CLOSTEROVIRUSES (CLOSTEROVIRIDAE)

History

The name closterovirus was first coined in the early 1970s for a taxonomic group of plant viruses characterized by an elongated, very flexuous particle morphology and an open particle structure. The name is derived from the Greek ‘kloster’ meaning thread or filament. The group was approved by the International Committee on Taxonomy of Viruses in 1976.

Very early, it was observed that closteroviruses showed some variations in a number of their properties, such as genome size, cytopathology, tissue tropism and relation with insect vectors. This initially led to several attempts to divide the group. In the early 1990s, the availability of genomic sequences led to the splitting of the group to separate the trichoviruses from the true closteroviruses. Due to their localization in the phloem and the difficulties in particle purification, closteroviruses have traditionally been a genus which, with few exceptions, has attracted relatively little attention. However, recent years have seen a remarkable increase in the number of publications on closteroviruses. This can probably be explained by two factors: the significant improvement brought to these studies by the introduction of molecular biology techniques, and the emergence of a number of ‘new’ closteroviruses, apparently linked to the explosion of insect vector populations. As a result, characterization and sequence data are now accumulating at a rapid rate. This has already led to a further splitting of the Closterovirus genus, in order to accommodate in a separate taxon the species with a divided genome (Crinivirus). The two genera (Closterovirus and Crinivirus) are now included in the recently created Closteroviridae family.

Regardless of whether monopartite or bipartite, closterovirus genomes have a similar structure and contain unique genes coding for a homologue of the cellular heat shock protein (HSP) 70 proteins and a divergent copy of the coat protein. These genes are present in all species sequenced to date. Closteroviruses are phloem-associated and induce the formation of characteristic cytopathic structures in infected cells. Transmission is semipersistent by a variety of vectors, i.e. aphids, whiteflies, and pseudococcid or coccid mealybugs. Some can be transmitted (with difficulty) by mechanical inoculation. These properties are common to all species and should be regarded as characterizing traits of the family.

Taxonomy and Classification

The major problem in dealing with the taxonomy of members of the Closteroviridae is that, for many of these viruses, only limited information is available. This is mostly due to the fact that particles are difficult to purify because of low virus titer in the plants (linked with their phloem localization). The particles are also labile and they tend to break or aggregate during purification.

Closterovirus taxonomy has rapidly evolved during the last few years as nucleotide sequence data began to accumulate. Major differences in genome size and organization, biological properties (type of vector, host range) and the results of phylogenetic analysis, suggested that the family Closteroviridae be divided into at least two genera (Table 1 ):

• •Genus Closterovirus, typified by beet yellows virus (BYV), comprised of species with monopartite genomes, transmitted by aphids, mealybugs or whiteflies.
• •Genus rinivirus, typified by lettuce infectious yellows virus (LIYV), comprised of whitefly- transmitted species with bipartite genomes.

Table 1.

Tentative grouping, particle length and coat protein molecular weight of members of the Closteroviridae family

