MOLECULAR ANALYSIS OF THE CUCUMBER NECROSIS VIRUS GENOME

Cucumber necrosis virus (CNV), a recently established member of the tombusvirus group, is a simple isometric virus which encapsidates a small (4.7 kb) monopartite, positive polarity RNA genome. The nucleotide sequence of CNV predicts the synthesis of 5 proteins with molecular weights of 33, 92, 41, 21, and 20 kDa. The 92 kDa protein could arise from readthrough of the amber terminator of the open reading frame (ORF) for the 33 kDa protein and the 21 and 20 kDa proteins from different 3' terminal nested ORFs. In vitro translation studies using wheat germ extracts demonstrated that each of the proteins predicted from the CNV sequence can be translated from virion RNA with the exception of the 92 kDa protein. The 92 kDa protein was not detected even when exogenous tRNA from a number of sources was added to supplement the system. Highly infectious transcripts from full-length CNV cDNA clones were synthesized and used in mutagenesis studies to determine the modes of expression and the functions of CNV-encoded proteins. In vitro and in vivo studies using transcripts with mutated AUG codons for the 21 and 20 kDa nested ORFs demonstrated that CNV generates a bifunctional subgenomic mRNA which can encode both the 21 and 20 kDa proteins. A synthetic transcript containing an altered AUG codon for the 20 kDa protein produces only mild symptoms suggesting that the 20 kDa protein may be involved, either directly or indirectly, in the severe necrosis typical of CNV-infected Nicotiana clevelandii. Finally, amino acid sequence comparisons between the CNV 92 kDa protein and the putative replicases of representative members of several plant virus groups (luteo-, carmo-, necro-, diantho-, maize chlorotic mottle) and the animal virus Hepatitis C suggest that these viruses should be classified together in a supergroup distinct from the proposed alpha- and picornavirus supergroups.