Sex hormone regulation of innate immunity in the female reproductive tract: the role of epithelial cells in balancing reproductive potential with protection against sexually transmitted pathogens

Abstract

The immune system in the female reproductive tract (FRT) does not mount an attack against HIV or other sexually transmitted infections (STI) with a single endogenously produced microbicide or with a single arm of the immune system. Instead, the body deploys dozens of innate antimicrobials to the secretions of the female reproductive tract. Working together, these antimicrobials along with mucosal antibodies attack many different viral, bacterial and fungal targets. Within the FRT, the unique challenges of protection against sexually transmitted pathogens coupled with the need to sustain the development of an allogeneic fetus have evolved in such a way that sex hormones precisely regulate immune function to accomplish both tasks. The studies presented in this review demonstrate that estradiol and progesterone secreted during the menstrual cycle act both directly and indirectly on epithelial cells and other immune cells in the reproductive tract to modify immune function in a way that is unique to specific sites throughout the FRT. As presented in this review, studies from our laboratory and others demonstrate that the innate immune response is under hormonal control, varies with the stage of the menstrual cycle, and as such is suppressed at mid-cycle to optimize conditions for successful fertilization and pregnancy. In doing so, a window of STI vulnerability is created during which potential pathogens including HIV enter the reproductive tract to infect host targets.

Fig. 1

Schematic of the mucosal immune system throughout the human female reproductive tract. As seen in the drawing on the left side, the vagina and ectocervix are lined with squamous epithelial cells. Columnar cells are present throughout the upper FRT including the endocervix, uterine endometrium, and fallopian tubes. Panel a–d are confocal photomicrographs showing the distribution of immune cells throughout out the uterus. Frozen sections were directly stained with three fluorescently tagged monoclonal antibodies: an epithelial cell specific antibody (Cy3-labeled clone BerEP4, red color panels a–d), anti-CCR5 (FITC-labeled clone 2D7, green color panels a–d), and anti-CXCR4 (Cy5 labeled 12G5, blue color panels a–d). Panel a shows epithelial gland extending from the myometrial interface (far left) to the luminal epithelium (right-hand side). Panels b through d show higher magnification fields from the same region. The luminal epithelium, in contrast to the adjacent glands, retains a relatively intense BerEP4 expression (red color panels a and d). CCR5 expression is prominent on the lymphoid aggregates located in the stratum basalis immediately adjacent to the myometrium (‘LA’ in panel b). Epithelial expression of CCR5 is low in the stratum basalis and increases in the proximal third of the stratum functionalis. Panel f: This photograph plate consists of a lymphoid aggregate in the uterine endometrium at the late proliferative stage of the menstrual cycle. The three fluorochromes are T cells (Cy3-anti-CD3, red), B cells (FITC-anti-CD19, green), and macrophages (Cy5-anti CD14, blue). Panel e: Vaginal squamous cells expressing GalCer (green) is expressed on parabasal epithelial cells (p) and in the surface regions of the cornified layer. Panel g: CD8+ T cells present in both the submucosa and the squamous epithelium. Panel h: CD14 expression is found on both stroma and squamous epithelium macrophages. Panel i: CD1a-positive dendritic cells (DC) are present in the squamous epithelium (blue). Panel a–d; Panel f; Panel e, g, h, i.

Fig. 2

Estradiol inhibits transepithelial resistance (TER) in the human uterine epithelial cell line ECC-1. ECC-1 cells were grown in media supplemented with stripped FBS until maximal TER of about 2000 Ω/well in cell inserts was obtained. Estradiol (5 × 10⁻⁸ m) and/or progesterone (1 × 10⁻⁷ m) were added to the basolateral compartment of some inserts, and TER was measured after 48 hr. Four wells per group; **significantly different (P < 0.01) from control.

Fig. 3

Effect of estradiol and poly (I:C) on TER of primary uterine epithelial cells. Human uterine epithelial cells were cultured in cell inserts in media supplemented with stripped FBS until maximal TER was obtained. The media was changed and some inserts were cultured with 5 × 10⁻⁸ m estradiol for 24 hr and then an additional 24 hr with estradiol with or without apical treatment with poly (I:C) and TER was measured. Six wells per group; **significantly different (P < 0.01) from control.

Fig. 4

Estradiol stimulates secretion of SLPI by ECC-1 cells. ECC-1 cells were grown to confluence and high TER in cell inserts. The media was changed and some inserts were cultured with 5 × 10⁻⁸ m estradiol and/or progesterone (1 × 10⁻⁷ m) for 48 hr. Apical conditioned media were recovered, centrifuged to remove cell debris, and SLPI was measured by ELISA. Four wells per group; **significantly different (P < 0.01) from control or progesterone.

Fig. 5

Estradiol inhibits LPS-induced IL-6 secretion by uterine epithelial cells. Monolayer cultures of uterine epithelial cells at high TER were treated with 5 × 10⁻⁸ m estradiol for 48 hr in the basolateral compartment of cell inserts with some wells receiving an apical treatment of ultra pure lipopolysaccharide (LPS; 100 mg/mL) after 24 hr. Apical conditioned media were recovered, centrifuged to remove cell debris, and IL-6 was measured by ELISA. LPS stimulated a significant increase in IL-6 secretion, which was mostly abrogated by pre-treatment with estradiol. Four wells per group; **significantly different (P < 0.01) from control, estradiol and estradiol + LPS. Adapted from Ref. [12].

