Species: Subterranean clover stunt virus
P. W. G. Chu
CSIRO Plant Industry, GPO Box 1600, Canberra, ACT, 2601, Australia
H. J. Vetten
BBA, Institut f. Pflanzenvirologie, Mikrobiologie und biologische Sicherheit, Messeweg 11-12 D-38104 Braunschweig, Germany
Host Range and Symptomatology
Transmission by Vectors
Transmission through Seed
Transmission by Grafting
Transmission by Dodder
Nucleic Acid Hybridization
Stability in Sap
Properties of Particles
Properties of Infective Nucleic Acid
Relations with Cells and Tissues
Ecology and Control
The virus is the type member of the genus Nanovirus, (family Nanoviridae). The virus genome consists of at least 6 distinct circular single-stranded (ss) DNA components each of about 1 kb in size with each encapsidated in a separate icosahedral particle of 17-19 nm diameter. Each DNA component appears to possess only one major gene and to contain all signals required for replication and gene expression. The virus particle contains a major protein of Mr 19,000.
The natural host range seems restricted to legumes. Like all other nanoviruses, it is transmitted only by aphids in a non-propagative circulative manner and for which a virus-encoded helper factor appears necessary.
The virus causes severe reduction in growth and premature death of affected subterranean clover plants (Fig.1, Fig.2) with individual plant yield losses of up to 65% (Grylls & Butler, 1956; O'Loughlin, 1958; Anonymous, 1976). Infected plants are much less competitive and are more susceptible to root and crown rots, resulting in high mortality rates. As crop infections tend to be clumped and plants infected early usually die, concentration of SCSD in certain parts of a paddock can lead to bare ground or a displacement of clover by weeds. Field studies showed that when the incidence of SCSD was 13-80%, the forage yield of total plant matter was reduced by 30-76% with a corresponding reduction in yield of the clover component of 35-81%. Clover seed yields were also reduced by 58-94%. A disease incidence of 6-20% early in the growing season reduces seedling regeneration in the following season by 30-48% and increases the areas of weed and bare ground by 7-43%.
The virus also infects peas and beans and peas (Fig.4, Fig.5), causing a disease known as top yellows, tip yellows or leafroll (Letham, 1982). During the 1950s and 1960s, the virus was responsible for major losses of 86 to 100% in vegetable and seed crops of beans and peas throughout southern Australia annually (Smith, 1966; Grylls, 1972; Letham, 1982). Repeated crop failures and lack of resistant cultivars have resulted in reduced cropping in these regions and increased the need for spraying against the insect vector. Other potentially important susceptible or alternative hosts include annual Medicago species (O'Loughlin, 1958), lupins, lucerne, lentil, chickpea and soybean.
Milk vetch dwarf virus (MDV), a closely related nanovirus, is reported only from Japan. By contrast, Faba bean necrotic yellows virus (FBNYV), a close relative of MDV, occurs in many Near Eastern, Middle Eastern, and North African countries as well as in Ethiopia and Spain.
The relative severity of disease symptoms on different hosts depends on the virus isolate rather than the host and, with mild isolates, infected plants show a partial recovery. Experimentally infected lucerne and lupins are symptomless.
Comparing distinct regional virus isolates (Chu et al., 1989) showed that most could be classified into one of three groups based on the severity of their symptoms on different hosts:
Isolates causing severe stunting that occur mainly in wetter temperate regions of the South Eastern regions of New South Wales and Victoria. These isolates induce a severe shortening of internodes and petioles, a reduction in leaf size and numbers and a consequent reduction in stem or stolon length and number. Plants affected by these isolates usually develop severe chlorosis of leaf margins, puckering in young leaves followed by reddening (especially in subterranean clover) of leaf margins, and sometimes a dark greening of leaf blades in matured leaves and subsequently premature browning and drying off of old leaves.
Isolates causing a moderate reduction in internode, petiole and stolon lengths without any obvious reduction in leaf number so that plants become bunchy in appearance. Infected plants usually develop yellowish-green cupped leaves with intermittent vein-clearing without leaf reddening but older leaves may continue to develop more severe chlorosis and subsequently dry off.
Isolates producing mild yellowing in young leaves initially but leaves later become either symptomless or show only a mild yellowing with intermittent vein-clearing but with no obvious reduction in any growth parameters. These isolates appear to be found mainly in the drier Mediterranean regions of South Eastern South Australia and Western Victoria.
ISEM, Western blotting, dot-blotting and agar gel double diffusion tests are suitable only for detecting virus in purified preparations but and are less sensitive than ELISA. No monoclonal antibodies to the virus have been produced (Chu et al., 1995)
All known virus isolates are related serologically and cross-hybridise with each other. Phenotypically distinct isolates may differ in symptom severity, host range, serological reactivity and the electrophoretic mobility of their DNA components in polyacrylamide gel electrophoresis. New South Wales isolates A, F and AA infect all 54 lines of subterranean clover tested, but a Tasmanian isolate SCS6 was unable to infect 14 of these lines. Only some isolates infect lupin. Cross-protection experiments performed with subterranean clover indicated that a mild virus isolate can protect plants against severe symptoms from infection by a severe isolate (Chu et al., 1995).
