The Plant Viruses: Polyhedral Virions and Bipartite RNA Genomes

The Plant Viruses Volume 5 POLYHEDRAL VIRIONS AND BIPARTITE RNA GENOMES

THE VIRUSES Series Editors HEINZ FRAENKEL-CONRA T, University of California

Berkeley, California

ROBERT R. WAGNER, University of Virginia School of Medicine Charlottesville, Virginia

THE VIRUSES: Catalogue, Characterization, and Classification Heinz Fraenkel-Conrat

Other volumes in the series:

THE BACTERIOPHAGES Volumes 1 and 2 • Edited by Richard Calendar

THE BUNYAVIRIDAE Edited by Richard M. Elliott

THE CORONA VIRIDAE Edited by Stuart G. Siddell

THE INFLUENZA VIRUSES Edited by Robert M. Krug

THE P APOV A VIRIDAE Volume 1 • Edited by Norman P. Salzman Volume 2 • Edited by Norman P. Salzman and Peter M. Howley

THE PARAMYXOVIRUSES Edited by David W. Kingsbury

THE PARVOVIRUSES Edited by Kenneth 1. Berns

THE PLANT VIRUSES Volume 1 • Edited by R. 1. B. Francki Volume 2 • Edited by M. H. V. Van Regenmortel and Heinz Fraenkel-Conrat Volume 3 • Edited by Renate Koenig Volume 4 • Edited by R. G. Milne Volume 5 • Edited by B. D. Harrison and A. F. Murant

THE REOVIRIDAE Edited by Wolfgang K. Joklik

THE RETROVIRIDAE Volumes 1-4 • Edited by Jay A. Levy

THE RHABDOVIRUSES Edited by Robert R. Wagner

THE TOGA VIRIDAE AND FLA VIVIRIDAE Edited by Sondra Schlesinger and Milton J. Schlesinger

THE VIROIDS Edited by T. O. Diener

A complete listing of volumes in this series appears at the back of this volume.

The Plant Viruses Volume 5 POLYHEDRAL VIRIONS AND BIPARTITE RNA GENOMES

Edited by

B. D. HARRISON University of Dundee Dundee, United Kingdom

and

A. F. MURANT Scottish Crop Research Institute Dundee, United Kingdom

Springer Science+Business Media, LLC

L i b r a r y of Congress C a t a l o g i n g - i n - P u b l i c a t i o n Data

Polyhedra l v i r i o n s and b i p a r t i t e RNA genomes / e d i t e d by B.D. Harr ison and A . F . Murant

p. era. — (The p lan t v i ruses ; v . 5) (The v i r u s e s ) Inc ludes b i b l i o g r a p h i c a l re fe rences and index. ISBN 978-1-4899-1774-4 1 . RNA v i r u s e s . 2 . P lant v i r u s e s . I . H a r r i s o n , B. D. (Bryan D.)

I I . Murant , A. F. I I I . S e r i e s . I V . S e r i e s : The v i r u s e s . QR395.P65 1996 5 7 6 ' . 6 4 8 3 — d c 2 0 96-13480

CIP

ISBN 978-1-4899-1774-4 ISBN 978-1-4899-1772-0 (eBook) DOI 10.1007/978-1-4899-1772-0

© Springer Science+Business Media New York 1996 Originally published by Plenum Press, New York in 1996 Softcover reprint of the hardcover 1st edition 1996

10 9 8 7 6 5 4 3 2 1

A l l rights reserved

No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfi lming, recording, or otherwise, without written permission from the Publisher

Contributors

G. Adam, Universitat Hamburg, Institut fur Angewandte Botanik, D-20309, Hamburg, Germany

G. Boccardo, Istituto di Fitovirologia Applicata del Consiglio Nazionale delle Ricerche, 10135 Torino, Italy

D. J. F. Brown, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, United Kingdom

G. A. de Zoeten, Department of Botany and Plant Pathology, Michigan State University, East Lansing, Michigan 48824

S. A. Demler, Department of Botany and Plant Pathology, Michigan State University, East Lansing, Michigan 48824

J. P. Fulton, Department of Plant Pathology, University of Arkansas, Fayetteville, Arkansas 72701

R. C. Gergerich, Department of Plant Pathology, University of Arkansas, Fayetteville, Arkansas 72701

R. W. Goldbach, Department of Virology, Agricultural University, 6709 PD Wageningen, The Netherlands

D. J. Hagedorn, Department of Plant Pathology, University of Wisconsin, Madison, Wisconsin 53706

R. I. Hamilton, Pacific Agriculture Research Centre, Agriculture and AgriFood Canada, Vancouver, British Columbia, Canada V6T lX2

K. F. Harris, Department of Entomology, Texas A&M University, College Station, Texas 77843

B. D. Harrison, Department of Biological Sciences, University of Dundee, Dundee DDI 4HN, United Kingdom

A. T. Jones, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, United Kingdom

V. Lisa, Istituto di Fitovirologia Applicata del Consiglio Nazionale delle Ricerche, 10135 Torino, Italy

G. P. Martelli, Dipartimento di Protezione delle Piante, Universita degli Studi di Bari, 7126 Bari, Italy

v

vi CONTRIBUTORS

M. A. Mayo, Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom

A. F. Murant, Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom

W. M. Robertson, Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom

D. J. Robinson, Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom

H. A. Scott, Department of Plant Pathology, University of Arkansas, Fayetteville, Arkansas 72701

R. Stace-Smith, Pacific Agriculture Research Centre, Agriculture and AgriFood Canada, Vancouver, British Columbia, Canada V6T lX2

J. H. Tremaine, Pacific Agriculture Research Centre, Agriculture and AgriFood Canada, Vancouver, British Columbia, Canada V6T lX2

D. 1. Trudgill, Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom

R. A. Valverde, Department of Plant Pathology and Crop Physiology, Louisiana State University, Baton Rouge, Louisiana 70803

J. Wellink, Department of Molecular Biology, Agricultural University, 6709 PD Wageningen, The Netherlands

Preface

This fifth volume in the series The Plant Viruses, dealing with viruses with bipartite genomes, completes the coverage of viruses with isometric particles and genomes consisting of single-stranded, positive-sense RNA: viruses that have tripartite and monopartite genomes of this kind were dealt with in Volumes 1 and 3, respectively. How close are the affinities among the viruses within the groupings distinguished in this way? All those with tripartite genomes are considered to be sufficiently closely related to be included in the family Bromoviridae, whereas the monopartite-genome viruses covered in Volume 3 clearly are a much more diverse collection. Affinities among the viruses with bipartite genomes are considered in Chapter 1 of this volume, along with the possible origins, advantages, and disadvantages of these genomes. The conclusion reached from this assessment is that the bipartitegenome viruses fall into four categories, those within each category having closer affinities with viruses not included in this book than with viruses in the other categories. No evidence was found that possession of a bipartite genome gives a virus overwhelming advantages over viruses of other sorts. More probably, any advantages are largely balanced by disadvantages, and bipartite genomes may be best considered simply as an alternative design for the hereditary material of a virus.

Taking the view that no great similarities exist between bipartitegenome viruses in general, the viruses are dealt with genus by genus, with several chapters being used to describe different aspects of those genera that have been studied in the most detail. To give a rounded account, the membership of each genus together with the molecular and biological properties, ecology, and control of typical members are described and discussed.

This volume has some advantages over early volumes in the series because the pace of advance of virology in the intervening years has been rapid and much additional knowledge has been gained. For example, the complete nucleotide sequence is known for at least one member of five of the six genera considered in this volume. Analysis of these sequences has en-

vii

viii PREFACE

abled a more soundly based taxonomy to be adopted. Indeed, throughout the volume we have used the names and classifications found in the Sixth Report of the International Committee on Taxonomy of Viruses (1995), in which generic names have the ending -virus (e.g., Nepovirus) and family names have the ending -idae (e.g., Comoviridae).

Modern approaches have done much more than simply to clarify relationships among viruses. The types, arrangement, and expression strategy of viral genes are now well understood for some of the viruses described here, and functions can be assigned to many of the gene products. Much has been learned about the ways different viruses replicate and pass from cell to cell, the multiple roles of individual virus-coded proteins are becoming better recognized, and infectious transcripts are now available for several of the viruses, so allowing reverse genetics to be used to study and define these roles. In addition, the creation of new types of virus resistance by transforming plants with virus-related nucleotide sequences has become a popular topic for research. The reader will find examples of this rich variety of modern work throughout the volume. Finally we would like to acknowledge the willing cooperation of the authors of individual chapters and their efforts to make the volume interesting and authoritative.

Dundee

B. D. Harrison A. F. Murant

Contents

Chapter 1

Plant Viruses with Bipartite RNA Genomes and Polyhedral Particles: Diversity and Affinities

B. D. Harrison and A. F. Murant

I. Introduction .............................................. 1 II. Recognition of the Existence of Bipartite RNA Genomes ..... 2

III. Diversity and Affinities of Plant Viruses with Isometric Particles and Bipartite RNA Genomes . . . . . . . . . . . . . . . . . . . . . . . 3

IV. Possible Origins of Bipartite RNA Genomes ................. 5 V. Advantages and Disadvantages of Bipartite RNA Genomes ... 8

VI. Dependence, Satellitism, and Multipartite Genomes ......... 11 References ..................................................... 13

Chapter 2

Comoviruses: Identification and Diseases Caused

R. A. Valverde and J. P. Fulton

I. Identification of Comoviruses .............................. 17 A. Members of the Genus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 B. Criteria for Grouping or Distinguishing Comovirus Species 18

II. Comoviruses and the Diseases They Cause .................. 23 A. Andean Potato Mottle Virus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 B. Bean Pod Mottle Virus .................................. 24 C. Bean Rugose Mosaic Virus .............................. 24 D. Broad Bean Stain Virus........ ........ ....... . . ....... .. 24 E. Broad Bean True Mosaic Virus.... ......... ....... ....... 25 F. Cowpea Mosaic Virus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 G. Cowpea Severe Mosaic Virus ............................ 26

ix

x CONTENTS

H. Glycine Mosaic Virus ................................... 27 I. Pea Mild Mosaic Virus .................................. 27 J. Quail Pea Mosaic Virus ................................. 27 K. Radish Mosaic Virus .................................... 28 1. Red Clover Mottle Virus ................................ 28 M. Squash Mosaic Virus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 N. Ullucus Virus C ........................................ 29

References ..................................................... 29

Chapter 3

Comoviruses: Molecular Biology and Replication

R. W. Goldbach and J. Wellink

I. Introduction .............................................. 35 II. Composition of Virus Components ......................... 36

A. General Description .................................... 36 B. Separation of Particle Components ...................... 38 C. Particle Proteins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 D. Virion Structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 E. Assembly .............................................. 42

III. Genome Structure and Organization ........................ 42 A. Terminal Structures .................................... 42 B. The Nucleotide Sequence of Comoviral RNA Species ..... 43

IV. Translation ............................................... 44 A. Initiation of Translation ................................ 44 B. Mechanism of Translation of RNA-2 ... . . . . .... . . . . . . . . . . 46

V. Processing of the CPMV Polyproteins ....................... 47 A. Processing in Vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 B. Processing in Vitro ..................................... 47 C. The Cleavage Sites ..................................... 50 D. The Proteinase ......................................... 50

VI. Replication ............................................... 51 A. Cellular Location of Comoviral RNA Replication ......... 51 B. Viral Proteins Involved in CPMV Replication .. . . . . . . . . . . . 52 C. Signals on the RNA Molecules .......................... 57 D. Replication of the CPMV RNA Species Is Linked to Their

Translation ............................................ 59 E. A Model for CPMV RNA Replication .................... 60

VII. Intercellular Transport of Comoviruses . . . . . . . . . . . . . . . . . . . . . . 64 A. Cell-to-Cell Movement Using Virus-Induced Tubules ..... 64 B. The 48-kDa Protein Induces Tubules on Pro top lasts . . . . . . . 66 C. General Remarks ....................................... 66

VIII. Concluding Remarks ...................................... 67 References ..................................................... 68

CONTENTS xi

Chapter 4

Comoviruses: Transmission, Epidemiology, and Control

R. C. Gergerich and H. A. Scott

I. Transmission and Epidemiology ............................ 77 A. Modes of Virus Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 B. Weeds as Primary Sources of Infection ................... 86 C. Beetle Ecology ......................................... 87

II. Control ................................................... 89 A. Use of Virus-Free Seed .................................. 89 B. Use of Virus-Free Propagation Material. . . . . . . . . . . . . . . . . . . 90 C. Control of Beetle Vectors ............................... 90 D. Breeding for Resistance ................................. 91

References ..................................................... 93

Chapter 5

Nepoviruses: General Properties, Diseases, and Virus Identification

A. F. Murant, A. T Tones, G. P. Martelli, and R. Stace-Smith

I. Introduction and General Properties ........................ 99 II. Host Ranges and Diseases Caused .......................... 104

A. General Features ....................................... 104 B. Diseases Caused . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

III. Detection, Diagnosis, and Quantitative Assay ............... 121 A. Electron Microscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 B. Serology ............................................... 122

References ..................................................... 127

Chapter 6

Nepoviruses: Molecular Biology and Replication

M. A. Mayo and D. J. Robinson

I. Introduction .............................................. 139 II. Properties of Virus Particles ................................ 140

A. Purification ............................................ 140 B. Particle Size and Structure .............................. 141 C. Sedimentation Properties ............................... 145 D. Isopycnic Centrifugation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 E. Electrophoretic Properties ............................... 146 F. Particle Composition ................................... 147 G. Forces Stabilizing Particles... . . . . ..... . . .. . . . . . .... . . . .. 147

III. Particle Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 A. Preparation ............................................ 148

xii CONTENTS

B. Number and Sizes....... ......... .......... .. ......... . 148 IV. Genomic RNA............................................ 149

A. Sizes .................................................. 149 B. Biochemical Features ................................... 150 C. RNA Packaging........................................ 151 D. Sequences ............................................. 151 E. Noncoding Regions........ ......... .......... .......... 152 F. Sequence Homologies between Viruses.. . ......... . ...... 153

V. Coding Regions and Genome Expression .................... 154 A. Assignment of Function to Genome Parts..... ......... .. 154 B. Domains in Protein Sequences .......................... 155 C. Expression of N epovirus Genomes ....................... 157

VI. Properties of Putative Gene Products. . . . . . . . . . . . . . . . . . . . . . . . 159 A. Products of RNA-2 ..................................... 159 B. Products of RNA-l ..................................... 164

VII. Replication ............................................... 169 A. RNA Polymerase ....................................... 169 B. Intermediates of RNA Replication ....................... 169 C. Recombination ......................................... 170 D. Sites of Virus Replication ............................... III E. Control of Replication .................................. 1 II

VIII. The Impact of Molecular Biology on Nepovirus Classification 172 IX. Satellites of Nepoviruses ................................... 173

A. Type B Satellites ....................................... 174 B. Type D Satellites ....................................... 176

References ..................................................... 177

Chapter 7

N epoviruses: Transmission by Nematodes

D. J. F. Brown, D. L. Trudgill, and W M. Robertson

I. Introduction .............................................. 187 II. The Vector Nematodes. . . ....... . .. ....... .. ........ ....... 188

A. Taxonomy and Biology.................................. 188 B. Structure of the Feeding Apparatus ...................... 188 C. Feeding Behavior ....................................... 190 D. Host Response ......................................... 191

III. Relations between Nepoviruses and Their Nematode Vectors 192 A. Criteria for Demonstrating Nematode Transmission ...... 192 B. Test Procedures ........................................ 192 C. Vector-Virus Associations .............................. 193 D. Specificity ............................................. 196 E. Vector Efficiency ....................................... 199

CONTENTS xiii

F. Sites of Virus Retention. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 201 G. Transmission: Ingestion, Retention, and Inoculation ...... 202

IV. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 References ..................................................... 203

Chapter 8

Nepoviruses: Ecology and Control

B. D. Harrison and A. F. Murant

I. Introduction .............................................. 211 II. Virus Ecology and Epidemiology ............................ 211

A. Naturally Occurring Hosts .............................. 211 B. Occurrence and Population Dynamics of Vector

Nematodes ............................................ 212 C. Patterns of Disease Outbreaks. . . . . . . . . . . . . . . . . . . . . . . . . . . 215 D. Natural Transmission through Seed and Pollen ........... 216 E. Interplay of Virus, Vector, and Host Plant Factors ......... 218

III. Control ................................................... 219 A. Removing Virus Sources ................................ 219 B. Agronomic Methods .................................... 220 C. Application of Nematicides to Soil ...................... 220 D. Virus-Resistant Cultivars ............................... 221 E. Integrated Control ...................................... 223

IV. Concluding Remarks ...................................... 224 References ..................................................... 224

Chapter 9

Fabaviruses: Broad Bean Wilt and Allied Viruses

V Lisa and G. Boccardo

I. Introduction .............................................. 229 II. Hosts and Symptomatology ................................ 230

A. Natural Hosts and Economic Importance. . . . ... . . . . .... .. 230 B. Experimental Host Range ............................... 231 C. Mode of Transmission .................................. 233

III. Cytopathology ............................................ 234 IV. Properties of Particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 237

A. Purification ............................................ 237 B. Types of Particle ....................................... 238 C. The Genome ........................................... 238 D. The Coat Proteins.. .... . . . .. .. . . . . ... . . . . . ... . . .... . . .. 241

V. Relationships ............................................. 241

xiv CONTENTS

A. Relationships within the Genus ......................... 241 B. Affinities with Other Viruses . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

VI. Diagnosis ................................................. 244 VII. Ecology and Control ....................................... 245

References ..................................................... 246

Chapter 10

Dianthoviruses: Properties, Molecular Biology, Ecology, and Control

R. 1. Hamilton and r. H. Tremaine

I. Introduction .............................................. 251 II. Host Range, Symptoms, and Geographical Distribution . . . . . .. 252

III. Virion Properties .......................................... 254 A. Structure .............................................. 254 B. Infectivity-Dilution Curves ............................. 258 C. Virion Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 259 D. Aggregation. . . . . . . . . . . .. . . . . . . . ... . . . . . . . . ... . . . . . . . . .. 260 E. Serological Properties ................................... 261 F. Electrophoretic Mobility ................................ 262

IV. Molecular Biology ......................................... 263 A. Pseudorecombinants .................................... 263 B. dsRNA Species ......................................... 265 C. Genome Strategy and Gene Function .................... 265 D. Sequence Relationships ................................. 270

V. Cytopathology ............................................ 271 VI. Ecology and Control ....................................... 272

A. Ecology ................................................ 272 B. Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274

VII. Concluding Remarks ...................................... 275 References ..................................................... 276

Chapter 11

Raspberry Bushy Dwarf Idaeovirus

A. T. Tones, M. A. Mayo, and A. F. Murant

I. Introduction .............................................. 283 II. Biological Properties ....................................... 284

A. Geographical Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 284 B. Natural and Experimental Transmission ................. 285 C. Disease Symptoms and Effects in Rubus ................. 286 D. Symptoms in Herbaceous Test Plants .................... 289

III. Particle Properties ......................................... 291

CONTENTS xv

A. Purification of Virus Particles ........................... 291 B. Properties of Virus Particles ............................. 291 C. Composition of Virus Particles .......................... 291

IV. Molecular Biology ......................................... 292 A. Nucleotide Sequences .................................. 292 B. Sequence Features of the Viral Proteins .................. 294

V. Detection and Control ..................................... 296 A. Detection in Plants ..................................... 296 B. Therapy ............................................... 298 C. Control in Crops ....................................... 298

VI. Relationships with Other Viruses ........................... 299 References ..................................................... 299

Chapter 12

Pea Enation Mosaic Enamovirus: Properties and Aphid Transmission

S. A. Demler, G. A. de Zoeten, G. Adam, and K. F. Harris

I. Introduction .............................................. 303 II. Biological Properties ....................................... 304

A. Experimental Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 304 B. Symptoms ............................................. 304 C. Cytopathology ......................................... 305

III. Particle Composition and Properties ........................ 307 A. Physicochemical Properties of the Particles . . . . . . . . . . . . . .. 307 B. Coat Protein ........................................... 311 C. Nucleic Acid. ... . . . . . . . .. . . . . . . . . .. . . . .... . . . . .... . . . .. 312 D. RNA Species Composing the Genome and Their

Association with the Sedimenting Components. . . . . . . . . .. 312 IV. Molecular Biology ......................................... 314

A. Molecular Organization of RNA-l ....................... 314 B. Molecular Organization of RNA-2 ....................... 315 C. The PEMV Paradox: Bipartite Genome or Mixed Infection? 316 D. Analogy with Luteovirus Helper-Dependent Virus

Complexes ............................................. 318 E. RNA-3 ................................................ 320 F. RNA Replication and Encapsidation ..................... 321

v. Aphid Transmission ....................................... 321 A. Occurrence of Isolates That Are Transmissible or

Nontransmissible by Aphids ............................ 321 B. Electrophoresis of Virions ............................... 322 C. Coat Proteins Involved in Aphid Transmission ........... 323 D. Genomic Determinants for Aphid Transmissibility ....... 323

xvi CONTENTS

E. Strain Groups .......................................... 329 VI. Virus-Vector Interactions .................................. 330

A. Transmission Characteristics ............................ 330 B. Fate of Virus in the Vector .............................. 331 C. Does PEMV Replicate in the Insect Vector? .............. 336

References ..................................................... 338

Chapter 13

Pea Enation Mosaic Enamovirus: Ecology and Control

D. J. Hagedorn

I. Introduction .............................................. 345 II. Ecology ................................................... 346

A. Host Range ............................................ 346 B. Vector Relations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348 C. Environmental Influences ............................... 349

III. Control ................................................... 350 A. Vector Control ......................................... 350 B. Overwintering Host Control ............................ 351 C. Host Plant Resistance .................................. 352

References ..................................................... 353

Index.......................................................... 357

CHAPTER 1

Plant Viruses with Bipartite RNA Genomes and Polyhedral Particles Diversity and Affinities

B. D. HARRISON AND A. F. MURANT

I. INTRODUCTION

Of the 47 genera of plant viruses recognized by the International Committee on Taxonomy of Viruses (Murphy et a1., 1995), 33 have genomes composed of single-stranded, positive-sense RNA. Of these 33 genera, 9 have bipartite genomes, 19 have isometric particles, and 6 have both characteristics. This volume therefore deals with an important fraction of the known types of plant viruses, namely, members of the genera Comovirus, Dianthovirus, Enamovirus, Fabavirus, Idaeovirus, and Nepovirus.

B. D. HARRISON • Department of Biological Sciences, University of Dundee, Dundee DDl 4HN, United Kingdom. A. F. MURANT • Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom.

2 B. D. HARRISON AND A. F. MURANT

II. RECOGNITION OF THE EXISTENCE OF BIPARTITE RNA GENOMES

Recognition that some plant viruses have bipartite RNA genomes came about by a series of (sometimes faltering) steps, made against the background of evidence that tobacco mosaic virus (TMV) has a monopartite genome: the predominant 300-nm length of its rod-shaped nucleoprotein particles (Williams and Steere, 1951) and the association of infectivity with intact RNA molecules of 2 million Da that can be extracted from them (Gierer, 1957). However, early work by Rice et al. (1955) showed that the isometric particles of squash mosaic comovirus (SMV) sedimented as three components (T, M, and B) with sedimentation coefficients of 56 S, 88 S, and 111 S, respectively. The particles differed in RNA content (0, 27, and 35%, respectively) but not in size (Mazzone et al., 1962). The first attempts to determine the infectivity of particles of the individual components did not provide clear-cut results, doubtless because of the difficulty in obtaining preparations of each component free from the others by the methods then available. However, T component was found to have little if any infectivity (Rice et al., 1955). Also, Wood and Bancroft (1965) made the important discovery that mixtures of M and B components of either cowpea mosaic comovirus (CPMV) or bean pod mottle comovirus were considerably more infective than preparations of either M or B separately. The effect was virus specific: Infectivity was not increased when M component of one of the viruses was mixed with B component of the other.

At about this time, work on another previously little-studied plant virus, tobacco rattle tobravirus (TRV), showed that its rod-shaped nucleoprotein particles are of two predominant lengths (L and S) and contain RNA molecules of two sizes (L-RNA and S-RNA, respectively). When the two types of particle were separated by sucrose density-gradient centrifugation, ability to produce local lesions in inoculated leaves was associated with the long (L) particles or L-RNA only (Harrison and Nixon, 1959a,b). Further work showed that these lesions contained infectious L-RNA, that they did not contain detectable S-RNA or nucleoprotein particles of either length unless the inocula also contained S particles or S-RNA, and that S-RNA controlled the serological specificity of both Land S particles (Lister, 1966; Frost et al., 1967; Siinger, 1968). At first, it seemed that the infectivity of CPMV was likewise associated with particles of the fastest sedimenting component (B component) (Van Kammen, 1967). However, subsequent work, aided by advances in the techniques available for fractionating preparations of virus particles or RNA, led to the conclusion that both M particles (or their RNA) and B particles (or their RNA) are needed for lesion production in inoculated leaves (Bruening and Agrawal, 1967; Van Kammen, 1968; Van Kammen and Van Griensven, 1970). The CPMV system therefore differed somewhat from the tobravirus one. Further work has shown that other comoviruses behave like CPMV, whereas other tobraviruses behave like TRV. Thus different

BIPARTITE RNA GENOMES 3

genera of viruses have bipartite genomes with different kinds of interaction between the genome parts. Subsequent research has greatly increased the number of plant viruses known to have bipartite RNA genomes and has revealed further diversity among their genetic systems.

III. DIVERSITY AND AFFINITIES OF PLANT VIRUSES WITH ISOMETRIC PARTICLES AND BIPARTITE RNA GENOMES

The six virus genera that have isometric particles and bipartite, positivesense, single-stranded RNA genomes fall into four categories, depending on how the genetic information is divided between the two genomic RNA molecules and how this information is expressed. The first and largest category contains the comoviruses, nepoviruses, and fabaviruses, which are grouped together in the family Comoviridae. The larger genomic RNA molecule (RNA-I) of these viruses specifies the viral RNA replicase, whereas the smaller RNA (RNA-2) specifies the coat protein(s) and movement protein (Table I). Each RNA species is translated to produce a polyprotein that is cleaved into functional proteins, largely by the action of a viral protease that is encoded by RNA-I. Other similarities include a small virus-encoded protein that is covalently attached to the 5' end, and a polyadenylate sequence at the 3' end of each RNA species. However, comoviruses are transmitted by beetles, whereas fabaviruses are transmitted by aphids, and many nepoviruses have soil-inhabiting nematode vectors (though some spread in association with pollen). Also, whereas particles of definitive nepoviruses typically contain only one species of protein, those of comoviruses, fabaviruses, and a few atypical nepoviruses contain two protein species.

Genera in the Comoviridae have their closest affinities, as assessed by genetic constitution, encoded amino acid sequences, and mode of gene expression, not with other viruses that have bipartite RNA genomes and isometric particles but with viruses that have monopartite RNA genomes, or elongated particles, or both. For example, with parsnip yellow fleck sequivirus (PYFV), which has isometric particles and a monopartite genome, the types and arrangement of genes in the RNA, and the amino acid sequences of the viral proteins, are similar to those of members of the Comoviridae (Turnbull-Ross et a1., 1993). Obvious similarities also exist between viruses in the family Comoviridae and members of the family Potyviridae, which have flexuous filamentous particles; within this family, the similarities are greatest to the bymoviruses, which have a bipartite genome (Koenig and Huth, 1988; Kashiwazaki et a1., 1990, 1991; Davidson et a1., 1991), but are fungus-transmitted.

Only one idaeovirus, raspberry bushy dwarf virus (RBDV), is described. It differs from members of the Comoviridae in several ways. It does not produce a T component, its nucleoprotein particles are labile, its two genomic RNA species are not polyadenylated and do not have a covalently bonded

TA

BL

E I

. S

ome

Cha

ract

eris

tics

of

Pla

nt V

irus

Gen

era

Hav

ing

Isom

etri

c P

arti

cles

and

Bip

arti

te G

enom

es o

f S

ingl

e-S

tran

ded

RN

A

Par

ticl

e L

ocat

ion

of g

enes

enc

odin

g

No.

di

amet

er

Coa

t M

ovem

ent

Mod

e of

G

enus

sp

ecie

sa

(nm

) P

olym

eras

e pr

otei

n(s)

fu

ncti

on

Exp

ress

ion

syst

em

natu

ral

tran

smis

sion

Com

ovir

us

15

28

-30

R

NA

-l

RN

A-2

R

NA

-2

Tw

o po

lypr

otei

ns c

leav

ed b

y vi

ral

pro-

Bee

tles

te

asel

s)

Nep

ovir

us

28 (

+8)

2

8-3

0

RN

A-l

R

NA

-2

RN

A-2

A

s co

mov

irus

es

Nem

atod

es o

r po

llen

F

abav

irus

3

28

-30

R

NA

-l

RN

A-2

R

NA

-2

As

com

ovir

uses

A

phid

s (n

onpe

rsis

tent

) Id

aeov

irus

1

33

RN

A-l

R

NA

-2

RN

A-2

b D

irec

t an

d vi

a su

bgen

omic

RN

A

Via

pol

len

Dia

ntho

viru

s 3

(+1)

3

2-3

5

RN

A-l

R

NA

-l

RN

A-2

A

s id

aeov

irus

V

ia s

oil

Ena

mov

irus

1

25 a

nd 2

8 R

NA

-l a

nd

RN

A-l

R

NA

-2

Dir

ect,

by

fram

eshi

ft t

rans

lati

on,

via

Aph

ids

(per

sist

ent)

R

NA

-2

subg

enom

ic R

NA

and

by

stop

cod

on

read

-thr

ough

"Dat

a fr

om t

he S

ixth

Rep

ort

of t

he I

nter

nati

onal

Com

mit

tee

on T

axon

omy

of V

irus

es (

Mur

phy

et a

l., 1

995)

. Fi

gure

s in

par

enth

eses

are

num

ber

of t

enta

tive

spe

cies

. bT

enta

tive

ass

ignm

ent.


protein at their 5' ends, and its coat protein is translated from a subgenomic RNA (Natsuaki et a1., 1991; Ziegler et a1., 1992). Indeed, its strongest affinities are with viruses with tripartite RNA genomes, such as alfalfa mosaic virus and ilarviruses, which are members of the Bromoviridae (Ziegler et a1., 1993). Like ilarviruses, RBDV is transmitted from plant to plant in association with pollen.

Dianthoviruses form the third category. Their coat protein gene is in RNA-1 (unlike viruses in the two previous categories) and is expressed from a subgenomic RNA (Table I). The viral polymerase is thought to be expressed by frameshift translation of RNA-I, as in luteoviruses, but the closest sequence similarities of both the viral polymerase and the coat protein are to the corresponding proteins of carmoviruses, a genus with monopartite genomes (Xiong and Lommel, 1989).

The last category is represented by a single virus, pea enation mosaic enamovirus (PEMV), which, unlike the viruses in the first three categories, is transmitted by aphids in the persistent manner. It differs from all other plant viruses studied in that each genome segment contains a polymeraselike gene and can replicate in protoplasts independently of the other segment. However, RNA-l contains the coat protein gene, whereas RNA-2 contains information necessary for systemic movement of both RNA molecules (Table I), so that both RNA species are needed to establish a particle-producing systemic infection. The two genomic RNA species have different affinities, RNA-l with luteovirus RNA, and RNA-2 with umbravirus RNA (Demler and De Zoeten, 1991; Demler et a1., 1993).

From these simple comparisons, it is clear that plant viruses with isometric particles and single-stranded, positive-sense RNA genomes are an assortment of different types. Their genomes differ in gene content and arrangement, and in translation strategy. Their biological characteristics are equally diverse, with symptoms ranging from mosaics or yellowing to necrosis or enations, and modes of transmission involving aphids, beetles, nematodes, or pollen.

IV. POSSIBLE ORIGINS OF BIPARTITE RNA GENOMES

The occurrence of bipartite RNA genomes among viruses with diverse kinds of genes, gene arrangements and expression strategies, and contrasting biological characteristics, supports the idea that multipartite RNA genomes have arisen on different occasions and in different viral lineages. How bipartite genomes have arisen is a matter for conjecture, but might have occurred in a variety of ways.

1. Division of a monopartite genome into two parts. 2. Acquisition of an extra genome part by a virus with a monopartite

genome.


3. Divergence of a copy of a viral genome in cultures that retain the original form, to a point where the altered form confers benefits on the culture and becomes an additional genome part.

4. Association of nucleic acid molecules that contain genes with complementary functions.

5. Association of two dissimilar viruses followed by complementary loss of function, resulting in interdependence of the two defective genomes.

It is unlikely that the bipartite genomes of the six genera dealt with in this volume have all arisen in just one of these ways, and seems more probable that different families/genera have arisen by different routes. For example, the genomes of Comoviridae resemble a split version of the genome of the sequivirus, PYFV (Turnbull-Ross et a1., 1993). The order of the genes in the genomes of members of the Comoviridae seems to be the same as that in PYFV RNA, with the RNA-2 of Comoviridae being equivalent to the 5' part of the PYFV genome and RNA-1 being equivalent to the 3' part. In addition, there are obvious sequence similarities among the equivalent encoded proteins. Several of the amino acid motifs found in the proteins of members of the Comoviridae and PYFV can also be found in animalinfecting picornaviruses, such as poliovirus, which have monopartite genomes. These similarities, together with those in genome structure and in method of gene expression via polyproteins, are strong evidence that all these viruses are representatives of the same evolutionary lineage (Goldbach, 1986). The question of whether the monopartite or the bipartite genome is the ancestral form remains open. However, the occurrence of monopartite but not bipartite genome picornalike viruses in invertebrates and vertebrates, and of monopartite as well as bipartite genome forms in plants, tends to support a monopartite genome ancestry. Moreover, the relative frequency of occurrence, and number and variety of different species of Comoviridae recognized, compared with the infrequent occurrence and paucity of recognized species of Sequiviridae, suggest that bipartite genomes may represent more successful, and perhaps more recently evolved, forms of plant-adapted picornaviruses.

Similar considerations apply when the idaeovirus, RBDV, is compared with the viruses that have the greatest affinity with it, namely members of the Bromoviridae (Ziegler et a1., 1993). This comparison shows that the 5' and 3' parts of RBDV RNA-1 are equivalent to RNA-1 and RNA-2, respectively, in the tripartite genome of members of the Bromoviridae. RBDV RNA-2 is equivalent to RNA-3 of Bromoviridae. RBDV seems more similar to ilarviruses in particle structure and mode of natural spread (via pollen) than to other members of the Bromoviridae, but little sequence information is available for ilarviruses, and the closest known sequence similarities are to alfalfa mosaic alfamovirus. As to whether the bipartite or tripartite genome is the more derived form, it is tempting to speculate that the bipartite


idaeovirus form is ancestral and that its rarity compared with the number and variety of species, and relative frequency of occurrence, of the tripartite genome Bromoviridae are evidence of the more recent evolution of a range of better-adapted forms. There is no evidence of a monopartite genome virus that could resemble an idaeovirus ancestor, but the above discussion on the origin(s) of Comoviridae suggests that its discovery would not be a surprise.

Dianthoviruses have their closest affinities with carmoviruses, but differ in having the movement protein gene on a separate RNA species (Lommel et a1., 1988; Osman et a1., 1986). As already noted, the similarities between dianthoviruses and carmoviruses are in their replicases and coat proteins (Xiong and Lommel, 1989). In contrast, the RNA-2-encoded movement protein of dianthoviruses has little resemblance to any carmovirus protein, and comparisons with carmovirus sequences give no clue to the origin of dianthovirus RNA-2. Perhaps it has been acquired from a source unrelated to carmoviruses. The closest analogy to such an event is found in beet necrotic yellow vein furovirus (BNYVV), RNA-1 and RNA-2 of which can together infect plants and produce particles. However, when RNA-3 is added to the culture, systemic invasion of root and shoot tissues is much more extensive and transmission by the fungal vector is enhanced (Tamada and Abe, 1989). Thus in this system, RNA-3 aids virus spread but is not essential. Dianthovirus RNA-1 can replicate and produce nucleoprotein particles in the absence of RNA-2 (Osman and Buck, 1987). Acquisition of RNA-2 could therefore confer similar advantages and could have happened in the same way as acquisition of RNA-3 of BNYVV.

The bipartite genome of PEMV, the sole member of the genus Enamovirus, is very different from those of the other five genera considered here. Each genomic RNA species can replicate in protoplasts in the absence of the other and, in contrast to other bipartite genomes, the 5' and 3' ends of the different genome segments are not shared (Demler et a1., 1993, 1994). Examination of the deduced amino acid sequences of PEMV proteins suggests that RNA-1 represents the genome of a defective luteovirus, whereas RNA-2 resembles that of an umbravirus. Indeed, the interaction between the two RNA species is closely related to that found in luteovirus-dependent virus complexes (Murant, 1993). An example is the complex of two viruses that jointly cause carrot motley dwarf disease, namely carrot red leaf luteovirus (CaRLV) and carrot mottle umbravirus (CMoV). CaRLV is confined to phloem tissue and is transmissible in the persistent manner by aphids but not by inoculation with sap. CMo V is sap-transmissible and invades mesophyll cells but seems not to produce virions and relies for aphid transmission on its RNA becoming packaged in CaRLV coat protein in source plants infected with both viruses (Waterhouse and Murant, 1983). The difference between the PEMV and carrot motley dwarf systems is that, unlike CaRLV, PEMV RNA-1 seems incapable of spreading from cell to cell within the phloem in the absence of RNA-2. Moreover, the presence of RNA-2 enables PEMV RNA-1 not only to spread within the phloem but also to invade


mesophyll tissues, with the result that both RNA species are transmissible by inoculation with sap from virus-infected plants (Demler et a1., 1994). PEMV therefore seems to have resulted from the association of two unrelated viruses, followed by partial disablement of one or both of the ancestral genomes to produce two mutually dependent RNA species. The fact that only one enamovirus is known suggests that its genome is not the result of a common molecular event, although virus complexes such as those involved in the prevalent carrot motley dwarf and groundnut rosette diseases (Watson et a1., 1964j Hull and Adams, 1968) provide ample opportunity for it to occur.

Although speculative, the foregoing discussion leads to the conclusion that bipartite genomes have arisen in different ways at different times and in different viral lineages. The processes involved in the production of bipartite genomes are of course complemented by events that may lead to evolutionary change in viral genomes of any type: gene mutation, gene duplication and divergence, gene shuffling, capture of nucleotide sequences, unequal crossing over, and so forth. The genera dealt with in this volume therefore are a diverse assemblage of viruses, some of which share little other than the characteristics referred to in the title.

V. ADVANTAGES AND DISADVANTAGES OF BIPARTITE RNA GENOMES

The existence of many plant viruses with bipartite RNA genomes, scattered apparently among different evolutionary lineages, suggests that such genomes have some advantages. Conversely, the occurrence of plant viruses with genomes of other types indicates that any advantages that bipartitegenome viruses may have are not overwhelming, may not apply in all circumstances, and may be accompanied by disadvantages. We discuss these possible advantages and disadvantages below. Some relate to virus infection, replication, spread through the plant, and transmission from plant to plant, whereas others concern genetic stability and variation, and adaptability to environmental change. Further information, together with more detailed consideration of some aspects of these questions, is provided by Lane (1979) and Chao (1994).

Possible advantages of bipartite genomes include the following:

1. The genomic nucleic acid molecules are smaller, allowing them to be replicated faster than larger nucleic acid molecules. Although this is plausible, evidence from experiments on virus replication in protoplasts is inadequate to support or refute it.

2. The timing and amount of translation of different viral genes can be differentially controlled. Although this is probably correct, some monopartite genome viruses have evolved alternative ways of doing it.


3. Smaller genomic nucleic acid molecules can be packaged in smaller nucleoprotein particles than can larger molecules. It is suggested that this process is likely to be simpler and more efficient (faster, less prone to error), but we know of no good evidence for this proposition.

4. Smaller nucleoprotein particles are likely to be more readily transported than larger ones from cdl to cell through plasmodesmata. This may apply to Comoviridae, particles of which are transported from cell to cell through tubules composed of the viral movement protein. However, the monopartite genome sequivirus, PYFV, which has 33-nm diameter particles, nevertheless produces virion-containing tubules that pass through plasmodesmata.

5. Smaller genomic nucleic acid molecules present smaller targets than larger ones for inactivation by physical, chemical, or enzymic agents. Moreover, a bipartite genome facilitates the elimination of deleterious mutations because wild-type molecules are likely to be more common among smaller nucleic acid molecules than among larger ones.

6. A bipartite genome facilitates the positive selection of better-fitted mutants because advantageous mutations are less likely to be accompanied by deleterious ones in a smaller nucleic acid molecule than in a larger one.

7. Pseudorecombination (reassortment of the genome segments of different parental isolates) enhances the amount of variation stemming from a few mutations while maintaining the mutual adaptation of genes within the same genome segment. For example, the coat proteins and movement protein of comoviruses are mutually adapted for tubule-mediated cell-to-cell transport of virions and are both encoded by the same genome segment. Bipartite genomes therefore seem likely to adapt to environmental change more rapidly than monopartite genomes of the same size.

8. Bipartite genomes that arise by association of two unrelated viruses may possess combinations of genes that confer novel properties. For example, association of the luteoviruslike RNA-l of PEMV with the umbraviruslike RNA-2 enables RNA-l to invade mesophyll tissue and enables particles containing RNA-l to be produced in greater amounts than would be those of a typical phloem-limited luteovirus. The presence of RNA-I-containing particles in mesophyll tissue also enables them to be acquired and inoculated by aphids in shorter feeding periods than are necessary for luteovirus particles.

Possible disadvantages of bipartite genomes include the following:

1. Both genome segments must be replicated and expressed to give a full infection. For example, comovirus RNA-l can replicate in protoplasts and single cells, but nucleoprotein particles are not produced and cellto-cell movement does not occur in the absence of RNA-2. The


chance of the two genome segments reaching the same infectible site is less than that of one genomic nucleic acid molecule doing so, especially when the two segments are packaged in different virus particles. This difficulty is probably largely avoided when the two genome segments are packaged in the same virus particle, as in dianthoviruses.

2. Viral RNA polymerase must recognize both genome segments. This is usually achieved by occurrence of the same terminal nucleotide sequences in both segments, especially at their 3' ends. In nepoviruses, there is evidence (Le Gall et a1., 1995) that 3' sequence identity can be produced by recombination between genome segments. PEMV, however, has different 3' sequences in RNA-1 and RNA-2, but also has a specific polymerase gene in each genome segment. This seems to be a less economical system.

3. Viral coat protein must recognize and package both genome segments, implying that both RNA species share the sequence or structure that is necessary for recognition.

4. For viruses that pass from cell to cell as RNA, not as virus particles, both genome segments must be recognized by the viral movement protein. This seems not to be a problem in practice because, in those systems studied in sufficient detail, movement proteins bind to all viral single-stranded RNA species tested (Citovsky et a1., 1990).

5. Vectors must deliver both genome segments, and therefore (except for dianthoviruses) the two kinds of virus particle, to the same infectible site. Although this requirement would seem likely to be a considerable disadvantage, and no doubt vectors often deliver only one particle type, the difficulty is in practice overcome in several kinds of transmission system, perhaps in many instances because the high virus concentrations reached in infected plants make them potent sources of inoculum. Among the viruses considered here, vectors range from aphids to beetles and nematodes, transmission type may be noncirculative or circulative, and some viruses are transmitted to plants via pollen.

To summarize, it seems clear that the advantages of bipartite genomes, as compared with monopartite ones, are essentially balanced by the disadvantages in a considerable variety of plant viruses. The main advantages seem to be in enabling genomes with sizes larger than about 8 kb to be packaged in particles of 30-nm diameter or less, in efficient regulation of gene expression, and in enabling greater genetic variation to occur within virus populations. The main disadvantages are the need for different RNA species to reach the same site to produce a full infection and for vector transmission systems that deliver several particles to one site. However, it is difficult to point to particular circumstances in which bipartite RNA genome viruses have proved to be markedly more or less successful than other plant viruses,


and they may simply represent an alternative way in which infectious nucleic acids can survive in nature.

The relative infrequency of examples of bipartite-genome, singlestranded RNA viruses that infect vertebrates or invertebrates is intriguing. Key factors may be the different methods of infection, involving receptors and the need for the complete viral genome to be contained in one virion, the lack of vector systems suitable for delivering two or more virus particles to the same infectible site, and the transient nature of viremia in infected hosts, which therefore limits the availability of virus particles to bloodsucking vectors.

VI. DEPENDENCE, SATELLITISM, AND MULTIPARTITE GENOMES

Table II lists some plant virus systems in which RNA species interact in a variety of ways. Where the dependence between two RNA species is mutual, they are considered to be the components of a bipartite genome. Where the dependence is for replication and is unilateral, as in nepovirus satellite RNA or tobacco necrosis satellite virus, the term satellite seems warranted. However, in the other systems listed, the status of some individual components is equivocal. For example, the luteovirus CaRLV and the umbravirus CMo V can each replicate and invade plants independently and are given different names, although CMoV is totally dependent, for spread by aphids, on its RNA being packaged in the coat protein of the aphid-transmissible CaRLV (Murant et a1., 1985). Here, the ability of CMoV RNA to replicate and cause systemic symptoms on its own probably justifies the designation "dependent virus."

But how is RNA-3 of the furovirus BNYVV to be described? RNA-1 plus RNA-2 can replicate, infect plants, and be transmitted occasionally by the fungal vector, Polymyxa betae. However, as already explained, although RNA-3 is not an essential component of the system, in practice it is so important that virtually all field isolates of the virus contain it (Tamada and Abe, 1989; Tamada et a1., 1989). RNA-3 can therefore be regarded either as a helpful satellite RNA or as a genome part. Perhaps it is in the course of evolving from a satellite RNA into a genomic RNA.

Even more complex is the role of the satellite RNA associated with groundnut rosette disease (Murant, 1990; Murant and Kumar, 1990). Groundnut rosette assistor luteovirus (GRAV) and groundnut rosette umbravirus (GRV) interact in essentially the same way as CaRLV and CMoV. However, GRAV does not mediate the aphid transmission of GRV unless the satellite RNA is also present in the source plants; no doubt as a consequence of this, the satellite RNA seems always to be associated with GRV in the field. The satellite RNA, which depends on GRV for its replication, is also largely responsible for inducing the symptoms of rosette disease. Thus, this satellite

TA

BL

E I

I. 'T

YPe

s of

Int

erac

tion

am

ong

Gen

omic

RN

A S

peci

es a

nd S

atel

lite

RN

A i

n P

lant

Vir

us S

yste

ms

No.

RN

A

No.

typ

es o

f N

o. t

ypes

of

No.

typ

es o

f S

yste

m

spec

ies

part

icle

coa

t R

NA

pol

ymer

ase

mov

emen

t pr

otei

n T

ype

of d

epen

denc

e

Tob

ravi

rus

RN

A-l

an

d R

NA

-2

2 M

utu

al

Nep

ovir

us R

NA

-l a

nd

RN

A-2

2

Mu

tual

N

epov

irus

and

sat

elli

te R

NA

3

Sat

elli

te R

NA

on

nep

ovir

us

RN

A-l

plu

s R

NA

-2

Bee

t ne

crot

ic y

ello

w v

ein

viru

s 2

Mu

tual

(B

NY

VV

) R

NA

-l a

nd

RN

A-2

B

NY

VV

and

its

RN

A-3

3

2 R

NA

-3 o

n B

NY

VV

RN

A-l

plu

s R

NA

-2;

RN

A-l

plu

s R

NA

-2

on

RN

A-3

T

obac

co n

ecro

sis

viru

s (T

NV

) 2

2 T

NV

sat

elli

te v

irus

on

TN

V

and

TN

V s

atel

lite

vir

us

Pea

ena

tion

mos

aic

viru

s 2

2 11

M

utu

al

RN

A-l

and

RN

A-2

C

arro

t re

d le

af v

irus

and

car

rot

2 2

21

Car

rot

mo

ttle

vir

us o

n ca

rrot

m

ott

le v

irus

re

d le

af v

irus

G

roun

dnut

ros

ette

ass

isto

r vi

rus

3 2

21

GR

V o

n G

RA

V;

GR

V s

atel

lite

(G

RA

V),

grou

ndnu

t ro

sett

e R

NA

on

GR

V;

GR

V o

n G

RV

vi

rus

(GR

V),

and

GR

V

sate

llit

e R

NA

sa

tell

ite

RN

A


RNA seems almost to merit the status of part of the CRV genome. However, it is not yet clear whether it functions through the activity of any translation product.

In this series of comparisons, we see the wide range of types of interdependence among infectious RNA molecules. They probably survive in any way or combination of ways open to them, so defying man's attempts to put them in a few mutually exclusive categories.

REFERENCES

Bruening, G., and Agrawal, H. 0., 1967, Infectivity of a mixture of cowpea mosaic virus ribonucleoprotein components, Virology 32:306.

Chao, 1., 1994, Evolution of genetic exchange in RNA viruses, in: The Evolutionary Biology of Viruses (S. S. Morse, ed.), pp. 233-250, Raven Press, New York.

Citovsky, V., Knorr, D., Schuster, G., and Zambryski, P., 1990, The P30 movement protein of tobacco mosaic virus is a single-strand nucleic acid binding protein, Cell 60:637.

Davidson, A D., Prols, M., Schell, J., and Steinbiss, H.-H., 1991, The nucleotide sequence of RNA 2 of barley yellow mosaic virus, r. Gen. Virol. 72:989.

Demler, S. A, and De Zoeten, G. A., 1991, The nucleotide sequence and luteovirus-like nature of RNA 1 of an aphid non-transmissible strain of pea enation mosaic virus, r. Gen. Virol. 72:1819.

Demler, S. A., Rucker, D. G., and De Zoeten, G. A., 1993, The chimeric nature of the genome of pea enation mosaic virus: The independent replication of RNA 2, r. Gen. Virol. 74:l.

Demler, S. A., Borkhsenious, O. N., Rucker, D. G., and De Zoeten, G. A., 1994, Assessment of the autonomy of replicative and structural functions encoded by the luteo-phase of pea enation mosaic virus, r. Gen. Virol. 75:997.

Frost, R. R., Harrison, B. D., and Woods, R. D., 1967, Apparent symbiotic interaction between particles of tobacco rattle virus, T. Gen. Virol. 1:57.

Gierer, A, 1957, Structure and biological function of ribonucleic acid from tobacco mosaic virus, Nature 179:1297.

Goldbach, R. w., 1986, Molecular evolution of plant RNA viruses, Annu. Rev. Phytopathol. 24:289.

Harrison, B. D., and Nixon, H. 1., 1959a, Separation and properties of particles of tobacco rattle virus with different lengths, T. Gen. Microbiol. 21:569.

Harrison, B. D., and Nixon, H. 1., 1959b, Some properties of infective preparations made by disrupting tobacco rattle virus with phenol, T. Gen. Microbiol. 21:59l.

Hull, R., and Adams, AN., 1968, Groundnut rosette and its assistor virus, Ann. Appl. Biol. 62:139.

Kashiwazaki, S., Minobe, Y., Omura, T., and Hibino, H., 1990, Nucleotide sequence of barley yellow mosaic virus RNA 1: A close evolutionary relationship with potyviruses, r. Gen. Virol. 71:278l.

Kashiwazaki, S., Minobe, Y., and Hibino, H., 1991, Nucleotide sequence of barley yellow mosaic virus RNA 2, r. Gen. Virol. 72:995.

Koenig, R., and Huth, w., 1988, RNA/cDNA hybridization and infectivity tests suggest that barley yellow mosaic virus isolate M has a bipartite genome, r. Phytopathol. 121:370.

Lane, 1. C., 1979, The RNAs of multipartite and satellite viruses of plants, in: Nucleic Acids in Plants (T. C. Hall and J. w. Davies, eds.), pp. 65-110, CRC Press, Boca Raton, Flo

Le Gall, 0., Candresse, T., and Dunez, J., 1995, Transfer of the 3' non-translated region of grapevine chrome mosaic virus RNA-l by recombination to tomato black ring virus RNA-2 in pseudorecombinant isolates, r. Gen. Virol. 76:1285.

Lister, R. M., 1966, Possible relationships of virus-specific products of tobacco rattle virus, Virology 28:350.


Lommel, S. A., Weston-Fina, M., Xiong, Z., and Lomonossoff, C. P., 1988, The nucleotide sequence and gene organization of red clover necrotic mosaic virus RNA-2, Nucleic Acids Res. 16:8587.

Mazzone, H. M., Incardona, N. 1., and Kaesberg, P., 1962, Biochemical and biophysical properties of squash mosaic virus and related macromolecules, Biochim. Biophys. Acta 55:164.

Murant, A. F., 1990, Dependence of groundnut rosette virus on its satellite RNA as well as on groundnut rosette assistor luteovirus for transmission by Aphis craccivora, r. Gen. Viral. 71:2163.

Murant, A. F., 1993, Complexes of transmission-dependent and helper viruses, in: Diagnosis of Plant Virus Diseases (R. E. F. Matthews, ed.), pp. 333-357, CRC Press, Boca Raton, F1.

Murant, A. F., and Kumar, I. K., 1990, Different variants of the satellite RNA of groundnut rosette virus are responsible for the chlorotic and green forms of groundnut rosette disease, Ann. Appl. BioI. 117:85.

Murant, A. F., Waterhouse, P. M., Raschke, J. H., and Robinson, D. J., 1985, Carrot red leaf and carrot mottle viruses: Observations on the composition of the particles in single and mixed infections, r Gen. Viral. 66:1575.

Murphy, F. A., Fauquet, C. M., Bishop, D. H. 1., Chabrial, S. A., Jarvis, A. w., Martelli, C. P., Mayo, M. A., and Summers, M. D. (eds.), 1995, Virus Taxonomy-Classification and Nomenclature of Viruses. Sixth Report of the International Committee on Taxonomy of Viruses, Springer-Verlag, Vienna. (Also in Arch. Viral., Supplementum 10)

Natsuaki, T., Mayo, M. A., Jolly, C. A., and Murant, A. F., 1991, Nucleotide sequence of raspberry bushy dwarf virus RNA-2: a bicistronic component of a bipartite genome, r. Gen. Viral. 72:2183.

Osman, T. A. M., and Buck, K. W., 1987, Replication of red clover necrotic mosaic virus RNA in cowpea protoplasts: RNA 1 replicates independently of RNA 2, r Gen. Viral. 68:289.

Osman, T. A. M., Dodd, S. M., and Buck, K. W., 1986, RNA 2 of red clover necrotic mosaic virus determines lesion morphology and systemic invasion in cowpea, r. Gen. Virol. 67:203.

Rice, R. v., Lindberg, C. D., Kaesberg, P., Walker, J. c., and Stahmann, M. A., 1955, The three components of squash mosaic virus, Phytopathology 45:145.

Sanger, H. 1., 1968, Characteristics of tobacco rattle virus. I. Evidence that its two particles are functionally defective and mutually complementing, Mol. Gen. Genet. 101:346.

Tamada, T., and Abe, H., 1989, Evidence that beet necrotic yellow vein virus RNA-4 is essential for efficient transmission by the fungus Polymyxa betae, r. Gen. Viral. 70:339l.

Tamada, T., Shirako, Y., Abe, H., Saito, M., Kiguchi, T., and Harada, T., 1989, Production and pathogenicity of isolates of beet necrotic yellow vein virus with different numbers of RNA components, r Gen. Viral. 70:3399.

Turnbull-Ross, A. D., Mayo, M. A., Reavy, C., and Murant, A. F., 1993, Sequence analysis of the parsnip yellow fleck virus polyprotein: Evidence of affinities with picornaviruses, r. Gen. Viral. 74:555.

Van Kammen, A., 1967, Purification and properties of the components of cowpea mosaic virus, Viralogy 31:633.

Van Kammen, A., 1968, The relationship between the components of cowpea mosaic virus. I. Two ribonucleoprotein particles necessary for the infectivity of CPMV, Viralogy 34:312.

Van Kammen, A., and Van Criensven, 1. J. 1. D., 1970, The relationship between the components of cowpea mosaic virus. II. Further characterization of the nucleoprotein components of CPMV, Viralogy 41:274.

Waterhouse, P. M., and Murant, A. F., 1983, Further evidence on the nature of the dependence of carrot mottle virus on carrot red leaf virus for transmission by aphids, Ann. Appl. BioI. 103:455.

Watson, M. A., Serjeant, E. P., and Lennon, E. A., 1964, Carrot motley dwarf and parsnip mottle viruses, Ann. Appl. BioI. 54:153.

Williams, R. c., and Steere, R. 1., 1951, Electron microscopic observations on the unit length of the particles of tobacco mosaic virus, r. Am. Chem. Soc. 73:2057.


Wood, H. A., and Bancroft, J. B., 1965, Activation of a plant virus by related incomplete nucleoprotein particles, Virology 27:94.

Xiong, Z., and Lommel, S. A. 1989, The complete nucleotide sequence and genome organization of red clover necrotic mosaic virus RNA-I, Virology 171:543.

Ziegler, A., Natsuaki, T., Mayo, M. A., Jolly, C. A., and Murant, A. F., 1992, The nucleotide sequence of RNA-l of raspberry bushy dwarf virus, 1. Gen. Virol. 73:3213.

Ziegler, A., Mayo, M. A., and Murant, A. F., 1993, Proposed classification of the bi-partite genomed raspberry bushy dwarf idaeovirus with tri-partite-genomed viruses in the family Bromoviridae, Arch. Virol. 131:483.

CHAPTER 2

Comoviruses: Identification and Diseases Caused R. A. VALVERDE AND J. P. FULTON

I. IDENTIFICATION OF COMOVIRUSES

A. Members of the Genus

The comoviruses are remarkably similar in many of their properties, such as the size and morphology of their particles and the sizes of their particle proteins and RNA components. They do not differ consistently in physical properties, such as thermal inactivation point (60-75 0C), longevity in vitro (a few weeks), and dilution endpoint (10-3-10-5). All comoviruses are good antigens and several are serologically related to others. All are, or are presumed to be, transmitted by leaf-feeding beetles. Their pathology, epidemiology, and control were reviewed by Stace-Smith (1981).

Because of these basic similarities, it is difficult to draw sharp lines of distinction between individual comoviruses. It could be argued that the genus contains a single virus species and that all members are simply strains. At the other extreme, all variants recognized could be individual virus species; however, the great variability in characteristics such as host reaction and serology make such a listing cumbersome and unrealistic. Specifying the individual members of the genus therefore becomes a matter of preference

R. A. VALVERDE • Department of Plant Pathology and Crop Physiology, Louisiana State University, Baton Rouge, Louisiana 70803. J. P. FULTON • Department of Plant Pathology, University of Arkansas, Fayetteville, Arkansas 7270l.

17

18 R. A. VALVERDE AND J. P. FULTON

and is obviously arbitrary. In the Fifth Report of the International Committee on Taxonomy of Viruses (ICTV), Francki et a1. (1991) distinguished 14 species of comoviruses. These are listed in Table I, together with the acronyms used in this chapter (Hull et a1., 1991). Cowpea mosaic comovirus (CPMV) strain SB is considered to be the type member.

Serology is the technique of choice for recognizing and distinguishing comoviruses. However, the range of variation among isolates considered to comprise one species is unclear in most instances, and further work on the similarities and differences among larger numbers of isolates might enable different comoviruses to be identified and distinguished with greater confidence. In practice, we are obliged to make decisions based on data for only one or two isolates, which do not necessarily typify the population as a whole.

B. Criteria for Grouping or Distinguishing Comovirus Species

1. Host Range and Symptoms

Host range and symptoms are the characteristics usually encountered first in identifying a comovirus. There are three different categories of host that may be of significance. These are the cultivated hosts in which the virus occurs, the wild hosts in which the virus occurs naturally, and the extended host range derived from mechanical transmission tests.

The host range of each comovirus is typically limited in extent, involving principally one plant family. With four comoviruses, the host range is of prime importance for identification. Ullucus virus C (UVC) occurs in Ullucus tuberosus (Basellaceae) and can be transmitted by mechanical means to Chenopodium amaranticolor, C. quinoa, C. murale, and Tetragonia expansa (Brunt and Jones, 1984. All known hosts of Andean potato mottle virus (APMV) are in the family Solanaceae with the exception of Gomphrena globosa (Fribourg et a1., 1979). Radish mosaic virus (RaMV) infects plants mainly in the Cruciferae, but local lesions are produced on a few species in the Solanaceae, Chenopodiaceae, and Cucurbitaceae (Campbell, 1964). Squash mosaic virus (SMV) infects mainly members of the Cucurbitaceae, but also infects symptomlessly a few species in the Hydrophyllaceae, Leguminosae, and Umbelliferae (Campbell, 1971). With each of these four viruses, the host range would essentially identify the virus, subject to verification by a serological test.

The principal hosts of all other reported comoviruses are species in the Leguminosae, and host range is much less useful for distinguishing and identifying these viruses. The host range can be misleading because individual isolates of a virus may differ in host range and in the severity of symptoms induced. The crop species in which the virus occurs and the geographic area of occurrence are often better clues to the identity of the virus than the

COMOVIRUSES: IDENTIFICATION AND DISEASES

TABLE I. The Comoviruses

Andean potato mottle (APMV) (Fribourg et a1., 1977, 1979) Strain C (Salazar and Harrison, 1978) Strain H (Salazar and Harrison, 1978) Strain B (Avila et a1., 1984)

Bean pod mottle (BPMV) (Zaumeyer and Thomas, 1948; Semancik, 1972) J-lO (Moore and Scott, 1971)

Bean rugose mosaic (BRMV) (Gamez, 1972, 1982) Ampollado (Galvez et a1., 1977) Mosaico-em-Desenho (Lin et a1., 1981b)

Broad bean stain (BBSV) (Gibbs and Smith, 1970) Pea green mottle (Valenta et a1., 1969) MF virus (Devergne and Cousin, 1966) Pea seed-borne symptomless (Musil et a1., 1983; Kowalska and Beczner, 1980)

Broad bean true mosaic (BBTMV) (Gibbs and Paul, 1970) Echtes Ackerbohnenmosaik (Quantz, 1953)

Cowpea mosaic (CPMV) (Van Kammen and De Jager, 1978) Cowpea yellow mosaic (Chant, 1959) Cowpea mosaic virus-SB (Agrawal, 1964)

Cowpea severe mosaic (CPSMV) (De Jager, 1979) Cowpea mosaic virus-Vs (Agrawal, 1964) Cowpea mosaic virus-Vu (Agrawal, 1964) Cowpea mosaic virus-Arkansas (Shepherd, 1964)

Glycine mosaic (GMV) (Bowyer et a1., 1980) Glycine mosaic-GW (Bowyer et a1., 1980)

Pea mild mosaic (PMiMV) (Clark, 1972) Quail pea mosaic (QPMV) (Moore, 1973; Moore and Scott, 1981)

Bean curly dwarf mosaic (Meiners et a1., 1977) Radish mosaic (RaMV) (Campbell, 1973)

Radish enation mosaic (Campbell and Tochihara, 1969) Red clover mottle (RCMV) (Sinha, 1960; Valenta and Marcinka, 1971)

Pea symptomless virus (Mahmood et a1., 1972) Squash mosaic (SMV) (Freitag, 1956; Campbell, 1971)

Cucurbit ring mosaic (Freitag, 1956) Muskmelon mosaic (Anderson, 1954; Lindberg et a1., 1956) Latent muskmelon virus (Anderson, 1954; Lindberg et a1., 1956) Watermelon stunt (Nelson et a1., 1965; Nelson and Knuhtsen, 1973)

Ullucus virus C (UVe) (Brunt et a1., 1982; Brunt and Jones, 1984)

19

experimental host range based on mechanical transmission. Thus, keys based on host reactions for the identification of comoviruses infecting legumes (Hampton et a1., 1978) are unlikely to be satisfactory.

In general, little importance can be attached to different host reactions of viruses that have many common hosts. Thus, cowpea mosaic virus (CPMV) and cowpea severe mosaic virus (CPSMV) are serologically distinct, occur in distinct geographic areas, and are considered to be separate viruses (Swaans and Van Kammen, 1973), but some isolates of CPSMV produce a faint mottle in cowpea (Vigna unguiculata), others produce severe mosaic and distortion, and yet others produce a bright yellow mottle suggestive of CPMV. At first, it


seemed that the two viruses could be distinguished by their reactions in Chenopodium amaranticolor, in which both produced local lesions but only CPMV was thought to cause systemic infection (Chant, 1962j Van Kammen and De Jager, 1978). Lin et al. (1981a), however, found several isolates of CPSMV that produced systemic symptoms in C. amaranticolor. Similarly, whereas most strains of bean pod mottle virus (BPMV) remain localized in C. quinoa, Moore and Scott (1971) encountered an isolate that produced a bright yellow systemic mottling in this host. Nelson et al. (1965) described an isolate of SMV that caused systemic infection of watermelon, whereas the type isolate does not infect watermelon. Many other examples exist of variation in host reaction among isolates of a comovirus.

2. Cross-Protection

Cross-protection has not been extensively studied among comoviruses. Demski (1969) demonstrated cross-protection between three distinguishable isolates of SMV, but not between SMV and members of three other virus genera: tobacco ringspot nepovirus, cucumber mosaic cucumovirus, and watermelon mosaic 2 potyvirus. Similarly, Mahmood et al. (1972) demonstrated cross-protection between pea symptomless virus and red clover mottle virus (RCMV). However, though the type isolate of BPMV protected Phaseolus vulgaris cv. Black Valentine from infection by a necrotic local lesion isolate, CPSMV, bean rugose mosaic virus (BRMV), and quail pea mosaic virus (QPMV) did so too (T. P. Fulton, unpublished data). Thus crossprotection occurs among different comoviruses as well as among strains of a single comovirus.

In contrast, Valenta et al. (1969) obtained no evidence of protection between broad bean stain virus (BBSV) (pea green mottle strain) and broad bean true mosaic virus (BBTMV), and found that protection between BBSV and RCMV was incomplete. However, because of the nature of their trials and the fact that the hosts used were systemically susceptible to all the viruses, their tests would probably not have detected low levels of protection. Thus, although cross-protection may be of some help in the recognition and distinction of comoviruses, it is unlikely to provide reliable guidance.

3. Vector Relations

Most comoviruses are transmitted by leaf-feeding beetles (see Chapter 4), but vector specificity is low. Probably each comovirus can be transmitted by any beetle species that feeds on the appropriate host.

4. Compatibility between Genome Segments

Both the middle and bottom nucleoprotein components (or their RNA constituents) are required for infection (Van Kammen, 1968j Wood, 1972). It is

COMOVIRUSES: IDENTIFICATION AND DISEASES 21

also possible to obtain infection when middle and bottom components of closely related strains are mixed. Indeed, pseudorecombinants have been obtained by mixing middle and bottom components of closely related but distinguishable isolates of individual comoviruses (Bruening, 1969; De Jager and Van Kammen, 1970; Moore and Scott, 1971). However, not all possible pseudorecombinants are viable. For example, Kassanis et a1. (1973) found that the middle and bottom components of an isolate of RaMV from kale (Brassica oleracea) and a distinguishable isolate from turnip (Brassica rapa) would complement each other but neither would complement the opposite component of the type isolate from radish (Raphanus sativus). In contrast, in those examples studied, pseudorecombinants could not be produced between distinct members of the group (Govier, 1975).

5. Cytopathology

ultrastructural studies with a limited number of comoviruses have indicated that the cytological changes they induce are virtually identical (Kitajima et a1., 1974; Kim and Fulton, 1971, 1972; Van der Sheer and Groenewegen, 1971). The most evident effects are the production of cytoplasmic inclusions consisting of a great number of membranous vesicles and virus particles. Virus particles also occur in tubules situated between the plasmalemma and the cell wall or embedded in cell wall protrusions. The tubules are involved in cell-to-cell movement of virus (Chapter 3). No striking differences in cytopathology were evident between the comoviruses CPMV and CPSMV, except when either of these viruses occurred in mixed infection in bean (Phaseolus vulgaris) with bean yellow mosaic potyvirus (BYMV) (Carr and Kim, 1983); cells infected with both CPSMV and BYMV exhibited intranuclear inclusions not associated with either virus alone or with a mixed infection of CPMV and BYMY. These authors cited this as a cytological marker specific for CPSMV in mixed infection with BYMV and as justification for recognizing the distinction of CPSMV from CPMY. Whether other markers of this type could be found is not known.

6. Serology

All comoviruses are good antigens and high titer antisera are available for all of them. Serology is the most commonly used tool for recognizing and distinguishing members. In fact, serological similarities and differences are the basis for grouping or separating isolates of the viruses. All other characteristics are considered merely supportive.

All comoviruses tested are serologically related to one or more other comoviruses. UVC, for example, reacts weakly with CPMV antiserum, but does not react with antisera to 11 other comoviruses (Brunt et a1., 1982). In contrast, CPSMV reacts weakly with antisera to BPMV, QPMV, BRMV, APMV, RaMV, RCMV; and CPMV (Bruening, 1978).


In spite of the extensive use of serology, quantitative standards for distinguishing members have not been specified. A few studies have employed serologically specific electron microscopy (Rosemeyer et a1., 1981), immunofluorescence (Alvarez and Campbell, 1978), and enzyme-linked immunosorbent assay (ELISA) (Nolan and Campbell, 1984), but most have utilized immunodiffusion tests (Fulton and Scott, 1979; Lin et a1., 1981a, 1984), which, because high-titer antisera are available, provide useful information. Interpretations are based on the patterns of precipitation in the agar (Fulton and Scott, 1979). Isolates of the same virus develop discrete, well-defined, curved bands that coalesce between wells. Small, sharply defined spurs are produced between strains of the same virus. Diffuse, ill-defined areas of precipitation are evident when antigen and antiserum are of two distinct but related comoviruses.

Although useful local lesion hosts are available for some of the comoviruses, serology is usually the technique of choice for virus detection and assay. Some comoviruses are seed-borne in some crops, and seed lots can be assayed by ELISA (Nolan and Campbell, 1984; Franken et a1., 1990), dot immunobinding (Lange et a1., 1989), autoradiography (Powell and Schlegel, 1970), or immunofluorescence (Alvarez and Campbell, 1978). Ghabrial and Schultz (1983) used ELISA to detect viruliferous beetles in field collections. Field-collected plant samples are also readily tested for comoviruses by ELISA or immunodiffusion.

7. Other Methods

Other diagnostic tools such as light microscopy of viral inclusions (Christie and Edwardson, 1986), electrophoretic analysis of viral doublestranded RNA (dsRNA) (Valverde et a1., 1990), molecular hybridization, and polymerase chain reactions (PCR) (Hull, 1993) have not yet been used widely for comovirus diagnosis. However, the availability of viral cDNA clones and sequence data is likely to make the molecular hybridization and PCR methods practical.

8. Final Comments on Differentiation and Grouping of Comoviruses

Some individual comoviruses tend to have a restricted geographical distribution (Bruening, 1978), which is probably related to the occurrence of their natural hosts and beetle vectors. BPMV is apparently restricted to southern and eastern United States. In contrast, CPSMV, a virus that is distantly related to BPMV and has a similar host range, is reported from southern United States, Puerto Rico, Mexico, El Salvador, Costa Rica, Venezuela, Surinam, and Brazil. It probably occurs throughout North and South America within the range of its beetle vectors. Phaseolus beans are widely cultivated in Central America, where BRMV and QPMV are more prevalent


in crops than BPMV. The widespread occurrence of CPSMV is no doubt related to the presence of environmentally adaptable natural hosts. A variety of factors may restrict one virus geographically or allow another to spread over a much greater area of the world. This suggests that the entire biological situation regarding hosts, vectors, and viruses should be considered when defining a distinct comovirus.

II. COMOVIRUSES AND THE DISEASES THEY CAUSE

This section deals with the diseases caused by individual comoviruses, the symptoms they induce in naturally infected plants, their geographic distribution, and strain variation. For more detailed information on experimental and diagnostic hosts, see CMI/AAB Descriptions of Plant Viruses. Vectors and virus-vector relations are reviewed in Chapter 4.

A. Andean Potato Mottle Virus

APMV (Fribourg et a1., 1977, 1979) induces a mild to severe mottling in potato (Solanum spp.), depending on the cultivar. Severely affected cultivars may exhibit stunting and deformation as well as top necrosis. The virus occurs naturally only in the Andean region of South America including Chile, Bolivia, Colombia, and Peru (Avila et a1., 1984). Only solanaceous plants are susceptible except as indicated below. Chlorotic spotting, veinclearing, and mosaic are typical of infected plants such as Lycopersicon chilense, Nicandra physaloides, Nicotiana bigelovii, and Nicotiana c1evelandii. Local lesion hosts have not been reported for this virus. The virus is readily transmitted by mechanical means and in the field probably is disseminated by handling foliage or by plant-to-plant contact. APMV is transmitted by the beetle Diabrotica viridula, and other species of Diabrotica are suspected vectors (Avila et a1., 1984). Salazar and Harrison (1978) recognized three strains of APMV based on host reactions. The type and C strains were serologically indistinguishable, whereas strain H gave a spur when compared with these two in agar double-diffusion tests. Strains C and H but not the type strain infected Gomphrena globosa systemically but without inducing symptoms. Avila et a1. (1984) isolated a fourth strain, designated strain B, from potatoes in Brazil. This strain produced severe mosaic and crinkle in potato. In other solanaceous hosts, symptoms were more severe than those produced by the other strains. Recently, Valverde and Black (1993) reported a comovirus from tabasco pepper (Capsicum frutescens) in Central America. This virus was transmitted by D. balteata, had a host range restricted to the Solanaceae, and gave a strong reaction with antiserum to APMV (R. A. Valverde, unpublished data); it may be a distinct strain of APMV.


B. Bean Pod Mottle Virus

BPMV (Semancik, 1972) is widespread in soybean in the southern and eastern United States. The virus was originally described from bean (Phaseolus vulgaris) (Zaumeyer and Thomas, 1948), although it is not as much of a threat to beans as to soybeans (Glycine max). In soybean, BPMV causes mild-tosevere mottling of leaves; severe strains cause puckering and leaf distortion. Mottling of pods and seeds may also occur. The disease causes a modest reduction in the yield of soybeans (Hopkins and Mueller, 1984; Myhre et a1., 1973). When BPMV occurs as a mixed infection with soybean mosaic potyvirus, however, the yield loss is much greater (Ross, 1968) and symptoms are more severe (Anjos and Ghabrial, 1991). In Phaseolus bean, BPMV induces a mild-to-severe mottle; pods are often severely malformed and develop a severe mottle characterized by large blotches of dark green tissue. Wild hosts include the legumes Lespedeza sp., Stizolobium deeringianum, Desmodium paniculatum, and Trifolium incarnatum. The host range of BPMV is largely confined to the Leguminosae. The bean cultivar Pinto is the local lesion host of choice. The virus is highly antigenic and most isolates are indistinguishable in gel double-diffusion serological tests, though, when extensive collections of isolates from soybeans were tested, a few gave a spur with type antigen (Moore and Scott, 1981). No major symptom variants are described.

C. Bean Rugose Mosaic Virus

In susceptible Phaseolus bean varieties, BRMV causes a severe mosaic with leaf distortion and stunting (Gamez, 1972, 1982). Pods are severely malformed. BRMV is largely confined to the Leguminosae. Phaseolus vulgaris cv. Top Crop is a good local lesion host (Gamez, 1982). The virus is reported from several areas of Central and South America. Isolates may differ in host range and symptom severity. A very severe strain from El Salvador is termed "ampollado" (Galvez et a1., 1977), but most bean cultivars grown in Central America are resistant, and this strain does not present a serious threat to crops. In gel double-diffusion tests with BPMV antiserum, the ampollado strain produces a diffuse straight band of precipitate in contrast with the sharply defined curved band caused by the type isolate. The virus is readily transmitted by leaf-feeding beetles. Lin et a1. (1981b) described a strain of BRMV from Brazil called "mosaico-em-desenho," which forms a spur with type BRMV in agar double-diffusion tests.

D. Broad Bean Stain Virus

BBSV gets its name from the necrotic staining it causes in the seed coat of broad bean (Vicia faba) (Gibbs et a1., 1968; Gibbs and Smith, 1970). Leaves show a bright chlorotic mottling with some distortion. The host range of


BBSV is restricted to the Leguminosae. Phaseolus vulgaris cv. Tendergreen is a local lesion host (Gibbs and Smith, 1970). Various strains of this virus occur commonly in Europe, North Africa, and Australia (Moghal and Francki, 1974). Other strains of this virus have been isolated from pea (Pea sativum), including pea green mottle virus, which produces very mild symptoms in pea and broad bean, "mosaique de Ie feve," which produces somewhat more severe symptoms in pea and broad bean, and pea seed-borne symptomless virus. Valenta and Gressnerova (1966) and Valenta et a1. (1969) indicated that these three strains are closely related serologically. Serological comparisons by Musil et a1. (1983) confirmed a close relationship between pea green mottle virus, pea seed-borne symptomless virus, and BBSV. In Europe, not only BBSV but also two other comoviruses, BBTMV and RCMV, are common in broad bean, pea, and/or clover (Trifolium spp.). Isolates of all three viruses differ in host range, severity of symptoms, and serology. Host reactions are therefore not reliable for distinguishing BBSV from BBTMV or RCMV. Serological tests are essential, preferably with antisera to all three viruses.

E. Broad Bean True Mosaic Virus

BBTMV, also known as Echtes Ackerbohnenmosaik-Virus (Quantz, 1953; Gibbs and Paul, 1970) is reported from Europe and North Africa. It causes a severe mosaic in broad bean and pea. Shoot necrosis occurs in cool conditions. Some varieties of Phaseolus bean can be infected experimentally. The reported host range for this virus is confined to legumes. No local lesion host is known. BBTMV is serologically related to CPMV (Jones and Barker, 1976). Isolates of BBTMV from Europe and North Africa are serologically indistinguishable (Gibbs et a1., 1968; Jones and Barker, 1976).

F. Cowpea Mosaic Virus

CPMV, the type member of the comovirus group, typically causes a bright yellow mosaic in cowpea with some distortion and stunting (Van Kammen and De Jager, 1978). The host range is rather limited; few hosts are known outside the Leguminosae. Phaseolus vulgaris cv. Pinto and C. quinoa are suitable local lesion hosts. CPMV particles reach a high titer in infected plants, are stable, easily purified and highly antigenic. CPMV has been the subject of detailed molecular biological research (see Chapter 3). Two comoviruses from cowpea, CPMV and CPSMV, were assumed to be strains of the same virus until Swaans and Van Kammen (1973) showed that they are distinct comoviruses. CPMV is reported from Nigeria (Chant, 1959), Kenya (Bock, 1971), Philippines (Talens, 1979), and Iran (Kaiser et a1.; 1968). Although there are reports of its occurrence in North and South America


(Agrawal, 1964; McLaughlin et al., 1977; Kvicala et al., 1970), we doubt their reliability. Thus, although Kvicala et al. (1970) reported the occurrence of CPMV in Cuba, they provided no serological evidence to indicate whether the virus was CPMV or another comovirus. The report of CPMV in the United States by McLaughlin et al. (1977) refers to a virus infecting an experimental plant on which a field-collected beetle had fed. However, CPMV was being handled in the same greenhouse and the virus was never recovered from field-collected plants. The Sb isolate of CPMV reported from Surinam by Agrawal (1964) may have been introduced in seed from Africa without subsequently becoming established in the environment. In recent years we have tested many samples of virus-infected cowpea from North and South America but none contained CPMV. We also supplied antisera to collaborators in the United States, Mexico, Guatemala, El Salvador, Panama, Costa Rica, Colombia, Venezuela, and Brazil, of whom many recorded CPSMV but none detected CPMV.

The apparent absence of CPMV from North and South America is surprising because there has been ample opportunity for its introduction in seed from Africa. Moreover, efficient beetle vectors are present in America, although they differ from those in Africa.

G. Cowpea Severe Mosaic Virus

As indicated above, CPSMV occurs commonly in cowpea in southern United States, Caribbean islands, Central America, and northern South America. The virus is apparently well established and endemic throughout its range, and it no doubt also has wild hosts. For example, Phaseolus lathyroides and Vigna vexillata are widespread weeds in Central and South America that are commonly infected with CPSMV (Alconero and Santiago, 1973; Lima and Nelson, 1977; Valverde et al., 1982a). The symptoms caused by CPSMV (De Jager, 1979) in cowpea range from mild mottling to severe mosaic, stunting, and leaf distortion depending on the isolate of the virus. Only leguminous plants are reported as natural hosts. Chenopodium amaranticolor and Phaseolus vulgaris cv. Pinto may be used for local lesion assay. In addition to differences in symptom severity, isolates also exhibit minor antigenic variations. Isolates from different geographic regions typically can be distinguished by the production of a small spur in agar doublediffusion tests. For example, an isolate from southern United States can be distinguished in this way from one from Mexico (Rocha and Fulton, 1985). Isolates from El Salvador, Costa Rica, Venezuela, and Brazil can also be distinguished from one another. Lin et al. (1981a, 1984) identified four serotypes. Throughout warm temperate and tropical America, field incidence of CPSMV may reach 100% and such severe infestations can result in 50% reduction in fresh plant weight and in number and weight of pods (Debrot and Benitez de Rojas, 1967) and up to 80% loss in seed yield (Lima and Nelson, 1977; Valverde et al., 1982b).


A severe disease caused by CPSMV in soybean is reported from Puerto Rico (Thongmeearkom and Goodman, 1976; Thongmeearkom et a1., 1978) and Brazil (Anjos and Lin, 1984). The virus is inefficiently transmitted by beetle vectors to soybean and the disease is reported only near cowpea plantings with a high incidence of CPSMV. Although infected soybean plants are severely stunted, the limited incidence of CPSMV in soybean growing near infected cowpea plantings suggests that it is not a serious threat to soybean crops.

H. Glycine Mosaic Virus

GMV (Bowyer et a1., 1980) was isolated from naturally infected wild Glycine clandestina and G. tabacina plants with light and dark green mosaic and deformation of the leaf margins at several locations in Australia. Although never found naturally in cultivated plants, GMV infects soybean (G. max), producing vein yellowing and mosaic. Many Phaseolus vulgaris cultivars too can be infected, with local lesions developing in some and systemic mottle in others. A mild mottle is produced in pea. The only nonlegume hosts are Chenopodium amaranticolor and C. quinoa, both of which develop local lesions, severe systemic vein yellowing, and leaf distortion. The type isolate of GMV from New South Wales reacted weakly with antisera to BBTMV, BBSV, and SMV. In addition, the GW strain from Queensland reacted weakly with antisera to CPMV, CPSMV, BPMV, and RCMV. The type isolate and the GW strain could be distinguished from each other by formation of a small spur in agar double-diffusion serological tests. As this virus has never been reported from cultivated plants, it has no known economic importance.

I. Pea Mild Mosaic Virus

PMiMV has been isolated only once, from pea seed in New Zealand (Clark, 1972). Mechanically inoculated pea seedlings develop systemic vein clearing followed by vein necrosis and downward curling of the leaves and later by a mild mosaic. PMiMV is reported to infect only legume species. Limited serological studies suggest that PMiMV is distinct from other comoviruses: it is antigenically related to BBSV and CPSMV but not to BPMV, CPMV, or RCMV. Because of the limited nature of these tests and the single report of its occurrence, PMiMV can be only tentatively distinguished as an individual member of the comovirus group.

J. Quail Pea Mosaic Virus

In bean and soybean, QPMV produces a mosaic indistinguishable from that produced by BPMV. The virus was first obtained from a wild legume

28 R. A. VALVERDE AND ,. P. FULTON

(quail pea, Strophostyles helvola) (Moore, 1973), and later occasionally from soybean growing in the same area of Arkansas (Moore and Scott, 1981). QPMV infects plants mainly in the family Leguminosae, although some species in the Chenopodiaceae and Convolvulaceae can be infected experimentally. Phaseolus vulgaris cv. Pinto is satisfactory for local lesion assays of the type strain. A virus called bean curly dwarf mosaic in El Salvador (Meiners et al., 1977) and Costa Rica (Hobbs, 1981) is considered a strain of QPMV because it is closely related serologically.

K. Radish Mosaic Virus

RaMV produces a variety of symptoms, including leaf distortion, mosaic, ringspot, and veinal necrosis in many crucifers (Campbell, 1964, 1973). Hosts in the Solanaceae, Chenopodiaceae, and Cucurbitaceae give local lesion reactions. Chenopodium amaranticolor is a satisfactory local lesion host. RaMV is economically important in turnip and radish, and is reported from the United States (California), Japan, Europe, and North Africa, (Campbell and Tochihara, 1969; Koenig and Fischer, 1981). Isolates from California and Japan are serologically indistinguishable from each other (Campbell and Tochihara, 1969) but are distinguishable from the European isolate in gel double-diffusion tests (Stefanac and Mamula, 1972). RaMV is distantly serologically related to CPMV, BPMV, and SMV.

1. Red Clover Mottle Virus

Although RCMV (Valenta and Marcinka, 1971) can infect many legumes, it is found mainly in red clover (Trifolium pratense) (Sinha, 1960). Systemic vein clearing in this host may be followed by a general chlorosis, mottling, chlorotic spots and rings, and some leaf distortion. Few species outside the Leguminosae are susceptible, except Chenopodium amaranticolor, C. quinoa, and Gomphrena globosa, in which local lesions are produced. RCMV is confined to most parts of Europe except the south. Isolates of RCMV from northern Europe may differ in host range, symptoms, and serological specificity (Bos and Maat, 1965; Lapchic et al., 1975; Gerhardson and Lindsten, 1973). For example, Mahmood et al. (1972) described an isolate of RCMV from pea that infected pea, red clover, and broad bean symptomlessly and named it the pea symptomless strain. Like the type strain, this strain induced local lesions in G. globosa, C. amaranticolor, and C. quinoa, and the two were closely related serologically.

Most isolates of RCMV react weakly with CPMV and BPMV antisera and more strongly with BBSV antiserum. There is some question of whether all isolates of RCMV and BBSV should be considered together as a single serogroup (Mahmood et al., 1972).


M. Squash Mosaic Virus

SMV causes severe diseases in a variety of cucurbits including squash (Cucurbita spp.), cantaloupe (Cucumis melo), and cucumber (Cucumis sativus) (Freitag, 1956). Symptoms include bright yellow vein clearing, mosaic, and leaf distortions and deformities. In squash and cucumber, fruits are severely deformed. The natural host range of SMV is limited to the Cucurbitaceae and Chenopodiaceae (Campbell, 1971 j Lockhart et al., 1982h however, it also infects plants in five other families. Early Yellow Summer Crookneck squash is reported to give chlorotic local lesions after mechanical inoculation. The virus has been reported from California and several areas of southern United States, Israel (Campbell, 1971), Morocco (Lockhart et al., 1982), and Japan (Yoshida et al., 1980). Because the virus is commonly seedborne, it probably occurs worldwide, wherever susceptible cucurbits are grown. Early studies indicated that SMV did not infect watermelon, but Nelson et al. (1965) isolated strains that infect this host. Nelson and Knuhtsen (1973), who compared isolates from different locations in the United States, recognized six biotypes based on host reactions. However, only two strains could be distinguished by spur formation in agar double-diffusion serological tests.

N. Ullucus Virus C

UVC (Brunt et al., 1982 j Brunt and Jones, 1984) is commonly present in Ullucus tuberosus (Basellaceae), a vegetatively propagated food crop grown in the Andean region of South America. U. tuberosus plants infected with UVC are symptomless. When found in nature in plants showing severe symptoms, UVC is usually associated with one or two other viruses. The experimental host range is restricted. In addition to U. tuberosus, only Chenopodium amaranticolor, C. quinoa, C. murale, and Tetragonia expansa could be infected. Although clearly a member of the comovirus group, and distantly related only to CPMV, it did not react with antisera to 11 other comoviruses (Brunt and Jones, 1984).

REFERENCES

Agrawal, H., 1964, Identification of cowpea mosaic virus isolates, Meded. LandbHoogesch, Wageningen 64:1.

Aiconero, R., and Santiago, A., 1973, Phaseolus lathyroides as a reservoir of cowpea mosaic virus in Puerto Rico, Phytopathology: 63:120.

Alvarez, M., and Campbell, R. N., 1978, Transmission and distribution of squash mosaic virus in seeds of cantaloupe, Phytopathology 68:257.

Anderson, C. W., 1954, Two muskmelon mosaic virus strains from Central Florida, Phytopathology 44:371.


Anjos, J. R. N., and Ghabrial, S. A., 1991, Studies on the synergistic interactions between soybean mosaic virus (SMV) and two comoviruses in mixed infections in soybeans, Phytopathology 81:1167.

Anjos, J. R. N., and Lin, M. T., 1984, Bud blight of soybeans caused by cowpea severe mosaic virus in central Brazil, Plant Dis. 68:405.

Avila, A. c., Salazar, 1. F., Ortega, M., and Daniels, J., 1984, A new strain of Andean potato mottle virus from Brazil, Plant Dis. 68:997.

Bock, K. R., 1971, Notes on East African plant virus diseases. 1. Cowpea mosaic virus, E. Afr. Agric. For. r 37:60.

Bos, 1., and Maat, D. Z., 1965, A distinctive strain of the red clover mottle virus in The Netherlands, Neth. r Plant Pathol. 71:8.

Bowyer, J. W., Dale, J. 1., and Behncken, G. M., 1980, Glycine mosaic virus: A comovirus from Australian native Glycine species, Ann. Appl. Biol. 95:385.

Bruening, G., 1969, The inheritance of top component formation in cowpea mosaic virus, Virology 37:577.

Bruening, G., 1978, Comovirus group, CMI/AAB Descriptions of Plant Viruses No. 199. Brunt, A. A., and Jones, R. A. c., 1984, Ullucus virus C, CMI/AAB Descriptions of Plant Viruses

No. 277. Brunt, A. A., Barton, R. J., Phillips, S., and Jones, R. A. C., 1982, Ullucus virus C, a newly

recognized comovirus infecting Ullucus tuberosus (Basellaceae), Ann. Appl. Biol. 101:73. Campbell, R. N., 1964, Radish mosaic virus, a crucifer virus serologically related to strains of

bean pod mottle virus and to squash mosaic virus, Phytopathology 54:1418. Campbell, R. N., 1971, Squash mosaic virus, CMI/AAB Descriptions of Plant Viruses No. 43. Campbell, R. N., 1973, Radish mosaic virus, CMI/AAB Descriptions of Plant Viruses No. 121. Campbell, R. N., and Tochihara, H., 1969, Radish enation mosaic virus is a strain of radish

mosaic virus, Phytopathology 59:1756. Carr, R. J., and Kim, K. S., 1983, Ultrastructure of mixed plant virus infection: Bean yellow

mosaic virus with cowpea severe mosaic virus or cowpea mosaic virus in bean, Virology 124:338.

Chant, S. R., 1959, Viruses of cowpea, Vigna unguiculata 1. (Walp.), in Nigeria, Ann. Appl. Biol. 47:565.

Chant, S. R., 1962, Further studies on the host range and properties of Trinidad cowpea mosaic virus, Ann. Appl. Biol. 50:159.

Christie, R. G., and Edwardson, J. R., 1986, Light microscopic techniques for detection of plant virus inclusions, Plant Dis. 70:273.

Clark, M. F., 1972, Purification and some properties of a comovirus group virus isolated in New Zealand, N. Z. r Agric. Res. 15:846.

Debrot, E., and Benitez de Rojas, C. E., 1967, El virus del mosaico del frijol, Vigna sinensis Endl. (cowpea mosaic virus) en Venezuela, Agronomia Trop. 17:3.

De Jager, c. P., 1979, Cowpea severe mosaic virus, CMI/AAB Descriptions of Plant Viruses No. 209.

De Jager, c. P., and Van Kammen, A., 1970, The relationship between the components of cowpea mosaic virus. III. Location of genetic information for two biological functions in the middle component of CPMV, Virology 41:281.

Demski, J. W., 1969, Local reaction and cross protection for strains of squash mosaic virus, Phytopathology 59:251.

Devergne, J. c., and Cousin, R., 1966, Le virus de la mosalque de la feve (MF) et les symptomes d'ornamentation sur graines, Ann. Epiphyties 17(numero hors scrie):147.

Francki, R. I. B., Fauquet, c., Knudson, D. 1., and Brown, 1. (eds.), 1991, Classification and Nomenclature of Viruses. Fifth Report of the International Committee on Taxonomy of Viruses, Springer-Verlag, Vienna. (Also in Arch. Virol., Suppl. 2.)

Franken, A. A., Maat, D. Z., and Kamminga, G. c., 1990, Detection of squash mosaic virus in seed of melon (Cucumis melo) by enzyme-linked immunosorbent assay (ELISA), Neth. r Plant Pathol. 96:91.


Freitag, J. H., 1956, Beetle transmission, host range, and properties of squash mosaic virus, Phytopathology 46:73.

Fribourg, C. E., Jones, R. A. C., and Koenig, R., 1977, Andean potato mottle, a new member of the cowpea mosaic virus group, Phytopathology 67:969.

Fribourg, C. E., Jones, R. A. C., and Koenig, R., 1979, Andean potato mottle virus, CMII AAB Descriptions of Plant Viruses No. 203.

Fulton, J. P., and Scott, H. A., 1979, A serogrouping concept for legume comoviruses, Phytopathology 69:305.

Galvez, G. E., Cardenas, M., Kitajima, E. W., Diaz, J. A., and Nieto, M. P., 1977, Purification, serology, electron microscopy and properties of the ampollado strain of bean rugose mosaic virus, Turrialba 27:343.

Gamez, R., 1972, Los virus del frijol en Central America. ll. Algunas propiedades y transmission por crisomelidos del virus del mosaico rugoso del frijol, Turri alba 22:249.

Gamez, R., 1982, Bean rugose mosaic virus, CMI/AAB Descriptions of Plant Viruses No. 246.

Gerhardson, B., and Lindsten, K., 1973, Red clover mottle virus and red clover necrotic mosaic virus in Sweden, Phytopathol. Z. 76:69.

Ghabrial, S. A., and Schultz, E J., 1983, Serological detection of bean pod mottle virus in bean leaf beetles, Ceratoma trifurcata, Phytopathology 73:480.

Gibbs, A. J., and Paul, H. 1., 1970, Echtes Ackerbohnenmosaik-Virus, CMI/AAB Descriptions of Plant Viruses No. 24.

Gibbs, A. J., and Smith, H. G., 1970, Broad bean stain virus, CMI/AAB Descriptions of Plant Viruses No. 29.

Gibbs, A. J., Giussani-Belli, G., and Smith, H. G., 1968, Broad bean stain and true broad bean mosaic viruses, Ann. Appl. Biol. 61:99.

Govier, D. A., 1975, Complementation between middle and bottom components of broad bean stain virus and Echtes Ackerbohnenmosaik-Virus, J. Gen. Viral. 28:373.

Hampton, R., Beczner, 1., Hagedorn, D., Bos, 1., Inouye, T., Barnett, 0., Musil, M., and Meiners, J., 1978, Host reactions of mechanically transmissible legume viruses of the northern temperate zone, Phytopathology 68:989.

Hobbs, H. A., 1981, Transmission of bean curly dwarf mosaic virus and bean mild mosaic virus by beetles in Costa Rica, Plant Dis. 65:491.

Hopkins, J. D., and Mueller, A. J., 1984, Effect of bean pod mottle virus on soybean, Glycine max, yield, J. Bcon. Bntomol. 77:943.

Hull, R., 1993, Nucleic acid hybridization procedures, in: Diagnosis of Plant Virus Diseases (R. E. E Matthews, ed.), pp. 253-271, CRC Press, Boca Raton, F1.

Hull, R., Milne, R. G., and Van Regenmortel, M. H. v., 1991, A list of proposed standard acronyms for plant viruses and viroids, Arch. Viral. 120:151.

Jones, A. T., and Barker, H., 1976, Properties and relationships of broad bean stain virus and Echtes Ackerbohnenmosaik-Virus, Ann. Appl. Biol. 83:231.

Kaiser, J. W., Danesh, D., Okhovat, M., and Mossamebi, H., 1968, Regional pulse improvement project. Diseases of pulse crops (edible legumes) occurring in Iran, Iran J. PI. PathoI. 4:2.

Kassanis, B., White, R. E, and Woods, R. D., 1973, Genetic complementation between middle and bottom components of two strains of radish mosaic virus, J. Gen. Viral. 20:277.

Kim, K. S., and Fulton, J. P., 1971, Tubules with virus-like particles in leaf cells infected with bean pod mottle virus, Virology 43:329.

Kim, K. S., and Fulton, J. P., 1972, Fine structure of plant cells infected with bean pod mottle virus, Virology 49:112.

Kitajima, E. W., Tascon, A., Gamez, R., and Galvez, G. E., 1974, Ultrastructural studies on bean leaf tissues infected with two strains of bean rugose mosaic virus, Turrialba 24:393.

Koenig, R., and Fischer, H. U., 1981, A Moroccan radish mosaic virus isolate from turnip, Brassica rapa var. rapa, Plant Dis. 65:758.

Kowalska, C., and Beczner, 1., 1980, Characterization of a seed-borne virus in Pisum sativum, Tag.-Ber. Akad. Landwirtsch. Wiss. DDR 184:353.


Kvicala, B. A., Smrz, J., and Blanco, N., 1970, Some properties of cowpea mosaic virus isolated in Cuba, Phytopathol. Z. 69:233.

Lange, 1., Jomantor, A., and Heide, M., 1989, Testing seeds for viruses by dot immunobinding (DBII directly on plain paper, Tidsskr. Planteavl 93:93.

Lapchic, 1. G., Kuznetzova, 1. 1., Melniczenko, V. S., Vocelko, S. K., and Sjedin, A. A., 1975, Red clover mottle virus in Ukraine, Phytopathol. Z. 82:339.

Lima, J. A. A., and Nelson, M. R., 1977, Etiology and epidemiology of mosaic of cowpea in Ceara, Brazil, Plant Dis. Rep. 61:864.

Lin, M. T., Anjos, J. R. M., and Rios, G. P., 1981a, Serological grouping of cowpea severe mosaic isolates from Central Brazil, Phytopathology 71:435.

Lin, M. T., Gamez, R., and Kitajima, E. W., 1981b, Bean "mosaico-em-desenho" virus is a member of the bean rugose mosaic virus serogroup, Fitopatol. Brasil. 6:293.

Lin, M. T., Hill, J. H., Kitajima, E. W., and Costa, C. 1.,1984, Two new serotypes of cowpea severe mosaic virus, Phytopathology 74:58l.

Lindberg, G. D., Hall, D. H., and Walker, J. c., 1956, A study of melon and squash mosaic viruses, Phytopathology 46:489.

Lockhart, B. E. 1., Ferji, Z., and Hafidi, B., 1982, Squash mosaic virus in Morocco, Plant Dis. 66:119l.

Mahmood, K., Horsten, J., and Peters, D., 1972, Pea symptomless virus, a newly recognized strain of red clover mottle virus, Neth. f. Plant Pathol. 78:204.

McLaughlin, M. R., Thongmeearkom, P., Milbrath, G. M., and Goodman, R. M., 1977, Isolation and some properties of a yellow subgroup member of cowpea mosaic virus from Illinois, Phytopathology 67:844.

Meiners, J. P., Waterworth, H. E., Lawson, R. H., and Smith, F. E., 1977, Curly dwarf mosaic disease of beans from El Salvador, Phytopathology 167:63.

Moghal, S. M., and Francki, R.1. B., 1974, Occurrence and properties of broad bean stain virus in South Australia, Aust. f. Biol. Sci. 27:34l.

Moore, B. J., 1973, Quail pea mosaic virus: A new member of the comovirus group, Plant Dis. Rep. 57:31l.

Moore, B. J., and Scott, H. A., 1971, Properties of a strain of bean pod mottle virus, Phytopathology 61:83l.

Moore, B. J., and Scott, H. A., 1981, Quail pea mosaic virus, CMI/AAB Descriptions of Plant Viruses No. 238.

Musil, M., Valenta, v., Kowalska, c., Weatroszak, I., and Beczner, 1., 1983, Serological properties of some comoviruses, Biologia (Bratisl.I38:23l.

Myhre, D. 1., Pitre, H. N., Haridasan, M., and Hesketh, J. D., 1973, Effect of bean pod mottle virus on yield components and morphology of soybeans in relation to soil water regimes: A preliminary study, Plant Dis. Rep. 57:1050.

Nelson, M. R., and Knuhtsen, H. K., 1973, Squash mosaic virus variability: Review and serological comparisons of six biotypes, Phytopathology 63:920.

Nelson, M. R., Matejka, J. c., and McDonald, H. H., 1965, Systemic infection of watermelon by a strain of squash mosaic virus, Phytopathology 55:1362.

Nolan, P. A., and Campbell, R. N., 1984, Squash mosaic virus detection in individual seeds and seed lots of cucurbits by enzyme-linked immunosorbent assay, Plant Dis. 68:97l.

Powell, C. C., and Schlegel, D. E., 1970, The histological localization of squash mosaic virus in cantaloupe seedlings, Virology 42:123.

Quantz, 1., 1953, Untersuchungen liber ein samenlibertragbares Mosaikvirus der Ackerbohne (Vicia iabal, Phytopathol. Z. 20:42l.

Rocha, M. A., and Fulton, J. P., 1985, Some properties of a cowpea severe mosaic isolate from Tabasco, Mexico, Turrialba 34:237.

Rosemeyer, M. E., Bemis, W. P., Nelson, M. R., and Wheeler, R. E., 1981, Transmission electron microscopic serology of squash mosaic virus, Phytopathology 71:252.

Ross, J. P., 1968, Effect of single and double infections of soybean mosaic and bean pod mottle viruses on soybean yield and seed characters, Plant Dis. Rep. 52:344.


Salazar, L. F., and Harrison, B. D., 1978, Particle properties and strains of Andean potato mottle virus, f. Gen. Virol. 39:171.

Semancik, J. S. 1972, Bean pod mottle virus, CMI/AAB Descriptions of Plant Viruses No. 108. Shepherd, R. J., 1964, Properties of a mosaic virus of cowpea and its relationship to the bean pod

mottle virus, Phytopathology 54:466. Sinha, R. C., 1960, Red clover mottle virus, Ann. Appl. Biol. 48:742. Stace-Smith, R., 1981, Comoviruses, in: Handbook of Plant Virus Infections. Comparative

Diagnosis (E. Kurstak, ed.), pp. 170-195, Plenum Press, New York. Stefanac, Z., and Mamula, D., 1972, A strain of radish mosaic occurring in turnip in Yugoslavia,

Ann. Appl. Biol. 69:229. Swaans, H., and Van Kammen, A., 1973, Reconsideration of the distinction between the severe

and yellow strains of cowpea mosaic virus, Neth. f. Plant Pathol. 79:257. Talens, L. T., 1979, Cowpea viruses in the Philippines.!. Identity of a mosaic causing virus in

cowpea Vigna unguiculata cultivar Red, Philipp. f. Crop. Sci. 4:37. Thongmeearkom, P., and Goodman, R. M., 1976, A severe disease of soybeans caused by an

isolate of cowpea mosaic virus, Proc. Am. Phytopathol. Soc. 3:209. Thongmeearkom, P., Paschal, P., and Goodman, R. M., 1978, Yield reduction in soybeans

infected with cowpea mosaic virus, Phytopathology 68:1549. Valenta, v., and Gressnerova, M., 1966, Serological relationships among members of the cowpea

mosaic virus group, Acta Virol. 10:182. Valenta, v., and Marcinka, K., 1971, CMI/AAB Descriptions of Plant Viruses No. 74. Valenta, V., Gressnerova, M., Marcinka, K., and Nermut, M. V., 1969, Some properties of pea

green mottle virus, a member of the cowpea mosaic group, isolated in Czechoslovakia, Acta Virol. 13:422.

Valverde, R. A., and Black L. L., 1993, Properties of an isometric virus from Capsicum frutescens, Phytopathology 83:1372.

Valverde, R. A., Moreno, R., and Gamez, R., 1982a, Incidence and some ecological aspects of cowpea severe mosaic virus in two cropping systems in Costa Rica, Turrialba 32:29.

Valverde, R. A., Moreno, R., and Gamez, R., 1982b, Yield reductions in cowpea (Vigna unguiculata) infected with cowpea severe mosaic virus in Costa Rica, Thrrialba 32:89.

Valverde, R. A., Nameth, S. T., and Jordan, R. L., 1990, Analysis of double-stranded RNA for plant virus diagnosis, Plant Dis. 74:255.

Van der Sheer, C., and Groenewegen, J., 1971, Structure in cells of Vigna unguiculata infected with cowpea mosaic virus, Virology 46:493.

Van Kammen, A., 1968, The relationship between the components of cowpea mosaic virus.!. Two nucleoprotein particles necessary for the infectivity of CPMV, Virology 32:312.

Van Kammen, A., and De Jager, C. P., 1978, Cowpea mosaic virus, CMI/AAB Descriptions of Plant Viruses No. 197.

Wood, H. A., -1972, Genetic complementation between the two nucleoprotein components of cowpea mosaic virus, Virology:49:592.

Yoshida, K., Goto, T., Nemoto, M., and Tsuchizaki, T., 1980, Squash mosaic virus isolated from melon (Cucumis melo 1.) in Hokkaido, Ann. Phytopathol. Soc. Tpn. 46:349.

Zaumeyer, W. J., and Thomas, H. R., 1948, Pod mottle, a virus disease of beans, T. Agric. Res. 77:81.

CHAPTER 3

Comoviruses: Molecular Biology and Replication R. W. GOLDBACH AND J. WELLINK

I. INTRODUCTION

Presently, the genus Comovirus comprises 14 definitive species (Table I), which are characterized by having a genome composed of two molecules of single-stranded RNA, each separately encapsidated in icosahedral shells built from two different protein species. The type species, cowpea mosaic virus (CPMV), is the most thoroughly studied with respect to genome structure, replication, translation strategy, and virus-host interactions. As a consequence, most of this chapter will be dedicated to this virus. Indeed, CPMV has lent itself well for molecular analyses because it is conveniently propagated in its natural host, Vigna unguiculata, giving yields as high as 1-2 gJkg of infected leaf material. Moreover, CPMV produces non structural proteins in considerable quantities, a property that has greatly facilitated the investigation of its translational expression. Furthermore, the availability of an efficient host protoplast system, full-length cDNA clones for the production of desired mutants, and specific antibodies against all viral gene products have allowed in-depth analyses of crucial processes during the infection cycle. Therefore, despite the limited number of laboratories in which it has been used as a subject for molecular studies, CPMV is at present among the

R. W. GOLDBACH • Department of Virology, Agricultural University, 6709 PD Wageningen, The Netherlands. J. WELLINK • Department of Molecular Biology, Agricultural University, 6709 PD Wageningen, The Netherlands.

3S

36 R. W. GOLDBACH AND J. WELLINK

TABLE 1. Comovirus Species and Reported Genome Sizes

RNA sizes (in nucleotides)b

Speciesa Acronym RNA-l EMBU RNA-2 EMBU

Cowpea mosaic virus (type species) (CPMV) 5889 X00206 3481 XOO729 Bean pod mottle (BPMV) 3662 M62738 Cowpea severe mosaic (CPSMV) 5957 M83830 3732 M83309 Red clover mottle (RCMV) 6033 X64886 3543 M14913

aNo information is available for the following comovirus species: Andean potato mottle IAPMV), bean rugose mosaic IBRMV), broad bean stain IBBSV), broad bean true mosaic IBBTMV), glycine mosaic IGMV), pea mild mosaic IPMiMV), quail pea mosaic IQPMV), radish mosaic IRaMV), squash mosaic ISMV), or ullucus C IUVe).

bExclusive of polylA) tract. CEMBL database accession number.

best-characterized plant viruses. This chapter will summarize and discuss how studies on capsid structure, genomic organization, translational strategy, replication, and cell-to-cell movement have contributed to our present knowledge of the molecular biology of comoviruses.

II. COMPOSITION OF VIRUS COMPONENTS

A. General Description

Comoviruses have small, icosahedral particles with a diameter of approximately 28 nrn. Purified preparations consist of two, sometimes three, distinguishable types of particle, which are separated in velocity or buoyant density gradients (Rice et a1., 1965; Mazzone et a1., 1962; Bruening and Agrawal, 1967). These types of particle are designated as bottom (B), middle (M), and top (T) component, corresponding to their place in the centrifuge tube. The Band M components are nucleoprotein particles, each containing a segment of the single-stranded, bipartite RNA (denoted RNA-1 and RNA-2, respectively), whereas T component consists of empty protein shells (Fig. 1).

The protein capsids of B, M, and T components are similar, consisting of 60 copies of each of two different capsid proteins (Wu and Bruening, 1971), and the observed differences in sedimentation coefficient and density are exclusively due to differences in RNA content. Sedimentation coefficients reported for these components are 111-127 S (B), 91-100 S (M), and 51-60 S (T) (Mazzone et a1., 1962; Bruening and Agrawal, 1967; Van Kammen, 1967; Bruening, 1977, 1978; Brunt et a1., 1984).

Both Band M components, or their RNA species, are necessary for infectivity (Van Kammen, 1968; De Jager, 1976), which indicates that the genetic information essential for virus multiplication is distributed between the two genome segments. RNA-1 appears to code for all proteins needed for

COMOVIRUSES: MOLECULAR BIOLOGY 37

B M T

~ VPg

~ VPg

An An • • RNA-1 RNA-2

T

infection

FIGURE 1. Composition of comovirus particles. Preparations of comovirus particles consist of three different centrifugal components, denoted bottom (B), middle (M), and top (T) components. These components have identical icosahedral protein capsids with a diameter of 28 nm. Band M components each contain a segment of the bipartite RNA genome. T component consists of empty protein shells. Both Band M components, or their RNA molecules, are required for infectivity. The RNA molecules each possess a VPg molecule (_) and a 3' poly(A) tail.

viral RNA replication, whereas RNA-2 codes for the two capsid proteins as well as proteins responsible for cell-to-cell movement. As a consequence, a certain amount of independence can be attributed to the B component (Goldbach et a1., 1980). T component does not seem to have a specific function in virus infectivity and may be regarded as a side product of the viral assembly process. The amount of T component produced varies greatly among different comoviruses and even among different isolates of the same virus.

The genome of comoviruses consists of two species of positive-sense, single-stranded RNA. The sizes of the RNA molecules differ slightly among the various comovirus species. For some species [CPMV, cowpea severe mosaic virus (CPSMV), and red clover mottle virus (RCMV)], complete nucleotide sequences are available (Table I).

The genomic RNA molecules of all comoviruses so far tested have a small protein covalently linked to their 5' ends (denoted VPg: Viral Protein genome-linked) and a poly (A) tract at their 3' ends (Fig. 1), structural properties that comoviruses share with some other plant viruses (e.g., nepoviruses, potyviruses, sobemoviruses, and luteoviruses) and with the picornaviruses and caliciviruses of animals.

In addition to RNA, the M and B components contain spermidine (approximately 200 molecules/particle) and traces of putrescine and spermine (Bruening, 1977; Nickerson and Lane, 1977). These polyamines are probably closely associated with the RNA molecules.


B. Separation of Particle Components

Separation of Band M components is important for the generation of pseudorecombinant viruses by mixing B component from one virus (strain) with M component from another virus (strain). This type of experiment allows the assignment of the encoded functions to the individual genome segments. Separation of components is also needed to obtain pure RNA-l and RNA-2, which can be used for in vitro translation studies. This is certainly true for those comoviruses for which no infective cDNA clones are available. Two convenient techniques for large-scale separation of the nucleoprotein particles are available, namely velocity gradient centrifugation in sucrose gradients and equilibrium centrifugation in density gradients.

When CsCI equilibrium gradients are utilized, one should be aware that at (even slightly) alkaline pH the B components of several comoviruses split into two forms that differ in buoyant density but still have the same biological activity (Bruening, 1977). These forms are designated Bu (B-upper, buoyant density 1.42 g/ml) and BJ (B-Iower, buoyant density 1.47 g/ml), and their existence has been suggested to be due to differential replacement of the polyamines by cesium ions, thereby causing differences in buoyant density (Bruening, 1977; Virudachalam et a1., 1985).

C. Particle Proteins

1. Comoviruses Have Two Different Capsid Proteins

Comoviruses are distinguished from most other small isometric plant viruses in having two types of capsid protein, which occur within the capsids in a ratio of 1:1 (Wu and Bruening, 1971; Geelen, 1974). Reported molecular weights, as derived from electrophoretic mobilities in sodium dodecyl sulfate (SDS)-polyacrylamide gels, range for the larger capsid protein between 37,000 and 43,000 and for the smaller capsid protein between 18,000 and 26,000, depending on the virus and on the gel system used (Wu and Bruening, 1971; Geelen, 1974; Blevings and Stace-Smith, 1976; Oxelfelt, 1976; Bruening, 1978; Brunt et a1., 1984). In various reports, the molecular weights of the CPMV (strain Sb) capsid proteins have been estimated (by electrophoresis) to be approximately 37,000 and 23,000, hence these proteins have been historically denoted as VP37 and VP23, respectively (Rottier et al., 1979; Franssen et al., 1982). However, the precise localization of the coding region of these proteins on CPMV RNA-2 has enabled their molecular weights to be calculated exactly (Van Wezenbeek et a1., 1983; Franssen, 1984). Whereas the experimental estimate for the molecular weight of the small capsid protein (VP23) agrees rather well with the calculated molecular weight (23,930), there is a striking discrepancy between the experimental and calculated


values for the large capsid protein (37,000 vs. 41,720). However, to avoid any confusion when reviewing the literature, the proposed nomenclature for the CPMV proteins, based on their apparent sizes (VP37 and VP23) (Franssen et a1., 1982), will be followed here.

2. The Small Capsid Protein Is Heterogeneous in Size

The small capsid protein (but not the large one) of almost all comoviruses tested so far is heterogeneous in size (Bruening, 1977). This heterogeneity, which is caused by a specific though still obscure processing reaction, results in most cases in the occurrence of two major size classes, one of which probably corresponds with the intact protein. As a consequence, in most comovirus isolates two electrophoretic forms of virus particle can be distinguished, a slow and a fast migrating form, each containing both Band M components (Bancroft, 1962; Agrawal, 1964; Semancik, 1966; Geelen, 1974; Siler et a1., 1976). With CPMY, the slow form has been shown to contain the large (intact) version of VP23, whereas the fast form exclusively contains the trimmed version, which is approximately 3000 Da smaller (Geelen et a1., 1972). The conversion of intact to trimmed VP23 proceeds with aging of the virus particles and occurs exclusively at the C-terminus (Niblett and Semancik, 1969; Franssen, 1984; Kridl and Bruening, 1984). Indeed, Edman degradation of unfractionated protein gave only one unambiguous amino-terminal sequence for CPMV VP23 (Van Wezenbeek et a1., 1983; Franssen, 1984). Based on results obtained with carboxypeptidase Y, Franssen et a1. (1986) concluded that the conversion from intact to trimmed VP23 is caused by the (single step?) release of a carboxy-terminal peptide of 23 amino acids, having a molecular weight of 3157. The size of this peptide agrees with the difference in apparent molecular weight of the two forms of VP23; moreover, the fact that it contains two lysine and four arginine residues (Franssen, 1984) may explain the observed difference in charge of the slow and fast electrophoretic forms of CPMV particles. Amino acid analysis of the large capsid protein VP37 demonstrated that this protein is blocked at the N-terminus, probably by an N-acetylated methionine (Bruening, 1981).

3. Glycosylation of the Capsid Proteins

There is only one report documenting glycosylation of comoviral capsid proteins. Using gas liquid chromatography and mass spectrometry, Partridge et a1. (1974) identified glucosamine, glucose, galactosamine, and minor amounts of galactose and mannose in the capsid of CPMV (in total 1.90 g carbohydrate per 100 g of capsid protein). Strikingly, no carbohydrate compounds were found in the capsid of bean pod mottle virus (BPMV). Indeed, the primary structure of both VP37 and VP23 of CPMV reveal the presence of (respectively 3 and 2) potential sites for N-type glycosylation (Asn-x-Thr and


Asn-x-Ser) (Goldbach and Van Kammen, 1985). Whether these sites indeed are used for glycosylation remains to be determined. Assuming that the capsid is composed of 60 copies each of VP37 and VP23, the data of Partridge et al. (1974) indicate that each pair of capsid proteins contain together, on average, two molecules of glucosamine and one molecule each of glucose and galactosamine, respectively.

D. Virion Structure

As discussed elsewhere in this volume (see Chapters 1 and 6), the genomes of comoviruses closely resemble in structure and organization those of picornaviruses of animals. Moreover, these viruses exhibit significant sequence homology in their replication proteins (see also Section VI), providing strong evidence that comoviruses and picornaviruses, although having very distinct host ranges, are evolutionarily related by common ancestry (Franssen et al., 1984c; Goldbach, 1986; Goldbach and Wellink, 1988). Though comoviruses have two capsid proteins whereas picornaviruses have four, and though there is no homology in the primary structure of the capsid proteins between comoviruses and picornaviruses, their virions have the same basic geometry. Virion structures have been determined to atomic resolution (2.8-3.5 A maps) for two comoviruses (CPMV, BPMV) and six picornaviruses (reviewed in Rossmann and Johnson, 1989). These analyses revealed that the two comoviral proteins together form three distinct globular structures ("13-barrels") that are equivalent to the three l3-barrels formed by VP1, VP2, and VP3 of picornaviruses. The large comoviral capsid protein (which comprises two l3-barrels) corresponds to VP2 and VP3, whereas the small capsid protein corresponds with VP1, i.e., it forms pen tamers around the fivefold symmetry axes (Chen et al., 1989; Stauffacher et al., 1987) (Fig. 2). Thus the capsids of comoviruses and picornaviruses are very similar in architecture, consisting of 60 copies each of three different 13-barrels arranged in a pseudo T = 3 lattice. Both groups of viruses appear to have evolved from the T = 3 structure, as found in the more primitive plant and insect viruses, as a result of triplication of a protocapsid gene coding for a single l3-barrel.

There are of course differences in the capsid morphology of picornaviruses and comoviruses, probably because picornaviruses need cell receptor binding sites and have co evolved with the animal immune system, whereas comoviruses do not need cell receptors but must be transmissible by beetle vectors. Thus the comovirus subunits lack the long N- and C-terminal extensions present in the picornaviral proteins, and there is no counterpart for VP4. The large loops at the surface of the picornaviral capsid proteins, making up the main epitopes in the virus particle, are lacking in comoviruses, which have much smoother surfaces (King et al., 1991).

The similarity in basic geometry of the comoviral and picornaviral cap-


PICORNAVIRUS

r"H:>l

COMOVIRUS

FIGURE 2. A comparison of the pseudo T = 3 picornavirus and comovirus capsids. Each trapezoid represents a l3-barrel, as schematically drawn at the left. In the picornavirus capsid, the three capsid proteins, VP1, VP2, and VP3, each contain a l3-barrel. The comoviral capsid is similar to the picornaviral capsid except that two of the l3-barrels (corresponding to picornaviral VP2 and VP3) are covalently linked to form a single protein, the large (L) capsid protein. The small (S) comoviral capsid protein corresponds to VPl. (Figure kindly provided by Dr. O. Le Gall.)

sids has been elegantly exploited by Porta et al. (1994) who have shown that CPMV particles can be used as an "epitope-presenting system" for the production of vaccines against, e.g., picornaviruses. Thus an epitope derived from foot-and-mouth disease virus (FMDV), consisting of 25 amino acids (known as the "FMDV loop"), was inserted in CPMV VP23 at a site selected on the basis of the known three-dimensional structure of the two proteins (Usha et a1., 1993). To this end, a synthetic oligonucleotide sequence encoding the FMDV loop was inserted into a full-length cDNA clone of RNA-2. Electron microscopy revealed the presence of virus particles in cowpea leaves that were inoculated with an in vitro transcript of this mutant clone, mixed with a transcript of a full-length cDNA clone of RNA-I. This indicates that the foreign sequence did not interfere with the ability of the carrier virus to assemble. The CPMV virions produced were shown to possess the antigenic property of the inserted FMDV epitope (Usha et al., 1993). Likewise, a neutralizing epitope located on human immunodeficiency virus-1 (HIV-1) protein gp 41 was incorporated in the CPMV VP23 sequence, yielding a CPMV recombinant that induced HIV-neutralizing antibodies in mice (McLain et al., 1995).


E. Assembly

The process of virus assembly in comoviruses remains obscure, largely because they disassemble irreversibly, so that in vitro assembly studies are impossible. In vitro translation experiments (see also Section IV) indicate that the two capsid proteins of CPMV are released from a common precursor protein of approximately 60 kDa which is encoded by RNA-2. This capsid protein precursor has also been detected in vitro (Wellink et a1., 1987b). Therefore, it is tempting to assume that comovirus particles are assembled from such precursor proteins and that the two mature capsid proteins are generated by a final proteolytic cleavage only after the assembly of a so-called procapsid structure. The great advantage of such a scenario would be that the particle can be built up from 60 strictly equal building blocks.

However, when expressed in transformed tobacco plants, this 60-kDa capsid protein precursor was shown to accumulate to considerable amounts (2 pg/g of wet tissue) but did not assemble into viruslike capsids (Nida et a1., 1992). It is not known whether this is due to the way in which the precursor was expressed or whether proteolytic processing into the mature capsid proteins is a prerequisite for assembly. The existence of empty T-component particles in infected cells indicates that encapsidation of RNA is not essential for assembly into capsids. Furthermore, no signals on the RNA that might be involved in the encapsidation process of comoviral RNA have been identified. It has been suggested that VPg acts as a signal for encapsidation, but experimental data supporting this idea are lacking.

III. GENOME STRUCTURE AND ORGANIZATION

A. Terminal Structures

The Single-stranded RNA molecules of comoviruses are positive-stranded, i.e., they are infective in the absence of any viral proteins and are directly translatable in vitro. The genomic RNA species of all comoviruses investigated so far [Andean potato mottle virus (APMV), broad bean true mosaic virus (BBTMV), BPMV, CPMV, CPSMV, squash mosaic virus (SMV), RCMV] are characterized by two structural features, namely a poly(A) tail at the 3' end and a protein (VPg) at the 5' end (El Manna and Bruening, 1973; Daubert and Bruening, 1979; Daubert et a1., 1978; Stanley et al., 1978). These two structural features are also found in the genomic RNA species of the nepoviruses (see Chapter 6), potyviruses, luteoviruses, and sobemoviruses and, outside the plant virus world, in the picornaviruses and caliciviruses.

The poly(A) tails of both RNA-l and RNA-2 of CPMVare heterogeneous in length (Ahlquist and Kaesberg, 1979), that of RNA-2 ranging from 20 to 400 residues, with a mean value of 167, whereas that of RNA-l is significantly shorter, ranging from 10 to 170 residues, with a mean value of 87. These


poly(A) tails are not synthesized by a poly(A) polymerase, as is polyadenylated cellular mRNA. No polyadenylation signal (AAUAAA) is found in either RNA-lor RNA-2; moreover, 5'-terminal poly(U) stretches have been identified in the minus-strands of replicative-form (RF) molecules (Lomonossoff et al., 1985), indicating that the poly(A) tail is included in the normal replication process.

The comoviral VPg is encoded by RNA-I; reported sizes, deduced from electrophoretic mobility, range between 3000 and 6000 Da. Sequence information revealed that the VPgs of three comoviruses (CPMV; CPSMV, RCMV) are almost identical and consist of 28 amino acids (Wellink et al., 1986). The VPg of CPMV is linked to the 5'-terminal uridine of the genomic RNA molecules by the f3-0H group of the serine residue located at the N-terminal end of the protein (Jaegle et a1., 1987). Removal of VPg by digestion with proteinase K does not influence the infectivity of the comoviral RNA molecules (Stanley et a1., 1978). The possible function of this peculiar protein as a primer for RNA replication will be discussed in Section VI.

B. The Nucleotide Sequence of Cornoviral RNA Species

For CPMV; CPSMV, and RCMV; the complete nucleotide sequences of both genomic RNA species are available (Table I). Excluding the poly(A) tails, the RNA-l molecules have 5850-6050 nucleotides, and the RNA-2 molecules have 3450-3750 nucleotides. The RNA molecules each contain a single open reading frame (ORF) covering more than 90% of their sequence. This agrees with the observations from translation experiments that expression of both RNA molecules involves the production of large primary translation products ("polyproteins") from which the functional proteins are derived by proteolytic processing. For CPMV the ORF of RNA-l runs from a start codon at position 207 to a (UAG) stop codon at position 5805 (Lomonossoff and Shanks, 1983) (Fig. 3), and can code for a protein of 209,663 Da. The ORF in RNA-2 of CPMV runs from a start codon at position 161 to a (UAA) stop codon at position 3299, corresponding to a protein of 115,823 Da (Van Wezenbeek et a1., 1983). These theoretical values for the primary translation products agree well with the apparent molecular weights of both the in vitro and in vivo translation products obtained from these RNAs. For all comoviruses tested, a second in-frame AUG codon in RNA-2 apparently also initiates translation, at least in vitro (Goldbach and Krijt, 1982).

In addition to the ORFs, both genomic RNA molecules have 5'- and 3'-terminal nontranslated regions (NTRs). The 5' NTR ranges between 161 and 450 nucleotides, whereas the 3' NTR ranges from 85 to 470 nucleotides, depending on the virus. It is to be expected that these sequences contain important regulatory signals such as binding sites for ribosomes, for the viral replicase, and possibly for the capsid proteins. Indeed, the NTRs of CPMV, CPSMV, BPMV; and RCMV begin with the same VPg-UAUUAAAAU ...

44 R. W. GOLDBACH AND f. WELLINK

RNA-2 RNA-1 512 3299 207 5805

VP 161 mp structural co-pro hel? VPg pro pol VPg : 58kD/48kD VP37 VP23 An 32kD 58kD 24kD 87kD An

I /'-... '-... OM OG OS OS OM OG , ,

105kD • 200kD

95kD 32kD • 170kD

58kD 60kD 84kD • 87kD

48kD 60kD 60kD • 110kD

VP37 VP23 58kD • 112kD

VPg 24kD .-- 87kD

FIGURE 3. Genetic organization and translational expression of the CPMV genome. Open reading frames in the RNA molecules are indicated with open bars and VPg with a black square. Nucleotide positions of start and stop co dons are indicated. Abbreviations: mp, movement protein; pro, proteinase; co-pro, co-factor required for proteinase; hell, putative helicase; pol, RNA-dependent RNA polymerase.

structure (Chen and Bruening, 1992a), which forms part of two conserved stem/loop structures (c. Pleij, personal communication) (Fig. 4). Moreover, because both RNA species of a single comovirus presumably undergo the same interactions with viral or cellular proteins, one might expect the binding sites for these proteins to be identifiable as shared nucleotide sequences. Indeed, for individual comoviruses, the 5' termini of RNA-l and RNA-2 contain regions with high sequence homology and so do the 5' and 3' termini (see Section VI.C).

