Synergistic Activation of the Mc2r Promoter by FOXL2 and NR5A1 in Mice

Abstract

Forkhead box protein L2 (FOXL2) is the earliest ovarian marker and plays an important role in the regulation of cholesterol and steroid metabolism, inflammation, apoptosis, and ovarian development and function. Mutations and deficiencies of the human FOXL2 gene have been shown to cause blepharophimosis-ptosis-epicanthus inversus syndrome as well as premature ovarian failure. Although Foxl2 interacts with steroidogenic factor 1 (Nr5a1) and up-regulates cyp19a1a gene transcription in fish, FOXL2 represses the transcriptional activity of the gene that codes for steroidogenic acute regulatory protein (Star) in mice. Most of the recent studies have heavily focused on the FOXL2 target genes (Star and Cyp19a1) in the ovaries. Hence, it is of importance to search for other downstream targets of FOXL2 and for the possibility of FOXL2 expression in nonovarian tissues. Herein, we demonstrate that the interplay between FOXL2 and NR5A1 regulates Star and melanocortin 2 receptor (Mc2r) gene expression in mammalian systems. Both FOXL2 and NR5A1 are expressed in ovarian and adrenal gland tissues. As expected, FOXL2 represses and NR5A1 enhances the promoter activity of Star. Notably, the promoter activity of Mc2r is activated by FOXL2 in a dose-dependent manner. Surprisingly, we found that FOXL2 and NR5A1 synergistically up-regulate the transcriptional activity of Mc2r. By mapping the Mc2r promoter, we provide evidence that distal NR5A1 response elements (−1410 and −975) are required for synergistic activation by FOXL2 and NR5A1. These results suggest that the interplay between FOXL2 and NR5A1 on the Mc2r promoter functions as a novel mechanism for regulating MC2R-mediated cell signaling as well as steroidogenesis in adrenal glands.

Introduction

Steroidogenic factor 1 (NR5A1, also known as SF1) plays a crucial role in the regulation of steroid hormone biosynthesis as well as in the endocrine development of both the gonads and adrenal glands [1–3]. Several genes, including Cyp19a1, Star, and melanocortin 2 receptor (Mc2r), have been identified as targets of NR5A1 [4–6]. Regulation of these genes involves the concerted action of NR5A1 with multiple transcription factors with which it can synergize, such as EGR1 [7], GATA4 [8], SOX9 [9], and WT1 [10]. Several transcriptional coactivators, such as nuclear receptor coactivator 1 [11], CREB-binding protein/p300 [12], transcriptional intermediary factor 2 [13, 14], and β-catenin [15], have been reported to interact with NR5A1 and very likely participate in NR5A1-mediated gene activation. On the other hand, factors such as nuclear receptor corepressor 1 [16], NR0B1 [17], and DDX20 [18] appear to play an inhibitory role by limiting NR5A1 function. Mounting evidence indicates that SUMOylation inhibits and phosphorylation activates NR5A1 [19, 20], while recent structural analyses have revealed that phospholipids can activate NR5A1 ligands [21, 22]. Recent extensive studies have found that NR5A1 is associated with developmental disorders and birth defects, such as adrenal agenesis and aplasia [23], androgen insensitivity syndrome [24], and ovarian insufficiency [25]. Therefore, NR5A1 is not only important for steroidogenesis but is also essential for organ development in the adrenal gland and gonads.

Forkhead box protein L2 (FOXL2), a single-exon gene encoding a forkhead transcription factor, is the earliest ovarian marker and plays an important role in the regulation of cholesterol and steroid metabolism, inflammation, apoptosis, and ovarian development and function [26]. FOXL2 is a nuclear protein that is highly conserved in vertebrates and is mainly present in fetal and adult extraocular muscles (i.e., superior rectus) and ovarian follicular cells [27, 28]. The GEO database also shows that the mRNA expression of human FOXL2 is detected in many normal tissues (from the most abundant to the least: ovary, pituitary gland, uterus, thymus, adrenal gland, and trachea), cell lines (such as hematopoietic stem cells, activated macrophages, colon cancers, splenocytes, and leukemia cells), and soft tissue sarcomas (such as fibrosarcoma and liposarcoma). Foxl2 mRNA is also expressed in the pituitary, mainly in gonadotroph and thyrotroph cells, suggesting a role in organogenesis [29]. Mutations and deficiencies of the human FOXL2 gene have been shown to cause blepharophimosis-ptosis-epicanthus inversus syndrome as well as premature ovarian failure. Foxl2 knockout mouse models suggest that FOXL2 plays an essential role in normal morphological change and differentiation in granulosa cells [30, 31]. This further supports the idea that FOXL2 is one of the earliest known markers of ovarian differentiation. A recent report demonstrates that deletion of Foxl2 in adult ovarian follicles up-regulates the transcriptional activity of testis-specific genes including Sox9, suggesting that testicular development is actively repressed by FOXL2 throughout the life of females [32]. This information further explains why FOXL2 is highly expressed in ovarian tissues but not in testes.

