Plant Viruses and Their Classification SUMMARY Classification of biological organisms forms a basic framework for much of their further studies. The classification of viruses poses several problems because of their non-sexual reproduction and inherent variation. The International Committee on the Taxonomy of Viruses has formalized and oversees the classification of viruses, but some problems still remain/ The taxa and rules used in virus classification and nomenclature are described in this chapter and the difficulties in delimiting viral species and strains are discussed in detail. The main focus is on viruses of Angiosperms, but viruses that infect Gymnosperms, Pteridophytes, Algae and Fungi are also described. It is important to purify viruses to determine characters which are used in classification, and methods used in plant virus purification are discussed.
There are numerous classification systems for living organisms. For the purposes of this book I will use what is termed the kingdom Plantae (Cavalier-Smith, 2004) and divide it into informal groups, angiosperms (flowering plants), gymnosperms (conifers), pteridophytes (ferns), bryophytes (mosses and liverworts), and green algae; I will also consider fungi. Viruses are found in most of these groups, but those in angiosperms, especially crop plants, have attracted most attention. In this chapter I will discuss how viruses are classified. I will also describe some features of viruses of lower plants and fungi but will give details of the individual taxa of viruses of angiosperms in Appendix A.
I. CLASSIFICATION OF VIRUSES A. Historical Aspects In all studies of natural objects, humans have an innate desire to name and to classify. Virologists are no exception. Virus classification, as with all other classifications, arranges objects showing similar properties into groups and, even though this is a totally artificial and humandriven activity without any natural base, it does have certain properties: It gives a structured arrangement of the organisms so that the human mind can comprehend them more easily. ● It helps with communication between virologists. ●
Plant Virology, Fifth Edition. © 2014 2012 Elsevier Inc. All rights reserved.
It enables properties of new viruses to be predicted. It could reveal possible evolutionary relationships.
In theory, it is possible to consider the problems of naming and classifying viruses as separate issues. In practice, however, naming soon comes to involve classification. Early workers generally gave a virus a name derived from the host plant in which it was found together with the most conspicuous disease symptom, for example, TMV1. Viruses were at first thought of as stable entities, and each disease condition in a particular host species was considered to be due to a different virus. However, by the early 1930s three important facts began to be recognized: (i) an individual virus can exist as different strains, which may cause very different symptoms in the same host plant; (ii) different viruses may cause very similar symptoms on the same host plant; and (iii) some diseases may be caused by a mixture of two unrelated viruses. Johnson (1927) and in subsequent work stressed the need for using some criteria other than disease symptoms and host plants for identifying viruses. He suggested that a virus should be named by adding the word virus and a number to the common name for the host in which it was first found, for example, tobacco virus 1 for TMV. Johnson and Hoggan (1935) compiled a descriptive key based on five characters: modes of transmission, natural or differential hosts, longevity in vitro, thermal death point, and distinctive or specific symptoms. About 50 viruses were identified and placed in groups. Smith (1937) outlined a scheme in which the known viruses or virus diseases were divided into 51 groups. Viruses were named and grouped according to the generic name of the host in which they were first found. Successive members in a group were given a number. For example, TMV was Nicotiana virus 1, and there were 15 viruses in the Nicotiana virus group. Viruses that were quite unrelated in their basic properties were put in the same group. Although Smith’s list served for a time as a useful catalog of the known viruses, it could not be regarded as a classification. Holmes (1939) published a classification based primarily on host reactions and methods of transmission. He used a Latin binomial–trinomial system of naming. 1
Acronyms of virus names are shown in Appendix D.
For example, TMV became Marmor tabaci, Holmes (Marmor meaning marble in Latin). His classification was based on diseases rather than the viruses, and thus 53 of the 89 plant viruses considered by Holmes fell in the genus Marmor, which contained viruses, known even at that time to differ widely in their properties. Between 1940 and 1966, various schemes were proposed either for plant viruses only or for all viruses. None of these schemes was adopted by any significant number of virologists. It became increasingly apparent that a generally acceptable system of nomenclature and classification could be developed only through international cooperation and agreement, with the opinions of a majority of working virologists being taken into account. At the International Congress for Microbiology held in Moscow in 1966, the first meeting of the International Committee for the Nomenclature of Viruses was held, consisting of 43 people representing microbiological societies of many countries. An organization was set up for developing an internationally agreed taxonomy and nomenclature for all viruses. Rules for the nomenclature of viruses were laid down. The subsequent development of the organization, now known as the International Committee for Taxonomy of Viruses (ICTV), has been summarized (Matthews, 1983, 1985a,b); the ICTV has presented nine reports (Wildy, 1971; Fenner, 1976; Matthews, 1979, 1982; Francki et al., 1991; Murphy et al., 1995; Van Regenmortel et al., 2000; Fauquet et al., 2005; King et al., 2012) and published intermediate reports in the Archives of Virology. In recent years, ICTV has also had a Web presence providing lists of the currently recognized taxa and information on the Executive Committee, Subcommittees, and Study Groups. Templates and other information to assist in writing and submitting taxonomic proposals have also been provided. Since 2008, the Virus Data subcommittee has overseen the development and maintenance of an official ICTV Web site (http://www.ictvonline.org/) that now provides a central point of reference for all ICTV matters. A separate Web site (http://talk.ictvonline.org/) is used to host taxonomic proposals and allows for comment and discussion to which all virologists are invited to contribute. The main features of the agreed nomenclature and taxonomy as they apply to plant viruses are considered in the following sections. Virus taxonomy differs from other types of biological classification because ICTV not only regulates a Code of Nomenclature but also considers and approves the creation of virus taxa (currently orders, families, subfamilies, genera, and species). Priority of publication is not the determining factor.
B. Systems for Classification Organisms may be classified in two general ways. One is the classic monothetic hierarchical system applied by
Linnaeus to plants and animals. This is a logical system in which decisions are made as to the relative importance of different properties, and are then used to place a taxon in a particular phylum, order, family, genus, and so on. Maurin et al. (1984) proposed a classification system of this sort that embraces viruses infecting all kinds of hosts. While such systems are convenient to set up and use, there is, as of yet, no sound basis for them as far as viruses are concerned. The major problem with any such system is that we have no scientific basis for rating the relative importance of all the different characters involved. For example, is the kind of nucleic acid (DNA or RNA) more important than the presence or absence of a bounding lipoprotein membrane? Is the particle symmetry of a small RNA virus (helical rod or icosahedral shell) more important than some aspect of genome strategy during virus replication? An alternative to the hierarchical system was proposed by Adanson (1763). He considered that taxa were best derived by considering all available characters. He made a series of separate classifications, each based on a single character, and then examined how many of these characters divided the species in the same way. This gave divisions based on the largest number of correlated characters. The method is laborious and has not been much used until recently because up to 60 or more equally weighted independent qualitative characters are needed to give satisfactory division (Sneath, 1962; Harrison et al., 1971) (Table 2.1). Although the availability of computers has renewed interest in this kind of classification so far, they have not really been used for this to any great extent. In practice, some weighting of characters is inevitably involved even if it is limited to decisions as to which characters to include in consideration and which to leave out. The main advantages of an Adansonian classification at present may be to confirm groupings arrived at in other ways and to suggest possibly unsuspected relationships that can then be checked by further experimental work. The rapid accumulation of complete nucleotide sequence information for many viruses in a range of different families and groups is having a profound influence on virus taxonomy in at least three ways: (i) most of the virus families delineated by the ICTV, mainly on morphological grounds, can now be seen to represent clusters of viruses with a relatively close evolutionary origin; (ii) with the discovery of previously unsuspected genetic similarities between viruses infecting different host groups, the unity of virology is now quite apparent. The stalling approach of some plant virologists to the application of families, genera, and species to plant viruses is no longer tenable; (iii) genotypic information is now more important for many aspects of virus taxonomy than phenotypic characters. However, there are some limitations to the use of genotype (sequence) data alone. An important one is that with the present state of knowledge it is very difficult, or
Chapter | 2 Plant Viruses and Their Classification
TABLE 2.1 Descriptors Used in Virus Taxonomy
TABLE 2.1 (Continued) Descriptors Used in Virus Taxonomy
I. Virion properties
A. Morphology properties of virions
1. Presence or absence
II. Genome organization and replication
3. Presence or absence of an envelope or peplomers
1. Genome organization
4. Capsomeric symmetry and structure
2. Strategy of replication of nucleic acid
B. Physical properties of virions
3. Characteristics of transcription
1. Molecular mass
4. Characteristics of translation and post-translational processing
2. Buoyant density
5. Sites of accumulation of virion proteins, site of assembly, site of maturation and release
3. Sedimentation coefficient 4. pH stability
6. Cytopathology, inclusion body formation
5. Thermal stability
III. Antigenic properties
6. Cation (Mg2+, Mn2+, Ca2+) stability
1. Serological relationships
7. Solvent stability
2. Mapping epitopes
8. Detergent stability
IV. Biological properties
9. Radiation stability
1. Host range, natural and experimental
C. Properties of the genome
2. Pathogenicity, association with disease
1. Type of nucleic acid, DNA or RNA
2. Strandedness: single stranded or double stranded
3. Linear or circular
4. Sense: positive, negative, or ambisense
5. Number of segments
6. Size of genome or genome segments
7. Presence or absence and type of 5′ terminal cap
8. P resence or absence of 5′ terminal covalently linked polypeptide
9. P resence or absence of 3′ terminal poly(A) tract (or other specific tract)
3. Tissue tropisms, pathology, histopathology 4. Mode of transmission in nature 5. Vector relationships 6. Geographic distribution
10. Nucleotide sequence comparisons D. Properties of proteins 1. Number 2. Size 3. F unctional activities (especially virion transcriptase, virion reverse transcriptase, virion hemagglutinin, virion neuraminidase, virion fusion protein) 4. Amino acid sequence comparisons E. Lipids 1. Presence or absence 2. Nature (Continued)
impossible, to predict phenotypic properties of a virus based on sequence data alone. For example, if we have two viral nucleotide sequences differing in a nucleotide at a single site, we could not, in most cases, deduce from this information alone that one led to mosaic disease and the other to lethal necrosis in the same host. However, with increasing understanding of gene function, we could now decide from the nucleotide sequences alone which of two rhabdoviruses replicated in plants and insects and which in vertebrates and insects. A further limitation is the increasing recognition of recombination between the genomes of viruses from two (or more) different taxa (see Chapter 7, Section IX, B) which leads to the problem of how to place the recombinant. Any classification of viruses should be based not only on evolutionary history, as far as this can be determined from the genotype, but should also be useful in a practical sense. Most of the phenotypic characters used today in virus classification will remain important even when the nucleotide sequences of most viral genomes have been determined.
C. Families, Genera, and Species At a meeting in Mexico City in 1970, the ICTV (then called the ICNV) approved the first taxa for viruses (Matthews, 1983). There were two families and type genera for these, plus 22 genera not placed in families for viruses infecting vertebrates, invertebrates, or bacteria. Sixteen taxa designated as groups were approved for viruses infecting plants. While other virologists subsequently moved quite rapidly to develop a system of families and genera for the viruses, plant virologists clung to the notion of groups. Some plant virologists had especial difficulty with the species concept believing that it should not be applied to viruses. Their main reason is that because viruses reproduce asexually the criterion of reproductive isolation cannot be used as a basis for defining virus species (Milne, 1988). The main building block of a biological classification is the species. The pragmatic view of Davis and Heywood (1963) in discussing angiosperm taxonomy was that: “There is no universally correct definition (of a species) and progress in understanding the species problem will only be reached if we concentrate on the problem of what we shall treat as a species for any particular purpose.” The species concept applied to plants is also discussed by Wagner (1984). In day-to-day practice, virologists use the concept of a “virus” as being a group of fairly closely related strains, variants, or pathovars. A virus defined in this way is essentially a species in the sense suggested by Davis and Heywood, and defined by the ICTV. In 1991, the ICTV accepted the concept that viruses exist as species, adopting the following definition (Van Regenmortel, 1990): A viral species is a polythetic class of viruses that constitutes a replicating lineage and occupies a particular ecological niche.
A “polythetic class” is one whose members have several properties in common, although they do not necessarily all share a single defining common property. Thus, members of a virus species are defined collectively by a consensus group of properties in contrast to higher viral taxa which are “universal” classes which are defined by properties that are necessary for membership (Ball, 2005). To illustrate the controversies in classification the use of the word “polythetic” (which is not widely understood) has been challenged (Gibbs and Gibbs, 2006) followed by a repost (Van Regenmortel et al., 2013). Gibbs and Ohshima (2010) noted that the search for a way to define species has followed two paths, the semantic way illustrated by the definition above accepted by the ICTV and the pragmatic of Adams et al. (2005) who used molecular data tempered by evolutionary considerations. Formal acceptance of a new species can only be done through the ICTV, but it has been suggested that a combination of the semantic and pragmatic approaches, combined with evolutionary insight
may give a more robust definition (Gibbs and Ohshima, 2010). The species has formed the basis of modern virus classification being firmed up in subsequent ICTV reports, especially the seventh and subsequent reports in which a “List of species-demarcating criteria” is provided for each genus. This enables viruses to be differentiated as species and tentative species. Guidelines to the demarcation of a virus species are given in Van Regenmortel et al. (1997). With the species forming the basis of the classification system, they can be grouped into other taxa on various criteria. To date, the taxonomic levels of order, family, and genus have been defined by the ICTV (Table 2.2) and it is likely that there will be pressure for further higher and intermediate taxa. For example, Hull (1999) argued that the development of an overall classification for viruses and elements that replicate by reverse transcription necessitates creating the taxa of class and suborder. As noted by Ball (2005), other groupings such as superfamily, clade, and quasispecies may communicate useful information or important concepts but are not recognized by the ICTV. A detailed discussion of virus classification, the currently accepted taxa, and how the ICTV operates is given in Fauquet (1999) and by Van Regenmortel et al. (2000), Fauquet et al. (2005), and King et al. (2012).
1. Virus Species The delineation of the kinds of virus that exist in nature, that is to say virus species, is a practical necessity, especially for diagnostic purposes relating to the control of virus diseases. The Association of Applied Biologists (AAB) has produced a series of over 400 descriptions of plant viruses and plant virus groups and families. Each description was written by a recognized expert for the virus or group (genus) in question and the editors use common-sense guidelines devised by themselves to decide whether a virus described in the literature is a new virus, or merely a strain of a virus that has already been described. When they published a new virus description, they were, in effect, delineating a new species of virus. However, although the descriptions are being updated online (www.dpvweb. net), the ICTV ultimately determines the authenticity of each new species. The criteria for delineating virus species are listed in Table 2.2 Viruses are now recognized as “species” or “tentative species” which are viruses that have not yet been sufficiently characterized to ensure that they are distinct and not strains of an existing virus or do not have the full characteristics of the genus to which they have been assigned. Of the 1325 plant virus species listed in the ICTV ninth report and those ratified by ICTV members in 2012 (Adams and Carstens, 2012), 1016 are true species and 309 tentative species.
Chapter | 2 Plant Viruses and Their Classification
TABLE 2.2 Criteria Demarcating Different Virus Taxa I. Order Common properties between several families including:
considered a collection of quasispecies. The concept of quasispecies is discussed in more detail in Section III. Various characters are considered in designating a strain. These are described in more detail in relation to virus variation in Sections IV and V.
Biochemical composition Virus replication strategy
3. Naming of Viruses (Species)
Particle structure (to some extent)
Questions of virus nomenclature have generated more heat over the years than the much more practically important problems of how to delineate distinct virus species. At an early stage in the development of modern viral taxonomy, a proposal was made to use cryptograms to add precision to vernacular names of viruses (Gibbs et al., 1966). The proposal was not widely adopted, except for a time, among some plant virologists. Some virologists favor using the English vernacular name as the official species name. Using part of a widely known vernacular name as the official species name may frequently be a very suitable solution, but it could not always apply (e.g., with newly discovered viruses). Other virologists have favored serial numbering for viruses (species). The experience of other groups of microbiologists is that, while numbering or lettering systems are easy to set up in the first instance, they lead to chaos as time passes and changes have to be made in taxonomic groupings. The ICTV rules about naming a species are: (i) a species name shall consist of as few words as practicable but be distinct from names of other taxa; (ii) species names shall not consist only of a host name and the word “virus”; (iii) a species name must provide an appropriately unambiguous identification of the species; (iv) numbers, letters, or combinations thereof may be used as species epithets where such numbers and letters are already widely used (e.g., Potato virus Y). However, newly designated serial numbers, letters, or combinations thereof are not acceptable alone as species epithets. If a number or letter series is in existence it may be continued. The idea of Latinized binomial names for viruses was supported by the ICTV for many years but never implemented for any viruses. The proposal for Latinization (Kuhn et al., 2010) has now been modified to one for the general use of non-Latin binomials, e.g., tobacco mosaic Tobamovirus, Fiji disease Fijivirus, and lettuce necrotic yellows rhabdovirus (Van Regenmortel et al., 2010); this proposal has been rejected by the ICTV Executive Committee.
General genome organization II. Family Common properties between several genera including: Biochemical composition Virus replication strategy Nature of particle structure Genome organization III. Subfamily Common properties between several genera but used only to solve complex hierarchical problems IV. Genus Common properties within a genus including: Virus replication strategy Genome size, organization, and/or number of segments Sequence homologies (hybridization properties) Vector transmission V. Species Common properties within a species including: Genome rearrangement Sequence homologies Serological relationships Vector transmission Host range Pathogenicity Tissue tropism Geographical distribution Modified from Fauquet (1999).
2. Virus Strains and Isolates A common problem is to determine whether a new virus is truly a new species or a strain of an existing species. Conversely, what was considered to be a strain may, on further investigation, turn out to be a distinct species. This is due to the population structure of viruses which, because of continuous production of errors in replication, can be
4. Acronyms or Abbreviations Abbreviations of virus names have been used for many years to make the literature more easy to read and more succinct to present. The abbreviation is usually in the form of an acronym using the initial letters of each word
in the virus name. As the designation of the acronym was by the author of the paper, it was leading to much overlap and confusion. For instance, AMV was used to designate Alfalfa mosaic virus and Arabis mosaic virus and could also justifiably be used for a number of viruses such as Abutilon mosaic virus, Agropyron mosaic virus, Alpina mosaic virus, Alstromeria mosaic virus, Alternantha mosaic virus, Aneilema mosaic virus, or Anthoxanthum mosaic virus. Therefore, in 1991, the Plant Virus Section of the ICTV initiated a rationalization of plant virus acronyms (Hull et al., 1991) and has subsequently updated the list regularly in subsequent reports. The designation of the abbreviations is based on the following principles: Abbreviations should be as simple as possible. An abbreviation must not duplicate any other previously coined and still in current usage. ● The word “virus” in a name is abbreviated as “V”. ● The word “viroid” in a name is abbreviated as “Vd”. ● ●
A set of guidelines is laid out in Fauquet and Mayo (1999). Although these, and the acronyms derived from them, are not officially sanctioned by the ICTV, the acronyms are used in the seventh and subsequent reports. The rules for plant virus abbreviations are outlined in Box 2.1.
D. Plant Virus Genera, Subfamilies, Families, and Orders There is no formal definition for a genus, but it is usually considered as “a population of virus species that share common characteristics and are different from other populations of species.” Each genus contains a “type species” which is defined (Mayo et al., 2002): A type species is a species whose name is linked to the use of a particular genus name. The genus so typified will always contain the type species
The formal criteria for demarcating a genus are listed in Table 2.2; the practicalities of criteria are discussed in Section II. Currently, there are 92 genera of plant viruses recognized of which 10 are “unassigned” to a specific family (Adams and Carstens, 2012: King et al., 2012). Genera are named either after the type species, e.g., Caulimovirus after Cauliflower mosaic virus or are given a descriptive name, often from a Greek or Latin word, for a major feature of the genus, e.g., Closterovirus from the Greek κλωστηρ (kloster) which is spindle or thread, descriptive of the virus particle shape and Geminivirus from the Latin geminus meaning twins to describe the particles. Similarly, genera are grouped together into families and subfamilies on common characteristics (Table 2.2). There are 21 families which have plant virus members; those of the family Reoviridae are in two subfamilies, and there is one subfamily in the family Secoviridae
BOX 2.1 Rules for Virus Abbreviations or Acronyms Abbreviations should be as simple as possible. An abbreviation must not duplicate any other previously coined and still in current usage. ● 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” and “symptomless” as “SL”. ● Abbreviations for single words should not normally exceed two letters. ● Where a particular combination of letters has been adopted for a particular plant, subsequent abbreviations for viruses of that host should use the same combination. ● The second (or third) letter of a host plant abbreviation is in lower case, e.g., Ab for Abutilon. ● When several viruses have the same name and are differentiated by a number, the abbreviation will have a hyphen between the letters and number, e.g., Plantain virus 6 is abbreviated as “PlV-6.” ● When viruses end by a letter, this letter is added to the end of the abbreviation without a hyphen, e.g., Potato virus X is abbreviated as “PVX”. ● When viruses are distinguished by their geographical location, a minimum number of letters (two or three) are added to the abbreviation using a hyphen, e.g., Tomato yellow leaf curl virus from Thailand is “TYLCV-Th”. ● When a virus name comprises a disease name and the word “associated virus,” these are abbreviated as “aV”, e.g., Grapevine leafroll associated virus 2 is abbreviated as “GLRaV-2”. A set of guidelines is laid out in Fauquet and Mayo (1999). Although these, and the acronyms derived from them, are not officially sanctioned by the ICTV, the acronyms are used in the ICTV reports. ● ●
(King et al., 2012). The family is either named after the type member genus, e.g., Caulimoviridae names after the genus Caulimovirus, or given a descriptive name, as with the genus, for a major feature of the family, e.g., Geminiviridae, descriptive of the virus particles. There are three orders of viruses that contain families in which there are plant virus genera (King et al., 2012). The Mononegavirales contains two families of plant rhabdoviruses, the Picornavirales and the Tymovirales contain two subfamilies and three families of plant viruses, respectively. The current classification of plant viruses is given in Appendix A and shown diagrammatically in Figure 2.1.
E. Use of Virus Names The ICTV sets rules, which are regularly revised, on virus nomenclature and the orthography of taxonomic names
Chapter | 2 Plant Viruses and Their Classification
Virus taxa infecting plants (part 2)
Virus taxa infecting plants (part 1)
Geminiviridae Curtovirus Mastrevirus Topocuvirus
Caulimovirus Cavemovirus Petuvirus Soymovirus
Closterovirus / Ampelovirus
Badnavirus Tungrovirus Crinivirus
Babuvirus: 6 virions Nanovirus: 8 virions
ssRNA (-) and (+/–)
All other genera Tymovirales
Comovirinae Cheravirus Sadwavirus Torradovirus
Bromoviridae Alfamovirus llarvirus Oleavirus Anulavirus Bromovirus Cucumovirus llarvirus
Luteoviridae Polemovirus Sobemovirus Tombusviridae
Ourmiavirus Umbravirus 100 nm
FIGURE 2.1 Taxa of viruses infecting plants. From King et al. (2012), with permission of the publishers.
(Pringle, 1998; Van Regenmortel et al., 2000; King et al., 2012). The last word of a species is “virus” and suffix (ending) for a genus is “…virus,” for a subfamily is “… virinae,” for a family is “… viridae,” and for an order is “… virales.” In formal taxonomic usage, the virus order, family, subfamily, genus, and species names are printed in italics (or underlined) with the first letter being capitalized; other words in species names are not capitalized unless they are proper nouns or parts of proper nouns. Also, in formal use, the name of the taxon should precede the name being used, e.g., “the family Caulimoviridae,” “the genus Mastrevirus,” and “the species Potato virus Y.” In informal use, the family, subfamily, genus, and species names are written in lower case Roman script, the taxon does not include the formal suffix and the taxonomic unit follows the name being used, e.g., “the caulimovirus family,” “the mastrevirus genus,” and the “potato virus Y species.” In even less formal circumstances, but widely used, the taxonomic unit is omitted and the taxon for higher taxa can be in the plural, e.g., “caulimoviruses,” “mastreviruses,” and “potato virus Y.” Suggestions for taxon-specific suffixes in informal usage are discussed in Vetten and Haenni (2010). As noted above (Section I, C, 3), there is a proposal for
using non-Latinized binomials, e.g., tobacco mosaic tobamovirus, Fiji disease fijivirus, and lettuce necrotic yellows rhabdovirus (Van Regenmortel et al., 2010) but this is currently not accepted by the ICTV. However, this orthography, especially for species, has led to confusion amongst many virologists who have difficulty in distinguishing between the species name which is an abstract concept and the virus (vernacular) name which is a real tangible object that can cause disease, be transmitted, be purified, etc. (Van Regenmortel, 2006). One major cause of the problem is that the species and vernacular name are often the same. This difficulty is discussed in depth by Kuhn and Jahrling (2010). Informal usage arises from practicalities and can lead to the adoption of more formal use. For instance, the genus Badnavirus was not adopted in 1991 but was used widely in the literature and was adopted in the 1995 ICTV report. However, the year 2000 report limited its use to certain DNA viruses with bacilliform particles excluding RTBV. As will be apparent in this book, it is necessary to distinguish the reverse transcribing DNA viruses that have isometric particles from those that have bacilliform particles; thus there is the informal usage with “caulimoviruses” for the former and “badnaviruses” for the latter.
F. Endogenous Plant Viral Sequences The discovery of plant pararetroviral [viruses that replicate by reverse transcription (see Chapter 7, Section VII)] sequences integrated into the host genome [termed endogenous pararetroviral sequences (EPRVs)] raises several questions concerning taxonomy and nomenclature. These integrated sequences are usually fragmented and rearranged, but in some cases the sequences are such that episomal infection by viable virus can be activated (see Chapter 7 Section IX, B, 3). The major questions are as follows: Should these sequences be classified as viruses? The replication of most retroelements have an endogenous integrated phase in their replication cycle and many of these are classified as viruses, e.g., the family Retroviridae and the families Pseudoviridae and Metaviridae are classified as viruses. This raises the question as to whether the fragmented, rearranged sequences should be included in the classification as most of these would be inactive, not capable of causing an episomal infection. Many of the Pseudoviridae and Metaviridae species are inactive as are some members of the subfamily Orthoretrovirinae. The ICTV considers that retrotransposons are viruses for the purpose of classification and nomenclature (King et al., 2012) and thus, the plant EPRVs should also be treated as viruses. What information should be used in classification? The information available on EPRVs is DNA sequences, albeit from fragments of a complete viral genome. The sequence of the polymerase (reverse transcriptase and RNaseH) which has several conserved motifs is used in phylogenetic analyses to develop a universal classification system for retroelements (Van Regenmortel et al., 1990; Xiong and Eikbush, 1990). This raises the question as to whether the nucleotide or derived amino acid sequence should be used; the amino acid sequence would be affected by mutations in the integrated sequence. How should EPRVs be named? As noted in Section I, A, the classification and naming of a virus should communicate its properties to other people. The two important features about EPRVs are that their name should indicate that they are integrated sequences and whether they are capable of being activated to give episomal infection. The classification of EPRV is the subject of much discussion as shown in papers by Staginnus et al. (2009) and Geering et al. (2010). Staginnus et al. (2009) proposed a system that named EPRVs on whether or not they had an episomal form and a functional or nonfunctional EPRV. The Geering et al. (2010) proposal was based on a phylogenetic analysis of the derived amino acids from the EPRV sequence corresponding to the CaMV polymerase from amino acid L269 to R672 which contains seven conserved motifs. Endogenous viral sequences that could not be conceptually translated because of mutations were then added to the
alignment and selectively realigned. This analysis (Figure 2.2) identified groups which mostly fitted the grouping of episomal viruses. Endogenous forms were identified by the prefix “e,” e.g., eTVCV (endogenous Tobacco vein clearing virus) but replication competent forms were not distinguished from incompetent forms.
G. Subviral Agents (King et al., 2012) The ICTV rules concerning the classification of viruses also apply to the classification of viroids. The taxa are designated for species the suffix “viroid,” for genera the suffix “-viroinae,” and for families the suffix “-viroidae.” Satellites are not classified as viruses but are assigned an arbitrary classification as seems useful to workers in the particular fields.