IV. TRANSLATION

A. Initiation of Translation

Comoviruses employ a "polyprotein processing" translation strategy, i.e., each genomic RNA species is functionally monocistronic, their primary translation products being proteolytically processed to give several mature, functional proteins. Also, the comoviral RNA species are structurally peculiar as messenger RNA molecules. At the 3' end, they have a poly(A) taillike normal mRNA, but at the 5' end, instead of a cap structure, they have a small protein (VPg). For poliovirus it has been shown that VPg is not present on viral RNA isolated from polysomes in infected cells (Nomoto et a1., 1977; Hewlett et a1., 1976), but it is not known whether this is also true for comoviruses. In reticulocyte lysates it has been shown that CPMV VPg is

COMOVIRUSES: MOLECULAR BIOLOGY

U A I A e A A

e-G U-G A-U A-U A-U AA II A-U G A U-G A-U

U-AI IG-el A-U G-e

'j-AAAAGCGAACGU-AAA

vPg. CPMV RNA-1

45

A A I U U U A

e-G U-G A-U A-U A-U AA II A-U G A U-G G-e

~:~I I~:gl j-AAAAGCGAACGU-GAA

VPg. CPMV RNA-2

FIGURE 4. Secondary structure prediction of the 5' regions of CPMV RNA-l and RNA-2. The solid line represents sequences conserved among different comoviruses. The length of hairpin I varies among the different viruses.

stable when linked to RNA, but is rapidly degraded when the RNA is removed (De Varennes et a1.,1986J. However, the presence ofVPg seems not to be important for the translation of the CPMV RNA molecules, because in vitro transcripts, which lack VPg, are infective and are translated efficiently in vitro (Vos et a1., 1988aJ.

The initiation of translation on comoviral RNA must be different from that of normal eukaryotic mRNA where the cap has an important role. It is not known whether initiation of translation on the comoviral RNA molecules requires a free 5' end or whether internal initiation can take place on the 5' NTR as has been described for picornaviruses (Sonenberg, 1991J. Preliminary data suggest that for efficient translation of CPMV RNA a free 5' end is required (J. Wellink and G. Kroon, unpublished resultsJ.

In CPMV, RNA-I has a 5' NTR of 206 nucleotides and RNA-2 has a 5' NTR of 160 nucleotides. The first 50 nucleotides of both NTRs are very similar, but the remainder of the sequence is widely divergent. Both 5' NTRs are relatively poor in G residues (15.5 % J, and, except for the two conserved stem/loop structures (Fig. 4J, no extensive secondary structures have been predicted to occur. The conserved domain in the NTR probably plays an important role in replication of the RNA molecules. Whether these sequences also playa role in translation is not clear.

The 5' NTR of CPMV RNA-2 contains an AUG codon at position 115 which starts a 2-kDa ORF that ends with a stop codon at nucleotide position 175. Although the RNA-2 molecules of other comoviruses studied do not contain this small ORF, it could playa role in the translation of CPMV RNA (see Section IV.BJ.


B. Mechanism of Translation of RNA-2

The RNA-2 molecules of all comoviruses tested are translated in vitro into two carboxy-coterminal polyproteins (Pelham, 1979; Goldbach and Krijt, 1982). For CPMV this has also been shown to occur in infected cells (Rezelman et a1., 1989; Holness et a1., 1989). Translation of CPMV-2 is initiated at the AUG codons at positions 161 for the 105-kDa protein (Vos et a1., 1984; Holness et a1., 1989) and 512 for the 95-kDa protein (Wellink et a1., 1993b). According to Kozak (1984), the AUG codon at position 161 is in a suboptimal context for initiation of translation (A at -3 and U at +4), and, indeed, Verver et a1. (1991) showed that in vitro some of the ribosomes fail to initiate at this AUG and proceed to scan until they reach the AUG at position 512, which has a good context (G at -3 and +4). All comoviral RNA-2 species sequenced so far have a suboptimal AUG codon at the beginning of the large ORF.

It is also possible that ribosomes bypass AUG 161 on CPMV RNA-2 by initiating at AUG lIS, terminating at nucleotide 175, and then reinitiating at AUG 512. Evidence has been obtained that, in vitro, ribosomes can initiate at AUG 115, although it has a suboptimal context (U at -3 and A at +4) (Wellink et a1., 1993b). Furthermore, a mutant RNA that lacks this AUG produces relatively more 105-kDa protein than 95-kDa protein, indicating that AUG 115 influences translation initiation on AUG 161 (Wellink et a1., 1993b).

By constructing dicistronic messengers, Verver et a1. (1991) showed that the 161-512 sequence of RNA-2 is able to mediate internal initiation of translation in reticulocyte lysates and wheat germ extracts. This would be another mechanism for production of the 95-kDa protein. Using a different approach, Thomas et a1. (1991) obtained similar results. These authors also reported that eukaryotic initiation factor 4F (eIF-4F) stimulated RNA-2 translation. Normally, eIF-4F activity is linked to the translation of capped mRNAs, but the authors speculate that in CPMV the unwinding activity of eIF-4F is needed for initiation on RNA-2. The initiation of translation of CPMV RNA-2 on both AUGs showed a different sensitivity to the Mg2+ concentration (Thomas et a1., 1991). This result suggests that either the secondary structure around each site is different or initiation occurs by two different mechanisms.

Belsham and Lomonossoff (1991) used dicistronic messsengers in an animal cell transient expression system, but found no evidence for internal initiation in vivo. Nevertheless, in vitro the 161-512 sequence of RNA-2 is sufficient to allow internal binding of ribosomes. This sequence is very rich in U residues (37.2 %), and no extensive secondary structures have been predicted to occur. Indeed, large stretches of the sequence are probably in a single-stranded form, and possibly this absence of secondary structure favors the internal initiation on CPMV RNA-2.


V. PROCESSING OF THE CPMV POLYPROTEINS

A. Processing in Vivo

Upon translation of the comoviral RNA species, large polyproteins are produced that are processed at specific sites into several intermediate and final cleavage products. This process has been studied extensively for CPMV both in vivo (cowpea protoplasts) and in vitro (reticulocyte lysates). Processing data about polyproteins encoded by other comoviruses are scarce (Beier et a1., 1981; Gabriel et a1., 1982), but sequence comparisons suggest that processing follows pathways similar to those found for CPMV.

Cowpea mesophyll protoplasts are readily inoculated with CPMV and, upon labeling with 35S-methionine, virus-specific proteins with apparent sizes of 170,112,110,87,84,60,58,37,32, and 23 kDa can easily be detected (Rottier et a1., 1980; Rezelman et a1., 1980). The combined molecular weight of these proteins greatly exceeds the coding capacity of the two genomic RNA molecules, and this indicates that several of these proteins are precursors of others. The 37- and 23-kDa proteins were found to be encoded by RNA-2, and immunological tests showed that they represent the capsid proteins (Rottier et a1., 1980). Comparison of the various RNA-I-encoded proteins by peptide mapping and further analysis of extracts of infected protoplasts by immunological techniques have resulted in the processing model for the RNA-l polyprotein shown in Fig. 3 (Rezelman et a1., 1980; Zabel et a1., 1982; Goldbach et a1., 1982; Goldbach and Rezelman, 1983; Dorssers et a1., 1983; Peters et a1., 1992b; Wellink et a1., 1987b).

The existence of the RNA-2-encoded 58- and 48-kDa proteins was first postulated on the basis of in vitro translation experiments (see next section). Later, in experiments with antipeptide antibodies, both proteins were detected in infected protoplasts and plants (Wellink et a1., 1987a; Rezelman et a1., 1989; Holness et a1., 1989). By incubating protoplasts in the presence of 2 mM ZnC12 to inhibit proteinase activity, it also has been possible to detect the 105- and 95-kDa polyproteins (Rezelman et a1., 1989) and the 60-kDa capsid protein precursor (Wellink et a1., 1987a) (see Fig. 3). N-terminal amino acid sequence analysis of the purified capsid proteins, the VPg, and the 170-, 110-, 87-, 84-, 60-, and 58-kDa proteins isolated from infected protoplasts, confirmed the processing model shown in Fig. 3 (Zabel et a1., 1984; Franssen et a1., 1986; Wellink et a1., 1986) and revealed that cleavages in the CPMV polyproteins occur at specific Gln/Gly, Gln/Ser, and GIn/Met sites.

B. Processing in Vitro

Early in vitro translation studies showed that the CPMV RNA molecules are translated into large proteins (Pelham and Jackson, 1976; Davies

48 R. W. GOLDBACH AND f. WELLINK

et a1., 1977). Pelham (1979), in an extensive study on the translation of the CPMV RNA molecules in reticulocyte lysates, was the first to show that proteolytic processing occurred in this system, presumably by a product encoded by RNA-I. The 200-kDa polyprotein was rapidly cleaved into 170-and 32-kDa proteins. This cleavage required ATP and reducing agents and was inhibited by reagents that modify cysteine residues (Pelham, 1979). The 105- and 95-kDa polyproteins of RNA-2 were cleaved into 58- or 48-kDa proteins and the 60-kDa precursor to the capsid proteins (Pelham, 1979; Franssen et a1., 1982). Subsequent studies showed that the proteolytic activity responsible for this cleavage is also produced in protoplasts inoculated with RNA-1 only (Franssen et a1., 1982), and studies with specific antibodies provided evidence that the 32-kDa protein played an essential part in this cleavage (Franssen et a1., 1984b).

In later in vitro translation studies it was found that upon prolonged incubation the 170-kDa protein is further processed into 110-, 87-, 84-, 60-, 58-, and 24-kDa proteins (Fig. 3) (Franssen et a1., 1984a; Peng and Shih, 1984). Since all these cleavage products are very similar to the viral proteins produced by CPMV in vivo (Franssen et a1., 1984a), the lysate system is very suitable for studying the various steps in this process in more detail. The only exception is the cleavage of the RNA-2-encoded 60-kDa protein into the two capsid proteins VP37 and VP23, which has only been observed in some reticulocyte lysate preparations. This cleavage was found to be very sensitive to the hemin concentration in the lysate (Bu and Shih, 1989).

The availability of Escherichia coli expression vectors and efficient in vitro transcription systems has made it possible to study the proteinase and the processing pathways of the CPMV polyproteins in great detail (Garcia et a1., 1987; Verver et a1., 1987; Vos et a1., 1988b; Dessens and Lomonossoff, 1991, 1992; Peters et a1., 1992a,b). By mutational analysis it was shown that the 24-kDa protein is the proteinase responsible for all cleavages in the viral polyproteins (Garcia et a1., 1987; Verver et a1., 1987; Vos et a1., 1988b). For the cleavages in the RNA-2 polyprotein this was shown by constructing a hybrid RNA-1/RNA-2 in which the 24-kDa coding region was placed in frame with a part of the RNA-2 polyprotein coding region including the cleavage sites. In these hybrid proteins the 24-kDa proteinase was able to carry out cleavages at all sites present in the construct (Garcia et a1., 1987; Vos et a1., 1988b). However, for efficient in-trans processing of the GIn/Met site in the RNA-2 polyprotein (which is the normal situation in infected cells), the 32-kDa protein was found to be also required as cofactor (Vos et a1., 1988b). The available evidence suggests that after cleavage the 32-kDa protein associates with the 58-kDa domain of the 170-kDa protein. This complex is then able to cleave the GIn/Met site in the RNA-2 polyprotein. The 32- and 170-kDa proteins do not necessarily have to be derived from the same 200-kDa molecule, but translation from separate RNA molecules must be simultaneous in order to form a functional complex (Peters et a1., 1992a). This observation suggests that activity of the complex is dependent on a certain conformation


and that interaction between the 32- and 170-kDa proteins has to take place during translation.

Remarkably, the 32-kDa protein is also involved in the processing of the RNA-1 polyprotein. Release of the 32-kDa protein from the 200-kDa polyprotein by the 24-kDa proteinase is always very rapid and can occur as soon as the ribosomes have finished translation of the 24-kDa coding region (Franssen et a1., 1984a). Further cleavages of the 170-kDa protein occur only very slowly, a situation that is also apparent in vivo because the 170-kDa protein is the most abundant non structural protein in infected cells. However, when the 170-kDa protein is translated from a mutant RNA that lacks the 32-kDa coding region, it is efficiently further processed into 112-, 110-, 87-, 84-, 60-, 58-, and 24-kDa proteins (Peters et a1., 1992a). Thus the presence of the 32-kDa protein somehow prevents further cleavages of the 170-kDa protein. Again this is probably achieved by association of the 32-kDa protein with the 58-kDa domain in the 170-kDa protein, thereby inducing a conformation that makes the 170-kDa protein less prone to further cleavages.

Studies with cleavage site mutants have revealed that all cleavages in the RNA-1 polyprotein occur most efficiently in cis (Peters et a1., 1992b). The 170-kDa protein can be cleaved at three different sites into three sets of proteins, 58 kDa + ll2 kDa, 60 kDa + llO kDa, and 84 kDa + 87 kDa (see Fig. 3). In the lysate the 60-kDa protein is stable and cannot be further cleaved in trans into the 58-kDa protein and VPg (Peters et a1., 1992b). The 84-kDa protein can be cleaved into either 60-kDa + 24-kDa or 58-kDa + 26-kDa proteins. The 26-kDa protein does not accumulate and is rapidly cleaved into VPg and the 24-kDa protein. Studies by Dessens and Lomonossoff (1992) have shown that cleavage of the llO-kDa protein at the 24-kDa/87-kDa junction is extremely inefficient. Cleavage at this site is greatly enhanced by sequences upstream of the 24-kDa proteinase domain, suggesting that the 87-kDa protein arises only through direct processing of the 170-kDa protein.

The 32-kDa/170-kDa and the 84-kDa/87-kDa cleavages can occur in trans, but this is not an efficient process (Vos et a1., 1988a; Peters et a1., 1992b). A remarkable observation is that the cleavage efficiencies vary considerably between different lysate preparations (Franssen et a1., 1984a; Peng and Shih, 1984; Vos et a1., 1988b; Peters et a1., 1992b). Whether this is caused by differences in the number of molecules that are misfolded or by the presence of undefined factors in the lysate that playa role in the processing is not clear.

Processing has also been studied with a transient expression system in cowpea protoplasts. Results obtained with this system confirm that the 32-kDa protein has a regulatory role in the processing of the 170-kDa protein and that the llO-kDa protein is a stable product (Van Bokhoven et a1., 1993a). Recently, processing of the ll2-kDa protein in lysates and protoplasts has been compared. In some in vitro experiments the ll2-kDa protein was found to be stable (S. A. Peters et a1., unpublished data), whereas in other experiments cleavage of the ll2-kDa protein occurred, but only at the 24-kDa/87-


kDa junction (Dessens and Lomonossoff, 1992). In protoplasts, cleavage of the 112-kDa protein occurred via two alternative pathways either into the VPg and 1l0-kDa proteins or into the 26- and 87-kDa proteins (S. A. Peters et a1., unpublished data). The 26-kDa protein was probably further cleaved into the VPg and 24-kDa proteins.

C. The Cleavage Sites

Partial N-terminal amino acid analysis has revealed that cleavages in the CPMV polyproteins occur at GIn/Met (2 sites), Gln/Ser (2 sites), and Gln/Gly (2 sites) dipeptide sequences (Fig. 3). When sequences surrounding the cleavage sites were compared, it was found that Ala or Pro are present at position -2 and that five of the six sites have Ala at position -4 (Wellink et a1., 1986).

Cleavage sites in polyproteins from other comoviruses, as determined by amino acid sequence analysis or by comparison with the CPMV cleavage sites, are very similar to the CPMV sites. All 20 cleavage sites examined have GIn at position -I, while at position -4, 13 sites have Ala and 4 sites have Val, and at position -2, 16 sites have Ala or Pro. For CPSMV and RCMV, Gln/ Ala and Gln/Thr cleavage sites have been reported (Shanks and Lomonossoff, 1992; Shanks et a1., 1986; MacFarlane et a1., 1991; Chen and Bruening, 1992a,b).

At several cleavage sites in the CPMV polyproteins the effect of mutations on the in vitro cleavage efficiency was investigated. When the Gly at position + 1 in the Gln/Gly cleavage site between the capsid proteins was changed into Ala, Ser, or Met (amino acids that are present at this position at other sites), cleavage was almost completely abolished (Vos et a1., 1988b). Apparently cleavage site requirements for this site are very strict. A different picture emerges from experiments in which mutations were introduced at the Gln/Ser site between the RNA-I-encoded 32- and 170-kDa proteins. When changed into a His/Met dipeptide sequence, this site was still efficiently cleaved; even after insertion of four amino acid residues between the His and the Met, cleavage still occurred (Peters et a1., 1992b). These results indicate that the cleavage site requirements for this efficient in cis cleavage are not stringently determined by the Gln/Ser dipeptide sequence. Apparently, during synthesis of the RNA-l polyprotein, folding of this polypeptide chain drives the active site of the 24-kDa proteinase and the cleavage site together and favors a rapid intramolecular cleavage.

D. The Proteinase

Franssen et a1. (1984c) and Argos et a1. (1984) showed that there is considerable amino acid sequence homology between the 3C proteinase .of picornaviruses and the 24-kDa proteinase of CPMV. Later it was found that a


considerable number of RNA viruses contain 3C-like proteinases that show homology with a group of cellular serine proteinases (Bazan and Fletterick, 1988; Gorbalenya et ai., 1989a). However, instead of a serine, the viral enzymes contain a cysteine at their active site. Based on these sequence homologies, two models were constructed that predicted the positions of the other two amino acids that would form the catalytic center of these proteinases (Bazan and Fletterick, 1988; Gorbalenya et a1., 1989a. The results of sitedirected mutagenesis studies on the 24-kDa proteinase of CPMV by Dessens and Lomonossoff (1991) strongly favor the model proposed by Gorbalenya et a1. (1989a) in which the His-40, the Glu-7S, and the Cys-166 form the catalytic triad.

VI. REPLICATION

A. Cellular Location of Comoviral RNA Replication

Characteristic cytopathological structures appear in the cytoplasm of CPMV-infected cells (Assink et a1., 1973; De Zoeten et a1., 1974; Hibi et a1., 1975). These structures consist of large arrays of membranous vesicles surrounded by amorphous electron-dense material. CPMV RNA-l is able to replicate in cowpea protoplasts by itself, independently from RNA-2, and in such RNA-I-infected protoplasts, proliferation of membranes and electrondense material is also found (Rezelman et a1., 1982). Hence, it appears that the induction of cytopathological structures is a RNA-I-encoded function and represents a prerequisite for viral RNA replication. Indeed, by autoradiography performed on sections of these cells and on isolated cytopathological structures, De Zoeten et al. (1974) provided evidence that replication of CPMV RNA is associated with such membranous vesicles. It has been established that the electron-dense structures contain the bulk of the nonstructural proteins encoded by CPMV RNA-1 (Wellink et ai., 1988), but it is not known whether these structures, like the membranous vesicles, have a specific function in viral RNA replication.

Electron microscopic examination of insect cells in which the 200-kDa polyprotein was produced with the help of a baculovirus vector revealed the presence of numerous membranous vesicles and electron-dense structures, resembling the characteristic cytopathological structures found in CPMVinfected cowpea cells (Van Bokhoven et al., 1992). The induction of membranous vesicles and electron-dense structures therefore appears to be a property of the RNA-I-encoded proteins per se, independent of viral RNA replication taking place or the occurrence of plant factors. Interestingly, the membranous vesicles were also observed in insect cells in which the RNA-lencoded 60-kDa protein alone was produced, whereas in cells in which the 87- or llO-kDa proteins were produced, such structures were absent. Immunogold labeling of these cells using anti-VPg serum and protein A-gold re-


vealed that the 60-kDa protein was associated with these vesicles (Van Bokhoven et a1., 1992). These observations suggest that the 60-kDa protein is responsible for the induction of membrane proliferation in insect cells and may also have that role in plant cells. Furthermore, the association of the 60-kDa protein with these vesicles strengthens the suggestion made previously (Eggen and Van Kammen, 1988) that the 60-kDa protein has a role in anchoring replication complexes to membranes. For poliovirus, the induction of membranous vesicles and attachment of replication complexes to these vesicles has been attributed to protein 2C, which has extensive sequence identity with the 60-kDa protein of CPMV (Bienz et a1., 1987). These results support the notion that the CPMV 60-kDa protein and poliovirus protein 2C fulfill analogous functions in viral RNA replication.

B. Viral Proteins Involved in CPMV Replication

In this section the role of viral proteins in CPMV RNA replication will be discussed, especially the role of the viral RNA polymerase, the nucleotide binding motif (NTBM)-containing protein, VPg, and the RNA-2-encoded 58-kDa protein. The 32- and 24-kDa proteins too are essential for replication because they are responsible for release of functional replicative proteins from the polyproteins (see Section V).

1. The RNA Polymerase

The RNA-dependent RNA polymerases (RdRp) of animal viruses, plant viruses, and bacteriophages have characteristic conserved motifs in their amino acid sequences (Kamer and Argos, 1984; Poch et a1., 1989; Koonin, 1991). One prominent block is S/GTXXXTXXXNT/S (in which X may be any amino acid) followed 21 to 52 amino acids downstream by a second block consisting of a highly conserved GDD sequence embedded in a stretch of hydrophobic amino acid residues (Kamer and Argos, 1984; Franssen et a1., 1984c). The GDD consensus sequence is thought to be at or near the catalytic site of the polymerase molecule. Indeed, single amino acid substitutions at the glycine in the GDD sequence of the viral RNA polymerases of QI3 and poliovirus resulted in partial or complete loss of polymerase activity (Inokuchi and Hirashima, 1987; Jablonski et a1., 1991). Based on these comparisons, the RNA-I-encoded 87-kDa protein could be the RdRp; however, several precursors of the 87-kDa protein, namely the 110-, 112-, and 170-kDa proteins, are also present in CPMV-infected cells and could represent alternative forms of the RdRp.

Replication of CPMV RNA is associated with the vesicular membranes of the virus-induced cytopathological structures in infected cells, and, accordingly, RNA polymerase activity has been detected in the crude membrane fraction of infected leaves (Zabel et a1., 1974). Moreover, the crude


membrane fraction of CPMV-infected cowpea leaves actually harbors two functionally different RdRp activities: one that is encoded by the host and another that is virus-specific (Dorssers et a1., 1982, 1983). In the membrane fraction of uninfected cowpea leaves small amounts of (host) RdRp activity can be detected. In infected cowpea leaves this activity is increased at least 20-fold and represents more than 95% of the total RNA polymerase activity in the crude membrane fraction, masking almost completely the viral RdRp activity. The host RdRp can be stripped from the membranes by washing with a Mg2+ -deficient buffer, leaving the virus-specific RdRp activity associated with the membranes. The host RdRp, a monomeric protein of 130 kDa, transcribes endogenous plant RNA and viral RNA into small ( - )-sense RNA molecules (Dorssers et a1., 1982; Van der Meer et a1., 1983). The physiological significance of the increase of the 130-kDa host polymerase in CPMVinfected cells remains unresolved. However, the increase of 130-kDa polymerase activity is not essential for viral RNA replication because CPMV RNA replication in cowpea protoplasts is not accompanied by an increased production of 130-kDa protein (Van der Meer et a1., 1984).

The virus-specific RdRp from CPMV-infected cells is capable of fully elongating nascent viral RNA chains that have already been initiated in vivo and thus displays the features of a viral RNA replicase. In crude membrane fractions deprived of the host RdRp, the RNA products synthesized by the viral RdRp are all (+ )-stranded and are predominantly found as replicative form (RF) RNA (Dorssers et a1., 1983). The CPMV RdRp can be released from the membranes by treatment with Triton X-IOO (Dorssers et a1., 1984) and retains its RNA-elongating activity, indicating that a membranous environment is not essential.

Analysis of the protein composition of the purified RdRp revealed that the 1l0-kDa protein encoded by RNA-l was the only detectable viral protein in the preparation. Moreover, the llO-kDa protein cosedimented with the RNA polymerase activity during purification. From these results it has been concluded that the RNA-I-encoded 1l0-kDa protein, which consists of the 24-kDa protease and the 87-kDa (polymerase domain) protein, is the active viral RNA polymerase in CPMV RNA replication. This is in contrast with poliovirus where the protein corresponding to the 1l0-kDa protein, 3CD, has no polymerase activity, whereas the protein homologous to the 87 -kDa protein, 3D, is the active enzyme (Van Dyke and Flanegan, 1980). Two hostencoded proteins of 68 and 57 kDa have been detected in highly purified CPMV RdRp preparations, but it remains to be established whether these proteins are subunits or contaminants of the viral RdRp.

So far it has not been possible to isolate a CPMV RdRp activity from infected plants that is able to transcribe added template RNA. As an alternative approach to obtain a template-dependent CPMV-specific polymerase activity and in order to be able to study individual CPMV replication proteins, two heterologous expression systems have been employed, E. coli (Richards et a1., 1989) and insect cells (Van Bokhoven et a1., 1990, 1991, 1992).


Extracts of E. coli or insect cells producing the putative CPMV polymerase and its precursors were assayed for polymerase activity using polY(A) or CPMV RNA as template and oligo(U) as primer. However, none of the extracts contained such activity, despite a wide variety of conditions used for the assay. In sharp contrast, polioviral polymerase produced in both expression systems exhibited easily detectable activity in the same assays (Richards et a1., 1989; Van Bokhoven et a1., 1991, 1992). There are two main possibilities to explain the lack of activity of the CPMV polymerase: (1) CPMV polymerase requires a plant (host) factor for activity, and (2) CPMV polymerase cannot use oligo(U) as a primer in RNA synthesis (despite the high degree of homology to the poliovirus polymerase) or is not able to function on exogenous RNA (template + primer).

To investigate whether the CPMV polymerase needs a plant host factor for activity, a transient expression vector system was used in cowpea protoplasts (Van Bokhoven et a1., 1993a). DNA copies of the regions of CPMV RNA-l encoding the 87-, 110-, 170-, and complete 200-kDa proteins were each inserted in a DNA vector under control of the 35S promoter of cauliflower mosaic virus. In cowpea protoplasts transfected with the vectors, large amounts of the expected CPMV-specific proteins were synthesized. These proteins exhibited the same characteristic activities (proteolytic processing, the induction of cytopathological structures, and NTP-binding) manifested by the viral proteins produced in protoplasts inoculated with RNA-I. Nevertheless, extracts of such protoplasts were still inactive in replicating exogenous template.

However, intact cowpea protoplasts expressing the entire 200-kDa (but not those expressing the 170-,110-, or 87-kDa) protein-encoding sequence of RNA-l were able to support replication of co-inoculated RNA-2. It was concluded that all the viral replication proteins, including the viral polymerase, are functionally intact upon their synthesis in cowpea protoplasts. Apparently, expression of the complete 200-kDa coding region of RNA-l is required for CPMV RNA replication.

The most likely explanation for the lack of success in obtaining a cellfree, template-dependent replicase preparation is the inability of the CPMV polymerase to function in poly(A)/oligo(U) assays that are used to detect such activity. Possibly, the CPMV polymerase cannot use oligo(U) as a primer or cannot function on any added template-primer combination because translation and replication are closely linked (see Section VI.D). The ability of poliovirus polymerase to use poly(A)/oligo(U) may be a fortuitous property that is not shared by the CPMV polymerase. The oligo(U)-primed activity of the poliovirus polymerase has no specificity toward poliovirus RNA as template and certainly does not mimic the initiation of replication events that occur in vivo. VPg precursors might have a role in this process (see Section VI.E). Also, complete replication of poliovirus (Le., negative-stran~ and positive-strand synthesis) has only been accomplished with extracts of uninfected HeLa cells in which the input poliovirus RNA was first translated (Molla et a1., 1991). Thus it may be that in vitro CPMV replicase activity can


only be observed under conditions that first allow translation of the RNA template.

2. NTBM-Containing Proteins

A second group of proteins conserved among plus-strand RNA viruses contain a GXXXGKS/T nucleotide binding motif (NTBM) (Walker et a1., 1982; Gorbalenya et a1., 1985). For comoviruses, this conserved motif is present in the 58-kDa protein (and its precursors) encoded by RNA-1 (Fig. 3). Computer-aided sequence comparison among the various NTBM proteins has led to their clustering into three major groups; the NTBM-containing proteins of alphalike viruses, of (picornalike) picornaviruses and comoviruses, and of (picornalike) potyviruses (Gorbalenya et a1., 1988, 1989b; Hodgman, 1988). Strikingly, the NTBM-containing proteins of alphalike viruses and potyviruses bear resemblance to helicases that are related to eukaryotic translation initiation factor eIF-4A, whereas NTBM-containing proteins of the picornaviruses and comoviruses appear to be related to Simian virus-40 (SV40) large T antigen, a protein containing RNA and DNA helicase activity (Gorbalenya and Koonin, 1989; Gorbalenya et a1., 1990; Lain et a1., 1989). These sequence comparisons led Lain et a1. (1990, 1991) to test viral NTBM proteins for helicase activity, and indeed they were able to show that the cylindrical inclusion protein of plum pox (poty)virus contains a RNA helicase activity dependent on the hydrolysis of ATP to ADP.

The functional importance of the putative ribonucleotide binding site in the 58-kDa domain was tested in a covalent affinity labeling assay (S. Peters, unpublished observations). Covalent binding of chemically modified ATP to the 60- and 84-kDa proteins (both being precursors to the 58-kDa protein) was detected, and this appeared to be specific, because binding of chemically modified GTP was not observed. In other experiments, specific ATP binding was detected by using affinity chromatography on ATP-agarose (Peters et a1., 1994). Moreover, by using the transient expression system in protoplasts it was shown that a Lys to Thr amino acid substitution in the A-site of the NTBM, which is suggested to be involved in binding of the phosphate moiety of the ribonucleotide, resulted in a decreased ATP-binding capacity of the 84-kDa protein, whereas an Asp to Pro amino acid replacement in the B-site of the NTBM did not affect the ATP-binding properties of this protein. In addition, both mutations have been shown to be lethal to the virus (Peters et a1., 1994). It therefore appears that the NTBM in the 58-kDa domain is involved in an ATP-consuming function that is essential for viral RNA replication.

However, no proof that the NTBM-containing proteins of comoviruses have helicase activity has been obtained so far. The role of the 60-kDa protein in the induction of vesicles and in attaching the replication complex to the membranes has been discussed in Section v.A. It is not known whether the NTBM is essential for these activities.


3. VPg

Several groups of plus-strand RNA viruses have been shown to possess a genome-linked protein (VPg). The function of VPg has not been definitively established. For poliovirus it has been suggested that VPg is involved in RNA replication as a primer (Wimmer, 1982; Takegami et al., 1983; Crawford and Baltimore, 1983) and/or as a nuclease during hairpin-primed synthesis (Tobin et al., 1989). In a membranous replication complex, endogenous poliovirus VPg can be uridylylated in vitro to VPg-pUpU, which can then be extended to form larger RNA molecules (Takegami et al., 1983). This suggests that the uridylylated protein may serve as a primer for RNA-dependent RNA synthesis. So far it has not been possible to detect uridylylated VPg (precursors) in CPMV-infected cells.

Initially, the RNA-I-encoded 60-kDa protein was proposed to be the direct precursor for VPg (Zabel et al., 1982; Goldbach et al., 1982). However, later studies on processing of putative VPg precursors in vitro revealed that trans cleavage of the Gln/Ser site between the 58 kDa and the VPg is extremely inefficient (Peters et al., 1992a). Furthermore, kinetic studies on the processing of the 170-kDa protein have shown that the 60- and the 58-kDa proteins appear to be produced simultaneously, a pattern of accumulation inconsistent with their having a precursor-product relationship (Dessens and Lomonossoff, 1992; Peters et al., 1992b). Moreover, large amounts of the 60-kDa protein are found in fractions prepared from CPMV-infected cowpea leaves (Peters et al., 1992a), and this does not point toward a VPg precursor function for the 60-kDa protein. In line with the finding that in vitro cleavages in the RNA-l polyprotein occur most efficiently in cis, the 112-kDa processing intermediate (VPg + 110 K) was found to function as a VPg precursor either directly or via a 26-kDa (VPg + 24 K) processing intermediate (S. A. Peters et al. unpublished data).

If VPg is involved in a protein-primed mechanism to initiate RNA replication, then the 112-kDa viral replicase precursor is a likely candidate to start this event via cis cleavage into VPg and the 110-kDa protein. Subsequently or maybe concomitantly, VPg becomes uridylylated and might serve as a primer from which RNA is elongated by the 110-kDa viral replicase. Because the serine residue that links the VPg to the RNA is part of a cleavage site, it is also possible that release of VPg at its N-terminus and linkage to RNA are coupled reactions. It has been suggested that VPg also serves as a signal for encapsidation of virion RNA (Reuer et al., 1990); however, here too the experimental proof is lacking for CPMV.

4. The RNA-2-Encoded 58-kDa Protein

RNA-2 must be replicated by the RNA-I-encoded replication proteins. Deletion analysis has shown that large parts of the coding region of RNA-2 are not essential for replication. However, mutants in which translation of


the 105-kDa protein was disrupted replicated very poorly (Wellink and Van Kammen, 1989; Holness et a1., 1989; Rohll et a1., 1993). Further analysis showed that the (approx. 10 kDa long) N-terminal part of the 58-kDa protein of RNA-2, rather than its nucleotide sequence, is required for replication of RNA-2 (Van Bokhoven et a1., 1993b). The N-terminus of the 58-kDa protein probably functions in cis only, because replication of mutants that are not able to produce 58-kDa protein could not be complemented by co-inoculation with normal CPMV RNA.

A screening of protein databases did not reveal any significant amino acid homology of the 58-kDa protein with other known proteins. The N-terminus of the 58-kDa protein is not conserved among the comoviruses BPMV, CPSMV, RCMV, and CPMV, except for the presence of many hydrophobic and aromatic amino acid residues (Chen and Bruening, 1992a). Hydrophobic and aromatic amino acid motifs are often found in members of the highly heterologous "family" of RNA-binding proteins (Kenan et a1., 1991). Therefore, the 58-kDa protein encoded by RNA-2 may be involved in RNA binding, perhaps binding specifically to RNA-2, and enabling it to be recognized as a template for the replication proteins encoded by RNA-I. The requirement for the 58-kDa protein in RNA-2 replication implies that only initiation of translation at nucleotide 161, resulting in the synthesis of the 105-kDa polyprotein, will lead to replication of RNA-2. Replication of RNA-2 will not follow upon initiation of translation at nucleotide 512, which results in the synthesis of the 95-kDa polyprotein. The alternative translation initiation sites at nucleotides 161 or 512 may provide a regulatory switch for either (1) replication of RNA-2, or (2) the production of the capsid proteins and the 48-kDa movement protein. The relative amounts of the 58- and 48-kDa proteins in CPMV-infected plants suggest that initiation of translation occurs more often at nucleotide 512 than at nucleotide 161 (Rezelman et a1., 1989). Most translation events on RNA-2 will thus not be followed by replication. This is not necessarily surprising since RNA-2 must be translated 120 times to produce sufficient coat proteins to encapsidate one progeny molecule each of RNA-l and RNA-2. Also, RNA-l needs to be translated more often than it is replicated because the production of one progeny molecule of each genomic RNA requires at least four molecules of VPg and consequently the synthesis of four 200-kDa polyproteins.

C. Signals on the RNA Molecules

Because both comovirus genomic RNA species are produced by the same RNA-I-encoded replication machinery, it may be expected that they share features that have a function in RNA replication. Indeed, their 5' NTRs have extensive sequence homology and so do their 3' NTRs. The first 50 nucleotides in the two 5' leader sequences of CPMV show 86% homology and are able to fold into two similar stem-loop structures (Fig. 4). The last 65 nucleo-


tides preceding the poly(A) tail show 82 % homology and also are able to fold into two similar stem-loop structures (Fig. 5) (Eggen et a1., 1989a). Especially striking is a stretch of 11 nucleotides, UUUUAUUAAAA, in the 3' ends of both viral RNA molecules. This sequence is complementary to a stretch of 11 bases in the 5' ends of the molecules (allowing one G-U base pairing), indicating that the (+)- and (- I-strands of both RNA molecules have very similar stretches of nucleotide sequence at their 3' ends. This sequence may represent a recognition sequence for the viral RNA replicase. Indeed, in vitro mutagenesis of this sequence in hairpin n at the 3' end of RNA-1 severely reduced virus infectivity (Fig. 5) (Eggen et a1., 1989a).

By deletion analysis, Rohll et a1. (1993) showed that signals required for RNA-2 replication are located in the 5'-terminaI524 nucleotides and in the 3'-terminal 151 nucleotides. Site-directed mutagenesis of the two minor stems of the hairpin I structure in the 3' NTR (Fig. 5) indicate that the structure of the molecule rather than its sequence is the important determinant for efficient replication. The right arm of the structure seems to be particularly important in determining the level of RNA accumulation (Rohll et a1., 1993).

I

IT (t.G,-0.5 keal/mol)

'-A' ,iJ A\ :U A: \A A/ , ,

:U - A :U - A IU - A. iu - A

CPMV RNA-1

IT (t. G'-1.1ke~l/mol)

CPMV RNA-2

FIGURE 5. Secondary structure predictions of the 3' regions of CPMV RNA-1 and RNA-2 consisting of the noncoding regions immediately preceding the poly(A) tail and some residues of the poly(A) tail. The solid line in I represents an identical structure in the two RNA molecules, except for a U-A to CoG change in hairpin B (shown by open circles). The interrupted line in hairpin II represents a conserved stretch of 11 nucleotides at the 3' ends of RNA-1 and RNA-2 which might represent a signal in viral RNA replication. Thermodynamic parameters were calculated according to Freier et al. (1986). A program developed for the prediction of pseudoknots in RNA (Abrahams et al., 1990) and the Zuker program (Zuker and Stiegler, 1981) were used.


Exchange of the 5' and 3' noncoding regions of RNA-2 with those of RNA-I maintained the capacity of these RNAs to be replicated, in trans (Van Bokhoven et al., I993a). Thus, however important the 5' and 3' non coding regions of RNA-2 may be in RNA replication, they do not determine ability to be replicated, in trans. Furthermore, these results make it very likely that only the homologous terminal sequences of the two RNA molecules are important for replication.

In vitro mutagenesis studies have shown that RNA-I without a poly(A) tail but with five Cs at the 3' end is not infective, whereas RNA-I with only four A residues followed by five nonviral nucleotides is infective, although poorly (Eggen et al., I989a). Sequence analysis has shown that plants infected with transcripts having poly(A) tails of about 50 nucleotides contain progeny RNA with poly(A) tails of variable length (Eggen et al., I989b). Therefore, the poly(A) tail cannot be copied exactly from the poly(U) stretch during plusstrand synthesis. The extension could be the consequence of RNA polymerase slipping on a poly(U) template. Another possibility is that the poly(A) tail is extended by a terminal nucleotidyl transferase.

D. Replication of the CPMV RNA Species Is Linked to Their Translation

Van Bokhoven et al. (1993b) found that defined mutants of RNA-I could not be replicated by the viral functions provided by wild-type RNA-I. These results suggest that CPMV RNA-I functions as a template for RNA replication only if replication proteins are synthesized from the same RNA molecule, i.e., replication of RNA-I occurs in cis only.

If there is a tight linkage between translation and replication for CPMV RNA-I, the question arises how replication of RNA-2 (which must occur in trans) is achieved. Remarkably, Van Bokhoven et al. (1993b) have also reported a linkage between replication and translation for RNA-2. They found that the N-terminal part of the 58-kDa protein of RNA-2, which is required for replication of RNA-2 (see Section VI.B.4), probably functions in cis only, because replication of mutants that are not able to produce 58-kDa protein could not be complemented by co-inoculation with wild-type CPMV RNA. Thus it seems that replication of RNA-2 depends on translation of the 58-kDa polypeptide from the very same RNA molecule, suggesting that the translating ribosomes transport the N-terminal domain of the 58-kDa protein (contained in the 105-kDa polyprotein) to the 3' end of the RNA molecule. Then a ribonucleoprotein complex occasionally may be formed, consisting o·f the 105-kDa polyprotein, viral RNA, and possibly ribosomal factor(s). This complex could then be recognized by the RNA-I-encoded replicative machinery to start negative-strand RNA synthesis. For RNA-I, the observed linkage between translation and replication may be effected


similarly by transportation of the replication proteins to the 3' end of the RNA (see Section VI.E).

A remarkable observation is that RNA-2 mutants containing long 3' NTRs replicate very poorly (Van Bokhoven et a1., 1993b). It is possible that the stability of these RNA molecules is reduced. Alternatively, the linkage between translation and replication might be hampered by such long 3' NTRs interfering with the formation of a ribonucleoprotein complex as described above.

E. A Model for CPMV RNA Replication

1. Formation of the Replication Complex

The principal conclusions of the preceding sections have been incorporated in a model (Fig. 6) for CPMV RNA replication, which proposes an important role for proteolytic processing in the establishment of a functional replication complex and so accounts for the observed linkage between translation and replication. When a RNA-l molecule is translated in the infected cell, a 200-kDa polyprotein is synthesized, which is rapidly cleaved into the 32- and 170-kDa proteins (Fig. 6a). The 32- and 170-kDa proteins are associated with each other by interaction of the 32-kDa protein with the hydrophobic domain of the 58-kDa protein that is contained within the 170-kDa protein. The same protein-protein interaction inhibits further proteolytic processing of the 170-kDa protein into smaller cleavage products (Peters et a1., 1992a). In these cells, the 60-kDa domain of the 170-kDa protein induces proliferation of membranes, possibly from the rough endoplasmic reticulum, which will lead eventually to the formation of the vesicular membranes characteristic of CPMV-infected cells. It should be noted that at this stage most, if not all, of the 32- and 170-kDa proteins may not be associated with membranes but reside probably in the electron-dense structures that appear in the cytoplasm of the cell early in infection.

The replication of RNA-l depends on translation of the replicative proteins from the very same RNA molecule. Hence, the synthesized proteins must recognize the 3' end of the RNA from which they were translated. How this is accomplished is not known, but a host factor, possibly a ribosomal protein, may be involved in this process. The involvement of such a factor in viral RNA replication is also found with bacteriophage QI3 (Blumenthal and Carmichael, 1979) and possibly also with cucumber mosaic virus (Quadt et a1., 1991). Following translation of CPMV RNA-I, a ribonucleoprotein complex may occasionally be formed, consisting of the interacting viral 32-and 170-kDa proteins, viral RNA, and possibly a ribosomal factor(s), which becomes attached to membranous vesicles (Fig. 6b). The anchoring to membranes is mediated by the hydrophobic domain within the 58-kDa protein of RNA-I, which is also the site of interaction with the 32-kDa protein. It is