Virus	Particle length(nm)	Mol.wt(× 103)
Genus Closterovirus
A. Definitive species
Aphid-transmitted
Beet yellows virus (BYV)	1250–1450	22
Beet yellow stunt virus (BYSV)	1250–1400	24
Carnation necrotic fleck virus (CNFV)	1400–1500	23.5
Wheat yellow leaf virus (WYLV)	1600–1850	?
Burdock yellows virus (BuYV)	1600–1750	?
Citrus tristeza virus (CTV)	1900–2000	22
Carrot yellow leaf virus (CYLV)	1600	?
Mealybug-transmitted
Grapevine leafroll-associated virus 3 (GLRaV-3)	1800–2100	43
Little cherry virus (LChV)		46
Vector unknown
Grapevine leafroll-associated virus 2 (GLRaV-2)	1800	24
B. Tentative species
Aphid-transmitted
Clover yellows virus (CYV)	1700–1800	?
Dendrobium vein necrosis virus (DVNV)	1850	?
Heracleum virus 6 (HV6)	1400–1660	?
Mealybug-transmitted
Pineapple mealybug wilt-associated virus 1 (PMWaV-1)	?	?
Pineapple mealybug wilt-associated virus 2 (PMWaV-2)	?	?
Sugarcane mild mosaic virus (SMMV)	1500–1600	?
Whitefly-transmitted
Beet pseudo-yellows virus (BPYD)	1500–1800	?
Diodea vein chlorosis virus (DVCV)	?	?
Cucurbit chlorotic spot virus (CCSV)	1250–1450	?
Vector unknowm
Grapevine leafroll associated virus 1 (GLRaV-1)	2200	39
Grapevine leafroll associated virus 4 (GLRaV-4)	1800	36
Grapevine leafroll associated virus 5 (GLRaV-5)	1400–1700	35
Grapevine leafroll associated virus 6 (GLRaV-6)	1800	36
Grapevine leafroll associated virus 7 (GLRaV-7)	1500–1700	37
Festuca necrosis virus (FNV)	1700	?
Alligator weed stunting virus (AWSV)	1700	?
Megakepasma mosaic virus (MeMV)	?	?
Genus Crinivirus
A. Definitive species
Abutilon yellows virus (AYV)	800–850	?
Cucurbit yellow stunting disorder virus (CYSDV)	825–900	?
Lettuce chlorosis virus (LCV)	800–850	33
Lettuce infectious yellows virus (LIYV)	650–700	28
Sweet potato sunken vein virus (SPSVV)	850	29
Tomato chlorosis virus (ToCV)	800–850	32
Tomato infectious chlorosis virus (TICV)	850–900	32

Particle Structure and Composition

Although resembling that of some other flexuous filamentous plant viruses, such as tricho-, capillo- and vitiviruses, the very flexuous and open particle structure is still the most common and conspicuous trait of members of the Closteroviridae family (Fig. 1 ). In fact, in preliminary studies, this is often the only indication that a given virus might be a member of this family. Particles have, in general, a consistent diameter of about 12 nm, with a length varying between 650–900 and 2000 nm, depending on the virus. The RNA content is usually 5–6%. For some viruses ( BYV, citrus tristeza virus (CTV), LIYV, beet yellows stunt virus (BYSV), little cherry virus (LChV)), two types of coat protein (CP) subunits have been identified: the genuine CP and a CP homologue that probably arose by gene duplication. With BYV and CTV, the duplicate protein coats an extremity of the virions, forming a distinct structure, for which the terms ‘rattlesnake’ or ‘heterodimeric’ have been coined. Given that the presence of a CP duplicate is found in all closteroviruses so far sequenced, such an unusual particle structure may well be a general characteristic of the family.

Figure 1.

Particles of grapevine leafroll associated virus 2 (GLRaV-2) observed by electron microscopy. Original magnification approximately ×250000.

In addition to the end-to-end aggregation and breakage problems encountered during purification, it should be noted that closterovirus particles may assume slightly different dimensions depending on the negative stain used for electron microscopy, further complicating the determination of their size. In the case of CTV, smaller than full-length particles have been detected, which could encapsidate subgenomic or defective interfering RNAs.

The pitch of the primary helix, which can usually be measured from the obvious crossbanding of the particle, is in the 3–3.8 nm range, giving estimated ratios of four nucleotides per protein subunit and of about ten protein monomers per turn of the helix. The CPs of most members of the family fall in a restricted size range, usually 22–30 kDa. Notable exceptions are the CPs of mealybug-transmitted viruses, such as LChV or the grapevine leafroll associated viruses (GLRaVs), that have molecular masses of up to 46 kDa. An unusual property of the CP of some closteroviruses, such as BYV and carnation necrotic fleck virus (CNFV), is that they are devoid of tryptophan, which explains their high (1.4–1.8) A ₂₆₀:A ₂₈₀ absorbance ratios. However, the CP of CTV does contain a significant amount of tryptophan (A ₂₆₀:A ₂₈₀ of 1.21–1.22). The particles of several virus species such as CTV are not stable in CsCl solutions but are stable in Cs₂SO₄, with a buoyant density of 1.24–1.27 g ml⁻¹.

