Metformin inhibits StAR expression in human endometriotic stromal cells via AMPK-mediated disruption of CREB-CRTC2 complex formation.

ORIGINAL E n d o c r i n e

ARTICLE R e s e a r c h

Metformin Inhibits StAR Expression in Human Endometriotic Stromal Cells via AMPK-Mediated Disruption of CREB-CRTC2 Complex Formation Jia-Ning Xu, Cheng Zeng, Yan Zhou, Chao Peng, Ying-Fang Zhou, and Qing Xue Department of Obstetrics and Gynecology, Peking University First Hospital, Beijing 100034, People’s Republic of China Context: Endometriosis is an estrogen-dependent disease affecting reproductive women. Metformin could have a therapeutic effect on endometriosis through regulation of local estrogen production. Objects: The aim of this study was to investigate the molecular and cellular mechanism by which metformin regulates StAR expression in human endometriotic stromal cells (ESCs). Methods: ESCs derived from ovarian endometriomas were cultured with metformin and prostaglandin E2 (PGE2). StAR mRNA was measured by quantitative PCR; pregnenolone, progesterone, and estrogen production were measured by ELISA kits; steroidogenic acute regulatory protein (StAR), AMP-activated protein kinase, cAMP response element binding protein (CREB), and CREBregulated transcription coactivator 2 (CRTC2) protein expression were measured by Western blot assay; and CRTC2 translocation and its association with CREB were assessed by coimmunoprecipitation assay and CRTC2-CREB complex binding by a chromatin immunoprecipitation assay. Results: 1) StAR mRNA levels in ESCs are 264 times higher than those in endometrial cells. 2) Metformin downregulates the StAR mRNA expression (maximum 31.7%) stimulated by PGE2 (2.4fold) in ESCs. 3) PGE2 induces CRTC2 translocation and enhances its association with CREB to form a transcription complex that binds to the StAR promoter region. 4) Metformin prevents the nuclear translocation of CRTC2 by increasing AMP-activated protein kinase phosphorylation. This inhibits transcription of StAR by disrupting formation of the CREB-CRTC2 complex, involved in activation of the StAR promoter cAMP response element. Conclusions: We have demonstrated a detailed mechanistic analysis of StAR expression regulated by metformin in ESCs. Our data highlight a role for CRTC2 in the mechanism by which metformin inhibits StAR expression. (J Clin Endocrinol Metab 99: 2795–2803, 2014)

ndometriosis is an estrogen-dependent, chronic inflammatory disease affecting 6% to 10% of women of reproductive age (1). Theories proposed for the development of endometriosis include retrograde menstruation, immune system deficiency (2), an inflammatory environment (3), local estrogen synthesis (4), and epigenetic events (5). However, the precise mechanism remains unclear. A large body of evidence proves that high levels of steroidogenic acute regulatory protein (StAR) and aromatase

E

are expressed in endometriotic stromal cells (ESCs), giving rise to local estrogen production (6, 7). In the de novo estrogen biosynthesis (6), StAR facilitates the entry of cholesterol into the mitochondrion, where cholesterol is converted to pregnenolone by P450 side-chain cleavage (P450scc), which is then converted to progesterone via 3-hydroxysteroid dehydrogenase type 2 (HSD3B2). P450c17 catalyzes progesterone to androstenedione, and aromatase is responsible for the conversion of androstene-

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received February 28, 2014. Accepted May 2, 2014. First Published Online May 13, 2014

Abbreviations: AMPK, AMP-activated protein kinase; ChIP, chromatin immunoprecipitation; CRE, cAMP response element; CREB, CRE binding protein; CRTC, cAMP-regulated transcriptional coactivator; Ct, comparative threshold cycle; DAPI, 4⬘,6-diamidino-2-phenylindole; EM, endometrial cell; ESC, endometriotic stromal cell; HSD17B1, 17␤-hydroxysteroid dehydrogenase type 1; HSD3B2, 3-hydroxysteroid dehydrogenase type 2; MEK, MAPK kinase; P450scc, P450 side-chain cleavage; P450c17, 17-hydroxylase/17–20-lyase; PGE2, prostaglandin E2; qPCR, quantitative PCR; siRNA, small interfering RNA; StAR, steroidogenic acute regulatory protein.

doi: 10.1210/jc.2014-1593

J Clin Endocrinol Metab, August 2014, 99(8):2795–2803

jcem.endojournals.org

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 January 2015. at 07:17 For personal use only. No other uses without permission. . All rights reserved.

2795

2796

Xu et al

Metformin Inhibits StAR Expression in ESCs

dione to estrone, which is further converted to the biologically active estradiol by the enzyme HSD17B1 (17␤-hydroxysteroid dehydrogenase type 1). StAR plays a vital role in steroidogenesis by providing a continuous supply of cholesterol for estradiol production (8). Previous studies have demonstrated that prostaglandin E2 (PGE2)-induced upregulation of StAR gene activity involves phosphorylation of cAMP response element (CRE) binding protein (CREB), recruitment of CREB binding protein, acetylation of histone, and binding of CCAAT/enhancer binding protein to the StAR promoter (9 –11). Metformin, which is widely used as an antidiabetes agent (12), has been demonstrated to exert direct effects on the ovary, inhibiting basal and insulin-stimulated 17-estradiol and progesterone production by human granulosa cells (13, 14) and breast cancer cells (15). Direct effects of metformin on endometriosis are supported by animal and clinical studies, indicating that it could be an effective therapy for this disease (16, 17), as its possible modulatory effect on local steroid production. A family of CREB coactivators has been recently identified as transducers of CREB activity (18, 19). CREB associates with these cAMP-regulated transcriptional coactivators (CRTCs) to form a complex involved in gene expression (20, 21). Studies have demonstrated the involvement of CRTCs (principally CRTC2) in the regulation of cAMP-responsive genes, primarily CYP19A1 promoter activity in MCF-7 cells (22). Furthermore, metformin inhibits the nuclear translocation of CRTC2, which is a direct downstream target of AMP-activated protein kinase (AMPK) family members activated by metformin (15, 23). The aim of this study was to investigate the molecular and cellular mechanism by which metformin regulates StAR expression in human ESCs. We propose that metformin inhibits StAR expression by interacting with CRTC2 and CREB via the AMPK signaling pathway, thereby influencing estrogen production in endometriosis.

