Method of specifying vaccine components for viral quasispecies

Many RNA viruses do not have a single, representative genome but instead form a "quasispecies"-a set of related viral variants that coexist in field populations and even within single infected individuals (reviewed in Holland, et al. 1992 Curr Top Microbiol Immunol 176:1-20, Smith, et al. 1997 J Gen Virol 78:1511-1519, Domingo, et al. 1985 Gene 40:1-8, Domingo, et al. 1995 Molecular Basis of Virus Evolution 181-191, Duarte, et al. 1994 Infect Agents Dis 3:201-214). The emergence of immunologically distinct members of a viral quasispecies through mutation and subsequent immune selection is called "antigenic drift." Antigenic drift is thought to be important in HIV infection and the continuing seasonal influenza epidemics,especially because immunity generated against one viral variant rapidly selects for escape variants. Attributed to antigenic drift are the moderately high failure rate and the short-lived efficacy of influenza vaccines (Wilson and Cox 1990 Annu Rev Immunol 8:737-771), the failure of synthetic foot-and-mouth disease virus vaccines (Taboga, et al. 1997 J Virol 71:2606-2614), and the current failure of recombinant HIV vaccines to provide complete protection against field strains of the virus (Berman, et al. 1997 J Inf Dis 176:384-397). An algorithm for determining the viral antigenic protein variants to be used to construct vaccines designed to immunize against variable viral populations (quasispecies) is described. The method entails analyzing multiple nucleotide sequences of viral proteins and identifying those variants that provide selective advantage to the virus. Examples are given for influenza A hemagglutinin 3 and HIV-1 gp120.