Comparative Plant Virology, Second Edition

COMPARATIVEPLANT VIROLOGYSECOND EDITION

ELSEVIER science &technology booksCompanion Web Site:http://www.elsevierdirect.com/companions/9780123741547Comparative Plant Virology, Second Edition,by Roger HullRevision to Fundamentals of Plant Virology written by R. MatthewsResources for Professors:• Image bank• Virus profilesTOOLS ALLTEACHINGFOR YOURNEEDStextbooks.elsevier.comACADEMICPRESSTo adopt this book for course use, visit http://textbooks.elsevier.com

COMPARATIVEPLANTVIROLOGYSECOND EDITIONROGER HULLEmeritus FellowDepartment of Disease and Stress BiologyJohn Innes CentreNorwich, UKAMSTERDAM • BOSTON • HEIDELBERG • LONDONNEW YORK • OXFORD • PARIS • SAN DIEGOSAN FRANCISCO • SINGAPORE • SYDNEY • TOKYOAcademic Press is an imprint of Elsevier

Cover Credits:BSMV leaf — Mild stripe mosaic; Symptom of BSMV in barley. Image courtesy ofA.O. Jackson.BSMV genome: The infectious genome (BSMV) is divided between 3 species of positivesense ssRNA that are designated a, b, and g. Image courtesy of Roger Hull.BSMV particles. Image courtesy of Roger Hull.Diagram showing systemic spread of silencing signal: The signal is generated in the initiallyinfected cell (bottom, left hand) and spreads to about 10–15 adjacent cells where it is amplified.It moves out of the initially infected leaf via the phloem sieve tubes and then spreadsthroughout systemic leaves being amplified at various times. Image courtesy of Roger Hull.Elsevier Academic Press30 Corporate Drive, Suite 400, Burlington, MA 01803, USA525 B Street, Suite 1900, San Diego, California 92101-4495, USA84 Theobald’s Road, London WC1X 8RR, UKThis book is printed on acid-free paper.Copyright # 2009, Elsevier Inc. All rights reserved.No part of this publication may be reproduced or transmitted in any form or by anymeans, electronic or mechanical, including photocopy, recording, or any informationstorage and retrieval system, without permission in writing from the publisher.Permissions may be sought directly from Elsevier’s Science & Technology RightsDepartment in Oxford, UK: phone: (þ44) 1865 843830, fax: (þ44) 1865 853333, E-mail:permissions@elsevier.co.uk. You may also complete your request on-line via the Elsevierhomepage (http://elsevier.com), by selecting “Customer Support” and then “ObtainingPermissions.”Library of Congress Cataloging-in-Publication DataHull, Roger, 1937-Comparative plant virology / Roger Hull. –2nd ed.p. cm.ISBN 978-0-12-374154-7 (hardcover : alk. paper) 1. Plant viruses. I. Title.QR351.H85 2009579.2 0 8–dc222008040333British Library Cataloguing in Publication DataA catalogue record for this book is available from the British LibraryISBN 13: 978-0-12-374154-7For all information on all Elsevier Academic Press publicationsvisit our Web site at www.elsevierdirect.comPrinted in China09 10 9 8 7 6 5 4 3 2 1

ContentsPreface xiiiList of AbbreviationsxvSection IINTRODUCTION TO PLANTVIRUSESChapter 1. What Is a Virus?I. Introduction 3II. History 3III. Definition of a Virus 9A. How Viruses Differ from Other PlantPathogens 9B. Are Viruses Alive? 13IV. Classification and Nomenclature of Viruses 13A. Virus Classification 13B. Families, Genera, and Species 14C. Naming Viruses (Species) 15D. Acronyms or Abbreviations 16E. Plant Virus Classification 17F. Virus Strains 17G. Use of Virus Names 19V. Viruses of Other Kingdoms 20VI. Summary 21Chapter 2. Overview of Plant VirusesI. Introduction 23II. Economic Losses Due to Plant Viruses 24III. Virus Profiles 24IV. Macroscopic Symptoms 25A. Local Symptoms 25B. Systemic Symptoms 261. Effects on Plant Size 262. Mosaic Patterns and Related Symptoms 263. Yellow Diseases 284. Leaf Rolling 285. Ring Spot Diseases 286. Necrotic Diseases 287. Developmental Abnormalities 288. Wilting 299. Recovery from Disease 2910. Genetic Effects 29C. The Cryptoviruses 29D. Diseases Caused by Viral Complexes 29E. Agents Inducing Virus-Like Symptoms 30V. Histological Changes 30A. Necrosis 30B. Hypoplasia 30C. Hyperplasia 321. Cell Size 322. Cell Division in Differentiated Cells 323. Abnormal Division in Cambial Cells 32VI. Cytological Effects 32A. Effects on Cell Structures 321. Nuclei 322. Mitochondria 333. Chloroplasts 334. Cell Walls 335. Cell Death 34B. Virus-Induced Structures in theCytoplasm 341. Accumulations of Virus Particles 342. Aggregates of Virus-Encoded Proteins 353. Caulimovirus Inclusions 35C. Why Inclusion Bodies? 37D. Cytological Structures 37VII. The Host Range of Viruses 38A. Limitations in Host Range Studies 38B. Patterns of Host Range 39v

viC. The Determinants of Host Range 391. Initial Events 392. Expression and Replication 393. Cell-to-Cell Movement 404. Stimulation of Host-Cell Defences 40VIII. Viruses in Other Kingdoms 40IX. Summary 40Chapter 3. Agents That Resemble or AlterPlant Virus DiseasesI. Viroids 43A. Classification of Viroids 44B. Pathology of Viroids 441. Macroscopic Disease Symptoms 442. Cytopathic Effects 443. Location of Viroids in Plants 454. Movement in the Plant 455. Transmission 456. Epidemiology 45C. Properties of Viroid RNAs 451. Sequence and Structure 452. Replication 473. Recombination Between Viroids 494. Interference Between Viroids 49D. Molecular Basis for Biological Activity 50E. Diagnostic Procedures for Viroids 50II. Phytoplasma 50III. Satellite Viruses and Satellite RNAs 51A. Satellite Plant Viruses (A-type) 52B. Satellite RNAs (satRNAs) 531. Large Satellite RNAs (B-type) 532. Small Linear Satellite RNAs (C-type) 533. Small Circular Satellite RNAs (D-type) 544. Satellite-Like RNAs 55a. A Satellite RNA of Groundnut RosetteVirus (GRV) 55b. Ancillary RNAs of Beet Necrotic YellowVein Virus (BNYVV) 565. The Molecular Basis for SymptomModulation 56C. Satellite DNAs 57D. Discussion 58IV. Defective and Defective Interfering NucleicAcids 58A. Group 1: Single Deletion D RNAs 60B. Group 2: Multiple Deletion D and DI RNAs 60C. Defective DNAs Associated with DNAViruses 60V. Viruses of Other Kingdoms 60VI. Summary 61CONTENTSChapter 4. Plant Virus Origins andEvolutionI. Introduction 63II. Virus Evolution 64A. Origins of Viruses 64B. Virus Variation 65C. Types of Evolution 651. Microevolution and Macroevolution 652. Sequence Divergence orConvergence 673. Modular Evolution 674. Sources of Viral Genes 67a. Replicases 67b. Proteinases 68c. Coat Proteins 70d. Cell-to-Cell Movement Proteins 71e. Suppressors of Gene Silencing 71D. Selection Pressures for Evolution 711. Adaptation to Niches 712. Maximising the Variation 713. Controlling the Variation 72a. Muller’s Ratchet 73b. Muller’s Ratchet and PlantViruses 734. Role of Selection Pressure 735. Selection Pressure by HostPlants 74E. Timeline for Evolution 741. Nonconstant Rates of Evolution 742. Estimated Rates of Evolution 74III. Evidence for Virus Evolution 75A. Geminiviruses 75B. Closteroviruses 75C. Luteoviruses 75IV. Coevolution of Viruses with Their Hosts andVectors 80V. Viruses of Other Kingdoms 80VI. Summary 80Section IIWHAT IS A VIRUS MADE OF?Chapter 5. Architecture and Assembly ofVirus ParticlesI. Introduction 85II. Methods 86

CONTENTSviiA. Chemical and Biochemical Studies 86B. Methods for Studying Size and Fine Structureof Viruses 861. Hydrodynamic Measurements 862. Electron Microscopy 873. X-Ray Crystallography 874. Neutron Small-Angle Scattering 875. Atomic Force Microscopy 876. Mass Spectrometry 887. Serological Methods 888. Stabilising Bonds 88III. Architecture of Rod-Shaped Viruses 88A. Introduction 88B. Structure of TMV 891. General Features 892. Virus Structure 90C. Assembly of TMV 921. Properties of the Coat Protein 922. Assembly of TMV Coat Protein 923. Assembly of the TMV Rod 92a. Assembly in vitro 92b. Assembly in vivo 94IV. Architecture of Isometric Viruses 94A. Introduction 94B. Possible Icosahedra 94C. Clustering of Subunits 97D. Quasiequivalence 97V. Small Icosahedral Viruses 97A. Subunit Structure 97B. Virion Structure 981. T ¼ 1 Particles 982. Other Particles Based on T ¼ 1Symmetry 98a. Bacilliform Particles Based on T ¼ 1Symmetry 98b. Geminiviruses 993. T ¼ 3 Particles 99a. Bacilliform Particles Based on T ¼ 3Symmetry 100b. Pseudo T ¼ 3 Symmetry 1004. T ¼ 7 Particles 100C. The Arrangement of Nucleic Acid WithinIcosahedral Viruses 1001. RNA Structure 1002. Interactions Between RNA and Protein inSmall Isometric Viruses 100D. Stabilisation of Small Isometric Particles 1011. Protein-RNA Stabilisation 1012. Protein-Protein Stabilisation 1013. Protein-Protein þ Protein-RNAStabilisation 101VI. More Complex Isometric Viruses 101VII. Enveloped Viruses 101VIII. Assembly of Icosahedral Viruses 102A. Bromoviruses 102B. RNA Selection During Assembly of PlantReoviruses 102IX. General Considerations 103X. Viruses of Other Kingdoms 104XI. Summary 104Chapter 6. Plant Viral GenomesI. Introduction 105II. General Properties of Plant ViralGenomes 105A. Information Content 106B. Economy in the Use of Genomic NucleicAcids 106C. The Functions of Viral Gene Products 1071. Functional Proteins 107a. Proteins That Initiate Infection 107b. Proteins That Replicate the ViralGenome 108c. Proteins That Process Viral GeneProducts 108d. Proteins That Facilitate Viral MovementThrough the Host 109e. Overcoming Host Defence Systems 109f. Proteins That Facilitate the Host to HostMovement of Viruses 109D. Nucleic Acids 1091. Multipartite Genomes 1092. Nucleic Acid Structures 1093. Noncoding Regions 109a. End-Group Structures 109b. 5 0 and 3 0 Noncoding Regions 112c. Intergenic Regions 112III. Plant Viral Genome Organisation 112A. Structure of the Genome 112B. Recognising Activities of ViralGenes 1141. Location of Spontaneous or ArtificiallyInduced Mutations 1142. Recombinant Viruses 1153. Expression of the Gene in a TransgenicPlant 1154. Hybrid Arrest and Hybrid SelectionProcedures 1155. Sequence Comparison with Genes of KnownFunction 1166. Functional Regions Within a Gene 116

viiiCONTENTSIV. Viruses of Other Kingdoms 116V. Summary 116Chapter 7. Expression of Viral GenomesI. Stages in Virus Infection Cycle 117II. Virus Entry and Uncoating 119A. Virus Entry 119B. Uncoating 1191. Uncoating of TMV 1192. Uncoating of Brome Mosaic Virus andSouthern Bean Mosaic Virus 1193. Uncoating of Turnip Yellow MosaicVirus 1214. Uncoating Other Plant Viruses 122III. Initial Translation of Viral Genome 122IV. Synthesis of mRNAs 123A. Negative-Sense Single-Stranded RNAViruses 123B. Double-Stranded RNA Viruses 123C. DNA Viruses 1241. Caulimoviridae 1242. Geminiviridae 125V. Plant Viral Genome Strategies 125A. The Eukaryotic Translation SystemConstraints 125B. Virus Strategies to Overcome EukaryoticTranslation Constraints 1261. Strategy 1. Polyproteins 1262. Strategy 2. Subgenomic RNAs 1293. Strategy 3. Multipartite Genomes 1314. Strategy 4. Splicing 1315. Strategy 5. Translation of Both Viral andComplementary Strands (Ambisense) 1316. Strategy 6. Internal Initiation 1317. Strategy 7. Leaky Scanning 133a. Two Initiation Sites on One ORF(Two Start) 133b. Overlapping ORFs 133c. Two or More Consecutive ORFs 1338. Strategy 8. Non-AUG Start Codon 1339. Strategy 9. Transactivation 13310. Strategy 10. Translational (Ribosome)Shunt 13411. Strategy 11. Read-Through Proteins 13412. Strategy 12. Frameshift Proteins 135C. Control of Translation 1361. Cap but No Poly(A) Tail 1362. Poly(A) Tail but No Cap 1363. Neither Cap nor Poly(A) Tail 1364. Cap Snatching 1365. 5 0 UTR 137D. Discussion 137VI. Viruses of Other Kingdoms 137VII. Summary 137Chapter 8. Virus ReplicationI. Host Functions Used by Plant Viruses 139II. Methods for Studying Viral Replication 140III. Replication of Plus-Sense Single-Stranded RNAViruses 140A. Viral Templates 140B. Replicase 1431. RNA-Dependent RNA Polymerase 1432. Helicases 1433. Methyl Transferase Activity 1444. Organisation of Functional Domains in ViralORFs 144C. Sites of Replication 146D. Mechanism of Replication 147E. Discussion 147IV. Replication of Negative-Sense Single-StrandedRNA Viruses 152V. Replication of Double-Stranded RNAViruses 152VI. Replication of Reverse TranscribingViruses 153A. Introduction 153B. Reverse Transcriptase 154C. Replication of “Caulimoviruses” 1541. Replication Pathway 1542. Inclusion Bodies 155VII. Replication of Single-Stranded DNAViruses 156A. Geminivirus Replication 156B. Geminivirus Rep Proteins 156VIII. Faults in Replication 158A. Mutation 158B. Recombination 1591. DNA Virus Recombination 1592. RNA Virus Recombination 1593. Recombination and Integrated ViralSequences 161IX. Viruses of Other Kingdoms 161X. Summary 163

Section IIIHOW DO PLANT VIRUSESWORK?Chapter 9. Virus-Host Interactions — PlantLevelI. Movement and Final Distribution 167A. Intracellular Movement 168B. Intercellular Movement 1681. Plasmodesmata 1682. Movement Proteins (MPs) 1693. What Actually Moves 1754. Cell-to-Cell Movement of Viroids 1755. Complementation 1766. Rate of Cell-to-Cell Movement 176C. Systemic Movement 1761. Steps in Systemic Movement 1762. Form in Which Virus Is Transported 1793. Rate of Systemic Movement 1794. Movement in the Xylem 180D. Final Distribution in the Plant 180E. Outstanding Questions on Plant VirusMovement 181II. Effects on Plant Metabolism 181A. Nucleic Acids and Proteins 181B. Lipids 182C. Carbohydrates 182D. Photosynthesis 184E. Respiration 184F. Transpiration 184G. Low-Molecular-Weight Compounds 184III. Processes Involved in Symptom Production 185A. Sequestration of Raw Materials 185B. Effects on Growth 186C. Effects on Chloroplasts 186D. Mosaic Symptoms 186E. Role of Membranes 187IV. Other Kingdoms 188V. Summary 188Chapter 10. Virus-Plant Interactions:1. Molecular LevelI. Introduction 191II. Host Responses to Inoculation 192A. Immunity 192CONTENTSB. Subliminal Infection 195C. Nonpermissive Infection 1951. Local Infection 195a. Host Protein Changes in the HypersensitiveResponse 197b. Local Acquired Resistance 1982. Systemic Infection 1983. Systemic Acquired Resistance 1984. Programmed Cell Death 200D. Permissive Infection 2001. Systemic Host Response 2002. Virus Genes Involved 200III. Interactions Between Viruses 202A. Interactions Between Related Viruses 202B. Interactions Between Unrelated Viruses 2031. Complete Dependence for Disease 2032. Incomplete Dependence for Disease 2033. Synergistic Effects on VirusReplication 2034. Effects on Virus Movement 203C. Interactions Between Viruses and Other PlantPathogens 203IV. Viruses of Other Kingdoms 204V. Summary 204Chapter 11. Virus-Plant Interactions:2. RNA SilencingI. Introduction 207II. Mechanism of Silencing 208A. The Basic Pathway 208B. Components of the System 2081. dsRNA 2082. Dicer 2093. Products of Dicer 2094. RISC 211C. Results of the System 211III. Systemic Silencing 211IV. Overcoming Silencing 211A. Suppression of Silencing 2111. Protein Suppressors of Silencing 2112. Nucleic Acid Suppressors ofSilencing 214B. Avoidance of Silencing 214V. Silencing and Symptoms 214A. Recovery 215B. Dark Green Islands and Mosaics 216C. miRNA 216D. siRNA Effects 216ix

xCONTENTSE. Synergistic Effects 216F. Other Activities of SilencingSuppressors 217VI. Transcriptional and Translational Repression 218VII. Evolutionary Aspects 218VIII. RNA Silencing in Animal and OtherViruses 218IX. Summary 218Section IVPLANT VIRUSES INAGRICULTURE AND INDUSTRYChapter 12. Plant-to-Plant MovementI. Introduction 223II. Transmission via Plant Material 223A. Mechanical Transmission 223B. Seed Transmission 224C. Pollen Transmission 225D. Vegetative Transmission 225E. Grafting 225III. Transmission by Invertebrates 225A. Relationships Between Plant Viruses andInsects 228B. Nonpersistent Transmission by Insects 2311. Features of NonpersistentTransmission 2312. Virus-Vector Relationships 231a. Direct Capsid Interaction 232b. Indirect Interaction Involving HelperComponents 232C. Persistent Transmission by Insects 2351. Circulative Viruses 235a. Features of Circulative Virus: VectorInteraction 235b. Dependent Transmission 2372. Propagative Viruses 2373. Thrip Transmission of Tospoviruses 238D. Virus Transmission by Beetles 238E. Nematode Transmission of Viruses 2391. Features of Nematode Transmission 2392. Virus-Nematode Relationships 239IV. Fungal Transmission of Viruses 240V. Viruses of Other Kingdoms 242VI. Summary 242Chapter 13. Plant Viruses in the Field: Diagnosis,Epidemiology, and EcologyI. Diagnosis 245A. Introduction 245B. Methods Involving Biology of the Virus 2461. Indicator Hosts 2462. Host Range 2463. Methods of Transmission 2474. Cytological Effects 2475. Mixed Infections 247C. Methods That Depend on Physical Propertiesof the Virus Particle 2471. Stability and PhysicochemicalProperties 2472. Electron Microscopy 247D. Methods That Depend on Properties of ViralProteins 2491. Serology 2492. Types of Antisera 2493. Methods for Detecting Antibody-VirusCombination 249a. ELISA Procedures 249b. Serologically Specific ElectronMicroscopy 255c. Electrophoretic Procedures 256d. Dot Blots 256E. Methods That Involve Properties of the ViralNucleic Acid 2561. Type and Size of Nucleic Acid 2562. Cleavage Patterns of DNA 2573. Hybridization Procedures 2574. Dot Blots 2575. Polymerase Chain Reaction 2576. DNA Microarray 259F. Decision Making on Diagnosis 260II. Epidemiology and Ecology 260A. Epidemiology of Viruses in Agriculture 2611. Primary Infections 2612. Secondary Spread 264B. Plant Viruses in the Natural Environment 267C. Emergence of New Viruses 267III. Viruses of Other Kingdoms 268IV. Summary 268Chapter 14. Conventional ControlI. Introduction 269II. Avoiding Infection 271

CONTENTSxiA. Removal of Sources of Infection 271B. Virus-Free Seed 271C. Virus-Free Vegetative Stocks 271D. Modified Agronomic Practices 272E. Quarantine Regulations 273III. Stopping the Vector 274A. Air-Borne Vectors 2741. Insecticides 2752. Insect Deterrents 2753. Agronomic Techniques 275B. Soil-Borne Vectors 2771. Nematodes 2772. Fungi 277IV. Protecting the Plant 277A. Protection by a Plant Pathogen 277B. Antiviral Chemicals 278V. Conventional Resistance to Plant Viruses 280A. Introduction 280B. Genetics of Resistance to Viruses 281C. Tolerance 281D. Use of Conventional Resistance forControl 2821. Immunity 2822. Field Resistance 2823. Tolerance 282VI. Strategies for Control 283VII. Viruses of Other Kingdoms 283VIII. Summary 283Chapter 15. Transgenic Plants and VirusesI. Transgenic Protection Against Plant Viruses 285A. Introduction 285B. Natural Resistance Genes 285II. Pathogen-Derived Resistance 286A. Protein-Based Protection 2861. Transgenic Plants Expressing a Viral CoatProtein 2862. Other Viral Proteins 286B. Nucleic Acid–Based Protection 2871. RNA-Mediated Protection 2882. Molecular Basis of RNA-MediatedProtection 2883. Sequences for RNA-MediatedProtection 2894. Ribozymes 2895. Relationship Between NaturalCross-Protection and Protection inTransgenic Plants 2896. Transgenic Protection by Satellite and DINucleic Acids 290C. Other Forms of Transgenic Protection 290D. Field Releases of Transgenic Plants 2901. Potential Risks 2902. Field Performance 293III. Possible Uses of Plant Viruses for GeneTechnology 293A. DNA Viruses as Gene Vectors 2931. Caulimoviruses 2942. Geminiviruses 294B. RNA Viruses as Gene Vectors 294C. Viruses as Sources of Control Elements forTransgenic Plants 2951. DNA Promoters 2952. RNA Promoters 2953. Translation Enhancers 295D. Viruses for Producing Vaccines 2951. Vaccines Using Plant Virus Vectors 2962. Viruses for Presenting HeterologousPeptides 296a. Cowpea Mosaic Virus (CPMV) 296b. Tobacco Mosaic Virus (TMV) 298E. Viruses in Functional Genomics ofPlants 298F. Plant Viruses in Nanotechnology 298IV. Viruses of Other Kingdoms 300V. Summary 301Appendix: Profiles 303Index 363

This page intentionally left blank

PrefaceThis book has been developed from Fundamentalsof Plant Virology written by R. E. F.Matthews in 1992. Since then major advanceshave been made in the understanding of themolecular biology of viruses, how they functionand how they interact with their hosts.This has revealed similarities and differencesbetween viruses infecting members of the differentkingdoms of living organisms, plants,animals, fungi, and bacteria. In this changingenvironment of teaching virology, this bookdoes not just deal with plant viruses alone butplaces them in context in relation to viruses ofmembers of other kingdoms.This book has been written for students ofplant virology, plant pathology, virology, andmicrobiology who have no previous knowledgeof plant viruses or of virology in general.An elementary knowledge of molecular biologyis assumed, especially of the basic structures ofDNAs, RNAs, and proteins, of the genetic code,and of the processes involved in protein synthesis.As some of these students may not havea grounding in the structure and function inplants including the main subcellular structuresfound in typical plant cells, these featureswhich are important to the understanding ofhow viruses interact with plants are illustrated.In each chapter there is a list of further readingto enable the student to explore specific topicsin depth.The fifteen chapters in this book can bedivided into four major sections that form alogical progression in gaining an understandingof the subject. The first four chapters introduceplant viruses describing: what is a virus,giving an overview of plant viruses, discussingother agents that cause diseases that resembleplant virus diseases, and considering factorsthat are involved in virus evolution. The pointsraised in this latter chapter are equally relevantto viruses of other kingdoms. The next fourchapters deal with what viruses are made of.The chapter on virus architecture and assemblyis also very relevant to viruses of other kingdomsas are the major points raised in chapterson plant virus genome organization, genomeexpression, and genome replication. The nextsection on how do plant viruses work is morespecific to plant viruses and highlights differencesand similarities between virus interactionswith plant, animal, and bacterial hosts.xiii

xivPREFACEThese interactions are described at the plantlevel (movement of the virus within the plantand effects on plant metabolism) and at themolecular level including a chapter devoted tothe newly understood host defence system ofRNA silencing. The last four chapters deal withplant viruses and agriculture and industry. Thedescription of how plant viruses move betweenhosts which often involves specificmolecular interactions leads into discussion ofthe epidemiology of viruses in the field andhow they are controlled. The last chapter is onthe use of recombinant DNA technology incontrolling viruses and also in using them commerciallyin, for instance, the pharmaceuticaland nanotechnology industries.A unique feature of this book is a series of“profiles” on 32 plant viruses that feature in thetext. These profiles describe briefly the majorproperties of the viruses including their taxonomicposition, their biology, their particles,and their genomes. References are given toenable students to acquire even more informationon these targeted viruses.I am very grateful to a large number of colleaguesfor their helpful discussion on varioustopics and in providing material prior to publication.I am especially indebted to John Carr, AndyJackson, Mark Stevens, and Peter Waterhouse fortheir helpful comments on various sections of thebook and on providing illustrative material. Myeternal gratitude goes to my wife who hastolerated “piles of paper” around the house andwho has given me continuous encouragement.Roger HullNorwich, UKJuly, 2008

List of Abbreviations3’OHÅAABDPVADPARATPcDNACIDdDpD RNA/DNADI RNA/DNAdsDNAdsRNAEDTAELISAEMERGTPGMHC-ProHEL3’ hydroxyl groupAngström (10 10 meter)Association of Applied BiologistsDescriptions of Plant VirusesAdenosine diphosphateAberrante ratioAdenosine triphosphateComplentary (or copy) DNACylindrical inclusionDNA-dependent DNApolymeraseDefective RNA or DNADefective interfering RNA orDNADouble-stranded DNADouble-stranded RNAEthylene diamino tetra-acetic acidEnzyme-linked inmuno-sorbentassayElectron MicroscopeEndoplasmic reticulumGuanosine triphosphateGenetically modifiedHelper component proteaseHelicaseHRHSPICRICTVIRESISEMkbkDaLRRMabMPmRNAMiRNAMTRMWNBSNInmORFPCDPCRPDRHypersensitive responseHeat-shock proteinInter-cistronic regionInternational Committee on theTaxonomy of VirusesInternal ribosome entry siteImmunoabsorbent electronmicroscopyKilobaseKilodaltonLeucine-rich repeatMonoclonal antibodyMovement proteinMessenger RNAMicroRNAMethyl transferaseMolecular weightNucleotide binding siteNuclear inclusionNanometerOpen reading frameProgrammed cell deathPolymerase chain reaction (IC-PCR, immune-capture PCR; RT-PCR, reverse transcription PCR)Pathogen-derived resistancexv

xviLIST OF ABBREVIATIONSPolPR(protein)PROPTGSRb(protein)RdRpRFRFLPRIRISCRNAiRTSASARPolymerasePathogenesis-related (protein)ProteasePost-transcriptional genesilencingRetinoblastoma (protein)RNA-directed RNA PolymeraseReplicative formRestriction fragment lengthpolymorphismReplicative intermediateRNA-induced silencing complexRNA interferingReverse transcriptaseSalicylic acidSystemic acquired resistanceSDS-PAGESELSg RNASi RNAssDNASSEMssRNATAVTGBTGStRNAUTRVIGSVPgSodium dodecyl sulphatepolyacrylamide gelelectrophoresisSize exclusion limitSubgenomic RNASmall interfering RNASingle-stranded DNASerologically-specific electronmicroscopySingle-stranded RNATransactivatorTriple gene blockTranscriptional gene silencingTransfer RNAUntranslated regionVirus-induced gene silencingVirus protein genome-linked

S E C T I O N IINTRODUCTION TO PLANTVIRUSES

This page intentionally left blank

C H A P T E R1What Is a Virus?This chapter discusses broad aspects of virology and highlights how plant viruses have led thesubject of virology in many aspects.O U T L I N EI. Introduction 3II. History 3III. Definition of a Virus 9IV.Classification and Nomenclatureof Viruses 13V. Viruses of Other Kingdoms 20VI. Summary 21I. INTRODUCTIONPlant viruses are widespread and economicallyimportant plant pathogens. Virtually allplants that humans grow for food, feed, andfiber are affected by at least one virus. It is theviruses of cultivated crops that have been moststudied because of the financial implications ofthe losses they incur. However, it is also importantto recognise that many “wild” plants arealso hosts to viruses. Although plant virusesdo not have an immediate impact on humansto the extent that human viruses do, the damagethey do to food supplies has a significant indirecteffect. The study of plant viruses has ledthe overall understanding of viruses in manyaspects.II. HISTORYAlthough many early written and pictorialrecords of diseases caused by plant virusesare available, they are do not go back as far asrecords of human viruses. The earliest knownwritten record of what was very likely a plantComparative Plant Virology, Second Edition3Copyright # 2009, Elsevier Inc. All rights reserved.

4 1. WHAT IS A VIRUS?virus disease is a Japanese poem that was writtenby the Empress Koken in A.D. 752 andtranslated by T. Inouye:In this villageIt looks as if frosting continuouslyFor, the plant I sawIn the field of summerThe colour of the leaves were yellowingThe plant, which has since been identified asEupatorium lindleyanum, has been found to besusceptible to Tobacco leaf curl virus, whichcauses a yellowing disease.In Western Europe in the period from about1600 to 1660, many paintings and drawingswere made of tulips that demonstrate flowersymptoms of virus disease. These are recordedin the Herbals of the time and some of the earliestin the still-life paintings of artists such asAmbrosius Bosschaert. During this period,blooms featuring such striped patterns wereprized as special varieties, leading to the phenomenonof “tulipomania” (Box 1.1).Because of their small genomes, viruses haveplayed a major role in elucidating many of theconcepts in molecular biology, and the study ofplant viruses has produced several of the majorfindings for virology in general. The major stepsin reaching the current understanding of virusesare shown in the timeline in Figure 1.1.Details of these “breakthroughs” can be foundin Hull (2002; plant viruses), Fenner, (2008; vertebrateviruses), and Ackermann (2008; bacterialviruses). Plant viruses played a major role indetermining exactly what a virus was. In the latterpart of the nineteenth century, the idea that infectiousdisease was caused by microbes was wellestablished, and filters were available that wouldnot allow the known bacterial pathogens to passthrough. In 1886, Mayer (see Figure 1.2A)described a disease of tobacco that he calledMosaikkrankheit, which is now known to be causedby the Tobacco mosaic virus (TMV). Mayer demonstratedthat the disease could be transmitted tohealthy plants by inoculation with extracts fromdiseased plants. A major observation was madein 1892 by Iwanowski, who showed that sap fromtobacco plants displaying the disease describedby Mayer was still infective after it had beenpassed through a bacteria-proof filter candle.However, based on previous studies, it wasthought that this agent was a toxin. Iwanowski’sexperiment was repeated in 1898 by Beijerinck(see Figure 1.2B), who showed that the agent multipliedin infected tissue and called it contagiumvivum fluidum (Latin for “contagious living fluid”)to distinguish it from contagious corpuscularagents (Figure 1.2C).Beijerinck and other scientists used the termvirus to describe the causative agents of suchtransmissible diseases to contrast them withbacteria. The term virus had been used moreor less synonymously with bacteria by earlierworkers, but as more diseases of this sort werediscovered, the unknown causative agentscame to be called “filterable viruses.” Similarproperties were soon after reported for someviruses of animals (e.g., the filterable nature ofBOX 1.1TULIPOMANIATulips were introduced into the Netherlands in the late sixteenth century. Bulbs that produced ”brokencoloured”flowers were in great demand and created a rapidly expanding market, leading tohyperinflation.(continued)I. INTRODUCTION TO PLANT VIRUSES

II. HISTORY5BOX 1.1(continued)Semper Augustus tulip with flower colour break (one of the most favoured varieties)One bulb cost 1,000 Dutch florins (guilders) in 1623, and by 1635, 6,000 florins. To understand thevalue of this, one Viceroy tulip bulb was exchanged for goods that were valued at almost 2,400 florins:4 tons of wheat (448 florins) 4 barrels of beer (3 florins)8 tons of rye (558 florins) 2 barrels of butter (192 florins)4 fat oxen (480 florins) 1,000 lbs cheese (120 florins)8 fat pigs (240 florins) 1 bed with accessories (100 florins)12 fat sheep (120 florins) 1 silver goblet (60 florins)2 hogsheads of wine (70 florins)By 1636 there was much speculation, and futures were being taken out on these bulbs. In early1637 one bulb was valued at 10,000 florins, but a few weeks later, the bubble burst and many peoplewere left bankrupt. It was not until the 1920s that the viral aetiology of tulip flower breaking wasdiscovered and that the symptoms were caused by an aphid-transmitted potyvirus. Today, 100 florinsis equivalent to about U.S. $30,000.I. INTRODUCTION TO PLANT VIRUSES

6 1. WHAT IS A VIRUS?FIGURE 1.1Plant752 AD Plant virus in Japanesepoem1600-1637 Tulipomania1886 Meyer Transmission ofTMV1892 Iwanowski Filterability ofTMV1898 Beijerink Viruses as anentity1900-1935 Descriptions ofmany viruses1935 Purification of TMV1936 TMV contains pentosenucleic acid1939 EM TMV rod-shapedparticles1951 TYMV RNA in proteinshell1956 Virus particles made ofidentical protein subunit1955/56 Infectious nature ofTMV RNA1962 Structure of isometricparticles1983 Structure of TBSV to 2.9Å1960 Sequence of TMV coatprotein1980 Sequence of CaMV DNAgenome1982 Sequence of TMV RNAgenome1984 Infectious transcripts ofmulticomponent BMV1986 Transgenic protection ofplants against TMV1996 Recognition of RNAsilencing1997 Recognition of virussuppressors of silencingAnimalPrehistory1350 BC Smallpox recorded in Egypt1796 Jenner developed smallpox vaccineRecognition of viral entity1898 Filterability of PV and FMDVBiological age1900– Descriptions of many viruses1901 Mosquito transmission of YFVBiophysical/biochemical age1940 VACV contains DNA1949 PV grown in cultured cells1985 Structure of poliovirus to 2.9ÅMolecular age1979 Sequence of PV VPg1970 Recognition of reverse transcriptase1981 Infectious transcript of PV1981 Sequence of poliovirus RNAgenomeBacteria1915. Filterability of phage1915- Descriptions of manyvirusesEarly 1920s Infection cycleunderstood1940-1970 Phage genetics1978 Infectious transcript ofQβAbbreviations: BMV, Brome mosaic virus; CaMV, Cauliflower mosaic virus; FMDV, Foot and mouthdisease virus; PV, Poliovirus; TBSV, Tomato bushy stunt virus; TMV, Tobacco mosaic virus; TYMV,Turnip yellow mosaic virus; VACV, Vaccinia virus; YFV, Yellow fever virus.Timeline of development of virology.I. INTRODUCTION TO PLANT VIRUSES

II. HISTORY7FIGURE 1.2 A. Adolf EduardMayer (1843–1942); B. Martinus WillemBeijerinck (1851–1931); C. Page from labjournal of W.M. Beijerinck from 1898relating to TMV. A and B courtesy ofthe historical collection, AgriculturalUniversity, Wageningen, Netherlands;C. (# Kluyver Institute) Courtesy CuratorKluyver Laboratory Collection,Delft School of Microbiology Archive,Delft University of Technology.ABCI. INTRODUCTION TO PLANT VIRUSES

8 1. WHAT IS A VIRUS?the agent causing foot and mouth disease in1898) and of bacteria in 1915. Over the courseof time, the word filterable has been dropped,leaving just the term virus.As shown in the timeline in Figure 1.1, in thesubsequent development of virology, many ofthe studies ran in parallel for viruses of plants,vertebrates, invertebrates, and bacteria. In fact,when viewed overall, there is evidence of muchcross-feeding between the various branches ofvirology. However, there were differencesmainly due to the interactions that theseviruses have with their hosts. For instance, vertebratesproduce antibodies that counterviruses, whereas plants, invertebrates, and bacteriado not. Another factor that has contributedto advances is the simplicity of the systemexemplified by studies on bacteriophage beinglinked to studies on bacterial genetics.The development of plant, and other, virologycan be considered to have gone through fivemajor (overlapping) ages. The first two, Prehistoryand Recognition of viral entity, were justdescribed. After these two came the Biologicalage, between 1900 and 1935, when it was determinedthat plant viruses were transmitted byinsects and that some of these viruses multipliedin, and thus were pathogens of, insects ina manner similar to some viruses of vertebrates.One of the constraints to plant virology was thelack of a quantitative assay, until Holmes in1929 showed that local lesions produced insome hosts after mechanical inoculation couldbe used for the rapid quantitative assay of infectivevirus. This technique enabled properties ofviruses to be studied much more readily andpaved the way for the isolation and purificationof viruses a few years later.The Biochemical/Physical age started in theearly 1930s. The high concentration at whichcertain viruses occur in infected plants andtheir relative stability was crucial in the firstisolation and chemical characterisation ofviruses because methods for extracting andpurifying proteins were not highly developed.In 1935, Stanley announced the isolation of thisvirus in an apparently crystalline state but consideredthat the virus was a globulin containingno phosphorus. In 1936, however, Bawden andhis colleagues described the isolation fromTMV-infected plants of a liquid crystalline nucleoproteincontaining nucleic acid of the pentosetype. Around 1950, Markham and Smith showedthat the RNA of Turnip yellow mosaic virus wasencapsidated in a protein shell and was importantfor biological activity. This led to the classicexperiments of Gierer, Schramm, Fraenkel-Conrat,and Williams in the mid-1950s that demonstratedthe infectivity of naked TMV RNA andthe protective role of the protein coat.In parallel with these biochemical studies,physical studies in the late 1930s using X-rayanalysis and electron microscopy confirmed thatTMV had rod-shaped particles and obtainedaccurate estimates of the size of the rods. Attentionturned to the structure of these particles,and in 1956, Crick and Watson suggested thatthe protein coats of small viruses are made upof numerous identical subunits arrayed eitheras helical rods or as a spherical shell with cubicsymmetry. This led to Caspar and Klug (1962)formulating a general theory that delimited thepossible numbers and arrangements of the proteinsubunits forming the shells of the smallerisodiametric viruses (see Chapter 5). Our recentknowledge of the larger viruses with more complexsymmetries and structures has come fromelectron microscopy using negative-stainingand ultrathin-sectioning methods.The current Molecular age started in about1960 when the full sequence of 158 amino acidsin the coat protein of TMV was determined.The sequence of many naturally occurringstrains and artificially induced mutants wasalso determined at about the same time. Thiswork made an important contribution to establishingthe universal nature of the genetic codeand to our understanding of the chemical basisof mutation. This age continued with thesequencing of representatives of most, if notI. INTRODUCTION TO PLANT VIRUSES

III. DEFINITION OF A VIRUS9all, virus genera leading to a greater understandingof how viruses function and interactwith their hosts. The results from these studiesare described in detail in this book and in thesuggested further reading.III. DEFINITION OF A VIRUSA. How Viruses Differ from OtherPlant PathogensIn the size of their nucleic acids, virusesrange from a monocistronic mRNA in the satellitevirus of tobacco necrosis virus (STNV) to agenome larger than that of the smallest cells(Figure 1.3). A biologically more meaningfulway of comparing genome sizes is to considerthe information content—that is, the numberof genes that they contain; some examples aregiven in Table 1.1. Before attempting to definewhat viruses are, we must consider brieflyhow they differ from other entities such as cellularparasites, plasmids, and transposablegenetic elements. The three simplest kinds ofparasitic cells are the Mycoplasmas, the Rickettsiae,and the Chlamydiae.Mycoplasmas and related organisms are notvisible by light microscopy. They are 150–300 nmin diameter with a bilayer membrane but nocell wall, and they contain RNA along withribosomes and DNA. They replicate by binaryfission, and some that infect vertebrates can begrown in vitro. Their growth is inhibited by certainantibiotics. Some mycoplasmas are plantpathogenic (see Chapter 3).The Rickettsiae, for example, the agent oftyphus fever, are small, nonmotile bacteria,usually about 300 nm in diameter. They havea cell wall, plasma membrane, and cytoplasmwith ribosomes and DNA strands. They areobligate parasites and were once thought to berelated to viruses, but they are definitely cellsbecause they multiply by binary fission, andthey contain enzymes for ATP production.The Chlamydiae, for example, the agent thatcauses psittacosis, include the simplest knowntype of cell. They are obligate parasites thatgrow by infecting eukaryotic cells and lack anenergy-generating system. They are as smallas, or smaller than, many viruses. Chlamydiaehave two phases to their life cycle. Inside hostcells they take on an intracellular replicativeform (termed the reticulate body) and rely onthe host cell energy-yielding system; outsidethe cell they survive by forming infectious elementarybodies about 300 nm in diameter, whichis smaller than some pox viruses. Chlamydiaecan be grown only where their host cells growand cannot be propagated in bacterial culturemedia.Several criteria do and do not distinguish allviruses from all cells (see Table 1.2).Plasmids are autonomous extrachromosomalgenetic elements found in many kinds of bacteria.They consist of closed circular DNA. Some canbecome integrated into the host chromosomeand replicate with it. Some viruses that infect prokaryoteshave properties like those of plasmidsand, in particular, the ability to integrate into thehost cell chromosome. However, viruses differfrom plasmids in the following ways:1. Normal viruses have a particle with astructure designed to protect the geneticmaterial in the extracellular environmentand to facilitate entry into a new host cell.2. Virus genomes are highly organised forspecific virus functions of no known valueto the host cell, whereas plasmids consist ofgenetic material that is often useful for thesurvival of the cell.3. Viruses can kill cells or cause disease in thehost organism, but plasmids cannot.Transposons, or mobile genetic elements(sometimes called “jumping genes”), aresequences of DNA that can move around to differentpositions within the genome of a singlecell, a process termed transposition. Two types ofmobile genetic elements exist, based on theirI. INTRODUCTION TO PLANT VIRUSES

10 1. WHAT IS A VIRUS?ACFIGURE 1.3 Size comparison of different organisms. A. Organisms classified according to genome size. The vertical axisgives an approximate indication of numbers of species within the size range of each group. B. Size comparison among abacterium, several viruses, and a viroid. C. Comparison of size of rhinovirus and a pinhead. A. Modified from Hinegardner[1976; in Molecular Evolution, (F.J. Ayala, Ed.), pp. 179-199, Sinauer, Sunderland, MA]; B. With kind permission fromSpringer Science þ Business Media: Arch. Virol., Interference between proflavine treated reovirus and related and unrelatedviruses, vol. 15, 1965, pp. 200–2009, E. Zalan; Arch. Virol., Die Interferenz zwischen dem Polyoma-virus and dem Stomatitisvesicularis-Virusin der Maus, vol. 15, 1965, pp. 210-219, D. Falke; Arch. Virol., Properties of a new attenuated type 3 poliovirus,vol. 15, 1965, pp. 220-233, J. Šimon. C. http://web.uct.ac.za/depts/mmi/stannard/linda.html.Bmechanism of transposition. Class I mobilegenetic elements, or retrotransposons, move inthe genome by being transcribed to RNA and thenback to DNA by reverse transcriptase. Class IImobile genetic elements move directly from oneposition to another within the genome using atransposase to “cut and paste” them within thegenome. In many properties, retrotransposonsI. INTRODUCTION TO PLANT VIRUSES

III. DEFINITION OF A VIRUS11TABLE 1.1Information Content of Genomes of Various OrganismsType of OrganismExampleSize ofGenomeNumber of Genes(Open Reading Frames)Higher plant Rice 3.9 10 8 kbp >37,000Vertebrate Human 3.3 10 9 kbp 20,000–25,000Invertebrate Drosophila 1.2 10 8 kbp 13,400Yeast 1.2 10 7 kbp 5,770Eubacteria Escherichia coli 4.6 10 6 kbp 4,377Mycoplasma Mycoplasma genitalium 5.8 10 5 kbp 485Large virus infectingVaccinia virus 190 kbp 250vertebratesLarge virus infecting chlorellalikeParamecium bursarum Chlorella virus 1 330 kbp 697algaeLarge virus infectingAutographa californica multiple133.9 kbp 150invertebratesnucleopolyhedrosisSmall virus infecting angiosperms Tobacco mosaic virus 6395 nt 4Smallest known virus Tobacco necrosis satellite virus 1239 nt 1TABLE 1.2Criteria That DistinguishViruses from Cells1. Lack of continuousmembrane separatingvirus from host duringreplication2. Absence of proteinsynthesisingsystem3. Contain either DNA orRNA4. Replication is by synthesisof a pool of componentsand not by binary fissionDistinguishing Criteria for VirusesCriteria That Do NotDistinguish Virusesfrom Cells1. Size2. Nature and size ofgenome3. Contain both DNA andRNA4. Absence of rigid cellenvelope5. Obligate cell parasitism6. Absence of energyyieldingsystem7. Complete dependenceon host cell for aminoacidsresemble retroviruses, and they are classified asMetaviruses and Pseudoviruses. However, thereis debate as to whether these are really viruses inthe strictest sense. We can now define a virus, asshown in Box 1.2.To be identified positively as a virus, an agentmust normally be shown to be transmissible andto cause disease in at least one host. One of thebasic tenets of pathology is that to prove that a diseaseis caused by a certain infectious agent, onemust fulfill Koch’s postulates, which weredevised for bacteria; modifications of the postulateshave been suggested to account for specificproperties of viruses (Table 1.3). Today, however,it is not always possible to fulfill these postulatesfor viruses. For instance, plant cryptovirusesrarely cause detectable disease and are not transmissibleby any mechanism except through seedsor pollen. Usually, it is satisfactory to show a clearassociation of the viral genome sequence with thedisease after eliminating the possibility of jointinfection with another virus.I. INTRODUCTION TO PLANT VIRUSES

12 1. WHAT IS A VIRUS?BOX 1.2DEFINITION OF A VIRUSA virus is a set of one or more nucleic acid templatemolecules, normally encased in a protectivecoat or coats of protein or lipoprotein, thatis able to organise its own replication onlywithin suitable host cells. Within such cells,virus replication is (1) dependent on the host’sprotein-synthesising machinery, (2) organisedfrom pools of the required materials rather thanby binary fission, (3) located at sites that are notseparated from the host cell contents by a lipoproteinbilayer membrane, and (4) continuallygiving rise to variants through several kinds ofchange in the viral nucleic acid.TABLE 1.3Bacteria1. Demonstrate that theagent is regularly foundin the diseased host2. Cultivate the agent on asuitable medium3. Reproduce the disease inthe host by reintroducingthe cultured agent4. Reisolate the agent fromthe artificially infectedhosta Rivers (1937).Koch’s Postulates for Bacteriaand VirusesViruses a1. Isolation of virus fromdiseased host2. Cultivate virus inexperimental host or hostcells3. Prove lack of largerpathogens4. Produce comparabledisease in original hostspecies or in related ones5. Reisolate the virusThe structure and replication of viruses havethe following features.1. The infectious nucleic acid may be DNA orRNA (but never both) and be single- ordouble-stranded. If the nucleic acid is singlestrandedit may be of positive or negativesense. (Positive sense has the sequence thatwould be used as an mRNA for translationto give a virus-coded protein.)2. The mature virus particle may containpolynucleotides other than the genomicnucleic acid.3. Where the genetic material consists of morethan one nucleic acid molecule, each may behoused in a separate particle or all may belocated in one particle.4. The genomes of viruses vary widely in size,encoding between 1 and about 250proteins. Plant viral genomes are at thesmall end of this range, mostly encodingbetween 1 and 12 proteins. The plant viruscodedproteins may have functions in virusreplication, in virus movement from cell tocell, in virus structure, and in transmissionby invertebrates or fungi. Animal andbacterial viruses may contain more genesassociated with their interactions with theirhosts.5. Viruses undergo genetic change. Pointmutations occur with high frequency as aresult of nucleotide changes brought about byerrors in the copying process during genomereplication. Other kinds of genetic changemay be due to recombination, reassortment ofgenome pieces, loss of genetic material, oracquisition of nucleotide sequences fromunrelated viruses or the host genome.I. INTRODUCTION TO PLANT VIRUSES

IV. CLASSIFICATION AND NOMENCLATURE OF VIRUSES136. Enzymes specified by the viral genome maybe present in the virus particle. Most of theseenzymes are concerned with nucleic acidsynthesis.7. Replication of many viruses takes place indistinctive virus-induced structures in thecell.8. Some viruses share with certain nonviralnucleic acid molecules the property ofintegration into host-cell genomes andtranslocation from one integration site toanother.9. A few viruses require the presence ofanother virus for their replication.B. Are Viruses Alive?This question is asked very frequently. Thedefinitions of a living organism vary widely,with the most accepted one being “A livingorganism has cellular structure and is manifestby growth through metabolism, reproduction,and the power of adaptation to the environmentthrough changes that originate internally.”While viruses reproduce and adapt,they are not cellular and do not metabolise;they rely on their host cell metabolism. Thus,technically they are not living organisms andthe term virus life cycle should not be used; virusreplication cycle describes the making of a newvirus particle from an input particle.IV. CLASSIFICATIONAND NOMENCLATURE OFVIRUSESIn all studies of natural objects, humans seemto have an innate desire to name and to classifyeverything. It has been said that taxonomy is“the craft of making dead things look alive.”Virologists are no exception. Virus classification,as with all other classifications, arranges objectswith similar properties into groups, and eventhough this may be a totally artificial andhuman-driven activity without any natural base,it does have certain properties:• It gives a structured arrangement of theorganisms so that the human mind cancomprehend them more easily.• It helps with communication amongvirologists and between virologists andother interested parties.• It enables properties of new viruses to bepredicted.• It could reveal possible evolutionaryrelationships.In theory, it is possible to consider the problemsof naming and classifying viruses as separateissues. In practice, however, naming sooncomes to involve classification.From the 1930s to 1960s, various classificationsystems were proposed for plant (and other)viruses. This led to much confusion, and at theInternational Congress for Microbiology, held inMoscow in 1966, the first meeting of the InternationalCommittee for the Nomenclature of Viruseswas held. An organisation was set up for developingan internationally accepted taxonomy andnomenclature for all viruses. Rules for the nomenclatureof viruses were laid down. This committeedeveloped into the International Committee forTaxonomy of Viruses (ICTV), which has sinceproduced eight reports, the most recent beingFauquet et al. (2005). These reports give the definitivedescriptions of the various taxa of viruses.A. Virus ClassificationA detailed list of the criteria used for virusclassification and taxonomy is given in Murphyet al. (1995). The criteria come under four majorheadings: virion properties, such as size andshape, type of genome, properties of proteins;genome organisation and replication; antigenicproperties; and biological properties, such ashost range, pathogenicity, and transmission.I. INTRODUCTION TO PLANT VIRUSES

14 1. WHAT IS A VIRUS?A problem arises as to how much weight isput onto each character. In practice, the natureand sequence of the genomic nucleic acid arethe major characters that are used, but other properties,such as particle shape and composition,antigenic relationships, and biology, are alsoconsidered to be important. Any classificationof viruses should be based not only on evolutionaryhistory, as far as this can be determinedfrom the genotype, but should also beuseful in a practical sense. Most of the phenotypiccharacters used today in virus classificationwill remain important even when thenucleotide sequences of most viral genomeshave been determined.B. Families, Genera, and SpeciesThe main building block of a biological classificationis the species. In day-to-day practice,virologists use the concept of a “virus” asbeing a group of fairly closely related strains,variants, or pathovars. A virus defined in thisway is essentially a species in the sense suggestedfor angiosperms and defined by the ICTV.In 1991, the ICTV accepted the concept thatviruses exist as species, adopting the followingdefinition:A viral species is a polythetic class of viruses thatconstitutes a replicating lineage and occupies a particularecological niche. [Polythetic denotes a taxonomicgroup classified on the basis of several characters,as opposed to a monothetic group.]The species has formed the basis of modernvirus classification being established insubsequent ICTV reports, especially the seventhand eighth, in which a List of Species-DemarcatingCriteria is provided for each genus. Thisenables viruses to be differentiated as speciesand tentative species, which are viruses thathave not yet been sufficiently characterised toensure that they are distinct and not strains ofan existing virus or do not have the full characteristicsof the genus to which they have beenassigned. Of the 1,037 plant viruses listed inthe eighth ICTV report, 751 are true speciesand 286 are tentative species. Further studieswill provide enough data to classify the tentativespecies. A common problem is determiningwhether a new virus is truly a new species or astrain of an existing species. Conversely, whatwas considered to be a strain may, on furtherinvestigation, turn out to be a distinct species.This is due to the population structure of virusesthat, because of continuous production of errorsin replication, can be considered a collection ofquasi-species. The concept of quasi-species isdiscussed in more detail following.With the species forming the basis of the classificationsystem, they can be grouped into othertaxa on various criteria. To date, the taxonomiclevels of order, family, and genus have beendefined by the ICTV, and it is likely that therewill be pressure for further higher and intermediatetaxa. No formal definition for a genusexists, but it is usually considered “a populationof virus species that share common characteristicsand are different from other populations ofspecies.” Currently, 80 genera of plant virusesare recognised. In some cases—such as the Rhabdoviridae—numerousviruses are recognised thatobviously belong to that family but for whichthere is not enough information to place themeither in existing genera or for creating newgenera; these viruses are listed as “unassigned.”Genera are named either after the type species—for example, Caulimovirus after Cauliflower mosaicvirus—or are given a descriptive name, oftenfrom a Greek or Latin word, for a major featureof the genus—for example, Closterovirus, fromthe Greek klostr (kloster), which is a spindleor thread, or that describes the virus particleshape, such as Geminivirus, from the Latin geminus,meaning “twins.”Similarly, genera are grouped together intofamilies on common characteristics (Table 1.4).There are 17 families recognised for plantviruses; some, such as Reoviridae and Rhabdoviridae,are in common with animal virus families.I. INTRODUCTION TO PLANT VIRUSES

IV. CLASSIFICATION AND NOMENCLATURE OF VIRUSES15TABLE 1.4Criteria Demarcating DifferentVirus TaxaI OrderCommon properties between several families including:Biochemical compositionVirus replication strategyParticle structure (to some extent)General genome organisationII FamilyCommon properties between several genera including:Biochemical compositionVirus replication strategyNature of particle structureGenome organisationIII GenusCommon properties with a genus including:Virus replication strategyGenome size, organisation, and/or number ofsegmentsSequence homologies (hybridisation properties)Vector transmissionIV SpeciesCommon properties within a species including:Genome arrangementSequence homologies (hybridisation properties)Serological relationshipsVector transmissionHost rangePathogenicityTissue tropismGeographical distributionSeventeen of the genera have not yet beenassigned to families and are termed “floatinggenera.” The acquisition of further data on thesefloating genera, together with changing attitudeson virus classification, will no doubt lead to thedesignation of further plant virus families. Thefamily is either named after the type membergenus—for example, Caulimoviridae, named afterthe genus Caulimovirus—or given a descriptivename, as with the genus, for a major feature ofthe family—for example, Geminiviridae, whichdescribes the virus particles.Only three orders have been accepted thusfar by the ICTV. The Mononegavirales contains,among other families, the Rhabdoviridae, whichcontains two plant virus families. In practice,genome nucleic acid sequence data are increasinglybeing used to delimit genera, species, andstrains (Figure 1.4). A detailed discussion ofvirus classification, the currently accepted taxa,and how the ICTV operates are provided inFauquet et al. (2005).C. Naming Viruses (Species)Questions of virus nomenclature have generatedmore heat over the years than the muchmore practically important problems of how todelineate distinct virus species. When a familyor genus is approved by the ICTV, a type speciesis designated. Some virologists favour usingthe English vernacular name as the official speciesname. Using part of a widely known vernacularname as the official species name mayfrequently be a very suitable solution, but it couldnot always apply (e.g., with newly discoveredviruses). Other virologists favour serial numberingfor viruses (species). The experience of othergroups of microbiologists is that, although numberingor lettering systems are easy to set up inthe first instance, they lead to chaos as timepasses and changes must be made in taxonomicgroupings. The idea of Latinized binomial namesfor viruses was supported by the ICTV for manyyears but never implemented for any viruses.In successive editions of the ICTV reports,virus names in the index have been listed bythe vernacular name (usually English) followedby the family or genus name—for example,I. INTRODUCTION TO PLANT VIRUSES

16 1. WHAT IS A VIRUS?ABCFIGURE 1.4 Differentiation of taxa by pairwise identities of sequences of variants of A. RT/RNaseH nucleotidesequences of Banana streak virus isolates; B. Nucleic acid sequences of the L1 gene of members of the Family Papillomaviridae;C. Amino acid sequences of coat proteins of potyviruses. A. With kind permission from Springer Scienceþ Business Media:Arch. Virol., The diversity of Banana streak virus isolates in Uganda, vol. 150, 2005, pp. 2407-2420, G. Harper; B. From VirusTaxonomy, 8 th Report of the National Committee on the Taxonomy of Viruses, Fauquet et al., p. 5, Copyright Elsevier (2005);C. Reichmann et al. (Journal of General Virology 73, 1–16, 1992).tobacco mosaic Tobamovirus, Fiji disease Fijivirus,and Lettuce necrotic yellows rhabdovirus.This method for naming a plant virus is becomingincreasingly used in the literature.D. Acronyms or AbbreviationsAbbreviations of virus names have been usedfor many years to make the literature easier toread and more succinct to present. The abbreviationis usually in the form of an acronym usingthe initial letters of each word in the virus name.As the designation of the acronym was by theauthor of the paper, it was leading to muchoverlap and confusion. For instance, amongplant viruses, AMV was used to designateAlfalfa mosaic virus and Arabis mosaic virus andcould also justifiably be used for Abutilon mosaicvirus, Agropyron mosaic virus, Alpina mosaic virus,Alstromeria mosaic virus, Alternantha mosaic virus,Aneilema mosaic virus, or Anthoxanthum mosaicvirus. Therefore, in 1991 the Plant Virus sectionof the ICTV initiated a rationalisation of plantvirus acronyms and has subsequently updatedthe list regularly in ICTV reports (Box 1.3).There are no efforts to create a commonacronym system for viruses from differentkingdoms. Thus, CMV can mean CucumberI. INTRODUCTION TO PLANT VIRUSES

IV. CLASSIFICATION AND NOMENCLATURE OF VIRUSES17BOX 1.3RULES FOR VIRUS ABBREVIATIONS OR ACRONYMS• Abbreviations should be as simple as possible.• An abbreviation must not duplicate any otherpreviously coined term or one still in use.• The word virus in a name is abbreviated as V.• The word viroid in a name is abbreviated as Vd.• M is usually used for “mosaic” and Mo for“mottle.”• The word ringspot is abbreviated as RS andsymptomless as SL.• Abbreviations for single words should notnormally exceed two letters.• Where a particular combination of letters hasbeen adopted for a particular plant,subsequent abbreviations for viruses of thathost should use the same combination.• The second (or third) letter of a host plantabbreviation is in lowercase—for example,Ab for Abutilon.• When several viruses have the same nameand are differentiated by a number, theabbreviation will have a hyphen between theletters and number—for example, Plantainvirus 6 is abbreviated as PlV-6.• When viruses end with a letter, the letter isadded to the end of the abbreviation withouta hyphen—for example, Potato virus X isabbreviated PVX.• When viruses are distinguished by theirgeographical location, a minimum number ofletters (two or three) are added to theabbreviation with a hyphen—for example,Tomato yellow leaf curl virus from Thailand isTYLCV-Th.• When a virus name comprises a disease nameand the words associated virus, these areabbreviated aV—for example, Grapevine leafrollassociated virus 2 is abbreviated GLRaV-2.A set of guidelines is laid out in Fauquet andMayo (1999). Although these and the acronymsderived from them, are not officially sanctionedby the ICTV, the acronyms are used in the ICTVreports.mosaic virus (of plants), Canine minute virus (ofvertebrates), or Clo Mor virus (of invertebrates).Thus, acronyms have to be taken in context.E. Plant Virus ClassificationThe current classification of plant viruses isshown in Figure 1.5.F. Virus StrainsA virus species is not a uniform populationbecause in each infected cell, a wide range ofvariants is present. This situation is termed aquasi-species (Box 1.4).The quasi-species concept makes it difficultto strictly define a strain. However, one mustdescribe variants within a species and, in reality,take a pragmatic approach. Characters haveto be weighed up as to how they would contributeto making subdivisions and to communication,not only between virologists but alsoto plant pathologists, extension workers, farmers,and many other groups. An example isthe luteovirus Beet western yellows virus(BWYV), which has a wide host range, includingsugar beet in the United States. For manyyears, Beet mild yellows virus, which infectedsugar beet in Europe, was regarded as a strainof BWYV. Confusion arose when it was discoveredthat the European luteovirus that wasmost closely related to BWYV did not infectsugar beet but was common in the oilseed rapecrop. This caused many problems in explainingI. INTRODUCTION TO PLANT VIRUSES

18 1. WHAT IS A VIRUS?FIGURE 1.5 Classification of plant viruses. From Virus Taxonomy, 8 th Report of the National Committee on the Taxonomyof Viruses, Fauquet et al., p. 18, Copyright Elsevier (2005).I. INTRODUCTION TO PLANT VIRUSES

IV. CLASSIFICATION AND NOMENCLATURE OF VIRUSES19BOX 1.4QUASI-SPECIESA quasi-species is a population structure in whichcollections of closely related genomes are subjectedto a continuous process of genetic variation,competition, and selection. Usually, thedistribution of mutants or variants is centredon one or several master sequences. The selectionequilibrium is meta-stable and may collapseor change when an advantageous mutantappears in the distribution.In this case, the previous quasi-species will besubstituted by a new one characterised by a newmaster sequence and a new mutant spectrum.The stability of a quasi-species depends on thecomplexity of the genetic information in theviral genome, the copy fidelity on replication ofthe genome, and the superiority of the mastersequence.A quasi-species has a physical, chemical, andbiological definition. In the physical definition, aquasi-species can be regarded as a cloud insequence space, which is the theoretical representationof all the possible variants of a genomicsequence. For an ssRNA virus of 10 kb, thesequence space is 410,000. Thus, the quasispeciescloud represents only a very small proportionof the sequence space and is constrainedby the requirements of gene and nucleic acidfunctions. Chemically, the quasi-species is arated distribution of related nonidentical genomes.Biologically, a quasi-species is the phenotypicexpression of the population, mostlikely dominated by that of the master sequence.to farmers that the BWYV in their overwinteringoilseed rape crop would not infect theirbeet crop the next year.G. Use of Virus NamesThe ICTV sets rules, which are regularlyrevised, on virus nomenclature and the orthographyof taxonomic names (see the eighth ICTVreport). The last word of a species is virus, andthe suffix (ending) for a genus name is -virus.For a subfamily, it is -virinae; for a family, it is-viridae; and for an order, it is -virales. In formaltaxonomic usage, the virus order, family, subfamily,genus, and species names are printedin italics (or underlined), with the first letterbeing capitalized; other words in species namesare not capitalized unless they are propernouns or parts of proper nouns. Also, in formaluse, the name of the taxon should precedethe name being used—for example, the familyCaulimoviridae, the genus Mastrevirus, and thespecies Potato virus Y. An example of classification,nomenclature, and orthography is shownin Box 1.5.In informal use, the family, subfamily, genus,and species names are written in lowercaseRoman script, the taxon does not include theformal suffix, and the taxonomic unit followsthe name being used—for example, the caulimovirusfamily, the mastrevirus genus, and thepotato virus Y species. In even less formal circumstances,but still widely used, the taxonomicunit is omitted and the taxon for higher taxa canbe in the plural—for example, caulimoviruses,mastreviruses, and potato virus Y.I. INTRODUCTION TO PLANT VIRUSES

20 1. WHAT IS A VIRUS?BOX 1.5EXAMPLE OF VIRUSCLASSIFICATION,NOMENCLATURE ANDORTHOGRAPHYTaxa Example SuffixOrder Mononegavirales -viralesFamily Rhabdoviridae -viridaeSubfamily-virinaeGenus Nucleorhabdovirus -virusSpecies Sonchus yellownet virusAcronym SYNVInformal usage arises from practicalities andcan lead to the adoption of more formal use.For instance, the genus Badnavirus was notadopted in 1991 but was used widely in the literatureand was adopted in the 1995 ICTVreport. However, the year 2000 report limitedits use to certain DNA viruses with bacilliformparticles excluding Rice tungro bacilliform virus.As will be apparent in this book, it is necessaryto distinguish the reverse transcribing DNAviruses that have isometric particles from thosethat have bacilliform particles; the informalusage will be caulimoviruses for the former andbadnaviruses for the latter.V. VIRUSES OF OTHER KINGDOMSThe eighth report from the ICTV (Fauquetet al., 2005) noted over 2,700 accepted, tentative,and unassigned virus species classified into 3orders, 73 families (4 of these divided into subfamilies),and 287 genera. Most of these taxonomicgroupings at the genus level arespecific to viruses of plants, vertebrates, invertebrates,or prokaryotes, but some genera ofviruses infect more than one kingdom. Theoverall classification is based on genome type,some very obvious differences exist betweenthe genome types of plant, vertebrate, invertebrate,and prokaryotic viruses (Table 1.5).TABLE 1.5Numbers of Virus Species in Various KingdomsPlant Vertebrate Invertebrate Prokaryote Fungi and AlgaeNumber % Total Number % Total Number % Total Number % Total Number % TotaldSDNA 0 0 263 28.4 150 60.7 203 62.5 127 67.9sSDNA 198 19.1 66 7.1 27 10.9 92 28.3 0 0RT 66 6.4 62 6.7 24 9.7 0 0 13 7.0dSRNA 48 4.6 75 8.1 38 15.4 24 7.4 33 17.6sS-RNA 48 4.6 227 24.5 0 0 0 0 0 0sSþRNA 677 65.3 234 25.2 8 3.2 6 1.8 14 7.5Total 1,037 927 247 325 187Data from Fauquet et al. (2005), using numbers of assigned, unassigned, and tentative virus species.I. INTRODUCTION TO PLANT VIRUSES

VI. SUMMARY• Plant viruses are important pathogens.• The study of plant viruses has madeimportant contributions to theunderstanding of viruses in general—forexample, the recognition of viruses aspathogens, the structure of virus particles,and the infectious nature of RNA.• This chapter defines a virus, contrasts it withsimilar agents, and discusses how virusesare classified.ReferencesAckermann, H.-W. (2008). [History of virology]—Bacteriophage.Encyclopedia of virology, Vol. 2, 442–449.Caspar, D.L.D. and Klug, A. (1962). Physical principles inthe construction of plant viruses. Cold Spring HarborSymp. Quant. Biol. 27, 1–24.Fauquet, C.M. and Mayo, M.A. (1999). Abbreviations forplant virus names—1999. Arch Virol. 144, 1249–1272.Fauquet, C.M., Mayo, M.A., Maniloff, J., Desselberger, U.,and Ball, L.A. (2005). Virus taxonomy. Eighth Report ofthe International Committee on Taxonomy of Viruses. Elsevier,San Diego.Fenner, F.J. (2008) History of viruses/Vertebrate viruses.Encyclopedia of Virology, Vol. 2, 455–458.VI. SUMMARYHull, R. (2001). Matthew’s plant virology, 4th ed. AcademicPress, San Diego.Murphy, F.A., Fauquet, C.M., Bishop, D.H.L., Ghabrial, S.A.,Jarvis, A.W., Martelli, G.P., Mayo, M.A., and Summer,M.D. (1995). Virus taxonomy. Sixth Report of the InternationalCommittee on Taxonomy of Viruses. Springer-Verlag,Wien, Austria.Rivers, T.M. (1937). Viruses and Koch’s postulates. J. Bacteriol.33, 1–12.Further Reading21Fauquet, C.M. (2008). Taxonomy, classification, and nomenclatureof viruses. Encyclopedia of Virology, Vol.5,9–23.Grandbastien, M.-A. (2008). Retrotransposons of plants.Encyclopedia of Virology, Vol. 4, 428–435.Hull, R. (2002) Matthew’s plant virology, 4th ed. AcademicPress, London.Hull, R. (2008). History of virology: Plant viruses. Encyclopediaof Virology, Vol. 2, 450–454.Levin, H.L. (2008). Metaviruses. Encyclopedia of Virology,Vol. 3, 301–310.Pavord, A. (1999). The tulip. Bloomsbury, London.Sanjuan, R. (2008). Quasispecies. Encyclopedia of Virology,Vol. 4, 359–366.Van Regenmortel, M.H.V. Virus species. Encyclopedia ofVirology, Vol. 5, 401–405.Voytas, D.F. (2008). Pseudoviruses. Encyclopedia of Virology,Vol. 4, 352–358.I. INTRODUCTION TO PLANT VIRUSES

This page intentionally left blank

C H A P T E R2Overview of Plant VirusesPlant viruses are important both economically and as model systems to explore the interactionsof viruses and their host cells. They have some properties in common with viruses of otherkingdoms and some properties specific to the properties of their plant hosts.O U T L I N EI. Introduction 23II.Economic Losses Due to PlantViruses 24III. Virus Profiles 24IV. Macroscopic Symptoms 25V. Histological Changes 30VI. Cytological Effects 32VII. The Host Range of Viruses 38VIII. Viruses in Other Kingdoms 40IX. Summary 40I. INTRODUCTIONMost of the plant virus names that are commonlyused today include terms that describean important symptom in a major host or thehost from which the virus was first described.Some viruses, under appropriate conditions,may infect a plant without producing any obvioussigns of disease. Others may lead to rapiddeath of the whole plant. Between these extremes,a wide variety of diseases can be produced.Virus infection does not necessarily causedisease at all times in all parts of an infectedplant. We can distinguish six situations inwhich obvious disease may be absent:1. Infection with a very mild strain of thevirus2. A tolerant host3. “Recovery” from disease symptomsin newly formed leaves (see Chapter 11)4. Leaves that escape infection because oftheir age and position on the plantComparative Plant Virology, Second Edition23Copyright # 2009, Elsevier Inc. All rights reserved.

24 2. OVERVIEW OF PLANT VIRUSES5. Dark green areas in a mosaic pattern (seeChapter 11)6. Plants that are infected with cryptic virusesII. ECONOMIC LOSSES DUE TOPLANT VIRUSESOne of the main driving forces for thedetailed studies of plant viruses is the impactthat the resulting diseases have on crop productivityworldwide, although it is difficult toobtain firm data on the actual losses themselves.The losses due to fungal and bacterial pathogensare well documented, and lists of loss estimatesattributable to specific named fungi or bacteriaare easily obtainable. In these compendia, thelosses due to viruses are often lumped togetherin categories such as “Virus Diseases,” “AllOther,” or “Miscellaneous” diseases. However,viruses are responsible for far greater economiclosses than are generally recognised. This lack ofrecognition are due to several factors, especiallytheir insidious nature. Virus diseases are frequentlyless conspicuous than those caused byother plant pathogens, and they last much longer.This are especially true for perennial cropsand those that are vegetatively propagated.One further problem with attempting to assesslosses due to virus diseases on a global basis isthat most of the data are from small comparativetrials that do not necessarily give information thatcan be used for more global estimates of losses.This is due to factors such as variation in lossesby a particular virus in a particular crop from yearto year, variation from region to region and climaticzone to climatic zone, differences in lossassessment methodologies, identification of theviral aetiology of the disease, variation in the definitionof the term “losses,” and complicationswith other loss factors.In addition to the obvious detrimental effectssuch as reduced yields and visual product quality,virus infections often do not induce noticeableTABLE 2.1 Some Types of Direct and IndirectDamage Associated with Plant Virus InfectionsReduction in growthYield reduction (including symptomless infection)Crop failureReduction in vigourIncreased sensitivity to frost and droughtIncreased predisposition to attack by other pathogensand pestsReduction in quality or market valueDefects of visual attraction: size, shape, colourReduced keeping qualityReduced consumer appeal: grading, taste, texture,compositionReduced fitness for propagationCost of attempting to maintain crop healthCultural hygiene on farm, including vector controlProduction of virus-free propagation materialsChecking propagules and commodities on export/import(quarantine programmes)Eradication programmesBreeding for resistanceResearch, extension, and education.From Waterworth and Hadidi (1998).disease but influence their effects on plants in avariety of more subtle ways. Table 2.1 identifiessome of the ways that viruses can damage cropplants. We can see that the effects of virus infectionextend into areas far beyond the actualreduction in yield and quality. Loss estimates donot take account of these indirect factors.In spite of all these limitations, various collectionsof loss data (Table 2.2) have been compiled.Various newly emerging virus problems incrops are being exacerbated by changes in agriculturalpractices, global trade, and the climate(see Anderson et al., 2004).III. VIRUS PROFILESAppendix A provides profiles of 32 plantviruses that feature strongly in this book. Theseprofiles give you basic information on theseimportant viruses and are referred to in thisand subsequent chapters.I. INTRODUCTION TO PLANT VIRUSES

IV. MACROSCOPIC SYMPTOMS25TABLE 2.2Some Examples of Crop Losses Due to VirusesCrop Virus Countries Loss/YearRice Tungro SE Asia $1.5 10 9Ragged stunt SE Asia $1.4 10 8Hoja blanca S. and C. America $9.0 10 6Barley Barley yellow dwarf UK £6 10 6Wheat Barley yellow dwarf UK £5 10 6Potato Potato leafroll UK £3-5 10 7Potato virusYPotato virus XSugar beet Beet yellows UK £5-50 10 6Beet mild yellowsCitrus Citrus tristeza Worldwide £9-24 10 6Cassava African cassava mosaic Africa $2 10 9Many crops Tomato spotted wilt a Worldwide $1 10 9Cocoa Cocoa swollen shoot Ghana 1.9 10 8 trees ba Data from Prins and Goldbach (1996); references to other data given in Hull and Davies (1992).b Number of trees eradicated over about 40 years.IV. MACROSCOPIC SYMPTOMSSymptoms on plants may be local on theinoculated leaves and/or systemic on spreadto other parts of the plant from inoculatedleaves.A. Local SymptomsLocalised lesions that develop near the siteof entry on leaves are not usually of any economicsignificance but are important forbiological assay. Three types of local responseto infection can result: necrotic local lesions inwhich the infected cells die (Profile 14; seeAppendix) and that vary from small pinpointsto large irregular spreading necrotic patches;chlorotic local lesions in which the infectedcells lose chlorophyll and other pigments; andring spot lesions that typically consist of a centralgroup of dead cells beyond which developone or more superficial concentric rings ofnecrotic or chlorotic cells with normal green tissuebetween them (Profile 6; see Appendix).Some viruses in certain hosts show no visiblelocal lesions in the intact leaf, but when the leafis cleared in ethanol and stained with iodine,“starch lesions” may become apparent (seeChapter 9).Viruses that produce local lesions wheninoculated mechanically onto leaves may notdo so when introduced by other means. Forexample, Beet yellows virus produces necroticlocal lesions on Chenopodium capitatum, but itdoes not do so when the virus is introducedby the aphid Myzus persicae feeding on parenchymacells.I. INTRODUCTION TO PLANT VIRUSES

26 2. OVERVIEW OF PLANT VIRUSESB. Systemic SymptomsVarious symptoms often appear in combinationin particular diseases, and the pattern ofdisease development for a particular host-viruscombination often involves a sequential developmentof different kinds of symptoms.1. Effects on Plant SizeReduction in plant size is the most generalsymptom induced by virus infection (seeBox 2.1). The degree of stunting is generallycorrelated with the severity of other symptoms,particularly where loss of chlorophyll from theleaves is concerned. Stunting is usually almostentirely due to reduction in leaf size and internodelength. Leaf number may be littleaffected.In perennial deciduous plants, such asgrapes, there may be a delayed initiation ofgrowth in the spring. Root initiation in cuttingsfrom virus-infected plants may be reduced, asin chrysanthemums. In vegetatively propagatedplants, stunting is often a progressive process.For example, virus-infected strawberry plantsand tulip bulbs may become smaller in eachsuccessive year.2. Mosaic Patterns and RelatedSymptomsOne of the most common obvious effects ofvirus infection is the development of a patternof light and dark green areas, which creates amosaic effect in infected leaves. The detailednature of the pattern varies widely for differenthost-virus combinations. In dicotyledons,the areas that make up the mosaic are generallyirregular in outline. For example, onlytwo shades of colour—like dark green and apale or yellow-green—may be involved, (seeProfiles 3 and 14; see Appendix), or there maybe many different shades of green, yellow,or even white, as with Turnip yellow mosaicvirus (TYMV) in Chinese cabbage (Profile 18;see Appendix). The junctions between areas ofdifferent colour may be sharp, and such diseasesresemble quite closely the mosaics producedby inherited genetic defects in thechloroplasts. Abutilon mosaic virus is a goodexample of this type (Profile 7; see Appendix).In mosaic diseases that infect herbaceousplants, usually a fairly well-defined sequencein the development of systemic symptomsoccurs. The virus moves up from the inoculatedleaf into the growing shoot and intopartly expanded leaves. In these leaves, thefirst symptoms are a “clearing” or yellowingof the veins; it is suggested that this symptomis an optical illusion. However, chlorotic veinbandingis a true symptom and may be veryfaint or may give striking emphasis to thepattern of veins [see Cauliflower mosaic virus(CaMV), Profile 4, and Beet yellows virus (BYV),Profile 5; see Appendix]. Vein-banding maypersist as a major feature of the disease.No mosaic pattern develops in leaves thatare past the cell division stage of leaf expansionwhen they become infected (about 4–6 cm longfor leaves such as tobacco). These leavesbecome uniformly paler than normal. In theoldest leaves to show mosaic, a large numberof small islands of dark green tissue usuallyappear against a background of paler colour(for molecular aspects of dark green islands,see Chapter 11). The mosaic areas may be confinedto the youngest part of the leaf blade—that is, the basal and central region. Althoughconsiderable variation in different plants mayappear, successively younger systemicallyinfected leaves show, on the average, mosaicsconsisting of fewer and larger areas. Themosaic pattern is laid down at a very earlystage of leaf development and may remainunchanged, except for general enlargement,for most of the life of the leaf.In monocotyledons, a common result of virusinfection is the production of stripes or streaksof tissue lighter in colour than the rest of the leaf.I. INTRODUCTION TO PLANT VIRUSES

IV. MACROSCOPIC SYMPTOMS27BOX 2.1GROUNDNUT ROSETTE DISEASEGroundnut rosette is an important disease of groundnuts(peanut) (Arachis hypogaea) in West and East Africa. It istransmitted by the aphid Aphis craccivora in a circulativenonpropagative manner. It causes the leaves to becomelight green or chlorotic, stunts the plant, and reducesyield. The figure here compares a disease plant (upper)with a healthy plant (lower).The disease is caused by a complex of twoviruses—an umbravirus, Groundnut rosette virus(GRV), and a luteovirus, Groundnut rosette assistorvirus (GRAV)—and a satellite RNA (sGRV).Umbraviruses do not encode a coat protein,and that for GRV is provided by GRAV. sGRVis essential for this encapsidation and for thesymptom production by GRV. GRV’s role inthis complex is to potentiate the replication ofsGRV. Thus, in the final complex, all three entitiesinteract as shown in the diagram.The shades of colour vary from pale green toyellow or white, and the more or less angularstreaks or stripes run parallel to the length of theleaf (see Banana streak virus (BSV), Profile 4, andMaize streak virus, Profile 7). The developmentof the stripe diseases found in monocotyledonsfollows a similar general pattern to that foundfor mosaic diseases in dicotyledons.I. INTRODUCTION TO PLANT VIRUSES

28 2. OVERVIEW OF PLANT VIRUSESA variegation or “breaking” in the colour ofpetals commonly accompanies mosaic or streaksymptoms in leaves. The breaking usually consistsof flecks, streaks, or sectors of tissue witha colour different from normal (see Box 1.1).The breaking of petal colour is frequently dueto loss of anthocyanin pigments, which revealsany underlying colouration due to plastid pigments.Flower colour-breaking may sometimesbe confused with genetic variegation, but it isusually a good diagnostic feature for infectionby viruses that produce mosaic symptoms. In afew plants, virus-induced variegation has beenvalued commercially. As with the developmentof mosaic patterns in leaves, colour-breaking inthe petals may develop only in flowers that aresmaller than a certain size when infected. Virusinfection may reduce pollen production anddecrease seed set, seed size, and germination.Fruits that form on plants showing mosaicdisease in the leaves may show a mottling—forexample, zucchinis infected with Cucumbermosaic virus (CMV; Profile 3). In other diseases,severe stunting and distortion of fruit may occur.Seed coats of infected seed may be mottled.3. Yellow DiseasesViruses that cause a general yellowing of theleaves are not as numerous as those that causemosaic diseases, but some, such as the virusesthat cause yellows in sugar beet, are of considerableeconomic importance. The first sign ofinfection is usually a clearing or yellowing ofthe veins in the younger leaves followed by ageneral yellowing or reddening of the leaves[see Barley yellow dwarf virus, Profile 8 (seeAppendix); Rice tungro bacilliform virus (RTBV),Profile 4; and BYV, Profile 5]. This yellowingmay be slight or severe. No mosaic is produced,but in some leaves sectors of yellowedand normal tissue may appear.4. Leaf RollingVirus infection can result in leaf rolling,which is usually upward but occasionallydownward (see Potato leaf roll virus, Profile 8).Pronounced epinasty (downward bending ofleaves) may sometimes be a prominent feature.5. Ring Spot DiseasesRing spots are a pattern of concentric ringsand irregular lines on the leaves and sometimesalso on the fruit (see Tobacco ring spot virus, Profile6, and Papaya ring spot virus, Profile 10; seeAppendix). The lines may consist of yellowedtissue or may be due to death of superficiallayers of cells, giving an etched appearance. Insevere diseases, complete necrosis through thefull thickness of the leaf lamina may occur.Ring spot patterns may also occur on otherorgans, such as bulbs and tubers (Profile 15;see Appendix).6. Necrotic DiseasesThe death of tissues, organs, or the whole plantis the main feature of some diseases. Necrotic patternsmay follow the veins as the virus moves intothe leaf (Profile 16; see Appendix). In some diseases,the entire leaf is killed. Necrosis mayextend fairly rapidly throughout the plant. Forexample, with joint infections of Potato virus X(PVX) and Potato virus Y (PVY) in tomatoes,necrotic streaks appear in the stem (see Box 11.3in Chapter 11). Necrosis spreads rapidly to thegrowing point, which is killed, and subsequentlyall leaves may collapse and die. Wilting of theparts that are about to become necrotic often precedessuch systemic necrotic disease.7. Developmental AbnormalitiesBesides being generally smaller than normal,virus-infected plants may show a wide range ofdevelopmental abnormalities. Such changesmay be the major feature of the disease ormay accompany other symptoms. For example,uneven growth of the leaf lamina is often foundin mosaic diseases. Dark green areas may beraised to give a blistering effect, and the marginof the leaf may be irregular and twisted.I. INTRODUCTION TO PLANT VIRUSES

IV. MACROSCOPIC SYMPTOMS29In some diseases, the leaf blade may be more orless completely suppressed—for example, intomatoes infected with CMV (Profile 3).Some viruses cause swellings in the stem,which may be substantial in woody plants—for example, in Cocoa swollen shoot virus (CSSV)disease. Another group of growth abnormalitiesis known as enations, which are outgrowthsfrom the upper or lower surface ofthe leaf, usually associated with veins.Viruses may cause a variety of tumour-likegrowths. The most studied tumours are thoseproduced by Wound tumor virus (WTV). In asystemically infected plant, external tumoursappear on leaves or stems where wounds aremade. In infected roots they appear spontaneously,beginning development close to cells inthe pericycle that are wounded when developingside roots break through the cortex. Stemdeformation such as stem splitting and scarringis caused by some viruses in some woodyplants. Virus infection of either the rootstockor scion can cause necrosis and/or failure ofthe graft union. One of the unusual symptomsof BSV in some Musa cultivars is that the fruitbunch emerges from the side of the pseudosteminstead of the top of it. This is due tonecrosis of the cigar leaf.8. WiltingSome virus infections induce wilting of theaerial parts, leading to the death of the wholeplant (see Citrus tristeza virus, Profile 5).9. Recovery from DiseaseNot uncommonly, a plant shows disease symptomsfor a period, and then new growth appearsin which symptoms are milder or absent,although virus is still present. This commonlyoccurs with Nepovirus infections. Many factorsinfluence this recovery phenomenon. The stageof development at which a plant is infected mayhave a marked effect on the extent to which symptomsare produced. The environment can alsoaffect recovery from disease, as can host speciesor variety and virus strain. The molecular aspectsof disease recovery are discussed in Chapter 11.10. Genetic EffectsInfection with Barley stripe mosaic virus(BSMV) induces an increase in mutation ratein Zea mays and also a genetic abnormalityknown as an aberrant ratio (AR). This AR effectwas observed only when the original pollenparent was infected and showing virus symptomson the upper leaves. The AR effect isinherited in a stable manner in plants in whichthe virus can no longer be detected, with a lowfrequency of reversion to normal ratios.C. The CryptovirusesCryptoviruses escaped detection for manyyears because most of them cause no visiblesymptoms or, in a few situations, very mildsymptoms. They are not transmissible mechanicallyor by vectors, but they are transmitted efficientlyin pollen and seed. They occur in verylow concentrations in infected plants. Nevertheless,they have molecular characteristics thatmight be expected of disease-producing viruses.The genome of cryptoviruses consists oftwo dsRNA segments, and these viruses sharesome properties with the reoviruses. There isno indication, other than the low concentrationat which they occur, as to why they causesymptomless infection.D. Diseases Caused by Viral ComplexesThe joint infection of two viruses can causesymptoms that are more severe than those ofeither of the two viruses; this is termed synergism.An example is tomato streak caused bythe joint infection with PVX and PVY, whichwas just discussed (see Box 11.3).Some virus diseases are caused by the complexesof two (or more) viruses, each contributingfeatures to the disease. Rice tungro diseaseis caused by the joint infection of RTBV andI. INTRODUCTION TO PLANT VIRUSES

30 2. OVERVIEW OF PLANT VIRUSESbiological, physical, and chemical agents. Theseinclude phytoplasmas, spiroplasmas, rickettsialikeorganisms (all three of these are discussedin Chapter 3), bacteria, toxins produced byarthropods, nutritional deficiencies, high temperatures,hormones (weed killers), and insecticidedamage and air pollutants. Furthermore,genetic variants, some of which may be causedby transposons, can resemble mosaic and leafvariegations caused by viruses. Thus, care mustbe taken in diagnosing the viral aetiology of anew problem.V. HISTOLOGICAL CHANGESFIGURE 2.1 Diagram showing interaction of Rice tungrobacilliform virus and Rice tungro spherical virus in tungro diseasecomplex.Rice tungro spherical virus (RTSV; Figure 2.1).RTBV causes severe symptoms in rice but isnot leafhopper transmitted on its own; RTSVis leafhopper transmitted but causes few, ifany, symptoms. The complex is leafhoppertransmitted with RTSV, giving the transmissionproperties and RTBV the severe disease symptoms.Similarly, umbraviruses require a luteovirusfor aphid transmission. An even morecomplicated complex is show in groundnutrosette disease (see Box 2.1).E. Agents Inducing Virus-LikeSymptomsDisease symptoms, similar to those producedby viruses, can be caused by a range ofThe basic structure of a plant and a plant cellis shown in Box 2.2. The macroscopic symptomsinduced by viruses frequently reflecthistological changes within the plant. Thesechanges are of three main types—necrosis,hypoplasia, and hyperplasia—that may occursingly or together in any particular disease.A. NecrosisNecrosis, as a macroscopic symptom, wasdiscussed previously. It may be general ormay be limited to specific tissues, such as localisedareas of the roots (e.g., Tobacco necrosisvirus), internally in immature tomato fruitswith late infection by Tobacco mosaic virus(TMV), or limited to the phloem, as in potatoleaf roll disease.B. HypoplasiaLeaves with mosaic symptoms frequentlyshow hypoplasia (localised retarded growth frequentlyleading to thinner areas on leaves) inthe yellow areas. The lamina is thinner than inthe dark green areas, and the mesophyll cellsare less differentiated with fewer chloroplastsand fewer or no intercellular spaces (Figure 2.2).I. INTRODUCTION TO PLANT VIRUSES

ABOX 2.2PLANT STRUCTUREBCuticleUpperepidermisPalisademesophyllcellVeinSpongymesophyllcellLowerepidermisCuticleCBundlesheathcellXylemDEpidermisPithCuticleCortexVascularbundleSieve tubeelementXylemPhloemPhloemCompanioncellMitochondrionChloroplastCentralvacuoleCell wallEGolgi complexSmoothendoplasmicreticulumPlasmamembraneFMitochondrionCytoplasmPlasmodesmaMicrotubules(part ofcytoskeleton)RoughendoplasmicreticulumNucleusMicrotubules(part ofcytoskeleton)Fig. Basic structure of plant tissues. A. Diagrammatic dicotyledonous plant; B. Diagram of a section of aleaf showing the various tissues; C. Details of the cell types in a vein. D. Diagram of a section of a stem;E. Structure of a plant cell; F. Structure of an animal cell.

32 2. OVERVIEW OF PLANT VIRUSES1. Cell SizeVein-clearing symptoms are due, with someviruses at least, to enlargement of cells nearthe veins. The intercellular spaces are obliterated,and since there is little chlorophyllpresent, the tissue may become abnormallytranslucent.2. Cell Division in Differentiated CellsSome viruses such as PVX may produceislands of necrotic cells in potato tubers. Thetuber may respond with a typical wound reactionin a zone of cells around the necrotic area.Starch grains disappear, and an active cambiallayer develops. Similarly, in a white halo zonesurrounding necrotic local lesions inducedby TMV in N. glutinosa leaves (Profile 14), celldivision occurred in mature palisade cells.3. Abnormal Division of Cambial CellsThe vascular tissues appear to be particularlyprone to virus-induced hyperplasia. For example,in the CSSV diseased shoots, abnormalamounts of xylem tissue are produced, but thecells appear structurally normal. In crimson cloverinfected by WTV, abnormal development ofphloem cambium cells is present, and phloemparenchyma forms meristematic tumour cells.FIGURE 2.2 Hypoplasia in tobacco cells infected withTMV. A. Section through palisade cells of a dark green areaof a leaf with mosaic showing that the cells are essentiallynormal; B. Section through a nearby yellow-green areashowing that the cells are large and undifferentiated inshape. (Courtesy of P.H. Atkinson)The major anatomical effect of Apple stemgroovingvirus in apple stems is the disappearanceof the cambium in the region of the groove. Normalphloem and xylem elements are replacedby a largely undifferentiated parenchyma.C. HyperplasiaHyperplasia is growth of abnormally largecells or excessive cell division.VI. CYTOLOGICAL EFFECTSA. Effects on Cell Structures1. NucleiMany viruses have no detectable cytologicaleffects on nuclei, but others give rise to intranuclearinclusions of various sorts and mayaffect the nucleolus or the size and shape ofthe nucleus, even though they appear not toreplicate in this organelle. Here are someexamples:• In pea leaves and pods infected with Peaenation mosaic virus (PEMV), particlesI. INTRODUCTION TO PLANT VIRUSES

VI. CYTOLOGICAL EFFECTS33accumulate first in the nucleus, and then thenucleolus disintegrates. PEMV also causesvesiculation in the perinuclear space.• Crystalline platelike inclusions can be seenby light microscopy in the nuclei of cellsinfected with several potyviruses.• Geminivirus infection can cause markedhypertrophy of the nucleolus, which maycome to occupy three-quarters of thenuclear volume. Fibrillar rings ofdeoxyribonucleoprotein appear, andmasses of virus particles accumulate inthe nucleus.2. MitochondriaThe long rods of Tobacco rattle virus may beassociated with the mitochondria in infectedcells (see Figure 2.3A), as are the isometric particlesof Broad bean wilt virus-1 (see Figure 2.3B).3. ChloroplastsInfection with TYMV induces small peripheralvesicles (Figure 2.4) and other changes inand near the chloroplasts, as well as manyother cytological changes in the chloroplasts,most of which appear to constitute a structuraland biochemical degeneration of the organelles.In many infections, the size and number ofstarch grains seen in leaf cells are abnormal.In mosaic diseases, there is generally speakingless starch than normal, but in some diseases(e.g., sugar beet curly top and potato leaf roll)excessive amounts of starch may accumulate.Similarly in local lesions induced by TMV incucumber cotyledons, chloroplasts becomegreatly enlarged and filled with starch grains.4. Cell WallsThe plant cell wall tends to be regardedmainly as a physical supporting and barrierstructure. In fact, it is a distinct biochemicaland physiological compartment that contains asubstantial proportion of the total activity of certainenzymes in the leaf. Three kinds of abnormalityhave been observed in or near the wallsof virus-diseased cells:• Abnormal thickening, due to the depositionof callose, may occur in cells near the edge ofvirus-induced lesions.• Cell wall protrusions involving theplasmodesmata have been reported forseveral unrelated viruses. These are ofteninvolved in cell-to-cell movement of viruses(see Chapter 9).• Depositions of electron-dense material betweenthecellwallandtheplasmamembrane(sometimes called paramural bodies) mayextend over substantial areasofthecellwall.FIGURE 2.3 Association of virusparticles with mitochondria. A. Sectionof Nicotiana clevelandii showing mitochondrionwith fringe of Tobacco rattlevirus particles. [From Reichmann et al.(Journal of General Virology 73, 1–16,1992)] B. Section of N. clevelandiishowing particles of Broad bean wiltvirus-1 arranged between mitochondria.[From Hull and Plaskitt (1974;Intervirology 2, 352–359, S. Karger AG,Basel).]ABI. INTRODUCTION TO PLANT VIRUSES

34 2. OVERVIEW OF PLANT VIRUSESFIGURE 2.4 TYMV-induced peripheralvesicles in the chloroplasts ofChinese cabbage. A. General view;B. Details of the boxed vesicles in Ashowing continuity of the chloroplastmembranes in the vesicle and fibrillarmaterial inside the vesicle. Reprintedfrom Virology, 59, T. Hatta and R.E.F.Matthews, The sequence of early cytologicalchanges in Chinese cabbageleaf cells following systemic infectionwith turnip yellow mosaic virus, pp.37–50, Copyright (1974), with permissionfrom Elsevier.AB5. Cell DeathDrastic cytological changes occur in cells asthey approach death. These changes have beenstudied by both light and electron microscopy,but they do not tell us how virus infection actuallykills the cell.B. Virus-Induced Structures in theCytoplasmVarious types of virus-induced structuresare found in the cytoplasm of infected plants.They are either accumulations of virus particles,aggregates of virus-encoded proteins, or modificationsof the cell organelles associated withvirus replication (often termed viroplasms; seeBox 2.3).1. Accumulations of Virus ParticlesVirus particles may accumulate in an infectedcell in sufficient numbers and exist under suitableconditions to form three-dimensional crystallinearrays. These may grow into crystalslarge enough to be seen with the light microscope,or they may remain as small arrays thatBOX 2.3VIROPLASMSAs we will see in Chapter 8, virus replicationoften involves membranes that causes cellulardisturbance. This disturbance can be seenas clustering of membranes and/or associatedorganelles that are visible under theelectron microscope (and sometimes eventhe light microscope)—for instance, in TMVinfections. These different structures aresometimes called viroplasms, although thisterm is also used to describe CaMV andrhabdovirus inclusion bodies.can be detected only by electron microscopy.The ability to form crystals within the host celldepends on properties of the virus itself and isnot related to the overall concentration reachedin the tissue or to the ability of the purifiedvirus to form crystals. For example, TYMVI. INTRODUCTION TO PLANT VIRUSES

VI. CYTOLOGICAL EFFECTS35can readily crystallize in vitro. It reaches highconcentration in infected tissue but does notnormally form crystals there. By contrast, Satellitetobacco necrosis virus occurring in muchlower concentrations frequently forms intracellularcrystals.Rod-shaped viruses belonging to variousgroups may aggregate in the cell into more orless ordered arrays. In tobacco leaves showingtypical mosaic symptoms caused by TMV,leaf-hair and epidermal cells over yellowgreenareas may almost all contain crystallineinclusions of virus particles, whereas those infully dark green areas contain none (seeFigure 2.5A). Sometimes long, curved, fibrousinclusions, or spike-like or spindle-shapedinclusions made up largely of virus particles,can be seen by light microscopy. Differentstrains of the virus may form different kindsof paracrystalline arrays. Most crystalline inclusionshave been found only in the cytoplasm,but some have been detected in nuclei.Many small icosahedral viruses form crystallinearrays in infected cells (see Figure 2.5B),sometimes in ordered structures such astubules. Plant cells infected with viruses thatbelong to the Reoviridae or Rhabdoviridae frequentlycontain masses of virus particles in regulararrays in the cytoplasm. With somerhabdoviruses the bullet-shaped particles accumulatein more or less regular arrays in theperinuclear space (Figure 2.5C).2. Aggregates of Virus-Encoded ProteinsPotyviruses induce the formation of characteristiccylindrical inclusions in the cytoplasmof infected cells (Box 2.4).3. Caulimovirus InclusionsTwo forms of inclusion bodies (also termedviroplasms) have been recognised in thecytoplasm of plants infected with CaMV andother “caulimoviruses” (see Figure 2.6). Bothforms contain virus particles. Electron-denseinclusions are made up of open reading frame(ORF) VI product and are considered to be thesites of virus synthesis and assembly (see Chapter8). Electron-lucent inclusion bodies are madeup of ORF II product, one of the proteinsinvolved in aphid transmission (see Chapter 12).FIGURE 2.5 Crystals of virus particlesin infected leaves. A. Freezeetchedpreparation of TMV crystalwithin a tobacco mesophyll cellshowing herringbone arrangement ofparticles; bar ¼ 1 mm. [This articlewas published in J. Ultrastr. Res., 54,J.H. Willison, The hexagonal latticespacing of intracellular crystallinetobacco mosaic virus, 176–182, CopyrightElsevier (1976).] B. Transversesection of tubes composed of particlesof Broad bean wilt virus-1 in Nicotianaclevelandii leaf; bar ¼ 250 nm. [FromHull and Plaskitt (1974; Intervirology2, 352–359, S. Karger AG, Basel).] C.Thin section of maize leaf showingparticles of Maize mosaic virus apparentlybudding through the innernuclear membrane (INM). Cy, cytoplasm;N, nucleus; ONM, outernuclear membrane. Bar ¼ 0.3 mm.[From McDaniel et al. (1985; Phytopathology75, 1167–1172).]ABCI. INTRODUCTION TO PLANT VIRUSES

36 2. OVERVIEW OF PLANT VIRUSESBOX 2.4POTYVIRAL INCLUSION BODIESPotyviruses induce the formation of characteristic cylindrical inclusions in the cytoplasm of infectedcells. The most striking feature of these inclusions viewed in thin cross-section is the presence of acentral tubule from which radiate curved “arms” to give a pinwheel effect. Reconstruction fromserial sections shows that the inclusions consist of a series of plates and curved scrolls with a finelystriated substructure with a periodicity of about 5 nm. The bundles, cylinders, tubes, and pinwheelsseen in section are aspects of geometrically complex structures (Figure).Fig. A. Scrolls induced in Nicotiana clevelandii by an unidentified potyvirus; B. Laminated aggregates inducedby Statice virus Y in Chenopodium quinoa; C. Scrolls and laminated aggregates induced by Turnip mosaic virus inN. clevelandii; D. Scrolls and short curved laminated aggregates in N. tabacum infected with PVY. Bars ¼200 nm. [From Lesemann (1988; in The plant viruses, Vol. 4, The filamentous plant viruses, R.G. Milne, Ed., pp. 179–235,Plenum Press, New York).]Pinwheel inclusions originate and develop in association with the plasma membrane at sites lyingover plasmodesmata. The central tubule of the pinwheel is located directly over the plasmodesmata,and it is possible that the membranes may be continuous from one cell to the next. The core and thesheets extend out into the cytoplasm as the inclusion grows. Later in infection, they may become dissociatedfrom the plasmodesmata and come to lie free in the cytoplasm. Virus particles may be intimatelyassociated with the pinwheel arms at all times and particularly at early stages of infection.The 71 kDa virus-coded protein that is found in these inclusions is part of the viral replicase.I. INTRODUCTION TO PLANT VIRUSES

VI. CYTOLOGICAL EFFECTS37FIGURE 2.6 Electron micrographsof CaMV inclusion bodies in infectedturnip leaves immunogold-labelledand anti-P62 (ORF VI product) antiserum.A. Electron-dense inclusion bodieswith gold particles preferentially labellingthe matrix (bar ¼ 200 nm); B. Cellshowing heavily labelled electrondenseinclusion body (filled-in star)and unlabelled electron-lucent inclusion(open star; bar ¼ 1 mm); C. Electron-lucentinclusion (bar ¼ 500 nm).[This article was published in Virology,185, A.M. Espinoza, V. Medina, R. Hull,P.G. Markham, Cauliflower mosaic virusgene II product forms distinct inclusionbodies in infected plant cells, 337–344,Copyright Elsevier (1991).]C. Why Inclusion Bodies?Viruses can go through numerous rounds ofreplication, which if allowed in an uncontrolledmanner could result in cell death. Thus, it islikely that aggregation of a virus is a mechanismof removing it from the replication pool. Also,the aggregates of virus gene products are consideredto be a means of sequestering potentiallydeleterious proteins from the cellmetabolism. For instance, the potyviral genomeis expressed from a polyprotein, which givesrise to as many copies of protease and replicasemolecules as of coat protein molecules (seeChapter 7). It is likely that the proteases andreplicases could be deleterious to the cell, sothey are sequestered into inclusion bodies.D. Cytological StructuresSome normal structures in cells could bemistaken for virus-induced effects—for example,crystalline or membrane-bound inclusionsin plastids. Phosphorus-deficient bean leavesI. INTRODUCTION TO PLANT VIRUSES

38 2. OVERVIEW OF PLANT VIRUSESshow degenerative changes in the chloroplastslike those seen in some virus infections. Similarly,in sulphur-deficient Zea mays, chloroplastsmay contain many osmiophilic granulesand small vesicles.Nuclei of healthy cells sometimes containcrystalline structures that might be mistakenfor viral inclusions. Such virus-induced effectsas disorganisation of membrane systems, presenceof numerous osmiophilic granules, anddisintegration of organelles are similar to normaldegenerative processes associated withaging or degeneration induced by other agents.VII. THE HOST RANGE OFVIRUSESSince the early years of the last century, plantvirologists have used host range as a criterionfor attempting to identify and classify viruses.In a typical experiment, the virus under studywould be inoculated by mechanical means to arange of plant species that would then beobserved for the development of virus-like diseasesymptoms. Back-inoculation to a hostknown to develop disease might be used tocheck for symptomless infections. In retrospect,it can be seen that reliance on such a proceduregives an oversimplified view of the problem ofvirus host ranges. Over the past few years, ourideas of what we might mean by “infection”have been considerably refined, and some possiblemolecular mechanisms that might make aplant a host or a nonhost for a particular virushave emerged.The term host is sometimes used ratherloosely. Technically, it is defined as “an organismor cell culture in which a given virus canreplicate.” This would mean that a plant speciesin which the virus can replicate in the initiallyinfected cell (subliminal infection) is a host.However, this is impractical, and in this bookthe term local host is used for a species in whichthe virus is restrained to the inoculated leaf,and systemic host is used for a species in whichthe virus spreads from the inoculated leafto other, but not necessarily all, parts of theplant.A. Limitations in Host Range StudiesAlmost all the plant viruses so far describedhave been found infecting species among theangiosperms. Only a minute proportion ofthe possible host-virus combinations has beentested experimentally. The following arithmeticindicates the scale of our ignorance. In a majorstudy on host range, 24 viruses were tested on456 angiosperm species revealing 1,312 newhost-virus combinations, or 12 percent of thosetested. There may be about 250,000 species ofangiosperms, and over 1,000 plant viruses havebeen recorded. If the 12 percent rate applied onaverage to all these plants and viruses, thenthere may be more than 27 10 6 new compatiblehost-virus combinations awaiting discovery.In relation to this figure, the number ofcombinations already tested must be almostnegligible.Our present knowledge of the occurrenceand distribution of viruses among the variousgroups of plants is both fragmentary andbiased. There are four probable reasons for this.• Plant virologists who work on diseases asthey occur in the field have been primarilyconcerned with viruses that cause economiclosses in cultivated plants. They haveusually been interested in other plant speciesonly to the extent that they might be actingas reservoirs of a virus or its vector affectinga cultivated species. Thus, until fairlyrecently all the known plant viruses wereconfined to the angiosperms. Within thisgroup, most of the known virus hosts areplants used in agriculture or horticulture orare weed species that grow in cultivatedareas.I. INTRODUCTION TO PLANT VIRUSES

VII. THE HOST RANGE OF VIRUSES39• It seems likely that widespread and severedisease in plants due to virus infection islargely a consequence of human agriculturalmanipulations. Under natural conditions,viruses are probably closely adapted to theirhosts and cause very little in the way ofobvious disease (see Chapter 4). Thus, casualinspection of plants growing in their naturalhabitat may give little indication of theviruses that might be present.• The selection of “standard” test plants forviruses is to a great extent governed by thosespecies that are easy to grow in glasshousesand to handle for mechanical and insectvector inoculation.• The genera and species chosen for a hostrange study may not form a taxonomicallybalanced selection. Most virologists whowork in the north temperate zone use mainlyfestucoid grasses in host range studies,whereas in other parts of the world,nonfestucoid groups predominate in theflora and in agricultural importance.B. Patterns of Host RangeIn spite of the preceding limitations, somegeneral points can be made. Different virusesmay vary widely in their host range. At oneextreme BSMV is virtually confined to barleyas a host in nature. At the other, CMV caninfect more than 1,300 species in more than100 botanical families, and Tomato spotted wiltvirus (TSWV) has a host range of over 800 speciesbelonging to 80 botanical families.C. The Determinants of Host RangeOn the basis of present knowledge there arefour possible stages where a virus might beblocked from infecting a plant and causingsystemic disease: during initial events—theuncoating stage; during attempted replicationin the initially infected cell; during movementfrom the first cell in which the virus replicated;and by stimulation of the host’s cellulardefenses in the region of the initial infection.1. Initial EventsThe initial event for any infection is the recognitionof a suitable host cell or organelle. Bacterialviruses and most of those infectingvertebrates and invertebrates have specific proteinson their surface that act to recognise a proteinreceptor on the surface of a susceptible hostcell. The surface proteins on plant rhabdo-, reo-,and tospo-viruses may have such a cell recognitionfunction that is unlikely to be of use inplants but is likely to be important in recognisingreceptors of insect vectors. No evidenceexists for plant cell recognition receptors on thesurface of any of the ssRNA plant viruses; however,there are receptors on the surface of RNAviruses that have a circulative interaction withtheir biological vector (see Chapter 12). The evidenceavailable for these small viruses suggeststhat host range is usually a property of the RNArather than the protein coat. When it has beentested, the host range of a plant virus is the samewhether intact virus or the RNA is used as inoculum.Surface recognition proteins may be of littleuse to a virus in the process of infecting aplant because they must enter cells throughwounds on the plant surface. Various lines ofevidence suggest that there is little or no hostspecificity in the uncoating process. Thus, bothTMV and TYMV are uncoated as readily innonhosts as in host species.2. Expression and ReplicationIt has already been noted that the majority ofplant viruses have single-stranded (þ)-strandgenomes and, as will be discussed in Chapter7, the first events on entering a cell are thatthe virus particles are uncoated and the genomicRNA is translated. Following inoculationof TMV to plant species considered as nonhosts,viral RNA has been found in polyribosomes;furthermore, TMV particles uncoat andexpress their RNA in Xenopus oocytes.I. INTRODUCTION TO PLANT VIRUSES

40 2. OVERVIEW OF PLANT VIRUSESHowever, evidence exists that specific viralgenes involved in replication may also beinvolved as host range determinants. Varioushost proteins have been found in replicationcomplexes of several viruses (see Chapter 8).The replication of many viruses takes place inassociation with particular cell organelles, andtherefore recognition of a particular organelleor site within the cell by a virus (or by somesubviral component or product) must be a frequentoccurrence. Plant viruses may haveevolved a recognition system basically differentfrom that of viruses of vertebrates and invertebratesthat normally encounter and recognisetheir host cells in a liquid medium or at aplasma membrane surface. This recognitioncan be effected in two ways: by the virusgenome-encoding products that recognise specificcell components or by complementationbetween two coinfecting viruses. The latter caseis exemplified by squash leaf curl disease in theUnited States, which is caused by two distinctbut highly homologous bipartite geminiviruses.The host range of one virus is a subset of theother. Virus replication is involved in thehost restriction of one of the viruses, and thereplication of the restricted virus is rescued intrans by coinfection with the nonrestrictedvirus. Sequence analysis revealed that therestricted virus had a 13-nucleotide deletion inthe common region. In other respects, thesequences of the two common regions werealmost identical.3. Cell-to-Cell MovementAs we will see in Chapter 9, plant virusesmove from cell to cell through cytoplasmic connections(plasmodesmata), and this processrequires a functional cell-to-cell movementprotein. Thus, this is one of the factors thatdetermines whether a particular virus can giverise to readily detectable virus replication in agiven host species or cultivar. The situationwhen a virus can replicate in the initiallyinfected cell but not move to adjacent cells istermed subliminal infection. A functional cellto-cellmovement protein may not be the onlyrequirement for systemic movement of a virus,and other gene products may be involved atthis stage.4. Stimulation of Host-Cell DefensesPlants contain various defense systemsagainst viruses, such as RNA silencing andhypersensitive response (see Chapters 10 and11) which the virus must overcome to establisha successful infection.VIII. VIRUSES IN OTHERKINGDOMSMany obvious differences that are related tothe natures of the organisms are apparent invirus infections in members of different kingdoms.Factors such as host range are easier tounderstand than others.IX. SUMMARY• Plant viruses cause considerable economiclosses.• Local symptoms occur in inoculated leavesand are usually either necrotic or chloroticlesions or ringspots• Systemic symptoms include mosaics andmottles, general chorosis, distortions ofleaves and stems, wilting and stunting, andnecrosis.• Virus infection of plants causes varioushistological changes of both plant cells andintracellular structure.• Viruses can form recognisable changes to thecell, including aggregates of virus particleand viral proteins.• Plant viral host ranges are controlled byvarious factors, including no virus-planthost cell receptor recognition system andlittle host specificity in the initial uncoatingI. INTRODUCTION TO PLANT VIRUSES

IX. SUMMARY41process; mechanisms that block virusreplication in the cells where the virus firstgained entry; the ability of the virus to movefrom the first infected cell; and the hostdefence system.ReferencesAnderson, P.K., Cunningham, A.A., Patel, A.G., Morales,F.J., Epstein, P.R., and Daszak, P. (2004). Emerging infectiousdiseases in plants: pathogen pollution, climatechange, and agrotechnology drivers. Trends Ecol. Evolut.19, 535–544.Hull, R. and Plaskitt, A. (1974). The in vivo behaviour ofBroad bean wilt virus and three of its strains. Intervirol.2, 352–359, S. Karger AG, Basel.Hull, R. and Davies, J.W. (1992). Approaches to nonconventionalcontrol of plant virus diseases. Crit. Rev. Plant Sci.11, 17–33.Prins, M. and Goldbach, R. (1996). RNA-mediated resistancein transgenic plants. Arch Virol. 141, 2259–2276.Waterworth, H.E. and Hadidi, A. (1988). Economic lossesdue to plant viruses. In Plant virus disease control(A. Hadidi, R.K. Khetarpal, and H. Koganezawa, Eds.),pp. 1–13. APS Press, St. Paul.Further ReadingHull, R. (2002). Matthew’s plant virology. Academic Press,San Diego.I. INTRODUCTION TO PLANT VIRUSES

This page intentionally left blank

C H A P T E R3Agents That Resemble or AlterPlant Virus DiseasesVarious agents that resemble virus infections can induce diseases in plants, and other agentscan alter virus diseases. The former include viroids and phytoplasma, and the latter comprisesmall nucleic acids associated with plant pathogenic viruses. Essentially, most fall into twogroups—viroids and phytoplasma—that can replicate autonomously and satellite nucleic acidsand defective interfering (DI) nucleic acids that require a functional virus for their replication.O U T L I N EI. Viroids 43II. Phytoplasma 50III.Satellite Viruses and SatelliteRNAs 51IV.Defective and Defective InterferingNucleic Acids 58V. Viruses of Other Kingdoms 60VI. Summary 61I. VIROIDSVarious important virus-like diseases inplants have been shown to be caused by pathogenicRNAs known as viroids, which have thefollowing basic properties:• Viroids are small circular molecules, a fewhundred nucleotides long, with a highdegree of secondary structure.• Viroids do not code for any polypeptidesand replicate independently of anyassociated plant virus.• Viroids are the smallest known selfreplicatinggenetic unit.Viroids are of practical importance as theycause several economically significant diseasesand are of general biological interest as beingamong the smallest known agents of infectiousdisease. The most studied viroid is Potato spindleComparative Plant Virology, Second Edition43Copyright # 2009, Elsevier Inc. All rights reserved.