Materials and Methods Primary cell culture and experimental agents ESCs and endometrial cells (EMs) were derived from the cyst walls of ovarian endometriomas (n ⫽ 21) and normal endometrial tissues (n ⫽ 8), respectively, which were collected immediately after surgery. The age range of the subjects was 26 to 43 years. All samples were confirmed histologically, and cells digested from the tissue were cultured using a protocol previously described by Ryan et al (24) with minor modifications. Briefly, endometriotic or endometrial tissues were digested with collagenase (1 mg/mL; Sigma-Aldrich) and deoxyribonuclease I (0.1 mg/mL; Sigma-Aldrich). Epithelial cells were removed by filtration of stromal cells through a 75-␮m sieve. Stromal cells were then harvested and cultured in DMEM/F-12 containing 10%


fetal bovine serum (Gibco) to confluence. All cells were passed 1 or 2 times, and we used second-generation ESCs for the following assays. Tissues were collected after obtaining written informed consent, and the study protocol was approved by the Institutional Review Board of Peking University. After serum starvation in serum-free DMEM/F-12 medium for 16 hours, cells were treated with experimental agents at various concentrations and for different times as indicated in the figure legends. The experimental agents were PGE2 (Sigma-Aldrich), 1,1dimethylbiguanide hydrochloride (metformin; Sigma-Aldrich), compound C (an AMPK inhibitor; EMD Millipore), and PD98059 (a MAPK kinase [MEK]/ERK inhibitor; Sigma-Aldrich).

Pregnenolone, progesterone, and estradiol assays Steroid levels in the media were measured after treatments of cells. Briefly, cells were treated with or without PGE2 and metformin for the indicated times in the figure legends. Conditioned culture media were collected. Pregnenolone, progesterone, and estradiol concentrations were determined using ELISA kits (progesterone and estradiol from Cayman Chemical; pregnenolone from Alpha Diagnostic International Inc) according to the manufacturer’s protocol.

RNA extraction and quantitative analysis by realtime PCR Total RNA was extracted from cultured primary stromal cells in 100-mm plates using TRIzol reagent (Life Technologies) and quantified using a NanoDrop 1000 spectrophotometer (Thermo Scientific). Two micrograms of total RNA were used to generate cDNA using High-Capacity RNA-to-cDNA kits (Life Technologies) according to the manufacturer’s instructions. Real-time quantitative PCR (qPCR) was performed using an ABI 7500 sequence detection system and the ABI TaqMan gene expression system (Life Technologies) for StAR and GAPDH. The primers for StAR were forward 5⬘-CTGAGCAGAAGGGTGTCATCAG-3⬘, reverse 5⬘-AGTTTGGTCTTAGAGGGACTTCCA-3⬘, and probe 5⬘-CACTTGCATGGTGCTTCACCCGTTG3⬘. The GAPDH primers were forward 5⬘-GAAGGTGAAGGTCGGAGTC-3⬘, reverse 5⬘-GAAGATGGTGATGGGATTTC-3⬘, and probe 5⬘-CAAGCTTCCCGTTCTCAGCC-3⬘. Other steroidogenic enzymes involved in de novo estrogen biosynthesis were also measured by qPCR. The primers were as follows: P450scc, forward 5⬘-GCTGCATGGGACGTGATTTT-3⬘ and reverse 5⬘-GAGGATGCCACGGTAATCGT-3⬘; HSD3B2, forward 5⬘-AAGGAGATCAGGGCCTTGGA-3⬘ and reverse 5⬘-TGGCAGGCTCTTTTCAGGAA-3⬘; P450c17, forward 5⬘-CTGGCCAAGGAGGTGCTTATT-3⬘ and reverse 5⬘-GAGTCAGCGAAGGCGATACC-3⬘; P450aro, forward 5⬘-CACATCCTCAATACCAGGTCC-3⬘ and reverse 5⬘-CAGAGATCCAGACTCGCATG-3⬘; and HSD17B1, forward 5⬘-GGGAGCGTGGGAGGATTG-3⬘ and reverse 5⬘-CGCACTCGATCAGGCTCAA3⬘. Cycling conditions were 1 cycle at 95°C for 5 minutes followed by 40 cycles at 95°C for 10 seconds, 59°C for 15 seconds, and 72°C for 20 seconds. The comparative threshold cycle (Ct) method (7) was used to analyze the results. All samples were normalized to GADPH mRNA levels.

