Estrogen receptor-alpha (ER-alpha) and defects in uterine receptivity in women

Abstract

Endometriosis is a disorder that affects 5% of the normal population but is present in up to 40% of women with pelvic pain and/or infertility. Recent evidence suggests that the endometrium of women with endometriosis exhibits progesterone insensitivity. One endometrial protein that fluctuates in response to progesterone is the estrogen receptor-alpha (ER alpha), being down-regulated at the time of peak progesterone secretion during the window of implantation. Here we demonstrate that the biomarker of uterine receptivity, beta 3 integrin subunit, is reduced or absent in some women with endometriosis and that such defects are accompanied by inappropriate over-expression of ER alpha during the mid-secretory phase. Using a well-differentiated endometrial cell line we showed that the beta 3 integrin protein is negatively regulated by estrogen and positively regulated by epidermal growth factor (EGF). By competing against estrogen with various selective estrogen receptor modulators (SERMs) and estrogen receptor agonists and antagonists, inhibition of expression of the beta 3 integrin by estrogen can be mitigated. In conclusion, we hypothesize that certain types of uterine receptivity defects may be caused by the loss of appropriate ER alpha down-regulation in the mid-secretory phase, leading to defects in uterine receptivity. Such changes might be effectively treated by timely administration of the appropriate anti-estrogens to artificially block ER alpha and restore normal patterns of gene expression. Such treatments will require further clinical studies.

Figures

Figure 1

Immunohistochemistry of ER-alpha in normal cycling women from the proliferative with inset showing luminal epithelium (A), early secretory (B) and mid-secretory (C) endometrium. Other samples from the mid-secretory endometrium from women with endometriosis shows increased and inappropriate expression of the ER-alpha protein in glandular and stromal compartments (D and E) compared to normal mid-secretory endometrium (C). Bar = 50 μm.

Figure 2

Box and whisker plots are shown depicting the semi-quantitative immunohistochemical staining (HSCORE ± SD) for ER-alpha in normal women throughout the menstrual cycle in proliferative, early, mid and late secretory glandular (A) and luminal (B) epithelium and in stroma (C). In a separate analysis, glandular and luminal ER-alpha HSCOREs were compared between normal fertile mid-secretory endometrium and similar samples in women with endometriosis and suspected defects in uterine receptivity (D). Significance was based on ANOVA with Bonferroni correction; * = p th, 25th, 50th, 75th, and 90th percentile distribution of the data and standard error bars are not shown in this figure.

Figure 3

Western blot for beta 3 integrin expression in Ishikawa cells treated with no treatment (con), or three days of estradiol (E2; 10-8M), estradiol plus the 10-6 M pure antagonist ICI 182,780 (E2 + ICI), epidermal growth factor (EGF; 10 ng/ml), EGF plus the EGF receptor antagonist antibody C225 (EGF + C225), EGF + ICI, EGF + E2 + ICI and ICI alone.

Figure 4

Western blot showing the effect of estrogen and anti-estrogens on the inhibition of the beta 3 integrin subunit in Ishikawa cells. In control cells, the beta 3 integrin was expressed while 10 -8 M estradiol reduced that expression significantly. E3 (estriol) was less potent than estradiol (E2) at inhibiting beta 3. Of the SERMS, in cells treated with E2 plus equimolar concentrations of other anti-estrogens and SERMs, a differential competitive antagonism occurred with the most potent antagonism by ICI 182,780. Raloxifene had an anti-estrogen effect on this well-differentiated endometrial cell line as well. Con = Control; E2 = estradiol; E3 + estriol; ICI = ICI 182,780, C = clomiphene citrate; T = tamoxifen; R = raloxifene.