I RNA-II

~ ~'" ~ {ljj -'5 {ljj3

® IRNA-21 /' 3 ~ ~

- '5 ~ ~ t

membranous vc!Wcies declron·dcnsc stnK:lwtS

@

- ~~~~~~~~~~~~ --. ® ©

~M~MM s ~~M ~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~

@

FIGURE 6. A model for CPMV RNA replication. (al Translation of the viral RNAs. The 200-kDa polyprotein encoded by RNA-l is cleaved during its synthesis into the 32-kDa and 170-kDa proteins. The 32-kDa protein remains associated with the 170-kDa protein by interaction with the hydrophobic domain of the 58-kDa protein that is contained within the 170-kDa protein, thereby inhibiting further processing of the 170-kDa protein. The RNA-I-encoded proteins are kept together in the cytoplasm of the infected cell and will induce the formation of electrondense structures and membranous vesicles. Occasionally, the 32-kDa/170-kDa proteins are arrested at the 3' end of the RNA-l molecule, possibly by virtue of a host factor (represented by wavy lines I. Initiation of translation of RNA-2 at nucleotide position 161 will result in synthesis of the 105-kDa polyprotein. The 58-kDa protein, which is contained in the 105-kDa polyprotein, or the entire 105-kDa polyprotein may bind to the 3' end of RNA-2, possibly also with the involvement of a host protein. Subsequently, the 32-kDa/170-kDa protein complex recognizes the ribonucleoprotein complex at the 3' end of RNA-2. (bl Docking of the prereplication complex in vesicular membranes. The hydrophobic domain of the RNA-I-encoded 58-kDa protein enables the template RNA and 32-kDa/170-kDa proteins to become membrane-bound. The membrane association of the 58-kDa protein abolishes the 32-kDa/170-kDa interaction. The 170-kDa protein is then further processed and a functional RNA replication complex is formed. (cl Synthesis of complementary (-I-strand RNA. VPg, represented by a filled circle, is likely to be involved in initiation of RNA synthesis. Elongation of the initiated (-I-strand RNA involves the RNA polymerase activity of the llO-kDa protein and the RNA helicase activity of the RNA-2-encoded 58-kDa protein. (dl Termination of (-I-strand RNA synthesis and replication complex formation for (+ I-strand RNA synthesis. A new set of 32-kDa/170-kDa proteins is recruited to the 5' end of the (+ 1-RNA strand, where they recognize a host protein bound to a 5' stem-loop structure. A membrane-bound replication complex may then be formed, just as has been described for (-I-strand RNA synthesis. (el Synthesis of genomic RNAs. When (+ I-strand RNA synthesis proceeds, the host protein may be translocated to the newly synthesized 5' stemloop structure. A new set of 32-kDa/170-kDa proteins is attracted and formation of yet another replication complex occurs. Translocation of the host protein is repeated several times, thus allowing multiple (+ I-strand RNAs to be synthesized on one (-I-strand RNA template. (Figure kindly provided by Dr. H. Van Bokhoven.1


proposed that the association of the hydrophobic domain with membranes thus abolishes the 32-kDa/58-kDa protein interaction, thereby triggering further proteolytic processing of the 170-kDa protein and enabling the completion of the formation of a replication complex (Fig. 6c). The functional membrane-bound replication complex consists of RNA-l template associated with viral replication proteins (cleavage products) and possibly a (ribosomal) host factor(s).

The formation of a replication complex for RNA-2 may be mediated in an analogous manner, requiring the binding of the RNA-2-encoded 58-kDa polypeptide at the 3' end of the RNA molecule. As with RNA-I, a ribonucleoprotein complex may then be formed, consisting of RNA-2, the RNA-2-encoded 58-kDa protein (or its precursor), and possibly a ribosomal protein(s). This complex is recognized by the 32-kDa/170-kDa proteins of RNA-l and subsequently directed to the membranous vesicles. The formation of a functional RNA-2 replication complex then proceeds in the same way as for RNA-I.

The possible involvement of a ribosomal host protein in the formation of the membrane-bound replication complex is speculative and needs to be proved. It has been found recently that CPMV RNA replication can occur in protoplasts of a wide variety of plant species (Wellink et a1., 1993a), indicating that any host proteins that may be involved in viral RNA replication are well conserved during evolution. However, this observation does not exclude the possibility that host proteins may be engaged at other stages of viral RNA replication, such as membrane proliferation or RNA initiation.

2. CPMV RNA Synthesis

The mechanism of initiation of ( - )-strand CPMV RNA synthesis is still one of the most obscure steps in CPMV viral RNA replication. Because VPg is found at the 5' end of each progeny RNA molecule, it is very likely that VPg or one of its precursors is involved in the initiation step. By analogy with poliovirus, it is most likely that uridylylated VPg, or one of its precursors, functions as a primer in RNA synthesis (Takeda et a1., 1986). However, the hairpin primer model of Flanegan and co-workers (Young et a1., 1985; Tobin et a1., 1989), which is based on the activity of a host terminal uridylyl transferase activity, cannot be excluded as yet. Once initiation of CPMV RNA synthesis has occurred, RNA elongation proceeds, and this very likely involves the RNA polymerase activity of the llO-kDa protein (Dorssers et a1., 1984). The putative RNA helicase activity of the 58-kDa protein encoded by RNA-l (or a precursor) may also be required at this stage for unwinding secondary structures in the template RNA and/or preventing the formation of double-stranded structures.

Once the synthesis of the complementary (- )-strand RNA has been completed the (- )-strand can be used as a template to produce multiple


progeny (+ )-strands. However, at some point during synthesis of the (-)strand, probably during initiation, VPg must be attached to the 5' end of the newly synthesized ( - )-strand and, consequently, proteolytic processing of a new 200-kDa polyprotein must occur to synthesize VPg-linked (+ )-strand RNA. Hence, for each new strand of VPg-linked RNA to be synthesized, an entire new set of replicative proteins is produced. Therefore, Goldbach and Van Kammen (1985) proposed that the viral replication proteins are used only once and that an entire new set of RNA-I-encoded proteins (32-170 kDa) is recruited at the 3' end of the (- )-strand RNA template. Because the (-)strand RNA template is not translated, the formation of a replication complex for (+ )-strand RNA synthesis must proceed differently than for (-)strand RNA synthesis. This raises the question of how the replication proteins encoded by RNA-l recognize the (- )-strand RNA as a template for (+)strand RNA synthesis. A clue as to how this might be accomplished is provided by the important observation by Andino et al. (1990) that the synthesis of poliovirus (+ )-strand RNA was dependent on a stem-loop structure at the 5' end of the (+ )-sense RNA. It was found that the association of a host protein(s) with this 5' stem-loop structure and the subsequent binding of poliovirus proteins 3C and 3D, or their precursor protein 3CD, was required for (+ )-strand RNA synthesis. Likewise, CPMV ( + )-strand RNA synthesis may also be dependent on a host protein, which may have been used already for (- )-strand synthesis, that recognizes a structure at the 5' end of the (+ )-strand RNA (Fig. 6d and e). This ribonucleoprotein structure is then recognized by the 32-170 kDa proteins of RNA-l and, upon proteolytic processing of the 170-kDa protein, a functional replication complex can be formed. Initiation of ( + )-strand RNA synthesis may then proceed in a similar way to ( - )-strand synthesis, using as a template the (-)-RNA that is in close proximity with the activated replication proteins. Evidence that the 5' end of ( + )-sense RNA is involved in viral (+ )-strand RNA synthesis has also been obtained for brome mosaic virus, a member of the alphaviruslike supergroup (Pogue and Hall, 1992). It was suggested by these authors that the binding of host proteins to structures at the 5' end of ( + )-sense RNA may be a general feature in the ( + )-strand RNA synthesis of many other viruses. Interestingly, structural comparison of the 5' NTRs of the viral RNA molecules of the comoviruses CPMV, RCMV, BPMV, and CPSMV reveals a striking conservation of two 5'-terminal stem-loop structures, whereas other parts of the 5' NTRs have diverged widely (Fig. 4).

In Section IV it was mentioned that the 170-kDa protein can be cleaved into three different sets of proteins, 58 kDa/112 kDa, 60 kDa/110 kDa, and 84 kDa/87 kDa. The 112-,110-, and 87-kDa proteins each contain the conserved polymerase domain. Therefore it is possible that CPMV uses these alternative cleavages of the 170-kDa protein to make different replication complexes that might provide the virus with a mechanism to regulate RNA synthesis, perhaps at the level of template selection.

At present the actual link between translation and replication of CPMV


RNA is not clear. It seems reasonable to assume that it involves the formation of the complex that initiates negative-strand synthesis. One of the replication proteins involved in this process may only function in cis. This may be difficult to reconcile with replication of RNA-2 (by RNA-I-encoded proteins) unless one assumes that the RNA-2-encoded 58-kDa (105-kDa) protein can convert a cis-acting protein to a trans-acting one. Alternatively, as proposed above, ribosomes are involved because they "transport" either the replicative proteins or an essential host factor (of ribosomal origin?) to the 3' end of the RNA. It is also possible that RNA has to interact with ribosomes in order to make it a suitable template for replication. Irrespective of the actual mechanism, the advantage of this translation-linked mode of replication is that it results in a positive selection of replication-competent mutants.

VII. INTERCELLULAR TRANSPORT OF COMOVIRUSES

A. Cell-to-Cell Movement Using Virus-Induced Tubules

Upon initial entry into the plant cell, the virus particle dissociates and replication of the genome starts to produce progeny virus. Systemic infection of the host plant depends completely on the ability of the (progeny) virus to move from the initially infected cell into neighboring cells. It is generally accepted that plant viruses use the plasmodesmata for cell-to-cell movement (Gibbs, 1976). However, because of the small size exclusion limit of plasmodesmata, they must be modified to allow virus progeny to pass. In recent years, several studies have revealed the involvement in this process of virusencoded movement proteins. Two different types of movement mechanism have been identified (Hull, 1989; Goldbach et a1., 1990; Maule, 1991). One mechanism involves the transport of the viral genome in a "nonvirion" form assisted by its movement protein as exemplified by tobacco mosaic virus (TMV). The other mechanism involves the movement of complete virus particles and this mechanism is exemplified by the comoviruses. In this case the desmotubule is removed and specific virus-induced tubules are assembled in the plasmodesmata, through which virus particles move from one cell to the other (Fig. 7) (Van Lent et a1., 1991).

CPMV RNA-l is able to replicate independently from RNA-2 in isolated protoplasts, but RNA-2 is required for systemic infection of a plant (Goldbach et a1., 1980; Rezelman et a1., 1982). This indicates that the proteins encoded by RNA-2 are involved in virus spread through the plant. CPMV RNA-2 codes for two sets of proteins, the overlapping 58-kDa/48-kDa proteins and the capsid proteins (Fig. 3). Using infective transcripts, Wellink and Van Kammen (1989) showed that deletions in the coding regions of the capsid

COMOVIRUSES: MOLECULAR BIOLOGY

---------virus particles

ER

ER ---... _-... "'-

65

FIGURE 7. (a) Electron micrograph and (b) a schematic interpretation of the tubular structure induced by CPMV and involved in cell-to-cell movement of virus particles. ER, endoplasmic reticulum; PM, plasmomembrane

proteins and the 58-kDa/48-kDa proteins prevented systemic infection in cowpea plants. Apparently the capsid proteins and the 58-kDa/48-kDa proteins are all required for cell-to-cell movement of the virus, and CPMV probably moves as virus particles.

The 48-kDa protein has been detected in the pellet fraction of extracts of CPMV-infected protoplasts and in the culture medium (Wellink et a1., 1987b), whereas the 58-kDa protein was found to be present mainly in the cytoplasmic fraction (Rezelman et a1., 1989). The 58-kDa/48-kDa proteins and the capsid proteins were located more precisely in sections of infected plants and protoplasts by immunogold labeling. Using antibodies against the 58-kDa/48-kDa proteins, Van Lent et a1. (1990) showed that the 58-kDa and/ or 48-kDa proteins were located in the virus-induced tubular structures that are formed in the plasmodesmata of CPMV-infected plant cells. These tubular structures contain virus particles, as was confirmed by labeling with antiCPMV serum (Van Lent et a1., 1991). Similar results were obtained for RCMV (Shanks et a1., 1989). The essential requirement of the 58-kDa/48-kDa proteins for cell-to-cell movement, as found by deletion analysis and the localization studies, provide firm evidence that these tubular structures are involved in cell-to-cell movement of CPMV.


B. The 48-kDa Protein Induces Tubules on Protoplasts

Structures remarkably similar to those formed in intact cells were also found in infected cowpea protoplasts, which possess neither cell walls nor plasmodesmata (Van Lent et a1., 1991). The virus-containing tubules extended into the medium and were enveloped by the plasma membrane. Apparently the tubule has a functional polarity, being formed within the protoplast toward an external target, which in tissue would be the neighboring cell to be infected.

This finding made it possible to study the effects of mutations in the 58-kDa/48-kDa proteins and the capsid proteins on the formation of these structures. It was found that a mutant with a deletion in the 58-kDa/48-kDa proteins as well as a mutant that failed to produce the 48-kDa protein but still produced the 58-kDa protein were not able to form tubules (Kasteel et a1., 1993). This suggests that at least the 48-kDa protein is essential for this process. A mutant that failed to produce both coat proteins was still capable of forming tubules, but of course these did not contain virus particles (Kasteel et a1., 1993). Therefore, the coat proteins do not have a role in the formation of the tubule wall.

Using a transient expression system based on the cauliflower mosaic virus 35S promoter, it was shown that expression of only the 48-kDa protein sufficed for the formation of (empty) tubules (Wellink et a1., 1993a). Whether the 48-kDa protein, which can induce a protoplast to form structures up to 20 f.Lm long, is the sole component of the tubule wall or whether host components are also involved remains to be established. Protoplasts from plant species that are not hosts for CPMV, like tomato, carrot, and barley, are, upon expression of the 48-kDa protein, also capable of forming tubules (Wellink et a1., 1993a). This indicates that tubule formation is not restricted to species that are hosts of CPMV and that putative host components involved in the process must be strongly conserved proteins (e.g., cytoskeletal proteins).

C. General Remarks

As already mentioned, at least two different mechanisms exist for movement of virus progeny from cell to cell. One type allows the movement of a viral RNA protein complex for which the coat protein is not required as has been described for tobamoviruses (Atabekov and Dorokhov, 1984; Takamatsu et a1., 1987), and the other allows the movement of virus particles through tubular structures as described above for comoviruses.

Malyshenko et a1. (1988) have reported the complementation of transport of the RNA-l of RCMV by sunn-hemp mosaic tobamovirus. Apparently the tobamovirus movement protein can facilitate movement of unencapsidated RCMV RNA-I, indicating that the tobamovirus-type of transport mechanism is not very specific. There is no evidence for complementation in


the reverse combination, but it seems probable that the transport mechanism involving tubules is much more specific and does not allow passage of heterologous viruses.

The experimental evidence indicates that the 48-kDa protein induces formation of the tubular structures involved in cell-to-cell movement of CPMV. Whether the tubules assemble in the cytoplasm and subsequently associate with and extend through plasmodesmata or assemble directly in the plasmodesmata is not clear. Also, a possible role for the RNA-2-encoded 58-kDa protein, which is also involved in RNA-2 replication (see Section VI), in dissolving or removing the desmotubules in existing plasmodesmata or in the induction of newly formed, modified plasmodesmata cannot be ruled out.

It is also still an open question whether virus particles actually move through the tubule or whether the "top" of the tubule "dissolves" in the neighboring cell, thus releasing the virus particles in the cytoplasm. Later in infection, callose/cell wall material is deposited against the tubules (Van Lent et a1., 1990). This may be a defense reaction of the plant to confine the virus in the tubules. If this were achieved, it could be a mechanism by which a plant could prevent virus movement. The induction of tubules in protoplasts from nonhost plant species is in line with this reasoning (Wellink et a1., 1993a).

Tubular structures also are found in tissues infected with caulimoviruses, nepoviruses, or tospoviruses (Perbal et a1., 1993; Roberts and Harrison, 1970; Wieczorek and Sanfac;on, 1993; Kormelink et a1., 1984). The CPMV 48-kDa protein shows limited amino acid sequence homology with the movement protein of caulimoviruses but not with the putative movement protein of nepoviruses (Koonin et a1., 1991). Remarkably, comovirus and nepovirus movement proteins may be distantly related to the HSP90 protein, a chaperone protein known to interact with the cytoskeletal framework (Koonin et a1., 1991; Sanchez et a1., 1988), suggesting that comovirus and nepovirus movement proteins have chaperonelike activity.

VIII. CONCLUDING REMARKS

The data presented in this chapter illustrate the progress in applying molecular and plant cell biological techniques that has led, during the past 10 years, to a rather detailed knowledge of the molecular biology of comoviruses, especially of the type species, CPMV. The availability of full-length cDNA clones of both genome segments of CPMV, from which infective transcripts can be produced in vitro, has allowed the production of desired mutants, which have assisted in the further unraveling of processes such as translational initiation, proteolytic processing of the viral polyproteins, the viral RNA replication process, and cell-to-cell movement. Crystallography studies have led to detailed understanding of the molecular structure of the


comoviral protein capsid. Many questions still remain unresolved, however, and certainly need further study. For instance, a true template-dependent in vitro RNA replication system is still lacking and, as a consequence, both the template requirements and the possible involvement of host proteins in this process remain largely unresolved. This is partly due to the observed strict interdependence of translation and replication during the viral multiplication cycle as discussed in Section VI. Another major drawback in the study of comoviruses so far has been the lack of a suitable transformable host system, which would lend itself to complementation and virus-host interaction studies. This has been the consequence of the narrow host ranges of most comoviruses, which are usually restricted to a few legume species, for which easy transformation and regeneration protocols have not yet been developed. This certainly holds for the best-studied examples of comoviruses, including CPMV. It might therefore be wise to extend comovirus research to viruses like, e.g., Andean potato mottle virus, which includes solanaceous plants such as potato and Nicotiana clevelandii in its host natural range, to further unravel the molecular aspects of comovirus-host interactions.

ACKNOWLEDGMENTS. The authors wish to thank Hans van Bokhoven and Jan van Lent for their critical comments and helpful suggestions, Kees Pleij for the RNA folding analysis presented in Figs. 3 and 5, George Lomonossoff for communicating results prior to publication, Piet Kostense for making some of the illustrations, and Rina Hartman for preparing the manuscript. The research on cowpea mosaic virus at the Departments of Molecular Biology and Virology at the Wageningen Agricultural University is supported by the Netherlands Foundations for Chemical and Biological Research (SON/SLW) with financial aid from the Netherlands Organization for the Advancement of Pure Research (NWO).

REFERENCES

Abrahams, J. P., Van den Berg, M., Van Batenburg, E., and Pleij, c., 1990, Prediction of RNA secondary structure, including pseudoknotting, by computer simulation, Nucleic Acids Res. 18:3035.

Agrawal, H., 1964, Identification of cowpea mosaic virus isolates, Meded. Landbouwhogesch. Wageningen 64(5):l.

Ahlquist, P., and Kaesberg, P., 1979, Determination of the length distribution of poly A at the 3'-terminus of the virion RNAs of EMC virus, poliovirus, rhinovirus, RAV-61 and CPMV and of mouse globin mRNA, Nucleic Acids Res. 7:1195.

Andino, R., Rieckhof, G. E., and Baltimore, D., 1990, A functional ribonucleoprotein complex forms around the 5' end of poliovirus RNA, Ce1163:369.

Argos, P., Kamer, G., Nicklin, M. J. H., and Wimmer, E., 1984, Similarity in gene organization and homology between proteins of animal picornaviruses and plant comovirus suggest a common ancestry of these virus families, Nucleic Acids Res. 12:725l.

Assink, A. M., Swaans, H., and Van Kammen, A., 1973, The localization of virus-specific doublestranded RNA of cowpea mosaic virus in subcellular fractions of infected Vigna leaves, Virology 53:384.


Atabekov, J. G., and Dorokhov, Y. 1., 1984, Plant virus-specific transport function and resistance of plants to viruses, Adv. Virus Res. 29:313.

Bancroft, J. B., 1962, Purification and properties of bean pod mottle virus and associated centrifugal and electrophoretic components, Virology 16:419.

Bazan, J. F., and Fletterick, R. J., 1988, Viral cysteine proteases are homologous to the trypsin-like family of serine proteases: Structural and functional implications, Proc. Natl. Acad. Sci. USA 85:7872.

Beier, H., Issinger, D. G., Deuschle, M., and Mundry, K. W., 1981, Translation of the RNA of cowpea severe mosaic virus in vitro and in cowpea protoplasts, r. Gen. Virol. 54:379.

Belsham, G. J., and Lomonossoff, G. P., 1991, The mechanism of translation of cowpea mosaic virus middle component RNA: No evidence for internal initiation from experiments in an animal cell transient expression system, r. Gen. Virol. 72:3109.

Bienz, K., Egger, D., and Pasamontes, 1., 1987, Association of polioviral proteins of the P2 genomic region with the viral replication complex and virus-induced membrane synthesis as visu<llized by electron microscopic immunocytochemistry and autoradiography, Virology 160:220.

Blevings, S., and Stace-Smith, R., 1976, In vivo and in vitro effects on the electrophoretic forms of particles of broad bean true mosaic virus, r. Gen. Virol. 31:199.

Blumenthal, T., and Carmichael, G. G., 1979, RNA replication: Function and structure of QI3-replicase, Annu. Rev. Biochem. 48:525.

Bruening, G., 1977, Plant covirus systems: Two component systems, in: Comprehensive Virology, Vol. lllH. Fraenkel-Conrat and R. R. Wagner, eds.), pp. 55-141, Plenum Press, New York.

Bruening, G., 1978, Comovirus group, CMI/AAB Descriptions of Plant Viruses No. 199. Bruening, G., 1981, Biochemistry of plant viruses, in: The Biochemistry of Plants, A Comprehen

sive Treatise, Vol. 6, Proteins and Nucleic Acids IA. Marcus, ed.), pp. 571-631, Academic Press, New York.

Bruening, G., and Agrawal, H. 0., 1967, Infectivity of a mixture of cowpea mosaic virus ribonucleoprotein components, Virology 32:306.

Brunt, A. A., Phillips, S., Barton, R. J., and Jones, R. A. C., 1984, Ullucus virus C, CMI/AAB Descriptions of Plant Viruses No. 177.

Bu, M., and Shih, D. S., 1989, Inhibition of proteolytic processing of the polyproteins of cowpea mosaic virus by hemin, Virology 173:348.

Chen, X., and Bruening, G., 1992a, Nucleotide sequence and genetic map of cowpea severe mosaic virus RNA 2 and comparisons with RNA 2 of other comoviruses, Virology 187:682.

Chen, X., and Bruening, G., 1992b, Cloned DNA copies of cowpea severe mosaic virus genomic RNAs: Infectious transcripts and complete nucleotide sequence of RNA 1, Virology 191:607.

Chen, Z., Stauffacher, c., Li, Y., Schmidt, T., Bomu, W., Kamer, G., Shanks, M., Lomonossoff, G., and Johnson, J. E., 1989, Protein-RNA interactions in an icosahedral virus at 3.0 A resolution, Science, NY 245:154.

Crawford, N. M., and Baltimore, D., 1983, Genome-linked protein VPg of poliovirus is present as free VPg and VPG-pUpU in poliovirus-infected cells, Proc. Natl. Acad. Sci. USA 80:7452.

Daubert, S. D., and Bruening, G., 1979, Genome-associated proteins of comoviruses, Virology 98:246.

Daubert, S. D., Bruening, G., and Najarian, R. C., 1978, Protein bound to the genome RNAs of cowpea mosaic virus, Eur. r. Biochem. 92:45.

Davies, J. W., Aalbers, A. M. J., Stuik, E. J., and Van Kammen, A., 1977, Translation of cowpea mosaic virus RNA in a cell-free extract from wheat germ, FEBS Lett. 77:265.

De Jager, c. P., 1976, Genetic analysis of cowpea mosaic virus mutants by supplementation and reassortment tests, Virology 70:151.

Dessens, J. T., and Lomonossoff, G. P., 1991, Mutational analysis of the putative catalytic triad of the cowpea mosaic virus 24K protease, Virology 184:738.

Dessens, J. T., and Lomonossoff, G. P., 1992, Sequence upstream of the 24K protease enhances cleavage of the cowpea mosaic virus B-RNA-encoded polyprotein at the junction between the 24K and 87K proteins, Virology 189:225.


De Varennes, A., Lomonossoff, G. P., Shanks, M., and Maule, A. L 1986, The stability of cowpea mosaic virus VPg in reticulocyte lysates, ,. Gen. Virol. 67:2347.

De Zoeten, G. A., Assink, A. M., and Van Kammen, A., 1974, Association of cowpea mosaic virus-induced double-stranded RNA with a cytopathological structure in infected cells, Virology 59:341.

Dorssers, 1., Zabel, P., Van der Meer, L and Van Kammen, A, 1982, Purification of a hostencoded RNA-dependent RNA-polymerase from cowpea mosaic virus-infected cowpea leaves, Virology 116:236.

Dorssers, 1., Van der Meer, J., Van Kammen, A., and Zabel, P., 1983, The cowpea mosaic virus RNA replication complex and the host-encoded RNA-dependent RNA-polymerase-template complex are functionally different, Virology 125:155.

Dorssers, 1., Van der Krol, S., Van der Meer, J., Van Kammen, A., and Zabel, P., 1984, Purification of cowpea mosaic virus RNA replication complex: Identification of a virus-encoded no,ooodalton polypeptide responsible for RNA chain elongation, Proc. Natl. Acad. Sci. USA 81:1951.

Eggen, R., and Van Kammen, A, 1988, RNA replication in comoviruses, in: RNA Genetics, Vol. 1 (P. Ahlquist, J. Holland, and E. Domingo, eds.l, pp. 49-69, CRC Press, Boca Raton, F1.

Eggen, R., Verver, J., Wellink, J., De Jong, A, Goldbach, R., and Van Kammen, A., 1989a, Improvements of the infectivity of in vitro transcripts from cloned cowpea mosaic virus cDNA: Impact of terminal nucleotide sequences, Virology 173:447.

Eggen, R., Verver, J., Wellink, J., PleiL K., Van Kammen, A., and Goldbach, R., 1989b, Analysis of sequences involved in cowpea mosaic virus RNA replication using site-specific mutants, Virology 173:456.

EI Manna, M. M., and Bruening, G., 1973, Polyadenylate sequences in the ribonucleic acids of cowpea mosaic virus, Virology 56:198.

Franssen, H., 1984, Proteolytic processing of the primary translation products of cowpea mosaic virus RNAs, PhD Thesis, Agricultural University Wageningen.

Franssen, H., Goldbach, R., Broekhuysen, M., Moerman, M., and Van Kammen, A., 1982, Expression of middle-component RNA of cowpea mosaic virus: In vitro generation of a precursor to both capsid proteins by a bottom-component RNA-encoded protease from infected cells, ,. Virol. 41:8.

Franssen, H., Goldbach, R., and Van Kammen, A., 1984a, Translation of bottom component RNA of cowpea mosaic virus in reticulocyte lysate: Faithful proteolytic processing of the primary translation product, Virus Res. 1:39.

Franssen, H., Moerman, M., Rezelman, G., and Goldbach, R., 1984b, Evidence that the 32,000-Dalton protein encoded by bottom-component RNA of cowpea mosaic virus is a proteolytic processing enzyme, J. Virol. 50:183.

Franssen, H., Leunissen, L Goldbach, R., Lomonossoff, G., and Zimmern, D., 1984c, Homologous sequences in non-structural proteins from cowpea mosaic virus and picornaviruses, EMBO J. 3:855.

Franssen, H., Roelofsen, w., and Goldbach, R., 1986, Mapping of the coding regions for the capsid proteins of cowpea mosaic virus on the nucleotide sequence of middle component RNA, Chinese J. Virol. 2:46.

Freier, S. M., Kierzek, R., Jaeger, J. A., Sugimoto, N., Caruthers, M. H., Neilson, E., and Turner, O. H., 1986, Improved free-energy parameters for predictions of RNA duplex stability, Proc. Natl. Acad. Sci. USA 83:9373.

Gabriel, C. L Derrick, K. S., and Shih, D. S., 1982, The synthesis and processing of the proteins of bean pod mottle virus in rabbit reticulocyte lysates, Virology 122:476.

Garcia, J. A., Schrijvers, 1., Tan, A, Vos, P., Wellink, J., and Goldbach, R., 1987, Proteolytic activity of the cowpea mosaic virus encoded'24K protein synthesized in Escherichia coli, Virology 159:67.

Geelen, J. 1. M. c., 1974, Structuur en eigenschappen van cowpea-mozaiekvirus, PhD Thesis, Agricultural University Wageningen.


Geelen, J. 1. M. c., Van Kammen, A, and Verduin, B. J. M., 1972, Structure of the capsid of cowpea mosaic virus. The chemical subunit: Molecular weight and number of subunits per particle, Viralogy 49:205.

Gibbs, A J., 1976, Viruses and plasmodesmata, in: Intercellular Communication in Plants: Studies on Plasmodesmata IB. E. S. Gunning and A. W. Robards, eds.), pp. 149-164, SpringerVerlag, Berlin.

Goldbach, R. w., 1986, Molecular evolution of plant RNA viruses, Annu. Rev. Phytopatho1. 24:289.

Goldbach, R., and Krijt, J., 1982, Cowpea mosaic virus-encoded protease does not recognize primary translation products of M-RNAs from other comoviruses, r Vira1. 43:115l.

Goldbach, R., and Rezelman, G., 1983, Orientation of the cleavage map of the 200 kilo dalton polypeptide encoded by the bottom-component RNA of cowpea mosaic virus, r Vira1. 46:614.

Goldbach, R., and Van Kammen, A., 1985, Structure, replication and expression of the bipartite genome of cowpea mosaic virus, in: Molecular Plant Virology, Vol. 21J. W. Davies, ed.l, pp. 83-120, CRC Press, Boca Raton, F1.

Goldbach, R., and Wellink, J., 1988, Evolution of plus strand RNA viruses, Interviralogy 29:260. Goldbach, R., Rezelman, G., and Van Kammen, A, 1980, Independent replication and expression

of B-component RNA of cowpea mosaic virus, Nature 286:297. Goldbach, R., Rezelman, G., Zabel, P., and Van Kammen, A., 1982, Expression of the bottom

component RNA of cowpea mosaic virus: Evidence that the 60 kilodalton VPg-precursor is cleaved into single VPg and a 58 kilodalton polypeptide, r Vira1. 42:360.

Goldbach, R., Eggen, R., De Jager, C., Van Kammen, A., Van Lent, J., Rezelman, G., and Wellink, J., 1990, Genetic organisation evolution and expression of plant viral RNA genomes, in: Recognition and Response in Plant Virus Interactions Nato ASI Series: Cell Biology Vol. 41 IR. S. S. Fraser, ed.l, pp. 147-162, Springer-Verlag, Berlin.

Gorbalenya, A. E., and Koonin, E. v., 1989, Viral proteins containing the purine NTP-binding sequence pattern, Nucleic Acids Res. 17:8413.

Gorbalenya, A. E., Blinov, V. M., and Koonin, E. v., 1985, Prediction of nucleotide-binding properties of virus-specific proteins from their primary structure, Molec. Genetika 11:30.

Gorbalenya, A. E., Koonin, E. V., Donchenko, A. P., and Blinov, V. M., 1988, A conserved NTPmotif in putative helicases, Nature 333:22.

Gorbalenya, A. E., Donchenko, A P., Blinov, V. M., and Koonin, E. V., 1989a, Cysteine proteases of positive strand RNA viruses and chymotrypsin-like serine proteases, FEBS Lett. 243:103.

Gorbalenya, A E., Koonin, E. V., Donchenko, A P., and Blinov, V. M., 1989b, A novel superfamily of nucleoside triphosphate-binding motif containing proteins which are probably involved in duplex unwinding in DNA and RNA replication and recombination, FEBS Lett. 235:16.

Gorbalenya, A E., Koonin, E. V., and Wolf, Y. I., 1990, A new superfamily of putative NTPbinding domains encoded by genomes of small DNA and RNA viruses, FEBS Lett. 262:145.

Hewlett, M. J., Rose, J. K., and Baltimore, D., 1976, 5' -Terminal structure of poliovirus polysomal RNA is pUp, Prac. Nat1. Acad. Sci. USA 73:327.

Hibi, T., Rezelman, G., and Van Kammen, A., 1975, Infection of cowpea mesophyll protoplasts with cowpea mosaic virus, Viralogy 64:308.

Hodgman, T. c., 1988, A new superfamily of replicative proteins, Nature 333:22. Holness, C. 1., Lomonossoff, G. P., Evans, D., and Maule, A. J., 1989, Identification of the

initiation codons for translation of cowpea mosaic virus middle component RNA using sitedirected mutagenesis of an infectious cDNA clone, Viralogy 172:31l.

Hull, R., 1989, The movement of viruses in plants, Annu. Rev. Phytopatho1. 27:213. Inokuchi, v., and Hirashima, A., 1987, Interference with viral infection by defective RNA

replicase, r Vira1. 61:3946. Jablonski, S. A., Luo, M., and Morrow, C. D., 1991, Enzymatic activity of poliovirus RNA

polymerase mutants with single amino acid changes in the conserved YGDD amino acid motif, r Vira1. 65:4565.


Jaegle, M., Wellink, J., and Goldbach R., 1987, The genome-linked protein of cowpea mosaic virus is linked to the 5'-terminus of Band M RNA by a phosphodiester linkage to serine, J. Gen. Virol. 68:627.

Kamer, G., and Argos, P., 1984, Primary structural comparison of RNA-dependent polymerases from plant, animal and bacterial viruses, Nucleic Acids Res. 12:7269.

Kasteel, D., Wellink, J., Verver, J., Van Lent, J., Goldbach, R., and Van Kammen, A., 1993, The involvement of cowpea mosaic virus M RNA encoded proteins in tubule formation, T. Gen. Virol. 74:172l.

Kenan, D. J., Query, C. c., and Keene, J. D., 1991, RNA recognition: Towards identifying determinants of specificity, Trends Biochem. Sci. 16:214.

King, A. M. Q., Lomonossoff, G. P., and Ryan, M. D., 1991, Picornaviruses and their relatives in the plant kingdom, Semin. Virol. 2:1l.

Koonin, E. V., 1991, The phylogeny of RNA-dependent RNA polymerases of positive-strand RNA viruses, J. Gen. Virol. 71:2197.

Koonin, E. V., Mushegian, A. R., Ryabov, E. v., and Dolja, V. v., 1991, Diverse group of plant RNA and DNA viruses share related movement proteins that may possess chaperone-like activity, J. Gen. Virol. 72:2895.

Kormelink, R., Storms, M., Van Lent, J., Peters, D., and Goldbach, R., 1994, Expression and subcellular location of the NSM protein of tomato spotted wilt virus (TSWV), a putative viral movement protein, Virology 200:56.

Kozak, M., 1984, Compilation and analysis of sequences upstream from the translational start site in eukaryotic mRNAs, Nucleic Acids Res. 12:857.

Kridl, J. c., and Bruening, G., 1984, Comparison of capsids and nucleocapsids from cowpea mosaic virus-infected cowpea protoplasts and seedlings, Virology 129:369.

Lain, S., Riechmann, J. 1., Martin, M. T., and Garcia, G. A., 1989, Homologous potyvirus and flavivirus belonging to a superfamily of helicase-like proteins, Gene 82:357.

Lain, S., Riechmann, J. 1., and Garcia, J. A., 1990, RNA helicase: A novel activity associated with a protein encoded by a positive strand RNA virus, Nucleic Acids Res. 18:7003.

Lain, S., Martin, M. T., Riechmann, J. 1., and Garcia, J. A., 1991, Novel catalytic activity associated with positive-strand RNA virus infection: Nucleic acid-stimulated ATPase activity of the plum pox potyvirus helicase-like protein, J. Virol. 65:l.

Lomonossoff, G. P., and Shanks, M., 1983, The nucleotide sequence of cowpea mosaic virus B-RNA, EMBO r. 2:2253.

Lomonossoff, G., Shanks, M., and Evans, D., 1985, The structure of cowpea mosaic virus replicative form RNA, Virology 144:35l.

MacFarlane, S. A., Shanks, M., Davies, J. W., Zlotnick, A., and Lomonossoff, G. P., 1991, Analysis of the nucleotide sequence of bean pod mottle virus middle component RNA, Virology 183:405.

Malyshenko, S. I., Lapchic, 1. G., Kondakova, O. A., Kuznetzova, 1. 1., Taliansky, M. E., and Atabekov, J. G., 1988, Red clover mottle comovirus B-RNA spreads between cells in tobamovirus-infected tissues, J. Gen. Virol. 69:407.

Maule, A., 1991, Virus movement in infected plants, Crit. Rev. Plant Sci. 9:457. Mazzone, H. M., Incardona, N. 1., and Kaesberg, P., 1962, Biochemical and biophysical properties

of squash mosaic virus and related macromolecules, Biochim. Biophys. Acta 55:164. McLain, 1., Porta, c., Lomonossoff, G., and Dimmock, N. J., 1995, Human immunodeficiency

type I virus neutralizing antibodies raised to a gp41 peptide expressed on the surface of a plant virus, AIDS Res. Human Retroviruses 11:327.

Molla, A., Paul, A. V., and Wimmer, E., 1991, Cell-free de novo synthesis of poliovirus, Science, NY 254:1647.

Niblett, C. 1., and Semancik, J. S., 1969, Conversion of the electrophoretic forms of cowpea mosaic virus in vivo and in vitro, Virology 38:685.

Nickerson, K. W., and Lane, 1. c., 1977, Polyamine content of several RNA plant viruses, Virology 81:455.


Nida, D. 1., Anjos, J. R., Lomonossoff, G. P., and Ghabrial, S. A., 1992, Expression of cowpea mosaic virus coat protein precursor in transgenic tobacco plants, J. Gen. Virol. 73:157.

Nomoto, A., Detjen, D., Pozzatti, R., and Wimmer, E., 1977, The location of the poliogenome protein in viral RNAs and its implication for RNA synthesis, Nature 268:208.

Oxelfelt, P., 1976, Biological and physiochemical characteristics of three strains of red clover mottle virus, Virology 74:73.

Partridge, J. E., Shannon, 1. M., Gumpf, D. J., and Colbaugh, P., 1974, Glycoprotein in the capsid of plant viruses as a possible determinant of seed transmissibility, Nature 247:391.

Pelham, H. R. B., 1979, Synthesis and proteolytic processing of cowpea mosaic virus proteins in reticulocyte lysates, Virology 96:463.

Pelham, H. R. B., and Jackson, R. J., 1976, An efficient mRNA dependent translation system from reticulocyte lysates, Eur. J. Biochem. 67:247.

Peng, X. X., and Shih, D. S., 1984, Proteolytic processing of the proteins translated from the bottom component RNA of cowpea mosaic virus. The primary and secondary cleavage reactions, J. Biol. Chern. 259:3197.

Perbal, M. c., Thomas, C. 1., and Maule, A. J., 1993, Cauliflower mosaic virus gene 1 product (PI) forms tubular structures which extend from the surface of infected protoplasts, Virology 195:281.

Peters, S. A., Voorhorst, W. G. B., Wery, J., Wellink, J., and Van Kammen, A., 1992a, A regulatory role for the 32kD protein in proteolytic processing of cowpea mosaic virus polyproteins, Virology 191:81.

Peters, S. A., Voorhorst, W. G. B., Wellink, J., and Van Kammen, A., 1992b, Processing of VPgcontaining polyproteins encoded by the B RNA from cowpea mosaic virus, Virology 191:90.

Peters, S. A., Verver, J., Nollen, E. A. A., Van Lent, J. W. M., Wellink, J., and Van Kammen, A., 1994, The NTP-binding motif in cowpea mosaic virus B polyprotein is essential for viral replication, J. Gen. Virol. 75:3167.

Poch, 0., Sauvaget, T., Delaren, M., and Tordo, N., 1989, Identification of four conserved motifs among the RNA-dependent polymerase encoding elements, EMBO J. 8:3867.

Pogue, G. P., and Hall, T. C.,1992, The requirement for a 5' stem-loop structure in brome mosaic virus replication supports a new model for viral positive-strand RNA initiation, J. Virol. 66:674.

Porta, c., Spall, V. E., Loveland, J., Johnson, J. E., Barker, P. J., and Lomonossoff, G. P., 1994, Development of cowpea mosaic virus as a high-yielding system for the presentation of foreign peptides, Virology 202:949.

Quadt, R., Rosdorff, H. J. M., Hunt, T. M., and Jaspars, E. M. J., 1991, Analysis of the protein composition of alfalfa mosaic virus RNA-dependent RNA polymerase, Virology 182:309.

Reuer, 0., Kuhn, R., and Wimmer, E., 1990, Characterization of poliovirus clones containing lethal and nonlethal mutations in the genome-linked protein: VPg, J. Virol. 64:2967.

Rezelman, G., Goldbach, R., and Van Kammen, A., 1980, Expression of bottom component RNA of cowpea mosaic virus in cowpea protoplasts, J. Virol. 36:366.

Rezelman, G., Franssen, H. J., Goldbach, R. W., Ie, T. S., and Van Kammen, A., 1982, Limits to the independence of bottom component RNA of cowpea mosaic virus, J. Gen. Virol. 60:335.

Rezelman, G., Van Kammen, A., and Wellink, J., 1989, Expression of cowpea mosaic virus M RNA in cowpea protoplasts, J. Gen. Virol. 70:3043.

Rice, R. V., Lindberg, G. D., Kaesberg, P., Walker, 1- C., and Stahmann, M. A., 1965, The three components of squash mosaic virus, Phytopathology 45:145.

Richards, O. C., Eggen, R., Goldbach, R., and Van Kammen, A., 1989, High-level synthesis of cowpea mosaic virus RNA polymerase and protease in Escherichia coli, Gene 78:135.

Roberts, I. M., and Harrison, B. D., 1970, Inclusion bodies and tubular structures in Chenopodium amaranticolor plants infected with strawberry latent ringspot virus, J. Gen. Virol. 7:47.

Rohll, J. B., Holness, C. 1., Lomonossoff, G. P., and Maule, A. J., 1993, 3 /-Terminal nucleotide sequences important for the accumulation of cowpea mosaic virus M RNA, Virology 193:672.


Rossmann, M. G., and Johnson, J. E., 1989, Icosahedral RNA virus structure, Annu. Rev. Biochern. 58:533.

Rottier, P. J. M., Rezelman, G., and Van Kammen, A., 1979, The inhibition of cowpea mosaic virus replication by actinomycin D, Virology 92:299.

Rottier, P. J. M., Rezelman, G., and Van Kammen, A., 1980, Protein synthesis in cowpea mosaic virus-infected cowpea protoplasts: Detection of virus-related proteins, r Gen. Virol. 51:359.

Sanchez, E. R., Redmond, T., Scheerer, L. L., Bresnick, E. H., Welch, W. J., and Pratt, W. B., 1988, Evidence that the 90-kilodalton heat shock protein is associated with tubulin-containing complex in L cell cytosol and in intact PtK cells, Mol. Endocrinol. 2:756.

Semancik, J. S., 1966, Studies on electrophoretic heterogeneity in isometric plant viruses, Virology 30:968.

Shanks, M., and Lomonossoff, G. P., 1992, The nucleotide sequence of red clover mottle virus bottom component RNA, r Gen. Virol. 73:2473.

Shanks, M., Stanley, J., and Lomonossoff, G. P., 1986, The primary structure of red clover mottle virus middle component RNA, Virology 155:697.

Shanks, M., Tomenius, K., Clapham, D., Huskisson, N. S., Barker, P. J., Wilson, I. G., Maule, A. J., and Lomonossoff, G. P., 1989, Identification and subcellular localization of a putative cellto-cell transport protein from red clover mottle virus, Virology 173:400.

Siler, D. J., Babcock, J., and Bruening, G., 1976, The electrophoretic mobility and enhanced infectivity of a mutant of cowpea mosaic virus, Virology 71:560.

Sonenberg, N., 1991, Picornavirus RNA translation continues to surprise, Trends Genet. 7:105. Stanley, J., Rottier, P., Davies, J. W., Zabel, P., and Van Kammen, A., 1978, A protein linked to the

5' -termini of both RNA components of the cowpea mosaic virus genome, Nucleic Acids Res. 5:4505.

Stauffacher, C. v., Usha, R., Harrington, M., Schmidt, T., Hosur, M., and Johnson, J. E., 1987, The structure of cowpea mosaic virus at 3.5A resolution, in: Crystallography in Molecular Biology (D. Moras, J. Drenth, B. Strandberg, D. Suck, and K. Wilson, eds.), pp. 293-308, Plenum Press, New York.

Takamatsu, N., Ishikawa, M., Meshi, T., and Okada, Y., 1987, Expression of bacterial chloramphenicol acetyltransferase gene in tobacco plants mediated by TMV-RNA, EMBO r 6:307.

Takeda, N., Kuhn, R. J., Yand, C-F., Takegami, T., and Wimmer, E., 1986, Initiation of poliovirus plus-strand RNA synthesis in a membrane complex of infected HeLa cells, r Virol. 60:43.

Takegami, T., Kuhn, R. J., Anderson, C. W., and Wimmer, E., 1983, Membrane-dependent uridylylation of the genome-linked protein VPg of poliovirus, Proc. Natl. Acad. Sci. USA 80:7447.

Thomas, A. A. M., Ter Haar, E., Wellink, J., and Voorma, H. 0., 1991, Cowpea mosaic virus middle component RNA contains a sequence that allows internal binding of ribosomes and that requires eukaryotic initiation factor 4F for optimal translation, r Virol. 65:2953.

Tobin, G. J., Young, D. c., and Flanegan, J. B., 1989, Self-catalyzed linkage of poliovirus terminal protein VPg to poliovirus RNA, Cell 59:511.

Usha, R., Rohll, J. B., Spall, V. E., Shanks, M., Maule, A. J., Johnson, J. E., and Lomonossoff, G. P., 1993, Expression of an animal virus antigenic site on the surface of a plant virus particle, Virology 197:366.

Van Bokhoven, H., Wellink, J., Usmany, M., Vlak, J. M., Goldbach, R., and Van Kammen, A., 1990, Expression of plant virus genes in animal cells: High level synthesis of cowpea mosaic virus B-RNA encoded proteins with baculovirus expression vectors, J. Gen. Viral. 71:2509.

Van Bokhoven, H., Mulders, M., Wellink, 1-, Vlak, J. M., Goldbach, R., and Van Kammen, A., 1991, Evidence for dissimilar properties of comoviral and picornaviral RNA polymerases, r Gen. Virol. 72:567.

Van Bokhoven, H., Van Lent, J. W. M., Custers, R., Vlak, J. M., Wellink, J., and Van Kammen, A., 1992, Synthesis of the complete 200K polyprotein of cowpea mosaic virus B RNA in insect cells, r Gen. Virol. 73:2775.


Van Bokhoven, H., Verver, J., Wellink, J., and Van Kammen, A., 1993a, Protoplasts transiently expressing the 200kD coding sequence of cowpea mosaic virus B RNA support replication of M RNA, r. Gen. Viro1. 74:2233.

Van Bokhoven, H., Le Gall, 0., Kasteel, D., Verver, J., Wellink, J., and Van Kammen, A., 1993b, Cis- and trans-acting elements in cowpea mosaic virus RNA replication, Virology 195:377.

Van der Meer, J., Dorssers, 1., and Zabel, P., 1983, Antibody-linked polymerase assay on protein blots: A novel method for identifying polymerases following SDS-polyacrylamide gel electrophoresis, EMBO r. 2:233.

Van der Meer, J., Dorssers, 1., Van Kammen, A., and Zabel, P., 1984, The RNA-dependent RNA polymerase of cowpea is not involved in cowpea mosaic virus RNA replication: Immunological evidence, Virology 132:413.

Van Dyke, T. A., and Flanegan, J. B., 1980, Identification of poliovirus polypeptide P63 as a soluble RNA-dependent RNA polymerase, r. Viro1. 35:732.

Van Kammen, A., 1967, Purification and properties of the components of cowpea mosaic virus, Virology 31:633.

Van Kammen, A., 1968, The relationship between the components of cowpea mosaic virus. I. Two ribonucleoprotein particles necessary for the infectivity of CPMV, Virology 34:312.

Van Lent, J. w. M., Wellink, J., and Goldbach, R., 1990, Evidence for the involvement of the 58K and 48K proteins in the intercellular movement of cowpea mosaic virus, r. Gen. Viro1. 71:219.

Van Lent, J., Storms, M., Van der Meer, E, Wellink, J., and Goldbach, R., 1991, Tubular structures involved in movement of cowpea mosaic virus are also formed in infected cowpea protoplasts, r. Gen. Viro1. 72:2615.

Van Wezenbeek, P., Verver, J., Harmsen, J., Vos, P., and Van Kammen, A., 1983, Primary structure and gene organisation of the middle-component RNA of cowpea mosaic virus, EMBO r. 2:94l.

Verver, J., Goldbach, R., Garcia, J. A., and Vos, P., 1987, In vitro expression of a full-length DNA copy of cowpea mosaic virus B RNA: Identification of the B RNA encoded 24-kd protein as a viral protease, EMBO r. 6:549.

Verver, J., Le Gall, 0., Van Kammen, A., and Wellink, J., 1991, The sequence between nucleotides 161 and 512 of cowpea mosaic virus M RNA is able to support internal initiation of translation in vitro, r. Gen. Viro1. 72:2339.

Virudachalam, R., Harrington, M., Johnson, J. E., and Markley, J. 1., 1985, lH, 13C, and 31p nuclear magnetic resonance studies on cowpea mosaic virus: Detection and exchange of polyamines and dynamics of the RNA, Virology 141:43.

Vos, P., Verver, J., Van Wezenbeek, P., Van Kammen, A., and Goldbach, R., 1984, Study of the genetic organisation of a plant viral RNA genome by in vitro expression of a full-length DNA copy, EMBO 1- 3:3049.

Vos, P., Jaegle, M., Wellink, J., Verver, J., Eggen, R., Van Kammen, A., and Goldbach, R., 1988a, Infectious RNA transcripts derived from full-length DNA copies of the genomic RNAs of cowpea mosaic virus, Virology 165:33.

Vos, P., Verver, J., Jaegle, M., Wellink, J., Van Kammen, A., and Goldbach, R., 1988b, Two viral proteins involved in the proteolytic processing of the cowpea mosaic virus polyproteins, Nucleic Acids Res. 16:1967.

Walker, J. E., Saraste, M., Runswick, M. J., and Gay, N. J., 1982, Distantly related sequences in the a- and b-subunits of ATP-synthetase, myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold, EMBO r. 1:945.

Wellink, J., and Van Kammen, A., 1989, Cell-to-cell transport of cowpea mosaic virus requires both the 58K/48K proteins and the capsid proteins, r. Gen. Virol. 70:2279.

Well ink, J., Rezelman, G., Goldbach, R., and Beyreuther, K., 1986, Determination of the proteolytic processing sites in the polyprotein encoded by the bottom-component RNA of cowpea mosaic virus, r. Virol. 59:50.

Wellink, J., Jaegle, M., Prinz, H., Van Kammen, A., and Goldbach, R., 1987a, Expression of the middle component RNA of cowpea mosaic in vivo, r. Gen. Virol. 68:2577.


Wellink, J., Jaegle, M., and Goldbach, R., 1987b, Detection of a novel protein encoded by the bottom-component RNA of cowpea mosaic virus, using antibodies raised against a synthetic peptide, ,. Viral. 61:263.

Wellink, J., Van Lent, J., and Goldbach, R., 1988, Detection of viral proteins in cytopathic structures in cowpea protoplasts infected with cowpea mosaic virus, ,. Gen. Viral. 69:751.

Wellink, J., Van Lent, J. W. M., Verver, J., Sijen, T., Goldbach, R. W., and Van Kammen, A., 1993a, The cowpea mosaic virus M RNA-encoded 48K protein is responsible for induction of tubular structures in protoplasts, ,. Virol. 67:3660.

Wellink, J., Verver, J., and Van Kammen, A., 1993b, Mutational analysis of AUG codons of cowpea mosaic virus M RNA, Biochimie 75:741.

Wieczorek, A., and Sanfa~on, H., 1993, Characterization and subcellular location of tomato ringspot nepovirus putative movement protein, Virology 194:743.

Wimmer, E., 1982, Genome-linked proteins of viruses, Cell 28:199. Wu, G.-J., and Bruening, G., 1971, Two proteins from cowpea mosaic virus, Virology 45:596. Young, D. C., Tuschall, D. M., and Flanegan, J. B., 1985, Poliovirus RNA-dependent RNA

polymerase and host cell protein synthesize product RNA twice the size of poliovirion RNA in vitro, ,. Virol. 54:256.

Zabel, P., Weenen-Swaans, H., and Van Kammen, A., 1974, In vitro replication of cowpea mosaic virus RNA. I. Isolation and properties of the membrane-bound replicase, J. Viral. 14:1049.

Zabel, P., Moerman, M., Van Straaten, F., Goldbach, R., and Van Kammen, A., 1982, Antibodies against the genome-linked protein VPg of cowpea mosaic virus recognize a 60,000 dalton precursor polypeptide, J. Viral. 41:1083.

Zabel, P., Moerman, M., Lomonossoff, G., Shanks, M., and Beyreuther, K., 1984, Cowpea mosaic virus VPg: Sequencing of radiochemically modified protein allows mapping of the gene on B RNA, EMBO J. 3:1629.

Zuker, M., and Stiegler, P., 1981, Optimal computer folding of large RNA sequences using thermodynamics and auxiliary information, Nucleic Acids Res. 9:133.

CHAPTER 4

Comoviruses: Transmission, Epidemiology, and Control R. C. GERGERICH AND fl. A. SCOTT

1. TRANSMISSION AND EPIDEMIOLOGY

A. Modes of Virus Transmission

l. Beetle Transmission

Most comoviruses have been shown to be transmitted efficiently by leaffeeding beetles in the families Chrysomelidae, Coccinellidae, Curculionidae, or Meloidae. Table I lists the comoviruses, with acronyms used in this chapter, and the beetle species that have been shown to transmit them. Beetle transmission of glycine mosaic virus (GMV), pea mild mosaic virus (PMiMV), and ullucus virus C (UVe) has not been demonstrated. Comoviruses can be acquired or transmitted by beetles immediately upon initiation of feeding, although higher frequencies of transmission occur with prolonged feeding. It was once thought that virus transmission by beetles is a simple mechanical process involving contamination of the mouthparts of the beetle (Smith, 1965). More recent evidence, however, indicates a complex type of interaction between plant viruses and their beetle vectors (Fulton et a1., 1980j Gergerich and Scott, 1991). The interaction between beetles and the viruses they transmit has been characterized as circulative because many

R. C. GERGERICH AND H. A. SCOTT • Department of Plant Pathology, University of Arkansas, Fayetteville, Arkansas 72701.

77

Vir

us I

Acr

onym

)

And

ean

pota

to m

ott

le v

irus

IA

PMV

) B

ean

pod

mo

ttle

vir

us

IBPM

V)

Bea

n ru

gose

mos

aic

viru

s IB

RM

V)

Bro

ad b

ean

stai

n v

irus

IB

BSV

)

Bro

ad b

ean

tru

e m

osai

c vi

rus

IBB

TM

V)

TA

BL

E 1

. B

eetl

e V

ecto

rs o

f C

om

ov

iru

ses

Vec

tor

spec

ies

Dia

brot

ica

viri

dula

Cer

otom

a tr

ifurc

ata

Col

aspi

s fla

vida

C

olas

pis

lata

D

iabr

otic

a ba

ltea

ta

Dia

brot

ica

unde

cim

punc

tata

how

ardi

E

pica

uta

vitt

ata

Epi

lach

na v

ariv

esti

s C

erot

oma

rufj.

com

is

Dia

brot

ica

adel

pha

Dia

brot

ica

balt

eata

A

pion

aes

tivu

m

Api

on a

ethi

ops

Api

on v

orax

Si

tona

lin

eatu

s A

pion

aet

hiop

s A

pion

vor

ax

Sito

na l

inea

tus

Ref

eren

ces

Aba

d an

d S

alaz

ar,

unpu

blis

hed

lin

Avi

la e

t aI

., 19

84)

Ros

s 11

963)

; W

alte

rs 1

1964

); W

alte

rs a

nd L

ee 1

1969

); H

orn

et

ai.

1197

0);

Moo

re a

nd S

cott

119

71);

Wal

ters

et

ai.

1197

2);

San

derl

in 1

1973

); F

ulto

n an

d S

cott

119

74);

Mil

brat

h et

ai.

1197

5);

Pat

el a

nd P

itre

119

76)

Ho

rn e

t ai

. 11

970)

H

orn

et

ai.

1197

0)

Ho

rn e

t ai

. 11

970)

R

oss

1196

3);

Hor

n et

ai.

1197

0);

Mil

brat

h et

al.

1197

5)

Pat

el a

nd P

itre

119

71)

Ful

ton

and

Sco

tt 1

1974

) G

amez

119

72);

Car

tin

and

Gam

ez 1

1973

) G

amez

119

72);

Car

tin

and

Gam

ez 1

1973

) G

amez

119

72);

Car

tin

an

d G

amez

119

73)

Coc

kbai

n et

ai.

1197

5)

Coc

kbai

n et

ai.

1197

5)

Coc

kbai

n et

ai.

1197

5)

Coc

kbai

n et

ai.

1197

5)

Coc

kbai

n et

ai.

1197

5)

Coc

kbai

n et

al.

1197

5)

Coc

kbai

n et

ai.

1197

5)

Cow

pea

mos

aic

viru

s (C

PMV

)

Cow

pea

seve

re m

osai

c vi

rus

(CPS

MV

)

Gly

cine

mo

saic

vir

us (

GM

V)

Pea

mil

d m

osa

ic v

irus

(PM

iMV

)

Aca

lym

ma

vitt

ata

Cer

otom

a tr

ifur

cata

D

iabr

otic

a ba

ltea

ta

Dia

brot

ica

unde

cim

punc

tata

how

ardi

D

iabr

otic

a vi

rgife

ra

Nem

atoc

erus

ace

rbus

O

othe

ca m

utab

ilis

P

aral

uper

odes

qua

tern

us

Aca

lym

ma

vitt

ata

Cer

otom

a ar

cuat

a

Cer

otom

a as

trof

asci

ata

Cer

otom

a ru

fico

rnis

Cer

otom

a tr

ifur

cata

Dia

brot

ica

adel

pha

Dia

brot

ica

balt

eata

D

iabr

otic

a sp

ecio

sa

Dia

brot

ica

unde

cim

punc

tata

how

ardi

D

iabr

otic

a vi

rgife

ra

Epi

lach

na v

ariv

esti

s G

ynan

drob

roti

ca v

aria

bili

s Sy

sten

a sp

. U

nk

no

wn

U

nk

no

wn

Jans

en a

nd S

tapl

es (

1971

) Ja

nsen

and

Sta

ples

(19

71)

Jans

en a

nd S

tapl

es (

1971

) Ja

nsen

and

Sta

ples

(19

71)

Jans

en a

nd S

tapl

es (

1971

) W

hitn

ey a

nd G

ilm

er (

1974

) C

han

t (1

959)

; B

ock

(197

1);

Wh

itn

ey a

nd G

ilm

er (

1974

) W

hitn

ey a

nd G

ilm

er (

1974

) Ja

nsen

and

Sta

ples

(19

71)

Deb

rot

and

Roj

as (

1967

); C

ost

a et

al.

(197

8);

Anj

os a

nd L

in (

1984

); L

in e

t al

. (1

984)

V

alve

rde

et a

l. (1

978)

D

ale

(194

9, 1

953)

; D

ebro

t an

d R

ojas

(19

67);

Per

ez a

nd

Cor

tes-

Mon

Hor

(1

970)

; K

vica

la e

t al

. (1

973)

; V

alve

rde

et a

l. (1

978,

198

2)

Sm

ith

(19

24);

Wal

ters

an

d B

arne

tt (

1964

); J

anse

n an

d S

tapl

es (

1970

a,

1971

); S

ande

rlin

(19

73);

Ful

ton

and

Sco

tt (

1974

); M

cLau

ghli

n et

al.

(197

8)

Val

verd

e et

al.

(197

8)

Jans

en a

nd S

tapl

es (

1971

); V

alve

rde

et a

l. (1

982)

L

in e

t al

. (1

984)

Ja

nsen

and

Sta

ples

(19

70a,

197

1)

Jans

en a

nd S

tapl

es (

1971

) Ja

nsen

and

Sta

ples

(19

70b)

; F

ult

on

and

Sco

tt (

1974

); V

alve

rde

et a

l. (1

978)

V

alve

rde

et a

l. (1

978)

G

amez

, un

publ

ishe

d (i

n F

ulto

n et

al.,

197

5)

(con

tinu

ed)

Vir

us (

Acr

onym

)

Qua

il p

ea m

osai

c vi

rus

(QPM

V)

Rad

ish

mos

aic

viru

s (R

aMV

)

Red

clo

ver

mo

ttle

vir

us

(RC

MV

) S

quas

h m

osai

c vi

rus

(SM

V)

Ull

ucus

Vir

us C

(U

Ve)

TA

BL

E I

. (C

onti

nued

)

Vec

tor

spec

ies

Cer

otom

a ru

flcor

nis

Cer

otom

a tr

ifurc

ata

Dia

brot

ica

adel

pha

Dia

brot

ica

balt

eata

D

iabr

otic

a un

deci

mpu

ncta

ta h

owar

di

Epi

lach

na v

ariv

esti

s G

ynan

drob

roti

ca v

aria

bilis

P

aran

apia

caba

wat

erho

usei

D

iabr

otic

a un

deci

mpu

ncta

ta

unde

cim

punc

tata

E

pitr

ix h

irti

penn

is

Phy

llot

reta

cru

cife

rae

Phy

llot

reta

sp.

P

hyll

otre

ta s

trio

lata

A

pion

apr

ican

s A

pion

var

ipes

A

caly

mm

a th

iem

ei t

hiem

ei

Aca

lym

ma

triv

itta

ta

Aca

lym

ma

vitt

ata

Atr

anch

ya m

enet

ries

i A

ulac

opho

ra f

emor

alis

D

iabr

otic

a ba

ltea

ta

Dia

brot

ica

long

icor

nis

Dia

brot

ica

unde

cim

punc

tata

how

ardi

D

iabr

otic

a un

deci

mpu

ncta

ta

unde

cim

punc

tata

D

iabr

otic

a vi

rgife

ra

Epi

lach

na a

dmir

abil

is

Epi

lach

na c

fuys

omel

ina

Epi

lach

na s

pars

aori

enta

lis

Unk

now

n

Hob

bs (

1981

) M

oore

(19

73)

Hob

bs (

1981

) H

obbs

(19

81)

Mei

ners

et

al.

(197

7)

Mei

ners

et

al.

(197

7)

Hob

bs (

1981

) H

obbs

(19

81)

Ref

eren

ces

Cam

pbel

l an

d C

olt

(196

7);

Frei

tag,

unp

ubli

shed

(in

Wal

ters

, 19

69)

Frei

tag,

unp

ubli

shed

(in

Wal

ters

, 19

69)

Frei

tag,

unp

ubli

shed

(in

Wal

ters

, 19

69)

Cam

pbel

l an

d C

olt

(196

7)

Toc

hiha

ra (

1968

) G

erha

rdso

n an

d P

ette

rson

(19

74)

Ger

hard

son

and

Pet

ters

on (

1974

) L

astr

a (1

968)

F

reit

ag (

1941

a,b,

195

6)

Sit

terl

y (1

960)

; S

tone

r (1

963)

Y

oshi

da e

t al

. (1

980)

Y

oshi

da e

t al

. (1

980)

S

itte

rly

(196

0)

Sto

ner

(196

3)

Sit

terl

y (1

960)

; S

tone

r (1

963)

F

reit

ag (

1941

a,b,

195

6)

Sto

ner

(196

3)

Yos

hida

et

al.

(198

0)

Coh

en a

nd N

itza

ny (

1963

); L

ockh

art

et a

l. (1

982)

Y

oshi

da e

t al

. (1

980)

COMOVIRUSES: TRANSMISSION 81

beetle-transmitted viruses move into the hemolymph of their vectors after ingestion (Freitag, 1956 j Slack and Scott, 1971). In vector species such as the bean leaf beetle (Cerotoma trifurcata) and the spotted cucumber beetle (Diabrotica undecimpunctata howardi), southern bean mosaic sobemovirus (SBMV) moves from the digestive system into the circulatory system through the midgut (Wang et a1., 1994). However, the comovirus bean pod mottle virus (BPMV) does not enter the hemolymph of these species, although it is efficiently transmitted by them (Wang et a1., 1992). Moreover, neither SBMV nor BPMV move into the hemolymph of the Mexican bean beetle (Epilachna varivestis), an efficient vector of these viruses (Wang et a1., 1992). As SBMV and BPMV are similar in size and shape, this selective movement through the midgut wall suggests a specific interaction between the virus and some component of the midgut that is required for virus movement through the midgut wall. Because ingested comoviruses are not always found in the hemolymph of their beetle vectors, it appears that virus circulation in the beetle is not invariably an important part of the transmission process, and that in some cases the relation between the beetle and the virus is similar to that found in the foregut-borne type of transmission described for viruses transmitted in the semipersistent manner by aphids (anthriscus yellows virusj Murant et a1., 1976) and leafhoppers (maize chlorotic dwarf virusj Nault and Ammar, 1989). The gradual decrease in the amount of BPMV in beetles during test feeding following virus acquisition (Ghabrial and Schultz, 1983) is taken to indicate that the virus does not multiply in the beetle.

The transmission of comoviruses by beetles is characterized by short acquisition and inoculation access periods, an apparent lack of a latent period, and short persistence times in beetles that are actively feeding (Table II). Beetles can acquire virus very quickly from infected tissue, and where attempts have been made to determine the minimum time required for virus acquisition, it has been found that virus can be acquired on the first bite (Fulton et a1., 1980).

Virus transmission occurs as a result of feeding activity during which the beetle regurgitates virus-containing fluid. Plant viruses that are not trans-

TABLE II. Characteristics of Vector Transmission of some Comoviruses

Minimum Minimum acquisition inoculation

Virus Vector access period access period Persistence Reference

CPSMV Cerotoma <5 min <1 hr Up to 6 days Dale (1953) ru{icornis

SMV Acalymma <5 min <5 hr Up to 17 days Freitag (1956) trivittata

BBTMV Apion vorax <0.5 hr <1 day Up to 8 days Cockbain et al. (1975)

82 R. C. GERGERICH AND H. A. SCOTT

mitted by beetles are similar to beetle-transmissible viruses in that they may or may not circulate in the beetle (Wang et a1., 1992). Moreover, they are deposited with regurgitant on the leaf surface (Scott and Fulton, 1978) but are not transmitted, apparently because a factor(s) in beetle regurgitant selectively prevents their transmission (Gergerich et a1., 1983). Ribonuclease (RNase), which is found in high concentrations in beetle regurgitant, appears to prevent transmission of all but the beetle-transmissible viruses (Gergerich et a1., 1986). Several lines of evidence suggest that the beetle-transmissible viruses escape the effects of RNase at beetle feeding wounds because they are transported in xylem more readily than non-beetle-transmissible viruses (Gergerich and Scott, 1988). It is thought that the presence of RNase at the feeding wound prevents virus infection in cells damaged by beetle feeding, and that particles of beetle-transmissible viruses enter the exposed xylem elements and then initiate infection in unwounded cells near the wound, but far enough away to escape the inhibitory effects of the RNase.

The efficiency of transmission of comoviruses by beetles depends greatly on the species of beetle vector and the species of host plant. Species of leaffeeding beetles differ in the efficiency with which they transmit the same comovirus, but some transmission usually occurs, provided that the beetle feeds vigorously on the test plant and that the test plants are sufficiently susceptible to the virus. Cockbain et a1. (1975) obtained 43% transmission of broad bean true mosaic virus (BBTMV) with Apion vorax, but only 2 % or less with A. aethiops, A. aestivum, and Sitona lineatus. Similarly, Cerotoma ruficornis transmitted bean rugose mosaic virus (BRMV) at a frequency of nearly 80%, whereas Diabrotica balteata and D. adelpha transmitted it at frequencies of near 20% and 10%, respectively (Gamez, 1972 j Cartin and Gamez, 1973).

The efficiency of comovirus transmission by beetles may also be affected by the plant species used as source and test plants. After having acquired cowpea severe mosaic virus (CPSMV) from infected cowpea, Cerotoma trifurcata transmitted it to healthy cowpea with 100% efficiency, but to soybean with only 24% efficiency. In similar tests done with Diabrotica undecimpunctata howardi, 72 % of the plants became infected with CPSMV when cowpea was used as the acquisition and test host, but no transmission occurred when the test host was soybean (Jansen and Staples, 1970a). Natural infection of soybean by CPSMV has been reported (Dale, 1949 j Thongmeearkom and Goodman, 1976 j McLaughlin et a1., 1978 j Anjos and Lin, 1984), but the incidence of virus in soybean growing near infected cowpea fields was low (Thongmeearkom et a1., 1978), even though the beetle vector was present. A similar host effect may explain why BPMV, widespread in soybeans in southern and eastern United States, has not been reported in cowpea in this region. Cowpea is susceptible to mechanical inoculation of some isolates of BPMV and is readily colonized by beetle species that are potential vectors.

The inability of beetles to transmit comoviruses to hosts known to be susceptible to mechanical inoculation may be caused by low levels of resis-


tance to the virus in these host plants. Gergerich et a1. (1991) demonstrated that bean leaf beetles transmit CPSMV and cowpea mosaic virus (CPMV) efficiently to Monarch cowpea but only inefficiently to Black Valentine bean (Phaseolus vulgaris), even though both hosts are susceptible to rub-inoculation with virus-containing sap. However, rub-inoculation tests with diluted inoculum revealed that Monarch cowpea is more susceptible to these viruses than Black Valentine bean. This type of test may provide a better comparison with beetle inoculation because apparently comoviruses are transmitted much less efficiently by beetles than by mechanical inoculation. Gergerich et a1. (1991) suggested that a "gross-wound inoculation" technique that mimics beetle feeding damage might be more suitable than the traditional rub-inoculation technique for screening plants for virus resistance. Gross-wound inoculation of plants is accomplished by punching a disk from a leaf with the edge of a glass cylinder that was previously wetted with the virus inoculum (Gergerich et a1., 1983); the plant is then evaluated for virus infection.

The length of time that a beetle remains viruliferous after feeding on a virus-infected plant depends on the beetle species involved. In comparative tests, Epilachna varivestis rarely transmitted CPSMV beyond 1 day, whereas Cerotoma trifurcata transmitted the same virus for several days at high frequencies, and some transmission was recorded after 6-8 days (Fulton and Scott, 1974). Gamez (1972) showed that BRMV was transmitted by C. rujicornis for 9-10 days, but for only 1-3 days when Diabrotica balteata served as the vector.

Retention of ability to transmit virus is also related to the activity of the beetle after acquisition feeding. Freitag (1956) collected overwintering Diabrotica undecimpunctata undecimpunctata in February from weeds not known to be susceptible to squash mosaic virus (SMV) and showed that they transmitted the virus to healthy squash plants. Similarly, BPMV is retained by a low percentage of Cerotoma trifurcata during a dormant, overwintering period of several months (Walters et a1., 1972; Mueller and Haddox, 1980). Although such long retention times suggest multiplication of the virus in the beetle, there is no direct evidence to support this hypothesis.

There are reports of comoviruses being transmitted by vectors other than beetles, for example, the transmission of CPMV by grasshoppers and thrips (Whitney and Gilmer, 1974) and transmission of SMV by grasshoppers (Stoner, 1963). These reports, however, need to be confirmed. In an effort to repeat the work of Whitney and Gilmer (1974), Allen and Van Damme (1981) used flower bud thrips (Megalurothrips s;ostedti) in attempts to transmit CPMV to cowpea, but, based on the lack of virus symptoms in test plants, concluded that this species is unlikely to be a significant vector of this virus.

2. Seed Transmission

Seed transmission is characteristic of several members of the comovirus group (Table III) and is important in the epidemiology of these viruses. For


TABLE III. Transmission of Comoviruses through Seed

Virus Host Seed transmission (%)a Selected reference

BPMV Glycine max 0.10 Lin and Hill (1983) BBSV Vicia faba 1.3-40 Gibbs et al. (1968) BBTMV Vicia faba 2.1-15 Quantz (1953) CPMV Vigna unguiculata 5 Gilmer et al. (1974) CPSMV Vigna sesquipedalis 8 Dale (1949)

Vigna unguiculata 5.8-55 Anderson (1957) SMV Chenopodium quinoa 20 Lockhart et al. (1985)

Chenopodium murale 23 Lockhart et al. (1985) Citrullus vulgaris 1.5 Nelson and Knuhtsen (1973a) Cucumis melo 2.8-93.5 Rader et al. (1947) Cucumis flexuosus Rader et al. (1947) Cucurbita maxima 1.5-5.3 Grogan et al. (1959) Cucurbita mixta 0.3 Grogan et al. (1959) Cucurbita moschata 11 Nolan and Campbell (1984) Cucurbita pepo 2.2-12.1 Middleton (1944)

aFigures represent range of percent transmission reported.

those comoviruses not known to be seed-transmitted, tests have either been limited or not done (Stace-Smith, 1981). The extremely low frequencies of seed transmission recorded with several comoviruses require that large numbers of seeds be used in transmission tests. For example, early efforts to demonstrate transmission of BPMV in soybean seed failed (Skotland, 1958; Ross, 1963; Schwenk and Nickell, 1980), but using a large number of seeds, Lin and Hill (1983) reported 0.1 % seed transmission (7 of 6976 seeds). If fewer seeds had been used in this trial, the seed-borne nature of this virus might not have been apparent.

Virus transmission through the seed is usually determined by observing symptoms in progeny seedlings or by directly assaying seeds or embryos. However, progeny seedlings infected with comoviruses may not show obvious symptoms, as with several of the nepoviruses (Shepherd, 1972) or symptoms may be expressed only under certain environmental conditions (Bennett, 1969; Russo et a1., 1982). Therefore, when checking for seed transmission of comoviruses, it is important to test seedlings either by sap inoculation to suitable indicator plants or by using a sensitive serological technique, such as the enzyme-linked immunosorbent assay (ELISA) (Lockhart et a1., 1985). If visual detection of symptoms is used to evaluate seed transmission, then the identity of the virus in each seedling showing symptoms must be confirmed and some attempt should be made to ascertain whether symptomless seedlings are virus-infected.

The direct detection of virus in seeds may not be a reliable method for determining the potential for seed transmission. Schwenk and Nickell (1980) detected BPMV by gel double-diffusion serological tests in seeds from infected soybean plants. However, they were unable to find the virus in seed-


lings grown from such seed. Autoradiography (Powell and Schlegel, 1970), immunofluorescence, and immunodiffusion tests (Alvarez and Campbell, 1978) were not sensitive enough to detect SMV in cantaloupe seeds but were suitable for detecting virus in seedlings 3-6 days old. The virus was detected, in embryos from seeds that had imbibed water, by assays on systemic and local lesion hosts, but a small percentage of virus infection in the embryos was missed by this method (Alvarez and Campbell, 1978). In contrast, when ELISA was used to detect SMV in embryos from imbibed seeds, no embryos identified as virus-free produced virus-infected plants, whereas several embryos identified as infected produced virus-free plants (Nolan and Campbell, 1984). Russo et al. (1982) successfully used immunosorbent electron microscopy (ISEM) to detect broad bean stain virus (BBSV) in seeds, whereas Cockbain et al. (1976) failed to detect BBSV by mechanical inoculation of test plants or by gel double-diffusion serological tests of embryo and/or seed coat homogenates. Because of their extreme sensitivity, ELISA and ISEM may overestimate the actual percentage of seed transmission. They are, however, useful for determining whether a seed lot came from plants that were virusinfected.

Comoviruses may infect seed through either the pollen or the ovule. BBSV was detected in pollen from infected broad bean (Vicia jabal and was transmitted through the pollen to seeds (Vorra-Urai and Cockbain, 1977). SMV, however, appears to be a superficial contaminant on the surface of pumpkin pollen (Nolan and Campbell, 1984), and such contaminated pollen did not give rise to virus-infected seed (Alvarez and Campbell, 1978). SMV, BBSV, and BBTMV have been shown to infect progeny seedlings through the ovule (Alvarez and Campbell, 1978; Vorra-Urai and Cockbain, 1977).

The effect of virus strain on seed transmission of comoviruses is best illustrated by the two serologically distinct forms of SMV. Isolates belonging to serotype 1 were transmitted through seed in pumpkin, squash, cantaloupe, honeydew melon, and watermelon, whereas transmission through seed of isolates from serotype 2 was limited to pumpkin and squash (Nelson and Knuhtsen, 1973a). Field samples of SMV from cantaloupe and muskmelon belonged only to serotype I, indicating the importance of transmission through seed in the occurrence of this virus in these crops (Nelson and Knuhtsen, 1973a,b).

Transmission through seed has been shown to be of critical importance in the epidemiology of diseases caused by comoviruses because it provides an overwintering mechanism, as well as a dispersal mechanism, for the virus. Although the frequency of seed transmission in commercial seed lots is usually less than 1 %, which in itself is not high enough to cause a significant yield reduction, the presence of an efficient beetle vector can result in sufficient secondary spread to cause serious crop loss. In eastern Scotland, despite the use of seed infected with BBSV and BBTMV in commercial bean fields, no detectable spread of the virus to healthy bean was observed (Jones, 1978). In southern England, however, where Apion vorax, an efficient vector of these


viruses, is present, secondary spread of the virus resulted in up to 90% of the plants being infected with BBSV or BBTMV (Cockbain et a1., 1975).

3. Vegetative Propagation

Transmission through vegetative propagation via tubers is important in the epidemiology of Andean potato mottle virus (APMV) and UVC, which infect potato and Ullucus tuberosus, respectively. Both viruses are widespread in the Andean region of South America where these crops have been vegetatively propagated for well over 1000 years (Brunt and Jones, 1984j

Fribourg et a1., 1977 j Hooker, 1981).

4. Plant-to-Plant Contact

The ease with which comoviruses are mechanically transmitted suggests that some comoviruses may spread in nature by natural foliar contact and cultural operations. Fribourg et a1. (1977) reported that APMV in potato is readily transmitted by plant-to-plant contact and that this is probably the most important means by which the virus spreads in potato fields. Other comoviruses such as BPMV (Zaumeyer and Thomas, 1957) and UVC (Brunt et a1., 1982) have been reported to spread in the field by foliar contact, but the importance of such spread in the epidemiology of these viruses has not been established.

B. Weeds as Primary Sources of Infection

Wild hosts have been reported for several comoviruses, but their role in the occurrence and spread of these viruses has not been established. The ability of several comoviruses to infect annual and perennial weeds has been demonstrated by mechanical inoculation to experimental test plants. However, such tests are meaningless epidemiologically unless the virus can be detected in naturally infected field samples.

CPSMV has been isolated from wild legume hosts such as Phaseolus lathryoides in Puerto Rico (Alconero and Santiago, 1973) and Brazil (Lima and Nelson, 1977), Vigna vexillata in Costa Rica (Valverde et a1., 1982), and Desmodium canescens in south central United States (McLaughlin et a1., 1978), and it has been suggested that these plants are potentially important reservoirs of the virus. Isolation of a comovirus from a wild plant, however, does not necessarily mean that this plant is involved in the overwintering and spread of the virus. The beetle vector must feed well enough upon the wild host to be able to transmit the virus from it to the crop plant of interest.

SMV is commonly found in squirting cucumber, Ecballium elaterium, a weed present throughout the year in commercial cucurbit fields in Israel, yet the virus has not been isolated from susceptible cultivated cucurbits grown


in the same area (Cohen and Nitzany, 1963). In explanation, Cohen and Nitzany (1963) postulated that the usual insecticide treatments for cucurbits effectively controlled the beetle vector, Epilachna chrysomelina, thus preventing spread of this virus from the wild host into the cultivated crop. In Morocco, wild Chenopodium spp. have been observed to be naturally infected with SMV, and high frequencies of seed transmission (20-23%) have been demonstrated (Lockhart et a1., 1985). However, no beetle vectors have been identified that can transmit the virus from Chenopodium spp. to cultivated cucurbits.

In other instances, weeds may play an important role in the epidemiology of comoviruses in the presence of a suitable beetle vector. It is suspected that Desmodium spp. are important in the epidemiology of BPMV in southern United States (Moore et a1., 1969). Up to 31 % of these perennial legumes along the margins of Louisiana soybean fields were infected with BPMV (Horn et a1., 1970). These weeds may serve as a food source for Cerotoma trifurcata (Isely, 1930), which transmits BPMV efficiently from Desmodium paniculatum to soybean (Walters and Lee, 1969). Although weed hosts appear to playa role in the epidemiology of BPMY, their importance relative to overwintering beetles (Walters et a1., 1972; Mueller and Haddox, 1980) and virus-infected seed (Lin and Hill, 1983) as sources of BPMV for infection of soybean has not been critically evaluated.

C. Beetle Ecology

The importance of beetles in the spread of comoviruses in a particular cropping system is affected by many variables that are related to the biology, ecology, and population density of the beetle vector. In temperate ecosystems, beetles go into diapause during the winter months; this diapause results in low populations in the spring and distinctive population peaks during the growing season. In the tropics, however, where crops are planted and harvested continuously, most beetles do not go into diapause and are ready to invade the following season's crops as soon as plants emerge. Thus, overwintering in diapause, emergence patterns, and spring dispersal of beetles relate only to temperate agroecosystems.

Perhaps the best-studied beetle-host system is that of Cerotoma trifurcata (an important vector of BPMV) on soybeans in midwestern United States. Isely (1930) observed that C. trifurcata adults overwinter along the edges of woods and under field trash. Boiteau et a1. (1979) and Jeffords et a1. (1983) examined the emergence patterns of overwintering C. trifurcata and found that beetles emerged in April and May (Illinois) and April through July (North Carolina). Upon emergence, adults move from sites of hibernation to alfalfa, sweet clover, and nonleguminous wild plants where they feed until soybean, the major host for oviposition, is available (Helm et a1., 1983; Waldbauer and Kogan, 1976a). Beetles then colonize early planted soybean


and oviposition in the soil begins immediately. Waldbauer and Kogan (1976b) found three peaks of adult abundance in central Illinois soybean fields: the overwintering colonizers and two summer generations, the second of which forms the overwintering population.

Dispersal of beetles and the comoviruses they transmit appears to occur over relatively short distances. Field observations of the incidence of BBSV and BBTMV in field beans (Vicia faba 1. minor) suggested that Apion vorax did not spread these comoviruses between fields that were a few hundred meters apart (Cockbain et a1., 1975). In a study of flight behavior of the chrysomelid beetle Cerotoma trifurcata, Boiteau et a1. (1979) concluded that flights of bean leaf beetles in soybean fields are very short (less than 30 m), and that longer flights occur only during dispersal of beetles from overwintering sites in the spring and return of beetles to overwintering sites in the fall. These results suggest that within-field dispersal of BPMV by bean leaf beetles occurs over relatively short distances.

The relation between population density peaks of C. trifurcata and the spatial and temporal distribution of BPMV-infected soybean plants in Arkansas was investigated by Hopkins and Mueller (1983). The distribution of BPMV-infected plants in the field was random at infection frequencies of <10%, became aggregated at frequencies of 10 to 40%, and approached uniformity at infection frequencies of >40%. The peak in BPMV transmissions occurred between the density peaks of the second and third beetle populations, approximately 12 weeks after soybean planting. Assuming an economic injury threshold for BPMV of 20-40% infection of the plant population, Hopkins and Mueller (1983) recommended that control measures be directed at the second beetle population peak because it is after this peak that the greatest increase in BPMV-infected soybean plants occurred.

Several studies of the population dynamics of leaf-feeding beetles in tropical agroecosystems have determined the effect of plant monoculture versus polyculture on the abundance of beetle pests. Apparently the traditional practice of intercropping beans and cowpeas with other crops may provide a cultural method of control for beetle pests. In a study over three seasons of monocrops, dicrops, and tricrops of maize, bean, and squash in Costa Rica, Risch (1980a) found that whenever an intercrop contained at least one nonhost plant for a given beetle species, the population of that species per host plant was significantly reduced relative to the population in the monocrop. Combined cultures of bean and banana or cowpea and banana had only one third the number of Diabrotica balteata and Cerotoma ruficornis compared to bean or cowpea monocultures (Risch, 1980b). Further studies (Risch, 1981) showed that beetles tended to migrate more from polycultures that included a nonhost plant. Experimental evidence indicated that beetles avoided shaded host plants and that beetles in polycultures spent less time on nonhost than on host plants, which resulted in greater movement of beetles in mixed crops. Thus, cultural methods of pest control are real possibilities, although shade-tolerant legumes may be needed if polyculture


is to be used successfully. Risch (1980b) thought that planting mixed crops over a large area in consecutive seasons would lead to a gradual decline in beetle populations over time because the effect of decreased beetle abundance toward the end of one season would be reflected in a lowered population at the beginning of the next season.

Very few comparative studies on the incidence of beetle-transmitted viruses in mixed crops have been reported. Gonzalez (1978) showed that the incidence of CPSMV in cowpea was lower in a diculture of corn and cowpea than in a cowpea mono culture during both the rainy and dry seasons. In contrast, Valverde et al. (1982) showed that the incidence of CPSMV was not statistically different in cowpea mono culture than in a diculture of cowpea and corn, even though the total beetle population was greater in the monoculture than in the diculture. They postulated that the discrepancy between these two studies occurred because the incidence of virus and the population density of beetle vectors were much greater in their study than in that of Gonzalez (1978).

II. CONTROL

Effective control programs for the diseases caused by comoviruses must be based on a thorough knowledge of the source(s) from which the virus is introduced into the crop, the beetle species that transmit the virus, and the interaction of the crop host with the virus. The three main approaches that have been used to control comoviruses include prevention through exclusion of the virus, eradication or reduction in numbers of beetle vectors, and breeding for resistance to the virus.

A. Use of Virus-Free Seed

The use of virus-free seed is an effective control measure for viruses, such as SMV, that are primarily introduced in this way. Virus-free seed can be collected from fields known to be free of SMV, or seed lots can be tested using various methods (see Section I.A.2) to ensure that the seed lot is virus-free or has an acceptably low incidence of infected seed. Cockbain et al. (1976) suggested that inspection of seed crops for symptoms of BBSV or BBTMV at the seedling stage and again at the end of flowering is probably the most practical way to identify crops that produce seed lots carrying little or no virus. Jones (1978) concluded that the production of Vicia faha seed free from BBSV and BBTMV could be readily accomplished in eastern Scotland and other areas free from Apion vorax, an efficient vector of the virus. Middleton (1944) suggested that because a higher percentage of SMV is transmitted by seeds that are light and poorly filled than by normal seed, careful winnowing would remove the seeds carrying virus and reduce the percentage transmis-


sion in a seed lot. In contrast, Cockbain et a1. (1976) thought that the removal of light (small) seeds from V faba seed lots infected with BBTMV or BBSV did not necessarily decrease disease incidence.

Eradication of virus from infected seed by seed treatment has not been successful with comoviruses, probably because these viruses infect the embryo, as has been shown with SMV (Alvarez and Campbell, 1978). Cockbain et a1. (1976) showed that heat or chemical treatments of seed lots infected with BBSV or BBTMV were ineffective in reducing seed transmission. Middleton (1944) reported no decrease in seed transmission of SMV in squash after 3 years of storage.

B. Use of Virus-Free Propagation Material

Control methods for viruses, such as APMV and UVC, that occur in vegetatively propagated hosts include the production of virus-free tubers. The production of seed potato tubers free of APMV is accomplished by roguing APMV-infected plants in seed potato fields (Fribourg et a1., 1977) or through routine serological testing for APMV during seed potato production (Avila et a1., 1984). Stone (1982) obtained virus-free clones of Ullucus tuberosus by meristem-tip culture and chemotherapy. Although she was working with a complex of viruses, including UVC, she felt that greatly improved yields could be obtained by growing virus-free plants.

C. Control of Beetle Vectors

Direct control of beetle vectors by spraying entire fields with insecticide has not been shown to be economic, but the use of trap crops to attract beetle vectors, with subsequent killing of the vectors on the limited area of trap crop, has been suggested as a possible control measure for BPMV in soybean (Newsom et a1., 1975) and CPSMVin cowpea (Gay et a1., 1973). The existence of a winter period when soybean or other plant hosts are unavailable to the beetles and the necessity for dispersal from overwintering sites provide an opportunity for disruption of the soybean-G. trifurcata association. In Louisiana, Newsom et a1. (1975) recommended early planting of a soybean trap crop of 10% or less of the total soybean production area in order to attract G. trifurcata. Properly timed applications of insecticides to these plantings would then kill emerging beetles before they could disperse into the main planting. Boiteau et a1. (1979) reported, however, that this practice is not useful in North Carolina because the delayed planting date for the main soybean crop reduces yields and increases damage by other pests, particularly lepidopterans. In addition, the 3-month period of beetle emergence would require frequent insecticide applications over a long period of time, and numerous trap crop plantings would be required because these beetles over-


winter in a variety of habitats. Because beetle emergence is not initiated by the presence of a suitable host for oviposition, late planting of soybean over a wide geographical area may have a significant effect on C. trifurcata populations. If the time span between beetle emergence and the availability of soybean for oviposition is too great, then the overwintered beetles will die before they have a chance to oviposit. Jeffords et al. (1983) felt that poor synchronization of beetle emergence and soybean planting dates in Illinois in 1981 resulted in a very low beetle population, whereas high beetle populations were observed in years when beetle emergence was well synchronized with that of the soybean crop.

D. Breeding for Resistance

Breeding for immunity, resistance, or tolerance may be the only effective way to control those comoviruses that are not controlled by elimination of the primary sources of infection and that are spread efficiently within a crop. Sources of resistance have been identified for some of the more economically important comoviruses in legumes. Wells and Deba (1961) identified 22 of 458 cowpea accessions as immune to CPMV in the field, based on growth of individual plants and on the severity of leaf symptoms following mechanical inoculation. Robertson (1965) showed that the local lesion reaction of cowpea varieties to inoculation with CPMV was correlated with their resistance or immunity to systemic infection: cowpea varieties that developed chlorotic lesions on the inoculated primary leaves became systemically infected and were classified as susceptible; those that developed necrotic lesions did not become systemically infected and were resistant; and those that gave no reaction on primary leaves were immune to infection. Using this technique, Robertson (1965) identified 16 of 79 cowpea varieties as immune to CPMV, and Beier et al. (1977) identified 65 of 1031 lines of cowpea as operationally immune. Robertson (1965) found that some cowpea lines were immune to one strain of CPMV but resistant to another strain. Comparative studies of available strains of CPMV on described resistant or immune varieties of cowpea are needed to develop a better basis for breeding cowpea cultivars resistant to CPMV.

The molecular basis of the resistance to CPMV in the cowpea cv. Arlington has been studied in some detail. This cultivar was identified by Beier et al. (1977, 1979) as the only one of 1031 cowpea lines that was resistant to CPMV at the protoplast level: the virus concentration reached only 1-10% of that in the susceptible cv. Blackeye 5. This form of resistance was not displayed toward CPSMV, and so seemed to be CPMV-specific. In crosses between cv. Arlington and cv. Blackeye 5, the resistance was inherited as a single dominant character. The resistance affected accumulation of the virus in the cell and not, as with resistance in cowpea to tobacco mosaic virus (Sulzinski and Zaitlin, 1982), cell-to-cell movement. Sanderson et al. (1985)


and Ponz et a1. (1987) showed that CPMV resistance in cv. Arlington protoplasts is associated with the presence of a protein that acts as an inhibitor of the proteolytic processing reactions by which the polyproteins transcribed from the CPMV genomic RNA molecules are cleaved to form the functional proteins (see Chapter 3). This work was reviewed by Bruening et al. (1987).

Immunity to CPSMV was identified in 4 of 169 cowpea accessions from the International Cowpea Disease Nursery (Ibadan, Nigeria) when these accessions were tested against three isolates of CPSMV that differed in their country of origin, pathogenicity, and serology (Fulton and Allen, 1982). The cultivar Macaibo, originally described as immune to a Brazilian isolate of CPSMV by Lima and Nelson (1977), was found to be resistant to six other isolates of CPSMV (Fulton and Allen, 1982). These studies indicate that sources of resistance to CPSMV are available for use in cowpea breeding and that this resistance may be effective against different CPSMV isolates.

Resistance to BRMV in the form of immunity or local lesion hypersensitive reaction has been identified in common bean (Machado and Pinchinat, 1975). Zaumeyer and Thomas (1957) identified resistance to BPMV in the form of a hypersensitive reaction in green-podded varieties of snapbean and most dry beans.

Resistance to BPMV has not been found in soybean (Walters, 1970j Scott et a1., 1974), although Scott et a1. (1974) identified resistance to BPMVwithin the genus Glycine and suggested that interspecific crosses might be used to introduce resistance to BPMV into commercial soybean varieties. Similarly, although no resistance to SMV was identified in cucumber, melon, or squash, hypersensitive resistance to SMV occurred in two accessions of Cucumis metuliferus, and this may be a potential source of SMV resistance for commercial cucurbits through interspecific crosses (Provvidenti and Robinson, 1974).

Inoculation of seedlings by conventional mechanical transmission techniques is the traditional method for detecting resistance to plant viruses. Little is known about detection of resistance when the virus is transmitted by the beetle vector, although there are suggestions that beetle transmission results in marked differences in plant susceptibility when compared to mechanical transmission. Jansen and Staples (1970a) showed that there was a difference between soybean and cowpea as source or test plants in beetle transmission trials with CPSMV even though the two plant species are equally susceptible to mechanical transmission. These results indicate that host differences, which are not related to susceptibility by mechanical inoculation, may affect the ability of beetles to transmit certain viruses and that the traditional method of screening for resistance by means of mechanical inoculation may not detect some forms of resistance to vector-inoculated virus.

The use of beetle transmission in trials to evaluate resistance to vectorborne virus would obviously result in a cumbersome screening procedure. However, results of experiments using gross-wound inoculation (see Section


I.A.1), which mimics beetle feeding (Gergerich et a1., 1991), suggest that this inoculation procedure may be useful in identifying resistance to infection by vector-borne virus. This inoculation technique is less efficient than mechanical inoculation of Carborundum-dusted leaves and it more closely matches the efficiency of virus transmission by leaf-feeding beetles. Cerotoma trifurcata transmits CPSMV efficiently from Black Valentine bean to Monarch cowpea, but not from Black Valentine bean to Black Valentine bean, even though both plants are susceptible to mechanical inoculation (Gergerich et a1., 1991). Transmission trials using the gross-wound inoculation technique with purified virus in the presence of regurgitant or RNase result in the same type of host effect, i.e., Monarch cowpea is much more susceptible to infection by CPSMV than Black Valentine bean. These results suggest that the gross-wound inoculation technique might identify heretofore undetected resistance to beetle-inoculated viruses. This procedure of screening may prove to be particularly useful for detection of resistance in crop plants in which no resistance has been demonstrated with conventional screening procedures, such as BPMV in soybean (Walters, 1970; Scott et a1., 1974).

REFERENCES

Alconero, R., and Santiago, A., 1973, Phaseolus lathyroides as a reservoir of cowpea mosaic virus in Puerto Rico, Phytopathology 63:120.

Allen, D. J., and Van Damme, P., 1981, Thrips transmission of cowpea yellow mosaic virus, Trop. Agric. 58:18l.

Alvarez, M., and Campbell, R. N., 1978, Transmission and distribution of squash mosaic virus in seeds of cantaloupe, Phytopathology 68:257.

Anderson, C. W., 1957, Seed transmission of three viruses in cowpea, Phytopathology 47:515. Anjos, J. R. N., and Lin, M. T., 1984, Bud blight of soybeans caused by cowpea severe mosaic virus

in Central Brazil, Plant Dis. 68:405. Avila, A. C., Salazar, 1. E, Ortega, M., and Daniels, J., 1984, A new strain of Andean potato

mottle virus from Brazil, Plant Dis. 68:997. Beier, H., Siler, D. J., Russell, M. 1., and Bruening, G., 1977, Survey of susceptibility to cowpea

mosaic virus among protoplasts and intact plants from Vigna sinensis lines, Phytopathology 67:917.

Beier, H., Bruening, G., Russell, M. 1., and Tucker, C. 1., 1979, Replication of cowpea mosaic virus in protoplasts isolated from immune lines of cowpeas, Virology 95:165.

Bennett, C. W., 1969, Seed transmission of plant viruses, Adv. Virus Res. 14:22l. Bock, K. R., 1971, Notes on east African plant virus diseases. I. Cowpea mosaic virus, East Afr.

Agric. For. ,. 37:60. Boiteau, G., Bradley, J. R., and Van Duyn, J. W., 1979, Bean leaf beetle: Emergence patterns of

adults from overwintering sites, Environ. Entomol. 8:427. Bruening, G., Ponz, E, Glascock, C., Russell, M. 1., Rowhani, A., and Chay, C., 1987, Resistance

of cowpeas to cowpea mosaic and tobacco ringspot virus, in: Plant Resistance to Viruses ID. Evered and S. Harnett, eds.l, pp. 23-32, Ciba Foundation Symposium No. 133, Wiley, Chichester.

Brunt, A. A., and Jones, R. A. c., 1984, Ullucus virus C, CMljAAB Descriptions of Plant Viruses No. 277.

Brunt, A. A., Barton, R. J., Phillips, S., and Jones, R. A. c., 1982, Ullucus virus C, a newly recognized comovirus infecting Ullucus tuberosus IBasellaceael, Ann. Appl. Biol. 101:73.


Campbell, R. N., and Colt, w. M., 1967, Transmission of radish mosaic virus, Phytopathology 57:502.

Cartin, 1. F., and Gamez, R., 1973, Transmission of two new strains of bean rugose mosaic virus by chrysomelid insects, in: Abstracts, 13th Annual Meeting of the American Phytopathology Society Caribbean Division, San Jose, Costa Rica (p. 121, American Phytopathological Society, St. Paul, Minnesota.

Chant, S. R., 1959, Viruses of cowpea, Vigna unguiculata (1.) Walp., in Nigeria, Ann. Appl. Biol. 47:565.

Cockbain, A. J., Cook, S. M., and Bowen, R., 1975, Transmission of broad bean stain virus and Echtes Ackerbohnenmosaik-Virus to field beans (Vicia fabal by weevils, Ann. Appl. Biol. 81:33l.

Cockbain, A. J., Bowen, R., and Vorra-Urai, S., 1976, Seed transmission of broad bean stain virus and Echtes Ackerbohnenmosaik-Virus in field beans (Vicia faba), Ann. Appl. Biol. 84:32l.

Cohen, S., and Nitzany, F. E., 1963, Identity of viruses affecting cucurbits in Israel, Phytopathology 53:193.

Costa, C. 1., Lin, M. T., Kitajima, E. W., Santos, A. A., Mesquita, R. C. M., and Freire, F. R., 1978, Cerotoma arcuata (Oliv.I, urn crisomelideo vector do mosaico da Vigna no Brasil, Fitopatol. Bras. 3:8l.

Dale, W. T., 1949, Observations on a virus disease of cowpea in Trinidad, Ann. Appl. Biol. 36:327. Dale, W. T., 1953, The transmission of plant viruses by biting insects, with particular reference

to cowpea mosaic, Ann. Appl. Biol. 40:384. Debrot, E., and Rojas, C. E. B., 1967, EI virus del mosaico del frijol, Vigna sinensis Endl. (cowpea

mosaic virus) en Venezuela, Agran. Trap. 17:3. Freitag, J. H., 1941a, A comparison of the transmission of four cucurbit viruses by cucumber

beetles and by aphids, Phytopathology 31:8. Frietag, J. H., 1941b, Insect transmission, host range, and properties of squash-mosaic virus,

Phytopathology 31:8. Freitag, J. H., 1956, Beetle transmission, host range, and properties of squash mosaic virus,

Phytopathology 46:73. Fribourg, C. E., Jones, R. A. C., and Koenig, R., 1977, Andean potato mottle, a new member of the

cowpea mosaic virus group, Phytopathology 67:969. Fulton, J. P., and Allen, D. J., 1982, Identification of resistance to cowpea severe mosaic virus,

Trap. Agric. ('IIinidad) 59:66. Fulton, J. P., and Scott, H. A., 1974, Virus-vectoring efficiencies of two species of leaf-feeding

beetles, Ann. Proc. Am. Phytopathol. Soc. 1:159. Fulton, J. P., Scott, H. A., and Gamez, R., 1975, Beetle transmission of legume viruses. in:

'IIopical Diseases of Legumes (T. Bird and K. Maramorosch, eds.I, pp. 123-131, Academic Press, New York.

Fulton, J. P., Scott, H. A., and Gamez, R., 1980, Beetles, in: Vectors of Plant Pathogens (K. Maramorosch and K. Harris, eds.), pp. 115-132, Academic Press, New York.

Gamez, R., 1972, Los virus del frijol en Centroamerica. II. Algunas propiedades y transmision por crisomelidos del virus del mosaico rugoso del frijol, Thrrialba 22:249.

Gay, J. D., Johnson, A. W., and Chalfant, R. B., 1973, Effects of a trap-crop on the introduction and distribution of cowpea virus by soil and insect vectors, Plant Dis. Rep. 57:684.

Gergerich, R. C., and Scott, H. A., 1988, Evidence that virus translocation and virus infection of non-wounded cells are associated with transmissibility by leaf-feeding beetles, J. Gen. Viral. 69:2935.

Gergerich, R. C., and Scott, H. A., 1991, Determinants in the specificity of virus transmission by leaf-feeding beetles, in: Advances in Disease Vector Research, Vol. 8 (K. F. Harris, ed.I, pp. 1-13, Springer-Verlag, New York.

Gergerich, R. C., Scott, H. A., and Fulton, J. P., 1983, Regurgitant as a determinant of specificity in the transmission of plant viruses by beetles, Phytopathology 73:936.

Gergerich, R. C., Scott, H. A., and Fulton, J. P., 1986, Evidence that ribonuclease in beetle regurgitant determines the transmission of plant viruses. J. Gen. Virol. 67:367.


Gergerich, R. C., Scott, H. A., and Wickizer, S. 1., 1991, Determination of host resistance to beetle transmission of plant viruses, Phytopathology 81:1326.

Gerhardson, B., and Petterson, J., 1974, Transmission of red clover mottle virus by clover shoot weevils, Apion spp., Swedish ,. Agric. Res. 4:161.

Ghabrial, S. A., and Schultz, F. J., 1983, Serological detection of bean pod mottle virus in bean leaf beetles, Phytopathology 73:480.

Gibbs, A. J., Giussani-Belli, G., and Smith, H. G., 1968, Broad bean stain and true broad-bean mosaic viruses, Ann. Appl. Biol. 61:99.

Gilmer, R. M., Whitney, W. K., and Williams, R. J., 1974, Epidemiology and control of cowpea mosaic in Western Nigeria, in: Proceedings of the 1st IITA Grain Legume Improvement Workshop, p. 269, International Institute of Tropical Agriculture, Ibadan, Nigeria.

Gonzalez, C. E., 1978, Identidad, transmision por insectos crisomelidos y epifitiologia de virus del frijol de costa (Vigna unguiculata 1. Walpl en Costa Rica, Tesis, Ing. Agric. Universidad de Costa Rica.

Grogan, R. G., Hall, D. H., and Kimball, K. A., 1959, Cucurbit mosaic viruses in California, Phytopathology 49:366.

Helm, C. G., Jeffords, M. R., Post, S. 1., and Kogan, M., 1983, Spring feeding activity of overwintered bean leaf beetles (Coleoptera: Chrysomelidael on nonleguminous hosts, Environ. Entomol. 12:321.

Hobbs, H. A., 1981, Transmission of bean curly dwarf mosaic virus and bean mild mosaic virus by beetles in Costa Rica, Plant Dis. 65:491.

Hooker, W., 1981, The potato, in: Compendium of Potato Diseases (W. Hooker, ed.I, pp. 1-2, American Phytopathological Society, St. Paul, MN.

Hopkins, J. D., and Mueller, A. J., 1983, Distribution of bean pod mottle virus in Arkansas soybean as related to the bean leaf beetle, Cerotoma trifurcata, (Coleoptera: Chrysomelidael population, Environ. Entomol. 12:1564.

Horn, N. 1., Newsom, 1. D., Carver, R. G., and Jensen, R. 1., 1970, Effects of virus diseases on soybeans in Louisiana, Louisiana Agric. 13:12.

Isley, D., 1930, The biology of the bean leaf-beetle, Bull. Arkansas Agric. Exp. Stn. No. 248. Jansen, W. P., and Staples, R., 1970a, Effect of cowpeas and soybeans as source or test plants of

cowpea mosaic virus on vector efficiency and retention of infectivity of the bean leaf beetle and the spotted cucumber beetle, Plant Dis. Rep. 54:1053.

Jansen, W. P., and Staples, R., 1970b, Transmission of cowpea mosaic virus by the Mexican bean beetle, ,. Econ. Entomol. 63:1719.

Jansen, W. P., and Staples, R., 1971, The specificity of transmission of cowpea mosaic virus by species within the sub-family Galerucinae, family Chrysomelidae, r. Eeon. Entomol. 64:365.

Jeffords, M. R., Helm, C. G., and Kogan, M., 1983, Overwintering behavior and spring colonization of soybean by the bean leaf beetle (Coleoptera: Chrysomelidael in Illinois, Environ. Entomol. 12:1459.

Jones, A. T., 1978, Incidence, field spread, seed transmission and effects of broad bean stain virus and Echtes Ackerbohnenmosaik-Virus in Vicia faba in eastern Scotland, Ann. Appl. Biol. 88:137.

Kvicala, B. A., Smrz, J., and Blanco, N., 1973, A beetle-transmitted virus disease of cowpea in Cuba, FAD Plant Prot. Bull. 21:27.

Lastra, R., 1968, Occurrence of cucurbit viruses in Venezuela, Plant Dis. Rep. 52:171. Lima, J. A. A., and Nelson, M. R., 1977, Etiology and epidemiology of mosaic of cowpea in Ceara,

Brazil, Plant Dis. Rep. 61:864. Lin, M. T., and Hill, J. H., 1983, Bean pod mottle virus: occurrence in Nebraska and seed

transmission in soybeans, Plant Dis. 67:230. Lin, M. T., Hill, J. H., Kitajima, E. W., and Costa, C. 1., 1984, Two new serotypes of cowpea severe

mosaic virus, Phytopathology 74:581. Lockhart, B. E. 1., Ferji, Z., and Hafidi, B., 1982, Squash mosaic virus in Morocco, Plant Dis.

66:1191.


Lockhart, B. E. 1., Jebbour, F., and Lennon, A. M., 1985, Seed transmission of squash mosaic virus in Chenopodium spp., Plant Dis. 69:946.

Machado, P. F. R., and Pinchinat, A. M., 1975, Herencia de la reaccion del frijol comun a la infeccion por el virus del mosaico rugoso, Turrialba 25:418.

McLaughlin, M. R., Thongmeearkom, P., Goodman, R. M., Milbrath, G. M., Ries, S. M., and Royse, D. J., 1978, Isolation and beetle transmission of cowpea mosaic virus (severe subgroup) from Desmodium canescens and soybeans in Illinois, Plant Dis. Rep. 62:1069.

Meiners, J. P., Waterworth, H. E., Lawson, R. H., and Smith, F. F., 1977, Curly dwarf mosaic disease of beans from El Salvador, Phytopathology 67:163.

Middleton, J. T., 1944, Seed transmission of squash-mosaic virus, Phytopathology 34:405. Milbrath, G. M., McLaughlin, M. R., and Goodman, R. M., 1975, Identification of bean pod

mottle virus from naturally infected soybeans in Illinois, Plant Dis. Rep. 59:982. Moore, B. J., 1973, Quail pea mosaic virus: A new member of the comovirus group, Plant Dis.

Rep. 57:31l. Moore, B. J., and Scott, H. A., 1971, Properties of a strain of bean pod mottle virus, Phytopathol

ogy 61:83l. Moore, B. J., Scott, H. A., and Walters, H. J., 1969, Desmodium paniculatum, a perennial host of

bean pod mottle virus in nature, Plant Dis. Rep. 53:154. Mueller, A. J., and Haddox, A. W., 1980, Observations on seasonal development of bean leaf

beetle, Cerotoma trifurcata (Forster), and incidence of bean pod mottle virus in Arkansas soybean, J. Georgia Entomol. Soc. 15:398.

Murant, A. F., Roberts, I. M., and Elnagar, S., 1976, Association of virus-like particles with the foregut of the aphid Cavariella aegopodii transmitting the semi-persistent viruses anthriscus yellows and parsnip yellow fleck, J. Gen. Virol. 31:47.

Nault, 1. R., and Ammar, E. D., 1989, Leafhopper and planthopper transmission of plant viruses, Annu. Rev. Entomol. 34:503.

Nelson, M. R., and Knuhtsen, H. K., 1973a, Squash mosaic virus variability: Epidemiological consequences of differences in seed transmission frequency between strains, Phytopathology 63:918.

Nelson, M. R., and Knuhtsen, H. K., 1973b, Squash mosaic virus variability: Review and serological comparisons of six biotypes, Phytopathology 63:920.

Newsom, 1. D., Jensen, R. 1., Herzog, D. C., and Thomas, J. W., 1975, Apestmanagement system for soybeans, Louisiana Agric. 18:10.

Nolan, P. A., and Campbell, R. N., 1984, Squash mosaic virus detection in individual seeds and seed lots of cucurbits by enzyme-linked immunosorbent assay, Plant Dis. 68:97l.

Patel, V. c., and Pitre, H. N., 1971, Transmission of bean pod mottle virus to soybean by the striped blister beetle, Epicauta vittata, Plant Dis. Rep. 55:628.

Patel, V. C., and Pitre, H. N., 1976, Transmission of bean pod mottle virus by bean leaf beetle and mechanical inoculation to soybeans at different stages of growth, J. Georgia Entomol. Soc. 11:289.

Perez, J. E., and Cortes-Monllor, A., 1970, A mosaic virus of cowpea from Puerto Rico, Plant Dis. Rep. 54:212.

Ponz, F., Glascock, C. B., and Bruening, G., 1987, An inhibitor of polyprotein processing with the characteristics of a natural virus resistance factor, Molec. Plant-Microbe Interact. 1:25.

Powell, C. C., and Schlegel, D. E., 1970, Factors influencing seed transmission of squash mosaic virus in cantaloupe, Phytopathology 60:1466.

Provvidenti, R., and Robinson, R. W., 1974, Resistance to squash mosaic virus and watermelon mosaic virus 1 in Cucumis metuliferus, Plant Dis. Rep. 58:735.

Quantz, 1., 1953, Untersuchungen uber ein samenubertragbares Mosaikvirus der Ackerbohne (Vicia faba), Phytopathol. Z. 20:42l.

Rader, W. E., Fitzpatrick, H. F., and Hildebrand, E. M., 1947, A seed-borne virus of muskmelon, Phytopathology 37:809.

Risch, S. J., 1980a, The population dynamics of several herbivorous beetles in a tropical


agroecosystem: The effect of intercropping corn, beans and squash in Costa Rica, J. Appl. Ecol. 17:593.

Risch, S. J., 1980b, Fewer beetle pests on beans and cowpeas interplanted with banana in Costa Rica, Thrrialba 30:229.

Risch, S. J., 1981, Insect herbivore abundance in tropical monocultures and polycultures: An experimental test of two hypotheses, Ecology 62:1325.

Robertson, D. G., 1965, The local lesion reaction for recognizing cowpea varieties immune from and resistant to cowpea yellow mosaic virus, Phytopathology 55:923.

Ross, J. P., 1963, Transmission of bean pod mottle virus in soybeans by beetles, Plant Dis. Rep. 47:1049.

Russo, M., Savino, v., and Vovlas, c., 1982, Virus diseases of vegetable crops in Apulia XXVIII. Broad bean stain, Phytopathol. Z. 104:115.

Sanderlin, R. S., 1973, Survival of bean pod mottle and cowpea mosaic viruses in beetles following intrahemocoelic injections, Phytopathology 63:259.

Sanderson, J. L., Bruening, G., and Russell, M. L., 1985, Possible molecular basis of immunity of cowpeas to cowpea mosaic virus, in: Cellular and Molecular Biology of Plant Stress (T. L. Key and T. Kosuge, eds.l, pp. 401-412, UCLA Symp. Molec. Cell. Bioi. New Ser. No. 22, Liss, New York.

Schwenk, F. W., and Nickell, C. D., 1980, Soybean green stem caused by bean pod mottle virus, Plant Dis. 64:863.

Scott, H. A., and Fulton, J. P., 1978, Comparison of the relationship of southern bean mosaic virus and the cowpea strain of tobacco mosaic virus with the bean leaf beetle, Virology 84:197.

Scott, H. A., Van Scyoc, J. v., and Van Scyoc, C. E., 1974, Reactions of Glycine spp. to bean pod mottle virus, Plant Dis. Rep. 58:19l.

Shepherd, R. J., 1972, Transmission of viruses through seed and pollen, in: Principles and Techniques in Plant Virology IC. Kado and H. Agrawal, eds.l, pp. 267-292, Van Nostrand Reinhold Co., New York.

Sitterly, W. R., 1960, A new insect vector of squash mosaic virus, Plant Dis. Rep. 44:134. Skotland, C. B., 1958, Bean pod mottle virus of soybeans, Plant Dis. Rep. 42:1155. Slack, S. A., and Scott, H. A., 1971, Hemolymph as a reservoir for the cowpea strain of southern

bean mosaic virus in the bean leaf beetle, Phytopathology 61:538. Smith, C. E., 1924, Transmission of cowpea mosaic by the bean leaf beetle, Science 60:268. Smith, K. M., 1965, Plant virus-vector relationships, Adv. Virus Res. 11:6l. Stace-Smith, R., 1981, Comoviruses, in: Handbook of Plant Virus Infections. Comparative

Diagnosis IE. Kurstak, ed.1, pp. 171-195, Elsevier/North-Holland Biomedical Press, New York.

Stone, O. M., 1982, The elimination of four viruses from Ullucus tuberosus by meristem-tip culture and chemotherapy, Ann. Appl. Biol. 101:79.

Stoner, W. N., 1963, A mosaic virus transmitted by beetles and a grasshopper, Phytopathology 53:890.

Sulzinski, M. A., and Zaitlin, M., 1982, Tobacco mosaic virus replication in resistant and susceptible plants: In some resistant species virus is confined to a small number of initially inoculated cells, Virology 121:12.

Thongmeearkom, P., and Goodman, R. M., 1976, A severe disease of soybeans caused by an isolate of cowpea mosaic virus, Proc. Am. Phytopathol. Soc. 3:209.

Thongmeearkom, P., Paschal, E. H., and Goodman, R. M., 1978, Yield reductions in soybeans infected with cowpea mosaic virus, Phytopathology 68:1549.

Tochihara, H., 1968, Radish enation mosaic virus, Ann. Phytopathol. Soc. Japan 34:129. Valverde, R., Moreno, R., and Gamez, R., 1978, Beetle vectors of cowpea mosaic virus in Costa

Rica, Thrrialba 28:90. Valverde, R., Moreno, R., and Gamez, R., 1982, Incidence and some ecological aspects of cowpea

severe mosaic virus in two cropping systems in Costa Rica, Thrrialba 32:29.


Vorra-Urai, S., and Cockbain, A. T., 1977, Further studies on seed transmission of broad bean stain virus and Echtes Ackerbohnenmosaik-Virus in field beans (Vicia iabal, Ann. Appl. Biol. 87:365.

Waldbauer, G. P., and Kogan, M., 1976a, Bean leaf beetles: Bionomics and economic role in soybean agroecosystems, in: World Soybean Research, Proceedings of the World Soybean Research Conference 1975 (1. D. Hill, ed.I, pp. 619-628, Interstate Printers and Publishers, Danville, Illinois.

Waldbauer, G. P., and Kogan, M., 1976b, Bean leaf beetle: Phenological relationship with soybean in Illinois, Environ. Entomol. 5:35.

Walters, H. T., 1964, Transmission of bean pod mottle virus by bean leaf beetles, Phytopathology 54:240.

Walters, H. T., 1969, Beetle transmission of plant viruses, Adv. Virus Res. 15:339. Walters, H. J., 1970, Bean pod mottle virus disease of soybeans, Arkansas Farm Res. 19:8. Walters, H. T., and Barnett, O. W., 1964, Bean leaf beetle transmission of Arkansas cowpea mosaic

virus, Phytopathology 54:911. Walters, H. T., and Lee, F. N., 1969, Transmission of bean pod mottle virus from Desmodium

paniculatum to soybean by the bean leaf beetle, Plant Dis. Rep. 53:411. Walters, H. J., Lee, F. N., and Tackson, K. E., 1972, Overwintering of bean pod mottle virus in bean

leaf beetles, Phytopathology 62:808. Wang, R. Y., Gergerich, R. C., and Kim, K. S., 1992, Noncirculative transmission of plant viruses

by leaf-feeding beetles, Phytopathology 82:946. Wang, R. Y., Gergerich, R. C., and Kim, K. S., 1994, Entry of ingested plant viruses into the

hemocoel of the beetle vector Diabrotica undecimpunctata howardi, Phytopathology 84:147.

Wells, D. G., and Deba, R., 1961, Sources of resistance to the cowpea yellow mosaic virus, Plant Dis. Rep. 45:878.

Whitney, W. K., and Gilmer, R. M., 1974, Insect vectors of cowpea mosaic virus in Nigeria, Ann. Appl. Biol. 77:17.

Yoshida, K., Goto, T., Nemoto, M., and Tsuchizaki, T., 1980, Squash mosaic virus isolated from melon (Cucumis melo 1.1 in Hokkaido, Ann. Phytopathol. Soc. Japan 46:349.

Zaumeyer, W. T., and Thomas, H. R., 1957, A monographic study of bean diseases and methods for their control, U.S. Dept. Agricultural Technical Bulletin No. 868.

CHAPTER 5

Nepoviruses: General Properties, Diseases, and Virus Identification A. F. MURANT, A. T. JONES, G. P. MARTELLI,

AND R. STACE-SMITH

I. INTRODUCTION AND GENERAL PROPERTIES

One of the select number of plant viruses studied in the prenucleoprotein era of virology was first obtained in the United States from tobacco plants with a ringspotting disease (Fromme et a1., 1927). This virus, later known as tobacco ringspot virus (TRSV), induced necrotic rings in inoculated leaves of tobacco and in the first leaves to be invaded systemically; leaves produced subsequently showed little evidence of infection, although they too contained the virus (Wingard, 1928). The virus could infect a wide range of test plants, and so also could other similar viruses described later, such as tobacco ringspot virus No.2 [= tomato ringspot virus (ToRSV)] (Price, 1936), tomato black ring virus (TBRV) (Smith, 1946L arabis mosaic virus (ArMV) (Smith and Markham, 1944L raspberry ringspot virus (RRSV) (Cadman, 1956; Harrison, 1958), and strawberry latent ringspot virus (SLRSV) (Lister, 1964). These viruses

A. F. MURANT AND A. T. JONES • Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom. G. P. MARTELLI • Dipartimento di Protezione delle Piante, Universita degli Studi di Bari, 7126 Bari, Italy. R. STACE-SMITH • Pacific Agriculture Research Centre, Agriculture and Agri-Food Canada, Vancouver, British Columbia, Canada V6T 1X2.

99

100 A. F. MURANT ET AL.

were found to have isometric particles 25-30 nm in diameter, which, on closer examination, were seen to be icosahedral (Steere, 1956; Harrison and Nixon, 1960). Moreover, each virus typically produced, in addition to nucleic acid-free "empty" shells (T component), at least two kinds of nucleoprotein particle (M and B components) differing in sedimentation coefficient and density but not in diameter or antigenicity (Steere, 1956; Debrot, 1964; StaceSmith et a1., 1965; Murant et a1., 1972; Schneider et a1., 1972). These different nucleoproteins proved to contain two differently sized species of RNA (Diener and Schneider, 1966; Murant et a1., 1972) that together make up the genome of each virus (Harrison et a1., 1972a,b).

How these ringspot viruses were spread from plant to plant in nature remained a mystery for many years, but over a short period in the 1950s, it became evident that several of them were carried in soil (Wagnon and Breece, 1955; Cadman, 1956; Harrison, 1956, 1957). In 1958, Hewitt and co-workers incriminated a soil-inhabiting dagger nematode (Xiphinema index) as the vector of grapevine fanleaf virus (GFLV) in the United States, and a search for other nematode vectors soon followed. As a result, allied nematode species occurring in Europe and North America were shown to transmit ArMY, RRSV, SLRSV, TBRV, and ToRSV (Breece and Hart, 1959; Harrison and Cadman, 1959; Jha and Posnette, 1959; Harrison et a1., 1961; Taylor, 1962; Lister, 1964). Soon afterwards, transmission through seed, first reported for TRSV and ToRSV, respectively, by Henderson (1931) and Kahn (1956), was found to be characteristic of the nematode-transmitted ringspot viruses (Lister, 1960; Lister and Murant, 1967).

Because these viruses shared many physical and biological properties, Cadman (1963) introduced the collective term nepovirus (nematode vectors, polyhedral particles) for them. Subsequently, these viruses were placed by Harrison et a1. (1971) in a distinct taxonomic group, for which the name nepovirus (now genus Nepovirus) was retained. Since then, more than 25 additional viruses have been assigned to the nepoviruses, either as definitive or tentative members. A full alphabetical listing is given in Table I, along with the acronyms that are recommended by the International Committee on Taxonomy of Viruses (ICTV) (Hull et a1., 1991).

The characteristics now regarded as typical of a nepovirus include: mosaic, mottle, and ringspot symptoms, often followed by "recovery" (see Section II); broad host range; nematode vector; transmissibility through seed; isometric particle morphology (ca 28 nm, hexagonal outlines, some particles penetrated by stain); three sedimenting components; a single particle protein (ca 55 kDa); and a bipartite ssRNA genome. However, although for classification in a taxonomic group or genus a virus should possess many of the properties regarded as typical, it need not possess all of them. Moreover, although all properties are supposed to be regarded as of equal importance for classification, some are in practice given a higher weighting than others. Thus the morphology of the particles and their coat protein composition are given a high weighting. Table I of Chapter 6 (this volume) separates the

NEPOVIRUSES: GENERAL PROPERTIES 101

definitive and tentative nepoviruses on the basis of coat protein composition: those with a single-coat protein of molecular weight ca 55 kDa are regarded as definitive nepoviruses, and those with two or three smaller coat proteins or whose protein composition is unknown are regarded as tentative nepoviruses. To give special weighting to this property rather than others is necessarily a subjective decision, especially considering that two tentative members, cherry rasp leaf virus (CRLV) and SLRSV, are known to have nematode vectors. However, at least with lucerne Australian symptomless virus (LASV), Rubus Chinese seed-borne virus (RCSV), satsuma dwarf virus (SDV), and SLRSV, there is additional justification for this "tentative" placing. They each have two coat protein species, of 40-47 and 21-27 kDa, in which respect they resemble comoviruses and fabaviruses more closely than nepoviruses. Possibly as a result of this, negatively stained particles of SLRSV and RCSV resemble those of fabaviruses more closely than those of nepoviruses (see Chapter 6, Section II.B).

In contrast, the possession of a nematode vector is given a lower weighting, perhaps because it is a more difficult property to determine (see Chapter 7, this volume, for a review of transmission of nepoviruses by nematodes). Therefore, paradoxically, fewer than half the viruses now classified as nepoviruses have been shown to have nematode vectors (Table I). In many instances, this is because the necessary research has not been done. However, for at least two members, cherry leaf roll virus (CLRV) and blueberry leaf mottle virus (BLMV), there is good experimental evidence that they are not transmitted by nematodes (Jones et a1., 1981; Childress and Ramsdell, 1986; see Chapter 7).

The seed transmissibility of nepoviruses was reviewed by Murant (1983). Seed transmission is an important feature in the epidemiology of many nematode-transmitted nepoviruses, especially of those, such as RRSV and TBRV, that are retained for only a few weeks by their vectors, and which therefore rely on persistence in seeds of wild plants for survival over winter or through periods of fallow (Murant and Taylor, 1965; Murant and Lister, 1967; see Chapter 8, this volume). Infection of the seed can occur through both the pollen and the ovule. With most viruses, particles carried in the pollen seem unable to infect the pollinated plant, but there is good circumstantial evidence that at least CLRV (Mircetich et a1., 1980) and BLMV (Childress and Ramsdell, 1987; Boylan-Pett et a1., 1991) may have evolved in this direction instead of associating with a nematode vector. This topic is discussed in detail in Chapter 8.

The two genomic RNA species of nepoviruses have molecular weights (x 10-6) of ca 2.8 (RNA-I) and 1.4-2.3 (RNA-2). Murant (1981) subdivided the definitive nepoviruses according to (1) the sizes of their RNA-2 molecules, and (2) their serological relationships, if any, with other members. The listing given in Table I of Chapter 6 is based on this approach. Further study may enable other serological sub clusters to be formed, but there do not seem to be the same extensive serological relationships among nepoviruses as are found

TA

BL

E I

. N

epov

irus

es:

Cul

tiva

ted

Pla

nts

Aff

ecte

d, M

ode

of T

rans

mis

sion

, an

d G

eogr

aphi

cal

Dis

trib

utio

n

CM

II A

AB

Des

crip

tion

M

ode

of

Cul

tiva

ted

plan

ts i

nfec

ted

Geo

grap

hica

l V

irus

(ac

rony

m)

no.

or r

ef.

tran

smis

sion

a (t

hose

mos

t af

fect

ed e

cono

mic

ally

in

bol

d)

dist

ribu

tion

Ara

bis

mos

aic

(ArM

V)

16

N

Bla

ck c

urra

nt,

cele

ry,

cher

ry,

cucu

mbe

r,

Eur

ope,

Jap

an,

New

el

derb

erry

, fo

rsyt

hia,

gra

pevi

ne,

hop,

Z

eala

nd,

USS

R

lett

uce,

nar

ciss

us,

oliv

e, r

aspb

erry

, re

d cu

rran

t, r

huba

rb,

stra

wbe

rry,

tam

aril

lo

Arr

acac

ha A

(A

VA

) 21

6 U

A

rrac

acha

P

eru

Arr

acac

ha B

(AV

B)

270

U

Arr

acac

ha,

oca,

pot

ato

Peru

, B

oliv

ia

Art

icho

ke A

egea

n ri

ngs p

ot

Ran

a et

al.

(198

5)

U

Art

icho

ke

Gre

ece,

Tur

key

(AA

RSV

) A

rtic

hoke

Ita

lian

lat

ent

(AIL

V)

176

N

Art

icho

ke,

chic

ory,

gra

pevi

ne,

pela

rgon

ium

B

ulga

ria,

Gre

ece,

S.

Ital

y A

rtic

hoke

vei

nban

ding

(A

VB

V)

285

U

Art

icho

ke

S. I

taly

A

rtic

hoke

yel

low

rin

gspo

t 27

1 P

Art

icho

ke,

broa

d be

an,

card

oon,

Fre

nch

bean

, G

reec

e, I

taly

(A

YR

SV)

toba

cco

Blu

eber

ry l

eaf

mot

tle

(BLM

V)

267

P B

lueb

erry

, gr

apev

ine

N.

Am

eric

a C

assa

va A

mer

ican

lat

ent

(CsA

LV)

Wal

ter

et a

l. (1

989)

U

C

assa

va

Bra

zil,

Guy

ana

Cas

sava

gre

en m

ottl

e (C

GM

V)

Len

non

et a

l. (1

987)

U

C

assa

va

Sol

omon

Isl

ands

C

herr

y le

af r

oll

(CLR

V)

80,3

06

P B

irch

, bl

ackb

erry

, ch

erry

, de

lphi

nium

, E

urop

e, N

. A

mer

ica,

do

gwoo

d, e

lder

berr

y, e

lm,

lila

c, o

live

, N

ew Z

eala

nd,

USS

R

rasp

berr

y, r

huba

rb,

wal

nut

Che

rry

rasp

lea

f (C

RLV

) 15

9 N

A

pple

, ch

erry

, pe

ach,

ras

pber

ry

N.

Am

eric

a C

hico

ry y

ello

w m

ottl

e (C

YM

V)

132

U

Cel

ery,

chi

cory

, le

ttuc

e, p

arsl

ey

S. I

taly

C

ocoa

nec

rosi

s (C

NV

) 17

3 U

C

ocoa

W

. A

fric

a C

rim

son

clov

er l

aten

t (C

CLV

) K

ente

n et

al.

(198

0)

U

Cri

mso

n cl

over

E

urop

e, N

. A

mer

ica

Cyc

as n

ecro

tic

stu

nt

(CN

SV)

Kus

unok

i et

al.

(198

6)

U

Cyc

as r

evol

uta,

gla

diol

us

Japa

n D

ogw

ood

mos

aic

(DoM

V)

Bar

nett

et

al.

(198

9)

U

Com

us fl

orid

a U

nite

d S

tate

s G

rape

vine

Bul

gari

an l

aten

t (G

BLV

) 18

6 U

G

rape

vine

B

ulga

ria,

Por

tuga

l G

rape

vine

chr

ome

mos

aic

(GC

MV

) 10

3 U

C

eler

y, g

rape

vine

H

unga

ry,

Eng

land

G

rape

vine

fan

leaf

(G

FLV

) 28

N

G

rape

vine

W

orld

wid

e

Gra

pevi

ne T

hnis

ian

ring

spot

O

uert

ani

et a

l. (1

992)

U

G

rape

vine

T

unis

ia

(GT

RSV

) H

ibis

cus

late

nt r

ings

pot

(HLR

SV)

233

U

Hib

iscu

s N

iger

ia

Luc

erne

Aus

tral

ian

late

nt (

LALV

) 22

5 U

L

ucer

ne,

whi

te c

love

r A

ustr

alia

, N

ew Z

eala

nd

Luc

erne

Aus

tral

ian

sym

ptom

less

R

emah

et

al.

(198

6)

U

Luc

erne

A

ustr

alia

(L

ASV

) M

ulbe

rry

ring

spot

(M

RSV

) 14

2 N

M

ulbe

rry

Japa

n M

yrob

alan

lat

ent

ring

spot

16

0 U

M

yrob

alan

plu

m,

peac

h, s

wee

t ch

erry

S.

W.

Fra

nce

(MLR

SV)

Oli

ve l

aten

t ri

ngsp

ot (

OLR

SV)

301

U

Oli

ve

S. I

taly

P

each

ros

ette

mos

aic

(PR

MV

) 15

0 N

B

lueb

erry

, gr

apev

ine,

pea

ch

N.

Am

eric

a P

otat

o bl

ack

ring

spot

(PB

RSV

) 20

6 U

P

otat

o P

eru

Pot

ato

U (

PVU

) Jo

nes

et a

l. (1

983)

U

P

otat

o P

eru

Ras

pber

ry r

ings

pot

(RR

SV)

6,19

8 N

A

rtic

hoke

, bl

ack

curr

ant,

che

rry,

goo

sebe

rry,

E

urop

e, T

urke

y, U

SSR

gr

apev

ine,

nar

ciss

us,

rasp

berr

y, r

ed c

urra

nt,

stra

wbe

rry

Rub

us C

hine

se s

eed-

born

e (R

CSV

) B

arba

ra e

t al

. (1

985)

U

R

ubus

sp.

C

hina

S

atsu

ma

dwar

f (S

DV

) 20

8 U

S

atsu

ma

oran

ge

Japa

n, T

hrke

y S

traw

berr

y la

tent

rin

gspo

t (S

LRSV

) 12

6 N

B

lack

cur

rant

, ce

lery

, el

derb

erry

, gr

apev

ine,

E

urop

e na

rcis

sus,

oli

ve,

peac

h, r

aspb

erry

, re

d cu

rran

t, r

ose,

str

awbe

rry

Tob

acco

rin

gspo

t (T

RSV

) 17

,309

N

A

nem

one,

ash

, ap

ple,

bla

ckbe

rry,

blu

eber

ry,

N.

Am

eric

a br

oad

bean

, ch

erry

, cu

curb

its,

dog

woo

d,

eggp

lant

, el

derb

erry

, F

renc

h be

an,

glad

iolu

s, g

rape

vine

, ir

is,

pela

rgon

ium

, so

ybea

n, A

mer

ican

spe

arm

int,

tob

acco

T

omat

o bl

ack

ring

(TB

RV

) 38

N

A

rtic

hoke

, be

et,

cele

ry,

fors

ythi

a, g

rape

vine

, E

urop

e le

ek,

lett

uce,

nar

ciss

us,

onio

n, p

each

, po

tato

, pr

ivet

, ra

spbe

rry,

rob

inia

, st

raw

berr

y, s

wed

e, t

urni

p T

omat

o ri

ngsp

ot (

ToR

SV)

18,2

90

N

App

le,

cym

bidi

um,

glad

iolu

s, g

rape

vine

, N

. A

mer

ica,

Jap

an,

hydr

ange

a, n

arci

ssus

, pe

larg

oniu

m,

Pru

nus

USS

R

spp.

, ra

spbe

rry,

red

curr

ant,

str

awbe

rry,

to

bacc

o T

omat

o to

p ne

cros

is (

ToT

NV

) B

ancr

oft

(196

8)

U

Tom

ato

Uni

ted

Sta

tes

aN,

nem

atod

e ve

ctor

rep

orte

d (s

ee C

hapt

er 7

); P,

tra

nsm

itte

d in

pol

len

to t

he p

olli

nate

d pl

ant

(see

Cha

pter

8);

U,

vect

or u

nkno

wn.


in some other plant virus genera, such as the luteoviruses. Serologically related nepoviruses have RNA-2 molecules of the same size, and they also usually have nematode vectors belonging to the same genus (see Chapter 7, this volume).

II. HOST RANGES AND DISEASES CAUSED

A. General Features

Most nepoviruses have extremely wide natural and experimental host ranges. Table I shows that nepoviruses cause disease in a wide range of cultivated plants throughout the world, although the geographical distribution of most individual nepoviruses is somewhat restricted. The noteworthy exception is GFLY, which seems, along with its vector (Xiphinema index), to have been disseminated around the world with grapevine planting material carried by the early colonisers.

GFLV is undoubtedly the nepovirus of greatest economic importance, because of its worldwide distribution, because infected grapevines may suffer yield losses in excess of 50% (Rudel, 1985), and because the virus is difficult to eradicate from sites of virus outbreaks. Other nepoviruses, though more restricted in distribution, can have devastating effects on crops where they occur. In Britain, RRSV, with or without TBRV, or ArMV, with or without SLRSV, may kill raspberry and strawberry plants, or render them worthless, in sizeable areas of crop. In North America, TRSV can cause total losses of soybean crops and ToRSV can have very serious effects on almond, cherry, nectarine, and peach. The "blackline" disease of English walnut caused by CLRV is "the most important factor limiting walnut production in some regions and is a serious threat to walnut production in California" (Mircetich et a1., 1980).

Nepoviruses often cause few or no obvious symptoms when they infect wild plants, and this adaptation suggests that they should be regarded primarily as parasites of wild plants. Nepoviruses may also infect some cultivated plants symptomlessly, but the presence of viruliferous nematodes in a soil becomes evident-and economically important-when a crop is grown that reacts sensitively to infection. The resulting diseased plants occur in characteristic patches that may range from a few square meters to a few hectares in extent, reflecting the horizontal distribution of the nematode vector. When disease results from the use of infected planting material and not from spread by nematodes, affected plants are of course scattered through the crop in a more random manner. The same is true of those nepoviruses such as CLRV and BLMY, which seem not to have nematode vectors but spread in association with pollen and infect the pollinated mother plant (see Chapter 8).


Plants that are sensitive to nepoviruses often respond with some kind of leaf flecking or mottling, as with ArMV, RRSV, SLRSV, and TBRV in many cultivars of raspberry and strawberry, the precise nature of the symptom depending on the cultivar, time of year, and other factors. Ringspot symptoms (e.g., those caused by RRSV in Malling Jewel raspberry) are relatively infrequent in cultivated plant species, but are a common response of herbaceous test plants, especially of Nicotiana species. Other types of symptom found are vein yellowing or vein netting (as with GFLV in grapevine and ArMV in elderberry), leaf curling or rolling (as with RRSV in Norfolk Giant raspberry or CLRV in cherry), and "rasp leaf"-the production of narrow leaves with enations on their undersides-as with CRLV, ToRSV, or myrobalan latent ringspot virus (MLRSV) in cherry. In many plants, such as strawberry and raspberry, the foliar symptoms that develop in the spring may fade in midsummer, to reappear perhaps in the autumn but more usually in the following spring. A remarkable feature of experimental infection by nepoviruses of many herbaceous test plant species, especially of Nicotiana, is the phenomenon known as "recovery": after the acute phase of symptom development, the subsequently produced leaves are symptomless, though they contain virus, albeit in somewhat lower concentration than in the symptom-bearing leaves.

A fuller description, with illustrations, of the main types of symptom that nepoviruses cause is given by Murant (1981). The wide range of symptom types, the diversity of reaction of different cultivars of the same species, and the similarity of symptoms that different nepoviruses may induce in the same species make identification of the viruses difficult or impossible from symptoms alone. Tests on useful indicator hosts must be supported by the other diagnostic methods outlined in Section III of this chapter. Infection and symptom development in herbaceous indicator test plants is often greatly influenced by the growing conditions before and after inoculation, the nepovirus isolate, and, with some nepoviruses, the presence or absence of satellite RNAs. However, all the known nepoviruses infect Chenopodium quinoa, some inducing obvious symptoms in the inoculated and/or uninoculated leaves. Many nepoviruses also infect C. amaranticolor and/or C. murale, some inducing local lesions in these hosts. These Chenopodium species therefore have been used commonly for virus detection and assay and/or virus propagation. Other useful test species include Cucumis sativus, Nicotiana c1evelandii, and Phaseolus vulgaris.

B. Diseases Caused

It is impossible in the space available here to give detailed descriptions of all the diseases caused by nepovirusesj this section of this chapter gives brief accounts, with key references, of the diseases caused by individual nepoviruses in their most important crop hosts. For the reactions of important


diagnostic species, consult the references below and in Table I. Table I also shows which viruses are known to have nematode vectors (for detailed information, consult Chapter 7) and which are transmitted to the pollinated plant (see Chapter 8).

1. Arabis Mosaic Virus (ArMV)

Most isolates of ArMV are serologically similar, although those from hop (Humulus lupulus) and one from barley in Switzerland differ more than most others from the type strain. However, isolates can differ considerably in host range and symptomatology (Bock, 1966; Jones et a1., 1989). All tested isolates, including these serological variants from hop and barley, are transmitted by Xiphinema diversicaudatum (Harrison and Cadman, 1959; Jha and Posnette, 1959; Valdez et a1., 1974; D. J. F. Brown andA. T. Jones, unpublished data). This nematode also transmits SLRSV, and the two viruses may occur together in soils.

ArMV has a very wide natural host range. In addition to the cultivated plants listed in Table I, it is also reported from many wild plants, including Capsella bursa-pastoris, Lamium amplexicaule, Mentha arvensis, Polygonum aviculare, Sambucus nigra, Senecio vulgaris, Stellaria media, and Taraxacum officinale.

In red raspberry (Rubus idaeus) (Murant, 1970), ArMV causes chlorotic mottling, vein yellowing, or yellow speckling, depending on the cultivar. In sensitive cultivars the young canes are stunted ("yellow dwarf") and produce little or no fruit. Some cultivars are immune to some isolates but may be infected by others (Jones et a1., 1989).

ArMV induces a wide range of symptoms in strawberry (Fragaria x ananassa), including "mosaic" and "yellow crinkle," depending on the cultivar and virus strain (Lister, 1970). Stunting may be slight or so severe as to kill the plants within 1 or 2 years of infection. Leaves are commonly twisted, cupped, or crinkled and may show chlorotic mottle, yellow spots, blotchy mosaic, or streaks. Leaf symptoms tend to fade in midsummer.

In hop (Thresh et a1., 1972), ArMV causes "bare bine" (shoots are few and weak with small leaves ) and, in the cultivar Fuggle, "split leaf blotch" (leaves have yellow blotches and tend to split). Subsequent development of "nettlehead" (shoots stiff and erect and failing to climb, leaves with vein clearing and enations on the undersides) seems to be associated with the presence of a satellite RNA of molecular weight ca 75,000 (Davies and Clark, 1983; see Chapter 6).

In some barley cultivars, an ArMV isolate from Switzerland induced pronounced leaf yellowing (P. Giigerli, D. J. F. Brown, and A. T. Jones, unpublished data). In association with viruses of the prunus necrotic ringspot type, ArMV induces "rasp leaf" symptoms in cherry (Prunus cerasus) (Cropley, 1961b). ArMV has a very wide experimental host range. Schmelzer (1963a) reported infection of 93 species in 28 families.


2. Arracacha Virus A (AVA)

AVA (Jones and Kenten, 1978) is reported only from arracacha (Arracacha xanthorrhiza; Umbelliferae), an important root crop in Peru. The virus has not been returned to healthy arracacha, but infection in the field is associated with pronounced yellow mosaic of the young leaves. No strains are reported. Experimentally, AVA infected 38 of 63 species tested in 10 of 12 plant families, inducing distinct symptoms in many instances.

