Skip to main content
NCBI home page
Molecular Endocrinology logoLink to Molecular Endocrinology
. 2009 Feb 5;23(5):649–661. doi: 10.1210/me.2008-0412

FSH and FOXO1 Regulate Genes in the Sterol/Steroid and Lipid Biosynthetic Pathways in Granulosa Cells

Zhilin Liu 1, Michael D Rudd 1, Inmaculata Hernandez-Gonzalez 1, Ignacio Gonzalez-Robayna 1, Heng-Yu Fan 1, Anthony J Zeleznik 1, JoAnne S Richards 1
PMCID: PMC2675958  PMID: 19196834

Abstract

The forkhead box transcription factor FOXO1 is highly expressed in granulosa cells of growing follicles but is down-regulated by FSH in culture or by LH-induced luteinization in vivo. To analyze the function of FOXO1, we infected rat and mouse granulosa cells with adenoviral vectors expressing two FOXO1 mutants: a gain-of-function mutant FOXOA3 that has two serine residues and one threonine residue mutated to alanines rendering this protein constitutively active and nuclear and FOXOA3-mutant DNA-binding domain (mDBD) in which the DBD is mutated. The infected cells were then treated with vehicle or FSH for specific time intervals. Infection of the granulosa cells was highly efficient, caused only minimal apoptosis, and maintained FOXO1 protein at levels of the endogenous protein observed in cells before exposure to FSH. RNA was prepared from control and adenoviral infected cells exposed to vehicle or FSH for 12 and 24 h. Affymetrix microarray and database analyses identified, and real time RT-PCR verified, that genes within the lipid, sterol, and steroidogenic biosynthetic pathways (Hmgcs1, Hmgcr, Mvk, Sqle, Lss, Cyp51, Tm7sf2, Dhcr24 and Star, Cyp11a1, and Cyp19), including two key transcriptional regulators Srebf1 and Srebf2 of cholesterol biosynthesis and steroidogenesis (Nr5a1, Nr5a2), were major targets induced by FSH and suppressed by FOXOA3 and FOXOA3-mDBD in the cultured granulosa cells. By contrast, FOXOA3 and FOXOA3-mDBD induced expression of Cyp27a1 mRNA that encodes an enzyme involved in cholesterol catabolism to oxysterols. The genes up-regulated by FSH in cultured granulosa cells were also induced in granulosa cells of preovulatory follicles and corpora lutea collected from immature mice primed with FSH (equine choriogonadotropin) and LH (human choriogonadotropin), respectively. Conversely, Foxo1 and Cyp27a1 mRNAs were reduced by these same treatments. Collectively, these data provide novel evidence that FOXO1 may play a key role in granulosa cells to modulate lipid and sterol biosynthesis, thereby preventing elevated steroidogenesis during early stages of follicle development.

FOXO1 may play a key role in granulosa cells to modulate lipid and sterol biosynthesis, thereby preventing elevated steroidogenesis during early stages of follicle development.

The FOXO family of transcription factors is evolutionarily conserved from Drosophila melanogaster and Caenorhabditis elegans to man (1,2). In mammals, FOXO1, FOXO3, and FOXO4 are differentially expressed in specific tissues but also exhibit overlapping expression patterns (3). Functionally, FOXO factors have been linked to cell survival, cell longevity, metabolic homeostasis, and apoptosis (4). These diverse effects of FOXO factors have been documented by targeted deletion of the genes in vivo, as well as by overexpression of mutant forms of FOXO1 and FOXO3 in different cell types in culture (5,6,7,8,9,10). Each approach has revealed the complexity of genes regulated by FOXO factors. This appears to be mediated, in part, by the ability of FOXO factors to act not only as transcriptional DNA binding factors but also as coregulatory molecules for a wide range of other transcription factors including nuclear hormone receptors, CCAAT enhancer binding protein-β, β-catenin, and SMAD proteins (11,12,13,14,15).

Targeted deletion of each gene in vivo has documented that each FOXO factor exhibits a distinct phenotype. Foxo4 null mice are normal (3). Foxo3 null mice exhibit premature ovarian failure due to the global exit of primordial follicles from the resting pool. This response is mediated primarily by the disruption of FOXO3 in oocytes and their premature maturation (5,8). That FOXO3 exerts its major effects in the ovary by regulating oocyte function and restricting exit of follicles from the resting pool is supported by the observations that the phenotype of mice in which Pten has been conditionally knocked out in oocytes mimics that of the Foxo3 null mice (16). Foxo1 null mice are embryonic lethal with marked evidence of vascular defects, precluding analyses of FOXO1 functions in tissues of adult mice (3). More recently, mice with floxed alleles of the Foxo1, Foxo3, and Foxo4 genes have been generated, allowing cell-specific disruption of each or all three genes in selective cell types (6,7). Results from these studies have indicated that the effects of Foxo1/3/4 factors are cell context and tissue specific. Remarkably, genes regulated by FOXO factors in endothelial cells present in lung are distinctly different for genes regulated in endothelial cells present in thymocytes (7).

