The hormonal environment and estrogen receptor signaling alters Chlamydia muridarum infection in vivo

Abstract

Genital Chlamydia is the most common bacterial sexually transmitted infection in the United States and worldwide. Previous studies indicate that the progression of chlamydial infection is influenced by various factors, including the female sex hormones estrogen and progesterone. Sex hormone levels naturally fluctuate in women throughout their menstrual cycle. Varying concentrations of estrogen and progesterone may impact the progression of chlamydial infection and the host's immune response to Chlamydia. Estrogen signals through estrogen receptors (ERs), ERα and ERβ. These receptors are similar in structure and function, but are differentially expressed in tissues throughout the body, including the genital tract and on cells of the immune system. In this study, we used ovariectomized (OVT) BALB/c mice to investigate the impact of long-term administration of physiologically relevant concentrations of estrogen (E2), progesterone (P4), or a combination of E2/P4 on the progression of and immune response to C. muridarum infection. Additionally, we used ERα and ERβ knockout C57/BL6 mice to determine the how ERs affect chlamydial infection and the resulting immune response. Estrogen exposure prevented C. muridarum infection in vaginally infected OVT mice exposed to E2 alone or in combination with P4, while OVT or Sham mice exposed to hormone free, P4 or depo-medroxyprogesterone acetate shed similar amounts of chlamydiae. The hormonal environment also altered T cell recruitment and IFNϵ production the genital tracts of infected OVT and Sham mice on day 10 post infection. The absence of ERα, but not ERβ, in ER knockout mouse strains significantly changed the timing of C. muridarum infection. ERαKO mice shed significantly more chlamydiae at day 3 post infection and resolved the infection faster than WT or ERβKO animals. At day 9 post infection, flow cytometry showed that ERαKO mice had more T cells present and targeted RNA sequencing revealed increased expression of CD4 and FOXP3, suggesting that ERαKO mice had increased numbers of regulatory T cells compared to ERβKO and WT mice. Mock and chlamydia-infected ERαKO mice also expressed more IFNϵ early during infection. Overall, the data from these studies indicate that sex hormones and their receptors, particularly ERα and ERβ, differentially affect C. muridarum infection in murine models of infection.

Keywords: chlamydia; estrogen receptor; estrogen receptor knockout mice; long-term hormone delivery; ovariectomized mouse model.

Figure 1

Measurement of E2 and P4 from serum. The concentration of 17-β estradiol (A) and progesterone (B) in mouse sera was determined by ELISA or LC MS/MS, respectively as described in the Methods. E2: 17-β estradiol, P4: progesterone, E2/P4: combination of 17-β estradiol and progesterone, SHAM: sera from surgical sham mice treated with DMPA. Data shown represent the average hormone concentration ± standard deviation in pooled sera from 16 mice collected across 2 independent experiments.

Figure 2

Sex hormones alter bacterial vaginal shedding and immune response to C muridarum (Cm) in infected ovariectomized mice. (A). EB shedding was monitored by chlamydial titer assays from C muridarum infected, hormone-treated mice for 21 days pi. Data shown represent the average (n= 12-29 mice/group/day) IFU/group ± SEM. (B) Flow cytometry analysis was performed to detect CD3 positive T cells in the genital tracts of mock- or Cm infected, hormone-treated OVT mice at day 10pi. Data shown represent the average (n= 12 mice/group) percentage of total cells measured ± SEM. Asterisks indicate significant differences (P ≤ 0.05). *Cm vs mock, **HF (Cm), ***P4 (Cm), vs experimental group. (C) The concentration of IFNϵ was detected in vaginal swab samples on day 3pi by ELISA. Data shown represent the average (n= 7 mice/group) IFNϵ concentration ± SEM. Asterisks (*) indicate significant differences between Cm and mock infection (P ≤ 0.05).

Figure 3

Estrogen receptors alter bacterial shedding, pathology, and the T cell response to C muridarum (Cm) in infected C57Bl/6 mice. (A). EB shedding was monitored by chlamydial titer assays from C muridarum infected C57Bl/6 WT, ERαKO or ERβKO mice for 21 days pi. Data shown represent the average (n= 24 mice/group/day) IFU/group ± SEM. (B). Gross genital tract pathology was assessed using the scoring criteria in Table 1 on day 9 or 21pi. Data shown represent the average pathology score ± SEM/group. (n=4-21mice/group/day) An asterisk (*) indicates significant difference from Cm-ERαKO D9pi, **Significantly different from Cm-ERαKO D21pi. (C) Flow cytometry analysis was performed to detect CD3 positive T cells in the genital tracts of mock or Cm-infected, hormone-treated OVT mice at day 9pi. Data shown represent the average (n=12 mice/group) percentage of total cells measured ± SEM. Asterisks indicate significant differences (P ≤ 0.05). *Cm vs mock, **ERαWT Cm vs Cm group, ***ERβWT Cm vs Cm group. (D-F). Fold changes in expression of genes determined to be significantly different using the CLC Bio Genomics Workbench RNAseq deferential expression analysis tool. (D). Comparisons between Cm-infected ERαKO and WT whole genital tracts (GT), or cervix, horn and ovary tissues. (E). Comparisons between Cm-infected ERαKO and ERβKO GTs. (F). Comparisons between Cm-infected ERαKO cervix, horn and ovary tissues. (G). The concentration of IFNϵ detected in vaginal swab samples on day 3pi by ELISA. Data shown represent the average (n= 4-9 mice/group) IFNϵ concentration ± SEM. Asterisks (*) indicate significant differences between Cm and mock infection (P ≤ 0.05).