44 3. AGENTS THAT RESEMBLE OR ALTER PLANT VIRUS DISEASEStuber viroid (PSTVd). Viroid names are abbreviatedto initials with a “d” added to distinguishthem from abbreviations for virus names.A. Classification of ViroidsBased on the sequence and predicted structuresof their RNAs, the present 29 known viroidsare classified into two families: thePospiviroidae and the Asunviroidae; each familyhas several genera (Table 3.1).B. Pathology of Viroids1. Macroscopic Disease SymptomsViroids infect both dicotyledonous and monocotyledonousplants. As a group, there is nothingthat distinguishes the disease symptoms producedby them from those caused by viruses.Their symptoms include stunting, mottling, leafdistortion, and necrosis and range from theslowly developing lethal disease in coconutpalms caused by Coconut cadang-cadang viroid(CCCVd) to the worldwide symptomless infectionof Hop latent viroid (HLVd).2. Cytopathic EffectsVarious effects of viroid infection on cellularstructures have been reported. For example, insome infections changes have been observedin membranous structures called plasmalemmasomes.Several workers have described pronouncedcorrugations and irregular thicknessin cell walls of viroid-infected tissue. VariousTABLE 3.1Classification of Viroids aFamily Genus Viroid Abbreviation Size (nt)Pospiviroidae Pospiviroid Potato spindle tuber PSTVd 356–359Chrysanthemum stunt CSVd 355–356Citrus exocortis CEVd 369–373Tomato apical stunt TASVd 360Tomato planta macho TPMV 360Hostuviroid Hop stunt HSVd 296–303Cocadviroid Coconut cadang-cadang CCCVd 246Citrus viroid IV CVd-IV 284Hop latent HLVd 256Apscaviroid Apple scar skin ASSVd 329–331Apple dimple fruit ADFVd 306Citrus viroid III CVd-III 294Grapevine yellow speckle viroid 1 GYSVd-1 367Grapevine yellow speckle viroid 2 GYSVd-2 363Coleviroid Coleus blumei viroid 1 CbVd-1 248–251Coleus blumei viroid 2 CbVd-2 301–302Coleus blumei viroid 3 CbVd-3 361–364Avsunviroidae Avsunviroid Avocado sunblotch ASBVd 247Pelamoviroid Peach latent mosaic PLMVd 337–338Chrysanthemum chlorotic mottle CChMVd 399a Only some representative viroid species listed. For more recent full lists, see Fauquet et al. (2005) and Flores et al. (2005).I. INTRODUCTION TO PLANT VIRUSES

I. VIROIDS45degenerative abnormalities have been found inthe chloroplasts of viroid-infected cells.3. Location of Viroids in PlantsUsing confocal laser scanning microscopyand transmission electron microscopy in conjunctionwith in situ hybridisation, both Citrusexocortis viroid (CEVd) and CCCVd were foundin vascular tissues as well as mesophyll cells.From experiments involving fractionatingcomponents of viroid-infected cells, it has becomegenerally accepted that most viroids are locatedin the nucleus. The main exception is Avocadosun blotch viroid (ASBVd), which is found in chloroplasts.Within nuclei, PSTVd and CCCVd arelocated in nucleoli, whereas CEVd accumulatesto higher concentrations in the nucleoplasm.4. Movement in the PlantViruses with defective coat proteins andnaked RNAs move slowly through the plant bycell-to-cell movement (see Chapter 9). By contrast,viroids move rapidly from cell to cell of ahost plant in the manner of competent viruses.The cell-to-cell movement is via plasmodesmataand is mediated by specific sequences or structuralmotifs. Long-distance movement of viroidsis almost certainly through the phloem. The relativeresistance of viroid RNA to nuclease attackprobably facilitates their long-distance movement.It is also possible that viroid particlesundergoes RNA translocation while bound tosome host protein.5. TransmissionViroids are readily transmitted by mechanicalmeans in most of their hosts. Transmissionin the field is probably mainly by contaminatedtools or similar means. This ease of transmissionof an RNA molecule in the presence ofnucleases is probably due to viroid secondarystructure and to the complexing of viroids tohost components during the transmission process.Several viroids have been shown to bepollen and seed transmitted in tomato, potato,and grapes.6. EpidemiologyThe main methods by which viroids arespread through crops are by vegetative propagation,mechanical contamination, and pollenand seed. The relative importance of thesemethods varies with different viroids andhosts. For example, vegetative propagation isdominant for PSTVd in potatoes and Chrysanthemumstunt viroid in chrysanthemums.Mechanical transmission is a significant factorfor others, such as CEVd in citrus and HSVd inhops. Seed and pollen transmission are factorsin the spread of ASBVd in avocados.For most viroid diseases, the reservoir ofinoculum appears to be within the crop itself,which raises the question as to where theviroid diseases came from. The evidence suggeststhat many viroid diseases are of relativelyrecent origin. None of the recognisedviroid diseases was known to exist before1900, and many were first described since1940. The sudden appearance and rapidspread of a new viroid disease can probablybe accounted for by viroids being readilytransmitted by mechanical means and manymodern crops being grown as large-scalemonocultures. Thus, from time to time aviroid present in a natural host and probablycausing no disease might escape into a nearbysusceptible commercial crop and spread rapidlywithin it. If the viroid and crop planthad not evolved together, disease would bea likely outcome. There is direct evidence forsuch a sequence of events with the tomatoplanta macho disease in Mexico.C. Properties of Viroid RNAsThe properties of the RNAs of the twofamilies of viroids are summarised in Table 3.2.1. Sequence and StructureThe nucleotide sequences of most membersof the viroid group and those of numerousviroid variants are now known. They range insize from 246 to 375 nucleotides (see Table 3.1).I. INTRODUCTION TO PLANT VIRUSES

46 3. AGENTS THAT RESEMBLE OR ALTER PLANT VIRUS DISEASESTABLE 3.2Properties of Viroid RNAsPospiviroidaeAvsunviroidaeShape of molecule Circular CircularSecondary structure Rodlike, about 37 nm long BranchedCentral conserved region Present AbsentSite of replication Nucleus ChloroplastMechanism of replication Rolling circle, asymmetrical pathway Rolling circle, symmetrical pathwayReplication enzymeNuclear DNA-dependent RNApolymerase IIChloroplastic DNA-dependent RNApolymeraseCleavage of product of replication Host nuclear RNase Hammer-head ribozymeMembers of both families have circular RNAmolecules, but those of the Pospiviroidae arerodlike, whereas those of the Avsunviroidae arebranched (Figure 3.1).It should be remembered that these structureshave been derived either from computerpredictions or from in vitro experiments andthat, in vivo, viroids may be associated withABCFIGURE 3.1 Structure of viroids. A. Three-dimensional representation of viroid molecule [Reprinted from Matthews’Plant Virology, 4th ed., R. Hull, Viroids, Satellite Viruses and Satellite RNAs, pp. 593–626, Copyright (2002), with permissionfrom Elsevier]; B. Secondary structure of PSTVd showing the functional domains; C. Proposed secondary structure of Peachlatent mosaic viroid. The predicted self-cleavage sites are indicated, (þ) strand filled arrow, ( ) strand open arrow. [This articlewas published in Encyclopedia of Virology, R. Owens (A. Granoff and R.G. Webster, Eds.), Viroids, pp. 1928–1937,Copyright Elsevier (1999).]I. INTRODUCTION TO PLANT VIRUSES

I. VIROIDS47host proteins and have other structures. However,other evidence points to at least a partialrod-shaped structure in vivo in that viableduplications or deletions preserve this type ofstructure. Viroids have tertiary structure thatis thought to be important in interactions withhost proteins.The predicted rodlike structures of the Pospiviroidaehave five structural-functional domainsthat are common to all members (Figure 3.1B;see Box 3.1). These were thought to have specificfunctions, but the situation is now consideredto be more complex. For instance,symptom expression is thought to be controlledby determinants located within the T L , P, V,and T R domains.2. ReplicationEven if it was assumed that the three outof-phasepotential reading frames were fully utilized,viroids do not contain enough informationto code for an RNA replicase. It is now generallyaccepted that viroids are not translated to giveBOX 3.1DOMAINS INPOSPIVIROIDAE RNAThe domains on PSTVd are illustrated inFigure 3.1.C (central domain): This contains a centralconserved region (CCR) of about 95nucleotides.P (pathogenicity domain): Implicated inpathogenesis.T L and T R (left and right terminaldomains): Implicated in replication.V (variable domain): Varies between speciesoften showing

48 3. AGENTS THAT RESEMBLE OR ALTER PLANT VIRUS DISEASESFIGURE 3.2 Models for viroid replication;see text for details. [This article was publishedin Sem. Virol., 8, R. Flores, F. Di Serio, andC. Hernandez, Viroids: the non-coding genomes,pp. 65–73, Copyright Elsevier (1997).]• Processing of the long products of rolling circlereplication of members of the Pospiviroidae togive monomeric (þ) RNAs is effected by a hostRNase activity in the nucleolus.• Cleavage of the long replication productsof members of the Asunviroidae is by ahammerhead ribozyme (see Box 3.2)• Circularisation of the monomericmolecules of nuclear viroids is thoughtto be calalysed by host RNA ligase.It is unclear if a chloroplastic RNAligase exists for members of theAvsunviroidae or whether the reaction isautocatalytic.I. INTRODUCTION TO PLANT VIRUSES

I. VIROIDS49BOX 3.2HAMMERHEAD RIBOZYMESA hammerhead ribozyme is a small catalytic RNA motif that catalyzes self-cleavage reaction. Its namecomes from its secondary structure, which resembles a carpenter’ hammer (see figure). The hammerheadribozyme is involved in the replication of the Avsunviroidae viroid and some satellite RNAs.Fig. Hammerhead structures formed byAvsunviroidae that cleave replication ( )-sense and (þ)-sense replicationintermediates. Conserved nucleotides are boxed and cleavage site indicated by arrowhead. This article was publishedin Sem. Virol., 8, R. Flores, F. Di Serio, and C. Hernandez, Viroids: the non-coding genomes, pp. 65–73,Copyright Elsevier (1997).All hammerhead ribozymes contain three base-paired stems and a highly conserved core of residuesrequired for cleavage. The cleavage reaction proceeds by an attack of a 2 0 hydroxyl oxygen of acatalytic site cytosine on the phosphorus atom attached to the 3 0 carbon of the same residue, breakingthe sugar phosphate backbone and producing a 2 0 ,3 0 cyclic phosphate. As for protein ribonucleases,a metal ion bound in the active site (Mg þþ ) stabilises the ionized form of the 2 0 hydroxyl oxygen,promoting the catalytic attack.Artificial ribozymes can be made to enable RNA to be cleaved at a predetermined site. Thehammerhead motif is prepared in vitro by synthesising two separate strands and combining themtogether to give rise to the self-cleavage reaction.3. Recombination Between ViroidsThe nucleotide sequence data make it highlyprobable that recombination events have takenplace in the past between different viroids, presumablyduring replication in mixed infections.For example, Tomato apical stunt viroid appearsto be a recombinant viroid made up of mostof the sequence of a CEVd-like viroid but withthe T2 domain replaced by a T2 domain froma PSTVd-like viroid.4. Interference Between ViroidsInoculation of tomato plants with a very mildstrain of PSTVd gives substantial protectionI. INTRODUCTION TO PLANT VIRUSES

50 3. AGENTS THAT RESEMBLE OR ALTER PLANT VIRUS DISEASESagainst a second inoculation with a severe strainapplied two weeks after the first. However, themild strain must be established first. This protection,termed cross-protection (see Chapter 10),occurs not only between strains of a particularviroid but also between different viroids. Forexample, when PSTVd RNA transcripts from acloned PSTVd DNA are inoculated together withHSVd RNA, the PSTVd reduces the level ofHSVd RNA in infected plants. Plants inoculatedwith dual transcripts—two copies of a severePSTVd strain linked to two of HSVd—developPSTVd symptoms and only PSTVd progenyRNA can be detected.D. Molecular Basis for BiologicalActivityBecause of their very small size, their autonomousreplication, the known structure of manyvariants, and the lack of any viroid-specific polypeptides,it has been hoped by many workers thatviroids might provide a simple model system thatwould give insights as to how variations in thestructure of a pathogen modulate disease expression.Very small changes in nucleotide sequencemay give rise to dramatic changes in the kind ofdisease induced by a viroid, and, therefore, diseaseinduction must involve specific recognitionof the viroid sequence by some host macromolecule(s).Until the nature of the host macromolecule(s)is known, the interpretation ofcorrelations between nucleotide sequence andbiological properties of viroids will remain speculative.At present, the only practicable biologicalproperties that can be observed are infectivityand severity of disease, although viroid-bindingproteins are starting to be recognised.Sequence comparison between viroid variantsand site-directed mutagenesis has identifiedthe P domain but has shown theimportance of viroid structure, with many mutationsrendering it noninfectious. Comparisonswith host RNA sequences reveal similaritiesbetween viroid sequences and ribosomal RNAprocessing sequences, suggesting the potentialfor interference. Some host proteins have beenfound to bind to viroids, though their role inpathogenesis has not yet been elucidated. A proteinof 602 amino acids from tomato has beensuggested to be involved either in the transportof PSTVd RNA into or from the nucleus or inthe formation of the replication complex. Thediscovery of viroid-specific siRNAs has led tothe suggestion that viroids could be inducers ofRNA silencing (see Chapter 11 for RNA silencing).It is thought that RNA silencing couldmediate the pathogenicity of viroids.E. Diagnostic Procedures for ViroidsSince viroids produce no specific proteins,the immunological methods applied so successfullyto viruses cannot be used for the diagnosisof diseases that they cause. Similarly, because nocharacteristic particles can be detected, electronmicroscopic techniques are inappropriate. Forthese reasons diagnostic procedures have beenconfined to biological tests, gel electrophoresis,PCR, and nucleic acid hybridisation tests.II. PHYTOPLASMASome important virus-like diseases arecaused by bacteria that lack a cell wall and areobligate parasites of the phloem tissue. Thereare two genera of such organisms: Phytoplasmaand Spiroplasma.Phytoplasma were formerly known asMycoplasma-like organisms, or MLOs. Theyhave a pleomorphic or filamentous shape witha diameter of less than 1 mm and very smallgenomes. Taxonomically they belong to theorder Acholeplasmatales; the genus name of Phytoplasmais not yet formally recognised and iscurrently at Candidatus stage.Two common symptoms of phytoplasmainfection are phyllody—the production of leaflikestructures in place of the floral parts andyellowing of the leaves due to the presence ofthe pathogen in the phloem disrupting theI. INTRODUCTION TO PLANT VIRUSES

III. SATELLITE VIRUSES AND SATELLITE RNAS51transport of carbohydrates. Many phytoplasma-infectedplants have a bushy appearancedue to changes in growth patterns thatinduce proliferation of axillary side shoots andincrease in size of internodes, giving witches’broom symptoms. This symptom can be ofcommercial use—for instance, it can enablethe production of poinsettia plants that havemore than one flower. Phytoplasma arerestricted to phloem tissue, and it is consideredthat the symptoms of infected plants are due todisturbance of phloem functions, includingupsetting the hormone balance as well asinducing host defence proteins.Phytoplasma are mainly transmitted by leafhoppers(family Cicadellidae) and planthoppers(family Fulgoridae) in a propagative circulativemanner (see Chapter 12 for virus-vector relationships).They infect the vector and thus arepathogens of both the plant and insect. Phytoplasmaare also transmitted by graftinginfected material into healthy plants and bytaking cuttings from infected plants.The genome sizes of sequenced phytoplasmarange from 530 kb to 1,350 kb, encoding between671 and 754 genes; they also contain plasmids.They lack many of the genes essential for cellmetabolism, such as those for de novo synthesisof amino acids, fatty acids, or nucleotides, andthus they rely on the uptake of nutrients fromboth their plant and insect hosts.Diagnosis of phytoplasma infection is usuallyby molecular techniques such as PCR. Control ismost frequently by breeding and deployment ofresistant varieties and by control of the insectvector. For long-term valuable perennials suchas trees, the application of tetracycline hasproved effective but very expensive.Spiroplasma resemble phytoplasma in manyrespects but have a distinct helical morphology.They can also be cultured in vitro, requiring arich culture medium, and they grow well at30 o C but not at 37 o C. Taxonomically theybelong to the order Entomoplasmatales, genusSpiroplasma. The Spiroplasma genome size is780–2,200 kb and, as with Phytoplasma, lacksmany of the genes requires for full cell metabolism.They contain genes that confer helicityand motility, but these properties are notrequired for pathogenicity or transmission.Two important diseases caused by spiroplasmaare citrus stubborn disease due to infectionby Spiroplasma citri and corn stunt causedby Spiroplasma kunkelii. The symptoms of citrusstubborn are shorter and broader cuppedleaves and the development of witches’ broomsymptoms. Fruiting is suppressed and fruitquality affected. In corn stunt infections theplants are stunted, have broad chlorotic stripeson the leaves, and have poor filling of the ears.Both these diseases are transmitted by leafhoppers,and diagnosis and control are similar tothat for phytoplasma.III. SATELLITE VIRUSES ANDSATELLITE RNASSatellites are subviral agents that lack thegenes encoding functions that are necessaryfor their replication. They depend for their multiplicationon coinfection of a host cell with ahelper virus. The genomes of satellites differsubstantially or totally from those of theirhelper virus.Two major classes of satellite agents can be distinguishedaccording to the source of the coat proteinused to encapsulate the nucleic acid. Insatellite viruses, the satellite nucleic acid codesfor its own coat protein. In satellite RNAs orDNAs, the nucleic acid becomes packaged in proteinshells made from coat protein of the helpervirus. Satellite viruses and satellite nucleic acidshave the following properties in common:• Their genetic material is a nucleic acidmolecule of small size. The nucleic acid isnot part of the helper virus genome andusually has little sequence similarity to itapart from the terminal regions.• Replication of the nucleic acid is dependenton a specific helper virus.I. INTRODUCTION TO PLANT VIRUSES

52 3. AGENTS THAT RESEMBLE OR ALTER PLANT VIRUS DISEASES• The agent affects disease symptoms, at leastin some hosts.• Replication of the satellite interferes to somedegree with replication of the helper.• Satellites are replicated on their own nucleicacid template.Both RNA and DNA viruses have satellitesassociated with them—those of RNA virusesbeing RNA and those of DNA viruses beingDNA. Satellite viruses and nucleic acids canbe categorised into six groups based on theirproperties (Table 3.3). Not included in the tableare defective RNAs and DNAs, which havesequences derived from those of the helpervirus; these are described in detail later in thischapter. The coat-dependent RNA repliconsare included because their sequences differfrom those of the helper virus.Several satellite RNAs associated with a particulargroup of viruses have been shown tohave viroid-like structural properties. Theseagents have been termed virusoids, but thisterm is no longer favoured. There appears tobe no taxonomic correlation between theviruses that are associated with satellites, andsatellitism seems to have arisen a number oftimes during virus evolution. Some virusesare associated with more than one satellite,and satellites can even require a second satelliteas well as the helper virus for replication.A. Satellite Plant Viruses (A-type)Satellite tobacco necrosis virus (STNV) wasthe first satellite virus to be recognised; theterm satellite virus was coined to denote the17 nm isometric particles associated with the30 nm Tobacco necrosis virus (TNV) isometricparticles. The helper Necrovirus TNV replicatesindependently of other viruses and normallyinfects plant roots in the field. There is significantspecificity in the relationship between satelliteand helper. Strains of both viruses havebeen isolated, and only certain strains of thehelper virus will activate particular strains ofthe satellite.Both STNV and TNV are transmitted by thezoospores of the fungus Olpidium brassicaeTABLE 3.3Satellite Viruses, Nucleic Acids, and Coat-Dependent RNA RepliconsSatellite TypeSatelliteHelperVirus aIndependentReplicationDirects ProteinSynthesisCoatProteinSatellite virus genomic RNA(A type)Messenger satellite RNA(B type)Nonmessenger satellite RNA(C type)Nonmessenger satellite RNA(D type)Satellite DNACoat-dependent RNArepliconSTNV genomicRNABsatTomBRVRNATNV No Yes OwnTomBRV No No HelpervirusSCMV RNA CMV No No HelpervirusSTRSV RNA TRSV No No HelpervirusBegomovirusDNAbNo No HelpervirusST9aRNA Yes Yes Helpervirusa Abbreviations of viruses: CMV ¼ Cucumber mosaic virus; TNV ¼ Tobacco necrosis virus; TomBRV ¼ Tomato black ring virus;TRSV ¼ Tobacco ringspot virus.I. INTRODUCTION TO PLANT VIRUSES

III. SATELLITE VIRUSES AND SATELLITE RNAS53(Chapter 12). Transmission depends on anappropriate combination of four factors: satelliteand helper virus strains, race of fungus,and species of host plant.The complete nucleotide sequence of STNVRNA was one of the first viral sequences to bedetermined and has no significant similaritywith that of the TNV genome. The STNVgenome contains only one open reading frame(ORF): that for its coat protein. The 5 0 terminusof STNV is unlike that of most ss (þ)-senseRNA plant viruses (see Chapter 6 for RNA terminalstructures). There is neither a 5 0 cap nor aVPg, and the 5 0 termination is 5 0 -ppApGpUp-.The 3 0 -terminal region can be folded to give atRNA-like structure with an AUG anticodon,but there is no evidence that this structure canaccept methionine.Relatively little is known about the replicationof STNV in vivo, but it is widely assumedthat the RNA replication must be carried outby an RNA-dependent RNA polymerasecoded for, at least in part, by the helper virus.It is assumed that replication is through a (–)strand in a manner similar to that of (þ)-strand RNA viruses (see Chapter 8). Replicationof STNV substantially suppresses TNVreplication, and it is thought likely that thisinvolves competition for the replicase. Thepresence of STNV in the inoculum reducesthe size of the local lesions produced by thehelper virus, possibly by reduction in TNVreplication.B. Satellite RNAs (satRNAs)As Table 3.3 shows, there are three classes ofsatellite RNAs. These do not encode their owncoat protein and are encapsidated in the helpervirus coat protein. They are grouped on size,the large ones with mRNA activity beingtermed B type and the smaller ones withoutmRNA activity being divided on form: thosewith linear molecules (C type) and those withcircular molecules (D type).1. Large Satellite RNAs (B-type)As with the other satellite RNAs, B-typesatellites are dependent on the helper virusfor replication and encapsidation, but in contrastto the other types they direct protein synthesisboth in vitro and in vivo from a singleORF. The RNAs of the B-type satellites thusfar studied are 0.7 kb or larger. Most of the B-type satellites (given the prefix “Bsat”) arefound associated with nepoviruses, and onehas been found associated with a potexvirus.The B satellite of Tomato black ring virus(BsatTBRV) is an RNA of 1,374 nucleotides thatis packaged in varying numbers in particlesmade of the helper virus coat protein, givingrise to a series of components of differing buoyantdensity in solutions of CsCl. Like the helpervirus, the satellite RNA is transmitted throughthe seed and also by the nematode vector ofthe virus. The satellite does not appear to affectreplication of the helper virus or to modifysymptoms, except that the number of locallesions on Chenopodium amaranticolor is reduced.BsatTBRV RNA contains one ORF with acoding capacity for a protein of 419–424 aminoacids, has a VPg at the 5 0 terminus, and is polyadenylatedat the 3 0 end (see Chapter 6 forRNA terminal structures). Mutagenesis studiessuggest that this protein region and the 5 0 and3 0 untranslated regions are all necessary forthe replication of the Bsat. It is most likely thatB satellites replicate in the same manner as thehelper virus.2. Small Linear Satellite RNAs (C-type)C-type satellites have linear RNA moleculesthat are generally smaller than 0.7 kb and donot have mRNA activity. They are found associatedwith cucumoviruses, tombusviruses,and carmoviruses and can cause severe diseaseoutbreaks (Box 3.3).There are many natural variants of the satelliteassociated with Cucumber mosaic virus(satCMV), all differing in sequence from CMVI. INTRODUCTION TO PLANT VIRUSES

54 3. AGENTS THAT RESEMBLE OR ALTER PLANT VIRUS DISEASESBOX 3.3SEVERE DISEASE IN TOMATO CAUSED BY A SATELLITEOF CUCUMBER MOSAIC VIRUS (CMV)In 1972, a devastating outbreak of a lethalnecrotic disease of field-grown tomatoesoccurred in the Alsace region of France. It wasdiscovered that CMV was involved in thisoutbreak, but it was not clear why necrosisoccurred instead of the usual fern-leaf symptoms.An additional small RNA, called satCMVor CARNA5, that was not part of the viralgenome was identified and when added toCMV isolates caused lethal necrotic disease intomatoes. Similar outbreaks in tomatoes insouthern Italy have been shown to be due to anecrogenic isolate of satCMV. However, mostCMV satellites attenuate the symptoms, and theoverall effect depends on the combination of satelliteRNA and helper virus strain.itself. They vary in length from about 330 to 390nucleotides. It is probable that most satelliteRNA preparations consist of populations ofmolecules with closely related sequences alldependent on the CMV genome. Packaging inCMV coat protein enables the satellite RNA tobe transmitted by aphids that transmit CMV.Like the helper genomic CMV RNAs, theyare all capped with M 7 Gppp at their 5 0 terminiand have a hydroxyl group at the 3 0 end (seeChapter 6 for RNA terminal structures). A 3 0 -terminal tRNA-like structure is possible, butattempts to aminoacylate the RNA have beenunsuccessful. The RNA has a high degree ofsecondary structure, which may account forthe stability of the RNA both in vitro andin vivo and for its relatively high specificinfectivity.The carmovirus Turnip crinkle virus (TCV)supports a family of satellite RNAs. Two ofthe satellites, D and F, do not affect symptoms,while C is virulent, intensifying TCV symptomsin turnip. The virulent satellite C is of particularinterest, since, unlike other knownsatellites, it appears to be a recombinantbetween D satellite and a defective interferingRNA (see following) of the helper virus TCV(Box 3.4). As with B satellites, it is thought thatC satellites replicate in the same manner as thehelper virus.3. Small Circular Satellite RNAs (D-type)D-type satellite RNAs are small, about 350–450 nucleotides, and occur as circular as wellas linear molecules. They are found associatedwith nepoviruses, luteoviruses, and sobemoviruses.Several nepoviruses support the replicationof satellite RNAs, which becomeencapsulated in particles made of the helpervirus coat protein. As well as the large B-typesatellites described previously, small ones withcircular RNA molecules also exist. The satelliteof Tobacco ringspot virus (sTRSV) consists of asmall RNA species of about 359 nucleotides ina protein shell identical to that of the helpervirus. Twelve to 25 satellite RNA moleculesbecome packaged in a single particle. The satellitecannot replicate on its own, and it interfereswith the replication of TRSV. There are noORFs, and the RNA is not translated in vitro.The satellite RNA associated with the luteovirusCereal yellow dwarf virus (sRPV) reduces theaccumulation and symptom production of thehelper virus in oats. This satellite also has multimericlinear and circular forms together with ahammerhead ribozyme sequence (see Box 3.2).I. INTRODUCTION TO PLANT VIRUSES

III. SATELLITE VIRUSES AND SATELLITE RNAS55BOX 3.4TURNIP CRINKLE VIRUS SATELLITE CFig. Structure of TCV satellite RNA C showing the variousdomains. From Hull (2002).TCV satellite C is 355 bases (see figure). The189 bases at the 5 0 end are homologous to theentire sequence of a smaller nonvirulent TCV satelliteD RNA. The rest of the RNA C molecule(166 bases in the 3 0 region) is nearly identical insequence to two regions at the 5 0 end of the TCVhelper genome. Thus, in a structural sense atleast, TCV satellite RNA C is a hybrid between asatellite RNA and a DI RNA. When plants areinoculated with a mixture of TCV, satellite RNAD, and satellite RNA C transcripts containingnonviable mutations in the 5 0 domain, recombinantsatellite RNAs are recovered. Sequenceanalysis around 20 recombinant junctions supportsa copy-choice model for this recombination.In this model, while in the process of replicating(–) strands, the replicase can leave the templatetogether with the nascent (þ) strand and canreinitiate synthesis at one of two recognitionsequences on the same or a different template.Four viruses described from Australia,which belong in the Sobemovirus genus, havebeen shown to contain small RNAs that occurin both circular and linear forms. The smallRNAs have been shown to have the biologicalproperties of satellite RNAs. The helper viruscan replicate independently, while the smallRNA cannot. The satellite RNAs, which rangein size from about 325 to 390 nucleotides, areencapsidated in both linear and closed circularforms. They show no sequence similarity withthe helper RNAs.There is quite strong evidence that D satellitesare replicated by a rolling circle mechanism asjust described for viroids. The cleavage ofthe multimeric forms appears to be autolytic bya phospho-transfer reaction between an adenylateresidue and the 2 0 hydroxyl of a neighboringcytidylate, to give rise to a 5 0 hydroxyl terminaladenosine and a 3 0 cytosine 2 0 ,3 0 -cyclic phosphodiestergroup. The viroid-like satellite RNAsare predicted to form hammerhead-shapedself-cleavage domains that resemble those ofASBVd.4. Satellite-Like RNAsMost satellites are not essential for thebiological success of the helper virus. However,it is a relatively small evolutionary step fromthis situation to one where the satellite isinvolved in the disease spread and expression.Since such a molecule cannot be described asnonessential for the helper virus, it cannot technicallybe considered a satellite; hence, they aregrouped together as “satellite-like RNAs.”a. A Satellite RNA of Groundnut RosetteVirus (GRV). The satellite associated withGRV (satGRV) is a linear ssRNA of 895–903nucleotides that relies on GRV for its replication.Although different variants of sGRV containup to five potential ORFs in either theI. INTRODUCTION TO PLANT VIRUSES

56 3. AGENTS THAT RESEMBLE OR ALTER PLANT VIRUS DISEASES(þ)- or (–)-sense strands, none of these ORFs isessential for replication of the satellite RNA.As discussed in Box 2.1, satGRV is involvedwith the two viruses causing groundnut rosettedisease and is essential for encapsidation andsymptom production within the complex.b. Ancillary RNAs of Beet Necrotic YellowVein Virus (BNYVV). Field isolates ofBNYVV contain four or five RNA components,but laboratory isolates maintained by mechanicalinoculation of Chenopodium quinoa or Tetragoniaexpansa may lack any or all of the threesmaller RNAs. Thus, the BNYVV genome comprisesRNAs 1 and 2 (Profile 2; see Appendix)together with RNAs 3, 4, and 5 that resemblesatellite RNAs. Each of the satellite-like RNAsencodes one ORF and contributes to the fieldpathology of the parent virus. RNA3 contributesto a large degree to the pathology of thevirus, whereas RNA4 greatly increases the efficiencyof virus transmission by its vector.RNA5 seems to play a role in both the symptomproduction and transmission.5. The Molecular Basis for SymptomModulationAs noted in Box 3.3, satellite RNAs can beresponsible for outbreaks of severe disease inthe field. In addition, because of their small sizeand the availability of many sequence variants,either isolated from nature or produced in thelaboratory, they are amenable to detailedmolecular study. For these reasons it has beenhoped that studies correlating molecular structureof the satellite RNAs with their effects ondisease in the plant would give us some insightinto the molecular basis of disease induction.Results so far have been difficult to interpret.This is not surprising, since multiple factorsare involved, including the strain of the helpervirus, the coat protein of the helper virus(TCV and sat RNA C), the host (satCMV), thestrain of satellite, specific sequences within thesatellite RNA, and perhaps the environment.This is exemplified by comparison of the nucleotidesequences of CMV satellite RNAs, suggestingthat only a few nucleotide changesmay be necessary to change the host responseand that different kinds of disease response(e.g., yellowing or necrosis) may be associatedwith different domains of the satellite sequence.The domain for chlorosis in tomato is in the 5 0185 nucleotides of these RNAs, while thedomainfornecrosisisinthe3 0 150 nucleotides.The sequence of the WL1-sat, which attenuatesCMV symptoms on tomatoes, differs from allnecrogenic satellite RNAs at three nucleotidespositions in the conserved 3 0 region of the RNA.This has implications in the use of satellite RNAsas a control measure to mitigate symptom expression(see Chapter 14) in that relatively few mutationscould convert a symptom attenuationsatellite strain to a necrogenic strain.A satellite RNA of particular interest in relationto disease modulation is the virulent satelliteRNA of TCV illustrated in Box 3.4. Otheravirulent strains of the satellite (e.g., strain F)lack the 3 0 sequence derived from the helperTCV genome. The 3 0 sequence derived fromTCV contains a region essential for infectivityand a larger overlapping region determiningvirulence. The domain of satellite homologycontains regions affecting monomer accumulationand modulating symptom expression.The only general conclusion to be drawn isthat changes in a disease induced by the presenceof a satellite RNA depend on changes inits nucleotide sequence. So far, there is no convincingevidence that such changes aremediated by a polypeptide translated from thesatellite RNA. Indeed, differences in diseasemodulation occur between RNAs that haveno significant ORFs. Disease modulation ismost probably brought about by specific macromolecularinteractions between the satelliteRNA and (1) helper virus RNAs, (2) host RNAs,(3) virus-coded proteins, (4) host proteins, or(5) any combination of 1–4. Until the kinds ofI. INTRODUCTION TO PLANT VIRUSES

III. SATELLITE VIRUSES AND SATELLITE RNAS57interaction that are important in disease inductionhave been established on a molecular basis, differencesin nucleotide sequence between relatedsatellites will remain largely uninterpretable.C. Satellite DNAsDNA extracted from preparations of monopartitebegomoviruses often contains moleculesthat are not the viral genomic DNA (Figure 3.3).Some of these are considered to be satelliteDNA and others defective interfering DNAs.The most studied satellite DNA is termedDNAß, which forms a complex with most, ifnot all, monopartite begomoviruses, causingsignificant disease symptoms. DNAßs areapproximately half the size of the genomicDNA (Figure 3.3), have a 5 0 stem-loop structure,one ORF (ßC1), an A-rich region, and a 3 0sequence conserved between all DNAßs. Theyshow no significant homology to their helperbegomovirus, on which they are dependent forreplication, encapsidation, and movement bothwithin and between hosts. They enhance the titreof the helper virus and are often involved withsymptom production that involves ORF ßC1.The replication of DNAß is considered to beby the rolling circle mechanism used by itshelper virus (see Chapter 8 for geminivirus replication).The replication of one DNAß speciesis supported by several begomoviruses, butthe symptom enhancement is only found inthe original helper virus/DNAß combination.The first satellite to be recognised to be associatedwith a DNA virus was a 262 nucleotidecircular ssDNA associated with the begomovirusTomato leaf curl virus (ToLCV) from northernAustralia. Replication of ToLCV-sat isdependent on the helper virus replication-associatedprotein (see Chapter 8 for geminivirusreplication-associated protein), and the satelliteDNA is encapsidated by the helper virus coatprotein. It has no significant ORFs, and the onlysignificant sequence similarity to that of theFIGURE 3.3 Genome organisation of begomoviruses and their associated subviral components. All components are circularbut are shown as linear maps for clarity. The common region (CR) is highly conserved between DNA-A and DNA-Bcomponents of a particular begomovirus. The satellite-conserved region (SCR) is highly conserved between all knownDNA-ß components. Genes encoding the coat protein (CP), replication-associated protein (Rep), transcriptional transactivatorprotein (TrAP), replication enhancer (REn), movement protein (MP), and nuclear shuttle protein (NSP) are indicated.This article was published in Virology, 344, R.W. Briddon and J. Stanley, Subviral agents associated with plant singlestrandedDNA viruses, pp. 198–210, Copyright Elsevier (2006).I. INTRODUCTION TO PLANT VIRUSES

58 3. AGENTS THAT RESEMBLE OR ALTER PLANT VIRUS DISEASEShelper virus is in two short motifs in separateputative stem-loop structures: TAATATTAC,which is conserved in all geminiviruses, andAATCGGTGTC, which is identical to a putativereplication-associated protein bindingmotif in ToLCV (Figure 3.3).D. DiscussionSatellite viruses and RNAs form specific associationswith their helper viruses. This raises twointerlinked questions: How did they evolve? andWhat is their natural function? It is thought thatsatellite viruses arose from an independent virusby degenerative loss of functions in a mixedinfection with the helper virus. It seems likelythat satellite RNAs arose from at least three differentlineages, as shown in Figure 3.4.As to function, there are two possibilities:Satellites are either “molecular parasites” orthey have a beneficial role in the biology ofthe helper viruses. As molecular parasites,satellites compete for the replication machinerywith the helper virus. However, this may notbe to the disadvantage of the helper virus.FIGURE 3.4 Hypothetical schemes for the derivation ofsatellite RNAs. [From Hull (2002).]It is possible that the attenuation of symptomsby many satellites could enhance the survivalof the host to the benefit of the helper virus.The variants that induce more severe symptoms,such as some CARNA 5 isolates, wouldbe selected against under natural conditionsbecause they kill their host and thus limit thespread of the helper virus.IV. DEFECTIVE ANDDEFECTIVE INTERFERINGNUCLEIC ACIDSTruncated, and often rearranged, versions ofgenomic viral nucleic acids are associated withmany plant and animal RNA viruses and someplant DNA viruses. These subviral nucleic acidsare termed defective interfering (DI) if theymodulate the helper virus replication or defective(D) if they have no effect. They are dispensablefor helper virus propagation, completelydependent on the helper virus for their replication,and are derived mainly by one or morepremature termination and reinitiation eventsduring the replication of the helper virus genomicnucleic acid. These subviral nucleic acidscontain all the cis-acting elements necessary forreplication by the replication system of the parentvirus but usually lack genes required forencapsidation and movement.DI nucleic acids can cause major alterationsto the normal disease progression of the helpervirus. Most moderate the symptoms of thehelper virus but some (e.g., TCV) can enhancesymptom severity. The D and DI nucleic acidsand viruses of plants are listed in Table 3.4.The deletion patterns in the DI and D RNAsand viruses fall into two groups: Group 1, inwhich the defective RNA is derived from a singleinternal deletion, and Group 2, which consistsof a mosaic of the parental viral genome(Table 3.4 and Figure. 3.5).I. INTRODUCTION TO PLANT VIRUSES

IV. DEFECTIVE AND DEFECTIVE INTERFERING NUCLEIC ACIDS59TABLE 3.4 Examples of Defective and Defective Interfering Nucleic Acids (Modified from Hull, 2002)Family/DeletedGenus Virus a SegmentType of DefectiveElementCommentsRNA VirusesReoviridae WTV RNAs 2 and 5 D virus Loss of vector transmissionTospovirus TSWV L RNA DI RNA Encodes polymerase proteinM RNA D virus Loss of viral envelope and probable loss ofvector transmissionBromoviridae BBMV RNA2 DI RNA Exacerbates symptoms in some hosts, encodesviral polymerase.CMV RNA3 D RNA Deletion in 3a protein; encodes coat protein.Tobravirus TRV RNA2 D and DI Virus Vector transmission eliminates DIFurovirus SBWMV RNA2 D Virus Loss of vector transmissionRNAs 3 and 4 D Virus Loss of vector transmission and ability to infectrootsTombusviridae TCV Various DI RNA See textDNA VirusesBegomovirus CLCuMV See DNA-1 inFig. 3.3DI DNASee texta Abbreviations for viruses: BBMV ¼ Broad bean mottle virus; CLCuMV ¼ Cotton leaf curl Multan virus; CMV¼ Cucumbermosaic virus; SBWMV ¼ Soil-borne wheat mosaic virus; TCV ¼ Turnip crinkle virus; TRV ¼ Tobacco rattle virus; TSWV ¼ Tomatospotted wilt virus; WTV¼ Wound tumor virus.FIGURE 3.5 Defective and defectiveinterfering RNAs. A. Group 1 singledeletionD RNA. CMV genomic RNA3(top) and the D RNA 3ß derived from it(bottom; the region deleted from the 3aORF is indicated. [This article was publishedin Sem. Virol., 7, M.V. Graves,J. Pogany, J., and J. Romero, Defectiveinterfering RNAs and defective virusesassociated with multipartite RNA virusesof plants, pp. 399–408, CopyrightElsevier (1996).] B. Group 2 DI RNA.The upper part shows the genome organisationof TBSV. Below are two size classesof DI RNAs with the portions derivedfrom the TBSV genomic RNA indicatedby shaded boxes with the deleted interveningregions shown as lines. [This articlewas published in Sem. Virol., 7, K.A.White, Formation and evolution of Tombovirusdevective interfering RNAs, pp.409–416. Copyright Elsevier (1996).]ABI. INTRODUCTION TO PLANT VIRUSES

60 3. AGENTS THAT RESEMBLE OR ALTER PLANT VIRUS DISEASESA. Group 1: Single Deletion D RNAsThe defective RNAs associated with CMVhave no apparent effect on symptom productionor virus accumulation. They are derivedfrom RNA3 (Figure 3.5A) by a single deletionthat removes a segment of the 3a ORF butmaintains the reading frame downstream ofthe deletion. This leaves a defective cell-to-cellmovement protein and a functional coat protein.CMV D RNAs accumulate in various Nicotianaspecies, but in tomato, zucchini squash,and muskmelon, the D RNAs only accumulatein inoculated tissue but do not move systemically.Furthermore, the D RNA accumulatesand is encapsidated in both inoculated cotyledonsand leaves of tomato and zucchini squashand accumulates but is not encapsidated onlyin the inoculated cotyledons of muskmelon.This indicates that host and tissue specificityis involved in replication, cell-to-cell movement,systemic movement, and encapsidationof CMV D RNAs.B. Group 2: Multiple Deletion D andDI RNAsMultiple deletion DI and D RNAs are characteristicof several members of the Tombusviridaefamily (Table 3.4 and Figure 3.5). These DI RNAsare about 10 to 20 percent the size of the parentgenome and have common structural featuresthat retain terminal regions and an internal segmentbut lack all coding capacity (Figure 3.5B).DI RNAs can be generated de novo by highmultiplicitypassages of transcripts from clonedhelper virus infectious cDNA. Although the DIRNA can represent 60 percent of virus-specificRNA in leaf extracts, less than 5 percent of theencapsidated RNA is DI RNA. The presenceof DI RNAs in infected plants greatly reduceshelper virus accumulation and persistentlyattenuates symptoms. Three mechanisms havebeen suggested for DI RNAs to give theseeffects: competition with the helper virus fortransacting factors necessary for replication,specific interaction with virus-encoded products,and activation of the RNAi defence system.It is possible that more than one of thesemechanisms operates in a helper virus/DIRNA interaction.C. Defective DNAs Associated withDNA VirusesBesides having satellite DNAs, some monopartitebegomoviruses also have DI DNAs.These ameliorate the symptoms and reducevirus accumulation of the helper virus. Thesecircular ssDNA molecules have been termedDNA-1 and are about half the size of the helperbegomovirus. They encode the Rep gene (seeChapter 8 for geminivirus replication) and canreplicate autonomously; however they dependon the helper begomovirus for encapsidationand movement both within and betweenplants. Thus, technically they are neither satellitesnor true DI DNA, as they do not fit thestrict definitions of either. It has been suggestedthat they are similar to the nanovirus componentencoding the Rep and have been modifiedto overcome the size constraints of geminivirusencapsidation.V. VIRUSES OF OTHER KINGDOMSMany other agents cause virus-like diseasesin other kingdoms. For instance, prions wereoriginally considered to be similar to viroidsbut are now recognised to be completely differentprotein entities. Spiroplasms have beenassociated with transmissible spongiformencepatholopies.Satellite viruses have been found in associationwith viruses of both vertebrates andinvertebrates. Hepatitis delta virus (HDV) is aI. INTRODUCTION TO PLANT VIRUSES

VI. SUMMARY61satellite of Hepatitis B virus (HBV). Like itshelper virus, HDV has an RNA genome, butit encoded only one protein: the delta antigenthat encapsidates it. It requires HBV for it replication.In invertebrates Chronic bee-paralysisassociated satellite virus has an RNA genomeencoding just its coat protein and depends onChronic bee-paralysis virus for its replication.Satellite RNAs and DNAs have only beenfound associated with plant viruses, but D andDI RNAs and DNAs are found with both plantand animal viruses.VI. SUMMARY• The effects of various agents resemble oralter plant virus diseases.• Viroids are small pathogenic moleculesthat do not encode any proteins and arethe smallest known self-replicating nucleicacid.• Phytoplasma are obligate parasites of plantsthat lack a cell wall.• Satellite viruses encode their own coatprotein but depend on the helper virus forreplication.• Various forms of satellite RNAs and DNAsexist, and each requires the helper virus forreplication, although their nucleic aciddiffers from that of the helper virus.• Defective and defective interferingnucleic acids are derived from that of thehelper virus by deletion and thus havesequence similarities to that of the helper virus.• DI RNAs and DNAs modulate thesymptoms caused by the parent virus.ReferencesFauquet, C.M., Mayo, M.A., Maniloff, J., Desselberger, U.,and Ball, L.A. (2005). Virus taxonomy. Eighth Reportof the International Committee on Taxonomy of Viruses.Elsevier, San Diego, California.Flores, R., Hernández, C., Martinéz de Alba, A.E., Daròs J.-A.,and Di Serio, F. (2005). Viroids and viroid-host interactions.Annu. Rev. Phytopathol. 43, 117–139.Hull, R. (2001). Matthews’ plant virology. Academic press,San Diego.Further ReadingBové, J.M., Renaudin, J., Saillard, C., Foissac, X., and Garnier,M. (2003). Spiroplasma citri, a plant pathogenic mollicure:Relationships with its two hosts, the plant and the leafhoppervector. Annu. Rev. Phytopathol. 41, 483–500.Briddon, R.W. and Mansoor, S. (2008). Beta ss DNA satellites.Encyclopedia of Virology, Vol. 1, 314–320.Christensen, N.M., Axelsen, K.B., Nicolaisen, M., andSchulz, A. (2005). Phytoplasmas and their interactionswith hosts. Trends in Plant Science. 10, 526–533.Ding, B. and Itaya, A. (2007). Viroid: A useful model forstudying the basic principles of infection and RNA biology.Molec. Plant Microbe Interact. 20, 7–20.Flores, R. and Owens, R.A. (2008). Viroids. Encyclopedia ofVirology, Vol. 5, 332–341.Hogenhout, S.A., Oshima, K., Ammar, E.-D., Kabizawa, S.,Kingdom, H.N., and Namba, S. (2008). Photoplasmas:bacteria that manipulate plants and insects. Molec. plant.Pathol. 9, 403–423.Lee, I.-M., Vavis, R.E., and Gundersen-Rindal, D.E. (2000).Phytoplasma: Phytopathogenic mollicutes. Annu. Rev.Microbiol. 54, 221–255.Palukaitis, P., Rezaian, A., and Garcia-Arenal, F. (2008).Satellite RNAs, satellite DNAs, and satellite viruses.Encyclopedia of Virology, Vol. 4, 526–534.Rocheleau, L. and Pelchat, M. (2006). The subviral RNAdatabase: A toolbox for viroids, the hepatitis delta virus,and satellite RNAs research. BMC Microbiology 6:24.http://www.biomedcentral.com/1471-2180/6/24.Roux, L. (2008). Defective interfering viruses. Encyclopedia ofVirology, Vol. 2, 1–4.Simon, A.E., Roosinck, M.J., and Havelda, Z. (2004). Plantvirus satellite and defective interfering RNAs: Newparadigms for a new century. Annu. Rev. Phytopathol.42, 415–437.Taylor, J.M. (2008). Hepatitis delta virus (Deltavirus). Encyclopediaof Virology, Vol. 2, 375–376.Westhof, E. and Lescoute, A. (2008). Ribozymes. Encyclopediaof Virology, Vol. 4, 475–480.I. INTRODUCTION TO PLANT VIRUSES

This page intentionally left blank

C H A P T E R4Plant Virus Origins and EvolutionLike other living entities, viruses substantially resemble the parent during their replication, butthey can change to create new types, or strains. This inherent variation enables viruses to adaptto new and changing situations. Over longer periods of time, new viruses evolve, but theremust have been a time at which the archetypical virus arose.O U T L I N EI. Introduction 63II. Virus Evolution 64III. Evidence for Virus Evolution 75IV.Coevolution of Viruses withTheir Hosts and Vectors 80V. Viruses of Other Kingdoms 81VI. Summary 81I. INTRODUCTIONFor many groups of organisms, an understandingof the evolutionary pathways can beobtained from the fossil record over geologicaltime, but viruses do not leave the conventionalform of fossils; insects preserved inamber and seeds from archaeological sitesmay have some evidence of viruses. However,the increasing molecular information that isaccruing on plant (and other) viruses isrevealing what can be termed molecular fossilinformation. Even so, our knowledge of thepathways of virus evolution is quite fragmentary,although there is no doubt that viruseshave undergone, and continue to undergo,evolutionary change that is sometimes quiterapid.Comparative Plant Virology, Second Edition63Copyright # 2009, Elsevier Inc. All rights reserved.

64 4. PLANT VIRUS ORIGINS AND EVOLUTIONII. VIRUS EVOLUTIONIn discussing evolution of viruses, we mustrecognise that it is distinct from the evolutionof virus diseases. A new disease may be theconsequence of the “evolution” of the causalvirus, but it can also result from no change inthe virus. For instance, a new disease can resultfrom the movement of an “old” virus into anew situation. It is likely that the epidemics ofswollen shoot in cacao in West Africa, maizestreak in Africa, and tungro in rice in SoutheastAsia were due to the spread of the viruses fromasymptomatic natural hosts into either a newspecies in that area or a changed agronomicsituation.As just noted, viruses do not form conventionalfossils, but their genome sequences containmolecular fossils. Analysis of nucleic acidsand proteins of bacteria, fungi, and higherorganisms is providing much information onrelatedness and even on evolutionary pathways.The high turnover rate of viral genomescoupled with their potential for great variationwould seem to preclude this source of informationon viral evolution. However, most basicfunctions are controlled by highly conservedsequences that are essential, say, for enzyme catalyticsites; significant variation in these motifswould be lethal for the organism. Such motifsare conserved in viruses and form an importantsource of information for gaining an insight intoviral evolution and even origins.Because of this variation and high replicationrates, viruses are the most rapidly evolvinggenetic agents of all biological entities. However,this has to be viewed in the light of the currentenvironment in which they evolve. One canconsider that there are two “ages” of virusevolution: the long “prehuman age” when theenvironment for viruses changed slowly andthe recent “human age” when the environment(e.g., human populations and movement, growingmonocultures of crops and movement ofcrop species) is changing rapidly and whenviruses are probably evolving more rapidly.Most, if not all, of the available data on virusevolution relates to the latter “age.”A. Origins of VirusesFor many years, three hypotheses for the originsof viruses (Box 4.1) were accepted. However,the drawbacks to each of thesehypotheses and modern thinking are that thebasic development of the “virus world” mayhave predated the common ancestors of thethree major domains of Bacteria, Archaea, andEukarya. It is suggested that the virus worldBOX 4.1HYPOTHESES ON THEORIGIN OF VIRUSESThe virus-first hypothesis: Viruses are descendedfrom primitive precellular lifeforms. This hypothesis was based on theideas that the earliest prebiotic polymerswere RNA, which had enzymatic properties,such as that shown by ribozymes (seeBox 3.2). These prebiotic RNAs later parasitisedthe earliest cells.The reduction hypothesis. Virusesdeveloped from the normal constituents ofcells. The suggestion is that viruses arosefrom some cell constituent that escapedthe normal control mechanism and becameself-replicating entities. Examples of thenormal cell constituent include transposableelements and mRNAs.The escape hypothesis. Viruses arederived from degenerate cells that eventuallyparasitised normal cells. This has beensuggested as a possible origin for large vertebrateviruses, such as poxviruses.I. INTRODUCTION TO PLANT VIRUSES

II. VIRUS EVOLUTION65might even have played a significant role in theorigin and evolution of these domains.In considering the evolution from ancientforms of viruses to viruses of modern organisms,it must be remembered that in geologicaltimes five major great extinctions occurred,when up to 90 percent of living organisms diedout. If viruses did arise early in evolution, theymust have passed through these extinctionsand subsequently diversified after each one.Further support for an ancient origin ofpresent-day viruses is the geographic distributionof some groups. For instance, tymoviruseshave no known means of intercontinental spreadother than by humans. Different tymovirusesoccur in Europe, North and South America, Asia,Africa, and Australia, which were still one landmassat the end of the Jurassic period. Thus, thepresent-day tymoviruses are thought to havediverged from a common stock; the continentsdrifted apart between 138 and 80 millionyears ago.B. Virus VariationThe two main forms of variation of virusesare mutation (nucleotide changes) and recombination(nucleic acid sequence rearrangement).(See Figure 4.1.) Mutations (see Chapter 8)usually occur during nucleic acid replicationas a result of lack of proofreading of the newlysynthesised strand; they can also be inducedby external agents such as radiation. Singlebase changes that occur in a coding regionmay lead to replacement of one amino acid byanother in the protein product, the introductionof a new stop codon that results in early terminationof translation, and a shorter polypeptideor replacement of a codon that has eithergreater or lesser usage in the particular host.Deletion or addition of a single nucleotide in acoding region will give rise to a frameshift,with consequent downstream amino acidchanges. Such deletions or additions will usuallybe lethal unless compensated for by a secondchange (addition or deletion) that restoresthe original reading frame. Nucleotide changesin noncoding regions will vary in their effects,depending on the regulatory or recognitionfunctions of the sequence involved. There is nodoubt that a single base change giving rise to asingle amino acid substitution in the proteinconcerned is a frequent source of virus variabilityunder natural conditions in vivo. Mutationsthat involve more than one nucleotide changecan give rise to more major changes in theencoded protein or in the regulatory sequence.Recombination occurs in viruses with genomesconsisting of either DNA or RNA.(Recombination in both DNA and RNA virusesis discussed in detail in Chapter 8.) Recombinationalevents can lead to changes in the genomesof both DNA and RNA viruses and can lead todeletions, additions, or exchanges of sequencebetween two genomes. Deletions often give riseto defective or defective interfering nucleicacids, which are discussed in Chapter 3.A further form of recombination, termedpseudo-recombination or reassortment, occursin viruses that have their genomes dividedbetween several nucleic acid strands. In jointinfections between two strains of such viruses,new variants can arise by the encapsidation ofgenome segments from the two strains. Thisreassortment or shuffling of genomic segmentsis found in most divided genome virusesincluding bromoviruses, geminiviruses, andorthomyxoviruses (e.g., influenza virus).C. Types of EvolutionThe long-term evolution of viruses was justdiscussed. Now we will examine some of thefactors involved in virus evolution in moderneukaryotic cells.1. Microevolution and MacroevolutionThe variation of viruses gives the materialon which selection pressures can act, whichresults in their evolution. The different formsof variation have different importance in theI. INTRODUCTION TO PLANT VIRUSES

66 4. PLANT VIRUS ORIGINS AND EVOLUTIONWild type ORFM D D Q S R M L Q T L A G V N LatggacgatcaatccaggatgctgcagactctggccggggtgaacctgSilent (third base) mutationM D D Q S R M L Q T L A G V N LatggacgatcaatccaggatgctgcaaactctggccggggtgaacctgPoint mutation (missense)M D D Q S R M L K T L A G V N LatggacgatcaatccaggatgctgaagactctggccggggtgaacctgPoint mutation (nonsense)M D D Q S R M L stopatggacgatcaatccaggatgctgtagactctggccggggtgaacctgSingle base deletion leading to frame-shiftM D D Q S R M L R L W P G stopatggacgatcaatccaggatgctg-agactctggccggggtgaacctgAFIGURE 4.2 The apparent relative importance of somesources of genetic novelty that influence virus evolution.[From Gibbs and Keese (1995; in Molecular basis for virusevolution, A.J. Gibbs, C.H. Calisher, and F. Garcia-Arenal,Eds., pp. 76–90, reprinted with permission of CambridgeUniversity Press).]BFIGURE 4.1 Variation in a sequence. A. Mutation. Thetop lines show the wild type ORF with the nucleotidesequence in lowercase and the amino acid sequence in uppercase.Mutations are shown by red nucleotides and deletion by–. The mutations focus on the codon cag encoding glutamine(Q); the third base mutation does not change the amino acid; achange in the first base of the codon (c > a) changes the polarglutamine to lysine (K), which has a positive charge; changingthe c to a t introduces a stop codon; deleting the c leads to aframeshift and different downstream amino acids. B. A simplemodel of recombination between the donor virus (green)and the acceptor (red). The box indicates a region of homologybetween the donor and acceptor. During transcriptionof the donor the polymerase switch strands in the homologyregion, giving the recombinant.level or type of evolution (Figure 4.2). Thus,strains are differentiated mainly by mutationsand small insertions or deletions that areselected, changing the master sequence in thequasi-species cloud (for quasi-species, seeBox 1.3). This can be termed microevolution.Larger and more radical changes caused byrecombination and/or acquisition of newgenes, termed macroevolution, lead to the generationof new genera or families. This essentiallystarts a new quasi-species cloud that isselected upon in further microevolutionarydiversification.Because of the population structure of avirus isolate, microevolution can be a continuousprocess. However, extremely highmutation rates do not necessitate rapid evolution.Evidence exists that some virus genera(e.g., begomoviruses) appear to be diversifyingrapidly at present, whereas others (e.g., tobamoviruses)appear to be more stable. It is likelythat, as with the evolution of higher organisms,viruses go through a stage of relatively rapiddiversification when presented with a changingenvironment and then enter a relatively stablephase where they have adapted to the newenvironment(s). The selection pressures are discussedlater in this chapter.I. INTRODUCTION TO PLANT VIRUSES

II. VIRUS EVOLUTION67Macroevolution is a much rarer and a stepwiseprocess. The great majority of recombinationevents between viral sequences or thoseleading to the acquisition of new genes will belethal or deleterious to the virus. However, itis the very rare event that leads to the formationof a viral genome with new properties thatenable it to be more successful than its progenitors.Examples of this will be discussed later inthis chapter.Microevolution and macroevolution are notnecessarily independent systems. It is quitepossible that microevolutionary changes ofone virus could lead to the formation of “hotspots”for recombination with sequences inother viruses. Similarly, not all recombinationevents will lead to major changes. Recombinationbetween near homologous sequences willcreate new sequence combinations that haveonly minor differences to the parentalsequences. Thus, microevolution and macroevolutionare all a matter of degree.2. Sequence Divergence or ConvergenceSequence similarity between two genes doesnot necessarily indicate evolutionary relationship(homology). Without other evidence, itmay be impossible to establish whethersequence similarity between two genes is dueto a common evolutionary origin or to convergence.There is much debate as to whether theevolution of viral RNA-dependent RNA polymerasesis convergent or divergent.3. Modular EvolutionA process of modular evolution was initiallyproposed for DNA bacteriophage but is nowconsidered to also apply to all viruses. It is suggestedthat viruses have evolved by recombinationalrearrangements or reassortments ofinterchangeable elements or modules (Box 4.2).The basic module is for replicating thegenome, and it consists of the replicase and, inat least the case of reverse transcribing viruses,the coat protein. To this are added modules thatenable the basic replicon to fill its “niche”—thatis, to provide functions for the virus to infect itshost and to move between hosts.4. Sources of Viral GenesThe current understanding is that a significantpart of viral evolution involves theacquisition and exchange of functional modulesthat make up the virus genome. As to theoverall origins of genes or modules, two theoriespredominate: a common origin by a molecular“big bang” and continuous creation.Mechanisms by which new coding sequencescan be generated in a continuous creation scenarioinclude “overprinting,” in which an existingnucleotide sequence is translated de novoin a different reading frame or from noncodingORFs and gene duplication (for example, inClosteroviruses).a. Replicases. RNA replicases. The structureof core RNA replicases is discussed in Chapter8, where it is pointed out that they compriseseveral functional units: the RNA-dependentRNA polymerase (RdRp), a helicase, and amethyl-transferase; the RdRps and helicasesfall into several supergroups or superfamilies(see Boxes 8.2 and 8.3). There is disagreementas to whether a common origin exists for eachof these basic activities giving a monophyleticevolutionary pathway or whether their evolutionis polyphyletic from several origins. However,it has been suggested that the earliestform of nucleic acid was RNA, which wouldindicate RNA replication is very ancient.Notwithstanding the divergence of opinionabout the monophyletic or polyphyletic originsof the component parts of core RNA replicases,reasonable evidence exists that their arrangementin modern viral genomes has arisenby modular shuffling of these components(Figure 4.3).Reverse transcriptase. The replication of genomesof members of the several virus families,including the Retroviridae and Caulimoviridae, isI. INTRODUCTION TO PLANT VIRUSES

68 4. PLANT VIRUS ORIGINS AND EVOLUTIONBOX 4.2VIRUS MODULESModules are defined as interchangeable genetic elements, each of which carries out a particularbiological function; a basic modular structure for a plant virus is shown in the figure.These modules, or parts of them, can be interchanged, which enables their independent evolutionunder a wide variety of selective conditions. Such modular mobility can overcome the evolutionaryconstraints that would occur if all the modules had to coevolve within a single genomic unit.The following are the essentials of modular evolution:• The product of evolution is a favourable combination of modules selected to work optimallyindividually and together to fill a particular niche.• Joint infection of the host by two or more viruses is essential for the assembly of newcombinations of modules. The viruses do not necessarily have to give full systemic infection ofthe host; they just have to replicate in the same cell. This can lead to changes in virus host range ofthe recombinant virus when compared with the donor viruses.• Viruses in the same “interbreeding” population can differ widely in any characteristic becausethese are aspects of the function of individual modules.• Evolution acts primarily at the level of the individual module and not at the level of the intact virus.Selection of modules is for a good execution of function, retention of the appropriate regulatorysequences, and functional compatibility with most, if not all, other modules in that genome.by reverse transcriptase (RT), which converts anRNA template to DNA (see Chapter 8). Severalarguments point to the basic elements of RTbeing very ancient. The suggestion that RNApreceded DNA evolutionarily, coupled withsequence similarities between RdRps and RTs,has been taken to indicate that RT is among theearliest of enzymes. The amino acid homologiesbetween the RTs of all retro-elements suggestthat the enzyme has evolved only once.DNA replicases. The DNA > DNA replicationof various plant virus groups is described forgeminiviruses in Chapter 8. These viruses usehost DNA replicase functions, and becausethey replicate in differentiated cells, the cell cyclemust be modified. Thus, one of the viral genefunctions is to bind to retinoblastoma (Rb) proteins,which directs the cell into the S phase.It is likely that the Rb-binding protein(s) has beenacquired by the viruses from host sources.b. Proteinases. Three classes of virus-codedproteinases are found among plant viruses(Box 4.3). The functions of these were identifiedby comparison with amino acid sequencesfrom plants and animals, hence the termschymotrypsin-like and papain-like. This wouldsuggest a host origin. In most cases, the proteaseis a unique gene product, but it is sometimesassociated with a protein with anotherI. INTRODUCTION TO PLANT VIRUSES

II. VIRUS EVOLUTION69FIGURE 4.3 Conservation and variability inthe arrangement of “core” genes in proteinsinvolved in replication of (þ)-strand RNA viruses.POL 1, 2, 3, RNA-dependent RNA polymerase ofsubgroups 1, 2, and 3, respectively; HEL 1, 2, 3,(putative) helicase of superfamily 1, 2, and3, respectively; S-PRO, serine chymotrypsin-likeproteinase; C-PRO, cysteine chymotrypsin-likeproteinase; P-PRO, papain-like proteinase; MTR1, 2, methyl transferases of types 1 and 2, respectively;Boxes are not to scale. [From Koonin, E.V.,Dolja, V.V., and Morris, T.J. (1993). Evolutionand taxonomy of positive-strand RNA viruses:implications of comparative analysis of aminoacid sequences. Biochem. Mol. Biol. 28, 357–430,reprinted by permission of publisher (Taylor &Francis Ltd., http://www.tandf.co.uk/journals).]function—for example, the HC-Pro of potyviruses,in which it is associated with the insecttransmission helper factor (see Chapter 12) andalso the suppressor of host gene silencing (seeChapter 11). Whether this represents a multiplefunction for the original proteinase or theacquisition of various functional modules intothe proteinase molecule is unknown.I. INTRODUCTION TO PLANT VIRUSES

70 4. PLANT VIRUS ORIGINS AND EVOLUTIONBOX 4.3VIRUS CODED PROTEINASESFour classes of virus-coded proteinases are currentlyrecognised: serine, cysteine, aspartic, andmetallo-proteinases, named usually after theircatalytic site. Three of these four types of proteinaseare found among plant viruses.Proteinases encoded by some plant virusesVirus groupCaulimoviridaePotyviridaeComoviridaeTymovirusClosterovirusViralproteaseAspartateSerine, cysteine, serine-like aSerine-likeCysteineCysteine, aspartatea Has a cysteine at its active site but a structure similarto a serine proteinase.Serine proteinases are also termed 3C proteases(from their expression in picornaviruses)or chymotrypsin-like proteases. Most have a catalytictriad of amino acids (His, Asp, Ser), but insome, the Ser is replaced by Cys. The serine residueis unusually active and acts as a nucleophileduring catalysis by donating an electronto the carbonyl carbon of the peptide bond tobe hydrolyzed. An acyl serine is formed, and aproton is donated to the departing amyl groupby the active-site histidine residue. The acylenzyme is then hydrolyzed, the carboxylic acidproduct is released, and the active site is regenerated.Serine proteases cleave primarily atGln-Gly, Gln-Ser, Gln-Ala, and Gln-Asn.Cysteine proteinases, also known as papainlikeor thiol proteinases, have a catalytic dyadcomprising Cys and His residues in close proximitythat interact with each other. During proteolysis,the Cys sulfhydryl group acts as anucleophile to initiate attack on the carbonyl carbonof the peptide bond to be hydrolyzed. Anacyl enzyme is formed through the carbonyl carbonof the substrate and the sulfhydryl group ofthe active-site His. The carbonyl carbon is thenhydrolyzed from the thiol group of the proteinase,and the active-site residues are regenerated.Aspartic or acid proteinases are composed ofa catalytic dyad of two Asp residues. They mostlikely do not form covalent enzyme-substrateintermediates and are thought to operate by anacid-base catalysis.Viral proteinases are highly specific for theircognate substrates, a specificity that dependson the three-dimensional structures of both theproteinase and substrate. For instance, Cowpeamosaic virus proteinase does not recognise primarytranslation products of mRNAs from othercomoviruses.c. Coat Proteins The structure of the virusparticle is determined by the structure of thecoat protein subunit (see Chapter 5). Becauseof packing considerations, proteins making uprod-shaped particles have to be wedge shaped,whereas those making up spherical particleshave to be conical or have a three-dimensionalwedge shape. The shape of these proteins isdetermined by the secondary and tertiarystructures of their polypeptide chains. The coatproteins of most, if not all, viruses with isometricparticles have a basic structure comprisingeight antiparallel ß sheets, termed the ß barrelor jellyroll motif (see Chapter 5). This suggestsa single origin for the isometric virus coat protein.However, the coat protein of Sindbis virusI. INTRODUCTION TO PLANT VIRUSES

II. VIRUS EVOLUTION71has close relationships to chymotrypsin-likeproteinases, which would indicate a separateorigin.d. Cell-to-Cell Movement Proteins. Thecell-to-cell movement of many plant viruses ispotentiated by one or more movement proteins(MP; see Chapter 9). At least one type of theseproteins—the “30K superfamily,” which includesthe Tobacco mosaic virus (TMV) MP—appears tofunction by gating the plasmodesma open toenable the viral infection unit to pass to an uninfectedcell. A plant paralogue of TMV MP (a transcriptionfactor, the KNOTTED1 homeodomain)has been recognised. Thus, it is likely that thisfunctional module has been acquired by virusesfrom the plant genome. As there are several waysin which cell-to-cell movement is effected, it islikely that different genes have been acquiredby various viruses. For instance, the triple geneblock of potexviruses and hordeiviruses wouldappear to be a combination of three differentgenes rather than the separation of functionaldomains of one gene product.e. Suppressors of Gene Silencing. The plant(and other organisms) gene silencing defencesystem against “foreign” nucleic acids isdescribed in Chapter 11. Most, if not all, plantviruses have genes that suppress this defencesystem, and some viruses of vertebrates do aswell. The origin of these suppression genes isnot known, but there is a possibility that theyhave been acquired from the host genome.One time that plants and animals have to tolerate“foreign” nucleic acids is during the fertilisationprocess. Proteins involved in thefertilisation process might be the source of theviral silencing suppression gene(s).D. Selection Pressures for EvolutionVirus evolution, like that of other organisms,is a trade-off between costs and benefits and isdriven by the impact of selection pressures onthe inherent variation. Much of the molecularinformation on viruses comes from laboratoryexperiments, which, although they can providean insight into the mechanisms involved, cannotgive a picture of what happens in the “naturalsituation.” In this section, we examinesome of the factors that drive and limit the evolutionof plant viruses.1. Adaptation to NichesA virus is closely adapted to specific hostsand within those hosts to specific cellular sites.These can be thought of as ecological niches.There are two sorts of niches: macro-nichesand micro-niches. Macro-niches are the hostsand geographic situations in which the viruscan establish a successful infection. Thus,viruses that infect animals require differentgenes (modules) for spreading from cell to cellthan those that infect plants (see Chapter 9).An example of geographic situations is ricetungro disease, the occurrence of which isdetermined by the distribution of its insect vector,the rice green leafhopper, Nephotettix virescens(Figure 4.4). Micro-niches are the siteswithin the host cells in which the virus replicatesand moves from cell to cell and are usuallydefined by organelles, membranes, andthe cytoskeleton (see Chapters 8 and 9).2. Maximising the VariationEarlier in this chapter, we discussed thesources of variation, mutation, pseudo-recombination,and recombination that provide the materialupon which selection pressures can act. Therapid replication of viruses and the lack of proofreadingor repair mechanisms give the potentialfor much mutagenic variation in both RNA andDNA viruses. Thus, selection pressures operateon quasi-species, which comprise a dominantsequence and a multitude of minor variantsequences (see Box 1.4). Three factors contributeto the success of recombination, giving rise tonew viruses: mixed infections, high levels of viralreplication, and increased host range of the vector.I. INTRODUCTION TO PLANT VIRUSES

72 4. PLANT VIRUS ORIGINS AND EVOLUTIONFIGURE 4.4 Distribution of rice tungrodisease; the circle shows the distributionof the main leafhopper vector(Nephotettix virescens), and the diamondsshow the sites from which the two Ricetungro bacilliform virus strains wereobtained. [This article was published inEncyclopedia of Virology, vol. 2, R. Hull,Rice tungro disease, p. 483, CopyrightElsevier (2008).]3. Controlling the VariationViruses are absolutely dependent on theirhosts for their propagation and survival. Anyviral variant that damages the host significantlywould be selected against, and thus the evolutionarypressure is toward an equilibrium statebetween the virus and its host. There are veryfew examples of the impact of factors that controlvariation in the natural situation, as mostviruses that are studied cause overt diseases inorganisms of interest to man. In most cases theseare new interactions between the virus and thehost, and there has not been sufficient time forevolution to select an equilibrium. One exampleof the natural situation (Table 4.1) is found in astudy of viruses in 92 plants of the wild speciesPlantago lanceolata. Eight viruses were found, fiveof which caused no symptoms in P. lanceolata.In light of this high level of variation, someconstraints must be present that control thepreservation of the identity of a virus speciesor strain.TABLE 4.1Viruses of Plantago lanceolataPlantain virus X Potexvirus SymptomsPlantain virus A Rhabdovirus SymptomsRibgrass mosaic virus Tobamovirus Slight symptomsPlantago virus 4 Caulimovirus No symptomsPlantago virus 5 ? Necrovirus No symptomsPlantago virus 6 ? Tombusvirus No symptomsPlantago virus 7 ? Potyvirus No symptomsPlantago virus 8 ? Carlavirus No symptomsData from J. Hammond (1980). PhD Thesis, University ofEast Anglia, UK.I. INTRODUCTION TO PLANT VIRUSES

II. VIRUS EVOLUTION73a. Muller’s Ratchet. It is likely that mostmutations in a viral genome will be either neutralor deleterious. Mutations that lead to lossof a critical function would not be propagatedin a population unless they were complementedby other members of that population. However,mutations that caused only a decrease in“fitness” would be more likely to be retained.Theoretical considerations have led to the conceptof “Muller’s ratchet” (Box 4.4).BOX 4.4MULLER’S RATCHETAND VIRUSESTheoretical considerations have led to theconcept of Muller’s ratchet, which indicatesthat high mutation rates can have significantimpact on populations, especially ifthey are small in size. The concept ofMuller’s ratchet is that if the average mutationis deleterious, there will be a drift todecrease of population fitness leading to“mutational meltdown” (see Lynch et al.,1993). Muller’s ratchet is particularly applicableto small populations, and for manyviruses, transmission and infection form abottleneck in which the population is small.Back mutations at the specific site of adeleterious mutation or compensatorymutations are likely to occur at a lower ratethan forward mutations. In populations ofhigher organisms, this drift is limited bysex which recreates, through geneticexchange, genomes with fewer or no mutations.Obviously, this process does notoccur in the conventional sense in viruses,but it is likely that recombination or geneticreassortment within the quasi-species couldplay a part in controlling the effects ofMuller’s ratchet.b. Muller’s Ratchet and Plant Viruses. Evidenceexists that Muller’s ratchet has the followingeffects.• Effects on plant virus populations. Thetobamoviruses in herbarium samples andliving samples of Nicotiana glauca in NewSouth Wales, Australia, covering a periodfrom 1899 to 1993 were analyzed. Before 1950,many plants were infected with both TMVand Tobacco mild green mosaic virus (TMGMV),but after that date only TMGMV was found.In experimental joint infections of N. glauca,TMV accumulated to about 10 percent of thelevel of that in single inoculations; the level ofTMGMV was not affected. It was concludedthat TMV colonised N. glauca in New SouthWales earlier or faster than TMGMV, but injoint infections the latter virus caused adecrease of the TMV population below thethreshold at which deleterious mutationswere eliminated.• Methods to prevent it. Nematode transmissionof Tobacco rattle virus (TRV) serves as abottleneck to clear the virus population ofdefective interfering (DI) RNAs (for DI RNA,see Chapter 3). These DI RNAs, derivedfrom RNA2, have a modified coat proteingene and interfere with viral replication. Itwas suggested that TRV RNA2 and the DIRNA are encapsidated in cis by theirencoded coat proteins, which are,respectively, functional and nonfunctional intransmission, eliminating the DI RNA at thetransmission bottleneck.Variants that reduce the fitness of the viruspopulation have to be removed. From their success,it is obvious that plant viruses can overcomethe constraint of Muller’s ratchet, although themechanism is not known for most viruses.4. Role of Selection PressureFunctional viral genes will be retained onlyif they are needed for survival of the virus.For example, RNAs 3 and 4 of Beet necroticI. INTRODUCTION TO PLANT VIRUSES

74 4. PLANT VIRUS ORIGINS AND EVOLUTIONyellow vein virus (Profile 2) appear to sufferdeletions when not under the requirement tobe transmitted by the vector.5. Selection Pressure by Host PlantsConditions within a given host species orvariety will exert pressure on an infecting virusagainst rapid and drastic change. Viral genomesand gene products must interact inhighly specific ways with host macromoleculesduring virus replication and movement. Thesehost molecules, changing at a rate that is slowcompared with the potential for change in avirus, will act as a brake on virus evolution.The evolution of variants in populations ofTMV originating from an in vitro transcript of acDNA clone was studied by assessing the abilityto cause necrotic lesions (nl) in Nicotiana sylvestris.The proportion of nl variants in tobacco,tomato, Solanum nigrum, and Petunia hybridawas similar. However, in Physalis floridiana therewas strong selection pressure against nl variants,which were reduced to almost undetectablelevels. Nevertheless, in all hosts tested large andapparently random changes in the proportionof nl variants in individual plants were observed,showing that viral populations can evolve rapidlyon a timescale of days. Perhaps these variationsin a strain population occur during theearly stages of infection in new host species.Over a period of many transfers in the samespecies, the populations may stabilise. This ideais supported by the fact that TMV obtained fromW.M. Stanley in the 1930s and independentlysubcultured many times over decades in tobaccovarieties in the United States and Germany wasshown in both countries to have the same coatprotein amino acid sequence.E. Timeline for EvolutionThe direction and timing of the evolution ofmost organisms is derived from the stratigraphicalposition of the conventional fossil ancestors.However, it is not possible to use this approachfor the molecular fossils of viruses. The onlypotential sources of information are the molecularclocks of change of sequence, but theseusually do not give the direction of change.1. Nonconstant Rates of EvolutionThe molecular clock hypothesis, whichassumes a constant rate of change in sequenceover evolutionary time, has often been used inthe interpretation of phylogenetic “trees.” However,there is considerable uncertainty as towhether molecular clocks can be applied to theevolution of viral genomes. Furthermore, differentproteins, or parts of a protein coded by a viralgenome, may evolve at different rates. In addition,some noncoding sequences in the genome maybe highly conserved, particularly those recognitionsequences essential for genome replication.2. Estimated Rates of EvolutionThe rate of point mutation for RNA viruseshas been estimated to be approximately 10 -3 –10 -4 per nucleotide per round of replication withsome variation between different viruses contrastingwith estimates of about 10 -11 for DNApolymerases (see Figure 4.5). However, it is verydifficult to relate mutation rates to the actualrates of change in viruses that might be occurringin the field at present or over past evolutionarytime because of (1) variation in rates ofchange in different parts of a viral genome, (2)uneven rates of change over a time period, and(3) lack of precise historical information.FIGURE 4.5 Error rates of transcription within andbetween RNA and DNA. [From Hull (2002).]I. INTRODUCTION TO PLANT VIRUSES

III. EVIDENCE FOR VIRUSEVOLUTIONAs just noted, evidence for virus evolutionmust come from the study of present-dayviruses in present-day hosts. It is the comparisonsof genome organisations and the detailsof sequences that are giving evidence for theevolutionary pathways of many viruses. Themajor factors involved can be shown by threeexamples of plant viruses but can also beapplied to many other virus groups.A. GeminivirusesThe DNA > DNA replication of geminivirusesshould result in a less mutagenic variationthan RNA > RNA replication (seeFigure 4.5). However, there is evidence for significantgenomic variation in at least two geminivirusgenera: the begomoviruses and themastreviruses. Sequence analysis of three Maizestreak virus (MSV) isolates obtained by serialpassage of a wild-type isolate through a semiresistantcultivar of maize indicated that theoriginal infection had a quasi-species structurewith mutations distributed throughout thegenome. Mutation frequencies were estimatedto be between 3.8 10 -4 and 10.5 10 -4 , levelssimilar to those found with RNA > RNA replication.The mutagenic variation of geminivirusesis unexpected in view of theproofreading normally associated with DNA >DNA replication and may reflect a lack ofpostreplicative repair. Geminivirus DNA doesnot appear to be methylated, as its replicationis inhibited by methylation. Thus, althoughthese viruses use the host system for their replication,the normal host mechanisms for mismatchrepair probably do not function. Thereis increasing evidence for natural recombinationwithin geminiviruses, on occasions leadingto new diseases (Box 4.5).III. EVIDENCE FOR VIRUS EVOLUTIONB. ClosterovirusesThe genome organisations of members of thefamily Closteroviridae are described in Profile 5 inthe Appendix. These viruses have the largest(þ)-strand ssRNA genomes among plant viruses.It has been suggested that the closterovirusgenome arose from a common ancestor, withrearrangement of that genome and acquisitionof other modules by recombination (Box 4.6).Recombination is an essential part of thisproposed evolutionary pathway. Analyses ofmultiple species of Citrus tristeza virus (CTV)defective RNAs show that they appear to havearisen by recombination of a subgenomicRNA (sgRNA) with distant parts from the 5 0end of the CTV genome. It is suggested thatclosteroviruses are able to use the sgRNAand/or their promoter signals for modularexchange and rearrangement of their genomes.In spite of the potential for variation, a studyshowed that mild CTV isolates maintained indifferent citrus hosts, from several geographicallocations (Spain, Taiwan, Colombia, Florida,and California) and isolated at different timeswere remarkably similar indicating a high degreeof evolutionary stasis in some CTV populations.C. Luteoviruses75The organisation of the (þ)-sense ssRNA genomesof members of the family Luteoviridae isshown in Profile 8 in the Appendix. The genomeorganisation of the viruses in the families Luteoviridaeand Tombusviridae (Profile 16) and the genusSobemovirus comprises two basic modules: the 5 0replicase proteins and the 3 0 proteins, which includethe virion coat protein. Phylogenetic analyzesof the RdRp and the coat protein suggest that therehave been gene transfer events between the modulesof members of this supergroup and suggestedtwo main clusters termed the enamo and carmoclusters. Likely gene transfers were recognisedboth between and within these main clusters.I. INTRODUCTION TO PLANT VIRUSES

76 4. PLANT VIRUS ORIGINS AND EVOLUTIONBOX 4.5EXAMPLE OF GEMINIVIRUS EVOLUTIONTomato yellow leaf curl disease spread and diversified rapidly through the Mediterranean basinduring the twentieth century, as shown in the figure, which gives the first dates that the diseasewas recognised in various countries.Three viruses are currently known to cause the disease: Tomato yellow leaf curl virus (TYLCV; seeProfile 7), Tomato yellow leaf curl Sardinia virus (TYLCSV), and Tomato yellow leaf curl Sudan virus(TYLCSDV); these viruses are all begomoviruses but differ in sequence.Various recombinants have now been detected between these viruses. TYLCV has recombined withTYLCSV to give a new virus, Tomato yellow leaf curl Malaga virus, which has subsequently recombinedwith TYLCSDV and also formed further recombinants with its parents. Both TYLCV and TYLCSV havealso recombined with TYLCSDV. Thus, there is rapid variation in this group of viruses.The current classification of the Luteoviridaerecognises three genera: the Luteovirus, whichfalls in the carmo cluster, and the Polerovirusand Enamovirus, which are placed in the enamocluster. The proposed evolutionary pathway isshown in Box 4.7.There is further considerable evidence forrecombination within the Luteoviridae. One candistinguish three types of events: recombinationwithin a gene, recombination of large parts ofthe genome within a genus, and recombinationbetween large parts of the genome betweengenera.Thus, recombination appears to be rampantboth within the Luteoviridae and between membersof this family and those of some othergroups of small ssRNA viruses. There is alsoevidence that Potato leaf roll virus (PLRV) canrecombine with host sequences; the 5 0 -terminal119 nucleotides of some RNAs of a Scottish isolateof PLRV are very similar to an exon oftobacco chloroplast DNA.I. INTRODUCTION TO PLANT VIRUSES

BOX 4.6CLOSTEROVIRUS EVOLUTIONThe progenitor of the alphavirus supergroup has been proposed to comprise a complex of genesencoding methyl transferase (MTR), helicase (HEL), papain-like protease (P-PRO), polymerase(POL) (for description of these modules in the replicase see Chapter 8, Section IV, B), and capsid protein.The proposed evolutionary pathyway is shown in the Figure:Fig. A tentative scenario for the evolution of closteroviruses. CPe designates an elongated particle capsidprotein; CPH, capsid protein homolog; HEL1, RNA helicase of superfamily 1; HSP70r, HSP70-related protein;MTR 1, type 1 methyl transferase; POL 3, polymerase of supergroup (see Box 8.2) 3; P-PRO, papain-like protease.[From Dolja et al. (1994; Annu. Rev. Phytopathol. 32, pp. 261–285); Reprinted, with permission, from the Annual Review ofPhytopathology Volume 32 #1994 by Annual Reviews www.annualreviews.org.](continued)I. INTRODUCTION TO PLANT VIRUSES

78 4. PLANT VIRUS ORIGINS AND EVOLUTIONBOX 4.6(continued)This evolutionary pathway comprises the following steps:1. Deletion of P-PRO from the core of the replication genes.2. Substitution of the postulated alphavirus type capsid protein by a capsid protein capable offorming elongated virions.3. Invention of the frameshift mechanism of POL expression (see Chapter 7, Section V, B, 12).4. Acquisition of the heat-shock protein HSP70 from the cellular genome.5. Duplication of the capsid protein and the functional switch for one of the tandem copies tofacilitate aphid transmission.6. Insertion of long coding sequences between the MET and HEL cistrons.7. Secondary acquisition of the leader P-PRO, perhaps from a potyvirus or a related virus.8. Additional diversification and acquisition of the 3’-terminal genes.9. Split of the genome into two components giving the crinivirus (a genus in the Closteroviridae witha divided genome) genome organisation.It is suggested that steps 1 and 2 occured early in evolution and gave a common ancestor of thewhole tobamovirus cluster. The order of the other events is rather arbitrary.BOX 4.7PROPOSED EVOLUTION OF THE LUTEOVIRIDAEThe most likely model for the origin of the genomes of the luteovirus and polerovirus genomes (Profile8) (Figure) suggests that recombination arose by strand switching at subgenomic RNA start sitesduring RNA replication in cells jointly infected by the two parental viruses. For the derivation of theluteovirus genome, the sgRNA start site on diantho-like viruses has homology to that of poleroviruses.Recombination at this site would create a hybrid virus with dianthovirus polymerase andpolerovirus coat protein and neighbouring genes. A recombination event at the sgRNA start sitedownstream of ORF5 would give the complete luteovirus genome organisation. A single recombinationbetween the 5 0 region of a sobemovirus and the 3 0 part of a luteovirus followed by prematuretermination would give the polerovirus genome organisation.(continued)I. INTRODUCTION TO PLANT VIRUSES

III. EVIDENCE FOR VIRUS EVOLUTION79BOX 4.7(continued)Fig. Model for the origin of genomes of luteoviruses and poleroviruses. Solid black lines represent viral genomicRNA; dashed lines indicate sub-genomic RNAs; Boxes indicate the genes; blue shading, genes with sequence similarityto umbra-diantho- and carmo-viruses; green. Sequence similarity fo sobemoviruses. Grey boxes representputative origins of replication and sub-genomic RNA promoters. POL, RNA-dependent RNA polymerase; PRO?,putative protease; CP, coat protein; MP?, putative movement protein; AT, read-through domain of coat proteingene. Pink line shows the proposed path of replicase as it switched strands during copying viral RNA in a mixedinfection. From Miller et al. (1997; Plant Dis. 81, 700-710).I. INTRODUCTION TO PLANT VIRUSES

80 4. PLANT VIRUS ORIGINS AND EVOLUTIONFIGURE 4.6Interactions between a virus, its host, and its vector.IV. COEVOLUTION OF VIRUSESWITH THEIR HOSTS ANDVECTORSFahrenholtz’s rule postulates that parasitesand their hosts speciate in synchrony (Eichler,1948). Thus, there is a prediction that phylogenetictrees of parasites and their hosts should be topologicallyidentical. In view of the known wide hostranges of many present-day plant viruses, it is notto be expected that Fahrenholtz’s rule will be followedclosely for viruses and their hosts. Nevertheless,it is now widely accepted that viruseshave had a long evolutionary history and havecoevolved with their host organisms. The conceptof coevolution does not imply that viruses haveevolved to a state of higher complexity followingtheir plant hosts in this respect. On the contrary,the evidence available at present shows that thelargest and most complex virus infecting photosyntheticorganisms is found in the simplest host,a Chlorella-like green alga.When considering coevolution, one mustconsider the interactions between the virus andthe host, between the virus and its vector, andbetween the host and the vector (Figure 4.6).These interactions can be complex and are alsoaffected by external factors such as climate andcompetition from other organisms. However,in the natural situation it is essential for thevirus to be able to replicate efficiently but notaffect its host significantly and also to be efficientlytransmitted to other hosts and establishinfections therein.I. INTRODUCTION TO PLANT VIRUSES

V. VIRUSES OF OTHER KINGDOMSThere is extensive molecular evidence thathas been used to suggest evolutionary pathwaysfor many groups of viruses. The examples justgiven are from plant viruses, but the same principlesapply to viruses from other kingdoms.VI. SUMMARY• Viruses do not produce conventional fossils,but viral sequences contain “molecularfossils.”• Viruses probably originated early in theevolution of living organisms.• Virus evolution is through variation andselection.• Virus variation is due to mutations that cancreate new virus strains and some speciesand recombination that gives rise to newvirus genera and families.• Viral genomes are composed of functionalmodules that can be interchanged byrecombination.• There are various selection pressures onvirus evolution and, as viruses require livingVI. SUMMARYhosts for their replication, the ultimate stateis an equilibrium state between the virus andits host.• Viruses have coevolved with their hosts andtheir vectors.ReferencesEichler, W. (1948). Some rules on ectoparasitism. Ann. Mag.Nat. Hist. Ser. 12, 588–598.Lynch, M., Bürger, R., Butcher, D., and Gabriel, W. (1993).The mutational meltdown in asexual populations.J. Hered. 84, 339–344.Further Reading81Forterre, P. (2008). Origin of viruses. Encyclopedia of Virology,Vol. 3, 472–478.Hull, R (2002). Matthew’s plant virology. Academic Press, SanDiego.Karasev, A.V. (2000). Genetic diversity and evolution ofclosteroviruses. Annu. Rev. Phytopathol. 38, 293–324.Lovisolo, O., Hull, R., and Rosler, O. (2003). Coevolution ofviruses with hosts and vectors and possible paleontology.Adv. Virus Res. 62, 326–379.Roossinck, M.J. (Ed.) (2008). Plant virus evolution. Springer,Berlin.Villarreal, L.P. (2008). Evolution of viruses. Encyclopedia ofVirology, Vol. 2, 174–184.I. INTRODUCTION TO PLANT VIRUSES

This page intentionally left blank

S E C T I O N IIWHAT IS A VIRUS MADE OF?

This page intentionally left blank

C H A P T E R5Architecture and Assemblyof Virus ParticlesThe genomic nucleic acids of plant viruses are protected by virus-coded proteins which formdefined virus particles. The particles of most viruses have a simple structure, but others aremore complex.O U T L I N EI. Introduction 85II. Methods 86III. Architecture of Rod-Shaped Viruses 88IV. Architecture of Isometric Viruses 94V. Small Icosahedral Viruses 97VI. More Complex Isometric Viruses 101VII. Enveloped Viruses 101VIII. Assembly of Icosahedral Viruses 102IX. General Considerations 103X. Viruses of Other Kingdoms 104XI. Summary 104I. INTRODUCTIONKnowledge of the detailed structure of virusparticles is an essential prerequisite to ourunderstanding of many aspects of virology—for example, how viruses survive outside thecell, how they infect and replicate within thecell, and how they are related to one another.Knowledge of virus architecture has increasedgreatly in recent years, due both to moredetailed chemical information and to the applicationof more refined electron microscopic,optical diffraction, and X-ray crystallographicprocedures.Many different terms are used in virus structure(see Box 5.1). Figures 1.3 and 1.5 indicateComparative Plant Virology, Second Edition85Copyright # 2009, Elsevier Inc. All rights reserved.

86 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLESBOX 5.1TERMS USED IN VIRUS STRUCTURECapsid: The closed shell or tube of a virusCapsomere: The clusters of subunits on thecapsid as seen in electron micrographs; alsotermed morphological subunitEncapsidation (or encapsulation): The processof enclosing the viral genomic nucleic acidin virus-encoded protein usually to form a virusparticleMulticomponent virus: Infectious virusgenome divided between several nucleic acidsegments that are separately encapsidatedNegative stain: The virus particle is embeddedin an electron dense material such as phosphotungicacid or uranyl acetate, which formsa dense background against which the virus particleappears translucent in the electron microscope.This method is capable of providinginformation about structural details often finerthan those visible in thin sections, replicas, orshadowed specimens and has the advantage ofspeed and simplicityNucleocapsid: The inner nucleoprotein coreof membrane-bound virusesNucleoprotein: A complex between the viralgenomic nucleic acid and virus-encoded protein,which may or may not have a definedstructureParticle (virus particle): The virus genomeenclosed in a capsid and for some viruses alsoa lipid membraneProtein subunit: Individual virus-encodedprotein molecule that makes up the capsomereor nucleoprotein; also termed structural subunitVirion: The mature virus that may be membranebound; term interchangeable with virusparticlethe range of sizes and shapes found amongplant viruses. The three basic types of virusparticle are rod-shaped, isometric (spherical),and complex.II. METHODSA. Chemical and Biochemical StudiesKnowledge of the size and nature of the viralnucleic acid and of the proteins and othercomponents that occur in a virus particle isessential to an understanding of its architecture.Chemical and enzymatic studies may givevarious kinds of information about virus structure.For example, the fact that carboxypeptidaseA removed the terminal threonine fromintact Tobacco mosaic virus (TMV) indicated thatthe C-terminus of the polypeptide was exposedat the surface of the virus. For viruses withmore complex structures, partial degradationby chemical or physical means—for example,removal of the outer envelope—can be used toestablish where particular proteins are locatedwithin the particle. Studies on the stability of avirus under various pH and ionic or other conditionsmay give clues as to the structure andthe nature of the bonds holding the structuretogether.B. Methods for Studying Size and FineStructure of Viruses1. Hydrodynamic MeasurementsThe classic procedure for hydrodynamicmeasurements is to use the Svedberg equation(see Schachman, 1959). Laser light scatteringII. WHAT IS A VIRUS MADE OF?

