Differences in the Genital Microbiota in Women Who Naturally Clear Chlamydia trachomatis Infection Compared to Women Who Do Not Clear; A Pilot Study

Abstract

In vitro studies indicate IFNγ is central to Chlamydia trachomatis (Ct) eradication, but its function may be compromised by anaerobes typically associated with bacterial vaginosis (BV), a frequent co-morbidity in women with Ct. Here we investigated the associations between natural clearance of cervical Ct infection, the vaginal microbiome, and the requirements for IFNγ by evaluating the vaginal microbial and cytokine composition of Ct treatment visit samples from women who cleared Ct infection in the interim between their Ct screening and Ct treatment visit. The pilot cohort was young, predominantly African American, and characterized by a high rate of BV that was treated with metronidazole at the Ct screening visit. The rate of natural Ct clearance was 23.6% by the Ct treatment visit (median 9 days). 16S rRNA gene sequencing revealed that metronidazole-treated women who had a Lactobacillus spp.-dominant vaginal microbiota (CST 2 or 3) at the Ct treatment visit, were more prevalent in the Ct clearing population than the non-clearing population (86% v. 50%). L. iners (CST2) was the major Lactobacillus spp. present in Ct clearers, and 33% still remained anaerobe-dominant (CST1). Vaginal IFNγ levels were not significantly different in Ct clearers and non-clearers and were several logs lower than that required for killing Ct in vitro. An expanded panel of IFNγ-induced and proinflammatory cytokines and chemokines also did not reveal differences between Ct clearers and non-clearers, but, rather, suggested signatures better associated with specific CSTs. Taken together, these findings suggest that BV-associated bacteria may impede Ct clearance, but a Lactobacillus spp.-dominant microbiome is not an absolute requirement to clear. Further, IFNγ may be required at lower concentrations than in vitro modeling indicates, suggesting it may act together with other factors in vivo. Data also revealed that the vaginal bacteria-driven inflammation add complexity to the genital cytokine milieu, but changes in this microbiota may contribute to, or provide cytokine biomarkers, for a shift to Ct clearance.

Keywords: Chlamydia trachomatis; bacterial vaginosis; interferon-γ; microbiome; natural clearance; proinflammatory cytokines; women.

Figure 1

Study design. At the Ct screening visit, BV-positive was defined as an Amsel score of ≥3 of 4 by clinical assessment. All patients who were BV-Amsel positive after a full BV screen (n=20), or who were symptomatic for BV but with partial Amsel assessment (n=9), were prescribed metronidazole. All were compliant. For the Ct-treatment visit, when participants were enrolled, BV-positive was defined as a Nugent score of 7-10 by Gram stain. Abbreviations: MTZ, metronidazole; AZI, azithromycin.

Figure 2

Composition and structure of the vaginal microbiota at the Ct treatment/research enrollment visit (n=54). (A) Heatmap of the most abundant bacterial taxa identified by 16S rRNA V4 sequencing of vaginal swabs from Ct clearing and non-clearing women. Women are categorized by BV status using Amsel (0-4), Nugent (green, 0-4; yellow, 4-6; red, 7-10), and CST (1-3). Metronidazole treatment at the Ct treatment visit is noted (–, untreated; +, treated; blank, NA; blue, metronidazole; purple, Metrogel; gray, treated for BV but unknown if this was metronidazole or metronidazole gel). Ct status is + and red for Ct positive and – and blue for Ct clearer. The scale bar demonstrates relative abundance of each species. Results indicate that an optimal Lactobacillus-dominated microbiome is not necessary to clear Ct. (B) Principal coordinates analysis demonstrating clustering into three distinct CSTs. (C) Pie chart depicting the portion of Ct clearers and non-clearers that were clinically BV negative at the screening visit or their BV resolution status following metronidazole treatment; those women that did not adhere to metronidazole treatment were omitted.

Figure 3

Paired cervical and vaginal microbiome samples. Heatmap of 25 most abundant bacterial taxa identified by 16S rRNA gene V4 sequencing of vaginal or cervical swabs from Ct clearing and non-clearing women. Matched samples are representative of the three CST groups and demonstrate the similarities in relative abundance of organisms between the vaginal and cervical environments within a woman. Chlamydia trachomatis is a dominant organism in the cervical samples of Ct non-clearers.

Figure 4

Associations of Ct clearance and CSTs with genital inflammation. (A) Log₁₀-transformed IFNγ levels plotted by Ct clearance indicate clearers have slightly higher levels of IFNγ than non-clearers. (B, C) PCA performed over the log₁₀ transformed cytokine values show a large overlap between Ct clearers and non-clearers in cytokines (B). When colored by CST the PCA performed over the log transformed cytokine values shows a distinction between the Lactobacillus spp.-dominated CST2 and CST3 which associate more with IP-10, IL-6, and RANTES compared to the BV-like CST1 (C).