II. CRITERIA USED FOR CLASSIFYING VIRUSES There are two interconnected problems involved in attempting to classify viruses. First, related viruses must be placed in genera and higher taxa. For this purpose, the more stable properties of the virus, such as amount and kind of nucleic acid, particle morphology, and genome strategy, are most useful. The second problem is to be able to distinguish between related viruses and give some assessment of degrees of relationship within a group. Properties of the virus for which there are many variants are more useful for this purpose. These include symptoms, host range, nucleotide sequence of specific regions of the genome, and amino acid composition of specific proteins, usually the coat protein (CP). Certain properties, such as serological specificity and amino acid sequences may be useful both for defining groups and for distinguishing viruses within groups. Two further problems must be borne in mind when considering the criteria to be used for virus classification. As noted above, one is the problem of weighting— the relative importance of one character as compared to another. Some weighting of characters is inevitable, whatever system is used to place viruses in taxa. The second problem concerns the extent to which a difference in one character depends on a difference in another. For example, a single-base change in the nucleic acid may result in an amino acid replacement in the CP, which in turn alters serological specificity and electrophoretic mobility of the virus. The same base change in another position in a codon may not affect the gene product at all. Many of the earlier criteria were related to phenotypic characters. As the complete nucleotide sequences of more and more viruses are being determined, virus genotype is becoming increasingly important. What weight is to be
Chapter | 2 Plant Viruses and Their Classification
(B) 100 100
76 D10 D4
100 D4 51 61
98 80 D15 95
78 100 D45 100 D45
SceTy3v DineGypV PVCV RTBV-type RTBV-WB Osat V-A Osat V-B Osat V-C SCBMorV SCBlmV ComYMV DBV CiyMV CSSV TaBV BSOLV KTSV BSMysV BSGfV BRRV SbCMV PCSV SVBV CERV CaMV HRLV DMV-Holland MiMV DMV-D10 FMV CsVMV TVCV LycEPRV-Lh9 LycEPRV-Lh1 LycEPRV-Lh5 LycEPRV-Lc2 LycEPRV-Lh6 LycEPRV-Lh4 LycEPRV-Lh8 LycEPRV-Lh7 LycEPRV-Le1 NtEPRV-V3 NtEPRV-V14 NtEPRV-E21 NtEPRV-E22 NtEPRV-V9 NtEPRV-V4 NtEPRV NtEPRV-V6
SceTy3v DineGypV PVCV
SbCMV BRRV PCSV SVBV CaMV 100 100 HRLV 100 CERV 100 99 DMV-Holland MiMV 100 DMV-D10 98 EMV 100 RTBV-type RTBV-WB 100 Osat V-C Osat V-A 100 Osat V-B TaBv 100 CSSV CiYMV DBV 100 100
CsVMA LycEPRV-Lh9 LycEPRV-Lht 100 LycEPRV-Lh5 100 LycEPRV-Lh7 LycEPRV-Le1 100 100 TVCV LycEPRV-Le2 LycEPRV-Lh8 LycEPRV-Lh6 LycEPRV-Lh4 100 NtEPRV-V9 95 NtEPRV-V4 NtEPRV-E22 96 NtEPRV 100 NtEPRV-V6 NtEPRV-V3 100 NtEPRV-V14 96 NtEPRV-V21 100
ComYMV SCBMorV SCBImV BSMysV BSOLV KTSV 98 BSGfV
FIGURE 2.2 Phylogenetic trees of endogenous viruses of the family Caulimoviridae with statistical support using different methods: (A) Parsimony analysis without indels: strict consensus of 23 equally parsimonious trees (Consistency Index 0.2923, Retention Index 0.5636, and Length 10.389) with bootstrap percentages above nodes and Decay Indices (DP) below key nodes; (B) Bayesian inference of sequence data partitioned into RT–RNaseH domains, with indels: a randomly selected phylogram, with posterior probabilities above nodes. Abbreviations and sources of sequence are listed in Geering et al. (2010). From Geering et al. (2010), with permission of the publishers.
given to sequence data in virus classification? A geneticist may say that the evolutionary history of a virus and therefore its relationships with other viruses are completely defined, as far as we will be able to know it, by its nucleotide sequence. This is certainly true in a theoretical sense, but there are also practical aspects to be considered as discussed in Section I, B. The overall properties used in classifying a virus are listed in Table 2.1. Properties that are useful for characterizing strains of a virus are discussed in Chapter 8, Section III.
A. Structure of the Virus Particle The importance of virus structure in the classification of viruses is summarized in Figure 2.1. With isometric viruses, particle morphology, as revealed by electron microscopy, has not proved as generally useful as with the rods. This is mainly because many isometric particles lie in the same size range (about 25–30 nm in diameter) and are of similar appearance unless preparations and photographs
of high quality are obtained (Hatta and Francki, 1984). Where detailed knowledge of symmetry and arrangement of subunits has been obtained by X-ray analysis or highresolution electron microscopy, these properties give an important basis for grouping isometric viruses. For large viruses with a complex morphology, the structure of the particle as revealed by electron microscopy gives valuable information indicating possible relationships.
B. Physicochemical Properties of Virus Particles Physicochemical properties and stability of the virus particle have sometimes played a part in classification. These properties are discussed in Chapter 13, Section II. Noninfectious empty viral protein shells and minor noninfectious nucleoproteins are characteristic of certain groups (e.g., tymoviruses). The existence of these components reflects the stability of the protein shell in the absence of the full-length viral RNA.
The sequence organization and expression strategy of viral genomes, as discussed in Chapter 6, is now of prime importance for the placing of viruses into families, and genera, or for the establishment of a new genus or family.
1. New Sequencing Techniques (reviewed by Metzker, 2010; Studholme et al., 2011; Radford et al., 2012) The development of new rapid, inexpensive next generation high-throughput technologies sequencing (also termed deep-sequencing) over the last 10 years or so is changing the ways we think about the application of sequences to plant virology. These techniques can be applied not only to partially purified and characterized viruses but also to the detection of potential new viruses in crude preparations by metagenomics (see Chapter 1, Section I). These new technologies rely on a combination of template preparation, sequencing, and imaging and on the analysis of the sequence data by assembly methods and genome alignment. They provide a vast amount of data that can be used for a wide range of purposes including virus classification, identification of new viruses, diagnostics, and ecological studies.
2. Use of Sequence Data in Classification As indicated above, the analysis and comparison of sequence data are playing an increasing role in virus classification. Most of these analyses and comparisons are undertaken using computer programs. For any analysis, there are three major questions to be asked: (1) Should the analysis be on the nucleotide sequence or the amino acid sequence? (2) Should the analysis involve the whole genome or specific region(s)? and (3) What computer programs should be used? The answers to these questions depend upon what is being addressed. For instance, does the investigator want to know how this virus should be classified, is the virus unique enough to form a new taxon, how does it fit into the existing taxonomy, and/or are there possible evolutionary relationships with other taxa? Initially, when there were few differences in the difficulty of sequencing viral proteins (usually the CP) and viral nucleic acid (especially RNA) both proteins and nucleic acid sequences were used for classification. With the advent of Sanger sequencing and its automation and the recent next generation sequencing much more emphasis is being put on the use of the genome sequence for classifying viruses. This provides a more refined way of finding differences as, because of the redundancy of the genetic code, the nucleic acid sequence may change but the protein sequence remains the same. However,
C. Properties of Viral Nucleic Acids
16 14 12 10 8 6 4 2
50 60 70 % Homology
FIGURE 2.3 Demarcation between the extent of amino acid sequence homologies in CPs among distinct individual potyviruses (left-hand distribution) and between strains of the same virus (right-hand peak). The 136 possible pairings between 17 strains of eight distinct viruses were analyzed. The homologies between distinct viruses had a mean value of 54.1% and a standard deviation of 7.29%, while the homologies between strains of individual viruses showed a mean of 94.5% and standard deviation of 2.56%. The dashed curves show that all values for distinct viruses and strains fall within ±3 standard deviations from their respective mean values. From Shukla and Ward (1988), with permission of the publishers.
this throws up the problem of what differences in nucleic acid sequence are significant enough for consideration in classification. In most cases, sequence analysis used in classification has involved specific regions of the plant viral genome. CP amino acid sequence homologies have been used to distinguish between distinct potyviruses and strains of these viruses (Figure 2.3) and to estimate degrees of relationship within the group (Figure 2.4). Dendrograms indicating relationships within a virus group have also been based on amino acid composition of the CPs, for example, with the tobamoviruses (Gibbs, 1986). A dendrogram based on peptide patterns obtained from in vitro translated 126-kDa proteins of eight tobamoviruses was very similar to that obtained with CPs (Fraile and García-Arenal, 1990). Once a set of viruses such as the tobamoviruses has been delineated based on CP amino acid composition, new isolates may sometimes be readily placed as a strain of an existing virus (Creaser et al., 1987). Using a statistical procedure known as principal component analysis, Fauquet et al. (1986) showed that groupings of 134 plant viruses and strains obtained using the amino acid composition of their CPs (Figure 2.5) correlated well with the groups established by the ICTV (Matthews, 1982). However, phylogenetic relationships based on one protein may not be similar to one based on another protein. This is suggestive of recombination between viruses (see below). For nucleic acid data differentiators between species can be sequence identity of the whole viral genome (e.g., less than 89% full-length sequence of species within
Chapter | 2 Plant Viruses and Their Classification
BaYMV WSMV TVMV SbMV-V TEV-NAT TEV-HAT SCMV-SC MDMV-B MDMV-A SorgMV JGMV MDMV-KS1 ZYMV-NAT ZYMV-C ZYMV-F WMV-2FC WMV-2 SbMV-N PStV BCMV-NVRS BCMV-3 BCMV-5 PWV-TB PWV-S PWV-K PWV-M PWV-K PRSV-P PRSV-W PepMov PVY-N PVY-10 PVY-D PVY-43 PVY-18 PVY-1 LMV PSbMV PPV-R PPV-AT PPV-NAT PPV-D BYMV-G BYMV-S BYMV-CS CIYVV
FIGURE 2.4 Relationship between some potyviruses and their strains. The dendrogram is based on the amino acid sequence of the core of the viral CPs, omitting the more variable terminal sequences. The core regions were aligned, and the percentages of difference in each sequence (that is, their distance) were calculated. Gaps were counted as a 21st amino acid. The distances were then converted into the dendrogram using the neighbor joining method of Saitou and Nei (1986). The divisions in the horizontal scale represent 10% differences. The broken vertical line indicates a possible boundary between virus species and strains. From Matthews (1991), with permission of the publishers.
a genus of geminiviruses), sequence identity within individual components of a multicomponent virus (e.g., RNA1 of furoviruses species sharing 58–74% identity and RNA2 46–80% identity) or to specific regions of the genome (e.g., species within genera of the family Caulimoviridae being differentiated by having less than 80% nucleotide
identity in the RT–RNaseH region of the genome). However, using specific regions of the genome may not detect “new” viruses that arise by recombination between two other viral genomes (Section II, C, 3). The recent ICTV Reports on Virus Taxonomy (King et al., 2012) list “Species demarcation criteria in each genus.” These criteria include nucleic acid sequence and amino acid sequence derived from the genome sequence data differentiation for many of the appropriate genera (those which contain more than one species), but there is no consistency between the various genera. For some genera, it is the whole nucleotide sequence identity and for others it is nucleotide sequence identity within specific genome regions, usually that encoding the CP and/or the polymerase. The use of amino acid sequence is usually for the CP and the polymerase. There are various arguments for choosing specific regions and gene products; for example, the CP influence various phenotypic characters of the virus (e.g., transmission, movement within the host) and the polymerase is probably the most conserved gene in the virus.
3. Problems in Using Sequence Data for Classification There are several potential problems in using sequence data for classification. The most obvious is the veracity of the data, e.g., is the sequence from the virus that is being considered, is the virus preparation contaminated, is the sequence reading correct, are the correct parameters being used in aligning and analysis of the sequences? One major problem is that there are an increasing number of cases of sequences showing that “new” viruses have arisen from recombination events between two other viruses. This raises the currently unresolved question as to how should these viruses be classified? The usual approach is pragmatic as exemplified by the luteoviruses (Mayo and D’Arcy, 1999) which were first formally recognized as a group comprising the RPV, RMV and SGV strains of BYDV, BWYV, and SbDV in the second ICTV report (Fenner, 1976). Over the next few years, groupings of these viruses increased in a variable manner, depending upon the criteria used. In the sixth report (Murphy et al., 1995), the previous “luteovirus group” was named as the genus Luteovirus and the species divided into two subgroups typified by BYDV-PAV and PLRV. By this time, many of the luteoviral genomes had been sequenced, and it was becoming very apparent that there needed to be a rethink on the classification of this group (D’Arcy and Mayo, 1997). Not only were there differences in genome organizations but there were different polymerase enzymes in different species, some having a “carmovirus-type” polymerase and others a “sobemovirus-type” polymerase (see Chapter 7 for polymerases). Based mainly on these sequence data, the group was divided into three genera that were contained
FIGURE 2.5 Three-dimensional diagram illustrating factors 2, 3, and 4 of a principal components analysis of 122 data sets of plant virus CPs compared by their amino acid composition. The three axes contain 62% of the information. The positions of the viruses on axis 2 are indicated by the sizes of the circles. The numbers within the circles are codes for individual viruses. From Fauquet et al. (1986), with permission of the publishers.
in the family Luteoviridae (Mayo and D’Arcy, 1999; Van Regenmortel et al., 2000). The question of the classification of BYDV was revisited by Miller et al. (2002) who noted that the structural and movement proteins conferring aphid transmission and phloem-limitation properties resemble those of the Luteoviridae whereas genes involved in replication and cis-acting signals more closely resemble those of the Tombusviridae. They reached the pragmatic conclusion that classification of BYDV at the family level should remain “always in the eye of the beholder.” Natural reassortment of genomic segments can lead to new combinations as shown by members of the genus Ourmiavirus in which the three RNA segments appear to be derived from different sources. RNA1, which encodes the RdRp, shows closest affinity to fungal viruses of the family Narnaviridae; the product of RNA2, the movement
protein, shows evolutionary relationship to the genus Tombusviridae; and the product of RNA3, the CP, showed similarity to several plant and animal viruses (Rastgou et al., 2009). Thus, it appears that ourmiavirus genomes are a chimera of segments derived from viruses of different kingdoms. A recent example of recombination is BBLV in which phylogenetic analysis shows a close relationship to members of the Partitiviridae, but its genome organization is very similar to members of the Totiviridae (Martin et al., 2011); this virus has not yet been formally classified.
4. Computing Programs for Analysis of Molecular Data Computers are one of the major tools in analysis of viral sequences and there is a wide range of programs that
Chapter | 2 Plant Viruses and Their Classification
address specific questions. The main uses are comparisons of two or more sequences, analysis of gene families, including functional predictions and the estimation of evolutionary relationships among viruses. Another tool is databases of viral sequences [e.g., GenBank at www.ncbi.nlm.nih.gov, the EBI (EMBL) database at www .ebi.ac.uk/embl, and the Descriptions of Plant Viruses at www.dpvweb.net]. To answer the question of whether a “new” virus is really new and, if so, does its sequence differ significantly from other viruses, there are programs for searching databases for pairwise alignments and for multiple sequence alignments (http://en.wikipedia.org/wiki/List_ of_sequence_alignment_software). One of the main problems in using these programs is deciding on gap penalties. For the analysis of the possible functions of gene products predicted from viral sequences, the amino acid sequence is used in database searches to detect conserved sequence motifs. Various gene families that relate to plant viruses are discussed in Chapters 7, 8, 10, and 12. The possible relationships between taxa higher than species are usually analyzed using phylogenetic methods and programs. These can give some leads as to possible evolutionary routes. It is important to understand the input and output of the various phylogenetic programs so that the correct one is chosen to address the question being posed. Details of these programs are beyond the scope of this book and the reader is referred to reviews such as McCormack and Clewley (2002), Baldauf (2003), and Schmitt and Barker (2009), to books (Salemi and Vandamme, 2003; Wiley and Lieberman, 2011), or to a local expert. Figure 2.6 shows examples of phylogenetic trees of relationships within and between genera of the family Potyviridae and between the CP and polymerase genes of genera of the family Tombusviridae and other taxa. One continuing danger in the use of computer analyses of molecular data is summed up by the saying “garbage in, garbage out.”
D. Properties of Viral Proteins The properties of viral proteins, and in particular the amino acid sequences, have been of great importance in virus classification at all levels: for delineating strains, as discussed in Section IV; for viruses, as discussed in this section; and for indicating evolutionary relationships between families and groups of viruses. The use of amino acid sequences for demarcating between species is discussed above. Among other uses of the properties of viral proteins in classification is the functional equivalence of CPs which may not always be reflected in amino acid sequence similarities. Thus, the CPs of AMV and TSV are required to activate the genomic
RNAs to initiate infection (Box 7.3). The CPs of these two viruses are able to activate each other’s genome, recognizing the same sequence of nucleotides in the RNAs, but there is no obvious amino acid sequence similarity between them (Cornelissen and Bol, 1984). An important goal for using properties of viral proteins in classification is to know the three-dimensional structure of the protein of at least one member of a group. This then allows amino acid substitutions in different viruses in a group to be correlated with biological function. This was achieved for TMV CP and six viruses related to it (Altschuh et al., 1987). The amino acid sequence homologies of these seven tobamoviruses ranged from about 28% to 82%. Twenty-five residues are conserved in all seven sequences (Figure 2.7). Twenty of these conserved residues are concentrated in two locations in the molecule: at low radius in the TMV rod near the RNA-binding site (36–41, 88–94, and 113–120) and at high radius forming a hydrophobic core (61–63 and 144–145). Where viruses within this set of seven differ in sequence, the differences are often complementary. For example, among buried residues, a change to a large side chain in one position may be compensated by a second change to a smaller one in a neighboring amino acid. This study with tobamoviruses, while clearly delineating functionally critical regions of the molecule, did not lead to any clear evidence of particular evolutionary relationships within the group. Other viral gene products have been used in taxonomy. As described in Box 7.2, there are major groups of RNAdependent RNA polymerases. However, a reevaluation of these suggested that this criterion was insufficient to support evolutionary groupings of RNA viruses (Zanotti et al., 1996).
E. Serological Relationships Serological methods for determining relationships and their limitations are discussed in Chapter 13, Section III, A. In the past, these methods have been the most important single criterion for placing viruses in related groups. Members of some groups may be all serologically related (e.g., tymoviruses), whereas, in others, only a few may show any serological relationship (e.g., nepoviruses). In the groups of rod-shaped viruses defined primarily on particle dimensions, many viruses within groups are serologically related, and no serological relationship has been established between groups. It seems probable that serological tests will remain a useful criterion upon which virus groups are based. Serological tests may also be used to estimate degrees of relationship between viruses in a group (Figure 2.8). For both tombusviruses and tymoviruses, there is no correlation between the serological differentiation index
(B) Coat protein
Ag RG M M oM V V V
V TriM V SM SC
99 TFMV LYSV LMoV BYMV CIY VV TE V T P VM YM VA V V
MV BrS MV WEq V ONM V WSM
T CB VB PV MV
V V FM V PP SP oM K V LM MV JY SV NY V ScaM TuMV BiMoV PVY SCMoV PepSMV VerVY P ep M PTV oV PV W V PM V
DsMV BaRM V BtMV FreMV PeMoV BBrMV PSbMV V AWM V M MW V PRS V M MV p Da PLD V M PV MV V iV TM Ch W
SV CB V M M SP
SMV V CLL V WM V M WV PV EA Y FV
BC M T CA elM V BC BMV V M ZYM NV V ZaM MV
CSV JGMV CaYSV PenMV SCMV SrMV MD MV N OY DV SY DV SV
Hepatitis E virus
Carmovirus Necrovirus Machiomovirus
FIGURE 2.6 Phylogenetic analyses of plant virus families. (A) Unrooted phylogenetic tree based on the codon-aligned nucleotide sequences of the polyproteins of fully sequenced members of the family Potyviridae. The tree uses the majority of the polyprotein (from the 6K1 cistron to the end of the CP). One representative sequence was chosen for each species. The analysis was done in MEGA4 (maximum composite likelihood distances) and the numbers on major branches indicate percentage of bootstrap support out of 10,000 bootstrap replications. (B) Phylogenetic analysis of CP and polymerase proteins of genera of the family Tombusviridae and relationships to other taxa. The colored boxes encompass viruses with the two different morphological types of virions, smooth (purple) versus those with a protruding domain (yellow). Protein sequences were aligned and dendrograms constructed using ClustalX2. Genera of the family Tombusviridae are in bold whereas nonmembers are in regular type. The CP of PPV (family Potyviridae) was used as an outgroup. BYDV-PAV is a luteovirus (family Luteoviridae) and SBMV is a sobemovirus. Striped jack nervous necrosis virus (SJNNV) is the type member of the genus Betanodavirus in the family Nodaviridae. Polymerase dendrogram. Hepatitis virus C-1 (HCV-1) and bovine viral diarrhea virus (BVDV-1) are members of the genera Hepacivirus and Pestivirus respectively in the family Flaviviridae. PEMV-2 is an umbravirus in the family Luteoviridae and Hepatitis E virus is a hepevirus in the family Hepeviridae. The polymerase of TMV (family Virgaviridae) was used as an outgroup for generation of the polymerase dendrogram. From King et al. (2012), with permission of the publishers.
Chapter | 2 Plant Viruses and Their Classification
(SDI) values as illustrated in Figure 2.8 and estimates of genome homology. For tobamoviruses, however, there are clear correlations between SDI values, amino acid sequences, and estimates of genome homology (Koenig and Gibbs, 1986). Serological methods can be used to designate a set of virus strains as constituting a new virus within an established group, for example, PMMoV virus in the genus Tobamovirus (Pares, 1988). When it comes to defining degrees of relationship within a group, borderline situations will sometimes be found. In that circumstance, additional criteria will be needed. Using a broadly cross-reactive antiserum against the viral CP core, Shukla et al. (1989b) showed that potyviruses transmitted by mites or whiteflies (now in different genera) were serologically related to a definitive potyvirus with aphid vectors. Distinct potyviruses have often been
FIGURE 2.7 Conserved amino acid residues in the CPs of seven tobamoviruses. The α-carbon chain tracing of one subunit is shown viewed down the disk axis. The positions are marked for hydrophobic (■) and hydrophilic (●) residues, which are invariant in all seven viruses. From Altchuuh et al. (1987), with permission of the publishers.
5.0 – 7.5
.25) 3.0 (3
4.75 (4.5) 1.75 (2 .0)
(2.75) 3.25 ) (1.75 2.25 225 3.0
TBSV- BS 3
PLCV 0 6.
5.75 – 9.0
CvbRSV FIGURE 2.8 The use of serology to estimate degrees of relationship between viruses in a group. The diagram illustrates a classification of 10 tombus viruses with distances representing the mean serological differentiation index of reciprocal tests (RT-SDI). RT-SDI values have been rounded to the nearest 0.25 and, when in order to represent the relationships in two dimensions, the “observed” and “diagrammatic” (in parentheses) RT-SDI values differ, the two values are shown. CIRV, TVN, and CybRSV are only distantly related to one another and to the other tombusviruses, which form a central cluster. In a multidimensional system, these three viruses would have to be arranged in planes above and below that of the central cluster. The arrows indicate the average distance of these viruses from the central cluster. The arrows indicate the average distance of these viruses from the central cluster as a whole, but not from individual viruses. TVN has been renamed NRV. From Koenig and Gibbs (1986), with permission of the publishers.
difficult to define because serological relationships have been found to be complex and often inconsistent. Shukla et al. (1989a) applied a systematic immunochemical analysis to some members of this group. This method involved the use of overlapping peptide fragments to define the parts of the protein combining with antibodies in particular antisera and make it possible to develop both virus-specific and group-specific antisera.
F. Biological Properties in the Plant The use of biological properties such as host range and symptoms in particular host plants as major criteria led to considerable confusion in the identification and classification of viruses. As knowledge of physical and chemical properties of viruses increased, the biological properties were rated as much less important. However, biological properties must still be given some importance in classifying viruses. Macroscopic symptom differences will often reveal the existence of a different strain of a virus where no other criterion, except a full nucleotide sequence comparison, will do so. The detailed study of the cytology of infected cells made possible by electron microscopy has shown that many groups of viruses cause characteristic inclusion bodies or other cytological abnormalities in the cells they infect (Chapter 4, Section IV). Ultrastructural changes in cells usually appear to be much more stable characteristics of virus groups than macroscopic symptoms. Certain tombusviruses can be distinguished from other members of the group by their cytopathological effects, especially in Chenopodium quinoa (Russo et al., 1987). Cross-protection in the plant has been used as a useful indicator of relationship between viruses (Chapter 14, Section IV, A).
G. Methods of Transmission As discussed in Chapter 12, most viruses have only one type of vector, and usually all the viruses within a group have the same type of vector. This character has been considered important in defining new genera within, for example, the original Potyvirus and Geminivirus groups. Under current classification, members of different genera within the family Potyviridae are transmitted by aphids, whiteflies, eriophyid mites, or plasmodiophorales and those of different genera of the family Geminiviridae by leafhoppers, treehoppers, or whiteflies. In general, the type of vector appears to be a stable character that is useful in delineating major groups of viruses. However, within a group of viruses or virus strains some may be transmitted efficiently by a vector species, some inefficiently, and some not at all. Details of the way in which a virus is transmitted by a vector (e.g., nonpersistent or persistent aphid-transmitted viruses, or on the surface or within the spores of a fungus)
may provide further criteria for the grouping of viruses. However, under certain conditions a virus culture may lose the ability to be transmitted by a vector, for example, WTV (Chapter 12, Section III, B, 2, 3).
H. Demarcation Criteria As noted above, virus species are polythetic and need multiple demarcation criteria to reliably delineate different species as listed in the ICTV reports. The usual criteria for the demarcation of genera are sequence comparisons that provide a quantitative measure of divergence. Pairwise sequence identity profiles frequently resolve into peaks which delineate genera, species, and even strains (Figure 2.9). Criteria for demarcating higher taxa are broader and take into account specific unique features, e.g., the pseudoT = 3 virion architecture of the order Picornavirales (Le Gall et al., 2008). An example of the criteria for demarcation is that of species in the family Potyviridae (King et al., 2012): Genome sequence relatedness: different species have CP aa sequence identity less than about 80%; and nt sequence identity less than 76% either in the CP or over the whole genome. ● Differences in polyprotein cleavage sites. ● Host range and key host reactions; lack of cross-protection. ● Different inclusion body morphology. ● Antigenic properties: serological relatedness may help in distinguishing species. ●
I. Discussion As noted in the introduction to this chapter, the classification system for viruses, though fully justified, is without a natural base. This is primarily because there is no time-related information on the evolution of, and relationships between, virus species and genera. Thus, one cannot distinguish with certainty between convergent and divergent evolution. There is further discussion on virus evolution in Chapter 8. Notwithstanding these limitations, an effective system for classifying plant viruses has been developed over the last 40 years or so. The development of this system has given rise to controversies and, no doubt, there will be others in the future (Bos, 1999, 2000; Pringle, 1999; Van Regenmortel, 1999; Van Regenmortel et al., 2013). This system is fulfilling most of the four criteria that I listed in Section I, A but, as pointed out above, there are some reservations on evolutionary relationships. The system is dynamic and is continuously being modified and refined to take account of new research findings. New virus genera are being created and genera grouped together on common features into families. For instance, the grouping of luteoviruses into three genera in one family (described in Section
Chapter | 2 Plant Viruses and Their Classification
Number of sequences
12 10 8
52– 54– 56– 58– 60– 62– 64– 66– 68– 70– 72– 74– 76– 78– 80– 82– 84– 86– 88– 90– 92– 94– 96– 98– 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 80 92 94 96 98 100
Nucleotide sequence identity
80 Number of comparisons
FIGURE 2.9 Differentiation of taxa by pairwise identities of sequences of variants of: (A) RT–RNaseH nucleotide sequences of BSV isolates; (B) nucleic acid sequences of the L1 gene of members of the family Papillomaviridae; and (C) amino acid sequences of CPs of potyviruses. Part A from Harper et al. (2005), Part B from Shukla and Ward (1988), and Part C from Fauquet et al. (2005), with permission of the publishers.
