Altered utilization of splice sites and 5'' termini in late RNAs produced by leader region mutants of simian virus 40.

We compared the 5' termini and splices of the late 16S and 19S RNAs synthesized by wild-type simian virus 40 and five mutants containing deletions in their late leader region. All mutants produced more unspliced 19S RNA than did wild-type virus, and in two mutants, unspliced 19S RNA constituted more than 60% of the total 19S species. The other three mutants each utilized predominantly a different one of the three spliced species of 19S mRNA. All mutants also produced decreased quantities of 16S mRNA, indicating that they may be defective for splicing both late RNAs. None of the 5' termini of the 16S and 19S RNAs made by the five mutants predominated as in those made by the wild type. Some of the mutant 5' termini were the same as those used by the wild type, whereas others were different. Although present, the major 5'-end positions used by the wild type were frequently not used as major sites by the mutants. In addition, mutants with very similar deletion endpoints synthesized RNAs with different 5' ends. Thus, downstream mutations have a pronounced effect on the location of 5' ends of the late RNAs, and there is no obvious involvement of a measuring function in the placement of 5' ends. For all mutants and wild-type virus, the 5' termini used for 16S and 19S RNAs showed major differences, with some degree of correlation found between the 5' ends and the internal splices of specific mRNA species. A model for the regulation of simian virus 40 late gene expression is presented to explain these findings.     

Heterogeneity of the 5'' terminus of late mRNA induced by a viable simian virus 40 deletion mutant.

d1-1811 is a viable simian virus 40 deletion mutant which lacks the DNA region corresponding to the major capping site of the late viral RNA. The exact size of the deletion (40 base pairs) was determined by comparison of the mutant DNA sequence with the wild-type simian virus 40 (strain 776) DNA sequence. Although d1-1811 forms somewhat smaller plaques, the amount of viral RNA late after infection was not significantly reduced compared with that of the wild type. Virus-specific, polyadenylate-containing, 32P-labeled late RNA was purified from the cytoplasm and enzymatically degraded to characterize the 5' terminus. The cap-containing oligonucleotides were isolated, and their structures were analyzed by further digestion. Instead of a single cap structure, we found a variety of capped 5' termini, with adenosine caps occurring much more frequently than guanosine caps. Nevertheless, there was a remarkable homology between both types of terminal sequences. Conceivably, the minor cap population present in wild-type simian virus 40 late mRNA may correspond to the collection of capped termini identified in the d1-1811 late mRNA . Cellular cytoplasmic RNA shows a similar pattern of cap structures, but the relative abundance is quite different.     

Adenovirus transcription. II. RNA sequences complementary to simian virus 40 and adenovirus 2DNA in AD2+ND1- and AD2+ND3-infected cells.

The genomes of the two nondefective adenovirus 2/simian virus 40 (Ad2/SV 40) hybrid viruses, nondefective Ad2/SV 40 hybrid virus 1 (Ad2+ND1) and nondefective hybrid virus 3 (Ad2+ND3), WERE FORMED BY A DELETION OF ABOUT 5% OF Ad2 DNA and insertion of part of the SV40 genome. We have compared the cytoplasmic RNA synthesized during both the early and late stages of lytic infection of human cells by these hybrid viruses to that expressed in Ad2-infected and SV40-infected cells. Separated strands of the six fragments of 32P-labeled Ad2 DNA produced by cleavage with the restriction endonuclease EcoRI (isolated from Escherichia coli) and the four fragments of 32P-labeled SV40 DNA produced by cleavage with both a restriction nuclease isolated from Haemophilus parainfluenzae, Hpa1, and EcoRI were prepared by electrophoresis of denatured DNA in agarose gels. The fraction of each fragment strand expressed as cytoplasmic RNA was determined by annealing fragmented 32P-labeled strands to an excess of cellular RNA extracted from infected cells. The segment of Ad2 DNA deleted from both hybrid virus genomes is transcribed into cytoplasmic mRNA during the early phase of Ad2 infection. Hence, we suggest that Ad2 codes for at least one "early" gene product which is nonessential for virus growth in cell culture. In both early Ad2+ND1 and Ad2+ND3-infected cells, 1,000 bases of Ad2 DNA adjacent to the integrated SV40 sequences are expressed as cytoplasmic RNA but are not similarly expressed in early Ad2-infected cells. The 3' termini of this early hybrid virus RNA maps in the vicinity of 0.18 on the conventional SV40 map and probably terminates at the same position as early lytic SV40 cytoplasmic RNA. Therefore, the base sequence in this region of SV40 DNA specifies the 3' termini of early messenger RNA present in both hybrid virus and SV40-infected cells.     

