G Protein-coupled Estrogen Receptor Mediates the Up-regulation of Fatty Acid Synthase Induced by 17β-Estradiol in Cancer Cells and Cancer-associated Fibroblasts*

Background: Fatty acid synthase (FASN) is a key lipogenic enzyme regulated by various factors, including estrogens. Results: GPER mediates FASN expression and activity induced by estrogens in cancer cells. Conclusion: Fatty acid biogenesis is regulated by estrogens through GPER. Significance: GPER may be included among the transduction mediators involved by estrogens in regulating FASN expression and activity. Activation of lipid metabolism is an early event in carcinogenesis and a central hallmark of many tumors. Fatty acid synthase (FASN) is a key lipogenic enzyme catalyzing the terminal steps in the de novo biogenesis of fatty acids. In cancer cells, FASN may act as a metabolic oncogene, given that it confers growth and survival advantages to these cells, whereas its inhibition effectively and selectively kills tumor cells. Hormones such as estrogens and growth factors contribute to the transcriptional regulation of FASN expression also through the activation of downstream signaling and a cross-talk among diverse transduction pathways. In this study, we demonstrate for the first time that 17β-estradiol (E2) and the selective GPER ligand G-1 regulate FASN expression and activity through the GPER-mediated signaling, which involved the EGF receptor/ERK/c-Fos/AP1 transduction pathway, as ascertained by using specific pharmacological inhibitors, performing gene-silencing experiments and ChIP assays in breast SkBr3, colorectal LoVo, hepatocarcinoma HepG2 cancer cells, and breast cancer-associated fibroblasts. In addition, the proliferative effects induced by E2 and G-1 in these cells involved FASN as the inhibitor of its activity, named cerulenin, abolished the growth response to both ligands. Our data suggest that GPER may be included among the transduction mediators involved by estrogens in regulating FASN expression and activity in cancer cells and cancer-associated fibroblasts that strongly contribute to cancer progression.

gen-responsive elements located within the promoter of target genes (3)(4)(5). In addition, an increasing number of evidence has recently demonstrated that the G protein-coupled receptor named GPER functions as an estrogen receptor in normal and cancer cells (6 -9). Indeed, GPER is widely distributed in neural, placental, hearth, prostate, hepatic, bone, vascular epithelial, lymphoid, and reproductive tissues as well as in breast, endometrial, ovarian, and thyroid carcinomas (10 -14). Several studies, including our own (15)(16)(17)(18)(19)(20)(21), have shown that GPER mediates estrogen (17␤-estradiol) signals activating the epidermal growth factor receptor (EGFR)/ERK/AP1 transduction pathway (22)(23)(24)(25)(26)(27)(28). In this context, it has been reported that GPER stimulates through G␣ s the cAMP pathway and through G␤␥ the Src activity, which leads to the shedding of heparin binding EGF and the activation of EGFR (12). As a consequence, several signaling cascades such as ERK, PI3K, and phospholipase C are engaged in the stimulation of downstream biological responses, including gene expression changes, cell proliferation, and migration (6,14). One main metabolic change in cancer cells is represented by an altered lipogenic pathway such as an increased synthesis of fatty acids, which are important substrates in the energy production, building blocks of cellular membranes, intracellular second messengers, and anchorage for membrane proteins (29). Free fatty acids derive from both the diet and de novo synthesis, which is catalyzed in lipogenic tissues by fatty acid synthase (FASN) that is able to generate palmitate from malonyl-CoA and acetyl-CoA in the presence of NADPH (29,30). In normal cells, FASN expression is relatively low and occurs in liver and adipose tissues mainly through nutritional signals; conversely, in cancer cells, FASN levels are elevated and independent of nutritional signals (31). FASN has been strongly associated with cell proliferation, aggressiveness, and metastasis in different types of tumors and considered predictive of poor prognosis in diverse malignancies (32). Although the mechanisms involved in the up-regulation of FASN in tumor cells remain to be completely understood, an intricate interplay between estrogen signaling and FASN function has been found in breast tumors (33). In the present study, we demonstrate for the first time that E2 regulates FASN expression and function through GPER in different types of cancer cells that do not express ERs. On the basis of our results, GPER signaling may be included among the transduction pathways by which E2 triggers fatty acid biogenesis, which strongly contributes to the development and aggressive features of diverse tumors.
