To better understand genetic diversity of mammalian reoviruses, we studied sequence variability in the S3 gene segment of 17 field-isolate reovirus strains and prototype strains of the three reovirus serotypes. Strains studied were isolated over a 37-year period from different mammalian hosts and geographic locations. A high degree of variability was observed in the nucleotide sequences of the S3 gene, whereas the deduced amino acid sequences of the S3 gene product, sigma NS, were highly conserved. When variability among the S3 nucleotide sequences was analyzed using pairwise comparisons, we found that 5' and 3' noncoding regions were significantly more conserved than the remainder of the gene. This high degree of sequence conservation was also observed within the first 15 nucleotides of the 5' coding region. Phylogenetic analyses showed that multiple alleles of the S3 gene cocirculate and that genetic diversity in the S3 gene does not correlate with host species, geographic locale, or date of isolation. Phylogenetic trees constructed from variation in the S3 sequences are distinct from those previously generated from sequences that encode attachment protein sigma 1, core protein sigma 2, and outer capsid protein sigma 3, which supports the hypothesis that reovirus gene segments reassort in nature. These findings suggest that reovirus gene segments are well-adapted to mammalian hosts and that reovirus evolution has reached an equilibrium.

Mice infected with reovirus develop abnormalities in glucose homeostasis. Reovirus strain type 3 Abney (T3A) was capable of systemic infection of nonobese diabetic (NOD) mice, an experimental model of autoimmune diabetes. Reovirus antigen was detected in pancreatic islets of T3A-infected mice, and primary cultures of pancreatic islets from NOD mice supported T3A growth. Significantly fewer T3A-infected animals compared to uninfected controls developed diabetes. However, despite the alteration in diabetes penetrance, insulitis was evident in T3A-infected mice. These results suggest that viral infection of NOD mice alters autoimmune responses to ␤-cell antigens and thereby delays development of diabetes.

Reovirus mRNAs are efficiently translated within host cells despite the absence of 3 polyadenylated tails. The 3 nontranslated regions (3NTRs) of reovirus mRNAs contain sequences that exhibit a high degree of gene-segment-specific conservation. To determine whether the 3NTRs of reovirus mRNAs serve to facilitate efficient translation of viral transcripts, we used T7 RNA polymerase to express constructs engineered with full-length S4 gene cDNA or truncation mutants lacking sequences in the 3NTR. Full-length and truncated s4 mRNAs were translated using rabbit reticulocyte lysates, and translation product 3 was quantitated by phosphorimager analysis. In comparison to full-length s4 mRNA, translation of the s4 mRNA lacking the 3NTR resulted in a 20 to 50% decrease in 3 produced. Addition to translation reactions of an RNA oligonucleotide corresponding to the S4 3NTR significantly enhanced translation of full-length s4 mRNA but had no effect on s4 mRNA lacking 3NTR sequences. Translation of s4 mRNAs with smaller deletions within the 3NTR identified a discrete region capable of translational enhancement and a second region capable of translational repression. Differences in translational efficiency of full-length and deletion-mutant mRNAs were independent of RNA stability. Protein complexes in reticulocyte lysates that specifically interact with the S4 3NTR were identified by RNA mobility shift assays. RNA oligonucleotides lacking either enhancer or repressor sequences did not efficiently compete the binding of these complexes to full-length 3NTR. These results indicate that the reovirus S4 gene 3NTR contains a translational operator sequence that serves to regulate translational efficiency of the s4 mRNA. Moreover, these findings suggest that cellular proteins interact with reovirus 3NTR sequences to regulate translation of the nonpolyadenylated reovirus mRNAs.