Western blot analysis Cells were washed in ice-cold PBS and lysed in ice-cold radioimmunoprecipitation assay buffer (1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% SDS) containing 1% protease


doi: 10.1210/jc.2014-1593

inhibitor cocktail (Amresco) and phosphatase inhibitors (Nanjing KeyGen Biotech). Lysates were cleared by centrifuging at 12 000 rpm for 20 minutes. A bicinchoninic acid (BCA) protein assay was then used to quantify protein concentrations (KeyGen Biotech). Thirty micrograms of protein extract were subjected to 10% SDS-PAGE and subsequently electrotransferred to a nitrocellulose membrane. Membranes were incubated for 1 hour at room temperature with Tris-buffered saline (2mM Tris-HCl [pH 8.0] and 15mM NaCl [pH 7.6]) with 0.1% Tween 20 containing 5% nonfat dry milk powder or BSA powder to saturate nonspecific binding sites. Primary antibodies were incubated at the following concentrations: anti–phospho-AMPK␣ (Thr172), 1:1000; AMPK␣, 1:1000; StAR, 1:1000, anti–phospho-CREB(Ser133), 1:500; anti-CREB, 1:1000; (2535, 2603, 8449, 9198, and 9197; all from Cell Signaling Technology); and anti-CRTC2, 1:200 (46272X; Santa Cruz Biotechnology). Anti–␤-actin antibody or anti-GAPDH antibody, 1:1000 (TA-09 and TA-08; ZSGB-BIO) were used as a loading control. An enhanced chemiluminescence detector was used (Thermo Scientific). All protein bands were quantified by densitometry using ImageJ software (National Institutes of Health).


2797

Small interfering RNA knockdown Primary ESCs were cultured in the growth medium as described above to achieve approximately 50% to 60% confluence at the time of transfection. Transfection was performed using a small interfering RNA (siRNA) (Life Technologies) against AMPK␣1 (5⬘-CACGAGGCCAAGAUCCGCUACUACA-3⬘) or a nontargeting negative control siRNA (low GC content; Life Technologies) at a final concentration of 100nM using Lipofectamine RNAiMAX (Life Technologies). Thirty-six hours after transfection, cells were serum starved for 12 hours, treated with relevant reagents, and processed for real-time PCR or Western blot analysis.

Immunofluorescence imaging Primary ESCs were plated directly onto coverslips (24 ⫻ 24 mm). Cells were serum starved for 16 hours and then treated with the experimental agents for the indicated time. Cells were washed with PBS and fixed in 4% paraformaldehyde for 15 minutes. After removal of the fixation solution and washing with PBS, slides were filled with detergent solution containing 0.5%

Figure 1. Differences in StAR mRNA expression between ESCs and EMs. A) StAR mRNA levels in primary ESCs were markedly higher (264-fold) than those in EMs (n ⫽ 6; **, P ⬍ .01, t test). B) After starvation overnight, ESCs were treated with or without metformin (M; 10␮M or 100␮M) for 24 hours and then cultured with or without PGE2 (P; 1␮M) for another 24 hours. StAR mRNA was measured by semiquantitative PCR and expressed as a fold change in expression relative to control values (no treatment) and normalized to GAPDH mRNA expression. Metformin reduced basal StAR mRNA expression (maximum 36%) in a dose-dependent manner. Metformin downregulated StAR mRNA expression (maximum 31.7%), which was upregulated by PGE2 (2.4-fold), in a dose-dependent manner. Values are the mean ⫾ SEM (n ⫽ 4; **, P ⬍ .01, ANOVA). C) After starvation overnight, ESCs were treated with or without metformin (M; 100␮M) for 24 hours and then cultured with or without PGE2 (P; 1␮M) for another 24 hours. The other steroidogenic enzymes involved in de novo estrogen synthesis and regulated by cAMP were also measured by qPCR, normalized to GAPDH mRNA expression (n ⫽ 4; *, P ⬍ .05, ***P ⬍ .001, ANOVA). Abbreviation: CON, control.


2798

Xu et al


Triton X-100 at room temperature. Before adding antibody, nonspecific antibody binding sites were blocked by incubating the cells for 30 minutes in blocking buffer (Bioworld Technology). After this step, the cells were incubated with primary antibodies for 1 hour at room temperature and overnight at 4°C with anti-CRTC2 (1:100; Santa Cruz Biotechnology) diluted in 5% BSA/PBS. Secondary antibodies (1:100 Alexa Fluor 488 goat antimouse IgG; Life Technologies) were incubated for 1 hour at room temperature. Cell nuclei were stained with 4⬘,6-diamidino2-phenylindole (DAPI) (1:300) at room temperature for 5 minutes. Cover glasses were applied by using fluorescence mounting medium (Bioworld Technology). Fluorescence was visualized and captured by a FluoView FV500 confocal laser scanning microscope (Olympus) at ⫻600 magnification. CRTC2 staining appeared green, and DAPI staining appeared blue; the presented images are representative of most the cells examined for each treatment.

Coimmunoprecipitation assay ESCs were lysed with nondenaturing lysis buffer (Applygen Technologies) with 1% protease inhibitors and incubated on ice for 20 minutes. The cell lysates were then centrifuged in a refrigerated microcentrifuge at 12 000g for 15 minutes at 4°C. The supernatants were used either directly for immunoprecipitation or stored at ⫺80°C. For immunoprecipitation, equal amounts of protein (500 ␮g) were first immunoprecipitated with antiCRTC2 (Santa Cruz Biotechnology) at 4°C for 4 to 6 hours, followed by the addition of protein A agarose (Roche) and incubation at 4°C overnight. The immunoprecipitates were collected by centrifugation, washed 3 times with PBS containing 0.5% protease inhibitors, and eluted with SDS-PAGE sample buffer. Immunoprecipitates were then analyzed by Western blotting as described above.