II, C, 3) has helped significantly with the understanding of these viruses. Other virus groups are being, or are likely to be, merged. For instance, it is becoming increasingly apparent that AMV shares many properties with the ilarviruses and there is beginning to be a strong case for merging the Ilarvirus and Alfamovirus genera. However, if this occurs recognition will have to be given to the differences in particle structure between the two groups. One large grouping that needs further consideration is that of the seven floating genera of (+)-strand ssRNA viruses. With the increasing masses of information, there is beginning to be pressure for the creation of higher taxa. The example of viruses (and other elements) that involve reverse transcription in their replication is noted in Section I, C. This would bring together viruses from different kingdoms, especially eukaryotic and even elements that might not be regarded as viruses. There are, of course, virus groups such as the Reoviridae and Rhabdoviridae that span the plant and animal kingdoms. The real need now is to develop systems that reflect commonalities in viruses with (+)-strand ssRNA genomes. As will be discussed further in Chapters 7 and 8, viruses are faced with various problems
to overcome in infecting their host, and common solutions have been found for plant- and animal-infecting viruses. This should be reflected in the classification.
III. STRAINS OF VIRUSES A virus species is not a uniform population but shows a considerable amount of variation. Some of these variants show phenotypic differences in the infected host and are termed strains. Thus a strain is always a variant, but a variant is most often not a strain. Another term frequently used is an isolate which is experimental material corresponding to an instance of a given virus; isolates of a given virus may or may not differ. When viewed at the molecular level the variation within a species is given the term quasispecies.
A. Quasispecies (reviewed by Domingo et al., 2006)
1. The Concept of Quasispecies The concept of quasispecies is fundamental to the understanding of virus variation and evolution; it is discussed
in detail by Eigen (1993), Domingo (1999), Smith et al. (1997), and Domingo et al. (2006). The term quasispecies describes a type of population structure in which collections of closely related genomes are subjected to a continuous process of genetic variation, competition, and selection. Usually, the distribution of mutants or variants is centered on one or several master sequences. The selection equilibrium is metastable and may collapse or change when an advantageous mutant appears in the distribution. In this case, the previous quasispecies will be substituted by a new one characterized by a new master sequence and a new mutant spectrum. The stability of a quasispecies depends upon the complexity of the genetic information in the viral genome, the copy fidelity on replication of the genome, and the superiority of the master sequence. A quasispecies has a physical, chemical, and biological definition. In the physical definition, a quasispecies can be regarded as a cloud in sequence space which is the theoretical representation of all the possible variants of a genomic sequence. For a ssRNA virus of 10 kb, the sequence space is potentially 4 × 104. Thus, the quasispecies cloud represents only a very small proportion of the sequence space and is constrained by the requirements of gene and nucleic acid functions. Chemically, the quasispecies is a rated distribution of related nonidentical genomes. Biologically, a quasispecies is the phenotypic expression of the population, most likely dominated by that of the master sequence.
2. Quasispecies in Nature The concept of quasispecies has developed since the original ideas. The variants in a quasispecies population appear not to act independently, there being increasing evidence for complementation between their gene functions (Sardanyés and Elena, 2010) and so even apparently replication compromised molecules can input into the phenotype of the population with mutations occurring during every replication cycle, the population is dynamic. Models show that there is a critical mutation rate, the error-threshold, beyond which the master sequence is lost (Manrubia et al., 2010) and the quasispecies population is dominated by mutants which may function as a population by complementation (Sardanyés and Elena, 2010). Such models suggest that a virus with low replication rate and high mutational robustness might survive better than a virus with high replication rate and low mutational robustness (Sardanyés et al., 2008).
3. Plant Virus Quasispecies Analyses of passaged infectious clones of both RNA and DNA plant viruses through susceptible hosts have shown that quasispecies populations develop (Kim et al., 2005; Ge et al., 2007; Acosta-Leal et al., 2008; Jeske et al., 2010).
However, there is evidence for some constraints on the sizes of the population. For example, Acosta-Leal et al. (2008) showed that the population of BNYVV is more homogeneous in susceptible plants than in plants containing the Rz1-allele which produces asymptomatic root-limited infections. In the bipartite crinivirus (ToCV), the population of RNA1 component (encoding replication-associated proteins) is more complex than that of the RNA2 component (encoding encapsidation—systemic movement—and insect transmission-relevant proteins) (Lozano et al., 2009). This indicates that, in multicomponent viruses, function can generate large differences in the genetic structure of the different genomic segments. Roossinck (2011) suggested that mutualistic viruses have narrower quasispecies populations than pathogenic viruses due to more stringent selection pressure in coevolution.
B. Virus Strain The conventional definition of a virus species is given in Section I, C and, as discussed in the previous section, a species is essentially a cloud of variants in sequence space. This makes a strict definition of a strain difficult, if not impossible. However, one has to describe strains within a species and in reality, one has to take a pragmatic approach. Characters have to be weighed up as to how they would contribute to making subdivisions and to communication, not only between virologists but also to plant pathologists, extension workers, farmers, and many other groups. This can be illustrated by two examples. BWYV is a luteovirus with a wide host range including sugar beet in the United States and for many years BMYV, which infects sugar beet in Europe, was regarded as a strain of BWYV (Waterhouse et al., 1988). Confusion arose when it was realized that the European luteovirus that is most closely related to BWYV does not infect sugar beet but is common in the oilseed rape crop. This caused many problems in explaining to farmers that the BWYV (beet western yellows virus) in their overwintering oilseed rape crop would not infect their beet crop the next year! In an analysis of nucleotide sequences of 38 isolates of BYMV and BMYV from Europe, Iran, and the USA, de Miranda et al. (1995) identified three distinct sequence groups. The first contained isolates that could infect both oilseed rape and beet, the second infected only oilseed rape, and the third only sugar beet. It was suggested that groups 1 and 2 be named BWYV (possibly BWYV-1 and BWYV-2) and the third group, BMYV. The second example is RTBV that has two major strains or isolate groups, one in South East Asia and the other in the Indian subcontinent (Druka and Hull, 1998). These two strains differ in nucleotide sequence by about 25% and at the amino acid level by about 12–40% depending on open reading frame (ORF) (Hull, 1996).
Chapter | 2 Plant Viruses and Their Classification
Thus, by the criteria used for potyviruses (Section IV, A, 2) and those of several other genera, these should be considered as two distinct species. However, the strains cause the same symptoms and cross-interact with their transmission helper virus, RTSV, in a similar manner. Thus, it would be confusing to farmers and extension workers to give them different names, but it is useful to virologists to indicate that they are different strains.
IV. CRITERIA FOR THE RECOGNITION OF STRAINS AND SPECIES A virus species might be defined simply as a collection of strains with similar properties. Sometimes we wish to ask whether two similar virus isolates are identical or not; on other occasions we will have to decide whether two isolates are different virus species or strains of the same species. Two kinds of properties are available for the recognition and delineation of virus strains—structural criteria based on the properties of the virus particle itself and its components and biological criteria based on various interactions between the virus and its host plant and its insect or other vectors. These criteria were discussed in Section II in relation to the general problem of virus classification. Serological properties are based on the structure of the viral protein or proteins, but because of their practical importance, serological criteria are considered separately in Section IV, B.
A. Structural Criteria Structural criteria comprise the properties of the viral nucleic acids and proteins. The properties used as criteria for strain differentiation are described in detail in Hull (2002) and are summarized below.
1. Nucleic Acids The nucleic acid criteria include: – Sequence differences in the viral genome—The use of this character can cause problems with multicomponent viruses when the genomic segments might be under different selection pressures. Methods for assessing nucleic acid relationships include nucleotide sequences, hybridization, heterogeneity mapping, restriction fragment length polymorphism, singlestrand conformation polymorphisms, and microarrays. – Size differences between genomes—This can be due to additional genes or differences in size of coding or noncoding regions. An example is RNA2 of the TCM strain of TRV which is considerably larger than that of other strains that have been sequenced, for example, PSG, due partly to a repetition of 1099 3′ nucleotides from RNA1, which includes a 16-kDa ORF, and
partly to a 29-kDa ORF that was unique to this RNA2 (Angenent et al., 1986, 1989). There are some limitations on the use of sequence data for delimiting strains. For instance, a given base substitution, deletion, or addition may have very different effects on the protein coded for depending on a number of circumstances. The following factors may be important: – Because the genetic code is degenerate, many base substitutions cause no change in the amino acid being coded for. For example, in the TMV strain L11A, derived from the L strain, there are 10 base substitutions, 7 of which occur in the third position of in-phase codons and do not influence amino acid sequence (Nishiguchi et al., 1985). – A given base substitution may result in change to an amino acid of very similar properties, which causes very little change in the protein. Alternatively, the change may be to a very different kind of amino acid (e.g., from an aliphatic side chain to an aromatic one), giving rise to a viable protein with changed physical properties or to a nonfunctional mutant that does not survive. – A single- or double-base deletion or addition will cause a frameshift mutation with greater or lesser effect depending on whether it is near the beginning or the end of the gene, whether a second change (addition or deletion) brings the reading frame back to the original, and how many proteins are coded for by the section of nucleic acid in question. – A more general problem in using base sequence data for classification is that some parts of the genome, and some gene products, may have multiple functions. Some parts of the genome may code for a single poly peptide, but others may code for more than one; and some polypeptides may have more than one function. Some parts of the genome may have both coding and control or recognition functions; other parts may have just control functions. Furthermore, mutations in one gene may affect the production of another. For example, mutations in the presumed polymerase gene of TMV mutant LIIA cause reduced synthesis of the cell-to-cell movement protein (30 kDa), thus reducing efficiency of movement of this strain (Watanabe et al., 1987). Thus, even if we know the full base sequences for the nucleic acids of a set of virus strains, we would be unwise to use these sequences to establish degrees of relationship without other information. It was once thought that a virus classification scheme based only on nucleotide sequence would be the ultimate aim (Gibbs, 1969). It is now apparent that the significance to be placed on nucleic acid base sequence data can be judged from a biological point of view only in conjunction with knowledge of the
organization of the genome and the functions and interactions of its parts and products. The use of infectious clones and molecular biological techniques is helping in the determination of the association between sequence and biological variation.
2. Structural Proteins The CP or proteins and other structural proteins found in viruses are very important, both for the viruses and for virologists wishing to delineate viruses and virus strains. The CPs of the small RNA viruses must have evolved to give a satisfactory balance between three important functions. 1. The ability to self-assemble around the RNA; mutants are known in which this function is defective even at normal temperatures. For example, strain PM2 of TMV cannot form virus rods with RNA. The protein aggregates in vitro at pH 5.2 to form long, open, flexuous, helical structures rather than compact rods (Zaitlin and Ferris, 1964); 2. Stability of the intact particle inside the cell, and during transmission to a fresh host plant. 3. The ability to disassemble to the extent necessary to free the RNA for transcription and translation. A variant that could not carry out this function would not survive in nature. For example, Bancroft et al. (1971) described a mutant of CCMV induced by nitrous acid that was unable to be uncoated in the cell, and was therefore noninfectious in spite of the fact that RNA isolated by the phenol procedure from the virus was highly infectious. For the small RNA viruses, the CP is of particular importance for the delineation of viruses and virus strains. Besides the intrinsic properties of this protein (size, amino acid sequence, and secondary and tertiary structure), many other measurable structural properties of the virus depend largely or entirely on the CP. These include serological
specificity, architecture of the virus, electrophoretic mobility, cation binding, and stability to various agents. Thus, ideas on relationships within groups of virus strains, based on properties dependent on the CP, may be rather heavily biased. On the other hand, if mutations in the non-CP genes have occurred more or less at the same rate as in the CP during the evolution of strains in nature, then such views on relationships may be reasonably well based. The Potyvirus genus will serve to illustrate the use of the properties of CPs in the delineation of virus strains. Potyviruses have been one of the most difficult virus groups to study taxonomically. The group contains more than 14% of all the known plant virus species. The viruses infect a wide range of host plants and exist in nature as many strains or pathotypes differing in biological properties such as host range and disease severity. It has been considered by some workers that strains of potyviruses may form a continuous spectrum between two or more otherwise distinct viruses, making delineation of viruses and groups of strains difficult or impossible. However, comparisons between the amino acid sequences of the CPs of several viruses and many strains indicate that this approach may provide a useful basis for taxonomy within the group (Shukla and Ward, 1989a,b). Analysis of the 136 possible pairings between a set of viruses and strains revealed a clear-cut bimodal distribution, with distinct viruses having an average sequence homology of 54%, while strains averaged 95% (Figure 2.9C). These data give no support for the “continuous spectrum” idea among the potyviruses. Distinct viruses show major differences in length of their CPs (Figure 2.10). Major differences in amino acid sequence were near the N-termini, with high homology in the C-terminal half of the proteins. On the other hand, strains have very similar N-termini. Two exceptions to this pattern appear to reflect the misplacing of certain potyvirus isolates on the basis of previous data. Serological tests suggested that PeMV and PVY were
PPV-D JGMV-JG PWV-TB PVY-D Pe MV SMV-V TVMV TEV-NAT
150 200 Aminio acid residue number (PPV-D)
FIGURE 2.10 Schematic diagram showing the location of amino acid sequence differences between seven distinct members of the Potyvirus group and PeMV. The sequences were compared with strain D of PVY, the type member. PeMV is very similar to PVY in its CP sequence. From Shukla and Ward (1988), with permission of the publishers.
Chapter | 2 Plant Viruses and Their Classification
only distantly related, yet the sequence data (Figure 2.10) clearly indicate that PeMV should be considered a strain of PVY. SMV-N and SMV-V, formerly considered to be strains of SMV, but when shown to have a sequence homology of 58%, were considered as two distinct viruses (Shukla and Ward, 1988). In both cases and in others, a pragmatic approach has to be taken in assigning species status. The projecting (P) domain of Tombusvirus CPs has a more variable amino acid sequence than the structural (S) domain (Hearne et al., 1990). Thus, greater variability at the exposed surface may be a feature of both rod-shaped and icosahedral plant viruses.
3. Nonstructural Proteins The nonstructural virus-coded proteins are not much used in delineating strains. Mayo et al. (1982) could detect no difference in the tryptic peptides obtained from the VPg of different strains of RpRSV or TBRV. Some strains of TMV may differ widely in the amino acid sequence of their 30-kDa proteins, as discussed by Atabekov and Dorokhov (1984). Four structural proteins of the two serotypes of PYDV—SYDV and CYDV—fell into two groups based on peptide mapping. Proteins M and N differed little between the strains, whereas M2 and G were significantly different (Adam and Hsu, 1984).
4. Proportion of Particle Classes The proportion of particles with differing sedimentation rates found in purified virus preparations or in crude extracts may vary quite widely with different strains of a virus or members of a virus genus. Variation of three kinds can be distinguished: 1. In relative amounts of top component (empty protein shells). For example, among the tymoviruses, the proportion of empty shells to viral nucleoprotein is usually in the range of 1:2 to 1:5 for TYMV and 10:1 to 15:1 for OkMV. Even quite closely related strains may vary in this property, for example, strains of RCMV (Oxelfelt, 1976). For some multipartite viruses, the proportion of top component has been shown to depend on a function of one RNA species. 2. The proportion of nucleic acid components encapsulated may vary in different strains of viruses with multipartite genomes, for example, AMV (van VlotenDoting et al., 1968). Again, nucleoprotein proportions may be under the control of a particular RNA species. 3. Abnormal particle classes may be produced by particular strains. Thus, Hull (1970) described an isolate of AMV producing considerable amounts of particles longer than the B component. It should be recognized that the proportion of particle classes can be affected by factors other than the strain
of virus. These include time after infection, host species, environmental conditions, system of culture (for example, the proportion of TYMV empty protein shells is higher in infected protoplasts than in whole leaf tissue), and isolation procedure.
5. Other Structural Features a. Architecture of the Virus Particle Related viruses will be expected to have very similar size, shape, and geometrical arrangement of subunits. However, significant differences in particle morphology have been found within groups of related strains. Differences in rod length are frequent between strains of rod-shaped (helical) viruses such as TRV (Cooper and Mayo, 1972) and BSMV (Chiko, 1975). Sometimes the variation in architecture appears to be “abnormal” even though the strain of virus is a viable one. Thus, the packing of the CP of the Dahlemense strain of TMV involves a periodic perturbation of the helix (Caspar and Holmes, 1969). Some AMV strains contain abnormally long particles that have the normal diameter but contain more than one RNA molecule (Hull, 1970; Heijtink and Jaspars, 1974). b. Electrophoretic Mobility The electrophoretic mobility of a virus depends in the first place on the amino acid composition of the CP and second on the three-dimensional structure, which affects the availability of ionizing groups. Mobility is also dependent on the ions present in the buffer used. Isolates of PEMV that differed in aphid transmissibility also differed in electrophoretic behavior (Hull, 1977). c. Stability and Density Among the small RNA viruses, differences in stability and density have been used to differentiate virus strains. The RNA content of the virus may vary with strain and thus affect buoyant density in strong salt solutions (Lot and Kaper, 1976). However, differences in the CP most commonly lead to a difference in stability or density.
B. Serological Criteria The nature of antigens and antibodies, the basis for serological tests, and their advantages and limitations are discussed in Chapter 13, Section III. This section considers the use of serological criteria to delineate viruses and virus strains.
1. Some General Considerations a. Presence or Absence of Serological Relationship Serological tests provide a useful criterion for establishing if two virus isolates are related or not. Any of the tests
described in Chapter 13 can be applied, but most commonly some modification of the precipitation reaction or ELISA tests is used. Provided adequate precautions are taken, serological tests can be valuable for placing viruses into groups. If two virus isolates show some degree of serological relationship, it is highly probable that they will have many other properties in common and belong in the same virus group. There are a few unexplained exceptions. Various examples are known of viruses that undoubtedly belong in the same group but that show no serological cross-reactivity—for example, TYMV and EMV in the Tymovirus genus. In making tests for serological relationships, there are several potential sources of error. – Presence in viral antisera of antibodies reacting with host constituents such as the abundant protein ribulose 1′,5′-bisphosphate carboxylase. – Nonspecific precipitation of host materials in crude extracts. – Nonspecific precipitation of viral antigens, especially at high concentrations. – Contamination of antigen preparations with other viruses. – Virus altered during isolation. It should always be borne in mind that virus may be altered during isolation in a way that can affect its serological specificity. – Nonreciprocal positive reactions. To demonstrate that two viruses are serologically unrelated, reactive antisera must be prepared against each of the viruses under test. It must be shown that each reacts with its own antiserum, but gives no reaction with the heterologous antiserum. This reciprocal test is necessary since the viruses might in fact be related, but one may occur in too low a concentration in the extracts to give any positive reaction. Negative one-way tests are of little value. As discussed in the next section, it is preferable to use high-titer antisera to demonstrate a lack of serological relationship. – Isolates taken from the field may be mixtures of several different serotypes (Dekker et al., 1988). These considerations apply particularly to the use of polyclonal antisera. The use of MAbs avoids several of these problems, but they have limitations of their own as discussed in Chapter 13, Section III, D, 2. b. Degrees of Serological Relationship i. Among a Group of Virus Strains A considerable amount of experimental work has been directed towards determining degrees of relatedness within groups of strains and in attempts to correlate serological properties with other biological and chemical characteristics. Delineation of virus strains is a particularly important aspect of any program designed to produce resistant varieties of a host species.
TABLE 2.3 Some Serological Procedures Used for the Delineation of Viruses and Virus Strains Procedure
Virus or Virus Group
Tymo-, Koenig (1981) tombus-, como-, tobamo-, potex-, carla-, potyviruses
Adams and Barbara (1982)
Radial double diffusion in agar, ELISA, and SSEM
Rao et al. (1982)
Quantitative rocket immunoelectrophoresis
Berger and Toler (1983)
Electroblot immunoassay Tymo-, tombus-, Burgermeister and como-, nepoKoenig (1984) , tobamo-, potex-, carla-, potyviruses Direct or indirect ELISA
Ilarviruses and AMV
Halk et al. (1984)
Various ELISA procedures
Huss et al. (1987)
Indirect protein A– sandwich ELISA
Tobamoviruses and virus strains
Hughes and Thomas (1988)
Dekker et al. (1988)
Weidemann and Maiss (1996) and Hoffmann et al. (1997)
If two isolates of a virus are identical, they will respond identically when cross-reacted with each other’s antisera, whatever form of serological test is applied. If, however, they are related but distinct, some degree of cross-reaction will be observed, at least with polyclonal antisera, but the reactions may not be identical. Various types of serological tests can be used to identify and distinguish virus strains. Examples are given in Table 2.3. When a group of only two or three virus isolates is to be considered, it is a relatively simple matter, provided technical precautions are observed, to determine whether the isolates are unrelated serologically, whether they are identical, or whether they show differing degrees of relationship. Using the same set of isolates and the same antisera, quite reproducible results can be obtained, to indicate, for example, that strains A and B are closely related and that both are more distantly related to strain C.
Chapter | 2 Plant Viruses and Their Classification
However, when large numbers of related strains are tested, the situation may become quite complex and less and less meaningful as more strains are considered in relation to one another. ii. Experimental Variables There are a number of important experimental variables that can affect the estimated degree of serological relationship between viruses and strains. These include the following: (i) a major source of experimental variation is the variability in antisera, both in successive bleedings from the same animal and in sera from different individuals. The proportion of cross-reacting antibody present in a series of bleedings taken over a period of months from a single animal may vary widely (Koenig and Bercks, 1968); (ii) the extent to which antisera cross-react to two virus strains is usually correlated with the antibody content of the serum. Sera of low titers show lower cross-reactivity and those with high titers show greater cross-reactivity. Thus, to detect serological differences between closely related strains using polyclonal antisera it is preferable to use antisera of fairly low titer. To demonstrate distant serological relationships, it may be necessary to use high-titer antisera; (iii) many virus preparations used for immunization and for antibody assay may contain varying amounts of free CP or CP in various intermediate states of aggregation or in a denatured state. CP in the intact virus may lose amino acids through proteolysis. Antibodies reactive with CP in these various forms may or may not indicate the same sort of relationships as antibodies against intact virus. The method used to detect and assay cross-reacting and strain-specific antibodies may affect the apparent degree of relationship. Examples are given in the references listed in Table 2.3. c. The Serological Differentiation Index In spite of all the variables, useful assessment of degrees of serological relationship can be obtained by testing successive bleedings from many animals and pooling the results. Quantitative measurements of degrees of serological relationship have been carried out using precipitation titers and ELISA tests (Jaegle and Van Regenmortel, 1985; Clark and Barbara, 1987). The extent of serological crossreactivity can be expressed by a SDI (Van Regenmortel and von Wechmar, 1970) (see Chapter 13, Section III, A, 4, d). The SDI values are equal to the difference in those titers expressed as negative log 2. For example, such replicated comparisons have been made for sets of tobamoviruses (Van Regenmortel, 1975) and tymoviruses (Koenig, 1976). Table 2.4 shows a comparison of the SDIs obtained from ELISA and precipitin tests. There were differences in the reciprocal SDIs found by ELISA for pairs of viruses, and the average of these values did not correspond closely to those found by precipitin SDIs. Nevertheless, both kinds of test show clearly that CGMMV is substantially different from the other tobamoviruses.
Clark and Barbara (1987) describe a more refined statistical procedure for calculating SDIs from ELISA tests that is capable of discriminating reliably among virus strains that differ by as little as 0.2 SDI.
2. The Role of Virus Components in Serological Reactions There is no good evidence that plant viral ssRNA can elicit RNA-specific antibodies. Antibodies formed in response to injection with a plant virus react only with the virus CP, either in the intact virus or as various partial degradation products of the intact protein shell. A formal demonstration of the role of the protein was made by Fraenkel-Conrat and Singer (1957). They carried out mixed reconstitution experiments between serologically distinct strains of TMV. The artificial hybrid virus had the serological type of the protein used to coat the RNA, but the progeny following infection had protein of the type from which the RNA was obtained. Nevertheless, the viral RNA may play some secondary role in stimulating production of antibodies against the viral protein shell. Intact TYMV is substantially more immunogenic than the apparently identical empty protein shell, which contains no RNA (Marbrook and Matthews, 1966). This difference was found in rabbits and mice using several routes of injection. The difference persisted throughout the time course of the primary response and was also found in the response to a second injection. Isolated TYMV RNA injected at the same time did not augment the immunogenicity of the empty protein shells. Artificial empty protein shells produced from the infectious virus in vitro were no more immunogenic than the natural empty shells. Noninfectious TYMV nucleoprotein was just as immunogenic as infectious virus. Thus, it was concluded that the enhanced immunogenicity of the nucleoprotein must be due to the physical presence of the RNA inside the particle. Whether this enhanced immunogenicity of the viral nucleoprotein is a general feature of plant viruses remains to be determined. TMV appears to be more immunogenic than either protein rods or subunits (Marbrook and Matthews, 1966; Loor, 1967). The mechanism by which the ssRNA within the virus stimulates immunogenicity is not yet understood. dsRNAs can be immunogenic. dsRNA antisera react with dsRNA but not dsDNA or ss nucleic acids. The antisera lack specificity for particular ds nucleic acids (see Chapter 13, Section IV, A, 2). There are several reasons why we would expect intact viruses (such as TMV and TYMV) and protein subunits or subviral aggregates prepared from them to differ in the antigenic sites they possess: – Some antibody-combining sites on the intact virus may be made up of parts of the exposed surface of two or
TABLE 2.4 Serological Differentiation Indices (SDI) for Pairs of Tobamoviruses Calculated from ELISA and from Precipitin Tests Tobamoviruses
SDI from ELISA
Average SDI from Precipitin Test
% Sequence Similarity in Viral CPs
From Jaegle and Van Regenmortel (1985).
more subunits. Such a site would not exist in isolated subunits. – Subunits probably have characteristic combining sites, which are masked when the subunits are packed into the intact shell. – Conformational changes occur when the subunits aggregate, so that the configuration of the exposed surface of the packed subunit may not be the same as when it exists as a monomer. Examples of all these phenomena are known among plant viruses and their protein subunits.
3. Procedures Used for Delineating Viruses and Strains a. Assay Methods The various serological methods that are used in the detection and assay of viruses are discussed in Chapter 13, Section III, B. Most of these procedures have been used for delineating viruses and virus strains. Some examples are listed in Table 2.3. This list reflects the fact that ELISA procedures have become the most popular for the delineation of viruses and strains. b. Monoclonal Antibodies The advantages and disadvantages of using MAbs for assay, detection, and diagnosis of viruses are summarized in Chapter 13, Section III, D. The outstanding value of MAbs in the delineation of virus strains is that their
molecular homogeneity ensures that only one antigenic determinant is involved in a particular reaction. The high specificity of this single interaction is not swamped in a large number of other interactions as with a polyclonal antiserum. Provided a MAb can be found that recognizes a small antigenic difference between two virus strains, then very fine distinctions can be made in a reproducible manner. Although the ability of different MAbs to distinguish single amino acid exchanges may vary widely, some may be able to do so (Al-Moudallal et al., 1982). However, there are several limitations in the use of MAbs: (i) there is usually no immunoprecipitation between MAbs and viral protein monomers; (ii) MAbs are often sensitive to minor conformation changes in the antigen such as may be caused by detergent or by binding of antigen to an ELISA plate (Dekker et al., 1987); (iii) among a set of virus strains the relative reactivity of different MAbs may vary considerably. For example, TMV strain 06 differs from type TMV by residue exchanges at positions 9, 65, and 129 of the CP but reacted like TMV with MAb-a, more strongly than TMV with MAb-c, and not at all with MAb-b; (iv) MAbs may be heterospecific, that is, they may frequently react more strongly with other antigens than with the virus used for immunization. The reaction of strain 06 with MAb-c, which is stronger than that with TMV, the strain used as the immunogen, illustrates this phenomenon. If, during the selection of hybridoma clones for the isolation of MAbs, the clones are tested only with the strain of virus used as immunogen, MAbs with low affinity for
Chapter | 2 Plant Viruses and Their Classification
this strain may go undetected and be discarded. Among such MAbs may be clones that would be very useful for the detection of other strains. Thus, when searching for strain-specific MAbs it is important to screen clones against a panel of structural relatives of the immunogen. Another potential limitation of MAbs in the delineation of strains can occur if two strains have an identical antigenic determinant in common. If, by chance, the MAb specificity is directed against this determinant, the strains will appear identical even though they may have substantially different determinants elsewhere in the molecule. For example, strain Y-TAMV is a member of the ToMV group of strains that has an 18% difference in CP amino acid sequence compared with TMV. The two viruses are readily distinguished by polyclonal antisera but not by some MAbs. These limitations highlight the importance of generating diverse panels of MAbs for the delineation of viruses and strains. Experiments with ToMV further illustrate the problem. Strains of this virus are considered to be serologically quite uniform. Ten MAbs raised against the virus reacted in an identical manner with 15 ToMV strains and isolates but two of them cross-reacted with TMV and RMV (Dekker et al., 1987).
