Pre-infection frequencies of equine herpesvirus-1 specific, cytotoxic T lymphocytes correlate with protection against abortion following experimental infection of pregnant mares

In general, vaccines containing inactivated equine herpesvirus-1 (EHV-1) fail to prevent abortion in pregnant mares following infection with a virulent strain of EHV-1. We have tested the hypothesis that resistance to EHV-1-induced abortion in pregnant mares is associated with high frequencies of EHV-1 specific, major histocompatibility complex (MHC) class I-restricted, cytotoxic T lymphocytes (CTL) in the circulation. To test this theory, three groups of pregnant mares were assembled with varying backgrounds of infection or vaccination in an attempt to mimick the immune status of the general population. Group 1 mares (n=9) were untreated controls selected at random. Group 2 mares (n=5) were vaccinated three times intramuscularly with inactivated EHV-1. Group 3 mares (n=3) had been infected with EHV-1 on four previous occasions. The frequency of CTL in blood leucocytes was measured by limiting dilution analysis at three time points; at the beginning of pregnancy (approximately 28 weeks before infection) in the Group 2 and Group 3 mares (4–7 weeks of gestation) (Group 1 was unavailable for sampling) and then 2 weeks before (30–40 weeks of gestation) and 3 weeks after experimental infection in all the mares. Serum samples were collected to monitor complement fixing (CF) antibody titres. Mares in all three groups were infected experimentally with EHV-1 strain Ab4/8 by the intranasal route after which they were monitored clinically to determine the outcome of pregnancy and samples were collected to determine the duration of nasopharyngeal shedding and cell-associated viraemia. The untreated control mares showed low pre-infection CTL. After experimental infection, they all seroconverted, aborted and demonstrated expected clinical and virological signs. Some vaccinated mares (3/5) had elevated titres of CF antibody prior to their first vaccination. All the vaccinated mares seroconverted after vaccination and exhibited higher CTL frequencies than controls before infection. Four of the five foaled normally. The multiply infected mares had low CF antibody titres prior to infection and showed neither seroconversion nor clinical or virological signs after infection. All multiply infected mares exhibited high frequencies of CTL before infection and they all foaled normally. The CTL frequencies observed differed significantly from the expected frequencies in the control and multiply infected groups at 2 weeks pre-infection (P=0.034) and between the foaling and aborting mares at 2 weeks pre-infection (P=0.005) and 3 weeks post-infection (P=0.015). The results show a positive correlation between the number of virus-specific CTL in the peripheral blood of pregnant mares and their protection against abortion induced by EHV-1 infection. Therefore, as indicated by this study, rational approaches to the development of new vaccines for EHV-1 should stimulate cytotoxic immune responses and develop virus-specific CTL as pre-requisites for protection against abortion.