Cell Cultures-The SkBr3 breast cancer cells were maintained in RPMI 1640 (Invitrogen) without phenol red, supplemented with 10% fetal bovine serum (FBS) and 100 g/ml penicillin/streptomycin. The LoVo colorectal adenocarcinoma cells and the LNCaP prostate cancer cells were maintained in RPMI 1640 with phenol red, supplemented with 10% FBS and 100 g/ml penicillin/streptomycin. The hepatocarcinoma cells HepG2 and the MCF-7 breast cancer cells were cultured in DMEM (Dulbecco's modified Eagle's medium) with phenol red, supplemented with 10% FBS and 100 g/ml penicillin/streptomycin. All cell lines were grown in a 37°C incubator with 5% CO 2 . Cancer-associated fibroblasts (CAFs) were extracted as described previously (25) and maintained in a mixture of MEDIUM 199 and HAM'S F-12 (1:1) supplemented with 10% FBS. Primary cells cultures of breast fibroblasts were characterized by immunofluorescence. Briefly, cells were incubated with human anti-vimentin (V9) and human anti-cytokeratin 14 (LL001); all antibodies were from Santa Cruz Biotechnology, DBA (Milan, Italy). In addition, we used antifibroblast-activated protein ␣ antibody (H-56), also purchased from Santa Cruz Biotechnology, DBA, for fibroblast activation characterization (data not shown).
Gene Expression Studies-Total RNA was extracted using TRIzol commercial kit (Invitrogen) according to the manufacturer's instructions. RNA was quantified spectrophotometrically, and its quality was checked by electrophoresis through agarose gels stained with ethidium bromide. Only samples that were not degraded and showed clear 18 S and 28 S bands under ultraviolet light were used for real-time PCR.
Total cDNA was synthesized from the RNA by reverse transcription using the murine leukemia virus reverse transcriptase (Invitrogen) following the protocol provided by the manufacturer. The expression of selected gene was quantified by realtime PCR using Step One TM sequence detection system (Applied Biosystems, Inc., Milan, Italy), following the manufacturer's instructions. Gene-specific primers were designed using Primer Express software (version 2.0, Applied Biosystems, Inc.) and are as follows. FASN and the ribosomal protein 18 S, which was used as a control gene to obtain normalized values: FASN (human), 5Ј-CATCCAGATAGGCCTCATAGAC-3Ј (forward) and 5Ј-CTCCATGAAGTAGGAGTGGAAG-3Ј (reverse); 18 S (human, mouse), 5Ј-GGCGTCCCCCAACTTCTTA-3Ј (forward) and 5Ј-GGGCATCACAGACCTGTTATT-3Ј (reverse). Assays were performed in triplicate, and the results were normalized for 18 S expression and then calculated as fold induction of RNA expression. For all experiments, cells were switched to medium without serum 24 h before treatments. FASN expression was evaluated also using semiquantitative RT-PCR, as described previously (34).
Gene Silencing Experiments and Plasmids-Cells were plated onto 10-cm dishes, maintained in serum-free medium for 24 h, and then transfected for additional 24 or 48 h before treatments with a control vector or an independent shRNA sequence for each target gene using FuGENE 6 (Roche Diagnostics). The shRNA plasmid for EGFR was purchased from SABioscience Corp. (Frederick, MD). Short hairpin constructs against human GPER (shGPER) were generated and used as described previously (7). The plasmid DN/c-fos, which encodes a c-Fos mutant that heterodimerizes with c-Fos dimerization partners but does not allow DNA binding (35), was a kind gift from Dr. C. Vinson (National Institutes of Health, Bethesda, MD). The expression vector for FLAG-tagged human GPER has been described (15). It was used to generate the GPER rescue vector containing silent mutations in the shRNA-targeted sequence: codons 293-297 were changed to CCGTGTAAACAAAGT. The expression Immunostaining Assay-Fifty percent confluent cultured SkBr3, LoVo, and HepG2 cells, and CAFs grown on coverslips were serum-deprived for 24 h and treated for 18 h with 1 nM E2. Then cells were fixed in 4% paraformaldehyde, permeabilized with 0.2% Triton X-100, washed three times with PBS, and incubated overnight with a mouse primary antibody against FASN (1:500). After incubation, the slides were extensively washed with PBS and incubated with propidium iodide (1:1000 Sigma-Aldrich) for cell nuclei detection and donkey antimouse IgG-FITC (1:250; purchased from Santa Cruz Biotechnology). The Leica AF6000 Advanced Fluorescence Imaging System supported by quantification and image processing software Leica Application Suite Advanced Fluorescence (Leica Microsystems CMS) were used for experiment evaluation.