Several genes, including Gnrhr, Atf3, Nr5a2, Cyp19a1, and Star, have been identified as FOXL2 target genes [33–37]. In the sheep ovary, FOXL2 plays an important role in cholesterol metabolism and steroidogenesis by up-regulating the expression of CYP19A1 [35]. Foxl2 interacts with Nr5a1 in fish and up-regulates cyp19a1 gene transcription [36]. However, FOXL2 represses the promoter activity of the Star gene by binding to the Star promoter in mice [37]. Further future studies will be required to explain this contradiction in the repression of cholesterol synthesis and the up-regulation of estrogen synthesis. Moreover, a recent report has shown that human FOXL2 represses the CYP19A1 gene in granulosa cell differentiation [38]. More evidence is required to explain this contradiction of FOXL2 function on the same gene between species. Recently, FOXL2 expression has been detected in testicular tumor cells that resemble normal ovarian granulosa cells, suggesting that FOXL2 plays an important role in tumorigenesis [39]. However, recent studies of FOXL2 in ovarian granulosa cell tumors suggest that FOXL2 may act as a tumor suppressor [40, 41]. Thus, targeting FOXL2 proteins may lead to prevention and treatment of steroidogenesis- and apoptosis-mediated defects and disorders.

There have been many studies since FOXL2 was discovered as a master regulator in female sex differentiation and development. Most of them have focused on the FOXL2 target genes such as Star and Cyp19a1 in the ovaries. Hence, it was quite tempting to search for other downstream targets of FOXL2 and for the possibility of FOXL2 expression in nonovarian tissues. Herein, we investigate the function of FOXL2 in adrenal cortex, which shares a common embryological heritage with the gonad and kidney: all are derived from components of the urogenital ridge. MC2R is the key receptor involved in the adrenocorticotropic hormone (ACTH)-mediated signaling, which has been shown to be the major endocrine signaling system in adrenal cortex. The binding of ACTH to MC2R promotes the activation of PKA and MAPK-dependent signaling cascades that collectively initiate the adrenal-specific, steroidogenic transcriptional network. The ACTH-MAPK-ERK pathway eventually phosphorylates NR5A1, which is required for the promoter binding activity of NR5A1 [6]. We and others have previously reported that CDK7, in addition to ERK, can phosphorylate NR5A1 in adrenocortical cancer cells [20, 42]. In this study, we identified a novel functional link between FOXL2 and NR5A1 in the transcriptional regulation of the mouse Mc2r gene. Both Nr5a1 and Foxl2 mRNAs are expressed in ovary and adrenal gland with Foxl2 being expressed at higher levels in the ovary as compared to the adrenal glands. FOXL2 dramatically enhances the transcriptional activity on the Mc2r promoter. The synergy between these two proteins is mediated by distal NR5A1 response elements on the Mc2r promoter. These findings validate a novel role of FOXL2 in the regulation of Mc2r gene expression in the adrenal gland.

Materials and Methods

Reagents

All the cell culture reagents and protein-A agarose were purchased from Invitrogen (Carlsbad, CA). Antibodies used include: NR5A1 (Western blot; Upstate Biochemistry Inc., Charlottesville, VA), NR5A1 (immunohistochemistry; a generous gift from Dr. K. Morohashi, National Institute for Basic Biology, Myodaiji-cho, Japan), FOXL2 (generous gift from Dr. Reiner A. Veitia, Reproduction et Physiopathologie Obstétricale, Hôpital Cochin, Paris, France). Luciferase (LUC) activity was measured using the Dual Luciferase Assay System (Promega, Madison, WI), and Ni²⁺-NTA (nitrilotriacetic) agarose was purchased from QIAGEN (Valencia, CA).

DNA Constructs

Mouse Nr5a1 cDNA was PCR amplified by using forward primer 5′-TCGTAAGCTTATGGACTACTCGTACGACGAG-3′ and reverse primer 5′-ACGAGGATCCTCAAGTCTGCTTGGCCTGCAG-3′, digested with HindIII/BamHI, and then ligated into the same sites of pcDNA3.1(+) to create pcDNA3-Nr5a1 expression plasmid. The murine Mc2r promoter (−2000-bp) was PCR amplified by using forward primer 5′-ACGACCCGGGACATTAACTTTGAATTAGGGTGG-3′ and reverse primer 5′-TCGTCCCGGGCTGAAGTAGGATCTTTCTCGGC-3′, digested with XmaI (SmaI), and then ligated into the same sites of pGL3 to create Mc2r promoter LUC plasmid. The mutated NR5A1 binding site at −1410 with the core sequence CAAGGTGA mutated to CATTTTTA and the mutated NR5A1 binding site at −975 with the core sequence GAAGGTCA mutated to GATTTTTA were generated by PCR-based mutagenesis (QuikChange Lightning site-directed mutagenesis kit; Stratagene, La Jolla, CA). His-Foxl2-pcDNA3 and CMV-SID-Foxl2-pcDNA3 plasmids were generously provided by Dr. Colin Clay (Colorado State University, Fort Collins, CO) [33]. Mouse Star LUC reporter plasmid was kindly provided by Dr. Kenneth Escudero (Texas A&M University, Kingsville, TX). Gal4RE-TATA LUC plasmid was kindly provided by Dr. Ron Koenig (University of Michigan, Ann Arbor, MI). pM (Gal4 DBD) and pVP16 (VP16 AD) plasmids were from Clontech (Mountain View, CA). The mammalian two-hybrid kit containing pFR-Luc, pcmv-BD, and pcmv-AD plasmids were purchased from Stratagene. Gal4RE-SV40 LUC plasmid was from Promega. All the constructs were verified by nucleotide sequencing.