Translation initiation at a downstream AUG occurs with increased efficiency when the upstream AUG is located very close to the 5' cap.

The major late 16S mRNA species of simian virus 40 encodes both a 61-amino-acid protein, LP1, and the major virion protein, VP1. Although the initiation signal GCCAUGG is usually utilized at high efficiency, at least one-third of 40S ribosomal subunits bypass it when it is present on the major 16S mRNA of simian virus 40 (S. A. Sedman, P. J. Good, and J. E. Mertz, J. Virol. 63:3884-3893, 1989). The LP1 translation initiation codon is situated 10 bases from the 5' end of this mRNA. To determine whether the short length of the untranslated leader of this mRNA affects the efficiency of translation initiation at the LP1 initiation signal, monkey cells were transfected with plasmids which encode major late 16S-like mRNAs with 5' untranslated regions (UTRs) of 6 or 44 bases. Decreasing the length of the 5' UTR from 44 to 6 bases resulted in a 30% decrease in translation initiation at the LP1 AUG and a threefold increase in synthesis of VP1. When the VP1 open reading frame was replaced with the chloramphenicol acetyltransferase open reading frame, the reduction in 5' UTR length resulted in a 70% decrease in translation initiation at the LP1 AUG and a 30% increase in chloramphenicol acetyltransferase synthesis. Therefore, ribosomes bypass an AUG codon more efficiently when it is located very close to the 5' end of the mRNA.     

Evidence of methylation of B77 avian sarcoma virus genome RNA subunits.

B77 avian sarcoma virus RNA was labeled with (methyl-3H) methionine under conditions that prevent non-methyl incorporation of 3H radioactivity into purine rings. From the determined values for the extent of methylation of 4S RNA isolated from infected chicken embryo cells, it was estimated that 30 to 40S RNA subunits that results from heat denaturation of the 60 to 70S RNA contain approximately 21 methyl groups, of which 14 to 16 are present at internal positions as N6 -methyladenosine residues. In addition, each of the virion RNA subunits appears to contain about two methyl groups in the "capped" 5' -terminal structure m7G(5')ppp(5') gm. These properties are consistent with the hypothesis that the 30 to 40S genome RNA os oncornaviruses also serves an mRNA function in infected cells.     

Simian virus 40-rabbit beta-globin recombinants lacking late mRNA splice sites express cytoplasmic RNAs with altered structures.

Deletions were introduced at exon-intron boundaries in the late region of a simian virus 40-beta-globin cDNA recombinant to study the role of splicing in the formation of simian virus 40 late cytoplasmic RNAs. The recombinant was used as a wild type because it allowed characterization of mutant RNAs expressed from defective genomes in the presence of comparable RNAs contributed by the coinfecting helper virus. Removal of a 17-base pair segment at map position 0.76, which included a portion of the leader sequence implicated in the splicing of the major 16S mRNA, prevented expression of 16S-type mRNA. The same mutant accumulated cytoplasmic 19S-type mRNA, but the assortment of the 5' ends of these mRNAs differed from the assortment of the wild-type counterparts. Another mutant that lacks nucleotide sequences implicated in the splicing of the major 16S mRNA and one of the principal 19S-type RNAs accumulated a 16S-type mRNA with a previously undetected leader splice, and assortment of 19S mRNAs with new or normally underrepresented splices, and even a species of unspliced cytoplasmic 19S mRNA.