Genome Structure

Members of the Closteroviridae family have either monopartite or bipartite single-stranded RNA genomes of positive polarity. The genomic RNAs of BYV, CTV and LIYV are infectious, and have messenger activity in vitro. The genomic RNAs of BYV, CTV, LChV and LIYV do not appear to include a poly (A) tract, but terminate with a heteropolymeric sequence devoid of conspicuous features other than the frequent presence of potential hairpins near the 3′ end. Indirect evidence suggests that the 5′ end of the genome of family members is capped.

Genome Organization and Expression: Affinities with Other Virus Taxa

To date, complete genomic sequences have been determined for only five viruses in the family: BYV, LIYV, LChV and two isolates of CTV (CTV-T36, a decline isolate from Florida, and CTV-VT, a yellows isolate from Israel), while the 3′-terminal sequence is available for beet yellows stunt virus (BYSV), grapevine leafroll-associated virus 2 (GLRaV-2), grapevine leafroll-associated virus 3 (GLRaV-3), and tomato infectious chlorosis virus (TICV). Relevant accession numbers are: (X73476: BYV), (RNA1 U15440, RNA2 U15441: LIYV), (Y10237: LChV), (U16304: CTV-T36), (U56902: CTV-VT), (U51931: BYSV), (Y14131: GLRaV-2).

The 15500 nt genome of BYV shows an arrangement of nine open reading frames (ORFs) (Fig. 2 ). The largest 5′-proximal ORF, 1a, encodes a large polyprotein with calculated molecular mass of 295 kDa, containing domains characteristic of papain-like proteinase (P-PRO), methyltransferase (MT) and helicase (HEL). The downstream ORF 1b encodes a 48 kDa putative RNA-dependent RNA polymerase (POL), which is probably expressed via a +1 ribosomal frameshifting, as a 348 kDa ORF 1a/1b fusion protein. ORF 2 encodes a 6.4 kDa protein with membrane-binding properties, homologous to the small hydrophobic proteins of potex- and carlaviruses. ORF 3 codes for a 65 kDa homologue of the cellular HSP 70 heat shock proteins. The proteins of this family are molecular chaperones found in all cell types, and are implicated in protein–protein interactions. The 65 kDa protein of BYV has ATPase activity associated with the conserved N-terminal domain. However, it apparently lacks the ability to interact with unfolded protein chains, another property typical of cellular HSP 70 proteins. Recent experimental work indicates that this protein is essential for cell-to-cell movement of the virus. ORF 4 overlaps ORF 3 and encodes a 64 kDa protein with sequence similarity to proteins of the HSP 90 family of heat shock proteins. ORFs 6 and 5 code for the capsid protein (22 kDa) and its diverged duplicate (24 kDa), respectively. ORFs 7 and 8 code for proteins that do not have sequence similarity to any proteins in the database, with the exception of the homologous proteins of CTV. ORF 2, 3, 4, 5 and 6, coding for the small hydrophobic protein, the HSP 70 homologue, the HSP 90 homologue, the diverged copy of the CP and the CP itself, constitute a five-gene module present in all species sequenced so far.

Figure 2.

Genomic organization of some members of the Closteroviridae family. BYV, beet yellows virus; BYSV, beet yellow stunt virus; CTV, citrus tristeza virus; LChV, little cherry virus; LIYV, lettuce infectious yellows virus; pro, domain characteristic of papain-like proteinases; MT, methyltransferase domain; HEL, helicase domain; POL, RNA-dependent RNA polymerase domain; HSP 70, protein homologue of the cellular HSP 70 heat shock protein; CP, coat protein; CPd, diverged duplicate of coat protein; RBP, RNA-binding protein.