II. METHODS87has been used to determine the radii of severalapproximately spherical viruses with a highdegree of precision. Virus particles bind water,and this method gives an estimate of thehydrated diameter, whereas the Svedbergequation and electron microscopy give a diameterfor the dehydrated virus. Comparison ofthe two approaches can give a measure of thehydration of the virus particle.2. Electron MicroscopyMeasurements made on electron micrographsof isolated virus particles, or thin sectionsof infected cells, offer very convenientestimates of the size of viruses. For some of thelarge viruses and for the rod-shaped viruses,such measurements may be the best available,but they are subject to significant errors, suchas magnification errors or flattening of theparticles.With helical rod-shaped particles, the centralhollow canal is frequently revealed bynegative staining. With the small sphericalviruses, which often have associated withthem empty protein shells, it was assumed bysome workers that stained particles showing adense inner region represented empty shellsin the preparation. However, staining conditionsmay lead to loss of RNA from a proportionof the intact virus particles, allowing stainto penetrate, while stain may not enter someempty shells. Stains differ in the extent towhich they destroy or alter a virus structure,and the extent of such changes depends closelyon the conditions used. The particles of mostviruses that are stabilised by protein:RNAbonds are disrupted by some negative stainsat pHs above 7.0.Some aspects of the structure of the envelopedviruses, particularly bilayer membranes, canbe studied using thin sections of infected cellsor of a pellet containing the virus. Cryo-electronmicroscopy, which involves the extremely rapidfreezing of samples in an aqueous medium,allows the imaging of symmetrical particles inthe absence of stain and under conditions thatpreserve their symmetry. This technique hasbeen widely used for isometric viruses. Forthese particles, the images have to be reconstructedto give a three-dimensional object froma two-dimensional image using a range ofapproaches, leading to information that complementsthat from X-ray crystallography and alsoallows the detailed analysis of viruses that arenot amenable to crystal formation.3. X-Ray CrystallographyFor viruses that can be obtained as stablecrystalline preparations, X-ray crystallographycan give accurate and unambiguous estimatesof the radius and structure of the particles inthe crystalline state. The technique is limitedto viruses that are stable or can be made stablein the salt solutions necessary to producecrystals.4. Neutron Small-Angle ScatteringNeutron scattering by virus solutions is amethod by which low-resolution informationcan be obtained about the structure of smallisometric viruses and in particular about theradial dimensions of the RNA or DNA andthe protein shell. The effects of different conditionsin solution on these virus dimensions canbe determined readily. The method takes advantageof the fact that H 2 O-D 2 Omixturescanbeused that match either the RNA or the proteinin scattering power. Analysis of the neutron diffractionat small angles gives a set of data fromwhich models can be built.5. Atomic Force MicroscopyTMV particles, when dried on glass substrates,assemble into characteristic patternsthat can be studied using atomic force microscopy.In highly orientated regions, the particlelength was measured as 301 nm and the widthas 14.7 nm; the latter measurement shows intercalationof packed particles. The particles arenot flattened and their depth was 16.8–18.6 nm.II. WHAT IS A VIRUS MADE OF?

88 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLES6. Mass SpectrometryAs well as measuring the mass of viral proteinsand particles, mass spectrometry can beused to identify posttranslational modificationsof viral protein such as myristoylation, phosphorylation,and disulfide bridging. When thistechnique is used in conjunction with X-ray crystallography,the mobility of the capsid can bestudied. Similarly, nuclear-magnetic-resonancespectroscopy can detect mobile elements on thesurface of virus particles.7. Serological MethodsThe reaction of specific antibodies withintact viruses or dissociated viral coat proteinshas been used to obtain information that isrelevant to virus structure. For instance, theterminal location of the minor coat protein onthe flexuous rod-shaped particles of closterovirusesand criniviruses was recognised usingpolyclonal antibodies (Figure 5.1).8. Stabilising BondsThe primary structures of viral coat proteinsand nucleic acids depend on covalent bonds.In the final structure of the simple geometricviruses, these two major components are heldtogether in a precise manner by a variety ofnoncovalent bonds. Two kinds of interactionare involved: protein:protein and protein:nucleic acid. In addition, small molecules suchas divalent metal ions (Ca 2þ in particular)may have a marked effect on the stability ofsome viruses. Knowledge of these interactionsis important for understanding the stability ofthe virus in various environments, how itmight be assembled during virus synthesis,and how the nucleic acid might be releasedfollowing infection of a cell. The stabilisinginteractions are hydrophobic bonds, hydrogenbonds, salt linkages, and various other longandshort-range interactions.III. ARCHITECTURE OFROD-SHAPED VIRUSESA. IntroductionCrick and Watson (1956) put forward ahypothesis concerning the structure of smallviruses, which has since been generally confirmed.Using the then recent knowledge thatthe viral RNA was enclosed in a coat of proteinand that the naked RNA was infectious, theyassumed that the basic structural requirementFIGURE 5.1 “Rattlesnake” tails onCitrus tristeza virus particles. A. Particlegold-labelled with p27 (minor coatprotein) antibodies; B. Particles goldlabelledwith coat protein (p25) antibodies.Bars ¼ 100 nm. [From Febreset al. (1996; Phytopathology 86, 1331–1335).]II. WHAT IS A VIRUS MADE OF?

III. ARCHITECTURE OF ROD-SHAPED VIRUSES89for a small virus was the provision of a shell ofprotein to protect its ribonucleic acid. They consideredthat the protein coat might be mademost efficiently by the virus that controlledthe production in the cell of a large number ofidentical small protein molecules, rather thanin one or a few very large ones.They pointed out that if the same bondingarrangement is to be used repeatedly in theparticle, the small protein molecules wouldaggregate around the RNA in a regular manner.There are only a limited number of waysin which the subunits can be arranged. Thestructures of all the geometric viruses are basedon the principles that govern either rod-shapedor spherical particles.In rod-shaped viruses, the protein subunitsare arranged in a helical manner. There is notheoretical restriction on the number of proteinsubunits that can pack into a helical array inrod-shaped viruses.B. Structure of TMV1. General FeaturesThe particle of TMV is a rigid helical rod,300 nm long and 18 nm in diameter(Figure 5.2A). The composition of the particleis approximately 95 percent protein and 5 percentRNA. It is an extremely stable structure,having been reported to retain infectivity inCentral canalA B CFIGURE 5.2 Structure of TMV. A Electron micrograph of negatively stained TMV particles; bar ¼ 100 nm. [From Hull(2002).] B. Drawing showing relationship of the RNA and protein subunits; note that the RNA shown free of protein couldnot maintain that configuration in the absence of protein; C. Photograph of a model of TMV with the major dimensionsindicated. B. and C. Reprinted from Adv. Virus Res., 7, A. Klug and D.L.D Caspar, The structures of small viruses,pp. 225–325, Copyright (1961), with permission from Elsevier.II. WHAT IS A VIRUS MADE OF?

90 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLESnonsterile extracts at room temperature for atleast 50 years; however, the stability of nakedTMV RNA is no greater than that of any otherssRNA. Thus, stability of the virus with respectto infectivity is a consequence of the interactionsbetween neighbouring protein subunits andbetween the protein and the RNA.X-ray diffraction analyses have given usa detailed picture of the arrangement of theprotein subunits and the RNA in the virusrod. The particle comprises approximately2,130 subunits that are closely packed in aright-handed helical array. The pitch of thehelix is 2.3 nm, and the RNA chain is compactlycoiled in a helix following that of theprotein subunits (Figure 5.2B, C). There arethree nucleotides of RNA associated with eachprotein subunit, and there are 49 nucleotidesand 16 1 /3 protein subunits per turn. The phosphatesof the RNA are at about 4 nm fromthe rod axis. In a proportion of negativelystained particles, one end of the rod (the 5 0end) can be seen as concave, and the otherend (the 3 0 end) is convex. A central canal witha radius of about 2 nm becomes filled withstain in negatively stained preparations of thevirus (Figure 5.2A).2. Virus StructureBecause intact TMV does not form threedimensionalregular crystalline arrays in solution,fibre diffraction methods were used tosolve the structure to a resolution of 2.9 Å andto produce a model for the virus (Figures 5.2Cand 5.3).The following are the important features ofthis model:• The outer surface. The N- and C-termini of thecoat protein are at the virus surface(Figure 5.3B). However, the very C- terminalresidues 155–158 were not located on thedensity map and are therefore assumed to besomewhat disordered.• The inner surface. The presence of the RNAstabilises the inner part of the proteinsubunit in the virus so that its positioncan be established. The highest peak inthe radial density distribution is at about2.3 nm.• The RNA binding site (Figure 5.3C). Thebinding site is in two parts, being formedby the top of one subunit and the bottomof the next. The three bases associatedwith each protein subunit form a “claw”that grips the left radial helix of the topsubunit.• Electrostatic interactions involved in assemblyand disassembly. Pairs of carboxyl groupswith anomalous pK values (near pH 7.0)are present in TMV. These groups mayplay a critical role in assembly anddisassembly of the virus. There are threesites where negative charges from differentmolecules are juxtaposed in subunitinterfaces. These create an electrostaticpotential that could be used to drivedisassembly.A low-radius carboxyl-carboxylate pairappears to bind calciumA phosphate-carboxylate pair that alsoappears to bind calciumA high-radius carboxyl-carboxylate pair inthe axial interface. This cannot bindcalcium but can bind a proton and thustitrate with an anomalous pK.• Water structure. Water molecules aredistributed throughout the surface of theprotein subunit, both on the inner and outersurfaces of the virus and in the subunitinterfaces.• Specificity of TMV protein for RNA. TMVprotein does not assemble with DNA even ifthe origin of assembly sequence (seefollowing) is included; thus, the specificityfor RNA must involve interactions made bythe ribose hydroxyl groups because all threebase-binding sites could easily accommodatethymine.II. WHAT IS A VIRUS MADE OF?

III. ARCHITECTURE OF ROD-SHAPED VIRUSES91FIGURE 5.3 Interactions between TMVsubunits and RNA. A. From Figure 5.2B withthe square indicating the region shown inpanel B; the subunits marked * are the twosubunits in panel B. B. Secondary structureof coat protein subunits. The four a-helicesare indicated (LS, left-slewed; LR, left radial;RS, right-slewed; RR, right radial); N is thesubstantially obscured N-terminus; C is theC-terminus. C. Enlargement of the interactionof RNA with protein. The backbone structureof the protein subunits and the three RNAnucleotides (labelled 1, 2 and 3), representedas GAA, are illustrated. [This article waspublished in J. Molec. Biol., 208, K.Namba,R. Pattanayek, and G. Stubbs, Visualisationof protein-nucleic acid interactions in a virus.Refined structure of intact tobacco mosaicvirus at 2.9 Å resolution by X-ray fiber diffraction,pp. 307–325, Copyright Elsevier (1989).]ACENTRAL CANALBα-HelicesN-terminusC-terminusOUTERFACEOFVIRUSCII. WHAT IS A VIRUS MADE OF?

92 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLESC. Assembly of TMVTMV particles can be disassembled intocoat protein subunits and RNA by dialysingvirus preparations against alkaline buffers andseparating the protein from the nucleic acidby ammonium sulphate precipitation. This hasled to much study of the factors involved inin vitro virus disassembly and reassembly.1. Properties of the Coat ProteinThe coat protein comprises 158 amino acids,giving it a molecular weight of 17–18 kDa.Fibre diffraction studies have determined thestructure to 2.9 Å resolution (Figure 5.3B). Theprotein has a high proportion of secondarystructure with 50 percent of the residues formingfour a-helices and 10 percent of the residuesin ß-structures, in addition to numerousreverse turns. The four closely parallel or antiparallela-helices make up the core of the subunit,and the distal ends of the four helicesare connected transversely by a narrow andtwisted strip of ß-sheet. The central part of thesubunit distal to the ß-sheet is a cluster of aromaticresidues forming a hydrophobic patch.2. Assembly of TMV Coat ProteinThe protein monomer can aggregate in solutionin various ways depending on pH, ionicstrength, and temperature. The major formsare summarised in Figure 5.4A. Experimentswith monoclonal antibodies show that bothends of stacked disks expose the same proteinsubunit surface. Thus, each two-layer unit inthe stack must be bipolar (i.e., facing in oppositedirections). The existence of these variousaggregates has been important both for ourunderstanding of how the virus is assembledand also for the X-ray analysis that has led toa detailed understanding of the virus structure.The helical protein rods that are produced atlow pH are of two kinds: one with 16 1 /3 subunitsper turn of the helix, as in the virus,and one with 17 1 /3. In both of these forms theprotein subunit structure is very similar to thatin the virus.3. Assembly of the TMV Roda. Assembly in Vitro. The classic experimentsof Fraenkel-Conrat and Williams (1955)showed that it was possible to reassemble intactvirus particles from TMV coat protein and TMVRNA. Since then, a detailed understanding hasbeen gained on the assembly of the virus. Thethree-dimensional structure of the coat proteinis known in atomic detail, and the completenucleotide sequence of several strains of thevirus and related viruses is known. The systemtherefore provides a useful model for studyinginteractions during the formation of a macromolecularassembly from protein and RNA.Four aspects of rod assembly must be considered:the site on the RNA where rod formationbegins; the initial nucleating event that beginsrod formation; rod extension in the 5 0 direction;and rod extension in the 3 0 direction. The currentunderstanding shows the following features.TheAssemblyOriginintheRNA.Coatproteindoes not begin association with the viral RNAin a random manner, and the origin of assemblyis between 900 and 1,300 nucleotides from the3 0 end. Nucleotide sequences near the initiatingsite can form quite extensive regions of internalbase-pairing, as is illustrated in Figure 5.4B. Loop1 in Figure 5.4B, with the sequence AAGAA-GUCG, combines first with a double disk of coatprotein. The base sequence in the stem of loop 1is not critical, but its overall stability is importantas is the small loop at the base of the stem,The Initial Nucleating Event. TMV rod assemblyis initiated by the interaction between adouble disk of coat protein (see Figure 5.4C,steps 1 and 2) with the long 5 0 tail and the short3 0 tail of the RNA looping back down the axialhole. The transition between the double diskand helix is mainly controlled by a switchingmechanism involving the abnormally titratingcarboxyl groups. At low pH, the protein canform a helix on its own because the carboxylII. WHAT IS A VIRUS MADE OF?

III. ARCHITECTURE OF ROD-SHAPED VIRUSES93ABCFIGURE 5.4 Assembly of TMV. A. Effects of pH and ionic strength on the formation of some aggregation states of TMVcoat protein. The forms mentioned in the text that are involved in the assembly of the virus are identified. [Reprinted by permissionfrom Macmillan Publishers Ltd: Nature (London) New Biol. 229, 37–42 from Durham et al., copyright 1971.] B. Proposedsecondary structure of TMV origin of assembly extending from bases 5290 to 5527 of the viral sequence. [Reprinted bypermission from Macmillan Publishers Ltd., from Zimmern (1983; EMBO J. 2, 1901–1907).] C. Model for the assembly ofTMV; the stages are described in the text. [Courtesy P.J.G. Butler.]II. WHAT IS A VIRUS MADE OF?

94 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLESgroups become protonated. It is thought thatthe inner part of the two-layered disk acts as apair of jaws to bind specifically to the originof-assemblyloop in the RNA and, in the process,converts each disk of the double disk intoa “lock washer” and forming a protohelix(Figure 5.4C, step 3).Rod Extension in the 5 0 Direction. Following initiationof rod assembly, there is rapid growth ofthe rod in the 5 0 direction “pulling” the RNAthrough the central hole until about 300 nucleotidesare coated (Figure 5.4C, steps 4–8). It isgenerally agreed that rod extension is faster inthe 5 0 than in the 3 0 direction by the addition ofdouble disks. It is uncertain how the 5 0 capstructure of the RNA is encapsidated. Disassemblyby ribosomes (see Chapter 7) and in vivowould suggest that the structure at the extreme5 0 end might differ from that over most of thevirus particle.Rod Extension in the 3 0 Direction. While 5 0extension of the rod is rapid and by addition ofdouble disks, 3 0 extension is much slower. It isthought that 3 0 extension of the assembling rodsis by the addition of small A protein aggregates.b. Assembly in Vivo. Evidence exists thatthe process involved in the initiation of assemblyoutlined earlier is almost certainly usedin vivo. However, there is no evidence thatestablishes the method by which the TMV rodelongates in vivo, but there is no reason to supposethat it differs from the mechanism that hasbeen proposed for in vitro assembly.types: tetrahedral (2:3), octahedral (4:3:2), andicosahedral (5:3:2); most isometric virus particleshave icosahedral symmetry. The basic icosahedronhas 60 identical subunits arrangedidentically on the surface of a sphere and displayfivefold, threefold, and twofold rotationalsymmetry (Figure 5.5). A shell made up ofmany small identical protein molecules makesmost efficient use of a virus’s genetic material.B. Possible IcosahedraIn developing potential structures for sphericalviruses a problem arises with the limitationof 60 protein subunits. A very basic infectiousgenome would code for a capsid protein(about 1,200 nucleotides) and a polymerase(about 2,500 nucleotides), which would requirea minimum internal “hole” of radius about 9nm. A 60 subunit icosahedron made of coatprotein of about 20–30 kDa (the usual size of aviral coat protein) would have an internal“hole” of about 6 nm and thus would not besufficiently large to encapsidate most viral genomes.Caspar and Klug (1962) enumerated allthe possible icosahedral surface lattices andthe number of structural subunits involved.The basic icosahedron (Figure 5.5), with 12groups of 5 structural subunits (termed pentamers)(or 20 3), giving a total of 60 structuralsubunits, can be subtriangulated (Box 5.2).IV. ARCHITECTURE OFISOMETRIC VIRUSESA. IntroductionFrom crystallographic considerations, Crickand Watson concluded that the protein shellof a small “spherical” or isometric virus couldbe constructed from identical protein subunitsarranged with cubic symmetry, which has threeFIGURE 5.5 The regular icosahedron. The 5-fold, 3-fold,and 2-fold symmetry axes are identified.II. WHAT IS A VIRUS MADE OF?

IV. ARCHITECTURE OF ISOMETRIC VIRUSES95BOX 5.2BASIC ICOSAHEDRAL SYMMETRY STRUCTUREThe basic icosahedron of 60 structural subunitscan be subtriangulated according to theformula:T ¼ PAf 2 ,where T is called the triangulation number.Parameter f is based on the fact that the basictriangular face can be subdivided by lines joiningequally spaced divisions on each side.Such a sheet can be folded down to give thebasic icosahedron by cutting out one trianglefrom a hexagon (e.g., cross-hatching) and thenjoining the cut edges to give a vertex with fivefoldsymmetry.However, if each vertex is joined to anotherby a line not passing through the nearest vertex,other triangulations of the surface are obtained.In the simplest case, the “next but one” verticesare joined.Thus, f is the number of subdivisions of eachside and is the number of smaller trianglesformed. There is another way in which subtriangulationcan be made, and this is represented byP. It is easier to consider a plane network ofequilateral triangles.This gives a new array of equilateral triangles,and the plane net can be folded to givethe solid shown in Figure 5.6A by removingthe shaded triangle from each of the originalvertices and then folding to give a vertex with5-fold symmetry.II. WHAT IS A VIRUS MADE OF?

96 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLESAFIGURE 5.6 Icosahedral symmetry.A. Ways of subtriangulation ofthe triangular faces of the basic icosahedronshown in Figure 5.5; 1, Thebasic icosahedron with T ¼ 1; 2, withT ¼ 4; 3, with T ¼ 3; 4, with T ¼ 12.[From Caspar and Klug (1962),# Cold Spring Harbor LaboratoryPress.] B. Structure of T ¼ 3 particlesof TYMV with reconstructed densitydistribution on particles showing 1,2-fold axis; 2, 5-fold axis; 3, 3-foldaxis. [This article was publishedin J. Molec. Biol., 72, J.E. Mellemaand L.A. Amos, Three-dimensionalimage reconstruction of turnip yellowmosaic virus, pp. 819–822, CopyrightElsevier (1972).]BThe process outlined in Box 5.2 leads to expansionof the particle by adding groups of six subunits(hexamers) in predetermined positionsbetween the pentamer clusters (Figure 5.6A,panels 1–4). This gives a series of potential spheresof increasing size depending on the T number. Thepreceding formula gives limitations for the valuesof T, and the most common ones found in virusesare listed in Table 5.1. From this table it can be seenthat with increasing particle size the number ofTABLE 5.1Triangulation Numbers and Sizes of Isometric Virus ParticlesTriangulationNumber TNumber ofSubunitsNumber ofPentamersNumber ofHexamersApproximateDiameter (nm)Example1 60 12 0 17 Satellite tobacco necrosis virus30 Cowpea mosaic virus a28 Poliovirus a3 180 12 20 25–30 Tomato bushy stunt virus30 Flock House virus25 Norwalk virus27 Enterobacteria phage MS24 240 12 30 40 Nudaurelia capensis o virus40 Sindbis virus core7 420 12 60 50 Cauliflower mosaic virus45 Simian virus 4013 780 12 120 75 Wound tumor virus75 Bluetongue virus corea Pseudo T ¼ 3 (see text).II. WHAT IS A VIRUS MADE OF?

V. SMALL ICOSAHEDRAL VIRUSES97pentamers remains constant (12) and that thenumber of hexamers increases.C. Clustering of SubunitsThe actual detailed structure of the virussurface will depend on how the physical subunitsare packed together. For example, threeclustering possibilities for the basic icosahedronare as follows:In fact many smaller viruses are based on theP ¼ 3, f ¼ 1, T ¼ 3 icosahedron. In this structure,the structural subunits are commonlyclustered about the vertices to give pentamersand hexamers of the subunits. These arethe morphological subunits seen in electronmicrographs of negatively stained particles(Figure 5.6B, 1–3).D. QuasiequivalenceIn the basic T ¼ 1 icosahedron each of thesubunits is in the same equivalent position tothe adjacent subunits. However, when furthersubunits are introduced to form hexamers, it isnot possible for them to occupy equivalent positions.Caspar and Klug (1962) assumed that notall the chemical subunits in the shell need to bearrayed in a strictly mathematically equivalentway. They also assumed that the shell is heldtogether by the same type of bonds throughoutbut that the bonds may be deformed in slightlydifferent ways in different nonsymmetryrelatedenvironments. This is termed quasiequivalence.Quasiequivalence would occurin all icosahedra except the basic structure(Fiure 5.5). Basically, pentamers give the3-dimensional curvature, 12 giving a closedicosahedron, whereas hexamers give a 2-dimensional curvature (or tubular structure).Thus, the conceptual basis of quasiequivalenceis the interchangeable formation of hexamersand pentamers by the same protein subunit.V. SMALL ICOSAHEDRAL VIRUSESA. Subunit StructureThe coat protein subunits of most small icosahedralviruses are in the range of 20–40 kDa;some are larger but fold to give effective “pseudomolecules”within this range. In contrast torod-shaped viruses, the subunits of most smallicosahedral viruses have a relatively high proportionof ß-sheet structure and a low proportionof a-helix and most have the same basicstructure. This comprises an eight-strandedantiparallel ß sandwich, often termed a ß-barrel,which is shown schematically in Figure 5.7A.The overall shape is a 3-dimensional wedgewith the B-C, H-I, D-E, and F-G turns beingat the narrow (interior) end. Most variationbetween the subunit sizes occurs at the N- andC-termini and between the strands of ß-sheet atthe broad end of the subunit. Figure 5.7B showsthe ß-barrel structure for the coat proteins ofthree viruses that have icosahedral particles.It is the detailed positioning of the elementsof the ß-barrel and of the N- and C-terminithat give the flexibility to overcome the quasiequivalenceproblems. The coat protein subunitscomprise one, two, or three structural domains;the S (shell) domain; the R domain (randombinding);and the P (protruding) domain; allviruses have the S domain. The R domain issomewhat of a misnomer but defines an N-terminal region of the polypeptide chain thatassociates with the viral RNA. As it is random,no structure can be determined by X-ray crystallography.The P domain gives surface protuberanceson some viruses.II. WHAT IS A VIRUS MADE OF?

98 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLESABFIGURE 5.7 Structure of protein subunits of icosahedral viruses. A. Basic structure of the ß-barrel with the 8 ß-sheetslabelled B-I; the N- and C-termini are shown; the cylinders represent conserved helices. [Reprinted from Encyclopedia ofvirology, Vol. 5, Johnson and Spier, pp. 393–400, Copyright Elsevier (2008).] B. Diagrammatic representation of the backbonefolding of TBSV, SBMV, and STNV. [Reprinted from J. Molec. Biol., 165, M.G., C. Abad-Zapatero, M.R.N. Murty, L. Liljas,A. Jones, and B. Strandberg, Structural comparisons of some small spherical plant viruses, pp. 711–736, Copyright (1983),with permission from Elsevier.]B. Virion StructureAt present, we can distinguish seven kinds ofstructure among the protein shells of small icosahedral-or icosahedral-based plant viruseswhose architecture has been studied in sufficientdetail. These are T ¼ 1 particles, bacilliform particlesbased on T ¼ 1, geminate particles based onT ¼ 1, T ¼ 3 particles, bacilliform particles based onT ¼ 3, pseudo T ¼ 3particles,andT¼ 7particles.1. T = 1 ParticlesThe satellite viruses are the smallest knownplant viruses, with a particle diameter of about17 nm and capsids made up of 17–21 kDa polypeptides.The structure of Satellite tobacco necrosisvirus (STNV) was the first to be solved and shownto be made of 60 protein subunits of 21.3 kDaarranged in a T ¼ 1 icosahedral surface lattice.The general topology of the polypeptide chainof the coat protein (Figure 5.7C) shows an Sdomain but no P domain. Three different sets ofmetal ion binding sites (probably Ca 2þ ) have beenlocated. These link the protein subunits together.2. Other Particles Based on T = 1 Symmetrya. Bacilliform Particles Based on T = 1Symmetry. Some virus particles—for instance,those of Alfalfa mosaic virus (AMV; see Profile 3in the Appendix)—are bacilliform with roundedends separated by a tubular section. The structureof these particles is based on icosahedralsymmetry. The rounded ends would have constraintsof icosahedra with the three-dimensionalcurvature determined by 12 pentamers,6 at each end. The two-dimensional structureof the tubular section would be made up of hexamers.Thus, in these virus particles the hexamersare not interspersed between the pentamers.Purified preparations of AMV contain fournucleoprotein components present in majorII. WHAT IS A VIRUS MADE OF?

V. SMALL ICOSAHEDRAL VIRUSES99FIGURE 5.8 Structure of AMV particles.A. Sizes (in nm) of the four main classes(components) of particles. [From Hull(2002).] B. Geodestix models showing theproposed structures of the components ofAMV shown in panel A. [This article waspublished in Virology, 37, R. Hull, G.J.Mills, and R. Markham, Studies on alfalfamosaic virus. II. The structure of the viruscomponents, pp. 416–428, CopyrightElsevier (1969).]AABamounts (bottom, B; middle, M; top b, Tb; and topa, Ta) that each contain an RNA species of definitelength. Three of the four major components arebacilliform particles, 19 nm in diameter and differinglengths. The fourth (Ta) is normally spheroidalwith a diameter of 19 nm (Figure 5.8A).It is suggested that the tubular structure ofthe bacilliform particles is based on a T ¼ 1 icosahedroncut across its 3-fold axis with rings ofthree hexamers (18 coat protein monomers)forming the tubular portion (Figure 5.8B).b. Geminiviruses. Geminiviruses containssDNA and one type of coat polypeptide. Particlesin purified preparations consist of twinnedor geminate icosahedra (see Profile 7). The finestructure of Maize streak virus (MSV) particleswas determined by cryoelectron microscopy,and three-dimensional image reconstructionshows that they consist of two T ¼ 1 icosahedrajoined together at a site where one morphologicalsubunit is missing from each, giving a total of 22morphological subunits in the geminate particle.3. T = 3 ParticlesThe isometric particles of many plant viruseshave T ¼ 3 structure. The major structuralfeatures found in many of these viruses are exemplifiedin Tomato bushy stunt virus (TBSV; Profile16) (Figure 5.7B) and Turnip yellow mosaic virus(Profile 18). The structure of TBSV has been determinedcrystallographically to 2.9Å resolution andthe virus shown to contain 180 protein subunitsarranged to form a T ¼ 3icosahedralsurfacelattice,with prominent dimer clustering at the outsideof the particle, the clusters extending to aradius of about 17 nm. The detailed structureshows that coat protein has two different largescaleconformational states in which the anglebetween the P and S domains differ by about20 o ; the conformation taken up depends onwhether the subunit is near a quasi dyad (2-foldII. WHAT IS A VIRUS MADE OF?

100 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLESaxis) or a true dyad in the T ¼ 3 surface lattice.Further into the particle the R domain forms aflexible arm that winds with the equivalentdomain of adjacent subunits to form an interlockingnetwork. Thus, the detailed external andinternal structures differ.a. Bacilliform Particles Based on T = 3Symmetry. As noted in the preceding sectionon bacilliform particles, it has been suggestedthat the structure of bacilliform particles is basedon icosahedral symmetry. Rice tungro bacilliformvirus (Profile 4) has bacilliform particles of 130 30 nm, which are thought to be based on aT ¼ 3 icosahedral symmetry cut across its 3-foldaxis.b. Pseudo T = 3 Symmetry. Comoviruses(Profile 6), and most likely nepoviruses, fabaviruses,and sequiviruses together with picornaviruses,have icosahedral particles formedof one, two, or three coat protein species. If allthe coat protein species are considered as oneprotein, the symmetry would appear to beT ¼ 1. However, the larger polypeptides ofviruses with one or two protein species formto give “pseudomolecules,” and for each virusthe structure can be considered to be made upof effectively three “species.” This then gives apseudo T ¼ 3 symmetry.4. T = 7 ParticlesCauliflower mosaic virus (CaMV) has a verystable isometric particle about 50 nm in diameter(Profile 4) containing dsDNA. The circulardsDNA is encapsidated in subunits of a proteinprocessed from a 58 kDa precursor to several products,the major ones being approximately 37 and42 kDa. Electron microscopy shows a relativelysmooth protein shell with no structural features.Calculations from MW of the coat protein andthe amount of protein in the virus suggested thatit may have a T ¼ 7 icosahedral structure, whichwas supported by cryoelectron microscopy to aresolution of about 3 nm.C. The Arrangement of Nucleic AcidWithin Icosahedral VirusesFor most icosahedral viruses, it is not possibleto gain a detailed picture of the arrangementof the nucleic acid within the particles, as it doesnot form an ordered structure that can be identifiedby current techniques. In some cases, thereis some ordered structure. For example, nearly20 percent of the RNA in the comovirus Beanpod mottle virus particles bind to the interior ofthe protein shell in a manner displaying icosahedralsymmetry. The RNA that binds issingle-stranded, and interactions with the proteinare dominated by nonbonding forces withfew specific contacts. Following are some generalpoints about the arrangement of nucleicacid within these particles.1. RNA StructureIt is probable that the RNA inside manysmall icosahedral viruses has some doublehelicalstructure. For example, about 75 percentof the RNA of Turnip yellow mosaic virus is ina-helical form, and about 95 percent of Cowpeachlorotic mottle virus has an ordered secondarystructure. Computer predictions of secondarystructure for 20 viral RNAs indicate that thegenomes of icosahedral viruses had higherfolding probabilities than those of helicalviruses. When the folding probability of a viralsequence is compared with that of a randomsequence of the same base composition, theviral sequences are more folded. These resultssuggest that base sequence plays some partin the way in which ssRNA genomes foldwithin the virus.2. Interactions Between RNA and Proteinin Small Isometric VirusesTwo types of RNA-protein interaction in thesmall isometric viruses can occur, dependingon whether basic amino acid side chains, usuallyin the mobile R domain, or polyaminesneutralize charged phosphates of the RNA.II. WHAT IS A VIRUS MADE OF?

D. Stabilisation of Small IsometricParticlesSmall isometric virus particles are stabilisedin three ways: by protein-RNA interactions,by protein-protein interactions, and by a combinationof the two.1. Protein-RNA StabilisationThe particles of alfamo- and cucumo-virusesare stabilised by electrostatic interactionsbetween basic amino acid side chains and RNAphosphate groups. These viruses are unstable inhigh salt concentrations dissociating to proteinand RNA. One feature is that these viruses areunstable in some negative stains used for electronmicroscopy.2. Protein-Protein StabilisationThe particles of some viruses, such as comoviruses,are very stable, and virus preparationshave “empty” particles that do not containRNA. These particles are stabilised by stronginteractions between protein subunits with littleor no involvement of the RNA.3. Protein-Protein + Protein-RNAStabilisationThe particles of viruses, such as bromoviruses,and sobemoviruses, are relatively stable at lowpHs but become unstable above pH 7. As thepH is raised between 6.0 and 7.0, the particles ofBrome mosaic virus swell from a radius of 13.5nm to more than 15 nm. This is due to the protonationof carboxyl-carboxylate groups on adjacentprotein subunits, which interact at low pH butrepel each other at higher pH. The swollen particleis then just stabilised by protein-RNA interactionsand becomes salt sensitive. A similarsituation is found in Southern bean mosaic virus(SBMV) with the addition of divalent cation, usuallyCa 2þ , links between adjacent subunits. Forswelling of the particles to occur the pH has tobe raised, and there must be a divalent cation chelator,such as EDTA, present.VII. ENVELOPED VIRUSESVI. MORE COMPLEX ISOMETRICVIRUSESPhytoreoviruses have distinctly angularparticles about 65–70 nm diameter that contain12 pieces of dsRNA. Six different proteins arepresent in the particle of Rice ragged stunt virus(Profile 11; see Appendix), and seven are presentin the capsid of Rice dwarf virus (RDV).Unlike most animal Reoviruses, which have triple-shelledcapsids, the particles of phytoreovirusesconsist of an outer shell of protein andan inner core containing protein. However,there is no protein in close association withthe RNA. The particles are readily disruptedduring isolation by various agents. Under suitableconditions, subviral particles can be produced,which lack the outer shell and revealthe presence of 12 projections at the fivefoldvertices of an icosahedron.The structure of RDV has been derived to25 Å resolution using cryoelectron microscopyand image reconstruction. This revealedtwo distinct icosahedral shells: a T ¼ 13 outershell, 700 Å diameter, composed of 260 trimericclusters of P8 (46 kDa), and an inner T ¼ 1shell, 567 Å diameter and 25 Å thick, made upof 60 dimers of P3 (114 kDa). The T ¼ 1coreplays a critical role in the organisation of thequasiequivalence of the T ¼ 13 outer capsidand probably guides the assembly of the outercapsid.VII. ENVELOPED VIRUSES101Two families of plant viruses with envelopedparticles exist. Rhabdoviridae is a familyof viruses whose members infect vertebrates,invertebrates, and plants. The virus particles(see Profile 13 in Appendix and Figure 2.5C)have a complex structure. Rhabdoviruses fromwidely differing organisms are constructed onthe basic plan shown in Figure 5.9A.II. WHAT IS A VIRUS MADE OF?

102 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLESABFIGURE 5.9 A. Model for rhabdovirus structure showing the proposed three-dimensional relationship between the viralproteins and the lipid layer. [From Cartwright et al. (1972; J. Virol. 10, 256–260; DOI 10.1145/567752.567759, reproduced withpermission from American Society for Microbiology).] B. Model showing structure of TSWV particle. The three nucleoproteincomponents are shown inside the particle with glycoprotein spikes (G1 and G2) extending through the lipid envelope. [FromKormelink (2005; Association of Applied Biologists, Descriptions of Plant Viruses, No. 412).]Some animal rhabdoviruses may be bulletshaped,but most, and perhaps all plantmembers, are rounded at both ends to give abacilliform shape. The particles comprise a helicalribonucleoprotein core made up of the viralgenome and the N protein and the viral polymerasesurrounded by the matrix (M) protein.This is enveloped in a membrane throughwhich glycoprotein (G) spikes protrude as surfaceprojections. It is suggested that the structureof the M protein layer is based on halficosahedral rounded ends as in other bacilliformparticles and that the G proteins relate tothis structure.Tospovirus particles (Profile 17; see Appendix)are spherical with a diameter of 80–110 nmand comprise a lipid envelope encompassingthe genomic RNAs, which are associated withthe N protein as a nucleoprotein complex. Theviral polymerase protein is also contained withinthe particle. The lipid envelope contains twotypes of glycoproteins (Figure 5.9B).VIII. ASSEMBLY OF ICOSAHEDRALVIRUSESA. BromovirusesThe protein subunits of several bromovirusescan be reassembled in vitro to give avariety of structures. In the presence of viralRNA, the protein subunits reassemble to formparticles indistinguishable from native virus.However, the conditions used are nonphysiologicalin several respects. The assembly mechanismof bromoviruses is thought to involvethe carboxyl-carboxylate pairs just mentioned.B. RNA Selection During Assemblyof Plant ReovirusesEvery plant reovirus particle appears to containone copy of each genome segment. Thus,there is a significant problem as to the macromolecularrecognition signals that allow one,II. WHAT IS A VIRUS MADE OF?

IX. GENERAL CONSIDERATIONS103and one only, of each of 10 or 12 genome segmentsto be incorporated into each particleduring virus assembly. For example, the packagingof the 12 segments of RDV presumablyinvolves 12 different and specific protein-RNAand/or RNA-RNA interactions. Thus, theremust be two recognition signals: one that specifiesa genome segment as viral rather than hostand one that specifies each of the 12 segments.The 5 0 - and 3 0 -terminal domains are fully conservedbetween the genome segments and it issuggested that this determines the recognitionof viral rather than host RNA.In Reovirus the (–)-sense strands are synthesisedby the viral replicase on a (þ)-sensetemplate that is associated with a particulatefraction. These and related results led tothe proposal that dsRNA is formed within thenascent cores of developing virus particlesand that the dsRNA remains within these particles.If true, this mechanism almost certainlyapplies to the plant reoviruses. It implies thatthe mechanism that leads to selection of a correctset of 12 genomic RNAs involves the ssplus strand. It is thought that selection isdirected by base-paired inverted repeats foundin the genomic RNAs. It may be that othervirus-coded “scaffold” proteins transientlypresent in the developing core are involved inRNA recognition rather than, or as well as,the three proteins found in mature particles.IX. GENERAL CONSIDERATIONSPlant viruses, because they occur at muchhigher concentrations in infected plants andcan be relatively easily purified, have providedmuch of the information on basic virus structure.The basic concepts are not limited to plantviruses but apply equally to viruses from otherkingdoms.The virus particle protects the viral genome outsidethe cell and is often involved in transmission(see chapter 12). It is also likely that the formationof particles soon after viral replication sequestersthe viral nucleic acid, thus controlling the impactof its expression on the host cell.The idea of quasiequivalence in the bondingbetween subunits in icosahedral protein shellswith T > 1 as put forward by Caspar and Klugin 1962 is still useful but has required modificationin the light of later developments. Viruseshave evolved at least two methods by which asubstantial proportion of the potential nonequivalencein bonding between subunits indifferent symmetry related environment canbe avoided: (1) by having quite different proteinsin different symmetry-related positionsas in the reoviruses and the comoviruses, and(2) by developing a protein subunit with twoor more domains that can adjust flexibly indifferent symmetry positions, as in TBSV andSBMV.The high-resolution analysis of the structuresof isometric viruses is showing that,although the outer surface of the shell showsquasiequivalent icosahedral symmetry, thismay not extend into the inner parts. There isfrequently interweaving, especially of theC-terminal regions of the polypeptide chain,which forms an internal network.The many negatively charged phosphategroups on the nucleic acid within a virus aremutually repelling. To produce a sufficientlystable virus particle, these charges, or at leastmost of them, need to be neutralized. Structuralstudies to atomic resolution have revealedthree solutions to this problem:1. In TMV, the RNA is closely confinedwithin a helical array of protein subunits.Two of the phosphates associated witheach protein subunit are close toarginine residues. However, theelectrostatic interactions between proteinandRNAarebestconsideredascomplementarity between two electrostaticsurfaces.II. WHAT IS A VIRUS MADE OF?

104 5. ARCHITECTURE AND ASSEMBLY OF VIRUS PARTICLES2. In TBSV and a number of other icosahedralviruses, a flexible basic amino-terminalarm with a histone-like composition projectsinto the interior of the virus interactingwith RNA phosphates. Additionalphosphates are neutralized by divalentmetals, especially Ca 2þ .3. In TYMV, where there is little interpenetrationof RNA and protein, the RNA phosphates areneutralized by polyamines and Ca 2þ ions.In the more complex particles with structurescomprising several layers of proteins, there isincreasing evidence for interactions betweenthe layers. It seems likely that these interactionsdrive the structural arrangements of proteinsubunits in adjacent layers.X. VIRUSES OF OTHER KINGDOMSMany of the basic principles of virus structurewere determined for plant viruses but arealso applicable to viruses of all kingdoms.XI. SUMMARY• The three basic virus particle shapes arerod-shape, isometric (spherical), andcomplex.• Rod-shaped particles have helical symmetry,and most isometric particles haveicosahedral symmetry.• The assembly of the rod-shaped particles ofTMV starts with the interaction of the origin ofassembly (OAS) with a double disk of coatprotein converting the double disk into a “lockwasher,” continues with the further additionof double disks ("lock washers") to the 5 0 sideof the OAS, and is completed by the addition ofcoat protein subunits to the 3 0 side of the OAS.• Considerations of icosahedral symmetry putconstraints on the structural possibilities oficosahedral viruses.• The bonding between structural subunits oficosahedra larger than the basic icosahedroninvolves quasiequivalent interactions.• Variations on the basic icosahedral symmetrylead to bacilliform and geminate particles.• Isometric virus particles are stabilised byprotein-RNA interactions, protein-proteininteraction, and divalent cation binding;viruses differ in the forms of particlestabilisation.• Complex virus structures involve bothhelical and icosahedral symmetry, two ormore shells of icosahedral symmetry, andenveloping in cellular membranes.ReferencesCaspar, D.L.D. and Klug, A. (1962). Physical principles inthe construction of regular viruses. Cold Spring HarborSymp. Quant. Biol. 27, 1–24.Crick, F.H.C. and Watson, J.D. (1956). Structure of smallviruses. Nature (London) 177, 473–475.Fraenkel-Conrat, H. and Williams, R.C. (1955). Reconstitutionof active tobacco mosaic virus from its inactive proteinand nucleic acid components. Proc. Natl. Acad. Sci.USA. 41, 690–698.Schachman, H.K. (1959). Ultracentrifugation in biochemistry.Academic Press, New York.Further ReadingChiu, W., Chang, J.T., and Rixon, F.J. (2008). Cryoelectronmicroscopy. Encyclopedia of Virology, Vol. 1, 603–613.Hull, R. (2002). Matthew’s plant virology. Academic Press,San Diego.Johnson, J.E. and Speir, J.A. (2008). Principles [of virusstructure]. Encyclopedia of Virology, Vol. 5, 393–400.Luo, M. (2008). Nonenveloped [viruses]. Encyclopedia ofVirology, Vol. 1, 200–203.Navaratnarajah, C.K., Warrier, R., and Kuhn, R.J. (2008).Enveloped [viruses]. Encyclopedia of Virology, Vol. 1,193–199.Peremyslov, V.V., Andreev, I.A., Prokhnevsky, A.I., Duncan,G.H., Taliansky, M.E., and Dolja, V.V. (2004). Complexmolecular architecture of beet yellow virus particles.Proc. Natl. Acad. Sci. USA 101, 5030–5035.Speir, J.A. and Johnson, J.E. (2008). Non-enveloped virusstructure. Encyclopedia of Virology, Vol. 5, 380–392.II. WHAT IS A VIRUS MADE OF?

C H A P T E R6Plant Viral GenomesViral genomes contain the information for the replication and expression of genes necessary forthe functioning of the virus at the right time and in the right place.O U T L I N EI. Introduction 105II.General Properties of Plant ViralGenomes 105III. Plant Viral Genome Organisation 112IV. Viruses of Other Kingdoms 116V. Summary 116I. INTRODUCTIONOne of the main features of the last two or threedecades has been the great explosion of sequencedata on virus genomes. The first plant viralgenome to be sequenced was the DNA of CaMVin 1980, followed by the RNA of TMV in 1982.By 2007 full genomic sequences were availablefor at least one member of the 80 genera of plantviruses and of many species within most of thegenera. This plethora of data has led to muchcomparison between sequences often going intofine detail beyond the scope of this book. However,the comparisons, coupled with the use ofinfectious clones of viruses, have given valuableinformation for use in mutagenesis experimentsto elucidate the functions of various gene productsand noncoding regions. In this chapter,we examine the various gene products thatviruses encode and, in general terms, the genomeorganisations from which they are expressed.II. GENERAL PROPERTIES OFPLANT VIRAL GENOMESBasically, the viral genome, regardless ofwhat kingdom it infects, comprises codingregions that express the proteins required forthe viral infection cycle (initial infection,Comparative Plant Virology, Second Edition105Copyright # 2009, Elsevier Inc. All rights reserved.

106 6. PLANT VIRAL GENOMESmovement through the host, interactions withthe host, and movement between hosts) andnoncoding regions that control the expressionand replication of the genome; control sequencescan also be found in the coding regions.A. Information ContentIn theory, the same nucleotide sequence in aviral genome could code for up to 12 polypeptides.There could be an open reading frame(ORF) in each of the three reading frames ofboth the positive (þ)- and negative (–)-sensestrands, giving six polypeptides. Usually anORF is defined as a sequence commencing withan AUG initiation codon and capable of expressinga protein of 10 kDa or more. If each ofthese ORFs had a leaky termination signal, theycould give rise to a second read-through polypeptide.However, in nature, there must besevere evolutionary constraints on such multipleuse of a nucleotide sequence, because evena single base change could have consequencesfor several gene products. However, two overlappinggenes in different reading frames dooccasionally occur, as do genes on both (þ)-and (–)-sense strands. Read-through and frameshiftproteins are quite common and aredescribed in Chapter 7.The number of genes found in plant virusesranges from 1 for the satellite virus, STNV, to 12for some closteroviruses and some reoviruses.Most of the single-stranded (ss) (þ)-sense RNAgenomes code for about four to seven proteins.In addition to coding regions for proteins, genomicnucleic acids contain nucleotide sequenceswith recognition and control functions that areimportant for virus replication and expression.B. Economy in the Use of GenomicNucleic AcidsViruses make very efficient use of the limitedamount of genomic nucleic acids they possess.Eukaryote genomes may have a content ofintrons that is 10–30 times larger than that ofthe coding sequences. Like prokaryote cells,most plant viruses lack introns, but some donot (see Chapter 7). Plant viruses share withviruses of other host kingdoms several otherfeatures that indicate very efficient use of thegenomic nucleic acids:• Coding sequences are usually very closelypacked, with a rather small number ofnoncoding nucleotides between genes.• Coding regions for two different genes mayoverlap in different reading frames [e.g., inPotato virus X (Profile 9; see Appendix) andTurnip yellow mosaic virus (Profile 18)], or onegene may be contained entirely withinanother in a different reading frame [e.g., theLuteovirus genome (Profile 8) and Tomatobushy stunt virus (Profile 16)].• Read-through of a “leaky” terminationcodon may give rise to a second, longerread-through polypeptide that is coterminalat the amino end with the shorter protein.This is quite common among the virusgroups with ss (þ)-sense RNA genomes.Frameshift proteins in which the ribosomeavoids a stop codon by switching to anotherreading frame have a result that is similar toa “leaky” termination signal; these aredescribed in Chapter 7.• A single gene product may have more thanone function. For example, the coat proteinof Maize streak virus (MSV) has a protectivefunction and is involved in insect vectorspecificity, cell-to-cell transport of the virus,nuclear transport of the viral DNA, andpossibly symptom expression and control ofreplication.• Functional introns have been found in severalgeminiviruses and in Rice tungro bacilliformvirus (RTBV; Profile 4). Thus, mRNA splicing,a process that can increase the diversity ofmRNA transcripts available and therefore thenumber of gene products, may be a featurecommon in viruses with a DNA genome.II. WHAT IS A VIRUS MADE OF?

II. GENERAL PROPERTIES OF PLANT VIRAL GENOMES107• A functional viral enzyme may use a hostcodedprotein in combination with a viruscodedpolypeptide (see Chapter 8).• Regulatory functions in the nucleotidesequence may overlap with codingsequences (e.g., the signals for subgenomicRNA synthesis in TMV).• In the 5 0 and 3 0 noncoding sequences of thessRNA viruses, a given sequence ofnucleotides may be involved in more thanone function. For example, in genomic RNA,the 5 0 -terminal noncoding sequences mayprovide a ribosome recognition site and atthe same time contain the complementarysequence for a replicase recognition site inthe 3 0 region of the (–) strand.C. The Functions of Viral Gene ProductsThere are two types of viral gene products:structural and functional. Structural gene productscomprise the coat proteins described inChapter 5 and, for complex viruses, scaffoldproteins that direct the formation of the viralcoat. Few, if any, examples of scaffold proteinsin plant viruses exist.1. Functional ProteinsTable 6.1 lists the presence of functional proteinsfor viruses of the various kingdoms (seealso Box 4.2).a. Proteins That Initiate Infection. Plants,animals, and bacteria differ in the surfaces thatthey present to an incoming virus (see Box 2.2).The cells that are first infected in bacteria andanimals usually exist in a liquid medium. Thus,viruses reach the cell by simple diffusion, andthe initial interaction with the virus is at the cellmembrane surface. Entry into these cells is usuallymediated by cell surface receptors and, inthe case of some bacteriophages, by specialisedvirus mechanisms. In contrast, the plant surfaceis a waxy cuticle, and each cell is surroundedby a cellulose cell wall. Thus, the incomingvirus has to pass these two barriers by variousforms of mechanical damage, and there are nospecialised cell surface receptors. There are alsodifferences in bacteria, animals, and plants inthe ways that viruses spread from the initiallyinfected cell(s) to surrounding cells. Bacteriaare unicellular, and cells in an animal body usuallyhave no or few cytoplasmic connectionsbetween them. Thus, the initially infected bacterialor animal cells normally release virus frominfected cells and infect further cells by interactingwith surface recognition sites. In contrast,although plant cells have a cellulose cell wall,there are cytoplasmic connections (plasmodesmata)between the cells. A plant virus movesfrom cell to cell via plasmodesmata and vasculartissue throughout almost the entire plant(see Chapter 9). Thus, as far as a virus isTABLE 6.1Functional Viral ProteinsVirusProtein Plant Vertebrate Invertebrate Fungus ProkaryoteInitiate infection –/þ a þ þ þ þReplicate viral genome þ þ þ þ þProcess gene product þ þ þ þ þMovement through host þ – – Check CheckMovement host to host þ – – Check Checka See text.II. WHAT IS A VIRUS MADE OF?

108 6. PLANT VIRAL GENOMESconcerned, one can consider a bacterial colonyor an animal as a collection of cells in a liquidmedium from which a virus has to exit througha cell membrane and enter a new cell throughthe membrane. As far as a virus is concerned,a plant is a single cell. However, plant virusesthat circulate in their insect vectors have tocross various barriers, such as the gut and salivarygland accessory cell walls. These barriersare crossed by receptor-mediated mechanisms(see Chapter 12).Within plant cells, the situation may be quitedifferent. There is increasing evidence for theinvolvement of membranes in virus replicationand in cell-to-cell movement. Thus, it is likelythat there are receptor sites on intracellularmembranes that are targeted by viral proteinsrather than virus particles. Another aspect ofinitiating and establishing virus infection isovercoming host defences. Various viral geneproducts are involved in this and are describedin Chapters 10 and 11.b. Proteins That Replicate the ViralGenome. It is now generally accepted that allviruses, except satellite viruses (see Chapter 3),code for one or more proteins that have anenzymatic function in nucleic acid synthesis,either genomic nucleic acid or mRNAs or both.The general term for these enzymes is polymerase.As there is some inconsistency in the literaturein relation to the terms used for differentpolymerases the terms used in this book areoutlined in Box 6.1.Some viral gene products are involved ingenome replication in a nonenzymatic way.For instance, the 5 0 VPg protein (see later in thischapter) found in some virus genera is thoughtto act as a primer in RNA synthesis, thus havinga nonenzymatic role in RNA synthesis.Alfalfa mosaic virus (AMV) coat protein has anessential role in the initiation of infection bythe viral RNA, possibly by priming (–)-strandsynthesis. This protein is discussed in moredetail in Box 8.5.c. Proteins That Process Viral Gene Products.Viruses in which the whole genome ora segment of the genome is first transcribedinto a single polyprotein usually encode oneBOX 6.1TERMS USED TO DESCRIBE POLYMERASESRNA-dependent RNA polymerase (RdRp). Catalyzestranscription of RNA from an RNAtemplate.RNA-dependent DNA polymerase (reversetranscriptase: RT). The enzyme coded, for example,by members of the Retroviridae and Caulimoviridae,which copies a full-length viral RNA intogenomic DNA.Replicase. The enzyme complex that makescopies of an entire RNA genome and the subgenomicmRNAs. Replicase enzymes often haveeither various functional domains or are madeup of virus-encoded subunits with differentfunctions. For instance, RNA ! RNA replicationcan involve methyl transferase and helicaseactivities, as well as the actual polymerase itself.Transcriptase. RdRp found as a functionalpart of the virus particle as in the Rhabdoviridaeand Reoviridae.There are no examples of viruses that replicateDNA directly from DNA using DNAdependentDNA polymerase (DdDp) amongplant viruses. In the Geminiviridae the viral geneproduct(s) associate(s) with the host DdDp.II. WHAT IS A VIRUS MADE OF?

II. GENERAL PROPERTIES OF PLANT VIRAL GENOMES109or more proteinases that process the polyproteininto functional proteins. These aredescribed in more detail in Box 4.3.d. Proteins That Facilitate Viral MovementThrough the Host. As noted previously,plant viruses differ from those of other kingdomsin that the virus moves from cell to cellthough cytoplasmic connections (plasmodesmata).For many plant viruses, one or more specificvirus-coded protein(s) is required for thiscell-to-cell movement and for systemic movementwithin the host plant (see Chapter 9).e. Overcoming Host Defence Systems. Aswe will see in Chapter 11, plants have a nucleicacid that targets the defence system. Successfulviruses encode one or more protein(s) that suppressthis defence system. Similarly, animaland bacterial viruses have ways of overcomingthe cognate host defence mechanism.f. Proteins That Facilitate the Host to HostMovement of Viruses. Gene products havebeen identified as essential for successful transmissionfrom plant to plant by invertebrate vectorsand possibly by fungal vectors (seeChapter 12).D. Nucleic AcidsAs shown in Table 1.5, plant viral genomescan comprise dsDNA, ssDNA, dsRNA, and ss(–)-sense RNA, but most commonly ss (þ)-senseRNA. In this section we examine some of theproperties of the ss RNA genomes.1. Multipartite GenomesOf the 80 genera of plant viruses, 33 havetheir genome in two or more separate piecesof different sizes, called multipartite genomes.In most genera the individual pieces are encapsidatedin separate particles (multicomponentviruses; Box 5.1). (See Brome mosaic virus(BMV) in Profile 3.)2. Nucleic Acid StructuresIn the intact virus particle, the three-dimensionalarrangement of the RNA is partly orentirely determined by its association with thevirus protein or proteins (Chapter 5). In solution,dsDNA has a well-defined secondary structureimposed by base-pairing and base-stacking inthe double helix; viral ssRNAs have no such regularstructure but under appropriate conditionscontain numerous short helical regions of intrastrandhydrogen-bonded base-pairing interspersedwith ss regions. However, as will bediscussed in Chapter 11, one of the host defencesystems targets dsRNA above a certain length,so there would be selection against extensivesecondary structure of free RNA in vivo.RNA molecules can fold into complex threedimensionalshapes and structures to performtheir diverse biological functions. The most prevalentof these forms is the pseudoknot, which, inits simplest manifestation, involves the loop of astem-loop structure base-pairing with a sequencesome distance away (Figure 6.1). The involvementof pseudoknots has been recognised in agreat variety of functions of viral RNAs, includingcontrol of translation by –1 frameshifting(see Chapter 7), by read-through of stop codons,by internal ribosome entry sites (see Chapter 7),and by translational enhancers.3. Noncoding Regionsa. End-Group Structures. Many plant viralssRNA genomes contain specialised structuresat their 5 0 and 3 0 termini. These are shown inthe genome maps in the Profiles.The 5 0 CapMany mammalian cellular messenger RNAsand animal virus messenger RNAs have a methylatedblocked 5 0 -terminal group of the form:m 7 G 5N ppp 5N X ðmÞ pY ðmÞ p ...where X (m) and Y (m) are two methylated bases.Some plant viral RNAs have this type of5 0 end, known as a “cap,” but in the knownII. WHAT IS A VIRUS MADE OF?

110 6. PLANT VIRAL GENOMESFIGURE 6.1 RNA pseudoknots. A. Secondary structure.The dotted lines indicate the base-pair formation of thenucleotides from the hairpin loop with the complementaryregion at the 5 0 side of the hairpin. B. and C. Schematic folding.D. Three-dimensional folding, showing the quasi-continuousdouble-stranded helix. The stem regions (S1 andS2) and the loop regions (L1 and L2) are indicated;L1 crosses a deep groove and L2 a shallow groove. [Modifiedfrom Deiman and Pleig (1997; Semin. Virol. 8, 166–175).]plant viral RNAs the bases X and Y are notmethylated. The capping activity is virus codedand differs from the host capping activity. Cappingactivity has been identified in several plantviruses including the 126 kDa TMV protein andin protein 1a encoded by RNA-1 of BMV. Theseactivities methylate GTP using S-adenosylmethionine(AdoMet) as the methyl donor, theguanyl transferase and transferase activity beingspecific for guanine position 7, forming a covalentcomplex with m 7 GTP.5 0 Linked ProteinMembers of several plant virus groups have arelatively small protein (c 3.5–24 kDa) covalentlylinked to the 5 0 end of the genome RNA. Theseare known as VPg’s (short for virus protein, genomelinked). All VPg’s are coded for by the virusconcerned, and for most viruses the viral genecoding for the VPg has been identified. If a multipartiteRNA genome possesses a VPg, all thegenomic RNAs will have the same proteinattached. The VPg is attached to the genomicRNAs by a phosphodiester bond between theß-OH group of a serine or tyrosine residue locatedat the NH 2 terminus of the VPg and the 5 0 -terminaluridine residue of the genomic RNA(s). VPgs areinvolved in virus replication (see Box 8.6).3 0 Poly(A) TractsPolyadenylate sequences have been identifiedat the 3 0 terminus of the messenger RNAsof a variety of eukaryotes. Such sequences havebeen found at the 3 0 terminus of several viralRNAs that can act as messengers. The lengthof the poly(A) tract may vary for differentRNA molecules in the same preparation, andsuch variation appears to be a general phenomenon.Internal poly(A) tracts are found inbromoviruses and hordeiviruses.3 0 tRNA-Like StructuresThe 3 0 termini of the RNA genomes of severalplant viruses have the property of acceptingand binding specific host tRNAs throughan ester linkage. The accepting activity is alsopresent in the ds replicative form of the viralRNA and in this state is resistant to RNaseattack, thus demonstrating that the amino acidaccepting activity is an integral part of the viral(þ)-sense ssRNA.The tRNA-like structures at the 3 0 termini ofa variety of plant viral RNAs form pseudoknots(Figure 6.2). These structures are involved inthe regulation of RNA replication (see Boxes8.4 and 8.5).Barley stripe mosaic virus (BSMV) RNAs areunusual in that they have an internal poly(A)sequence between the end of the coding regionII. WHAT IS A VIRUS MADE OF?

II. GENERAL PROPERTIES OF PLANT VIRAL GENOMES111FIGURE 6.2 tRNA-like structures at the 3 0 end of plant viral RNAs. A. TYMV RNA, valine aminoacylation; B. BMVRNA, tyrosine aminoacylation; C. TMV, histidine aminoacylation. Acc, aminoacyl acceptor stem; T, Tc-arm; D, D-arm;A/C, anticodon arm. [From Dreher (1999; Annu. Rev. Phytopathol. 37, 151–174), reprinted, with permission, from the AnnualReview of Phytopathology, Volume 37 # 1999 by Annual Reviews. www.annualreviews.org.] D. Three-dimensional model ofcanonical tRNA. [From Dumas et al. (1987; J. Biomol. Struct. Dyn. 4, 707–728).]II. WHAT IS A VIRUS MADE OF?

112 6. PLANT VIRAL GENOMESand the 3 0 tRNA-like sequence that acceptstyrosine.Complementary 5 0 and 3 0 SequencesThe RNA genome segments of members ofthe tospoviruses and tenuiviruses have complementarysequences at the 5 0 and 3 0 ends thatenable the termini of the RNAs to anneal toform “panhandle structures.” These sequencesare conserved across the genome segments ofmembers of each genus. Similar structures arefound in the genomes of vertebrate- and invertebrate-infectingBunyaviridae.b. 5 0 and 3 0 Noncoding Regions. The 5 0 and3 0 noncoding regions control both translationand replication. As we will see in Chapter 7,these two regions interact in the initiation oftranslation of, at least, the 5 0 open readingframes (ORFs). The 3 0 noncoding region is thesite of initiation of (–)-strand RNA synthesisand the 5 0 noncoding region (the 3 0 end of (–)-strand RNA) is the site of initiation of (þ)-strand synthesis (see Chapter 8).c. Intergenic Regions. Sequences in intergenicregions are also involved in both RNA synthesisand the translation of downstream ORFs.The initiation of synthesis of subgenomic RNAsis often in these regions and these RNAs are themessengers for translation of non-5 0 ORFs inmany viruses, as we will see in Chapter 7.An increasing number of interactions are occurringbetween terminal and internal sequenceregions in the control of expression of the genomicinformation from (þ)-strand RNAs. It is likely thatsimilar interactions will be found that control theexpression of (–)-strand and dsRNA genomesand the genomes of ss and ds DNA viruses.III. PLANT VIRAL GENOMEORGANISATIONThe genome organisations of representativeplant viruses are shown in the Profiles in theAppendix.A. Structure of the GenomeSeveral questions must be answered todetermine the structure of a viral genome:1. Is the genomic nucleic acid single- or doublestranded,DNA or RNA, linear or circular?2. How many pieces of nucleic acid make upthe basic infectious genome? How manypieces make up the genome of a virusfunctioning in the natural situation?3. What is the full nucleic acid sequence of theinfectious genome?4. What are the structures of the 5 0 and 3 0termini of a linear nucleic acid?5. How many ORFs does the genome contain?The answer to this question raises two morequestions:a. What is an ORF? The simple definition isthat it is a piece of the (þ)-sense mRNAcoding for a protein of more than 10 kDawith an AUG translation start codon andan appropriate stop codon. However,there are an increasing number of proteinsof less than 10 kDa that are being shownto be functional. Factors to be consideredwith such proteins include whether theyhave highly improbable amino acidcomposition or whether they havesequence similarity between smallORFs in several viruses which mayindicate that they are functional (e.g.,the 6 kDa ORF3 of Citrus tristeza virus;Profile 5).ORFs of significant size representingpossible proteins of 100 amino acids ormore occur in the (–)-sense strands ofseveral viruses that are normally regardedas being (þ) stranded (e.g., TMV, AMVRNA1, and RNA2). There is no evidencethat any of these have functionalsignificance. However, there is no reason,in principle, why functional ORFsshould not occur in the (–)-sense strand.Such ORFs are found in the geminiviruses,tenuiviruses, and tospoviruses (Profiles7, 12, and 17, respectively; see Appendix).II. WHAT IS A VIRUS MADE OF?

III. PLANT VIRAL GENOME ORGANISATION113ORFs do not necessarily start with theconventional AUG start codon. An AUUstart codon has been recognised for ORFI ofRTBV (see Box 7.5) and a CUG start codonfor the capsid protein of Soil-borne wheatmosaic virus. This phenomenon raises thequestion of the definition of an ORF.Conventionally, it starts with an AUGcodon and stops with one of the three stopcodons. If non-AUG start codons are morewidely used than believed, an ORF shouldbe a largish in-frame region without a stopcodon.b. Is the ORF functional? Some of the ORFsrevealed by the nucleotide sequence willcode for proteins in vivo, whereas othersmay not. The functional ORFs can beunequivocally identified by finding therelevant protein in infected cells or byin vitro translation studies using viralmRNAs (Box 6.2).Virus-coded proteins, other than thosefound in virus particles, may be difficultto detect in vivo, especiallyiftheyoccurinvery low amounts and are only transientlyexpressed during a particular phase of thevirus replication cycle. However, a batteryof methods is now available for detectingvirus-coded proteins in vivo and matchingthese with the ORFs in a sequenced viralgenome. In particular, the nucleotidesequence information gives a preciseestimate of the size and amino acidcomposition of the expected protein.Knowledge of the expected amino acidsequence can be used to identify thein vivo product either from a partialamino acid sequence of that productor by reaction with antibodies raisedagainst either a synthetic polypeptidethat matches part of the expected aminoacid sequence or against the ORF or partthereof expressed in, say, E. coli.6. What are the regulatory and recognitionsignals for expression of the mRNAs? Theseare found in various parts of the genome,BOX 6.2IN VITRO TRANSLATION SYSTEMSIn vitro translation systems are derived fromcells that have a high rate of protein synthesis.Here are the most commonly used systems:• E. coli cell-free system: The prokaryoticin vitro system derived from Escherichia colitranslates some monocistronic viral RNAswith fidelity but produces equivocal resultswith other RNAs.• The eukaryotic rabbit reticulocyte lysatesystem.• The eukaryotic wheat embryo system.Thesesystemscanbeusedforanalysisofclonedgenes [using coupled transcription:translation.]Some in vivo systems, such as the Toadoocyte system, have also been used for analysingplant viral RNAs. However, in vivo systems canhave problems with proteins that are toxic to thecells and when the expressed protein is rapidlyprocessed by intracellular proteases.Details of in vitro systems can be found in Jagus, R.(1987). Translation in cell-free systems. In Methodsin enzymology (S.L. Berger and A.R. Kimmel, Eds.),Vol. 152, pp. 267–296. Academic Press, San Diego.II. WHAT IS A VIRUS MADE OF?

114 6. PLANT VIRAL GENOMESparticularly the 5 0 and 3 0 noncodingsequences and between ORFs inpolycistronic RNAs. Regulatory sequencesmay also be found in coding regions.7. What are the mRNAs for ORFs ofpolycistronic RNA viral genomes and DNAviruses? It may be difficult to establishwhether a viral RNA of subgenomic size is afunctional mRNA or merely a partlydegraded or partly synthesised piece ofgenomic RNA.Quite often, genuine viral subgenomicmRNAs are encapsidated along with thegenomic RNAs. These can then be isolatedfrom purified virus preparations andcharacterised. When the sequence of thegenomic nucleic acid is known, twotechniques can be used to locate preciselythe 5 0 terminus of a presumed subgenomicRNA. In the S1 nuclease protectionprocedure, the mRNA is hybridized with acomplementary DNA sequence that coversthe 5 0 region of the subgenomic RNA.The ss regions of the hybridized moleculeare removed with S1 nuclease. The DNAthat has been protected by the mRNA isthen sequenced. In the second method,primer extension, a suitable ss primermolecule is annealed to the mRNA.Reverse transcriptase is then used toextend the primer as far as the 5 0 terminusof the mRNA and the DNA produces issequenced.B. Recognising Activities of Viral GenesBefore information on the sequence ofnucleotides in viral genomes became availableand before the advent of in vitro translationsystems, there were two ways to determinethe activities of viral genes: identification ofproteins in the virus particle and classic virusgenetics. These approaches are still relevant.Classic genetic studies have identified manybiological activities of viral genomes. Thediscovery of viruses with the genome dividedbetween two or three particles opened up thepossibility of locating specific functions on particularRNA species by comparing the physicaland biological properties of reassortments ofthe various components with those of the parentviruses after inoculation to appropriatehosts. This approach has been used on bothnatural and artificially induced mutants.The advent of a wide range of technologies—such as sequencing, mutagenesis, and recombinantDNA—has provided many approaches toaddressing the question of viral gene function.A few of these give unequivocal proof of function,whereas others are more or less stronglyindicative of a particular function. There aretwo basic groups of methods. In the first, whichmay not be generally applicable, the naturalgene product produced in infected tissue isisolated, and its activity is established by directmethods. Some virus-coded proteins besidescoat proteins have functions that can be identifiedin in vitro tests, such as the use of in vitrotranslation products to detect protease activity.Another approach is to express the viral genein either a prokaryotic system such as E. coli orin a eukaryotic system such as baculovirus vectorsin insect cells or a yeast vector in yeast.The eukaryotic systems are preferred as theyprovide posttranslational modification not foundin prokaryotic systems. Viral proteins producedin such systems can be used for in vitro experimentssuch as demonstrating the aphid transmissionhelper component activity of theproduct of CaMV ORF II (see Chapter 12). Thesecond group of methods involves, directly orindirectly, the use of recombinant DNA technology.This consists of the following approaches.1. Location of Spontaneous or ArtificiallyInduced MutationsKnowledge of nucleotide sequences in naturalvirus variants allows a point mutation to belocated in a particular gene, even if the proteinproduct has not been isolated. In this way, theII. WHAT IS A VIRUS MADE OF?

III. PLANT VIRAL GENOME ORGANISATION115changed or defective function can be allocatedto a particular gene. The temperature-sensitivemutant of TMV, known as LS1, serves as anexample. At the nonpermissive temperature, itreplicates and forms virus particles normallyfound in protoplasts and infected leaf cells butis unable to move from cell to cell in leaves.A nucleotide comparison of the LS1 mutantand the parent virus showed that the LS1mutant had a single base change in the 30kDa protein gene that substituted a serine fora proline. This was a good indication, but notdefinitive proof, that the 30 kDa protein isinvolved in cell-to-cell movement.The genomes of many DNA viruses andcDNAs to many RNA viruses have been clonedand the DNA or transcripts thereof shown to beinfectious. There are numerous examples ofexperiments in which point mutations, deletions,or insertions have been used to elucidatethe function(s) of the gene produced by themodified ORF. The introduction of definedchanges in particular RNA viral genes to studytheir biological effects, and thus define genefunctions, is commonly known as reversegenetics. This approach has been of majorimportance in understanding gene functions.2. Recombinant VirusesRecombinant DNA technology can be usedto construct viable viruses from segments ofrelated virus strains that have differing propertiesand thus to associate that property witha particular viral gene. For example, variousviable recombinants were constructed containingparts of the genome of two strains ofTMV, only one of which caused necrotic locallesions on plants such as Nicotiana sylvestris,which contain the N 0 gene. Infection withthese recombinants indicated that the viralfactor responsible for the necrotic responsein N 0 plants is coded for in the coat proteingene.One further application of recombinants isthe tagging of gene products with fluorescentor other probes that report where in the plantor protoplast that gene product is beingexpressed or accumulates. By using videoimaging, the sequence of events involved inthe functioning of the gene product can berecorded. This approach has been used tostudy the movement of plant viruses aroundthe cell.3. Expression of the Gene in a TransgenicPlantAs with mutagenesis of infectious clonedgenomes of viruses, the technique of transformingplants with viral (and other) sequenceshas had a major impact on understandingviral genes and control functions, and thereare numerous examples of their expression intransgenic plants. The basic features of thetechnique are that a construct comprising thegene of interest, a promoter (often the 35S promoterof CaMV), and a transcriptional terminatorsequence are introduced into suitableplant material. The gene of interest can betagged with a marker, frequently a fluorescentdye, to enable its location in the transformedplant.4. Hybrid Arrest and Hybrid SelectionProceduresHybrid arrest and hybrid selection procedurescan be used to demonstrate that a particularcDNA clone contains the gene for aparticular protein. In hybrid arrest, the clonedcDNA is hybridized to mRNAs, and themRNAs are translated in an in vitro system.The hybrid will not be translated. Identificationof the missing polypeptide defines the gene onthe cDNA.In the hybrid select procedure, the cDNAmRNAhybrid is isolated and dissociated. ThemRNA is translated in vitro to define theencoded protein. In appropriate circumstances,these procedures can be used to identify genefunction—for example, identifying the proteasegene in Tobacco vein mottling virus.II. WHAT IS A VIRUS MADE OF?

116 6. PLANT VIRAL GENOMES5. Sequence Comparison with Genesof Known FunctionAs noted earlier in this chapter, sequencecomparisons can be used to obtain evidencethat a particular ORF may be functional. Thesame information may also give strong indicationsas to actual function. For example, usingsequence information amino acid sequencesimilarities were found between the gene foran RNA-dependent RNA polymerase (RdRp)in poliovirus and proteins coded for by severalplant viruses. This similarity implied quitestrongly that these plant virus-coded proteinsalso have a polymerase function. The conservedamino acid sequences (motifs) of RdRpsand many other viral gene products aredescribed at the appropriate places in this book(e.g., for RdRps see Box 8.2).6. Functional Regions Within a GeneSpontaneous mutations and deletions can beused to identify important functional regionswithin a gene. However, mutants obtained bysite-directed mutagenesis and deletions constructedin vitro can give similar informationin a more systematic and controlled manner.For example, the construction and transcriptionof cDNA representing various portions of theTobacco etch virus genome, translation in vitro,and testing of the polypeptide productsshowed that the proteolytic activity of the 49kDa viral proteinase lies in the 3 0 -terminalregion. The amino acid sequence in this regionsuggested that it is a thiol protease related inmechanism to papain.However, care must be taken with thisapproach. Many functions depend on thethree-dimensional structure of the protein andmutations, not at the active site, may have asecondary effect on the protein structure.IV. VIRUSES OF OTHERKINGDOMSMost of the points made in this chapter applyto viruses of other kingdoms. Differences, suchas gene products that control the initiation ofinfection, have been explained in the text.V. SUMMARY• Viruses are very efficient in the use of thelimited amount of genomic nucleic acid thatthey possess.• Viral genomes contain coding sequences andsequences that control the expression of theviral genome.• Viral genomes encode proteins required forsuccessful infection, including proteinsinitiating infection (not most plant viruses)and proteins that replicate the viral genome,process viral gene products, facilitatemovement through the host, overcome hostdefence systems, and facilitate movementbetween hosts.• The 5 0 and 3 0 regions have structures that areimportant for genome expression andreplication.Further ReadingHull, R. (2002). Matthews’ plant virology. Academic Press,San Diego.Chung, B.Y., Miller, W.A., Atkins, J.F., and Firth, A.E.(2008). An overlapping essential gene in the Potyviridae.Proc. Natl. Acad. Sci. USA 105, 5897–5902.II. WHAT IS A VIRUS MADE OF?

C H A P T E R7Expression of Viral GenomesViral genomes are expressed via mRNAs. Eukaryotic cells pose constraints on how the informationon an mRNA is expressed. Viruses have a variety of ways of overcoming theseconstraints.O U T L I N EI. Stages in Virus Infection Cycle 117II. Virus Entry and Uncoating 119III. Initial Translation of ViralGenome 122IV. Synthesis of mRNAs 123V. Plant Viral Genome Strategies 125VI. Viruses of Other Kingdoms 137VII. Summary 137I. STAGES IN VIRUSINFECTION CYCLEThe infection cycle of a virus, be it of a plant,vertebrate, invertebrate, or bacterium, hasseven stages (see Figure 7.1):1. Virus initial entry into the cell (discussed inChapter 12).2. Genome uncoating (discussed in thischapter).3. Production of mRNAs. As willbe shown in this chapter, the routeused for the production of mRNAsdepends on the nature of the viralgenome.4. Translation of the viral geneticinformation from the mRNAs. Someof this information is expressed earlyand some late in the infection cycle,depending on when the product isrequired.Comparative Plant Virology, Second Edition117Copyright # 2009, Elsevier Inc. All rights reserved.

118 7. EXPRESSION OF VIRAL GENOMESFIGURE 7.1 The seven stages of avirus infection cycle. The square boxrepresents the plant cell wall and the circlewithin it, the plasma-membrane.II. WHAT IS A VIRUS MADE OF?

II. VIRUS ENTRY AND UNCOATING1195. Replication of the viral genome usingat least some of the factors expressedfrom the viral genome (discussed inChapter 8).6. Assembly of the progeny virions (discussedin Chapter 5).7. Release of virus from initially infected celland infection of adjacent cells (discussed inChapter 9).However, we must bear in mind that theseseven stages are not completely separate andsequential and that they are closely integratedand coordinated.II. VIRUS ENTRY ANDUNCOATINGA. Virus EntryAs we will see in Chapter 12, plant virusesrequire damage to the cuticle and cell wallto be able to enter a plant cell. There is noevidence for a specific entry mechanism suchas plasma membrane receptor sites or endocytoticuptake as with viruses of vertebratesand invertebrates, and it is generally consideredthat entry is accomplished by “bruteforce”.B. UncoatingOnce in the initially infected cell, the virusparticle has to be uncoated to release thegenomic nucleic acid. There is a dichotomyin the structural stabilisation of viruses in thatthe particles have to be stable enough to protectthe viral genome when being transportedoutside the host but must be able to presentthe genome to the cellular milieu for the firststages in replication. The uncoating processdepends on the chemical bonds that stabilisethe virus particles (see Chapter 5). This hasbeen studied for the rod-shaped Tobaccomosaic virus (TMV) and for several isometricviruses.1. Uncoating of TMVUsing TMV radioactively labeled in the proteinor the RNA or in both components, thefollowing has been observed:1. Within a few minutes of inoculation, about10 percent of the RNA may be released fromthe virus retained on the leaf.2. Much of the RNA is degraded, but some arestill full-length RNA molecules.3. The early stages of uncoating of the particlesdo not appear to depend on preexisting orinduced enzymes.4. The process is not host specific, at least inthe early stages.However, there is a fundamental difficulty withall such experiments. Concentrated inoculamust be used to provide sufficient virus foranalysis, but this means that large numbers ofvirus particles enter cells rapidly. Thus, it isimpossible to know which among these particlesactually establish an infection.To initiate infection, TMV RNA must beuncoated, at least to the extent of allowing thefirst ORF to be translated. By various in vitroexperiments, it was shown that TMV isuncoated by a process termed cotranslationaldisassembly (Box 7.1).2. Uncoating of Brome Mosaic Virus andSouthern Bean Mosaic VirusThe isometric particles of Brome mosaic virus(BMV) and Southern bean mosaic virus (SBMV)are stabilised by carboxyl-carboxylate bondsand by protein-RNA interactions; SBMV particlesare also stabilised by Ca 2þ (see Chapter 5).Above pH 7, the carboxyl-carboxylate bondsprotonate, and after the removal of Ca 2þ fromSBMV, the particles of both viruses swell, beingjust stabilised by protein-RNA interactions.These swollen particles can be uncoated byII. WHAT IS A VIRUS MADE OF?

120 7. EXPRESSION OF VIRAL GENOMESBOX 7.1COTRANSLATIONAL DISASSEMBLY OF TMVThe first event in cotranslational disassembly of TMV is that the structure of the virion relaxes so the5 0 terminus of the RNA is accessible to a ribosome. The 68 nucleotide 5 0 leader sequence, which lacksG residues, interacts more weakly with coat protein subunits than do other regions of the genome.As discussed in Chapter 5, TMV particles are stabilised by carboxylate interactions, which becomeprotonated at slightly alkaline pHs. This allows a ribosome to attach to the 5 0 leader sequenceand then to move down the RNA, displacing coat protein subunits as it moves. The ribosomepartially-stripped-rodcomplexes are termed striposomes (Figure A), which have been found bothin vivo and in vitro.Fig.A. Electron micrographs of striposome complexesshowing ribosomes clustered around oneend (the 5 0 end) of TMV rods. [This article waspublished in Virology, 137, T.M. Wilson, Cotranslationaldisassembly of tobacco mosaic virus in vitro,pp. 255–265, Copyright Elsevier (1984).]Having initiated translation, ribosomes proceed along TMV RNA, translating the 5 0 ORF and the126/183 kDa replicase protein (see Profile 14 in the Appendix for TMV genome map) and displacing(continued)II. WHAT IS A VIRUS MADE OF?

II. VIRUS ENTRY AND UNCOATING121BOX 7.1(continued)coat protein subunits. When the ribosomes reach the stop codon of the 126/183 kDa ORF, they disengage.This raises the question of how the 3 0 quarter of the particle is disassembled. It is consideredthat the replicase performs this task in a 3 0 !5 0 direction in synthesising the (–)-strand replication intermediatein a process termed coreplicational disassembly from the 3 0 end.Thus, TMV is uncoated in a bidirectional manner, using the cotranslational mechanism for the5 0 !3 0 direction, yielding the replicase that disassembles the rest of the particle in the 3 0 !5 0 direction,showing that disassembly and replication are coupled processes. The process happens rapidly withthe whole capsid uncoated within about 30 minutes (Figure B).Fig.B. Time scale of bidirectionaldisassembly and assemblyof TMV particle invivo. [From Wu and Shaw(1996), Proc. Natl. Acad. Sci.USA, 93, 2981–2984, Copyright(1996) National Academyof Sciences, U.S.A.]Bcotranslational disassembly in vitro translation.However, mutants of another bromovirus thatdid not swell under alkaline conditions showedthat swelling was not necessarily required forcotranslational disassembly, leading to the suggestionthat there is a pH-dependent structuraltransition in the virion, other than swelling,which enables the RNA to be accessible to thetranslation system. The proposed model, whichis similar to those of some vertebrate and insectviruses, postulates that the N termini of the fivesubunits in the pentameric capsomere undergoa major structural transition from the interiorto the exterior of the virion. This provides a channelthrough which the RNA passes to be accessiblefor translation. However, the 5 0 end ofthe RNA must be released, which suggests thatit is located in association with a pentamericcapsomere.3. Uncoating of Turnip YellowMosaic VirusThe isometric particles of Turnip yellowmosaic virus (TYMV) do not cotranslationallydisassemble in the in vitro translation systemII. WHAT IS A VIRUS MADE OF?

122 7. EXPRESSION OF VIRAL GENOMESjust described for BMV and SBMV. In vitrostudies show that under various nonphysiologicalconditions, such as pH 11.5 or freezingand thawing, the RNA can escape from TYMVparticles without disintegration of the proteinshell. When Chinese cabbage leaves are inoculatedwith TYMV, a significant proportion ofthe inoculum is uncoated after two minutes,with empty virus particles and low-MW proteinbeing formed following RNA release. Atleast 80 to 90 percent of this uncoating takesplace in the epidermis. This uncoating processis not confined to known hosts of TYMV.4. Uncoating Other Plant VirusesThe requirements in the first stages of infectionwith other types of genomes are differentto those of the (þ)-sense ssRNA viruses.Viruses with dsRNA or (–)-sense ssRNA haveto transcribe their genome to give mRNA.These viruses carry the viral RNA-dependentRNA polymerase (RdRp) in the virus particleand, presumably, transcription is an earlyevent. It is not known if this occurs within thevirus particle, possibly in a relaxed structure,or if the viral genome is released into the cell.However, it is most likely that this processtakes place in an environment that is protectedfrom cellular nucleases and that it is coupled totranslation of the mRNA.The dsDNA genomes of members of theCaulimoviridae must be transported to thenucleus, where they are transcribed to mRNAby the host RNA-dependent RNA polymerase(see later in this chapter). The coat protein ofCauliflower mosaic virus (CaMV) has a nuclearlocalisation signal that will presumably targetthe particle into the nucleus. Particles of somecaulimoviruses and badnaviruses are particularlystable, capable of resisting phenol, andnothing is known about how they disassemble.The ssDNA genomes of members of theGeminiviridae also have to be transported tothe nucleus so they can be replicated beforebeing transcribed to give mRNAs. Nuclearlocalisation signals have been recognised insome geminiviral proteins, but nothing isknown about how the particles uncoat.III. INITIAL TRANSLATION OFVIRAL GENOMEViral genomes are expressed from mRNAsthat are either the nucleic acid (þ)-sense ssRNAviruses or transcripts from the (–)-sense ordsRNA, or from ds or ss DNA viruses. Baltimore(1971) pointed out that the expression ofall viral genomes, be they RNA or DNA, ss ords, (þ)- or ( )-sense, converges on the mRNAstage (Figure 7.2). As we will see later in thischapter, expression of the viral mRNA facesvarious constraints imposed by the eukaryotictranslation system.FIGURE 7.2 Routing of viral genome expressionthrough mRNA. Route I is transcription of dsDNA usuallyby the host DNA-dependent RNA polymerase; route II istranscription of ssDNA to give a dsDNA template for I(e.g., geminiviruses); route III is transcription of dsRNA,usually by virus-coded RdRp (e.g., reoviruses); route IV isreplication of (þ)-strand RNA via a (–)-strand template byvirus-coded RdRp—the viral (þ) strand is often the templatefor early translation (the (þ)-strand RNA viruses);route V is transcription of (–)-strand RNA viral genomeby virus-coded RdRp (e.g., tospoviruses); route VI isreverse transcription of the RNA stage of retro- andpararetroviruses leading to a dsDNA template for mRNAtranscription (for pararetroviruses the input viral dsDNAcan be the template. [From Baltimore (1971).]II. WHAT IS A VIRUS MADE OF?

IV. SYNTHESIS OF MRNAS123IV. SYNTHESIS OF MRNASThe (þ)-sense ssRNAs of many genera ofplant viruses can act as mRNAs directly onentry into the host cell. For viruses with othertypes of genome, mRNAs have to be synthesisedat some stage of the infection cycle.A. Negative-Sense Single-StrandedRNA VirusesAll viruses with a (–)-sense ssRNA genomecarry the viral RdRp in their virus particles.Thus, one of the early events on entry into ahost cell is the transcription of the viral genometo (þ)-sense RNA required for both translationof the viral genetic information and as an intermediatefor replication.Plant rhabdoviruses, like those infecting vertebrates,possess a genome consisting of a singlepiece of (–)-sense ssRNA, with a length inthe range of 11 to 13 kb and encoding six proteins,one more than animal rhabdoviruses(for genome organisation of plant rhabdoviruses,see Profile 13 in the Appendix). Theplant rhabdoviruses appear to be expressed ina manner similar to animal rhabdoviruses suchas Vesicular stomatitis virus (VSV), which hasbeen studied much more extensively. ForVSV, the active transcribing complex consistsof the RNA genome tightly associated withthe N protein, and the polymerase made up ofthe phospho-protein (P) and the large (L) protein.This complex starts transcribing (þ)-senseRNA at a single entry site at the 3 0 end of thegenome and transcribes the leader RNA thatis transported to the nucleus, where it inhibitshost cell transcription. The complex then transcribesthe mRNA for the N protein, which iscapped during synthesis by the polymerase.At the end of the N gene, and of all genes, isthe sequence 5 0 -AGUUUUUUU-3 0 (element I),which signals termination and polyadenylationof the mRNA. This intergenic sequence alsocomprises a short untranscribed sequence (elementII) and the start site for transcription ofthe next mRNA (element III). Similar sequencesare found in plant rhabdoviruses.Thus, the viral genes are transcribed separatelyfrom the 3 0 end and are transcribed indecreasing amounts (N > P > sc4 > M > G > L).This is an efficient way of regulating geneexpression, as the genes that are located at the3 0 end are those that are required in greatestamounts.The genomes of tospoviruses comprise threessRNA segments (see Profile 17). L RNA is (–)-sense and monocistronic encoding the viralRdRp. The mRNA is transcribed from the virionRNA by the virion-associated polymerase. Theother two RNAs have an ambisense genearrangement with one ORF in the viral strandand one in the complementary strand. The twoORFs are separated by an AU-rich intergenicregion of variable length. For both RNAs thevirion-sense ORF is expressed from a subgenomic(sg) RNA transcribed from the complementaryRNA and the complementary-sense ORF froman sgRNA transcribed from the virion RNA.The intergenic region between the ambisenseORFs is predicted to form stable hairpin structuresthat are suggested to control the terminationof transcription of sgRNAs. However, asnoted in the following section, this should beconsidered with circumspection. The formationof tospovirus mRNAs involves cap snatching(see later in this chapter).B. Double-Stranded RNA VirusesPlant members of the Reoviridae family areplaced in three genera: Phytoreovirus with 12dsRNA genome segments and Fijivirus andOryzavirus each with 10 dsRNA genome segments.The genome organisation of the OryzavirusRice ragged stunt virus is described inProfile 11. Most of the dsRNA segments aremonocistronic, but segment 4 is bicistronicand segment 8 has three ORFs.II. WHAT IS A VIRUS MADE OF?

124 7. EXPRESSION OF VIRAL GENOMESThe plant reoviruses, like their counterpartsthat infect vertebrates and insects, contain atranscriptase that uses the RNA in the particleas template to produce ssRNA copies. In animalreoviruses, this occurs in subviral particlescomprising part of the capsid, the polymerase,and the dsRNAs. Early in infection only(þ)-sense ssRNAs are synthesised, which act asmRNAs. Later, (–)-sense strands are synthesisedleading to viral replication (see Chapter 8). It islikely that a similar series of events occurs in theplant reoviruses, especially when they multiplyin their insect vectors.C. DNA VirusesThe synthesis of mRNAs from either thedsDNA members of the Caulimoviridae or thessDNA members of the Geminiviridae and thenanoviruses does not involve a virus-codedenzyme but is performed by the host DNAdependentRNA polymerase II located in thenucleus. This synthesis is initiated by a viralpromoter.1. CaulimoviridaeThe genome organisations of the Caulimoviridaegenera are described in Profile 4. Most ofthe detailed studies have been performed onCauliflower mosaic virus (CaMV).As we will see in Chapter 8, there are twophases in the nucleic acid replication cycle ofCaMV: the nuclear phase of transcription andthe cytoplasmic phase of gene expression andreverse transcription. In the first, the dsDNA ofthe infecting particle moves to the cell nucleus,where the overlapping nucleotides at the gapsare removed, and the gaps are covalently closedto form a fully dsDNA. These minichromosomesform the template used by the host DNA-dependentRNA polymerase to transcribe two RNAs of35S and 19S, as indicated in Profile 4. As wellas promoters for these two mRNAs, the viralDNA also has signals for the polyadenylatedtermination of transcription. The 35S promotercomprises the core promoter upstream of thetranscription start site and various control elementsboth upstream and downstream of thestart site (Box 7.2).Other caulimovirus promoters also show similarmodular structures. The promoters of someviruses require sequences downstream of thetranscription start site for maximum expression.An example of this is the Rice tungro bacilliformvirus (RTBV) promoter, which requires anenhancer located in the first 90 nucleotides ofthe transcript for efficient transcription. Thereare two subelements in this enhancer region,one being position and orientation independentand the other being position dependent.Some promoters are specific to the vasculartissue. In that of RTBV the region between 164and 100 nucleotides upstream of the transcriptionstart site is essential for vascular tissueexpression, and deletion leads to specificityin the epidermis. This tissue specificity is notsurprising for RTBV, as the virus itself isphloem-limited.Most of the caulimovirus promoters act inboth monocot and dicot plant species eventhough the parent virus is restricted in hostrange. The CaMV 35S promoter has been usedfor the expression of transgenes in many dicotand monocot plant species and is considereda good, strong constitutive promoter. It hasalso been shown to be active in bacteria, yeast,animal HeLa cells, and Xenopus oocytes.The caulimovirus 19S promoter has beenmuch less studied. Only that of CaMV has beenanalysed and shown to be weaker than the 35Spromoter when tested in transgenic constructs.This is in contrast to virus infections leadingto comparable levels of the 35S and 19S RNAsand the product of the gene encoded by the19S RNA, the ORFVI product, being the mostabundant viral protein. The core 19S promotercan be strongly activated by the 35S promoterenhancer elements, but no enhancer elementshave been detected for the promoter itself.The caulimovirus 35S and 19S RNAs are 3 0coterminal and share a polyadenylation signal.The signal motif, AAUAAA, is found upstreamII. WHAT IS A VIRUS MADE OF?

V. PLANT VIRAL GENOME STRATEGIES125BOX 7.2THE CAULIFLOWER MOSAIC VIRUS 35S PROMOTERThe 35S RNA is the major transcript from CaMV(see Profile 4). Its expression is controlled by a promoterthat is a sequence of the viral genomic DNA.The core promoter is characterised by what istermed a “TATA box,” about 25 nucleotidesupstream of the transcription start site. A detailedanalysis of the CaMV promoter revealedthat it has a modular nature with subdomainsconferring patterns of tissue-specific expression.The two major domains are domain A (–90 toþ8; numbering relative to transcription start siteat þ1), which is important for root-specificexpression, and domain B (–343 to –90), whichis mainly involved in expression in the aerialparts of the plant. The region of the A domainbetween –83 and –63 contains an as-1 (activationsequence-1)-like element that is important forthe root-specific expression. The as-1 element ispresent in several nonviral promoters and canbe recognised in many of the caulimoviruspromoters, where it is important. The B domaincomprises five subdomains, B1 to B5, eachconferring specific expressions patterns indeveloping and mature leaves. A 60-nucleotideregion downstream of the transcription start sitealso enhances gene expression. Thus, in the fullpromoter these domains and subdomains actcoordinately and synergistically to give theconstitutive expression of the CaMV 35S promoter.Plant nuclear factors have been identifiedthat bind to various regulatory regions in thispromoter.of the transcription termination or cleavage sitebut downstream of the transcription initiationsite.2. GeminiviridaeThe circular ssDNA genomes of membersof the Geminiviridae have ORFs both in thevirion-sense and complementary-sense orientations(see Profile 7). All geminiviruses employthe same basic strategy to transcribe their genomesin that it is bidirectional from the longintergenic region and terminating diametricallyopposite in the short intergenic region. However,there are differences between the generain the details of transcription.The long intergenic region (which also containsthe common region involved in replication)is about 300 nucleotides. Each of the virionsenseand complementary-sense transcripts hasa characteristic eukaryotic RNA polymerase IIpromoter sequence, with transcription beinginitiated 20–30 bp downstream of the TATAbox motifs. For at least some of the promoters,there are upstream sequences that enhancepromoter activity. The C1 promoter for the Repprotein (see Chapter 8 for Rep protein) lies inthe common region and overlaps the origin of(þ)-strand DNA synthesis; this illustrates theclose interactions between transcription and replication.The short AT-rich intergenic region containsthe polyadenylation sites for the virion- andcomplementary-sense RNAs, which overlap sothat they share a few 3 0 nucleotides.V. PLANT VIRAL GENOMESTRATEGIESA. The Eukaryotic Translation SystemConstraintsIt is generally accepted that the eukaryoticprotein-synthesising system translates the informationfrom viral mRNAs. This translationII. WHAT IS A VIRUS MADE OF?

126 7. EXPRESSION OF VIRAL GENOMESsystem has various features and controls all ofthe following:• Plant cellular mRNAs have a cap—aninverted and methylated GTP at the 5 0terminus [m 7 G(5 0 )ppp(5 0 )N] and a poly(A)tail at the 3 0 terminus.• In most circumstances, mRNAs contain asingle open reading frame (ORF).• Translation is initiated at an AUG startcodon, the context of which controls theefficiency of initiation.• The cap, 5 0 untranslated region, the codingregion, the 3 0 untranslated region, and thepoly(A) tail all have potential to influencetranslational efficiency and mRNA stability.However, the feature of the single ORF (bullettwo above) presents major difficulties forviruses (Box 7.3).B. Virus Strategies to OvercomeEukaryotic Translation ConstraintsOn current knowledge, there are at least 12strategies by which RNA viral genomes andtranscripts from DNA viruses ensure that alltheir genes are accessible to the eukaryotic protein-synthesisingsystem and overcome theproblem outlined in Box 7.3. The strategies fallinto three groups (Figure 7.3):• Making the viral genomic RNA or segmentthereof effectively monocistronic by bringingany downstream AUG to the 5 0 end. This isdone by either having a single ORFexpressing a polyprotein that issubsequently cleaved to give thefunctional proteins (strategy 1) or bydividing up the viral genome to givemonocistronic RNAs either duringexpression (strategies 2, 4, and 5) orpermanently (strategy 3);• Avoiding the constraints of the 5 0 AUG.This can be done in several ways(strategies 6 to 10).• Maximising the information expressed from aviral RNA by bringing together two adjacentORFs to give two proteins, one from the 5 0 ORFand the other from both ORFs. Thus, thesecond protein comprises the upstreamprotein in its N-terminal region, theC-terminal region being from thedownstream ORF (strategies 11 and 12).1. Strategy 1. PolyproteinsIn this strategy the coding capacity of theRNA for more than one protein, and sometimesfor the whole genome, is translated from a singleORF. The polyprotein is then cleaved atspecific sites by one or more virus-codedBOX 7.3VIRUS M RNA TRANSLATION PROBLEMIn the scanning model for translation, the 40Sribosomal subunit binds to the 5 0 cap and translocatesto the first AUG in a suitable context,where it forms the 80S ribosome that translatesonly that ORF immediately downstream fromthe 5 0 region of an mRNA; at the stop codon ofthis ORF, the ribosomes dissociate. Thus, ORFsbeyond this point normally remain untranslated.Viral genomes, except those of satellite viruses,encode two or more proteins and therefore arepresented with a problem of how to expresstheir downstream proteins in the eukaryoticsystem. Much of the variation in the way geneproducts are translated from viral RNA genomesappears to have evolved to meet thisconstraint.II. WHAT IS A VIRUS MADE OF?

V. PLANT VIRAL GENOME STRATEGIES127FIGURE 7.3 Diagram illustratingthe 12 strategies that viruses havefor overcoming the constraints of theeukaryotic translation system. A. Diagrammaticviral genome with threeopen reading frames; B. The 12 strategiesdescribed in the text.ABproteinases (see Box 4.3) to give the final geneproducts. Some viruses use one type of proteinase,others two or three.The use of the polyprotein strategy is exemplifiedby the potyviruses, which have genomesof approximately 10 kb that contain asingle ORF for a polyprotein of about 3,000 to3,300 amino acids (see Profile 10). The polyproteinis cleaved to give 10 proteins (Figure 7.4),using three virus-coded proteases.The 35 kDa P1 is a serine protease thatcleaves itself from the polyprotein at Phe-Ser.The 52 kDa HC-Pro is a papain-like cysteineprotease that cleaves at its C-terminal Gly-Gly.II. WHAT IS A VIRUS MADE OF?

128 7. EXPRESSION OF VIRAL GENOMESFIGURE 7.4 Schematic for theprocessing of the potyvirus polyprotein.The top part shows the potyvirusgenome with 5 0 VPg and 3 0poly(A) sequence and the gene productswithin the polyprotein. Belowthis is shown the cascade of processingof the polyprotein. The primaryevents are probably cotranslationaland autocatalytic, yielding precursorsand mature products. For geneproducts, see Profile 10. [FromRiechmann et al. (Journal of GeneralVirology 73, 1–16, 1992).]The 27 kDa protease domain of the NIa regionis a serine protease responsible at most, ifnot all, for the other cleavages at Gln-(Ser/Gly). Thus, the P1 and HC-Pro proteases actautocatalytically and the NIa protease acts bothtrans- and autocatalytically. The cleavage eventstake place at different rates yielding intermediateproducts that have different properties to thefinal products.There are both advantages and disadvantagesto the polyprotein processing strategy.Apart from overcoming the non-5 0 start codonproblem, there are advantages both in the factthat several functional proteins are producedfrom a minimum of genetic information andalso in the potential for regulating the processingpathway. This is shown in the differencesin the cleavage sites for the NIa proteaseand the influence that the surrounding residuescan have on the rates of cleavage. Similarly,the requirement for the processing at theC-terminus of the HC-Pro before P1 is cleavedfrom the product most probably represents acontrol mechanism.The major disadvantage is that it is difficultto visualise how the polyprotein strategyof the potyviruses can be efficient. The coatprotein gene is at the 3 0 end of the genome(see Figure 7.4). Thus, for every molecule ofthe coat protein produced, a molecule of allthe other gene products must be made. Sinceabout 2,000 molecules of coat protein areneeded to encapsidate each virus particle butprobably only one replicase molecule to produceit, this appears to be a very inefficient procedure.Indeed, large quantities of several geneproducts, apparently in a nonfunctional state,accumulate in infected cells (see Box 2.4). Nevertheless,the potyviruses are a very successfulgroup. There are many member viruses, andthey infect a wide range of host plants. Otherviruses using polyprotein processing haveadditional devices that can avoid this problem.Comoviruses have their two coat proteins on aseparate genome segment (see Profile 6). Theredoes not appear to be any massive accumulationof noncoat gene products in cells infectedwith these viruses.II. WHAT IS A VIRUS MADE OF?

V. PLANT VIRAL GENOME STRATEGIES1292. Strategy 2. Subgenomic RNAsSubgenomic RNAs (sgRNA) are synthesisedduring viral replication from a genomic RNA thatcontains more than one ORF, giving 5 0 -truncated,3 0 coterminal versions of the genome. This thenplaces the ORFs that were originally downstreamat the 5 0 end of the mRNA, as shown in theexpression of TMV (Figure 7.5A).FIGURE 7.5 Subgenomic RNAs.A. sgRNAs in the expression ofTMV. The top line shows the genomeorganisation of TMV (see Profile 14);below is the expression of the fourORFs, the 126K protein being translateddirectly from the genomic RNAand read through to give the 183Kprotein; the 32k movement proteinORF3 and the 17K coat protein ORF4are translated from sgRNAs. [FromHull (2002).] B. Expression of CTV.The top line is the genome organisation(see Profile 5). Below are theways that the gene products areexpressed by translation of the 5 0ORF directly from the genomic RNA,frameshift with the 5 0 ORF, proteolyticprocessing of these productsand expression of ORFs 2–11 fromsgRNAs. [From Karasev and Hilf(1997; in Filamentous viruses of woodyplants (P.L. Monette, Ed.), pp. 121–131, Research Signpost, Trivandrum).]ABII. WHAT IS A VIRUS MADE OF?

130 7. EXPRESSION OF VIRAL GENOMESWhen several genes are present at the 3 0 endof the genomic RNA, a family of 3 0 collinearsgRNAs may be produced; Citrus tristeza virus(CTV) has a nested set of at least 9 sgRNAs (seeFigure 7.5B). sgRNAs may be encapsidated (e.g.,BMV; see Profile 3) and can cause uncertainty asto what comprises an infectious genome.At least four models have been proposed forthe synthesis of sgRNA from the genomic RNA(Box 7.4). The first and second mechanisms havebeen proposed for plant viruses.The use of sgRNAs is widespread in plantviruses as a strategy to obviate the limitationsof eukaryotic translation. The synthesis of theseBOX 7.4SUBGENOMIC RNASAND THEIR PROMOTERSA subgenomic RNA (sgRNA) is a 5 0 truncatedversion of the genomic RNA that places anORF that was originally downstream of the 5 0ORF at the 5 0 end (see Figure 7.2B, panel 2). Thisenables the downstream ORF to be translated.sgRNAs are produced during the replication ofthe genomic RNA, and at least four models havebeen suggested for their synthesis:1. De novo internal initiation on the full-length(–)-strand of the genome during (þ)-strandsynthesis.2. Premature termination during (–)-strandsynthesis of the genome, followed by the useof the truncated nascent RNA as a templatefor sgRNA synthesis.3. Initiation on the full-length (–) strand primedby a short leader from the 5 0 end of thegenomic DNA during (þ)-strand synthesis.This has been found for coronaviruses.4. Intramolecular recombination during (–)-strand synthesis, in which the replicasejumps from the subgenomic RNA start siteon the full-length (þ) strand and reinitiatesnear the 5 0 end of the genome. This also hasbeen found for coronaviruses.The simplest model for de novo internal initiationof sgRNAs necessitates the replicase recognisinga sequence upstream of the sgRNA 5 0end. This is termed the subgenomic promoter.Brome mosaic virus (BMV) has a tripartitegenome (Profile 3). The coat protein ORF isdownstream on the bicistronic RNA-3 and isexpressed from sgRNA-4. A subgenomic promoterfor RNA-4 is in an intergenic regionupstream of its 5 0 end. The subgenomic promoterof BMV comprises two major parts. A “core” promoterof 20 nucleotides (–20 to þ1) upstream ofthe subgenomic initiation site (the subgenomicinitiation site is designated as þ 1 and – numbersare 5 0 of that site as read on the positive-strandRNA) is the smallest region capable of promotingsgRNA synthesis with low accuracy at a basallevel; the core promoter is predicted to form ahairpin structure. Three “enhancer” regions, onedownstream of the start site and two upstream,provide accuracy of replication initiation andcontrol yields of sgRNAs. The fully functionalsubgenomic promoter encompasses about 150nucleotides.Premature termination of (–)-strand synthesisis effected by either cis or trans long-distanceinteractions between a region just upstream ofthe subgenomic promoter and another regionof the viral nucleic acid. This can either be onthe same nucleic acid molecule as that givingrise to the sgRNA (cis interaction) as exemplifiedby Tomato bushy stunt virus or on another genomicfragment of a split genome virus (trans interaction;e.g., Red clover necrotic mosaic virus).II. WHAT IS A VIRUS MADE OF?

