Prolonged exposure to the candidate microbicide C31G differentially reduces cellular sensitivity to agent re-exposure.

Comparative assays of in vitro cytotoxicity using nonoxynol-9 (N-9) and the candidate microbicides C31G and sodium dodecyl sulfate (SDS) demonstrated that these agents, which are, respectively, characterized as nonionic, amphoteric, and anionic surfactants, differed in their concentration-dependent effects on cell viability, especially after prolonged exposure. We hypothesized that differences in cellular sensitivity may have been due, in part, to cellular changes induced by long-term exposure to each agent. To examine this possibility, HeLa cells were exposed to N-9, C31G, or SDS for extended periods of time and subsequently reassessed for sensitivity to each of these agents. Following 10 continuous days of C31G exposure, HeLa cells were less sensitive to a subsequent C31G exposure compared to cells that had not undergone long-term C31G treatment. Interestingly, long-term C31G exposure also changed subsequent sensitivity to N-9 but not SDS. In contrast, prolonged exposure to either N-9 or SDS did not reduce sensitivity to re-exposure. The effect of long-term C31G exposure was both concentration-dependent and transient, as treated cells reverted to pre-exposure sensitivity in a time-dependent manner following the cessation of C31G exposure. Lipid analyses of cells exposed to C31G for extended durations revealed altered phospholipid profiles relative to C31G-naïve cells. Experiments examining the individual components of C31G demonstrated the involvement of the amine oxide moiety in reductions in cellular sensitivity. These studies, which provide new information concerning the cytotoxicity of surfactant microbicides, suggest that cervicovaginal epithelial cells may have greater in vivo tolerance for products containing C31G through unique interactions between C31G and components of the cellular membranes.

Toxicity, inflammation, and anti-human immunodeficiency virus type 1 activity following exposure to chemical moieties of C31G.

C31G, which has potent activity against the human immunodeficiency virus type 1 (HIV-1) and an established record of safety in animal studies and human trials, is a microbicidal agent comprised of a buffered equimolar mixture of two amphoteric, surface-active agents: an alkyl amine oxide (C14AO) and an alkyl betaine (C16B). Studies of long-term in vitro exposure to C31G and its constituents have suggested that the components of C31G may contribute differentially to its toxicity and efficacy. In the present studies, in vitro assays of cytotoxicity and anti-HIV-1 activity demonstrated that C16B was slightly less cytotoxic compared to either C31G or C14AO, whereas the anti-HIV-1 activities of C31G and its individual constituents were similar. In the murine model of cervicovaginal microbicide toxicity, in vivo exposure to C14AO resulted in severe cervical inflammation followed by a delayed disruption of the columnar epithelium. In contrast, exposure to C16B caused severe cervical epithelial disruption and a secondary, less intense inflammatory response. These results demonstrate that (i) there are both mechanistic and temporal differences in toxicity associated with the components of C31G not necessarily predicted by in vitro assessments of cytotoxicity and (ii) contributions of each component to the anti-HIV-1 activity of C31G appear to be equal. In addition, these findings indicate that direct and indirect mechanisms of in vivo toxicity can be observed as separate but interrelated events. These results provide further insight into the activity of C31G, as well as mechanisms potentially associated with microbicide toxicity.

Magnetic resonance imaging to determine the distribution of a vaginal gel: before, during, and after both simulated and real intercourse.

To provide effective contraception and protection against sexually transmitted disease, vaginal gels should maximally cover the cervical os and the vaginal epithelium before, during and after intercourse. To non-invasively monitor the intravaginal distribution of an applied intravaginal gel, we performed high-resolution magnetic resonance imaging (MRI) of the female pelvis before, during and after both real and simulated sexual intercourse. We sought to determine whether simulated intercourse with a plastic phallus could be used as a surrogate for real intercourse for such experiments. Dilute gadolinium chelate solution was mixed with Gynol-II gel and introduced intravaginally to volunteer female human subjects using a conventional applicator. MRI was performed at 1.5 Tesla with a surface coil. Imaging of the female pelvis was performed: (1) immediately after insertion of the gel; (2) during real intercourse with a male partner (2 subjects) or simulated intercourse with a plastic phallus (4 subjects); and (3) after completion of real or simulated intercourse. Subjects were studied after application of both 3 mL and 5 mL of vaginal gel. Measurements of gel thickness covering the vaginal mucosa were made digitally using electronic calipers. The bolus of gel is initially located in the upper vaginal canal, superior to the urogenital diaphragm. Both real and simulated intercourse dramatically increases the spread of gel to the lower vagina. The cervix appears to be adequately covered with gel both before and after intercourse. Increasing the volume of the gel increases initial vaginal mucosal coverage but also increases leakage from the introitus. No statistically significant differences in vaginal mucosal coverage were found between patients having undergone real vs. simulated intercourse, or on post-intercourse scans of 3 mL versus 5 mL. MRI is a sensitive, reproducible means of tracking the spread of intravaginal medications.