Cell Culture and Transfection

COS7 and HepG2 cells were maintained in Dulbecco modified Eagle's medium (DMEM) in the presence of 10% fetal bovine serum and antibiotics (GIBCO/Invitrogen, Carlsbad, CA) in humidified air containing 5% CO₂ at 37°C. Y1 cells were maintained in DMEM supplemented with 7.5% horse serum and 2.5% fetal bovine serum and antibiotics in humidified air containing 5% CO₂ at 37°C. After incubation, the cells were transfected using Fugene HD Transfection Reagent (Roche, Indianapolis, IN). Approximately 45–48 h after transfection, the cells were harvested. Luciferase activity was measured and normalized with respect to Renilla activity. All the experiments were performed three times in triplicate.

Immunoprecipitation Assay

Y1 cells (2 × 10⁶) were seeded onto 10-cm plates. After 24 h, cells were harvested and lysed in lysis buffer (40 mM HEPES, 120 mM sodium chloride, 10 mM sodium pyrophosphate, 10 mM sodium glycerophosphate, 1 mM ethylenediaminetetraacetic acid (EDTA), 50 mM sodium fluoride, 0.5 mM sodium orthovanadate, 1% Triton X-100) containing protease inhibitor cocktail (Sigma), followed by rotation for 1 h at 4°C to solubilize the proteins. Soluble protein was collected and immunoprecipitated with the indicated antibody overnight. Protein-A agarose beads were added to the protein lysates for 2 h in the cold room. The beads were centrifuged and washed at least three times with the lysis buffer. For Ni²⁺-bead pull-down assays, Ni²⁺-NTA agarose was used to precipitate His-tagged FOXL2 from cell lysates. Proteins were eluted by boiling in 50 μl of 2× Laemmli sample buffer, resolved by 8% SDS-PAGE, and processed for immunoblotting as described below.

Immunoblotting

Protein lysates were allowed to rotate at 4°C for 30 min, and the protein contents of the high-speed supernatant were determined using the Bradford protein assay (Bio-Rad Laboratories, Inc., Hercules, CA). Equivalent quantities of protein (20–45 μg) were resolved on polyacrylamide-SDS gels, transferred to nitrocellulose membrane (Bio-Rad), and immunoblotted with specific antibodies. The results were visualized using the Supersignal West Dura Extended Duration Substrate kit (Pierce Chemical Co., Rockford, IL).

Tissues from Experimental Animals

Ovaries and adrenal glands were obtained from normal adult mice. All the animal studies have been approved by the University of Michigan Animal User and Care Committee. Mouse adult testis cDNA was purchased from Zyagen (San Diego, CA).

RT-PCR and Real-Time PCR

Ovaries and adrenal glands were frozen in a dry-ice ethanol bath and stored at −80°C. Total RNA from ovaries and adrenal glands were extracted using TRIzol reagent and treated with DNase (Ambion, Austin, TX) to remove any residual genomic DNA; the RNA content was quantified by ultraviolet spectrometry. One microgram of total RNA was used to synthesize cDNA using the iScript kit (Bio-Rad) according to the manufacturer's recommended protocol. The final cDNA product was purified and eluted in 50 μl of Tris-EDTA buffer using PCR purification columns (QIAGEN). Two primers (5′-AAGCCCCCGTACTCGTACGTGGCGCTCATC-3′ and 5′-GTAGTTGCCCTTCTCGAACATGTC-3′) were used to amplify 241-bp Foxl2 fragments. Two primers (5′-CCTGAACAACCACAGCCTCGTAAAGGAC-3′ and 5′-CTGCATGCTCAGGGCCCGCACCTCCACC-3′) were used to amply 147-bp mouse Nr5a1 fragment. Two primers (5′-CATCACCATCTTCCAGGAGCGA-3′ and 5′-GGTGAAGACACCACTAG-3′) were used to amplify 88-bp mouse glyceraldehydes-3-phosphate dehydrogenase (Gapdh) fragments. For quantitative real-time PCR analysis of mRNA transcript abundance, cDNA was combined with 2× SYBR green PCR master mix (Applied Biosystems, Foster City, CA), and gene-specific primers in the ABI 7500 thermocycler system (Applied Biosystems). All the data were normalized to Gapdh as an internal housekeeping standard.