4. Antigenic Sites Involved in the Serological Delineation of Viruses and Strains Because of the crucial role they play in inter-subunit bonding, the sides of protein subunits that make up the shells or rods of a particular virus might be expected to be fairly constrained in the extent to which amino acid replacements would allow the subunit to remain functional. This would be particularly so with rod-shaped viruses (and also certain isometric viruses) in which RNA–protein interactions are also important. One might expect much less constraint on that part of the protein subunit that makes up the surface of the virus, and that would therefore also provide the antigenic sites of the intact virus. This expectation has been confirmed for members of the Potyvirus genus. Biochemical and immunological evidence suggested that the N-terminal 29 amino acids of TEV are hydrophilic and located at or near the virus surface (Allison et al., 1985). Mild proteolysis by trypsin of the particles of six distinct potyviruses showed that the N- and C-terminal regions of the CPs are exposed at the particle surface. Trypsinization removed 30–67 residues from the N-terminus and 18–20 from the C terminus, the length removed depending on the virus (Shukla et al., 1988). This proteolysis left a fully assembled, infectious virus particle containing protein cores consisting of 216 or 218 amino acids. Electroblot immunoassays with polyclonal antisera showed that potyvirus-specific antigenic sites are
located in the trypsin-resistant core protein region. Thus, antibodies to the dissociated core protein should react with most potyviruses. On the other hand, the surface-located N-terminus is the only large region in the CP that is unique to a particular potyvirus, and most virus-specific antibodies should react with this region. This fits with the amino acid sequence data, which shows that the N-terminal region is the most variable in potyvirus CPs. It has been known for some time that potyviruses become partly degraded on storage. The use of partially degraded virus as an immunogen or in antigenic analyses may account for many of the contradictory reports in the literature concerning serological relationships among the potyviruses (Shukla et al., 1988). Shukla et al. (1989c) developed the following simple procedure to remove cross-reacting group-specific antibodies. The virus-specific N-terminal region of the CP of one potyvirus was removed using lysylendopeptidase and the truncated CP was then coupled to cyanogen bromide–activated Sepharose. By passing antisera to different potyviruses through such a column, the cross-reacting antibodies were bound. Antibodies that did not bind reacted only with the homologous virus and its strains, as judged by electroblot immunoassays. The practical utility of this procedure has been demonstrated for a group of 17 potyvirus isolates infecting maize, sorghum, and sugarcane in Australia and the United States whose taxonomy was in a confused state (Shukla et al., 1989d). The results demonstrated that the 17 strains belong to four distinct potyviruses, for which the names JGMV, MDMV, SCMV, and SrMV were proposed. Electroblot immunoassays using native and truncated CPs (minus the N-terminus) can be used to screen MAbs to determine whether they are group- or virus-specific. This procedure was used to distinguish MAbs that were virus-specific from those that reacted with 2 or more, and sometimes all 15, of the potyviruses tested (Hewish and Shukla, reported in Shukla and Ward, 1989b). Using panels of MAbs to ACMV, ICMV, OLCV, and TYLCV epitope, profiles have been obtained for various whitefly-transmitted geminiviruses and serotypes thereof (Swanson et al., 1992; Konaté et al., 1995; Harrison et al., 1997; Ogbe et al., 2003; Wu et al., 2012). These profiles reveal MAbs that react with several viruses and those that are virus- or even strain-specific. The epitope profiles of 12 begomoviruses are illustrated in Figure 2.11.
5. Production of Antibodies Against Defined Antigenic Determinants Geysen et al. (1984) described a procedure for the rapid concurrent synthesis of hundreds of peptides on solid supports which had sufficient purity to be used in ELISA tests. Using sets of such peptides and antisera against a
Virus and source
MAb (SCR no.) 12 14 17 18 20 22 23 27 29 32 33 52 53 54 55 56 58 60 62 66
ACMV, Kenya OLCV, lvory Coast EACMV, Madagascar TLCV, Burkina Faso TYLCV, Senegal ICMV, India TLCV, India TYLCV, Thailand CLCuV, Pakistan BGMV, Puerto Rico TGMV, Brazil CLCrV, USA
Cassava Okra Cassava Tobacco Tomato Cassava Tobacco Tomato Cotton Bean Tomato Cotton
0 1 2 3 4
FIGURE 2.11 Epitope profiles of 12 begomoviruses from six hosts, illustrating the differences between viruses from the same host in different continents, and similarities among viruses from different hosts in the same continent. Strengths of reaction range from imperceptible (0) to very strong (4). From Harrison and Robinson (1999), with permission of the publishers.
virus, immunologically important amino acid sequences on the virus could be closely defined. This procedure has been successfully applied to the analysis of both polyclonal sera and MAbs raised against potyviruses (Shukla et al., 1989a). This work opened up the possibility of using synthetic peptides corresponding to defined antigenic sites as immunogens to generate group-specific, virus-specific, and perhaps some strain-specific serological probes. Intrinsic and extrinsic factors affecting antigenic sites in relation to the prediction of important amino acid sequences are discussed in Berzofsky (1985) and the use of recombinant antibodies for virus strain discrimination is reviewed by Boonham and Barker (2000). The ability to be able to express single-chain variable fragments from cloned cDNAs on the surface of phage particles (Chapter 13, Section III, E) gives another approach to obtaining highly specific antibodies.
6. Antibodies Against Nonstructural Proteins Variation in nonstructural proteins can sometimes be used to define strains. For instance, Chang et al. (1988) used the serological reactions of nuclear inclusion proteins to study relationships between a set of potyviruses and potyvirus strains.
7. Other Uses of Strain-Specific Antisera Besides the use of serological methods for establishing relationships between plant viruses, strain-specific antisera provide very useful reagents for various kinds of experiments. For example, antisera specific for ToMV strains have been used to monitor the effectiveness of the protection given by infection of tomatoes with mild strains of ToMV to superinfection with wild strains (Cassells and Herrick,
1977), and to study the mechanism of cross-protection (Barker and Harrison, 1978). Strain-specific antisera were used to show that, when tobacco leaf protoplasts were doubly infected with two TMV strains, some progeny rods contained a mixture of both CPs (Otsuki and Takebe, 1976). Antibodies specific for TMV strains were used to study the conditions under which phenotypically mixed rods of TMV could be formed in vivo and in vitro (Atabekova et al., 1975; Taliansky et al., 1977). Purcifull et al. (1973) used strain-specific antisera for several potyviruses to show that the protein found in the inclusion bodies induced by each strain was distinct, unrelated to the viral CP, and independent of host species in which the virus was grown. The site of initiation and direction of TMV assembly were elegantly confirmed by Otsuki et al. (1977) using strain-specific antibody.
C. Biological Criteria 1. Symptoms a. Macroscopic Symptoms Symptom differences are of prime importance in the recognition of mutant strains. However, the extent of differences in disease symptoms may be a quite unreliable measure of the degree of relatedness between different members of a group of strains. Symptoms produced by different virus strains in the same species and variety of host plant may range from the symptomless “carrier” state, through mosaic diseases of varying degrees of severity, to lethal necrotic disease. The diseases produced by a given set of strains in one host plant may not be correlated at all with the kinds of disease produced in another host species. Most viruses, including many of widespread occurrences such as TMV, PVX, PVY, AMV, and CMV, occur as numerous strains in
Chapter | 2 Plant Viruses and Their Classification
nature. Many “new” viruses have been described primarily based on symptoms and other biological properties, which have turned out later to be a strain of one of these commonly occurring viruses. Some viruses appear to have given rise to relatively few strains as judged by symptoms, for example, PLRV in potato varieties. A set of defined cultivars that give differential local lesion responses may provide a particularly useful and rapid method for delineating strains among field isolates of a virus. However, the important influence of environmental conditions on local lesion responses must be controlled. A virus causing severe disease is often said to be more “virulent” than one causing mild disease. From what has been said in other sections, it should be apparent that the description can only be applied to a given strain of the virus inoculated into a particular variety of host plant in a specific manner and growing under particular environmental conditions. As described in Chapter 4, Section VIII, A, the presence of satellite viruses, RNAs or DNAs, or of defective or defective interfering (DI) nucleic acids can alter symptom expression, and care should be taken to establish that these are not present in the infection. A named variety of host plant, especially a long-established one, may come to vary considerably in its reaction to a given strain of virus, due, for example, to the fact that seed merchants in different localities may make different selections for propagation. This may add a further complication to the identification of strains by means of symptoms produced on named cultivars. Nevertheless, a systematic study of symptoms produced on several host species or varieties under standard conditions may help considerably to delineate strains among large numbers of field isolates of a virus. b. Cytological Effects The cytological changes induced by different strains of a virus are often readily distinguished. Differences are of three kinds: (i) in the effects on cell organelles; (ii) in the virus-induced structures within the cell; or (iii) in the distribution or aggregation state of virus particles within the cell. Such differences may be of increasing importance in the delineation of viruses and virus strains. However, other factors may cause variation in the extent of differences between strains. For example, various strains in the stock culture of TYMV have markedly different effects on chloroplasts in cells of systemically infected leaves (see Figure 10.25), but these differences may be much less marked or nonexistent in the infected cells of local lesions. Different strains of TuMV show differences in the morphology of their cylindrical inclusions (McDonald and Hiebert, 1975). Ultrastructural changes in both nucleus and cytoplasm of oat cells infected with BYDV strains differed between
strains that were specific for a particular aphid vector and those that were not (Gill and Chong, 1979). Different strains of AMV may differ markedly in the way in which virus particles form aggregates within infected cells (Hull et al., 1970; Wilcoxson et al., 1975). The characteristic viroplasms found in cells infected with caulimoviruses (see Figure 7.22) may vary with different strains (Givord et al., 1984; Stratford et al., 1988). The variation may be associated with differences in gene II and the proportions of the products of genes II and IV. Mixed infections with two variants of BYDV in oats gave rise to altered patterns of effects in vascular tissue, including a predisposition for the xylem to become infected (Gill and Chong, 1981).
2. Host Range and Host Plant Genotype Host ranges of viruses generally are discussed in Chapter 4, Section V. Many strains of a virus may have very similar host ranges, but others may differ considerably. Similar responses of a set of host plant genotypes to two viruses may provide good evidence that they are related strains (Schroeder and Provvidenti, 1971). On the other hand, a loss in ability to infect a particular host may be brought about by a single mutation. Dahl and Knight (1963) studied 12 mutants isolated from ToMV that had been treated with nitrous acid. One of these strains had lost the capacity to infect tomato. Strains of a virus that have different host ranges often produce different disease symptoms on some common host. This is not always so. For example, four strains of TMV that were not clearly distinguishable by symptoms on Nicotiana tabacum or on common varieties of Lycopersicon esculentum could be differentiated by their host ranges on a set of Lycopersicon hosts, including two varieties of L. esculentum and three selections of L. peruvianum (McRitchie and Alexander, 1963). Strains of PVX have been grouped according to their reactions to a range of host plant genotypes (Cockerham, 1970).
3. Methods of Transmission Different arthropod vector species or different biotypes of a single species may differ in their transmission of various strains of the same virus. Differences may be of the following kinds: – In the percentage of successful transmissions, for example, MDMV strains by aphid species (Louie and Knoke, 1975). – In minimum acquisition time by the vector, for example, MDMV in aphid vectors (Thongmeearkom et al., 1976). – In the length of the latent period, for example, strains of PEMV in its aphid vector (Bath and Tsai, 1969). – In the time that the vectors remain infective (Thongmeearkom et al., 1976).
– Some strains may not be transmitted at all by particular vectors, for example, strains of PYDV and leafhopper species (Black, 1941). Patterns of transmissibility by three aphid species have allowed large numbers of field variants of BYDV found in North America to be placed into five main groups (Rochow, 1979), which were then recognized as different species. The quite stable groupings have facilitated studies on the distribution of virus variants both geographically and in successive seasons. If one strain of a virus is transmissible by mechanical means all others usually are too. However, there are reports of marked variation in mechanical transmissibility depending on both host clone and virus strain, for example, AMV in alfalfa (Frosheiser, 1969). Defective strains may occur in which the RNA is not coated or is incompletely coated with protein. Such strains will not be mechanically transmissible except under conditions where they are protected from attack by nucleases.
another in water pimpernel Samolus parviflorus (Bennett, 1955). Within a set of isolates that are undoubtedly related strains all possibilities may exist—reciprocal cross- protection of varying degrees of completeness, unilateral cross-protection, and no cross-protection, as was found for strains of TSV in tobacco (Fulton, 1978). The other factor that may make cross-protection tests ambiguous is that there can be quite strong interference between some unrelated viruses (Bos, 1970). Most experiments on cross-protection have been carried out using mechanical transmission. Cross-protection may also occur in the plant with viruses transmitted in a persistent manner by insect vectors. Thus, Harrison (1958) found that infection with a mild strain of PLRV protected plants against infection with a severe strain introduced by the aphid vector Myzus persicae. Cross-protection also occurs in viroids. The role of RNA silencing in cross-protection is discussed in Chapter 9, Section V, H, 1.
4. Cross-Protection The mechanism of cross-protection is discussed in Chapter 14, Section IV, A. Early observations on the interactions between virus strains led to the development of the concept of crossprotection. It was shown by McKinney (1929) that tobacco plants infected with a green mosaic virus (TMV) developed no further symptoms when inoculated with a yellow mosaic virus. Salaman (1933) found that tobaccos inoculated with a mild strain of PVX were immune from subsequent inoculation with severe strains of the virus, even if inoculated only after 5 days. They were not immune to infection with the unrelated viruses, TMV and PVY. This phenomenon, which has been variously called crossprotection, antagonism, or interference, was soon found to occur very commonly among related virus strains. It is most readily demonstrated when the first strain inoculated causes a fairly mild systemic disease and the second strain causes severe symptoms or necrotic local lesions. Development of such lesions can be readily observed and a quantitative assessment can be made. Interference between related strains can also be demonstrated by mixing the two viruses in the same inoculum and inoculating to a host that gives distinctive lesions for one or both of the two viruses or strains. For a time, cross-protection tests were given considerable weight in establishing whether two virus isolates were related strains or not, but subsequent developments have indicated the need for caution. Among a group of strains that on other grounds are undoubtedly related, some may give complete cross-protection, while with other combinations protection may be only partial. Some virus strains do not appear to cross-protect at all. Thus, none of the strains of BCTV protect against one
5. Virus Productivity Different strains of a virus may vary widely in the amount of virus produced in a given host under standard conditions. For example, the common strain of TMV is the most productive; other naturally occurring strains vary over a range down to about one-tenth that of common TMV when productivity was measured as the number of local lesions produced from inocula made from extracts of single local lesions produced in N. tabacum cultivar Xanthi nc (Veldee and Fraenkel-Conrat, 1962). Chemically induced mutants also vary widely in productivity, and all are less productive than common TMV. Some of these strains cause severe symptoms in certain hosts, but there is no correlation between severity of disease and productivity. Productivity appears to be a genetically stable character since it remains fairly constant for a given mutant when tested after successive transfers. Chemical mutation quite frequently increases the severity of disease produced, but rarely, if ever, increases the productivity. From a type culture of TMV, Kassanis (quoted in Matthews, 1991) isolated strains causing slowly spreading bright yellow local lesions, usually without systemic spread, in White Burley tobacco Virus content, of these yellow lesions was extremely low. Such strains are difficult to maintain in the laboratory and would never survive in nature.
6. Specific Infectivity Bawden and Pirie (1956) showed that infectivity per unit weight of purified type TMV was greater than that of a Datura strain when tested in N. glutinosa. There is some evidence suggesting that the viral CP may be involved in differences in specific infectivity at least between different
Chapter | 2 Plant Viruses and Their Classification
viruses. Thus, Fraenkel-Conrat and Singer (1957) found that RMV had only about 5% of the specific infectivity of common TMV. However, when RMV RNA was reconstituted with common TMV protein, the specific infectivity was about four times higher than the RMV preparation that provided the RNA. Reconstituted TMV usually has a lower specific infectivity than the intact virus. The reason for the increase when the RMV RNA was coated with type TMV protein is not known, but might be due to the relative ease with which intact RMV and the RMV RNA reconstituted with type protein are uncoated in vivo.
7. Genome Compatibility The possibility of carrying out viability tests with mixtures of components from different isolates of viruses with multipartite genomes provides a functional biological test of relationship. Such tests were carried out with TRV strains by Sänger (1969). Only 2 of the 20 combinations he tested gave a functional interaction. Members of the Nepovirus genus show various degrees of compatibility in genetic reassortment experiments (Randles et al., 1977). Rao and Francki (1981) found that the RNAs 1, 2, and 3 of three strains of CMV were interchangeable in all combinations. However, with TAV, a distinct virus in the Cucumovirus genus, only RNA3 could be exchanged with those of the CMV strains. Similarly, only RNA3 could be successfully exchanged between two members of the Bromovirus genus—BMV and CCMV (Allison et al., 1988). The incompatibility of RNAs 1 and 2 of these viruses is presumably due to how their gene products interact (see Chapter 16, Section II, D, 1). Genome compatibility can be tested in a more direct fashion when the gene products can be isolated and their function is known. For example, Goldbach and Krijt (1982) showed that the protease coded for by CPMV did not process the primary translation products of other comoviruses. The transcriptase activities found in the particles of two rhabdoviruses (LNYV and BNYV) did not carry out transcription with the heterologous virus (Toriyama and Peters, 1981).
8. Activation of Satellites Particular isolates of TNV will support the replication of some STNVs but not others (Uyemoto and Gilmer, 1972). Similarly, among the cucumoviruses and the small satellite RNAs found in association with them, some viruses support the replication of particular satellite RNAs while others do not (Chapter 5, Section II, B).
D. Discussion The extent to which virus species have been clearly delineated varies widely among the different genera and families
of viruses. There are dangers in formalizing virus species or virus groups before a sufficient number and diversity of strains have been investigated. For example, at a stage when only about seven tymoviruses were known, two subgroups were suggested on the basis of serological relationships and RNA base composition (Gibbs, 1969; Harrison et al., 1971). Since then further tymoviruses have been discovered with intermediate characteristics (Koenig and Givord, 1974). For some groups, such as the potyviruses, “a common set or pattern of correlating stable properties” has emerged that can allow the grouping of virus strains into species with some degree of confidence. The relative importance, or weight, to be placed on different properties of a virus for purposes of classification remains a difficult problem. An adequate understanding of the significance to be placed on the various properties may come only when we have a detailed knowledge of the structure of the viral genome, the polypeptides it codes for and their functions, and the regulatory or other roles of any translated or untranslated regions in the genome. Even with such knowledge difficulties will remain. For example: i. Disease induction, which is a complex process, has been shown for some viruses to depend on the functions of two or more viral genes. ii. Various possible mechanisms are now known whereby a single mutation could have effects on two or more functions. iii. A single gene product may have two or more functions, differing in importance, for the virus infection cycle. Thus from a practical point of view it may be an oversimplification to establish relationships between viruses and strains within a family or group solely on the basis of nucleotide sequences. As there is no formal definition of a strain, a pragmatic approach has to be taken. In considering use of the various possible criteria for the delineation of virus strains, we must bear in mind that, from a strictly genetic point of view, complete nucleotide sequence data would be sufficient to establish relationships between strains; however, as argued above this is probably an oversimplification. Small changes in nucleotide sequence can have very different phenotypic effects. At one extreme a single-base change in the CP gene could give rise to changes in several of the phenotypic properties noted earlier. On the other hand, several base changes might give rise to no phenotypic effects at all. For practical purposes, phenotypic characters such as host range, disease symptoms, and insect vectors must usually be given some weight in delineating and grouping virus strains. One further consideration in the delimitation of virus strains is how to differentiate them from virus isolates. A common mistake is to name different virus isolates as strains when there are no real differences between them.
When a sufficient number of strains have been examined, nucleotide sequence relationships may be used to delineate subgroups of strains of a virus. For example, based on competition hybridization tests, 30 strains of PSV could be divided into two groups with little homology between them, but extensive homology within groups (Diaz-Ruiz and Kaper, 1983). Similarly, CMV strains have been subgrouped (Section V, C, 4).
V. CORRELATIONS BETWEEN CRITERIA FOR CHARACTERIZING VIRUSES AND VIRUS STRAINS In the preceding sections, I have surveyed the various criteria that can be used to delineate variation among virus isolates. How can we use these criteria to decide whether a particular isolate is identical to another isolate or a related variant or strain, or whether it is a distinct virus? This is a question of considerable practical importance, because the recognition and identification of virus strains may be most important for effective virus control. In addition, the virological literature is cluttered with inadequate descriptions of virus isolates. These are frequently described as new viruses or new strains, particularly if they are found in a new host or a different country, when adequate study might show they were very probably identical to some virus already described. The definition of a virus species is given in Section I, C.
A. Criteria for Identity There is only one criterion that will establish that two virus isolates are identical—the identity of the complete sequence of their genome nucleic acids. The development of rapid, inexpensive sequencing techniques is enabling this to be a practical approach. Also to be taken into account is determining what to sequence in a quasispecies population. For most practical purposes, the following criteria would be sufficient to establish provisional identity of two virus isolates: (i) identity of size, shape, and of any substructure, of the virus particle as revealed by appropriate electron microscopy; (ii) serological identity in adequate tests; (iii) identical disease symptoms and host ranges for a set of indicator hosts and genotypes; and (iv) identical transmission, especially with respect to any arthropod, nematode, or fungal vectors. The presumption of identity would be greatly reinforced by information on some aspect of nucleic acid sequence, for example, identical sequences in a particular region of the genome or identity as judged by heterogeneity mapping.
B. Strains and Viruses The broad questions of virus classification are dealt with in Section I. Here we will consider the problems involved
in using the various properties outlined earlier in this chapter to define virus strains, to group them, and to decide whether an isolate is a strain or a different virus. One method is to take a quantitatively determined set of characteristics such as the amino acid composition of the CP. Statistical procedures and computer analysis are then used to derive a classification with degrees of relationship indicated. Computer analysis is particularly useful for handling large amounts of numerical data as was used, for example, to derive Figure 2.4 from amino acid sequences. However, a classification based on computer analysis is no more objective than other ways of making a classification. It will depend on the personal judgments and selections made by the taxonomist providing the data. In the Adansonian approach, all the known characters are given weight in determining groupings. This approach has become popular with the widespread availability of computers but there are significant limitations. For example, as noted earlier, many of the fairly easily measured characters of a small virus depend on properties of the CP. Thus, differences in the CP may be given undue weight. Similarly, symptom differences between two strains could be emphasized, merely by recording differences on an extended host range. On the other hand, the hierarchical system involves making arbitrary decisions about which characters are the most important. There are serious objections to applying such a system, without some modification, especially when we are considering classification within a group of related viruses. The most useful characters will be different within different virus groups. Thus, the CP of STNV, being the only gene product of this virus, should be given more substantial weight than would the CP of a virus with, say, 10 genes. Similarly, particle morphology may be most useful for those groups such as the Rhabdoviridae that possess a complex structure. The best course is probably the pragmatic one of considering all known properties within a group of variants and weighting them in a common-sense manner in relation to the overall properties of the group in question. When strains arise in the stock culture of a virus in the laboratory as they do with such viruses as TMV and TYMV, we can be reasonably sure that they will be closely related to the parental strain—often arising from a single mutation. Phenotypic differences in most properties will usually be small, but may sometimes be large as with the TMV strains, such as PM1, that produce defective CPs and no intact virus. Virus isolates collected in the field, perhaps from different host species in different countries, may appear related on the basis of some properties and unrelated on others. The only generalization that can be made at present is that closely related strains will differ in only a few properties, while distantly related strains will differ in many.
Chapter | 2 Plant Viruses and Their Classification
The extent to which different properties show correlations varies widely in the different groups of viruses.
C. Correlations for Various Criteria From a purely genetic point of view, the relationships between a set of virus strains can be assessed precisely if we know the differences in nucleotide sequences between their genomes. However, from the virological point of view, other factors must be taken into consideration. For example: (i) nucleotide changes that are silent, that is, lead to no change in the structure or function of the virus, are usually of little interest. However, one must remember that with accumulating knowledge, what are considered as nonfunctional nucleotides at one time may in the future be highly significant. Furthermore, the expression of a viral genome often involves secondary and/or tertiary nucleic acid interactions between often distant parts of the genome (see Chapter 6). (ii) Particular gene functions may be of particular ecological and therefore practical significance, for example, mutations in a viral gene that affects insect vector specificity; and (iii) when large numbers of field isolates have to be typed over a short time interval, only rapid diagnostic methods are practicable. The confidence with which particular criteria can be used depends in part on the extent to which they correlate with other criteria. This section gives a brief overview of these problems.
1. Host Responses Where a group of strains are fairly closely related, host responses may provide the best, or even only practicable, criteria for establishing strain types. For example, Mosch et al. (1973) found that 18 isolates of TMV from greenhouse tomato crops could be placed in three groups depending on their pathogenicity for a set of Lycopersicon esculentum clones. There were no differences in certain physical properties (buoyant density and sedimentation coefficient) and only small individual differences in CP composition. These did not correlate with the pathogenicity groups. When a virus of economic importance such as AMV is highly variable, the classification of large numbers of field isolates must usually depend primarily on symptoms and host range on a standard set of indicators (Crill et al., 1971; Hajimorad and Francki, 1988).
2. Vector Transmission Among three isolates of BYDV there was a correlation between closeness of serological relationship and transmission by aphid vectors (Aapola and Rochow, 1971). Pead and Torrance (1989) found that MAbs could be used
to type the three major vector-specific strain groups (now species) of BYDV. On the other hand, an isolate of PLRV that was poorly transmitted by aphids was indistinguish able serologically from readily transmitted isolates (Tamada et al., 1984). There was no correlation between serological relatedness and the ability of English populations of Longidorus attenuatus to transmit different isolates of TBRV (Brown et al., 1989).
3. Multipartite Genomes The ability of multicomponent viruses to complement each other successfully provides a powerful functional criterion indicating relationship. However, this property may not correlate closely with the physical properties of the virus particle or other properties of the virus. For example, certain viruses that have been considered as strains of CPMV (Swaans and van Kammen, 1973) did not successfully complement each other in mixed infection experiments (van Kammen, 1968). Successful complementation has been shown to occur not only between already wellrecognized strains but also between viruses thought to be distinct members of the same group. Such results further complicate the use of complementation tests as a criterion of relationship. For example, Bancroft (1972) demonstrated successful complementation between BMV and CCMV. These are both in the Bromovirus genus, but they have almost totally different host ranges and appear unrelated serologically. The Ilarvirus and Alfamovirus genera have a tripartite genome and a separately encapsulated CP cistron. If the three genomic RNAs are used for infection, the CP RNA, or some CP itself, is required for infectivity (discussed in Box 7.3). The CP or the CP gene of some ilarviruses, for example, TSV, will activate the RNAs of AMV and the reverse combination is also active. However, mixtures of the three-genome RNAs from the two viruses do not complement each other (van Vloten-Doting, 1975; Gonsalves and Fulton, 1977), and there is no sequence similarity between the corresponding RNA segments. Transgenic tobacco plants expressing the TSV CP gene are resistant to infection with TSV but susceptible to AMV. They can be infected with AMV RNAs 1, 2, and 3, demonstrating that the endogenously produced TSV CP can activate the AMV genome, even though it does not protect against this virus (van Dun et al., 1988). The CP of AMV nucleoproteins is specifically removed by the addition of AMV RNA. Similarly, the nucleoproteins of ilar viruses may lose their protein when free viral RNA is added. There is reciprocity in this reaction between certain ilar viruses and AMV (van Vloten-Doting, 1975; Gonsalves and Fulton, 1977). These results have led some workers to suggest that AMV should be placed in the Ilarvirus genus (Section II, I).
