Chorionic gonadotropin regulates prostaglandin E synthase via a phosphatidylinositol 3-kinase-extracellular regulatory kinase pathway in a human endometrial epithelial cell line: implications for endometrial responses for embryo implantation

Abstract

Successful implantation necessitates modulation of the uterine environment by the embryo for a specific period of time during the menstrual cycle. Infusion of chorionic gonadotropin (CG) into the oviducts of baboons to mimic embryo transit induces a myriad of morphological, biochemical, and molecular changes in the endometrium. Endometrial epithelial cells from both baboons and humans when stimulated by CG in vitro, activates a cAMP-independent MAPK pathway leading to prostaglandin E(2) (PGE(2)) synthesis. This study shows that in the human endometrial cell line, HES, CG, acting via its G-protein coupled receptor, phosphorylates protein kinase B, c-Raf, and ERK1/2 in a phosphatidylinositol 3-kinase (PI3K)-dependent manner. Furthermore, ERK1/2 phosphorylation is independent of the signaling paradigms of Galpha(s), Galpha(I), and epidermal growth factor receptor (EGFR) transactivation, typical of gonadal cells, indicating an alternative signaling pattern in the endometrium. After phosphorylation by CG, ERK1/2 translocates to the nucleus in a time-dependent manner. Downstream of ERK1/2, CG activates the nuclear transcription factor, Elk1, also in a PI3K-MAPK-dependent manner. Lastly, we show that in HES cells, this pathway regulates the expression of the microsomal enzyme PGE(2) synthase (mPTGES), a terminal prostanoid synthase responsible for PGE(2) synthesis. CG regulates the mPTGES promoter and also induces mPTGES synthesis in HES cells via the PI3K-ERK1/2 pathway. We suggest that this alternative PI3K-ERK-Elk pathway activated by CG regulates prostaglandin production by the endometrial epithelium and serves as an early trigger to prepare the endometrium for implantation.

Figures

Figure 1

Detection and silencing of LHCGR- and CG-induced signaling in HES cells. A, HES and CHO-LH cells were plated on six-well plates at a visual confluence of 80%. HES cells were transfected with the shRNA cassettes targeted against the LHCGR or scrambled siRNA controls. Cells were then lysed and total protein was extracted from the cells after transfection. Western blot analysis was carried out to detect levels of the 80-kDa LHCGR protein. Note the decrease in LHCGR expression upon transfection with siRNA cassette (lane 3). B, HES cells were plated on six-well plates at a visual confluence of 80%. Cells were transfected with shRNA cassettes or the scrambled sequence as a control. Thirty-six hours after transfection, they were treated with or without 10 IU/ml of CG or 10% FBS for 10 min. The immunoblot shows p- and t-ERK1/2 levels detected in 30 μg of lysate from the cells separated by SDS-PAGE. The histograms represent the normalization to t-ERK1/2. Note that ERK1/2 phosphorylation by CG (and not by FBS) is significantly reduced upon silencing the LHCGR with the siRNA sequence. Data are expressed as the mean sem of three different experiments done in triplicate. *, P < 0.05. Cntrl, Control; CT, control; Scr, scrambled (Scrmbld) siRNA sequence; siRNA, siRNA specific against LHCGR.

Figure 2

ERK1/2 phosphorylation by CG is independent of Gαs/Gαi/EGFR activation in HES cells. HES cells (A, C, and E) and CHO-LH cells (B, D, and F) were grown to 60% of visual confluence. HES (A) and CHO-LH (B) cells were treated with 10 IU/ml CG for 10 min or 10 or 100 ng/ml Ctx for 1 h. HES (C) and CHO-LH (D) cells were treated for 10 min with 10 IU/ml CG after pretreatment with 100 or 200 ng/ml Ptx-Gi inhibitor for 1 h. HES cells (E) and CHO-LH cells (F) were treated for 10 min with 10 IU/ml CG after pretreatment with 100 nm AG1478, an EGFR inhibitor, for 1 h. Thirty micrograms of the total cell lysates from each experiment were isolated and analyzed using Western blot analysis for levels of p-ERK1/2. T-ERK1/2 was used as a loading control, and the corresponding histograms represent the normalization t-ERK1/2. Data are expressed as the mean ± sem of three different experiments done in triplicate. *, P < 0.05. Note that treatment with Ctx failed to induce ERK1/2 phosphorylation in the HES cells in contrast to the CHO-LH cells. Additionally, both Ptx and AG1478 failed to inhibit CG-induced ERK1/2 phosphorylation in HES cells but significantly decreased ERK1/2 phosphorylation in CHO-LH cells.

