- Volume 86, Issue 9, 2005

Persistent infections of the uterine cervix with ‘high-risk’ human papillomavirus (HPV) are now recognized as necessary for the development of cervical cancer. Among them, HPV types 16 and 18 exhibit numerous variants associated with different risks for cervical cancer development. In this study, the questions of whether different HPV type 18 variants exhibit changes in early gene transcription and the molecular mechanisms underlying these differences were investigated. It was shown that, indeed, type 18 variants exhibited singular differences in E6 transcripts in vivo. Higher levels of the E6*I transcript were detected regularly in clones harbouring the African variant, as opposed to low levels of this transcript detected in clones containing the reference clone (Asian–Amerindian), where significantly higher levels of full-length E6 transcript were usually observed. As a direct consequence, higher levels of p53 protein were found in the presence of African E6, as opposed to the low levels of p53 observed with the Asian–Amerindian E6. These variations in consequence affected the levels of cellular proteins regulated by p53, such as Bax. Similar changes in the relative levels of E6 transcripts were observed when tumours containing type 18 E6 variants were analysed. The different ability of cells containing variant E6 genes to form tumours in nude mice was suggested by the fact that tumour volumes were considerably higher when cells expressed the Asian–Amerindian E6. Mutagenesis analysis of the reference clone showed that a C491A change reverts the phenotype. These results suggest that different splicing patterns of E6 within HPV type 18 variants may possibly have biological implications in viral tumorigenesis.

Multiple synonymous substitution mutations in the Wheat streak mosaic virus P3 cistron did not affect translation in vitro but rendered the virus incapable of systemic infection. Multiple synonymous substitutions in the cylindrical inclusion cistron did not alter infectivity or in vitro translation. To assess replication and movement phenotypes, P3 mutations were placed in context with a GUS reporter gene. GUS activity measured in barley protoplasts 36 h post-transfection indicated that mutants with synonymous substitutions in P3 retained the ability to replicate at 22–80 % of wild-type levels. Almost no GUS activity was detected in protoplasts transfected with a P3 frame-shift mutant. Histochemical GUS assays conducted 3 days post-inoculation (p.i.) revealed genomes with multiple synonymous substitutions in P3, which were able to establish infection foci limited to small clusters of cells that increased in size only slightly by 5 days p.i. Infection foci produced by wild-type Wheat streak mosaic virus-expressing GUS were much larger at 3 days p.i. and had coalesced by 5 days p.i. No GUS activity was detected in plants inoculated with the frame-shift mutant bearing GUS. Three of four mutants, each with a single synonymous substitution in the 3′-proximal half of the P3 cistron, were wild-type with respect to systemic infectivity. A model RNA secondary structure obtained for the region was disrupted by the debilitating single mutation but not by the other three single mutations. Collectively, these results identify an internal RNA sequence element in the P3 cistron that affects both replication and movement of the viral genome.

Microarrays derived from Solanum tuberosum expressed sequence tags were used to test the hypothesis that genetically distinct enveloped viruses elicit unique changes in Nicotiana benthamiana gene expression. The results of our study, which included Sonchus yellow net virus (SYNV), a plant rhabdovirus that replicates in the nucleus of infected cells, and Impatiens necrotic spot virus (INSV), a plant bunyavirus that replicates in the cytoplasm, were consistent with this hypothesis. Statistically significant changes (P⩽0·01) in the expression of 275, 2646 and 4165 genes were detected in response to INSV at 2, 4 and 5 days post-inoculation (d.p.i.), respectively. In contrast, 35, 665 and 1458 genes were expressed differentially in response to SYNV at 5, 11 and 14 d.p.i., respectively. The microarray results were verified by Northern hybridization using a subset of these genes as probes. Notably, INSV, but not SYNV, induced expression of small heat-shock protein genes to high levels. In contrast to SYNV, infection by INSV resulted in downregulation of all histone genes, of which the downregulation of histone 2b expression to very low levels was confirmed by Northern hybridization. The expression of a putative WRKY transcription factor at 11 d.p.i., but not at 5 or 14 d.p.i., in SYNV-infected tissue suggested that the temporal response to virus infection was identified readily using our experimental design. Overall, infection by INSV resulted in larger fold changes in host gene expression relative to infection by SYNV. Taken together, the present data demonstrate differential responses of a common host to two genetically distinct viruses.

Prion diseases involve conversion of host-encoded cellular prion protein (PrPC) to a disease-related isoform (PrPSc). Using recombinant human β-PrP, a panel of monoclonal antibodies was produced that efficiently immunoprecipitated native PrPSc and recognized epitopes between residues 93–105, indicating for the first time that this region is exposed in both human vCJD and mouse RML prions. In contrast, monoclonal antibodies raised to human α-PrP were more efficient in immunoprecipitating PrPC than PrPSc, and some of them could also distinguish between different PrP glycoforms. Using these monoclonal antibodies, the physical association of PrP glycoforms was studied in normal brain and in the brains of humans and mice with prion disease. It was shown that while PrPC glycoforms can be selectively immunoprecipitated, the differentially glycosylated molecules of native PrPSc are closely associated and always immunoprecipitate together. Furthermore, the ratio of glycoforms comprising immunoprecipitated native PrPSc from diverse prion strains was similar to those observed on denaturing Western blots. These studies are consistent with the view that the proportion of each glycoform incorporated into PrPSc is probably controlled in a strain-specific manner and that each PrPSc particle contains a mixture of glycoforms.