4. General Nucleotide Sequence Similarities Using hybridization techniques, there may be complete lack of significant base sequence homology between viruses that on other grounds, such as morphology of the particle and serology, are certainly related (Zaitlin et al., 1977). At the other extreme, Bol et al. (1975) described four strains of AMV with well-characterized differences in biological tests that were virtually indistinguishable in nucleic acid hybridization tests. Strains of CMV are divided into two subgroups based on serology and nucleic acid hybridization, as discussed by Rizzo and Palukaitis (1988). Of 39 strains examined by nucleic acid hybridization, 30 belong in subgroup I and 9 in subgroup II. RNAs belonging to the two subgroups can be reassorted to yield viable recombinants. The RNAs 1 and 2 of representatives of the two groups have been sequenced and compared (Rizzo and Palukaitis, 1988, 1989). Different regions of the RNAs varied in the extent of sequence homology (from 62% to 81%). Strains within the two subgroups cannot be distinguished by the usual nucleic acid hybridization techniques. However, Owen and Palukaitis (1988) used molecular heterogeneity mapping to place 13 of the CMV strains into two groups based on their ability to hybridize to two representative strains. Molecular heterogeneity mapping could distinguish strains within the two groups.
5. 3′ Noncoding Nucleotide Sequences Another approach for discriminating between distinct potyviruses and strains has been explored by Frenkel et al. (1989). They compared the 3′ noncoding nucleotide sequences of 13 potyviruses and found that viruses that were distinct on other grounds had 3′ noncoding sequences of different lengths (189–475 nucleotides). The degree of sequence similarity ranged from 39% to 53%. Such values are comparable to that obtained when the 3′ untranslated regions from unrelated potyviruses are compared and they are probably in the range expected for chance matching. By contrast, the 3′ untranslated regions of sets of viruses recognized on other criteria as related strains were very similar in length and in nucleotide sequence homology (83–99%). WMV-2 and SGMV-N were found to have 78% homology and on this basis were considered strains of the same virus.
6. Serological Relationships Relationships determined by serological methods might, by chance, correlate quite well with any other properties. However, it is reasonable to expect that they may show some correlation with those criteria that also depend on some property of the CP.
Correlations have been reported between degree of relatedness, measured by cross-protection tests, and serological relatedness (e.g., PVX strains, Matthews, 1949 and BYDV isolates, Aapola and Rochow, 1971). On the other hand, there was no correlation between serological relatedness within a group of TNV isolates and their ability to support the replication of three differing isolates of STNV (Kassanis and Phillips, 1970), nor was there any correlation between serological relatedness and symptoms in tobacco for TRSV (Gooding, 1970). MAbs raised against strains of PVX reacted in a complex manner with the strains in different groups based on the reaction of host varieties (Torrance et al., 1986). Nevertheless, a resistance breaking strain could be identified. A series of 10 MAbs raised against PLRV failed to differentiate between strains that caused different symptoms in indicator hosts (Massalski and Harrison, 1987). Traditionally, viruses and strains within the Potyvirus genus have been very difficult to delineate. This has been not only because of the large number of viruses involved but because different tests for relationship gave different answers. The work of Shukla and his colleagues has gone some way toward establishing a sound basis for classification of virus isolates belonging to this group. For example, earlier work suggested that cross-protection did not distinguish some isolates that, on other criteria such as serology, were considered to be separate viruses. Shukla et al. (1989d), using antibodies directed against the N-terminal part of the CP, showed that potyviruses infecting maize, sorghum, and sugarcane in Australia and the United States comprised four distinct viruses. The earlier cross-protection tests fell neatly into place on this basis. Similarly, the kind of cytoplasmic inclusions found with these isolates supported the idea of four distinct viruses. Difficulties remain, however, because some unexpected serological cross-reactions occur between viruses that on other soundly based criteria are regarded as distant members of the Potyvirus genus. The antigenic site for these cross-reactions may reside with a few common amino acids close to the N-terminus of the CP (Shukla and Ward, 1989a,b).
7. Nonstructural Proteins Yeh and Gonsalves (1984) used antisera raised against the inclusion body proteins of two potyviruses to confirm that they were related strains of one virus rather than two distinct viruses. Thornbury and Pirone (1983) showed that the helper component protein of two different potyviruses were serologically distinct. There was no serological relationship between the 35-kDa protein coded for by AMV and the corresponding proteins of three other viruses with a tripartite genome (van Tol and van Vloten-Doting, 1981).
Chapter | 2 Plant Viruses and Their Classification
VI. VIRUSES OF GYMNOSPERMS, PTERIDOPHYTES, ALGAE, AND FUNGI A. Viruses of Gymnosperms A disease of Cycas revoluta has been shown to be due to a virus with a bipartite genome and other properties that place it in the Nepovirus group (Hanada et al., 1986; Han et al., 2002). The virus was readily transmitted by mechanical inoculation to various Chenopodium species. It was also transmitted through the seed of these species. There have been a few reports of pines being infected experimentally with viruses from angiosperms (Fulton, 1969; Jacobi et al., 1992). There are also few reports of naturally occurring virus-like diseases in other gymnosperms but the viral nature of the diseases has not been demonstrated (Schmelzer et al., 1966). For example, a putative cryptic virus has been reported from Pinus sylvestris (Veliceasa et al., 2006). However, because of the presence of substances such as tannins, there are technical difficulties in attempting to isolate viruses from gymnosperms. Retrotransposons are found in conifers (Friesen et al., 2001) with gypsy and copia-like retroelements forming a major component of the gymnosperm genome.
B. Viruses of Ferns No viruses have been reported from bryophytes but several viruses have been reported from ferns. A virus with particles like those of a Tobravirus was found in hart’s tongue fern (Phyllitis scolopendrium (L.) Newn) by Hull (1968); this virus has not been further characterized. A virus infecting Japanese holly fern (Cyrtomium falcatum) has unique properties (Valverde and Sabanadzovic, 2009). This virus has a bipartite RNA genome contained in quasispherical particles 30–40 nm diameter. Both RNAs have been sequenced (Figure 2.12). The genome organization of RNA1 (6228 nt) resembles that of RBDV idaeovirus whereas that of RNA2 (3007 nt) does not resemble any other plant virus. This virus was transmitted by grafting and through the spores of infected plants. A virus that reacted with cucumber mosaic virus antiserum has been described from Adiantum pedatum (Nameth and Steininger, 1997), a full-length Ty3/gypsytype retrotransposon from A. capillus-veneris (Nozue et al., 1997) and a putative potyvirus from Dryopteris felix-mas (Neinhaus et al., 1974). Cheo (1972) reported the replication of TMV in eight fern species.
C. Viruses of Algae An increasing number of viruses (termed virioplankton) have been found infecting both marine and freshwater
eukaryotic algae. By far the greatest numbers of studied viruses have very large particles, double-stranded DNA genomes, and are placed in the family Phycodnaviridae. There are also some algal viruses with smaller particles and RNA genomes.
1. Large Algal Viruses (Phycodnaviridae) (reviewed by Dunigan et al., 2006; Wilson et al., 2009; Van Etten et al., 2010a,b) The family Phycodnaviridae comprises a diverse and rapidly expanding collection of large icosahedral, dsDNA viruses that infect algae. Some, or even many, of these virus species play a major role in the global marine carbon and sulfur cycles. The marine phytoplankton (microalgae) form the base of the marine food web and their photosynthetic activities provide an important carbon sink that influences the global carbon cycle and even climate. Conservative estimates suggest there are somewhere between 100,000 and several million species of algae and that only approximately 40,000 have been identified. Members of the Phycodnaviridae infect these organisms which are considered to be the most important microorganisms in maintaining the balance of life on the planet (Wilson et al., 2009). The most obvious effect of their lytic infection is the disintegration of major algal blooms though they also probably have less obvious impacts on the population dynamics of phytoplankton. Members of the Phycodnaviridae are currently grouped into six genera (named after the hosts they infect): Chlorovirus, Coccolithovirus, Prasinovirus, Prymnesiovirus, Phaeovirus, and Raphidovirus (King et al., 2012), and
JHFMoV RNA1 (~6.2 kb) RNA2 (~3.0 kb)
RBDV RNA1 (~5.5 kb) RNA1 (~2.2 kb)
M T R
FIGURE 2.12 Schematic representation of the genomic organization of JHFMoV compared with RBDV (genus Idaeovirus). The same shading/pattern of positive genome products indicates similar function (MTR, methyltransferase; Hel, helicase; RdRp, RNA-dependent RNA polymerase; MP, movement protein; CP, coat protein). From Valverde and Sabanadzovic (2009), with permission of the publishers.
some of their properties are listed in Table 2.5. The genomes of representatives of the genera Chlorovirus, Phaeovirus, and Coccolithovirus have been sequenced. Two species, the chlorovirus Paramecium bursaria chlorella virus 1 (PBCV-1) (reviewed by Van Etten, 2003; Yamada et al., 2006) and the putative prasinovirus Ostreococcus tauri virus 5 (OtV5) (Derelle et al., 2008) have been studied in most detail. The particles of PBCV-1 are 165–190 nm diameter with the outer capsid comprising 1692 capsomeres arranged in a T = 169d skew icosahedral lattice (see Chapter 3, Section IV for icosahedral symmetry). The particle vertices each have a pocket on the inside which might contain enzymes used for the initial stages of infection and a spike-like structure on the outside of the capsid (Cherrier et al., 2009). The particles have an internal membrane that is required for infection. The genome is closed-linear double-stranded DNA of 330 kbp (Table 2.5) with the two strands being joined by hairpin bends. As with all the sequenced phycodnaviruses, the genome codes for a large number of proteins including many that are unexpected for a virus, e.g., ornithine decarboxylase, hyaluronan synthase, a potassium ion channel protein, and most, if not all, machinery to glycosylate the viral glycoproteins (Van Etten et al., 2010a,b); it also codes for tRNAs and appears to have “captured” some host genes. The PBCV-1 virion contains more than 110 different virus-encoded proteins. PBCV-1 initiates infection by attaching rapidly, specifically, and irreversibly to the chlorella cell wall, probably at a unique virus vertex. This is immediately followed by degradation of the host wall at the point of contact by a virus-encoded and packaged enzyme(s). Following host cell wall degradation, the viral internal membrane is thought to fuse with the host membrane, leading to entry of the viral DNA and virion-associated proteins into the cell, leaving an empty virus capsid attached to the cell wall. This process triggers a rapid depolarization of the host membrane (probably triggered by a virus-encoded potassium channel located in the virus internal membrane) which may function to prevent infection by a second virus. These initial events in PBCV-1 infection are discussed in detail by Thiel et al. (2010). It is likely that the viral DNA and DNA-associated proteins quickly move to the nucleus and early transcription is detected within 5–10 min post infection (p.i.). Shortly after infection, host chromosomal DNA begins to be degraded and the host is reprogrammed to transcribe viral RNAs. Viral DNA replication begins 60–90 min after infection followed by transcription of late genes. Assembly of virus capsids begins in localized regions in the cytoplasm, called virus assembly centers, approximately 2–3 h p.i. These become prominent at 3–4 h p.i. and by 5–6 h p.i., the cytoplasm becomes filled with infectious progeny virus particles. By 6–8 h p.i., localized lysis of the host cell releases progeny with approximately
1000 particles being released per cell, of which about 30% are infectious. OtV5 infection causes lysis of Ostreococcus tauri, the smallest eukaryotic free-living photosynthetic organism known. The icosahedral particles of OtV5 have a diameter of 122 nm, the complete particle containing a central region of uneven electron-dense material (Derelle et al., 2008). The linear dsDNA genome of 186.234 kbp has terminal inverted repeats and contains 268 ORFs; 22% of the predicted proteins are similar to proteins to which a function has been attributed including various enzymes involved in DNA replication. There are several tRNAencoding sequences and also 6l host-like DNA sequences; the latter suggests horizontal transfer but divergence from the host genes indicate that this was not a recent event. The genome did not encode for any site-specific endonucleases which is in accord with the lack of degradation of the host genome on infection. Thus members of the family Phycodnaviridae have several unusual features including the size of the genome and the number of proteins encoded. The fact that they encode many of the enzymes involved in DNA replication and expression suggests that they have limited reliance on the host machinery and might be considered to run counter to the definition of a virus (see Chapter 1, Section II).
2. Small Algal Viruses Skotnicki et al. (1976) described a virus infecting the eukaryotic alga Chara australis. The virus (CAV) has rod-shaped particles and some other properties like those of tobamoviruses. However, its genome is much larger (11 kb, rather than 6.4 kb for TMV), and about 7 kb of the genome has been sequenced, revealing other relationships (Matthews, 1991): (i) the CP of CAV has a composition closer to BNYVV and TRV than to TMV; (ii) the GDDpolymerase motif of CAV is closest to that of BNYVV; and (iii) the two GKT nucleotide-binding motifs found in CAV are arranged in a manner similar to that found in potexviruses. Thus, CAV appears to share features of genome organization and sequences found in several groups of rod-shaped viruses infecting angiosperms. It appears to have strongest affinity with the Furovirus group and has no known angiosperm host. For these reasons it is most unlikely that CAV originated in a recent transfer of some rod-shaped virus from an angiosperm host to Chara. The marine toxic bloom-forming alga, Heterosigma akashiwo, is lysed by a single-stranded RNA virus H. akashiwo RNA virus (HaRNAV) (Tai et al., 2003). The 25 nm icosahedral particles of this virus contain a genome of 8.6 kb which encodes a single ORF with a genome organization similar to picornaviruses (Lang et al., 2004). This virus is the type member of the family Marnaviridae. Shotgun sequencing of whole genomes reverse-transcribed
Ectocarpus siliculousus virus 1 (EsV-1)
Micromonas pusilla virus SP1 (MpV-SP1)
Chrysochomulina brevifilum virus PW1 (CbV-PW1)
Heterosigma akashiwa virus 01 (HaV01)
Haptophycaceae (aka Prymnesiophyaceae) spp
Micromonas pusilla Pyramimonas orientalis
See King et al. (2012). FW, freshwater; MW, marine/coastal water; ND, not determined. PFU/cell. Data from Dunigan et al. (2006) and King et al. (2012).
Emiliania huxleyi virus 86 (EhV- Emiliania huxleyi 86)
Chlorella NC64A Chlorella Pbi
Paramecium bursaria chlorella virus 1 (PBCN-1)
Known Host Range
150–350 Open circular, singlestranded regions
313–379 Closed-linear dsRNA hairpin termini
190 + 25 nm spikes on vertices
Genome Size (kbp) and Latent Period (h) Conformation
Particle Diameter (nsm)
TABLE 2.5 Taxonomy and General Characteristics of Members of Family Phycodnaviridae
from RNA isolated from marine sources including phytoplankton has revealed a wide range of putative RNA viruses, some of which appear to be related to known higher plant virus taxa such as the comoviruses and tombusviruses (Culley et al., 2006; Culley and Steward, 2007; Lang et al., 2009). A dsRNA virus has been found in the alga Micromonas pusilla with a genome was composed of 11 segments ranging between 0.8 and 5.8 kb (Brussaard et al., 2004). This is classified as the type (and only) member of the genus Minoreovirus in the family Reoviridae (King et al., 2012). Straight (~280 nm) and flexuous (~700–900 nm) virus particles have been found associated with dieback symptoms in the brown alga Eklonia radiata in New Zealand (Easton et al., 1997).
virus (King et al., 2012) has similarities to those of carmo viruses (TCV and CarMV) (Preisig et al., 2000). It is suggested that sequence similarities to replicases of potexviruses are shared by the M-dsRNA segment of Sclerotinia sclerotiorum debilitation-associated RNA virus (SsDRV) and Botrytis virus F (genus Mycoflexivirus) (Howitt et al., 2006; Xie et al., 2006) and that that the L-dsRNA segment of SsDRV is related to Hepatitis E virus and Rubi-like viruses (Liu et al., 2009). As noted above, Sclerotinia sclerotium hypovirulence-associated DNA virus 1 is phylogenetically related to geminiviruses. The genomes of some of the fungal viruses appear to be either unencapsidated or are found in pleomorphic vesicles. These perhaps reflect that they are transmitted by anastomosis between fungal hyphae or by zoospores and thus are not exposed to the external environment (Nuss, 2005).
D. Viruses of Fungi (Also Known as Mycoviruses)
1. Phenotypic Effects of Fungal Viruses
Representatives of 16 genera in 14 viral families infect fungi (Table 2.6). As can be seen from Table 2.6, many of these contain dsRNA genomes. As noted in Table 8.1, fungi appear to have a higher proportion of dsRNA genomes than viruses from other kingdoms. The genomes of members of the genus Hypovirus were originally considered to be dsRNA as this form of RNA predominated in the pleomorphic vesicles in infected fungi. However, only one RNA strand is employed in mRNA transcription and ssRNA can initiate infection. It is thought that the dsRNA is a replicative intermediate and that low amounts of the (+)-strand ssRNA is due to RNA silencing (Lin et al., 2007). A number of fungal viruses are listed as unassigned in King et al. (2012). Most have either dsRNA or (+)-strand ssRNA genomes. However, one (Sclerotinia sclerotium hypovirulence-associated DNA virus 1) has an ssDNA genome (Yua et al., 2010). Based on its Rep sequences, this virus is phylogenetically related to geminiviruses but is distinct both in its genome organization and particle morphology. Some of the fungal viruses are in unique genera (e.g., Barnavirus and Hypovirus) and others share genera with viruses from other kingdoms (e.g., Mitovirus shared with viruses infecting plants, vertebrates, and invertebrates, and Edornavirus shared with viruses infecting plants and oomycetes). Parts of the genomes of some mycoviruses show similarities to parts of genomes of viruses from other kingdoms. For example, phylogenetic analysis shows an apparent relationship between the L-dsRNA of the virus causing hypovirulence in the chestnut blight fungus, Cryphonectria parasitica, and the potyvirus genome (Koonin et al., 1991). The C-terminal half of the genome of Diaporthe RNA virus 1 (DRV1) which is an unassigned
In many cases, mycoviruses are cryptic and cause little or no apparent effects on the fungal host. In some cases, mycovirus infection can cause phenotypic effects in the fungal host. Several mycoviruses cause degeneration of edible fungi; for example, infection with La France isometric virus and Mushroom bacilliform virus induces severe symptoms in Agaricus bisporus (Tavantzis et al., 1980; Romaine et al., 1986; Reville et al., 1994) and a dsRNA mycovirus causes degeneration of the oyster mushroom (Pleurotus ostreatus) (Qiu et al., 2010). Virus infection can attenuate the virulence of some plant pathogenic fungi; this property, termed hypovirulence, has raised the possibility of using mycoviruses for the biocontrol of fungal diseases (reviewed by Nuss, 2005). Hypovirulence has been reported for infections by several families of mycoviruses that infect a range of plant pathogenic fungi (Table 2.7) To study the molecular basis of hypovirulence, Allen et al. (2003) monitored changes in the transcriptional profile of C. parasitica on infection by Cryphonectria hypovirus 1-EP713 (CHV1-EP713) using cDNA microarrays. Transcriptional alterations were found in 13% of the unique cDNAs (132 upregulated and 163 downregulated) representing a broad spectrum of biological functions including stress responses, carbon metabolism, and transcriptional regulation. They suggested that hypovirus infection results in persistent reprogramming of a significant portion of the fungal transcriptome. As can be seen from Table 2.7, several viruses may be associated with hypovirulence of a specific fungus which raises the question as to whether the viruses interact. For example, infection of C. parasitica, by CHV1-EP713 or by Mycoreovirus1-Cp9B21 (MyRV1-Cp9B21) results
(reviewed by Ghabrial and Suzuki, 2009)
Drosophila melanogaster copia virus Penicillum chrysogenum virus Vicia faba endornavirus Atkinsonella hypoxylon virus Rosellinia necratix virus Mycoreovirus1 Saccharomyces cerevisiae virus L-A Helminthosporium victoriae virus 109S
Sclerotinia sclerotinium debilitation-associated RNA Capsidless virus Botrytis virus F Mushroom bacilliform virus Cryphonectria hypovirus 1 Saccharomyces 20S RNA narnavirus Cryphonectria mitovirus 1
No true virions
No true virions
Further data can be found in King et al. (2011); Megabirnaviridae is a new family (Adams and Carstens, 2012) and details of the type species are in Chiba et al. (2009). See text for more details.
Saccharomyces cerevisiae TY1 virus
Saccharomyces cerevisiae TY3 virus
TABLE 2.6 Virus Infecting Fungia
Size (kb or kbp)
in different hypovirulence phenotypes. CHV1-EP713 depresses pigmentation and conidiation drastically in contrast to MyRV1-Cp9B21 which has little effect on these processes (Sun et al., 2006). Double infection by these two viruses gave a phenotype similar to that of single infections of CHV1-EP713 but with further decreased levels of host conidiation and vegetative growth and increased levels of MyRV1-Cp9B21 genomic dsRNA accumulation. This suggested a synergistic interaction between the two viruses (see Chapter 9 Section V, I for synergism) and indicated that CHV1-EP713 p29 might be an RNA silencing suppressor (Sun et al., 2006). Three dsRNA segments, designated L-, M- and S-, were detected in Sclerotinia sclerotiorum strain EP-1PN (Li et al., 1999). The L- and M- segments represent two distinct viruses, Sclerotinia sclerotiorum RNA virus L (SsRV-L) and Sclerotinia sclerotiorum debilitation-associated virus (SsDRV) respectively (Xie et al., 2006; Liu et al., 2009). These two viruses can replicate independently with SsRV-L showing little or no adverse effects whereas SsDRV shows a debilitation phenotype. It is not yet known if there is an interaction between the two viruses in joint infections (Liu et al., 2009).
2. Mycoviruses and RNA Silencing (reviewed by Nuss, 2011) RNA silencing and suppression of silencing with higher plant viruses is discussed in Chapter 9. Silencing and silencing suppression have also been shown with fungal viruses (for details see Nuss, 2011). Most of the advances in understanding the fungal system have been made using the hypovirus/C. parasitica system where RNA silencing is a robust antiviral defense response which reduces hypovirus symptoms but also produces a new population of potentially bioactive virus-derived small (vs) RNAs. Hypovirus CHV1/EP713 p29 is a silencing suppressor (Segers et al., 2006) and shows strong similarities with the potyvirus silencing suppressor HC-Pro (Choi et al., 1991). p29 interferes with transcriptional induction of the host RNA silencing defense (Nuss, 2011). The host antiviral RNA silencing response on C. parasitica contributes to hypervirus RNA recombination (Zhang and Nuss, 2008; Nuss, 2011). In hypovirus infections there is accumulation of significant levels of DI RNAs generated from the viral genomic RNA by recombination deletion events. The fungal Dicer and Argonaute genes, dcl2 and agl2, that are induced in response to virus infection and required for the antiviral defense response, significantly contribute to hypovirus RNA recombination and DI RNA production. The distribution of vsRNAs is nonrandom along the hypovirus CHV1/EP713 genome which is suggested to relate to the relative abundance of DI RNAs compared to genomic RNA (Nuss, 2011).
E. Discussion The existence of a (+)-sense ssRNA virus infecting the genus Chara suggests an ancient origin for this type of virus. Other than this example, the meager information about viruses infecting photosynthetic eukaryotes below the angiosperms can tell us very little about the age and course of evolution among the plant viruses. The cycads are regarded as living fossils, being in the record since early Mesozoic times. However, the Nepovirus found in Cycas revoluta is quite likely to have originated in a modern angiosperm, since it readily infects Chenopodium spp. The Phycodnavirus PBCV-1 infecting a Chlorella-like alga is much more likely to be of ancient origin. However, based on structure they do not appear to be possible primitive viruses. They are much larger and more complex than any known viruses infecting angiosperms, with a genome of about 300 kbp and at least 50 structural proteins (Meints et al., 1986).
VII. PLANT VIRUS PURIFICATION There are no generally applicable rules for virus purification. Procedures that are effective for one virus may not work with another apparently similar virus. Even different strains of the same virus may require different procedures for effective isolation. A great deal has been written about purity and homogeneity as they apply to plant viruses. In a chemical sense, there is no such thing as a pure plant virus preparation. Even if a preparation contained absolutely no low- or high-molecular-weight host constituents (which is most unlikely), there are other factors to be considered: i. Most preparations almost certainly consist of a mixture of infective and noninfective virus particles. The latter will probably have one or more breaks in their nucleic acid molecules most of which will have occurred at different places in different particles. ii. Most virus preparations almost certainly consist of a mixture of mutants even though the parent strain greatly predominates. Such mutants will differ in the base sequence of their RNA or DNA in at least one place. If the mutation is in the cistron specifying the CP then this may also differ from the parent strain. iii. Purified preparations of many viruses can be shown to contain one or more classes of incomplete, noninfective particles. iv. The charged groups on viral proteins and nucleic acids will have ions associated with them. The inorganic and small organic cations found in the purified virus preparation will depend very much on the nature of the buffers and other chemicals used during isolation. v. Some of the larger viruses appear to cover a range of particle sizes having infectivity.
Disease Chestnut blight
Chestnut blight Dollar spot disease of turfgrass Dutch elm disease Victoria blight of oats
White root rot
Rice blast Diaporthe disease of stone fruit Scab disease of small grain
Cryphonectria parasitica (Ascomycete)
Cryphonectria nitschkei (Ascomycete)
Sclerotinia homoeocarpa (Ascomycete)
Ophiostoma ulmi, O. novo-ulmi (Ascomycete)
Helminthosporium victoriae (Ascomycete)
Rosellinia necatrix (Ascomycete)
Magnaporthe oryzae (Ascomycete)
Diaporthe ambigua (perjuncta) (Ascomycete)
Fusarium graminearum (Ascomycete)
Cryphonectria parasitica mycoreovirus1 (CpMRV-1)
Reoviridae/Mycoreovirus (group 1)
Rosellinia necatrix megabirnavirus 1 (RnMBV1)c
Diaporthe RNA virus (DRV)
Magnaporthe oryzae virus 1 (MoV1)
Rosellinia necatrix virus 1 (RnV1)
Rosellinia necatrix mycoreovirus 3 (RnMYRV-3)
Helminthosporium victoriae virus 145S (HvV-145S)
No particles observed
Isometric 35 nm diameter
Isometric 50 nm diameter
Isometric 35–40 nm diameter
Isometric 50 nm diameter— spikes on vertices
Isometric 40 nm diameter
Isometric 40 nm diameter
ds RNA, 1 segment 7.5 kbp; RdRp related to Cryphonectria viruses and BYMV
ssRNA, 1 segment 4.1 kbp; RdRp related to tombusvirusesd
dsRNA, 4 segmentsu
dsRNA, 2 segments
dsRNA, 2 segments
dsRNA, 12 segments
dsRNA, 4 segments
dsRNA, 1 segment c 5 kbp
dsRNA, 1 segment 2.6 kbpb
dsRNA, 1 segment 2.6 kbpb
ds RNA, 4 segments
dsRNA, 11 segments
ssRNA, 2.3–2.7 kb
Possibly ssRNA, 9.8–12.7 kb
Isometric 40 nm diameter
Isometric 50 nm diameter— spikes on vertices
Associated with mitochondria—no particles observed
No particles observed but vesicles
Reoviridae/Mycoreovirus (group 2)
Helminthosporium victoriae virus 190S (HvV-190S)
Ophiostoma mitovirus (OMV) various strains
Ophiostoma mitovirus 3a (OMV-3a)
Cryphonectria nitschkei chrysovirus 1 (CnV1)
Cryphonectria parasitica mitovirus 1 (CMV1)
Cryphonectria parasitica virus 1 (CHV1)
TABLE 2.7 Mycovirus-Mediated Hypovirulence of Plant Pathogenic Fungi
Data from Nuss (2005) and King et al. (2012). a ND = not determined; RdRp = RNA-dependent RNA polymerase. b Genome is possibly ssRNA. c Described by Chiba et al. (2009); classification by Adams and Carstens (2012). d Moleleki et al. (2011). e Urayama et al. (2010). f Wu et al. (2007).