Mammalian Two-Hybrid Assay

The mammalian vector pM encoding the GAL4-DBD is designed to generate a fusion protein with a bait polypeptide. The mammalian vector pVP16 encoding the activation domain (AD) derived from the VP16 protein of herpes simplex virus is designed to express a fusion protein linked to a target polypeptide. The Nr5a1 and Foxl2 cDNAs were subcloned from the pcDNA3.1(+) expression constructs into mammalian expression vectors pM and pVP16 to generate pM-Nr5a1 (bait) and pVP16-Foxl2 (target), respectively. Both plasmids Gal4RE-TATA-Luc and Gal4RE-SV40-Luc contain sequence encoding LUC and five copies of the GAL4 DNA-binding site. The plasmids pM-Nr5a1 and pVP16-Foxl2 were cotransfected with Gal4RE-TATA-Luc or Gal4RE-SV40-Luc into HepG2 cells using Fugene HD (Roche). Approximately 45–48 h after transfection, the cells were harvested. Luciferase activity was measured and normalized with Renilla activity. All the experiments were performed three times in triplicate.

Statistical Analysis

Statistical analyses were performed using the Student t-test or a one-way ANOVA when more than two groups were compared. After the ANOVA analysis, the post hoc multiple comparisons were performed by using the Tukey honestly significant difference test to determine the statistical difference among the subgroups. For each test, P < 0.05 and P < 0.001 were considered significant (*) and very significant (**), respectively.

Results

Foxl2 mRNA and FOXL2 Protein Expression in Ovaries and Adrenal Glands

Foxl2 mRNA expression in the ovaries and adrenal glands of the mouse was assessed using RT-PCR and real-time RT-PCR. Total RNA was extracted from the ovaries and adrenal glands of adult mice. As shown in Figure 1A, RT-PCR revealed Foxl2 transcripts in both the ovaries and adrenal glands of adult mice with Foxl2 being expressed at higher levels in the ovary as compared to the adrenal glands. Both the ovaries and adrenal glands expressed transcripts of Nr5a1 and Gapdh (housekeeping gene). Unlike the ovary and adrenal gland, the PCR product of Foxl2 was absent in testes. Real-time RT-PCR analysis was performed to measure Nr5a1 and Foxl2 expression in ovaries and adrenal glands (Fig. 1B). Nr5a1 mRNA levels were similar in both ovaries and adrenal glands. Foxl2 mRNA levels in the ovaries were about 40–50 times those found in adrenal glands. As was found in the adrenal glands of adult mice, Y1 cells (mouse adrenocortical cancer cells) also expressed low level of Foxl2 mRNA and abundant Nr5a1 mRNA (Fig. 1C). To determine whether FOXL2 proteins are endogenously expressed in Y1 cells, expression vectors encoding wild-type Foxl2 or empty vectors were transfected into Y1 cells. As shown in Figure 1D, when wild-type Foxl2 was transfected, a 45-kDa protein was detected, corresponding to the predicted size. When empty vector was transfected, less FOXL2 protein was detected. We next investigated whether FOXL2 proteins was endogenously expressed in adrenal glands of adult mice. As shown in Figure 1E, FOXL2 protein was detected in both the adrenal glands and ovaries with much more abundant FOXL2 expression in the ovaries than in the adrenal glands. As expected, both tissues expressed NR5A1 proteins. These results indicate that adrenal glands and Y1 cells endogenously express FOXL2 protein.

Fig. 1.

Foxl2 expression in adrenal gland and Y1 cells. A) RT-PCR analysis of Foxl2 and Nr5a1 expression from mouse ovaries, adrenal glands, and testes. Ovaries, adrenal glands, and testes were obtained from adult animals. Total RNA was extracted and reverse transcribed before PCR analysis and fractionation using electrophoresis. The 241-bp PCR product of Foxl2 and the 147-bp PCR product of Nr5a1 are present in the ovaries and adrenal glands, with Foxl2 predominantly expressed in the ovaries. Unlike the ovary and adrenal gland, the product of Foxl2 is absent in testes. Glyceraldehyde-3-phosphate dehydrogenase (Gapdh) serves as a control to indicate the presence of cDNA in each sample. B) Real-time RT-PCR analysis was performed to measure Nr5a1 and Foxl2 mRNA expression with Gapdh as an internal control in the ovary and adrenal. Each point represents the average of three experiments, each with triplicate samples. Error bars indicate the standard errors. **P < 0.001. C) RT-PCR analysis of Foxl2 and Nr5a1 expression from Y1 cells. D) Y1 cells were transfected without or with His-Foxl2 expression plasmid and cell lysates were immunoblotted with anti-FOXL2 antibody. Endogenous 45-KDa FOXL2 protein is detected in Y1 cells (left lane). Cell lysates were probed with a β-actin antibody to control for equal loading. E) Total lysates from the adrenal glands and ovaries of adult mice were immunoblotted with anti-FOXL2 antibody. Cell lysates were probed with a NR5A1 antibody for positive control. Cell lysates were probed with a β-actin antibody to control for equal loading.

