Expression of the G protein-coupled estrogen receptor (GPER) in endometriosis: a tissue microarray study

The G protein-coupled estrogen receptor (GPER) is thought to be involved in non-genomic estrogen responses as well as processes such as cell proliferation and migration. In this study, we analyzed GPER expression patterns from endometriosis samples and normal endometrial tissue samples and compared these expression profiles to those of the classical sex hormone receptors.

A tissue microarray, which included 74 samples from different types of endometriosis (27 ovarian, 19 peritoneal and 28 deep-infiltrating) and 30 samples from normal endometrial tissue, was used to compare the expression levels of the GPER, estrogen receptor (ER)-alpha, ER-beta and progesterone receptor (PR). The immunoreactive score (IRS) was calculated separately for epithelium and stroma as the product of the staining intensity and the percentage of positive cells. The expression levels of the hormonal receptors were dichotomized into low (IRS < 6) and high (IRS > =6) expression groups.

The mean epithelial IRS (+/−standard deviation, range) of cytoplasmic GPER expression was 1.2 (+/−1.7, 0–4) in normal endometrium and 5.1 (+/−3.5, 0–12) in endometriosis (p < 0.001), of nuclear GPER 6.4 (+/−2.6, 0–12) and 6.8 (+/−2.9, 2–12; p = 0.71), of ER-alpha 10.6 (+/−2.4, 3–12) and 9.8 (+/−3.0, 2–12; p = 0.26), of ER-beta 2.4 (+/−2.2; 0–8) and 5.6 (+/−2.6; 0–10; p < 0.001), and of PR 11.5 (+/−1.7; 3–12) and 8.1 (+/−4.5; 0–12; p < 0.001), respectively. The mean stromal IRS of nuclear GPER expression was 7.7 (+/−3.0; 2–12) in endometrium and 10.8 (+/−1.7; 6–12) in endometriosis (p < 0.001), of ER-alpha 8.7 (+/−3.1; 2–12) and 10.6 (+/−2.4; 2–12; p = 0.001), of ER-beta 1.8 (+/−2.0; 0–8) and 5.4 (+/−2.5; 0–10; p < 0.001), and of PR 11.7 (+/−0.9; 8–12) and 10.9 (+/−2.0; 3–12; p = 0.044), respectively. Cytoplasmic GPER expression was not detectable in the stroma of endometrium and endometriosis. The observed frequency of high epithelial cytoplasmic GPER expression levels was 50% (n = 30/60) in the endometriosis and none (0/30) in the normal endometrium samples (p < 0.001). High epithelial cytoplasmic GPER expression levels were more frequent in endometriomas (14/20, 70%; p = 0.01), as compared to peritoneal (9/18, 50%) or deep-infiltrating endometriotic lesions (7/22, 31.8%). The frequency of high stromal nuclear GPER expression levels was 100% (n = 74/74) in endometriosis and 76.7% (n = 23/30) in normal endometrium (p < 0.001). The frequency of high epithelial nuclear GPER expression levels did not differ between endometriosis and normal endometrium.

The present data indicate a unique GPER expression pattern in endometriosis, especially in endometriomas as compared to the normal endometrium. The overexpression of GPER in endometriotic lesions suggests a potential role for GPER in the hormonal regulation of endometriosis, which should be taken into consideration for future hormonal treatment strategies.