Parvovirus H-1 expression: mapping of the abundant cytoplasmic transcripts and identification of promoter sites and overlapping transcription units.

The 5.2-kilobase (kb) genome of the autonomous parvovirus H-1 was transcribed in the rightward direction, yielding steady-state polyadenylated transcripts of 4.8, 3.2, and 2.9 kb. Detailed mapping of these transcripts demonstrated that the H-1 genome contained two overlapping transcription units: the larger unit extended from 4 map units (5' end) to 96 map units (3' end), and the smaller unit extended from 40 map units (5' end) to 96 map units (3' end). The 4.8- and 3.2-kb transcripts were derived from the larger transcription unit and differed by a 1,500-nucleotide segment (10 to 40 map units) which was present in the 4.8-kb transcript but was spliced from the 3.2-kb transcript. The 2.9-kb transcript, the most abundant of the three known H-1 transcripts, was derived from the smaller transcription unit. The sequence at each initiation site was consistent with the presence of a class II (RNA polymerase II) promoter, and cell-free transcription of parvovirus H-1 restriction fragments containing either promoter resulted in transcription of the correct DNA strand and produced 5' ends identical to those seen in vivo. All three transcripts contained a small but heterogeneous splice at 45 to 47 map units. Minor differences in splicing at this site may result in the synthesis of different viral proteins.