, if the latter mechanism does operate, sucrose is unlikely to be the senescence promoter involved. Subsequent sections will explore the extent to which the influences of roots on shoots can be explained by changes in amounts and actions of one or more of the five main classes of plant hormone (cytokinins, gibberellins, abscisic acid, auxin, ethylene). Each will be dealt with separately, except for auxin, which although present in both xylem and phloem sap (Hall and

PLANT HORMONES AND COMMUNICATION

123

Medlow, 1974) has attracted relatively little attention. The most common experimental approach has been to apply a stress to the roots in attempts t o alter qualitatively or quantitatively the input or output of hormones by the roots and to link these with developmental effects of the stress.

V. CYTOKININS A . INTRODUCTION AND EARLY RESEARCH

Chemical and physiological characterization of these purine hormones is not straightforward. Most possess a side-chain of five carbons at the N6 position that can become conjugated by glycosylation. Amongst several biological properties, cytokinins must promote cell division, e.g. in soybean callus, to satisfy the definition. Other recognized effects of exogenously applied cytokinins include the retardation of leaf senescence and the promotion of stornatal opening, lateral bud outgrowth, shoot extension and seed germination (Horgan, 1984). Chemically, most cytokinins can be classified into three groups, each based on one of three bases, namely zeatin, dihydrozeatin and isopentenyl adenine. The bases exist free and as ribosides (i.e. ribofuranosyls), glucosides (i.e. glucopyranosyls of base o r riboside) o r ribotides (i.e. phosphate derivatives of ribosides). At least 30 such compounds have been identified by gas chromatography/mass spectrometry (GC-MS), although only a few plant species have been comprehensively analysed (McGaw, 1987). Positive identification of zeatin riboside and other cytokinins in xylem sap by GC-MS has been reported for Acer (Horgan et al., 1973), Lycopersicon esculentum (van Staden and Menary, 1976) and Phaseolus vulgaris (Palmer and Wong, 1985), amongst other plant species. The demanding technology needed for such analyses, coupled with the difficulty of obtaining internal standards for GC-MS, and the problem of deciding which one out of so many cytokinins to measure, may explain the decline in the number of papers published on cytokinins and root-shoot relationships in recent years. Radio- or enzyme linked-immunoassays are a convenient and rapid alternative to GC-MS but kits can be obtained commercially only for zeatin-based cytokinins. Furthermore, crossreactivity with related conjugates requires their prior separation by chromatography before quantification is reliable (Incoll et al., 1990). Xylem sap can contain a range of cytokinins (Hall et al., 1987); this increases the complexity of the analyses considerably. In the preceding era of the bioassay, plant science laboratories could obtain the requirements for cytokinin measurements easily and relatively cheaply. This accessible technology underpins a sizeable literature (van Staden and Davey, 1979) inspired by the idea that cytokinins are carried by transpiration in physiologically meaningful amounts from the many

124

M. B . JACKSON

thousands of root tips present on a typical root system (Miller, 1916) to recipient shoot tissues. Much of the published work contains measurements of cytokinin activity in unpurified or crudely purified samples of xylem sap, most often obtained from de-topped plants, and in ways criticised unfavourably in the preceding Section 1I.B. Such measurements cannot distinguish between the presence of cytokinin precursors and fully active cytokinins. At best they provide estimates of net cytokinin-like activity remaining after the contrary effects of any inhibitory substances present. These are useful physiological measures, in principle, but they are likely to overestimate net cytokinin concentrations because the diluting effect of the transpiration stream has, for the most part, been ignored. As early as 1943, Went deduced that it was unlikely that auxin from roots sustains the shoot. By the late 1950s it had become accepted that other hormones, most notably the cytokinins and gibberellins, with different physiological properties from auxin, were probably active in plants. The ability of applications of kinetin, a synthetic cytokinin, to retard senescence in detached (i.e. rootless) leaves (Richmond and Lang, 1957; Mothes et al., 1959; Mothes and Engelbrecht, 1963) and to stimulate stem extension, was reminiscent of the effects of roots on shoots. Earlier, Went and Bonner (1943) had shown that coconut milk (liquid endosperm of Cocos nuciferu) could overcome the inhibition of stem extension caused by removing the roots from dark-grown tomato seedlings. Coconut milk is now known to contain zeatin and zeatin riboside (Loeffler and van Overbeek, 1964; van Staden and Drewes, 1975) and to promote cell division in tissue cultures, one of the physiological tests for the presence of cytokinins. These early results suggested that cytokinins might be formed in roots and influence shoots after transport in xylem sap. Accordingly, in 1962, Kulaeva (1962) reported cytokinin activity, measured by the ability to retard leaf senescence, to be present in crude xylem sap from Nicotiana rusrica. This paper was followed by a similar report of senescence-retarding and cell divisionpromoting activity in two fractions of xylem sap of sunflower partially purified by paper chromatography (Kende, 1964,1965). At about the same time, extracts from root rips of sunflower were found to be especially rich in cytokinin activity (Weiss and Vaadia, 1965), and decapitated root systems were shown to be capable of net production of cytokinins over 24 h (Henson and Wareing, 1976). Later, aseptically cultured roots were shown to be capable of making cytokinins independently of shoots (Griffaut, 1977; Chen and Petschow, 1978; van Staden and Smith, 1979; Butcher et al., 1988). Jesko (Jesko and Vizarova, 1980; Jesko, 1981) detected pulses of cytokinin output from successive nodal roots of Zea mays as they emerge, in association with temporary increases in photosynthesis. Thus, roots became to be recognized as rich sources of cytokinins, and it is widely believed they are the main source of these hormones in plants. Of course, roots are not uniquely capable of making cytokinins. Shoot tissues can also produce these

PLANT HORMONES AND COMMUNICATION

125

hormones (Wang and Wareing, 1979) and even enrich xylem sap with them (van Staden and Dimalla, 1980). Many experiments have used environmental stress applied to roots as a means of decreasing their cytokinin production and assessing its importance in shoot development. An early paper along these lines was that of Itai and Vaadia (1965), who found that drought stress promoted leaf senescence in association with a marked fall in the cytokinin activity of sap bleeding from de-topped sunflowers. However, possible links between cytokinin output from roots and normal events in shoots, such as spring bud break or leaf senescence in monocarpic plants, has also been investigated. This is reviewed next. B. DEVELOPMENT IN UNSTRESSED PLANTS

1. Spring sap in woody plants

Osmotically driven xylem sap flowing vigorously from the cut surface of freshly de-topped trees in late winter or early spring is the so-called spring sap. The flow precedes shoot growth and leaf emergence and thus is undiluted by transpiration and consequently rich in cytokinin activity (Luckwill and Whyte, 1968; Reid and Burrows, 1968). This activity is thought to promote bud burst, stem extension and early leaf expansion. In support of this possibility, Jones (1973) found cytokinin activity in spring sap of apple that co-chromatographed with zeatin riboside. When sap was applied to rootless shoots in vitro, at concentrations estimated at 0.5 p ~it ,substituted for the missing roots by stimulating extension and delaying leaf senescence. Authentic zeatin riboside had similar effects at 0.5 p ~In.more recent work, cytokinin concentrations in sap, obtained from excised shoots by air displacement, have been found to increase in February, before buds start to grow (Tromp and Ovaa, 1990), suggesting a causal role for the cytokinins. These levels (100-300 n ~ were ) sustained until leaf emergence in May, before declining substantially in July, and remaining low until the following spring. It is by no means certain that the cytokinins originated in the roots and Tromp and Ovaa suggest bark as a source, where hydrolysis of storage proteins also commences in February. Dilution rather than reduced production by roots may explain the decreases in concentration seen once the leaves emerged and transpiration commenced. Immunoassays of HPLCpurified samples for isopentenyl-adenine, zeatin, zeatin riboside and ribotide showed zeatin to predominate except in April, when the riboside was dominant (Tromp and Ovaa, 1990). There are other reports confirming the presence of cytokinins in xylem sap of woody species detected by GC-MS (Purse et al., 1976; Waseem et al., 1991) or bioassay (Ahokas, 1984) but measuring cytokinin delivery rates from roots and testing the physiological significance of the cytokinins has been non-existent. Thus, the case for believing that cytokinins in spring sap play a role in controlling early shoot

126

M. B. JACKSON

growth is based almost entirely on the fact that they are present in the sap at high concentrations at the time buds start to grow in the spring. 2. Leaf senescence and related processes Sitton et al. (1967a) reported that cytokinin activity bioassayed in sap exuding from de-topped sunflower plants increased during exponential plant growth, but dropped by 90% when roots stopped growing and leaves began to senesce, thus suggesting a causal relationship. Similarly, Davey and van Staden (1976) reported a fall in zeatin riboside concentration in tomato plants during flower formation and thus prior to leaf senescence. Nooden and co-workers have sought stronger evidence of causality in studies of leaf senescence in soybean (Clycine max L.) (reviewed in NoodCn et al., 1990a). Four cytokinins active in retarding leaf senescence (zeatin, zeatin riboside, dihydrozeatin and dihydrozeatin riboside) were quantified using thin-layer chromatography, column chromatography and radioimmunoassays, before and during the time leaves senesced (NoodCn et a l . , 1990b). Concentrations measured were compared to those needed to retard senescence in an explant assay in which the cytokinins were administered, in the presence of minerals, at concentrations found in the xylem sap (Mauk et al. , 1990). Concentrations of zeatin riboside and other cytokinins in xylem sap from pressurized roots of de-topped plants were shown to fall dramatically at the time pods were fully extended (a reduction of 89%) and to remain at this low value (approximately 10-20n~) until leaves became yellow, before rising again slightly (Table 111). The early fall in cytokinin concentration was thought to contribute to the leaf senescence because of its precipitous nature, because it preceded senescence, and because concentrations of zeatin fed through the transpiration stream in the senescence TABLE I11 Effect of stage of development on zeatin riboside in xylem sap of de-topped soybean plants. A comparison is made of concentrations in sap and calculated concentrations after adjusting to a common flow rate. Delivery rates from roots are also shown Stage of development Pods 1 cm long Pods fully extended Early-mid pod fill Late pod fill Leaves and pods yellow

Concentration Delivery rate Sap flow" Original" (nmol per concentration at 518 p1 per (k.1per 50 min) (kM) 50 min (FM) 50 min) 518 1925 1162 550

63.04 7.2 11.2 16.0

63.04 26.8 25.1 17.0

32.7 13.9 13.0 8.8

400

20.0

15.4

7.9

"Results from NoodCn et al. (1990b). Sap was obtained by pressurizing detached roots for 50 min at 100 kPa.

PLANT HORMONES AND COMMUNICATION

127

retardation assay (10 nM needed for minimum activity, with strong effects at 4.6 FM) were not dissimilar from those found in sap from de-topped plants ( 1 1 - 6 3 n ~ ,Table 111). The character of this evidence is in line with the requirements outlined in Table I for establishing a physiological role for a hormone. It is supported by other results from plants with pods removed at strategic times to influence both senescence and sap cytokinins. In all, the work of Nooden and colleagues is some of the most precise and quantitative available to support the view that cytokinins from roots can control the timing of natural leaf senescence. Unfortunately, when these results with soybean are re-examined in the light of guidelines for analysing hormones in xylem sap given in Section II.B, they appear less compelling. Problems arise because of dilution effects. If it is assumed that sap from de-topped roots will dilute the hormone in proportion to sap flow, cytokinin concentrations are more fairly compared after re-calculating for the same sap flow rate. In the experiments of NoodCn e t a f . (1990b) sap flows changed markedly between different stages in plant development when the sap was collected for assay. Thus, concentrations can only be compared validly after such re-calculations. This can readily be done using the original data of NoodCn etaf. (1990b). The outcome shows that the fall in concentration of zeatin riboside at the time pods are extended fully and prior to leaf senescence is less abrupt (by 57.5% rather than 88.5%Table 111) than originally thought and falls only gradually thereafter, to about 24% of the original value, by the time the plants are highly senescent. Calculations for zeatin, dihydrozeatin and dihydrozeatin riboside give a similar picture (data not shown). A further difficulty is that even by taking this approach, the diluting effect of whole-plant transpiration is ignored. According to Nooden et al. (1990b), transpiration averaged over 24 h was approximately 13 times faster than sap flow from de-topped plants used by the cytokinin analyses. Thus, in whole plants, the expected concentration of zeatin riboside in the transpiration stream prior to the time pod extension was complete (i.e. before senescence began) would be 13 times smaller, i.e. of the order of 5nM rather than 6 3 n ~ Five . nanomolar is only half the minimum concentration needed to show any retardation of senescence in the explant assay and almost 1000 times less than that needed to slow down senescence to the level seen on intact plants (Mauk et a f . , 1990). Thus, the likely concentrations of cytokinins in the xylem sap of intact plants are probably insufficient to retard leaf senescence based on tests with detached leaves. Such tests would, in any case, have been sounder if based on reproducing the delivery rates of cytokinins. If delivery rates of zeatin riboside per whole shoot are calculated for the plants used by Nooden et al. (1990b), i.e. concentration x sap flow rate, the amounts delivered from the roots are almost unchanged at the time original sap concentrations fell so dramatically (Table HI). Instead, delivery of zeatin riboside and other cytokinins decreased later during development, and gradually, with delivery

128

M. B. JACKSON

at terminal senescence being about 24% of that from roots of young plants. Overall, these studies still have some way to go to before a convincing case is established for cytokinin regulation of leaf senescence on soybean. In contrast to the above studies, Heindle et al. (1982) have calculated fluxes of major cytokinins during soybean development in the field. In absolute terms the amounts were much smaller than those calculated from Nooden et al. (1990b) given in Table 111. However, they show a general decline with time, especially between full bloom and seed set, i.e. somewhat too early to be obviously implicated in leaf senescence control. Furthermore, the results also show a very small output of cytokinins from the roots of young pre-flowering plants when leaves were young and far from senescent. This observation, and the measurements of cytokinin flux and of responses to applied cytokinins published by Carlson et al. (1987), suggest that high hormone output from the roots of soybean may be more closely connected with successful seed set than with retarding senescence. Removing pods when fully expanded delays leaf senescence in soybean plants and also increases the levels of some cytokinins in xylem sap no matter how they are calculated (NoodCn et al., 1990b). Others have also reported that changes in reproductive development can affect sap cytokinins. For example, Beever and Woolhouse (1973) found delivery rates of cytokininlike activity to increase many-fold following flower induction by short days in Perilla frutescens, even though root growth was inhibited. Later, cytokinin delivery fell to a very low level when the leaves became senescent. In contrast, a very striking 75-90% decrease in the delivery rate of bioassayed cytokinin from roots of Xanthium stromarium was recorded by Henson and Wareing (1976) after just one flowering-inducingshort day. However, a lack of attempts to probe the physiological significance of these various and varied findings makes it difficult to make confident conclusions concerning physiological significance.The target requirements are measurements of (1) hormone delivery rates from roots to shoots, (2) delivery rates per gram of source roots to a unit weight or area of recipient shoot tissue, and (3) tests of physiological activity based on supplying the appropriate cytokinins to leaves at realistic delivery rates per unit leaf area. Recent work with sugar cane (Saccharum spp. hybrid) points the way forward (Meinzer etal., 1991). Delivery rates of zeatin riboside per unit leaf area from roots of unstated size decayed in an exponential manner as the plants grew larger and stomata closed. The closing of stomata is an early indicator of leaf senescence. This approach is clearly a marked improvement on all that has gone before.

C . ROOT EXCISION STUDIES

The assumption underlying such work has been that cytokinin output is proportional to the bulk of the root system. Thus, is should prove possible to

PLANT HORMONES AND COMMUNICATION

129

link decreased cytokinin output with any changes in shoot behaviour caused by pruning or volume restriction. There are several substantial studies linking the formation of new roots on de-rooted leaf or shoot cuttings with long-term survival and increases in exudation of cytokinin-like activity. Using a senescence-retarding assay, Wheeler (1971) found increases in cytokinin activity in the water in which rooting of bean leaves was taking place at about the time new roots were forming. Similarly, Engelbrecht (1972), using the tobacco callus assay, showed increases in cytokinin activity in petioles 4 days after detaching bean leaves from the plant. At this time, new roots had formed but had not emerged from the petiole. Wang et al. (1977) confirmed by GC-MS that root formation in excised bean leaves does coincide with increased cytokinin content of the leaf, notably of zeatin glucoside. Bollmark et al. (1985) also reported only small amounts of zeatinand isopentenyl adenine-like cytokinins in detached leaves until adventitious roots emerged. Because applying dilute benzyladenine (10-8-10-9 M) inhibited the formation of adventitious roots, Bollmark et al. (1985) suggested that endogenous cytokinins from the new roots may explain why further root initiation ceases once the first set of roots has formed. These four pieces of experimental evidence show that the presence of roots is conducive to high levels of extractable cycokinin. However, convincing measurements of delivery rates of cytokinins from roots to shoots in xylem sap following removal and regeneration of roots are notably absent from the literature. There has been much testing of the possibility that applying cytokinins to shoots can replace the effect of missing roots. The earliest example of such an approach is that of Kulaeva (1962). Although her results are mostly anecdotal, the experimental design is a model of conceptual clarity and completeness. Excised leaves of Nicotiana rustica placed in water senesced rapidly, while steam girdling of leaves on whole plants to prevent outward phloem transport while retaining xylem links between the leaf and the roots, did not promote senescence. Thus, the promoting effects of leaf excision were related to severing inputs from the roots (negative message) rather than the result of any accumulation of senescence promoter normally exported in phloem (accumulative message) and from which excised leaves might suffer. Thus, xylem sap appeared to carry a senescence inhibitor. This inhibitor was found, by Kulaeva, to possess the properties of a cytokinin since the senescence-delaying influence of retaining xylem connections between roots and leaf could be reproduced either with kinetin or with applications of crude xylem sap. Support for a senescence-inhibiting and cytokinin-like role for xylem sap was provided by Mothes and Engelbrecht (1963), who found that rapid chlorosis, caused by shading the distal portion of detached tobacco leaves, or by heating them to approximately 49°C for 3 min, was prevented if the leaf was first allowed to root at the petiole, o r was treated with kinetin. The absence of roots also interferes with flower

130

M. B. JACKSON

development, e.g. in stored cuttings of Vitis vinifera, where the absence of roots has been linked with the failure of flower buds to develop. This failure to form flowers can be prevented either by warming the medium to encourage early adventitious rooting, or by supplying kinetin (Mullins, 1967). A somewhat complicated example of the possible impact of cytokinins from roots was provided by Treharne et al. (1970), who showed that a halving of the root system in Phaseolus vulgaris strongly diminished the stimulation in photosynthesis per unit leaf area that could be induced under a wide range of light intensities by partial defoliation; applying 20 mg 1-' kinetin fully compensated for the missing roots. Similarly, indoleacetic acid (IAA) and gibberellic acid treatments induce aerial stolons of Solanurn andigena on underground shoots, but only if roots are present. However, applying the cytokinin benzyladenine can substitute for the missing roots in this system (Wooley and Wareing, 1972). More straightforwardly, Richards and Rowe (1977) found that the inhibiting effect on leaf expansion and leaf production or lateral shoot production caused by removing 50% of the roots from peach seedlings (Prunus persica) could be more than compensated for by applying benzyladenine. Several other papers have reported that inhibition of shoot growth o r photosynthesis caused by removing a sizeable proportion of the roots can be overcome with exogenous cytokinin treatments (Carmi and Koller, 1978; Carmi and Heuer, 1981). Unfortunately, the persuasiveness of this sort of evidence is seriously eroded by observations that plants not suffering a loss of roots also respond strongly and positively to applications of cytokinins (e.g. Richards and Rowe, 1977; Jackson and Campbell, 1979; Carmi, 1986). It is argued by Carmi (1986) that the exogenous synthetic cytokinin is active because it compensates for a fall in endogenous cytokinin output believed to result from the smaller root system of plants given exogenous cytokinin! Carmi (1986) has also made similar deductions from measurements of photosynthesis, which increased in plants of Phaseolus vulgaris with whole roots systems that were given benzyladenine. However, the explanation that it occurs because the applied cytokinins reduced output of root cytokinins as a result of inhibition of root growth has not been demonstrated by direct measurement. The explanation is particularly difficult to accept in Carmi's experiments since the shoot systems of his plants possess an artificially contrived excess of roots, and thus of putative cytokinin input, which was achieved by removing all shoot tissue except for the primary leaves. Apply benzyladenine to these plants would be unlikely to return the root :shoot ratio to that of ordinary unpruned plants, and yet growth promotion was obtained. A more likely explanation for the positive responses of shoot systems to exogenous cytokinins is simply that responsiveness to cytokinins is an intrinsic property of shoots and has little to do with how much natural cytokinin is supplied by roots. Therefore, in species where plants with an initially full complement of roots can respond positively to exogenous cytokinin, responsiveness following root excision is

PLANT HORMONES AND COMMUNICATION

131

a poor test of the physiological importance of any associated decrease in root cytokinins. That cytokinin output from roots may be of little consequence is suggested by the observation in oak seedlings (Quercus rubra) that removing root tips, the most likely sites of intensive cytokinin biosynthesis, increased rather than decreased cytokinin in xylem sap (Carlson and Larson, 1977), and that similar treatment to Lolium perenne stimulated rather than slowed shoot growth (James and Hutto, 1972). In the experiments of Carmi and van Staden (1983), a large decrease in cytokinin levels in leaves and petioles measured 8 days after removing two-thirds of the roots of decapitated bean plants was not accompanied by any statistically significant change in leaf area, weight or soluble protein. Misgivings are compounded by a report that the decline in endogenous zeatin and zeatin riboside which takes place in de-rooted leaves of Helianthus annuus kept in water for 3 days, may be reversed by supplying nitrate-containing nutrient solution directly to leaf tissue, thus by-passing the need for roots (Salama and Wareing, 1979). In similar work, Wang and Wareing (1979) found not only that lateral buds on decapitated shoots of Solanum andigena could grow for up to 30 days in the absence of roots, but also that during this time cytokinin levels in the shoot did not decline and may even have increased during this period of rootless growth. Overall, the promise of early papers has not been fulfilled and de-rooting experiments have not succeeded in demonstrating convincingly that cytokinins from roots explain the positive effects of roots on shoot development, leaf longevity and photosynthesis. In future experiments, transgenic plants that overexpress genes for cytokinin biosynthesis (see Smart et al., 1991) may help, particularly if reciprocal grafts between roots and shoots are made with the wild-type, or expression is regulated through linking the structural cytokinin genes to a root-specific promoter.

D. RESPONSES TO MINERAL NUTRIENT SHORTAGE

A shortage of mineral nutrients in the rooting medium inhibits shoot growth and promotes senescence. Several authors have tried to attribute these responses to decreased cytokinin production by the stressed roots. Kulaeva (1962) showed that leaf senescence, induced in 55-day-old Nicotiana rustica plants by nitrate shortage at the roots, could be overcome with a foliar application of kinetin, but not with a range of other organic compounds including adenine and IAA, or with minerals applied directly to the leaves. These results imply that senescence of older leaves is promoted by a lack of cytokinin production by mineral-starved roots (negative message), although mediation by an accumulation message cannot be ruled out. Subsequent research has repeatedly shown decreased concentrations, o r delivery rates

132

M. B. JACKSON

of bioassayed cytokinin activity in xylem sap, from de-topped roots systems when nutrient deficiency, especially of nitrogen, is imposed for several days (Cucurbita pep-Goring and Mardanov, 1976; Helianthus annuusWagner and Michael, 1971; Salama and Wareing, 1979; Solanurn tuberosurn-Sattelmacher and Marshner, 1978). An exception to these findings has recently been shown in root exudates from de-topped Urtica dioica plants grown in sand given various amounts of nitrate ferilization (Fubeder et al., 1988). In sunflower, much of the cytokinin activity cochromatographs with zeatin and zeatin riboside and has been tentatively identified as such by single ion current GC-MS (Salama and Wareing, 1979). Unfortunately, Salama and Wareing (1979) found that depleted concentrations of cytokinins within the foliage of nitrate-starved plants could be rectified simply by supplying nitrate directly to the leaves, rather than through the roots, by-passing the need for root cytokinins as second messengers. Thus, diminished supplies of root cytokinins to the shoots of mineral-starved plants may not contribute significantly to slow shoot growth, enhanced senescence and decreased foliar cytokinin levels (Horgan and Wareing, 1980; Darrall and Wareing, 1981), even where external applications of cytokinins have given some symptom relief (e.g. Horgan and Wareing, 1980). Clearly, detailed time-course studies over hours rather than days are needed in this kind of work. In any future experiments it would be important to establish when nutrient shortage at the roots reduces cytokinin flux in the xylem sap in relation to the time when (1) the physiology and cytokinin levels change in the leaves and (2) when mineral levels decrease in the foliage. Kuiper and Staal(l987) and Kuiper et al. (1989) have gone some way towards producing such information by showing, in Plantago major, that diluting the mineral supply to the roots reduces both shoot growth and extractable zeatin and zeatin riboside levels in roots and shoots, within a day and before mineral concentrations changed in the shoot as a whole. They also showed that shoot growth could be restored almost to normal for several days by applying benzyladenine, which presumably compensated for a low cytokinin output from the mineral-starved roots. When Plantago plants were put back into mineral-sufficient solutions, mineral concentrations returned to normal in the shoots several days before growth rates and endogenous cytokinin concentrations increased (Kuiper et al., 1989), implying that shoot growth was not limited by mineral levelsperse. Analyses of cytokinin and minerals in growing parts rather than in the whole shoot, and tests on the effect of benzyladenine on nutrient levels and on the behaviour of unstressed plants, would have bolstered this interpretation. These results obtained by withholding mineral nutrients have given support to the notion that cytokinin export from roots is depressed by the stress. A shortage of nitrogen seems to be more effective than a lack of phosphorus (Coleman et al. , 1990). Whether this decrease in cytokinins is of much importance in maintaining normal shoot growth and leaf longevity is

PLANT HORMONES AND COMMUNICATION

133

less certain. Much of the available information remains rudimentary and benefits little from the opportunities created by modern analytical methods. Salutary papers by Radin and Eidenbock (1984) and Salim (1991) remind us of the importance of water in the root :shoot relationships of mineralstarved plants. In Gossypium hirsutum, phosphorus deficiency increased the hydraulic conductivity of the leaves to water intake. This in turn decreased daytime leaf water potentials that may have limited leaf expansion (Radin and Eidenbock, 1984). In all work that attempts to ascribe roles to cytokinins in stressed plants, more account should probably be taken of possible hydraulic influences.

E. EFFECTS OF OTHER STRESSES APPLIED TO ROOTS

When roots are subjected to drought, flooding, heat, cold, salinity, parasitism, etc., shoot growth slows, stomata may close and leaf senescence is promoted. Experimental approaches similar to those already discussed for mineral shortage and root excision or constriction have been used extensively to implicate diminished cytokinin supply from roots. Regulation of stomatal closure is of particular concern here because drought stress is often linked to stomatal closure. Exogenous cytokinins have been known to promote opening or inhibit closure of stomata in some but not all plants (Livne and Vaadia, 1965; Meidner, 1967; Luke and Freeman, 1968). Early results also indicated that exogenous application of cytokinin could ameliorate leaf senescence promoted by drought stress (Shah and Loomis, 1965).

1. Drought, salinity and biotic stresses Itai and Vaadia (1965) were perhaps the first to publish correlations between droughting and decreased cytokinin in xylem sap from de-topped plants. However, this first paper contained only very preliminary information, and described soybean bioassays of bleeding sap from sunflower, collected over 72 h, with little or no replication. Later papers substantiated the finding and extended it to salinity stress (Jtai et al., 1968) and heat stress (Itai et al., 1973). The measurements were of concentrations in xylem sap with no sap flow rates given to allow calculation of cytokinin delivery. However, it is probably safe to assume that, because sap flow from de-topped plants would have been slowed by drought, cytokinin delivery rates would have been even more strongly reduced than the 70% loss of concentration reported. Incoll and Jewer (1987) cited ten reports of water shortage or salt stress decreasing cytokinin concentrations in leaves or roots, but Itai and Vaadia (1971) was their most recent citing of analyses of xylem sap from droughted plants. However, the latter paper places considerable doubt on the necessary involvement of root cytokinins in drought-induced decreases in these

134

M. B. JACKSON

hormones in xylem sap or in leaves. Itai and Vaadia (1971) showed that using a fan for only 30 min to wilt leaves on intact, well-watered plants reduced cytokinin concentrations in leaves and in bleeding xylem sap, in the absence of stress at the roots. Furthermore, droughting detached leaves had a similar effect and cytokinin levels recovered to 73% of their original concentration when the detached leaves were re-hydrated. The need for stress at the roots to mediate these effects was therefore obviated. Experiments by Aharoni et a f . (1977) and Ong (1978) also lead to the conclusion that cytokinin levels in leaves can change in response to water shortage independently of roots. A comparison of endogenous zeatin, zeatin riboside, isopentenyladenine and isopentenyladenosine concentrations in the roots and shoots of salt-resistant and salt-tolerant lines of barley (Kuiper et al., 1990) throw additional doubt on the notion that decreased cytokinin levels explain damage symptoms caused by root stress. In this work large decreases in cytokinins were restricted to the salt-resistant barley plants rather than salt-intolerant ones (Kuiper et al., 1990). One of the consequences of the diminished cytokinin content in leaves of droughted or osmotically stressed plants could be the promotion of stomatal closure. When leaves wilt, cytokinins may decrease (Itai and Vaadia, 1971), and this might enhance stomatal closure along with increases in abscisic acid (see Section VI1.B). However, such a mechanism cannot help explain how stomata close with little or no attendant decrease in leaf hydration (Bates and Hall, 1981). Here, transmission of the influence of drying roots to shoots is more likely to involve a hormonal message. The idea that this may be a negative message in the form of decreased export of cytokinins from roots received support from the split root experiments of Blackman and Davies (1985) and related work (Davies et al., 1986). These authors found that stomatal closure could be induced in the absence of shoot water deficits by drying only one of the two half-root systems. But, when Incoll et al. (1990) imposed drought stress that was gentle enough to induce some closing of stomata within 3-6 days, without any attendant fall in water potentials of leaves of Phaseolus vulgaris, there was no associated decrease in the concentration of zeatin, zeatin riboside or six other cytokinins in the xylem sap. The sap was obtained by pressurizing the roots of de-topped plants for 5 min and assaying by radioimmunoassay after HPLC purification (Ray, 1989). In other work, osmotic stress applied to roots failed to bring about a large decrease in the concentration of zeatin riboside in roots of a droughtintolerant tomato species, while a decrease in this cytokinin was seen in stressed roots of the most tolerant species (Solanurn pennelli) (Pillay and Beyl, 1990). Thus, a further uncertainty is added to the idea that droughting injury is associated causally with decreased cytokinin delivery by roots. There is some evidence that biotic stresses on roots results in decreased cytokinin levels in xylem sap. Drennan and El Hiweris (1979) reported a 90-95% decrease in the concentration of cytokinin activity in bleeding sap,

PLANT HORMONES AND COMMUNICATION

135

using a tobacco pith callus assay, following infection of Sorghum vulgare with the parasitic root-infecting weed Striga hermonthica. Since the decrease would almost certainly have been accompanied by reduced sap flow rates and whole-plant transpiration rates, an extremely large decrease in cytokinin flux is indicated. The sap was partially purified by thin-layer chromatography prior to assay. Xylem sap from cultivars of Sorghum with greater tolerance to Striga contained the most cytokinin. Drennan and El Hiweris (1979) were able to show an increase in growth in dry weight and leaf area of parasitized Sorghum by applying benzyladenine at weekly intervals to the soil to compensate for the small cytokinin supply from the roots. Infection of the roots of Eucalyptus marginata by Phytophthora cinnamomi, a possible cause of Eucalyptus die-back, was shown by Cahill et al. (1986) to reduce concentrations of zeatin and isopentenyl adenine within 3 days in sap obtained from de-topped root systems pressurized to 10 kPa. After 14 days infection, concentrations in a susceptible type were 26% of those of controls, while in a field-resistant strain of Eucalyptus marginata, no decrease in cytokinin levels were seen. Root infection of Douglas fir (Pseudotsuga menziesii) by at least one type of ectomycorrhizal fungus (Thelephora) has been shown to reduce the flux of zeatin riboside in xylem sap (Coleman et al., 1990) from pressurized, de-topped root systems. However, on a root weight basis, no difference in cytokinin output was seen.

2. Soil waterlogging Waterlogging asphyxiates roots, thereby preventing growth, cell division and the aerobic respiration on which most root metabolism depends. This in turn influences shoot development, leading to a wide range of morphogenetic responses, such as epinastic curvature of petioles, hypertrophic stem swelling, adventitious rooting, aerenchyma formation, stomata1 closure, slow shoot extension and premature leaf senescence (Jackson and Drew, 1984; Jackson, 1990). Tomato and sunflower have been studied most by those wishing to understand how oxygen deprivation in roots leads to rapid developmental changes in the shoots. A diminution in cytokinin supply from roots to shoots caused by soil waterlogging was first shown by Carr and Reid (1969), who found smaller concentrations in bleeding sap of Helianthus annuus plants decapitated at intervals during 4 days of flooding. The decrease was rather small over the first 3 days (11-29%), with a much larger decrease on the fourth day of flooding (62%) in association with marked root death. Sap flow from decapitated plants also declined during flooding; the published rates allow cytokinin delivery to be calculated. These show that delivery declined by 22-25% during the first 2 days of waterlogging (by which time shoot growth had stopped). After 3 days, flooding delivery was 37% below normal (when lower leaves were starting to yellow). After 4 days (when lower leaves were clearly yellowing) cytokinin output to the shoots was 94% that of

136

M. B. JACKSON

well-drained plants. Burrows and Carr (1969) showed that applying kinetin to young leaves taken from plants flooded for 96 h and kept in the dark delayed their senescence. Although this result suggests that leaves were suffering from a shortage of endogenous cytokinins supplied by roots, other results argue against this. For example, removing leaves from non-flooded plants failed to promote their rapid senescence, even though the leaves were obviously separated from a source of root cytokinins (Burrows and Carr, 1969). More recently, Neuman etal. (1990) and Smit etal. (1990), using immunoassays and a pressure of 0.3 MPa to generate sap flows from de-topped plants that are likely to have approached those of whole plant transpiration, confirmed that decreases in cytokinin flux result from oxygen deficiency. Zeatin riboside flux from roots of Phaseolus vulgaris and a hybrid poplar (Populus trichocarpa x P. deltoides) exposed to severe oxygen deprivation for 3,24 or 48 h decreased by at least two-thirds (Smit et al., 1990 and Fig. 6) in association with loss of root tip activity assessed in terms of nuclear

. .t

Sao concentration

Delivery rate

Leaf concentration

8ol , I-

-

-E0

0.6

-&

0)

0)

0

f

E

0.4

8

$

E

~

3c

c

I N 0.2

0

40

N

3

48

0

3

24

4

4a

Time (h)

Time (h)

Time (h)

(a)

(b)

(4

Fig. 6. Effect of oxygen shortage at the roots of Poplar on (a) concentration of zeatin riboside in xylem sap, (b) delivery rate from the roots in xylem sap, and (c) concentration in growing leaves (means and standard errors of 3-6 replicates). Oxygen shortage was imposed by replacing the air supply to nutrient solution with nitrogen gas for up to 48 h. Xylem sap (5 ml) was obtained from de-topped plants by pressurizing roots to 0.3 MPa, equivalent to leaf water potential of plants of approximately -0.3MPa. The experiment shows a marked decrease in cytokinin delivery and xylem sap concentration after 3 h or more of oxygen deprivation but no associated change in leaf cytokinin levels. Redrawn from Smit el al. (1990).

PLANT HORMONES AND COMMUNICATION

137

divisions. However, this decrease in delivery did not change the concentration of zeatin riboside activity in the leaves (see also Smit et a f . , 1990). Furthermore, dihydrozeatin levels in the leaves of Phaseofus vufgaris were doubled rather than depressed by 24 h of root hypoxia, even though sap concentrations were reduced by two-thirds (delivery rates were not stated). This evidence does little to strengthen the case for the involvement of cytokinins from roots in the stomata1 closure and slower leaf growth brought about by poor root aeration. The inability of leaf discs from Phaseolus plants with hypoxic roots to expand or re-open their stomata in response to exogenous zeatin riboside, while those from unstressed plants were able to give strong positive responses (Neuman et af., 1990), confirmed that other more active factors (i.e. positive messages) present in xylem are probably more influential than a decrease in root cytokinin. Other lines of research also suggest that decreases in cytokinin supply from roots are unlikely to be key elements in bringing about morphological changes in the shoot system. Drew et af. (1979) were unable to reproduce the senescence-promoting effects of root anaerobiosis in Hordeurn vulgare by removing root tips, the putative sources of root cytokinins. In these plants, inorganic nitrogen supply appeared to be more important than cytokinins, since senescence was more successfully reversed by nitrogen fertilization via a small number of aerated roots than by applying cytokinins. Cytokinins from roots also look unlikely regulators of epinastic curvatures that develop in waterlogged Lycopersicon esculenturn. Although applications of benzyladenine at rather high concentrations (10-15 mg I-l) can largely prevent the response, especially if gibberellins are also included (Selman and Sandanam, 1972; Railton and Reid, 1973; Jackson and Campbell, 1979), removing roots, and thus the source of root cytokinins, does not reproduce the effect of soil flooding (Jackson and Campbell, 1975b, 1976a). Furthermore, the presence of a second well-aerated root system growing above the flooded roots in damp peat, that is a potentially rich source of endogenous cytokinins, failed to inhibit epinasty (Jackson and Campbell, 1979). The opposite effect would have been expected if the upper root system was making good an endogenous hormone deficiency. In contrast to the results with epinastic curvatures, the slow rate of stem extension by flooded plants was successfully overcome by the presence of well-aerated upper roots. This effect could conceivably have been mediated by an output of root cytokinins since foliar sprays containing benzyladenine have a similar effect (Railton and Reid, 1973; Jackson and Campbell, 1979). However, since nonwaterlogged plants were even more responsive to hormone application than flooded ones, a safer conclusion is that flooding merely reduced the vigour and capacity of the shoots to elongate in response to cytokinin supply. Stomata1 closure in waterlogged plants is of particular interest since it can occur within 24 h, without the mediation of marked or prolonged water deficits in the leaves (Jackson et al., 1978; Bradford, 1982). Bradford (1983)

138

M. B. JACKSON

and Zhang and Davies (1986) believe that a decrease in cytokinins is involved because applications of benzyladenine can open the stomata of flooded plants of Pisum sativum and tomato (Jackson and Campbell, 1979). However, Neuman et al. (1990) found that zeatin riboside and three other cytokinins over a wide range of concentrations were unable to re-open stomata of leaf discs taken from Phaseolus or Populus plants after 24 h root anoxia. This indicates that stomata were probably not closed by cytokinin shortage. Cytokinin treatments can overcome an inhibitory effect of soil flooding on the net rate of carbon dioxide fixation that is independent of stomata1 apertures. The effect is seen at various internal concentrations of carbon dioxide (Bradford, 1983). Maintenance of ribulose bisphosphate activity could explain the result since it has been known for many years that activity of this key photosynthetic enzyme can be increased with exogenously applied cytokinins (Treharne et al., 1970). This subject merits further work. Overall, there seems to be little doubt that oxygen-deficient roots export much less cytokinin to the shoots than do well-aerated ones. The evidence for believing this has a significant impact on shoot development remains unconvincing. F. CONCLUSIONS

The literature on cytokinins in xylem sap and its possible importance for shoot development is much larger than that for other hormones. It seeks to establish that cytokinin output from roots is a requirement for healthy shoot development and that decreases in output caused by root stresses (negative message) are detrimental to the shoot system. There are serious shortcomings in the way most of the measurements have been made. Bioassays rather than physicochemical methods or immunoassays have usually been used, and scant attention has been given to the effects of changes in sap flow rates on concentration. However, there seems little doubt that cytokinin levels in xylem sap are subject to large changes during normal development and that they alter considerably as a result of changes to the environmental conditions around the roots, and as a result of events in shoot development such as flowering. However, the physiological importance of these changes in cytokinins moving to the shoot in xylem sap has yet to be established, despite much experimentation.

VI. GIBBERELLINS A. INTRODUCTION

Gibberellins (GAS), in common with other hormones, elicit a wide range of responses when administered to plants, although considerable specificity of response is inherent in different species and tissues at various stages of

PLANT HORMONES AND COMMUNICATION

139

development. Evidence of the kind advocated in Table I for hormonal roles is restricted primarily to stem or leaf extension and to germination of some cereal grains. There is also evidence that GAS are important in delaying leaf senescence, especially in concert with cytokinins (NoodCn, 1986). The gibberellins, defined by chemical structure rather than biological activity (cf. cytokinins), are diterpenoid acids and carry either 19 o r 20 carbon atoms associated with an ent-gibberellane four-ring structure (Sponsel, 1987). Approximately 60 different GAS have been found in higher plants, at least 16 of which are present in the seeds of Phaseolus vulgaris. It is fortunate for those wishing to study their physiology that only a small number of GAS possess biological activity p e r se, although these active forms are related to many others which occur earlier or later in the biosynthetic pathway. The pathway commences with mevalonic acid, which gives rise to many intermediates that include geranylgeranylpyrophosphate (GGPP), the C20 precursor for all plant diterpenes. GGPP is cyclized by ent-kaurene synthetase and a subsequent series of hydroxylations and oxidations to generate the first gibberellin (GA12aldehyde) from which all others are derived. These include biologically active G A I , GA3, GA4 and GA,. G A I is likely to be the primary active G A in vegetative tissue of many species. A hormonal role for endogenous gibberellins has been established with greatest certainty in stem extension. The evidence is firmly based on a thorough knowledge of the pathways of gibberellin biosynthesis, gained through the application of analytical chemistry and manipulation of G A levels using either dwarf mutants, with defects at certain steps in the biosynthesis of GAS (MacMillan, 1987), or specific inhibitors of gibberellin biosynthesis, most notably paclobutrazol (Lenton et al., 1987). Evidence implicating gibberellins from roots in shoot development is similar to that already outlined for the cytokinins. It suffers from many of the same deficiencies identified in the cytokinin work and has no new approaches to offer except for the use of gibberellin-deficient mutants in reciprocal grafting experiments between mutant and wild-type stocks and scions. Although the literature is too small to warrant the extensive subdivisions used to assess root cytokinins, it will be reviewed along similar lines. B. STUDIES ON UNSTRESSED PLANTS

The identification of G A activity in roots and in the media used to culture excised roots of Lycopersicon esculentum for over five years, and over many sub-cultured generations, provides some of the earliest evidence that roots are capable of producing GAS independently of the shoot (Butcher, 1963). Twenty-five years later, these GAS were identified by GC-MS as G A I and GA3 (Butcher et al., 1988). Concentrations of GAS in intact plants have been reported to be greater in roots than shoots (Michniewicz and Kriesel,

140

M. B. JACKSON

1972; Faull et al., 1974), but the opposite has also been found (Kaufman et al., 1976). Some of the root gibberellins may find their way into the transpiration stream or spring sap flow (Dathe et al., 1982) and, thus, up to the shoots (Sitton et al., 1967b). Indeed, there may be considerable cycling of GAS between roots and shoots, with complex interconversions between active and inactive GAS taking place (Hoad and Bowen, 1968; Crozier and Reid, 1971). In the earliest work, G A activity was detected in unpurified bleeding sap and in paper chromatographed sap from de-topped Pisum sativum, Lupinus albus and Impatiens glandulifera plants using four different bioassays (Carr et al., 1964). The authors were careful to test sap at a range of dilutions and found growth-inhibitory activity in some undiluted samples and gibberellin activity in all samples at various sap concentrations. Rough calculations of gibberellin flux from roots to shoots of 10-30 pg per plant per day were judged sufficient to stimulate shoot extension when calibrated against responses of the shoot to GA3. Using the dwarf-pea epicotyl elongation bioassay and the barley endosperm test, Phillips and Jones (1964) also detected GA-like activity in extracts of 500 ml samples of bleeding sap from Helianthus annuus and found that the concentration decreased with time up to 24 h after removing the shoot. The danger inherent in bioassays of unpurified sap is illustrated in the analyses of Skene (1967). These showed that GA-like activity in the xylem sap of Vitis vinifera, measured by the barley endosperm test, was inhibitory to any GAS also present. The inhibitor was thought to be abscisin I1 (now abscisic acid, see Section VIII). This paper also illustrates the commonly held, but erroneous, notion that the concentration of GAS measured in slowly flowing bleeding sap of de-topped plants remains undiminished by the much faster flow of whole-plant transpiration. In assuming this, Skene (1967) over-estimated both the likely flux of hormone into the shoot and, thus, its probable impact on shoot growth. Although the presence of gibberellins in xylem sap has been shown many tmies (grape vine-Skene, 1967; sunflower-Sitton et al., 1967; tomatoSembdner et al., 1968; Douglas fir-Lavender et al., 1973; walnut-Dathe et al., 1982), assumptions concerning its physiological significance must be tempered by the strong likelihood that younger shoot parts, especially growing zones and developing seeds, may be self-sufficient in GAS. Jones and Phillips (1966) demonstrated that while roots (apical 3 4 mm only) were capable of producing GAS, young leaves and the apical buds of sunflower were also able to produce GAS, as assessed by the levels of GA-like activity in agar on which excised plant parts were placed. Indeed, almost all subsequent studies of G A biosynthetic pathways have been made on tissues other than roots (Phinney et al., 1986; MacMillan, 1987). In peas, GA1 is the most active endogenous G A in promoting stem extension. Mutants in which the conversion of GAzo to G A I by a 3p-hydroxylase enzyme is blocked (the le mutant) are dwarf (Ingram et al., 1984) unless given exogenous GA1. The

141

PLANT HORMONES AND COMMUNICATION

3P-hydroxylase gene appears to be expressed mainly in shoot tips rather than in roots since grafting an le mutant shoot on to a non-mutant stock comprising roots and mature shoot tissue does not promote stem extension (Lockard and Grunwald, 1970; McComb and McComb, 1970; Reid et al., 1983). In pea, at least, this seems to rule out the ability of the normal root system to supply sufficient G A I , or its precursors, to influence stem extension. A different result has been obtained with dwarf tomato mutants, generated by ethylmethanesulphonate mutagenesis, which are defective at steps in the G A biosynthetic pathway either before or after ent-kaurene. Grafting onto wild-type stocks comprising roots and some stem reversed the dwarf growth habit of the scion (Zeevaart, 1983). This suggests that the root

1

I

1.5

I

2.0

I

1

I

2.5

I

3.0

Concentration of GA, (log,, pg g-’ fresh wt)

Fig. 7. Relationship between final length of first leaf of seedlings of wheat (Triricum aesfivurn, cv. Maris Huntsman) and different levels of endogenous gibberellin A l in the expansion zone obtained by supplying different amounts of the gibberellin biosynthesis inhibitor paclobutrazol. The linear response to loglo changes in endogenous GA1 shows that leaf extension would be much more responsive to small increases in GAI when existing endogenous GA1 concentrations are small. Endogenous levels of GAI would normally be similar to the highest concentration shown on this diagram, suggesting that shoot extension is normally well-buffered against modest changes in GA concentrations. From Lenton ef al. (1987).

142

M. B. JACKSON

system can supply the shoots with physiologically significant amounts of GA. However, if shoots of non-mutant plants are dependent upon GAS from roots, non-mutant shoots should be dwarfed by grafting on to mutant roots; however, such dwarfing does not occur (Zeevaart, 1983). The explanation for this seemingly contradictory result is probably that wild-type shoots are self-sufficient in GAS. The GA-deficient shoots are, however, much more responsive t o very small increases in GA. In turn this is related to the very small amounts of endogenous G A in the mutant shoots. Using different amounts of paclobutrazol to manipulate endogenous G A levels in Triticum aestivum, Lenton et al. (1987) showed that the smaller the internal concentration of endogenous G A , the more sensitive the shoot is to G A supply (Fig. 7). This implies that the growth of shoots containing a normal content of GAS (e.g. 5-6ngg-' fresh wt) are strongly buffered against changes in G A levels of the magnitude that root systems can be expected to bring about. These findings also place in doubt claims that developmental events in shoot apices, such as monocarpic senescence in peas, or the switch from juvenile to adult leaf form in Hedera helix, are causally related to associated decreases in the flux of root GAS (Frydman and Wareing 1973; Proebsting et al., 1978).

C . EFFECTS OF ROOT EXCISION AND ENVIRONMENTAL STRESSES APPLIED TO ROOTS

The balance of evidence presented in Section V1.B favours the view that the supply of GAS from the roots of healthy plants is not very important for shoot growth. If, indeed, this is the case, it must follow that effects on shoot elongation ,and possibly other processes brought about by stresses imposed on roots, cannot be consequences of a negative G A message. This could explain why Horgan and Wareing (1980) were unable to overcome the inhibiting effects of nitrogen or phosphorus starvation on shoot growth by administering GA3 to the shoots of Betula pendula or Acerpseudoplatanus, and why applications of GA1 or GA3 fail to reverse the inhibiting effects of de-rooting on shoot extension in juvenile Hedera helix (Frydman and Wareing, 1973). It is also clear that stress, such as water shortage, can depress endogenous G A levels when applied to detached leaves, thus bypassing the roots (Aharoni et al., 1977). Nevertheless, some papers have sought to implicate reduced output of root GAS in shoot responses to perturbations to the root system. For example, Holm and Key (1969) and Crozier and Reid (1971) reported that inhibition of shoot extension in seedlings of Glycine max or Phaseolus coccineus, caused by root excision, could be partially overcome by applying GA3 (lo-' M), although in the former case the hormone was only active when administered in the presence of a cytokinin. Soil flooding has also been thought to depress shoot growth

PLANT HORMONES AND COMMUNICATION

143

by interfering with GA production by roots. In support of this, concentrations of G A (GA3 equivalents) in partially purified xylem sap of de-topped tomato plants has been found to be substantially decreased by 1,2 or 3 days of soil waterlogging in association with a less dramatic decline in the concentrations in the roots and shoots and a slowing of stem extension (Reid et af., 1969; Reid and Crozier, 1971). However, although applying GA3 to shoots stimulated stem extension in waterlogged plants, well-drained plants were equally responsive during the first 2 days, suggesting they were not more GA-deficient than their non-flooded counterparts. There are several other disconcerting mismatches in the results of Reid and Crozier (1971) that detract from the notion that depressed G A levels in xylem sap are responsible for the slower shoot growth of waterlogged plants. For example, G A levels in leaves of waterlogged plants returned almost to control levels after 3 days flooding although xylem sap G A remained very low. Also, the responsiveness of stem extension to exogenous CAI was similar in control and flooded plants for 3 days, even though G A levels in sap and foliage and stem extension rates were all much smaller in flooded plants for most of this time. Selman and Sandanam (1972) and Jackson and Campbell (1979) found that stem extension-flooded tomato plants were actually less responsive to hormone treatments containing GA3, suggesting that the constraint on shoot extension may be on G A responsiveness rather than on G A levels. In Reid and Crozier’s work (Reid and Crozier, 1971) there was also a large difference between the amount of GA3 needed to be applied to the plants exogenously (1 pg per plant) to promote stem extension and the amounts of G A in shoots of untreated non-flooded plants (7.5 ng GA3 equivalents per plant). In more recent work with rooted Popufus cuttings (Neuman et al., 1990), 24 h in oxygen-deficient nutrient solutions inhibited cell division in roots and reduce stem extension by about 25% in association with 22-36% less C A I and GA3 in the shoot tip, when analysed by GC-MS. Neuman et af. (1990) considered these changes to be physiologically insignificant. On balance, evidence for implicating a lack of root GAS in the slow rates of shoot elongation in flooded plants is not very persuasive.

D. CONCLUSIONS

There can be little doubt that roots are able to make GASindependently of the shoot. But it is also apparent that shoot tissues, or at least young growing parts and also seeds, are vigorous producers of GAS, and that 3phydroxylase activity required to convert GA2o into physiologically active G A l may be largely restricted to the shoots. Shoots may, therefore, be self-sufficient in active GAS, although the position regarding older, nongrowing parts remains to be clarified. Grafting experiments using GAdeficient mutants also indicate that shoots are independent of root GAS, or

144

M. B. JACKSON

their precursors, at least for the regulation of shoot elongation. Physiological studies of stressed plants indicate that the output of GAS by roots is suppressed by stress. Experimental support for the belief that this contributes in any significant way to slowing shoot growth under these conditions is poor at present.

VII. ETHYLENE A. INTRODUCTION

Ethylene is the only known gaseous plant hormone. It is produced by all plants, and, in common with the other hormones, has a wide spectrum of effects on development, with specificity of response being determined by the species or by the position, age or physiological state of the target cells. Evidence that satisfies the requirements given in Table I for hormonal action is strong for leaf, flower and fruit abscission, fruit ripening, accelerated underwater extension by many aquatic and semi-aquatic plants, swelling by roots and shoots in response to small mechanical pressures, development of intercellular gas space (aerenchyma) in poorly aerated roots, and epinastic leaf curvatures in flooded plants (Beyer etal., 1984; Jackson, 1985a,c, 1987). In most higher plants, ethylene is derived from methionine, which is regenerated in a four-stage sulphur-conserving cycle driven by ATP and a transamination step. One component of the cycle, S-adenosylmethionine (SAM), is converted to 5’-methylthioadenosine and l-aminocyclopropane1-carboxylic acid (ACC) by a rate-limiting step catalysed by the enzyme ACC synthase. ACC is the immediate precursor of ethylene and is oxidized to the gas by the highly expressed “ethylene-forming enzyme” (EFE), also known as ACC oxidase. The availability of substances that inhibit one or other of these last two steps in ethylene biosynthesis has allowed much testing of the involvement of ethylene in developmental processes. The most widely used inhibitors are aminoethoxyvinylglycine (AVG), which interferes with ACC synthase, and cobalt ions or anaerobiosis, which can block the oxidation of ACC. Research with ethylene also benefits from the availability of inhibitors that interfere with the action of ethylene, most notably silver nitrate and norbornadiene. The simple gas chromatographic equipment needed to measure ethylene with a sensitivity similar to that of the plant itself (<0.01 pl1-l) is a further aid to precise and replicated assays. Extraction of the gas is straightforward and is usually accomplished by applying partial vacuum, or by relying on simple outward diffusion. However, such methods have some shortcomings. For example, production is most often estimated using excised segments enclosed in small sealed containers to accumulate the gas. This can cause artifacts associated with extra ethylene formed in response to wounding (Jackson and Campbell,

PLANT HORMONES AND COMMUNICATION

145

1976b; Morgan et al., 1990). One way to avoid this is to make measurements during the 20-30 min lag period that often precedes the onset of the wound reaction (Jackson et al., 1978). Vacuum extraction can also be problematic (author’s own observations) because of the artificially high values generated by ethylene produced by tissues and efficiently entrapped within them during immersion in the extraction medium (usually saturated ammonium sulphate). These problems can largely be overcome by flowing air over the part of the intact plant that is of interest and either measuring the trace amounts of ethylene present with highly sensitive photoacoustic laser equipment (Voesenek et al., 1990), or by accumulating the ethylene in a trap until sufficient is available for conventional analysis by flame ionization gas chromatography (Bassi and Spencer, 1979). Unfortunately, studies of root-shoot relationships are only beginning to benefit from these improved techniques. The literature on ethylene involvement in root to shoot communication is small compared to that for cytokinins, GAS and ABA. Furthermore, it is limited to the study of flooding and centres on attempts to explain how the stress gives rise to marked epinastic curvatures that commence in the leaves within less than 24 h. Unlike the findings obtained with the other hormones, the movement of a precursor (ACC) from roots to shoots is involved in addition to that of the hormone itself. Studies of waterlogging have provided the first, and by far the clearest example of a physiologically active positive message moving from the roots to the shoots of stressed plants.

B. FLOODING

The idea that flooding soil leads to ethylene-mediated responses in the shoot can be traced back to Turkova (1944), who recognized the similarity getween the effects of ethylene on shoots of potato or tomato plants and certain morphological effects of flooding. The connection was strengthened about 25 years later when Smith and Russell (1969) found that flooded soil can accumulate considerable amounts of ethylene, and when Musgrave et al. (1972) demonstrated that submergence of plant tissue raises internal ethylene concentrations substantially as a result of entrapment arising from the extremely slow diffusion rate of ethylene in water. Kawase (1972) was the first to measure elevated concentrations of ethylene in gas extracted from shoot tissues above the water line. Other similar reports soon followed (e.g. El Beltagy and Hall, 1974; Smith and Jackson, 1974; Tang and Kozlowski, 1982) thus raising the prospect that ethylene, as a positive message, may move from inundated parts to responsive tissues above. Likely consequences of ethylene enrichment for shoot development include slow leaf extension, adventitious rooting, stem swelling, outgrowth of lenticels, and epinastic leaf curvature (Jackson and Drew, 1984). A causal involvement of ethylene in epinastic curvature was indicated by the

146

M. B. JACKSON

inhibition of epinastic curvature achieved by supplying flooded plants with inhibitors of ethylene action such as silver nitrate, di-iodohydroxybenzoic acid and carbon dioxide (Jackson and Campbell, 1976a; Bradford and Dilley, 1978; Wilkins et al., 1978). That ethylene can move from roots to shoots was demonstrated by Zimmerman et al. (1931), who succeeded in promoting epinastic leaf curvatures in tomatoes by exposing only the roots to ethylene. Jackson and Campbell (1975a) found that 2 p1 I-’ bubbled through nutrient solutions was sufficient to bring about the effect within approximately 3 h and demonstrated, with radiolabelled ethylene, that direct transfer of the gas can occur. Steam-girdling to collapse non-stelar tissues of the stem inhibited this movement, indicating that transport was mostly through intercellular spaces of the stem, although the transpiration stream can also transport ethylene (Eklund, 1992). However, several lines of evidence point away from ethylene transport per se as a major functional link between roots and shoots. The transport pathway is highly inefficient due to extensive radial diffusion (Zeroni et al., 1977) and not all soils accumulate large amounts of ethylene when flooded. Furthermore, the roots of flooded plants are unlikely to make much ethylene because anaerobic conditions arrest biosynthesis of the gas (see Section V1I.A). This suggests there may not always be a rich source of ethylene in or around the roots of flooded plants to supply the shoot. An alternative mechanism for enriching shoots with ethylene was indicated by experiments in which plants of Lycopersicon esculentum were grown in nutrient solution depleted of oxygen with flowing nitrogen gas. Despite the absence of soil ethylene, and of any stimulation of root ethylene production, the plants quickly developed epinastic leaf curvatures in close association with increased concentrations of ethylene in the petioles and other shoot tissues (Jackson and Campbell, 1975b, 1976a). The effect could not be reproduced simply by removing roots, indicating that a negative or accumulation messages were not responsible. Instead, a positive message passing from anaerobic roots that stimulated ethylene production in aerobic shoot tissue was proposed (Jackson and Campbell, 1975b, 1976a). The occurrence of faster ethylene production in leaves of plants with anoxic roots has since been confirmed, and occurs co-incidentally with epinastic leaf curvature (Bradford and Dilley , 1978; Jackson et al., 1978). Other species also form more ethylene in shoot tissues when the root environment is waterlogged (Phaseolus vulgaris-Wadmanvan Schravendijk and van Andel, 1986; Rumex palustris-Voesenek et al., 1990). Experiments using plants with vertically divided root systems indicated that a positive message from anaerobic roots, promoting ethylene production and epinastic curvatures, can be directed to one or other of the lowest leaves of the plant, through the xylem, by exposing only one of the half-root systems to anaerobic conditions (Jackson and Campbell, 1975b; Jackson, 1980). Re-orientation, to the upright, of the normally horizontal shoot extension that characterizes the tomato mutant “diageotropica” is

PLANT HORMONES AND COMMUNICATION

147

another response to ethylene that can be reproduced by depriving the roots of oxygen; reversion to a more orthodox stance in the anaerobically treated plants is associated with faster ethylene production by the shoot (Jackson, 1979). Because anoxia can temporarily interpose abnormally high levels of resistance to water uptake by roots (Mees and Weatherly, 1957), the possibility is sometimes entertained that the extra ethylene production in shoots of flooded plants results from stress in shoot tissues caused by water shortage, rather than from a positive message delivered by the anoxic roots. However, although this could perhaps be a factor in flooded Phaseolus vulgaris (Wadman-van Schravendijk and van Andel, 1986), where leaves wilt in response to flooding, the shoots of flooded tomato plants develop epinastic curvatures and produce increased amounts of ethylene in association with higher, rather than lower, leaf water potentials (Jackson et al., 1978). A very small and temporary decrease observed during the first few hours of flooding tomato plants is unlikely to be of significance because water shortage in intact plants is not very effective at promoting ethylene production (Morgan et a f . , 1990). The clue that the message transported in the transpiration stream was a precursor of ethylene came from observations of abnormally rapid ethylene production by roots returned to air after a period without oxygen (Jackson et a l . , 1978). Using fruit, Adams and Yang (1979) explained the effect in terms of a rapid oxidation to ethylene of the precursor ACC that accumulated during anoxia. The accumulation of ACC in anoxic roots seems to be attributable both to a blocking of oxidation to ethylene, and to an increase in ACC synthesis probably brought about by increased ACC synthase production and activity (Wang and Arteca, 1992). Bradford and Yang (1980a) found that some of this accumulated ACC can enter the bleeding xylem sap of flooded tomato plants exuding from plants de-topped after waterlogging for 12-72 h. They proposed ACC as the positive message promoting shoot ethylene production and petiole epinasty. The effect was also obtained when root systems were treated anaerobically after the shoot was removed, indicating that the ACC originated in roots and not in shoots. To help test if the amounts of ACC in xylem sap were sufficient to explain the rates of ethylene production and petiole epinasty, delivery rates of ACC were calculated by multiplying concentration in bleeding sap by sap flow rates. Fluxes of 0.5 nmol h-' per root system into a shoot bearing 5-7 leaves were found after 12 h flooding, increasing to 3.0 nmol h-' after 48 h. When excised whole shoots were fed ACC at similar rates, a flux of 2nmolh-' was sufficient to induce some epinasty, although more was needed to raise ethylene production. Evidence that roots and shoots are actually enriched with ACC was not provided, but recent work has shown this does indeed take place after 5 h (roots) and 24 h (shoots) (Wang and Arteca, 1992). One possible problem with the estimates of delivery rates of ACC is that the sap

148

M. B. JACKSON

flows used in the calculations were much slower than those of whole-plant transpiration. If ACC in xylem sap is not diluted in strict proportion to the rate of water flow, these calculations could be erroneous when extrapolated to whole plants. However, a test using pressure vessels to vary the flow of sap through detached root systems has shown that ACC is diluted by increasing rates of sap flow in both flooded and non-flooded plants (Fig. 8a). The dilution is roughly in proportion to sap flow rate since ACC delivery was higher in flooded than in non-flooded plants and similar over a range of sap flow rates, including those approximating to whole-plant transpiration (Fig. 8b). These results confirm that large increases in the output of ACC do take place in response to a few hours soil waterlogging (Fig. 8b). The ACC measurements were made using gas chromatography and nitrogen/ phosphorous detection and confirmed by GC-MS (Hall et af., 1992). The superior specificity and sensitivity of this technique over the conventional indirect assays for ACC have revealed an ACC flux from roots to shoots even in non-waterlogged plants.

T

E

-120

.

g3

Flooded

W

c 90

0 .c . ' ! c.

' -

C 60

8 0

b..

.... Flooded =.

30

t 0

1

2

3

4

Fig. 8. Effect of increasing the flow of sap through detached root systems of tomato plants grown in well-drained soil or in soil flooded for 12h on (A) the concentration of 1aminocyclopropane-I-carboxylicacid (ACC) in xylem sap and (B) the delivery rate of ACC from the cut stump. Arrows show transpiration rates of comparable intact plants. Means of four replicates. From M. A. Else, W. J. Davies and M. B. Jackson (unpublished results).

PLANT HORMONES AND COMMUNICATION

149

The results, thus far, lend persuasive, quantitative support to the hypothesis that, at least in tomato plants, anoxic roots export sufficient ACC to raise ethylene production in the shoots and promote epinastic curvatures. Similar studies with the tomato mutant “diageotropica” indicated that increased ethylene production in shoots in response to root anaerobiosis could also be explained by ACC transported from the roots (Bradford and Yang, 1980b). The hypothesis that ACC is the active message emanating from oxygen-starved roots has received further support from experiments with the ethylene biosynthesis inhibitors amino-oxyacetic acid (AOA), aminoethoxyvinylglycine (AVG) and cobalt chloride (Bradford et al., 1982). AOA and AVG interfere with the conversion of SAM to ACC catalysed by ACC synthase, and when administered to anoxic roots prevent increases in ACC flux in xylem sap. The attendant inhibition of increased ethylene production and epinasty in the shoots provides an elegant demonstration of root-derived ACC as a physiologically active positive message. This role was highlighted by showing that when the ACC is prevented from being oxidized to ethylene by administering cobalt chloride, increases in epinasty and ethylene production are again suppressed, even though ACC entry into the shoot proceeds unabated (Bradford et al., 1982). C. CONCLUSIONS

Studies of the physiology of flooding effects on shoot development, of the biochemistry of ethylene biosynethesis, and the measurements of ethylene and ACC, together form a compatible whole that strongly supports a role for ACC exported from anoxic roots as a positive message that influences shoot development and morphology. The approaches used to quantify the message, and to test it physiologically, conform to many of the requirements of good practice outlined in Sections II.B, 1I.C and Table I. It is unfortunate that the work has been almost entirely restricted to tomato plants. There is a pressing need to extend the findings to other species.