The 19296 nt genome of CTV-T36 (the largest known plant virus genome) contains 12 ORFs potentially encoding at least 17 protein products. The general gene arrangement is similar to that of BYV, with two gene blocks or modules easily identified. The first block is the replication-associated complex (including the MT, HEL and POL domains, with the POL expressed via a +1 ribosomal frameshifting) preceded by a similar papain-like protease (apparently duplicated in the case of CTV). The second block is the five-gene module. Outside of these modules, other genes in CTV and BYV differ. In addition to a supplementary copy of the leader protease, CTV has four unique genes that do not have counterparts in the BYV genome, including a 3′-terminal ORF encoding a 23 kDa protein which probably has RNA-binding activity.

The genome of LIYV consists of two RNA components. The 8118 nt RNA1 contains three ORFs that include the domains for a papain-like protease, methyltransferase, helicase and RNA polymerase (expressed via a +1 ribosomal frameshift). The 7193 nt RNA2 includes six ORFs, which include the five-gene module. However, LIYV is distinct from monopartite closteroviruses in several ways: its genome consists of two RNAs; the order of the capsid protein gene and the gene for its diverged copy is reversed as compared to BYV and CTV (the same is also observed in the mealybug-transmitted LChV, GLRaV-3 and in the whitefly-transmitted sweet potato sunken vein virus (SPSVV)). The diverged copy (52 kDa) is also much larger than the CP (28 kDa) and could be the result of a triplication of the CP gene, with the CP gene followed by two fused, highly diverged duplicates. In addition, LIYV includes ORFs that are not related to ORFs found in the other viruses.

Members of the Closteroviridae share common and original expression mechanisms. In all members of the family for which sufficient information is available, the POL domain is encoded by a separate ORF overlapping with the 3′-terminal portion of the large upstream ORF, including the MT and HEL domains. It has been suggested that both ORFs are expressed as a large fusion protein resulting from a +1 ribosomal frameshifting. Such a +1 ribosomal frameshifting is a unique feature among positive-strand plant viruses. In vitro translation experiments have demonstrated the autocatalytic cleavage of the N-terminal portion of the ORF 1a product of BYV that contains the papain-like proteinase domain. Furthermore, site-directed mutagenesis of the predicted catalytic amino acid residues abolished the proteolysis, thus corroborating the identity of the protease. With BYV, CTV, BYSV and LIYV, the products of the ORFs downstream, ORFs 1a and 1b, are expressed via the formation of a nested set of 3′-coterminal subgenomic RNAs (sgRNA). Therefore, the genome expression of these viruses (and likely of all members of the family) is based on polyprotein processing, translational frameshifting and multiple sgRNA generation, thus resembling that of coronaviruses. However, unlike coronaviruses, the closterovirus polymerase belongs to the Sindbis virus-like lineage of polymerases.

Another common feature of members of the family is the presence of the gene coding for the HSP 70-related protein. This gene, first reported for BYV, has now been identified in 12 additional closteroviruses (CTV, LChV, BPYV, BYSV, CCSV, CNFV, GLRaV-1, GLRaV-2, GLRaV-3, GLRaV-7, PMWaV-1, PMWaV-2) and five criniviruses (CYSDV, TICV, LIYV, LCV, ToCV). In the cell, the molecular chaperones of the HSP 70 family perform several functions, including protein folding, protein trafficking and secretion. Like some HSP 70 proteins, the 65 kDa protein of BYV binds to microtubules in vitro, this binding being stimulated by ATP hydrolysis. Several functions have been postulated for this HSP 70 homologue: mediation of cell-to-cell movement through plasmodesmata via interactions with the cellular translocation machinery (such as cytoskeletal or secretional proteins); involvement in the assembly of multisubunit complexes for genome replication and/or subgenomic RNA synthesis, or assembly of viral particles. Recent evidence, showing that BYV mutants in which the expression of the HSP 70 protein is abolished are restricted to inoculated cells, demonstrates the role of this protein in cell-to-cell movement of the virus.