Results from overexpression of constitutively active FOXO1 in which two critical serine residues and one threonine residue have been mutated to alanines (FOXOA3) indicate that, in the cell types examined, each cell exhibits a few common genes but also specific responses to FOXOA3 (9,17). For example, FOXO1 acts via FOXO1 (insulin) response elements (IREs) to induce transcription of Cdkn1b (p27KIP) and Igfbp1 (9,18). However, many other effects appear to occur independently of IREs (11,13). Moreover, microarray analyses of cells overexpressing FOXOA3 or FOXOA3 containing a mutant DNA-binding domain (mDBD) documented further that many effects of FOXO1 occurred independently of DNA binding to an IRE. Whereas cell apoptosis was linked to FOXOA3, the FOXOA3-mDBD mimicked other FOXO1 effects but not apoptosis (9). Moreover, because these results were obtained in a Pten null cell line, some, but not all, conclusions may be appropriate for other cell types. Specifically, the roles of FOXO factors appear to be cell context specific. For example, overexpression of FOXOA3 in pancreatic cells selectively blocks proliferation but also suppresses glucose metabolism, leading the authors to conclude that FOXO1 in these cells acts as a linchpin between nutrient sensing and β-cell turnover (19). Moreover, the metabolic diapause and genes affected in these cells were distinct from those associated with diapause in C. elegans. In muscle cells and tissue, the effects of FOXO are contextual and dependent on the stage of muscle cell differentiation (17). Overexpressing FOXO1 in myoblasts promotes fusion and does not stimulate apoptosis, in myotubes it promotes atrophy, and in muscle it switches metabolism from carbohydrate oxidation to fatty acids as the major energy source (17). In liver, overexpression of FOXOA3 promotes glucose production but suppresses lipogenesis (20,21). Conversely, disrupting Foxo1 in liver reduces glucose production (22,23).

In the mammalian ovary, FOXO1 is highly and selectively expressed in granulosa cells of growing follicles (3,24). In these cells in vivo, Foxo1 expression is hormonally induced by FSH and estradiol and rapidly down-regulated in response to LH/human (h) choriogonadotropin (CG) during the process of luteinization. In cultured granulosa cells, FSH and IGF-I rapidly phosphorylate FOXO1 via activation of the phosphatidylinositol 3-kinase/AKT signaling pathway, leading to its exclusion from the nucleus. In cultured granulosa cells FSH also rapidly down-regulates the expression of the Foxo1 gene. These results indicate that FSH and, more potently, LH in vivo act to suppress the functions and expression of FOXO1 (24). By contrast, recent studies indicate that FOXO1 is a negative regulator of FSH-mediated proliferation and differentiation (10). Thus, in granulosa cells there appears to be an FSH receptor FOXO1 regulatory loop.

Because the genes and cellular functions regulated by FOXO1 in granulosa cells during follicular development remain to be determined, we sought to identify specific FOXO1 target genes by using a gain-of-function approach in which a constitutively active form of FOXO1 would be predicted to enhance the expression of genes induced by endogenous FOXO1 and repress genes normally suppressed by FOXO1. To accomplish this goal, we used an adenoviral vector expressing a constitutively active, nuclear form of FOXO1 in which three serines are mutated to alanines (designated FOXOA3) in cultured rat and mouse granulosa cells (9,13). In addition, increasing evidence is documenting that FOXO1 acts not only as a transcription factor binding to consensus FOXO1 binding sites in the promoters of target genes but also as a coregulatory molecule that modifies the functional activity of other transcription factors including nuclear receptors, SMAD proteins, and β-catenin (11,12,13,15). Therefore, we also tested the functional activity of a DBD mutant of FOXOA3 (FOXOA3-DBD) in cultured rat and mouse granulosa cells. Using these two mutant forms of FOXO1 we sought to identify specific target genes that were either dependent or independent of the DBD and which either enhanced or suppressed specific effects of FSH on granulosa cell function. Based on microarray analyses, RT-PCR analyses, and in vivo studies, we identified one distinct class of genes that is regulated by FOXOA3 and FOXOA3-mDBD in granulosa cells. Specifically, genes controlling the biosynthesis of cholesterol, other sterols, and steroidogenic genes, were markedly reduced by the FOXO1 mutants, indicating that endogenous FOXO1 may impact cholesterol biosynthesis in vivo, a major function of granulosa cells and luteal cells.

Results

Adenoviral-mediated expression of FOXOA3 and FOXOA3-mDBD in rat granulosa cells is time and dose dependent

To determine potential functions of FOXO1 in cultured rat granulosa cells, the adenoviral vectors expressing FOXOA3 and its DBD mutant FOXOA3-mDBD were used (13). These vectors were chosen to distinguish genes regulated by FOXO1 as a transcription factor vs. effects of FOXO1 as a coregulatory molecule that are independent of DNA binding activity. To analyze the temporal and dose-dependent expression of endogenous FOXO1 and adenoviral-expressed FOXOA3 and FOXOA3-DBD, granulosa cells were harvested into defined medium, washed once, and plated on serum-coated dishes in serum-free defined medium at 0.5 × 106 cells per well in 12-well culture dishes. Cells were infected with the adenoviral vectors for 4 h in serum-free-defined medium as described in the figure legends. Cells were then washed and treated with hormones as described in the figure legends. Expression of FOXOA3 increased progressively from 3–12 h after infection and declined by 24 h (Fig. 1A). FOXOA3-mDBD was also elevated at 12 h and declined at 24 h. In contrast, the levels of endogenous FOXO1 protein were low in the green fluorescent protein (GFP)-infected control cells treated with FSH/testosterone (T).

