This study describes the preclinical development of a matrix-type silicone elastomer vaginal ring device designed to provide controlled release of UC781, a non-nucleoside reverse transcriptase inhibitor. Testing of both human- and macaque-sized rings in a sink condition in vitro release model demonstrated continuous UC781 release in quantities considered sufficient to maintain vaginal fluid concentrations at levels 82–860-fold higher than the in vitro IC50 (2.0 to 10.4 nM) and therefore potentially protect against mucosal transmission of HIV. The 100-mg UC781 rings were well tolerated in pig-tailed macaques, did not induce local inflammation as determined by cytokine analysis and maintained median concentrations in vaginal fluids of UC781 in the range of 0.27 to 5.18 mM during the course of the 28-day study. Analysis of residual UC781 content in rings after completion of both the in vitro release and macaque pharmacokinetic studies revealed that 57 and 5 mg of UC781 was released, respectively. The pharmacokinetic analysis of a 100-mg UC781 vaginal ring in pig-tailed macaques showed poor in vivo–in vitro correlation, attributed to the very poor solubility of UC781 in vaginal fluid and resulting in a dissolution-controlled drug release mechanism rather than the expected diffusion-controlled mechanism.

Fluctuations in susceptibility to HIV or SHIV during the menstrual cycle are currently not fully documented. To address this, time point of infection was determined in 19 adult female pigtail macaques vaginally challenged during their undisturbed menstrual cycles with repeated, low-dose SHIVSF162P3 exposures. Eighteen macaques (95%) first displayed viremia in the follicular phase, as compared to 1 macaque (5%) in the luteal phase (p<0.0001). Due to a viral eclipse phase, we estimated a window of most frequent virus transmission between days 24-31 of the menstrual cycle, in the late luteal phase. Thus, susceptibility to vaginal SHIV infection is significantly elevated in the second half of the menstrual cycle when progesterone levels are high, and when local immunity may be low. Such susceptibility windows have been postulated before but not definitively documented. Our data support findings of higher susceptibility to HIV in women during progesterone-dominated periods including pregnancy and contraceptive use.

BACKGROUND: The best current animal model for HIV infection and evaluation of antiviral compounds is the Simian-human immunodeficiency virus (SHIV)/macaque system. There are multiple recombinant SHIVs available, but these viruses have limitations in evaluating combination drug strategies for prevention. Drug combinations that target reverse transcriptase (RT, either nRTI or nnRTI) and envelope (entry or fusion inhibitors) have to be tested separately, which does not permit the assessment of additive, synergistic, or antagonistic effects of ARV combinations. We describe construction of a dual SHIV containing both HIV RT and a CCR5-specific HIV envelope gene in a simian immunodeficiency virus backbone. METHODS: The RT Env SHIV molecular clone was constructed using RT SHIV and SHIV162p3 sequences as templates to generate RT Env SHIV. RT Env SHIV was expanded in vitro in CD8-depleted macaque peripheral blood mononuclear cells (PBMC). Recombinant virus was used to infect a rhesus macaque (4.3 x 10(4) tissue culture infectious dose [TCID(50)], intravenously [IV]). A second passage in a macaque by IV transfer of 10 ml of blood obtained from the first infection was also done. The in vivo adapted virus stock from these macaques was used to produce high titer stocks in vitro and used to rectally infect an additional macaque. RESULTS: Peak viral load reached 6 x 10(5) vRNA copies/ml in plasma in both IV-exposed macaques and remained detectable in the one animal for 16 weeks after infection. A viral stock (1.68 x 10(4) TCID(50)) derived from the second macaque passage has been produced in CD8-depleted rhesus PBMC and was successfully used to demonstrate mucosal transmission. The resulting RT Env SHIV retained the sensitivity to HIV RT and entry inhibitors of its parental viruses. CONCLUSIONS: The objective of this study was to develop and characterize a SHIV recombinant virus for evaluating the efficacy of ART and microbicide products that target both HIV RT and/or Env-mediated entry. RT Env SHIV can productively infect macaques by both the IV and mucosal route, making it a valuable tool for transmission studies.

Animal models for research on susceptibility to HIV are currently not available. Here we explore whether a macaque model of repeated low-dose rectal or vaginal virus challenges could be employed. We tested the hypothesis that susceptibility to Simian HIV is not merely stochastic in this model but rather is associated with identifiable host factors. Forty macaques required a median of 3.5 SHIVSF162P3 challenges for infection. We studied the association of their susceptibility with 13 predisposing plasma cytokines/chemokines (RANTES, Eotaxin, monocyte chemoattractant protein (MCP)-1, IL-7, MIP-1beta, TNF-alpha, MIP-1alpha, granulocyte colony-stimulating factor, IL-8, interferon-gamma, IL-17, IL-1beta, IL-6). Higher plasma RANTES, IL-8, and Eotaxin were associated with lower susceptibility, that is, higher resistance to infection. In a group of macaques with low IL-8 and RANTES, a median 3 exposures were required to infect; whereas, when either IL-8 or RANTES were high, a median 12 exposures were required. Thus, susceptibility was associated with identifiable discrete host factors and was not stochastic. In addition, the macaque model identified key human resistance factors (RANTES, Eotaxin), but also revealed a novel association with resistance (IL-8). Future direct evaluation of these or other factors in the animal model may be beneficial for developing new immunomodulation strategies for HIV prevention.