A further peculiarity of members of the Closteroviridae is the duplication of the capsid protein gene. To date, this is the only known case of such a duplication in viruses with elongated particles. The capsid proteins of BYV and CTV, and their homologues, show a high degree of sequence conservation and, according to the pattern of conserved amino acid residues, the duplicate copies probably retain the general spatial folding and some crucial properties of the CPs. Interestingly, in LIYV, LChV, GLRaV-2 and GLRaV-3, the CP gene and its tandem copy have the opposite order as compared to BYV and CTV, and the similarity is much lower.

The large size of the Closteroviridae genomes has been proposed to be due to the acquisition of foreign coding sequences via RNA recombination (protease, HSP 70 protein) and gene duplication, which may also explain the significant variability in genome organization observed between genera and among members of the same genus. Phylogenetic reconstructions based on POL, HEL and MT amino acid sequence comparisons indicate that the Closteroviridae should be regarded as members of the ‘alpha-like’ supergroup of viruses, their closest affinities being with the family Bromoviridae. Sequence comparisons show that the CPs of BYV and CTV are related (24% homology). These proteins belong to a lineage that includes the CPs of many other filamentous plant viruses (carla-, potex-, capillo-, tricho-, viti- and potyviruses) but they are not closely related to the CPs of members of any of these groups.

Host Range and Geographic Distribution

Host ranges of individual members of the Closteroviridae vary considerably but are usually rather restricted. While BYV was reported to infect over 120 species in 15 families, many other viral species have host ranges limited to a single botanical family. This is the case of CNFV, CTV, wheat yellow leaf virus (WYLV), LChV, and the GLRaVs. Very different geographic distribution situations are reported, depending on the individual virus considered: while some viruses cover essentially the totality of their host’s geographic range (e.g. CTV, BYV, CNFV, GLRaVs), others have rather restricted distributions.

Virus–Host Relations: Cytopathic Effects

Some closterovirus, but not crinivirus species, are transmitted by sap inoculation, though with great difficulty and low efficiency, most likely because of their phloem-restricted condition. The cells of infected plants show two types of cytopathic effects: nonspecific subcellular changes accompanying chlorotic or necrotic reactions (e.g. membrane proliferation, vesiculation of chloroplasts, degeneration and vesiculation of mitochondria, accumulation of osmiophilic granules) and specific reactions, in the form of inclusion bodies. Two types of inclusions are observed: aggregates of virions and membranous vesicles, or combinations of the two. Virions occur either in crossbanded aggregates made up of stacked tiers of particles packed side by side, or in loose wavy paracrystalline aggregates, or, more typically, in irregular bundles intermingled with single or clustered membranous vesicles containing a network of fine fibrils. Inclusion of the latter kind are one of the hallmarks of members of the Closteroviridae, and are often referred to as BYV-type inclusion bodies. They are limited to the phloem of infected cells, i.e. sieve tubes, companion cells and phloem parenchyma, but occasionally, as with BYV, can also be found in the mesophyll and epidermis near local lesions. The fibrillar material contained within the membranous vesicles has been interpreted as possible double-stranded RNA, thus suggesting a role for these structures in viral replication, a contention that needs experimental verification.

Virus Transmission

BYV, the type species of the Closterovirus genus, has aphid vectors. However, other types of vectors (mealybugs, whiteflies) have been reported for members of the same genus. In many cases, the mode of transmission or the vector has not been determined (Table 1). Studies have mostly dealt with aphid transmission of BYV and CTV. Information on the transmission of other viruses is still very limited and, in many instances, transmission has only been demonstrated under laboratory conditions. In the case of BYV, CTV and other closteroviruses, aphid transmission is semipersistent, i.e. long acquisition and transmission periods (minimum feeding time for acquisition and transmission in the range of 15 and 30 min, respectively), relatively long retention of the virus (up to 2–3 days) and absence of a latency period. The virus does not appear to be retained through molts or to be transmitted to the vector progeny. In general, relatively short acquisition or transmission feeding periods result in lower efficiency of transmission, this being correlated with the length of these periods (up to 24 h for acquisition, up to 6 h for transmission). The molecular bases of the semipersistent mode of aphid transmission are currently not understood. In particular, the localization of virions in viruliferous aphids is not known, nor are the reasons for the long retention period. On the other hand, the close association of the viruses with phloem tissues accounts for some of the transmission characteristics, such as the length of the acquisition and transmission feeding periods and the lack of transmission during brief probings. One hypothesis, which remains to be verified, is that the CP duplicate may facilitate the interaction between virions and the insect's foregut.