Figure 1.

Figure 1

Open in a new tab

Time- and dose-dependent expression of FOXO1 and adenoviral expressed FOXOA3 and FOXOA3-mDBD. A, Granulosa cells from estradiol-treated immature rats were plated in defined serum-free DMEM-F12 (12-well plates) that were precoated with 1% FBS. The cells were infected 5–6 h after plating with 10 MOIs of each adenoviral vector: GFP (control), constitutively active FOXO1 (A3; FOXOA3; FKHR:AAA) or a DNA-binding mutant (mDBD, FOXOA3-mDBD, FKHR;HRAAA) that were obtained from Dr. William Sellers (Harvard University) and have been described previously (9,13). The cells were washed 4 h after infection and cultured in the presence or absence of hormones [FSH (100 ng/ml) and testosterone (10 ng/ml)] for 3, 6, 12, or 24 h as indicated. The Western blot is representative of three separate experiments. B, Granulosa cells were harvested and plated on serum-coated coverslips. The cells were then infected and treated with hormones, as above. Cell morphology, adenoviral infection efficiency, and apoptosis were analyzed by differential interference contrast imaging, immunofluorescent staining for GFP (green), and cleaved caspase 3 (red). C, Levels of endogenous FOXO1 and dose-dependent expression of FOXOA3 and FOXOA3-mDBD after infection with 2.5, 5, and 10 MOIs were analyzed by Western blotting in the absence or presence of FSH/T as indicated. DIC, Differential interference contrast; NT, not treated; WT, wild type.

Granulosa cells infected with the control vector and treated with FSH/T for 3, 12, and 24 h exhibited a cobblestone appearance that is typically observed in these cells exposed to either FSH or forskolin. The FOXOA3- and FOXOA3-mDBD-infected cells treated with FSH/T appeared similar to the control cells at 3 h but did not retain the cobblestone appearance at 12 h and 24 h. Rather, the FOXO mutant-expressing cells appeared more flattened. Immunofluorescent imaging indicated that most cells (∼90%) were infected with the GFP-tagged-FOXO expressing adenoviral vectors (green) but showed low levels of cleaved caspase 3 (red; 7 h and 28 h), used as an indicator of apoptosis. Thus, at 7 h and 28 h after infection when levels of FOXOA3 and FOXOA3-mDBD are elevated (Fig. 1A), cellular apoptosis was apparent in only a small number of cells (<5%, Fig. 1B).

To determine the dose-dependent increases in expression of the adenoviral expressed FOXO factors and how these related to endogenous levels of FOXO1, rat granulosa cells were infected with 0, 2.5, 5, and 10 multiplicities of infection (MOIs) of FOXOA3 and FOXOA3-mDBD (Fig. 1C). Noninfected, nontreated control granulosa cells exhibited elevated levels of endogenous FOXO1 (Fig. 1C, NT), whereas the addition of FSH/T dramatically depleted the cells of this factor, confirming previous studies (Fig. 1C, FSH/T) (24). FOXOA3 and FOXOA3-mDBD at 5–10 MOIs restored the levels of FOXO in these cells to that observed for endogenous FOXO1 before exposure to FSH/T (Fig. 1C). Thus, the levels of FOXOA3 and FOXOA3-mDBD that were achieved in these studies were maintained at a near physiological level as observed in granulosa cells of small growing follicles.

FOXOA3 and FOXOA3-mDBD impact the expression of similar and distinct sets of FSH-regulated genes in granulosa cells

Because relatively little is known about the functional roles of FOXO1 in granulosa cells, we chose to use a gain-of-function approach in which a constitutively active FOXOA3 would be predicted to enhance the expression of genes induced by endogenous FOXO1 and repress genes normally suppressed by FOXO1. By using a mutant in which the DBD is altered and impaired and which does not readily active IRE response elements (9,13), we predicted that those genes regulated by FOXO1 by mechanisms that are independent of binding to IREs would be identified as well. Accordingly, we infected rat granulosa cells with the adenoviral vectors expressing either FOXOA3 or FOXOA3-mDBD to identify FOXO-specific target genes that were either dependent or independent of the DBD and which either enhanced or suppressed specific effects of FSH on granulosa cell function. For this we harvested and plated rat granulosa cells as described above, infected them with FOXOA3 or FOXOA3-mDBD adenoviral vectors (10 MOIs) for 4 h, and then cultured the cells in the absence or presence of FSH for 12 and 24 h. At these times, total RNA was prepared and submitted to the Microarray Core Facility at Baylor College of Medicine (BCM) for analyses using the Affymetrix Rat Gene Chips. Data were analyzed by Bioconductor Software as previously described (25).