Chromatin immunoprecipitation and real-time PCR Chromatin immunoprecipitation (ChIP) was performed using a ChIP assay kit (Pierce Chromatin Prep Module; Thermo Scientific) according to the manufacturer’s instructions. Briefly, cells were cross-linked by 1% formaldehyde for 10 minutes at room temperature and collected in PBS containing 1% protease inhibitors. The cross-linked cells were then lysed and enzymatically digested by micrococcal nuclease to shear genomic DNA. The following antibodies were used for immunoprecipitation: polyclonal CRTC2 (Santa Cruz Biotechnology), monoclonal CREB (Cell Signaling Technology), and polyclonal rabbit IgG


(Santa Cruz Biotechnology). Protein/DNA complexes were eluted from the beads and treated with proteinase K solution at 65°C for 2.5 hours. Real-time PCR was performed on the purified DNA in the presence of SYBR Master Green Mix (Life Technologies) using the primers 5⬘-CACAAACGGCCAAAGCAG-3⬘ and 5⬘-GAAGGCTGTGCATCATCATC-3⬘ for amplification of the StAR promoter region. The ⌬⌬Ct method was used to analyze real-time ChIP results. In brief, real-time qPCR was performed with a primer set targeting a positive locus and a primer set targeting a negative locus. Each ChIP DNA fraction’s Ct value was normalized to the input DNA fraction Ct value for the same qPCR assay (⌬Ct) to account for differences in chromatin sample preparation: ⌬Ct[normalized ChIP] ⫽ Ct[ChIP] ⫺ (Ct[input] ⫺ log2[input dilution factor]). The difference between the ChIP fraction Ct values (⌬⌬Ct) of the normalized experimental sample (S2) and the control sample (S1) was then determined: ⌬⌬Ct[S2 ⫺ S1] ⫽ ⌬Ct[S2:normalized ChIP] ⫺ ⌬Ct[S1:normalized ChIP]. Finally, the differential occupancy fold change (linear conversion of the second ⌬⌬Ct to yield a fold change in site occupancy) was calculated: fold change in occupancy ⫽ 2(⫺⌬⌬Ct[S2⫺S1]). The real-time PCR products were resolved in 2% agarose gel for evaluation.

Statistical analysis All data are presented as the mean ⫾ SEM. Experiments comparing 2 groups were analyzed using the two-tailed Student’s t test. When more than 2 groups were compared, one-way ANOVA followed by Tukey’s multiple-comparison test was used. Asterisks indicate statistically significant differences: *, P ⬍ .05; **, P ⬍ .01; and ***, P ⬍ .005.

Results StAR mRNA expression in ESCs and EMs TaqMan-based real-time and semiquantitative RTPCR assays were used to quantify mRNA levels of StAR in the 2 types of cell. StAR mRNA levels in primary ESCs were markedly higher (264-fold) than those in primary EMs (Figure 1A; n ⫽ 6; P ⬍ .01, Student’s t test). Metformin attenuates StAR mRNA expression and acts on other steroidogenic enzymes Metformin reduced basal StAR mRNA expression, with a maximal effect (36%) observed at 100␮M. Met-

Figure 2. Effects of metformin on the production of steroid hormone. After starvation overnight, ESCs were treated with or without metformin (M; 100␮M) for 24 hours and then cultured with or without PGE2 (P; 1␮M) for an additional 48 hours. Conditioned media were collected and assayed for pregnenolone (A), progesterone (B), and estradiol (C) by indicated ELISA kit. Results were analyzed from 3 separate experiments using samples from different women and expressed as a fold change compared with control (CON) production. Values are the mean ⫾ SEM (n ⫽ 5; *, P ⬍ .05, **, P ⬍ .01, ANOVA).


doi: 10.1210/jc.2014-1593

formin also downregulated the StAR mRNA expression (maximum 31.7%) that was stimulated by PGE2 (2.4fold) (Figure 1B; n ⫽ 4; P ⬍ .01, ANOVA). In addition, PGE2 significantly stimulated the p450aro (3.9-fold)


2799

mRNA level, whereas induction of P450scc, P450c17, HSD3B2, and HSD17B1 mRNA was modest (1.1- to 1.9-fold; Figure 1C; n ⫽ 4; P ⬍ .005, ANOVA). Metformin attenuated the mRNA levels of these genes

Figure 3. Metformin inhibits StAR expression via the AMPK signaling pathway in ESCs. A) After starvation overnight, ESCs were treated with or without metformin (M; 100␮M) for 5, 10, or 30 minutes or 1, 2, 4, or 8 hours. Whole-cell lysates were prepared and subjected to SDS-PAGE and analyzed by Western blotting with the indicated antibodies. Vehicle-treated samples were included for all time points to demonstrate that the activation of the target protein (AMPK or MEK/ERK) was not due to time in culture. Levels of AMPK or MEK/ERK phosphorylation were normalized to the total amount of AMPK or MEK/ERK and standardized by vehicle control (n ⫽ 3): p-AMPK/AMPK (⫹M):p-AMPK/AMPK (vehicle control) or p-ERK/ERK (⫹M):p-ERK/ERK (vehicle control). B) After starvation overnight, ESCs were preincubated with an AMPK inhibitor (compound C [CC]; 5␮M) or a MEK/ERK inhibitor (PD98059 [PD]; 25␮M) for 1 hour and then treated with or without metformin (M; 100␮M) for 24 hours, after which they were cultured with or without PGE2 (P; 1␮M) for another 24 hours. A TaqMan-based real-time PCR was conducted (n ⫽ 3; *, P ⬍ .05; **, P ⬍ .01; ***, P ⬍ .005, ANOVA). C) AMPK␣1 siRNA (S1) and negative control siRNA (NC) were transfected to ESCs for 36 hours, after which the cells were serum starved for 12 hours. Upper panel, Western blot assay for the knockdown effect of AMPK␣1 siRNA after transfection for 48 hours. Middle and lower panels, Metformin did not inhibit StAR protein or mRNA expression after AMPK siRNA treatment. For the Western blot assay, cells were treated with or without metformin for 24 hours, after which they were cultured with or without PGE2 for another 48 hours. For the mRNA assay, cells were treated with or without metformin for 24 hours then cultured with or without PGE2 for another 24 hours (n ⫽ 3; *, P ⬍ .05, ANOVA). Abbreviation: CON, control.


