Microarray assay for evaluation of the genetic stability of modified vaccinia virus Ankara B5R gene

Stability of the HSV-2 US-6 Gene in the del II, del III, CP77, and I8R-G1L Sites in Modified Vaccinia Virus Ankara After Serial Passage of Recombinant Vectors in Cells

The modified vaccinia virus Ankara (MVA), a severely attenuated strain of vaccinia virus, is a promising vector platform for viral-vectored vaccine development because of its attributes of efficient transgene expression and safety profile, among others. Thus, transgene stability in MVA is important to assure immunogenicity and efficacy. The global GC content of the MVA genome is 33%, and GC-rich sequences containing runs of C or G nucleotides have been reported to be less stable with passage of MVA vectors in cells. The production of recombinant MVA vaccines requires a number of expansion steps in cell culture, depending on production scale. We assessed the effect of extensive passage of four recombinant MVA vectors on the stability of the GC-rich herpes simplex type 2 (HSV-2) US6 gene encoding viral glycoprotein D (gD2) inserted at four different genomic sites, including the deletion (del) II and del III sites, the CP77 gene locus (MVA_009–MVA_013) and the I8R-G1L intergenic region. Our data indicate that after 35 passages, there was a reduction in gD2 expression from del II, del III and CP77 sites. Sequencing analysis implicated US6 deletion and mutational events as responsible for the loss of gD2 expression. By contrast, 85.9% of recombinant plaques expressed gD2 from the I8R-G1L site, suggesting better accommodation of transgenes in this intergenic region. Thus, the I8R-G1L intergenic region may be more useful for transgene insertion for enhanced stability.

Microarray hybridization for assessment of the genetic stability of chimeric West Nile/dengue 4 virus

Genetic stability is an important characteristic of live viral vaccines because an accumulation of mutants can cause reversion to a virulent phenotype as well as a loss of immunogenic properties. This study was aimed at evaluating the genetic stability of a live attenuated West Nile (WN) virus vaccine candidate that was generated by replacing the pre-membrane and envelope protein genes of dengue 4 virus with those from WN. Chimeric virus was serially propagated in Vero, SH-SY5Y human neuroblastoma and HeLa cells and screened for point mutations using hybridization with microarrays of overlapping oligonucleotide probes covering the entire genome. The analysis revealed several spontaneous mutations that led to amino acid changes, most of which were located in the envelope (E) and non-structural NS4A, NS4B, and NS5 proteins. Viruses passaged in Vero and SH-SY5Y cells shared two common mutations: G(2337) C (Met(457) Ile) in the E gene and A(6751) G (Lys(125) Arg) in the NS4A gene. Quantitative assessment of the contents of these mutants in viral stocks indicated that they accumulated independently with different kinetics during propagation in cell cultures. Mutant viruses grew better in Vero cells compared to the parental virus, suggesting that they have a higher fitness. When tested in newborn mice, the cell culture-passaged viruses did not exhibit increased neurovirulence. The approach described in this article could be useful for monitoring the molecular consistency and quality control of vaccine strains.

Roles of vaccinia virus genes E3L and K3L and host genes PKR and RNase L during intratracheal infection of C57BL/6 mice

The importance of the 2'-5' oligoadenylate synthetase (OAS)/RNase L and double-stranded RNA (dsRNA)-dependent protein kinase (PKR) pathways in host interferon induction resulting from virus infection in response to dsRNA has been well documented. In poxvirus infections, the interactions between the vaccinia virus (VV) genes E3L and K3L, which target RNase L and PKR, respectively, serve to prevent the induction of the dsRNA-dependent induced interferon response in cell culture. To determine the importance of these host genes in controlling VV infections, mouse single-gene knockouts of RNase L and PKR and double-knockout mice were studied following intratracheal infection with VV, VVΔK3L, or VVΔE3L. VV caused lethal disease in all mouse strains. The single-knockout animals were more susceptible than wild-type animals, while the RNase L(-/-) PKR(-/-) mice were the most susceptible. VVΔE3L infections of wild-type mice were asymptomatic, demonstrating that E3L plays a critical role in controlling the host immune response. RNase L(-/-) mice showed no disease, whereas 20% of the PKR(-/-) mice succumbed at a dose of 10(8) PFU. Lethal disease was routinely observed in RNase L(-/-) PKR(-/-) mice inoculated with 10(8) PFU of VVΔE3L, with a distinct pathology. VVΔK3L infections exhibited no differences in virulence among any of the mouse constructs, suggesting that PKR is not the exclusive target of K3L. Surprisingly, VVΔK3L did not disseminate to other tissues from the lung. Hence, the cause of death in this model is respiratory disease. These results also suggest that an unanticipated role of the K3L gene is to facilitate virus dissemination.