Vector ranges vary from rather wide to restricted, depending on the virus: BYV has been shown to be transmitted by 23 aphid species (Myzus persicae and Aphis fabae being the major natural vectors), CTV by seven species (Toxoptera citricida and Aphis gossypii being the most efficient vectors) and a number of other viruses, such as CNFV and WYLV, by a single species.

Some of the closterovirus species and all those of the Crinivirus genus are transmitted semipersistently by whiteflies: namely, BPYV, TICV and CCSV, transmitted by Trialeurodes vaporariorum; LIYV, CYSDV and SPSVV, transmitted by Bemisia tabaci; LCV, transmitted by Bemisia argentifolii; diodia vein chlorosis virus (DVCV) and abutilon yellow virus (AYV), transmitted by Trialeurodes abutilonea. Persistence and specificity of transmission of whitefly-transmitted viruses in their respective vectors has been used as a biological character for virus species differentiation. For example, LIYV is retained by viruliferous vectors for a maximum of 3 days, while LCV and CYSDV can be retained for 4 and 9 days, respectively.

Pseudococcid (Planococcus, Pseudococcus, Phenacoccus, Saccharicoccus and Dysmicoccus) and coccid (Pulvinaria) mealybugs have been reported to transmit five different closteroviruses (Table 1). The mode of transmission, determined so far only for GLRaV-3, is semipersistent, and may not be vector-specific, as this virus can be transmitted by both pseudococcid (Planococcus and Pseudococcus) and coccid (Pulvinaria) mealybug genera.

Diseases and Their Economic Significance

The most frequent type of symptoms induced by members of the Closteroviridae family are of the yellowing type, frequently accompanied by phloem necrosis. Phloem localization of these viruses probably accounts for these types of symptoms, which are also caused by luteoviruses, another group of phloem-restricted viruses. The economic impact varies widely with the virus: while some are known to affect weeds only, many others have a significant detrimental impact on crop plants. These include BYV, CTV, CNFV, whitefly-transmitted viruses like BPYV, LIYV, and the GLRaVs.

BYV is responsible for serious losses of beet and spinach. Together with beet mild yellowing and beet western yellows luteoviruses, this virus is responsible for the so-called yellowing diseases of beet. In some European countries, losses of sugar beet yield have been estimated to be as high as 15–30% in some years. This reduction is mostly due to decreased root yields, with only marginal effects on the sugar content. Severity of losses correlates with the time of infection and the susceptibility of the affected cultivar. In artificial infection trials, yield losses as high as 40–60% have been recorded, but late season infection has only marginal effects on the yield.

CTV is undoubtedly one of the major citrus pathogens, recognized as one of the top ten viruses with the highest economic impact in eight out of ten wide geographic areas of the world. It is estimated that over the past 50 years or so, CTV has caused the loss of over 80 million trees worldwide. The name ‘tristeza’ is derived from a condition leading to a quick decline of sweet orange and numerous other commercial citrus species grafted on sour orange rootstocks, a symptom due to a scion–rootstock combination problem, with no decline appearing on plants grafted on tolerant rootstock. The most devastating disease induced by CTV is probably stem pitting of a number of citrus species, such as grapefruits, some sweet orange varieties, and sour and sweet limes. Stem pitting is usually accompanied by loss of plant vigour, decline and reduced fruit size. Other major types of diseases caused by CTV include vein clearing of sour lime and seedling yellows, a severe stunting disorder developed by some varieties when infected at an early stage. The pattern of symptoms induced by CTV is complicated by large differences in species and varietal susceptibility and by the existence of an extremely large biological variability between CTV strains.