V. PLANT VIRAL GENOME STRATEGIES131RNAs involves close interlinks with viral replicationand having strong controlling systems.The subgenomic promoter may having elementsin both intergenic and coding regions,the latter suggesting that the position may controlexpression of the promoter.3. Strategy 3. Multipartite GenomesViruses with multipartite genomes have theinformation required for the virus infectioncycle divided between two or more nucleic acidsegments. This is found for both DNA andRNA plant viruses. For the (þ)-sense ssRNAviruses, this strategy places the gene at the 5 0end of each RNA segment, and thus it is opento translation (e.g., RNAs 1 and 2 of BMV; seeProfile 3).Of the 80 plant virus genera, 33 have multipartitegenomes. In most of these, the genomesegments are encapsidated in separate particles,such viruses being termed multicomponent.(see Box 5.1). Members of the Reoviridae,and possibly the Partitiviridae, have all theirmultipartite genome segments in one particle.4. Strategy 4. SplicingThe production of mRNAs from DNA ineukaryotes involves splicing, which removesinternal noncoding sequences and can givevarious versions of an mRNA. Two of thefamilies of plant viruses with DNA genomes,the Caulimoviridae and Geminiviridae, use splicingin the production of mRNAs, a process that,at least in the caulimoviruses, opens up downstreamORFs.ORF4 of Rice tungro bacilliform virus (RTBV)is expressed from an mRNA spliced from the35S RNA (see Profile 4). The splice removesan intron of about 6.3 kb and brings a shortORF (sORF) into frame with ORF4. The splicedonor and acceptor sequences correspond toplant splice consensus sequences.The circular ssDNA genome of Maize streakvirus has four ORFs, two being expressed fromtranscripts, V1 and V2, in the virion sense, andtwo, C1 and C2, from transcripts in the complementarysense (see Profile 7 for genomeorganisation). The C transcripts are of lowabundance, and a splicing event fuses ORFC1 to C2.5. Strategy 5. Translation for BothViral and Complementary Strands(Ambisense)Some of the genome segments of the tospovirusesand tenuiviruses encode two proteins,with one ORF in the virion sense and the otherin the complementary sense (see Rice stripevirus and Tomato spotted wilt virus (TSWV) inProfiles 12 and 17, respectively). Thus, one ofthe proteins is expressed from complementary-senseRNA. This, called the ambisenseexpression strategy, was discussed previouslyin this chapter and is another means by whichviruses overcome the eukaryotic translationconstraints.6. Strategy 6. Internal InitiationAs we noted previously, most eukaryoticmRNAs are 5 0 -capped and translation starts atthe 5 0 end. However, in some cases, the 5 0 nontranslatedregions or other regions of uncappedviral (and host) mRNAs contain internal ribosomeentry sites (IRES), allowing eukaryoticribosomes to be loaded onto the mRNA substantiallydownstream of the 5 0 end. It is suggestedthat the IRES forms a complex secondary/tertiary structure to which ribosomes andtransacting factors bind. Although this internalinitiation strategy was first demonstrated foranimal viruses, an increasing number of plantviruses (e.g., a crucifer-infecting tobamovirus,a luteovirus, and a nepovirus) are also beingshown to employ it.It is considered that the IRES strategyenables a potentially inefficient mRNA (nocap and long 5 0 UTR) to be translated efficientlyand might also provide translational controlso gene products can be expressed at theappropriate time.II. WHAT IS A VIRUS MADE OF?

132 7. EXPRESSION OF VIRAL GENOMESBOX 7.5EXPRESSION OF RICE TUNGRO BACILLIFORMVIRUS (RTBV)The genome of RTBV contains four ORFs (see Profile 4). The dsDNA genome is transcribed to give a35S RNA, which is spliced to form the mRNA for ORF IV. Thus, the expression of ORFs I–III facesthe problems of translation in a eukaryotic system. RTBV has a long leader sequence (more than 600nucleotides) that contains 12 short ORFs. Long ORF I has an AUU start codon, and the next approximately1,000 nucleotides have only two AUG codons in any reading frame: those for ORFs II and III.ORFs I, II, and III each overlap the next by one nucleotide having a “stop/start” signal of AUGA.This has led to the development of the following model for the expression of RTBV ORFs I–III. (Fig.)Fig. Expression of the firstthree ORFs of RTBV. Topline shows genome organisationof first three ORFs.Lower panel shows translationshunt with much pf theleader sequence folded into ahairpin followed by the non-AUG start codon for ORF Iand “weak” start for ORF II.The translation process isdescribed below. [From Hull(2002).]The long leader sequence of RTBV is bypassed by a ribosome shunt mechanism by forming anelongate hairpin conformation (Fig.). In the shunt process the ribosome enters at the 5 0 cap, which isessential; translocates to sORF A; and shunts across the stable hairpin from the shunt “donor” sitejust downstream of sORF A to the shunt “landing” site just upstream of ORF I. This places the40S ribosome subunits at the AUU start codon of ORF I, a start codon that is about 10 percent as efficientas an AUG codon. It is suggested that only some of the 40S ribosome subunits associate with60S ribosomes and initiate translation at the ORF I start codon, the rest translocating to the nextAUG, which is the start codon for ORF II but is in a poor context. The model suggests that only someof these 40S ribosome subunits initiate here, the remainder passing to the next AUG, which is at thestart of ORF III and in a good context.II. WHAT IS A VIRUS MADE OF?

V. PLANT VIRAL GENOME STRATEGIES1337. Strategy 7. Leaky ScanningLeaky scanning is where the 40S ribosomalsubunits start scanning from the 5 0 end of theRNA but do not all start translation at the firstAUG. Some, or all, pass the first AUG and starttranslation at downstream ORFs. In some cases,the 40S subunits of the 80S ribosomes fail todisengage at a stop codon, and they reinitiateat a downstream start codon. There are threeforms of leaky scanning:a. Two Initiation Sites on One ORF (TwoStart). The genome of Cowpea mosaic virus(CPMV) is divided between two RNA species,the shorter of which, M-RNA, codes for twoC-terminal collinear polyproteins of 105 and95 kDa initiated from two in-frame AUGcodons (see Profile 6 for CPMV genome organisation).The AUG for the 95 kDa protein is in amore favourable context (G at positions –3 andþ4) than is that for the 105 kDa protein (A at –3and U at þ4).b. Overlapping ORFs. The genomes of membersof the Luteovirus and Polerovirus genera containsix ORFs, the 3 0 ones of which are expressedfrom sgRNAs (see Profile 8 for genome organisations).ORF4 (17 kDa) is contained in a differentreading frame within ORF3 (coat protein).The translation of ORF4 fits very well withleaky scanning from the translation initiationof ORF3, as the context of the ORF3 AUG (U atposition –3 and A at þ4) is unfavourable.c. Two or More Consecutive ORFs. Theexpression of ORFs I, II, and III of RTBV isanother example of leaky scanning due to thelack of AUG codons in upstream regions (seeBox 7.5).The translation of non-5 0 ORFs by leakyscanning usually results from the AUG startcodon in the upstream ORF being in a poorcontext. A lack or dearth of AUG codons inthe upstream region can enhance the effect.However, care must be taken in interpretingin vitro translation information as evidence forleaky scanning. Parameters such as translationsystem and conditions, especially the presenceof divalent cations, can affect the expressionfrom non-5 0 ORFS.8. Strategy 8. Non-AUG Start CodonSome viral ORFs appear to start with acodon that is not the conventional AUG startcodon; the initiation at these ORFs is inefficient.ORF1 of RTBV is AUU (Box 7.5).9. Strategy 9. TransactivationThe dsDNA genome of Cauliflower mosaicvirus (CaMV) has six closely spaced functionalORFs (I–VI; see Profile 4 for genome organisation)and is transcribed to give a more-thangenome-lengthRNA, the 35S RNA, and anmRNA (19S) for ORF VI. Although there issome evidence for the use of spliced RNAs forexpressing some of the ORFs, it appears thatsome of the ORFs are expressed from the longRNA. Most of the downstream ORFs are not,or are poorly, expressed in protoplasts or transgenicplants unless the product of ORF VI ispresent. This gene product is termed a transactivator(TAV) and is thought to facilitateinternal initiation.For reinitiation at a downstream ORF, thereshould not be overlapping of long ORFs, andit is particularly efficient when the first ORFis about 30 codons long. The polar effects ofthe insertion of stem-loop structures, whichwould inhibit the translocation of ribosomes,into polycistronic mRNAs and the specificityfor nonoverlapping ORFs indicated that transactivationcauses enhanced reinitiation of ribosomes.The optimal 30-codon length for thefirst ORF is just long enough to emerge from atranslating ribosome and suggests that theTAV acts directly or indirectly on the translatingor terminating ribosome.The transactivating function locates to the centralone-third of the TAV protein. TAV associatesII. WHAT IS A VIRUS MADE OF?

134 7. EXPRESSION OF VIRAL GENOMESwith polysomes, with an 18 kDa ribosomal orribosome-associated protein from plants andyeast and also with the 60S ribosomal subunitprotein L 18 (RPL18) from Arabidopsis thaliana.10. Strategy 10. Translational (Ribosome)ShuntShort ORFs (sORFs; defined arbitrarily ashaving fewer than 50 codons and no knownfunction for the product) in a leader sequencecan interfere with translation of a downstreamORF. In the translational ribosome shunt mechanism,initially scanning ribosomes are transferreddirectly from a donor to an acceptorsite without linear scanning of the interveningregion, thus avoiding sORFs.The 35S RNAs of members of the Caulimoviridaehave long and complex leader sequencesranging in length from about 350 to more than750 nucleotides and contain between 3 and19 sORFs. They fold to give complex stem-loopstructures. Translational shunting has beenproposed as the mechanism by which theseconstraints to translation are bypassed. Fortranslational shunting, there must be shuntdonor and acceptor sites and a well-definedstructure to bring these together. Most of thestudies on these features have been on CaMVand RTBV (see Box 7.5).11. Strategy 11. Read-Through ProteinsThe termination codon of the 5 0 gene may be“leaky” and allow a proportion of ribosomes tocarry on translation to another stop codondownstream from the first, giving rise to a second,longer, functional polypeptide. This istermed read-through or stop-codon suppression,resulting in a protein that has the same sequenceas the upstream protein in its N-terminal portionand a unique sequence in its C-terminalportion. An example is shown in Figure 7.5A,where the amber stop codon of TMV ORF1(which encodes a 126 kDa protein) is readthroughinto ORF2 to give a protein of 183 kDa.The 183 kDa protein has the same N-terminalamino acid sequence as the 126 kDa protein buta unique C-terminal sequence.The read-through strategy is found in severalplant virus genera. The proteins that are producedby read-through are either replicases(e.g., TMV) or extensions to the coat protein(e.g., luteoviruses) thought to be involved intransmission vector interactions (see Chapter 12).In read-through the stop codon is read by asuppressor tRNA instead of the ribosome beingreleased by the eukaryotic release factor complex.Essentially, competition exists betweenthe release factor complex and the suppressortRNAs. Two main factors are involved in readthroughof stop codons: the context of the stopcodon and the suppressor tRNAs involved.Stop codons have different efficiencies of termination(UAA > UAG > UGA), and also thefirst, and possibly the second, nucleotide 3 0 ofthe stop codon acts as an important efficiencydeterminant. In plant viruses either amber(UAG) or opal (UGA) stop codons are readthrough;there are no examples of read-throughof the natural ochre (UAA) stop codon. Unlikeretroviruses, no structural requirements forstop codon suppression in plant systems exist.The synthesis of a read-through proteindepends primarily on the presence of appropriatesuppressors tRNAs. Thus, two normaltRNAs tyr from tobacco plants were shown topromote UAG read-through during the translationof TMV RNA in vitro. The tRNA tyr musthave the appropriate anticodon, shown to beGcA, to allow effective read-through. tRNA glnwill also suppress the UAG stop codon.The proportion of read-through protein producedmay be modulated by sequence contextof the termination codon, by long-distanceeffects, and by the availability of the suppressortRNA. It is reasonable to suggest that about 1to 10 percent of the times that a ribosome reachesa suppressible stop codon result in read-through.As well as overcoming the constraints of theeukaryotic translation system, read-through ofstop codons provides a mechanism for theII. WHAT IS A VIRUS MADE OF?

V. PLANT VIRAL GENOME STRATEGIES135control of the expression of gene products.Transmission helper factors have to beincorporated in the virus capsid, but it is probablynot necessary to have them on all coat proteinsubunits. Thus, it could be more efficientfor the virus expression system if 1 to 10 percentof the coat protein subunits also containthe transmission factor. Similarly, the viral replicasecomprises several functional domains(described in Chapter 8) that are probablyrequired in different amounts and even at differenttimes. The production of two proteinsthat both contain some of the domains andone, the read-through protein, that also containsthe other domains would give controlover the availability of these functions.12. Strategy 12. Frameshift ProteinsAnother mechanism by which two proteinsmay commence at the same 5 0 AUG is by aswitch of reading frame before the terminationcodon of the 5 0 ORF to give a second, longer,“frameshift” protein. This translational frameshiftevent allows a ribosome to bypass the stopcodon at the 3 0 end in one reading frame andswitch to another reading frame so translationcan continue to the next stop codon in thatreading frame. A frameshift is illustrated inFigure 7.6.At the frameshift site, ribosomes change theirreading frame either one nucleotide in the 5 0direction (–1 frameshift) or one nucleotide inFIGURE 7.6 Translational frameshift. The ribosomebypasses the stop codon in frame 0 by switching backone nucleotide to frame –1 at a UUUAC sequence beforecontinuing to read in frame –1 to give the fusion or frameshiftprotein. [From Hull (2000).]the 3 0 direction (þ1 frameshift). The frameshiftprocess gives two proteins (the frame andframeshift proteins) that are identical from theN-terminus to the frameshift site but differentbeyond that point. The frame protein is alwaysproduced in greater quantity than the frameshiftprotein. Frameshift obviously occurs whereORFs overlap and may be at any place withinthat overlap.The frameshift strategy is found in severalplant virus genera and, in all cases that areknown, involves the replicase. Most instancesof frameshift are in the –1 direction (as shownin Figure 7.6); only those of the Closteroviridaeare in the þ1 direction (Figure 7.5B).For a –1 frameshift three features are needed: aheptanucleotide sequence, termed the “slippery”or “shifty” sequence at the frameshift site; astrongly structured region downstream of theframeshift site; and a spacer of four to ninenucleotides between the slippery sequence andthe structured region.The slippery sequence comprises two homopolymerictriplets of the type XXX YYY Z (X ¼ A,G, U; Y ¼ A, U; Z ¼ A, C, U). Upon reachingthis heptanucleotide sequence, the two ribosome-boundtRNAs that are in one readingframe (X.XXY.YYZ) shift by one nucleotide inthe 5 0 direction (XXX.YYY.Z), retaining twoout of the three base-paired nucleotides withthe viral RNA. This mechanism was deducedfor retroviruses, but the evidence points to asimilar mechanism in plant RNA viruses.The strongly structured regions are separatedby the spacer region from the frameshiftpoint and are either hairpins or pseudoknots(see Chapter 6 for pseudoknots). It is consideredthat the structure causes the ribosome topause, thereby initiating frameshift.Frameshifting in the þ1 direction requires arun of slippery bases and a rare or “hungry”codon or termination codon. A downstreamstructures region is not necessary but may befound, as has been suggested for Beet yellowsvirus (BYV). The following mechanism has beenII. WHAT IS A VIRUS MADE OF?

136 7. EXPRESSION OF VIRAL GENOMESsuggested for the þ1 frameshift between Lettuceinfectious yellows virus ORF1a and ORF1b:K ðprotein 1aÞ K ðprotein 1bÞLIYV RNA 5 0 ...AAAG ...3 0 slippage 5 0 ...AAAG ...3 0jjj ----------> jjtRNAlys 3 0 ...UUU ...5 0 3 0 ...UUU ...5 0As with the read-through strategy, most ofthe studies on the proportion of translationevents that result in as read-through proteininvolve in vitro systems. These can give frameshiftrates as high as 30 percent; in vivo ratesof 1 to 5 percent are more likely. Similar tothe read-through strategy, frameshift givescontrol of the production of functional domainsin proteins.C. Control of TranslationVarious mechanisms for the control of expressionof viral genomes have been described inthe preceding section. These relate primarilyto mechanisms that viruses have developed toovercome the problem of the limitation oftranslation of mRNAs in eukaryotic systems tothe 5 0 ORF. Among other features that controlor regulate the translation of eukaryotic hostmRNAs are the various noncoding regions(untranslated regions, or UTRs), which includethe termini of the RNAs, the 5 0 terminus beingcapped and the 3 0 terminus having a poly(A)tail. Also involved in the control and efficiencyof translation are the 5 0 leader sequence and the3 0 noncoding region. In eukaryotic mRNAs, thereis coordinated interaction between the 5 0 and 3 0UTRs and even evidence for circularisedmRNAs. Only some plant viral RNAs are cappedand have poly(A) tails. The majority either have acap or a poly(A) tail or have neither, althoughthese may be translated very efficiently.1. Cap but No Poly(A) TailThe genome of TMV is capped but lacksa poly(A) tail. The structure of the 3 0 UTR iscomplex being composed of five pseudoknotscovering a 177 base region. The 3 0 -terminal twopseudoknots form the tRNA-like structure thatis involved in virus replication (see Figure 6.2and Box 8.4). The remaining three pseudoknotsmake up the upstream pseudoknot domain thatis conserved in the tobamoviruses. This domainappears to functionally substitute for a poly(A)region in promoting interactions between the5 0 and 3 0 termini and enhancing translation initiationin a cap-dependent manner. A 102 kDahost protein binds to the pseudoknot domainand also to the 5 0 UTR. It is likely that this proteinis involved in bringing the 5 0 and 3 0 endstogether in the manner shown for eukaryoticmRNAs.2. Poly(A) Tail but No CapPotyviruses are polyadenylated at their 3 0 termini,but the 5 0 terminus is attached to a VPg(see Chapter 6 for VPg). The 5 0 UTR of Tobaccoetch virus confers cap-independent enhancementof translation of reporter genes by interactionsbetween the leader and the poly(A) tail.Two centrally located cap-independent regulatoryelements that promote cap-independenttranslation have been identified in the 143-nucleotide leader sequence.The VPg of the potyvirus Turnip mosaic virus(TuMV) interacts with the eukaryotic translationalinitiation factor eIF(iso)4E of Arabidopsisthaliana, and wheat (Triticum aestivim). eIF(iso)4E binds to cap structures of mRNAs and playsan important role in regulating the initiation oftranslation.3. Neither Cap nor Poly(A) TailMany plant viruses have neither a 5 0 cap nor a3 0 poly(A) tail, but these mRNAs are expressedvery effectively and thus can be considered tohave translation enhancement mechanisms.4. Cap SnatchingAll negative-strand RNA viruses with segmentedgenomes use a mechanism, termedII. WHAT IS A VIRUS MADE OF?

VII. SUMMARY137“cap snatching” to initiate transcription of theirmRNAs; this applies to viruses both from plants(tospoviruses and tenuiviruses) and from vertebratesand invertebrates (e.g., bunyaviruses andinfluenzaviruses). In this process, cap structurescomprising about 12–20 5 0 nucleotides arecleaved from host mRNAs by a virus-encodedendonuclease and are then used to prime transcription.Cap snatching has been demonstratedfor TSWV and the tenuivirus Maize stripe virus,both of which can snatch caps from positivesenseRNA viruses.5. 5 0 UTRAs well as being involved in the enhancementof translation initiation, the 5 0 UTR of TMV alsoenhances the efficiency of translation. The 67nucleotide 5 0 UTR, termed the O sequence, dramaticallyenhances translation of downstreamgenes in both plant and animal cells; in constructsin transgenic plants it enhanced translationby four- to sixfold. The O sequence hasreduced secondary structure and a 25 base poly(CAA) region, which mutagenesis indicated isthe primary element for in vivo translationalenhancement. The 36 base 5 0 leader sequencefrom AMV RNA 4 also enhances translation, asdoes the 84-nucleotide leader sequence of PVXgenomic RNA.D. DiscussionThe preceding descriptions show the greatdiversity of mechanisms that viruses use toexpress the information required for their functionfrom what are often compact genomes.Viral mRNAs have to compete with hostmRNAs for the translation machinery withoutcausing significant damage to the normal functioningof the cell. The diversity of mechanismsovercomes constraints imposed on viruses usingtheir host translational machinery. Most virusesuse more than one of the strategies outlined inthis chapter to express their genetic information,as shown by TMV, CTV, and RTBV.These mechanisms can be viewed in twoways: specifically overcoming the constraintsof the eukaryotic system—say, in the requirementfor a cap for translation initiation andthe translation of only the 5 0 ORF—and the controlof translation so the right product is in theright place in the right quantity at the righttime. The two uses of the mechanisms cannotbe separated. For instance, in many casesthe frameshift and read-through mechanismsprovide different functions of the replicationcomplex, the upstream one from the shorterprotein containing the helicase and cappingactivities and the downstream one the replicase.The regulation is not only by the recognisedORFs but can be by noncoding sequences andpossibly by short ORFs that normally mightnot be considered. There are many examples ofthe coat protein gene being expressed more efficientlythan that for the replicase; more coat proteinis required than the replicase.VI. VIRUSES OF OTHERKINGDOMSMost of the features in the expression ofplant genomes are applicable to viruses ofmembers of other eukaryotic kingdoms. Translationmechanisms in prokaryotes differ fromthose in eukaryotes, and thus there are differentconstraints on translation of viral genomesin these hosts.VII. SUMMARY• All viral infection cycles must producemRNA from which to express the viralgenome.• The uncoating of many plant viruseswith (þ)-strand ssRNA genomes isintegrated with the initial translationprocess.II. WHAT IS A VIRUS MADE OF?

138 7. EXPRESSION OF VIRAL GENOMES• There are constraints in the eukaryotictranslation system to the expression ofpolycistronic viral genomes.• Viruses have various strategies to overcomethe translation constraints.• There are strong controls on translation sothe right protein is expressed in the rightamount at the right place and at the righttime.ReferenceBaltimore, D. (1971). Expression of animal virus genomes.Bacteriol. Rev. 35, 235–241.Further ReadingDreher, T.W. and Miller, W.A. (2006). Translational controlin positive strand RNA plant viruses. Virology 344,185–197.Dolja, V.V., Kreuze, J.F., and Valkonen, J.P.T. (2006). Comparativeand functional genomics of closteroviruses.Virus. Res. 117, 38–51.Hanley-Bowdoin, L., Settlage, S.B., Orozco, B.M., Nagar, S.,and Robertson, D. (1999). Geminiviruses: Models forplant DNA replication, transcription and cell cycleregulation. Crit. Rev. Plant Sci. 18, 71–106.Hull, R. (2000). Matthew’s plant virology. Academic Press,San Diego.López-Lastra, M., Rivas, A., and Barria, M.I. (2005). Proteinsynthesis in eukaryotes: The growing biological relevanceof cap-independent translation initiation. Biol. Res. 38,121–146.Miller, W.A. and Koev, G. (2000). Synthesis of subgenomicRNAs by positive-strand RNA viruses. Virology273, 1–8.Rothnie, H.M., Chen, G., Fütterer, J., and Hohn, T. (2001).Polyadenylation in rice tungro bacilliform virus: cisactingsignals and regulation. J. Virol. 75, 4148–4198.Ryabova L., Park, H.S., and Hohn, T. (2004). Control oftranslation reinitiation on the cauliflower mosaic virus(CaMV) polycistronic. RNA Biochem. Soc. Transact. 32,592–596.Shaw, J.G. (1999). Tobacco mosaic virus and the study ofearly events in virus infection. Phil. Trans. R. Soc. Lond. B.354, 603–611.II. WHAT IS A VIRUS MADE OF?

C H A P T E R8Virus ReplicationOne of the major features of viruses is their ability to replicate their genomic nucleic acid, oftento high levels, in cells in which there are normally strict limits on the production of new nucleicacid molecules. Some viruses do this by adapting the existing cellular machinery, and othersreplicate their nucleic acid by mechanisms not widely used in host cells.O U T L I N EI. Host Functions Used by PlantViruses 139II.III.IV.Methods for Studying ViralReplication 140Replication of Plus-SenseSingle-Stranded RNA Viruses 140Replication of Negative-SenseSingle-Stranded RNA Viruses 152V. Replication of Double-StrandedRNA Viruses 152VI.VII.Replication of ReverseTranscribing Viruses 153Replication of Single-StrandedDNA Viruses 156VIII. Faults in Replication 158IX. Viruses of Other Kingdoms 161X. Summary 163I. HOST FUNCTIONS USED BYPLANT VIRUSESLike all other viruses, plant viruses are intimatelydependent on the activities of the host cellfor many aspects of replication, which includethe following:• Components for virus synthesis. Viruses useamino acids and nucleotides synthesised byhost-cell metabolism to build viral proteinsand nucleic acids. Certain other, morespecialised, components found in someviruses—for example, polyamines, are alsosynthesised by the host.Comparative Plant Virology, Second Edition139Copyright # 2009, Elsevier Inc. All rights reserved.

140 8. VIRUS REPLICATION• Energy. The energy required for thepolymerisation involved in viral protein andRNA synthesis is provided by the host cell,mainly in the form of nucleoside triphosphates.• Protein synthesis. Viruses use the host cell’sprotein-synthesising system for the synthesisof viral proteins using viral mRNAs. Manyviruses also depend on host enzymes for anyposttranslational modification of theirproteins—for example, glycosylation.• Nucleic acid synthesis. Almost all viruses codefor an enzyme or enzymes involved in thesynthesis of their nucleic acids, but they maynot contribute all the polypeptides involved.For example, in the first phase of thereplication of caulimoviruses, the viral DNAenters the host-cell nucleus and is transcribedinto RNA form by the host’s DNA-dependentRNA polymerase II. In most, if not all, RNAviruses, the replication complex comprisesthe viral RNA-dependent RNA polymerase(RdRp), several other virus-coded activities,and various host factors. ssDNA viruses alterthe cell cycle constraints on the host DNAreplication system. These aspects will bedeveloped in greater detail in this chapter.• Structural components of the cell. Structuralcomponents of the cell, particularlymembranes, are involved in virusreplication. For example, viral nucleic acidsynthesis usually involves a membraneboundcomplex. If we count a membrane asa compartment, eukaryotic cells have at least20 compartments. In their replication, plantviruses have adapted in a variety of ways tothe opportunities provided by thisintracellular metabolic diversity.II. METHODS FOR STUDYINGVIRAL REPLICATIONBecause of the involvement of host systemsand the close integration with other stages ofthe infection cycle, it is generally accepted thata full picture of viral replication can only beobtained from in vivo systems. However,owing to their complexity, in vivo systems areextremely difficult to establish, and many ofthe questions of detailed interactions and functionsmust be addressed by in vitro systems.One of the major constraints to studying thereplication of plant viruses is the establishmentof a system where the replication events aresynchronous. In inoculated whole plants, mostcells are infected sequentially, and so at anyone time a virus will be at different stages ofreplication in different cells. The threeapproaches to studying plant virus replicationare in vivo systems, manipulation of the viralgenome, and in vitro systems (Box 8.1).III. REPLICATION OF PLUS-SENSESINGLE-STRANDED RNA VIRUSESThe basic mechanism of replication of (þ)-sense RNA genomes is that the virus-encodedreplicase synthesises a complementary (–)strand, using the (þ) strand as a template, andthen new (þ) strands are synthesised from the(–)-strand template. Synthesis of new RNA isfrom the 3 0 to 5 0 ends of the templates. Replicationoccurs in a replication complex that comprisesthe templates, newly synthesised RNA,the replicase, and host factors.A. Viral TemplatesTwo kinds of RNA structures have beenisolated from viral RNA synthesising systems.One, known as replicative form (RF; Figure 8.1A),is a fully base-paired ds structure, whose role isnot certain. For example, it may represent RNAmolecules that have ceased replicating. The other,called replicative intermediate (RI), is only partly dsand contains several ss tails (nascent productstrands; Figures 8.1A and B).The ds nature of RFs and RIs is apparent whenthese molecules have been isolated from infectedII. WHAT IS A VIRUS MADE OF?

III. REPLICATION OF PLUS-SENSE SINGLE-STRANDED RNA VIRUSES141BOX 8.1METHODS FOR STUDYING PLANT VIRUS REPLICATIONA. In Vivo SystemsPlant viruses have been studied using severalin vivo systems:1. Protoplasts. Protoplasts are isolated plantcells that lack the rigid cellulose walls found inintact tissue. Metabolically active protoplastsisolated from leaf cells have been infected witha range of viruses that replicate in a near-synchronousmanner. This enables one-step virusgrowth experiments to be carried out, an importantkind of experiment that has long been availableto those studying viruses of bacteria andmammals.2. Temperature manipulation. The synchrony ofinfection in the young systemically infected leafcan be greatly improved by manipulating thetemperature. The lower inoculated leaves of anintact plant are maintained at normal temperatures( 25–30 o ), while the upper leaves are keptat 5–12 o . Under these conditions, systemic infectionof the young leaves occurs, but replicationdoes not. When the upper leaves are shifted toa higher temperature, replication begins in afairly synchronous fashion.3. Yeast. Yeast, Saccharomyces cerevisiae, is a single-cellorganism for which there is a considerableresource of classical and molecular genetics.Although yeast is the host for several viruses andvirus-like agents, no ssRNA virus is known tonaturally infect it. However, yeast has been shownto support the replication of some (þ)-strandssRNA viruses, including Brome mosaic virus andFlockhouse virus, and has been used to advancethe understanding of how RNA viruses replicate.B. Manipulating Viral Genomes1. Virus mutagenesis. Infectious cDNA or DNAclones are available for many viruses and openthe possibility to make specific mutations, theeffects of which on virus replication can be studiedin intact plants.2. Viral reporter systems. By manipulatingcloned viral genomes reporter molecules, usuallyfluorescent proteins can be attached tospecific viral gene products or expressed separatelyfrom the viral genome. This enables thevirus to be studied in intact plants in real timeand for details to be obtained on the exactlocation of the gene function being studied.Radioactively labelled virus precursors areoften used for identifying intermediates ofreplication.3. Metabolic inhibitors. Inhibitors of certainspecific processes in normal cellular metabolismhave been widely applied to the study ofvirus replication. Three have been of particularimportance: actinomycin D, which inhibitsDNA-dependent RNA synthesis but not RNAdependentRNA synthesis; cycloheximide,which is used as a specific inhibitor of proteinsynthesis on 80S cytoplasmic ribosomes; andchloramphenicol, which inhibits protein synthesison 70S ribosomes (e.g., in chloroplastsand bacteria).C. In Vitro Systems1. In vitro replication systems. There have beenvarious attempts to isolate competent replicationcomplexes from virus-infected plant material.The main problems are (1) the difficulty ofseparating complexes of proteins and nucleicacids from normal cell constituents; (2) membranesare an integral part of the replicationcomplexes of (þ)-strand RNA viruses, and thetechnology for isolating such components is notyet well developed; and (3) uninfected plantcells contain an endogenous RdRp, which isoften enhanced on virus infection.(continued)II. WHAT IS A VIRUS MADE OF?

142 8. VIRUS REPLICATIONBOX 8.1(continued)2. Primer extension. Properties of replicationcomplexes can be studied by adding nucleotide triphosphatesunder the appropriate conditions andassessing the resulting products from extension ofprimed strands on the existing template. The productscan be analysed by incorporating a labellednucleotide triphosphate (radioactive or fluorescentlabel) or by probing the product with a labelledprobe. This approach can be used to study theoptimum condition for the replicase enzymes.3. Enzyme activities. The enzymes involvedin replication have been purified by standardprotein and enzyme purification techniques,including size exclusion chromatography andion exchange chromatography. The properties ofthese enzymes have been studied by standardenzymatological techniques and other techniquessuch as activity gels.4. Protein-protein and protein-nucleic acid interactions.Virus replication often involves proteinproteinand protein-nucleic acid interactions.These have been studied by techniques such asthe yeast two-hybrid system, the yeast threehybridsystem, by cross linking using treatmentwith chemicals or with UV irradiation, by gelmobility shifts, and by sandwich blots.3FIGURE 8.1 Replication of (þ)-strandssRNA. A. Forms of association betweenpositive- and negative-sense viral RNA; left,replicative form (RF); right, replicativeintermediate (RI). [Redrawn from Matthews(2002).] B. Possible structures of RI; left,semiconservative; right, conservative.[Redrawn from Buck (1999; Phil. Trans. R.Soc. Lond. B. 354, 613–627).]ABcells. The nature of the structures in the infectedcell is unknown but is of great significance inview of the importance of dsRNA in the plantdefense response (see Chapter 11). It is likely thatthese molecules are essentially single-strandedin vivo, the strands being kept apart either bycompartmentalisation or by proteins bound tothem. The (þ) and (–) forms of the viral genomecontain the signals that control both the specificityand timing of their replication.It is likely that many of the control signalsare at the 3 0 ends of the template strands,though it should be recognised that theremight be long-distance signals elsewhere inthe template, and/or the 3 0 end interacts withother regions in the RNA. For (–)-strand synthesis,the 3 0 terminus of the (þ) strand hasbeen found to be important. As described inChapter 6, the three basic structures in plantviral RNA 3 0 termini are tRNA-like structures,poly(A) tails, and non-tRNA heteropolymericsequences. There is no real correlation betweenthe 3 0 terminal structure and the supergroupingof RdRps (described following), which mightindicate similar roles for dissimilar 3 0 termini.The 5 0 termini of the (þ) strands (i.e., the 3 0termini of the (–)-strand templates for (þ)-strand synthesis) are much more variable andII. WHAT IS A VIRUS MADE OF?

III. REPLICATION OF PLUS-SENSE SINGLE-STRANDED RNA VIRUSES143offer no obvious directions as to the specificpriming of (þ)-strand synthesis. It is generallythought that the specificity is provided by the(–)-strand initiation and synthesis and that the(–)- and (þ)-strand synthesis are coupled.B. ReplicaseThree or more virus-coded enzymatic activitiescan be involved in the replication of (þ)-strand RNA viruses: the RNA-directed RNApolymerase (RdRp), a helicase, and a methyltransferaseactivity. These are collectivelyknown as the viral replicase, but sometimesthis term is used (incorrectly) for the RdRp.1. RNA-Dependent RNA PolymeraseThe RdRp catalyzes the synthesis of RNAusing an RNA template. Two features have madethis enzymic activity difficult to study. First, it isusually associated with membrane structures inthe cell, and, on isolation, the enzyme(s) oftenbecome(s) unstable. Second, tissues of healthyplants may contain in the soluble fraction of thecell low amounts of a host enzyme with similaractivities. The amounts of such enzyme activitymay be stimulated by virus infection. Variousproperties of viral RdRps are outlined in Box 8.2.The three-dimensional structure of the RdRpof poliovirus (supergroup 1) has been determinedby X-ray crystallography (Figure 8.2).The overall structure of the enzyme is similarto those of other polymerases (DNA-dependentDNA polymerase, DNA-dependent RNA polymerase,and reverse transcriptase) that havebeen likened to a right hand. The palm domaincontains the catalytic core and is similar to thatof the other three polymerases. The thumb andfinger domains differ from those of the otherpolymerases. Using the neural net PHD (for“Predict at Heidelberg”) computer method thesecondary structure of the RdRps of BMV,TBSV, and TMV has been predicted and comparedwith the poliovirus RdRp structure. Thisanalysis indicated that the RdRps of these supergroup2 and 3 viruses have a similar structure tothe supergroup 1 poliovirus enzyme, each containinga region unique to RdRps.RdRp activity is also present in uninfectedplants and at one time was thought to beinvolved in RNA virus replication. This activityhas been isolated as a cDNA clone encoding a128 kDa protein.2. HelicasesHelicases are polynucleotide-dependentnucleoside triphosphate (NTP) phosphatasesthat possess ssDNA- and/or RNA-displacingactivity. They play a pivotal role in genomereplication and recombination by displacingcomplementary strands in duplex nucleic acidsand possibly removing secondary structurefrom nucleic acid templates. Some helicasesrequire a 3 0 flanking single strand of nucleicacid and others a 5 0 flanking single strand;these are known as 3 0 –5 0 and 5 0 –3 0 helicases,respectively. Some properties of viral helicasesare outlines in Box 8.3.Several plant virus genera appear to lack thecharacteristic NTP-binding motifs of helicases.Several possible reasons for this have beensuggested:1. The NTP-binding motifs may have divergedso much that they are not recognisable fromthe primary amino acid sequence.2. The viral polymerase may have unwindingactivity.3. Unwinding may be effected by a helixdestabilisingprotein that uses the energy ofstoichometric binding to ssRNA to melt theduplex in the absence of NTP hydrolysis.4. The virus may coopt a host helicase.Some plant viruses have additional helicaseactivities located elsewhere in their genomes.It is thought that the additional helicases inbenyviruses, hordeiviruses, and potexvirusesare involved in cell-to-cell movement.II. WHAT IS A VIRUS MADE OF?

144 8. VIRUS REPLICATIONBOX 8.2SOME PROPERTIES OF VIRAL RD R PSRdRps have various characteristic amino acid sequence motifs. The most conserved of these is a Gly-Asp-Asp (GDD) motif, which is flanked by segments of mainly hydrophobic amino acids and isinvolved with Mg 2þ binding. This suggests a structure of two antiparallel ß-strands connected by ashort, exposed loop containing the GDD motif. Further alignments of RdRp sequences show eightmotifs, which has led to the classification of RdRps into three “supergroups”:Supergroup 1 is sometime called the “picornavirus (-like)” supergroup.Supergroup 2 is the “carmovirus (-like)” supergroup.Supergroup 3 is the “alphavirus (-like)” supergroup.The supergroups extend across viruses that infect vertebrates, plants, and bacteria, and there arerepresentatives of plant viruses in each supergroup (Table). The members of each group have severalproperties in common.Supergroups of RdRpsSupergroup 1(Picornavirus supergroup)Supergroup 2(Carmovirus supergroup)Supergroup 3(Alphavirus supergroup)Comovirus Aureusvirus Alfamovirus PotexvirusPolerovirus Carmovirus Bromoviru TobamovirusPotyvirus Luteovirus Closterovirus TobravirusSobemovirus Tombusvirus Hordeivirus Tymovirus1. Supergroup 1 members are characterised by usually having one genome segment that has a 5 0VPg and expresses its genetic information as a polyprotein.2. Members of supergroups 2 and 3 have one to several genome segments, the RNA is often capped,and individual genes are translated.3. Methyl Transferase ActivityThe methyl transferase activity leading to 5 0capping of RNAs is described in Chapter 6.4. Organisation of Functional Domains inViral ORFsNot all virus genera have the three functionaldomains. Some genera, such as Tobamovirusand Cucumovirus, do indeed have allthree, but others, such as Comovirus and Potyvirus,lack the methyl transferase domain; thegenomes of these viruses do not have a m 7 G5 0 cap and therefore would not require thisactivity. Yet, others, like the Tombusviridae andthe Bromovirus genus, lack the helicase domain(Box 8.3).For all the viral genomes that express aspolyproteins or fused protein (frameshift orII. WHAT IS A VIRUS MADE OF?

III. REPLICATION OF PLUS-SENSE SINGLE-STRANDED RNA VIRUSES145ABCFIGURE 8.2 Structures of polymerases showing common features. A. Dengue virus RdRp. [From Yap et al. (2007; J. Virol.81, 4753–4765).] B. Moloney murine leukaemia virus reverse transcriptase. [This article was published in Structure, 12,D. Das and M.M. Georgiadis, The crystal structure of the monomeric reverse transcriptase from Moloney murine leukemiavirus, pp. 819–829, Copyright Elsevier Cell Press (2004).] C. A right hand for comparison.read-through), the domains appear to be in theorder (N-terminal to C-terminal) methyl transferase,helicase, and the RdRp; in divided genomes,it is not possible to allocate the order. However, afeature of many of the virus genera is that themethyl transferase and helicase domains arephysically separated from the RdRp domain.This can be by the methyl transferase and helicasedomains being in one ORF and the RdRpbeing in a separately expressed ORF (e.g., Bromovirus),by them being in two adjacent ORFs separatedby either a frameshift or read-throughtranslational event (e.g., Luteovirus), or thembeing on a polyprotein and separated by proteaseactivity (e.g., Potyvirus). However, in theCapilloviruses and Marafiviruses, the proteasedomain lies between the methyl transferase andthe other two domains. In the Trichoviruses, Vitiviruses,and Idaeoviruses, the three domainsappear not to be separated.For many viruses, it appears that the methyltransferase and helicase domains are on a singleprotein. However, although these two activitiesare expressed on the same ORF of Beetyellows virus, probing extracts from infectedplants with monoclonal antibodies indicatesthat in vivo the 295 kDa protein is processedto a 63 kDa protein containing the methyltransferase domain and a 100 kDa protein containingthe helicase domain.II. WHAT IS A VIRUS MADE OF?

146 8. VIRUS REPLICATIONBOX 8.3SOME PROPERTIES OF VIRAL HELICASESBased on conserved amino acid sequence motifs, helicases have been grouped into a number ofsuperfamilies. Five superfamilies have been recognised, three of which have representatives in(þ)-strand RNA viruses. Superfamilies I and II have seven conserved motifs, whereas superfamilyIII has three motifs. Two motifs common to all three superfamilies are variants of ATP-bindingmotifs and have the conserved sequences GXXXXGKT/S and MMMMD, where X is an unspecifiedamino acid and M is a hydrophobic residue. Most members of superfamilies I–III are 3 0 –5 0 helicases.The superfamily designation for various plant virus genera is given in the Table.Helicase superfamilies and plant virus generaSupergroup I Supergroup II Supergroup III No motifAlfamovirus Bymovirus Comovirus CarmovirusBromovirus Potyvirus Sequivirus LuteovirusClosterovirusPolerovirusCucumovirusSobemovirusHordeivirusTombusvirusPotexvirusTobamovirusTobravirusTymovirusThe crystal structure for the hepatitis C virus RNA helicase (superfamily II) has been determinedand the mechanism for unwinding duplex RNA suggested. The structure comprises three domainsforming a Y-shaped molecule. The RNA-binding domain is separated from the NTPase and otherdomain by a cleft into which ssRNA could be modeled. It is suggested that a dimer form of this proteinunwinds dsRNA by passing one strand through the channel formed by the clefts of the twomolecules and by passing the other strand outside the dimer. Because of the conserved motifsbetween the various superfamilies, it is likely that many of the features determined for the hepatitisC virus helicase are applicable to this enzyme from plant viruses.C. Sites of ReplicationPlant cells contain a range of membranes (seeBox 9.1). It is generally assumed that the replicationof (þ)-strand RNA viruses involves associationwith cellular membranes. In many virusinfections there is perturbation of membranestructures, frequently leading to the formationof vesicles, which in the proven cases have beenshown to be associated with replication complexes.A variety of membranes are involvedin vesicle formation (Table 8.1). There doesII. WHAT IS A VIRUS MADE OF?

TABLE 8.1 Example of Membranes PossiblyAssociated with (þ)-Strand RNA Virus ReplicationMembraneEndoplasmic reticulumChloroplast outer membraneVacuolar membranePeroxisomeMitochondrianot appear to be any correlation between themembrane involved and the supergroups ofthe replicase proteins or even with the virusfamily.D. Mechanism of ReplicationIII. REPLICATION OF PLUS-SENSE SINGLE-STRANDED RNA VIRUSESVirus aBMV. TMVTYMV, AMVCMVTBSVCIRV, TRVa Virus abbreviations: AMV, Alfalfa mosaic virus; BMV, Bromemosaic virus; CIRV, Carnation Italian ring spot virus; CMV,Cucumber mosaic virus; TBSV, Tomato bushy stunt virus;TMV, Tobacco mosaic virus; TRV, Tobacco rattle virus; TYMV,Turnip yellow mosaic virus.The RF and RI RNAs are considered to provideevidence on the mechanism of RNA replication.It is suggested that the RF could arisefrom the initial synthesis of a (–) strand on a(þ)-strand template. RI RNAs usually containmore (þ) strands than (–) strands, which istaken to indicate that each is a single (–) strandto which several (þ) strands are attached(Figure 8.1B).There are two hypotheses as to the mechanismof (þ)-strand synthesis. The semiconservativemechanism involves total displacementof the newly synthesised strand by the oncomingstrand (Figure 8.1B, left-hand structure),and in the conservative mechanism, it is suggestedthat the duplex RNA is only transientlyunwound at the growing end of the nascentstrands (Figure 8.1B, right-hand structure).The majority of evidence supports the semiconservativemechanism.E. Discussion147Boxes 8.4. 8.5, and 8.6 detail the replicationof three plant virus groups and show how(þ)-strand viruses replicate. The evidence forthe involvement of membranes is incontrovertible,but the reasons why different viruses usedifferent membranes (Table 8.1) are not yetunderstood.The replication complexes comprise severalvirus-coded proteins with different functions.These are assembled onto the relevant membraneby a membrane-binding protein ordomain that then interacts with the other components.The coordination of assembly andthe composition of the replication complexesare controlled not only by the interactionsbetween the component proteins (and nucleicacids) but also by the way they are expressed.Thus, in some cases, some of the componentproteins are expressed from frameshift orread-through, and in other cases, they areexpressed from a polyprotein processed in adefined manner.Various host proteins are involved in replicationcomplexes. The involvement of translationinitiation factors (Boxes 8.4–8.6) is of particularinterest in indicating coordination between translationand replication. However, these two functionsoperate along the template RNA indifferent directions, translation being 5 0 > 3 0 andreplication 3 0 > 5 0 , and thus, there must becontrols to prevent interference between the twoprocesses. Other cell processes, such as phosphorylation,play a role in control of replication.Most of the RNA elements involved in replicationoperate in cis, showing that the templateRNA is an integral part of the replication complex.The elements at the 3 0 end of the templateRNA that initiate (–)-strand synthesis appearto be well defined. This is in contrast to thoseat the 5 0 end of the genomic RNA, whichinitiate (þ)-strand synthesis. It seems likely that(–)-strand and (þ)-strand synthesis are highlycoordinated and that once the (þ)-strandII. WHAT IS A VIRUS MADE OF?

148 8. VIRUS REPLICATIONBOX 8.4REPLICATION OF TOBACCO MOSAIC VIRUSIn protoplasts, the synthesis of (–) strands ofTMV RNA ceases 6–8 hours after inoculation,whereas (þ)-strand synthesis continues for a further10 hours. The 5 0 ORF of TMV encodes a 126kDa protein, the stop codon of which is readthrough to give a 183 kDa protein (see Profile14 for genome organisation); thus, both proteinshave the same N-terminal sequence, the 183 kDaprotein having a unique C-terminal sequence.The proteins are in a 1:1 ratio, and both havethe methyl transferase and helicase motifs; the183-kDa protein also contains the RdRp motif.Efficient replication requires the 183 kDa proteinto form a heterodimer with the 126 kDa protein,which is probably bound to the template RNA.However, there may be still other functions forthe 126 kDa protein. As noted in Chapter 7, itis produced in about ten times the amount ofthe read-through product, the 183 kDa protein,yet the two proteins form a 1:1 heterodimer.There are various cis-acting factors involved inthe replication of TMV RNA. The 3 0 untranslatedregion (UTR) can be folded into three structuraldomains (see Figure 6.2), a 3 0 domain mimickinga tRNA acceptor branch, an analogue of a tRNAanticodon branch, and an upstream domain comprisingthree pseudoknots, each containing twodouble-helical segments. It is thought that thesecondary structure rather than the primarystructure is important in binding the replicase.Sequences at the 5 0 end of TMV RNA are alsoimportant for replication. Large deletions in the5 0 region and deletion of nucleotides 2–8 abolishedreplication, but other small deletions inthe 5 0 UTR did not. This suggests that the 5 0 replicasebinding site may be complex.Various host proteins have been found associatedwith TMV replication complexes. Theseinclude a 56 kDa protein that is immunologicallyrelated to the 55 kDa (GCD10) subunit of translationinitiation factor eIF3 from yeast and wheatgerm (56 kDa) and that interacts specifically withthe methyl transferase domain shared by the 126and 183 kDa TMV proteins. The association of aGCD10-like protein with the TMV replicationcomplex suggests that, in vivo, replication andprotein synthesis may be closely connected.The cellular site for TMV replication is associatedwith cytoplasmic inclusions or viroplasmsthat enlarge to form what are termed “X bodies”composed of aggregate tubules embedded in aribosome-rich matrix.The use of the green fluorescent protein(GFP) as a reporter for TMV expression identifiedirregular shaped structures in cells thatcontained the viral replicase (and movementprotein); these structures were derived from theendoplasmic reticulum (ER). Using in situ hybridisationand immunostaining, it was shown thatTMV RNA, the viral replicase, and the viralmovement protein colocalised at the ER, includingthe perinuclear endoplasmic reticulum; thesemolecules and associated with ER-related vesicles.template has been “captured” by the in vivoreplication complex, the full round of RNA replicationwill occur. The lack of (þ)-strand synthesisin most in vitro replication systemsindicates either that an important factor is lostduring extraction or that there are conformationalconstraints imposed by the location ofthe complex in vivo.II. WHAT IS A VIRUS MADE OF?

III. REPLICATION OF PLUS-SENSE SINGLE-STRANDED RNA VIRUSES149BOX 8.5REPLICATION OF BROMOVIRUSESA. Brome Mosaic VirusIn barley protoplasts infected in vitro with BMV, RNAs 1 and 2 (see Profile 3 for genome organisation)are detected 6 hours after infection. All four RNAs are present at 10 hours, and maximumRNA synthesis is from 16 to 25 hours. Virus particle formation is greatest between 10 and 25 hoursafter inoculation.BMV replication complex contains proteins 1a and 2a that have the motifs for methyl transferase(1a) and RdRp (2a). Protein 1a binds to protein 2a, the interaction being between a 115 amino acidregion at the N-terminus of protein 2a and a 50 kDa region of protein 1a encompassing the helicasedomain. The yeast two-hybrid system has shown further features of interactions between and withinthe 1a and 2a proteins. The direct interaction between these two proteins is stabilised by the presenceof the centrally conserved RdRp domain of 2a. There are both intramolecular interactions betweenthe capping and helicase domains of protein 1a and intermolecular 1a-1a interactions involvingthe N-terminal 515 residues of that protein. This has led to a model for the assembly of BMV replicase(Fig.); it is suggested that this model is applicable to other members of the Bromoviridae.Fig. Working model for theassembly of BMV 1a-2acomplex. The putative intramolecularinteraction in 1aprevents the formation ofthe 1a-2a complex until theintermolecular 1a-1a interactionhas occurred. The 2aprotein interacts with thehelicase-like domain of 1athrough its N-terminus.[From O’Reilly et al. (1998;J. Virol. 72, 7160–7169).]The replication complex obviously also involves BMV RNA. Cis-acting signals are located in the5 0 and 3 0 UTRs and in the intercistronic region (ICR) of RNA3 sites, where (–)-strand, (þ)-strand, andsg RNA synthesis are initiated. The 3 0 UTR of BMV is highly structured, mimicking tRNAs and forminga pseudoknot (see Chapter 6). The 3 0 134 bases of BMV RNAs containing the tRNA-like sequence(Figure 6.2B) have been identified as the minimum sequence requirement for in vitro (–)-strand production.Initiation in vitro is primer independent and is at the penultimate C residue of the 3 0 -CCA.This indicates that the terminal A residue is not templated but is added to the (þ) strands by tRNAnucleotidyl transferase after replication. The tRNA-like sequence contains the replicase-binding site,(continued)II. WHAT IS A VIRUS MADE OF?

150 8. VIRUS REPLICATIONBOX 8.5(continued)the specificity for the interaction with the replicase being determined by a stem-loop structure in thetRNA-like domain.A transition between initiation to elongation of (–)-strand synthesis occurs when nascent RNAs of10 nucleotides or longer remain associated with the replication complex; it can then be extended intofull-length RNAs whereas shorter RNAs were released from the complex.The 5 0 UTRs of bromovirus RNAs share sequence similarity, the most notable region matchingthe box B recognition sequence of RNA polymerase III promoters and thus also the conservedTCC loop of tRNAs; there is also a box B consensus in the intercistronic region of RNA3. It is suggestedthat this sequence might have a role in interacting with host factors.BMV RNA3 encodes two proteins, the downstream one being expressed from a sg RNA. As wellas the box B consensus sequence just described, BMV RNA3 has an oligo (A) tract in its intercistronicregion.Various host proteins have been found associated with BMV replication complexes, including aprotein antigenically related to the wheat germ eukaryotic translation initiation factor 3 (eIF-3) thatbinds strongly and specifically to BMV protein 2a. The development of the yeast system for studyingBMV replication has led to the identification of further host proteins that appear to be involved.Ultrastructural changes induced by BMV involve a proliferation and modification of parts of theER. Double-label immunofluorescence and confocal microscopy of BMV-infected barley protoplastsand in yeast cells show that 1a and 2a proteins colocalise throughout infection to defined cytoplasmicspots adjacent to, or surrounding, the nucleus and that BMV-specific RNA synthesis coincideswith these spots. The BMV replication complexes were tightly associated with markers for endoplasmicreticulum and not with the medial Golgi or later compartments of the cellular secretory pathway.In yeast cells, protein 1a localises to the endoplasmic reticulum in the absence of protein 2a.As it interacts with the 2a protein, it recruits that protein to the endoplasmic reticulum, and thusit is considered to be a key organiser of the replication complex.B. Alfalfa Mosaic VirusThe genome organisation of Alfalfa mosaic virus (AMV) is similar to that of BMV (see Profile 3),but there is a requirement of coat protein, either from the virus particle or expressed from sgRNA4 for replication. One to three coat protein dimers bind through their N-terminal amino acidsto a specific site near the 3 0 terminus of RNAs 1–3. There is a homologous sequence at the 3 0 terminiof the three genomic RNAs.Most of the basic features of the replication of AMV are similar to those just described for BMV.The major differences are the structure of the 3 0 terminus of the genomic RNAs and the necessity forbinding of coat protein for genome activation. Binding of coat protein to the 3 0 end of AMV RNAinhibits the production of (–)-strand RNA by interfering with the formation of the pseudoknot similarto that of BMV. It is proposed that the coat protein binding has two roles, first in very early stagesof virus replication, most probably in translation, and later in infection in shutting off (–)-strandRNA synthesis.II. WHAT IS A VIRUS MADE OF?

III. REPLICATION OF PLUS-SENSE SINGLE-STRANDED RNA VIRUSES151BOX 8.6REPLICATION OF POTYVIRUSESAlthough most of the protein products of potyvirusesare involved in some way in viral replication,there is a set of core replicationproteins: the CI helicase, the 6K 2 , the NIa VPgproteinase, and the NIb RdRp. As can be seenfrom the genome map (Profile 10), these form ablock of proteins. These proteins are analogousin terms of gene order and sequence motifs tothe Poliovirus 2C, 3A, 3B, 3C, and 3D proteins,respectively. Thus, it is likely that there are similaritiesin functions and in the replicationstrategy.The CI proteins of potyviruses have aminoacid motifs indicative of helicases and RNAbinding; NTPase and helicase activities havebeen demonstrated. The protein is membraneassociated and mutagenesis demonstrated thatit is involved in virus replication. The NIa processingproduct comprises two regions: the N-terminal VPg and the C-terminal protease; theproperties of the VPg are described in Chapter6. The NIb protein has an RdRp sequence motif,and nucleic acid binding has been demonstratedfor this protein.Various studies have revealed a series ofinteractions between several of these potyviralreplication-associated proteins. The NIb interactswith NIa, the interaction being betweenthe protease region for some viruses and theVPg region for others. The NIa and VPg proteinsstimulate the NIb-associated RNA polymeraseactivity, the stimulation being mainly attributableto the VPg.Thus, a picture is being built up of the compositionand assembly of potyviral replicationcomplexes. This involves both interactionsbetween the gene products and control of theprocessing of the polyprotein. Three polyproteinscontaining the 6K protein have beendetected in Tobacco etch virus (TEV)-infectedcells, CI/6K, 6K/VPg, and 6K/NIa (VPg plusproteinase).It is thought that the VPg is involved in theinitiation of RNA synthesis in a manner similarto that proposed for picornaviruses. It is likelythat host proteins and other potyviral proteinsare also recruited into the complex. The VPginteracts with the host translational eukaryoticinitiation factor (iso) 4E and the Nib protein witha host poly(A) binding protein. The 6K 2 proteinassociates with large vesicular structuresderived from endoplasmic reticulum (ER). Oninfection with TEV, the ER network appears tocollapse into discrete aggregated structures,and viral RNA in replication complexes is associatedwith ER-like membranes.Building on the interactions between the proteinsit is suggested that the replication complexassembles on ER-like membranes, initially by thebinding of the 6K protein to those membranes.This also brings in the NIa protein, as the membrane-bindingactivity of the 6K protein overridesthe nuclear localisation signal of NIa. NIabrings the viral RNA into the complex by itsRNA-binding ability. Then the 6K-NIa complexrecruits the NIb product through the NIa-NIbinteraction, delivering the polymerase to theRNA. The controlled processing of the potyviralpolyprotein enables the elements of the complexto be assembled and then released when theyhave completed their function. Thus, NIa, havingrecruited NIb, is released from the complex,and its nuclear-localisation takes over targetingit to the nucleus.II. WHAT IS A VIRUS MADE OF?

152 8. VIRUS REPLICATIONIV. REPLICATION OF NEGATIVE-SENSE SINGLE-STRANDEDRNA VIRUSESPlant rhabdoviruses resemble animal rhabdovirusesin having large membrane-boundparticles that contain a single species of (–)-sensessRNA (see Profile 13 for general descriptionand genome organisation). Basically, the (–)-strand virion RNA is associated with thenucleocapsid protein (N) to form coiled nucleocapsid;a large protein (L), considered to be thereplicase, is also associated with the nucleocapsid.The nucleocapsid is encased in the matrix(M) protein to form a core that, in turn, is envelopedin a membrane to form the bacilliform particle.As with animal rhabdoviruses, the (–)-strand genome of plant rhabdoviruses has twofunctions: as the template for transcription ofmRNAs for individual genes (described inChapter 7) and as the template for replicationvia a full-length (þ) strand [(–)-sense ssRNA >(þ)-sense ssRNA > (–)-sense ssRNA]. The polymerasecomplex undertakes both functions.All vertebrate rhabdoviruses replicate andassemble in the cytoplasm as do some plant rhabdoviruses(the cytorhabdoviruses), but otherplant rhabdoviruses (the nucleorhabdoviruses)replicate in the nucleus. The replication and transcriptionof both genera of plant rhabdovirusesthat lead to the formation of nucleocapsid coresoccur in viroplasms, which are discrete bodiesformed from accumulations of proteins andnucleic acids. The new nucleocapsid coresmature by budding through cellular membranesto acquire the outer viral membrane.When examining rhabdoviruses in the cell, itmust be remembered that the outer nuclearmembrane is contiguous with the endoplasmicreticulum (ER). Thus, nucleorhabdovirusesbudding through the inner nuclear membraneinto the perinuclear space (see Figure 2.5C)may further be included in vesicles derivedfrom the outer membrane and be found in thecytoplasm. Similarly, cytorhabdoviruses thatassociate with the ER may affect the outernuclear membrane, giving an appearance of anuclear involvement. The replication cycle ofplant nucleorhabdoviruses and cytorhabdovirusesis outlined in Figure 8.3.It must be remembered that plant rhabdovirusesalso replicate in their insect vectors.The membrane involved in the maturation ofthe virus particle depends on the insect celltype in which replication is taking place. Thereplication of tospoviruses resembles that ofother bunyaviruses, with the particles maturingin association with the Golgi stack membranes.V. REPLICATION OF DOUBLE-STRANDED RNA VIRUSESPlant members of the Reoviridae family haveeither 10 or 12 dsRNA genome segments,depending on the genus. The genome organisationof the 10 segments of Rice ragged stunt virusis outlined in Profile 11.The replication is dsRNA > (þ)-sense ssRNA> dsRNA. Plant reoviruses replicate in the cytoplasm,as do those infecting mammals. Followinginfection, densely staining viroplasmsappear in the cytoplasm of both infected plantcells and those of various tissues in the insectvector. Much of the viroplasm is made up ofprotein—most likely viral proteins. Viral RNAappears to be synthesised in the viroplasm,where the mature particles are assembled.The mature particles then migrate into thecytoplasm. Little is known about the detailedmolecular aspects of plant reovirus replication,but it is likely to be similar to that of animalreoviruses. It must also be remembered that aswith (–)-sense ssRNA viruses just discussed,the dsRNA viruses also replicate in theirinsect vector. The problem with packagingthe dsRNAs of reoviruses is discussed inChapter 5.II. WHAT IS A VIRUS MADE OF?

VI. REPLICATION OF REVERSE TRANSCRIBING VIRUSES153FIGURE 8.3 Models for the replication cycle of A, plant nucleorhabdoviruses, and B, cytorhabdoviruses. [This articlewas published in Encyclopedia of virology, A.O. Jackson, M. Goodin, I. Moreno, J. Jackson, and D.M. Lawerence (A. Granoffand R.G. Webster, Eds.), Plant rhabdoviruses, pp. 1531–1541, Copyright Elsevier Academic Press (1999).]VI. REPLICATION OF REVERSETRANSCRIBING VIRUSESA. IntroductionThe Caulimoviridae is the only family of plantviruses with dsDNA genomes, and it comprisessix genera that differ in genome organisationbut have essentially the same replication methods.Most experimental work has been carriedout on Cauliflower mosaic virus (CaMV) and Ricetungro bacilliform virus (RTBV). The replicationof members of the Caulimoviridae is DNA >RNA > DNA and thus resembles that of retroviruses;however, it does differ from that of retrovirusesin several important points:• The replication does not involve integrationinto the host genome for transcription of theRNA. This is done from an episomalminichromosome.• The virus does not encode an integrasegene.• The virion DNA is circular dsDNA and notthe linear DNA with long terminal repeatscharacteristic of retroviruses.• The DNA phase of the replicationcycle is encapsidated rather than theRNA phase, which is encapsidated inretroviruses.Thus, the Caulimoviridae are known aspararetroviruses.II. WHAT IS A VIRUS MADE OF?

154 8. VIRUS REPLICATIONAs with retroviruses, the replication cycle ofpararetroviruses has two phases: a nuclear phasewhere the viral DNA is transcribed by hostDNA-dependent RNA polymerase and a cytoplasmicphase where the RNA product oftranscription is reverse transcribed by virusencodedRNA-dependent DNA polymerase orreverse transcriptase (RT) to give DNA. In retroviruses,the RT activity is part of the pol gene,which also includes the RNase H activity thatremoves the RNA moiety of the RNA:DNA intermediateof replication. The pol gene is part ofthe gag-pol polyprotein that is cleaved by anaspartate proteinase, the gag being analogous tocoat protein. In pararetroviruses, the reversetranscriptase and RNaseH activities are closelyassociated. In badnaviruses, the coat proteinand pol are expressed from the same ORF, butin caulimoviruses they are expressed from separateORFs (see genome maps in Profile 4). Allplant pararetroviruses encode an aspartateproteinase.B. Reverse TranscriptaseThe structure of retrovirus RT is shown inFigure 8.2B, and it is thought that pararetrovirusRT has a similar structure; the structurehas been likened to a right hand as previouslydescribed for RdRp. RT has a characteristicmotif of tyrosine-isoleucine-aspartic acidasparticacid (YIDD), and several amino acidmotifs identify the RNase H domain; these arefound in caulimoviruses.C. Replication of “Caulimoviruses”1. Replication PathwayThe replication pathway is outlined inFigure 8.4. In the first phase of replication, thedsDNA of the infecting particle moves to thecell nucleus, where the overlapping nucleotidesat the gaps are removed, and the gapsare covalently closed to form a fully dsDNA.The covalently closed DNA associates withhost histones to form minichromosomes thatare the template used by the host enzyme,DNA-dependent RNA polymerase II, to transcribetwo RNAs of 19S and 35S, as describedin Chapter 7.The two polyadenylated RNA speciesmigrate to the cytoplasm for the second phaseof the replication cycle that takes place in inclusionbodies (see Figure 2.6). The 19S RNA is themRNA for gene VI that is translated in largeamounts to produce the inclusion body protein.To commence viral DNA synthesis on the35S RNA template, a plant methionyl tRNAmolecule forms base-pairs over 14 nucleotidesat its 3 0 end with a site on the 35S RNAcorresponding to a position immediately downstreamfrom the D1 discontinuity in the a-strandDNA (see following). The viral reverse transcriptasecommences synthesis of a DNA (–)strand and continues until it reaches the 5 0 endof the 35S RNA, with the RNase H activityremoving the RNA moiety of the RNA:DNAduplex, giving what is termed “strong stop”DNA. At this point, a switch of the enzyme tothe 3 0 end of the 35S RNA is needed to completethe copying. The switch is made possible bythe 180-nucleotide direct repeat sequence at eachend of the 35S RNA, which enables the 3 0 endof the strong stop DNA to hybridize with the3 0 end of the 35S RNA. When the template switchis completed, reverse transcription of the 35SRNA continues up to the site of the tRNAprimer, which is displaced and degraded to givethe D1 discontinuity in the newly synthesisedDNA (Figure in Profile 4).The rest of the used 35S template is removedby an RNase H activity. In this process, twopolypurine tracts (PPT) of the RNA are leftnear the position of discontinuities D2 and D3(Figure in Profile 4) in the second DNA strand(the (þ)-strand). Synthesis of the second(þ)strand of the DNA then occurs, initiating atthese two RNA primers. The growing (þ)strandII. WHAT IS A VIRUS MADE OF?

VI. REPLICATION OF REVERSE TRANSCRIBING VIRUSES155FIGURE 8.4 Diagram of the replicationcycle of CaMV. [From Hull (2002).]has to pass the D1 gap in the (–) strand, whichagain involves a template switch.The presence of RT activity in inclusion bodiesand virus particles and of replication intermediatesin virus particles indicates that, as with retroviruses,the reverse transcription of CaMV occursin particle-like proviral structures.2. Inclusion BodiesCaMV (and other caulimoviruses) inducecharacteristic inclusion bodies in the cytoplasmof their host cells. There are two forms of inclusionbodies: electron-dense ones that are madeup of ORF VI product (see Figure 2.6A) andII. WHAT IS A VIRUS MADE OF?

156 8. VIRUS REPLICATIONelectron-lucent ones (see Figure 2.6B) that aremade up of ORF II product; virus particles arefound in both types of inclusion bodies. Theelectron-dense inclusion bodies are the site forprogeny viral DNA synthesis and for theassembly of virus particles. As we will see inChapter 12, the electron-lucent inclusion bodiesare involved in aphid transmission of the virus.Most virus particles are retained within theinclusion bodies.VII. REPLICATION OF SINGLE-STRANDED DNA VIRUSESThe replication of members of the Geminiviridaeis ssDNA > dsDNA > ssDNA. Most of thegeminivirus genera have monopartite genomes,but many of the begomoviruses have bipartitegenomes; however, DNA A of bipartite begomovirusescontains all the information necessaryfor virus replication, except for the geneson DNA B encoding proteins involved inmovement of particles to the nucleus.Geminivirus particles usually accumulate inthe nucleus, and with some, such as Maizestreak virus (MSV), large amounts of virus accumulatethere. In some infections, fibrillar rings,which must be part of a spherical structure,appear in the nucleus.A. Geminivirus ReplicationGeminiviruses use the rolling-circle mechanismof replication similar to that describedfor viroids in Chapter 3. This is a two-step process.In the first phase, the ss (þ) strand is thetemplate for the synthesis of (–) strand to generatea ds, replicative form (RF). This RF actsas the template for both transcription, asdescribed in Chapter 7, and (þ)-strand synthesisgenerating free ssDNA. The priming of(–)-strand synthesis is usually by an RNAmolecule that is generated through RNApolymerase or DNA primase activity. (þ)-strand synthesis is primed by a site-specificnick in the (þ) strand of the RF.The elucidation of the replication cycle hasrevealed several aspects of the normal cellcycle, since geminivirus replication dependson many host functions. Of especial interest isthat geminiviruses replicate in differentiatedcells that are in the G phase and have shutdown most of their DNA replication activities.Thus, geminiviruses reactivate the replicationactivities that they require and convert the cellback to S phase (Box 8.7). The geminivirus replicationcycle is outlined in Figure 8.5.Minus-strand synthesis of MSV occurs in thenucleus and is primed by a small RNA oligonucleotidecomplementary to the 3 0 intergenicregion, which is extended by host DNA-dependentDNA polymerase.The priming of geminivirus (þ)-strand synthesisis through a DNA cleavage at a specificsite in vivo in the intergenic common region.The geminivirus Rep protein is a site-specificendonuclease that nicks and ligates (þ)-strandviral DNA at the same position in vitro.B. Geminivirus Rep ProteinsThe Rep protein is the only geminiviral proteinthat is essential for replication. In begomoviruses,Rep is encoded by ORF C1, and inmastreviruses, it is expressed from ORFs C1:C2 through a spliced mRNA; unspliced RNAgives RepA from ORF C1 (for geminivirusgenome organisation, see Profile 7).Rep and RepA are multifunctional proteinsin that they do the following:1. Localise within the nucleus.2. Have specific DNA recognition sites.3. Have site-specific endonuclease and ligationactivity for (þ)-strand viral DNA (seepreceding).4. Have ATP/GTPase activity.II. WHAT IS A VIRUS MADE OF?

VII. REPLICATION OF SINGLE-STRANDED DNA VIRUSES157BOX 8.7GEMINIVIRUS CONTROL OF CELL CYCLEGeminiviruses replicate in differentiated plant cells in which host DNA replication has ceased. Theviral replication is dependent on host DNA replication factors, and thus the cell cycle has to be modified.The viral Rep protein is involved in this modification. Rep proteins (or RepA of mastreviruses)bind to retinoblastoma (Rb) proteins from a variety of sources including plants.Animal Rb proteins regulate cell growth most probably through control of the transition of the G0/G1 into S phase of the cell cycle. It is thought that the plant analogues of Rb proteins have a similarfunction. Various animal DNA viruses control their host cell cycle through the binding of a virusencodedprotein with the host Rb protein through a LXCXE motif. Most geminivirus Rep (and RepA)proteins have this LXCXE motif and bind Rb proteins from various sources. Thus, the suggestion isthat the binding inhibits the Rb protein control that maintains the host cell in the G phase of the cellcycle, enabling it to return to S phase and produce the factors required for viral replication. However,for this to occur, Rep must be expressed from the incoming virus. Therefore, there must be enoughcapability in the newly infected cell to initiate (–)-strand synthesis to give the dsDNA for transcriptionof the mRNA for Rep. The C4 protein of Beet curly top virus also controls the G2/M checkpoint.GFig. The left part of the diagram shows the cell cycle: M, mitosis phase; G1, (growth) interphase; G0, quiescentphase; S, DNA synthesis phase; G2, interphase. The top box to the right shows how the retinoblastoma (RBR) proteininteracts with the E2F protein, preventing it from binding to its binding site that is necessary for transcriptionleading to the G1 phase. The bottom right-hand box shows how the viral Rep (or RepA) protein prevents the RBRprotein from interacting with E2F, thus enabling E2F to bind to the DNA, overcoming the G1/S phase checkpoint,and initiating transcription.II. WHAT IS A VIRUS MADE OF?

158 8. VIRUS REPLICATIONFIGURE 8.5 Diagram of the replication of a geminivirus. [Kindly provided by J. Stanley. From Hull (2002).]5. Activate the promoter for the coat proteingene mRNA.6. Interact with retinoblastoma proteins (seeBox 8.7).7. For the begomoviruses, Rep can repress itsown promoter and can stimulate theexpression of the proliferating cell nuclearantigen.VIII. FAULTS IN REPLICATIONThe two main ways by which faults arise inreplication are by mutation and recombination.A. MutationAs noted in Chapter 4, replication mutationscan be base substitutions, base additions, orbase deletions. In discussing mutations, onehas to distinguish between mutation frequencyand mutation rate. Mutation frequency is theproportion of mutants (averaged for an entiresequence or specific for a defined site) in agenome population. Mutation rate is the frequencyof occurrence of a mutation event duringgenome replication.The rate of mutational errors depends on themode of replication, the nucleotide sequencecontext, and environmental factors. As shownII. WHAT IS A VIRUS MADE OF?

VIII. FAULTS IN REPLICATION159in Figure 4.5, nucleic acids that replicateDNA!DNA have much lower mutation ratesthan those that replicate by other pathways. Thisis because DNA-dependent DNA polymerasehas a proofreading ability that checks that thecorrect nucleotide has been added, whereas theother polymerases (DNA-dependent RNA polymerase,RdRp, and RT) do not. The crystal structuresof RNA replicases and RT do not revealthe 5 0 to 3 0 exonucleolytic proofreading domainpresent in DNA-dependent DNA polymerases.B. RecombinationRecombination is the formation of chimericnucleic acid molecules from segments thatwere previously separated on the same moleculeor are present in different parental molecules.It usually, but not always, takes placeduring replication and can be a repair mechanismfor aberrations resulting from mutation.In many of the experiments on recombination,the design is to restore an important functionby recombination between two nucleicacids with lost or depleted function. In thisapproach there is strong selection for therecombination event that may distort measurementof recombination frequency. A more realisticpicture of the “natural” situation is given ifone performs the experiments under reducedor nonselective conditions. Thus, although it isrecognised that the rates of recombination,especially that of RNA, are high there are fewestimates as to the actual values under “natural”conditions. The mechanisms of recombinationbetween DNA and RNA viruses have bothdifferences and similarities.1. DNA Virus RecombinationThe two basic forms of recombination inDNA viruses are homologous recombination,which occurs between two DNA sequences thatare the same or very similar at the crossoverpoint, and nonhomologous or illegitimate recombination,which occurs at sites where thereis either microhomology or no obvious homology;the latter usually happens during doublestrandbreak repair. In animal and bacterialviruses, nonhomologous recombination is arare event and is usually mediated by a virusorhost-encoded protein. Homologous recombinationcan require specific host or viral proteinsbut can also be due to template switching duringreplication.Recombination is common among geminivirusesand is a major driving force in the evolutionof this virus family (e.g., see Box 4.5). Ithas been found both within and between geminivirusspecies, and there is strong evidence forboth homologous and nonhomologous recombination.Recombination is also common inCaMV and probably in all the Caulimoviridae.Both DNA and RNA recombination have beenimplicated in CaMV.2. RNA Virus RecombinationThree classes of RNA recombination havebeen recognised (Box 8.8). Various mechanismshave been proposed for RNA recombination.The most widely accepted is the replicasedriventemplate switching model, whichinvolves four elements: three RNAs [the primaryRNA template (donor strand), the strandsynthesised from the primary strand (nascentstrand), and the acceptor strand] and the replicasecomplex (Figure 8.6).Synthesis of the nascent strand on the donorstrand is halted or slowed temporarily, whichenables either the RdRp or nascent strand tointeract with the acceptor strand, leading totemplate switching. Thus, there are two typesof signal on the donor or nascent strand: one(pausing or arrest signal) that halts the RdRpbut from which it can escape and one (terminatorsignal) that releases the RdRp from theRNAs. It is thought that these signals may besimilar to those involved in template switchingII. WHAT IS A VIRUS MADE OF?

BOX 8.8RNA RECOMBINATIONThere are three forms of RNA recombination.Class 1, termed similarity-essential recombination,has substantial sequence similarity betweenparental RNAs. There can be two types of products:precise and imprecise recombinants. ForBMV, five features that influence this form ofrecombination have been identified:• The length of sequence identity of thecommon region between the donor andacceptor RNA needs to be 15 nucleotides ormore to effect efficient homologouscrossovers.• The extent of sequence identity is important.• The AU content of the common region with 61–65 percent AU supporting frequenthomologous recombinations.• The relative position of the AU-rich region.• The inhibitory effect of GC-rich sequences onupstream hotspot regions while present onthe acceptor RNA. Thus, a recombinationhotspot has a GC-rich common regionfollowed by an AU-rich region. It is thoughtthat the AU-rich region causes the replicaseto pause and promotes RdRp slippage,leading to the incorporation of nontemplatednucleotides; nontemplated nucleotides arefound in BMV recombinants.Fig. Three classes of RNA recombination. ReplicasemediatedRNA synthesis after the template-switch eventisshownbyanarrow.Thehairpinstructureshownonthe acceptor RNA symbolically represents various RNAstructures that are required for class 2 and 3 recombination.[This article was published in Virology, 235, P.D.Nagy and A.E. Simon, New insights into the mechanismsof RNA recombination, pp. 1–9, Copyright Elsevier(1997).]Class 2 recombination, similarity-nonessentialrecombination, occurs when there are no similarregions between the parents. It is thought that featuressuch as transesterification, RdRp bindingsites, and secondary structure play a role in therecombination event. In BMV this occurs at about10 pecent of the rate of class 1 recombination.Class 3 recombination, similarity-assistedrecombination, combines features from both classes1 and 2 recombinations. In this class, thereare sequence similarities between the parentalRNAs, but additional RNA determinants ononly one of the parental RNAs are required forefficient recombination.II. WHAT IS A VIRUS MADE OF?

IX. VIRUSES OF OTHER KINGDOMS161by DNA-dependent DNA polymerases and RTand to be the sequence and/or secondary structure(e.g., stable hairpin) of the donor ornascent RNA.Recombination is thought to occur in most, ifnot all, RNA viruses. The evidence for recombinationis chimeric molecules, defective (D), anddefective interfering (DI) molecules, which arediscussed in Chapter 3. However, someviruses, such as the bromoviruses and tombusviruses,appear to have higher rates of recombinationthan others do, and these have been thesubjects for detailed studies.Mutagenesis of the replicase also gives someinformation on the mechanism. Mutationswithin the helicase domain of BMV 1a proteinincreased the frequency of recombination andshifted the recombination sites to energeticallyless stable parts of the heteroduplex. This suggeststhat the mutations reduced the processivityof the replication complex facilitatingtemplate switching at higher frequencies. Onthe other hand, mutations in the RdRp (2a protein)decreased the frequency of recombination.FIGURE 8.6 Models for replicase-mediated templateswitching during recombination of BMV and of Turnip crinklevirus (TCV). A. Heteroduplex-mediated recombination between(þ) strands of BMV RNAs 1 and 3. B. Recombinationbetween satellite RNAs associated with TCV. The sequenceof the required hairpin motif is shown. C. Recombinationevents between the identical regions of BMV RNAs 2 and 3.Recombination is favoured when GC-rich and AU-rich arelocated as shown. [This article was published in Virology, 235,P.D. Nagy and A.E. Simon, New insights into the mechanismsof RNA recombination, pp. 1–9, Copyright Elsevier (1997).]3. Recombination and Integrated ViralSequencesSome Caulimoviridae and Geminiviridae viralsequences have been found to be naturallyintegrated into plant genomes. It is not altogethersurprising that nuclear-located viralDNA sequences are inserted into the host chromosomesby illegitimate recombination. However,cases exist in which episomal infectioncan arise from such integrated sequences, asexemplified in Box 8.9.IX. VIRUSES OF OTHERKINGDOMSThe replication of plant viruses is very similarto that of viruses in other kingdoms. Thereare a few differences, which were discussed inthis chapter.II. WHAT IS A VIRUS MADE OF?

162 8. VIRUS REPLICATIONBOX 8.9EPISOMAL INFECTION FROM INTEGRATED BANANASTREAK VIRUS (BSV) SEQUENCESIn certain banana (Musa) cultivars, there hasbeen apparently spontaneous outbreaks of BSV,especially during tissue culture and breedingprogrammes. The best studied situation is withvariety Obino l’Ewai (AAB genome) and the tetraploidsfollowing crossing with variety Calcutta4 (AA genome). The evidence forepisomal infections that arise from integratedsequences is as follows:• Many of the tetraploid lines had up to 100percent infection after crossing symptomlessparent plants. Tissue culture plantlets ofObino l’Ewai from symptomless motherplants had lower rates of infection, and thoseof Calcutta 4 no infection.• PCR of total DNA from Obino l’Ewai usingBSV primers and Southern blotting of thatDNA probed with BSV sequences gavepositive results even though no virus couldbe detected by immunoelectron microscopyor by immune-capture PCR.• Sequencing of genomic clones from Obinol’Ewai revealed a complex insert of BSV (seefollowing), the sequence of which was >99percent homologous to that of the episomalvirus.• The cloned products prepared by thesequence-specific amplification polymorphism(S-SAP) approach fell intothree classes, one of which comprised Musasequence interfacing BSV sequence.• Fluorescent in situ hybridisation revealed amajor and a minor locus.• Fibre-stretch hybridisation, in whichchromosomes are denatured, spread, andthen hybridized with fluorescent probesshowed that the integrants were complex (seefollowing).The picture that has been derived for BSV inObino l’Ewai is that there are two integrationsites. In each, there are tandem replicates of BSVsequence interspersed with Musa sequence. Themajor locus comprises six repeats over 150 kb,and the minor locus three repeats over 50 kb.The interface of the BSV and Musa sequence is atthe same site as the 5 0 end of the 35S RNA transcript.From there, a segment of BSV sequence isuninterrupted for about 5 kbp. There then followsa region of short BSV-derived sequences inreverse and forward orientation, followed by asegment of BSV in reverse orientation. Downstreamof this is a further region of BSV in forwardorientation covering the part of the genome missingfrom the first segment. The model is that the7.4 kbp episomal viral genome is derived fromthe integrant by two recombination events (Fig.),which are triggered by the stresses induced bycrossing and/or tissue culture.(continued)II. WHAT IS A VIRUS MADE OF?

X. SUMMARY163BOX 8.9(continued)Fig. Model for recombination fromintegrated BSV sequence to giveepisomal virus. A. Structure of anintegrated BSV sequence in MusaObino l’Ewai nuclear genomicDNA. Filled arrows and numbersrepresent the BSV sequence andtheir directions relative to theepisomal virus. X is a rearrangedassortment of short BSV sequences;open arrows represent directrepeats. B. Hypothetical intermediateafter a first recombination eventbetween 280 bp direct repeats atBSV 5530–5810. C. Episomal BSVsequence produced after a secondrecombination between 98 bp directrepeats at BSV 7265–7363. PredictedORFs are shown. [This articlewas published in Virology, 255,T. Ndowora, G. Dahal, D. LaFleur,G. Harper, R. Hull, N.E. Olszewski,and B. Lockhart, Evidence that badnavirusinfection in Musa can originatefrom integrated pararetroviralsequences, pp. 214–220, CopyrightElsevier (1999).]X. SUMMARY• A range of host functions includingnucleotides and amino acids and structuralcomponents of the cell are used by viruses.• Plant virus replication is studied using bothin vivo and in vitro systems.• (þ)- and (–)-sense single-stranded RNAviruses and double-stranded RNAviruses are replicated by a virusencoded polymerase. There are variouscis-acting control signals on the viralnucleic acid.• Plant viruses that replicate using a viruscodedreverse transcriptase are termedII. WHAT IS A VIRUS MADE OF?

164 8. VIRUS REPLICATIONpararetroviruses. Their replication is basicallysimilar to that of animal retroviruses butdiffers in several important points.• Plant single-stranded DNA viruses use thehost DNA replicase. They encode a proteinthat alters the cell cycle from the quiescent tothe DNA synthesis stage.• Faults in viral nucleic acid replication arecaused by mutation and recombination.Further ReadingBujarski, J.J. (2008). Recombination. Encyclopedia of Virology,Vol. 4, 374–381.Cann, A.J. (2008). Replication of viruses. Encyclopedia ofVirology, Vol. 4, 406–411.Hull, R. (2002). Matthew’s plant virology. Academic Press,San Diego.Nagy, P.D. (2008). Yeast as a model host to explore planthostinteractions. Annu. Rev. Phytopathol. 46, 217–242.For RNA virus replicationBeauchemin, C., Boutet, N., and Laliberté, J.-F. (2007).Visualisation of the interaction between the precursorsof VPg, the viral protein linked to the genome of Turnipmosaic virus, and the translation eukaryotic initiation factoriso 4E in planta. J. Virol. 81, 775–782.Jakubiec, A. and Jupin, I. (2007). Regulation of positivestrandRNA virus replication: The emerging role ofphosphorylation. Virus Res. 129, 73–79.Nagy, P.D. and Pogany, J. (2006). Yeast as a model host todissect functions of viral and host factors in tombusvirusreplication. Virology 344, 211–220.Noueiry, A.O. and Ahlquist, P. (2003). Brome mosaic virusRNA replication: Revealing the role of the host in RNAvirus replication. Annu. Rev. Phytopathol. 41, 77–98.Reichert, V.L., Choi, M., Petrillo, J.E., and Gehrke, L. (2007).Alfalfa mosaic virus coat protein bridges RNA andRNA-dependent RNA polymerase in vitro. Virology364, 214–226.Sanfaçon, H. (2005). Replication of positive-strand RNAviruses in plants: contact points between plant and viruscomponents. Can. J. Bot. 83, 1529–1549.For rhabdovirusesJackson, A.O., Dietzgen, R., Goodin, M.M., Bragg, J.N., andDeng, M. (2005). Biology of plant rhabdoviruses. AnnuRev. Phytopathol. 43, 623–660.Jackson, A.O., Dietzgen, R.G., Fang, R.-X-, Goodin, M.M.,Hogenhout, S.A., Deng, M., and Bragg, J.N. (2008). Plantrhabdoviruses. Enclopedia of Virology, Vol. 4, 187–196.For geminivirus replicationGutierrez, C. (2002). Strategies for geminivirus DNA replicationand cell cycle interference. Physiol. Molec. PlantPathol. 60, 219–230.Hanley-Bowdoin, L., Settlage, S.B., and Robertson, D.(2004). Reprogramming plant gene expression: A prerequisiteto geminivirus DNA replication. Molec. PlantPathol. 5, 149–156.For pararetrovirus replicationHohn, T. and Richert-Pöeggeler, K.R. (2006). Replication ofplant pararetroviruses. In Recent advances in DNA virusreplication (K.L. Hefferon, Ed.), pp. 289–319. ResearchSignpost 37/661, Kerala, India.II. WHAT IS A VIRUS MADE OF?

S E C T I O N IIIHOW DO PLANTVIRUSES WORK?

This page intentionally left blank

C H A P T E R9Virus-HostInteractions — Plant LevelTo induce a disease, the virus must spread throughout and replicate in much of the plant. Atthis stage, the viral genome and the host genome confront one another, with the virus attemptingto establish infection and the host attempting to resist it.O U T L I N EI. Movement and Final Distribution 167II. Effects on Plant Metabolism 181III.Processes Involved in SymptomProduction 185IV. Viruses of Other Kingdoms 188V. Summary 188I. MOVEMENT AND FINALDISTRIBUTIONAs outlined in Table 9.1, a plant responds inmany different ways to the introduction of avirus into the initial cell. (We will examine totalresistance in Chapter 10 and recovery in Chapter11.) For local and systemic infection, virus movementis closely coupled with virus replicationand is a dynamic regulated cascade of events.The full systemic infection of a plant isshown in Figure 9.1. From the initially infectedcell, the virus moves locally to adjacent cellsand then to the vascular system, enabling fullsystemic spread to distal parts of the plant(for relevant details of plant anatomy, see Box2.2). An exception to this is that initial cell-tocellmovement may be bypassed in phloemlimitedviruses, which are injected directly intothe phloem by their vector (see Chapter 12).Comparative Plant Virology, Second Edition167Copyright # 2009, Elsevier Inc. All rights reserved.

168 9. VIRUS-HOST INTERACTIONS — PLANT LEVELTABLE 9.1 Types of Response by Plantsto Inoculation with a VirusIMMUNE (nonhost). Virus does not replicate in protoplastsor in the initially inoculated cells of the intact plant.Inoculum virus may be uncoated, but no progeny viralgenomes are produced.INFECTIBLE (host). Virus can infect and replicate inprotoplasts.Resistant (extreme hypersensitivity). Virus multiplicationis limited to initially infected cells because of anineffectual virus-coded movement protein, giving riseto subliminal infection. Plants are field resistant.Resistant (hypersensitivity). Infection limited by a hostresponse to a zone of cells around the initially infectedcell, usually with the formation of visible necroticlocal lesions. Plants are field resistant.Susceptible (systemic movement and replication)Sensitive. Plants react with more or less severe disease.Tolerant. There is little or no apparent effect on the plant,giving rise to latent infection.A. Intracellular MovementIn Chapter 8, we saw the association of replicationof viruses with cell membrane systems.Plant viruses use the cytoskeleton and membranesystems (Box 9.1) to move from the sitesof replication to the periphery of the cell toenable infectious units to pass to adjacent cells.Details of the process of intracellular movementof viruses are difficult to separate fromthose of intercellular movement, but the generalpicture is that various virus proteins suchas movement proteins (see below) and coatproteins are involved. The actual proteins usedin the process differ between different virusgroups, but the basic process appears to be thata complex of viral nucleic acid and proteinbinds to microtubules and/or microfilamentsand is translocated to the plasmodesmata; thecomplex, including cytoskeleton components,may also bind to the endoplasmic reticulum.B. Intercellular MovementThe cell-to-cell (or short-distance) movementis from the initially infected cell(s), which areusually epidermal or mesophyll cells, to thevascular bundle. In the majority of cases thereare two major barriers to movement: movementfrom the first infected cell and movement out ofparenchyma cells into vascular tissues.1. PlasmodesmataSince the virus cannot cross the cell walldirectly, it must use plasmodesmata, whichare cytoplasmic connections between adjacentcells (Box 9.2). However, plant virus particles,or even free, folded viral nucleic acids, are toolarge to pass through unmodified plasmodesmata(Figure 9.2). Thus, the plasmodesmatalsize exclusion limit (SEL) has to be increased,and viral movement proteins (MPs) facilitatethis.Of special note to virus movement are plasmodesmatabetween the bundle sheath andphloem parenchyma, between the phloemparenchyma and companion cells, and betweenthe companion or intermediary cells and thesieve elements, giving different tissue boundaries,at least for some viruses. Plasmodesmatabetween the bundle sheath and phloem parenchymadiffer from mesophyll plasmodesmatain that they require additional modificationbefore some viruses can enter the phloem ofminor veins. The systemic infection of severalviruses, such as Brome mosaic virus (BMV),appears to be controlled by the bundlesheath–vascular bundle interface. This tissueboundary also appears to be demonstrated byphloem-limited viruses (e.g., luteoviruses) thatare unable to spread across the bundle sheathto mesophyll cells (see following). Plasmodesmatabetween companion cells and sieve elementshave a special structure comprising asingle pore on the sieve element wall and abranched arrangement in the adjoining companioncell wall.III. HOW DO PLANT VIRUSES WORK?

I. MOVEMENT AND FINAL DISTRIBUTION169FIGURE 9.1 Diagram showing thespread of TMV through a mediumyoung tomato plant. The inoculatedleaf on the left is marked by hatching,and systemically infected leaves areshown in black. [From Ann. Appl. Biol.21, G. Samuel, The movement oftobacco mosaic virus within the plant,90–111, Copyright Wiley-Blackwell(1934).]There is considerable variation in plasmodesmatalSELs, depending on factors such astype of plasmodesma and the sink-source transitionof photoassimilate production, and measurementshave ranged from less than 10 kDato more than 50 kDa. Plasmodesmata occur ingroups on cell walls, and it must not beassumed that those within one group are allthe same. Similarly, they will vary with timeand possibly with condition, such as virusinfection. Thus, the system of interconnectionsbetween plant cells is dynamic and changingwith the different functional demands that areplaced on this symplastic system.2. Movement Proteins (MPs)Most plant viruses encode MPs, which havethree functional characteristics:• They are associated with and/or havethe ability to increase the SEL ofplasmodesmata.• They have the ability to bind to eitherssRNA or ssDNA.• They have the ability to transport themselvesand/or viral nucleic acid from cell to cell.The wide variety of virus-encoded movementproteins fall into four main types. Type 1III. HOW DO PLANT VIRUSES WORK?

170 9. VIRUS-HOST INTERACTIONS — PLANT LEVELBOX 9.1PLANT ENDOMEMBRANE AND CYTOSKELETAL SYSTEMSThe cell contains a membrane and cytoskeleton system extending through the cytoplasm and linkingvarious organelles and other features.a. The Endoplasmic ReticulumThe endoplasmic reticulum (ER) system is a pleomorphic and multifunctional organelle found inall eukaryotic cells offering a large membrane surface with different functional domains. At least 16types of ER domain have been recognised (Fig.). Of these, seven are currently recognised as having,or potentially having, an involvement with plant viruses.Fig. Schematic diagram of a plant cell depicting the various forms of the ER domain. MT, microtubule; PM,plasma membrane; TV, transport vesicle; TGM, trans-Golgi network. [From Plant J. 11, A. Staehelin, The plantER: A dynamic organelle composed of a large number of discrete functional domains, pp. 1151–1165, Copyright Wiley-Blackwell (1997).](continued)III. HOW DO PLANT VIRUSES WORK?

I. MOVEMENT AND FINAL DISTRIBUTION171BOX 9.1(continued)The outer membrane of the nuclear envelope is continuous with the ER and has membraneboundribosomes. The outer nuclear membrane joins the inner membrane at nuclear pore complexes(2 in Fig.) that mediate the directed transport of proteins and nucleic acids between the nucleus andcytoplasm. Plant rhabdoviruses mature by budding through either the inner nuclear membrane orthe ER (see Chapter 8). The outer membrane of the nuclear envelope is also the major microtubuleorganising centre (MTOC) in plant cells (4 in Fig.). The two classical types of ER are the rough (sheet)and smooth ER, which are distinguished by the presence (rough) or absence (smooth) of attachedribosomes (5 and 6 in Fig.).As we saw in Chapter 8, the membrane of tospoviruses is derived by budding into the Golgi cisternaethough the replication and expression of the viral genome takes place in the cytoplasm. Thetransport of proteins from the ER to the Golgi bodies is by transport vesicles from the transitionalER (7 in Fig.). The ER has a structural association with filaments in plant cells suggesting actinbindingdomains (11 in Fig.). The seventh domain with viral associations is the part of the ER thatpasses through plasmodesmata (16 in Fig.).b. The CytoskeletonThe two main elements of the plant cytoskeleton are the microtubules and the microfilaments.Microtubules are made up of tubulin. In plant cells, the a and ß tubulins that form the a/ß dimmers,which are the basic units for constructing microtubules and g tubulin, are found in association withall microtubule arrays. Associated with microtubules are specific proteins, microtubule-associatedproteins (MAPs), some of which are mechano-chemical motors that move various components alongmicrotubules. Other MAPs modulate the assembly of microtubules at the MTOCs (4 in Fig.) and thedisassembly at their distal end; this forms a “treadmill” that moves any protein associated with themicrotubule from the MTOC end to the distal end. These functions are probably involved in cytoplasmicstreaming and intracellular transport of molecules and macromolecules.Microfilaments are composed primarily of actin, a helical assembly of globular units (G-actin)that form a ropelike filament (F-actin). The endomembrane system is surrounded by a sheath ofcytoskeleton, a primary constituent of which is F-actin. This endomembrane sheath confers bothbasic structure and the structural dynamism to the ER and is suggested to be the framework on,and within which, many of the metabolic reactions within the cell are organised and operate. Thesheath links the ER to various organelles, such as the plasma membrane, via various molecules.Actin is associated with plasmodesmata and is suggested to be involved in regulation of the sizeexclusion limit.is exemplified by TMV and has been studied inmost detail. The MP of 30 kDa (P30; see Profile14 for TMV genome organisation) has a predictedsecondary structure suggesting a seriesof ß-elements flanked by an a-helix at eachend. The MP binds to the viral RNA, meltingthe tertiary and secondary structure to form aribonucleoprotein (RNP) complex of 1.5–3 nmdiameter. The MP also modifies the structure ofthe plasmodesmata to give a diameter of 3–4 nm.III. HOW DO PLANT VIRUSES WORK?

172 9. VIRUS-HOST INTERACTIONS — PLANT LEVELTransport of the RNP complex to and throughthe plasmodesmata is considered to be anactive process involving interaction with actinmicrofilaments and possibly the ER. Variousfunctional domains have been identified onTMV P30, including targeting the MP to theplasmodesmata, increasing the SEL and bindingssRNA.BOX 9.2PLASMODESMATAPlasmodesmata are cytoplasmic connections through the wall of adjacent plant cells and form animportant route for communication between the cells. They regulate cell-to-cell communication, thusenabling the differentiation of plant organs and tissues. Developmental changes in their structure,frequency, and size exclusion limit (SEL) can lead to establishment of symplastic domains, withinwhich the metabolism and functions of cells is probably synchronised.The basic structure of plasmodesmata consists of two concentric membrane cylinders—theplasma membrane and the endoplasmic reticulum (appressed ER or desmotubule)—that traversethe cellulose walls between adjacent plant cells (Fig. panel A). The annulus between the two membranecylinders gives continuity of the cytosol between cells. High-resolution electron microscopyreveals proteinaceous particles about 3 nm in diameter embedded in both the plasmamembraneand the appressed ER and connected by spokelike extensions. The spaces between these protein particlesare thought to form tortuous microchannels about 2.5 nm in diameter. Injection of dyes of variousmolecular radii indicates that these microchannels have a basal SEL allowing passive diffusionof molecules of about 1 kDa.AFig. The structure of plasmodesmata. A. Diagram of the structure of a simple plasmodesma. [From Ghoshroy et al.(1997; Annu. Rev. Plant Physiol. Plant Mol. Biol. 48, 27–50). Reprinted, with permission, from the Annual Review of PlantPhysiology and Plant Molecular Biology, Volume 48 # 1997 by Annual Reviews www.annualreviews.org.](continued)III. HOW DO PLANT VIRUSES WORK?

I. MOVEMENT AND FINAL DISTRIBUTION173BOX 9.2(continued)CB[Electron micrographs of sections showing B. primary plasmodesmata and C. secondary plasmodesmata [(c)].ML, middle lamella of cell wall; Ca, central cavity in a plasmodesma. [From Ding et al. (1992; Plant Cell 4, 915–928).]Plasmodesmata come in many varieties (Fig. panel B). Primary plasmodesmata formed in a newcell wall are simple but then undergo modification to give complex structures (secondary plasmodesmata)with branched channels and a conspicuous central cavity that are formed in a basipetalpattern as leaves undergo expansion growth. Of especial note to virus movement are plasmodesmatabetween the bundle sheath and phloem parenchyma, between the phloem parenchyma andcompanion cells, and between the companion or intermediary cells and the sieve elements, givingdifferent tissue boundaries, at least for some viruses.III. HOW DO PLANT VIRUSES WORK?

174 9. VIRUS-HOST INTERACTIONS — PLANT LEVELFIGURE 9.2 The relative sizes of some plant virus particles(above) compared with the size of a plasmodesma(below): CaMV, Cauliflower mosaic virus; CPMV, Cowpeamosaic virus; CTV, Citrus tristeza virus; LNYV, Lettucenecrotic yellows virus; PVY, Potato virus Y; TMV, Tobaccomosaic virus; TSWV, Tomato spotted wilt virus; WTV, Woundtumor virus. [From Gibbs (1976; in Intercellular communicationin plants: Studies on plasmodesmata, B.E.S. Gunning andR.W. Robards, Eds., pp. 149–164, Springer-Verlag, Berlin).]In several groups of viruses (type 2)—forinstance, the hordeiviruses and the potexviruses—the MP activity is found in three overlappingor closely located ORFs known as the triplegeneblock (TGB) encoding TGBp1, TGBp2,and TGBp3 (see Profiles 1 and 9, respectively,in Appendix for genome organisations).In type 3, the potyviruses appear not to havea dedicated MP or MPs but involve severalvirus-coded proteins that also have other functionsto that of cell-to-cell movement (see Profile10 for potyvirus genome organisation).The viral RNA forms a complex with the coatprotein and the HC-Pro; both these proteinsincrease the SEL of plasmodesmata. The CIprotein is a helicase and is found near plasmodesmataearly in infection. It is considered to bean ancillary protein to the movement processprobably facilitating delivery to and alignmentof the HC-Pro/Coat protein/viral RNA complexat the plasmodesmatal pore.In type 4 proteins, some viruses, such ascomoviruses, caulimoviruses, and tospoviruses,produce tubules that extend through plasmodesmataor cell walls (Figure 9.3). Electronmicroscopy shows that those of comovirusesand caulimoviruses contain virus particles, andthose of tospoviruses contain presumed nonenvelopednucleocapsids. The 58K/48K MP ofCowpea mosaic virus (CPMV) is encoded by theRNA 2 (see Profile 6 for CPMV genome organisation),the 48K moiety forming virioncontainingtubules (35 nm diameter) replacingthe desmotubules. CaMV MP is the productof ORFI (P1), a 38 kDa protein forming thetubule and having nucleic acid binding activityin vitro.The model for this form of virus movement(Figure 9.4) is that MPs localise to the plasmodesmata,where they induce the removal ofthe desmotubule and assemble into tubulesextending unidirectionally into the adjacentplant cell. Virions assembled in the cytoplasmare escorted to the tubular structures throughinteractions with their MP and are then transportedto the adjacent cell.Thus, there are two basic strategies for cell-tocellmovement (Figure 9.4). In strategy 1 (types1, 2, and 3), the plasmodesmata are gated openat the infection front and close after, thus maintainingthe integrity of normal intercellular communication.In most, if not all, cases theinfection unit that passes through is a nucleoproteincomplex and not the discrete virus particle.In strategy 2 (type 4), the plasmodesmataused for viral transport are permanently modified,usually by the movement protein forminga tubule through which virus particles move(except in the case of tospoviruses).III. HOW DO PLANT VIRUSES WORK?