FOXL2 Interacts with NR5A1

Previously, FOXL2 was found to interact through the forkhead domain with the ligand-binding domain of NR5A1 to form a heterodimer in fish [36] and in human granulosa cells [43]. In order to confirm the interaction of mouse FOXL2 and NR5A1, we first performed mammalian two-hybrid assays. Nr5a1 and Foxl2 were subcloned into the pM vector and pVP16 vector to produce the pM-Nr5a1 and pVP16-Foxl2, respectively. The pM-Nr5a1 and pVP16-Foxl2 were transfected along with Gal4RE-TATA-Luc or Gal4RE-SV40-Luc into HepG2 cells. As shown in Figure 2A, the interaction of NR5A1 and FOXL2 promotes higher LUC activity as compared with the controls. Similar results are observed by using pFR-Luc in COS7 cells (Fig. 2B).

Fig. 2.

Physical interaction between FOXL2 and NR5A1. A) FOXL2 and NR5A1 interact as revealed by mammalian two-hybrid assays. HepG2 cells were cotransfected with Gal4RE-TATA-Luc (left) or Gal4RE-SV40-Luc (right), pM, pVP16, pM-Nr5a1, or pVP16-Foxl2 as indicated. Cells were assayed for reporter activity 48 h after transfection. Relative LUC activity (fold activation) was calculated and plotted. B) FOXL2 and NR5A1 interaction as revealed by the mammalian two-hybrid assay. COS7 cells were cotransfected with pFR-Luc, pCMV-BD (BD), pCMV-AD (AD), BD-Nr5a1, or AD-Foxl2 as indicated. Cells were assayed for reporter activity 48 h after transfection. Relative LUC activity (fold activation) was calculated and plotted. C) Y1 cells were transfected with pcDNA3 vector or His-Foxl2 as described in Materials and Methods. Forty-eight hours after transfection, cell lysates were subjected to Ni²⁺-bead pull down, followed by anti-NR5A1 or anti-FOXL2 immunoblotting. Expression of NR5A1 and FOXL2 in the total lysates (WCL) was confirmed by Western blotting. Note that NR5A1 was specifically coimmunoprecipitated by FOXL2 (lower right panel). *P < 0.05; **P < 0.001.

The interaction between FOXL2 and NR5A1 was further investigated by a coimmunoprecipitation assay. A His-tagged Foxl2 expression plasmid was transfected in Y1 cells for 48 h, and the His-tagged FOXL2 was precipitated by Ni²⁺-bead pull-down. As shown in Figure 2C, His-tagged FOXL2 (His-FOXL2) was expressed in Y1 cells and was immunoprecipitated from the whole-cell lysate and Ni²⁺ eluate. Endogenous NR5A1 was present in the WCL and was coimmunoprecipitated with His-tagged FOXL2 (Ni²⁺ eluate). From the results of the mammalian two-hybrid assay and coimmunoprecipitation in Y1 cells, we can conclude that FOXL2 and NR5A1 interact with each other in vivo.

FOXL2 Is a Repressor of the Star Promoter but an Activator of the Mc2r Promoter

A previous report has shown that wild-type FOXL2 is a repressor of the Star promoter [37]. Because NR5A1 is a strong activator of Star promoter and can bind to FOXL2, we asked whether FOXL2 influences NR5A1-mediated transcriptional activity on the Star promoter. We chose liver carcinoma HepG2 cells for this study of NR5A1- and FOXL2-mediated transcriptional activity because they did not express endogenous NR5A1 and FOXL2. As shown in Figure 3A, NR5A1 alone activated Star gene transcription in a LUC assay using HepG2 cells. Interestingly, FOXL2 decreased the NR5A1-activated Star expression in a dose-dependent manner when it was cotransfected with Nr5a1. In Figure 3B, FOXL2 alone reduced Star promoter activity. However, NR5A1 partially blocked the repressive activity of FOXL2 even at high doses of NR5A1. This indicates that FOXL2 repressor activity was sufficiently robust. Therefore, FOXL2 and NR5A1 had opposite effects on Star promoter activity as seen in Figure 3C.

Fig. 3.

FOXL2 antagonizes NR5A1-mediated Star promoter activity. A) Dose-dependent suppression of NR5A1-mediated enhancement of Star promoter activity by FOXL2. HepG2 cells were cotransfected with Star-Luc, 100ng of Nr5a1 plasmid, and different amount of Foxl2 plasmid. The total amount of transfected plasmids was adjusted to 0.5 μg with empty vectors. B) Dose-dependent rescue of FOXL2-mediated suppression of Star promoter activity by NR5A1. HepG2 cells were cotransfected with Star-Luc, 10 ng of Foxl2 plasmid, and different amount of Nr5a1 plasmid. The total amount of transfected plasmids was adjusted to 0.5 μg with empty vectors. C) FOXL2 represses NR5A1-mediated Star promoter activity. HepG2 cells were cotransfected with Star-Luc and 100 ng Nr5a1 plasmid, 10 ng Foxl2 plasmid, or a combination of the two. The total amount of transfected plasmids was adjusted to 0.5 μg with empty vectors. Relative LUC activity (fold activation) was calculated and plotted. All the experiments were performed three times in triplicate. Error bars indicate the standard errors. *P < 0.05; **P < 0.001.