Fig. 6

Estradiol inhibits IL-1β mediated secretion of HBD2 and IL-8 and the inhibition occurs although involvement of the estrogen receptor. ECC-1 cells were preincubated with estradiol, the pure ER antagonist ICI182780, or a combination of steroid and antagonist for 72 hr before stimulation with IL-1β (5 ng/mL). Apical and basolateral conditioned medium were collected following 24-hr stimulation and analysed for HBD2 and IL-8 protein secretion by ELISA. The results are shown as the mean ± SEM. *Significantly different (P < 0.05) from control. **Significantly different (P < 0.01) from control. ***Significantly different (P < 0.001) from control. Gray bars, apical conditioned medium; black bars, basolateral conditioned medium. Adapted from Ref. [50].

Fig. 7

Estradiol decreases the production of HBD2 and elafin in primary vaginal squamous epithelial cells. Vaginal secretions and cells were recovered from volunteers using an Instead™ Menstrual Cup (Instead, La Jolla, CA) following insertion for 1 hr. Cells were cultured overnight in a 96-well plate before treatment with estradiol (5 × 10⁻⁸ m) for 48 hrs. Supernatants were recovered, centrifuged, and analysed for HBD2, elafin, and MIP3α production by ELISA (**P < 0.01; ***P < 0.001).

Fig. 8

Schematic of estradiol regulation of innate immune function by human epithelial cells in the upper and lower female reproductive tract. In the uterus, estradiol enhances the secretion of antimicrobial factors and reduces the secretion of induced pro-inflammatory mediators. In contrast to the uterus, estradiol inhibits both the constitutive and induced secretion of antimicrobials in the vagina.

Fig. 9

Estradiol stimulates secretion of HGF by human uterine stromal fibroblasts. Uterine fibroblasts were cultured to confluence in 24-well plates and 1 × 10⁻⁸ m estradiol or control media was added on Day 0 and every 2 days thereafter with media change. The conditioned media was recovered every 2 days, centrifuged, and supernatants analysed for HGF by ELISA. HGF secretion increased with estradiol treatment and with ongoing culture. Four wells per group; **, significantly different from control on the day shown (P < 0.01).

Fig. 10

Estradiol augments LPS-induced IL-1β levels in monocytes (a). Peripheral blood monocytes were incubated with indicated concentrations of estradiol for 24 hr and then stimulated with 10 ng/mL LPS for an additional 12 hr. Supernatants were collected from these cultures. IL-1β production was quantified by ELISA. (b) Estradiol inhibits interleukin-1 receptor type I (IL-1RtI) protein expression. Whole cell lysates were generated from ECC-1 cells incubated with various concentrations of estradiol for 72 hr. Proteins were resolved by 10% SDS-PAGE and detected with an anti-IL-1RtI antibody. Individual bands were scanned, and the intensity was quantified by computer analysis. The amount of IL-1RtI was normalized to GAPDH levels and plotted as a percentage of the control (representative of two experiments).^,

Fig. 11

Estradiol increases DC-SIGN expression on dendritic cells. Immature DC were derived from human monocytes with IL-4 and GM-CSF for 7 days in the presence or absence of physiological concentrations of estradiol (10⁻⁷ m). (a) Overlay histograms for DC-SIGN expression. The histograms depict DC-SIGN expression by control DC (green line) and DC generated with estradiol (red line). The black line represents staining with matched isotype control antibody. (b) Averaged mean fluorescence intensity (MFI) values for DC-SIGN expression (n = 3 donors) are shown as % MFI relative to control DC. *Indicates P < 0.05 compared to control DC.

Fig. 12

Effect of estradiol and TGFβ on NK cell immune function. To examine the effect of estradiol on chemokine expression, endometrial tissue sections were incubated with different concentrations of estradiol as indicated for 48 hr, then snap frozen and stored at −80°C. Chemokines CXCL11 (a) and CXCL10 (b) were measured following the isolation of total RNA using TRIzol. Quantitative real-time PCR was used to determine the relative fold expression of each gene compared with medium only. In other studies, endogenous TGFβ suppression of poly (I:C)-induced interferon-γ (IFN-γ) production by uterine NK cells was measured (c and d). Uterine cells were isolated from a hysterectomy patient and cultured with media, IL-12 and IL-15, or poly (I:C) as indicated for 18 hr in the presence of blocking anti-TGFβ monoclonal antibodies (αTGFβ), control IgG (IgG), or media only (media). Uterine NK cells were then analysed for intracellular IFN-γ production by flow cytometry (c). IL-12 and IL-15 in combination and poly (I:C) increase the percent of IFN-γ+ producing NK cells and antibody to TGFβ enhances the number of uterine NK cells induced by poly (I:C) that produce IFN-γ. (d) The data expressed in terms of fold increase in IFN-γ+ NK cells. Adapted from Ref. [125] (bottom) and adapted from Ref. [130].

Fig. 13

Estradiol induces antibacterial activity in ECC-1 epithelial cells. Human uterine epithelial cells were treated with or without two concentrations of estradiol for 48 hr. The apical conditioned media was recovered after 2 days, centrifuged and an aliquot of supernatants was then incubated with Staphylococcus aureus for 1 hr, and the bacteria were cultured overnight. Control colony-forming units (CFU) refers to the colonies grown in media in the absence of uterine epithelial cells. The inhibition seen with no estradiol represents the bacteria that grew out in the presence of antimicrobials constitutively produced by the cells. Estradiol by itself had no effect on bacterial CFU (not shown). Four wells per group; **significantly different (P < 0.01) from control. Adapted from Ref. [12].