VIII. ABSCISIC ACID A . INTRODUCTION

Abscisic acid (ABA) is a sesquiterpene derived, like all plant terpenes, from mevalonic acid. More immediately, ABA is produced following the cleavage of a C ~ carotenoid, O violaxathin, to form xanthoxin and then ABA aldehyde, the immediate precursor of ABA. The pathway is known to operate in roots (Parry and Horgan, 1992). As with the GAS, biochemical analysis of deficiency mutants (“wilty mutants”) has helped to establish the biosynthetic pathway (Taylor, 1991), but only after many years of

150

M. B. JACKSON

uncertainty and contradiction. Chemical identification of ABA in plant extracts was achieved in the mid-1960s. This was the culmination of efforts to discover hormones regulating bud dormancy and leaf abscission, hence the earlier names of dormin and abscisin I1 (Addicott ef uf., 1968). Subsequent research has shown that ABA is not much involved in either of these processes, but there is persuasive evidence for believing that the (+) form of the cispuns-isomer plays a role in imposing dormancy in newly formed seeds and in promoting stornatal closure (Milborrow, 1984). Leaf expansion may also be regulated by endogenous ABA (Dale, 1988), along with a range of other effects for which evidence of endogenous regulation is less certain (Trewavas and Jones, 1991). Increases in ABA arc commonly seen in plants subjected to environmental stress, leading to the widely held view, pioneered by Wright (Wright and Hiron, 1969), that ABA is a ‘stress’ hormone. Physiological studies of ABA have benefited from the successful development of straighforward physicochemical and immunological assays. Electron capture gas chromatography of methylated derivatives is sensitive and suitable for routine work, especially if preceded by HPLC purification, and if a radiolabelled internal standard and a chromatography injection standard, such as ethyl-ABA, are used (e.g. Hall and Jackson, 1986). Immunoassays can be even more rapid and sensitive (Weiler, 1980), and recent developments using monoclonal antibodies and a scintillation proximity reagent (Whitford and Croker, 1991) allow hundreds of analyses to be processed dependably within a few days. The most reliable results are those from procedures which have been verified by GC-MS. Unlike the literature for cytokinins and GAS, the majority of papers involving ABA and rootshoot relationships are less than 10 years old and most ABA measurements have been made with physicochemical or immunological methods rather than with bioassays. The ABA literature is also distinctive in taking more fully into account the possible contributions of changed water and mineral relations, and in adopting a critical and quantitative approach to delineating a role for root-sourced ABA. By far the largest literature concerns the stornatal closure and slowed leaf expansion that commonly accompanies soil drying.

B. WATER DEFICIENCY AND STOMATAL CLOSURE

An extensive literature supports the idea that a loss of hydration in one or more parts of the plant promotes stomatal closure by increasing the amount of ABA arriving at the vicinity of stornatal guard cells, or perhaps entering them (Reid and Wample, 1985; Hartung and Baier, 1990). Current thinking centres on the belief that apoplastic ABA is more important than symplastic ABA, since receptor sites for the hormone are probably present on the outer

PLANT HORMONES AND COMMUNICATION

151

surface on the guard cell plasmalemma (Hartung, 1983). Increases of only 1.2 x loT5fmol ABA per guard cell have been claimed to be necessary to effect some closure, although sensitivity to ABA differs between plants (Schurr and Gollan, 1990), between leaves of different ages (Atkinson et al., 1989) or cultivars (Loveys, 1984b), and is subject to some environmental regulation (Tardieu and Davies, 1992) mediated by ABA itself (Atkinson et al., 1989).

1. Physiological SigniJcance of ABA in shoot apoplast and xylem sap Evidence that stomatal aperture is related causally to the concentration of ABA in the xylem sap of leaves or shoots is quite strong, but published data need to be examined carefully to be sure that the increases in ABA are not simply a result of less dilution as closing stomata slow transpiration rates. Loveys (1984b) found that in thoroughly watered, glasshouse-grown Vitis vinifera, concentrations of ABA in xylem sap, obtained after centrifuging lengths of excised stem, were too dilute for physiological significance (approximately 5 X lop5mM) and stomatal apertures varied little throughout the day. However, in well-watered field-grown plants, xylem sap ABA was more concentrated and increased during the first few hours of each photoperiod to 3 X 1 0 - 4 m ~ ,by which time leaf water potentials had decreased by 1MPa and stomata had begun to close after several hours of steadily increasing apertures. The increase in ABA preceded a slowing in transpiration and thus represents a potential cause rather than a consequence of stomatal closure. In support of this, feeding ABA by transpiration to detached field-grown leaves at concentrations similar to the highest levels found in xylem sap of field-grown plants resulted in rapid partial stomatal closure. Especially convincing are calculations of ABA uptake rates from stem xylem into the leaves of plants growing in the field. Uptake was calculated to increase to a peak at the time stomata began to close, while simulating these delivery rates by feeding ABA to detached, transpiring leaves achieved approximately 30-50% stomatal closure (Loveys, 1984b). Much of the ABA in the xylem sap of well-watered field grown plants may simply have been there as a result of a redistribution of leaf ABA from symplast to apoplast as water potentials fell, but Loveys (1984b) considered the roots also to be a possible source. However, in droughted Vitis vinifera, only about 50% of the increase in foliar ABA could be accounted for by uptake from shoot xylem (Loveys and During, 1984), indicating that in more severely stressed plants, enhanced synthesis by leaf tissue also takes place. In droughted apricot trees (Prunus armenica), Loveys et al. (1987) found ABA in xylem sap to be only 5% of that required to close stomata when ABA was tested for activity on detached leaves. However, this relates well to the observation that, in this species, stomata do not close readily under droughted conditions. A similar picture may also apply to almond trees (Prunus dulcis) (Wartinger et al., 1990). Since the work of Loveys, others

152

M. B . JACKSON

(e.g. Hartung el a f . ,1990) have also linked changes in ABA in xylem sap of leaves or shoots to the extent of stomatal closure, especially in field-grown plants. In droughted, mechanically impeded and irrigated maize, the coupling between leaf conductance and ABA concentration in sap extruded from detached leaves by 5 kPa pressure, was far closer than that between leaf conductance and bulk leaf ABA (Tardieu e t a f . ,1992a,b,c). Curvilinear relationships between averaged xylem sap ABA concentrations and stomatal apertures at mid-day were seen in several treatments over a number of days, and also for individual plants within these treatments and sampled on the same day. Over most of the range of leaf conductances the relationship with xylem sap ABA was linear and steep. However, if the reasonable assumption is made that transpiration flow dilutes ABA in an approximately linear fashion, the increases in ABA concentrations in maize leaf xylem sap can be more than accounted for arithmetically by decreased dilution resulting from slower leaf conductance (Table IV). Thus, the distinction between cause and effect with respect to xylem ABA concentrations and stomatal apertures is clouded in this particular analysis. Further difficulties have been raised by Munns and King (1988), who found that the concentration of ABA in sap forced from the base of severed leaves of intact laboratory-grown Triticum aestivum plants by gentle ) pressures to the roots, was enriched with ABA 50-fold (to 5 x 1 0 - 8 ~ by droughting that was severe enough to partially close the stomata. However, this ABA concentration was 100 times too dilute to close stomata in tests where synthetic ABA was administered to detached wheat leaves transpiring at similar rates to whole plants (on a unit area basis). Xylem sap ABA may also be inadequate to explain stomatal closure in droughted Phaseofus vufgaris (Trejo and Davies, 1991), indicating the presence of some other stomata-closing agent. Munns and King (1988) reported that xylem sap of wheat may contain such a substance. The claim was based on finding anti-transpirant activity in xylem sap surviving passage down an immunoaffinity column intended to remove ABA. However, the effectiveness of the column was not actually demonstrated and subsequent work (Munns et a f . , 1991) has shown that anti-transpirant activity not attributable to ABA was an artifact generated by freezing the sap at -20°C for several days prior to testing. When Zhang and Davies (1991) repeated and extended Munns and King’s approach, using maize, all the anti-transpirant activity of xylem sap was attributable to its ABA content when tested in a detached maize-leaf transpiration assay, and on stomatal apertures of Commelina communis epidermal strips. Their results benefitted from prior microfiltration of xylem sap to remove particulate material that otherwise can block xylem vessels. In their detached leaf assay, Zhang and Davies (1991) found anti-transpirant activity of ABA solutions and of sap from droughted or well-watered plants generated a reasonably consistent response curve to the ABA they contained when plotted on a log scale (Fig. 9a). The upper range of ABA

TABLE IV Relationship between stomatal conductance and A BA concentrations in xylem sap from leaves of Zea mays grown in irrigated, non-compacted soil or non-irrifated, compacted soil Irrigated, non-compacted soil ABA concentration in extracted xylem sap (Fmol m-2)

15.9 19.8 19.8 29.7 35.7 39.6 39.6 43.7 51.6 73.4

-0

Non-irrigated, compacted soil

Leaf conductance (mol m-'s-l)

ABA concentration in sap flowing at 1mol m-2 s-'

ABA concentration in extracted xylem sap (pmol m-2)

Leaf conductance (mol m-2 s-')

ABA concentration in sap flowing at 1mol m-'s-'

0.19 0.17 0.13 0.18 0.22 0.20 0.18 0.14 0.17 0.17

3.0 3.4 2.6 5.5 7.9 7.9 7.1 6.3 8.8 12.6

47.6 55.6 63.5 63.5 99.2 111.1 111.1 126.9 131.0 138.9 151.0 198.4

0.083 0.063 0.097 0.029 0.098 0.022 0.040 0.011 0.063 0.012 0.034 0.009

4.0 3.5 6.2 1.8 9.7 2.4 4.4 1.4 8.3 1.7 5.1 1.8

Data indicate that increases in ABA concentration in sap extracted from leaves that resulted from non-imgationkompaction can be more than accounted for by less dilution as transpiration slowed in response to stomatal closure. These data are taken from a more extensive set of results of Tardieu er al. (1992bFig. 5b). From these data, concentrations of ABA were normalized to leaf conductances of 1 mol m-Zs-'.

5:5 3: 0 XJ

52

M

m

%U

8K

5z

0

2

0 2

154

M. B. JACKSON

2.5

-- 2.0 I

u)

NI

-E

-

1.5

E

c 0 .c

E

.E 1.c c

u)

E

t

0.5

1 1 1 1 1 1 1 1 a 1 0

0.01

0.1

1

10

0

0.05 0.1

0.15

0.1

Abscisic acid ( p ~ )

Fig. 9. Effect on transpiration rates of detached leaves of maize (Zeu mays) of supplying xylem sap obtained from droughted (0)or well-watered ( A ) maize plants containing known amounts of naturally produced ABA, or (0)xylem sap after much of the natural ABA was removed by an immunoaffinity column, or (0)synthetic ABA at known concentrations in water. (a) Log plot that includes all the original data. (b) Linear plot after the two highest ABA concentrations were removed. The results indicate that anti-transpirant activity of extracted maize xylem sap is largely explicable by its ABA content, when tested on excised leaves. From Zhang and Davies (1991).

concentrations chosen by Zhang and Davies is somewhat higher than those normally present in xylem sap. If these high values are removed, a plot of the anti-transpirant activity against ABA on a linear scale (Fig. 9b) gives a clearer picture of the relationship. However, as discussed earlier, the increased ABA in xylem sap of field-grown plants over the range associated with increased stornatal closure could equally have been a result of stornatal closure and slower transpiration, rather than their cause. Thus, evidence causally linking stomatal closure with increases in ABA in leaf xylem sap is sometimes equivocal. This is surprising since ABA is a highly active promoter of stomatal closure and, as yet, no other more effective constituent of leaf xylem sap has been identified. One technical reason for the difficulty may be that too much attention has been given to ABA concentration in sap, which is subject to transpiration-linked variability. Changes in delivery rate into the leaf would be a more reliable measure of ABA loading of the stomata. Where this has been examined

PLANT HORMONES AND COMMUNICATION

155

specifically (see Loveys, 1984b, and Schurr and Gollan, 1990), the results have indicated marked increases in ABA input into leaves by xylem transport when stomata are closing. 2. Mechanisms raising apoplastic ABA in leaves of droughted plants The most likely signal from the roots of droughted plants that can raise xylem sap ABA in leaves is a negative hydraulic message causing some leaf dehydration. This could inhibit plasmalemma ATPases, resulting in slower outward proton transport and, consequently, a more alkaline apoplast. In turn, the diffusion gradient of membrane-permeable undissociated ABA into the alkaline apoplast would be enhanced, thereby promoting efflux. This redistribution of existing leaf ABA to the apoplast may then have been reinforced by the well-known increased synthesis of ABA in leaf cells that, in an unexplained way, results from a loss of leaf hydration (Wright and Hiron, 1969; Pierce and Rashke, 1980). Movement of apoplastic ABA to the exterior of symplastically isolated guard cells may be achieved by diffusion; or by mass flow in leaf xylem sap, which is also more alkaline in water-deficient leaves (Hartung et al., 1988). Transpiration flow would probably deposit xylem ABA at the guard cells (Cornish and Zeevaart, 1985b) or nearby epidermal cells, since these are sites of intense water evaporation (Aston and Jones, 1976). However, there is considerable evidence that drying roots can also close stomata and raise ABA levels in leaves in the absence of changes to leaf hydration. This implies that a non-hydraulic message is conveyed from roots to shoots. The next section reviews evidence for this and for believing that the message is ABA exported from the roots in the transpiration stream.

Evidence for drying roots as a source of apoplastic ABA. The need to hypothesize a non-hydraulic message from roots to shoots that closes stomata comes from observations that plants growing in drying soil can close their stomata with little or no loss of water potential or turgor in the responding leaves (Meyer and Gingrich, 1964; Bates and Hall, 1981; Sharp and Davies, 1985). In plants with root systems separated vertically or horizontally into two, droughting one set of roots can close stomata even though leaf hydration is sustained by water taken up from the second, non-droughted set (Blackman and Davies, 1985). Partial dehydration of only a few roots appears to be needed in laboratory-grown plants (Neales et al., 1989), although in the field, most of the root system must be stressed (Tardieu et al., 1992b). Thus, closure can, seemingly, be a response to water deficits in the roots even when the leaves are fully turgid. This possibility has been elegantly confirmed by experiments involving pressurizing roots of droughted Helianthus annuus or Triticum aestivum plants just enough to maintain leaf water potentials and turgor at levels close to those of wellwatered plants. Despite the absence or more negative leaf water potentials when droughted plants are grown in this way, stomata still close (Passioura

156

M. B . JACKSON

and Munns, 1984; Gollan et al., 1986). Explaining the transmission of this stimulus for stomata1 closure in terms of increased output of ABA from drying roots has been a considerable experimental challenge, and is described below. Since the work of Davison (1965), ABA or substances with its physiological properties have been known to be present in sap obtained from de-topped root systems (Davison, 1965; Skene, 1967; Lenton et al., 1968; Davison and Young, 1973) and roots may supply at least some of this since they also contain ABA, especially at the tips (Rivier et al., 1977; Bottger, 1978) and produce the hormone in excised cultures (Tietz, 1975; Hartung and Abou-Mandour, 1980). Roots, like shoots, can increase their ABA content in response to water shortage or osmotic stress (Barr, 1973; Milborrow and Robinson, 1973; Walton et al., 1976; Rivier et al., 1983, but see Cohen et al., 1978) and within 1h (Ribaut and Pilet, 1991). Using excised maize roots, and three different osmotica, Lachno (1983) showed that the rise can take place independently of the shoot system and that some of this ABA finds its way into xylem sap. Lachno found that 18h at -507 kPa (200 mol mP3mannitol) more than doubled the ABA concentration in root tissue of various ages, including older parts where, in intact plants, water influx might be more extensive overall (Sanderson et al., 1988), and thus effectively carry much of the extra ABA into the xylem. When roots were exposed for 2 h to sufficient polyethylene glycol 6000 to impose a modest osmotic stress, ABA concentrations in the xylem sap were much increased (Fig. 10). Similar amounts of sap (5-8 g per 100 roots) and root tissue (10 g fresh weight) from stressed and unstressed tissue were used to generate these data (Lachno, 1983), which support the view that roots can increase their content of free ABA in response to moderate dehydrating stress independently of the shoot system, and can transfer considerable amounts of the hormone to xylem sap. Cornish and Zeevaart (1985a,b) came to similar conclusions after finding that despite stem-girdling to inhibit ABA export from shoots, ABA in roots of intact Xanthium stromarium plants was much increased when water deficient. Excised roots of Xanthium were also responsive to drying to 70% of normal fresh weight, or to submerging in an osmoticum. The increases in ABA were seen within 2 h and could be halted by re-hydration (Cornish and Zeevaart, 1985a,b). Concentrations of ABA in roots, on a dry weight basis, increase with decreasing soil water content, and roots lower down the soil profile become increasingly enriched with the hormone as the soil dries (Zhang and Davies 1989a). Since these increases in root ABA take place in the absence of decreased leaf hydration, it is unlikely that the roots are recipients of ABA transported from water-stressed leaves. However, this cannot be entirely ruled out since water relationships are usually only checked for young fully expanded leaves. It is important that this possibility is examined more carefully. Older leaves, especially, could make a contribution (Zhang and Davies, 1989b) because they are more

PLANT HORMONES AND COMMUNICATION

157

2.1 -

-,

1.8 -

0)

0

c Q I E

1.51.2 -

5.

.E

sa

-

0.9 -

0.6 -

Control

0.3 0

-100 -200 -300 -400 -500 -600 Osmotic stress (kPa)

Fig. 10. Abscisic acid concentration in xylem exudate from isolated roots of Zea mays collected for 12 h following osmotic stress imposed by polyethylene glycol 6000. The results indicate that stressed roots are able to enrich xylem sap with ABA independently of the shoot. From Lachno (1983).

prone to large hydraulic resistances or loss or stomatal regulation, and therefore may lose turgor before younger leaves (Atkinson et al., 1989). There is evidence that some of the extra ABA they may make would be transported to roots in the phloem and could enter the xylem sap, and thus be recycled (Shindy et al., 1973; Hoad, 1975,1978; Loveys, 1984a). Wolf et al. (1990) calculated that in osmotically stressed Lupinus albus 45% of the ABA present in the xylem sap originated in shoot tissues. Building on the above findings, much experimentation has sought to demonstrate that increases in ABA from drying roots in xylem sap are sufficiently large and timely to effect stomatal closure. Zhang et al. (1987) reported that the leaf epidermis of Commelina communis was enriched with ABA in association with stomatal closure when one-half of a divided-root system was droughted sufficiently long (>3-4 days) to raise root ABA concentrations, while the other half was watered sufficiently well to prevent loss of leaf water potential or turgor. The authors concluded that sufficient ABA moved from the drying half-root-system to reduce leaf conductances in unstressed leaves. Why roots in drying soil are not hydrated by roots that are still moist is unclear (see Saab and Sharp, 1990; Blum and Johnson, 1992). The mechanism by which water-borne ABA moves from zones of low water potential in drying roots to parts with higher water potentials is also uncertain. However, despite this, Zhang and Davies (1987) demonstrated, by direct testing, that ABA applied externally to roots, albeit at high concentrations, can indeed move to the shoots of Commelina from roots deprived of water.

158

M. B. JACKSON

In maize also, soil drying for up to 20 days causes increasing amounts of stomatal closure in the absence of marked changes in leaf turgor and leaf water potential. The closure has been linked to increases in ABA concentration in 1ml samples of xylem sap allowed to exude naturally from the stump of decapitated root systems over 3 h (Zhang and Davies, 1990b). The maximum ABA concentration, almost 180 nM, was seen after 16 days drought; at other times ABA concentration was between 4 0 - 6 0 n ~ .A further study showed up to 6 0 0 n ~ABA in xylem sap in a single plant ~ to roots of well-watered (Zhang and Davies, 1990a). Suplying 10 p , ABA plants was found to raise ABA in root exudate from 26 to 66 nM, which was sufficient to reduce stomatal conductances by 58% (Zhang and Davies, 1990b).This calibration exercise indicates that increases in concentration in droughted plants would probably be enough to cause some stomatal closure. Zhang and Davies (1990a) have evidence that the samples of sap they analysed from droughted plants were indeed captured transpiration sap rather than concentrated sap drawn slowly through the roots by osmosis. Thus, the concentrations are thought to be those extant in the transpiration stream of whole plants. If account is taken arithmetically of the slower transpiration arising from smaller leaf conductances, increases in concentration of xylem exudate shown by Zhang and Davies (1990b) are 2-3 times greater than those which can be accounted for by less dilution. This is evidence of increased output by droughted roots and of a modest positive ABA message from drying roots. The increased output from the roots presumably resulted from extra synthesis in the roots. A component of the effect may also be attributable to a larger amount of root relative to the shoot since the root : shoot ratio increases in response to water shortage (Sharp and Davies, 1985). This possibility highlights the advisability of expressing estimates of hormone levels in the transpiration stream in terms of output per unit amount of root system delivered to a known amount of recipient shoot tissue. As yet no such values are available for any experimental system. Evidence of increases in ABA in xylem root exudate accompanying stomatal closure of droughted plants with fully hydrated leaves has also been obtained for Helianthus annuus by Zhang and Davies (1989b) and Neales el al. (1989). In the former paper, ABA concentrations in xylem sap were found to increase from 0.028 nM to 0.078 nM after 6 days drought, and before any of the leaves suffered a loss of turgor or water potential and at a time when stomatal conductances over all leaves were reduced, on average, from 5 mm s-' to 3 mm s-' . When the effect of reduced dilution arising from the smaller leaf conductances is accounted for, the amount of ABA in the xylem sap, and presumably arising from the roots, appears to have been doubled by the drought, thus indicating a small positive ABA message. McLeod and Neales (1990) are the only authors to have calculated delivery rates of ABA from water-deficient roots. They briefly report that 3 h in polyethylene

PLANT HORMONES AND COMMUNICATION

159

glycol (- 1MPa osmotic stress) reduced delivery by 37% in one experiment and increased it by 137% in another. Changes in the output of ABA directly from the roots was indicated since defoliation had little effect on the results (Neales and McLeod, 1991).

3. Conclusions Results show clearly that stomata on certain leaves can close, at least partially, in response to soil drying, but without themselves first suffering from a negative hydraulic message. The balance of evidence indicates that apoplastic ABA is an important regulator of this stomatal closure, and that drying roots may enrich this hormone pool with physiologically active amounts of the hormone moving in xylem sap. Unsettling aspects that merit further research include: (1) the absence of published delivery rates of ABA from the roots-these are needed to demonstrate a genuine positive message; (2) the small scale of the increases in xylem sap ABA concentration druing the early stages of droughting and stomatal closure, once dilution effects are accounted for; (3) the possibility that in some experiments, at the time stomata first start to close, extra ABA may have originated in older, potentially water-deficient leaves, rather than in the roots (but see Zhang and Davies 1990a); (4)the possibility that ABA may not be the only active promoter of stomatal closure in xylem sap; and (5) uncertainty concerning the extent to which roots in the drier parts of the soil profile may be hydrated by roots in wetter parts (Blum and Johnson, 1992), thus blunting the sensitivity of the system to drying in the upper soil profile (Saab and Sharp, 1990). The impact of changes in the root :shoot ratio on ABA in xylem sap also merits further investigation.

C. WATER DEFICIENCY AND LEAF EXPANSION

Studies of stomatal closure during soil drying are often interwoven with experiments on leaf expansion. Gowing et ul. (1990) demonstrated, in a divided root system experiment, that droughting Malus x domesticu inhibits leaf growth in the absence of reduced leaf turgor. The effect was shown to be the result of a growth-inhibiting factor moving to the leaves from the roots, since re-wetting the drying roots, or removing them, resulted in a recovery of leaf expansion. Drying roots also inhibit leaf extension in wheat and lupin, even when the root system is pressurized slightly to maintain normal xylem and leaf water potentials (Kuang et al., 1990; Passioura and Gardner, 1990). Other measurements support these observations (e.g. Ludlow et al., 1989; Blum et al., 1991), indicating that a positive message, such as ABA, from drying roots is involved. In support of this interpretation, the ABA biosynthesis inhibitor, fluridone, applied to Zea mays seedlings grown at low water potentials (-1.6MPa), promoted leaf extension (Saab et al., 1990),

160

M. B. JACKSON

and Zhang and Davies (1990a) found a negative log-linear relationship between leaf extension in Zea mays or Helianthus annuus and ABA concentrations in xylem sap centrifuged from leaves or shoots or exuding from the cut surface of de-topped plants. Since even the oldest leaves did not suffer from any loss of water potential, and, thus, are unlikely to be the source of the extra ABA, Zhang and Davies concluded that the ABA was a positive message directly from drying roots. Feeding ABA through a few roots, in amounts that raised xylem sap ABA to levels similar to those of droughted plants, inhibited leaf growth (Zhang and Davies, 1990b). Leaf expansion may be more sensitive than stomatal closure to non-hydraulic messages such as ABA from drying roots (Saab and Sharp, 1990). D. SOIL FLOODING

The literature has a similar pattern to that describing drought responses but it is less extensive and is characterized by a paucity of data linking ABA in the shoot apoplast with stomatal behaviour or leaf growth. In contrast to the effects of soil drying, responses of shoots to flooding can take place within a few hours. This prompt reaction should assist in identifying cause and effect more clearly by avoiding potential complications from more general changes in growth pattern (e.g. in root :shoot ratio). Evidence for ABA as a positive message from oxygen-deficient roots is less convincing than that obtained from droughting studies. More use has been made of ABA-deficient mutants in flooding studies than in research into the root-shoot relationships of droughted plants. It has been known for many years that soil flooding can close stomata (e.g. Moldau, 1973; Coutts, 1981). In some circumstances closure is associated with increased bulk ABA concentrations in leaves which experience severe water deficits (e.g. Hiron and Wright, 1969; Shaybany and Martin, 1977; Sivakumaran and Hall, 1978; Wadman-van-Schravendijk and van Andel, 1985). These leaf water deficits are thought to be brought about by decreases in hydraulic conductivity of the roots during the early stages of oxygen starvation (Kramer and Jackson, 1954; Mees and Weatherley, 1957; Smit and Stachowiak, 1988). But, as in the case of droughted plants, stomata have often been found to close in leaves not suffering dehydration and lowered water potentials (Pereira and Kozlowski, 1977; Jackson et al., 1978; Bradford and Hsaio, 1982; Jackson and Kowalewska, 1983; Zhang and Davies, 1986,1987; Reece and Rhia, 1991), although a small and temporary reduction in leaf water potential ( ~ 0 . MPa 1 for an hour or so) can sometimes precede a loss of leaf conductance (Jackson et al., 1978), especially under conditions of high evaporative demand (Neuman and Smit, 1991). This mild and short-lived water defict might conceivably initiate closure through a redistribution of leaf ABA in favour of the apoplast; however in tomato plants Bradford and Hsaio (1982) showed that brief water deficits do

PLANT HORMONES AND COMMUNICATION

161

not cause long-term changes in leaf water relationships. Once stornatal closure is underway, it is sustained for several days in association with increased foliar ABA (Jackson and Drew, 1984; Jackson 1985b; Jackson and Hall, 1987; Neuman and Smit, 1991), despite leaf water potentials that are either less negative (Jackson et al., 1978) or similar to those of wellaerated control plants (Neuman and Smit, 1991). A causal link between stomatal closure and the increased ABA is indicated in tests with detached leaves of Pisum sativum, which partially closed their stomata when supplied with sufficient ABA to raise foliar levels to those of flooded plants (Jackson and Hall, 1987). In similar feeding experiments using detached Phaseofus vulgaris leaves, stomata did not close until bulk ABA levels were approximately double those of plants growing in hypoxic conditions (Neuman and Smit, 1991), suggesting an enhanced responsiveness to ABA in flooded plants. The involvement of ABA in the stornatal response to flooding is also indicated by very small changes in foliar ABA and stornatal apertures that take place when ABA deficient mutants of Pisum sativum (“wilty”) or Lycoperiscon esculentum (“flacca”) are subjected to oxygen deficient conditions at the roots (Jackson and Hall, 1987; Jackson, 1990, 1991). In plants showing leaf water deficits, much of the extra ABA is most likely to be formed in the leaves as a result of a negative hydraulic message that increases biosynthesis. Where leaf water deficits are absent, the roots could be a source, as seems likely in droughted plants. Evidence for and against this hypothesis is outlined below. 1. Evidence against ABA as a positive message f r o m roots Abscisic acid production is unlikely to be increased in roots by anaerobic conditions because the generally accepted biosynthetic pathway involves oxygenations. Abscisic acid contains atoms of oxygen at four positions. These become labelled with isotope when plant material is incubated in atmospheres containing lXO2.The pattern of labelling varies depending on the tissue and duration of exposure (Zeevaart et al., 1991). In roots, which are deficient in epoxy-carotenoids, “ 0 2 labelling of ABA takes place in the carboxyl group and at two positions on the ring (Creelman et al., 1987), indicating that molecular oxygen is needed for the final step in ABA biosynthesis and for carotenoid synthesis even in the short term. Thus, it is not surprising that when roots are waterlogged or subjected to anaerobic conditions, ABA concentrations are either unchanged or decrease (Jackson et al., 1988; Jackson, 1991; Shih-Ying and Van Toai, 1991). Predictably, therefore, pea plants grown in aerated or oxygen-deficient nutrient solutions decreased their leaf conductances in association with increased foliar ABA, but without attendant increases in root ABA (Jackson et al., 1988). Removing the shoot from oxygen-deficient roots for several hours, in an attempt to retain hormone in the root system, also failed to increase ABA concentrations. If stomata on flooded plants close in response to ABA from

162

M. B. JACKSON

the roots, it can be expected that grafting root stocks of an ABA-deficient mutant such as the “flacca” tomato onto wild-type shoots should suppress the effect. Although “flacca” roots are known to have an impaired ability to form ABA from its precursors, ABA aldehyde and ABA alcohol (Duckham et al., 1991), the effect on the wild-type scion is insubstantial. Similarly, grafting a wild-type root system onto an ABA-deficient shoot does not restore the stomatal and ABA responses to flooding (Jackson, 1990,1991). The possibility that the partial rather than total oxygen shortage that roots experience during the first few hours of flooding stimulates ABA production in roots seems unlikely, since detailed time-course measurements by Zhang and Davies (1987) showed no statistically significant increase in roots during the first 18h of flooding (there was a decrease if bound ABA is taken into account), even though stomata had begun to close within 8 h. There are few published measurements of ABA in xylem sap, the presumptive carrier of ABA from oxygen-deficient roots to the leaves. Sap from de-topped roots of Populus hybrid cuttings pressurized to 0.3 kPa were not statistically different after 3-48h without oxygen, although delivery rates of the hormone, calculated by the authors from unstated flow sap rates, showed a decrease of approximately 30% after 3 h and 77% after 48 h. The value of calculating delivery rates is shown here very clearly. In contrast to this pattern, ABA increased in the leaves after 3 h and returned to normal after 48 h, thus breaking any obvious link between ABA delivery rates from roots and leaf ABA levels (Smit et al., 1990). Similarly, decreases in ABA concentrations in roots and bleeding sap from de-topped barley plants after 2-7 days flooding were found by R. Munns and colleagues (personal communication), even though the stomata were closing and leaf extension was slowing in association with increases in bulk leaf ABA and with an absence of leaf water deficits. The discussion above has centred on stomatal closure, but Smit and colleagues have also examined leaf expansion, which is slowed by flooding, probably as a consequence of stiffening cell walls (Zhang and Davies, 1986) that can occur without loss of water potential (Smit et al., 1990; Neuman and Smit, 1991). A positive message was implicated by finding that hypoxic treatment slowed leaf expansion over and above that obtained by girdling the stem. Starting the stress at the beginning of the photoperiod rather than at the end was found to be more effective. This was interpreted in terms of transpiration pulling in more of the putative positive message, although such a view does not fit well with the probability that message delivery would be unaffected by sap flow rate (Fig. 2). When tested in a leaf growth bioassay, xylem sap from the de-topped roots of Phaseolus vulgaris and Populus inhibited the expansion of leaf discs, with sap from plants with hypoxic roots being the more effective. The amount of ABA in sap from hypoxic plants was adequate to explain the effect, but only by postulating a second factor that enhanced ABA activity in the stressed plants (Smit et al., 1990; Neuman

PLANT HORMONES A N D COMMUNICATION

163

and Smit, 1991). This indicates that a positive message is operating to slow leaf growth. It is unfortunate that the tests were made using sap at approximately twice the presumed concentration in the transpiration stream. Derooting and leaf excision studies in Pisum sativum (Jackson and Kowalewska, 1983) also indicate a positive message from roots that is toxic to leaves. Abscisic acid, ethanol and ACC were eliminated as likely causes, leaving chemical identity uncertain. 2. Evidence supporting A BA as a positive message from roots Despite the likely need for oxygen and non-reducing conditions to sustain ABA biosynthesis, there is evidence that increased amounts of the hormone pass to the shoots of flooded plants from the roots. Increases of loo%, 500% and 700% in ABA concentrations of roots of flooded pea plants were found by Zhang and Davies (1987) during the second, third and fourth photoperiods, during which time stomata were partially closed and bulk leaf ABA was increasing. Although these changes were preceded by stomatal closure, and thus cannot explain the initiation of stomatal closure, they could help to sustain closure. Since the leaves were not water deficient, the authors suggested that the flooded roots were the source of the extra ABA, but no supporting evidence with measurement of xylem sap ABA was provided. However, Neuman and Smit (1991) provide weak evidence that xylem sap is enriched with ABA in Phaseolus vulgaris treated hypoxically for 3-4 h under conditions of high evaporative demand. They found ABA in sap driven from de-topped roots by 3kPa pressure was increased from 4.9 to 1 0 . 8 n ~ . However, without sap flow rates it is impossible to eliminate the risk that this was simply a result of less dilution as a result of slower sap flow. Treating roots hypoxically for 3-4 h after removing the shoot to eliminate the leaves as a source of hormone gave only a small increase in xylem sap ABA (4.7 nM well-aerated, 6.4 n~ hypoxic) , suggesting that a significant contribution from the foliage rather than from the roots may well take place in intact plants. Clearer results have been obtained with sap from de-topped tomato plants sufficently pressurized to create sap flows approximating to transpiration rates of intact flooded o r well-drained plants. Using this approach, flooding for 24 h or more prior to de-topping increased both the concentration of ABA in xylem sap and the delivery rate from the roots (M. A . Else et a l . , unpublished; Fig. 2). At this time, in the youngest fully open leaves, stomata were partially closed, bulk ABA concentrations were increased, and water potentials were less negative than controls. These findings support the notion that flooded roots enrich the shoots of tomato plants with ABA. They also provide the basic data needed for tests of the physiological significance of ABA in xylem sap.

3. Reconciliation of findings The apparently contradictory findings in Sections VII1.D. 1 and VIII.D.2, may be reconciled with the hypothesis that most, if not all, the increases in

164

M. B. JACKSON

ABA in the roots, xylem sap or shoot originated in the shoot. It is possible that at least some of the extra ABA in leaves and sap may represent an accumulation of ABA no longer exported in the phloem to active sinks such as root tips, shoot apices and undeveloped leaves, the growth of which is suppressed by soil flooding. There is ample evidence that when phloem transport is active, ABA and other hormones are transported out of leaves to the roots and shoot tips (Hocking et al., 1972; Shindy et a l . , 1973). It is also known that ABA accumulates in leaves in amounts sufficient to close stomata when phloem export is blocked by girdling (Setter et al., 1981; Henson, 1984, but see Smith and Dale, 1988). The lack of actively growing sinks in flooded plants may have the same effect. In support of this accumulation message theory, detached leaves from well-aerated plants of Pisum sativurn incubated with their petioles in water close their stomata after a delay of 1 day. This occurs in association with increases in ABA compared with leaves on well-aerated intact plants and in the absence of any loss of water content (Jackson and Hall, 1987). Against the idea are results of Smit et al. (1990) and Reece and Rhia (1991) showing that stem-girdling could not fully reproduce the effect of flooding root systems of larch, white spruce or poplar. Instead of an accumulative message, these authors favour a positive message from the roots to explain stomatal closure, but provide little other supporting evidence. A negative hydraulic message cannot be ruled out as the instigator of the increase in sap and leaf ABA since leaf water potentials were not measured in all leaves and could have been more negative in older leaves as a result of early increases in root hydraulic resistance. In these circumstances, any additional ABA generated by water-deficient older leaves could circulate throughout the plants via interchange between phloem and xylem. This recirculation can occur in droughted (Hoad, 1975, 1978) and osmoticallystressed plants (Wolf et al., 1990). Such a process could explain why ABA concentrations in xylem sap forced through de-topped roots of Phaseolus vulgaris starved of oxygen for 3-4 h were somewhat higher if the shoot was removed at the end rather than at the start of the stress (Neuman and Smit, 1991). E. VARIOUS OTHER STRESSES

Drought and flooding are not alone in causing stomatal closure or decreased leaf expansion in the absence of reduced leaf hydration or water potentials. Nitrate starvation in tomato and barley can also have these effects (Chapin et al., 1988). The responses are linked with increases in ABA concentration of leaves and, albeit later, the roots (Chapin, 1991) and in xylem sap flowing at unstated rates (Krauss, 1978), raising the possibility of a positive message. However, leaf growth and stomatal aperatures were also strongly reduced when plants of the ABA-deficient tomato “flacca” were starved of nitrogen.

PLANT HORMONES AND COMMUNICATION

165

This result largely discounts ABA involvement. A complication, that has ramifications for all studies of stress responses, is that nitrogen (and also phosphorus) starvation enhances the responsiveness of stomata to ABA, in addition to raising ABA levels in leaves (Radin et al., 1982). Cooling roots of Phaseolus vulgaris from 23°C to 10°C can slow leaf growth and close stomata. Within 30 min leaf water potentials decrease and ABA levels increase, presumably in response to a negative hydraulic message arising from enhanced root hydraulic resistance (Dale et al., 1991). However, high leaf ABA slows leaf growth and stomata1 closure is sustained long after leaf water potentials return to normal (4h). Assuming the extra ABA retarded leaf growth and closed stomata, the question of its origin remains. Smith and Dale (1988) measured a doubling of the ABA concentration in bleeding sap from cooled roots. Sap flows were not stated but could be expected to be slower in cooled plants because of greater hydraulic resistance. Hence, lack of dilution is a more likely explanation than extra synthesis by cooled roots. Smith and Dale seriously over-estimated ABA delivery rates to the shoots by multiplying concentrations in slow moving xylem exudates by whole-plant transpiration rates. However, even these over-estimates appeared to be too small to explain the increases in leaf ABA caused by root cooling. A more probable explanation is that the extra leaf ABA is an accumulation message, resulting from reduced export from leaves to growing zones. The halving of leaf ABA concentrations that takes place within 2 h of re-warming roots is also compatible with this interpretation (Dale er al., 1991). Similarly, raising of leaf ABA by excising roots (Dale et al., 1991) is compatible with the accumulation message hypothesis. There appears to be a case for believing that a positive message from mechanically impeded roots slows growth and reduces leaf conductances, although the chemical identity of any message is obscure at present. These responses by the shoot can occur in cereal seedlings in advance of losses in leaf water potential, photosynthetic rate and phosphorus uptake (Masle and Passioura, 1987), indicating a non-hydraulic message. Passioura and Gardner (1990) grew wheat in drying soil that was either compacted or uncompacted and maintained leaf turgors at normal levels by pressurizing the roots. Despite this pressure treatment, leaf extension was slowed by soil drying, the effect beginning at higher water contents in the compact soil. This suggests that shoots were responding non-hydraulically to a mechanical constraint on root growth imposed by the more compacted soil. Phosphorus shortage, as a negative message, was ruled out by supplying additional phosphorus, which failed to relieve the effects of impedance. The probability that ABA is the message passing from mechanically impeded roots is not strong, since such roots have not been found to contain abnormally large amounts of the hormone (Lachno, 1983; Moss et al., 1988). Where larger amounts of ABA have been found in mechanically impeded plants, it has been in association with water deficiency either in the shoot (Stypa et al.,

166

M. B . JACKSON

1987) or in the root (Tardieu el al., 1992a,c). The previously mentioned findings of Masle and Passioura (1987) and Passioura and Gardner (1990) contrast uncomfortably with those of Goss and Russell (1980) and Iijima and Kono (1992), which show little or no decreases in shoot growth when roots of Hordeum vulgare o r Zea mays are stunted by mechanical impedance under conditions of good aeration and nutrient availability. F. CONCLUSIONS

Evidence that apolastic ABA can regulate stomatal apertures appears strong. Similarly, it is clear that leaves of droughted or flooded plants can close their stomata in association with increases in ABA while remaining fully turgid. The nature of the message linking this behaviour with the stressed roots is still not entirely clear. For droughted plants there is a wealth of evidence linking stomatal closure (and slow leaf expansion) with increases in ABA concentrations in xylem sap. However, the extent to which this indicates increased output by roots is sometimes clouded by (1) dilution changes resulting from slower transpiration; (2) output of ABA from older leaves suffering from water deficits; and (3) the possibility of as yet unidentified inhibitors of stomatal opening or of leaf growth being present in xylem sap. There is no strong evidence for a positive ABA message from the roots of plants stressed by mineral shortage, mechanical impedance or cold. More emphasis needs to be placed on quantifying hormone delivery rates, on addressing the involvement of accumulative messages, and on quantifying the extent to which messages cycle between roots and shoots.

IX. FINAL REMARKS A large number of well-executed experiments indicate close operational links between roots and shoots. They lead, plausibly, to the conclusion that roots influence shoot developent in ways other than through mineral nutrition and hydration control, although there is little evidence that roots regulate closely the root:shoot dry weight ratio. The marked effects of newly initiated roots on lateral bud growth, shoot extension and photosynthesis are especially convincing, as is the ability of roots to maintain non-senescent foliage, particularly in slowly growing mineral-deficient plants. The promoting, or inhibiting, influence of roots on flowering also illustrates the remarkable power of roots to guide shoot development. In addition, there is much evidence of rapid and marked effects of environmentally stressed roots on shoot behaviour that appear to be independent of mineral or water supply. Such evidence leads, inevitably, to the possibility that hormone traffic between roots and shoots forms the mechanistic basis for the morphological and functional control that roots can exert on shoots.

PLANT HORMONES AND COMMUNICATION

167

Unfortunately, many of the experiments designed to reveal such a functional link are not persuasive. One problem is the surprising lack of care in quantifying, realistically, the amounts of hormone transferred from roots to shoots. Thus, the question posed by the title of the review cannot be answered with an unqualified “yes”! In looking for hormones moving into shoots from the roots, many authors report only concentrations in sap obtained from de-topped root systems or in other ways. By themselves, such data say little about the actual amounts moving in the intact plant because concentrations are not those extant in the intact plant, and are in any case subject to variation that is independent of hormone flux. Information concerning the relative amounts of root and shoot involved in generating or receiving a putative hormone message is also lacking, although highly desirable for the sound interpretation of results. An accurate estimate of hormone flux is a prerequisite for dependable experimental testing of the physiological impact of a putative hormone message. There is evidence that some hormonal messages may be other than one of the five principal hormones (e.g. Satoh et al., 1992). This possibility would bear closer examination. A great deal of the research has concentrated on the possible role of cytokinins from roots in regulating a wide range of shoot processes, especially leaf senescence in plants with stressed root systems. Although basic physiology strongly suggests that roots have such effects independently of any regulation through minerals and water, the case for believing that root cytokinins are involved remains surprisingly weak, despite an extremely large literature. Support for the view that the output of GASfrom roots is of physiological significance for the shoot is also weak. The most convincing evidence for hormonal regulation of shoots by roots comes from studies of plants suffering from soil flooding or soil drying. An especially convincing case exists for believing that a positive message, in the form of increased ACC transport from oxygen-deficient roots to well-aerated shoots, raises shoot ethylene production to physiologically active levels. However, it is unfortunate that most of the research has been limited to only one species. There is also convincing evidence that exposing roots to drying soil can raise ABA in the leaf apoplast sufficiently to close stomata and slow leaf expansion. That some of this ABA is generated by the roots themselves appears probable, although substantial increases in delivery from drying roots into the shoot have yet to be demonstrated unequivocally. There has been only limited recognition of the role of roots as sinks for shoot hormones. By operating in this way, a root system can be expected to regulate shoot hormone levels by controlling rates of export. The impact of the balance between production in the shoots and export to the roots merits more attention. In redressing the balance, more account could usefully be taken of the extent to which hormones are circulated between roots and shoots in both phloem and xylem transport pathways. Future studies are

168

M. B. JACKSON

likely to be increasingly quantitative, taking advantage of modern physicochemical and immunological methods, and of computer-based modelling methods to encapsulate the hormone economy of the whole plant when particular changes in development are taking place.

ACKNOWLEDGEMENTS I thank D r G. V. Hoad and Mr M. A . Else for critically reading early drafts of the manuscript, Mrs J. M. Llewellyn for typing and Mr N. L. Smith for help with the figures.

REFERENCES Adams, D. 0. and Yang, S. F. (1979) Ethylene biosynthesis: identification of 1-aminocyclopropane-1-carboxylicacid as an intermediate in the conversion of methionine to ethylene. Proceedings of the National Academy of Sciences, USA 76, 170-174. Addicott, F. T., Carns, H . R., Cornforth, J. W., Lyon, J. L., Milborrow, B. V. Ohkuma, K . , Ryback, G., Smith, 0. E., Thiessen, W. E . and Wareing, P. F. (1968). Abscisic acid: a proposal for the re-designation of abscisin I1 (dormin). In “Biochemistry and Physiology of Plant Growth Substances” (F. Wightman and G . Setterfield, eds), pp. 1527-1530. Runge Press, Ottawa. Aharoni, N . , Blumenfeld, A. and Richmond, A. E. (1977). Hormonal activity in detached lettuce leaves as affected by leaf water content. Plant Physiology 59, 1169-1173. Ahokas, H. (1984). Cytokinins in the spring sap of curly birch (Betula pendula f. carelica) and the non-curly form. Journal of Plant Physiology 118, 33-39. Aston, M. J. and Jones, M. M. (1976). A study of the transpiration surfaces o f A v e n a sterilis L. var. Algerian leaves using monosilisic acid as a tracer for water movement. Planta 130, 121-129. Atkinson, C. J., Davies, W. J. and Mansfield, T. A. (1989). Changes in stomata1 conductance in intact ageing wheat leaves in response to abscisic acid. Journal of Experimental Botany 40, 1021-1028. Baier, M. and Hartung, W. (1991). Movement of abscisic acid across the plasma membrane of phloem elements of Plantago major. Journal of Plant Physiology 137,297-300. Barr, M. L. (1973). Biosynthesis of abscisic acid: incorporation of radioactivity from 2-C mevalonic acid by intact roots. Plant Physiology 51 (Suppl.), 47. Bassi, P. K . and Spencer, M. S. (1979). A cuvette design for the measurement of ethylene production and carbon dioxide exchange by intact shoots under controlled environmental conditions. Plant Physiology 64, 488-490. Bates, L. M. and Hall, A. E. (1981). Stomata1 closure with soil water depletion is not associated with changes in bulk leaf water status. Oecologia 50, 62-65. Beever, J . E. and Woolhouse, H. W. (1973). Increased cytokinin from root system of Perilla fructescens and flower and fruit development. Nature New Biology 46, 31-32. Benson, R. J . , Boyer, J. S. and Mullet, J. E. (1988). Water deficit-induced changes in abscisic acid, growth, polysomes, and translatable RNA in soybean hypocotyls. Plant Physiology 88, 289-294.

PLANT HORMONES AND COMMUNICATION

169

Bernier, G. and Kinet, J. M. (1986). The control of flower initiation and development. In “Plant Growth Substances 1985”. (M. Bopp, ed.), pp. 293-302. Springer-Verlag, Berlin. Beyer, E. M. Jr., Morgan, P. W. and Yang, S. F. (1984). Ethylene. In “Advanced Plant Physiology” (M. B. Wilkins, ed.), pp. 111-126. Pitman, London. Biddington, N. L. and Dearman, A. S. (1984). Shoot and root growth of lettuce seedlings following root pruning. Annals of Botany 53,663-668. Blackman, P. G . and Davies, W. J. (1985). Root to shoot communication in maize plants of the effects of soil drying. Journal of Experimental Botany 36, 39-48. Blaikie, S. J. and Mason, W. K. (1990). Correlation of growth of the root and shoot systems of white clover after a period of water shortage and or defoliation. Australian Journal of Agricultural Research 41, 891-900. Blum, A. and Johnson, J. W. (1992). Transfer of water from roots into dry soil and the effects on wheat water relations. Plant and Soil 145, 141-150. Blum, A., Johnson, J. W., Ramser, E. L. and Tollner, E. W. (1991). The effect of a drying top soil and a possible non-hydraulic root signal on wheat growth and yield. Journal of Experimental Botany 42, 1225-1232. Bollmark, M., Kubat, B. and Eliasson, L. (1985). The role of cytokinins in the regulation of adventitious root formation. Physiologia Plantarum 64 (abstr. 97), 19a. Bottger, M. (1978). Levels of endogenous indole-3-acetic acid and abscisic acid during the course of the formation of lateral roots. Zeitschrift fur v a n zenphysiologie 86, 283-286. Bradford, K. J. (1982). Regulation of shoot responses to root stress by ethylene, abscisic acid, and cytokinin. In “Plant Growth Substances 1982” (P. F. Wareing, ed.), pp. 599408. Academic Press, London. Bradford, K. J. (1983). Involvement of plant growth substances in the alteration of leaf gas exchange of flooded tomato plants. Plant Physiology 73,48@483. Bradford, K. J. and Dilley, D. R. (1978). Effects of root anaerobiosis on ethylene production, epinasty and growth of tomato plants. Plant Physiology 61, 5 0 6 509. Bradford, K. J. and Hsiao, T. C. (1982). Stomata1 behaviour and water relations of waterlogged tomato plants. Plant Physiology 70, 150&1513. Bradford, K. J. and Yang, S. F. (1980a). Xylem transport of l-aminocyclopropane-lcarboxylic acid, an ethylene precursor, in waterlogged plants. Plant Physiology 65,322-326. Bradford, K. J. and Yang, S. F. (1980b). Stress-induced production in the ethylenerequiring tomato mutant diageotropica. Plant Physiology 65, 327-330. Bradford, K. J., Hsiao, T. C . and Yang, S. F. (1982). Inhibition of ethylene synthesis in tomato plants subjected to anaerobic root stress. Plant Physiology 70, 1503-1507. Burrows, W. J. and Carr, D. J. (1969). Effects of flooding the root system of sunflower plants on the cytokinin content of the xylem sap. Physiologia Plantarum 22, 1105-1112. Butcher, D. N. (1963). The presence of gibberellins in excised tomato roots. Journal of Experimental Botany 14, 272-280. Butcher, D. N., Appleford, N. E. J . , Hedden, P. and Lenton, J. R. (1988). Plant growth substances in root cultures of Lycopersicon esculentum. Phytochemistry 27, 1575-1578. Buttrose, M. S. (1966). The effect of reducing leaf area on the growth of roots, stems and berries of Gordo grapevines. Vitis 5,455-464. Buttrose, M. S. and Mullins, M. G. (1968). Proportional reduction in shoot growth

170

M. B. JACKSON

of grape vines with root systems maintained at constant relative volumes by repeat pruning. Australian Journal of Biological Sciences 21, 1095-1101. Cahill, D. M., Weste, G . M. and Grant, B. R. Changes in cytokinin concentrations in xylem extrudate following infection of Eucalyptus marginata Donn ex Sm with Phytophthora cinnamomi. Plant Physiology 81, 1103-1 109. Cannell, R. Q. and Jackson, M . B. (1981). Alleviating aeration stresses. In “Modifying the Plant Environment to Reduce Crop Stress” ( G . Arkin and H. M. Taylor, eds), pp. 141-192. American Society of Agricultural Engineers, St Joseph, MO. Canny, M. J . and McCully, M . E. (1988). The xylem sap of maize roots: its collection, composition and formation. Australian Journal of Plant Physiology 15,557-566. Carlson, D. R., Dyer, D. J., Cotterman, D. C. and Durley, R. C. (1987). The physiological basis for cytokinin induced increases in pod set in 1x93-100 soybeans. Plant Physiology 84, 233-239. Carlson, W. C. and Larson, M. M. (1977). Changes in auxin and cytokinin activity in roots of red oak (Quercus rubra) seedlings during lateral root formation. Physiologia Plantarum 41, 162-166. Carmi, A . (1986). Effects of cytokinins and root pruning on photosynthesis and growth. Photosynthetica 20, 1-8. Carmi, A. and Heuer, B. (1981). The role of roots in control of bean shoot growth. Annals of Botany 48, 519-527. Carmi, A. and Koller, D. (1978). Effects of the roots on the rate of photosynthesis in primary leaves of bean (Phaseolus vulgaris L.). Photosynthetica 12, 178184. Carmi, A. and Koller, D. (1979). Regulation of photosynthesis activity in the primary leaves of bean (Phaseolus vulgaris L.) by materials moving in the water conducting system. Plant Physiology 64, 285-288. Carmi, A. and van Staden, J. (1983). Role of roots in regulating the growth rate and cytokinin content in leaves. Plant Physiology 73, 7 6 7 8 . Carr, D. J. and Reid, D. M. (1969). The physiological significance of the synthesis of hormones in roots and of their export to the shoot system. In “Biochemistry and Physiology of Plant Growth Substances” (F. Wightman and G. Setterfield, eds), pp. 1169-1185. Runge Press, Ottawa. Carr, D. J., Reid, D. M. and Skene, K. G. M . (1964). The supply of gibberellins from the root to the shoot. Planta 63, 382-392. Chailakhian, M. Kh. (1961). Principles of ontogenesis and physiology of flowering in higher plants. Canadian Journal of Botany 39, 1817-1841. Chapin, 11, F. S. (1990). Effects of nutrient deficiencyon plant growth: evidence for a centralized stress-response system. In “Importance of Root to Shoot Communication in the Responses to Environmental Stress” (W. J. Davies and B. Jeffcoat, eds), pp. 135-148. British Society for Plant Growth Regulation, Bristol. Chapin, 11, F. S., Walter, C. H. S. and Clarkson, D. T. (1988). Growth response of barley and tomato to nitrogen stress, and its control by abscisic acid, water relations and photosynthesis. Planta 173, 352-366. Chen, C.-M. and Petschow, B. (1978). Cytokinin biosynthesis in cultured, rootless tobacco plants. Plant Physiology 62,861-865. Chibnall, A . C. (1939). “Protein Metabolism in the Plant”. Yale University Press, New Haven (p. 266 in reprinted edition, 1964). Chibnall, A. C. (1954). Protein metabolism in rooted runner-bean leaves. The New Phytologist 53, 31-37. Chibnall, A . C. and Wiltshire, G. H. (1954). A study with isotopic nitrogen of

PLANT HORMONES AND COMMUNICATION

171

protein metabolism in detached runner-bean leaves. The New Phytologist 53, 38-43. Cohen, D. B., Dumbroff, E. B. and Webb, D. P. (1978). Seasonal patterns of abscisic acid in roots of Acer saccharum. Plant Science Letters 11, 35-39. Coleman, M. D., Bledsoe, C. S. and Smit, B. A. (1990). Root hydraulicconductivity and xylem sap levels of zeatin riboside and abscisic acid in ectomycorrhizal douglas fir seedlings. The New Phytologist 115, 275-284. Cornish, K. and Zeevaart, J. A. D. (1985a). Abscisic acid accumulation by roots of Xanthium strumarium L. and Lycopersicon esculentum Mill. in relation to water stress. Plant Physiology 79, 653-658. Cornish, K. and Zeevaart, J. A . D. (1985b). Movement of abscisic acid into the apoplast in response to water stress in Xanthium strumarium L. Plant Physiology 78, 623-626. Coutts, M. P. (1981). Effects of waterlogging on water relations of actively growing and dormant Sitka spruce seedlings. Annals of Botany 47, 747-753. Creelman, R. A., Gage, D. A., Stults, J. T. and Zeevaart, J. A. D. (1987). Abscisic acid biosynthesis in leaves and roots of Xanthium strumarium. Plant Physiology 85, 726-732. Crozier, A. and Reid, D. M. (1971). Do roots synthesize gibberellins? Canadian Journal of Botany 49,967-975. Dale, J. E. (1988). The control of leaf expansion. Annual Review of Plant Physiology and Plant Molecular Biology 39, 267-295. Dale, J. E., Stacciarini, E. and Sattin, M. (1990). The effects of root cooling on leaf growth. In “Importance of Root to Shoot Communication in the Responses to Environmental Stress” (W. J. Davies and B. Jeffcoat, eds), pp. 149-161. British Society for Plant Growth Regulation, Bristol. Darrall, N. M. and Wareing, P. F. (1981). The effect of nitrogen nutrition on cytokinin activity and free amino acids in Betula pendula Roth. and Acer pseudoplatanus L. Journal of Experimental Botany 32,369-379. Das Gupta, D. K. (1972). Developmental physiology of sugar-beet. 111. Effects of decapitation, defoliation, and removing part of the root and shoot on subsequent growth of sugar beet. Journal of Experimental Botany 23,93-102. Dathe, W . , Sembdner, G . , Yamaguchi, I . and Takahashi, N. (1982). Gibberellins and growth inhibitors in spring bleeding sap, roots and branches of Juglans regia L. Plant and Cell Physiology 23, 115-123. Davey, J. E. and van Staden, J. (1976). Cytokinin translocation: changes in zeatin and zeatin-riboside levels in root exudate of tomato plants during their development. Planta 130, 6S72. Davidson, R. L. (1969). Effect of root/leaf temperature differentials on root/shoot ratios of some pasture grasses and clover. Annals of Botany 33, 561-569. Davies, W. J . and Jeffcoat, B. (eds) (1990). “Importance of Root to Shoot Communication in the Responses to Environmental Stress”. British Society for Plant Growth Regulation, Bristol. Davies, W. J., Metcalf, J., Lodge, T. A. and da Costa, A. R. (1986). Plant growth substances and the regulation of growth under drought. Australian Journal of Plant Physiology 13, 105-125. Davison, R. M. (1965). Some properties of a plant growth inhibitor in xylem sap of woody species. Australian Journal of Biological Science 18, 475486. Davison, R. M. and Young, H. (1973). Abscisic-acid content of xylem sap. Planta 109,95-98. De Ropp, R. S. (1946). Studies in the physiology of leaf growth. 11. The influence of roots on the growth of leaves and stems in rye. Annals of Botany 10,353-359. Drennan, D. S. H. and El Hiweris, S. 0. (1979). Changes in growth regulating

172

M. B. JACKSON

substances in Sorghum vulgare infected by Striga hermonthica. I n “Proceedings of the Second Symposium on Parasitic Weeds” (Musselman, L. J., Worsham, A. D. and Eplee, R. E., eds), pp. 144-155. North Carolina State University, Raleigh, NC. Drew, M. C., Sisworow, E. J. and Saker, L. R. (1979). Alleviation of waterlogging damage to young barley plants by application of nitrate and a synthetic cytokinin, and comparison between the effects of waterlogging, nitrogen deficiency and root excision. The New Phytologist 82,301-314. Duckham, S. C., Linforth, R. S. T. and Taylor, I. B. (1990). ABA synthesis in roots and shoots. I n “Importance of Root to Shoot Communication in the Responses to Environmental Stress” (W. J. Davies and B. Jeffcoat, eds), pp. 328-331. British Society for Plant Growth Regulation, Bristol. Eklund, L. (1993). Ethylene dynamics in Picea abies. Plant Growth Regulation (in press). El-Beltagy, A. S. and Hall, M. A. (1974). Effect of water stress upon endogenous ethylene levels in Vicia faba. The New Phytologist 73, 47-60. Else, M. A . , Davies, W. J . and Jackson, M. B. (1991). The deliveryofhormones and other solutes from roots to shoots of tomato plants: contamination of xylem sap in early samples from de-topped plants. Journal of Experimental Botany 42 (Suppl.; Abstr. P2.41), 16. Engelbrecht, L. (1972). Cytokinins in leaf-cuttings of Phaseolus vulgaris L. during their development. Biochemie und Physiologie der Pflanzen 163, 335343. Farrar, S. C. and Farrar, J . F. (1985). Fluxes of carbon compounds in leaves and roots of barley plants. In “Regulation of Sources and Sinks in Crop Plants”, Monograph No. 12 (B. Jeffcoat, A. Hawkins and A. D. Stead, eds), pp. 67-84. British Plant Growth Regulator Group, Bristol. Faull, K. F., Coombe, B. G. and Paleg, L. G. (1974). Extraction and characterization of gibberellins from Hordeum vulgare L., seedlings. Australian Journal of Plant Physiology 1 , 183-198. Ferguson, A. R. (1980). Xylem sap from Actinidia chinensis: apparent differences in sap composition arising from the method of collection. Annals of Botany 46, 791-801. Frydman, V. M. and Wareing, P. F. (1973). Phase change in Hedera helix L. 11. The possible role of roots as a source of shoot gibberellin-like substances. Journal of Experimental Botany 24, 1139-1148. Fubeder, A., Wagner, B. and Breck, E. (1988). Quantification by ELISA of cytokinins in root-pressure exudates of Urtica dioica plants grown under different nitrogen levels. Botanica Acta 101, 214-219. Galston, A. W. (1948). On the physiology of root initiation in excised asparagus stem tips. American Journal of Botany 35,281-287. Gollan, T., Passioura, J. B. and Munns, R. (1986). Soil water status effects the stomata1 conductance of fully turgid wheat and sunflower leaves. Australian Journal of Plant Physiology 13,459-464. Goring, H. and Mardanov, A. A. (1976). Influence of nitrogen deficiency on K/Ca ratio and cytokinin content of pumpkin seedlings. Biochemie und Physiologie der Pflanzen 170,261-264. Goss, M. J. and Russell, R. S. (1980). Effects of mechanical impedance on root growth in barley (Hordeum vulgare L.). 111. Observations on the mechanism of the response. Journal of Experimental Botany 31, 577-588. Gowing, D. J. G., Davies, W. J. and Jones, H. G. (1990). A positive signal as an indicator of soil drying in apple, Malus x dornestica Borkh. Journal of Experimental Botany 41, 1535-1540.

PLANT HORMONES AND COMMUNICATION

173

Griggs, P., Stuchbury, T. and Wang, T. L. (1988). Dihydrozeatin riboside from Phaseolus sap; quantification by radioimmunoassay and GC-MS. Phytochemistry 27, 1583-1588. Griffaut, B. (1977). Mise en evidence d’une syntheses de cytokinines dans des racines isolees de Lupin (Lupinus albus L.) cultivees in vitro. Compte Rendus Academie des Sciences, Skrie D, Paris 225, 331-334. Hall, K. C. and Jackson, M. B. (1986). Abscisic acid in straw from autumn-sown wheat. Journal of the Science of Food and Agriculture 37, 219-222. Hall, K. C., Else, M. A. and Jackson, M. B. (1993). Determination of 1aminocyclopropane-1-carboxylicacid (ACC) by gas chromatography using a nitrogen detector. Plant Growth Regulation (in press). Hall, P. J., Badenoch-Jones, J., Parker, C. W., Letham, D. S. and Barlow, B. A. (1987). Identification and quantification of cytokinins in the xylem sap of mistletoes and their hosts in relation to leaf mimicry. Australian Journal of Plant Physiology 14, 429438. Hall, S. M. and Medlow, G. C. (1974). Identification of IAA in phloem and root pressure saps of Ricinus communis L. by mass spectrometry. Planta 119, 257-261. Hameed, M. A . , Reid, J. B. and Rowe, R. N. (1987). Root confinement and its effect on the water relations, growth and assimilate partitioning of tomato (Lycopersicon esculentum Mill.). Annals of Botany 59, 685-692. Harper, J. L. (1977). “Population Biology of Plants”. Academic Press, London. Hartung, W. (1983). The site of action of abscisic acid at the guard cell plasmalemma of Valerianella locusta. Plant, Cell and Environment 6, 427-428. Hartung, W. and Abou-Mandour, A. (1980). Abscisic acid in root cultures of Phaseolus coccineus L. Zeitschrift fur PJanzenphysiologie 97,65-69. Hartung, W. and Baier, M. (1990). Content and transport of abscisic acid (ABA) in guard cells. Biochemie und Physiologie der PJanzen 186,341-349. Hartung, W. and Slovik, S. (1991). Tansley review No. 35. Physicochemical properties of plant growth regulators and plant tissue determine their distribution and redistribution: stomatal regulation by abscisic acid. The New Phytologist 119, 361-382. Hartung, W., Radin, J. W. and Hendrix, D. L. (1988). Abscisicacidmovement into the apoplastic solution of water-stressed cotton leaves. Plant Physiology 86, 908-913. Hartung, W., Heilmeier, Wartinger, A., Kettemann, I . and Schulze, E.-D. (1990). Ionic content and abscisic acid relations of Anastatica hierochuntica L. under arid conditions. Israel Journal of Botany 39, 373-382. Heindle, J. C., Carlson, D. R., Brun, W. A. and Brenner, M. L. (1982) Ontogenetic variation of four cytokinins in soybean root pressure exudate. Plant Physiology 70, 1619-1625. Henson, I. E. (1984). Evidence for a role for abscisic acid in mediating stornatal closure by destructing translocation from leaves of pearl millet (Pennistum americanum [L.] Leeke). Journal of Experimental Botany 35, 1419-1432. Henson, I. E. and Wareing, P. F. (1976). Cytokinins in Xanthium strumarium L.: distribution in the plant and production in the root system. Journal of Experimental Botany 27, 1268-1278. Herold, A. (1980). Regulation of photosynthesis by sink activity-the missing link. The New Phytologist 86, 131-144. Hiron, R. W. P. and Wright, S. T. C. (1973). The role of endogenous abscisic acid in the response of plants to stress. Journal of Experimental Botany 24,769-781. Hoad, G . V. (1975). Effect of osmotic stress on abscisic acid levels in the xylem sap of sunflower (Helianthus annuus L.). Planta 124, 5-29.