To determine whether the Mc2r promoter is regulated by FOXL2, wild-type Foxl2 was cotransfected with the −2000-bp Mc2r promoter-LUC reporter plasmid into HepG2 (MC2R-negative) and Y1 (MC2R-positive) cells. Forty-eight hours after transfection, Mc2r promoter activity was determined by measuring LUC activity in cell lysates. To our surprise, wild-type FOXL2 enhanced basal Mc2r promoter activity in a dose-dependent manner (Fig. 4, A and B) in both cell types.

Fig. 4.

Synergistic up-regulation of Mc2r gene expression by FOXL2 and NR5A1. A) HepG2 cells were cotransfected with Mc2r-Luc and different amount of Foxl2 plasmid. B) Y1 cells were cotransfected with Mc2r-Luc and different amount of Foxl2 plasmid. C) HepG2 cells were cotransfected with Mc2r-Luc and 100 ng Nr5a1 plasmid, 20 ng Foxl2 plasmid, or a combination of the two. D) HepG2 cells were cotransfected with Mc2r-Luc, 100 ng of Nr5a1 plasmid and different amount of Foxl2 plasmid. E) Nr5a1 alone (100 ng) or together with WT (wild-type) Foxl2 (10 ng) or SID (dominant negative) Foxl2 (10 ng) expression plasmids were cotransfected into HepG2 cells with Mc2r-Luc (300 ng/well). Relative LUC activity (fold activation) was calculated and plotted. The experiments were performed three times in triplicate. Error bars indicate the standard errors. *P < 0.05; **P < 0.001.

FOXL2 and NR5A1 Synergistically Activate the Mc2r Promoter

Prior studies have indicated that NR5A1 is a transcriptional activator of the Mc2r promoter [6, 20]. Thus, we examined whether FOXL2 can enhance NR5A1-activated Mc2r gene expression. To test this hypothesis, we cotransfected Nr5a1 and Foxl2 with Mc2r-LUC reporter gene and determined LUC activity 48 h posttransfection. Although NR5A1 alone induced a 20-fold increase and FOXL2 alone induced a 3-fold increase in LUC activity, coexpression of NR5A1 and FOXL2 resulted in a 60-fold synergistic activation (Fig. 4, C and D). Next, we examined whether this synergy between NR5A1 and FOXL2 in the Mc2r promoter activation was the result of FOXL2-mediated action. We cotransfected Nr5a1 and wild-type Foxl2 (WT Foxl2) or dominant negative Foxl2 (SID-Foxl2) with Mc2r promoter plasmid in HepG2 cells. As shown in Figure 4E, while WT FOXL2 and NR5A1 promoted synergy effect, SID-FOXL2 significantly reduced this synergistic effect.

Minimal Mc2r Promoter Region Responsive to FOXL2-NR5A1 Activation

Because the −2000-bp mouse Mc2r promoter contains three NR5A1 response elements (Fig. 5A), the Mc2r promoter was truncated to determine the minimal region that is important for transcriptional activation by NR5A1 and FOXL2. In these studies, we chose liver carcinoma HepG2 cells because they do not express endogenous NR5A1 and FOXL2. The −2000-bp and −1680-bp promoters contain three NR5A1 response elements (−1410 bp, −975 bp, and −95 bp). These two promoters of different length showed similar transcriptional activity and synergy effect by NR5A1 and FOXL2. As shown in Figure 5B, deletion of the distal NR5A1 response element (−1410 bp), as shown in the −1320 promoter, resulted in a significant loss of NR5A1-mediated transcriptional activity in HepG2 cells. This indicates that the distal NR5A1 response element is important for NR5A1 action on Mc2r promoter. Interestingly, the synergy effect by NR5A1 and FOXL2 on the −1320-bp promoter is still maintained. Truncation of the two distal NR5A1 response elements (−1410 bp and −975 bp) resulted in almost complete loss of transcriptional activity by NR5A1 and synergy effect by NR5A1 and FOXL2. We obtained a similar result by using human choriocarcinoma JEG3 cells that contain most of the enzymes for steroidogenesis but do not express NR5A1 and FOXL2 (Fig. 5C). These results suggest that the distal NR5A1 response elements (−1410 bp and −975 bp) are required for both NR5A1-mediated transcriptional activation and NR5A1-FOXL2-mediated synergistic activation.

Fig. 5.

Regions of Mc2r promoter important for synergistic transcriptional up-regulation by FOXL2 and NR5A1. A) Putative NR5A1 binding sites in different Mc2r promoter constructs. B) HepG2 cells were cotransfected with mouse Mc2r promoter deletion constructs and either 100 ng of Nr5a1 alone, 20 ng Foxl2 alone, or 100 ng Nr5a1 and 20 ng Foxl2 expression plasmids. The total amount of transfected plasmids was adjusted to 0.5 μg with empty vectors. Luciferase activities were measured 48 h after transfection. The relative LUC activity (fold activation) was calculated and plotted. The experiments were performed three times in triplicate. The error bars indicate the standard errors. C) JEG3 cells were cotransfected with mouse Mc2r promoter deletion constructs and either 100 ng Nr5a1 alone, 20 ng Foxl2 alone, or 100 ng Nr5a1 and 20 ng Foxl2 expression plasmids. The total amount of transfected plasmids was adjusted to 0.5 μg with empty vectors. Luciferase activities were measured 48 h after transfection. Relative LUC activity (fold activation) was calculated and plotted. All the experiments were performed three times in triplicate. Error bars indicate the standard errors. *P < 0.05; **P < 0.001.

