Stimulating the GPR30 estrogen receptor with a novel tamoxifen analogue activates SF-1 and promotes endometrial cell proliferation

Abstract

Estrogens and selective estrogen receptor (ER) modulators such as tamoxifen are known to increase uterine cell proliferation. Mounting evidence suggests that estrogen signaling is mediated not only by ERalpha and ERbeta nuclear receptors, but also by GPR30 (GPER), a seven transmembrane (7TM) receptor. Here, we report that primary human endometriotic H-38 cells express high levels of GPR30 with no detectable ERalpha or ERbeta. Using a novel tamoxifen analogue, STX, which activates GPR30 but not ERs, significant stimulation of the phosphatidylinositol 3-kinase (PI3K) and mitogen-activated protein kinase (MAPK) pathways was observed in H-38 cells and in Ishikawa endometrial cancer cells expressing GPR30; a similar effect was observed in JEG3 choriocarcinoma cells. STX treatment also increased cellular pools of phosphatidylinositol (3,4,5) triphosphate, a proposed ligand for the nuclear hormone receptor SF-1 (NR5A1). Consistent with these findings, STX, tamoxifen, and the phytoestrogen genistein were able to increase SF-1 transcription, promote Ishikawa cell proliferation, and induce the SF-1 target gene aromatase in a GPR30-dependent manner. Our findings suggest a novel signaling paradigm that is initiated by estrogen activation of the 7TM receptor GPR30, with signal transduction cascades (PI3K and MAPK) converging on nuclear hormone receptors (SF-1/LRH-1) to modulate their transcriptional output. We propose that this novel GPR30/SF-1 pathway increases local concentrations of estrogen, and together with classic ER signaling, mediate the proliferative effects of synthetic estrogens such as tamoxifen, in promoting endometriosis and endometrial cancers.

Figure 1

Primary human ectopic H-38 endometrial cells express GPR30, SF-1, and CYP19A1, but no ERs. A, relative expression of GPR30 and nuclear ERs in primary H-38 endometrial cancer cells, and other cell lines including breast carcinoma cell lines: MCF-7, MDA-MB-231 (MDA), and T47-D; endometrial cancer cell line: Ishikawa; choriocarcinoma placental cell line: JEG3; primary endometrial cells: normal endometrium (Endo) and endometriotic cells (H-38). B, localization of endogenous GPR30 (calnexin, endoplasmic reticulum membrane marker) in H-38 cells, determined by immunocytochemistry and confocal microscopy. C, relative expression of SF-1 and Cyp19a1 in T47-D, Ishikawa, normal endometrium, and H-38 cells.

Figure 2

Tamoxifen and the tamoxifen analogue STX activate PI3K and MAPK pathways in H-38 cells. A, chemical structures of STX and OHT. B, phospho-ERK1/2 (pERK1/2) and phospho-Akt (pAkt) were detected by immunoblotting using serum-depleted H-38 primary cells treated with negative control inhibitors [1 μM; 100 nM wortmannin (W) or 10 μM U0126 (U), 30 min pretreatment before adding STX], or with positive control drugs (EGF or insulin); total ERK1/2 and Akt show loading controls.

Figure 3

STX fails to bind, activate, or antagonize nuclear ERs. A, 17βE2 (open symbols) or STX (filled symbols) binding to hERα (top) or hERβ (bottom) was measured by competition binding assays; fluorescence polarization readout using fluorescently labeled E2. B, JEG3 (ER-negative) cells transfected with hERα (top) or Ishikawa (ERβ-positive) cells (bottom) were assayed for ERE TATA-luciferase reporter activity following 24 h off drug treatment (drug and dose indicated). Gen, genistein.