A total of 2001probes (∼4.4% of total) showed changes of 4-fold or greater between RNA from FSH-treated cells compared with FOXOA3-FSH or FOXOA3-mDBD-FSH-treated cells at 24 h. Of these, 310 gene probes exhibited overlapping patterns between FOXOA3 and the DNA binding mutant. It is also important to note that many other genes were independently regulated by either FOXOA3 (of 372, 315 increased and 57 decreased) or FOXOA3-mDBD (of 1319, 776 increased and 543 decreased), and in each case more genes were up-regulated than down-regulated. Because FOXOA3 contains an intact, functional DBD some of the genes selectively regulated by FOXOA3 are presumed to be regulated by the direct binding of FOXOA3 to transcriptional regulatory regions of these genes. By contrast, because the FOXOA3-mDBD lacks a functional IRE DBD, its effects are presumed to be mediated by its actions as a coregulatory molecule or by its binding to alternative DNA regions (9,13). Thus, these two factors control overlapping as well as distinct genes in granulosa cells. When gene annotation analyses of these probes were carried out using NIH-DAVID (26), uniquely regulated gene patterns were recognized. The most significantly enriched gene ontology terms (P <1E-04) at the biological process level 3 are listed in Fig. 2.

Figure 2.

Figure 2

Open in a new tab

Microarray analyses of genes regulated by FOXOA3 and FOXOA3-mDBD in rat granulosa cells. Cells were harvested, infected, and hormonally treated for 12 and 24 h as in Fig. 1A. Data obtained at 24 h were analyzed by computer-based programs to determine genes selectively regulated by FOXOA3 and FOXOA3-mBDB. A major class of genes regulated by the FOXO mutants included those in the fatty acid, sterol, and steroidogenic biosynthetic pathways.

FOXOA3 and FOXOA3-mDBD impact fatty acid, cholesterol (sterol), and steroid biosynthesis

Based on the microarray analyses of the 310 genes regulated by both FOXOA3 and its DBD mutant in granulosa cells, we determined that genes regulating lipid biosynthesis were among those most dramatically up-regulated by FSH and down-regulated by FOXOA3 and FOXOA3-mDBD at 24 h (Fig. 2). Within the lipid synthesis category, the genes selectively down-regulated in the adenoviral-infected cells belonged to the fatty acid, cholesterol/sterol biosynthetic, and steroidogenic pathways (Fig. 2). Real-time RT-PCR verified that essentially all genes analyzed within the cholesterol biosynthetic pathway were increased by FSH, and this increase (as well as basal levels; data not shown) was decreased by FOXOA3 and FOXOA3-mDBD. Real-time RT-PCR also showed that the induction of the genes by FSH was more dramatic than that observed by the microarray approach (Fig. 3). Because all of the genes within the cholesterol pathway are controlled by specific transcription factors, known as sterol element-binding transcription factors (Srebf1 and Srebf2)(also known as Srebp1 and Srebp2) (27,28), we searched the microarray database, and verified by real-time RT-PCR, that expression of these transcription factors was also increased by FSH/T and reduced in the FOXO1 adenoviral-infected cells (Fig. 3). Other known targets of Srebf1/2 include Fasn (fatty acid synthase), Elov6, a fatty acid elongase (29), and Scd1/2, stearoyl coenzyme A desaturases 1 and 2, rate-limiting enzymes in the biosynthesis of monounsaturated fatty acids (30,31). Expression of Elov6 and Fasn was also reduced in the adenoviral-infected cells (Figs. 2 and 4, respectively) as were Scd1 (Fig. 4) and Scd2 (−2.5-fold based on microarray analyses).

Figure 3.

Figure 3

Open in a new tab

Genes in the cholesterol and steroid biosynthetic pathway are up-regulated by FSH and down-regulated by FOXOA3 and FOXOA3-mDBD. A, The schematic illustrates genes controlling the cholesterol biosynthetic pathway and selected genes in the steroidogenic pathway as well as specific transcription factors that impact expression of these genes. Genes analyzed by real-time RT-PCR and compared with data from the microarray analyses are denoted by black boxes. B, Real-time (red bars) and microarray data (blue bars) are presented for selected genes regulated by FSH, FOXOA3, and FOXOA3-mDBD at 24 h.

Figure 4.

Figure 4

Open in a new tab

Time (panel A)- and dose (panel B)-dependent effects of FOXO1 mutants on sterol, fatty acid, and steroidogenic genes. A, Cells were cultured and infected as in Fig. 1A. Total RNA was prepared at selected time intervals after FSH/T treatment and analyzed by real time RT-PCR for selected genes regulating cholesterol and steroid biosynthesis. B, Granulosa cells were cultured without (NT, white bar) or with FSH/T (black bar) and increasing doses of FOXOA3 and FOXOA3-mDBD (0, 3.75, 7.5, and 15 MOIs) (gray bars). The FOXO1 mutants caused repression of selected genes, Srebf1, Hmgcs1,Hmcgr, Cyp51, and Ldlr even at the lowest dose as well as induction of Cyp27a1, respectively, in FSH-treated rat granulosa cells. NT, No treatment. Letters denote significant differences (P < 0.05) between treatment groups with “a” exhibiting the highest activity and decreasing levels with alphabetical order.