2800

Xu et al


except for HSD3B2 (Figure 1C; n ⫽ 4; P ⬍ .05, ANOVA). Effects of metformin on the production of steroid hormone in ESCs Levels of pregnenolone, progesterone, and estradiol were higher in PGE2-treated ESCs compared with vehicle control. Treatment of ESCs with metformin significantly decreased the estradiol production and made a slight reduction of pregnenolone in culture media (Figure 2, A–C; n ⫽ 5; P ⬍ .05 and P ⬍ .01, ANOVA). Metformin inhibits StAR expression via the AMPK signaling pathway in ESCs To identify the signaling events involved in the regulation of StAR expression by metformin, we investigated the phosphorylation of molecules in a subsequent signaling pathway. The AMPK pathway was phosphorylated gradually after metformin stimulation, peaking at 2 to 4 hours


(Figure 3A; n ⫽ 3). We then observed that PGE2 increased StAR mRNA expression and that metformin reduced this induction in ESCs, whereas addition of an AMPK inhibitor (compound C) reversed the reduction by metformin (Figure 3B; n ⫽ 3; P ⬍ .01, ANOVA). To confirm the role of AMPK in the regulation of StAR mRNA expression by metformin, we used siRNA to knock down the AMPK␣1 protein. Using ImageJ software to analyze immunoblotting bands, we first demonstrated that AMPK␣1 protein expression was reduced by almost 50%. And transfection of ESCs with siAMPKa1 reversed (not further suppressed) metformin downregulation of StAR mRNA and protein (Figure 3C; n ⫽ 3; P ⬍ .01, ANOVA). Inhibition of StAR expression by metformin is MEK/ERK-independent Treatment of ESCs with metformin for indicated times induced MEK/ERK phosphorylation after 10 minutes (Figure 3A; n ⫽ 3). To examine the effect of ERK phosphorylation on StAR gene expression, we cotreated ESCs with or without PGE2, metformin, and a MEK inhibitor (PD98059; 25␮M). As noted previously, compound C reversed metformin-decreased StAR expression, whereas PD98059 had no such effect on StAR mRNA expression (Figure 3B; n ⫽ 3; P ⬍ .05, P ⬍ .01, and P ⬍ .005, ANOVA).

Figure 4. Metformin inhibits nuclear translocation of CRTC2 by activation of AMPK in ESCs. After starvation overnight, ESCs were treated with or without metformin (M) for 24 hours, after which they were cultured with or without PGE2 (P) for another 24 hours. Top left, CRTC2 (green) located in both the cytoplasm and the nucleus under resting conditions. Top right, PGE2 induced nuclear translocation of CRTC2. Bottom left, Treatment with 100␮M metformin prevented PGE2mediated nuclear translocation of CRTC2. Bottom right, Prevention of CRTC2 nuclear translocation by metformin was weakened by preculturing ESCs with an AMPK inhibitor (compound C [CC]). Inset shows nucleus-specific DAPI immunofluorescence (blue) (n ⫽ 3). Abbreviation: CON, control.

Metformin limits nuclear translocation of CRTC2 by activation of AMPK in ESCs CRTC2 has been reported to be translocated into the nucleus, where it forms a transcription complex with CREB (20). Increased phosphorylation and hence activation of AMPK have been shown to inhibit nuclear translocation of CRTC2 in human breast stromal cells (22). We employed immunofluorescence and confocal microscopy to observe CRTC2 translocation. As previously demonstrated, CRTC2 was present in both the cytoplasm and the nucleus in unstimulated cells (Figure 4, top left). Treatment with PGE2 resulted in translocation of CRTC2 to the nucleus (Figure 4, top right). Ad-


doi: 10.1210/jc.2014-1593

dition of metformin altered the subcellular localization of CRTC2, which was partly confined to the cytoplasm (Figure 4, bottom left). In addition, preincubation with an AMPK inhibitor (compound C) limited the effect of metformin on CRTC2 translocation (Figure 4, bottom right; n ⫽ 3). Metformin disrupts the interaction between CREB and its coactivator CRTC2 and prevents them binding to the StAR promoter region To determine whether CREB and CRTC2 were physically associated in PGE2-treated human ESCs and influenced by metformin treatment, coimmunoprecipitation assays were conducted. First, we demonstrated that there


2801

was no difference between the protein basal levels of CRTC2 and CREB in EMs and ESCs (Figure 5A; n ⫽ 3) and no marked change in the expression of either protein after treatment (Figure 5B; n ⫽ 3). We then observed an increase in complex formation between CRTC2 and CREB with PGE2 treatment (1–24 hours), which was significantly decreased with metformin treatment (12– 48 hours) (Figure 5C; n ⫽ 3). To further investigate the regulation by metformin of CREB-CRTC2 complex recruitment to the StAR promoter region, ChIP assays were conducted in conjunction with real-time PCR. The proximal StAR promoter region containing a CRE half-site (10) of 109 bp was amplified. Using a ⌬⌬Ct method to analyze the real-time ChIP results, we found that PGE2 stimulated the binding of CRTC2 (2 ⫾ 0.3-fold) and CREB (3.8 ⫾ 0.9-fold) to the StAR promoter region. The effect of PGE2 on CRTC2 and CREB was blocked by metformin, with decreases of 35% and 54%, respectively (Figure 5D; n ⫽ 3; P ⬍ .05, ANOVA).