3. Arracacha Virus B (AVB)

AVB (Jones and Kenten, 1983), is reported from arracacha (Arracacha xanthorrhiza), oca (Oxalis tuberosa), and potato (Solanum tuberosum) in Peru and Bolivia. It causes no symptoms in experimentally infected potato; attempts to return it to healthy arracacha and oca failed. AVB has a wide experimental host range and is seed-borne in C. quinoa and potato. Strain 0, which occurs in oca and potato, is serologically distinguishable from strain T found in arracacha. Strain T does not infect potato. Strain 0 usually induces less severe symptoms in herbaceous plants than does strain T.

4. Artichoke Aegean Ringspot Virus (AARSV)

AARSV is the name we now propose for a virus isolated from artichoke (Cynara scolymus) in Greece and Turkey and formerly identified as a serotype of RRSV (Rana et al., 1985). Greek and Turkish isolates are serologically distinguishable but share 73 % sequence homology, whereas their homology with the English strain of RRSV is only 7% (Robinson and Clark, 1987). Naturally infected artichokes are either symptomless or display mild chlorotic mottling or yellow blotches, according to the variety (Rana et al., 1985). Search for the vector was unsuccessful (Roca et al., 1986). AARSV has a moderately wide experimental host range, infecting 22 species out of 40 inoculated in 7 dicotyledonous families.

5. Artichoke Italian Latent Virus (AlLV)

The type strain of AILV, originally isolated in Italy (Majorana and Rana, 1970) but reported also from Bulgaria (Jankulova et al., 1978), is serologically distinguishable from the Greek strain (Rana and Kyriakopoulou, 1982). The type strain is transmitted by Longidorus apulus (Roca et al., 1975) and the Greek strain by L. fasciatus (Roca et al., 1982). Infections in artichoke (Cynara scolymus) are usually symptomless, although stunting and generalized yellowing of the plants may be induced, especially by the Greek strain.

In chicory (Cichorium intybus), AlLV causes chlorotic mottling of the leaves, often accompanied by bright yellow spots (Vovlas et al., 1971). The virus has also been isolated from pelargonium (Pelargonium spp.) with se-


vere leaf malformations and from grapevines with fanleaflike symptoms. AILV also infects many wild plants, often symptomlessly (e.g., Crepis neglecta, Helminthia echioides, Hypochaeris aetnensis, Lactuca virosa, Urospermum dalechampii, and Lamium amplexicaule) (Quacquarelli et a1., 1976). The experimental host range of AlLV is wide, including 63 species in 12 dicotyledonous families (see, among others, Savino et a1., 1976).

6. Artichoke Vein Banding Virus (AVBV)

AVBV is of little economic importance. It is restricted in nature to artichoke, inducing in some cultivars transient chlorotic discolorations along the veins. The experimental host range of AVBV is moderately wide, including 20 species out of 71 inoculated in 7 dicotyledonous families (Gallitelli et a1., 1978).

7. Artichoke Yellow Ringspot Virus (AYRSV)

Isolates of AYRSV of Italian and Greek origin differ in host range but not serologically (Rana et a1., 1980). AYRSV is transmitted through seeds (up to 100%) and pollen of artificially and naturally infected weeds. In Nicotiana c1evelandii, transmission occurs also to pollinated plants (Kyriakopoulou et a1.,1985).

In artichoke (Cynara scolymus) and cardoon (e. cardunculus) (Rana et a1., 1980), AYRSV symptoms range, according to the season and the cultivar, from mild yellowing or a few scattered yellow blotches on the leaves to intense chrome yellow rings, lines, and oak leaf patterns, often followed by necrosis. Affected plants are stunted and their yield decreased. Their distribution in the field is erratic or in small groups.

Natural infection occurs also in tobacco (chlorotic or necrotic bands, rings, and oakleaf patterns), French bean and broad bean (Vicia laba) (stunting, diffuse yellowing, and leaf malformation), and in many wild plants, including Stellaria media, Beta maritima, Chrysanthemum segetum, Geranium molle, Papaver rhoeas, and Nicotiana glauca, inducing chlorotic to yellow rings, mottling, and, occasionally, malformation of the leaves (Kyriakopoulou et a1., 1985). The experimental host range is wide, including 56 species in 11 dicotyledonous families (Rana et a1., 1980).

8. Blueberry Leaf Mottle Virus (BLMV)

BLMV strain Ml from blueberry (Vaccinium corymbosum) in Michigan was shown (Ramsdell and Stace-Smith, 1979) to be closely serologically related to strain NY from grapevine (Vitis labrusca) in New York State (Uyemoto et a1., 1977). The virus seems to be restricted to these hosts in nature. There is a lack of evidence for a nematode vector (Childress and


Ramsdell, 1986) and the pattern of field spread suggests pollen transmission by bees (Childress and Ramsdell, 1987 j Boylan-Fett et a1., 1991).

Infected blueberry (cv. Rubel) plants exhibit general dieback of stems, leaf distortion, and mottling. The single infected grapevine (cv. Concord) plant observed showed delayed budbreak, differential growth of shoots, and pale green leaves. Fruit clusters were small with many aborted berries. Experimentally, BLMV infected a limited range of commonly used herbaceous test plants, comprising 23 species in 7 dicotyledonous families (Ramsdell and Stace-Smith, 1979 j Uyemoto et a1., 1977).

9. Cassava American Latent Virus (CsALV)

This virus was found in several cultivars of cassava (Manihot esculenta) from South America (Brazil, Guyana) that were also infected with cassava common mosaic potexvirus (Walter et a1., 1989). Cassava plants infected experimentally with CsALV alone showed no symptoms although the virus could be detected in the plants by enzyme-linked immunosorbent assay (ELISA). Apart from cassava (Euphorbiaceae), CsALV infected species from only two other families, Chenopodiaceae and Solanaceae.

10. Cassava Green Mottle Virus (CGMV)

CGMV is reported only from the Solomon Islands, infecting cassava (Manihot esculenta) (Lennon et a1., 1987). Mechanically inoculated cassava plants initially show systemic necrotic spotting and a mild green mottle. This is followed by partial recovery to give almost symptomless systemic infection. The virus has a wide experimental host range, including 30 species in 12 families. No serological relationship was detected to any of 18 nepoviruses.

11. Cherry Leaf Roll Virus (CLRV)

CLRV occurs in many wild and cultivated, mostly woody, plant species. It symptomlessly infects olive (Olea europaea) (Savino and Gallitelli, 1981) and rhubarb (Rheum rhaponticum) (Tomlinson and Walkey, 1967) and induces chlorotic ringspot, leaf patterns, yellow vein netting, or yellow spotting in species of Betula (birch) (Schmelzer, 1972aj Cooper and Atkinson, 1975), Euonymus (Larson et a1., 1990), Juglans (walnut) (Savino et a1., 1977 j

Cooper, 1979 j Mircetich et a1., 1980 j Mircetich and Rowhani, 1984), Sambucus (elderberry) (Schmelzer, 1966 j Hansen and Stace-Smith, 1971 j Jones and Murant, 1971), Rubus (raspberry) (Jones and Wood, 1978), and in Comus florida (dogwood) (Waterworth and Lawson, 1973) and Ptelea trifoliata (Schmelzer, 1972b). It causes "mosaic" and dieback in Ulmus americana (elm) (Swingle et a1., 1943), "leaf roll" and plant death in cherry (Cropley, 1961aj Schimanski et a1., 1975), and "blackline," a major disease associated

no A. F. MURANT ET AL.

with decline, in English walnut trees propagated on Juglans hindsii or Paradox (J. hindsii x J. regia) seedling rootstocks (Mircetich et a1., 1980; Mircetich and Rowhani, 1984). When present with ArMV, it causes ringspot of lilac (Syringa vulgaris) (Novak and Lanzova, 1975).

Many strains are reported. Isolates from different natural host species are distinguishable serologically, but those from a single host species are usually not. Serologically indistinguishable isolates may differ in virulence. All CLRV strains tested have a wide experimental host range, infecting species in more than 36 plant families (Schmelzer, 1966; Horvath, 1979).

12. Cherry Rasp Leaf Virus (CRLV)

Isolates of CRLV show minor differences in symptom severity in herbaceous hosts but appear to be antigenically similar (Hansen et a1., 1974; Jones et a1., 1985). In nature, the virus infects a narrow range of plant species. In addition to the plants listed in Table I, CRLV has been isolated from a few species of perennial weed growing in cherry and apple orchards, including Taraxacum officinale, Plantago major, and Balsamorhiza sagittata (Hansen et a1., 1974).

The most characteristic symptom of the disease in cherry is "rasp leaf": prominent enations appear between the lateral veins and along the midrib on the underside of the leaf. Affected leaves are narrow, folded, and deformed. Newly infected trees develop enations, usually on the lower leaves first, and the virus then spreads within the tree only slowly, especially within mature trees. Affected spurs and branches may die, giving the tree an open, bare appearance. Trees that are affected at an early stage remain stunted, produce few fruits, and some are killed (Nyland, 1976).

In apple (Malus sylvestris), CRLV induces "flat apple" disease (Parish, 1977): affected branches produce fruits that are about half as long as they are wide; the calyx lobes are prominent, the calyx basin is broad, and the stem cavity is shallow. Affected fruits are small and have prominent lenticels, particularly when immature. Leaves on affected branches are long and narrow and appear brittle and dry. A few years after infection, diseased trees become dwarfed and have a dense, bushy appearance.

The virus has also been recovered from stunted peach (Prunus persica) (Hansen et a1., 1974) and from symptomless red raspberry (Jones et a1., 1985; R. Stace-Smith, unpublished data). Experimentally, the virus infected 22 out of 24 commonly used herbaceous species, usually causing symptomless or mild infections (Hansen et a1., 1974).

13. Chicory Yellow Mottle Virus (ChYMV)

In chicory, Ch YMV causes bright yellow mottle, ringspots, and line patterns in the leaves (Vovlas et a1., 1971). Affected plants are commonly unmarketable. In the field they are distributed at random or in small groups.

NEPOVIRUSES: GENERAL PROPERTIES III

The virus is transmitted through a low proportion of chicory seeds, which may account in part for the recurrent infections in this crop (Vovlas, 1973). Naturally infected lettuce (Lactuca sativa) and celery (Apium graveolens) plants show yellow mottling of the leaves, whereas parsley (Petroselinum crispum) reacts primarily with malformation of the leaves.

The experimental host range of Ch YMV is moderately wide, including 32 species in 12 dicotyledonous families. The reactions of some hosts depend on whether the inoculum contains satellite RNA (Piazzolla and Rubino, 1984; see also Chapter 6, this volume). For example, Nicotiana glutinosa develops chlorotic or necrotic concentric, targetlike ringspots when infected with satellite-free cultures of Ch YMV but is infected symptomlessly when satellite RNA is also present.

14. Cocoa Necrosis Virus (CNV)

In cocoa (Theobroma cacao), CNV (Kenten, 1972) induces leaf necrosis, defoliation, and dieback, often causing death of seedlings infected young. Leaves from plants in the chronic phase of infection may develop irregular translucent areas along the veins (Thresh and Tinsley, 1959; Owusu, 1971). In a limited host range investigation, CNV infected 15 of 30 species in 7 of 11 dicotyledonous families (Kenten, 1972).

15. Crimson Clover Latent Virus (CCLV)

CCLV has been found only in crimson clover (Trifolium incarnatum) (Kenten et a1., 1980) in which it occurs symptomlessly. Seed-borne infection (>90%) has been detected in crimson clover seed originating from Europe and North America. No strains are reported. CCLV cannot be transmitted directly from crimson clover to herbaceous plants; the virus must first be partially purified. Experimentally, CCLV inoculum in Chenopodium quinoa sap infected only 3 of 27 species from 7 plant families.

16. Cycas Necrotic Stunt Virus (CNSV)

CNSV, found in Cycas revoluta in Japan, is of note as being the first virus from a gymnosperm to have been characterized (Kusunoki et a1., 1986; Hanada et a1., 1986). The only other plant in which it has been found in nature is cultivated gladiolus (Gladiolus sp.) (K. Hanada, personal communication). Although not yet tested against antisera to all other nepoviruses, CNSV seems from its properties to be a distinct member of the genus. Infected Cycas revoluta plants show dwarfing and twisting of young leaves and chlorotic or necrotic spots on mature leaves; the plants decline progressively year by year, and severely affected plants may be killed.

The experimental host range of an isolate from C. revoluta was restricted to 11 species in 4 families: Aizoaceae, Amaranthaceae, Cheno-


podiaceae, and Cycadaceae (Kusunoki et al., 1986). An isolate from gladiolus seems to have a wider host range, including some species of Solanaceae (K. Hanada, personal communication).

17. Dogwood Mosaic Virus (DoMV)

This virus is reported only from wild plants of dogwood (Cornus florida) growing in localized areas of South Carolina (Barnett et al., 1989). It is distantly serologically related to ArMV and.GFLV, but seems sufficiently distinct to be regarded as a separate nepovirus. The natural vector is unknown; DoMV was not transmitted by a Scottish population of Xiphinema diversicaudatum (D. J. F. Brown, unpublished data, cited by Barnett et al., 1989). DoMV infected plants in 10 families, but caused only mild symptoms in some hosts.

18. Grapevine Bulgarian Latent Virus (GBLV)

Several serologically distinguishable strains of GBLV are known. The best studied are the type strain from Bulgaria (Martelli et al., 1977) and the Portuguese strain (Gallitelli et al., 1983). The only known natural host is grapevine (Vitis vinifera), several cultivars of which are infected symptomlessly, although symptoms may appear when GBLV occurs in mixed infections with other viruses (Dimitrijevic, 1985). Its effect on the yield and quality of the grapes is unknown. The experimental host range is narrow, comprising 10 species in 4 dicotyledonous families (Martelli et al., 1977; Gallitelli et al., 1983).

19. Grapevine Chrome Mosaic Virus (GCMV)

Isolates of GCMV, although serologically very similar to or indistinguishable from one another, are of two major types, chromogenic and distorting, according to the symptoms induced in grapevines (Vitis spp.). In European grapes (Vitis vinifera) and several American rootstock hybrids, chromogenic strains cause discolorations of varying type and intensity in the leaves, which, in extreme cases, turn uniformly chrome yellow or whitish and may develop interveinal or marginal necrosis. Distorting strains cause chlorotic mottling and malformation of the leaves, usually accompanied by deformation of shoots and canes such as double nodes, short internodes, or fasciations. With most isolates, affected vines lack vigor, bear little or no crop, and tend to decline and die within a few years of infection. Infection also causes physiological disturbances: altered nitrogen metabolism and growth hormone production, diminished photosynthetic activity, and decreased sugar and pigment content (Martelli et al., 1986). Diseased plants, especially those with chrome-yellow symptoms, have a patchy distribution in the field. Vines interplanted in these patches become diseased and show


symptoms within 1 or 2 years. In celery (Apium graveolens) (Hollings et a1., 1969), GCMV was found associated with yellow vein disease. The experimental host range is rather narrow, including 14 species in 5 dicotyledonous families (Martelli and Quacquarelli, 1972).

20. Grapevine Fanleaf Virus (GFLV)

The'many biological variants (Le., chromogenic and distorting strains) of GFLV are remarkably uniform serologically, though distantly related to ArMV (Dias and Harrison, 1963). However, a strain recently isolated in Tunisia is serologically distinguishable both from ordinary GFLV isolates and from ArMV (Savino et a1., 1985). In nature, GFLV infects only Vitis spp. European grapes and American hybrid rootstocks are both infectible, showing syndromes that, according to the infecting strain, are known as:

1. Fanleaf: Leaves may be variously and severely distorted, mottled, asymmetrical, puckered, and with acute denticulations. Canes and shoots show abnormal branching, double nodes, short internodes, fasciations, and zigzag growth. Bunches are reduced in number and size, ripen irregularly, have imperfectly developed ("shot") berries, and poor berry setting. Affected vines may decline slowly or, rarely, die quickly. Symptoms develop early in the spring, remaining visible throughout the vegetative season, although they may fade in midsummer.

2. Yellow mosaic: Bright chrome-yellow discolorations develop early in the spring and may affect all vegetative parts of the vine (leaves, canes, tendrils, flower bunches); they range from a few scattered yellow spots, sometimes appearing as rings or lines, to variously extended mottling or blotching or yellowing of the entire leaf. Malformations of leaves and canes are usually mild. Fruit clusters are small and with "shot" berries. In hot climates, summer foliage has the normal green color but the yellowed spring growth turns whitish.

"Vein banding," a syndrome formerly attributed to infection by a particular strain of GFLV (Goheen and Hewitt, 1962), is now considered to be caused primarily by grapevine yellow speckle viroid(s) (reviewed by Martelli, 1993).

Physiological alterations of GFLV-infected vines consist of deranged metabolism of protein, carbohydrate, and phenol; decreased transpiration rate, photosynthetic, and acid phosphatase activity; reduced pigment and sugar contents; and increased peroxidase activity (see review by Martelli et a1., 1986). GFLV infections are detrimental in many ways, causing progressive decline and death of the vines, decreased yield (average losses of 67% over a 6-year period were reported in 1985 by Rudel), poor fruit quality, shortening of the productive life of the vineyard, low proportion of graft "take," reduced rooting ability of propagating material, and decreased resistance to adverse climatic conditions (Bovey et a1., 1974). Experimentally, the virus infected some 50 species in 7 dicotyledonous families.


21. Grapevine Tunisian Ringspot Virus (GTRSV)

GTRSV is of no economic importance. It was isolated in Tunisia from a vine with mild mottling and leaf deformation (Ouertani et a1., 1992). The experimental host range is rather restricted, including 14 species out of 19 inoculated in 6 dicotyledonous families.

22. Hibiscus Latent Ringspot Virus (HLRSV)

HLRSV (Brunt et a1., 1980) is reported only from Hibiscus rosa-sinensis in Nigeria. Infected plants rarely show symptoms, but very young leaves are sometimes faintly chlorotic. No strains are reported. Experimentally, HLRSV infected 22 of 73 species tested in 7 of 20 plant familiesj most were infected symptomlessly or showed only indistinct symptoms.

23. Lucerne Australian Latent Virus (LALV)

LALV is reported only from Australasia infecting lucerne (Medicago sativa) (Blackstock, 1978 j Jones et a1., 1979) and white clover (Trifolium repens) (Forster and Morris-Krsinich, 1985). It is usually symptomless in these hosts but sometimes induces faint systemic chlorotic or necrotic linepatterns. LALV is seed-borne in lucerne but not in white clover.

The lucerne (L) and white clover (We) isolates are distinguishable serologically and in behavior and symptomatology in some hosts. LALV-L does not systemically infect white clover, and LALV-WC does not systemically infect lucerne. Australian and New Zealand isolates of LALV-L also differ in reactions in some test plants.

24. Lucerne Australian Symptomless Virus (LASV)

LASV (Remah et a1., 1986), originally described as a strain of LALV (Blackstock, 1978), is reported only from Australia, infecting lucerne (Medicago sativa) symptomlessly. It has a narrow experimental host range and occurs symptomlessly or induces only faint symptoms in most herbaceous hosts. It is seed-borne in Chenopodium quinoa. It is difficult to maintain in culture during winter months (Remah et a1., 1986).

25. Mulberry Ringspot Virus (MRSV)

In mulberry (Moms alba), its only known natural host, MRSV causes mosaic and ringspot symptoms. Leaf enations are associated with some isolates. In a limited host range study, the virus infected 10 species in 5 dicotyledonous families (Tsuchizaki et a1., 1971 j Tsuchizaki, 1975).

NEPOVIRUSES: GENERAL PROPERTIES llS

26. Myrobalan Latent Ringspot Virus (MLRSV)

This virus, isolated from symptomless myrobalan plum trees (Prunus cerasifera), is transmissible by grafting to several Prunus species. Plum (P. domestica) is infected symptomlessly, whereas peach (P. persica) reacts with mottling, mild distortion of the leaves and apical rosetting, and sweet cherry (P. avium) develops enations on the undersides of the leaves (Dunez et a1., 1971). Experimentally, the range of herbaceous hosts is narrow, including 18 species (several with latent infection) in 6 dicotyledonous families (Dunez et a1., 1971).

27. Olive Latent Ringspot Virus (OLRSV)

A virus isolated from symptomless olive (Olea europaea) trees. Experimentally, OLRSV infected 7 species in 5 dicotyledonous families (Savino et a1., 1983).

28. Peach Rosette Mosaic Virus (PRMV)

Isolates of PRMV are similar in their host reactions and only one isolate from grapevine was serologically distinguishable from others tested (Dias and Cation, 1976). In nature, PRMV infects relatively few crop and weed hosts. In addition to the cultivated plants listed in Table I, the virus was detected in Rumex crispus, Solanum carolinense, and Taraxacum officinale but not in 13 other weed species associated with diseased grapevines (Ramsdell and Myers, 1978).

Affected peach trees show delayed foliation, chlorotic mottling, and distortion to the early formed leaves and shortening of the internodes to produce a rosette appearance ("rosette mosaic"). Normal-looking branches are interspersed with affected ones. Leaf chlorosis is evident early in the growing season and the chlorotic areas vary in size, shape, and color intensity. Leaves formed later in the season are of near-normal size and are darker green than normal leaves (Klos, 1976).

In grapevine (Vitis labrusca cv. Concord), PRMV induces delayed dormancy breaking, late and uneven bloom, small and uneven berry clusters, leaf deformity and mottling, and vines that are lighter green than normal. Cane growth is short and crooked, giving the vines an umbrellalike growth habit. After infection for several years, vines become unproductive and may die (Ramsdell and Myers, 1978).

In blueberry, PRMV has been observed only on the cultivars Jersey and Berkley (Ramsdell and Gillett, 1981), causing leaves to become strap-shaped or crescent-shaped. Symptoms were not uniformly distributed over the entire bush. A few terminal leaves were narrow and elongated but none of the leaves showed chlorotic or necrotic lesions.


Experimentally, PRMV infects a range of commonly used herbaceous test plants. In many hosts, the inoculated leaves become infected but remain symptomless (Dias and Cation, 1976).

29. Potato Black Ringspot Virus (PBRSV)

PBRSV is reported only from Peru, infecting cultivated potato. The type isolate was obtained from plants showing necrotic spotting of young tip leaves (Salazar and Harrison, 1977h another isolate was obtained from plants with "calico" symptoms (Fribourg, 1977). However, no calico symptoms developed in potato experimentally infected with the calico isolate (Salazar and Harrison, 1978b) but many potato cultivars and Solanum species developed systemic necrotic ringspots (Salazar and Harrison, 1978a). Most plants infected from the tuber showed no symptoms (Salazar and Harrison, 1978a). The type and calico isolates differ only slightly serologically and in symptoms in some hosts (Salazar and Harrison, 1978b). Experimentally, PBRSV infected 40 of 41 species tested in 11 plant families; most developed severe systemic symptoms.

30. Potato Virus U (PVU)

PVU is a little-studied virus from Peru (Jones et a1., 1983). Its particles sediment as three major components, which suggests that there are two nucleic acid species. It has been found naturally only in a single potato (Solanum tuberosum) plant, which showed bright yellow leaf markings ("calico"). Experimentally it was transmitted to 44 species in 7 plant families, but in several potato cultivars tested, infection was confined to inoculated leaves or reached only a very low concentration in uninoculated leaves.

31. Raspberry Ringspot Virus (RRSV)

Isolates of RRSV that have been most studied belong to two serotypes, Scottish and English. These serotypes are serologically only distantly related to some isolates from grapevine in Germany (Vuittenez et a1., 1970; Jones et a1., 1994) and cherry in Switzerland (Vuittenez et a1., 1970; Rana et a1., 1985; Jones, 1985), and these latter viruses ought perhaps to be regarded as distinct viruses. Isolates of the Scottish serotype can occur in soils together with the beet ringspot serotype of TBRV, which has the same nematode vector, Longidorus elongatus. RRSV occurs in a very wide range of dicotyledonous, and some monocotyledonous species. In addition to the cultivated plants listed in Table I, it is reported from many wild plants, including Caps ella bursa-pastoris and Stellaria media.

In red raspberry (Murant, 1970), RRSV induces vein yellowing, chlorotic ringspots or flecks, or leaf-curling symptoms depending on the cultivar. Leaf symptoms may fade in midsummer but return in the autumn. Plants of


sensitive cultivars are stunted and die within 2-3 years of the first appearance of symptoms; other cultivars merely show decreased vigor. Some raspberry cultivars are immune to some strains of the virus (Jones et a1., 1989).

In strawberry (Lister, 1970), RRSV induces chlorotic spots or rings or generalized chlorosis, depending on the cultivar, the symptoms tending to fade in midsummer. The plants become progressively stunted and die prematurely. In cherry, in association with viruses of the prunus necrotic ringspot type, RRSVinduces rasp leaf symptoms (Cropley, 1961b; Anon., 1963). In grapevine, an English serotype was associated with fanleaf-type symptoms (Jones et a1., 1994).

Experimentally, RRSV has been shown to infect species in more than 14 dicotyledonous families, inducing symptoms in many (Murant, 1978). However, an English serotype from grapevine in Germany induced no symptoms or, at best, very faint transient symptoms in herbaceous test plants (Jones et a1., 1994).

32. Rubus Chinese Seed-Borne Virus (RCSV)

RCSV (Barbara et a1., 1985) is reported from only a single symptomless seedling grown from seed of an unidentified Rubus species sent from China. Experimentally, RCSV infected 23 of 39 species in 6 of 8 plant families, most of them symptomlessly.

33. Satsuma Dwarf Virus (SDV)

Isolates of SDV (Usugi and Saito, 1979) are closely related serologically. The major disease caused by SDV is stunting, leaf malformation, and rosetting of satsuma orange (Citrus unshiu). Leaves produced in the spring are down-curled in the shape of an inverted boat. Viruses related to SDV are associated with citrus mosaic, navel orange infectious mottling, and natsudaidai dwarf diseases. Many citrus plants show only transient symptoms. The virus is restricted in nature to citrus, but it can be transmitted mechanically to a relatively wide range of herbaceous species.

34. Strawberry Latent Ringspot Virus (SLRSV)

Isolates of SLRSV from the United Kingdom are serologically very similar, but Italian isolates from olive and peach (Savino et a1., 1979; Belli et a1., 1980), raspberry (Vegetti et a1., 1979; A. F. Murant, unpublished data), and grapevine (Credi et a1., 1981) were distinguishable from the type strain and, some of them at least, from each other by spur formation in gel diffusion serological tests. SLRSV occurs in soils together with ArMV, which has the same nematode vector, Xiphinema diversicaudatum. SLRSV has a wide natural host range, infecting, in addition to the cultivated plants listed in Table I, many wild plants including CapseJ1a bursa-pastoris, Lamium am-


plexicaule, Mentha arvensis, Sambucus nigra, Senecio vulgaris, Stellaria media, and Taraxacum officinale.

In red raspberry (Murant, 1970), SLRSV often occurs together with ArMV and the plants show symptoms not conspicuously different from those caused by ArMV alone. However, when found alone in cv. MaIling Jewel (Taylor and Thomas, 1968), SLRSV induced severe stunting and yellow blotches and speckles on the leaves of primo canes with poor development of the lateral shoots on fruiting canes. Some cultivars are immune.

Similarly, the reactions of strawberry (Lister, 1970) to infection with SLRSV are largely unknown because it has usually been found together with ArMV; however, plants of cv. Cambridge Vigour graft-inoculated with SLRSV were stunted and had yellow blotches on the leaves. Isolates of SLRSV were found associated with a rosetting disease of peach in Italy (Belli et al., 1980) and mixed infections of SLRSV and prune dwarf virus induced a severe decline disease of peach in France (Scotto La Massese et al., 1973). In the grapevine hybrid rootstock 106/8, SLRSV was associated with chlorotic mottling, asymmetry, and malformation of the leaves and obvious stunting of the plants (Credi et al., 1981). In some olive cultivars in Italy (Marte et al., 1986) and Portugal (Henriques et al., 1992), SLRSV induces malformation of leaves and fruits, whereas in other cultivars the infection is symptomless. Leaves of celery infected with SLRSV were crinkled, distorted, and decreased in size ("strap-leaf") (Walkey and Mitchell, 1969). Roses (Rosa spp.) infected with SLRSV were stunted and their leaves were distorted and had yellow angular flecks (Harrison, 1967; Ikin and Frost, 1976). The experimental host range of SLRSV is very wide. Schmelzer (1969) reported infection of 126 species in 27 families, most of them symptomlessly.

35. Tobacco Ringspot Virus (TRSV)

Four serological variants of TRSV have been identified from tobacco and one from watermelon (Citrullus vulgaris) (Gooding, 1970). Variants differing mainly in symptom expression have also been reported. TRSV occurs in both annual and perennial crops and causes severe disease problems in those regions of North America where the nematode vector also occurs. In addition to the cultivated plants listed in Table I, the virus infects many biennial and perennial weed hosts, often symptomlessly (Tuite, 1960; Rush and Gooding, 1970).

In tobacco, the virus causes concentric rings and line patterns of chlorotic and necrotic tissue on the leaves. When located in interveinal regions, the spots are circular and the lines are necrotic; when centered near larger veins, the spots are irregular and the symptoms follow the veins and their branches. New leaves on systemically infected plants may show no obvious symptoms, although they contain the virus (recovery). Severely affected plants may nevertheless be dwarfed with small leaves of poor quality (Lucas, 1975).


Of the many diseases caused by TRSV, "bud blight" of soybean (Glycine max) is the most severe and causes the greatest losses (Sinclair and Shurtleff, 1975). Plants infected when less than 5 weeks old have shortened internodes, fewer nodes, and are severely stunted. Plants infected when more mature show milder symptoms. The most striking symptom is curving of the terminal bud, with other buds on the plant later becoming brown and brittle.

In blueberry, TRSV causes "necrotic ringspot" disease (Lister et a1., 1963; Converse and Ramsdell, 1982). Sensitive cultivars are stunted and unproductive and show extensive twig dieback, with new leaves produced in the spring showing chlorotic or necrotic spots, rings, or line patterns. Severely affected leaves are reduced in size and deformed. Cultivars that are less susceptible may show tip die back followed by recovery.

TRSV is prevalent in cucurbits in Texas (McLean and Meyer, 1961) and Wisconsin (Sinclair and Walker, 1956), causing stunting, leaf mottling and malformation, and decreased fruit set and size. As the plant matures, leaves become tattered and necrotic and internodes are shortened, producing a compact plant with brittle leaves and stems. Many affected plants tend to recover. Experimentally the virus infected species in more than 17 dicotyledonous and monocotyledonous families (Price, 1940).

36. Tomato Black Ring Virus (TBRV)

TBRV isolates are of two major serotypes: one contains the type (Smith, 1946), lettuce ringspot (Smith and Short, 1959), and potato bouquet (Kohler, cited in Harrison, 1958) isolates; the other contains the beet ringspot (Harrison, 1958) and potato pseudo-aucuba (Bercks, 1962) isolates. Isolates serologically similar to the beet ringspot strain sometimes occur together in soils with isolates of the Scottish serotype of RRSV, which share the same nematode vector, Longidorus elongatus.

In nature, TBRV infects a very wide range of monocotyledonous and dicotyledonous species. In addition to the cultivated plants listed in Table I, TBRV is reported from many wild plants, including Caps ella bursa-pastoris, Cerastium vulgatum, Geranium dissectum, Lamium amplexicaule, Myosotis arvensis, Polygonum aviculare, P. convolvulus, Spergula arvensis, Stellaria media, Veronica agrestis, and V persica.

In red raspberry (Murant, 1970), TBRV causes chlorotic mottling, ringspotting, or leaf curling depending on the cultivar, with some stunting and decrease of yield; fruit may be deformed ("crumbly") as a result of abortion of some of the drupelets and overdevelopment of others. There is much more severe disease when RRSV is also present. Some raspberry cultivars are immune to some TBRV isolates (Jones et a1., 1989).

In strawberry (Lister, 1970), TBRV causes chlorotic spots and rings or more extensive areas of chlorosis, depending on the cultivar or time of year. Leaves produced later in the season may be symptomless, but in subsequent years symptoms return and the plants become progressively dwarfed and


eventually die. When RRSV is also present, disease is much more severe and death of the plants is accelerated.

TBRV also causes ringspotting diseases of beet (Harrison, 1957) and lettuce (Smith and Short, 1959), "bouquet" and "pseudo-aucuba" diseases of potato (Solanum tuberosum) (GehringandBercks,1956; Harrison, 1958), and unnamed diseases of grapevine (Stellmach, 1970), leek (Allium porrum) and onion (A. cepa) (Calvert and Harrison, 1963), and swede (Brassica napus) and turnip (E. rapa) (Harrison, 1957). "Black ring" of tomato (Smith, 1946) seems of no economic importance.

Experimentally, TBRV infects a very wide range of commonly used herbaceous test plants. Schmelzer (1963b) reported infection of 76 species in 25 families of dicotyledonous plants.

37. Tomato Ringspot Virus (ToRSV)

ToRSV causes serious diseases of perennial crops in United States and Canada. The virus has been disseminated to many countries in infected plant material, but natural spread is largely confined to North America. The type strain was originally isolated from tomato in eastern United States, but, despite the name, natural infections in tomato are rare. Many variants, giving slightly different symptoms in herbaceous host plants, and several serologically distinguishable isolates are known (Gooding, 1963; Bitterlin and Gonsalves, 1988).

In nature, the virus occurs mostly in perennial crops. In addition to the cultivated plants listed in Table I, ToRSV is reported from many wild plants, including Stellaria media, Lamium amplexicaule, Taraxacum officinale, Oxalis corniculata, Plantago major, Fragaria virginian a, Rumex acetosella, and Trifolium pratense (Powell et a1., 1982).

Symptoms of ToRSV in red raspberry depend on the cultivar, the duration of the infection, and the time of year. Plants normally show no symptoms in the season of infection, but in the following year some leaves of the primocanes show yellow rings, line patterns, or vein chlorosis. Chronic symptoms are delayed foliage in the spring, chlorosis of leaves on the fruiting canes, and fruit that is malformed and crumbly (Stace-Smith, 1984).

"Yellow bud mosaic," a serious disease of almond (Prunus amygdalus), nectarine (P. persica var. nectarina), peach, plum, and sweet cherry, is also caused by ToRSV. Newly infected peach or nectarine trees develop yellow blotches or spots. Blotching is accompanied by leaf distortion, necrosis, and retardation of growth of some buds, producing tufts of pale yellow leaves. "Stem pitting" is commonly associated with ToRSV infection of peach and cherry in eastern United States (Smith et a1., 1973).

ToRSV is implicated in the etiology of "brownline disease" of prune in North and South America (Cummins and Gonsalves, 1986; Auger, 1989) and is the major cause of "apple union necrosis and decline." The symptoms are pitting, invagination, and necrosis in the woody cylinder at the graft union.


Symptoms are thought to be caused by the presence of the virus in the rootstock but not in the scion of diseased trees (Stouffer et a1., 1977); cultivars propagated on MM106 clonal rootstock are particularly sensitive. In grapevine, ToRSV causes "yellow vein" disease in California (Gooding and Hewitt, 1962) and decline symptoms in northern United States and Canada. Some cultivars show only slight mottling of the foliage, increased cane size, and a slight-to-moderate fruit set; others develop a pronounced chlorotic mottling of the foliage, shortened internodes, severe stunting, and fruit clusters with berries of varying size and maturity (Dias, 1977; Gilmer and Uyemoto, 1972). Experimentally, the virus has infected species in more than 35 dicotyledonous and monocotyledonous families.

38. Tomato Top Necrosis Virus (ToTNV)

This virus has been little studied; the original culture has been lost and no isolates are currently available. ToTNV was first designated "tomato ringspot" because it caused a distinctive ringspot disease of tomato (Samson and Imle, 1942), but was later renamed (Bancroft, 1968) to distinguish it from ToRSV.

Natural infection with ToTNV was reported only in tomato and a few weed species (Solanum carolinense, Nicandra physaloides, and Datura stramonium) growing in or near tomato fields. Infected tomato plants showed curling and necrosis of the more actively growing shoots, and brown necrotic streaks and rings on the petioles and adjacent stems of affected leaves. Fruits that were young at the time of infection showed corky, superficial rings (Samson and Imle, 1942).

Experimentally, ToTNV infected several species of Amaranthaceae, 1 of Martyniaceae, and 19 species or varieties of Solanaceae out of the 78 species representing 27 plant families that were tested (Samson and Imle, 1942). Another virus isolate had a similar but not identical host range (Bancroft, 1968).

39. Other Possible Members

Among several imperfectly described viruses that are possible candidates for membership of the genus is one associated with cherry rosette disease in Switzerland (Brown et a1., 1994).

III. DETECTION, DIAGNOSIS, AND QUANTITATIVE ASSAY

A. Electron Microscopy

The particles of nepoviruses have diameters of ca 25-28 nm and, except in uranyl stains, have strongly angular (usually hexagonal) outlines (Harrison


and Nixon, 1960). As a rule, the particles are not damaged by the commonly used negative stains (phosphotungstate, uranyl acetate, uranyl formate, ammonium molybdate) when used at the standard concentrations and pH values (Milne, 1984). Details of the surface structure are not readily discernible even in good micrographs. The particles are of two or three types: those penetrated by negative stain, thought to correspond to the noninfective T component, and those partially or not penetrated by the stain, which possibly correspond to the M and B centrifugal components (Murant, 1981). The relative proportions of the three particle types depend on the virus strain and also on the history of the particle preparation. All types of particle are present in concentrated purified virus preparations and are readily seen in leaf dips or in crushed tissue extracts from infected herbaceous hosts, provided that they are not in the recovery phase. Leaf dips from naturally infected woody plants are seldom successful because of low virus concentration. At their vertices, SLRSV and RCSV particles are penetrated by negative stain (Murant, 1981; Barbara et a1., 1985), a feature that may be a result of their atypical particle properties (see Chapter 6, this volume).

With some nepoviruses, notably CLRV, SLRSY, TRSV, and TBRV, tubular structures containing rows of virus particles can be seen in preparations from crushed apical meristems (Walkey and Webb, 1968) or, less commonly, in tissues of young systemically infected leaves of artificially or naturally infected hosts. These tubules (which are not specific to nepoviruses) are ca 40-50 nm in diameter, have walls 5-6 nm thick, sometimes with a banded substructure, and may be up to 4 /-lm long and contain up to 200 particles (Walkey and Webb, 1968, 1970; Hicks, 1985). Virus-containing tubules were frequently observed in extracts of apical leaves of glasshouse-grown cuttings of grapevines naturally infected with GFLV and ToRSV (Corbett and Podleckis, 1985).

B. Serology

Host range and symptomatology, and the morphology of virus particles in crude sap extracts, may suggest the presence of a nepovirus, but there is no substitute for serological tests as a means of identification. Table I in Chapter 6 (this volume) shows that some nepoviruses are serologically related. No serological relationships have been detected, using polyclonal antisera, between members of different serological clusters; even within some of the clusters, especially those based on ArMY, CLRY, and TBRY, relationships may be very distant and may be detectable only by using high-titered antiserum. It is therefore advisable to make serological tests with antisera to each of the major serotypes.

Most nepoviruses are moderately good immunogens and rabbit antisera with titers of 1/200 to 1/2000 in gel double-diffusion tests may be obtained by a variety of immunization procedures. Any of the standard serological tests


that are commonly used in plant virology are applicable to nepoviruses, but the following are the most generally useful.

l. Double Diffusion in Gels

The Ouchterlony double-diffusion test in agar or agarose gels is simple, inexpensive, and relatively quick (precipitin lines develop within 6-24 hr) and has been widely used for identifying nepoviruses, for determining the degree of relatedness among them, and for detecting and determining differences between strains or serotypes of the same nepovirus. The concentration of nepovirus particles in sap from naturally infected hosts (woody perennials, shrubs, and vines in particular) is usually too low to give a visible precipitation line (Thomas, 1980), though reactions were obtained with GBLV in sap of naturally infected grapevine leaves (Martelli et al., 1977) and with CLRV in sap from cherry buds (Cropley, 1960). Reactions are more readily obtained with clarified concentrated extracts of naturally infected host leaves (e.g., GFLV in grapevine) (Vuittenez et al., 1964) or with undiluted crude sap from glasshouse-grown herbaceous hosts showing fresh symptoms. With many nepoviruses, the sap from inoculated leaves of Chenopodium quinoa gives good results. For more precise serological work (i.e., assessing relationships between viruses, titrating antisera), the use of purified concentrated virus preparations is recommended.

2. Agglutination Tests

When inert carriers of relatively large size (e.g., red blood cells, polystyrene latex particles) are coated with antigen or antibody, they agglutinate (form visible clumps) when exposed to homologous reactants (antibodies or antigens, respectively). Two such tests have been used for detecting nepoviruses.

a. Latex Agglutination (LA)

In this test, the carrier particles are polystyrene latex spheres sensitized with antibodies either directly (Bercks and Querfurth, 1969; Abu Salih et al., 1968b) or after previous coating with staphylococcal protein A (Querfurth and Paul, 1979; Torrance, 1980). The latter method, called PALLAS (protein A-coated latex-linked antiserum), permits the use of low-titered antisera. The LA test is simple, inexpensive, and quick [visible flocculation appears within 5-10 min in the procedure used by Abu Salih et al. (1968b)] and the latex-antibody conjugates may be cold-stored for more than 3 years without loss of activity (Bercks and Querfurth, 1969). LA tests are 25- to WOO-fold more sensitive than precipitin tests in tubes or gels, according to estimates by different authors reviewed by Van Regenmortel (1982).


Several nepoviruses have been detected by LA in sap of artificially infected (Abu Salih et a1., 1968b; Torrance, 1980) and naturally infected (Bercks, 1967, 1968) plants. The test was used for routine field-detection of nepoviruses in grapevine in Germany in the late 1970s (H. 1. Paul, personal communication), but now has been superseded by enzyme immunoassay (Walter et a1., 1984).

b. Passive Hemagglutination (PHA)

The method followed with plant viruses is to coat tanned sheep erythrocytes with the purified ,,-globulin fraction of the antiserum and then allow the cells to react with the homologous antigen. Abu Salih et a1. (1968a,b) found that the PHA test could detect RRSV in sap of Nicotiana c1evelandii and ArMV in purified virus preparations with a sensitivity up to 40 times greater than the LA test and up to 1000 times greater than precipitin tests in tubes or gels. The PHA test detected as little as 50 ng/ml of RRSV and was only about ten times less sensitive than infectivity assay for detecting RRSV in raspberry (H. S. Abu Salih, A. F. Murant, and M. J. Daft, unpublished data).

With some hosts, satisfactory results in LA and PHA tests may depend on the use of special extraction media. With grapevine, for instance, the addition of 1 % caffeine or 2.5% nicotine to extraction buffers was reported to be critical (Bercks and Querfurth, 1969). With raspberry, too, 2.5% nicotine is usually used to prevent leaf tannins from precipitating the virus particles.

3. ELISA

Since its first application in plant virology (Clark and Adams, 1977), ELISA has gained favor progressively so as to become the technique of choice for large-scale routine serological detection. Technical procedures were reviewed by Clark and Bar -Joseph (1984) and Van Regenmortel and Dubs (1993).

ELISA is about 1000-5000 times more sensitive than double diffusion in gels (Jankulova et a1., 1982) and 200 times more sensitive than LA (Thomas, 1980). Mowat (1986, and unpublished data) found ELISA to be much more sensitive and reliable than infectivity assay for detecting TBRV in narcissus. ELISA detected less than 30 ng/ml of ArMV and less than 10 ng/ml of RRSV in purified virus preparations (Clark and Adams, 1977) and has been widely used to detect nepoviruses.

Monoclonal antibodies to GFLV were successfully used in ELISA to detect the virus in different grapevine organs (Huss et a1., 1986) and for distinguishing virus isolates (Huss et a1., 1987). Using ArMV as immunogen, Frison and Stace-Smith (1992) investigated the possibility of raising broadspectrum monoclonal antibodies that could be used for quarantine purposes. Although extended cross-reactivity among different nepoviruses (ArMV,


GFLV, RRSV, CLRV, and TBRV) was observed in several hybridoma lines, none of the monoclonals was able to recognize all the nepoviruses tested.

For nepovirus detection, both "direct" and "indirect" ELISA procedures (for details, see Clark and Bar-Joseph, 1984) have proved equally sensitive (Walter et a1., 1984), although the high serological specificity of direct ELISA may result in failure to detect some strains, as with ToRSV (Stace-Smith, 1984); indirect ELISA, which has lower specificity, may be preferable in these instances. The advantages of indirect ELISA (high sensitivity and low background) for the detection of GFLV in grapevine sap were retained by using F(ab')2 antibody fragments, without the need to employ two different antisera (Rowhani, 1992). A blocking procedure (Powell and Derr, 1983) may be useful in ELISA for assessing the degree of relatedness between nepoviruses. Alkaline phosphatase (Bovey et a1., 1982) or horseradish peroxidase (Kolber et a1., 1981) have each been used with equal success as enzyme markers. However, both are reported to give nonspecific reactions under some conditions. Thus, using alkaline phosphatase in conventional direct ELISA, Mink et a1. (1985) found that extracts of tip leaves of rapidly growing healthy apple shoots gave positive reactions in summer, though not in spring. Using horseradish peroxidase in indirect ELISA for detecting ArMV, RRSV, SLRSV, and TBRV, Jones and Mitchell (1987) found that extracts of roots but not shoots of Cucumis sativus, Chenopodium quinoa, and Petunia hybrida seedlings transplanted into nematode-infested soil to act as bait plants gave very high absorbance (A4S0 ) values, whether or not the plants became infected with virus. Values were very much lower in undisturbed healthy seedlings unless their leaves or cotyledons were first rubbed with a finger wetted with buffer or plant sap. This nonspecific reaction was apparently caused by natural substrate-oxidizing activity in roots that was stimulated by mechanical injury to tops or roots during transplanting.

The extraction method and the extraction buffer may affect the sensitivity of ELISA in some hosts. With grapevines, for instance, grinding tissues with a mortar and pestle or with an ingenious crusher in a plastic bag (Giigerli, 1984) gave more reproducible results than using the Polliihne roller machine (Bovey et a1., 1985). Also with grapevine, Engelbrecht (1982) found it necessary to add 1-2% nicotine to standard extracting buffer to obtain reliable readings, whereas Kearns and Mossop (1984) used high-molarity phosphate buffer containing thioglycollic acid and sodium diethyldithiocarbamate, and Walter and Etienne (1987) used Tris-HCI buffer containing 0.8% NaCI and 2 % polyvinyl pyrrolidone.

4. Dot-Immunobinding Assay (DIA)

In the DIA, sap expressed from infected tissues is spotted on a nitrocellulose membrane and exposed in succession to the virus-specific antiserum, enzyme-labeled antirabbit immunoglobulin G, and the substrate, which develops a colored reaction (Van Regenmortel and Dubs, 1993). In


different hosts, DIA enabled TRSV and ToRSV to be detected with sensitivities comparable to those of ELISA (Powell, 1987).

5. Immunoelectron Microscopy

The two main techniques are antibody coating, often referred to as "decoration," and immunosorbent electron microscopy (ISEM) (Milne and Lesemann, 1984). A combination of ISEM and decoration is perhaps the most satisfying way of identifying viruses serologically because it provides the visible proof of particles specifically recognized by the antibodies. For nepovirus detection, ISEM has been found to be as sensitive as ELISA (Savino et a1., 1981; Bovey et a1., 1982) or more sensitive (1. Torrance, 1979, quoted in Murant, 1981; Thomas, 1980). ISEM was used to detect ArMV, GBLV, GCMV, and GFLV in extracts of naturally infected grapevine (Russo et a1., 1982) and ArMY, GCMV, GFLY, RRSV, SLRSV, and TBRV in extracts of viruliferous nematodes (Roberts and Brown, 1980). Antibody coating is less sensitive than ISEM for virus detection, but its sensitivity may be increased by specific labeling of the bound antibodies with protein A-gold complex (Louro and Lesemann, 1984).

6. Newer Techniques

Double-stranded RNA preparations made from infected plants (Dodds et a1., 1984) give characteristic electrophoretic patterns that may be of diagnostic value, as with RRSV (Jones et a1., 1986) and ToRSV (Kurppa and Martin, 1986) in raspberry. However, for woody plants generally, this method seems unlikely to be sufficiently convenient, reliable, or sensitive to provide a means of screening stocks for infection. Molecular techniques such as nucleic acid hybridization and polymerase chain reaction (PCR) technology (reviewed by Hull, 1993) are increasingly applied to nepovirus detection and identification. Cloned complementary DNA probes to genomic RNAs of ArMV (Jelkmann et a1., 1988), ToRSV and ArMV (Hadidi and Hammond, 1989), TBRV and GCMV (Bretout et a1., 1989), and, more recently, GFLVand its satellite RNA (Fuchs et a1., 1991; Saldarelli et a1., 1993) enabled detection of homologous sequences in tissue extracts of different hosts. In these reports, dot blot hybridization was done with radioactive probes, but equally satisfactory results were reported when a digoxigenin -labeled probe was used to detect GFLV in total nucleic acid extracts from infected grapevine samples (Gemmrich et a1., 1993).

PCR technology, combined or not with immunocapture (IC) (Jansen et a1., 1990), is already a useful complement of, and may become a substitute for, ELISA because the detection of target nucleic acid is extremely efficient. For instance, reverse transcription PCR was reported to detect ArMV RNA in extracts of 5 mg of grapevine or 1 mg of C. quinoa leaf (Ipach et a1., 1992) and could detect as little as 128 fg of GFLV RNA (Rowhani et a1., 1993). IC/PCR


applied to sap expressed from GFLV-infected grapevine leaves or cortical scrapings gave excellent and reproducible results with a much lower limit of detection than ELISA. Dilutions beyond which virus was not detected were reported to be 1:1280 (ELISA) and 1:20,140 (Ie/peR) for cortical scrapings (Brandt and Himmler, 1993), and 1:3200 (ELISA) and 1:80,000 (Ie/peR) for leaf extracts (Nolasco and De Sequeira, 1993a).

By combining Ie/peR with restriction fragment length polymorphism or single-stranded conformation polymorphism analysis, Nolasco and De Sequeira (1993b) showed that a remarkable variation in genome constitution occurred in GFLV from different plants in the same vineyard. They suggested that this technique is a valuable tool for the quick identification of naturally occurring virus variants without the need for sequencing.

REFERENCES

Abu Salih, H. S., Murant, A. E, and Daft, M. J., 1968a, Comparison of the passive haemagglutination and bentonite flocculation tests for serological work with plant viruses, f. Gen. Virol. 2:155.

Abu Salih, H. S., Murant, A. E, and Daft, M. J., 1968b, The use of antibody-sensitized latex particles to detect plant viruses, f. Gen. ViroI. 3:299.

Anon., 1963, Virus diseases of fruit trees, Rep. East Malling Res. Stn. 1962:27. Auger, J., 1989, Tomato ringspot virus (TomRSV) associated with brownline disease of prune

trees in Chile, Acta Hortic. 235:197. Bancroft, J. B., 1968, Tomato top necrosis virus, Phytopathology 58:1360. Barbara, D. J., Ashby, S. C., and McNamara, D. G., 1985, Host range, purification and some

properties of rubus Chinese seed-borne virus, Ann. AppI. BioI. 107:45. Barnett, O. W., Reddick. B. B., Burrows, P. M., and Baxter, 1. W., 1989, Characterization of

dogwood mosaic nepovirus from Comus florida, Phytopathology 79:951. Belli, G., Fortusini, A., and Vegetti, G., 1980, Properties of a strain of strawberry latent ringspot

virus associated with a rosetting disease of peach in northern Italy, Acta Phytopathol. Acad. Sci. Hung. 15:113.

Bercks, R., 1962, Serologische Dberkreuzreaktionen zwischen Isolaten des Tomatenschwarzringflecken-Virus, PhytopathoI. Z. 46:97.

Bercks, R., 1967, Vorkommen und Nachweis von Viren in Reben, Weinberg und Keller 14:151. Bercks, R., 1968, Dber den Nachweis des Himbeerringflecken-Virus (raspberry ringspot virus) in

Reben, PhytopathoI. Z. 62:169. Bercks, R., and Querfurth, G., 1969, Weitere methodische Untersuchungen uber den Latextest

zum serologischen Nachweis pflanzenpathogener Viren, Phytopathol. Z. 65:243. Bitterlin, M. W., and Gonsalves, D., 1988, Serological grouping of tomato ringspot virus isolates:

Implications for diagnosis and cross-protection, Phytopathology 78:278. Blackstock, J. McK., 1978, Lucerne transient streak and lucerne latent, two new viruses of

lucerne, Aust. f. Agric. Res. 29:291. Bock, K. R., 1966, Arabis mosaic and Prunus necrotic ringspot viruses in hop (Humulus lupulus

1.), Ann. AppI. BioI. 57:131. Bovey, R., Brugger, J. J., Simon, J.-1., and Jaquinet, A., 1974, La selection sanitaire de la vigne en

Suisse romande, Rev. Suisse Vitic. Arboric. Hortie. 6:77. Bovey, R., Brugger, J. J., and Gugerli, P., 1982, Detection of fanleaf virus in grapevine tissue

extracts by enzyme-linked immunosorbent assay (ELISA) and immunoelectron microscopy, (IEM), in: Proceedings of the 7th Meeting International Council for the Study of Viruses and


Virus Diseases of the Grapevine, Niagara Falls 1980 (A. J. McGinnis, ed.) pp. 259-275, Agriculture Canada, Vineland, Ontario, Canada.

Bovey, R., Giigerli, P., and Brugger, J. J., 1985, Some improvements in the detection of grapevine fanleaf virus by enzyme· linked immunosorbent assay (ELISA) and immunoelectron microscopy (IEM), in: Proceedings of the 8th Meeting International Council for the Study of Viruses and Virus Diseases of the Grapevine, Bari 1984, (G. P. Martelli and W. B. Hewitt, eds.), pp. 208-209, Mediterranean Phytopathological Union, Bologna, Italy.

Boylan-Pett, W., Ramsdell, D. C., Hoopingarner, R. A., and Hancock, J. F., 1991, Honeybee foraging behavior, in-hive survival of infectious pollen·borne blueberry leaf mottle virus and transmission of the virus in highbush blueberry, Phytopathology 81:1407.

Brandt, S., and Himmler, G., 1993, Detection of grapevine fanleaf virus from woody material by using immunocapture polymerase chain reactions, in: Extended Abstracts of the 11th Meeting of the International Council for the Study of Viruses and Virus Diseases of the Grapevine, Montreux, 1993, p. ISO, Federal Agricultural Research Station of Changins, Nyon, Switzerland.

Breece, J. R., and Hart, W. H., 1959, A possible association of nematodes with the spread of peach yellow bud mosaic virus, Plant Dis. Rep. 43:989.

Bretout, c., Candresse, T., Le Gall, 0., Brault, v., Ravelonandro, M., and Dunez, J., 1989, Virus and RNA-specific molecular hybridization probes for two nepoviruses, Acta Hortie. 235:23l.

Brown, D. J. F., Grunder, J., Hooper, D. J., Klingler, J., and Kunze, P., 1994, Longidorus arthensis sp. n. (Nematoda: Longidoridae) a vector of cherry rosette disease caused by a new nepovirus in cherry trees in Switzerland, Nematologica 40:133.

Brunt, A. A., Barton, R. J., Phillips, S., and Lana, A. 0., 1980, Hibiscus latent ringspot virus, a newly recognised virus from Hibiscus rosa-sinensis (Malvaceae) in western Nigeria, Ann. Appl. Biol. 96:37.

Cadman, C. H., 1956, Studies on the etiology and mode of spread of Scottish raspberry leaf curl disease, r. Hortic. Sci. 31:11l.

Cadman, C. H., 1963, Biology of soil-borne viruses, Annu. Rev. Phytopathol. 1:143. Calvert, E. L., and Harrison, B. D., 1963, Outbreaks of tomato black ring virus in onion and leek

crops in Northern Ireland, Hortic. Res. 2:115. Childress, A. M., and Ramsdell, D. c., 1986, Lack of evidence for a nematode vector of blueberry

leaf mottle virus (BBLMV), Acta Hortic. 186:87. Childress, A. M., and Ramsdell, D. C., 1987, Bee-mediated transmission of blueberry leaf mottle

virus via infected pollen in highbush blueberry, Phytopathology 77:167. Clark, M. F., and Adams, A. N., 1977, Characteristics of the microplate method of enzyme

linked immunosorbent assay for the detection of plant viruses, r. Gen. Virol. 34:475. Clark, M. F., and Bar-Joseph, M., 1984, Enzyme immunosorbent assays in plant virology, in:

Methods in Virology (K. Maramorosch and H. Koprowski, eds.), Vol. 7, pp. 51-58, Academic Press, New York.

Converse, R. H., and Ramsdell, D. c., 1982, Occurrence of tomato and tobacco ringspot viruses and of dagger and other nematodes associated with cultivated highbush blueberries in Oregon, Plant Dis. 66:710.

Cooper, J. I., 1979, Virus Diseases of Trees and Shrubs, Institute of Terrestrial Ecology, Cambridge.

Cooper, J.I., and Atkinson, M. A., 1975, Cherry leaf roll virus causing a disease of Betula spp. in the United Kingdom, Forestry 48:193.

Corbett, M. K., and Podleckis, E. v., 1985, Membrane-associated spherical particles in extracts and tissues of virus-infected grapevines, Phytopathol. Mediterr. 24:157.

Credi, R., Babini, A. R., Betti, L., Bertaccini, A., and Gelli, C., 1981, A distinctive isolate of strawberry latent ringspot virus from grapevines in Italy, Phytopathol. Mediterr. 20:56.

Cropley, R., 1960, Serological detection of a virus in cherry trees with a leaf roll disease, Nature 188:875.

Cropley, R., 1961a, Cherry leaf-roll virus, Ann. Appl. Biol. 49:524.


Cropley, R., 1961b, Viruses causing rasp-leaf and similar diseases of sweet cherry, Ann. Appl. Biol. 49:530.

Cummins, J., and Gonsalves, D., 1986, Constriction and decline of the "Stanley" prune associated with tomato ringspot virus, J. Am. Soc. Hortic. Sci. 111:315.

Davies, D. 1., and Clark, M. F., 1983, A satellite-like nucleic acid of arabis mosaic virus associated with hop nettlehead disease, Ann. Appl. Biol. 103:439.

Debrot, E. A, 1964, Studies on a strain of raspberry ringspot virus occurring in England, Ann. Appl. Biol. 54:183.

Dias, H. F., 1977, Incidence and geographic distribution of tomato ringspot virus in DeChaunac vineyards in the Niagara peninsula, Plant Dis. Rep. 61:24.

Dias, H. F., and Cation, D., 1976, The characterization of a virus responsible for peach rosette mosaic and grape decline in Michigan, Can. J. Bot. 54:1228.

Dias, H. F., and Harrison, B. D., 1963, The relationship between grapevine fanleaf, grapevine yellow mosaic and arabis mosaic viruses, Ann. Appl. Biol. 51:97.

Diener, T. 0., and Schneider, I. R., 1966, The two components of tobacco ringspot virus: Origin and properties, Virology 29:100.

Dimitrijevic, B., 1985, Reactions of host plants to grapevine Bulgarian latent virus in mixed infections with other viruses from grapevine, Phytopathol. Mediterr. 24:12.

Dodds, J. A, Morris, T. J., and Jordan, R. 1., 1984, Plant viral double-stranded RNA, Annu. Rev. Phytopathol. 22:151.

Dunez, J., Delbos, R., Desvignes, J.-C., Marenaud, C., Kuszala, J., and Vuittenez, A., 1971, Mise en evidence d'un virus de type ring spot sur Prunus cerasifera, Annales de Phytopathologie (numero hors serie):1l7.

Engelbrecht, D. J., 1982, Indexing grapevines for grapevine fanleaf virus by enzyme-linked immunosorbent assay, in: Proceedings of the 7th Meeting International Council for the Study of Viruses and Virus Diseases of the Grapevine, Niagara Falls 1980 (A. J. McGinnis, ed.), pp. 277-282, Agriculture Canada, Vineland, Ontario, Canada.

Forster, R. 1. S., and Morris-Krsinich, B. AM., 1985, A distinct strain of lucerne Australian latent virus in white clover in northern New Zealand, Ann. Appl. Biol. 107:449.

Fribourg, C. E., 1977, Andean potato calico strain of tobacco ringspot virus, Phytopathology 67:174. Frison, E. A., and Stace-Smith, R., 1992, Cross-reacting and heterospecific monoclonal anti

bodies produced against arabis mosaic nepovirus, ,. Gen. Virol. 73:2525. Fromme, F. P., Wingard, S. A, and Priode, C. N., 1927, Ringspot of tobacco: An infectious disease

of unknown cause, Phytopathology 17:321. Fuchs, M., Pinck, M., Etienne, 1., Pinck, 1., and Walter, B., 1991, Characterization and detection

of grapevine fanleaf virus by using cDNA probes, Phytopathology 81:559. Gallitelli, D., Rana, G. 1., and Di Franco, A., 1978, n virus della scolorazione perinervale del

carciofo, Phytopathol. Mediterr. 17:1. Gallitelli, D., Savino, V., and De Sequeira, O. A., 1983, Properties of a distinctive strain of

grapevine Bulgarian latent virus, Phytopathol. Mediterr. 22:27. Gehring, F., and Bercks R., 1956, Untersuchungen iiber das Bukett- und Pseudo-Aucuba-Virus

der Kartoffel, Phytopathol. Z. 27:215. Gemmrich, A. R., Link, G., and Seidel, M., 1993, Detection of grapevine fanleaf virus (GFLV) in

infected grapevines by non-radioactive nucleic acid hybridization, Vitis 32:237. Gilmer, R. M., and Uyemoto, J. K., 1972, Tomato ringspot virus in "Baco Noir" grapevines in

New York, Plant Dis. Rep. 56:133. Goheen, A. C., and Hewitt, W. B., 1962, Vein banding, a new virus disease of grapevines, Am. J.

Enol. Vitic. 13:73. Gooding, G. V., 1963, Purification and serology of a virus associated with the grape yellow vein

disease, Phytopathology 53:475. Gooding, G. Y., 1970, Natural serological strains of tobacco ringspot virus, Phytopathology

60:708. Gooding, G. Y., and Hewitt, W. B., 1962, Grape yellow vein: Symptomatology, identification and the

association of a mechanically transmissible virus with the disease, Am. J. Enol. Vitic. 13:193.

130 A. E MURANT ET AL.

Giigerli, P., 1984, Une methode simple pour Ie broyage de tissu vegetal, Rev. Suisse Vi tic. Arboric. Hortic. 16:87.

Hadidi, A., and Hammond, R. W., 1989, Construction of molecular clones for identification and detection of tomato ringspot and arabis mosaic viruses, Acta Hortic. 235:223.

Hanada, K., Kusunoki, M., and Iwaki, M., 1986, Properties of virus particles, nucleic acid and coat protein of cycas necrotic stunt virus, Ann. Phytopathol. Soc. Japan 52:422.

Hansen, A. J" and Stace-Smith, R., 1971, Properties of a virus isolated from golden elderberry, Sambucus nigra aurea, Phytopathology 61:1222.

Hansen, A. T., Nyland, G., McElroy, E D., and Stace-Smith, R., 1974, Origin, cause, host range and spread of cherry rasp leaf disease in North America, Phytopathology 64:721.

Harrison, B. D., 1956, Soil transmission of Scottish raspberry leaf-curl disease, Nature 178:553.

Harrison, B. D., 1957, Studies on the host range, properties and mode of transmission of beet ringspot virus, Ann. Appl. Biol. 45:462.

Harrison, B. D., 1958, Relationship between beet ringspot, potato bouquet and tomato black ring viruses, J. Gen. Microbiol. 18:450.

Harrison, B. D., 1967, The transmission of strawberry latent ringspot virus by Xiphinema diversicaudatum (Nematoda), Ann. Appl. BioI. 60:405.

Harrison, B. D., and Cadman, C. H., 1959, Role of a dagger nematode (Xiphinema sp.) in outbreaks of plant diseases caused by arabis mosaic virus, Nature 184:1624.

Harrison, B. D., and Nixon, H. L., 1960, Purification and electron microscopy of three soil-borne plant viruses, Virology 12:104.

Harrison, B. D., Mowat, W. P., and Taylor, C. E., 1961, Transmission of a strain of tomato black ring virus by Longidorus elongatus (Nematoda), Virology 14:480.

Harrison, B. D., Finch, T. T., Gibbs, A. T., Hollings, M., Shepherd, R. T., Valenta, v., and Wetter, C., 1971, Sixteen groups of plant viruses, Virology 45:356.

Harrison, B. D., Murant, A. E, and Mayo, M. A., 1972a, Evidence for two functional RNA species in raspberry ringspot virus, J. Gen. Virol. 16:339.

Harrison, B. D., Murant, A. E, and Mayo, M. A., 1972b, Two properties of raspberry ringspot virus determined by its smaller RNA, J. Gen. Virol. 17:137.

Henderson, R. G., 1931, Transmission of tobacco ringspot by seed of petunia, Phytopathology 21:225.

Henriques, M.1. C., Rei, E T., Leitiio, E A., Serrano, T. E, and Potes, M. E, 1992, Virus diseases of Olea europaea L. cultivars. I. Immunodiagnosis of strawberry latent ringspot virus, Phytopathol. Mediterr. 31:127.

Hewitt, W. B., Raski, D. T., and Goheen, A. C., 1958, Nematode vector of soil-borne fanleaf virus of grapevines, Phytopathology 48:586.

Hicks, R. G. T., 1985, An electron microscope study of tubules in homogenates of Chenopodium quinoa infected with strawberry latent ringspot virus, Phytopathol. Z. 112:359.

Hollings, M., Stone, O. M., and Martelli, G. P., 1969, Celery yellow vein and Hungarian chrome mosaic viruses, Rep. Glasshouse Crops Res. Inst. 1968:102.

Horvath, T., 1979, New artificial hosts and non-hosts of plant viruses and their role in the identification and separation of viruses. XIII. Nepovirus group (CLRV subgroup): Cherry leaf roll virus, Acta Phytopathol. Acad. Sci. Hung. 14:319.

Hull, R., 1993, Nucleic acid hybridization procedures, in: Diagnosis of Plant Virus Diseases (R. E. E Matthews, ed.), pp. 253-271, CRC Press, Boca Raton, FL.

Hull, R., Milne, R. G., and Van Regenmortel, M. H. V., 1991, A list of proposed standard acronyms for plant viruses and viroids, Arch. Virol. 120:151.

Huss, B., Walter, B., Etienne, L., and Van Regenmortel, M. H. v., 1986, Grapevine fanleaf virus detection in various grapevine organs using polyclonal and monoclonal antibodies, Vitis 25:178.

Huss, B., Muller, S., Sommermeyer, G., Walter, B., and Van Regenmortel, M. H. v., 1987, Grapevine fanleaf virus monoclonal antibodies: Their use to distinguish different isolates, J. Phytopathol. 119:358.


Ikin, R., and Frost, R. R., 1976, Virus diseases of roses. II. Strawberry latent ringspot virus R/1:2.6 or 2 x 1.6/38: S/S:S/Ne, Phytopathol. Z. 87:205.

Ipach, U., Altmayer, B., and Eichhorn, K. W., 1992, Detection of arabis mosaic virus using the polymerase chain reaction (PCR), Vitis 31:213.

Jankulova, M., Savino, v., Gallitelli, D., Quacquarelli, A., and Martelli, G. P., 1978, Isolation of artichoke Italian latent virus from the grapevine in Bulgaria, in: Proceedings of the 6th Meeting International Council for the Study of Viruses and Virus Diseases of the Grapevine, Cordova 1976, Monograf Instituto Nacional de Investigaciones Agrarias (No. 18), pp. 143-148, Instituto Nacional de Investigaciones Agrarias, Madrid, Spain.

Jankulova, M., Eskenazy, M., Bakardzhieva, N., and Georgieva, P., 1982, ELISA for the quantitative determination of grapevine fanleaf virus, Z. PfJanzenkr. PfJanzenschutz 89:18.

Jansen, R. W., Siegl, G., and Lemon, S. L., 1990, Molecular epidemiology of human hepatitis A virus defined by an antigen-capture polymerase chain reaction, Proc. Natl. Acad. Sci. USA 87:2867.

Jelkmann, W., Maiss, E., Breyel, E., and Casper, R., 1988, Production and use of cDNA clones from arabis mosaic virus, Ann. Appl. BioI. 113:483.

Jha, A., and Posnette, A. F., 1959, Transmission of a virus to strawberry plants by a nematode (Xiphinema sp.), Nature 184:962.

Jones, A. T., 1985, Antigenic variants of raspberry ringspot virus (RRV), Rep. Scott. Crop Res. Inst. 1984:181.

Jones, A. T., and Mitchell, M. J., 1987, Oxidising activity in root extracts from plants inoculated with virus or buffer that interferes with ELISA when using the substrate 3;3;5',5'tetramethylbenzidine, Ann. Appl. BioI. 111:359.

Jones, A. T., and Murant, A. F., 1971, Serological relationship between cherry leaf roll, elm mosaic and golden elderberry mosaic viruses, Ann. Appl. BioI. 69:11.

Jones, A. T., and Wood, G. A., 1978, The occurrence of cherry leaf roll virus in red raspberry in New Zealand, Plant Dis. Rep. 62:835. .

Jones, A. T., Forster, R. L. S., and Mohamed, N. A., 1979, Purification and properties of Australian lucerne latent virus, a seed-borne virus having affinities with nepoviruses, Ann. Appl. BioI. 92:49.

Jones, A. T., McElroy, F. D., and Brown, D. J. F., 1981, Tests for transmission of cherry leaf roll virus using Longidorus, Paralongidorus and Xiphinema nematodes, Ann. Appl. BioI. 99:143.

Jones, A. T., Mayo, M. A., and Henderson, S. J., 1985, Biological and biochemical properties of an isolate of cherry rasp leaf virus from red raspberry, Ann. Appl. BioI. 106:1Ol.

Jones, A. T., Abo El-Nasr, M. A., Mayo, M. A., and Mitchell, M. J., 1986, Association of dsRNA species with some virus-like diseases of small fruits, Acta Hortic. 186:63.

Jones, A. T., Mitchell, M. J., and Brown, D. J. F., 1989, Infectibility of some new raspberry cultivars with arabis mosaic and raspberry ringspot viruses, and further evidence for variation in British isolates of these two nepoviruses, Ann. Appl. BioI. 115:57.

Jones, A. T., Brown, D. 1. F., McGavin, W. J., Riidel, M., and Altmayer, B., 1994, Properties of an unusual isolate of raspberry ringspot virus from grapevine in Germany and evidence for its possible transmission by Paralongidorus maximus, Ann. Appl. BioI. 124:595.

Jones, R. A. c., and Kenten, R. H., 1978, Arracacha virus A, a newly recognised virus infecting arracacha (Arracacha xanthorrhiza; Umbelliferae) in the Peruvian Andes, Ann. Appl. Biol. 90:85.

Jones, R. A. c., and Kenten, R. H., 1983, Arracacha virus B, CMI/AAB Descriptions of Plant Viruses No. 270.

Jones, R. A. c., Fribourg, C. E., and Koenig, R., 1983, A previously undescribed nepovirus isolated from potato in Peru, Phytopathology 73:195.

Kahn, R. P., 1956, Seed transmission of the tomato-ringspot virus in the Lincoln variety of soybeans, Phytopathology 46:295.

Kearns, C. G., and Mossop, D. W., 1984, Detection of nepoviruses of Vitis vinifera in New Zealand using enzyme-linked immunosorbent assay (ELISA), N. Z. r. Agric. Res. 27:431.


Kenten, R. H., 1972, The purification and some properties of cocoa necrosis virus, a serotype of tomato black ring virus, Ann. Appl. Biol. 7l:119.

Kenten, R. H., Cockbain, A. J., and Woods, R. D., 1980, Crimson clover latent virus-a newly recognised seed-borne virus infecting crimson clover (Trifolium incarnatum), Ann. Appl. Biol. 96:79.

Klos, E. J., 1976, Rosette mosaic, in: Virus Diseases and Noninfectious Disorders of Stone Fruits in North America, Agric. Handbook 437, pp. 135-138, U.S. Dept. Agriculture, Washington, DC.

Kolber, M., Lehoczky, J., Pacsa, S., and Kobza, S., 1981, Detection of grapevine fanleaf virus (GFV) with ELISA technique in the leaf samples collected in various phenophases and in deep frozen leaf extracts, Kertgazdasiig 13:1l.

Kurppa, A., and Martin, R. R., 1986, Use of double-stranded RNA for detection and identification of virus diseases of Rubus species, Acta Hortic. 186:5l.

Kusunoki, M., Hanada, K., Iwaki, M., Chang, M. U., Doi, Y., and Yora, K., 1986, Cycas necrotic stunt virus, a new member of nepoviruses found in Cycas revoluta. Host range, purification, serology and some other properties, Ann. Phytopathol. Soc. Japan 52:302.

Kyriakopoulou, P. E., Rana, G. 1., and Roca, F., 1985, Geographic distribution, natural host range, pollen and seed transmissibility of artichoke yellow ringspot virus, Ann. lnst. Phytopathol. Benaki, Athens 14:139.

Larson, R. C., Gergerich, R. c., and Kim, K. S., 1990, Characterization and ultrastructural studies of a nepovirus from Euonymus, Phytopathology: 80:134.

Lennon, A. M., Aiton, M. M., and Harrison, B. D., 1987, Purification and properties of cassava green mottle, a previously undescribed virus from the Solomon Islands, Ann. Appl. Biol. 110:545.

Lister, R. M., 1960, Transmission of soil-borne viruses through seed, Virology 10:547. Lister, R. M., 1964, Strawberry latent ringspot: A new nematode-borne virus, Ann. Appl. Biol.

54:167. Lister, R. M., 1970, Nematode-borne viruses as pathogens in strawberry, in: Virus Diseases of

Small Fruits and Grapevines (N. W. Frazier, ed.), pp. 32-42, University of California Press, Berkeley, CA.

Lister, R. M., and Murant, A. F., 1967, Seed-transmission of nematode-borne viruses, Ann. Appl. Bioi. 59:49.

Lister, R. M., Raniere, 1. c., and Varney, E. H., 1963, Relationships of viruses associated with ringspot diseases of blueberry, Phytopathology 53:103l.

Louro, D., and Lesemann, D.-E., 1984, Use of protein A-gold complex for specific labelling of antibodies bound to plant viruses. I. Viral antigens in suspensions, J. Virol. Methods 9:107.

Lucas, J. B., 1975, Ringspot, in: Diseases of Tobacco (G. B. Lucas, ed.), pp. 523-530, Biological Consulting Associates, Raleigh, NC.

Majorana, G., and Rana, G. 1., 1970, Un nuovo virus latente isolato da carciofo in Puglia, Phytopathol. Mediterr. 9:193.

Marte, M., Gadani, E, Savino, V., and Rugini, E., 1986, Strawberry latent ringspot virus associated with a new disease of olive in Central Italy, Plant Dis. 70:17l.

Martelli, G. P. (ed.), 1993, Graft-Transmissible Diseases of Grapevines. Handbook for Detection and Diagnosis, FAO Publication Division, Rome.

Martelli, G. P., and Quacquarelli, A., 1972, Hungarian chrome mosaic of grapevine and tomato black ring: Two similar but unrelated plant viruses, Annales de Phytopathologie (numero hors serie):123.

Martelli, G. P., Gallitelli, D., Abracheva, P., Savino, V., and Quacquarelli, A., 1977, Some properties of grapevine Bulgarian latent virus, Ann. Appl. Biol. 85:5l.

Martelli, G. P., Graniti, A., and Ercolani, G. 1., 1986, Nature and physiological effects of grapevine diseases, Experientia 42:933.

McLean, D. M., and Meyer, H. M., 1961, A survey of cucurbit viruses in the Lower Rio Grande Valley of Texas: Preliminary report, Plant Dis. Rep. 45:137.


Milne, R. G., 1984, Electron microscopy for the identification of plant viruses in in vitro preparations, in: Methods in Virology (K. Maramorosch and H. Koprowski, eds.), Vol. 7, pp. 87-120, Academic Press, New York.

Milne, R. G., and Lesemann, D.-E., 1984, Immunosorbent electron microscopy in plant virus studies, in: Methods in Virology (K. Maramorosch and H. Koprowski, eds.), Vol. 8, pp. 85-101, Academic Press, New York.

Mink, G. I., Howell, W. E., and Fridlund, P. R., 1985, Apple tip leaf antigens that cause spurious reactions with tomato ringspot virus antisera in enzyme-linked immunosorbent assay, Phytopathology 75:325.

Mircetich, S. M., and Rowhani, A., 1984, The relationship of cherry leafroll virus and blackline disease of English walnut trees, Phytopathology 74:423.

Mircetich, S. M., Sanborn, R. R., and Ramos, D. E., 1980, Natural spread, graft-transmission and possible etiology of walnut blackline disease, Phytopathology 70:962.

Mowat, W. P., 1986, Methods of virus indexing for virus-tested narcissus stocks, Acta Hortic. 177:217.

Murant, A. F., 1970, Soil-borne viruses and diseases in Rubus, in: Virus Diseases of Small Fruits and Grapevines (N. W. Frazier, ed.), pp. 132-148, University of California Press, Berkeley, CA.

Murant, A. F., 1978, Raspberry ringspot virus, CMI/AAB Descriptions of Plant Viruses No. 198. Murant, A. F., 1981, Nepoviruses, in: Handbook of Plant Virus Infections and Comparative

Diagnosis (E. Kurstak, ed.), pp. 197-238, Elsevier/North Holland, Amsterdam. Murant, A. F., 1983, Seed and pollen transmission of nematode-borne viruses, Seed Sci. Technoi.

11:973. Murant, A. F., and Lister, R. M., 1967, Seed-transmission in the ecology of nematode-borne

viruses, Ann. Appi. BioI. 59:63. Murant, A. F., and Taylor, C. E., 1965, Treatment of soils with chemicals to prevent transmission

of tomato blackring and raspberry ringspot viruses by Longidorus elongatus (de Man), Ann. Appi. BioI. 55:227.

Murant, A. F., Mayo, M. A., Harrison, B. D., and Goold, R. A., 1972, Properties and RNA components of raspberry ringspot virus, r. Gen. Viroi. 16:327.

Nolasco, G., and De Sequeira, O. A., 1993a, Immunocapture polymerase chain reaction (IC/PCR) in the diagnosis of grapevine fanleaf virus (GFLV) in grapevine field samples, in: Extended Abstracts 11th Meeting International Council for the Study of Viruses and Virus Diseases of the Grapevine, Montreux, 1993, pp. 158-159, Federal Agricultural Research Station of Changins, Nyon, Switzerland.

Nolasco, G., and De Sequeira, O. A., 1993b, Genome diversity of field isolates of grapevine fanleaf virus (GFLV) analysed by single stranded conformation (SSCP) and restriction fragment length (RFLP) polymorphisms, in: Extended Abstracts 11th Meeting International Council for the Study of Viruses and Virus Diseases of the Grapevine, Montreux, 1993, pp. 31-32, Federal Agricultural Research Station of Changins, Nyon, Switzerland.

Novak, J. B., and Lanzova, J., 1975, Etiology of two virus diseases of lilac (Syringa vulgaris 1.) occurring in Czechoslovakia, BioI. Plantarum 17:226.

Nyland, G., 1976, Cherry rasp leaf, in: Virus Diseases and Noninfectious Disorders of Stone Fruits in North America, Agric. Handbook 437, pp. 219-221, U.S. Dept. of Agriculture, Washington, DC.

Ouertani, R., Savino, v., Minafra, A., Boscia, D., Castellano, M. A., Martelli, G. P., and Greco, N., 1992, Properties of a previously undescribed grapevine nepovirus from Tunisia, Arch. Viroi. 126:107.

Owusu, G. K., 1971, Cocoa necrosis virus in Ghana, Trop. Agric. Trinidad 48:133. Parish, C. 1., 1977, A relationship between flat apple disease and cherry rasp leaf disease,

Phytopathology 67:982. Piazzolla, P., and Rubino, 1., 1984, Evidence that the low molecular weight RNA associated with

chicory yellow mottle virus is a satellite, Phytopathoi. Z. 111:199.


Powell, C. A., 1987, Detection of three plant viruses by dot-immunobinding assay, Phytopathology 77:306.

Powell, C. A., and Derr, M. A., 1983, An enzyme-linked immunosorbent blocking assay for comparing closely related virus isolates, Phytopathology 73:660.

Powell, C. A., Forer, 1. B., and Stouffer, R. F., 1982, Reservoirs of tomato ringspot virus in fruit orchards, Plant Dis. 66:583.

Price, W. c., 1936, Specificity of acquired immunity from tobacco-ringspot diseases, Phytopathology 26:665.

Price, W. c., 1940, Comparative host ranges of six plant viruses, Am. r. Bot. 27:530. Quacquarelli, A., Rana, G. 1., and Martelli, G. P., 1976, Weed hosts of plant pathogenic viruses in

Apulia, Polioprivredna Znanstvena Smotra 39:561. Querfurth, G., and Paul, H. 1., 1979, Protein A-coated latex-linked antisera (PALLAS): New

reagents for a sensitive test permitting the use of antisera unsuitable for the latex test, Phytopathol. Z. 94:282.

Ramsdell, D. C., and Gillett, J. M., 1981, Peach rosette mosaic virus in highbush blueberry, Plant Dis. 65:757.

Ramsdell, D. C., and Myers, R. 1., 1978, Epidemiology of peach rosette mosaic virus in a Concord grape vineyard, Phytopathology 68:447.

Ramsdell, D. c., and Stace-Smith, R., 1979, Blueberry leaf mottle, a new disease of highbush blueberry, Acta Hortic. 95:37.

Rana, G. 1., and Kyriakopoulou, P. E., 1982, Artichoke Italian latent and artichoke mottled crinkle viruses in artichoke in Greece, Phytopathol. Mediterr. 21:101.

Rana, G. 1., Gallitelli, D., Kyriakopoulou, P. E., Russo, M., and Martelli, G. P., 1980, Host range and properties of artichoke yellow ringspot virus, Ann. Appl. Biol. 96:177.

Rana, G. 1., Castrovilli, S., Gallitelli, D., and Kyriakopoulou, P. E., 1985, Studies on two serologically distinct raspberry ringspot virus strains from artichoke, Phytopathol. Z. 112:222.

Remah, A., Jones, A. T., and Mitchell, M. J., 1986, Purification and properties of lucerne Australian symptomless virus, a new virus infecting lucerne in Australia, Ann. Appl. Biol. 109:307.

Roberts, 1. M., and Brown, D. J. F., 1980, Detection of six nepoviruses in their nematode vectors . by immunosorbent electron microscopy, Ann. AppI. BioI. 96:187. Robinson, D. J., and Clark, J., 1987, Genome sequence homology among strains of raspberry

ringspot nepovirus, Rep. Scott. Crop Res. Inst. 1986:172. Roca, F., Martelli, G. P., Lamberti, F., and Rana, G. 1., 1975, Distribution of LongidoIUs attenu

atus Hooper in Apulian artichoke fields and its relationship with artichoke Italian latent virus, Nematol. Mediterr. 3:91.

Roca, F., Rana, G. 1., and Kyriakopoulou, P. E., 1982, LongidoIUs fasciatus Roca et Lamberti vector of a serologically distinct strain of artichoke Italian latent virus in Greece, Nematol. Mediterr. 10:65.

Roca, F., Rana, G. 1., and Kyriakopoulou, P. E., 1986, Studies on Longidoridae (Nematoda, Dorylaimida) and raspberry ringspot spread in some artichoke fields in Greece, Nematol. Mediterr. 14:251.

Rowhani, A., 1992, Use of F(ab'lz antibody fragments in ELISA for detection of grapevine viruses, Am. r. Enol. Vitic. 42:38.

Rowhani, A., Chay, C., Golino, D. A., and Falk, B. W., 1993, Development of a polymerase chain reaction technique for the detection of grapevine fanleaf virus in grapevine tissue, Phytopathology 83:749.

Rudel, M., 1985, Grapevine damage induced by particular virus-vector combinations, Phytopathol. Mediterr. 24:183.

Rush, M. C., and Gooding, G. v., 1970, The occurrence of tobacco ringspot virus strains and tomato ringspot virus in hosts indigenous to North Carolina, Phytopathology 60:1756.

Russo, M., Martelli, G. P., and Savino, V., 1982, Immunosorbent electron microscopy for detecting sap-transmissible viruses of grapevine, in: Proceedings of the 7th Meeting of the Interna-


tional Council for the Study of Viruses and Virus Diseases of the Grapevine, Niagara Falls 1980 (A. J. McGinnis, ed.), p. 251-257, Agriculture Canada, Vineland, Ontario, Canada.

Salazar, L. F., and Harrison, B. D., 1977, Two previously undescribed potato viruses from South America, Nature 265:337.

Salazar, 1. F., and Harrison, B. D., 1978a, Host range and properties of potato black ringspot virus, Ann. Appl. Biol. 90:375.

Salazar, 1. F., and Harrison, B. D., 1978b, The relationship of potato black ringspot virus to tobacco ringspot and allied viruses, Ann. Appl. Biol. 90:387.

Saldarelli, P., Minafra, A., and Walter, B., 1993, A survey of grapevine fanleaf nepovirus isolates for the presence of satellite RNA, Vitis 32:99.

Samson, R. W., and Imle, E. P., 1942, A ring-spot type of virus disease of tomato, Phytopathology 32:1037.

Savino, v., and Gallitelli, D., 1981, Cherry leafroll virus in olive, Phytopathol. Mediterr. 20:202. Savino, V., Gallitelli, D., Jankulova, M., and Rana, G. 1., 1976, A comparison of four isolates of

artichoke Italian latent virus (AlLV), Phytopathol. Mediterr. 16:4l. Savino, V., Quacquarelli, A., Gallitelli, D., Piazzolla, P., and Martelli, G. P., 1977, n virus

dell'accartocciamento fogliare del ciliegio nel noce. I. Identificazione e caratterizzazione, Phytopathol. Mediterr. 16:96.

Savino, V., Barba, M., Gallitelli, G., and Martelli, G. P., 1979, Two nepoviruses isolated from olive in Italy, Phytopathol. Mediterr. 18:135.

Savino, V., Russo, M., and Martelli, G. P., 1981, Osservazioni sulla selezione sanitaria e risanamento della vite in Italia meridionale e su alcuni metodi di diagnosi virologica, in: Atti del 3° Simposio Internacionale sulla Selezione Cion ale della Vite, Venezia, 1981, pp. 302-309, Instituto Sperimentale per la Viticoltura, Conegliano, Italy.

Savino, V., Gallitelli, D., and Barba, M., 1983, Olive latent ringspot virus, a newly recognised virus infecting olive in Italy, Ann. Appl. Biol. 103:243.

Savino, V., Cherif, c., and Martelli, G. P., 1985, A natural serological variant of grapevine fanleaf virus, Phytopathol. Mediterr. 24:29.

Schimanski, H.-H., Schmelzer, K., and Albrecht, H.-J., 1975, Die Spatbliihende Traubenkirsche (Prunus serotina Ehrh.) als naturlicher Wirt des Kirschenblattroll-Virus, Arch. Phytopathol. P{1anzenschutz 11:329.

Schmelzer, K., 1963a, Untersuchungen an Viren der Zier- und Wildgeholze. 2. Mitteilung. Virosen an Forsythia, Lonicera, Ligustrum und Laburnum, Phytopathol. Z. 46:105.

Schmelzer, K., 1963b, Untersuchungen an Viren der Zier- und Wildgeholze. 3. Mitteilung. Virosen an Robinia, Caryopteris, Ptelea und anderen Gattungen, Phytopathol. Z. 46:235.

Schmelzer, K., 1966, Untersuchungen an Viren der Zier- und Wildgeholze. 5. Mitteilung. Virosen an Populus und Sambucus, Phytopathol. Z. 55:317.

Schmelzer, K., 1969, Das latente Erdbeerringflecken-Virus aus Evonymus, Robinia und Aesculus, Phytopathol. Z. 66:1.

Schmelzer, K., 1972a, Das Kirschenblattroll-Virus (cherry leaf-roll virus) aus der Birke (Betula pendula Roth), Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. 127:10.

Schmelzer, K., 1972b, Nachweis der Verwandtschaft zwischen Herkiinften des Kirschenblattroll-Virus (cherry leaf-roll virus) und dem Ulmenmosaik-Virus (elm mosaic virus), Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. 127:140.

Schneider, I. R., Hull, R., and Markham, R., 1972, Multidense satellite of tobacco ringspot virus: A regular series of components of different densities, Virology 47:320.

Scotto La Massese, C., Marenaud, C., and Dunez, J., 1973, Analyse d'un phenomene de degenerescence du pecher dans la vallee de l'Eyrieux, C. R. Hebd. Seances Acad. Agric. France 59:327.

Sinclair, J. B., and Shurtleff, M. C., 1975, Bud blight, in: Compendium of Soybean Diseases U. B. Sinclair and M. C. Shurtleff, eds.), pp. 44-45, American Phytopathological Society, St. Paul, MN.

Sinclair, J. B., and Walker, J. C., 1956, A survey of ring spot on cucumber in Wisconsin, Plant Dis. Rep. 40:19.


Smith, K. M., 1946, Tomato black-ring: A new virus disease, Parasitology 37:126. Smith, K. M., and Markham, R., 1944, Two new viruses affecting tobacco and other plants,

Phytopathology 34:324. Smith, K. M., and Short, M. E., 1959, Lettuce ringspot: A soil-borne virus disease, Plant Pathol.

8:54. Smith, S. H., Stouffer, R. F., and Soulen, D. M., 1973, Induction of stem pitting in peaches by

mechanical inoculation with tomato ringspot virus, Phytopathology 63:1404. Stace-Smith, R., 1984, Red raspberry virus diseases in North America, Plant Dis. 68:274. Stace-Smith, R., Reichmann, M. E., and Wright, N. S., 1965, Purification and properties of

tobacco ringspot virus and two RNA-deficient components, Virology 25:487. Steere, R. 1., 1956, Purification and properties of tobacco ringspot virus, Phytopathology 46:60. Stellmach, G., 1970, Tomato black ring in Vitis, in: Virus Diseases of Small Fruits and Grape

vines (N. W. Frazier, ed.), pp. 234-236, University of California Press, Berkeley, CA. Stouffer, R. F., Hickey, K. D., and Welsh, M. F., 1977, Apple union necrosis and decline, Plant Dis.

Rep. 61:20. Swingle, R. U., Tilford, P. E., and Irish, C. F., 1943, A graft transmissible mosaic of American elm,

Phytopathology 33:1196. Taylor, C. E., 1962, Transmission of raspberry ringspot virus by Longidorus elongatus (de Man)

(Nematoda: Dorylaimidae), Virology 17:493. Taylor, C. E., and Thomas, P. R., 1968, The association of Xiphinema diversicaudatum (Mico

letzky) with strawberry latent ringspot and arabis mosaic viruses in a raspberry plantation, Ann. Appl. Biol. 62:147.

Thomas, B. J., 1980, The detection by serological methods of viruses infecting the rose, Ann. Appl. Biol. 94:91.

Thresh, J. M., and Tinsley, T. w., 1959, The viruses of cacao, Technical Bulletin of the West African Cocoa Research Institute No.7.

Thresh, J. M., Pitcher, R. S., McNamara, D. G., and Ormerod, P. J., 1972, The spread and control of nettlehead and related diseases of hop, Rep. East Malling Res. Stn. 1971:155.

Tomlinson, J. A., and Walkey, D. G. A., 1967, The isolation and identification of rhubarb viruses occurring in Britain, Ann. Appl. Biol. 59:415.

Torrance, 1., 1980, Use of protein A to improve sensitisation of latex particles with antibodies to plant viruses, Ann. Appl. Biol. 96:45.

Tsuchizaki, T., 1975, Mulberry ringspot virus, CMI/AAB Descriptions of Plant Viruses No. 142. Tsuchizaki, T., Hibino, H., and Saito, Y., 1971, Mulberry ringspot virus isolated from mulberry

shOWing ringspot symptom, Ann. Phytopathol. Soc. Japan: 37:266. Tuite, J., 1960, The natural occurrence of tobacco ringspot virus, Phytopathology 50:296. Usugi, T., and Saito, Y., 1979, Satsuma dwarf virus, CMI/AAB Descriptions of Plant Viruses

No. 208. Uyemoto, J. K., Taschenberg, E. F., and Hummer, D. K., 1977, Isolation and identification of a

strain of grapevine Bulgarian latent virus in Concord grapevine in New York State, Plant Dis. Rep. 61:949.

Valdez, R. B., McNamara, D. G., Ormerod, P. J., Pitcher, R. S., and Thresh, J. M., 1974, Transmission of the hop strain of arabis mosaic virus by Xiphinema diversicaudatum, Ann. Appl. Biol. 76:113.

Van Regenmortel, M. H. V., 1982, Serology and Immunochemistry of Plant Viruses, Academic Press, New York.

Van Regenmortel, M. H. v., and Dubs, M.-C., 1993, Serological procedures, in: Diagnosis of Plant Virus Diseases (R. E. F. Matthews, ed.), pp. 159-214, CRC Press, Boca Raton, F1.

Vegetti, G., Belli, G., Cinquanta, S., and Soncini, c., 1979, Identificazione e prime caratteristiche di un ceppo di SLRV (strawberry latent ringspot virus) isolato da coltivazioni di lamp one in Lombardia, Riv. Patol. Veg. S IV 15:51.

Vovlas, c., 1973, Trasmissione per seme del virus della maculatura gialla della cicoria, Phytopathol. Mediterr. 12:104.


Vovlas, c., Martelli, G. P., and Quacquarelli, A., 1971, Le virosi delle piante ortensi in Puglia. VI. n complesso delle maculature anulari della cicoria, Phytopathol. Mediterr. 10:244.

Vuittenez, A., Munck, M.-C., and Kuszala, J., 1964, Mise en evidence du virus de la degenerescence infectieuse par test serologique et observation des particules caracteristiques dans les extraits de feuilles de vigne malades, C. R. Hebd. Seances Acad. Sci. Paris 259:2156.

Vuittenez, A., Kuszala, J., Rudel, M., and Bruckbauer, H., 1970, Detection et etude serologique du virus latent des taches annulaires du haisier (strawberry latent ringspot), du virus des anneaux noirs de la tomate (tomato black ring) et du virus des taches annulaires du framboisier (raspberry ringspot) chez des vignes du Palatinat, Ann. Phytopathol. 2:279.

Wagnon, H. K., and Breece, J. R., 1955, Evidence of retention of peach yellow bud mosaic virus in soil, Phytopathology 45:696.

Walkey, D. G. A., and Mitchell, J., 1969, Studies on a "strap leaf" disease of celery caused by strawberry latent ringspot virus, Plant Pathol. 18:167.

Walkey, D. G. A., and Webb, M. J. w., 1968, Virus in plant apical meristems, r. Cen. Viral. 3:31l. Walkey, D. G. A., and Webb, M. J. w., 1970, Tubular inclusion bodies in plants infected with

viruses of the NEPO type, r. Cen. Viral. 7:159. Walter, B., and Etienne, B., 1987, Detection of the grapevine fanleaf virus away from the period of

vegetation, r. Phytopathol. 120:355. Walter, B., Vuittenez, A., Kuszala, J., Stocky, G., Burckard, J., and Van Regenmortel, M. H. v.,

1984, Detection serologique des virus du court-noue de la vigne par Ie test ELISA, Agranomie 4:527.

Walter, B., Ladeveze, I., Etienne, L., and Fuchs, M., 1989, Some properties of a previously undescribed virus from cassava: Cassava American latent virus, Ann. Appl. Biol. 115:279.

Waterworth, H. E., and Lawson, R. H., 1973, Purification, electron microscopy, and serology of the dogwood ringspot strain of cherry leaf roll virus, Phytopathology 63:14l.

Wingard, S. A., 1928, Hosts and symptoms of ring spot, a virus disease of plants, r. Agric. Res. 37:127.

CHAPTER 6

Nepoviruses: Molecular Biology and Replication M. A. MAyo AND D. J. ROBINSON

I. INTRODUCTION

As with many groups of plant viruses, our knowledge of nepoviruses at the biochemical and molecular levels has accumulated in several phases. In the first phase, virus particles were characterized by physical means. Nepoviruses were shown to have polyhedral particles of about 28 nm diameter that sedimented at several different rates. Explanation for these features followed largely because of the development of electrophoretic methods for separating and sizing protein and nucleic acid molecules. These approaches showed nepovirus particles to have the rather unusual symmetry of T = 1 and to contain two sizes of RNA molecule. Further work showed these RNA molecules to be parts of a bipartite genome. The most recent phase has been the application of cDNA cloning and sequencing methods, which have revealed the primary structures of the genomes and gene products of several nepoviruses. These approaches, coupled with mutagenesis of cloned DNA, have immense potential for tackling the major challenge so far unachieved, which is to obtain an understanding at the molecular level of how these viruses function in the biological processes of infection, multiplication, and transmission. In this chapter we aim to review inf<;>rmation that has come from these different phases of nepovirus molecular biology with particular

M. A. MAYO AND D. J. ROBINSON • Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom.

139

140 M. A. MAYO AND D. J. ROBINSON

emphasis on the most recent developments. The earlier phases of the subject are described in reviews by Murant (1981L Francki et a1. (1985), and Martelli and Taylor (1989). Brief overviews of the genus Nepovirus are given by Mayo (1994) and Goldbach et a1. (1995).

One result of the recent molecular biological work has been an increased confidence in the taxonomic placement of nepoviruses. Nepoviruses are currently classified as the genus Nepovirus in the family Comoviridae, which also includes the genera Comovirus and Fabavirus (Mayo and Martelli, 1993; Goldbach et a1., 1995). At a higher level, nepoviruses, along with viruses from a number of families, fall into a "Picornaviruslike supergroup" (Koonin and Dolja, 1993). Table I lists the virus species assigned, either definitively or tentatively, to the genus Nepovirus, together with the acronyms used throughout this chapter.

II. PROPERTIES OF VIRUS PARTICLES

A. Purification

The wide host ranges characteristic of nepoviruses offer considerable scope for the choice of propagation host. The plant species most commonly used for growing virus for purification include Chenopodium quinoa, Cucum is sativus, Nicotiana benthamiana, N. clevelandii, Petunia hybrida, Phaseolus vulgaris, and Vigna unguiculata.

Most nepoviruses have stable particles, whose purification poses few problems. Clarification with 8.5% (v/v) butan-l-01 (Gooding, 1963) or with a mixture of equal volumes of butan-1-o1 and chloroform (Steere, 1956) has been widely used and is effective for many of the viruses in the genus. However, there are exceptions. For example, Stace-Smith (1966) found that tomato ringspot virus (ToRSV) particles were largely destroyed by treatment with butan-1-o1 or chloroform, and devised a method in which host plant constituents were removed by freezing and thawing the extracts, and the virus particles were recovered by precipitation with 15% (w/v) ammonium sulfate. This method has also been used for cherry leaf roll virus (CLRV) and peach rosette mosaic virus (PRMV). Another gentle method of clarification that avoids the use of organic solvents employs a suspension of bentonite to coacervate impurities, and has been used for grapevine Bulgarian latent virus (GBLV) (Martelli et a1., 1977), arabis mosaic virus (ArMV), strawberry latent ringspot virus (SLRSV) (Savino et a1., 1979), and lucerne Australian latent virus (LALV) (Jones et a1., 1979), although with variations in the concentration and method of purification of the bentonite. The method of clarification chosen may affect the composition of the preparation obtained. For example, empty protein shells occur in preparations of cocoa necrosis virus (CNV) clarified by ammonium sulfate precipitation, but not in those treated with butan-l-ol, which apparently destroys empty particles (see Section II.G).

NEPOVIRUSES: MOLECULAR BIOLOGY AND REPLICATION 141

Whichever method of clarification is adopted, further purification and concentration are achieved by differential centrifugation or by precipitation with polyethylene glycol (PEG), or by a combination of the two. However, virus particles may be lost by precipitation with excessive concentrations of PEG (Jones et a1., 1979), and an injudicious choice of resuspending buffer can cause particles to aggregate (Jones et a1., 1985). Yields of particles of most nepoviruses are in the range 10 to 50 mg/kg leaf tissue.

B. Particle Size and Structure

Particles of most nepoviruses have angular hexagonal outlines and diameters of 26 to 28 nm when negatively stained in ammonium molybdate or sodium phosphotungstate (e.g., Fig. 1a). However, when stained with uranyl salts or methylamine tungstate, the particles of tobacco ringspot virus (TRSV) appear considerably larger, are less angular, and seem to be collapsed or damaged (Roberts, 1988).

Polyhedral crystals of TRSV (Heuss et a1., 1981) and bipyramidal crystals of ArMV (Takemoto et a1., 1985) have been examined at resolutions down to 3.3 A and 10 A, respectively. Both have similar unit cell dimensions. Their particle-to-particle distances are 280 A and 274 A, respectively, which corresponds with the diameters observed by electron microscopy. No difference in crystallization or crystal morphology between particles of different sedimenting components was found for either virus.

The size of particles in suspension can be calculated from the diffusion coefficient (D), but accurate measurement of D is difficult. Values of 1.6 x 10-7 cm2 sec1 are quoted for TRSV and ToRSV (Tremaine and Stace-Smith, 1968). Estimates for several nepoviruses were given by Murant et a1. (1981), who derived D from the molecular weight of RNA and protein components of T, M, and B particles, from an assumed particle structure, and from the sedimentation coefficients of the components. The estimates were between 1.42 x 10-7 cm2 sec1 and 1.52 x 10-7 cm2 sec1 for tomato black ring virus (TBRV), ToRSV, TRSV, CLRY, myrobalan latent ringspot virus (MLRSV), ArMY, and raspberry ringspot virus (RRSV) but only 1.27 x 10-7 cm2 sec-1 for SLRSV. Particle diameters of about 28-30 nm for the first group and 34 nm for the tentative nepovirus SLRSV were calculated from these estimates.

Evidence that SLRSV particles are larger than RRSV particles was also obtained from electrophoresis of particles in polyacrylamide gel. In these experiments, the proportionate decrease in electrophoretic mobility caused by increasing the strength of polyacrylamide gel was greater for SLRSV particles than for RRSV particles (Mayo et a1., 1974). However, the diameter of negatively stained SLRSV particles is given as 28-30 nm by Bellardi and Gelli (1984).

Little or no surface detail is apparent on negatively stained particles of most nepoviruses, although some was apparently observed on those of one

TAB

LE I

. M

olec

ular

and

Par

ticl

e P

rope

rtie

s of

Nep

ovir

uses

Coe

ffic

ient

of

sedi

men

tati

on

of p

arti

cles

(S)

Mol

ecul

ar w

eigh

t

Vir

us n

ame

and

acro

nym

a T

M

B

R

NA

-1b

RN

A-2

b

Def

init

ive

mem

bers

(si

ngle

maj

or c

oat

prot

ein

of c

a 55

kD

a)

a. R

NA

-2 m

ol.

wt.

1.3

-1.5

x 1

(J6, p

rese

nt i

n M

and

B c

ompo

nent

s, M

com

pone

nt 8

6-9

3 S

A

rabi

s m

osai

c (A

rMV

)1

53

93

120

2.4

(2.8

) 1.

4 (1

.3)

Arr

acac

ha A

(A

VA

) 50

92

12

5 2.

5 1.

4 A

rtic

hoke

Aeg

ean

ring

spot

(A

AR

SV)2

2.

4 1.

4 C

assa

va A

mer

ican

lat

ent

(CsA

LV)

(2.5

) (1

.4)

Dog

woo

d m

osai

c (D

oMV

)l

(2.9

) (1

.4)

Gra

pevi

ne f

anle

af (

GFL

V)1

50

86

12

0 73

42

3774

P

otat

o bl

ack

ring

spot

(PB

RSV

)3

49

84

117

2.5

1.5

Ras

pber

ry r

ings

pot

(RR

SV)2

50

92

13

0 2.

4 (2

.8)

3928

T

obac

co r

ings

pot

(TR

SV)3

53

91

12

6 2.

4 (2

.8)

1.4

(1.3

) b.

R

NA

-2 m

ol.

wt.

1.4

-1.6

x 1

(J6, p

rese

nt i

n M

com

pone

nt o

nly,

M c

ompo

nent

85-

101

S A

rtic

hoke

Ita

lian

lat

ent

(AlL

V)4

55

96

12

1 2.

4 1.

5 (1

.7)

Coc

oa n

ecro

sis

(CN

V)4

54

10

1 12

9 C

rim

son

clo

ver

late

nt

(CC

LV)

2.2

1.6

Cyc

as n

ecro

tic

stu

nt

(CN

SV)

85

112

(2.5

) (1

.5)

Gra

pevi

ne c

hrom

e m

osai

c (G

CM

V)4

92

11

7 72

12

4441

M

ulbe

rry

ring

spot

(M

RSV

) 50

96

12

6 2.

6 1.

4 (1

.5)

Coa

t pr

otei

n (x

lO-3

)

54

53

54

57

54

56

59

54

57

54

60

52

65

57

Oli

ve l

aten

t ri

ngsp

ot (

OL

RSV

) 52

97

13

2 2.

7 1.

4 58

Pol

ypro

tein

enc

oded

by

RN

A-I

e (x

lO-3

)

(250

)

253

(200

) (2

25)

250

RN

A-2

e (x

lO-3

)

(115

+ 1

05)

131

124

(116

)

146

Tom

ato

blac

k ri

ng (

TBR

V)4

55

97

12

1 73

56

4662

57

25

4 15

0 c.

R

NA

-2 m

ol.

wt.

1.9

-2.2

x 1

(J6,

pres

ent

in M

com

pone

nt o

nly,

M c

ompo

nent

109

-128

S,

som

etim

es b

arel

y se

para

ted

from

B

com

pone

nt

Art

icho

ke y

ello

w r

ings

pot

(AY

RSV

) B

lueb

erry

lea

f m

ott

le (

BLM

V)5

2.

3 53

12

0 12

8 2.

4 1.

9 (1

.9)

2.2

53

54

-oj>. r,.J ~ ;> ~ o ~ ~

':-<

e5 2 Vl o Z

Z

Cas

sava

gre

en m

ottl

e (C

GM

V)

(2.9

) (2

.3)

trI

53

"d

Che

rry

leaf

rol

l (C

LRV

) 11

5 12

8 2.

4 (2

.8)

2.1

(2.3

) 54

(2

50)

(165

) ~

Chi

cory

yel

low

mot

tle

(ChY

MV

) 51

11

6 12

8 2.

4 2.

0 54

~

Gra

pevi

ne B

ulga

rian

lat

ent

(GB

LV)5

52

12

0 12

7 2.

2 2.

1 54

C

Il

Gra

pevi

ne T

unis

ian

ring

spot

vir

us

2.4

2.0

59

trI ~

(GT

RSV

) ~

Hib

iscu

s la

ten

t ri

ngsp

ot (

HL

RSV

) 51

11

4 13

2 2.

5 2.

2 54

0 t"

' L

ucer

ne A

ustr

alia

n la

tent

(LA

LV)

56

128

133

2.4

(2.8

) 2.

1 (2

.5)

55

trI

Myr

obal

an l

aten

t ri

ngsp

ot (

MLR

SV)

50

109

126

2.6

(2.8

) 1.

9 (2

.0)

53

() c::

Pea

ch r

oset

te m

osai

c (P

RM

V)

52

115

134

2.5

2.2

57

t"' :>

Pot

ato

U (

PVU

) 55

11

7 13

5 58

:;.:

I

Tom

ato

ring

spot

(To

RSV

) 53

11

9 12

7 82

14

7273

58

24

4 20

7 5

Ten

tati

ve m

embe

rs (

coat

pro

tein

s m

ulti

ple

or

unkn

own)

t"

'

d.

Two

coat

pro

tein

s of

ca

45 a

nd 2

5 kD

a 0 C"

l L

ucer

ne A

ustr

alia

n sy

mpt

omle

ss (

LASV

) 13

0 (2

.5)

(1.4

) 40

, 26

><

Rub

us C

hine

se s

eed-

born

e (R

CSV

)6

1.4

47

,25

:>

S

atsu

ma

dwar

f (S

DV

) 11

9 12

9 1.

9 1.

7 4

2,2

1

Z

0 S

traw

berr

y la

ten

t ri

ngsp

ot (

SLR

SV)6

55

97

13

0 2.

6 (2

.9)

3824

4

3,2

7

(250

) 99

:;.:

I

e. Tw

o o

r th

ree

coat

pro

tein

s of

ca 2

1-28

kD

a tr

I "d

Arr

acac

ha B

(AV

B)

126

2.5

1.3

26

,22

t:

()

Art

icho

ke v

ein-

band

ing

(AV

BV

) 56

92

12

4 2.

4 1.

4 27

.5,

24.5

, 2

2

~ C

herr

y ra

sp l

eaf

(CR

LV)

56

96

12

8 2.

0 1.

5 2

6,2

3,2

1

(200

) (1

02)

(5

f. C

oat p

rote

in c

ompo

sitio

n un

know

n Z

T

omat

o to

p ne

cros

is (

ToT

NV

) 52

10

2 12

6

aAcr

onym

s fo

llow

Hul

l et

al.

1199

1) a

s re

com

men

ded

by I

eTV

. V

iruse

s fo

llow

ed b

y th

e sa

me

supe

rscr

ipt

num

ber

are

sero

logi

cally

rel

ated

. bV

a1ue

s be

twee

n 1

and

3 ar

e M

, x 1

0-6 ,

tho

se in

bra

cket

s be

ing

for

dena

ture

d R

NA

; va

lues

of

>200

0 ar

e nu

mbe

rs o

f nu

cleo

tides

exc

ludi

ng t

he p

olY

IA)

tail,

de

term

ined

by

nucl

eotid

e se

quen

cing

. cV

alue

s in

bra

cket

s w

ere

dete

rmin

ed b

y in

vit

ro tr

ansl

atio

n; o

ther

val

ues

wer

e de

duce

d fr

om n

ucle

otid

e se

quen

ces.

~


FIGURE 1. (a) Electron micrograph of particles of TRSV negatively stained with 2 % ammonium molybdate, pH 7. Bar, 100 nm. (b) Electron micrograph of particles of SLRSV negatively stained with 2% sodium phosphotungstate, pH 6. Bar, 50 nm. (c) Electron micrograph of freeze-dried particles of TRSV shadowed at 45° with uranium. Bar, 50 nm.


strain of TRSV and of ArMV (Chambers et a1., 1965; Agrawal, 1967). Stained particles are of three types: wholly penetrated by stain, partially penetrated, or not penetrated. These three types of staining have often been linked with the different nucleic acid content of the three sedimenting components T, M, and B (see Section II.C) (Debrot, 1964; Murant et a1., 1968). However, stain penetration is not necessarily related to nucleic acid content because, for example, the proportion of TRSV particles penetrated increased with the pH of the stain (Davison and Francki, 1969) and preparations of B component of TBRV have been observed to contain all three types of particle (M. A. Mayo, H. Barker, and I. M. Roberts, unpublished results). Some particles of the tentative nepoviruses, SLRSV and Rubus Chinese seed-borne virus (RCSV), appear to be penetrated by negative stain at their vertices (Murant, 1981; Barbara et a1., 1985) (Fig.1b). This may be related to their atypical coat protein compositions (see Section III.B).

Shadowed, freeze-dried particles of TRSV appear slightly smaller than negatively stained ones, with mean diameters of 24.5-25 nm (Roberts, 1988). However, it was possible to observe structural detail on such particles, with many particles exhibiting fivefold or threefold symmetry (Fig. 1c). The images were consistent with a structure consisting of 60 subunits clustered in 12 pen tamers in aT = 1 lattice. Similar observations were made on cherry rasp leaf virus (CRLV), SLRSV, and TBRV particles, although the size and shape of the 12 morphological units differed between viruses (Roberts, 1984). A T = 1 structure can accommodate protein subunits with molecular weights of 52,000 to 65,000 (see Section III.B), unlike the T = 4 structure that was proposed in earlier work (Chambers et a1., 1965; Agrawal, 1967; Chu and Francki, 1979).

C. Sedimentation Properties

Sedimentation resolves particles of most nepoviruses into three components. These are named, from their positions in sucrose density gradients, top (T; 49-58 S), middle (M; 84-128 S), and bottom (B; 112-134 S). Sedimentation coefficients for the components of individual viruses are listed in Table I. With some nepovirus isolates, additional minor sedimenting components are observed. by analytical ultracentrifugation (e.g., TRSV: Schneider, 1971; GBLV: Gallitelli et a1., 1983), and are correlated with the presence of satellite RNA (see Section IX).

The proportions of the three components in purified preparations differ among nepoviruses, some lacking T or M (e.g., ToRSV: Piazzolla et a1., 1985; Puffinberger and Corbett, 1985) or both [e.g., lucerne Australian symptomless virus (LASV): Remah et a1., 1986]. The proportions in preparations of any one isolate tend to be reproducible. Different isolates of the same virus may be similar (ToRSV: Stace-Smith, 1984), or very different (TRSV: Schneider and Diener, 1966; CLRV: Jones, 1985) in the proportion of M to B. The relative


amounts of T, M, and B in preparations of some nepoviruses may depend on the method of purification (see Section II.A) or the host and/or season of the year in which the virus is propagated (Schneider et a1., 1974; Ladipo and De Zoeten, 1972). Moreover, the relative amount of B-component particles in purified RRSV increased with increasing interval between inoculation and purification (Barker, 1980) (see also Section VII.E).

D. Isopycnic Centrifugation

Centrifugation to equilibrium in cesium chloride solutions also resolves particles of most nepoviruses into three main components. T-component particles have densities of about 1.3 g/cm3 and contain little or no RNA; the densities of M- and B-component particles are related to their sedimentation coefficients and reflect their RNA content (Murant et a1., 1972). The presence of satellite RNA in particles of some nepoviruses results in several additional density components (see Section IX). B-component particles of mulberry ringspot virus (MRSV) (Hibi, 1986), olive latent ringspot virus (OLRSV) (Savino et a1., 1983), RRSV (Mayo et a1., 1973), and TRSV (Stace-Smith, 1985) can be resolved into two density components under some conditions (see Section IV. C).

E. Electrophoretic Properties

The isoelectric points (pI) of several nepoviruses are rather low: pH 4.0 for grapevine fan1eafvirus (GFLV), 4.5 for MLRSV, 4.7 forTRSV (Hewitt et a1., 1970; Dunez et a1., 1976; Stace-Smith, 1985). However, that of blueberry leaf mottle virus (BLMV) is pH 7.5 (Ramsdell and Stace-Smith, 1983). Particles of RRSV and TBRV at pH 5.8 (M. A. Mayo, unpublished observations) and of SLRSV at pH 6.8 (Mayo et a1., 1974) are negatively charged, suggesting that they too have relatively acidic pI. With RRSV (Jones et a1., 1989), CLRV (Walkey et a1., 1973) and TBRV (M. A. Mayo, unpublished observations), strains may differ in electrical charge. Attempts to correlate these charge differences between RRSV and TBRV strains with their specificity for different species of nematode vector were unsuccessful (B. D. Harrison and M. A. Mayo, unpublished observations).

M and B components of ToRSV migrated electrophoretically at the same rate in sucrose gradients at pH 8.0 (Schneider et a1., 1974). However, preparations of ArMV and SLRSV subjected to electrophoresis at pH 7.0 in cellulose acetate were resolved into B component and T component, the B component being the more negatively charged (Clark, 1976). Treatment of B-component particles with alkaline buffers caused them to sediment and migrate electrophoretically like T-component particles. Thus, encapsidated RNA seems to be able to modify the surface charge of particles of some nepoviruses. How-


ever, these results need to be treated with caution. Although fresh preparations of particles of dogwood mosaic virus (DoMV) gave a single band in density gradient electrophoresis, two bands were observed after storage or treatment of the preparation with ethylene diamine tetra-acetate (EDTA) (Barnett et a1., 1989).

F. Particle Composition

The ultraviolet absorption spectra of preparations of T component are typical of proteins with little or no nucleic acid, having A260:A280 ratios less than one. In contrast, the spectra of Band M components are typical of nucleoprotein, with A260:A280 ratios of 1.4-1.8, depending on their RNA content (Stace-Smith et a1., 1965; Murant et a1., 1972). There are no reports of any extra components such as polyamines or lipids in nepovirus particles. Particle proteins of TRSV (Chu and Francki, 1979) and RRSV (D. J. Robinson, unpublished results) did not react with periodic acid-Schiff's reagent in tests that gave a positive result with amounts of carbohydrate equivalent to about two hexose residues per protein subunit.

C. Forces Stabilizing Particles

Because T-component particles are found in preparations of most nepoviruses and apparently do not result from damage to nucleoprotein component particles, it is unlikely that nucleic acid plays an essential role in stabilizing nepovirus particles. Indeed, the particle protein of ArMV expressed in transgenic plants can assemble to form viruslike particles that are indistinguishable from those of natural T component. Assembly into viruslike particles was also observed when the protein was expressed in insect cells, but not when expressed in bacterial cells, which suggests that although no other virus-specified component is required for assembly, eukaryotic but not prokaryotic cells contain something that is essential (Bertioli et a1., 1991).

The way in which TRSV particles dissociated in, or resisted dissociation by, various agents, in particular sodium dodecyl sulfate (SDS), led Kaper (1973) and Boatman and Kaper (1976) to suggest that nepovirus particles are largely stabilized by protein-protein interactions. Circular dichroism studies with chicory yellow mottle virus (Ch YMV) led Piazzolla et a1. (1977 a) to similar conclusions, although they suggested that minor interactions occur between tyrosyl and/or tryptophanyl residues and the RNA bases. In some circumstances, and with some nepoviruses, T- and M-component particles are less stable than B-component particles and these interactions may therefore playa stabilizing role. Moreover, T-, M-, and B-component particles of TRSV were degraded at different rates by a one-phase phenol treatment, which suggests that their stabilities differ; M- and B-component particles


treated in this way gave rise to novel ribonuclease-sensitive components (Quacquarelli et a1., 1971).

When nucleoprotein particles of ChYMV (Quacquarelli et a1., 1972), GFLV, or RRSVbut not those of ArMV (Quacquarelli et a1., 1976) are frozen in certain solvents, the particles release RNA and are converted to artificial T component, suggesting that protein-protein bonds are more stable than protein-RNA bonds. Similarly, when particles of ChYMV, ArMV, RRSV, GFLV, TBRV, or grapevine chrome mosaic virus (GCMV) were heated, RNA was released at lower temperatures than those at which protein was denatured (Piazzolla et a1., 1977b). This was interpreted as evidence that the hydrogen-bonded RNA was released more easily than the hydrophobically bonded protein subunits were separated.

III. PARTICLE PROTEINS

A. Preparation

There have been no studies with solutions of native particle protein molecules in either a monomeric or an oligomeric form. TRSV protein has been prepared by treatment with 1 M HCI at room temperature (Stace-Smith et a1., 1965) or by heating at 55-60 DC in 0.1 M phosphate, pH 8 (Chu and Francki, 1979). Protein of TRSV (Chu and Francki, 1979) or TBRV (M. J. Farmer, unpublished results) made by heating virus particles at pH 8 was not soluble in aqueous buffers unless it was S-carboxymethylated or oxidized with performic acid, but it was soluble in 1 % SDS or in 8 M urea. The amino acid sequences of nepovirus particle proteins deduced from the viral nucleotide sequences are relatively hydrophobic.

B. Number and Sizes

1. Definitive Nepoviruses

Polypeptides extracted from particles of definitive nepoviruses (clusters a, b, and c in Table I) form a single major band in SDS-polyacrylamide gel electrophoresis (PAGE). The apparent molecular weight of this polypeptide is usually between 54,000 (54 kDa) and 57 kDa, although a few estimates are outside this range. The estimated size of the polypeptide may vary between strains of the same virus (e.g., RRSV: Jones et a1., 1989). Treatment with hot 8 M urea or carboxymethylation, or electrophoresis in different strengths of polyacrylamide gel, did not affect the estimated size of TRSV particle protein (Mayo et a1., 1971). However, the particle protein of RRSV, though not those of TBRV or TRSV, migrated as two components in gels containing a discontinuous buffer system unless the protein had been treated with iodoace-


tamide. The faster-moving component was apparently an artifact caused by intramolecular S-S bond formation (Acosta and Mayo, 1990b).

Sometimes, small amounts of smaller polypeptides are observed (e.g., Mayo et a1., 1971); this led Chu and Francki (1979) to propose that the particle protein of TRSV, and by implication that of other definitive nepoviruses, was a tetramer or fourfold repeat of units with a molecular weight of about 13 kDa. However, this hypothesis is not compatible with subsequent data on the nucleotide sequence of the region of RNA-2 that encodes coat protein (see Section V.B), and has been discounted.

2. Tentative Nepoviruses

In SDS-PAGE, coat proteins of LASV, SLRSV, satsuma dwarf virus (SDV), and RCSV (cluster d in Table I) separate into two species of 40 to 47 kDa and 21 to 27 kDa, which, at least for SLRSV (Mayo et a1., 1974) and SDV (Usugi and Saito, 1979), are present in approximately equimolar amounts.

Particles of arracacha virus B (AVB), artichoke vein-banding virus (AVBV), and CRLV (cluster e in Table I) contain two or three protein species of 21 to 27.5 kDa. The three proteins of CRLV are in roughly equimolar amounts (Jones et a1., 1985) and they are affected in different ways when treated with cystamine or hydroxyethyl disulfide, which suggests that they are not modified versions of a single protein species (1. Lane, unpublished results). The 26- and 22-kDa proteins of AVB are in approximate proportions of 3:1 (Kenten and Jones, 1979).

Tomato top necrosis virus (ToTNV) (Bancroft, 1968) is listed as a tentative nepovirus in Table I (cluster f). However, the original culture is lost and it is impossible to determine its precise affinities.

IV. GENOMIC RNA

A. Sizes

Preparations of nepovirus particles typically contain two genomic RNA species. Additional species found in some nepoviruses are regarded as satellite RNA (see Section IX). When examined by electrophoresis or sedimentation in nondenaturing solvents, the apparent molecular weights of the genomic RNA species are about 2.5 x lO6 (RNA-I) and between 1.4 x lO6 and 2.3 x lO6 (RNA-2) (Table I). The values for denatured RNA are usually slightly greater (e.g., RNA-1 = 2.8 x lO6) (Table I). There may be differences in the sizes of RNA-2 between strains of the same virus (e.g., TBRV: Stobbs and Van Schagen, 1985). Where RNA-l and RNA-2 are rather similar in size, as with ToRSV, they may not always be resolved by gel electrophoresis (Piazzolla et a1., 1985). For several nepoviruses, the RNA molecular weight can also be calculated from sequence data (see Table I).


B. Biochemical Features

1. Polyadenylation

RNA of all nepoviruses tested binds to oligo(dT)-cellulose in solutions of high ionic strength (Mayo et a1., 1979a) and can act as a template for the synthesis of cDNA by reverse transcriptase when primed with oligo(dT). These properties indicate the presence of poly(A). Sequence data show that the poly(A) sequence is not internal, as it is in brome mosaic virus RNA-3 (Ahlquist et a1., 1981), but at the 3' end. Oligonucleotide maps of TBRV RNA indicate that the poly(A) sequence is polydisperse in length (C. Fritsch, unpublished results).

2. Genome-Linked Protein (VPg)

The infectivity for plants of potato black ringspot virus (PBRSV) RNA (Salazar and Harrison, 1978) and the infectivities of TRSV or TBRV RNA for plants or isolated protoplasts (Harrison and Barker, 1978) were greatly decreased or abolished by treatment with protease. Also, after radioiodination by the chloramine T method, RNA of several nepoviruses is much more radioactive than RNA from tobacco mosaic tobamovirus, tobacco rattle tobravirus, or tomato bushy stunt tombusvirus, even when rigorously purified by methods designed to disrupt non covalent interactions. Moreover, most of this radioactivity can be rendered soluble in 70% ethanol by treatment of the labeled RNA with proteases (Mayo et a1., 1979b, 1982). These findings suggest that nepovirus RNA molecules are covalently linked to protein.

The bound protein (VPg) does not interfere with translation in vitro (Chu et a1., 1981; Koenig and Fritsch, 1982). When translation of TBRV RNA was allowed to initiate in the presence of an inhibitor of elongation, the RNA fragments protected by ribosomes from nuclease digestion remained bound to the VPg (Koenig and Fritsch, 1982). By analogy with other viruses (Daubert and Bruening, 1985), VPg is probably at the 5' end of the RNA molecules, and this is consistent with data from sequencing experiments.

The apparent molecular weight, estimated by gel electrophoresis, of the VPg-oligonucleotides released by ribonuclease treatment from RNA of TRSV and RRSV (Mayo et a1., 1979b, 1982), TBRV (Koenig and Fritsch, 1982; Mayo et a1., 1982), or CLRV (Hellen and Cooper, 1987) was 4000-6000. However, the VPgs of GFLV (Pinck et a1., 1991) and TBRV (Hemmer et a1., 1995) have been sequenced and shown to consist of 24 and 27 amino acid residues, respectively, with molecular weights of 2900-3100. The VPgs of different nepoviruses can be distinguished by peptide mapping; those on the genomic and satellite RNA species of TBRV gave similar maps (Mayo et a1., 1982). Furthermore, an antiserum against the VPg of GFLV does not react with the VPg of the closely related ArMV (Margis et a1., 1993).


C. RNA Packaging

All definitive nepoviruses have M-component particles that contain one molecule of RNA-2 and B-component particles that contain one molecule of RNA-I. In addition, nepoviruses in cluster (a) of Table I, which have RNA-2 molecules with molecular weights about half that of RNA-I, have a second type of B-component particle that contains two molecules of RNA-2 (TRSV: Diener and Schneider, 1966; RRSV: Murant et a1., 1972; cassava American latent virus (CsALV): Walter et a1., 1989; MRSV: Hibi, 1986; GFLV: Quacquarelli et a1., 1976). Indeed, Rezaian and Francki (1973) estimated that 85 to 90% of the RNA-2 synthesized in tissue infected with TRSV was packaged in B-component particles. Gel electrophoretic analysis of RNA extracted from UV-irradiated B component of RRSV showed a decrease in the amount of RNA-2 and an increase in RNA with a mobility close to that of RNA-l (Mayo et a1., 1973). Irradiation did not affect the mobility of RNA-2 from treated M-component particles. These observations were interpreted as having resulted from intermolecular cross-linking of RNA-2 molecules inside B-component particles that contained two molecules of RNA-2.

Under appropriate conditions, B-component particles of nepoviruses in cluster (a) of Table I can be partially resolved by isopycnic centrifugation into two subcomponents. B-component particles of RRSV formed one band when centrifuged at 25°C to equilibrium in CsCl solutions that contained phosphate buffer (Murant et a1., 1972) but formed two bands when centrifuged at 2 °C in CsCl solutions containing phosphate (Mayo et a1., 1973) or EDTA (M. A. Mayo, unpublished results). Although it was not possible to separate these components completely, the more dense particles were found to contain proportionately more RNA-l (M. A. Mayo, unpublished results), which suggests that the separation was related to the RNA composition of the particles. However, the separation was not observed at 20°C (Mayo et a1., 1973) or in Tris-buffered solutions (M. A. Mayo, unpublished results).

The presence of satellite RNA in particles of some nepoviruses greatly increases the number of kinds of nucleoprotein particle that can be constructed, and most of these are found in purified preparations (see Section IX). Thus, nepovirus protein shells appear to be able to encapsidate a variety of combinations of RNA up to a maximum content that is close to the molecular weight of RNA-I. At this maximum, the particles contain about 44% RNA.

D. Sequences

Complete nucleotide sequences of both genome parts of four nepoviruses, GCMV, GFLV, TBRV, and ToRSV, have been determined, as have those of RNA-2 of RRSV and SLRSV (Table IT). Each comprises a single long open reading frame (ORF) flanked by 5'- and 3'-terminal noncoding regions

152 M. A. MAYO AND D. r. ROBINSON

TABLE II. Nucleotide Sequence Data on Nepovirus Genomic RNA Species and the Sizes of their Terminal Noncoding Regions

Acc. No.a 5'-NCRb 3'-NCRC Reference

RNA-l BLMV 1390 Bacher et a1. (1994b) CLRV S84124 -1500 Scott et a1. (1992) GCMV X15346d 215,218 241 Le Gall et a1. (1989) GFLV D00915d 242 241 Ritzenthaler et a1. (1991) TBRV-S D00322d 260 301 Greif et a1. (1988) ToRSV Ll9655d 77, 440 1543 Rott et a1. (1995)

RNA-2 ArMV DlO086 -210 Bertioli et a1. (1991)

X55460 Steinkellner et a1. (1992) BLMV 1390 Bacher et a1. (1994b) CLRV S84125 Scott et a1. (1992)

S84126 S63537 -1500 Scott et a1. (1993)

GCMV X15163 d 217 252 Brault et a1. (1989) GFLV X16907d 7,232 212 Serghini et a1. (1990) RRSV S46011 d 206 397 Blok et a1. (1992) SLRSV X75165 552 Everett et a1. (1994)

X77466d 669,732 482 Kreiah et a1. (1994) TBRV-S X04062d 287 301 Meyer et a1. (1986) TBRV-E X80831 d 299 287 Le Gall et a1. (1995a) ToRSV D12477d 77, 440 1550 Rott et a1. (1991b) TRSV L09205 Buckley et a1. (1993)

aAccession numberls) in the EMBL and GenBank sequence databases. bNumber of residues preceding the initiation codon of the large ORF. Two numbers indicate that there are 2 in-frame AUG sequences; that more likely to be functional is shown in bold.

'Number of residues between the first termination codon in the large ORF and the start of the polYIA) tail

dComplete sequence known.

(NCR). The sizes of these RNA species and of their potential translation products are given in Table I. Partial sequences of several other nepovirus RNA species have also been obtained (Table II). The polypeptide sequences encoded by these sequenced molecules are discussed in Section VI.

E. Noncoding Regions

The 5' - and 3' -terminal NCR of most nepovirus RNA species are between 200 and 600 residues long, and the corresponding regions of RNA-l and RNA-2 of the same virus are mostly similar in length (Table II). The length of the 5' -NCR is uncertain in several instances because of the existence of more than one in-frame AUG codon. The most striking example is the 5' -NCR of GFLV RNA-2 in which the first in-frame AUG is at position 8, but there is a second at position 233, which would give a 5' -NCR similar in


length to that of RNA-I. Similarly, both RNA species of ToRSV would have 5'-NCRs of 440 nucleotides (nt) if their second in-frame AUG co dons were used. In each of these instances, the second AUG is in a better Kozak context for initiation of translation (Serghini et a1., 1990; Rott et a1., 1991b). The 5'-NCR of GCMV RNA-1 has two consecutive AUG co dons, the second of which is in the better context (Le Gall et a1., 1989). In vitro translation of SLRSV RNA-2 yields two products of 99 and 96 kDa (Hellen et a1., 1991), which is consistent with initiation from both in-frame AUG codons (Kreiah et a1., 1994). Apart from these exceptions, there is no reason to believe that the first in-frame AUG in other nepovirus RNA sequences is not the initiation site for translation.

The terminal regions of RNA-1 and RNA-2 of the same virus are very similar in nucleotide sequence. In the 5'-NCR, sequence identity ranges from 68% in GCMV (Brault et a1., 1989) to 100% in ToRSV (Rott et a1.,1991a). Moreover, in ToRSV the high level of sequence homology continues through the first few hundred nucleotides of the coding regions. At the 3' end, the NCRs in RNA-1 and RNA-2 of GCMV (Brault et a1., 1989), ToRSV (Rott et a1., 1991a), TBRV (Greif et a1., 1988), BLMV (Bacher et a1., 1994b), and CLRV (Scott et a1., 1992) are identical or virtually so. In contrast, the 3'-NCRs of RNA-1 and RNA-2 of GFLV are only 80% identical (Ritzenthaler et a1., 1991). Surprisingly, comparison of isolates of CLRV from birch and rhubarb showed that, although the whole 1500 nt 3'-NCR is shared by RNA-1 and RNA-2 of the birch isolate, only the terminal 700 nt are identical with those of the rhubarb isolate (Scott et a1., 1992).

Le Gall et a1. (1995a) found several structural features in the 5'-NCRs of TBRV and GCMV. A potential stem-loop structure at the 5' terminus is followed by a pyrimidine-rich region and two or three direct repeats of another potential stem-loop structure.

F. Sequence Homologies between Viruses

The 3'-NCRs in RNA-2 of GFLV, TBRV, and GCMV are about 60% identical in sequence, with several patches of complete identity (Serghini et a1., 1990), but the sequences of the 3'-NCRs in RNA-2 of ToRSV (Rott et a1., 1991a), RRSV (Blok et a1., 1992), BLMV (Bacher et a1., 1994b), and SLRSV (Everett et a1., 1994) have little in common with these three. A similar pattern is observed in comparisons among 3'-NCRs of RNA-I. There are no reports of extended regions of sequence homology between the genomes of different nepoviruses at the 5' end, and although Fuchs et a1. (1989) discerned a consensus among the extreme 5' -terminal sequences of several nepovirus genomic and satellite RNA species, sequences determined subsequently have not conformed to it.

Homologies between the proteins encoded by the genomes of different nepoviruses are dealt with in Section VI, but some additional RNA homolo-


gies have been detected in nucleic acid hybridization experiments. RNA from GCMV, CNV, and artichoke Italian latent virus (AlLV) (which are each distantly related serologically to TBRV), but not that from several other nepoviruses, hybridized with cDNA to RNA-lor RNA-2 of two strains of TBRV (Dodd and Robinson, 1984). Similarly, cDNA probes specific for RNA-1 and RNA-2 of GFLV both reacted with nucleic acid extracts from plants infected with the serologically related ArMV, but not with extracts from plants infected with several unrelated nepoviruses (Fuchs et a1., 1991). Curiously, a probe for ArMV RNA-2 was reported to cross-hybridize with RNA-1 but not RNA-2 of SLRSV (Hadidi et a1., 1992).

V. CODING REGIONS AND GENOME EXPRESSION

A. Assignment of Function to Genome Parts

Each of the two RNA molecules of the nepovirus genome encodes one large polypeptide (Section IV.D; Table I) that is cleaved to release functional proteins (see Section V.C.3). In genetic terms, each RNA therefore represents a physical linkage group. In early experiments to locate genetic determinants in the nepovirus genome, this property was exploited by inoculating plants with heterologous mixtures of RNA-1 and RNA-2 from distinctive strains (e.g., Harrison et a1., 1972). The formation in this way of virus isolates with new combinations of genetic material was termed "pseudo recombination" to distinguish it from recombination in which new nucleic acid molecules are formed. The isolates obtained were described as pseudorecombinant isolates or, latterly, pseudorecombinants.

The serological specificity of pseudorecombinants was that of the virus strain contributing the RNA-2, showing that, as with most bipartite genome viruses, the coat protein is encoded by the smaller of the genome parts. Similarly, the transmissibility by Longidorus elongatus of RRSV isolates of the English serotype (Harrison et a1., 1974) or TBRV isolates of the Scottish serotype (Harrison and Murant, 1977b) was inherited by pseudorecombinants containing their RNA-2. In contrast, the behavior of viruses in their host plants was mostly determined by RNA-I. For RRSV, the type of systemic symptom induced in Chenopodium quinoa, the ability of one strain to multiply in the raspberry cultivar Lloyd George, and the ability of others to infect noninoculated tissues of Phaseolus vulgaris were linked to the source of the RNA-I. There was also evidence of a more complex control of symptom induction. Determinants on RNA-1 and RNA-2 were shown to act in a supplementary manner to control the type of local lesion induced in C. amaranticolor, and a determinant on RNA-1 seemed to be epistatic to one on RNA-2 in the control of systemic symptoms in Petunia hybrida (Harrison et a1., 1974). One possible mechanism of such an epistatic effect would involve cleavage of the polyprotein encoded by RNA-2 by the protease en-


coded by RNA-l (see Section v.e). Transmissibility of isolates of RRSV and TBRV through seed was mainly determined by RNA-I, although RNA-2 had an additional but smaller influence (Hanada and Harrison, 1977).

Further evidence for the assignment of gene functions to genome parts came from experiments in which protoplasts were inoculated with each part separately. Whereas in protoplasts inoculated with RNA-2 there was no apparent replication of the inoculum, in protoplasts inoculated with RNA-l alone infective progeny RNA-l did accumulate (Robinson et a1., 1980). This result showed that enzymes required for replication are encoded by RNA-I. Also, because TBRV RNA is very poorly infective unless it carries an intact VPg molecule (Mayo et al., 1982), the accumulation of infective RNA-l suggested that VPg and the prot eases needed to release it from the polyprotein are also encoded by RNA-I. These deductions have since been confirmed by direct sequencing of RNA-l and VPg (see Section VI.B.3). Furthermore, because in contrast to the results from protoplast experiments no accumulation of RNA-l was detectable in plants inoculated with RNA-I, it seemed likely that at least one of the determinants of virus movement between cells was carried on RNA-2.

Although pseudorecombinants were formed with various combinations of RNA-l and RNA-2 from different strains of RRSV (Harrison et al., 1972) or CLRV (Haber and Hamilton, 1980; Jones and Duncan, 1980), not all such combinations from strains of TBRV belonging to different serotypes were infective. A mixture of RNA-l from TBRV-G (German serotype) and RNA-2 from TBRV-A (Scottish serotype) was infective in plants but the reciprocal mixture was not (Randles et al., 1977). However, when protoplasts were inoculated with the mixture of TBRV-A RNA-l and TBRV-G RNA-2, T-component particles lacking nucleic acid were formed after 2 days in culture (Mayo and Barker, 1983a). Thus, TBRV-G coat protein had been translated from replicated TBRV-G RNA-2 and protease had been translated from replicated TBRV-A RNA-I, but neither RNA had become encapsidated.

In general, pseudorecombinants are formed only when the RNA-l and RNA-2 molecules are from strains of the same virus. Although viable pseudorecombinants were obtained between TBRV and GCMV (Doz et al., 1980; C. Oncino, personal communication), these two viruses are closely related serologically (see Section VIII).

B. Domains in Protein Sequences

The amino acid sequences of nepovirus polyproteins contain several domains that resemble those to which putative functions have been assigned in proteins of other viruses. The ORF in RNA-l encodes RNA-dependent RNA polymerase, protease, and NTP-binding domains (see Fig. 2a). The arrangement of the domains in the RNA-I polyprotein resembles closely that in the RNA-l polyprotein of cowpea mosaic comovirus (CPMV) (Fig. 2a) and

156

a) RNA-1

TBRV GCMV GFLV ToRSV

CPMV

b) RNA-2

TBRV GCMV GFLV

RRSV

ToRSV

SLRSV

CPMV

I NTB I

M. A. MAYO AND D. J. ROBINSON

/I I I I

VPg I

pro I

pol I

II I

MP CP

CP

CP

CP1 CP2

MP CP1 I CP21

FIGURE 2. Diagram of the cleavages known to occur in the polyproteins encoded by RNA-1 or RNA-2 of several nepoviruses. NTB indicates the location of putative nucleotide triphosphate binding activity; pro indicates the location of putative protease activity; VPg indicates the location of the genome-linked protein; pol indicates the location of putative RNA polymerase activity; CP indicates the coat protein(sl; and MP indicates the location of putative movement protein activity.

to a lesser extent that in the polyproteins of potyviruses and picornaviruses (Koonin and Dolja, 1993).

Coat protein and VPg can be obtained from purified virus particles, and the determination of the terminal amino acid sequences of these proteins has been used to locate them in the polyproteins. VPg is located centrally between the putative NTP-binding proteins and proteases encoded by GFLV RNA-1 orTBRV RNA-1 (Pinck et a1., 1991 j Hemmer et a1., 1995). Coat protein is the C-terminal part of the RNA-2 polyprotein of GCMV (Brault et a1., 1989), GFLV (Serghini et a1., 1990), TBRV (Demangeat et a1., 1991), ArMV (Bertioli et a1., 1991), RRSV (Blok et a1., 1992), ToRSV (Buckley et a1., 1993), and BLMV (Bacher et a1., 1994b). Figure 2 shows maps of the functional