To determine the temporal effects of FSH as well as the dose-dependent effects of FOXOA3 and FOXOA3-mDBD, additional rat granulosa cells were cultured, infected with adenoviral vectors, and treated with or without FSH. As shown in Fig. 4A, FSH induced and FOXOA3 repressed the expression of selected sterol (Srebf1, Hmgcr, Mvk, Sqle, Lss, Cyp5 , and Ldlr), fatty acid (Fasn and Scd1), and steroidogenic (Star and Cyp11a1) genes within 24 h. Uniquely, FSH induction of Star occurred rapidly by 3 h and declined progressively from 6–12 h, whereas levels of Cyp11a1 mRNA were highest at 24 h. Of note also, the basal levels of several of these genes were also down-regulated in the infected cells at 3 and 6 h (Fig. 4A). These data provide the first documentation in granulosa cells that FSH regulates essentially all genes controlling the cholesterol biosynthetic pathway and some within the fatty acid synthesis pathway, including the transcriptional regulators Srebf1 and Srebf2.

Not all genes were repressed by expressing FOXOA3 or FOXOA3-mDBD in granulosa cells. One gene that was induced in a time- and dose-dependent manner by both FOXOA3 and FOXOA3-mDBD was Cyp27a1 (Figs. 3 and 4, A and B). Importantly, this enzyme is critical for reducing cholesterol levels in cells via the generation of 27-hydroxycholesterol, an oxysterol inhibitor of Hmgcr (32,33). In the same samples Srebf1, Hmgcs1, Hmgcr, Lss, Cyp51, and Ldlr were markedly induced by FSH but suppressed by the FOXO1 mutants, with the effects of FOXOA3 being generally more potent than those of the DNA binding mutant (Fig. 4B). As might be predicted from the results in Fig. 4, A and B, the synthesis of progesterone was reduced in media samples collected from the FOXO1 mutant-infected cells (Fig. 4B and Table 1). Thus, one specific role of endogenous FOXO1 in granulosa cells may be to prevent premature up-regulation of the cholesterol biosynthetic pathway that is essential for endogenous steroidogenesis.

Table 1.

Effects of FOXOA3 and FOXOA3-mDBD on FSH-Stimulated Production of cAMP and Progesterone

Treatments (MOI) cAMP (pmol/ml) Progesterone (ng/ml)
NT 2.3 (2.9, 1.8) ND
FSH/T 105.5 (114, 97) 17.6 ± 1.0
FSH/T, A3, 3.75 69.5 (72, 66) 3.8 ± 0.12
FSH/T, A3, 7.50 57.5 (56, 59) 2.6 ± 0.17
FSH/T, A3, 15
39.0 (38, 40)
1.9 ± 0.06
NT 4.4 (4.2, 4.5) ND
FSH/T 113 (111, 115) 16.5 ± 0.05
FSH/T, mDBD, 3.75 150 (154, 146) 9.8 ± 0.07
FSH/T, mDBD, 7.50 140 (147, 133) 7.9 ± 0.17
FSH/T, mDBD, 15.0 115.5 (113, 118) 5.5 ± 0.35
Open in a new tab

Media samples were collected from the cultures as described in Fig. 4B, boiled, and analyzed for cAMP and progesterone. FOXOA3, but not FOXOA3-mDBD, dose dependently reduced levels of cAMP in the culture medium collected 24 h after FSH treatment. Progesterone concentrations were reduced to a greater extent by FOXOA3 than by FOXOA3-mDBD. cAMP is representative of two separate experiments. Progesterone is representative of two separate experiments run in triplicate. ND, Not determined; NT, not treated. 

As shown herein and reported previously, FSH induces the expression of aromatase (Cyp19) (34), steroid acute regulatory protein (Star) (35), and cholesterol-side chain cleavage cytochrome P450 (Cyp11a1) in granulosa cells (36,37) (Figs. 2, 4, and 5 and data not shown). The induced expression of these genes was also suppressed in the FOXOA3 and FOXOA3-mDBD infected cells (Figs. 2, 4, and 5). When we searched the database for key transcription factors that regulate these genes, we noted and confirmed by real-time RT-PCR that the expression of Nr5a1 (also known as Sf1), Nr5a2 (also known as Lrh1), and Gata4 was repressed (Fig. 5 and data not shown) (38,39,40). By striking contrast, FSH repressed and FOXOA3 increased expression of Nr0b1 (DAX), a known antagonist of nuclear receptors (41), whereas FOXOA3-mDBD induced expression of Nr0b2 (short heterodimer partner), an antagonist expressed at high levels in liver but also in the testis (42). By contrast, levels of mRNAs for other transcription factors (Sp1, Cebpb, Creb, Usf1) that impact steroidogenic gene expression did not change by microarray analyses (data not shown).

Figure 5.

