The vaginal microbiota of women living with HIV on suppressive antiretroviral therapy and its relation to high-risk human papillomavirus infection

Abstract

Background: Few studies have investigated the vaginal microbiota (VM) in women living with HIV (WLWH) in the context of high-risk human papillomavirus (HR-HPV) infection, even though WLWH are at an increased risk of HPV-related malignancies, including cervical cancer. To explore the impact of HIV and HPV infection on the VM in WLWH, we determined the prevalence of HR-HPV infection and cervical cytologic abnormalities in a cohort of 44 WLWH and 39 seronegative-women (SNW), characterized the vaginal microbiota by 16S sequencing, assessed genital inflammation and systemic immune activation by multiplex bead assay and flow cytometry, respectively. Finally, we explored relationships between bacterial richness and diversity, the top 20 bacterial genera, genital inflammation and systemic immune activation.

Results: We found that HR-HPV prevalence was similar between WLWH and SNW. High-grade squamous intraepithelial lesions (HSIL) were only detected in WLWH negative for HR-HPV infection. In regression analyses, no risk factors were identified. Women co-infected with HIV and HR-HPV had the highest level of systemic immune activation, and these levels were significantly different compared with SNW without HR-HPV infection. Lactobacillus iners was the dominant Lactobacillus species in WLWH and SNW alike.

Conclusion: We found no evidence of differences in vaginal microbial richness and diversity, microbial community structure, and genital inflammation by HIV, HPV, or HIV and HPV status.

Keywords: Antiretroviral therapy; Cervical cytology; Genital inflammation; High-risk HPV; Human immunodeficiency virus; Human papillomavirus; Immune activation; Lactobacillus iners; Vaginal microbiota; Vaginal pH.

Fig. 1

Plasma IL-6 and CD8 + T cell activation are highest in women living with HIV with HR-HPV infection. Legend: Scatter plot showing the median and interquartile range of A. The log₁₀ plasma IL-6 concentrations and B. The frequency of CD8 + HLADR + CD38 + (%) in seronegative women (SNW) and women living with HIV (WLWH) with or without HPV infection. Kruskal–Wallis test was used to compare groups and p values were adjusted for multiple comparisons using Dunn's multiple comparisons test. *p < 0.05 (statistical significance). Only significant adjusted p values are shown. Statistical analysis and graphs were done in GraphPad Prism 9. Abbreviations: CD: cluster of differentiation, HLA: human leukocyte antigen, HIV: human immunodeficiency virus, HPV: human papillomavirus, HPVP: HPV positive, HPVN: HPV negative, KWp = Kruskal–Wallis p, ml = milliliter, pg = picogram, SNW: seronegative women, WLWH: women living with HIV, %: frequency

Fig. 2

The vaginal microbiota of women living with HIV and seronegative women with or without HR-HPV infection. Legend: A. Boxplots showing the median and interquartile range of two alpha diversity metrics: richness (number of observed species) and shannon (richness and evenness). Groups were compared using Kruskal–Wallis test and p values adjusted for multiple comparisons (KWp = 0.933 for richness, and KWp = 0.936 for shannon). No difference in alpha diversity was observed when stratifying by HIV status, HPV status or both. B .Clustering of microbial communities was visualized using principal coordinate analysis (PCoA, Bray–Curtis dissimilarity) and differences were assessed by PERMANOVA after testing for homogeneity of dispersions (betadisper). Pairwise comparisons revealed that microbial communities of WLWH with HPV infection (WLWH HPVP) were marginally different from those of WLWH without HPV infection (WLWH HPVN, R² = 0.049, p = 0.047), with no other difference being observed (SNW HPVN vs SNW HPVP, R² = 0.024, p = 0.472; WLWH HPVN vs SNW HPVN R² = 0.026, p = 0.162; WLWH HPVP vs SNW HPVP, R² = 0.029, p = 0.678). C. Heatmap showing the top 20 genera stratified by HIV and HPV status. The heatmap was generated using ampvis2 in R. Relative abundances were centered log-ratio transformed. Abbreviations: HIV: human immunodeficiency virus, HPV: human papillomavirus, HPVN: HPV negative, HPVP: HPV negative, PCoA: principal coordinate analysis, PERMANOVA: Permutational multivariate analysis of variance, R² = R squared, SNW: seronegative women, WLWH: women living with HIV

Fig. 3

Differentially expressed taxa in women living with HIV and seronegative women with or without HPV infection as determined by Linear Discriminant Analysis Effect Size (LEfSe). Legend: Linear discriminant analysis plots showing differentially discriminant taxa identified in A. WLWH and SNW irrespective of HPV infection; B. HPVN and HPVP women irrespective of HIV status; C. WLWH only; and D. HPVN women only. LEfSe analyses were performed at https://huttenhower.sph.harvard.edu/galaxy/, with the following parameters: LDA threshold 3 or higher, alpha value for factorial Kruskal–Wallis test 0.05 and alpha value for pairwise Wilcoxon test 0.05. The ASV table was filtered prior to LEfSe analysis, removing ASVs not seen more than 5 times in at least 5% of samples, to avoid including ASVs with small means and outliers. Abbreviations: ASVs: amplicon sequences variants, LDA: Linear Discriminant Analysis, HIV: human immunodeficiency virus, HPV: human papillomavirus, HPVP: HPV positive, HPVN: HPV negative, SNW: seronegative women, WLWH: women living with HIV