174

M. B . JACKSON

Hoad, G. V. (1978). Effect of water stress on abscisic acid levels in white lupin (Lupinus albus) fruits, leaves and phloem exudate. Planta 142,287-290. Hoad, G. V. and Bowen, M. R. (1968). Evidence for gibberellin-like substances in phloem exudate of higher plants. Planta 82, 22-32. Hocking, T. J., Hillman, J. R. and Wilkins, M. B. (1972). Movement of abscisic acid in Phaseolus vulgaris plants. Nature 235, 124-125. Holm, R. E. and Key, J. L. (1969). Hormonal regulation of cell elongation in the hypocotyl of rootless soybean: an evaluation of the role of DNA synthesis. Plant Physiology 44, 1295-1301. Horgan R. (1984). Cytokinins. In “Advanced Plant Physiology” (M. B. Wilkins, ed.), pp. 53-75. Pitman, London. Horgan, J. M. and Wareing, P. F. (1980). Cytokinins and the growth responses of seedlings of Betulapendula Roth. and Acerpseudoplatanus L. to nitrogen and phosphorous deficiency. Journal of Experimental Botany 31,525-532. Horgan, R., Hewett, E. W . , Purse, J. G., Horgan, J. M. and Wareing, P. F. (1973). Identification of a cytokinin in sycamore sap by gas chromatography-mass spectrometry. Plant Science Letters 1, 321-324. Humphries, E. C. (1958). Effects of removal of part of the root system on subsequent growth of the root and shoot. Annals of Botany 22,251-257. Humphries, E. C. (1959). Effects of mutilation of the root on subsequent growth. Scientific Horticulture 14, 42-48. Humphries, E. C. (1963). Dependence of net assimilation rate on root growth of isolated leaves. Annals of Botany 27, 175-183. Humphries, E. C. and Thorne, G . N. (1964). The effect of root formation on photosynthesis of detached leaves. Annals of Botany 28, 391400. Hunt, R. C. (1975). Further observations on root-shoot equilibria in perennial rye-grass (Lolium perenne). Annals of Botany 39, 745-755. Iijima, M. and Kono, Y. (1992) Development of golgi apparatus in the root cap cells of maize (Zea mays L.) as affected by soil compaction. Annals of Botany 70, 207-212. Incoll, L. D. and Jewer, P. C. (1987). Cytokinins and the water relations of whole plants. In “Cytokinins: Plant Hormones in Search of a Role?” (R. Horgan and B. Jeffcoat, eds), pp. 85-97. British Plant Growth Regulator Group, Bristol. Incoll, L. D., Ray, J. P. and Jewer, P. C. (1990). Do cytokinins act as root to shoot signals? In “Importance of Root to Shoot Communication in the Responses to Environmental Stress’’, Monograph No. 21, (W. J. Davies and B. Jeffcoat, eds) pp. 185-199. British Society for Plant Growth Regulation, Bristol. Ingestad, T. (1982). Relative addition rate and external concentration; driving variables used in plant nutrition research. Plant, Cell and Environment 5, 443-453. Ingestad, T. and Lund, A.-B. (1979). Nitrogen stress in birch seedlings. I. Growth technique and growth. Physiologia Plantarum 45, 137-148. Ingestad, T. and Lund, A.-B. (1986). Theory and techniques for steady state mineral nutrition and growth of plants. Scandinavian Journal of Forestry Research 1, 439-443. Ingram, T. J., Reid, J. B., Murfet, I . C., Gaskin, P., Willis, C. L. and MacMillan, J. (1984). Internode length in Pisum. The le-gene controls the 3P-hydroxylation of gibberellin to gibberellin Al and internode growth in Pisum sativum L. Planta 160, 455-463. Itai, C. and Vaadia, Y. (1965). Kinetin-like activity in root exudate of water-stressed sunflower plants. Physiologia Plantarum 18, 941-944.

PLANT HORMONES AND COMMUNICATION

175

Itai, C. and Vaadia, Y. (1971). Cytokinin activity i n water-stressed shoots. Plant Physiology 47, 87-90. Itai, C., Richmond, A. and Vaadia, Y . (1968). The role of root cytokinins during water and salinity stress. Israel Journal of Botany 17, 187-195. Itai, C., Ben-Zioni, A. and Ordin, L. (1973). Correlative changes in endogenous hormone levels and shoot growth induced by short heat treatments to the root. Physiologia Plantarum 29, 355-360. Jackson, M. B. (1979). Is the diageotropic tomato ethylene deficient? Physiologia Plantarurn 46, 347-351. Jackson, M. B. (1980). Aeration in the nutrient film technique of glasshouse crop production, and the importance of oxygen, ethylene and carbon dioxide. Acta Horticulturae 98, 61-78. Jackson, M. B. (1981). Morphological consequences for flooded, oxygen-deficient plants of transport processes between roots and shoots. In “Mechanisms of Assimilate Distribution and Plant Growth Regulators” (J. Kralovic, ed.), pp. 16(&182. Slovak Society of Agriculture, Food and Forestry, Slovak Academy of Science, Bratislava. Jackson, M. B. (198%). Ethylene and plant responses to soil waterlogging and submergence. Annual Review of Plant Physiology 36, 145-174. Jackson M. B. (198%). Responses of leafed and leafless peas to soil waterlogging. In “The Pea Crop. A Basis for Improvement” (P. B. Hebblethwaite, M. C. Heath and T. C. Dawkins, eds), pp. 163-172. Butterworths, London. Jackson, M. B. (1985~).Ethylene and the responses of plants to excess water in their environment. In “Ethylene and Plant Development” (J. A. Roberts and G. A. Tucker, eds), pp. 241-265. Butterworths, London. Jackson, M. B. (1987). A structured evaluation of the involvement of ethylene and abscisic acid in plant responses to aeration stress. In “Hormone Action in Plant Development-A Critical Appraisal” (G. V. Hoad, J. R. Lenton, M. B. Jackson and R. Atkin, eds), pp. 189-199. Butterworths, London. Jackson, M. B . (1990). Communication between the roots and shoots of flooded plants. In “Importance of Root to Shoot Communication in the Responses to Environmental Stress”. Monograph No. 21 (W. J . Davies and B. Jeffcoat, eds), pp. 115-133. British Society for Plant Growth Regulation, Bristol. Jackson, M. B. (1991). Regulation of water relationships in flooded plants by ABA from leaves, roots and xylem sap. In “Abscisic Acid, Physiology and Biochemistry” (W. J. Davies and H. G. Jones, eds), pp. 217-226. Bios, Oxford. Jackson, M. B. and Campbell, D. J. (1975a). Movement of ethylene from roots to shoots, a factor in the responses of tomato plants to waterlogged soil conditions. The New Phytologist 74, 397-406. Jackson, M. B. and Campbell, D. J. (197%). Ethylene and waterlogging effects in tomato. Annals of Applied Biology 81, 102-105. Jackson, M. B. and Campbell, D. J. (1976a). Waterlogging and petiole epinasty in tomato: the role of ethylene and low oxygen. The New Phytologist 76,21-29. Jackson, M. B. and Campbell, D. J. (1976b). Production of ethylene by excised segments of plant tissue prior to the effect of wounding. Planta 129, 273274. Jackson, M. B. and Campbell, D. J. (1979). Effects of benzyladenine and gibberellic acid on the responses of tomato plants to anaerobic root environments and to ethylene. The New Phytologist 82,331-340. Jackson, M. B. and Drew, M. C. (1984). Effects of flooding on growth and metabolism of herbaceous plants. In “Flooding and Plant Growth” (T. T. Kozlowski, ed.), pp. 47-128. Academic Press, Orlando, FL. Jackson, M. B. and Hall, K. C. (1987). Early stomata1 closure in waterlogged pea

176

M. B. JACKSON

plants is mediated by abscisic acid in the absence of foliar water deficits. Plant, Cell and Environment 10, 121-130. Jackson, M. B. and Kowalewska, A. K. B. (1983). Positive and negative messages from roots induce foliar desiccation and stomata1 closure in flooded pea plants. Journal of Experimental Botany 34, 493-506. Jackson, M. B., Gales, K. and Campbell, D. J. (1978). Effect of waterlogged soil conditions on the production of ethylene and on water relationships. Journal of Experimental Botany 29, 183-193. Jackson, M. B., Young, S . F. and Hall, K. C. (1988). Are roots a source of abscisic acid for the shoots of flooded plants? Journal of Experimental Botany 39, 1631-1637. Jacobs, W. P. (1959). What substance normally controls a given biological process? 1. Formulation of some rules. Developmental Biology 1, 527-533. James, D. B. and Hutto, J. M. (1972). Effects of tiller separation and root pruning on the growth of Lolium perenne L. Annals of Botany 36, 485-495. Jesko, T. (1972). Removal of all nodal roots initiating the extension growth in Sorghum saccharatum (L.) Moench. I. Effect of photosynthetic rate and dark respiration. Photosynthetica 6, 51-56. Jesko, T. (1981). Inter-organ control of photosynthesis mediated by emerging nodal roots in young maize plants. In “Structure and Function of Plant Roots” (R. Brouwer, 0. Gasparikova, J. Kolek and B. C. Loughman, eds), pp. 367-371. Martinus Nijhoff/Dr W. Junk, The Hague. Jesko, T. and Vizarova, G. (1980). Changes of free endogenous cytokinins during transitorily increased photosynthetic rate initiated by formation of the first two whorls of nodal roots in Zea mays. Photosynthetica 14, 83-85. Jesko, T., Heinrichova, K. and Lukacovic, A. (1971). Increase in photosynthetic activity during the formation of the first node roots and first tiller in Sorghum saccharatum (L.) Moench. Photosynthetica 5,233-240. Jones, 0. P. (1973). Effects of cytokinins in xylem sap from apple trees on apple shoot growth. Journal of Horticultural Science 48, 181-188. Jones, R. L. and Phillips, I. D. J. (1966). Organs of gibberellin synthesis in lightgrown sunflower plants. Plant Physiology 41, 1381-1386. Jordan, W. R. and Skoog, F. (1971). Effects of cytokinins on growth and auxin in coleoptiles of de-rooted Avena seedlings. Plant Physiology 48, 97-99. Kaufman, P. B., Ghosheh, N. S . , Nakosteen, L., Pharis, R. P., Durley, R. C. and Morf, W. (1976). Analysis of native gibberellins in the internode, nodes, leaves, and inflorescence of developing Avena plants. Plant Physiology 58, 131-1 34. Kawase, M. (1972). Effects of flooding on ethylene concentration in horticultural plants. Journal of the American Society for Horticultural Science 97,584588. Kende, H. (1964). Preservation of chlorophyll in leaf sections by substances obtained from root exudate. Science 145, 1066-1067. Kende, H. (1965). Kinetin-like factors in root exudate of sunflowers. Proceedings of the National Academy of Sciences, USA 53,1302-1307. Killingbeck, K. T. (1990). Leaf production can be de-coupled from root activity in the desert shrub ocotillo (Fouquieria splendens Engelm.). American Midland Naturalist 124, 124-129. King, R. W. (1976). Implications for plant growth of the transport of regulatory compounds in phloem and xylem. In “Transport and Transfer Processes in Plants” (J. B. Passioura, ed.), pp. 415-443. Academic Press, New York. Koch, Dr (1876). Untersuchungen uber Bacterien. Die Aetiologie der MilzbrandKrankheit, begrundet auf die Entwicklungseschichte des Bacillus anthracis. Beitrage zur Biologie der manzen 2, 277-310.

PLANT HORMONES AND COMMUNICATION

177

Kramer, P. J. and Jackson, W. T. (1954). Causes of injury to flooded tobacco plants. Plant Physiology 29, 241-245. Krauss, A. (1978). Tuberization and abscisic acid content in Solanum tuberosum as affected by nitrogen nutrition. Potato Research 21, 183-193. Krizek, D. T., Carmi, A., Mirecki, R. N., Snyder, F. W. and Bunce, J. A. (1985). Comparative effects of soil moisture stress and restricted root zone volume on morphogenetic and physiological responses of soybean (Glycine max (L) Merr.). Journal of Experimental Botany 36,25-38. Kuang, J. B., Turner, N. C. and Henson, I. E. (1990). Influence of xylem water potential on leaf elongation and osmotic adjustment of wheat and lupin. Journal of Experimental Botany 41, 217-221. Kuiper, D. and Kuiper, P. J. C. (1988). Phenotypic plasticity in physiological perspective. Acta Oecologia (Oecologia Plantarum) 9, 43-59. Kuiper, D. and Staal, M. (1987). The effects of exogenously applied plant growth substances on the physiological plasticity in Plantago major spp. pleiosperma: responses of growth, shoot to root ratio and respiration. Physiologia Plantarum 69,651-658. Kuiper, D., Schuit, J. and Kuiper, P. J. C. (1989). Effects of internal and external cytokinin concentrations on root growth and shoot to root ratio of Plantago major ssp. pleiosperma at different nutrient conditions. In “Structural and Functional Aspects of Transport in Roots” (B. C. Loughman et al., eds), pp. 183-188. Kluwer Academic, Dordrecht. Kuiper, D., Schuit, J. and Kuiper, P. J. C. (1990). Actual cytokinin concentrationsin plant tissue as an indicator for salt resistance in cereals. Plant and Soil 123, 243-250. Kulaeva, 0. N. (1962). The effect of roots on leaf metabolism in relation to the action of kinetin in leaves. Soviet Plant Physiology 9, 182-189. Lachno, D. R. (1983). Abscisic acid and indole-3-acetic acid in maize roots subject to water, salt and mechanical stress. In “Growth Regulators in Root Development” (M. B. Jackson and A. D. Stead, eds), pp. 37-54. British Plant Growth Regulator Group, Wantage. Lavender, D. P., Sweet, G. B., Zaerr, J. B. and Hermann, R. K. (1973). Spring shoot growth in Douglas-fir may be initiated by gibberellins exported from the roots. Science 182,839. Lawrence, D. K. and Strangeways, E. (1985). The role of sucrose in the regulation of photosynthesis in winter wheat. In “Regulation of Sources and Sinks in Crop Plants”, Monograph No. 12. (B. Jeffcoat, A. F. Hawkins and A. D. Stead, eds), pp. 5 1 4 6 . British Plant Growth Regulator Group, Bristol. Lenton, J. R., Bowen, M. R. and Saunders, P. F. (1968). Detection of abscisic acid in the xylem sap of willow (Salix viminalis L.) by gas-liquid chromatography. Nature 220, 8 6 8 7 . Lenton, J. R., Hedden, P. and Gale, M . D. (1987). Gibberellin insensitivity and depletion in wheat4onsequences for development. I n “Hormone Action in Plant Development. A Critical Appraisal” (G. V. Hoad, J. R. Lenton, M. B. Jackson and R. K. Atkin, eds), pp. 145-160. Butterworths, London. Livne, A. and Vaadia, Y. (1965). Stimulation of transpiration rate in barley leaves by kinetin and gibberellic acid. Physiologia Plantarum 18, 658-664. Lockard, R. G. and Grunwald, C (1970). Grafting and gibberellin effects on the growth of tall and dwarf peas. Plant Physiology 45, 160-162. Loeffler, J. E. and van Overbeek, J. (1964). Kinin activity in coconut milk. I n “Regulateurs Naturels de la Croissance Vegetal” (J. P. Nitsch, ed.), pp. 77-82. CRNS, Paris.

178

M. B. JACKSON

Loveys, B. R. (1984a). Abscisic acid transport and metabolism in grape vine (Vitis vinifera). The New Phytologist 98, 575-582. Loveys, B. R. (1984b). Diurnal changes in water relations and abscisic acid in field-grown Vitis vinifera cultivars. 111. The influence of xylem-derived abscisic acid on leaf gas exchange. The New Phytologist 98,563-573. Loveys, B. R. and During, H. (1984). Diurnal changes in water relations and abscisic acid in field-grown Vitis vinifera cultivars. 11. Abscisic acid changes under semi-arid conditions. The New Phytologist 97, 3 7 4 7 . Loveys, B. R. and Robinson, S. P. (1987). Abscisic acid synthesis and metabolism in barley leaves and protoplasts. Plant Science 49, 23-30. Loveys, B. R., Robinson, S. P. and Downton, W. J. S. (1987). Seasonal and diurnal changes in abscisic acid and water relations of apricot leaves (Prunus armenica L.). The New Phytologist 107, 15-27. Luckwill, L. C. and Whyte, P. (1968). Hormones in the xylem sap of apple trees. In “Plant Growth Regulators”, SCI Monograph 31, pp. 87-101. Society of Chemical Industry, London. Ludlow, M. M., Sommer, K . J., Flower, D. J., Ferraris, R. and So, H. B. (1989). Influence of root signals resulting from soil dehydration and high soil strength on growth of crop plants. Current Topics in Plant Biochemistry and Physiology 8, 81-99. Luke, H. H. and Freeman, T. E. (1968). Stimulation of transpiration by cytokinins. Nature 217, 873-874. MacMillan, J. (1987). Gibberellin-deficient mutants of maize and pea and the molecular action of gibberellins. I n “Hormone Action in Plant Development. A Critical Appraisal” (G. V. Hoad, J. R. Lenton, M. B. Jackson and R. K. Atkin, eds), pp. 73-87. Butterworths, London. Made, J. and Passioura, J. B. (1987). The effect of soil strength on the growth of young wheat plants. Australian Journal of Plant Physiology 14, 643-656. Mauk, C. S., Brinker, A. M. and NoodCn, L. D. (1990). Probing monocarpic senescence and pod development through manipulation of cytokinin and mineral supplies in soybean explants. Annals of Botany 66, 191-201. McComb, A. J. and McComb, J. A. (1970). Growth substances and the relationship between phenotype and genotype in Pisum sativum. Planta 91, 235245. McCully, M. E. (1990). Bridges for nutrient re-cycling in the roots of maize. Australian Society of Plant Physiologists Annual Meeting, Combined Abstracts, p. 69. McDonald, A. J. S., Lohammer, T. and Ericsson, A. (1986). Growth responses to step decreases in nutrient availability in small birch (Betula pendula Roth). Plant, Cell and Environment 9, 427432. McGaw, B. A. (1987). Cytokinin biosynthesis and metabolism. In “Plant Hormones and their Role in Plant Growth and Development” (P. J. Davies, ed.), pp. 76-93. Martinus Nijhoff, Dordrecht. McLeod, A. and Neales, T. (1990). Are leaves a source of ABA in the xylem sap of “droughted” sunflower plants? Australian Society of Plant Physiologists Annual General Meeting, Combined Abstracts, Abstr. 158. McMichael, B. L. and Quisenberry, J . E. (1991). Genetic variation for root-shoot relationships among cotton germplasm. Environmental and Experimental Botany 31, 461470. Mees, G. C. and Weatherley, P. E. (1957). The mechanism of water absorption by roots. 11. The role of hydrostatic pressure gradients across the cortex. Proceedings of the Royal Society Series B 14, 381-391. Meidner, H. (1967). The effect of kinetin on stomata1 opening and the rate of intake

PLANT HORMONES AND COMMUNICATION

179

of carbon dioxide in mature primary leaves of barley. Journalof Experimental Botany 18, 556561. Meinzer, F. C., Grantz, D . A. and Smit, B. (1991). Root signals mediate coordination of stomatal and hydraulic conductance in growing sugarcane. Australian Journal of Plant Physiology 18,329-338. Meyer, R. E. and Gingrich, J. H. (1964). Osmoticstress: effectsof itsapplication t o a portion of wheat root systems. Science 144, 1463-1464. Michniewicz, M. and Kriesel, K. (1972). Dynamics of gibberellin-like substances in the development of buds, newly formed shoots and adventitious roots of willow cuttings (Salix viminalis L.). Acta Societutis Botanicorum Poloniae XLI, 301-310. Miginiac, E. and Sotta, B. (1985). Organ correlations affecting flowering in relation to phytohormones. Biologia Plantarum 27, 373-381. Milborrow, B. V. (1984). Intibitors. I n “Advanced Plant Physiology” (M. B. Wilkins, ed.), pp. 76-110. Pitman, London. Milborrow, B. V. and Robinson, D . R . (1973). Factors affecting the biosynthesis of abscisic acid. Journal of Experimental Botany 24, 537-548. Miller, E. C. (1916). The root systems of agricultural plants. Journalof the American Society of Agronomy 8, 129-154. Milligan. S. R. and Dale, J. E. (1988). The effects of root treatmentson growth of the primary leaves of Phaseolus vulgaris L.: General features. The New Phytologist 108, 27-35. Moldau, H. (1973). Effects of various water regimens on stornatal and mesophyll conductances of bean varieties. Photosyntheticu 7, 1-7. Monyo, J. H. and Whittington, W. J. (1970). Genetic analysis of root growth in wheat. Journal of Agricultural Science 74, 329-338. Morgan, P. W., Chuan-Jin, H., D e Greef, J. A . and De Proft, M. P. (1990). Does water deficit stress promote ethylene synthesis by intact plants? Plant Physiology 94, 1616-1624. Moss, G . I . , Hall, K. C. and Jackson, M. B. (1988). Ethylene and the response of roots of maize (Zea mays L.) to physical impedance. The New Phytologist 109, 30S311. Mothes, K. and Engelbrecht, L. (1956). Nitrogen metabolism in leaf cuttings. Flora 143,428-432. Mothes, K. and Engelbrecht, L. (1963). O n the activity of a kinetin-like root factor. Life Sciences 11, 852-857. Mothes, K., Engelbrecht L. and Kulajewa, 0. (1959). Uber die Wirkung des kinetins aut stickstoffverteilung und Eioweissynthese in isolierten Blattern. Flora 147, 445-464. Mullins, M. G . (1967). Morphogenetic effects of roots and some synthetic cytokinins in Vitis vinifera L. Journal of Experimental Botany 18, 206214. Munns, R. (1990). Chemical signals moving from roots to shoots: the case against ABA. I n “Importance of Root to Shoot Communication in the Responses to Environmental Stress” (W. J. Davies and B. Jeffcoat, eds), pp. 175-183. British Society for Plant Growth Regulation, Bristol. Munns, R. and King, R . (1988). Abscisic acid is not the only stornatal inhibitor in the transpiration stream of wheat plants. Plant Physiology 88, 703-708. Munns, R., Passioura, J. B . , Milborrow, B. V., James, R. and Close, T. J . (199 1). A new inhibitor of transpiration. I n “Abstracts of the Australian Society of Plant Physiologists”, Abstr. P75. Musgrave, A , , Jackson, M. B. and Ling, E. (1972). Callitriche stem elongation is controlled by ethylene and gibberellin. Nature New Biology 238, 93-96. Neales, T. F. and Incoll, L. D. (1968). The control of leaf photosynthesis rate by the

180

M. B. JACKSON

level of assimilate concentration in the leaf: a review of the hypothesis. Botanical Review 34, 107-125. Neales, T. F. and McLeod, A. L. (1991). Do leaves contribute to the abscisic acid present in the xylem sap of “droughted” sunflower plants? Plant, Cell and Environment 14, 979-986. Neales, T. F., Masia, A., Zhang, J . and Davies, W. J. (1989). Theeffectsof partially drying part of the root system of Helianthus annuus on the abscisic acid content of the roots, xylem sap and leaves. Journal of Experimental Botany 40, 1113-1120. Neuman, D. S. and Smit, B. A. (1991). The influence of leaf water status and ABA on leaf growth and stomata of Phaseolus seedlings with hypoxic roots. Journal of Experimental Botany 42, 1499-1506. Neuman, D. S., Rood, S. B. and Smit, B. A. (1990). Does cytokinin transport from root-to-shoot in the xylem sap regulate leaf responses to root hypoxia? Journal of Experimental Botany 41, 1325-1333. Nonhebel, H. M., Hillman, J. R., Crozier, A. and Wilkins, M. B. (1985). Metabolism of [14C]indole-3-aceticacid by the cortical and stelar tissues of Zea mays L. shoots. Planta 164, 105-109. NoodCn L. D. (1986). Synergism between gibberellins and cytokinins in delaying leaf senescence in soybean explants. Plant and Cell Physiology 27, 577-579. Nooden, L. D. (1985). Regulation of soybean senescence. In “World Soybean Research Conference 111” (R. Shibles, ed.), pp. 891-899. Westview Press, Boulder and London. Nooden, L. D., Guiamet, J. J., Singh, S . , Letham, D. S . , Tsjui, J. and Schneider, M. J. (1990a). Hormonal control of senescence. In “Plant Growth Substances 1988” (R. P. Pharis and S. B. Rood, eds), pp. 537-546. Springer-Verlag, Berlin. Nooden, L. D., Singh, S. and Letham, D. S. (1990b). Correlation of xylem sap cytokinin levels with monocarpic senescence in soybean. Plant Physiology 93, 3s39. Ong, H. T. (1978). Roles of hormones in the responses of excised tomato cotyledons to mannitol-induced water stress. Biologia Plantarum 20, 318-323. Osborne, D. J . , Jackson, M. B. and Milborrow, B. V. (1972). Physiological properties of abscission accelerator from senescent leaves. Nature New Biology 240, 98-101. Palmer, M. V. and Wong, 0. C. (1985). Identification of cytokinins from xylem exudate of Phaseolus vulgaris L. Plant Physiology 79,296-298. Parker, C. (1984). The influence of Striga species on sorghum under varying nitrogen. In “Proceedings of the 3rd International Symposium on Parasitic Weeds” (C. Parker et al., eds), pp. 90-98. ICARDA, Aleppo, Syria. Parry, A. D. and Horgan, R. (1992). Abscisic acid biosynthesis in roots. I. The identification of potential abscisic acid presursors and other carotenoids. Planta 187, 185-191. Parthier, V. B. (1964). Proteinsynthese in grunen Blattern. 11. AminosaureInkorporation in die Proteine subzellularer Strukturen in abhangigkeit vom physiologischen Zustand der Blatter (Blattalter). Flora 154, 230-244. Passioura, J . B. and Gardner, P. A. (1990). Control of leaf expansion in wheat seedlingsgrowing in drying soil. Australian Journalof Plant Physiology 17,149-157. Passioura, J. and Munns, R. (1984). Hydraulic resistance of plants. 11. Effect of root medium and time of day in barley and lupin. Australian Journal of Plant Physiology 11,341-350. Pereira, J. S. and Kozlowski, T. T. (1977). Variations among angiosperms in response to flooding. Physiologia Plantarum 42, 181-192.

PLANT HORMONES AND COMMUNICATION

181

Peterson, T. A., Reinsel, M. D. and Krizek, D. T. (1991). Tomato (Lycopersicon esculentum Mill., cv. “Better Bush”) plant response to root restriction. Journal of Experimental Botany 42, 1233-1240. Phillips, I. D. J. and Jones, R. L. (1964). Gibberellin-like activity in bleeding sap of root systems of Helianthus annuus detected by a new dwarf pea epicotyl assay and other methods. Planta 63, 265r278. Phinney, B. O . , Freeling, M., Robertson, D. S., Spray, C. R. and Silverthorne, J. (1986). Dwarf mutants of maize-the gibberellin pathway and its molecular future. In “Plant Growth Substances 1985” (M. Bopp, ed.), pp. 55-64. Springer-Verlag, Berlin. Pierce, M. L. and Rashke, K. (1980). Correlation between loss of turgor and accumulation of abscisic acid in detached leaves. Planfa 148, 174-182. Pillay, I. and Beyl, C. (1990). Early responses of drought-resistant and droughtsusceptible tomato plants subjected to water stress. Journal of Plant Growth Regulation 9, 213-219. Pitman, M. G. (1972). Uptake and transport of ions in barley seedlings. 111. Correlations between transport to the shoot and relative growth rate. Australian Journal of Biological Research 25, 905-919. Proebsting, W. M., Davies, P. J. and Marx, G. A . (1978). Photoperiod-induced changes in gibberellin metabolism in relation to apical growth and senescence in genetic lines of peas (Pisum safivum L.). Planta 141, 231-238. Purse, J. G., Horgan, R., Horgan, J. M. and Wareing, P. F. (1976). Cytokinins in sycamore spring sap. Planta 132, 1-8. Radin, J. W. (1984). Stomata1 responses to water stress and to abscisic acid in phosphorus-deficient cotton plants. Plant Physiology 76, 392-394. Radin, J. W. and Eidenbock, M. P. (1984). Hydraulic conductance as a factor limiting leaf expansion in phosphorus-deficient cotton plants. Plant Physiology 75, 372-377. Radin, J. W., Parker, L. L. and Guinn, G. (1982). Water relations of cotton plants under nitrogen deficiency. V. Environmental control of abscisic acid accumulation and stomata1 sensitivity to abscisic acid. Plant Physiology 70,1066-1070. Railton, I. D. and Reid, D. M. (1973). Effects of benzyladenine on the growth of waterlogged tomato plants. Planta 111, 261-266. Ray, J . P. (1989). “A study of endogenous cytokinins and abscisic acid in whole plants of Phaseolus vulgaris L. during deficits in soil water”. PhD thesis, University of Leeds. Reece, C. F. and Rhia, S. J. (1991). Role of root systems of eastern larch and white spruce in response to flooding. Plant, Cell and Environment 14, 29-34. Reid, D. M. and Burrows, W. J. (1968). Cytokinin and gibberellin-like activity in the spring sap of trees. Experimentia 24, 185-187. Reid, D. M. and Crozier, A. (1971). Effects of waterlogging on the gibberellin content and growth of tomato plants. Journal of Experimental Botany 22, 39-48. Reid, D. M. and Wample, R. L. (1985). Water relations and plant hormones. In “Hormonal Regulation of Development 111: Role of Environmental Factors, Encyclopedia of Plant Physiology” (R. P. Pharis and D. M. Reid, eds), Vol. 11, pp. 513-578. Springer-Verlag, Berlin. Reid, D. M . , Crozier, A. andHarvey, B. M. R . (1969). Theeffectsoffloodingon the export of gibberellins from root to shoot. Planta 89, 376-379. Reid, J. B., Murfet, I. C. and Potts, W. C. (1983). Internode length in Pisum. 11. Additional information on the relationship and the action of the loci Le, La, Cry, Na and Lm. Journal of Experimental Botany 34, 349-364. Ribaut, J.-M. and Pilet, P.-E. (1991). Effects of water stress on growth, osmotic

182

M. B. JACKSON

potential and abscisic acid content of maize roots. Physiologia Plantarum 81, 156-162. Richards, D. (1978). Root-shoot interactions: functional equilibria for nutrient uptake in peach (Prunuspersica L. Batsch). Annals of Botany 42,1039-1043. Richards, D. (1981). Root-shoot interactions in fruiting tomato plants. In “Structure and Function of Plant Roots” (R. Brouwer, 0. Gasparikova, J. Kolek, and B. C. Loughman, eds), pp. 373-380. Martinus Nijhoff/Dr W. Junk, The Hague. Richards, D. and Rowe, R. N. (1977). Effects of root restriction, root pruning and 6-benzylaminopurine on the growth of peach seedlings. Annals of Botany 41, 729-740. Richmond, A. E. and Lang, A. (1957). Effect of kinetin on protein content and survival of detached Xanthium leaves. Science 125,650-651. Rivier, L., Milon, H. and Pilet, P. E. (1977). Gas chromatographic-mass spectrometric determinations of abscisic acid levels in the cap and apex of maize roots. Planta 134, 2S27. Rivier, L., Leonard, J.-F. and Cottier, J. P. (1983). Rapid effect of osmotic stress on the content and exodiffusion of abscisic acid in Zea mays roots. Plant Science Letters 31, 133-137. Ruff, M. S . , Krizek, D. T., Mirecki, R . M. and Inouye, D. W. (1987). Restricted root zone volume: influence on growth and development of tomato. Journalof the American Sociew for Horticultural Science 112, 763-769. Saab, I. N. and Sharp, R. E. (1990). Variation in the sensitivity of root and shoot responses to soil drying in maize. In “Importance of Root to Shoot Communication in the Responses to Environmental Stress”, Monograph 21 (W. J. Davies and B. Jeffcoat, eds), pp. 271-277. British Society for Plant Growth Regulation, Bristol. Saab, I. N., Sharp, R. E., Pritchard, J. and Voetburg, G. S. (1990). Increased endogenous abscisic acid maintains primary root growth and inhibits shoot growth at low water potentials. Plant Physiology 93, 1329-1336. Sabanek, J.,Tan,H. M. andKutacek,M. (1985). Growthcorrelative roleofrootson effects of exogenous IAA upon the growth of epicotyls of pea seedlings. Acta Universitatis Agriculturae 33, 5-7. Salama, A. M. S. El-D. A. and Wareing, P. F. (1979). Effects of mineral nutrition on endogenous cytokinins in plants of sunflower (Helianthus annuus L.). Journal of Experimental Botany 30, 971-981. Salim, M. (1991). Changes in water conducting properties of plant roots by nutrition and salt stress. Journal of Agronomy and Crop Science-Zeitschrift fur acker und Pj7anzenbau 166, 285-287. Sanderson, J., Whitbread, F. C. and Clarkson, D. T. (1988). Persistent cross-walls reduce the axial hydraulic conductivity in the apical 20 cm of barley seminal root axes: implications for the driving force for water movement. Plant, Cell and Environment 11,247-256. Sattelmacher, B. and Marshner, H. (1978). Nitrogen nutrition and cytokinin activity in Solanurn tuberosum. Physiologia Plantarum 42,185-189. Satoh, S . , Iizuka, A . , Nakamura, N. and Fujii, T. (1992). Proteins and carbohydrates in xylem sap from squash root. Plant Cell Physiology 33, 841-847. Schulze, E.-D. and Ehleringer, J. R. (1984). The effect of nitrogen supply on growth and water use efficiency of xylem tapping mistletoes. Planta 162, 268-275. Schurr, U. and Gollan, T. (1990). Composition of xylem sap of plants experiencing root water stress-a descriptive study. In “Importance of Root to Shoot Communication in the Responses to Environmental Stress” (W. J . Davies and

PLANT HORMONES AND COMMUNICATION

183

B. Jeffcoat, eds), pp. 201-214. British Society for Plant Growth Regulation, Bristol. Schwabe, W. W. and Al-Doori, A. H. (1973). Analysis of a juvenile-like condition affecting flowering in the blackcurrant (Ribes nigrum). Journal of Experimental Botany 4, 969-981. Selman, I. W. and Sandanam, S. (1972). Growth responses of tomato plants in non-aerated water culture to foliar sprays of gibberellic acid and benzyladenine. Annals of Botany 36, 837-848. Sembdner, G., Weiland, J., Aurich, 0. and Schreiber, K. (1968). Isolation, structure and metabolism of a gibberellin glucoside. In “SCI Monograph No. 31Plant Growth Regulators”, pp. 70-86. Society for Chemical Industry, London. Setter, T. L., Brun, W. A. and Brenner, M. L. (1981). Abscisic acid translocation and metabolism in soybeans following de-podding and petiole girdling. Plant Physiology 67,774-779. Shah, C. B. and Loomis, R. S. (1965). Ribonucleic acid and protein metabolism in sugar beet during drought. Physiologia Plantarum 18, 240-254. Sharp, R. E. and Davies, W. J. (1985). Root growth and water uptake by maize plants in drying soil. Journal of Experimental Botany 36, 1441-1456. Shaybany, B. and Martin, G. C. (1977). Abscisic acid identification and its quantitation in leaves of Juglans seedlings during waterlogging. Journal of the American Society for Horticultural Science 102, 300-302. Shih-Ying, H. and Van Toai, T. T. (1991). Abscisic acid induces anaerobiosis tolerance in corn. Plant Physiology 97, 593-597. Shindy, W. W., Asmundson, C. M., Smith, 0. E. and Kumamoto, J. (1973). Absorption and distribution of high specific radioactivity 2-’4C-abscisic acid in cotton seedlings. Plant Physiology 52, 443-447. Sitton, D., Itai, C. and Kende, H. (1967a). Decreased cytokinin production in the roots as a factor in shoot senescence. Planta 73,296300. Sitton, D., Richmond, A. andvaadia, Y. (1967b). On the synthesisof gibberellinsin roots. Phytochemistry 6, 1101-1105. Sivakumaran, S. and Hall, M. A . (1978). Effects of age and water stress on endogenous levels of plant growth regulators in Euphorbia luthyrus L. Journal of Experimental Botany 29, 195-205. Skene, K. G . M. (1967). Gibberellin-like substances in root exudate of Vitis vinifera. Planta 74, 250-262. Smart, C. M., Scofield, S. R., Bevan, M. W. and Dyer, T. A. (1991). Delayed leaf senescence in tobacco plants transformed with tmr, a gene for cytokinin production in Agrobacterium. The Plant Cell 3, 647-656. Smit, B. A. and Stachowiak, M. (1988). Effects of hypoxia and elevated carbon dioxide concentration on water flux through Populus roots. Tree Physiology 4, 153-1 65. Smit, B. A., Neuman, D. S. and Stachowiak, M. L. (1990). Root hypoxia reduces leaf growth. Role of factors in the transpiration stream. Plant Physiology 92, 1021-1028. Smith, P. G. andDale, J. E. (1988). Theeffectsofrootcoolingandexcision treatments on the growth of primary leaves of Phaseolus vulgaris L. Rapid and reversible increases in abscisic acid content. The New Phytologist 110, 293-300. Smith, K. A. and Jackson, M. B. (1974). Ethylene, waterlogging and plant growth. Report of the Agricultural Research Council Letcombe Laboratory, 1973, pp. 6s7.5. Smith, K. A. and Russell, R. S. (1969). Occurrence of ethylene, and its significance in anaerobic soil. Nature 234, 148-149.

184

M. B. JACKSON

Smith, H. and Wareing P. F. (1964). Gravimorphism in trees. 3. The possible implication of a root factor in the growth and dominance relationships of the shoots. Annals of Botany 28,297-309. Sponsel, V. M. (1987). Gibberellin biosynthesis and metabolism. In “Plant Hormones and their Role in Plant Growth and Development” (P. J. Davies, ed.), pp. 43-71. Martinus Nijhoff, Dordrecht. Stypa, M., Nunezbarrios, A., Barry, D. A., Miller, M. H. and Mitchell, W. A. (1987). Effects of subsoil bulk-density, nutrient availability and soil moisture on corn root-growth in the field. Canadian Journal of Soil Science 67,293-308. Tang, Z. C. and Kozlowski, T. T. (1982). Physiological, morphological and growth responses of Platanus occidentalis seedlings to flooding. Plant and Soil 66, 243-255. Tardieu, F. and Davies, W. J. (1992). Stomata1 response to abscisic acid is a function of current plant water status. Plant Physiology 98, 540-545. Tardieu, F., Katerji, N., Bethenod, O., Zhang, J. and Davies, W. J. (1992a). Maize stomatal conductance in the field; its relationship with the soil and plant water potentials, mechanical constraints and ABA concentration in the xylem sap. Plant, Cell and Environment 14, 121-126. Tardieu, F., Zhang, J. and Davies, W. J. (1992b). What information is conveyed by an ABA signal from maize roots in drying field soil? Plant, Cell and Environment 15, 185-192. Tardieu, F., Zhang, J., Katerji, N., Bethenod, O., Palmer, S. and Davies, W. J. (1992~).Xylem ABA controls the stomatal conductance of field-grown maize subject to soil compaction or soil drying. Plant, Cell and Environment 15, 193-198. Taylor, I . B. (1991). Genetics of ABA biosynthesis. In “Abscisic Acid, Physiology and Biochemistry” (W. J. Davies and H. G. Jones, eds), pp. 23-37. Bios, Oxford. Thornley, J. H. M. (1972). A balanced quantitative model for root :shoot ratios in vegetative plants. Annals of Botany 36,431-441. Tietz, A. (1975). Influence of the culture method on the phytohormone excretion of pea roots. Biochemie und Physiologie der fylanzen 167,371-378. Topa, M. A. and McLeod, K. W. (1986). Responses of Pinus clausa, Pinusserotina and Pinus taeda seedlings to anaerobic solution culture. 11. Changes in tissue nutrient concentration and net acquisition. Physiologia Plantarum 68, 532539. Treharne, K. J., Stoddart, J. L., Pughe, J., Paranjothy, K. and Wareing, P. F. (1970). Effects of gibberellins and cytokinins on the activity of photosynthetic enzymes and ribosomal RNA synthesis in Phaseolus vulgaris L. Nature 228, 129-131. Trejo, C. L. and Davies, W. J. (1991). Drought-induced closure of Phaseolus vulgaris L. stomata precedes leaf water deficit and any increase in xylem ABA concentration. Journal of Experimental Botany 42, 1507-1515. Trewavas, A. J. (1986). Resource allocation under poor growth conditions. A major role for growth substances in developmental plasticity. In “Plasticity in Plants” (D. H. Jennings and A. J. Trewavas, eds), pp. 31-76. Biochemical Society, Colchester. Trewavas, A. J. and Jones, H. G. (1991). An assessment of the role of ABA in plant development. In “Abscisic Acid, Physiology and Biochemistry” (W. J. Davies and H. G. Jones, eds), pp. 169-188. Bios, Oxford. Tromp, J. and Ovaa, J. C. (1990). Seasonal changes in the cytokinin composition of xylem sap of apple. Journal of Plant Physiology 136, 606-610. Tschaplinski, T. J. and Blake, T. J. (1985). Effects of root restriction on growth

PLANT HORMONES AND COMMUNICATION

185

correlations, water relations and senescence of alder seedlings. Physiologia Plantarum 64, 167-176. Turkova, N. S. (1944). Growth reactions in plants under excessive watering. Comptes Rendus (Doklady) de I'Academie des Sciences de I'URSS 42, 8790. Vanden Driessche, R. and Wareing, P. F. (1966). Nutrient supply, dry matter production and nutrient uptake of forest tree seedlings. Annals of Botany 30, 657-672. van Staden, J. and Davey, J. E. (1979). The synthesis, transport and metabolism of endogenous cytokinins. Plant, Cell and Environment 2, 93-106. van Staden, J. and Dimalla, G. G. (1980). The production and utilization of cytokinins in rootless, dormant almond shoots maintained at low temperatures. Zeitschrift fur Pfanzenphysiologie 103, 121-129. van Staden, J. and Drewes, S. E. (1975). Identification of zeatin and zeatin riboside in coconut milk. Physiologia Plantarum 34, 106-109. van Staden, J. and Menary, R. C. (1976). Identification of cytokinins in the xylem sap of tomato. Zeitschrift fur Pfanzenphysiologie 78,262-265. van Staden J. and Smith, A. R. (1979). The synthesis of cytokinins in excised roots of maize and tomato under aseptic conditions. Annals of Botany 42, 75 1-753. van Staden, J., Carmi, A . , Forsyth, C. and Krizek, D. T. (1987). Cytokinin-like activity in the roots and shoots of tomatoes following reciprocal grafts between normal and dwarf genotypes. Suid Afrikaans Tydskrif Vir Plantkunde 53, 27C278. Voesenek, L. A. J. C., Harren, F. J. M., Bogemann, G. M., Blom, C. W. P. M. and Reuss, J. (1990). Ethylene production and petiole growth in Rumex plants induced by soil waterlogging. Plant Physiology 94, 1071-1077. Wadman-van-Schravendijk, W. and van Andel, 0. M. (1985). Interdependence of growth, water relations and abscisic acid level in Phaseolus vulgaris during waterlogging. Physiologia Plantarum 63,215-220. Wadman-van Schravendijk, H. and van Andel, 0. M. (1986). The role of ethylene during flooding of Phaseolus vulgaris. Physiologia Plantarum 66,257-264. Wagner, von H. and Michael, G. (1971). The influence of varied nitogen supply on the production of cytokinins in the roots of sunflower plants. Biochemie und Physiologie der Pfanzen 162, 147-158. Walton, D. C., Harrison, M. A. and Ccite, P. (1976). The effects of water stress on abscisic acid levels and metabolism in roots. Planta 131, 141-144. Wang, T. L. and Wareing, P. F. (1979). Cytokinins and apical dominance in Solanum andigena: lateral shoot growth and endogenous cytokinin levels in the absence of roots. The New Phytologist 82, 19-28. Wang, T. L., Thompson, A. G. and Horgan, R. (1977). A cytokinin glucoside from the leaves of Phaseolus vulgaris L. Planta 138, 285-288. Wang, T.-W. and Arteca, R. N. (1992). Effects of low 0 2 root stress on ethylene biosynthesis in tomato plants (Lycopersicon esculentum Mill cv Heinz 1350). Plant Physiology 98, 97-100. Wareing, P. F. and Nasr, T. (1961). Gravimorphism in trees I. Effects of gravity on growth and apical dominance in fruit trees. Annals of Botany 25, 321-340. Wartinger, A., Heilmeier, H . , Hartung, W. and Schulze, E.-D. (1990). Daily and seasonal courses of leaf conductance and abscisic acid in the xylem sap of almond trees [Prunus dulcis (Miller) D. A. Webb] under desert conditions. The New Phytologist 116, 581-587. Waseem, M., Phipps, J., Carbonneau, R. and Simmonds, J. (1991). Plant growth substances in sugar maple (Acer saccharum Marsh) spring sap. Identification

186

M. B. JACKSON

of cytokinins, abscisic acid and an indolic compound. Journal of Plant Physiology 138, 489493. Weiler, E. (1980). Plant hormone immunoassay. Physiologia Plantarum 54, 230234. Weiss, C. and Vaadia, Y. (1965). Kinetin-like activity in root apices of sunflower plants. Life Sciences 4, 1323-1326. Wellensiek, S. J. (1985). Silene armeria. In “Handbook of Flowering”. (A. H. Halevy, ed.), Vol. IV, pp. 32S330. CRC Press, Boca Raton, FL. Went, F. W. (1938a). Specific factors other than auxin affecting growth and root formation. Plant Physiology 13, 55-80. Went, F. W. (1938b). Transplantation experiments with peas. American Journal of Botany 25, 4455. Went, F. W. (1943). Effect of the root system on tomato stem growth. Plant Physiology 18, 51-65. Went, F. W. and Bonner, D. M. (1943). Growth factors controlling stem growth in darkness. Archives of Biochemistry 1, 439-452. Wheeler, A. W. (1971). Auxins and cytokinins exuded during formation of roots by detached primary leaves and stems of dwarf french beans (Phaseolus vulgaris L.). Planta 98, 128-135. Whitford, P. N. and Croker, S. J. (1991). A homogeneous radioimmunoassay for abscisic acid using a scintillation proximity assay technique. Phytochemical Analysis 2, 134136. Wilkins, H., Alejar, A. A. and Wilkins, S. M. (1978). Some effects of halogenated hydroxybenzoic acids on seedling growth. In “Opportunities for Chemical Plant Growth Regulation”, Monograph 21, pp. 83-94. British Crop Protection Council, Croydon. Wilson, J. B. (1988). A review of evidence on the controls of root : shoot ratios, in relation to models. Annals of Botany 61, 4 3 H 4 9 . Wolf, O., Jeschke, W. D. and Hartung, W. (1990). Long distance transport of abscisic acid in NaC1-treated intact plants of Lupinus albus. Journal of Experimental Botany 41, 593-600. Woolley, D. J. and Wareing, P. F. (1972). The role of roots, cytokinins and apical dominance in the control of lateral shoot form in Solanum andigena. Planta 105, 33-42. Wright, S. T. C. and Hiron, R. W. P. (1969). (+)-Abscisic acid, the growth inhibitor induced in detached leaves by a period of wilting. Nature 224, 7 19-720. Zeevaart, J. A. D. (1983). Environmental control of plant development and its relation to hormones. In “MSU-DOE Plant Research Laboratory Annual Report”, p. 159. Michigan State University, Ann Arbor, MI. Zeevaart, J. A. D, Rock, C. D., Fantauzzo, F., Heath, T. G. and Gage, D. A. (1991). Metabolism of ABA and its physiological implications. In “Abscisic Acid Physiology and Biochemistry” (W. J. Davies and H. G. Jones, eds), pp. 39-52. Bios, Oxford. Zeroni, M., Jerie, P. H . and Hall, M. A . (1977). Studies on the movement and distribution of ethylene in Vicia faba L. Planta 134, 119-125. Zhang, J. and Davies, W. J. (1986). Chemical and hydraulic influences on the stomata of flooded plants. Journal of Experimental Botany 37, 1479-1491. Zhang, J. and Davies, W. J . (1987). ABA in roots and leaves of flooded pea plants. Journal of Experimental Botany 38, 649-659. Zhang, J. and Davies, W. J. (1989a). Abscisic acid produced in dehydrating roots may enable the plant to measure the water status of the soil. Plant, Cell and Environment 12, 73-81.

PLANT HORMONES AND COMMUNICATION

187

Zhang, J. and Davies, W. J. (1989b). Sequential response of whole plant water relations to prolonged soil drying and the involvement of xylem sap ABA in the regulation of stomatal behaviour of sunflower plants. The New Phytologist 113,167-174. Zhang, J. and Davies, W. J. (1990a). Does ABA in the xylem control the rate of leaf growth in soil-dried maize and sunflower plants? Journal of Experimental Botany 41, 1125-1132. Zhang, J. and Davies, W. J. (1990b). Changes in concentration of ABA in xylem sap as a function of changing soil water status can account for changes in leaf conductance and growth. Plant, Cell and Environment 13,277-285. Zhang, J. and Davies, W. J. (1991). Antitranspirant activity in xylem sap of maize plants. Journal of Experimental Botany 42, 317-321. Zhang, J . , Schurr, U. and Davies, W. J. (1987). Control of stomatal behaviour by abscisic acid which apparently originates in the roots. Journal of Experimental Botany 38, 1174-1181. Zimmerman, P. W., Hitchcock, A. E. and Crocker, W. (1931). The movement of gases into and through plants. Contributions of the Boyce Thompson Institute of Plant Research 3 , 313-320. Zobel, R. W. (1986). Rhizogenetics (root genetics) of vegetable crops. HortScience 21,956-959.

This Page Intentionally Left Blank

Second-hand Chloroplasts: Evolution of Cryptomonad Algae

G. I . McFADDEN Plant Cell Biology Research Centre. School of Botany. University of Melbourne. Parkville VIC 3052. Australia

I . Introduction

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

I1. Overview of Cryptomonad Features

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

I11. The Nucleomorph . . . . . . . . . . . . . . . . A . Nucleus-like Organelle . . . . . . . . . . . B . DNA in the Nucleomorph . . . . . . . . . C . Eukaryotic Ribosomes around the Nucleomorph D . Origin of the Nucleomorph . . . . . . . . . E . Isolation of the Nucleomorph . . . . . . . . IV . The Chloroplast . . . . . . . . . . . . A . Chloroplast Membranes . . . . . B . Storage Product . . . . . . . . C. Photosynthetic Pigments . . . . D . Chloroplast Genome . . . . . .

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

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

. . . . . . . . . .

190 192 192 192 195 196 200 203

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

208 208 208 208 210

V . Cryptomonads as Endosymbionts: Parasites of Cryptomonads and Endosymbionts of Cryptomonads . . . . . . . . . . . . .

213

VI . Second-hand Chloroplasts in Other Algae VII . Role of the Nucleomorph

. . . . . . . . . . .

. . . . . . . . . .

214

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

216

VIII . Summary . . . . . . . . . . . . . . . . . . . . . . . . . IX . Taxonomic Appendix Acknowledgements

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

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

218 219 220

References . . . . . . . . . . . . . . . . . . . . . . . . 220 Advances in Botanical Research Vol . 19 Copyright 01993 Academic Press Limited

ISBNO-12-005919-3

All rights of reproductionin any form reserved

190

G. I. McFADDEN

I.

INTRODUCTION

Oxygenic photosynthesis is believed to have evolved at least 3500 million years ago, when prokaryotes were the principal life form (Schopf and Walter, 1982; Krishna Rao et al., 1985; Walsh and Lowe, 1985). Since then, the autotrophic way of life has become widespread among prokaryotes and eukaryotes and has radically altered the biosphere through the increase of oxygen concentration. The global success of photosynthetic organisms attests to the selective advantage of being able to synthesize food from inorganic nutrients and light energy. It follows that natural selection could well favour any means by which the capacity for photosynthesis could be acquired by heterotrophs. The theory of endosymbiosis proposes that green algae, and their descendants the land plants, acquired photosynthetic capacity by engulfing a photosynthetic prokaryote and retaining it within (Margulis, 1981). A wealth of morphological and biochemical evidence (reviewed in CavalierSmith, 1982; Gray and Doolittle, 1982; Taylor, 1987; Gray, 1988, 1989) indicates that either a cyanobacterium (Pace et al., 1986; Giovanni et al., 1988; Turner et al., 1989; Morden and Golden, 1991) or a chlorophyll b-containing prokaryote (Morden and Golden, 1989), was “captured” by a nucleated cell and retained for the purpose of carrying out photosynthesis. The captured prokaryote has subsequently been reduced to an endosymbiont, the chloroplast, by reduction of its genome through transfer (or duplication) of the majority of its genes into the host’s nucleus (Weeden, 1981; Harrington and Thornley, 1982). The inner of the two chloroplast membranes is proposed to represent the plasma membrane of the prokaryote endosymbiont, while the outer membrane is proposed to derive either from the phagocytotic vacuole of the host (Dodge, 1979; Whatley et al., 1979; Whatley and Whatley, 1981) or the outer membrane of the endosymbiont (Cavalier-Smith, 1987). While endosymbiosis of a photosynthetic prokaryote can explain the origin of green algae and subsequently the land plants, the origin of photosynthetic capacity in other algae is much less certain. The diverse morphology and biochemistry of algal chloroplasts have prompted speculation that different algal chloroplasts arose from separate endosymbiotic events, perhaps involving as many as three (Raven, 1970; Larkum and Barrett, 1983; Morden and Golden, 1991), four (Sagan, 1967) or even six (Mereschkowsky, 1910) different prokaryotic endosymbionts. Furthermore, it has been proposed that several algal groups acquired their chloroplasts “secondhand” by engulfing photosynthetic eukaryotes, as opposed to photosynthetic prokaryotes (Dodge, 1979; Whatley et al., 1979; Gibbs, 1981a; Whatley and Whatley, 1981; Whatley, 1989). These evolutionary scenarios involve two sequential endosymbiotic events, one between a nucleated cell and a prokaryote to create a chloroplast-containing eukaryote, and a second

EVOLUTION OF CRYPTOMONAD ALGAE

191

between another phagotrophic eukaryote and the photosynthetic eukaryote (Fig. 1). In cryptomonad algae several lines of molecular and morphological evidence combine to indicate that a heterotrophic flagellate engulfed a red alga to acquire a chloroplast. This review weighs that evidence and examines the symbiotic relationship that apparently involves at least four cells, two prokaryotes and two different eukaryotes, amalgamated into a single cell.

photosynthetic prokaryote phagotrophic eukaryote

photosynthetic eukaryote

2nd phagotrophic eukaryote

alga with second-hand chloroplast

Fig. 1 . The second hand chloroplast hypothesis. All chloroplasts (C) are proposed to ultimately derive from one or more prokaryotic endosymbionts. The primary endosymbiosis results in a double membrane-bound compartment with a prokaryotic genome-the chloroplast. The chloroplasts of some algae are believed to have been acquired by a secondary endosymbiosis of a photosynthetic eukaryote. The chloroplast-containing eukaryote is hijacked by a second phagotrophic eukaryote (nucleus N’). In cryptomonads, the nucleus (N) and cytoplasm of the photosynthetic eukaryote are apparently retained.

192

G. I. McFADDEN

11. OVERVIEW OF CRYPTOMONAD FEATURES The first cryptomonad was described by Ehrenberg in 1832. Currently there are approximately 50 genera and some 200 species described in the Division Cryptophyta, but the systematics of the group is in a state of flux (see Section IX: Taxonomic Appendix). Cryptomonads are abundant in marine and freshwater habitats. Cells are typically ovoid to bean-shaped, and somewhat flattened laterally. The two flagella, which can both bear tubular hair-like appendages known as mastigonemes or “flimmers”, are inserted into an anteriolateral invagination known as the gullet. Cells lack a rigid wall but are covered in an elaborate array of scales and pellicular plates. A characteristic feature is the presence of “ejectisomes” ; extrusive organelles of unknown function (Anderson, 1962; Wehrmeyer, 1970). The single, ramified mitochondrion has flat cristae (Santore and Greenwood, 1977). The nucleus is typically located in the posterior of the cell and mitosis involves partial breakdown of the nuclear envelope and an open, barrel-shaped spindle without centrioles (see Meyer and Pienaar, 1984, for references). Sexual reproduction is not clearly established but may occur (Hill and Wetherbee, 1986; Kugrens and Lee, 1988). Further details of cryptomonad fine structure can be found in previous reviews (Gantt, 1980; Oakley and Santore, 1982; Gillott, 1990).

111. THE NUCLEOMORPH A . NUCLEUS-LIKE ORGANELLE

A key feature of the cryptomonads is a small organelle termed the nucleomorph. This structure, first described in 1974 (Greenwood, 1974), is proposed to be the vestigial nucleus of the eukaryotic endosymbiont (Greenwood et al., 1977; and see Fig. 1). The nucleomorph is located in the periplastidal space, a cytoplasmic compartment between the chloroplast envelope and the extension of the rough endoplasmic reticulum that envelops the chloroplast (Fig. 2). Within the periplastidal space are particles resembling eukaryotic (80s) ribosomes and starch grains (Sepenswol, 1973; Gillott and Gibbs, 1980). All cryptomonads with a chloroplast (even nonpigmented chloroplasts-leucoplasts) contain a nucleomorph, and those species having two chloroplasts have two nucleomorphs. The heterotrophic cryptomonad genus, Goniomonas, has no chloroplast and no nucleomorph (Mignot, 1965; Hill, 1991a). The nucleomorph exhibits many characteristics of a eukaryotic nucleus, albeit a very small one. The nucleomorph is round to ovoid with a diameter of about 1 pm (Figs. 3 and 4). It is delimited by a double membrane envelope with several pores reminiscent of nuclear membrane pores (Gillott

EVOLUTION OF CRYPTOMONAD ALGAE

193

Fig. 2. Layout of cryptomonad cell. The cryptomonad apparently contains a photosynthetic eukaryotic endosymbiont, and the nucleomorph and periplastidal space are the vestiges of the endosymbiont’s nucleus and cytoplasm (compare with Fig. 1). The chloroplast is enveloped by four membranes. Between the two inner and the two outer membranes is a compartment known as the periplastidal space, within which is a small nucleus known as the nucleomorph. The two inner membranes are probably homologous to the envelopes of other chloroplasts. The outer two membranes are termed the chloroplast endoplasmic reticulum (CER). The outermost membrane is continuous with the endoplasmic reticulum-nuclear envelope and bears ribosomes. The third membrane, counting out from the chloroplast, could represent the plasma membrane of the eukaryotic endosymbiont, in which case the endosymbiont is actually within the lumen of the host cell’s endoplasmic reticulum.

and Gibbs, 1980; Morrall and Greenwood, 1982; Santore, 1982). The contents of the nucleomorph are distinguished into three components: a background matrix; several electron-dense, spherical to rod-shaped structures; and a relatively large granular zone of medium electron density that resembles a nucleolus (Figs. 3 and 4). Nucleomorph division is dissimilar to normal nuclear division in that no spindle is visible; the nucleomorph

194

G . I. McFADDEN

Fig. 3. Longitudinal section of a cryptomonad (Kornrnu cuuduru) showing the main nucleus (Nu) and the nucleomorph (Nm). The chloroplast (Chl) contains stacks of thylakoids with electron-dense material in the lumina and a pyrenoid (Py). The mitochondrion (Mi) has flat cristae. Scale bar = 1 pm.

EVOLUTION OF CRYPTOMONAD ALGAE

195

Fig. 4. The nucleomorph of Plugioselmis pulusrris. Between the inner and outer pairs of membranes surrounding the chloroplast is the periplastidal space (asterisk) which contains eukaryotic ribosomes and starch grains (not shown). Also located within the periplastidal space is the nucleomorph, a miniature nucleus complete with a double membrane interrupted by pores (arrowheads), a nucleolus-like zone (open arrow). and unidentified electron-dense bodies (small arrows). Scale bar = 0.5 pm.

division apparently occurs by pinching in two (McKerracher and Gibbs, 1981; Morrall and Greenwood, 1982). Chromatin is not distinguishable within the nucleomorph during the division process (McKerracher and Gibbs, 1981; Morrall and Greenwood, 1982). The nucleomorph is the first structure to divide in the cell, followed by the chloroplast, then the main nucleus. B.

DNA IN THE NUCLEOMORPH

Several histochemical studies demonstrate the presence of DNA in the nucleomorph, further confirming its nucleus-like status. Staining with the (DAPI) shows DNA-binding fluorochrome 4’-6-diamino-2-phenylindole fluorescence in the cryptomonad nucleomorph (Hansmann et al., 1985; Ludwig and Gibbs, 1985). The DNA content of the nucleomorph was estimated at approximately lo8 or lo9 daltons (Ludwig and Gibbs, 1985). Immunogold techniques have localized DNA to the matrix of the

196

G . I. McFADDEN

nucleomorph and to a part of the nucleolus-like zone (Hansmann et al., 1986). The anti-DNA antibody did not bind to the electron-dense globules in the nucleomorph, so the composition and role of these structures remains enigmatic.

C. EUKARYOTIC RIBOSOMES AROUND THE NUCLEOMORPH

Since the nucleomorph appeared to be a eukaryotic nucleus within a subcompartment of the cryptomonad cell, it was relevant to enquire: (1) were the ribosome-like particles present in the periplastidal space around the nucleomorph really ribosomes; (2) were they prokaryotic or eukaryotic ribosomes; and (3) were these ribosomes perhaps encoded by nucleomorph DNA? Once again, histochemical techniques were employed. Using the enzyme-gold method with a ribonuclease (Bendayan, 1981), Hansmann (1988) demonstrated the presence of RNA in the periplastidal space, supporting the idea that the 22nm particles within it were ribosomes. The RNase-gold also indicated RNA within the nucleolus-like portion of the nucleomorph, suggesting that transcriptionally active ribosomal RNA (rRNA) genes could be present in this zone of the nucleomorph. If the nucleomorph and periplastidal space of the cryptomonad are the vestiges of the nucleus and cytoplasm of a eukaryotic endosymbiont, then the rRNA genes in the nucleolus-like portion of the nucleomorph and the rRNAs in the periplastidal ribosomes should share sequence identity with the other eukaryotes. The author has been able to localize either eukaryotic or prokaryotic rRNAs at the ultrastructural level using techniques of highresolution in situ hybridization (McFadden et al., 1990; McFadden, 1991; see also Fig. 5). Probing of cryptomonad cells with eukaryotic- and prokaryotic-specific rRNA clones shows that the eukaryotic rRNAs are present in the periplastidal space (Figs. 6 and 7; see also McFadden, 1990a,b). Prokaryotic rRNAs are found in the chloroplast (Fig. 8) and also in the mitochondrion (McFadden, 1991). The in situ hybridization technique also detects transcripts of rRNAs in the nucleolus (McFadden et af., 1988, 1990; McFadden, 1989,1991). The nucleolus-like zone of the nucleomorph labels with the eukaryotic rRNA probe (Figs. 9a,b), indicating that rRNA genes are possibly being transcribed therein (McFadden, 1990a,b). Microscopy has thus demonstrated that the nucleomorph has four characteristics of a eukaryotic nucleus: a double membrane with pores, DNA, self-replication, and a nucleolus-like zone in which eukaryotic rRNAs are apparently transcribed. The nucleomorph is surrounded by eukaryotic ribosomes in the periplastidal space which is isolated from the rest of the cell by the chloroplast endoplasmic reticulum (CER). Are the rRNAs in the periplastidal ribosomes encoded by DNA in the nucleolus-like zone of the nucleomorph? No

EVOLUTION O F CRYPTOMONAD ALGAE

197

ultrathin resin section

Fig. 5 . Diagrammatic representation of the in situ hybridization procedure used to localize either prokaryotic or eukaryotic rRNA (McFadden, 1991). A biotin-labelled probe is synthesized from a clone of either a prokaryotic or eukaryotic rRNA gene. The probe is hybridized to complementary nucleic acid sequences (rRNAs) in the ribosomes at the surface of the section. The biotin tag is then detected with an antibody to biotin developed in goats. These goat IgGs are then detected with rabbit-anti-goat IgGs conjugated to colloidal gold markers. The diagram is not to scale and is not intended to represent the way in which the antibodies recognize their determinants.