Whitefly-transmitted viruses are the widespread agents of emerging diseases causing major losses to a number of crops, such as lettuce, sugar beet and cucurbits. A clear-cut estimate of yield losses caused by viral infections is complicated by the difficulty of correctly identifying the agent, because field symptoms (stunting, yellowing and reduced vigour) can also be attributed to the whitefly vectors. In documented cases of severe LIYV infection, losses of up to 80% have been reported from southwestern USA. Epidemic virus outbreaks have been estimated to induce, in a single season, losses of lettuce, sugarbeet, melon and squash worth $US8 million. The recently observed increase in the impact of these viruses appears to be directly linked to the explosion of whitefly vector populations during the last few years, which may have been favoured by a number of factors, i.e. the widespread use of synthetic organic insecticides, resistance to pesticides, changing climatic conditions, intensified agricultural practices and international distribution of plant material contaminated by whitefly populations.

Leafroll disease is one of the most economically relevant diseases of the grapevine and occurs worldwide. It causes yield losses of 10–70%, lowers the sugar content, soluble solids and phenolic compounds of the berries, and reduces graft take and rooting ability. The etiology of the disease is being unravelled, for at least three (GLRaV-1, GLRaV-3, and GLRaV-7) of the seven serologically distinct closteroviruses associated with it have been experimentally proven to be genuine leafroll agents. GLRaV-2 is the cause of both leafroll and a graft incompatibility condition.

Virus Epidemiology and Control

Transmission by vectors may not be the major means of long-distance closterovirus dispersal, which is primarily due to distribution of contaminated propagating material through trading. This is obviously the case for CTV, CNFV and GLRAVs, and the cause of the transfer of BYV from Europe to the USA. In regions where a given virus is absent, quarantine is the best protective measure. In infected areas, infections can be controlled through the production and use of virus-free propagative material, along with control of insect vector populations and destruction of virus reservoirs. When possible, these measures can be coupled with the use of resistant or tolerant varieties or rootstocks. A search for natural resistance has been successful for CTV, but not for GLRaVs, and is currently underway for criniviruses.

The most popular detection techniques for certification of propagation and planting material include bioassays on susceptible hosts (indexing) and enzyme-linked immunosorbent assays (ELISA). Detection of double-stranded RNAs associated with viral infection has also been used, though often inconclusively. The potential of molecular techniques (hybridization, polymerase chain reaction (PCR)) is increasingly being exploited. PCR appears of particular interest when targeting the conserved HSP70 coding region.

In most cases, it has been possible to obtain virus-free plants through the use of meristem tip culture, thermotherapy or a combination of both. A special mention should be made of the use of crossprotection to control tristeza. This strategy is based on the preinoculation of trees with mild CTV isolates to protect them from the expression of symptoms caused by more severe superinfecting isolates. Crossprotection is not without risk, but with CTV it has been used commercially and, in general, with great success in a number of countries, to protect several million trees.

Future Perspectives

Members of the Closteroviridae family are important pathogens of plants but, despite the recent remarkable advances, information on their genomic organization and interactions with hosts and vectors is still limited. Stumbling blocks are the large genome size of these viruses and the difficulties encountered with their characterization and purification. As molecular data has accumulated, the taxonomy of this formerly heterogeneous group of viruses, previously based on morphological criteria, has evolved towards more phylogenetically based groupings. In some crops, such as grapevine, new viruses continue to be discovered. At the same time, in the western USA, an explosion of the whitefly populations and establishment of the new species B. argentifolii, has led to the discovery of several new diseases and viruses. Comprising both very old diseases of high incidence and emerging viruses, the field of closterovirology is rapidly becoming a very active and dynamic research area. The recent development of synthetic inocula using cDNA clones from several of these viruses offers new opportunities for the understanding of the relationships with their hosts and vectors. At the same time, molecular biology techniques have greatly improved our ability to detect and characterize these viruses, which should rapidly translate into increased capability to fight the diseases they cause.

See also:

CAPILLOVIRUSES; TRICHOVIRUSES; PLANT VIRUS DISEASE – ECONOMIC ASPECTS.