Immune responses to modified-live and recombinant African horse sickness virus vaccines

Abstract

There have been numerous reports of vaccinated horses that contract fatal African horse sickness due to African horse sickness virus (AHSV) serotypes that are included in the current commercial vaccine used in southern Africa, which emphasizes the importance of thorough characterization of the equine immune response to AHSV. In particular, there are concerns about possible interference between vaccine strains in the polyvalent vaccine which led us to hypothesise that the administration of individual AHSV serotypes could induce a better immunity to individual serotypes than that achieved with the current polyvalent vaccines. There is also little published information describing the half-life of maternally derived neutralising antibody in foals to the nine AHSV serotypes. This is important for revising and developing vaccination protocols for foals of vaccinated mares. Given the lethality of both natural and experimental AHSV infections in horses, several aspects of immunity induced by different types of AHSV vaccines were evaluated. The neutralising antibody response of horses (foals) immunized with a commercial modified live virus (MLV) AHSV vaccine was evaluated and compared to the immune response elicited to monovalent MLV AHSV serotypes. Foals were immunized with either the polyvalent AHSV vaccine, or one of four monovalent vaccines containing individual AHSV serotypes 1, 4, 7 and 8. There were marked differences in the immunogenicity of individual virus serotypes contained in the vaccine. Foals more consistently seroconverted to AHSV 1 and responses to other serotypes were highly variable, and often weak or not detected. The serotype-specific responses of foals given the monovalent MLV vaccines were similar to those of foals given the polyvalent preparation suggesting that there is no apparent enhanced immune response through the administration of a monovalent vaccine as opposed to the polyvalent vaccine. Furthermore, the immunogenicity of individual AHSV serotypes contained in the commercial MLV vaccine varies remarkably. Neutralising antibody titres to the 9 known serotypes of AHSV were determined in a cohort of brood mares that were regularly vaccinated with the MLV AHSV vaccine, and the passive transfer and rate of decay of maternal antibody to the individual virus serotypes in their foals were measured. Similar to the data obtained from immunized foals, there was marked variation in the neutralising antibody response of the mares to individual AHSV serotypes even after repeated vaccination. This was mirrored in the duration of maternally-derived antibodies in their respective foals. In an effort to further characterize cellular immune (CMI) responses to AHSV, the immunity in horses induced by an experimental canarypox virus vectored recombinant (ALVAC®-AHSV4) vaccine was characterised. The detection of VP2/VP5 specific IFN-γ responses was assessed by enzyme-linked immune spot (ELISpot) assay and clearly demonstrated that all ALVAC®-AHSV4 vaccinated horses developed significant IFN-γ production compared to unvaccinated horses. Flow cytometry demonstrated that this vaccine induced mainly CD8+ T-cells, able to recognize multiple T-cell epitopes throughout all of VP2 and only the N-terminus portion of VP5. In summary, the antibody and cellular response of horses to different AHSV vaccines was evaluated and compared. The results are relevant to the design of more efficacious AHSV vaccines and to identification of protective immunity in horses to this virus.