domains in polyproteins of those nepovirus RNA species for which complete nucleotide sequences are known. In general, the maps are very similar to those of CPMV polyproteins (Fig. 2). The maps of RNA-1 of GCMV, GFLV, and ToRSV are essentially similar to that ofTBRVRNA-1 (Le Gall et a1.,1989; Ritzenthaler et a1., 1991; Rott et a1., 1995). The RNA-2 maps are more diverse because of differences in length of the sequences to the N -terminal side of the coat protein, which have potential coding capacities ranging from 68 kDa (RRSV) to 149 kDa (ToRSV). For TBRV and GCMV (Demangeat et a1., 1991), and GFLV (Margis et a1., 1993), it has been shown thatin vitro and in vivo the RNA-2 polyprotein is cleaved into three products. For other nepoviruses it is not known what cleavage products other than coat protein are formed.

Figure 2b also illustrates the difference between SLRSV RNA-2, which encodes two coat proteins, and RNA-2 of definitive nepoviruses. The map for SLRSV RNA-2 closely resembles that of CPMV RNA-2. This is discussed further in Section VIII.

C. Expression of N epovirus Genomes

1. Translation in Vitro

Although nepovirus genomes resemble those of comoviruses (e.g., CPMV) and potyviruses [e.g., tobacco etch virus (TEV)] in encoding large polyproteins, those of CPMV (Goldbach et a1., 1981) and TEV (Dougherty and Hiebert, 1980) are readily cleaved during translation in vitro, whereas nepovirus polyproteins are at most only partially cleaved (e.g., Demangeat et a1., 1990). Nevertheless, either by direct radiolabeling or by immunodetection it has been possible to deduce several of the cleavage sites in polyproteins of TBRV and GCMV (Demangeat et a1., 1990, 1991, 1992) and GFLV (Margis and Pinck, 1992; Margis et a1., 1993, 1994). The RNA-2 polyproteins of GFLV (Morris-Krsinich et a1., 1983) and TRSV (Forster and Morris-Krsinich, 1985) were cleaved only when RNA-1 translation products were present, which showed that RNA-1 encodes the protease. Moreover, the cleavage was relatively specific: GFLV polyprotein was not cleaved by the RNA-1 translation products of either TRSV (Morris-Krsinich et a1., 1983) or TBRV (Demangeat et a1., 1991). However, the GCMV polyprotein was cleaved by TBRV protease, which presumably happens when pseudorecombinants are formed between GCMV RNA-2 and TBRV RNA-l and must reflect relatively similar sequence specificities of the proteases (see Section V.C.3).

2. Production of Proteins in Vivo

Except for the proteins in virus particles (coat protein and VPgL there is little experimental information about the production of nepovirus proteins in infected cells. Experiments to detect proteins specifically involved in the


multiplication of RRSV in infected protoplasts showed that up to 24 infection-specific proteins were formed (Acosta and Mayo, 1993). Several of these appeared to be related to one another as charge isomers or variously modified forms of one protein, but the likelihood of partial cleavage and the appearance of infection-specific, host-coded proteins made simple conclusions difficult to reach. Moreover, it has been shown recently that GFLV and TBRV proteins, which would be termed partial cleavage products because they could be further cleaved, are functional in the partially cleaved state (see Section VI.B.2). An alternative approach, in which proteins from infected cells are allowed to react in immunoblots with antisera specific to one of the final cleavage products, has yielded clearer results for some of the proteins of TBRV (Demangeat et a1., 1992) and GCMV (Hibrand et a1., 1992) (see Section VI).

3. Cleavage Sites in Polyproteins

Table III lists the dipeptide sites at which various nepovirus polyproteins are known to be cleaved. Although the same dipeptide is cleaved at different sites in some polyproteins (e.g., TBRV, SLRSV), 11 different dipeptides are known to be cleaved in the nepovirus polyproteins studied so far. Moreover, in the polyproteins of GFLV, four different dipeptides are cleaved and none is cleaved at more than one site. This wide range of different cleavage sites in nepovirus polyproteins illustrates the danger of trying to predict probable sites from what is known about the cleavage sites of the polyproteins of CPMV or picornaviruses. It is not known what determines whether cleavage

TABLE III. Cleavage Sites in Nepovirus Polyproteins

Virus Cleaved dipeptide Proteins separateda Reference

GFLV R-G MP-CP Serghini et al. (1990) C-A 28K-38K Margis et al. (1993) COS NTB-VPg Pinck et al. (1991) G-E VPg-pro Pinck et al. (1991)

TBRV K-A MP-CP Demangeat et al. (1992) K-A NTB-VPg Hemmer et al. (1995) K-S VPg-pro Hemmer et al. (1995)

GCMV R-A MP-CP Brault et al. (1989) ArMV R-G MP-CP Bertioli et al. (1991) RRSV C-A MP-CP Blok et al. (1992) CLRV Q-S MP-CP Scott et al. (1993) SLRSV S-G MP-CPl Kreiah et al. (1994)

S-G CPI-CP2 Kreiah et al. (1994) ToRSV Q-G MP-CP Sanfa~on (1995) TRSV C-A MP-CP Buckley et al. (1993) BLMV N-S MP-CP Bacher et al. (1994b)

acp, coat protein; MP, putative movement protein; NTB, protein containing an NTP-binding domain; pol, polymerase; pro, protease.


occurs at one particular dipeptide site rather than at other similar sites, nor how a single protease can cleave diverse dipeptides.

VI. PROPERTIES OF PUTATIVE GENE PRODUCTS

A. Products of RNA-2

l. Coat Protein

The amino acid sequences of the coat proteins of nine definitive nepoviruses and of SLRSV have been determined. In pairwise comparisons among the coat proteins, made by using CLUSTALV, software for multiple sequence alignment (Higgins et a1., 1992), only two pairs of viruses had coat proteins that were more than 30% identical and the viruses in each pair were serologically related. These viruses were GCMV and TBRV (57% identical) and ArMV and GFLV (69% identical). No similarities were detectable between any of the coat proteins of the nine definitive nepoviruses and either of those of SLRSV (tentative nepovirus, cluster d, Table I).

Figure 3 shows an alignment based on secondary structure predictions of the coat proteins of nine definitive nepoviruses obtained by using the HSSP

program (Rost and Sander, 1993, 1994). Many l3-sheets and some a-helical domains were predicted. By reference to the model for the secondary structure of CPMV coat proteins (Rossmann and Johnson, 1989), it was possible to organize the predicted l3-sheets into three sets of eight (131, 132, 133, in Fig. 3), which could make a T = 1 structure particle. The letters attached to the l3-sheets in Fig. 3 are based on this assumption.

In Fig. 3 the sequences are grouped according to the clusters listed in Table I. The strong similarities between ArMV and GFLV (342 residues in common) and between GCMV and TBRV (227 residues in common) were unsurprising, but the alignment shows that there are also some amino acid residues common to viruses in either cluster a (47 residues) or cluster c (44 residues). Nine residues were the same in all nine sequences.

Figure 3 also shows that, although most of the coat protein sequences can be aligned at their Nand C termini, ToRSV coat protein has an extended C-terminal sequence, which would not greatly alter the supposed structure of the virus particle. In contrast, CLRV coat protein seems to lack sequence at the C terminus, which includes two l3-sheet regions. The CLRV sequence also lacks a tripeptide resembling the F/L-Y/W-G motif present in all the other sequences and previously thought to be highly conserved (Blok et a1., 1992). It will be of considerable interest to see how the coat proteins of other nepoviruses, and indeed of other strains of CLRV, fit the alignment in Fig. 3. In similar alignment tests, Le Gall (personal communication) has detected the similarities shown in Fig. 3 and also has detected some similarity be-

Ar~V

GfL

v RR

SV

TRSY

TB

RV

GC~Y

BLM

Y To

RSV

(LRV

A ro

W

GFLV

RR

SV

TRSV

TS

RV

Go.W

SL

. ... V

To

RSV

CLRV

ArM

Y G

FtV

RR

SV

TRSV

TB

RV

GCMV

BL

MV

ToRS

V (L

RV

( ~IB]

[ Q

]

C

• ]

( p

IC

]

[ ~I

O J[

Q

J

GL

AG

RG

SVQ

I' QAG

----[Y

~TLOL

~OOM

YSGF

SGIQ

YElr

A~YM

~FKV

Y[RY

PAN~

TITW

VMS.

O 0

SSIS

T--

-T

GLAG

RGVI

YI"

, QA

N--

--RY

GT

LNI"O

MIS

OFKG

VQYE

IT

A LV

M"

FKIV

IRLP

ANA

T

TWVM

S '0

0 SR

ITA

---S

AY

EvDP

LHLL

YYES

VO

" •

LAG

----

fL

RIO

" KA

AIFO

SAVW

R RO

LK

" IK

MRl

TAAT

SC

S V

LCA

C"

0 AA

TKT

---G

AY

TVVP

OPTC

(GTL

SI'K

••

KKG

----

KH G

HO

lo

AIM

DVGG

LHSQ

AK

VN

FT

VRM

I<APR

NA

A

SIA

(T..

0

LPAL

GNE(

P

iI;I~~

:I~:

:::

.~~i

S:~~

~'~~

•• IB~

~~::

lIu:

tfta

llrt

.::~

SGLIAOTSIAHVVQCWVPKOATKG----RVIAINLREOIATSONLVKYIAKGLIHPOLKLRM

TVG!FV~SIGICC~FG~SKYYEGO--T

GGSW

QEGT

EAAY

LGKV

TCAK

OAKG

G----

TL

TLOI

IKE(

KSQN

LLRY

K RQ

CFLH

GKLR

LRCF

IPT

FC

SMN

(SL'

FG

o O

SNV

LG---

A

SRIT

PGM

HwSA

VTPF

KCAA

EAAE

NTl

RWSL

RS II

SESG

TOAW

T RE

LPTT

FL V

TGTl

AMS

lolA

LGLV

(' FN

° --

SKLl

O--

-T

I I'

• I

I I

• •

I

(~Ie]

[ pl

F ]

(~IG

] [

~ ]

( ~Iot ]

A

SP--

4YTL

$VPH

WLL

HHKN

G-rT

-SC

DLD

YG

ELC

I§FG

A rT

F€SP

KLH

FTCL

TIK

EL

AA

I---

---F

WEL

YA

EFE-

AA

-KSH

GK

-AO

P--V

YTLS

VPHw

LIHH

KLG-

TF -S

CEID

YG

EL(

WFK

STIF

ESPR

LHFT

CL T

KE

LAA"

--

----A

WEl

YAE

L E

-EA

-TSF

LGK

-IT

A--S

LVTG

LPNT

VwAT

ROT-

SEVE

wOIO

LAAV

C FF

ALEO

TFGY

MOF

LIYV

LR

ITAY

A.

----

--IYV

SFHV

OWT-

Q--

ESM

LATL

I PS

E --

MFE

LPT

KVFM

LKOA

O--

VII-

EWQF

NYGE

L T

L(

NWAN

VA T

QPTL

YFF

VAS

T

VTM

AA'

----

--C

lVTM

HVOM

r. -PY

IORF

E LN

--

g-

-wm

'oD

lvD

-ls -

~-flllI'lli!lHa'g~GY(VPK~FmARS T

ATK-

----

-FfiW

FYTR

GE

-KL

-VR(

,LAE

-

g--I~cDivDSK

-~-vtr

' , ~GFSKPT~mpWSAQVT------vm.LAQGOEIA-iIGLATR--

ALP

[ -EV

CNQL

PNFV

CPIS

EK -$

V -FE

FDLD

loIi

~NL

FQTS

KGFA

OPVl

LVY UD~LPASO~-

----

-YlC

EVCI

KSA

-LH

-A TS

VAt~

-

SFPV

KLAS

LLPT

EVIS

LAOG

P-VV

-TW

TFOI

G GL

YYSE

GAYA

RPKI

YFLV

LS

VPAE

A --

----

FTY

QLL

FEO

H-T

F-SN

SFG

A-

FPS-

-TLG

QNIo

IPQR

VFPL

$NPL

ER-N

FSFS

IolG

E L

TM

HPQA

TAFE

OVQF

ILYV

LNT

VA(A

AE

GHIL

wQIo

IKOO

ANAP

YE-L

Q-LP

VVPK

--I

• I

I I

I

(JIl

l]

[ ~l

8 ]

(jll

C]

[ Q

]

( ~Z

O ]

C

o2e

]

~FIYSLOAFNGSL

-KFLTIPPLEYiATSAYK1-VSLL~TL-

-V-OGTHKVYNFNNTLLSYYLGIG~VKGKVHVCSP(TYGIVLR~SEW

"LV

FDPG

VFN

GK

F-Q

FLTC

PPIF

F AV

TALR

.-AGL

TL

VP--~-VGTTKVYNLNSTLVSCVLGMG

VRGR

VHI(A

PIFY

SIVL

W

SEW

•

FVW

PPKP

TDIS

L-LK

EVW

GPYR

FT

GTE

AK

E.-F

AIM

AI

--L-H

GQQI

VRTF

PRVV

AAHF

RSW

T VR

MSI

QEVS

SIFL

TGTY

M

VSE

MTW

PIQL

GOTF

A-ID

RYYE

AKEI

K GS

TSIo

lL.

ISYN

F PV

--K-

HSKK

HAIS

YSRA

VMSR

NLGW

S IS

GSVK

SVSS

-LF(

TASF

FP

W

QPLS

mI!E

AR':I!T!}-LNL~K~AV(!-

TYAM

Il -FPVSLAAKL-Q~

SIV

TmI!S

LE::J

]KjJ-IEIm~R!}4AI~AGSINF -P

lSFA

VQ

QK

-K

PIL

SLPH

FFO

CR

LPL

O.

o 'F

SFtL

GRSS

KREN

HIGI

N!SA

R-VV

-S~N

TFYS

FPAA

YTQL

LQSV

Gfrl

IVVQ

TGSK

AISC

E~FL

ILQP

V

PFIT

LPIfI

FNRl

O-IG

Y

0 TE

IDlT

STPA

PNA

-YRL

L LS

T--V

IS

MST

LNAN

QALL

RFFQ

GSN

L.

IKK

IGTA

LTTC

SLLl

SlRH

--

----

----

OG

AR

-LD

0

'ATF

AQ

GSF

PSTI

N-V

NLG

EP

RTVL

-T

APIT

SFHQ

AALS

YFIS

YG

Tr

LYKIGSGLVQvOIALA~WH

I •

FIG

UR

E 3

. A

lign

men

t of

nep

ovir

us c

oat

prot

eins

by

the

prog

ram

HSS

P (R

ost

and

San

der,

199

3).

at a

nd 13

sho

w

the

pred

icte

d al

pha

heli

ces

and

beta

she

ets,

res

pect

ivel

y. T

he

lett

ers

on

the

l3-s

heet

s ar

e in

thre

e se

ts (1

31,1

32, an

d 133

) an

d ar

e de

sign

ated

by

anal

ogy

wit

h th

e kn

own

stru

ctur

e of

com

ovir

us c

oat p

rote

ins

(Ros

sman

n an

d Jo

hnso

n,

1989

). R

esid

ues

that

are

the

sam

e in

one

or

mor

e of

the

thr

ee c

lust

ers

are

show

n in

rev

erse

con

tras

t; -

show

s ga

ps in

sert

ed b

y th

e pr

ogra

m to

ach

ieve

ali

gnm

ent;

* i

ndic

ates

whe

re a

res

idue

is c

omm

on to

all

nin

e se

quen

ces;

I i

ndic

ates

a p

osit

ion

whe

re s

even

or

eigh

t of

the

nin

e se

quen

ces

have

th

e sa

me

resi

due.

0::

o ~ ?> ~ o ~ I:l ~

':-'

~ o SS

Z

C/l o Z

.\rlo

lV

GeLV

~RSV

TRSV

TB

RV

GCMV

SL

MV

ToRS

V CL

RV

ArM

V

GFLV

RR

$V

TRSV

TB

RV

GCMV

BL

MV

ToRS

V CL

RV

Ar'l

V

GFLV

RR

SV

TRSV

TB

RV

GCMV

BL

MV

ToRS

V CL

RV

Ar'l

V

GFLV

RR

SV

TRSV

TB

RV

GC)lV

9L

)lV

ToRS

V (L

RV

c '"

] ( ,

'F]

( ~'G]

( ~I

~ ]

~ll

J NGVT-NNWNQLFKYPCCY-I-EEDGilEIRSPYHRTPLRLIDAQSASSFTSTLNFYAIS~!A

'~GETAKMPVVVQI~E!A--------

NGTT

-IoID

WNE

LFKY

PGV

Y-V

-HD

GS

EVKI

RSPY

HRTP

ARLL

).GQS

Q-RD

MSS

LNFY

AIA

L\'~~GETAQLPIVVQI

EIV

----

---

-~-A

T -AOL

AOIV

TRKH

WI -

V -K

SNE

I EV

OL Y

CPYG

ENPT

FTGQ

ANGK

---

-;>F I!V

fIKLGf

rVGiI~

DSVGIF

~;:'Ir

HI GL

TGVY

KNPT

LH

ECEA

PPTL

RQVL

WGP

HQI-M

-HGD

G E!

AIKT

RLHS

AATT

EEGF

GR--

----

-LGI

LPlS

~IAi

UAHV

GSYE

;:!V

HI

TW

R--

----

---

TVPE-TDQLAQHSHV~-V-.SlQ--~FSTA

-------1GNSGSAFY\!ISTLCA~ETGSml!D<mE-----

---

OT

n ---lEQLSQMP~-C -C>a!1JVSSLK~ATANFGOSGAR[IW-

----

--l/

TP

'lSS

a»tE

SX

LIl

iLl!

1lL

mo-

----

----

OKTA

-HNL

EQAL

RL~C

R-IP

iG~S

!RIQ

T,~QREQ!FNTGVQLV

--------IY

AVGGPHGAQAISAPYQ!HIHIQ

--------

KDAS

-LTL

ETAY

QR,"

YIL

A-

G

SLPI

ST,'

AT--------SFL~DMLRLElFAI,\G?FSPKDNKAKYQ

Y

HIE

----

----

DN

ADGV

EFRE

LIKI

" V

LlS-

G

GE S

LPLN

AI'

GH

TSTR

DRG

PTM

A---

-----V

ClV

SGV

VA

PKD

CSA

PYR

Y

RV

E--

----

---

I I

I I

II

I C

~1B]

( ~lC]

( Q

]

C

~)O]

L-PD

lSV

PSFP

--N--OY

F~DF

SSFT

ViEE

Y ---

----

--V

----iSRF~~ SS

--TTS

~lG~

FAHM

IACH

liiL

IL!

R-P

DlS

lPSF

E--D

--OY

F O

FSEF

Tl'

EE

I---

----

--E

----

' SR

FF.

TS--N

TC

SMGE

FA

AMIA

CH

L

LQ

S-G

DR

SVG

SA--

----

----

RIN

NIA

' L

VF

----

----

-N--

--I

GR

IE'

IAAA

GKY

NY

TS

LL

FS

VT

I IT

P-

OSQV

HPPI

oIFS

SS--E

LY

LTNL

KP'

NT

G--

----

---V

VN

FDI

GYI

'AS

--K

OA

lAS

lSWLV

AAT

V

IS

ANPG

LCRE

INY

--~

--:!1

FAW

CL L

ECLO

NSKA

SP--

----

---r

-KV

Kla

IC(l

.SS

--,jtiVKUFelWL:JJIDliltl~!Tr~

----

-AOP

PMC-

-R--:J!NYDQRFAWFTLLRPPOPKlSKlLKlT-

---LaVCW!AY--\3EA~FWM~~:trL~

E -E

GO

EPR

PI -

-GN

Vra

L

TISE

MTN

-LTR

F --

----

---Q

----lJ

ARE~

VL --

PGQT

VTIo

IRRN

AIENLlRS~EV~FI

LVEG

V"

TIA

G--

E--

SFRNFKIOOWKF---------E----W.~R.

LD--O

KSEV

LLRQ

HP'

LLIS

S F

I F

FO

TQl"

VIAT

--R

--L

L

SF

SGFA

PVV

PEA

-----P

RTW

IoI--

--I.C

R

• KV

--EGH

LKIo

IEAH

P'

RLVA

S l

I I

I I

I I

I ••

• [

~lE

] [

~IF

J (

pIG

]

l-EG--CFGKSSIVTITKLCGOKATG--MOGAS------RVCA

LQNI

oIGC

-ETE

LYIG

NYAI

NPNT

A-L-

-SLY

SRWlA

~KLD

K L-

NT-

-eFG

KSS

V

TITK

LVG

OK

AM

G--l

OG

PS---

---H

VFA

IQK

LEG

-TTE

LLV

GN

FA

'NPN

TR-F

--SLY

SRW

MA

KlOQ

P-

NT-

-TlG

ESK

LK

YIoI

QYLY

HTAT

ENFF

GDQA

-----

-TRG

IIDQ

DG

F-Tr

OIA

CG

OFF

, RV

G--L

--PG

EVER

LG

YSSN

R-

SK--Q

AQAQ

E V

SITT

NY

RDW

GA

Y--W

QG

QA

------

RIY

DLR

RTEA

-EIP

IFLG

SYA

, PS

GA

-L--G

K-Q

NY

VR

SIV

N

WHPA--cLGKQl~LK~~INNEH--IGDTH------CYNSLSNTH--S~~KA--OEA~EIQSPD

L-NKGTSFKDL~IS~5~5TIGE HH

GEF-

----

-HLG

GPLS

SSL-

A~E~-

T-pCTS~RVETAH

L-NV--AHIVPTATMQILTAV~VGNA--ETNGSQILQSWIVPVSQVFEK-EVEIolOLTDYPGF~GG

-IG

AOHD

QPYI

DIACGN

L-HT

--TAG

NHKG

SFSA

RLAF

GV

EE--

IEQ

TS---

---TV

QPL

VG

ACE

ARI

PVEF

KTY

TGY

GP

-P--

GSM

EPYI

YVRL

TQ

F-A

$--S

LTTP

TTY

VQ

lVH

KF'

T

TS

----

----

----

----

----

----

--S

ES

yLT

KL

RH

AS

TV

-L--N

LovV

VAGL

GGFM

I

I I

[ ~I

H ]

[~ll

J

ISMKM

L~ILCKPRGNFEURTCFKV

. IK

VL'

LC

KPRP

GFS

Y

TSFP

V ,

SIA

E!'

5FEV

L-SM

N Y

STIK

VT

. O

IVA

l"VCL

RPK

-$IK

RS

ATLF

FPQI

FYIo

ININ

VRVH

PGFE

LYGR

SITP

LRI

~KLVDRLSVNVILQEGFSFYGPSVKHFKKEVGTPSATLGTNNPVGRPPENVDTGGPGGQYAAALQAAQQAGKNPFGRG

RLRS

G II

I

Z

tT1

"d o ;::; :;<

:I c:: (J'J

tT

1 ~ 5 r tT

1 ()

c:: r ~ i::l o r o C'l

0-< ~ :;<

:I tT

1 "d

t::

()

~ (3

Z §


tween the sequences of the coat proteins of definitive nepoviruses, SLRSV, and comoviruses.

A comparison between the C-terminal amino acid sequence of coat protein extracted from purified particles of TBRV with that predicted by the nucleotide sequence suggested that the terminal nine amino acids had been removed in vivo at some stage (Demangeat et al., 1992). In immunoblots of protein extracted from plants at intervals after infection, the protein reacting with antibodies to TBRV particles had an estimated size of 59 kDa in samples from leaves a few days after infection but of 57 kDa in samples taken after a further week. This suggests that the coat protein was affected by proteases in the plant cells after the particles were formed and that perhaps during purification any terminal nonapeptide remaining attached to the coat protein molecules was readily lost. The coat protein of GCMV lacks the C-terminal 10 amino acids that correspond to this readily detached sequence in TBRV coat protein (Fig. 3). Demangeat et al. (1992) speculated that this terminal sequence was on the outside of the virus particles and might playa role in the transmission of the particles by nematode vectors. The C-terminal amino acid in coat protein from purified particles of GFLV is valine, which corresponds to that predicted from the nucleotide sequence (Serghini et al., 1990).

2. "Movement Protein"

The RNA-2 polyproteins of TBRV and GFLV are cleaved into three products by viral protease. Meyer et al. (1986) suggested that the protein adjacent to the coat protein had weak sequence similarities to movement proteins (MP) of some tobamoviruses and to the putative MP of CPMV. These similarities have also been discussed by Melcher (1990) and Koonin et al. (1991), but it is unlikely that clear evidence as to function can be deduced from such sequence comparisons.

More suggestive (d. Demangeat et al., 1992) is the similarity in location in the genome of the putative MP of nepoviruses and that of CPMV (Wellink and Van Kammen, 1989). Moreover, a striking similarity between comoviruses and nepoviruses is that cells infected with either type of virus produce tubules, many of which contain particles. The putative comovirus MP is associated with these tubules (Shanks et al., 1989; Van Lent et al., 1990) and recent work has shown that the analogous protein of ToRSV also is associated with tubules (Wieczorek and Sanfac;on, 1993).

A multiple alignment of the putative MPs of ArMV, GCMV, GFLV, RRSV, TBRV, and ToRSV indicated the existence of a conserved proline residue in a hydrophobic region (P motif); such a proline-containing motif is also detectable in the putative MPs of caulimoviruses, comoviruses, and capilloviruses (Mushegian, 1993). However, there was little other sequence similarity shared by all nepovirus MPs. Pairwise comparisons among MPs are more revealing. Figure 4 shows DIAGON plots that compare the amino acid sequences of the RNA-2 polyproteins of GFLV, ToRSV, RRSV, and TBRV.

NEPOVIRUSES: MOLECULAR BIOLOGY AND REPLICATION

> CIl a: a:

b)

SOD

.

a)

, , , / .

TBRV

500

, ,

,.

1000

, .

.'

~' ,,"

,

.

.

1000

. .'

/ ,

ToRSV

1000

> ...J U.

500 C)

c)

1500

/' , , , , .

500

, .

163

1000

500

1000

, . . . 1000 . , . / .

. 500

TBRV

FIGURE 4. DIACON plots showing comparisons of the polyproteins encoded by RNA-2 of (a) GFLV and ToRSV, (b) RRSV and TBRV, and (c) GFLV and TBRY. The thick diagonal lines show where COMPARE (Devereux et a1., 1984) registered a match (stringency = 17; window of 30 residues!; the thin vertical and horizontal lines show the boundary between the coat protein (to the top and right) and the sequence to the N-terminal side of the coat protein. The numbers are those of amino acids from the N termini of the polyproteins.

The coat proteins of all four viruses are equally similar, but the MP parts of their polyproteins are not. As noted by Rott et a1. (1991b), the MPs of GFLV and ToRSV show relatively strong similarities (Fig. 4a). The same is true for the RRSV and TBRV MPs (Blok et a1., 1992) (Fig. 4b). However, the MPs of GFLV and TBRV (Fig. 4c) and ToRSV and RRSV (not shown) show no similarity in comparable plots. Because GFLV and ToRSV are transmitted by Xiphinema spp_, whereas RRSV and TBRV are transmitted by Longidorus spp_, it has been tentatively suggested that the MP plays a role in vector specificity (Blok et a1., 1992; Mayo et a1., 1994). However, SLRSV too is transmitted by Xiphinema spp. but there is little similarity in sequence between the MPs of SLRSV and GFLV (Everett et a1., 1994; Kreiah et a1., 1994)_


3. N-terminal Protein

The N-terminal proteins in the RNA-2 polyproteins of different nepoviruses are diverse in size. Those of TBRV and GCMV are about 40 to 50 kDa (Demangeat et a1., 1991), but those of GFLV (Margis et a1., 1993) and, putatively, RRSV (Blok et a1., 1992) are about 28 kDa. That of ToRSV is 104 kDa (Wieczorek and Sanfayon, 1993) and contains a substantial threefold sequence repeat (Rott et a1., 1991b). There is no corresponding protein in this position in the genome of CPMV (Fig. 2). The nepovirus proteins are dissimilar in sequence (e.g., Fig. 4) and there is little to suggest their function. Sanfayon (1995) has suggested that the region may be involved in RNA replication. The 44-kDa N-terminal protein of GCMV was found in both soluble and membrane fractions from inoculated leaves of Chenopodium quinoa and, like the MP, was detected only transiently between 2 and 5 days after inoculation (Hibrand et a1., 1992).

B. Products of RNA-l

1. Polymerase

The 159 amino acid residues of the GFLV, TBRV, GCMV, and ToRSV polypeptides illustrated in Fig. Sa contain the recognized consensus sequences of RNA-dependent RNA polymerases (D-x3-[F,Y,W,L,C,A]-Xlo.llD-~-[S,T,M]-G-x3-T-x3-[N,E]-xn-[G,S]-D-D) (Koonin and Dolja, 1993) and the polypeptides of about 80 kDa that contain these motifs are assumed to be polymerases. These are the most similar among polypeptides of different nepoviruses. Comparisons by CLUSTALV of regions slightly larger than those shown in Fig. 5 gave 81 % identity between the TBRV and GCMV polypeptides and 45 to 53 % identity between any other pairs among the TBRV, GCMV, GFLV, and ToRSV polypeptides. Of the 52 residues in this region that were common to all four nepovirus sequences, 45 were also present in the corresponding region of the polymerase of CPMV.

2. Protease

The main parts of the protease sequences of the above four nepoviruses are shown in Fig. 5b. They are detectably similar but contain only two conserved tripeptides, H-Q-A and D-C-G. However, the Hand G residues of these tripeptides are part of a motif characteristic of viral cysteine proteases (H-xn-E/D-xn-C-G-~-G-~-G-xn-H-xn-G) (Rott et a1., 1995) (marked - in Fig. 5b) in which the H (underlined), E, and C residues are thought to form the catalytic triad of the enzyme (Gorbalenya et a1., 1989). The final H of the motif is present in ToRSV protease but is replaced by L in the other nepoviral proteases. This replacement may contribute to differences in dipeptide specificity between different proteases (Sanfayon, 1995). Margis and Pinck (1992)


a) polymerase _. GFLV -SDVGYNCDYKAFDGLITEQILSTIADMINAGY----RDPVGNRQRKNLLLAICGRLSICGNQVYATEAGIPSGCAL ~ KTNEAINCDYSGFDGLLNAQYIECIAKMINRLYALSGESevQQAqRYNMLMALVGRYAFVGPEVYKVNCGLPSGFAL GChfV KTNEAINCDYSGFDGLLTPQLVEMMAKMINRLYLRSGESEVMQAQRLNMIMALCGRYALVGTQYYKVNCGLPSGFAL ToRSV KNSVALNCDYSRFDGLLNYQAYVHIVNFINKLY-----NDEHSIVRGNLLMANYGRWSVCGQRVFEVRAGMPSGCAL

GFLV ~ GChfV ToRSV

I •••• ····1· I ••• • ·111·1 •• I ··1 ·1··· •• . -TVVLNSIFKELUlRYCFKKIVPPVYKECFDRCVVLITYGDDNVFTVAQSVMQYFTGDALIOoQ4AKLGYTITDGKDK5L TVVVNSVFNEILIRYAYKKLAPKPERNRFNQYVCLLVYGDDNLISVSPSIASIfTGEAIRITLAEKKVKITDGSDKDA TVVMNSIFNEILIRYAYKTLAPTPEKNSFGINVCLLVYGDDNLISVSPAVASWFTGEAIRVTLAEKRIKITDGSDKDA TVIINSLFNEMLIRYVYRITVPRPLVNNFKQEVCLIVYGDDNLISIKPDTMKYFNGEQIKTILAKYKVTITDGSDKNS ··11··1·1·1·1·· II • I· ··1 ·····1 III • *1 I I· I ••••••

b) protease _

GFLV QGEHEELVTELYVYC--DGVKKLISTCWFKGRSLLMTRHQALAVPIGNEIEVIYAD--GTTKK-LVWPGR TBRV QAGDGLLPAARFVCCYLSTGGGFVSANQYKNKSVRMTRHQALRFQEGEQLTVIFSSTGESQLIRWHKYHM GChfV QAGNGLLPASRLCVAIYGPRGYFISGMQYKNKCVMMTRHQAQSLNEGDELSVVFASTGESMMIRFHAYHI ToRSV QFNESHAVNMLVRID--LPDGNIlSACRFRGKSLALT~LTIPPGAKIHIVYTDNNGNTKAPLTHFFQ

GFLV TBRV GChfV ToRSV

GFLV TBRV GChfV ToRSV

• I II· II I I 1·1··· • 11111 -QEDGNCKGFVEF-PENELVVFEHPHLLTLPIKYEKYFVDDADRQISPNVAVKCCVARLE-----------REEP----------GSEIVTWlAPSLPSLSPDLKDLFLEDKEVDLPNHFKTIGYVLRVDNTAF------RENV----------GSEVVCWLAPSLPQLPCDLKGLFLEDAEVELPSNFKSMGYVLRQDSNAF------PTGPNGEHFLRFFNGTEVCIYSHPQLSALPGAPQNYFLKDVEK-ISGDIAIKGCGIKLGRTSVGECVGYK

*1 *··1 ·1 • I II I I - . _.DGIPQFHFWSKYATARSEVHTLKDEGGGNVYQNKIRRYIVYAHEAKKYDCGALAVAVIQGIPKVIAMLVSG ----HYDLLDTYAAVDKTPLPLKGVVGNELYLHEIPEKITFHYESRNDDCGMIILCQIKGKMRVVGMLVAG ----HYDTLDTYAAVDKTPLVLKGYNGDDLYIHEIPEKIVFHYESRNNDCGMLLTCQLSGKMKYVGMLVAG DNEPVLNHWRAVAKVRTTKITIDNYSEGGDYSNDLPTSIISEYVNSPEDCGALLVAHLEGGYKIIGMHVAG

I • I • I I· ••• I I· 1111· ·1·

c) NTP-blndlng protein GFLV WVYIFGASQSGKTTIANSIIIPALLEEMNLPKSSVYSRPKTGGFWSGYARQACVKVDDFYAIE TBRV WIYLFGQRHCGKSNFMATLDN-ALAKHFGLPNTTAY-RNCKDSFFSGYSGQTFFHVDDLSSVK GChfV WIYLWGPSHCGKSNFMDVLGM-ALCKHFDLPYTVCG-RNYKDSFFSGYMGQTIMEIDDLSSIK ToRSV WVYLYGGPRCGKSLFAQSFMN-AAVDFMGTTVDNCYFKNARDDFWSGYRQEAICCVDDLSSCE

·1·1· ··1 I I • I I· ••• II 1··1 I

d) 'Protease co-factor (CPhfV 32K analogue) GFLV FDVTMAPYLQHLASAHSILKKIWEKLSEWMESLKSKASLALEYMRQHAIFALGAMVIGGYVVLVEKVLIAAKII TBRV FRECIKMIHKELGCAMELIEYMIKKVKDWYNSMLEKLHCGLATLGTYAMYALAILLGCGLTTLLERCIGGAGIL GChfV FREALTTIKFELGYAMELVEVLIARVKSWFDTLLAKIDHALASLGKWACYALGILLGIGLCNLIETIIGGHGML ToRSV FDDTIGKWIPKLLGATQKIEELWRWSLEWAQNMSKKLDVSLRVLRGSALVGYGLLLVSGILYFAEQLLRSFGLL

* I •• I I • I I • I· I • III II ·1 I· I I II

e) N-termlnal protein

GFLV RAARRSAACKKYRAKRALAEFEAIVQSERLDQLKTGFQYVLPAPK TBRV RLSRKYAALTARVRAKRAAARELREKELFLETQDLLNAPLLPPME GChfV KLSRKYADL TAQVRARRAAARDLRAKEIYLEIVDLLGAPLLSIPQ ToRSV RKAAKYAAFAARKKAAAVAAQKARAEAPRLAAQKAAIAKILRDRQ

I I I • * • I·

FIGURE 5. Alignments of regions of sequence similarity from different parts of the polyproteins encoded by RNA-l of four nepoviruses. The regions are parts of (a) the putative RNA polymeraseSj (b) the putative proteasesj (c) the putative NTP-binding proteinsj (d) the putative protease cofactorsj (e) the N-terminal proteins .• indicates residues thought to belong to the consensus sequence of polymerase (a) or protease (bl; * indicates residues identical in all four sequenceSj I indicates residues that are similar in all four sequences.


showed that changing the L residue in the protease of GFLV to H by mutagenesis of cDNA clones in expression vectors resulted in a loss of protease activity. The reciprocal experiment with ToRSV (Sanfayon, 1995) showed that a change of H to L also caused a loss of protease activity, as did changing either the C residue to I or the L to H in TBRVprotease (Hemmer et a1., 1995).

Detailed study of the protease of GFLV expressed by transcription and translation of a cDNA clone showed it not to belong to serine, cysteine, aspartic, or metalloprotein classes of proteases (Margis et a1., 1991). As in the proteases of TEV and CPMY, cysteine is thought to replace serine in a trypsinlike family of serine proteases (d. Bazan and Fletterick, 1988). With the GFLV protease, changing C to L led to a loss of proteolytic activity, whereas a change of C to S had no effect (Margis and Pinck, 1992); however, the latter change altered the catalytic triad to D-S-G, characteristic of the chymotrypsin superfamily of proteases. In comparisons between the entire sequences of their proteases, as for the other proteins, TBRV and GCMV were relatively similar (68% identity); GFLV and ToRSV proteases were 35% identical and other pairwise comparisons resulted in values of about 20% identity.

The proteases of nepoviruses appear to be functional, at least some of the time, when joined to neighboring polypeptides. Thus, TBRV polypeptides comprising the combinations VPg-protease-polymerase or protease-polymerase were able to cleave polypeptides in trans (Hemmer et a1., 1995). Indeed, the protease-polymerase polypeptide was found to accumulate in TBRV-infected plants (Demangeat et a1., 1992). With GFLY, Margis et a1. (1994) showed that, whereas protease alone was more effective than the combination of VPg-protease at cleaving the RNA-2 polyprotein, the VPgprotease was the more effective at cleaving at the N terminus of the RNA-l polyprotein. Thus, whether or not protease is attached to other proteins may determine which of the various possible proteolytic activities is favored at a particular time in virus multiplication. It has also been suggested that cellular factors have an effect on protease activity.

3. VPg

The amino acid sequences of the VPg of GFLV (Pinck et a1., 1991) and TBRV (Hemmer et a1., 1995) have been determined directly, which means they can be located precisely in the RNA-1 polyprotein. The sequence of the polyprotein of GCMV RNA-1 is sufficiently similar to that of TBRV RNA-1 for the sequence of GCMV VPg to be predicted, and Rott et a1. (1995) deduced the sequence of ToRSV VPg in a similar way. Alignment of the sequences (Fig. 6) shows that with the insertion of only one gap in two of the sequences, it is possible to detect a consensus sequence of E/D-xI1_21-Y-xI2rR-N-xI4_sl-R.

The sequences of the VPgs of three comoviruses (CPMV, cowpea severe mosaic virus, and bean pod mottle virus) are much more similar to each other than are the nepovirus VPgs (Chen and Bruening, 1992b). However, compari-


+1-

GFLV SEPRLEERYSPRNRFV-SRISKIRG V3

TBRV AQQKSSSQEGGYRARNIPIHH 5/1

GCMV AHSVYSADGGDRGYRSRNIPINH 5/2

ToRSV STIPSGSYAD-VYNARNMTRVF 3/1

FIGURE 6. Alignment of the known sequences of VPg of GFLV and TBRV with those deduced from the polyprotein sequences encoded by RNA-l of GCMV and ToRSV. The boxed residues form the suggested consensus sequence; + / - indicates the proportion of positive and negative charges in the VPg sequence.

son of the comovirus and nepovirus sequences (Mayo and Fritsch, 1994) suggests a consensus, which links viruses in the different genera, of E/D-XII_3j-

Y-xI3j-N-XI4_Sj-R. GFLV RNA is linked to VPg by a phospho diester bond between the

5' -phosphate and the -OH group of the N -terminal serine of VPg. Serine is present in all the VPg sequences at or near the N terminus and may be involved in the link between all nepovirus RNAs and their respective VPgs. The sequences of all nepovirus VPgs show the molecules to be basic and hydrophilic.

One role for VPg is suggested by the results of experiments in which nepovirus RNA was treated with proteases and assayed for infectivity. The infectivity of RNA of TRSV, TBRV, or ToRSV (Harrison and Barker, 1978; Mayo et a1., 1982) or LASV (Remah et a1., 1986) was abolished by the treatment, whereas that of ArMY, SLRSY, or RRSV RNA was decreased to some extent, but not abolished (Mayo et a1., 1982). In particular, the infectivity of RRSV RNA was rarely decreased more than twofold by protease treatment, although treatment with pronase caused a greater loss of infectivity than did treatment with proteinase K. Koenig and Fritsch (1982) reported that protease treatment decreased the size of TBRV VPg to about half (estimated to be about 16 amino acids) that of untreated VPg. It seems likely that the piece of RRSV VPg remaining after protease treatment is partly functional and that pronase removes more of the VPg molecule than does proteinase K. These results suggest that VPg enhances infectivity of nepovirus RNA, perhaps by protecting the 5' extremity of the RNA from nucleolytic degradation. This idea was supported by the finding that the relative infectivity of proteasetreated RRSV RNA could be enhanced by adding larger than normal amounts of bentonite to inocula (Barker and Mayo, 1982). Another proposal, made for picornaviruses but possibly of wider relevance, is that the VPg functions in


RNA replication as a primer for RNA synthesis (Takeda et a1., 1986); yet other roles, such as in particle assembly, have been suggested, tl;1ough with no convincing evidence. Protease treatment of either RNA-lor RNA-2 of TBRV abolished the infectivity of mixed RNA inocula in which the other component was untreated (Harrison and Barker, 1978) and any model of VPg function must be able to account for this observation.

4. NTP-Binding Protein

The regions of the nepovirus RNA-1 polyproteins that contain motifs characteristic of NTP-binding proteins are shown in Fig. 5c. These are relatively small parts of proteins of about 50 kDa (see Fig. 2). As with the proteases, the NTP-binding proteins of TBRV and GCMV are about 67% identical in a region slightly larger than that shown, whereas among other pairs of viruses the proteins had about 30 to 40% of identical residues.

This protein is thought to act as a helicase that "unwinds" the secondary structure of nucleic acids (Gorbalenya and Koonin, 1989), as demonstrated for the analogous protein of plum pox potyvirus (Lain et a1., 1990). Rott et a1. (1995) pointed out that a sequence feature common to NTP-binding proteins of several viruses, including ToRSV, is a stretch of hydrophobic residues near the N-terminus. These are also present in the NTP-binding proteins of GCMV (ILMIAAAIILVLV), TBRV (ILMIAAALILILV), and GFLV (LLLTLVAILLLISAAY), although the GFLV sequence is interrupted by Sand T. Sanfayon (1995) proposed that the NTP-binding protein acts as a membrane anchor protein, as apparently does the analogous protein of CPMV (Goldbach et a1., 1982). This hydrophobic domain may therefore serve this purpose.

5. Protease Cofactor

Rott et a1. (1995) have identified a consensus among nepoviruses within the N-terminal 50-kDa protein (Fig. 5d). Except for those of GCMV and TBRV, which are 54 % identical, the other polypeptides are between 20 and 26% identical. Several of the residues common to all four nepoviruses are also found in the N-terminaI32-kDa protein of CPMV RNA-I. This protein is thought to act as a protease cofactor (Vos et a1., 1988), and a similar role has been proposed for this part of the nepovirus polyprotein (Ritzenthaler et a1., 1991; Rott et a1., 1995).

6. N-Terminal Region of the Polyprotein

Rott et a1. (1995) have shown that a region near the N terminus of the RNA-1 polyprotein of ToRSV is very similar to a region near the N terminus of the RNA-2 polyprotein. In other nepoviruses whose genomes have been completely sequenced, there is no obvious similarity between the N termini of the RNA-1 and RNA-2 polyproteins. However, the N termini of the RNA-1 polyproteins of TBRV and GCMV (59% identical) and of ToRSV and TBRV


(23% identical) are detectably similar (Fig. 5). Comparisons between other pairs gave 13% identity scores. However, there is much more similarity among the sequences in amino acid content. In the region shown in Fig. 5e, all the sequences are alanine-rich (8 to 19 out of 45 residues) and are relatively basic (net charge = +4 to + 11). The region is presumably a domain within the N-terminal ca 50-kDa protein, which may have protease co-factor activity, but there is no evidence as to the function of this domain or why it should be duplicated in the RNA-l and RNA-2 polyproteins of ToRSV.

VII. REPLICATION

A. RNA Polymerase

Peden et al. (1972) detected RNA polymerase activities in both soluble and particulate fractions from TRSV-infected cucumber cotyledons. Although the enzymological properties of the extracts were those expected for an enzyme copying an RNA template, activity was neither dependent on added RNA nor specific for TRSV RNA. The product was apparently doublestranded, as judged by its resistance to pancreatic ribonuclease. The soluble activity was first detected 2 days after inoculation, reached a peak the next day, and then decreased rapidly. No activity was detected in either soluble or particulate fractions from uninfected plants. The TRSV-induced activities each differed in physical properties from otherwise similar activities induced by cucumber mosaic virus (CMV) infection. Thus, the soluble TRSV-induced activity sedimented in sucrose gradients as a broad peak with a size range of 120 to 180 kDa, compared with a sharp peak at 123 kDa for the CMV-induced soluble enzyme, and the particulate TRSV-induced activity was less readily solubilized than its CMV-induced analogue. These differences between enzymes induced by two different viruses provide circumstantial evidence that each is at least partly virus-codedj but the role, if any, of the TRSV-induced polymerase(s) in virus RNA replication is unknown.

Mayo and Barker (1983b) found that addition of actinomycin D to Nicotiana tabacum protoplasts immediately after inoculation with particles of TRSV resulted in a decrease in the proportion of protoplasts that became infected. This effect, which diminished when addition of the drug was delayed until 3 hr after inoculation with virus, suggests that transcription of the host genome is required at an early stage in virus replication, perhaps to provide a component of the replicase.

B. Intermediates of RNA Replication

The detection of high-molecular-weight double-stranded RNA species of the sizes expected for replicative forms of RNA-l and RNA-2 has been reported in extracts from plants infected with TRSV (Schneider et al., 1974),


RRSV (Jones et al., 1986), ToRSV (Kurppa and Martin, 1986), and GCMV (Le Gall et al., 1988). Thus, it seems likely that replication of nepovirus RNA involves the synthesis of full-length minus-strand RNA copies as intermediates.

Le Gall et al. (1988) studied the structures at the termini of the GCMV double-stranded RNA species. They were unable to label any of the 5' ends, and suggested that this indicated that VPg molecules were attached to both positive and negative strands. A poly(A) sequence was found at the 3' end of one strand, presumably the positive strand, of each species. The sequences at the 3' ends of the other strands were complementary to the sequences at the 5' ends of the respective genomic RNA species, except that there was no residue corresponding to the 5'-terminal U in either instance.

C. Recombination

Sequence comparisons among the predicted polyprotein translation products of RNA-2 of an English isolate (E) and a Scottish isolate (S) of TBRY, and of GCMV, suggest that recombination has played a part in the evolution of these viruses. The level of amino acid identity between the two TBRV isolates is relatively constant all along the sequence, whereas between TBRV-E and GCMV there is more identity in the MP domain than elsewhere. Comparisons of nucleotide residues at silent third positions of codons gave a similar pattern of identities, suggesting that there has been an exchange of regions of the RNA among virus isolates in this cluster (Le Gall et al., 1995a).

The existence of long regions of identical sequence at the 3' ends of RNA-1 and RNA-2 in both ToRSV and CLRV led Rott et al. (1991a) and Scott et al. (1992), respectively, to suggest that high-frequency recombination might be involved in maintaining the identity. Rott et al. (1991a) proposed a mechanism in which negative-strand RNA synthesis was initiated at the 3' end of RNA-l only, although it is not clear what feature of RNA-l could provide the specific polymerase recognition signal that would be required. The negative strand corresponding to RNA-2 would be synthesized after template switching at the junction of the identical and unique sequences. A similar mechanism, involving initiation of positive-strand RNA synthesis on only one of the negative-strand species, could account for conservation of the identical sequences at the 5' end.

Experimental evidence for recombinational exchange of 3'-NCRs was obtained by Le Gall et al. (1994, 1995b). After a pseudorecombinant possessing RNA-1 of GCMV and RNA-2 of TBRV had been passaged a few times in plants, both RNA species were found to have the 3'-NCR sequence of GCMV.

However, for BLMV, which has sequences at the 3' ends of RNA-l and RNA-2 that differ at only 4 of 1400 positions, Bacher et al. (1994a) were unable to detect recombinants in a virus population that had been main-


tained by serial mechanical passage for many years. In 45 cDNA clones that were partially sequenced, several mutations relative to the consensus sequence were observed, but none of the clones contained residues characteristic of both RNA species at the marker positions. They concluded that recombination, if it occurred at all, was a relatively rare event and certainly not frequent enough to account for maintenance of the 3'-terminal identity.

D. Sites of Virus Replication

Cells infected with nepoviruses characteristically contain cytoplasmic inclusion bodies that consist of irregular membranous structures, probably derived from the Golgi apparatus and/or the endoplasmic reticulum, together with granular material and ribosomes (Harrison and Murant, 1977a). These inclusion bodies were first observed by electron microscopy in ArMVinfected cells by Gerola et a1. (1965) and have subsequently been found in cells infected with many other members of the genus. It is often difficult to tell whether any of the particles that can be seen in the inclusion bodies are virus particles, but in some instances stacked linear arrays of viruslike particles occur in or at the periphery of the inclusion bodies, as, for example, with AYRSV (Russo et a1., 1978). The inclusion bodies of SLRSV contain masses of hollow structures resembling empty shells of virus coat protein and, in their outer parts, rows of virus particles enclosed in membranous tubules (Roberts and Harrison, 1970). Similar virus-containing tubules are characteristic of infections with many nepoviruses and are probably involved in cell-to-cell movement (see Section VI.A.2). In tobacco protoplasts infected with RRSV, virus antigen revealed by staining with fluorescent antibody was generally distributed throughout the cytoplasm but with a more intensely stained area, probably corresponding to the inclusion body (Barker and Harrison, 1977). Although none of these observations provides direct evidence that nepovirus inclusion bodies are the sites of virus replication, the similar structures in comovirus-infected cells have been shown to contain doublestranded virus RNA (De Zoeten et a1., 1974" presumably an intermediate in the replication process. Thus there is strong circumstantial evidence that nepovirus replication too takes place in the inclusion bodies.

E. Control of Replication

There is some evidence for differential control of the replication of RNA-1 and RNA-2. In Nicotiana benthamiana infected with RRSY, the relative proportions of B to M components or of RNA-1 to RNA-2 were greater in systemically infected leaves showing few symptoms (recovered leaves) than in similar leaves showing more marked symptoms. There was, however, little change in the proportion of RNA-1 to RNA-2 during the


course of virus multiplication in a particular symptom-bearing leaf (Barker, 1980). In N. c1evelandii protoplasts, accumulation of M and B components of RRSV decreased markedly or ceased at about 60 hr after infection, although accumulation of T component continued up to at least 91 hr after infection. In contrast, in N. tabacum protoplasts, accumulation of all three components continued undiminished up to at least 73 hr after infection, and there was little change in the ratio of T to B components during the course of the infection (Acosta and Mayo, 1990a).

VIII. THE IMPACT OF MOLECULAR BIOLOGY ON NEPOVIRUS CLASSIFICATION

Nepoviruses have been subdivided in various ways, but the simplest scheme, used in Table I, follows that proposed by Murant (1981). Definitive and tentative nepoviruses are separated by the number and size of their coat proteins, and the definitive nepoviruses are divided into three clusters according to (1) the size of RNA-2 and (2) serological relationships. Neither the serological clustering nor comparisons among the sequences of the coat proteins conflict with the clustering on the basis of RNA-2 size.

As might be expected from their serological relationship, the coat protein sequences of ArMV and GFLV are relatively similar, as are those of TBRV and GCMV (Section VI.A.I). The other sequences available are too divergent for relationships among them to be detected and so contribute little information of taxonomic value. However, comparisons between coat protein sequences of GCMV and different serotypes of TBRV do raise some taxonomic questions. The coat protein sequence shown for TBRV in Fig. 3 is that of an isolate of the Scottish serotype (S), but this sequence is only about 62% identical to that of the coat protein of an English (E) serotype isolate (Le Gall et a1., 1995a). Since the coat proteins of each serotype of TBRV are about 52 % (E) or 56% (S) identical in sequence to that of GCMY, the two serotypes are almost as different from each other as they are from GCMY. These results suggest that the serotypes of TBRV might be assigned the status of virus species, although the strong similarity between the MPs of GCMV and TBRV-E (Section VILe) might suggest otherwise. The relatively close relationship among these three viruses was also suggested by the formation (in some combinations) of pseudorecombinants between viruses from the two TBRV serotypes (Randles et a1., 1977 j Mayo and Barker, 1983a), and between GCMV and TBRV-S (Doz et a1., 1980 j C. Oncino, personal communication) (see Section V.A).

The RNA-2 sequences of the cluster c (Table 1) viruses so far examined all contain unusually long (> 1300 nt) 3' NCRs (see Table II). This may prove to be a feature of these virusesj but the coding sequence, at least of ToRSV RNA-2, is also longer than those in RNA-2 of cluster a or cluster b viruses (Table I).


Some viruses, including SLRSV, which was one of the original group of six viruses called "nepoviruses" by Cadman (1963), are classified as "tentative" nepoviruses because they have more than one type of coat protein molecule (clusters d and e in Table I) or because the coat protein composition is unknown (cluster f). The nucleotide sequence of RNA-2 of SLRSV and the sequences of the encoded proteins further emphasize the difference between SLRSV and definitive nepoviruses. Figure 2 shows that the arrangement of the genes for the two coat protein molecules resembles that in comovirus RNA. However, although the larger coat proteins of four comoviruses are 35% identical and the smaller proteins are 22% identical (Chen and Bruening, 1992a), the sequences of SLRSV coat proteins have no detectable homology with the corresponding comovirus proteins (Everett et a1., 1994). Nevertheless, predictions of the secondary structures of SLRSV coat proteins have suggested a similarity with the coat proteins of CPMV (Le Gall, personal communication). Thus, the classification of SLRSV and perhaps other viruses in cluster d as nepoviruses or comoviruses, or indeed fabaviruses, may need to be reexamined. Cluster e nepoviruses have particles that appear to contain three types of coat protein. These viruses perhaps should be distinguished from nepoviruses, comoviruses and fabaviruses, but until some sequence data are available little can be decided.

IX. SATELLITES OF NEPOVIRUSES

Satellites are defined as "a type of sub-viral agent which comprises nucleic acid molecules that depend for their productive multiplication on coinfection of a host cell with a helper virus." Furthermore, "satellite nucleic acids contain substantial nucleotide sequence distinct from that in the genomes of either helper virus or host" (Mayo et a1., 1995, p. 487). Nepoviruses have figured prominently in the literature on satellites, not least because satellites of nepoviruses were the second (Schneider, 1969) and third (Murant et a1., 1973) examples of plant virus satellites to be found. Indeed, of the 24 plant viruses known to have satellites, eight are nepoviruses. Possibly this is because nepovirus coat proteins readily form empty protein shells, in which satellite RNA can be encapsidated without affecting the packaging of virus RNA. However, most nepoviruses do not have satellite RNA and none is known for comoviruses, which also form empty T-component particles.

Some satellites form particles containing satellite-coded protein, but most, including those of nepoviruses, are encapsidated in the coat protein of their helper viruses. Nepovirus satellites are of two types: type B mRNA satellites or type D circular RNA satellites (Mayo et a1., 1995). Table IV lists these satellites. There is no apparent pattern about which viruses support which type of satellitej some isolates of ArMV and Ch YMV support type B satellites, whereas other isolates of either virus support type D satellites.


TABLE IV. Satellites of Nepoviruses

Database Encoded accession polypeptide

Nepovirus number Size (nt) size (kDa)a Reference

B type satellites Arabis mosaic D00664 1104 39 Liu et al. (1990) Chicory yellow mottle D00686 1145 39 Rubino et al. (1990) Grapevine Bulgarian latent -1800 ND Gallitelli et al. (1983) Grapevine fanleaf D00442 1114 37 Fuchs et al. (1989) Myrobalan latent ringspot -l300 (-45) Fritsch et al. (1984) Strawberry latent ringspot X69826 1118 36 Kreiah et al. (1993) Tomato black ring X00978 1375 48 Hemmer et al. (1987)

D type satellites Arabis mosaic M21212 300 b Kaper et al. (1988) Chicory yellow mottle D00685 457 b Rubino et al. (1990) Tobacco ringspot M14879 359 _b Buzayan et al. (1986)

aValues calculated from nucleotide sequences except for that in parentheses which was established from the electrophoretic mobility of the in vitro translation product of the RNA; ND, not determined.

bNo translation product.

A. Type B Satellites

1. Satellite-Encoded Protein

Type B satellite RNA molecules act as mRNA for proteins of between 39 and 48 kDa (Table IV). These satellite RNA molecules resemble the genomic RNA molecules of their helper viruses in being linked to the helper-coded VPg at the 5' end and in having a 3' poly(A) tail. However, there is little if any sequence homology between the terminal NCRs of satellite and helper virus RNA molecules (Fritsch and Mayo, 1989).

The ORF in satellite RNA of TBRV (Hemmer et a1., 1993) and GFLV (Hans et a1., 1993) must be intact for the satellite RNA to be replicated. The satellite molecules act efficiently as messenger RNA in in vitro translation (Fritsch et a1., 1984). However, there is no evidence as to the function(s) of the satellite-coded proteins (Fritsch et a1., 1993). The satellite proteins all have positively charged domains at each end of the molecule and have net positive charges of between 24 and 34 (Fritsch et a1., 1993). There are slight sequence similarities between parts of the satellite proteins of TBRY, GFLV, ArMV, and ChYMV (Fritsch et a1., 1993).

2. Effects on Symptoms

The presence of type B satellites in cultures of nepoviruses usually does not affect the symptoms induced by the viruses (Fritsch et a1., 1993). Only the ArMV B-type satellite exacerbated the symptoms of helper virus infection and then only in 3 of the 42 hosts of ArMV tested; in 10 of the 42 hosts,


symptoms were less severe when satellite was present (Liu et a1., 1991). Neither do type B satellites much modify the multiplication of their helper viruses (Fritsch et a1., 1993). Thus they appear to be ideally adapted to their "parasitic" mode of existence.

Type B satellite RNA molecules are packaged in virus protein shells individually or in multiples or sometimes together with a molecule of RNA-2. Virus particles of cultures of nepoviruses that contain these satellites are therefore often heterogeneous in sedimentation rate and in buoyant density (Murant and Mayo, 1982) (see Sections II.C and II.D). TBRV satellite RNA (Murant and Mayo, 1982) and ArMV satellite RNA (Liu et a1., 1991) are transmitted by nematodes, presumably as a result of being packaged in TBRV coat protein. TBRV satellite RNA is transmitted together with TBRV through the seed of infected hosts (Hanada and Harrison, 1977).

3. Sequence Similarities among Satellites

Satellite RNA can occur in cultures of TBRV of either of the two serotypes. Satellites of isolates Land S (Scottish serotype) encode proteins that are about 90% identical in sequence, as do satellites of isolates G and C (German serotype), but the maximum identity between satellite proteins of TBRV isolates of different serotypes is about 60%. This suggests that satellites of TBRV have diverged together with the strains of the helper virus. The satellites of ArMV and GFLV encode proteins that are 84% identical, whereas the coat proteins of the two viruses are 69% identical (Fritsch et a1., 1993).

4. Specificity of Helper-Satellite Interaction

Satellites of TBRV multiply only when the helper virus is from the same serotype as the isolate from which the satellite was obtained (Murant and Mayo, 1982). However, a pseudorecombinant comprising RNA-l from an isolate of the G serotype and RNA-2 of an isolate of the S serotype supported the multiplication of a satellite of the G isolate but not of a satellite of the S isolate (Murant and Raschke, 1982). This suggests that the specificity for supporting satellite multiplication depends on a function encoded by RNA-I. A pseudorecombinant comprising RNA-l of GCMV and RNA-2 of TBRV-S was able to support the multiplication of a satellite of TBRV-S in experiments made using transcript satellite RNA from cloned cDNA (c. Oncino, personal communication) although apparently not when satellite RNA extracted from virus particles was used (Doz et a1., 1980).

A satellite of ArMV multiplied only with ArMV strains that were closely related serologically to the lilac strain from which the satellite originated. In contrast, a satellite of GFLV did multiply when inoculated together with ArMV even though ArMV is only distantly related to GFLV (Hans et a1., 1992).


B. Type D Satellites

These satellites are less than 500 nt long and occur as circular as well as linear molecules. They also differ from the RNA of their helper viruses in having neither a VPg nor a poly(A) tract and in not being mRNA. Nepovirus satellites of this type are encapsidated as linear molecules (which may be multimeric), but circular forms are present in infected cells. Three of the eight type D satellites listed by Mayo et a1. (1995) are associated with nepoviruses. The satellites differ in length (Table IV), but their sequences contain regions of strong similarity. The first 60 nt at the 5' ends are about 75% identical, and this region, together with the 6 nt at the 3' end, forms a "hammerhead" structure involved in the cleavage of multimeric satellite RNA. Another region of strong homology among all three satellites is in the center of the molecules (Rubino et a1., 1990).

Type D satellites replicate by a "rolling circle" mechanism in which polymerase copies the circular template to make a multimeric, linear product. This RNA is cleaved in a self-catalyzed reaction to yield linear mono-

transcription

8 • ( -)

r ligation cleavage 1 (+) (-)

r cleavage

ligation

1

(+)

transcription

FIGURE 7. The symmetrical model for the replication of D-type satellite RNA. (After Bruening et a1., 1991.)


meric forms. The circular template may be of either polarity and the linear monomers ligate to form circular template molecules, so that the model is symmetrical (Bruening et a1., 1991) (Fig. 7).

Different structures are involved in the self-cleavage of (+)- and (-)sense strands. A "hammerhead" structure is involved in the cleavage of ( + )sense RNA, but cleavage of (- )-sense RNA occurs at a different place and involves a different structure (Bruening et a1., 1991). Both reactions involve autocatalytic cleavage of phosphodiester bonds rather than an enzymatic hydrolysis and do not involve protein.

In general, type D satellites have more effect on helper virus symptoms than do type B satellites. TRSV satellite RNA greatly attenuates the symptoms of TRSV infection, presumably because the satellite competes with virus RNA for replication. Plants transformed to express TRSV satellite RNA were protected against challenge inoculation with TRSV (Gerlach et a1., 1987). In contrast, in hop plants infected with ArMY, the severe symptom of nettlehead disease is associated with the presence of a type D satellite RNA (Davies and Clark, 1983).

ACKNOWLEDGMENTS. We thank J. 1. Cooper, C. Fritsch, O. Le Gall, C. Oncino, D. M. Rochon, M. E. Rott, and H. Sanfayon for access to unpublished information, and the Scottish Office Agriculture and Fisheries Department for financial support.

REFERENCES

Acosta, 0., and Mayo, M. A., 1990a, Unusual electrophoretic properties of the coat protein of raspberry ringspot nepovirus, Interviralogy 31:31.

Acosta, 0., and Mayo, M. A., 1990b, Accumulation of different types of raspberry ringspot nepovirus particle in infected Nicotiana protoplasts, 1. Cen. Viral. 71:713.

Acosta, 0., and Mayo, M. A., 1993, Polypeptide synthesis induced in Nicotiana clevelandii protoplasts by infection with raspberry ringspot virus, Arch. Viral. 19:145.

Agrawal, H. 0., 1967, The morphology of arabis mosaic virus, 1. Ultras tru ct. Res. 17:84. Ahlquist, P., Dasgupta, R., and Kaesberg, P., 1981, Near identity of 3' RNA secondary structure in

bromoviruses and cucumber mosaic virus, Cell 23:183. Bacher, J. W., Warkentin, D., Ramsdell, D., and Hancock, J. F., 1994a, Selection versus recom

bination: What is maintaining identity in the 3' termini of blueberry leaf mottle nepovirus RNAI and RNA2? 1. Cen. Vira1. 75:2133.

Bacher, J. W., Warkentin, D., Ramsdell, D., and Hancock, J. F., 1994b, Sequence analysis of the 3' termini of RNAI and RNA2 of blueberry leaf mottle virus, Virus Res. 33:145.

Bancroft, J. B., 1968, Tomato top necrosis virus, Phytopathology 58:1360. Barbara D. J., Ashby, S. c., and McNamara, D. G., 1985, Host range, purification and some

properties of rubus Chinese seed-borne virus, Ann. AppJ. BioI. 107:45. Barker, H., 1980, Studies on the Behaviour of Raspberry Ringspot Virus in Plant Leaves and

Pratoplasts, PhD. Thesis, University of Dundee. Barker, H., and Harrison, B. D., 1977, Infection of tobacco mesophyll protoplasts with raspberry

ringspot virus alone and together with tobacco rattle virus, 1. Cen. Viral. 35:125. Barker, H., and Mayo, M. A., 1982, A possible role for the genome-linked protein of raspberry

ringspot virus, Report of the Scottish Crap Research Institute, 1981, p. 109.


Barnett, O. W., Reddick, B. B., Burrows, P. M., and Baxter, L. W., 1989, Characterization of dogwood mosaic nepovirus from Comus florida, Phytopathology 79:95l.

Bazan, J. F., and Fletterick, R. L 1988, Viral cysteine prot eases are homologous to the trypsin-like family of serine proteases: Structural and functional implications, Proc. Natl. Acad. Sci. USA 85:7872-

Bellardi, M. G., and Gelli, c., 1984, Infezione del virus della maculatura anulare latente della fragola (SLRV) nel narciso in Italia, Informatore Fitopatol. 34:49.

Bertioli, D. L Harris, R. D., Edwards, M. L., Cooper, J. I., and Hawes, W. S., 1991, Transgenic plants and insect cells expressing the coat protein of arabis mosaic virus produce empty virus-like particles, J. Gen. Virol. 72:1I:iOl.

Blok, V. c., Wardell, L Jolly, C. A., Manoukian, A., Robinson, D. L Edwards, M. L., and Mayo, M. A., 1992, The nucleotide sequence of RNA-2 of raspberry ringspot nepovirus, J. Gen. Virol. 73:2189.

Boatman, S., and Kaper, J. M., 1976, Molecular organization and stabilizing forces of simple RNA viruses. IV. Selective interference with protein-RNA interactions by use of sodium dodecyl sulfate, Virology 70:l.

Brault, v., Hibrand, L., Candresse, T., Le Gall, 0., and Dunez, J., 1989, Nucleotide s~quence and genetic organization of Hungarian grapevine chrome mosaic nepovirus RNA2, Nucleic Acids Res. 17:7809.

Bruening, G., Passmore, B. K., Van Tol, H., Buzayan, J. M., and Feldstein, P. A., 1991, Replication of a plant virus satellite RNA: Evidence favors transcription of circular templates of both polarities, Mol. Plant-Microbe Interactions 4:219.

Buckley, B., Silva, S., and Singh, S., 1993, Nucleotide sequence and in vitro expression of the capsid protein gene of tobacco ringspot virus, Virus Res. 30:335.

Buzayan, J. M., Gerlach, W. L., Bruening, G., Keese, P., and Gould, A. R., 1986, Nucleotide sequence of satellite tobacco ringspot virus RNA and its relationship to multimeric forms, Virology 151:186.

Cadman, C. H., 1963, Biology of soil-borne viruses, Annu. Rev. Phytopathol. 1:143. Chambers, T. c., Francki, R. 1. B., and Randles, J. W., 1965, The fine structure of Gladiolus virus,

Virology 25:15. Chen, X., and Bruening, G., 1992a, Nucleotide sequence and genetic map of cowpea severe

mosaic virus RNA 2 and comparisons with RNA 2 of other comoviruses, Virology 187:682. Chen, X., and Bruening, G., 1992b, Cloned DNA copies of cowpea severe mosaic virus genome

RNA: Infectious transcripts and complete nucleotide sequence of RNA-I, Virology 191:607. Chu, P. W. G., and Francki, R. 1. B., 1979, The chemical subunit of tobacco ringspot virus coat

protein, Virology 93:398. Chu, P. W. G., Boccardo, G., and Francki, R. 1. B., 1981, Requirement of a genome-associated

protein of tobacco ringspot virus for infectivity but not for in vitro translation, Virology 109:428.

Clark, M. F., 1976, Electrophoretic heterogeneity of the sedimenting components of arabis mosaic virus, J. Gen. Virol. 32:33l.

Daubert, S. D., and Bruening, G., 1985, Detection of genome-linked proteins of plant and animal viruses, in: Methods in Virology, Vol. 8 (K. Maramorosch and H. Koprowski, eds.), pp. 347-379, Academic Press, New York.

Davies, D. L., and Clark, M. F., 1983, A satellite-like nucleic acid of arabis mosaic virus associated with hop nettlehead disease, Ann. Appl. Biol. 103:439.

Davison, E. M., and Francki, R. 1. B., 1969, Observations on negatively stained tobacco ringspot virus and its two RNA-deficient components, Virology 39:235.

Debrot, E. A., 1964, Studies on a strain of raspberry ringspot virus occurring in England, Ann. Appl. Bioi. 54:183.

Demangeat, G., Greif, C., Hemmer, 0., and Fritsch, C., 1990, Analysis of the in vitro cleavage products of the tomato black ring virus RNA-I-encoded 250K polyprotein, J. Gen. Vira1. 71:1649.

Demangeat, G., Hemmer, 0., Fritsch, C., Le Gall, 0., and Candresse, T., 1991, In vitro processing


of the RNA-2-encoded polyprotein of two nepoviruses: Tomato black ring virus and grapevine chrome mosaic virus, r. Gen. Virol. 72:247.

Demangeat, G., Hemmer, 0., Reinbolt, y., Mayo, M. A., and Fritsch, c., 1992, Virus-specific proteins in cells infected with tomato black ring nepovirus: Evidence for proteolytic processing in vivo, r. Gen. Viral. 73:1609.

Devereux, y., Haeberli, P., and Smithies, 0., 1984, A comprehensive set of sequence analysis programs for the VAX, Nucleic Acids Res. 12:387.

De Zoeten, G. A., Assink, A. M., and van Kammen, A., 1974, Association of cowpea mosaic virus-induced double-stranded RNA with a cytopathological structure in infected cells, Virology 59:341.

Diener, T. 0., and Schneider, I. R., 1966, The two components of tobacco ringspot virus nucleic acid: Origin and properties, Virology 29:100.

Dodd, S. M., and Robinson, D. y., 1984, Nucleotide sequence homologies among RNA species of strains of tomato black ring virus and other nepoviruses, r. Gen. Viral. 65:1731.

Dougherty, W. G., and Hiebert, E., 1980, Translation of potyvirus RNA in a rabbit reticulocyte lysate: Reaction conditions and identification of capsid protein as one of the products of in vitro translation of tobacco etch and pepper mottle viral RNAs, Virology 101:466.

Doz, B., Macquaire, G., Delbos, R., and Dunez, y., 1980, Caracteristiques et role du RNA 3, RNA satellite du virus des anneaux noirs de la tomate, Ann. Virol. {Institut Pasteur) 131E:489.

Dunez, y., Delbos, R., and Dupont, G., 1976, Myrobalan latent ringspot virus, CMI/AAB Descriptions of Plant Viruses No. 160.

Everett, K. R., Milne, K. S., and Forster, R. 1. S., 1994, Nucleotide sequence of the coat protein genes of strawberry latent ringspot virus: Lack of homology to the nepoviruses and comoviruses, r. Gen. Virol. 75:1821.

Forster, R. 1. S., and Morris-Krsinich, B. A. M., 1985, Synthesis and processing of the translation products of tobacco ringspot virus in rabbit reticulocyte lysates, Virology 144:516.

Francki, R. I. B., Milne, R. G., and Hatta, T., 1985, Atlas of Plant Viruses, Vol. II, pp. 23-37, CRC Press, Boca Raton, F1.

Fritsch, C., and Mayo, M. A., 1989, Satellites of plant viruses, in: Plant Viruses, Vol. I. Structure and Replication (c. 1. Mandahar, ed.), pp. 289-321, CRC Press, Boca Raton, F1.

Fritsch, c., Koenig, I., Murant, A. F., Raschke, y. H., and Mayo, M. A., 1984, Comparisons among satellite RNA species from five isolates of tomato black ring virus and one isolate of myrobalan latent ringspot virus, r. Gen. Viral. 65:289.

Fritsch, C., Mayo, M. A., and Hemmer, 0., 1993, Properties of the satellite RNA of nepoviruses, Biochimie 75:561.

Fuchs, M., Pinck, M., Serghini, M. A., Ravelonandro, M., Walter, B., and Pinck, 1., 1989, The nucleotide sequence of satellite RNA in grapevine fanleaf virus, strain Fl3, r. Gen. Virol. 70:955.

Fuchs, M., Pinck, M., Etienne, 1., Pinck, 1., and Walter, B., 1991, Characterization and detection of grapevine fanleaf virus by using cDNA probes, Phytopathology 81:559.

Gallitelli, D., Savino, v., and De Sequeira, O. A., 1983, Properties of a distinctive strain of grapevine Bulgarian latent virus, Phytopathol. Mediterr. 22:27.

Gerlach, W. 1., Llewellyn, D., and Haseloff, J" 1987, Construction of a plant disease resistance gene from the satellite RNA of tobacco ringspot virus, Nature 328:802.

Gerola, F. M., Bassi, M., and Betto, E., 1965, Some observations on the shape and localization of different viruses in experimentally infected plants, and on the fine structure of the host cells. I. Arabis mosaic virus in Chenopodium am aranticolor, Caryologia 18:353.

Goldbach, R. W., Schilthuis, J, G., and Rezelman, G., 1981, Comparison of in vivo and in vitro translation of cowpea mosaic virus RNAs, Biochem. Biophys. Res. Commun. 99:89.

Goldbach, R., Rezelman, G., Zabel, P., and Van Kammen, A., 1982, Expression of the bottomcomponent RNA of cowpea mosaic virus. Evidence that the 60-kilodalton VPg precursor is cleaved into single VPg and a 58-kilodalton polypeptide, r. Virol. 42:630.

Goldbach, R., Martelli, G. P., and Milne, R. G., 1995, Family Com ovirida e, in: Virus Taxonomy-Classification and Nomenclature of Viruses: Sixth Report of the International


Committee on Taxonomy of Viruses (F. A. Murphy, C. M. Fauquet, D. H. L. Bishop, S. A. Ghabrial, A. W. Jarvis, G. P. Martelli, M. A. Mayo and M. D. Summers, eds.l, pp. 341-347, Springer-Verlag, Vienna (also in Arch. Viral. Suppl. 101.

Gooding, G. v., 1963, Purification and serology of a virus associated with the grape yellow vein disease, Phytopathology 53:475.

Gorbalenya, A. E., and Koonin, E. v., 1989, Viral proteins containing the purine NTP-binding sequence pattern, Nucleic Acids Res. 17:8413.

Gorbalenya, A. E., Donchenko, A. P., Blinov, V. M., and Koonin, E. v., 1989, Cysteine proteases of positive strand RNA viruses and chymotrypsin-like serine proteases. A distinct protein superfamily with a common structural fold, FEBS Lett. 243:103.

Greif, C., Hemmer, 0., and Fritsch, C., 1988, Nucleotide sequence of tomato black ring virus RNA-I, r Gen. Viral. 69:1517.

Haber, S., and Hamilton, R. I., 1980, Distribution of determinants for symptom production, nucleoprotein component distribution and antigenicity of coat protein between the two RNA components of cherry leaf roll virus, J. Gen. Viral. 50:377.

Hadidi, A., Piazzolla, P., Savino, V., and Clark, M. F., 1992, Arabis mosaic virus: Detection by SP6-generated cRNA probes and possible recombination with strawberry latent ringspot virus, Acta Hortic. 308:49.

Hanada, K., and Harrison, B. D., 1977, Effects of virus genotype and temperature on seed transmission of nepoviruses, Ann. Appl. BioI. 85:79.

Hans, F., Fuchs, M., and Pinck, L., 1992, Replication of grapevine fanleaf virus satellite RNA transcripts in Chenopodium quinoa protoplasts, J. Gen. Viral. 73:2517.

Hans, F., Pinck, M., and Pinck, L., 1993, Location of the replication determinants of the satellite RNA associated with grapevine fanleaf nepovirus (strain F131, Biochimie 75:597.

Harrison, B. D., and Barker, H., 1978, Protease-sensitive structure needed for infectivity of nepovirus RNA, r Gen. Viral. 40:711.

Harrison, B. D., and Murant, A. F., 1977a, Nepovirus group, CMI/AAB Descriptions of Plant Viruses No. 185.

Harrison, B. D., and Murant, A. F., 1977b, Nematode transmissibility of pseudo-recombinant isolates of tomato black ring virus, Ann. Appl. Biol. 86:209.

Harrison, B. D., Murant, A. F., and Mayo, M. A., 1972, Two properties of raspberry ringspot virus determined by its smaller RNA, J. Gen. Viral. 17:137.

Harrison, B. D., Murant, A. F., Mayo, M. A., and Roberts, I. M., 1974, Distribution of determinants for symptom production, host range and nematode transmissibility between the two RNA components of raspberry ringspot virus, J. Gen. Viral. 22:233.

Hellen, C. U. T., and Cooper, J. I., 1987, The genome-linked protein of cherry leaf roll virus, J. Gen. Viral. 68:2913.

Hellen, C. U. T., Yuanyi, L., and Cooper, J. I., 1991, Synthesis and proteolytic processing of arabis mosaic nepovirus, cherry leaf roll nepovirus and strawberry latent ringspot nepovirus proteins in reticulocyte lysate, Arch. Viral. 120:19.

Hemmer, 0., Meyer, M., Greif, c., and Fritsch, c., 1987, Comparison of the nucleotide sequences of five tomato black ring virus satellite RNAs, r Gen. Viral. 68:1823.

Hemmer, 0., Oncino, c., and Fritsch, C., 1993, Efficient replication of the in vitra transcripts from cloned cDNA of tomato black ring virus satellite RNA requires the 48K satellite RNAencoded protein, Viralogy 194:800.

Hemmer, 0., Greif, c., Duforcq, P., Reinbolt, J., and Fritsch, c., 1995, Functional characterization of the proteolytic activity of the tomato black ring virus RNA-I-encoded 120 K protein, Viralogy 206:362.

Heuss, K. L., Murthy, M. R. N., and Argos, P., 1981, Crystallization of tobacco ringspot virus, J. Mol. Biol. 153:1161.

Hewitt, W. B., Martelli, G., Dias, H. F., and Taylor, R. H., 1970, Grapevine fanleaf virus, CMII AAB Descriptions of Plant Viruses No. 28.

Hibi, T., 1986, Physical and chemical properties of several plant viruses, in: Praceedings of the


International Symposium on Virus Diseases of Rice and Leguminous Craps in the Tropics, Tsukuba, 1985, p. 86.

Hibrand, L., Le Gall, 0., Candresse, T., and Dunez, J., 1992, Immunodetection of the proteins encoded by grapevine chrome mosaic nepovirus RNA2, f. Gen. Viral. 73:2093.

Higgins, D. G., Bleasby, A. J., and Fuchs, R., 1992, CLUSTALV: Improved software for multiple sequence alignment, CABIOS 8:189.

Hull, R., Milne, R. G., and Van Regenmortel, M. H. v., 1991, A list of proposed standard acronyms for plant viruses and viroids, Arch. Viral. 120:151.

Jones, A. T., 1985, Cherry leaf roll virus, CMI/AAB Descriptions of Plant Viruses No. 306. Jones, A. T., and Duncan, G. H., 1980, The distribution of some genetic determinants in the two

nucleoprotein particles of cherry leaf roll virus, f. Gen. Viral. 50:269. Jones, A. T., Forster, R. L. S., and Mohamed, N. A., 1979, Purification and properties of Australian

lucerne latent virus, a seed-borne virus having affinities with nepoviruses, Ann. Appl. BioI. 92:49.

Jones, A. T., Mayo, M. A., and Henderson, S. J., 1985, Biological and biochemical properties of an isolate of cherry rasp leaf virus from red raspberry, Ann. Appl. BioI. 106:101.

Jones, A. T., Abo El-Nasr, M. A., Mayo, M. A., and Mitchell, M. J., 1986, Association of dsRNA species with some virus-like diseases of small fruits, Acta Hortie. 186:63.

Jones, A. T., Mitchell, M. J., and Brown, D. J. F., 1989, Infectibility of some new raspberry cultivars with arabis mosaic and raspberry ringspot viruses and further evidence for variation in British isolates of these two nepoviruses, Ann. Appl. BioI. 115:57.

Kaper, J. M., 1973, Arrangement and identification of simple isometric viruses according to their dominating stabilizing interactions, Virology 55:299.

Kaper, J. M., Tousignant, M. E., and Steger, G., 1988, Nucleotide sequence predicts circularity and self-cleavage of 300-ribonucleotide satellite of arabis mosaic virus, Biochem. Biophys. Res. Commun. 154:318.

Kenten, R. H., and Jones, R. A. C., 1979, Arracacha virus B, a second isometric virus infecting arracacha (Arracacia xanthorrhiza; UmbelliferaeJ in the Peruvian Andes, Ann. Appl. BioI. 93:31.

Koenig, 1., and Fritsch, c., 1982, A protein linked at the 5' end of satellite and genomic tomato black ring virus RNAs: Study of in vitra translation after protease treatment, f. Gen. Viral. 60:343.

Koonin, E. V., and Dolja, V. v., 1993, Evolution and taxonomy of positive-strand RNA viruses: Implications of comparative analysis of amino acid sequences, Crit. Rev. Biochem. Mol. BioI. 28:375.

Koonin, E. v., Mushegian, A. R., Ryabov, E. v., and Dolja, V. v., 1991, Diverse groups of plant RNA and DNA viruses share related movement proteins that may possess chaperone-like activity, f. Gen. Viral. 72:2895.

Kreiah, S., Cooper, J. 1., and Strunk, G., 1993, The nucleotide sequence of a satellite RNA associated with strawberry latent ringspot virus, f. Gen. Viral. 74:1163.

Kreiah, S., Strunk, G., and Cooper, J. 1., 1994, Sequence analysis and location of capsid proteins within RNA 2 of strawberry latent ringspot virus, f. Gen. Viral. 75:2527.

Kurppa, A., and Martin, R. R., 1986, Use of double-stranded RNA for detection and identification of virus diseases of Rubus species, Acta Hortie. 186:51.

Ladipo, J. L., and De Zoeten, G. A., 1972, Influence of host and seasonal variation on the components of tobacco ringspot virus, Phytopathology 62:195.

Lain, S., Riechmann, J. L., and Garcia, J. A., 1990, RNA helicase: A novel activity associated with a protein encoded by a positive strand RNA virus, Nucleic Acids Res. 18:7003.

Le Gall, 0., Candresse, T., and Dunez, J., 1988, Nucleotide sequence of the 3' ends of the doublestranded RNAs of grapevine chrome mosaic nepovirus, f. Gen. Viral. 69:423.

Le Gall, 0., Candresse, T., Brault, v., and Dunez, J., 1989, Nucleotide sequence of Hungarian grapevine chrome mosaic nepovirus RNAl, Nucleic Acids Res. 17:7795.

Le Gall, 0., Candresse, T., and Dunez, J., 1994, RNA recombination is probably a common


phenomenon in nepoviruses, in: Proceedings of Workshop on Genetic Recombination and Defective Interfering Particles in RNA Viruses, Madrid, 1994, p. 70, Instituto Juan March de Estudio e Investigaciones, Madrid.

Le Gall, 0., Lanneau, M., Candresse, T., and Dunez, L 1995a, The nucleotide sequence of the RNA-2 of an isolate of the English serotype of tomato black ring virus: RNA recombination in the history of nepoviruses, r. Gen. Virol. 76:1279.

Le Gall, 0., Candresse, T., and Dunez, J., 1995b, Transfer of the 3' non-translated region of grapevine chrome mosaic virus RNA-1 by recombination to tomato black ring virus RNA-2 in pseudorecombinant isolates, r. Gen. Virol. 76:1285.

Liu, Y. Y., Hellen, C. U. T., Cooper, J. I., Bertioli, D. L Coates, D., and Bauer, G., 1990, The nucleotide sequence of a satellite RNA associated with arabis mosaic nepovirus, r. Gen. Virol. 71:1259.

Liu, Y. Y., Cooper, J. I., Edwards, M. 1., and Hellen, C. U. T., 1991, A satellite RNA of arabis mosaic nepovirus and its pathological impact, Ann. Appi. Bioi. 118:577.

Margis, R., and Pinck, 1., 1992, Effects of site-directed mutagenesis on the presumed catalytic triad and substrate-binding pocket of grapevine fanleaf nepovirus 24-kDa proteinase, Virology 190:884.

Margis, R., Viry, M., Pinck, M., and Pinck, 1., 1991, Cloning and in vitro characterization of the grapevine fanleaf virus proteinase cistron, Virology 185:779.

Margis, R., Ritzenthaler, c., Reinbolt, J., Pinck, M., and Pinck, 1., 1993, Genome organization of grapevine fanleaf nepovirus RNA 2 deduced from the 122K polyprotein P2 in vitro cleavage products, J. Gen. Virol. 74:1919.

Margis, R., Viry, M., Pinck, M., Bardonnet, N., and Pinck, 1., 1994, Differential proteolytic activities of precursor and mature forms of the 24 K proteinase of grapevine fanleaf nepovirus, Virology 200:79.

Martelli, G. P., and Taylor, C. E., 1989, Distribution of viruses and their nematode vectors, Adv. Dis. Vector Res. 6:15l.

Martelli, G. P., Gallitelli, D., Abracheva, P., Savino, v., and Quacquarelli, A., 1977, Some properties of grapevine Bulgarian latent virus, Ann. Appl. Bioi. 85:51.

Mayo, M. A., 1994, Nepoviruses, in: Encyclopedia of Virology (R. G. Webster and A. Granoff, eds.), pp. 901-907, Academic Press, London.

Mayo, M. A., and Barker, H., 1983a, Defective multiplication of a pseudorecombinant of tomato black ring virus in tobacco protoplasts, r. Gen. Viral. 64:649.

Mayo, M. A., and Barker, H., 1983b, Effects of actinomycin D on the infection of tobacco protoplasts by four viruses, J. Gen. Viral. 64:1775.

Mayo, M. A., and Fritsch, C., 1994, A possible consensus sequence for VPg of viruses in the family Comoviridae, FEBS Lett. 354:129.

Mayo, M. A., and Martelli, G. P., 1993, New families and genera of plant viruses, Arch. Viral. 133:496. Mayo, M. A., Murant, A. F., and Harrison, B. D., 1971, New evidence on the structure of

nepoviruses, r. Gen. Virol. 12:175. Mayo, M. A., Harrison, B. D., Murant, A. F., and Barker, H., 1973, Cross-linking of RNA induced

by ultraviolet irradiation of particles of raspberry ringspot virus, r. Gen. Viral. 19:155. Mayo, M. A., Murant, A. F., Harrison, B. D., and Goold, R. A., 1974, Two protein and two RNA

species in particles of strawberry latent ringspot virus, J. Gen. Viral. 24:29. Mayo, M. A., Barker, H., and Harrison, B. D., 1979a, Polyadenylate in the RNA of five nepo

viruses, J. Gen. Virol. 43:603. Mayo, M. A., Barker, H., and Harrison, B. D., 1979b, Evidence for a protein covalently linked to

tobacco ringspot virus RNA, r. Gen. Viral. 43:735. Mayo, M. A., Barker, H., and Harrison, B. D., 1982, Specificity and properties of the genome

linked proteins of nepoviruses, J. Gen. Viral. 59:149. Mayo, M. A., Robertson, W. M., Legorboru, F. J., and Brierley, K. M., 1994, Molecular approaches

to an understanding of the transmission of plant viruses by nematodes, in: Advances in Molecular Nematology (F. Lamberti, C. De Giorgi, and D. McK. Bird, eds.) pp. 277-293, Plenum Press, New York.


Mayo, M. A., Berns, K., Fritsch, c., Jackson, A. 0., Kaper, J. M., Leibowitz, M. J., and Taylor, J. M., 1995, Satellites, in: Virus Taxonomy-Classification and Nomenclature of Viruses: Sixth Report of the International Committee on Taxonomy of Viruses (F. A. Murphy, C. M. Fauquet, D. H. 1. Bishop, S. A. Ghabrial, A. W. Jarvis, G. P. Martelli, M. A. Mayo, and M. D. Summers, eds.), pp. 487-492, Springer-Verlag, Vienna (also in Arch. Viral., Suppl. 10).

Melcher, U., 1990, Similarities between putative transport proteins of plant viruses, 1- Gen. Vira1. 71:1009.

Meyer, M., Hemmer, 0., Mayo, M. A., and Fritsch, c., 1986, The nucleotide sequence of tomato black ring virus RNA-2, 1- Gen. Vira1. 67:1257.

Morris-Krsinich, B. A. M., Forster, R. 1. S., and Mossop, D. W., 1983, The synthesis and processing of the nepovirus grapevine fanleaf virus proteins in rabbit reticulocyte lysate, Viralogy 130:523.

Murant, A. F., 1981, Nepoviruses, in: Handbook of Plant Virus Infections. Comparative Diagnosis (E. Kurstak, ed.), pp. 197-238, Elsevier/North-Holland, Amsterdam.

Murant, A. F., and Mayo, M. A., 1982, Satellites of plant viruses, Annu. Rev. Phytopatho1. 20:49. Murant, A. F., and Raschke, J. H., 1982, Specificity of the dependence of satellite RNA on tomato

black ring virus (TBRV), Rep. Scottish Crap Res. Inst., 1981, p. 107. Murant, A. F., Taylor, C. E., and Chambers, J., 1968, Properties, relationships and transmission of

a strain of raspberry ringspot virus infecting raspberry cultivars immune to the common Scottish strain, Ann. App1. Bio1. 61:175.

Murant, A. F., Mayo, M. A., Harrison, B. D., and Goold, R. A., 1972, Properties of virus and RNA components of raspberry ringspot virus, 1- Gen. Vira1. 16:327.

Murant, A. F., Mayo, M. A., Harrison, B. D., and Goold, R. A., 1973, Evidence for two functional RNA species and a "satellite" RNA in tomato black ring virus, 1- Gen. Vira1. 19:275.

Murant, A. F., Taylor, M., Duncan, G. H., and Raschke, J. H., 1981, Improved estimates of molecular weight of plant virus RNA by agarose gel electrophoresis and electron microscopy after denaturation with glyoxal, 1- Gen. Viral. 53:321.

Mushegian, A. R., 1993, The putative movement domain encoded by nepovirus RNA-2 is conserved in all sequenced nepoviruses, Arch. Viral. 135:437.

Peden, K. W. C., May, J. T., and Symons, R. H., 1972, A comparison of two plant virus-induced RNA polymerases, Virology 47:498.

Piazzolla, P., Guantieri, v., Vovlas, c., and Tamburro, A. M., 1977a, Circular dichroism studies of chicory yellow mottle virus, 1- Gen. Vira1. 37:359.

Piazzolla, P., Gallitelli, D., and Quacquarelli, A., 1977b, The behaviour of some isometric plant viruses heated in vitra as determined by particle stabilizing forces, 1- Gen. Viral. 37:373.

Piazzolla, P., Savino, v., Castellano, M. A., and Musci, D., 1985, A comparison of grapevine yellow vein virus and a grapevine isolate of tomato ringspot virus, Phytopathol. Mediterr. 24:44.

Pinck, M., Reinbolt, J., Loades, A. M., Le Ret, M., and Pinck, 1., 1991, Primary structure and location of the genome-linked protein (VPg) of grapevine fanleaf nepovirus, FEBS Lett. 284:117.

Puffinberger, C. W., and Corbett, M. K., 1985, Euonymus chlorotic ringspot disease caused by tomato ringspot virus, Phytopathology 75:423.

Quacquarelli, A., Schneider, I. R., and Diener, T. 0., 1971, Partial degradation of tobacco ringspot virus in a single-phase phenol system: Formation of new products, Viralogy 45:664.

Quacquarelli, A., Piazzolla, P., and Vovlas, c., 1972, Freezing in the production of artificial top component of chicory yellow mottle virus, 1- Gen. Vira1. 17:147.

Quacquarelli, A., Gallitelli, D., Savino, v., and Martelli, G. P., 1976, Properties of grapevine fanleaf virus, 1- Gen. Vira1. 32:349.

Ramsdell, D. c., and Stace-Smith, R., 1983, Blueberry leaf mottle virus, CMI/AAB Descriptions of Plant Viruses No. 267.

Randles, J. w., Harrison, B. D., Murant, A. F., and Mayo, M. A., 1977, Packaging and biological activity of the two essential RNA species of tomato black ring virus, f. Gen. Vira1. 36:187.

Remah, A., Jones, A. T., and Mitchell, M. T., 1986, Purification and properties of lucerne


Australian symptomless virus, a new virus infecting lucerne in Australia, Ann. Appl. Biol. 109:307.

Rezaian, M. A., and Francki, R. 1 B., 1973, Replication of tobacco ringspot virus 1. Detection of a low molecular weight double-stranded RNA from infected plants, Virology 56:238.

Ritzenthaler, C., Viry, M., Pinck, M., Margis, R., Fuchs, M., and Pinck, 1., 1991, Complete nucleotide sequence and genetic organization of grapevine fanleaf nepovirus RNA1, J. Gen. Virol. 72:2357.

Roberts, 1 M., 1984, Freeze-drying and high resolution shadowing of plant virus particles for structural studies, in: Proceedings of the 8th European Congress on Electron Microscopy, Budapest, Hungary, Vol. 3 (A. Csanady, P. Rohlich, D. Szab6, eds.l, p. 1805, The Programme Committee of the Eighth European Congress on Electron Microscopy, Budapest, Hungary.

Roberts, 1 M., 1988, The structure of particles of tobacco ringspot nepovirus: Evidence from electron microscopy, J. Gen. Virol. 69:1831.

Roberts, 1 M., and Harrison, B. D., 1970, Inclusion bodies and tubular structures in Chenopodium amaranticolor plants infected with strawberry latent ringspot virus, J. Gen. Viral. 7:47.

Robinson, D. J., Barker, H., Harrison, B. D., and Mayo, M. A., 1980, Replication of RNA-1 of tomato black ring virus independently of RNA-2, J. Gen. Virol. 51:317.


Rost, B., and Sander, C., 1993, Improved prediction of protein secondary structure by use of sequence profiles and neural networks, Proc. Natl. Acad. Sci. USA 90:7558.

Rost, B., and Sander, c., 1994, Combining evolutionary information and neural networks to predict protein secondary structure, Proteins 19:55.

Rott, M. E., Tremaine, J. H., and Rochon, D. M., 1991a, Comparison of the 5' and 3' termini of tomato ringspot virus RNA 1 and RNA 2: Evidence for RNA recombination, Virology 185:468.

Rott, M. E., Tremaine, J. H., and Rochon, D. M., 1991b, Nucleotide sequence of tomato ringspot virus RNA-2, J. Gen. Virol. 72:1505.

Rott, M. E., Gilchrist, A., Lee, 1., and Rochon, D., 1995, Nucleotide sequence of tomato ringspot virus RNA I, J. Gen. Virol. 76:465.

Rubino, 1., Tousignant, M. E., Steger, G., and Kaper, J. M., 1990, Nucleotide sequence and structural analysis of two satellite RNAs associated with chicory yellow mottle virus, J. Gen. Viral. 71:1897.

Russo, M., Martelli, G. P., Rana, G. 1., and Kyriakopoulou, P. E., 1978, The ultrastructure of artichoke yellow ringspot virus infections, Microbiologica 1:81.

Salazar, 1. F., and Harrison, B. D., 1978, Host range and properties of potato black ringspot virus, Ann. Appl. Biol. 90:375.

Sanfa~on, H., 1995, Nepoviruses, in: Pathogenesis and Host Specificity in Plant Diseases, Vol. III. Viruses and Viraids (R. P. Singh, U. S. Singh, and K. Kohmoto, eds.l, pp. 129-141, Pergamon Press, Oxford.

Savino, v., Barba, M., Gallitelli, G., and Martelli, G. P., 1979, Two nepoviruses isolated from olive in Italy, Phytopathol. Mediterr. 18:135.

Savino, V., Gallitelli, D., and Barba, M., 1983, Olive latent ringspot virus, a newly recognized virus infecting olive in Italy, Ann. Appl. Biol. 103:243.

Schneider, 1 R., 1969, Satellite-like particle of tobacco ringspot virus that resembles tobacco ringspot virus, Science 166:1627.

Schneider, I. R., 1971, Characteristics of a satellite-like virus of tobacco ringspot virus, Virology 45:108.

Schneider,I. R., and Diener, T. 0.,1966, The correlation between the proportions of virus-related products and the infectious component during the synthesis of tobacco ringspot virus, Virology 29:92.

Schneider, I. R., White, R. M., and Civerolo, E. 1., 1974, Two nucleic acid-containing components of tomato ringspot virus, Virology 57:139.

Scott, N. W., Cooper, J. 1, Liu, Y. Y., and Hellen, C. U. T., 1992, A 1.5 kb sequence homology in


3' -terminal regions of RNA-l and RNA-2 of a birch isolate of cherry leaf roll nepovirus is also present, in part, in a rhubarb isolate, f. Gen. Virol. 73:481.

Scott, N. W., Cooper, J. I., and Edwards, M. L., 1993, The identification, cloning and sequence analysis of the coat protein coding region of a birch isolate (121 of cherry leaf roll nepovirus, Arch. Virol. 131:209.

Serghini, M. A., Fuchs, M., Pinck, M., Reinbolt, J., Walter, B., and Pinck, L., 1990, RNA2 of grapevine fanleaf virus: Sequence analysis and coat protein cistron location, f. Gen. Virol. 71:1433.

Shanks, M., Tomenius, K., Clapham, D., Huskison, N., Barker, P., Wilson, 1. G., Maule, A. J., and Lomonossoff, G. P., 1989, Identification and subcellular localisation of a putative cell-to-cell transport protein from red clover mottle virus, Virology 173:400.

Stace-Smith, R., 1966, Purification and properties of tomato ringspot virus and an RNA-deficient component, Virology 29:240.

Stace-Smith, R., 1984, Tomato ringspot virus, CMI/AAB Descriptions of Plant Viruses. No. 290. Stace-Smith, R., 1985, Tobacco ringspot virus, CMI/AAB Descriptions of Plant Viruses. No. 309. Stace-Smith, R., Reichmann, M. E., and Wright, S. A., 1965, Purification and properties of

tobacco ringspot virus and two RNA-deficient components, Virology 25:487. Steere, R. L., 1956, Purification and properties of tobacco ringspot virus, Phytopathology 46:60. Steinkellner, H., Weinhausl, A., Laimer, M., Da Camara-Machado, A., and Katinger, H., 1992,

Identification of the coat protein gene of arabis mosaic nepovirus and its expression in transgenic plants, Acta Hartic. 308:37.

Stobbs, L. W., and Van Schagen, J. G., 1985, Relationship between grapevine Joannes-Seyve virus and tomato blackring virus, Can. f. Plant Pathol. 7:37.

Takeda, M., Kuhn, R. J., Yang, C. E, Takegami, T., and Wimmer, E., 1986, Initiation of poliovirus plus-strand RNA synthesis in a membrane complex of infected He La cells, f. Virol. 60:43.

Takemoto, Y., Nagahara, Y., Fukuyama, K., Tsukihara, T., and Iwaki, M., 1985, Crystallization and preliminary characterization of arabis mosaic virus, Virology 145:191.

Tremaine, J. H., and Stace-Smith, R., 1968, Chemical composition and biophysical properties of tomato ringspot virus, Virology 35:102.

Usugi, T., and Saito, Y., 1979, Satsuma dwarf virus, CMI/AAB Descriptions of Plant Viruses No. 208.

Van Lent, J., Wellink, J., and Goldbach, R., 1990, Evidence for the involvement of the 58K and 48K proteins in the intercellular movement of cowpea mosaic virus, f. Gen. Virol. 71:219.

Vos, P., Verver, J., Jaegle, M., Wellink, J., and Goldbach, R., 1988, Two viral proteins involved in the proteolytic processing of the cowpea mosaic virus polyprotein, Nucleic Acids Res. 16:1967.

Walkey, D. G. A., Stace-Smith, R., and Tremaine, J. H., 1973, Serological, physical, and chemical properties of strains of cherry leaf roll virus, Phytopathology 63:566.

Walter, B., Ladeveze, 1., Etienne, L., and Fuchs, M., 1989, Some properties of a previously undescribed virus from cassava: Cassava American latent virus, Ann. Appl. BioI. 115:279.

Wellink, J., and Van Kammen, A., 1989, Cell-to-cell transport of cowpea mosaic virus requires both the 58K/48K proteins and the capsid proteins, f. Gen. Virol. 68:2577.

Wieczorek, A., and Sanfa~on, H., 1993, Characterization and subcellular localization of tomato ringspot nepovirus putative movement protein, Virology 194:734.

CHAPTER 7

N epoviruses: Transmission by Nematodes D. J. F. BROWN, D. L. TRUDGILL,

AND W. M. ROBERTSON

I. INTRODUCTION

Nepoviruses share many biological and molecular characteristics such as symptom types, extensive host ranges, transmissibility through seed, isometric 28-nm-diameter particles, bipartite single-stranded RNA (ssRNA) genomes, and a single coat protein species of about 60 kDa. Some nepoviruses are transmitted by longidorid nematodes, a property that is reflected in the name nepovirus, which stems from nematode-transmitted viruses with polyhedral particles. However, this property has been established only for about one third of the viruses currently classified as nepoviruses (Murant, 1989), and for many others the mode of transmission is unknown. Many nepoviruses, including most of those transmitted by nematodes, have been shown to be transmitted through seed and pollen (Murant, 1983). Those nepoviruses known to have nematodes as vectors are transmitted by species of Longidorus or Xiphinema (Brown, 1989) or, in one instance, Paralongidorus (Jones et al., 1994). Only for tobacco ringspot virus (TRSV) have other animals (several species of arthropod) been implicated as natural vectors (Dunleavy, 1957; Komuro and Iwaki, 1968; Messieha, 1969; Rani et al., 1969; Schuster, 1963).

D. J. F. BROWN, D. L. TRUDGILL, AND W. M. ROBERTSON • Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom.

187

188 D. J. F. BROWN ET AL.

Recent advances in our knowledge of vector nematodes and their relations with the viruses they transmit are reviewed here. Reported associations between longidorid nematodes and nepoviruses are examined in relation to the concept of specificity and to the methods and criteria used to establish nematode transmission. Also, recent information on the structure and feeding behavior of vector nematodes as they affect plant hosts and mechanisms of virus retention is examined and discussed.

II. THE VECTOR NEMATODES

A. Taxonomy and Biology

Longidorus, Xiphinema, and Paralongidorus are closely related genera in the order Dorylaimida, family Longidoridae. Three other orders of nematodes (Aphelenchida, Triplonchida, and Tylenchida) have species that feed on plants, but Paratrichodorus and Trichodorus, in the order Triplonchida, are the only other genera containing species that are specific vectors of plant viruses (Brown et a1., 1989b; Ploeg et a1., 1990). The associations between members of the Triplonchida and tobraviruses are outside the remit of this chapter.

Longidorus, Xiphinema, and Paralongidorus spp. are relatively large (up to 12 mm long), soil-inhabiting nematodes that feed ectoparasitically on plant roots. They have long mouth stylets and can feed deeply within root tips where many species induce galls containing large cells with dense cytoplasm. Most species have wide host ranges, long life-cycles (several months to 5 years), and slow rates of multiplication. The morphology, biology, and ecology of the Longidoridae were comprehensively reviewed by Lamberti et a1. (1975) and detailed standard descriptions of the more important virus vector species have been published: L. attenuatus (Brown and Boag, 1977); L. elongatus (Hooper, 1973); L. macrosoma (Brown and Boag, 1975); X. americanum sensu lato (Siddiqi, 1973); X. diversicaudatum (Pitcher et a1., 1974); X. index (Siddiqi, 1974).

B. Structure of the Feeding Apparatus

Vector species of all three genera have a long (60-250 j.Lm), hollow, hypodermic needlelike stylet that enables them to feed deeply within plant roots yet remain ectoparasitic. The stylet (Figs. 1 and 2A) is in two parts. The anterior part (odontostyle) is used to penetrate the plant root and has an opening at the tip. The posterior part (odontophore or stylet extension) contains nerve tissues in close proximity to the food canal; this enables feeding nematodes to discriminate between sites deep within plant roots (Robertson, 1976; Robertson and Wyss, 1983; Trudgill, 1976; Trudgill et a1.,

NEPOVIRUSES: TRANSMISSION 189

s .. B ~

~ p

FIGURE 1. A female Xiphinema index; B, esophageal bulb; S, odontostyle; P, odontophore IS + P = stylet); V, vulva. Bar, 200 ~m.

1991). The stylet protractor muscles are attached to the rear of the odontophore (in Xiphinema spp. there are characteristic flanges for muscle attachment) and the length and rigidity of the odontophore enables the odontostyle to be almost totally protracted.

Longidorid nematodes are characterized by the presence of a typical Dorylaimoid esophagus (Fig. 1), consisting of a long narrow anterior part that connects the stylet to a prominent muscular cylindrical bulb. The esophageal bulb provides the pumping action during feeding that withdraws the contents of the plant cells and forces them into the gut against the hydrostatic pressure of the body. The esophageal bulb also contains three large gland cells. The duct of the dorsal gland cell and an associated gland nucleus opens into the food canal just anterior to the esophageal bulb and its contents are passed forward into the plant during feeding (Robertson and Wyss, 1983; Robertson et a1., 1985; Towle and Doncaster, 1978; Trudgill, 1976; Trudgill et a1., 1991). A pair of subventral gland ducts and gland nuclei are situated about halfway along the bulb and, in some instances, there appears to be a second pair of subventral gland duct openings, but without gland nuclei, at the posterior end of the bulb. The ducts of the two subventral gland cells open into the pump chamber and during feeding their contents are probably passed backward through a one-way valve into the gut. The absence of the second pair of subventral gland nuclei helps to separate longidorids from other Dorylaims (Hooper, 1975).

190 D. 1. F. BROWN ET AL.

~o --

c FIGURE 2. [A) Xiphinema index feeding on a root of Ficus carica; 0, esophageal bulb; S, stylet. Bar, 100 /-Lm. (B) Longitudinal section through the lumen of the food canal (F) in the odontophore of Xiphinema index showing particles of GFLV (V) adsorbed to the food canal wall. Bar, 200 nrn. (e) Transverse section of the'stylet and guiding sheath of Longidorus macrosoma with particles (V) of RRSV. Bar, 200 nm.

C. Feeding Behavior

Those Longidorus spp. studied all have a similar feeding behavior. A feeding site is selected in the region of the root meristem or in the zone of elongation, and several nematodes may exploit an individual root tip as a feeding site (Fig. 2A). The odontostyle is progressively inserted until it is almost completely protracted and the tip is several cells deep. The next stage


is a period of relative inactivity during which saliva from the dorsal gland cell is thought to be injected, over a period of 30-50 min. This is followed by active ingestion which usually continues for several hours, interrupted at approximately hourly intervals for about 3 min during which time the nematode is believed to again inject saliva (Robertson et a1., 1985; Towle and Doncaster, 1978). A volume equivalent to that of more than 40 root tip cells may be extracted by a single nematode feeding for 1 hr (Robertson et a1., 1985), the cytoplasm and most of the organelles being removed. Older galls contain groups of empty cells in which the interconnecting walls have been holed, presumably as a direct or indirect effect of the nematode's saliva (Bleve-Zacheo et a1., 1977; Robertson et a1., 1985).

Xiphinema spp. that induce root tip galls show two distinct types of feeding behavior. Most frequently, nematodes feed on a column of progressively deeper cells. As each cell is penetrated, the contents are removed by short periods of ingestion (a few seconds to a few minutes) interspersed with brief pauses (1-10 sec) during which saliva from the dorsal gland cell is rapidly injected into the plant. The dorsal gland cell in Xiphinema spp. contains an extensive duct system, absent in Longidorus spp., which facilitates this rapid salivation. The second type of feeding behavior usually occurs only in deeper cells and resembles feeding by Longidorus spp.: a long period (15-60 min) of inactivity during which saliva is thought to be injected, generally followed by an even longer period (1-3 hr) of continuous ingestion (Trudgill and Robertson, 1982). During these prolonged periods of ingestion, a globular structure has been observed at the stylet tip and the cytoplasm of the cell penetrated by the stylet tip appears to remain intact. This latter type of feeding resembles that of several sedentary, endoparasitic, tylenchid nematodes that have stylets with much narrower lumina than longidorids and that use secretions from the dorsal gland cell to form special feeding tubes in the host cell. The probable effect of these feeding tubes is to limit the materials ingested to the liquid fraction (Wyss and Zunke, 1986).

D. Host Response

On plants growing in agar, X. diversicaudatum and x. index feed on most parts of the root, whereas L. elongatus feed only at root tips (Trudgill and Robertson, 1982). Root tips fed upon by L. elongatus, X. diversicaudatum, or X. index, but not those fed upon by X. americanum sensu lato, cease growing and swell to form characteristic galls. Root tip galls induced by x. index on good hosts, such as fig and grapevine, contain groups of enlarged, multinucleate cells with dense cytoplasm (Rumpenhorst and Wei scher, 1978; Vovlas et a1., 1978; Weischer and Wyss, 1976; Wyss, 1981; Wyss et a1., 1980, 1988). Similar changes are caused by X. diversicaudatum and the root tip galls also contain increased amounts of DNA, RNA, and protein (Griffiths et a1., 1982; Griffiths and Trudgill, 1983; Griffiths and Robertson, 1984).


Root tip galls induced by Longidorus spp. contain large cells with enlarged, ameboid-shaped nuclei. Multinucleate cells have not been found, but the swollen cells contain increased amounts of DNA (Bleve-Zacheo et a1., 1977; Griffiths et a1., 1982; Griffiths and Robertson, 1983).

III. RELATIONS BETWEEN NEPOVIRUSES AND THEIR NEMATODE VECTORS

A. Criteria for Demonstrating Nematode Transmission

Trudgill et a1. (1983) proposed the following set of criteria for assessing the transmission of plant viruses: (1) it must be shown that the virus has infected the bait plant systemically; (2) the tests should be conducted with handpicked nematodes and should include such controls that the nematode is shown unequivocally to be the vector; and (3) the nematode and the virus isolate should be identified at the commencement and conclusion of each test. All the published records of longidorid nematodes transmitting nepoviruses have been reevaluated according to these criteria (Trudgill et a1., 1983), and those associations that satisfy them are listed in Table I under the heading "specific vector." Those for which the published evidence is considered inadequate are listed under the heading "nonvalidated association."

B. Test Procedures

Many factors may affect the frequency of transmissions obtained in laboratory and glasshouse experiments (McGuire, 1982; Taylor, 1980; Taylor and Brown, 1981). In tests to establish a species as a virus vector, the nematodes must feed on the virus source plant and on the bait plants; the nematodes must ingest the virus from the source plants; the virus must infect the bait plant; and possible virus contamination of the bait plant root system or transmission by alternative vectors must be avoided. Trudgill and Brown (1978a) described test procedures that met these criteria and provided a means of assessing the effectiveness with which each stage of the transmission process has been accomplished. Their tests are conducted with small numbers of nematodes in small plastic pots placed in temperature-controlled boxes to obviate soil moisture and temperature fluctuations (Taylor and Brown, 1974). Root tip galls, when formed, are counted to assess the extent of feeding on the roots of the source and bait plants. The proportion of Longidorus spp. ingesting virus from the source plants may be estimated by crushing groups of nematodes and inoculating the resulting suspension directly to suitable indicator plants (Trudgill and Brown, 1978a; Yassin, 1968). However, this "slash-testing" technique detects only infective virus in the nematode gut and underestimates the proportion of nematodes ingesting virus; moreover it is unsuitable for Xiphinema spp. Immunosorbent electron micro-


scopy (ISEM), which is much more sensitive than slash-testing, can be used for detecting virus both in Longidorus and in Xiphinema spp. (Roberts and Brown, 1980). However, it was not possible to detect virus in the heads of vector species by using ISEM and the detection of virus in nematodes by ISEM or slash-testing is not indicative that the virus is transmissible by the nematode. Specifically retained virus particles have been detected only by electron microscopy of thin sections through the specific sites of virus retention in vector species.

In the procedures described by Trudgill et a1. (1983), handpicked nematodes are used throughout and appropriate procedures and controls are used to detect systemic infection of the bait plants and to ensure the absence of alternative vectors. The proportion of nematodes transmitting virus in any test may be estimated by using the maximum-likelihood equation of Gibbs and Gower (1960). Provided that the number of nematodes tested is sufficient, these procedures can also be used to provide evidence of the probable inability of a nematode species to transmit a virus isolate.

C. Vector-Virus Associations

Since Hewitt et a1. (1958) demonstrated that X. index is a natural vector of grapevine fanleaf virus (GFLV), more than 40 plant virus-vector combinations have been reported. Table I shows in the column headed "specific vector" that, of ca 300 species in the Longidoridae, only eight species of Longidorus and six of Xiphinema have been confirmed as vectors of viruses (Brown, 1989; Brown and Halbrendt, 1992; Brown et a1., 1994a,b). In addition, circumstantial evidence from field association and natural spread of virus and from limited laboratory experimentation suggests that Paralongidorus maximus is a natural vector of an atypical strain of raspberry ringspot virus (RRSV) in vineyards in the Palatinate region of Germany (Jones et a1., 1994). Except for mulberry ringspot virus (MRSV), transmitted by L. martini in Japan, all the known associations involving longidorid nematodes and nepoviruses have been described from Europe or North America (Brown, 1989; Brown et a1., 1994a,b), although several of the nematode species occur in other regions of the world. Several vector-virus associations have widespread geographical distributions, for example, the North American nepoviruses with their vector species, X. index with GFLV, and X. diversicauda tum with arabis mosaic virus (ArMV). In contrast, most associations in Europe have relatively small, well-defined distributions within a country, for example, L. apulus and L. fasciatus with artichoke Italian latent virus (AlLV) in Italy and Greece, respectively, and L. arthensis with cherry rosette virus (CRV) in the Arth region of Switzerland.

Some nepoviruses can be transmitted by more than one distinct vector (e.g., the Scottish and English strains of RRSV are each transmitted by both L. elongatus and L. macrosoma), and some vectors are able to transmit more than one distinct virus [e.g., X. diversicaudatum transmits both ArMV and

TA

BL

E I

. N

emat

ode-

Tra

nsm

itte

d N

epov

irus

es w

ith

Th

eir

Spe

cifi

c V

ecto

r S

peci

es

and

Oth

er R

epor

ted

Ass

ocia

tion

s w

ith

Lon

gido

rid

Nem

atod

es

Nep

ovir

usa

Asi

an

Long

idor

us

Mul

berr

y ri

ngsp

ot

Eur

opea

n Lo

ngid

orus

A

rtic

hoke

Ita

lian

lat

ent!

Che

rry

rose

tteb

Ras

pber

ry r

ings

pot

Tom

ato

blac

k ri

ng!

Par

alon

gido

rus

Ras

pber

ry r

ings

pot

Xip

hine

ma

Ara

bis

mos

aic2

Che

rry

leaf

roll

Gra

pevi

ne c

hrom

e m

osai

c!

Gra

pevi

ne f

anle

afc,

2 S

traw

berr

y la

tent

rin

gspo

tb

Acr

onym

MR

SV

AlL

V

CR

Y

RR

SV

TB

RV

RR

SV

ArM

V

CL

RV

GC

MV

G

FLV

SL

RSV

Spec

ific

vec

tor

L. m

arti

ni (Y

agita

and

Kom

uro,

197

2)

L. a

pulu

s (R

ana

and

Roc

a, 1

975;

Roc

a et

a1.

, 19

75;

Lam

bert

i an

d B

leve

-Zac

heo,

197

9).

L. f

asci

atus

(R

oca

and

Lam

bert

i, 1

981;

Roc

a et

a1.,

198

2)

L. a

rthe

nsis

(B

row

n et

a1.,

199

4a)

L. e

long

atus

(T

aylo

r, 19

62)

L. m

acro

som

a (H

arri

son,

196

4)

L. a

tten

uatu

s (H

arri

son,

196

4)

L. e

long

atus

(H

arri

son

et a

1.,

1961

)

P. m

axim

us (

Jone

s et

a1.,

199

4)

X.

dive

rsic

auda

tum

(H

arri

son

and

Cad

man

, 19

59;

Jha

and

Pos

nett

e, 1

959)

Unk

now

n

Unk

now

n X

. in

dex

(Hew

itt

et a

1.,

1958

) X

. di

vers

icau

datu

m (

Lis

ter,

196

4)

Non

vali

date

d as

soci

atio

ns

Non

e

Non

e

Non

e L.

cae

spiti

cola

(V

alde

z, 1

972)

L.

lep

toce

phal

us (

Val

dez,

197

2)

L. p

rofu

ndor

um (

Fri

tzsc

he a

nd K

egle

r, 19

68)

X.

dive

rsic

auda

tum

(F

ritz

sche

and

Keg

ler,

1968

) N

one

Non

e

L. c

aesp

itico

la (

Val

dez,

197

2)

P. m

axim

us (

McE

lroy

et

a1.,

1976

) X

. ba

keri

(Iw

aki

and

Kom

uro,

197

4)

X.

coxi

(Fr

itzs

che,

196

4)

X.

inde

x (F

ritz

sche

and

Thi

ele,

197

9)

X.

coxi

(F

ritz

sche

and

Keg

ler,

1964

) X

. di

vers

icau

datu

m (

Fri

tzsc

he a

nd K

egle

r, 19

64)

X.

vuit

tene

zi (F

legg

, 19

69)

X.

inde

x (M

ali

et a

1., 1

975)

X

. it

alia

e (C

ohn

et a

1.,

1970

) P.

max

imus

(M

cElr

oy e

t a1

., 19

76)

X.

coxi

(P

utz

and

Stoc

ky,

1970

)

Am

eric

an

Long

idor

us

Pea

ch r

oset

te m

osai

c X

iphi

nem

a C

herr

y ra

sp l

eafb

Pea

ch r

oset

te m

osai

c P

otat

o bl

ack

ring

spot

3

Tob

acco

rin

gspo

t3

Tom

ato

ring

spot

PRM

V

CR

LV

PRM

V

PBR

SV

TR

SV

ToR

SV

L. d

iade

ctur

us (

Eve

leig

h an

d A

llen

, 19

82)

X.

amer

ican

um s

ensu

lato

(N

ylan

d et

a1.,

196

9)

X.

amer

ican

um s

ensu

str

icto

(B

row

n an

d H

albr

endt

, 19

92)

X.

calif

orni

cum

(B

row

n an

d H

albr

endt

, 19

92)

X.

rive

si (

Bro

wn

and

Hal

bren

dt,

1992

) X

. am

eric

anum

sen

su la

to (

Klo

s et

a1.,

196

7)

Unk

now

n

X.

amer

ican

um s

ensu

lato

(F

ulto

n, 1

962)

X

. am

eric

anum

sen

su s

tric

to (

Bro

wn

and

Hal

bren

dt,

1992

) X

. ca

lifor

nicu

m (

Hoy

et

a1.,

1984

) X

. ri

vesi

(B

row

n an

d H

albr

endt

, 19

92)

X.

amer

ican

um s

ensu

lat

o (B

reec

e an

d H

art,

19

59)

X.

amer

ican

um s

ensu

str

icto

(B

row

n an

d H

albr

endt

, 19

92)

X.

bric

olen

sis

(Bro

wn

and

Hal

bren

dt,

1992

) X

. ca

lifo

rnic

um (

Hoy

et

a1.,

1984

) X

. ri

vesi

(Fo

rer

et a

1., 1

981;

Bro

wn

and

Hal

bren

dt,

1992

)

aVir

uses

wit

h t

he

sam

e su

pers

crip

t nu

mbe

r ar

e se

rolo

gica

lly

rela

ted.

bT

enta

tive

nep

ovir

uses

.

Non

e

Non

e

Non

e X

. am

eric

anum

sen

su la

to (

e.

E. F

ribo

urg

and

P. J

atal

a, i

n J

ones

, 19

81)

X.

coxi

(va

n H

oof,

197

1)

X.

bake

ri (

Con

vers

e, 1

977)

X.

brev

icol

le (

Fri

tsch

e an

d K

egle

r, 19

68)

cGFL

V a

nd i

ts v

ecto

r, X

. in

dex,

occ

ur i

n m

ost

of t

he w

orld

's p

rinc

ipal

vit

icul

ture

are

as (

Bro

wn,

198

9; M

uran

t, 1

989;

Tay

lor,

197

2; T

aylo

r B

row

n, 1

981)

.


strawberry latent ringspot virus (SLRSV), and L. elongatus transmits both RRSV and tomato black ring virus (TBRV)]. However, with two reported exceptions, nepoviruses are in general not transmitted by nematode species of more than one genus.

The exceptions are RRSV and peach rosette mosaic virus (PRMV). Two serologically distinct strains of RRSV are transmitted in Germany by L. macrosoma and P. maximus, respectively. PRMV has been reported to be transmitted in Michigan by x. americanum sensu lata (Klos et a1., 1967) and in Ontario by both x. americanum sensu lata and L. diadecturus (Allen et a1., 1982, 1984; Eveleigh and Allen, 1982). The only population of L. diadecturus shown to be a vector of PRMV was from an orchard with PRMVinfected peach trees, in Mersea Township, Essex County, Ontario (Allen, 1986; Allen et a1., 1982; Eveleigh and Allen, 1982). No other populations of this species, which is widespread in North America (Robbins and Brown, 1991), have been shown to transmit PRMV. Allen et a1. (1984) reported that x. americanum and L. diadecturus occurred together at this orchard, that both were vectors of PRMY, but that the x. americanum-group nematodes were much less efficient. These X. americanum-group nematodes differed morphologically from those from Michigan, reported by Klos et a1. (1967) as vectors of PRMV. The original Ontario orchard site has been fumigated and L. diadecturus have not been observed in soil samples collected subsequently. Also, no virus isolates are available from the numerous virus transmission tests made with L. diadecturus (W. R. Allen, personal communication). It seems possible that in this one orchard L. diadecturus was associated with an atypical isolate of PRMV, and that this species is not a vector of PRMV at other locations.

Although X. italiae was reported as a vector of GFLV in Israel (Cohn et a1., 1970), relatively large numbers of nematodes (50-200 individuals) were used in the transmission tests. The possibility exists that small numbers of X. index, a species present in association with GFLV in Israel, may have inadvertently been used in these tests. Subsequently, Martelli (1975) claimed to have obtained a single transmission of GFLV in one experiment with X. italiae. However, despite the widespread occurrence of this nematode species and a similar widespread occurrence of GFLV in viticultural areas in the Mediterranean region (Brown and Taylor, 1987), no unequivocal evidence has been published to show that x. italiae is a vector of GFLV. On the contrary, in an extensive survey of viticultural areas in southern Italy (Catalano et a1., 1992), X. index transmitted GFLV in 117 of 119 samples, whereas x. italiae, which was present in 41 samples, did not transmit GFLV. We conclude that it is unlikely that x. italiae is a specific vector of GFLV.

D. Specificity

It is apparent from Table I (column headed "specific vector") that, among nepovirus-nematode vector associations that are well established (i.e., natural association of virus and vector nematode species at field sites, and labora-


tory evidence that the nematode species is a vector of the virus), there is a pattern of specificity between the viruses and their vectors. Reports that populations of vector species differ in their ability to transmit different isolates of a virus (Brown, 1985, 1986a; Brown and Taylor, 1981; Brown and Trudgill, 1983; Brown et a1., 1989a, 1994b,c; McGuire, 1982; Van Hoof, 1971) provide further evidence for the concept of a specific relationship between nepoviruses and their vectors.

Harrison et a1. (1961) and Cadman (1963) reported that the more two strains of a nepovirus differed serologically (a function of the virus coat protein), the more likely they were to be transmitted by different nematode species. Subsequently, evidence was obtained that vector-virus specificity is determined by particular characters of the virus coat protein (Harrison, 1964; Harrison et a1., 1974a,b). However, the same nematode species may transmit minor serological and/or symptomatological variants of a single virus, the occurrence of which in mixed infections at a single site is now well established (Brown, 1989; Brown et a1., 1994c; Jones et a1., 1989; Rudel et a1., 1983; Stellmach and Berres, 1985; Taylor and Brown, 1976). The type Scottish strain of RRSV is transmitted by L. elongatus and this species also transmits two minor antigenic variants of RRSV recovered from raspberry cultivars immune to the type Scottish strain. These three variants all occur in raspberry plantations in Scotland in association with L. elongatus (Murant et a1., 1968; Jones et a1., 1989). Also, two symptomatological variants of ArMV have been recovered from populations of X. diversicaudatum from Norway and Scotland, and transmission tests with individual nematodes from a viruscarrying population of X. bricolensis from a raspberry plantation in Washington state revealed the natural occurrence of three antigenic variants of tomato ringspot virus (ToRSV) (Brown et a1., 1994c).

In contrast, Brown et a1. (1989a) described English and German isolates of TBRV that differed in transmissibility by an English population of L. attenuatus, though they could not be distinguished from each other by spur formation in gel diffusion serological tests. The explanation for this is not known.

The reports listed in Table I as "nonvalidated associations" between nepoviruses and nematode species are, as mentioned above, regarded by us as inadequate. These reports, usually based on laboratory experiments, would, if true, indicate a marked lack of specificity between some nepoviruses and their vector nematodes; for example, transmission of ArMV by six species (in three genera) and of RRSV by seven species (in three genera) of longidorid nematode. Various explanations have been made for apparent nonspecific transmissions. Taylor and Robertson (1969, 1975) found unattached virus particles in the buccal capsule of Longidorus and Xiphinema spp. and suggested that this might occasionally result in nonspecific transmission of virus. McNamara (1978) offered an alternative explanation for apparent nonspecific transmissions. He showed that virus particles in the bodies of nematodes entangled in, or in nematode feces adhering to, roots of bait plants could be potent sources of contamination in laboratory tests. He concluded


that unequivocal evidence for nematode transmission requires clear demonstration of systemic infection of the bait plant.

Evidence that a nepovirus may be transmitted in a nonspecific manner by a longidorid nematode was reported for an isolate of PRMV; which was transmitted by one of 52 handpicked groups of ten L. breviannulatus (Allen, 1986) and by one of 46 groups of two L. elongatus (Allen and Ebsary, 1988). The bait plants in each experiment were shown to be systemically infected with PRMV and the virus transmission criteria proposed by Trudgill et al. (1983) (see Section III.A) were fulfilled. Allen and Ebsary (1988) concluded that the viruses had been transmitted in a nonspecific manner. Such transmission probably occurs only in laboratory experiments in which nematodes are quickly transferred from virus-infected to healthy plants; there appears to be little opportunity for such nonspecific transmission under natural conditions.

The position with North American nepoviruses is especially confusing. Between 1959 and 1969, X. americanum was reported as a vector of cherry leafroll virus (CRLV), PRMV, ToRSV; and TRSV (Breece and Hart, 1959; Fulton, 1962; Klos et al., 1967; Nyland et al., 1969). Subsequently, Lamberti and Bleve-Zacheo (1979) reviewed the X. americanum group, included the descriptions of 15 new species, and recognized 25 morphologically similar, parthenogenetic species. An additional 19 species subsequently have been described, to provide a current total of 44 species. Of these species, 21 are present in North America and 12 are considered indigenous (Robbins and Brown, 1991; Robbins, 1993).

All the earlier reports of virus transmission by X. americanum must now be referred to as X. americanum sensu lato. However, Forer et al. (1981) and Hoy et al. (1984) have reported X. rivesi andX. californicum as vectors of ToRSV. More recently, Brown et al. (1994b) tested three populations of X. americanum sensu stricto and one population each of X. bricolensis, x. californicum, and X. rivesi for their ability to transmit cherry rasp leaf virus (CRLV), TRSV, and two serologically distinguishable strains of ToRSV. All four viruses were transmitted by X. californicum and X. rivesi, the latter being the most efficient vector; X. bricolensis transmitted ToRSV strain PBL more efficiently than ToRSV strain PSP and did not transmit CRLV or TRSV. Nematodes from each of the X. americanum sensu stricto populations transmitted TRSV but not ToRSV-PBL; in addition, those from Arkansas and California transmitted ToRSV-PSP and those from Pennsylvania transmitted CRLV. The transmission of CRLV; TRSV; and ToRSV by x. americanum, X. californicum, and x. rivesi contrasts with the very narrow specificity of transmission that exists between indigenous European nepoviruses and their vector species. However, the findings that X. bricolensis transmits only ToRSV (and one strain more efficiently than another), and that populations of x. americanum differ in their ability to transmit CRLV and strains of ToRSV, suggest that there is some specificity of virus transmission by these nematodes. This vector-virus specificity is evident at the population level and


with serologically distinguishable virus strains, and appears similar to that reported for populations of X. diversicaudatum and serologically distinguishable strains of ArMV and SLRSV (Brown, 1985, 1986a). Brown et al. (1993) suggested that differences between distinguishable strains of ToRSV in their transmissibility by a population of x. californicum (Hoy et al., 1984) and between different populations of X. americanum-group nematodes, in their ability to transmit TRSV and strains of ToRSV, may result from specificity of transmission of "local" isolates of virus by "local" populations of nematode species. Therefore, specificity of transmission between North American nepoviruses and their nematode vector species may be more complex and subtle than observed with European viruses and their specific vector nematode species.

Identification of the parthenogenetically reproducing members of the x. americanum group is based on relatively minor morphological and/or small morphometrical differences. Nevertheless, morphological characteristics and morphometrics enabled four populations used by Brown et al. (1994b) to be distinguished from one another and to be identified, by their close similarity with the published descriptions, with the four species X. americanum sensu stricto, X. bricolensis, x. californicum, and x. rivesi. Genetic discontinuities among these same populations were confirmed by Vrain et al. (1992), who compared their DNA by restriction fragment length polymorphism analyses.

After studying populations of several X. americanum-group species, Halbrendt and Brown (1992,1993) reported that species indigenous to North America had only three juvenile stages, whereas another species found in Europe, in common with most Nematoda, had four. Also, all populations used in the study by Brown et al. (1994b) and shown to be virus vectors had only three juvenile stages. X. americanum-group species found in Europe have not been reported to be naturally associated with nepoviruses. If it could be shown to be generally true that nonvector species have four juvenile stages, whereas vector species have three, this might provide a rapid, objective method for distinguishing virus-vector populations of these morphologically similar species.

E. Vector Efficiency

The number of transmissions obtained in experiments, usually considered as representing the virus transmission efficiency of the vector species, is affected by numerous factors, including the age and kind of plant used, temperature, soil moisture, size of pot, and numbers of nematodes used (Taylor and Brown, 1981). Minimum periods reported for Xiphinema species to acquire viruses from infected plants refer to access periods and not to feeding periods. Teliz et al. (1966) found that 1 hr was sufficient for x. americanum sensu lata to acquire ToRSY, but that the amount of transmis-


sion was not proportional to the length of the acquisition access period during the first 48 hr. In separate experiments, the frequency of transmission increased when access time to the bait plant was increased, with 100% transmission being achieved when the access period was 4 days. In general, increasing both the acquisition and inoculation access times results in increased frequency of transmission. However, the virus transmission frequency also depends on the number of nematodes used in each test replicate and the number that gain access to and feed on the plant roots. Adult and juvenile stages of vector nematodes are capable of transmitting their associated viruses; for example, all developmental stages of X. americanum sensu lato and x. index were found to transmit ToRSV and GFLV, respectively (Raski and Hewitt, 1963; Teliz et al., 1966). However, ArMV and SLRSV were found to be transmitted more frequently by female than by male X. diversicaudatum (Harrison, 1967; Taylor and Thomas, 1968), although this may be due to differences in the frequency of feeding rather than to inherent differences in efficiency of transmission. Also, viruses do not pass through the egg and they are shed when the juvenile stages of the vector nematodes pass through a molt (Harrison and Winslow, 1961; Taylor and Raski, 1964)

Xiphinema spp. are usually efficient vectors of their associated viruses. In vineyards, rapid spread of GFLV occurs even when populations of X. index are almost undetectable (M. Riidel, personal communication; J. Klingler, personal communication). Similarly, ArMV (hop strain) may spread in soils containing less than one X. diversicaudatum per 200 g soil (Pitcher, 1975). Trudgill et al. (1981) demonstrated in mini-pot experiments that about 80% of X. diversicaudatum transmitted ArMV and 30% transmitted SLRSV. X. americanum sensu lata is also an efficient vector, with 3-5% of individual nematodes transmitting TRSV during a 24-hr period of access to the bait plant (McGuire, 1964a). Individual X. americanum sensu lata were capable of transmitting virus to more than one plant without reacquisition (McGuire, 1964b), and both TRSV and ToRSV can be transmitted together by one nematode (Fulton, 1967). X. americanum sensu lata retained TRSV for up to 1 year when stored at low temperatures without a host (Bergeson et al., 1964); X. index in moist, plant-free soil retained GFLV for 8 months (Taylor and Raski, 1964; Taylor, 1968); and x. diversicaudatum retained viable ArMV and SLRSV for 8 months, more than sufficient time for the virus to overwinter in its vector (Taylor, 1972).

In contrast, the effectiveness with which Longidorus spp. transmit their associated viruses is more variable. Trudgill et al. (1981) estimated that TBRV (English strain) was transmitted in laboratory tests by 27% of L. attenuatus, RRSV and TBRV (Scottish strains) by 2 % and 8% of L. elongatus, respectively, and RRSV (English strain) by 5% of L. macrosoma. However, efficiency of transmission can be substantially affected by the choice of virus isolate; for example, Brown et al. (1989a) found that a population of L. attenuatus from England transmitted seven isolates of TBRY; with frequencies ranging from 3 to 78%. Also, the natural interactions between virus,


vector, and plant host can substantially influence the frequency of transmission by the vector nematode. For example, between 2 and 15 % of individual L. arthensis recovered from the root zone of diseased cherry trees growing in Switzerland transmitted CRY (Brown et a1., 1994a), whereas between 14 and 49% of L. fasciatus recovered from the root zone of diseased artichokes growing in Greece transmitted AlLV (D. J. F. Brown, unpublished data). Apparently under natural conditions L. fasciatus is a more frequent vector of its naturally associated nepovirus than is L. arthensis with its associated nepovirus.

F. Sites of Virus Retention

In the absence of a host, viruses seem to be less persistent in Longidorus than in Xiphinema species. For example, RRSV and TBRV remain transmissible for only about 8 weeks in L. elongatus, whereas ArMV and GFLV remain transmissible in their vector Xiphinema species for 8 months or more (Murant and Taylor, 1965; Taylor, 1968, 1972).

The specific sites of retention have been revealed by electron microscopy of thin sections through the feeding apparatus of nematodes exposed to virus-infected plants. In X. diversicaudatum carrying ArMV or SLRSV, in X. index carrying GFLV, and in X. americanum sensu lato carrying ToRSV, the virus particles were found to be associated with the cuticle lining the lumen of the odontophore and the esophagus (Fig. 2B) (McGuire et a1., 1970; Raski et a1., 1973; Robertson, 1975; Taylor and Robertson, 1970). In contrast, the viruses transmitted by species of Longidorus are associated with the odontostyle, and early investigations with L. elongatus and Scottish strains of RRSV and TBRV revealed particles between the odontostyle and the guiding sheath (Taylor and Robertson, 1969). However, subsequent studies revealed particles in association with the inner surface of the odontostyle (Trudgill et a1., 1981), a feature that had been observed in L. macrosoma carrying RRSV (Fig. 2C) (English strain) and L. apulus carrying AILV (Taylor and Robertson, 1975; Taylor et a1., 1976). Despite extensive searches, no virus like particles have been found in the esophageal (salivary) glands or surrounding tissue (W. M. Robertson, unpublished data).

Taylor and Robertson (1975) suggested that virus particles may be retained for short periods in the lumen of the stoma of Longidorus and Xiphinema nematodes. Viruslike particles were observed in the anterior mouth parts of L. elongatus, that had previously been allowed to feed on plants infected with ArMY, but the nematodes did not transmit the virus (Taylor and Robertson, 1969). Also, clumps of virus particles, present in sufficient quantity to provide an infectious inoculum when the nematodes next fed, have been observed by electron microscopy in the triangular section of the lumen immediately anterior to the guide ring in L. macrosoma and x. diversicaudatum (Taylor and Robertson, 1975).


G. Transmission: Ingestion, Retention, and Inoculation

Transmission comprises a sequence of processes. Virus particles must be (1) ingested from a source plant, (2) associate with and (3) later dissociate from the site of retention in the vector, and (4) then be introduced into the bait plant. During ingestion only a small proportion of virus particles is retainedj most particles are passed into the gut. Brown (1986b) demonstrated that the ability of populations of X. diversicaudatum to retain and transmit isolates of ArMV and SLRSV is an inherited character. Harrison et a1. (1974a) suggested that the surface of the virus particle plays a key role in the retention proceSSj the RNA-2 segments of the virus genomes of both RRSV and TBRV, which include the virus coat protein genes, were shown to contain the genetic determinants for vector transmissibility (Harrison et a1., 1974aj Harrison and Murant, 1978).

Trudgill and Brown (1978b) found that inefficient transmission by L. macrosoma of RRSV, especially of the Scottish type strain, was linked with failure of the virus to dissociate from the specific sites of retention from within the vector (Fig. 2C). Moreover, with European Longidorus species, casual observation suggests that the more efficient the nematode species as a vector, the smaller the number of virus particles observed by electron microscopy at the specific sites of retention (W. M. Robertson, unpublished data). No such correlation has been found with virus-vector Xiphinema species, and this may imply fundamental differences in the mechanisms of recognition between Longidorus andXiphinema species and their associated viruses. A fast-binding (retention)/slow-release (dissociation) mechanism could present an ecologically advantageous system for successful virus transmission by vector nematodes (Mayo et a1., 1994), and this may occur with Xiphinema vector species.

Taylor and Brown (1981) speculated that surface charges may be involved, and suggested that pH changes during salivation cause dissociation of virus particles from the site of retention, and hence lead to transmission. Robertson and Henry (1986) found that, at the sites of retention, the wall of the food canal can be stained for carbohydrate and that particles of ArMV retained within the odontophore of X. diversicaudatum were surrounded by a matrix of carbohydrate-containing material. These results suggest that the protein coat of the virus particles may have lectinlike properties, with carbohydrates being involved in both retention and release. Examination of electron micrographs of nepoviruses specifically retained in vector longidorids revealed thin linking structures between individual virus particles, and between virus particles and the wall of the food canal of the nematode (Robertson and Wyss, 1983 j Mayo et a1., 1994). Comparison of amino acid sequences of proteins encoded by the genomic RNA of nepoviruses also indicates a correlation between a virus gene product and transmission. Blok et a1. (1992) reported that with GFLV and ToRSV (viruses with Xiphinema vectors) stretches of identical or nearly identical amino acid sequence were


present in the polyprotein encoded by the RNA-2 of each virus, immediately upstream of the coat protein gene; the same was true of TBRV and RRSV (viruses with Longidorus vectors). However, the stretches of sequence shared by either pair of these viruses were not common to all four. It has been speculated that these proteins might attach to structures in the nematodes that differ between the two genera (Mayo et a1., 1994).

IV. CONCLUSIONS

A striking feature of nematode-virus interactions is their specificity. Each nepovirus is transmitted by only one or a few nematode species. Moreover, there are differences in transmission efficiency between different isolates of a virus (especially isolates that differ serologically) and between different populations of the vector nematodes.

Another striking feature is that Xiphinema vector species remain able to transmit their associated viruses for many months, even years, whereas Longidorus vector species retain their associated viruses for only a few weeks. This suggests that fundamental differences may exist in the nature of the vector-virus association between the two nematode genera.

The specificity of the association between the vector and the virus appears to be determined by properties of the virus particle proteins. Although some clues are beginning to emerge about the regions of these proteins that might be responsible, the precise mechanism(s) involved in the retention of virus particles at, and their subsequent release from, specific sites in the vector are unknown. Carbohydrates, perhaps secreted from the dorsal gland cell in the esophageal bulb or perhaps coming from the plant, might be involved in this process. Moreover, current information does not rule out the possibility that, as with tobraviruses (Mayo et a1., 1994), a viruscoded helper factor may also be involved.

REFERENCES

Allen, W. R., 1986, Effectiveness of Ontario populations of Longidorus diadecturus and L. breviannulatus as vectors of peach rosette mosaic and tomato black ring viruses, Can. f. Plant Pathol. 8:49.

Allen, W. R., and Ebsary, B. A., 1988, Transmission of raspberry ringspot, tomato black ring and peach rosette mosaic viruses by an Ontario population of Longidorus elongatus, Can. f. Plant Pathol. 10:1.

Allen, W. R., Van Schagen, J. G., and Eveleigh, E. S., 1982, Transmission of peach rosette mosaic virus to peach, grape, and cucumber by Longidorus diadecturus obtained from diseased orchards in Ontario, Can. f. Plant Pathol. 4:16.

Allen, W. R., Van Schagen, J. G., and Ebsary, B. A., 1984, Comparative transmission of the peach rosette mosaic virus by Ontario populations of Longidorus diadecturus and Xiphinema americanum (Nematoda:Longidoridae), Can. f. Plant Pathol. 6:29.

Bergeson, G. B., Athow, K. L., Laviolette, F. A., and Thomasine, M., 1964, Transmission, movement and vector relationships of tobacco ringspot virus in soybean, Phytopathology 54:723.


Bleve-Zacheo, T., Zacheo, G., Lamberti, F., and Arrigoni, 0., 1977, Cell wall breakdown and cellular response in developing galls induced by Longidorus apulus, Nematol. Medit. 5:305.

Blok, V. c., Wardell, J., Jolly, C. A., Manoukian, A., Robinson, D. J., Edwards, M. 1., and Mayo, M. A., 1992, The nucleotide sequence of RNA-2 of raspberry ringspot nepovirus, J. Gen. Viral. 73:2189.

Breece, J. R., and Hart, W. H., 1959, A possible association of nematodes with the spread of peach yellow bud mosaic virus, Plant Dis. Rep. 43:989.

Brown, D. J. F., 1985, The transmission of two strains of strawberry latent ringspot virus by populations of Xiphinema diversicaudatum (Nematoda:Dorylaimidal, Nematol. Medit. 13:217.

Brown, D. J. F., 1986a, The transmission of two strains of arabis mosaic virus from England by populations of Xiphinema diversicaudatum from ten countries, Rev. Nematol. 9:82.

Brown, D. J. F., 1986b, Transmission of virus by the progeny of crosses between Xiphinema diversicaudatum (Nematoda:Dorylaimoideal from Italy and Scotland, Rev. Nematol. 9:71.

Brown, D. J. F., 1989, Viruses transmitted by nematodes, EPPO/OEPP Bull. 19:453. Brown, D. J. F., and Boag, B., 1975, LongidoIUs macrasoma, Commonw. Inst. Helminthol. Descr.

Plant Parasitic Nematodes, Vol. 5, No. 67. Brown, D. J. F., and Boag, B., 1977, LongidoIUs attenuatus, Commonw. Inst. Helminthol. Descr.

Plant Parasitic Nematodes, Vol. 7, No. 101. Brown, D. J. F., and Halbrendt, J. M., 1992, The virus vector potential of Xiphinema americanum

and related species, J. Nematol. 24:584. Brown, D. J. F., and Taylor, C. E., 1981, Variazioni nella trasmissione di virus tra popolazioni di

nematodi vettori Longidoridae, Atti della Soc. Ital. Nematol., Giorante Nematol., Firenze, 1979, 191.

Brown, D. J. F., and Taylor, C. E., 1987, Comments on the occurrence and geographical distribution of longidorid nematodes in Europe and the Mediterranean region, Nematol. Medit. 15:333.

Brown, D. J. F., and Trudgill, D. 1., 1983, Differential transmissibility of arabis mosaic and strawberry latent ringspot viruses by three populations of Xiphinema diversicaudatum (Nematoda:Dorylaimidal from Scotland, Italy and France, Rev. Nematol. 6:229.

Brown, D. J. F., Murant, A. F., and Trudgill, D. 1., 1989a, Differences between isolates of the English serotype of tomato black ring virus in their transmissibility by an English population of LongidoIUs attenuatus (Nematoda: Dorylaimoideal, Rev.Nematol. 12:51.

Brown, D. J. F., Ploeg, A. T., and Robinson, D. J" 1989b, A review of reported associations between TrichodoIUS and ParatrichodoIUS species (Nematoda: Trichodoridael and tobraviruses with a description of laboratory methods for examining virus transmission by trichodorids, Rev. Nematol. 12:235.

Brown, D. J. F., Halbrendt, J. M., Robbins, R. T., and Vrain, T. c., 1993, Transmission of nepoviruses by Xiphinema americanum-group nematodes, J. Nematol. 25:349.

Brown, D. J. F., Grunder, J., Hooper, D. J., Klingler, J., and Kunz, P., 1994a, LongidoIUs arthensis sp.n. (Nematoda:Longidoridael, a vector of cherry rosette disease caused by a new nepovirus in cherry trees in Switzerland, Nematologica 40:133.

Brown, D. J. F., Halbrendt, J. M., Jones, A. T., Vrain, T. c., and Robbins, R. T., 1994b, Transmission of three North American nepoviruses by populations of four distinct Xiphinema americanum-group species (Nematoda: Longidoridael, Phytopathology 84:646.

Brown, D. J. F., Vrain, T. C., Jones, A. T., Robertson, W. M., Halbrendt, J. M., and Robbins, R. T., 1994c, Xiphinema bricolensis-a natural vector of three serologically distinguishable strains of tomato ringspot nepovirus, J. Nematol. 26:94.