Chromatin Immunoprecipitation (ChIP) Assay-SkBr3, LoVo, and HepG2 cells, and CAFs were grown in 10-cm dishes to 70 -80% confluence, shifted to serum-free medium for 24 h, and then treated with vehicle, 1 nM E2, and G-1 for 3 h. There-after, cells were cross-linked with 1% formaldehyde and sonicated. Supernatants were immunocleared with sonicated salmon DNA/protein A-agarose (Upstate Biotechnology, Inc., Lake Placid, NY) and immunoprecipitated with the anti-Fos antibody or nonspecific IgG (Santa Cruz Biotechnology). Pellets were washed, eluted with a buffer consisting of 1% SDS and 0.1 mol/liter NaHCO 3 , and digested with proteinase K. DNA was obtained by phenol/chloroform extraction and precipitated with ethanol. A 4-l volume of each sample was used as template to amplify an AP1-containing region corresponding to Ϫ1606/Ϫ1596 located in the 5Ј-flanking region of FASN gene by real-time PCR (Applied Biosystems, Milan, Italy). The primers used were as follows: 5Ј-CTGGCAGCCAGGGCCA-3Ј (forward) and 5Ј-GCTGTGGTTGACGCACGG-3Ј (reverse). To verify the specificity of c-Fos recruitment at the AP1 site, we also performed ChIP assay using the following primers: 5Ј-ACGCTCATTGGCCTGGG-3Ј (forward) and 5Ј-TGGCTC-CCTCTAGGCCGG-3Ј (reverse), which amplify the estrogen target gene SREBP-1c containing region corresponding to Ϫ189/Ϫ171 located in the 5Ј-flanking region of the FASN gene (36). In particular, it was shown that the binding to the SREBP-1c site occurs in an ER-dependent manner upon estrogen stimulation (36). Real-time PCR data were normalized with respect to unprocessed lysates (input DNA). Input DNA quantification was performed by using 4 l of the template DNA. The relative antibody-bound fractions were normalized to a calibrator that was chosen to be the basal, untreated sample. Final results were expressed as percent differences with respect to the relative inputs.
Proliferation Assays-For quantitative proliferation assays, 1 ϫ 10 4 SkBr3, LoVo, and HepG2 cells, and CAFs were seeded in 24-well plates in regular growth medium. Cells were washed once they had attached and further incubated in a medium supplemented with 2.5% charcoal-treated FBS. Ligands were added at this point; medium was changed every day (with ligands and cerulenin were applicable). On day 6 (after 5 days of treatment), cells were trypsinized and counted using Countess TM automated cell counter (purchased from Invitrogen).
Migration Assay-Migration assays were performed using Boyden chambers (Costar Transwell, 8-mm polycarbonate membrane). Cells were seeded in the upper chambers. E2 and G-1 alone or in combination with cerulenin were added to the medium without serum in the bottom wells. After 24 h, cells on the bottom side of the membrane were fixed and counted.
FASN Enzymatic Activity Assay-FASN activity in whole cells was measured by the incorporation of [1,2-4 C]acetate (PerkinElmer Life Sciences) into fatty acids. Cells were plated in six-well plates at 3 ϫ 10 5 cells per dish and incubated overnight. The next day after 12 h of starvation, cells were treated with vehicle, 1 nM E2, 1 M cerulenin, and 1 nM E2 ϩ 1 M cerulenin overnight and then incubated with 0.5 Ci/ml [1,2-14 C]acetate for 8 h. Cells were washed and harvested in 1ϫ PBS and [1,2-14 C] incorporated lipids were extracted with chloroform/methanol (1:4). After centrifugation at 12,000 ϫ g for 10 min, the lower phase containing radiolabeled lipids was counted by scintillation counter. FASN activity was calculated as nmol/mg total protein/min, and variations were reported as fold respect to the vehicle-treated cells. Each experiment was repeated at least in triplicate.
