Transcription map of the B genome component of tomato golden mosaic virus and comparison with A component transcripts

In a previous study, the bipartite genome of tomato golden mosaic virus (TGMV) was shown to be transcribed into at least six polyadenylated RNAs (G. Sunter, W.E. Gardiner, and D. M. Bisaro, 1989, Virology 170, 243-250). Two of these, a 1.3-kb complementary sense and a 0.9-kb viral sense transcript, were mapped to the B genome component of this geminivirus. The results of more detailed primer extension and S1 nuclease protection experiments presented here define the limits of the single transcription unit corresponding to the 0.9-kb RNA which spans the BR1 open reading frame (ORF). The data also demonstrate that complementary sense TGMV RNAs are more complex than indicated by our earlier studies. Analysis of the 1.3-kb BL1-specific RNA indicates that it is actually a family of distinct transcripts with different start sites. Three transcripts have 5' ends that map near the common region of DNA B and all of these start sites lie upstream of the BL1 ORF. Similar analysis of the 1.6-kb complementary sense AL1 RNA indicates that a complex set of transcripts also map to the analogous region of genome component A. Four transcripts have 5' ends that map near the common region but only one of these start sites is upstream of the initiation codon for the AL1 open reading frame (ORF). None of the transcripts appear to be processed. The possible significance of multiple transcripts in these regions of the TGMV genome is discussed, and the common region-proximal transcription units of the A and B genome components are compared.