Figure 5

Open in a new tab

FSH receptor, FSH target genes, and selected other genes are differentially regulated by FOXOA3 and FOXOA3-mDBD. A, Cells were cultured and infected with 10 MOI adenoviral vectors as in Fig. 1. Real-time RT-PCR analyses document that FSH receptor (Fshr), aromatase (Cyp19), and LH receptor (Lhcgr) mRNAs are increased by FSH (black bar) and reduced by FOXOA3 and FOXOA3-mDBD (gray bars). Nr5A1 (data not shown) and Nr5a2, transcription factors that impact Cyp19 promoter were down-regulated by FOXOA3-mDBD but not by FOXOA3. By contrast, transcriptional antagonists of Nr5a1 and Nr5a2, Nr0b1 and Nr0b2, were differentially up-regulated or down-regulated by the FOXO mutants. B, Other genes of interest were also up-regulated differentially by the mutant FOXO factors. NT, No treatment. Representative of four separate experiments. Letters denote significant differences (P < 0.05) between treatment groups with “a” exhibiting the highest activity and decreasing levels related to alphabetical order.

FOXOA3 and FOXOA3-mDBD impact the expression of differentiation-related and novel genes

To determine whether the responses to FSH were related to changes in the expression of the FSH receptor, we analyzed the levels of Fshr mRNA (Fig. 5). Both time- and dose-dependent decreases in Fshr mRNA occurred in the infected cells, but the magnitude of the decreases in Fshr mRNA levels was not as dramatic as that of many other genes (Figs. 3 and 4 and data not shown), including Cyp19, Cyp11a1, and Lhcgr, which are known targets of FSH action (Figs. 4A and 5) (43). Thus, an additional role of endogenous FOXO1 appears to be to prevent premature differentiation responses to FSH, in part, by monitoring levels of FSH receptors. Importantly, FSH stimulation of cAMP accumulation in media samples of cultured granulosa cells was reduced 30–60% in response to increasing doses of FOXOA3 but showed no change in response to increasing amounts of FOXOA3-mDBD (Fig. 4B and Table 1). Because FOXOA3 and its DBD mutant both exerted potent negative regulatory effects on the cholesterol and steroidogenic biosynthetic gene expression, it seems unlikely that the reduced levels of cAMP per se is the major mechanism controlling expression of these genes. By contrast, reduced cAMP levels are likely to impact genes selectively regulated by FOXOA3 compared with those altered by FOXOA3-mDBD (Fig. 2).

Microarray data also indicated that FOXOA3 and the mDBD mutant altered other cell-signaling pathways. Specifically, expression of some IGF pathway components that were down-regulated by FSH were increased approximately 5-fold (Irs2). Interestingly, Irs2 is regulated by FOXO1 in hepatocytes (20,22). Other genes were reduced approximately 5-fold (Igfbp5), even in the absence of FSH. Although FOXO1 has been associated with cell apoptosis, cell cycle arrest, and the expression of Cdkn1a and Cdkn1b, these cellular pathways and genes were not the major ones altered by either FOXOA3 or FOXOA3-mDBD in the granulosa cells (Fig. 5). Moreover, evidence for apoptosis was minimal in the cells expressing FOXOA3, at levels similar to those of endogenous FOXO1 (Fig. 1). Expression of Klf5, a member of the SP1 family of transcription factors, was also induced by FOXOA3 and FOXOA3-mDBD but potently down-regulated by FSH (Fig. 5). Members of the fibroblast growth factor (FGF) family (Fgf13 and Fgf9) were selectively increased by FOXOA3 or FOXOA3-mDBD, respectively (Fig. 5). Lastly, FOXOA3 and its DNA binding mutant suppressed expression genes involved in glucose metabolism, namely Pfkm and Pfkp (Fig. 5).

Expression of Foxo1 and sterol biosynthetic genes in murine granulosa cells in vivo

We next sought to determine the temporal expression patterns of Foxo1 and the sterol biosynthetic genes in granulosa cells during follicle development and luteinization in vivo. For this, granulosa cells were isolated from immature mice before and after hormonal stimulation with equine (e) CG to induce the growth of preovulatory follicles and hCG to initiate ovulation and luteinization. As shown in Fig. 6A, expression of Foxo1 mRNA increased in response to eCG but decreased in response to hCG, confirming earlier studies in the rat and mouse (24). The decrease in Foxo1 mRNA was associated with the selective down-regulation of immunoreactive FOXO protein in granulosa cells of preovulatory and ovulatory mouse follicles but not small follicles (Fig. 6B). Therefore, the increase in Foxo1 mRNA in response to eCG alone reflects the increase in Foxo1 mRNA that occurs selectively in small follicles but not in antral follicles. Western blot analyses confirmed that the high levels of FOXO1 protein present in granulosa cells of immature mice decreased rapidly within 1 h after treatment with eCG and remained low at 24 h after treatment (Fig. 6C). The decline in FOXO1 protein was associated with increased levels of phosphorylated FOXO1 that were observed within 0.5 h and that persisted to 4 h after eCG treatment. In these same extracts the phosphorylation of AKT increased at 0.5 h and persisted to 24 h whereas levels of AKT were unchanged. Thus, there is rapid phosphorylation and inactivation as well as degradation of FOXO1 in granulosa cells of antral follicles in response to eCG in vivo (Fig. 6, B and C). Because E2F1 is a potential transcriptional regulator of Foxo1 (44), the levels of E2f1 mRNA were also analyzed and shown to change in a pattern identical to that of Foxo1 (Fig. 6A).