I. MOVEMENT AND FINAL DISTRIBUTION175FIGURE 9.3 Virus particles intubules. A. Plasmodesma containing atubule with virus particles in a sectionof a zinnia leaf infected with the caulimovirus,Dahlia mosaic virus; v¼ virusparticles. [This article was published inVirology, 37, E.W. Kitajima and J.A.Lauritis, Plant virions in plasmodesmata,pp. 681–685, Copyright Elsevier(1969).] B. Tubules extending from a protoplastinfected with CPMV and probedwith an antibody to the 48/58 kDa viralgene product linked to a fluorescenceprobe; bar ¼ 5 mm. C. Electron micrographof CPMV tubule in a partially purifiedfraction; bar ¼ 100 nm. [FromKasteel (1999; PhD Thesis, Universityof Wageningen, The Netherlands).]ABCAFIGURE 9.4 Models for plant virus intracellular movement. A. TMV. Viral genomic RNA complexed with MP movesalong microtubules from ER-associated sites of replication and protein synthesis (viral factories) to establish additional viralfactories at other ER sites. From these sites the RNA-MP complex associates with actin filaments and is delivered to theplasmodesmata. B. Viruses that form MP tubules. The viral genome encapsidated in virus particles moves through MPcontainingtubules that pass through highly modified plasmodesmata. CP, coat protein; ER, endoplasmic reticulum; MF,microfilament; MP, movement protein; MT, microtubule; PD, plasmodesmata; vRNA, viral RNA genome. [Modified fromLazarowitz and Beachy (1999; Plant Cell, 11, 535–548).]B3. What Actually MovesFor many viruses, movement proteins are notthe only viral gene product involved in cell-to-cellmovement (Table 9.2). For strategy 1, shown inFigure 9.4, five different combinations of proteinshave been recognised. Group 1 viruses requireonly the movement protein, whereas group2 viruses also require the viral coat protein. ThessDNA begomoviruses (group 3) require anuclear shuttle protein as well as the movementprotein. Groups 4 and 5 viruses require multiplegenes for movement.4. Cell-to-Cell Movement of ViroidsThe rapid movement of Potato spindle tuborveroid from cell to cell is mediated by a sequencespecificor structural motif (see Chapter 3).III. HOW DO PLANT VIRUSES WORK?

176 9. VIRUS-HOST INTERACTIONS — PLANT LEVELTABLE 9.2Viral Factors Involved in the Local and Systemic Spread of Plant VirusesVirus Viral Factor Known or Possible Site of EffectTMV MP Cell-to-cell movementCoat proteinPossible for entrance to CC/SE complex126/183 kDa proteins Entrance into CC and possible entry into and exit from SETEV Coat protein, CI Cell-to-cell movementHC-ProPossible entry into or exit from SEVPgPossible entry into or exit from SETEV, Tobacco etch virus; TMV, Tobacco mosaic virus; CI, cytoplasmic inclusion; VPg, genome-linked protein; CC, companioncell, SE, sieve element.5. ComplementationThe presence of a virus with a compatibleMP can complement the cell-to-cell spread ofanother virus in an apparent nonpermissivehost. For example, Southern cowpea mosaic virus(a sobemovirus with isometric particles) doesnot accumulate in leaves of bean (Phaseolus vulgaris)when inoculated by itself, but it doeswhen coinoculated with Sunn-hemp mosaic virus(a tobamovirus with rod-shaped particles).Complementation can also lead to the invasionof cells outside the bundle sheath by phloemlimitedviruses. For instance, joint infectionwith a mesophyll cell-infecting virus can leadto luteoviruses also infecting mesophyll cells.6. Rate of Cell-to-Cell MovementTo measure the movement out of the initiallyinfected cell, the epidermis was stripped fromN. sylvestris at various times after inoculationwith TMV; this showed that virus moved intothe mesophyll in 4 hours at 24–30 o C. Tobaccorattle virus microinjected into trichome cells ofN. clevelandii took about 4 hours to move out ofinoculated cells. The rates of subsequent cellto-cellspread vary with leaf age, with differentcell types, and in different directions withinthe leaf. The measured rates range from 4 to 13mm/hour (roughly one cell every 1 to 3 hours).There are generally fewer plasmodesmataper unit area on the vertical walls of mesophyllcells than on the walls that are more or lessparallel to the leaf surface. Furthermore, theretend to be lines of mesophyll cells linked efficientlytogether and ending in contact with aminor vein. Viruses may spread more rapidlyalong such routes than in other directionswithin the mesophyll.C. Systemic Movement1. Steps in Systemic MovementThe long-distance transport is through theplant vascular tissue, usually the phloemsieve-tubes (see Box 2.2 for plant structure;details of vein structure are given in Box 9.3),following the flow of metabolites from thesource leaves to the sink leaves or tissues. Then,further cell-to-cell movement establishes systemicinfection of the young leaves.There are five distinct and consecutive stepsfrom localised cell-to-cell movement of a virusto full systemic invasion of the plant (Figure 9.5).1. The first event of the movement of a virusinto the vascular system is to pass thebundle sheath cells. It is likely that virusesenter the bundle sheath cells from themesophyll by the cell-to-cell movementprocesses just described. However, there isstrong evidence that a different process isIII. HOW DO PLANT VIRUSES WORK?

BOX 9.3VEIN AND PHLOEM STRUCTUREVeins are defined as minor or major, based on their structure, location, branching pattern, and function. Fordicotyledons, major veins are enclosed in parenchyma tissue, forming a rib rising above the leaf surface andbranch, usually no more than twice. They function in long-distance transport of water, inorganic nutrients,photoassimilates, and other organic matter and are thought to be involved with photoassimilate unloading.Minor veins that are are embedded in mesophyll cells are the result of three or more branchings from the firstorderveins and function in loading of photoassimilates in mature leaves. Veins have also been divided intofive classes, classes I to III being major veins and IV and V being minor veins (Fig.). The difference betweenmajor and minor veins is less apparent in monocotyledons. The structure of veins and vein-associated cellschanges as the leaf develops from a sink to a source. This can have a significant effect on virus movement.ABCFig. See legend on next pageIII. HOW DO PLANT VIRUSES WORK?

178 9. VIRUS-HOST INTERACTIONS — PLANT LEVELBOX 9.3(continued)Fig. Veins in plant leaves. A. Major veins in a leaf. B. Enlargement of boxed area in A showing minor veins.C. Schematic model of sink source transition in a developing leaf (enlargement of boxed area in B). The blueregion indicates the exporting (source) part of the leaf and the yellow the importing (sink) part. The numbersshow the vein class and the arrows the direction of movement of photoassimilates. [From Roberts et al. (1997; PlantCell 9, 1381–1396).]The vascular tissue is surrounded by the bundle sheath and comprises parenchyma (in majorveins), sieve tubes with companion cells, and xylem elements (see Box 2.2).FIGURE 9.5 Cellular route for systemic movement of plant viruses. Leaf veins used by viruses for systemic transportshown in red and yellow; those not used shown in green. Blue dotted lines indicate cellular boundaries that viruses areunable to cross—for example, sink/source transition zones and apical meristem. 1 and 2. Viral infection initiates withmechanical damage (red jagged arrow) of mesophyll cells of a lower source leaf. The virus spreads from cell to cell andreaches the vascular system, where it is trafficked toward systemic organs but not locally. 3. To enter the phloem, the virusmust cross from mesophyll cells (ME) through bundle sheath cells (BS) and phloem parenchyma (VP) into companion cells(CC) and then into the sieve elements (SE) of the phloem. This movement involves viral MPs (blue arrows) and is throughplasmodesmata (pink circles). Additional factors (black arrows) are required for movement BS > VP > CC > SE. 4 and 5.Once within the SEs, virus passes out of the inoculated leaf and spreads rapidly upward using the internal, adaxial (red)phloem, and slowly downward in the external, abaxial (yellow) phloem. 6, 7, and 8. Virus enters the sink, but not thesource, zones, unloading only from major veins (classes 1–3, indicated in red). 9. Virus does not enter the apical meristem.[From Waigmann, E., Ueki, S., Trutnyeva, K., et al. (2004). The ins and outs of nondestructive cell-to-cell and systemic movementof plant viruses. Crit. Rev. Plant. Sci. 23, 195–250, reprinted by permission of the publisher (Taylor & Francis Ltd.,http://www.tandf.co.uk/journals).]III. HOW DO PLANT VIRUSES WORK?

I. MOVEMENT AND FINAL DISTRIBUTION179required for the infection unit to pass out ofthe bundle sheath cell into the vascularparenchyma; little is known about thisprocess. Luteoviruses are introduced into thevascular system by their insect vector and donot move into mesophyll cells. However, asjust mentioned, this lack of movement can becomplemented by a coinfection with a viruscapable of crossing the bundle sheath. Thus,the bundle sheath forms a major boundaryfor systemic virus movement.2. Having crossed the vascular parenchyma, thevirus reaches the vascular parenchyma/companion cell boundary. The companioncells supply the enucleate sieve elements withmost cellular maintenance functions throughspecialised plasmodesmata. There is evidencethatsome,butnotall,viralmovementproteins facilitate crossing this boundary.3. When the infection unit penetrates the sieveelement tubes of inoculated leaves, it canmove throughout the plant. In many plantspecies there are two structural types ofphloem: external (abaxial, on the undersideof the leaf), which transport metabolitesdownward to the roots, and internal(adaxial, on the upper side of the leaf),which transport to the upper leaves andapex of the plant. Many viruses aretransported in both directions, slowlydownward in the external phloem andrapidly upward in the internal phloem.4. Virus exit from the phloem is not simply areverse process of the entry into the phloembut is probably by a different mechanism.For example, TMV loads into both majorand minor veins in the source leaves butunloads only from the major veins (class IIIand larger; see Box 9.3) in the sink leaves ofN. benthamiana.5. It is likely that the passage of virus across thebundle sheath cell of the importingleaf involves the same mechanisms asthose in the exporting leaf. Thus, luteovirusesare retained within the vascular tissue.2. Form in Which Virus Is TransportedMost viruses require coat protein for longdistancetransport, but some, like Barley stripemosaic virus (BSMV), are capable of long-distancemovement without a coat protein; BSMVforms a nucleoprotein complexed with TGB1nonstructural protein. It is not clear if all theviruses that require coat protein for longdistancetransport move in the sieve elementsas virus particles. There is good evidence forvirus particle involvement for some such asTMV, but for others it is difficult to distinguishbetween a requirement of coat protein for cellto-cellmovement necessary for phloem loadingand after unloading, from that for actual sieveelement transport.Other viral proteins are important for vascular-dependentaccumulation, including theMP of TMV, the CMV gene 2b, and the HC-Pro of potyviruses. However, care must betakeninattributingproteinfunctiondirectlyto long-distance spread, as they may beinvolved in, for example, suppression of hostdefense mechanisms.3. Rate of Systemic MovementThe time at which infectious material movesout of the inoculated leaf into the rest of the plantvaries widely, depending on such factors as hostspecies and virus, age of host, method of inoculation,and temperature. After transmission byaphids, Barley yellow dwarf virus may move outof the inoculated leaf within 12 hours. TMVmoves out of tobacco leaves 32–48 hours aftermechanical inoculation, and Cucumber mosaicvirus can spread systemically 24–30 hours afterinoculation. In the classic experiments of Samuel(Figure 9.1), one terminal leaflet of a tomatoplant was inoculated with TMV and the spreadof virus with time then followed by cutting upsets of plants into many pieces at various times.He incubated the pieces to allow any smallamount of virus present to increase and thentested for the presence of virus by infectivity.III. HOW DO PLANT VIRUSES WORK?

180 9. VIRUS-HOST INTERACTIONS — PLANT LEVELVirus moved first to the roots and then to theyoung leaves. It was some time before the middle-agedand older leaves became infected. Invery young plants, older leaves did not becomeinfected, even after several months.Once virus enters the phloem, movementcan be very rapid. Values of about 1.5–8 cm/hour have been recorded for TMV. However,when examined in detail, systemic invasion bya virus is affected by a complex of factors andespecially the source-sink status of individualleaves and parts of leaves.The time at which unrelated viruses movefrom the inoculated leaf of the same individualhost plant may be different. For example, intobacco plants inoculated with a mixture ofPotato virus X (PVX) and Potato virus Y (PVY),PVY moved ahead of PVX and could beisolated alone from the tip of the plant.4. Movement in the XylemFew viruses move long distance through thexylem. For example, Southern bean mosaic virus(SBMV) and other sobemoviruses move as virusparticles through dead tissue, which implicatesxylem vessels. For the sobemoviruses, movementin the xylem has been correlated with transmissionby beetles.D. Final Distribution in the PlantWhether or not a virus will move systemicallyat all depends on events in or near thelocal infection. If a virus forms a local lesion,it may be retained to only a small group ofcells. However, some viruses “escape” fromthe lesions and spread systemically. It is oftenassumed that viruses that do move systemicallybecome fairly evenly distributed throughoutthe plant, but, in fact, this seldom happens.Several factors can result in a very unequal distribution,including vascular connectionsbetween the initially infected leaf and the restof the plant, host genes, viral genes, the hostdefense system and environmental conditions;these and other factors affect the rate andextent of virus movement through the plant.A common situation is shown in Figure 9.1,where the first systemically infected leaf isabove the inoculated leaf (the same phyllotaxis),and the young sink leaves become systemicallyinfected before older leaves to whichthe virus spreads by cell-to-cell movement. Asleaves undergo sink-source transition in photoassimilateimport, there is a progressive basipetaldecline in the amount of photoassimilateand virus entering the lamina so that in moremature sink leaves only the base of the leafbecomes infected.Many of the viruses giving a general systemicinfection apparently recover from infectionin newly produced young leaves or evengo through cycles of recovery and reinfection(see Chapter 11). In a plant that is infected witha systemic virus for a length of time, the concentrationof virus may not be uniform in differentorgans. With most mosaic-type virusesthat have been investigated, virus reaches amuch higher concentration in the leaf laminathan in other parts of the plant.The distribution of some viruses is limited tocertain tissues. Luteoviruses and other phloemlimitedviruses are found usually only inphloem parenchyma, companion cells, andsieve elements. Dark green areas in the mosaicpatterns of diseased leaves usually contain verylittle virus compared with yellow or yellowgreenareas. This has been found for virusesthat differ widely in structure and that infectboth mono- and dicotyledonous hosts. Thephenomenon may therefore be a fairly generalone for diseases of the mosaic type. Possiblereasons for the low concentration of virus indark green areas are discussed in Chapter 11.Most viruses do not invade apical meristematictissues. With many virus-host combinations,there appears to be a zone of variablelength (usually about 100 mm but up to 1,000mm) near the shoot or root tip that is free ofvirus or that contains very little virus. It seemsIII. HOW DO PLANT VIRUSES WORK?

II. EFFECTS ON PLANT METABOLISM181likely that the plant defense system of genesilencing (see Chapter 11) is enhanced in meristematiccells, and this inhibits the presenceand replication of the virus.E. Outstanding Questions on PlantVirus MovementHere are some of the questions that remainto be answered about cell-to-cell movement ofviruses:How do MPs gate plasmodesmata open atthe infection front?What are the forms of infection unit thatpass through plasmodesmata, and whatare the forces that move them from theinfected to the uninfected cell?How do plasmodesmata close or return tonormal signaling control after the passageof the infection unit?Do viruses pass through existingplasmodesmata, or are new onesformed? If they pass through existingplasmodesmata, are certain onesamenable to gating?What are the features of plasmodesmata thatdetermine tissue specificity andsymplastic domains?There are also many questions to be answeredabout the details of long-distance transport:What are the exact routes of movement fromthe phloem parenchyma to the sieveelements? Are there different routesspecific for individual viruses orgroups of viruses? If so, what arethe factors that determine individualroutes?What is the sieve element route from thesource leaf to the sink leaf? As notedearlier in this chapter, there is someevidence for switching from the externalphloem from the source leaf to theinternal phloem leading to the sink leaf.Does this occur for all systemicallymoving viruses in all plant species?What are the forms in which the infectionunits are transported? Many virusesrequire coat protein for long-distancetransport, but do they form particles?What are the factors that lead to the infectionunits being unloaded from the sieveelements?These last two questions are difficult to answer,as it may be only a minority of the infectionunit that is unloaded and establishes systemicinfection. Furthermore, for individual virusesthere may be differences in the preferredunloading (and loading) form under differentconditions.II. EFFECTS ON PLANTMETABOLISMAll of the various macroscopic and microscopicsymptoms of disease that are discussedin Chapter 2 must originate from biochemicalaberrations induced by the virus.A. Nucleic Acids and ProteinsIt is widely assumed that the small RNA viruseshave little effect on host-cell DNA synthesis,but there are very few, if any, definitiveexperiments addressing this question. However,they do have an effect on ribosomal RNA synthesisand the concentration of ribosomes thatdiffers with the virus, strain of virus, time afterinfection, and the host and tissue concerned. Inaddition, 70S and 80S ribosomes may be affecteddifferently. For example, in Chinese cabbageleaves that are chronically infected with Turnipyellow mosaic virus (TYMV), the concentration of70S (chloroplast) ribosomes in the yellow-greenislands in the mosaic is greatly reduced comparedto that in dark green islands in the sameleaf; there is little effect on the concentration ofIII. HOW DO PLANT VIRUSES WORK?

182 9. VIRUS-HOST INTERACTIONS — PLANT LEVELcytoplasmic ribosomes in such yellow-greenislands of tissue. The extent of this reductiondepends on the strain of TYMV, and it alsobecomes more severe with time after infection.Loss of 70S ribosomes more or less parallels theloss of chlorophyll, with “white” strains causingthe most severe loss.A somewhat different result is obtained if theeffect of TYMV infection with time in a youngsystemically infected leaf is followed. Chloroplastribosome concentration falls markedlyas the virus concentration reaches a maximum.About the same time, there is a significantincrease in cytoplasmic ribosome concentration,which is mainly due to the stunting effect ofinfection. On the other hand, if the effectsof virus infection on these components for theplant as a whole are considered, a differentpicture emerges. Infection reduces both cytoplasmicand chloroplast ribosomes.The coat protein of a virus such as TMV cancome to represent about half the total proteinin the diseased leaf. This can occur withoutmarked effects on the overall content of hostproteins. Most other viruses multiply to a muchmore limited extent. Effects on host protein synthesisare not necessarily correlated withamounts of virus produced. A reduction in theamount of the most abundant host protein—ribulose bisphosphate carboxylase-oxygenase(rbcs or rubisco)—is one of the most commoneffects of viruses that cause mosaic and yellowingdiseases.TMV infection has been estimated to reducehost protein synthesis by up to 75 percent duringthe period of virus replication. Infectiondoes not alter the concentration of host polyadenylatedRNA or its size distribution, whichsuggests that infection may alter host proteinsynthesis at the translation stage rather thaninterfering with transcription.Virus infection can result in some host genesbeing shut-off, some induced, and others notaffected (Box 9.4). In this, plant viruses resemblein some respects various animal virusesbut differ in that all three phenomena are associatedwith the same virus. These observationsraise many questions. Does the virus shut-offthe expression of genes that could be considered“competitive” but enhance others thatmay be helpful? Is this phenomenon restrictedto cotyledons, or is it a general mechanism atthe infection front throughout the plant? If itis general, could it be associated with symptomexpression?B. LipidsThe sites of virus synthesis within the cellalmost always contain membrane structures(see Box 9.1). TYMV infection alters the ultrastructureof chloroplast membranes, and rhabdovirusand tospovirus particles obtain theirouter membrane by budding through somehost-cell membrane.C. CarbohydratesSome viruses appear to have little effect oncarbohydrates in the leaves, whereas othersmay alter both their rate of synthesis and rateof translocation. These changes may be illustratedin a simple manner.Leaves of Cucurbita pepo that have beeninoculated several days previously withCucumber mosaic virus that does not causenecrotic local lesions are harvested in the morningor after some hours in darkness, decolourised,and treated with iodine, which stainsstarch. Cryptic local lesions may show up asdark-staining areas against a pale background,indicating a block in carbohydrate translocation.On the other hand, if the inoculatedleaves are harvested in the afternoon after aperiod of photosynthesis, decolourised, andstained with iodine, local lesions, termed starchlesions, may show up as pale spots againstthe dark-staining background of uninfectedtissue. Thus, virus infection can decrease therate of accumulation of starch when leaves areIII. HOW DO PLANT VIRUSES WORK?

II. EFFECTS ON PLANT METABOLISM183BOX 9.4EFFECT ON CELLULAR GENES OF INFECTION WITHPEA SEED-BORNE MOSAIC VIRUSIn a study on the mRNAs of various host genes at the infection front of Pea seed-borne mosaic virus inpea cotyledons (Fig.), three situations were identified:ABCDEFFig. Induction and shut-off of cellular genes associated with Pea seed-borne mosaic virus (PSbMV) infection identifiedby in situ hybridization on near-consecutive section of PSbMV-infected pea tissue. Positive detection of specificRNAs is shown by dark staining. A. Diagrammatic representation of area of pea seed with box indicatingregion shown in other panels. B. Detection of genomic viral RNA. C. Detection of (–)-strand viral RNA showingregions where virus is replicating. D–F. Detection of HSP70, polyubiquitin, and lipoxigenase RNAs, respectively.Arrow in F shows recovery from shut-off (restoration of lipoxigenase expression) within the affected area. [Thisarticle was published in Virology, 243, M. Aranda and A. Maule, Virus-induced host gene shutoff in animals and plants,pp. 261–267, Copyright Elsevier (1998).]1. Inhibition of host gene expression. The expression of at least 11 host genes was suppressed—forexample, lipoxigenase 1 (Fig., panel F).2. Induction of host gene expression. The expression of the heat shock protein HSP70 and ofpolyubiquitin was induced in association with viral replication (Fig. panels D and E,respectively).3. No effect on host gene expression—for example, actin and ß-tubulin.Probing the infection front with antibodies showed that host protein accumulation was also affectedin a similar manner.III. HOW DO PLANT VIRUSES WORK?

184 9. VIRUS-HOST INTERACTIONS — PLANT LEVELexposed to light. A detailed analysis of infectedC. pepo cotyledons showed that activity of someenzymes (e.g., sucrose synthase and ATPdependentphosphofructokinase) rose, that ofother enzymes (e.g., ADP glucose pyrophorylaseand rubisco) dropped, and that of others(e.g., chloroplastic fructose bisphosphataseand hydroxypyruvate kinase) were not affectedat all.Infection with some viruses, such as Beet yellowsvirus, can induce damage to the phloemthat restricts translocation of photoassimilates.These accumulate in the leaf lamina and possiblygive rise to some of the symptoms like yellowing,thickened leaves, and leaf rolling (seeProfile 5).From the few diseases that have been examinedin any detail, it is not possible to makevery firmly based generalisations about othercarbohydrate changes, but the following maybe fairly common effects:• A rise in glucose, fructose, and sucrose invirus-infected leaves• A greater rise in these sugars caused by mildstrains of a given virus compared withsevere strains• Effects of infection on mesophyll cells,not yet understood, may reducetranslocation of carbohydrates out of theleaves.D. PhotosynthesisVirus infection usually affects the process ofphotosynthesis, with reduction in carbon fixationbeing the most common effect in leavesshowing mosaic or yellows diseases. This reductionusually becomes detectable some days afterinfection.Photosynthetic activity can be reduced bychanges in chloroplast structure, reduced contentof photosynthetic pigments or rubisco, orreduction in specific proteins associated withthe particles of photosystem II. However, suchchanges appear to be secondary, occurringsome time after infection when much virus synthesishas already taken place.Overall, during the period of rapid virusreplication, infection may cause a diversion ofthe early products of carbon fixation away fromsugars and into pathways that lead moredirectly to the production of intermediates forthe synthesis of nucleic acids and proteins.The most general result of virus infection is areduction in photosynthetic activity, but theremay be no overall effect in relatively earlystages of infection. Any reduction in photosynthesisis likely to arise from a variety of biochemicaland physical changes. The relativeimportance of different factors varies with thedisease.E. RespirationFor many host-virus combinations wherenecrosis does not occur, there is a rise in respirationrate, which may begin before symptomsappear and continue for a time as diseasedevelops. In chronically infected plants, respirationis often lower than normal. In host-viruscombinations where necrotic local lesionsdevelop, there is an increase in respiration asnecrosis develops. This increase is accountedfor, at least in part, by activation of the hexosemonophosphate shunt pathway.F. TranspirationIn chronically virus-infected leaves, transpirationrate and water content are generallylower than in corresponding healthy tissues.The reported effects over the first one to twoweeks after inoculation vary.G. Low-Molecular-Weight CompoundsThere are numerous reports on the effectsof virus infection on concentration of lowmolecular-weightcompounds in various partsIII. HOW DO PLANT VIRUSES WORK?

III. PROCESSES INVOLVED IN SYMPTOM INDUCTION185of virus-infected plants. The analyses give riseto large amounts of data, which vary with differenthosts and viruses and are impossible tointerpret in relation to virus replication. Thefollowing are some of the effects:• A consistent increase in one or both of theamides, glutamine, and asparagine.A general deficiency in soluble nitrogencompounds compared with healthy leavesmay occur during periods of rapid virussynthesis.• Phosphorus is a vital component of allviruses and as such may come to representabout one-fifth of the total phosphorus intheleaf.Inspiteofthis,thereislittleornoeffect of infection on host phosphorusmetabolism.• Virus infection frequently involves yellowmosaic mottling or a generalised yellowingof the leaves. Such changes are obviouslydue to a reduction in leaf pigments,such as carotene and xanthophyll. Thereduction in amount of leaf pigmentscan be due either to an inhibition ofchloroplast development or to thedestruction of pigments in maturechloroplasts. The first effect probablypredominates in young leavesthat are developing as virus infectionproceeds.• Virus infection usually appears to affect onlythe vacuolar anthocyanin pigments inflowers. The pigments residing inchromoplasts may not be affected.In summary, the physiological and biochemicalchanges most commonly found invirus-infected plants are a decrease in rate ofphotosynthesis, often associated with a decreasein photosynthetic pigments, chloroplastribosomes, and rubisco; an increase in respiratoryrate; an increase in the activity of certainenzymes, particularly polyphenoloxidases; anddecreased or increased activity of plant growthregulators.Many of the changes in host plant metabolismare probably secondary consequences ofvirus infection and not essential for virus replication.A single gene change in the host, or asingle mutation in the virus, may change analmost symptomless infection into a severe disease.Furthermore, metabolic changes inducedby virus infection are often nonspecific. Similarchanges may occur in disease caused by cellularpathogens or following mechanical or chemicalinjury. In many virus diseases, the generalpattern of metabolic change appears to resemblean accelerated aging process. Because ofthese similarities, rapid diagnostic proceduresbased on altered chemical composition of thevirus-infected plant must be used with considerablecaution.III. PROCESSES INVOLVED INSYMPTOM INDUCTIONThere are various processes involved in theinduction of disease. At the present time weare unable to implicate specific virus- or hostcodedproteins with the initiation of any ofthese processes.A. Sequestration of Raw MaterialsThe diversion of supplies of raw materialsinto virus production, thus making host cellsdeficient in some respect, is an obvious mechanismby which a virus could induce diseasesymptoms. Sometimes an increase in severityof symptoms is associated with increased virusproduction. However, in well-nourishedplants, there is no general correlation betweenamount of virus produced and severity of disease.Except under conditions of specific preexistingnutritional stress, it is unlikely that theactual sequestration of amino acids and othermaterials into virus particles has any directconnection with the induction of symptoms.III. HOW DO PLANT VIRUSES WORK?

186 9. VIRUS-HOST INTERACTIONS — PLANT LEVELThe rate of virus replication in individualcells could be an important factor in influencingthe course of events. Very high demand for keyamino acids or other materials over a very shortperiod, perhaps of a few hours, could lead toirreversible changes with major long-termeffects on the cell and subsequently on tissuesand organs. Another consideration is the possibilitythat the “switching off” of host genes atthe infection front (see Box 9.3) is a general phenomenonand that this could lead to depletionof key proteins at an important time.B. Effects on GrowthVirus infection has various effects on thegrowth of plants. The stunting of growth couldbe due to a change in the activity of growthhormones, a reduction in the availability ofthe products of carbon fixation by direct effectson chloroplast structure or translocation offixed carbon, and a reduction in uptake ofnutrients.Most plant viruses belonging to the Reoviridaeinduce galls or tumours in their plant hostsbut not in the insect vectors, in which they alsomultiply. There is a clear organ or tissue specificityfor the different viruses. For example,tumours caused by Wound tumor virus (WTV)predominate on roots and, to a lesser extent,stems. We can be reasonably certain that somefunction of the viral genome induces tumourformation, but we are quite ignorant as tohow this is brought about. Wounding playsan important role as an inducer or promoterof tumours caused by WTV. Hormonesreleased on wounding may play some part inthis process.In leaves showing mosaic disease, the darkgreen islands of tissue frequently show blisteringor distortion. This is due to the reduced cellsize in the surrounding tissues and the reducedsize of the leaf as a whole. The cells in the darkgreen island are much less affected and maynot have room to expand in the plane of thelamina. The lamina then becomes convex orconcave to accommodate this expansion.C. Effects on ChloroplastsTYMV infection in Chinese cabbage causesclumping of chloroplasts and some otherchanges, such as the formation of large vesicles(sickling) and fragmentation. In the sickling processa large, clear vesicle appears in chloroplasts,the chlorophyll-bearing structures being confinedto a crescent-shaped fraction of the chloroplastvolume. The vesicle is bounded by amembrane that appears to arise from stromalamellae. Red and blue light are equally effectiveinducers of sickling, which does not occur in thedark.D. Mosaic SymptomsMosaic symptoms are very common in plantvirus infections. They only develop in sinkleaves to which the virus has spread by longdistancemovement. There are various factorsinvolved in mosaic symptoms.Virus strain. In some infections, such as TMVin tobacco, the disease in individual plantsappears to be produced largely by a singlestrain of the virus. However, occasional brightyellow islands of tissue in the mosaic containdifferent strains of the virus, which probablyarise by mutation and, during leaf development,come to exclude the original mild strainfrom a block of tissue.Leaf age at time of infection. By inoculatingplants at various stages of growth, it has beendemonstrated for TYMV and TMV that thetype of mosaic pattern that develops in a leafat a particular position on the plant dependsnot on its position but on its stage of developmentwhen infected by the virus. There is a criticalleaf size at the time of infection abovewhich mosaic disease does not develop. Thiscritical size is about 1.5 cm (length) for tobaccoleaves infected with TMV.III. HOW DO PLANT VIRUSES WORK?

III. PROCESSES INVOLVED IN SYMPTOM INDUCTION187Gradients in size of colour islands. Although thesize of macroscopic islands in a mosaic is veryvariable, there tends to be a definite gradientup the plant. Leaves that were younger whensystemically infected tend to have a mosaicmade up of larger islands of tissue. Even withinone leaf there may be a relationship betweenboth the number and size of islands and theage of different parts of the lamina as determinedby frequency of cell division.Patterns in the macroscopic mosaic. Patterns inthe macroscopic mosaic are often so jumbledthat an ontogenetic origin for them cannot bededuced. Occasionally, patterns that are clearlyderived from the apical initials have beenobserved. For example, Chinese cabbage leaves(or several successive leaves) have been observedto be divided about the midrib intotwo islands of tissue—one dark green and theother containing a uniform virus infection.The microscopic mosaic. The overall mosaic patternmay be made up of microscopic mosaics. InChinese cabbage leaves infected with TYMV,areas in the mosaic pattern that macroscopicallyappear to be a uniform colour may be found onmicroscopic examination to consist of mixed tissuein which different horizontal layers of themesophyll have different chloroplast types.A wide variety of mixed tissues can be found.For example, in some areas, both palisade andthe lower layers of the spongy tissue may consistof dark green tissue, whereas the central zoneof cells in the lamina is white or yellow green.This situation may be reversed, with the centrallayer being dark green and the upper and lowerlayers consisting of chlorotic cells. These areas ofhorizontal layering may extend for several millimeters,or they may be quite small, gradingdown to islands of a few cells or even one cellof a different type. The junction between islandsof dark green cells and abnormal cells in themicroscopic mosaic is often very sharp.Genetic control of mosaics. In spite of the factthat the most striking effects of TYMV infectionin Brassica spp. are on the chloroplasts, theresponse of these organelles to infection isunder some degree of nuclear control. Certainvarieties of Brassica rapa respond to all thestrains isolated from B. chinensis with a milddiffuse mottle. In reciprocal crosses betweenB. rapa and B. chinensis, all the progeny gave aB. chinensis type of response to the strains.The nature of dark green tissue. Dark greenislands in the mosaic pattern are cytologicallyand biochemically normal as far as has beentested. They contain low or zero amounts ofinfectious virus and no detectable viral proteinor viral dsRNA. Dark green islands are resistantto superinfection with the same virus orclosely related viruses.Various factors can influence the proportionof leaf tissue that develops as green islands ina mosaic. These include leaf age, strain of virus,season of the year, and removal of the lowerleaves on the plant. The dark green islands oftissue may not persist in an essentially virusfreestate for the life of the leaf. “Breakdown”leading to virus replication usually takes placeafter a period of weeks or after a sudden elevationin temperature. Dark green islands are furtherdiscussed in Chapter 11.E. Role of MembranesTypical leaf cells are highly compartmentalised,and there is an increasing awareness of thevital roles that membranes play in the functioningof normal cells. Virus replication involves theinduction of new or modified membrane systemswithin infected cells. Both animal and bacterialviruses alter the structure and permeability propertiesof host cell membranes. Infection of insectvector cells by plant viruses may alter the physicalproperties of the plasma membrane.• Various virus-induced responses, some ofwhich were discussed in earlier sections,indicate ways in which changes in membranefunction may have far-reaching consequencesfor the plant: Virus infection may affectIII. HOW DO PLANT VIRUSES WORK?

188 9. VIRUS-HOST INTERACTIONS — PLANT LEVELstomatal opening (stomata are the breathingpores in plant leaves). Net photosynthesis maybe limited by stomatal opening, which in turndepends on guard cell membrane function.• A virus-induced increase in stomataresistance leads to increased internal ethyleneconcentration followed by an epinasticresponse. The change in stomatal opening isprobably induced by changes in properties ofthe guard cell plasma membranes.• Sucrose accumulation in infected leavesappears to depend on membranepermeability changes rather than alteredphloem function.• Virus-induced starch accumulation inchloroplasts may be due in part at least toalterations in permeability of chloroplastmembranes.• There is little doubt that ethylene is involvedin the induction of some symptoms of virusdisease.• Living cells maintain an electrochemicalpotential difference across their plasmamembrane, which is internally negative.There is an interdependence between thispotential difference and ion transport acrossthe membrane. Infection of Vigna sinensiscells by Tobacco ringspot virus alters theirtransmembrane electropotentials.IV. OTHER KINGDOMSIn the previous chapters we noted the similaritiesbetween plant viruses and those ofother kingdoms in the expression and replicationof their genomes. However, there aremajor differences between how plant virusesand those of other kingdoms move aroundtheir host, which reflect differences in the cellularstructure of the organism. As shown inBox 9.2, plant cells are connected by cytoplasmiclinks (plasmodesmata), whereas thoseof animals and bacteria are separate beingsurrounded by a liquid medium (see alsoBox 2.2). Thus, as far as a virus is concerned,a plant can be considered as a single cell,whereas an animal or bacterium is a collectionof cells in which the virus has to passacross the outer cell membrane into the liquidmedium and enter through the cell membraneof the next cell.There are similarities and differences in theeffects that plant viruses have on the hostmetabolism when compared with animal andbacterial viruses. These reflect similarities inthe basic host metabolism, such as synthesis ofnucleotides and amino acids, and in the translationmachinery. Most of the differences focus oncarbohydrate production, with plants fixing carbonthrough photosynthesis and animals andbacteria acquiring carbohydrates from externalsources.There are basic differences in the processesinvolved in symptom production, again reflectingdifferences in the hosts. In plants many ofthe symptoms are associated with metabolicperturbations or damage to cell organelles,especially chloroplasts. In animals and bacteriathere is some metabolic perturbation, but manyof the symptoms are associated with damage tothe cells themselves.V. SUMMARY• The three phases in the establishment offull systemic infection of plants areintracellular movement, intercellularmovement, and systemic movement of theinfection unit.• Intracellular movement of the infectionunit involves the cell membrane andcytoskeleton system and is integrated tovirus replication.• Intercellular movement is via cytoplasmicconnections through the cell wall(plasmodesmata) and involves virusencodedmovement proteins (MPs). The MPsIII. HOW DO PLANT VIRUSES WORK?

V. SUMMARY189either temporally increase the size exclusionlimit of the plasmodesmata or form tubulesthrough the plasmodesmata.• The five steps in systemic movement aremovement across the bundle sheath into thevascular system, movement into thecompanion cells, movement into the phloemsieve tubes (or xylem vessels for a fewviruses), exiting from the phloem, andmovement in the newly infected leaf fromthe vascular system across the bundle sheathinto the mesophyll.• The movement across the bundle sheath andinto and out of the companion cells involvesmodifying various types of plasmodesmata.The movement in the phloem follows thesource-sink flow of normal plant metabolites(e.g., carbohydrates).• The final distribution of a virus in a plantusually does not involve all tissues.Frequently, viruses are excluded frommeristematic tissues.• Virus infections have considerable effects oncell metabolism such as photosynthesis,respiration, and transpiration. The effectsinclude increasing the activity of someenzymes, decreasing the activity of others,and not affecting the activity of yet others.• Symptom induction is primarily by theperturbation of the cell metabolism anddamage to cell organelles such aschloroplasts.Further ReadingHarries, P.A. and Nelson, R.S. (2008). Movement of virusesin plants. Encyclopedia of Virology, Vol. 3, 348–355.Hull, R. (2002). Matthew’s plant virology. Academic Press,San Diego.Lucas, W.J. (2006). Plant viral movement proteins: Agentsfor cell-to-cell trafficking of viral genomes. Virology344, 169–184.Verchot-Lubicz, J., Ye, C-M., and Bamunusinghe, D. (2007).Molecular biology of potexviruses: Recent advances.J. Gen. Virol. 88, 1643–1655.Waigmann, E., Ueki, S., Trutnyeva, K., and Citovsky, V.(2004). The ins and outs of nondestructive cell-to-celland systemic movement of plant viruses. Crit. Rev. PlantSci. 23, 195–250.III. HOW DO PLANT VIRUSES WORK?

This page intentionally left blank

C H A P T E R10Virus-Plant Interactions:1. Molecular LevelThe ultimate interactions between a virus and its host occur at the molecular level where thevirus genome meets the host genome. It is these interactions that dictate the success of the virusinfection and the symptoms that result.O U T L I N EI. Introduction 191II. Host Response to Inoculation 192III. Interactions Between Viruses 202IV. Viruses of Other Kingdoms 204V. Summary 204I. INTRODUCTIONThe application of methods based on DNAtechnology for studying the role of viral geneproducts in disease induction is providingincreasing amounts of information on the interactionsinvolved in the full virus infection of aplant. Important procedures include the constructionof infectious clones of viruses, sitedirectedmutagenesis of the viral genome,switching genes between viruses and virusstrains, the construction of transgenic plants thatexpress only one or a few viral genes, and theisolation and sequencing of genes from hostplants, especially Arabidopsis thaliana. Even moreimportant are the recent conceptual advances inthe interplay between the viral and host genomes.Essentially, the host is attempting torestrict the virus infection and the virus isattempting to overcome these restrictions.Comparative Plant Virology, Second Edition191Copyright # 2009, Elsevier Inc. All rights reserved.

192 10. VIRUS-PLANT INTERACTIONS: 1. MOLECULAR LEVELII. HOST RESPONSES TOINOCULATIONThe term host is defined in Chapter 2, but theterms for describing the various kinds ofresponses to inoculation with a virus thatplants display have been used in ambiguousand sometimes inconsistent ways. The currentdefinitions used by plant virologists are givenin Table 9.1; some of these differ from those inother branches of virology. For example, latentused in reference to bacterial or vertebrateviruses usually indicates that the viral genomeis integrated into the host genome. There arevery few plant viruses that integrate into thehost genome, and the integrant is activated togive an episomal infection (Box 8.9); unlike bacterialand vertebrate integrating viruses, integrationis not an essential part of the plantviral replication cycle. Thus, the three basicresponses of a plant to inoculation with a virusare total immunity; nonpermissive infection, inwhich the plant reacts to contain the virusinfection; and permissive infection, in whichthe plant does not contain the infection.It had been assumed for many years thatvirus-host cell interactions must involve specificrecognition or lack of recognition between hostand viral macromolecules. Interactions mightinvolve activities of viral nucleic acids, or specificvirus-coded proteins, or host proteins thatare induced or repressed by viral infection. Theeffects of virus infection on host proteins are discussedin Chapter 9. An increasing understandingof interactions involved in the induction ofdisease is accruing from the study of variousforms of virus resistance and the realisation ofhow similar some of these are to responses toattack by other microbes and by pests.Natural resistance to viruses in plants is conferredby two types of genes: dominant genesand recessive genes (Table 10.1), as well as theinherent defence against foreign nucleic acidsdiscussed in Chapter 11. The majority of dominantresistance genes are monogenic andresemble genes that confer resistance to otherpathogens in the “gene-for-gene” process inwhich the host R gene interacts with the pathogen’sAvr gene (Box 10.1). Most recessiveresistance genes interfere with the virus replicationcycle, preventing the expression orreplication of the viral genome.A. ImmunityMost plant species are resistant to most plantviruses; this is termed nonhost resistance.TABLE 10.1Plant Resistance Gene to VirusesHostVirusGene Species R type a Genotype Virus b Avr Type of resistanceRx1 Potato CC-NBS-LRR Dominant PVX Coat protein ImmunityTm-1 Tomato - Dominant ToMV Replicase ImmunityN Tobacco TIR-NBS-LRR Dominant TMV Replicase/helicase HRTm-2 Tomato CC-NBS-LRR Dominant ToMV Movement protein HRN’ Tobacco - Dominant TMV Coat protein HRPat-1 Lycopersicum spp. Translation factoreIF4ERecessive PVY VPg Immunitya See Box 10.2 for types of R gene.b Virus abbreviations: PVX, Potato virus X; PVY, Potato virus Y; TMV, Tobacco mosaic virus; ToMV, Tomato mosaic virus.III. HOW DO PLANT VIRUSES WORK?

II. HOST RESPONSES TO INOCULATION193BOX 10.1FLOR’S GENE-FOR-GENE MODEL FOR RESISTANCEOF A PLANT TO A PATHOGEN OR PESTIn his studies on the inheritance of the resistanceof flax (Linum usitatissimum) to the flax rust pathogen(Melampsora lini), Flor (1971) revealed theclassic “gene-for-gene” model in which the hostresistance (R) gene interacts with the pathogen’savirulence (Avr) gene.The model proposes that for resistance tooccur, complementary pairs of dominant genes,one in the host and the other in the pathogen,are required. A loss or alteration in the host Rgene or in the pathogen Avr gene leads to diseaseor compatibility (Fig). Basically, the interactionbetween the R and Avr genes leads to both a localand systemic signal cascade. The local signallingcascade triggers a host response that containsthe pathogen infection to the primary site; the systemiccascade primes defence systems in otherparts of the plant.The gene-for-gene hypothesis of Flor (1971). Thereaction of the host with either dominant, R, or recessive,r, gene to the pathogen with either dominantavirulence, AA, or recessive avirulence, Aa, gene isindicated as: þ, susceptible; , resistant.HostphenotypeAvirulentAAPathogen phenotypeAvirulentAaVirulentaaSusceptible rr þ þ þResistant RrResistant RRþThere are likely to be many mechanisms butthese are known in a few cases where thereare both resistant and susceptible varieties ofa plant species whose genetics can be studied.Extreme resistance to Potato virus X (PVX) inpotato is provided by the dominant Rx1 andRx2 genes, which are located on chromosomesXII and V, respectively. Rx resistance is elicitedby the viral coat protein in a strain-specificmanner, and, by comparison of PVX sequencesand mutagenesis, amino acid residue 121 in thecoat protein gene was recognised as the majordeterminant of resistance-breaking activity.This extreme resistance is not associated witha hypersensitive response (HR).The primary structure of the Rx1 and Rx2gene products is similar to the LZ-NBS-LRRclass of R genes (Box 10.2). Thus, Rx is similarto R genes that give resistance to bacteria andfungi by conferring an HR; in contrast, phenotypicanalysis showed that Rx-mediated resistanceis independent of an HR. However,transgenic expression of the Rx gene showedthat there is the potential for an Rx-mediatedHR but that this potential is not realised whenthe coat protein is expressed from the PVXgenome during viral infection.The multiplication of ToMV in tomatoplants is inhibited by the presence of the Tm-1gene, the inhibition being more effective in thehomozygote (Tm-1/Tm-1) than in the heterozygote(Tm-1/þ). Tm-1 resistance is expressed inprotoplasts even in the presence of actinomycinD and thus can be classed as being extreme.III. HOW DO PLANT VIRUSES WORK?

T194 10. VIRUS-PLANT INTERACTIONS: 1. MOLECULAR LEVELBOX 10.2Numerous plant R genes have been identified, cloned, and sequenced. Many of them share strikingstructural similarities, and they are now placed into five groups based on their structures (Fig.); asixth group is of R genes that do not fit the other five groups. Many have a leucine-rich repeat region(LRR) and a nucleotide-binding site (NBS). Other common features include a serine-threonine kinasedomain, a leucine zipper or structures found in insects (the developmental gene Toll of Drosophila),and the immune response gene TIR from mammals.Fig. Structure of plant R genes: Kin ¼ serine/threonine kinase; LRR ¼ leucine-rich repeats; NBS ¼ nucleotidebinding site; TIR ¼ Toll and interleukin receptor domain. Below each structure are given some examples of Rgenes, those controlling viruses being boxed; N against TMV; Rx against PVX; Sw5 against Tomato spotted wiltvirus; HRT against Turnip crinkle virus; RCY1 against Cucumber mosaic virus; Tm-2 against Tomato mosaic virus.Others are examples of R genes controlling fungi (Mla against powdery mildew, Erysiphe graminis f. sp.hordei; L6 against flax rust, Melampsora lini; Cf2 against Cladosporium fulvum), bacteria (Rps2, Rps4, and Pto againstPseudomonas syringae pv. tomato; Xa21 against Xanthomonas oryzae pv oryzae), against nematodes (Gpa2 againstcyst nematode; Mi against root-knot nematodes and the aphid Macrosiphum euphorbiae).Plant R genes seem to encode receptors that interact directly or indirectly with elicitors producedby the pathogen Avr genes. It is likely that the LRR is the pathogen recognition domain, and recognitionof the Avr gene product prompts a signal transduction cascade, the precise mechanisms ofwhich are poorly understood but that possibly involve salicylic acid, jasmonic acid, and ethylene.It is interesting to note that plant R genes against viruses have the same structure as those againstfungi, bacteria, nematodes, and aphids. Genome mapping shows that R genes against all thesepathogens cluster on the chromosome.III. HOW DO PLANT VIRUSES WORK?

Comparison of the sequences of a virulent withan avirulent strain of TMV revealed two basesubstitutions resulting in amino acid changesin the replicase 130 and 180 kDa proteins (Glnto Glu at position 979 and His to Tyr at 984).Mutagenesis of infectious transcripts suggestedthat the two concomitant base substitutions,and possibly also the resulting amino acidchanges, were involved in the recognition ofthis Avr gene by the tomato Tm-1 gene.Immunity to many potyviruses is given byrecessive genes (Table 10.1). The interactionbetween the host eukaryotic translation initiationfactor 4E (eIF4E) and its isoform [eIF(iso)4E] and the viral VPg (see Chapter 6 for VPg)is central to this form of resistance (Box 10.3).This interaction is important in the translationand replication of potyviral genomes (see Box8.6). The resistance gene in plants correspondsto eIF4E or eIF(iso)4E, mutations of which preventthe interaction with the VPg. Thus, examplesexist of immunity being controlled both bydominant and recessive genes.II. HOST RESPONSES TO INOCULATIONB. Subliminal InfectionAs noted in Box 9.1, some viruses can replicatein the initially infected cell but cannotspread to adjacent cells because they cannotpass through the plasmodesmata; this is termedsubliminal infection and can be shown by theability of the virus to replicate in protoplasts ofa plant species but not in the whole plant.C. Nonpermissive Infection195The host may respond to virus infectioneither by containing the virus to the inoculatedleaf (local infection) or by reacting to systemicinfection by the virus.1. Local InfectionA number of virus-host combinations inducean HR around the site of infection, whichshows as local lesions (figure in Profile 14).The localisation of virus replication in tissuenear the site of infection is important inBOX 10.3INTERACTION BETWEEN POTYVIRAL VPG ANDEUKARYOTIC TRANSLATION INITIATION FACTORThe 5 0 -linked protein, VPg, (see Chapter 6) ofseveral potyviruses binds to the eukaryotictranslation initiation factor 4E (eIF4E) or to itsisoform [eIF(iso)4E] in yeast two-hybrid andin vitro binding systems. In the eukaryotic translationinitiation complex, eIF4E binds the 5 0 capof mRNAs, bringing it into close proximity to3 0 poly(A) sequence, which is bound to the complexby the poly(A)-binding protein. It isthought that the binding of the VPg by eIF4E isinvolved in the translation and/or replicationof the potyviral genome. Potyviruses differ intheir ability to use eIF4E isoforms from a givenhost plant with some requiring one specific isoformfor their replication cycle and others usingseveral isoforms. Mutations in the translationinitiation factor confer recessive resistance.These mutations are nonconservative substitutionsof a few amino acids clustering in twoneighbouring regions of the eIF4E 3D structurelocated near the cap-binding pocket and at thesurface of the protein.III. HOW DO PLANT VIRUSES WORK?

196 10. VIRUS-PLANT INTERACTIONS: 1. MOLECULAR LEVELagriculture and horticulture as the basis forfield resistance to virus infection.Genes that induce an HR in intact plants orexcised leaf pieces fail to do so when isolate protoplastsare infected. In the whole plant, virusparticles are found in, but restricted to, theregion immediately surrounding the necroticlesion. The N gene of tobacco (e.g., cv SamsunNN) that confers resistance to TMV (Box 10.4)belongs to the TIR-NB-LRR class of R genes(Box 10.2). The response is elicited by the helicasedomain of the TMV replicase.The N’ gene, originating from Nicotiana sylvestris,controls the HR directed against mosttobamoviruses, except U1 (vulgare) and OMstrains of TMV, which move systemically andproduce mosaic symptoms in N’-containingplants. The TMV coat protein gene is involvedBOX 10.4THE N GENE OF TOBACCOThe N gene was isolated by transposon taggingusing the maize activator transposon and characterised.The gene encodes a protein of 131.4 kDawith an amino-terminal domain similar to that ofthe cytoplasmic domain of the Drosophila Toll proteinand the interleukin-1 receptor in mammals, anucleotide binding site, and 14 imperfect leucinerichregions (Fig.). The N gene is expressed fromtwo transcripts, N S and N L , via alternative splicingpathways. The N S transcript codes for the fulllengthN protein and is more prevalent beforeand for 3 hours after TMV infection. The N L transcriptcodes for a truncated N protein (N tr ), lacking13 of the 14 leucine-rich repeats and is more prevalent4 to 8 hours after infection. A TMV-sensitivetobacco variety transformed to express the N proteinbut not the N tr protein is susceptible to TMV,whereas transgenic plants expressing both N Sand N L transcripts are completely resistant. However,the ratio of N S to N L mRNAs before and afterTMV infection is critical, as the expression of eitherone mRNA alone or the two at a 1:1 ratio givesincomplete resistance. It is suggested that the relativeratio of the two N messages is regulated byTMV signals.The HR response of the N gene to TMV is temperaturesensitive, being inactivated above 28 o C.It is suggested that at higher temperatures theinteraction between the viral elicitor and the hostsurveillance mechanism that leads to HR isweakened.Fig. Schematic diagram of the N and N tr proteinsshowing the various domains. CD, putative cytoplasmicdomain of N with sequence similarity to Toll, interleukin-1R,and MyD88; NBS, putative nucleotidebindingsite; LRR, leucine-rich repeat region consistingof 14 imperfect tandem leucine-rich repeats. The N trprotein arises from alternative splicing of the chromosomalN gene. [From Proc. Natl. Acad. Sci. USA., 92,S.P. Dinesh-Kumar, S. Whitham, D. Choi, R. Hehl,C. Corr, and B. Baker, Transposon tagging of tobaccomosaic virus resistance gene N: Its possible role inthe TMV-N-mediated signal transduction pathway,pp. 4175–4180, Copyright (1995) National Academy ofSciences, U.S.A.]III. HOW DO PLANT VIRUSES WORK?

II. HOST RESPONSES TO INOCULATION197in the induction of the N’ gene HR. The elicitoractive site covers approximately 600Å 2 of theright face of the coat protein a-helical bundleand comprises 30 percent polar, 50 percentnonpolar, and 20 percent charged residues.The N’ gene specificity is also dependent onthe three-dimensional fold of the coat protein,as well as on specific surface features withinthe elicitor active site.Two allelic genes in tomato, Tm-2 and Tm-2 2 ,give an HR to certain strains of ToMV (seeTable 10.2). As well as being determined by virusstrain, the HR is also dependent on genotype oftomato and on environmental conditions, especiallytemperature. The response can vary froma very mild necrotic lesion giving apparent subliminalinfection, through the normal necroticlocal lesion to systemic necrosis. The 30 kDamovement protein (MP) of TMV has been identifiedas being the inducer of the HR in both Tm-2and Tm-2 2 plants.HR in potato to PVX is controlled by theNb gene, which has been mapped to a resistancegene cluster in the upper arm ofchromosome V. The Nb avirulence determinantwas mapped to the PVX 25 kDa gene encodingthe MP.a. Host Protein Changes in the HypersensitiveResponse. One of the earliest detectableevents in the interaction between a planthost and a pathogen that induces necrosis is arapid increase in the production of ethylene,which is a gaseous plant stress hormone. In thehypersensitive response to viruses, there is anincreased release of ethylene from leaves. The factthat ethepon (a substance that releases ethylene)introduced into leaves with a needle can mimicthe changes associated with the response of SamsunNN (containing the N gene) to TMV is goodevidence that ethylene is involved in the initiationof this HR. An early burst of ethylene productionis associated with the virus-localising reaction,but the increase in ethylene production is notdetermined by the onset of necrosis but by amuch earlier event.The HR involves a series of complex biochemicalchanges at and near the infection siteTABLE 10.2Genetic Interactions Between ToMV-Resistant Tomato Plants and Strains of the Virus aVirus genotypeHost genotype b 0 1 2 2 2 1.2 1.2 2Wild type M c M M M M MTm-1 R M R R M MTm-2* R R M R M RTm-2 2 * R R R M R MTm-1/Tm-2 R R R R M RTm-1/Tm-2 2 R R R R R MTm-2/Tm-2 2 R R R R R RTm-1/Tm-2/Tm-2 2 R R R R R Ra Modified from Fraser [1985; in Mechanisms of resistance to plant disease (R.S.S. Fraser, Ed.), pp. 62–79, Martinus Nijhoff/Junk,Dordrecht].b Plants with genotype marked*May show local and variable systemic necrosis rather than mosaic when inoculated with virulent strains.c M ¼ systemic mosaic; R ¼ resistance.III. HOW DO PLANT VIRUSES WORK?

198 10. VIRUS-PLANT INTERACTIONS: 1. MOLECULAR LEVELthat include the accumulation of cytotoxic phytoalexins,the deposition of callose and lignin inthe cell walls, and the rapid death of plant cellsforming the necrotic lesion. The regulation ofHR is equally complex, involving interplay ofmany potential signal transducing moleculesincluding reactive oxygen species, ion fluxes,G proteins, jasmonic and salicylic acids, proteinphosphorylation cascades, activation of transcriptionfactors, and protein recycling by thepolyubiquitin system.b. Local Acquired Resistance. A highdegree of resistance to TMV develops in a 1 to2 mm zone surrounding TMV local lesions inSamsun NN tobacco. The zone increases in sizeand resistance for about 6 days after inoculation.Greatest resistance develops in plantsgrown at 20–24 o C; resistance is not found inplants grown at 30 o C.2. Systemic InfectionOn occasions, the necrosis induced by virusinfection is not limited to local lesions butspreads. This is usually from expanding locallesions that reach veins and result in systemiccell death. The systemic necrosis can rangefrom necrosis in a few areas of upper leavesor sporadic necrotic spots mixed with mosaicsymptoms to widespread necrosis leading todeath of the plant. The systemic necrotic symptomsare dependent on host genotype, virusstrain, and environmental conditions. Ring spotsymptoms, in which necrotic rings spread in anapparently diurnal manner are described inChapter 2.3. Systemic Acquired ResistanceSystemic acquired resistance (SAR) is awhole-plant resistance that occurs in responseto an earlier localised exposure to a pathogen,especially (but not only) one that causes tissuenecrosis. It is the activation of defences in uninfectedparts of the plant and gives a long-lastingand broad-based resistance. Plant virusesas well as bacteria and fungi induce SAR, usuallyfollowing the development of necrotic locallesions.One of the responses of plants to hypersensitivityis the production of a family of solubleproteins, termed pathogenesis-related (PR) proteins.PR proteins have been identified in arange of plant species and shown to be inducedby a variety of microbial infections (viruses,viroids, bacteria, and fungi) and by treatmentwith certain chemical elicitors, notably salicylicacid (SA) and acetylsalicylic acid (aspirin).Their production inhibits the systemic infectionby fungi and bacteria, but no specific PR proteinappears to directly affect subsequent infectionsby viruses. The induction of PR proteinsby HR due to fungi and bacteria requires theaccumulation of endogenous SA. The applicationof SA affects the replication of someviruses but not others.Thus, a conundrum arose in that the inductionof PR proteins by the HR response toa virus was not associated with the SAR tothat virus. However, the HR induced byTMV infection of N-gene tobacco has manyfeatures in common with HRs caused by fungiand bacteria. The reaction is mediated by asustained burst of reactive oxygen species(ROS) followed by first local and then systemicaccumulation of SA. Study of thisconundrum revealed an alternative pathwaythatconfersSARtovirusesbutnottobacteriaor fungi (Box 10.5). Therefore, the SAR pathwayin plants differs between virus andfungi/bacteria.The action at a distance involved in SARpresumably requires the translocation of somesubstance or substances. Substantial evidenceexists that transport of a resistance-inducingmaterial is involved. For example, when themidrib of an upper tobacco leaf is cut, resistancedoes not develop in the portion ofthe lamina distal to the cut. The nature of thematerial that migrates is unknown, as is theactual mechanism of resistance in the resistantIII. HOW DO PLANT VIRUSES WORK?

BOX 10.5PATHWAY OF SYSTEMIC ACQUIRED RESISTANCETO PLANT VIRUSESThe HR-induced systemic acquired resistance by TMV infection of N-gene tobacco has many featuresin common with that caused by fungi and bacteria through pathogenesis-related (PR) proteins. Thereaction is mediated by a sustained burst of reactive oxygen (ROS) followed by first local and thensystemic accumulation of salicylic acid (SA). However, the SA-induced resistance to TMV replicationin tobacco is inhibited by salicylhydroxamic acid (SHAM), which does not inhibit the SA-inducedsynthesis of PR proteins; this suggests that there are two branches in the pathway to SA-inducedresistance (Fig.) One branch leads to the production of PR proteins that confer resistance to fungiand bacteria, and the other induces resistance to viral replication and movement.Fig. Pathways for systemic acquired resistance induced by viruses and by fungi and bacteria.SHAM is a relatively selective competitive inhibitor of the alternative oxidase (AOX) in the mitochondrialelectron flow pathway in plants. The SHAM-sensitive pathway, induced by SA and potentiallyby AOX, is critical in the early stages of N-gene mediated resistance to TMV in tobacco.Salicylic acid also increases the turnover of the host RNA-dependent RNA polymerase (RdRp),which could be involved in RNA silencing.Thus, the current model for the HR-induced SAR of TMV in N-gene tobacco is that the interactionbetween the viral avirulence gene (the helicase domain of the RdRp) and the N gene triggers the productionof salicylic acid, which opens up two pathways. One increases the activity of the AOX in themitochondria, which leads to the increased expression of other (unknown) defence genes in thenucleus. The other increases the RNA silencing defence system.III. HOW DO PLANT VIRUSES WORK?

200 10. VIRUS-PLANT INTERACTIONS: 1. MOLECULAR LEVELuninfected leaves. The migrating materialmight involve SA, ethylene, jasmonic acid,nitrous oxide, or even small peptides such assystemin.4. Programmed Cell DeathMulticellular organisms have mechanismsfor eliminating developmentally misplaced orunwanted cells or sacrificing cells to preventthe spread of pathogens. This is termed programmedcell death (PCD), or apoptosis; apoptosisis a specific case of PCD with a distinctset of physiological and morphological features.Although much of the work on PCDhas been done in animal systems, there isincreasing interest in this process in plants.The HR response to plant pathogens has variousfeatures in common with PCD. Certain animalviruses can inhibit PCD. It will beinteresting to see if plant viruses have similarproperties.D. Permissive Infection1. Systemic Host ResponseAs described in Chapter 2, there is a widerange of systemic symptoms induced by viruses,the most common and characteristic of which isthe mosaic symptom. The mosaic symptom comprisesareas of the leaf showing various degreesof chlorosis, together with areas that remaingreen and are termed “dark green islands.” Thedark green, light green, and even chloroticpatches that make up mosaics range from relativelylarge (e.g., TMV, TYMV, and AbMV) to relativelysmall, giving a fine mosaic (e.g., CPMV incowpea, AMV in tobacco). These areas are oftendelimited by the vein structure of the leaf, givingstreak or stripe symptoms in monocotyledons.The development of mosaic symptoms isdescribed in more detail in Chapter 9.Very little is known about the detailedmolecular biology of the plant in the developmentof mosaic symptoms. However, it is clearthat the RNA silencing host defence systemplays a significant role (see Chapter 11).2. Virus Genes InvolvedBecause virus genomes are relatively easy tomanipulate, much more is known about virusdeterminants of symptoms. This can be shownby two examples. By making in vitro recombinantsbetween various strains of Cauliflowermosaic virus (CaMV), the input of various partsof the genome to symptom production weredetermined (Figure 10.1). The effects of someof the genes, such as the movement protein(ORF1 product), on virus spread were obvious,but that of others (e.g., ORFI/II, movementprotein þ aphid transmission factor and ORFV,replicase on stunting) are not clear.Specific amino acid substitutions and deletionsin the coat protein of TMV affect the productionof chlorotic symptoms. Mutants that retainedthe C-terminus of the coat protein induce thestrongest chlorotic symptoms in tobacco in bothexpanded and developing leaves. The chloroticsymptom formation is related to the concentrationof TMV capsid proteins, which form aggregatesin infected cells but do not accumulate inchloroplasts. In contrast to this, in infections withYSI/1, a naturally occurring chlorotic mutant ofthe U1 strain of TMV, coat protein, is found inthe chloroplasts associated with the thylacoidmembrane fraction in both the stroma and membranefractions of the chloroplasts of infectedcells. The coat protein of YSI/1 differs from thatof U1 in two nucleotides, one of which gives anAsp to Val change at amino acid 19 is responsiblefor the chlorotic phenotype. The coat protein ofanother natural chlorotic TMV mutant, flavum,also has a substitution at amino acid 19, but thistime Asp to Ala. However, these severe chloroticsymptoms are unusual in TMV infections andonly associated with mutants. In natural infections,the chlorotic element of the mosaic isusually light green and is not accompaniedby the accumulation of coat protein bodies orwith coat protein in chloroplasts.III. HOW DO PLANT VIRUSES WORK?

II. HOST RESPONSES TO INOCULATION201FIGURE 10.1 Location of CaMV genome domains containing strain-specific symptom determinants. Arrows withRoman numerals represent viral genes. The outer black boxes show the regions determining symptom differences betweentwo CaMV strains; the dotted line is a regions whose specific role was not determined. [This article was published in Virology,172, R. Stratford and S.N. Covey, Segregation of cauliflower mosaic virus symptom genetic determinants, pp. 451–459,Copyright Elsevier (1989).]Even though these examples identify symptomdeterminants in viral genomes, they donot reveal the interactions between the viraland host genome that lead to the expressionof specific symptoms. The in planta role ofsome of the protein products of viral genesinvolved in disease induction may be verydifficult to study for several reasons:• The proteins may be present in very lowconcentration, as a very few molecules percell of a virus-specific protein could block ordepress some host-cell functions.• Such proteins may only be present in theinfected cells for a short period relative to thatrequired for the completion of virus synthesis.• The virus-specified polypeptide may formonly part of the active molecule in the cell.• The virus-specified polypeptide may bebiologically active only in situ, such as in themembrane of some particular organelle.• Many macroscopic disease symptoms maybe due to quite unexpected side effects ofvirus replication. Here is a hypotheticalexample. Consider the 49 kDa proteinase (22kDa VPg þ 27 kDa nuclear inclusion a; seeProfile 10 for genome map) coded for byTobacco etch virus. The amino acid sequencesthat function as substrate recognition signalshave been identified. In the usual host undernormal conditions, this proteinase canaccumulate to high levels within infectedcells without causing cell death. This mustmean that the proteinase does notsignificantly deplete the amount of any vitalhost protein. Suppose that we change toanother host species or to differentenvironmental conditions. In the newsituation, some host-coded protein that isessential for cell function might be sensitiveto cleavage, with a new pattern of diseasedeveloping as a consequence.III. HOW DO PLANT VIRUSES WORK?

202 10. VIRUS-PLANT INTERACTIONS: 1. MOLECULAR LEVELHowever, as we will see in Chapter 11, oneviral gene product, the suppressor of RNAsilencing, obviously plays a major part insymptom production. The final expression ofmany symptoms reflects the balance betweenRNA silencing and the suppression of thisdefence mechanismIII. INTERACTIONS BETWEENVIRUSESThe joint infection of a plant with two (ormore) viruses can lead to responses that differfrom those of infection by the individualviruses. The response can depend on whetherthe viruses are related or unrelated andwhether the infection is simultaneous orsequential.A. Interactions Between Related VirusesSequential infection by strains of viruses, oreven related viruses, can lead to the secondvirus being suppressed. This has been termedcross protection, which is defined as the protectionconferred on a host by infection withone strain of a virus that prevents infection bya closely related strain of that virus. Forinstance, tobacco plants inoculated with a mildstrain of PVX are immune from subsequentinoculation with severe strains of the virus,even if inoculated after only 5 days later. Theyare not immune to infection with the unrelatedviruses, TMV and PVY. This phenomenon,which has also been called antagonism, orinterference, occurs very commonly amongrelated virus strains. It is most readily demonstratedwhen the first strain inoculated causesa fairly mild systemic disease and the secondstrain causes necrotic local lesions or a severedisease. Interference between related strainscan also be demonstrated by mixing the twoviruses in the same inoculum and inoculatingto a host that gives distinctive lesions for oneor both of the two viruses or strains.Strains of Alfalfa mosaic virus (AMV) differ inthe aggregation bodies that their particles formin infected cells. Electron microscopy was usedto investigate the cross-protection interactionsbetween two strains of AMV. When the challengingstrain was inoculated at the same timeas the protecting strain or a short time after(about 4 hours), the two aggregation typeswere found side-by-side merging into eachother in the same cell. When there was a longerinterval between inoculation of the protectingand challenging strains (about 7 hours), thetwo strains were found in separate parts ofthe cytoplasm of the same cell and after aninterval of about 10 hours in separate cells.Only when cross-protection was complete asassessed by back inoculation to an indicatorhost could the aggregation bodies of the challengingstrain not be found.Several mechanisms have been proposed forcross-protection. However, the most rationalexplanation of this phenomenon is the RNAsilencing mechanism induced by plant virusinfection (see Chapter 11). Cross-protectionwith mild virus strains is used as a controlmeasure (see Box 14.4).Coinoculation can lead to concurrent protectionin which there is a reduction in challengingvirus infection rate and/or titre induced by aprotecting virus that does not accumulate orinduce symptoms in that host plant. The cowpealine Arlington shows no symptoms after inoculationwith Cowpea mosaic virus (CPMV), andno infectivity or accumulation of capsid antigencan be detected. Coinoculation of Arlington withCPMV and Cowpea severe mosaic virus (CPSMV)reduces the numbers of CPSMV-inducedlesions. Inoculation of an isogenic line derivedfrom Arlington with CPMV also protects itagainst infection with Cherry leaf roll virus andSouthern Cowpea mosaic virus. This protection iselicited by CPMV RNA 1.III. HOW DO PLANT VIRUSES WORK?

B. Interactions Between UnrelatedViruses1. Complete Dependence for DiseaseFor some virus combinations, there is completedependence of one virus on the other forits replication. The dependent virus is termed asatellite virus and is fully described in Chapter 3.2. Incomplete Dependence for DiseaseThis situation exists between two viruseswhere both are normally associated with arecognised disease in the field. For example,the important tungro disease of rice is normallycaused by a mixture of RTBV, a reverse transcribingDNA virus, and RTSV an RNA virus.As described in Chapter 2, RTSV on its own istransmitted by the rice green leafhopper butcauses few or no symptoms. RTBV causessevere symptoms in rice, but no vectors areknown for this virus on its own; it requiresthe presence of RTSV for transmission. Thus,in the disease complex, RTSV gives the transmissionand RTBV most of the symptoms.Most, if not all, umbraviruses are associatedwith luteoviruses that provide their insecttransmission. An even more complex systemis that of groundnut rosette disease whichinvolves three agents, an umbravirus (GRV), aluteovirus (GRAV), and a satellite RNA (GRVsat-RNA) as described in Box 2.1.3. Synergistic Effects on Virus ReplicationJoint infection of tobacco plants with PVXand PVY is characterised by severe veinalnecrosis in the first systemically infected leaves(see Box 11.3). Leaves showing this synergisticreaction contain up to 10 times as much PVXas with single infections but only the sameamount of PVY. Ultrastructural studies andfluorescent antibody staining show that bothviruses are replicating in the same cells andthat the increased production of PVX is due toan increase in virus production per cell ratherIII. INTERACTIONS BETWEEN VIRUSESthan an increase in the number of cells supportingPVX replication. The level of PVX (–)-strandRNA increases disproportionately to that of(þ)-strand RNA in doubly-infected tissues suggestingthat the synergism involves an alterationin the normal regulation of the relativelevels of the two RNA strand polarities duringviral replication. This is due to the activity ofthe HC-Pro product of PVY, which is asuppressor of RNA silencing (see Chapter 11).4. Effects on Virus MovementAs discussed in Chapter 9, infection andsystemic movement by one virus in a particularhost may complement the cell-to-cell andsystemic movement of an unrelated virus thatnormally would not move from the initiallyinfected cells in that host. Similarly, a fully systemicallyinfecting virus can complement themovement of a tissue-restricted virus (e.g.,phloem-limited virus) out of that tissue.C. Interactions Between Viruses andOther Plant Pathogens203Virus infection can affect resistance to fungaland bacterial infection of plants. For example,infection with Phytophthora infestans develops lessrapidly in potato plants infected with a number ofviruses and infection of a hypersensitive tobaccocultivar with TMV induced systemic and longlivedresistance against Phytophthora parasitica,Peronospora parasitica, and Pseudomonas tabaci.The development of resistance of this sort probablyinvolves the PR proteins discussed earlier inthis chapter. Indeed, fungicidal compounds havebeen isolated from plants reacting with necrosisto virus infection.On the other hand, virus infection may increasethe susceptibility of a plant to fungal infection. Forexample, sugar beet plants in the field infectedwith Beet mild yellows virus have greatly increasedsusceptibility to Alternaria infection.Alternatively, fungal infections can affect thesusceptibility to viral infections. For example,III. HOW DO PLANT VIRUSES WORK?

204 10. VIRUS-PLANT INTERACTIONS: 1. MOLECULAR LEVELpinto bean leaves were heavily inoculated withthe uredinial stage of the rust fungus, Uromycesphaseoli, on one half-leaf and then later withTMV over the whole leaf. Subsequent estimationsof the amounts of TMV showed the presenceof up to 1,000 times as much virusinfectivity in the rusted as in the nonrustedhalf-leaves.Other fungi may induce resistance or apparentresistance to viral infection. Xanthi tobaccoplants that had been injected in the stem with asuspension of spores of Peronospora tabacina producedfewer and smaller necrotic local lesionswhen inoculated with TMV three weeks later.IV. VIRUSES OF OTHERKINGDOMSThere are both similarities and differences inthe interactions of viruses with plant and animalhosts reflecting the fact that, although thecells of both are eukaryotic, their distributionwithin the organism and their interconnectivitydiffer. Thus, nonhost resistance to viruses inanimals can be attributed to the possibility thatthe virus can not enter the initial cell as it cannot interact with the cell surface receptor(s).As noted in Chapter 12, plant viruses do notinteract with cell surface receptors but have tobe introduced directly into the cell (usually bya vector); thus, nonhost resistance reflects thefact that the virus can not replicate in that cell.The ways that the outcome of successful infectionis viewed in animals (see Dimmock et al.2007, Chapter 14) differ from those of plant virusinfections. This makes it difficult to comparethese outcomes in detail, but as noted above,reflect differences in organisational differencesin the two organisms. However, at the singlecell level there are likely to be many similaritiesin the molecular interactions involved in theexpression and replication of the viral genomes.The differences in interactions of viruses withplants and bacteria reflect the fact that the latterare usually single celled organisms and thattheir cell organisation is prokaryotic.V. SUMMARY• Three basic responses to inoculation ofa plant with a virus: total immunity,nonpermissive infection, permissiveinfection.• Extreme immunity is usually either nonhostresistance or conferred by one or moregenes which may be either dominant orrecessive.• Nonpermissive infection is the containingof the virus to the inoculated leaf orpreventing full systemic infection of theplant.• Local infection is usually containedby a hypersensitive response (HR) whichis usually controlled by a single gene.• Systemic infection may be controlled by asignal moving from a local response(systemic acquired resistance: SAR). TheSAR pathway in response to virus infectiondiffers from that to fungal or bacterialinfection.• Both viral and host genes are involved inthe establishment of permissive infectionand symptom production. Among themost important viral genes is/are that(those) that suppress RNA silencing(see Chapter 11).• Viruses can interact to give synergisticeffects where the overall symptoms are moresevere than those of the individual virusescombined.ReferencesDimmock, N.J., Easton, A.J., and Leppard, K.N (2007). Introductionto modern virology. Blackwell Publishing, Oxford, UK.Flor, H.H. (1971). Current status of the gene-for-gene concept.Annu. Rev. Phytopathol. 9, 275–296.III. HOW DO PLANT VIRUSES WORK?

V. SUMMARY205Further ReadingCaranta, C. and Dogimont, C. (2008). Natural resistanceassociated with recessive genes. Encyclopedia of Virology,Vol. 4, 177–186.Foster, G.D., Johansen, I.E., Hong, Y., and Nagy, P.D.(2008). Plant virology protocols: From viral sequence to proteinfunction, 2nd ed., Humana Press, Forlag.Kang, B.-C., Yeam, I., and Jahn, M.M. (2005). Geneticsof plant virus resistance. Annu. Rev. Phytopathol. 43,581–621.Maule, A.J., Caranta, C., and Boulton, M.I. (2007). Sourcesof natural resistance to plant viruses: Status and prospects.Molec. Plant Pathol. 8, 223–231.Moffett, P. and Klessig, D.F. (2008). Natural resistance associatedwith dominant genes. Encyclopedia of Virology,Vol. 4, 170–176.Robaglia, C. and Caranta, C. (2006). Translation initiationfactors: A weak link in plant RNA virus infection. TrendsPlant Sci. 11, 40–45.Ryals, J.A., Neuenschwander, U.H., Willits, M.G., Molina,A., Steiner, H.Y., and Hunt, M.D. (1996). Systemicacquired resistance. Plant Cell, 8, 1809–1819.Singh, D.P., Moore, C.A., Gilliland, A., and Carr, J.P. (2004).Activation of multiple antiviral defence mechanisms bysalicylic acid. Molec. Plant Pathol. 5, 57–63.Soosaar, J.L.M., Burch-Smith, T.M., and Dinesh-Kumar, S.P.(2005). Mechanisms of plant resistance to viruses. NatureReviews: Microbiology 3, 789–798.III. HOW DO PLANT VIRUSES WORK?

This page intentionally left blank

C H A P T E R11Virus-Plant Interactions:2. RNA SilencingRNA silencing is a major defence system in plants that successful viruses must overcome.O U T L I N EI. Introduction 207II. Mechanism of Silencing 208III. Systemic Silencing 211IV. Overcoming Silencing 211V. Silencing and Symptoms 214VI. Transcriptional and TranslationalAu9 Repression 218VII. Evolutionary Aspects 218VIII.RNA Silencing in Animal and OtherViruses 218IX. Summary 218I. INTRODUCTIONAs we will see in Chapter 15, attempts weremade to confer protection against viral infectionby transforming plants to express viral geneproducts. However, the results showed variousinconsistencies, especially with protection beinggiven by constructs that would express atranscript but not the viral protein. These andother observations were in accord with anincreasing number of cases in which transformationwith homologues of endogenous plantgenes led to both the transgene and endogenousgene expression being cosuppressed. The cosuppressionis due to either transcriptional genesilencing (TGS) or posttranscriptional geneComparative Plant Virology, Second Edition207Copyright # 2009, Elsevier Inc. All rights reserved.

208 11. VIRUS-PLANT INTERACTIONS: 2. RNA SILENCINGsilencing (PTGS) or, possibly, a combination ofthe two. This led to the realisation that plantshave a defence system against “foreign” nucleicacids and subsequently this, or a similar defencesystem, has been recognised in vertebrates, invertebrates,and fungi. As well as being termed PTGSin plants, it has been called RNA interference(RNAi) in animals and gene quelling in fungi.Since the mechanism is similar in all organisms,this book uses the term RNA silencing.The RNA silencing system not only providesdefence against viruses but also against activationof transposons and transgenes. Furthermore, itis involved extensively in developmentalcontrol through the microRNA (miRNA) pathway(see following). Thus, it is a generic endogenoussystem, one function of which is torespond to virus infections.II. MECHANISM OF SILENCINGA. The Basic PathwayA common feature of RNA silencing is that itinvolves highly structured or double-stranded(ds) RNA, which is an unusual molecule ineukaryotic cells. The basic pathway is shownin Figure 11.1. The dsRNA is cleaved into smallinterfering (si) RNA fragments of 20–25 basepairs by a ribonuclease called DICER. Thestrands of the siRNA fragments are separatedto give the guide strand, which is complementaryto the target mRNA (the RNA that gavethe original dsRNA) and the antiguide strand(helper strand), which is further degraded.The guide strand is then incorporated intothe RNA-induced silencing complex (RISC),which targets it to the cognate mRNA, forminga duplex with that RNA. The cognate mRNAstrand complementary to the bound siRNA iscleaved by the slicer activity of RISC to givefurther siRNA fragments. This leads to threecontrol pathways: further cleavage of mRNAs,translational repression, and transcriptionalFIGURE 11.1 RNA silencing pathways. The red lineindicates the guide strand and the green line the antiguide(helper) strand and the target mRNA.repression. All three pathways are involved inthe plant response to virus infections.B. Components of the System1. dsRNAAs just noted, the silencing system targetsseveral pathways that have dsRNA or highlystructured RNA as a common starting element(Figure 11.2).RNA Viruses. ssRNA viruses replicate via acomplementary RNA (see Chapter 8) and thusgo through a dsRNA stage (the replicative formor intermediate). Furthermore, the secondarystructure of their single-stranded genomes cancontain significant regions of base-pairing thatmay or may not be accessible to DICER; it is likelythat both the replicative form and the secondarystructure are targets for RNA silencing. The genomesof dsRNA viruses are, by their nature,double-stranded. The extensive secondary structureof viroids (see Chapter 3) is a dsRNA targetfor the silencing system.III. HOW DO PLANT VIRUSES WORK?

II. MECHANISM OF SILENCING209FIGURE 11.2 Sources of dsRNAfor the silencing pathway. A. ReplicatingssRNA virus; note both thereplicative intermediate and secondarystructure in the ss form can be targets.B. Replicating dsRNA virus. C.Retro- and pararetro-virus. The secondarystructure in the terminalrepeats (TR) can be a target. D. ReplicatingssDNA virus. The overlapping3 0 ends of transcripts from complementarystrands can be targets. E. Viroids.F. Structure of a hairpinconstruct for transformation intoplant to give protection against thevirus (see Chapter 15).ABCEFDDNA Viruses. The genomes of DNA viruses(e.g., geminiviruses) are transcribed to givemRNAs (see Chapter 8), which are likely to havesignificant secondary structure. The reversetranscribingDNA viruses—for example, caulimoviruses—replicatevia RNA, which also hassecondary structure (see Chapter 8).Aberrant RNAs are transcripts that have faultsdue to, say, truncation of transcription or falsetranscription, giving ds regions and RNAs madedouble-stranded by host RNA-dependent RNApolymerase.Transposons and Transgenes. Retrotransposonsreplicate by reverse transcription and thus havean RNA stage with secondary structure—forexample, stem loops—in the terminal repeats.Transgenes are expressed through mRNA or aberrantRNAs and also may have inverted repeats.Priming miRNA (pri-miRNA) is transcribedfrom intergenic and intronic regions of the hostgenome as stem-loop structure with imperfectbase-pairing (Figure 11.3A) and with complementarityto one or more host mRNA. It is processedto give miRNA.2. DicerDicer and its homologues are RNaseIIIenzymes that cleave dsRNA (see Box 11.1).3. Products of DicersiRNA. Twenty to 25 nucleotide dsRNA fullybase-paired with a 2 nucleotide 3 0 overhang(Figure 11.3B). It is complementary only to thedsRNA from which it arose.miRNA. Twenty-one to 23 nucleotide dsRNAproduced from priming miRNA.III. HOW DO PLANT VIRUSES WORK?

210 11. VIRUS-PLANT INTERACTIONS: 2. RNA SILENCINGFIGURE 11.3 A. Structure of primiRNA.B. Diagram of the structure ofan siRNA showing a 19–21 base-pairRNA duplex with 2 nucleotide 3 0 overhangs:OH, 3 0 hydroxyl; P, 5 0 phosphate.APOHHOBPBOX 11.1DICERS AND RISCSDicerDicers are a 200 kDa family of proteins that usuallycontain an ATPase/RNA helicase domain, aPAZ domain, two RNaseIII domains, and a C-terminal dsRNA binding domain. The distancebetween the two RNase III domains determinesthe size of the siRNA that it produces. Dicer progressivelycleaves dsRNA at 21–25 bp intervalsto generate siRNA with 2-nt 3 0 overhangs andphosphorylated 5 0 termini (Figure 11-3B). It alsocleaves pri-miRNA to give 20–25 bp fragmentswith blunt ends. There are various dicers thathave distinct roles; those from plants are calleddicer-like (DCL).Dicer Product Function of ProductDCL1 miRNA DevelopmentalcontrolDCL2 22nt siRNA Stress-relatednatural defencetranscriptionDCL3 24nt siRNA Methylation ofcognate DNADCL4 21nt siRNA PTGS; systemicspread of signalHumans and Caenorhabditis each have one dicerthat processes both dsRNA (to give siRNAs) andpri-miRNA (to give miRNAs). Drosophila hastwo dicers: Dcr-1, which mainly producesmiRNA, and Dcr-2, which mainly producessiRNA.RISCBesides the si- or mi-RNA, the RNA-inducedsilencing complex (RISC) includes members ofthe Argonaute (AGO) protein family and variousaccessory factors. The AGO proteins are 100kDa and have a PAZ domain and a PIWI domain,which is related to RNaseH endonuclease andfunctions in slicer activity. Various AGO proteinsare probably associated with the variations of thebasic silencing pathway.Arabidopsis encodes 10 putative AGOs, withAGO1 being involved in the RISC. VertebrateRISC has 4 AGOs (1–4), and Drosophila RISC hasAGO-1 involved in the miRNA pathway andAGO-2 involved in the siRNA pathway. AGO-2also processes the antiguide strand during RISCassembly.III. HOW DO PLANT VIRUSES WORK?

IV. OVERCOMING SILENCING2114. RISCRISC is a multiprotein complex containingeither si- or mi-RNA that cleaves target mRNAs(see Box 11.1).C. Results of the System1. Target RNA degradation: Processing ofthe cognate viral RNA or mRNA targetedby si- or mi-RNA in the RISC (Figure 11.1)2. Translational repression: Prevention oftranslation of mRNA due to binding bysiRNA (or miRNA)3. DNA methylation: Epigenetic effectcaused by methylation of host (orviral) DNA because of the presence ofsiRNA (see also Section VI).III. SYSTEMIC SILENCINGThe silencing of a virus can spread systemicallythroughout the plant after induction inthe initially infected cell. It is thought that theviral primary 21 nt siRNA produced by thedicer/RISC system in the initially infected cellmoves through plasmodesmata to about 10–15surrounding cells either with or ahead of theinfection front (Figure 11.4). The host RNAdependentRNA polymerase amplifies thesiRNAs to give ds forms, which are then processedinto secondary siRNAs. These thenmove to surrounding cells and are amplified.This reiterative process moves the siRNA signalto the vascular tissue from where it, possiblyassociated with small-RNA bindingproteins, is distributed to other parts of theplant. From there, the signal spreads throughthe systemic leaves, once again in a reiterativemanner. This systemic movement of the silencingsignal ahead of viral movement primes thedefence response prior to the arrival of thevirus.IV. OVERCOMING SILENCINGThe RNA silencing defence system is a majorproblem facing a virus, especially one with anRNA genome. Thus, viruses have evolved arange of mechanisms to overcome this defencesystem, including suppression of the RNAsilencingpathway, avoidance of the defence system,and possibly outwitting the defence system.A. Suppression of SilencingViruses and viroids have developed twobasic strategies to suppress RNA silencing:encoding a protein that interferes with thesilencing pathway or through a nucleic acidmediatedmechanism.1. Protein Suppressors of SilencingAn increasing number of plant (and animal)viral gene products have been identified assuppressors of RNA silencing (Table 11.1).These proteins often also have other functionsand frequently are identified as being “pathogenicitydeterminants.” However, these otherfunctions, such as replication enhancement,determining virus movement, and symptomproduction, can be attributed to the suppressionof silencing. Some viruses, such as Citrustristeza virus (see Profile 5 for genome organisation),have more than one gene productinvolved in suppression of silencing; these proteinsare thought to have different modes ofaction in silencing suppression.A common feature of many of the suppressorsthat have been characterised is that theybind dsRNA, some binding both long and shortdsRNAs and others just binding ds siRNA(Table 11.1). The p19 of tombusviruses (seeBox 11.2), the p20 of closteroviruses, the p15of Peanut clump virus, the TGB1 of Barley stripemosaic virus, and the HC-Pro of potyviruseseach just bind ds siRNA and inhibit an intermediatestep of RNA silencing. It is thought thatIII. HOW DO PLANT VIRUSES WORK?

212 11. VIRUS-PLANT INTERACTIONS: 2. RNA SILENCINGFIGURE 11.4 Systemic spread of silencing signal. The signal is generated in the initially infected cell (bottom, left hand)and spreads to about 10–15 adjacent cells, where it is amplified. It moves out of the initially infected leaf via the phloemsieve tubes and then spreads throughout systemic leaves being amplified at various times. P, plasmodesma.these silencing suppressors inhibit the assemblyof RISC by competing more efficiently forthe ds siRNA than the RISC assembly complex.HC-Pro, p19, and p21 suppressors bind 21-ntsiRNA duplexes more efficiently than 24-ntsiRNA duplexes; however, p21 and HC-Prorequire a 2-nt 3 0 end overhang, whereas p19does not. The aureusvirus p14 and Turnip crinklevirus coat protein bind both long and shortdsRNA, which could suggest inhibiting at anearlier stage in the silencing pathway.Some evidence exists that not all suppressorsfunction by binding ds RNA and thereby inhibitingthe assembly of RISC. For instance, Cucumbermosaic virus (CMV)-encoded 2b suppressor proteininteracts with Argonaute 1 protein in RISCand inhibits its cleavage activity, and the PolerovirusP0 triggers ubiquitin-mediated proteolysisIII. HOW DO PLANT VIRUSES WORK?

IV. OVERCOMING SILENCING213TABLE 11.1Some Viral RNA Silencing SuppressorsVirus Family Virus Suppressor(s) Other Functions MechanismCarmovirusClosterovirusCucumovirusHordeivirusTurnip crinklevirusCitrus tristezavirusCucumbermosaic virusBarley stripemosaic virusPecluvirus Peanut clumpvirusPolerovirus Beet westernyellows virusPotyvirus Potato virus Y;Tobacco etchvirusTombusvirusTospovirusCarnationItalianringspotvirusTomato spottedwilt virusPlant (þ)-Strand RNA VirusesP38 Coat protein Binds both long and si dsRNAP20, P23, CP Replication enhancer,Nucleic acid binding,Coat protein respectively2bgbHost-specific movement,symptomsReplication enhancer;movement; seedtransmission;pathogenicitydeterminantSuppress intra- and intercellularsilencing, Suppress intracellularsilencing, Suppress intercellularsilencing respectivelyInteracts with ARGONAUTE 1Binds si dsRNAP15 Movement Binds both long and si dsRNAP0 Pathogenicity determinant Targets ARGONAUTE 1HCProP19Movement; polyproteinprocessing; aphidtransmission;pathogenicitydeterminantMovement; pathogenicitydeterminantBinds si dsRNABinds si dsRNAPlant (–)-Strand RNA VirusNSs Pathogenicity determinant Not determinedPlant DNA VirusGeminivirus(Begomovirus)Tomato yellowleafcurl virusAC2Transcription activatorproteinActivates host suppressorPlant Reverse-Transcribing VirusesCaulimovirusCauliflowermosaic virusP6Transactivator of genomeexpression; symptomseverity determinantNot determinedNodavirusFlock Housevirus;NodamuravirusAnimal (þ)-Strand RNA VirusB2 Plaque formation dsRNA bindingContinuedIII. HOW DO PLANT VIRUSES WORK?

214 11. VIRUS-PLANT INTERACTIONS: 2. RNA SILENCINGTABLE 11.1(Continued)Virus Family Virus Suppressor(s) Other Functions MechanismAnimal (–)-Strand RNA VirusOrthobunya-virus La Crosse virus NSs Not determinedOrthomyxo-virus Influenzavirus ANS1Poly(A)-binding; inhibitorof mRNA transport; PKRinhibitordsRNA bindingAnimal DNA VirusAdenovirus Adenovirus VA1 RNA PKR inhibitor Dicer bindingPoxvirus Vaccinia virus E3L PKR inhibitor dsRNA bindingLentivirusSpumavirusHumanimmunodeficiencyvirus-1Primate foamyvirus - 1Animal Reverse-Transcribing VirusTatTranscriptionalactivatorInteracts with dicerTas Transcriptional activator Not determinedof Argonaute 1. Suppression by the begomovirus(geminivirus) AC2 gene product requires transactivationof host suppressor(s). Thus, there maybe various mechanisms for the viral suppressionof silencing.2. Nucleic Acid Suppressors of SilencingAs we saw in Chapter 3, viroids do not encodeany proteins. Yet, the finding of viroid-specificsiRNAs shows that their highly structured RNAsare processed by the silencing pathway. That viroidssuccessfully infect plants indicates that theymust be able to suppress the silencing. It is suggestedthat the secondary structure may alsohave the property of suppressing silencing ofthe replication of, at least, some viroids.B. Avoidance of SilencingFor some viruses no suppressor of silencinghas yet been identified. The susceptible stagein the viral replication cycle to the silencingdefence system is when dsRNA is exposed atstages such as RNA replication or translation.Thus, it is thought that some viruses avoidexposure to the defence system by replicatingin inaccessible sites, such as vesicles. Furthermore,if a virus replicates and expresses rapidlyand then safely encapsidated its genome, itmay outcompete the defence system.V. SILENCING AND SYMPTOMSSilencing and suppression of silencing has amajor influence on the symptoms that virusesproduce in plants. It must be remembered thata productive infection is a balance betweensilencing and suppression of silencing. If therewas no silencing or full suppression of silencing,it is most likely that the plant would diesoon after infection. As pointed out in Chapter4, rapid plant death is a selective disadvantageto the virus. On the other hand, if there was nosuppression of silencing, the virus would notbe able to establish an infection.III. HOW DO PLANT VIRUSES WORK?

V. SILENCING AND SYMPTOMS215BOX 11.2TOMBUSVIRUS SILENCING SUPPRESSORThe silencing suppressor of several tombusviruses is a protein of 19 kDa (p19). It is responsible forTomato bushy stunt virus pathogenesis. The crystal structure of p19 of a related virus, Carnation Italianringspot virus, has been determined and shows that the protein homodimer acts as a molecular calliperbinding a specific size (21nt) of siRNA duplex (Fig.).Fig. The structure of the p19 silencing suppressor bound tosiRNA. The p19 dimer binds one face of an siRNA duplex(brown). Contact between the “core” of the p19 dimer andthe RNA phosphate groups contributes to the protein’s highaffinity for dsRNA, while a pair of tryptophan residues (red;Trp42 and Trp39) in the “reading head” measure siRNAlength. Because each p19 monomer (blue and green) contributea reading head, the protein has been described as a “molecularcalliper” that sizes up the dsRNA so as to bind best tocanonical siRNAs. [This article was published in Curr. Biol.14, P.D. Zamore, Plant RNAo: how a viral silencing suppressorinactivates siRNA, pp. R198–200, Copyright Elsevier (2004).]Although some symptoms are attributable tosilencing and suppression not all are. As outlinedin Chapter 9, the replication, movement,and accumulation of viruses in plants can causeupsets to the physiology and metabolism of aplant. A good example is the yellowing symptomscaused primarily by the presence ofvirus in the phloem upsetting the starch-sugarbalance.A. RecoveryAs noted in Chapter 2, a virus-infected plant,especially with Nepoviruses, may show severeviral symptoms on the inoculated and first systemicallyinfected leaves; however, new growthappears in which symptoms are milder orabsent. For other viruses, such as Alfalfa mosaicvirus, the virus content in the plant can increaseIII. HOW DO PLANT VIRUSES WORK?

216 11. VIRUS-PLANT INTERACTIONS: 2. RNA SILENCINGtermed “dark green islands,” and they areresistant to subsequent virus infection. It wouldappear that they result from one or a smallgroup of cells in which there is silencing butno suppression and that divide to form adiscrete tissue zone.FIGURE 11.5 Concentration and specific infectivity ofAlfalfa mosaic virus harvested from whole tobacco plants atdifferent times after inoculation. Zero lesions are assumedfor time 0. Curve A, amount of purified virus nucleoprotein(mg/kg total leaf wet weight); curve B, number of locallesions induced by sap inoculation; curve C, number oflocal lesions when purified sap samples were equalisedspectrophotometrically. [This article was published in Virology,15, C.W. Kuhn and J.B. Bancroft, Concentration andspecific infectivity changes of alfalfa mosaic virus duringsystemic infection, pp. 281–288, Copyright Elsevier (1961).]and decrease (Figure 11.5) with concomitantchanges in symptoms sometimes in a cyclicalmanner. These recovery phenomena likelyreflect changes in the balance between silencingand suppression.B. Dark Green Islands and MosaicsOne of the most common symptoms of virusinfection is a mosaic of light and dark greenareas on the leaves (Chapters 2 and 9). Themosaic pattern develops in leaves that areundergoing cell division and leaf expansionand can include well-defined dark green areasin which there is no detectable virus; these areC. miRNAAs well as suppressing siRNA silencing,protein suppressors can affect the miRNApathway. Not all suppressors impact upon thispathway, and not all miRNAs are affected.However, it is likely that the perturbation ofthe miRNA pathway leads to some symptomproduction. There is evidence for this in thatsome transgenic plants expressing suppressorsof silencing have the phenotype of virusinfectedplants even though there is no viruspresent.Symptom differences between strains havebeen attributed to differences in the interactionsbetween silencing protein and the miRNApathway. The 2b protein of mild strains ofCMV is a strong suppressor of RNA silencingbut interacts weakly with miRNAs and thus isa weak inducer of symptoms.D. siRNA EffectsTransgenic tomato plants expressing portionsof Potato spindle tuber viroid containedviroid-specific siRNAs and had a phenotypesimilar to the symptoms of viroid infection.This suggests that the siRNAs might be responsiblefor the symptom production possibly bytargeting plant genes for RNA silencing.E. Synergistic EffectsThe silencing suppressor of one virus canaffect the replication of another virus in ajoint infection leading to synergistic effects(Box 11.3).III. HOW DO PLANT VIRUSES WORK?

V. SILENCING AND SYMPTOMS217BOX 11.3SYNERGISMSynergism is when the effects of the combinedinfection by two viruses are more severe thanthat of either of the constituent viruses alone. Itcan be caused by the infection with two virusesthat belong to the same genus or by two virusesfrom different genera.The Uganda variant of African cassava mosaicvirus (UgACMV) caused a severe pandemic incassava in Uganda in the 1990s (see Profile 7).It devastated the country’s cassava production,leading to annual losses estimated of more thanUS$60 million between 1992 and 1997. Thedisease has subsequently spread to other surroundingcountries. UgACMV is caused bysynergism between two begomoviruses: Africancassava mosaic virus (ACMV) and East Africancassava mosaic virus (EACMV). The synergisticeffect is mediated by the differential andcomplementary suppression of silencing bythe AC4 gene of ACMV and the AC2 gene ofEACMV.The joint infection of tobacco plants with PVXand PVY is characterised by severe veinal necrosisin the first systemically infected leaves (Fig).Leaves showing this synergistic reaction containup to 10 times as much PVX as with single infectionsbut only the same amount of PVY. Ultrastructuralstudies and fluorescent antibodystaining show that both viruses replicate in thesame cells and that the increased production ofPVX is due to an increase in virus productionper cell rather than an increase in the numberof cells supporting PVX replication. The levelof PVX (–)-strand RNA increased disproportionatelyto that of (þ)-strand RNA in doublyinfected tissues, suggesting that the synergisminvolves an alteration in the normal regulationof the relative levels of the two RNA strandpolarities during viral replication. The HC-Prosilencing suppressor from PVY is responsiblefor the increase in the amount of PVX.F. Other Activities of SilencingSuppressorsIn Chapter 10, we saw some forms of resistancethat plants can show to viruses. Thereare various complex interactions betweenRNA silencing and its suppression and someof these other forms of virus resistance. Forinstance, potyviral HC-Pro suppressorenhances other forms of host resistance to someviruses and reduces it to other viruses. In somecases, these changes in resistance involved aIII. HOW DO PLANT VIRUSES WORK?

218 11. VIRUS-PLANT INTERACTIONS: 2. RNA SILENCINGsalicylic acid (SA)-dependent pathway,whereas in others an SA-independent pathway,and it could be reversed by the addition of theCMV 2b suppressor.VI. TRANSCRIPTIONAL ANDTRANSLATIONAL REPRESSIONFigure 11.1 shows that there are two otherpathways besides the siRNA/RISC-directedcleavage of mRNA—namely, translationalrepression and transcriptional repression (seealso Section II, B, 3). In translational repressionthe (–) strand of the siRNA binds to the mRNApreventing translation. In the context of a virusinfection, it is difficult to differentiate this frommRNA cleavage, as in the absence of activesuppressors, there will only be viral genomein the initially infected cell(s), and any suppressorwill operate earlier at the RISC or pre-RISCstage. Transcriptional repression could resultfrom the siRNA inducing DNA methylationor remodelling of the chromatin. There is circumstantialevidence for methylation of bothgeminiviral DNA and of modification of chromatinstructure.viral counterdefence. As the prime target forsilencing is dsRNA, there must be selectionagainst excessive secondary structure in viralgenomes. Thus, there must be a balancebetween the needs for secondary structure,such as replication and translational control,packaging in isometric particles, and thepotential exposure to the silencing pathway.VIII. RNA SILENCING IN ANIMALAND OTHER VIRUSESThe silencing pathway appears to be presentin most, if not all, eukaryotes, and it seems as ifit controls viruses in both invertebrate and vertebratehosts. In vertebrates silencing may notbe as important as the immune response tovirus infection; on the other hand, its importancemay not be fully recognised. Severalviruses of both invertebrates and vertebratesencode suppressors of silencing (Table 11.1),which indicates that this defence system operatesin these hosts. Since the silencing pathwayis not found in prokaryotes, there appears to beno such defence system against bacterialviruses.VII. EVOLUTIONARY ASPECTSIt is suggested that RNA silencing has a longevolutionary history. The diversity of viral suppressorprotein sequences indicates that thiscounterdefence strategy has arisen severaltimes. Some could have been derived and modifiedfrom endogenous host suppressors andothers from nucleic acid–binding proteins. It isobvious that the selection pressure for suppressorproteins has been paramount in the developmentof most successful viruses.As suggested earlier, evolution of thesilencing/suppressor system has been toachieve a balance between host defence andIX. SUMMARY• The RNA-silencing defence system is foundin eukaryotes.• The defence system is activated by dsRNAthat is cleaved to give small RNA moleculesthat target the cognate RNA.• There are two main sources of dsRNA:perfectly base-paired from viruses,transposons, transgenes, and aberrations intranscription, which are processed to givesmall interfering (si)RNA; and imperfectlybase-paired priming-micro (pri-mi)RNA,which is processed to give micro (mi)RNAsthat control host development.III. HOW DO PLANT VIRUSES WORK?

IX. SUMMARY219• The dsRNAs are processed through threepathways that have a common basic sectionthat cleaves the dsRNA and targets the si- ormiRNA to its cognate mRNA.• The three pathways lead to degradation ofmRNA, translational repression of mRNA,and transcriptional repression. Defenceagainst RNA viruses is usually by the firstpathway and against DNA viruses by thefirst and third pathways. The secondpathway is frequently used by miRNA.• In plants the silencing response spreadssystemically and primes the defence aheadof the virus infection front.• Viruses encode suppressors of silencing thatact as a counterdefence against the silencingsystem.• Silencing and suppression are also found tobe associated with viruses of invertebratesand vertebrates.Further ReadingBisaro, D.M. (2006). Silencing suppression by geminivirusproteins. Virology 344, 158–168.Bortolamiol, D., Pazhouhandeh, M., Marrocco, K., Genschil,P., and Ziegler-Graff, V. (2007). The polerovirus F boxprotein P0 targets ARGONAUTE1 to suppress RNAsilencing. Current Biology 17, 1615–1621.Díaz-Pendón, J.A. and Ding, S.W. (2008). Direct and indirectroles of viral suppressors of RNA silencing in pathogenesis.Annu. Dev. Phytopathol. 46, 303–326.Itaya, A., Zhong, X., Bundschub, R., Qi, Y., Wang, Y.,Takeda, R., Harris, A.R., Molina, C., Nelson, R.S., andDing, B. (2007). A structured viroid RNA serves as asubstrate for dicer-like cleavage to produce biologicallyactive small RNAs but is resistant to RNA-inducedsilencing complex-mediated degradation. J. Virol. 81,2980–2994.Lakatos, L., Csorba, T., Pantaleo, V., Chapman, E.J., Carrington,J.C., Liu, Y.-P., Dolja, V.V., Calvino, L.F.,López-Moya, J., and Burgyán, J. (2006). Small RNAbinding is a common strategy to suppress RNA silencingby several viral suppressors. EMBO J. 25, 2768–2780.Lewsey, M., Robertson, F.C., Canto, T., Palukaitis, P., andCarr, J.P. (2007). Selective targeting of miRNA-regulatedplant development by a viral counter-silencing protein.Plant J. 50, 240–252.Li, F. and Ding, S.-W. (2006). Virus counterdefense: Diversestrategies for evading the RNA-silencing immunity.Annu. Rev. Microbiol. 60, 503–531.MacDiarmid, R. (2005). RNA silencing in productive virusinfections. Annu. Rev. Phytopathol. 43, 523–544.Olson, K.E., Keene, K.M., and Blair, C.D. (2008). An innateimmune response to viruses. Encyclopedia of Virology.Padmanabhan, M.S. and Dinesh-Kumar, S.P. (2008). Virusinduced gene silencing (VIGS). Encyclopedia of Virology,Vol. 5, 375–379.Shams-Bakhsh, M., Canto, T. and Palukaitis, P. (2007).Enhanced resistance and neutralisation of defenseresponses by suppressors of RNA silencing. VirusResearch 130, 103–109.Szittya, G., Dalmay, T., and Burgyan, J. (2008). Genesilencingpathway. Encyclopedia of Virology, Vol. 4,141–148.Trinks, D., Rajeswaran, R., Shivaprasad, P.V., Akbergenov,R., Oakeley, E.J., Veluthambi, K., Hohn, T., and Pooggin,M.M. (2005). Suppression of RNA silencing by a geminivirusnuclear protein, AC2, correlates with transactivationof host genes. J. Virol. 79, 2517–2527.Voinnet, O. (2005). Induction and suppression of RNAsilencing: Insights from viral infections. Nature ReviewsGenetics 6, 206–220.Xie, Q. and Guo, H.-S. (2006). Systemic antiviral silencing inplants. Virus Res. 118, 1–6.Zhang, Z.-H., Xie, Q., and Guo, H.-S. (2007). Antiviraldefense in plants. Plant Viruses 1, 21–26.Zhang, X., Yuan, Y.-R., Pei, Y., Lin, S.-S., Tuschl, T., Patel, D.J., and Chua, N.H. (2006). Cucumber mosaic virusencoded2b suppressor inhibits Arabidopsis Argonaute1cleavage activity to counter plant defense. Genes andDevelopment 20, 3255–3268.III. HOW DO PLANT VIRUSES WORK?