Cadman, C. H., 1963, Biology of soil-borne viruses, Annu. Rev. Phytopathol. 1:43. Catalano, 1., Savino, V., and Lamberti, F., 1992, Presence of grapevine fanleaf nepovirus in

populations of longidorid nematodes and their vectoring capacity, Nematol. Medit. 20:67. Cohn, E., Tanne, E., and Nitzany, F. E., 1970, Xiphinema italiae, a new vector of grapevine fanleaf

virus, Phytopathology 60:181. Converse, R. H., 1977, Rubus virus diseases important in the United States, HortScience 12:471. Dunleavy, J. M., 1957, The grasshopper as a vector of tobacco ringspot virus in soybean, Phyto

pathology 47:681.


Eveleigh, E. S., and Allen, W. R., 1982, Description of Longidorus diadecturus n.sp. (Nematoda: Longidoridae), a vector of the peach rosette mosaic virus in peach orchards in southwestern Ontario, Canada, Can. r. Zool. 60:112.

Flegg, J. J. M., 1969, Tests with potential nematode vectors of cherry leaf-roll virus, Rep. East Malling Res. Stn. 1968:155.

Forer, 1. B., Hill, N., and Powell, C. A., 1981, Xiphinema rivesi, a new tomato ringspot vector, Phytopathology 71:874.

Fritzsche, R., 1964, Untersuchungen uber die Virusubertragung durch Nematoden, Wissenschaftliche Zeitschrift der Universitiit Rostock Reihe Mathematik und Naturwissenschaften 13:343.

Fritzsche, R., and Kegler, H., 1964, Die Dbertragung des Blattrollvirus der Kirsche (cherry leafroll virus) durch Nematoden, Naturwissenschaften 51:299.

Fritzsche, R., and Kegler, H., 1968, Nematoden als vektoren von viruskrankheiten der obstgewachse, TagBer. dt. Akad. LandwWiss. Berl. 97:289.

Fritzsche, R., and Thiele, S., 1979, Eignung von Xiphinema-Herkunften aus der UdSSR und der DDR zur Dbertragung des Arabis-Mosaik-Virus, Nachrichtenbl. Pflanzenschutz DDR. 33:103.

Fulton, J. P., 1962, Transmission of tobacco ringspot virus by Xiphinema americanum, Phytopathology 52:375.

Fulton, J. P., 1967, Dual transmission of tobacco ringspot and tomato ringspot virus by Xiphinema americanum, Phytopathology 57:535.

Gibbs, A. J., and Gower, J. c., 1960, The use of a multiple-transfer method in plant virus transmission studies-some statistical points arising in the analysis of results, Ann. Appl. Biol. 48:75.

Griffiths, B. S., and Robertson, W. M., 1983, Nuclear changes induced by the nematode Longidorus elongatus in root-tips of ryegrass Lolium perenne, Histochem. r. 15:927.

Griffiths, B. S., and Robertson, W. M., 1984, Nuclear changes induced by the nematode Xiphinema diversicaudatum in root-tips of strawberry, Histochem. r. 16:265.

Griffiths, B. S., and Trudgill, D. 1., 1983, A comparison of the generation times and rates of growth of Xiphinema diversicaudatum and Longidorus elongatus on a good and a poor host, Nematologica 29:78.

Griffiths, B. S., Robertson, W. M., and Trudgill, D. 1., 1982, Nuclear changes induced by the nematodes Xiphinema diversicaudatum and Longidorus elongatus in root-tips of perennial ryegrass Lolium perenne, Histochem. r. 14:719.

Halbrendt, J. M., and Brown, D. J. F., 1992, Morphometric evidence for three juvenile stages in some species of Xiphinema americanum sensu lato, r. Nematol. 24:305.

Halbrendt, J. M., and Brown, D. J. F., 1993, Aspects of biology and development of Xiphinema americanum and related species, r. Nematol. 25:355.

Harrison, B. D., 1964, Specific nematode vectors for serologically distinctive forms of raspberry ringspot and tomato black ring viruses, Virology 22:544.

Harrison, B. D., 1967, The transmission of strawberry latent ringspot virus by Xiphinema diversicaudatum (Nematoda), Ann. Appl. Biol. 60:405.

Harrison, B. D., and Cadman, C. H., 1959, Role of a dagger nematode (Xiphinema sp.) in outbreaks of plant disease caused by arabis mosaic virus, Nature 184:1624.

Harrison, B. D., and Murant, A. F., 1978, Nematode transmissibility of pseudorecombinant isolates of tomato black ring virus, Ann. Appl. Biol. 86:209.

Harrison, B. D., and Winslow, R. D., 1961, Laboratory and field studies on the relation of arabis mosaic virus to its nematode vector Xiphinema diversicaudatum (Micoletzky), Ann. Appl. Biol. 49:621.


Harrison, B. D., Murant, A. F., Mayo, M. A., and Roberts, I. M., 1974a, Distribution of determinants for symptom production, host range and nematode transmissibility between the two RNA components of raspberry ringspot virus, r. Gen. Virol. 22:233.


Harrison, B. D., Robertson, W. M., and Taylor, C. E., 1974b, Specificity of retention and transmission of viruses by nematodes, T. Nematol. 6:155.

Hewitt, W. B., Raski, D. J., and Goheen, A. c., 1958, Nematode vector of soil-borne fanleaf virus of grapevine, Phytopathology 48:586.

Hooper, D. J., 1973, Longidorus elongatus, Commonw: Inst. Helminthol. Descr. Plant Parasitic Nematodes, Vol. 2, No. 30.

Hooper, D. J., 1975, Virus vector nematodes-taxonomy and general introduction, in: Nematode Vectors of Plant Viruses (F. Lamberti, C. E. Taylor, and J. w. Seinhorst, eds.), pp. 1-14, Plenum Press, New York.

Hoy, J. W., Mircetich, S. M., and Lownsbery, B. F., 1984, Differential transmission of prunus tomato ringspot virus strains by Xiphinema californicum, Phytopathology 74:332.

Iwaki, M., and Komuro, Y., 1974, Viruses isolated from narcissus (Narcissus spp.) in Japan. V. Arabis mosaic virus, Ann. Phytopathol. Soc. Tapan 40:344.

Jha, A., and Posnette, A. F., 1959, Transmiss~on of a virus to strawberry plants by a nematode (Xiphinema sp.), Nature 184:962.

Jones, A. T., Mitchell, M. J., and Brown, D. J. F., 1989, Infectibility of some new raspberry cultivars with arabis mosaic and raspberry ringspot viruses and further evidence for variation in British isolates of these two viruses, Ann. Appl. Biol. 115:57.

Jones, A. T., Brown, D. J. F., McGavin, W., Riidel, M., and Altmayer, B., 1994, Properties of an unusual isolate of raspberry ringspot virus from grapevine in Germany and evidence for its possible transmission by Paralongidorus maxim us, Ann. Appl. Biol. 124:283.

Jones, R. A. c., 1981, The ecology of viruses infecting wild and cultivated potatoes in the Andean region of South America, in: Pests, Pathogens and Vegetation (J. M. Thresh, ed.), pp. 89-107, Pitman, London.

KIos, E. J., Fronek, F., Knierim, J. A., and Cation, D., 1967, Peach rosette mosaic transmission and control studies. Q. Bull. Mich. St. Univ. Exp. Stn. 49:287.

Komuro, Y., and Iwaki, M., 1968, [Bean yellow mosaic virus and tobacco ringspot virus isolated from crotalaria (Crotalaria spectabilislJ, Ann. Phytopathol. Soc. Tapan 34:7.

Lamberti, F., and Bleve-Zacheo, T., 1979, Studies on Xiphinema americanum sensu lato with description of fifteen new species (Nematoda, Longidoridae), Nematol. Medit. 7:5l.

Lamberti, F., Taylor, C. E., and Seinhorst, J. W., 1975 (eds.), Nematode Vectors of Plant Viruses, Plenum Press, New York.

Lister, R. M., 1964, Strawberry latent ringspot: A new nematode-borne virus, Ann. Appl. Biol. 54:167.

Mali, V. R., Vanek, G., and Bojnansky, V., 1975, Transmission by nematodes of some grapevine viruses occurring in Czechoslovakia and Hungary, Polnohospodarstvo 3:l.

Martelli, G. P. 1975, Some features of nematode-borne viruses and their relationships with the host plants, in: Nematode Vectors of Plant Viruses (F. Lamberti, C. E. Taylor, and J. W. Seinhorst, eds.), pp. 223-252, Plenum Press, New York.

Mayo, M., Robertson, W. M., Legorburu, F. J., and Brierley, K. M., 1994, Molecular approaches to an understanding of the transmission of plant viruses by nematodes, in: Advances in Molecular Plant Nematology (F. Lamberti, C. De Giorgi, and D. McK. Bird, eds.), pp. 277-293, Plenum Press, New York.

McElroy, F. D., Brown, D. J. F., and Boag, B., 1976, The virus-vector and damage potential, morphometrics and distribution of Paralongidorus maximus, T. Nematol. 9:122.

McGuire, J. M., 1964a, Efficiency of Xiphinema americanum as a vector of tobacco ringspot virus, Phytopathology 54:799.

McGuire, J. M., 1964b, Serial transfer of Xiphinema americanum as a tool for studying transmission of tobacco ringspot virus, Phytopathology 54:900.

McGuire, J. M., 1982, Nematode transmission of viruses, in: Nematology in the Southern Region of the United States (R. D. Riggs, ed.), pp. 190-192, Southern Cooperative Series Bulletin No. 276, University of Arkansas, Fayetteville.

McGuire, J. M., Kim, K. S., and Douthit, L. B., 1970, Tobacco ringspot virus in the nematode Xiphinema americanum, Virology 42:212.


McNamara, D. G., 1978, Studies on the Ability of the Nematode Xiphinema diversicaudatum (Micol.) to Transmit Raspberry Ringspot Virus and to Survive in Plant-Free Soil, PhD. Thesis, University of Reading, Reading, England.

Messieha, M., 1969, Transmission of tobacco ringspot virus by thrips, Phytopathology 59:943.

Murant, A. E, 1983, Seed and pollen transmission of nematode-borne viruses, Seed Sci. Technoi. 11:973.

Murant, A. E, 1989, Nepoviruses, in: Plant Protection and Quarantine, Vol. II. Selected Pests and Pathogens of Quarantine Significance (R. P. Khan, ed.), pp. 44-57, CRC Press, Boca Raton, F1.

Murant, A. E, and Taylor, C. E., 1965, Treatment of soil with chemicals to prevent transmission of tomato black ring and raspberry ringspot viruses by Longidorus elongatus (De Man), Ann. Appi. Bioi. 55:227.

Murant, A. E, Taylor, C. E., and Chambers, J. C., 1968, Properties, relationships and transmission of a strain of raspberry ringspot virus infecting raspberry cultivars immune to the common Scottish strain, Ann. Appi. Bioi. 61:175.

Nyland, G., Lownsbery, B. E, Lowe, B. K., and Mitchell, J. E, 1969, The transmission of cherry rasp leaf virus by Xiphinema americanum, Phytopathology 59:111.

Pitcher, R. S., 1975, Chemical and cultural control of nettlehead and related virus diseases of hop, in: Nematode Vectors of Plant Viruses (E Lamberti, C. E. Taylor, and J. W. Seinhorst, eds.1, pp. 447-448, Plenum Press, New York.

Pitcher, R. S., Siddiqi, M. S., and Brown, D. J. E, 1974, Xiphinema diversicaudatum, Commonw. Inst. Helminthoi. Descr. Plant Parasitic Nematodes, Vol. 4, No. 60.

Ploeg, A. T., Brown, D. J. E, and Robinson, D. J., 1990, The association between species of Trichodorus and Paratrichodorus vector nematodes and serotypes of tobacco rattle tobravirus, Ann. Appi. Bioi. 121:619.

Putz, C., and Stocky, G., 1970, Premieres observations sur une souche de strawberry latent ringspot virus trans mise par Xiphinema coxi Tarjan et associee a une maladie du framboisier en Alsace, Ann. Phytopathoi. 2:329.

Rana, G. 1., and Roca, E, 1975, Trasmissione con nematodi del virus latente italiano del carciofo (AILVI, Atti II CongI. Intern. Studi Carciofo, Bari 1974:855.

Rani, S., Verma, H. N., and Verma, G. S., 1969, A virus disease of Petunia hybrida, Plant Dis. Rep. 53:903.

Raski, D. J., and Hewitt, W. B., 1963, Plant parasitic nematodes as vectors of plant viruses, Phytopathology 53:39.

Raski, D. J., Maggenti, A. R., and Jones, N. 0., 1973, Location of grapevine fanleaf and yellow mosaic virus particles in Xiphinema index, J. Nematol. 5:208.

Robbins, R. T., 1993, Distribution of Xiphinema americanum and related species in North America, J. Nematol. 25:344.

Robbins, R. T., and Brown, D. J. E, 1991, Comments on the taxonomy, occurrence and geographical distribution of Longidoridae (Nematoda) in North America, Nematologica 37:395.

Roberts, I. M., and Brown, D. J. E, 1980, Detection of six nepoviruses in their nematode vectors by immunosorbent electron microscopy, Ann. Appi. Bioi. 96:187.

Robertson, W. M., 1975, Ultrastructure of nematode vectors of plant viruses with reference to their feeding apparatus, Rep. Scott. Hortic. Res. Inst. 1974:76

Robertson, W. M., 1976, A possible gustatory organ associated with the odontophore in Longidorus leptocephalus and Xiphinema diversicaudatum, Nematologica 21:443.

Robertson, W. M., and Henry, C. E., 1986, An association of carbohydrate with particles of arabis mosaic virus retained within Xiphinema diversicaudatum, Ann. Appl. Bioi. 109:299.

Robertson, W. M., and Wyss, U., 1983, Feeding processes of virus-transmitting nematodes, in: Current Topics in Vector Research (K. E Harris, ed.1, pp. 271-295, Praeger, New York.

Robertson, W. M., Trudgill, D. 1., and Griffiths, B. S., 1985, Feeding of Longidorus elongatus and L. leptocephalus on root-tip galls of the perennial ryegrass Lolium perenne, Nematologica 30:222.

208 D. J. E BROWN ET AL.

Roca, E, and Lamberti, E, 1981, Longidorus fasciatus, sp.n. from Greece and Italy, Nematol. Medit. 9:175.

Roca, E, Martelli, G. P., Lamberti, E, and Rana, G. L., 1975, Distribution of Longidorus attenuatus Hooper in Apulian artichoke fields and its relationship with artichoke Italian latent virus, Nematol. Medit. 3:91.

Roca, E, Rana, G. L., and Kyriakopoulou, P. E., 1982, Longidorus fasciatus Roca et Lamberti vector of a serologically distinct strain of artichoke Italian latent virus in Greece, Nematol. Medit. 10:65.

Rudel, M., Alebrand, M., and Altmayer, B., 1983, Untersuchungen uber den Einsatz der ELlSATest zum Nachweis verschiedener Rebeviren, Die Wein-Wissenschaft 38:177.

Rumpenhorst, H. J., and Weischer, B., 1978, Histopathological and histochemical studies on grapevine roots damaged by Xiphinema index, Rev. Nematol. 1:217.

Schuster, M. E, 1963, Flea beetle transmission of tobacco ringspot virus in the Lower Rio Grande Valley, Plant Dis. Rep. 47:510.

Siddiqi, M. S., 1973, Xiphinema americanum, Commonw. lnst. Helminthol. Descr. Plant Parasitic Nematodes, Vol. 2, No. 29.

Siddiqi, M. S., 1974, Xiphinema index, Commonw. lust. Helminthol. Descr. Plant Parasitic Nematodes, Vol. 3, No. 45.

Stellmach, G., and Berres, R., 1985, Investigations on mixed infections of nepoviruses in Vitis spp. and Chenopodium quinoa Willd. by means of ELISA, Phytopathol. Medit. 24:125.

Taylor, C. E., 1962, Transmission of raspberry ringspot virus by Longidorus elongatus (de Man) (Nematoda: Dorylaimida), Virology 17:493.

Taylor, C. E., 1968, The association of ringspot viruses with their nematode vectors, C. R. 8th Symp. lnt. Nematologie, Antibes 1965:109.

Taylor, C. E., 1972, Nematode transmission of plant viruses, Pest Articles and News Summaries (PANS) 18:269.

Taylor, C. E., 1980, Nematodes, in: Vectors of Plant Pathogens (K. E Harris and K. Maramorosch, eds.), pp. 375-416, Academic Press, New York.

Taylor, C. E., and Brown, D. J. E, 1974, An adaptable temperature controlled cabinet, Nematol. Medit. 2:171.

Taylor, C. E., and Brown, D. J. E, 1976, The geographical distribution of Xiphinema and Longidorus nematodes in the British Isles and Ireland, Ann. Appl. Biol. 84:383.

Taylor, C. E., and Brown, D. J. E, 1981, Nematode-virus interactions, in: Plant Parasitic Nematodes, Vol. III (B. M. Zuckerman and R. A. Rohde, eds.), pp. 281-301, Academic Press, New York.

Taylor, C. E., and Raski, D. J., 1964, On the transmission of grape fanleaf by Xiphinema index, Nematologica 10:489.

Taylor, C. E., and Robertson, W. M., 1969, The location of raspberry ringspot and tomato black ring viruses in the nematode vector, Longidorus elongatus (De Man), Ann. Appl. Biol. 64:233.

Taylor, C. E., and Robertson, W. M., 1970, Sites of virus retention in the alimentary tract of the nematode vectors, Xiphinema diversicaudatum (Micol.) and X. index (Thorne & Allen), Ann. Appl. Biol. 66:375.

Taylor, C. E., and Robertson, W. M, 1975, Acquisition, retention and transmission of viruses by nematodes, in: Nematode Vectors of Plant Viruses (E Lamberti, C. E. Taylor, and J. W. Seinhorst, eds.), pp. 253-276, Plenum Press, New York.

Taylor, C. E., and Thomas, P. R., 1968, The association of Xiphinema diversicaudatum (Micoletzky) with strawberry latent ringspot and arabis mosaic viruses in a raspberry plantation, Ann. Appl. Biol. 62:147.

Taylor, C. E., Robertson, W. M, and Roca, E, 1976, Specific association of artichoke Italian latent virus with the odontostyle of its vector, Longidorus apulus, Nematol. Medit. 4:23.

Teliz, D., Grogan, R. G., and Lownsbery, B. E, 1966, Transmission of tomato ringspot, peach yellow bud mosaic and grape yellow vein viruses by Xiphinema americanum, Phytopathology 56:658.


Towle, A., and Doncaster, C. C., 1978, Feeding of Longidorus caespiticola on ryegrass, Lolium perenne, Nematologica 24:277.

Trudgill, D. L., 1976, Observations on the feeding behaviour of Xiphinema diversicaudatum, Nematologica 22:417.

Trudgill, D. L., and Brown, D. J. F., 1978a, Frequency of transmission of some nematode-borne viruses, in: Plant Disease Epidemiology IP. R. Scott and A. Bainbridge, eds.), pp. 283-289, Blackwells Scientific Publications, London.

Trudgill, D. L., and Brown, D. J. F., 1978b, Ingestion, retention and transmission of two strains of raspberry ringspot virus by Longidorus macrosoma, f. NematoI. 10:85.

Trudgill, D. L., and Robertson, W. M., 1982, Feeding and salivation behaviour of Xiphinema diversicaudatum and Longidorus elongatus, Nematologica 28:177.

Trudgill, D. L., Brown, D. J. F., and Robertson, W. M., 1981, A comparison of the effectiveness of the four British virus vector species of Longidorus and Xiphinema, Ann. Appl. BioI. 99:63.

Trudgill, D. L., Brown, D. J. F., and McNamara, D. G., 1983, Methods and criteria for assessing the transmission of plant viruses by longidorid nematodes, Rev. N/3matoI. 6:133.

Trudgill, D. L., Robertson, W. M., and Wyss, U., 1991, Feeding behaviour of Xiphinema diversicaudatum, Rev. N/imatol. 14:107.

Valdez, R. B., 1972, Transmission of raspberry ringspot virus by Longidorus caespiticola, L. leptocephalus and Xiphinema diversicaudatum and of arabis mosaic virus by L. caespiticola and X. diversicaudatum, Ann. AppI. BioI. 71:229.

Van Hoof, H. A, 1972, Viruses transmitted by Xiphinema species in the Netherlands, Neth. f. Plant PathoI. 77:30.

Vovlas, N., Inserra, R. N., and Martelli, G. P., 1978, Modificazioni anatomiche indotte da Xiphinema index e Meloidogyne incognita in radici di un ibrido di Vitis vinifera x V rotundifolia, NematoI. Medit. 6:67.

Vrain, T. c., Wakarchuk, D. A, Levesque, C. A, and Hamilton, R. I., 1992, Intraspecific rDNA restriction fragment length polymorphism in the Xiphinema americanum group, Fundam. AppI. NematoI. 15:563.

Weischer, B., and Wyss, U., 1976, Feeding behaviour and pathogenicity of Xiphinema index on grapevine roots, Nematologica 22:319.

Wyss, U., 1981, Ectoparasitic root nematodes: Feeding behaviour and plant cell responses, in: Plant Parasitic Nematodes, Vol. III lB. M. Zuckerman and R. A Rohde, eds.), pp. 325-351, Academic Press, New York.

Wyss, U., and Zunke, U., 1986, Dynamics of giant cell formation, in: Cyst Nematodes IF. Lamberti, c. E. Taylor, and J. W. Seinhorst, eds.), pp. 147-162, Plenum Press, London.

Wyss, U., Lehmann, H., and Jank-Ladwig, R., 1980, Ultrastructure of modified root-tip cells in Ficus carica, induced by the parasitic nematode Xiphinema index, 1. Cell Sci. 41:193.

Wyss, U., Robertson, W. M., and Trudgill, D. L., 1988, Oesophageal bulb function of Xiphinema index and associated root cell responses, assessed by video enhanced light microscopy, Rev. NematoI. 11:253.

Yagita, H., and Komuro, Y., 1972, [Transmission of mulberry ringspot virus by Longidorus martini Merny], Ann. PhytopathoI. Soc. Japan 38:275.

Yassin, A. M., 1968, Transmission of viruses by Longidorus elongatus, Nematologica 14:419.

CHAPTER 8

Nepoviruses: Ecology and Control B. D. HARRISON AND A. F. MURANT

I. INTRODUCTION

Previous chapters have described the general and molecular characteristics of nepoviruses, their importance as pathogens, and their transmission by nematodes. This chapter deals with their survival and spread in field conditions and with ways of minimizing their incidence and so preventing or decreasing losses of crop. Nepoviruses typically are transmitted by nematodes and through seed and pollen. Thus their incidence and importance depend on a range of kinds of interaction between the viruses, host plants, vector nematodes, and environmental factors. Various aspects of the subject were reviewed by Lamberti et al. (1975), Harrison (1977), Taylor (1980) and ~urant (1981, 1983)

II. VIRUS ECOLOGY AND EPIDEMIOLOGY

A. Naturally Occurring Hosts

Nepoviruses are noted for their wide host ranges, which typically include more than half the species tested and contain members of more than 20

B. D. HARRISON • Department of Biological Sciences, University of Dundee, Dundee DDl 4HN, United Kingdom. A. F. MURANT • Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom.

211


plant families. Table I in Chapter 5 lists many naturally infected perennial species, which are crop plants, such as grapevine, cherry, hop, raspberry, and strawberry, or ornamentals, such as forsythia, privet, and narcissus. Other naturally infected species are cultivated annuals, such as soybean, celery, and lettuce. However, nepoviruses also infect many wild plants, including woody perennials such as Sambucus nigra and many species of annual weed (Harrison, 1957, 1958a,b; Tuite, 1960; Murant and Lister, 1967; Hansen et a1., 1974; Ramsdell and Myers, 1978; Powell et a1., 1982). Moreover, the same plant species may be susceptible to several nepoviruses (Chapter 5 and Table I). A result of this ability to infect a range of species is that land infested with viruliferous vector nematodes is likely to carry virus hosts, whether crop or wild species, at all times except immediately after soil cultivations. Indeed, several nepoviruses can be considered to be pathogens that infect primarily wild species but also infect crop species when they are planted at virusinfested sites. However, there are exceptions to this general rule. For example, grapevine fanleaf virus (GFLV) is rarely found in any species other than grapevine (Hewitt et a1., 1970) and the hop strain of arabis mosaic virus (ArMV) seems to be confined to hop (Thresh and Pitcher, 1978).

B. Occurrence and Population Dynamics of Vector Nematodes

The nepoviruses known to have longidorid nematode vectors are listed in Chapter 7, where the relations between the viruses and their vectors are discussed. Points of importance for nepovirus epidemiology are that each virus is typically transmitted by only one or a few closely related species and that viruliferous nematodes can remain infective for weeks or months but do not retain the virus through the molt or pass it on to their progeny.

The occurrence of nematode-transmitted nepoviruses is therefore greatly influenced by the distribution of their specific nematode vectors. Thus, whereas Longidorus elongatus, the natural vector of the Scottish serotypes of raspberry ringspot virus (RRSV) and tomato black ring virus (TBRV), occurs in most parts of Britain, L. macrosoma is found in south central England in association with the English serotype of RRSV, and L. attenuatus occurs mainly in East Anglia along with the English serotype of TBRY. Xiphinema diversicaudatum occurs in several parts of Britain except northern Scotland and is found most frequently in south and southwest England (Taylor and Brown, 1976). The distributions of ArMV and strawberry latent ringspot virus (SLRSV) follow the same pattern as that of X. diversicaudatum, which transmits them both. Indeed, the two viruses are liable to be found at the same sites (Taylor and Thomas, 1968). Similarly, RRSV and TBRV are frequently both detected by bait tests on soil from individual L. elongatus-infested fields in Scotland, and their incidences tend to vary in parallel (Harrison, 1958b). At such sites, both viruses typically are found in

NEPOVIRUSES: ECOLOGY AND CONTROL 213

TABLE 1. Occurrence in Europe of 4 Nepoviruses in 12 Naturally Infected Plant Species

Virus

ArMV RRSV SLRSV TBRV

Crop plants Grapevine (Vitis vinifera) + + + + Lettuce (Lactuca sativa) + + Raspberry (Rubus idaeus) + + + + Strawberry (Fragaria x ananassa) + + + +

Ornamentals Forsythia intermedia + + Ligustrum vulgare + + Narcissus pseudonarcissus + + + +

Wild perennials Robinia pseudo-acacia + + Sambucus nigra + + + +

Annual weeds Caps ella bursa-pastoris + + + + Lamium amplexicaule + + + Stellaria media + + + +

the weed flora, even when the crop being grown is susceptible to only one of them.

The distribution of longidorid vector species in Western Europe was summarized by Brown and Taylor (1987) and that in North America was recorded by Robbins and Brown (1991). Vector species differ in their soil preferences. For example, whereas X. diversicaudatum is found mainly on clay and organic fen soils, L. elongatus occurs on light loams and L. attenuatus on sandy soils. At an infested site, the nematodes tend to be patchily distributed (Harrison and Winslow, 1961; Harrison et a1., 1961; Taylor and Thomas, 1968). Harrison and Winslow (1961) deduced that patches of uncultivated land infested with X. diversicaudatum may increase in radius by about 30 cm a year, and a patch they studied was still identifiable about 30 years later (Fig. 1) (Taylor et a1., 1994).

Different vector species also differ in their host preferences. Xiphinema index multiplies well only on grapevine and fig (Brown and Coiro, 1985), whereas X. diversicaudatum has a much wider host range, which includes many woody species and also herbaceous ones (Thomas, 1970). L. elongatus multiplies on strawberry and some other herbaceous species but not on raspberry (Taylor, 1967), to which it nevertheless transmits RRSV and TBRV. Longidorids may need a year or more to complete their life cycles in cool temperate climates, although less in warmer ones. Individuals may live for at least 2 years (Flegg, 1968). In Scotland, Taylor (1967) found that population densities of L. elongatus increased two- to fourfold per year on good hosts but gradually declined on raspberry; most of the nematodes occurred in the top

so

40 -

l

30 -

I

(/)

'0

"- en >-

20

10 o

1960

x. di

.".,s

/cau

diltu

m I

200g

N1

_ >1

60

_ >

80

.:·.-4

> 4

0

t ~;,,

---I

> 2

0

,:-;:

:, . .:

) >

2

I""';

--';-'

__ 'C".;:~

'"",

\",:~

c::::

J <

2

Hed

ge

~

tv

1991

X. d

ivllr

sica

iJda

tum

I 20

0g

_>

50

_>

25

IZ.E!

:l > 1

2

e;:

~-:l

>

6

Ibt'

~1

> 2

c::::

J <

1

Hed

ge

FIG

UR

E 1

. D

istr

ibu

tio

n o

f Xip

hine

ma

dive

rsic

auda

tum

in a

woo

dlan

d ad

jace

nt to

a h

edge

row

in 1

960

and

1991

. (F

rom

T

aylo

r et

ai.,

199

4. R

epro

duce

d w

ith

per

mis

sion

.)

t-J ~

~

~ ~ ::z:; ~

en o Z

;,. a ?> !'Ii ~

c::: ~ ~


20 cm of soil, but some penetrated to a depth of at least 60 cm. In comparison, X. index reached a depth of at least 3 m in vineyards in California (Raski et a1., 1965). The picture that emerges from these studies is of vectors that are slow to invade soil and that multiply at a modest rate but nevertheless produce very persistent infestations.

C. Patterns of Disease Outbreaks

In crops, infected plants typically occur in patches, which coincide with the horizontal distribution of vector nematodes (Harrison and Cadman, 1959; Harrison and Winslow, 1961; Harrison et a1., 1961). These patches recur in successive crops and most of them enlarge only slowly, at speeds consistent with the rate of migration (e.g., 30 cm/year), or movement by cultivation, of the nematodes. In some instances, however, virus spread across a field is more rapid. This can happen when a perennial crop that produces a large root system is planted in a field already infested throughout with vector nematodes, and the virus is introduced in a small proportion of the planting material. In these circumstances, the hop strain of ArMV can apparently be transmitted by X. diversicaudatum from one hop plant to its neighbor, and then within a year can be translocated through the plant to the other side of its root system, a distance of several meters (Thresh and Pitcher, 1978).

A third kind of distribution of virus-infected plants is found when a proportion of the planting material is infected but the field is free of vector nematodes. When this happens, infected plants are scattered through the planting. Where the planting material is free of vector nematodes, the virus spreads no further. However, where rooted plants infested with vector nematodes are planted and the nematodes become established in the new planting, slow spread of the virus can be expected. The worldwide occurrence of GFLV and its vector, X. index, in vineyards probably has resulted from the introduction of both with planting stock. Any introduction of virus in planting material, with or without vector nematodes, can disseminate a nepovirus over long distances and is of particular importance with vegetatively propagated species such as grapevine, raspberry, and narcissus but can also apply to virus-infected seed, including weed seed.

Tobacco ringspot virus (TRSV), although transmitted by nematodes of the x. americanum group (see Chapter 7), is reported also to be transmitted in some soybean crops by aerial vectors, namely thrips (Messieha, 1969) and spider mites (Thomas, 1969). Spread of the bud blight disease caused by TRSV into soybean crops seems to occur largely from the crop margins adjacent to areas where wild hosts eire infected, is too rapid to be explained by nematode transmission, and occurs in fields in which x. americanum could not be found (Bergeson et a1., 1964). The above accessory vectors may be responsible for this spread.


D. Natural Transmission through Seed and Pollen

Seed transmission has been reported for at least 19 nepoviruses (Murant, 1983) and probably occurs with all of them in at least some hosts. The seed may be infected by virus introduced either'through the ovule or through pollen, though the frequency of infection tends to be greatest when both are infected (Lister and Murant, 1967). The proportion of seedlings infected in these ways can vary greatly but is often large and may approach 100% (Murant, 1983). Frequency of transmission from infected mother plants to progeny seedlings depends on the host species and genotype, the virus and virus strain, the time of infection of mother plants relative to flowering, and the environmental conditions. For example, TBRV was transmitted to 83 % of progeny seedlings of Caps ella bursa-pastoris but to only 2 % of those of Malling Exploit raspberry. Conversely, RRSV was transmitted to only 3% of progeny of C. bursa-pastoris but to 18 % of those of Malling Exploit raspberry (Lister and Murant, 1967). In the weed species Stellaria media, an isolate of the Scottish serotype of TBRV was transmitted to 32-42 % of seedlings, but an isolate of the German serotype was transmitted to only 0-7%; and whereas ArMV was more frequently seed-transmitted when mother plants were grown at 14°C (34%) than at 22 °C (11%), the reverse was true for SLRSV (0 and 29%, respectively) (Hanada and Harrison, 1977). In soybean, TRSV was seed-transmitted at high frequency when the mother plants were inoculated at least 2 weeks before the seeds were set, but at much lower frequencies when the interval was shorter (Owusu et a1., 1968).

Pollen from parents infected with RRSV or TRSV germinated less well than that from healthy parents and competed poorly with virus-free pollen during fertilization (Lister and Murant, 1967; Yang and Hamilton, 1974). Moreover, the presence of TBRV in either male or female parents somewhat delayed and decreased seed germination in the weed Spergula arvensis (Lister and Murant, 1967), and parental infection with CLRV had a large detrimental effect on seed germination in birch (Betula pendula) (Cooper et a1., 1984). These apparently ecologically disadvantageous effects are offset by others, with the result that transmission through seed and pollen is of great importance in nepovirus survival and spread. First, in several species, seedlings infected through seed are symptomless (Lister and Murant, 1967), presumably because their meristematic tissue was invaded before seed germination and they are therefore in the recovery phase (Benda and Naylor, 1958) of infection. Support for this idea was obtained by Hanada and Harrison (1977) who grew symptomless TBRV-infected tobacco and Chenopodium quinoa seedlings at 33°C for the third and fourth weeks after germination to free their meristems from virus; when the plants were subsequently returned to about 20°C, they developed the severe "shock" symptoms that are the characteristic first response to systemic infection. These findings carry the implication that seedlings (including weed seedlings) infected through seed are well able to compete with their virus-free counterparts. Second, virus-


infected seed can retain the ability to produce infected seedlings for long periods. For example, 6-year-old seed of Stellaria media produced seedlings with 5 % and 11 % infection by RRSV and ArMV, respectively (Lister and Murant, 1967). Dormant seeds, especially weed seeds, are therefore important potential reservoirs of nepoviruses. The extent of such reservoirs was estimated by Murant and Lister (1967), who collected soil from the sites of nepovirus outbreaks in Britain, dried it to kill any vector nematodes, and then allowed dormant weed seeds to germinate after remoistening the soil. Seedlings of 11 of 20 weed species were infected with TBRV at eight sites, whereas seedlings of only 1 of 11 species were found to be infected with ArMV at six sites. TBRV occurred in up to 30% of Stellaria media seedlings, but the incidence was usually lower. Moreover, only a few infected seedlings were found of some species, such as Spergula arvensis and Senecio vulgaris, in which the virus is readily seed-transmitted following artificial inoculation. Clearly, dormant weed seeds are important virus reservoirs at sites of nepovirus outbreaks and are potential vehicles for dissemination of the viruses to other sites.

For more than half the viruses currently classified as nepoviruses, the vector is unknown (Chapter 5; Murant, 1983, 1989). In many instances this is because the necessary research has not been done, but at least a few nepoviruses appear to spread without the aid of a nematode vector. For example, cherry leaf roll virus (CLRV) does not occur in patches in crops as would be expected if transmission were through soil, and an early claim for transmission by nematodes in laboratory experiments (Fritzsche and Kegler, 1964) was not confirmed in very thorough tests by Jones et al. (1981). Moreover, CLRV isolates from different genera of natural hosts (Betula, Comus, Tuglans, Prunus, Rheum, Rubus, Sambucus, Ulmus) are serologically different (Jones, 1976; Cooper and Edwards, 1980), whereas the serological variability of most other nepoviruses is correlated with vector specificity and not with host genus. This strongly suggests that CLRV does not have a nematode vector and depends for transmission on some highly genus-specific plant factor. Seed transmission of CLRV is known to occur in many natural and experimental hosts (Murant, 1983) and the virus is known to enter seed from pollen in elm (Callahan, 1957) and birch (Cooper, 1976). This has led to suggestions (Cooper and Atkinson, 1975; Cooper, 1976; Jones, 1976) that CLRV carried in pollen might spread to infect the pollinated plant. Strong circumstantial evidence for this was presented by Mircetich et al. (1980), who found that CLRV spreads in walnut (Juglans regia) in California, c'ausing a black line to develop in woody tissue at the interface between the infected scion cultivar and CLRV-immune rootstock. Obviously, immunity of the rootstock rules out spread through the soil. Moreover, spread was observed only when the trees were old enough to flower, and the rate of spread in different cultivars was correlated with the extent to which the pollen shedding of infected cultivars coincided with the receptivity of pistils of healthy cultivars.


With blueberry leaf mottle virus (BLMV), too, there is no evidence for a nematode vector (Childress and Ramsdell, 1986) but good evidence for spread from one blueberry (Vaccinium corymbosum) plant to another in pollen carried by bees (Childress and Ramsdell, 1987; Boylan-Pett et al., 1991). BLMV infectivity associated with pollen carried by contaminated bees was maintained for at least 10 days. Infected pollen was passed from bee to bee in a hive, between bees from nearby hives, and occasionally between bees from hives 600 m apart. However, most virus spread in blueberry is likely to be to plants close to virus sources. Grapevine Bulgarian latent virus (GBLV), which is serologically related to BLMY, seems another possible candidate for transmission in association with pollen. A fourth is artichoke yellow ringspot virus (AYRSV), which experimentally was transmitted by pollination to healthy plants of Nicotiana clevelandii (Kyriakopoulou et al., 1985). Spread through pollen seems likely to be more important with viruses that, like CLRY, infect long-lived perennial plants than with viruses that infect annuals, which might become infected only towards the end of their life span.

So far, infection of plants pollinated with virus-carrying pollen has been reported only for viruses that seem to lack nematode vectors: no evidence was obtained that RRSV or TBRV are transmitted in raspberry or strawberry pollen to the plant pollinated, though they were transmitted to seed (Lister and Murant, 1967). However, the possibility that some nepoviruses might spread in both ways cannot be excluded.

E. Interplay of Virus, Vector, and Host Plant Factors

The ecology and epidemiology of nepoviruses is the result of the subtle interplay of viral, vector, and plant factors. For example, raspberry is infected by RRSV and TBRV in Scotland but is a poor host of their vector, L. elongatus, which in turn multiplies freely on some grass species that are not hosts of the viruses. The most important hosts for survival and spread of the viruses at a site are therefore those, such as the weed Stellaria media, that are good hosts of L. elongatus and are readily infected by the viruses, and in which a large percentage of progeny seedlings become infected through seed (Lister and Murant, 1967; Taylor, 1967). The complementarity of seed transmission and nematode transmission is well seen in this system. Thus in experiments in glasshouse conditions, a noninfective population of L. elongatus in field soil collected from an outbreak area became infective when seed of various weed species was permitted to germinate; conversely, an infective L. elongatus population became noninfective when all germinating seedlings were removed (Murant and Lister, 1967). Moreover, seed transmission in weed species in field conditions seems commoner for nepoviruses that persist in their vector for only a few weeks, such as RRSV in L. elongatus, than for those that persist for many months, such as ArMV and SLRSV in X. diversicaudatum, despite the fact that both ArMV and SLRSV have wide host ranges and are


readily seed-transmitted when tested experimentally in glasshouse conditions. A further illustration of the complementarity of nematode and seed transmission is seen in the natural dissemination of nepoviruses. Nematodes can transmit viruses from plant to plant but, unlike aerial vectors, do not disseminate viruses efficiently from place to place. Instead, nepoviruses are carried to new sites in seeds and become established when such seeds germinate in soil already infested with noninfective vector nematodes. With those nepoviruses, such as CLRV and BLMV, that are transmitted in pollen to the plant pollinated, this ability substitutes for nematode transmission, but with the difference that it is confined to plants that are of the same species as the virus source.

Studies with pseudorecombinant isolates of RRSV and TBRV showed that whereas the genetic determinant for nematode transmissibility is in RNA-2 (Harrison et a1., 1974; Harrison and Murant, 1977), seed transmissibility is largely determined by RNA-1 (Hanada and Harrison, 1977). Each genome segment is therefore subject to different but complementary selection pressures in field conditions, as well as for compatibility with the other segment.

III. CONTROL

In general, plant viruses are controlled by eliminating virus sources, preventing transmission from plant to plant, and growing virus-resistant cultivars. Nepoviruses are no exception to this rule, and methods based on each of these approaches are available.

A. Removing Virus Sources

An important contribution to nepovirus control can be made by using virus-free planting material, especially for establishing long-lived vegetatively propagated perennial crops. Schemes operating in many countries are designed to test mother plants for freedom from nepovirus infection, to propagate them in conditions that will prevent infection, and to certify the resulting stocks as superior planting material. For example, nepovirus-free grapevines are produced on a large scale in California (Nyland and Goheen, 1969), and a well-developed scheme for production and certification of raspberry stocks has operated in Scotland for many years. Where no virus-free mother plants of a cultivar are available, they can often be produced by heat treatment. Many grapevine cultivars were freed from GFLV in this way (Galzy, 1961). For viruses that are seed-borne in crop plants, such as TRSV in soybean (Desjardins et a1., 1954) and SLRSV in parsley (Bos et a1., 1979; Hanson and Campbell, 1979), seed stocks should be checked for freedom from infection.


Nepovirus-free planting material can be used effectively only when the fields to be planted are free from virus-carrying nematodes. Preplanting tests of the status of fields are therefore desirable. Also, removal of naturally infected crop or weed plants from a field, and replanting with healthy stock, is usually ineffective because nepoviruses are retained by their vectors for long periods and may also survive in dormant weed seeds. Attempts to remove the reservoirs of RRSV and TBRV in a strawberry plantation by weed control with the herbicide chloroxuron were largely ineffective in the short term (Taylor and Murant, 1968), but herbicide treatment might be beneficial if done with more effective chemicals or if continued for several years.

Potential quarantine problems posed by nepoviruses were reviewed by Murant (1989). He considered that the extent to which they pose an actual problem was not clear, but that much may be achieved by banning the importation of soil and of all rooted plants with soil attached, prohibiting the importation of propagating material or seed likely to contain specific nonindigenous nepoviruses unless it is accompanied by a certificate to show they were not detected in normally reliable tests, and testing imported material to ensure that it is indeed free from infection. This set of precautions would greatly decrease the chance of importing nepoviruses but may be difficult to impose in practice in view of the continuing expansion of international trade in plants and seeds.

B. Agronomic Methods

These are of most use for controlling nepoviruses that lack weed hosts. For example, incidence of the hop strain of ArMV in a subsequently planted hopyard was greated decreased by fallowing the land for 2 years after removing a previous infected crop (McNamara et a1., 1973). Infective X. diversicaudatum presumably lose their charge of virus during this period, and the virus seems to have no field host other than hop. The same treatment is, however, less effective for GFLV in grapevine, because grape roots and x. index both can survive for at least 5 years after grubbing an infected vineyard (Raski et a1., 1965).

When crops such as grapevine are established with rooted planting material, precautions should be taken to ensure that vector nematodes are not introduced with the plants. Root systems can be washed free of soil and, preferably, should also be dipped in nematicide.

C. Application of Nematicides to Soil

Nematicides are expensive to apply to soil and are therefore uneconomic for controlling nepovirus infection in most annual crops. They have greater potential for perennial plantation crops, providing that they reduce the


nematode population to very low levels «1 % of the initial numberJ throughout the depth of soil penetrated by roots. Despite these tough requirements, beneficial effects have been recorded, especially after applying soil fumigants such as dichloropropane-dichloropropene (D-DJ, dichloropropene (Telone), or methyl bromide, but nonfumigant nematicides have proved to be less useful. The efficacy of fumigants depends on the soil type, its moisture and organic matter contents, the soil temperature, and whether or not the soil surface is sealed, for example, with water or by covering it with plastic sheeting (Thomason and McKenry, 1975). They also have the advantage of killing growing plants and dormant seeds. In British conditions, D-D was most effective when applied in autumn and sealed in the soil over the winter. For example, the incidence of ArMV in strawberry was decreased from 65 to 3% by an autumn preplanting application of D-D that killed >99% of X. diversicaudatum in soil to a depth of 30 cm (Harrison et a1., 1963). In a woodland soil, a X. diversicaudatum population was also decreased by an autumn D-D treatment to <1 % of its initial density, and the treatment was effective to a depth of at least 70 cm. The X. diversicaudatum populations at such sites recovered only slowly, and it was concluded that, provided the nematicide decreases the x. diversicaudatum population to <1 nematode/2 liters soil, the treatment should not need to be repeated for several years (Harrison et a1., 1963).

Murant and Taylor (1965), in a 4-year field trial, found that D-D killed 95% of L. elongatus and prevented transmission of RRSV and TBRV to a subsequently planted strawberry crop, even though the nematode numbers increased about fourfold each year after the strawberries were planted. Nematodes in untreated control plots lost infectivity after an overwintering fallow but regained it in the spring from weed seedlings germinating from infected seed and transmitted the viruses to 36% of the strawberry plants in the first season. In the same experiments, soil treatment with the fungicide pentachloronitrobenzene, which had been shown to abolish the infectivity of soils in pot experiments (Cadman and Harrison, 1960J, also gave useful control.

Fumigation with methyl bromide, D-D, or dichloropropene before planting has also proved to be useful for control of L. elongatus in narcissus fields (Mowat, 1980) and of X. index in vineyards with shallow soil (Vuittenez, 1958), but not where the vineyard soil is deep and grapevine roots and x. index at depths of 1 m or more can survive the treatment and persist for at least 5 years (Hewitt et a1., 1962).

D. Virus-Resistant Cultivars

Where virus-resistant cultivars exist or genes for virus resistance are known, a cheap and effective method of nepovirus control is available. For example, in raspberry, which in Britain is subject to infection by four nepo-


viruses, dominant major genes for resistance to the three commonest of these viruses have been found (Jennings, 1964) (Table II). Each resistance is expressed as immunity to infection by graft or nematode inoculation. Where used in the field, such resistant cultivars have in most instances given excellent results. However, resistance-breaking strains of virus have appeared in a few fields (Table II), and the existence of such strains limits the practical value of resistant cultivars. The best studied of the resistancebreaking isolates is the LG strain of RRSV, which infects raspberry cultivars such as Lloyd George that are immune to the common Scottish strain. Although readily transmitted by L. elongatus, the LG strain has not become widespread, probably because it is poorly seed-transmitted in weeds as compared with other strains, and so is at a selective disadvantage except where resistant raspberry cultivars are planted (Murant et al., 1968). Seed transmissibility and the ability to infect resistant cultivars are both controlled by RNA-1 (Harrison et al., 1974), which by analogy with other nepoviruses presumably has a single open reading frame. Hence the very mutations that confer resistance-breaking possibly also decrease seed transmissibility.

Unfortunately, there are few examples of comparable resistances of other crops to nepoviruses. Programs to screen germ plasm for resistance may therefore be worthwhile. Another approach is to use vector-resistant cultivars or rootstocks. For example, Bouquet (1981) reported that Vitis rotundifolia was not infected by nematode-inoculated GFLV, although it could be infected by graft inoculation. This suggests the possibility of incorporating vector resistance gene(s) from V rotundifolia into commercially acceptable grapevine rootstocks. In contrast, prospects for control of RRSV and TBRV by the use of vector-resistant plants seem unpromising. L. elongatus does not multiply on raspberry (Taylor, 1967), yet it transmits both RRSV and TBRV to raspberry in field conditions and presumably delivers virus inoculum to root

TABLE II. Susceptibility of Red Raspberry (Rubus idaeus) Cultivars to Four Nepovirusesa

Virusb

Cultivar ArMV RRSV SLRSV TBRV

Glen Clova + + + Glen Prosen -(RB) -(RB) • • Leo + Lloyd George -(RB) -(RB) Malling Exploit + + + + Malling Jewel -(RB) + + + Norfolk Giant -(RB) + + aData from Murant 11987) and Jones et al. (1989) b+, Susceptible; -, immune to infection by graft inoculation with stock cultures of the viruses; (RB), resistance-breaking virus isolates found in the field; ., not tested.


cells even if it does not feed to any great extent. Another possibility, for crop plants such as soybean, is to search for resistance to seed transmission. This would prevent the introduction of TRSV into soybean crops with the seed used to establish them.

Finally, some progress has been made towards producing cultivars with transgenic resistance to nepoviruses. Two approaches have been tried. In the first, tobacco plants were transformed with a DNA copy of the coat protein gene of ArMV and were shown to produce particles resembling the RNA-free top component of the virus. The concentration of ArMV coat protein antigen in one of the transformed plant lines was as great as in ArMV-infected Chenopodium quinoa, but the transgenic plants grew normally nevertheless (Bertioli et a1., 1991). When they were inoculated with ArMV particles or ArMV RNA, the inoculated leaves developed necrotic lesions similar to those in control plants. However, the noninoculated tip leaves, instead of developing the systemic necrotic spots, line patterns, and leaf distortion usually induced by ArMV, remained symptomless and contained little ArMV infectivity (Bertioli et a1., 1992). Similar results were obtained with tobacco plants transformed with the coat protein gene of grapevine chrome mosaic virus (GCMV) (Brault et a1., 1993). Coat protein-mediated resistance therefore shows promise but has not yet been applied to a species in which the viruses cause an economically important disease, and the efficacy and durability in field conditions of the resistance conferred are unknown. However, the resistance to ArMV is effective against nematode-inoculated virus (Cooper et a1., 1994), and plants transformed with the two coat protein genes of SLRSV or with the gene encoding the smaller protein alone are resistant both to sap-inoculated and to nematode-inoculated SLRSV (Kreiah et a1., 1996).

The second approach consists of transformation with the DNA copy of a satellite RNA that attenuates symptoms of the helper virus. In work with a small (359-nucleotide) satellite RNA of TRSV, Gerlach et a1. (1987) found that satellite-transformed plants produced small amounts of functional satellite RNA. When these transformed plants were inoculated with satellitefree TRSV, the lesions that developed in inoculated leaves were less necrotic than those caused by TRSV in control plants, large amounts of the satellite RNA were produced, and systemic symptoms, which are normally severe, were almost completely absent. This approach is therefore promising too, but no information is available on its practical application in the field.

E. Integrated Control

Although several of the control measures outlined in previous sections may be effective on their own, the best results are often achieved by combining different approaches. For example, the scheme used to produce virustested stocks of narcissus in Scotland includes the following elements


(Mowat, 1980). Mother plants are selected for freedom from virus by examination for virus like symptoms, serological tests, and inoculation of indicator plants with leaf extracts. Totally infected cultivars are freed from virus by heat treatment or meristem-tip culture. The selected mother plants are then multiplied by advanced propagation methods in a protected environment in presterilized soil. Subsequent propagation is done in fields selected for relative freedom from vector nematodes but nevertheless is fumigated with dichloropropene. The growing crop is inspected for virus like symptoms and any suspect plants plus a selection of others are tested serologically. Crops. that pass all the tests are certified as planting material.

The most appropriate combination of control measures must be tailored to the crop, the virus to be controlled, and the circumstances. For instance, patchy outbreaks of RRSV in raspberry can be controlled by removing diseased plants, treating soil in the affected area with nematicide, and replanting with a virus-resistant cultivar.

IV. CONCLUDING REMARKS

It is clear from this survey that the ecology and epidemiology of nepoviruses involve a fascinating series of adaptations and interactions, which contrast with those applicable to plant viruses that are spread by aerial vectors. Survival of a nepovirus at a site typically involves its transmission by nematodes to wild species and crop plants, and its persistence in vector nematodes and dormant seeds. Spread to other sites involves transport of the virus and/or vector by man with planting material, or dissemination of virusinfected weed seed to sites that are already infested with virus-free populations of vector nematodes. Transmission by nematodes and transmission through seed are complementary elements of the survival systems of many nepoviruses. However, a few nepoviruses are spread in pollen to the pollinated plant instead of by nematodes.

Control of nepoviruses exploits the knowledge gained from the ecological studies and is centered on (1) eliminating the viruses from sites of disease outbreaks (a demanding task), (2) preventing introduction of the viruses and their vectors, especially with planting material, and (3) planting virusresistant cultivars, where they are available and commercially acceptable.

REFERENCES

Benda, G. T. A., and Naylor, A. W., 1958, On the tobacco ringspot disease. III. Heat and recovery, Am. r. Bot. 45:33.

Bergeson, G. B., Athow, K. 1., Laviolette, F. A., and Thomasine, M., 1964, Transmission, move· ment, and vector relationships of tobacco ringspot virus in soybean, Phytopathology 54: 723.

Bertioli, D. J., Harris, R. D., Edwards, M. 1., Cooper, J. I., and Hawes, W. S., 1991, Transgenic plants and insect cells expressing the coat protein of arabis mosaic virus produce empty virus· like particles, r. Cell. Viral. 72:1801.


Bertioli, D. J., Cooper, J. I., Edwards, M. 1., and Hawes, W. S., 1992, Arabis mosaic nepovirus coat protein in transgenic tobacco lessens disease severity and virus replication, Ann. Appl. Biol. 120:47.

Bos, 1., Huttinga, H., and Maat, D. Z., 1979, Parsley latent virus, a new and prevalent seedtransmitted, but possibly harmless virus of Petroselinum crispum, Neth. r. Plant Pathol. 85:125.

Bouquet, A., 1981, Resistance to grape fanleaf virus in muscadine grape inoculated with Xiphinema index, Plant Dis. 65:791.

Boylan-Pett, W., Ramsdell, D. c., Hoopingarner, R. A., and Hancock, J. F., 1991, Honeybee foraging behavior, in-hive survival of infectious, pollen-borne blueberry leaf mottle virus and transmission of the virus in highbush blueberry, Phytopathology 81:1407.

Brault, V., Candresse, T., Le Gall, 0., Delbos, R. P., Lanneau, M., and Dunez, J., 1993, Genetically engineered resistance against grapevine chrome mosaic nepovirus, Plant Mol. Biol. 21:89.

Brown, D. J. F., and Coiro, M. I., 1985, The reproductive capacity and longevity of Xiphinema index (Nematoda: Dorylaimida) from three populations on selected host plants, Rev. Nematol. 8:171.

Brown, D. J. F., and Taylor, C. E., 1987, Comments on the occurrence and geographical distribution of longidorid nematodes in Europe and the Mediterranean region, Nematol. Medit. 15:333.

Cadman, C. H., and Harrison, B. D., 1960, Studies on the behavior in soils of tomato black ring, raspberry ringspot, and arabis mosaic viruses, Virology 10:1.

Callahan, K. 1., 1957, Pollen transmission of elm mosaic virus, Phytopathology 47:5. Childress, A. M., and Ramsdell, D. c., 1986, Lack of evidence for a nematode vector of blueberry

leaf mottle virus (BBLMV), Acta Hortic. 186:87. Childress, A. M., and Ramsdell, D. C., 1987, Bee-mediated transmission of blueberry leaf mottle

virus via infected pollen in highbush blueberry, Phytopathology 77:167. Cooper, J. I., 1976, The possible epidemiological significance of pollen and seed transmission in

the cherry leaf roll virus/Betula spp. complex, Mitteilungen aus der Biologische Bundesanstalt /iir Land und Forstwirtschaft, Berlin-Dahlem 170:17.

Cooper, J. I., and Atkinson, M. A., 1975, Cherry leaf roll virus causing a disease of Betula spp. in the United Kingdom, Forestry 48:193.

Cooper, J. I., and Edwards, M.-1., 1980, Cherry leaf roll virus in Tuglans regia in the United Kingdom, Forestry 53:41.

Cooper, J. I., Massalski, P. R., and Edwards, M.-1., 1984, Cherry leaf roll virus in the female gametophyte and seed of birch and its relevance to vertical virus transmission, Ann. Appl. Biol. 105:55.

Cooper, J. I., Edwards, M. 1., Rosenwasser, 0., and Scott, N. W., 1994, Transgenic resistance genes from nepoviruses: Efficacy and other properties, New Zealand r. Crop Hortic. Sci. 22:129.

Desjardins, P. R., Latterell, R. 1., and Mitchell, J. E., 1954, Seed transmission of tobacco-ringspot virus in Lincoln variety of soybean, Phytopathology 44:86.

Flegg, J. J. M., 1968, Life-cycle studies of some Xiphinema and Longidorus species in southeastern England, Nematologica 14:197.

Fritzsche, R., and Kegler, H., 1964, Die Dbertragung des Blattrollvirus der Kirsche (cherry leafroll virus) durch Nematoden, Naturwissenschaften 51:299.

Galzy, R., 1961, Confirmation de la nature virale du court-noUl~ de la vigne par des essais de thermotherapie sur les cultures in vitro, C. R. Acad. Sci. Paris 253:706.

Gerlach, W. 1., Llewellyn, D., and Haseloff, J., 1987, Construction of a plant disease resistance gene from the satellite RNA of tobacco ringspot virus, Nature 328:802.

Hanada, K., and Harrison, B. D., 1977, Effects of virus genotype and temperature on seed transmission of nepoviruses, Ann. Appl. Biol. 85:79.

Hansen, A. J., Nyland, G., McElroy, F. D., and Stace-Smith, R., 1974, Origin, cause, host range and spread of cherry rasp leaf disease in North America, Phytopathology 64:721.


Hanson, C. M., and Campbell, R. N., 1979, Strawberry latent ringspot virus from "Plain" parsley in California, Plant Dis. Rep. 63:142.

Harrison, B. D., 1957, Studies on the host range, properties and mode of transmission of beet ringspot virus, Ann. Appl. Biol. 45:462.

Harrison, B. D., 1958a, Raspberry yellow dwarf, a soil-borne virus, Ann. Appl. Biol. 46:22l. Harrison, B. D., 1958b, Further studies on raspberry ringspot and tomato black ring, soil-borne

viruses that affect raspberry, Ann. Appl. Biol. 46:57l. Harrison, B. D., 1977, Ecology and control of viruses with soil-inhabiting vectors, Annu. Rev.

Phytopathol. 15:33l. Harrison, B. D., and Cadman, C. H., 1959, Role of a dagger nematode (Xiphinema sp.) in

outbreaks of plant diseases caused by arabis mosaic virus, Nature 184:1624. Harrison, B. D., and Murant, A. F., 1977, Nematode transmissibility of pseudo-recombinant

isolates of tomato black ring virus, Ann. Appl. Biol. 86:209. Harrison, B. D., and Winslow, R. D., 1961, Laboratory and field studies on the relation of arabis

mosaic virus to its nematode vector Xiphinema diversicaudatum (Micoletzky), Ann. Appl. Biol. 49:62l.


Harrison, B. D., Peachey, J. E., and Winslow, R. D., 1963, The use of nematicides to control the spread of arabis mosaic virus by Xiphinema diversicaudatum (Micol.), Ann. Appl. Biol. 52:243.

Harrison, B. D., Murant, A. F., Mayo, M. A., and Roberts, I. M., 1974, Distribution of determinants for symptom production, host range and nematode transmissibility between the two RNA components of raspberry ringspot virus, ,. Gen. Virol. 22:233.

Hewitt, W. B., Goheen, A. C., Raski, D. J., and Gooding, G. V., 1962, Studies on virus diseases of the grapevine in California, Vitis 3:57.

Hewitt, W. B., Martelli, G., Dias, H. F., and Taylor, R. H., 1970, Grapevine fanleaf virus, CMII AAB Descriptions of Plant Viruses No. 28.

Jennings, D. 1., 1964, Studies on the inheritance in the red raspberry of immunities from three nematode-borne viruses, Genetica 35:152.

Jones, A. T., 1976, Serological specificity of isolates of cherry leaf roll virus from different natural hosts, Agriculturae Conspectus Scientificus 39:527.

Jones, A. T., McElroy, F. D., and Brown, D. J. F., 1981, Tests for transmission of cherry leaf roll virus using Longidorus, Paralongidorus and Xiphinema nematodes, Ann. Appl. Biol. 99:143.

Jones, A. T., Mitchell, M. J., and Brown, D. J. F., 1989, Infectibility of some new raspberry cultivars with arabis mosaic and raspberry ringspot viruses, and further evidence for variation in British isolates of these two nepoviruses, Ann. Appl. Biol. 115:57.

Kreiah, S., Edwards, M. 1., Hawes, W. S., Jones, A. T., Brown, D. J. F., McGavin, W. G., Prehaud, C., and Cooper, J. 1., 1996, Coat protein constituents from strawberry latent ringspot virus expressed in transgenic tobacco lessened the rate of systemic invasion following inoculation by nematode vectors or mechanical inoculation, Bur. J. Plant Pathol. (in press).

Kyriakopoulou, P. E., Rana, G. 1., and Roca, F., 1985, Geographic distribution, natural host range, pollen and seed transmissibility of artichoke yellow ringspot virus, Ann. Inst. Phytopathol. Benaki (Athens) 14:139.

Lamberti, F., Taylor, C. E., and Seinhorst, J. w. (eds.), 1975, Nematode Vectors of Plant Viruses, NATO Ser. A: Life Sciences, Vol. 2, Plenum Press, London.

Lister, R. M., and Murant, A. F., 1967, Seed-transmission of nematode-borne viruses, Ann. Appl. Bioi. 59:49.

McNamara, D. G., Ormerod, P. J., Pitcher, R. S., and Thresh, J. M., 1973, Fallowing and fumigation experiments on the control of nettlehead and related virus diseases of hop, Proc. 7th Br. Insectic. Fungic. Conf. 2:597.

Messieha, M., 1969, Transmission of tobacco ringspot virus by thrips, Phytopathology 59:943.


Mircetich, S. M., Sanborn, R. R., and Ramos, D. E., 1980, Natural spread, graft-transmission and possible etiology of walnut blackline disease, Phytopathology 70:962.

Mowat, W. P., 1980, The production of virus-free narcissus stocks in Scotland, Acta Hortic. 109:513.

Murant, A. E, 1981, The role of wild plants in the ecology of nematode-borne viruses, in: Pests, Pathogens and Vegetation (J. M. Thresh, ed.), pp. 237-248, Pitman, London.

Murant, A. E, 1983, Seed and pollen transmission of nematode-borne viruses, Seed Sci. Technol. 11:973.

Murant, A. E, 1987, Raspberry yellow dwarf and associated diseases of Rubus caused by arabis mosaic and strawberry latent ringspot viruses, in: Virus Diseases of Small Fruits (R. H. Converse, ed.), Agriculture Handbook No. 631, p. 204, US Department of Agriculture, Washington, DC.

Murant, A. E, 1989, Nepoviruses, in: Plant Protection and Quarantine, Vol. II, Selected Pests and Pathogens of Quarantine Significance (R. P. Kahn, ed.), pp. 43-57, CRC Press, Boca Raton, F1.

Murant, A. E, and Lister, R. M., 1967, Seed-transmission in the ecology of nematode-borne viruses, Ann. Appl. Biol. 59:63.

Murant, A. E, and Taylor, C. E., 1965, Treatment of soils with chemicals to prevent transmission of tomato blackring and raspberry ringspot viruses by Longidorus elongatus (de Man), Ann. Appl. Biol. 55:227.

Murant, A. E, Taylor, C. E., and Chambers, J., 1968, Properties, relationships and transmission of a strain of raspberry ringspot virus infecting raspberry cultivars immune to the common Scottish strain, Ann Appl. Biol. 61:175.

Nyland, G., and Goheen, A. C., 1969, Heat therapy of virus diseases of perennial plants, Annu. Rev. Phytopathol. 7:331.

Owusu, G. K., Crowley, N. c., and Francki, R. 1. B., 1968, Studies of the seed-transmission of tobacco ringspot virus, Ann. Appl. Biol. 61:195.

Powell, C. A., Forer, 1. B., and Stouffer, R. E, 1982, Reservoirs of tomato ringspot virus in fruit orchards, Plant Dis. 66:583.

Ramsdell, D. c., and Myers, R. 1., 1978, Epidemiology of peach rosette mosaic virus in a Concord grape vineyard, Phytopathology 68:447.

Raski, D. J., Hewitt, W. B., Goheen, A. C., Taylor, C. E., and Taylor, R. H., 1965, Survival of Xiphinema index and reservoirs of fanleaf virus in fallowed vineyard soil, Nematologica 11:349.

Robbins, R. T., and Brown, D. J. E, 1991, Comments on the taxonomy, occurrence and distribution of Longidoridae (Nematoda) in North America, Nematologica 37:395.

Taylor, C. E., 1967, The multiplication of Longidorus elongatus (de Man) on different host plants with reference to virus transmission, Ann. Appl. Biol. 59:275.

Taylor, C. E., 1980, Nematodes, in: Vectors of Plant Pathogens (K. E Harris and K. Maramorosch, eds.), pp. 375-416, Academic Press, New York.

Taylor, C. E., and Brown, D. J. E, 1976, The geographical distribution of Xiphinema and Longidorus nematodes in the British Isles and Ireland, Ann. Appl. Biol. 84:384.

Taylor, C. E., and Murant, A. E, 1968, Chemical control of raspberry ringspot and tomato black ring viruses in raspberry, Plant Pathol. 17:171.

Taylor, C. E., and Thomas, P. R., 1968, The association of Xiphinema diversicaudatum (Micoletzky) with strawberry latent ringspot and arabis mosaic viruses in a raspberry plantation, Ann. Appl. Biol. 62:147.

Taylor, C. E., Brown, D. J. E, Neilson, R., and Jones, A. T., 1994, The persistence and spread of Xiphinema diversicaudatum in cultivated and uncultivated biotopes, Ann. Appl. Biol. 124:469.

Thomas, C. E., 1969, Transmission of tobacco ringspot virus by Tetranychus sp., Phytopathology 59:633.

Thomas, P. R., 1970, Host status of some plants for Xiphinema diversicaudatum (Micol.) and their susceptibility to viruses transmitted by this species, Ann. Appl. Biol. 65:169.


Thomason, I. T., and McKenry, 1975, Chemical control of nematode vectors of plant viruses, in: Nematode Vectors of Plant Viruses (F. Lamberti, C. E. Taylor, and T. W. Seinhorst, eds.), pp. 423-439, Plenum Press, London.

Thresh, T. M., and Pitcher, R. S., 1978, The spread of nettlehead and related virus diseases of hop, in: Plant Disease Epidemiology (P. R. Scott and A. Bainbridge, eds.), pp. 291-298, Blackwell, London.

Tuite, T., 1960, The natural occurrence of tobacco ringspot virus, Phytopathology 50:296. Vuittenez, A., 1958, Activite comparee des fumigants, des insecticides et des divers produits

appliques au traitement du sol sur les contaminations par la degenerescence infectieuse de la vigne, C. R. Seances Acad. Agric. Fr. 44:90l.

Yang, A. F., and Hamilton, R. I., 1974, The mechanism of seed transmission of tobacco ringspot virus in soybean, Virology 62:26.

CHAPTER 9

Fabaviruses: Broad Bean Wilt and Allied Viruses v. LISA AND G. BOCCARDO

I. INTRODUCTION

Broad bean wilt virus (BBWV), first isolated from broad bean (Vicia faba) in Australia (Stubbs, 1947), is the type member of the genus Fabavirus. Fabaviruses are aphid-transmitted viruses infecting a wide range of plants worldwide. Cooper (1994) noted that this genus includes a number of poorly characterized isolates with different common names and from many different plant species. We shall try to clarify this rather confused situation. BBWV isolates were divided by Uyemoto and Provvidenti (1974) into two serotypes (I and 11). The type strain belongs to serotype I. Nasturtium ringspot virus (NRSV) (Smith, 1950), petunia ringspot virus (PeRSV) (Rubio-Huertas, 1959), parsley virus 3 (PV3) (Frowd and Tomlinson, 1970, 1972), and P.o. pea streak virus (Kim and Hagedorn, 1959) are isolates or strains closely related to BBWV serotype I. BBWV serotype II was first detected in Plantago lanceolata in the United States (Uyell1-0to and Provvidenti, 1974), and later found in several other plant species. A possible third serotype was reported from Australia (Shukla and Gough, 1983), and an isolate from Cynara scolymus was reported to differ from both recognized serotypes: Migliori (1993) named it artichoke French latent virus (AFLV). Lamium mild mosaic virus (LMMV) (Lovisolo, 1957) was isolated in the United Kingdom from Lamium album

v. LISA AND G. BOCCARDO • Istituto di Fitovirologia Applicata del Consiglio Nazionale delle Ricerche, 10135 Torino, Italy.

229

230 V. LISA AND G. BOCCARDO

and a few other species of Labiatae (Lovisolo, 1957; Lisa et a1., 1982), but has not been reported since. It is distantly serologically related to BBWV, but distinct from it, and is regarded as another fabavirus.

BBWV and LMMV have isometric particles, which are hexagonal in outline and about 30 nm in diameter, and upon density gradient ultracentrifugation, sediment as three components (top, T; middle, M; and bottom, B). Two species of single-stranded positive-sense RNA (RNA-l and RNA-2), both necessary for infectivity, are separately encapsidated in Band M particles, respectively, whereas T particles do not contain nucleic acid. The viral capsids are composed of two species of polypeptide.

On the basis of physicochemical characters, BBWV was at one time considered a possible comovirus (Matthews, 1982), although it is apparently not serologically related to comoviruses and is transmitted by aphids, not by beetles. However, following proposals (Doel, 1975; Lisa et a1., 1982; Russo et a1., 1973) that BBWV and LMMV should be separated from other plant viruses, a new taxonomic group, fabavirus, was approved by the International Committee on Taxonomy of Viruses (Francki et a1., 1991). The name derives from the Linnean name of broad bean, Vicia faba. The genus Fabavirus is now included, together with the genera Como virus and Nepovirus, within the family Comoviridae (Murphy et a1., 1995).

Various aspects of the ecology, biology, pathology, and other characteristics of fabaviruses have been reviewed (Cockbain, 1983, 1984; Edwardson and Christie, 1991; Cooper, 1994). This chapter is intended as a comprehensive update of the available knowledge on these viruses.

II. HOSTS AND SYMPTOMATOLOGY

A. Natural Hosts and Economic Importance

The natural host range of BBWV includes a few monocotyledons and many dicotyledons: Edwardson and Christie (1991) listed 177 species in 39 families. The majority of hosts are herbaceous annuals, but some are woody perennials, for example, Catalpa bignonioides (Schmelzer and Wolf, 1969), Cornus florida (Scott and Barnett, 1984), Syringa vulgaris (Fraser and Conroy, 1963), and Vitis vinifera (Castrovilli et a1.,1985). BBWVhas also been isolated from overwintering perennial ornamentals, such as rhizomatous iris (Bailiss et a1., 1975), Leonurus cardiaca (Bruckart and Lorbeer, 1976), Begonia semperflorens (Lockhart and Betzold, 1982), and Linaria vulgaris (Rist and Lorbeer, 1989), which can therefore be important virus reservoirs, along with wild plants and weeds (Taylor and Stubbs, 1972; Uyemoto and Provvidenti, 1974; Schmelzer et a1., 1975; Bruckart and Lorbeer, 1976; Gracia and Feldman, 1976; Milicic et a1., 1976; Schmelzer and Stahl, 1977; Schmelzer et a1., 1977). BBWV has been reported most frequently in horticultural crops, but it also occurs in ornamentals such as Ajuga reptans (Shukla and Gough, 1983), Digitalis lanata (Schumann, 1963), Narcissus tazetta (Iwaki and Komuro,

FABAVIRUSES 231

1972), and Limonium sinuatum (Hein et al., 1977). Hosts of BBWV of economic importance or from which isolates of interest have been obtained are listed in Table I. Symptoms on field-infected plants range from mottle, mosaic, ringspots, distortion, wilting, apical necrosis, and stern necrosis to symptomless infection.

Generally, BBWV has been considered a virus of minor economic importance (Taylor and Stubbs, 1972; Conti and Masenga, 1977; Polak, 1981; Scott and Barnett, 1984; Castrovilli et al., 1985), and this might be why it has attracted relatively little attention. However, it can cause serious epidemics in restricted areas in some crops. Schroeder and Provvidenti (1970) and Provvidenti et al. (1984) described destructive diseases of lettuce and spinach caused by BBWV in New York State. Weidemann et al. (1975) and Marrou et al. (1976) reported similar diseases with almost total losses of the same crops in Germany and southern France, respectively. BBWV is one of the commonest viruses in broad bean, cowpea, and pea crops in central and eastern China (Xu et al., 1988), where it was also detected in spinach (Tian et al., 1982). Infections on Phaseolus vulgaris have occurred in the United States, Italy, and China (Provvidenti, 1983; Lisa et a1., 1986; V. Lisa, unpublished data). A severe outbreak of BBWV in sweet pepper (Capsicum frutescens) was reported from the Hiroshima prefecture of Japan (Imoto, 1975).

Losses caused by BBWV in affected crops ranged from about 2 to 25 % in broad bean, depending on the age of the plants at the time of infection (Makkouk et al., 1990), and were up to 20% in artichokes (Migliori et a1., 1987). From the Mediterranean island of Linosa, infections of broad bean were recorded for 3 consecutive years with incidences of 10% in the first survey, increasing to 80% in the last (Rosciglione and Cannizzaro, 1977).

As mentioned, LMMV has been found only once, infecting Lamium album, L. purpureum, and two species of Marrubium (Lovisolo, 1957); since then there has been no report of natural infection (Lisa et al., 1982). Recently, though, LMMV has been again isolated from Lamium orvala in Germany (D.-E. Lesemann, personal communication).

B. Experimental Host Range

Fabaviruses are easily sap-transmitted and their experimental host range is wide. Many hosts, especially solanaceous plants, react to BBWV infection by developing ringspot symptoms. Broad bean and Nicotiana clevelandii are suitable hosts for maintaining most BBWV isolates (Taylor and Stubbs, 1972). Chenopodium quinoa is a good diagnostic and propagation host, giving chlorotic local lesions, leaf epinasty, and apical necrosis with most isolates so far reported, the exception being an Egyptian isolate from pea (Kishtah et a1., 1978), which gives only local lesions. No appreciable difference in host range has been observed between isolates belonging to serotypes I and II (Uyemoto and Provvidenti, 1974; Migliori et a1., 1988; Makkouk et a1., 1990), although variation in host range and symptomatology were reported among


TABLE I. Selected Natural Host Range of BBWV: Species of Economic Importance or Hosts of Isolates of Interest

Geographical Host distribution Serotypea References

Aiuga reptans Australia I, III? Shukla and Gough (1983) Begonia semperfiorens USA I Lockhart and Betzold (1982) Beta vulgaris Czechoslovakia Polak (1981) Bouvardia spp. Italy II Vaira et al. (1992) Capsicum annuum Italy, France, I Giunchedi (1972); Boccardo and

Morocco Conti (1973); Rougier and Marchoux (1974); Lockhart and Fischer (1977); Gracia and Gutierrez (1982) Davino et al. (1989)

Catharanthus IOseus Australia II Shukla et al. (1980) Comus florida USA II Scott and Barnett (1984) Cucurbita pepo Italy I V. Lisa, G. Dellavalle,

G. D' Agostino (unpublished data)

Cynara scolymus Italy, France I, III? Russo and Rana (1978); Migliori et al. (1988); Migliori (1993)

Daucus carota Germany Taylor and Stubbs (1972) HelleboIUs vesicarius UK II Murant and Roberts (1977) Lactuca sativa USA, France I, II Uyemoto and Provvidenti

(1974); Marrou et al. (1976) Limonium sinuatum Germany Hein et al. (1977) Lupinus spp. Germany Schmidt and Schubert (1979) Lycopersicon esculentum France Marrou et al. (1974) Narcissus tazetta Japan Iwaki and Komuro (1972) Phaseolus vulgaris Germany, I, II Schmidt (1981); Provvidenti

USA, Italy, (1983); Lisa et al. (1986); China V. Lisa (unpublished data)

Petroselinum crispum Germany, UK Taylor and Stubbs (1972) Petunia hybrida Spain Taylor and Stubbs (1972) Phytolacca americana Italy O. Lovisolo et al. (1995) Pisum sativum Worldwide I, II Taylor and Stubbs (1972); Xu

et al. (1988) Plantago lanceolata USA, I, II, III? Uyemoto and Provvidenti

Argentina, (1974); Gracia and Feldman Italy (1976); Lovisolo et al. (1995)

Pogostemon patchouli Japan III? Natsuaki et al. (1994) Solanum melongena France I Rougier and Marchoux (1974) Spinacia oleracea Worldwide I, II Taylor and Stubbs (1972); Xu

et al. (1988) Tropaeolum maius Worldwide Taylor and Stubbs (1972) Vida faba Worldwide I, II Taylor and Stubbs (1972); Xu

et al. (1988) Vigna sinensis France, China Selassie-Gebre et al. (1976); Xi

et al. (1982) Vitis vinifera South Africa Castrovilli et al. (1985)

aAs stated by the authors or inferred from the text.

FABAVIRUSES 233

isolates of different origin, regardless of the serotype involved (Gracia and Feldman, 1976; Selassie-Gebre et a1., 1976; Shukla et a1., 1980; Vega et a1., 1980; Migliori et a1., 1988; Makkouk et a1., 1990). The host range of LMMV too is wide, but differs somewhat from that of BBWV (Lovisolo, 1957; Lisa et a1., 1982).

C. Mode of Transmission

All fabavirus isolates or strains so far tested are aphid-transmitted in the nonpersistent manner, with minimum acquisition access and inoculation feeding times varying from 15 sec to 10 min and from 30 sec to 24 hr, respectively. Stubbs (1960) found that Myzus persicae, allowed short inoculation probes of not more than 30 sec, transmitted BBWV to up to six of ten successively probed plants; Aphis craccivora was a much less efficient vector. Boccardo and Conti (1973), working with NRSV, reported that frequency of transmission by M. persicae was greater after acquisition feeding times of 1 min than of 10 min, whereas the reverse was true for Aphis nasturtii. However, this could be more a reflection of the aphid feeding behavior than an intrinsic property of the virus. About 20 other aphid species, including Acyrthosiphon onobrychis, A. pisum, Aphis craccivora, A. fabae, A. nasturtil, Cavariella aegopodii, Hyperomyzus lactucae, Macrosiphon euphoribiae, M. pisi, and M. solanifolii have been reported as vectors of BBWV (type and NRSV strains) (Stubbs, 1960; Taylor and Stubbs, 1972; Karl et a1., 1972; Boccardo and Conti, 1973; Marrou et a1., 1976; Gracia and Gutierrez, 1982; Makkouk et a1., 1990; Edwardson and Christie, 1991). M. persicae is the most efficient vector of the type strain of BBWV and of the majority of other isolates belonging to both serotypes I and II (Frowd and Tomlinson, 1972; Taylor and Stubbs, 1972; Gracia and Feldman, 1976; Marrou et a1., 1976; Makkouk et a1., 1990). Transmission frequencies appear somewhat influenced by the plant species used as donor and test host (Schmelzer, 1960; Conti and Boccardo, 1973). However, AFLV from France is efficiently transmitted by Capitophorus horni, but apparently not by M. persicae, Aulacorthum solani, or Brachycaudus cardui (Migliori et a1., 1984, 1987, 1988). There is no report of attempted transmission of BBWV isolates or strains or of LMMV with possible vectors other than aphids.

BBWV was not transmitted by three Cuscuta species (Taylor and Stubbs, 1972) and is reportedly not transmitted through the seeds of broad bean (Taylor and Stubbs, 1972), pepper (Boccardo and Conti, 1973), cowpea (Selassie-Gebre et a1., 1976), French bean (Provvidenti, 1983), and Thunbergia alata (Provvidenti and Hoch, 1985). However, Makkouk et a1. (1990) reported that a Syrian isolate of BBWV was transmitted through the seeds of broad bean at a very low frequency (0.4-0.6%, depending on the stage of inoculation).

LMMV differs from BBWV in being better transmitted by Cryptomyzus alboapicalis than by M. persicae (Lovisolo, 1957). However, this may not be

234 v. LISA AND G. BOCCARDO

very meaningful, because, as mentioned above, at least one virus of this genus is not transmitted by M. persicae.

III. CYTOPATHOLOGY

There is an extensive literature on the cytopathological effects of BBWV and its strains NRSV, PeRSV, and PV3, but most work refers to isolates belonging to serotype I. Many BBWV isolates induce prominent cytoplasmic inclusions easily detected by light microscopy in leaf epidermal cells of different hosts, but especially in Vicia faba. The inclusions are amorphous, spindle-shaped, or crystalline (Rubio-Huertos, 1968; Juretic et al., 1970; Milicic et al., 1974; Russo and Martelli, 1975; Christie and Edwardson, 1977). These three types of inclusion appear to correspond, respectively, to the masses of convoluted membranes, tubular arrays of virus particles, and crystallized virus particles observed by electron microscopy as detailed below.

Thin sections of infected tissue reveal virus particles scattered in the cytoplasm, although they are difficult to distinguish from ribosomes. Virus particles may also aggregate in a number of different ways: (1) as singlewalled tubules (Fig. 1) formed from rolled-up, close-packed, two-dimensional hexagonal arrays; (2) as more complex hollow crystalline forms (Fig. 2); (3) as three-dimensional crystals (Fig. 3); and (4) as unstructured but highly localized masses of particles (Fig. 4). The different types of aggregate often occur together in infected cells. Type 1 aggregates, long tubules about 80 nm in diameter with nine particles in cross-section (Fig. 1b,c), are typical of many BBWV isolates (Rubio-Huertos, 1968; Sahambi et a1., 1973; Hull and Plaskitt, 1974; Weidemann et al., 1975; Russo and Martelli, 1975; Kishtah et al., 1978; Vega et al., 1980). With the PV3 strain, sheets of virus particles may form scrolls or tubules 200-250 nm in diameter, with 25-30 particles in crosssection (Frowd and Tomlinson, 1972; Hull and Plaskitt, 1974). Some isolates, such as the type strain of BBWV, the NRSV isolate of Smith (1950), and others, all of serotype I, did not produce tubules (Sahambi et al., 1973; Hull and Plaskitt, 1974; Saric et al., 1984; Castrovilli et al., 1985); these were, however, detected in Lathyrus odoratus and Nicotiana benthamiana cells infected by an artichoke isolate possibly belonging to serotype II (Migliori et al., 1988). Hollow crystalline forms (type 2 above) were found with a few isolates, all belonging to serotype I, in association with tubules and solid crystals (Kishtah et al., 1978; Russo et al., 1979; Vega et al., 1980) or with crystals only (Saric et al., 1984). Type 3 crystalline aggregates may occur alone (Hein et a1., 1977; L.-X. Feng and D.-E. Lesemann, personal communication). In mesophyll cells of C. quinoa infected with the ATCC PV 131 serotype II isolate, large numbers of type 3 crystalline aggregates were found (Fig. 3), especially in necrotic tissues where desiccation had occurred. Type 4 aggregates consisted largely of "empty" capsids without densely stained

FABAVIRUSES 235

b 2

FIGURE 1. Thin sections of broad bean mesophyll cells infected with an Argentine isolate of BBWV, serotype I. (Courtesy of J. Vega.) (a) Cross-sectioned tubules in the cytoplasm, lining the tonoplast. Scale bar, 500 nm. (b) Tubules as in (a), enlarged. Nine virus particles appear in each ring. Scale bar, 200 nm. (c) Longitudinal section of tubules. Scale bar, 200 nm.

FIGURE 2. Quadrangular hollow structures in cross-section, also consisting of virus particles. Scale bar, 300 nm.


3

4

FIGURES 3 and 4. Thin sections of Chenopodium quinoa mesophyll cells infected with BBWV serotype II isolate ATCC PV 131.

FIGURE 3. Three-dimensional crystals composed mostly of "empty" viral capsids. Scale bar, 300 nm. (Courtesy of R. G. Milne and V. Masenga.)

FIGURE 4. Unstructured mass of virus particles (VP), convoluted membranes (arrows), and vesicles (arrow heads). Scale bar, 400 nm. (Courtesy of R. G. Milne and v. Masenga.)

interiors (R. G. Milne and v. Masenga, personal communication). As with comoviruses and nepoviruses, virus particles sometimes occur in single file in tubules and passing through plasmodesmata (Hull and Plaskitt, 1974; Saric et a1., 1984; Castrovilli et a1., 1985; Francki et a1., 1985, 1991).

Fabavirus infection does not appear to alter cell organelles in specific ways and wall outgrowths or paramural bodies have not been reported, but large masses of convoluted membranes and vesicles, similar to those induced by comoviruses and nepoviruses, are commonly observed in the cytoplasm (Fig. 4). The only report of nuclear inclusions (Gracia and Feldman, 1976)

FABAVIRUSES 237

refers to a crystalline structure observed in tissues infected with an isolate of BBWV serotype II. Nonstructured aggregates of "empty" viral capsids were also noted in nuclei of cells infected with the ATCC PV 131 serotype II isolate (R. G. Milne and V. Masenga, personal communication). LMMV particles have been seen only in the cytoplasm of infected cells and the virus did not induce any specific cytopathological alteration (Lisa et al., 1982).

IV. PROPERTIES OF PARTICLES

A. Purification

Although the concentration of BBWV particles in infected cells appears to be relatively high, and in crude sap of herbaceous test plants the virus is moderately stable (Taylor and Stubbs, 1972), virus purification may be troublesome, at least with some isolates. This may be another reason why these viruses are relatively little studied. Irrespective of the extraction buffer used, fabavirus particles tend to aggregate or to adsorb to membranes or cell debris and precipitate in the early stages of the purification process.

Extraction solutions have usually been supplemented with varying amounts of antioxidants and sometimes with nonionic detergents, in order to solubilize host membranes. Crude leaf extracts have usually been clarified with chloroform and/or butan-l-01 and the virus particles concentrated by cycles of differential centrifugation; the particles are further purified by sucrose velocity density gradient ultracentrifugation, by CsCI isopycnic ultracentrifugation, or by zone electrophoresis (Taylor et al., 1968; Taylor and Stubbs, 1972; Frowd and Tomlinson, 1972; Boccardo and Conti, 1973; Uyemoto and Provvidenti, 1974; Doel, 1975; Lisa et al., 1982; Castrovilli et al., 1985; Migliori et al., 1988). Addition to the extraction buffer of sucrose (which could be substituted by glycerol, ethylene glycol, or glucose) to a final concentration of 25% appeared to reduce particle aggregation and allow better yields (Doel, 1975). Particles of some isolates could be purified by a method devised for cucumber mosaic cucumovirus by Lot et al. (1972), in which virus is concentrated from clarified sap by precipitation with polyethylene glycol (PEG) (mol. wt. 6000) and the nonionic detergent Triton X-IOO is added to the resuspension buffer (Migliori et al., 1984). Addition of 2.5% Triton X-IOO to the crude sap, concentration with PEG, differential centrifugation cycles, and sucrose velocity density gradient centrifugation were used to purify an isolate from grapevine (Castrovilli et al., 1985). Particles of isolates belonging to BBWV serotype II were partially purified by addition of 5% Triton X-IOO and concentrated by differential and sucrose density gradient velocity ultracentrifugations. Because fabavirus particles tend to aggregate, and therefore to be pelleted by low-speed centrifugation, care must be taken to recover the particles from this phase by repeated extractions with appropriate buffer (V. Lisa, quoted in Migliori et al., 1988).


B. Types of Particle

Upon velocity sucrose density gradient or CsCI isopycnic ultracentrifugation, preparations of particles of BBWV and LMMV separate into three distinct components: top (T), consisting of capsids devoid of nucleic acid, and middle (M) and bottom (B), each encapsidating a distinct RNA species. In this respect, fabavirus particles appear to resemble those of comoviruses more than those of nepoviruses, whose B particles may contain either a single molecule of RNA-lor two of RNA-2 (Francki et a1., 1985, 1991; Murphy et a1., 1995). Particles of all three components of fabaviruses are isometric, roughly hexagonal in outline, and about 30 nm in diameter (Fig. 5). Estimates of sedimentation coefficient (so20 w)' calculated from analytical ultracentrifuge Schlieren diagrams using the graphical method of Markham (1960), ranged from 56 to 63 S (T), 93 to 100 S (M), and 113 to 126 S (B) for different BBWV isolates (Taylor and Stubbs, 1972; Sahambi et a1., 1973; Doel, 1975; Bailiss et a1., 1975; Scott and Barnett, 1984; Castrovilli et a1., 1985; Xu et a1., 1988; Migliori et a1., 1988). The buoyant densities of particles of BBWV type isolate, NRSV and PeRSV in CsCI containing the detergent Igepon T 73 were 1.39-1.40 glml for M component and 1.44 glml for B component (Doel, 1975). In CsCI without detergent, the buoyant density values for a broad bean isolate belonging to serotype II were 1.301 (T) and 1.379 (M) g/ml; the B component separated into two classes with densities of 1.409 and 1.464 glml (Xu et a1., 1988).

C. The Genome

The genomes of both BBWV and LMMV consist of two species of positive-sense, single-stranded RNA (RNA-1 and RNA-2), both necessary for infectivity, encapsidated in Band M particles, respectively. However, it is not known whether these RNA molecules, like those of comoviruses and nepoviruses, possess a covalently linked protein (VPg) at their 5/ termini or if

FIGURE 5. Purified M component particles of NRSV (BBWV serotype I). Scale barr 100 nm. (Courtesy of R. G. Milne.)

FABAVIRUSES 239

instead they have capped structures. Nor is it known if they possess adenylic homopolymers at their 3' termini, as do the RNA molecules of comoviruses and nepoviruses. The genetic functions of the two RNA species of fabaviruses have not been studied, and it is not known whether they are fully interdependent or whether RNA-1 possesses limited independence from RNA-2, as with comoviruses IGoldbach et a1., 1980; Franssen et a1., 1982) and nepoviruses IRobinson et a1., 1980). Doel(1975) reported that, with BBwv, NRSV, and PeRSV, M and B components each induced some local lesions when inoculated to C. quinoa, though far fewer than did mixtures of the two components. In retrospect, we can consider that probably there was slight residual cross-contamination of the separated M and B components, since no replication of RNA-2 in the absence of RNA-1 has ever been reported within the ComoviIidae.

Digestion of phenol-extracted RNA from particles of BBWV IATCC PV 131, serotype n), NRSV IBBWV serotype I), and LMMV with up to 200 fJ.g/ml of proteinase K at 37 °C did not significantly alter the number of local lesions induced on C. quinoa, as compared with untreated RNA IG. Boccardo and V. Lisa, unpublished data). This suggests that VPg, if present in these viruses, is, as with comoviruses, not required for infectivity IFrancki et a1., 1985).

LMMV genomic RNA was transcribed into complementary DNA IcDNA) using oligoldTh2 to prime a first-strand reaction driven by avian myeloblastosis virus reverse transcriptase IG. Boccardo, M. d'Aquilio and S. Antoniazzi, unpublished data). This suggests that these RNA molecules, like those of comoviruses and nepoviruses IFrancki et al., 1991; Murphy et al., 1995), possess polYIA) tails at their 3' termini. However, when either oligoldTh2 or random hexanucleotides were used to prime the reaction, much more cDNA corresponding to RNA-2 than to RNA-1 was obtained Isee Fig. 6b), perhaps because the latter possesses a less easily denaturable secondary structure. In Northern hybridization experiments, LMMV cDNA reacted only with the homologous RNA species and not with RNA molecules of either serotype of BBWV I Fig. 6b), indicating that little, if any, nucleotide sequence homology exists between BBWV and LMMV IG. Boccardo, M. d' Aquilio, and S. Antoniazzi, unpublished data).

BBWV M component contains 25-26% RNA IRNA-2), of estimated molecular weight Ix 106) 1.3-1.7; the B component contains 35% RNA IRNA-1) of molecular weight Ix 106) 2.0-2.6ITaylor and Stubbs, 1972; Doel, 1975; Lisa et al., 1982; Castrovilli et al., 1985). The relatively large variations in reported molecular weight may reflect differences in the techniques used: in most instances, the RNA samples were coelectrophoresed with the standards under nondenaturing conditions. Probably the most accurate published estimates were obtained for two BBWV isolates from artichoke by Migliori et al. (1988) using agarose-gel electrophoresis after denaturing the RNA samples with 50% formamide: they reported values of 2.1 x 106 and 1.3 x 106 for RNA-1 and RNA-2, respectively. The estimated molecular weights of LMMV RNA-l and RNA-2 were 2.23 x 106 and 1.59 x 106, respec-


b

FIGURE 6. Agarose gel electrophoresis of denatured (McMaster and Carmichael, 1977) RNA and Northern blot of the same gel, with LMMV cDNA as probe. (a) RNA samples from (from left to right): alfalfa mosaic alfamovirus, tobacco mosaic tobamovirus, Gibco-BRL RNA ladder, LMMV, NRSV (BBWV serotype I), and ATCC isolate PV 131 (BBWV serotype II). (b) RNA bands from the gel in (a) were transferred to nylon membranes by capillary action, fixed under UV light, and probed by standard methods (Marzachi et al., 1992) with LMMV cDNA. Note that whereas the two genomic RNA species seem to be present in almost equimolar amounts (a), cDNA against LMMV RNA-l appears to have been produced in much smaller amounts than cDNA to LMMV RNA-2. No cross-reaction of LMMV cDNA with genomic RNA species of either serotype of BBWV is detectable.

tively, under nondenaturing conditions (Lisa et al., 1982). M. d'Aquilio and G. Boccardo (unpublished data) incubated preparations of native genomic RNA from LMMV, NRSV (BBWV serotype I), and BBWV ATTC PV 131 (serotype II) for 60 min at 50 DC with glyoxal and dimethylsulfoxide (McMaster and Carmichael, 1977), and electrophoresed them in horizontal agarose gel slabs as described by Lisa et al. (1988), together with a commercial singlestranded RNA ladder (Gibco BRL). The results (Fig. 6a; Table II) indicate that the nucleic acids of the three fabaviruses differ slightly in size.

The molar percentage base ratios of RNA-1 of BBWV, NRSV, and PeRSV were similar, all containing 26% guanine, 30% adenine, 17.5% cytosine, and

FABAVIRUSES

TABLE II. The Molecular Sizes of the Genomic RNA Species of Fabaviruses, Determined

under Denaturing Conditionsa

Virus

BBWV serotype I BBWV serotype II LMMV

Molecular size (kb)b

RNA-l

6.29 (2.13) 6.60 (2.26) 6.35 (2.15)

RNA-2

3.85 (1.30) 4.11 (1.39) 4.11 (1.39)

aM. d'Aquilio and G. Boccardo (unpublished data). For experimental procedure, see text.

bValues in parentheses are mol. wt. (x 106).

241

26.5% uracil; BBWV RNA-2 was also essentially identical in base composition (Doel, 1975).

D. The Coat Proteins

The protein shells of BBWV and LMMV are built of apparently equimolar amounts of two distinct polypeptides, estimated by polyacrylamide gel electrophoresis to be (x 103 ) 23-27 and 41-45 in size (Doel, 1975; Lisa et a1., 1982; Castrovilli et a1., 1985; Migliori et a1., 1988). Doel (1975) judged it unlikely that the larger polypeptides were dimeric forms of the smaller. Moreover, tryptic digests of purified but unfractionated BBWV, NRSV, and PeRSV capsid proteins each consisted of approximately 70 oligopeptides, a figure consistent with the presence of two polypeptides of different amino acid sequence (Doel, 1975). Though the coat proteins of comoviruses are reportedly somewhat different in size from those of fabaviruses (Table III), the corresponding proteins comigrate when coelectrophoresed in polyacrylamide gels (Doel, 1975; Lisa et a1., 1982).

V. RELATIONSHIPS

A. Relationships within the Genus

BBWV type strain is serologically indistinguishable from, or very closely related to, a number of strains or isolates including NRSV, first described in the United Kingdom and later found in several European countries. These include PeRSV, PV3, and the P.O. pea streak virus (Taylor and Stubbs, 1972; Frowd and Tomlinson, 1972; Sahambi et a1., 1973; Doel, 1975; Lisa et a1., 1982). In New York State, Uyemoto and Provvidenti (1974) found some isolates that were serologically distinct from those previously described and

TA

BL

E I

II.

Co

mp

aris

on

of

Som

e P

rope

rtie

s of

BB

WV

, C

omov

irus

es,

Nep

ovir

uses

, an

d S

LR

SV

Sed

imen

tati

on c

oeff

icie

nt,

Est

imat

ed m

ol.

wt.

See

d S

20,w

(Sv

edbe

rgs)

R

NA

-1

RN

A-2

C

oat

pro

tein

s P

redo

min

ant

Vir

uses

tr

ansm

issi

on

Vec

tors

T

M

B

(x

106

) (x

106

) (x

103

) cy

topl

asm

ic i

nclu

sion

s

BB

WV

R

are

Aph

ids

56-6

3 9

3-1

00

11

3-12

6 2

.0-2

.6

1.4-

1.7

23-2

7; 4

1-4

5

Ves

icle

s, t

ubul

es,

tubu

lar

viru

s cr

ysta

ls

Com

ovir

uses

a Y

es

Bee

tles

5

1-6

0

91-1

03

112-

130

2.0-

2.7

1.3-

1.8

21

-23

;37

-48

V

esic

les,

tub

ules

, vi

rus

crys

tals

N

epov

irus

esa

Yes

N

emat

od

es

49

-56

8

4-1

20

11

7-13

4 2.

1-2.

8 1.

4-2.

4 5

2-6

0

Tub

ules

, di

ffus

e in

clus

ions

SL

RsV

a,b

Yes

N

emat

od

es

58

94

126

2.6-

2.9

1.4-

1.6

29

;44

T

ubul

es,

vesi

cles

aMur

phy

et a

l. 11

995)

. bM

uran

t 11

974)

.

FABAVIRUSES 243

formed spurs towards the wells containing the heterologous viruses in agar gel double-diffusion tests; they designated the original virus as serotype I and the variants as serotype II. Isolates close to BBWV type strain from Australia were grouped in serotype I and those close to the ATCC PV 131 lettuce isolate in serotype II (Uyemoto and Provvidenti, 1974). Most BBWV isolates so far reported can be grouped in this way. In Table I, a number of BBWV isolates are listed according to their serotype, where known. Isolates belonging to serotype I seem to be prevalent in Europe, although isolates of serotype II have been reported (Murant and Roberts, 1977; Migliori et a1., 1988; Vaira et a1., 1992). Isolates of serotype II have been more commonly reported from Australia, North America, and China (Table I). Francki et a1. (1991) listed BBWV serotype II as a separate virus within the fabavirus taxonomic group (now genus, Murphy et a1., 1995). Recognized differences between the two serotypes refer only to serological properties (Uyemoto and Provvidenti, 1974). Should other characteristics be found different, it would perhaps be worth using the name of BBWV for isolates belonging to serotype I and the name plantago mosaic virus (Gracia and Feldman, 1976) for isolates belonging to serotype II. An isolate of a possible third serotype has been reported in Australia from Ajuga reptans. In agar gel immunodiffusion tests with antisera to serotypes I and II, isolates belonging to these two serotypes formed spurs with this antigen (Shukla and Gough, 1983), and similar results were obtained with an isolate from patchouli (Pogostemon patchouli) in Japan, named Patchouli mild mosaic virus (PaMMY; Natsuaki et a1., 1994). AFLV, isolated from artichoke in France and transmitted by Capitophorus horni, was reported to differ from BBWV isolates belonging to both serotypes I and II and from LMMV (Migliori, 1993). Isolates differing from BBWVI and II, PaMMY and LMMV have been reported in northern Italy from Plantago lanceolata and Phytolacca americana (Lovisolo et a1., 1995). However, a direct comparison of the properties of these isolates appears necessary before it can be decided whether they are better considered separate fabaviruses or strains of already recognized ones.

LMMV appears to differ from any BBWV strain so far reported. In agar gel immunodiffusion tests, it was distantly related to the BBWV type strain and to NRSV (both belonging to serotype I), with serological differentiation indices of 9 and 10, respectively, and unrelated to isolate ATCC PV 131, belonging to BBWV serotype II (Lisa et a1., 1982). These results were confirmed by Migliori et a1. (1988).

B. Affinities with Other Viruses

Table III compares the properties of fabaviruses with those of some other plant viruses. The closest affinities of fabaviruses are with comoviruses (Bruening, 1978) and nepoviruses (Harrison and Murant, 1977; Murant, 1981), which likewise have bipartite ssRNA genomes (see also Chapters 3 and 6). These three genera are placed together in the family Comoviridae (Murphy

244 v. LISA AND G. BOCCARDO

et a1., 1995). However, whereas fabaviruses have aphid vectors, comoviruses are transmitted by beetles, and nepoviruses typically have nematode vectors (though for many of them the vector is unknown and a few seem to be transmitted in association with pollen). Definitive nepoviruses also differ from fabaviruses in having only one capsid protein species, whereas comoviruses resemble fabaviruses in possessing two species of capsid protein, which comigrate with those of fabaviruses in polyacrylamide gel electrophoresis.

Two capsid protein species of similar sizes to those of fabaviruses and comoviruses are produced by a few viruses that are not yet taxonomically well defined but are tentatively classified as nepoviruses (see Chapter 6, Table I, part d, and also Chapter 5). The best studied of these tentative nepoviruses is strawberry latent ringspot virus (SLRSV) (Murant, 1974), which has a nematode vector. The others are rubus Chinese seed-borne virus (RCSV) (Barbara et a1., 1985 L which is serologically related to SLRSV, lucerne Australian symptomless virus (LASV) (Remah et a1., 1986) and satsuma dwarf virus (SDV) (Usugi and Saito, 1979); the vectors and modes of transmission of these three viruses are unknown. No serological relationship has been detected between BBWV and SLRSV or LASY, and no tests for serological relationship to fabaviruses have been done with the other two viruses.

Nepoviruses and comoviruses have a small protein (VPg) covalently linked at the 5' termini of their genomic RNA molecules. It is not known whether fabaviruses have such a VPg, but if so it would resemble those of comoviruses in being unnecessary for infectivity (see Section IV.C). Comoviruses, nepoviruses, and fabaviruses induce rather similar cytopathological effects and are easily sap-transmitted. BBWV and LMMV differ from comoviruses in having a larger experimental host range and in normally producing only moderate concentrations of virus particles in infected tissues. Nepoviruses, including the tentative ones, are commonly seed-transmitted; most comoviruses are too, at low but significant frequencies (Stace-Smith, 1981). In contrast, BBWV is reportedly not seed-transmitted in at least four hosts (Taylor and Stubbs, 1972; Boccardo and Conti, 1973; Selassie-Gebre et a1., 1976; Provvidenti, 1983; Provvidenti and Hoch, 1985); however, a very low percentage of seed transmission of a Syrian isolate has been detected in broad bean (Makkouk et a1., 1990).

Comoviruses and nepoviruses have significant similarities to Potyviridae and Picornaviridae in genome structure, organization, and expression, but very little is known about these features for fabaviruses. However, in general, fabaviruses seem to resemble comoviruses more closely than nepoviruses.

VI. DIAGNOSIS

Diagnostic methods for the detection of BBWV and LMMV have not been applied on a large scale, because these viruses so far have not been found

FABAVIRUSES 245

to cause major diseases in vegetatively propagated plants and are not, or only infrequently, associated with diseases that are prevalent in crops of economic value. Diagnosis is mainly based on conventional methods, such as host response, serology, and electron microscopy. Antisera to a number of virus strains and isolates have been raised and serological methods widely applied. Virus concentration in naturally infected plants is seldom high enough to allow direct testing by immunodiffusion methods, thus requiring transmission of the virus to test plants. Many isolates can be detected easily in crude sap of test plants by agar gel immunodiffusion. In this test, LMMV and some BBWV isolates tend to precipitate around the antigen well, hindering the formation of the precipitin line. This effect can be reduced by using agarose in citrate buffer rather than standard media such as agar or agarose in phosphate-buffered saline (Lisa et al., 1982). Enzyme-linked immunosorbent assay (ELISA) permitted a clear distinction of a BBWV serotype II isolate from NRSV and PV3 (A. J. Cockbain, personal communication). An indirect platetrapping ELISA procedure was reliably applied in testing large numbers of field samples for epidemiological studies (Rist and Lorbeer, 1989). Virus particles can easily be seen in infective sap by electron microscopy of negatively stained samples. Immunosorbent electron microscopy (ISEM) or ISEM plus decoration (Milne and Luisoni, 1977) can be applied to crude sap of fieldinfected plants.

VII. ECOLOGY AND CONTROL

BBWV has adapted widely to different hosts, having been found occurring naturally in almost 200 species, including several species of cultivated crop plant (Edwardson and Christie, 1991). The virus has been transmitted experimentally by at least 20 aphid species, including some that commonly infest vegetable crops. As seed transmission is infrequent and the virus is not retained for long in the vector, survival of BBWV under natural conditions probably depends on infection of perennial hosts, at least in temperate regions with cold winters. Thus, many of the wild hosts of BBWV may have epidemiological importance in acting as sources of virus for infection of crops. Plantago lanceolata and P. major are possibly the most important overwintering hosts of the virus (Provvidenti, 1983). P. lanceolata was commonly found infected with BBWV around pepper fields where the virus was present (Lovisolo et al., 1995). Infected Linaria vulgaris growing close to lettuce fields affected by BBWV was possibly a primary source of virus for the crop (Rist and Lorbeer, 1989).

Control of BBWV in crops presents all the obvious difficulties posed by nonpersistent aphid-transmitted viruses with wide natural host ranges. Roguing of weeds, insecticide spraying, and avoidance of sowing during periods when plants are prone to aphid infestation have all been suggested as ways to control BBWV in broad bean crops in Australia (Stubbs, 1947; Fraser and Conroy, 1963).


BBWV-resistant or tolerant genotypes have been sought in the United States for lettuce, spinach, and French bean. A number of accessions of Lactuca sativa and a few of L. virosa were tolerant; a selection of spinach and a large number of bean lines or cultivars were resistant (Schroeder and Provvidenti, 1970; Provvidenti, 1983; Provvidenti et a1., 1984). No source of resistance to BBWV in broad bean has yet been reported (Cockbain, 1983).

ACKNOWLEDGMENTS. We are grateful to all those mentioned in the text for contributing data, materials, and unpublished work, and to R. G. Milne, J. I. Cooper, and the late R. I. B. Francki for helpful discussion and revision of the English text.

REFERENCES

Bailiss, K. W., Brunt, A. A., and Dale, W. T., 1975, The natural occurrence of broad bean wilt virus in rhizomatous iris and spinach, Plant Pathol. 24:60.

Barbara, D. J., Ashby, S. C., and McNamara, D. G., 1985, Host range, purification and some properties of rubus Chinese seed·borne virus, Ann. Appl. Biol. 107:45.

Boccardo, G., and Conti, M., 1973, Purification and properties of nasturtium ringspot virus, Phytopathol. Z. 78:14.

Bruckart, W. L., and Lorbeer, J. W., 1976, Cucumber mosaic virus in weed hosts near commercial fields of lettuce and celery, Phytopathology 66:253.

Bruening, G., 1978, Comovirus group, CMI/AAB Descriptions of Plant Viruses No. 199. Castrovilli, S., Savino, v., Castellano A. M., and Engelbrecht, D. J., 1985, Characterization of a

grapevine isolate of broad bean wilt virus, Phytopathol. Medit. 24:35. Christie, R. G., and Edwardson, R. J., 1977, Light and electron microscopy of plant virus

inclusions, Fla. Agric. Exp. Stn. Monograph Series No.9. Cockbain, A. J., 1983, Viruses and virus-like diseases of Vicia faba L., in: The Faba Bean (Vicia

faba L.). A Basis for Improvement (P. D. Hebblethwaite, ed.), pp. 421-445, Butterworths, London.

Cockbain, A. J., 1984, Faba beans in China: Some facts and personal observations, particularly concerning virus diseases, in: Vicia faba: Agronomy, Physiology and Breeding (P. D. Hebblethwaite, T. C. K. Dawkins, M. C. Heath, and G. Lockwood, eds.), pp. 185-191, Nijhoff and Junk Publishers, Dordrecht, The Netherlands.

Conti, M., and Boccardo, G., 1973, Investigations on two aphid borne viruses of pepper, Comptes Rendus des Troisiemes Journees de Phytiatrie et de Phytopharmacie Circum-Mediterraneennes, Sassari, 1971:20.

Conti, M., and Masenga, v., 1977, Identification and prevalence of pepper viruses in Northwest Italy, Phytopathol. z. 90:212.

Cooper, J.L, 1994, Fabaviruses, in: Encyclopedia of Virology (R. G. Webster and A. Granoff, eds.), pp. 451-454, Academic Press, New York.

Davino, M., Areddia, R., Polizzi, G., and Grimaldi, V., 1989, Osservazioni su una butteratura delle bacche di peperone in Sicilia, La Difesa delle Piante 12(1-2):65.

Doel, T. R., 1975, Comparative properties of type, nasturtium ringspot and petunia ringspot strains of broad bean wilt virus, r Cen. Virol. 26:95.

Edwardson, J. R., and Christie, R. G., 1991, Fabaviruses, in: CRC Handbook of Viruses Infecting Legumes (J. R. Edwardson and R. G. Christie, eds.) pp. 263-274, CRC Press, Boca Raton, FL.

Francki, R.L B., Milne, R. G., and Hatta, T., 1985, Atlas of Plant Viruses, Vol. II, CRC Press, Boca Raton, FL.

FABAVIRUSES 247

Francki, R. I. B., Fauquet, C. M., Knudson, D. 1., and Brown, F. (eds.), 1991, Classification and Nomenclature of Viruses. Fifth Report of the International Committee on Taxonomy of Viruses, Springer-Verlag, Vienna. (also in Arch. Virol. Suppl. 2).

Franssen, H., Goldbach, R., Broekhuijsen, M., Moerman, M., and Van Kammen, A., 1982, Expression of middle-component RNA of cowpea mosaic virus: In vitro generation of a precursor to both capsid proteins by a bottom-component RNA-encoded protease from infected cells, J. Virol. 41:8.

Fraser, 1., and Conroy, R. J., 1963, Vascular wilt disease of broad beans and other plants, Agric. Caz. New South Wales 74:225.

Frowd, J. A., and Tomlinson, J. A., 1970, Nasturtium ringspot virus in parsley crops in Britain, Plant Dis. Rep. 54:734.

Frowd, J. A., and Tomlinson, J. A., 1972, Relationships between a parsley virus, nasturtium ringspot virus and broad bean wilt virus, Ann. Appl. BioI. 72:189.

Giunchedi, 1., 1972, Ricerche virologiche sulla butteratura del peperone, Inf. Fitopatol. 22(4):3. Goldbach, R., Rezelman, G., and Van Kammen, A., 1980, Independent replication and expression

of B-component RNA of cowpea mosaic virus, Nature 286:297. Gracia, 0., and Feldman, J. M., 1976, Studies of weed plants as sources of viruses. IV. Broad bean

wilt virus in Sysimbrium irio and Plantago lanceolata in Argentine, Phytopathol. Z. 85:227.

Gracia, 0., and Gutierrez, 1. S., 1982, Broad bean wilt virus in pepper crops in Argentina, Phytopathol. Medit. 21:107.

Harrison, B. D., and Murant, A. F., 1977, Nepovirus group, CMI/AAB Descriptions of Plant Viruses No. 185.

Hein, A., Lesemann, D. E., and Querfurth, G., 1977, Das broad bean wilt-Virus in LimoniumKulturen in Suddeutschland, Phytopathol. Z. 89:340.

Hull, R., and Plaskitt, A., 1974, The in vivo behavior of broad bean wilt virus and three of its strains, Intervirology 2:352.

Imoto, M., 1975, Studies on mosaic disease of sweet pepper (Capsicum frutescens). 2. Occurrence of mosaic disease of sweet pepper and kinds of its causal viruses in Hiroshima prefecture, Bull. Hiroshima Prefectural Agric. Exp. Stn. 36:57.

Iwaki, M., and Komuro, Y., 1972, Viruses isolated from narcissus (Narcissus spp) in Japan. III. Cucumber mosaic virus, tobacco rattle virus and broad bean wilt virus, Ann. Phytopathol. Soc. Tpn. 38:137.

Juretic, N., Milicic, D., and Schmelzer, K., 1970, Zur Kenntnis des Ringmosaik-Virus der Kapuzinerkresse (nasturtium ringspot virus) und seiner Zell-einschlusskiirper, Acta Bot. Croat. 29:17.

Karl, E., Schmelzer, K., and Wolf, P., 1972, Untersuchungen zur Dbertragbarkeit des Ringmosaik-Virus der Kapuzinerkresse (nasturtium ringspot virus) durch Blattlausarten, Arch. PfJanzenschutz 8:269.

Kim, W. S., and Hagedorn, D. J., 1959, Streak-inciting viruses of canning pea, Phytopathology 49:656.

Kishtah, A. A., Russo, M., Tolba, M. A., and Martelli, G. P., 1978, A strain of broad bean wilt virus isolated from pea in Egypt, Phytopathol. Medit. 17:157.

Lisa, V., Luisoni, E., Boccardo, G., Milne, R. G., and Lovisolo, 0., 1982, Lamium mild mosaic virus: A virus distantly related to broad bean wilt, Ann. Appl. BioI. 100:467.

Lisa, v., Dellavalle, G., and Peressini, S., 1986, Broad bean wilt virus in Phaseolus vulgaris in Friuli-Venezia Giulia, Inf. Fitopatol. 36(4):31.

Lisa, V., Milne, R. G., Accotto, G. P., Boccardo, G., Caciagli, P., and Parvizy, R., 1988, Ourmia melon virus, a virus from Iran with novel properties, Ann. Appl. BioI. 112:291.

Lockhart, B. E. 1., and Betzold, J. A., 1982, Broad bean wilt virus in begonia in Minnesota, Plant Dis. 66:72.

Lockhart, B. E. 1., and Fischer, H. U., 1977, Some properties of an isolate of broad bean wilt virus associated with a field disease of peppers in Morocco, Phytopathol. Z. 88:209.

248 V LISA AND G. BOCCARDO

Lot, H., Marrou, J., Quiot, J. B., and Esvan, C., 1972, Contribution a l'etude du virus de la mosalque du concombre ICMV). II. Methode de purification rapide du virus, Ann. Phytopathol. 4:25.

Lovisolo, 0., 1957, Virus e piante spontanee. 1. "Mosaico lieve del Lamium," nuovo virus di tipo maculatura anulare, Bollettino Staz. Pat. Veg. Roma 15:89.

Lovisolo, 0., Caciagli, P., Lise, V, and Conti, M., 1995, Ecology and biodiversity of broad bean wilt virus in northern Italy. Proc. 8th Conf. Virus Dis. Vegetables, Prague, 1995:73.

Makkouk, K. M., Kumari, S. G., and Bos, 1., 1990, Broad bean wilt virus: Host range, purification, serology, transmission characteristics, and occurrence in faba bean in West Asia and North Africa, Neth. J. Plant Pathol. 96:291.

Markham, R., 1960, A graphical method for the rapid determination of sedimentation coefficients, Biochem. J. 77:516.

Marrou, J., Duteil, M., and Schmidt, R., 1974, Identification de plusieurs virus capables de provoquer un syndrome necrotique sur tomate, Ann. Phytopathol. 6:228.

Marrou, J., Duteil, M., Leclant, E, Cadilhac, B., and Esvan, C., 1976, Mise en evidence du virus du fletrissement de la feve Broad bean wilt (BBWV), dans des laitues presentant des symptomes de marbrure, Ann. Phytopathol. 8:461.

Marzachi, c., Antoniazzi, S., D'Aquilio, M., and Boccardo, G., 1992, Molecular cloning of the segmented RNA genome of ourmia melon virus, Riv. Patol. Veg. 2:3.

Matthews, R. E. E, 1982, Classification and Nomenclature of Viruses. Fourth Report of the International Committee on Taxonomy of Viruses, Karger, Basel, Switzerland lalso in Intervirology 17.

McMaster, G. K., and Carmichael, G. G., 1977, Analysis of single- and double-stranded nucleic acids on polyacrylamide and agarose gels by using glyoxal and acridine orange, Proc. Natl. Acad. Sci. USA 74:4835.

Migliori, A., 1993, Artichoke French latent virus IAFLV): A new fabavirus transmitted by Capitophorus horni, Petria 3:59.

Migliori, A., Lot, H., Pecaut, P., Duteil, M., and Rouze-Jouan, J., 1984, Les virus de l'artichaut. 1. Mise en evidence de trois virus dans les cultures franc;:aises d'artichaut, Agronomie 4:257.

Migliori, A., Homo, E., Corre, J., Marzin, H., Legal, V, and Curvale, J. P., 1987, Repartition, frequence et nuisibilite des virus chez l'artichaut en Bretagne, Pepinieristes-HorticulteursMaraichers, Rev. Hortic. 274:29.

Migliori, A., Rana, G. 1., Piazzolla, P., Gourret, J. P., and Rubino, 1., 1988, Caracterisation d'une nouvelle souche du virus du fletrissement de la feve isolee de l'artichaut en France, Agronomie 8:201.

Milicic, D., Wrischer, M., and Juretic, N., 1974, Intracellular inclusion bodies of broad bean wilt virus, Phytopathol. Z. 80:127.

Milicic, D., Juretic, N., Plese, N., and Wrischer, M., 1976, Some data on cell inclusions and natural hosts of broad bean wilt virus, Acta Bot. Croat. 35:17.

Milne, R. G., and Luisoni, E., 1977, Rapid immune electron microscopy of virus preparations, in: Methods in Virology, Vol. 61K. Maramorosch and H. Koprowski, eds.), pp. 265-281, Academic Press, New York.

Murant, A. E, 1974, Strawberry latent ringspot virus, CMI/AAB Descriptions of Plant Viruses No. 126.

Murant, A. E, 1981, Nepoviruses, in: Handbook of Plant Virus Infections. Comparative Diagnosis IE. Kurstak, ed.), pp. 197-238, Elsevier/North-Holland, Amsterdam.

Murant, A. E, and Roberts, 1. M., 1977, An isolate of broad bean wilt virus serotype II from Helleborus vesicarius, Ann. Rep. Scott. Hortic. Res. Inst. 23:83.

Murphy, EA., Fauquet, C. M., Bishop, D. H. 1., Ghabrial, S. A., Jarvis, A. W., Martelli, G. P., Mayo, M. A., and Summers, M. D.leds.), 1995, Virus Taxonomy-Classification and Nomenclature of Viruses. Sixth Report of the International Committee on Taxonomy of Viruses, Springer-Verlag, Vienna lalso in Arch. Virol., Suppl. 10).

Natsuaki, K. T., Tomaru, K., Ushiku, S., Ichikawa, Y., Sugumura, Y., Natsuaki, T., Okuda, S., and

FABAVIRUSES 249

Teranaka, M., 1994, Characterization of two viruses isolated from patchouli in Japan, Plant Dis. 78:1094.

Polak, I., 1981, Studies of virus diseases of sugar beet and its resistance to these diseases in Czechoslovakia, in: Principles of Plant Resistance to Diseases and Pests, Proc. COMECON Symp., Cluj·Napoca 1979:9.

Provvidenti, R., 1983, Reaction of bean cultivars to broad bean wilt virus, Bean Improvement Cooperative 26:69.

Provvidenti, R., and Hoch, H. c., 1985, Foliar ringspot disease of Thunbergia alata caused by broad bean wilt virus, Plant Dis. 69:726.

Provvidenti, R., Robinson, R. w., and Shail, J. w., 1984, Incidence of broad bean wilt virus in lettuce in New York State and sources of resistance, Hortie. Sci. 19:569.

Remah, A., Jones, A. T., and Mitchell, M. J., 1986, Purification and properties of lucerne Australian symptomless virus, a new virus infecting lucerne in Australia, Ann. Appl. Biol. 109:307.

Rist, D. L., and Lorbeer, J. W., 1989, Occurrence and overwintering of cucumber mosaic virus and broad bean wilt virus in weeds growing near commercial lettuce fields in New York, Phytopathology 79:65.

Robinson, D. J., Barker, H., Harrison, B. D., and Mayo, M. A., 1980, Replication of RNA-1 of tomato black ring virus independently of RNA-2, J. Gen. Virol. 51:317.

Rosciglione, B., and Cannizzaro, G., 1977, Reperimento del virus dell'avvizzimento della fava nell'isola di Linosa, Phytopathol. Medit. 16:140.

Rougier, J., and Marchoux, G., 1974, Nouveaux virus, diagnostiques sur piment et aubergine dans Ie sud-est de la France en 1973, Ann. Phytopathol. 6:227.

Rubio-Huertos, M., 1959, Nota previa sobra un virus de tipo "ringspot" encontrado en Petunia hybrida Vilm., Microbiol. Espanola 12:325.

Rubio-Huertos, M., 1968, Further studies on ultrastructure of plants infected with petunia ringspot virus, Protoplasma 65:465.

Russo, M., and Martelli, G. P., 1975, Light and electron microscopy of broad bean wilt virus intracellular inclusions, J. Submicrosc. Cytol. 7:335.

Russo, M., and Rana, G. L., 1978, Occurrence of two legume viruses in artichoke, Phytopatol. Medit. 17:212.

Russo, M., Martelli, G. P., and Vovlas, C., 1973, Le virosi delle piante ortensi in Puglia. XlV. 11 virus dell'avvizzimento della fava su fava e peperone, Phytopathol. Medit. 12:61.

Russo, M., Kishtah, A. A., and Martelli, G. P., 1979, Unusual intracellular aggregates of broad bean wilt virus particles, J. Gen. Virol. 43:453.

Sahambi, H. S., Milne, R. G., Cook, S. M., Gibbs, A. J., and Woods, R. D., 1973, Broad bean wilt and nasturtium ringspot viruses are related, Phytopathol. Z. 76:158.

Saric, A., Stefanac, Z., and Wrischer, M., 1984, Two rare types of particle aggregates in infections caused by a Yugoslavian isolate of broad bean wilt virus, Phytopathol. Medit. 23:88.

Schmelzer, K., 1960, Untersuchungen tiber das Ringmosaik-Virus der Kapuzinerkresse, Z. Pfl. Krankh. 67:193.

Schmelzer, K., and Stahl, P., 1977, Spontane Wirtspflanzen des Ackerbohnenwelke-Virus. 2. Mitteilung, Zbl. Bakt. II 132:123.

Schmelzer, K., and Wolf, P., 1969, Nachweis des Ringmosaik-Virus der Kapuzinerkresse (nasturtium ringspot virus) in Trompetenbaum (Catalpa bignonioides Walt.) und Mohre (Daucus carata L.), Zbl. Bakt. II 123:557.

Schmelzer, K., Gippert, R., Weisenfels, M., and Beczner, I., 1975, Spontane Wirtspflanzen des Ackerbohnenwelke-Virus (broad bean wilt virus). 1. Mitteilung, Zbl. Bakt. II 130:696.

Schmelzer, K., Stahl, P., and Gebhart, c., 1977, Spontane Wirtspflanzen des AckerbohnenwelkeVirus. 3. Mitteilung, Zbl. Bakt. II 132:209.

Schmidt, H. E., 1981, The diagnosis of viruses of legume crops in German Democratic Republic as a prerequisite for breeding broad bean (Vicia faba L.), bean (Phaseolus vulgaris L.), pea (Pisum sp.), and lupin (Lupinus sp.) for resistance to virus diseases, in: Principles of Plant Resistance to Diseases and Pests, Proc. COMECON Symp., Cluj-Napoca 1979:25.


Schmidt, H. E., and Schubert, 1.,1979, Das Ackerbohnenwelke-Virus (broad bean wilt virus) als Schaderreger bei Lupinen (Lupinus spec.), Arch. PhytopathoI. u. Pflanzenschutz, Berlin 15:169.

Schroeder, W. T., and Provvidenti, R., 1970, A destructive blight of Spinacia oleracea incited by a strain of the broad bean wilt virus, Phytopathology 60:1405.

Schumann, K., 1963, Untersuchungen zur Charakterisierung und Identifizierung der Erreger des Digitalis-Mosaik. II. Das Ringmosaik-Virus der Kapuzinerkresse, PhytopathoI. Z. 48:135.

Scott, S. W., and Barnett, O. W., 1984, Some properties of an isolate of broad bean wilt virus from dogwood (Comus florida), Plant Dis. 68:983.

Selassie-Gebre, K., Marchoux, G., and Quiot, J. B., 1976, Methode d'identification du virus du fIetrissement de la feve (broad bean wilt virus: BBWV) isole de Vigna sinensis (1.) Savi, Ann. PhytopathoI. 8:33l.

Shukla, D., D, and Gough, K. H., 1983, Tobacco streak, broad bean wilt, cucumber mosaic and alfalfa mosaic viruses associated with ringspot of Ajuga reptans in Australia, Plant Dis. 67:22l.

Shukla, D. D., Teakle, D. S., and Gough, K. H., 1980, Periwinkle, a latent host for broad bean wilt and cucumber mosaic viruses in Australia, Plant Dis. 64:802.

Smith, K. M., 1950, Some virus diseases of ornamental plants, 1. Roy. Hortic. Soc. 75:350. Stace-Smith, R., 1981, Comoviruses, in: Handbook of Plant Virus Infections. Comparative

Diagnosis (E. Kurstak, ed.), pp. 171-195, Elsevier/North-Holland, Amsterdam. Stubbs, 1. 1., 1947, A destructive vascular wilt virus disease of broad bean (Vicia faba 1.) in

Victoria, 1. Dept. Agric. Victoria 46:323. Stubbs, 1. 1., 1960, Aphid transmission of broad bean wilt virus and comparative transmission

efficiency of three vector species, Aust. 1. Agric. Res. 11:734. Taylor, R. H., and Stubbs, 1. 1., 1972, Broad bean wilt virus, CMI!AAB Descriptions of Plant

Viruses No. 8l. Taylor, R. H., Smith, P. R., Reinganum, C, and Gibbs, A. J., 1968, Purification and properties of

broad bean wilt virus, Aust. 1. BioI. Sci. 21:929. Tian, W. H., Wang, X. E, Pei, M. Y., and Xie, D. Z., 1982, Studies on stunt mosaic disease of

spinach, Acta PhytophyI. Sin. 9:153. Usugi, T., and Saito, Y., 1979, Satsuma dwarf virus, CMI!AAB Descriptions of Plant Viruses

No. 208. Uyemoto, J. K., and Provvidenti, R., 1974, Isolation and identification of two serotypes of broad

bean wilt virus, Phytopathology 64:1547. Vaira, A. M., Lisa, v., and Luisoni, E., 1992, Diffusione di due ceppi di tomato spotted wilt virus

in Liguria, Inf. FitopatoI. 42(2):59. Vega, J., Gracia, 0., and Feldman, J. M., 1980, Characteristics of argentine broad bean wilt virus:

Cytopathological aspects, PhytopathoI. Z. 99:242. Weidemann, H. 1., Lesemann, D., Paul, H. 1., and Koenig, R., 1975, Das Broad Bean Wilt-Virus

als Ursache fur eine neue Vergilbungskrankheit des Spinats in Deutschland, PhytopathoI. Z. 84:215.

Xi, Z. X., Xu, S. H., and Mang, K. Q., 1982, An isolate causing systemic necrosis mosaic on cowpea in Peking suburb. I. Host range, purification and electron microscopy, Acta PhytopathoI. Sin. 12:38.

Xu, Z. G., Cockbain, A. J., Woods, R. D., and Govier, D. A., 1988, The serological relationships and some other properties of isolates of broad bean wilt virus from faba bean and pea in China, Ann. AppI. BioI. 113:287.

CHAPTER 10

Dianthoviruses: Properties, Molecular Biology, Ecology, and Control R. I. HAMILTON AND J. H. TREMAINE

I. INTRODUCTION

The genus Dianthovirus consists of three definitive species: carnation ringspot virus (CRSV) (the type species), red clover necrotic mosaic virus (RCNMV), and sweet clover necrotic mosaic virus (SCNMV) (Francki et a1., 1991). A possible fourth species is furcraea necrotic streak virus (FNSV), which is serologically related to RCNMV and hybridizes with cDNA clones to each of the two RCNMV genomic RNA species (Morales et a1., 1992). The genus name is derived from Dianthus, the generic name of carnation (D. caryophyllus), which is the most common natural host of CRSV. Dianthovirus particles are isometric, 33 nm in diameter, and sediment as a single species with a sedimentation coefficient (S20 w) of about 133 S at pH 5.0. They contain a single capsid protein with molecular weight of approximately 37 x 103 and two major genomic RNA species with molecular weights of approximately 1.5 x 106 (RNA-I) and 0.5 x 106 (RNA-2). Dianthovirus particles are stable and easily purified with yields up to 100 mg/kg of infected tissue, and are thus well suited to studies of virus structure and replication. However, di-

R. I. HAMILTON AND J. H. TREMAINE • Pacific Agriculture Research Centre, Agriculture and Agri-Food Canada, Vancouver, British Columbia, Canada V6T 1X2.

251

252 R.L HAMILTON AND J. H. TREMAINE

anthoviruses have not been studied as intensively as other plant viruses with bipartite genomes, probably because they do not usually cause serious crop losses. Two comprehensive reviews on the properties of the dianthoviruses have been published within the last few years (Hiruki, 1987; GiesmanCookmeyer et al., 1995) that attest to recent interest in this genus of plant viruses.

Noordam et al. (1951) were probably the first to isolate what is now known as CRSV from carnation showing mosaic symptoms, and named it carnation mosaic virus. Kassanis (1955) studied a carnation virus that was similar if not identical to that of Noordam et al. (1951) but preferred the name CRSV to describe the symptoms induced in seedling carnation and to avoid confusion with a different virus named carnation mosaic virus by American workers. Using serology, host range, thermal inactivation point, longevity in vitro, and electron microscopy, Kassanis (1955) distinguished CRSV from carnation mottle virus, another isometric virus commonly found in carnation. Hollings and Stone (1970) found that CRSV did not react with antisera to any of 35 other isometric viruses.

The first description of symptoms in red clover (Trifolium pratense) and alfalfa (Medicago sativa) infected by RCNMV was by Musil and Matisova (1967) in Czechoslovakia. Musil (1969a,b) isolated RCNMV and distinguished it from other clover viruses by host range, physical properties, electron microscopy, and serology. SCNMV (Hiruki et al., 1984b; Hiruki, 1986b) was first isolated from sweet clover (Melilotus officinalis) exhibiting systemic mosaic, ringspots, and veinal necrosis near Grand Prairie and Athabasca in Alberta, Canada (Hiruki, 1986a). FNSV, the causal virus of "macana" or necrotic streak of fique (Furcraea macrophylla and F. cabuya), a major fiber crop in Colombia, South America, was first reported in the Department of Antioquia in western Colombia, but it is now distributed throughout the main fique-production areas of the Andean regions of Colombia (Dabek and Castano, 1978).

The first report pointing to a bipartite genome for CRSV was the observation of two distinct RNA components in density gradient centrifugation of the products of sodium dodecyl sulfate (SDS) dissociation of virus particles (Tremaine and Ronald, 1976). Ragetli and Elder (1977) reported a similar result on SDS treatment of clover primary leaf necrosis virus, first described as a new virus but now considered a strain of RCNMV (Rao and Hiruki, 1985) and designated here as RCNMV-Can. Dodds et al. (1977) established that both RNA species of CRSV were required for infectivity, thus proving the bipartite nature of the dianthovirus genome.

II. HOST RANGE, SYMPTOMS, AND GEOGRAPHICAL DISTRIBUTION

CRSV has caused serious problems in carnations grown by vegetative propagation (Kassanis, 1955; Hollings and Stone, 1970; Lommel et al., 1983).

DIANTHOVIRUSES 253

Symptoms in naturally infected carnation and sweet william (D. barbatus) include curling or distortion of leaves, occasionally with tip necrosis (Hollings and Stone, 1965). Flowers are often distorted, of poor quality, and are thus unmarketable. The experimental host range of CRSV includes over 130 species in 25 dicotyledonous families (Tremaine and Dodds, 1985). With the establishment of virus-free certified plant propagation material and the use of serological indexing and strict sanitation procedures, CRSV has been virtually eliminated from commercial carnation plantings in western Europe (Sparnaaij, 1983), North America (Lommel et a1., 1983), and Israel (Mor, 1983), but a high proportion (33-53 %) of Sim, Mediterranean, and spray types of carnation were infected in commercial production nurseries near Bogota, Colombia (Valenzuela and Pizano, 1992).

CRSV has been mechanically transmitted to herbaceous hosts from pear (Pyrus communis) trees with stony pit disease, from apple (Malus sylvestris) and sour cherry (Prunus cerasus) trees with decline syndromes (Richter et a1., 1978; Kleinhempel et a1., 1980), and from plum (Prunus domestica) trees (Casper, 1976). It has also been detected serologically in petals, leaves, fruits, roots, and cambium tissues from apple trees with apple spy decline and pear trees with pear stony pit and vein yellows diseases. CRSV was also mechanically transmitted to some apple virus indicator plants as evidenced by reisolation on herbaceous hosts. However, the significance of CRSV in the etiology of fruit tree diseases remains uncertain.

Weeds, too, were naturally infected with CRSV in these CRSV-infected orchards. Stellaria media was the species most frequently infected, and in tests with 1000 plants, CRSV infection occurred in 8 % at some individual sites, and in 32 and 70% at two sites in a 5-year-old nursery containing CRSV-infected apple trees. CRSV infection was detected in 21 of 40 Urtica urens plants and in 4 of 40 Poa annua plants in another CRSV-infected apple orchard (Kleinhempel et a1., 1980). CRSV-infected plants of all three species were symptomless. CRSV infection of S. media in vineyards has also been reported (Rudel et a1., 1977). The role of weeds as sources of virus for infection of tree fruits and grapes remains to be determined.

RCNMV infects red clover, alfalfa, white clover (Trifolium repens), and sweet clover growing as field crops or in pastures, and has been reported in northern Europe, Australia, New Zealand, and Canada (Hollings and Stone, 1977). In naturally infected red clover, the leaves show a mild to severe mosaic, with mild to severe necrosis of the veins accompanied by leaf distortion, and the plants show moderate to severe stunting in the winter (Musil, 1969a,b). In the summer these symptoms are wholly or partially masked, particularly in newly infected plants. Inoculated red clover plants show similar symptoms but local necrotic lesions are prominent. Sweet clover plants show mild to severe systemic mosaic symptoms, probably dependent on the season in which the plants were infected or observed. Okuno et a1. (1983), using controlled growth chambers maintained at 17 DC and 26 DC in a study of pseudorecombinants of three strains of RCNMV, determined that temperature was critical in obtaining reproducible symptoms for use as

254 R. I. HAMILTON AND J. H. TREMAINE

genomic markers. The wide variation in symptomatology and host range reported for RCNMV by Musil (1969a), Gerhardson and Lindsten (1973), Hollings and Stone (1977), Ragetli and Elder (1977), and Hiruki (1987) was probably due in part to studies being done with plants at different temperatures or with less strictly regulated temperatures. However, virus strain differences (Hiruki, 1987), differences in susceptibility of plant cultivars to RCNMV (Bowen and Plumb, 1979), and differences in inoculation technique were probably involved as well.

The natural host of SCNMV is sweet clover. The virus was not detected in alsike clover (Trifolium hybridum), red clover, or crown vetch (Coronilla varia) growing on sites containing SCNMV-infected sweet clover (Hiruki, 1986a), but a serologically distinct strain was subsequently isolated from alfalfa showing very mild chlorosis and mild stunting in the same area (Inouye and Hiruki, 1985; Pappu et a1., 1988). The geographical distribution of both strains appears to be limited to the province of Alberta, Canada.

Fique plants infected for a prolonged time with FNSV are stunted and the leaves show chlorotic streaks on both sides, which later coalesce and become necrotic. Chronically infected plants usually die (Dabek and Castano, 1978). The natural host range is restricted to fique, but the virus can be mechanically inoculated to F. selloa var. marginata and sisal (Agave sisalana).

Although natural infections are limited to a few species, the experimental host ranges of CRSV, RCNMV, and SCNMV include many plant species in at least 25 dicotyledonous families (Hollings and Stone, 1970, 1977; Hiruki et a1., 1984b; Pappu et a1., 1988). RCNMV does not infect carnation or sweet william, but CRSV infects crimson clover (Trifolium incarnatum) systemically (H. W. J. Ragetli, unpublished results). The alfalfa isolate of SCNMV could be distinguished from the sweet clover isolate by its inability to infect tomato (Lycopersicon esculentum) (Pappu et a1., 1988). Plants naturally infected by dianthoviruses typically exhibit mild mosaic symptoms that are often alleviated at temperatures above 20 cC.

III. VIRION PROPERTIES

A. Structure

Dianthovirus particles are probably composed of either a single RNA-l molecule (molecular weight 1.5 x 106) or three RNA-2 molecules (molecular weight 0.5 x 106), encapsidated by 180 protein subunit molecules of 37 to 39 kDa arranged in aT = 3 structure (Fig. 1) The weight of such particles would thus be 8.2 to 8.5 MDa. The particle weight of CRSV determined by sedimentation-diffusion is 7.1 MDa (Kalmakoff and Tremaine, 1967). The larger values seem more consistent with those for the tombusviruses, which are similar in size but have a monopartite genome.

Most freshly prepared dianthovirus particle preparations contain a sin-

DIANTHOVIRUSES 255

B

FIGURE 1. Structure of dianthovirus virions. (A) Electron micrograph of purified RCNMV particles in negative stain. The particles are approximately 33 nm in diameter. (B) Model of dianthovirus virus particle based on crystallographic analysis of turnip crinkle and tomato bushy stunt tombusviruses. Individual capsid subunits are labeled A, B, and C, depending on their packing environment. Note the granular nature of the surface of the virus particles shown in (A), which is presumably due to the protruding domains (P) of the capsid subunits. (After Giesman-Cookmeyer et a1., 1995.)


gle 37- to 39-kDa protein component (Barton, 1981; Lommel, 1983; Hiruki et a1., 1984b; Tremaine and Dodds, 1985). However, some preparations stored at 4 °C contain 36-kDa and 34-kDa proteins, presumably formed by proteolytic degradation of the 37- to 39-kDa protein. This limited proteolysis occurs to a greater extent with CRSV than with RCNMV (Barton, 1981; Tremaine and Dodds, 1985; J. H. Tremaine, unpublished results). The amino acid compositions of four strains of CRSV reported by Kalmakoff and Tremaine (1967) and Tremaine et a1. (1984) (Table I) were of preparations containing a single 38-kDa protein. Results of similar analyses (T. H. Tremaine, unpublished results) of RCNMV-Can (Ragetli and Elder, 1977) and RCNMV-V from Victoria, Australia (Gould et a1., 1981), here designated as RCNMV-Aus, are also shown in Table I. There are no major differences in amino acid composition between the two RCNMV strains or among the four CRSV strains. The amino acid compositions of the individual dianthoviruses have many differences but are basically similar.

TABLE I. Number of Amino Acid Residues in Capsid Proteins of Two Strains (Aus and Can) of Red Clover Necrotic Mosaic Virus (RCNMV)

and Four Strains (R, N, A, K) of Carnation Ringspot Virus (CRSV)

RCNMV CRSV

Amino Acid Ausa Cana Rb Nb Ab Kc

Lys 16 14 IS 16 IS 14 His 2 2 2 2 3 2 Arg 18 21 IS 17 16 16 Asx 32 31 33 35 34 34 Thr 34 32 38 38 38 37 Ser 38 37 39 38 38 37 Glu 28 26 22 23 23 23 Pro 18 23 19 19 17 20 Gly 25 28 23 24 23 20 Ala 30 24 24 26 27 24 Cys 5 6 2 3 3 3 Val 34 30 36 30 32 36 Met 3 3 6 4 6 7 Ile 14 16 16 IS 14 16 Leu 24 26 27 27 27 26 Tyr 11 11 IS IS IS 16 Phe 12 14 13 12 12 12 Trp nri nr 3 3 4 4 Total 344 344 348 347 347 347

aJ. H. Tremaine (unpublished data). liTremaine et ai. (1984). cKalmakoff and Tremaine (1967). dm, Not reported.

DIANTHOVIRUSES 257

