Search

Complete sequences for the intergenic regions of the genome of human respiratory syncytial virus were obtained by dideoxynucleotide sequencing using synthetic oligonucleotides. These experiments established that the 10 respiratory syncytial viral genes are arranged, without additional intervening genes, in the order 3' 1C-1B-N-P-M-1A-G-F-22K-L 5'. For the first nine genes, the exact gene boundaries were identified by comparison of the genomic sequences with previously determined mRNA sequences. The intergenic regions varied in length from 1 to 52 nucleotides and lacked any obvious conserved features of primary or secondary structure except that each sequence ended (3' to 5') with an adenosine residue. The exact start site of the 10th gene, the L gene, was not determined. However, RNA blot hybridization using a synthetic oligonucleotide designed from the genomic sequence mapped the L gene to within 54 nucleotides of the end of the penultimate 22K gene. The lack of conservation of chain length and nucleotide sequence for the respiratory syncytial viral intergenic regions, together with the complexity of the genetic map, contrasts with previous observations for other nonsegmented negative-strand viruses.

A transcriptional map for human respiratory syncytial virus was determined by measuring the kinetics of viral gene inactivation in response to UV irradiation. Monolayer cell cultures of respiratory syncytial virus-infected HEp-2 cells were exposed to UV light, and residual viral RNA synthesis was monitored both by gel electrophoresis and by hybridization to dot blots of cloned cDNAs of the 10 known viral genes. Target sizes for the 10 individual viral genes were calculated relative to the UV sensitivity of intracellular viral genome replication. Target size analysis indicated that the 10 viral genes were transcribed as a single transcriptional unit and that the transcription of an individual gene was dependent on the prior transcription of all the preceding genes. The order of gene transcription (with nomenclature according to encoded proteins) was determined to proceed from the promoter as follows: 14K, 11K, N, P, M, 9.5K, 36K, F, 24K, L.

A full-length cDNA clone for the 24S subgenomic mRNA of the vaccine strain (HPV77) of rubella virus has been isolated from a cDNA library made from the RNAs of infected cells. Starting from the first Met start codon, the 24S mRNA codes for a precursor protein of 1063 amino acids (aa). This precursor encodes a capsid protein of 300 aa, and two envelope proteins, E1 (481 aa) and E2 (282 aa). Both the E1 and E2 proteins are preceded by a stretch of 21 hydrophobic aa, characteristic of a signal peptide, and each has three putative glycosylation sites in the polypeptide chains. Comparison between the structural proteins of the vaccine and the wild-type (wt; M33) strains of rubella virus, revealed that the E2 protein of the vaccine strain differs, in its apparent Mr, by approx. 3 kDa, from the wt strain. The difference could be due to decreased glycosylation of the vaccine strain E2 protein, as revealed by [3H]mannose incorporation studies. Five single-aa changes in the structural proteins occurred during the attenuation process, one each in the capsid and the E1 protein and three in the E2 protein. The change of Thr-412----Ile in the E2 protein results in the loss of a putative glycosylation site at Asn-410, which offers a plausible explanation for decreased glycosylation of the E2 protein from the vaccine strain of rubella virus.