This page intentionally left blank

S E C T I O N IVPLANT VIRUSES INAGRICULTURE ANDINDUSTRY

This page intentionally left blank

C H A P T E R12Plant-to-Plant MovementBeing obligate parasites, viruses depend for survival on being able to spread from one susceptibleindividual to another fairly frequently.O U T L I N EI. Introduction 223II. Transmission via Plant Material 223III. Transmission by Invertebrates 225IV. Fungal Transmission of Viruses 240V. Viruses of Other Kingdoms 242VI. Summary 242I. INTRODUCTIONBecause viruses cannot penetrate the intactplant cuticle and the cellulose cell wall(Figure 12.1B), plants have a barrier to infection.This problem is overcome either by avoiding theneed to penetrate the intact outer surface (e.g.,in seed transmission or by vegetative propagation)or by some method involving penetrationthrough a wound in the surface layers, such asin mechanical inoculation and transmission byinsects. There is considerable specificity in themechanism by which any one virus is naturallytransmitted.II. TRANSMISSION VIA PLANTMATERIALA. Mechanical TransmissionMechanical inoculation involves the introductionof infective virus or viral RNA into awound on the plant’s surface. When virus establishesitself successfully in the cell, infectionoccurs. This form of transmission occurs naturallywith a few viruses such as Tobacco mosaicvirus (TMV) and Potato virus X (PVX) that arevery stable and reach high concentrations inthe plant. TMV can readily contaminate hands,Comparative Plant Virology, Second Edition223Copyright # 2009, Elsevier Inc. All rights reserved.

224 12. PLANT-TO-PLANT MOVEMENTABFIGURE 12.1 Leaf surface structure. A. Scanning electron micrographs of surface of Nicotiana glutinosa leaf before (leftpanel) and after (right panel) mechanical inoculation. Note leaf hairs in left panel that are broken in right panel. [From Hull(2002).]B. Diagrammatic representation of cross-section of the upper part of a leaf showing the barriers to virus infection.[From Eglinton and Hamilton (1967; Science 156, 1322–1335).]clothing, and implements and can be spread byworkers and, for instance, birds in tobacco andtomato crops. TMV may be spread mechanicallyby tobacco smokers because the virus is commonlypresent in processed tobacco leaf. Forexample, a survey showed that all 37 brands ofcigarette and 60 out of 64 smoking tobaccoscontained infectious TMV.Mechanical transmission is of great importancefor many aspects of experimental plantvirology, particularly for the assay of viruses,often by local lesion production; in the propagationof viruses for purification; in host rangestudies; in diagnosis; and in the study of theearly events in the interaction between a virusand susceptible cells. Mechanical inoculationis usually done by grinding up infected leaftissue in a buffer—usually a phosphate bufferthat contains additives that control nucleasesand polyphenols—incorporating an abrasivesuch as celite or carborundum, and thenrubbing the extract gently on the leaves of therecipient plant. The gentle application woundsthe leaf surface without causing cell death(Figure 12.1A).B. Seed TransmissionAbout one-seventh of the known plantviruses are transmitted through the seed of atleast one of their infected host plants. Seedtransmission provides a very effective meansof introducing virus into a crop at an early stage,giving randomised foci of primary infectionthroughout the planting. Thus, when someother method of transmission can operate toIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

spread the virus within the growing crop, seedtransmission may be of very considerable economicimportance. Viruses may persist in seedfor long periods so commercial distribution ofa seed-borne virus over long distances mayoccur. Seed transmission rates vary from lessthan 1 to 100 percent, depending on virus andhost.Two general types of seed transmission canbe distinguished. With TMV in tomato, seedtransmission is largely due to contaminationof the seedling by mechanical means. Theexternal virus can be readily inactivated by certaintreatments eliminating all, or almost all,seed-borne infection. In the second and morecommon type of seed transmission, the virusis found within the tissues of the embryo. Thedeveloping embryo can become infected eitherprior to fertilisation by infection of the gametes(indirect embryo invasion or gametic transmission)or by direct invasion after fertilisation.Generally speaking, for infection of the embryofrom the mother plant, the earlier the plant isinfected, the higher the percentage of seed thatwill transmit the virus. Obviously, for indirectembryo invasion by pollen, the infection takesplace at pollination.The direct route of seed infection from themother plant poses problems in that symplasticconnection is severed at meiosis. To infect theembryo, the virus has to reach either the floralmeristems, which are beyond the limits of normallong-distance movement in the phloem(see Chapter 9), or the embryo itself. The routeof direct embryo infection of peas by Pea seedbornemosaic virus has been examined in detail(Box 12.1).C. Pollen TransmissionSome viruses are transmitted from plant toplant via pollen. As with seed transmission,two mechanisms appear to operate in pollentransmission: gametic infection of the embryoand direct infection of the mother plant.III. TRANSMISSION BY INVERTEBRATESD. Vegetative PropagationVegetative propagation is an important horticulturalpractice, but it is also, unfortunately,a very effective method for perpetuating andspreading viruses. Economically importantviruses spread systemically through most vegetativeparts of the plant. A plant once systemicallyinfected with a virus usually remainsinfected for its lifetime. Thus, any vegetativeparts taken for propagation, such as tubers,bulbs, corms, runners, and cuttings, will normallybe infected.E. GraftingGrafting is essentially a form of vegetativepropagation in which part of one plant (thescion) grows on the roots (the stock) of anotherindividual. Once organic union has been established,the stock and scion become effectively asingle plant. Where either the rootstock or theindividual from which the scion is taken isinfected systemically with a virus, the graftedplant as a whole will become infected if bothpartners in the graft are susceptible. Graftingmay succeed in transmitting a virus whereother methods fail.III. TRANSMISSION BYINVERTEBRATES225Many plant viruses are transmitted fromplant to plant in nature by invertebrate vectors,members of the Insecta and Arachnida classes ofthe Arthropoda, and the Dorylaimida order ofthe Nematoda. Box 12.2 shows the orders of theInsecta that transmit plant viruses. Six of theorders contain insects that feed by chewing.The Homoptera feed by sucking sap from plantsand are numerically the most important subordercontaining plant virus vectors. Figure 12.2shows three of the most common vectors ofplant viruses: aphids, leafhoppers, and whitefly.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

BOX 12.1DIRECT EMBRYO INFECTION OF PEAS BY PEASEED-BORNE MOSAIC VIRUS (PSbMV)The route by which PSbMV reached pea seeds has been studied in detail by comparing a variety(Vedette) in which the virus is seed transmitted with one (Progretta) in which it is not (Figure).ACBDFig. The pathway to seed transmission of Pea seedbornemosaic virus (PSbMV) in pea. A and B. Analysisof the distribution of PSbMV in longitudinalsections through immature pea seed by immunochemistryusing a monoclonal antibody to thevirus coat protein shows that a cultivar-virusinteraction, which is permissive for seed transmission(e.g., with Pisum sativum cv. Vedette in A)results in widespread accumulation of the virusin the testa tissues. In contrast, in the nonpermissiveinteraction (e.g., with cv. Progretta in B) virusenters the seed through the vascular bundle but isunable to invade the adjacent testa tissues extensively.In both cases there is a gradual reductionin the amount of accumulated virus after invasionsuch that in cv. Progretta only patches (asterisks)of infected tissue remain detectable. Systematicanalyses of the immature seeds of different ageshave identified the routes (red arrows) of virusinvasion in the two cultivars (illustrated diagrammaticallyin C for cv. Vedetta and in D for cv. Progretta).The most consistent observation from allthese studies is that the virus must reach themicropylar area of the testa for seed transmissionto occur, a location providing the closest point ofcontact (arrowhead in A) between the testa tissuesand the embryonic suspensor. In the nonpermissiveinteraction the virus appears to be blocked(denoted by red squares) in its ability to spreadinto and/or replicate in the nonvascular testa tissues.E, embryo proper; F, funiculus; M, micropylarregion; S, suspensor; T, testa; V, vascularbundle. Bar marker ¼ 500m. [This article was publishedin Trends Microbiol. 4, A.J. Maule and D.Wang, Seed transmission of plant viruses: A lessonin biological complexity, pp. 153–158, CopyrightElsevier (1996).]The virus moves through the testa of the immature seed after fertilisation and must reachthe micropylar region of the seed for embryo infection to occur. The micropyle is in close contact withthe base of the embryonic suspensor, which functions as a conduit for nutrient flow to support growthof the embryo. The suspensor is the route by which the virus invades the embryo itself, but it degradesas part of the seed development programme. This leaves a “window of opportunity” for embryo infection;this window of opportunity is taken by the virus in Vedette but not in Progretta. However, there isno symplastic connection between maternal and embryonic tissue, and it is still unknown how thevirus crosses from the maternal testa cells to the embryonic suspensor.

III. TRANSMISSION BY INVERTEBRATES227BOX 12.2VIRUS TRANSMISSION BY INSECTASeven of the 29 orders in the living Insecta feeding on living green land plants are vectors of plantviruses and are listed here with the approximate number of vector species in parentheses:1. Orthoptera—chewing insects; some feed on green plants (27).2. Dermaptera—chewing insects; a few feed on green plants (1).3. Coleoptera—chewing insects; many feed on green plants; see table.4. Lepidoptera—chewing insects; larvae of many feed on green plants (4).5. Diptera—larvae of a few feed on green plants (2).6. Thysanoptera (thrips)—some are rasping and sucking plant feeders (10).7. Hemiptera—feed by sucking on green plantsSuborder Heteroptera, Families Myridae, and Piesmatidae ( 4)Suborder “Homoptera” a , see Table.Distribution of plant virus vectors among selected Homoptera and Coleoptera families. [From Nault(1997; Ann. Entomol. Soc. Am. 90, 521–541).]Order, Suborder, FamilyCommon Nameof Insect GroupNo. SpeciesDescribedNo. VectorSpeciesNo. VirusesTransmittedHomopteraAuchenorrhynchaCicadidae Cicada 3,200 0 0Membracidae Treehopper 4,500 1 1Cercopidae Spittlebug 3,600 0 0Cicadellidae Leafhopper 15,000 49 31Fulgoroidea Planthopper 19,000 28 24SternorrhynchaPsyllidae Psyllid 2,000 0 0Aleyroididae Whitefly 1,200 3 43Aphididae Aphid 4,000 192 275Pseudococcidae Mealybug 6,000 19 10ColeopteraChrysomelidae Leaf beetle 20,000 48 30CoccinellidaeLadybird3,500 2 7beetleCucurlionidae Weevil 36,000 10 4Meloidae Blister beetle 2,100 1 1a “Homoptera” is a widely used generic term for several suborders of the Hemiptera.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

228 12. PLANT-TO-PLANT MOVEMENTABCFIGURE 12.2 Major vectors of plant viruses. A. Aphid (Myzus persicae); note the stylet penetrating the leaf. B. Leafhopper(Circulifer tenellus). C. Whitefly (Bemisia tabaci).A. Relationships Between Plant Virusesand InsectsThe transmission of viruses from plant toplant by invertebrate animals is of considerableinterest from two points of view. First,such vectors provide the main method ofspread in the field for many viruses that causesevere economic loss. Second, there is muchbiological and molecular interest in the relationshipsbetween vectors and viruses, especiallyas some viruses have been shown tomultiply in the vector. Such viruses can beregarded as both plant and animal viruses.Even for those that do not multiply in the animalvector, the relationship is usually morethan just a simple one involving passive transportof virus on some external surface of thevector (“the flying pin”). Transmission byinvertebrate vectors is usually a complex phenomenoninvolving specific interactionsbetween the virus, the vector, and the hostplant, coupled with the effects of environmentalconditions. Most of the detailed studies onvirus transmission and virus-vector relationshipshave been made using aphids. Many ofthe features described following for aphidtransmission are applicable to transmissionby vectors from other insect orders.As a general rule, viruses that are transmittedby one type of vector are not transmittedby any of the others. This specificity is notonly at the level of vector type, family, genus,or species but can be even at the level of biotype.There are two basic interactions between virusesand their biological vector. They may be takenup internally within the vector, termed persistent,internally borne or circulative, or they maynot pass to the vector’s interior, in which casethey are termed nonpersistent, externally borne,or noncirculative (Box 12.3).IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRYBOX 12.3RELATIONSHIPS BETWEEN PLANT VIRUSES AND THEIR VECTORSThere are two major types of interaction between a virus and its vector: nonpersistent and persistent. Features of the interactionsare outlined in the table here, and the pathways that the viruses have with their vector are shown in the figure.Site invectorExternally– borneRetentiontime(halflife)InternallyborneVirus Transmission GroupType oftransmissionNonpersistentlytransmittedstylet-borneNonpersistentlytransmittedforegut-borne(semipersistent)PersistentcirculativePersistentpropagativeVirusproductinteractingwithvectorCapsid þ/helperfactorCapsid þ/helperfactorAcquisitiontime(max.dose)Seconds tominutesMinutes tohoursHours todaysHours todaysTranstadialpassageTransmission CharacteristicsVirus in vectorhaemolymphLatentperiodVirus Transovarialmultiplies transmissionin vectorMinutes No No No No NoHours No No No No NoDays toweeksWeeks tomonthsYes Yes Hours to NoNodaysYes Yes Weeks Yes OftenIII. TRANSMISSION BY INVERTEBRATES(continued)229

IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRYBOX 12.3(continued)Figure. Mechanism of transmission of plant viruses by arthropods with piercing-sucking mouthparts. The general anatomy of the alimentarysystem and the salivary system is shown; the areas relevant to virus transmission are labeled. One inset shows a detailed view of the distalend of the mouthparts where the food canal (FC) and the salivary canal (SC) empty into a common space. One current model of transmissionof stylet-borne (nonpersistent, noncirculative) viruses suggests that the transmissible virus is retained at the distal tip of the stylets and thenreleased by salivary secretions as the insect salivates during feeding. A second inset shows a detailed view of the foregut-borne (semipersistent,noncirculative) with virus particles attached to the cuticle lining of the foregut, a region that would include the sucking pump (SUP), pharynx(PX), and esophagus (ES). Notice that the virus is embedded in a matrix material attached to the cuticle. The origin or composition of the matrixmaterial is unknown. The circulative nonpropagative viruses pass through the foregut into the anterior midgut (AM), posterior midgut (PM), andthen into the hindgut (HG). They do not infect the gut cells but are transported through the posterior midgut and hindgut cells and releasedinto the haemocoel (body cavity). Current information indicates that these viruses specifically associate with the accessory salivary glands(ASG) and are transported across the ASG cells and then released into the salivary canal (SC). The circulative propagative viruses will infectmidgut cells and subsequently infect other tissues. These viruses ultimately associate with the principal salivary glands (PG) and possibleASG prior to their release into the SC. SP, salivary pump. [This article was published in Encyclopedia of virology, Vol. 1. S.M. Gray and D.M.Rochon (A. Granoff and R. Webster, Eds.), Vectors of plant viruses, pp. 1899–1910, Copyright Elsevier (1999).]230 12. PLANT-TO-PLANT MOVEMENT

III. TRANSMISSION BY INVERTEBRATES231Essentially, there are three stages in thetransmission cycle:1. The acquisition phase, in which thevector feeds on the infected plantand acquires sufficient virus fortransmission.2. The latent period, in which the vectorhas acquired sufficient virus but is not able totransmit it. For externally borne viruses, thereis little or no latent period.3. The retention period is the length of timeduring which the vector can transmit thevirus to a healthy host.B. Nonpersistent Transmissionby Insects1. Features of NonpersistentTransmissionOf the over 300 known aphid-borne viruses,most are nonpersistent. The following virusgenera have definite members transmitted ina nonpersistent manner: Alfamovirus, Caulimovirusby Myzus persicae, Closterovirus, Cucumovirus,Fabavirus, Macluravirus, and Potyvirus.These genera include viruses with helical andisometric particles and with DNA and RNAmono-, bi-, and tripartite genomes. There are noknown nonpersistent viruses transmitted byleafhoppers.Nonpersistently transmitted viruses areacquired rapidly from plants, usually in a matterof seconds. During this time, aphid’s styletdoes not usually penetrate beyond the epidermalcells, and when it penetrates beyondthe epidermis into the mesophyll and vasculartissue, the transmission rate declines rapidly.The initial host-finding behaviour of aphidsis short probing, thought to sample the epidermalcells 1 sap, and fits very well with thismechanism. Since the sampling is especiallybrief on nonhosts for the aphid, the vectors ofnonpersistent viruses are often noncolonisersof that species.With a nonpersistent virus there is little orno latent period, and aphids begin to lose theability to infect immediately after the acquisitionfeed. The rate at which infectivity is lostdepends on many factors, including temperatureand whether they are held on plants orunder some artificial condition.Different strains of the same virus may differin the efficiency with which they are transmittedby a particular aphid species. Some strainsmay not be transmitted by aphids at all. Differentstrains of the same nonpersistent virus donot usually interfere with each other’s transmission,as is sometimes found with propagativeviruses.Aphid species vary widely in the number ofdifferent viruses they can transmit. At oneextreme, M. persicae is known to be able to transmita large number of nonpersistent viruses,whereas other aphids transmit only one virus.These differences in part reflect the extent towhich different aphid species have been tested,but there is no doubt that real differences in versatilityoccur. Among species that transmit agiven virus, one species may be very much moreefficient than another. For instance, marked differenceswere found in the efficiency with whichPotato virus Y was transmitted by different specieseven when acquisition feed and test feedtimes were standardised (Figure 12.3). This canreflect the initial feeding behaviour of differentaphid species on the test plant species.2. Virus-Vector RelationshipsAs just noted, when they alight on a leaf,aphids may make brief probes into the leaf—often less than 30 seconds. Thus, the initialbehaviour of such aphids on reaching a leaf isideally suited to rapid acquisition of a nonpersistentvirus. Sap sampling on a virus-infectedplant will contaminate the stylet tips, the foodcanal, and the foregut. These sites have beenfavoured for the retention site of virus that willbe injected subsequently into a healthy plantfollowing another exploratory probing feed.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

232 12. PLANT-TO-PLANT MOVEMENTFIGURE 12.3 Relative efficiencies of three aphis speciesin transmitting Potato virus Y after defined acquisition feedingtimes. [Reprinted with permission from Bradley andRideout (1953; Can. J. Zool. 31, 333–341).]The weight of evidence favours the food canalin the maxillae as the site where infective virusis retained during nonpersistent transmission,but it must be remembered that this evidenceidentifies sites of accumulation but does notgive any indication as to whether that virus istransmitted.There are two phases to the interactioninvolved in nonpersistent virus transmission:retention of the virus at a specific site andrelease of the virus. All nonpersistently transmittedviruses have a simple structure of nucleicacid encapsidated in simple icosahedral or rodshapedparticles by one or more coat proteinspecies. Thus, it is the capsid protein that isavailable for any interactions. Two forms ofinteraction have been identified in the retentionphase: one in which there appears to be directinteraction between the virus capsid and thesite of retention in the aphid and the other inwhich a nonstructural virus-encoded protein isinvolved. This nonstructural protein is termeda helper component, helper factor, or aphid transmissionfactor.a. Direct Capsid Interaction. It is thoughtthat Alfalfa mosaic virus and Cucumber mosaicvirus (CMV) transmission involves direct linksbetween the capsid and the binding site withinthe aphid vector. The prime evidence is thatpurified virus can be transmitted from artificialfeeding systems without the addition of otherproteins or factors. Most of the detailed evidenceis for CMV. The efficiency of aphidtransmission of heterologously assembled particlesbetween the genomes and capsid proteinsof a highly aphid-transmitted (HAT) and apoorly aphid-transmitted (PAT) strain of CMVsegregated with the source of coat protein.The amino acid differences between the coatproteins of HAT and PAT strains of CMV areassociated with both vector transmissibilityand virion stability.The minor coat protein of Closteroviruses,which encapsidates only a teminal part of theviral genome to give “rattlesnake” particles(see Figure 5.1), is thought to be involved inaphid transmission.b. Indirect Interaction Involving HelperComponents. The interactions involving helpercomponents of two groups of viruses, the potyvirusesand caulimoviruses, have been studiedin detail. The helper components of potyvirusesare encoded by the virus and have the followingproperties in relation to virus transmission:• The helper factor of one potyvirus may ormay not permit the aphid transmission ofanother potyvirus when tested in an in vitroacquisition system. Thus, there is somespecificity in the phenomenon.• The helper component must be acquired bythe aphid either during or before virusacquisition. If it is provided after the virus,there is no transmission.• Potyvirus helper components have MWs inthe range of 53 kDa and 58 kDa. They arecleaved from the polyprotein encoded by thevirus as a product that, as well as helperIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

III. TRANSMISSION BY INVERTEBRATES233component activity, has various otheractivities including being a protease. Thus, itis termed HC-Pro.• Purified helper component can be used tofacilitate the transmission of potyviruses byfeeding through artificial membranes.• The biologically active form of helpercomponent appears to be a dimer.By studying the effects of mutations in the coatprotein and helper component on aphid transmissionof various potyviruses, a picture appearsof some of the molecular interactions involved.The current hypothesis is that the helper componentforms a bridge between the virus capsidand the aphid stylet (Figure 12.4A). Close to theN-terminus of the coat protein is the amino acidFIGURE 12.4 Models for the interactionsbetween viruses, transmissionhelper factors, and vector. A. Possibleinteractions between the HC-Pro, theaphid’s stylets, and the potyviral coatprotein. (1). Position of the virion particlesclose to the apical section of thefood canal. (2). A model assuming anassociation between two molecules ofHC-Pro. Note that one molecule ofthe HC-Pro is bound to a “receptor”on the stylet, while the second HC-Pro molecule is bound to the coat proteinsubunit. (3) A model assumingthat a dimer is needed to bind to the“receptor” on the stylet. Both HC-Promolecules are linked to coat proteinsubunits. (4) A proposed structuralbinding between the PTK motif ofHC-Pro and the DAG motif foundon the N-terminus of the coatprotein subunit. [This article was publishedin Virus-insect-plant interactions,B. Raccah, H. Huet, and S. Blanc(K.F. Harris, O.P. Smith, and J.E. Duffus,Eds.), Potyviruses, pp. 181–206,Copyright Elsevier (2001).] B. Interactioninvolved in the aphid transmissionof Cauliflower mosaic virus (CaMV). Thepanel on the left is from the figure inBox 12.3. The rest of the figure showsdiagrammatically the interactionsbetween a CaMV particle and theaphid’s stylet (see text). N and C arethe N-terminal and C-terminal regionsof P2, respectively.ABIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

234 12. PLANT-TO-PLANT MOVEMENTtriplet DAG (aspartic acid-alanine-glycine) that isimportant for transmissibility. Mutagenesisshows that both the DAG sequence itself andthe context of the surrounding amino acidsaffects transmissibility. Biochemical and immunologicalanalyses indicate that this N-terminalregion is located on the external surface of thevirus particle. Two important regions havebeen identified in HC-Pro. One, characterisedby the amino acid sequence PKT (prolinelysine-threonine),appears to be involved in theinteraction with the capsid protein. The other,termed the KITC (lysine-isoleucine-threoninecysteine)region, appears to be involved withthe HC-Pro retention on the aphid’s stylets.Little is known about the mechanisms ofrelease of nonpersistent viruses from the siteof binding in the aphid’s stylet, but three theorieshave been proposed:1. The mechanical transmission theorysuggests that the virus is simply inoculatedby the stylet.2. In the ingestion-egestion theory, release iseffected by regurgitation and salivation.3. Since the food and salivary canals of thestylets fuse near the tip of the maxillarystylet, nonpersistently stylet-borne virusescould be released by saliva alone.Cauliflower mosaic virus (CaMV), and presumablyother caulimoviruses, requires a helper component(or aphid transmission factor) when beingtransmitted by M. persicae. The CaMV helper componentsystem has the following properties:• As with potyviruses, the helper componentmust be acquired by the aphid either duringor before virus acquisition.• Helper components of other caulimovirusescan complement defects in CaMV helpercomponent.• The CaMV helper component system involvestwo noncapsid proteins: the 18 kDa product ofORFII (P2) and the 15 kDa product of ORFIII(P3); (for CaMV gene map, see Profile 4).• In infected cells, P2 is found in crystallineelectrolucent inclusion bodies (seeFigure 2.6) and P3 in association with virusparticles.• P2 interacts very strongly with microtubuleswith binding domains in two regions: onenear the N-terminus and the other near theC-terminus.Thus, the CaMV helper component system ismore complex that that of potyviruses. Thevirus can be transmitted from an in vitro acquisitionsystem containing baculovirus-expressedP2 and sap from a plant infected with a P2-defective isolate but not from P2 þ purifiedvirus; however, the virus can be transmittedwhen P3 is added to the purified virus. Secondarystructure predictions of P2 suggest twodomains: the N-terminus being predominantlyb-sheet and the C-terminus predominantly a-helix; the two domains are separated by a predictedrandom structure. The 61 amino acid C-terminal domain interacts with partially purifiedvirus and with the 30 N-terminal aminoacids of P3. Mutations of the N-terminal domainabolish its ability to facilitate transmission butdo not affect its ability to bind to semipurifiedvirus. This leads to a model for how the CaMVhelper system operates (Figure 12.4B). P3, whichforms a tetramer, binds to the virus capsid composedmainly of P4, with the C-terminus of P2binding to P3. The bridge is completed by theN-terminus of P2 binding to a nonglycosylatedprotein embedded in the chitin matrix of thecommon food/salivary duct of the aphid stylet(Figure 12.4B). The role of the microtubule bindingactivity of P2 is unknown, but it is noted thatthe microtubule binding domains overlap theP3 and aphid binding domains.As with nonpersistent viruses, nothing isknown about the molecular details of virusrelease from the vector, but as the virus isretained in the common food/salivary duct atthe tip of the aphid’s stylet, it is thought thatrelease could be effected by the aphid’s saliva.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

C. Persistent Transmission by InsectsThe main features of persistent transmissionare summarised in Box 12.3. Viruses transmittedin this manner are usually transmittedby one or a few species of aphid. Yellowingand leafrolling symptoms are commonly producedby infection with persistently transmittedviruses. Viruses that are internally bornein their aphid vectors may replicate in the vector(propagative) or may not (circulative). Foran aphid to become a transmitter by either typeof relationship, the virus has to be ingestedfrom the infected plant and reach the salivaryglands, usually via the hemolymph, to beegested into the healthy plant. Thus, it has topass at least two barriers: the gut wall and thewall of the salivary glands.1. Circulative Virusesa. Features of Circulative Virus: VectorInteraction. Circulative viruses are usuallyphloem-limited, and thus the vector must feedfor a longer time to acquire the virus (seeBox 12.3). Luteoviruses are the most studiedof the circulative (persistent) viruses for whichthere is no demonstration of replication in thevector.The minimum acquisition time can be as littleas 5 minutes but is usually several hours.This is followed by a latent period of at least12 hours, after which the virus can be transmittedwith an inoculation access time of 10 to30 minutes. The aphids then remain capableof transmitting for at least several days.As just noted, persistently transmittedviruses have to cross at least the gut and salivarygland barriers. Particles of Cereal yellow dwarfvirus-RPV (CYDV-RPV) associate only with thecell membranes of the hindgut of the aphidvector Rhopalosiphon padi. It is suggested thatthe particles entered the hindgut cells by endocytosisinto coated pits and coated vesicles andaccumulated in tubular vesicles and lysosomes(Figure 12.5A).III. TRANSMISSION BY INVERTEBRATES235Particles are then released into the haemocoelby fusion of the tubular vesicles with thebasal plasmalemma. Aphid salivary glandscomprise two principal glands and two accessoryglands. Potato leafroll virus particles havebeen seen in the basal lamina and plasmalemmainvaginations of accessory salivarycells (Figure 12.5B). Particles were also foundin tubular vesicles in the cytoplasm near salivarycanals and in coated pits connected tothe canal membrane. The basal lamina andthe basal plasmalemma function as independentbarriers to transmission of different luteovirus-aphidcombinations. From these studiesthe route that luteoviruses take across the twobarrier tissues in their aphid vector wouldappear to be by incorporation into coated vesiclesand transport across the cell(s). Thus, themain sites of interaction for the virus particlesis with the plasma membrane on the gut sideof the gut wall cells and with two plasmamembranes on the haemocoel side of the salivarygland accessory cells, which suggests areceptor-mediated interaction.Because purified luteoviruses can be aphidtransmitted from in vitro acquisition, it is likelythat no noncapsid proteins are involved. The capsidcomprises the major capsid protein and aminor amount of a larger protein translated viaa read-through of the coat protein stop codon(see Chapter 7). Particles containing just the majorcoat protein without any read-through proteinare not transmissible, which led to the widespreadassumption that the read-through portionwas required for aphid transmissibility. However,there is no clear picture of the luteoviruscomponent of the receptor-mediated recognition.Several aphid proteins of Mr ranging from31 to 85 kDa have been shown to interact withpurified luteoviruses in vitro. Antisera raisedagainst two of these proteins, P31 and P44,react specifically with extracts of accessorysalivary glands from vector aphids, suggestingthat these proteins might be involved in luteovirus-specificrecognition at this site.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

ABFIGURE 12.5 Models for interactions and transcellular transport of luteoviruses in aphid vectors. A. Transcellulartransport through aphid gut epithelium. Visualisation of endocytotic- and exocytotic-associated ultrastructure supportsreceptor-mediated endocytosis as a mechanism regulating vector-specific luteovirus acquisition. Based on this model, luteovirusesrecognised at the gut-cell apical plasmalemma (APL) bind to the membrane (1) initiating virus invagination (2) intocoated pits (CP). Coated pits bud off the APL as virus-containing coated vesicles (CV), which transport the virus (3) tolarger uncoated vesicles, called receptosomes (RS), which act to concentrate the virus (4). Tubular vesicles containing linearaggregates of virus form on the receptosomes (5) transport the virus to the basal plasmalemma (BPL) and fuse with the BPL,allowing release of the virus from the cell (6). Luteoviruses can then diffuse through the gut basal lamina (BL) and into thehaemocoel (7). Eventually, receptosomes (endosomes) mature into lysosomes (L), and any virus particles remaining in thelysosome are probably degraded. MT, microtubules. B. Luteovirus interactions with accessory salivary glands (ASG) determiningvector-specific transmission. Luteoviruses in the haemocoel first encounter the extracellular basal lamina (BL) surroundingthe ASG. The BL acts as a selective barrier to luteovirus transmission. Depending on the aphid biotype and thespecific luteovirus, the virus particles may be prevented from penetrating the BL (A) or may diffuse through (B, C) tothe basal plasmalemma (BPL). A second selective barrier occurs at the BPL. Luteovirus particles not recognised at theBPL remain outside the cell in the pericellular space (B). Luteoviruses recognised by putative virus receptors (C) on theBPL (1) are encytosed by coated pits (2) and accumulate into tubular vesicles (TV) in the cytoplasm (3). The TV adjacentto the microvilli-lined canals formed by the apical plasmalemma (APL) bud off coated vesicles (CVs) (4) containing individualvirions. The CVs transport the virus (5) to the canals, fuse with the APL (6), forming coated pits (CP), and release thevirus into the canal lumen allowing transport of luteovirus out of the aphid along with salivary secretions. PLI, plasmalemmainvagination; SD, salivary duct. [From Gildow (1999; in The Luteoviridae, H.H. Smith and H. Barker, Eds.,pp. 88–112, CAB International, Wallingford, UK).]

III. TRANSMISSION BY INVERTEBRATES237Another binding aphid protein, which interactswith many luteoviruses and other viruses,is the 60 kDa symbionin or GroEL from theendosymbiotic bacterium Buchneria spp. Thisprotein, found readily in aphid hemolymph, isa member of the molecular chaperone familythat is responsible for stabilising the structureof proteins. The interaction of luteoviruses withsymbionin is determined by the read-throughdomain of the minor capsid protein. The luteovirus-symbionininteraction is essential for theretention of the virus in the hemolymph.b. Dependent Transmission. As with certainnonpersistent viruses, some persistentviruses require a helper factor—in this case, avirus—to be present in the plant before aphidtransmission can occur. For persistent virusesdependent on another virus, it is the presenceof the virus itself in a mixed infection that providesthe assistance. This type of dependenttransmission is due to phenotypic mixingtogether during replication of the two virusesin the plant, resulting in the encapsidation ofthe genome of one virus in coat protein subunitsof the other virus.Umbraviruses do not encode a coat protein.For their aphid transmission they associatewith a helper luteovirus, which is presumedto supply the coat protein and thus aphidtransmission properties. Each definitive Umbravirusspecies is associated with a specific luteovirus.These systems have the followingcharacteristics:• Both viruses are transmitted in a circulativenonpropagative manner.• The dependent virus (umbravirus) is saptransmissible, but the helper is not.• The dependent virus is only transmitted byaphids from source plants that contain bothviruses; in other words, aphids alreadycarrying helper virus cannot transmit thedependent virus from plants infected onlywith this virus.• Evidence from a variety of experimentsindicates that the dependent virus istransmitted by the aphid vector only whenits RNA is packaged in a protein shell madeof the helper virus protein.This phenotypic mixing must take place indoubly-infected plants. Groundnut rosette virusdepends on its satellite RNA as well as onGroundnut rosette assistor virus for transmissionby Aphis craccivora (see Box 2.1).2. Propagative VirusesPropagative viruses can be considered to beviruses of the insect that have become adaptedto plants. Two plant virus families, Reoviridaeand Rhabdoviridae, and the Tenuivirus and Marafivirusgenera contain members that replicate intheir leafhopper vectors. Such replication usuallyhas little effect on the hoppers. However, from thevirus point of view, replication in the vector hastwo important consequences: Once they acquirevirus, the vectors normally remain infective forthe rest of their lives, and replication in the vectoris often associated with transovarial passage ofthe virus, thus giving it a means of survival overwinter that is quite independent of the host plant.With viruses that replicate in their vectors, thereis usually a high degree of specificity betweenvector and virus or even strains of a virus.Many of the virus:host interactions resemblethose found in animal viruses, with some genesthat adapt the virus to either animals or plants.In the plant reoviruses, particular genome segmentscode for gene products required for replicationin the insect but not in the plant. Therhabdo- and tospoviruses need glycoproteinspikes for infecting insects but not for plants.On the other hand, the sc4 gene product inplant rhabdoviruses (see Profile 13 for genomemap) that facilitates cell-to-cell spread in plantsis not found in animal rhabdoviruses.Whereas aphids are vectors of many of thepersistent circulative viruses, most of the persistentpropagative viruses are leafhopper- orIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

238 12. PLANT-TO-PLANT MOVEMENTplanthopper-transmitted. However, severalmembers of the Rhabdoviridae replicate in theiraphid vector, including Sowthistle yellow veinvirus (SYVV).The latent period of SYVV in the vector islong and depends strongly on temperature.Characteristic bacilliform particles have beenobserved in the nucleus and cytoplasm of cellsin the brain, subesophageal ganglion, salivaryglands, ovaries, fat body, mycetome, and muscle.Virus particles appear to be assembled inthe nucleus. The virus can be serially transmittedfrom aphid to aphid by injection ofhemolymph, and infection is associated withincreased mortality of the aphids. Decreasedlife span varies with different virus isolates.However, since infected aphids live throughthe period of maximum larviposition, theintrinsic rate of population growth was hardlyaffected. The virus is transmitted through theegg of Hyperomyzus lactucae, about 1 percent oflarvae produced being able to infect plants.Continuous passage of SYVV in the aphid bymechanical inoculation gives rise to isolatesthat have lost the ability to infect the plant host.3. Thrip Transmission of TospovirusesTransmission of tospoviruses by thrips hasseveral unusual features. Only the first and earlysecond larvae stages can acquire the virus andthe competence to acquire decreases with age ofthe larvae. Tomato spotted wilt virus (TSWV) canbe acquired or transmitted by first instar nymphsof Frankliniella occidentalis in feeding periods of asshort as 5 minutes, but the median acquisitionaccess period on infected Impatiens plants is morethan 100 minutes. The median latent period varieswith temperature being 84 hours at 27 o Cand171 hours at 20 o C. Individuals may retain infectivityfor life, but their ability to transmit may beerratic. The virus is not passed through the egg.As with other internally borne persistentlytransmitted viruses, tospoviruses have to pass severalbarriers in the vector, which suggests that thereis/are a receptor-mediated mechanism(s). TSWVis enveloped with spikes made up from twovirus-coded glycoproteins, extending from theenvelope (see Profile 17). Passage of TSWVthrough plants can result in envelope-deficientisolates. Feeding F. occidentalis on plantsinfected with wild-type and an envelopedeficientisolate showed that the thrips onlybecame infected when they acquired intactvirus particles. These and other experimentssuggest that the viral glycoproteins bind to areceptor in the vector’s midgut. Two proteinsfrom F. occidentalis have been shown to bindto TSWV glycoproteins, one of which is localisedto the larval thrip midgut and the otherpresent throughout the thrip’s body.A detailed study of the route that TSWV takesthrough F. occidentalis showed that the first infectionswere found in the midgut (Mg1) regionabout 24 hours postacquisition (hpa). Theseinfections increased in intensity but remainedrestricted to the Mg1 epithelium for some time.In late larval stage, it spread to the circularand longitudinal midgut muscle tissues. By theadult stage, the visceral muscle tissues of themidgut and foregut were infected. Infection ofthe salivary glands was first observed 72 hpa,and at the same time, the ligaments connectingthe midgut with the salivary glands becameinfected. There was no evidence for TSWV ineither the haemocoel or the midgut basal lamina.It appeared that the virus reached the salivaryglands through the ligaments connecting Mg1to the salivary glands. This is a different routeto that conventionally proposed for persistentviruses, which is movement through the haemocoelfrom the gut cells to the salivary glands.D. Virus Transmission by BeetlesLeaf-feeding beetles have chewing mouthpartsand do not possess salivary glands. Theyregurgitate during feeding, which bathes themouthparts in sap. This regurgitant will containvirus if the beetle has fed on an infected plant.Beetles can acquire virus after a single biteIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

III. TRANSMISSION BY INVERTEBRATES239and can infect a healthy plant with one bite.However, beetle transmission is not a purelymechanical process. There is a high degree ofspecificity between beetle vector and virus,and some very stable viruses, such as TMV,are not transmitted by beetles. The viruses thatare transmitted belong to the Tymovirus, Comovirus,Bromovirus, and Sobemovirus genera.Sometimes one beetle species will transmit aparticular virus with high efficiency, while arelated species does so inefficiently. It is suggestedthat the regurgitant fluid of the beetlescontains an inhibitor that prevents the transmissionof non-beetle-transmitted viruses butdoes not affect those that are transmitted.There is good evidence that the inhibitor is anRNase.E. Nematode Transmission of Viruses1. Features of Nematode TransmissionTwo genera of plant viruses are transmittedby nematodes. Nepoviruses are transmitted byspecies in the genera Xiphinema and Longidorus,and tobraviruses are transmitted by species ofTrichodorus and Paratrichodorus. All three tobravirusesare nematode transmitted, but onlyabout one-third of the nepoviruses are transmittedby these vectors. With the exception ofTobacco ring spot virus (TRSV), which is reportedto also have aphid vectors, none of the virusesin these two genera is known to have invertebratevectors other than nematodes; somenepoviruses are pollen transmitted.Nematodes are difficult vectors to deal withexperimentally because of their small size andtheir rather critical requirements with respectto soil moisture content, type of soil, and, to alesser extent, temperature. To overcome theseproblems, five criteria have been proposed forestablishing the nematode vectoring of viruses.• Infection of a bait plant must be demonstrated.• Experiments should be done with handpickednematodes.• Appropriate controls should be included toshow unequivocally that the nematode is thevector.• The nematode should be fully identified.• The virus should be fully characterised.A common method for detecting nematodetransmission has been to set out suitable “bait”plants (such as cucumber) in a sample of thetest soil. These plants are grown for a time toallow any viruliferous nematodes to feed onthe roots and transmit the virus and for anytransmitted virus to replicate. Extracts fromthe roots and leaves of the bait plants are theninoculated mechanically to a suitable range ofindicator species (see Chapter 13).2. Virus-Nematode RelationshipsThe nematode transmission of a virus hasbeen divided into seven discrete but interrelatedprocesses: ingestion, acquisition, adsorption,retention, release, transfer, and establishment.Ingestion is the intake of virus particles fromthe infected plant, and although it does notrequire a specific interaction between nematodeand virus, it needs a specific interaction betweenthe nematode and plant. In the acquisitionphase, the ingested virus particles are retainedin an intact state, and specific features on thesurface of the particle are recognised by receptorsites in the nematode feeding apparatus leadingto adsorption. Once adsorbed, infectious particlescan be retained in the nematode for monthsor even years but not after moulting. Release ofthe virus particles is thought to occur by achange in pH caused by saliva flow when thenematode commences feeding on a new plant.In the transfer and establishment phases, thevirus particles are placed in the healthy plantcell and start replicating and causing infection.There is specificity in the relationshipsbetween nematodes and the viruses they transmitwith often an apparent unique associationbetween the virus isolate and the vector species.There are some cases of different virus isolatesIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

240 12. PLANT-TO-PLANT MOVEMENTsharing the same vector species or, conversely,one particular virus isolate being transmittedby several nematode species. There are 13 trichodoridspecies known to be tobravirus vectors,but only one or two of these transmitseach tobravirus. There is a substantial degreeof specificity between the nematode vector andthe tobravirus serotype. Several nepovirusesare transmitted by more than one vector species,but there can be differences in the observationsunder laboratory and field conditions.Once acquired, viruses may persist in transmissibleform in starved Longidorus for up to12 weeks, in Xiphinema for about a year, andmuch more than a year in Trichodorus. Transmissiondoes not appear to involve replicationof the virus in the vector. Plant virus particleshave never been observed within nematodecells. Consistent with this is the fact that no evidencehas been obtained for virus transmissionthrough eggs of nematode vectors.Specificity of transmission does not appearto involve the ability to ingest active virus,since both transmitted and nontransmittedviruses have been detected within individualsof the same nematode species. Sites of retentionof virus particles within nematodes have beenidentified by electron microscopy of thin sections.Nepovirus particles are associated withthe inner surface of the odontostyle of variousLongidorus species and with the cuticular liningof the odontophore and esophagus of Xiphenemaspecies. Tobravirus particles have beenobserved absorbing to the cuticular lining ofthe esophageal lumen.The genetic determinants for the transmissibilityof the nepoviruses Raspberry ringspot virusand Tomato black ring virus are encoded byRNA2, which expresses, among other proteins,the viral coat protein (see Profile 6 for nepovirusgenome organisation).By making reciprocal pseudo-recombinantsbetween a nematode transmissible and a nontransmissibleisolate of the tobravirus Tobaccorattle virus (TRV), it was shown that transmissibilitysegregated with RNA2. As noted inProfile 15, tobravirus RNA2 is variable in size,and, besides encoding the viral coat protein,encodes one to three nonstructural proteins.A recombinant virus, in which the coat proteingene of a nematode nontransmissible isolate ofPea early browning virus (PEBV) was replacedwith that of a highly nematode transmissibleisolate of TRV, was not transmitted by nematodes,which indicated that more than one ofthe RNA2 genes was involved. Mutations inthe 29 kDa and the 23 kDa nonstructural genesof PEBV both abolished nematode transmissionwithout affecting particle formation, as didremoval of the C-terminal mobile region ofthe coat protein. It is suggested that the nonstructuralproteins may be transmission helpercomponents analogous to those in some aphidand leafhopper virus transmission systems.IV. FUNGAL TRANSMISSIONOF VIRUSESSeveral viruses have been shown to be transmittedby soil-inhabiting fungi or protists. Theknown vectors are members of the class Plasmodiophoromycetesin the division Myxomycota orin the class Chytridiomycetes in the divisionEumycota. Both classes include endoparasitesof higher plants. Species in the chytrid genusOlpidium transmit viruses with isometric particles,while species in two plasmodiophorusgenera, Polymyxa and Spongospora, transmitrod-shaped or filamentous viruses.The two chytrid vectors, Olpidium brassicaeand O. bornavirus, are characterised by havingposteriorly uniflagellate zoospores, while thoseof the three plasmodiophoral vectors, Polymyxagraminis, P. betae, and Spongospora subterranean,are biflagellate. All five species are obligateparasites of plant roots and have similardevelopment stages (Figure 12.6). They surviveIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

IV. FUNGAL TRANSMISSION OF VIRUSES241FIGURE 12.6 Life cycle of a plasmodiophoralfungus. On the lefthandside is the diploid stage in rootcells; on the right-hand side is thehaploid stage in root hairs. Betweenare the phases in the soil whereplant-to-plant virus transmission canoccur. [Reprinted from Matthews’plant virology, 4th ed., R. Hull, Transmission1: by invertebrates, nematodesand fungi, pp. 485–532,Copyright (2002), with permissionfrom Elsevier.]between crops by resting spores that producezoospores that infect the host. The zoosporesform thalli in the host cytoplasm. In the earlystages of infection, the cytoplasm of thalli isseparated from the host cytoplasm by a membrane,but later the thallus forms a cell wall.The entire thallus is converted into vegetativesporangia or resting spores.Various degrees of host specificity exist inboth the chytrid and plasmodiophoral vectors,with some isolates having a wide host rangeand others a narrow host range. The wide hostrange isolates tend to be better virus vectorsthan are the narrow ones.Two types of virus-fungal vector relationshipshave been recognised, termed in vitro and in vivo.The in vitro virus-vector relationship is foundbetween the isometric viruses of the Tombusviridaeand two Olpidium species. Virions from thesoil water adsorb onto the surface of the zoosporemembrane and are thought to enter the zoosporecytoplasm when the flagellum is “reeledin.” It is unknown how the virus passes fromthe zoospore cytoplasm to the host cytoplasm,but it is thought that this occurs early in fungalinfection of the root. Reciprocal exchange of thecoat proteins of Tomato bushy stunt virus (nottransmitted by O. bornavirus) and Cucumbernecrosis virus (CNV; transmitted by O. bornavirus)showed that the coat protein is involved in theuptake of the virus by the zoospore. One aminoacid in the coat protein of CNV is important forIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

242 12. PLANT-TO-PLANT MOVEMENTtransmissibility, and binding studies showedthat this is associated with recognition of thevirus by O. bornavirus zoospores.The model for the in vivo virus-vector relationshipis demonstrated by the interactionsbetween Beet necrotic yellow vein virus (BNYVV)and Polymyxa betae. The virus is within the zoosporeswhen they are released from the vegetativesporangia or resting spores and infects thenew host when these zoospores establish theirown infection of the root. The processes of virusacquisition and release by the zoospores areunknown. The read-through domain from thecoat protein (see Profile 2 for genome organisationof BNYVV) is implicated in the fungal transmission.BNYVV RNAs 3 and 4 also have anindirect effect on the transmission, most likelythrough controlling factors such as spread andaccumulation of the virus in the root system.V. VIRUSES OF OTHER KINGDOMSIn moving to a new host, viruses of vertebratesand invertebrates do not have to crossbarriers such as the cuticle and cell wall thatface plant viruses. As the respiratory and digestivetracts comprise very large areas of livingcells surrounded just by the plasmamembrane,most viruses of vertebrates are transmitted bythe respiratory and faecal-oral route. Some,loosely termed arboviruses (e.g., rhabdoviruses,bunyaviruses, and flaviviruses), aretransmitted by arthropods in a manner similarto that of propagative persistent plant viruses,the usual vectors being blood-sucking insectssuch as mosquitoes. From the insects’ point ofview, the vectors of arboviruses and the analogousplant viruses are the alternate host, thevertebrate, and plant. There is mother-to-childvertical transmission (analogous to seed transmission)of some viruses of vertebrates andinvertebrates. Bacterial viruses spread eitherthrough cell division or through the surroundingliquid medium.VI. SUMMARY• Plant viruses must cross two barriers—thecuticle and the cell wall—before they caninfect a plant; this is done by mechanicaldamage.• The plant virus can be introduced eitherfrom plant material or by a biological vector.• Introduction from plant material can be bymechanical damage (e.g., breaking leafhairs), through seed or pollen, or by graftingor vegetative propagation.• Biological vectors are invertebrates(arthropods or nematodes) and fungi andprotests.• Each plant virus is usually transmitted innature by just one of the precedingmethods.• Plant viruses have very specific and intricateinteractions with their biological vectors.• There are two basic interactions with insectvectors: nonpersistent or stylet borne, inwhich the virus interacts with the insect’smouth parts, and persistent or circulative, inwhich the virus passes though the insect’sgut wall into the haemocoel and then intothe salivary glands, from where it is injectedinto the plant.• The interactions involved the virus coat andin some cases additional virus geneproducts.• Interactions with nematode and fungalvectors are also detailed and involve thevirus coat protein and sometimes additionalviral factors.Further ReadingBlanc, S. (2008). Vector transmission of plant viruses.Encyclopedia of Virology, Vol. 5, 274–281.Brown, D.J.F., Robertson, W.M., and Trudgill, D.L. (1995).Transmission of viruses by plant nematodes. Annu.Rev. Phytopathol. 33, 223–249.Brown, J.K. and Czosnek, H.K. (2002). Whitefly transmissionof plant viruses. Adv. Bot. Res. 36, 65–100.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

VI. SUMMARY243Campbell, R.N. (1996). Fungal transmission of plantviruses. Annu. Rev. Phytopathol. 34, 87–108.Gray, S. and Gildow, F.E. (2003). Luteovirus-aphid interactions.Annu. Rev. Phytopathol. 41, 539–566.Hogenhaout, S.A., Ammar, E.D., Whitfield, A.E., andRedinbaugh, M.G. (2008). Insect vector interactions withpersistently transmitted plant viruses. Annu. Rev. Phytopathol.46, 327–359.Hull, R. (2002). Matthews’ Plant Virology. Academic Press,San Diego.Maule, A.J. and Wang, D. (1996). Seed transmission of plantviruses: A lesson in biological complexity. Trends Microbiol.4, 153–158.Ng., J.C.K. and Falk, B.W. (2006). Virus-vector interactionsmediating nonpersistent and semipersistent transmissionof plant viruses. Annu. Rev. Phytopathol. 44, 183–212.Ng., J.C.K. and Perry, K.L. (2004). Transmission of plantviruses by aphid vectors. Molec. Plant Pathol. 5, 505–511.Reisen, W.K. (2008). Vector transmission of animal viruses.Encyclopedia of Virology, Vol. 5, 268–273.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

This page intentionally left blank

C H A P T E R13Plant Viruses in the Field:Diagnosis, Epidemiology, andEcologyPlant viruses are a significant agricultural problem and cause major losses. In any diseasesituation it is important to know what virus is causing the problem, where it comes from,and how it spreads before developing control measures.O U T L I N EI. Diagnosis 245II. Epidemiology and Ecology 260III. Viruses of Other Kingdoms 268IV. Summary 268A. IntroductionI. DIAGNOSISThere are three basic situations in which thetechniques for recognising and identifying avirus are needed: diagnosis of a viral infectionin the field to determine whether it is causedby a known or unknown virus; detection of aknown virus, usually in an epidemiological orquarantine situation; and assay for a knownvirus—for instance, when purifying or manipulatingit.When appraising the relative merits of differentmethods, the following important factorsmust be considered:• What question is being addressed? Is onejust determining if the plant is virus infected,what virus is infecting the plant, or whatstrain of the virus is infecting the plant?Comparative Plant Virology, Second Edition245Copyright # 2009, Elsevier Inc. All rights reserved.

246 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGY• Sensitivity: How small an amount of viruscan be measured or detected?• Accuracy and reproducibility• Numbers of samples that can be processedin a given time by one operator• Cost and sophistication of the apparatus andmaterials needed• The degree of operator training required• Adaptability to field conditionsIt must always be remembered that diseasedplants in the field may be infected by more thanone virus. Thus, an early step in diagnosis for anunknown disease must be to determine whethermore than one virus is involved. The methodsinvolved in assay, detection, and diagnosis canbe placed in four groups according to theproperties of the virus on which they depend:biological activities, physical properties of thevirus particle, properties of viral proteins, andproperties of the viral nucleic acid.B. Methods Involving Biologyof the VirusBiological methods for the assay, detection,and diagnosis of viruses are much more timeconsumingthan most other methods nowavailable. Nevertheless, they remain very important.For diagnosis, in most circumstances onlyinoculation to an appropriate host species candetermine whether a particular virus isolatecauses severe or mild disease. However, thisgroup of methods does have a major drawback,especially when used on a new virus or in quarantinesituations. It raises the possibility that theproduction of infected plants might be a sourceof infection for local crops even in the face ofthe strictest containment conditions.1. Indicator HostsDisease symptoms on plants in the field arefrequently inadequate on their own to give apositive identification. This is particularly sowhen several viruses cause similar symptoms,as do yellowing diseases in beet; when a singlevirus, such as Cucumber mosaic virus, is veryvariable in the symptoms it causes; or whenboth of these factors are relevant in a singlehost. Thus, since the early days of plant virology,searches have been made for suitable speciesor varieties of host plant that will giveclear, characteristic, and consistent symptomsfor the virus or viruses being studied, usuallyunder glasshouse conditions. Many good indicatorspecies have been found in the generaNicotiana, Solanum, Chenopodium, Cucumis, Phaseolus,Vicia, and Brassica. Certain plant speciessuch as Chenopodium amaranticolor, C. quinoa,and Nicotiana benthamiana react to a wide rangeof viruses.Testing large numbers of samples usingindicator hosts requires glasshouse facilitiesthat may be occupied for weeks or longer.However, the actual manipulations involvedin mechanical inoculation may take less timeper sample than many other methods of testing,and many attempts have been made tostreamline the procedures for sap extractionand inoculation. As just noted, care must betaken to keep inoculated test plants under suitablecontainment to minimise escape of thevirus(es) to the outside environment.2. Host RangeHost range is an important criterion in diagnosis.However, there are various factors thatmust be considered:• In many of the reported host range studies,only positive results have been recorded.• Absence of symptoms following inoculationof a test plant has not always beenfollowed by back inoculation to an indicatorspecies to test for masked infection.• The manner of inoculation may well affectthe results. Mechanical inoculation hasalmost always been used in extensive hostrange studies because of its convenience, butmany plants contain inhibitors of infectionIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

I. DIAGNOSIS247that prevent mechanical inoculation to thespecies, or from it, or both.• In studying large numbers of species, it isusually practicable to make tests only underone set of conditions, but it is known that agiven species may vary widely insusceptibility to a virus, depending on theconditions of growth.• Quite closely related strains of a virus maydiffer in the range of plants they will infect.Host range data may apply only to the virusstrain studied.3. Methods of TransmissionThe different methods of virus transmission,discussed in Chapter 12, may be useful diagnosticcriteria. Their usefulness may depend on theparticular circumstances. For example, a viruswith an icosahedral particle that is transmittedthrough the seed and by nematodes is veryprobably a nepovirus. On the other hand, sucha virus transmitted mechanically and by theaphid Myzus persicae might belong to any oneof several groups.4. Cytological EffectsCytological effects (see Chapter 2) detectableby light microscopy can sometimes be usedeffectively to supplement macroscopic symptomsin diagnosis. The light microscope has anumber of advantages when the inclusions arelarge enough to be easily observed:• It is a readily available instrument.• Specimen preparation techniques can besimple.• There is a wide field of view, thus allowingmany cells to be readily examined.• A variety of cytochemical procedures areavailable.5. Mixed InfectionsSimultaneous infection with two or evenmore viruses is not uncommon. Such mixedinfections may complicate a diagnosis basedon biological properties alone, especially if thehost response is variable. However, severalpossible differences in biological propertiesmay be used to separate the viruses:• If one virus is confined to the inoculated leafin a particular host, while the other movessystemically• If a host can be found that only one of theviruses infects• If the two viruses cause distinctive locallesions in a single host• If the two viruses have different methods oftransmission—for example, by differentspecies of invertebrate vectorsOn the other hand, difficulties may arise insorting out certain diseases in the field thatresult from a more or less stable associationbetween two or more viruses. For instance,groundnut rosette disease is caused by twoviruses and a satellite (see Box 2.1).C. Methods That Depend on PhysicalProperties of the Virus Particle1. Stability and Physicochemical PropertiesA virus has certain independently measurableproperties of the particle that depend onits detailed composition and architecture.These properties can be useful for identificationand as criteria for establishing relationships.The most commonly measured properties aredensity, sedimentation coefficient and diffusioncoefficient, and ultraviolet absorption spectrum(Box 13.1).2. Electron MicroscopyKnowledge of the size, shape, and any surfacefeatures of the virus particle is a basic requirementfor virus identification. Electron microscopycan provide this information quickly and,in general, reliably. For the examination of virusparticles in crude extracts or purified preparations,a negative-staining procedure is now usedIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

BOX 13.1ULTRAVIOLET SPECTRA OF VIRUSESThe ultraviolet absorption spectrum of a virus is a combination of the absorption spectra of the nucleicacid and the coat protein (Figure A). Nucleic acids have maximum absorption at about 260 nm and atrough at about 230 nm, whereas the absorption spectrum for proteins peaks at about 280 nm andtroughs at about 250 nm. The specific absorption for nucleic acids (20–25 OD units/mg/ml/ at260 nm) is much greater than that for proteins (about 1 OD unit/mg/ml/ at 280 nm). The combinationgives an absorption spectrum for a virus that peaks at about 260 nm and troughs at about 230–240 nm(Figures A, B).ABFig. Ultraviolet absorption spectra. A. Comparative absorption spectra of 1 mg/ml RNA, a virus containing 20%RNA (1 mg/ml) and a typical protein (10 mg/ml); note the shoulder on the protein spectrum at 290 nm due toabsorbance by tryptophan. [From Hull (1985; in Virology, a practical approach, B.W.J. Mahy, Ed., pp. 1–14, IRL Press,Oxford).] B. Absorption spectra of a suspension of (a) Turnip yellow mosaic virus (TYMV) RNA; (b) TYMV nucleoproteinparticles (33% RNA); (c) Cucumber mosaic virus particles (19% RNA); (d) Tobacco mosaic virus particles (5% RNA);and (e) TYMV protein particles (top component). All suspensions contain the same weight of material except (a),which contained half as much as the others. [From Gibbs and Harrison (1976; Plant virology, the principles, EdwardArnold, London).]Thus, this property depends mainly on the ratio of nucleic acid to protein in the virus; the aminoacid composition of the protein can also influence the spectrum. Many unrelated viruses with similarcompositions have similar absorption spectra. Ultraviolet absorption provides a useful assay for purifiedvirus preparations, provided other criteria have eliminated the possibility of contamination withnonviral nucleic acids or proteins, especially ribosomes. Measurements of A 260 may be unreliable forrod-shaped viruses that vary substantially in their degree of aggregation from one preparation toanother, thus leading to changes in the amount of light scattering for a given virus concentration.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

I. DIAGNOSIS249almost universally. Of particular use for rapidresults is the epidermal strip technique, in whichthe broken surface of a strip of epidermis isremoved from the leaf is wiped throughnegative stain on the electron microscope grid.Commonly used negative stains are sodiumphosphotungstate, ammonium molybdate, oruranyl acetate, depending on the stability of thevirus to these stains.Approximate particle dimensions can bedetermined. This is particularly useful for rodshapedviruses for which particle length distributionscan be obtained. Surface features maybe seen best in isolated particles on the grid.Depending on size and morphology, a virusmay be tentatively assigned to a particular taxonomicgroup. However, some small icosahedralviruses cannot be distinguished from membersof unrelated groups on morphology alone.For many viruses, examination of thin sectionsby electron microscopy is a valuable procedurefor detecting virus within cells andtissues, but this has its limitations such as differentiatingvirus particles from normal cellconstituents. The large, enveloped viruses, theplant reoviruses, and the rod-shaped virusescan usually be readily distinguished becausetheir appearance in thin sections generally differsfrom that of any normal structures.D. Methods That Depend on Propertiesof Viral ProteinsSome of the most important and widely usedmethods for assay, detection, and diagnosisdepend on the surface properties of viral proteins.For most plant viruses, this means theprotein or proteins that make up the viral coat.Different procedures may use the protein in theintact virus, the protein subunits from disruptedvirus, or proteins expressed fromcloned cDNA or DNA in a system such asE. coli or insect cells. More recently nonstructuralproteins coded for by a virus have beenused in diagnosis.1. SerologySerological procedures are based on theinteraction between a protein or proteins(termed the antigen) in the pathogen with antibodiesraised against them in a vertebrate. Thecomponents of a serological reaction are shownin Box 13.2.2. Types of AntiseraThere are two basic types of antisera: polyclonal,which contain antibodies to all the availableepitopes on the antigen, and monoclonal(Mab), which contain antibodies to one epitope.There is much discussion as to which is the bestfor diagnosis, but this will depend on whatquestion the diagnostician is addressing.Monoclonal antisera are much more specificthan polyclonal antisera and can be used to differentiatestrains of many pathogens. On theother hand, specificity can be a disadvantage,and a variant of the pathogen may not bedetected.The bringing together of two technologies,the production of single-chain antibodies, andthe display of recombinant proteins on the surfaceof bacteriophage (phage display) haveresulted in the ability to produce a large rangeof molecules with the properties of single-typeantibodies similar to Mabs without involvinginjection of animals.3. Methods for Detecting Antibody-VirusCombinationA wide variety of methods has been developedfor demonstrating and estimating combinationbetween antibodies and antigens. Themost widely used are the enzyme-linkedimmunoabsorbent assay (ELISA), immunoabsorbentelectron microscopy (ISEM), and “dotblots” that employ either polyclonal or monoclonalantibodies.a. ELISA Procedures. Many variations of thebasic procedure have been described (Box 13.3),IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

250 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGYBOX 13.2COMPONENTS OF A SEROLOGICAL REACTION1. Antibodies: An antibody is a molecule that binds to a known antigen. Antibodies are secreted byB lymphocytes. Structurally they are composed of one or more copies of a characteristic unit that canbe visualised as forming a Y shape (Fig.).Fig. IgG antibody molecule. Arrows indicate specific sites where the molecule is cleaved by the enzymes papain andpepsin to give the Fab, Fc, and (FabN) 2 fragments. L, light chains; H, heavy chains. The four polypeptide moleculesare joined by disulfide bridges. The two antigen-combining sites are made up from the variable regions of the L andH chains. [This article was published in Serology and immunochemistry of plant viruses, M.H.V. Van Regenmortel,Copyright Elsevier (1982), Academic Press, New York.]Each Y contains four polypeptides: two identical copies of the heavy chain and two identical copies ofthe light chain joined by disulfide bonds. Antibodies are divided into five classes—IgG, IgM, IgA, IgE, andIgD—based on the number of Y-like units and the type of heavy-chain polypeptide they contain.IgG molecules have three protein domains. Two of the domains, forming the arms of the Y, are identicaland are termed the Fab domain. They each contain an antigen-binding site at the end, making theIgG molecule bivalent. The third domain, the Fc domain, forms the stem of the Y. The three domainsmay be separated from each other by cleavage with the protease papain. The Fc region binds proteinA, a protein obtained from the cell wall of Staphylococcus aureus, with very high affinity. This propertyis used in several serological procedures.(continued)IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

I. DIAGNOSIS251BOX 13.2(continued)The N-terminal regions of both the light and heavy chains in the arms of the Y-shaped IgG moleculecomprise very heterogeneous sequences. This is known as the variable (V) region. The C-terminal regionof the light chains and the rest of the heavy chains form the constant (C) region. The V regions of one heavyand one light chain combine to form one antigen binding site. The heterogeneity of the V regions providesthe structural basis for the large repertoire of binding sites used in an effective immune response.2. Antigens: Antigens are usually fairly large molecules or particles consisting of, or containing, proteinor polysaccharides that are foreign to the vertebrate species into which they are introduced. Most havea molecular weight greater than 10 kDa, although smaller peptides can elicit antibody production.There are two aspects to the activity of an antigen. First, the antigen can stimulate the animal to produceantibody proteins that will react specifically with the antigen. This aspect is known as the immunogenicityof the antigen. Second, the antigen must be able to combine with the specific antibodyproduced. This is generally referred to as the antigenicity of the molecule.Large molecules are usually more effective immunogens than small ones. Thus, plant viruses containingprotein macromolecules are often very effective in stimulating specific antibody production;the subunits of a viral protein coat are much less efficient.It is specific regions on antigens, termed epitopes, which induce and interact with specific antibodies. Epitopescan be composed of amino acids, carbohydrates, lipids, nucleic acids, and a wide range of syntheticorganic chemicals. About 7–15 amino acids at the surface of a protein may be involved in an antigenic site.There are several different grouping of epitopes:The continuous epitope is a linear stretch of amino acids.The discontinuous epitope is formed from a group of spatially adjacent surface residues broughttogether by the folding of the polypeptide chain or from the juxtaposition of residues from two or moreseparate peptide chains.Cryptotopes are hidden epitopes revealed only on dissociation or denaturation of the antigen.Neotopes are formed by the juxtaposition of adjacent polypeptides (e.g., adjacent viral coat proteinsubunits).Metatopes are epitopes present on both the dissociated and polymerised forms of the antigen.Neutralisation epitopes are specifically recognised by antibody molecules able to neutralise the infectivityof a virus.The epitope type giving rise to a monoclonal antibody (MAb) and the relative proportions of differentepitopes recognised by a polyclonal antiserum can affect the outcome of certain serological tests.For instance, an antibody to a cryptotope is unlikely to recognise an antigen in DAS-ELISA but is likelyto in a Western blot.3. Interaction Between Antibodies and Antigens. The interaction between an epitope and antibody isaffected by both affinity and avidity. Affinity is a measure of the strength of the binding of an epitopeto an antibody. As this binding is a reversible bimolecular interaction, affinity describes the amount ofantibody-antigen complex that will be found at equilibrium. High-affinity antibodies perform better inall immunochemical techniques, not only because of their higher capacity but also because of thestabilityof the complex. Avidity is a measure of the overall stability of the complex between antibodies and(continued)IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

252 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGYBOX 13.2(continued)antigens and is governed by three factors: the intrinsic affinity of the antibody for the epitope, thevalency of the antibody and antigen, and the geometric arrangement of the interacting components.Titre is a relative measure of the concentration of a specific antibody in an antiserum. It is often used todefine the dilution end point of the antiserum for detection of an antigen. Thus, as the sensitivities of variousserological tests differ, the apparent titre is applicable only to the test under discussion.The serological differentiation index (SDI) is a measure of the serological cross-reactivity of two antigens.It is the number of two-fold dilution steps separating the homologous and heterologous titres.The homologous titre is that of the antiserum with respect to the antigen used for immunising the animal,while the heterologous titre is that with respect to another related antigen.BOX 13.3ELISA PROCEDURESDirect Double-Antibody Sandwich MethodThe principle of the direct double-antibody sandwich procedure is summarised in the Figure (A). Thetechnique is widely used but suffers two limitations:1. It may be very strain specific. For discrimination between virus strains, this can be a useful feature, butfor routine diagnostic tests, it means that different viral serotypes may escape detection. This highspecificity is almost certainly due to the fact that the coupling of the enzyme to the antibody interfereswith weaker combining reactions with strains that are not closely related.2. It requires a different antivirus enzyme-antibody complex to be prepared for each virus to be tested.AFig. Immune detection of viruses. A. Principle of the ELISA technique for plant viruses (direct double-antibodysandwich method). (1) The gamma globulin fraction from an antiserum is allowed to coat the surface of wells in apolystyrene microtitre plate. The plates are then washed. (2) The test sample containing virus is added and combinationwith the fixed antibody is allowed to occur. (3) After washing again, enzyme-labeled specific antibody is allowedto combine with any virus attached to the fixed antibody. For instance, alkaline phosphatase is linked to the antibodywith glutaraldehyde.) (4) The plate is again washed and enzyme substrate is added. The colourless substrate p-nitrophenylphosphate (open circle) gives rise to a yellow product with alkaline phosphatase (filled circle), which can beobserved visually in field applications or measured at 405 nm using an automated spectrophotometer. [Modified fromClark and Adams (1977; J. Gen. Virol. 34, 475–483).](continued)IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

BOX 13.3(continued)Indirect Double-Antibody Sandwich MethodsIn the indirect procedure, the enzyme used in the final detection and assay step is conjugated to an antiglobulinantibody. For example, if the virus antibodies were raised in a rabbit, a chicken antirabbit globulinmight be used. Thus, one conjugated globulin preparation can be used to assay bound rabbitantibody for a range of viruses.Many variations of these procedures are possible (Figure (B)).Direct proceduresindirect proceduresEEShort procedures withpreincubation of twoingredients.EEEE= Virus= Enzyme= intact virus specific antibody= Fab of F(ab) 2 fragment of virus specific antibody= antibodies specific for IgG or Fc fragmentsEEE(i) Black and white symbols indicate antibodies derived from two different animal species.(ii) Shaded areas indicate a reaction between the Fc portion of a virus-specific antibodyand an Fc-specific antibody(iii) Reagents added in order from bottom of diagram except where vertical bars indicatetwo reagents are preincubated together before addition to the plate.EBFig.B. Diagrammatic representation of some variants of ELISA. [Modified from Koenig and Paul (1982; J. Virol. Methods5, 113–125).]The direct double-antibody sandwich method is the most convenient for the routine detection ofplant viruses in situations where strain specificity and very low virus concentrations cause no problems.Dot ELISASeveral procedures have been developed that use nitrocellulose or nylon membranes as the solid substratefor ELISA tests. For the final colour development, a substrate is added that the enzyme linked tothe IgG converts to an insoluble coloured material. An example of a dot blot assay is given in Figure (C).(continued)IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

254 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGYBOX 13.3(continued)Fig.C. The sensitivity of a dot immunobindingassay. V ¼ purified Strawberry pseudo mild yellowedge virus; I ¼ crude sap from infected leaves;and H ¼ crude sap from healthy leaves. The conjugatedenzyme was alkaline phosphatase andthe substrate for colour development was FastRed TR salt. [From Yoshikawa et al. (1986; Ann.Phytopathol. Soc. Jpn. 52, 728–731).]CThe main advantages of the method are speed (three hours to complete a test), low cost, and the smallamount of reagents required. Dot blot immunoassays may be particularly useful for routine detection ofvirus in seeds or seed samples, especially for laboratories where an inexpensive and simple test isneeded. However, endogenous insect enzymes may interfere with tests using insect extracts.Immuno-Tissue PrintingAs an extension to dot ELISA, immuno-tissue printing involves applying the cut surface of a leaf or stem(or any other plant organ) to a nitrocellulose membrane and revealing the presence of an antigen (sayviral coat protein) by immunoprobing.Tissue printing has several advantages:1. It gives detailed information on the tissue distribution of a virus.2. Extraction of sap from leaves in which a virus has limited tissue distribution leads to the dilution ofthe virus with sap from uninfected cells. Since this technique samples the contents of each cell on thecut surface individually there is increased sensitivity.3. The technique is easily applicable to field sampling; tissue blots can be made in the field, and there isno need to collect leaf samples for sap extraction in the laboratory.with the objective of optimising the tests forparticular purposes. The method is very economicalin the use of reactants and readilyadapted to quantitative measurement. It canbe applied to viruses of various morphologicaltypes in both purified preparations and crudeextracts. It is particularly convenient whenlarge numbers of tests are needed. It is verysensitive, detecting concentrations as low as1–10 ng/ml.As with other diagnostic tests, it is necessaryto define and optimise the time of samplingand the tissue to be sampled to achieve a reliableroutine detection procedure. The main factorlimiting the number of tests that can be doneusing ELISA procedures is the preparation oftissue extracts.When many hundreds of field samples have tobe processed, it is often necessary to store themfor a time before ELISA tests are carried out.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

I. DIAGNOSIS255Storage conditions may be critical for reliableresults. For this reason, conditions need to beoptimised for each virus and host.b. Serologically Specific Electron Microscopy.A combination of electron microscopyand serology can provide rapid diagnosis andreveal features of a virus. The method offers adiagnostic procedure based on two propertiesof the virus: serological reactivity with the antiserumused and particle morphology. Variousterms have been used to describe the process:serologically specific electron microscopy(SSEM), immunosorbent electron microscopy(ISEM), solid-phase immune electron microscopy,and electron microscope serology. Basicallythere are two approaches. Virus particlescan be trapped on an electron microscope gridpreviously coated with an antiserum. The particlescan be then negatively stained and reactedwith another antiserum that decorates particlesof the virus against which the antiserum isdirected (Figure 13.1A). The other method is,after being adsorbed onto the EM grid, thevirus particles are coated with a virus-specificantibody to which gold particles have beenattached (Figure 13.1B).This method has many advantages:1. The result is usually clear in the form ofvirus particles of a particular morphology,and thus false positive results are rare.2. Sensitivity may be of the same order aswith ELISA procedures and may be 1,000times more sensitive for the detection ofsome viruses than conventional EM.3. When the support film is coated with anantibody, much less host backgroundmaterial is bound to the grid.4. Antisera can be used without fractionationor conjugation, low-titre sera can besatisfactory, and only small volumes arerequired.5. Very small volumes (0.1 ml) of virus extractmay be sufficient.6. Antibodies against host components arenot a problem inasmuch as they do notbind to virus.7. One antiserum may detect a range ofserological variants (on the other hand, theuse of monoclonal antibodies may greatlyincrease the specificity of the test).8. Results may be obtained within 30minutes.ABFIGURE 13.1 Immune specific electron microscopy (ISEM). A. A natural mixture of two potyviruses from a perennialcucurbit, Bryonia cretica. The antiserum used has decorated particles of only one of the viruses in the mixture. One particlenear the centre is longer than normal and decorated for only part of its length. This particle probably arose by end-to-endaggregation between a particle of each of the two viruses. Bar ¼ 100 nm. [From Milne (1991; in Electron microscopy of plantpathogens, K. Mendgen and D.E. Lesemann, Eds., pp. 87–102, Springer-Verlag, New York).] B. Gold particle labeling of virusparticles. Rice tungro bacilliform virus (RTBV) particles were treated first with anti-RTBV protease rabbit antiserum and thenwith gold-labeled goat antirabbit serum. Bar ¼ 200 nm. [This article was published in Virology, 205, J.M. Hay, F. Grieco,A. Druka, M. Pinner, S.-C. Lee, and R. Hull, Detection of rice tungro bacilliform virus gene products in vivo, pp. 430–437,Copyright Elsevier (1994).]IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

256 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGY9. When decoration is used, unrelatedundecorated virus particles on the grid arereadily detected (Figure 13.1A).10. Different proteins on a particle can bedecorated (e.g., the two coat proteins onclosteroviruses; see Figure 5.1).11. Prepared grids may be sent to a distantlaboratory for application of virus extractsand returned to a base laboratory forfurther steps and EM examination.The following are some disadvantages of theprocedure:1. It will not detect virus structures too smallto be resolved in the EM (e.g., coat proteinmonomers).2. Sometimes the method works inconsistentlyor not at all for reasons that are not wellunderstood.3. It involves the use of expensive EMequipment, which requires skilled technicalwork and is labour-intensive. For thesereasons it cannot compete with, say, ELISAfor large-scale routine testing.4. When quantitative results are required,particle counting is laborious, variability ofparticle numbers per grid square may behigh, and control grids are required.c. Electrophoretic Procedures. Electrophoresisin a suitable substrate separates proteinsaccording to size and net electric charge at thepH used. Polyacrylamide gel electrophoresis ina medium containing sodium dodecyl sulfate(SDS-PAGE) is commonly used. The positionand amount of the proteins can then be visualisedby a nonspecific procedure such as stainingor by a specific procedure such as immunoassay(termed Western blotting).d. Dot Blots. Protein dot blots resemble Westernblotting, except that the constituent proteinsin the sample are not separated. The samplesare spotted onto a solid matrix such as nylon ornitrocellulose and probed with an antiserum towhich a reporter group has been attached (seeFigure C in Box 13.3). Reporter groups can beeither enzymes that give a colour reaction witha specific chemical or a fluorescent compound.Unlike ELISA, the colour product must beinsoluble.Similar to dot blots, there are commerciallyavailable kits that use the same technique aspregnancy testing kits, in which the antibodyis immobilised on a strip and the sample isdiffused along the strip.E. Methods That Involve Propertiesof the Viral Nucleic AcidGeneral properties of a viral nucleic acid,such as whether it is DNA or RNA, double-(ds) or single-stranded (ss), or if it consistsof one or more pieces, are fundamental forallocating an unknown virus to a particularfamily or group. However, with the exceptionof dsRNA, these properties are usually of littleuse for routine diagnosis, detection, or assay.The ability to make DNA copies (cDNA) ofparts or all of a plant viral RNA genome hasopened up many new possibilities. The nucleotidesequence of the DNA copy can be determined,but this is far too time-consuming tobe considered as a diagnostic procedure, exceptin special circumstances.Basically, the four approaches to usingnucleic acids for detection and diagnosis ofviruses are the type and molecular sizes of thevirion-associated nucleic acids; the cleavagepattern of viral DNA or cDNA; hybridisationbetween nucleic acids; and the polymerasechain reaction.1. Type and Size of Nucleic AcidDouble-stranded RNAs are associated withplant RNA viruses in two ways: The plant reoviruses,and cryptoviruses have genomes thatconsist of dsRNA pieces, and in tissues infectedwith ssRNA viruses, a ds form of the genomeRNA accumulates that is twice the size of theIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

I. DIAGNOSIS257genomic RNA. This is known as the replicativeform (Figure 8.1). These dsRNAs have beenused for diagnosis following characterisationof the ds forms by PAGE.2. Cleavage Patterns of DNACleaving cDNAs of RNA genomes and thegenomes of DNA viruses at specific sites withrestriction enzymes and determining the sizesof the fragments by PAGE are possible proceduresfor distinguishing viruses in a particulargroup. For instance, isolates of Cauliflower mosaicvirus could be distinguished on the restrictionendonuclease patterns of the virion DNA.3. Hybridisation ProceduresThese procedures depend on the fact that ssnucleic acid molecules of opposite polarity andwith sufficient similarity in their nucleotidesequence will hybridise to form a ds molecule.The theory concerning nucleic acid hybridisationis complex. The Watson and Crickmodel for the structure of dsDNA showedthat the two strands were held together byhydrogen bonds between specific (complementary)bases—namely, adenine and thymidine,cytosine and guanine. This interaction of basepairingis the basis of all molecular hybridisation.Essentially, there are two stages, disruptionof the base-pairing (termed melting ordenaturing the nucleic acid) and reinstatementof base-pairing (termed reassociation, renaturation,or hybridisation). Among the factors thataffect denaturation are temperature, nucleicacid composition, type and concentration of saltin the buffer, the pH of the buffer, presence oforganic solvents, and base-pair mismatch. Factorsthat affect reassociation include temperature,salt concentration, base-pair mismatch,the length of the nucleic acid fragments, the typeof nucleic acid (DNA and/or RNA), the concentrationof nucleic acid, and the presence of variousanionic polymers.For diagnosis of virus by hybridisation, theunknown nucleic acid is termed the target andthe known nucleic acid is termed the probe.In many of the systems the target is immobilisedon a solid matrix and the probe appliedin liquid. The probe comprises sequences complementaryto the target sequences to which areporter system, either radioactive or nonradioactive,is attached to reveal when hybridisationhas taken place. There are three basic types ofnonradioactive reporter: those that directlymodify bases in the probe DNA, those thatattach precursors (e.g., horseradish peroxidaseor alkaline phosphatase) to the probe DNA orRNA, and those that incorporate labeled precursors(e.g., biotinylated nucleotides) into theprobe. The detection of the reporter group isusually by an enzyme that gives a colouredproduct or a luminescent compound on reactionwith a substrate. The coloured producthas to be insoluble (unlike the coloured productof the ELISA reaction).Excess probe hybridised to the target immobilizedon a solid matrix has to be removedbefore detection. This is done by washing atselected temperatures and salt concentrations.At this stage relatedness of the probe to thetarget can be ascertained by using different“stringencies” that relate to the effect of hybridisationand/or wash conditions on the interactionbetween complementary nucleic acidsthat may be incompletely matched. The use ofdifferent stringencies is one of the more powerfultools of the hybridisation technology. Thereare various hybridisation formats (Box 13.4).4. Dot BlotsDot blots, as just described for proteins, arewidely used for detecting nucleic acids usinglabeled probes (see Box 13.4).5. Polymerase Chain ReactionDNA fragments of interest can be enzymaticallyamplified in vitro by the polymerasechain reaction (PCR). The technique involvesthe hybridisation of synthetic complementaryoligonucleotide primers to the target sequenceIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

258 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGYBOX 13.4HYBRIDISATION FORMATSSouthern BlottingSee standard molecular biology text books.Dot Blot HybridisationDot blot hybridisation is now the most commonlyused procedure for testing of large numbers ofsamples. These are the main steps in dot blothybridisation:1. A small amount of sap is extracted from theplant under test.2. The viral nucleic acid is denatured by heatingor, if it is DNA, by alkali treatment.3. A spot of the extract is applied to a membrane.4. The membrane is baked or exposed toultraviolet light to bind the nucleic acid firmlyto it.5. Nonspecific binding sites on themembrane are blocked by incubation in aprehybridisation solution containing a protein,usually bovine serum albumin or non-fatdried milk, and small ss fragments of anunrelated DNA, together with salt and otheringredients.6. Hybridisation of a labeled probe nucleic acid tothe test nucleic acid bound to the substrate.7. Washing off excess (unhybridised) probe andestimation of the amount of probe bound by amethod appropriate to the kind of label usedfor the probe. The prehybridisation step(about two hours) and the hybridisation step(overnight) are carried out in heat-sealableplastic bags in a water bath at about 65 C.The technique is now widely used in plantvirology. For instance, a dot blot technique hasbeen successfully applied to screening large numbersof potato plants in a programme of breedingpotatoes for resistance to several viruses.A nonradioactive dot blot system using minimalequipment was developed for the routine diagnosisof a range of insect-transmitted viruses. Themethod has been used to assess relationshipsbetween tombusviruses, but some unexpectedcross-hybridisations were observed. A sensitivenonradioactive procedure has been developedfor detecting Bamboo mosaic virus and its associatedsatellite RNA in meristem-tip culturedplants.Tissue Print HybridisationThis is similar to immuno-tissue printing asdescribed in Box 13.3 but using labeled nucleicacid probes. Probes can be specifically designedfor detecting (þ) or (–) strands or specific partsof the genome. Tissue print hybridisation is ofespecial use for viroids that do not have proteinsthat can be detected immunologically. Viroidshave been detected by molecular hybridisationof imprinted membranes. An adaptation of theprocedure, termed squash-blot (or swat blot), hasbeen designed to assay Maize streak virus in singleleafhopper vectors squashed directly onto thenitrocellulose filter.In Situ HybridisationIn situ hybridisation can give information ofthe distribution of the target nucleic acid withina cell. The loci at which Banana streak virus (BSV)is integrated into the banana chromosomes wereidentified by in situ hybridisation. As an extensionof this procedure, fibre stretch hybridisation,in which denatured chromosomal DNAis stretched out on slides before hybridisation,gives information on the detailed structure ofthe integrated locus; this is described for BSV inBox 8.9.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

I. DIAGNOSIS259and synthesis of multiple copies of complementaryDNA of the sequence between theprimers using heat-stable DNA polymerase.The process goes through a series of amplificationcycles, each consisting of melting the dstemplate DNA molecules in the presence ofthe oligonucleotide primers and the four deoxyribonucleotidetriphosphates at high temperature(melting), hybridisation of the primerswith the complementary sequences in the templateDNAs at lower temperature (annealing),and extension of the primers with DNA polymerase(DNA synthesis). During each cycle,the sequence between the primers is doubledso that after n cycles, a 2 n amplification shouldbe obtained. Usually the reaction is of 30–50cycles.As PCR is based on DNA, it is not directlyapplicable to most plant viruses that haveRNA genomes. However, a cDNA can be madeto the desired region of the RNA genome usinga primer and reverse transcriptase and thisused as the initial template. This procedure,now widely used, is termed RT-PCR. A furtherrefinement is to couple PCR with the captureof the virus particles by immobilised antibodies,termed immune capture PCR (IC-PCR orIC-RT-PCR).PCR (and RT-PCR) has proved to be a verypowerful tool for virus detection and diagnosis.It can be used to directly produce a DNA productof predicted size that can be confirmed bygel electrophoresis. The choice of primers canbe used to distinguish between strains of avirus or, with primers containing a variety ofnucleotides at specific positions (degenerateprimers), be used for more generic determinations.Strains can also be distinguished byamplifying a region that has differences inrestriction endonuclease sites. PCR is widelyused in producing probes for hybridisation byincorporating reporter nucleotides in the reaction.To illustrate the widespread applicabilityto the detection and diagnosis of plant viruses,we will look at some examples; these are byno means a comprehensive collection of theuses to which this technique can be applied.As described in Box 8.9, Banana streak virus(BSV) sequences are frequently found integratedin the host genome. This precludes the use ofstraightforward PCR for diagnosis of episomalinfections of this virus, but IC-PCR can be usedin which the episomal virus particles are capturedand the host chromosomal DNA (containingthe integrant) is removed before PCR. Usingan IC-RT-PCR method in a single closed tube,Tomato spotted wilt virus can be detected in leafand root tissue, and Citrus tristeza virus and Plumpox virus can be detected in plant tissue andsingle aphid samples.In an RFLP analysis of 10 Potato virus Yisolates representative of four symptomaticgroups, the whole genomes were each amplifiedin two fragments by RT-PCR. Using sevenrestriction enzymes, three RFLP groups weredetermined in the 5 0 fragment and two in the3 0 fragment that correlated with the biologicalcharacters. One group of isolates appeared tohave resulted from a recombination event.Degenerate primers to highly conservedregions have been used to detect whiteflytransmittedgeminiviruses. Strains of the leafhopper-transmittedgemini virus Wheat dwarfvirus can be differentiated by using universaland strain-specific promoters. A highly sensitiveprocedure for the early detection of Beetnecrotic yellow vein virus (BNYVV) in plant, soil,and vector samples involves PCR and digoxigeninlabeling. PCR is being widely used inthe detection and identification of phytoplasmasin both plant and insect samples.6. DNA MicroarrayThe principle of microarrays or DNA chips isthe hybridisation of fluorescently labeled targetsequences to probe sequences spotted onto a solidsurface, usually a glass microscope slide. TotalRNA from the infected plant is converted toDNA by RT-PCR, and the cDNA is labeled byreaction with a fluorescent dye. Probes to manyIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

260 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGYdifferent viruses or variants can be spotted ontothe glass slide, and hybridisation can reveal jointinfections with more than one virus.F. Decision Making on DiagnosisThe application of biological, physical, andmolecular techniques has given a large “toolbag”for detection and diagnosis of plantviruses. This emphasizes made in the introductionthat it is important to identify the questionto be addressed. If one wishes to determine if aplant is virus infected—say, for quarantine purposes—onedoes not necessarily need a sophisticatedtechnique that identifies a virus strain. Onthe other hand, if one is studying the durabilityof a potential resistance gene (or transgene), it isvery useful to have an understanding of therange of variation of the virus. Thus, one hasto select the best technique for what is wanted.In making this decision various points have tobe taken into account, including the following.The sensitivity required. There is much discussionabout the relative sensitivity of detectionprocedures. However, the sensitivity of manyof the serological and nucleic acid–based testsis adequate for most purposes, so one shoulduse the system that is most convenient. For easeand speed of operation, and low cost, dot blotsbased on either an immunological test ornucleic acid hybridisation have a lot to offer.There are field kits available for testing for,say, potato viruses that are based on the sametechnology as home pregnancy kits. However,for each virus situation it is advisable to comparetests to see which is the most appropriate.The number of samples. Where large numbersof samples have to be handled, the followingfactors will be important in choosing a test procedure:specificity; sensitivity; ease and speed ofoperation; and cost of equipment, consumablesupplies, and labour.The material being sampled. In many cases,and especially in trees, the distribution of virusis not uniform. Thus, one has to be careful withthe taking of samples, and it is advisable to takeseveral samples from each tree.The reliability of the technique. A techniquecan be unreliable in two ways. False negativescan result from sampling part of the plant notcontaining virus, inhibition of the reaction bya plant constituent, or limitation in the materialsbeing used (e.g., mismatch in a primer forPCR). False positives can be due to plant constituents,especially when testing new plantspecies.The equipment and expertise available. The routinereliable detection of most viruses usuallydoes not require expensive equipment. Thereare some exceptions, such as the detection ofBSV described earlier in this chapter, whichrequires either ISEM or IC-PCR. However, it isimportant to have a good, reliable supply of consumablesand means for storing them withoutdeterioration. Similarly, most of the basic techniquesare relatively easy to learn, but it is importantthat they are learned properly so potentialsources of error can be recognised.There are international guidelines (FAO/IPGRI Technical Guidelines for the Safe Movementof Germplasm) that have been drawn up byexpert panels to assist with international germplasmmovement. These detail the currentideas on the safest and simplest tests for ensuringthat plant propagules do not contain theviruses that are known to infect that species. Itshould always be remembered that one cannot,of course, test for unknown viruses.II. EPIDEMIOLOGY AND ECOLOGYTo survive, a plant virus must have (1) oneor more host plant species in which it can multiply,(2) an effective means of spreading to andinfecting fresh individual host plants, and (3) asupply of suitable healthy host plants to whichit can spread. The actual situation that exists forany given virus in a particular locality, or onthe global scale, will be the result of complexIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

II. EPIDEMIOLOGY AND ECOLOGY261BOX 13.5ECOLOGY ANDEPIDEMIOLOGYFIGURE 13.2 Some of the interactions involved in thenatural infection of a plant by a virus. [This article was publishedin Semin. Vinol. 2, The movement of viruses withinplants R. Hull, pp. 89–95, Copyright Elsevier (1997).]interactions between many physical andbiological factors, the major ones of which areshown in Figure 13.2.An understanding of the ecology and epidemiologyof a virus in a particular crop andlocality is essential for the development ofappropriate methods for the control of the diseaseit causes. As with most other obligateparasites, the dominant ecological factors to beconsidered are usually the way viruses spreadfrom plant to plant and the ways that other factorsinfluence such spread. It is important todistinguish between the terms epidemiologyand ecology, which in many papers have beenused interchangeably (Box 13.5). In this chapter,we consider them both together.A. Epidemiology of Virusesin AgricultureTaking plant viruses as a whole, the flying,sap-sucking groups of insect vectors, particularlythe aphids, are by far the most importantagents of spread. There are two stages in theinfection of a crop: primary infection andEcology describes the factors influencing thebehaviour of a virus in a given physical situation.These factors include host range, tissuetropism, pathogenesis, and host responses. Itis a fundamental concept based on the relationalproperties of the virus and is not a propertyof the environment.Epidemiology is the study of the determinants,dynamics, and distribution of viral diseases inhost populations. It includes a dimensionalaspect of the factors determining the spread of avirus into a given situation. The two terms ofteninterlink, but each has a specific meaning.subsequent secondary spread. These areaffected by biological factors such as host range,susceptibility of the crop, and behaviour of thevector, by cultural factors such as time of plantingand size of field and by physical factors suchas weather and high level air currents. Althoughthese factors all interlink and often present acomplex situation, there are several basic pointsthat can be made.1. Primary InfectionsThe source of primary infections variesdepending on whether the crop is grown on aseasonal basis or whether there is overlappingcropping. If grown on a seasonal basis, thevirus must have an overwintering (or oversummering)host. This can be a wild species, a vegetativeorgan such as a potato or a seed for aseed-transmitted virus, or a vector in whichthe virus persists (e.g., a nematode).The initial spread into the crop by an insectvector depends on the virus-vector relationshipand on the feeding behaviour of the insect onthat crop. As noted in Chapter 12, crop plantsIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

262 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGYare often infected with nonpersistent viruses byaphids that do not colonise the plant. This isbecause the aphids feed for very short timeswhen seeking a host but for sufficiently longto infect the plant. When an aphid finds a suitableplant, it settles for a longer feed and thusis more likely to transmit a persistent virus,such as a luteovirus. Aphid behaviour is alsoaffected by the growth status of the plant, asthey seek the contrasting colours between theplant and the background soil. Thus, whencrop leaves have met within and between rows,aphids tend to land initially on plants at theedge of the crop, thus giving the “edge effect”of primary infections. Frequently, primaryinfection of insect-borne viruses usually showsas scattered patches over the field (Figure 13.3).Primary infections with nematode- and fungal-transmittedviruses reflect the distributionof the vector in the soil. Because of ploughingand other cultural practice, the infectionpatches are often elongate (Figure 13.4).If there are no overwintering hosts in thatregion, viruses can be brought in by insectvectors carried on high, level air currents fromdistant regions. Under appropriate climaticconditions, a continuous long-distance journeyFIGURE 13.3 Aerial view of a sugar beet field withpatches of plants infected by yellowing viruses and gradientsof infection spreading from the patches. (Courtesy ofIACR Broom’s Barn Research Centre.)may not be uncommon. Geostrophic airstreamsare air movements at altitudes of about 1,000 mor more moving along the isobars of a relativelylarge-scale weather system. Such airstreamshave almost certainly led to the masstransport of winged aphids from Australia toNew Zealand, a distance over sea of about2,000 km. Lettuce necrotic yellows virus and severalvector species of aphids have probablybeen introduced in this way. Not only theviruses but also potential aphid vectors are stillmoving into new continents. Thus, Metopalophiumdirhodum, a vector for Barley yellow dwarfvirus (BYDV), was first recorded in Australiain 1985 and was almost certainly a fairly recentarrival. Leafhopper vectors may also travellong distances. For example, large numbers ofMacrosteles fascifrons may be blown each springfrom an overwintering region about 300 kmnorth of the Gulf of Mexico through the midwesternUnited States and into the prairie provincesof Canada (Figure 13.5). BYDV and cerealaphids follow similar routes.Commonly, in tropical countries there isoverlap between the old maturing crop andthe newly planted one; the old crop is a sourceof primary infection of the new crop. This isthought to be the cause of the great increaseof outbreaks of rice tungro disease in SoutheastAsia associated with the “green revolution.”An analogous situation is where there is a “rollingfront” of crop planting over a large regionassociated with movement of seasonal rainfall.The spread of groundnut rosette disease insub-Saharan Africa is thought to be associatedwith the moving seasonal rainfall in the intertropicconvergence meteorological zone.The way a particular crop is grown andcultivated in a particular locality and the waysland is used through the year in the area mayhave a marked effect on the incidence of a virusdisease in the crop. Many diverse situationsarise, some examples of which will illustratethe kinds of factors involved.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

II. EPIDEMIOLOGY AND ECOLOGY263FIGURE 13.4 Relationship betweendensity of nematode infestation andoutbreak of virus disease. Left: Populationcontours of Xiphinema diversicaudatum.Right: Outbreak of Strawberrylatent ring spot virus in a 10-year-oldraspberry plantation. [From Taylorand Thomas (1968; Ann. Appl. Biol.62, 147–157, Wiley-Blackwell).]Planting date. A clear-cut example of the waytime of sowing seed affects virus incidenceoccurs with the winter wheat crop in southernAlberta. In a normal season, the percentageinfection with streak mosaic is markedlydependent on the time at which the seed issown. For sowing dates earlier than September,the spring-sown crop, carrying disease, overlapswith the autumn-sown crop. High air temperaturesmay cause a dramatic reduction insome aphid vector populations. If the plantingdate for the autumn crop is delayed until suchconditions prevail, much less virus spreadmay take place in a crop.Crop rotation. The kind of crop rotationpracticed may have a marked effect on the incidenceof viruses that can survive the winter inweeds or volunteer plants. With certain crops,volunteer plants that can carry viruses may survivein high numbers for considerable periods.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

264 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGYFIGURE 13.5 Hypothesis to account for long-distance aphid movement and a severe outbreak of Maize dwarf mosaicvirus (MDMV) in corn in Minnesota in 1977. The midcontinental and southern distributions of Johnsongrass, the major wildhost of MDMV, are shown relative to a major drought in May and June 1977. The large arrow indicates the path of low-leveljet winds (up to 80 km/hr) on 2 July 1977. The smaller dashed arrow indicates the path of a cold front that caused the lowleveljet wind to become diffuse and triggered thunderstorm activity moving through Minnesota and into Wisconsin. [FromZeyen et al. (1987; Ann. Appl. Biol. 111, 325–336, Wiley-Blackwell).]Soil cultivation. Soil cultivation practices mayaffect the spread and survival of viruses in thesoil or in plant remains. As just noted, nematodeand fungal vectors may be spread bymovement of soil during cultivation. The physicalmovements of soil tillage and harvestingoperations are the major means of spread ofBNYVV.Nurseries and production fields as sources ofinfection. Nurseries, especially where they havebeen used for some years, may act for themselvesas important sources of virus infection.2. Secondary SpreadMany of the factors associated with primaryinfection control the secondary spread ofviruses. Thus, nonpersistent viruses are mainlyspread by the aphids that are not seen on thecrop—those passing through on host-seekingflights. Much of the spread of persistent virusesIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

II. EPIDEMIOLOGY AND ECOLOGY265is by aphids colonising the crop, even walkingfrom plant to plant.The cultivation of a single crop, or at least avery dominant crop, over a wide area continuouslyfor many years may lead to major epidemicsof virus disease, especially if an airbornevector is involved. Soil-borne vectors canalso be important from this point of view—forexample, with Grape vine fan leaf virus in vineyards,where the vines are cropped for manyyears. Monocropping may also lead to a buildupof crop debris and the proliferation of weedsthat become associated with the particular crop.Weather conditions such as rainfall, temperature,and wind can play a major role in thesecondary development of a virus disease inthe field. This is mostly by affecting the populationsize and behaviour of the vector but alsocan impact upon the crop development andits susceptibility to virus infection. This canlead to wide annual variation in the incidenceof a virus disease in an annual crop, as shownin Figure 13.6. Overall, the factors involved inboth primary infection and secondary spreadare extremely complex and can be illustratedby the epidemiology of luteoviruses (Box 13.6).FIGURE 13.6 Variation of virus incidence with year and location. Mean monthly incidence of sugar beet yellows diseasein different regions of England in years with greatly differing amounts of spread. [Data supplied by G.D. Heathcote(Broom’s Barn Experimental Station) for representative crops sown annually in March or April in eastern England, in wetterareas in western England, and in cooler areas in northern England.] The dashed lines indicate the mean incidence ofyellows at the end of September as calculated for the entire English crop. [This article was published in Ann. Appl. Biol.111, R.J. Zeyen, E.L. Stromberg, and E.L. Kuehnast, Long-range aphid transport hypothesis for maize dwarf mosaic virus:History and distribution in Minnesota, pp. 325–336. Copyright Elsevier (1987).]IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

266 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGYBOX 13.6COMPLEXITIES OF EPIDEMIOLOGYThe factors and interactions involved in the ecology and epidemiology of viruses in crops areextremely complex. This can be illustrated by the processes involved in the spread of luteoviruseswithin and between crops (Fig.). The luteoviruses are among the most studied of virus groups as faras epidemiology is concerned but the complexities shown in the Figure are just as applicable to othercrop situations. There are virtually no data on the epidemiology of viruses in noncrop natural systems.Fig. Epidemiological process of luteovirus spread showing the various routes of infection and the relevant differencesbetween potato and other crops. [From Robert (1999; in The Luteoviridae, H.G. Smith and H. Barker, Eds.,pp. 221–231. CAB International, Wallingford, UK).]IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

B. Plant Viruses in the NaturalEnvironmentThere is not much information on the impactthat virus infections might have on plants inthe natural ecosystem. As noted in Chapter 4,a virus that kills its host plant with a rapidlydeveloping systemic disease is much less likelyto survive than one that causes only a mild ormoderate disease that allows the host plant tosurvive and reproduce effectively. There isprobably a natural selection in the field againststrains that cause rapid death of the host plant,and thus viruses in the wild plants are unlikelyto show many overt symptoms (see Table 4.1).Even so, there is some evidence that virusinfection can reduce the competitive and reproductiveability of a plant species, thus impactingupon the community structure. Forinstance, in California, BYDV infection reducednative grass growth, survival, and fecundity,thus affecting its fitness in competition withintroduced grasses.II. EPIDEMIOLOGY AND ECOLOGY267Another example of international trade leadingto the spread of an important disease isexemplified by the spread of BNYVV, whichcauses the important rhizomania disease insugar beet (Figure 13.7).Important new diseases are also arisingfrom interactions between the genomes of twoviruses by either recombination or by synergisminduced by suppressors of gene silencing(See Box 11.3).C. Emergence of New VirusesAs well as distributing viruses around theworld, humans have moved crop species tonew countries, often with disastrous consequencesas far as virus infection is concerned.Plant species that were relatively virus-free intheir native land may become infected withviruses that have long been present in thecountries to which they were moved for commercialpurposes. Thus, cacao was transferredfrom the Amazonian jungle to West Africa latein the late 1800s, and since then major commercialproduction has developed there. Theswollen shoot disease caused by Cacao swollenshoot virus, first reported in cacao in 1936,was very probably transmitted from naturalWest African tree hosts of the virus by themealybug vectors that are indigenous to theregion.FIGURE 13.7 Spread of beet rhizomania caused by Beetnecrotic yellow vein virus through Europe and worldwide.[Data courtesy of Dr. M. Stevens (Broom’s Barn ResearchCentre).]IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

268 13. PLANT VIRUSES IN THE FIELD: DIAGNOSIS, EPIDEMIOLOGY, AND ECOLOGYIII. VIRUSES OF OTHERKINGDOMSDiagnostics used for detecting viruses ofvertebrates and invertebrates are very similarto those described previously for plant viruses.The epidemiology of vertebrate and invertebrateviruses differs significantly from that ofplant viruses. The basic difference is that vertebratesand invertebrates can move around andplants (usually) cannot. Therefore, much ofthe natural spread of plant viruses is by invertebratevectors; these are involved in some animaldisease conditions, such as bluetonguevirus disease of cattle and sheep that is spreadby midges. In the natural situation, it is themovement of infected vertebrates into the closeproximity to healthy ones that is an importantfactor in the development of epidemics.Humans play a major role in the spread of cropand farm animal viruses and in the worldwidedistribution of viruses.IV. SUMMARY• Four major properties of viruses are usedin diagnostics, biological properties,physical properties, protein properties,and nucleic acid properties. Most moderndiagnostics are based on the latter twoproperties.• It is important to identify the reason for thediagnosis before selecting the technique(s) tobe used.• Diagnostic techniques using the virusprotein properties focus mainly onserological techniques targeted at the viruscoat protein.• Diagnostics using the viral nucleic acidproperties focus mainly on hybridisationtechniques and the polymerase chainreaction.• Invertebrate vectors are among the mostimportant factors in the epidemiology ofplant viruses. They introduce primaryinfections (as does seed transmission) anddisseminate the virus in secondary spread.• Other factors involved in epidemiologyinclude agronomic practices and weatherconditions.Further ReadingAgindotan, B. and Perry, K.L. (2007). Microarray detectionof plant RNA viruses using randomly primed andamplified complementary DNAs from infected plants.Phytopathology 97, 119–127.Cooper, I. and Jones, R.A.C. (2006). Wild plants and viruses:Under-investigated ecosystems. Adv. Virus Res. 67, 1–47.Fargette, D., Konaté, G., Fauquet, C., Muller, E., Peterschmitt,M., and Thresh, J.M. (2006). Molecular ecologyand emergence of tropical plant viruses. Annu. Rev.Phytopathol. 44, 235–260.Foster, G.D., Johansen, I.E., Hong, Y., and Nagy, P.D.(2008). Plant Virology Protocols: From viral sequence to proteinfunction, 2nd ed. Humana Press, Totowa, N.J.Hull, R. (2002). Matthews’ plant virology. Academic Press,San Diego.Jeger, M.J., Seal, S.E., and Van den Bosch, F. (2006). Evolutionaryepidemiology of plant virus disease. Adv. VirusRes. 67, 163–203.Koenig, R., Lesemann, D.-E., Adam, G., and Winter, S.(2008). [Diagnostic techniques] for plant viruses. Encyclopediaof Virology, Vol. 2, 18–28.Reynolds, D.R., Chapman, J.W., and Harrington, R. (2006).The migration of insect vectors of plant and animalviruses. Adv. Virus Res. 67, 453–517.Vainiopää, R. and Leinikki, P. (2008). [Diagnostic techniques]:Molecular and serological approaches. Encyclopedia ofVirology, Vol.2,29–36.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

C H A P T E R14Conventional ControlA range of control measures are used to mitigate the considerable losses that plant virusescause to crops.O U T L I N EI. Introduction 269II. Avoiding Infection 271III. Stopping the Vector 274IV. Protecting the Plant 277V. Conventional Resistance to PlantViruses 280VI. Strategies for Control 283VII. Viruses of Other Kingdoms 283VIII. Summary 283I. INTRODUCTIONThe use of fungicidal chemicals to protectcrop plants from infection or minimise invasionis an important method for the control of manyfungal diseases. No such direct method for thecontrol of virus diseases is yet available. Mostof the procedures that can be used effectivelyinvolve measures designed to reduce sourcesof infection inside and outside the crop, to limitspread by vectors, and to minimise the effect ofinfection on yield. Generally speaking, suchmeasures offer no permanent solution to avirus disease problem in a particular area. Controlof virus disease is usually a running battlein which organisation of control procedures,care by individual growers, and cooperationamong them is necessary year after year. Thefew exceptions are where a source of resistanceto a particular virus has been found in, or successfullyincorporated into, an agriculturallyuseful cultivar. This is becoming of increasingimportance with the development of transgenicprotection of plants against viruses that areComparative Plant Virology, Second Edition269Copyright # 2009, Elsevier Inc. All rights reserved.