Fig. 6 . Transmission electron micrograph (longitudinal section of Komma caudara) of a cryptomonad showing the nucleomorph (Nm) and the periplastidal space (arrows) within which is a starch grain (S). The chloroplast (Chl) and pyrenoid (Py) are also visible. The main nucleus (Nu) is toward the cell posterior. Scale bar = 0.5 pm.

Fig. 7. Localization of eukaryotic rRNA in a section similar to Fig. 6. The markers indicate the presence of eukaryotic rRNAs in the main cytoplasm and the nucleolus of the main nucleus. In addition, the periplastidal space (arrows) is clearly positive for eukaryotic rRNA (arrows). Scale bar = 0.5 pm.

Fig. 8. Localization of prokaryotic rRNA in a section similar to those shown in Figs. 6 and 7. The ribosomes in the chloroplast stroma are heavily labelled with the prokaryotic probe, but the periplastidal space (arrows) and the main nucleocytoplasmic area are not labelled. Scale bar = 0.5 pm.

Fig. 9. Localization of eukaryotic rRNA in the nucleolus-like zone of the nucleomorph of Kornmu cuudutu. The labelling is of equivalent intensity to the labelling in the nucleolus of the main nucleus (No). Scale bars = 0.25 pm.

firm proof is yet forthcoming but the answer is very likely to be in the affirmative. The only other eukaryotic rRNA genes located in cryptomonads thus far occur in the main nucleus (McFadden, 1990b). In order to reach the periplastidal space, rRNAs from the main nucleus would need to cross the two membranes of the CER. Certain short RNAs apparently cross

200

G. I. McFADDEN

the mitochondria1 membrane (Chang and Clayton, 1987; MarCchalDrouard et al., 1988), but no mechanisms of membrane translocation for rRNAs is yet known. D. ORIGIN OF THE NUCLEOMORPH

1. Autogenous or exogenous origin Taken together, the morphological and histochemical data outlined above are congruent with the presence of a vestigial photosynthetic eukaryote within the cryptomonad, and support the idea of a secondary endosymbiosis as represented in Fig. 1. The microscopical data do not, however, discount the possibility that the nucleomorph and periplastidal space are of autogenous rather than exogenous origin. Autogenous origin, where the nucleomorph is derived from the main nucleus and partioned off into the periplastidal space by the CER, is difficult to explain (Cavalier-Smith, 1986), but examples of genome partitioning such as the micro- and macronuclei of ciliates do occur (Lynn and Small, 1990). Small blebs of dinoflagellate nuclei, reminiscent of the cryptomonad nucleomorph, are sometimes partitioned off by evaginations of the nuclear envelope (Lewis and Burton, 1988), but whether these blebs contain any DNA or have any function is undetermined. Bogorad (1975, 1982) discussed the possible autogenous origin of mitochondria and plant chloroplasts by partitioning of the nuclear genome, but he did not include the cryptomonad chloroplast and nucleomorph in his discussion. 2. Cryptomonads have two different sets of rRNA genes The definitive test is to compare gene sequences from the nucleomorph with gene sequences from the main nucleus. If the nucleomorph is autogenously derived, the sequences can be expected to be similar, although some divergence might have occurred if the nucleomorph “broke away” a long time ago. Conversely, if the nucleomorph is indeed the genome of a foreign cell, then the gene sequences could be expected to be radically divergent. The latter is exactly what was found (Douglas et al., 1991a). Since the nucleomorph apparently contained rRNA genes, these were the logical sequences to compare. Moreover, the expanding database of rRNA gene sequences could be cross-checked to sift through the eukaryotes for possible relatives of the putative symbionts. But how to get the sequences out of the two nuclei? A means of separating the nucleomorphs from the main nuclei was not available, so Douglas et al. (1991a) employed the polymerase chain reaction (PCR; Saiki et al., 1988). They reasoned that if the 16s-like rRNA genes of the nucleomorph and nucleus were divergent, then they might be different in size. Using primers that direct amplification of the eukaryotic 16s-like rRNA gene, Douglas et al. (1991a) amplified total cryptomonad

EVOLUTION OF CRYPTOMONAD ALGAE

201

DNA and obtained two products. These PCR products were cloned and sequenced, and the two sequences aligned. Alignment revealed that both were obviously eukaryotic 16S-like rRNAs, but they shared only 70% positional identity, indicating highly divergent sequences. It has been previously demonstrated that two different rRNA sequences can occur in a single cell (Gunderson et al., 1987a), but the divergence is considerably less and one sequence is clearly a derivative of the other (Enea and Corredor, 1991). Secondary structure models indicate that the two cryptomonad rRNAs would have distinct structures and that one is not a derivative of the other (Douglas et al., 1991a). The shorter 16s-like gene was assigned to the nucleus on the basis that it alone could be amplified from DNA prepared from isolated cryptomonad nuclei. By a process of elimination, the longer sequence was assumed to derive from the nucleomorph. Proof that both cryptomonad sequences are active genes was obtained by positive Northern hybridization using sequence-specific synthetic oligonucleotides (Douglas et al., 1991b). The nucleus-specific probe hybridized to an RNA of 1800 bp. Two nucleomorph-specific probes were used, and each probe hybridized to a different RNA, one of 1450 bp and one of 250 bp. The nucleomorph transcript is apparently discontinuous, and Douglas et al. (1991a) postulate that post-transcriptional excision of a 166 bp spacer from variable region V2 is the probable explanation. Discontinuities in nucleocytoplasmic 23s-like rRNAs of certain insects, the chloroplast 23S-like rRNAs of plants and green algae (Appels and Honeycutt, 1986), and the mitochondria1 16s-like rRNAs of a ciliate (Schnare et al., 1986) have been previously reported, but the nucleomorph rRNA is the first known bipartite nucleocytoplasmic 16S-like rRNA.

3. The nucleomorph is derived f r o m a red algal nucleus Ph ylogenetic trees, based on comparison of the cryptomonad nucleus and nucleomorph rRNA sequences determined by Douglas et al. (1991a) with other eukaryote sequences, reveal several interesting affiliations (Fig. 10). The nuclear sequence and nucleomorph sequence are clearly not closely related. The nuclear sequence branches with the rhizopod Acanthamoeba, and these two form a sister group to the land plants and green algae. The nucleomorph sequence affiliates with two red algal sequences (Fig. lo), supporting the theory that the eukaryotic endosymbiont in cryptomonads was a red alga (Dodge, 1979; Whatley et al., 1979; Gillott and Gibbs, 1980; Gibbs, 1983). 4. Are cryptomonads the ancestors of chromophyte algae? An important finding of the rRNA sequence phylogenetic analyses (Perasso et al., 1990; Douglas et ul., 1991b; Eschbach et al., 1991b) is that the cryptomonad nuclear sequence does not branch together with members of the chromophyte algae (chlorophyll c-containing algae including diatoms and chrysophytes but excluding dinoflagellates; see Fig. 10). It had

202

G. I . McFADDEN

previously been suggested that cryptomonads were perhaps the ancestors of the chromophytes because they also possessed chlorophyll c, four membranes surrounding the chloroplasts (chloroplast ER), and tubular mastigonemes on the flagella (Dodge, 1979; Cavalier-Smith, 1986, 1989). The chromophytes would thus derive from cryptomonads by loss of the nucleomorph, periplastidal ribosomes, and phycobilins (Coombs and Greenwood, 1976; Dodge, 1979; Gibbs, 1981b; Cavalier-Smith, 1986, 1989). In addition, the oomycetes, hyphochytrids and thraustochytrids (all of which produce heterokont zoospores with tubular mastigonemes) were held not to be true fungi but merely chromophytes that had lost their chloroplasts (Cavalier-Smith, 1986, 1989). The evolutionary progression would thus be cryptomonads + chromophytes + zoosporic fungi (Cavalier-Smith, 1986).

Fig. 10. Phylogenetic tree showing the relationship of the two cryptomonad rRNA sequences toother eukaryote sequences (redrawn from Douglaseral., lYYlb, with permission). The two cryptomonad sequences are not closely related. The sequence believed to derive from the nucleomorph branches with the red algae (Gracilaria and Gracilariopsi,s), while the sequence from the main nucleus is affiliated with Acantharnoeba. A clade containing the chromophyte algae (Ochrornonas and Skeletonemu) and the oomycete (Achlyia) is remote to the cryptomonad nuclear sequence, refuting the suggestion that chromophytes and oomycetes are descendants of a cryptomonad. The scale bar represents 10 changes per 100 nucleotides.

EVOLUTION OF CRYPTOMONAD ALGAE

203

Phylogenetic trees based on both molecular and morphological characters group chromophytes and zoosporic fungi together (Gunderson, et al., 1987b; Bhattacharya and Druel, 1988; Beakes, 1989; Whatley, 1989; Ariztia etal., 1991; Bhattacharya etal., 1991; Williams, 1991), but a recent analysis indicates that zoosporic fungi diverged before the acquisition of a chloroplast produced the chromophyte lineage (Williams, 1991). Several cryptomonad characteristics make it unlikely that chromophytes are their descendants. First, cryptomonad mitochondria have flat cristae while those of chromophyte mitochondria are tubular (Dodge, 1979). Second, the cryptomonad periplast is unique and not at all similar to the surface coverings of chromophytes. Finally, cryptomonads are anisokont while chromophytes (with the exception of prymnesiophytes) are classically heterokont (Moestrup, 1982). From the available morphological and molecular data it would seem that the chromophytes and zoosporic fungi are probably not descendants of cryptomonads. This means that CER and tubular mastigonemes are either shared primitive characters or have evolved convergently in cryptomonads and chromophytes. Cavalier-Smith (1986) argues that C E R probably only evolved once, but host E R is associated with the endosymbiont in the ciliate Myrionecta (see Section V), suggesting that a CER-like arrangement of membranes may not be so unique. It also seems doubtful that the tubular appendages on cryptomonad flagella are homologues of those on chromophyte flagella (Moestrup, 1982). Cryptomonads have tubular appendages on both flagella, so they are not heterokont like most of the chromophytes and zoosporic fungi. Moreover, the cryptomonad appendages are bipartite whereas the chromophyte flagella appendages are tripartite (Moestrup, 1982). Similarly, chlorophyll c could have perhaps evolved twice from chlorophyll a (Jeffrey, 1989), or perhaps both cryptomonad and chromophyte chloroplasts developed from closely related primary prokaryotic endosymbionts containing chlorophyll c (Larkum, 1991; see also Section 1V.D).

E. ISOLATION OF THE NUCLEOMORPH

I . Separation of nucleomorph and nuclear DNA While the PCR analysis reveals two distinct sets of rRNA genes in cryptomonads, definitive proof that one set is from the nucleomorph and the other from the nucleus is not yet available. Such proof could come from either in situ hybridization or blot hybridizations using isolated nucleomorphs and nuclei. Attempts to separate nuclear DNA from nucleomorph DNA by isopycnic centrifugation, based on differences in buoyant density, have been without success (Hansmann et al., 1987). Two low-density bands contained circular chromosomes of approximate size 130 kb and 43 kb that are believed to represent chloroplast and mitochondria1 DNA respectively (Hansmann et

204

G. I. McFADDEN

Fig. 11. Longitudinal section of Rhodomonas salina showing the nucleomorph (arrow) within the pyrenoid (Py). The pyrenoid has an invagination, which is lined by the two innermost chloroplast membranes, within which the nucleomorph is positioned. Scale bar = 1 pm.

af., 1987). The major high-density band contained nuclear DNA, but no band corresponding to nucleomorph DNA could be found. Recently, a lateral approach to isolation of the nucleomorph was developed using a cryptomonad in which the nucleomorph was encapsulated in the pyrenoid (Fig. 11). Hansmann and Eschbach (1990) lysed crypto-

Fig. 12. Electron micrograph of isolated pyrenoid/nucleomorph complexes from Rhodornonas salina. Each pyrenoid contains a nucleomorph. The preparation is 99.8% pure with respect to nuclear contamination. Scale bar = 1 pm. *

206

G . I . McFADDEN

Fig. 13. Higher magnification of isolated pyrenoid-nucleomorph complexes showing the pyrenoid (Py) encasing the nucleomorph (Nm). Scale bar = 0.5 p,m.

monads and banded the lysate on Percoll step gradients to isolate the pyrenoids, still containing their nucleomorphs. Nucleomorph/pyrenoid complexes isolated from Rhodornonas salina using the protocol of Hansmann and Eschbach (1990) are shown in Figs. 12 and 13. DAPI fluorometry of the isolated nucleomorphs shows between 1.3Mb and 2.8Mb of DNA per nucleomorph (Hansmann and Eschbach, 1990). If the 2.8Mb examples represent DNA doubling prior to nucleomorph division, and one assumes that the nucleomorph contains a diploid set of chromosomes, then the haploid size of the nucleomorph genome is 0.7Mb-about 1130th the size of the cyanobacterium Anacysfis nidulans or 1/7th the size of Escherichia coli. If the nucleomorph is indeed the vestige of a red algal nucleus, it would seem to be drastically reduced in DNA content. The separation of nucleomorph and nuclear DNA will allow construction of a nucleomorph genomic library from which it should be possible to isolate nucleomorph genes.

EVOLUTION O F CRYPTOMONAD ALGAE

207

2. A diploid nucleomorph with three short, linear chromosomes An electrophoretic karyotype of the nucleomorph has been established by pulsed-field gel electrophoresis of isolated nucleomorphs and shows three linear chromosomes of 195 kb, 225 kb and 240 kb (Eschbach et al., 1991a). How the three linear chromosomes are segregated during division of the nucleomorph, in the absence of a detectable spindle, is an intriguing question. The total of 660 kb for the three chromosomes is in good agreement with the DAPI estimates and is congruent with a diploid nucleomorph. Southern blotting of the isolated chromosomes with probes for either 16slike or 23s-like rRNA genes indicates that each chromosome carries at least one set of rRNA genes (Eschbach et al., 1991a). A dot blot of nuclear and nucleomorph DNA from Rhodomanas salina probed with rDNA of pea (Jorgensen et al., 1987) also indicates presence of rRNA genes in the nucleomorph DNA (Fig. 14). It should now be possible to probe isolated nucleomorph DNA with the oligonucleotide probes believed to be specific for either nuclear or nucleomorph DNA (Douglas et al., 1991a) to test whether the putative nucleomorph sequence represented in Fig. 10 is indeed derived from the nucleomorph.

Fig. 14. Slot blot of nucleomorph and total Rhodomonas salina DNA probed with rRNA genes of pea (2.6 kb BamHl fragment; Jorgensen et al., 1987). The nucleomorph DNA is positive for rRNA genes.

208

G. I. McFADDEN

IV. THE CHLOROPLAST A. CHLOROPLAST MEMBRANES

Most cryptomonads have a single chloroplast, but species of the genera Cryptomonas and Campylomonas have two. The chloroplast is bounded by four membranes-an inner pair of membranes comprising the plastid envelopes, and two outer membranes, known as the chloroplast endoplasmic reticulum (CER), which are an extension of the endoplasmic reticulum/ nuclear envelope (Figs. 2 , 3 , 4 ) . The outermost CER membrane of cryptomonads bears ribosomes, as does the outermost CER membrane in chromophytes (Gibbs, 1979, 1981b). The cryptomonad thylakoids are typically organized in pairs but loose stacks of several thylakoids occur. (Figs. 3,4,6, and 11). The thylakoids are relatively thick and contain electron-dense material (Figs. 3, 4, 6, and 11). B. STORAGE PRODUCT

No starch is stored in the chloroplast, but starch grains are found in the periplastidal space (Fig. 6). This location of starch is consistent with cryptomonads harbouring a red algal endosymbiont (Gillott and Gibbs, 1980), since red algae store starch in their cytoplasm. However, the cryptomonad starch contains amylose (an a(1,4)-linked linear glucan) and amylopectin (an a(174,6)-branched glucan) and thus resembles starch of land plants (Antia et al., 1979; McCracken and Cain, 198l), while red algal starch (termed floridean starch) was initially reported to contain no amylose (Percival, 1968; Craigie, 1974). Further investigations have revealed the presence of amylose in unicellular (McCracken et al., 1980) and multicellular (McCracken and Cain, 1981) red algal starch, and the so-called floridean starch is probably restricted to certain multicellular red algae. Hence, the presumably primitive unicellular red algae-which are perhaps the ancestors of cryptomonad chloroplasts, periplastidal space and nucleomorph-probably had amylose. C. PHOTOSYNTHETIC PIGMENTS

A unique combination of light-harvesting pigments distinguishes the cryptomonad chloroplast from other algae. In addition to chlorophylls a and c2,the cryptomonads have phycobiliproteins. Depending on the species, either phycoerythrin or phycocyanin is present (Hill and Rowan, 1989). Cryptomonad phycobiliproteins are apparently homologues of phycobiliproteins present in red algae and cyanobacteria (Glazer and Apell, 1977; Wehrmeyer, 1983; Guard-Friar et al., 1986; Reith and Douglas, 1990).

EVOLUTION OF CRYPTOMONAD ALGAE

209

However, the cryptomonad phycobiliproteins are not organized into phycobilisomes like those of red algae and cyanobacteria (Glazer, 1982). Moreover, only one phycobiliprotein occurs in each cryptomonad, while two or three are present in red algae and cyanobacteria. Another essential difference is in phycobiliprotein location. The phycobilisomes of red algae and cyanobacteria are attached to the stromal side of the thylakoid membrane (Glazer, 1982), but cryptomonad phycobiliprotein occurs within the lumen of the thylakoid membranes (Gantt et al., 1971; Dwarte and Vesk, 1982; Spear-Bernstein and Miller, 1987) and is probably attached to the lumenal side of the thylakoid membrane (Ludwig and Gibbs, 1989a). The unique electron-dense nature of the cryptomonad thylakoid lumina (Figs. 3, 4, 6 and 11) is attributed to the phycobiliprotein therein (Dodge, 1969; Gantt et al., 1971; Rhiel et al., 1985) and absence of phycobilisomes may allow the stacking of thylakoids in cryptomonads (Spear-Bernstein and Miller, 1985), a chloroplast feature not observed in red algae or cyanobacteria. The cryptomonad phycobiliprotein is known to act as the major light antenna (Haxo and Fork, 1959; Gantt, 1979), but exactly how the excitation energy is transferred to photosystem I1 in the absence of the usual intermediary structures in the phycobilisome is unknown. Since the phycobiliprotein is on the lumenal or “wrong” side of the thylakoid, it was suggested that perhaps cryptomonad thylakoids are “inside-out,’ and that energy transfer to photosystem I1 would thus be possible (Gantt e t a l . , 1971; Gantt, 1979, 1980). However, freeze-fracture analysis indicates normal topology for cryptomonad thylakoid membrane proteins (Spear-Bernstein and Miller, 1985), and it must be assumed that energy transfer from an antenna on one side of the membrane to a reaction centre on the other side is occurring. Possibly the intralumenal arrangement of the light antenna compensates for the lack of phycobilisomes (LichtlC et al., 1987; Ludwig and Gibbs, 1989a), or, alternately, phycobilisomes are required for energy transfer if the antenna is not within the lumen. In red algae, the presumed endosymbiont in cryptomonads, the linker peptides of the phycobilisome are nuclear-encoded (Grossman et al., 1986) so the linker protein genes, were they present in cryptomonads, would be expected to reside in the nucleomorph. Ludwig and Gibbs (1989a) suggested that perhaps the linker peptides were lost from nucleomorph, and that a mechanism for translocation of P-phycoerythrin to the thylakoid lumen had previously evolved to compensate for the loss in ability to form phycobilisomes. To further confound the issue, cryptomonad p-phycoerythrin, which is chloroplast genome-encoded, lacks the standard leader peptide that directs proteins from the stromal space into the thylakoid lumen (Howe and Wallace, 1990; Reith and Douglas, 1990). It is interesting in this respect that the a-phycoerythrin subunits of cryptomonads are apparently not related to the a-phycoerythrin subunits of red algae (Sidler et al., 1985, 1987), and

210

G. I. McFADDEN

these may perhaps be involved in transporting the P-phycoerythrin subunit across the thylakoid membrane (Reith and Douglas, 1990).

D. CHLOROPLAST GENOME

Analysis of the plastid genome from cryptomonads could be expected to help identify the eukaryotic endosymbiont, and perhaps even the primary prokaryotic endosymbiont. Determination of the content, arrangement, coding patterns and copy number of the various genes in the chloroplast genome has helped in establishing evolutionary relationships among land plants and, to a lesser extent, among the algae (Palmer, 1985,1990; Cattolico, 1986; Cattolico and Loiseaux-de Goer, 1989).

I. Structure of the chloroplast chromosome Physical mapping of a cryptomonad chloroplast genome reveals a circular chromosome of 118 kb containing two small inverted repeats (Douglas, 1988) which is thus superficially similar to chloroplast chromosomes of land plants and most algae. Surprisingly though, the gross structure is not equivalent to red algal chloroplast chromosome structure. Two red algal chloroplast chromosomes have been mapped (Li and Cattolico, 1987; Shivji 1990), and they are both different from the cryptomonad. One red alga, Crifithsia pac@ca, has only one set of rRNA genes and hence no inverted repeat whatsoever (Li and Cattolico, 1987). The other species, Porphyra yezoensis, has inverted repeats, but they are situated in the chromosome in the reverse orientation to cryptomonads with respect to the large singlecopy region containing rbcLS (operon encoding the large and small subunits of Rubisco) and tufA (protein synthesis elongation factor Tu). Since the inverted repeats in the green algae are oriented in the opposite direction to those of their descendants, the land plants (Palmer, 1985), it is difficult to determine the significance of the anomalous repeat orientation in cryptomonads and red algae. Moreover, since the loss of one repeat is known to have occurred in Pisum sativum and Vicia faba (Palmer and Thompson, 1982), it is possible that cryptomonad chloroplast chromosomes are derived from a Grifithsia-like ancestor with the same repeat orientation that has since lost one repeat. Another explanation could be that the inverted repeat has evolved independently in red algae and cryptomonads. Palmer and Thompson (1982) propose that the inverted repeat serves to stabilize the chloroplast genome from rearrangement and prevent recombinational gene loss. Such a beneficial trait could well have evolved convergently and the differences in inverted repeat content argue for parallel development. Difficulties in rationalizing chloroplast chromosome structure are also apparent when comparing cryptomonads with chromophytes. The inverted repeats of the two brown algae Pylaiella littoralis (Loiseaux-de Goer et al.,

EVOLUTION OF CRYPTOMONAD ALGAE

21 1

1988) and Dictyota dichotoma (Kuhshel and Kowallik, 1987) and the chrysophyte Ochromonas dancia (Cattolico and Loiseaux-de Goer, 1989) have the same orientation as cryptomonads with respect to rbcL but not psbA. In two diatoms (Odontella and Coscindiscus) and the xanthophyte Vaucheria (Kowallik, 1989), the inverted repeat is oriented in the same way with respect to psbA (D1-reaction centre polypeptide of photosystem 11). psbA is actually contained within the inverted repeat of Coscindiscus (Kowallik, 1989). The rbcLS genes of Vaucheria have not been mapped. In another chromophyte, Olisthodiscus luteus, the 23s rRNA genes are proximal to the single-copy region containing psbA (Reith and Cattolico, 1986; Delaney and Cattolico, 1989), as occurs in cryptomonads. However, rbcLS is contained within the inverted repeat Olisthodiscus luteus (Reith and Cattolico, 1986; Delaney and Cattolico, 1989). Rearrangements of chloroplast genes are common (Palmer, 1985), and even occur within members of the same genus of chromophytes (Kowallik, 1990), so comparison of the gross morphology of chloroplast chromosomes is perhaps premature given the small data set for algae. Once more detailed maps for a wider selection of algal taxa are available, it should be possible to draw more conclusive comparisons, as has been possible with land plants (Palmer, 1990).

2. Chloroplast gene sequences Several protein, tRNA, and rRNA genes have been sequenced from the cryptomonad chloroplast chromosome (Douglas and Durnford, 1989, 1990a,b; Douglas et al., 1991a; Douglas, 1991; Douglas and Turner, 1991). Like land plants and other algae, the cryptomonad tRNA"" and tRNAA'" genes are located in the spacer between 16s and 23s rRNAs (Douglas and Durnford, 1990a). The cryptomonad tRNA1Ie and tRNAAl" genes are similar to those of other algae in that they lack introns (Douglas and Durnford, 1990a) characteristic of the land plant chloroplast lineage (Manhart and Palmer, 1990). Like all chloroplast tRNA"" and tRNAAlagenes, the cryptomonad genes do not encode the CCA 3' termini (Douglas and Durnford, 1990a). These results indicate that the inverted repeats of chloroplast chromosomes in land plants and algae (including cryptomonads) are all homologous and that they could well be derived from the rRNA operon of bacterial endosymbionts, since the latter usually contain tRNA"" and tRNAAlagenes (Sprinzl et al., 1989). 3. Rubisco phylogeny Genes for the large and small subunits of ribulose-l,5-bisphosphate carboxylase/oxygenase (rbcL and rbcS) have now been sequenced from a wide variety of organisms including cryptomonads (Douglas and Durnford, 1989; Douglas et al., 1991a). Phylogenetic trees based on comparison of rbcL and rbcS sequences cluster cryptomonads with red algae (Douglas and Durnford, 1989; Valentin and Zetsche, 1990a,b; Douglas et al., 1991b;

212

G . I. McFADDEN

Morden and Golden, 1991), again supporting the hypothesis that cryptomonads contain a red algal endosymbiont. The cryptomonad sequence shares more positional identity with the unicellular red algae Porphyridium aerugineum than with Cyanidium caldarium, which is an unusual thermoacidophyllic unicellular red alga (Valentin and Zetsche, 1990b). Rubisco from chromophytes is similar to that from red algae (Newman et a l . , 1989), and phylogenetic trees derived from rbcL and rbcS sequences group the chromophytes with red algae and cryptomonads (Douglas et al., 1991a; Morden and Golden, 1991). The other major cluster contains all the green or chlorophyll b-containing chloroplasts (i.e. euglenoids, green algae and the land plants; Douglas et al., 1991a; Morden and Golden, 1991). To explain this dichotomy of rbcL and rbcS sequences, it has been postulated that chloroplasts have polyphyletic origins. The chlorophyll b line of chloroplasts is suggested to derive from a cyanobacterial ancestor, while the red algal, cryptomonad and chromophyte chloroplasts would seem to derive from P-purple bacteria, as represented by the chemolithotroph Alcaligenes eutrophus (Boczar et al., 1989; Douglas et al., 1991a; Morden and Golden, 1991). In Alcaligenes eutrophus (formerly known as Hydrogenomonas), carbon fixation is driven by oxidation of molecular hydrogen. Alcaligenes eutrophus lacks a photosystem, so postulating this bacterium as a progenitor for chloroplasts in cryptomonads, red algae and chromophytes implies that oxygenic photosynthesis has evolved twice. Also indicated would be the parallel evolution of chlorophyll a and phycobiliproteins, since these pigments are not known in P-purple bacteria. The other, seemingly unlikely, alternative would be that a common ancestor of p-purple bacteria and cyanobacteria possessed oxygenic photosynthesis, and that after establishment of an endosymbiotic relationship leading to the non-chlorophyll b line, remaining P-purple bacteria underwent secondary loss of the ability to use water as an electron donor. 4. Chloroplast rRNA and protein gene phylogeny Phylogenies based on chloroplast DNA sequences other than rbcLS do not invariably support the polyphyletic origin of chloroplasts. Sequence of the quinone-binding, photosystem 11-associated D1 protein (psbA) from a red alga has a high level of positional identity when compared to psbA of land plants, but also contains a seven amino acid insertion at the carboxyl terminus that is characteristic of cyanobacteria (Maid et al., 1990). Phylogenies inferred from chloroplast 16s rRNA gene sequences do not suggest a @-purpleprogenitor for any chloroplasts (Douglas and Turner, 1991). Rather, all chloroplasts are monophyletic and have a cyanobacterium as a sister group (Douglas and Turner, 1991). There is, however, a deep divergence in the chloroplast lineage separating red algal, cryptomonad, Euglena and chromophyte chloroplasts into one clade, and chlorophyll 6-containing chloroplasts (with the exception of Euglena) into another. The

EVOLUTION OF CRYPTOMONAD ALGAE

213

discrepancy between the phylogeny inferred from rbcL sequences and that inferred from 16s rRNA data remains unresolved. If the 16s rRNA phylogeny is correct, the similarities between rbcL sequences of Alcaligenes eutrophus and non-chlorophyll 6-containing chloroplasts could be explained in several ways: (1) displacement of the cyanobacterial endosymbiont’s rbcLS operon with an Alcaligenes-like operon by horizontal gene transfer before, during, or after the establishment of the endosymbiosis; (2) existence of an unknown prokaryote with a photosystem, phycobilins, and an A . eutrophus-like rbcLS operon; (3) convergent evolution of the rbcLS sequences in A. eutrophus and non-chlorophyll b-containing chloroplasts; (4) the possibility that all plastids derive from a cyanobacterium with two different rbcL genes, a different one of which was subsequently lost in the two divergent lines; (5) that rbcL in rhodophytes and chromophytes is derived from the mitochondrion (proposed to be derived from a purple bacterium) by interorganelle transfer; and (6) failure of the rbcLS-based trees to predict the correct phylogeny of bacteria and chloroplasts, perhaps due to biases in A T content of the genomes examined (Lockhart et al., 1992). It should also be noted that an alternative phylogeny, inferred from structure function and dispersion of Rubisco, places the chemosynthetic bacteria as descendants of cyanobacteria and a sister group of chloroplasts (McFadden et al., 1986).

+

V.

CRYPTOMONADS AS ENDOSYMBIONTS, PARASITES OF CRYPTOMONADS AND ENDOSYMBIONTS OF CRYPTOMONADS So, naturalists observe, a flea Hath smaller fleas that on him prey And these have smaller fleas to bite ’em And so proceed ad infiniturn

Jonathan Swift (1733) Cryptomonads exist as endosymbionts within the cells of ciliates and dinoflagellates. The gymnostome ciliate Myrionecta rubra (formerly Mesodinium rubrum or Cyclotyrichium meunieri), which commonly causes nontoxic red tides, contains numerous cryptomonads (Hibberd, 1977; Oakley and Taylor, 1978). Curiously, the cryptomonad endosymbiont is sometimes subdivided into separate membrane-bound compartments in the ciliate. In one type of compartment are found several chloroplasts, each with its

214

G. I . McFADDEN

periplastidal space and nucleomorph, as well as the mitochondria and surrounding cytoplasm presumably derived from the main cryptomonad cytoplasm. The main cryptomonad nucleus along with more mitochondria is found in another type of compartment. Both types of compartment have a single smooth membrane dividing the endosymbiont cytoplasm from the host cytoplasm. In some places, a layer of host rough E R overlays the dividing membrane. Exactly how the chloroplasts and motochondria are maintained in a compartment remote from their nucleus is a mystery. The cryptomonad appears to be a permanent endosymbiont and has lost its flagellar apparatus, ejectisomes, and periplast. Several dinoflagellates have been shown to contain cryptomonads (Wilcox and Wedermayer, 1984, 1985; Larsen, 1988; Schnepf and Elbrachter, 1988; Schnepf etal., 1989; Farmer and Roberts, 1990). That the cryptomonads are true endosymbionts, and not merely incompletely digested prey cells, has not been unequivocally established (Larsen, 1988; Fields and Rhodes, 1991). The dinoflagellates involved feed by myzocytosis, a process in which the dinoflagellate seems to suck out the contents of the prey cell through a tube known as the peduncle (Schnepf and Elbrachter, 1988). The cryptomonad “endosymbionts” in the dinoflagellates frequently lack their nucleus and sometimes only the chloroplast is found in the dinoflagellates, and it is possible that the dinoflagellate sometimes fails to ingest all the organelles of the prey cell. Chloroplasts obtained in this manner have been dubbed “cleptochloroplasts” (Schnepf et a f . , 1989) and it is possible that they represent an early stage in the establishment of a true endosyrnbiont-derived organelle. A kinetoplast flagellate known as Proteromonas steinii parasitizes several species of Cryptomonas (Ettl and Moestrup, 1980; Ettl, 1984). The parasite occupies up to half the cryptomonad cell and is released either as a cyst or numerous swarmers when the host cell eventually dies (Ettl and Moestrup, 1980). Some chloroplast-containing cryptomonads can contain bacterial endosymbionts. The bacteria occur in the main cytoplasmic compartment and can be in a peribacterial membrane, a specialized vacuole, or free in the cytoplasm (Kugrens and Lee, 1990; Schnepf and Melkonian, 1990). The bacterial endosymbionts contain bacteriophage-like particles (Schnepf and Melkonian, 1990).

VI. SECOND-HAND CHLOROPLASTS IN OTHER ALGAE Do any other algae have second-hand chloroplasts? Indirect acquisition of photosynthetic capacity has been proposed for euglenoids, dinoflagellates, chromophytes and Chforarachnion.A variety of evolutionary schemes postulate photosynthetic eukaryotic endosymbionts as chloroplast progenitors

EVOLUTION OF CRYPTOMONAD ALGAE

215

in these algae (see Gibbs, 1978, 1981a, 1985, 1990; Dodge, 1979; Whatley and Whatley, 1981; Sitte, 1990). The chloroplasts of these algae have three, or sometimes four, membranes surrounding them, and these extra membranes have been proposed to represent either the plasma membrane of the endosymbiont o r the phagocytotic membrane of the host, or both in the case of those chloroplasts having four limiting membranes (Fig. 2). While the hypotheses for chromophyte and dinoflagellate chloroplast origins are still highly speculative, the chloroplasts of Chlorurachnion and the dinoflagellate Lepidodinium viridae are very likely to derive from green algae. As outlined in Section 111.D.4, the chromophyte chloroplasts are probably not derived from cryptomonad chloroplasts by loss of the nucleomorph and periplastidal ribosomes. Another explanation could be that chromophyte chloroplasts are derived from a red algal endosymbiont that subsequently lost the phycobiliproteins, which would explain the similarities in chromophyte and red algal ribulose-1,5-bisphosphate carboxylases. However, other features, including the accessory pigments and storage products in chromophytes, are dissimilar to red algae, so the origin of the chromophyte chloroplast remains uncertain. Dinoflagellates typically have three membranes delimiting the chloroplast and possibly acquired their chloroplasts from a chromophyte (Tomas and Cox, 1973; Gibbs, 1981a)-a case of a “hand-me-down” chloroplast. The euglenoid chloroplast shares certain features, including chlorophyll b , with green algae (Palmer, 1985), but sequences of nuclear-encoded 16S-like rRNA ally the euglenoids with kinetoplasts (parasitic flagellates including trypanosomes, Crithidia and leishmanias; Wolters, 1991; see also Fig. 10). Euglenoids may be flagellates that engulfed either a green algal cell (Gibbs, 1978, 1985, 1990), or some other eukaryotic alga (Douglas and Turner, 1991), which is now reduced to a chloroplast with three membranes. Chlorarachnion and the dinoflagellate Lepidodinium viridae almost certainly contain a eukaryotic endosymbiont, as they both contain a nucleuslike structure analogous to the cryptomonad nucleomorph. Chlorarachnion is a reticulopodial amoeba that contains several green chloroplasts containing chlorophylls a and b (Hatakeyama et al., 1991). Like cryptomonads, the chloroplasts of Chlorarachnion are surrounded by four membranes and between the inner and outer pairs is a periplastidal space with eukaryoticsize ribosomes and a nucleus-like structure (Hibberd and Norris, 1984) that contains DNA (Ludwig and Gibbs, 1987, 1989b). The Chlorarachnion nucleomorph apparently encodes rRNAs (McFadden, 1989), but sequences are not yet available. It seems most likely that Chlorarachnion acquired its chloroplast by engulfing a green alga (possibly a prasinophyte), and that the nucleomorph is the homologue of the green algal nucleus. It will be most interesting to determine the coding function of the Chlorarachnion nucleomorph DNA and compare the location of chloroplast protein genes in Chlorarachnion with the green algae and land plants.

216

G. I. McFADDEN

The dinoflagellate Lepidodinium viridae is different from typical dinoflagellates in that it contains green chloroplasts with chlorophylls a and b (Watanabe et al., 1987). Like cryptomonads and Chlorarachnion, the Lepidodinium viridae chloroplasts have four membranes with a periplastidal space and a nucleus-like organelle surrounded by ribosome-like particles (Watanabe et al., 1987). The presence of DNA in the nucleus-like structure has not been demonstrated but Lepidodinium viridae seems likely to contain a reduced green alga. From these preliminary observations it would seem that conversion of phagotrophic protists to an autotrophic life-style by engulfing and retaining photosynthetic eukaryotes may have occurred several times during evolution. Protists would seem to have a propensity for “kidnapping” autotrophic cells and putting them to work as slaves (Cavalier-Smith and Lee, 1985), or, conversely, algae may have a propensity for invading phagotrophs and remaining permanently inside the host cell. Either way, distantly related protists seem to have acquired chloroplasts of the same type by harbouring closely related eukaryotic endosymbionts. It is therefore important to distinguish between the original evolution of chloroplasts and their subsequent dissemination among a variety of protists by secondary endosymbioses.

VII. ROLE OF THE NUCLEOMORPH What is the role of the nucleomorph and why does it persist? The nucleomorph is situated in a subcompartment of the cryptomonad cell and is surrounded by eukaryotic ribosomes. Preliminary indications are that nucleomorph DNA encodes rRNAs for the surrounding ribosomes (see Section 111) and it is likely that nucleomorph DNA could encode ribosomal proteins and housekeeping genes. But what is the ultimate purpose of this second set of eukaryotic translation machinery in the periplastidal space? Presumably, the nucleomorph DNA encodes proteins essential to the chloroplast. The chloroplast chromosome has the capacity to encode roughly 100 protein genes, so the many hundreds of other chloroplast proteins must be encoded elsewhere-either in the nucleus or nucleomorph. Only one cryptomonad chloroplast protein, an a-subunit of phycoerythrin (cpeA), has thus far been demonstrated to be encoded outside the chloroplast. In red algae and cyanobacteria cpeA is usually located on the chloroplast chromosome adjacent cpeB (the P-subunit of phycoerythrin), but in cryptomonads the cpeA gene is apparently absent from the chloroplast genome (Reith and Douglas, 1990). A cpeA gene isolated from a library of total cryptomonad DNA (Jenkins et al., 1990) contains features typical of a “nuclear” gene (CAT and TATA boxes, polyadenylation site, 66% G + C content, and an N-terminal transit sequence), but it has not yet been

EVOLUTION OF CRYPTOMONAD ALGAE

217

determined whether cpeA resides in the nucleus or in the nucleomorph of the cryptomonad, or even in both. Available evidence suggests that the cryptomonad chloroplast originated from a red alga, so it might be expected that the nucleomorph encodes many of the same chloroplast proteins as does the red algal nucleus. In modern day red algae the nucleus presumably encodes the majority of chloroplast proteins, but thus far only the genes for phycobilisome linker polypeptides have been demonstrated to be nuclear-encoded (Egelhoff and Grossman, 1983; Grossman et al., 1986). Since cryptomonads lack phycobilisomes, they probably lack these genes altogether. Other likely candidates for nucleomorph-encoded chloroplast proteins might be ferredoxin, cytochrome and chlorophyll alc binding protein, since these proteins, or their homologues, are usually nuclear-encoded in plants and several different algae. Clearly the nucleomorph has been retained by all the chloroplastcontaining cryptomonads. Several herbivorous molluscs and planktonic ciliates are known to sequester algal chloroplasts (Hinde, 1983; Stoecker et al., 1987). The chloroplasts carry out photosynthesis for a limited period but eventually degenerate, probably through lack of essential components normally produced by the nucleocytoplasm of the alga. The molluscs periodically renew the chloroplasts by further grazing (Hinde, 1983). The cryptomonad flagellate host would seem to have obviated the need to renew its chloroplasts by retaining the algal nucleus in the form of the nucleomorph. In addition to providing some essential component(s) for the biogenesis or maintenance of the chloroplasts, the nucleomorph may have a pivotal role in the endosymbiotic relationship. The nucleomorph and periplastidal cytoplasm are located between the primary producer (the chloroplast) and the consumer (the host flagellate cell). Significantly, this intermediary zone is where excess photosynthate is stored (as starch in the periplastidal space), and screening of a nucleomorph library for starch synthatases and hydrolases may be a fruitful exercise. It is also noteworthy that the cryptomonad nucleomorph is invariably closely associated with the pyrenoid (Figs. 3 and 6), even being encased within the latter in some species (Figs. 11-13). This association even extends to the nucleomorph of Chlorarachnion reptans, which is also situated in a notch of the pyrenoid (Hibberd and Norris, 1984). The role of pyrenoids is not clear, but they are usually close to the site of starch deposition, perhaps further implicating the nucleomorph in starch metabolism. In the unicellular red alga Rhodella, a protrusion of the nucleus invades the pyrenoid (Evans, 1970; Patrone et al., 1991), so perhaps the pyrenoid/ nucleus association in cryptomonads is carried over from the eukaryotic endosymbiont. The Rhodella pyrenoid appears to be involved in mitosis, apparently encasing the dividing nucleus. The cryptomonad nucleomorph

218

G . I. McFADDEN

might be sited near the pyrenoid to ensure that the daughter nucleomorphs are segregated with the two halves of the chloroplast during division. The periplastidal compartment probably serves as a transit zone, through which there is bidirectional traffic of a variety of substances. Photosynthate must go from the chloroplast to the host; numerous metabolites (including ATP) must be conveyed from the mitochondrial-containing host compartment to the periplastidal compartment and chloroplast. All of this traffic must also cross the two pairs of membranes that separate the three compartments. Gibbs (1979, 1981b) proposed that the ribosomes on the CER surface might translate proteins into the lumen between the CER membranes, and that vesicles might convey molecules across the periplastidal space to the chloroplast. According to this hypothesis, nuclearencoded chloroplast proteins in cryptomonads should bear an N-terminal signal peptide rather than a transit peptide. Nucleomorph-encoded chloroplast proteins, on the other hand, would be expected to have a transit peptide. Sequencing of nuclear-encoded chlorophyll alc fucoxanthin binding proteins (fcp) of diatoms, which have a CER but no nucleomorph or periplastidal ribosomes, reveals the presence of a possible N-terminal signal peptide (Grassman et al., 1990). In vitro studies with the diatom fcp clones should establish whether these chloroplast proteins are first directed into the lumen of the E R en route to the chloroplast. The nucleomorph contains only 660kb of DNA and must be much reduced in comparison to the original red algal nucleus. Similarly, the periplastidal space is estimated to perform a mere 3% of the translation in the cell (Sitte and Baltes, 1989). Since many functions would be duplicated in a cell with two nucleocytoplasmic compartments, it is reasonable to assume that the nucleomorph has lost many redundant genesmitochondrial proteins are an obvious case. It is also possible that nucleomorph genes have been transferred to the main nucleus.

VIII.

SUMMARY

Cryptomonads are apparently chimaeras derived from a phagotrophic flagellate and a eukaryotic, chloroplast-containing cell. The cell contains the genomes of at least four different evolutionary lineages: mitochondrial DNA (probably from a purple non-sulphur eubacterium), chloroplast DNA (from a photosynthetic eubacterium?), main nucleus (phagotrophic flagellate), and nucleomorph (red algal nucleus?). The identity of the phagotrophic flagellate that engulfed the endosymbiont is unknown. Similarities in flagellar apparatus between cryptomonads and trichomonads (parasitic flagellates) have been noted (Roberts et al., 1981), but phylogenetic trees based on 23S-like rRNA sequences do not ally cryptomonads and Trichomonas (Perasso et al., 1989).

EVOLUTION OF CRYPTOMONAD ALGAE

219

The eukaryotic endosymbiont is most likely to have been a red alga, as first proposed several years ago (Dodge, 1979; Whatley et a[., 1979). This conclusion is based on several lines of evidence: (1) the presence of phycobiliproteins in both cryptomonads and red algae; (2) the similarities of cryptomonad and red algal rbcLS sequences; (3) the similarity of cryptomonad and red algal chloroplast 16s rRNA sequences; (4) the similarity of the nucleomorph 16s-like rRNA gene sequence to the nucleocytoplasmic 16s-like rRNA of red algae; and ( 5 ) the presence of starch in the cryptomonad periplastidal space and starch in the cytoplasm or red algae. The identity of the primary prokaryotic endosymbiont that gave rise to the chloroplast of the red algae is undetermined but seems most likely to have been a cyanobacterium. Further study of red algae and cryptomonads will hopefully reveal whether the chloroplasts of red algae, cryptomonads and chromophytes are derived from the same primary cyanobacterial endosymbiont as the chloroplasts of green algae and land plants. The endosymbiosis between cryptomonad and red alga probably developed from a predator-prey relationship. Digestion of the prey cells by the flagellate was postponed in order to allow the engulfed cell to photosynthesize. By holding the photosynthetic cell captive, the phagotroph was able to benefit from the autotrophic ability of its prey. Eventually, a permanent endosymbiotic relationship was established. The endosymbiont’s mitochondria became redundant and were lost. Its nuclear genome became much reduced, perhaps through transfer of genes to the nucleus of the host cell. The diversity and distribution of cryptomonads suggest that the endosymbiosis was not recently established, although it apparently post-dates the divergence of the red algal lineage.

IX. TAXONOMIC APPENDIX Several of the cryptomonads used in laboratory studies reviewed here have been reclassified. The strain designated as Cryptomonas 0 has been described in the new genus Guillardia as Guillardia theta D. Hill et Wetherbee (Hill and Wetherbee, 1989a). The strain designated Cryptomonas (D remains undescribed but is considered an ally of species in the genus Teleaulax (Hill, 1991b). The species first described as Cryptomonas salina Wislouch and subsequently moved to Chroomonas salina (Wislouch) Butcher, and then to Pyrenomonas salina (Wislouch) Santore has been transferred to Rhodomonas salina (Wislouch) Hill et Wetherbee (Hill and Wetherbee, 1989b). Similarly, Cryptomonas maculata Butcher and Cryptomonas abbreviata Butcher, which were transferred to Pyrenomonas as Pyrenomonas maculata (Butcher) Santore and Pyrenomonas abbreviata (Butcher) Santore are now known as Rhodomonas maculata Butcher ex Hill et Wetherbee and Rhodomonas abbreviata Butcher ex Hill et Wetherbee

220

G . I. McFADDEN

(Hill and Wetherbee, 1989b). Chroomonas caudata Geitler has been transferred to the new genus Komma to become Komma caudata (Geitler) Hill (Hill, 1991~).The chloroplast-lacking Cyathomonas is considered synonymous with Goniomonas (Hill, 1991a).

ACKNOWLEDGEMENTS I am grateful to Professor Adrienne Clarke for placing the resources of the Plant Cell Biology Research Centre at my disposal. Dr Martha Ludwig kindly read the manuscript. D r David Hill provided the algal cultures, Figs. 4 and 11, and was a “mine of wisdom” regarding cryptomonads. Mr Greg Adcock isolated the pyrenoid/nucleomorph complexes and made the slot blot in Fig. 14. The work was supported by a grant from the Australian Research Council.

NOTE ADDED IN PROOF Definitive evidence that the red algal-like rRNA gene of cryptomonads is located in the nucleomorph was recently published. Using D N A from isolated nucleomorphs in PCR experiments, Maier et al. (1991) were able to amplify and clone the endosymbiont 16s-like rRNA gene of Pyrenomonas salina. Phylogenetic trees incorporating the endosymbiont gene sequence indicate that it is related to the red algae. As final proof that cryptomonads are composed of one eukaryotic cell operating inside another eukaryotic cell, it is now necessary to demonstrate that the nucleomorph gene encodes rRNAs for ribosomes in the periplastidal space, and that the nuclear gene encodes rRNAs for the host cytoplasm.

REFERENCES Anderson, E. (1962). A cytological study of Chilomonasparamecium with particular reference to the so-called trichocysts. Journal of Protozoology 9, 380-395. Antia, N. J., Cheng, J. Y., Foyle, R. A. J. and Percival, E. (1979). Marine cryptomonad starch from autolysis of glycerol-grown Chroomonas salina. Journal of Phycology 15,57-62. Appels, R. and Honeycutt, R. L. (1986). rDNA: evolution over a billion years. In “DNA Systematics” (S. K . Dutta, ed.), pp. 81-135. CRC Press, Boca Raton, FL. Ariztia, E. V . , Anderson, R. A. and Sogin, M. L. (1991). A new phylogeny for chromophyte algae using 16s-like rRNA sequences from Mallomonas papillosa (Synurophyceae) and Tribonema aequale (Xanthophyceae).Journal of Phycology 27,428-436. Bhattacharya, D. and Druel, L. D. (1988). Phylogenetic comparisons of the small subunit ribosomal DNA sequence of.Costaria costafa (Phaeophyta) with those

EVOLUTION OF CRYPTOMONAD ALGAE

221

of other algae, vascular plants, and Oomycetes. Journal of Phycology 24, 539-543.

Bhattacharya, D., Stickel, S. K. and Sogin, M. L. (1991). Molecular phylogenetic analysis of actin geneic regions from Achlya bisexualis (Oomycota) and Costaria costata (Chromophyta). Journal of Molecular Evolution 33, 525536.

Beakes, G. W. (1989) Oomycete fungi: their phylogeny and relationship to chromophyte algae. In “The Chromophyte Algae: Problems and Perspectives”, The Systematics Society Association Special Volume No. 38 (J. C. Green, B. S. C. Leadbeater and W. L. Diver, eds), pp. 325-342. Clarendon Press, Oxford. Bendayan, M. (1981). Ultrastructural localization of nucleic acids by use of enzymegold complexes. Journal of Histochemistry and Cytochemistry 29, 531-541. Boczar,B. A., Delaney,T. P. andCattolico,R. A. (1989). Genefortheribulose-1,5bisphosphate carboxylase small subunit protein of the marine chromophyte Olkthodiscus luteus is similar to a chemoautotrophic bacterium. Proceedings of the National Academy of Sciences, USA 86, 499H999. Bogorad, L. (1975). Evolution of organelles and eukaryotic genomes. Separation of genes for chloroplast ribosomes in two genomes suggest principles of organelle biology. Science 188, 891-898. Bogorad, L. (1982). Regulation of intracellular gene flow in the evolution of eukaryotic genomes. In “On the Origins of Chloroplasts” (J. A. Schiff, ed.), pp. 277-295. Elsevier/North Holland, New York. Cattolico, R. A. (1986). Chloroplast evolution in algae and land plants. Trends in Ecology and Evolution 1, 64-67. Cattolico, R. A., and Loiseaux-de Goer, S. (1989). Analysis of chloroplast evolution and phylogeny: a molecular approach. In “The Chromophyte Algae: Problems and Perspectives”, The Systematics Society Association Special Volume No. 38 (J. C. Green, B. S. C. Leadbeater and W. L. Diver, eds), pp. 85-100. Clarendon Press, Oxford. Cavalier-Smith, T. (1982). The origin of plastids. Biological Journal of the Linnean Society 17, 289-306. Cavalier-Smith, T. (1986). The kingdom Chromista: origin and systematics. In “Progress in Phycological Research” (F. Round and D. J. Chapman, eds), Vol. 4, pp. 309-347. Biopress Ltd., Bristol. Cavalier-Smith, T. (1987). The simultaneous symbiotic origin of mitochondria, chloroplasts, and microbodies. Annals of the New York Academy of Sciences 53, 55-71.

Cavalier-Smith, T. (1989). The kingdom Chromista. In “The Chromophyte Algae: Problems and Perspectives”, The Systematics Society Association Special Volume No. 38 (J. C. Green, B. S. C. Leadbeater and W. L. Diver, eds), pp. 381408, Clarendon Press, Oxford. Cavalier-Smith, T. and Lee, J. J. (1985). Protozoa as hosts for endosymbioses and the conversion of symbionts into organelles. Journal of Protozoology 32, 376379.

Chang, D. D. and Clayton, D. A. (1987). A mammalian mitochondria1 RNA processing activity contains nucleus-encoded RNA. Science 235, 1178-1 185. Coombs, J. and Greenwood, A. D . (1976). Compartmentation of the photosynthetic apparatus. In “The Intact Chloroplast”, Topics in Photosynthesis Volume 1(J. Barber, ed.), pp. 1-51. ElseviedNorth Holland Biomedical Press, Amsterdam. Craigie, J. S. (1974). Storage products. In “Algal Physiology and Biochemistry” (W. D. P. Stewart, ed.), pp. 206305. Blackwells, Oxford. Delaney, T. P. and Cattolico, R. A. (1989). Chloroplast ribosomal DNA

222

G. I. McFADDEN

organization in the chromophytic alga Olisthodiscus luteus. Current Genetics 15,221-229. Dodge, J. D. (1969). The ultrastructure of Chroomonas mesostigmatica Butcher (Cryptophyceae). Archiv fur Mikrobiologie 69, 266-280. Dodge, J. D. (1979). The phytoflagellates: fine structure and phylogeny. In “Biochemistry and Physiology of the Protozoa” (M. Levandowsky and J. M. Hutner, eds), pp. 7-57. Academic Press, New York. Douglas, S. E. (1988). Physical mapping of the plastid genome from the chlorophyll c-containing alga Cryptomonas @. Current Genetics 14, 591-598. Douglas, S. E. (1991). Unusual organization of a ribosomal protein operon in the plastid genome of Cryptomonas @: evolutionary considerations. Current Genetics 19, 289-294. Douglas, S. E. and Durnford, D. G. (1989). The small subunit of ribulose-1,5bisphosphate carboxylase is plastid-encoded in the chlorophyll c-containing alga Cryptomonas @. Plant Molecular Biology 13, 13-20. Douglas, S. E. and Durnford, D.G. (1990a). Sequence analysis of the plastid rDNA spacer region of the chlorophyll c-containing alga Cryptomonas @. DNA Sequence-Journal of DNA Sequencing and Mapping 1,55-62. Douglas, S. E. and Durnford, D. G. (1990b). Nucleotide sequence of the genes for ribosomal protein S4 and tRNAArg from the chlorophyll c-containing alga Cryptomonas @. Nucleic Acids Research 18, 1903. Douglas, S. E. and Turner, S. (1991). Molecular evidence for the origin of plastids from a cyanobacterium-like ancestor. Journal of Molecular Evolution 33, 267-273. Douglas, S. E., Durnford, D. G. and Morden, C. W. (1991a). Nucleotide sequence of the gene for the large subunit of ribulose-l,5-bisphosphatecarboxylase/ oxygenase from Cryptomonas @. : Evidence supporting the polyphyletic origin of plastids. Journal of Phycology 26, 500-508. Douglas, S. E., Murphy, C. A., Spencer, D. F. and Gray, M. W. (1991b). Molecular evidence that cryptomonad algae are evolutionary chimeras of two phylogenetically distinct unicellular eukaryotes. Nature 350, 148-151. Dwarte, D. and Vesk, M. (1982). A freeze-fracture study of cryptomonad thylakoids. Protoplasma 117, 130-141. Egelhoff, T. and Grossman, A. R. (1983). Cytoplasmic and chloroplast synthesis of phycobilisome polypeptides. Proceedings of the National Academy of Sciences, USA 80,3339-3343. Ehrenberg, C. G. (1832). Uber dei Entwicklung und Lebensdauer der Infusionthiere; nebst fernen Beitragen zu einer Vergleichung ihrer organischen System. Abhandlungen der Koniglichen Akademie der Wissenschaften in Berlin 1832, 1-154. Enea, V. and Corredor, V. (1991). The evolution of plasmodia1stage-specific rRNA genes is dominated by gene conversion. Journal of Molecular Evolution 32, 183-186. Eschbach, S . , Hofmann, J. B., Maier, U. G., Sitte, P. and Hansmann, P. (1991a). A eukaryotic genome of 660 kb: electrophoretic karyotype of nucleomorph and cell nucleus of the cryptomonad alga. Pyrenomonas salina. Nucleic Acids Research 19, 1779-1781. Eschbach, S., Wolters, J. and Sitte, P. (1991b). Primary and secondary structure of the nuclear small subunit ribosomal RNA of the cryptomonad Pyrenomonas salina as inferred from the gene sequence: evolutionary implications. Journal of Molecular Evolution 32, 247-252. Ettl, H. (1984). Uber einen intrazellularen Parasiten bei Cryptomonas (Cryptophyceae), 11. Plant Systematics and Evolution 148, 165-167.

EVOLUTION OF CRYPTOMONAD ALGAE

223

Ettl, H. and Moestrup, 8.(1980). Uber einen intrazellularen Parasiten bei Cryptomonas (Cryptophyceae),Plant Systematics and Evolution 135,211-226. Evans, L. V. (1970). Electron microscopical observation on a new red unicell, Rhodella maculata gen. nov., sp. nov. British Phycological Journal 5, 1-13. Farmer, M. A. and Roberts, K . R. (1990). Organelle loss in the endosymbiont of Gymnodinium acidotum (Dinophyceae). Protoplasma 153, 178-185. Fields, S. D. and Rhodes, R. G. (1991). Ingestion and retention of Chroomonas spp. (Cryptophyceae) by Gymnodinium acidotum (Dinophyceae). Journal of Phycology 27,525-529. Gantt, E. (1979). Phycobiliproteins of Cryptophyceae. In “Biochemistry and Physiologyof Protozoa” (M. LevandowskyandS. H. Hunter,eds),Vol. 1, pp. 121137. Academic Press, New York. Gantt, E. (1980). Photosynthetic cryptophytes. In “Phytoflagellates” (E. Cox, ed.), pp. 381-405. Elsevier, Amsterdam. Gantt, E., Edwards, M. R. and Provasoli, L. (1971). Chloroplast structure of the Cryptophyceae. Evidence for phycobiliproteins within the intrathylakoidal spaces. Journal of Cell Biology 48, 28tL290. Gibbs, S. P. (1978). The chloroplasts of Euglena may have evolved from symbiotic green algae. Canadian Journal of Botany 56,2882-2889. Gibbs, S. P. (1979). The route of entry of cytoplasmically synthesized proteins into chloroplasts of algae possessing chloroplast ER. Journal of Cell Science 35, 253-266. Gibbs, S. P. (1981a). The chloroplasts of some algal groups may have evolved from endosymbiotic eukaryotic algae. Annals of the New York Academy of Sciences 361, 193-208. Gibbs, S. P. (1981b). The chloroplast endoplasmic reticulum: structure, function, and evolutionary significance. International Review of Cytology 72, 49-99. Gibbs, S. P. (1983). The cryptomonad nucleomorph: Is it the vestigial nucleus of a eukaryotic endosymbiont? In “Endocytobiology 11. Intracellular Space as an Oligogenetic System”, Proceedings of the 2nd International Colloquium of Endocytobiology, Tubingen, Germany (H. E. A. Schenk and W. Schwemmler, eds), pp. 987-992. Walter de Gruyter, Berlin and New York. Gibbs, S. P. (1985). The chloroplasts of Euglena may have evolved from symbiotic green algae. In “The Origin of Eukaryotic Cells”, Benchmark Papers in Systematics and Evolutionary Biology, Vol. 9 (B. D. Dyer and R. Obar, eds), pp. 177-183. Van Nostrand Reinhold, New York. Gibbs, S. P. (1990). The evolution of algal chloroplasts. In “Experimental Phycology”, Vol. 1, Cell Walls and Surfaces, Reproduction and Photosynthesis (W. Wiessner, D. G. Robinson and R. C. Starr, eds), pp. 145-157. Springer-Verlag, Berlin. Gillott, M. A. (1990). Phylum Cryptophyta (Cryptomonads). In “Handbook of Protoctista” (L. Margulis, J. 0. Corliss, M. Melkonian and D. J . Chapman, eds). pp. 139-151. Jones and Bartlett, Boston. Gillott, M. A. and Gibbs, S. P. (1980). The cryptomonad nucleomorph: its ultrastructure and evolutionary significance. Journal of Phycology 16, 558-568. Giovannoni, S. J . , Turner, S . , Olsen, G . J., Barns, S . , Lane, D. J. and Pace, N. R. (1988). Evolutionary relationships among cyanobacteria and green chloroplasts. Journal of Bacteriology 170, 3584-3592. Glazer, A . N. (1982). Phycobilisomes: structure and dynamics. Annual Review of Microbiology 35, 173-198. Glazer, A. N. and Appell, G. S. (1977). A common evolutionary origin for the biliproteins of cyanobacteria, Rhodophyta, and Cryptophyta. Federation of European Microbiological Societies, Microbiology Letters 1, 113-1 16.

224

G. I . McFADDEN

Gray, M. W. (1988). Organelle origins and ribosomal RNA. Biochemistry and Cell Biology 66, 325-348. Gray, M. W. (1989). The evolutionary origin of organelles. Trends in Genetics 5 , 294-299. Gray, M. W. and Doolittle W. F. (1982). Has the endosymbiont hypothesis been proven? Microbiological Reviews 46, 1-42. Greenwood, A. D. (1974). The Cryptophyta in relation to phylogeny and photosynthesis. In “Electron Microscopy 1974” (J. V. Sanders and D. J. Goodchild, eds), pp. 566-567. Australian Academy of Sciences, Canberra. Greenwood, A. D., Griffiths, H. B. and Santore, U. J. (1977). Chloroplasts and cell compartments in Cryptophyceae. British Phycological Journal 12, 119. Grossman, A. R., Lemaux, P. G. and Conley, P. B. (1986). Regulated synthesis of phycobilisome components. Photochemistry and Photobiology 44,827-837. Grossman, A. R., Mandoori, A. and Snyder, D. (1990). Light-hawesting proteins of diatoms. Their relationship to the chlorophyll a/b binding proteins of higher plants and their mode of transport into plastids. Molecular and General Genetics 224, 91-100. Guard-Friar, D., Eisenberg, B. L., Edwards, M. and McColl, R. (1986). Immunochemistry on cryptomonad biliproteins. Plant Physiology 80, 38-42. Gunderson, J. H., Sogin, M. L., Wollett, G., Hollingdale, M., de la Cruz, V. F., Waters, A. P. and McCutchan, T. F. (1987a). Structurally distinct, stagespecific ribosomes occur in Plasmodium. Science 238, 933-937. Gunderson, J. H., Elwood, H., Ingold, A , , Kindle, K. and Sogin, M. L. (1987b). Phyletic relationships between chlorophytes, chrysophytes, and oomycetes. Proceedings of the National Academy of Sciences, USA 84, 582S5827. Hansmann, P. (1988) Ultrastructural localization of RNA in cryptomonads. Protoplasma 146, 81-88. Hansmann, P. and Eschbach, S. (1990). Isolation and preliminary characterization of the nucleus and the nucleomorph of a cryptomonad. Pyrenomonas salina. European Journal of Cell Biology 52, 373-378. Hansmann, P., Falk, H. and Sitte, P. (1985). DNA in the nucleomorph of Cryptomonas demonstrated by DAPI fluorescence. Zeitschrift fur Naturforschung 4 0 ,~933-935. Hansmann, P . , Falk, H., Scheer, U. and Sitte, P. (1986). Ultrastructural localization of DNA in two Cryptomonas species by use of a monoclonal DNA antibody. European Journal of Cell Biology 42, 152-160. Hansmann, P., Maerz, M . and Sitte, P. (1987). Investigation on genomes and nucleic acids in cryptomonads. Endocytobiosis and Cell Research 4, 289-295. Harrington, A. and Thornley, A. L. (1982). Biochemical and genetic consequences of gene transfer from endosymbiont to host genome. Journal of Molecular Evolution 18, 287-292. Hatakeyama, N., Sasa, T., Watanabe, M. M. andTakaichi, S. (1991). Structure and pigment composition of Chlorarachnion sp. Journal of Phycology 27 (Suppl.),

A154. Haxo, F. T. and Fork, D. C. (1959). Photosynthetically active accessory pigments of cryptomonads. Nature 184, 1051-1052. Hibberd, D. J. (1977). Observations on the ultrastructure of the cryptomonad endosymbiont of the red-water ciliate Mesodinium rubrum. Journal of the Marine Biological Association U K 57, 45-61. Hibberd, D. J. and Norris, R. E. (1984). Cytology and ultrastructure of Chlorarachnion reptans (Chlorarachniophyta divisio nova, Chlorarachniophyceae classis nova). Journal of Phycology 20, 310-330. Hill, D. R. A. (1991a). Diversityof heterotrophiccryptomonads. In “The Biology of

EVOLUTION OF CRYPTOMONAD ALGAE

225

Free-living Flagellates”, The Systematics Society Association Special Volume No. 43 (D. J. Patterson and J. Larsen, eds), pp. 235-240. Clarendon Press, Oxford. Hill, D. R. A. (1991b). A revised circumscription of Cryptomonas (Cryptophyceae) based on examination of Australian strains. Phycologia 30, 170-188. Hill, D. R. A. (1991~).Chroomonus and other blue-green cryptomonads. Journal of Phycology 27, 133-145. Hill, D. R. A. and Rowan, K. S. (1989). The biliproteins of the Cryptophyceae. Phycologia 28, 455-463. Hill, D. R. A. and Wetherbee, R. (1986). Proteomonas sulcata gen. et sp. nov. (Cryptophyceae), a cryptomonad with two morphologically distinct and alternating forms. Phycologia 25, 521-543. Hill, D. R. A. and Wetherbee, R. (1989a). Guillardia theta gen. et sp. nov. (Cryptophyceae). Canadian Journal of Botany 68,1873-1876. Hill, D. R. A. and Wetherbee, R. (1989b). A reappraisal of the genus Rhodomonas (Cryptophyceae). Phycologia 28, 143-158. Hinde, R. (1983). Retention of algal chloroplasts by molluscs. In “Algal SymbiosisA Continuum of Interaction Strategies” (L. J. Goff, ed.), pp. 97-107. Cambridge University Press, Cambridge and New York. Howe, C. J. and Wallace, T. P. (1990). Prediction of leader peptide cleavage sites for polypeptides of the thylakoid lumen. Nucleic Acids Research 18, 3417. Jeffrey, S. W. (1989). Chlorophyll c pigments and their distribution in the chromophyte algae. In “The Chromophyte Algae: Problems and Perspectives”, The Systematics Society Association Special Volume No. 38 (J. C. Green, B. S. C. Leadbeater and W. L. Diver, eds), pp. 13-36. Clarendon Press, Oxford. Jenkins, J . , Hiller, R. G . , Speirs, J. and Godovac-Zimmerman, J. (1990). A genomic clone encoding a cryptophyte phycoerythrin a-subunit. Evidence for three a-subunits and an N-terminal membrane transit sequence. FEBS Letters 273, 191-194. Jorgensen, R. A., Cueller, R. E., Thompson, W. F. and Kavanagh, T. A. (1987). Structure and variation in rDNA of pea. Characterization of a cloned repeat and chromosomal rDNA variants. Plant Molecular Biology 8, 3-12. Kowallik, K. (1989). Molecular aspects and phylogenetic implications of plastid genomes of certain chromophytes. In “The Chromophyte Algae: Problems and Perspectives”, The Systematics Society Association Special Volume No. 38 (J. C. Green, B. S. C. Leadbeater and W. L. Diver, eds), pp. 101-124. Clarendon Press, Oxford. Kowallik, K. (1990). Properties of chromophytic plastic genomes. Journal of Phycology 26 (Suppl.), Abstr. no. 52. Krishna Rao, K., Cammack, R. and Hall, D. 0. (1985). Evolution of light energy conversion. In “Evolution of Prokaryotes” (K. H. Schleifer and E. Stackenbrandt, eds), pp. 143-173. Academic Press, New York. Kugrens, P. and Lee, R. E . (1988). Ultrastructure of fertilization in a cryptomonad. Journal of Phycology 24, 385-393. Kugrens, P. and Lee, R. E. (1990). Ultrastructure evidence for bacterial incorporation and myxotrophy in the photosynthetic cryptomonad Chroomonas pochmanni Huber-Pestalozzi (Cryptomonadida). Journal of Protozoology 37, 263-267. Kuhshel, M. and Kowallik, K. V. (1987). The plastome of a brown alga, Dictyota dichotoma. 11. Location of structural genes coding the large subunit of ribulose-1,5-bisphosphatecarboxylasetoxygenase and for polypeptides of photosystems I and 11. Molecular and General Genetics 207, 361-368.