Statistical Analysis-Statistical analysis was performed using analysis of variance followed by Newman-Keuls' testing to determine differences in means. p Ͻ 0.05 was considered as statistically significant.

E2 and G-1 Induce FASN Expression in ER-negative Cancer
Cells-To provide novel insights into the mechanisms by which estrogens may regulate FASN gene in cancer cells, we began the present study evaluating FASN expression upon exposure to E2 and the GPER ligand G-1 in breast SkBr3, colorectal LoVo, and hepatocarcinoma HepG2 tumor cells, and in CAFs that lack the classical ERs but express GPER (supplemental Fig. 1). In time course experiments, E2 and G-1 showed the ability to increase the mRNA expression of FASN, as evaluated by real-time PCR (Fig. 1, A-D) and using a semiquantitative PCR (data not shown) (34). The up-regulation of FASN mRNA was paralleled by increased FASN protein levels upon exposure to E2 and G-1 (Fig. 1E-L), as also evaluated by immunofluorescence studies (Fig. 2).
GPER/EGFR/ERK/c-Fos/AP1 Signaling Mediates FASN Expression Induced by Estrogens-These results prompted us to evaluate the role exerted by GPER signaling in the up-regula-  DECEMBER 21, 2012 • VOLUME 287 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 43239 tion of FASN by E2. Silencing GPER in all cells used, E2 and G-1 did not induce FASN expression, suggesting that GPER mediates this effect. The requirement for GPER and the specificity of the GPER knockdown were further emphasized by the fact that the co-transfection of a shRNA-resistant version of GPER ("GPER rescue") restored the response. (Fig. 3, A, C, E, and G). As GPER activation triggers EGFR signaling (14,15), we next demonstrated that the increase of FASN protein levels induced by E2 and G-1 requires EGFR as determined through gene silencing experiment (Fig. 4, A, C, E, and G). Confirming this observation, E2 and G-1 lost the ability to up-regulate FASN protein expression using the EGFR inhibitor AG1478 as well as the MEK inhibitor PD98059 (Fig. 5, A-D). The latter finding was nicely supported by EGFR activation and the rapid ERK phosphorylation induced by E2 and G-1 in SkBr3, LoVo, and HepG2 cancer cells, and in CAFs (Fig. 5, E-H). As the GPER/ EGFR/ERK transduction signaling triggers c-Fos expression (15)(16)(17)(18)(19)(20)(21)(22), we first ascertained this response to E2 and G-1 (Fig. 6,  A-H) and then we determined that c-Fos is recruited to the AP1 site located within the promoter sequence of FASN (Fig. 6, I-L).

GPER Regulates Fatty Acid Synthase Expression
Amplifying a FASN promoter region containing the SREBP-1c site, which is an ER-mediated estrogen target gene (36), we did not observe the recruitment of c-Fos (data not shown), hence indicating the specificity of its binding to the AP1 site. Moreover, using a dominant-negative variant of c-fos the induction of FASN by E2 and G-1 was no longer evident (Fig. 6, M-P), further confirming the role played by c-Fos in this biological response. Taken together, these findings indicate that the GPER/EGFR/ERK/c-Fos/AP1 transduction pathway mediates the transcription of FASN induced by E2 and G-1 in our model system.

DISCUSSION
FASN is a key lipogenic enzyme that plays a relevant role in cancer pathogenesis and development (33). Accordingly, FASN expression has been found elevated in numerous types of cancer (44 -46) and detected in a most intense manner in carcinomas with higher risk of recurrence and death (29), hence delineating its functional nature of a metabolic oncogene. As it concerns the regulation of FASN levels, steroid hormones, growth factors (for example, EGFR and ERBB2) and the PI3K-Akt signaling pathways were shown to modulate FASN expression (31,(47)(48)(49)(50). However, how FASN is up-regulated in the first place in normal or preneoplastic cells to prime tumorigenesis is currently unclear, and the specific cytotoxicity of FASN inhibition in cancer cells as well as its role in chemotherapeutic resistance remain to be clarified. Overall, the aforementioned data regarding the FASN-dependent fatty acid synthesis in cancer cells make this enzyme as a suitable target for cancer treatment, mainly considering that the silencing of FASN expression inhibits the proliferation and induces apoptosis in cancer cells (37,38). In this regard, it is worth nothing that the pharmacologic inhibitor of FASN activity, cerulenin, induced a selective cytotoxicity in cancer cells by decreasing fatty acid synthesis, which delayed the progression of breast, ovarian, and prostate human cancer xenografts and suppressed liver metastasis in a colon cancer xenograft model (39 -43).