Discussion

Figure 5. Metformin inhibits the interaction between CREB and its coactivator CRTC2 and prevents them binding to the StAR promoter region. A) There was no difference between CRTC2 and CREB protein expression in EMs and ESCs (n ⫽ 3). B) After starvation overnight, cells were treated with or without metformin (M) for 24 hours, after which they were cultured with or without PGE2 (P) for another 24 hours. None of the treatments altered the basal levels of CRTC2 or CREB protein (n ⫽ 3). C) After starvation overnight, cells were treated for different times. The upper panel represents short-period treatment of cells that were incubated with or without metformin for 11 hours and then cultured with or without PGE2 for another 1 hour; the lower panel represents long-period treatment of cells that were incubated with or without metformin for 24 hours and then cultured with or without PGE2 for another 24 hours. After all treatments, cell lysates were collected then immunoprecipitated with anti-CRTC2 antibody and analyzed by Western blotting with anti-CREB antibody and phosphorylated CREB (p-CREB) antibody. Association between p-CREB (the activated form of CREB) or CREB and CRTC2 was increased by PGE2 treatment; however, metformin attenuated the interaction (n ⫽ 3). D) After starvation overnight, cells were treated with or without metformin for 11 hours and then cultured with or without PGE2 for another 1 hour. Cells were harvested and subjected to a ChIP assay with CRTC2 and CREB antibody and IgG as a control. Metformin prevented the binding of CRTC2 and CREB to the StAR promoter region, whereas PGE2 enhanced their binding. Upper panel, Electrophoretic analysis of qPCR products. Bottom panel, ChIP qPCR analysis (n ⫽ 3; *P ⬍ .05, ANOVA). Abbreviations: CON, control; IB, immunoblot; IP, immunoprecipitation.

In this study, we have shown that PGE2 stimulated StAR transcription by enhancing the binding of both CREB and its coactivator CRTC2 to the StAR promoter region, where they stimulate CRE activation. Furthermore, metformin prevented the nuclear translocation of CRTC2 by increasing AMPK phosphorylation, weakening the association between CREB and CRTC2 as a transcription complex (Figure 6). PGE2 increases intracellular cAMP levels via the PGE2 receptor 2 (EP2). Activation of protein kinase A results in phosphorylation of CREB, allowing its binding to a CRE on the promoter region of the StAR gene (10). Our data show that PGE2 induced the translocation of CRTC2 and enhanced its association with CREB as a transcription complex that binds to the StAR promoter region. Our findings indicated that PGE2 induced the expression of StAR, P450scc, HDS3B2, HSD17B1, and


2802

Xu et al


Figure 6. Model of downregulation by metformin of StAR expression in ESCs. PGE2 directly increases intracellular cAMP levels via the PGE2 receptor 2 (EP2). Activation of protein kinase A (PKA) results in phosphorylation of cAMP response element binding protein (p-CREB), which allows it to bind to a CRE on the promoter region of the StAR gene. We showed that PGE2 induced the translocation of CRTC2 and enhanced its association with CREB to form a transcription complex that binds to the StAR promoter region. We also demonstrated that metformin treatment led to an increase in AMPK phosphorylation that prevented retention of CRTC2 in the cytosol. This inhibited transcription of StAR by disrupting formation of the CREB-CRTC2 complex, which activates the CRE on the StAR promoter. CDS, coding sequence.

P450aro to catalyze the conversion of cholesterol to estrogen in ESCs. PGE2 also stimulated pregnenolone and estrone production through additional upregulation of StAR and P450aro, the key enzymes of de novo estrogen biosynthesis. In addition, metformin can not only downregulate StAR mRNA expression but also have an inhibitory effect on the expression of other steroidogenic enzymes except for HSD3B2. A decrease in the conversion of cholesterol to pregnenolone and other steroid hormones by metformin proved its actual biological effect. And by assessing whether the same mechanism applies to additional enzymes in the local estrogen production pathway enhanced the relevance of this study. After oral administration, plasma levels of metformin range between 10⫺4M and 10⫺5M (1 ␮g/mL) (25). We found that the decrease in StAR expression induced by metformin was dose-dependent and was greatest at a concentration of 10⫺4 M to 10⫺5 M. This differs from other studies (26, 27) conducted with extremely high concentrations of metformin that would never be reached therapeutically. We also found that the AMPK pathway was gradually phosphorylated after treatment with metformin, reaching a peak by 2 to 4 hours, which is similar to the findings of a previous study (28). In our study, activation of the AMPK signaling pathway is necessary for the reduction of StAR mRNA and protein expression by metformin. However, it has been reported that, in human granulosa cells, metformin negatively regulated production of aromatase, another key enzyme in steroidogenesis, by activating the MEK/ERK pathway (13). Interestingly, we found that metformin induced ERK phosphorylation, but inhibition of the MEK/ERK pathway


had no effect on blocking metformin-decreased StAR expression. Activation of MEK/ERK pathway may play a role in other aspects of metformin’s function. The discrepancy may be due, in part, to differences in the cells or ligands used, and further investigations are necessary to resolve this disparity and explore the specific mechanism. The activation of transcription factors is so rapid that appropriate treatment times for the following assay had to be selected carefully. We demonstrated that metformin treatment interrupted the association between CRTC2 and CREB from 12 to 48 hours while not altering the basal level of these 2 proteins, which indicates that complex formation was affected and not the total levels of the proteins. For the assay of binding ability, we used real-time PCR, a semiquantitative method, to analyze ChIP results, which is more accurate and direct than electrophoretic analysis. We found that metformin prevented CRTC2 and CREB binding to the StAR promoter region, with decreases of 35% and 54%, respectively, which was enhanced by PGE2. Three major sites in the body produce estrogen in women with endometriosis. First, estradiol secreted by the ovary is transported to endometriotic tissue. Second, cholesterol is converted to estradiol locally. Third, aromatase in adipose tissue and the skin catalyzes the conversion of circulating androstenedione to estrone, which is subsequently converted to estradiol (5). The first 2 sources of estrogen are mainly responsible for the development of endometriosis. In addition to our study, it has been found that PGE2-stimulated aromatase activity was suppressed by metformin in ESCs (29). Thus, metformin could be expected to suppress estrogen levels in endometriotic tissues. Interestingly, metformin also inhibited FSH, insulinstimulated progesterone, and estradiol production in granulosa cells (14). In this way, metformin may provide an effective treatment for endometriosis through suppression of both ovarian and local production of estrogens. It has been shown that metformin promotes progesterone receptor expression in endometrial cancer cells (30). This sheds light on progestin resistance in endometriosis. It may also promote the development of novel treatment strategies aimed at molecular targets involved in signaling pathways, chromatin remodeling, and gene regulation in endometriosis. We have demonstrated a direct effect of metformin, but there may be other mechanisms responsible for its action. Can metformin inhibit cell proliferation or prevent apoptosis through such a signaling pathway? In addition, it remains possible that the responses of endometriotic epithelial cells or other types of endometriotic cell to metformin may differ from the present results. In summary, we have demonstrated regulation of StAR expression by metformin in ESCs. These findings have