Sequence analysis of PCR products of DNA-S from different virus isolates found significant nucleotide sequence variations between them in both coding and non-coding regions. The non-coding region of DNA-S is the most variable, differing by up to 34% between isolates whereas coding regions differed by up to 15% resulting in up to 10% amino acid changes (Boevink et al., 1995). This may reflect the variable serological reactivity of the isolates. There was no association between sequence variation in DNA-S and specific differences in biological properties and in the geographic origins of the isolates.
PCR amplification of the DNA from different isolates using a conserved set of primers to the common region, identified a polymorphism in the size of some minor bands in addition to a constant major band of c. 1 kb for every isolates, suggesting the presence of different sized subgenomic DNAs in these isolates. Virus isolates may also be characterised by the presence or absence of specific satellite-like DNA components. However, the biological variation of virus isolates could not be related to a specific size or intensity of the subgenomic DNAs and/or the presence or absence of a specific satellite-like DNA (Chu et al., 1995).
These results indicate the very variable nature of the virus and it may exist as a broad continuum of closely related isolates. The large number of genomic components suggests the huge potential for the virus to change through genetic reassortment. The complete sequence analysis of the genomic DNAs of several isolates may determine the full extent of variation and the taxonomic relationships amongst them.
Based upon the analysis of the available sequence information, 6 to 8 distinct circular viral ssDNA components are thought to be integral parts of the nanovirus genome (Timchenko et al., 1999, 2000). In the 6 genomic DNAs described from SCSV, the virus shares an overall nucleotide sequence identity of about 55% with the three other known nanovirus species (Faba bean necrotic yellows virus [FBNYV], Milk vetch dwarf virus [MDV] and Banana bunchy top virus [BBTV]) and takes an intermediate position between BBTV and the two very closely related nanovirus species FBNYV and MDV (Katul et al., 1998; Randles et al., 2000). The most highly conserved gene product of nanoviruses is the master replication initiator (M-Rep) protein, which, in SCSV, is encoded by DNA-R (formerly C8; Timchenko et al., 2000). The M-Rep protein of the virus shares amino acid sequence identities of about 83% and 55% with those of FBNYV (and MDV) and BBTV, respectively (Timchenko et al., 2000). Further significant levels of amino acid sequence identity among protein homologs encoded by all known nanoviruses (> 45%) have been observed only for the putative nuclear shuttle protein encoded by DNA-N (formerly DNA C4 in SCSV). In all the other gene products, including the coat protein (CP), the virus shares significant levels of amino acid sequence identity with those of FBNYV and MDV (> 44%) but not with those of BBTV (< 25%) (Katul et al., 1998; Randles et al., 2000).
Antisera to the virus and to FBNYV cross-react weakly with each other in Western immuno-blots and immuno-electron microscopy, but not at all in DAS-ELISA (Katul et al., 1993; Chu et al., 1995). Whereas the majority of 19 monoclonal antibodies to FBNYV reacted with MDV, only one of them reacted with the SCSV isolates F and I, indicating that the serological relationship between FBNYV and MDV is close but remote between SCSV and FBNYV (Katul et al., 1993; Franz et al., 1996). This is confirmed by amino acid sequence identities of 53.5 to 55% for the coat proteins of SCSV and FBNYV (Katul et al., 1998), and of 83% for those of FBNYV and MDV (Sano et al., 1998).
Buoyant density: 1.24 g.cm-3 in caesium sulphate and 1.34 g.cm-3 in caesium chloride.
Particle weight: Approximately 1.6 million daltons.
Extinction coefficient: 3.6 at A260 nm (1 mg/ml, 1 cm light path) (corrected for light-scattering).
A260/A280: 1.35 (corrected for light-scattering) (Chu & Helms, 1988)
Purified virus particle preparations contain up to eight circular ssDNA components (identified previously as C1-C8), each of about 1 kb in size and thought to be encapsidated individually in particles (Fig.7A and Fig.8). However, only six of these appear to be integral components of the virus genome (Boevink et al., 1995; Timchenko et al., 2000). By analogy with Faba bean necrotic yellows virus (Timchenko et al., 1999) and Banana bunchy top virus (Horser et al., 2001), the two Rep-encoding DNAs, C2 and C6, are regarded as satellite-like DNAs.
DNA prepared from purified particles also contains linearised forms (Fig.7B) of the circular DNA components that probably result from degradation during virus purification or DNA extraction. The DNA is about 23% of the particle weight. In addition to the two satellite-like DNAs, subsidiary nucleic acids are likely to include primers required for dsDNA synthesis upon infection (Hafner et al., 1997a).
