The relationship between vaginal microbiota composition, HPV infection status, and associated cytokine profiles and immune cell subsets.

Abstract

Studying the interplay between vaginal microbiota, HPV, cytokines, and immune cell subsets is crucial for understanding female reproductive health and the risk of cervical cancer. A healthy vaginal microbiota, primarily composed of Lactobacillus species, helps maintain an acidic environment that inhibits pathogens and supports local immune function. Disruptions in this microbiota can influence HPV susceptibility and the immune response, as specific cytokines play key roles in inflammation and immune cell recruitment. Variations in cytokine profiles can influence the clearance or persistence of HPV infections, highlighting the importance of immune cell subsets, such as dendritic cells and T cells, in responding to the virus. The aims of this PhD dissertation were to investigate the relationship between vaginal microbiota composition, HPV infection status, and the associated cytokine profiles and immune cell subsets, by evaluating (i) the epidemiology of Human Papillomavirus (HPV) infection, including its interplay with STIs and BV, and their impact on matrix metalloproteinases (MMPs) in South African women (Chapter 2); (ii) the interplay between high-risk HPV infection, mucosal immune responses, and vaginal microbiota in HIV-negative South African women who were also coinfected with STIs and/or BV and took part in the CAPRISA 083 study (Chapter 3); and the immune parameters related to the response to Metronidazole treatment, specifically analyzing changes in microbiota composition and immune cell frequencies. Methods This study involved an assessment of 267 HIV-uninfected women with laboratory-confirmed STIs, candidiasis, and/or BV within the CAPRISA 083 cohort. The participants, aged 18-40, were recruited at the Prince Cyril Zulu Communicable Diseases Clinic in Durban, South Africa, and enrolled in the CAPRISA 083 study. HIV rapid testing and point-of-care STI screening using the GeneXpert (Cepheid) and standard PCR multiplex assays were conducted at each visit. Vaginal swabs were obtained for Gram staining to diagnose candidiasis and BV using Nugent’s criteria, with scores of 0-3 considered BV negative, 4-6 intermediate, and 7-10 positive. HPV genotyping was performed using the Roche Linear Array on menstrual cup pellets. Screening for STIs and BV utilized the Xpert® CT/NG assay, while genomic DNA was extracted from vaginal swabs with the PowerSoil DNA Isolation Kit (MoBio, CA, USA). Concentrations of five matrix metalloproteinases (MMPs) and 48 cytokines were measured in menstrual cup supernatants through multiplexed ELISA. Cervical T- and dendritic-cell subsets and monocytes were measured in endocervical cyto brushes samples using multiparameter flow cytometry, and the composition of vaginal microbial communities was evaluated using 16S rRNA gene sequencing. All statistical analyses were performed using R software version 4.3.1 (www.r-project.org). To compare characteristics between HPV-negative and HPV-positive groups, Fisher's exact test and Pearson's chi-squared tests were applied for categorical variables. For continuous variables, such as age and number of sexual partners, the Wilcoxon rank-sum test was utilized. Descriptive data for continuous variables were summarized using medians and interquartile ranges, while categorical data were presented as frequencies and percentages. Results As reported in chapter 2, HPV was detected in 34% of women (83 out of 243), with a median age of 23 years (interquartile range [IQR] 21-27 years). Among those infected, low-risk (lr) HPV (71%, 59/83) and high-risk (hr) HPV (54.2%, 45/83) were common. Hr-HPV was significantly more prevalent in women with STIs and/or BV compared to those without these conditions (p = 0.029). In multivariable analysis, BV was linked to higher odds of hr-HPV detection (odds ratio [OR]: 2.64, 95% confidence interval [CI]: 1.02-6.87, p = 0.046). Additionally, strains associated with the Gardasil®9 vaccine were more commonly found in women diagnosed with STIs and/or BV (55.2%, 32/58) compared to those without (24%, 6/25; p = 0.009). Among women with STIs and/or BV, HPV detection was significantly correlated with increased concentrations of MMP-10 (b = 0.55, 95% CI 0.79-1.01; p = 0.022). In the workr eported in chapter 3, the presence of any HPV strains was inversely related to TRAIL, CTACK, MIG, and IFN-α2 concentrations, but this significance diminished for TRAIL and IFN-α2 after adjusting for age, Lactobacillus dominance, and STI presence. In addition, the multivariable analysis revealed that women with any HPV infection had significantly reduced frequencies of CCR5+ CD4+ T cells and HLADR+ CD14-CD123+ plasmacytoid dendritic cells. Oncogenic HPV specifically correlated with increased frequencies of HLADR+ CD14- CD123+ cells. In multivariate analyses adjusting for age and STIs, women with Lactobacillus dominance and any HPV infection showed higher levels of MIP-1β, RANTES, and IL-7, but lower CTACK levels compared to those without HPV. The in vivo experiments reported in Chapter 4 demonstrated that MDZ treatment modestly reduced BV-associated bacteria while rare colonization by Lactobacillus species (excluding L. iners) was observed; additionally, by weeks 6 and 12, there was a significant increase in CCR5+ CD4+ T cells and plasmacytoid dendritic cells compared to pre-treatment levels. Moreover, MDZ non-responders exhibited significantly higher frequencies of activated CD4 T cells and monocytes than responders, underscoring the impact of MDZ treatment on immune cell profiles and microbial community dyna. Conclusion In conclusion, this study highlights the complex interplay between HPV infection, BV, and immune responses in women. In Chapter 2, we observed a notable prevalence of HPV, particularly high-risk strains, in younger women, with significant associations found between hr-HPV and the presence of STIs or BV. The multivariable analysis further established BV as a significant risk factor for hr-HPV detection. Additionally, the presence of HPV was linked to increased MMP-10 concentrations among women with STIs and/or BV, indicating a potential inflammatory response. Findings from Chapter 3 reveal that HPV infections were associated with altered immune cell frequencies and cytokine profiles, suggesting that HPV may modulate local immune responses. Finally, Chapter 4’s findings on MDZ treatment indicate its potential to influence microbial communities and immune cell dynamics, particularly in relation to BV. Collectively, these results underscore the need for continued research into the interactions between HPV, bacterial communities, and immune responses to inform better clinical management strategies. Collectively, the work reported in this thesis underscores the significant associations between HPV infection, bacterial vaginosis, and immune responses, highlighting the impact of these factors on cervical health and the potential implications for clinical management strategies.