Botrytis cinerea (Ascomycete)Gray mold
Many diseases and wide Unclassified host range
Sclerotinia sclerotiorum (Ascomycete)
Rhizoctonia disease of potato
Rhizoctonia solani (Basidiomycete)
TABLE 2.7 (Continued) Mycovirus-Mediated Hypovirulence of Plant Pathogenic Fungi
Botrytis mitovirus 1 (BMV-1)
Sclerotinia sclerotiorum hypovirulence-associated DNA virus 1 (SaHADV-1)
Rhizoctonia virus M2 (RVM2)
No particles observed
Geminate isometric particles
No particles observed Associated with both cytoplasm and mitochondria
dsRNA, 1 segment 3.0 kbpf
ssDNA, 2166 nt, Rep related to geminiviruses
dsRNA, 1 segment 3.6 kbp
Chapter | 2 Plant Viruses and Their Classification
vi. A variable proportion of the virus particles may be altered in some way during isolation. Enzymes may attack the CP. For example, extracts of bean (Phaseolus vulgaris) contain a carboxypeptidaselike enzyme that removes the terminal threonine from TMV CP (Rees and Short, 1965). Proteolysis at defined sites may give rise, during isolation of the virus, to a series of CP molecules of less than full size (e.g., with SNMoV, Chu and Francki, 1982, and BaYMV, Ehlers and Paul, 1986). CPs may undergo chemical modification when leaf phenols are oxidized (Pierpoint et al., 1977). More complex viruses, such as the reoviruses, may lose part of their structure during isolation (Hatta and Francki, 1977). Thus, for plant viruses, purity and homogeneity are operational terms defined by the virus and the methods used. A virus preparation is pure for a particular purpose if the impurities, or variations in the particles present, do not affect the particular properties being studied or can be taken account of in the experiment. Effective isolation procedures have now been developed for many plant viruses. Rather than describe these in detail, I shall consider, in general terms, the problems involved in virus isolation. Detailed protocols for isolation procedures for a number of viruses are given in numerous publications including Hull (1985), Walkey (1991), Stace-Smith and Martin (1993), Dijkstra and de Jager (1998), and in various chapters in Foster and Taylor (1998).
A. Choice of Plant Material 1. Assay Host During the development of an isolation procedure, it is useful, but not always possible, to be able to assay fractions for infectivity. Of course, this is best done with a local lesion host. Great accuracy usually is not necessary in the preliminary assays, but reliability and rapid development of lesions are a great advantage. If no local lesion host is available, then assays must be done using a systemic host. Assays by the injection of insect vectors sometimes have been used where mechanical transmission is impossible. Electron microscopy or, if antisera or probes are available, dot blot ELISA or dot blot nucleic acid hybridization can be used to follow the progress of purification.
2. Propagation Host The choice of host plant for propagating a virus may be of critical importance for its successful isolation. Various points have to be considered in the choice of a propagating host: The host plant should be easy to grow, preferably from seed. However, care should be taken that there is no
seed-transmitted virus such as SoMV, which is highly transmitted in seed of Chenopodium spp. ● The virus should reach a high concentration in the host. ● The host should not contain high amounts of substances such as phenolic materials, organic acids, mucilages and gums, certain proteins, and enzymes, particularly ribonucleases that can inhibit or irreversibly precipitate the virus. For example, many stone fruit viruses were very difficult to isolate from their natural hosts, since most members of the Rosaceae contain high concentrations of tannins in their leaves. Discovery of alternative non-rosaceaous hosts, for example, cucumber (Cucumis sativus L.), has allowed the isolation of several such viruses. ● The host plant constituents should be easily separable from the virus during purification. For instance, certain legumes, e.g., Canavalia ensiformis, contain large amounts of Rubisco (fraction 1 protein), often in an aggregated form that can copurify with rod-shaped viruses. In practice, species of Chenopodium, Cucumis, Nicotiana, Petunia, Phaseolus, and Vigna have been found suitable for the propagation of a large number of viruses. The plant species used, the conditions under which it is grown, and the time at which it is harvested should be chosen to maximize the starting concentration of infectious virus. For many viruses, concentration rises to a peak after a few days or weeks and then falls quite rapidly (Figure 2.13). Sometimes the distribution of virus within the plant is so uneven that it is worthwhile to dissect out and use only those tissues showing prominent symptoms. Viruses frequently occur in much lower concentration in the midrib than in the lamina of the leaf. If the midrib and petiole are large, it may pay to discard them. In special situations, dissection of tissue is almost essential, for example, with WTV and other viruses causing tumors where the virus is found associated with the tumor tissue. Another reason for harvesting only certain parts of the infected plant may be to avoid high concentrations of inhibitory substances or materials that adsorb to the virus and are later difficult to remove. Such materials frequently occur in lower concentration in new young growth. In certain hosts, virus can only be isolated from such tissue. Similarly, root tissue may sometimes provide more favorable starting material than leaves (Ford, 1973), although virus concentration is usually lower in roots. For some viruses, flowers may provide a suitable source of virus at high concentration. The possibility that the host used to culture a virus may already harbor another virus or become infected with one must always be borne in mind. Contamination of greenhouse-grown plants with unwanted viruses is not at all uncommon. Strains of TMV, PVX, and TNV may be particularly prevalent, especially in greenhouses that have
8 10 12
8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 Leaf number
FIGURE 2.13 Change in concentration of PVA with age in leaves of a single Samsun tobacco plant inoculated on leaf number 1. Virus concentration was measured serologically. From week 15 onward, older leaves had died. Numbers beside graphs show weeks after infection. From Bartels (1954), with permission of the publishers.
been used for virus work for some time. It is not necessarily sufficient to use a host that is only a local lesion host for such contaminating viruses (e.g., N. glutinosa for TMV). Very small amounts of such a resistant virus may become differentially concentrated during isolation of a second virus.
3. Extraction Medium Once infected plant cells are broken, and the contents released and mixed, the virus particles find themselves in an environment that is abnormal. Thus, it is often necessary to use an artificial extraction medium designed to preserve the virus particles in an infectious, intact, and unaggregated state during the various stages of isolation. The conditions that favor stability of purified virus preparations may be different from those needed in crude extracts or partially purified preparations (Brakke, 1963). Moreover, different factors may interact strongly in the extent to which they affect virus stability. The main factors to be considered in developing a suitable medium are as follows. a. pH and Buffer System Many viruses are stable over a rather narrow pH range, and the extract must be maintained within this range. As most viruses have an acid isoelectric point, buffer solutions with pH values in the range of 7–8 should prevent them from precipitating. However, the structural integrity of some viruses (e.g., bromoviruses) is lost on exposure to pHs in this range. The use of a buffer of pH 5 will precipitate many normal plant proteins and is advantageous if the virus itself is not precipitated. Buffers commonly used for virus extraction include borate, citrate, phosphate, and Tris.
b. Metal Ions and Ionic Strength Some viruses require the presence of divalent metal ions (Ca2+ or Mg2+) for the preservation of infectivity and even for the maintenance of structural integrity. Ionic strength may be important. Some viruses fall apart in media of ionic strength below about 0.2 M, while others are unstable in media above this molarity. AMV particles may be precipitated by Mg2+ concentrations above 1 mM and degraded by concentrations above 0.1 M (Hull and Johnson, 1968). For some viruses, NaEDTA may be included to minimize aggregation by divalent metal ions. On the other hand, NaEDTA disrupts certain viruses. Decisions on the pH, presence of chelating agent, and ionic strength should take account of the factors stabilizing the virus particle (see Chapter 3, Section V). c. Reducing Agents and Substances Protecting Against Phenolic Compounds Reducing agents such as ascorbic acid, cysteine hydrochloride, 2-mercaptoethanol, sodium sulfite, or sodium thioglycollate are frequently added to extraction media. Dithiothreitol (Cleland’s reagent) is a useful reducing agent as it has little tendency to be oxidized by air. These materials assist in preservation of viruses that readily lose infectivity through association with products resulting from oxidation of plant extracts and also may reduce adsorption of host constituents to the virus. Phenolic materials especially may cause serious difficulties in the isolation and preservation of viruses. Several methods have been used more or less successfully to minimize the effects of phenols on plant viruses during isolation: (i) cysteine or sodium sulfite added to the extraction medium both probably act by inhibiting the phenol oxidase and by combining
Chapter | 2 Plant Viruses and Their Classification
with the quinone (Pierpoint, 1966); (ii) polyphenoloxidase is a copper-containing enzyme. Two chelating agents with specificity for copper, diethyldithiocarbamate, and potassium ethyl xanthate have been used to obtain infectious preparations of several viruses (e.g., PNRSV; Barnett and Fulton, 1971); the former is also a reducing agent (iii) materials that compete with the virus for phenols have sometimes been used. For example, Brunt and Kenten (1963) used various soluble proteins and hide powder to obtain infective preparations of CSSV from cocoa leaves. Synthetic polymers containing the amide link required for complex formation with tannins have been used effectively to bind these materials. The most important of these is polyvinyl pyrrolidone (PVP). d. Additives That Remove Plant Proteins and Ribosomes Many viruses lose infectivity fairly rapidly in vitro. One reason for this loss may be the presence of leaf ribonucleases in extracts or partly purified preparations. Dunn and Hitchborn (1965) made a careful study of the use of magnesium bentonite as an additive in the isolation of various viruses. They found that, under appropriate conditions, contamination of the final virus product with nucleases was reduced or eliminated. In addition, ribosomes, 19S protein, and green particulate material from fragmented chloroplasts were readily adsorbed by bentonite, provided Mg2+ concentration was 10−3 M or greater. However, variation occurs in the activity of different batches of bentonite, and the material must be used with caution as some viruses are degraded in its presence. Charcoal may be used to adsorb and remove host materials, particularly pigments. Subsequent filtration to remove charcoal may lead to substantial losses of virus adsorbed in the filter cake. NaEDTA at 0.01 M in pH 7.4 buffer will cause the disruption of most ribosomes, preventing their cosedimentation with the virus. This substance can be used only for viruses that do not require divalent metal ions for stability. e. Enzymes Enzymes have been added to the initial extract for various purposes. Thus, Adomako et al. (1983) used pectinase to degrade mucilage in extracted sap of cocoa leaves prior to precipitation of CSSV. Improved yield of a virus limited to phloem tissue was obtained when fibrous residues were incubated with Driselase and other enzymes (Singh et al., 1984). This material contains pectinase and cellulose and presumably aids in the release of virus that would otherwise remain in the fiber fraction. The enzymes also digest materials that would otherwise coprecipitate with the virus. Jones et al., (1991) found that digestion of rice tissue with Celluclast gave more reliable yields of the two rice tungro viruses than did digestion with Driselase.
Treatment with trypsin at an optimum concentration can markedly improve the purification of TuMV (Thompson et al., 1988). f. Detergents and Other Additives Nonionic detergents such as Triton X-100 or Tween 80 are often used in the initial extraction medium to assist in release of virus particles from insoluble cell components and to dissociate cellular membranes that may contaminate or occlude virus particles. However, detergents should not be used with enveloped viruses. The particles of some viruses, such as caulimoviruses, may be contained within inclusion bodies that have to be disrupted. Hull et al. (1976) showed that caulimovirus inclusion bodies were disrupted on treatment with urea and Triton X-100.
4. Extraction Procedure Freezing of the plant tissue, say by liquid N2, before extraction enables disruption of vascular tissues releasing phloem-limited viruses and facilitates subsequent removal of host materials. For some viruses, however, freezing may have a deleterious effect. A variety of procedures are used to crush or homogenize the virus-infected tissue. These include (i) a pestle and mortar, which are useful for small-scale preparations, (ii) various batch-type food blenders and juice extractors, which are useful on an intermediate scale, and (iii) roller mills, colloid mills, and commercial meat mincers, which can cope with kilograms of tissue. For long, fragile, rodshaped viruses, grinding in a pestle and mortar may be the safest procedure to minimize damage. For instance, BYV particles were broken using a blender (Bar-Joseph and Hull, 1974). The addition of acid-washed sand greatly improves the efficiency of extraction. If an extraction medium is used, it is often necessary to ensure immediate contact of broken cells with the medium. The crushed tissue is usually expressed through muslin or cheesecloth.
5. Preliminary Isolation of the Virus a. Clarification of the Extract In the crude extract, the virus is mixed with a variety of cell constituents that lie in the same broad size range as the virus and that may have properties that are similar in some respects. These particles include ribosomes, Rubisco (fraction I) proteins from chloroplasts, which has a tendency to aggregate, phytoferritin, membrane fragments, and fragments of broken chloroplasts. Also present are unbroken cells, cell wall fragments, all the smaller soluble proteins of the cell, and low-molecular-weight solutes. The first step in virus isolation is usually designed to remove as much of the macromolecular host material
as possible, leaving the virus in solution. The extraction medium may be designed to precipitate ribosomes or to disintegrate them. The extract may be subject to some treatment such as heating to 50–60°C for a few minutes, freezing and thawing, acidification to a pH of less than 5, or the addition of K2HPO4 to coagulate much of the host material. The treatment has to be chosen on a case-bycase basis as it may damage the virus. For some viruses, organic solvents such as chloroform give very effective precipitation of host components. For others, the extract can be shaken with n-butanol-chloroform, which denatures much host material. The treated extract is then subjected to centrifugation at fairly low speed (e.g., 10–20 min at 5000–10,000 g). This treatment sediments cell debris and coagulated host material. With the butanol-chloroform system, centrifugation separates the two phases, leaving virus in the aqueous phase and much denatured protein at the interface. It should be noted that although some viruses can withstand the butanol-chloroform treatment, quite severe losses may occur with others. Chloroform alone gives a milder treatment than does a butanol-chloroform mixture. Of course, solvents and detergents cannot be used for viruses that have a membrane. It must be remembered that there are safety considerations to be taken into account in using large volumes of these organic solvent. For many viruses, it pays to carry out the isolation procedure at 4°C and as fast as possible once the leaves have been collected. On the other hand, some viruses occur in membrane-bound structures or other structures within the cell. It may take time for the virus to be released from these after the leaf extract is made. A low-speed centrifugation soon after extraction may then result in much virus being lost in the first pellet, but, on the other hand, this can be used to concentrate the virus. b. Concentration of the Virus and Removal of Low-Molecular-Weight Materials i. High-Speed Sedimentation. Centrifugation at high speed for a sufficient time will sediment the virus. Provided the particular virus is not denatured by the sedimentation, it can be brought back into solution in an active form. This is a very useful step, as it serves the double purpose of concentrating the virus and leaving behind low-molecularweight materials. However, high-speed sedimentation is a physically severe process that may damage some particles (e.g., some reoviruses, Long et al., 1976). Following highspeed sedimentation, some viruses remain as characteristic aggregates when the pellets are resuspended. The virus particles in these aggregates may be quite firmly bound together (Tremaine et al., 1976). If host membranes are involved in the binding, a nonionic detergent may help to release virus particles. Sedimentation of viruses occurring in very low concentration may result in very poor recoveries. The major process causing losses appears to be the
dissolving and redistribution of the small pellet of virus as the rotor comes to rest (McNaughton and Matthews, 1971). Resuspending particles from the surface of the pellet can lead to preferential losses of more slowly sedimenting components (Matthews, 1981). ii. Precipitation with Polyethylene Glycol. Hebert (1963) showed that certain plant viruses could be preferentially precipitated in a single-phase polyethylene glycol (PEG) system, although some host DNA may also be precipitated. Since that time, precipitation with PEG has become one of the commonest procedures used in virus isolation. The exact conditions for precipitation depend on pH, ionic strength, and concentration of macromolecules. Its application to the isolation of any particular virus is empirical, although attempts have been made to develop a theory for the procedure (Juckes, 1971). PEG precipitation is applicable to many viruses, even fragile ones. For example, the procedure gave a good yield of intact particles of CTV, a fragile rod-shaped virus (Lee et al., 1987). It has the advantage that expensive ultracentrifuges are not required, although differential centrifugation is often used as a second step in purification procedures. iii. Density Gradient Centrifugation. Many viruses, particularly rod-shaped ones, may form pellets that are very difficult to resuspend. Density gradient centrifugation offers the possibility of concentrating such viruses without pelleting. A density gradient is illustrated in Figure 13.5. The following modification of the density gradient procedure may sometimes be used even with angle rotors for initial concentration of a virus without pelleting at the bottom of the tube. A cushion of a few milliliters of dense sucrose is placed in the bottom of the tube, this being overlaid with a column of low-density sucrose about 2 cm deep, and the rest of the tube is filled with clarified virus extract. Under appropriate conditions, virus can be collected from the region of the interface between the two sucrose layers. In a similar application, a cushion of medium density sucrose is placed in the bottom of the tube before centrifugation; this prevents contamination of pellets with chlorophyll material. Density gradient centrifugation is used in the isolation procedure for many viruses. iv. Salt Precipitation or Crystallization. Salt precipitation was commonly employed before high-speed centrifuges became generally available. It is still a valuable method for viruses that are not inactivated by strong salt solutions. Ammonium sulfate at concentrations up to about one-third saturation is most commonly used, although many other salts will precipitate viruses or give crystalline preparations. After standing for some hours or days, the virus is centrifuged down at low speed and resuspended in a small volume of a suitable medium.
Chapter | 2 Plant Viruses and Their Classification
v. Precipitation at the Isoelectric Point. Many proteins have low solubility at or near their isoelectric points. Isoelectric precipitation can be used for viruses that are stable under the conditions involved. The precipitate is collected by centrifugation or filtration and is resuspended in a suitable medium. vi. Dialysis. Dialysis through cellulose membranes can be used to remove low-molecular-weight materials from an initial extract and to change the medium. It is more usually employed to remove salt following salt precipitation or crystallization, or following density gradient fractionation in salt or sucrose solutions, as discussed in the next section.
6. Further Purification of the Virus Preparation Virus preparations taken through one step of purification and concentration will still contain some low- and high-molecular-weight host materials. Further purification steps can remove more of these. The procedure to be used will depend very much on the stability of the virus, the scale of the preparation, and the purpose for which the preparation is required. Sometimes highly purified preparations can be obtained by repeated application of the same procedure. For example, a preparation may be subjected to repeated crystallization or precipitation from ammonium sulfate, or may be given several cycles of high- and low-speed sedimentation. The latter procedure leads to the preferential removal of host macromolecules because they remain insoluble when the pellets from a high-speed sedimentation are resuspended. Losses of virus often occur with this procedure, either because some, or all, of the virus itself also becomes insoluble or because insufficient time is allowed for the virus to resuspend. This is a particularly a difficulty with some rod-shaped viruses. Losses may also occur from virus resuspending before the supernatant fluid is removed. Such losses may be severe with very small pellets, where the surface-to-volume ratio is high. Generally, during an isolation it is useful to apply at least two procedures that depend on different properties of the virus. This is likely to be more effective in removing host constituents than repeated application of the same procedure. a. Density Gradient Centrifugation One of the most useful procedures for further purification and for assay, particularly of less stable viruses, is density gradient centrifugation which was developed by Brakke (1951, 1960). This technique has proved to be highly versatile and has been widely used in the fields of virology and molecular biology. It has largely replaced the analytical ultracentrifuge in studies on viruses. A centrifuge tube is partially filled with a solution having a decreasing density from the bottom to the top of the tube. For plant viruses, sucrose is commonly used to form the gradient, and the virus solution is layered on top of the gradient. With gradients formed with cesium salts, the virus particles
may be distributed throughout the solution at the start of the sedimentation or they may be layered on top of the density gradient. Details of density gradient centrifugation are described in Chapter 13, Section II, B, 2. The various forms of density gradient centrifugation can give some indication of the purity of the preparation. They also allow a correlation between particles and infectivity to be made, and frequently reveal the presence of noninfective virus-like particles or multicomponent viruses. Bands of a single component may spread more widely in the gradient than is apparent from a trace of optical density. Several cycles of density gradient centrifugation may be necessary to obtain components reasonably free of mutual contamination. Work with multicomponent viruses has shown that it may be extremely difficult to obtain one component completely free of another, even by repeated density gradient fractionation. Density gradients prepared from the nonionic medium Nycodenz were used effectively for further purification of several viruses (Gugerli, 1984). Brakke and Dayly (1965) showed that zone spreading of a major component by nonideal sedimentation can cause zone spreading of a minor component that the major component overlaps. The way in which zones are removed from the gradient may have a marked effect on the extent of cross-contamination. Strong solutions of salts such as CsCl are also effective gradient materials for viruses that are sufficiently stable. Successive fractionation in two different gradients may sometimes give useful results. The effective buoyant density and the stability of a virus in strong CsCl solutions may depend markedly on pH and on other ions present (Matthews, 1974). Viruses that are unstable in CsCl may be stable in Cs2SO4 (Hull, 1976). When a virus preparation is subjected to density gradient centrifugation in CsCl or Cs2SO4, multiple bands may be formed. These may be due to the presence of components containing differing amounts of RNA, as is found with TYMV (Keeling et al., 1979). Virus particles containing uniform amounts of nucleic acid may sometimes form multiple bands in CsCl or other salt gradients because of such factors as differential binding of ions (Hull, 1977; Noort et al., 1982). b. Gel Filtration Filtration through agar gel or Sephadex may offer a useful step for further purification of viruses that are unstable to the pelleting involved in high-speed centrifugation. However, such a step will dilute the virus. c. Immunoaffinity Columns Monoclonal antiviral antibodies can be bound to a support matrix such as agarose to form a column that will specifically bind the virus from a solution passed through the column. Virus can then be eluted by lowering the pH. Clarified
plant sap may destroy the reactivity of such columns (Ronald et al., 1986), while low pH may damage the virus. To avoid such treatment, de Bortoli and Roggero (1985) developed an electrophoretic elution technique. The use of such columns is probably justified only in special circumstances. d. Chromatography Chromatographic procedures have occasionally been used to give an effective purification step for partially purified preparations. McLean and Francki (1967) used a column of calcium phosphate gel in phosphate buffer to purify LNYV. OBDV was purified using cellulose column chromatography (Banttari and Zeyen, 1969), while Smith (1987) used fast protein liquid chromatography to separate the two electrophoretic forms of CPMV. e. Further Concentration of the Virus and Removal of Contaminants At various stages in the isolation of a virus, it is necessary to concentrate virus and remove salts or sucrose. For viruses that are stable to pelleting, high-speed centrifugation is commonly employed for the concentration of virus and the reduction of the amount of low-molecular-weight material. Ordinary dialysis is used for removal or exchange of salts. For unstable viruses, some other procedures are available. A dry gel powder (e.g., Sephadex) can be added to the preparation, or the preparation in a dialysis bag can be packed in the dry gel powder and placed in the refrigerator for a period of hours, while water and ions are absorbed through the tubing by the Sephadex. Ultrafiltration under pressure through a membrane can be used to concentrate larger volumes and to remove salts. Pervaporation, in which virus solution in a dialysis bag is hung in a draft of air, may lead to loss of virus due to local drying on the walls of the tube and to the concentration of any salts that are present.
7. Storage of Purified Viruses Storage of purified preparations of many plant viruses for more than a few days may present a problem. It is often best to avoid long-term storage by using the preparations as soon as possible after they are made. Under the best of conditions, most viruses except TMV lose infectivity on storage at 4°C in solution or as crystalline preparations under ammonium sulfate. Such storage allows fungi and bacteria to grow and contaminate preparations with extraneous antigens and enzymes. Addition of low concentrations of sodium azide, thymol, or EDTA will prevent growth of microorganisms but EDTA may affect virus structure. Preparations may be stored in liquid form at low temperature by the addition of an equal volume of glycerol. By careful attention to the additives used in the medium (a suitable buffer plus some protective protein, sugar, or polysaccharide), it may be possible to retain
infectivity in deep-frozen solutions or as frozen dried powders for fairly long periods (Fukumoto and Tochihara, 1984). Preparations to be used for analytical studies on protein or nucleic acid are best stored as frozen solutions, after the components have been separated. In solutions containing more than about 10 mg/ml of viruses such as TYMV, virus particles interact quite strongly and spontaneously degrade, especially if the ionic strength is low.
8. Discussion and Summary Different viruses vary over a 10,000-fold range in the amount of virus that can be extracted from infected tissue (from 0.4 or less to 4000 μg/g fresh weight). They also vary widely in their stability to various physical, chemical, and enzymatic agents that may be encountered during isolation and storage. For these reasons, isolation procedures have to be optimized for each virus, or even each strain of a virus. Important factors for the successful isolation of a virus are (i) choice of host plant species and conditions for propagation that will maximize virus replication and minimize the formation of interfering substances; and (ii) an extraction medium that will protect the virus from inactivation or irreversible aggregation. Most viruses can be isolated by a combination of two or more of the following procedures: high-speed sedimentation, density gradient fractionation, precipitation using PEG, salt precipitation or crystallization, gel filtration, and dialysis. Density gradient fractionation of the purified preparation, with combined physical and biological examination of particles in fractions from the gradient, can usually best achieve positive identification of the infectious virus particle or particles. When a new virus or strain is being isolated, it is essential to back-inoculate the isolated virus to the original host to demonstrate that it, and it alone, is in fact the cause of the original disease. Attention must also be paid to the precise conditions under which purified viruses are stored, as many viruses may lose infectivity and undergo other changes quite rapidly following their isolation. As noted earlier, in a chemical sense there is no such thing as a pure virus preparation. Purity and homogeneity are operational terms. A virus preparation is pure for a particular purpose if the impurities or inhomogeneities in it do not interfere with the objectives of the experiment.
REFERENCES Aapola, A.I.E., Rochow, W.F., 1971. Relationships among three isolates of barley yellow dwarf virus. Virology 46, 127–141. Acosta-Leal, R., Fawley, M.W., Rush, C.M., 2008. Changes in the intraisolate genetic structure of beet necrotic yellow vein virus populations associated with plant resistance breakdown. Virology 376, 60–68.
Chapter | 2 Plant Viruses and Their Classification
Adam, G., Hsu, H.T., 1984. Comparison of structural proteins from two potato yellow dwarf viruses. J. Gen. Virol. 65, 991–994. Adams, A.N., Barbara, D.J., 1982. The use of F(ab′)2-based ELISA to detect serological relationships among carlaviruses. Ann. Appl. Biol. 101, 495–500. Adams, M.J., Carstens, E.B., 2012. Ratification vote on taxonomic proposals to the International Committee on Taxonomy of Viruses. Arch. Virol. 157, 1411–1422. Adams, M.J., Antoniw, J.F., Fauquet, C.M., 2005. Molecular criteria for genus and species discrimination within the family Potyviridae. Arch. Virol. 150, 459–479. Adanson, M., 1763. Familles des Plantes, vol. 1. Vincent, Paris. Adomako, D., Lesemann, D.E., Paul, H.L., Owusu, G.K., 1983. Improved methods for the purification and detection of cocoa swollen shoot virus. Ann. Appl. Virol. 103, 109–116. Allen, T.D., Dawe, A.L., Nuss, D.L., 2003. Use of cDNA microarrays to monitor transcriptional responses of the chestnut blight fungus Cryphonectria parasitica to infection by virulence-attenuating hypoviruses. Eukar. Cell 2, 1253–1265. Allison, R.F., Dougherty, W.G., Parks, T.D., Willis, L., Johnston, R.E., et al., 1985. Biochemical analysis of the capsid protein gene and capsid protein of tobacco etch virus: N-terminal amino acids are located on the virion’s surface. Virology 147, 309–316. Allison, R.F., Janda, M., Ahlquist, P., 1988. Infectious in vitro transcripts from cowpea chlorotic mottle virus cDNA clones and exchange of individual RNA components with brome mosaic virus. J. Virol. 62, 3581–3588. Al-Moudallal, Z., Briand, J.P., Van Regenmortel, M.H.V., 1982. Monoclonal antibodies as probes of the antigenic structure of tobacco mosaic virus. EMBO J. 1, 1005–1010. Altschuh, D., Lesk, A.M., Bloomer, A.C., King, A., 1987. Correlation of coordinated amino acid substitutions with function in viruses related to tobacco mosaic virus. J. Mol. Biol. 193, 693–707. Angenent, G.C., Linthorst, H.J.M., van Belkum, A.F., Cornelissen, B.J.C., Bol, J.F., 1986. RNA2 of tobacco rattle virus strain TCM encodes an unexpected gene. Nucleic Acids Res. 14, 4673–4682. Angenent, G.C., Verbeek, H.B.M., Bol, J.F., 1989. Expression of the 16K cistron of tobacco rattle virus in protoplasts. Virology 169, 305–311. Atabekov, J.G., Dorokhov, Y.L., 1984. Plant virus-specific transport function and resistance of plants to viruses. Adv. Virus Res. 29, 313–364. Atabekova, T.I., Taliansky, M.E., Atabekov, J.G., 1975. Specificity of protein–RNA and protein–protein interaction upon assembly of TMV in vivo and in vitro. Virology 67, 1–13. Baldauf, S.L., 2003. Phylogeny for the faint of heart: a tutorial. Trends Genet. 19, 345–351. Ball, L.A., 2005. The universal taxonomy of viruses in theory and practice. ICTV Eighth Report, pp. 3–7. Bancroft, J.B., 1972. A virus made from parts of the genomes of brome mosaic and cowpea chlorotic mottle viruses. J. Gen. Virol. 14, 223–228. Bancroft, J.B., McLean, J.D., Rees, M.W., Short, M.N., 1971. The effect of an arginyl to a cysteinyl replacement on the uncoating behaviour of a spherical plant virus. Virology 45, 707–715. Banttari, E.E., Zeyen, R.J., 1969. Chromatographic purification of the oat blue dwarf virus. Phytopathology 59, 183–186. Bar-Joseph, M, Hull, R., 1974. Purification and partial characterization of sugar beet yellows virus. Virology 62, 552–562. Barker, H., Harrison, B.D., 1978. Double infection, interference and superinfection in protoplasts exposed to two strains of raspberry ringspot virus. J. Gen. Virol. 40, 647–658.