270 14. CONVENTIONAL CONTROLdiscussed in Chapter 15. Even with conventionaland transgenic resistance, protectionmay not be permanent when new strains ofthe virus arise that can cause disease in a previouslyresistant cultivar.Correct identification of the virus or virusesinfecting a particular crop is essential for effectivecontrol measures to be applied. Of major importancein designing a strategy for controllinga virus in a specific crop is an understandingof the epidemiology of that virus that wasdiscussed in Chapter 13. This enables diseaseoutbreaks to be forecasted (Box 14.1).The three major approaches to conventionalcontrol of plant viruses are the removal oravoidance of sources of infection, protectingplants from systemic infection, and deploymentof resistance.BOX 14.1DISEASE FORECASTINGAn understanding of the epidemiology and ecologyof some major crop virus diseases has led toprocedures for forecasting potential epidemics.This is very useful in implementing control measures.There are two main approaches to forecasting:monitoring the progress of a diseaseand developing mathematical models.Monitoring Virus Disease ProgressMany large-scale farmers routinely monitor theircrops and apply control measures at an appropriatetime. However, as virus diseases take severaldays or even weeks to show symptoms afterinfection, the application of control measuresbased on symptom appearance can be too late.It also depends on the correct diagnosis of thedisease and knowledge of how it is spread.Mathematical ModellingThere are an increasing number of mathematicalmodels directed at forecasting the outcome ofthe spread of a disease into an agronomic situation.Basically, the two types of models areprediction models, to predict a possible epidemic,and simulation models, to understandthe factors that give rise to and control a givensituation. A model is developed to answerspecific questions, and there is no general modelto predict the potential and outcome of allpotential viral epidemics. In developing amodel, as many factors as can be predicted aretaken into account. These include, knowledgeon the virus, its vector, virus-vector interaction,type of crop, the cropping system, and variousenvironmental factors that can impact on thesebiological factors. A good model enables oneto make strategic management decisions onwhether the problem is going to be significantand, if so, when and how to deal with it.The efficiency of prediction from even a goodmodel is only limited by the amount and reliabilityof the data fed into it. The data must beobtained from various sources and collated. Anexample is a model for predicting virus yellowsdisease of sugar beet in the United Kingdom thatis based on the preceding winter weather (especiallythe number of frost days) and the dateswhen the aphid vectors begin their springmigration and region in which the beet is beinggrown (eastern, western, and northern regions).This model has been refined to allow for thenumbers of migrating Myzus persicae, the majorvector.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

II. AVOIDING INFECTIONA. Removal of Sources of InfectionIt is obvious that there will not be a virusproblem if the crop is free of virus whenplanted and when there is no source of infectionnear enough to allow it to spread into thecrop. Sources of infection are discussed inChapter 13.To eliminate these sources, it may be worthwhileto remove infected plants (rogue) from acrop. If the spread is occurring rapidly fromsources outside the crop, roguing the crop willhave no beneficial effect. In certain situations,roguing may increase disease incidence by disturbingvectors on infected plants. In manycrops, newly infected plants may be acting assources of virus for further vector infectionbefore they show visible signs of disease.Most of the successful eradication schemeshave been on tree crops. The following areamong the factors that dictate success:• Relatively small numbers of infected treesand infection foci• Low rate of natural spread• Good data on extent and distribution ofinfection• Rapid, reliable, and inexpensive diagnosticprocedure for the virus and resources forrapid and extensive surveys and treeremoval.Two examples of roguing and eradicationschemes are given in Box 14.2.B. Virus-Free SeedWhere a virus is transmitted through theseed, such transmission may be an importantsource of infection, since it introduces the virusinto the crop at a very early stage, allowinginfection to be spread to other plants whilethey are still young. In addition, seed transmissionintroduces scattered foci of infectionII. AVOIDING INFECTIONthroughout the crop. Where seed infection isthe main or only source of virus, and wherethe crop can be grown in reasonable isolationfrom outside sources of infection, virus-freeseed may provide a very effective means forcontrol of a disease.Lettuce mosaic virus is a good example ofcontrolling a virus problem through clean seed.Crops grown from virus-free seed in Californiahad a much lower percentage of mosaic at harvestthan adjacent plots grown from standardcommercial seed. To obtain effective controlby the use of virus-free or low-virus seed, a certificationscheme is necessary, with seed plantsbeing grown in appropriate isolation.C. Virus-Free Vegetative Stocks271For many vegetatively propagated plants,the main source of virus is chronic infection inthe plant itself. With such crops, one of themost successful forms of control has involvedthe development of virus-free clones—that is,clones free of the particular virus under consideration.Two problems are involved. First, avirus-free line of the desired variety with goodhorticultural characteristics must be found.When the variety is 100 percent infected,attempts must be made to free a plant or partof a plant from the virus. Second, havingobtained a virus-free clone, a foundation stockor “mother” line must be maintained virus free,while other material is grown up on a sufficientlylarge scale under conditions where reinfectionwith the virus is minimal or does nottake place. These stocks are checked that theyare “virus free” (e.g., below a set level ofdetected virus) and are then used for commercialplanting.As a plant is usually infected with a virus forlife, various techniques have to be used to freethem of virus, including heat therapy in whichthe plant is kept at a temperature usually in therange of 35 to 40 o C for periods of weeks, meristemtip culture, taking advantage of the factIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

272 14. CONVENTIONAL CONTROLBOX 14.2ROGUING AND ERADICATION CONTROLOF PLANT VIRUSESBanana Bunchy Top Virus (BBTV)One of the most successful examples of diseasecontrol by roguing of infected crop plants hasbeen the reduction in incidence of BBTV inbananas in eastern Australia. Legislation toenforce destruction of diseased plants and abandonedplantations was enacted in the late 1920s.Within about 10 years, the campaign was effectiveto the point where bunchy top disease wasno longer a limiting factor in production. Thesuccess of the scheme was attributed to theabsence of virus reservoirs other than bananas,together with a small number of wild bananas;knowledge that the primary source of the viruswas planting material and that spread wasby aphids; cultivation of the crop in small,discrete plantations rather than as a scatteredsubsistence crop; strict enforcement of stronggovernment legislation; and the cooperation ofmost farmers.Cocoa Swollen Shoot Virus (CSSV)Cocoa swollen shoot disease (CSSD) is caused by(CSSV). The disease was discovered in Ghana in1936 and is one of the most devastating scourgesof cocoa, in the 1940s threatening to wipe out thecocoa industry in what is now Ghana. A massivenationwide eradication campaign began in 1946after it had been shown that the swollen shootand dieback disease was caused by a virus thatis spread from tree to tree by several species ofmealybugs (Pseudococcidae). The eradicationcampaign has continued to the present time,but there have been serious interruptions anddiscontinuities. Thus, financial resources andpersonnel who could have otherwise been usedto improve the standard of husbandry, raisecocoa production, and make improvements ofother crops in the agricultural sector have beendiverted into eradicating the disease by cuttingout diseased and at times neighbouring “contact”trees. By the 1980s, more than 190 milliontrees had been removed, but despite massiveexpenditure, swollen shoot was more prevalentin Ghana than ever before. Among the main problemswere the financial and logistic problemsin mounting and sustaining such a large and complexeradication programme, lack of cooperationfrom farmers who were reluctant to lose severalyears’ production, and lack of detailed epidemiologicalinformation. Since then the mainapproach to control is by trying to find and deployresistance to the virus.that most viruses do not invade the plant meristem(see Chapter 9), and chemotherapy, bytreating select plants with antiviral compoundssuch as an analogue of guanosine (ribavirin,also called virazole) in combination within vitro tissue culture. Such techniques are onlyused on elite material of high-cost crops suchas soft fruit and flowers. It is very importantto include long-term virus testing into theprogramme for producing virus-free motherplants and also to maintain the nuclear stockin a virus-free environment.D. Modified Agronomic PracticesVirus infection can be reduced by modifyingagronomic practices such as breaking the infectionwhere one major susceptible annual cropIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

II. AVOIDING INFECTION273or group of related crops is grown in an areaand where these are the main hosts for a virusin that area by ensuring that there is a periodwhen none of the crop is grown. A good exampleof this is the control of planting date of thewinter wheat crops in Alberta to avoid overlapwith the previous spring- or winter-sown crop(Figure 14.1). This procedure, together withelimination of volunteer wheat and barleyplants and grass hosts of Wheat streak mosaicvirus before the new winter crop emerges, cangive good control in most seasons.Other approaches include changing plantingdates to avoid young plants being exposed tomajor migrations of the insect vector and usingclose-spaced planting to reduce the attractivenessto flying aphids.E. Quarantine RegulationsMost agriculturally advanced countries haveregulations controlling the entry of plant materialto prevent the introduction of diseases andpests not already present. Many countries nowhave regulations aimed at excluding specificviruses and their vectors, sometimes from specificcountries or areas. The setting up of quarantineregulations and providing effectivemeans for administering them is a complexproblem. Economic and political factors frequentlyhave to be considered. Quarantinemeasures may be well worthwhile with certainviruses, such as those transmitted throughseed, or in dormant vegetative parts such asfruit trees and bud wood.FIGURE 14.1 Wheat streak mosaicdisease cycle. Preventing the infectionof winter wheat in the autumn isthe key to controlling this disease insouthern Alberta. Dark area periodduring which effective control can normallybe achieved; broken hatchedbands, problems presented by volunteerseedlings, early-seeded winterwheat, and/or late-maturing springwheat or barley; arrows, transfer ofvirus by windblown mites. (Diagramcourtesy of T.G. Atkinson.)IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

274 14. CONVENTIONAL CONTROLThe value of quarantine regulations willdepend to a significant degree on the previoushistory of plant movements in a region. Forexample, active exchanges of ornamental plantsbetween the countries of Europe have beengoing on for a long period, leading to analready fairly uniform geographical distributionof viruses infecting this type of plant.On the other hand, the European Plant ProtectionOrganisation found it worthwhile to setup quarantine regulations against fruit treeviruses not already recorded in Europe.In spite of many countries having regulationsdesigned to prevent the entry of damagingviruses, they can spread internationallyvery rapidly. A good example is the rhizomaniadisease of sugar beet, shown in Figure 13.7.III. STOPPING THE VECTORAs described in Chapter 12, plant viruses areusually transmitted by arthropod vectors, butsome are transmitted by fungal vectors, andothers, particularly Tobacco mosaic virus (TMV)and Tomato mosaic virus (ToMV), may be transmittedmechanically (by “human vectors”).Once TMV or ToMV enters a crop like tobaccoor tomato, it is very difficult to prevent itsspread during cultivation and particularly duringsuch processes as tying-up of plants. Controlmeasures consist of treatment of implementsand washing of the hands. Workers’ clothingmay become heavily contaminated with TMVand thus spread the virus by contact.A. Air-Borne VectorsBefore control of virus spread by air-bornevectors can be attempted, it is necessary to identifythe vector. This information has sometimesbeen difficult to obtain. Not uncommonly, it isan occasional visitor rather than a regular coloniserthat is the main or even the only vector ofa virus. Furthermore, some aphid species aremore efficient vectors than others. For instance,the brown citrus aphid (Toxoptera citricida) isamuch more efficient vector of Citrus tristeza virus(CTV) than is the melon aphid (Aphis gossypii;see Figure 14.2).FIGURE 14.2 Comparative increaseof Citrus tristeza virus infectionin field situations when vectored bythe brown citrus aphid (Toxoptera citricida),an efficient vector, and themelon aphid (Aphis gossypii), a lessefficientvector. Data for the brown citrusaphid were taken from test plots inCosta Rica and the Dominican Republic.Data for the melon aphid weretaken from surveys and experimentalplots in Spain, Florida, and California.Initial infection levels were less than1%. Note “stairstep” progression ininfection with the melon aphid, whichis believed to correspond to periodicheavy aphid migrations. [FromGarnsey et al. (1998; in Plant virus diseasecontrol, A. Hadidi, R.K. Khetarpal,and H. Koganezawa, Eds., pp. 639–658, APS Press, St. Paul, MN).]IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

III. STOPPING THE VECTOR2751. InsecticidesThe application of insecticides is currentlyone of the main ways of controlling insect pestsof plants. To prevent an insect from causingdirect damage to a crop, it is necessary only toreduce the population below a damaging level.Control of insect vectors to prevent infection byviruses is a much more difficult problem, asrelatively few winged individuals may causesubstantial spread of virus. Contact insecticideswould be expected to be of little use unless theywere applied very frequently. Persistent insecticides,especially those that move systemicallythrough the plant, offer more hope for viruscontrol. Viruses are often brought into cropsby winged aphids, and these may infect a plantduring their first feeding, before any insecticidecan kill them. When the virus is nonpersistent,the incoming aphid, when feeding rapidly,loses infectivity anyway, so killing it withinsecticide will not make much difference toinfection of the crop from the outside. On theother hand, an aphid bringing in a persistentvirus is normally able to infect many plants,so killing it on the first plant will reducespread.As far as subsequent spread within the cropis concerned, similar factors should operate.Spread of a virus that is nonpersistent shouldnot be reduced as much by insecticide treatmentas a persistent virus where the insectrequires a fairly long feed on an infected plant.Thus, spread of the persistent Potato leaf rollvirus (PLRV) in potato crops was substantiallyreduced by appropriate application of insecticides,but spread of the nonpersistent Potatovirus Y (PVY) was not.As discussed in Box 14.1, disease forecastingdata can be an important factor in the economicuse of insecticides. Sometimes a long-termprogramme of insecticide use aimed primarilyat one group of viruses will help in the controlof another virus. Thus, the well-timed use ofinsecticides in beet crops in England, aimedmainly at reducing or delaying the incidenceof yellows diseases (Beet yellows virus and Beetmild yellows virus), has also been a major factorin the decline in the importance of Beet mosaicvirus in this crop. A warning scheme to sprayagainst the vectors of beet yellows viruses wasinitiated in the United Kingdom in 1959 and isbased on monitoring populations of aphids incrops from May until early July.As well as the problems just described, theremay be other adverse biological and economicconsequences related to the use of insecticides,including development of resistance by thetarget insect to the insecticide, resurgence ofthe pest once the insecticide activity has wornoff, and possible effects on humans and otheranimals in the food chain.2. Insect DeterrentsThe application of various chemicals ormaterials can deter aphids from landing on orfeeding on crop plants. Spraying mineral oilson plants affects the feeding behaviour ofaphids and leafhoppers and can give some protectionagainst nonpersistent viruses. Derivativesprepared from the pheromone (E)-ß-farneseneand related compounds have been shown tointerfere with the transmission of PVY by Myzuspersicae in glasshouse experiments. Laying aluminiumstrips on the ground between crop rowsrepels aphids coming into the crop throughreflecting UV light.3. Agronomic TechniquesA tall cover crop will sometimes protect anundersown crop from insect-borne viruses.For example, cucurbits are sometimes grownintermixed with maize. It is thought thatincoming aphids land on the barrier crops, feedbriefly, and either stay there or fly away.A major approach is the use of crops bredfor resistance to insect pests. Sources of resistancehave been found against most of the airbornevector groups. The basis for resistanceIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

276 14. CONVENTIONAL CONTROLto the vectors is not always clearly understood,but some factors have been defined. In generalterms, there are two kinds of resistance relevantto the control of vectors. First, nonpreferenceinvolves an adverse effect on vector behaviour,resulting in decreased colonisation, and second,antibiosis involves an adverse effect onvector growth, reproduction, and survival aftercolonisation has occurred. These two factorsmay not always be readily distinguished. Somespecific mechanisms for resistance are stickymaterial exuded by glandular trichomes, suchas those in tomato; heavy leaf pubescence insoybean; A-type hairs on Solanum betrhaultiithat when ruptured, entrap aphids with theircontents and B-type hairs on the same host,which entangle aphids, making them strugglemore and so rupture more A-type hairs; inabilityof the vector to find the phloem in Agropyronspecies; and interference with the abilityof the vector to locate the host plant. For example,in cucurbits with silvery leaves, there was adelay of several weeks in the development of100 percent infection in the field with Cucumbermosaic virus (CMV) and Clover yellow vein virus.This effect may be due to aphids visiting plantswith silvery leaves less frequently because oftheir different light-reflecting properties.There may be various limitations on theuse of vector-resistant cultivars: Sometimessuch resistance provides no protection againstviruses. For example, resistance to aphid infestationin cowpea did not provide any protectionagainst Cowpea aphid-borne mosaic virus.If a particular virus has several vector species,or if the crop is subject to infection with severalviruses, breeding effective resistance against allthe possibilities may not be practicable, unlessa nonspecific mechanism is used (e.g., tomentoseleaves). Perhaps the most serious problemis the potential for new vector biotypes toemerge following widespread cultivation of aresistant cultivar, as may happen followingthe use of insecticides (Box 14.3).BOX 14.3INSECT BIOTYPES OVERCOMING PLANT RESISTANCEThis breakdown of plant resistance to insects iswell illustrated by the history of the rice brownplanthopper (Nilaparvata lugens) (BPH). Withthe advent of high-yielding rice varieties inSoutheast Asia in the 1960s and 1970s, the riceBPH and Rice grassy stunt virus, which it transmits,became serious problems. Cultivars containinga dominant gene (Bph1) for resistance tothe hopper were released about 1974. Withinabout three years, resistance-breaking populationsof the hopper emerged. A new, recessiveresistance gene (bph2) was exploited in cultivarsreleased between 1975 and 1983. They weregrown successfully for a few years until a newhopper biotype emerged that overcame theresistance. A study of the adaptation of threecolonies of N. lugens to rice cultivars containingdifferent resistance genes showed that the bhp1and bhp3 resistance genes were overcome morereadily by colonies that had been exposed forabout 10 years to those genes. However, rice cultivarIR64 which contained bph1 and some minorresistance genes showed a greater durability ofresistance than other cultivars. DNA markers toBPH resistance genes are being used in breedingprogrammes.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

B. Soil-Borne VectorsMost work on the control of viruses transmittedby nematodes and fungi has centredon the use of soil sterilization with chemicals.However, several factors make general andlong-term success unlikely:• Huge volumes of soil may have to betreated.• A mortality of 99.99 percent still leaves manyviable vectors.• The use of some of the chemicalsinvolved has been banned in certaincountries, and such bans are likely to beextended.In any event, chemical control can be justifiedeconomically only for high-return crops orcrops that can remain in the ground for manyyears. However, some recent advances in nematodecontrol procedures may be applicable tothe control of viruses that they transmit andmay be adaptable to the control of fungustransmittedviruses.1. NematodesThere are four basic strategies for nematodecontrol:1. Exclusion or avoidance usually byquarantine.2. Reduction of the initial population densityby cultural approaches such as use of cleanplanting stock or crop rotation with a breakcrop of a species that is not a host for thetarget nematode, by chemical nematicides,by biological tactics such as introducingbiological agents antagonistic to nematodesand organic amendments, or by the use ofnematode-resistant crop varieties that willreduce nematode populations.3. Suppression of nematode reproduction bychemicals, organic amendments, andcertain natural and transgenic resistancetraits.IV. PROTECTING THE PLANT4. Restriction of the current or future cropdamage by nematode resistance. However,tolerant cultivars will reduce crop damagedue to nematode feeding but will not reducethe chances of virus infection.2. FungiFungus-transmitted viruses are importantin two agronomic situations: nutrient or aquaticsystems, and fields. The major control measuresare the use of three types of chemicals:surfactants, heavy metals, and sometimes fungicidesfor use in nutrient or aquatic systems;soil amendments and fungicides to control thefungal vector in the soil; and soil partial sterilantsor disinfectants to reduce the active andresting spore stages of fungal vectors in thefield. In general, attempts to control infectionwith viruses having fungal vectors by applicationof chemicals to the soil have usually notbeen successful.IV. PROTECTING THE PLANTEven if sources of infection are available andthe vectors are active, a third kind of controlmeasure is available: protecting inoculatedplants from developing systemic disease. Thereare essentially three approaches that have beenused to protect plants, using a mild strain of thevirus (termed cross-protection or mild strainprotection), the use of chemicals, and geneticprotection (conventional resistance and transgenicresistance).A. Protection by a Plant Pathogen277Inoculation of plants with either a mild virusstrain or with satellite RNA (termed crossprotection)has been used to protect againstsevere virus strains. The phenomenon of crossprotectionis described in Chapter 10. Infectionof a plant with a strain of virus causing onlyIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

278 14. CONVENTIONAL CONTROLmild disease symptoms (the protecting strain;also known as the mild, attenuated, hypovirulent,or avirulent strain) may protect it frominfection with severe strains (the challengingstrain). Thus, plants might be purposelyinfected with a mild strain as a protective measureagainst severe disease.Although such a procedure could be worthwhileas an expedient in very difficult situations,it is not to be recommended as ageneral practice for the following reasons:• So-called mild strains often reduce yield byabout 5 to 10 percent.• The infected crop may act as a reservoir ofvirus from which other more sensitivespecies or varieties can become infected.• The dominant mild strain of virusmay change to a more severe type in someplants.• Serious disease may result from mixedinfection when an unrelated virus isintroduced into the crop.• For annual crops, introduction of a mildstrain is a labour-intensive procedure.• The genome of the mild strain mayrecombine with that of another virus,leading to the production of a new virus.In spite of these difficulties, the procedurehas been used successfully, at least for a time,with some crops. A suitable mild isolate shouldhave the following properties:• It should induce milder symptoms in allthe cultivated hosts than isolatescommonly encountered and should notalter the marketable properties of the cropproducts.• It should give fully systemic infections andinvade most, if not all, tissues.• It should be genetically stable and not giverise to severe forms.• It should not be easily disseminated byvectors to limit unintentional spread to othercrops.• It should provide protection against thewidest possible range of strains of thechallenging virus.• The protective inoculum should be easy andinexpensive to produce, check for purity,provide to farmers, and apply to the targetcrops.Mild protecting strains are produced fromnaturally occurring variants, from randommutagenesis, or from directed mutagenesis ofsevere strains. The control of CTV providesthe most successful example for the use ofcross-protection (Box 14.4).Satellite viruses and RNAs are described inChapter 3, and, as far as potential biocontrolagents, fall into three categories: those thatenhance the helper virus symptoms, those thathave no effect, and those that reduce the helpervirus symptoms. The last one has potential as acontrol agent. Most of the work has focussed onthe satellites of CMV with some successes in fieldapplication, especially against the necrogenicsatCMV (CARNA5) described in Chapter 3.However, there has been concern over thedurability of using satellites as biocontrol agents.There is a wide range of necrogenic and othervirulent strains of satCMV. Passage of a benignsatellite of CMV through Nicotiana tabacum ledto the satellite rapidly mutating to a pathogenicform and mutations of a single or a few basescan change a nonnecrogenic variant to a necrogenicone (Figure 14.3). Necrogenic variants ofthe CMV satellite have a greater virulence thannonnecrogenic variants, but, as they depress theaccumulation of the helper virus more than dononnecrogenic variants, the necrogenic variantsare not so efficiently aphid transmitted.B. Antiviral ChemicalsConsiderable effort has gone into a searchfor inhibitors of virus infection and multiplicationthat could be used to give direct protectionto a crop against virus infection in the way thatIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

IV. PROTECTING THE PLANT279BOX 14.4CONTROL OF CITRUS TRISTEZA VIRUS BY CROSS-PROTECTIONWorldwide, Citrus tristeza virus (CTV) is themost important virus in citrus orchards. In the1920s, after its introduction to South Americafrom South Africa, the virus virtually destroyedthe citrus industry in many parts of Argentina,Brazil, and Uruguay. The application of crossprotectionby inoculation with mild CTV isolatesin Brazil proved to be successful particularlywith Pera oranges, with more than 8 milliontrees being planted in Brazil by 1980. Protectioncontinues in most individual plants through successiveclonal generations. However, in an eightyearassessment of the ability of four mildisolates to suppress severe CTV isolates inValencia sweet orange on sour orange rootstockin Florida, about 75 percent of the mild-strainprotected trees had severe symptoms comparedwith about 85 percent of the unprotected trees.The use of the same isolates gave better protectionof Ruby Red grapefruit on sour orange rootstock.Thus, there are differences in theresponses of the scion:rootstock combination,but it is also important to have a compatiblemild strain. The search for improved attenuatedstrains of the virus continues, and the techniqueis being adopted in other countries.FIGURE 14.3 Alignment of the 55 3 0 -terminal residues of the CMV satellite RNA variants mutated from a necrogenicform towards a nonnecrogenic one or vice versa. The arrows indicate the positions found to be determinant for necrogenicity.[From Jacquemond and Tepfer (1998; in Plant virus disease control, A. Hadidi, R.K. Khetarpal, and H. Koganezawa,Eds., pp. 94–120, APS Press, St. Paul, MN).]IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

280 14. CONVENTIONAL CONTROLfungicides protect against fungi. There hasbeen no successful control on a commercialscale by the application of antiviral chemicalsdue to these major difficulties:• An effective compound must inhibit virusinfection and multiplication withoutdamaging the plant. This is a major problem,as virus replication is so intimately boundup with cell processes and any compoundblocking virus replication is likely to havedamaging effects on the host.• An effective antiviral compound would needto move systemically through the plant if itis to prevent virus infection by invertebratevectors.• A compound acting systemically wouldneed to retain its activity for a reasonableperiod. Frequent protective treatmentswould be impracticable. Many compoundsthat have some antiviral activity areinactivated in the plant after a time.• For most crops and viruses, the compoundwould need to be able to be produced on alarge scale at an economic price. This mightnot apply to certain relatively small-scale,high-value crops, such as greenhouse orchids.• For use with many crops, the compoundwould have to pass food and drugregulations. Many of the compounds thathave been used experimentally would not beapproved under such regulations.Because of these difficulties, there are only afew cases of the use of chemicals to producevirus-free stock plants.V. CONVENTIONAL RESISTANCETO PLANT VIRUSESA. IntroductionWhen one considers the advantages anddisadvantages of the control measures justdescribed, it is obvious that the use of cropplants that are resistant to viruses is likely tobe the most promising approach. Thus, formany years plant breeders have been attemptingto produce virus-resistant varieties. Thereare two sources of resistance gene: natural onesfrom sexually compatible species and nonconventionalones introduced by genetic modification;the latter is discussed in Chapter 15.There are three types of genes that plant breedersconsider for control of plant viruses: thoseconferring immunity, those conferring field resistance,and those conferring tolerance (Box 14.5).Some molecular aspects of these virus:plantinteractions are discussed in Chapter 10.The following points concerning the effectsof host genes on the plant’s response to infectionemerge from many different studies:• Both dominant and recessive Mendeliangenes may have effects. However, whilemost genes known to affect host responsesare inherited in a Mendelian manner,cytoplasmically transferred factors maysometimes be involved.• There may or may not be a gene doseeffect.• Genes at different loci may have similareffects.• The genetic background of the host mayaffect the activity of a resistance gene.• Genes may have their effect with all strainsof a virus or with only some.• Some genes influence the response to morethan one virus.• Plant age and environmental conditions mayinteract strongly with host genotype toproduce the final response.• Route of infection may affect the hostresponse. Systemic necrosis may developfollowing introduction of a virus by graftinginto a high-resistant host that does not allowsystemic spread of the same virus followingmechanical inoculation.• Resistance originally thought to be to thevirus may be really to the vector.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

V. CONVENTIONAL RESISTANCE TO PLANT VIRUSES281BOX 14.5TYPES OF RESISTANCE TO A PLANT VIRUSThere are three main types of resistance andimmunity to a particular virus:1. Immunity involves every individual of thespecies; little is known about the basis forimmunity, but it is related to the question ofthe host range of viruses discussed inChapter 2.2. Cultivar resistance describes the situationwhere one or more cultivars or breeding lineswithin a species show resistance, whereasothers do not.3. Acquired or induced resistance is present whereresistance is conferred on otherwisesusceptible individual plants followinginoculation with a virus.Some authors have considered that immunityand cultivar resistance are based on quite differentunderlying mechanisms. However, studieswith a bacterial pathogen in which only onepathogen gene was used show that for this classof pathogen at least the two phenomena havethe same basis.B. Genetics of Resistance to VirusesResistance to viruses in many crop viruscombinations is controlled by a single dominantgene (Table 14.1). However, this maymerely reflect the fact that most resistant cultivarswere developed in breeding programmesaimed at the introduction of a single resistancegene. Furthermore, incomplete dominance mayTABLE 14.1 Summary of Number of VirusResistance Genes Reported.Resistance GeneMonogenicOligo- orPolygenicDominant 81 10Recessive 43 20Incompletely dominant 15 6(Nature unknown) — 4Total number ofresistance genes139 40From Khetarpal et al. (1998; in Plant virus disease control,A. Hadidi, R.K. Khetarpal, and H. Koganezawa, Eds.,pp. 14–32, APS Press, St. Paul, MN).be a reflection of gene dosage or be due to environmentalfactors. Some specific examples ofdominant and recessive genes for resistancewere shown in Table 10.2. The current reportsabout the mechanisms of resistance by eithertotal immunity or hypersensitive responsewere also discussed in Chapter 10.In several plant species, the resistance virusresistance genes are clustered to specific locion the chromosomes; in this, virus resistanceresembles that for fungi. For instance, in Pisumsativum, the resistances to the lentil strain of Peaseed-borne mosaic virus, Bean yellow mosaic virus,Watermelon mosaic virus-2, Clitoria yellow veinvirus, and Bean common mosaic virus NL-8 strain(all potyviruses) are controlled by tightlylinked recessive genes on chromosome 2.C. ToleranceThe classic example of genetically controlledtolerance is the Ambalema tobacco variety.TMV infects and multiplies through the plant,but in the field, infected plants remain almostnormal in appearance. This tolerance is due toa pair of independently segregating recessiveIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

282 14. CONVENTIONAL CONTROLgenes, r m1 and r m2 , and perhaps to others, aswell with minor effects. On the other hand,tolerance to Barley yellow dwarf virus is controlledby a single dominant gene in barley,with different alleles giving different degreesof tolerance.D. Use of Conventional Resistancefor ControlA review of the consideration in a breedingprogramme for resistance to an importantvirus—that causing rhizomania of sugar beet—isgiveninScholtenandLange(2000).Manyoftheaspects that they discuss are applicable to breedingprogrammes for resistance to other viruses.In this section, we examine the application of conventionalresistance to the control of viruses.1. ImmunityAlthough many searches have been made,true immunity against viruses and viroids,which can be incorporated into useful crop cultivars,is a rather uncommon phenomenon.2. Field ResistanceWhere suitable genes can be introduced intoagriculturally satisfactory cultivars, breedingfor resistance to a virus provides one of the bestsolutions to the problem of virus disease. However,there are two major problems. It hasproved difficult to find resistance genes in speciesthat are sexually compatible with the cropspecies. There have been widespread searchesin wild species for such genes, and techniquesfor wide crosses, such as embryo rescue, havebeen used. Chemical and radiation mutagenesisof the crop plant has also been used to provideuseful resistanceThe second problem has been the durabilityof resistance. How long can the gene bedeployed successfully before a resistancebreaking (virulent) strain of the virus emerges?Of 87 host-virus combinations for which resistancegenes have been found, more than 75percent of those tested were overcome by virulentvirus isolates. Fewer than 10 percent of theresistance genes have remained effective whentested against a wide range of virus isolatesover a long period. However, some of the virulentisolates were found only in laboratory testsrather than field outbreaks.The costs of a breeding programme must beweighed against the possible gains in cropyield. Many factors are involved, such as theseriousness of the virus disease in relation toother yield-limiting factors; the “quality” ofthe available resistance genes—for example,resistance genes against CMV are usually“weak” and short-lived, which may be due, atleast in part, to the many strains of CMV thatexist in the field; the importance of the crop(compare, for instance, a minor ornamentalspecies with a staple food crop such as rice);and crop quality. Good virus resistance thatgives increased yields may be accompanied bypoorer quality in the product, as happenedwith some TMV-resistant tobacco cultivars.The difficulties in finding suitable breedingmaterial are compounded when there arestrains of not just one but several viruses to consider.Cowpeas in tropical Africa are infected toa significant extent by at least seven differentviruses. In such circumstances, a breedingprogramme may utilise any form of genetic protectionthat can be found. Sources of resistance,hypersensitivity, or tolerance have been foundfor five of the viruses. However, several of theseviruses have different strains or isolates thatmay break resistance to other isolates. There is,of course, the further problem of combiningthese factors with multiple resistance to fungaland bacterial diseases. For example, geneticresistance to TMV, cyst nematodes, root-knotnematodes, and wildfire from Nicotiana repandahas been incorporated into N. tabacum.3. ToleranceWhere no source of genetic resistance can befound in the host plant, a search for tolerantIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

VIII. SUMMARY283varieties or races is sometimes made. However,tolerance is not nearly as satisfactory a solutionas genetic resistance for several reasons:• The infected tolerant plants may act as areservoir of infection for other hosts. Thus, itis bad practice to grow tolerant and sensitivevarieties together under conditions wherespread of virus may be rapid.• Large numbers of virus-infected plants maycome into cultivation. The geneticconstitution of host or virus may change togive a breakdown in the tolerant reaction.• The deployment of tolerant varietiesremoves the incentive to find immunity tothe virus until the tolerance breaks down inan “out of sight, out of mind” attitude.• Virus infection may increase susceptibility toa fungal disease (see Chapter 10).However, tolerant varieties may yield verymuch better than standard varieties wherevirus infection causes severe crop losses andwhere large reservoirs of virus exist under conditionswhere they cannot be eradicated. Thus,tolerance has, in fact, been widely used. Cultivarsof wheat and oats commonly grown inthe midwestern United States have probablybeen selected for tolerance to BYDV in an incidentalmanner because of the prevalence ofthe virus.VI. STRATEGIES FOR CONTROLThree kinds of situations are of particularimportance: annual crops of staple foods suchas grains and sugar beet that are either grownon a large scale or are subsistence crops andthat under certain seasonal conditions may besubject to epidemics of viral disease; perennialcrops, mainly fruit trees with a big investmentin time and land, where spread of a virusdisease, such as citrus tristeza or plum pox,may be particularly damaging; and high-valuecash crops such as tobacco, tomato, cucurbits,peppers, and a number of ornamental plantsthat are subject to widespread virus infections.With almost all crops affected by viruses, anintegrated and continuing programme of controlmeasures is necessary to reduce crop lossesto acceptable levels. Such programmes willusually need to include elements of all threekinds of control measure just discussed. Indeveloping strategies for the integratedapproach, it is essential to have a full understandingof the disease, its epidemiology andecology, and the pathogen, its genetic makeupand functioning and its potential for variation.VII. VIRUSES OF OTHERKINGDOMSSome of the ways of controlling animalviruses, such as avoidance of infected individuals,are the same as those described previouslyfor plant viruses. However, plants donot have an innate immune system, so controlby immunisation is not a viable approach. Aswe will see in Chapter 15, there is an analogousapproach in plants to immunisation in that theycan be transformed to activate the RNA silencingdefence system. Also, they can be transformedto produce antibodies that have beenshown to mitigate some viruses.Although there are examples of geneticresistance to viruses infecting vertebrates, moreeffort is put into control by immunisation andchemoprophylaxis. There are some examplesof breeding virus resistance genes into invertebrates(e.g., shrimps).VIII. SUMMARY• There are four basic approaches tocontrolling plant virus diseases: avoidinginfection, stopping the vector, protecting theplant, and breeding for resistance.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

284 14. CONVENTIONAL CONTROL• The first two approaches involve agronomicpractices such as using clean plantingmaterial, changing the planting time, andusing insecticides against vectors.• Insecticides are better at preventing thespread of viruses with a persistentinteraction with their vector than those witha nonpersistent interaction.• Plants can be protected by inoculating themwith a mild strain of the virus (crossprotection).This is only viable with highcostperennial crops.• Breeding for resistance is considered to bethe best approach but has the difficulties ofsources of resistance genes in sexuallycompatible species and the durability ofresistance.Further ReadingBrown. J.K. (2008). Plant resistance to viruses. Encyclopediaof Virology, Vol. 4, 164–169.Caranta, C. and Dogimont, C. (2008). Natural resistanceassociated with recessive genes. Encyclopedia of Virology,Vol. 4, 170–176.Hull, R. (2002). Matthew’s plant virology. Academic Press,San Diego.Jones, R.A.C. (2006). Control of plant virus diseases. Adv.Virus Res. 67, 205–244.Lecoq, H., Moury, B., Desbiez, C., Palloix, A., and Pitrat, M.(2003). Durable virus resistance in plants through conventionalapproaches: A challenge. Virus Res. 100, 31–39.Moffett, P. and Klessig, D.F. (2008). Natural resistance associatedwith dominant genes. Encyclopedia of Virology,Vol. 4, pp. 170–176, Elsevier Academic Press, Oxford.Scholten, O.E. and Lange, W. (2000). Breeding for resistanceto rhizomania in sugar beet: A review. Euphytica, 112,219–231.Ziebell, H. (2008). Mechanics of cross-protection. CABReview: Perspectives in Agriculture, Veterinary Science,Nutrition and Natural Resources 3, No. 049, 1–13 (www.cababstractsplus.org/cabreviews).IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

C H A P T E R15Transgenic Plants and VirusesGenetic modification technology has opened up many possibilities for protecting plants againstvirus infection and for using plant viruses as gene vectors.O U T L I N EI. Transgenic Protection Against PlantViruses 285II. Pathogen-Derived Resistance 286III.Possible Uses of Plant Viruses forGene Technology 293IV. Viruses of Other Kingdoms 300V. Summary 301I. TRANSGENIC PROTECTIONAGAINST PLANT VIRUSESA. IntroductionIt is now possible to introduce almost anyforeign gene into a plant and obtain expressionof that gene. In principle, this should makeit possible to transfer genes for resistanceor immunity to a particular virus, across species,genus, and family boundaries. Furthermore,genes can be designed to interfere withdirectly, or induce the host to interfere with,the virus infection cycle. Several approaches toproducing transgenic plants resistant to virusinfection are being actively explored.There are essentially three sources of transgenesfor protecting plants against viruses: naturalresistance genes; genes that are derivedfrom viral sequences, giving what is termedPathogen-derived resistance (PDR); and genesfrom various other sources that interfere withthe target virus.B. Natural Resistance GenesMolecular aspects of genes found in plantspecies that confer resistance to various virusesComparative Plant Virology, Second Edition285Copyright # 2009, Elsevier Inc. All rights reserved.

286 15. TRANSGENIC PLANTS AND VIRUSESare discussed in Chapter 10. When a resistancegene has been identified, it can be isolated andtransferred to another plant species. The Rx1gene that gives extreme resistance to Potatovirus X (PVX) has been isolated from potatoand transformed into Nicotiana benthamianaand N. tabacum, where it provides resistanceto the virus. Similarly, the N gene, found naturallyin N. glutinosa, and that confers hypersensitiveresistance to TMV, gives resistance toTMV when transferred to tomato. Much of thespecificity of R genes is determined by the leucinerepeat regions (see Box 10.2); this regioncan be manipulated in vitro to give yet furthersources of resistance.II. PATHOGEN-DERIVEDRESISTANCESince the mid-1980s, PDR has attractedmajor interest and is the main method bywhich transgenic protection is being producedagainst viruses in plants. The concept of PDRis explained in Box 15.1. In this rapidly expandingsubject, there are various terminologicalproblems. The main one is whether to term thisphenomenon resistance or protection, since thereactions of various forms of transgenic protectiongive a great range of responses, varyingfrom delay in symptom production forjust a few days to complete immunity. Thisbook uses protection wherever possible, but insituations where it has been used widely (suchas pathogen-derived resistance), resistance isused.Currently, there are two basic molecularmechanisms by with PDR is thought to operate.In some systems the expression of an unmodifiedor a modified viral gene product interfereswith the viral infection cycle (called proteinbasedprotection). The second mechanism doesnot involve the expression of a protein product(called nucleic acid–based protection).A. Protein-Based ProtectionAs noted in Box 15.1, the first demonstrationof PDR involved the expression of TMV coat protein.Since then, there have been many examplesof the use of this coat protein–mediated protection;the phenomenon is often referred to as“coat protein–mediated resistance.” The expressionof other viral gene products also gives protectionto a greater or lesser extent against thetarget virus.1. Transgenic Plants Expressing a ViralCoat ProteinThe sequences encoding viral coat proteinsare the most widely used for conferring protectionin plants because this gene was used in thefirst example of this approach and because coatprotein genes are relatively easy to identify andto clone. These are some of the properties of theprotection given by of TMV coat protein:• The higher the amount of virus inoculum,the lower the protection afforded by TMVcoat protein.• There is a positive correlation between thelevel of protections and the sequence similaritybetween the transgene coat protein and that ofthe challenge virus. For instance, TMV coatprotein gives better protection against Tomatomosaic virus, which has 82 percent sequenceidentity than against Ribgrass mosaic virus,which has 45 percent identity.• TMV coat protein does not protect againstinoculation with viral RNA.The evidence suggests that the resistance conferredby the coat protein is an early event inthe virus infection cycle, possibly by affectingcotranslational disassembly of the incomingvirus particles (see Box 7.1 for an explanationof cotranslational disassembly).2. Other Viral ProteinsAmong other virus-encoded proteins thathave been explored experimentally to giveIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

II. PATHOGEN-DERIVED RESISTANCE287BOX 15.1PATHOGEN-DERIVED RESISTANCEThe ideas leading up to the concept of pathogenderivedresistance for plant viruses are encapsulatedas a general concept in a paper bySandford and Johnson (1985). They suggested thatthe transgenic expression of pathogen sequencesmight interfere with the pathogen itself termingthis concept parasite-derived resistance. Since then,several names have been used for this approachincluding nonconventional protection, transgenic resistance,and engineered virus resistance, but thegenerally accepted term is now pathogen-derivedresistance (PDR).The basic idea arising out of Sanford andJohnson’s concept is that, if one understands themolecular interactions involved in the functioningof a pathogen, mechanisms can be devised for interferingwith them. Although this concept applies toall pathogens and invertebrate pests, it has mainlybeen used against plant viruses because of their relativelysimple genomes. In developing the conceptit was recognised that the interactions of interestoccur at all stages of the virus infection cycle andthat they can potentially be interfered with in variousways—for example, by blocking the interactionor by decoying one or more of the moleculesinvolved in the interaction. This then led to the ideaof the overall strategy as being one of attackingspecific viral “targets” with specific molecular“’bullets.” However, in practice, much of thedevelopment of this approach was done withoutdetailed knowledge of the precise molecularmechanisms involved, and analysis of theseresults has thrown light on several new mechanisms.Perhaps the most important is the genesilencing phenomenon described in Chapter 11.The first demonstration of PDR against plantviruses was by Powell-Abel et al. (1986), whoshowed that the expression of Tobacco mosaicvirus (TMV) coat protein in tobacco plants protectedthose plants against TMV. They showedthat transgenic plants expressing TMV coat proteineither escaped infection following inoculationor developed systemic disease symptomssignificantly later than plants not expressingthe gene. Plants that showed no systemic diseasedid not accumulate TMV in uninoculated leaves.Transgenic plants produced only 10 to 20 percentas many local lesions as controls wheninoculated with a strain of TMV causing locallesions. The idea that transgenic plants resist initialinfection rather than subsequent replicationwas suggested by results obtained using transgenicXanthi nc tobacco plants, in which fewerlocal lesions were produced than on controlplants. However, the lesions that did developwere just as big as on control leaves, indicatingthat once infection was initiated, there was nofurther block in the infection cycle.protection are both complete or modified cell-tocellmovement proteins and replicase proteins.The rationale behind using modified proteinsis that they would block an essential stage ofthe infection cycle. Varying results have beenobtained, with some constructs giving reasonableprotection. However, follow-up experimentshave shown that the protection could bebypassed by another virus complementing thedefect.B. Nucleic Acid–Based ProtectionThree potential forms of protection basedon the expression of viral RNA sequences havebeen recognised: that induced by the viral RNAIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

288 15. TRANSGENIC PLANTS AND VIRUSESBOX 15.2UNEXPECTED RESULTS WITH COAT PROTEINPROTECTIONIn the early days of developing pathogen-derivedresistance (see Box 15.1), it was thought that a viralprotein had to be expressed to confer the protection.However, some unexpected results werefound, especially with controls designed not toexpress the protein (this shows the importance ofcontrols!). For instance, there was no correlationbetween the protection and the expression ofPotato virus Y (PVY) or Potato leaf roll virus coatproteins in potato. The untranslatable coat proteingene of Tobacco etch virus, Tomato spotted wilt virus,and PVY gave higher levels of protection thaneither full-length or truncated translatable constructs.This led to the realisation that the protectionwas RNA-mediated and was an importantstep in the understanding of the RNA silencingdefence system.sequence, that induced by the expression of satelliteRNAs, and that in which ribozymes aretargeted to viral genomes. Early in the developmentof coat protein–mediated protection, therewere some unexpected observations (Box 15.2).These and other observations suggested that theprotection, at least in these cases, was mediatedby nucleic acid rather than by protein.1. RNA-Mediated ProtectionAs no promoterless transgenes have beenshown to confer protection, it must be assumedthat either the RNA transcript or a protein thatis encoded give the protection. In a plant that istransformed with a construct that does not givea protein, any protection is obviously due tothe RNA. When a plant is transformed with aconstruct designed to produce a viral protein,it can often be difficult to distinguish betweenprotection due to the expression of the proteinitself or due to the RNA transcript. However,there are various features of the protection thattend to be characteristic for RNA-mediatedprotection:• There is no direct correlation between RNAexpression levels and the level of protection.• Usually, no transgene-encoded protein canbe detected, and the steady state of thetranscript in inoculated plants is often in lowamounts.• The protection is usually narrow and againststrains of the virus that have very similarsequences to that of the transgene.• Unlike coat protein–mediated protection, theprotection is not overcome by inoculatingRNA.• Also, unlike coat protein–mediatedprotection, RNA-mediated protection is notdose dependent and operates at high levelsof inoculum.• The insert in the host genome often comprisesmultiple copies of the transgene, particularlywith direct repeats of coding regions.• Copies of the transgene may be truncatedand/or in an antisense orientation.• Transgene sequences and sometimes theirpromoter(s) may be methylated.2. Molecular Basis of RNA-MediatedProtectionWhen transcript levels have been examined,three general classes of resistance phenotypehave been recognised:1. Plants that are fully susceptible. These plantshave low to moderate levels of transgenetranscription and steady-state RNA.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

II. PATHOGEN-DERIVED RESISTANCE2892. Plants that become infected and then recover.These have moderate to high levels oftransgene transcription and steady-stateRNA in uninfected plants but low-levelsteady-state RNA in recovered tissues.3. Plants that are highly resistant. These plantshave high levels of transgene expression butlow steady-state levels.The recovery phenomenon associated withlow steady states of transgene RNA in recoveredtissues and the low steady-state RNAlevels in highly resistant plants, coupled withthe narrow range of protection against viruseswith homologous sequences to the transgeneare strongly indicative of homology-dependentRNA silencing, which is described in detail inChapter11.Because of the variation in response, largenumbers of independent transformants shouldbe tested, not only to obtain lines with the bestprotection characteristics but also to rule outthe possibility that protection is not given by aparticular construct.3. Sequences for RNA-Mediated ProtectionRNA-mediated protection has been givenby a range of sequences from viral genomes.In many cases, it has resulted from attemptsto transform plants with the viral genes justdescribed or from constructs designed to produceantisense RNA, which is intended toblock translation of the viral mRNA.It is the constructs that are important asthey need to transcribe to give dsRNA, whichinitiates the silencing pathway. The hairpinconstruct comprising sequences that are transcribedto give (þ) strand and (–) strand ofthe target (viral) RNA separated by a spacer(Figure 11.2F) are the most efficient.4. RibozymesAs described in Box 3.2, ribozymes arecatalytic RNAs that can cleave at specific sitesin complementary target RNAs. Since the ribozymehas to be complementary to the targetviral sequence, it can be considered to be anantisense RNA. Thus, any effect can be difficultto distinguish from that of RNA silencing. Incorporationof a ribozyme into an antisense RNA toTMV gave no significant advantage over theantisense RNA itself but constructs directedagainst Plum pox virus that containing a hammerheadribozyme gave stronger protectionthan the ordinary antisense RNA construct.5. Relationship Between Natural Cross-Protection and Protection in TransgenicPlantsThe mechanism for transgenic protectionagainst a virus infection, especially coat protein–mediatedprotection, has been comparedwith natural cross-protection or mild-strainprotection (see Chapter 10). There are severalsimilarities that have been used to support theidea:• In both situations, the degree of resistancedepends on the inoculum concentration,with high concentrations reducing theobserved resistance.• Both are effective against closely relatedstrains of a virus, less against distantlyrelated strains, and not at all againstunrelated viruses.• In some circumstances, cross-protectioncan be substantially overcome whenRNA is used as inoculum rather thanwhole virus.On the other hand, there appear to be some differencesbetween natural cross-protection andcoat protein–induced resistance. When crossprotectionbetween related strains of a virus isincomplete, the local lesions produced may bemuch smaller than in control leaves. This indicatesreduced movement and/or replication ofthe superinfecting strain.It is quite possible that there are severalmechanisms that give cross-protection. One ofthem is likely to involve the RNA silencing hostdefence system and thus, to resemble RNAmediatedprotection.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

290 15. TRANSGENIC PLANTS AND VIRUSES6. Transgenic Protection by Satelliteand DI Nucleic AcidsThe general nature of satellite RNAs isdescribed in Chapter 3, and the ability of somesatellite RNAs to attenuate the symptoms of thehelper virus was also discussed there. Defectiveinterfering (DI) nucleic acids are described inChapter 10. They are mutants of viral genomesthat are incapable of autonomous replicationbut contain sequences that enable them to bereplicated in the presence of the parent helpervirus. In many cases, they are amplified at theexpense of the parent virus, ameliorating thesymptoms induced by that virus.Transgenic plants expressing either satelliteor DI RNA are less severely diseased wheninoculated with the helper virus. However,the use of satellite RNAs in transgenic plantsto protect against the effect of virus infectionhas both advantages and disadvantages. Theprotection afforded is not affected by the inoculumconcentration, as it is with viral coat proteintransformants. The losses that do occur intransgenic plants because of slight stunting willaffect only the plants that become naturallyinfected in the field. Furthermore, the resistancemay be stronger in transgenic plants thanin plants inoculated with the satellite. Inoculationis not needed each season, and the mutationfrequency is lower. Nevertheless, thereare distinct risks and limitations with the satellitecontrol strategy. The satellite RNA couldcause virulent disease in another crop speciesor could mutate to a form that enhances diseaserather than causing attenuation (see Chapter14). Another risk is the reservoir of virus availableto vectors in the protected plants. Last, thesatellite approach will be limited to thoseviruses for which satellite RNAs are known.C. Other Forms of Transgenic ProtectionVarious forms of protection against viruseshave been shown for a variety of transgenesthat are not derived from viruses themselves.For instance, the ß-1,3-glucanases are proteinsbelieved to be part of the constitutive andinduced defence system of plants against fungalinfection. Unexpectedly, plants deficientin these enzymes due to expression of an antisenseRNA show markedly reduced lesionnumber and size on inoculation of Nicotianatabacum Havana 425 with TMV and reducedseverity and delay of symptoms of TMV intransgenic N. sylvestris.Plants do not have an immune system likethat of animals, in which specific antibodyproteins are formed in response to an infection.Transgenic plants can produce antibodies(Box 15.3): The expression in plants of a single-chainFv antibody derived from a panel ofmonoclonal antibodies against African cassavamosaic virus coat protein reduced the infectionincidence and delayed symptom development.In mammalian systems, interferons areeffective antiviral molecules. When one of thecomponents of the virus-inhibiting pathway,2 0 -5 0 oligoadenylate synthetase was expressedin potato plants, it gave protection againstPVX. The virus concentration in transgenicplants was lower than it was in plants expressingPVX coat protein.D. Field Releases of Transgenic Plants1. Potential RisksThere have been concerns expressed aboutthe release and use of plants modified bygenetic manipulation. This has led to plantsproduced by this means being treated in a differentmanner from those produced by conventionalbreeding techniques and being subject tospecific regulatory structures. The concerns arein two areas: potential risk to human and otheranimal health and potential risk to the environment.There are strong controls over the use ofnew sources of human food and animal feed,and, as plant viruses are a normal componentof human and animal diet, little problem isIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

II. PATHOGEN-DERIVED RESISTANCE291BOX 15.3ANTIBODIES IN PLANTSIt has long been assumed that plants could notproduce antibodies. However, plants transformedwith constructs expressing either singlegamma (heavy) chains (see Box 13.2 for antibodystructure) or single kappa (light) chains and thencrossed yield a functional antibody comprisingmore than 1 percent of the leaf proteins. Thetwo variable chains of an antibody can beexpressed in E. coli as a fusion protein by joiningthem with a flexible linker protein. These can beput into a phage display system, and a verylarge number of combinations of the two chainscan be expressed and then selected. Suitableclones can then be transformed into plants.foreseen here. However, there has been a longtermdebate over the use of the Cauliflowermosaic virus (CaMV) 35S promoter in transformationconstructs (Box 15.4).A major consideration for the use of plantlines transgenically protected against viruses isthe possible environmental risks. The area ofvirus-protecting transgenes that has attractedspecific interest is the use of virus sequences.The basic question is “What is the risk of anyinteractions that might arise between a virus orvirus-related sequence integrated in the hostgenome and another virus super-infecting thatplant?” Three scenarios are considered: heteroencapsidation,recombination, and synergism.Heteroencapsidation involves the superinfectionof a plant expressing the coat proteinof virus A by the unrelated virus B, the expressionof the virus A coat protein not protectingthe plant against virus B. The risk is that theBOX 15.4USE OF THE CAULIFLOWER MOSAIC VIRUS (CaMV) 35SPROMOTER IN TRANSGENIC PLANTSThe CaMV 35S promoter is one of the mostwidely used promoters for transforming plantsand is used in most currently released geneticallymodified (GM) crop plants. The similarityin replication between CaMV and retroviruses(see Chapter 8) has led to suggestions that thispromoter might give rise to some risks. One ofthe suggested scenarios is that on eating GMcrops, the 35S promoter could integrate into thehuman genome leading to the activation of cancers.This argument has been countered bypointing out that humans have been eating brassicasinfected with CaMV for many hundreds ofyears and that no problem has been identified.Furthermore, infected plants contain more than1,000 times as many copies of the promoter asdo GM plants, and the human digestive systembreaks down plant DNA to small pieces. Thisillustrates the arguments about GM crops andthe use of viral sequences in producing them.For further information, see Hull et al. (2000).IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

292 15. TRANSGENIC PLANTS AND VIRUSEScoat protein of virus A might encapsidate thegenome of virus B, thereby conferring on itother properties, such as different transmissioncharacteristics (Figure 15.1). Heteroencapsidationby transgenically expressed coat proteinhas been reported for closely related viruses,such as Cucumber mosaic virus and Alfalfa mosaicvirus, and between potyviruses but only occursFIGURE 15.1 Hypothetical scheme to illustrate heteroencapsidation.The plant on the left is transformed withthe coat protein of virus A, a rod-shaped virus transmittedby aphids; it is protected against virus A (top right) but notagainst virus B, an isometric virus transmitted by whitefly(middle right). In the heteroencapsidation situation, thegenome of virus B is encapsidated by the coat protein ofvirus A expressed in the transgenic plant and thus the heteroencapsidantwould be transmissible by aphids (bottomright). As noted in the text, this situation has not beenfound with unrelated viruses.between closely related viruses, the particles ofwhich have similar forms of stabilisation.Recombination between the transgene andsuperinfecting virus might lead to a new virus.As described in Box 4.5, there are examplesof “new” viruses naturally arising from recombination.The question is, would the deploymentof plants transgenic in viral sequenceslead to any more recombination than occurs inthe natural situation?As described in Chapter 8, synergistic interactionbetween two unrelated viruses arepotentiated by distinct virus sequences. Thus,there is a possibility that the effect of a superinfectingvirus could be exacerbated by a viraltransgene. However, it appears that synergismis due to interactions between suppressors ofRNA silencing (see Box 11.3), so the use of suchgenes should be avoided in transgenic constructdesign.The construct design is important in minimisingpotential risk. Most of the precedingscenarios depend on a viral protein or largetranscript being produced in the transgenicplant. The use of the RNA silencing protectionapproach leads to the production of relativelyshort transcripts that are processed to siRNAand thus minimise any potential risk. If a viralprotein has to be produced, understandingthe molecular interactions involved in thepotential risk situations can lead to methodsfor the “sanitising” of the transgene. For example,as described in Chapter 12, aphid transmissionof potyviruses involves an amino acidtriplet (asp, ala, gly; DAG) in the coat protein,the mutation of which blocks aphid transmissiondoes not affect the protection offeredby the transgene. Similarly, understanding ofthe factors involved in recombination will leadto transgene constructs that lessen the possibilityof new molecules being formed between thetransgene and a superinfecting virus.In all of these risk assessments, it is importantto compare the transgenic situation withthe nontransgenic situation. Thus, there is theIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

III. POSSIBLE USES OF PLANT VIRUSES FOR GENE TECHNOLOGY293possibility of the described above potentialrisks occurring in mixed natural infectionsbetween viruses.2. Field PerformanceTesting the field performance of transgenicplants is essentially no different from testingplant lines that have been obtained by traditionalbreeding. The testing objectives include evaluatingthe plant appearance, typeness, growth vigour,yield, and quality. Of especial importance isto assess the stability and durability of the protectingtransgene under these conditions. Twomain factors can affect stability and durability:possible climatic effects on the expression of thetransgene and the presence of protection-breakingstrains or isolates of the virus that are presentin the viral ecosystem but were not recognised inthe initial glasshouse tests.The use of transgenic protection in papaya toPapaya ring spot virus has been a great successin controlling the disease in Hawaii and maintainingpapaya production there. However,attempts to transfer the technology to Thailandshowed the necessity of having the constructtargeted to the local strains of the virus.III. POSSIBLE USES OFPLANT VIRUSES FOR GENETECHNOLOGYA. DNA Viruses as Gene VectorsIn the early 1980s, there was considerableinterest in the possibility of developing plantviruses as vectors for introducing foreign genesinto plants. At first, interest centred on thecaulimoviruses, the only plant viruses withdsDNA genomes, because cloned DNA of theviruses was shown to be infectious. Interest laterextended to the ssDNA geminiviruses and thento RNA viruses when it became possible toreverse transcribe these into dsDNA, whichcould produce infectious RNA transcripts.These are the main potential advantages of aplant virus as a gene vector:• The virus or infectious nucleic acid could beapplied directly to leaves, thus avoiding theneed to use transformation technologies andthe consequent difficulties in plantregeneration.• It could replicate to high copy number.• There would be no “position effects” ofinsertion into a site in the plantchromosomal DNA.• The virus could move throughout the plant,thus offering the potential to introduce agene into an existing perennial crop, such asorchard trees.Such a virus vector would have to be able tocarry a nonviral gene (or genes) in a way thatdid not interfere with replication or movementof the genomic viral nucleic acid. Ideally, itwould also have the inability to spread fromplant to plant in the field, providing a naturalcontainment system; induction of very mild orno disease; a broad host range, which wouldallow one gene vector to be used for many speciesbut would be a potential disadvantage interms of safety; and maintenance of continuousinfection for the lifetime of the host plant.These are the major general limitations in theuse of plant viruses as gene vectors:• They are not inherited in the DNA of thehost plant, and therefore genes introducedby viruses cannot be used in conventionalbreeding programmes.• Plants of annual crops would have to beinoculated every season unless there was avery high rate of seed transmission.• The foreign gene introduced with the viralgenome may be lost quite rapidly byrecombination or other means with the virusreverting to wild type.• It would be necessary to use a virus thatcaused minimal disease in the crop cultivar.The virus used as vector might mutate toIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

294 15. TRANSGENIC PLANTS AND VIRUSESproduce significant disease or be transmittedto other crops that were susceptible.Infection in the field with an unrelatedsecond virus might cause very severedisease.1. CaulimovirusesThere appear to be several constraints to theuse of CaMV as a gene vector. These includethe packing capacity of the CaMV particle,the amount of viral DNA that can be removedwithout affecting the functioning of the genome,and the interactions between different parts ofthe genome in expression and replication.Removal of nonessential regions of the genomeshould enable about 1,000 bp of sequence to beinserted, but it is not certain as to whether allthis sequence is really nonessential.2. GeminivirusesMuch attention has been focused on thegeminiviruses as potential gene vectors becauseof their DNA genomes and because the smallsize of the genomes makes them convenient forin vitro manipulations. Nevertheless, this smallsize may restrict the amount of viral DNA thatcan be deleted. However, this is counterbalancedby the fact that for some geminiviruses aviable coat protein and encapsidation are notnecessary for successful inoculation by mechanicalmeans or for systemic movement throughthe plant.There are other potential difficulties. Recombinationcan occur to give parental-type molecules.Most geminiviruses are restricted mainlyto the phloem and associated cells. However,the wide host range of the geminiviruses (comparedwith the caulimoviruses) makes them ofconsiderable interest. The fact that some membersinfect cereal crops would be particularlyuseful except they are not seed transmitted andare mechanically transmitted only with difficulty.In any event, inoculations on the scaleneeded for cereal crops would be impractical.B. RNA Viruses as Gene VectorsThe ability to manipulate RNA virus genomesby means of a cloned cDNA intermediatehas opened up the possibility of using RNA aswell as DNA viruses as gene vectors. In principlethe known high error rate in RNA replication(see Figure 4.5) might place a limitation onthe use of RNA viruses. The experimental evidenceto date suggests that mutation may notbe a major limiting factor, at least in the shortterm. Viruses with isometric and rod-shapedparticles have been studied as potential vectors,but those with rod-shaped particles have betterpotential because there is less constraint on theamount of nucleic acid that can be inserted. Fourbasic strategies have been used for RNA virusgene vectors (see Figure 15.2).In the gene replacement strategy, a viralgene is removed and replaced with the geneof interest. This has been attempted with severalRNA viruses with varying success, butbecause of the integrated and coordinatedexpression of most viral genomes, it does notappear to be a general viable approach.Instead of removing a viral gene, in the geneinsertion strategy, the gene of interest isFIGURE 15.2 Comparison of strategies used toexpress foreign genes (black boxes) from differentviruses; white boxes indicate viral genes. The epitopemethod involves translational fusion (a) of a smallsequence inside the coat protein gene or translationalread through (b) of an amber stop codon (*) at the 3 0end. [From Scholthof et al. (1996; Ann. Rev. Phytopathol.34, 299–323. Reprinted, with permission, from theAnnual Review of Phytopathology, Volume 34 # 1996 byAnnual Reviews www.annualreviews.org).]IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

III. POSSIBLE USES OF PLANT VIRUSES FOR GENE TECHNOLOGY295inserted usually attached to a viral subgenomicpromoter. This enables the gene of interest tobe expressed in a manner similar to downstreamviral genes (see Chapter 7 for subgenomicRNAs). Viable gene vector systemshave been established for several rod-shapedRNA viruses [e.g., Tobacco mosaic virus (TMV),Potato virus X (PVX), and Tobacco rattle virus(TRV)] whose structure can accommodate theextra nucleic acid from the insert. One of theadvantages of TRV is that it expresses foreignproteins efficiently in roots. One problem withthis approach is that if the subgenomic promoterfor the gene of interest is the same asthat for the normal viral gene, there is a likelihoodof recombination leading to loss of theinsert. For the TMV vector, this was overcomeby developing a hybrid vector containingsequences from two tobamoviruses: TMV-U1and Odontoglossum ring spot virus (ORSV). In thisvector, the gene of interest is expressed from theTMV coat protein subgenomic promoter and thecoat protein from the ORSV promoter.In the epitope presentation strategy the epitope-encodingsequence is inserted into a locationin the viral coat protein gene so it ispresented on the surface of the virus particle.This will be described in more detail following.For the complementation system, the geneof interest is inserted into a virus that is “disarmed”through the removal of an essentialgene. This vector is then inoculated to plantstransgenic in the “missing” gene, which complementsthe defect. Infectious clones of defectiveinterfering (DI) RNAs may be used asvectors.C. Viruses as Sources of ControlElements for Transgenic PlantsCertain plant viral nucleic acid sequenceshave been found to have useful activity ingene constructs as promoters of DNA andRNA transcription and as enhancers of mRNAtranslation.1. DNA PromotersTranscription of CaMV DNA gives rise toa 19S and a 35S mRNA (Chapter 7). The 19Sand 35S promoters are both strong constitutivepromoters, but the 35S promoter is muchmore effective than the 19S in several systems.For example, expression of the a-subunit ofß-conglycinin in petunia plants under controlof the 35S promoter was 10–50 times greaterthan that from the 19S promoter. A variant(enhanced) 35S promoter that contains a tandemduplication of 250 bp of upstream sequencesgives about a tenfold increase in transcriptionalactivity.Several of the promoters from other DNAviruses, such as badnaviruses and geminiviruses,have been shown to have activity intransformation constructs but are not used aswidely as the CaMV 35S promoter.2. RNA PromotersThe subgenomic promoters for several RNAviruses have now been identified. These areused in the RNA virus vectors just describedand may prove useful for gene amplification.3. Translation EnhancersUntranslated leader sequences of severalviruses have been shown to act as very efficientenhancers of mRNA translational efficiency bothin vitro and in vivo and in prokaryotic and eukaryoticsystems (see Chapter 7). For instance, theTMV leader sequence (the O sequence) enhancestranslation of almost every mRNA in every systemthat has been tested.D. Viruses for Producing VaccinesThere are two approaches to using plantviruses to produce vaccines and other materialsof medical and veterinary interest: expressingthe target vaccine gene using a plant viral vectoror presenting the epitope on the coat proteinof a plant virus (Figure 15.3).IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

296 15. TRANSGENIC PLANTS AND VIRUSESFIGURE 15.3 Plant-derived vaccine research strategies.[This article was published in Vaccine 23, C. Arntzen, S.Plotkin, and B. Dodet, Plant-derived vaccines and antibodies:Potential and limitations, pp. 1753–1756, CopyrightElsevier (2005).]1. Vaccines Using Plant Virus VectorsTarget virus proteins have been expressedfrom plant RNA virus vectors. For instance,hepatitis B surface antigen and Norwalk viruscapsid proteins expressed in Nicotiana benthamianaand tomato assembled into virus-like particlesand conferred immunogenicity in animalswhen applied orally.2. Viruses for Presenting HeterologousPeptidesThe production of small peptides is requiredfor several reasons, including acting as epitopesfor vaccines and biologically active peptides.As described in Box 13.2, epitopes are patchesof amino acids that adopt specific conformations.Free peptides can act as epitopes, but theimmunogenicity is enhanced by presentationon the surface of a macromolecular assembly.One approach to presenting the peptidesequence in the correct conformation is to incorporateit into a viral coat protein sequencein such a way that they are exposed on the surfaceof the virus particle. The virus particle canthen be used as a vaccine. There are severaladvantages to doing this with plant viruses:• The virus can be produced in large amountsand in less-developed countries where thetechnology for animal virus vaccineproduction may be limited.• Such vaccines may be given orally as part ofthe normal food supply.• The virus will not infect humans or animalsand thus is completely inactive.• The system is not subject to contaminationby other virulent animal pathogens.A potential disadvantage is that the highrate of mutation of RNA viruses could resultin the deletion or loss of inserted sequences,especially as they would not be under selectionpressure; however, experience is indicating thatthis problem may not be as significant as at firstfeared. Several plant viruses have been used forthe presentation of foreign peptides.a. Cowpea Mosaic Virus (CPMV). The structureof the particles of CPMV has been solvedto atomic resolution. The capsid comprisestwo types of protein: the L protein that hastwo ß-barrel (see Chapter 5 for capsid proteinstructure) domains and the S protein thathas one ß-barrel domain. Analysis of the threedimensionalstructure suggested that loopsbetween the ß-strands would be suitable forthe insertion of sequences to be expressed asepitopes, as these loops are not involved in contactsbetween protein subunits. The ßB-ßC loopof the S protein is highly exposed and was usedfor most of the insertions (Figure 15.4); someinsertions have been made in other loops.Early studies on inserting sequences at theßB-ßC loop site gave guidelines for constructionIV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

FIGURE 15.4 Generation of chimeric Cowpea mosaic virus particles. Foreign sequences are inserted into the gene for the Scoat protein on RNA2. Both RNA1 and RNA2 are translated into polyproteins and undergo a cascade of cleavages, whosesites and final products are shown. RNA2 (bearing the heterologous sequence) needs to be coinoculated with RNA1(unmodified) to initiate infection in cowpea plants. S protein harbouring a foreign peptide in its ßB-ßC loop of the ß-barrelstructure (see Chapter 5) and native L protein assemble at 60 copies each into icosahedral virus particles on which the foreigninsert is expressed around the 5 0 -fold axes of symmetry. [From Porta and Lomonossoff (1998; Rev. Medical Virol. 8, 25–41).]

298 15. TRANSGENIC PLANTS AND VIRUSESof viable, genetically stable chimeras. Theseincluded: (i) foreign sequences should beinserted as additions and not replacements ofthe CPMV sequence; (ii) sequence duplicationshould be avoided as this led to loss of insertby recombination; (iii) the precise site of insertionwas important for maximising growth ofchimeras. Understanding of these guidelinesgave a standard procedure for inserting foreignDNA into the ßB-ßC loop of the S protein.Chimeras with inserts for the sequence for upto 38 amino acids have been successfully madein which the presence of foreign sequences didnot significantly affect the ability of the modifiedvirus to replicate. Various epitopes have beeninserted, including ones from Human rhinovirus14, Human immunodeficiency virus type 1, andCanine parvovirus.b. Tobacco Mosaic Virus (TMV). With thedevelopment of infectious cDNA clones toTMV, it was possible to use a self-replicatingsystem in plants. Fusion of a foreign sequenceto the C-terminus of the coat protein preventedparticle assembly. To overcome this problem,the insert was placed after an amber stop codonat the C-terminus of the coat protein gene(Figure 15.5A) so it could be expressed as aread-through protein. Particles were assembledwith about 5 percent of the coat protein subunitsexpressing the inserted sequence. Replacementof two amino acids on a surface loopnear the C-terminus of the coat protein gaveparticles with 100 percent of the subunits containingthe insert (Figure 15.5B), as did insertsinto another part of the C-terminal region notinvolved in particle assembly (Figure 15.5C).E. Viruses in Functional Genomicsof PlantsThe gene silencing induced by virus infectionof plants is described in Chapter 11.In virus-induced gene silencing (VIGS) a geneincorporated into a virus vector—say, a TMVbasedvector, a PVX-based vector, a TRV-basedvector, or, for monocots, a Barley stripe mosaicvirus–based vector—will generate dsRNA andthus silence a homologous gene in a plant(Figure 15.6).In selecting a vector for a specific purpose,the virus must be able to infect that host.Among the advantages of this approach togenomics is as the virus construct is inoculatedto seedlings or mature plants, it overcomes theproblem with insertional mutagenesis (theother main approach) of identifying geneswhose disruption is lethal before the plant hasdeveloped. With the VIGS approach, the lethalitywould be apparent from the death of themature plant that had been inoculated. VIGSis also simple, rapid, and allows for a highthroughputscreening. However, as describedin Chapter 11, viruses can suppress gene silencing,so the effect on the plant gene may be onlytemporary. The VIGS phenomenon was initiallyshown for transgenes and has now beendemonstrated for a wide range of plant genes(Figure 15.7).F. Plant Viruses in NanotechnologyPlant viruses, such as CPMV, have a numberof features that can be exploited for nanoscalebiomaterial fabrication. These include theordered structure of the capsids, the detailedknowledge of the molecular biology of thevirus, the ease of inoculation and high yield,and the lack of biological hazards to humans.The structure of the capsid provides regularlyspaced attachment units for a nanoscaffold.The approach is similar to that of presentationof epitopes and involves decorating surfaceresidues such as lysine, tyrosine, and carboxylategroups with functional compounds to formthe nanoscaffold.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

FIGURE 15.5 Production of chimeric Tobacco mosaic virus (TMV) particles in planta. Foreign oligonucleotide sequencesare introduced at one of three positions, labeled A, B, and C, in the gene of TMV coat protein, which is expressed fromthe most 3 0 of the viral subgenomic RNAs. In vitro transcripts of the altered full-length genomic cDNA are inoculated ontotobacco plants. The resulting recombinant TMV coat proteins are represented as ribbon drawings with the numbers, indicatingthe insertions site. Upon assembly of these coat proteins, chimeric rod-shaped virions are formed on which the foreignpeptides are differentially displayed and distributed; a maximum of 5% of the coat proteins present an insert in position A;all of the coat proteins are modified in positions B and C. [From Porta and Lomonossoff (1998; Rev. Medical Virol. 8, 25–41).]

300 15. TRANSGENIC PLANTS AND VIRUSESABFIGURE 15.6 Examples of VIGSconstructs showing different approaches.A. Tobacco rattle virus (seeProfile 15 for TRV genome organisation).For the VIGS tests the RNA1construct and the modified RNA2-construct containing the target insertare inserted into an Agrobacteriumtumefaciens plasmid, which is theninoculated into the target species. Thetop diagram shows the genome organisationof a cDNA clone of RNA1;Lb is left border; Rb is right border;35S, Cauliflower mosaic virus 35S promoter.The box contains informationon RNA2. The top line shows thegenome organisation; the middle lineshows the agroinoculation plasmidPTV00 in which ORFs P2b and P2care removed; the arrow indicates theduplicated coat protein promoter andthe triangle, a multiple cloning sitefor insertion of the VIGS sequence targetinga specific host gene. The bottomline is a construct for agroinoculationto obtain gene silencing in roots. [Datafrom Valentine et al. (2004; PlantPhysiol. 136, 3999–4009).] B. Barleystripe mosaic virus constructs for VIGSin monocots; (see Profile 1 for BSMVgenome organisation). BSMV infectionrequires three RNA species (a, ß, g),cDNAs of which are cloned behind aT7 promoter and transcripts areexpressed in Escherichia coli for inoculationto plants. The top two lines showthe genome organisation of RNAs aand ß. The bottom line shows RNA gwith the site modified to enable insertionof the VIGS target sequence. [Datafrom Halzberg et al. (2002; Plant J. 30,315–327).]IV. VIRUSES OF OTHERKINGDOMSViruses are also being used to modify vertebrates.There is an increasing number of casesof viral vectors (such as adenoviruses and retroviruses)being used to deliver therapeuticgenes into the host genome for gene therapy.Unlike plant viruses, the animal virus, togetherwith the gene of interest, integrates into thehost genome. Promoter sequences, such asthose from Simian virus 40 and Human herpesvirus5 (cytomegalovirus), have been used widelyin constructs for transforming animals.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

V. SUMMARY301FIGURE 15.7 Some examples of VIGS experiments. A. Tomato leaf infected with Tobacco rattle virus (TRV); B. PDS genesilenced in tomato using a TRV construct; photobleached tissue indicates regions of silencing; C. TRV:Prf silenced tomatoleaf treated with an avirulent Pseudomonas syringae strain; visible disease symptoms are the result of loss of resistancebecause of the suppression of the Prf gene. [From Burch-Smith et al. (2004; The Plant J. 39, 734–746, Wiley-Blackwell).]V. SUMMARY• Genetic modification technology has openedthe possibility of conferring protection inplants against viruses.• The most commonly used sequencesfor protecting plants are viral sequences,either coding for a viral protein whichinterferes with the replication cycle ofthe target virus, or a noncoding sequence,which primes the RNA silencing defencesystem.• Plants containing viral transgenes have beenreleased to the field but are subject to strongregulations.• Plant viruses also provide sequences such aspromoters and translation enhancers forconstructs used for introducing other genesinto plants.• The coat proteins of plant viruses have beenengineered to express epitopes againstanimal viruses or nanoscaffolds fornanotechnology.• In virus-induced gene silencing (VIGS),plant viral genomes are engineered tocontain sequences that would silence(through RNA silencing) target hostsequences and are of use in functionalgenomics.ReferencesAbel, P.P., Nelson, R.S., De, B., Hoffmann, N., Rogers, S.G.,Fraley, R.T., and Beachy, R.N. (1996). Delay of diseasedevelopment in transgenic plants that express the tobaccomosaic virus coat protein gene. Science 232, 738–743.Hull, R., Covey, S.N., and Dale, P. (2000). Genetically modifiedplants and the 35S promoter: Assessing the risksand enhancing the debate. Microb. Ecol. Health Dis. 12, 1–5.Sanford, J.C., and Johnston, S.A. (1985). The concept ofparasite-derived resistance. J. Theor. Bio. 113, 395–405.Further ReadingDietzgen, R.G. and Mitter, N. (2006). Transgenic genesilencing strategies for virus control. Austr. Plant Pathol.35, 605–618.Fuchs, M. (2008). Engineering resistance. Encyclopedia ofVirology, Vol. 4, 156–163.Gleba, Y., Marillonnet, S., and Klimyuk, V. (2008). Plantvirus vectors (gene expression systems). Encyclopedia ofVirology, Vol. 4, 229–237.Padmanabhan, M.S. and Dinesh-Kumar, S.P. (2008). Virusinducedgene silencing (VIGS). Encyclopedia of Virology,Vol. 5, 375–379.Rybicki, E.P. (2008). Vaccine production in plants. Encyclopediaof Virology, Vol. 5, 221–225.Steinmetz, N.F., Lomonossoff, G.P., and Evans, D.J. (2006).Cowpea mosaic virus for material fabrication: Addressablecarboxylate groups on a programmable nanoscaffold.Langmuir 22, 3488–3490.Young, M. Willits, D., Uchida, M., and Douglas, T. (2008).Plant viruses as biotemplates for materials and their usein nanotechnology. Annu. Rev. Phytopathol. 46, 361–384.IV. PLANT VIRUSES IN AGRICULTURE AND INDUSTRY

This page intentionally left blank

Appendix: ProfilesThese profiles give brief details of selected plant viruses that either have great economic significanceand/or have contributed to the understanding of viral functions and that are discussed inthis book. Where there are two or more species in the same taxonomic group, they are listed underthe appropriate family or genus so comparisons can be made within that group. These profiles areof only 32 of the more than 1,000 plant virus species. Further information on these viruses includesreviews, where available, and the Association of Applied Biologists Descriptions of Plant Viruses,which describes much of the biology of each virus (see www.dpvweb.net).INDEX1. Barley stripe mosaic virus2. Beet necrotic yellow vein virus3. Family Bromoviridae (Brome mosaic virus, Cucumber mosaic virus, Alfalfa mosaic virus)4. Family Caulimoviridae (Cauliflower mosaic virus, Banana streak virus, Rice tungro bacilliform virus)5. Genus Closterovirus (Citrus tristeza virus, Beet yellow virus)6. Family Comoviridae (Cowpea mosaic virus, Tobacco ringspot virus)7. Family Geminiviridae (Tomato yellow leaf curl virus, African cassava mosaic virus, Abutilon mosaicvirus, Bean golden mosaic virus, Cotton leaf curl disease, Maize streak virus)8. Family Luteoviridae (Barley yellow dwarf virus-PAV, Potato leafroll virus)9. Potato virus X10. Genus Potyvirus (Potato virus Y, Papaya ringspot virus, Plum pox virus, Tobacco etch virus)11. Rice ragged stunt virus12. Rice stripe virus13. Sonchus yellow net virus14. Tobacco mosaic virus15. Tobacco rattle virus16. Family Tombusviridae (Tomato bushy stunt virus, Carnation mottle virus)17. Tomato spotted wilt virus18. Turnip yellow mosaic virus303

304 PROFILESProfile No. 1BARLEY STRIPE MOSAIC VIRUS (BSMV)ClassificationType species of the Hordeivirus genus.Symptoms and Host RangeBSMV naturally infects barley (Hordeum vulgare) and wheat (Triticum aestivum). The symptomsin barley depend on strain and range, from a mild stripe mosaic (Fig. left panel; courtesy of A.O.Jackson) to lethal necrosis; crop losses can be 20% or more. (Fig. right panel shows infection of asusceptible variety compared with that of a resistant variety; courtesy of A.O. Jackson.) It has anartificial host range of more than 250 monocot and dicot species.StrainsThere are several strains causing different symptoms in barley, wheat, and oats.Geographic DistributionWorldwide

TransmissionPROFILE NO. 1.No natural vector known. BSMV is highly seed transmitted in barley and is pollen-borne. It ismechanically transmissible.ParticlesBSMV has tubular rod-shaped particles 22 nm indiameter and of 2–4 lengths (100–150 nm), dependingon strain.Genome and Genome Organisation305The infectious genome is divided between threespecies of positive sense ssRNA that are designateda, ß, and g. Their sizes vary between strains,with some containing sequence duplications. EachRNA has a 5’ cap, a 3’ tRNA-like structure, and apoly(A) sequence internal from the 3’ end.represents the 5’ cap, the 3’ t-RNA-like structure, An3’ shows a poly(A) sequence, methyltransferase domain, helicase domain, RdRp domain. RNAa encodes a single replicase subunitprotein (aa); RNAß encodes four proteins: ßa, the virus coat protein and TGB1, TGB2, and TGB3that for the triple gene cell-to-cell movement proteins; RNAg encodes two protein: ga, whichtogether with aa forms the complete replicase and gb, which affects the pathogenicity of the virus.TGB1, TGB2, TGB3, and gb are expressed from subgenomic RNAs.NotesInfection with BSMV can cause mutation in host plants.Further InformationAtabekov, J.G. and Novikov, V.K. (1989). Barley stripe mosaic virus. Association of Applied Biologists descriptions of plantviruses, No. 344.Bragg, J.N., Lim, H.-S., and Jackson, A.O. (2008). Hordeiviruses. Encyclopedia of Virology, Vol. 2, 459–466.

306 PROFILESClassificationProfile No. 2BEET NECROTIC YELLOW VEIN VIRUS (BNYVV)BNYVV is the type species of the genus Benyvirus.Symptoms and Host RangeBNYVV causes rhizomania disease of sugar beet (Beta vulgaris).The most characteristic symptom is root stunting andproliferation of the lateral roots (Fig). In early infections,plants are often stunted, and can wilt and die; in later infections,virus may reach the leaves, causing yellow and necroticvein symptoms. The host range is very narrow.[Figure courtesy of Association of Applied Biologists Descriptionsof Plant Viruses, No. 144.]StrainsThere are three major strain groups of BYNVV (A, B, andP-type) based on nucleic acid sequence and RNA composition.Geographic DistributionWorldwide in most countries in which sugar beet is grown.TransmissionBYNVV is transmitted by the plasmodiophorid vector, Polymyxa betae (see Chapter 12 for fungustransmission). It is mechanically transmissible to experimental hosts and is not seedtransmitted.ParticlesBNYVV is a multicomponent virus with rigidrod-shaped particles; predominant lengths of 390,265, 105, 90, and 80 nm; and diameters of 20 nm.Each segment of the multipartite genome is encapsidatedin separate particle by a single coat proteinspecies of 21 kDa. [Figure courtesy of M. Stephens.]

Genome and Genome OrganisationPROFILE NO. 2.The infectious genome comprises two species of positive sense ssRNA (RNA1 and 2), but in naturalinfections three other RNA species (RNA 3, 4, and 5) are found that contribute to the biology ofthe virus. Each of the RNAs has a 5’ cap and is polyadenylated at the 3’ end.307*5’ cap; methyl transferase; protease; helicase; RdRp; the 3’ Poly-A; ¼ read-through.RNA1 encodes a single polypeptide of 237 kDa that is the replicase; RNA2 has 6 ORFs: ORF2Ais the 21 kDa coat protein that reads through into ORF2ART, the read-through protein beinginvolved in virus assembly and fungus transmission, ORFs 2B, 2C, and 2D form the triple blockinvolved in cell-to-cell movement, ORF2E encodes a 14 kDa protein, which is a silencing suppresser;RNA3 has three ORFs, and RNAs 4 and 5 each contain one ORF; ORFs 2B, 2C, 2D, 2E, and thetwo 3’ ORFs in RNA3 are expressed from subgenomic RNAs.NotesRhizomania is the most severe virus disease of sugar beet, limiting production in many areas.Further InformationSchirmer, A., Link, D., Cogmat, V., Moury, B., Beuve, M., Meunier, A., Bragard, C., Gilmer, D., and Lemaire, O. (2005). Phylogeneticanalysis of isolates of Beet necrotic yellow vein virus collected worldwide. J. Gen. Virol. 86, 2897–2911.Stevens, M., Lui, H.-Y., and Lemaire, O. (2006). The viruses, in The sugar beet book (A.P. Draycott, Ed.), Blackwells Publishing,Oxford, UK, pp. 256–285.Tamada, T. (2002). Beet necrotic yellow vein virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 144.

308 PROFILESBrome mosaic virus (BMV)ClassificationProfile No. 3FAMILY BROMOVIRIDAEGENUS BROMOVIRUSBMV is the type species of the genus Bromovirus in the family Bromoviridae.Symptoms and Host RangeCauses mosaic and brown streaks in leaves (Fig.). Natural host range limited to familyPoaceae; experimental host range includes Nicotiana benthamiana and several Chenopodiumspp., in which it produces local lesions. BMV will also replicate in yeast,Saccharomyces cerevisiae. [Figure shows symptoms of BMV in barley; courtesy of Wooleyand Kao (2004), Descriptions of Plant Viruses, No. 405.]StrainsNo strains reported.Geographic DistributionWorldwideTransmissionNatural transmission not determined; Xiphenema nematodes shown to transmitexperimentally.ParticlesIsometric 27 nm diameter, T ¼ 3 symmetry. Threeparticles components, one containing a molecule ofRNA1, one containing a molecule of RNA2, and onecontaining one molecule each of RNA3 and RNA4.[Figure shows surface structure of BMV particle; thisarticle was published in Encyclopedia of Virology,Bujarsky (B.W.G. Mahy and M. van Regenmortel, Eds.),Bromoviruses, No. 638. Copyright Elsevier (2008).

Genome and Genome OrganisationPROFILE NO. 3.The infectious genome is divided between three species of positive sense ssRNA; RNA1, 3.23 kb;RNA2, 2.87 kb; RNA3, 2.12 kb; the subgenomic RNA4 (1.2 kb) encoding coat protein is alsoencapsidated. The 5’ end of each RNA is capped, the 3’ end has a t-RNA-like structure (see Chapter 6).309represents the 5’ cap, the 3’ t-RNA-like structure, methyl transferase domain, helicasedomain, RdRp domain. RNAs 1 and 2 encode the replicase; RNA3 is bi-cistronic encoding thecell-to-cellmovement protein and the coat protein that is expressed by subgenomic RNA4.NotesBMV has been extensively studied as a model for RNA replication, gene expression, and virionassembly of positive-strand RNA viruses.Further InformationBujarski, J.J. (2008). Bromoviruses. Encyclopedia of Virology, original 638.Kao, C.C. and Sivakumaran, K. (2000). Brome mosaic virus, good for an RNA virologist’s basic needs. Molec. Plant Pathol. 1,91–97.Wang, X. and Ahlquist, P. (2008). Brome mosaic virus. Encyclopedia of Virology, Vol. 1, 381–385.Wooley, R.S. and Kao, C.C. (2004). Brome mosaic virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 405.Cucumber mosaic virus (CMV)GENUS CUCUMOVIRUSClassificationType member of genus Cucumovirus in family Bromoviridae.

310 PROFILESSymptoms and Host RangeCMV in zucchini CMV distortion of zucchini CMV, causing fernleaf intomato (from www.avdrc.org)Causes mosaic symptoms in leaves of many dicot species and chlorotic streaks in monocot species,deformation of leaves (fernleaf in tomato) and of fruits, and sometimes severe necrosis. Verybroad host range including >1,300 species in >100 monocot and dicot plant families; isolated fromnatural infections of >500 species.StrainsSeveral strains differing in symptoms, host range, transmission, serology, and nucleotidesequence. Nucleotide sequence and serology give two subgroups, I and II; subgroup I divided intoIA and IB on nucleotide sequence.Geographic DistributionCMV is found worldwide.TransmissionTransmitted by >80 spp. of aphid (in 33 genera) in the nonpersistent manner (see Chapter 12).Mechanically transmissible and also seed transmitted in some plant spp.

ParticlesPROFILE NO. 3.Isometric 28–30 nm diameter, T ¼ 3 symmetry.Three particles components, one containing a moleculeof RNA1, one containing a molecule of RNA2,and one containing one molecule each of RNA3and RNA4. The Figure shows particles of CMV negativelystained in uranyl acetate, bar marker ¼ 100nm. [Courtesy of Palukaitis and Garcia-Arenal(2003), Descriptions of Plant Viruses, No. 400.]311Genome and Genome OrganisationThe infectious genome is divided between threespecies of positive sense ssRNA; RNA1, 3.36 kb;RNA2, 3.05 kb; RNA3, 2.16 kb; the subgenomicRNA4, (1.03 kb) encoding coat protein is alsoencapsidated. The 5’ end of each RNA is capped,the 3’ end has a t-RNA structure (see Chapter 6).represents the 5’ cap, the 3’ t-RNA structure, methyl transferase domain, helicase domain,RdRp domain. RNAs 1 and 2 encode the replicase; RNA3 is bi-cistronic encoding the cell-to-cellmovement protein and the coat protein that is expressed by subgenomic RNA4.NotesCMV is an economically important virus worldwide. Combined with a satellite RNA, it causesa severe necrotic disease of tomato (see Box 3.3).Further InformationGarcia-Arenal, F. and Palukaitis, P. (2008). Cucumber mosaic virus. Encyclopedia of Virology, Vol. 1, 614–619.Palukaitis, P. and Garcia-Arenal, F. (2003). Cucumber mosaic virus. Association of Applied Biologists Descriptions of PlantViruses, No. 400.

312 PROFILESAlfalfa mosaic virus (AMV)GENUS ALFAMOVIRUSClassificationAMV is the type and only member of the genus Alfamovirus in the family Bromoviridae.Symptoms and Host RangeAMV infections are economically important inalfalfa (Medicago sativa), where they can causemosaics and mottles; often infection is symptomlessbut can render the plants susceptible to frost. It isfound naturally in more than 150 species. The experimentalhost range is more than 600 species. Intobacco (Nicotiana tabacum) it causes necrotic andchlorotic local lesions and a systemic mottle (Figure).StrainsThere are numerous strains of AMV that differ inthe symptoms that they produce.Geographic DistributionWorldwideTransmissionAMV is transmitted by aphids in the nonpersistent manner. It is also transmitted mechanicallyand by seed.ParticlesThere are four major classes of particles calledbottom component (B), middle component (M),top component b (Tb), and top component a (Ta).B, M, and Tb are bacilliform and contain genomicRNAs 1, 2, and 3, respectively; Ta component containstwo molecules of subgenomic RNA4 and hastwo forms: bacilliform Ta-b and spheroidal Ta-t particles.The bacilliform particles (Fig) are made up ofa single coat protein species (24.3 kDa) and are 19nm wide and have lengths of 56 nm (B), 43 nm(M), 30 nm (Tb), and 30 nm (Ta-b). The Figureshows a preparation of AMV negatively stainedin uranyl acetate; bar marker is 50 nm. (See alsoFigure 5.8.)

Genome and Genome OrganisationPROFILE NO. 3.The infectious genome is divided between three species of positive sense ssRNA; RNA1, 3.64kb; RNA2, 2.59 kb; RNA3, 2.14 kb; the subgenomic RNA4 (0.88 kb) encoding coat protein isalso encapsidated. The 5’ end of each RNA is capped, the 3’ end has a t-RNA-like structure (seeChapter 6).313represents the 5’ cap, the 3’ t-RNA-like structure, methyl transferase domain, helicasedomain, RdRp domain. RNAs 1 and 2 encode the replicase; RNA3 is bi-cistronic encoding thecell-to-cell movement protein and the coat protein that is expressed by subgenomic RNA4.NotesAMV requires its coat protein for replication. This is supplied either by the capsid of the incomingvirus or by the presence of RNA4 in RNA preparations.Further InformationBol, J.F. (2003). Alfalfa mosaic virus: Coat protein-dependent initiation of infection. Molec. Plant Pathol. 4, 1–8.Bol, J.F. (2008). Alfalfa mosaic virus. Encyclopedia of Virology, Vol. 1, 81–86.Jaspars, E.M.J. and Bos, L. (1980). Alfalfa mosaic virus. Association of Applied Biologists Description of Plant Viruses, No. 229.

314 PROFILESProfile No. 4FAMILY CAULIMOVIRIDAEGENUS CAULIMOVIRUSCauliflower mosaic virus (CaMV)ClassificationType species of the genus Caulimovirus in the family Caulimoviridae.Symptoms and Host RangeCaMV in cauliflower Mosaic in Symptoms of CaMV strains in turnipcauliflower leaf leaves; healthy in centre.CaMV infection causes a mosaic and mottle disease in many Cruciferae. The central pictureshows chlorotic vein-banding. The natural host range is limited to the family Cruciferae; somestrains also infect Nicotiana clevelandii and Datura stramonium on artificial infection.StrainsCaMV is a variable virus with many strains that vary in virulence from very severe to very mild(Figure right hand panel).Geographic DistributionWorldwide

TransmissionCaMV is transmitted by aphids in the nonpersistent manner using a virus-encoded helper protein(see Chapter 12). The virus is mechanically transmissible but is not seed transmitted.ParticlesPROFILE NO. 4.315CaMV particles negatively stained in Reconstruction of CaMV surface structure.ammonium molybdate. Bar marker ¼ 50 nm. From Cheng et al (1992; Virology 186, 655–668).The virus has isometric particles (T ¼ 7) 50 nm in diameter (see Chapter 5). The Figure panelsshow negatively stained CaMV preparation and the T ¼ 7 structure.Genome and Genome OrganisationEach particle contains a single circular molecule of dsDNA of about 8 kbp. Most strains havethree discontinuities in the DNA, two in one strand and one in the other; one strain has onlyone in each strand. Replication is by reverse transcription (see Chapter 8).

316 PROFILESFurther InformationThe double complete circles represent thedsDNA genome with the positions of the discontinuitiesmarked D1, D2, and D3. The boxedarrows show the positions of the promoters.Outside the double circles the positions of the35S and 19S transcripts are shown by the arrowheadmarking the 3’ ends. Inside the double circlethe arced boxes represent ORFs. ORFproduct functions: 1, cell-to-cell movement; 2,aphid transmission; 3, DNA-binding protein; 4,coat protein; 5, protease, reverse transcriptase,RNaseH; 6, translational transactivator; 7unknown.NotesCaMV was the first plant virus shown to replicateby reverse transcription. The 35S promoteris widely used in constructs for planttransformation.Haas, M., Bureau, M., Geldereich, A., Yot, P., and Keller, M. (2002). Cauliflower mosaic virus: Still in the news. Molec. PlantPathol. 2, 419–430.Hohn, T. (2008). [Caulimoviridae] Molecular biology. Encyclopedia of Virology, Vol. 1, 464–468.Schoelz, J.E. (2008). [Caulimoviridae] General features. Encyclopedia of Virology, Vol. 1, 457–463.Shepherd, R.J. (1981). Cauliflower mosaic virus. Association of Applied Biologists Descriptions of Plant Viruses, No.243.Banana streak virus (BSV)GENUS BADNAVIRUSClassificationA species in the genus Badnavirus in the family Caulimoviridae.Symptoms and Host RangeCauses chlorotic streaks in leaves, which becomenecrotic with time; the expression of these leafsymptoms can often be periodic. Also may causestunting of the plant, constriction of the emergingbunch, and detachment and splitting of outersheaths of pseudostem. Host range limited toMusaceae.

StrainsVery variable with several viruses (BSV-Mys, BSV-GF, BSV-IM, BSV-OL) causing similarsymptoms.Geographic DistributionBSV probably occurs wherever bananas are grown.TransmissionTransmitted by mealybugs, most probably in a semipersistent manner. Transmitted in cultivationmainly by planting infected suckers. The virus is not mechanically transmissible.ParticlesPROFILE NO. 4.Bacilliform particles 30 nm in diameter and 130–150 nm in length; longer particles are found withsome strains. The figure shows a BSV preparationnegatively stained in uranyl acetate; bar marker ¼100 nm.317Genome and Genome OrganisationdsDNA circular molecule of about 7.4 kbpwith one discontinuity in each strand. Replicatedby reverse transcription (see Chapter 8).Viral sequences integrate into the host genomeand in some cases can be activated to giveepisomal infection (see Box 8.9).The double complete circles represent thedsDNA genome. Outside the double circles thearced boxes represent ORFs. ORF product functions:1, unknown; 2, unknown; 3, a polyproteincleaved to give the cell-to-cell movement protein,the coat protein, the aspartate proteaseand the replicase comprising reverse transcriptaseand RNase H.

318 PROFILESNotesBSV was the first plant virus shown to have sequences integrated in the host genome that areactivatable to give episomal infection.Further InformationGeering, A.D.W. and Thomas, J.E. (2002). Banana streak virus. Association of Applied Biologists Descriptions of Plant Viruses,No. 390.Harper, G., Hull, R., Lockhart, B., and Olszewski, N. (2002). Viral sequences integrated into plant genomes. Annu. Rev.Phytopathol. 40, 119–136.Rice tungro bacilliform virus (RTBV)ClassificationGENUS TUNGROVIRUSType species of the genus Tungrovirus in the family Caulimoviridae.Symptoms and Host RangeCauses stunting of the plant, red or orange colouring of the leaves,and reduction in tiller number. When complexed with Rice tungro sphericalvirus (RTSV) the symptoms are more severe. It has a limited hostrange mainly in the Poaceae.StrainsThe genomes of RTBV isolates from Southeast Asia differ from thosefrom the Indian subcontinent in that the latter have a deletion in a noncodingregion. There are strains of the Southeast Asian type that differin response to resistance genes in rice.Geographic DistributionRTBV occurs in South East and East Asia and in the Indian subcontinent(see Figure 4.4).TransmissionTransmitted by leafhoppers only when RTSV has been acquired previously or at the same time.Transmission is in a semipersistent manner by several leafhopper species, the most importantbeing the rice green leafhopper, Nephotetix virescens.

ParticlesBacilliform particles 30 nm in diameter and 130–150 nm in length very similar to those ofBanana streak virus shown previously.Genome and Genome OrganisationdsDNA circular molecule of about 8.0 kbpwith one discontinuity in each strand. Replicatedby reverse transcription (see Chapter 8).The double complete circles represent thedsDNA genome with the discontinuities markedD1 and D2. Inside the double circles the arcedboxes represent ORFs. ORF product functions: 1,(P24) unknown; 2, (P12) unknown; 3, (P194) a polyproteincleaved to give the cell-to-cell movementprotein, the coat protein, the aspartate proteaseand the replicase comprising reverse transcriptaseand RNase H; 4, (P48) the transactivator protein.Outside the double circles, the positions of the35S RNA transcript and the splices mRNA forORF4 are shown, with arrowheads marking the3’ ends.NotesRTBV complexes with RTSV to give the verysevere tungro disease of rice. RTBV has various interesting molecular features such as the non-AUG start codon for ORF1 and the transactivator properties of the product of ORF4 that aredescribed in this book.Further InformationPROFILE NO. 4.Hull, R. (2004). Rice tungro bacilliform virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 406.319

320 PROFILESProfile No. 5GENUS CLOSTEROVIRUSCITRUS TRISTEZA VIRUS (CTV)ClassificationA member of the genus Closterovirus in the family Closteroviridae.Symptoms and Host RangeThe severity of tristeza or quick decline depends on the virus strain, the citrus species, and the rootstockonto which it has been grafted; it is important on commercial varieties of citrus on sour orange(C. aurantium) rootstocks. Severe strains cause rapid wilting of the scion (see left-hand Figure), stempitting of the trunk, and honeycombing immediately below the bud union; other milder symptomsinclude vein yellowing (right-hand Figure). The host range is mainly limited to the plant family Rutaceae.[Figures from Moreno et al. (2008; Molec. Plant Pathol. 9, 251–268).]StrainsCTV has many strains that show a wide variety of symptoms.Geographic DistributionWorldwide wherever citrus is grown.