226

G. I. McFADDEN

Larkum, A. W. D. (1991). The evolution of chlorophylls. In “Chlorophylls” (H. Scheer, ed.), pp. 367-383. CRC Press, Boca Raton, FL. Larkum, A. W. D. and Barrett, J. (1983). Light-harvesting processes in algae. Advances in Botanical Research 10, 1-221. Larsen, J . (1988). An ultrastructural study of Amphidinium poecilochroum (Dinophyceae), a phagotrophic dinoflagellate feeding on small species of cryptophytes. Phycologia 27, 366-377. Lewis, J. and Burton, P. (1988). A study of newly excysted cells of Gonyaulax polyedra (Dinophyceae) by electron microscopy. British Phycological Journal 23. 4940. Li, N. and Cattolico, R. A. (1987). Chloroplast genome characterization in the red alga Griffithsia pacifica. Molecular and General Genetics 209, 343-351. LichtlC, C., Duval, J. C. and Lemoine, Y. (1987). Comparative biochemical, functional and ultrastructural studies of photosystem particles from a Cryptophycea: Cryptomonas rufrescens; isolation of an active phycoerythrin particle. Biochimica et Biophysica Acta 894, 76-90. Lockhart, D. J., Beanland, T. J., Howe, C.J. and Larkum, A. W. D. (1992). Sequence of Prochloron didemni atpBE and the inference of chloroplast origins. Proceedings of the National Academy of Science, USA (in press). Loiseaux-de Goer, S . , Markowicz, Y . , Dalmon, J. and Audren, H. (1988). Physical maps of the two circular plastid DNA molecules of the brown alga Pylaiella littoralis (L.) Kjellm. Current Genetics 14, 155-162. Ludwig, M. and Gibbs, S. P. (1985). DNA is present in the nucleomorph of cryptomonads: further evidence that the chloroplast evolved from a eukaryotic endosymbiont. Protoplasma 127, 9-20. Ludwig, M. and Gibbs, S . P. (1987). Are the nucleomorphs of cryptomonads and Chlorarachnion the vestigial nuclei of eukaryotic endosymbionts? Annals of the New York Academy of Sciences 501, 198-21 1. Ludwig, M. and Gibbs, S. P. (1989a). Localization of phycoerythrin at the lumenal surface of the thylakoid membrane in Rhodomonas lens. Journal of Cell Biology 108, 875-884. Ludwig, M. and Gibbs, S . P. (1989b). Evidence that the nucleomorphs of Chlorarachnion reptans (Chlorarachniophyta) are vestigal nuclei: morphology, division and DNA-DAPI fluorescence. Journal of Phycology 25, 385-394. Lynn, D. H . and Small, E. B. (1990). Phylum Ciliophora. In “Handbook of Protoctista” (L. Margulis, J. 0. Corliss, M. Melkonian and D. J. Chapman, eds), pp. 498-523. Jones and Bartlett, Boston, MA. Maid, U., Valentin, K. and Zetsche, K. (1990). The psbA-gene from a red alga resembles those from Cyanobacteria and Cyanelles. Current Genetics 17, 255-259. Maier, U-G., Hofmann, J . B., Eschbach, S . , Wolters, J. and Igloi. G. L. (1991) Demonstration of nucleomorph-encoded eukaryotic small subunit ribosomal RNA in cryptomonads. Molecular and General Genetics 230, 155-160. Manhart, J. R. and Palmer, J. D. (1990). The gain of two chloroplast tRNA introns marks the green algal ancestors of land plants. Nature 345, 26&270. Marechal-Drouard, L., Weil, J.-H. and Guillemaut, P. (1988). Import of several tRNAs from the cytoplasm into the mitochondria in bean Phaseolus vulgaris. Nucleic Acids Research 16,47774788. Margulis, L. (1981). “Symbiosis in Cell Evolution”. Freeman & Co, San Francisco, CA. McCracken, D. A. and Cain, J. R. (1981). Amylose in floridean starch. New Phytologist 88, 67-71. McCracken, D. A., Nadakavukaren, M. J. and Cain, J, R. (1980). A biochemical

EVOLUTION OF CRYPTOMONAD ALGAE

227

and ultrastructural evaluation of the taxonomic position of Glaucosphaera vacuolata Korsh. The New Phytologist 86, 39-44. McFadden, G. I . (1989). In situ hybridization in plants: from macroscopic to ultrastructural resolution. Cell Biology International Reports 13, 3-21. McFadden, G. I . (1990a). Origin of algal plastids from eukaryotic endosymbionts. In “Insitu Hybridization” (N. Harris and D. Wilkinson, eds), Society for Experimental Biology Seminar Series, pp. 143-156. Cambridge University Press, Cambridge. McFadden, G. I . (1990b). Evidence that cryptomonad chloroplasts evolved from photosynthetic eukaryotic endosymbionts. Journal of Cell Science 95, 303308. McFadden, G. I. (1991). In situ hybridization for electron microscopy: molecular cytology goes ultrastructural. In “Electron Microscopy of Plant Cells“ (J. L. Hall and C. R. Hawes, eds), pp. 219-255. Academic Press, London. McFadden, G. I . , Bonig, I. and Clarke, A. E . (1990). Double label in situ hybridization for electron microscopy. Transactions of the Royal Microscopical Society 1, 68S688. McFadden, G. I . , Cornish, E. C . , Bonig, I . and Clarke, A . E. (1988). A simple fixation and embedding method for use in hybridization histochemistry of plants. Histochemical Journal 20, 575-586. McFadden, B. A., Torrez-Ruiz, J., Daniell, H. and Sarjoni, G. (1986). Interaction, functional relations and evolution of large and small subunits in Rubisco from Prokaryota and Eukaryota. Philosophical Transactions of the Royal Society of London, B Series 313, 347-358. McKerracher, L. and Gibbs, S. P. (1981). Cell and nucleomorph division in the alga Cryptomonas. Canadian Journal of Botany 60, 2440-2452. Mereschkowsky, J . (1910). Theorie der zwei Plasmaarten als Grundlage der Symbiogenesis, einer neuen Lehre von Enstehung der Organismen. Biologisches Centralblatt 30, 353-367. Meyer, S. R. and Pienaar, R. N. (1984). Mitosis and cytokinesis in Chroomonas africana Meyer and Pienaar (Cryptophyceae). South African Journal of Botany 3, 320-330: Mignot, J. P. (1965). Etude ultrastructurale de Cyathomnonas truncata From. (Flagelle Cryptomonadine). Journal of Microscopy 4, 239-252. Moestrup, 0. (1982). Flagellar structure in algae: a review, with new observations particularly on the Chrysophyceae, Phaeophyceae (Fucophyceae), Euglenophyceae and Reckertia. Phycologia 21,427-528. Morden, C. W. and Golden, S. S. (1989). psbA genes indicate common ancestory of prochlorophytes and chloroplasts. Nature 337,382-385. Morden, C. W. and Golden, S. S. (1991) Sequence analysis and phylogenetic reconstruction of the genes encoding the large and small subunits of ribulose1,5-bisphosphate carboxylaseioxygenase from the chlorophyll b-containing prokaryote Proclorothrix hollandica. Journal of Molecular Evolution 32,379395. Morrall, S. and Greenwood, A. D. (1982). Ultrastructure of nucleomorph division in species of Cryptophyceae and its evolutionary implication. Journal of Cell Science 54, 311-328. Newman, S. M., Derocher, J. andcattolico, R. A. (1989). Analysisof chromophytic and rhodophytic ribulose-l,5-bisphosphatecarboxylase indicates extensive structural and functional similarities among evolutionarily diverse algae. Plant Physiology 91, 939-946. Oakley, B. R. and Santore, U . J. (1982). Crytophyceae: Introduction and bibliography. In “Selected Papers in Phycology 11” (J. R. Rosowski and B. C.

228

G . I. McFADDEN

Parker, eds), pp. 682-686. Phycological Society of America Book Division, Lawrence, KS. Oakley, B. R. and Taylor, F. J. R. (1978). Evidence for a new type of endosymbiotic organization in a population of the ciliate Mesodinium rubrum from British Columbia. BioSystems 10, 361-369. Pace, N. R. Olsen, G . J. and Woese, C. R. (1986). Ribosomal RNA phylogeny and the primary lines of evolutionary descent. Cell 45, 325-326. Palmer, J . D. (1985). Evolution of chloroplast and mitochondrial DNA in plants and algae. In “Molecular Evolutionary Genetics” (R. J . McIntyre, ed.), pp. 131240. Plenum Press, New York. Palmer, J.D. (1990). Contrasting modes and tempos of genome evolution in land plant organelles. Trends in Genetics 6, 115-120. Palmer, J. D. and Thompson, W. F. (1982). Chloroplast DNA rearrangements are more frequent when a large inverted repeat sequence is lost. Cell 29, 537-

550. Patrone, L. M., Broadwater, S. T. and Scott, J. L. (1991). Ultrastructure of vegetative and dividing cells of the unicellular red algae Rhodella violacea and Rhodella maculata. Journal of Phycology 27, 242-253. Percival, E. (1968). Marine algal carbohydrates. Oceanography and Marine Biology Annual Review 6, 137-161. Perasso, R . , Baroin, A. Qu, L. H., Bachellerie, J . P. and Adoutte, A. (1989). Origin of the algae. Nature 339, 142-144. Raven, P. H. (1970). A multiple origin for plastids and mitochondria. Science 169, 64 1-646. Reith, M. and Cattolico, R. A. (1986). The inverted repeat of Olisthodiscus luteus ctDNA contains genes for both subunits of RuBPcase and the 32,000 d Q3 protein: phylogenetic implications. Proceedings of the National Academy of Science, USA 88, 8599-8603. Reith, M. and Douglas, S. (1990). Localization of P-phycoerythrin to the thylakoid lumen of Cryptomonas does not involve a signal peptide. Plant Molecular Biology 15,585-592. Rhiel, R. E., Morschel, E. and Wehrmeyer, W. (1985). Correlation of pigment deprivation and ultrastructural organization of thylakoid membranes in Cryptomonas maculata following nutrient deficiency. Protoplasma 129,62-73. Roberts, K. R., Stewart, K. D. and Mattox, K . R. (1981). The flagellar apparatus of Chilomonas paramecium (Cryptophyceae) and its comparison with certain zooflageilates. Journal of Phycology 17, 159-167. Sagan, L. (1967). On the origin of mitosing cells. Journal of Theoretical Biology 14, 225-274. Saiki, R. K . , Gelfland, D. H., Stoffel, S., Scharf, S. J . , Higuchi, R., Horn, G. T., Mullis, K. B. and Erhlich, H. A. (1988). Primer-directed enzymatic amplification of DNA. Science 239,487491. Santore, U. J. (1982). The distribution of the nucleomorph in Cryptophyceae. Cell Biology International Reports 6, 1055-1063. Santore, U. J. and Greenwood, A. D. (1977). The mitochondrial complex in Cryptophyceae. Archiv fur Mikrobiologie 112, 207-218. Schnare, M. N., Heinonen, T. Y. K., Young, P. G. and Gray, M. W. (1986). A discontinuous small subunit ribosomal RNA in Tetrahymena pyriformis mitochondria. Journal of Biological Chemistry 261, 5187-5193. Schnepf, E. and Elbrachter, M. (1988). Cryptophycean-like double membranebound chloroplast in the dinoflagellate, Dinophysis Ehrenb.: evolutionary, phylogenetic and toxicological implications. Botanica Acta 101, 196203.

EVOLUTION OF CRYPTOMONAD ALGAE

229

Schnepf, E. and Melkonian, M. (1990). Bacteriophage-like particles in endocytotic bacteria of Cryptomonas (Cryptophyceae). Phycologia 29, 33S343. Schnepf, E., Winter, S. and Mollenhauer, D. (1989). Gymnodinium aeruginosum (Dinophyta):a blue-green dinoflagellate with a vestigial, anucleate, cryptophycean endosymbiont. Plant Systematics and Evolution 164 75-91. Schopf, J. W. and Walter, M. R. (1982). Origin and early evolution of cyanobacteria: the geological evidence. In “The Biology of Cyanobacteria” (N. G. Carr and B. A. Whitten, eds), pp. 543-564. Blackwell Scientific Publications, Oxford. Sepenswol, S. (1973). Leucoplast of the cryptomonad Chilomonas paramecium. Experimental Cell Research 76, 395409. Shivji, M. S. (1990). Organization of the chloroplast genome in the red alga Porphyra yezoensis. Current Genetics 19, 49-54. Sidler, W., Kumpf, B., Suter, F . , Morriset, F . , Wehrmeyer, W. and Zuber, H. (1985). Structural studies on cryptomonad biliprotein subunits. Two different a-subunits in Chroomonas phycocyanin-645 and Cryptomonas phycoerythrin545. Biological Chemistry Hoppe-Seyler 366, 233-244. Sidler, W., Nutt, H . , Kumpf, B., Zuber, H . . Wehrmeyer, W. and Brenzel, A. (1987). Structural principles of cryptomonad biliproteins and cyanobacterial phycobiliproteins. In “Progress in Photosynthesis Research” (J. Biggins, ed.), Vol. 2, pp. 233-244. Martinus Nijhoff, Dordrecht. Sitte, P. (1990). Phylogenetische Aspekte der Zellevolution. Biologische Rundschau 28, 1-18. Sitte, P. and Bakes, S. (1989). Morphometric analysis of two cryptomonads. Quantitative evaluation of fine-structural changes in an endosymbiotic system. In “Endocytobiology IV”. 4th International Colloquium on Endocytobiology and Symbiosis (P. Nardon, V. Gianinazzi-Pearson, A. M. Grenier, L. Margulis and D. c. Smith, eds), pp. 229-233. INRA, Paris. Sprinzl, M . , Hartman, T., Weber, J . , Blank, J. and Zeidler, R. (1989). Compilation of tRNA sequences and sequences of tRNA genes. Nucleic Acids Research (Suppl.) 17. rl-172. Spear-Bernstein, L. and Miller, K. R. (1985). Are the photosynthetic membranes of cryptophyte algae inside out? Protoplasma 129, 1-9. Spear-Bernstein, L. and Miller, K. R. (1987). Immunogold localization of the phycobiliprotein of a cryptophyte alga to the intrathylakoidal space. In “Progress in Photosynthesis Research” (J. Biggens, ed.), Vol. 2, pp. 309-312. Martinus Nijhoff Publishing, Dordrecht. Stoecker, D. K., Michaels, A. E. and Davis, L. J . (1987). Large proportions of marine planktonic ciliates found to contain functional chloroplasts. Nature 326,79&792. Taylor, F. J . R. (1987). An overview of the status of evolutionary cell symbiosis theories. Annals of the New York Academy of Sciences 503, 1-16. Tomas, R. N. and Cox, E . R. (1973). Observations on the symbiosis of Peridinium balticum and intracellular algal. I. Ultrastructure. Journal of Phycology 9, 304-323. Turner, S . , Burger-Wiersma, T., Giovannoni, S. J., Mur, L. R. and Pace, N. R. (1989). The relationship of a prochlorophyte Prochlorothrix hollandica to green chloroplasts. Nature 337, 38C382. Valentin, K. and Zetsche, K. (1990a). Rubisco genes indicate a close phylogenetic relation between the plastids of Chromophyta and Rhodophyta. Plant Molecular Biology 15, 575-584. Valentin, K. and Zetsche, K. (1990b). Structure of Rubisco operon from the

230

G . I. McFADDEN

unicellular red alga Cyanidium caldarium: evidence for a polyphyletic origin of the plastids. Molecular and General Genetics 22, 425-430. Walsh, M. M. and Lowe, D. R. (1985). Filamentous microfossils from the 3,500Myr-old Onverwacht Group, Barberton Mountain Land, South Africa. Nature 314,53(3-532. Watanabe, M. M., Takeda, Y . , Sasa, T., InoLye, I . , Suda, S., Sawaguchi, T. and Chihara, M. (1987). A green dinoflagellate with chlorophylls a and b: morphology, fine structure of chloroplast and chorophyll composition. Journal of Phycology 23,382-389. Weeden, N. F. (1981). Genetic and biochemical implications of the endosymbiotic origin of the chloroplast. Journal of Molecular Evolution 17, 133-139. Wehrmeyer, W. (1970). Zur feinstruktur der Chloroplasten einiger photoautotrophen Cryptophyceen. Archiv fur Mikrobiologie 71, 367-383. Wehrmeyer, W. (1983). Phycobiliproteins and the phycobiliprotein organization in the photosynthetic apparatus of cyanobacteria, red algae, and cryptophytes. In “Proteins and Nucleic Acids in Plant Systematics” (U. Jensen and D. E. Fairbrothers, eds), pp, 143-167. Springer-Verlag, Berlin. Whatley, J. M. (1989). Chromophyte chloroplasts-a polyphyletic origin? In “The Chromophyte Algae: Problems and Perspectives”, The Systematics Society Association Special Volume No. 38 (J. C. Green, B. S. C. Leadbeater and W. L. Diver, eds), pp. 125-144, Clarendon Press, Oxford. Whatley, J. M. and Whatley, F. R. (1981). Chloroplast evolution. New Phyfologisr 87,233-247. Whatley, J. M., John, P. and Whatley, F. R. (1979). From extracellular to intracellular: the establishment of mitochondria and chloroplasts. Proceedings of the Royal Society of London, Series B 204, 165-187. Wilcox, L. W. and Wedermayer, G . J. (1984). Gymnodinium acidotum Nygaard (Pyrrophyta), a dinoflagellate with an endosymbiotic cryptomonad. Journal of Phycology 20,236242. Wilcox, L. W. and Wedermayer, G. J. (1985). Dinoflagellate with blue-green chloroplasts derived from an endosymbiotic eukaryote. Science 227,192-194. Williams, D. M. (1991). Phylogenetic relationships among the Chromista: a review and preliminary analysis. Cladisitics 7, 141-156. Wolters, J. (1991). The troublesome parasites-molecular and morphological evidence that Apicomplexa belong to the dinoflagellate-ciliate clade. BioSystems 25,75433.

The Gametophyte-Sporophyte Junction in Land Plants

ROBERTO LIGRONE Dipartimento di Biologiu Vegetule, Universita di Napoli, Viu Foria 223, 1-80139 Nupoli, Italy JEFFREY G. DUCKETT School of Biological Sciences, Queen Mary and Westjield College, University of London, Mile End Road, London E l 4 N S , U K

and KAREN S. RENZAGLIA School of Biological Sciences, Box 23590A, East Tennessee State University, Johnson City, TN 37614, USA

I.

Introduction

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

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

234 235 253 275

The Taxonomic Significance of the Placenta in Bryophytes and Implications for Phylogeny . . . . . . . . . . . . . . . .

283

IV.

Pteridophytes . . . . . . . . . . . . . . . . . . . . . . .

295

V.

Seedplants . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . .

301 306 307

11.

111.

Bryophytes . . . . . . . . . . . . . A. Mosses (Bryopsida) . . . . . . B. Liverworts (Hepatopsida) . . . C. Anthocerotes (Anthocerotopsida)

Advances in Botanical Research Vol. 19 ISBN (b12-00.5919-3

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

232

Copyright 01993Academic Press Limited All rights of reproduction in any form reserved

232

R. LIGRONE ef a / .

I. INTRODUCTION The life-cycle of land plants characteristically involves an embryonic phase during which the sporophyte is associated with the parental gametophyte and has no direct contact with the substratum (Graham, 1985; Duckett and Renzaglia, 1988a). This phase is very short in lower tracheophytes, where the young sporophyte rapidly breaks free from the confines of the gametophyte and develops a root and an aerial photosynthetic system, thus becoming independent. Likewise, seed plants have a relatively short embryonic phase that terminates with seed germination after a dormant period of variable duration. By contrast, the sporophyte of bryophytes is permanently associated with the gametophyte and therefore retains structural if not functional dependence on the latter throughout its life-span. During development the embryo utilizes conspicuous amounts of organic compounds for growth and respiration. Moreover, in most cases, the embryo accumulates abundant reserves, mostly in the form of starch, proteins and lipids, that are utilized later in development. It is widely maintained that the bulk of these substances are of exogenous origin, the embryo being barely able to produce them autonomously. There is abundant evidence from physiological studies (Proctor, 1977; Courtice et a [ . , 1978;Thomas et af.,1978,1979; Browning and Gunning, 1979a,b,c; Caussin et a f . , 1983; Renault et al., 1989) that the sporophyte of bryophytes needs organic nutrients from the gametophyte for normal growth and development, although it contains chlorophyll and, notably in mosses and anthocerotes, may have well-organized photosynthetic tissues (Duckett and Renzaglia, 1988b). The same is presumably true for the embryos of pteridophytes, although no direct evidence is presently available (DeMaggio, 1963; Sheffield and Bell, 1987; Bell, 1989). The situation is more complex in seed plants, as here the reduced gametophyte grows and is fertilized within the parental sporophyte. The embryo is nourished by the gametophyte for a short time, thereafter it starts taking up nutrients directly from the old sporophyte. The dependence of the embryo on exogenously supplied nutrients is reflected by a series of morphological specializations that presumably serve to facilitate nutrient translocation. In bryophytes and most pteridophytes the embryo forms a basal or lateral organ, the foot, that penetrates the gametophyte tissue and is thought to be active in nutrient uptake. In addition, pteridophytes and seed plants possess a suspensor that orientates and keeps the embryo in intimate contact with the nutritive tissue of the gametophyte (Gifford and Foster, 1989). The suspensor is generally small and short-lived in pteridophytes. Conversely, the embryo of seed plants has a relatively large and long-lived suspensor but never forms a foot (Wardlaw, 1965; Gifford and Foster, 1989). At very early stages in development the

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

233

embryo of mosses and most liverworts forms a suspensor-like basal structure. Most frequently this consists of one or few cells and collapses with the subsequent development of the foot. The interface between the sporophyte foot and parental gametophyte in bryophytes and pteridophytes is commonly referred to as the placenta (Pate and Gunning, 1972; Gunning and Pate, 1974; Ligrone and Gambardella, 1988a,b). Generally placental cells have dense cytoplasm rich in mitochondria, endoplasmic reticulum and ribosomes. Most frequently they present a wall-membrane apparatus typical of transfer cells (Pate and Gunning, 1972; Gunning and Pate, 1974). A true placenta is lacking in seed plants, as the embryo does not form a foot, but transfer cells may be found in several different sites such as the interface between the gametophyte and its parental sporophyte, as well as in the suspensor, cotyledons, endosperm and integumentary endothelium (Gunning and Pate, 1969a, 1974; Pate and Gunning, 1972; Gunning, 1977). The formation of an embryo has been widely recognized as one of the most distinctive features of land plants, unknown in chlorophyll a bcontaining algae, with the possible exception of Coleochaete (Graham et al., 1991). Red algae are the only other group in which post-fertilization development may produce an intercalated multicellular diploid phase, the carposporophyte, that remains associated with the gametophyte by means of highly specialized placental structures (Bold and Wynne, 1985). There is no doubt, however, that red algae and land plants have but a very remote common ancestry and similarities in their life-cycles must be due to parallel evolution. The term embryophytes for land plants has therefore received increasing favour in the last years (Mishler and Churchill, 1984, 1985; Bremer, 1985; Crane, 1985; Bremer et al., 1987), although it is still debated whether, in a cladistic context, embryophytes may be considered a monophyletic group (Sluiman, 1985). The placenta has a critical role in the life-cycle of land plants as it connects and integrates two organisms subject to different environmental and genetic constraints. Regardless of whether the land plant sporophyte and its nutritional and developmental association with the gametophyte originated de novo, through a delay in zygotic meiosis (Bower, 1908), or secondarily from a free-living organism (Fritsch, 1945), one of the very first steps must have been the evolution of a placental tissue. Since then the placenta has been evolving over millions of years under the selective pressures that have produced all the extant, and extinct, groups of land plants. This review presents a comparative analysis of the placenta in land plants, including the first detailed systematic survey of bryophytes encompassing several groups not studied before, together with comments on pteridophytes, where comparative data are virtually non-existent, and on placental analogues in seed plants. Far from exhaustive, this is primarily an attempt to

+

R. LIGRONE

234

et a1

draw the attention of plant morphologists, physiologists and taxonomists to a topic of considerable relevance to the phylogeny, taxonomy and functional biology of land plants.

11.

BRYOPHYTES

The bryophytes present a particularly conspicuous placental region, probably as a consequence of the inability of their sporophyte to become autonomous. The nutrient relationships of the two generations have been reviewed recently (Ligrone and Gambardella, 1988a) and will not be discussed in detail here. Ultrastructural studies performed in recent years have revealed remarkable diversity in the organization of the placenta (Ligrone and Gambardella, 1988b). Unfortunately, the paucity of comparative data and in particular a complete absence of information for several major groups, notably the Andreaeidae and Tetraphidales among the mosses and the Jungermanniales and Metzgeriales among the liverworts, have hitherto prevented any wide-ranging conclusions. The most clear-cut feature that distinguishes the sporophytegametophyte junction in different bryophyte groups is the presence or absence of wall labyrinths in the placental cells of each generation (Tables I, I11 and V). Further distinctive attributes are the cell wall structure, notably the shape and arrangement of wall ingrowths, and certain cytoplasmic features of placental cells such as plastid morphology (Tables 11, IV and VI). Some of these characteristics may undergo changes during sporophyte development. For example, in mature sporophytes the wall labyrinth of placental transfer cells is partially or completely obliterated by deposition of new wall material (Browning and Gunning, 1979a; Gambardella, 1987; Gambardella and Ligrone, 1987). It is therefore important that different developmental stages be examined and corresponding stages be considered when comparing different species or groups. This is not an easy task; the sporophytes develop at different times of the year, according to the species, and in liverworts the phase of active sporophyte growth is usually very short. Moreover, attempts to obtain sporophytes under laboratory conditions have been successful only in a very limited number of species. In this account two main stages of sporophyte development are distinguished. Stage 1, pre-meiotic, is characterized by proliferative activity in the sporangial tissue. Stage 2, post-meiotic, is characterized by the presence of maturing spores in the still intact sporangium. The highest demand for exogenous nutrients by the sporophyte, presumably associated with maximum activity of the placenta, has been observed during the first stage of development (Proctor, 1977; Browning and Gunning, 1979b). When not otherwise indicated, the ultrastructural observations reported here refer to that stage. The new information presented in this review encompasses 17 species of mosses, 26 of liverworts and 9 of anthocerotes, with representatives from all the major orders of bryophytes (Tables I-VI).

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

235

Nomenclature for genera and species follows Corley et a f .(1981) for mosses, Grolle (1983) for liverworts, and Stotler and Crandall-Stotler (1977) or Hasegawa (1988) for anthocerotes. A. MOSSES (Bryopsida)

The first division of the zygote within the archegonium is transverse and produces an upper, epibasal cell and a lower, hypobasal cell. According to Roth (1969), in all mosses, except Sphagnum, the epibasal cell generates an elongate flattened embryo with a lenticular apical cell, whereas the hypobasal cell undergoes a few randomly orientated divisions producing a small mass of cells. Subsequently the embryo forms an intercalary meristem which is responsible for the development basipetally of the foot and the lowermost part of the seta, whereas cells differentiating acropetally from the meristem produce the upper seta and the capsule base (Crandall-Stotler, 1984). In Sphagnum the embryo does not form an intercalary meristem. As a consequence, the seta is lacking and the foot is entirely derived from the activity of the hypobasal cell (Roth, 1969). In some bryalean genera, e.g. Archidium and Ephemerum (Crandall-Stotler, 1984), the seta meristem is functional for only a few division cycles and consequently a very small foot and short seta are developed. In most cases the hypobasally derived cells usually degenerate at an early stage of sporophyte development and either disappear or in a few moss genera (Table I) form a small necrotic appendage visible at the tip of the foot (Roth, 1969; HCbant, 1975; Ligrone and Gambardella, 1988a). 1 . Andreaeidae This subclass consists of the genus Andreaea, with 50 to 100 species (Schultze-Motel, 1970), and the recently described genus Andreaeobryum, represented by the single species A . macrosporum (Steere and Murray, 1976; Murray, 1988). Anatomical characteristics of recently discovered antheridia and sporophytes of Takakia (Davison et a f . , 1989; Mcfarland et a f . , 1989; Smith, 1990) indicate that this is a primitive moss with clear relationships with the Andreaeidae, rather than a liverwort. For this reason the placental organization of Takakia is discussed in this section. The sporophyte of Andreaea consists of a capsule, a very short seta and a conical foot (Fig. 1). After fertilization the archegonial stalk proliferates to form a parenchymatous tissue surrounding the foot. Following spore maturation the sporophyte is elevated by a pseudopodium, formed by elongation of archegonial and stem cells (Roth, 1969; Murray, 1988). The foot has a parenchymatous structure, with no trace of conducting tissue. The internal cells contain abundant lipid reserves in the form of osmiophilic droplets of varying size (Fig. 2 ) . The epidermal cells are slightly larger and have denser cytoplasm with numerous evenly scattered vacuoles of small sizes (Fig. 3). In the lower part of the foot the epidermal cells exhibit prominent wall labyrinths along their outer tangential walls (Fig. 3).

TABLE I Characteristic features of the placenta in mosses Distribution of transfer cells

Distribution of wall ingrowths

Reference' Sporophyte Gametophyte

Sporophyte

Gametophpe

Time of ingmwih development

lngrowth morphology

Sporophyte Gametophyte Sporophytc Gametophyte Conducting Foot shape tissue in fwt

ANDREAEIDAE Tolioloo rcrorophyllo

16

+

Andreoco rupmrrk Hedw

1

+

Andreoeo rorhii Web. & Mohr Andreuobryum marrosponm Sleeve & B. Murr. SPHAGNmAE Sphagnumfmbnarum Wik. Sphagnum fdlaKlinggr. Sphagnum subnitow Rws & warnst. Sphagnum curprdotum Ehrh.

1 15

11 11

+ +

I

s,ngle layer O"k* 1ang.b

NIA

Early

NIA

NIA

Early

NIA

walls

Single layer outer tang. walls

Smgle layer 0"ter tang.

Labyrinth. NIA Labynnth. coarse NIA

}

Hydroids

-

Sporophyte

NIA

Early

N/A

NIA

NIA

NIA

NIA

Early

NIA

?

NIA

7

Elongate

Slightly elongate

l h c k . nameous

+ thin areas with plasmodesmara

Thick

7

Tlick and thin NIA

areas with plasmodesmata

~

BRYIDAE Polytrichales Polyrnchum formmum Hedw. Polynchum commune Hedw Polynrchum pil$e..n Hedw Dowoni? superb0 Grev. Dendroltgornchumdcndrordrs

13 13 14

+

+ +

Broth. Pogonoaun dordes P.Beauv Pqonomm neesii (C.MuU.)

10 I

Dozy Olrgontchum hercrnicum Lam.

1

+ + + + +

Atrichurn undulorum P.

1

+

1

+

&Dc

7 7

outer tang walls

Labynnth, NIA coarse

Hydroids. leptoidr

Elongate tapertog

Tetraphidales TeonphLs pellucidn Hedw

Several layem thick-walled cells

Single layer thick-

walled cells Outer tang. walls 4

loner tang walls

Early

Early

lateral in middle

Labyrinth. Shon. coarse Hydroidr maw

Elongate tapering

Bryales

Bmbaumiinsae BuxboumiapzpenBest

12

+

Diphwrumfoliorum Mohr.

1

I

Single layer thckwalled cells

2-3 inner cell layen

Beau".

Single layer outer tang Walk

Inner tang. walls t outer cell layers

Early

Early

Labynnth

Early

Early

Labyrmth less eXte"Sl"e

Labyrinth, fine Labyrinth. coarse

Hydroids

Elongate parenchyma

Elongate tapenng

Collapsed foot up

coarse

walls

Gametophyte

cells at

i"P

~

1 1

Other wall features

Archidiiicae rmerrimm Min.

Archi&--

2

+

+

outer tang. walls.

Inner tang walls

1

+

+

Single layer cuter tang. Walls

lnncr tang walls t outer aU layers

Early

1

+

-

single layer outer tang.

NIA

Ealy

1

+

+

Single layer outer tang.

Inner tang. walls + outer cell layers

Earl?

Inner tang. walls + 0"kr cell lavers

Early

Inner tang. walls + 0)utercell layers

Early

inner rang. walls f outer cell layers

Early

Inner tang. W a l k + outer cell layers

Early

Outer tang. walls t Inner cells layers Outer tang. walk inner cell layers

Inner tang walls

Early

Inner tang. walls

Early

Single layer uuter Img

Inner tang. W d I * + ""tcrlrrll layers

Fsrl)

Single laycr outer tanp walls

Inner Idag Wdll. +

Early

Smgic laycr outcr tang.

Inner 1img. Willlb + OuterlwU b y m Inner t ~ g wila .

Dicranineae Laxobryum glnucum Angstr.

W&

waus

1

+

+

Single layer OUter tang walls

1

+

Single layer O U t a tang

f

waus

I2

+

+

Single layer outer tang waus

4

i

+

Single layer outer tang W2.b

17

t

f

9

+

+

5. x

t

4

10

I

I

1

+

wall.

+

1

h

t

I

+

+ +

waur Slnglc layer OUlCT tanp.

walls

I

I

10

t

BUlbOUS

Slnglc layer outer lung. w . 1

f

""tcrccll lnycrr

+

rjrly

€ally

outcrccll layerr

Labyrinlh. hbyrinth

murc

lnncr tang. walls I ""trr ccu lilyerr

Early

lnncr tang walk I

Earl)

Hvnnincnc ""1crICcII 1aycrr

.

.-

long.

.

10

"References. I , DEW; 2, Brown and L e m o n (1985); 3, Browningand Gunning (1979a); 4, Chauhan (1990); 5. Chauhan and Lal (lWgl), 6. Ejm6 and Suirc (lW7): 7. HCbant (19751.8.La1 and Chauhan (1981):9. La1 and Narang (1985); 10, Ligrone and Cambardella (19a8a); 11. Ligrane and Renzaglia (1989);12. Ligrone el d.(1982b); 13. Maier (1%7): Maier and Maicr (1972); 15. Murray 11988); 16. Remagha d nl (1991);17. WrncLe and %hulr (1975). btang = tangential.

TABLE I1 Cytoplasmic features of the nlacental cells in mosses Plastids Sporophyte Shape ANDREAEIDAE Takakia cerarophylla Andreaea rupestris Andreaea rothii Andreaeobryum macrosporum

[

?

SPHAGNIDAE Sphagnum fimbriatum Sphagnum fallax Sphagnum subnirens

Sphagnum curpidatum

Elongate Ovoid

Membrane system

r.gr. ?

Pleomorphic p.r

Plastoglohuli Starch

++

1.g.

{

Gametophyte

{

++ ?

+

I

{

-

{

Shape

Membrane system

Pleomorphic

r.gr.

Ovoid

r.gr.

{

Cytoplasmic lipid droplets

Plastoglobuli

+++

{

Starch

++

-I ?

Sporophyte

Gametophyte

++

+ ++

++

{

?

?

?

?

?

-

Ovoid

n.gr.

+I -1

Ovoid

n.gr. p.r.

11

BRYIDAE

?

-

I

Polytrichales Polyrrichum formosum Polyrrichum commune Polyrrichum piliferum Dowsonia superba Dendroligotrichum dendroides Pogonarum aloides Pogonafum neesii Oligorrichum hercinicum Atrichum undulatum

Pleomorphic r.gr. p.r.

Tetraphidalcs Tetraphis pellucida

Ovoid

Bryales Buxbaumiincae Buxbaumia piperi Diphyscium foliosum

Pleomorphic p.1. Pleomorphic m.gr.

+

-

-

+

+I-

++

+I+]

?

++

+

I Plcomorphi Ovoid

n.gr.

+ + -

-

Pleomorphic

r.gr.

t

-

Sphcroidal Spheroidal

pr0.b. Irregular

-

+

~

+

-

+

+

-

Archidiineae Archidium tenerrimiim

Pleomorphic

m.gr.

+

-

Spheroidal

r.gr.

Dicranineae Leucohryum glaucum Dicranum majus Blindia acura Cladophascum gymnomitrioides

Pleomorphic Pleomorphic Pleomorphic Pleomorphic

p.r. m.gr. m.gr. m.gr.

+

-

Spheroidal Spheroidal Spheroidal Spheroidal

r.gr.

-

-

r.gr. n.gr. r.gr.

-

Fissidentineae Fissidens crassipes

Pleomorphic

r.gr.+ p.r.

+

Pleomorphic

n.gr.

+

Pottiineae Timmiella harhuloides Phascum cuspidarum

Pleomorphic Pleomorphic

m.gr. r.gr.

+ +

-

Spheroidal Spheroidal

r.gr. r.gr.

Pleomorphic

n.gr. +p.r.

+

-

Pleomorphic

r.gr.

Funariineae Funaria hygromerrica Physcomifriumcoorgense Physcomirrium cyathicarpurn

1

Pleomorphij n.gr.

Bryincae Bryum capillore

Pleomorphic

Aulacomniurn palusrre Plagiomnium cuspidaturn Mnium hornurn

}

+

+ -

1

+

}

Pleomorphicl r.gr.

-

+

+

}

+

++ + + +

+ + + +

+ + +

+

-

+

+

+

1 1

+

}

+

-

-

Spheroidal

r.gr.

-

+

-

+

Ovoid Pleomorphic Pleomorphic

r.gr. + p.r. m.gr. m.gr. m.gr.

+

-

Pleomorphic Plcomorphic Spheroidal

r.gr.

+ + +

+ + ++

-

-

-

-

Leucodontineae Neckera crispa

Pleomorphic

p.r.

-

Spheroidal

r.gr.

+

+

-

-

Hypnincae Brachvfhecium vehtinum Isopterygium pulchellutn

Pleomorphic Pleomorphic

p.r.

-

Spheroidal Spheroidal Elongate

r.gr. r.gr.

+

p.r.

+

+ +

-

-

-

+ -

-

n.gr. r.gr.+ p.r.

+

+

Abbreviations: m.gr., massive grana; n.gr., normal grana, like those in leaves; r.gr., rudimentary grana, 2-3 thylakoids only; pro.b., prolamellar body; p.r.. peripheral reticulum.

240

R. LIGRONE et a/.

Figs. 1-4. The gametophyte-sporophyte junction in Andreaea rothii. Fig. 1. Light micrograph, longitudinal section, showing the conical foot and the absence of conducting tissues in the short seta. Fig. 2. The foot parenchyma cells contain abundant lipid droplets. Fig. 3. Wall labyrinths along the outer tangential walls of sporophyte placental cells. Fig. 4. Gametophyte placental cells with irregular wall thickenings (arrowed). Key to abbreviations used on figures: F, foot; G , gametophyte; H , hydroids; M, mitochondria; N, nucleus; P, plastids; S, sporophyte; SE, seta.

The gametophytic cells surrounding the foot also have dense cytoplasm rich in organelles, notably mitochondria. As in all the other mosses examined in this study, these tend to be larger than those in the foot cells. The gametophyte placental cells do not form wall ingrowths but present thickened wall areas with a loose fibrillar texture, interrupted by thinner areas crossed by plasmodesmata (Fig. 4).

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

24 1

Both sporophyte and gametophyte placental cells contain numerous small plastids with a rudimentary thylakoid system and are rich in plastoglobuli. Ultrastructural details of the placenta in Andreaeobryum have yet to be described. However, major morphological differences between Andreaea and Andreaeobryurn include the presence in the latter of a well-developed seta and a “bryoid” foot deeply penetrating the gametophyte tissue, which does not form a pseudopodium (Murray, 1988). Placental transfer cells are restricted to the sporophyte (Murray, 1988). The foot of Takakia is closely similar to that of Andreaeobryum (Renzaglia et al., 1992). This elongated tapering structure (Fig. 5 ) penetrates deeply into the broadened central conducting strand of the gametophyte stem. The inner foot region comprises a broad strand of hydroids extending lengthwise in the seta, but leptoids are absent. The epidermal cells of the foot exhibit well-developed wall ingrowths on their outer tangential walls (Fig. 6). Their dense cytoplasm contains abundant mitochondria, lipid droplets and elongated plastids with rudimentary grana and numerous plastoglobuli (Fig. 6). Starch grains are absent from the sporophyte placental cells but are common in the adjacent tissues, especially the cortex of the foot and gametophyte. Transfer cells are absent in the gametophyte placental cells. These have irregularly thickened walls lined with conspicuous deposits of electron-transparent material (Fig. 7).

2. Sphagnidae This subclass consists of the single genus Sphagnum, with 100 species. The sporophyte of Sphagnum lacks a distinct seta and develops at the apex of specialized gametophytic branches that elongate after spore maturation to form a pseudopodium (Roth, 1969). The foot is bulbous and parenchymatous (Fig. 8). An ultrastructural investigation of the placenta of Sphagnum jimbriatum and S . fallax at stage 2 of sporophyte development revealed that transfer cells were lacking in both the sporophyte and gametophyte (Ligrone and Renzaglia, 1989). This finding has been confirmed in S . subnitens and S . cuspidaturn at stage 1 of sporophyte development. The placental organization is almost identical in the four species. The sporophytic placental cells protrude from the bottom of the foot into a large, mucilage-containing space that separates the two generations (Fig. 8). These cells differ from the internal parenchyma cells of the foot, having denser cytoplasm and relatively small vacuoles (Fig. 9). Moreover, they apparently contain a vast number of plastids of very small sizes and irregular shape with few or no thylakoids, numerous plastoglobules and abundant peripheral reticulum (Fig. 10). These plastids are often aggregated in groups and presumably are sectional profiles of larger, extremely pleomorphic plastids. The nucleus generally lies in a central position and may have an irregular shape. The cells walls, relatively thin and uniform, have a compact fibrillar texture (Fig. 9).

242

R. LIGRONE

el a/

T H E GAMETOPHYTE-SPOROPHYTE JUNCTION

243

On the opposite side of the placenta are several layers of small gametophytic cells (Fig. 13). These have large vacuoles and dense cytoplasm mostly gathered at the periphery and rich in mitochondria, dictyosomes and sheets of rough endoplasmic reticulum (Fig. 11). The plastids are larger and less numerous than their sporophytic counterparts, and generally exhibit well-developed grana (Fig. 12). The gametophyte placental cells characteristically have thickened walls with a loose fibrillar texture, interrupted by pits with a high frequency of plasmodesmata (Fig. 14). In S. fimbriatum these thickened walls contain distinctive tubular structures arranged radially (Ligrone and Renzaglia, 1989). The gametophytic placental tissue contains abundant intercellular spaces that are continuous with the larger placental space. As the sporophyte matures the gametophyte cells closer to the foot degenerate (Fig. 13). Starch is generally lacking in both the sporophyte and gametophyte placental cells, whereas it is abundant in the adjoining parenchyma cells of both generations.

3. Bryidae This subclass includes all peristomate mosses and is divided into two groups: the nematodontous mosses, with peristome teeth composed of whole cells; and the arthrodontous mosses, with peristome teeth composed, primarily or exclusively, of articulated and fused cell plates (Edwards, 1984; Vitt, 1984). The former group traditionally includes the orders Polytrichales and Tetraphidales, whereas the latter includes all the most advanced mosses, classified in the single order Bryales by Vitt (1984) or in several orders by others (cf. Smith, 1978; Miller, 1979; Crosby, 1980; Corley et al., 1981). Typically in the Bryidae the sporophyte foot is highly elongate, conical in shape and penetrates the gametophyte stem tissue. It is partially or completely surrounded by the vaginula, a multilayered parenchymatous sheath derived from the proliferation of archegonial cells and the underlying stem tissue (Roth, 1969). Further details of the ultrastructure of the vaginula are given in Ligrone and Gambardella (1988a). In acrocarpous mosses the sporophyte develops at the apex of the gametophyte stem, which ceases growing since apical growth terminates in the production of archegonia. In pleurocarpous mosses the sporophyte

Figs. 5 7 . The gametophyte-sporophyte junction in Takakia ceratophylla. Fig. 5. Light micrograph, longitudinal section, showing the elongate foot with a central strand of hydroids. Fig. 6. Sporophyte placenta cell showing the wall labyrinth along the outer tangential wall and elongate plastids. Fig. 7. Gametophyte placental cell with electron-transparent inner wall material (arrowed). Figs. %lo. The gametophyte-sporophyte junction in Sphagnum cuspidaturn. Fig. 8. Light micrograph, longitudinal section, showing the bulbous foot at the apex of the pseudopodium (PS). Fig. 9. Sporophyte placental cell with thin walls, numerous vacuoles and small plastids and mitochondria. Fig. 10. Pleomorphic undifferentiated plastids in a sporophyte placental cell.

244

R. LIGRONE et al.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

245

develops at the apex of short lateral branches, and the main stem grows indefinitely. In some groups, e.g. the Polytrichales, the gametophyte is acrocarpous, but the old sporophytes may be displaced laterally due to subsequent innovative growth of new shoots. Frequently in acrocarpous mosses the foot tip penetrates the gametophyte central strand. This is due to intrusive growth of the foot and parallel sporophyte-induced proliferation and differentiation of the gametophytic tissue (Roth, 1969). Penetration of the central strand by the foot, however, seems to have no taxonomical significance as differences are found even within the same genus (Roth, 1969). The foot of Bryidae has the same histological organization as the seta. When the seta contains a central strand of conducting tissue (Hebant, 1977), this is also present in the foot. The lower part of the foot, however, differs from the seta in lacking a peripheral sterome and in the presence of highly specialized epidermal cells. In the Polytrichales the conducting strand contains both hydroids and leptoids, the latter being simpler in structure than in the gametophyte (HCbant, 1977,1979). Leptoids have also been reported in the foot of Funaria (Wiencke and Schulz, 1975), but generally they seem to be lacking in Bryales (Hebant, 1977, 1979). The foot (and seta) of many groups in the Bryales lacks a central strand of conducting cells, most likely as a consequence of reduction (Crosby, 1980; Schofield, 1985). In several acrocarpous genera, e.g. Polytrichum, Bryum, Funaria and Encalypta, apoplastic continuity between the central strands of the gametophyte and sporophyte may result from the degeneration of foot-tip cells (Roth, 1969; Hebant, 1975; Ligrone and Gambardella, 1988a).

3. Polytrichales The order Polytrichales comprises 20 genera and about 300 species placed by Smith (1971) in the single family Polytrichaceae. Representatives of seven genera have been examined ultrastructurally (Table I). The polytrichaceous mosses have a typical bryoid foot of elongate conical shape that penetrates the gametophyte stem tissue to a considerable depth. A conspicuous central strand of conducting tissue is present in both the gametophyte and sporophyte. As in the seta (Htbant, 1977), a system of intercellular spaces is present in the foot parenchyma (Ligrone and Gambardella, 1988a).

Figs. 11-14. The gametophyte-sporophyte junction in Sphagnum cuspidatum (cont.). Fig. 11. Gametophyte placental cell with characteristic wall thickenings. Fig. 12. Sporophyte placental cell; plastid with well-differentiated grana. Fig. 13. Longitudinal section through the gametophyte placental cells. Note the large intercellular spaces and the degenerating cells nearer the foot (top, arrowed). Fig. 14. Gametophyte placental cell; pitted region of the wall containing numerous plasmodesmata.

246

R. LIGRONE et al.

Figs. 15 and 16. The gametophyte-sporophyte junction in Polytrichales. Fig. 15. Atrichum undulatum; showing the sporophyte wall labyrinth and thickened tangential walls in the

gametophyte placental cells. Fig. 16. Pogonatum neesii; gametophyte placental cells with greatly thickened walls.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

247

The placental organization is substantially similar to that in Andreaea, with transfer cells in the foot only. The most prominent wall labyrinths are found in the epidermal cells but a wall-membrane apparatus typical of transfer cells may also be present in the two to three outermost layers of parenchyma cells (Fig. 17). The epidermal transfer cells have dense cytoplasm with numerous mitochondria and abundant endoplasmic reticulum (Fig. 15). The plastids are small, lack starch and have a central rudimentary thylakoid system, surrounded by conspicuous peripheral reticulum (Figs. 18and 19). The internal transfer cells tend to be more vacuolate and with less dense cytoplasm (Fig. 17). Several layers of specialized gametophyte cells surround the foot in Pogonatum and Oligotrichum, whereas a single layer is present in Atrichum (Fig. 15) and Polytrichum. The gametophyte placental cells have thickened walls with irregular outlines but never form a wall labyrinth. Cell wall thickening is most pronounced in Pogonatum (Fig. 16), with a consequent strong reduction of the cell lumen. A similar situation occurs in Oligotrichum. Only the inner tangential walls of the gametophyte placental cells become thickened in Atrichum (Fig. 15), whereas in Polytrichum thickwalled cells with dense cytoplasm are intermingled with larger thin-walled cells with prominent stacks of endoplasmic reticulum. Typically, the gametophyte placental cells have dense cytoplasm with small vacuoles and abundant endoplasmic reticulum. Their plastids may have a more highly organized thylakoid system and less peripheral reticulum than their sporophytic counterparts, but, unlike those in the vaginula, they contain little or no starch. The gametophyte cells closer to the foot often undergo cytoplasmic degeneration. Like Andreaea, the polytrichaceous mosses accumulate abundant lipid reserves in both the placental cells and adjoining parenchyma cells. 4. Tetraphidales The order Tetraphidales consists of two (Crosby, 1980) or three genera (Vitt, 1984), each with three o r four species. Only one of these, Tetraphis pellucida, has been examined ultrastructurally . The foot penetrates the gametophyte central strand and has a central strand of hydroids surrounded by elongate parenchyma cells, whereas leptoids are absent (Figs. 20 and 21). A well-developed wall labyrinth is present in the epidermal cells from very early stages in sporophyte development. In the lower part of the foot the epidermal cells are highly vacuolate and the bulk of the cytoplasm lies adjacent to the labyrinthine walls, whereas in the middle of the foot the epidermal cells contain very small vacuoles and most of their lumen is occupied by dense cytoplasm rich in mitochondria1 lipid droplets (Fig. 23) and chloroplasts with highly organized grana (Fig. 24). O n the opposite side of the placenta are gametophyte cells with thickened

248

R. LIGRONE et al.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

249

walls that may form large and coarse protuberances (Fig. 22). Conspicuous wall labyrinths are occasionally found on the lateral walls of the gametophyte cells adjoining the middle portion of the foot. The plastids in the gametophyte placental cells are of smaller and more irregular shape than those in the sporophyte. Their inner membrane system is rudimentary and small amounts of starch are sometimes present. Starch and lipid reserves are much more abundant in the adjoining cells of the vaginula. 5. Bryales Divided by Vitt (1984) into 15 suborders, 85 families and 765 genera, this is the largest and most diverse group of mosses. In sharp contrast to an extreme variability in the gametophyte and sporophyte morphology, the placental organization is remarkably uniform. Twenty species, belonging to nine different suborders, have been examined ultrastructurally (Table I), and several others have been examined by light microscopy only (see Roth, 1969, for a comprehensive review). All of these, with the exception of Dicranum (Fig. 25), exhibit transfer cells on both sides of the placenta (Figs. 26, 27, 31, 32 and 34). Labyrinthine walls are generally restricted to the single layer of epidermal cells in the foot and the adjoining layer of gametophyte cells, but isolated wall ingrowths are also frequent in gametophyte cells farther from the foot. Several layers of well differentiated transfer cells have been reported in the gametophyte placenta of Buxbaumia piperi (Ligrone et al., 1982a). In most cases the foot has a regular outline and is separated from the gametophyte by a placental space containing residues of collapsed cells of gametophytic origin (Fig. 25) and sometimes calcium oxalate crystals (Fig. 27). In some instances, however, interpenetration of the sporophyte and gametophyte placental tissue may increase the contact area between the two generations. For example, the elongate epidermal cells of Bryum are wedged among the adjoining gametophyte cells (Ligrone and Gambardella, 1988a), and in Diphyscium the foot is covered with multicellular tubular outgrowths that deeply penetrate the adjoining gametophyte tissue (Figs. 28-30). In general the wall labyrinths are equally well developed in both generations (Figs. 26 and 27) but sometimes are more elaborate in the sporophyte (e.g. Mnium, Aulacomnium, Fissidens) (Figs. 32 and 34). The opposite situation accrues in Diphyscium (Fig. 31). The wall ingrowths in Figs. 17-19. The gametophyte-sporophyte junction in Polytrichales (cont.). Fig. 17. Polytrichum formosum; several layers of sporophyte transfer cells. Fig. 18. Pogonatuni aloides; sporophyte placental cell plastids. Fig. 19. Oligotrichurn hercynicum; sporophyte placental cell plastid. Figs. 20 and 21, Light micrographs of the garnetophyte-sporophyte junction in Tetraphis pellucida. Fig. 20. Transverse section. Fig. 21. Longitudinal section.

250

R. LIGRONE ei al.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

251

the placental cells of the Bryales are thinner, longer and more convoluted than in the other orders of Bryidae (Fig. 33) and presumably this results in a proportional increase of the surface extension of the wall-membrane apparatus. In Funaria the wall labyrinths develop at a very early stage of sporophyte development, reaching the maximum extension well before capsule differentiation (Wiencke and Schulz, 1978; Browning and Gunning, 1979a). The same has been observed in several other species, such as Mnium hornum (Fig. 33), Brachythecium vefutinum and Fissidens crassipes (Fig. 34). As in the other orders of mosses, the placental cells in Bryales have dense cytoplasm rich in mitochondria and generally are not highly vacuolate. The presence of numerous vacuoles of small sizes instead of large vacuoles may protect the placental tissue against water stress (Oliver and Bewley, 1984), an eventuality that might be particularly prejudicial to sporophyte development. Plastids are somewhat variable in structure, although they are generally much less differentiated compared to those in other tissues (Duckett and Renzaglia, 1988b). A relatively well-developed inner membrane system has been observed in the plastids of sporophyte placental cells of Funaria (Browning and Gunning, 1979a), Physcomitrium (La1 and Chauhan, 1981) and Archidium (Brown and Lemmon, 1985). Massive granal stacks with few or no stroma thylakoids have been found in Timmieffa (Ligrone et a f . , 1982b), Mniurn (Fig. 35), Dicranum (Fig. 37), Cfadophascumand Diphysciurn. Highly pleomorphic plastids with abundant peripheral reticulum closely similar to the saccate mitochondria1 cristae in the same cells, and few or no thylakoids occur in the sporophyte placental cells of Bryum (Fig. 36; Ligrone and Gambardella, 1988a), Buxbaumia (Ligrone et a f . , 1982a), Fissidens and Brachythecium (Ligrone and Gambardella, 1988a). Plastids in gametophyte placental cells are either larger o r smaller than those in the sporophyte and almost always contain starch (Figs. 38-41), whereas this is much less frequent in the sporophytic counterparts (Figs. 35-37). The range in the ultrastructural characteristics of these plastids is illustrated in Figs. 38-41. Shape varies from spheroidal in Cfadophascum (Fig. 38), Diphyscium (Fig. 39) and Bfindia (Fig. 41) to pleomorphic in Fissidens. The thylakoid system is also variable, appearing as irregular arrays of membranes in Diphyscium (Fig. 39), small grana in Cfadophascum (Fig. 38) and large stacks in Bfindia (Fig. 40). The parenchyma cells of the vaginula almost invariably contain large amyloplasts and often also abundant lipid reserves. This feature is most Figs. 22-24. The gametophyte-sporophyte junction in Tetruphispellucidu (cont.). Fig. 22. Gametophyte placental cells with large, coarse wall protuberances (mowed). Fig. 23. Midregion of foot; wall labyrinth o n the outer walls. Fig. 24. Chloroplast in a sporophyte placental cell.

252

R. LIGRONE et al.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

253

pronounced in the vaginula cells of Diphyscium, which are comparable to endosperm cells of seed plants in the abundance of their reserve materials (Fig. 40). Starch is also abundant in the parenchyma cells of the foot. These cells are highly elongated longitudinally and present high concentrations of plasmodesmata in the cross walls. This may reflect a functional specialization in the acropetal transport of nutrients toward the sporangium. B. LIVERWORTS (Hepatopsida)

As in mosses, in all liverworts, except the Monocleales (Campbell, 1954a; K. S. Renzaglia, unpublished data) the first division of the zygote is transverse and produces a small epibasal and a large hypobasal cell.The destiny of these two cells appears to vary widely. However, the available information about early stages in embryogeny is far from conclusive in as much as original descriptions are relatively few and sometimes conflicting. It is generally agreed that in the Jungermanniales the second division occurs in the epibasal cell and produces a three-celled embryo (Smith, 1955; Schertler, 1979; Crandall-Stotler and Geissler, 1983; Schuster, 1984a). Of the two epibasally derived cells, the outermost one is the capsule initial whereas the other will form the seta and foot. The hypobasal cell undergoes few or no further divisions, forms a filament, referred to as the huustorium, and penetrates the underlying gametophyte tissue. The function of this structure seems to be confined to very early stages in sporophyte development, after which the foot becomes the primary absorptive organ and the haustorium generally becomes indistinct (Schuster, 1984a). In the Metzgeriales, as in the Jungermanniales, the hypobasal cell forms only a haustorial appendage and the foot has a epibasal origin (Clapp, 1912; McCormick, 1914; Campbell, 1916; Showalter, 1926, 1927a,b; Haupt, 1929b; Smith, 1955; Crandall-Stotler, 1981; Schuster, 1984a). The information available on embryogeny in the Calobryales is particularly poor and inconclusive. Early studies by Goebel (1891) and Campbell (1920) report embryo development in Cufobryumbfumei to follow the same pre-determination pattern as in the Jungermanniales and Metzgeriales, with the hypobasal cell producing a haustorium and the epibasal cell giving rise to the sporophyte proper. This type of embryogeny is currently referred to as typical of the whole group (Smith, 1955; Crandall-Stotler, 1981; Schuster, 1984a). Nevertheless, more recent accounts of early embryogeny in Hupfomitrium gibbsiue Steph. (Campbell, 1954b) and Cafobrymindicurn Udar et Figs. 25-27. Bryalean plancentas. Fig. 25. Dicranum majus; the gametophyte cells are thin-walled and lack wall ingrowths. Note the collapsed cells in the intraplacental space (arrowed). Fig. 26. Leucobryum gluucum; extensive wall labyrinths in the placental cells of both generations. Fig. 27. Cladophascum gymnomitrioides; calcium oxalate crystals in the intraplacental space (arrowed).

254

R. LIGRONE et al.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

255

Chandra (Mehra and Kumar, 1990) have reported a “quadrant stage” much like that in Marchantiales (see below). Moreover, according to these descriptions, no haustorium is formed, and since the cellular divisions following the first one are irregular there is no clear-cut distinction between epi- and hypo-basally derived parts of the embryo. Mehra and Kumar (1990) suggest that in the Calobryales, as in Marchantiales (see below), a filamentous and a quadrant type of early embryogeny may co-exist. In the Sphaerocarpales both the epibasal and hypobasal cell divide transversely. This results in a four-celled filamentous embryo in which each cell, by two successive vertical divisions, produces a tier of four cells (Smith, 1955). Two different patterns of embryo development, sometimes present in closely related taxa or even within the same genus, are recognized in the Marchantiales. In several genera, including Targionia (O’Keefe, 1915), Conocephalutn (Meyer, 1929), Reboulia (Dupler, 1922), Grimaldia (Meyer, 1931), Asterella (Haupt. 1929a) and Marchesta (=Neohodsgonia H. Pears.) (Campbell, 1954c), the zygote forms a four-celled filamentous embryo as in the Sphaerocarpales. In other genera, such as Corsinia (Meyer, 1912, 1914), Marchantia (McNaught, 1929) and Preissia (Haupt, 1926), the two daughter cells arising from the zygote divide by successive vertical walls, at right angles to each other, forming a “quadrant” type of embryo. No distinct haustorium is developed. The same is observed in Riccia, where most species have a quadrant type of embryo, but in certain species a four-celled embryo of the filamentous type is formed. In any case a seta and foot are lacking in this genus and all the cells of the embryo participate in the formation of a globose capsule that remains enclosed within the gametophyte tissue (Lewis, 1906; Pagan, 1932; Smith, 1955). The origin of the different parts of the mature sporophyte in the Sphaerocarpales and Marchantiales cannot be related with certainty, particularly in the filamentous type of embryo, to the epibasal or hypobasal cell, although a hypobasal origin of the foot and seta is often assumed (Campbell, 1954a; Smith, 1955; Schuster, 1984a). An alternative interpretation of embryogeny in these groups may be neotenic suppression of the hypobasal cell, implying that the first transverse division of the zygote is equivalent to the first transverse division of the epibasal cell in the other groups. This hypothesis may account for the lack of a haustorium in Sphaerocarpales, Marchantiales and Calobryales while maintaining a unitary basic pattern of early embryogeny in the whole of mosses and liverworts. Monoclea is apparently unique in the liverworts as, according to Campbell (1954a) and recently confirmed (K. S. Renzaglia, unpublished data), the Figs. 28-32. Bryalean placentas (cont.). Figs. 28-30. Light micrographs; longitudinal sections of the gametophyte-sporophyte junction in Diphysciurn foliosum. The massive foot is covered with multicellular tubular outgrowths. Fig. 31. Longitudinal section of the placenta of Diphysciurn. Wall ingrowths are more highly developed in the gametophyte cells. Fig. 32. Mniurn hornunz: wall ingrowths are more elaborate in the sporophyte.

256

R. LIGRONE

el

al.