Figure 6.

Figure 6

Open in a new tab

Expression of Foxo1 mRNA and protein and cholesterol biosynthetic genes in murine granulosa cells during follicular development, ovulation, and luteinization in vivo. Granulosa cells (gray bars) were harvested from ovaries obtained from immature mice (imm; d 23 of age) before and after injections of eCG (4 IU) treatment that stimulates preovulatory follicle development and from periovulatory follicles at selected time intervals (2–16 h) after hCG (4 IU) that stimulates ovulation. Ovaries (black bars) were also isolated from mice at 24 h and 48 h after hCG (4 IU) when corpora lutea were abundant. Ovaries (black bars) were also isolated from pregnant mice on d 8 and d 22 of gestation, times when corpora lutea are fully functional and regressing, respectively. Total RNA was prepared from granulosa cells or whole ovaries for real-time RT-PCR analyses. Expression of genes was normalized to L19. A, Real-time RT-PCR of endogenous Foxo1 and E2f1 mRNA in granulosa cells. B, Immunofluorescent localization of FOXO1 protein in follicles before and after treatment with eCG, 48 h, and eCG+hCG, 8 h. Note that immunoreactive FOXO1 is present at elevated levels in small growing follicles present in ovaries of immature, eCG-, and hCG- treated mice (yellow arrowheads). By contrast, FOXO1 rapidly declines in large antral preovulatory follicles of eCG- and hCG-treated mice (white arrowheads). C, Western blot of granulosa cell extracts at selected time intervals showing rapid and selective loss of FOXO1protein in granulosa cells in response to eCG. D, Real time RT-PCR of genes in the cholesterol biosynthetic and fatty acid pathways. Letters denote significant differences (P < 0.05) between treatment groups with “a” exhibiting the highest activity and decreasing levels with alphabetical order. preg, Pregnant.

In these same murine granulosa cell RNA samples, the levels of Cyp27a1 declined in a pattern similar to that of Foxo1 (Fig. 6D). By contrast, mRNAs encoding selective enzymes of the cholesterol biosynthetic pathway were increased 3- to 5-fold by eCG (Hmgcs1, Mvk, Sqle, Cyp51,Tm7sf2, and Dhcr24) and further by hCG (Hmgcs1, Sqle, Cyp51, and Tm7sf2). Most striking were the 35- to 100-fold elevated levels of Hmgcs1, Sqle, Cyp51, Tm7sf2 mRNAs in whole ovaries of mice 48 h after hCG and in functional corpora lutea of pregnant mice on d 8 of gestation. In addition, Scd1 was increased 10-fold by eCG and 100-fold by hCG and was high in functional corpora lutea present in ovaries of d 8 pregnant mice. By contrast, expression of genes in the cholesterol biosynthetic pathway were low in corpora lutea undergoing functional luteolysis on d 22 of pregnancy. Somewhat surprisingly, although expression of Srebf1 and Srebf2 mRNAs increased 3- to 4-fold by eCG, respectively, they were not further elevated by hCG or as a consequence of luteinization.

To determine whether FOXOA3 or FOXOA3-mDBD could impact transcription of the sterol biosynthetic enzymes in murine ovarian cells, granulosa cells were isolated from immature mice 24 h after treatment with eCG to increase the number of cells harvested. When these cells were infected with either FOXO1 mutant, the FSH-induced expression of sterol biosynthetic genes was reduced in a dose-dependent manner (Fig. 7) as observed in the rat cells (Fig. 4). Although not shown, Srebf1 was also increased by FSH and reduced by FOXOA3 but not by the DBD mutant. Likewise, in the murine cells as in the rat cells, the FOXO mutants increased expression of Cyp27a1 and Bmp2 mRNAs (Fig. 7).

Figure 7.

Figure 7

Open in a new tab

FOXOA3 and FOXOA3-mDBD down-regulate cholesterol biosynthetic, fatty acid genes in mouse granulosa cells in culture. Granulosa cells were isolated from immature mice 24 h after treatment with eCG and plated in serum-coated 24-well plates in serum-free, defined medium. The cells were infected with adenoviral vectors for 4 h (gray bars) as described for the rat granulosa cell cultures and treated without [not treated (NT, white bar)] or with forskolin (10 μm) and T (10 ng/ml) for 24 h (black bars). Total RNA was prepared for RT-PCR analyses as described in Materials and Methods. Letters denote significant differences (P < 0.05) between treatment groups with “a” exhibiting the highest activity and decreasing levels related to alphabetical order.