To further determine whether the distal NR5A1 response elements (−1410 bp and −975 bp) are required for NR5A1-FOXL2-mediated synergic activation, we next generated two distal NR5A1 response element mutations (1410M and 975M) in −2000-bp Mc2r promoter-LUC plasmids. Consistent with the previous results, mutations in distal NR5A1 response element regions reduced NR5A1-mediated promoter activity and decreased NR5A1-FOXL2-mediated synergic activation on Mc2r promoter in both HepG2 and JEG3 cells (Fig. 6, B and C). Similar results were observed in COS7 cells (data not shown). Thus, distal NR5A1 response elements (−1410 bp and −975 bp) are indeed required for both NR5A1-mediated transcriptional activation and NR5A1-FOXL2-mediated synergistic activation.

Fig. 6.

Distal NR5A1 response elements on Mc2r promoter are required for synergistic transcriptional up-regulation by FOXL2 and NR5A1. A) Mutations of distal NR5A1 response elements on Mc2r promoter. The mutated NR5A1 binding site at −1410 (1410M) with the core sequence CAAGGTGA mutated to CATTTTTA and the mutated NR5A1 binding site at −975 (975M) with the core sequence GAAGGTCA mutated to GATTTTTA were generated. B) HepG2 cells were cotransfected with mouse Mc2r promoter constructs (WT, 1410M, or 1410M + 975M) and either 100 ng Nr5a1 alone, 20 ng Foxl2 alone, or 100 ng Nr5a1 and 20 ng Foxl2 expression plasmids. The total amount of transfected plasmids was adjusted to 0.5 μg with empty vectors. Luciferase activities were measured 48 h after transfection. Relative LUC activity (fold activation) was calculated and plotted. Experiments were performed three times in triplicate. Error bars indicate standard errors. C) JEG3 cells were cotransfected with mouse Mc2r promoter constructs (WT, 1410M, or 1410M + 975M) and either 100 ng Nr5a1 alone, 20 ng Foxl2 alone, or 100 ng Nr5a1 and 20 ng Foxl2 expression plasmids. The total amount of transfected plasmids was adjusted to 0.5 μg with empty vectors. Luciferase activities were measured 48 h after transfection. Relative LUC activity (fold activation) was calculated and plotted. Experiments were performed three times in triplicate. Error bars indicate the standard errors. *P < 0.05; **P < 0.001.

Discussion

Nuclear hormone receptors mediate the transcriptional responses to a wide variety of physiological stimuli and thus function as important regulators of development, metabolism, and reproduction. For years, FOXL2 was studied mainly in ovarian tissues due to its important functions on ovarian development and maintenance. However, FOXL2 is expressed in nonovarian endocrine tissues, such as the adrenal gland. Herein, we show for the first time that in adrenal tissue FOXL2 acts as a transcriptional repressor of the Star gene but a transcriptional activator of the Mc2r gene. Although FOXL2 itself is a weak transcriptional activator compared to NR5A1 on Mc2r promoter, the FOXL2-NR5A1 complex dramatically enhances Mc2r promoter activity. In sum, our studies suggest that FOXL2 plays an important role in regulation of Mc2r transcriptional activity.

Importantly, we have demonstrated that Foxl2 mRNA is expressed not only, as expected, in mouse ovaries, but also in mouse adrenal glands and Y1 adrenocortical cancer cells. Our data further supports a previous report in the GEO database that FOXL2 mRNA expression in human adrenal gland is about 15-fold lower than that in human ovary. Although the level of Foxl2 mRNA expression in the adrenal glands is lower than in the ovaries, the expression of Foxl2 mRNA in adrenal glands and Y1 cells indicates a possible role of FOXL2 in adrenal gland development and steroid biosynthesis. Indeed, we observed that mouse FOXL2 interacts with mouse NR5A1, a master regulator in steroidogenesis and development of adrenal glands and gonads, and consequently activates NR5A1-mediated Mc2r promoter activity. Our finding that FOXL2 interacts with NR5A1 further validates the previous reports that Foxl2 up-regulates the cyp19a1 gene transcription by interacting with Nr5a1 in tilapia [36] and human FOXL2 interacts with NR5A1 and represses NR5A1-induced CYP17 transcription in granulosa cells [43]. Taken together, these data suggest that the interaction between FOXL2 and NR5A1 plays a decisive role in transcriptional activity of steroidogenic genes.