T H E GAMETOPHYTE-SPOROPHYTE JUNCTION

257

zygote undergoes free nuclear divisions to the 26-nucleate stage, after which partitioning produces a globose embryo. This description, however, contrasts with the illustration by Cavers (191 1) of a two-celled embryo followed by a quadrant stage, as well as with a description by Johnson (1904) of embryo development in Monoclea forsteri Hook. As in mosses, fertilization causes extensive proliferation in the gametophytic tissue underlying the embryo. The archegonial cells form a calyptra that surrounds the sporophyte until spore maturation. Additional protective structures are perianths or pseudoperianths, stem-derived structures such as perigynia, coelocaules and marsupia, or simple outgrowths of the thallus (Schuster, 1966, 1984a). In various Jungermanniales and Metzgeriales the receptacle tissue below fertilized archegonia contributes to the formation of the calyptra, that is therefore referred to as stem- or shoot-calyptra (Schuster, 1984a). 1. Jungermanniales The order Jungermanniales, the largest among the Hepatopsida with about 180 genera and 7500 species (Bold et al., 1989), comprises the “leafy liverworts”. These are also referred to as acrogynous liverworts in as much as their sporophyte develops at the apex of the main stem or lateral branches that, as in the acrocarpous mosses, cease growing. In contrast to the huge variability of gametophyte morphology, the sporophyte of the Jungermanniales is rather uniform and displays, in all instances, a distinct division into foot, seta and capsule. The foot is generally bulbous or spheroidal (Fig. 42) of relatively small size (20C300 km in diameter), and sometimes reduced to a few cells (e.g. in Lejeunea; Schuster, 1984a). In many instances a “collar” of cells develops at the junction with the seta. This probably reinforces the attachment of the sporophyte to the parental gametophyte. In some genera, e.g. Goebefobryum and Acrobolbus (Schuster, 1966), the collar is particularly bulky, and in extreme cases, as in Jackielfu (Schuster, 1984a), it consists of a massive sheath of multicellular filaments. The foot in Radulu (Figs. 53 and 54) recalls bryopsid mosses such as Bryum and Diphyscum in the presence of irregular radially elongated peripheral cells that interdigitate with the adjacent gametophytic cells. In some cases, e.g. Lejeuneu, Jubula, Frullania and Bryopteris, the foot remains surrounded by a tissue derived solely from the archegonial base, and does not penetrate the stem apex. Such a condition is probably related to the stalked structure of the archegonia in these genera (Crandall-Stotler and Guerke, 1980; Schuster, 1984a). More frequently, the foot is deeply Figs. 33 and 34. Bryalean placentas (cont.). Fig. 33. Oblique section through the thin, convoluted ingrowths in a gametophyte placental cell of Bryum cupillure. Fig. 34. The placenta in a young sporophyte of Fissidens crussipes. Note the early development of the wall labyrinths.

HerbeItaEeae Hwberra 'pp.

1

+

Lcpidoriaceae Kuriro rnchocldos Grolle Zoopsrr Irukiumw Horik

I 1

+ +

1 1

+ +

Cephalozraceae Cephpholorro brcupidnrv (Nees) Llrnpr. Cephphaloiro lunulifolm Durn

Jungerrnanniaceae Jungermannia gracrllrma Sm

1

t

Gyrnnornitriaceae Marruprlln funrkri Durn

1

+

1

I

+ +

Lophomleo hrrerophyllo (Schrad) Durn.

I

+

Radulaceae Rodulo complonrvo (L ) Durn

1

Scapantaceae Scoponio grveib Kaal Dtplophyllw, albrconr (L ) Durn.

j

GeOCalyC2Le*e

5

1

+ +

+

1 I 1

+

I

Riccordw mulnfido S Gray

~

1

1

~

+

t

~

~

~

~

Outermost fool a l l r . accarbnally

. N:A

Late

. N,A

on tareral cell,

Lahynnth. fine very dcnre

. N/A

.

-

.Bulbour

Thldtenln~01 mner laogntlal "all> auh

radially am"@ depovtr

N:A

NIA

Several l a y e r r of Ldlz

Late

Ourer tang. aod

Sereral laycn of

Late

lateral naUs

cells

Outer rang.bwalIr

,

NIA

Late

Lahynnth

Labynth

BdbO",

Coarse Late

Outertang *all5

Lahynnth

Lahynnth } Coane

Outerlang. nalk N:A N:A

N:4 N:A N:.4

NIA

NIA

NIA

X:A

N:A

NIA

NIA

NIA

W l

Ni.4

NIA

NIA

NIA

Lahynnth Coarse

Con,cal

Thm walk. inegular

plaimalernma

CALOBRYALES Colobrywn hlums Necs Hoplomrlnum hookui Neen

MARCHANTUDAE MONOCLEALES Monucleu gouxha Lmdb. SPHAEROCARPALES Sphaermwpm donne//,Aust Sphaerocarpos kzmw A w l . MARCHANTIALES Carrpineae Conpos monocorpos Prosk

+ I

+

+

+ +

+

I

I

+ +

4

TaC@O"lCiCW

Targorgronro h.vpophvllo L

+

Reboulw hemrrphnrnco Raddi

Outer

Several byerr of Fells

Outer rang and lateral walls

Several layers of cel- Late

Several layen of

+

outer tang

Several layers of d l S

+

Outer tang walls

+

} Outer

Wi1116

Late

15

tang. and lateral walls

+

} Early

}Early

cells

outerrang walls and inner parenchy-

} Labyrinth

} Labynnth }

Labyrinth

Labynnth

Transparent

Transparen1

} Late

} Labyrinth

Late

Labyrinth

Shon coarse

Late

Labyrinth Coax

Labyrioth Coarse

Late

Labyrinth

Labynnth

Late

Late

Labryinth

Labynnth

Late

Late

Labynnth

Labyrinth

} Very Late

Late

late

-

} Bulboua

-

Conlral

} Labyrinth }

} Spheroidal -

Spheroidal

Spheroidal

~

BUlhOUS

ma Monnta ondrogmo Evan,

Plogiochosmo rupesrrc Steph

1

+

+

Outer tang. wall

+

outer tang. wall

Several layers of

~

Bulboua

cells ~

Bulbous

TABLE IV Cytoplasmic features of placental cells in liverworts Plartids Fametophyte

Sporophyte Sham

Membrane system

Plastodobuli

Starch

Shaoe

Plwmorphic

r.gr.

+

+

Ellipsoidal

Spheroidal Ellipsoidal Pleomorphic

qr.

+ +

+

Cepholoria bicurpidatu Cephnloiio lunulifolio

Ellipsoidal Pleomorphic

n.gr.

Jungerrnanniaceae Jungermannia gracillima

Spheroidal Ellipsoidal

Membrane system

Plastoglobuli

Starch

Cytoplasmic lipid droplets Sporophyte Gametophyte . . .

JUNCERMANNIALES Herbertamae Herbem sp.

Ellipsoidal

Lepidoziaceae Kurrio Wchoclados Zoopsk liukiuenrk

r.gr.

Cephaloziaceae

Manuprlla funckii

Diplopkyllum albicons

Geacalycafeae Lophoeoleo heterophyllo

Ellipsoidal Ellipsoidal

*.g

+

Ellipsoidal

+

++ +

Ellipsoidal Ellipsoidal

+

n.gr.

Gymnomitriaceae Scapaniaceae Scaponzo grocilrr

+

++

+

+ +

+

Pleomorphic

n.gr.

Pleomorphic Pleomorphic

r.gr. r.gr.

Ellipsoidal Pleamarphic

n.gr.

Ellipsoidal

+

+

Elliposidal

r.gr.

Spheroidal

++

+

Elongate Pleomorphic

n.gr.

Spheroidal

Pleomorphic

n.gr.

Spheroidal

Pleomorphic

pr0.b.

Pleomorphic

Pleomorphic

qr.

Pleomorphic

Pleomorphic Pleomorphic Pleomorphic

r.gr.

+

Ellipsoidal Pleomorphic Pleomorphic

+

+

+

+

Radulaceae Rodulo complonoro

METZGERIALES Fossombroniaceae Fosrornbronio echinoto

+

Blasiaeeae B l a h pusillo

Pelliaceae Pollovicinio lyellii Pallovicinio indim Pellia endiviifolia Pellio epiphyllo A"e"raCeae Aneuro pinguk Cryprothollw mirobilir

Riccordio multifida

} }

+

*.

r.gr.

+

+

++ +

Pleomorphic Plwmorphic Plwmorphic

+

+

+

11 +

+ +

1:

+ +

+

Pleomorphic Spheroidal

'.*. r.gr

Sphered

'.

+

c

".*. '.gr.

Plmmorphic

}

".gr.

Plwrnorpbic Plmmorphic

Spheroidal

}

+

P.'.

}

".a.

Ovoid

'.*.

ELlipsoidal Plwmorpbic

r.-

Spheroidal Spheroidal Plwmorphic

spheroidal Spkroidal Plwmorphic

n.gr.

pro.b.

Discoid

0.gr.

r.*.

Elongate + Rbres Pltomorphic Plwmorphk

,.g.

Elongate Plwmorphic Plwrnorphic

Ovoid

Dismid Discoid

'.*.

+ +

+

'.*. + '.*.+

+ +

NIA

N/A

'gr. p.r.

w.

'.gr r.-n.gr

r.gr.

'.gr.

Dismid Discoid

'.*. r.g.

NIA

N/A

P.'.

Riccia sorocarpa

NIA

NIA

NIA

Abbreviations: we., oormal m a . like t h a v in leaves; r . ~ rudimentary . gram, 2-3thylaloids only; r , siogle thylaloids; pr0.b. prolameUar body; p.'. peripheral miarlum.

NIA

NIA

NIA

262

R. LIGRONE et al.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

263

embedded in the stem tissue (Fig. 42), although very little or no true penetration actually occurs, since the axial tissue surrounding the foot develops after fertilization (Schuster, 1984a). The cytology of the placental region, examined in 11 species belonging to eight different families (Table HI), is extremely uniform and highly distinctive. With the exception of Radula, elaborate wall labyrinths develop in the outermost cells of the foot. In this last genus the walls of the sporophytic placental cells are relatively thin and smooth. In the other 10 (Fig. 54) these cells produce thin and highly branched ingrowths of extremely dense wall material (Figs. 43-46). Less extensive wall labyrinths or isolated wall ingrowths are present in the parenchyma cells of the foot. No trace of wall ingrowths is observed in gametophytic placental cells of all eleven genera. The cytoplasmic organization of sporophytic placental cells is somewhat variable. In Zoopsis, Marsupella and Kurzia these cells contain one or a few large vacuoles of irregular shape and the nucleus is suspended in the centre by cytoplasmic strands connected with a peripheral layer of cytoplasm (Figs. 43 and 44). Most organelles, notably plastids and mitochondria, are associated with the wall labyrinth along the outer tangential walls. In Scapania, Diplophyllum, Lophocolea, Radula and Herberta the sporophytic cells have very small vacuoles and are rich in lipid deposits concentrated in proximity to the wall labyrinth (Figs. 45 and 46). The gametophytic placental cells, though lacking wall ingrowths, are clearly distinct from the parenchyma cells farther from the foot in having smaller vacuoles and denser cytoplasm, rich in mitochondria and other organelles. These cells form several concentric layers around the foot, the innermost of which degenerate and collapse during the phase of foot expansion (Figs. 43-45). An interesting specialization has been reported in Jubula, where the gametophyte cells within a 40pm range of the foot continue dividing as sporophyte growth proceeds to form small-celled filaments that extend towards the foot (Crandall-Stotler and Guerke, 1980). This also happens but to a more limited extent in Radula (Fig. 53). In every genus the inner tangential walls of gametophyte placental cells become thickened and may produce radially arranged deposits of dense material (Fig. 47) similar to those found in some Sphagnum species (Ligrone and Renzaglia, 1989), Andreaea, Takakia and Polytrichales. Plastid morphology in the placental cells of the Jungermanniales also is somewhat variable. The sporophyte cells contain numerous small plastids ranging in shape from spheroidal (Kurzia; Fig. 50), ellipsoidal or irregular (Herberta, Cephalozia, Zoopsis and Lophocolea; Fig. 49), to highly Figs. 35-37. Plastids in bryalean sporophyte placental cells. Fig. 35. Mniurn hornum; pleomorphic with massive grana. Fig. 36. Eryum cupillare; undifferentiated plastids intermingled with mitochondria. Fig. 37. Dicrunum mujus; ovoid with massive grana. Note the umbo-shaped mitochondria.

264

R. LIGRONE et

a1

Figs. 38-41. Plastids in bryalean gametophyte placental cells. Fig. 38. Cladophascum gymnomitrioides; starch grains surrounded by rudimentary grana. Fig. 39. Diphyscium foliosum. Fig. 40. Diphyscium foliosum; vaginula cell. Fig. 41. Blindia acuta.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

265

pleomorphic (Scapania and Diplophyllum; Figs. 51 and 52). The thylakoid system may be as extensive as in the leaf cells (Cephalotia, Lophocolea, Marsupella, Radula and Zoopsis; Figs. 46 and 49), or poorly differentiated (Scapania, Herberta and Diplophyllum; Figs. 51 and 52). Starch may be present (Lophocolea, Herberta, Cephalozia and Marsupella; Figs. 45 and 46) and is sometimes abundant (Kurzia; Figsvvvvv. 43 and 50), but in other genera (e.g. Scapania, Diplophyllum and Radula) is absent (Figs. 51 and 52). Plastids in gametophyte placental cells are generally larger than in the sporophyte, have a well developed thylakoid system and contain very little starch. The gametophyte plastids of Herberta are unusual in having abundant lipid inclusions associated with the thylakoid system (Fig. 48). 2. Mettgeriales The Metzgeriales, with about 20 genera and 550 species, are commonly referred to as the “simple thalloid liverworts”, the gametophytes lacking air chambers, air pores, pegged rhizoids and-with some exceptions, e.g. Blasia and Cavicularia (Renzaglia, 1982)-ventral scales (Bold el al., 1989). They are also called the anacrogynous liverworts, as sporophyte development normally does not terminate the growth of the gametophyte. The foot is generally conspicuous, of conical or spheroidal shape, and frequently bears a massive collar (Fig. 55). The placental region is highly variable, with all the possible combinations in the distribution of transfer cells (Table 111). In Blasia and Fossornbronia well-developed transfer cells are found in both generations (Figs. 56 and 57), generally forming several layers in the gametophyte. Transfer cells are restricted either to the sporophyte in Pallavicinia (Fig. 5 8 ) , or to the gametophyte in Riccardia (Fig. 59). The walls of the sporophytic placental cells in the latter genus are identical in appearance to their gametophytic counterparts in the Jungermanniales: both have thickened outer tangential walls containing radially arranged deposits of dense material (compare Figs. 59 and 60 with Fig. 47). In striking contrast, wall labyrinths are absent in Pellia (Fig. 61), Cryptothallus (Fig. 62) and Aneura. Here the placental cells of both generations are thin-walled save for the limited development of nacreous thickenings in the sporophytic cells of Aneura (Fig. 63). In these genera that lack wall ingrowths the plasmalemma of placental cells often presents an irregular outline along the tangential walls and may form short invaginations apparently not supported by wall material (Fig. 64). Similar invaginations are also found in parenchyma cells of the foot in Cryptothallus. In most species the wall ingrowths, when present, are short and coarse and form relatively simple wall labyrinths. By contrast, in Blasia (Fig. 57) and, to a lesser degree, in Fossombronia (Fig. 56), the sporophyte cells have highly branched wall ingrowths that form three-dimensional networks of

266

R. LIGRONE rl ul.

T H E GAMETOPHYTE-SPOROPHYTE JUNCTION

267

great complexity. Similarly elaborate wall labyrinths are common in marchantialean liverworts (see Section II.B.4), but are not found in the placenta of Jungermanniales. The placental cells also exhibit a notable diversity in their cytological organization. Generally they have dense cytoplasm rich in organelles and with numerous evenly scattered vacuoles of small sizes, but highly vacuolate placental cells are found in Pellia and Aneuru (Figs. 61 and 63). The sporophyte placental cells of Riccardia contain giant pleomorphic mitochondria, often associated with the nucleus (Fig. 65). Abundant sheets of rough endoplasmic reticulum characterize the gametophyte cells of Aneuru and Cryptothaffus(Fig. 66) and concentric sheets of endoplasmic reticulum surround lipid bodies in the sporophyte placental cells of Cryptothallus (Fig. 67). As in the Jungermanniales, the placental cells in the Metzgeriales contain numerous plastids of small sizes. In sporophytic cells, plastids with a welldeveloped thylakoid system are found in Bfusiu and Fossombroniu (Fig. 68). A less extensive thylakoid system is found in the plastids of Riccurdia (Fig. 69), and in Puffavicinia the thylakoids are associated with prolamellar body-like membranous arrays (Fig. 70). The thylakoid system is rudimentary in the sporophyte plastids of Pellia, Aneuru and in the early stages in Riccardia (Fig. 71) and Cryptothaffus(Fig. 72). At comparable stages in development, starch is absent in Fossombroniu, Blasia and Cryptothaffus,present in small amounts in Riccardia, and abundant in Aneura and Peffia.Plastids in gametophyte placental cells are much less variable and generally contain small starch grains and an inner membrane system of small grana connected by stroma thylakoids. Highly pleomorphic plastids occur in Aneura and Cryptothaffus(Fig. 73). As in other groups, the gametophyte cells closer to the foot, regardless of the presence of a wall labyrinth, degenerate precociously (Figs. 57 and 62). In most species, signs of cytoplasmic degeneration are visible in gametophyte placental cells soon after sporocyte differentiation, whereas the sporophyte placental cells show little changes until spore formation, although they show signs of cytoplasmic degeneration before the seta starts elongating. In Riccardia the wall labyrinths in gametophyte cells reach their maximum complexity after meiosis (Fig. 5 8 ) , at which time the wall thickenings develop in the sporophyte cells. A progressive increase in the thickness of sporophyte placental cell walls also occurs in Cryptothuffus along with capsule maturation. Figs. 42-44. The gametophyte-sporophyte junction in Jungermanniales. Fig. 42. Light micrograph; longitudinal section of the bulbous foot of Kurzia trichoclados. Fig. 43. Kurzia trichoclados; note the fine wall labyrinth and amyloplasts in the highly vacuolate sporophyte placental cells and collapsed gametophyte cells (arrowed) adjacent to the foot. Fig. 44. Marsupellafunckii; the cytology of the junction is virtually identical to that in Kurzia apart from the absence of starch in the sporophyte cells.

268

R. LIGRONE et ul

TH E GAMETOPHYTE-SPOROPHYTE JUNCTION

269

3. Calobryales The Calobryales are a small group of leafy liverworts that traditionally includes three genera, Haplomitrium, Calobryum and Takakia, each comprising a few species. The first two are considered as synonyms by Schuster (1984b), whereas the last one is now best placed in a separate group close to the Andreaeidae (see Section 1I.A. 1). The sporophyte of Haplomitrium and Calobryum is enclosed within a shoot calyptra until maturity and forms a massive seta terminating downwards with a large bulbous or obconical foot (Bartholomew-Began, 1991). Unlike the gametophyte stem, which contains a central strand of water-conducting dead cells (Hebant, 1977, 1979), both the seta and foot consist of homogeneous parenchyma. However, dead empty cells with electron-dense deposits associated with longitudinal walls have been observed occasionally in the foot of Calobryum blumei (Fig. 76). A wall labyrinth is present in both foot epidermal cells and adjoining gametophyte cells (Figs. 74 and 75). Smaller labyrinths or isolated wall ingrowths may occur in more peripheral gametophyte cells and in the outermost parenchyma cells of the foot. The wall ingrowths in sporophyte placental cells are generally longer and more highly branched than in the gametophyte. Placental transfer cells differentiate before proliferative divisions terminate in the capsule, that is long before the onset of meiotic division. The sporophyte placental cells seemingly contain a single, highly pleomorphic plastid in Haplomitrium (Fig. 78), whereas numerous spheroidal and pleomorphic plastids, frequently associated with the nucleus, are found in Calobryum (Fig. 77). In both instances these organelles have but a rudimentary inner membrane system consisting of small granal stacks and few or no stroma thylakoids. Gametophyte placental cells contain pleomorphic plastids with a better developed thylakoid system, characterized in Calobryum by the presence of granal stacks perpendicular to the long axis of the organelles (Fig. 79). Starch is lacking in gametophyte cells, whereas small starch grains are common in sporophyte cells of Calobryum (Fig. 77). Abundant starch deposits are found in both foot parenchyma and gametophyte cells farther from the foot. Both sporophytic and gametophytic placental cells contain numerous mitochondria, frequently in intimate association with plastids (Figs. 77 and 79). The mitochondria are distinctly larger in sporophytic than in gametophytic placental cells.

Figs. 45-47. The gametophyte-sporophyte junction in Jungermanniales (cont.). Fig. 45. Herberta sp; sporophyte transfer cells rich in lipid droplets. Collapsed garnetophyte cells are arrowed. Fig. 46. Lophocolea hererophyllu; sporophyte placental cell rich in lipid droplets. Fig. 47. Lophocolea; the gametophyte placental cell walls have tangential thickenings with radial deposits of dense material.

270

R. LIGRONE et al.

T H E GAMETOPHYTE-SPOROPHYTE JUNCTION

27 1

4. Marchantiidae The three orders Monocleales, Sphaerocarpales and Marchantiales are related by a series of morphological and developmental features, including a similar embryogeny, specialized oil body-containing cells (Schuster, 1984a) and blepharoplast morphology (Duckett et al., 1982, 1984; Brown et al., 1983; Carothers etal., 1983), and are classified together within the subclass Marchantiidae by Schuster (1984b). All three present a similar placental organization and will therefore be discussed together. The foot, inconspicuous in the Sphaerocarpales, is relatively large in most Marchantiales, and varies in shape from spheroidal (Sphaerocarpos, Targionia), bulbous (Corsinia, Lunularia, Reboulia), cup-shaped (Marchantia, Preissia) to obtuse-conical (Monoclea; Fig. 80) or conical-elongated (Conocephalum). The placenta has been studied in detail in Monoclea, two species of Sphaerocarpos and eight species of Marchantiales (Tables I11 and IV). All of these except Riccia, whose sporophyte lacks a foot, exhibit much the same organization, with well-differentiated transfer cells in both generations (Figs. 81-86). The wall labyrinths in sporophyte cells consist of highly branched and anastomosing ingrowths and attain a structural complexity not equalled elsewhere in liverworts except Blasia. In several species, e.g. Sphaerocarpos (Fig. 83), Reboulia (Fig. 854, Targionia (Gambardella, 1987), and Monoclea (Fig. 81), gametophyte transfer cells also form extensive wall labyrinths, whereas in others, e.g. Carrpos (Fig. 86), Dumortiera (Fig. 84) and Conocephalum (Ligrone and Gambardella, 1988a), they have but coarse and short wall ingrowths. As a rule, there are several layers of transfer cells in the gametophyte, the innermost ones degenerating early in sporophyte development (Fig. 84), and a single layer in the sporophyte. Small wall labyrinths or isolated wall ingrowths are frequent in the peripheral parenchyma cells of the foot in Reboulia and Conocephalum (Ligrone and Gambardella, 1988a). A study of placental development in Targionia hypophylla (Gambardella, 1987) has shown that the wall labyrinths develop after differentiation of sporocytes, i.e. much later than in mosses (Browning and Gunning, 1979a). Generally both sporophyte and gametophyte placental cells contain numerous plastids with a well-developed thylakoid system. Pleomorphic plastids with a scarcely developed inner membrane system have been observed in sporophyte placental cells of Mannia and Plagiochasma (Gambardella and de Lucia Sposito, 1981, 1983) and a rudimentary inner membrane system associated with prolamellar body-like structures occurs in

Figs. 4%5l. Plastids in jungermannialean placental cells. Fig. 48. Herbertu sp. gametophyte; ovoid and lipid bodies associated with the thylakoids. Fig. 49. Cephaloziu bicuspidata sporophyte: well developed grana. Fig. 50. Kurzia trichocludos sporophyte; ovoid amyloplasts. Fig. 51. Scuprmiu graci1i.r sporophytc; plcomorphic undifferentiated plastids.

272

R. LIGRONE et ul.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

273

sporophyte plastids of Conocephalum (Ligrone and Gambardella, 1988a). The plastid stroma in the gametophyte transfer cells of Reboulia contains a bundle of thin parallel fibres (Fig. 87; Ligrone and Gambardella, 1988a). As in Metzgeriales, the placental cells in Marchantiales contain large stacks of endoplasmic reticulum. Distinctive membrane-bound bundles of tubules are found in young gametophytic placental cells of Reboulia hemisphaerica var. macrocarpa Zodda (Zodda, 1934; Figs. 88 and 89). In crosssection the tubules exhibit hexagonal packing (Fig. 88). Prominent bundles of fibrillar material without a bounding membrane were detected in the var. macrocephala Zodda (Ligrone and Gambardella, 1988a). However, the gametophyte placenta cells in this taxon contain concentric arrays of giant cup-shaped mitochondria (Ligrone and Gambardella, 1988a). Giant pleomorphic mitochondria in intimate association with plastids are also found in the gametophyte placental cells of Carrpos (Fig. 86). Cytoplasmic degeneration starts in gametophyte placental cells during meiotic division, a process often taking several months in the Marchantiales. By contrast the sporophyte cells break down only after spore formation. As in mosses (Browning and Gunning, 1979a), the degenerating transfer cells of Marchantiales (Gambardella, 1987) and other groups obliterate their wall labyrinths by depositing new wall material in the interstices among wall ingrowths. It has been suggested that in mosses this process may help to prevent the flow of water towards the drying capsule after spore maturation (Browning and Gunning, 1979a). This is highly unlikely in liverworts, where obliteration of the wall labyrinth precedes the elongation of the seta, a process needing large amounts of water, particularly in the Jungermanniales and Metzgeriales. No transfer cells are found at the sporophyte-gametophyte junction in Riccia. The sporophyte is entirely enclosed in the gametophyte (Fig. 90) and consists of a spherical capsule with a single-layered wall, surrounded by a two-layered calyptra (Fig. 91). With capsule enlargement the cells of the inner layer of the calyptra collapse and, following spore formation, the walls of thallus cells facing the calyptra become thickened (Fig. 92). Despite the absence of a specialized placental tissue, abundant lipid reserves accumulate in the spores (Fig. 92), indicating the operation of an effective mechanism of nutrient transport towards the sporophyte. Most probably nutrients are

Fig. 52. Diplophyllum alhicans; pleomorphic plastid with a rudimentary thylakoid system in a sporophyte placental cell. Figs. 53 and 54. The gametophyte-sporophyte junction in Radula complanata. Fig. 53. Light micrograph, transverse section of the foot showing radially elongate outer cells. Fig. 54. Thin-walled sporophyte cells lacking ingrowths and thicker walled gametophyte cells. Figs. 55 and 56. The gametophyte-sporophyte junction in Metzgeriales. Fig. 55. Pellia epiphylla; light micrograph, longitudinal section showing the collar (arrowed). Fig. 56. Fossomhronia echinata a prominent labyrinth is present in both the sporophyte and gametophyte placental cells.

274

R. LIGRONE et al.

T H E GAMETOPHYTE-SPOROPHYTE JUNCTION

275

translocated across the whole sporophyte surface; the calyptra cells, that have dense cytoplasm rich in mitochondria, may play a major role in this process. C. ANTHOCEROTES (Anthocerotopsida)

The anthocerotes or hornworts comprise about 100 species that currently are classified into five or six genera, i.e. Phaeoceros, Notothylas, Folioceros, Anthoceros, Megaceros and Dendroceros (Schofield, 1985; Hasegawa, 1988). This group is distinguished from all the other embryophytes on the basis of a range of cytological, anatomical and developmental characteristics (Crandall-Stotler, 1980, 1984), and is considered by Schuster (1984~)as an independent evolutionary line of land plants. The archegonia develop from dorsal epidermal cells and at maturity protrude slightly above the thallus surface. The egg cell is bordered by cells of the thallus, in as much as an archegonial venter, in the strict sense, is lacking (Renzaglia, 1978). Unlike mosses and liverworts, the first division of the zygote is longitudinal and produces two cells that then divide transversely. A longitudinal wall perpendicular to the first one divides the embryo into four upper and four lower cells. The upper cells then divide transversely. The resulting 12-celled embryo consists of three tiers of four cells each. In Notothylas the foot arises from the lower tier only and the middle tier functions as a short-term meristem, whereas the cells of the upper tier serve as sporangial initials (Campbell, 1918; Renzaglia, 1978). In the other genera a massive bulbous foot derives from divisions in the lower two tiers, whereas the upper tier produces both a basal meristematic zone and acropetally differentiating sporangial tissues (Renzaglia, 1978; Crandall-Stotler, 1984). When embryogenesis begins, the thallus cells adjoining the archegonium divide and form an investing involucre around the sporophyte. In Notothylas the sporophyte remains enclosed in this involucre until maturation of all the spores is complete. In the other genera the sporophyte emerges from the involucre as soon as the first-formed spore mother cells complete meiosis, and continues growing for long periods because of activity of the basal meristem (Crandall-Stotler, 1984). Unlike mosses and liverworts, where the foot achieves complete cellular differentiation during specific stages in sporophyte development, in anthocerotes cellular proliferation and differentiation at the sporophyte-gametophyte junction proceeds continuously along with sporophyte growth throughout the life-span of the latter. Figs. 57-59. The gametophyte-sporophyte junction in Metzgeriales (cont.). Fig. 57. Blasia pusilla; prominent wall labyrinths in both generations. Fig. 58. Pullavicinia indica; sporophyte wall labyrinth and thin-walled gametophyte cells. Fig. 58. Riccafdia mulr@du; thick-walled sporophyte cells and gametophyte wall labyrinths.

Anrhoceros punc~oturL.

Anlhoreros Jormosar Haseg Anrhoceros gronulosa Haseg Phaeocrros / n e w husk.

Plmeorero.~curohntonus Proak. Foltoceror JuaJo.rlornirrBharadw

Nomrhdrv orhtorlorrr Sull. Noforhvlrv emperaro Haseg

Dmdrocem co~emoxu(Haseg. Dendroreros iovonicur Nees. Dmdroceros mherculons Haft Megocerorj?agellrrrrrSteph.

2 1 I

3 I 1

I 1 1 1 4 I

+ +

I

+ + + + +

+ + + + +

Adjacent to .t"tercell"lar

spacer

Early and C0"fl""O"I

Cornplcl labyrmths

Complex }labyrinths Complex labyrinths Complex }labyrinths Simple labyrinths Simple lahynnth

I

Bulbous

l l t m walled. branched

unicellular or multiceUular haustoria

0547

0.M7

}Bulbour BdbU

0 w.7

}Bulbous

0.uI 4

: :1

Ovoidal

0.zw 4 u.7-I 0

T A B L E VI Cytoplasmic features of placental cells in anthocerotes Plastids Gametopbyre Shape

Membrane

Sporophyte Plastoglobuli

Starch

Shape

system Anrhoceros pwcIuNs Anrhocernr fonnosoe A~irhoc~ror gronulosu

)pIeomorphic

Phoeocrror Iowa Phveocerm corobnionur Folmeerosfurrformu

/zyr tr.gr LPYI

Pleomorphic

c.gr.

/+

+

'pyr. Nororhvlos orbtculorb Norothvlar remperoro

Dindroceror

}Pleomarphic

C L ~ Y P ~ O ~ W

Drndroreros Dendroreror javonirur mhrrculom

Mega<
};L%&hir Spheroidal

kyr

1%

r.gr pyr

*

Abbreviations. n gr . normal grana; r.gr ,rudunentalygrana: pyr , pyrenoid.

Membrane

Plastaglabuli

Srarch

system

I+

1I-

1

Spheroidal pleomorphir

Spheroidal }pieumorpixuc Spheroidal pleomorphic ptxmd

I

Spheroidal pleomorphic

Spheroidal pleomorphic

l++

}w

I&

}r 8'

+

++

r.gr.

. ; ;}

+

).

gr-

qr.

I:+

++

1-

I++

I+

I-

+

Crystals m intercellular spaces

++

++

++ ++

++

+

278

R. LIGRONE et ul.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

279

The foot is ovoidal (Megaceros, Dendroceros; Fig. 93) or bulbous in the other four genera (Figs. 94 and 95) and generally 0.5-0.7mm in diameter except in Notothylas and Dendroceros (0.25-0.4 mm in diameter). The organization of the placenta is fundamentally the same in all the genera and comprises long and branched haustorial cells of sporophytic origin intermingled with gametophyte cells (Figs. 96,97,99 and 101). The two cellular types are separated by abundant intercellular spaces containing PASpositive mucilage, that enlarge locally forming wide lacunae. As in mosses and liverworts, the placental cells are highly specialized and clearly distinct from the other cells of the respective generation. The haustorial cells have uniform and relatively thin walls. They penetrate the gametophyte tissue by intrusive growth involving both cell division and elongation (Gambardella and Ligrone, 1987). The nucleus generally lies and divides far from the growing tip (Fig. 97), and this may branch with (Figs. 96 and 101) or without (Fig. 97) the formation of cross walls. The haustorial cells are highly vacuolated in Megaceros, whereas they contain less prominent vacuoles and more abundant cytoplasm in Anthoceros, Phaeoceros and Folioceros. Numerous vacuoles of small sizes occur in the haustorial cells of Notothylas (Fig. 101) and Dendroceros (Fig. 96). In all genera but Notothylas the haustorial cells contain large spheroidal or irregularly shaped plastids with a rudimentary thylakoid system, abundant plastoglobules, sometimes spherical starch grains and vesicles of various sizes (Vaughn et al., 1992; Figs. 97 and 98). The haustorial cells of Notothylas contain chloroplasts with a well-developed inner membrane system (Fig. 101) and sometimes a distinct pyrenoid, though less prominent than in other vegetative cells of the thallus. The gametophyte placental cells are much smaller than ordinary parenchyma cells of the thallus, have dense cytoplasm with several relatively small vacuoles and are rich in endoplasmic reticulum, dictyosomes and mitochondria which form prominent aggregates in Phueoceros carolinianus (Fig. 102). At maturity they present wall ingrowths of variable sizes, generally along the cell sides abutting intercellular spaces. Wall labyrinths of considerable complexity are found in Phaeoceros, Notothylus and Folioceros (Figs. 99 and 101), whereas in the other genera they are much simpler (Fig. 96). In all genera except Dendroceros the plastids lack starch and have rudimentary thylakoid systems and the pyrenoid is highly reduced or absent (Fig. 102). In the gametophyte placental cells of Dendroceros are

Figs. 60-63. The gametophyte-sporophyte junction in Mctzgeriales (cont.). Fig. 60. Riccardia mulfifidu;thickened tangential walls with radial deposits, sporophyte placental cell. Fig. 61. Pelliu epiphyllu; thin-walled placental cells in both generations. Fig. 62. Cryprofhullus mirubilis; wall ingrowths are absent. The intraplacental space contains collapscd garnetophyte cells (arrowed). Fig. 63. Aneuru pinguis: nacreous wall thickenings in the sporophyte placental cells.

280

R. LIGRONE et al.

T H E CAMETOPHYTE-SPOROPHYTE JUNCTION

281

chloroplasts with a well-developed thylakoid system but lacking a pyrenoid (Fig. 100). Dense bodies with a crystalline substructure are present in large amounts in the intercellular spaces, most notably in the large lacunae, of the placenta in Phaeoceros, Notothylas and Folioceros (Figs. 103-106). These structures react positively to protein stains (Marsh and Doyle, 1985) and are completely digested by pronase or pepsin (Fig. 104), hence they are assumed to be protein (Gambardella and Ligrone, 1987). In longitudinal or oblique sections the crystals consist of alternating dense and light bands. Transverse sections reveal a regular network of hexagonal subunits, approximately 15nm wide, with a light internal core (Fig. 103). Degradation of the crystals has been observed in Phaeoceros and Notothylas (Fig. IOS), notably at advanced stages in sporophyte development, and it has been suggested that these structures may serve as a source of nutrients for the sporophyte growth (Marsh and Doyle, 1985). The larger lacunae in Phaeoceros and Folioceros are formed by dissolution of gametophyte placental cells, a process that presumably is induced by the haustorial cells (Gambardella and Ligrone, 1987). Degeneration of gametophyte cells follows their symplasmic isolation from the adjoining cells, due to the severance of plasmodesmatal connections. Crystals are deposited in the vacuoles of gametophyte cells (Fig. 106) and are liberated in the placental lacunae as a consequence of cellular dissolution. Crystals are but rarely observed in the placenta of Megaceros and are not found in Anthoceros and Deridroceros. Both sporophyte and gametophyte placental cells of Dendroceros contain intravacuolar deposits of dense amorphous material (Fig. 98) that is assumed to be protein in as much as it is negative to the PATA test for carbohydrates and sensitive to digestion by pronase (Ligrone and Renzaglia, 1990). Similar protein deposits also occur in the parenchyma cells of the foot and virtually all the cells of the sporophyte capsule, including the spores. Overall, these observations suggest that protein is synthesized in the haustorial cells, presumably from precursors provided by gametophyte transfer cells, and is then transferred, via plasmodesmata, to the parenchyma cells of the foot and eventually to the cells of the growing capsule. Like the protein crystals in Phaeoceros and Folioceros, the amorphous protein deposits in the gametophyte placental cells of Dendroceros may function as a source of amino acids for the sporophyte, although they are never found free in placental spaces (Ligrone and Renzaglia, 1990). Figs. 64-67. The gametophyte-sporophyte junction in Metzgeriales (cont.). Fig. 64. Cryptorhallus mirabilis; highly irregular plasmalemma in a gametophyte placental cell. Fig. 65. Riccardia mulrificla: giant mitochondrion in a sporophyte placental cell. Fig. 66. Cryplorhullus mirabilis: grazing section of endoplasmic reticulum with abundant attached polysomes in a gametophyte placcntal cell. Fig. 67. Cryprothallus; concentric sheets o f R E R around a lipid body in a sporophyte placental cell.

282

R . LICKCINE or a/.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

283

The placental region develops very early and is well organized when the sporophyte, less than 1 mm long, is barely discernible with naked eye within the gametophyte thallus. In Phaeoceros abundant intercellular crystals are found in the placenta when the sporophyte emerges from the involucre, whereas wall labyrinths are formed in gametophyte placental cells during the subsequent phase of capsule elongation (Gambardella and Ligrone, 1987).

111. THE TAXONOMIC SIGNIFICANCE OF THE PLACENTA IN BRYOPHYTES AND IMPLICATIONS FOR PHYLOGENY Two main conclusions emerge from the comparative data described here for the first time: (a) placental organization is extremely diverse in the bryophytes as a whole; and (b) with some exceptions, it is highly constant within particular groups of bryophytes. The diagnostic features embrace not only the walls and cytoplasmic contents of the placental cells, but also the timing of their differentiation. Whereas wall characteristics render the placental regions of different groups almost absolutely distinctive, cytoplasmic features and especially plastid ultrastructure tend to be peculiar to particular genera. This remarkable diversity in placental morophology in bryophytes has no obvious (or even obscure) functional meaning. For example, the absence of transfer cells on one or both sides of the placenta can hardly be assumed to indicate a lower efficiency in solute translocation and/or greater nutritional autonomy of the sporophyte. Indeed, in taxa lacking placental wall ingrowths (e.g. Sphagnum and Pellia), the sporophytes are just as highly differentiated and spore production just as prolific as in species where these are well developed. Conversely a typical bryalean placenta is found in Archidium (Brown and Lemmon, 1985) in spite of the extreme reduction of the sporophyte in this genus (Snider, 1975). Similarly in three different bryalean orders, namely Buxbaumiales, Pottiales and Dicranales, the placental regions are just as highly differentiated in genera with rudimentary setae (Diphyscium,Phascum and Cladophascum, respectively) as in those with elongate setae. Although it is now widely accepted that transfer cell morphology is associated with intense short distance transport of solutes (Gunning and Pate, 1974), most of the evidence remains circumstantial and knowledge of the biochemical and physiological mechanisms involved in this process via Figs. 68-72. Plastids in metzgerialean sporophyte placental cells. Fig. 68. Fossombronia echinara; elongate chloroplasts. Fig. 69. Riccardia rnultifida post-meiosis; ovoid with small grana and starch grains adjacent to a giant mitochondrion. Fig. 70. Pallavicinia indica; prolamellar body with radiating grana. Fig. 71. Riccardia mult@da pre-meiosis; ovoid and pleomorphic with single thylakoids. Fig. 72. Cryptothallus; pleomorphic undifferentiated plastids.

284

R. LIGRONE et a1

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

285

the wall labyrinths has barely increased since the 1970s. Electrophysiological studies of the foot transfer cells in Polytrichum (Caussin et al., 1979, 1982, 1983; Renault et al., 1989) have demonstrated that protonelectrochemical gradients are generated through the wall-membrane apparatus and used to energize the uptake of solutes, particularly amino acids, from the apoplast. However, similar mechanisms also operate equally well in cells lacking wall ingrowths (Despeghel and Delrot, 1983; Kinraide et al., 1984). In the latter context it should be noted that the ultrastructure of thickened placenta walls with radial bands in Sphagnum, Polytrichales, Andreaea, Takakia, Jungermanniales and Riccardia is remarkably similar to that of the microspore intine in angiosperms (Charzynska et al., 1990; Murgia et al., 1991). Not only is the intine freely permeable to solutes, but recently the microspore has also been recognized as a type of transfer cell (Charzynska et a l . , 1990). The same kind of wall ultrastructure also characterizes vessel contact cells in the xylem of herbaceous angiosperms (Mueller and Beckman, 1984). Against a background of these functional considerations, the simplest explanation for the range of highly distinctive placental morphologies found in bryophytes is that each arose independently early in the evolution of each group. Once an effective transport system, with or without transfer cell morphology, had become established between the gametophyte and developing sporophyte, intense selection pressure for further change would disappear. Thus the ancestral condition would likely be retained more or less unchanged throughout the subsequent evolutionary history of each group. Clearly implicit in this hypothesis is the possibility that transfer cell morphology and other apparently similar placental features may not be homologous throughout bryophytes. Indeed, structural and temporal differences in wall ingrowth morphology and development strengthen this notion. Thus, with the exception of the Calobryales, early versus late differentiation of transfer cell morphology sets mosses clearly apart from hepatics. The fine dense wall labyrinths in Jungermanniales are very different from the coarse more transparent ingrowths in the Marchantiidae. Physiological studies are now needed to discover whether the temporal differences in placental transfer cell differentiation reflect differences in the availability of solutes destined for export from the gametophyte (see Gunning et al., 1968; Pate et al., 1970; Davey and Street, 1971; Gunning and Pate, 1974; Henry and Steer, 1980; for examples of the coincidence of wall Fig. 73. Cryptorhullus; highly pleomorphic undifferentiated plastid in a gametophyte placental cell. Figs. 74-76. The gametophyte-sporophyte junction in Calobryales. Fig. 74. Huplomitrium hooker;; wall ingrowths are more extensive in the sporophyte. Fig. 75. Calobryum blumei; a more cxtensive wall labyrinth in the sporophyte. Fig. 76. Calobryum blumei; dead cells with electron-dense deposits lining their walls in the central region of the foot.

286

R. LIGRONE

el

ul.

Figs. 77-79. Plastids in calobryalean placental cells. Fig. 77. Culobryum blumei sporophyte; pleomorphic plastids with starch and rudimentary grana. Fig. 78. Haplomitriurn hookeri sporophyte; a single highly pleomorphic plastid. Fig. 79. Calobryum blumei gametophyte; pleomorphic plastid with grana perpendicular to the long axis.

T H E G A M E I ’ O P ~ I Y I ‘ E - S P O R O P I ~ Y ‘JUNCTION r~

287

ingrowth development with the presence of solutes destined for export in vascular plants). Similarly, cytochemical investigations aimed at elucidating the biochemical basis for the structural differences between the labyrinths of different groups may help to establish whether or not these are homologous. Bearing in mind the difficulties in distinguishing between parallel evolution and homology, it is nevertheless possible to draw major taxonomic conclusions from placental morphology and it is highly pertinent to explore the implications for phylogeny. Three basic types of placenta may be identified in mosses: an Andreaeatype, present in the Andreaeales, Takakiales and Polytrichales, a Sphugnum-type, restricted to Sphagnales, and a bryalean type, characteristic of the Bryales and also present in the Tetraphidales. Each type is sharply distinct from the others since the different distribution or absence of transfer cells can be immediately recognized, at least at the ultrastructural level (Table I). Minor variants are found within each type or group, for example the radially elongate foot transfer cells of Bryum and the rhizoidlike outgrowths of Diphysciurn. On the whole, however, the placental organization is remarkably constant even in the exceedingly large group of the Bryales, where taxonomically distant genera present little or no appreciable differences. A greater intergeneric diversity is apparent in the Polytrichales, especially in the gametophyte, perhaps indicating that this order is of considerable antiquity. The highly distinctive placental organization of the Sphagnales reinforces the notion that this is the most divergent group in mosses (Crosby, 1980; Brown et al., 1982; Duckett et al., 1982, 1984; Robinson and Shaw, 1984; Ligrone and Renzaglia, 1989). On the other hand, the discovery that the Andreaeales and Polytrichales share a similar placenta places the former clearly apart from the Sphagnales (cf. the “Pseudopodiate line” devised by Kumar, 1984) and closer to bryalean mosses. The presence of the same basic type of placenta in all the Bryales (with the only known exception, Dicranum majus, almost certainly representing a derived condition) confirms this group, including the Buxbaumiales and Tetraphidales, as a monophyletic unit. The discovery of a bryalean placenta in Tetraphis, previously reported as having labyrinthine walls in the sporophyte only (Roth, 1969), is in line with the notion, based o n studies of spermatid morphology (Duckett et af., 1982) and peristome development (Edwards, 1984; Shaw and Anderson, 1988), of a closer phylogenetic relationship between the Tetraphidales and arthrodontous mosses rather than between either group and the nematodontous Polytrichales. Likewise, the Buxbaumiales possess an arthrodontous peristome (Edwards, lY84; Shaw et ml., 1987) and a typical bryalean placenta. If we assume that placental transfer cells are plesiomorphic in embryophytes (see above) and that the Andreaeales and Polytrichales are primitive to the Bryales (Crosby, 1980; Robinson and Shaw, 1984), then the

288

R. LIGRONE

et

(11.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

289

Andreaea-type of placenta appears to be the most likely ancestral type in mosses, with the other two types being derived (see also Ligrone and Renzaglia, 1989). This conclusion is strongly supported by the discovery of a placenta of the Andreaea-type in Takakia, a taxon that has long been classified in the Calobryales (Schuster, 1984b). Such a striking congruence between placental morphology and the anatomical features of the newly discovered antheridia and sporophyte in Takakia is a further confirmation of placental morphology as a taxonomically valuable character. Taxonomic and phyletic relationships between major groups in liverworts are more uncertain than in mosses. For example, it is still debated whether liverwort evolution has proceeded from leafy towards thalloid forms or vice versa. Schofield (1985) thinks of the gametophyte of a hypothetical ancestral liverwort as a flattened dorsiventral thallus with sex organs exposed on the dorsal surface, a model that is most closely approached by the Metzgeriales. Mishler and Churchill (1985) also believe that the ancestral type in liverworts is a simple thalloid gametophyte, though they construct a cladogram where the Metzgeriales and Jungermanniales are considered advanced and the Sphaerocarpales and Marchantiales primitive. Schuster (1984b) observes that the passage from a leafy to a thalloid habit has probably occurred several times independently in different liverwort groups, and suggests that the great diversity of gametophytic types may ultimately be derived back to a simple radial, leafless type. O n the basis of a wide range of characters, he recognizes a “primary dichotomy” between two subclasses, the Jungermaniidae (comprising the Jungermanniales and Metzgeriales) and the Marchantiidae (including the Sphaerocarpales and Marchantiales), with deviations from this generalization occurring principally in three taxa, i.e. Haplomitrium, Takakia and Monoclea. Placental morphology confirms the Jungermanniales and Marchantiidae as taxonomically well-delimited groups but sheds less light on possible inter-relationships between the Jungermanniales, Marchantiidae and the Calobryales. Unlike the ultrastructural uniformity in the appearance of the placental wall labyrinths in bryopsid mosses, their great variability in liverworts renders interpretation of homology, or lack of it, between transfer cells in different groups more problematic. The fine dense labyrinths in the sporophyte placenta of Jungermanniales are very different from the coarse electron-transparent ingrowths that characterize both generations in

Figs. 80-84. The garnetophyte-sporophyte junction in Marchantiidae. Figs. 80-82. Monoclea gottschei. Fig. 80. Light micrograph, longitudinal scction showing the obtuse conical foot. Fig. 81. Coarse electron-transparent wall labyrinth in garnetophyte placental cells. Fig. 82. Wall labyrinth in a sporophyte placental cell. Fig. 83. Sphaerocarpos texanus; wall labyrinths in placental cells of both generations. Fig. 84. Dumortiera hir.suta; short wall protuberances in the garnetophyte and extensive wall ingrowths in the sporophyte. Note the collapsed garnetophyte cells (arrowed).

290

R. LIGRONE et al.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

29 1

Marchantiidae. The ultrastructure of the calobryalean wall ingrowths is somewhat intermediate. The occurrence of sporophytic wall ingrowths and a tendency for the gametophytic placental cells to form thickened outer walls in 10 genera of the Jungermanniales ranging from putatively primitive (e.g. Herberta) to highly advanced (e.g. Zoopsis) (Schuster, 1984b) suggests that this is the ancestral condition in the group. The absence of transfer cell morphology in Radufa is almost certainly a derived condition perhaps related to interdigitation of the cells of the two generations instead of the clearly defined intraplacental lacuna seen in all the other genera. It would now be pertinent to discover whether this feature is peculiar to the Radulaceae, or if it also occurs in the allied families Porellaceae, Jubulaceae and Lejeunaceae. The uniformity in placental ultrastructure in the Marchantiidae which, with the exception of Riccia, extends throughout its three highly disparate orders Marchantiales, Sphaerocarpales and Monocleales, is clearly in accord with the widely recognized anatomical and development uniformity of this subclass (Brown and Lemmon, 1988, 1990; Crandall-Stotler, 1981; Schuster, 1984a,b). The occurrence in Carrpos of a placenta no less highly differentiated than in other members of the group weakens the case for a possible link with Riccia (Proskauer, 1961). Indeed, the absence of any trace of a placenta sets Riccia apart from all other archegoniates and from Coleochaete, the taxon generally considered to be the closest living algal relative to the embryophytes (Graham and McBride, 1979; Graham, 1982,1984,1985; Graham and Wilcox, 1983; Graham and Wedemayer, 1984; Graham and Taylor, 1986; Duckett and Renzaglia, 1988a; Delwiche et al., 1989; Vaughn et a f . ,1992). However, it must be stressed that the information for Riccia is based on a single species. The possible future discovery of the vestiges of a placenta elsewhere in the genus would be an unequivocal demonstration that the sporophyte of Riccia represents the end of a reduction series rather than an ancestral condition. The Metzgeriales are by far the most diverse group of liverworts and include several isolated entities whose affinities are highly problematic (Crandall-Stotler, 1981; Renzaglia, 1982; Schuster, 1984b; Brown and Lemmon, 1988; Carothers and Rushing, 1988). This diversity is almost certainly related to the extreme antiquity of the group, the fossils of which date back at least to the upper Devonian (Krassilov and Schuster, 1984). Although the wide range of placental morphologies further underlines the profound intersubordinal, interfamilial and even intergeneric discontinuities within the order, evaluation as to which of the variations represents Figs. 85 and 86. The gametophyte-sporophyte junction in Marchantiidae (cont.). Fig. 85. Reboulia hemisphaerica; more extensive wall ingrowths in the sporophyte. Fig. 86. Carrpos monocarpos; short wall protuberances in the gametophyte and normal chloroplasts in the placental cells of both generations.

292

R. LIGRONE

el

al.

T H E GAMETOPHYTE-SPOROPHYTE JUNCTION

293

the primitive condition is at present highly speculative. Perhaps the most plausible choice is the condition found in Blasia and Fossombronia (wall labyrinths in both generations) in as much as these are generally considered to be representatives of very old isolated taxa (Schuster, 1984b). Starting from the condition found in such genera as Blasia and Fossombronia, variant types might have evolved through the suppression of transfer cells in either the gametophyte (Pallavicinia) or the sporophyte (Riccardia) or both (Pellia, Aneura and Cryptothallus). The similarity in placental organization between the Calobryales and putatively primitive members of the Metzgeriales is in accord with Schuster’s (1984b) notion of a common, albeit extremely remote, ancestry for these two groups. On the other hand the fact that putatively primitive members of the Metzgeriales share a similar placental organization with the Marchantiidae might be indicative of a closer phyletic relationship between simple and complex thalloid liverworts than between either group and the Jungermanniales. This possibility is substantiated by ontogenetic data (Crandall-Stotler, 1981) and is closely in line with recently recognized affinities between Blasia and Marchantiidae. Most notable of these are marchantialean-like spermatozoids (Renzaglia and Duckett, 1987a,b), the presence of two rows of ventral scales, the location of the archegonia, sporophyte development, apical organization (Renzaglia, 1982) and monoplastidic spore mother cells (Brown and Lemmon, 1992). However, it must be noted that, as long as the foot is considered to have different embryological origins in the Metzgeriales and Marchantiidae groups (see Section I.B), the placentas in these two groups cannot be considered homologous and therefore are not strictly comparable in the context of phylogeny. The Jungermanniales are given a near ancestral position in liverwort phylogeny by Schuster (1979, 1984b), a conclusion that appears supported by recent comparative analyses of spermatogenesis (Duckett et al. , 1982, 1984; Renzaglia and Duckett, 1991). In accord with these data, the jungermannialean placenta might be the ancestral type in liverworts. This possibility is supported by the striking similarity with the Andreaea-type in mosses. In addition to transfer cells only in the sporophyte, the Jungermanniales, Andreaeales, Takakiales and Polytrichales share a similar substructure in gametophyte placental cells, namely thickened tangential walls with radially arranged deposits of dense material. This perhaps suggests a closer phyletic relationship between the Andreaeales-Takakiales Figs. 87-90. The gametophyte-sporophyte junction in Marchantiidae (cont.). Fig. 87. Rebouliu hernisphuerica; plastids with fibrillar bundles (arrowed) parallel to the long axis in a gametophyte placental cell. Fig. 88. Reboulia; longitudinal section through membrane-bound bundles of tubules in a gametophyte placental cell. Fig. 89. Reboulia; transverse sections through membrane-bound bundles of tubules. Fig. 90. Riccia sorocurpa; light micrograph, longitudinal section of a sporophyte containing spore tetrads. The capsule is completely enclosed in the gametophyte thallus. The archegonial neck is arrowed.

294

R. LIGRONE et al.

Figs. 91 and 92. The gametophyte-sporophyte junction in Marchantiidae; Riccia sorocarpa (cont.). Fig. 91. Section through the margin of a young sporophyte. C, two-layered calyptra; J , jacket and nurse cells; SMC, spore mother cell. Fig. 92. Section through the margin of a nearly mature sporophyte. Note the abundant lipid reserves in the spore tetrad (T), the collapsed inner layer of calyptra cells (arrowed) and the thickened walls of the thallus cells (G) facing the calyptra. Figs. 9>95. The gametophyte-sporophyte junction in Anthocerotes. Light micrographs, longitudinal sections. Fig. 93. Megaceros flagellark; ovoidal foot. Fig. 94. Anthoceros formosae: bulbous foot. Fig. 95. Notothylas temperata; smaller bulbous foot than Anthoceros.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

295

group and Jungermanniales than either the Marchantiidae or the Metzgeriales. This would imply a common origin for mosses and liverworts, from an ancestal form probably represented by an erect, radially symmetrical leafy gametophyte with terminal sporophytes. However, it should also be noted that the same distinctive wall substructure of dense radial bands also turns up in the sporophyte placental cells of Riccardia, a metzgerialean genus generally considered to be advanced (Schuster, 1984b), and in gametophyte placental cells of Sphagnum (Ligrone and Renzaglia, 1989). The possibility that this feature is merely associated with solute permeability rather than indicative of phyletic affinity should not be dismissed. In contrast to mosses and liverworts, the anthocerotes all exhibit the same, highly distinctive type of placenta, with minor variants mainly concerning plastid morphology and the presence, appearance and localization of protein deposits. This uniformity in placental organization emphasizes the homogeneity of this group and its separation from mosses and liverworts, as indicated by a wealth of anatomical, ultrastructural and developmental data (Crandall-Stotler, 1980, 1981, 1984; Duckett et a f . , 1982, 1984; Schuster, 1 9 8 4 ~ ;Carothers and Rushing, 1988; Duckett and Renzaglia, 1988a, 1989). The variations in placental morphology in anthocerotes appear to by very useful both for elucidating intergeneric affinities and for clarifying generic limits. The distinctive protein crystals found in Phaeoceros, Notothylas and Folioceros suggest close affinity between these genera. This contradicts the classification of anthocerotes recently proposed by Hasegawa (1988), where the family “Notothyladaceae”, comprising Notothyfas as the only genus, is separated from the family “Anthocerotaceae”, which includes Phaeoceros, Folioceros, Anthoceros and Megaceros. Recent ultrastructural studies of spermatozoid morphology and development have revealed close similarities between Phaeoceros and Notothyfus (Renzaglia and Duckett, 1988, 1989), although the lack of information on other genera presently precludes wider evaluation of the taxonomic relevance of male gamete microanatomy . The distinctive ultrastructural characteristics of the placenta in Dendroceros seemingly support the isolation of this genus in a separate taxonomical entity, e.g. the family Dendrocerotaceae as proposed by Hasegawa (1988).

IV. PTERIDOPHYTES By contrast to the wealth of comparative ultrastructural data on the gametophyte-sporophyte junction in bryophytes, knowledge of the embryology and the early stages in sporophyte differentiation in pteridophytes is limited exclusively to light microscope studies, the majority dating from the nineteenth century or the first half of the twentieth century and reiterated

296

R. LIGRONE et al

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

297

with little or no additional information in reviews and standard textbooks (Wardlaw, 1965; Bierhorst, 1971; Gifford and Foster, 1989). In these accounts it is usually tacitly assumed that the foot is the major site of exchange between gametophyte and sporophyte generations, but supporting experimental and or cytological data are lacking. More recent studies on the causal basis of the alternation of generations in pteridophytes (see reviews by Sheffield and Bell, 1987, 1989) have focused on the cytology of oogenesis and to a lesser extent on apospory and apogamy and do not consider the placental region. However, it is interesting to note that transfer cell morphology appears to be absent at the base of apogamous sporophytes in ferns whereas the cells initiating apogamous sporophytes in the moss Physcomitrium develop labyrinthine wall ingrowths even though a recognizable placenta is absent (Menon and Bell, 1981; La1 and Narang, 1985). Specific information on the placenta of pteridophytes is limited to a mention without micrographic details by Gunning and Pate (1974) of transfer cells in this region in Equisetum, two brief accounts of three ferns Polypodium, Adiantum (Gunning and Pate, 1969b) and Pteridium (Khatoon, 1986), and two very recent studies of Lycopodium (Peterson and Whittier, 1991; Duckett and Ligrone, 1992). Although these works clearly do not form a suitable basis for an incisive appraisal of the taxonomic and possible phylogenetic significance of the placenta in the different groups of pteridophytes and do not permit a comparison between these and other embryophytes, the fragmentary information they contain indicates a most fruitful area of enquiry for the future. In Polypodium and Adiantum wall ingrowths in both gametophyte and sporophyte develop very early, before the expansion of the first leaf, elongation of the root and differentiation of the first xylem: precisely the stage at which the sporophyte is most dependent on the gametophyte for nutrients (Gunning and Pate, 1969b). The ingrowths are more abundant and labyrinthine in the sporophyte in Adiantum and Pteridium, while the opposite occurs in Polypodium. The ingrowths are more highly developed on the tangential walls along the interface but also extend onto the lateral walls in the sporophytic cells in Adiantum and Polypodium. The sporophytic cells, but not those of the gametophyte, contain abundant starch, a situation probably reflecting sugar translocation from the gametophyte to the sporophyte. Unlike mosses and liverworts, where a clear-cut demarcation zone is generally interposed between the sporophyte and gametophyte, in ferns the

Figs. 96-9X. The garnetophyte-sporophyte junction in Anthocerotes (cont.). Fig. 96. Dendroceros tubercularis; gametophyte transfer cells and sporophyte haustorial cells. Note the chloroplasts in the former. Fig. 97. Dendroceros tubercularis; branched sporophyte haustorial cell. Note the undifferentiated plastids. Fig. 98. Megacerosjfagellaris;undifferentiated plastids in a sporophyte haustorial cell and crystals (arrowed) in the adjacent garnetophyte cell.

298

R . LIGRONE

ct

ul.

Figs. 99 and 100. The gametophyte-sporophyte junction in Anthocerotes (cont.). Fig. 99. Phaeoceros luevis; gametophyte transfer cells, sporophyte haustorial cells and placental lacunae (arrowed). Fig. 100. Dendroceros tuberculuris; plastid in a gametophyte transfer cell.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

299

cells of the two generations interdigitate and mucilage-containing intraplacental species appear to be lacking. Further differences between bryophytes on the one hand and ferns and Lycopodium (see below) on the other are an absence of dead and collapsed cells along the interface and the lack of any evidence for the renewal of the placental cells. These differences are perhaps related to the transient nature of sporophytic dependence on the gametophyte in pteridophytes. Intercellular spaces are lacking in the placental region of Lycopodium uppressum (Chapm.) Lloyd & Underw: the cells of the two generations lie close together or are separated by electron-dense intercellular material (Peterson and Whittier, 1991). As in bryophytes and seed plants (see Section V), symplastic isolation, initially established during the early stages in the differentiation of the axial row in the young archegonium (Bell, 1989), characterize all stages of sporophyte development in pteridophytes. In the foot region of Lycopodium uppressum there is little or no interdigitation of the cells of the two generations and the contiguous walls of both sporophyte and gametophyte develop coarse labyrinthine ingrowths of low electron opacity very similar ultrastructurally to those in the marchantialean placenta. Subsequently the interstices become occluded by dense amorphous wall material. The sporophyte-gametophyte junction in Lycopodium cernuum L. is somewhat different. Here the interface between the two generations, which develops ingrowths but to a more limited extent than in L. uppressurn, is not a special lateral development of the early embryo (i.e. a foot region) as in other species of Lycopodium (Goebel, 1905), but rather the lower part of the primary embryonic axis derived from the suspensor (Duckett and Ligrone, 1992). The homologue of the foot in L. cernuum, in terms of its lateral position and early appearance in sporophyte differentiation, is the protocorm. This juvenile structure lies outside the confines of the parent gametophyte and presumably derives its nutrition partly from photosynthesis and partly from an endophytic fungus. Of particular interest in the present context is that within the protocorm there develop schizogenous, mucilage-filled intercellular spaces closely similar to those in the placental region of bryophytes. These lacunae are the principal habitat of the mycobiont. Although the protocorm cells do not develop wall ingrowths, their invasion by the fungus is associated with the production of massive overgrowths of host cell wall material with a texture similar to that forming the wall thickenings in the gametophytic placental cells of Sphagnum and Jungermanniales. The major differences between the sporophyte-gametophyte junction in two species of Lycopodium on the one hand reflect the considerable antiquity of this genus, and on the other perhaps anticipate diversity in placental morphology in pteridophytes parallelling that in bryophytes. It will now be particularly interesting to discover whether wall ingrowths are also present

300

R. LIGRONE et al.

THE GAMETOPHYTE-SPOROPHYlE JUNCTION

301

along the margins of the well-developed suspensor present in many species of Selaginellu (Bierhorst, 1971) and to investigate the position and sites of nutrient exchange between gametophyte and sporophyte in the haustorialike foot region of Tmesipferis (Holloway, 1918).

V. SEED PLANTS The classic reviews on transfer cells (Gunning and Pate, 1969a, 1974; Pate and Gunning, 1972; Gunning, 1977) underline the cytoplasmic discontinuities between sporophyte and megagametophyte, between embryo and megagametophyte and between sporophyte and developing microspores (also see Charzynska et al., 1990; Murgia ef al., 1991; Pacini, 1990 for discussion of structural and functional interrelationships between tapetum and microspores) as posing special problems for the transport of solutes. Gunning and Pate go on to point out that, in view of the complex nutritional relationships, plus the fact that the new gametophyte and sporophyte are both parasitic on the parent sporophyte, it is not surprising to find transfer cells in these situations. Indeed, transfer cell morphology has now been described at virtually every site of probable solute exchange via the apoplast in the female reproductive organs of angiosperms (Table VII). In a few situations the zones of transfer cells face each other along the intergenerational apoplastic gap, but more commonly only the receptor surface bears the ingrowths. Sometimes the boundary between the two generations is a common wall lacking plasmodesmata, but more often these are separated by a space containing remains of dead or dying cells. The diverse locations and temporal differences in the development of wall ingrowths during the ontogeny of the various tissues provide perhaps the strongest circumstantial evidence of transfer cell activity in solute transport. As in the placentas of most bryophytes and pteridophytes, it seems to be an anatomical necessity that nutrients directed to the growing sporophyte have to pass through cells with wall ingrowths. Rather than reiterate the substance of the earlier reviews, this account focuses on some of the more significant discoveries since 1977, compares the present stage of knowledge of the gametophyte-sporophyte junction in seed plants with that in bryophytes and pteridophytes and points to areas urgently needing further study. As in bryophytes, in seed plants there are also virtually no physiological data to validate inferences about routes and timing of solute transport based almost entirely on the formation of wall labyrinths. It is, however, Figs. 101 and 102. The gametophyte-sporophyte junction in Anthocerotes (cont.). Fig. 101. Notothylus orbicularis; sporophyte haustorial cells and gametophyte transfer cells. Fig. 102. Phoeoceros carolinianus; mitochondria1 aggregates and a pleornorphic plastid in a gametophyte transfer cell. A small pyrenoid is arrowed.

