Association between statin use, the vaginal microbiome, and Gardnerella vaginalis vaginolysin-mediated cytotoxicity

Abstract

Background: Bacterial vaginosis (BV) is the leading dysbiosis of the vaginal microbiome. The pathways leading towards the development of BV are not well understood. Gardnerella vaginalis is frequently associated with BV. G. vaginalis produces the cholesterol-dependent cytolysin (CDC), vaginolysin, which can lyse a variety of human cells and is thought to play a role in pathogenesis. Because membrane cholesterol is required for vaginolysin to function, and because HMG-CoA reductase inhibitors (statins) affect not only serum levels of cholesterol but membrane levels as well, we hypothesized that statins might affect the vaginal microbiome.

Methods: To investigate the relationship between use of the statins and the vaginal microbiome, we analyzed 16S rRNA gene taxonomic surveys performed on vaginal samples from 133 women who participated in the Vaginal Human Microbiome Project and who were taking statins at the time of sampling, 152 women who reported high cholesterol levels but were not taking statins, and 316 women who did not report high cholesterol. To examine the effect of statins on the cytolytic effect of vaginolysin, the cholesterol-dependent cytolysin (CDC) produced by Gardnerella vaginalis, we assessed the effect of simvastatin pretreatment of VK2E6/E7 vaginal epithelial cells on vaginolysin-mediated cytotoxicity.

Results: The mean proportion of G. vaginalis among women taking statins was significantly lower relative to women not using statins. Women using statins had higher mean proportions of Lactobacillus crispatus relative to women with normal cholesterol levels, and higher levels of Lactobacillus jensenii relative to women with high cholesterol but not taking statins. In vitro, vaginal epithelial cells pretreated with simvastatin were relatively resistant to vaginolysin and this effect was inhibited by cholesterol.

Conclusions: In this cross-sectional study, statin use was associated with reduced proportions of G. vaginalis and greater proportions of beneficial lactobacilli within the vaginal microbiome. The negative association between statin use and G. vaginalis may be related to inhibition of vaginolysin function.

Fig 1. Microbial community profiles of women grouped by statin use.

Stacked bar plots showing vaginal microbial community profiles from 316 women who did not report high cholesterol and who were not taking statins, 152 women reported having high cholesterol but who were taking statins, and 133 women who were taking statins. The profiles are grouped by the most abundant species and are ordered by decreasing proportion of the dominant bacterium.

Fig 2. The proportion of G. vaginalis is lower in statin users.

A. Taxa that occurred in significantly different proportions in the vaginal microbiomes of statin users were detected by LEfSE analysis. Taxa significantly higher in women taking statins are in green and the species significantly lower (G. vaginalis) is in red. The chart on top compares vaginal microbiota from women with high cholesterol who were not taking statins to women taking statins and the lower chart compares vaginal microbiota from women with normal cholesterol who were not taking statins to women taking statins B. Boxplot of G.vaginalis proportions in subjects grouped based on ethnicity and sub-grouped based on statin use and normal versus high cholesterol, with whiskers that extend to the highest/lowest value within 1.5 times the interquartile range, outliers beyond the whiskers are plotted as points. The horizontal line in each box indicates the median. A Wilcoxon rank sum test with continuity correction was used to test whether the proportion of BV-associated bacteria followed the same distribution for groups of subjects (statin/high cholesterol no statin/no high cholesterol, African/European ancestry).

Fig 3. Statins reduce vaginolysin-mediated cytotoxicity.

A) VK2/E6E7 vaginal keratinocytes were incubated in control media or media containing 5 μg cholesterol / ml (Chol), 1 μg simvastatin / ml (Statin), simvastatin and cholesterol, or simvastatin and 1mM mevalonate for 48 hours. The cells were then left unchallenged or challenged with 10 μg VLY / mL for 1 hour. Trypan blue staining was performed to monitor rounding and permeabilization (observed as central darkening of the cells). B) VK2/E6E7 cells pre-treated with control, simvastatin, cholesterol, simvastatin and cholesterol, or simvastatin and mevalonate, and then challenged with 10 or 5 μg VLY / mL. Lactate dehydrogenase release assay was used to quantify permeabilization of the cells. * p<0.05, ** p<0.005 using one-way ANOVA with Tukey post-test for comparison of individual groups.