Among FOXL2 target genes, the Star gene is the most studied so far. STAR is important in steroidogenesis and cholesterol transport and is an important marker of granulosa cell differentiation. Many reports have clearly demonstrated that the Star gene is activated by NR5A1 [1] and is suppressed by FOXL2 [37]. A recent study has shown that FOXL2 shares a binding region in the Star promoter with NR5A1, suggesting that competition for binding sites may occur between NR5A1 (transcriptional activator) and FOXL2 (transcriptional repressor), further modulating Star gene activity [37]. Our findings (Fig. 3) further explain and support the previous studies. Our results indicate that FOXL2 repressor activity is sufficiently robust on Star promoter as NR5A1 could only partially block the repressive activity of FOXL2 even at high doses of NR5A1. Therefore, the competition between FOXL2 and NR5A1 for the Star promoter is crucial for steroidogenesis and granulosa cell differentiation.

In the current study, we demonstrate that FOXL2 and NR5A1 synergistically enhance Mc2r promoter activity. Two lines of evidence from the current study suggest that FOXL2 enhancement is likely mediated by physical interactions with NR5A1 rather than direct binding to FOXL2 response element(s), if present, on Mc2r promoter. First, NR5A1 coimmunoprecipitates with FOXL2. This further supports two previous reports that Foxl2 interacts through the forkhead domain with the ligand-binding domain of Nr5a1 in fish [36] and FOXL2 interacts with NR5A1 in human granulosa cells [43]. Second, the distal NR5A1 response elements on the Mc2r promoter are required for this enhancement. We were surprised to find that the distal (−1410 bp and −975 bp), not proximal, response elements for NR5A1 are indeed required for NR5A1-mediated transcriptional activation of the Mc2r promoter since most prior NR5A1 studies on Mc2r promoter were performed by using the proximal NR5A1 response element (−95 bp). Our findings are in agreement with a previous report that there is another positive NR5A1 regulatory element in the distal upstream region (between −1808 bp and −106 bp) of the Mc2r promoter [44].

In this study, we found that FOXL2 represses Star promoter activity, which is consistent with other reports [37, 42], but enhances Mc2r gene expression by synergistic interaction with NR5A1. Moreover, while Foxl2 up-regulates cyp19a1 gene transcription in fish [36], FOXL2 represses CYP19A1, CYP11A1, and CCND2 genes in human granulosa cells [38]. This indicates that the transcriptional activities of steroidogenic enzymes regulated by FOXL2 are likely to be dependent on cell type, promoter context, and cell-signaling pathways. This phenomenon also indicates that FOXL2 has a wide diverse role in regulating transcriptional activity of its target genes. Further studies indeed are required to dissect what machinery or complex is recruited by FOXL2 on promoters of FOXL2 target genes.

Our LUC assays in the present study show that a low dose of FOXL2 (10–20 ng plasmid) is enough to fully enhance NR5A1-induced activation of the Mc2r promoter activity in a synergistic manner. This might support the fact that the low level of FOXL2 protein expressed in the adrenal gland (40-fold lower than the level of FOXL2 in the ovary and 35-fold lower than the level of NR5A1 in the adrenal gland) is sufficient to execute its function. Many groups have observed similar findings. For example, a low dose of Foxl2 (10 ng plasmid) is enough to enhance Nr5a1 (100 ng plasmid) activation of the tilapia cyp19a1 gene [36], and even a lower dose of FOXL2 (2 ng plasmid) is sufficient to fully activate Cga gene expression in LβT2 cells [29]. Moreover, our current study and previous reports [37, 45] have also shown that low dose of FOXL2 (20 ng plasmid) is sufficient to repress NR5A1 (100 ng plasmid) activation of the Star promoter activity. Therefore, based on these results, low levels of FOXL2 protein are sufficient to fully exert its transcriptional activity on different genes. This phenomenon supports our hypothesis that the relatively low levels of FOXL2 (a 40-fold difference in the Foxl2 and Nr5a1 mRNAs) in adrenal gland and Y1 cells are sufficient to exert its effect in vivo.

Although the molecular mechanisms underlying the formation of and regulation by the FOXL2-NR5A1 complex remains unknown in adrenal gland, our study demonstrates the importance of FOXL2 as a novel molecular step in the regulation of Mc2r transcriptional activity. A previous report has demonstrated that FOXL2 is SUMOylated at lysine-25 via UBE21-mediated SUMOylation based on coimmunoprecipitation and promoter-LUC assays [45]. However, a more recent report has shown that other SUMO sites (lysine-114 and lysine-150) are involved in FOXL2′s stability, localization, and activity [46]. It is noteworthy that NR5A1 can be SUMOylated, and SUMOylation of NR5A1 represses its transcriptional activity [19, 20]. Therefore, future studies are necessary to determine how SUMO modification influences the interplay between FOXL2 and NR5A1 to differentially regulate target genes.

In summary, we have identified a novel relationship between FOXL2 and NR5A1 that contributes to the transcriptional regulation of Mc2r gene. Our study adds a new layer to the previous understanding of how FOXL2 functions to regulate adrenal and gonadal tissue differentiation and development as well as steroid biosynthesis via the early determinants of steroidogenic potential including MC2R and STAR.