Abstract Body

Background: Anti-microbial peptides (AMP) in vaginal fluid play a key role in vaginal mucosal immunity but dynamic effects have not been documented. Since penile-vaginal intercourse (PVI) is an important HIV transmission route, variation in AMP levels could affect HIV inhibition and likelihood of infection. This study documents levels of several AMPs – trappin, human beta-defensin 2 (HBD2), lactoferrin, secretory leukocyte protease inhibitor (SLPI), and lysozyme – at the time of PVI (compared to a neutral control for time levels) and assesses effects on in-vitro HIV inhibition. Methodology: 40 healthy women ages 18-39 years wore a vaginal tampon for 1 hour pre-PVI and again beginning 15 minutes after condom-protected PVI or during 1 hour of quiet time with a partner (without physical contact.) Tampons were frozen in 1 ml of PBS at the subjects’ homes until processing. Vaginal AMP levels were assessed by commercially available ELISA assays and normalized to total protein to control for inter-sample dilutional variation. HIV-1 BaL was incubated in 10% vaginal fluid (PVI or quiet time) and used to infect TZM-bl indicator cells. Infectivity was measured by luciferase activity and % HIV inhibition calculated by normalizing to the activity in untreated controls. A random-effects logistic regression model was used to control for multiple within-person comparisons. Results: Mean trappin levels were significantly greater at the time of PVI (1.99 mg/mg protein) versus quiet time (0.47 mg/mg protein; p=0.01) while mean levels of HBD2 (PVI = 136.36 ng/mg protein; quiet time = 265.16 ng/mg protein; p=0.04) and lactoferrin (PVI = 2.34 mg/mg protein; quiet time = 4.89 mg/mg protein; p=0.02) were significantly lower. Levels of SLPI and lysozyme did not significantly differ during PVI and quiet time. Controlling for PVI and menstrual phase, multivariable logistic regression showed that % HIV inhibition increased significantly with increased levels of HBD2 and lysozyme (p<0.001 and p=0.016, respectively). Conclusions: PVI is associated with reductions in some vaginal AMP and these changes are associated with decreases in in-vitro HIV inhibition. These data describe a potential role for intercourse-associated changes in vaginal mucosal immunity that has implications for understanding of HIV infection as well as topical HIV prevention products such as microbicides.