The G-protein coupled estrogen receptor (GPER/GPR30) is a gonadotropin receptor dependent positive prognosticator in ovarian carcinoma patients

Abstract

Follicle stimulating hormone receptor (FSHR) and luteinizing hormone receptor (LHCGR) were demonstrated to impact upon survival of patients suffering from epithelial ovarian cancer (EOC). Though structure wise the G-protein coupled estrogen receptor (GPER/GPR30) is related to FSHR/LHCGR, its prognostic impact in EOC remains controversial. We recently found that FSHR negative patients represent a specific EOC subgroup that may behave differently in respect to both treatment response and prognosis. Hence, the current study aimed to analyze how GPER may interact with the FSHR/LHCGR system in EOC and whether the prognostic significance of GPER in EOC cases (n=151) may be dependent on the FSHR/LHCGR immunophenotype of the tumor. Ovarian cancer cell lines were used to study how FSH and LH regulate GPER and whether GPER activation differentially affects in vitro cell proliferation in presence/absence of activated FSHR/LHCGR. In EOC tissue, GPER correlated with FSHR/LHCGR and was related to prolonged overall survival only in FSHR/LHCGR negative patients. Although GPER was found to be specifically induced by LH/FSH, GPER agonists (4-Hydroxy-Tamoxifen, G1) reduced EOC cell proliferation only in case of LH/FSH unstimulated pathways. To the same direction, only patients characterized as LHCGR/FSHR negative seem to gain from GPER in terms of survival. Our combined tissue and in vitro results support thus the hypothesis that GPER activation could be of therapeutic benefit in LHCGR/FSHR negative EOC patients. Further studies are needed to evaluate the impact of GPER activation on a clinical scheme.

Figure 1. Representative microphotographs of GPER expression in ovarian cancer are presented.

GPER showed a membrane as well as a cytoplasmic staining pattern in serous (A, A’), clear cell (B, B’), endometrioid (C, C’) and mucinous (D, D’) ovarian cancer specimens. In endometrioid cancers GPER was significantly lower than in mucinous (p = 0.01, *) or serous (p = 0.03, #) ones. Scale bars equal 100 µm and, box plots present GPER in relation to histological subtype (E). Significant observations derived from relevant Mann-Whitney U-tests.

Figure 2. GPER predicts favorable outcome in gonadotropin receptor negative EOC.

Prognostic significance of GPER was evaluated in subgroups of patients with or without expression of FSHR or/and LHCGR. Survival of patients whose tumors expressed GPER at high levels (solid lines) was compared to those with low GPER expression (dotted lines) by the log rank test and Kaplan-Meier survival plots were drawn. Remarkably, GPER predicted significantly more favorable outcome in subgroups classified as FSHR negative (A) and LHCGR negative (B). Stratification of EOC patients according the combined LH/FSH status revealed that only in case of a double negative immunophenotype GPER appears as a positive prognosticator (C).

Figure 3. GPER is up-regulated by gonadotropins in ovarian cancer cells.

(A) Screening of ovarian cancer cell lines for FSHR and LHCGR. Caov-3 cells expressed gonadotropin receptors at very low to non-detectable levels, while SKOV-3 was LHCGR positive and OVCAR-3 expressed FSHR. GPER was detected in all the three cell lines used in this study (A, B). OVCAR-3 cells were treated with FSH, which enhanced GPER protein expression (C) and gene transcription (D). SKOV-3 cells were treated with LH, which induced both GPER protein (E) and gene transcription (F). c: control, asterisks mark significant (p<0.05) observations as calculated using independent samples Student’s T-test (C, E) and the Rest2009 algorithm for gene transcription ratios (D, F). Scale bars in (B) equal 50 µm.

Figure 4. GPER up-regulation is dependent on the respective gonadotropin receptor.

siRNA mediated knockdown of FSHR (A, B) and LHCGR (C, D) was performed in order to evaluate whether the up-regulation of GPER is due to an effect specifically attributed to FSH/LH and FSHR/LHCGR. Gonadotropin treatment failed to up-regulate GPER in cells that had undergone FSHR (E) or LHCGR (F) silencing. c: control treated with the transfection reagent only, scr: control treated with an off target, scrambled siRNA. Stars mark significant (p<0.05) observations as calculated using independent samples Student’s T-test (E, F) and the Rest2009 algorithm for gene transcription ratios (B, D).

Figure 5. GPER signaling activators inhibit proliferation of ovarian cancer cells lines in an FSH and LH dependent manner.

FSHR positive (OVCAR-3) and negative (Caov-3) cells were exposed to GPER agonists (G1; 4-Hydroxy-Tamoxifen (OHT)) in the presence vs. absence of external FSH (A). In the FSH depleted setting G1 and OHT reduced OVCAR-3 cell proliferation by, respectively. Neither G1 nor OHT emerged to significantly affect OVCAR-3 proliferation in the presence of FSH. Caov-3 cells, being negative for GnRs, were sensitive to addition of GPER inducers regardless the presence of FSH, since both G1 and OHT significantly reduced median Caov-3 replication rates. Both G1 and OHT failed to reduce proliferation of LHCGR positive SKOV-3 cells in the presence of LH (B). In LH non stimulated conditions proliferation of SKOV-3 cells was significantly reduced when either G1 or OHT were added to the culture media. Doubling rates of LHCGR negative Caov-3 cells were reduced by G1 as well as by OHT regardless the presence of LH. Independent samples Student’s T-test was employed to test for differences between groups and significant (p<0.05) changes are indicated by stars (*).