Figure 3

PI3K-dependent activation of ERK1/2, Akt, and c-Raf by CG in HES cells. Sixty percent visually confluent HES cells (A and C) and CHO-LH cells (B and D) were treated for 10 min with 10 IU/ml CG after pretreatment with 1 or 20 μm LY29400 for 1 h. Thirty micrograms of proteins were separated by SDS-PAGE and probed using p-ERK1/2 antibody (A and B) or phosphorylated Akt (p-Akt) Ser473 antibody (C and D). E, Sixty percent visually confluent HES cells were treated for 10min with 10 IU/ml CG with or without pretreatment with 1 or 20 μm LY29400 for 1 h. Thirty micrograms of proteins were separated by SDS-PAGE and probed using phosphorylated c-Raf antibody. The blots were reprobed with t-ERK1/2 or β-actin as loading control, and the histograms represent the normalization to β-actin or t-ERK1/2. Note the PI3K-dependent phosphorylation of ERK1/2, Akt in both cell types, and of c-Raf in HES cells. Values are expressed as the mean ± sem (n = 3). *, P < 0.05.

Figure 4

A, Nuclear translocation of p-ERK1/2. CG induced nuclear translocation of p-ERK1/2 in HES cells: A, Fractionation of cellular components was carried out as indicated in Materials and Methods. Western blot analysis of p- and t-ERK1/2 in 50 μg nuclear (Nuc Ext) and 40 μg of cytoplasmic extracts (Cyt Ext) of HES cells treated with 10 IU/ml CG for the times indicated are shown. Lanes 1–5 represent increasing treatment times of 0, 2, 10, 30, and 60 min. B, Fifty micrograms of nuclear and 40 μg of cytoplasmic extracts were analyzed by Western blot using antibodies against c-Myc and β-tubulin (β-Tub) to verify for purity of the cellular fractions. C, Immunofluorescent imaging of p-ERK1/2 upon CG stimulation: 90% confluent HES cells were treated with 10 IU/ml CG for 0, 30, and 60 min. They were analyzed using an immunocytochemistry-compatible p-ERK1/2 antibody and visualized using confocal microscopy. The green indicates p-ERK1/2 staining in the upper panels, and the lower panels in blue show an overlay with nuclear staining by DAPI. The nuclear localization of p-ERK1/2 after treatment with CG at 30 and 60 min are indicated by the arrows. All images are at the same magnification; the bar (A), 20 μm. The inset represents the IgG control. The experiments have been performed in triplicate and the images are the best representation of the data.

Figure 5

Activation of the nuclear transcription factor Elk1 by CG. Sixty percent visually confluent HES cells were stimulated by 10 and 50 IU/ml of CG (lanes 2 and 3) for 10 min. In addition, cells were pretreated for 1 h with 1 or 20 μm PI3K inhibitor LY29400 (lanes 4, 5, 6) (A), or 1 or 10 μm MEK inhibitor, PD98059 (lanes 4, 5, 6) (B) before stimulation with 10 IU/ml CG for 10 min. After treatment, 30 μg of total cell lysate were analyzed by Western blot using an antibody against phosphorylated Elk1 (p-Elk1). The blots were reprobed with β-actin to confirm equal protein loading. The histograms represent the normalization to β-actin, and data are expressed as the mean sem of three different experiments done in triplicate. *, P < 0.05. Note the dose-dependent increase in Elk1 phosphorylation by CG mediated by the PI3K-ERK1/2 pathway.

Figure 6

mPTGES induction in response to CG stimulation. A, HES cells were cotransfected with either PTGES-pGl4 luciferase reporter plasmid and the Renilla luciferase or the pGl4 vector alone with the Renilla luciferase. Reporter activity was measured after the cells were subjected to stimulation by 10 IU/ml CG for 24 h, with or without a 1-h preincubation with 20 μm PI3K inhibitor LY29400 (LY) or 10 μm MEK inhibitor PD98059 (PD). Activity is represented as relative luciferase units (RLU) derived on normalization with renilla luciferase units. Data are expressed as the mean sem of three different experiments done in triplicate. *, P < 0.05. B, Immunofluorescent imaging of HES cells using an antibody against mPTGES: near confluent HES cells, grown on chamber slides, were treated with 10 IU/ml CG for 24 h, with or without a 1-h pretreatment with 20 μm LY29400 or a 1-h pretreatment with 10 μm PD98059 before CG treatment. The IgG control is shown in the inset. The cells were incubated and analyzed using a polyclonal antibody against mPTGES and visualized using confocal microscopy. The green color represents mPTGES and the bottom panels depict an overlay with DAPI, representing nuclear staining. The arrows indicate positive staining for mPTGES. All images are at the same magnification. Bar, 20 μm. CT, Control.