TABLE VII Occurrence and likely functions of transfer cells in the female reproductive organs of angiosperms. Modified and updated from Pate and Gunning (1972) and Gunning (1977) Organ or tissue

Embryo sac

Young embryo

Location

Probable function

Distribution

Recent references"

Egg cell, micropylar walls

Functions of synergids taken over by egg cell when synergids absent

Plumbag0

2,7,24

Antipodals, chalazal walls Synergids, micropylar walls The so-called filiform apparatus

Nutrition of embryo sac

Several genera

1,10,24

Nutrition of embryo sac

Several genera

Central cell, outer walls at micropylar pole chalazal region

Nutrition of embryo sac

Glycine, Scilla, Helianthus

5,6,7,16,17,23,26

Suspensor. outer walls

Nutrition of young embryo Nutrition of young embryo

Several genera

4,5,12,13,18,27

AIisma

3

Basal cell at micropylar pole

Chemotropic secretion, guidance of pollen tube

1,4,7,8,9,13,14, 15,19,20,24

Older embryo Nucellus Endosperm

Cotyledon epidermis, outer walls Epidermis near base of embryo Inner and outer faces Outer walls facing perisperm Aleurone, outer face

Integument

Inner face of inner

Nutrition of embryo

Some Leguminosae

21

Nutrition of embryo

Glycine, Triticum

5.22

Nutrition of embryo

Some Leguminosae and Cruciferae Mesern bryunthemuni

10,11,12

Nutrition of embryo Nutrition of embryo Facilitating viviparous germination and possibly salt exclusion Transfer to endosperm andlor embryo

21

Caryopses of Gramineae Rhizophoru

21

Some Leguminosae,

5

25

Glycine

"References: 1. Bhandari and Sachdeva (1983); 2. Bing-Quan Huang et al. (1990); 3, Bohdanowicz (1987): 4, Dute el al. (1989); 5 , Folsom and Cass (1986); 6. Folsom and Petersen (1984); 7. Kapil and Bhatnagar (1981); 8, Kennell and Horner (1985a); 9. Kennell and Horner (1985b); 10, Mansfield and Briarty (1990a); 11, Mansfield and Briarty (l990b); 12. Mansfield and Briarty (1991); 13. Mansfield er al. (1991); 14. Mogensen (1972); 15, Mogensen and Suthar (1979); 16, Newcomb (l973a); 17, Newcomb (l973b); 18, Newcomb and Fowke (1974); 19, Newcomb and Steeves (1971); 20, Olsen (1991); 21, Pate and Gunning (1972); 22, Smart and O'Brien (1983); 23, Tilton efal. (1984); 24. Willemse and Van Went (1984); 25, Wise and Juncosa (1989); 26, Yan eral. (1991); 27. Yeung and Clutter (1979).

304

R. LIGRONE et al.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

305

noteworthy that maize somatic embryos and embryos developing in vitro (Schel and Kieft, 1986; Fransz and Schel, 1991) appear to lack transfer cells. In Phoenix dactylifera L . differences in the acid phosphatases suggest that phosphate metabolism in the endosperm is independent of that in the cotyledon haustorium, and that acid phosphatases for endosperm phosphate metabolism are not secreted by the embryo nor by the cotyledon haustorium, but instead are stored in the endosperm (Sekhar and De Mason, 1989). There is now a pressing need to identify the specific nature of the metabolites undergoing import into young embryos via the various transfer cell zones (Table VII). Although relating to only two stages in development, the wealth of comparative information on the placenta in bryophytes forms the basis for wide-ranging taxonomic inferences. In angiosperms, by contrast, the situation is very different. A few very detailed developmental studies on a small number of genera, particularly Arabidopsis (Mansfield et a f . , 1991; Mansfield and Briarty, 1991), Capsefla (Schultz and Jensen, 1968a,b, 1969, 1971), Gfycine(Kennel1 and Horner, 1985a,b; Folsom and Cass, 1986; Dute et al., 1989) and Helianthus (Newcomb and Steeves, 1971; Newcomb, 1973a,b; Yan et al., 1991), reveal major differences in the extent and distribution of the wall ingrowths even between closely related genera. These relate partly to different patterns of embryo development and partly to differences in the location of seed storage reserves (e.g. cotyledons, perisperm or endosperm). Only one author (Mikeswell, 1990) attempts comparisons from a taxonomic standpoint. Mikeswell’s (1990) extensive survey of angiosperm families indicates that the differentiation of micropylar and chalaza1 haustoria from embryo sacs or endosperm is primarily confined to sympetalous plants with cellular endosperm and anatropous, unitegmic and tenuinucellate ovules. The presence of endosperm haustoria characterizes the subclass Asteridae, which includes the Plantaginales. The Plantaginaceae seems well aligned with families within the Scrophulariales. Mikeswell’s final conclusion, that the utilization of haustoria as an important embryological character in taxonomy would seem to be warranted, suggests the same could well be true for other transfer cell locations when data are available for more taxa. Ultrastructurally the vast majority of the wall ingrowths associated with the female reproductive tissues in angiosperms are coarse with a transparent matrix like those in the Marchantiidae. However, in rare instances, e.g. the suspensor of Gfycine(Dute et a f . ,1989) they are finer and electron-dense. A highly elaborate wall labyrinth occurs in the large basal cell of the suspensor Figs. 10>10h. The gametophyte-sporophyte junction in Anthocerotes (cont.). Crystals. Fig. 103. Folioceros fuciformis; intercellular. Fig. 104. Phaeoceros luevis; intercellular crystals after digestion with pepsin. Fig. 105. Nofothylasorbicularis; intercellular. Fig. 106. Folioceros fuciforrnis; intracellular.

306

R. LIGRONE

el

al.

of Afisma (Bohdanowicz, 1987).Thepresence of wall ingrowths is associated with cytoplasmic organization typical of transfer cells, namely abundant ribosomes, rough E R and numerous mitochondria. Very large megamitochondria have been described in young embryos of Capseffa(Schultz and Jensen, 1973) and Arabidopsis (Mansfield and Briarty, 1991). It is suggested that these may act as a reservoir for mitochondria1 DNA in relation to the rapid proliferation of mitochondria during embryo development. Although the cells along the gametophyte-sporophyte junction in angiosperms typically contain undifferentiated pleomorphic leucoplasts, with few reserve materials and rudimentary thylakoid systems, some highly unusual forms have also been described. Large plastoglobuli and prolamellar bodies characterize the endosperm plastids in Rhizophora (Wise and Juncosa, 1989). Prolamellar bodies also occur in the so-called “placental haustorium” plastids of Tropaeolum (Nagl and Kuhner, 1976). The suspensor by contrast contains plastids with an extremely dense stroma and scattered membranous vesicles. Much larger plastids with similar contents occur in the suspensor of Stelfaria (Newcomb and Fowke, 1974). Plastid tubules have been noted in the suspensors of Phaseolus and Pisum (Marinos 1970; Schnepf and Nagl, 1970). The current state of knowledge of the gametophyte-sporophyte interface in the gymnosperms is very similar to that for pteridophytes. In contrast to angiosperms, attempts to induce normal development of conifer zygotes and precotyledonary embryos in vitro have been unsuccessful (Gates and Greenwood, 1991, and literature cited therein) suggesting that a unique nutritional environment, probably involving continual variations in the physiological and chemical conditions, is required for embryo development. Although the considerable complexities of embryo development in gymnosperms are well documented at the light microscope level (Wardlaw, 1965), these have been totally ignored by electron microscopists. As far as we are aware the only report of wall ingrowths is in the basal plate wall between the oosphere cytoplasm and proembryo in Pinus (Gunning, 1977). For the future it would be interesting to discover whether or not ultrastructural differences between proembryos and embryos characterize the Gnetales, Cycads, Ginkgo and different families in the Coniferales in the same way that placental differences separate different groups of bryophytes.

ACKNOWLEDGEMENTS This review was made possible by a NATO Collaborative Grant to J . G . Duckett and K. S. Renzaglia and by a Guest Research Fellowship from the Royal Society of London enabling R. Ligrone to work at Queen Mary and Westfield College during 1989 and 1990. This financial support is most gratefully acknowledged. Collection of the specimens of Pogonatum neessii,

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

307

Herberta spp. Zoopsis liukiuensis, Pallavicinia indica, Calobryum blumei, Dumortiera hirsuta, Folioceros fuciformis, Dendroceros javanicus and D. tubercularis used in this study was made possible by a travel grant to J. G. Duckett from the Royal Society of London and by laboratory facilities arranged by Drs M. A . H. Mohamed and A. Nasrulhaq-Boyce in the Botany Department of the University of Malaya, Kuala Lumpur. Cladophascum gymnomitrioides was collected in Lesotho by J. G. Duckett under a British Council LINK between Queen Mary and Westfield College and the National University of Lesotho. The authors also thank D. K. Smith for providing live specimens of Takakia and R. C. Brown and B.E. Lemmon for allowing the use of their embedded material of Carrpos and Monoclea.

REFERENCES Bartholomew-Began, S. E. (199 1). “A morphogenetic re-evaluation of Haplomitrium Nees, Hepatophyta”. Bryophytorum Bibliotheca 41, 1 4 8 4 . Lubrecht and Cramer, Vaduz. Bell, P. R. (1989). The alternation of generations. Advances in Botanical Research 16, 55-93. Bhandari, N. N. and Sachdeva, A. (1983). Some aspects of organization and histochemistry of the embryo sac of Scilla sibirica Sato. Protoplasma 115,170-178. Bierhorst, D. W. (1971). “Morphology of Vascular Plants”. Macmillan, New York. Bing-Quan Huang, Russell, S. D., Strout, G. W. and Lian-Ju Mao (1990). Organization of isolated embryo sacs and eggs of Plumbago zeylanica (Plumbaginaceae) before and after fertilization. American Journal of Botany 77, 1401-1410. Bohdanowicz, J. (1987). Alisma embryogenesis: the development and ultrastructure of the suspensor. Protoplasma 137, 71-83. Bold, H. C., Alexopoulos, C. J. and Delevoryas, T. (1989). “Morphology of Plants and Fungi”. Harper & Row, New York. Bold, H. C. and Wynne, M. J. (1985). “Introduction to the Algae: Structure and Reproduction”. Prentice-Hall, New York. Bower, F. 0. (1908). “The Origin of a Land Flora”. Macmillan, London. Bremer, K. (1985). Summary of green plant phylogeny and classification. Cladistics 1. 369-385. Bremer’, K., Humphries, C. J., Mishler, B. D. and Churchill, S. P. (1987). On cladistic relationships in green plants. Taxon 36, 339-349. Brown, R. C. and Lemmon, B. E. (1985). Phylogenetic aspects of sporogenesis in Archidium. Micrographs in Systematic Botany from the Missouri Botanical Garden 11, 25-39. Brown, R. C. and Lemmon B. E. (1988). Sporogenesis in bryophytes. Advances in Bryology 3, 159-223. Brown, R. C. and Lemmon, B. E. (1990). Sporogenesis in bryophytes. In “Microspores: Evolution and Ontogeny” (S. Blackmore and R. B. Knox, eds), pp. 55-94. Academic Press, London. Brown, R. C., and Lemmon, B. E. (1992). Polar organizers in monoplastidic meiosis of hepatics (Bryophyta). Cell Motility and the Cytoskeleton 22, 72-79. Brown, R. C., Lemmon, B. E. and Carothers, Z . B. (1982). Spore wall development in Sphagnum lescurii. Canadian Journal of Botany 60,23962409.

308

R. LIGRONE ef al.

Brown, R. C., Carothers, Z. B. and Duckett, J. G. (1983). Comparative spermatogenesis in the Sphaerocarpales. 11. Blepharoplast structure in Geothallus tuberosus Campb. Bryologist 86,234-243. Browning, A. J . and Gunning B. E. S. (1979a). Structure and function of transfer cells in the sporophyte haustorium of Funaria hygrornetrica Hedw. I. The development and ultrastructure of the haustorium. Journal of Experimental Botany 30, 1247-1264. Browning, A. J. and Gunning B. E. S. (1979b). Structure and function of transfer cells in the sporophyte haustorium of Funaria hygrornetrica Hedw. 11. Kinetics of uptake of labelled sugars and localization of absorbed products by freezesubstitution and autoradiography. Journal of Experimental Botany 30, 12471264. Browning, A . J. and Gunning, B. E. S. (1979~).Structure and function of transfer cells in the sporophyte haustorium of Funaria hygrornetrica Hedw. 111. Translocation of assimilate into the attached sporophyte and along the seta of attached and excised sporophytes. Journal of Experimental Botany 30, 12651273. Campbell, D. H. (1916). The archegonia and sporophyte of Treubia insignis Goebel. American Journal of Botany 2, 199-210. Campbell, D. H. (1918). “The Structure and Development of Mosses and Ferns”. Macmillan, New York. Campbell, D. H. (1920). Studies on some East Indian Hepaticae, Calobryurn blumei N. ab E. Annals of Botany 34, 1-12. Campbell, E. 0. (1954a). The structure and development of Monoclea forsteri Hook. Transactions of the Royal Society of New Zealand 87, 243-244. Campbell, E. 0. (1954b). The structure and development of Calobryurn gibbsiae Steph. Transactions of the Royal Society of New Zealand 87, 245-254. Campbell, E. 0. (1954~).The structure and development of Marchesta areolata Camp. Transactions of the Royal Society of New Zealand 87, 255-262. Carothers, Z . B., Brown, R. C. and Duckett, J. G. (1983). Comparative spermatogenesis in the Sphaerocarpales. I. Blepharoplast structure in Sphaerocarpos and Riella. Bryologist 86, 97-105. Carothers, Z . B. and Rushing, A. E. (1988). Comparative morphology of the bryophyte blepharoplast. Advances in Bryology 3 , 95-134. Caussin, C., Despeghel, J. P., Faucher, M., Leger, A . and Bonnemain, J. L. (1979). Etude du mechanisme des echanges entre le gametophyte et le sporophyte chez les Bryophytes. Cornptes Rendus des Stances de I’Academie des Sciences Paris, Strie D 289, 1329-1334. Caussin, C., Despeghel, J . P. and Bonnemain, J . L. (1982). Absorption des acides aminks neutres par I’haustorium du sporophyte de Polytrichurn forrnosurn: aspects cinitiques et 6nergetiques du transport. Cornptes Rendus des Stances de I’Academie des Sciences Paris, Strie D 294, 725-730. Caussin, C., Fleurat-Lessard, P. and Bonnemain, J. L. (1983). Absorption of some amino acids by sporophytes isolated from Polytrichurn forrnosurn and ultrastructural characteristics of the haustorium transfer cells. Annals of Botany 51, 167-173. Cavers, F. (1911). The inter-relationships of the bryophytes. New Phytologist (Reprint 4), 1-203. Charzynska, M., Murgia, M. and Cresti, M. (1990). Microspore of Secale cereale as a transfer cell type. Protoplasrna 158, 2632. Chauhan, L. (1990). Ultrastructure of sporophyte-gametophyte junction in the moss Phascurn cuspidatum (Pottiaceae, Musci). In “Congress of East Asiatic Bryology” (T. Koponen, ed.), p. 28. Finnish Bryological Society, Helsinki.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

309

Chauhan, E. and Lal, M. (1987). Development of transfer cells in the haustoriumvaginula complex of Physcomitrium cyathicarpum Mitt. : an ultrastructural study, Journal of the Hattori Botanical Laboratory 63, 373-394. Chauhan, E. and Schraudolf, H. (1986). Ultrastructural studies on the placental region in Anthoceros punctatus L. Beitrage zur Biologie der Pfanzen 61, 357-372. Clapp, G. L. (1912). The life history of Aneura pinguis. Botanical Gazette 54, 177-1 93. Corley, M. F. V., Crundwell, A. C., Dull, R., Hill, M. 0.and Smith, A. J . E. (1981). Mosses of Europe and the Azores; an annotated list of species, with synonyms from the recent literature. Journal of Bryology 11,609-689. Courtice, G. R. M., Ashton, N. W. and Cove, D. J. (1978). Evidence for the restricted passage of metabolites into the sporophyte of the moss Physcomitrella patens (Hedw.) Br. Eur. Journal of Bryology 10, 191-198. Crandall-Stotler, B. (1980). Morphogenetic designs and a theory of bryophyte origins and divergence. Bioscience 30, 580-585. Crandall-Stotler, B. (1981). Morphologylanatomy of hepatics and anthocerotes. Advances in Bryology 1 , 315-398. Crandall-Stotler, B. (1984). Musci, hepatics and anthocerotes-An essay on analogues. In “New Manual of Bryology” (R. M. Schuster, ed.), Vol. 2, pp. 1093-1129. Hattori Botanical Laboratory, Nichinan. Crandall-Stotler, B. and Geissler, P. (1983). Sporophyte development and structure in Spruceanthus marianus (Gott.) Mizut., with special reference to capsule wall differentiation. Candollea 38, 105-124. Crandall-Stotler, B. and Guerke, W. R. (1980). Developmental anatomy of Jubula (Durn.) Hepaticae. Bryologist 83, 179-201. Crane, P. R. (1985). Phylogenetic analysis of seed plants and the origin of angiosperms. Annals of the Missouri Botanical Garden 72, 71C793. Crosby, M. R. (1980). The diversity and relationships of mosses. In “The Mosses of North America” (R. J. Taylor and A. E. Leviton, eds), pp. 115-129. Pacific Division, AAAS, San Francisco, CA. Davey, M . R. and Street, H. E. (1971). Studies on the growth in culture of plant cells. IX. Additional features of the fine structure of Acer pseudoplatanus L. cells cultured in suspension. Journal of Experimental Botany 22,90-95. Davison, P. G . , Smith, D. K. and McFarland, K . D. (1989). The discovery of antheridia in Takakia. American Botanical Society Bulletin 36, 65. Delwiche, C. F., Graham, L. and Thomson, N . (1989). Lignin-like compounds and sporopollenin in Coleochaete, an algal model for land plant ancestry. Science 245,399-401. De Maggio, A. E. (1963). Morphogenetic effects influencing the development of fern embryos. Journal of the Linnean Society (Botany) 58,361-376. Despeghel, J. P. and Delrot, S. (1983). Energetics of amino acid uptake by Viciafaba tissue. Plant Physiology 71, 1-6. Duckett, J. G., Carothers, Z. B. and Miller, C. C. J. (1982). Comparative spermatology and bryophyte phylogeny. Journal of the Hattori Botanical Laboratory 53, 107-125. Duckett, J. G., Carothers, Z. B. and Miller, C. C. J. (1984). Gametogenesis. In “New Manual of Bryology” (R. M. Schuster, ed.), Vol. 1 , pp. 232-275. Hattori Botanical Laboratory, Nichinan. Duckett, J. G. and Ligrone, R. (1992). A light and electron microscope study of the fungal endophytes in the sporophyte and gametophyte of Lycopodium cernuum L. with observations on the gametophyte-sporophyte junction. Canadian Journal of Botany 70, 58-72.

310

R. LIGRONE et al.

Duckett, J . G . and Renzaglia, K. S. (1988a). Cell and molecular biology of bryophytes: ultimate limits to the resolution of phylogenetic problems. Botanical Journal of the Linnean Society 98,225-246. Duckett. J. G. and Renzaglia, K. S . (1988b). Ultrastructure and development of plastids in bryophytes. Advances in Bryology 3, 33-93. Dupler, A. W. (1922). Early embryogeny of Reboulia hemisphaerica. Botanical Gazette 74, 143-157. Durand, E. J. (1908). The development of sexual organs and sporogoniuni of Marchantia polymorpha. Bulletin of the Torrey Botanical Club 35, 321-325. Dute, R. R., Peterson, C. M. and Rushing, A. E. (1989). Ultrastructural changes of the egg apparatus associated with fertilization and proembryo development of soybean, Glycine max (Fabaceae). Annals of Botany 64, 123-135. Edwards, S . R. (1984). Homologies and interrelationships of moss peristomes. In “New Manual of Bryology” (R. M. Schuster, ed.), Vol. 2, pp. 658-695. Hattori Botanical Laboratory, Nichinan. Eyme, J. and Suire, C. (1967). Au suject de I’infrastructure des cellules de la region placentaire de Mnium cuspidaturn Hedw. (Mouse bryale acrocarpe). Comptes Rendus des Stances de 1’Academie des Sciences Paris, Strie D 265,1788-1791. Folsom, M. W. and Cass, D. D. (1986). Changes in transfer cell distribution in the ovule of soybean after fertilization. Canadian Journal of Botany 64,965-972. Folsom, M. W. and Petersen, C. M. (1984). Ultrastructural aspects of the mature embryo sac of soybean, Glycine max Merc. Botanical Gazette 145, 1-10. Fransz, P. F. and Schel, J. H. N. (1991). An ultrastructural study on the early development of Zea mays somatic embryos. Canadian Journal of Botany 69, 858-865. Fritsch, F. E. (1945). Studies in the comparative morphology of the algae. IV. Algae and archegoniate plants. Annals of Botany 9, 1-29. Gambardella, R. (1987). Ultrastructure and development of the garnetophyte vaginula-sporophyte foot complex in the liverwort Targionia hypophylla L. Planta 172,431-438. Gambardella, R. and de Lucia Sposito, M. L. (1981-1982). Ultrastructure of the placental region in a liverwort Mannia androgyna. Delpinoa 23-24, 177-184. Gambardella, R. and de Lucia Sposito, M. L. (1983). Placenta ultrastructure in Plagiochasma rupestre (Forst.) Steph. (Marchantiales, Hepaticae). Giornale Botanic0 Ztaliano 117, 16C167. Gambardella, R. and Ligrone, R. (1987). The development of the placenta in the anthocerote Phaeoceros laevis L., Prosk. Planta 172, 439-447. Gambardella, R . , Ligrone, R. and Castaldo, R. (1981). Ultrastructure of the sporophyte foot in Phaeoceros. Cryptogamie, Bryologie et Lichtnologie 2, 177-1 84. Gates, J . C. and Greenwood, M. S. (1991). The physical and chemical environment of the developing embryo of Pinus. American Journal of Botany 78, 10021009. Gifford, E. M. and Foster, A. S. (1989). “Morphology and Evolution of Vascular Plants”. W. H. Freeman and Company, New York. Goebel, K. (1891). Morphologische und biologische Studien. IV. Uber javanische Lebermoose. Annales du Jardin botunique de Buitenzorg 9, 1-40. Goebel, K. (1905). “Organography of Plants, especially of the Archegoniatae and Sperrnatophyta, I1 Special Organography”. (I. B. Balfour, Trans.) Clarendon Press, Oxford. Graham, L. E. (1982). The occurrence, evolution and phylogenetic significance of parenchyma in Coleochaete Breb. (Chlorophyta). American Journal of Botany 69,447-454.

T H E GAMETOPHYTE-SPOROPHYTE JUNCTION

311

Graham, L. E. (1984). Coleochaete and the origin of land plants. American Journal of Botany 71, 603-608. Graham, L. E. (1985). The origin of the life cycle of land plants. American Scientist 73, 178-186. Graham. L. E., Delwiche, C. F. and Mishler, B. D. (1991). Phylogenetic connections between the “Green Algae” and the “Bryophytes”. Advances in Bryology 4, 213-244. Graham, L. E. and McBride, G. E. (1979). The occurrence and phylogenetic significance of a multilayered structure in Coleochaete spermatozoids. American Journal of Botany 66,887-894. Graham, L. E. and Taylor, C . , 111. (1986). The ultrastructure of meiospores of Coleochaete pulvinata (Charophyceae). Journal of Phycology 22, 299-307. Graham, L E. and Wedemayer, G. J. (1984). Spermatogenesis in Coleochaete pulvinata (Charophyceae): sperm maturation. Journal of Phycology 20, 302309. Graham, L. E. and Wilcox, L. W. (1983). The occurrence and phylogenetic significance of putative placental transfer cells in the green alga Coleochaete. American Journal of Botany 70, 113-120. Grolle, R. (1983). Hepatics of Europe including the Azores: an annotated list of species, with synonyms from the recent literature. Journal of Bryology 12, 403-459. Gunning, B. E. S. (1977). Transfer cells and their roles in transport of solutes in plants. Science Progress, Oxford 64, 539-568. Gunning, B. E. S. and Pate, J . S. (1969a). “Transfer Cells”. Plant cells with wall ingrowths, specialized in relation to short distance transport of solutes-their occurrence, structure and development. Protoplasma 68, 107-134. Gunning, B. E. S. and Pate, J. S. (1969b). Cells with wall ingrowths (transfer cells) in the placenta of ferns. Planta 87, 271-274. Gunning, B. E. S. and Pate, J . S. (1974). Transfer cells. I n “Dynamic Aspects of Plant Ultrastructure” (A. W. Robards, ed.), pp. 441-480. McGraw-Hill, London. Gunning, B. E. S . , Pate, J. S. and Briarty, L. G. (1968). Specialized “transfer cells” in minor veins of leaves and their possible significance in phloem translocation. Journal of Cell Biology 37, C7-12. Hasegawa, J. (1988). A proposal for a new system of the Anthocerotae, with a revision of the genera. Journal of the Hattori Botanical Laboratory 64,87-95. Haupt, A. W. (192g). Morphology of Preissia quadrata. Botanical -Gazette 82, 30-54. Haupt, A. W. (1929a). Studies in Californian Hepaticae. I . Asterella californica. Botanical Gazette 87, 302-318. Haupt, A. W. (1929b). Studies in Californian Hepaticae. 11. Fossombronia longiseta. Botanical Gazette 88, 103-108. HCbant, C. (1975). Organization of the conducting tissue-system in the sporophytes of Dawsonia and Dendroligotrichum (Polytrichales, Musci). Journal of the Hattori Botanical Laboratory 39, 235-254. Hebant, C. (1977). “The Conducting Tissues of Bryophytes”. Bryophytorum Bibliotheca 10, 1-157. J . Cramer, Vaduz. Hebant, C. (1979). Conducting tissues in bryophyte systematics. I n “Bryophyte Systematics” (G. C. S. Clarke and J. G. Duckett, eds), pp. 365-383. Academic Press, London. Henry, Y. and Steer, M. (1980). A re-examination of the induction of phloem transfer cell development in pea leaves Pisum sativum. Plant, Cell and Environment 3 , 377-380.

312

R. LIGRONE ef al.

Holloway, J. E. (1918). The prothallus and young plant of Tmesipteris. Transactions of the New Zealand Institution 50, 1-44. Johnson, D. S. (1904). The development and relationship of Monoclea. Botanical Gazette 38, 185-205. Kapil, R. N. and Bhatnagar, A. K. (1981). Ultrastructure and biology of female gametophyte in flowering plants. International Review of Cytology 70, 291341. Kelley, C. (1969). Wall projections in the sporophyte-gametophyte junction of Sphaerocarpos. Journal of Cell Biology 41, 91&914. Kennell, J. C. and Horner, H. T. (1985a). Influence of the soybean male-sterile gene (ms.) on the development of the female gametophyte. Canadian Journal of Genetics and Cytology 27,200-209. Kennell, J. C . and Horner, H. T. (1985b). Megasporogenesis and megagametogenesis in soybean, GIycine m a . American Journal of Botany 72,1553-1564. Khatoon, K. (1986). Occurrence of transfer cells in the sporophyte of Pteridium aquilinum L. Pakistan Journal of Botany 18, 9-13. Kinraide, T. B., Newman, I. A. and Etherton, B. (1984). A quantitative simulation model for H-amino acid cotransport to interpret the effects of amino acids on membrane potential and extracellular pH. Plant Physiology 76, 80G813. Krassilov, V. A. and Schuster, R. M. (1984). Paleozoic and mesozoic fossils. In “New Manual of Bryology”. (R. M. Schuster, ed.), Vol. 2, pp. 1172-1193. Hattori Botanical Laboratory, Nichinan. Kumar, S. S. (1984). An approach towards phylogenetic classification of mosses. Journal of the Hattori Botanical Laboratory 55, 219-226. Lal, M. and Chauhan, E. (1981). Transfer cells in the sporophyte-gametophyte junction of Physcomitrium cyathicarpum. Protoplasma 107, 79-83. Lal, M. and Narang, A. (1985). Ultrastructural and histochemical studies of transfer cells in the callus and apogamous sporophytes of Physcomitrium coorgense Broth. New Phytologist 100, 225-231. Lewis, C. E. (1906). The embryology and development of Riccia lutescens and Riccia crystallina. Botanical Gazette 41, 109-138. Ligrone, R. and Gambardella, R. (1988a). The sporophyte-gametophyte junction in bryophytes. Advances in Bryology 3,225-274. Ligrone, R. and Gambardella, R. (1988b). The ultrastructure of the sporophytegametophyte junction and its relationship to bryophyte evolution. Journal of the Hattori Botanical Laboratory 64, 187-196. Ligrone, R. and Renzaglia, K. S. (1989). The ultrastructure of the placenta in Sphagnum. New Phytologist 111, 197-201. Ligrone, R. and Renzaglia, K. S. (1990). The sporophyte-gametophyte junction in the hornwort, Dendroceros tubercularis Hatt. (Anthocerotophyta). New Phytologist 114, 497-505. Ligrone, R., Gambardella, R., Castaldo, R., Giordano, S. and de Lucia Sposito, M. L. (1982a). Gametophyte and sporophyte ultrastructure in Buxbaumia piperi Best (Buxbaumiales, Musci). Journal of the Hattori Botanical Laboratory 52, 465-499. Ligrone, R., Gambardella, R. and de Lucia Sposito, M. L. (1982b). Ultrastructure of the sporophyte foot-gametophyte vaginula complex in Timmiella barbuloides (Brid.) Moenk. Planta 154, 414425. Maier, K. (1967). Wandlabyrinthe im Sporophyten von Polytrichum. Planta 77, 108-126. Maier, K. and Maier, U. (1972). Localization of beta-glycerophosphatase and Mg+-activated adenosine triphosphatase in a moss haustorium, and the relation of these enzymes to the cell wall labyrinth. Protoplasma 75, 91-112.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

313

Mansfield, S. G . and Briarty, L. G. (1990a). Development of the free-nuclear endosperm in Arabidopsis thalian’a (L.)’ArabidopsisInformation Service 27, 53-64. Mansfield, S. G. and Briarty, L. G . (1990b). Endosperm cellularization in Arabidopsis thaliana L. Arabidopsis Information Service 27, 65-72. Mansfield, S. G. and Briarty, L. G . (1991). Early embryogenesis in Arabidopsis thaliana. 11. The developing embryo. CanadianJournal of Botany 69,461-476. Mansfield, S. G., Briarty, L. G. and Erni, S. (1991). Early embryogenesis in Arabidopsis thaliana I. The mature embryo sac. Canadian Journal of Botany 69,447460. Marinos, N. G. (1970). Embryogenesis of the pea (Pisum sativum). 11. An unusual type of plastid in the suspensor cells. Protoplasma 71, 227-233. Marsh, B. H. and Doyle, W. T. (1985). Intercellular protein crystals from the gametophyte-sporophyte junction of the hornwort Phaeoceros laevis Prosk. Protoplasma 129,223-226. McCormick, F. A. (1914). A study of Symphyogyna aspera. Botanical Gazette 58, 401-418. McFarland, K. D . , Smith, D. K. and Davison, P. G . (1989). Observations on antheridial and apical organization in Takakia ceratophylla. American Journal ofBotany (Suppl.) 76, 11. McNaught, H. L. (1929). Development of sporophyte on Marchantia chenopoda. Botanical Gazette 88, 400-416. Mehra, P. N. and Kumar, D. (1990). Some observations on the embryology of Calobryum indicum. Journal of the Hattori Botanical Laboratory 68,239-253. Menon, M. K . and Bell, P. R. (1981). Ultrastructural and cytochemical aspects of induced apogamy following abscissic acid pre-treatment of secondary moss protonema. Planta 151, 427-433. Meyer, K. (1912). Untersuchungen iiber den Sporophyt der Lebermoose. Bulletin Societe Imperiale des Naturalistes de Moscou 25, 263-286. Meyer, K. (1914). Untersuchungen iiber den sporophyte der Lebermoose 111. Berichte der Deutschen Botanischen Gesellschaft 32, 262-266. Meyer, K. (1929). Die Entwicklung des Sporogones bei Fegatella conica (Untersuchungen iiber den sporophyte der Lebermoose IV). Planta 13, 193-209. Meyer, K. (1931). Zur Entwicklungsgeschichte der Sporophyten einiger Marchantiales. Planta 13, 193-209. Mikeswell, J. (1990). Anatomy of terminal haustoria in the ovule of plantain (Plantago major L.) with taxonomic comparisons to other angiosperm taxa. Botanical Gazette 151,452-464. Mishler, B. D. and Churchill, S. P. (1984). A cladistic approach to the phylogeny of the “bryophytes”. Brittonia 36, 406424. Mishler, B. D. and Churchill, S. P. (1985). Transition to a land flora: phylogenetic relationships of the green algae and bryophytes. Cladistics 1, 305-328. Mogensen, H. L. (1972). Fine structure and composition of the egg apparatus before and after fertilization in Quercus gambelii: the functional ovule. American Journal of Botany 59, 931-941. Mogensen, H. L. and Suthar, H. K. (1979). Ultrastructure of the egg apparatus of Nicotiana tabacum (Solanaceae) before and after fertilization. Botanical Gazette 140, 168-179. Mueller, W. C. and Beckman, C. H. (1984). Ultrastructure of the cell wall of vessel contact cells in the xylem of tomato stems. Annals of Botany 53, 107-114. Murgia, M., Charzynska, M., Rougier, M. and Cresti, M. (1991). Secretory tapetum of Brassica oeracea L. : polarity and ultrastructural features. Plant Sexual Reproduction 4,28-35.

314

R. LIGRONE et al.

Murray, B. M. (1988). Systematics of the Andreaeopsida (Bryophyta): two orders with links to Takakia. Beiheft zur Nova Hedwigia 90,289-336. Nagl, W. and Kuhner, S. (1976). Early embryogenesis in Tropaeolum majus L.: diversification of plastids. Planta 133, 15-19. Newcomb, W. (1973a). The development of the embryo sac of sunflower Helianthus annuus before fertilization. Canadian Journal of Botany 51, 863-878. Newcomb, W. (1973b). The development of the embryo sac of sunflower Helianthus annuus after fertilization. Canadian Journal of Botany 51, 879-890. Newcomb, W. and Fowke, L. C. (1974). Stellaria media embryogenesis: the development and ultrastructure of the suspensor. Canadian Journal of Botany 52,607-614. Newcomb, W. and Steeves, T. A. (1971). Helianthus annuus embryogenesis: embryo sac wall projections before and after fertilization. Botanical Gazette 132,367-371. O’Keeffe, L. (1915). Structure and development of Targionia hypophylla. New Phytologist 14, 105-116. Oliver, M. J. and Bewley, J. D. (1984). Desiccation and ultrastructure in bryophytes. Advances in Bryology 2, 91-111. Olson, A. R. (1991). Post fertilization changes in ovules of Monotropa uniflora L. (Monotropaceae). American Journal of Botany 78,99-107. Pacini, E. (1990). Tapetum and microspore function. In “Microspores, Evolution and Ontogeny” (S. Blackmore and R. B. Knox, eds), pp. 213-237. Academic Press, London. Pagan, F. M. (1932). Morphology of the sporophyte of Riccia crystallina. Botanical Gazette 93,71-84. Pate, J. S. and Gunning, B. E. S. (1972). Transfer cells. Annual Review of Plant Physiology 23, 173-196. Pate, J. S., Gunning, B. E. S. and Milliken, F. F. (1970). Function of transfer cells in the nodal region of stems, particularly in relation to the nutrition of young seedlings. Protoplasma 71,313-334. Peterson, R. L. and Whittier, D. P. (1991). Transfer cells in the sporophytegametophyte junction of Lycopodium appressum. Canadian Journal of Botany 69,222-226. Proctor, M. C. F. (1977). Evidence on the carbon nutrition of moss sporophytes from 14C02 uptake and subsequent movement of labelled assimilate. Journal of Bryology 9, 375-386. Proskauer, J. (1961). On Carrpos I. Phytomorphology 11,359-378. Renault, S., Despeghel-Caussin, C . , Bonnemain, J. L. and Delrot, S. (1989). The proton electrochemical transmembrane gradients generated by the transfer cells of the haustorium of Polytrichum formosum and their use in the uptake of amino acids. Plant Physiology 90,913-920. Renzaglia, K. S. (1978). A comparative morphology and developmental anatomy of the Anthocerotophyta. Journal of the Hattori Botanical Laboratory 44, 31-90. Renzaglia, K. S. (1982). A comparative developmental investigation of the gametophyte generation in the Metzgeriales (Hepatophyta). Bryophytorum Bibliotheca 24, 1-253. Renzaglia, K. S., Carothers, Z . B. and Duckett, J . G. (1985). Comparative ultrastructural studies of spermatogenesis in the Metzgeriales (Hepaticae). I. The blepharoplast of Pallavicinia lyellii. American Journal of Botany 74, 53-64. Renzaglia, K. S. and Duckett, J. G . (1987a) Spermatogenesis of Blasiapusilla from antheridial initial through mature spermatozoid. Bryologist 90, 468-501. Renzaglia, K. S. and Duckett, J. G. (1987b). Comparative ultrastructural studies of

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

315

spermatogenesis in the Metzgeriales (Hepatophyta). 11. The blepharoplast of Blasia pusilla. American Journal of Botany 74, 53-64. Renzaglia, K. S. and Duckett, J . G . (1988). Different developmental processes underlie similar spermatozoid architecture in mosses, liverworts and hornworts. Journal of the Hattori Botanical Laboratory 64, 219-236. Renzaglia, K. S. and Duckett, J. G. (1989). Ultrastructural studies of spermatogenesis in the Anthocerotales. V. The posterior mitochondrion and nuclear metamorphosis in Notothylas and Phaeoceros. Protoplasma 51, 137-150. Renzaglia, K. S. and Duckett, J. G. (1991). Towards an understanding of the differences between the blepharoplasts of mosses and liverworts, and comparisons with hornworts, biflagellate lycopods and charophytes: a numerical analysis. New Phytologist 117, 187-208. Renzaglia, K. S., Smith, D. K . , McFarland, K. D. and Davison, P. G . (1992). Ultrastructure of the gametophyte/sporophyte junction in Takakia ceratophylla. American Journal of Botany (in press). Robinson, H. and Shaw, J. (1984). Considerations on the evolution of the moss operculum. Bryologist 87,293-296. Roth, D. (1969). Embryo und Embryotheca bei den Laubmoosen. Eine histogenetische und morphologische Untersuching. Bibliotheca Botanica 129, 1-49. Schel, J. H. N. and Kieft, H. (1986). An ultrastructural study of embryo and endosperm development during in vitro culture of maize ovaries (Zea mays). Canadian Journal of Botany 64,2227-2328. Schertler, M. M. (1979). Development of the archegonium and embryo in Lophocolea heterophyl la. Bryologist 82, 576-582. Schnepf, E., and Nagl, W. (1970). Uber einige Strukturbesonderheiten der Suspensorzellan von Phaseolus vulgaris. Protoplasma 69, 133-143. Schofield, W. B. (1985). “Introduction to Bryology”. MacMillan, New York. Schultz, S. R. and Jensen, W. A. (1968a). Capsella embryogenesis: the synergids before and after fertilization. American Journal of Botany 55,541-552. Schultz, S. R. and Jensen, W. A. (1968b). Capsella embryogenesis: the egg, zygote and young embryo. American Journal of Botany 55,807-819. Schultz, S . P. and Jensen, W. A. (1969). Capsella embryogenesis: the suspensor and basal cell. Protoplasma 67, 139-163. Schultz, P. and Jensen, W. A. (1971). Capsella embryogenesis: the chalaza1 proliferating tissue. Journal of Cell Science 8, 201-227. Schultze-Motel, W. (1970). Monographie der Laubmoosgattung Andreaea. I . Die costaten Arten. Willdenowia 6, 25-110. Schuster, R. M. (1966). “The Hepaticae and Anthocerotae of North America”, Vol. I. Columbia University Press, New York. Schuster, R. M. (1979). The phylogeny of the Hepaticae. I n “Bryophyte Systematics” ( G . C. S. Clarke and J. G . Duckett, eds), pp. 41-82. Academic Press, London. Schuster, R. M. (1984a). Comparative anatomy and morphology of the Hepaticae. I n “New Manual of Bryology” (R. M. Schuster, ed.), Vol. 2, pp. 760-891. Hattori Botanical Laboratory, Nichinan. Schuster, R. M. (1984b). Evolution, phylogeny and classification of the Hepaticae. In “New Manual of Bryology” (R. M. Schuster, ed.), Vol. 2, pp. 892-1070. Hattori Botanical Laboratory, Nichinan. Schuster, R. M. (1984~).Morphology, phylogeny and classification of the Anthocerotae. In “New Manual of Bryology” (R. M. Schuster, ed.), Vol. 2, pp. 1071-1092. Hattori Botanical Laboratory, Nichinan. Sekhar, K . N. C. and DeMason, D. A. (1989). Differential activity of acid

316

R. LIGRONE et al.

phosphatases from the endosperm and haustorium of date palm (Phoenix dactylijera) seeds. Canadian Journal of Botany 67, 1096-1 102. Shaw, J. and Anderson, L. E. (1988). Peristome development in mosses in relation to systematics and evolution. 11. Tetraphispellucida (Tetraphidaceae). American Journal of Botany 75, 1019-1032. Shaw, J., Anderson, L. E. and Mishler, B. D. (1987). Peristome development in mosses in relation to systematics and evolution. I. Diphyscium foliosum (Buxbaumiaceae). Memoirs of the New York Botanical Garden 45,55-70. Sheffield, E. and Bell, P. R. (1987). Current studies of the pteridophyte life cycle. Botanical Review 53,442-490. Showalter, A. M. (1926). Studies in the cytology of the Anacrogynae. 11. Fertilization in Riccardia pinguis. Annals of Botany 40, 713-726. Showalter, A. M. (1927a). Studies in the cytology of the Anacrogynae. 111. Fertilization in Fossombronia angulosa. Annals of Botany 41, 3 7 4 6 . Showalter, A. M. (1927b). Studies in the cytology of the Anacrogynae. IV. Fertilization in Pellia fabbroniana. Annals of Botany 41, 409-417. Sluiman, H. J. (1985). A cladistic evaluation of the lower and higher green plants (Viridiplantae).Plant Systematics and Evolution 149, 217-232. Smart, M. G. and O’Brien, T. P. (1983). The development of the wheat embryo in relation to the neighbouring tissues. Protoplasma 114, 1-13. Smith, A. J. E. (1978). “Moss Flora of Britain and Ireland”. Cambridge University Press, Cambridge. Smith, D. K. (1990). Sporophytes of Takakia discovered. Bryological Times 57/58, 1-4. Smith, G. L. (1955). “Cryptogamic Botany”. McGraw-Hill, New York. Smith, G. L. (1971). Conspectus of the genera of Polytrichaceae. Memoirs of the New York Botanical Garden 21, 1-83. Snider, J. A. (1975). Sporophyte development in the genus Archidium (Musci). Journal of the Hattori Botanical Laboratory 39, 85-104. Steere, W. C. and Murray, B. M. (1976). Andreaeobryum macrosporum, a new genus and species of Musci from Northern Alaska and Canada. Phytologia 33, 407410. Stotler, R. E. and Crandall-Stotler, B. (1977). A checklist of the liverworts and hornworts of North America. Bryologist 80, 405428. Thomas, R. J . , Stanton, D. S., Longendorfer, D. H. and Farr, M. E. (1978). Physiological evaluation of the nutritional autonomy of a hornwort sporophyte. Botanical Gazette 139, 306-311. Thomas, R. J., Stanton, D. S. and Grusak, M. A. (1979). Radioactive tracer study of sporophyte nutrition in hepatics. American Journal of Botany 66, 398-403. Tilton, V. R., Wilcox, L. E. and Palmer, R. G. (1984). Post-fertilization wandlabyrinthe formation and function in the central cell of soybean, Glycine max (L.) Merr. (Leguminosae). Botanical Gazette 145,334-339. Vaizey, J. R. (1888). On the anatomy and development of the sporogonium of the mosses. The Journal of Linnean Society 24, 262-285. Vaughn, K. C., Ligrone, R., Owen, H. A., Hasegawa, J., Campbell, E. O., Renzaglia, K. S. and Monge-Najera, J. (1992). The anthocerote chloroplast: a review. New Phytologist 120, 169-190. Vitt, D. H. (1984). Classification of the Bryopsida. In “New Manual of Bryology” (R. M. Schuster, ed.), Vol. 2, pp. 696759. Hattori Botanical Laboratory, Nichinan. Wardlaw, C. W. (1965). Physiology of embryonic development in cormophytes. In “Handbook of Plant Physiology XV/l” (W. Ruhland, ed.), pp. 843-965. -Springer-Verlag, Berlin.

THE GAMETOPHYTE-SPOROPHYTE JUNCTION

317

Wiencke, C. and Schulz, D. (1975). Sporophytenentwichlung von Funaria hygrometrica Sibth. I. Strukturelle Grundlagen der Wasser- und Nahrstoffaufnahme im Haustorium. Protoplasma 86, 107-117. Wiencke, C. and Schulz, D. (1978). The development of transfer cells in the haustorium of the Funaria hygrometrica sporophyte. Bryophytorum Bibliotheca 13,147-167. Willemse, M . T. M. and Van Went, J. L. (1984). The female gametophyte. In “Embryology of Angiosperms” (B. M. Johri, ed.), pp. 15S196. SpringerVerlag, Berlin. Wise, R. R. and Juncosa, A. M. (1989). Ultrastructure of the transfer tissues during viviparous seedling development in Rhizophora mangle (Rhizophoraceae). American Journal of Botany 76, 1286-1298. Yan, H., Yang, H.-Y. and Jensen, W. A. (1991). Ultrastructure of the developing embryo sac of sunflower (Helianthus annuus) before and after fertilization. Canadian Journal of Botany 69, 191-202. Yeung, E. C. and Clutter, M . E. (1979). Embryogeny of Phaseolus coccineus: the ultrastructure and development of the suspensor. Canadian Journal of Botany 57,120-136. Zodda, G. (1934). “Flora Italica Cryptogama. Pars IV. Bryophyta, Hepaticae”. Rocca S. Casciano.

This Page Intentionally Left Blank

AUTHOR INDEX

A Abeles, F.B., 72, 77 Abou-Mandour, A., 156,173 Adfim, A., 52, 77 Adams, D.O., 147,168 Adams, M.J., 89 Addicott, F.T., 150,168 Adoutte, A., 227 Aharoni, N., 134, 142,168 Ahokas, H., 125,168 Albersheim, P., 2, 3, 8,9,21, 31, 40, 52,69, 74, 75, 77, 78, 79, 80, 82, 83,84, 85, 86, 87,88, 89, 91, 92, 94, 95, 96, 98, 99,100,101 Albert, F., 97 Aldington, S., 2, 5, 22,41, 68, 74, 78 Al-Doori, A.H., 183 Alejar, A.A., 186 Alexopoulos, C.J., 257, 263,307 Altman, D.W., 78 Anderson, A.J., 24,40, 69, 77, 97, 99 Anderson, E., 192,220 Anderson, J.D., 30, 78,85,88 Anderson, L.E., 287,316 Anderson, R.A., 220 Antia, N.J., 208, 220 Apostol, I., 50, 51, 78 Appell, G.S., 208,223 Appels, R., 201,220 Appleford, N.E., 169 Ariztia, E.V., 203, 220 Arteca, R.N., 147, 185 Asamizu, T., 23, 78 Ashton, N.W., 232,309 Asmundson, C.M., 183 Aspinall, G.O., 59, 60, 78 Asselin, A,, 80 Aston, M.J., 155, 168 Atkinson, C.J., 151, 157, 168 Atkinson, M.M., 26, 49, 78, 79 Audren, H., 225

Aurich, O., 183 Awad, M., 72, 79 Ayers, A.R., 2,8, 11,20,73, 78,86

B Babbitt, J.K., 72, 79 Bachellerie, J.P., 227 Backe, M.A., 4, 79 Badenoch-Jones, J., 173 Badwey, J.A., 49, 79 Baenziger, J.U., 6,89 Baer, H.H., 93 Baev, N., 37, 79 Baier, M., 106, 150, 168, 173 Bailey, J.A., 8, 61, 77, 79, 90, 99, 101 Bailey, R.W., 6, 79 Baker, C.J., 26, 79 Baker, K.K., 90 Baldwin, E.A., 31, 79 Bakes, S., 218,229 Banfalvi, Z., 79 Barber, M.S., 28,34,35,43,73,78,80,97 Barbier-Brygoo, H., 44,80 Barlow, B.A., 173 Barns, S., 223 Baroin, A., 227 Barr, M.L., 156,168 Barrett, A.J., 4, 59, 80 Barrett, J., 190, 225 Barry, D.A., 184 Barthe, J.P., 80 Bartholomew-Began, S.E., 269,307 Basham, H.G., 24,25, 69,80 Bashan, Y., 69,80 Bassi, P.K., 145, I68 Bateman, D.F., 24, 25, 69,80 Bates, L.M., 134, 155,168 Bauer, W.D., 12, 14, 63, 80 Bauw, G., 94 Baydoun, E.A.-H., 6,44,60, 66, 75, 76,80

319

320

AUTHOR INDEX

Baynes, J.W., 86 Beakes, G.W., 203,220 Beanland, T.J., 225 Beardmore, M., 34,80 Beckman, C.H., 285,313 Beckman, J.M., 35,89 Beever, J.E., 107, 128,168 Beissman, B., 92 Bell, A.A., 8,80 Bell, J.N., 82,84, 90, 98, 99 Bell, P.R., 232,297,299,307,313,316 Bellincampi, D., 18, 80, 84, 87 BeMiller, J.N., 4, 101 Bendayan, M., 196,220 Benhamou, N., 43,61,80 Bennett, A.B., 72,88 Benson, R.J., I68 Ben-Zioni, A,, I75 Bergami, M., 83 Berger, N., 79 Bergstrom, G.C., 90 Bernasconi, P., 12,81 Bernier, G., 121,169 Bertram, R.E., 80 Bethenod, O., 184 Bevan, M.W., 183 Bewley, J.D., 251,314 Beyer, E.M. Jr., 144,169 Beyl, C., 134,181 Bezukladnikov, P.W., 5, 63,81 Bhandari, N.N., 303,307 Bhatnagar, A.K., 303,312 Bhattacharya, D., 203,220 Biddington, N.L., 104,169 Bierhorst, D.W., 297,301,307 Biggs, K.J., 88 Biggs, R.H., 31, 79 Biles, C.L., 72, 77 Bing-Quan Huang, 303,307 Birberg, W., 83 Bird, P.M., 8,34,81 Bishop, P., 98 Bishop, P.D., 4, 19, 20, 67, 72, 75, 81 Bisseling, T., 89 Black, W.C., 94 Blackman, P.G., 134, 155,169 Blaikie, S.J., 104, 169 Blake, D.A., 6, 81 Blake, T.J., 114, 115, I84 Blank, J., 229 Blaschek, W., 92 Bledsoe, C.S., 171 Blom, C.W.P.M., 185

Blum, A., 159,169 Blum, U., 4,81 Blumenfeld, A . , I68 Boczar, B.A., 212,220 Boffey, S.A., 67,81 Bogemann, G.M., 185 Bogorad, L., 200,220,221 Bohdanowicz, J., 303, 306,307 Bold, H.C., 233,257,263,307 Boller, T., 30, 31, 73,81, 83, 90, 94, 96, 98,100 Bollmann, J., 99 Bollmark, M., 129, 169 Bolwell, G.P., 12,81,82,86 Bonig, I., 226 Bonnemain, J.L., 232,285,308,314 Bonnen, A.M., 90 Bonner, B.A., 100 Bonner, D.M., 124,186 Booij, H., 84 Boon, J.J., 89 Bostock, R.M., 20,40, 52,82 Bottger, M., 156,169 Bowen, M.R., 140,174,177 Bower, F.O., 233,307 Bowles, D.J., 18, 19,82,86,100,101 Bowman, Y.J.L., 87 Boyer, J.S., 169 Bradford, K.J., 107, 110, 137, 138, 146, 147, 149, 160, 161,169 BrameLCox, P., 94 Branca, C., 80,87 Branca, C.A., 4, l8,82 Brecht, J.K., 72, 82 Breck, E., 172 Bremer, K., 233,307 Brenner, M.L., 173,183 Brenzel, A., 228 Brett, C.T., 60, 82 Briarty, L.G., 285, 303, 305, 306,311, 313 Brinker, A.M., 178 Brinkerhoff, L.A., 90 Broadwater, S.T., 227 Brodelius, P., 88 Brodelius, P.E., 96 Broekaert, W.F., 3,82 Brown, R.C., 237,251,271,283, 287, 291,293,307,308 Browning, A.J., 232,234,237, 251, 271,273,308 Bruce, R.J.,21, 28, 68,82 Brun, W.A., 173,183

AUTHOR INDEX

32 1

Chailakhian, M.Kh., 121,170 Chamberland, H., 80 Chang, D.D., 200,221 Chapin, F.S. 11, 164,170 Chappell, J., 8, 30, 31,83 Charzynska, M., 285, 301,308,313 Chauhan, E., 237,251, 276,309,312 Chauhan, L., 237,308 Chelf, P., 100 Chen, C.-M., 124, 170 Cheng, J.Y., 220 Cheong, J.-J., 5 , 9, 11,42,83 Chibnall, A.C., 117, 171 Chihara, M., 229 Chuan-Jin, H., 179 C Cahill, D.M., 109, 135,170 Churchill, S.P., 233, 289,307,313 Cain, J.R., 208,226 Clapp, G.L., 253,309 Clarke, A.E., 43, 90, 226 Callow, J.A., 34, 43, 49,82,86, 94 Clarkson, D.T., 170,182 Camardella, L., 82 Clayton, D.A., 200,221 Cammack, R., 225 Campbell, A.D., 30,82 Cline, K., 9, 74,83 Close, T.J., 179 Campbell, A.K., 92 Campbell, B.C., 67,82 Clutter, M.E., 303,317 Cohen, D.B., 156,171 Campbell, D.H., 253,275,308 Campbell, D.J., 130, 137, 143, 144, Coleman, M.D., 110, 132, 135,171 Collmer, A., 23, 68, 78,83 146,175, 176 Conley, P.B., 223 Campbell, E.O., 253,255,279,291, Constabel, F., 12, 83 308,316 Cannell, R.Q., 106, 170 Coombe, B.G., I72 Canny, M.J., 76, 82, 106,170 Coombs, J., 202,221 Cantrell, M.A., 93 Cooper, R.M., 68,69,83 Caplan, A.B., 94 Corcoran, M.R., 4,84 Carbonneau, R., 185 Cordewener, J., 38,84 Corley, M.F.V., 235, 243,309 Carlson, D.R., 107, 128,170, I73 Carlson, W.C., 131,170 Cornforth, J.W., 168 Cornish, E.C., 226 Carmi, A., 115, 119, 130, 131,170, Cornish, K., 155, 156,171 177,185 Corredor, V., 201,222 Carns, H.R., 168 Cosio, E.G., 42,84 Carothers, Z.B., 271,287,291, 293, C6te, P., 185 295,307,308,309,314 Cotterman, D.C., 170 Carr, D.J., 135, 136, 140,169,170 Cottier, J.P., 182 Cass, D.D., 303, 305,310 Castaldo, R., 237, 249, 251, 276,310, Cottrell, I.W., 59, 78 Courtice, G.R.M., 232,309 312 Cattolico, R.A., 210, 211,220,221, Cousson, A., 100 Coutts, M.P., 160, 171 225,227,228 Caussin, C., 285,308 Cove, D.J., 232,309 Cavalier-Smith, T., 190,200,202,203, Cox, E.R., 215,229 Craig, J.W.T., 78 216,221 Craigie, J.S., 208,221 Cavers, F., 257,308 Cervone, F., 25, 68, 70,71,80,82,83, Cramer, C.L., 54,82,84,86, 98,99 Crandall-Stotler, B., 235, 253, 257, 84,87 263,275,291,293,295,309,316 Chai, H.B., 52,83 Bryan, I.B., 91 Bryant, J.E., 81 Bucheli, P., 25,26, 33,82,85 Bulpin, P.V., 87 Bunce, J.A., 177 Burdett, A.N., 60, 82 Burger-Wiersma, T., 229 Burrows, W.J., 125, 136,169,181 Burton, P., 200,225 Butcher, D.N., 124, 139,169 Buttrose, M.S., 104, 115,169 Byrde, R.J.W., 101 Byrne, H., 100

322

AUTHOR INDEX

Crane, P.R., 233,309 Creelman, R.A., 161, 171 Cresti, M., 285, 301,308,313 Cribbs, D.H., 96 Crocker, W., 187 Croker, S.J., 150,186 Crosby, M.R., 243, 245,247,287,309 Crozier, A., 140, 142, 143,171,180, 181 Crundwell, A.C., 235,243,309 Cueller, R.E., 225 Curtis, W.R., 40, 86 D da Costa, A.R., 171 Dale, J.E., 108, 112, 115, 119, 150, 164, 165,171, 179,183 Dalmon, J., 225 Dalton, B.R., 4,81 Daniell, H., 226 Danks, M.L., 4,84 Darrall, N.M., 118, 132,171 Darvill, A . , 82, 86 Darvill, A.G., 3, 8, 9, 17, 21, 40, 75, 78,83, 84,85,88,89, 91, 92, 95, 96, 98, 99,100,101 Das Gupta, D.K., 171 Dathe, W., 140,171 Davey, J.E., 123, 126,171,185 Davey, M.R., 285,309 Davidson, R.L., 104, 113, 116,171 Davies, L.A., 85 Davies, P.J., 181 Davies, W.J., 105, 109, 110, 134, 138, 151, 152, 154, 155, 156, 157, 158,159,160, 162,168,169, 171,172,180,183, 184,186,187 Davis, K.R., 4, 22, 39,40, 68,69, 78, 84 Davis, L.J., 229 Davis, R.W., 30, 31, 77,86 Davison, P.G., 235, 241,309,313,315 Davison, R.M., 156,171 Dea, I.C.M., 87 Deakin, A.L., 34,85 Dean, J.F.D., 32, 33,85 Dearman, A.S., 104,169 Degra, L., 82,83 De Greef, J.A., 179 de la Cruz, V.F., 224 Delaney, T.P., 211,220,221 Delevoryas, T., 257,263,307 Dell, A., 84, 95, 96,101

De Lorenzo, G., 70,80,82,83,84,87 Delrot, S., 232, 285,309,314 Delseny, M., 97 de Lucia Sposito, M.L., 237,249,251, 259,271,310,312 Delwiche, C.F., 233,291,309,311 De Maggio, A.E., 232,309 DeMason, D.A., 305,315 DenariC, J., 93 De Proft, M.P., 179 Derocher, J., 227 De Ropp, R.S., 117,171 Desiderio, A., 80,87 Desjardins, A.E., 78 Despeghel, J.P., 285,308,309 Despeghel-Caussin, C., 232,285,314 Deuel, H., 17, 18,85 Deverall, B.J., 34,85 De Vries, S . , 84 de Vries, S.C., 27, 38,61,84 de Wit, P.J.G.M., 20, 24, 39, 60, 85, 98 Dey, P.M., 6,85 Diaz, C.L., 43,85 Dickerson, A.G., 32, 91 Dietrich, A., 53,85 Di Gregorio, S . , 94 Dildine, S.L., 98 Dilley, D.R., 146, 169 Dimalla, G.G., 125,185 Dixon, R.A., 2,12,21,23,39, 77,81, 82,85,86,93 Doares, S.H., 25, 78,82,85, 91 Dodge, J.D., 190, 201, 202,203, 209, 215,219,221 Doherty, H.M., 48, 86,100, 101 Doke, N., 12, 51, 52,83,86,88 Domard, A., 91 Doolittle, W.F., 190,223 Doubrava, N., 86, 95 Douglas, S., 208,209, 210, 216,228 Douglas, S.E., 200,201,202,207,210, 211, 212, 215,221,222 D’Ovidio, R., 83,84 Dow, J.M., 49,86 Downton, W.J.S., 178 Doyle, W.T., 281,313 Drennan, D.S.H., 134, 135,171 Drew, M.C., 135, 137, 145, 161,172, 175 Drewes, S.E., 124,185 Driguez, H., 93 Dron, M., 93

AUTHOR INDEX

Druel, L.D., 203,220 Duckett, J.G., 232, 237, 251,271,287, 291, 293, 295,297,299,307, 308,309,310,314,315 Duckham, S.C., 162,172 Dull, R., 235,243,309 Dumbroff, E.B., 171 Dunlop, D.S., 40,86 Dupler, A . W . , 255,310 Durand, E.J., 310 During, H . , 151,177 Durley, R.C., 170, 176 Durnford, D.G., 211,221,222 Dute, R.R., 303, 305,310 Duval, J.C., 225 Dwarte, D., 209,222 Dyer, D.J., 170 Dyer, T . A . , 183 E Eagles, G., 101 Ebel, J . , 8,42, 78,84,86,89, 98 Eberhard, S . , 27,86, 95 Eble, A.S., 4,86 Ecker, J . R . , 30, 31, 77,86 Edelbaum, O., 74,86 Edelmann, H.G., 14,27,62,86 Edwards, K . , 54,86 Edwards, M . , 66,87,223 Edwards, M . R . , 222 Edwards, S.R., 243,287,310 Egelhoff, T . , 217,222 Ehleringer, J.R., 182 Ehrenberg, C.G., 192,222 Eidenbock, M.P., 133,181 Eilert, U., 12, 83 Eisenberg, B.L., 223 Eklund, L., 146,172 El-Beltagy, A S . , 145, I72 Elbrachter, M., 214,228 El Hiweris, S.O., 134, 135,171 Eliasson, L., 169 Else, M . , 109,172 Else, M . A . , 173 Elstner, E.F., 96 Elwood, H . , 224 Elyakova, L.A., 5, 63,81 Emrnerling, M., 14, 15,87 Endre, G., 79 Enea, V., 201,222 Engelbrecht, L., 117, 124, 129,172, 179 English, P.D., 63, 69, 70,87

323

Ephritikhine, G., 80 Epperlein, M.M., 50, 87 Erhlich, H . A . , 228 Ericsson, A . , 178 Erni, S., 305,313 Eschbach, S., 201,204,206,207,222, 224 Esquerre-Tugaye, M.-T., 8, 29,46,80, 87, 96, 97 Etherton, B., 285,312 Ettl, H., 214, 222 Evans, L.V., 217,222 Ewings, D., 100 Eyme, J., 237,310 F Falk, H . , 224 Fantauzzo, F., 186 Fanutti, C., 66, 87 Farkas, T., 77 FarkaS, V., 56, 66, 87 Farmer, E.E., 2, 32, 53,87, 98 Farmer, M . A . , 214,222 Farr, M . E . , 232,316 Farrar, J.F., 118, 172 Farrar, S.C., 118, 172 Faucher, C., 93 Faucher, M., 285,308 F a d , K.F., 140,172 Fautz, E., 89 Feger, M., 89 Ferguson, A . R . , 110, 172 Ferraris, R., I78 Fielding, A . H . , 90, 101 Fields, S.D., 214,222 Filippini, F., 18, 27, 44, 87 Finelli, F., 79 Fischer, R.L., 72,88 Fletcher, J.S., 84 Fleurat-Lessard, P., 232,285,308 Flott, B.E., 95 Flower, D.J., 178 Fluhr, R . , 33, 94 Folsorn, M.W., 303,305,310 Fork, D.C., 209,224 Forsyth, C., 185 Foster, A . S . , 232, 297, 310 Fournier, J., 97 Fowke, L.C., 303,306,314 Foyle, R.A.J., 220 Fransz, P.F., 305,310 Freeling, M . , 181 Freeman, T.E., 133, 178

