SUMMARYMild proteolysis by trypsin of particles of six potyviruses (bean yellow mosaic virus, clover yellow vein virus, Johnson grass mosaic virus, passion-fruit woodiness virus, potato virus Y and watermelon mosaic virus II) revealed that the N-and C-terminal regions of their coat protein are exposed on the particles' surfaces. The enzyme treatment removed the N-terminal region (30 to 67 amino acids long, depending on the virus) and 18 to 20 amino acids from the C terminus of the coat proteins, leaving a fully assembled virus particle composed of coat protein cores consisting of 216 or 218 amino acid residues. These core particles were indistinguishable from untreated native particles in an electron microscope and were still infectious. The core particles lacked the virus-specific surface epitopes that are recognized by the bulk of the polyclonal antibodies raised against the whole virus particles. Epitopes thought to be groupspecific were located in the trypsin-resistant core protein region. The implications of these findings are discussed in relation to the similar surface location of the N-and Cterminal regions of the coat protein of other rod-shaped plant viruses and the observed common structural features displayed by isometric plant and animal viruses.

No abstract

SUMMARYAnalysis of the 136 possible pairings of the coat protein amino acid sequences from 17 strains of eight distinct potyviruses revealed a bimodal distribution of sequence homology. Distinct members of the group exhibited sequence homologies ranging from 38 to 71 ~ (average 54%) with major differences in the length and sequence of their N termini and high sequence homology in the C-terminal half of the coat proteins. In contrast strains of individual viruses exhibited sequence homologies of 90 to 99% (average 95 9/o) and had very similar N-terminal sequences. These findings cast doubt on the currently held 'continuum' hypothesis proposed to explain the unsatisfactory taxonomy of the potyvirus group. The coat protein sequence data, in combination with information on the nature of the potyvirus particle assembly, can be used to develop rationally designed, simple serological techniques that appear to be more useful and more easily applied than those properties previously used for potyvirus identification and classification.

There are two major requirements for potyvirus taxonomy; group-specific criteria and criteria which discriminate between distinct viruses and strains. This review discusses the relative merits in potyvirus taxonomy of molecular parameters, such as gene sequence data, nucleic acid hybridization, coat protein sequence data, or high-performance liquid chromatography peptide profiles, and phenotypic characteristics, such as particle morphology, host range, symptomatology, cross-protection, cytoplasmic inclusion morphology, and serology. Coat protein and gene sequence data are the most useful criteria, as they can be used to distinguish viruses from strains and to establish evolutionary relationships between groups of distinct potyviruses. This has led to the revised classification of some viruses and strains and to the clarification of previously conflicting and inconsistent biological properties. It has also provided a sound basis for subgrouping potyviruses. An analysis of the data supports the view that the potyvirus group, including the non-aphid-transmitted viruses, should be elevated to family status, that the vector transmission mode, which correlates with major sequence diversity, should define the four genera, and that distinct potyviruses correspond to species and their variants to strains.

The identification and classification of potyviruses has been in a very unsatisfactory state due to the large size of the group, the apparent vast variation among the members and the lack satisfactory taxonomic parameters that will distinguish distinct viruses from strains. In the past, use of classical methods, such as host range and symptomatology, cross-protection, morphology of cytoplasmic inclusions and conventional serology, revealed a "continuum" implying that the "species" and "strain" concepts cannot be applied to potyviruses. In contrast nucleic acid and amino acid sequence data of coat proteins has clearly demonstrated that potyviruses can be divided into distinct members and strains. This sequence data in combination with information of the structure of the potyvirus particle has been used to develop simple techniques such as HPLC peptide profiling, serology (using polyclonal antibody probes obtained by cross-adsorption with core protein from trypsin treated particles) and cDNA hybridization. These findings, along with immunochemical analyses of overlapping synthetic peptides have established the molecular basis for potyvirus serology; explained many of the problems associated with the application of conventional serology; and provided a sound basis for the identification and classification of potyviruses. As a result, the virus/strain status of some potyviruses has been redefined, requiring a change in the potyvirus nomenclature. These new developments necessitate a re-evaluation of the earlier literature on symptomatology, cross-protection, cytoplasmic inclusion body morphology and serology.

SUMMARYThe sequence of the 3' 1106 nucleotides of the watermelon mosaic virus 2 (WMV 2) genome has been determined. The sequence contains the complete coding region of the viral coat protein followed by a 3' untranslated sequence of 251 nucleotides. When these sequences were compared with the equivalent regions of the N strain of soybean mosaic virus (SMV-N), the coat protein coding regions were 82 ~ homologous, whereas the 3' untranslated sequences were 78~ homologous. Optimal alignment of the 3' untranslated regions of RNA from 13 strains of seven other distinct potyviruses revealed that the degree of homology between strains was in the range 83 to 99~. In contrast, the sequences from distinct viruses had identities in the range 39 to 53~, comparable to the level of identity found between the 3' non-coding regions of viruses from unrelated plant virus groups. On the basis of these results, WMV 2 and SMV-N could be regarded as strains of one virus. These results also suggest that the sequence of the 3' untranslated region of the potyvirus genome may be an accurate marker of genetic relatedness and could serve as an aid to identification and classification of potyviruses.

No abstract

SUMMARYAttempts to identify and classify distinct potyviruses and their strains have frequently been hampered by the presence of variable proportions of cross-reacting antibodies in antisera. Investigations of reactivities in electroblot immunoassays of 11 polyclonal antisera raised by injection of intact particles of potyviruses produced in different laboratories with 12 distinct potyviruses showed that such cross-reacting antibodies were directed towards the homologous core protein region of potyvirus coat proteins. A simple method was developed to obtain virus-specific antibodies using affinity chromatography. It involved removal of the surface-located, virus-specific Nterminal peptide region from particles of one potyvirus using lysyl endopeptidase, coupling of the truncated coat protein to cyanogen bromide-activated Sepharose gel, and passing antisera to different potyviruses through the column. Antibodies that did not bind to the column were found to be highly specific.