Gene Regulation and Quality Control in Murine Polyomavirus Infection

Abstract

Murine polyomavirus (MPyV) infects mouse cells and is highly oncogenic in immunocompromised hosts and in other rodents. Its genome is a small, circular DNA molecule of just over 5000 base pairs and it encodes only seven polypeptides. While seemingly simply organized, this virus has adopted an unusual genome structure and some unusual uses of cellular quality control pathways that, together, allow an amazingly complex and varied pattern of gene regulation. In this review we discuss how MPyV leverages these various pathways to control its life cycle.

Keywords: RNA decay; RNA editing; nuclear retention; quality control; transcription.

Figure 1

The murine polyomavirus (MPyV) genome. The genome shown is of strain NG59RA, which is 5327 base pairs in length. Early genes are in blue and late genes are in red. Transcripts are lines, with thicker regions denoting open reading frames and dotted lines introns. The replication origin and transcriptional control region is shown at the top of the genome. Late-strand transcripts can give rise to two miRNAs (small red line) that map to the early region and can influence viral and host gene expression.

Figure 2

A crowded arrangement at the ends of the early and late genes. The sequence shown is of the 3′-region of the early and late transcription units, with the early coding strand on top and the stop codons for large T antigen and the virion structural protein VP1 circled. Note the overlap of the polyadenylation signals, including the canonical AATAAA elements (yellow box). Cleavage and polyadenylation occur downstream of these elements, leading to early and late mRNAs that have the potential to overlap for at least 45 base pairs (bp) at their 3′-ends. Transcript overlap is essential for the viral life cycle.

Figure 3

The early–late switch is associated with poly(A) signal readthrough and double-stranded RNA (dsRNA) formation. See text for details of the regulation. At early times and before viral DNA replication (left), transcription occurs from both the early and late promoters. Early-strand RNAs are spliced to produce mRNAs for the early proteins. Late-strand transcripts are efficiently terminated and polyadenylated, but are unstable and produce only small amounts of late mRNAs and proteins. After the onset of DNA replication (right), transcription termination and polyadenylation become less efficient, allowing multigenomic transcripts to be produced. Giant transcripts are efficiently spliced to generate stable late mRNAs, but sequences antisense to early-strand transcripts can downregulate early genes.

Figure 4

Late pre-mRNA splicing. Giant transcripts serve as precursors to late mRNAs. Processing of VP1 mRNA is shown. In multigenomic transcripts, leader (L) exons splice to one another, removing genome-length introns. Then, a leader-body splice can occur, coincident with polyadenylation. This results in mRNAs with tandem non-coding late leader exons at their 5′-ends.

Figure 5

The late leader exon has regions of complementarity to 18S rRNA. While the biological consequence of this still remains unclear, there is striking complementarity to ribosomal RNA at two positions within the leader. We speculate that in late mRNAs containing tandem leaders in their 5′-untranslated regions (UTRs), these elements may serve to enhance the translation of late proteins.

Figure 6

Early-strand transcription start sites shift after the onset of viral DNA replication. (A) An expanded view of the intergenic region is shown, along with genome browser tracks showing the alignment of early-strand RNAs at several times after infection, as reported by us recently [20]. These data were confirmed using the 5′-rapid amplification of cDNA ends (RACE) analysis [20]. Note the dramatic shift from 5′-ends mapping to a specific site at early times (EE) to many upstream sites at later times (LE); (B) The intergenic region is depicted, along with a general cartoon of early-strand RNAs at early times (EE) and early-strand RNAs at late times (LE). Positions of large T antigen binding are shown, along with the palindromic core replication origin, the enhancer region and the late transcription start site region. Red stars denote AUG codons that could direct translation initiation. Those in LE but not EE transcripts are frameshifted relative to the early coding region.