Figure 4

STX stimulates SF-1 transcriptional activity and increases SF-1 phosphorylation in a GPR30-dependent manner. A, drug-dependent stimulation of the aromatase promoter II luciferase reporter (Aro-Luc) in JEG3 cells transfected with mSF-1 with expression confirmed (inset). B, aro-Luc reporter activation by SF-1 or SF-1 variants (S203A or A270W) after treating cells with 1 μM of the indicated drug. C, effects of MAP/ERK kinase (10 μM U0126) or PI3K [100 nM wortmannin (WM)] inhibitors (30 min pretreatment) on 1 μM STX stimulation of SF-1 transactivation (left). Following a 10-min STX treatment, phospho-SF-1 and phospho-ERK1/2 (pSF-1 and pERK1/2) were detected (right), with vehicle (−) and EGF (positive control) also shown. D, effects of siRNA on 1 μM drug-stimulated SF-1 activity (left) and phosphorylation (right) with duplicate independent experiments (+) shown. Control, no siRNA added; sicRNA, control siRNA; si-GPR30, siRNA against GPR30; SF-1, total SF-1 loading control. Representative data from three independent experiments are shown. In A, B, C (left), and D (left), cells were treated with indicated drugs for 4 h, 16 h following transfection; luciferase activity is mean values ± SD. In C (right) and D (right), cells were treated with the indicated drugs for 1 h.

Figure 5

STX increases PI3K/MAPK activity and PIP3 accumulation in Ishikawa cells stably expressing GPR30. A, drug-stimulated (4 h) Aro-Luc reporter activity in GPR30 stable Ishikawa cells with mSF-1 added was measured (left). Percent activation is normalized to activity observed with mSF-1 added and no drug treatment (100%). GPR30 mRNA expression was measured in GPR30-expressing stable Ishikawa cells (+ GPR30) and control cells (+ vector; right) B, phospho-Akt time course (1 μM drug) and dose response (15 min treatment) were detected by immunoblotting using cells transiently expressing GPR30. The inhibitor (100 nM WM) was added 30 min before 1 μM drug treatment for 15 min; total Akt shows loading controls and with negative controls as time (0) or vehicle (−). C, levels of [³²P]PI(3,4,5)P3 measured by high performance liquid chromatography in stable GPR30-expressing Ishikawa cells after vehicle or STX treatment (dose indicated, 15 min) after [³²P] orthophosphate labeling; peak shown confirmed to be glycero-inositol(1,3,4,5)P by de-acylated standard. D, dose response of STX or negative control vehicle (−) induced [³²P]PI(3,4,5)P3 peak accumulation normalized to total radioactivity (columns, mean of three independent experiments; bars, SD).

Figure 6

STX stimulation increases endogenous aromatase expression and proliferation in Ishikawa cells stably overexpressing GPR30 and model. A, endogenous Cyp19a1 expression measured after drug treatment (1 μM 17αE2, genistein, OHT, STX, or 100 nM 17βE2 for 16 h) in Ishikawa cells stably expressing GPR30. B, effects of GPR30 siRNA (si-GPR30) on Cyp19a1 expression in drug-stimulated (1 μM for 16 h) Ishikawa cells stably expressing GPR30; control, untransfected; si-GFP, control siRNA. C, following serum depletion and drug treatment (1 μM 17αE2, genistein, OHT, STX, or 100 nM 17βE2 for 16 h), cell proliferation was analyzed by FACS for control (stable vector) and GPR30 stably expressing (stable GPR30) Ishikawa cells. A to B, expression measured by quantitative PCR; columns, mean values; bars, SD; A to C, vehicle negative control (−) is indicated. D, proposed model of GPR30 and NR5A receptor signaling by estrogens. STX and estrogens bind and activate GPR30. GPR30 is reported to reside on either the plasma membrane (49) and/or on intracellular membranes (7), as depicted in the model. GPR30 activation by STX or OHT increases ERK1/2 and PI3K phosphorylation and elevates the cellular pools of PIP3. Collectively, these events activate NR5A nuclear receptor activity to promote increased expression of target genes such as aromatase.