Barnett, O.W., Fulton, R.W., 1971. Differential response of Prunus necrotic ringspot and tulare apple mosaic viruses to stabilizing agents. Virology 46, 613–619. Bartels, R., 1954. Serologische Untersuchungen überdas Verhalten des Kartoffel-A-virus in Tabacpflanzen. Phytopathology Z 21, 395–406. Bath, J.E., Tsai, J.H., 1969. The use of aphids to separate two strains of pea enation mosaic virus. Phytopathology 59, 1377–1380. Bawden, F.C., Pirie, N.W., 1956. Observations on the anomalous proteins occurring in extracts of plants infected with strains of tobacco mosaic virus. J. Gen. Microbiol. 14, 460–477. Bennett, C.W., 1955. Recovery of water pimpernel from curly top and the reaction of recovered plants to reinoculation with different virus strains. Phytopathology 45, 531–536. Berger, P.H., Toler, R.W., 1983. Quantitative immunoelectrophoresis of panicum mosaic virus and strains of St. Augustine decline. Phytopathology 73, 185–189. Berzofsky, J.A., 1985. Intrinsic and extrinsic factors in protein antigenic structure. Science 229, 932–940. Black, L.M., 1941. Specific transmission of varieties of potato yellowdwarf virus by related insects. Am. Potato J. 18, 231–233. Bol, J.F., Brederode, F.T., Janze, G.C., Rauh, D.K., 1975. Studies on sequence homology between the RNAs of alfalfa mosaic virus. Virology 65, 1–15. Boonham, N., Barker, I., 2000. Virus strain discrimination using recombinant antibodies. Dis. Markers 16, 95–97. Bos, L., 1970. The identification of three new viruses isolated from Wisteria and Pisum in the Netherlands and the problem of variation within the potato virus Y group. Neth. J. Plant Pathol. 76, 8–46. Bos, L., 1999. The naming of viruses: an urgent call to order. Arch. Virol. 144, 631–636. Bos, L., 2000. Structure and typography of virus names. Arch. Virol. 145, 429–432. Brakke, M.K., 1951. Density gradient centrifugation: a new separation technique. J. Am. Chem. Soc. 73, 1847–1848. Brakke, M.K., 1960. Density gradient centrifugation and its application to plant viruses. Adv. Virus Res. 7, 193–224. Brakke, M.K., 1963. Stabilization of brome grass mosaic virus by magnesium and calcium. Virology 19, 367–374. Brakke, M.K., Daly, J.M., 1965. Density-gradient centrifugation: nonideal sedimentation and the interaction of major and minor components. Science 148, 387–389. Brown, D.J.F., Murant, A.F., Trudgill, D.L., 1989. Differences between isolates of the English serotype of tomato black ring virus in their transmissibility by an English polpuation of Longidorus attennatus (Nematods: Dorylaimoidea). Rev. Nematol. 12, 51–56. Brunt, A.A., Kenten, R.H., 1963. The use of protein in the extraction of cocoa swollen-shoot virus from cocoa leaves. Virology 19, 388–392. Brussaard, C.P.D., Moordeloos, A.A.M., Sandaa, R.-A., Heldal, M., Bratbak, G., 2004. Discovery of a dsRNA virus infection the marine photosynthetic protest Micromonas pusilla. Virology 319, 280–291. Burgermeister, W., Koenig, R., 1984. Electro-blot immunoassay—a means for studying serological relationships among plant viruses? Phytopathol. Z 111, 15–25. Caspar, D.L.D., Holmes, K.C., 1969. Structure of Dahlemense strain of tobacco mosaic virus: a periodically deformed helix. J. Mol. Biol. 46, 99–133. Cassells, A.C., Herrick, C.C., 1977. The identification of mild and severe strains of tobacco mosaic virus in double inoculated tomato plants. Ann. Appl. Biol. 86, 37–46.
Cavalier-Smith, T., 2004. Only six kingdoms. Proc. R. Soc. Lond. B 271, 1251–1262. Chang, C.-A., Hiebert, E., Purcifull, D.E., 1988. Purification, characterisation and immunological analysis of nuclear inclusions induced by bean yellow mosaic virus and clover yellow vein, potyviruses. Phytopathology 78, 1266–1275. Cheo, P.C., 1972. The replication of tobacco mosaic virus in species of fern. Plant Dis. Reptr. 56, 1010–1011. Cherrier, M.V., Kostyuchenko, V.A., Xiao, C.A., Bowman, V.D., Battisti, A.J., et al., 2009. An icosahedral algal virus has a complex unique vertex decorated by a spike. Proc. Natl. Acad. Sci. USA 106, 11085–11089. Chiba, S., Salaipeth, L., Lin, Y.-H., Sasaki, A., Kanematsu, S., et al., 2009. A novel bipartite double-stranded RNA mycovirus from the white root rot fungus Rosellinia necatrix: molecular and biological characterization, taxonomic considerations and potential for biological control. J. Virol. 83, 12801–12812. Chiko, A.W., 1975. Evidence of multiple virion components in leaf-dip preparations of barley stripe mosaic virus. Virology 63, 115–122. Choi, G.H., Shapira, R., Nuss, D.L., 1991. Cotranslational autoproteolysis involved in gene expression from a double-stranded RNA genetic element associated with hypovirulence of the chestnut blight fungus. Proc. Natl. Acad. Sci. USA 88, 1167–1171. Chu, P.W.G., Francki, R.I.B., 1982. Detection of lettuce necrotic yellows virus by an enzyme-linked immunosorbent assay in plant hosts and the insect vector. Ann. Appl. Biol. 100, 149–156. Clark, M.F., Barbara, D.J., 1987. A method for the quantitative analysis of ELISA data. J. Virol. Methods 15, 213–222. Cockerham, G., 1970. Genetical studies on resistance to potato viruses X and Y. Heredity 25, 309–348. Cooper, J.I., Mayo, M.A., 1972. Some properties of the particles of three tobravirus isolates. J. Gen. Virol. 16, 285–297. Cornelissen, B.J.C., Bol, J.F., 1984. Homology between the proteins encoded by tobacco mosaic virus and two tricornaviruses. Plant Mol. Biol. 3, 379–384. Creaser, E.H., Gibbs, A.J., Pares, R.D., 1987. The amino acid composition of the coat protein of a tobamovirus from an Australian Capsicum crop. Aust. Plant Pathol. 16, 85–87. Crill, P., Hagedorn, D.J., Hanson, E.W., 1971. An artificial system for differentiating strains of alfalfa mosaic virus. Plant Dis. Rep. 55, 127–130. Culley, A.I., Steward, G.F., 2007. New genera of RNA viruses in subtropical seawater, inferred from polymerase gene sequences. Appl. Environ. Microbiol. 73, 5937–5944. Culley, A.L., Lang, A.S., Suttle, C.A., 2006. Metagenomic analysis of coastal RNA virus communities. Science 312, 1795–1798. D’Arcy, C.J., Mayo, M.A., 1997. Proposals for changes in luteovirus taxonomy and nomenclature. Arch. Virol. 142, 1285–1287. Dahl, D., Knight, C.A., 1963. Some nitrous acid-induced mutants of tomato atypical mosaic virus. Virology 21, 580–586. Davis, P.H., Heywood, V.H., 1963. Principles of Angiosperm Taxonomy. Oliver & Boyd, Edinburgh. de Bortoli, M., Roggero, P., 1985. Electrophoretic desorption of intact virus from immunoadsorbent. Microbiologica 8, 113–121. de Miranda, J., Stevens, M., de Bruyne, E., Smith, H.G., Bird, C., et al., 1995. Sequence comparison and classification of beet luteovirus isolates. Arch. Virol. 140, 2183–2200. Dekker, E.L., Doré, I., Porta, C., Van Regenmortel, M.H.V., 1987. Conformational specificity of monoclonal antibodies used in the diagnosis of tomato mosaic virus. Arch. Virol. 94, 191–203.
Dekker, E.L., Pinner, M.S., Markham, P.G., Van Regenmortel, M.H.V., 1988. Characterisation of maize streak virus isolates from different plant species by polyclonal and monoclonal antibodies. J. Gen. Virol. 69, 983–990. Derelle, E., Ferraz, C., Escande, M.-L., Eychenié, S., Cooke, R., et al., 2008. Life-cycle and genome of OtV5, a large DNA virus of the pelagic marine unicellular green alga Ostreococcus tauri. PLoS One 3, e2250. Diaz-Ruiz, J.R., Kaper, J.M., 1983. Nucleotide sequence relationships among thirty peanut stunt virus isolates determined by competition hybridization. Arch. Virol. 75, 277–281. Dijkstra, J., de Jager, C.P., 1998. Practical Plant Pathology: Protocols and Exercises. Springer-Verlag, Berlin. Domingo, E., 1999. Quasispecies. In: Granoff, A., Webster, R.G. (Eds.), Encyclopedia of Virology Academic Press, San Diego, CA, pp. 1431–1436. Domingo, E., Martin, V., Perales, C., Grande-Pérez, A., García-Arriaza, J., et al., 2006. Viruses as quasispecies: biological implications. Curr. Topics Microbiol. Immunol. 299, 51–82. Druka, A., Hull, R., 1998. Variation in rice tungro viruses: further evidence for two rice tungro bacilliform virus strains and possibly several rice tungro spherical virus variants. J. Phytopathol. 146, 175–178. Dunigan, D.D., Fitzgerald, L.A., Van Etten, J.L., 2006. Phycodnaviruses: a peek at genetic diversity. Virus Res. 117, 119–132. Dunn, D.B., Hitchborn, J.H., 1965. The use of bentonite in the purification of plant viruses. Virology 25, 171–192. Easton, L.M., Lewis, G.D., Pearson, M.N., 1997. Virus-like particles associated with dyeback symptoms in the brown alga Ecklonia radiata. Dis. Aquatic Org. 30, 217–222. Ehlers, U., Paul, H.-L., 1986. Characterisation of the coat proteins of different types of barley yellow mosaic virus by polyacrylamide gel electrophoresis and electro-blot immunoassay. J. Phytopathol. 115, 294–304. Eigen, M., 1993. Viral quasispecies. Sci. Am., 32–39. Fauquet, C.M., 1999. Taxonomy, classification and nomenclature of viruses. In: Granoff, A., Webster, R.G. (Eds.), Encyclopedia of Virology, second ed. Academic Press, New York, NY, pp. 1730–1756. Fauquet, C.M., Martelli, G.P., 1995. Up-dated ICTV list of names and abbreviations of viruses, viroids and satellites infecting plants. Arch. Virol. 140, 393–413. Fauquet, C.M., Mayo, M.A., 1999. Abbreviations for plant virus names— 1999. Arch. Virol. 144, 1249–1272. Fauquet, C.M., Dejardin, J., Thouvenel, J.-C., 1986. Evidence that the amino acid composition of the particle proteins of plant viruses is characteristic of the virus group: I. Multidimensional classification of plant viruses. Intervirology 25, 1–13. Fauquet, C.M., Mayo, M.A., Maniloff, J., Desselberger, U., Ball, L.A., 2005. Virus Taxonomy: Eighth Report of the International Committee on Taxonomy of Viruses. Elsevier, Amsterdam, p. 1259. Fenner, F., 1976. Classification and nomenclature of viruses—Second Report on the International Committee on Taxonomy of Viruses. Intervirology 7, 1–115. Ford, R.E., 1973. Concentration and purification of clover yellow mosaic virus from pea roots and leaves. Phytopathology 63, 926–930. Foster, G.D., Taylor, S.C. (Eds.), 1998. Plant Virology Protocols: From Virus Isolation to Transgenic Resistance Humana Press, Totowa, NJ. Fraile, A., García-Arenal, F., 1990. A classification of the tobamoviruses based on the comparison of their “126 K” proteins. J. Gen. Virol 71, 2223–2228.
Chapter | 2 Plant Viruses and Their Classification
Fraenkel-Conrat, H., Singer, B., 1957. Virus reconstitution: II. Combination of protein and nucleic acid from different strains. Biochim. Biophys. Acta 24, 540–548. Francki, R.I.B., Fauquet, C.M., Knudson, D.I., Brown, F., 1991. Classification and Nomenclature of Viruses. Fifth Report on the International Committee on Taxonomy of Viruses. Springer-Verlag, New York, NY. Frenkel, M.J., Ward, C.W., Shukla, D.D., 1989. The use of 3′-noncoding nucleotide sequences in the taxonomy of potyviruses: application to watermelon mosaic virus 2 and soybean mosaic virus N. J. Gen. Virol. 70, 2775–2783. Friesen, N., Brandes, A., Heslop-Harrison, J.S., 2001. Diversity, origin and distribution of retrotransposons (gypsy and copia) in conifers. J. Mol. Evol. 18, 1176–1188. Frosheiser, F.I., 1969. Variable influence of alfalfa mosaic virus strains on growth and survival of alfalfa and on mechanical and aphid transmission. Phytopathology 59, 857–862. Fukumoto, F., Tochihara, H., 1984. Effect of various additives on the long-term preservation of tobacco ringspot and radish mosaic viruses. Ann. Phytopathol. Soc. Jpn. 50, 158–165. Fulton, J.P., 1969. Transmission of tobacco ringspot virus to the roots of a conifer by a nematode. Phytopathology 59, 236. Fulton, R.W., 1978. Superinfection by strains of tobacco streak virus. Virology 85, 1–8. Ge, L., Zhang, J., Zhou, X., Li, H., 2007. Genetic structure and population variability of tomato yellow leaf curl China virus. J. Virol. 81, 5902–5907. Geering, A.D.W., Scharaschkin, T., Teycheney, P.-Y., 2010. The classification and nomenclature of endogenous viruses in the family Caulimoviridae. Arch. Virol. 155, 123–131. Geysen, H.M., Meloen, R.H., Barteling, S.J., 1984. Use of peptide synthesis to probe viral antigens for epitopes to a resolution of a single amino acid. Proc. Natl. Acad. Sci. USA 81, 3998–4002. Ghabrial, S.A., Suzuki, N., 2009. Viruses of plant pathogenic fungi. Ann. Rev. Phytopathol. 47, 353–384. Gibbs, A., 1986. Tobamovirus classification In: Van Regenmortel, M.H.V. Fraenkel-Conrat, H. (Eds.), The Plant Viruses, vol. 2 Plenum Press, New York, NY, pp. 167–180. Gibbs, A., Ohshima, K., 2010. Potyviruses and the digital revolution. Annu. Rev. Phytopathol. 48, 205–223. Gibbs, A.J., 1969. Plant virus classification. Adv. Virus Res. 14, 263–328. Gibbs, A.J., Gibbs, M.J., 2006. A broader definition of “the viral species”. Arch. Virol. 151, 1419–1422. Gibbs, A.J., Harrison, B.D., Watson, D.H., Wildy, P., 1966. What’s in a virus name? Nature 209, 450–454. Gill, C.C., Chong, J., 1979. Cytological alterations in cells infected with corn leaf aphid-specific isolates of barley yellow dwarf virus. Phytopathology 69, 363–368. Gill, C.C., Chong, J., 1981. Vascular cell alterations and predisposed xylem infection in oats by inoculations with paired barley yellow dwarf viruses. Virology 114, 405–414. Givord, L., Xiong, C., Giband, M., Koenig, I., Hohn, T., et al., 1984. A second cauliflower mosaic virus gene product influences the structure of the viral inclusion body. EMBO J. 3, 1423–1427. Goldbach, R., Krijt, J., 1982. Cowpea mosaic virus-encoded protease does not recognise primary translation products of mRNAs from other comoviruses. J. Virol. 43, 1151–1154. Gonsalves, D., Fulton, R.W., 1977. Activation of Prunus necrotic ringspot virus and rose mosaic virus by RNA4 components of some ilarviruses. Virology 81, 398–407.
Gooding, G.V., 1970. Natural serological strains of tobacco ringspot virus. Phytopathology 60, 708–713. Gugerli, P., 1984. Isopycnic centrifugation of plant viruses in Nycodenz® density gradients. J. Virol. Methods 9, 249–258. Hajimorad, M.R., Francki, R.I.B., 1988. Alfalfa mosaic virus isolates from Lucerne in South Australia: biological variability and antigenic similarity. Ann. Appl. Biol. 113, 45–54. Halk, E.L., Hus, H.T., Aebig, J., Franke, J., 1984. Production of monoclonal antibodies against three ilarviruses and alfalfa mosaic virus and their use in serotyping. Phytopathology 74, 367–372. Han, S.S., Karasev, A.V., Ieki, H., Iwanami, T., 2002. Nucleotide sequence and taxonomy of Cycas necrotic stunt virus. Arch. Virol. 147, 2207–2214. Hanada, K., Kusunoki, M., Iwaki, M., 1986. Properties of virus particles, nucleic acid and coat protein of cycas necrotic stunt virus. Ann. Phytopathol. Soc. Jpn. 52, 422–427. Harper, G., Hart, D., Moult, S., Hull, R., Geering, A., et al., 2005. The diversity of Banana streak virus isolates in Uganda. Arch. Virol. 150, 2407–2420. Harrison, B.D., 1958. Ability of single aphids to transmit both avirulent and virulent strains of potato leaf roll virus. Virology 6, 278–286. Harrison, B.D., Robinson, D.J., 1999. Natural genomic and antigenic variation in whitefly-transmitted geminiviruses (Begomoviruses). Annu. Rev. Phytopathol. 37, 369–398. Harrison, B.D., Finch, J.T., Gibbs, A.J., Hollings, M., Shepherd, R.J., et al., 1971. Sixteen groups of plant viruses. Virology 45, 356–363. Harrison, B.D., Liu, Y.L., Khalid, S., Hameed, S., Otim-Nape, G.W., et al., 1997. Detection and relationships of cotton leaf curl virus and allied whitefly-transmitted geminiviruses occurring in Pakistan. Ann. Appl. Biol. 130, 61–75. Hatta, T., Francki, R.I.B., 1977. Morphology of Fiji disease virus. Virology 76, 797–807. Hatta, T., Francki, R.I.B., 1984. Differences in the morphology of isometric particles of some plant viruses stained with uranyl acetate as an aid to their identification. J. Virol. Methods 9, 237–247. Hearne, P.Q., Knorr, D.A., Hillman, B., Morris, T.J., 1990. The complete genome structure and synthesis of infectious RNA from clones of tomato bushy stunt virus. Virology 177, 141–151. Hebert, T.T., 1963. Precipitation of plant viruses by polyethylene glycol. Phytopathology 53, 362. Heijtink, R.A., Jaspars, E.M.J., 1974. RNA contents of abnormally long particles of certain strains of alfalfa mosaic virus. Virology 59, 371–382. Hoffmann, K., Sackey, S.T., Maiss, E., Adomako, D., Vetten, H.J., 1997. Immunocapture polymerase chain reaction for the detection and characterization of cacao swollen shoot virus 1A isolates. J. Phytopathol. 145, 205–212. Holmes, F.O., 1939. Handbook of Phytopathogenic Viruses. Burgess, Minneapolis, MI. Howitt, R.L.J., Beever, R.E., Pearson, M.N., Forster, R.L.S., 2006. Genome characterization of a flexuous rod-shaped mycovirus, Botrytis virus X, reveals high amino acid identity to genes from plant “potex-like” viruses. Arch. Virol. 151, 563–579. Hughes, J.d’A., Thomas, B.J., 1988. The use of protein A–sandwich ELISA as a means for quantifying serological relationships between members of the tobamovirus group. Ann. Appl. Biol. 112, 117–126. Hull, R., 1968. A virus disease of Hart’s tongue fern. Virology 35, 333–335. Hull, R., 1970. Studies on alfalfa mosaic virus: IV. An unusual strain. Virology 42, 283–292.
Hull, R., 1976. The behavior of salt-labile plant viruses in gradients of cesium sulphate. Virology 75, 18–25. Hull, R., 1977. Particle differences related to aphid transmissibility of a plant virus. J. Gen. Virol. 34, 183–187. Hull, R., 1985. Purification, biophysical and biochemical characterisation of viruses with especial reference to plant viruses. In: Mahy, B.W.J. (Ed.), Virology: A Practical Approach IRL Press, Oxford, pp. 1–14. Hull, R., 1996. Molecular biology of rice tungro viruses. Annu. Rev. Phytopathol. 34, 275–297. Hull, R., 1999. Classification of reverse transcribing elements: a discussion document. Arch. Virol. 144, 209–214. Hull, R., 2002. Matthews’ Plant Virology. Academic Press, San Diego, CA. Hull, R., Johnson, M.W., 1968. The precipitation of alfalfa mosaic virus by magnesium. Virology 34, 388–390. Hull, R., Hills, G.J., Plaskitt, A., 1970. The in vivo behaviour of twentyfour strains of alfalfa mosaic virus. Virology 42, 753–772. Hull, R., Shepherd, R.J., Harvey, J.D., 1976. Cauliflower mosaic virus: an improved purification procedure and some properties of the virus particles. J. Gen. Virol. 31, 93–100. Hull, R., Milne, R.G., Van Regenmortel, M.H.V., 1991. A list of proposed standard acronyms for plant viruses and viroids. Arch. Virol. 120, 151–164. Huss, B., Muller, S., Sommermeyer, G., Walter, B., Van Regenmortel, M.H.V., 1987. Grapevine fanleaf virus monoclonal antibodies: their use to distinguish different isolates. J. Phytopathol. 119, 358–370. Jacobi, V., Castello, J.D., Flachmann, M., 1992. Isolation of tomato mosaic virus from red spruce. Plant Dis. 76, 518–522. Jaegle, M., Van Regenmortel, M.H.V., 1985. Use of ELISA for measuring the extent of serological cross-reactivity between plant viruses. J. Virol. Methods 11, 189–198. Jeske, H., Gotthardt, D., Kober, S., 2010. In planta cloning of geminiviral DNA: the true Sida micrantha mosaic virus. J. Virol. Methods 163, 301–308. Johnson, J., 1927. The classification of plant viruses. Res. Bull. Wis. Agric. Exp. Stn. 76, 1–16. Johnson, J., Hoggan, I.A., 1935. A descriptive key for plant viruses. Phytopathology 25, 328–343. Jones, M.C., Gough, K., Dasgupta, I., Subba Rao, B.L., Cliffe, J., et al., 1991. Rice tungro disease is caused by a RNA and a DNA virus. J. Gen. Virol. 72, 757–761. Juckes, I.R.M., 1971. Fractionation of proteins and viruses with polyethylene glycol. Biochim. Biophys. Acta 229, 535–546. Kassanis, B., Phillips, M.P., 1970. Serological relationship of strains of tobacco necrosis virus and their ability to activate strains of satellite virus. J. Gen. Virol. 9, 119–126. Keeling, J., Collins, E.R., Matthews, R.E.F., 1979. Behaviour of turnip yellow mosaic virus nucleoproteins under alkaline conditions. Virology 97, 100–111. Kim, T., Youn, M.Y., Min, B.E., Choi, S.H., Kim, M., et al., 2005. Molecular analysis of quasispecies of Kyuri green mottle mosaic virus. Virus Res. 110, 161–167. King, A.M.Q., Adams, M.J., Carstens, E.B., Lefkowitz, E.J., 2012. Virus Taxonomy: Ninth Report of the International Committee on Taxonomy of Viruses. Elsevier, Amsterdam, p. 1327. Koenig, R., 1976. A loop-structure in the serological classification system of tymoviruses. Virology 72, 1–5. Koenig, R., 1981. Indirect ELISA methods for the broad specificity detection of plant viruses. J. Gen. Virol. 55, 53–62.
Koenig, R., Bercks, R., 1968. Änderungen in heterologen Reaktionsvermögen von Antiseren gegen Vertroter der potato virus X-Gruppe im Laufe des Immunisierungsprozesses. Phytopathol. Z 61, 382–398. Koenig, R., Gibbs, A., 1986. Serological relationships among tombus viruses. J. Gen. Virol. 67, 75–82. Koenig, R., Givord, L., 1974. Serological interrelationships in the turnip yellow mosaic virus group. Virology 58, 119–125. Konaté, G., Barro, N., Fargette, D., Swanson, M.M., Harrison, B.D., 1995. Occurrence of whitefly-transmitted geminiviruses in crops in Burkino Faso, and their serological detection and differentiation. Ann. Appl. Biol. 126, 121–129. Koonin, E.V., Choi, G.H., Nuss, D.L., Shapira, R., Carrington, J.C., 1991. Evidence for a common ancestry of a chestnut blight hypovirulenceassociated double-stranded RNA and a group of positive-stranded RNA plant viruses. Proc. Natl. Acad. Sci. USA 88, 10647–10651. Kuhn, J.H., Jahrling, P.B., 2010. Clarification and guidance on the proper usage of virus and virus species names. Arch. Virol. 155, 445–453. Kuhn, J.H., Mahy, B.W.J., Martelli, G.P., Pringle, C., Rybicki, E.P., et al., 2010. A proposal to change existing virus species names to non-Latinized binomials. Arch. Virol. 155, 1909–1919. Lang, A.S., Culley, A.I., Suttle, C.A., 2004. Genome sequence and characterization of a virus (HaRNAV) related to picorna-like viruses that infects the marine bloom-forming alga Heterosigma akashiwo. Virology 320, 206–217. Lang, A.S., Rise, M.L., Culley, A.L., Steward, G.F., 2009. RNA viruses in the sea. FEMS Microbiol. Rev. 33, 295–323. Le Gall, O., Christian, P., Fauquet, C.M., King, A.M.Q., Knowles, N.J., et al., 2008. Picornavirales, a proposed order of positive-sense single-stranded RNA viruses with a pseudo-T = 3 virion structure. Arch Virol. 153, 715–727. Lee, R.F., Garnsey, S.M., Briansky, R.H., Goheen, A.C., 1987. A purification procedure for enhancement of citrus tristeza virus yields and its application to other phloem limited viruses. Phytopathology 77, 543–549. Li, G., Jiang, D., Wang, D., Zhu, B., Rimmer, R., 1999. Double-stranded RNAs associated with the hypovirulence of Sclerotinia sclerotiorum strain Ep-1PN. Prog. Nat. Sci. 9, 836–841. Lin, H., Lan, X., Laio, H., Parsley, T.B., Nuss, D.L., et al., 2007. Genome sequence, full length infectious cDNA clone, and mapping of viral double-stranded RNA accumulation determinant of hypovirus CHV1-EP721. J. Virol. 81, 1813–1820. Liu, H., Fu, Y., Jiang, D., Li, G., Xie, J., et al., 2009. A novel mycovirus that is related to the human pathogen hepatitis E virus and rubi-like viruses. J. Virol. 83, 1981–1991. Long, D.G., Borsa, J., Sargent, M.D., 1976. A potential artifact generated by pelleting viral particles during preparative ultracentrifugation. Biochim. Biophys. Acta 451, 639–642. Loor, F., 1967. Comparative immunogenicities of tobacco mosaic virus protein subunits and reaggregated protein subunits. Virology 33, 215–220. Lot, H., Kaper, J.M., 1976. Physical and chemical differentiation of three strains of cucumber mosaic virus and peanut stunt virus. Virology 74, 209–222. Louie, R., Knoke, J.K., 1975. Strains of maize dwarf mosaic virus. Plant Dis. Rep. 59, 518–522. Lozano, G., Grande-Pérez, A., Navas-Castillo, J., 2009. Populations of genomic RNAs devoted to the replication or spread of a bipartite virus differ in genetic structure. J. Virol. 83, 12973–12983.