324

AUTHOR INDEX

Gingrich, J.H., 155,179 Giordano, S., 237,249,251,312 Giovannoni, S.J., 190,223,229 Glazener, J.A., 28,88 Glazer, A.N.?208,209,223 44,45,49,56,57,58,59,60,61, 62, 63, 64,65, '66,68, 74, 75, 76, Godovac-Zimmerman, J., 225 78,80,86,88, 94, 95, 99, I01 Goebel, K., 253,299,310 Golden, S.S., 190,212,227 Frydman, V.M., 142,172 Fubeder, A . , 132,172 Goldstein, I.J., 6, 81 Fuchs, Y . , 32,88 Gollan, T., 109, 151, 155, 156, 172,182 Fiigedi, P., 83 Gollin, D.J., 6,88, 95, 100 Goodwin, J.C., 4, 89 Fukuda, M., 92 Goring, H., 132, 172 Funk, C., 12,88 Furuichi, N . , 24,88 Goss, M.J., 166,172 Fushtey, S.G., 24,88 Govers, F., 37,89 Gowing, D.J.G., 159, 172 Grab, D., 52,89 G Gadelle, A., 97 Graham, L., 291,309 Gage, D.A., 171,186 Graham, L.E., 232,233,291,310,311 Gale, M.D., 177 Grand, C., 98 Gales, K., 176 Granger, J.W., 80 Galston, A.W., 117, I72 Grant, B.R., 170 Gambardella, R., 233,234,235,237, Grantz, D.A., I79 243,245,249,251,259,271, Gray, M.W., 190,222,223,228 273,276,279,281,283,310,312 Green, E.D., 6,89 Gamble, H.R., 85,88 Green, T.R., 19, 75,89 Greenwood, A.D., 192,193,195,202, Gantt, E., 192,209,222 221,223,227,228 Garcia-Garrido, J.M., 88 Greenwood, M.S., 306,310 Garcia-Romera, I . , 69,88 Griffaut, B., 124,173 Gardner, J.M., 24,88 Gardner, P.A., 159, 165, 166,180 Griffiths, H.B., 223 Griggs, P., 110, 173 Garegg, P., 98 Grignon, C., 96 Garegg, P.J., 83, 96 Grisebach, H., 8,42, 53,86,89, 94 Gaskin, P., 174 Grolle, R., 235,311 Gates, J.C., 306,310 Gross, K.C., 85 Gauhe, A . , 93 Grossman, A . R . , 209,217,218,222, Gautier, C., 93 223 Gehri, A., 81 Gruber, T.A.,95 Geissler, P . , 253,309 Geissman, T.A., 84 Grunwald, C., 141,177 Gelfland, D.H., 228 Grusak, M.A., 232,316 Guard-Friar, D., 208, 217,223 Geraeds, C.C.J.M., 98 Guerke, W.R., 257,263,309 Gerrish, C., 85 Guern, J . , 80 Gholson, R.K., 90 Giigler, K., 88 Ghosheh, N.S., 176 Guiamet, J.J., 180 Gibbs, S.P., 190, 192, 195, 201, 202, Guillemaut, P., 226 208, 209, 215,218,222,223, Guinn, G., 181 226,227 Gunderson, J.H., 201,203,224 Gidley, M.J., 87 Gunning, B.E.S.,232,233,234,237, Gifford, E.M., 232,297,310 251,271, 273, 283, 285, 297, Gilboa-Garber, N., 80 301,302,303, 306,308,311,314 Gilkes, N . R . , 62,88 Gutteridge, J.M.C., 49, 90 Gillott, M.A., 192,201, 208,223 Frey, T., 84 Friend, J . , 91 Fritsch, F.E., 233,310 Fry, S.C.,4, 5 , 6 , 7 , 14, 15,25,27,29,

AUTHOR INDEX

H Hadwiger, L., 100, 101 Hadwiger, L.A., 35, 36,44, 55, 89, 92, 94 Hafez, A.M.A., 97 Hagendoorn, M.J.M., 12,20,89 Hahlbrock, K . , 8, 22, 39, 40, 83, 84, 85,89, 99 Hahn, M.G., 3, 8, 11,21,31,42,53, 78,83,89 Hahn, R., 83 Hahne, G., 26,89 Hall, A.E., 134, 155,168 Hall, D.O.,225 Hall, K.C., 119, 148, 150, 161, 164, 173,175, 176, 179 Hall, M.A., 62,88, 145, 160,172,183, 186 Hall, N.A., 7, 89 Hall, P.J., 123, 173 Hall, S.M., 122, 173 Halliwell, B., 49, 90 Halverson, L.J., 39, 90 Hameed, M.A., 115,173 Hammerschmidt, R., 6, 8, 20,29, 90 Hanke, D.E., 67,90 Hanna, R., 87 Hansmann, P., 195, 196,203,204,206, 222,224 Hardy, M.R., 7, 90 Hargreaves, J.A., 61, 90 Harmsen, H., 89 Harper, J.L., 113,173 Harren, F.J.M., 185 Harrington, A., 190,224 Harrison, M.A., 185 Hartman, T., 229 Hartung, W., 106, 110, 150, 151, 152, 155, 156, 168,173,185,186 Harvey, B.M.R., I81 Hasegawa, J . , 235, 275, 279,291, 295, 311,316 Hatakeyama, N., 215,224 Haug, A . , 99 Haupt, A.W., 253,255,311 Haxo, F.T., 209,224 Hayashi, T., 14,43,63,90,92 Heath, T.G., 186 Hebant, C., 235,237,245,269,311 Hedden, P . , 169,177 Hedrick, S.A., 54, 90, 98 Heidstra, R., 89 Heilmeier, H., 173, 185

325

Heindle, J.C., 107, 109, 128, 173 Heinonen, T.Y.K.,228 Heinrichova, K., 176 Heinstein, P.F., 78, 90 Heitefuss, R., 69, 96 Helgeson, J.P., 32,87 Hendrix, D.L., 173 Henis, Y., 80 Henry, Y., 285,311 Henson, I.E., 124, 128, 164,173,177 Hermann, R.K., 177 Herold, A . , 118,173 Heuer, B., 130,170 Hevesi, M., 77 Hewett, E.W., 174 Hibberd, D.J., 213, 215, 217,224 Higgins, V.J., 49, 52, 96 Higuchi, R., 228 Hill, D.R.A., 192, 208, 219, 220,224 Hill, M.O., 235,243,309 Hiller, R.G., 225 Hillman, J.R., 174,180 Hinch, J.M., 43,90 Hinde, R., 217,224 Hinton, D.R., 72, 90 Hiron, R.W.P., 150, 155, 160,173,186 Hislop, E.C., 25,90 Hitchcock, A.E., 187 Hoad, G.V., 106,140, 157,164,174 Hocking, T.J., 164,174 Hodge, S.K., 88 Hoffman, C., 94 Hofmann, J.B., 222 Holliday, M.J., 77, 90 Hollingdale, M., 224 Holloway, J.E., 301,312 Holm, R.E., 142,174 Honeycutt, R.L., 201,220 Hong, N . , 83 Hooykaas, P.J.J.,85 Hopper, D.G., 24, 90 Horgan, J.M., 132, 142,174,181 Horgan, R., 123, 149,174,180,181, 185 Horn, G.T., 228 Horn, M.A., 44, 45, 72,90 Horner, H.T., 303,305,312 Hoson, T., 14, 15,90 Howard, J . , 38, 91 Howe, C.J., 209,225 Howe, T.J., 225 Hsaio, T.C., 149, 160, 161,169 Huang, J-S., 78

326

AUTHOR INDEX

Huber, D.J., 72,82 Hughes, R.K., 32, 91 Huisman, W., 19, 98 Humphries, C.J., 233,307 Humphries, E.C., 115, 117, 118,174 Hunt, R.C., 104, 113,174 Hutto, J.M., 131,176 I Iijima, M., 166, 174 Iizuka, A., 182 Incoll, L.D., 109, 118, 123, 133, 134, 174, I79 Ingestad, T., 113, 116, 118, 119,174 Ingold, A., 224 Ingram, D.S., 94 Ingram, T.J., 140,174 Inouye, D.W., 182 Inouye, I., 229 Isaiah, H., 4, 91 Ishii, S., 33, 91 Ishii, T., 33, 91 Isobe, K., 100 Itai, C., 125, 133, 134,175, 183 Ito, Y . , 83,84

J

Jackson, M.B., 105, 106, 111, 112, 119, 130, 135, 137, 143, 144, 145, 146, 147, 150, 160, 161, 162, 163, 164,170,172, 173,175, 176,179,180,183 Jackson, W.T., 160,177 Jacobs, W.P., 112,176 James, D.B., 131, I76 James, R., 179 Janssens, R., 84 Jarvis, M.C., 53, 67, 91 Jeblick, W., 91, 92 Jeffcoat, B., 105,171 Jeffrey, S.W., 203,225 Jenkins, J., 225 Jennings, A.C., 85 Jensen, W.A., 303,305,315,317 Jerie, P.H., 186 Jeschke, W.D., 186 Jesko, T., 121, 124, 176 Jewer, P.C., 109, 133,174 Jin, D., 101 Jin, D.F., 21, 22,23, 91 John, P., 229 Johnson, D.S., 257,312 Johnson, J.W., I69

Johnson, M.A., 96 Jolles, P., 81 Jones, H.G., 150,172,184 Jones, M.M., 155,168 Jones, O.P., 125,176 Jones, R.L., 100, 140,176,181 Jordan, W.R., 117,176 Jorgensen, R.A., 207,225 Juncosa, A.M., 303, 306,317 K Kado, C.I., 24,88 Kapil, R.N., 303,312 Karnovsky, M.L., 49, 79 Katerji, N., 184 Kato, K., 63, 91 Kato, Y . , 14, 91,92, 94 Katou, K., 100 Kauffman, S., 26, 91 Kaufman, P.B., 140,176 Kauss, H., 35, 36, 49, 73, 91, 92, 99, 101 Kavanagh, T.A., 225 Kawase, M., 145,176 Keegstra, K., 80,87 Keen, N.T., 23, 24, 68, 74,83, 90, 91, 101

Keenan, P., 24, 91 Keil, M., 19, 91 Kelley, C., 259,312 Kende, H., 124,176, 183 Kendra, D.F:, 35,36,44,55, 92 Kennell, J.C., 303,305,312 Keon, J.P.R., 90 Keppler, L.D., 52, 92 Kettemann, I . , 173 Key, J.L., 142, 174 Khatoon, K . , 297,312 Kiefer, L.L., 14, 92 Kieft, H., 305,315 Kijne, J.W., 85 Killias, U., 69, 78 Killingbeck, K.T., 115,176 Kindle, K., 224 Kinet, J.M., 121, I69 King, R., 109, 152, I79 King, R.W., 108,176 Kinraide, T.B., 285,312 Kiraly, Z . , 77 Kirk, T.K., 100 Klambt, D., 80 Knight, M.R., 53, 92 Knopp, J.A., 78

AUTHOR INDEX

Knox, J.P., 99 Kobata, A . , 5 , 92 Koch, Dr., 112,176 Kodde, E., 24,85 Kogel, G., 38, 61, 92 Kogel, K.H., 92, 95 Kohle, H., 36, 91, 92,101 Koizumi, K., 7, 92 Koller, D., 119, 130, I70 Kombrink, E., 99 Kondorosi, A . , 79 Konno, H., 72, 92 Kono, Y . , 166,174 Konze, J . R . , 96 Kooiman, P., 7, 12, 92 Kormelink, F.J.M., 98 Kowalewska, A.K.B., 160,163,176 Kowallik, K . , 211,225 Kowallik, K.V., 211,225 Koyama, T., 66, 92 Kozlowski, T.T., 145, 160,180, 184 Kramer, P.J., 160,177 Krassilov, V.A., 291,312 Kratka, J . , 8, 92 Krauss, A., 164,177 Kriesel, K., 139, 179 Krishna Rao, K., 190,225 Krizek, D.T., 115,177,181,182,185 Kuang, J.B., 159,177 Kubat, B., 169 Kubodera, T., 91 KuC, J., 6, 8,20,21, 26, 28, 39, 77,82, 90, 92, 93 Kddela, V., 8, 92 Kugrens, P., 192,214,225 Kuhn, R., 4, 15, 93 Kuhner, S., 306,314 Kuhshel, M . , 211,225 Kuiper, D., 104, 105, 116, 132,177 Kuiper, P.J.C., 105,177 Kulaeva, O.N., 119, 124, 129, 131,177 Kulajewa, O., 179 Kumamoto, J., 183 Kumar, D., 255,313 Kumar, S.S., 287,312 Kumpf, B .,228 Kurantz, M.J., 24, 40, 93 Kurosaki, F., 4, 22, 23, 35, 53, 54, 93 Kutacek, M., 182 L Labavitch, J . , 30, 62, 93 Labavitch, J.M., 30,82,100

327

Lachno, D.R., 156, 165,177 Lafitte, G., 80 Laine, R.A., 82 Lal, M., 231,251, 297,309,312 Lamb, C.J., 2, 8, 54, 55,82, 84,85, 86, 89, 90, 93, 98, 99 Lamport, D.T.A.,29, 90, 93 Lane, D.J., 223 Lang, A . , 124,182 Larkum, A.W.D., 190, 203,225 Larsen, J., 214,225 Larson, B . , 99 Larson, M . M . , 131,170 Lavender, D.P., 140,177 Lawrence, D.K., 118,177 Lawton, M.A., 54,85, 93, 98 Leach, J.E., 29, 93 Lee, J.J., 216,221 Lee, R.E., 192, 214, 225 Lee, S.-C., 3, 20, 21, 93 Leger, A . , 285,308 Lemaux, P.G., 223 Lemmon, B.E., 237, 251, 283, 287, 291,293,307,308 Lemoine, Y., 225 Lenton, J.R., 139, 141, 142, 156,169, 177 Leonard, J.-F., 182 Lerouge, P., 37,43, 93 Letham, D.S., 173, 180 Lewis, C.E., 255,312 Lewis, J . , 200,225 Li, N . , 210,225 Lian-Ju Mao, 303,307 Lichtle, C., 209,225 Lieberman, M . , 78 LiCnart, Y . , 46, 93 Ligrone, R., 233,234,235,237,241, 243,245,249, 251, 259,263, 271,273,276, 279, 281,283, 287,289,291, 295,297,299, 309,310,312,316 Lindberg, B . , 98 Lindberg, G., 96 Lindner, W.A., 50, 94 Linforth, R.S.T., 172 Ling, E., 179 Livne, A . , 133, 177 Lockard, R.G., 141, 177 Lockhart, D.J., 213,225 Lodge, T.A., 171 Loeffler, J.E., 124,177 Lohammer, T., I78

328

AUTHOR INDEX

Maclachlan, G . A . , 100 Maclean, D.J., 24, 94 McLeod, A., 158,178 McLeod, A.L., 159,179,180 McLeod, K.W., 118,184 McMichael, B.L., 104, 178 MacMillan, J., 139, 140,174,178 Macmillan, J.D., 69, 95 McNaught, H.L., 255,313 McNeil, M., 17, 78, 84, 95,96,98, 99, 100 Maerz, M., 224 Maglothin, A., 87 Maid, U., 212,226 Maier, K., 237, 312 Maier, U., 237,312 Maier, U.G., 222 Maillet, F., 93 Makus, D.J., 81 Mandoori, A., 223 Maness, N.O., 94 Manhart, J.R., 211,226 Mansfield, J.W., 8,77, 79 Mansfield, S.G., 303, 305, 306,313 Mansfield, T.A., 168 Mardanov, A.A., 132,172 Marechal-Drouard, L., 200,226 Marfa, V., 86, 95 Margulis, L., 190,226 Marinelli, F., 21, 25, 94 Marinos, N.G., 306,313 M MarkoviE, O., 18, 97 Ma, R., 67, 94 Markowicz, Y., 225 McBride, G.E., 291,311 Marsh, B.H., 281,313 McColl, R., 223 Marshner, H., 110, 132,182 McComb, A.J., 141,178 Martin, D.J., 88 McComb, J.A., 141,178 Martin, G.C., 160,183 McCormick, F.A., 253,313 Martinez-Molina, E., 88 McCracken, D.A., 208,226 Marx, G.A., 181 McCully, M.E., 106,170,178 Masago, H., 101 McCutchan, T.F., 224 Masia, A, 180 McDonald, A.J.S., 113,178 McDougall, G.J., 4, 5 , 6, 7, 14, 15, 45, Masle, J., 165, 166,178 Mason, W.K., 104,169 56, 75, 78,88, 94, 95 Masuda, Y . , 14, 15,62, 90,96 McDougall, G.M., 63, 95 Masuta, C., 24, 48, 52, 94 McFadden, B.A., 213,226 Matama, M., 101 McFadden, G.I., 196, 197, 199,215, Matsuda, K., 14, 91, 92, 94 226 Matsui, H., 100 McFarland, 19,20, 95 McFarland, K.D., 235,241,309,313, Matsushita, J., 14, 91, 94 Matthews, K.J., 88 315 Mattoo, A.K., 78 McGaw, B.A., 123,178 Mattox, K.R., 228 McKerracher, L., 195,227 Mauch, F., 8,32,35,73,74,81, 94, 98 Maclachlan, G., 43, 56, 87, 90

Lois, A.F., 101 Loiseaux-de Goer, S., 210, 211,220, 225 Long, M., 90 Longendorfer, D.H., 232,316 Longman, D., 43,94 Loomis, R.S., 133,183 Lorences, E.P., 15, 17, 38, 75,88, 94 Lorz, H., 26,89 Lo Schiavo, F., 87 Loschiavo, F., 84 Loschke, D.C., 35,55,89 Lotan. T.. 33. 94 Loveys, B.R., 110, 112, 151, 155, 157, 178 Low, P.S., 78, 90 Lowe, D.R., 190,229 Luckwill, L.C., 125,178 Ludlow, M.M., 159, 178 Ludwig, C.H., 28, 98 Ludwig, M., 195,209,215,226 Lugtenberg, B.J.J., 85 Lukacovic, A , , 176 Luke, H.H., 133,178 Lund, A.-B., 113, 116, 119,174 Lynn, D.H., 200,226 Lyon, G., 78 Lyon, G.D., 21,52,84, 94 Lyon, J.L., 168

329

AUTHOR INDEX

Mauch-Mani, B., 94 Mauk, C.S., 126, 127, 178 Maurel, C . , 80 Mayer, J.E., 85 Mayer, M.G., 46, 94 Mazau, D . , 8,29,87, 97 Medlow, G.C., 122, I73 Mees, G.C., 147, 160,178 Mehra, P.N., 255,313 Meidner, H., 133,179 Meins, F . , 99,100 Meinzer, F.C., 107, 109, 128,177 Melchers, L.S., 85 Melkonian, M., 214,228 Melton, L.D., 17, 95 Menary, R.C., 123,185 Menon, M.K., 297,313 Mereschkowsky, J . , 190,227 Mertens, R., 96 Messiaen, J., 53, 95 Metcalf, J., 171 Meyer, K., 255,313 Meyer, R.E., 155,179 Meyer, S.R., 192,227 Michael, G., 107, 132,185 Michaels, A.E., 229 Michielsen, P . , 89 Michniewicz, M., 139,179 Miginiac, E., 121, 122,179 Mignot, J.P., 192,227 Mikeswell, J., 305,313 Milborrow, B.V., 150, 156,168,179, 180 Miller, C.C.J., 271,287,293,295,309 Miller, C.O., 4, 95 Miller, E.C., 124,179 Miller, K.R., 209,229 Miller, L., 69, 95 Miller, M.H., 184 Milligan, S.R., 115, I79 Milliken, F.F., 285,314 Milon, H., 182 Mirecki, R.M., 182 Mirecki, R.N., 177 Mishler, B.D., 233,287,289,307,311, 313,316 Mitchell, W.A., 184 Moerschbacher, B.M., 34, 95 Moesta, P., I01 Moestrup, 8.,203,214,222,227 Mogensen, H.L., 303,313 Mohnen, D . , 27, 86, 95, 99 Moldau, H . , 160,179

Mollenhauer, D., 228 Molloy, J.A., 78 Moloshok, T . , 19, 95 Moloshok, T . D . , 87 Monge-Najera, J., 279,291,316 Monyo, J.H., 104,179 Morden, C.W., 190, 212,222,227 Mod, W., 176 Morgan, P.W., 145, 147,169,179 Morrall, S . , 193, 195,227 Morriset, F . , 228 Morschel, E . , 228 Mort, A.J., 94 Morvan, C., 97 Morvan, H., 97 Moss, G.I., 165, 179 Mothes, K., 117, 124,129, 179 Moyer, M., 86 Mueller, W.C., 285,313 Muldoon, E.P., 90 Mullet, J.E., 168 Mullins, M.G., 115, 130,169,179 Mullis, K.B., 228 Munns, R., 109, 152, 155,172, 179, 180 Mur, L.R., 229 Murashige, T., 91 Murfet, I.C., 174,181 Murgia, M., 285,301,308,313 Murphy, C.A., 222 Murphy, D . L . , 85 Murray, B.M., 235, 237, 241,314,316 Musgrave, A . , 145, I79 Mussell, H., 24, 68, 95 Mutaftschiev, S., 15, 27,100

N Nadakavukaren, M.J., 226 Nagl, W . , 306,314,315 Nakahara, Y., 5, 95, 98 Nakayama, N., 78 Nakosteen, L., I76 Narang, A . , 237, 297,312 Nasr, T., 120,185 Neales, T., 158,178 Neales, T.F., 118, 155,158, 159,179, 180 Nealey, L.T., 14, 96 Nechaev, O.A., 100 Nelson, C.E., 55, 96 Neuman, D.S., 107,109,136,137, 138, 143, 160, 161, 162, 163, 164,180 Neumann, D.S., 183

330

AUTHOR INDEX

Nevins, D.J., 33, 96 Newcomb, W., 303,305,306,314 Newman, LA., 285,312 Newman, S.C., 212,227 Nilsson, K.G.I., 5, 96 Nishi, A., 78, 93 Nishitani, K., 33, 62, 96 Noguchi, M., 63, 91 Noll, U., 95 Nonhebel, H.M., 110,180 Nooden, L.D., 112, 126, 127, 128, 139, 178,180 Noronha-Dutra, A.A., 87 Norris, R.E., 215, 217,224 Northcote, D.H., 4,5, 59, 67, 80,81, 90, 99 Nothnagel, E.A., 3, 18, 21,22, 23, 78, 96 Novacky, A., 52, 92 Novick, D., 86 Nozue, M., 88 Nunezbarrios, A,, 184 Nuri, W., 95 Nutt, H . , 228 0 Oakley, B.R., 192,213,227 Oates, J.E., 101 O’Brien, T.P., 303,316 Ocampo, J.A., 88 Ogawa, T., 5, 83, 95, 98 Ohjuma, K., 168 Okada, Y., 92 Okamoto, H., I00 O’Keeffe, L., 255,314 Oki, L., 91 Okon, Y . , 80 Oliver, M.J., 251,314 Olsen, G.J., 223,227 Olson, A.R., 303,314 O’Neill, M., 17, 96 Ong, H.T., 134,180 Ordin, L., 175 Osborne, D.J., 119,180 Ossowski, P., 5 , 9, 96, 98 Ovaa, J.C., 110, 125,184 Owen, H.A., 279,291,316 P Pace, N.R., 190,223,227,229 Pacini, E., 301,314 Pagan, F.M., 255,314 Pagel, W., 69, 96

Paleg, L.G., 172 Palme, K., 80 Palmer, J.D., 210, 211, 215,226,227 Palmer, M.V., 123,180 Palmer, R.G., 303,316 Palmer, S., 184 Panabibres, F., 97 Paradies, I . , 32, 96 Paranjothy, K., 184 Parker, C., 106,180 Parker, C.W., 173 Parker, L.L., 181 Parry, A.D., 149,180 Parthier, V.B., 117, 180 Passioura, J.B., 109, 155, 159, 165, 166,172,178,179, 180 Pate, J.S., 233,283, 285, 297, 301, 302, 303,311,314 Patrick, A.D., 7, 89 Patrone, L.M., 217,227 Patterson, M.E., 79 Paus, F., 99 Pauze, F.J., 80 Paxton, J., 96 Pearce, G., 81,87, 98 Peever, T.L., 49, 52, 96 Pegg, G.F., 73, 96 PClissier, B., 46, 47, 87, 96 Pefia-CortQ, H., 19, 96 Perasso, R., 201,227 Percival, E., 208,220,227 Pereira, J.S., 160,180 Peters, B.M., 96 Petersen, C.M., 303,310 Peterson, C.M., 303,305,310 Peterson, R.L., 297, 299,314 Peterson, T.A., 115,181 Petrovics, G., 79 Petschow, B., 124,170 Peumans, W.J., 3,82 Pharis, R.P., 176 Phillips, I.D.J., 140, 176, 181 Phinney, B.O., 84, 140,181 Phipps, J., 185 Piatti, T., 12, 31, 32, 96 Pienaar, R.N., 192,227 Pierce, M.L., 155,181 Pilet, P.-E., 81, 156,181, 182 Pillay, I . , 134, 181 Pilnik, W., 18, 67, 72,97,100 Pilotte, A., 96 Pilotti, A., 83, 98 Pitman, M.G., 114,181

AUTHOR INDEX

Popperl, H., 84 Poten, F., 92 Potts, W.C., 181 Powers, M.J., 79 Prat, S., 96 Pressey, R., 31, 57,58,72, 78,83, 90, 97 Pridham, J.B., 6, 79 Priem, B . , 15, 61, 97 Prins, M., 89 Pritchard, J., 182 Proctor, M.C.F., 232,234,314 Proebsting, W.M., 142,181 Prome, J.C., 93 Proskauer, J., 291,314 Provasoli, L., 222 Pughe, J., 184 Purse, J.G., 125,174, I81 Purwin, C., 89

Qu, L.H., 227

Q

Quisenberry, J.E., 104, 178

R

33 1

Rhia, S.J., 160, 164, 181 Rhiel, R.E., 209,228 Rhodes, R.G., 214,222 Ribaut, J.-M., 156,181 Ricci, A . , 80 Rice, E.L., 84 Richards, D., 104, 115, 116, 130,182 Richmond, A., 175,183 Richmond, A.E., 124,168,182 Rickauer, M., 67, 97 Ride, J.P., 8,28,34,35,73,78,80,81,97 Rivier, L., 156, 182 Robbins, M.P., 81 Roberts, K., 99 Roberts, K.R., 214, 218,222,228 Robertsen, B . , 28,29, 39, 97 Robertsen, B.K., I00 Robertson, D.S., 181 Robinson, D.R., 156,179 Robinson, H., 287,315 Robinson, S.P., 110, 178 Roby, D., 30,35, 77,87, 97 Rocha-Sosa, M., 96 Roche, P., 93 Rock, C.D., 186 Rogers, K.R., 52, 97 Rombouts, F.M., 18, 67,72,97,100 Ronchi, V.N., 94 Rood, S.B., 180 Roseboom, P.H.M., 20,85 Roth, D., 235,241,243,245,249, 287, 315 Rougier M., 285,301,313 Rowan, K.S., 208,224 Rowe, R.N., 115,116,130,173,182 Roy, M.A., 79 Rubinstein, M., 86 Ruff, M.S., 115,182 Rumeau, D., 29,87, 97 Rush, J.S., 8, 21, 39, 93 Rushing, A.E., 291,295,303,305,308, 310 Russell, R.S., 145, 166, 172,183 Russell, S.D., 303,307 Ryals, J., 86 Ryan, C.A., 2, 19,20,36,59, 60, 75, 81,87,89, 95, 96, 98,100,101 Ryback, G., I68 Ryder, T.B., 54,84, 98

Raa, J., 99 Radin, J.W., 107, 133, 165,173,181 Railton, I.D., 137, 181 Ramser, E.L., 169 Ranucci, A , , 84 Rashke, K., 155,181 Raven, P.H., 190,228 Ray, J.P., 134,174,181 Ray, P.M., 62, 93 Read, N.D., 95 Redgwell, R.J., 5, 97 Reece, C.F., 160, 164,181 Reese, J.C., 94 Reid, D.M., 125, 135, 137, 140, 142, 143, 150,170,171, I81 Reid, J.B., 141,173,174,181 Reid, J.S.G., 87 Reinsel, M.D., 181 Reisener, H.J., 92, 95 Reith, M., 208,209,210,211, 216,228 Renault, S., 232, 285,314 Renwick, K.F., 88 Renzaglia, K.S., 232, 237, 241,243, 251, 263, 265, 275,276, 279, 281, 287, 289, 291,293, 295, 310,312,314,315,316 S Reuss, J., 185 Saab, I.N., 157, 159, 160,182 Rexova-Benkova, L., 18, 97 Sabanek, J., 117,282

332

AUTHOR INDEX

Sachdeva, A., 303,307 Sagan, L., 190,228 Saiki, R.K., 200,228 Saka, H., 33, 91 Sakai, K., 5, 98 Saker, L.R., 172 Salama, A.M.S. El-D.A., 110, 131, 132,182 Salim, M., 133, 182 Salvi, G., 80,82,83,84, 87 Sanchez-Serrano, J., 91, 96 Sandanam, S., 137,143, I83 Sanderson, J., 156,182 Santore, U.J., 192, 193,223,227,228 Sargent, J.A., 94 Saris, L., 84 Sarjoni, G., 226 Sarkanen, K.V., 28,98 Sasa, T., 224,229 Satoh, S., 182 Sattelmacher, B., 110, 132, 182 Sattin, M., 171 Saunders, P.F., 177 Sawaguchi, T., 229 Saxena, A., 88 Saxton, M.J., 87 Schafer, E., 89 Scharf, S.J., 228 Scheer, U., 224 Schel, J.H.N., 305,310,315 Schell, J., 80, 91 Schertler, M.M., 253,315 Schlumbaum, A., 73, 98 Schmelzer, E., 99 Schmidt, W.E., 42,84, 98 Schnare, M.N., 201,228 Schneider, M.J., 180 Schnepf, E., 214,228, 306,315 Schofield, W.B., 245,275,289,315 Schols, H.A., 69, 98 Schopf, J.W., 190,228 Schottens-Toma, I.M.J., 24, 98 Schraudolf, H., 276,309 Schreiber, K., I83 Schuch, W., 82,86 Schuit, J., 177 Schultz, S.R., 305, 315 Schultze-Motel, W., 235,315 Schulz, D., 237, 245, 251,317 Schulz, W., 99 Schulze, E.-D., 173,182, 185 Schurr, U., 109, 151, 155,182,187 Schuster, R.M., 253,255, 257,263,

269, 271, 275, 289,291,293, 295,312,315 Schwabe, W.W., 183 Scofield, S.R., 183 Scott, J.L., 227 Searle-Van Leeuwen, M.F., 98 Seitz, H.U., 14, 15, 87 Sekhar, K.N.C., 305,315 Sela, I., 86 Selman, I.W., 137, 143,183 Selvendran, R.R., 5,86, 97 Sembdner, G., 140,171,183 Sepenswol, S., 192,228 Sequeira, L., 93 Setter, T.L., 164,183 Shah, C.B., 133,183 Shannon, L., 91 Sharp, J.K., 5, 6, 9, 78, 98 Sharp, R.E., 155, 157, 158, 159,182, 183 Shaw, J., 287,315,316 Shaybany, B., 160,183 Sheffield, E., 232, 297,316 Sher, N., 86 Sherwood, R.T., 100 Shih-Ying, H., 161, 183 Shindy, W.W., 157, 164, I83 Shinshi, H., 50, 99 Shiuaev, V.N., 100 Shivji, M.S., 210,228 Showalter, A.M., 29,54,77,99,253,316 Sidler, W., 209,228 Siegrist, J., 73, 99 Silverthorne, J., 181 Simmonds, J., 185 Singh, S., 180 Sisworow, E.J., 172 Sitte, P., 215,218,222,224,228,229 Sitton, D., 126, 140,183 Sivakumaran, S., 160,183 Skare, N., 69, 99 Skene, K.G.M., 108,140,156,170, 183 Skoog, F., 95, 117,176 Slovik, S., 106, I73 Sluiman, H.J., 233,316 Small, E.B., 200,226 Smart, C.M., 131, I83 Smart, M.G., 303,316 Smidsr~d,O., 59, 99 Smit, B., 160,179,183 Smit, B.A., 109, 136, 137, 160, 161, 162,163, 164,171,180,183

AUTHOR INDEX

Smith, A.J.E., 235,243,309,316 Smith, A.R., 124,185 Smith, D.K., 235, 241,309,313,315, 316 Smith, G.L., 245,253,255,316 Smith, H., 120, 121,184 Smith, K.A., 145,183 Smith, O.E., 168,183 Smith, P.G., 108, 112, 119, 164, 165, 183 Smith, R.C., 44, 64, 66,88, 99 Smith, S.M., 92 Snider, J.A., 283,316 Snyder, D., 223 Snyder, F.W., 177 So, H.B., 178 Sogin, M.L., 220,224 Somlyai, G., 77 Sommer, K.J., 178 Somssich, I.E., 54, 55, 99 Sotta, B., 121, 122,179 Southwick, A., 86 Spear-Bernstein, L., 209,229 Speirs, J., 225 Spellman, M.W., 17, 99 Spencer, D.F., 222 Spencer, M.S., 145,168 Spikman, G., 39,60,85 Sponsel, V.M., 139,184 Spray, C.R., 181 Sprinzl, M., 211,229 Staal, M., 116, 132, I77 Stacciarini, E., 171 Stacey, G., 39, 90 Stacey, N.J., 62, 99 Stachowiak, M., 160,183 Stachowiak, M.L., 183 Stanton, D.S., 232,316 Steer, M., 285,311 Steere, W.C., 235,316 Steeves, T.A., 303, 305,314 Stekoll, M., 20, 21, 99 Stelzig, D.A., 96 Stevenson, T.T., 17, 99 Stewart, K.D., 228 Stickel, S.K., 220 Stipanovic, R.D., 78 Stoddart, J.L., 184 Stoddart, R.W., 5,99 Stoecker, D.K., 217,229 Stoffel, S., 228 Stotler, R.E., 235,316 Strand, L.L., 24,68, 95

Strange, R.N., 87 Strangeways, E., 118,177 Street, H.E., 285,309 Strong, F.M., 95 Strout, G.W., 303,307 Stuchbury, T., I73 Stults, J.T., 171 Stutz, E., 17, 18,85 Stypa, M., 165,184 Suda, S., 229 Suire, C., 237,310 Suter, F., 228 Suthar, H.K., 303,313 Svalheim, O., 29, 97 Sweet, G.B., 177 T Takahashi, N., I71 Takaichi, S., 224 Takasaki, S., 92 Takeda, F., 72,77 Takeda, Y., 229 Tal, N., 86 Talmadge, K.W., 80 Tan, H.M., 182 Tang, Z.C., 145,184 Tardieu, F., 110, 151, 152, 153, 155, 166,184 Tashiro, N., 93 Taylor, C., III., 291, 311 Taylor, F.J.R., 190,213,227,229 Taylor, I.B., 149,172,184 Templeton, M.D., 54, 55, 99 Tepper, C.S., 24,40, 99 Termote, F., 58,100 Terry, M.E., 62, I00 Terzi, M., 84, 87 Thain, J.F., 47, 48, 100,101 Thibaud, J.B., 96 Thiessen, W.E., 168 Thomas, J.R., 17,100 Thomas, R.J., 232,316 Thompson, A.G., I85 Thompson, N.S., 96 Thompson, W.F., 210,225,227 Thomson, N., 291,309 Thorne, G.N., 174 Thornley, A.L., 190,224 Thornley, J.H.M., 113,184 Thorpe, S.R., 86 Tietz, A., 156,184 Tiller, P.R., 101 Tilton, V.R., 303,316

333

334

AUTHOR INDEX

Tollner, E.W., 169 Tomas, R.N., 215,229 Tomiyama, K., 12,48,86,88, I00 Tommerup, I.C., 94 Tong, C.B., 30,100 Topa, M.A., 118,184 Toppan, A., 87,97 Torgov, V.I., 5,100 Torrez-Ruiz, J., 226 Toubart, P., 95, 100 Townsend, R.R., 7,90 Traas, T.P., 89 Tran Thanh Van, K., 15,27,100 Treharne, K.J., 130, 138,184 Trejo, C.L., 110, 152,184 Trewavas, A.J., 92, 95, 105, 150, 184 Tromp, J., 110, 125,184 Truchet, G., 93 Trudel, J., 80 Truelsen, T.A., 63, 100 Tschaplinski, T.J., 114,115,184 Tsjui, J., 180 Tsukumi, H., 72, 92 Tsurusawa, Y . , 93 Turkova, N.S., 145,185 Turner, N.C., 177 Turner, S., 190, 211, 212,215,222, 223,229

U Usov, A.I., 100 Usui, T., 5,100 V Vaadia, Y., 124, 125, 133, 134, 175, 177,183, 186 Vaizey, J.R., 316 Valent, B., 98 Valent, B.S., 3, 8,14,69, 78,100 Valentin, K., 211,212,226,229 van Andel, O.M., 146,147, 160,185 Van Boom, J., 84 Vance, C.P., 100 Van Cutsem, P., 95 Van Den Bulcke, M., 94 Vanden Driessche, R., 115,185 Vanderhoef, L.N., 18,100 Van Der Plas, L.H.W., 89 Van der Zandt, H., 84 Van Engelen, F., 84 Van Halbeek, H., 101 Van Kammen, A., 84,85,89 Van Montague, M., 94

van Overbeek, J., 124,177 van Staden, J., 115, 123, 124, 125, 126, 131,170,171, 185 Van Toai, T.T., 161,183 Van Went, J.L., 303,317 Varner, J.E., 99 Vaughn, K.C., 279,291,316 Venere, R.J., 90 Verduyn, R., 84 Verma, D.P.S., 63, 100 Vesk, M., 209,222 Vessey, J.C., 73, 96 Vitt, D.H., 243,247,249,316 Vizarova, G., 121, 124,176 Voesenek, L.A.J.C., 145, 146,185 Voetburg, G.S., 182 Vogeli, U., 31,81,83, 98, 100 Vogels, R., 84 Vonlanken, C., 83 Von Saltza, M.H., 95 Voragen, A.G.J., 98

W Wade, M., 83 Wadman-van-Schravendijk, H., 146, 147, 160,185 Wagner, B., 172 Wagner, von H., 107, 132,185 Walker-Simmons, M., 20, 22, 35, 36, 60, 96,100,101 Wallace, T.P., 209,225 Walsh, M.M., 190,229 Walter, C.H.S., 170 Walter, M.R., 190,228 Walton, D.C., 156, I85 Wample, R.L., 150, 181 Wang, M.-C., 101 Wang, T.L., 125, 129, 131, 173, 185 Wang, T.-W., 147,185 Ward, E., 86 Wardlaw, C.W., 232, 297, 306,316 Wareing, P.F., 110, 115, 118, 120, 121, 124, 125, 128, 130, 131, 132, 142,168, 171, 172, 173,174, 181,182,184,185,186 Wartinger, A., 151,173, 185 Waseem, M., 125,185 Watanabe, M.M., 216,224,229 Waters, A.P., 224 Weatherley, P.E., 147, 160, 178 Webb, D.P., 171 Weber, J., 229 Wedemayer, G.J., 291,311

335

AUTHOR INDEX

wedermayer, G.J., 214,230 Weeden, N.F., 190,229 Wehrmeyer, W., 192,208,228,229 Weigel, H., 59, 101 Weil, J.-H., 226 Weiland, J., 183 Weiler, E., 150, 186 Weiss, C., 124, 186 Wellensiek, S.J., 121, 186 Went, F.W., 117, 124,186 Wessels, J.G.H., 34, 101 West, C.A., 3, 8,20,21,22,23,28,68, 82,91, 93, 99, I01 Weste, G.M., 170 Wetherbee, R., 192, 219,220,224 Whatley, F.R., 190, 215,229 Whatley, J.M., 190, 201, 203,215, 219, 229 Wheeler, A.W., 129,186 Whistler, R.L., 4, I01 Whitbread, F.C., 182 Whitford, P.N., 150,186 Whittier, D.P., 297,299,314 Whittington, W.J., 104, 179 Whyte, P., 125, 178 Wickham, K.A., I01 Wiencke, C., 237,245,251,317 Wijesundera, R.L.C., 68,101 Wilcox, L.E., 303,316 Wilcox, L.W., 214,230,291,311 Wildon, D.C., 6, 20,100, I01 Wilkins, H., 146, I86 Wilkins, M.B., 174, 180 Wilkins, S.M., 186 Willemse, M.T.M., 303,317 Williams, D.M., 203,230 Willis, C.L., 174 Willmitzer, L., 91, 96 Wilson, J.B., 113, 186 Wiltshire, G.H., 171 Winter, S., 228 Wise, R.R., 303, 306,317 Woese, C.R., 227 Wolf, O., 106, 164,186

Wollett, G., 224 Wolters, J., 215,222,230 Wong, O.C., 123,180 Wong, Y .-S., 90 Woolhouse, H.W., 107, 128, 168 Woolley, D.J., 130, 186 Wright, S.T.C., 150, 155, 160,173, 186 Wyndaele, R., 63,100 Wynne, M.J., 233,307

Y Yamaguchi, I., 171 Yamashita, K., 92 Yamazaki, N., 3, 25, 78,101 Yan, H., 303, 305,317 Yang, H.-Y., 303,305,317 Yang, S.F., 100, 107, 110, 147, 149, 168,169 Yeung, E.C., 303,317 York, W.S., 4, 14, 56, 75, 92, 101 Yoshikawa, M., 42, 74, 91,101 Young, D.H., 36,48,49, 92, I01 Young, H., 156, 172 Young, P.G., 228 Young, R.E., 72, 79 Young, S.F., I76 Z Zacharius, R.M., 24,40, 93 Zaerr, J.B., 177 Zeevaart, J.A.D., 141, 142, 155, 156, 161,171, 186 Zeidler, R., 229 Zeroni, M., 146,186 Zetsche, K., 211,212,226,229 Zhang, J., 109, 110, 138, 152, 154, 156, 157, 158, 159, 160, 162, 163, 180,184,186,187 Ziegler, E., 46, 94 Zimmerman, P.W., 146,187 Zobel, R.W., I87 Zodda, G., 273,317 Zuber, H., 228

This Page Intentionally Left Blank

SUBJECT INDEX

A Abiotic elicitors, 21 Abscisic acid (ABA), 105-7, 145, 149-66 chemical identification, 150 evidence against ABA as positive message from roots, 161-3 evidence for drying roots as source of apoplastic ABA, 155 evidence supporting ABA as positive message from roots, 163 in xylem root exudate, 158 mechanisms raising apoplastic ABA in leaves of droughted plants, 155 miscellaneous stress effects, 164-6 physiological significance of in-shoot apoplast and xylem sap, 151-5 physiological studies, 150 reconciliation of findings, 1 6 3 4 soil flooding, 160-4 water deficiency and leaf expansion, 159-60 water deficiency and stomata1 closure, 150-9 Acanthamoeba, 201 Acer, 123 Acer pseudoplatanus, 142 N-Acetylglucosamine, 9 Acrobolbus, 257 S-Adenosylmethionine (SAM), 144 Adiantum, 297 Alcaligenes eutrophus, 212,213 Alder, 114 Alisma, 306 Almond trees; 151 Alnus glutinosa, 114 Amino-oxyacetic acid (AOA), 149 1-Aminocyclopropane-1-carboxylic acid (ACC), 30, 144, 145, 147, 148, 149

Aminoethoxyvinylglycine (AVG), 31, 32, 144, 149 Anacystis nidulans, 206 Andraeobryum, 235 Andreaea, 235, 247, 263,285, 287, 289, 293 Andreaea rothii, 240 Andreaeales, 287, 293 Andreaeidae, 235 Andreaeobryum, 241 Aneura, 265,267,293 Aneura pinguis, 279 Angiosperms, 305 transfer cells in, 302 Anthoceros, 275, 279, 281, 295 Anthocerotaceae, 295 Anthocerotes, 275-83 placental cell walls in, 276 placental cells in, 276 Apricot trees, 151 Arabidopsis, 305, 306 Arabinosyltransferase, 12 Archidium, 235, 251, 283 Aspergillus, 21 Aspergillus niger, 25, 68, 69 Asterella, 255 Asteridae, 305 ATP, 144 Atrichum, 247 Atrichum undulutum, 246 Auxin, 15, 27 AXX, 33 B Barley, 105, 114 Bean cell, 54 Beta vulgaris, 114 Betula pendula, 142 Biotic elicitors, 21 Biotic stresses, 133-5 Blackcurrant, 121

337

338

SUBJECT INDEX

Blasia, 265, 267, 293 Blasia pusilla, 275 Blasticidin S, 24 Blindia, 251 Blindia acuta, 264 Brachythecium, 251 Brachythecium velutinum, 251 Brown algae, 210 Bryales, 249-53 Bryidae, 243-5 Bryophytes, 233, 234-83 placenta in, 283-95 Bryopteris, 257 Bryum, 245, 251, 257, 287 Bryum capillare, 263 Buxbaumia, 251 Buxbaumia piperi, 249 Buxbaumiales, 287 C Ca2+concentration, 36, 53-4 Callose, 36 Calobryales, 255, 269, 289, 293 Calobryum, 269 Calobryum blumei, 253,269, 285, 286 Calobryum indictum Udar et Chandra, 253-5 Campylomonas, 208 Capsella, 305, 306 Carbohydrates, 46 Carrot cells, 23 Carrot protoplasts, 15 Carrpos, 273, 291 Carrpos monocarpos, 291 Casbene synthase, 21,22 Castor bean cell wall, 22 Cavicularia, 265 Cell wall composition, oligosaccharininduced changes in, 28-9 Cellulysin, 32 Cephalozia, 263, 265 Cephalozia bicuspidata, 271 Chaetomium globosum, 35 Chalcone synthase, 31, 54 Chenopodiurn polyspermum, 121 Chitin, 73 oligosaccharides of, 34-5 Chitin-derived oligomers, 28 Chitinases, 12, 32, 73 Chitosan, 28, 73 Chitosan oligosaccharides, 35-7 Chitosan oligosaccharins, 44 Chlorarachnion, 214-16

Chlorarachnion reptans, 217 Chloroplast, 208-13 chromosome, 210-1 1 gene sequences, 211 genome, 210-13 membranes, 208 photosynthetic pigments, 208-10 rRNA, 212-13 second-hand, 189-230 storage product, 208 Chloroplast endoplasmic reticulum (CER), 196,200,203,208,218 Chromophyte algae, 201-3 Cichorium intybus, 121, 122 Citrus “polygalacturonic acid”, 17 Cladophascum, 251,283 Cladophascum gymnomitrioides, 264 Cladosporium cucumerinum, 28, 29,39 Cladosporium fulvum, 24, 39,49, 68,

74 Cocos nucifera, 124 Coleochaete, 29 1 Colletotrichum lagenarium, 29, 46, 73 Colletotrichum lindemuthianum, 24, 29, 32,40, 52, 54,55,61, 68, 70 Commelina communis, 152, 157 Conocephalum, 255, 271, 273 Corsinia, 255, 271 Coscindiscus, 211 Cowpea pods, 25 Crithidia, 2 15 Cryptomonad algae evolution of, 189-230 see also Chloroplast; Nucleomorph Cryptomonads ancestors of chromophyte algae, 201-3 as endosymbionts, 213-14 cell layout, 192-3 overview of, 192 parasites of, 213-14 rRNA genes, 20&1 taxonomy, 219-20 Cryptomonas, 208, 214 Cryptothallus, 265,267, 283,285, 293 Cryptothallus mirabilis, 279, 28 1 Cucumber cell walls, 28, 29 Cucurbitapepo, 132 Cyanidium caldarium, 212 Cycloheximide, 24 Cytokinins, 123-38 development in unstressed plants, 125-8

SUBJECT INDEX

early research, 123-5 miscellaneous stresses applied to roots, 133-8 responses to mineral nutrient shortage, 131-3 root excision studies, 128-31 D 2,4-D, 14, 15, 63 Dendroceros, 275,279, 281, 295 Dendroceros tubercularis, 298 Dendrocerotaceae, 295 Dicranum, 249, 251 Dicranum majus, 263, 287 Dictyota dichotoma, 211 Diphyscium, 249,251,253, 283, 287 Diphyscium foliosum, 264 Diphyscum, 257 Diploid nucleomorph, 207 Diplophyllum, 263, 265 Diplophyllum albicuns, 273 Dodeca-a-( 1+4)D-galacturonide, 22 Douglas fir, 135 “Driselase”, 26 Drought, 133-5 Dumortiera, 271 E Elicitors abiotic, 21 biotic, 21 formation of pectic oligosaccharides as, 68-72 of phytoalexin synthesis, 2C-3 transport of, 76-7 Embryo, formation of, 233 Embryonic phase, 232 Embryophytes, 233 Encalypta, 245 Endosymbionts, cryptomonads as, 21314 Enzymes acting on chitin and chitosan, 73 direct effects of oligosaccharides, 568 Ephemerum, 235 Equisetum, 297 Erwinia carotovoru, 22, 68, 69 Erwinia chrysanthemi, 49 Erwinia rubrifaciens, 24 Escherichia coli, 206 Eschscholtziu, 31 Ethylene, 144-9

339

effect of flooding, 145-9 synthesis, 12 induction by pectic oligosaccharides, 30-2 “Ethylene-forming enzyme” (EFE), 144 Ethylene-inducing xylanase (EIX), 32, 33 Eucalyptus marginata, 135 Eukaryotic ribosomes around nucleomorph, 196-200

F FAXX, 33 Ferns, 297, 299 Ferulate, 33 Fissidens, 251 Fissidens crassipes, 25 1 Flax hypocotyls, 15 Flooding, 145-9, 1 6 M Flowering, 12@1 Folioceros, 275, 279, 281, 295 Folioceros fuciformis, 305 Fossombronia, 265, 267, 293 Fossombronia echinata, 273,283 Frullania, 257 L-Fucose-containing oligosaccharide, 15 2’-Fucosyl-lactose, 15 Funaria, 245, 251 Fungal cell walls, 12 components of, 8 Fungal infection, 20 Fungal oligo-p-glucans, 7-12 Fusarium, 21 Fusurium oxysporum, 61 Fusurium solani, 36 G GalAI2, 44 D-Galacturonic acid, 21 a-(1+4)-D-galacturonic acid, 22 Galacturonic acid residues, 20 a-D -galacturonidase, 31 Gametophyte-sporophyte junction in land plants, 231-317 Gas chromatographylmass spectometry (GC-MS), 123, 125 Gel-permeation chromatography, 19, 25, 31 Gibberellic acid, 130

340

SUBJECT INDEX

Gibberellins, 138-44 effects of root excision and environmental stresses applied to roots, 142-3 studies of unstressed plants, 139-42 a-(1+3),(1+4)-D-glucan, 20 p-(1+3),(1-+6)-glucan, 8-9, 12, 20,73 p-( 1+3)-linked D-glucan, 42 p-glucanase , 32 p-( 1+3)-D-glucanase, 22 P-D-glucanase, 7p-D-glucopyranose, 9 Glucosamine, 9 p-( 1+3)-linked D-glucose residues, 36 Glutathione, 50 Glutathione peroxidase, 50 Glyceollin, 50 Glycine, 305 Glycine rnax, 126, 142 Glycoprotein-derived oligosaccharins, 61-2 Glycoproteins, 20, 38 synthesis, 12 Goebelobryurn, 257 Goniornonas, 192 Gorse, 43 Gossypiurn hirsuturn, 107, 133 Grape vine, 115 Growth regulators, 12-17

H Haplomitrium, 269, 289 Haplornitriurn gibbsiae Steph., 253 Haplornitrium hookeri, 285 Haustorium, 253 Hedera helix, 142 Helianthus, 305 Helianthus annuus, 131, 132, 135, 140, 155, 158, 160 Heptasaccharide, 9 Herberta, 263, 265,269,271, 291 Hexadecadienoic acid, 37 Hexasaccharide, 9 Hordeurn vulgare, 114, 137 Hordeurn vufgare L. cv. Midas, 105 Hormones, 74-7 assessing developmental impact of messages, 111-12 criteria for implicating regulation of naturally occurring developmental phenomenon, 111

evidence for regulation of root:shoot ratio by roots, 112-16 hormone-like action of roots on shoots, 117-23 in root to shoot communication, 103-87 message concept, 106-12 quantifying messages in transpiration stream, 107-11 Hornworts, 275 HPLC, 125 HRGP, 29-31 Hydroxycinnamates, 29 6-Hydroxymellein, 21 Hydroxyproline-rich glycoproteins. See HRGP I Impatiens glandulifera, 140 Indoleacetic acid (IAA), 27-8, 50-1, 57-8,72, 130 In shoot apoplast, 151-5

J

Jackiella, 257 Jubula, 257 Jubulaceae, 291 Jungermaniidae, 289 Jungermanniales, 253,257-65,265, 267,273,285,289,293,299 K Kornrna caudata, 194, 197, 199 Kurzia, 263, 265 Kurzia trichoclados, 265, 271 L Land plants gametophyte-sporophyte junction in, 231-317 life-cycle of, 232 Leaf expansion and water deficiency, 159 Leaf senescence, 117-19, 126 Lectins, 43 Lejeunaceae, 291 Lejeunea, 257 Lepidodiniurn viridae, 215, 216 Lignification, 20, 28, 34 Lignin, 29 Lipid peroxidation, 52 Lipoxygenase, 52

34 1

SUBJECT INDEX

Liverworts, 233,253-75 anacrogynous, 265 placenta in, 258 placental cells in, 260 Lolium perenne, 131 Lophocolea, 263,265 Lophocolea heterophylla, 269 Lunularia, 271 Lupinus albus, 140, 157 Lycopersicon esculentum, 123, 137, 139, 161 Lycopodium, 297 Lycopodium appressum, 299 Lycopodium cernuum L., 299 Lysozyme, 12

M Macerase, 30 Magnaporthe grisea, 25, 26,33 Maize, 26, 114, 153, 154 Mannia, 271 Marchantia, 255, 271 Marchantiales, 255,271,273, 289,291 Marchantiidae, 271-5, 285, 291, 293, 305 Marchesta, 255 Marsupella, 263,265 Marsupella funckii, 267 Megaceros, 275,279,281, 295 Melons, HRGP biosynthesis in, 30 Membrane depolarization, 46 Messenger RNA (mRNA), 21, 29, 41, 54,55 6-Methoxymellein, 21 4-0-Methyl ether, 32 5’-Methylthioadenosine, 144 Metzgeriales, 257,265-7,267, 273, 289,291,293 Mineral nutrient shortage, 131-3 Mniurn, 251 Mnium hornurn, 251,263 Monoclea, 255, 289 Monoclea forsteri Hook, 257 Monocleales, 253, 271, 291 Mosses, 233,235-53 acrocarpous , 245 arthrodontous, 243 nematodontous, 243 placenta in, 236 placental cells in, 238 Myrionecta rubra, 213

N

Nicotiana , 119 Nicotiana rustica, 117, 124, 129, 131 Nicotiana tabacum, 50 Notothyladaceae, 295 Notothylas, 275,279, 281, 295 Notothylas orbicularis, 305 Nuclear DNA, separation of, 203-6 Nucleomorph, 192-207 derived from red algal nucleus, 201 DNA content, 195-6 electrophoretic karyotype of, 207 eukaryotic ribosomes around, 196200 isolation of, 203-7 nucleus-like characteristics, 192-5 origin of, 200-3 role of, 216-18 structure, 192 Nucleomorph DNA, 216 separation of, 203-6 Nutrient control theory, 112-16 shortcomings of, 113-16 0 Oak, 131 Ochromonas dancia, 211 Odontella, 211 Oligogalacturonides, 19, 22, 23, 27, 28, 44, 50, 51, 70 Oligo-P-glucans, 7-12, 22,42, 43 receptors for, 41-3 Oligo-P-(1+3),(1+6)-glucans, 58 Oligo-P-glucosides, 9, 10 Oligosaccharides, 2 direct effects on enzymes, 56-8 evidence for receptors, 41-6 fucose-free, 14 of chitin, 34-5 of chitosan, 35-7 of pectin, 17-32 purification and chemical characterization, 6-7 sequencing, 7 structure-activity relationships, 9 xyloglucan-derived, 12-17 Oligosaccharin-induced changes in cell wall composition, 28-9 Oligosaccharins, 1-101 artificial, 3 bioassays, 5-6 diversity of, 38 from N-linked glycoproteins, 37-8

342

SUBJECT INDEX

Oligosaccharins (cont.) Pectinase-inhibiting proteins (PGIPs), glycoprotein-derived, 61-2 61, 70-1 mechanism of formation and Pectinases, 17, 20, 25, 33 degradation, 62-74 Pectinmethylesterase, 26 membrane depolarization, 46 “Pectolyase”, 23 mode of action, 41-58 Pellia, 265, 267, 283, 293 movement within plant, 74-7 Pellia epiphylla, 273, 279 natural occurrence, 58-62 Pentasaccharide, 36 oligo-a-xylans as, 32-4 Perilla frutescens , 128 origin of concept, 2-3 Perilla ocimoides, 121 oxidative metabolism, 49-52 Petunia, 12 physiology of effects, 7 4 1 Phaeoceros, 275, 279, 281,283, 295 preparation, 3-5 Phaeoceros carolinianus, 279 method l:, 3 Phaeoceros laevis, 298, 305 method 2:, 3-4 Phascum, 283 method 3:, 4 Phaseolus, 61, 306 method 4:, 4-5 Phaseolus coccineus, 142 protein phosphorylation, 52-3 Phaseolus vulgaris, 54, 70, 117, 123, rapid effects of, 4 6 5 6 130, 134, 136, 137, 139, 146, receptors for, 43-6 147,161, 162, 164, 165 second messengers, 5 3 4 Phenylalanine ammonia lyase (PAL), successful host or successful 23, 31,32, 48, 54, 70 pathogen?, 39 Phloem, 106 synergism between, 4&1 Phoenix dactylifera L, 305 xylan-derived, 34 Photosynthesis, 120-1 Oligotrichurn, 247 Photosynthetic pigments, 208-10 Oligotrichurn hercynicum, 249 Phycobiliprotein, 209 Oligo-P-xylans as oligosaccharins, 32-4 P-Phycoerythrin, 209-10 Olisthodiscus luteus, 211 Physcomitriurn, 251, 297 Phytoalexins, 8-10, 2&3,26, 36, 40, 42,48,50, 53 P Phytophthora, 35,39 Pallavicinia, 265, 267, 293 Phytophthora cinnamomi, 135 Pallavicinia indica, 275, 283 Phytophthora infestans, 24, 40, 48, 51 Pea stem segments, 16, 27 Phytophthora megasperma, 40,42,50, Peach, 130 52,74 Pectic oligosaccharides, 17-32 Phytophthora megasperma f.sp. as elicitors of phytoalexin synthesis, glycinea, 46, 54 2&3 Phytophthora parasitica, 32 as growth regulators, 18 Phytophthoraparasitica var. nicotianae, 46 degradation in plant tissue, 72 Pinus, 306 formation as elicitors, 68-72 Pinus serotina, 118 formation as wound signals, 6 6 8 Pinus taeda, 118 formation in ripening fruit, 72 Pisum, 306 induction of ethylene synthesis, 30-2 Pisurn sativum, 138, 140, 161, 163, 164, morphogenesis-regulating activity, 210 27-8 Placenta, 233 natural occurrence, 59-61 in bryophytes, 283-95 Pectic oligosaccharins in liverworts, 258 hypersensitive response by, 23-6 in mosses, 236 transcription of protease inhibitor Placental cell walls in anthocerotes, 276 genes, 48 Placental cells Pectin lyase, 26 in anthocerotes, 276

SUBJECT INDEX

in liverworts, 260 in mosses, 238 Plagiochasrna, 271 Plagioselmis palustris, 195 Plant hormones. See Hormones Plantaginales, 305 Plantago, 116 Plantago major, 132 Pogonaturn ,247 Pogonatum neesii, 246 Polygalacturonic acid, 28 Polypodium, 297 Polytrichales, 245-7, 263, 285, 287, 293 Polytrichum, 245,247, 285 Polytrichum formosum, 249 Populus, 143, 162 Populus deltoides, 136 Populus trichocarpa, 136 Porellaceae, 291 Porphyridiurn aerugineum, 212 Potato protoplasts, 12 Potato tuber disks, 51 Preissia , 255, 271 Prolyl hydroxylase, 12 Pronase E, 23 “Protease inhibitor inducing factor” (PIIF), 19,48,75 Protease inhibitors, 19, 20 Protein gene phylogeny, 212-13 Protein phosphorylation, 52-3 Proterornonas steinii, 214 Prunus armenica, 151 Prunus dulcis, 151 Prunus persica, 130 Pseudomonas syringae, 49,69 Pseudotsuga menziesii, 135 Pteridium, 297 Pteridophytes, 233, 295-301 Puccinia, 61 Puccinia graminis, 38, 45 Pylaiella littoralis, 210 Pyrenoid-nucleomorph complexes, 206

Q

Quercus rubra, 131

R Radula, 257, 263, 265, 291 Radula complanata, 273 Radulaceae, 291 Reboulia, 255, 271,273 Reboulia hemisphaerica, 291, 293

343

Reboulia hemisphaerica var. macrocarpa Zodda, 273 Red algae, 201,202,217,219,233 Rhamnogalacturonans I and I1 (RG-I and RG-11), 17, 20, 26 L-Rhamnose, 21 Rhizobium, 43 Rhizobium meliloti, 37 Rhizophora, 306 Rhizopus stolonifer, 68 Rhodella , 217 Rhodomonas salina, 204-7 Ribes nigrum, 121 Ribonuclease A, 21 Ribosomal RNA (rRNA), 196, 200, 212-13,216,219 Ribulose-1 ,5-bisphosphate carboxylaset oxygenase, 211 Riccardia, 265,267,285, 293, 295 Riccardia multijida, 275,279,281, 283 Riccia, 255, 271, 273, 291 Riccia sorocarpa, 293 Rice blast pathogen, 25 RNA, 54, 196 Root excision studies, 128-31 Root:shoot ratio of plants, 112-16 Rubisco phylogeny, 211-12 Rubus, 15 Rudbeckia tricolour, 121 Rumex palustris, 146 S Saccharum spp. hybrid, 128 Salinity, 133-5 Salix viminalis, 120 Scapania, 263,265 Scapania gracilis, 271 Scrofularia arguta, 121 Scrophulariales, 305 Second-hand chloroplasts, 189-230 Second messengers, 53-4 Seed plants, 301-6 Selaginella, 301 Shoot extension, 120-1 Silene, 61, 121 Soil flooding, 145-9, 160-4 Soil waterlogging, 135-8 Solanum andigena, 130, 131 Solanum pennelli, 134 Solanum tuberosum, 132 Sorghum saccharum, 121 Sorghum vulgare, 135 Soybean, 21,36,44, 126, 128

344

SUBJECT INDEX

Sphaerocarpales, 255,271, 289, 291 Sphaerocarpos, 271 Sphagnidae, 241-3 Sphagnum, 235,241, 263,283,285, 287,295,299 Sphagnum cuspidatum, 241, 243,245 Sphagnum fallax, 241 Sphagnum fimbriatum, 241,243 Sphagnum subnitens, 241 Sporophyte. See Gametophytesporophyte junction Spring sap in woody plants, 125-6 Stellaria, 306 Stomata1 closure and water deficiency, 15&9 Striga hermonthica, 106, 135 Sugar beet, 114 Sugar cane, 128 Sycamore cells, 25 T Takakia, 235, 241,263,269, 285, 289 Takakia ceratophylla, 243 Takakiales, 293 Targionia, 255, 271 Targionia hypophylla, 271 Tetraphidales, 247-9 Tetraphis, 287 Tetraphis pellucida, 247, 249, 251 Thalictrum rugosum, 12 Timmiella, 25 1 Tmesipteris, 301 Tobacco leaf, 27 Tomato, 19, 36, 109 Transfer cells in angiosperms, 302 Trichoderma viride, 33 Trichomonas, 218 Triticum aestivum, 142, 152, 155 Tropaeolum , 306 U

Ulex europaeus, 43 Urtica dioica, 132 V Vaucheria, 211 Verticillium dahliae, 44, 50 Viciafaba, 210

Vigna, 25 Vitis vinifera, 115, 130, 140, 151 W Water deficiency and leaf expansion, 159-60 and stomata1 closure, 150-9 Willow, 120 Wound hormones, 18-20 Wound signals, 18-20, 6 6 8 non-transport of, 75-7

X Xanthium stromarium, 128, 156 Xanthomonas malevacearum, 24 XET, 65,66 XGS, 15 XG7, 16 XG8, 14, 16 XG9, 14-16, 33, 44, 45, 63, 64, 66, 73, 75 XG9n, 16 Xylan-derived oligosaccharins, 34 Xylanase, 26, 32, 33 p-(1--+4)-D-xylanases,32 Xylem, 107, 119 Xylem sap, 106-12, 123-6, 128, 133, 135, 148, 151-5, 158, 162 Xyloglucan, 41 sugar residues in, 13-14 Xyloglucan-derived oligosaccharides, 12-17 Xyloglucan endotransglycosylase, 56-7, 64 Xyloglucan oligosaccharides degradation of, 66 natural occurrence, 58-9 synthesis of, 62-6 transport of, 75 D-xylose, 21 Z Zeamays, 114, 121, 124, 153, 154, 159, 160 Zeatin, 131 Zeatin riboside, 126, 127, 131, 136 Zoopsis, 263,265,291