Characterization of Candidate Mutations for Use in a Live Attenuated West Nile Virus Vaccine
West Nile virus (WNV) is a mosquito-borne flavivirus that causes neurological disease and fatalities annually in the United States and Europe, but no human WNV vaccines have been licensed. Studies of WNV vaccine candidates and of vaccines for related flaviviruses indicate that a live attenuated vaccine based on the full-length WNV genome will likely induce the most robust protective immune response compared to other vaccination strategies. This dissertation describes investigation of potentially attenuating mutations in WNV to aid in candidate vaccine development. Single gene mutations were investigated in the envelope (E), NS4B, and NS5 proteins. Additionally, multigenic mutants were investigated that combined mutations in the NS1 protein with mutations in the E or NS4B proteins, as a vaccine with multigenic mutations should have fewer chances of reversion to virulence. The E mutation investigated, E138K, did not have a mouse attenuated phenotype, and virulence was associated with reversion to wild-type genotype. Of seven NS4B mutations investigated, only P54A and P54G mutants were significantly attenuated in mice. While the P54G mutant was genotypically stable, P54A was capable of reversion. In the NS5 protein, two single mutants were studied independently (K61A and E218A) and in combination as a double mutant. The NS5 single mutants were attenuated in mice, however, the double mutant had reduced attenuation associated with reversion at both mutated residues. For the multigenic mutants, strongly attenuating NS1 glycosylation site mutations (NS1mut) were combined with NS4B-C102S, NS4B-W103Y, or E-E138K. Although NS4B-C102S and NS4B-W103Y mutations did not independently attenuate WNV mouse virulence, the multigenic mutants retained the attenuated phenotype of the NS1mut. In comparison, the NS1mut+E-E138K mutant was not attenuated, and instead, the genotype showed evidence of reversion. Overall, NS4B-P54G and NS5-K61A mutations were the most strongly attenuated and stable mutations investigated, and thus, they should be considered during WNV vaccine development. Additionally, the NS1mut glycosylation site mutations should be stabilized with additional amino acid substitutions to continue to use this mutant in vaccine design. Overall, this dissertation describes the attenuating phenotype and stability of mutations in diverse regions of the WNV genome and provides insight into WNV rational vaccine development strategies.