The virus particle is presumably comprised of 60 chemical subunits of a single protein species (Chu & Helms, 1988) that in SDS-PAGE migrates as a 19- kDa band but has a Mol. Wt of approximately 18.6K as deduced from the major ORF on DNA-S (formerly DNA-C5) (Boevink et al., 1995). This protein is about 77% of the particle weight.
The viral genome consists of at least 6, structurally similar circular ssDNA species, each of c. 1 kb (Boevink et al., 1995; Timchenko et al., 2000). Each of these species contains one major ORF transcribed unidirectionally in the viral sense, and a non-coding region in which there are inverted repeat sequences potentially forming a stem-loop structure and encompassing the origin of replication. Each DNA species contains both a typical TATA box and a polyadenylation signal flanking the ORF, indicating the start and end of transcription, respectively (Boevink et al., 1995). The stem-loop sequence has been shown to be involved in nanovirus DNA replication (Hafner et al., 1997b; Timchenko et al., 1999).
By analogy with other nanoviruses (Timchenko et al., 1999; Vetten and Katul, 2001), the virus genome consists of at least the following ssDNA species and encodes the following proteins (Boevink et al., 1995; Timchenko et al., 2000) (Fig.8):
The sequences in the stem and loop are totally conserved between DNA-S, - N, and -U1 and vary only slightly between DNA-M, -C, and -R. They are all clearly distinct from the stem sequences of the satellite-like DNAs C2 and C6.
The degree of sequence conservation in the non-coding regions varies with the different DNA components. The non-coding regions of DNA-C and -S are most similar, sharing 258 conserved nucleotides, DNA-C, -N, -S, and -U1 share 170 identical nucleotides and DNA- C, -M, -N, -S, and -U1 share 152 nt.
The biochemical events involved in SCSV replication have not been determined. Available information for other nanoviruses suggests that replication is similar to that of geminiviruses in being completely dependent on the host cell's DNA replication enzymes, and occurs in the nucleus through double-stranded DNA intermediates by a rolling circle replication mechanism (Laufs et al., 1995; Bisaro, 1996; Hanley-Bowdoin et al., 2000). Upon inoculation, the nuclear localisation signals in the N-terminus of the coat protein ensure that the virus is translocated to the nucleus. Upon decapsidation of viral ssDNA, one of the first biosynthetic events is the synthesis of viral dsDNA (Chu et al., 1993a) with the aid of host DNA polymerase. As the virus DNAs have the ability to self-prime during dsDNA synthesis (Chu & Helms, 1988), it is likely that pre-existing primers are used for dsDNA replicative form (RF) synthesis, as has been shown for mastreviruses (Donson et al., 1984, 1987) and BBTV (Hafner et al., 1997a). From these dsDNAs (RF) forms, host RNA polymerase would then transcribe mRNAs encoding the M-Rep and other proteins required for virus replication. Viral DNA replication is initiated by the M-Rep protein that interacts with common sequence signals on all the genomic DNAs (Timchenko et al., 1999, 2000; Horser et al., 2001). Replication of the DNAs is by cellular enzymes, facilitated and enhanced by the action of Clink, a nanovirus-encoded cell cycle modulator protein (Aronson et al., 2000).
The virus is able to replicate in isolated pea and subterranean clover protoplasts (Chu et al., 1993b) and de novo synthesis of virus coat protein and DNA was detected by ELISA and nucleic acid hybridization using strand-specific probes, respectively. Nucleic acid hybridization showed that both complementary- and virion-sense DNAs accumulated in protoplasts to a maximum between days 3-7 and at day 7, respectively. The amount of virus coat protein decreased from day 0 to day 3 post-inoculation but increased thereafter over several days to reach a maximum at day 10. The kinetics of virus synthesis in protoplasts is similar to that observed for a geminivirus synthesis in protoplasts (Townsend et al., 1986; Woolston et al., 1989).
The virus and its aphid vector are endemic in most regions of southern Australia and are probably maintained all year round in a variety of perennial and annual legume crops and weeds growing over extensive areas. Infection occurs from late autumn to early spring in subterranean clover pastures, which is when pasture productivity is the chief factor limiting stock number and production per head (Grylls & Butler, 1959). High virus incidence in autumn is associated with large numbers of vectors during the summer on a variety of legume plants (Grylls & Butler, 1959; Grylls, 1972). A. craccivora transmits the virus over short distances from crop to crop (Grylls, 1972) and over long distances by migration (Johnson, 1957; Gutierrez et al., 1971). Climatic conditions have a major influence on colonization, establishment and dispersal of vector aphids (Gutierrez et al., 1974). Vector populations are thought to breed on various legume crops in the cooler, wetter coastal and highland areas of southern Australia during late spring and summer and, during autumn and winter, migrate to milder areas as far as northern New South Wales and southern Queensland where subterranean clover, annual Medicago species and other legume crops begin germinating. Annual aphid migration corresponds to annual subterranean clover germination and establishment in various parts of southern Australia. The aphid continues to breed on grain legume hosts, especially broad beans, until early winter (Grylls, 1972). A. craccivora multiplies rapidly under favourable conditions, sometimes reaching 11,000 aphids per plant and serious virus epidemics follow.