TransmissionThe virus is transmitted by aphids (mainly Toxoptera citricida and Aphis gossypii) in semipersistentmanner.ParticlesPROFILE NO. 5.Very flexuous rod-shaped particles about2,000 nm in length and 10 nm in diameter. Figure.Bar ¼ 100 nm [This figure was published inVirus Taxonomy, 7th Report of the International Committeeon the Taxonomy of Viruses, G.P. Martelli,A.A. Agranovsky, M. Bar-Joseph et al. (M.H.V.van Regenmortel, C.M. Fauquet, D.H.L. Bishopet al., Eds.), Closteroviridae, p. 943–964, CopyrightElsevier Academic Press, San Diego (2000).]321Genome and Genome OrganisationThe genome of CTV is (þ)-sense ssRNA of 19.3 kb. It has a 5’ cap and a 3’ hydroxyl group andcontains 12 ORFs.represents the 5’ cap, 3’OH, the 3’ hydroxyl group, methyl transferase domain, helicase domain,RdRp domain. The viral replicase is expressed from ORFs 1a and 1b, 1b reading through from 1a;ORF2 encodes a 33 kDa protein of unknown function; ORF3, a 6 kDa hydrophobic protein; ORF4 a65 kDa homologue of heat shock protein HSP70; ORF5, a 61 kDa protein involved in virion assembly;ORF6, the 27 kDa modified coat protein; ORF7, the 25 kDa coat protein; ORF8, an 18 kDa proteinand ORF9, a 13 kDa protein, both of unknown function; ORF10, a 10 kDa protein and ORF11, a 23 kDaprotein, both suppressors of silencing. ORFs 2–11 areexpressedfromsubgenomicRNAs(seeFigure7.5).NotesCTV is the most important virus of citrus and can be a serious constraint to production. It hasthe largest genome of any plant virus.Further InformationBar-Joseph, M. and Lee, R.F. (1989). Citrus tristeza virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 353.Bar-Joseph, M. and Dawson, W.O. (2008). Citrus tristeza virus. Encyclopedia of Virology, Vol. 1, 520–524.Dolja, V.V., Kreuze, J. F., and Valkonen, J.P.T. (2006). Comparative and functional genomics of closteroviruses. Virus Res.117, 38–51.Moreno, P., Ambrós, S., Albiach-Marti, M.R., Guerri, J., and Pena, L. (2008). Citrus tristeza virus: A pathogen that changedthe course of the citrus industry. Molec. Plant Pathol. 9, 251–268.

322 PROFILESBEET YELLOWS VIRUS (BYV)ClassificationBYV is the type member of the genus Closterovirus in the family Closteroviridae.Symptoms and Host RangeBYV causes vein clearing and vein yellowing symptoms (see right-hand Figure) in the youngleaves of infected plants, leading to yellowing and thickening (left-hand Figure) of older leaves.It has a moderate host range infecting more than 120 species.StrainsThere are several strains of BYV differing in the symptoms in beet and other hosts.Geographic DistributionWorldwide in sugar beet growing areas.TransmissionThe virus is transmitted by aphids (mainly Myzus persicae and Aphis fabae) in semipersistentmanner. It is mechanically transmissible but not seed transmitted.

ParticlesAs with CTV very flexuous rod-shaped particlesabout 1,250–1,450 nm long and 19 nm indiameter. They are composed of the genomicRNA and two species of coat protein, a majorone (p22) and a minor one (p24) associatedwith aphid transmission (see Figure 5.1).[Figure courtesy of M. Stevens.]PROFILE NO. 5.323Genome and Genome OrganisationThe genome of BYV is (þ)-sense ssRNA of 15.5 kb. It has a 5’ cap and a 3’ hydroxyl group andcontains 9 ORFs. represents the 5’ cap, 3’OH the 3’ hydroxy, methyl transferase domain,helicase domain, RdRp domain.ORFs 1a and 1b encode the replication enzymes; the two coat proteins are expressed from ORFsCp and CPd; the 6 K, HSP70, and 64 K are involved in cell-to-cell movement together with the twocoat proteins; the 20 K and 21 K products are thought to act as silencing suppressors.NotesBYV is one of the major viruses of sugar beet.Further InformationAgronovsky, A.A. and Lesemann, D.E. (2000). Beet yellows virus. Association of Applied Biologists Descriptions of PlantViruses, No 377.Dolja, V.V., Kreuze, J.F., and Valkonen, J.P.T. (2006). Comparative and functional genomics of closteroviruses. Virus Research117, 38–51.Peremyslov, V.V., Andreev, I.A., Prokhnevsky, A.I., Duncan, G.H., Taliansky, M.E., and Dolja, V.V. (2004). Complex moleculararchitecture of beet yellow virus particles. Proc. Natl. Acad. Sci. 101, 5030–5035.

324 PROFILESCowpea mosaic virus (CPMV)ClassificationProfile No. 6FAMILY COMOVIRIDAEGENUS COMOVIRUSType species of genus Comovirus in family Comoviridae.Symptoms and Host RangeCauses a mosaic, decrease in leaf area, and reduced flower production; yield loss of up to 75%.Limited host range mainly in family Leguminosae; experimental infection of Chenopodium amaranticolorgiving local lesions.StrainsSeveral strains differing mainly in symptoms have been described.Geographic DistributionNigeria, Tanzania, Japan, Surinam, Cuba, and the United States.TransmissionTransmitted by beetles, especially members of family Chrysomelidae (see Chapter 12). Mechanicallytransmissible; seed transmitted in cowpea.

ParticlesPROFILE NO. 6.325Purified preparation of CPMV negativelystained in 2% phosphotungstic acid, pH 6.8;bar ¼ 100 nm. [Courtesy of DescriptionsPlant Viruses, No. 378.]Model showing structure of CPMV.Source www.primidi.com/2006/03/20.htmlIsometric particles, 28–30 nm in diameter (left-hand picture) with T ¼ 1 (pseudo T ¼ 3) symmetry(right-hand picture) (See Chapter 5). Particles composed of two coat protein species. Three sedimentingcomponents, B containing B-RNA; M, M-RNA; and T, no RNA (empty) that show in electronmicrograph.Genome and Genome OrganisationInfectious genome divided between two molecules of positive-sense ssRNA with 5’ VPg and 3’Poly-A (see Chapter 6); M-RNA, 5.90 kb; B-RNA, 3.48 kb.represents the 5’ VPg, the 3’ Poly-A, the RdRp motif, the helicase motif. Each of thetwo RNAs is expressed as a polyprotein that is processed to give: B-RNA; a 32 K and 24 K protease,a 58 K nucleotide-binding protein, the VPg and the 78K replicase protein; M-RNA; the 58/48 Kcell-to-cell movement protein and the 37 K and 23 K coat proteins.NotesThe coat protein of CPMV is modified to make vaccines against animal viruses (see Chapter 15).

326 PROFILESFurther InformationPouwels, J., Carette, J.E., van Lent, J., and Wellink, J. (2002). Cowpea mosaic virus: Effects on host cell processes. Molec. PlantPathol. 2, 411–418.Lomonossoff, G. P. Cowpea mosaic virus. Encyclopedia of Virology, Vol. 1, 569–573.Van Kammen, A., van Lent, J., and Wellink, J. (2001). Cowpea mosaic virus. Association of Applied Biologists Descriptions ofPlant Viruses, No. 378.Tobacco ringspot virus (TRSV)GENUS NEPOVIRUSClassificationTRSV is the type member of the genus Nepovirus in the family Comoviridae. It belongs to subgroupA of the genus.Symptoms and Host RangeThe virus causes ring and line patterns in theleaves of tobacco (Nicotiana tabacum) (Fig.). It alsocauses a severe disease, bud blight of soybean inwhich the buds become brown and brittle, andnecrotic streaking appears on the stem. TRSV hasa wide experimental host range.StrainsThere are many variants of TRSV distinguishable on symptoms.Geographic DistributionThe virus is endemic in central and eastern North America and is found in crops (especiallyperennial plants) in various other places worldwide.TransmissionTRSV is transmitted by nematodes, Xiphenema spp. (see Chapter 12 for nematode transmission). Itis also mechanically and seed transmitted.

ParticlesPROFILE NO. 6.327The particles are isometric, 28 nm in diameter, T ¼1 (pseudo T ¼ 3) with the two genomic segmentsencapsidated separately in a capsid of 60 subunitsof the 57 kDa coat protein. Figure shows preparationof TRSV negatively stained in uranyl acetate; barmarker ¼ 100 nm. [Courtesy of Descriptions of PlantViruses, No. 309.]Genome and Genome OrganisationThe infectious genome is divided between two species of positive-sense ssRNA (RNA1 and 2),the 5’ end of each having a VPg and the 3’ end being polyadenylated.The diagram shows the genome organisation of Beet ringspot virus as a representative ofsubgroup A nepoviruses. represents the 5’ VPg, the 3’ Poly-A, the RdRp motif, the helicasemotif. Each of the two RNAs is expressed as a polyprotein that is processed to give: B-RNA;P1A and P1C are proteases, P1B is a nucleotide-binding protein, the VPg and P1E the replicaseprotein; M-RNA; P2A is the cell-to-cell movement protein and P2B the coat protein. The processingpathway is similar to that of CPMV/NotesThere are three subgroups of nepoviruses (A, B, C), which differ in genome organisation. Nepovirusesoften have satellite RNAs associated with them (see Chapter 3).Further InformationSanfaçon, H. (2008) Nepovirus. Encyclopedia of Virology, Vol. 3, 405–412.Stace-Smith, R. (1985). Tobacco ringspot virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 309.

328 PROFILESTomato yellow leaf curl virus (TYLCV)ClassificationProfile No. 7FAMILY GEMINIVIRIDAEGENUS BEGOMOVIRUSA species in the genus Begomovirus in the family Geminiviridae. TYLCV is a single componentbegomovirus (see ACMV following).Symptoms and Host RangeInfected plants are stunted with small chloroticleaves that have upward curling of the margins[Figure courtesy of Gafni (2003), Molec. PlantPathol. 4, 9–15.] The virus has a wide natural hostrange including Phaseolus vulgaris (common bean)and petunia.StrainsThere are numerous variants of this virus that, depending on sequence differences, are calledstrains of TYLCV (e.g., Israel, Iran, etc.) or different viruses (e.g., Tomato yellow leaf curl Sardiniavirus, Tomato yellow leaf curl Sudan virus).Geographic DistributionWorldwideTransmissionTransmitted in a circulative nonpropagative manner by the whitefly Bemisia tabaci (see Chapter12). Not transmitted by seed or mechanically.

ParticlesPROFILE NO. 7.329The particles are geminate, 20 30 nm. TheFigure shows negatively stained particles ofTYLCV; bar ¼ 100 nm. [Courtesy of Gafni (2003),Molec. Plant Pathol. 4, 9–15.]Genome and Genome OrganisationThe genome is a covalently closed circle ofssDNA, 2.79 kb. The diagram shows the genomeorganisation; the circle represents the ssDNA;the black box the common region; the arcedboxes within the circle are the ORFs, the functionbeing: VI, the coat protein; V2, cell-tocellmovement protein; C1, the Rep protein; C2,the transcriptional activator protein; C3, the replicationenhancer protein; C4, unknownfunction.NotesTYLCV is widespread and causes a severedisease of tomatoes. There is much recombinationbetween its variants giving rise to newstrains and viruses (see Chapter 4). Some otherbegomoviruses can cause similar symptoms intomato.Further InformationCzosnek, H. (1999). Tomato yellow leaf curl virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 368.Czosnek, H. (2008). Tomato yellow leaf curl virus. Encyclopedia of Virology, Vol. 5, 138–144.Gafni, Y. (2003). Tomato yellow leaf curl virus, the intracellular dynamics of a plant DNA virus. Molec. Plant Pathol. 4, 9–15.

330 PROFILESClassificationAFRICAN CASSAVA MOSAIC VIRUS (ACMV)A species in the genus Begomovirus in the family Geminiviridae. ACMV is a two componentbegomovirus. (See TYLCV previous)Symptoms and Host RangeInfection causes a severe mosaic and a 60 to 80%loss of yield. In nature the virus is only found incassava; the experimental host range is narrowand includes several solanaceous species, such asNicotiana spp.StrainsThere are several strains of ACMV differentiated on severity of symptoms.Geographic DistributionACMV is found in many sub-Saharan African countries.TransmissionTransmitted in a circulative nonpropagative manner by the whitefly Bemisia tabaci (see Chapter 12).ParticlesThe particles are geminate, 20 30 nm, and are similar to those shown above for TYLCU.

PROFILE NO. 7.331Genome and Genome OrganisationThe genome is divided between two moleculesof circular ssDNA of about 2.78 kb(DNA A) and 2.72 kb (DNA B). The diagramshows the genome organisation; the circlesrepresent the ssDNAs of components A andB; the black boxes, the common regions. Thearced boxes within the circles are the ORFs,the functions being: AV1, the coat protein;AV2, cell-to-cell movement protein; AC1, theRep protein; AC2, the transcriptional activatorprotein; AC3, the replication enhancer protein;AC4, unknown function; BV1, thenuclear shuttle protein; BC1, cell-to-cellmovement.AC1AC4AC2AAV2AV1NotesACMV is a major constraint to cassava cultivationin Africa. Interactions between it andclosely related viruses have caused verysevere epidemics (see Box 11.3).AC3BC1BBV1Further InformationBock, K.R. and Harrison, B.D. (1985). African cassava mosaic virus. Association of Applied Biologists Descriptions of PlantViruses, No. 297.Fargette, D., Konate, G., Fauquet, C., Muller, E., Peterschmitt, M., and Thresh, J.M. (2006). Molecular ecology and emergenceof tropical plant viruses. Annu. Rev. Phytopathol. 44, 235–260.

332 PROFILESSome Features of Other BegomovirusesABUTILON MOSAIC VIRUS (ABMV)A two-component begomovirus with a genome organisation similar to that of ACMV. It hascharacteristic symptoms as shown by those caused by a WestIndian isolate in Abutilon sellovianum var marmorata. [Figurecourtesy of Descriptions of plant viruses, No. 373.] Infected abutilonsare important as an ornamental plant.Further InformationJeske, H. (2000). Abutilon mosaic virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 373.BEAN GOLDEN MOSAIC VIRUS (BGMV)Type species of the genus Begomovirus. An important virus of beans (Phaseolus vulgaris) in Southand Central America causing a severe yellow or golden mottle of the leaves. It is a two-componentbegomovirus with a genome organisation similar to that of ACMV. It is transmitted by the whitefly,Bemisia tabaci.Further InformationGoodman, R.M. and Bird, J. Bean golden mosaic virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 192.Morales, F.J. (2008). Bean golden mosaic virus. Encyclopedia of Virology, Vol. 1, 295–300.COTTON LEAF CURL DISEASEAn important disease of cotton caused by a monopartite begomovirus and a satellite DNA ßand satellite DNA-1 (see Chapter 3 for DNA satellites). The disease causes severe losses in cottonin North Africa and the Indian subcontinent. The complex is transmitted by the whitefly, Bemisiatabaci. The genome organisation of the begomovirus component resembles that of TYLCV. On theIndian subcontinent, seven distinct begomovirus species have been associated with the disease.

PROFILE NO. 7.333Further InformationMansoor, S., Amin, I. and Briddon, R.W. (2008). Cotton leaf curl disease. Encyclopedia of Virology, Vol. 1, 563–568.Maize streak virus (MSV)GENUS MASTREVIRUSClassificationType species of the genus Mastrevirus in the family Geminiviridae.Symptoms and Host RangeThe first symptoms on a susceptible maize cultivar are circularpale spots. These develop into chlorotic veinal streaks, 0.5 to1.0 mm wide and a few mm to several cm long. The host rangeis restricted to the Graminae but it can infect >80 grass species.StrainsThere are nine major strains (MSV-A to MSV-I) that vary in severity of symptoms in maize; onlyMSV-A produces a severe disease in maize.Geographic DistributionMSV is found in sub-Saharan Africa and in the Indian Ocean islands of Madagascar, Mauritius,and La Réunion.

334 PROFILESTransmissionThe virus is transmitted by cicadellid leafhoppers in a persistent nonpropagative manner (seeChapter 12). The most important vector is Cicadulina mbila. Not transmitted mechanically or by seed.ParticlesGeminate particles 22 38 nm resembling those of begomoviruses.Genome and Genome OrganisationThe genome is a covalently closed circle ofssDNA, 2.6–2.8 kb.The circle represents the ssDNA; the black boxthe common region. The arced boxes within the circleshow the ORFs, the functions being: V1, cell-tocellmovement; V2, coat protein; C1 protein regulatesthe host DNA replication ability; C2 and C1:C2 (fusion protein) are replication associated proteins(see Chapter 8).NotesMSV is a major constraint to maize production in many African countries.Further InformationMartin, D.P. (2007). Maize streak virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 416.Martin, D.P., Shepherd, D.N., and Rybicki, E.P. (2008). Maize streak virus. Encyclopedia of Virology, Vol. 3, 263–271.

PROFILE NO. 8.Profile No. 8FAMILY LUTEOVIRIDAEGENUS LUTEOVIRUSBarley yellow dwarf virus—PAV (BYDV-PAV)335ClassificationType species of the genus Luteovirus in the family Luteoviridae.Symptoms and Host RangeStunting of plants and chlorosis or red/purple colour of leaves.Moderate natural host range limited to the family Poaceae causingyield losses of up to 30% in some cereal crops; also will infect somesolanaceous species on artificial inoculation. (Figure from www.ag.nsdu.edu.)StrainsSeveral variants of the virus that are classified as distinct species based mainly on their vector—for example, BYDV-PAV transmitted mainly by Rhopalosiphon padi and Macrosiphon avenae andBYDV-MAV, mainly by M. avenae.Geographic DistributionWorldwideTransmissionTransmitted by aphids in the persistent, circulative, nonpropagative manner (see Chapter 12). Notmechanically or seed transmitted.

336 PROFILESParticlesIsometric particles 25–30 nm in diameter with T ¼ 3 symmetry(Figure; Bar ¼ 20 nm). [Courtesy of Association of Applied BiologistsDescriptions of Plant Viruses, No. 32.] Two proteins involvedin the capsid (see Chapter 12).Genome and Genome OrganisationThe genome is a single molecule of positive-sense ssRNA, 5.68 kb, the 3’ end of which hasa hydroxyl group, (the genomes of other genera in the family Luteoviridae have a 5’ VPg; seeChapter 6).¼ RdRp motif, ¼ read-through, ¼ 3’ hydroxyl group. The genome encodes 6 ORFs thefunctions being: ORFs 1 and 2 viral replicase via a ribosomal frameshift (see Chapter 7); ORF3,major coat protein that forms a minor coat protein by read-through into ORF5 (see Chapter 7);ORF4, cell-to-cell movement; ORF6, function unknown; there are some minor ORFs within themajor ones that may regulate transcription late in replication.NotesThe luteoviruses cause serious losses to small grain cereal crops.Further InformationDomier, L.L. (2008). Barley yellow dwarf viruses. Encyclopedia of Virology, Vol. 1, 279–285.Domier, L.L. and D’Arcy, C.J.D. (2008). Luteoviruses. Encyclopedia of Virology, Vol. 3, 231–237.Miller, W.A., Liu, S. and Beckett, R. (2002). Barley yellow dwarf virus: Luteoviridae or Tombusviridae. Molec. Plant Pathol. 3,177–184.Rochow, W.F. (1970). Barley yellow dwarf virus. Association of Applied Biologists Description of Plant Viruses, No. 32.

Potato leafroll virus (PLRV)PROFILE NO. 8.GENUS POLEROVIRUS337ClassificationType species of the genus Polerovirus in the family Luteoviridae.Symptoms and Host RangePrimary infections of potato (Solanum tuberosum) cause pallor of tip leaves, which may becomerolled and erect. Secondary symptoms in plants grown from infected tubers are stunting and upwardrolling of leaflets, especially those on the lower leaves. Left-hand panel of Figure shows infected plantsurrounded by symptomless plants; right-hand panel is enlargement of infected plant showing leafrolling. (Source: www.apsnet.org) Causes considerable crop losses. Limited host range of about 20species.StrainsSeveral strains have been distinguished on symptom severity in potato and ease of aphidtransmission.Geographic DistributionWorldwide where potatoes are grown.TransmissionTransmitted by aphids in the persistent, circulative, nonpropagative manner (see Chapter 12). Notmechanically or seed transmitted.

338 PROFILESParticlesIsometric particles 25–30 nm in diameter with T ¼ 3 symmetry, similar to those for BYDV-PAV.Two proteins involved in the capsid (see Chapter 12).Genome and Genome OrganisationThe genome is a single molecule of positive-sense ssRNA, 5.68 kb, the 3’ end of which has aVPg.represents the 5’ VPg, thehelicasemotif. ¼ RdRp motif, ¼ read-through, ¼ 3’ hydroxylgroup. The genome encodes 6 major ORFs the functions being: ORF0, suppressor of RNA silencing;ORFs 1 and 2, viral replicase via a ribosomal frameshift (see Chapter 7); ORF3, major coat proteinthat forms a minor coat protein by read-through into ORF5 (see Chapter 7); ORF4, cell-to-cell movement;there are some minor ORFs within the major ones that may regulate transcription late inreplication.NotesAn important disease of potato.Further InformationHarrison, B.D. (1984). Potato leafroll virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 291.Taliansky, M., Mayo, M.A., and Barker, H. (2003). Potato leafroll virus: A classic pathogen shows some new tricks. Molec.Plant Pathol. 4, 81–89.

PROFILE NO. 9.Profile No. 9POTATO VIRUS X (PVX)ClassificationType member of the genus Potexvirus in the family Flexiviridae.339Symptoms and Host RangeNatural infections of PVX cause a mild mosaic in potato (Solanum tuberosum), a mottle intobacco (Nicotiana tabacum), and mosaic in tomato (Lycopersicum esculentum). It has a wide experimentalhost range of more than 240 species, mainly in the Solanaceae.StrainsThere are many minor variants of PVX.Geographic DistributionWorldwide in potato-growing areas.TransmissionMost field transmission is by mechanical contact. It is not seed transmitted.ParticlesPVX has flexuous rod-shaped particles 515 nm long and 13 nmwide [Figure; bar ¼ 100 nm]. [Courtesy of Brunt et al. (2000), inVirus Taxonomy. Seventh Report of the International Committee on theTaxonomy of Viruses, M.H.V. van Regenmortel et al., Eds., pp. 975–981, Academic Press, San Diego.] The genomic RNA is encapsidatedin a single species of coat protein.

340 PROFILESGenome and Genome OrganisationThe PVX genome is a single molecule of positive-sense ssRNA, 5.84 kb, the 5’ end of which hasa cap and the 3’ end is polyadenylated. The genome contains 5 ORFs.represents the 5’ cap, AAA shows a 3’ polyA sequence, methyl transferase domain, helicasedomain, RdRp domain. ORF1 encodes the replicase; ORFs 2, 3, and 4 form the triple geneblock for cell-to-cell movement; ORF5 encodes the coat protein. ORF1 is translated from the genomicRNA; ORFs 2–4 from subgenomic RNA 1 and ORF5 from subgenomic RNA2.NotesPVX is moderately important as a disease agent in potato and together with PVY is important incausing tomato streak disease.Further InformationKoenig, R. and Lesemann, D.-E. (1989). Potato virus X. Association of Applied Biologists Descriptions of Plant Viruses, No. 354.Ryu, K.H. and Hong, J.S. (2008). Potexvirus. Encyclopedia of Virology, Vol. 4, 310–313.

PROFILE NO. 10.Profile No. 10GENUS POTYVIRUSPOTATO VIRUS Y (PVY)ClassificationType species of the genus Potyvirus in the family Potyviridae.341Symptoms and Host RangeSymptoms vary according to host and strain of virus. Some strains are mild and cause a rugosemosaic (left-hand Figure); others cause a severe mosaic in potato with necrosis of tubers (righthandFigure). [Figures courtesy of Association of Applied Biologists Descriptions of Plant Viruses,No. 242.] Moderate natural host range mainly in the family Solanaceae; experimental host rangenearly 500 spp. In >70 genera of >30 plant families.StrainsThree main strain groups PVY O , PVY N , and PVY C, based on symptoms in Nicotiana tabacum cvSamsun, Solanum tuberosum ssp. tuberosum, and Physalis floridana.Geographic DistributionWorldwide in potato-growing areas.

342 PROFILESTransmissionTransmitted by aphids in the nonpersistent manner; requires a virus-coded transmission factor(see Chapter 12). Can be transmitted mechanically; there is no evidence for seed transmission.ParticlesLong flexible particles about 750 nm long and 11 nm wide.Figure shows negatively stained preparation of PVY. [Courtesyof Association of Applied Biologists Descriptions of PlantViruses, No. 242.]Genome and Genome OrganisationThe genome is a single molecule of positive-sense ssRNA, about 9.7 kb, the 5’ end of which hasa VPg and the 3’ end is polyadenylated (see Chapter 6). Potyviral genomes have very similargenome organisations—that of Tobacco etch virus is shown here.¼ VPg, ¼ Poly-A. ¼ helicase motif, ¼ RdRp motif. The potyvirus genome is expressedas a polyprotein that is processed to the gene products by proteinases (see Figure 7.4). The functionsof the gene products are: P1, serine proteinase; HC-Pro a cysteine proteinase, helper componentfor aphid transmission, suppressor of RNA silencing; P3 þ 6K, pathogenicity and host rangedeterminant, 71K, cytoplasmic inclusion (CI) body protein RNA helicase, cell-to-cell movement,RNA replication; 6K, membrane anchoring; VPg, 5’ virus-linked genome protein; 27K, cysteineproteinase, priming RNA synthesis; (VPg þ 27K proteins give the nuclear inclusion a - NIa);NIb, nuclear inclusion protein, RNA replicase; 30K, coat protein.

NotesPROFILE NO. 10.The potyvirus genus is the biggest genus of plant viruses with >110 species.343Further InformationKerlan, C. (2006). Potato virus Y. Association of Applied Biologists Description of Plant Viruses, No. 242.Kerlan, C. and Moury, B. (2008). Potato virus Y. Encyclopedia of Virology, Vol. 4, 288–301.López-Moya, J.J. and Garcia, J.A. (2008). Potyviruses. Encyclopedia of Virology, Vol. 4, 314–324.Other PotyvirusesPAPAYA RINGSPOT VIRUS (PRSV)PRSV infection causes significant losses in papaya (Caricapapaya). Leaves show mottling and distortion with ringspotson the fruit. The virus is found in most papaya growingcountries.There are several strains of PRSV that differ in severity ofsymptoms and in their genomic nucleic acid sequences.Further InformationGonsalves, D., Suzuki, J.Y., Tripathi, S., and Ferreira, S.A. (2008). Papaya ringspot virus. Encyclopedia of Virology, Vol. 4, 1–7.Purcifull, D., Edwardson, J., Hiebert, E., and Gonsalvez, D. (1984). Papaya ringspot virus. Association of Applied BiologistsDescriptions of Plant Viruses, No. 292.

344 PROFILESPLUM POX VIRUS (PPV)PPV causes Sharka disease or “plum pox” of Prunusspp (plum, apricot, peach, sour cherry, sweetcherry). Leaf symptoms are chlorotic ring spotsand oak-leaf patterns; fruit symptoms are shallowrings and discolouring. [Figure courtesy of Associationof Applied Biologists Descriptions of Plant Viruses,No. 410.] Premature fruit drop. The virus is foundin Europe, North Africa, Asia Minor, China, andSouth and North America.There are several strains of this important virusdiffering in symptoms in various hosts.Further InformationGlasa, M. and Candresse, T. (2005). Plum pox virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 410.Glasa, M. and Candresse, T. (2008). Plum pox virus. Encyclopedia of Virology, Vol. 4, 238–242.TOBACCO ETCH VIRUS (TEV)TEV causes mottling and necrotic etching of some varieties of tobacco (Nicotiana tabacum); inother varieties it is less severe. It also causes diseases of peppers and tomatoes. It has a wide experimentalhost range of more than 120 species. It is found in North, Central, and South America. Thestudy of TEV has contributed much to the understanding of the functioning of potyviruses.Further InformationPurcifull, D.E. and Hiebert, E. (1982). Tobacco etch virus. Association of Applied Biologists Descriptions of Plant Viruses,No. 258.

PROFILE NO. 11.Profile No. 11RICE RAGGED STUNT VIRUS (RRSV)ClassificationType species of the genus Oryzavirus in the family Reoviridae.345Symptoms and Host RangeInfected plants are stunted with white spindle-shaped enationson the back of twisted and ragged leaves. [Figure courtesy ofP. Waterhouse.] Delayed flowering and unfilled grains lead to yieldlosses; generally losses are 10 to 20%, but severe infections canresult in 100% loss. The host range is restricted to the genus Oryzaand some other members of the Poaceae.StrainsThere are three strains of RRSV, -India, -Philippines, and -Thailand.Geographic DistributionRRSV has been reported from China, India, Indonesia, Japan, Malaysia, the Philippines, SriLanka, Taiwan, and Thailand.TransmissionThe virus is transmitted by the rice brown planthopper, Nilopavarta lugens, in a circulative propagativemanner (see Chapter 12). Not mechanically or seed transmitted.

346 PROFILESParticlesIsometric, 75–80 nm in diameter (Fig. bar ¼ 50 nm);(courtesy of Descriptions of plant viruses, No. 248) with12 A-type surface spikes; the inner “subviral” core is57–65 nm in diameter and has 12 B-type spikes (seeChapter 5). Each particle contains a full complement ofgenome segments.Genome and Genome OrganisationTen segments of dsRNA ranging in size from 3.85 kbp to 1.16 kbp and totalling about 26 kbp.Each of the genomic segments is monocistronic, except segment 4, which is bicistronic. Thesegments code for: 1, B spike; 2, inner core capsid; 3, major core capsid; 4, RNA-dependent replicaseand unknown; 5, capping enzyme/guanyltransferase; 6, unknown; 7, nonstructural protein; 8, precursorprotease (major capsid); 9, vector transmission (spike); 10, nonstructural protein.NotesRRSV replicates in both plants and insects. It resembles many of the animal reoviruses.Further InformationGeijskes, R.J. and Harding, R.M. (2008). Plant reoviruses (Phytoreoviruses and Fijiviruses). Encyclopedia of Virology, Vol. 4,149–155.Milne, R.G., Boccardo, G., and Ling, K.C. (1982). Rice ragged stunt virus. Association of Applied Biologists Descriptions of PlantViruses, No. 248.

PROFILE NO. 12.Profile No. 12RICE STRIPE VIRUS (RSV)ClassificationType species of the genus Tenuivirus, not assigned to any family.347Symptoms and Host RangeGeneral stunting of the plant with chlorotic stripes and generalchlorosis on leaves (Figure); early infection gives significant lossof yield of rice due to few or no flower panicles being formed. Hostrange limited to monocotyledons, mainly Graminae.StrainsSome isolates differ in severity of symptoms.Geographic DistributionRSV causes important diseases in China, Japan, Korea, and Taiwan.TransmissionTransmitted by planthoppers in the circulative propagative manner. The main field vector is thesmall brown planthopper, Laodelphax striatellus.

348 PROFILESParticlesFilamentous nucleoprotein particles 500–2,000nm long and 8 nm wide (Figure). Some particlesmay appear branched or circular.Genome and Genome OrganisationThe ssRNA genome of RSV is divided into 4 segments, the largest (RNA1) being negative sense,and the other three (RNA 2–4) being ambisense (see Chapter 7 for ambisense).The product from the L segment is the RNA-dependent RNA polymerase (337 kDa). The functionsof the two products from RNA2 are unknown. The P35 from RNA3 is the nucleocapsid protein;P24 is possibly a suppressor of RNA silencing. The P20 from RNA4 is a major nonstructuralprotein; the function of P32 is unknown.NotesRSV replicates both in plants and insects. Tenuiviruses have some similarities with members ofthe Bunyaviridae, particularly those in the genus Phlebovirus.Further InformationRamirez, B.C. (2008). Tenuivirus. Encyclopedia of Virology, Vol. 5, 24–26.Toryama, S. (2000). Rice stripe virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 375.

PROFILE NO. 13.Profile No. 13SONCHUS YELLOW NET VIRUS (SYNV)ClassificationSpecies in genus Nucleorhabdovirus in the family Rhabdoviridae.349Symptoms and Host RangeInfected plants show stunting, mosaic symptoms,veinal necrosis, yellowing and crinkling ofleaves. Natural host range restricted to the Compositae;experimental host range includes Nicotianabenthamiana, [Figure courtesy of A. Jackson.] N. clevelandii,and Chenopodium quinoa, which is a locallesion host.StrainsNo strains known.Geographic DistributionCentral and South Florida.TransmissionTransmitted in a persistent manner by the aphid, Aphis coreopsidis in which it replicates.

350 PROFILESParticlesBacilliform particles 95 nm in diameter and 250 nm in length withsurface spikes. The Figure of an electron micrograph of a negativestained particle shows the striated inner core, envelope, and glycoproteinspikes. [This article was published in Encyclopedia of Virology,Vol. 4, A.O. Jackson, R.G. Dietzgen, R.-X. Fang, M.M. Goodin, S.A.Hogenhout, M. Deng, and J.N. Bragg (B.W.J. Mahy and M.H.V. vanRegenmortel, Eds.), Plant rhabdoviruses, pp. 187–196, Copyright Elsevier(2008).] For details of structure, see Chapter 5.Genome and Genome OrganisationOne component of negative-sense ssRNA, about 13.7 kb.SYNV encodes 6 proteins that are translated from separate transcripts. The functions of the geneproducts are: L, RNA-dependent RNA replicase; G, glycoprotein that forms the spikes; M1 andM2, structural matrix proteins; sc4, cell-to-cell movement protein; N, nucleocapsid protein. Fortranscription and more on gene functions, see Chapter 7.NotesSYNV replicates both in plants and insect and resembles animal retroviruses; the gene productsc4 is not found in animal rhabdoviruses and demonstrates the adaptation of the genome toplants.Further InformationJackson, A.O. and Christie, S.R. (1979). Sowthistle yellow net virus. Association of Applied Biologists Descriptions of PlantViruses, No. 205.Jackson, A.O., Dietzgen, R., Goodin, M.M., Bragg, J.N., and Deng, M. (2005). Biology of plant rhabdoviruses. Annu. Rev. Phytopathol.43, 623–660.Jackson, A.O., Dietzgen, R.G., Fang, R.-X-., Goodin, M.M., Hogenhout, S.A., Deng, M., and Bragg, J.N. (2008). Plant rhabdoviruses.Enclopedia of Virology, Vol. 4, 187–196.

PROFILE NO. 14.Profile No. 14TOBACCO MOSAIC VIRUS (TMV)ClassificationType species of the genus Tobamovirus.351Symptoms and Host RangeSymptoms of TMV in N. tabacumcv Samsun 3 weeks after infectionLocal lesions in N. tabacum NNgenotype 1 week after infectionTMV has a moderate host range including many solanaceous species. In Nicotiana tabacum cvSamsun the type strain causes initial vein clearing followed by a light and dark green mosaic, distortion,and blistering of systemically infected leaves (left-hand figure). Causes severe crop losses.In N. tabacum cv Samsun NN (right-hand figure) and N. glutinosa causes local lesions.[Figures courtesy of Association of Applied Biologists Descriptions of Plant Viruses, No. 370.]StrainsMany strains differing in symptoms and host range; the Ob strain overcomes the N generesistance.Geographic DistributionWorldwideTransmissionVery infectious, transmitted by contact; see Chapter 12.

352 PROFILESParticlesRigid rod-shaped particles 18 300 nm; see Figure 4.1. The particles are very stable.Genome and Genome OrganisationLinear (þ)-sense ssRNA, type strain 6395nt.5’ cap; methyl transferase; helicase; readthrough; RdRp; tRNA-like 3’ end. ORFs 1and 2 Replicase; ORF3 movement protein; ORF4 coat protein.NotesTMV was the first pathogen to be recognised as a virus (see Chapter 1).Further InformationKnapp, E. and Lewandowski, D.J. (2001). Tobacco mosaic virus, not just a single component virus anymore. Molec. PlantPathol. 2, 117–123.Lewarndowski, D.J. (2008). Tobamovirus. Encyclopedia of Virology, Vol. 5, 68–71.Van Regenmortel, M.H.V. (2008). Tobacco mosaic virus. Encyclopedia of Virology, Vol. 5, 54–59.Zaitlin, M. (2000). Tobacco mosaic virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 370.

PROFILE NO. 15.Profile No. 15TOBACCO RATTLE VIRUS (TRV)ClassificationType species of the genus Tobravirus.353Symptoms and Host RangeTRV causes systemic necrotic flecks on tobacco (Nicotiana tabacum)leaves, stunting of the shoots, and sometimes death. In potato (Solanumtuberosum) it causes an important disease, spraign or corky ringspot,with necrotic rings in the tuber flesh. [Figure courtesy of Associationof Applied Biologists Descriptions of Plant Viruses, No. 398.] The virushas a wide experimental host range of more than 400 species in morethan 50 dicot and monocot families.StrainsThere are many strains of TRV distinguished on symptoms in test plants.Geographic DistributionThe virus causes diseases in Europe, Japan, New Zealand, and North America.TransmissionTRV is transmitted by nematodes (Trichodorus and Paratrichodorus spp.; see Chapter 12 for nematodetransmission). It is also mechanically transmissible and seed transmitted.

354 PROFILESParticlesThe two genomic segments are each encapsidatedin rigid rod (tubular) particles of length 190 nm and50–115 nm (depending on isolate) and diameter 23 nm.Figure, bar ¼ 100 nm. [Courtesy of Association of AppliedBiologists Descriptions of Plant Viruses, No. 398.]Genome and Genome OrganisationThe genome is divided between two positive-sense ssRNA species (RNA1 and RNA2) each havinga 5’ cap and the 3’ hydroxyl group. The size of RNA2 differs between isolates.5’ cap; methyl transferase; helicase; readthrough; RdRp; ¼ 3’ hydroxyl group.RNA1 has 4 ORFs: ORF1 reads through into ORF2 to give the replicase protein (194 kDa); thethird ORF (P1a) is the 30 kDa cell-to-cell movement protein; the fourth ORF (P1b) expresses a 16kDa protein that suppresses RNA silencing. The 5’ ORF of RNA2 of all strains encodes the coatprotein; some strains have additional ORFs (P2b and P2c) that are involved in nematode transmissionof the virus. P1 a and b and P2 b and c are expressed from subgenomic RNAs.NotesTRV causes many natural diseases in a range of dicot and monocot plants.Further InformationMacFarlane, S.A. (2008). Tobravirus. Encyclopedia of Virology, Vol. 5, 72–75.Robinson, D.J. (2003). Tobacco rattle virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 398.

PROFILE NO. 16.Profile No. 16FAMILY TOMBUSVIRIDAEGENUS TOMBUSVIRUSTomato bushy stunt virus (TBSV)355ClassificationType member of the genus Tombusvirus in the family Tombusviridae.Symptoms and Host RangeDepending on host and virus strain range from mild to severe stunting, bushy growth, necrosis,chlorotic spots, and crinkling of leaves. The left-hand panel of the Figure shows distortion andnecrosis of tomato leaves; the right-hand panel shows deformation and chlorotic blotches ontomato fruit. [Courtesy of Association of Applied Biologists Descriptions of Plant Viruses, No. 382.]Restricted host range.StrainsThree major strains differentiated on host range and symptoms.Geographic DistributionEurope, North and South America, and North Africa.TransmissionNo vector known but evidence for soil transmission. Mechanically transmissible; seed transmittedin some hosts.

356 PROFILESParticlesIsometric 30 nm in diameter with T ¼ 3 symmetry (seeChapter 15). The Figure shows a preparation of TBSVnegatively stained in uranyl acetate; bar ¼ 50 nm. [Courtesyof Association of Applied Biologists Descriptions of PlantViruses, No. 382.]Genome and Genome OrganisationThe genome is a single molecule of positive-sense ssRNA, 4.78 kb, the 3’ end of which has ahydroxyl group (see Chapter 6).þ read-through of stop codon, ¼ RdRp motif, ¼ 3’ hydroxyl group. The genome encodes5 ORFs the functions being: ORF1 viral replicase with read-through of a weak stop codon (seeChapter 7); ORF2, coat protein; ORF3, cell-to-cell movement; ORF4, suppressor of RNA silencing;pX, function unknown.NotesTBSV is important in studies on virus structure and virus-host interactions.Further InformationLommel, S.A. and Sit, T.L. (2008). Overview [Tombusviridae]. Encyclopedia of Virology, Vol. 5, 145–150.Martelli, G.P., Russo, M., and Rubino, L. (2001). Tomato bushy stunt virus. Association of Applied Biologists Descriptions ofPlant Viruses, No. 382.Yamamura, Y. and Scholthof, H.B. (2005). Tomato bushy stunt virus: A resilient model system to study virus-plant interactions.Molec. Plant Pathol. 6, 491–502.

Carnation mottle virus (CarMV)PROFILE NO. 16.GENUS CARMOVIRUS357ClassificationCarMV is the type member of the genus Carmovirus in the family Tombusviridae.Symptoms and Host RangeThe virus causes a mild mottle and loss of vigour in carnations. Its natural host range isrestricted to Caryophyllaceae. The experimental host range is of more than 30 species in 15 dicotfamilies.StrainsThere are a few min variants of CarMV distinguished on symptoms in test plants.Geographic DistributionWorldwide wherever carnations are grown.TransmissionNo natural vectors known. It is distributed by vegetative propagation. The virus is mechanicallytransmissible but not seed transmitted.ParticlesCarMV particles are icosahedral, 30 nm in diameterwith T ¼ 3 symmetry. The Figure shows a preparation ofCarMV negatively stained in phosphotungstate; bar ¼100 nm. [Courtesy of Association of Applied BiologistsDescriptions of Plant Viruses, No. 7.]

358 PROFILESGenome and Genome OrganisationThe 4 kb genome of CarMV is a single species of positive-sense ssRNA, capped at its 5’ end andwith a 3’ hydroxyl group.5’ cap; read-through; RdRp; ¼ 3’ hydroxyl group. The genome has four ORFs. ORF1consists of two parts; the N-terminal part reading through to the C-terminal part to give the replicase.ORFs 2 and 3 are involved in cell-to-cell movement and ORF4 encodes the coat protein. ORFs2 þ 3 and ORF4 are expressed from subgenomic RNAs.NotesCarmoviruses such as CarMV and Turnip crinkle virus are important in molecular studies onsimple plant viruses and also on associated satellite RNAs.Further InformationHollings, M. and Stone, O.M. (1970). Carnation mottle virus. Association of Applied Biologists Descriptions of Plant Viruses,No. 7.Qu, F. and Morris, T.J. (2008). Carmovirus. Encyclopedia of Virology,Vol. 1, 453–456.

PROFILE NO. 17.Profile No. 17TOMATO SPOTTED WILT VIRUS (TSWV)ClassificationType species of genus Tospovirus in the family Bunyaviridae.359Symptoms and Host RangeTSWV has a very wide host range of >800 speciesfrom >80 botanical families, both dicotyledons andmonocotyledons. It also infects about 10 thrip species.The symptoms vary between and within hostspecies and range from chlorosis, mottling, stunting,and wilting to severe necrosis of leaves, stems, andfruits. The figure shows necrotic blotching and ringspotson tomato fruit. [Courtesy of Association ofApplied Biologists Description of Plant Viruses, No. 412.]StrainsThere are several strains of TSWV. Other tospoviruses have wide host ranges and can causesimilar symptoms.Geographic DistributionTSWV is found in temperate regions worldwide.TransmissionTransmitted by thrips in the circulative propagative manner (see Chapter 12). Frankinella occidentalisis an important vector especially in glasshouses. The virus is mechanically transmissible withdifficulty and is not seed transmitted.

360 PROFILESParticlesMembrane-bound quasi spherical particles 80–120 nm in diameterwith surface projections 5–10 nm in length (see Chapter 5). The Figureshows a negatively stained preparation of TSWV; bar ¼ 100 nm. [Courtesyof Association of Applied Biologists Descriptions of Plant Viruses, No. 412.]Genome and Genome OrganisationThe genome is divided between three segments of ssRNA; the L segmentis negative-sense 8.90 kb, the M and S segments are ambisense of4.82 and 2.92 kb, respectively (see Chapter 7 for ambisense).The product from the L segment is the RNA-dependent RNA polymerase (330 kDa). The P34(NSm) from the M segment is involved in cell-to-cell movement and the GP 1 & 2 form the glycoproteinprojection on the particle; the P34 (NSs) from the S segment is the suppressor of RNAsilencing and the N protein forms the nucleocapsid.NotesTSWV is an important plant virus economically. It replicates both in plants and insect andresembles animal bunyaviruses.Further InformationAdkins, S. (2000). Tomato spotted wilt virus—positive steps toward negative success. Molec. Plant Pathol 1, 151–157.Kormelink, R. (2005). Tomato spotted wilt virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 412.Pappu, H.R. (2008). Tospoviruses. Encyclopedia of Virology, Vol. 5, 157–162.

PROFILE NO. 18.Profile No. 18TURNIP YELLOW MOSAIC VIRUS (TYMV)ClassificationType species of the genus Tymovirus in the family Tymoviridae.361Symptoms and Host RangeCauses a bright yellow to white or yellow greenmosaic of leaves of Chinese cabbage (Brassica pekinensis).Limited host range mainly of Cruciferae.The Figure shows a white mosaic in turnip.StrainsThere are several strains of TYMV that differ in symptom severity and serologically.Geographic DistributionEuropeTransmissionTYMV is transmitted in a noncirculative manner by beetles. The virus is mechanically transmissiblebut not seed transmitted.

362 PROFILESParticlesTYMV has icosahedral particles, about 30 nm in diameter with T ¼ 3 symmetry. The Figure ofparticles negatively stained in uranyl acetate shows morphological subunits at the 5-fold axis(left-hand panel) and 2-fold axis (right-hand panel); bar ¼ 30 nm. [Courtesy of Association of AppliedBiologists Descriptions of Plant Viruses, No. 230.] The particles are made up of one coat protein speciesand are very stable, some particles containing the single genome species and others being “empty.”Genome and Genome OrganisationThe genome is a single molecule of positive-sense ssRNA, 6.32 kb with a 5’ cap and 3’ tRNAlikestructure.5’ cap; methyl transferase; 0 protease; helicase; RdRp; tRNA-like 3’ end. The genomehas three ORFs, ORF0 encoding a protein of 69 kDa that is involved in cell-to-cell movement andsilencing suppression; ORF1, a 206 kDa replicase protein that is processed between the helicaseand RdRP domains by the protease domain to give products of 141 and 66 kDa; ORF2 encodesthe 20 kDa coat protein.NotesStudy of the full and empty particles of TYMV led to the recognition of an infectious RNAgenome. Its replication involves the chloroplast membrane.Further InformationDreher, T.W. (2004). Turnip yellow mosaic virus: Transfer RNA mimicry, chloroplasts and a C-rich genome. Molec. PlantPathol. 5, 367–376.Haenni, A.-L. and Dreher, T.W. (2008). Tymoviruses. Encyclopedia of Virology, Vol. 5, 199–208.Matthews, R.E.F. (1980). Turnip yellow mosaic virus. Association of Applied Biologists Descriptions of Plant Viruses, No. 230.

IndexNote: Page numbers followed by f refer to figures; page number followed by t refer to tables; and b refer to boxes.AAbbreviations, of viruses, 16–17ABMV. See Abutilon mosaic virusAbutilon mosaic virus (ABMV), 16,26, 332Acholeplasmatales, 50ACMV. See African cassava mosaicvirusAcquired resistance, 281bAcronyms, of viruses, 16–17Actinomycin D, 141bActivities, of viral genes, 114–116Adaption, to niches, 71Addition, of virus nucleotides, 65African cassava mosaic virus (ACMV),217b, 290, 330, 331Aggregates, of virus-encodedproteins, 35Agricultureepidemiology of plant viruses in,261–266modifications for virus control,272–273, 275–276Agronomic practicefor stopping vectors, 275–276for virus control, 272–273Air-borne vectors, stopping of,274–276Alfalfa mosaic virus (AMV), 16expression of, 137genome of, 108, 112plant interactions of, 202, 215, 216fplant resistance to, 291profile of, 312, 313replication of, 149bstructure of, 98–99transmission of, 232Alfamovirus, 231Ambisense, 131AMV. See Alfalfa mosaic virusAncillary RNAs, of Beet necrotic yellowvein virus, 56Antagonism, 202Antibiosis, 275Antibodies, 250b, 290, 291bAntibody-virus combination,methods for detecting, 249–256Antigens, 250bAntisera, types of, 249Antiviral chemicals, 278–280Aphid transmission factor. See HelpercomponentAphid, virus transmission by, 225–240, 261, 262, 264f, 274–276Apoptosis. See Programmed celldeathApple stem grooving virus, 32Architecture, of virus particles, 83–104complex isometric viruses, 101enveloped viruses, 101–102general considerations, 103–104introduction, 85–86isometric viruses, 94–97methods, 86–88rod-shaped viruses, 88–94small icosahedral viruses, 97–101summary, 104viruses of other kingdoms, 104Arrest, hybrid, 115Arthropods, virus transmission by,229bArtificially induced mutation,location of, 114–115ASBVd. See Avocado sun blotch viroidAssemblyof TMV, 92–94, 120bof virus particles, 83–104general considerations, 103–104icosahedral viruses, 102–103introduction, 85–86methods, 86–88summary, 104viruses of other kingdoms, 104Atomic force microscopy, of viruses, 87A-type satellite viruses, 52–53Aureusvirus, 211Avirulence (Avr) genes, 193b, 194bAvocado sun blotch viroid (ASBVd),45, 47Avoidanceof crop infection, 271–274of RNA silencing, 214Avr genes. See Avirulence genesAvsunviroidae, 44, 45–47, 47–48BB type satellite RNAs, 53–57Bacilliform particles, 98–99, 100Bacteria, virus v., 4Badnavirus, 20, 154Bamboo mosaic virus, 258bBanana bunchy top virus (BBTV), 272bBanana streak virus (BSV), 26, 29, 162b,258b, 259, 316, 317, 318Barley stripe mosaic virus (BSMV), 29,39, 110, 179, 211, 300f, 304, 305Barley yellow dwarf virus (BYDV), 28,179, 262, 281, 335, 336BBTV. See Banana bunchy top virusBean common mosaic virus, 281Bean golden mosaic virus (BGMV), 332Bean yellow mosaic virus, 281Beet curly top virus, 157bBeet mild yellows virus, 17, 203, 275Beet mosaic virus, 275Beet necrotic yellow vein virus(BNYVV), 56, 73, 242, 259, 264,267, 267f, 306, 307Beet western yellows virus (BWYV), 17Beet yellows virus (BYV), 25, 26, 28,135, 144, 184, 275, 322, 323Beetle, virus transmission by, 238–239Begomoviruses, 57–58, 59t, 60, 66,76b, 156, 158, 175, 212Beijerinck, Martinus Willem, 4, 7f363

364 INDEXBenyviruses, 143Biochemical studies, for determiningarchitecture and assembly ofvirus particles, 86Biological activity, of viroids,molecular basis for, 50Biology, of virus, diagnostic methodsusing, 246–247BMV. See Brome mosaic virusBNYVV. See Beet necrotic yellow veinvirusBreeding, for resistance, 282Broad bean wilt virus-1, 33, 33f, 35fBrome mosaic virus (BMV), 101,119–121, 130b, 149b, 161, 161f,168, 308, 309Bromoviridae, 59t, 65, 101, 102, 144,149b, 161, 238BsatTBRV. See Tomato black ring virusBSMV. See Barley stripe mosaic virusBSV. See Banana streak virusBunyaviruses, 136BWYV. See Beet western yellows virusBYDV. See Barley yellow dwarf virusBYV. See Beet yellows virusCC type satellite RNAs, 53–57Cacao swollen shoot virus (CSSV), 267,272bCambial cells, abnormal division of,32CaMV. See Cauliflower mosaic virusCap5 0 , 109lack of, 136in virus without poly(A) tail, 136Cap snatching, 136–137Capilloviruses, 144Capsid interaction, 232Carbohydrates, plant level virus-hostinteractions effects on plantmetabolism of, 182–184Carmoviruses, 53CarMV. See Carnation mottle virusCarnation Italian ringspot virus, 215bCarnation mottle virus (CarMV),357, 358Cauliflower mosaic virus (CaMV), 122,201fcytological effects of, 35, 37fdiagnosis of, 257expression of, 133, 134as gene vectors, 294genome of, 105, 114, 115mRNA synthesis in, 124–125plant interaction of, 174, 200profile of, 314, 315–316replication of, 153, 155, 155f, 159structure of, 100symptoms of, 26in transgenic plants, 290, 291btransmission of, 232–234, 233fCaulimoviridae, 14, 35–36, 67, 70b,108b, 122, 124–125, 131, 134, 140,153–154, 154–156, 159, 161, 174,175f, 231, 232–234, 294CCCVd. See Coconut cadang-cadangviroidCell cycle, geminivirus control of,157bCell death, plant virus effects on, 34Cell division, of infected plants, 32Cell size, of infected plants, 32Cell structure, plant virus effects on,32–34Cell walls, plant virus effects on, 33Cell-to-cell movement, 40proteins facilitating, 109rate of, 176of viroids, 45, 175of virus in plant, 168–176Cell-to-cell movement proteins, 71Cellular genes, effect of Pea seed-bornemosaic virus infection on, 183bCereal yellow dwarf virus, 54, 235CEVd. See Citrus exocortis viroidChemical studies, for determiningarchitecture and assembly ofvirus particles, 86Chlamydiae, 9Chloramphenicol, 141bChloroplasts, plant virus effects on,33, 186Chronic bee-paralysis virus, 60Chrysanthemum stunt viroid (CSVd),45Circulative transmission. SeePersistent transmissionCirculative viruses, 235–237Citrus exocortis viroid (CEVd), 45, 49Citrus tristeza virus (CTV), 88fconventional control of, 274,278, 279bdiagnosis of, 259evolution of, 75expression of, 129f, 130, 137genome of, 112plant interactions of, 211profile of, 320, 321symptoms of, 29Classificationof viroids, 44of viruses, 13–14, 17Cleavage, of DNA, virus diagnosisby, 257Clitoria yellow vein virus, 281Closterovirus, 14, 67, 70b, 75, 88, 106,135, 211, 231, 232Clover yellow vein virus, 275Clustering, of isometric virussubunits, 97CMV. See Cucumber mosaic virusCNV. See Cucumber necrosis virusCoat protein, 70–71, 92, 286Coat protein-mediated resistance,286, 287, 288b, 289Coat-dependent RNA replicons,52, 52tCocoa swollen shoot virus (CSSV),29, 32Coconut cadang-cadang viroid(CCCVd), 44, 45Coevolution, of viruses with hostsand vectors, 80Comoviridae, 70b, 101, 126–128, 144,174, 238Comparison, of sequence to genewith known function, 116Complementary 5 0 and 3 0 sequences,of viral genome, 112Complementary strands, translationof, 131Complementation, in viralmovement, 176Complete dependence, on otherviruses, 203Complex isometric viruses, 101Control. See also Conventional controlof transgenic plants, viruses assources of, 295of vectors, 274–277Conventional control, 269–284avoiding infection, 271–274conventional resistance to plantviruses, 280–283introduction, 269–270plant protection, 277–280

stopping vectors, 274–277strategies for, 283summary, 283–284viruses of other kingdoms, 283Conventional resistance, to plantviruses, 280–283Convergence, of sequence, 67Coreplicational disassembly, 120bCoronaviruses, 130bCo-translational disassembly,119, 120bCotton leaf curl disease, 332Cowpea aphid-borne mosaic virus, 276Cowpea chlorotic mottle virus, 100Cowpea mosaic virus (CPMV), 70b, 133,174, 175f, 202, 296–298, 324, 325Cowpea severe mosaic virus (CPSMV),202CPMV. See Cowpea mosaic virusCPSMV. See Cowpea severe mosaic virusCriniviruses, 88Cropavoiding infection of, 271–274conventional resistance of, 280–283epidemiology of plant viruses in,261–266Crop losses, due to plant viruses, 24Crop rotation, virus spread and, 263Cross-protection, 49, 202, 277–278,279b, 289Cryptoviruses, 29Crystallography, of viruses, 87Crystals, of virus particles, 34–35CSSV. See Cacao swollen shoot virus;Cocoa swollen shoot virusCSVd. See Chrysanthemum stunt viroidCTV. See Citrus tristeza virusCucumber mosaic virus (CMV), 16conventional control of, 275, 278defective and defective-interferingnucleic acids of, 59f, 60diagnosis of, 246, 248bhost range of, 39plant interactions of, 179, 182, 212plant resistance to, 282, 291profile of, 309, 310, 311satellite RNAs of, 53–54, 54b, 56symptoms of, 28transmission of, 232Cucumber necrosis virus (CNV), 241Cucumoviruses, 53, 144, 231Cultivar resistance, 276, 281bCultivation, virus spread and, 263, 265INDEXCycloheximide, 141bCytological effects, of plant viruses,32–38cell structure effects, 32–34inclusion bodies, 37plant cytological structuresmistaken as, 37–38virus-induced structures incytoplasm, 34–36Cytological structures, of normalplant, 37–38Cytology, in virus diagnosis, 247Cytopathic effects, of viroids, 44–45Cytoplasm, virus-induced structuresin, 34–36Cytoskeletal system, of plant, 170bDD type satellite RNAs, 53–57Dahlia mosaic virus, 175fDark-green islands, 216Death, from viral infection, 34Defective nucleic acids, 58–60Defective-interfering nucleic acids,58–60, 73, 290Defense systems, of host, proteins forovercoming, 109Deletion, of virus nucleotides, 65Dependence, on other viruses, 203Dependent transmission, 237Detection, of antibody-viruscombination, 249–256Determinants, of host range, 39–40Deterrents, insect, 275Developmental abnormalities, ofinfected plants, 28–29Diagnosis, of plant viruses, 245–260decision making on, 260introduction, 245–246methods involving biology ofvirus, 246–247methods involving physicalproperties of virus particles,247–249methods involving properties ofviral nucleic acid, 256–260methods involving properties ofvirus proteins, 249–256Diagnostic procedures, for viroids, 50Dicer, 209–210, 210bDifferentiated cells, cell division in, 32Disassembly. See Assembly; UncoatingDisease. See also Plant virus diseases;Virus-plant interactionsforecasting of, 270bmanifestation of, 23–24monitoring progress of, 270brecovery from, 29satellite RNA modulation of, 56–57of viral complexes, 29–30Distribution, of virus in plant,167–181Divergence, of sequence, 67Division, of cells, of infected plants, 32DNA. See also Satellite DNAscleavage patterns of, virusdiagnosis by, 257defective, 60replication of, in single-strandedDNA viruses, 156–158viral, diagnosis methods involving,256–260DNA microarray, for virus diagnosis,259DNA promoters, for control oftransgenic plants, 295DNA replicases, 68DNA virusesdefective DNAs associated with, 60as gene vectors, 293–294mRNA synthesis in, 124–125recombination of, 159RNA silencing of, 208DNAß, 57–58Domains, of Pospiviroidae, 47bDot blot hybridization, 258bDot blot, virus diagnosis by, 256, 257Dot ELISA, 252bDouble-stranded DNA (dsDNA),124–125Double-stranded RNA (dsRNA), inRNA silencing, 208–209, 211–214Double-stranded RNA viruses,123–124, 152dsDNA. See Double-stranded DNAdsRNA. See Double-stranded RNAEEACMV. See East African cassavamosaic virusEast African cassava mosaic virus(EACMV), 217bEcology, of plant viruses, 260–267in agriculture, 261–266365

366 INDEXEcology, of plant viruses (Continued)emergence of new viruses, 267in natural environment, 267Economic losses, due to plant viruses,24Electron microscope serology. SeeSerologically specific electronmicroscopyElectron microscopy, of viruses, 87,247–249, 255–256Electrophoretic procedures, for virusdiagnosis, 256ELISA. See Enzyme-linkedimmunoabsorbent assayEmbryo infection, by Pea seed-bornemosaic virus, 226bEmergence, of new viruses, 267Enamovirus, 76End-group structures, of viralgenome, 109–112Endomembrane system, of plant,170bEndoplasmic reticulum, 170b, 172bEngineered virus resistance. SeePathogen-derived resistanceEntomoplasmatales, 51Entry, of virus, 119–122Enveloped viruses, 101–102Environment, plant viruses in, 267Enzyme-linked immunoabsorbentassay (ELISA), 249–256, 252bEpidemiologycomplexities of, 266bof plant viruses, 260–267in agriculture, 261–266emergence of new viruses, 267in natural environment, 267of viroids, 45Episomal infection, from integratedBanana streak virus sequences,162bEpitopes, 250bEradication, for plant virus control,271, 272bEscape hypothesis, of virus origin,64bEukaryotic translation initiationfactor, potyviral VPg interactionwith, 195bEukaryotic translation systemconstraintson viral genome expression,125–126virus strategies for overcoming,126–136Evolutionof plant virus, 63–82, 64–74coevolution of viruses withhosts and vectors, 80evidence for, 75–79introduction, 63origins of viruses, 64–65selection pressures for, 71–74summary, 81timeline for, 74types of, 65–71variation of viruses, 65, 71, 72–73viruses of other kingdoms, 81of RNA silencing, 218Expressionof other viral protein, by transgenicplant, 286in transgenic plant, 115of viral coat protein, by transgenicplant, 286of viral genome, 117–138discussion, 137initial translation, 122mRNA synthesis, 123–125strategies, 125–137summary, 137–138virus entry and uncoating,119–122virus infection cycle, 117–119viruses of other kingdoms, 137of virus, 39–40FFabavirus, 231Families, of viruses, 14–15, 19–20Faults, of virus replication, 158–161Fieldplant viruses in, 245–268diagnosis, 245–260epidemiology and ecology,260–267summary, 268viruses of other kingdoms, 268transgenic plant release into,290–293Field performance, of transgenicplants, 293Field resistance, 282Fijivirus, 15Final distribution. See DistributionFine structure, of virus, methods fordetermination of, 86–88Flockhouse virus, 141bFlor’s gene-for-gene model, for plantresistance to pathogen orpest, 193bForecasting, of disease, 270bFrameshift proteins, overcomingeukaryotic translationconstraints with, 135–136Functional domains, in viral ORF,144–145Functional genomics, of plants,viruses in, 298Functional proteins, of viral geneproducts, 107–109Fungus, virus transmission by,240–242, 262, 277Furovirus, 59tGGeminiviridae, 14, 33, 65, 75, 99, 106,108b, 112, 122, 125, 131, 156–158,159, 161, 259, 294Gene expression, in transgenicplant, 115Gene products. See Viral geneproductsGene silencing, suppression of, 71Gene technology, possible uses ofplant viruses in, 285–286,293–299DNA viruses as gene vectors,293–294RNA viruses as gene vectors,294–295vaccine production, 295–298viruses as sources of controlelements for transgenicplants, 295viruses in functional genomics ofplants, 298viruses in nanotechnology, 298–299Gene vectorsDNA viruses as, 293–294RNA viruses as, 294–295Genera, of viruses, 14–15, 19–20Genes. See also Avirulence genes;Resistance genes; Viral genescellular, effect of Pea seed-bornemosaic virus infection on, 183bfunctional regions within, 116

Genetic effects, of virus infection, 29Genetically modified plants. SeeTransgenic plantsGenetics, of resistance to viruses, 281Genomeexpression of, 117–138discussion, 137initial translation, 122mRNA synthesis, 123–125strategies, 125–137summary, 137–138virus entry and uncoating,119–122virus infection cycle, 117–119viruses of other kingdoms, 137manipulation of, for studying virusreplication, 141bof plant viruses, 105–116general properties of, 105–112introduction, 105organization of, 112–116proteins that replicate viralgenome, 108summary, 116viruses of other kingdoms, 116of viruses, 9–13, 65, 68bGenomics, of plants, viruses in, 298Grafting, virus transmission by, 225Grape vine fan leaf virus, 265GRAV. See Groundnut rosette assistorvirusGroundnut rosette assistor virus(GRAV), 27b, 55, 203, 237Groundnut rosette disease, 27b,203, 247Groundnut rosette virus (GRV), 27b,203, 237Growth, symptom induction effectson, 186GRV. See Groundnut rosette virusHHammerhead ribozyme, 49bHBV. See Hepatitis B virusHDV. See Hepatitis delta virusHelical structure, of viruses, 89, 90Helicases, 143, 144–145, 146bHelper component, indirectinteraction involving, 232–234Helper virus, 51–58, 58–60, 237Hepatitis B virus (HBV), 60Hepatitis delta virus (HDV), 60INDEXHeteroencapsidation, 291, 292fHeterologous peptides, viruses forpresenting, 296–298Histological changes, of infectedplants, 30–32History, of virus study, 3–9HLVd. See Hop latent viroidHop latent viroid (HLVd), 44Hordeiviruses, 71, 143, 174Host. See also Virus-host interactionsadaption to, 71hypersensitive response changesin, 197–198immunity, 192–195, 281b, 282indicator, 246inoculation response of, 192–202plant virus replication by usingfunctions of, 139–140proteins facilitating movementbetween, 109proteins facilitating movementthrough, 109proteins that overcome defensesystems of, 109selection pressures exerted by, 74virus coevolution with, 80virus equilibrium in, 72–73Host range, of viruses, 38–40,246–247Host-cell defenses, virus stimulationof, 40HR. See Hypersensitive responseHybrid arrest, 115Hybrid selection, 115Hybridization procedures, 257, 258bHydrodynamic measurements, ofviruses, 86–87Hyperplasia, 32Hypersensitive response (HR),192–195, 195–198, 198–200, 199bHypoplasia, 30–32Hypotheses, on virus origin, 64bIIcosahedral structure, of viruses,94–97Icosahedral virusesassembly of, 102–103small, 97–101ICTV. See International Committeefor Taxonomy of VirusesIdaeoviruses, replication of, 144367Immunity, of host, 192–195, 281b, 282Immunosorbent electron microscopy.See Serologically specificelectron microscopyImmuno-tissue printing, 252bIn situ hybridization, 258bIn vitro assembly, of TMV rod,92–94In vitro systems, for studying virusreplication, 141bIn vitro translation systems, 113bIn vivo assembly, of TMV rod, 94In vivo systems, for studying virusreplication, 141bInclusion bodies, 36b, 37, 155–156Inclusions, caulimovirus, 35–36Incomplete dependence, on otherviruses, 203Indicator hosts, for virus diagnosis,246Induced resistance, 281bInduction, of symptoms, processesinvolved in, 185–188Infection. See also Systemic infection;Virus-host interactionsavoidance of, 271–274blockade of, 39–40of embryo, by Pea seed-borne mosaicvirus, 226bepisomal, from integrated Bananastreak virus sequences, 162blocal, 195–198mixed, 247nonpermissive, 195–200permissive, 200–202with plant virus, 23–24primary, in crops, 261–264proteins that initiate, 107–108removal of sources of, 271secondary, in crops, 264–266subliminal, 40, 195Information content, of plant viralgenome, 106Initial events, of infection, 39Initial nucleating event, of TMV rodassembly, 92Initial translation, of viralgenome, 122Initiation sites, on ORF, 133Inoculation, 192–202. See also Plant-toplantmovementInsect deterrents, 275Insecticides, 275

368 INDEXInsectsbiotypes overcoming plantresistance, 276bnonpersistent transmission by, 228,229b, 231–234persistent transmission by, 228,229b, 235–237plant virus relationship with,228–231virus transmission by, 227bIntegrated viral sequences,recombination and, 161Interactions. See also Virus-hostinteractions; Virus-plantinteractionsbetween antibodies and antigens,250bwith capsid, 232between circulative virus andvector, 235–237helper component involvement in,232–234between plant virus and insect,228–231between potyviral VPg andeukaryotic translationinitiation factor, 195bbetween virus and nematode,239–240between virus and other plantpathogens, 203–204between viruses, 202–204Intercellular movement, of virus inplant, 168–176Interferencebetween viroids, 49–50between viruses, 202Intergenic regions, of viral genome, 112Internal initiation, overcomingeukaryotic translationconstraints with, 131–132Internal ribosome entry sites (IRES),131–132International Committee forTaxonomy of Viruses (ICTV), 13,14–15, 15–16, 16–17, 19–20Intracellular movement, of virus inplants, 168Invertebrates, transmission by, 225–240beetles, 238–239nematode, 239–240, 262, 263f, 277nonpersistent transmission byinsects, 228, 229b, 231–234persistent transmission by insects,228, 229b, 235–237plant virus and insectrelationships, 228–231IRES. See Internal ribosome entrysitesIsometric virusesarchitecture of, 94–97complex, 101stabilization of, 101uncoating of, 119–121, 121–122KKingdomsof viruses, 20of viruses other than plant viruses,40, 60–61, 81, 104, 116, 137,161–162, 188, 204, 218, 242,268, 283, 300Koch’s postulates, for defining virus,11, 12tLLeaf rolling, 28Leafhoppers, virus transmission by,225–240, 275Leaky scanning, overcomingeukaryotic translationconstraints with, 133Lettuce infectious yellows virus, 135Lettuce mosaic virus, 271Lettuce necrotic yellows rhabdovirus,15, 262Limitations, in host range studies,38–395 0 Linked protein, 110Lipids, plant level virus-hostinteractions effects on plantmetabolism of, 182Living organism, virus as, 13Local acquired resistance, 198Local host, 38Local infection, 195–198Local symptoms, of viruses, 25Locationof spontaneous or artificiallyinduced mutation, 114–115of viroids in plants, 45Low-molecular-weight compounds,plant level virus-hostinteractions effects on plantmetabolism of, 184–185Luteoviruses, 54, 75–79, 106, 133, 144,168, 176, 179, 180, 203, 235–237,265, 266bMMacluravirus, 231Macroevolution, 65–67Macroscopic symptomsagents inducing virus-likesymptoms, 30diseases caused by viralcomplexes, 29–30of viroids, 44of viruses, 25–30cryptoviruses, 29local symptoms, 25systemic symptoms, 26–29Maize dwarf mosaic virus (MDMV),264fMaize mosaic virus, 35fMaize streak virus (MSV)diagnosis of, 258bexpression of, 131genome of, 106profile of, 333, 334replication of, 156structure of, 99symptoms of, 26Maize stripe virus, 136Manipulation, of viral genome, forstudying virus replication, 141bMarafiviruses, 144, 237Mass spectrometry, of viruses, 88Mathematical modeling, of diseaseprogress, 270bMayer, Adolf Eduard, 4, 7fMDMV. See Maize dwarf mosaic virusMechanical transmission, 223–224Mechanismof replication, of plus-sense singlestrandedRNA viruses, 147of RNA silencing, 208–211Membraneof endoplasmic reticulum, 172bRNA virus replication associatedwith, 146–147symptom induction role of,187–188Metabolism, of plant, virus-hostinteractions effects on, 181–185

Methods, for determiningarchitecture and assembly ofvirus particles, 86–88chemical and biochemicalstudies, 86size and fine structure studies,86–88Methyl transferase, 144–145Microarray, DNA, for virusdiagnosis, 259Microevolution, 65–67Microfilaments, 170bMicroscopy, of viruses, 87, 247–249,255–256Microtubules, 170bmiRNA, in RNA silencing, 209,210f, 216Mitochondria, plant virus effects on, 33Mixed infections, in virus diagnosis,247Modified agronomic practice, forvirus control, 272–273, 275–276Modular evolution, 67Modules, of viruses, 68bMolecular basisof biological activity of viroids, 50of RNA-mediated protection,288–289of symptom modulation bysatellite RNAs, 56–57Molecular clock hypothesis, 74Molecular level virus-plantinteractions, 191–206host response to inoculation,192–202interactions between viruses,202–204introduction, 191summary, 204viruses of other kingdoms, 204Monitoring, of disease progress, 270bMonoclonal antisera, 249Mononegavirales, 15Mosaic patterns, 26–28, 186–187,200, 216Movement. See also Cell-to-cellmovement; Plant-to-plantmovementhost to host, proteins facilitating, 109of viroids in plants, 45of virus in plants, 167–181virus interaction effects on, 203in xylem, 180INDEXMovement proteins (MPs), 169–174MPs. See Movement proteinsmRNAsynthesis of, 123–125translation of, 126bviral gene expression through, 122MSV. See Maize streak virusMuller’s ratchet, 73Multipartite genomes, 109, 131Multiple deletion D-RNAs andDI-RNAS, 60Mutationas fault of virus replication,158–159location of spontaneous orartificially induced, 114–115of viruses, 65, 73, 74Mycoplasmas, 9NN gene, of tobacco, 196, 196b, 198–200,285Naming, of virus species, 15–16Nanotechnology, virus in, 298–299Natural environment, plant virusesin, 267Natural resistance genes, 285–286Necrosis, 28, 30Negative-sense single-stranded RNAviruses, 123, 152Nematode, virus transmission by,239–240, 262, 263f, 277Nepovirus, 29, 53, 54, 215, 239–240Neutron small-angle scattering, ofviruses, 87Niche adaption, 71Nomenclature, of viruses, 13–14, 14–15,15–16, 16–17, 17–19, 19–20Non-AUG start codon, overcomingeukaryotic translationconstraints with, 133Noncoding regions, of viral genome,109–112Nonconventional protection. SeePathogen-derived resistanceNonpermissive infection, 195–200Nonpersistent transmission, 228,229b, 231–234Nonpreference, 275Nuclei, plant virus effects on, 32–33Nucleic acid suppressors, of RNAsilencing, 214Nucleic acid-based protection, 286,287–290Nucleic acids. See also Defectivenucleic acids; Defectiveinterferingnucleic acids; RNAeconomy in use of, 106–107plant level virus-host interactionseffects on plant metabolismof, 181–182of satellite viruses, 51–58, 52twithin small icosahedral viruses, 100type and size, virus diagnosis by,256–257variation in, 65of viral genome, 12, 109–112diagnosis methods involving,256–260multipartite genomes, 109noncoding regions, 109–112nucleic acid structure, 109OOdontoglossum ring spot virus (ORSV),294Open reading frame (ORF)organization of functional domainsin, 144–145in overcoming eukaryotic systemconstraints on translation,126–136of plant viral genome, 106, 112,113, 114, 126of Rice tungro bacilliform virus, 132bOrders, of viruses, 14, 15, 19–20ORF. See Open reading frameOrganization, of plant viral genome,112–116genome structure, 112–114recognizing activities of viralgenes, 114–116Origins, of viruses, 64–65. See alsoEvolutionORSV. See Odontoglossum ring spot virusOrthomyxoviruses, 65Oryzavirus, 123Overlapping ORFs, 133PPapaya ringspot virus (PRSV), 28,293, 343Pararetroviruses, 153–154369

370 INDEXParasite-derived resistance. SeePathogen-derived resistancePartitiviridae, 131PathogenFlor’s gene-for-gene model forplant resistance to, 193bprotection by, 277–278virus interactions with, 203–204Pathogen-derived resistance (PDR),285, 286–293field release of transgenic plants,290–293nucleic acid-based protection, 286,287–290other forms of, 290protein-based protection, 286–287Pathogenesis-related (PR) proteins,198–200Pathology, of viroids, 44–45PCR. See Polymerase chain reactionPDR. See Pathogen-derived resistancePea early browning virus (PEBV), 240Pea enation mosaic virus (PEMV), 32Pea seed-borne mosaic virus (PSbMV),183b, 225, 226b, 281Peach latent mosaic viroid, 46fPeanut clump virus, 211PEBV. See Pea early browning virusPEMV. See Pea enation mosaic virusPeptides, heterologous, viruses forpresenting, 296–298Performance, of transgenic plants, 293Permissive infection, 200–202Persistent transmission, 228, 229b,235–237, 237–238Pest, Flor’s gene-for-gene model forplant resistance to, 193bPhloem, structure of, 177bPhotosynthesis, plant level virus-hostinteractions effects on, 184Physical properties, of virus particles,diagnosis methods involving,247–249Physiochemical properties, of virusparticles, 247Phytoplasma, 50–51Phytoreoviruses, 101, 123Plant. See also Transgenic plants;Virus-plant interactionsantibodies in, 290, 291bcytological structures of, 37–38endomembrane and cytoskeletalsystems of, 170bhistological changes induced byinfection of, 30–32insect resistance of, 276blocation of viroids in, 45metabolism of, virus-hostinteractions effects on, 181–185movement of viroids in, 45protection of, 277–280structure of, 34bvein and phloem structure of, 177bPlant level virus-host interactions,165–190effects on plant metabolism,181–185movement and final distribution ofvirus, 167–181processes involved in symptominduction, 185–188summary, 188–189viruses of other kingdoms, 188Plant material, transmission via,223–225Plant pathogens, virus v., 9–13Plant size, virus effects on, 26Plant virusconventional resistance to, 280–283diagnosis of, 245–260decision making on, 260introduction, 245–246methods involving biology ofvirus, 246–247methods involving physicalproperties of virusparticles, 247–249methods involving properties ofviral nucleic acid, 256–260methods involving properties ofvirus proteins, 249–256evolution of, 64–74coevolution of viruses withhosts and vectors, 80evidence for, 75–79introduction, 63origins of viruses, 64–65selection pressures for, 71–74summary, 81timeline for, 74types of, 65–71variation of viruses, 65, 71,72–73viruses of other kingdoms, 81expression of genome of, 117–138discussion, 137initial translation, 122mRNA synthesis, 123–125strategies, 125–137summary, 137–138virus entry and uncoating,119–122virus infection cycle, 117–119viruses of other kingdoms, 137in field, 245–268diagnosis, 245–260epidemiology and ecology,260–267summary, 268viruses of other kingdoms, 268in gene technology, 285–286,293–299DNA viruses as gene vectors,293–294RNA viruses as gene vectors,294–295vaccine production, 295–298viruses as sources of controlelements for transgenicplants, 295viruses in functional genomicsof plants, 298viruses in nanotechnology,298–299genome of, 105–116general properties of, 105–112introduction, 105organization of, 112–116proteins that replicate viralgenome, 108summary, 116viruses of other kingdoms, 116overview of, 23–42cytological effects, 32–38economic losses, 24histological changes of infectedplants, 30–32host range, 38–40introduction, 23–24macroscopic symptoms, 25–30summary, 40–41viruses in other kingdoms, 40transgenic protection against,285–286vector relationships with, 228–231,229b, 231–234, 239–240,261–264Plant virus diseases, agents thatresemble or alter, 43–62

defective and defective-interferingnucleic acids, 58–60, 73phytoplasma, 50–51satellite viruses and satellite RNAs,51–58, 203summary, 61viroids, 43–50viruses of other kingdoms, 60–61Planting date, virus spread and, 263Plant-to-plant movement, 221–244fungal transmission, 240–242,262, 277introduction, 223invertebrate transmission, 225–240plant material transmission,223–225summary, 242viruses of other kingdoms, 242Plasma membrane, 172b, 187–188Plasmids, 9Plasmodesmata, virus movementthrough, 168–169PLRV. See Potato leafroll virusPlum pox virus (PPV), 259, 289, 344Plus-sense single-stranded RNAviruses, 140–151Polerovirus, 76, 78b, 133, 212Pollen transmission, 225Polyclonal antisera, 249Polymerase chain reaction (PCR),virus diagnosis by, 257–259Polymerases, 108, 108bPolyproteins, overcoming eukaryotictranslation constraints with,126–128Poly(A) tail, 1363 0 Poly(A) tracts, 110Pospiviroidae, 44, 45–47, 47–48, 47bPosttranscriptional gene silencing. SeeRNA silencingPotato leafroll virus (PLRV), 28, 30, 76,235, 275, 337, 338Potato spindle tuber viroid (PSTVd), 43,45, 46f, 47, 47b, 49, 50Potato virus X (PVX)as gene vector, 294genome of, 106histological changes associatedwith, 32plant interaction of, 180, 202, 203plant resistance to, 193, 197,285, 290plant-to-plant movement of, 223INDEXprofile of, 339, 340symptoms of, 28, 29synergism of, 217btransmission of, 231, 232fPotato virus Y (PVY)conventional control of, 275diagnosis of, 259plant interaction of, 180, 202, 203plant resistance to, 288bprofile of, 341, 342, 343symptoms of, 28, 29synergism of, 217bPotexvirus, 53, 71, 143, 174Potyviral VPg, eukaryotic translationinitiation factor interaction with,195bPotyvirus, 4b, 33, 35, 36b, 68, 70b,126–128, 136, 144, 151b, 174, 179,195, 211, 231, 232–234, 291PPV. See Plum pox virusPR proteins. See Pathogenesis-relatedproteinsPrimary infection, in crops, 261–264Primary miRNA, in RNA silencing,209Profiles, of viruses, 24, 1–22. See alsospecific virus, e.g., Abutilon mosaicvirusProgrammed cell death, 200Promoter35S, 124, 125b, 290, 291bfor control of transgenic plants, 295of subgenomic RNAs, 130, 130bPropagation, virus transmission by,225Propagative viruses, 237–238Protection. See also Cross-protection;Resistancenucleic acid-based, 286, 287–290of plant, 277–280by plant pathogen, 277–278against plant viruses, transgenic,285–286protein-based, 286–287transgenic other forms of, 290Protein. See also Coat protein;Functional proteinshypersensitive response changesin, 197–198movement proteins, 169–174plant level virus-host interactionseffects on plant metabolismof, 181–182Rep proteins of geminivirus,156–158RNA interaction with, within smallicosahedral viruses, 100stabilization of, 101viraldiagnosis methods involving,249–256transgenic plant expression of,286–287Protein suppression, of RNAsilencing, 211–214Proteinases, 68–69Protein-based protection, 286–287Protists, virus transmission by,240–242PRSV. See Papaya ringspot virusPSbMV. See Pea seed-borne mosaic virusPseudo T = 3 symmetry, 100Pseudoknots, 109, 110fPsittacosis, 9PSTVd. See Potato spindle tuber viroidPVX. See Potato virus XPVY. See Potato virus YQQuarantine regulations, for viruscontrol, 273–274Quasiequivalence, of isometricviruses, 97Quasi-species, of viruses, 17–19371RR genes. See Resistance genesRaspberry ringspot virus, 240Rateof cell-to-cell movement, 176of plant virus evolution, 74of systemic movement, 179–180Raw materials, virus sequestration of,185–186RdRP. See RNA-dependent RNApolymeraseRDV. See Rice dwarf virusRead-through proteins, overcomingeukaryotic translationconstraints with, 134–135Recombinant viruses, 115Recombinationin evolution of plant virus, 75,76, 78b

372 INDEXRecombination (Continued)as fault of virus replication,159–161of transgenic plants, 292between viroids, 49of viruses, 65, 67Recovery, 29, 215–216Red clover necrotic mosaic virus, 130bReduction hypothesis, of virus origin,64bRegulations, quarantine, for viruscontrol, 273–274Release, of transgenic plant into field,290–293Removal, of infection sources, 271Reoviridae, 14, 35, 59t, 102–103, 106,108b, 123, 124, 131, 152, 186, 237Rep proteins, of geminivirus, 156–158Replicases, 67–68, 69f, 108b, 143–145Replicationof satellite DNAs, 57of satellite viruses and satelliteRNAs, 51, 52, 53synergistic effects of, 203of viral genome, 108of viroid RNA, 47–48of virus, 12, 13, 39–40, 139–164double-stranded RNAviruses, 152faults in, 158–161host functions used in, 139–140methods for studying, 140negative-sense single-strandedRNA viruses, 152plus-sense single-stranded RNAviruses, 140–151reverse transcribing viruses,153–156single-stranded DNA viruses,156–158summary, 163–164viruses of other kingdoms,161–162Replication pathway, ofcaulimoviruses, 154–155Replicative form (RF), 140–143,147, 156Replicative intermediate (RI),140–143, 147Repression, transcriptional andtranslational, 218Resistance. See also Pathogen-derivedresistanceconventional, 280–283field, 282insect biotypes overcoming, 276blocal acquired, 198of plantimmunity, 192–195, 281b, 282nonpermissive infection,195–200permissive infection, 200–202subliminal infection, 195systemic acquired, 198–200, 199btypes of, 281bResistance (R) genes, 192t, 193b, 194b,195–198, 280, 281, 282N gene of tobacco, 196, 196b,198–200, 285natural, 285–286Respiration, plant level virus-hostinteractions effects on, 184Response, to inoculation, 192–202immunity, 192–195, 281b, 282nonpermissive infection, 195–200permissive infection, 200–202subliminal infection, 195Results, of RNA silencing, 211Reticulate body, 9Retrotransposons, 9Retroviridae, 67, 108b, 154Reverse transcribing viruses,replication of, 153–156Reverse transcriptase, 9, 67, 154RF. See Replicative formRhabdoviridae, 14, 15, 35, 101–102,108b, 123, 152, 170b, 182, 237RI. See Replicative intermediateRibavirin, 271Ribgrass mosaic virus, 286Ribosome shunt, overcomingeukaryotic translationconstraints with, 134Ribozymes, 289Rice dwarf virus (RDV), 101, 102–103Rice grassy stunt virus, 276bRice ragged stunt virus (RRSV), 101,123, 152, 345, 346Rice stripe virus (RSV), 131, 347, 348Rice tungro bacilliform virus (RTBV), 20expression of, 131, 132b,133,134,137genome of, 106interactions of, 203mRNA synthesis in, 124profile of, 318, 319replication of, 153structure of, 100symptoms of, 28, 29, 30fRice tungro disease, 29, 71, 72f, 262Rice tungro spherical virus (RTSV), 29,30f, 203Rickettsiae, 9Ring spot diseases, 28RISC. See RNA-induced silencingcomplexRisks, of field release of transgenicplants, 290–293RNA. See also mRNA; Satellite RNAs;Subgenomic RNAdefective, 58–60defective-interfering, 58–60, 73diagnosis methods involving,256–260initial translation of, 122in plant reovirus assembly, 102–103recombination of, 160breplication ofin double-stranded RNAviruses, 152in negative-sense singlestrandedRNA viruses, 152in plus-sense single-strandedRNA viruses, 140–151in reverse transcribing viruses,153–156in viroids, 47–48within small icosahedralviruses, 100stabilization of, protein-RNA, 101TMV assembly origin in, 92, 93fTMV interaction with, 90, 91fof viral genome, 109–112, 112–114of viroids, 45–50RNA promoters, for control oftransgenic plants, 295RNA silencing, 207–220in animal and other viruses, 218evolutionary aspects of, 218introduction, 207–208mechanism of, 208–211overcoming of, 211–214summary, 218–219symptoms and, 214–218systemic silencing, 211transcriptional and translationalrepression, 218RNA virusesas gene vectors, 294–295protection from, 278, 279f

ecombination of, 159–161RNA silencing of, 208RNA-dependent DNA polymerase,108bRNA-dependent RNA polymerase(RdRP), 108b, 122, 123, 143,144–145, 144bRNA-induced silencing complex(RISC), 208, 210b, 211–214RNA-mediated protection, 288–289Rod, of TMV, assembly of, 92–94Rod-shaped viruses, architecture of,88–94introduction, 88–89TMV assembly, 92–94TMV structure, 89–91Rolling circle replication, 47–48, 55,57, 156RRSV. See Rice ragged stunt virusRSV. See Rice stripe virusRTBV. See Rice tungro bacilliform virusRTSV. See Rice tungro spherical virusRx1 gene, 285SSAR. See Systemic acquired resistancesat RNAs. See Satellite RNAsSatellite DNAs, 51–58Satellite RNAs (sat RNAs), 51–58B type, 53–57C type, 53–57D type, 53–57discussion of, 58molecular basis for symptommodulation, 56–57satellite-like RNAs, 55–56, 56–57transgenic protection by, 290Satellite tobacco necrosis virus (STNV),9, 12, 52–53, 98, 106Satellite viruses, 51–58, 203A-type, 52–53discussion of, 58protection from, 278, 279fSatellite-like RNAs, 55–56, 56–57Secondary infection, in crops,264–266Seed transmission, 224–225Seed, virus-free, 271Selection, hybrid, 115Selection pressures, for evolution,71–74controlling variation, 72–73INDEXby host plants, 74maximizing variation, 71Muller’s ratchet, 73niche adaption, 71role of, 73–74Sequencecomparison of, to gene withknown function, 116divergence or convergence of, 67integrated viral, recombinationand, 161for RNA-mediated protection, 289slippery, overcoming eukaryotictranslation constraints with,135–136of viroid RNA, 45–47Serological methods, of virus study,88, 249Serological reaction, components of,250bSerologically specific electronmicroscopy (SSEM), 255–256sgRNA. See Subgenomic RNAShort ORFs (sORFs), 134Shunt, translational, overcomingeukaryotic translationconstraints with, 134Silencing. See RNA silencingSilencing suppressors, other activitiesof, 217–218Sindbis virus, 70Single deletion D-RNAs, 60siRNA. See Small interfering RNASites, of replication, of plus-sense singlestrandedRNA viruses, 146–147Sizeof cell, of infected plants, 32of plant, virus effects on, 26of viral particles, methods fordetermination of, 86–88Slippery sequence, overcomingeukaryotic translationconstraints with, 135–136Small circular satellite RNAs, 53–57Small icosahedral viruses, 97–101nucleic acid arrangement within, 100stabilization of small isometricparticles, 101subunit structure of, 97virion structure of, 98–100Small interfering RNA (siRNA), inRNA silencing, 207, 208, 210f,211–214, 216373Small linear satellite RNAs, 53–57Snatching, of cap, 136–137Sobemoviruses, 54, 55, 75, 101, 238Soil-borne vectors, stopping of, 277Soil-borne wheat mosaic virus, genomeof, 113Solid-phase immune electronmicroscopy. See Serologicallyspecific electron microscopySonchus yellow net virus (SYNV),349, 350sORFs. See Short ORFsSouthern bean mosaic virus, 101,119–121, 180Southern blotting, 258bSouthern cowpea mosaic virus, 176Sowthistle yellow vein virus (SYVV),237, 238Species, of viruses, 14–15, 15–16,19–20Spherical viruses. See IsometricvirusesSpiroplasma, 50, 51Splicing, overcoming eukaryotictranslation constraints with, 131Spontaneous mutation, location of,114–115ssDNA, 125SSEM. See Serologically specificelectron microscopyStability, of virus particles, 247Stabilization, of small isometricparticles, 101Stabilizing bonds, of viruses, 88Start codon, non-AUG, overcomingeukaryotic translationconstraints with, 133STNV. See Satellite tobacco necrosisvirusStop-codon suppression, overcomingeukaryotic translationconstraints with, 134–135Stopping, of vectors, 274–277Strains, of viruses, 17–19Strawberry latent ringspot virus, 263fStriposomes, 120bStructure. See also Architectureof plant vein and phloem, 177bof plant viral genome, 112–114of RNA, within small icosahedralviruses, 100of small icosahedral virus, 97,98–100

374 INDEXStructure (Continued)of TMV, 89–91of viral nucleic acid, 109of viroid RNA, 45–47of virus, 86–88, 86bStudy methods, of virus replication,140Stylet-borne transmission. SeePersistent transmissionSubgenomic RNA (sgRNA)overcoming eukaryotic translationconstraints with, 129–131promoters of, 130, 130bSubliminal infection, 40, 195Subunitsof isometric virus, clustering of, 97of small icosahedral viruses,structure of, 97Sunn-hemp mosaic virus, 176Suppressionof gene silencing, 71of RNA silencing, 211–214of stop-codon, overcomingeukaryotic translationconstraints with, 134–135Suppressors, of silencing, otheractivities of, 217–218Symmetryicosahedral, 94–97T = 1, virus particle structureswith, 98–99T = 3, virus particle structureswith, 99–100T = 7, virus particle structureswith, 100Symptomsinduction of, 185–188modulation of, 56–57RNA silencing and, 214–218of viruses, macroscopic, 25–30virus-like, agents inducing, 30Synergistic effectsRNA silencing and, 216of transgenic plants, 292of virus replication, 203Synthesis, of mRNA, 123–125SYNV. See Sonchus yellow net virusSystemic acquired resistance (SAR),198–200, 199bSystemic host, 38Systemic host response, in permissiveinfection, 200Systemic infection, 167, 169f, 198Systemic movementform in which virus is transported,179movement in xylem, 180rate of, 179–180steps in, 176–179of virus in plants, 176–180Systemic silencing, 211Systemic symptomsdevelopmental abnormalities, 28–29genetic effects, 29leaf rolling, 28mosaic patterns and relatedsymptoms, 26–28necrotic diseases, 28plant size, 26recovery from disease, 29ring spot diseases, 28of viruses, 26–29wilting, 29yellow diseases, 28SYVV. See Sowthistle yellow vein virusTT = 1 virus particles, 98–99T = 3 virus particles, 99–100T = 7 virus particles, 100TBSV. See Tomato bushy stunt virusTCV. See Turnip crinkle virusTenuiviruses, 112, 237TEV. See Tobacco etch virusThrip, tospovirus transmissionby, 238Timeline, of plant virus evolution, 74Tissue print hybridization, 258bTMGMV. See Tobacco mild greenmosaic virusTMV. See Tobacco mosaic virusTobacco etch virus (TEV), 116, 136,151b, 176t, 201, 288b, 344Tobacco leaf curl virus, 4Tobacco mild green mosaic virus(TMGMV), 73Tobacco mosaic virus (TMV), 4, 8,73, 103assembly of, 92–94cell-to-cell movement proteins of,71chemical and biochemical studiesof, 86conventional control of, 274cytological effects of, 33, 34b, 35fdiagnosis of, 248bexpression of, 129, 129f, 134,136, 137as gene vector, 294genome of, 105, 109, 111f, 112, 115histological changes induced by,30, 32, 32fhost plant selection pressure on, 74host range of, 39microscopy of, 87plant interaction of, 169, 169f, 176t,179, 180, 182, 186, 200, 202,203–204plant resistance to, 193, 196, 196b,197–198, 198–200, 281, 282,285, 286, 287b, 289, 290plant-to-plant movement of,223, 225profile of, 351, 352replication of, 143, 148bstructure of, 89–91in transgenic plants, 295transmission of, 238uncoating of, 119–122vaccine production from, 298Tobacco, N gene of, 196, 196b,198–200, 285Tobacco necrosis virus, 30, 52–53Tobacco rattle virus (TRV), 33, 33f, 73,176, 240, 294, 300f, 301f, 353, 354Tobacco ringspot virus (TRSV), 28, 54,188, 239, 326, 327Tobacco vein mottling virus, 115Tobamovirus, 15, 66, 144, 196Tobravirus, 59t, 239–240ToLCV. See Tomato leaf curl virusTolerance, to plant virus, 281–282,282–283Tomato apical stunt viroid, 49Tomato black ring virus (BsatTBRV),53, 240Tomato bushy stunt virus (TBSV),103, 104expression of, 130bgenome of, 106profile of, 355, 356replication of, 143silencing suppressor of, 215bstructure of, 99–100transmission of, 241Tomato leaf curl virus (ToLCV), 57Tomato mosaic virus (ToMV), 197, 197t,274, 286

Tomato spotted wilt virus (TSWV), 39,131, 136, 238, 259, 288b, 359, 360Tomato yellow leaf curl disease, 76bTomato yellow leaf curl virus (TYLCV),328, 329Tombusviridae, 53, 59t, 60, 75, 144, 161,211, 215b, 241ToMV. See Tomato mosaic virusTospovirus, 59t, 102, 112, 123, 170b,174, 182, 237, 238Transactivation, overcomingeukaryotic translationconstraints with, 133–134Transcriptase, 9, 108bTranscription, 9Transcriptional gene silencing. SeeRNA silencingTranscriptional repression, 218Transgenes, in RNA silencing, 209Transgenic plantsCauliflower mosaic virus 35Spromoter in, 290, 291bcross-protection v. protection by,289expression of other viral proteinsby, 286field release of, 290–293gene expression in, 115viral coat protein expression by,286viruses and, 285–302pathogen-derived resistance,285, 286–293possible uses of plant viruses ingene technology, 285–286,293–299summary, 301transgenic protection againstplant viruses, 285–286viruses of other kingdoms, 300viruses as sources of controlelements of, 295Transgenic resistance. See PathogenderivedresistanceTranslationfor both viral and complementarystrands, 131control of, 136–137eukaryotic system constraints on,125–126, 126–136of viral genome, 122of viral mRNA, 126bTranslation enhancers, 295INDEXTranslation initiation factor, 147, 195bTranslation systems, in vitro, 113bTranslational repression, 218Translational shunt, overcomingeukaryotic translationconstraints with, 134Transmissionby aphids, 225–240, 261, 262, 264f,274–276by fungus, 240–242, 262, 277by invertebrates, 225–240beetles, 238–239nematode, 239–240, 262,263f, 277nonpersistent transmission byinsects, 228, 229b, 231–234persistent transmission byinsects, 228, 229b, 235–237plant virus and insectrelationships, 228–231via plant material, 223–225of viroids, 45in virus diagnosis, 247Transpiration, plant level virus-hostinteractions effects on, 184Transportation, of virus, 179Transposition, 9Transposons, 9, 209Trichoviruses, 1443 0 tRNA-like structures, 110, 111fTRSV. See Tobacco ringspot virusTRV. See Tobacco rattle virusTSWV. See Tomato spotted wilt virusTulipomania, 4, 4bTurnip crinkle virus (TCV), 54, 55b, 56,58, 161f, 211Turnip yellow mosaic virus (TYMV),26, 104cytological effects of, 34, 34fdiagnosis of, 248bexpression of, 136genome of, 106, 111fhost range of, 39plant interactions of, 181, 182, 186,187, 200profile of, 361, 362structure of, 99–100uncoating of, 121–122TYLCV. See Tomato yellow leaf curlvirusTymoviruses, 70b, 238TYMV. See Turnip yellow mosaicvirus375UUltraviolet spectra, of viruses, 248bUmbraviruses, 27, 29, 203, 237Uncoating, of virus, 119–1225 0 UTR, 137VVaccines, viruses in, 295–298Variation, of virus, 65, 71, 72–73Vector-resistant cultivars, 276Vectors. See also Gene vectorscirculative virus interaction with,235–237invertebrates, 225–240plant virus, in vaccines, 296plant virus relationships with,228–231, 229b, 231–234,239–240, 261–264stopping of, 274–277virus coevolution with, 80Vegetative propagation, virustransmission by, 225Vegetative stocks, virus-free, 271–272Vein, structure of, 177bVesicular stomatitis virus (VSV), 123VIGS. See Virus-induced genesilencingViral complexes, diseases caused by,29–30Viral gene productsfunctions of, 107–109proteins that facilitate host to hostmovement of viruses, 109proteins that facilitate viralmovement through host, 109proteins that overcome hostdefense systems, 109proteins that process viral geneproducts, 108–109Viral genes, 200–202activities of, 114–116sources of, 67–71Viral templates, of plus-sense singlestrandedRNA viruses, 140–143Virazole, 271Virion. See Virus particlesViroids, 43–50cell-to-cell movement of, 45, 175classification of, 44diagnostic procedures for, 50molecular basis for biologicalactivity, 50

376 INDEXViroids (Continued)pathology of, 44–45RNA properties of, 45–50siRNAs of, 214Virology, development of, 3–9Viroplasms, 34b, 35–36Virus. See also Plant virus;Recombinant virusesdefinition of, 1–22, 12balive status of, 13classification and nomenclature,13–14, 14–15, 15–16, 16–17,17–19, 19–20history, 3–9introduction, 3other plant pathogens v., 9–13summary, 21viruses of other kingdoms, 20emergence of new, 267in functional genomics ofplants, 298interactions between, 202–204in nanotechnology, 298–299origins of, 64–65pathogen interactions with, 203–204for presenting heterologouspeptides in vaccines, 296–298replication of, 139–164double-stranded RNAviruses, 152faults in, 158–161host functions used in, 139–140methods for studying, 140negative-sense single-strandedRNA viruses, 152plus-sense single-stranded RNAviruses, 140–151reverse transcribing viruses,153–156single-stranded DNA viruses,156–158summary, 163–164viruses of other kingdoms,161–162structure of, 86bmethods for determination of,86–88transgenic plants and, 285–302pathogen-derived resistance,285, 286–293possible uses of plant viruses ingene technology, 285–286,293–299summary, 301transgenic protection againstplant viruses, 285–286viruses of other kingdoms, 300transportation form of, 179ultraviolet spectra of, 248bvaccines produced from, 295–298variation in, 65, 71, 72–73Virus infection cycle, stages in,117–119Virus modules, 68bVirus particlesaccumulation of, 34–35architecture and assembly of,83–104complex isometric viruses, 101enveloped viruses, 101–102general considerations, 103–104icosahedral virus assembly,102–103introduction, 85–86isometric viruses, 94–97methods, 86–88rod-shaped viruses, 88–94small icosahedral viruses, 97–101summary, 104viruses of other kingdoms, 104physical properties of, diagnosismethods involving, 247–249Virus protein, genome linked (VPg),110, 195, 195bVirus replication cycle, 13Virus vectors, in vaccines, 296Virus-coded proteinases, 68–69Virus-encoded proteins, aggregatesof, 35Virus-first hypothesis, of virus origin,64bVirus-free seed, 271Virus-free vegetative stocks,271–272Virus-host interactions, plant level,165–190effects on plant metabolism,181–185movement and final distribution ofvirus, 167–181processes involved in symptominduction, 185–188summary, 188–189viruses of other kingdoms, 188Virus-induced gene silencing(VIGS), 298Virus-induced structures, incytoplasm, 34–36Virus-like symptoms, agentsinducing, 30Virusoids, 52Virus-plant interactionsmolecular level, 191–206host response to inoculation,192–202interactions between viruses,202–204introduction, 191summary, 204viruses of other kingdoms, 204RNA silencing, 207–220in animal and other viruses, 218evolutionary aspects of, 218introduction, 207–208mechanism of, 208–211overcoming of, 211–214summary, 218–219symptoms and, 214–218systemic silencing, 211transcriptional and translationalrepression, 218Vitiviruses, replication of, 144VPg. See Virus protein, genomelinkedVSV. See Vesicular stomatitis virusWWatermelon mosaic virus-2, 281Weather conditions, virus spreadand, 265, 265fWheat dwarf virus, 259Wheat streak mosaic virus, 272, 273fWhitefly, virus transmission by,225–240Wilting, 29Wound tumor virus (WTV), 29,32, 186WTV. See Wound tumor virusXX-ray crystallography, of viruses, 87Xylem, movement in, 180YYellow diseases, 28