Background: The potential for HLA antigens to elicit an allo-immune response deserves study, particularly as allo-immunization induces potent immune responses, and there is evidence that allo-immunization in macaques may protect against SIV. This study aims determine whether unprotected (UP) vaginal intercourse in women elicits allo-immune responses, and whether this correlates with protection from HIV infection of CD4⁺ cells.

Methods: Subjects—Serial bloods and detailed sexual history were taken from 4 HIV^- groups. Group I: monogamous partners (> 6 months [mos]) practising UP heterosexual intercourse (n = 36); Group II: females (n = 16) and males (n = 9) who had no (or protected) sexual intercourse (> 6 mos); Group III: women (n = 5) who practised UP oral sex and protected vaginal sex; and Group IV: homosexual females practicing UP oral sex. One way MLR of PBMC was performed from all subjects, stimulated with irradiated PBMC from steady partners and compared with that stimulated by PBMC from a 3^rd party. HLA class II typing was performed to ascertain whether any of the MLR reactions were likely to be weak due to fortuitous matching at HLA-DRB or –DQB1 between the responder and stimulator cells. For each volunteer, analysis of D32 CCR5 mutation was performed, flow cytometry to analysis of phenotypic expression of cell surface molecules, and in vitro infectivity of activated CD4⁺ T-cells with R5 and X4 strains of HIV-1. Statistics—Man-Whitney non-parametric for MLR, Student’s t-test for HIV infectivity.

Results: The mean (±sem) Stimulation Index (SI) in 31 females responding to male partners’ PBMC (10%) was 9.5 (±1.4) which was significantly greater than that responding to 3^rd party cells 5.4(±0.7) (p = 0.01). Males showed significant results but required a higher cell percentages. Group III showed a greater, but non-significant response than 3^rd party cells; there was no difference in female homosexuals. There was significant dose-dependent inhibition of transmission of the R5 strain of HIV-1 in females practising UP vaginal intercourse compared with their male partners (p < 0.05) and much greater than females or males practising protected sex for > 12 mos.

Conclusion: We have shown, for the first time, that mucosal allo-immunization occurs during intercourse. This suggests a physiological function that enhances immunity to sexually transmitted pathogens. This supports allo-immunization as a potential strategy for preventative and therapeutic vaccines against HIV.