The molecular weights of dianthovirus RNA species, estimated by polyacrylamide gel electrophoresis (PAGE) under denaturing conditions, are 1.5 x 106 (RNA-I) and 0.5 x 106 (RNA-2) for both RCNMV (Gould et a1., 1981) and CRSV (Lommel, 1983). By cDNA hybridization studies, Gould et a1. (1981) established that RCNMV RNA-l and RNA-2 are unique RNA species with little or no sequence homology. Lommel (1983) did not detect sequence homology between the RNA species of CRSV and RCNMV, but significant sequence homology has been reported at the amino acid level for several gene products of CRSV, RCNMV, and SCNMV (Ge et a1., 1992j Kendall and Lommel, 1992j Ryabov et a1., 1994). The significant sequence homology between RCNMV and FNSV (Morales et a1., 1992) suggests that FNSV may be a strain of RCNMV.

The nucleotide composition of unfractionated CRSV RNA is G, 26 % j A, 27%j C, 23%j and U, 24% (Kalmakoff and Tremaine, 1967). The lack of retention of CRSV or RCNMV RNA molecules on oligo(dT) cellulose columns indicates the absence of 3' poly(A) sequences (Lommel, 1983j Tremaine and Dodds, 1985), and no poly(A) tracts have been detected in dianthovirus genomes by nucleotide sequence analysis. The stem-loop regions at the 3' end of the dianthovirus RNA species, which are involved in the interaction with the virus-specific RNA polymerase, must be similar, if not identical, because pseudorecombinants made from RNA-1 of RCNMV and RNA-2 of CRSV (Lommel, 1983), and both possible pseudo recombinants of RCNMV and SCNMV (Okuno et a1., 1983), are infective, usually inducing local or systemic symptoms similar to those induced by RNA-1 of the parental virus.

CRSV preparations with native 38-kDa protein give a single band with a buoyant density of 1.366 g cm-3 in CsCI gradient centrifugation. Additional components with buoyant densities of 1.369 and 1.374 g cm-3 are present in virus preparations containing partially proteolyzed protein (Tremaine and Dodds, 1985). The major component of RCNMV has a buoyant density of 1.37 g cm -3 in CsCI, but a minor component with a greater density has also been observed (Gould et a1., 1981). This indicates that some viruslike particles contain a considerable amount of the lower-molecular-weight protein. Infectivity is associated with the single buoyant density component, which indicates that the particles containing one RNA-1 molecule have the same density as those containing three RNA-2 molecules, and therefore the size of RNA-2 must be one third that of RNA-I. Subgenomic RNA species that code for the coat protein may also be encapsidated in RCNMV particles (MorrisKrsinich et a1., 1983).

Direct evidence of the packaging of one molecule of RNA-1 in one particle type and three molecules of RNA-2 in another particle type is not obtainable because of the single buoyant density. Packaging of both RNA species in the same particle as occurs with Nodamura virus, a bipartite genome virus of invertebrates (Longworth, 1978), is improbable because the molar ratios of RNA-1:RNA-2 in CRSV are 1:1.7 to 1:2.8. Moreover, most of


the RNA released from CRSVparticles in 0.01 % SDS is RNA-2, and RNA-I is released at higher SDS concentrations (Tremaine and Ronald, 1976).

The isometric particles of CRSV; FNSV; RCNMV; and SCNMV are 34-35 nm in diameter (Tremaine and Ronald, 1976; Hatta and Francki, 1984; Hiruki et a1., 1984b; Morales et a1., 1992). In a comparative electron microscopy study of isometric particles from ten virus groups, Hatta and Francki (1984) distinguished RCNMV from all viruses except tomato bushy stunt tombusvirus (TBSV) on the basis of size (34.2 nmL round particle outline, and rough surface appearance. CRSV particles appear similar to those of RCNMV but the particle surface is less rough. Orlova et a1. (1980) noted that the capsomeric structure of CRSV is very similar to that of TBSV, which consists of 90 dimers and is characterized by 5:3:2 symmetry. In two-dimensional crystal specimens of CRSV the distance between particle centers in the hexagonal packing of particles is 34 to 35 nm.

B. Infectivity-Dilution Curves

Infectivity dilution data for CRSV (J. H. Tremaine and W. P. Ronald, unpublished results) provide another line of evidence for the independent packaging of RNA-I and RNA-2. These data were fitted to four equations: Eq. (I), a single-hit curve for one particle; Eq. (2), a two-hit curve for identical particles; Eq. (3), a two-hit curve for two distinct particle types A and B in equal quantities; and Eq. (4L a two-hit curve for two distinct particle types A and B in any ratio. These equations are:

y = N(I - eCx )

y = N[(I - ecx ) - cx(ecx )]

y = N(l - ecx )2

y = N(l - eQx)(I - ebx)

(1)

(2)

(3)

(4)

where y = number of lesions, N = maximum possible number of lesions, c =

total number of virus particles, x = dilution, and a and b are the numbers of the A and B particle types, respectively. After fitting the experimental data to the four equations, the summation of squared residuals was 21.5 for Eq. (I), 8.7 for Eq. (2), 6.0 for Eq. (3), and 1.9 for Eq. (4). The ratio of alb derived from the best fit to Eq. (4) was 4:1. The molar quantity of RNA-2 found on SDS dissociation of this virus preparation was 66% of that of RNA-I. If three RNA-2 molecules were packaged in a single particle, the ratio of particles containing RNA-I to particles containing RNA-2 would be 4.5 to 1. Therefore, the infectivity-dilution data are best approximated by Eq. (4), i.e., two distinct particle types are required for infection.

DIANTHOVIRUSES 259

C. Virion Stability

Dianthovirus particles have many characteristics in common with bromovirus particles, which are stabilized by RNA-protein interactions and pH-dependent protein-protein interactions (Kaper, 1975). However, some strains of CRSV are more stable than others, and RCNMV is more stable than CRSV. This variation of stability within members of the same virus group is not unexpected. For example, nitrous acid mutants of cowpea chlorotic mottle bromovirus have been isolated that are either much more stable or much less stable than the type virus (Lane, 1974).

CRSV particles swell slowly on change of pH from 5.0 to 7.5 (Tremaine and Ronald, 1976, and unpublished results; Kuhne and Eisbein, 1983). Three naturally occurring CRSV strains, designated N, A, and R (Tremaine et a1., 1976), differ in the proportion of virus particles swollen 1 hr after adjustment of the pH to 7.0 or 7.5. Most particles of all strains are swollen after 24 hr at pH 7.5. EDTA increases the rate of swelling, but Mg or Ca ions prevent it. When swollen CRSV-N particles are treated with ribonuclease, isometric particles of 22-nm diameter (probably T = 1 structures), ellipsoidal particles 22 x 33 nm and double-shelled particles of 33 nm diameter are formed (J. H. Tremaine and W. P. Ronald, unpublished results). Addition of trypsin to CRSV-N at pH 5.0 has no effect, but the 38-kDa protein is cleaved to 36-kDa and 34-kDa proteins at pH 7.0 to 8.0 and into 20-kDa to 26-kDa species at pH 9.0 (T. H. Tremaine and W. P. Ronald, unpublished results). The presence of contaminating plant proteases in some virus preparations is indicated by a limited proteolysis on storage (i.e., presence of 36-kDa and 34-kDa proteins), and this proteolysis is more evident in preparations stored at pH 7.0 than at pH 5.0 (Barton, 1981; J. H. Tremaine and W. P. Ronald, unpublished results). The purification of CRSV by ion exchange chromatography (Tremaine, 1961) or by CsCI density gradient centrifugation reduces proteolysis on subsequent storage.

Approximately 30% of the particles in a preparation of RCNMV-Can were swollen after 1 hr at pH 7.0 (J. H. Tremaine and W. P. Ronald, unpublished results). However, Lommel (1983) did not detect swelling of RCNMVAus at pH 7.0. Barton (1981) did not detect limited proteolysis of RCNMV in stored preparations at pH 7.0, but we have noted proteolysis on storage, as have Gould et al. (1981).

After 24 hr, approximately 50% of CRSV-N particles dissociate into RNA and protein components in 0.1 M Tris-HCI, pH 7.5,10 mM EDTA, and 1 M NaCI (J. H. Tremaine and W. P. Ronald, unpublished results). The virus particles can be reconstituted from RNA and protein components by dialysis against 0.1 M Tris-acetate buffer, pH 5.0, containing 1 mM CaCI2. The particles are heterogeneous and some geminate particles are formed. The protein did not assemble into empty capsids at pH values from 4 to 8. However, in the presence of a nucleating agent, sodium dextran sulfate (NDS), the protein formed misshapen particles of 20 to 33 nm diameter.

260 R. 1. HAMILTON AND J. H. TREMAINE

CRSV strains differ in the readiness with which their particles dissociate into RNA and protein after treatment at room temperature for 30 min with SDS at pH 7.0 (Tremaine and Ronald, 1976). Particles of strain N are almost completely dissociated by 0.4% SDS at pH 5.0 and by 0.1 % SDS at pH 7, but those of strains R and A were almost completely unaffected at pH 5 by SDS concentrations up to 15%. However, in 0.05% SDS at pH 7, some particles dissociated, some were unaffected, and others swelled to 49 nm diameter but remained resistant to higher SDS concentrations. RCNMV-Can particles, too, are resistant to dissociation by SDS at pH 5.0. Approximately 20% of the virus particles were swollen after 1 hr at pH 7.5 and 50% of the virus particles were dissociated in 1 % SDS (J. H. Tremaine and W. P. Ronald, unpublished results). Results from pseudorecombination experiments between CRSV-A and CRSV-N indicate that the capsid protein gene, which is located on RNA-I, probably controls sensitivity of CRSV to SDS (Dodds et a1., 1977).

Virions stabilized by RNA-protein interactions are very much like polyelectrolyte complexes when in the swollen condition. When increasing amounts of NDS were added to CRSV at pH 7.5, increasing amounts of RNA were liberated until at an NDS-virus ratio of 1:4 by weight, nearly all the virus particles were dissociated (Tremaine et a1., 1983). The acidic polyelectrolyte NDS has a greater affinity for the coat protein than does the RNA. Isometric viruslike particles (22 nm or 33 nm diameter) with characteristics of T = 1 and T = 3 particles, respectively, formed of NDS and the CRSV protein were found in these preparations by electron microscopy. NDS had no effect on unswollen virus particles at pH 5.

The addition of one part NDS to four parts RCNMV-Can by weight at pH 7.5 dissociated only 20% of the virus particles; 60% of the particles remained stable; and 20% formed particles that were greatly swollen (J. H. Tremaine and W. P. Ronald, unpublished results). All virus particles were dissociated at pH 8.25, but capsid protein subunits reassembled with NDS to form both T =

1 and T = 3 particles, which can be readily distinguished from virus particles by electron microscopy.

D. Aggregation

Particles of CRSV strains N, A, and R differ in their tendency to aggregate in suspension (Tremaine et a1., 1976). Strain A forms clusters of 12 virus particles and aggregates of linked clusters following high-speed centrifugation. This type of aggregation has been observed with a strain of tobacco necrosis satellite virus and with radish mosaic C comovirus (Kassanis and Woods, 1968; Kassanis et a1., 1973). Virus particles of strains R andN of CRSV aggregated at 25°C and 40 °C, respectively, but disaggregated to monomers at lower temperatures, and the effects of virus concentration and some chemicals on this endothermic process were studied by Kuhne et a1. (1983), Tremaine et a1. (1984), and Tremaine and Ronald (1985).

DIANTHOVIRUSES 261

E. Serological Properties

Tremaine et a1. (1976) readily detected serological differences between A and either N or R strains of CRSV in immunodiffusion tests. A serological reaction of partial identity between strains Nand R was consistently observed by immunodiffusion against CRSV-N antiserum at low dilution (1:2), and this result was confirmed by intragel adsorption tests (Tremaine et a1., 1984). A CRSV isolate from pear appeared identical to a carnation isolate (Kleinhempel et a1., 1980); however, extensive tests for serological differences were not done.

Musil and Gallo (1982) distinguished three serotypes of RCNMV: A (RCNMV-TpM34) and B (RCNMV-TpM48), both from Czechoslovakia (Musil, 1969b), and C (RCNMV-Sw) from Sweden (Gerhardson and Lindsten, 1973). Musil et a1. (1982) examined 34 RCNMV isolates from different regions of Czechoslovakia and found A, Band C serotypes separately and in mixtures. Their distribution was not limited to definite geographical areas. These serotypes had different electrophoretic mobilities and could be distinguished by immunoelectrophoresis with mixed antisera to serotypes A, B, and C.

RCNMV isolates G from England, S from Scotland, and Aus from Australia have strong cross-reactions with antiserum to serotype B (Hollings and Stone, 1977; Gould et a1., 1981). However, a serological analysis by the intragel adsorption test of six RCNMV isolates, Aus, Can, Eng (Bowen and Plumb, 1979), Sw (serotype C), TpM34 (serotype At and TpM48 (serotype B) clearly demonstrated that RCNMV-Can belongs to serotype B and that RCNMVAus and RCNMV-Eng are members of a new serotype, D (Rao et a1., (1987). Interestingly, adsorption of the RCNMV-Aus antiserum with homologous antigen or the heterologous RCNMV-Eng antigen completely removed the homologous antibodies, but the remaining antibodies were heterospecific, reacting with RCNMV-Can, -Sw, and -TpM48 but not with the homologous antigen. None of the antisera to the other viruses contained heterospecific antibodies that occur in some polyclonal antisera to tobacco mosaic tobamovirus (TMV) (Van Regenmortel, 1982).

Hiruki et a1. (1984a) found that direct, double antibody sandwich enzyme-linked immunosorbent assay (ELISA) failed to detect any cross-reaction between antigens of RCNMV, SCNMV, or CRSV and their antisera. In indirect ELISA, however, where the combination of bound antigen and antivirus antibody was detected by antiglobulin antibodies from a different animal species, an antiserum to RCNMV-Can gave weak cross-reactions with two other RCNMV strains and with CRSV and SCNMV, and an antiserum to CRSV gave somewhat weaker cross-reactions with all three strains of RCNMV and with SCNMV. Different results in direct and indirect ELISA are not unexpected because the conformation of plate-bound antigens is often more distorted than that of antibody-bound antigens, resulting in exposure of different epitopes in the protein for subsequent binding of antibodies and


giving a consequential increase in cross-reactivity (Van Regenmortel and Burckard, 1980). Cross-reactions between CRSV and RCNMV were much greater with RCNMV and CRSV antisera, respectively, in indirect ELISA when the dissociated capsid proteins were used as antigens (Tremaine and Dodds, 1985). Hiruki and Figueiredo (1985) found that two SCNMV monoclonal antibodies consistently cross-reacted with CRSV, RCNMV-Aus, and RCNMV-Eng, but cross-reacted poorly with RCNMV-Can and RCNMVTpM48 from Czechoslovakia.

Bercks and Querfurth (1972) detected weak cross-reactions between particles of some tymoviruses and one CRSV antiserum in gel diffusion tests in 0.25 M phosphate buffer. This may be attributable to formation of nonspecific precipitates in this buffer in the absence of NaCl (Altschuh and Van Regenmortel, 1983). Given the many differences in the properties of dianthoviruses and tymoviruses, it is highly unlikely that their capsids are serologically related.

F. Electrophoretic Mobility

The electrophoretic mobility of CRSV particles was determined in free boundary Tisehus electrophoresis in the pH range 4 to 8 (J. Kalmakoff and J. H. Tremaine, unpublished results). The virus moved as a single discrete boundary in the pH range 4 to 6.5 with an isoelectric point at pH 5.1. As the pH was increased to 7.0 and 7.5, the moving boundary spread to the anode and continued to spread when the polarity was reversed. Migration in agarose was similar. Kuhne et al. (1983) also obtained discrete, non spreading migration of CRSV in immunoelectrophoresis at pH 5.0 and 6.5 in 0.01 M phosphate buffers containing 0.001 M MgSO 4' Electrophoretic spreading may be a result of virion swelling, as indicated by electrophoresis studies with Ca or Mg ions at pH 7 to 8, or to changes in the capsid induced by limited proteolysis of the virus.

Gallo and Musil (1984a) determined the electrophoretic migration of serotypes A, B, and C of RCNMV in agarose containing phosphate at pH 7.2 and 8.0, or Tris-HCI, sodium diethylbarbiturate, or HEPES at pH 7.2 and 8.6, with buffer molarities of 0.1. 0.01 and 0.001 M. Migration at all molarities in HEPES was cathodic. Migration in most other buffers was anodic at 0.1 M, more slowly anodic at 0.01 M, but cathodic at 0.001 M. In isoelectric focusing in agarose containing 6.3% Pharmalyte 3-10, the isoelectric points of the A, B, and C serotypes were pH 5.0, 4.8, and 4.6 respectively, but the A and B serotypes each contained a minor component with an isoelectric point of pH 4.2 (Gallo and Musil, 1984b). The minor components may be virions in which the coat protein had been partially proteolyzed. These isoelectric points indicate that the virion should migrate anodically above pH 5 in agarose electrophoresis. The effect of buffer molarity on this migration may have been induced by swelling of the virion. On electrophoresis in Ionagar #2 in

DIANTHOVIRUSES 263

0.03 M phosphate buffer, pH 7.6, Hollings and Stone (1977) noted that five RCNMV isolates migrated to the cathode: C and A migrated slowly, B more rapidly, and G and S much more rapidly. Assuming considerable electroendosmotic flow in agar with a high sulfate content, the migration of A, B, and C is consistent with the results of Gallo and Musil (1984a) in phosphate buffers. The movement of G and S indicates they are similar to B but with a lower negative charge.

A comparative study of the electrophoretic mobility of seven isolates of the definitive dianthoviruses in agarose under nondenaturing conditions in 0.01 M phosphate, pH 7.0, showed considerable and reproducible variation between isolates of the same virus (Pappu and Hiruki, 1989). All viruses moved anodically: CRSV-N moved slowest; RCNMV-TpM34 and RCNMVTpM48 moved more rapidly but at identical rates; and SCNMV-38 (sweet clover isolate), CRSV-A, and RCNMV-Aus moved fastest and slightly ahead of SCNMV-59 (alfalfa isolate). These variations can probably be explained by differences in the amino acid composition of the capsid proteins through mutation, possibly through adaptation to a new host, as can be suggested for the origin of the alfalfa isolate of SCNMV.

IV. MOLECULAR BIOLOGY

A. Pseudorecombinants

Pseudorecombination experiments allow the assignment of various characteristics of the dianthoviruses to the appropriate genome segment. Formation of pseudorecombinants was thought to be possible for any combination of RNA-1 and RNA-2 within and among the definitive dianthoviruses (Okuno et a1., 1983), but there is now one reported exception (Rao and Hiruki, 1987L which will be discussed below.

The aggregation and serological properties of pseudorecombinants of the A and N strains of CRSV (Dodds et a1., 1977) and the Can and Sw strains of RCNMV and SCNMV (Okuno et a1., 1983) demonstrated that the coat protein cistron is located on RNA-I. Dianthoviruses and the unrelated pea enation mosaic enamovirus (Chapter 12) are the only segmented genome plant viruses in which the coat protein gene is located on the larger genome segment.

Okuno et a1. (1983) also studied the host response at 17 °C and 26°C and serological properties of pseudorecombinants derived from homologous and heterologous mixtures of RNA species from SCNMV; RCNMV-Can, and RCNMV-Sw. The two RNA species in the pseudorecombinants maintained their original electrophoretic mobilities, which differed slightly among those of the parental viruses. Systemic infection of sweet clover at 26°C was attributed to RNA-1 of SCNMV in pseudorecombinants of SCNMV and RCNMV-Sw. SCNMV, RCNMV-Can, and RCNMV-Sw did not infect white


clover at 26 ec, but pseudorecombinants containing RNA-l of RCNMV-Can and RNA-2 of SCNMV (Canl:S2) or RNA-l of RCNMV-Can and RNA-2 of RCNMV-Sw (Canl:Sw2) caused local infection in this host. White clover was not infected at either temperature by any pseudorecombinants containing SCNMV RNA-I. Pseudorecombinants SI:Can2 and Canl:S2 produced brownish rings on primary leaves of Red Kidney bean, a response that is intermediate between the whitish spots induced by RCNMV-Can and the rapid death following necrotic ringspots and veinal necrosis induced by SCNMV. These results clearly indicate that an interaction between RNA-l and RNA-2 is involved in the type of host response to infection by dianthoviruses.

Lommel (1983) obtained an enhancement of RNA-l infectivity with the pseudorecombinant of RCNMV RNA-l and CRSV RNA-2 (Rl:C2) but not with Cl:R2. On serial passage in a systemic host (Nicotiana clevelandii) the yield of virus from Rl:C2 gradually increased so that after 12 passages it was equal to those obtained from plants infected with the parental viruses. During these passages the electrophoretic mobility of the dsRNA of CRSV RNA-2 in Rl:C2 gradually decreased. The recombinant nature of Rl:C2 was established by cDNA hybridization, and further cDNA studies with SI nuclease indicated small sequence differences between CRSV RNA-2 and the RNA-2 of the stabilized Rl:C2. The host range of stabilized Rl:C2, including the inability to infect carnation, was identical to that of RCNMV. However, the symptoms induced in N. clevelandii by Rl:C2 differed from those of RCNMV and CRSV, suggesting an interaction between RNA-l and RNA-2.

The exception to the general observation that pseudorecombination is possible between any two dianthoviruses occurred with RNA-l of RCNMVTpM48 and RNA-2 of RCNMV-TpM34 (Rao and Hiruki, 1987). When this mixture was inoculated to Chenopodium quinoa, no viral particles were detected by immunosorbent electron microscopy using antisera to each of the parental viruses, the infectivity of the RNA mixture was not significantly different from that of each RNA alone, and no viral RNA could be detected in the inoculated plants by dot blot nucleic acid hybridization, suggesting that replication of viral RNA did not occur. Moreover, Northern hybridization analysis indicated that cDNA probes to unfractionated RNA of each virus hybridized slightly with the heterologous virus at 42 ec but only with the homologous viral RNA at 55 ec. These results, coupled with the facts that RCNMV-TpM34 and RCNMV-TpM48 belong to serotypes A and B, respectively (Musil and Gallo, 1982), and that RCNMV-Can, which also belongs to serotype B, forms pseudorecombinants with RCNMV-Sw and SCNMV, suggest that RCNMV-TpM48 is distinctly different from other RCNMV strains and that its classification as a strain of RCNMV, rather than as a separate dianthovirus, should be reconsidered. This suggestion must be assessed in light of the fact that the isolates of RCNMV-TpM34 and RCNMV-Tp48 used by Rao and Hiruki (1987) differed from those used by earlier workers in being able to infect, respectively, cowpea and Nicotiana

DIANTHOVIRUSES 265

c1evelandii. Possibly these isolates represent host range mutants that retained their serological specificity but have an altered movement protein that is not expressed or, if expressed, is not functional in these hosts.

B. dsRNA Species

dsRNA species are involved in the replication of ssRNA plant viruses and are also useful adjuncts in the characterization of RNA viruses and in their diagnosis (Dodds et a1., 1984). Two major dsRNA species and a minor one were detected by sucrose density gradient centrifugation of extracts from cowpea (Vigna unguiculata) infected with CRSV strains A and N (Dodds et a1., 1977) and in bean infected with several isolates of RCNMV (Osman and Buck, 1991a). They were not detected in extracts from virus-free plants, indicating that they were specific RNA species induced by dianthovirus infection. The sizes of the two largest CRSV dsRNA species (A, B) suggested that they corresponded to the two genomic ssRNA species, and this was confirmed by the finding that, after denaturation at high temperature, they migrated at the same rate as the corresponding genomic ssRNA species. However, the relationship of the smallest RNA to the other two could not be immediately established. The mobility of the largest dsRNA (A) of RCNMV was similar for all isolates, and its estimated size (4.4 kbp) suggested that it corresponded to genomic RNA-I. The mobility of dsRNA-B was also uniform, with an estimated size of 1.4 kbp, but that of dsRNA-C was 1.2-1.3 kbp, depending on the virus isolate. Subsequent Northern blot analysis, using cDNA to RNA-l plus RNA-2, or cDNA to RNA-l only, indicated that (1) dsRNA-A was derived from RNA-I, (2) dsRNA-B corresponds to a subgenomic RNA derived from RNA-I, and (3) dsRNA-C is the dsRNA form of RNA-2. Lommel (1983) found that the dsRNA counterpart of genomic CRSV RNA-l migrated faster than that of RCNMV, but the dsRNA-2 of CRSV migrated slower than that of RCNMV. This migration pattern of dsRNA paralleled differences in migration of the genomic RNA species of CRSV and RCNMV in nondenaturing gels. These results indicate that the genomes of definitive dianthoviruses are similarly organized and that they are distinct from those of other viruses (Morris, 1983).

C. Genome Strategy and Gene Function

1. Genome Organization

The dianthovirus genome consists of two distinct, positive-sense RNA species of about 4.5 kb (RNA-I) and 1.5 kb (RNA-2). The RNA-l molecules of different dianthoviruses are similar in length [3756 bases for CRSV-type, 3889 for RCNMV-Aus, and 3876 for SCNMV-59 (alfalfa isolate)]; the RNA-2 molecules have lengths of 1394 bases for CRSV, 1448 for RCNMV-Aus, and

266 R.I. HAMILTON AND J. H. TREMAINE

1446 and 1449 for SCNMV-38 (sweet clover isolate) and SCNMV-59, respectively. The complete nucleotide sequence of each of these genomes has been published (Ryabov et a1., 1994; Kendall and Lommel, 1992; Xiong and Lommel, 1989; Osman et a1., 1991a; Ge et a1., 1992, 1993). The genomic organization of RCNMV is shown in Fig. 2. Those of CRSV and SCNMV are similar and so probably is that of FNSY, although its genome has not been sequenced.

The 5' terminus of each genomic RNA species of RCNMV has a m7GpppA cap (Xiong and Lommel, 1989) and it is likely that those of the other dianthoviruses are similarly capped. The 3' termini of the genomic RCNMV RNA species and the other definitive dianthoviruses are not polyadenylated, but each has a nearly identical sequence of 27 nucleotides (Xiong and Lommel, 1989) that is capable of forming a stable stem-loop structure

RCNMV

RNA-l

813 831 Q K S LEO ~ . 9j 926

AAA~UCCCU~ G~ GA~~UA~ GC~~

_________ 24232427 3444

123 Jr;==:::::Ep~5?.7==~~~~t. ~=:iPQ;3~7c~51~-.,j? 3889 m7G ~r'g26 •

L-__ -'--.:.!8.:J31 subgenom;c RNA ~[~R ~.~pi3i7i~~p?-~? 27kDa? [

37 kDa capsid

88 kDa polymerase

84 1034

RNA-2 m7G Ht==:ip~3~5=~jl--.J'i' 1448

I I 35 kDa movement

FIGURE 2. Organization and expression of the RCNMV genome. Untranslated regions in RCNMV RNA-l and RNA-2 are depicted as solid lines and the ORFs are depicted as open rectangles. Vertical bars at the 5' ends of RNA-l and RNA-2 and near the 3' terminus of the p57 ORF identify a l3-nucleotide conserved sequence. The right-angle arrow identifies the probable location of the beginning of the capsid protein subgenomic RNA. The stem-loop at the 3' end of both viral RNA species represents the 3-terminal2 7 homologous nucleotides that are capable of forming a stable stem-loop structure. The diagonal line identifies the region of ribosomal frame shift, allowing readthrough of the p27 ORF. Rectangles below the RNA species represent virus-encoded polypeptides. The shaded areas in the proteins encoded by RNA-l identify domains of significant amino acid sequence similarity to counterparts in the carmoviruses, necroviruses, machlomoviruses, and tombusviruses. The diamond-shaped marker in the shaded region of the 88-kDa protein represents conserved polymerase motifs. The rectangles identified by R, a, S, and P in the 37-kDa capsid protein encoded by the subgenomic RNA of RNA-l represent the random (RNA-binding), arm, shell, and protruding domains, respectively. The shaded region in the movement protein encoded by RNA-2 identifies a motif that is conserved in the dianthoviruses and bromoviruses. (After Giesman-Cookmeyer et a1., 1995.)

DIANTHOVIRUSES 267

thought to be involved in recognition of the viral RNA replicase and in the initiation of the synthesis of (- )-strand RNA. The existence of this conserved sequence at the 3' ends of both RCNMV RNA species, coupled with the fact that pseudorecombinants can be formed by mixing RNA-1 and RNA-2 from different dianthoviruses, suggest that there are conserved sequences in the 3' ends of both RNA species of all dianthoviruses.

Three large open reading frames (ORF) have been identified in RNA-1 of the definitive dianthoviruses: a 5' -terminal ORF encoding a polypeptide of 27 kDa; an internal one encoding a polypeptide of 54-57 kDa (57 kDa in Fig. 2); and a 3'-terminal one that encodes the 37- to 39-kDa capsid protein (37 kDa in Fig. 2). A heptanucleotide sequence located immediately before the 27-kDa termination signal in RCNMV RNA-1 is followed by a 95-nucleotide intergenic region between the 27-kDa and 57-kDa ORF; this intergenic region contains a stable stem-loop structure (Xiong et a1., 1993b). These two stretches of sequence, first recognized in the gag-pol region of eukaryotic retroviruses (Varmus, 1988), are both essential to a ribosomal frame shifting event involved in the synthesis of the 88-kDa viral RNA polymerase (Kim and Lommel, 1994). Similar structures have been identified in RNA-1 of CRSV and SCNMV (Ryabov et a1., 1994) and others are thought to be similarly involved in the synthesis of the RNA polymerases of barley yellow dwarf (Veidt et a1., 1988) and potato leafrolliuteoviruses (Prufer et a1., 1992) and other plus-strand RNA viruses (Atkins et a1., 1990).

CRSV RNA-1 has a putative ORF, not detected in RNA-1 of the other dianthoviruses, downstream from the ORF for the capsid protein, and this ORF could encode a polypeptide of 10 kDa (Ryabov et a1., 1994). However, because it is followed by a noncoding region of only nine nucleotides, this ORF may not be expressed. Another possibility is that the amber termination codon for the 38-kDa capsid protein ORF is suppressed during translation, in which case a 51-kDa read-through protein composed of the capsid protein and the putative lO-kDa polypeptide would be synthesized.

Dianthovirus RNA-2 molecules appear to be monocistronic, encoding polypeptides of 33.8-36.5 kDa (35 kDa in Fig. 2).

2. Functions of Gene Products

Functions have been assigned to the major proteins encoded by all the dianthovirus ORFs with the exception of the putative lO-kDa polypeptide of CRSV. Results from pseudorecombination studies among the definitive dianthoviruses, coupled with the observation that the RCNMV RNA-1 molecule replicates in plant protoplasts and induces the production of typical virus particles (Osman and Buck, 1987), indicate that the genes for RNA polymerase and the capsid protein are located on RNA-1 (Dodds et a1., 1977; Okuno et a1., 1983). Similar conclusions were reached following in vitro translation of the genomic RNA species of RCNMV (Morris-Krsinich et a1., 1983; Lommel, 1983) and CRSV (J. H. Tremaine, unpublished results). Both


RNA species are required for systemic infection of plants (Gould et a1., 1981; Okuno et a1., 1983; Osman et a1., 1986; Paje-Manalo and Lommel, 1989), suggesting that the 35-kDa protein encoded by RNA-2 is the transport or movement protein involved in cell-to-cell movement and systemic infection of plants by dianthoviruses. Neither of the RCNMV genomic segments replicated independently in inoculated leaves of Nicotiana c1evelandii (PajeManalo and Lommel, 1989) nor did those of SCNMV in cowpea (Vigna unguiculata) (Hiruki et a1., 1992). These results do not constitute proof that RNA-l did not replicate in inoculated leaf cells because the low efficiency of mechanical inoculation would make it difficult to detect viral protein or RNA in the few cells infected. Moreover, RNA-l would probably not be able to move from inoculated cells to adjoining cells in the absence of the movement protein provided by RNA-2.

Experiments in which infective RNA-l transcripts of cloned DNA from capsid protein mutants, which lacked portions of the capsid protein gene at the 3' terminus, were co-inoculated with wild-type RNA-2 showed that the capsid protein gene is not required for cell-to-cell spread within inoculated leaves of N. benthamiana at 15°C or 25 dc. Systemic spread did occur, however, at the lower but not the higher temperature (Xiong et a1., 1993a). There were no differences between wild type and mutants in the type of symptom or its time of appearance in inoculated leaves. Although these mutant capsid proteins were synthesized and stable in vitro, no antigens corresponding to their predicted sizes were detected by Western blotting of extracts from inoculated or noninoculated leaves infected by these mutants, indicating that the mutant proteins were probably unstable and did not accumulate in infected plants. Neither N. benthamiana nor N. c1evelandii were systemically infected at 25°C, although occasional"oak-leaf" symptoms were observed at 4 weeks postinoculation in leaves of N. benthamiana immediately above the inoculated leaves, probably as a result of slow cell-tocell spread through the stem. These results are interesting because N. benthamiana is a host for a number of viruses that do not infect other Nicotiana species. In these instances, N. benthamiana may be unusually permissive to rapid intercellular movement of such viruses. The results also clearly indicate that movement of RCNMV, and presumably that of other dianthoviruses, is dependent on the interaction of viral and host genetic factors and environmental conditions.

Polypeptide products of in vitro translation of RCNMV RNA-l in a rabbit reticulocyte translation system corresponded to the predicted molecular mass of each ORF (Xiong and Lommel, 1989) with the exception of a 90-kDa polypeptide that was produced in very low amounts and did not correspond to any ORE This was considered to be probably a read-through product of the ORFs for the 27-kDa and 57-kDa polypeptides. Two polyclonal antisera to synthetic oligopeptides, one representing the C-terminal portion of the 27-kDa polypeptide and the other representing the C-terminal region of the 57-kDa polypeptide, and therefore the C-terminal region of the putative

DIANTHOVIRUSES 269

88-kDa read-through protein, precipitated not only their homologous polypeptides but also the 88-kDa polypeptide (Xiong et a1., 1993b), providing evidence that it is indeed a read-through protein. Although the 57-kDa polypeptide encoded in RNA-1 of RCNMV is produced in vitro, it could not be detected in vivo using an antibody to the C terminus of the 57/88-kDa polypeptide or antibody to a fusion protein including the 57-kDa protein (Kim and Lommel, 1994). Infective RNA transcripts of mutants that lack the ORF initiation codon for the 57-kDa protein or a downstream, in-frame methionine codon were unable to synthesize the protein in vitro, but they were indistinguishable from wild type in their ability to infect plants systemically (Kim and Lommel, 1994), suggesting that the 57-kDa polypeptide is an artifact of in vitro translation or, if produced in vivo, is not needed for systemic spread. The 88-kDa polypeptide contains conserved RNA-dependent RNA polymerase motifs, suggesting that it is the viral RNA polymerase (Koonin, 1991). The heptanucleotide sequence that facilitates ribosomal frameshifting was identified in RNA-1 immediately behind the amber termination codon for the 27-kDa ORF, and point and frameshift mutations within this sequence confirmed its involvement in in vitro synthesis of the 88-kDa polypeptide (Xiong et a1., 1993b). It is likely that the putative viral RNA polymerase is expressed in vivo by a -1 ribosomal frameshifting mechanism similar to that of coronaviruses (Brierley et a1., 1989) and retroviruses (Varmus, 1988). The 37-kDa capsid protein is expressed via a 3'-coterminal subgenomic RNA of 1.5 kb.

The RNA-2 species of dianthoviruses (Kendall and Lommel, 1992; Ge et a1., 1992; Lommel et a1., 1988; Osman et a1., 1986) is required for systemic infection of plants, suggesting that its 35-kDa protein product is the movement protein (MP) for dianthoviruses. Deletion mutants, lacking no more than 39 amino acids from the C-terminal end of the MP, moved as rapidly as wild type and induced typical symptoms on inoculated and systemically infected leaves in N. benthamiana, but mutants with larger deletions failed to move from cell to cell or systemically in this host (Xiong et a1., 1993a). The failure of a spontaneous mutant of RCNMV-TpM34, which induced necrotic rather than chlorotic local lesions in cowpea, to infect this host systemically was attributed to the deletion of one of a sequence of four A residues between nucleotides 790 and 793 of RNA-2 (Osman et a1., 1991b). Infective RNA transcripts of cloned DNA of the mutant RNA-2 that contained an inserted A residue in the sequence restored the ability of the combination of mutant RNA-2 and wild-type or mutant RNA-l to infect cowpea systemically. These results clearly prove that the 35-kDa protein controls cell-to-cell spread of dianthoviruses. The dianthovirus MP appears to be analogous in function to that of the tobamoviruses which accumulates in the plasmodesmata and modifies them to allow passage of molecules larger than the normal solutes (Lucas and Wolf, 1993). Tobamovirus MP molecules also bind ssRNA in vitro, and presumably in vivo, and it is thought that these MP-viral RNA complexes are the form in which some viruses move intercellularly in plants


(Hull, 1989; Citovsky et a1., 1992). RCNMV MP, which accumulates in cell wall fractions (Osman and Buck, 1991b) and perhaps in plasmodesmata, also binds ssRNA transcripts corresponding to the 5 '-terminal nucleotides of the RCNMV MP gene (Osman et a1., 1992; Giesman-Cookmeyer and Lommel, 1993), and MP mutants derived by alanine-scanning mutagenesis differed in the symptoms they induced and in their ability to spread intercellularly (Giesman-Cookmeyer and Lommel, 1993). Studies of RCNMV MP mutants enabled three MP domains to be identified. Some mutant MPs differed in their capacity to bind ssRNA; others differed in the degree of cooperativity of such binding; and a third class did not depend on these properties. In further studies (Fujiwara et a1., 1993), fluorescein-labeled dextran, co-injected into leaf mesophyll cells of cowpea with either wild-type RCNMV MP or movement-effective mutant MP, moved rapidly to adjoining cells but failed to move when injected alone or with movement-defective mutant MP. Similar results were obtained in experiments with labeled wild-type and movementeffective mutant MP or movement-defective mutant MP. These results strongly suggest that the RCNMV MP and viral RNA-MP complexes move via plasmodesmata.

D. Sequence Relationships

1. Among Dianthoviruses

The dianthoviruses have a very high degree of nucleotide and amino acid sequence similarity. The overall homology between the RNA-1 species of RCNMV-Aus and SCNMV is 80%, but when comparison is confined to their respective RNA polymerases, there is over 90% homology at the amino acid level, clearly suggesting the presence of highly conserved sequences in this polypeptide. Similar analyses indicate that the preframeshift 27-kDa protein of CRSV shares 71 and 63% amino acid sequence identity with the comparable proteins of RCNMV and SCNMV, respectively, and the CRSV 57-kDa polypeptide shares 79 and 76% sequence identity with the comparable proteins of RCNMV and SCNMV, respectively (Ryabov et a1., 1994). The least degree of homology was among the capsid proteins: about 74% between the aligned sequences of RCNMV and SCNMV (Ge et a1., 1993) but only 51-54 % between those of CRSV and either RCNMV or SCNMV (Ryabov et a1., 1994). Assuming that the three-dimensional structure of the CRSV capsid is similar to that of turnip crinkle tombusvirus, as determined by X-ray crystallography (Dolja and Koonin, 1991; Carrington et a1., 1987), the protruding domain of the capsid protein is much less conserved (34%) than the shell domain (58%).

Comparison of the nucleotides of the RNA-2 species of RCNMV and SCNMV indicates that they too are similar in organization. The RNA-2 species of RCNMV-Aus (Lommel et a1., 1988) and RCNMV-TpM34 (Osman

DIANTHOVIRUSES 271

et a1., 1991a) from Czechoslovakia (each of 1448 nucleotides) have considerable nucleotide sequence similarity with the two SCNMV strains (84.2 and 94.8%, respectively, with SCNMV-38, and 82.7 and 94.3%, respectively, with SCNMV-59) (Ge et a1., 1992). Comparison of the MPs encoded by these RNA-2 species indicates that the two SCNMV strains are more closely related to RCNMV-TpM34 than to RCNMV-Aus. The 34-kDa MP encoded by CRSV RNA-2 has 59.6 and 55.7% amino acid sequence identity with the corresponding proteins of RCNMV-Aus and RCNMV-TpM34, respectively. The N-terminal 230 amino acids are even more highly conserved, with sequence identities of 64.3 and 62.6%, respectively (Kendall and Lommel, 1992). These comparisons suggest that the organization of the RNA-2 species and the functional domains of the encoded protein may be similar among the dianthoviruses.

2. Between Dianthoviruses and Other Viruses

The Dianthovirus genus shares features with other genera of plant viruses. The 88-kDa RNA polymerase encoded in RNA-l has a high degree of sequence similarity to those of the Tombusvirus, Carmovirus, Luteovirus, Machlomovirus, and Necrovirus genera (Koonin and Dolja, 1993). Moreover, the amino acid sequences of the shell (S) and protruding (P) domains of the capsid protein, which are also encoded in RNA-I, are highly and moderately conserved, respectively, with those of the Tombusvirus and Carmovirus genera (Rochon et a1., 1991). There is a small amount of amino acid sequence identity between the MP encoded in RCNMV-Aus and the 3a protein (MP) of the bromoviruses, brome mosaic (23.6% over a 54-amino acid sequence) (Lommel et a1., 1988) and cowpea chlorotic mottle (19.7% over a 157-amino acid sequence) (Allison et a1., 1989). A similar low amount of sequence identity with the 3a proteins of these bromoviruses was found in the corresponding protein of RCNMV-TpM34 (Osman et a1., 1991a). However, similar degrees of relatedness were found in many unrelated proteins (Osman et a1., 1991a), and it remains to be determined if there is any functional significance to the presence of these related sequences in the movement proteins of dianthoviruses and bromoviruses.

V. CYTOPATHOLOGY

There is a paucity of information on the cytopathology of dianthovirus infections. In CRSV-infected sweet william and cowpea, large crystalline arrays of virus particles occurred in the cytoplasm, and the nuclei contained large aggregates of virus particles as well as tubular inclusions, often with virus particles on their surfaces (Weintraub et a1., 1975). Spherical inclusion bodies were also seen in nuclei of sweet william but not in cowpea. Inflated, electron-transparent mitochondria, clusters of proliferated endoplasmic re-


ticulum with dilated cisternae, and scattered CRSV-like particles were common in the cytoplasm of CRSV-infected leaf mesophyll cells in Chenopodium quinoa and Nicotiana megalosiphon (Koenig et a1., 1988); necrotic cells contained massive aggregates of these particles. Viruslike particles were also observed in nuclei, but the tubular inclusions observed in this organelle by Weintraub et a1. (1975) were not detected. RCNMV was present in leaves, stems, and roots of infected red clover (Hollings and Stone, 1977); membranous vesicles and patches of densely stained amorphous material were common in the cytoplasm (Francki et a1., 1985) and chloroplastic vesiculation was observed in red clover infected with RCNMV-Can (Ragetli and Elder, 1977). Amoeboid intracellular inclusions were seen in about 5 % of the cells in an experimental host (N. c1evelandii) but apparently not in red or white clover. Large amorphous inclusions seen by light microscopy in the cytoplasm of Red Kidney bean (Phaseolus vulgaris) infected with SCNMV were subsequently shown to be composed of large aggregates of virus particles when examined by electron microscopy (Hiruki et a1., 1984b). In fique infected with FNSV, viruslike particles were readily detected in root and leaf cells, often associated with large, electron-dense tubular inclusions (Morales et a1., 1992) and "aster" (flower)-like inclusions (Dabek and Castano, 1978).

VI. ECOLOGY AND CONTROL

A. Ecology

Little is known about the survival and spread of dianthoviruses in nature. The viruses are not transmitted by aerial vectors so far tested, although there is an unconfirmed report that the aphid Myzus persicae, but not the scale insect Saissetia caffea or the mealy bug Planococcus citri, can transmit FNSV, albeit inefficiently (Dabek and Castano, 1978). Given the highly contagious nature of the dianthoviruses, such transmission may be the result of contamination rather than of true vector transmission. None of the dianthoviruses is known to be transmitted by seed. Despite the fact that pollen from SCNMV-infected sweet clover is heavily contaminated with surfaceborne virus and is constantly present on the western flower thrips (Frankliniella occidentalis), which inhabit florets of sweet clover, no transmission of SCNMV was observed in several controlled experiments (Hiruki et a1., 1989). Rather, contact transmission as well as a soil-borne mode of transmission, without the aid of vectors, are clearly indicated by experimental evidence. Moreover, the natural host range of CRSV includes species in which the virus can be spread by vegetative propagation.

Of the three routes of spread, that in the soil is of current interest because of the controversial evidence for the role of soil-inhabiting organisms as vectors. Susceptible plants, transplanted into soil that previously contained infected plants or into sterilized soil to which infected plant

DIANTHOVIRUSES 273

tissues or virus suspensions had been added, were readily infected (Bowen and Plumb, 1979 j Brown and Trudgill, 1984 j Gerhardson and Insunza, 1979 j

Hollings and Stone, 1977 j Kegler and Kegler, 1981 j Lyness and Teakle, 1981). Particles of CRSV, RCNMV, and SCNMV are released from root cells into soil water (Hollings and Stone, 1977 j Kegler and Kegler, 1981 j Hiruki, 1986a) where they may retain infectivity for months. Presumably plants in the natural state can be inoculated by infective virions in the soil solution. Claims that the dianthoviruses are transmitted by longidorid nematodes (Fritzsche, 1968 j Fritzsche and Schmelzer, 1967 j Fritzsche et a1., 1979 j Kleinhempel et a1., 1980) or by species of the Chytridiomycete, Olpidium (Lange, 1977 j MacFarlane, 1981) have not been confirmed (Brown and Trudgill, 1984 j

Lyness and Teakle, 1981). The two last-named groups of investigators demonstrated soil transmission of CRSV or RCNMV to test plants in soil known not to contain the putative vectors. Transmission of CRSV to test plants in the presence of nematodes (Longidorus elongatus) was no greater than in their absence (Brown and Trudgill, 1984), but transmission of RCNMV to test plants was enhanced 2- to 20-fold in the presence of o. brassicae (Gerhardson and Insunza, 1979), suggesting perhaps that the fungus may playa role in initiating RCNMV infection. However, in controlled inoculations, the tobacco isolate of o. brassicae was not found in the roots of SCNMV-infected sweet clover seedlings nor was there a significant difference in the percentage of sweet clover seedlings infected by inocula that contained zoospores of o. brassicae and those that did not (Hiruki, 1986a). MacFarlane (1982) reported that zoospores of o. bornovanus (Sahtiyanci) Karling [= o. radicale Schwartz & Cook fide Lange & Insunzaj = o. cucurbitacearum Barr & Dias (Campbell and Sim, 1994)], probably more widespread in Britain than previously recognized, transmitted RCNMV to clover, leading to the tentative conclusion that the fungus is a vector. Clearly, further work is needed to determine if this or any other fungus is a vector of the virus or a factor in predisposing host plants to root infection.

An important consideration in all of the studies in which putative vectors are collected from the roots of infected plants is the possibility of contamination with infective virus particles present in soil water or on the root surface. The presence of infective dianthovirus particles in soil water and the isolation of CRSV from a pond surrounded only by grassland, from a stream that had no direct contact with agricultural fields (Yi et a1., 1992), and from a canal near a sewage plant (Koenig et a1., 1988) parallel reports of the isolation from river waters of other plant viruses such as an unidentified sugar beet virus (Tomlinson et a1., 1983a), tobamoviruses (Koenig, 1986, 1988 j

Tosic and Tosic, 1984), necroviruses (Tomlinson et a1., 1983b), potexviruses and carnation mottle virus (Koenig and Lesemann, 1985), and especially tombusviruses (Tomlinson and Faithfull, 1984 j Koenig and Lesemann, 1985 j

Koenig et a1., 1989). Thus, both for dianthoviruses and for tombusviruses, transport by water affords a distribution system that effectively contributes to their spread. Isolation of CRSV from a sewage canal does not necessarily


mean that the virus was originally present in sewage, but it would be interesting to determine if dianthoviruses remain infective after passage through the alimentary tracts of livestock (and migratory animals), as has been reported for TBSV in human feces (Tomlinson et a1., 1982) and for several plant viruses including TBSV in the feces of rabbits and/or mice (Kegler et a1., 1984). If so, an additional means of spread (e.g., in manure for fertilizer) is possible.

B. Control

Aside from the usual recommendation that proper sanitation be employed in plant propagation, the choice of additional control measures is dictated by the affected species. In vegetatively propagated plants such as carnation and the tree fruits, eradication of viruses in plant shoots and meristems by thermotherapy (Brierley, 1964) or meristem-tip culture (Stone, 1968j Kowalska, 1974) can be used to produce virus-free nuclear stocks as the basis for commercial propagation. Propagation from such stocks, coupled with adequate monitoring by suitable methods such as ELISA (Lommel et a1., 1983), should ensure continued production of virus-free plants. In California (Lommel et a1., 1983), and presumably in the rest of North America, CRSV infection of commercial carnation plantings is no longer of significance' but the virus has been reported occasionally in carnation in other parts of the world (Bennet and Milne, 1976j Bremer and Lahdenpera, 1981 j

Valenzuela and Pizano, 1992). Carnation and sweet william are probably the only herbaceous hosts of concern with respect to CRSV infection, but the role of CRSV in the etiology of disease in pome and stone fruits requires further study.

Control of RCNMV, especially because of its broader natural host range, would appear to be more difficult than that of CRSV. Because RCNMV is not seed-transmitted, seedlings of clovers and alfalfa presumably become infected by contact transmission or by soil-borne virus. Unfortunately, most of the red clover cultivars recommended for production in Britain are susceptible to natural infection by RCNMV, some (Hungoropoly and Teroba) showing about 45% natural infection (Bowen and Plumb, 1979). Similar results were obtained following mechanical inoculation of red and white clover cultivars commonly grown in Czechoslovakia (Musil et a1., 1979). The stability of RCNMV infectivity in soil would appear to preclude soil treatment as a method of eradicating the virus. However, planting a mixture of Hungoropoly clover (5 kg/hectare) and Italian ryegrass cv RvP (13 kg/hectare) resulted in 0.8% RCNMV infection compared to 8.9% for a pure stand of Hungoropoly (Lewis et a1., 1985), suggesting that a mixture of susceptible and immune crop species may significantly reduce the incidence of infection. However, genetic resistance or tolerance to RCNMV infection should be investigated for long-term disease control.

DIANTHOVIRUSES 275

VII. CONCLUDING REMARKS

Although the experimental host ranges of the dianthoviruses are very wide, the natural host ranges are limited to a few species. Application of the rapid and sensitive assay and identification procedures currently available for these viruses may result in their detection in other crops and weeds. The absence of symptoms in RCNMV-infected plants at higher temperatures may explain why more such reports have not been made to date.

Pseudorecombinants between definitive dianthoviruses have been produced experimentally, suggesting that genetic reassortment between these viruses could possibly occur under field conditions if a common host were present. There is no evidence for this, possibly because two dianthoviruses have not been found in the same host in the same place. In Canada, RCNMV was isolated from a single red clover plant near Vancouver (Ragetli and Elder, 1977), and in a systematic search in Alberta, C. Hiruki (unpublished results) failed to detect red clover, white clover, or alsike clover infected with RCNMV. Interestingly, however, the natural hosts of RCNMV include alfalfa and white clover in Australia (Gould et al., 1981; Lyness and Teakle, 1981) and sweet clover in Europe (Musil, 1969a). Perhaps a host plant such as alfalfa, which is naturally susceptible to RCNMV and SCNMV in single infections, would be susceptible to both viruses in mixed infections. If natural infection of the roots of field plants is via water-borne virus in soil, the opportunities for mixed infections with both RCNMV and SCNMV might be significant.

Further research is required to characterize the early events of virus infection as it occurs in nature, i.e., in roots apparently in the absence of a vector. Are there intrinsic properties of the dianthovirus virion that result in infection of root cells under conditions in which other plant viruses do not infect? The soil-borne nature and similar infection pattern of the dianthoviruses and tombusviruses, coupled with similarities in the primary sequences of the amino acids in their capsid proteins, suggest that common features may exist in the early stages of the interaction between these viruses and plant roots. Although the evidence for dianthovirus transmission by soilinhabiting fungi is equivocal, cucumber necrosis tombusvirus (CuNY) is naturally transmitted by zoospores of O. bornovanus, and a chimeric virus, containing the genome of TBSV modified by substitution of the capsid gene of TBSV by that of CuNY, was transmitted by O. bornovanus to Nicotiana clevelandii (McLean et al., 1994). The reciprocal chimeric virus was not transmitted by the fungus. CuNY probably diverged from an ancestral virus in its evolution and acquired genes for specific receptor sites, probably located in the protruding domain, which allowed specific recognition of the virus by O. bornovanus and its subsequent transmission.

A second aspect of the basic interaction of dianthoviruses and plant root systems is the route of distribution of viruses from the initially infected root cells. Does the dianthovirus MP play any role in this process? Is the ORF for


the MP translated in the roots of "nonhosts" (Paje-Manalo and Lommel, 1989), the leaf protoplasts of which support replication of RCNMV RNA-I? If the MP is expressed in "nonhost" plants, does it interact with the plasmodesmata as it does in susceptible hosts? Evidence is accumulating that indicates that viral MPs play a significant role in the host range of plant viruses (Mise et a1., 1993) and the availability of infective RNA transcripts for RCNMV (Xiong and Lommel, 1991) and SCNMV (Ge and Hiruki, 1993) provides a unique opportunity for research on genetic manipulation of MP genes and their subsequent interaction in different plant genotypes.

Utilization of plants transformed with dianthovirus genes should allow continued progress in the research on basic aspects of the interaction between viral and host genes that governs the responses of plants to infection by dianthoviruses. The physicochemical and biological properties of the dianthoviruses make them eminently suitable tools for future research on the basic aspects of disease induction by plant viruses.

REFERENCES

Allison, R. F., Janda, M., and Ahlquist, P., 1989, Sequence of cowpea chlorotic mottle virus RNAs 2 and 3 and evidence of a recombination event during bromovirus evolution, Virology 172:321.

Altschuh, D., and Van Regenmortel, M. H. v., 1983, Refutation of the alleged antigenic relationship between an isometric and a rod shaped plant virus, Int. 1. Microbio1. 1:13.

Atkins, J. F., Weiss, R. B., and Gesteland, R. F., 1990, Ribosome gymnastics-degree of difficulty 9.5, style 10.0, Cell 62:413.

Barton, R. J., 1981, Analysis of virus capsid polypeptides, Rep. Glasshouse Crops Res. Inst. 1980:151.

Bennet, P. R., and Milne, K. S., 1976, Carnation viruses in New Zealand, Acta. Hortic. 59:61. Bercks, R., and Querfurth, G., 1972, Serologische Beziehungen von cocksfoot mild mosaic virus

und carnation ringspot virus zu Viren der turnip yellow mosaic virus-Gruppe, Phytopatho1. Z. 75:215.

Bowen, R., and Plumb, R. T., 1979, The occurrence and effects of red clover necrotic mosaic virus in red clover (Trifolium pratenseJ, Ann. App1. Biol. 91:227.

Bremer, K., and Lahdenpera, M.-1., 1981, Virus diseases in carnation and chrysanthemum cuttings imported into Finland, Ann. Agriculturae Fenniae 20:214.

Brierley, 1., Digard, P., and Inglis, S. c., 1989, Characterization of an efficient coronavirus ribosomal frame shifting signal: requirement for an RNA pseudoknot, Cell 57:548.

Brierley, P., 1964, Heat cure of carnation viruses, Plant Dis. Rep. 48:143. Brown, D. J. F., and Trudgill, D. 1., 1984, The spread of carnation ringspot virus in soil with or

without nematodes, Nematologica 30:102. Campbell, R. N., and Sim, S. T., 1994, Host specificity and nomenclature of Olpidium borno

vanus (~ Olpidium radicaleJ and comparisons to O. brassicae, Can. 1. Bot. 72:1136. Carrington, J. c., Morris, T. J., Stockley, P. G., and Harrison, S. c., 1987, Structure and assembly

of turnip crinkle virus. IV. Analysis of the coat protein gene and implications of the subunit primary structure, 1. Mo1. Bio1. 194:265.

Casper, R., 1976, Natural occurrence of cucumber mosaic virus with plum pox virus (SharkaJ and prunus necrotic ringspot virus in plum, Mitt. Bio1. Bundesanst. Land-u.Forstwirtsch. 170:33.

DIANTHOVIRUSES 277

Citovsky, V, Wong, M.-1., Shaw, A. 1., Venkataram Prasad, B. V, and Zambryski, P., 1992, Visualization and characterization of tobacco mosaic virus movement protein binding to single-stranded nucleic acids, Plant Cell 4:397.

Dabek, A. J., and Castano, J. J., 1978, The occurrence, symptomatology, transmission and virus aetiology of macana disease of fique (Furcraea spp.1 in Colombia, South America, Phytopathol. Z. 92:57.

Dodds, J. A., Tremaine, J. H., and Ronald, W. P., 1977, Some properties of carnation ringspot virus single- and double- stranded ribonucleic acid, Virology 83:322.

Dodds, J. A., Morris, T. J., and Jordan, R. 1., 1984, Plant viral double-stranded RNA, Annu. Rev. Phytopathol. 22:15l.

Dolja, V V, and Koonin, D. V, 1991, Phylogeny of capsid proteins of small icosahedral RNA plant viruses, J. Gen. Viral. 72:148l.

Francki, R.I. B., Milne, R. G., and Hatta, T., 1985, Atlas of Plant Viruses, Vol. II, CRC Press, Boca Raton, F1.

Francki, R. I. B., Fauquet, C. M., Knudson, D. 1., and Brown, F. (eds.I, 1991, Classification and Nomenclature of Viruses: Fifth Report of the International Committee on Taxonomy of Viruses, Springer-Verlag, Vienna (also in Arch. Viral. Suppl. 21.

Fritzsche, R., 1968, bkologie und Vektoreignung von Longidorus macrosoma Hooper, BioI. Zentralbl. 87:139.

Fritzsche R., and Schmelzer, K., 1967, Dbertragbarkeit des Nelkenringflecken Virus durch Nematoden, Naturwissenschaften 54:498.

Fritzsche, R., Kegler, H., Thiele, S., and Gruber, G., 1979, Beitrag zur Epidemiologie und Dbertragung des Nelkenringflecken-Virus in Obstbestanden, Arch. Phytopathol. Pflanzenschutz 15:177.

Fujiwara, T., Giesman-Cookmeyer, D., Ding, B., Lommel, S. A., and Lucas, W. J., 1993, Cell-tocell trafficking of macromolecules through plasmodesmata potentiated by the red clover necrotic mosaic virus movement protein, Plant Cell 5:1783.

Gallo, J., and Musil, M., 1984a, Serotypes of red clover necrotic mosaic virus. III. Immunoelectrophoresis under different conditions, Acta Viral. 28:78.

Gallo, J., and Musil, M., 1984b, Isoelectric points of red clover necrotic mosaic virus serotypes, Acta Viral. 28:82.

Ge, Z., and Hiruki, c., 1993, The infectious transcripts of sweet clover necrotic mosaic virus bipartite genome constructed by the polymerase chain reaction, Prac. Japan Acad. Ser. B 69:113.

Ge, Z., Hiruki, C., and Roy, K. 1., 1992, A comparative study of the RNA-2 nucleotide sequences of two sweet clover necrotic mosaic virus strains, J. Gen. Virol. 73:2483.

Ge, Z., Hiruki, C., and Roy, K. 1., 1993, Nucleotide sequence of sweet clover necrotic mosaic dianthovirus RNA-I, Virus Res. 28:113.

Gerhardson, B., and Insunza, V, 1979, Soil transmission of red clover necrotic mosaic virus, Phytopathol. Z. 94:67.

Gerhardson, B., and Lindsten, K., 1973, Red clover mottle virus and red clover necrotic mosaic virus in Sweden, Phytopathol. Z. 76:67.

Giesman-Cookmeyer, D., and Lommel, S. A., 1993, Alanine scanning mutagenesis of a plant virus movement protein identifies three functional domains, Plant Cell 5:973.

Giesman-Cookmeyer, D., Kim, K.-H., and Lommel, S. A., 1995, Dianthoviruses, in: Pathogenesis and Host-Specificity in Plant Diseases: Histopathological, Biochemical, Genetic and Molecular Basis. Vol. 3: Viruses and Viroids (R. P. Singh, U. S. Singh, and K. Kohmoto, eds.1, Pergamon Press, Oxford pp. 157-176.

Gould, A. R., Francki, R. I. B., Hatta, T., and Hollings, M., 1981, The bipartite genome of red clover necrotic mosaic virus, Viralogy 108:499.

Hatta, T., and Francki, R. I. B., 1984, Differences in the morphology of isometric particles of some plant viruses stained with uranyl acetate as an aid to their identification, J. Viral. Methods 9:237.

278 R.I. HAMILTON AND J. H. TREMAINE

Hiruki, C., 1986a, Incidence and geographic distribution of sweet clover necrotic mosaic virus in Alberta, Plant Dis. 70:1129.

Hiruki, c., 1986b, Sweet clover necrotic mosaic virus, AAB Descriptions of Plant Viruses No. 321. Hiruki, C., 1987, The dianthoviruses: A distinct group of isometric plant viruses with bipartite

genome, Adv. Virus Res. 33:257. Hiruki, c., and Figueiredo, G., 1985, Monoclonal antibodies and their use in the study of

dianthoviruses, Acta Hortic. 164:217. Hiruki, c., Rao, A. 1. N., Furuya, Y, and Figueiredo, G., 1984a, Serological studies of di

anthoviruses using monoclonal and polyclonal antibodies, f. Gen. Virol. 65:2273. Hiruki, c., Rao, D. v., Chen, M. H., Okuno, T., and Figueiredo, G., 1984b, Characterization of

sweet clover necrotic mosaic virus, Phytopathology 74:482. Hiruki, c., Kudo, K., and Figueiredo, G., 1989, Transmission of sweet clover necrotic mosaic

virus, Proc. Tapan Acad. Ser. B 65:234. Hiruki, C., Kakuta, H., Ge, Z., Figueiredo, G., and Mizutani, J., 1992, Viral genome delivery into

detached and intact leaf tissues of Vigna unguiculata by RNA-coated gold particles using the improved particle gun, Proc. Tapan Acad. Ser B 68:183.

Hollings, M., and Stone, O. M., 1965, Investigations of carnation viruses. II. Carnation ringspot, Ann. Appl. Biol. 56:73.

Hollings, M., and Stone, O. M., 1970, Carnation ringspot virus, CMI/AAB Descriptions of Plant Viruses No. 21.

Hollings, M., and Stone, O. M., 1977, Red clover necrotic mosaic virus, CMI/AAB Descriptions of Plant Viruses No. 181.

Hull, R., 1989, The movement of viruses in plants, Annu. Rev. Phytopathol. 27:213. Inouye, N., and Hiruki, c., 1985, A new strain of sweet clover necrotic mosaic virus isolated

from alfalfa, Ann. Phytopathol. Soc. Tapan 51:82. Kalmakoff, J., and Tremaine, J. H., 1967, Some physical and chemical properties of carnation

ringspot virus, Virology 33:10. Kaper, J. M., 1975, The Chemical Basis of Virus Structure, Dissociation and Reassembly, North

Holland, Amsterdam. Kassanis, B., 1955, Some properties of four viruses isolated from carnation plants, Ann. Appl.

Biol. 43:103. Kassanis, B., and Woods, R. D., 1968, Aggregated forms of the satellite of tobacco necrosis virus,

f. Gen. Virol. 2:395. Kassanis, B., White, R. E, and Woods, R. E, 1973, Genetic complementation between middle and

bottom components of two strains of radish mosaic virus, f. Gen. Virol. 20:277. Kegler G., and Kegler, H., 1981, Beitriige zur Kenntnis der vektorlosen Dbertragung pflanzen

pathogener Viren, Arch. Phytopathol. Pflanzenschutz 17:307. Kegler, H., Kleinhempel, H., and Stanarius, A., 1984, Evidence of infectious plant viruses after

passage through the rodents alimentary tract, Arch. Phytopathol. Pfanzenschutz 20:189. Kendall, T. 1., and Lommel, S. A., 1992, Nucleotide sequence of carnation ringspot dianthovirus

RNA-2, f. Gen. Virol. 73:2479. Kim, K. H., and Lommel, S. A., 1994, Identification and analysis of the site of -1 ribosomal

frameshifting in red clover necrotic mosaic virus, Virology 200:574. Kleinhempel, H., Gruber, G., and Kegler, H., 1980, Investigations on carnation ringspot virus in

fruit trees, Acta Phytopathol. Acad. Sci. Hung. 15:107. Koenig, R., 1986, Plant viruses in rivers and lakes, Adv. Virus Res. 31:321. Koenig, R., 1988, Detection in surface waters of plant viruses with known and unknown natural

hosts, in: Viruses with Fungal Vectors (T. I. Cooper and M. J. C. Asher, eds.), pp. 307-313, Association of Applied Biologists, Wellesbourne, UK.

Koenig R., and Lesemann, D.-E., 1985, Plant viruses in German rivers and lakes. 1. Tombusviruses, a potexvirus and carnation mottle virus, Phytopathol. Z. 112:105.

Koenig, R., An, D., Lesemann, D.-E., and Burgermeister, W., 1988, Isolation of carnation ringspot virus from a canal near a sewage plant: cDNA hybridization analysis, serology and cytopathology, f. Phytopathol. 121:346.

DIANTHOVIRUSES 279

Koenig, R., Rudel, M., and Lesemann, D.-E., 1989, Detection of petunia asteroid mosaic, carnation ringspot and tobacco necrosis viruses in ditches and drainage canals in a grapevinegrowing area of West Germany, f. Phytopathol. 127:169.

Koonin, E. V., 1991, The phylogeny of RNA-dependent RNA polymerases of positive-stranded RNA viruses, f. Gen. Virol. 72:2197.

Koonin, E. V., and Dolja, V. V., 1993, Evolution and taxonomy of positive-strand RNA viruses: Implications of comparative analysis of amino acid sequences, Grit. Rev. Biochem. Mol. Biol. 28:375.

Kowalska, A., 1974, Freeing carnation plants from viruses by meristem-tip culture, Phytopathol. z. 79:30l.

KUhne, T., and Eisbein, K., 1983, Untersuchungen zur Stabilitat des Nelkenringflecken-Virus, Acta Phytopathol. Acad. Sci. Hung. 18:1Ol.

KUhne, T., Proll, E., and Reichenbacker, D., 1983, Aggregationseigenschaften des Nelkenring-flecken-Virus in vitro, Acta Phytopathol. Acad. Sci. Hung. 18:85.

Lane, 1. c., 1974, The bromoviruses, Adv. Virus Res. 19:15l. Lange, 1., 1977, Interrelations between fungi and viruses, Plantesygdomme i Denmark 1976 92:68. Lewis, G. C., Heard, A. 1., Gutteridge, R. A., Plumb, R. G., and Gibson, R. W., 1985, The effects of

mixing Italian ryegrass (Lolium multiflorum) with perennial ryegrass (L. perenne) or red clover (Trifolium pratense) on the incidence of viruses, Ann. Appl. Biol. 106:483.

Lommel, S. A., 1983, Genome and Replicative Organization of Plant Dianthoviruses and Arkansas Bee Virus, Ph.D. Thesis, University of California, Berkeley.

Lommel, S. A., McCain, A. H., Mayhew, D. E., and Morris, T. J., 1983, Survey of commercial carnation cultivars for four viruses in California by indirect enzyme-linked immunosorbent assay, Plant Dis. 67:53.

Lommel, S. A., Weston-Fina, M., Xiong, Z., and Lomonossoff, G. P., 1988, The nucleotide sequence and gene organization of red clover necrotic mosaic virus RNA-2, Nucleic Acids Res. 16:8587.

Longworth, J. F., 1978, Small isometric viruses of invertebrates, Adv. Virus Res. 23:103. Lucas, W. J., and Wolf, S., 1993, Plasmodesmata: The intercellular organelles of green plants,

Trends Gell Biol. 3:308. Lyness, E. W., and Teakle, D. S., 1981, Red clover necrotic mosaic virus isolated from Trifolium

repens and Medicago sativa in Victoria, Australasian Plant Pathol. 10:6. MacFarlane, I., 1981, Putative vectors of RCNMY, Rep. Rothamsted Exp. Stn. 1980:188. MacFarlane, I., 1982, Virus transmission by fungi: Red clover necrotic mosaic virus (RCNMV),

Rep. Rothamsted Exp. Stn. 1981:190. McLean, M. A., Campbell, R. N., Hamilton R. I., and Rochon, D. M., 1994, Involvement of the

cucumber necrosis virus coat protein in the specificity of fungus transmission by Olpidium bornovanus, Virology 204:840.

Mise, K., Allison, R. F., Janda, M., and Ahlquist, P., 1993, Bromovirus movement protein genes playa crucial role in host specificity, f. Virol. 67:2815.

Mor, Y., 1983, Carnation trends in Israel, Acta HortiG. 141:253. Morales, F., Castaiio, M., Calvert, 1., and Arroyave, J. A., 1992, Furcraea necrotic streak virus: An

apparent new member of the dianthovirus group, f. Phytopathol. 134:247. Morris, T. J., 1983, Virus-specific double-stranded RNA: Functional role in RNA virus infection,

in: Plant Infectious Agents: Viruses, Viroids, Virusoids and Satellites (H. D. Robertson, S. H. Howell, M. Zaitlin, and M. Malmberg, eds.), pp. 80-83, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

Morris-Krsinich, B., Forster, R. 1., and Mossop, D. W., 1983, Translation of red clover necrotic mosaic virus RNA in rabbit reticulocyte lysate: identification of the virus coat protein cistron on the larger RNA strand of the bipartite genome, Virology 124:349.

Musil, M., 1969a, Red clover necrotic mosaic virus, a new virus infecting red clover (Trifolium pratense) in Czechoslovakia, Biologica (Bratislava) 24:33.

Musil, M., 1969b, Serological properties of certain isolates of red clover necrotic mosaic virus, Acta Virol. 13:226.


Musil, M., and Gallo, J., 1982, Serotypes of red clover necrotic mosaic virus. I. Characterization of three serotypes, Acta Viral. 26:497.

Musil, M., and Matisova, J., 1967, A contribution to the knowledge of virus mosaic of red clover in Slovakia, Ochrana Rostlin N.S. 3:225.

Musil, M., Leskova, 0., and Smrz, J., 1979, Susceptibility and reaction of some Leguminosae to viruses occurring in Czechoslovakia, Sbornik Vedeckych Praci. 5/6:205.

Musil, M., Leskova, 0., and Gallo, J., 1982, Serotypes of red clover necrotic mosaic virus. II. Typing of 34 isolates, Acta Viral. 26:502.

Noordam, D., Thung, T. H., and Van der Want, J. P. H., 1951, Investigations on carnation mosaic. I. Tidischr. PI. Zeikt. 57:72.

Okuno, T., Hiruki, c., Rao, D. V, and Figueiredo, G. C., 1983, Genetic determinants distributed in two genomic RNAs of sweet clover necrotic mosaic, red clover necrotic mosaic and clover primary leaf necrosis viruses, r. Cen. Viral. 64:1907.

Orlova, E. V, Tsuprun, V 1., Kiselev, N. A., and Spaar, D., 1980, Structure of carnation ringspot virus and its crystal monolayers, Electran Micrasc. 2:602.

Osman, T. A. M., and Buck, K. W., 1987, Replication of red clover necrotic mosaic virus RNA in cowpea protoplasts: RNA 1 replicates independently of RNA 2, r. Cell. Viral. 68:289.

Osman, T. A. M., and Buck, K. W., 1991a, Double-stranded RNAs isolated from plant tissue infected with red clover necrotic mosaic virus correspond to genomic and sub genomic single-stranded RNAs, r. Cell. Viral. 71:945.

Osman, T. A. M., and Buck, K. W., 1991b, Detection of the movement protein of red clover necrotic mosaic virus in a cell wall fraction from infected Nicotiana clevelandii plants, r. Cen. Viral. 72:2853.

Osman, T. A. M., Dodds, S. M., and Buck, K. W., 1986, RNA 2 of red clover necrotic mosaic virus determines lesion morphology and systemic invasion of cowpea, r. Cen. Viral. 67:203.

Osman, T. A. M., Miller, S. J., Marriott, A. c., and Buck, K. W., 1991a, Nucleotide sequence of RNA-2 of a Czechoslovakian isolate of red clover necrotic mosaic virus, r. Cell. Viral. 72:213.

Osman, T. A. M., Ingles, P. J., Miller, S. J., and Buck, K. W., 1991b, A spontaneous red clover necrotic mosaic virus mutant with a truncated movement protein, r. Cen. Viral. 72:1793.

Osman, T. A. M., Hayes, R. J., and Buck, K. W., 1992, Cooperative binding of the red clover necrotic mosaic virus movement protein to single-stranded nucleic acids, r. Cell. Viral. 73:223.

Paje-Manalo, 1. 1., and Lommel, S. A., 1989, Independent replication of red clover necrotic mosaic virus RNA-1 in electroporated host and nonhost Nicotiana species protoplasts, Phytopathology 79:457.

Pappu, H. R., and Hiruki, C., 1989, Electrophoretic variability among dianthoviruses, Phytopathology 79:1253.

Pappu, H. R., Hiruki, c., and Inouye, N., 1988, A new serotype of sweet clover necrotic mosaic virus, Phytopathology 78:1343.

Prufer, D., Tacke, E., Schmitz, J., Kull, B., Kaufmann, A., and Rodhe, W., 1992, Ribosomal frameshifting in plants: A novel signal directs the -1 frameshift in the synthesis of the putative viral replicase of potato leafroll virus, Embo r. 11:111l.

Ragetli, H. W. J., and Elder, M., 1977, Characteristics of clover primary leaf necrosis virus, a new spherical isolate from Trifolium pratense, Can. r. Bot. 55:2122.

Rao, A. 1. N., and Hiruki, C., 1985, Clover primary leaf necrosis virus: A strain of red clover necrotic mosaic virus, Plant Dis. 69:959.

Rao, A. 1. N., and Hiruki, c., 1987, Unilateral compatibility of genome segments from two distinct strains of red clover necrotic mosaic virus, r. Cen. Viral. 68:19l.

Rao, A. 1. N., Ohki, S. T., Lakshman, D. K., and Hiruki, C., 1987, Identification of a new serotype and antigenic relationships among six isolates of red clover necrotic mosaic virus: Identification of a new serotype, Phytopathology 77:995.

Richter, J., Kleinhempel, H., Gruber G., and Kegler, H., 1978, Identifizierung eines Virus von

DIANTHOVIRUSES 281

Birnen mit Steinfruchtigkeit als Nelkenringflecken-Virus (carnation ringspot virus), Arch. Phytopathol. Pflanzenschutz 14:411.

Rochon, D. M., Johnston, J. C., and Riviere, C. J., 1991, Molecular analysis of the cucumber necrosis virus genome, Can. J. Plant Pathol. 6:142.

Rudel, M., Querfurth, G., and Paul, H. 1., 1977, Naturliches Vorkommen des carnation ringspot virus in Stellaria media (1.) Cyrill. aus Weinbergen erstmals nachgewiessen, Nachrichtenblatt des Deutschen Pflanzenschutzdienstes, Stuttgart: 29:59.

Ryabov, E. V., Generozov, E. Y., Kendall, T. 1., Lommel, S. A., andZavriev, S. K.,1994, Nucleotide sequence of carnation ringspot dianthovirus RNA-I, r. Gen. Virol. 75:243.

Sparnaaij, 1. D., 1983, Carnations in Europe, Acta Hortic. 141:26l. Stone, O. M., 1968, The elimination of four viruses from carnation and sweet william by

meristem-tip culture, Ann. Appl. Biol. 62:119. Tomlinson, J. A., and Faithfull, E. M., 1984, Studies on the occurrence of tomato bushy stunt

virus in English rivers, Ann. Appl. Biol. 104:485. Tomlinson, J. A., Faithfull,. E. M., Flewett, T. H., and Beards, G., 1982, Isolation of infective

tomato bushy stunt virus after passage through the human alimentary tract, Nature 300:637.

Tomlinson, J. A., Faithfull, E. M., and Seeley, N. D., 1983a, plant viruses in rivers, Rep. Natl. Vegetable Res. Stn. 1982:86.

Tomlinson, J. A., Faithfull, E. M., Webb, M. J. w., Fraser, R. S. S., and Seeley, N. D., 1983b, Chenopodium necrosis: A distinctive strain of tobacco necrosis virus isolated from river water, Ann. Appl. Biol. 102:135.

Tosic, M., and Tosic, D., 1984, Occurrence of tobacco mosaic virus in water of the Danube and Sava rivers, Phytopathol. Z. 110:200.

Tremaine, J. H., 1961, Removal of host antigens from plant virus preparations by ion exchange chromatography, Can. r. Bot. 39:1705.

Tremaine, J. H., and Dodds, J. A., 1985, Carnation ringspot virus, AAB Descriptions of Plant Viruses No. 308.

Tremaine, J. H., and Ronald, W. P., 1976, Differential effects of sodium dodecyl sulphate on strains of carnation ringspot virus, J. Gen. Virol. 30:299.

Tremaine, J. H., and Ronald, W. P., 1985, The effect of pH and some selected chemicals on the temperature-reversible aggregation of carnation ringspot virus, Phytopathology 75:467.

Tremaine, J. H., Ronald, W. P., and Valcic, A., 1976, Aggregation properties of carnation ringspot virus, Phytopathology 66:34.

Tremaine, J. H., Ronald, W. P., and McGauley, E. M., 1983, Effect of sodium dextran sulfate on some isometric plant viruses, Phytopathology 73:124l.

Tremaine, J. H., Ronald, W. P., and McGauley, E. M., 1984, Temperature-reversible aggregation of two strains of carnation ringspot virus, Phytopathology 74:16l.

Valenzuela, M., and Pizano, M., 1992, Distribution of carnation mottle virus and carnation ringspot virus in the Bogota Sabana, Acta Hortic. 307:22l.

Van Regenmortel, M. H. Y., 1982, Serology and Immunochemistry of Plant Viruses, Academic Press, New York.

Van Regenmortel, M. H. V., and Burckard, J., 1980, Detection of a wide spectrum of tobacco mosaic virus strains by indirect enzyme-linked immunosorbent assays, Virology 106:327.

Varmus, H., 1988, Retroviruses, Science 240:1427. Veidt, I., Lot, H., Scheidecker, D., Guilley, H., Richards, K., and Jonard, G., 1988, Nucleotide

sequence of beet western yellows virus RNA, Nucleic Acids Res. 16:9917. Weintraub, M., Ragetli, H. W. J., and Leung, E., 1975, Abnormal nuclear structures in mesophyll

cells infected with carnation ringspot virus, Phytomorphology 25:288. Xiong, Z., and Lommel, S. A., 1989, The complete nucleotide sequence and genome organization

of red clover necrotic mosaic virus RNA-I, Virology 171:543. Xiong, Z., and Lommel, S. A., 1991, Red clover necrotic mosaic virus infectious transcripts

synthesized in vitro, Virology 182:388.


Xiong, Z., Kim, K. H., Giesman-Cookmeyer, D., and Lommel, S. A., 1993a, The roles of the RCNMV capsid and cell-to-cell movement proteins in systemic infection, Virology 193:27.

Xiong, Z., Kim, K. H., Kendall, T. 1., and Lommel, S. A., 1993b, Synthesis of the putative red clover necrotic mosaic virus RNA polymerase by ribosomal frameshifting in vitro, Virology 193:213.

Yi, 1., Lesemann, D.-E., Koenig, R., Rudel, M., and Pfeilstetter, E., 1992, Isometric plant viruses found in ditches and streams in agricultural areas: recovery of previously found viruses and identification of hitherto unrecorded carmo- and tombusviruses including grapevine Algerian latent virus, r. Phytopathol. 134:121.

CHAPTER 11

Raspberry Bushy Dwarf Idaeovirus A. T. JONES, M. A. MAYO, AND A. F. MURANT

1. INTRODUCTION

In the 1930s and 1940s, Lloyd George became one of the most widely planted cultivars of red raspberry (Rubus idaeus) in Britain. It was grown for its flavor and quality and, in Scotland, for its resistance to soil-borne viruses. However, by the late 1950s, it had virtually disappeared from commercial production. Its rapid demise was attributed to the widespread incidence of "bushy dwarf" disease (Cadman and Harris, 1951), also known as "symptomless decline" (Cadman, 1952), which seriously decreased plant size, vigor, and fruit yield. Cadman (1961) detected in diseased plants a sap-transmissible virus that, because of its consistent association with the disease, he named raspberry bushy dwarf virus (RBDV). However, Barnett and Murant (1970) found that, on its own in young Lloyd George raspberry plants, RBDV did not induce obvious disease. Jones (1979) confirmed this by detailed observations of the growth, over 2 years, of RBDV-infected Lloyd George raspberry plants grown in pots in a gauze house. Nevertheless, he showed in the same experiment that, in mixed infection with black raspberry necrosis virus (BRNV), an aphid-borne virus, RBDV intensified the degeneration in vigor [characterized by stunting and proliferation (bushiness) of canes] caused by infection with

A. T. JONES, M. A. MAYO, AND A. F. MURANT • Scottish Crop Research Institute, Invergowrie, Dundee DD2 SDA, United Kingdom.

283

284 A. T. JONES ET AL.

BRNV alone. RBDV therefore does playa causal part in the total bushy dwarf syndrome.

Many of the red raspberry cultivars introduced in Britain in the 1970s and 1980s were, unlike cv. Lloyd George, resistant or immune to the strains of RBDV then prevalent (Jones, 1988), so that the attention given to the virus as a pathogen of Rubus decreased greatly during this period. However, fresh impetus was given to the study of RBDV by the discovery of a new and potentially damaging strain that could infect cultivars resistant or immune to the strain prevalent hitherto (Murant et a1., 1982; Barbara et a1., 1984; Jennings and Jones, 1989; Jones, 1993).

Barnett and Murant (1970) purified RBDV and showed that it had quasiisometric particles ca 33 nm in diameter. They produced an antiserum and showed that RBDV is serologically indistinguishable from the loganberry degeneration virus described by Legg (1960) and Ormerod (1970,1972). Jones et a1. (1982) demonstrated that RBDV is the causal agent of raspberry yellows disease, so that RBDV is also synonymous with raspberry yellows virus of Cadman (1952). The symptoms of raspberry line pattern disease reported from Poland (Basak, 1968) are very similar to those of raspberry yellows and, as the properties, symptomatology, and host range in herbaceous plants of raspberry line pattern virus are also very similar to those of RBDV (Basak, 1971), the two viruses are probably the same (Jones et a1., 1982).

Recent studies on the molecular biology of RBDV have shown it to possess a novel combination of properties, and it has been assigned, as the sole known member, to a new genus, Idaeovirus (Ziegler et a1., 1992; Mayo and Martelli, 1993; Murant and Mayo, 1995). Though RBDV has a bipartite genome, its closest affinities are with viruses that have tripartite genomes in the family Bromoviridae (Ziegler et a1., 1993).

II. BIOLOGICAL PROPERTIES

A. Geographical Distribution

RBDV probably occurs wherever red raspberry (R. idaeus, R. strigosus) or black raspberry (R. occidentalis) are grown. The virus has been reported from Australasia, eastern and western Europe, North and South America, South Africa, and the former Soviet Union (summarized by Murant, 1987). However, at least three variants of RBDV have been characterized and some are more restricted in their distribution.

1. Common or Scottish Isolates

The Scottish type isolate (S) is representative of isolates that have been identified in almost every country where raspberry is grown and are therefore sometimes referred to as the "common strain." They produce characteristic

RASPBERRY BUSHY DWARF IDAEOVIRUS 285

symptoms in herbaceous test plants and, under suitable environmental conditions, in sensitive raspberry cultivars. They are distinguished by a restricted Rubus host range, and many red raspberry cultivars are highly resistant, probably immune, to infection (Jones et a1., 1982). S isolates also fail to infect some hybrid berry cultivars (Jones et a1., 1982 j Barbara et a1., 1984).

2. Resistance-Breaking Isolates

Resistance-breaking (RB) isolates are indistinguishable from S isolates except that they are able to infect red raspberry cultivars and hybrid berry cultivars immune to S isolates (Barbara et a1., 1984 j Murant et a1., 1986). Indeed, they infect by grafting nearly all Rubus species and cultivars tested (Barbara et a1., 1984 j A. T. Jones, unpublished data). Because of this and the severe symptoms they may induce in some cultivars, some researchers have referred to RB isolates as "virulent" isolates (Barbara et a1., 1985). RB isolates seem restricted in their distribution to central and eastern Europe, Russia and Siberia, and, probably through importation of infected material, to isolated areas in England and some parts of western Europe (Barbara et a1., 1984 j

Murant, 1987 j Jones, 1993 j A. T. Jones, unpublished data).

3. Black Raspberry Isolates

Murant and Jones (1976) described two isolates of RBDV from the American black raspberry (R. occidentalis) (B isolates) that were serologically closely related to S isolates but distinguishable from them by spur formation in double-diffusion serological tests in agarose gel. This is the only evidence of serological variability in field isolates of RBDV. B isolates are more difficult to maintain in culture in herbaceous plants and their particles are more difficult to purify in quantity than S isolates. They are also difficult to detect in R. occidentalis extracts by inoculation to Chenopodium quinoa test plants but are readily detected by enzyme-linked immunosorbent assay (ELISA), suggesting that they may have a low specific infectivity (Murant and Jones, 1976 j Jones, 1986). Only B isolates were found in naturally infected R. occidentalis (A. T. Jones, unpublished data). However, more data are needed to determine whether B isolates are restricted to R. occidentalis and whether S isolates can occur in R. occidentalis in nature. B isolates have been found only in North America, the sole region where R. occidentalis is grown extensively.

B. Natural and Experimental Transmission

In red raspberry and black raspberry, RBDV is readily transmitted to progeny seedlings via either the pollen or the ovule of infected plants (Cadman, 1965 j Converse, 1973 j Murant et a1., 1974b transmission was greatest


(77%) when both parents were infected (Murant et a1., 1974). Cadman (1965) also reported that RBDV was transmitted to one of four raspberry plants pollinated with pollen from an infected plant. Transmission in association with pollen was confirmed in field and glasshouse studies by Murant et a1. (1974). Field plants became infected only when they were allowed to retain their flowering canes, and natural spread of RBDV seems to occur solely by this means. Healthy plants can be infected within the first two or three flowering seasons when planted close to virus sources (Converse, 1973 j Murant et a1., 1974 j Bulger et a1., 1990). In Canada, Bulger et a1. (1990) found that 84 % of Skeena raspberry plants were infected with RBDV by the fifth year of flowering.

Sdoodee and Teakle (1987) showed that Thrips tabaci is involved in the transmission of tobacco streak ilarvirus associated with pollen, apparently by inoculating the plants with virus carried by the pollen. This raises the question whether other so-called pollen-transmitted viruses, including RBDV, also are transmitted in this way. However, Bulger et a1. (1990) rarely found thrips in red raspberry flowers in the field experiment in Canada, though Thrips madronii and Frankliniella occidentalis were found commonly in flowers of wild R. parviflorus and R. procerus growing near the experimental plants. Furthermore, in laboratory experiments, C. quinoa plants did not become infected with RBDV when Thrips tabaci were caged on leaves dusted with pollen from RBDV-infected raspberry (Bulger et a1., 1990). Nevertheless, it cannot be concluded that insects or mites play no active or essential part in the transmission process. For example, it is not impossible that bees might serve not only to carry RBDV-infected pollen to healthy Rubus plants but, as a result of their foraging activities, to inoculate the virus to the plants.

Experimentally, RBDV is readily transmitted from Rubus by grafting to infectible Rubus species and cultivars. However, graft-inoculated plants usually need to go through a dormant season before the virus becomes systemic. RBDV is also transmissible from infected Rubus to herbaceous test plants by mechanical inoculation of sap extracts in either 2 % nicotine solution or 1 % polyethylene glycol (PEG) (mol. wt. 6000) (Barnett and Murant, 1970 j Murant, 1987). Mechanical transmission from RBDV-infected herbaceous plants to some infectible red raspberry cultivars is also possible by the same procedure, but with difficulty (Barnett and Murant, 1970 j Jones et a1., 1982).

C. Disease Symptoms and Effects in Rubus

1. Naturally Infected Plants

RBDV occurs in nature only in wild and cultivated Rubus species and cultivars. It infects cultivated red and black raspberry, blackberry (R. fruticosus), and hybrid berries such as boysenberry, loganberry, youngberry, and tayberry, and has been recorded in the wild red raspberry, R. parviflorus


(thimbleberry), R. phoenicolasius (wineberry), R. sachalinensis, and R. vulgatus ssp. buschii (summarized by Jones et a1., 1982; Credi et a1., 1986; Jones, 1986). When infecting alone, RBDV is often symptomless in Rubus plants, especially in Britain. However, in sensitive raspberry cultivars and under illdefined environmental conditions, it induces symptoms of yellows disease (Cadman, 1952; Jones and Wood, 1979; Jones et a1., 1982). This disease, seen most clearly in new cane growth (primocanes) of plants in late spring/early summer, is characterized by an initial yellowing of the main or minor veins of the lower leaves or leaflets. This yellowing progressively extends to other veins and ultimately to the interveinal lamina, sometimes affecting the whole leaf (Fig. 1). Some raspberry cultivars may also develop chlorotic/ yellow rings or line patterns in leaves (Daubeny et a1., 1978; Jones and Wood, 1979). Although the leaf symptoms persist, leaves produced in midsummer are, in general, symptomless and tend to obscure symptom-bearing leaves.

RBDV infection also causes "crumbly fruit" in some raspberry cultivars (Murant et a1., 1974; Daubeny et a1., 1978; Murant, 1987). This condition, arising from the abortion of some drupelets and the uneven development of others, frequently results in abnormally shaped fruit, which on picking may disintegrate into individual drupelets or drupelet clusters (Fig. 2). The same disease syndrome can be induced by other unrelated causes, such as poor pollination conditions, infection with other viruses (Taylor et a1., 1965; Daubeny et a1., 1970), or genetic aberrations (Murant et a1., 1973).

Additional symptoms have been attributed to RB isolates. In studies in England, Wilson et a1. (1983) and Barbara et a1. (1984) reported an association between infection with RBDV-RB and premature defoliation of fruiting canes, decreased vigor, leaf curling, necrosis and death of laterals, and increased winter kill in the red raspberry cvs. Glen Clova, Heritage and Malling Jewel, each of which is immune to S isolates of this virus (Jones et a1., 1982). Although this evidence is only circumstantial, it suggests that these severe symptoms are caused by infection with RBDV-RB (Barbara et a1., 1984).

2. Experimentally Infected Plants

Most Rubus species and cultivars commonly used as indicators for other Rubus viruses were either not infected or infected symptomlessly when graft-inoculated with RBDV (Jones et a1., 1982). The exceptions were R. henryi, which developed a transient chlorotic mottle unlike the severe tip necrosis and leaf curling caused by infection with other, unrelated, viruses (Jones, 1986), and R. molaccanus, which developed prominent chlorotic line patterns and interveinal yellowing (Jones et a1., 1982).

RBDV has been transmitted by graft inoculation from Rubus to perennial plants in other genera. It symptomlessly infected Pragaria vesca and Prunus mahaleb seedlings (Barnett and Murant, 1970) and induced pro-


FIGURE l. Leaves of RBDV-infected red raspberry showing, from upper left through to lower right, the progressive development of yellows disease symptoms.


FIGURE 2. Symptoms of crumbly fruit of red raspberry caused by RBDV infection.

nounced chlorotic vein banding and line patterns in Cydonia oblonga cv. C7-1 (Jones et a1., 1982).

D. Symptoms in Herbaceous Test Plants

RBDV has a moderately wide host range, but infects most hosts symptomlessly. By mechanical inoculation of RBDV-infective sap, Barnett and Murant (1970, 1971) infected 55 plant species from 12 dicotyledonous families. Chenopodium amaranticolor and C. quinoa are sensitive to infection and reliably develop systemic chlorotic spots and line patterns (Fig. 3); occasionally, inoculated leaves of these species may also develop chlorotic or necrotic lesions. Some other hosts develop symptoms erratically. For exam-


FIGURE 3. Chlorotic rings and line patterns caused by systemic infection with RBDV in a leaf of Chenopodium amaranticolor.

pIe, Phaseolus vulgaris cv. The Prince develops local pinpoint necrotic lesions without systemic invasion, but usually only in winter. Seedlings of some forms of Petunia hybrida segregate for reactions to RBDV inoculation; some are systemically infected without developing symptoms, whereas others develop large necrotic local lesions without systemic invasion. Tomato is immune to RBDV and is a useful host for separating other viruses from RBDV when they occur in mixed infections in Rubus (Barnett and Murant, 1970; Jones, 1977). Other commonly used herbaceous test plants such as Nicotiana c1evelandii, N. glutinosa, N. benthamiana, Spinacia oleracea, and Tetragonia expansa are systemically infected but develop no obvious symptoms (Barnett and Murant, 1970, 1971).


III. PARTICLE PROPERTIES

A. Purification of Virus Particles

Virus particles are best purified from leaf tissue of C. quinoa, as described by Barnett and Murant (1970), Murant (1976), or Murant et al. (1986). A satisfactory procedure is as follows. Pick inoculated and systemically infected leaves about 10 days after inoculation and extract each 100 g leaf tissue in 400 ml 0.05 M sodium phosphate buffer, pH 7.0, containing an antioxidant such as 0.01 M thioglycerol. Adjust the pH to 4.8 with 0.05 M citric acid and centrifuge at low speed. In different laboratories the virus may at this stage be in the pellet or supernatant fractions; the reason for this variation in behavior is not known. If the virus is in the pellet, recover it by suspension overnight in 0.05 M citrate buffer, pH 6.0, followed by low-speed centrifugation to remove insoluble material. If the virus is in the supernatant fluid following the pH 4.8 step, precipitate it by adding PEG, molecular weight 6000, to 8% (w/v) and NaCI to 0.8% (w/v), resuspending the pellets in citrate buffer. Concentrate the virus by differential centrifugation and further purify it by exclusion chromatography in columns of 2 % agarose beads and/or by centrifugation in sucrose density gradients.

B. Properties of Virus Particles

RBDV particles are about 33 nm in diameter and appear quasi-isometric (Fig. 4), because they tend to collapse and deform on the electron microscope grid. They appear more spherical in shape when aldehyde-fixed (I. M. Roberts and A. F. Murant, unpublished data). Particle preparations have A26o/A28o of 1.62, suggesting that RBDV particles contain about 24 % RNA. The particles are unstable and readily disrupt in the presence of 0.01 % sodium dodecyl sulfate, indicating that they are stabilized by protein-RNA linkages (Murant, 1976).

The particles sediment as a rather broad band in sucrose density gradients, with a sedimentation coefficient (s~o,wl of 115 Sin 0.05 M citrate buffer, pH 6. Formaldehyde-fixed virus particles, centrifuged to equilibrium at 20°C in solutions of CsCI or RbBr, form a band at a density of 1.37 g ml- 1

(Murant, 1975). Very rarely, preparations also contain particles ca 15 nm in diameter that sediment at ca 34 S and are serologically indistinguishable from the 33-nm-diameter particles.

C. Composition of Virus Particles

Purified preparations of RBDV particles typically yield a single major protein estimated by polyacrylamide gel electrophoresis to have a Mr of


FIGURE 4. Electron micrograph of a purified preparation of RBDV particles stained in uranyl formate. Bar, 100 nm.

about 30,000 (30 kDa). However, some preparations, even highly purified ones, contain minor amounts of smaller proteins.

The particles contain three species of ssRNA with Mr about 2.0 x 106

(RNA-I), 0.8 x 106 (RNA-2), and 0.3 x 106 (RNA-3) (Murant, 1975). The three RNA species are sometimes present in approximately equimolar amounts but, in some preparations and with some isolates, RNA-3 (or RNA material of about the same size) is often present in much greater molar concentration than RNA-l and RNA-2. The way in which the RNA molecules are packaged in the particles is unknown, but it is unlikely that all three RNA species are contained within the same particle.


A. Nucleotide Sequences

The complete nucleotide sequences have been determined for the three RNA species present in particles of isolate R15, an RB isolate. Figure 5 is a diagram of the genome structure of RBDV deduced from these sequences. RNA-3 is 946 nucleotides (nt) long and contains a single open reading frame (ORF), which encodes the 30-kDa coat protein (Mayo et a1., 1991). This predicted size corresponds to that of protein obtained from purified virus

RASPBERRY BUSHY DWARF IDAEOVIRUS

RNA-1 (5449 nt)

••• ...... 190 K

0000

D 12 K

(putative)

RNA-2 (2231 nt)

**

39 K

RNA-3 (946 nt)

30K (coat protein)

293

RNA-3

FIGURE 5. Diagram showing the genome structure of RBDV based on the RNA sequences. RNA molecules are represented by solid lines and the putative positions of the products of the ORFs are indicated as boxes. The positions of domains in the polyproteins are shown as " residues shared among putative transport proteins; ., methyl transferase; x, NTP-binding; and o RNA-dependent RNA polymerase. The dotted lines indicate that RNA-3 is derived from the 3'-terminal portion of RNA-2.

particles and that of the main polypeptide made by in vitro translation of RNA-3 (Murant et a1., 1986; Mayo et a1., 1991).

RNA-2 is 2231 nt long and its 3' half contains the entire sequence of RNA-3 (Natsuaki et a1., 1991); RNA-3 is thus a sub genomic mRNA. In addition, RNA-2 contains, in its 5' half, a second ORF which encodes a 39-kDa protein_ Polypeptides obtained by in vitro translation of RNA-2 appeared to be larger than this: translation in wheat germ extracts produced a 46-kDa protein (Murant et a1., 1986) and translation in reticulocyte lysate produced a 42-kDa protein (M. A. Mayo, unpublished data).


RNA-l is 5449 nt long and contains one large ORF, which encodes a 190-kDa protein (Ziegler et a1., 1992). This protein corresponds in size to the largest polypeptide made by in vitro translation (Murant et a1., 1986). A small ORF is also present in a different translation frame and overlapping the ORF for the 190-kDa protein (Fig. 5). This ORF encodes a 12-kDa polypeptide. Although no corresponding subgenomic RNA was detected in RNA extracted from RBDV-infected plants (Ziegler et a1., 1992) further investigation is needed in view of the discovery by Ding et a1. (1994) of a small functional ORF near the 3' end of RNA-2 of cucumber mosaic virus (CMV) which encodes a lO-kDa polypeptide and overlaps the ORF for the CMV polymerase.

There are similarities among the 5' - or 3' -terminal noncoding sequences of each of the three RNA species. At the 5' end of RBDV RNA, the first six to eight nucleotides are similar in all three RNA species, the consensus being UAUUU. At the 3' end, RNA-3 is identical to RNA-2 (being derived from it), and RNA-l and RNA-2 share sequence and structural similarities (Natsuaki et a1., 1991). The 3' -terminal 18 nucleotides are identical in RNA-l and RNA~2, and the terminal 71 nucleotides of each RNA can be folded into similar stem-loop structures [Fig. 6, which is a corrected version of Fig. 5 from Natsuaki et a1. (1991)].

The noncoding region between the two ORFs in RNA-2 contains the start of RNA-3 and presumably also contains sequences responsible for the initiation and promotion of its transcription. Like the comparable region of bicistronic RNAs of other viruses, this part of RBDV RNA-2 is rich in A and U residues (Natsuaki et a1., 1991). No dsRNA was found in infected tissue that could correspond to a double-stranded form of RNA-3 (Murant et a1., 1986), and RNA-3 is therefore probably transcribed from a minus-sense copy of RNA-2.

B. Sequence Features of the Viral Proteins

1. The 30-kDa Protein (Coat Protein)

The coat protein of RBDV shows no striking similarities with the coat proteins of other viruses. In comparisons made by using the program CLUS

TALV (Higgins et a1., 1992), RBDV coat protein was between 10% [alfalfa mosaic virus (AIMV)] and 19% [tobacco streak virus (TSV)] identical to coat proteins of viruses in the family Bromoviridae. An alternative to this sequence-based alignment program is to use secondary structure predictions to produce an alignment (Rost and Sander, 1993, 1994). The RBDV coat protein structure predicted by such an alignment was of a long N-terminal sequence containing two a-helices and followed by a region containing eight [3-sheets. If correct, this arrangement would resemble that of the coat proteins of many viruses that have isometric particles (Rossmann and Johnson, 1989). The N-terminal parts of the coat proteins of RBDV and of viruses in the


RNA-2

RNA-1

U A U A-U

U A C-G U G C G A-U A A U A G-U C-G C-G C-G U-G O-A G-C G-C C-G C-G G-C U-A C-G U-A C-G U-A G-C A-U A-U U-A

U - U - A - U - A - A - U - G - C - C - C - COH

U C U C A A U G-C A-U G A-U C-G U A U-A C-G U A C-G C-G C-G U C G-C C-G U-A C-G U-A G-C C-G C-G U-A A-U A-U U-A U-A

U -A-U-A-U-G - C-G - C - C - C - COH

FIGURE 6. Diagram showing the sequence similarities and putative stem-loop structures at the 3' ends of RBDV RNA-l and RNA-2.

Bromoviridae are relatively rich in basic amino acids, which are thought to be internal and to interact with the encapsidated RNA.

2. The 39-kDa Protein

RNA-2 of RBDV corresponds in size and gene content to RNA-3 of viruses in the family Bromoviridae. The ORF for the 39-kDa protein therefore corresponds in position to the ORF for the putative movement protein of, for example, AlMV (Van der Kuyl et al., 1991). A weak similarity was detected in multiple alignment tests between the 39-kDa protein and the putative or actual movement proteins of several viruses (U. Melcher, quoted in Natsuaki et al., 1991). In further alignment tests of this sort, Mushegian and Koonin (1993) have detected a match between part of the 39-kDa protein sequence (indicated in Fig. 5) and a motif conserved in the movement proteins of some other viruses, such as red clover necrotic mosaic dianthovirus and soil-borne wheat mosaic furovirus. These proteins were proposed to


belong to a 30-kDa superfamily of movement proteins (Mushegian and Koonin, 1993).

3. The 190-kDa Protein

The sequence of the 190-kDa protein has similarities with replicase proteins of viruses in the Sindbis-like supergroup (Ziegler et a1., 1992). There was good correspondence between parts of the 190-kDa protein and parts of the 183-kDa protein of tobacco mosaic virus, but the greatest similarity was with a hypothetical composite protein consisting of the 90-kDa protein of AIMV (encoded by RNA-2) attached to the C-terminus of the AIMV 125-kDa protein (which is encoded by RNA-I). Figure 7 shows a plot of this comparison made by using the program COMPARE in the GCG package (Devereux et a1., 1984). The three regions of greatest similarity (marked., x, and 0 in Fig. 5) correspond to the locations in the AIMV proteins of the methyl transferase, NTP-binding, and RNA-dependent RNA polymerase domains (Ziegler et a1., 1992). The first two domains occur in the 125-kDa protein of AIMV and the polymerase domain occurs in the 90-kDa protein of AIMV. RBDV RNA-l therefore corresponds in size and gene functions to RNA-l plus RNA-2 of AIMV and related viruses. Comparisons of the three domains of several viruses by using CLUSTALV suggested that RBDV is marginally more similar to viruses in the family Bromoviridae than to other viruses; the maximum similarity was to AIMV (Ziegler et a1., 1993).

V. DETECTION AND CONTROL

A. Detection in Plants

In Rubus, RBDV may induce yellowing, line patterns, or crumbly fruit, but such symptoms are unreliable for diagnosis because they may be due to infection with other disease agents or to other causes. Furthermore, RBDV often infects plants symptomlessly. Detection and diagnosis therefore depend on bioassays and/or serological tests.

RBDV can be reliably detected by one or other of the various forms of ELISA (Barbara et a1., 1984) and this is the preferred detection method for B isolates, which seem to have a low specific infectivity (Murant, 1987; A. T. Jones, unpublished data). Other studied isolates are readily detected by mechanical transmission to C. quinoa test plants. However, identification of RBDV in symptom-bearing herbaceous test plants is dependent on serological tests. As the distribution of RBDV in individual Rubus plants can be erratic (Barbara et a1., 1985), leaves/leaflets for mechanical transmission or ELISA should be taken and pooled from at least three different nodes.

Polyclonal antisera raised to S isolates have a moderate titer of ca 1/512 in gel double-diffusion tests (Barnett and Murant, 1970). RBDV antigen in


I I I

" " /

500 - , / AIMV -/ 90kD /

'"

" I I I

I I I

" 1000 -, -" " '/

/

" AIMV " 125kD

" 500 - -

,,-./

, /

/ /

I I I I

500 1000 1500

RBDV 190kD

FIGURE 7. Dot plot comparisons showing similarities between (lower part) the N-terminal part of the RBDV RNA-l product and the 125-kDa protein of AlMV and (upper part) the C-terminal part of the RBDV RNA-l product and the 90-kDa protein of AlMV (window 30 residues, stringency 17). The numbers on the axes indicate the numbering of the amino acid residues.

purified virus preparations or in infective sap of C. quinoa reacts with such antisera in microprecipitin tests and in agarose gel double-diffusion serological tests (Barnett and Murant, 1970). Spur formation in this latter test can be used to distinguish S or RB isolates from B isolates (Murant and Jones, 1976). Martin (1984) obtained monoclonal antibodies that were able to distinguish three different epitopes of a Canadian isolate of RBDV. However, these


monoclonal antibodies reacted in double antibody sandwich ELISA with all tested isolates from raspberry, including B, S, and RB isolates (A. T. Jones and W. McGavin, unpublished data).

B. Therapy

Unlike many other viruses infecting Rubus, RBDV is not readily eradicated from Rubus plants by heat treatment alone (Converse, 1973 j Murant et a1., 1974j Mellor and Stace-Smith, 1976j Lankes, 1995). However, heat treatment for several weeks at 36°C followed by propagation from shoot tips or apical meristems eliminated the virus from some plants (Converse, 1973 j

Murant et a1., 1974j Mellor and Stace-Smith, 1976j Lankes, 1995). Using only tissue culture of meristem tips, Theiler-Hedtrich and Baumann (1989) found it necessary to subculture up to three times to eliminate the virus. An alternative approach used with some success by Kudell and Buchenauer (1989) is to grow axillary bud cultures on media containing suitable concentrations of the antiviral compounds ribavirin and dodecyl-N-methylephedrinium bromide. Although RBDV was not detectable in the treated cultures, no information was given on whether the regenerated plants were tested for RBDY.

C. Control in Crops

Because RBDV is transmitted in association with infected pollen, the only methods of controlling it in crops are to grow resistant cUltivars or to plant healthy stock of infectible cultivars and grow them in isolation from possible sources of infection. Many red raspberry cultivars are resistant, possibly immune, to S isolates (the most widespread isolates worldwide) (Jones et a1., 1982). This resistance is conferred by the presence of a single dominant gene, Eu (Jones et a1., 1982, Murant et a1., 1982). However, almost all Eu-containing cultivars are infectible by grafting with RB isolates (Barbara et a1., 1984j A. T. Jones unpublished data). Resistance to graft inoculation with RB isolates was detected in only a very few red raspberry cultivars (Barbara et a1., 1984). Studies on this form of resistance in the North American cv. Haida indicated that its mode of inheritance was complex but seemed to depend on the presence of gene Eu, together with a second resistance component whose inheritance was probably multigenic (Jennings and Jones, 1989). In field studies in England, some raspberry cultivars, including some that did not contain gene Eu, were more resistant than others to natural infection with RB isolates (Wilson et a1., 1983 j Barbara et a1., 1984). This might suggest inherent resistance to field infection with RB isolates in some cultivars that are infectible by grafting. In the long term, effective control of the spread and effects of RB isolates in Rubus may be best achieved through


genetically engineered resistance produced by genes derived from RBDV (Jones, 1993).

VI. RELATIONSHIPS WITH OTHER VIRUSES

RBDV particles have a morphology very similar to that of ilarvirus particles. Moreover, the symptoms it induces in natural hosts and its natural transmission both through seed and to plants pollinated with infected pollen are further properties that RBDV shares with ilarviruses. Nevertheless, RBDV is serologically unrelated to all recognized ilarviruses tested (Barnett and Murant, 1970 j A. T. Jones unpublished data). Also, RBDV differs from ilarviruses in the sedimentation behavior of its particles, in the number and sizes of its RNA molecules, and in having a bipartite genome (Barnett and Murant, 1970 j Natsuaki et al., 1991 j Ziegler et al., 1992, 1993). Taken together, these properties distinguish RBDV from all other well-characterized viruses and justify its classification as the sole known member of the genus Idaeovirus (Murant and Mayo, 1995). The genus has not been assigned to a family (Mayo and Martelli, 1993), although Ziegler et al. (1993) suggested that, largely because of the molecular properties of RBDV, Idaeovirus should be classified in the family Bromoviridae, albeit as an atypical member of a family whose four recognized genera contain viruses with tripartite genomes.

REFERENCES

Barbara, D. J., Jones, A. T., Henderson, S. J., Wilson, S. C., and Knight, V. H., 1984, Isolates of raspberry bushy dwarf virus differing in Rubus host range, Ann. Appi. Bioi. 105:49.

Barbara, D. J., Ashby, S. C., and Knight, V. H., 1985, The occurrence and distribution of isolates of raspberry bushy dwarf virus in England, Ann. Appi. Bioi. 106:75.

Barnett, O. W., and Murant, A. F., 1970, Host range, properties and purification of raspberry bushy dwarf virus, Ann. Appi. Bioi. 65:435.

Barnett, O. W., and Murant, A. F., 1971, Differential hosts and some properties of raspberry bushy dwarf virus, Ann. Phytopathoi. numero hors serie: 129.

Basak, W., 1968, Line pattern disease of raspberry, Bull. Acad. Polan. Sci., Ser. Sci. Bioi. 16:307. Basak, W., 1971, Hosts and properties of raspberry line pattern virus, Bull. Acad. Polan. Sci., Ser.

Sci. Bioi. 19:681. Bulger, M. A., Stace-Smith, R., and Martin, R. R., 1990, Transmission and field spread of

raspberry bushy dwarf virus, Plant Dis. 74:514. Cadman, C. H., 1952, Studies in Rubus virus diseases. IV. Yellows diseases of raspberries, Ann.

Appi. Bioi. 39:495. Cadman, C. H., 1961, Mechanical transmission of viruses from raspberry, Rep. Scott. Hortic.

Res. Inst. 8:57. Cadman, C. H., 1965, Filamentous viruses infecting fruit trees and raspberry and their possible

mode of spread, Plant Dis. Rep. 49:230. Cadman, C. H., and Harris, R. V., 1951, Raspberry virus diseases: A survey of recent work, Rep.

East Malling Res. Stn. 1950:127.


Converse, R. H., 1973, Occurrence and some properties of raspberry bushy dwarf virus in Rubus species in the United States, Phytopathology 63:780.

Credi, R., Shier, 1. L., and Stace-Smith, R., 1986, Occurrence of raspberry bushy dwarf virus in native thimbleberry in British Columbia, Acta Hortic. 186:17.

Daubeny, H. A., Freeman, J. A., and Stace-Smith, R., 1970, Effects of virus infection on drupelet set of four red raspberry cultivars, f. Am. Soc. Hortic. Res. 95:730.

Daubeny, H. A., Stace-Smith, R., and Freeman, J. A., 1978, The occurrence and some effects of raspberry bushy dwarf virus in red raspberry, f. Am. Soc. Hortic. Res. 103:519.

Devereux, J., Haeberli, P., and Smithies, 0., 1984, A comprehensive set of sequence analysis programs for the VAX, Nucleic Acids Res. 12:387.

Ding, S.-W., Anderson, B. J., Haase, H. R., and Symons, R. H., 1994, New overlapping gene encoded by the cucumber mosaic virus genome, Virology 198:593.

Higgins, D. G., Blebs, A. J., and Fuchs, R., 1992, CLUSTALV: Improved software for multiple sequence alignment, CABIOS 8:189.

Jennings, D. 1., and Jones, A. T., 1989, Further studies on the occurrence and inheritance of resistance in red raspberry to a resistance-breaking strain of raspberry bushy dwarf virus, Ann. Appl. Biol. 114:317.

Jones, A. T., 1977, Partial purification and some properties of wineberry latent, a virus obtained from Rubus phoenicolasius, Ann. Appl. Biol. 86:199.

Jones, A. T., 1979, The effects of black raspberry necrosis and raspberry bushy dwarf viruses in Lloyd George raspberry and their involvement in raspberry bushy dwarf disease, f. Hortic. Sci. 54:267.

Jones, A. T., 1986, Advances in the study, detection and control of viruses and virus diseases of Rubus with particular reference to the United Kingdom, Crop Res. 26:127.

Jones, A. T., 1988, The influence of cultivating new raspberry varieties on the incidence of viruses in raspberry crops in the UK, Aspects Appl. Biol. 17:179.

Jones, A. T., 1993, Possibilities and problems for the control of viruses infecting Rubus and Ribes crops in Europe, Acta Hortic. 352:547.

Jones, A. T., and Wood, G. A., 1979, The virus status of raspberries (Rubus idaeus 1.1 in New Zealand, N. Z. f. Agric. Res. 22:173.

Jones, A. T., Murant, A. F., Jennings, D. L., and Wood, G. A., 1982, Association of raspberry bushy dwarf virus with raspberry yellows disease; reaction of Rubus species and cultivars, and the inheritance of resistance, Ann. Appl. Biol. 100:135.

Kudell, A. R., and Buchenauer, H., 1989, Elimination of raspberry bushy dwarf virus from axillary bud cultures of red raspberry cv. Lloyd George by antiviral compounds, ,. Phytopathol. 124:332.

Lankes, C., 1995, Control of raspberry bushy dwarf virus, Acta Hortic. 385:70. Legg, J. T., 1960, A virus-degeneration of loganberry, Rep. East Malling Res. Stn. 1959:102. Martin, R. R., 1984, Monoclonal antibodies define three different antigenic regions on raspberry

bushy dwarf virus, Can. f. Plant Pathol. 6:264. Mayo, M. A., and Martelli, G. P., 1993, New families and genera of plant viruses, Arch. Virol.

133:496. Mayo, M. A., Jolly, C. A., Murant, A. F., and Raschke, J. H., 1991, Nucleotide sequence of

raspberry bushy dwarf virus RNA-3, J. Gen. Virol. 72:469. Mellor, F. c., and Stace-Smith, R., 1976, Influence of heat therapy on rooting of, and elimination

of raspberry bushy dwarf virus, from shoot cuttings of red raspberry, Acta Hortic. 66:63. Murant, A. F., 1975, Some properties of the particles of raspberry bushy dwarf virus, Proc. Am.

Phytopathol. Soc. 2:116. Murant, A. F., 1976, Raspberry bushy dwarf virus, CMI/AAB Descriptions of Plant Viruses

No. 165. Murant, A. F., 1987, Raspberry bushy dwarf, in: Virus Diseases of Small Fruits (R. H. Converse,

ed.I, pp. 229-234, USDA Agriculture Handbook No. 631, Washington D.C. Murant, A. F., and Jones, A. T., 1976, Comparison of isolates of raspberry bushy dwarf virus from

red and black raspberries, Acta Hortic. 66:47.


Murant, A. F., and Mayo, M. A., 1995, Genus Idaeovirus, in: Virus Taxonomy-Classification and Nomenclature of Viruses: Sixth Report of the International Committee on Taxonomy of Viruses (F. A. Murphy, C. M. Fauquet, D. H. 1. Bishop, S. A. Ghabrial, A. W. Jarvis, G. P. Martelli, M. A. Mayo, and M. D. Summers, eds.), pp. 458-460, Springer-Verlag, Vienna (also in Arch. Virol. Suppl. 10).

Murant, A. F., Jennings, D. L., and Chambers, J., 1973, The problem of crumbly fruit in raspberry nuclear stocks, Hortic. Res. 13:49.

Murant, A. F., Chambers, J., and Jones, A. T., 1974, Spread of raspberry bushy dwarf virus by pollination, its association with crumbly fruit, and problems of control, Ann. Appl. Biol. 77:27l.

Murant, A. F., Jones, A. T., and Jennings, D. 1., 1982, Problems in the control of raspberry bushy dwarf virus, Acta Hortic. 129:77.

Murant, A. F., Mayo, M. A., and Raschke, J. H., 1986, Some biochemical properties of raspberry bushy dwarf virus, Acta Hortic. 186:23.

Mushegian, A. R., and Koonin, E. V., 1993, Cell-to-cell movement of plant viruses, insights from amino acid sequence comparisons of movement Pl"oteins and from analogies with cellular transport systems, Arch. Virol. 133:239.

Natsuaki, T., Mayo, M. A., Jolly, C. A., and Murant, A. F., 1991, Nucleotide sequence of raspberry bushy dwarf virus RNA-2: A bicistronic component of a bipartite genome, T. Gen. Virol. 72:2183.

Ormerod, P. J., 1970, A virus associated with loganberry degeneration disease, Rep. East Malling Res. Stn. 1969:165.

Ormerod, P. J., 1972, Rubus virus investigations. Loganberry, Rep. East Malling Res. Stn. 1971:127.


Rost, B., and Sander, c., 1993, Improved prediction of protein secondary structure by use of sequence profiles and neural networks, Proc. Natl. Acad. Sci. USA 90:7558.

Rost, B., and Sander, C., 1994, Combining evolutionary information and neural networks to predict protein secondary structure, Proteins 19:55.

Sdoodee, R., and Teakle, D. S., 1987, Transmission of tobacco streak virus by Thrips tabaci: A new method of plant virus transmission, Plant Pathol. 36:377.

Taylor, C. E., Chambers, J., and Pattullo, w.1., 1965, The effect of tomato black ring virus on the growth and yield of MaIling Exploit raspberry, Hortic. Res. 5:19.

Theiler-Hedtrich, R., and Baumann, G., 1989, Elimination of apple mosaic virus and raspberry bushy dwarf virus from infected red raspberry (Rubus idaeus 1.) by tissue culture, f. Phytopathol. 127:193.

Van der Kuyl, A. C., Neeleman, 1., and Bol, J. F., 1991, Deletion analysis of cis- and trans-acting elements involved in replication of alfalfa mosaic virus RNA3 in vivo, Virology 183:73l.

Wilson, S. C., Knight, V. H., and Barbara, D. J., 1983, Raspberry bushy dwarf virus and field infection of Malling Jewel, Plant Pathol. 32:357.

Ziegler, A., Natsuaki, T., Mayo, M. A., Jolly, C. A., and Murant, A. F., 1992, The nucleotide sequence of RNA-l of raspberry bushy dwarf virus, f. Gen. Viral. 73:3213.

Ziegler, A., Mayo, M. A., and Murant, A. F., 1993, Proposed classification of the bipartitegenomed raspberry bushy dwarf idaeovirus, with tripartite-genomed viruses in the family Bramoviridae, Arch. Viral. 131:483.

CHAPTER 12

Pea Enation Mosaic Enamovirus: Properties and Aphid Transmission S. A. DEMLER, G. A. DE ZOETEN, G. ADAM,

AND K. F. HARRIS

1. INTRODUCTION

Pea enation mosaic enamovirus (PEMV) has long been recognized to have a unique combination of properties and has been assigned to a monotypic group (Harrison et al., 1971), now given the generic name Enamovirus (de Zoeten and Demler, 1995). The generic characters include a bipartite, plussense, single-stranded RNA (ssRNA) genome, both RNA species (RNA-l and RNA-2) being needed to establish a systemic infection and each being encapsidated in separate morphologically distinct icosahedral particles. The virus is transmissible by aphids in the persistent (circulative) manner and by mechanical inoculation. Some isolates also contain a nonessential third RNA species, designated RNA-3.

PEMV has several attributes of special interest. These include the ultrastructural effects of infection on plant tissue and the ultrastructural features

S. A. DEMLER AND G. A. DE ZOETEN • Department of Botany and Plant Pathology, Michigan State University, East Lansing, Michigan 48824. G. ADAM • Universitat Hamburg, Institut fur Angewandte Botanik, D-20309, Hamburg, Germany. K. F. HARRIS • Department of Entomology, Texas A&M University, College Station, Texas 77843.

303

304 S. A. DEMLER ET AL.

found in viruliferous aphids. In addition, the particles of PEMV can be obtained in greater quantity than those of other viruses transmitted in the circulative manner, such as the luteoviruses. This has enabled the biochemical and structural features of PEMV to be studied more readily than those of luteoviruses, resulting in the identification of protein features involved in circulative aphid transmission.

Although PEMV has these advantages, there has been disagreement about the nature of its genome and the roles of individual particle types and RNA components in infection. Recent molecular biological studies have resolved many of these points of contention and have shown that PEMV RNA species interact in complex ways that have wider implications in plant virology. In this chapter we summarize and reanalyze the older information and integrate it with the results of recent studies.

II. BIOLOGICAL PROPERTIES

A. Experimental Hosts

Almost all the plant species that PEMV infects systemically are legumes. However, among nonleguminous plants, Nicotiana clevelandii, N. benthamiana, N. tabacum cv. White Burley, and Gomphrena globosa were found to be susceptible when inoculated mechanically. Local lesion hosts useful for quantitative infectivity assay include Chenopodium album, C. amaranticolor, C. quinoa, and aGalactia sp. (Hagedorn et a1., 1964; Izadpanah and Shepherd, 1966a; Gonsalves and Shepherd, 1972; Hull and Lane, 1973; Mahmood and Peters, 1973). Test plants differ in susceptibility to infection, and their response is affected by environmental conditions. Halfleaf or Latin square designs are therefore recommended for bioassays to minimise the effects of this variability (Mahmood and Peters, 1973). For virus purification, the most satisfactory source plant is Pisum sativum (especially varieties with bushy growth like Progress No.9), although purification from other sources has been reported (Izadpanah and Shepherd, 1966b; French et a1., 1973; Motoyoshi and Hull, 1974).

B. Symptoms

The symptoms associated with PEMV infection vary greatly, depending on the host, the virus isolate, and the environment. The virus induces the severest symptoms and reaches highest particle concentrations between 16 and 22°C. Infected pea plants initially show downward leaf curling at 5-7 days postinoculation (PI), and chlorotic or translucent spots appear on the leaves. Vivid chlorotic vein clearing often develops from 7 to 14 days PI. As the disease progresses, these foliar symptoms intensify, and the host develops growth malformations, including epinasty, severe stunting, rugosity,

PEMV: PROPERTIES AND TRANSMISSION 305

and loss of apical dominance. In severely affected plants, necrosis appears at the leaf tips and growing point. At 2-3 weeks PI, diagnostic blister or ridgelike enations appear, mainly on the underside of leaves and stipules, particularly along the leaf veins. Pods are distorted, undersized, and nonmarketable, and are often covered with wartlike outgrowths.

PEMV typically induces less severe symptoms in Nicotiana species than in leguminous hosts. N. c1evelandii and N. benthamiana develop a yellow, sometimes necrotic, mosaic, accompanied by varying amounts of curling and puckering of younger leaves, with stunting of the plant (Hagedorn et a1., 1964; Demler et a1., 1994b). In contrast, infections of N. tabacum cv. White Burley are largely symptomless (Motoyoshi and Hull, 1974). Enations have not been reported in Nicotiana spp.

C. Cytopathology

PEMV induces characteristic cytopathological effects that somewhat resemble those of luteoviruses. However, in contrast to luteoviruses, which are phloem-limited, PEMV occurs in most plant tissues and reaches greater concentrations. PEMV virions can therefore be detected easily by electron microscopy of dip preparations or by immunosorbent electron microscopy of crude sap from infected plants. In ultrathin sections, large accumulations of virus particles are found in the cytoplasm of all cell types and also in the nucleus, particularly in the region of the nucleolus (Fig. la) (Shikata and Maramorosch, 1966; Shikata et a1., 1966; Burgess et a1., 1974; Motoyoshi and Hull, 1974; Demler et a1., 1994a). It is not known whether virions are assembled within the nucleus or are transported to it from the cytoplasm. In general, virions occur randomly scattered or in loose aggregates, although some isolates produce massive paracrystalline arrays of virions in the cytoplasm (Shikata and Maramorosch, 1966; Shikata et a1., 1966; Demler et a1., 1994a).

The cytoplasm of PEMV-infected cells contains extensive networks of double-membraned vesicles enclosing fibrillar material (Fig. Ib) (de Zoeten et a1., 1972; Burgess et a1., 1974). These vesicles occur in all cell types and are particularly numerous in phloem tissue. Time-course studies revealed that the vesicles originate at the inner surface of the nuclear membrane and subsequently migrate to the perinuclear space. From here, they are extruded through the external nuclear membrane into the cytoplasm as doublemembrane-bounded clusters. The presence of these vesicles in both phloemrelated and superficial tissues, together with their distinctive double membranes, is diagnostic for PEMV infection.

In biochemical and ultrastructural studies, the replication of PEMV was found to be directly linked to the nucleus and to the vesicles originating from the nuclear membrane. Replication in nuclei was demonstrated by precursor incorporation in infected tissue in the presence of actinomycin D (AMD) and


(a) (b)

FIGURE 1. Cytopathological effects of PEMV infection in Pisum sativum leaf cells. N, nucleus; Ne, nuclear envelope; Cy, cytoplasm. (a) Immunogold detection of virions in the nucleus. The sections were treated with PEMV antibodies were detected with protein A conjugated to IS-nm gold particles. Note the large concentration of labeled viruslike particles (V) in the nucleolus. (b) Fibril-containing vesicles (Ve) associated with the nuclear envelope (Ne) and in the cytoplasm. These vesicles, which are the site of replication of PEMV RNA-I, originate within the nuclear membrane and are then extruded into the cytoplasm in clusters enclosed by a second membrane.

by the detection of minus-strand viral RNA by in situ hybridization with 1251_ labeled PEMV virion RNA (de Zoeten et a1., 1976). Also, PEMV-induced RNA-polymerase activity was detected in subcellular fractions enriched with either nuclei or the infection-induced vesicles (Powell, 1977). Furthermore, polymerase activity could be stimulated in isolated nuclei from healthy peas by adding PEMV RNA, but not by adding PEMV virions or by adding particles or RNA of other plant viruses (Powell and de Zoeten, 1977). The induction of polymerase activity was AMD-resistant and required all four nucleotides. Judging by hybridization and digestion experiments, the polymerase synthesized predominantly double-stranded RNA (dsRNA) and some positive-sense ssRNA. No PEMV virions were produced. Thus, PEMV RNA must be recognized specifically by the isolated nuclei.


Although PEMV is the only plant virus that has particles that contain plus-sense RNA and is known to replicate in the nucleus, similar vesicles are induced by some luteoviruses (Esau and Hoefert, 1972; Gill and Chong, 1976; Shepardson et a1., 1980). There are strong affinities between PEMV and luteoviruses, whose replication may likewise involve the nucleus. In addition, the vesicular structures associated with PEMV infection are often observed near or in plasmodesmata (de Zoeten and Gaard, 1983) and are abundant in phloem tissue and sieve elements, suggesting they may playa role in cell-to-cell or long-distance movement of the virus in plants.

Two other types of cytopathological structure are associated with PEMV infection (de Zoeten and Gaard, 1983). One is an electron-dense daggerlike tubular structure associated with plasmodesmata. The other is an elongated featherlike crystalline inclusion that occurs in epidermal cells, is associated with the perinuclear membrane or with the PEMV-specific vesicles, and is heavily studded with ribosomes. The functions of these structures are unknown.

III. PARTICLE COMPOSITION AND PROPERTIES

A. Physicochemical Properties of the Particles

Techniques for the purification of PEMV have been described by many workers (e.g., Bozarth and Chow, 1966; Gibbs et a1., 1966; Izadpanah and Shepherd, 1966b; Shepherd et a1., 1968; Gonsalves and Shepherd, 1972; Vovlas and Rana, 1972; Farro, 1973; French et a1., 1973; Hull and Lane, 1973; Mahmood and Peters, 1973; German and de Zoeten, 1975). Purified preparations of PEMV virions contain two main nucleoprotein components (Fig. 2), whose physicochemical properties (Hull, 1977b, 1981; Peters, 1982) are summarized in Table I. Different authors call the two components either middle (M) and bottom (B), or top (T) and bottom (B), based on their sedimentation rates. In the absence of any evidence for a nucleic acid-free third component, we prefer the terms T and B and will use them in this review. The amount of T component can vary considerably between isolates and can range from being nearly undetectable, as in the right panel of Fig. 2, to being the dominant nucleoprotein component (see, e.g., Hull and Lane, 1973).

Particles in purified preparations are isometric; estimated diameters range from 22 to 30 nm (Fig. 3) (Bozarth and Chow, 1966; Gibbs et a1., 1966; Izadpanah and Shepherd, 1966b; Musil et a1., 1970; Farro and Rassel, 1971; Vovlas and Rana, 1972; Hull and Lane, 1973). Although most authors report the same diameter for T and B particles, German and de Zoeten (1975) gave different values: 25 nm (T) and 28 nm (B). Moreover, there is other evidence that T and B particles are structurally dissimilar: B particles have hexagonal outlines, whereas T particles are irregular and ovoid in appearance (Fig. 3) (Gibbs et a1., 1966; Peters, 1982). The observed hexamer-pentamer clustering


B B

T T

A

10% 40% 10% 40%

FIGURE 2. Ultraviolet absorption traces obtained by centrifugation of purified PEMV particles (l-mg aliquots) through 10-40% sucrose gradients in 0.1 M sodium acetate buffer at pH 6.0. The tubes were centrifuged for 3 hr at 25,000 rpm in a Beckman SW28 rotor. Left trace, isolate AT- /)0;

right trace, isolate WSG. M, meniscus; T, T component; B, B component; A, virion aggregates. Note the difference in the relative amounts of T and B components in the two isolates.

TABLE I. Physical Properties of PEMV Virions

Property

Extinction coefficient at 260 nm (1 mg/ml, 1 cm lightpath)

Isoelectric point (pH)

Sedimentation coefficient, 820 w

(Svedbergs) , Density (g/cm3 )

Diffusion coefficient (cm2/sec x 107)

Virion mol. wt. (x 10-6)

Valuesa

7.2,1 7.52 (mixtures of T and B)

6-7,5-63

5-64

5.9,7.31 91-106 (T), 107-122 (B)5

In D20-sucrose: 1.36 (T and B)6 In Cs2S04: 1.38 (T and B)7 In CsCI: T degraded, 1.42-1.43 (B)6 By sedimentation: 1.89 (T and B)8 By electrophoresis: 1.76 (T); 1.55 (B)6 By sedimentation: 4.4 (T); 5.6 (B)6 From sequence data: 4.6-5.2 (T); 5.7 (B)9

aSuperscript numbers represent references as follows: 1. Hull (1977b); 2. Shepherd 11970); 3. Adam (1976); 4. Shepherd et al.I1968); 5. Hull /1981); 6. Hull and Lane 11973); 7. Hull (1976); 8. Bozarth and Chow 11966); 9. Demler and de Zoeten 11991), Demler et al. 11993).


FIGURE 3. Electron micrograph of purified PEMV particles. Note that although some particles have hexagonal outlines, others (corresponding to T component) are less regular or ovoid in appearance. Negatively stained with uranyl acetate. Magnification bar, 100 nm.

of protein subunits in B particles agrees well with the calculated number of 180 protein subunits per particle (Hull and Lane, 1973), assuming that the particles have T = 3 icosahedral symmetry (Caspar and Klug, 1962). In contrast, Hull and Lane (1973) calculated that T particles contain only 140-150 subunits, which would not allow construction of a protein shell according to the usual rules of quasiequivalence. However, a model was proposed that allows icosahedral symmetry with only 150 subunits when subunits are shared between adjacent hexamers (Hull and Lane, 1973j Hull, 1977b). This model can explain the differences in size and stability (see below) of T and B particles, but it lacks experimental confirmation. Nevertheless, the two kinds of particle clearly have different geometry.

T particles differ from B particles not only in structure but also in being more labile when exposed to differing regimens of salt concentration or pH (Bozarth et a1., 1969j Hull, 1977b). This differential stability, together with the absence of empty shells, suggests that PEMV particles are stabilized primarily by RNA-protein interactions (Hull, 1976). Hull (1977b) concluded that, at acidic pH, B particles are also stabilized by protein-protein inter-


PEMV RNA I 5705 nuc1eotides

84K FS 54K RT (AT)

34K t (Pro)

2IK(Cp) 33K

67K (Polymerase) 1 I

(a)

BWYV (Luteovirus) 5641 nucJeotides 74K RT (AT)

I 29K

(Pro)

+ 67K (Polymerase) 22K (Cp) 52K

I I I I I

66K FS 19K

PEMV RNA 2 4253 nucleotides

27K 33K FS

26K L ___ .

65K (Polymerase) 15K

(b)

RCNMV RNA 1 (Dianthovirus) 3889 nucleotides

FS 57K (polymerase)

27K 37K (Cp)

FIGURE 4. (a) Comparison of the genomic organizations of PEMV RNA-l and the luteovirus, BWYV. Shaded areas indicate regions of statistically significant amino acid sequence homology of the gene products, namely the putative protease core (Pro), the viral RNA-dependent RNA polymerase, the coat protein (Cp), and the 33-kDa readthrough domain involved in aphid transmission (AT). Note in PEMV RNA-l the lack of a 19-kDa ORF within the coat protein gene and the shorter 3' -terminal ORF. FS, regions involved in translational frameshifting; RT, regions involved in translational readthrough. (b) Comparison of the genomic organizations of RNA-2 of PEMV and RNA-l of red clover necrotic mosaic dianthovirus (RCNMV). Shaded areas depict regions of statistically significant amino acid sequence homology in the viral RNA-dependent RNA polymerases. Similar relationships exist with the polymerases of viruses in other genera: Tombusvirus, Carmovirus, NecIOvirus, Umbravirus, and Luteovirus (barley yellow dwarf-PAY type). FS, regions involved in translational frameshifting; Cp, coat protein.


actions that are lacking in T particles. The fact that PEMV coat protein can be assembled into particles with different structures may be important in enabling RNA molecules of different sizes to be encapsidated (see Section III.D). Such flexibility in requirements for packaging may allow the heterologous encapsidation of unrelated RNA species, and thereby foster the longterm association and coevolution of dissimilar viral RNA species.

B. Coat Protein

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis of preparations of PEMV particle proteins reveals a major component of 21-22 kDa (Hill and Shepherd, 1972; Hull and Lane, 1973). This corresponds closely to the estimate based on amino acid composition (199 amino acids, 21,800 Da) (Shepherd et a1., 1968). The amino acid sequence of the N terminus of this protein corresponds to the sequence predicted from the nucleotide sequence of the penultimate reading frame of PEMV RNA-l (Fig. 4a) (Demler and de Zoeten, 1991). This placement of the coat protein gene on RNA-l confirms the conclusions derived from pseudorecombination studies by Hull and Lane (1973). As with the coat proteins of other viruses

BMV PEMV

-RNA 1 -RNA 2

-RNA 3

(a)

PEMV -3 +3

(b)

PEMV -3 +3

-RNA 1 -RNA 2

·RNA3

FIGURE 5. Analysis of PEMV RNA by gel electrophoresis. (a) Separation in nondenaturing conditions: right track, PEMV RNA-I, RNA-2, and RNA-3; left track, brome mosaic virus (BMV) RNA, showing (top to bottom) RNA-I, RNA-2, RNA-3, and RNA-4. (b) Two left-hand tracks, gel electrophoresis of PEMV RNAs derived from isolates containing (+3) or freed from (-3) the PEMV satellite RNA. Two right-hand tracks, autoradiogram of same gel after exposure to a 32p_

labeled probe specific to the satellite RNA. Note the lack of homology to the two genomic RNA species.


with icosahedral particles, the PEMV protein is strongly basic at its N-terminus, with 17 of the first 42 residues being arginine or lysine. Some PEMV isolates possess, in addition to the 21-kDa protein, small amounts of a 54-kDa protein.

C. Nucleic Acid

PEMV virions contain two or three RNA species with sedimentation coefficients estimated at 34, 3D, and 12 S for RNA-I, RNA-2, and RNA-3, respectively (Fig. 5) (Gonsalves and Shepherd, 1972). Based on nucleotide sequence analysis, their molecular weights are 1.9 x 106 (5705 nucleotides) for RNA-I, 1.4 x 106 (4253 nucleotides) for RNA-2, and 2.4 x 105 (717 nucleotides) for RNA-3 (Demler and de Zoeten, 1991; Demler et a1., 1993, 1994b), values that are in close agreement with previous estimates (Gonsalves and Shepherd, 1972; Farro, 1973; Hull and Lane, 1973; German and de Zoeten, 1975; Adam et a1., 1979; Gabriel and de Zoeten, 1984).

The PEMV RNA species are not polyadenylated and cannot be aminoacylated at their 3' termini (German et a1., 1978). No m7G RNA cap structure has been detected (German, 1974, and unpublished results), although other types of cap structure cannot be discounted. However, a covalently linked 17.5-kDa protein was detected in preparations of PEMV RNA by iodination (Reisman and de Zoeten, 1982), suggesting the presence of a VPg, comparable to that of the comoviruses and nepoviruses (see Chapters 3 and 6, this volume), instead of a cap. As described in Section IY.C.l, there is a degree of replicative autonomy between the two genomic RNA species of PEMV, and it cannot be assumed that all RNA species encapsidated by PEMV coat protein possess a VPg.