Steroid hormones may have a role in the regulation of FASN expression in hormone-responsive tumors. For example, FASN expression was shown to contribute to the estrogen-driven response, which stimulated the proliferation in hormone-dependent endometrial cells (51). In MCF-7 breast cancer cells, FASN expression was influenced by E2 and progestins through the sterol receptor element binding protein 1 (SREBP-1) pathway as also observed in prostate cancer cells by androgens (52). In these studies, the activation of steroid receptors mediated the up-regulation of FASN as the antiandrogen bicalutamide, the antiprogestin mifepristone (RU486), and the antiestrogens 4-hydroxytamoxifen and faslodex (ICI 182780) inhibited the FASN response to the cognate ligands of hormone receptors (30,36,(53)(54)(55)(56). Nevertheless, the inhibition of MAPK and PI3K signaling pathways abolished FASN induction by steroids (32,51), suggesting that complex transduction mechanisms may contribute to the regulation of FASN expression.
In the context of these findings, our current results provide evidence regarding a new mechanism by which FASN may be regulated in a variety of tumor cells. We demonstrate that E2 and G-1 induce FASN expression and activity through the GPER-mediated signaling, which involves the EGFR/ERK/ c-Fos/AP1 transduction pathway. In particular, we show that the induction of FASN by E2 and G-1 is mediated by sequential events such as the rapid activation of ERK1/2 and the stimulation of c-Fos, which is then recruited to an AP1 site located within the FASN promoter sequence. Of note, FASN was required for important biological responses to E2 and G-1 such as cell proliferation and migration in cancer cells and CAFs lacking the classical ERs but expressing GPER.
Tumor progression is not achieved solely by cancer cells, but neoplastic epithelial cells coexist in carcinomas with several types of stromal cells that generate the microenvironment of the cancer cells (57). Among the stromal components, the most important type of cells recruited into the tumor mass are represented by fibroblasts, which, acquiring an activated phenotype, act as important regulators of the paracrine signals between stromal and cancer cells (58). In particular, the specialized group of fibroblasts, referred to as CAFs, actively contribute to the growth and invasion of tumor cells by providing a unique tumor microenvironment (59). In this regard, it has been reported that CAFs express a wide number of growth factors and extracellular matrix remodeling enzymes that promote the proliferation and invasion of tumor cells as well as angiogenesis and chemoresistance (60,61). In breast carcinoma, ϳ80% of stromal fibroblasts exhibit the activated phenotype that induces the proliferation of cancer cells at the metastatic sites, stimulating the tumor growth such as for the primary tumor (62). In addition, stromal fibroblasts may promote the local production of estrogens, which largely contribute to the progression of breast carcinomas through a signal cross-talk with many transduction pathways activated by growth factors (63). CAFs may trigger tumor progression also through further mechanisms as they facilitate the invasiveness of otherwise non-invasive cancer cells when co-injected into mice (64). Altogether, the aforementioned information do not recapitulate the complex interactions between the tumor epithelium and stromal cells as the intricated pathways leading to cancer progression still remain to be fully dissected. Interestingly, the present study demonstrates that GPER mediates the up-regulation of FASN by E2 and G-1 also in CAFs. In addition, using cerulenin, we demonstrated that the estrogen-induced proliferation and migration of CAFs involves FASN activity. These findings, together with our previous data showing that GPER is required for the migration of CAFs induced by E2 (23), further highlight the potential of estrogens to stimulate tumor progression through the GPER-mediated FASN expression and activity.
The present investigation provides novel insights into the molecular mechanisms by which the endogenous lipogenesis may exert an oncogenic role in the development of estrogensensitive tumors. In this regard, the lipogenic features of cancer cells through GPER may offer new avenues to identify and develop innovative therapeutic agents capable of successfully interfering with the initiation and progression of both primary and metastatic hormone-responsive tumors.