ELISA, PCR and nucleic acid hybridization are used for virus diagnosis. No cross-reactions occur with geminiviruses, luteoviruses and other legume- infecting viruses. ELISA detects the virus 4-5 days before the appearance of symptoms, which occurred at about 10 days after inoculation, whereas the nucleic acid hybridisation assay detects the virus 2-3 days before the onset of symptoms. PCR was is the most sensitive assay, detecting the virus 6-7 days before symptoms appeared.
There are no effective natural sources of virus resistance in either French bean (Smith, 1966) or subterranean clover (Chu et al., 1995). Attempts to breed subterranean clover lines with resistance to all SCSD isolates (Grylls & Peak, 1960) have been unsuccessful. Control measures include increasing seeding rates from 3.4-10 kg/ha to compensate for loss from virus infection soon after germination. The large number of different virus isolates, the extensive genetic variation among them, and the potential capacity for genetic recombination between the isolates makes successful classical breeding for virus resistance unlikely. Genetic engineering of resistance using virus-derived genes may provide a better hope of success.
Field identification of the disease is difficult due to the large variation of symptoms displayed, its similarity to other diseases and disorders, and mixed infections with other viruses. The virus can be confused with infections by luteoviruses such as Bean leaf roll virus and SDV, and in subterranean clover, with those of Subterranean clover mottle virus. Although characteristic symptoms as described above are often seen in subterranean clover pastures, usually the symptoms of the disease vary even within a field. Well spaced plants infected by a severe virus isolate early in the season usually develop a rosetted or stunted and bunchy appearance while those infected by a severe isolate late in spring or in a sheltered position develop a spindly appearance with marked chlorosis and puckering of young leaves. Plants infected by a mild isolate may develop mild yellowing or reddening or appear symptomless.
There are both biological similarities and differences between SCSV, Faba bean necrotic yellows virus and Milk vetch dwarf virus. These three legume-infecting nanoviruses are serologically related, share many common natural and experimental host plants and exhibit similar symptoms on many of these common host species (Chu et al., 1995; Katul et al., 1993; Franz et al., 1997). They also have a common aphid vector, A. craccivora. All the three viruses are highly variable (Chu et al., 1995; Katul et al., 1993, 1999; Franz et al., 1998) in terms of symptoms, other aphid vectors and host plant species as well as serological reactivity among isolates (Franz et al., 1996, 1997, 1998). Differentiation between these viruses is currently based firstly on their different geographical location, then serological analysis (e.g., DAS-ELISA using polyclonal antibodies), and finally confirmed by by sequence analysis of individual genomic DNAs (e.g., DNA-S). Coat protein amino acid sequence differences of > 15% are regarded as a species discriminator.
Effect of virus infection on stunting, reduced foliage and root growth in subterranean clover production.
Moderately severe virus infection on subterranean clover causing, from left to right, reddening of mature leaves, marginal chlorosis on young leaves and general chlorosis and significantly reduced leaf size.
Natural virus infection in a red clover pasture.
Effects of different virus isolates on the growth of broad bean plants. From left, uninfected control, very severe chlorosis with no growth (isolate A), severe stunting and chlorosis (isolate F), moderate stunting but with significantly reduced leaf size (isolate AA), mild stunting and normal growth form (isolate DD).
Effects of different virus isolates on the growth of pea plants. From left, uninfected control, moderate stunting but with prominent leafrolling and chlorosis and significantly reduced leaf size (isolate F), very severe stunting and chlorosis with no growth (isolate A).
Electron micrograph of virus particles in a purified preparation stained with 2% ammonium molybdate. Bar = 50 nm.
Electron micrographs of viral genomic DNA showing circular (A) and linear (B) molecules. Line tracings of the linear molecules (B) are included to indicate the ends of these molecules.
Diagram illustrating the putative genomic organisation of the virus and depicting the structure of the identified viral DNA components. Each DNA circle contains its designated name (former name in parentheses) and size, and the name/function and size of the encoded protein is given below each circle. Stars refer to the positions of the origin of replication of the DNAs containing the stem-loop structure. Arrows refer to the location and size of the ORFs and the direction of transcription.
Dark field microscopy showing histochemical staining for β-glucuronidase (GUS) activity in transverse stem section of a transgenic tobacco plant transformed with a GUS fusion construct containing the promoter region from the viral DNA-S.