D. RNA Species Composing the Genome and Their Association with the Sedimenting Components

The older literature contains contradictions on the number of RNA species composing the PEMV genome, their individual infectivity, and the nucleoprotein components in which they are encapsidated. Several factors have contributed to these ambiguities. The experimental approach most often used by earlier workers was to purify the two nucleoprotein components by repeated sucrose gradient ultracentrifugation and then to determine the infectivity of the separated and remixed components. Gibbs et a1. (1966), Izadpanah and Shepherd (1966b), Gonsalves and Shepherd (1972), and Mahmood and Peters (1973) concluded that each component is infectious alone, whereas Bozarth and Chow (1966) concluded that only the B particles are infectious. Hull and Lane (1973) and Adam et a1. (1979), using electrophoretic


fractionation of the two nucleoprotein components, concluded that both components are necessary for infection.

Hull and Lane (1973), Hull (1977bL and Adam et al. (1979) showed that it is impossible to obtain sufficiently pure components by repeated sucrose gradient centrifugation because of the small differences in S value between the T and B components. Although preparations of either component may appear homogeneous in sucrose gradients, they invariably contain the other component, as shown by electrophoresis or Northern blot analysis (S. Demler, unpublished results). The presence in some cultures of a deletion variant of RNA-l (see Sections Y.D. and Y.E.) may have diminished even further the differences in sedimentation rate between T and B components. In addition, PEMV particles tend to aggregate, producing faster sedimenting material (Fig. 2), which was shown by Adam (1976) to contain both T and B components. Although Hull and Lane (1973) and Adam et a1. (1979) obtained better separation by electrophoretic fractionation of the two nucleoprotein components, neither approach gave complete separation.

Attempts to separate RNA-l and RNA-2 by electrophoresis of RNA preparations also gave conflicting results. Both Hull and Lane (1973) and Gonsalves and Shepherd (1972) found that this method failed to separate the two RNA species completely. However, whereas Hull and Lane (1973) found that their preparations of RNA-l and RNA-2 were more infective when mixed than when separate, Gonsalves and Shepherd (1972) concluded that RNA-2 was infectious and that RNA-l was not.

The variable occurrence of RNA-3 in PEMV cultures seems to have further complicated attempts to assign the three different RNA species to the two nucleoprotein components. The occurrence of RNA-3 in virus particles presumably also interfered with attempts to determine the role of RNA-l and RNA-2 in infection. A close association between RNA-2 and T component was consistently found (Gonsalves and Shepherd, 1972; Hull and Lane, 1973; Adam et a1., 1979; Gabriel and de Zoeten, 1984L but reports on the RNA composition of B component were more variable. It is not clear in many of these studies whether RNA-3 occurred, and if so in what amounts, in the isolates examined. However, Hull and Lane (1973) and Hull (1976L using an isolate containing only RNA-l and RNA-2 with minimal amounts of RNA-3, reported the association of RNA-l with B component. In contrast, working with an isolate that contained significant amounts of RNA-3, Gonsalves and Shepherd (1972) suggested that two kinds of B-component particle exist, one containing RNA-l alone and the other containing both RNA-2 and RNA-3. This result was essentially confirmed by Adam et a1. (1979). Thus the data are consistent with a model in which T component consists of particles containing RNA-2, and B component can consist of two types of particle, one type containing RNA-l and the other (present only in isolates in which RNA-3 occurs) containing both RNA-2 and RNA-3. Only since the advent of molecular approaches has it been possible to determine the nature of RNA-I, RNA-2, and RNA-3 and the subtlety of their interactions.

314 s. A. DEMLER ET AL.


A. Molecular Organization of RNA-l

RNA-l (5705 nucleotides) contains five open reading frames (ORFs) (Fig. 4a) (Demler and de Zoeten, 1991). The expression of genetic information in this RNA relies on the use of ORF overlap and nonconventional translation strategies. In vitro translation studies indicate that RNA-l yields two main products of about 36 and 88 kDa, suggesting that the predicted products of the 34-kDa and the overlapping (out of frame) 84-kDa ORFs are produced by independent initiations of translation near the 5'-terminus (Gabriel and de Zoeten, 1984). In contrast, the internally encoded proteins corresponding to the 67-, 21-, and 33-kDa ORFs do not appear to be translated directly from RNA-I. Translation of the 84-kDa ORF was hypothesized to continue, by a frameshift, into the 67-kDa ORF, to yield a 147-kDa fusion protein. Both the 21- and 33-kDa ORFs are presumed to be translated from a single 1800-nucleotide subgenomic messenger RNA (see Fig. 8b), with the 33-kDa ORF being expressed by readthrough of the 21-kDa ORF to give a 54-kDa product.

The functions of most of the RNA-I-encoded proteins can be inferred by comparison with the proteins of other viruses. Thus the 67-kDa part of the 147-kDa frameshift product contains a series of sequence motifs that implicate this region as the heart of the viral RNA-dependent RNA polymerase (Habili and Symons, 1989; Koonin, 1991; Koonin and Dolja, 1993). The N-terminal 300 amino acids of this frameshift product (encoded by the 84-kDa ORF) constitute a predominantly hydrophobic region, which contains a series of domains indicative of a transmembrane protein. This might indicate insertion of the viral polymerase into the endoplasmic reticulum (ER) and/or nuclear envelope, which would be consistent with the evidence from ultrastructural and biochemical studies of PEMV replication. The 84-kDa region also contains a chymotrypsinlike motif (Demler and de Zoeten, 1991; Koonin and Dolja, 1993). Luteovirus genomic RNA molecules, too, have a protease sequence immediately upstream of the polymerase sequence. By analogy with picornalike viruses, the putative VPg of PEMV RNA-l may be encoded in the N-terminal region upstream of the protease.

The structural proteins of PEMV are encoded by the 21- and 33-kDa ORFs. The 21-kDa ORF encodes the main coat protein, as indicated by amino acid analysis, serological studies, and direct sequencing. As described in Section V.D., the 33-kDa readthrough domain is implicated in the aphid transmission of PEMV.

The genomic organization and sequence of PEMV RNA-l have a close but imperfect similarity to those of the luteovirus subgroup containing beet western yellows virus (BWYV) and potato leafroll virus (PLRV) (Fig. 4a) (Veidt et al., 1988; Mayo et al., 1989; Demler and de Zoeten, 1991; Vincent et al., 1991). The shaded areas in Fig. 4a represent regions of statistically significant amino acid sequence similarity to luteoviruses: the RNA-dependent RNA


polymerase, the putative protease within the 84-kDa ORF, the viral coat protein, and the coat protein readthrough domain. However, there are two key differences between PEMV RNA-1 and luteovirus RNA: (1) PEMV RNA-1 lacks, within the coat protein ORF, a 17- to 19-kDa ORF that is present in all luteoviruses examined so far and is suggested to represent a cell-to-cell movement protein (Tacke et a1., 1991, 1993); and (2) the coat protein readthrough domain of PEMV RNA-1 is 10 to 23 kDa smaller than that of luteoviruses. However, the comparable luteovirus domain undergoes proteolytic processing in vivo, which removes C-terminal fragments, leaving a portion of similar size to the PEMV domain (Bahner et a1., 1990; Martin et a1., 1990; Cheng et a1., 1994). Perhaps this processing event is unnecessary (or has been circumvented) in PEMV.

B. Molecular Organization of RNA-2

In all multipartite RNA genomes described to date, there are regions of homology at the termini of the different RNA species, which are thought to be critical for recognition of the viral template at the initiation of viral replication. In contrast, a striking feature of the PEMV RNA-2 sequence is the absence of homology to the 3'- and 5'-termini of PEMV RNA-l (Demler et a1., 1993). This was one of the first pieces of evidence that PEMV does not have a conventional multipartite genome.

PEMV RNA-2 (4253 nucleotides) contains four ORFs, each of which overlaps with another (Fig. 4b) (Demler et a1., 1993). In vitro translation of RNA-2 yields a prominent 42- to 45-kDa product, though a product of this size is not predicted by the nucleotide sequence. Using a series of in vitro transcripts of the 5' region of RNA-2, Demler et a1. (1993) showed that this product is encoded by the 33-kDa ORF. The cause of its aberrant electrophoretic mobility is unknown. In vitro translation of the 33-kDa ORF can continue, by a frameshift, into the 65-kDa ORF to produce a 94-kDa fusion protein. The overlap region of the 33- and 65-kDa ORFs contains a sequence resembling the frameshift signal described for barley yellow dwarf luteovirus (BYDV) and dianthoviruses (Brault and Miller, 1992; Miller et a1., 1988; Xiong and Lommel, 1989; Xiong et a1., 1993). No evidence was found for translation of the 26- or 27-kDa ORFs directly from RNA-2. In addition, several cDNA clones to PEMV RNA-2 contain a 3'-terminaI15-kDa ORF that lacks a stop codon (dashed box in Fig. 4b). The sequence of this region of RNA-2 is variable, and Demler et al. (1994a) showed that this ORF is not essential for PEMV infection.

Sequence comparisons have not enabled a function to be inferred for the product of the 33-kDa ORF, but remarkably the 65-kDa portion of the frameshift product contains sequence motifs typical of viral RNA-dependent RNA polymerases. This putative polymerase is distinct from that encoded by RNA-1 but is related to those of viruses in the genera Carmovirus, Tombus-


virus, Necrovirus, Dianthovirus, Umbravirus, and the Luteovirus BYDVPAY subgroup (Fig. 4b). Thus the polymerases of RNA-1 and RNA-2 of PEMV have affinities with those of viruses in different supergroups (Habili and Symons, 1989; Koonin, 1991; Koonin and Dolja, 1993).

The 3' location in PEMV RNA-2 of the overlapping 26- and 27 -kDa ORFs suggests that they are expressed from subgenomic messenger RNA. Northern blot analysis of infected tissue detected an RNA product of the predicted size, although its concentration in infected tissue was considerably lower than that of its counterpart derived from RNA-I. The amino acid sequence of the 27-kDa protein is suggestive of a cell-to-cell movement protein (Mushegian and Koonin, 1993). This idea is supported by the observation that nucleotide changes in the 26-kDa/27-kDa region affected the systemic invasion of pea plants (Demler et al., 1993, 1994a). However, because these two ORFs overlap, either or both of the encoded proteins may be responsible for this change in behavior. No evidence was found that RNA-2 encodes a coat protein (Demler et a1., 1993).

C. The PEMV Paradox: Bipartite Genome or Mixed Infection?

The occurrence of polymerase genes in both RNA-1 and RNA-2 raises the question whether PEMV has a divided genome or is a complex of two viruses. To circumvent the problems experienced in obtaining purified RNA-1 and RNA-2 by fractionating mixtures, infectious in vitro transcripts of RNA-1 and RNA-2 were prepared to allow reassessment of the independent capabilities of each RNA (Demler et a1., 1993, 1994a). Full-length eDNA clones of each RNA were assembled and modified using the polymerase chain reaction to incorporate a T7 RNA polymerase promoter element at the 5' terminus and a unique restriction site at the 3' terminus. Linearization with the appropriate restriction enzyme, followed by transcription with T7 RNA polymerase, generated full-length RNA molecules containing native 5' and 3' termini. These transcripts were then inoculated both to pea protoplasts and to whole plants to show which RNA was responsible for different biological functions and whether there are any shared functions that make it necessary for these two RNA species to remain associated in nature. The following sections summarize current understanding of the PEMV genome, and of the roles of RNA-l and RNA-2.

1. Replication.

The replicative competence of in vitro transcripts of PEMV RNA-1 and RNA-2 was assessed by inoculation of pea protoplasts (Demler et al., 1993, 1994a). Northern blot analyses showed that each RNA was capable of directing its own replication. Thus, the two polymerases identified by sequence analysis are each able to replicate their homologous RNA. Electron micros-


copy of infected protoplasts showed that RNA-l induced formation of the double-membrane-bounded replication complex. This is consistent with observations on three luteoviruses that have strong affinities with PEMV RNA-I: PLRV, BWYV, and the NY-RPV isolate of BYDV (Esau and Hoefert, 1972; Gill and Chong, 1976; Shepardson et a1., 1980). Protoplasts infected with PEMV RNA-2 exhibited very different cytopathological effects, characterized by proliferation and separation of the ER, and production of singlemembraned vesicles. It is not known whether these RNA-2-specific structures are involved in RNA-2 replication or whether RNA-l and RNA-2 in mixed infection replicate or accumulate at the same site. Protoplasts infected with mixtures of the in vitro transcripts of PEMV RNA-l and RNA-2 (or with RNA-l + RNA-2 derived from virions) were difficult to distinguish from those infected with RNA-l alone; the formation of the RNA-I-specific replication complex and its effects on ER and nuclear membrane appear to dominate the more subtle effects induced by RNA-2.

2. Particle Proteins and Assembly.

RNA-l is solely responsible for the expression of the structural proteins of PEMV (Demler et a1., 1993, 1994a). Northern blot analysis showed that RNA-I-infected pro top lasts contained the subgenomic RNA derived from the 3' -terminal region encoding the viral coat protein. Similarly, Western blot analyses of extracts of protoplasts infected with RNA-I, or RNA-l plus RNA-2, showed that viral coat protein was synthesized. Electron microscopy of sections of these protoplasts revealed PEMV-like particles, scattered in the cytoplasm and nucleus and particularly numerous around the nucleolus. More importantly, many RNA-I-infected protoplasts contained large crystalline aggregates of PEMV-like particles, which were found by immunogold labeling and in situ hybridization to contain PEMV coat protein and RNA-l. Thus, RNA-l alone was able to induce coat protein synthesis and virion assembly. Protoplasts infected with RNA-l plus RNA-2 did not contain these large crystalline aggregates, although they did contain large numbers of virions. Perhaps irregularities in the structure of the T particles prevent the crystals forming.

Protoplasts infected with RNA-2 alone did not contain PEMV-like particles or coat protein antigen. There is no evidence that PEMV particles contain any RNA-2-encoded proteins or that any RNA-2-encoded protein contains sequences also found in PEMV coat protein.

3. Systemic Invasion.

Although both genomic RNAs can infect protoplasts, their effects on whole plants are completely different. Mechanical inoculation with RNA-2 transcripts leads to systemic invasion at roughly the same speed as does inoculation with transcripts of RNA-l plus RNA-2 (Demler et a1., 1994a,b).


Thus, although some viruses require viral coat protein for systemic movement, RNA-2 is fully capable of establishing a systemic infection in the absence of coat protein. The symptoms expressed and speed of systemic invasion vary, depending on differences between individual infectious clones of RNA-2 and on environmental factors (Demler et a1., 1993, 1994a). The variation is particularly evident with variants that have mutations in the 26-kDa/27-kDa ORFs, a region that encodes the putative movement protein (Mushegian and Koonin, 1993). Most systemic infections of RNA-2 in P. sativum and N. benthamiana are symptomless. Plants display little growth malformation and develop a mild mosaic pattern at the edges of leaves at 2-3 weeks PI. Selected infectious clones induce mild mosaic and downward curling of infected leaves, but these symptoms are largely confined to the inoculated leaves. The ultrastructure of leaf cells infected with RNA-2 resembles that of infected protoplasts and includes ER proliferation and extensive production of single-membraned vesicles in the cytoplasm.

In contrast, mechanical inoculation of pea seedlings with RNA-l transcripts failed to establish systemic infection, as assessed by Northern blot analysis. However, inoculation of plants with RNA-2 plus RNA-l induced systemic invasion by both RNA species and typical symptoms of PEMV. In view of the luteoviruslike nature of RNA-I, its inability to establish a systemic infection in mechanically inoculated plants is not surprising, because luteoviruses are strictly phloem-limited and require an aphid vector to inoculate them into this tissue. However, preliminary experiments (S. A. Demler, unpublished data) suggest that RNA-l cannot induce systemic infection even when introduced by aphids. Aphids were allowed to feed through membranes on virion preparations purified from protoplasts infected with an RNA-l transcript (derived from an aphid-transmissible isolate) and then transferred to healthy seedlings. Of 20 plants inoculated, none was found to contain RNA-l when assayed by a Northern blot analysis that could detect one thousandth of the amount of RNA-l present in normal PEMV infections. In control tests in which the aphids were allowed to feed on virion preparations purified from protoplasts infected with an RNA-2 transcript as well as with the above RNA-l transcript, all the aphid-inoculated plants became infected systemically with RNA-l as well as with RNA-2.

D. Analogy with Luteovirus Helper-Dependent Virus Complexes

Current evidence suggests that PEMV is neither a conventional mixed infection nor a virus with a bipartite genome. The infectivity data suggest that the interaction of RNA-l and RNA-2 can best be described as a form of symbiosis between two defective viruses. RNA-l appears to be a defective kind of luteovirus RNA. It is able to replicate, form particles, and confer aphid transmissibility, but it fails to cause systemic infection not only when inoculated mechanically (as happens with luteoviruses) but also, in prelimi-


nary tests, when introduced by aphids. Perhaps this is because it lacks the 17-to 19-kDa ORF that in luteoviruses is suggested to represent a cell-to-cell movement protein. Another factor that could playa part is the small size of the coat protein readthrough extension compared with those found in luteoviruses. In contrast, RNA-2 can both replicate and cause systemic infection but does not encode a coat protein. The combination of RNA-l and RNA-2, however, has all these properties and forms a tightly linked infectious entity. RNA-l provides the coat protein and aphid transmissibility for both viruses, whereas RNA-2 directs their systemic invasion. These are the functions that provide the cohesive force to the symbiosis and give it characteristics more indicative of a virus with a bipartite genome. However, each RNA directs its own replication, a situation akin to a mixed infection. Other kinds of interaction between RNA-l and RNA-2 may remain to be discovered.

Although apparently unique among plant viruses, the PEMV system has some similarities to helper-dependent complexes involving members of the Luteovirus (BWYV subgroup) and Umbravirus genera (Falk and Duffus, 1981; Murant, 1993). In these complexes the luteovirus component is the helper for aphid transmissibility of the umbravirus, which is the dependent component. The umbravirus is an autonomously replicating RNA that is able in most instances to infect plants systemically but lacks a coat protein gene and so is unable to form particles that can be transmitted by aphids. The function of the luteovirus in these complexes is the same as that of RNA-l in the PEMV system, namely, to provide a coat protein for encapsidation of the RNA of the dependent component and so to confer aphid transmissibility. The role of the dependent RNA in these complexes is variable, but it may enhance the titer of the luteovirus, intensify its symptoms, alter its host range, and, in at least one instance (Falk et a1., 1979), confer limited mechanical transmissibility. In addition, in the two dependent viruses for which RNA sequence data have been obtained (BWYV-ST9-associated RNA and an Australian form of carrot mottle virus), there is strong sequence homology between the polymerase ORF and that of PEMV RNA-2 (Chin et a1., 1993; Murant et a1., 1995). However, unlike PEMV RNA-I, the luteovirus helper components of these complexes can each establish independent systemic infections. Furthermore, the luteovirus helper components remain essentially phloem-limited in mixed infections, whereas PEMV RNA-2 enables PEMV RNA-l to invade other tissues. Thus, PEMV may be an example of one of these complexes that has evolved into a more stable and somewhat more mutually dependent association of two unrelated RNA species.

It is noteworthy that PEMV also participates in a helper-dependent complex with bean yellow vein banding virus (BYVBV) (Cockbain et a1., 1986). In this complex, PEMV RNA-l probably provides the coat protein that encapsidates and confers aphid transmissibility on BYVBV RNA, which is presumably distinct from PEMV RNA-2 but,like it, lacks a coat protein gene. Bean leafrolliuteovirus too can serve as the transmission helper for BYVBV.


It is unknown whether PEMV RNA-2 and BYVBV RNA share nucleotide sequences.

E. RNA-3

As mentioned above, some PEMVisolates possess a third RNA (RNA-3) (Fig. 5), although the frequency of its occurrence in nature is unknown. Initially it was suggested that RNA-3 is an artifact, derived by cleavage of RNA-1 molecules to produce two fragments with the approximate sizes of RNA-2 and RNA-3 (German and de Zoeten, 1975; Adam et a1., 1979). RNA-3 did not seem to be needed for PEMV replication and did not affect symptoms in pea (Gonsalves and Shepherd, 1972; Hull and Lane, 1973).

A clue to the nature of RNA-3 came from studies of dsRNA in extracts of PEMV-infected pea tissue (Demler and de Zoeten, 1989). In a series of passages of PEMV isolate WSG by mechanical inoculation, increasing amounts of a dsRNA of Mr 0.7 X 106 were found, and a ssRNA species of Mr 0.3 X 106

began to be detected in preparations of virion RNA. Further work (Demler and de Zoeten, 1989; Demler et a1., 1994b) showed that RNA-3 has the properties of a satellite RNA: it occurs in some PEMV isolates but not others, and it is noninfectious on its own, but it replicates when inoculated together with PEMV RNA-2. No sequence similarity to PEMV RNA-lor RNA-2 was detected by Northern blot analysis. RNA-3 did not appear to affect the replication of PEMV RNA-lor RNA-2 and it had no effect on PEMV symptoms in pea, though it slightly decreased symptom severity in Nicotiana benthamiana (Demler et a1., 1994b).

Sequence analysis of RNA-3 showed that its ends have limited homology to those of RNA-2: 12 of the first 14 nucleotides at the 5' terminus and 7 of the final 8 nucleotides at the 3' terminus. No other homology was found with RNA-1 orRNA-2. A 27-nucleotide repeat of unknown function was also found. RNA-3 contains a number of small ORFs, but no corresponding products were found in in vitro translation tests. By using full-length infectious transcripts of RNA-3, its replication was shown to be totally dependent on RNA-2, which also mediates its systemic spread in plants (Demler et a1., 1994b). However, the encapsidation and aphid transmissibility of RNA-3 are controlled by RNA-I. Thus, RNA-3 is dependent on PEMV RNA-l for some functions and on PEMV RNA-2 for others.

Although satellites typically fail to confer any advantage on their helper viruses, RNA-3 may have an effect on particle stability. As already mentioned, B-component particles are more stable in some circumstances than T-component particles, and it seems possible that this applies not only to B-component particles containing RNA-l but also to those containing RNA-2 together with RNA-3.

The simultaneous dependence of a satellite on two viruslike agents has a precedent in the viruses associated with groundnut rosette disease, a


luteovirus-dependent complex (Reddy et a1., 1985a,b; Murant et a1., 1988; Murant, 1990; Murant and Kumar, 1990; Kumar et a1., 1991; Blok et a1., 1994). The groundnut rosette complex has three elements: (1) groundnut rosette assistor virus (GRAVL a typicalluteovirus related to PLRV; (2) groundnut rosette virus (GRV), an umbravirus; and (3) a satellite RNA. The satellite RNA depends on GRV for its multiplication in plants, and both the satellite RNA and GRV depend on GRAV for transmission by aphids. The satellite RNA is responsible for the characteristic symptoms of the disease, and it is also essential in mediating the GRA V-dependent aphid transmission of GRY. In RNA reassortment tests (de Zoeten et a1., 1994L GRV and PEMV RNA-2 each proved able to support the replication of the heterologous satellite RNA, and the GRV satellite was encapsidated in PEMV coat protein. In addition, the typical symptoms induced by the "yellow blotch mosaic" variant of the GRV satellite (Kumar et a1., 1991) were evident in co-infections with PEMV RNA-2, with or without PEMV RNA-I. These data increase the likelihood that PEMV is a virus complex evolutionarily related to these luteovirusdependent complexes.

F. RNA Replication and Encapsidation

Replication of PEMV proceeds via linear dsRNA complexes; replicative form (RF) and replicative intermediate (RI) RNA species have been isolated from infected plants. Three dsRNA species with Mr approximately twice that of each of the ssRNA species from purified virus particles have been identified (German and de Zoeten, 1975; Demler and de Zoeten, 1989, and unpublished results). The relative amounts of the three RF species in infected tissue differ from the relative amounts of the three ssRNA species in virions. In particular, in isolates containing only RNA-1 and RNA-2, dsRNA-2 is more abundant than dsRNA-1, although RNA-1 predominates over RNA-2 in virions.

V. APHID TRANSMISSION

A. Occurrence of Isolates That Are Transmissible or Nontransmissible by Aphids

One of the attractions of studying PEMV is the existence of isolates that differ in aphid transmissibility. Aphid nontransmissible (AT-) isolates have not been reported to occur independently in nature, but they commonly arise when aphid-transmissible (AT+) isolates are propagated serially by mechanical inoculation (French et a1., 1973; Tsai and Bath, 1974; Harris et a1., 1975). Tsai and Bath (1974) reported the elimination of aphid transmissibility after only three to four passages, whereas Adam et a1. (1979) required 11 passages.


Some cultures, such as the Wisconsin and P3 isolates, appear to be more stable genetically (Harris, 1979). For example, the Wisconsin isolate was mechanically passaged more than 20 times with no loss of aphid transmissibility. There are no reports of reversion of an AT- isolate to an AT+ phenotype. Obviously, to ensure that aphid transmissibility is not lost, serial mechanical transfers should be avoided. AT+ isolates are best maintained by serial aphid transmission with either Acyrthosiphum pisum or Myzus persicae. Occasional passage through an appropriate local lesion host followed by screening for aphid transmissibility may also be warranted.

B. Electrophoresis of Virions

Preparations of intact virions of AT+ and AT- isolates are largely indistinguishable from each other by analytical centrifugation or preparative sucrose gradient centrifugation, both producing T and B components. However, electrophoretic analyses provided the first evidence that linked aphid transmission of PEMV with a specific component of the capsid surface (Hull, 1977a,b; Adam et a1., 1979; Gabriel, 1983). When particles of AT- isolates were electrophoresed in low-percentage acrylamide gels (Fig. 6a), they separated into two discrete bands, corresponding to the T and B centrifugal components. The relative mobility of the two bands was affected by gel strength (pore size) but not by pH. This led Hull and Lane (1973) to conclude that separation of the two components depends on size, not charge, differ-

E c:

...... Ltl N

9

a b c

- e FIGURE 6. Densitometer scans of 3.4% polyacrylamide gels in which PEMV particle preparations were electrophoresed as described by Hull and Lane (1973). (a) AT- isolate WT (Hull and Lane, 1973). (b) The PEMV "variant" isolate of Hull and Lane (1973). (c) AT+ isolate TIi (Adam, 1978; Adam et a1., 1979). Note the multiple banding pattern of the AT+ isolate and the preponderance of T component relative to the B component in the "variant" isolate.


ences. In contrast, particles of AT+ isolates were remarkably heterogeneous in electrophoretic behavior (Fig. 6c) (Adam et al., 1979; Hull, 1977a; Gabriel, 1983), yielding 8 to 13 electrophoretic bands for B component and 10 to 11 for T component. Hull (1977a,b) and Adam (1978) thought that this multiple banding was caused by differences in particle size: Adam et al. (1979) estimated a diameter increment of 0.54 nm between the particles in successive electrophoretic components. Hull (1977a,b) and Adam et al. (1979) proposed that a stepwise increase in particle size might be caused by the incremental replacement of a fixed number of normal coat protein subunits by larger ones, as observed in variant strains of Q~ bacteriophage (Radloff and Kaesberg, 1973). It was suggested that this size heterogeneity is related to the presence of different numbers of surface projections on the Q~ particles. Bozarth and Chow (1966) reported that particles of AT+ isolates had such projections.

C. Coat Proteins Involved in Aphid Transmission

Differences between PEMV isolates in particle size and electrophoretic behavior reflect differences in particle protein composition. The particles of all isolates have the 21-kDa protein (Fig. 7) (Hull and Lane, 1973; Demler and de Zoeten, 1989). A minor electrophoretic component of 44 kDa (Hull, 1977 a) is probably a dimer of the 21-kDa protein. Hull (1977a) also detected a 28-kDa protein in an AT+ isolate, but this does not seem to have been found by others. Particles of AT+ isolates contain, in addition to the 21-kDa protein, a 54-kDa protein that constitutes about 5% of the total protein (Adam, 1976; Hull, 1977a; Adam et al., 1979; Gabriel, 1983; Demler et al., 1994c). It was therefore suggested that the 54-kDa protein may represent the aphid transmission determinant and may also be responsible for the size heterogeneity evident in the electrophoretic pattern of AT+ virions. In agreement with these observations, the particle proteins of AT+ and AT- isolates crossreacted with homologous and heterologous antisera (Adam, 1976), but antiserum to AT+ isolates contained an additional antibody population that did not react with particles of AT- isolates and led to spur formation in Ouchterlony tests (Clarke and Bath, 1977).

D. Genomic Determinants for Aphid Transmissibility

Western blot analysis (Gabriel, 1983) showed that antiserum directed against an AT- isolate (lacking the 54-kDa protein) nevertheless reacted with both the 21- and 54-kDa proteins of AT+ isolates. This result fits well with the subsequent finding, by sequence analysis, of an in-frame 33-kDa ORF immediately following the 21-kDa coat protein gene, and separated by a single UGA termination codon (Fig. 4a) (Demler and de Zoeten, 1991). A


Igase Igase

... MNASL'GDD~PDPGPQPPPP~PVGARF ••.

21K 33K Cp

54KRT

Igase I I

AT+ AT- Cp 54K

54K

31K

21K

N-Terminal Sequence

MPTRSRS

SPTPVGA

MPTRSRS

FIGURE 7. Demonstration of readthrough of the coat protein gene. Amino acid sequence analysis of the 21- and 54-kDa structural proteins (lane AT+, lower panel) showed that the N terminus of each protein is the same. Digestion of purified 54-kDa protein with the protease Igase (expected to recognize two PPSP tetrapeptides within the proline-rich region; arrows in upper panel) generated two main fragments of 31 and 22 kDa (right lane, lower panel). Analysis showed that the N-terminal amino acid sequence of the 22-kDa protein was the same as that of the 21-kDa coat protein, and that the N -terminal sequence of the 31-kDa fragment corresponded to the sequence at the second predicted proteolytic site.

read through fusion of the 21- and 33-kDa ORFs would give the 54-kDa protein associated with AT+ isolates.

In this respect, PEMV would resemble the luteoviruses, in which readthrough of the 21-kDa coat protein stop codon occurs in vitro and in vivo, and the readthrough domain can be detected serologically in virion preparations (Veidt et ai., 1988; Bahner et ai., 1990; Tacke et ai., 1990; Dinesh-Kumar et ai., 1992; Cheng et ai., 1994). Indeed, comparisons between PEMV RNA-1 and


luteovirus RNA reveal strong amino acid sequence homology in the products of the two 3'-terminal ORFs (the main coat protein and the readthrough domain) (Fig. 4a). However, the situation in PEMV is not identical to that in luteoviruses. For instance, nucleotide sequences surrounding the coat protein stop codons differ. All luteovirus genomic RNA species share the 12-nucleotide sequence AAA UAG GUA GAC in this region, whereas PEMV RNA-l has the sequence CUC UGA GGG GAC which, moreover, includes an opal instead of an amber stop codon. Another difference, already mentioned, is in the size of the readthrough domains: 33 kDa in PEMV RNA-l and about 50 kDa in luteoviruses.

Confirmation that the PEMV 54-kDa protein is the readthrough product has come from structural studies. The 21- and 54-kDa proteins were both found to contain th(! N-terminal sequence MPTRSRS (Fig. 7) (Demler et a1., 1994c). The origin of the C-terminal portion of the 54-kDa subunit was identified by sequencing the peptides produced by treatment with the protease Igase, which recognizes the tetrapeptides PPSP, PPTP, and PPAP, or the pentapeptide PAPSP. The sequence PPSP occurs twice within the prolinerich region of the readthrough domain close to the stop codon for the 21-kDa coat protein. Digestion of the 54-kDa protein with Igase yielded two peptides of 22 and 31 kDa. The 22-kDa peptide contained the N-terminal sequence MPTRSRS, whereas the 31-kDa fragment contained the N-terminal sequence SPTPVGA, which matches the sequence at the second predicted Igase cleavage site.

Initial attempts to relate production of the 54-kDa protein to RNA size and aphid transmissibility did not indicate a straightforward relation between these three properties. All PEMV isolates have RNA-2 of the same size. However, some AT- isolates have an RNA-l with a molecular weight about 0.12 x 106 smaller than normal (Adam et a1., 1979; Gabriel, 1983; Demler and de Zoeten, 1991), though others do not. Recent studies (Demler et a1., 1994c) have provided an explanation of this. All isolates that possess only the small RNA-l are AT-, but isolates having the full-length RNA-I, or RNA of both sizes, may be either AT+ or AT-. Starting with an AT+ isolate that had been propagated solely by aphid transmission, single-lesion isolates were obtained from inoculated leaves of C. quinoa. These isolates were mechanically inoculated to pea seedlings, which were subsequently used as virus sources in aphid transmission experiments. Only 8 of the 20 singlelesion isolates examined were AT+. Of the 12 remaining AT- isolates, eight contained full-length genomic RNA-I, and four contained the truncated RNA-l (Fig. 8). The truncated RNA-l of one isolate examined was found to lack 727 nucleotides in its 33-kDa ORF and therefore could not have produced the 54-kDa readthrough protein. This isolate (designated AT-<l) and AT+ and AT- single-lesion isolates with full-length RNA-l were used for the production of cDNA clones. The clones represented the sequence between the intergenic region immediately upstream from the 21-kDa coat protein ORF and the 3' terminus of RNA-I. These clones were then substituted for

326

RNA 1~-

(a)

AT

RNA 1-

sg 1-

(b)

c. quinoa local lesion isolates I I

AT+ AT- AT- AT- AT+ ~

S. A. DEMLER ET AL.

-RNA 1 -RNA 2

M

-RNA 1~

-sg 1~

FIGURE 8. (a) Denaturing gel electrophoresis of RNA preparations derived from PEMV isolates that differ in aphid transmissibility. The left two lanes represent aphid transmissible (AT+, NMT) and aphid nontransmissible (AT-, WSG) isolates propagated mechanically for 20 successive passages. Note the third "genomic" RNA midway between RNA-I and RNA-2 (designated RNA-Ill). The right three lanes represent single local lesion isolates derived from a culture of the NMT isolate maintained solely by aphid transmission. One of these isolates (AT-A) has the shortened (RNA-l.i) version of RNA-I and is not aphid-transmissible. The other two isolates have full-length RNA-I but only one of them is aphid-transmissible (AT+). (b) Autoradiogram of a Northern blot of total RNA from pea seedlings 7 days after infection with PEMV isolates that differ in aphid transmissibility. The blot was exposed to a probe specific for the coat protein gene in RNA-I. The probe reacted with two species: the genomic RNA-I and a -1800-nt subgenomic RNA (sgl) representing the genes for the coat protein and the readthrough domain. Note the decrease in size of the genomic and subgenomic RNA-I in the two deletion isolates (AT-A) relative to the AT+ and AT- isolates. M, mock-inoculated.

the corresponding region of a full-length AT- isolate transcription vector (designated pPERl) (Demler et a1., 1994a). When co-inoculated with fulllength transcripts of RNA-2 (also produced from an AT- isolate), all three of the RNA-l transcripts elicited wild-type PEMV symptoms and all retained the aphid transmissibility or nontransmissibility of their parent isolate.


Thus aphid transmissibility is determined by the 3' end of RNA-I. Further work, in which different types of chimeric RNA-1 molecule were produced, has localized the sequence that is critical for aphid transmissibility to a stretch of 110 nucleotides near the 3' end of the 33-kDa ORF (Fig. 9). Thus part of the coat protein readthrough domain confers aphid transmissibility. Absence of the readthrough domain, or specific sequence changes within it, abolish aphid transmissibility.

The sequence within the 33-kDa ORF of PEMV RNA-1 seems well suited to produce the 727-nucleotide deletion observed in isolate AT- <l (Fig. 10). The 5' end of the deletion lies within the region encoding a proline-rich N-terminal amino acid sequence, and the 3' end lies in a region encoding a smaller proline-rich sequence at the C-terminus, the junction point occurring immediately after seven consecutive C residues followed by seven consecutive G residues. Because proline is encoded by CCN triplets, both these regions are rich in cytidine. Current models of genetic recombination in viral RNA involve a three-step process (Lai, 1992). First, replication is interrupted

Parental Isolates:

54K RT

21 K 33K

Exchanges:

21K r )1

AT

21K AT~

54KRT

21K 33K AT

21K AT

I !t"Jt:ti1twi*$ I AT +

L....-_ ........... ____ 11 AT

L....-_ ........... __ ...... I ..... · ...... I AT +

FIGURE 9. Mapping of the region of PEMV RNA-l that is critical for aphid transmission. Fulllength cDNA clones were constructed that differed only in the origin of the coat protein gene and 33-kDa ORE These were derived from three single-lesion isolates: the AT-~ deletion isolate (diagonal lines), the aphid-transmissible AT+ isolate (shaded box), and the aphid-nontransmissible isolate AT- with full-length RNA-l (white box). Infectious transcripts of the recombinant isolates were synthesized in vitro and co-inoculated with RNA-2 to pea seedlings. The aphid transmissibility of each of the parental and recombinant isolates is listed to the right of each figure. The results show that a specific llO-nucleotide sequence in the 33-kDa ORF is required for aphid transmission.


5' ... TCCCGGGCCCCAAC CACCACCACCTCCACCCCCAAGTCCCA P G P Q P P P P P P P S PT

4K Prolinerich region

Cp 29K

--'--_----'1 ______ .1-1 3'

~ + -PEMV AT and AT

Full-length

727nt !J.

• Cp

.......------'- 3'

PEMV AT~

GCGGTACCGCCTCCA TTCCCCCCCGGGGGGGTT .. {345nt) .... 3' AV PP P FPPG G V

FIGURE 10. Proposed model for the production of the deletion in RNA-l of isolate AT- a. The 727-nucleotide deletion is flanked by two regions (boxed) that are each rich in cytidine residues and encode predominantly proline residues. The 3' flanking region is thought to form a stable hairpin structure during replication (7 consecutive C residues followed by 7 consecutive G residues). It is proposed that this hairpin interrupts replication and leads to detachment of the replication complex from the template. The C-rich nature of this nascent strand then allows it to reposition at the comparable sequence in the 5' flanking region, so that the replication resumes but with the deletion of the 727-nucleotide region in the 33-kDa ORE

by RNA secondary structure. Second, the replication complex detaches from its template with the nascent strand attached. Third, localized base pairing then repositions the complex and replication resumes. The sequence of PEMV RNA-1 contains all of the elements necessary for these events: a C-G hairpin occurring near the 3' terminus and two cytidine-rich regions capable of repositioning the complex. The frequency of occurrence of this deletion in nature is unknown.

The findings presented above confirm what has long been suspected: that isolates of PEMV contain a mixture of AT+ and AT- forms (Harris et a1., 1975; Hull, 1977a,b; Demler and de Zoeten, 1991). In theory, heterologous encapsidation of mixtures of AT- and AT+ variants could perpetuate the aphid transmission of AT- variants indefinitely. Indeed, heterologous encapsidation was detected in artificial mixed infections of AT- and AT+ isolates (Adam, 1978). However, the frequency with which AT- variants appear upon mechanical transmission suggests that AT- isolates have a selective advantage over AT+ ones. Indeed, AT- variants reach higher concentrations in plants than do AT+ variants, the AT-~ isolate consistently giving three to five times the particle yield of AT+ genotypes (French et a1., 1973; Hull, 1977b; Demler et a1., 1994c). Thus, replicative ability and aphid trans-


missibility constitute counteracting selection pressures, resulting in the maintenance of both AT+ and AT- variants in some AT+ cultures. With serial mechanical transmission, the selection for the AT+ genotypes would be relaxed and the equilibrium would shift toward AT- genotypes, resulting in the eventual elimination of aphid-transmissible forms.

E. Strain Groups

Isolates of PEMV differ little in host range and symptomatology, but differ considerably in aphid transmissibility and in the electrophoretic profile of their virions, properties that have been used to divide PEMV strains into four groups (Table II). Isolates in groups I and II give particle preparations with two homogeneous electrophoretic components. Group II contains a single isolate (Hull and Lane, 1973) in which T particles are the predominant nucleoprotein component. Particles of group IV isolates form multiple electrophoretic bands in polyacrylamide gels. Particles of group III isolates form more than two electrophoretic bands but fewer than are found with group IV isolates. Isolates in groups I and II are exclusively AT-, whereas isolates in groups III and IV are AT+.

It will be evident from the above discussion that the characteristics used to separate strains into these four groups may not be stable genetically. Group I isolates include forms with full-length or partially deleted RNA-I, and the former can give rise to the latter. Group III and IV isolates contain AT- variants that tend to become predominant in cultures maintained by mechanical inoculation (Adam et al., 1979; Demler et al., 1994a). The situation with RNA-2 is equally complex. Demler et al. (1993) found considerable variability among individual clones of RNA-2 in the extent to which they supported systemic spread, differences that were correlated with sequence differences in the 26- to 27-kDa ORF region. S. A. Demler (unpublished

TABLE II. Grouping of PEMV Isolates According to the Electrophoretic Behavior of Their Particlesa

Group Ib Group II

Bozarth and Chow! Hull and Lane "variant"3 Gonsalves and Shepherd2

WT3 WSG4 CNTs Tii (nt)6

Group III Group IV

E1547 Wis9

Hollings8 P32

GemblouxlO

NMT4 CATS Tii6

aFrom Adam (1978), Adam et al. (1979), Hull (1977b), Gabriel (1983). bSuperscript numbers represent references as follows: 1. Bozarth and Chow (1966); 2. Gonsalves and Shepherd (1972); 3. Hull and Lane (1973); 4. Gabriel (1983); 5. Tsai and Bath (1974); 6. Adam (1978), Adam et al. (1979); 7. Mahmood and Peters (1973); 8. Hull (1977b); 9. Izadpanah and Shepherd (1966b); 10. Farro and Rassel (1971).


results) also observed large differences over time in the relative amounts of RNA-l and RNA-2 in particles of supposedly the same isolate. There is little information on the response of RNA-2 to different selection pressures, such as that exerted by serial mechanical transmission. Isolates of PEMV are therefore typical of a quasi-species: a mixture of RNA molecules of variant genotypes (Domingo et a1., 1985). The position is further complicated by the degree of autonomy of each RNA as well as the contribution of each RNA to the RNA-I-RNA-2 interaction. Plainly, care must be taken, when comparing isolates of PEMV, to use appropriate methods for their maintenance and subculture.

VI. VIRUS-VECTOR INTERACTIONS

A. Transmission Characteristics

Several aphid species are reported to transmit PEMV, including Macrosiphum euphorbiae, Myzus persicae, M. ornatus, Acyrthosiphon pisum, A. solani, A. gossypii, and Aulacorthum solani (Osborn, 1935 j Ehrhardt and Schmutterer, 1964 j Schmutterer and Ehrhardt, 1964 j Harris, 1979 j Kennedy et a1., 1962 j Nault, 1967). Nault (1975) demonstrated that three species of aphids for which pea is a nonhost (M. avenae, Rhopalosiphum padi, and Schizaphis graminum) were also capable of transmitting PEMV. From an experimental and practical standpoint, the green peach aphid (M. persicaej, and particularly the pea aphid (A. pisum), are probably the most important vector species (see Chapter 13, this volume).

The aphid transmission of PEMV has been extensively analyzed, and the available data suggest that transmission of PEMV is circulative nonpropagative. Circulative transmission implies that following ingestion via the maxillary food canal the virus passes from the gut into the hemocoel, and from there into the salivary system, from which it is inoculated into plants in the form of virus-laden saliva ejected from the maxillary salivary canal.

The dynamics of aphid transmission of PEMV differ considerably from those of luteoviruses. The ability of PEMV to establish a true systemic infection, and its relatively high titer in infected tissues, may be responsible for its short acquisition and inoculation thresholds, which are unusually brief for a circulative virus. Estimated minimum acquisition periods for A. pisum range from 15 min for nymphs to 1-2 hr for adults (Simons, 1954 j Bath and Chapman, 1964 j Nault and Gyrisco, 1966 j Peters and Lebbink, 1973). A latent period of 10-11 hr was determined by Toros et a1. (1978), although other estimates range from 6-70 hr (Bath and Tsai, 1969 j French et a1., 1973 j

Sylvester and Richardson, 1966b j Chapman and Bath, 1968 j Thottappilly et a1., 1972). Most notably, the inoculation threshold of PEMV is remarkably short, with minimum estimates ranging from 7 sec to 2 min (Toros et a1., 1978 j Nault et a1., 1964 j Nault and Gyrisco, 1966 j Bath and Chapman, 1964 j


Peters and Lebbink, 1973; Bath and Tsai, 1969). These data suggest that aphids are able to inoculate PEMV during brief test probes into nonphloem tissues, which is more typical of non circulative viruses. Unlike the noncirculative viruses, however, PEMV still requires a latent period prior to inoculation. This ability to deliver inoculum and initiate infection in nonphloem tissues represents yet another selective advantage imparted by RNA-2 to the PEMV symbiosis. Toros et a1. (1978) have suggested that, relative to mesophyll tissues, the phloem is an inefficient and possibly insusceptible site for aphid inoculation (although phloem tissue does support viral replication).

As with other viruses, nymphs are more efficient vectors than adults (Chapman and Bath, 1968; Bath and Chapman, 1968). Viruliferous aphids remain inoculative following ecdysis (transstadial passage), and may retain inoculativity for a few days to as long as 4 weeks, depending on environmental conditions (Osborn, 1935; Chaudhuri, 1950; Simons, 1954; Heinze, 1959; Nault et a1., 1964; Ehrhardt and Schmutterer, 1965; Sylvester and Richardson, 1966a,b; Sylvester, 1967). For research purposes, aphids can be rendered viruliferous by feeding either on infected plants or through artificial membranes on purified virus (Thottappilly et a1., 1972; French et a1., 1974). Abdominal injection with infectious plant extracts, hemolymph, honeydew, or purified virus have also been employed (Nault et a1., 1964; Schmutterer and Ehrhardt, 1964; Richardson and Sylvester, 1965; Schmutterer, 1969; Clarke and Bath, 1973; Harris et a1., 1975). Richardson and Sylvester (1965) compared plant extracts, hemolymph, and honeydew as sources of inocula for injection and found highest frequencies of transmission with honeydew.

B. Fate of Virus in the Vector

PEMV was the first aphid-borne plant-pathogenic virus to be detected in insects by electron microscopy (Shikata et a1., 1966; de Zoeten et a1., 1972). Harris and associates later expanded these electron microscopic studies of PEMV to A. pisum that were rendered viruliferous either by feeding on infected plants (Harris and Bath, 1972) or by direct abdominal injection (Harris et a1., 1975). AT+ and AT- isolates were used to identify specific sites of interaction between virus and vector that are needed for transmission to occur, and the following scenario of the fate of PEMV in its aphid vector was outlined. These ideas were further expanded by Harris (1979, 1990). The reader is encouraged to make comparisons with similar studies on luteovirus circulative transmission (Gildow, 1987, 1990, 1993; Gildow and Gray, 1993).

In the digestive tract, foregut epithelial cells are devoid of detectable virus. The intact intima in this region of the gut apparently serves as a barrier excluding the passage of ingested virus into these cells; ingested materials are also absent in the space between the free borders of the cells and the intima. In contrast, dense concentrations of virions occur in the stomach, intestine, and hindgut lumina of aphids following a 24-hr acquisition-access


FIGURE 11. PEMV particles in the stomach (anterior midgut) lumen of a pea aphid, Acyrthosiphon pisum, rendered viruliferous by feeding on an infected Pisum sativum plant. Particles are loosely scattered in some areas, but tend to aggregate along the periphery of ingested food (IF) materials. Inset, enlargement of the outlined area. Virions in the gut lumen do not stain equally; some are quite electron-dense (a-arrow), others less so (b-arrows), and still others appear electron-transparent (c-arrow). Particles are polygonal, mainly hexagonal, in profile. Magnification bar, 0.5 /-Lm (inset, 0.1 /-Lm).

FIGURE 12. PEMV particles associated with the microvillous border of the intestine (posterior midgut) of an aphid fed on an infected plant. Note virions in an electron-dense matrix in the center of the photograph (a-arrow!, along the plasma membrane border (b-arrow), and within microvilli (c-arrows). Virions in the cytoplasm of microvilli are not surrounded by plasma membrane and appear identical to those outside intestinal cells.

period (Figs. 11 and 12). It is uncertain whether these virions represent ingested virus becoming concentrated in the stomach or progeny particles released from infected gut epithelial cells. These high concentrations of virus in the gut lumen also explain the high concentration of virions in honeydew (Richardson and Sylvester, 1965).

The stomach (Fig. 11) and intestine are the first regions where ingested PEMV particles contact the absorbing tissues of the vector. In the gut lumen, PEMV particles are observed in contact with the microvillous borders of intestinal epithelial cells (Fig. 12). Virions can also be detected inside the microvilli of these cells, which appear to be devoid of virus. No differences are evident between these particles and those outside the microvilli. Because


there was no evidence of plasma membrane-virion interactions that might be interpreted as viropexis or phagocytosis, it was suggested that PEMV virions enter gut epithelial cells by direct penetration.

After reaching the hemocoel, virions circulate until they reach the salivary system. Virions of an AT+ isolate were detected in the basal laminae of the accessory salivary glands and in the labyrinth of cisternae formed by extensive infolding and anastomosing of the plasma membranes of these cells (Figs. 13 and 14). Although the main salivary gland has also been suggested as a possible route for passage of virus particles, there is no evidence of virions entering it. These observations are therefore consistent with a membrane-mediated shuttling of virions from hemocoel to salivary duct lumina. Once in the lumina, it is thought that virions are expelled in the watery saliva released during the feeding process.

The differential response of aphids to AT+ and AT- virions offers an approach to examine the specificity of the virus-vector interaction. The gut-

FIGURE 13. PEMV particles in the basal lamina (BL) at the basal (distal) surface of an accessory salivary gland cell of a viruliferous aphid. Virions are thought to enter (arrow) the labyrinth of cisternae (formed by infolding of the basal plasmalemma) by attaching specifically to the plasmalemma and to be transported by membrane flow to the infolded apical (proximal) plasmalemma. The former basal plasmalemma and apical plasmalemma then fuse and open ("exo-cytosis"), allowing the virion to enter the canaliculus of a salivary duct cell. M, mitochondrion. Magnification bar, 0.2 /-Lm.


FIGURE 14. A group of six virions (arrow and inset) in a plasma membrane cisterna in the more central cytoplasm of an accessory salivary gland cell of a viruliferous aphid. M, mitochondrion. Magnification bar, 0.2 f1m (inset, 0.1 f1m).

hemocoel interface is a possible selective barrier to AT+ and AT- isolates of PEM~ but the available evidence does not support this. Examination of aphids that had acquired AT+ or AT- virions either by feeding or by injection showed that virions of each type of isolate were detectable in mesodermal cells, connective tissue cells, basophilic mesodermal cells, and fat cells (Harris, 1979). Thus AT- virions are capable of crossing the gut-hemocoel barrier. Indeed, AT+ and AT- isolates were found to be identical in their distribution in the aphid, with the notable exception of the salivary gland system. In contrast to what is found with AT+ isolates, no virions were found in any part of the salivary system of aphids fed on or injected with preparations of AT- isolates (Harris, 1979; Harris et a1., 1975). In addition, PEMV particles injected into the hemocoel took longer to reach the salivary gland basal lamina and were found there in greater numbers in inefficient transmission biotypes of A. pisum than in efficient ones. Together, these data demonstrate that differences in aphid transmissibility reflect differences in interaction of virus particles with the accessory salivary gland membranes.

In addition to observations on the circulation of PEMV virions in the aphid, Harris and co-workers found PEMV virions within a variety of


electron-dense, phagocytic structures and "viroplasmlike" structures in the cytoplasm of cells of the stomach and midgut of aphids embedded 24 hr after the start of the acquisition feed (Fig. 15). Whether these were true viroplasms or developing lysosomes is uncertain. Later, polymorphic structures occurred, containing numerous virions, either scattered or in aggregates, and bounded by a single membrane. Some of the inclusions enclosed myelinlike figures or had tubular projections, each of which also contained virions. Similar membrane-bounded viral inclusions were also observed in the cytoplasm of midgut epithelial cells, fat cells, connective tissue cells, and basophilic mesodermal cells and were tentatively identified as secondary and later stage lysosomes (Fig. 16).

Virions also enter a variety of cells of mesodermal origin, including connective tissue cells, fat cells, and basophilic mesodermal cells. Uptake by these cells occurs via phagocytosis, in endocytotic vacuoles, which are subsequently processed through the elaborate lysosomal apparatus. The concen-

Mv

FIGURE 15. PEMV particles in cytoplasmic inclusions in an intestinal, midgut epithelial cell of an aphid fed on a PEMV-infected source plant. Defined viral inclusions (a-arrows) are partially or completely surrounded by a membrane and may enclose multimembranous, myelinlike (My) structures (undigested lipids). The true nature of the "viroplasmlike" (b-arrows) inclusions is not known; however, evidence suggests that such viral inclusions represent stages in the cell's phagosome-lysosome system or vacuolar apparatus (Harris et a1., 1975). Note that the membranes surrounding the nucleus (N), the microvillous border of the cell (Mv), and themitochondria (M) appear to have disintegrated. Magnification bar, 1.0 f.Lm.

336

PV ' .. f " ... ~ . r ... . ., • • J

S. A. DEMLER ET AL.

FIGURE 16. A larger aggregate of PEMV particles and a partially digested secondary symbiote (SS) in a heterophagosome (secondary lysosome stage) in the cytoplasm of a mesodermal cell (possibly a connective tissue cell) in a PEMV-injected aphid are shown. Some virions appear as strings of particles (p-arrow) between adjacent membranes of the myelin figure within the lysosome. Note the single virion (arrow) and secondary symbiote (SS) in the lysosome in the lower left of the micrograph. PV, phagocytic vacuole. Magnification bar, 0.5 /-Lm.

tration of virions in these structures can be high (Fig. 15). Although entry into the lysosomal apparatus is a means by which the cells can eliminate foreign materials (including PEMV virions), the continuing presence of PEMV virions in mature lysosomes and telolysosomes suggests that they are not eliminated; rather, it seems that these structures may constitute a reservoir of virions that may help to explain how the aphid can retain the virus for such long periods (Adam et a1., 1980).

C. Does PEMV Replicate in the Insect Vector?

Since the first report of the detection of PEMV particles within cells of viruliferous aphids (Shikata et aI, 1966), it has been a matter of debate


whether the virus multiplies in its insect vector. The conventional opinion is that PEMV is non-propagative (for a summary the reader is referred to the reviews of Sylvester, 1969, and Harris, 1979; 1990). Several observations support this view. First, the transmission efficiency of viruliferous aphids decreases with time, although inoculativity can be boosted by additional acquisition feeding periods (Sylvester and Richardson, 1966a,b). Second, the inoculativity of aphids cannot be maintained by serial passage of hemolymph. Third, transmission efficiency is correlated with inoculum dose. The long persistence of PEMV in gut tissues has been cited as evidence of replication of PEMV in cells surrounding the gut tissues and release into the gut lumen. However, as suggested by Adam et a1. (1980), this observation could also reflect the release of ingested virus from lysosomal bodies.

Further evidence against PEMV multiplication in the aphid can be drawn from experiments with mixtures of AT+ and AT- isolates. Infections of plants with such mixtures can lead to heterologous encapsidation, which implies that multiplication of the virus has occurred; however, heterologous encapsidation was never observed when aphids were injected with mixtures of the two types of isolate (Adam, 1978). In addition, although PEMV antigen was detected in inoculated cells of aphid primary cultures (Adam and Sander, 1976), it was later found to have come from the inoculum (Adam et a1., 1980). Similarly, attempts by de Zoeten and Rettig (1972) to detect abnormal protein synthesis in viruliferous aphids were inconclusive. Moreover, Fargette et a1. (1982) measured the PEMV content of aphids by enzyme-linked immunosorbent assay at different times after acquisition and found no evidence of virus multiplication in the aphid.

However, these observations do not exclude the possibility of limited multiplication of PEMV in the aphid. Harris and Bath (1972), Harris et a1. (1975), and Harris (1979) detected virions in midgut cells, in which some other viruses multiply. Some midgut cells contain large concentrations of virus (Harris and Bath, 1972) and their nuclear membranes are completely degenerated or, where present, are indistinct, widely separated, and malformed, so that cell cytoplasm and nucleoplasm are not clearly separate. Whether these changes are caused by virus multiplication in the nucleus, as observed in the plant host, is unclear. However, no replication complex comparable to that found in pea plants was evident in aphid tissues (de Zoeten et a1., 1972). The presence of viroplasmlike structures too can be interpreted as evidence of PEMV multiplication. Although these have been interpreted as a component of the lysosomal apparatus, they are structurally distinct and specific to aphids that have ingested PEMV.

The recent recognition that RNA-l and RNA-2 of PEMV can each replicate on their own in plants and protoplasts complicates the issue of replication in the aphid, because now two independent processes must be sought. Only the replication of RNA-l could produce virions suitable for reintroduction into host plants. However, aphids producing only RNA-I-containing virions (which are unable to cause systemic infection of plants in the absence


of RNA-2) would be indistinguishable in transmission tests from those carrying a nonpropagative virus. Thus replication of RNA-I, even if it does occur, might go undetected. The same argument holds for aphids replicating RNA-2. Because RNA-2 fails to produce its own capsid protein, traditional transmission studies would fail to detect its replication. Though it is now generally accepted that PEMV is not able to multiply in aphid vectors, further evidence should be sought, using the more sensitive experimental approaches now available, before this possibility is completely ruled out.

ACKNOWLEDGMENTS G. Adam wishes to acknowledge the guidance and help of Prof. E. Sander and Prof. R. J. Shepherd that enabled him to achieve his results. S. A. Demler and G. A. de Zoeten were funded in part by the U. S. Department of Agriculture Competitive Research Grant No. 87-CRCR-I-2345 and by the National Science Foundation Grant No. DCB-9105334. K. F. Harris was funded in part by Texas Advanced Technology Competitive Research Grant No. 999902-169.

REFERENCES

Adam, G., 1976, Untersuchungen verschiedener Stiimme des Pea Enation Mosaik Virus im Hinblick auf ihre Ubertragbarkeit durch den Blattlausvektor Acyrthosiphon pisum (Hall.), Dissertation am Fachbereich Biologie der Universitiit Tiibingen, FRG.

Adam, G., 1978, Die gemeinsame Ubertragung eines vektoriellen und eines avektoriellen Stammes des Pea Enation Mosaik Virus durch die Blattlaus Acyrthosiphon pisum, Z. Pflkrankh. Pflschutz. 85:586.

Adam, G., and Sander, E., 1976, Isolation and culture of aphid cells for the assay of insecttransmitted plant viruses, Virology 70:502.

Adam, G., Sander, E., and Shepherd, R. J., 1979, Structural differences between pea enation mosaic virus strains affecting transmissibility by Acyrthosiphon pisum (Harris I, Virology 92:1.

Adam, G., Sander, E., and Lee, P. E., 1980, Inoculation of aphid primary cultures with an iodinated plant virus, in: Invertebrate Systems in Vitro (E. Kurstak, K. Maramorosch, and A. Diibendorfer, eds.I, pp. 483-492, Elsevier, North-Holland, Amsterdam.

Bahner, 1., Lamb, J., Mayo, M. A., and Hay, R. T., 1990, Expression of the genome of potato leafroll virus: Readthrough of the coat protein termination codon in vivo, J. Gen. Virol. 71:2251.

Bath, J. E., and Chapman, R. K., 1964, Transmission of pea enation mosaic virus by pea aphid adults and nymphs, Proc. North Central Branch. Entomol. Soc. Am. 20:100.

Bath, J. E., and Chapman, R. K., 1968, Influence of aphid stage on acquisition and inoculation phases of pea enation mosaic virus transmission, Ann. En tom 01. Soc. Am. 61:906.

Bath, J. E., and Tsai, J. H., 1969, The use of aphids to separate two strains of pea enation mosaic virus, Phytopathology 59:1377.

Blok, V. c., Ziegler, A., Robinson, D. J., and Murant, A. E, 1994, Sequences of 10 variants of the satellite-like RNA-3 of groundnut rosette virus, Virology 202:25.

Bozarth, R. E, and Chow, C. c., 1966, Pea enation mosaic virus: Purification and properties, Contrib. Boyce Thompson Inst. 23:301.

Bozarth, R. E, Wood, H. A., and Chow, C. c., 1969, Anomalous components of pea enation mosaic virus, Phytopathology 59:1019.

Brault, v., and Miller, W. A., 1992, Translational frameshifting mediated by a viral sequence in plant cells, Proc. Natl. Acad. Sci. USA 89:2262.


Burgess, J., Motoyoshi, E, and Fleming, E. N., 1974, Structural and autoradiographic observations of the infection of tobacco protoplasts with pea enation mosaic virus, Planta 119:247.

Caspar, D. 1. D., and Klug, A., 1962, Physical principles in the construction of regular viruses, Cold Spring Harbor Symp. Quant. Biol. 27:l.

Chapman, R. K., and Bath, J. E., 1968, The latent period of pea enation mosaic virus in three of its aphid vectors with emphasis on adult versus nymph comparisons, Phytopathology 58:494.

Chaudhuri, R. P., 1950, Studies on two aphid-transmitted viruses of leguminous crops, Ann. Appl. Biol. 37:342.

Cheng, S-1., Domier, 1. 1., and D' Arcy, C. J., 1994, Detection ofthe readthrough protein of barley yellow dwarf virus, Viralogy 202:1003.

Chin, 1. S., Foster, J., and Falk, B. w., 1993, The beet western yellows virus ST9-associated RNA shares structural and nucleotide sequence homology with carmo- and tombusviruses, Virology 192:473.

Clarke, R. G., and Bath, J. E., 1973, Transmission of pea enation mosaic virus by the pea aphid, Acyrthosiphon pisum, following virus acquisition by injection, Ann. Entomol. Soc. Am. 66:603.

Clarke, R. G., and Bath, J. E., 1977, Serological properties of aphid-transmissible and aphidnontransmissible pea enation mosaic virus isolates, Phytopathology 67:1035.

Cockbain, A. J., Jones, P., and Woods, R. D., 1986, Transmission characteristics and some other properties of bean yellow vein-banding virus, and its association with pea enation mosaic virus, Ann. Appl. BioI. 108:59.

Demler, S. A., and de Zoeten, G. A., 1989, Characterization of a satellite RNA associated with pea enation mosaic virus, r. Gen. Virol. 70:1075.

Demler, S. A., and de Zoeten, G. A., 1991, The nucleotide sequence and luteovirus-like nature of RNA 1 of an aphid non-transmissible strain of pea enation mosaic virus, T. Gen. Viral. 72:1819.

Demler, S. A., Rucker, D. G., and de Zoeten, G. A., 1993, The chimeric nature of the genome of pea enation mosaic virus: the independent replication of RNA 2, T. Gen. Virol. 74:l.

Demler, S. A., Borkhsenious, o. N., Rucker, D. G., and de Zoeten, G. A., 1994a, Assessment of the autonomy of replicative and structural functions encoded by the luteo-phase of pea enation mosaic virus, T. Gen. Viral. 75:997.

Demler, S. A., Rucker, D. G., Nooruddin, 1., and de Zoeten, G. A., 1994b, Replication of the satellite RNA of pea enation mosaic virus is controlled by RNA 2-encoded functions, r. Gen. Virol. 75:1399.

Demler, S. A., Rucker, D. G., and de Zoeten, G. A., 1994c, Expression and suppression of circulative aphid transmission in pea enation mosaic virus, Abstracts Annual Meeting Am. Soc. Viral., 1994, Madison, Wisconsin, USA, 219.

de Zoeten, G. A., and Demler, S. A., 1995, Genus Enamovirus, in: Virus TaxonomyClassification and Nomenclature of Viruses. Sixth Report of the International Committee on Taxonomy of Viruses IF. A. Murphy, C. M. Fauquet, D. H. 1. Bishop, S. A. Ghabrial, A. W. Jarvis, G. P. Martelli, M. A. Mayo, and M. D. Summers, eds.), pp. 384-387, Springer-Verlag, Vienna (also in Arch. Virol., Suppl. 10).

de Zoeten, G. A., and Gaard, G., 1983, Mechanisms underlying systemic invasion of pea plants by pea enation mosaic virus, Interviralogy 19:85.

de Zoeten, G. A., and Rettig, N., 1972, Plant and aphid protein patterns as influenced by pea enation mosaic virus, Phytopathology 62:1018.

de Zoeten, G. A., Gaard, G., and Diez, F. B., 1972, Nuclear vesiculation associated with pea enation mosaic virus-infected plant tissue, Virology 48:638.

de Zoeten, G. A., Powell, C. A., Gaard, G., and German, T. 1., 1976, In situ localization of pea enation mosaic virus double-stranded ribonucleic acid, Viralogy 70:459.

de Zoeten, G. A., Demler, S. A., Rucker, D. G., Ziegler, A., Robinson, D. J., and Murant, A. F., 1994, Replicative cross-support of the satellite RNAs of the groundnut rosette disease complex and pea enation mosaic virus, Abstracts Annual Meeting Am. Soc. Viral., 1994, Madison, Wisconsin, USA, 206.


Dinesh-Kumar, S. P., Brault, V., and Miller, W. A., 1992, Precise mapping and in vitro translation of a trifunctional subgenomic RNA of barley yellow dwarf virus, Virology 187:711.

Domingo, E., Martinez-Salas, E., Sobrino, F., De la Torre, J., Portela, A., Ortin, J., LopezGalindez, C., Perez-Brena, P., Villanueva, N., Najera, R., Van de Pol, S., Steinhauer, D., De Polo, N., and Holland, J., 1985, The quasispecies (extremely heterogeneous) nature of viral RNA genome populations: Biological relevance-a review, Gene 40:1.

Ehrhardt, P., and Schmutterer, H., 1964, Untersuchungen iiber die Beziehungen zwischen dem Enationenvirus der Erbse und seinen Vektoren. I. Ubertragungsversuche mit verschiedenen, an viruskranken Vicia faba-Pflanzen aufgewachsenen Blattlausarten, Z. Pflkrankh. Pflschutz. 71:381.

Ehrhardt, P., and Schmutterer, H., 1965, Untersuchungen iiber die Beziehungen zwischen dem Enationenvirus der Erbse und seinen Vektoren. III. Das Ubertragungsvermogen verschiedener Entwicklungsstadien von Acyrthosiphon pisum (Harr.) nach unterschiedlichen Saugzeiten an virusinfizierten Vicia faba-Pflanzen, Phytopathol. Z. 52:73.

Esau, K., and Hoefert, L. L., 1972, Development of infection with beet western yellows virus in the sugarbeet, Virology 48:724.

Falk, B. w., and Duffus, J. E., 1981, Epidemiology of helper-dependent persistent aphid transmitted virus complexes, in: Plant Diseases and Vectors: Ecology and Epidemiology (K. Maramorosch and K. F. Harris, eds.), pp. 161-179, Academic Press, New York.

Falk, B. W., Duffus, J. E., and Morris, T. J., 1979, Transmission, host range, and serological properties of the viruses that cause lettuce speckles disease, Phytopathology 69:612.

Fargette, D., Jenniskens, M.-T., and Peters, D., 1982, Acquisition and transmission of pea enation mosaic virus by the individual pea aphid, Phytopathology 72:1386.

Farro, A., 1973, Properties of the infective nucleic acid from pea enation mosaic virus, Phytopathol. Z. 76:11.

Farro, A., and Rassel, A., 1971, Caracteristiques en microscopie electronique d'un isolement du virus de la mosaique 'enation' du pois, Ann. Phytopathol. 3:53.

French, J. V., Bath, J. E., Tsai, J. H., and Thottappilly, G., 1973, Purification of pea enation mosaic virus from its pea aphid vector, Acyrthosiphon pisum, and aphid-transmission characteristics, Virology: 51:78.

French, J. v., Thottappilly, G., and Bath, J. E., 1974, Aphid-transmissibility of pea enation mosaic virus prepared by various purification regimes, Z. Pflkrankh. Pflschutz. 81:561.

Gabriel, C. J., 1983, Investigations into the in Vitro Translation and Aphid Transmission of Pea Enation Mosaic Virus, Ph.D. Thesis, University of Wisconsin, Madison, WI.

Gabriel, C. J., and de Zoeten, G. A., 1984, The in vitro translation of pea enation mosaic virus RNA, Virology 139:223.

German, T. L., 1974, Purification and Properties of the Replicative Forms and Replicative Intermediates of Pea Enation Mosaic Virus, Ph.D. Thesis, University of Wisconsin, Madison, WI.

German, T. L., and de Zoeten, G. A., 1975 Purification and properties of the replicative forms and replicative intermediates of pea enation mosaic virus, Virology 66:172.

German, T. L., de Zoeten, G. A., and Hall, T. C., 1978, Pea enation mosaic virus RNA contains no polyadenylate sequences and cannot be aminoacylated, Intervirology 9:226.

Gibbs, A. J., Harrison, B. D., and Woods, R. D., 1966, Purification of pea enation mosaic virus, Virology 29:348.

Gildow, F. E., 1987, Virus-membrane interactions involved in circulative aphid transmission of luteoviruses by aphids, CUII. Top. Vector Res. 4:93.

Gildow, F. E., 1990, Barley yellow dwarf virus-aphid interactions associated with virus transmission and vector-specificity, in: World Perspectives on Barley Yellow Dwarf (P. A. Burnett, ed.), pp. 111-122, CIMMYT, Mexico, D.F., Mexico.

Gildow, F. E., 1993, Evidence for receptor-mediated endocytosis regulating luteovirus acquisition by aphids, Phytopathology 83:270.

Gildow, F. E., and Gray, S. M., 1993, The aphid salivary gland basal lamina as a selective barrier


associated with vector-specific transmission of barley yellow dwarf luteoviruses, Phytopathology 83:1293.

Gill, C. C., and Chong, T., 1976, Differences in cellular ultrastructural alterations between variants of barley yellow dwarf virus, Virology 75:33.

Gonsalves, D., and Shepherd, R. T., 1972, Biological and physical properties of the two nucleoprotein components of pea enation mosaic virus and their associated nucleic acids, Virology 48:709.

Habili, N., and Symons, R. H., 1989, Evolutionary relationship between luteoviruses and other RNA plant viruses based on sequence motifs in their putative RNA polymerases and nucleic acid helicases, Nucleic Acids Res. 17:9543.

Hagedorn, D. T., Lane, R. E. c., and Ruppel, E. G., 1964, Host range of pea enation mosaic virus and use of Chenopodium album as a local-lesion host, Phytopathology 54:843.

Harris, K. F., 1979, Leafhoppers and aphids as biological vectors: Vector-virus relationships, in: Leafhopper Vectors and Plant Disease Agents (K. Maramorosch and K. F. Harris, eds.), pp. 217-308, Academic Press, New York.

Harris, K. F., 1990, Aphid transmission of plant viruses, in: Plant Viruses, Vol. 2, Pathology (C. 1. Mandahar, ed.), pp. 177-204, CRC Press, Boca Raton, F1.

Harris, K. F., and Bath, T. E., 1972, The fate of pea enation mosaic virus in its pea aphid vector Acyrthosiphon pisum (Harris), Virology 50:778.

Harris, K. F., Bath, T. E., Thottappilly, G., and Hooper, G. R., 1975, Fate of pea enation mosaic virus in PEMV-injected pea aphids, Virology 65:148.

Harrison, B. D., Finch, r. T., Gibbs, A. T., Hollings, M., Shepherd, R. T., Valenta, v., and Wetter, C., 1971, Sixteen groups of plant viruses, Virology 45:356.

Heinze, K., 1959, Versuche tiber die Haltbarkeit des Enationenvirus der Erbse (scharfes Adernmosaik) in der Erbsenblattlaus Acyrthosiphon onobrychis (B.d.F.), Anz. Schiidlingskunde 32:123.

Hill, T. H., and Shepherd, R. T., 1972, Molecular weights of plant virus coat proteins by polyacrylamide gel electrophoresis, Virology 47:817.

Hull, R., 1976, The behaviour of salt-labile viruses in gradients of caesium sulphate, Virology 75:18.

Hull, R., 1977a, Particle differences related to aphid transmissibility of a plant virus, r Gen. Virol. 34:183.

Hull, R., 1977b, Properties of an aphid-borne virus: Pea enation mosaic virus, in: Aphids as Virus Vectors (K. F. Harris and K. Maramorosch, eds.), pp. 137-162, Academic Press, New York.

Hull, R., 1981, Pea enation mosaic virus, in: Handbook of Plant Virus Infections. Comparative Diagnosis (E. Kurstak, ed.), pp. 239-256, Elsevier/North-Holland, Amsterdam.

Hull, R., and Lane, 1. C., 1973, The unusual nature of the components of a strain of pea enation mosaic virus, Virology 55:1.

Izadpanah, K., and Shepherd, R. T., 1966a, Galactia sp. as a local lesion host for the pea enation mosaic virus, Phytopathology 56:458.

Izadpanah, K., and Shepherd, R. T., 1966b, Purification and properties of the pea enation mosaic virus, Virology 28:463.

Kennedy, T. S., Day, M. F., and Eastop, V. F., 1962, A Conspectus of Aphids as Vectors of Plant Viruses, Commonwealth Institute of Entomology, London.

Koonin, E. v., 1991, The phylogeny of RNA-dependent RNA polymerases of positive-strand RNA viruses, r Gen. Virol. 72:2197.

Koonin, E. v., and Dolja, V. v., 1993, Evolution and taxonomy of positive-strand RNA viruses: Implications of comparative analysis of amino acid sequences, Crit. Rev. Biochem. Mol. Biol. 28:375.

Kumar, I. K., Murant, A. F., and Robinson, D. T., 1991, A variant of the satellite RNA of groundnut rosette virus that induces brilliant yellow blotch mosaic symptoms in Nicotiana benthamiana, Ann. Appl. Biol. 118:555.

Lai, M. M. c., 1992, RNA recombination in animal and plant viruses, Microbiol. Rev. 56:61.


Mahmood, K., and Peters, D., 1973, Purification of pea enation mosaic virus and the infectivity of its components, Neth. ,. Plant Pathol. 79:138.

Martin, R. R., Keese, P. K., Young, M. J., Waterhouse, P. M., and Gerlach, W. 1., 1990, Evolution and molecular biology of luteoviruses, Annu. Rev. Phytopathol. 28:34l.

Mayo, M. A., Robinson, D. J., Jolly, C. A., and Hyman, 1., 1989, Nucleotide sequence of potato leafrolliuteovirus RNA, J. Gen. Viral. 70:1037.

Miller, W. A., Waterhouse, P. M., and Gerlach, W. 1., 1988, Sequence and organization of barley yellow dwarf virus genomic RNA, Nucleic Acids Res. 16:6097.

Motoyoshi, F., and Hull, R., 1974, The infection of tobacco protoplasts with pea enation mosaic virus, J. Gen. Viral. 24:89.

Murant, A. F., 1990, Dependence of groundnut rosette virus on its satellite RNA as well as on groundnut rosette assistor luteovirus for transmission by Aphis craccivora, J. Gen. Virol. 71:2163.

Murant, A. F., 1993, Complexes of transmission-dependent and helper viruses, in: Diagnosis of Plant Virus Diseases (R. E. F. Matthews, ed.I, pp. 334-357, CRC Press, Boca Raton, F1.

Murant, A. F., and Kumar, I. K., 1990, Different variants of the satellite RNA of groundnut rosette virus are responsible for the chlorotic and green forms of groundnut rosette disease, Ann. Appl. Biol. 117:85.

Murant, A. F., Rajeshwari, R., Robinson, D. J., and Raschke, J. H., 1988, A satellite RNA of groundnut rosette virus that is largely responsible for symptoms of groundnut rosette disease, J. Gen. Virol. 69:1479.

Murant, A. F., Robinson, D. J., and Gibbs, M. J., 1995, Genus Umbravirus, in: Virus TaxonomyClassification and Nomenclature of Viruses. Sixth Report of the International Committee on Taxonomy of Viruses (F. A. Murphy, C. M. Fauquet, D. H. 1. Bishop, S. A. Ghabrial, A. W. Jarvis, G. P. Martelli, M. A. Mayo, and M. D. Summers, eds.I, pp. 388-391, Springer-Verlag, Vienna (also in Arch. Viral. Suppl. 101.

Mushegian, A. R., and Koonin, E. v., 1993, Cell-to-cell movement of plant viruses, Arch. Virol. 133:239.

Musil, M., Marcinka, K., and Ciampar, F., 1970, Some properties of pea enation mosaic virus, Acta Viral. 14:285.

Nault, 1. R., 1967, Inoculation of pea enation mosaic virus by the green peach, potato, and foxglove aphids, J. Econ. Entomol. 60:1586.

Nault,1. R., 1975, Tests for transmission of pea enation mosaic virus by oligophagous mustard and grain aphids, Phytopathology 65:496.

Nault, 1. R., and Gyrisco, G. G., 1966, Relation of the feeding process of the pea aphid to the inoculation of pea enation mosaic virus, Ann. Entomol. Soc. Am. 59:1105.

Nault, 1. R., Gyrisco, G. G., and Rochow, W. F., 1964, Biological relationship between pea enation mosaic virus and its vector, the pea aphid, Phytopathology 54:1269.

Osborn, H. T., 1935, Incubation period of pea mosaic in the aphid, Macrosiphum pisi, Phytopathology 25:160.

Peters, D., 1982, Pea enation mosaic virus, CMI/AAB Description of Plant Viruses No. 257. Peters, D., and Lebbink, G., 1973, The effect of oil on the transmission of pea enation mosaic

virus during short inoculation probes, Entomol. Exp. Appl. 16:185. Powell, C. A., and de Zoeten, G. A., 1977, Replication of pea enation mosaic virus RNA in

isolated pea nuclei, Proc. Natl. Acad. Sci. USA 74:2919. Powell, C. A., de Zoeten, G. A., and Gaard, G., 1977, The localization of pea enation mosaic

virus-induced RNA-dependent RNA polymerase in infected peas, Virology 78:135. Radloff, R. J., and Kaesberg, P., 1973, Electrophoretic and other properties of bacteriophage QI3:

The effect of a variable number of read-through proteins, ,. Virol. 11:116. Reddy, D. V. R., Murant, A. F., Duncan, G. H., Ansa, O. A., Demski, J. W., and Kuhn, C. W., 1985a,

Viruses associated with chlorotic rosette and green rosette diseases of groundnut in Nigeria, Ann. Appl. Biol. 107:57.

Reddy, D. V. R., Murant, A. F., Raschke, J. H., and Mayo, M. A., 1985b, Properties and partial


purification of infective material from plants containing groundnut rosette virus, Ann. Appl. Biol. 107:65.

Reisman, D., and de Zoeten, G. A., 1982, A covalently linked protein at the 5'-ends of the genomic RNAs of pea enation mosaic virus, J. Gen. Virol. 62:187.

Richardson, J., and Sylvester, E. S., 1965, Aphid honeydew as inoculum for the injection of pea aphids with pea-enation mosaic virus, Virology 25:472.

Schmutterer, H., 1969, Versuche zur Obertragung des Erbsenenationenvirus durch Injektion virushaltiger Hamolymphe von einer Blattlausart oder -form in die andere, Z. Pflkrankh. Pflschutz. 76:257.

Schmutterer, H., and Ehrhardt, P., 1964, Untersuchungen tiber die Beziehungen zwischen dem Enationenvirus der Erbse und seinen Vektoren. II. Versuche zur Obertragung des Virus durch Injektion von Haemolymphe in verschiedene Aphidenarten, Phytopathol. Z. 50:80.

Shepardson, S., Esau, K., and McCrum, R., 1980, Ultrastructure of potato leaf phloem infected with potato leafroll virus, Virology 105:379.

Shepherd, R. J., 1970, Pea enation mosaic virus, CMI/AAB Description of Plant Viruses No. 25. Shepherd, R. J., Wakeman, R. J., and Ghabrial, S. A., 1968, Preparation and properties of the

protein and nucleic acid components of pea enation mosaic virus, Virology 35:255. Shikata, E., and Maramorosch, K., 1966, Electron microscopy of pea enation mosaic virus in

plant cell nuclei, Virology 30:439. Shikata, E., Maramorosch, K., and Granados, R. R., 1966, Electron microscopy of pea enation

mosaic virus in plants and aphid vectors, Virology 29:426. Simons, J. N., 1954, Vector-virus relationships of pea-enation mosaic virus and the pea aphid

Macrosiphum pisi (Kalt.), Phytopathology 44:283. Sylvester, E. 5., 1967, Retention of inoculativity in the transmission of pea enation mosaic virus

by pea aphids as associated with virus isolates, aphid reproduction and excretion, Virology 32:524.

Sylvester, E. S., 1969, Virus transmission by aphids-a viewpoint, in: Viruses, Vectors and Vegetation (K. Maramorosch, ed.), pp. 159-173, Interscience, New York.

Sylvester, E. S., and Richardson, J., 1966a, "Recharging" pea aphids with pea enation mosaic virus, Virology 30:592.

Sylvester, E. 5., and Richardson, J., 1966b, Some effects of temperature on the transmission of pea enation mosaic virus and on the biology of the pea aphid vector, J. Eean. Entomol. 59:255.

Tacke, E., Priifer, D., Salamini, F., and Rohde, W., 1990, Characterization of a potato leafroll luteovirus subgenomic RNA: Differential expression by internal translation initiation and AUG suppression, ,. Gen. Virol. 71:2265.

Tacke, E., Priifer, D., Schmitz, J., and Rohde, W., 1991, The potato leafrolliuteovirus 17 kD protein is a single-stranded nucleic acid-binding protein, J. Gen. Virol. 72:2035.

Tacke, E., Schmitz, J., Profer, D., and Rohde, W., 1993, Mutational analysis of the nucleic acidbinding 17 kDa phosphoprotein of potato leafroll virus identifies an amphipathic a-helix as the domain for protein/protein interactions, Virology 197:274.

Thottappilly, G., Bath, J. E., and French, J. V., 1972, Aphid transmission characteristics of pea enation mosaic virus acquired from a membrane-feeding system, Virology 50:681.

Toros, S., Schotmann, C. Y. L., and Peters, D., 1978, A new approach to measure the LP50 of pea enation mosaic virus in its vector Acyrthosiphon pisum, Virology 90:235.

Tsai, J. H., and Bath, J. E., 1974, The loss of transmissibility of two pea enation mosaic virus isolates by the pea aphid, Acyrthosiphon pisum (Harris.), Proc. Am. Phytopathol. Soc. 1:115.

Veidt, I., Lot, H., Leiser, M., Scheidecker, D., Guilley, H., Richards, K., and Jonard, G., 1988, Nucleotide sequence of beet western yellows virus RNA, Nucleic Acids Res. 16:9917.

Vincent, J. R., Lister, R. M., and Larkins, B. A., 1991, Nucleotide sequence analysis and genomic organization of the NY-RPV isolate of barley yellow dwarf virus, J. Gen. Virol. 72:2347.


Vovlas, c., and Rana, G. 1., 1972, Le virosi delle piante ortensi in Puglia. VII. Lens esculenta Moench., ospite naturale del virus del mosaico con enazioni del pisello, Phytopathol. Medit. 11:97.

Xiong, Z., and Lommel, S. A., 1989, The complete nucleotide sequence and genome organization of red clover necrotic mosaic virus RNA-I, Virology 171:543.

Xiong, Z., Kim, K. H., Kendall, T. 1., and Lommel, S. A., 1993, Synthesis of the putative red clover necrotic mosaic virus RNA polymerase by ribosomal frameshifting in vitro, Virology 193:213.

CHAPTER 13

Pea Enation Mosaic Enamovirus: Ecology and Control D. J. HAGEDORN

I. INTRODUCTION

Pea enation mosaic enamovirus (PEMV) is important economically and very interesting scientifically. The disease it induces in peas was first discovered in New York State by Osborn (1935), who showed that the causal virus is aphid-transmitted in a persistent manner. The disease was named "pea enation mosaic" by Stubbs (1937), who described the symptoms in different varieties of pea (Pisum sativum), host range, insect transmission, temperature relations, and physical properties. Since these pioneering researches, many studies have been made on diverse scientific aspects of PEMV and the disease it incites. This chapter will present current knowledge on the ecology of PEMV and on the control of the pea enation mosaic disease.

PEMV is virtually worldwide in occurrence and may well be the cause of one of the most important virus diseases of peas in the United States. It also ranks high in worldwide importance (Hagedorn, 1974). In the United States, pea enation mosaic is important primarily in the northeast and northwest pea-growing areas. Severe pea enation mosaic epidemics were reported in the

D. J. HAGEDORN • Department of Plant Pathology, University of Wisconsin, Madison, Wisconsin 53706.

345

346 D. J. HAGEDORN

1950s in New York State, where estimates of disease incidence were nearly 90%. Pozdena et a1. (1955) reported troublesome pea yield reductions caused by PEMV in Czechoslovakia. A similar outbreak of pea enation mosaic in 1952 in West Germany was noted by Quantz (1952). Hagedorn (1958) found substantial amounts of PEMV-incited disease in 1957 pea crops in the Netherlands, Switzerland, West Germany, and England. Klein et a1. (1991) found that PEMV was one of the most predominant causes of a virus epidemic in chickpea (Cicer arietinum), lentil (Lens culinaris), and pea in eastern Washington and northern Idaho in 1990. Natural infection of lentil was also reported in Europe (Vovlas and Rana, 1972).

PEMV also causes a troublesome disease of field bean (Vicia faba) in Europe. Heathcote and Gibbs (1962) reported yield losses of up to 50% in England. In the former German Democratic Republic, PEMV was one of the most important viruses affecting field bean in studies covering an extended period (Schmidt, 1984, 1986; Schmidt et a1., 1988; Karl et a1., 1989).

Strainlike differences have been found when comparing isolates of PEMV used for aphid transmission studies. Osborn (1938) first reported that symptoms incited by a continuously aphid-transmitted PEMV isolate were conspicuously more severe than those produced by an isolate that was continuously transferred mechanically. McWhorter and Cook (1958) made a similar comparison and obtained similar results. In contrast, Ruppel and Hagedorn (1963) compared five mechanically transmitted PEMV isolates, under standardized conditions, with regard to physical properties, pea varietal reaction, and host symptomatology, and found few or no differences. More recent studies of a physical-chemical nature have established the characteristics of PEMV strain relationships. This research has been summarized by Hull (1977) and in Chapter 12 (this volume). Hampton (1984) stated that "most PEMV isolates are readily aphid-transmissible in nature, but only weakly transmissible by mechanical means."

II. ECOLOGY

A. Host Range

1. Natural Hosts

Pea enation mosaic is often more troublesome near the edges of pea fields, presumably because more PEMV-carrying aphids land in these locations. Initial symptoms on diseased peas (Stubbs, 1937) are plant stunting and foliage mosaic, followed by distinct foliage malformation and the development of enations (blisterlike tissue proliferations) on the underside of leaves and stipules and on pods. Chlorotic elongate, translucent areas, sometimes called "windows," often develop on diseased foliage. Significant pea yield

PEMV: ECOLOGY AND CONTROL 347

reduction often results. Symptoms in broad bean (V faba) are similar (Heathcote and Gibbs, 1962).

One of the most important and widespread natural host plants for PEMV is alfalfa (lucerne: Medicago sativa), although Farro and Vanderveken (1969) seem to be the only workers to have isolated PEMV from field-grown alfalfa. The virus was detected by enzyme-linked immunosorbent assay (ELISA) in a survey of viruses of alfalfa in western United States and Canada in 1986 (Rahman and Solberg, 1988). McWhorter and Cook (1958) reported circumstantial evidence that alfalfa served as an overwintering host of PEMV: early virus infections in pea fields in Oregon and Washington were most prevalent to the leeward of alfalfa fields, and the frequency of infection decreased as the distance from alfalfa fields increased.

McWhorter and Cook (1958) discovered that subterranean clover (Trifolium subterraneum) is a natural host for PEMV in Oregon, and they listed the following reported additional naturally occurring hosts: broad bean (Vicia faba), common vetch (V sativa), crimson clover (Trifolium incarnatum), field pea (P. arvense), hairy vetch (V villosa), lentil (Lens culinaris), narbonne vetch (V narbonensis), perennial pea (Lathyrus latifolius) spotted bur clover (Medicago arabica), sweet pea (Lathyrus odoratus), tangier pea (Lathyrus tingitanus), yellow lupin (Lupinus luteus), and yellow sweet clover (Melilotus officinalis). Many of these plants are reported as natural hosts, both in Europe and in the United States. McEwen and Schroeder (1956) considered alfalfa, common vetch, and white clover (Trifolium repens) to be probable overwintering hosts of PEMV in New York State, and Cockbain and Gibbs (1973) believed that kidney vetch (Anthyllis vulneraria) could be a natural source of PEMV in England, though probably less important than V sativa. There are no reported nonleguminous natural hosts of PEMV.

2. Experimental Hosts

The experimental host range of PEMV was studied by many legume virus scientists, primarily during the period of 1935-1973. The most comprehensive researches on this subject are those by Stubbs (1937), McEwen and Schroeder (1956), and Hagedorn (1957). Stubbs (1937) inoculated 16 plant species with pea aphids (Acyrthosiphon pisum) and established the following experimental hosts: broad bean, crimson clover, soybean, sweet pea, and yellow sweet clover. McEwen and Schroeder (1956) aphid-inoculated seedlings of 14 leguminous species including six alfalfa cultivars, two cultivars of red clover (Trifolium pratense), and two types of white clover. Two PEMV isolates were used, one from Washington State and one from New York State. These studies added seven more species to the experimental host range, by far the most important being the six cultivars of alfalfa tested. The other significant hosts discovered were common and hairy vetch, and ladino and white clovers (T. repens). Hagedorn (1957) mechanically inoculated 44 le-

348 D. J. HAGEDORN

gumes and observed symptoms in 27 species, but recovered PEMV from only 20. He considered that 11 of these species were new hosts of PEMV. Possibly the most significant were chickpea and blue lupin (Lupinus angustifolius). Izadpanah and Shepherd (1966) reported that the primary leaves of aGalactia sp. have limited usefulness for local lesion assays.

Hagedorn et al. (1964) inoculated PEMV to 48 plant species in 23 genera and 3 families, and found 20 species in 10 genera and 3 families to be susceptible, among which Nicotiana clevelandii was a new nonlegume systemic host and Chenopodium album was a useful local lesion host. Gonsalves and Shepherd (1972) used C. amaranticolor and C. quinoa as local lesion hosts. Other reported hosts are Gomphrena globosa (TwardowiczJakuszowa, 1969), Nicotiana benthamiana (Demler et al., 1993), and N. tabacum cv. White Burley (see Chapter 12, this volume). Thus the only reported nonleguminous hosts of PEMV are C. album, C. amaranticolor, C. quinoa (Chenopodiaceae), N. benthamiana, N. clevelandii, N. tabacum cv. White Burley (Solanaceae), and G. globosa (Amaranthaceae).

B. Vector Relations

PEMV is transmitted by several aphid species in a circulative (nonpropagative) manner. Kennedy et al. (1962) list the following vectors: pea aphid (Acyrthosiphon pisum), cotton aphid (Aphis gossypii), potato aphid (Macrosiphum euphorbiae), ornate aphid (Myzus ornatus), and green peach aphid (Myzus persicae). Researchers do not agree on which aphid is the most efficient vector (Chaudhuri, 1950 j Simons, 1954 j Bath and Chapman, 1966). Transmission of PEMV from pea to pea was reported by Nault (1975), who used three grain aphid species for which peas were not a host. By far the most important vector in nature is the pea aphid (A. pisum). This aphid often lays its eggs on alfalfa (and to a lesser extent on vetches and clover) in the fall, and PEMV is transmitted to peas and/or other legumes in the following season. In England in 1966,1967, and 1968, Cockbain and Gibbs (1973) found pea aphids carrying PEMV on common vetch in the spring, before winged aphids started to migrate.

All pea aphid instars transmit PEMV, nymphs more efficiently than adults. Nymphs can acquire the virus in 15 min, but adults may take 1-2 hr. Adult aphids usually require a temperature-dependent latent period of 4-70 hr, after which they can inoculate plants in 1-2 min (Bath and Chapman, 1968). After the latent period, pea aphids can inoculate via brief test probes, even into superficial tissues. The longer the inoculation feed, the more plants become infected (Nault and Gyrisco, 1966). The length of the retention period increases and the frequency of transmission decreases with the age of the aphid (Sylvester and Richardson, 1966b). Transmission efficiency can be influenced by the strain of A. pisum involved (Tsai et al., 1972).

PEMV is retained by aphids for up to 30 days and is retained after molting


(Clarke and Bath, 1973). Evidence for virus replication in the aphid is lacking, but virus particles are present in vector cells (see Chapter 12). Aphids can acquire the virus from purified preparations through Parafilm membranes (Thottappilly et a1., 1972) and become infective after injection with honeydew (Richardson and Sylvester, 1965). Virus has been found in individual aphids by using ELISA (Fargette et a1., 1982).

C. Environmental Influences

1. Disease Incidence

The incidence of PEMV in the pea crop is highly dependent on movement of infective aphids from overwintering source plants such as alfalfa, vetches, and clovers into pea fields. Because A. pisum is the major vector, there is particular interest in the natural activities of these insects. They develop alatae (winged forms), which move from their "mother" plants when this food source becomes less succulent. Shaefer (1938) found that less than normal fluid intake and associated increase of body wastes were associated with wing formation. If warm, humid, stormless days persist, great numbers of alatae develop and "fly" or float, with the aid of wind, to peas, and this results in high PEMV incidence. Conversely, cold (or hot), dry weather and cool, rainy conditions are not conducive to aphid multiplication and/or the development of alatae, and PEMV incidence is then relatively low.

McWhorter and Cook (1958) outlined the manner in which pea aphids spread PEMV in the field. Aphids become so numerous on alfalfa that the "excess population" migrates into nearby pea fields in May. Winged aphids may feed briefly on a pea plant, give birth to a few young aphids, and then fly on to other plants. If such aphids are carrying PEMV, each could inoculate several pea plants with the virus. Wind and rain storms commonly dislodge aphids, and they crawl up adjacent pea plants and inoculate the new host with PEMV. As the peas mature, a new generation of winged aphids develops and these insects move to later-sown pea fields. The spread of PEMV-carrying aphids from early to late peas results in the transfer of more and more virus as the season progresses, so that late-sown peas are the most severely affected. Pea enation mosaic incidence is also very dependent on aphid transmission efficiency. The effect of temperature (10, 20, or 30°C) on the transmission of PEMV by the pea aphid to sweet pea (Lathyrus odoratus) was investigated by Sylvester and Richardson (1966a). Temperature affected virus acquisition more than it did inoculation. Most of the reduction in virus acquisition occurred at 10 °C with acquisition periods of 3 hr or more. The effect of temperature on inoculation was largely associated with reduced transmission efficiency in inoculation access periods equal to or less than 30 min at 10 DC. The weighted mean period of retention of inoculativity was 29.5 days at 10 °C, 13.7 days at 20°C, and 4.3 days at 30 dc. Tsai and Bath (1970) also

350 D. J. HAGEDORN

studied temperature effects on transmission of PEMV by the pea aphid in the temperature range 10 to 30°C, and found that the most efficient transmission took place at 20 0c. These researchers also studied postinoculation temperature effects and noted that much more infection took place at greenhouse temperatures of 24-30°C than at 30-44°C.

2. Disease Development

Stubbs (1937) studied the effects of temperature on pea enation mosaic disease development in temperature-controlled glasshouses. Young inoculated plants were incubated at 20-22 °C until initial symptoms were noted, then groups of such plants were moved to each of four "temperature houses" kept at 12-14, 18-20, 22-24, and 28-30°C. Low temperatures (12-14°C) reduced the severity of the disease, largely because of less severe terminal bud distortion. Occurrence of leaf enations and transparent spots on the foliage was the same as at other temperatures. High temperatures (28-30°C) greatly increased the severity of the disease. Severe top necrosis developed during the first 10 days of exposure, and all infected plants died within 3 weeks. Typical disease symptoms were produced at 18-20 and 22-24°C. Hagedorn and Hampton (1975) observed less severe pea enation mosaic disease symptoms under growth chamber day-night temperatures of 32°C and 17 °C, respectively, than when the same pea cultivars were studied in the greenhouse at 22°C.

Mullen (1985), too, studied the effect of temperature on pea enation mosaic development. Plant growth chambers were used to grow pea cultivar Perfection 8221, under a 12-hrphotoperiod and at 16, 20, 24, and 28°C. Plants were inoculated at the two- to three-bifoliate leaf stage. At 16°C, symptoms of PEMV were very pronounced, but were much slower to develop than at other temperatures. The optimal temperature for disease development was 20°C, at which typical conspicuous symptoms were produced. At 24 °C, symptoms were reduced in extent and severity, the mosaic being less conspicuous than at lower temperatures. At the highest temperature of 28°C the pea plants were spindly and mosaic symptoms were much reduced. No attempt was made to explain the discrepancy between these results and those of Stubbs (1937).

III. CONTROL

A. Vector Control

Swenson et al. (1954) studied the effect in 1952 of insecticide applications to pea seed, soil, and pea foliage on the incidence of pea enation mosaic and other pea virus diseases in New York State. Treatment of pea seed with demeton (42%, O,O-diethyl-O-2[ethylmercapto]-ethyl thiophosphate) at 2.4


g per kg seed reduced PEMV incidence from 36 to 25 %. Soil application of 0.56 kg/hectare active ingredient lowered disease incidence to 20%, and one and three foliage applications at the same rate resulted in PEMV incidence of 21 and 16%, respectively. One and three foliage applications of parathion (25% O,O-diethyl O-p-nitrophenyl thiophosphate) at 0.28 kg/hectare reduced PEMV incidence to 22 and 12%, respectively. The three foliage applications of demeton or parathion resulted in reductions of all pea virus diseases of 62 and 55%, respectively. In 1953, parathion was used on large plots that were harvested by commercial processes. In three plantings, three parathion sprays reduced the relatively low virus disease incidence from 13.5 to 3.2%, from 29.0 to 8.4%, and from 2l.9 to 10.4%. These researchers concluded that a reduction of only 11.5% in disease incidence was of economic significance because the gain in yield was more than sufficient to pay for the cost of parathion applications. Where virus disease reduction reached 20%, there was very definite economic gain from increased yield.

Davis et al. (1961) reported on studies to control pea enation mosaic in peas with the insecticides parathion, demeton, and diazinon [O,O-diethyl-O(2-isopropyl-4-methyl-6-pyrimidinyl) phosphorothioate]. In 1955 and 1956, foliage applications were tested on Perfection and Pluperfect pea cultivars in four plantings in New York State locations. Three applications of parathion or diazinon at weekly intervals reduced the incidence of pea enation mosaic in most instances. Demeton seed treatment appeared to reduce pea stands. In both years, yield increases were usually obtained with the insecticidal treatment, but were not as large as the degree of virus disease control would indicate.

Even though these researches indicated promising results for PEMV disease control via aphid control, this approach has rarely, if ever, become a commercial practice. This may be because severe disease outbreaks are erratic in occurrence and are very difficult to foresee with accuracy.

B. Overwintering Host Control

Swenson and Hagedorn (1974), in a publication concerned with the management of aphid-borne legume viruses, recommended the control of these viruses, including PEMY, through the introduction of aphid resistance into perennial legume species that can act as overwintering hosts. This approach would provide a less fragmented approach to the problem of managing legume viruses than the traditional development of crop cultivars resistant to specific viruses, and it may well have a greater effect on the virus problem because aphid resistance would affect the prevalence of the entire complex of aphid-borne viruses. Several advantages to pea virus control through development of aphid-resistant perennial legumes were given, and examples of such research were indicated. For example, Harvey et al. (1972) were studying the development of alfalfa (lucerne) resistant to the pea aphid.

352 D. J. HAGEDORN

Wilcoxson and Peterson (1960) found that virus spread was correlated with susceptibility to the aphid in red clover cultivars and concluded that breeding aphid-resistant cultivars was a more practical approach to reducing virus spread than developing cultivars resistant to each of several viruses present in red clover. Breeding for PEMV resistance in cultivated host legume species has not become an active control effort. Thus, planting of peas or faba beans near alfalfa or clover fields is not recommended.

C. Host Plant Resistance

Several workers have examined the reaction of pea cultivars to PEMV. One of the most comprehensive early studies of this kind was reported by Stubbs (1937), who inoculated 34 pea cultivars in the greenhouse and found that all were susceptible to PEMY, with six giving 50% or more diseased plants. The most "tolerant" pea cultivars, those with less than 15% diseased plants, were Abundance, Referendum, Thomas Laxton, and Yellow Admiral. White and Raphael (1944) reported the reaction of 19 pea cultivars to PEMV in Tasmania; 12 remained healthy, but this research does not appear to have been pursued further. Smith and Dyer (1949) published results of a field test for reaction of 81 pea cultivars to PEMV. All cultivars were classified as susceptible, but 15 "seemed less adversely affected by the virus and remained fairly thrifty."

Schroeder (1951) grew 171 pea plant introductions (PIs) and 47 commercial cultivars in a field trellis planting and relied on natural spread of PEMV. He found all but one of these pea lines completely susceptible, the exception being PI 140295, in which 4 of 301 plants remained free of the disease. Subsequent progenies of these PEMV-resistant plants were designated G168. The enation symptom was widespread on all lines, but some did not develop other severe symptoms. The enation symptom was less obvious on G168, G169, G170, Bridger, Late W. R. Alaska, and Reg. W. R. Alaska. The cultivars Bonneville and Horal developed enations but none of the other symptoms. Schroeder and Barton (1958) studied the nature and inheritance of resistance to PEMV in pea. The resistant pea, G168, was not immune and most plants developed atypical symptoms consisting of faint mottle and chlorotic flecking; some plants, however, showed no symptoms. PEMV was recovered from inoculated plants regardless of symptom severity. No definitive conclusion was reached on the nature of resistance. However, crosses of G168 with commercial cultivars indicated that, genetically, resistance was conditioned by a single dominant gene En, which was not closely linked with Le, the gene for tallness, or R, the gene for round seed. The pea cultivars Perfected Freezer 60 and Surprise 60 were released as a result of these plant breeding efforts.

Ruppel and Hagedorn (1963) tested ten commercial pea cultivars and five resistant Geneva lines against five isolates of PEMV. Each virus isolate infected a high percentage of all commercial cultivars studied, and the Ge-


neva lines displayed the mild symptoms previously mentioned when inoculated with four of the PEMV isolates. The fifth isolate failed to infect four of the Geneva pea lines and produced mild symptoms on the other line.

Hagedorn and Hampton (1975) studied PEMV resistance among 43 commercial pea breeding lines from eight seed companies under greenhouse, high-temperature growth chamber, and field conditions. Degrees of resistance to PEMV among pea lines varied from zero to excellent. Nineteen breeding lines equaled or exceeded the PEMV resistance level of Perfected Freezer 60, the resistance source. Pea lines resistant to PEMV in one environment were usually resistant in the other two. Inoculation with a severe, mechanically inoculable PEMV strain produced the same reaction among pea lines as did natural inoculation by aphid vectors in the field. Later, Baggett and Hampton (1983) studied the variation in resistance in these pea lines conferred by En with inconclusive results.

Baggett (1976) released four new pea breeding lines that had been developed for multiple disease resistance-all were resistant to PEMV. They were designated Oregon M176, S423, S434, and S441 and included a range of maturity and sieve-size characteristics. Other government researchers have since developed and released pea lines and/or cultivars with excellent resistance to PEMV, but the word "immune" has not been used to describe them. These new PEMV-resistant peas are being used by commercial pea breeders in the development of new disease-resistant pea cultivars.

Muntyanu et al. (1985) studied 500 pea cultivars in Russia for reaction to PEMV in the field and found that 3 were highly resistant, 9 resistant, and 17 moderately resistant. In Germany, Schmidt et al. (1988) found quantitative PEMV resistance in faba bean inbreds, some of which were higher yielding. Baggett and Kean (1987) developed "Oregon 523" freezing pea, which was resistant to PEMV and the red clover vein mosaic virus. New York State researchers Weeden and Provvidenti (1988) found a marker locus, Adh-l, for resistance to PEMV in peas, that could be a practical marker for the En gene in breeding work. Aydin et al. (1987), working in Washington State, screened 29 lentil cultivars and accessions for resistance to PEMV and found two plant introductions (from India and Iran) that were tolerant. They suggested that these could be used for breeding improved cultivars.

REFERENCES

Aydin, H., Muehlbauer, F. J., and Kaiser, W. J., 1987, Pea enation mosaic virus resistance in lentil (Lens culinaris), Plant Dis. 7l:635.

Baggett, J. R., 1976, Oregon M176, S423, S434, S441 disease resistant pea breeding lines, Hortic. Sci. 11:620.

Baggett, J. R., and Hampton, R. 0., 1983, Pea enation mosaic virus: Variation in resistance conferred by En, The Pisum Newsletter 15:3.

Baggett, J. R., and Kean, D., 1987, "Oregon 523" freezing pea, Hortic. Sci. 22:330.

354 D. J. HAGEDORN

Bath, J. E., and Chapman, R. K., 1966, Efficiency of three aphid species in the transmission of pea enation mosaic virus, f. Econ. Entomol. 59:631.

Bath, J. E., and Chapman, R. K., 1968, Influence of aphid stage on the acquisition and inoculation phases of pea enation mosaic virus transmission, Ann. Entomol. Soc. Am. 61:906.

Chaudhuri, R. P., 1950, Studies on two aphid transmitted viruses of leguminous crops, Ann. Appl. Biol. 37:342.

Clarke, R. G., and Bath, J. E., 1973, Transmission of pea enation mosaic virus by the pea aphid, Acyrthosiphon pisum, following virus acquisition by injection, Ann. Entomol. Soc. Am. 66:603.

Cockbain, A. J., and Gibbs, A. J., 1973, Host range and overwintering sources of bean leaf roll and pea enation mosaic viruses in England, Ann. Appl. Biol. 73:177.

Davis, A. c., McEwen, E L., and Schroeder, W. T., 1961, Control of pea enation mosaic in peas with insecticides, f. Econ. Entomol. 54:161.

Demler, S. A., Rucker, D. G., and De Zoeten, G. A., 1993, The chimeric nature of the genome of pea enation mosaic virus: The independent replication of RNA 2, f. Gen. Virol. 74:1.

Fargette, D., Tenniskens, M. T., and Peters, D., 1982, Acquisition and transmission of pea enation mosaic virus by the individual pea aphid, Phytopathology 72:1386.

Farro, A., and Vanderveken, T., 1969, Proprietes d'une souche du virus de la mosaique enation du pois isolee en Belgique, Parasitic a 25:5.

Gonsalves, D., and Shepherd, R. J., 1972, Biological and physical properties of the two nucleoprotein components of pea enation mosaic virus and their associated nucleic acids, Virology 48:709.

Hagedorn, D. T., 1957, Host range and inoculation studies on pea enation mosaic virus, Phytopathology 47:14.

Hagedorn, D. T., 1958, Some observations on diseases of Pisum sativum in several European countries in 1957, Tijdschr. Pl. Ziekt. 64:263.

Hagedorn, D. T., 1974, Virus diseases of pea, Pisum sativum, American Phytopathological Society Monograph No.9.

Hagedorn, D. J., and Hampton, R. 0., 1975, Pea enation mosaic virus resistance among commercial breeding lines of Pisum sativum, Plant Dis. Rep. 59:895.

Hagedorn, D. T., Layne, R. E. c., and Ruppel, E. G., 1964, Host range of pea enation mosaic virus and use of Chenopodium album as a local lesion host, Phytopathology 54:843.

Hampton, R. 0., 1984, Pea enation mosaic virus, in: Compendium of Pea Diseases (D. J. Hagedorn, ed.1, pp. 32-33, American Phytopathological Society, St. Paul, MN.

Harvey, T. L., Hackerott, H. L., and Sorenson, E. L., 1972, Pea aphid-resistant alfalfa selected in the field, f. Econ. Entomol. 64:1661.

Heathcote, G. D., and Gibbs, A. T., 1962, Virus diseases in British crops of field beans (Vicia faba L. 1, Pl. Pathol. 11:69.

Hull, R., 1977, Properties of an aphid-borne virus: Pea enation mosaic virus, in: Aphids as Virus Vectors (K. E Harris and K. Maramorosch, eds.1, pp. 137-162, Academic Press, New York.

Izadpanah, K., and Shepherd, R. T., 1966, Galactia sp. as a local lesion host for the pea enation mosaic virus, Phytopathology 56:458.

Karl, E., Schmidt, H. E., Klein, W., Herbst, E., Schwanke, P., and Amme, M.,1989, Einschriinkung der Ausbreitung blattlausubertragbarer Viren in Bestiinden terminalinfloreszenter Ackerbohnen durch Vektorenbekiimpfung, Nachrichtenblatt Pflanzenschutz DDR 43:237.

Kennedy, T. S., Day, M. E, and Eastop, V. E, 1962, A Conspectus of Aphids as Vectors of Plant Viruses, Commonwealth Institute of Entomology, London.

Klein, R. E., Larsen, R. c., and Kaiser, W. T., 1991, Virus epidemic of grain legumes in Eastern Washington, Plant Dis. 75:1186.

McEwen, E L., and Schroeder, W. T., 1956, Host range studies on pea enation mosaic virus, Plant Dis. Rep. 40:11.

McWhorter, E P., and Cook, W. C., 1958, The hosts and strains of pea enation mosaic virus, Plant Dis. Rep. 42:51.


Mullen, J. B., 1985, Development of a multiple disease resistance screening protocol for Pisum sativum, Univ. Wisc. M.S. Res. Rep., 48 pp.

Muntyanu, S. K., Verderevskaya, T. D., Rozhkovan, V. V., Vetrova, E. G., and Vereshchaka, A. I., 1985, Evaluation of pea varieties for susceptibility to pea enation mosaic virus, Nauchnoteknicheskii Byulletin Vsesoyuznogo Selektsionno-geneticheskogo Instituta 4:57. [In Russian]

Nault,1. R., 1975, Tests for transmission of pea enation mosaic virus by oligophagous mustard and grain aphids, Phytopathology 65:496.

Nault, 1. R., and Gyrisco, G. G., 1966, Relation of the feeding process of the pea aphid to the inoculation of pea enation mosaic virus, Ann. Entomol. Soc. Am. 59:1185.

Osborn, H. T., 1935, Incubation period of pea mosaic in the aphid, Macrosiphum pisi, Phytopathology; 25:160.

Osborn, H. T., 1938, Studies on pea virus 1, Phytopathology 28:923. Pozdena, J., Soubodova, J., Petru, E., Limberk, J., and Blattny, c., 1955, A contribution to the

knowledge of pea mosaic in the Czechoslovakian Republic, Fol. Biol. 1:298. Quantz, 1., 1952, Untersuchungen iiber das Erbsenvirus 1 ("Enation"-Mosaik-Virus). I. Seine

Wirtspflanzen, Ausbreitung und Dberwinterung, Nachrbl. Dtsch. Pflanzenschutzd. (Braunschw.) 4:24.

Rahman, F., and Solberg, R. 1., 1988, Survey of viruses of alfalfa in the Western United States and Canada, in: Report of the 31st North American Alfalfa Improvement Conference, Beltsville, MD, June 1988.

Richardson, J., and Sylvester, E. S., 1965, Aphid honeydew as inoculum for the injection of pea aphids with pea enation mosaic virus, Virology 25:472.

Ruppel, E. G., and Hagedorn, D. J., 1963, Comparisons of pea enation mosaic virus isolates, Phytopathology 53:813.

Schmidt, H. E., 1984, Zum Schadausmass von Ackerbohnenvirosen in der Deutschen Demokratischen Republik und Schlussfolgerungen fiir die Bekampfung, Nachrichtenblatt Pflanzenschutz DDR 38:157.

Schmidt, H. E., 1986, Zur Virusanalyse bei Gemiise- und Kornerhiilsenfriichten in der DDR als Grundlage fiir die Bekampfung, Nachrichtenblatt Pflanzenschutz DDR 40:251.

Schmidt, H. E., Karl, E., and Meyer, U., 1988, Resistance of field bean (Vicia faba 1. ssp. minor (Peterm. em. Harz) Rothm.) to pea enation mosaic virus, Arch. Phytopathol. Pflanzenschutz 24:74.

Schroeder, W. T., 1951, An unusually heavy development of pea enation mosaic on 218 lines of pea in the field, Plant Dis. Rep. 35:156.

Schroeder, W. T., and Barton, D. W., 1958, The nature and inheritance of resistance to the pea enation mosaic virus in garden pea, Pisum sativum 1., Phytopathology 48:628.

Shaefer, C. W., 1938, Physiological conditions which produce wing development in the pea aphid, J. Agric. Res. 57:825.

Simons, J. N., 1954, Vector-virus relationships of pea enation mosaic virus and the pea aphid, Macrosiphum pisi (Kalt.), Phytopathology 44:283.

Smith, P. G., and Dyer, P. R., 1949, Results of a field test for resistance of pea varieties to the pea enation virus (pea virus 1). Plant Dis. Rep. 33:219.

Stubbs, M. W., 1937, Certain viroses of the garden pea, Pisum sativum, Phytopathology 27:242. Swenson, K. G., and Hagedorn, D. J., 1974, Management of aphid-borne legume viruses, Bulletin

of the Oregon State University Agricultural Station No. 615. Swenson, K. G., Davis, A. c., and Schroeder, W. T., 1954, Reduction of pea virus spread by

insecticide applications, J. Econ. Entomol. 47:490. Sylvester, E. S., and Richardson, J., 1966a, Some effects of temperature on the transmission of pea

enation mosaic virus and on the biology of the pea aphid vector, J. Econ. Entomol. 59:255. Sylvester, E. S., and Richardson, J., 1966b, "Recharging" pea aphids with pea enation mosaic

virus, Virology 30:592. Thottappilly, G., Bath, J. E., and French, J. V., 1972, Aphid transmission characteristics of pea

enation mosaic virus acquired from a membrane-feeding system, Virology 50:681.

356 D. J. HAGEDORN

Tsai, J. H., and Bath, J. E., 1970, Additional factors affecting the transmission of pea enation mosaic virus by the pea aphid, Ann. Entomol. Soc. Am. 63:1723.

Tsai, J. H., Bath, J. E., and Igbakwe, E. c., 1972, Biological and transmission characteristics of Acyrthosiphon pisum biotypes efficient and inefficient as vectors of pea enation mosaic virus, Ann. Entomol. Soc. Am. 65:1114.

Twardowicz-Jakuszowa, A., 1969, Badania nad wirusem ostrej mozaiki Grouchu jPisum virus 1 Smith), Zesz. probl. Postep. Nauk roln. 1969:169.

Vovlas, C., and Rana, G. 1., 1972, Le virose delle piante ortensi in Puglia. VII. Lens esculenta Moench., ospite naturale del virus del mosaico con enazioni del pisello, Phytopathol. Mediterr.11:97.

Weeden, N. F., and Provvidenti, R., 1988, A marker locus, Adh-l, for resistance to pea enation mosaic virus in Pisum sativum, J. Hered. 79:128.

White, N. H., and Raphael, T. D., 1944, The reaction of green pea varieties to downy mildew and two viruses, Tasmanian ,. Agric. 15:92.

Wilcoxson, R. D., and Peterson, A. G., 1960, Resistance of Dollard red clover to the pea aphid, Macrosiphum pisi, J. Econ. Entomol. 53:863.

INDEX

Alfalfa mosaic alfamovirus jAlMV), 6, 240, 294,296

Andean potato mottle comovirus jAPMV), 18-19, 23, 42, 68, 78, 86, 90

Aphid vectors of fabaviruses, 233-234 of pea enation mosaic enamovirus, 322,

330,345-350 Arabis mosaic nepovirus jArMV), 99-100,

102-106,112-113,117-118,122, 124-126, 140-142, 145-148, 150, 152, 154, 156, 158-162, 167, 171-172,174-177,193-194,197,199-202, 212-213, 216-218, 220-223

Arracacha A nepovirus lAVA), 102-103, 107, 142

Arracacha B, tentative nepovirus jAVB), 102-103, 107, 143, 149

Artichoke Aegean ringspot nepovirus jAARSV) , 102-103, 107, 142

Artichoke French latent fabavirus jAFLV), 229,233,243

Artichoke Italian latent nepovirus JAILV), 102-103, 107-108, 142, 154, 193-194, 201

Artichoke vein banding, tentative nepovirus jAVBV), 102-103, 108, 143, 149

Artichoke yellow ringspot nepovirus jAYRSV), 102-103, 108, 142, 171,218

Bacteriophage QP, 60, 323 Barley yellow dwarf luteovirus jBYDV), 267,

315, 317 Bean pod mottle comovirus jBPMV), 2, 19,

21-24, 27-28, 36, 39-40, 42, 63, 78,81-84,86-88,90,92-93,166

Bean rugose mosaic comovirus jBRMV), 19-20, 21, 24, 78, 82-83

Bean yellow mosaic potyvirus jBYMV), 21 Bean yellow veinbanding umbravirus

jBYVBV),319-320 Beet necrotic yellow vein furovirus jBNYW),

7,11 Beet western yellows luteovirus jBWYV), 310,

314,317,319 Beetle vectors of comoviruses, 77-83, 87-91

mechanism of transmission, 77, 81-83 vector species, 78-81 vector-virus relations, 81-83

Bipartite RNA genomes advantages and disadvantages, 8-11 affinities, 3-5, 11-13 discovery of, 2-3 possible origins, 5-8 recognition, 2-3 types, 1, 3-5

Black raspberry necrosis virus jBRNV), 283-284

Blueberry leaf mottle nepovirus jBLMV), 101-104, 108-109, 142, 152-153, 156, 158,160-161,170,218-219

Broad bean stain comovirus jBBSV), 19-20, 24-25,27-28,78,84-86,88-90

Broad bean true mosaic comovirus jBBTMV), 19-20, 27, 42, 78, 81-82, 84-86, 88-90

Broad bean wilt fabavirus jBBWV), 229-246 Brome mosaic bromovirus jBMV), 63, 150,

271 Bromoviridae, 5-7, 284, 294-296, 299 Bromoviruses, 63, 150, 259, 266, 271 Bymoviruses, 3

Caliciviruses, 37, 42 Capilloviruses, 162 Carmoviruses, 5, 7, 252, 266, 271, 273, 315

357

358

Carnation mosaic virus, original name for CRSY, 252

Carnation mottle carmovirus (CarMV), 252, 273

Carnation ringspot dianthovirus (CRSV), 251-254,256-267,270-274

Carrot motley dwarf disease, 8 Carrot mottle umbravirus (CMoV), 7, 11-12,

319 Carrot red leaf luteovirus (CaRLV), 7, 11-12 Cassava American latent nepovirus (CsALV),

102-103, 109, 142, 151 Cassava common mosaic potexvirus, 109 Cassava green mottle nepovirus (CGMV),

102-103, 109, 142 Caulimoviruses, 67, 162 Cherry leaf roll nepovirus (CLRV), 100-105,

109-110, 122-123, 125, 140-142, 146, 150, 152-153, 155, 158-161, 170, 194, 198, 216-217, 219

Cherry rasp leaf, tentative nepovirus (CRLV), 101-103, 11~ 14~ 14~ 14~ 195, 198

Cherry rosette, tentative nepovirus (CRV), 193-194, 201 i see alsa 121

Chicory yellow mottle nepovirus (Ch YMV), 102-103, 110-111, 142, 147-148

Clover primary leaf necrosis dianthovirus, strain of RCNMY, 252

Cocoa necrosis nepovirus (CNY), 102-103, 111,140,142-143,154

Camaviridae, 3, 6-7, 9, 230, 239, 243 Comoviruses, 1-13, 17-98, 101, 140, 155-

159,162,164,166-168,173,230, 239,242-244,260,311

beetle vector species, 78-81 control, 89-93 cytopathology, 21, 51 diagnosis, 18-23 diseases caused, 23-29 epidemiology, 86-89 genome organization and expression, 42-51 geographical distribution, 22-29 host range, 18-20,23-29 intercellular transport, 64-67, 91 members, 17-23 molecular biology, 35-68 particles, 36-38, 40-42 properties, 17-23 proteins

coat proteins, 36, 38-42, 44, 47-48 nonstructural proteins, 44, 46-67 polyproteins, 43-44, 46-51, 54, 56-57,

59-64 VPg,37,42-45,47,49-52, 54, 56-57,

61-63 pseudo-recombinants, 20-21, 38

INDEX

Comoviruses (cant.) relationships with other viruses, 1-13 replication, 51-64, 67-68 resistance to, 91-93 RNA

double-stranded, 53, 61-63 genomic, 36-38, 41-48, 51-64,67 nucleotide sequences, 36, 43-44

serology, 21-22 strains or variants, 22-29 symptoms, 18-20,23-29 transmission, 77-91

by beetles, 20, 23-24, 27, 77-83, 87-91 by grasshoppers, thrips, 83 mechanism in beetles, 77, 81-83 through seed and pollen, 83-86, 89-90 vector-virus relations, 81-83

Cowpea chlorotic mottle bromovirus (CCMV), 259, 271

Cowpea mosaic comovirus (CPMV), 2, 18-21,25-29,35-68,79,83-84,91-9~ 155-15~ 16~ 16~ 16~ 168, 173

Cowpea severe mosaic comovirus (CPSMV), 19-23, 25-27, 36-37, 42-43, 50, 57,63,79,81-84,86,89-93,166

Crimson clover latent nepovirus (CCLV), 102-103, Ill, 142

Cucumber mosaic cucumovirus (CMV), 60, 169,294

Cucumber necrosis tombusvirus (CuNY), 275 Cycas necrotic stunt nepovirus (CNSV), 102-

103, Ill, 142

Dependence, satellitism and multipartite genomes, 11-13

Dianthoviruses, 1-13,251-282,295,310, 315-316

control, 274 cytopathology, 271-272 diagnosis, 261-262 diseases caused, 252-254 epidemiology, 272-274 genome organization and expression, 265-

270 geographical distribution, 252-254 history, 252 host range, 252-254, 275 intercellular transport, 268-270 members, 251 molecular biology, 263-271 particles

aggregation of, 260 electrophoretic mobility of, 262 properties of, 251, 254-263 stability of, 259-260 structure of, 254-258, 259-260

INDEX

Dianthoviruses (cont.) proteins

coat protein, 255-256, 259-260, 266-268

nonstructural proteins, 266-270 pseudo-recombinants, 257, 260, 263-265,

267-268,275 relationships with other viruses, 1-13, 270-

271 RNA

double-stranded, 265 genomic, 257, 260, 263-266 nucleotide sequences, 266-267 subgenomic, 265-266, 269

serological properties, 261-262 strains or variants, 256, 259-265, 270-

271 symptoms, 252-254 transmission, 272-274

Dogwood mosaic nepovirus (DoMV), 102-103, 112, 142, 147

Echtes Ackerbohnenmosaik comovirus, synonym of BBTMY, 19, 25

Enamovirus, 1-13,303-356 aphid vectors, 348 comparison with helper-dependent virus

complexes, 316-320 control, 350-353 cytopathology

in plants, 305-307, 317 in vector insects, 331-337

diseases caused, 345-347 economic importance, 345-346 epidemiology, 349-350 genome organization and expression, 310,

314-316 geographical distribution, 345-347 host range, 304, 346-348 intercellular transport, 307, 315-319, 329 member, 303 molecular biology, 314-321 particles, 306-309, 311-313, 322-323,

329,331-337 properties, 303-304, 307-313 proteins

coat proteins, 310-312, 314-315, 317, 319,323-325

nonstructural proteins, 310, 314-316, 318-319

VPg, 312, 314 pseudo-recombinants, 321, 325-326 purification, 304 relationships with other viruses, 1-13,316-

320 replication, 305-307, 316-317 resistance to, 352-353

Enamovirus (cont.) RNA

359

double-stranded, 306, 320-321 genomic, 303, 306, 310-321, 325-330,

337-338 nucleotide sequences, 310, 314-316 subgenomic, 316-317

satellite RNA, 3{}3, 313, 320-321 strains or variants, 320-330, 346 symptoms, 304-305, 318, 320-321, 345-

348,350 transmission

by aphids, 303, 318-338, 345-349 experimental, 303, 347-348

vector nontransmissible isolates, 321-331, 333-334,337

virus-vector interactions, 330-338

Fabaviruses, 1-13, 101, 140, 173,229-250 aphid vectors, 233-234 control, 245-246 cytopathology, 234-237, 242, 244 diagnosis, 244-245 diseases caused, 229-232 economic importance, 231 epidemiology, 245-246 geographical distribution, 231-232 host range, 230-233, 245 inclusions, 234-236 intercellular transport, 236 members, 229 particles, 230, 234-238, 242, 245 properties, 230, 238 proteins

coat proteins, 241-242, 244 VPg, 238-239, 244

purification, 237 relationships with other viruses, 1-12, 242-

244 resistance to, 245-246 RNA

genomic, 230, 238-242 strains or variants, 229, 241, 243 symptoms, 230-233 transmission

by aphids, 233-234 through seed and pollen, 233, 244

Foot-and-mouth disease virus (FMDV), 41 Furcraea necrotic streak dianthovirus (FNSV),

251-252, 254, 257-258, 266, 272

Genome organization and expression of comoviruses, 42-51 of dianthoviruses, 265-270 of enamovirus, 310, 314-316 of idaeovirus, 292-296 of nepoviruses, 151-169

360

Glycine mosaic comovirus (GMV), 19,27, 77, 79

Grapevine Bulgarian latent nepovirus (GBLV), 102-103, 112, 123, 126, 140, 143, 145,174,218

Grapevine chrome mosaic nepovirus (GCMV), 102-103, 112-113, 126, 143, 148, 151-162, 164-170, 172, 175, 194,223

Grapevine fanleaf nepovirus (GFLV), 100, 102-105, 112-113, 122-127, 142, 148,150-154,156-168,172,174-176, 190, 193-196,200-202,212, 215,219-220,222

Grapevine Tunisian ringspot nepovirus (GTRSV), 102-103, 114, 143

Grapevine yellow speclde viroid (GYSVd), 113 Groundnut rosette assistor luteovirus

(GRAV), 11-12,321 Groundnut rosette disease, 8, 11-12, 320 Groundnut rosette umbravirus (GRV), 11-13,

321

Hibiscus latent ringspot nepovirus (HLRSV), 102-103, 114, 143

Human immunodeficiency virus (HIV-l), 41

Idaeovirus, 1-13,283-301 control, 298-299 diagnosis, 296-298 diseases caused, 283-284, 286-289 epidemiology, 285-286 genome organization and expression, 292-296 geographical distribution, 284 host range, 286-290 member, 284, 299 molecular biology, 292-297 particles, 291 proteins

coat protein, 291-295 nonstructural proteins, 293-297

purification, 291 relationships with other viruses, 1-13,284,

294-296,299 resistance to, 298-299 RNA

genomic, 292-294 nucleotide sequences, 292-294 subgenomic, 292-293

serology, 296-298 strains, 284-285 symptoms, 286-290 transmission

experimental, 286 through seed and pollen, 285-286 to pollinated plant, 286

Ilarviruses, 5-6

INDEX

Lamium mild mosaic fabavirus (LMMV), 229-231, 233, 238-241, 243-244

Lucerne Australian latent nepovirus (LALV), 102-103, 114, 140, 143,244

Lucerne Australian symptomless nepovirus (LASV), 101-103, 114, 143, 145, 149, 167

Luteoviruses, 5, 7, 11-12, 37, 42, 267, 271, 304,307,310,314-321,324-325, 330-331

Machlomoviruses, 266, 271 Mulberry ringspot nepovirus (MRSV), 102-

103, 114, 142, 146, 151, 193-194 Myrobalan latent ringspot nepovirus

(MLRSV), 102-103, 105, 115, 141, 143,174

Nasturtium ringspot fabavirus (NRSV), serotype I of BBWY, 229,233-234,239-241, 243

Necroviruses, 266, 271, 273, 316 Nematode vectors of nepoviruses, 100-104,

106-107, 112, 116-117, 119, 154, 163,175,187-209,212-215,218-221

biology, 188 feeding behavior, 190-191 geographical distribution, 193-196,212-

213 host preferences, 213 host response, 191-192 life cycles, 213 Longidorus spp., 107, 116, 119, 154, 163,

187-188, 190-198, 200-20~ 212-213,218,221-222

Paralongidorus sp, 187-188, 193-194, 196 Paratrichodorus, 188 population dynamics, 212-215 structure of the feeding apparatus, 188-

190 taxonomy, 188 1Tichodorus, 188 vector species, 194-195 virus transmission

criteria for demonstrating, 192 detection of virus in nematodes, 192-

193 efficiency, 199-201 ingestion, retention and inoculation,

202-203 sites of virus retention, 201-203 specificity, 163, 196-199 test procedures, 192-193 vector-virus associations, 193-196,212

Xiphinema spp., 100, 104, 106, 112, 117, 187-202,212-215,218,220-221

INDEX

Nepoviruses, 1-15,42,67,99-228,230, 239,242-244,312

classification, 1 72-1 73 control, 219-228

agronomic methods, 220 integrated control, 223-224 removing virus sources, 219 resistant cultivars, 221-223 transgenic resistance, 223 use of nematicides, 220-221

cytopathology, 171 diagnosis, 121-127 diseases caused, 102-121 economic importance, 102-104, 113 epidemiology, 211-219 genome organization and expression, 151-

169 geographical distribution, 102-103,212-

213 history, 99-100 host range, 102-121,211-213 intercellular transport, 162-163, 171 members, 100-103, 140, 142-143, 172-

173 molecular biology, 139-185 nematode vector species, 194-195 particles, 100, 121-122, 141-148 properties, 100-104 proteins

coat protein(sI, 101, 142-143, 148-149, 156-157,159-162,173,175,202

nonstructural proteins, 155-157, 162-169

polyproteins, 142-143, 154-159, 162-169

VPg, 150, 155-157, 166-168, 174, 176 pseudo-recombinants, 154-155, 175,202,

219,222 purification, 140-141 recombination, 1 70-1 71 relationships with other viruses, 1-12, 37,

140, 153-154, 159-173 replication, 169-172 resistance to, 221-223 RNA

double-stranded, 169-171 genomic, 101, 142-143, 149-172 nucleotide sequences, 151-152

satellite RNA, 145, 150-151, 173-177,223 serological relationships, 142-143 strains or variants, 106, 109-110, 112-

113, 116-119, 197,212 symptoms, 104-121, 175, 177 transmission

by nematodes, 100-103, 163, 175, 187-209,211-215,218-224

by thrips, spider mites, 187,215

Nepoviruses (cont.) transmission (cont.)

361

in seed and pollen, 100-103, 108-109, 175,211,216-219,224

to pollinated plant, 101-103, 108-109, 217-218

Nucleotide sequences of comoviruses, 36, 43-44 of dianthoviruses, 266-267 of enamovirus, 310, 314-316 of idaeovirus, 292-294 of nepoviruses, 151-152

Olive latent ringspot nepovirus (OLRSV), 102-103, 115, 142, 146

P.O. pea streak fabavirus, serotype I of BBWY, 229,241

Parsley virus 3 fabavirus (PV3), serotype I of BBWY, 229, 234, 239, 241

Parsnip yellow fleck sequivirus (PYFV), 6, 9 Pea enation mosaic enamovirus (PEMV), 1-

13, 263, 303-353; see also Enamovirus

Pea green mottle comovirus, strain of BBSY, 19

Pea mild mosaic comovirus (PMiMV), 19,27, 77, 79

Pea symptomless comovirus, strain of RCMY, 19-20

Peach rosette mosaic nepovirus (PRMV), 102-103,115-116,140,143,195-196,198

Petunia ringspot fabavirus (PeRSV), serotype I of BBWY, 229, 234, 241

Picornaviridae, 244 Picornavirus-like supergroup, 140 Picornaviruses, 6, 37, 40-41, 55, 156, 158,

168,314 Plantago mosaic fabavirus, serotype II of

BBWY, 243 Plum pox potyvirus, 55, 168 Poliovirus, 52-54, 63 Polymyxa betae, 11 Potato black ringspot nepovirus (PBRSV),

102-103, 116, 142, 150, 195 Potato leafrolliuteovirus (PLRV), 267, 314, 317 Potato U nepovirus (PVU), 102-103, 116, 142 Potexviruses, 109, 273 Potyviridae, 3, 244 Potyviruses, 21, 37, 42, 55, 156-157, 166,

168 Pseudo-recombinants, 9

of comoviruses, 20-21, 38 of dianthoviruses, 257, 260, 263-265, 267-

268, 275 of enamovirus, 321, 325-326 ofnepoviruses, 154-155, 175,202,219,222

362

Quail pea mosaic comovirus (QPMVI, 19-22, 27, 80

Radish mosaic comovirus (RaMVI, 18-19, 21,28,42,80,260

Raspberry bushy dwarf idaeovirus (RBDVI, 1-13,283-301; see also Idaeovirus

Raspberry ringspot nepovirus (RRSVI, 99-105, 107, 116-117, 119-120, 124-12~ 141-14~ 146-148, 150-16~ 167, 170-172, 190, 193-194, 196-197,200-203,212-213,216-218, 220-222,224

Red clover mottle comovirus (RCMVL 19-21, 25, 27-28, 36-37, 43, 50, 57, 63, 65-66, 80

Red clover necrotic mosaic dianthovirus (RCNMVL 251-276, 295, 310

Red clover vein mosaic carlavirus, 353 Rubus Chinese seed-borne nepovirus (RCSVI,

101-103,117,122,143,145,149,244

Satellite RNA, 11-13 ofnepoviruses, 145, 150-151, 173-177,223 of pea enation mosaic enamovirus, 303,

313,320-321 Satellite viruses, 11-13 Satsuma dwarf, tentative nepovirus (SDVL

101-103,117,143,149,244 Seed and pollen transmission

of comoviruses, 83-86, 89-90 of fabaviruses, 233, 244 of idaeovirus, 285-286 of nepoviruses, 100-103, 108-109, 175,

211,216-219,224 Sequiviridae, 6 Sindbis-like supergroup, 296 Sobemoviruses, 37, 42, 81 Soil-borne wheat mosaic furovirus, 295 Southern bean mosaic sobemovirus (SBMVI, 81 Squash mosaic comovirus (SMVI, 2, 18-20,

27-29, 42, 80-81, 83-87, 89-90, 92 Strawberry latent ringspot, tentative ne

povirus (SLRSVI, 99-106,117-118, 122, 125-126, 140-141, 143-146, 149,151-154,156-159,162-163, 167,171,173-174,194,196,199-202,212-213,216,218-219,222-223,242,244

INDEX

Sunn-hemp mosaic tobamovirus, 66 Sweet clover necrotic mosaic dianthovirus

(SCNMVI, 251-254, 257-258, 261-268,270-273,275-276

Tobacco etch potyvirus (TEVI, 157, 166 Tobacco mosaic tobamovirus (TMVI, 150,

240,261,291,296 Tobacco necrosis satellite virus (STNVI, 260 Tobacco rattle tobravirus (TRVI, 2, 150 Tobacco ringspot nepovirus (TRSVI, 99-100,

102-104,118-119,122,126,141-142, 144-145, 147-152, 157-158, 160-161, 167, 169-170, 174, 177, 187, 195, 198-200,215-216,219, 223

Tobacco streak ilarvirus (TSVI, 294 Tobamoviruses, 66, 150,240,261,269,273,

291, 296 Tobraviruses,2, 150,203 Tomato black ring nepovirus (TBRVI, 99-

105, 119-120, 122, 124-126, 141-142, 145-146, 148, 150-170,172,174-175,194,196-197,200-203, 212-213, 216-218, 220-222

Tomato bushy stunt tombusvirus (TBSVI, 150,255,258,274-275

Tomato ringspot nepovirus (ToRSVI, 99-100, 102-105, 120-122, 125-126, 140-141, 143, 145-146, 149, 151-153, 156-170, 172, 195, 197-202

Tomato top necrosis, tentative nepovirus (ToTNVI, 102-103, 121, 143, 149

Tombusviruses, 150,255,258,266,270-271,273-275,315

Tospoviruses, 67 Turnip crinkle tombusvirus (TCVI, 255,

270 Tymoviruses, 262

Ullucus C comovirus (Uvq, 18-19,21,29, 77,80,86,90

Umbraviruses, 5, 7, 11-13,316,319-321

Vectors: see Aphid vectors, Beetle vectors, Nematode vectors

Virus complexes, 7-13, 318-321

THE VIRUSES COMPLETE LISTING OF VOLUMES Series Editors HEINZ FRAENKEL-CONRAT, University of California

Berkeley, California

ROBERT R. WAGNER, University of Virginia School of Medicine Charlottesville, Virginia

THE ADENOVIRUSES Edited by Harold S. Ginsberg

THE ARENA VIRIDAE Edited by Maria S. Salvato

THE BACTERIOPHAGES Volumes 1 and 2 • Edited by Richard Calendar

THE BUNYAVIRIDAE Edited by Richard M. Elliott

THE CORONA VIRIDAE Edited by Stuart G. Siddell

THE HERPESVIRUSES Volumes 1-3 • Edited by Bernard Roizman Volume 4 • Edited by Bernard Roizman and Carlos Lopez

THE INFLUENZA VIRUSES Edited by Robert M. Krug

THE PAPOVAVIRIDAE Volume 1 • Edited by Norman P. Salzman Volume 2 • Edited by Norman P. Salzman and Peter M. Howley

THE PARAMYXOVIRUSES Edited by David W. Kingsbury

THE PARVOVIRUSES Edited by Kenneth I. Berns

THE PLANT VIRUSES Volume 1 • Edited by R. I. B. Francki Volume 2 • Edited by M. H. V. Van Regenmortel and Heinz Fraenkel-Conrat Volume 3 • Edited by Renate Koenig Volume 4 • Edited by R. G. Milne Volume 5 • Edited by B. D. Harrison and A. F. Murant

THE REOVIRIDAE Edited by Wolfgang K. Joklik

THE RETROVIRIDAE Volumes 1-4 • Edited by Jay A. Levy

THE RHABDOVIRUSES Edited by Robert R. Wagner

THE VIRUSES COMPLETE LISTING OF VOLUMES

THE TOGA VIRIDAE AND FLA VIVIRIDAE Edited by Sondra Schlesinger and Milton J. Schlesinger

THE VIROIDS Edited by T. O. Diener

THE VIRUSES: Catalogue, Characterization, and Classification Heinz Fraenkel-Conrat

The Plant Viruses: Polyhedral Virions and Bipartite RNA Genomes

Documents

Transcript of The Plant Viruses: Polyhedral Virions and Bipartite RNA Genomes