OPTIMIZATION OF AN IN VITRO MODEL OF BIOFILM FORMATION ON VAGINAL EPITHELIAL CELLS TO TEST STRATEGIES FOR PROTECTION AGAINST BACTERIAL VAGINOSIS

Abstract

Background: The composition of the vaginal microbiota (VMB) in the female genital tract (FGT) can impact the vaginal epithelium and protect against or increase risk of sexually transmitted viral infections. The VMB grows as a biofilm, a complex structure formed by bacteria for increased survival. When the VMB consists of a diverse bacterial community it correlates with pathogenic effects that lead to adverse health conditions and an increased risk of HIV infection. When the VMB contains Lactobacillus species, beneficial health effects and decreased susceptibility to infection are observed. The aim of this project is to optimize an in vitro model of biofilm formation for different bacteria associated with the VMB, identify the effects that biofilm has on vaginal epithelial cells and test biofilm treatment strategies. We hypothesize that a Lactobacillus biofilm will enhance barrier function and decrease cytotoxicity of vaginal epithelial cells whereas dysbiotic biofilm will decrease barrier function and induce cytotoxicity. We also hypothesize that various conditions, such as presence of estradiol and eubiotic short-chain fatty acids, will stimulate Lactobacillus biofilm growth and suppress dysbiotic biofilm growth in a vaginal epithelial cell model. Methods: For optimization of the biofilm model, VK2/E6E7 cells were grown in air-liquid interface (ALI) or liquid-liquid interface (LLI) cultures in presence or absence of L. crispatus, L. iners, G. vaginalis or P. bivia bacteria. Biofilm formation was assessed using FilmTracerTM SYPRO® Ruby biofilm matrix protein stain. Hormone effects were tested by adding estradiol (10-9 M) and progesterone (10-7 M) to culture media. Short-chain fatty acid (SCFA) effects were tested by adding lactic acid, acetic acid, succinic acid and butyric acid in varying concentrations to culture media. Enzyme effects were tested by adding sialidase to Vk2 cells before bacteria inoculation. Results: A novel in vitro model of biofilm formation on vaginal epithelial cells was created. Vk2 cells in ALI and LLI cultures remained viable in anaerobic conditions and showed mucin-1 production in aerobic and anaerobic conditions. Matrix protein staining provided a means to accurately visualize and quantify biofilm formation in this model. L. crispatus and L. iners biofilm growth maintained vaginal epithelial barrier integrity without cytotoxicity. G. vaginalis and P. bivia biofilm growth significantly reduced barrier integrity (p=0.0166, p=0.0115) and increased cytotoxicity (p=0.0024, p<0.0001). Estradiol significantly increased the growth of L. crispatus biofilm in the co-culture system (p<0.0001). Progesterone significantly increased G. vaginalis biofilm growth in the Vk2 cell co-culture (p=0.006). L. crispatus biofilm formation in the estradiol condition, G. vaginalis biofilm formation in the progesterone condition and P. bivia biofilm growth in the normal media condition were significantly decreased in the presence of sialidase (p<0.0001, p=0.0001, p=0.0380). Conclusion: A novel in vitro model of biofilm formation on a vaginal epithelial cell line that can be used to visualize and quantify biofilm growth was generated. This model was used to test various strategies for biofilm enhancement or dissociation. Estradiol enhanced beneficial Lactobacillus biofilm growth, while progesterone enhanced dysbiotic biofilm growth. Mucin- digesting enzyme sialidase was effective at dissociating all biofilms. This model can be used in the future to test different strategies of dysbiotic biofilm dissociation and enhancement of Lactobacillus biofilm in order to investigate treatments for Bacterial Vaginosis (BV) and reduce susceptibility to HIV transmission in women.