Chapter | 2 Plant Viruses and Their Classification
Manrubia, S.C., Domingo, E., Lázaro, E., 2010. Pathways to extinction: beyond the error threshold. Phil. Trans. R. Soc. B 365, 1943–1952. Marbrook, J., Matthews, R.E.F., 1966. The differential immunogenicity of plant virus proteins and nucleoproteins. Virology 28, 219–228. Martin, R.R., Zhou, J., Tzanetakis, J.E., 2011. Blueberry latent virus: an amalgam of the Partitiviridae and Totiviridae. Virus Res. 155, 175–180. Massalski, P.R., Harrison, B.D., 1987. Properties of monoclonal antibodies to potato leafroll luteovirus and their use to distinguish virus isolates differing in aphid transmissibility. J. Gen. Virol. 68, 1813–1821. Matthews, R.E.F., 1949. Studies on potato virus X. II. Criteria of relationships between strains. Ann. Appl. Biol. 36, 460–474. Matthews, R.E.F., 1974. Some properties of TYMV nucleoproteins isolated in cesium chloride density gradients. Virology 60, 54–64. Matthews, R.E.F., (Ed.), 1979. Classification and nomenclature of viruses. Third report of the international committee on taxonomy of viruses. Intervirology 12, 132–296. Matthews, R.E.F., 1981. Plant Virology, second ed. Academic Press, New York, NY. Matthews, R.E.F., (Ed.), 1982. Classification and nomenclature of viruses. Fourth report of the international committee on taxonomy of viruses. Intervirology 17, 1–199. Matthews, R.E.F., 1983. The history of viral taxonomy. In: Mathews, R.E.F. (Ed.), A Critical Appraisal of Viral Taxonomy CRC Press, Boca Raton, FL, pp. 1–35. Matthews, R.E.F., 1985a. Viral taxonomy. Microbiol. Sci. 2, 74–75. Matthews, R.E.F., 1985b. Viral taxonomy for the non-virologist. Annu. Rev. Microbiol. 39, 451–474. Matthews, R.E.F., 1991. Plant Virology, third ed. Academic Press, London. Maurin, J., Ackermann, H.W., Lebeurier, G., Lwoff, A., 1984. Un système des virus—1983. Ann. Inst. Pasteur Virol. 135E, 105–110. Mayo, M.A., D’Arcy, C.J., 1999. Family Luteoviridae: a reclassification of luteoviruses. In: Smith, H.G., Barker, H. (Eds.), The Luteoviridae CAB International, Wallingford, UK, pp. 15–22. Mayo, M.A., Barker, H., Harrison, B.D., 1982. Specificity and properties of the genome-linked proteins of nepoviruses. J. Gen. Virol. 59, 149–162. Mayo, M.A., Maniloff, J., Van Regenmortel, M.H.V., Fauquet, C.M., 2002. The type species in virus taxonomy. Arch. Virol. 147, 1271–1274. McLean, G.D., Francki, R.I.B., 1967. Purification of lettuce necrotic yellows virus by column chromatography on calcium phosphate gel. Virology 31, 585–591. McCormack, G.P., Clewley, J.P., 2002. The application of molecular phylogenetics to the analysis of viral genome diversity and evolution. Rev. Medic. Virol. 12, 221–238. McDonald, J.G., Hiebert, E., 1975. Characterization of the capsid and cylindrical inclusion proteins of three strains of turnip mosaic virus. Virology 63, 295–303. McKinney, H.H., 1929. Mosaic diseases in the Canary Islands, West Africa and Gibraltar. J. Agric. Res. 39, 557–578. McNaughton, P., Matthews, R.E.F., 1971. Sedimentation of small viruses at very low concentrations. Virology 45, 1–9. McRitchie, J.J., Alexander, L.J., 1963. Host-specific Lycopersicon strains of tobacco mosaic virus. Phytopathology 53, 394–398. Meints, R.H., Lee, K., van Etten, J.L., 1986. Assembly site of the virus PBCV-1 in a Chlorella-like green alga: ultrastructural studies. Virology 154, 240–245.
Metzker, M.L., 2010. Sequencing technologies—the next generation. Nat. Rev. Genet. 11, 31–46. Miller, W.A., Liu, S., Beckett, R., 2002. Barley yellow dwarf virus: Luteoviridae or Tombusviridae? Mol. Plant Pathol. 3, 177–183. Milne, R.G., 1988. Species concept should not be universally applied to virus taxonomy—but what to do instead? Intervirology 29, 254–259. Moleleki, N., Wingfield, M.J., Wingfield, B.D., Oreisig, O., 2011. Effect of diaporthe RNA virus 1 (DRV1) on growth and pathogenicity of different Diaporthe species. Eur J. Plant Pathol. 131, 261–268. Mosch, W.H.M., Huttings, H., Rast, A.T.B., 1973. Some chemical and physical properties of 18 tobacco mosaic virus isolates from tomato. Neth. J. Plant Pathol. 79, 104–111. Murphy, F.A., Fauquet, C.M., Bishop, 1995. Classification and Nomenclature of Viruses. Sixth Report of the International Committee on Taxonomy of Viruses. Springer-Verlag, New York, NY. Nameth, S.T., Steininger, J., 1997. Cucumber mosaic virus associated with leaf malformation and stunting in Adiantum pedatum. Plant Dis 81, 1333. Nienhaus, F., Mack, C., Schinzer, U., 1974. The isolation of viruses from fern plants. Z. Pflanzenkrankh. Pflanzensch 81, 533–537. Nishiguchi, M., Kikuchi, S., Kiho, Y., Ohno, T., Meshi, T., et al., 1985. Molecular basis of plant viral virulence: the complete nucleotide sequence of an attenuated strain of tobacco mosaic virus. Nucleic Acids Res. 13, 5585–5590. Noort, A., van den Drics, C.L.A.M., Pleij, C.W.A., Jaspars, E.M.J., Bosch, L., 1982. Properties of turnip yellow mosaic virus in cesium chloride solutions: the formation of high-density components. Virology 120, 412–421. Nozue, K., Kanegae, T., Wada, M., 1997. A full length Ty3/gypsy retrotransposon in the fern. Adiantum. J. Plant Res. 110, 495–499. Nuss, D.L., 2005. Hypovirulence: mycoviruses at the fungal–plant interface. Nat. Rev. Microbiol. 3, 632–642. Nuss, D.L., 2011. Mycoviruses, RNA silencing and viral RNA recombination. Adv. Virus Res. 80, 25–48. Ogbe, F.O., Atiri, G.I., Dixon, A.G.O., Thottappilly, G., 2003. Serological and biological variations of African cassava mosaic virus in Nigeria. Ann. Appl. Biol. 143, 203–213. Otsuki, Y., Takebe, I., 1976. Interaction of tobacco mosaic virus strains in doubly infected tobacco protoplasts. Ann. Microbiol. 127, 21. (abstract). Otsuki, Y., Takebe, I., Ohno, T., Fukuda, M., Okada, Y., 1977. Reconstitution of tobacco mosaic virus rods occurs bidirectionally from an internal initiation region: demonstration by electron microscopic serology. Proc. Natl. Acad. Sci. USA 74, 1913–1917. Owen, J., Palukaitis, P., 1988. Characterisation of cucumber mosaic virus. I. Molecular heterogeneity mapping of RNA3 in eight CMV strains. Virology 166, 495–502. Oxelfelt, P., 1976. Biological and physicochemical characteristics of three strains of red clover mottle virus. Virology 74, 73–80. Pares, R.D., 1988. Serological comparison of an Australian isolate of capsicum mosaic virus with capsicum tobamovirus isolates from Europe and America. Ann. Appl. Biol. 112, 609–612. Pead, M.T., Torrance, L., 1989. Some characteristics of monoclonal antibodies to a British MAV-like isolate of barley yellow dwarf virus. Ann. Appl. Biol. 113, 639–644. Pierpoint, W.S., 1966. The enzymic oxidation of chlorogenic acid and some reactions of the quinone produced. Biochem. J. 98, 567–580.
Pierpoint, W.S., Ireland, R.J., Carpenter, J.M., 1977. Modification of proteins during the oxidation of leaf phenols: reaction of potato virus X with chlorogenoquinone. Phytochemistry 16, 29–34. Preisig, O., Moleleki, N., Smit, W.A., Wingfield, B.D., Wingfield, M.J., 2000. A novel RNA mycovirus in a hypovirulent isolate of the plant pathogen Diaporthe ambigua. J. Gen. Virol. 81, 3107–3114. Pringle, C.R., 1998. The universal system of virus taxonomy of the International Committee on Taxonomy of Viruses (ICTV), including new proposals ratified since publication of the sixth report in 1995. Arch. Virol. 143, 203–210. Pringle, C.R., 1999. Editorial: virus nomenclature. Arch. Virol. 144, 1464–1466. Purcifull, D.E., Hiebert, E., McDonald, J.G., 1973. Immunochemical specificity of cytoplasmic inclusions induced by viruses in the potato Y group. Virology 55, 275–279. Qiu, L.Y., Li, Y.P., Liu, Y.M., Gao, Y.G., Qi, Y.C., et al., 2010. Particle and naked RNA mycoviruses in industrially cultivated mushroom Pleurotus ostreatus in China. Fungal Biol. 114, 507–513. Radford, A.D., Chapman, D., Dixon, L., Chantrey, J., Darby, A.C., et al., 2012. Application of next-generation sequencing technologies in virology. J. Gen. Virol. 93, 1853–1868. Randles, J.W., Harrison, B.D., Murant, A.F., Mayo, M.A., 1977. Packaging and biological activity of the two essential RNA species of tomato black ring virus. J. Gen. Virol. 36, 187–193. Rao, A.L.N., Francki, R.I.B., 1981. Comparative studies on tomato aspermy and cucumber mosaic viruses. VI. Partial compatibility of genome segments from the two viruses. Virology 114, 573–575. Rao, A.L.N., Hatta, T., Francki, R.I.B., 1982. Comparative studies on tomato aspermy and cucumber mosaic viruses. VII. Serological relationships reinvestigated. Virology 116, 318–326. Rastgou, M., Habibi, M.K., Izadpanah, K., Masenga, V., Milne, R.G., et al., 2009. Molecular characterization of the plant virus genus Ourmiavirus and evidence of inter-kingdom reassortment of viral genome segments as a possible route of origin. J. Gen. Virol 90, 2525–2535. Rees, M.W., Short, M.N., 1965. Variations in the composition of two strains of tobacco mosaic virus in relation to their host. Virology 26, 596–602. Reville, P.A., Davidson, A.D., Wright, P.J., 1994. The nucleotide sequence and genome organization of mushroom bacilliform virus— a single-stranded RNA virus of Agaricus bisporus (Lange) Imbach. Virology 202, 904–911. Rizzo, T.M., Palukaitis, P., 1988. Nucleotide sequence and evolutionary relationships of cucumber mosaic virus (CMV) strains: CMV RNA2. J. Gen. Virol. 69, 1777–1787. Rizzo, T.M., Palukaitis, P., 1989. Nucleotide sequence and evolutionary relationships of cucumber mosaic virus (CMV) strains: CMV RNA1. J. Gen. Virol. 70, 1–11. Rochow, W.F., 1979. Field variants of barley yellow dwarf virus: detection and fluctuation during twenty years. Phytopathology 69, 655–660. Romaine, C.P., Wach, M.P., Koons, K.C., Schlagnhaufer, B., 1986. Evidence suggests a viral etiology for LaFrance disease of the common mushroom. Phytopathology 76, 1121. Ronald, W.P., Tremaine, J.H., MacKenzie, D.J., 1986. Assessment of southern bean mosaic virus monoclonal antibodies for affinity chromatography. Phytopathology 76, 491–494. Roossinck, M.J., 2011. Changes in population dynamics in mutualistic versus pathogenic viruses. Viruses 3, 12–19.
Russo, M., Di Franco, A., Martelli, G.P., 1987. Cytopathology in the identification and classification of tombusviruses. Intervirology 28, 134–143. Saitou, N., Nei, M., 1986. Polymorphism and evolution of influenza virus genes. Mol. Biol. Evol. 3, 57–74. Salaman, R.N., 1933. Protective inoculation against a plant virus. Nature 131, 468. Salemi, M., Vandamme, A.-M., 2003. The Phylogenetic Handbook: A Practical Approach to DNA and Protein Phylogeny. Cambridge University Press, Cambridge, UK. Sänger, H.L., 1969. Functions of the two particles of tobacco rattle virus. J. Virol. 3, 304–312. Sardanyés, J., Elena, S.F., 2010. Error threshold in RNA quasispecies models with complementation. J. Theor. Biol. 265, 278–286. Sardanyés, J., Elena, S.F., Solé, R.V., 2008. Simple quasispecies models for the survival-of-the-flattest effect: the role of space. J. Theor. Biol. 250, 560–568. Schmelzer, K., Schmidt, H.E., Schmidt, H.B., 1966. Viruskrankheiten und Virusverdächtige Erscheinungen an Forstgehölzen. Arch. Forstwes 15, 107–120. Schmitt, I., Barker, F.K., 2009. Phylogenetic methods in natural product research. Nat. Prod. Rep. 26, 1585–1602. Schroeder, W.T., Provvidenti, R., 1971. A common gene for resistance to bean yellow mosaic virus and watermelon mosaic virus 2 in Pisum sativum. Phytopathology 61, 846–848. Segers, G.C., van Wezel, R., Zhang, X., Hong, Y., Nuss, D.L., 2006. Hypovirus papain-like protease p29 suppresses RNA silencing in the natural fungal host and in a heterologous plant system. Eukaryot. Cell 5, 896–904. Shukla, D.D., Ward, C.W., 1988. Amino acid sequence homology of coat proteins as a basis for identification and classification of the potyvirus group. J. Gen. Virol. 69, 2703–2710. Shukla, D.D., Ward, C.W., 1989a. Structure of potyvirus and coat proteins and its application in the taxonomy of the potyvirus group. Adv. Virus Res. 36, 273–314. Shukla, D.D., Ward, C.W., 1989b. Identification and classification of potyviruses on the basis of coat protein sequence data and serology. Arch. Virol. 106, 171–200. Shukla, D.D., Strike, P.M., Tracy, S.L., Gough, K.H., Ward, C.W., 1988. The N and C termini of the coat proteins of potyviruses are surface located and the N terminus contains the major virus-specific epitopes. J. Gen. Virol. 69, 1497–1508. Shukla, D.D., Tribbick, G., Mason, T.-J., Hewish, D.R., Geysen, H.M., et al., 1989a. Localization of virus-specific and group-specific epitopes of plant potyviruses by systematic immunochemical analysis of overlapping peptide fragments. Proc. Natl. Acad. Sci. USA 86, 8192–8196. Shukla, D.D., Ford, R.E., Tosic, M., Jilka, J., Ward, C.W., 1989b. Possible members of the potyvirus group transmitted by mites or whiteflies share epitopes with aphid-transmitted definitive members of the group. Arch. Virol. 105, 143–151. Shukla, D.D., Jilka, J., Tosic, M., Ford, R.E., 1989c. A novel approach to the serology of potyviruses involving affinity-purified polygonal antibodies directed towards virus specific N-termini of coat proteins. J. Gen. Virol. 70, 13–23. Shukla, D.D., Tosic, M., Jilka, J., Ford, R.E., Toler, R.W., et al., 1989d. Taxonomy of potyviruses infecting maize, sorghum and sugarcane in Australia and the United States as determined by reactivities of polyclonal antibodies directed towards virus-specific N-termini of coat proteins. Phytopathology 79, 223–229.
Chapter | 2 Plant Viruses and Their Classification
Singh, S.K., Anjaneyula, A., Lapierre, H., 1984. Use of pectino-cellulolytic enzymes for improving extraction of phloem-limited plant viruses as exemplified by the rice tungro virus complex. Agronomie 4, 479–484. Skotnicki, A., Gibbs, A., Wrigley, N.G., 1976. Further studies on Chara corallina virus. Virology 75, 457–468. Smith, D.B., McAllister, J., Casino, C., Simmonds, P., 1997. Virus “quasispecies”: making a mountain out of a molehill? J. Gen. Virol. 78, 1511–1519. Smith, K.M., 1937. A Text Book of Plant Virus Diseases. Churchill, London. Smith, T.A., 1987. The isolation of the two electrophoretic forms of cowpea mosaic virus using fast protein liquid chromatography. J. Virol. Methods 16, 263–269. Sneath, P.H.A., 1962. The construction of taxonomic groups. Symp. Soc. Gen. Microbiol. 12, 289–332. Stace-Smith, R., Martin, R.R., 1993. Virus purification in relation to virus diagnosis. In: Matthews, R.E.F. (Ed.), Diagnosis of Plant Viruses CRC Press, Boca Raton, FL, pp. 129–158. Staginnus, C., Iskra-Caruana, M.L., Lockhart, B., Hohn, T., RichertPöggeler, K.R., 2009. Suggestions for a nomenclature of endogenous pararetroviral sequences in plants. Arch. Virol. 154, 1189–1193. Stratford, R., Plastkitt, K.A., Turner, D.S., Markham, P.G., Covey, S.N., 1988. Molecular properties of Bari 1, a mild strain of cauliflower mosaic virus. J. Gen. Virol. 69, 2375–2386. Studholme, D.J., Glover, R.H., Boonham, N., 2011. Application of high-throughput DNA sequencing in phytopathology. Annu. Rev. Phytopathol. 49, 87–105. Sun, L., Nuss, D.L., Suzuki, N., 2006. Synergism between a mycoreovirus and a hypovirus mediated by the papain-like protease p29 of the proteotype hypovirus CHV1-EP713. J. Gen. Virol. 87, 3703–3714. Swaans, H., van Kammen, A., 1973. Reconsideration of the distinction between the severe and yellow strains of cowpea mosaic virus. Neth. J. Plant Pathol. 79, 257–265. Swanson, M.M., Brown, J.K., Poulos, B.T., Harrison, B.D., 1992. Genome affinities and epitope profiles of whitefly-transmitted gemini viruses from the Americas. Ann. Appl. Biol. 121, 285–296. Tai, V., Lawrence, J.E., Lang, A.S., Chan, A.M., Culley, A.L., et al., 2003. Characterization of HaRNAV, a single-stranded RNA virus causing lysis of Heterosigma akashiwo (Raphidophyaceae). J. Phycol. 39, 343–352. Taliansky, M.E., Atabekova, T.I., Atabekov, J.G., 1977. The formation of phenotypically mixed particles upon mixed assembly of some tobacco mosaic virus (TMV) strains. Virology 76, 701–708. Tamada, T., Harrison, B.D., Roberts, I.M., 1984. Variation among British isolates of potato leafroll virus. Ann. Appl. Biol. 104, 107–116. Tavantzis, S.M., Romaine, C.P., Smith, S.H., 1980. Purification and partial characterization of a bacilliform virus from Agaricus bisporus: a single stranded RNA mycovirus. Virology 105, 94–102. Thiel, G., Moroni, A., Dunigan, D., Van Etten, J.L., 2010. Initial events associated with virus PBCV-1 infection of Chlorella NC64A. Prog. Bot. 71, 170–183. Thompson, S., Fraser, R.S.S., Barnden, K.L., 1988. A beneficial effect of trypsin on the purification of turnip mosaic virus (TuMV) and other potyviruses. J. Virol. Methods 20, 57–64. Thongmeearkom, P., Ford, R.E., Jedlinski, H., 1976. Aphid transmission of maize dwarf mosaic virus strains. Phytopathology 66, 332–335.
Thornbury, D.W., Pirone, T.P., 1983. Helper components of two poty viruses are serologically distinct. Virology 125, 487–490. Toriyama, S., Peters, D., 1981. Differentiation between broccoli necrotic yellows virus and lettuce necrotic yellows virus by their transcriptase activities. J. Gen. Virol. 56, 59–66. Torrance, L., Larkins, A.P., Butcher, G.W., 1986. Characterisation of monoclonal antibodies against potato virus X and comparison of serotypes with resistance groups. J. Gen. Virol. 67, 57–67. Tremaine, J.H., Ronald, W.P., Valcic, A., 1976. Aggregation properties of carnation ringspot virus. Phytopathology 66, 34–39. Urayama, S., Kato, S., Suzuki, Y., Aoki, N., Le, M.T., et al., 2010. Mycoviruses related to chrysovirus affect vegetative growth in the rice blast fungus Magnaporthe oryzae. J. Gen. Virol. 91, 3085–3094. Uyemoto, J.K., Gilmer, R.M., 1972. Properties of tobacco necrosis virus strains isolated from apple. Phytopathology 62, 478–481. Valiceasa, D., Enunlu, N., Kos, P.B., Koster, S., Beuther, E., et al., 2006. Searching for a putative cryptic virus in Pinus sylvestris. Virus Genes 32, 177–186. Valverde, R.A., Sabanadzovic, S., 2009. A novel plant virus with unique properties infecting Japanese holly fern. J. Gen. Virol. 90, 2542–2549. van Dun, C.M.P., Overduin, B., van Vloten-Doting, L., Bol, J.F., 1988. Transgenic tobacco expressing tobacco streak virus or mutated alfalfa mosaic virus coat protein does not cross-protect against alfalfa mosaic virus infection. Virology 164, 383–389. Van Etten, J.L., 2003. Unusual life style of giant chlorella viruses. An. Rev. Genet. 37, 153–195. Van Etten, J.L., Gurnon, J.R., Yanai-Balser, G.M., Dunigan, D.D., Graves, M.V., 2010a. Chlorella viruses encode most, if not all, the machinery to glycosylate their glycoproteins independent of the endoplasmic reticulum and Golgi. Biochim. Biophys. Acta 1800, 152–159. Van Etten, J.L., Lane, L.C., Dunigan, D.D., 2010b. DNA viruses: the really big ones (Giruses). Annu. Rev. Microbiol. 64, 83–99. van Kammen, A., 1968. The relationship between the components of cowpea mosaic virus: I. Two ribonucleoprotein particles necessary for the infectivity of CPMV. Virology 34, 312–318. Van Regenmortel, M.H.V., 1975. Antigenic relationships between strains of tobacco mosaic virus. Virology 64, 415–420. Van Regenmortel, M.H.V., 1990. Virus species, a much overlooked but essential concept in virus classification. Intervirology 31, 241–254. Van Regenmortel, M.H.V., 1999. How to write the names of virus species. Arch. Virol. 144, 1041–1042. Van Regenmortel, M.H.V., 2006. Virologists, taxonomy and the demands of logic. Arch. Virol. 151, 1251–1255. Van Regenmortel, M.H.V., von Wechmar, M.B., 1970. A reexamination of the serological relationship between tobacco mosaic virus and cucumber virus 4. Virology 41, 330–338. Van Regenmortel, M.H.V., Burke, D.S., Calisher, C.H., Dietzgen, R.G., Fauquet, C.M., et al., 1990. Origin and evolution of retroelements based upon their reverse transcriptase sequences. EMBO J. 9, 3353–3362. Van Regenmortel, M.H.V., Bishop, D.H.L., Fauquet, C.M., Mayo, M.A., Maniloff, J., et al., 1997. Guidelines to the demarcation of virus species. Arch Virol. 142, 1505–1518. Van Regenmortel, M.H.V., Fauquet, C.M., Bishop, D.H.L., Carstens, E.B., Estes, M.K., et al., 2000. Virus Taxonomy: the seventh report of the International Committee on Taxonomy of Viruses. Academic Press, San Diego, CA.
Van Regenmortel, M.H.V., Burke, D.S., Calisher, C.H., Dietzgen, R.G., Fauquet, C.M., et al., 2010. A proposal to change existing virus species names to non-Latinized binomials. Arch. Virol. 155, 1909–1919. Van Regenmortel, M.H.V., Ackermann, H.-W., Calisher, C.H., Dietzgen, R.G., Horzinek, M.C., et al., 2013. Virus species polemics: 14 senior virologists oppose a proposed change to the ICTV definition of virus species. Arch. Virol. 158, 1115–1119. van Tol, R.G.L., van Vloten-Doting, L., 1981. Lack of serological relationship between the 35 K non-structural protein of alfalfa mosaic virus and the corresponding proteins of three other plant viruses with tripartite genomes. Virology 109, 444–447. van Vloten-Doting, L., 1975. Coat protein is required for infectivity of tobacco streak virus: biological equivalence of the coat proteins of tobacco streak and alfalfa mosaic viruses. Virology 65, 215–225. van Vloten-Doting, L., Kruseman, J., Jaspars, E.M.J., 1968. The biological function and mutual dependence of bottom component and top component of alfalfa mosaic virus. Virology 34, 728–737. Veldee, S., Fraenkel-Conrat, H., 1962. The characterisation of tobacco mosaic virus strains by their productivity. Virology 18, 56–63. Vetten, H.J., Haenni, A.-L., 2010. Taxon-specific suffixes for vernacular names. Arch. Virol. 151, 1249–1250. Wagner, W.H., 1984. A comparison of taxonomic methods in biosystematics. In: Grant, W.F. (Ed.), Plant Biosystematics Academic Press, Orlando, FL, pp. 643–654. Walkey, D.G.A., 1991. Applied Plant Virology, second ed. Chapman & Hall, London. Watanabe, Y., Morita, N., Nishiguchi, M., Okada, Y., 1987b. Attenuated strains of tobacco mosaic virus. Reduced synthesis of a viral protein with a cell-to-cell movement function. J. Mol. Biol. 194, 699–704. Waterhouse, P.M., Gildow, F.E., Johnstone, G.R., 1988. Luteovirus group. Commonwealth Mycological Institute/Association of Applied Biologists, Descriptions of Plant Viruses, No. 339. Weidemann, H.L., Maiss, E., 1996. Detection of the potato tuber necrotic ringspot strain of potato virus Y (PVYNTN) by reverse transcription and immunocapture polymerase chain reaction. J. Plant Dis. Protect. 103, 337–345. Wilcoxson, R.D., Johnson, L.E.B., Frosheiser, F.I., 1975. Variation in the aggregation forms of alfalfa mosaic virus strains in different alfalfa organs. Phytopathology 65, 1249–1254. Wildy, P., 1971. Classification and nomenclature of viruses. First report of the international committee on taxonomy of viruses. Monogr. Virol. 5, 1–65.
Wiley, E.O., Lieberman, B.S., 2011. Phylogenetics: Theory and Practice of Phylogenetic Systematics, second ed. Wiley-Blackwell, Hoboken, NJ. Wilson, W.H., Van Etten, J.L., Allen, M.J., 2009. The Phycodnaviridae: a story of how tiny giants rule the world. Curr. Topics Microbiol. Immunol. 328, 1–42. Wu, J.X., Shang, H.L., Xie, Y., Shen, Q.T., Zhou, X.P., 2012. Monoclonal antibodies against whitefly-transmitted tomato yellow leaf curl virus and their application in virus diagnosis. J. Integrat. Agric. 11, 263–268. Wu, M.D., Zhang, L., Li, G.Q., Jiang, D.H., Hou, M.S., et al., 2007. Hypovirulence and double-stranded RNA in Botrytis cinerea. Phytopathology 97, 1590–1599. Xie, J., Wei, D., Jiang, D., Fu, Y., Li, G., et al., 2006. Characterization of debilitation-associated mycovirus infecting the plant-pathogenic fungus Sclerotinia sclerotiorum. J. Gen. Virol. 87, 241–249. Xiong, Y., Eichbush, T.H., 1990. Origins and evolution of retroelements based on their reverse transcriptase sequences. EMBO J. 9, 3353–3362. Yamada, T., Onimatsu, H., Van Etten, J.L., 2006. Chlorella viruses. Adv. Virus Res. 66, 293–336. Yeh, S.-D., Gonsalves, D., 1984. Purification and immunological analyses of cylindrical-inclusion protein induced by papaya ringspot virus and watermelon mosaic virus 1. Phytopathology 74, 1273–1278. Yua, X., Li, B., Fu, Y., Jiang, D., Ghabrial, S.A., et al., 2010. A geminivirus-related DNA mycovirus that confers hypovirulence to a plant pathogenic fungus. Proc. Natl. Acad. Sci. USA 107, 8387–8392. Zaitlin, M., Ferris, W.R., 1964. Unusual aggregation of a nonfunctional tobacco mosaic virus protein. Science 143, 1451–1452. Zaitlin, M., Beachy, R.N., Bruening, G., 1977. Lack of molecular hybridization between RNAs of two strains of TMV. A reconsideration of the criteria for strain relationships. Virology 82, 237–241. Zanotti, F.M. de A., Gibbs, M.J., Gould, E.A., Holmes, E.C., 1996. A reevaluation of the higher taxonomy of viruses based on RNA polymerases. J. Virol. 70, 6083–6096. Zhang, X., Nuss, D.L., 2008. A host dicer is required for defective viral RNA production and recombinant virus vector RNA instability for a positive sense RNA virus. Proc. Natl. Acad. Sci. USA 105, 16749–16754.