doi: 10.1210/jc.2014-1593

implications for the use of metformin in the treatment of women with endometriosis.

Acknowledgments We appreciate Dr Yu Qi and Prof Ding-Fang Bu for their generous advice regarding the study. Address all correspondence and requests for reprints to: Ying-Fang Zhou and Qing Xue, No. 1 Xi’anmen Street, 100034 Beijing, China. E-mail: [email protected] and [email protected]. This work was supported by the National Natural Science Foundation of China (Grant 81270674) and the Natural Science Foundation of Beijing, China (Grant 7132204). J.-N.X. designed and performed experiments, collected and analyzed data, cultured cells, performed the literature review, and drafted the manuscript. C.Z. performed experiments, collected data, and cultured cells. Y.Z. performed experiments and cultured cells. C.P. collected data. Y.-F.Z. developed the concept and edited the manuscript. Q.X. developed the concept and designed experiments, administered the experiment, and edited the manuscript. All authors read and approved the final manuscript. Disclosure Summary: The authors have nothing to declare.

References 1. Giudice LC, Kao LC. Endometriosis. Lancet. 2004;364(9447): 1789 –1799. 2. Olive DL, Schwartz LB. Endometriosis. N Engl J Med. 1993; 328(24):1759 –1769. 3. Hirota Y, Osuga Y, Hirata T, et al. Activation of protease-activated receptor 2 stimulates proliferation and interleukin (IL)-6 and IL-8 secretion of endometriotic stromal cells. Hum Reprod. 2005;20(12): 3547–3553. 4. Attar E, Bulun SE. Aromatase and other steroidogenic genes in endometriosis: translational aspects. Hum Reprod Update. 2006; 12(1):49 –56. 5. Bulun SE. Endometriosis. N Engl J Med. 2009;360(3):268 –279. 6. Attar E, Tokunaga H, Imir G, et al. Prostaglandin E2 via steroidogenic factor-1 coordinately regulates transcription of steroidogenic genes necessary for estrogen synthesis in endometriosis. J Clin Endocrinol Metab. 2009;94(2):623– 631. 7. Xue Q, Lin Z, Yin P, et al. Transcriptional activation of steroidogenic factor-1 by hypomethylation of the 5⬘ CpG island in endometriosis. J Clin Endocrinol Metab. 2007;92(8):3261–3267. 8. Bulun SE, Lin Z, Imir G, et al. Regulation of aromatase expression in estrogen-responsive breast and uterine disease: from bench to treatment. Pharmacol Rev. 2005;57(3):359 –383. 9. Tsai SJ, Wu MH, Lin CC, Sun HS, Chen HM. Regulation of steroidogenic acute regulatory protein expression and progesterone production in endometriotic stromal cells. J Clin Endocrinol Metab. 2001;86(12):5765–5773. 10. Sun HS, Hsiao KY, Hsu CC, Wu MH, Tsai SJ. Transactivation of steroidogenic acute regulatory protein in human endometriotic stromal cells is mediated by the prostaglandin EP2 receptor. Endocrinology. 2003;144(9):3934 –3942. 11. Hsu CC, Lu CW, Huang BM, Wu MH, Tsai SJ. Cyclic adenosine 3⬘,5⬘-monophosphate response element-binding protein and CCAAT/enhancer-binding protein mediate prostaglandin E2-induced steroidogenic acute regulatory protein expression in endometriotic stromal cells. Am J Pathol. 2008;173(2):433– 441.


