Bacterial Vaginosis Is Associated with Loss of Gamma Delta T Cells in the Female Reproductive Tract in Women in the Miami Women Interagency HIV Study (WIHS): A Cross Sectional Study

Abstract

Bacterial vaginosis (BV) is the most common female reproductive tract infection and is associated with an increased risk of acquiring and transmitting HIV by a mechanism that is not well understood. Gamma delta (GD) T cells are essential components of the adaptive and innate immune system, are present in the female reproductive tract, and play an important role in epithelial barrier protection. GD1 cells predominate in the mucosal tissue and are important in maintaining mucosal integrity. GD2 cells predominate in peripheral blood and play a role in humoral immunity and in the immune response to pathogens. HIV infection is associated with changes in GD T cells frequencies in the periphery and in the female reproductive tract. The objective of this study is to evaluate if changes in vaginal flora occurring with BV are associated with changes in endocervical GD T cell responses, which could account for increased susceptibility to HIV. Seventeen HIV-infected (HIV+) and 17 HIV-uninfected (HIV-) pre-menopausal women underwent collection of vaginal swabs and endocervical cytobrushes. Vaginal flora was assessed using the Nugent score. GD T cells were assessed in cytobrush samples by flow cytometry. Median Nugent score was 5.0 and 41% of women had abnormal vaginal flora. In HIV uninfected women there was a negative correlation between Nugent score and cervical GD1 T cells (b for interaction = - 0.176, p<0.01); cervical GD1 T cells were higher in women with normal vaginal flora than in those with abnormal flora (45.00% vs 9.95%, p = 0.005); and cervical GD2 T cells were higher in women with abnormal flora than in those with normal flora (1.70% vs 0.35%, p = 0.023). GD T cells in the genital tract are protective (GD1) and are targets for HIV entry (GD2). The decrease in cervical GD1 and increase in GD2 T cells among women with abnormal vaginal flora predisposes women with BV to HIV acquisition. We propose to use GD T cell as markers of female genital tract vulnerability to HIV.

Fig 1. Comparison of percentages of endocervical GD1 T cells in HIV infected and uninfected women with normal and abnormal vaginal flora.

VF: vaginal flora. Normal VF is defined as Nugent score less than 4. Abnormal VF is defined as Nugent score of 4 or greater than 4. In HIV- women, frequency of GD1 T cells was higher in women with normal vaginal flora than in those with abnormal vaginal flora (45.00 vs 9.95, p = 0.005). Results from a 2 (HIV+, HIV-) x 2 (normal VF, abnormal VF) ANOVA showed that there was a nonsignificant main effect of abnormal vaginal flora, F(1,33) = 2.12, p = 0.155, but a statistically significant main effect of HIV status, F(1,33) = 10.34, p = 0.003. The interaction of BV and HIV status was statistically significant, F(1,33) = 7.05, p = 0.013. Further analysis of the interaction effect of BV and HIV status using pairwise comparisons with a Bonferroni correction showed that there was a statistically significant difference in GD1 T cells among HIV- women with abnormal vaginal flora compared to those without abnormal vaginal flora (p = 0.005), but not among HIV+ women (p = 0.424).

Fig 2. Interaction of Nugent score and HIV status predicting GD1 T cells.

Results of the regression model predicting endocervical GD1 T cells by HIV status, Nugent score, and the interaction of HIV status and Nugent score showed that for HIV- participants, GD1 T cells decreased as Nugent score increased, but for HIV+ women, as Nugent score increased, GD1 T cells decreased (b for interaction = 1.197, p < 0.001 for HIV- women)

Fig 3. Comparison of frequencies of endocervical GD2 T cells in HIV infected and uninfected women with normal and abnormal vaginal flora.

VF: vaginal flora. Normal VF is defined as Nugent score less than 4. Abnormal VF is defined as Nugent score of 4 or greater than 4. In HIV- women, frequency of GD2 T cells was higher in women with abnormal vaginal flora than in those with normal vaginal flora (1.79 vs 0.55, p = 0.023). A 2 (HIV+, HIV-) x 2 (normal VF, abnormal VF) ANOVA indicated that there was a nonsignificant main effect of vaginal flora, F(1,22) = .03, p = .876, but a nonsignificant main effect of HIV status, F(1,22) = .43, p = .520. The interaction of vaginal flora and HIV status was statistically significant, F(1,22) = 5.67, p = .028. The interaction effect was further analyzed using pairwise comparisons with a Bonferroni correction, which showed that there was a statistically significant difference in GD2 T cells among HIV- women with abnormal vaginal flora (Nugent score more than 4) compared to those with normal vaginal flora (p = 0.023), but not among HIV+ women (p = 0.171).