2803

12. Goodarzi MO, Bryer-Ash M. Metformin revisited: re-evaluation of its properties and role in the pharmacopoeia of modern antidiabetic agents. Diabetes Obes Metab. 2005;7(6):654 – 665. 13. Rice S, Pellatt L, Ramanathan K, Whitehead SA, Mason HD. Metformin inhibits aromatase via an extracellular signal-regulated kinasemediated pathway. Endocrinology. 2009;150(10):4794 – 4801. 14. Rice S, Elia A, Jawad Z, Pellatt L, Mason HD. Metformin inhibits follicle-stimulating hormone (FSH) action in human granulosa cells: relevance to polycystic ovary syndrome. J Clin Endocrinol Metab. 2013;98(9):E1491–E1500. 15. Brown KA, Hunger NI, Docanto M, Simpson ER. Metformin inhibits aromatase expression in human breast adipose stromal cells via stimulation of AMP-activated protein kinase. Breast Cancer Res Treat. 2010;123(2):591–596. 16. Oner G, Ozcelik B, Ozgun MT, Serin IS, Ozturk F, Basbug M. The effects of metformin and letrozole on endometriosis and comparison of the two treatment agents in a rat model. Hum Reprod. 2010; 25(4):932–937. 17. Soares SR, Martínez-Varea A, Hidalgo-Mora JJ, Pellicer A. Pharmacologic therapies in endometriosis: a systematic review. Fertil Steril. 2012;98(3):529 –555. 18. Iourgenko V, Zhang W, Mickanin C, et al. Identification of a family of cAMP response element-binding protein coactivators by genomescale functional analysis in mammalian cells. Proc Natl Acad Sci U S A. 2003;100(21):12147–12152. 19. Luo Q, Viste K, Urday-Zaa JC, et al. Mechanism of CREB recognition and coactivation by the CREB-regulated transcriptional coactivator CRTC2. Proc Natl Acad Sci U S A. 2012;109(51):20865–20870. 20. Altarejos JY, Montminy M. CREB and the CRTC co-activators: sensors for hormonal and metabolic signals. Nat Rev Mol Cell Biol. 2011;12(3):141–151. 21. Ravnskjaer K, Kester H, Liu Y, et al. Cooperative interactions between CBP and TORC2 confer selectivity to CREB target gene expression. EMBO J. 2007;26(12):2880 –2889. 22. Brown KA, McInnes KJ, Hunger NI, Oakhill JS, Steinberg GR, Simpson ER. Subcellular localization of cyclic AMP-responsive element binding protein-regulated transcription coactivator 2 provides a link between obesity and breast cancer in postmenopausal women. Cancer Res. 2009;69(13):5392–5399. 23. Tosca L, Chabrolle C, Uzbekova S, Dupont J. Effects of metformin on bovine granulosa cells steroidogenesis: possible involvement of adenosine 5⬘ monophosphate-activated protein kinase (AMPK). Biol Reprod. 2007;76(3):368 –378. 24. Ryan IP, Schriock ED, Taylor RN. Isolation, characterization, and comparison of human endometrial and endometriosis cells in vitro. J Clin Endocrinol Metab. 1994;78(3):642– 649. 25. Wiernsperger NF. Membrane physiology as a basis for the cellular effects of metformin in insulin resistance and diabetes. Diabetes Metab. 1999;25(2):110 –127. 26. Kim HG, Hien TT, Han EH, et al. Metformin inhibits P-glycoprotein expression via the NF-␬B pathway and CRE transcriptional activity through AMPK activation. Br J Pharmacol. 2011;162(5):1096–1108. 27. Jung JW, Park SB, Lee SJ, Seo MS, Trosko JE, Kang KS. Metformin represses self-renewal of the human breast carcinoma stem cells via inhibition of estrogen receptor-mediated OCT4 expression. PLoS One. 2011;6(11):e28068. 28. Lim JY, Oh MA, Kim WH, Sohn HY, Park SI. AMP-activated protein kinase inhibits TGF-␤-induced fibrogenic responses of hepatic stellate cells by targeting transcriptional coactivator p300. J Cell Physiol. 2012;227(3):1081–1089. 29. Takemura Y, Osuga Y, Yoshino O, et al. Metformin suppresses interleukin (IL)-1␤-induced IL-8 production, aromatase activation, and proliferation of endometriotic stromal cells. J Clin Endocrinol Metab. 2007;92(8):3213–3218. 30. Xie Y, Wang YL, Yu L, et al. Metformin promotes progesterone receptor expression via inhibition of mammalian target of rapamycin (mTOR) in endometrial cancer cells. J Steroid Biochem Mol Biol. 2011;126(3–5):113–120.


Metformin Suppresses Prostaglandin E2-Induced Cytochrome P450 Aromatase Gene Expression and Activity via Stimulation of AMP-Activated Protein Kinase in Human Endometriotic Stromal Cells.

Drospirenone induces decidualization in human eutopic endometrial stromal cells and reduces DNA synthesis of human endometriotic stromal cells.

Metformin anti-tumor effect via disruption of the MID1 translational regulator complex and AR downregulation in prostate cancer cells.

Activated Platelets Induce Estrogen Receptor β Expression in Endometriotic Stromal Cells.

MicroRNA-34a inhibits osteoblast differentiation and in vivo bone formation of human stromal stem cells.

in vitro Anti-Proliferative Effect of Adiponectin on Human Endometriotic Stromal Cells through AdipoR1 and AdipoR2 Gene Receptor Expression.

Apoptosis induction of human endometriotic epithelial and stromal cells by noscapine.

KAI1 Expression in Human Lung Carcinoma Cells.

Metformin inhibits expression and secretion of PEDF in adipocyte and hepatocyte via promoting AMPK phosphorylation.

Arcyriaflavin a, a cyclin D1-cyclin-dependent kinase4 inhibitor, induces apoptosis and inhibits proliferation of human endometriotic stromal cells: a potential therapeutic agent in endometriosis.

Rotavirus Genomic RNA Complex Forms via Specific RNA-RNA Interactions: Disruption of RNA Complex Inhibits Virus Infectivity.

Metformin inhibits ALK1-mediated angiogenesis via activation of AMPK.

Metformin inhibits mitochondrial complex I of cancer cells to reduce tumorigenesis.

MicroRNA expression analysis in endometriotic serum treated mesenchymal stem cells.

Metformin inhibits growth of lung adenocarcinoma cells by inducing apoptosis via the mitochondria-mediated pathway.

HBx inhibits CYP2E1 gene expression via downregulating HNF4α in human hepatoma cells.

Anti-Endometriotic Effects of Pueraria Flower Extract in Human Endometriotic Cells and Mice.

MicroRNA expression in bone marrow-derived human multipotent Stromal cells.

Transforming growth factor beta expression in human marrow stromal cells.

Functional expression of aryl hydrocarbon receptor on mast cells populating human endometriotic tissues.

Metformin inhibits hepatitis B virus protein production and replication in human hepatoma cells.

Metformin inhibits histone H2B monoubiquitination and downstream gene transcription in human breast cancer cells.

Exosomes derived from endometriotic stromal cells have enhanced angiogenic effects in vitro.

Metformin inhibits growth of human glioblastoma cells and enhances therapeutic response.