Discussion

The transcription factor FOXO1 is expressed at high levels in granulosa cells of growing follicles, but its expression is silenced as these cells terminally differentiate to nondividing, terminally differentiated luteal cells (24,45). Because most granulosa cells of growing follicles become apoptotic when follicles undergo atresia, a critical role for FOXO1 in this process in granulosa cells has been suggested by many but is yet to be proven (10). Likewise, because FOXO1 has been shown to induce Cdnk1b (p27KIP) (46) and decrease Ccnd2 (Cyclin D2) (10), FOXO1 may be a potential key regulator of cell cycle progression in these cells (10). In other cells, FOXO1 has been linked to oxidative stress, DNA repair, cell longevity, myoblast cell fusion, and muscle cell atrophy, as well as metabolic homoeostasis, by regulating intracellular production and metabolism of glucose and lipids (17,47,48). Conversely, targeted disruption of all Foxo factors (Foxo1, Foxo3, and Foxo4) causes thymic lymphomas and hemiangiomas (6,7). Thus, the effects of FOXO1 are complex and cell-context specific.

Our microarray data revealed that, in addition to the known FOXO targets, there are some novel granulosa cell-specific responses to FOXOA3 and FOXOA3-mDBD as well as FOXOA3 vs. FOXOA3-mDBD (Fig. 2). Specifically, expressing the FOXO1 mutants in granulosa cells profoundly reduced FSH-induced expression of genes controlling lipid, sterol, and steroid biosynthesis. These results document, for the first time, that in granulosa cells FSH selectively increases expression of the transcription factors, Srebf1 and Srebf2, that are known to regulate cholesterol and lipid biosynthetic genes (27,28). By contrast, FOXOA3 and FOXOA3-mDBD are potent transcriptional repressors of these genes. Because each enzyme of the cholesterol biosynthetic pathway is regulated, in part, by sterol element-binding transcription factor (SREBF)1 or SREBF2 (27) and because SREBF1 and SREBF2 can autoregulate their own expression (28), a decrease in either one or both of these transcription factors leads to suppression of the entire pathway. Indeed, we document that mRNA levels for each enzyme critical for cholesterol biosynthesis were reduced by the FOXO1 mutants in granulosa cells, most potently Hmgcs1, Hmcgr, Mvk, Sqle, Lss, Cyp51, and Tm7sf2. Thus, FOXO1 has the potential to control cholesterol biosynthesis and metabolism in granulosa cells of growing follicles, in part by regulating the expression of Srebf1 and Srebf2. By contrast, Cyp27a1, an enzyme that modifies cholesterol to generate oxysterol products capable of inhibiting Hmgcr, was induced selectively by FOXOA3 and FOXOA3-DBD in rat and mouse granulosa cells.

That granulosa cells do not acquire a high level of steroidogenic potential until after the LH/hCG surge further supports the notion that high levels of FOXO1 in conjunction with other factors, such as NR0B1 (DAX), may repress cholesterol biosynthesis in granulosa cells of growing follicles and that rapid phosphorylation as well as loss of Foxo1 expression during luteinization may allow the activity of this metabolic pathway to increase. Indeed, all cholesterol pathway genes examined, including Srebf1 and Srebf2, increased in response to eCG alone, and this was associated with the rapid phosphorylation and reduction of FOXO1 protein in granulosa cells of preovulatory follicles. That Srebf1 and Srebf2 did not increase further during luteinization indicates that regulation of these genes in vivo depends on additional regulatory molecules. Moreover, because the patterns of induction of the cholesterol biosynthetic genes are not entirely the same and seem quite diverse, factors in addition to SREBF1/2 likely impact their transcription during luteinization. For example, the delayed induction of Scd1 and the relatively higher expression levels of Tm7sf2 in luteinized ovaries of d 8 pregnant mice compared with the hCG primed immature mice indicate that factors other than the depletion of FOXO1 have a major impact on the expression of these genes. That levels of Cyp27a1 mRNA decreased in a manner identical to that of Foxo1 in luteinizing granulosa cells combined with the potent induction of Cyp27a1 expression by the FOXO1 mutants in cultured granulosa cells indicates that this gene is likely to be a direct FOXO1 target in granulosa cells. Foxo1, Cyp27a1, and Lxr1 are the mammalian orthologs of Daf16, Daf9, and Daf12 in C. elegans that impact reproduction, larval diapause, and longevity. These similarities raise the intriguing possibility that the interactions of these factors impact longevity in mammals and the worm by controlling specific metabolic pathways in gonadal cells as well as other organs (49,50).

These data extend previous observations that the effects of FSH in cultured granulosa cells can be enhanced by adenoviral expression of PKB/AKT (51). Because AKT can be phosphorylated and activated by FSH (52) and because pAKT phosphorylates and thereby inactivates FOXO1 (24,52), it is tempting to speculate that the effects of AKT in the adenoviral model (51) may be mediated, in part, by reducing the inhibitory effects of FOXO1 on selected sterol and steroidogenic genes. Our results also extend the pioneering studies by Eppig and colleagues (33) who show that cumulus cells, but not oocytes, express genes encoding the enzymes that control cholesterol biosynthesis, and that in cumulus cells, the expression of these genes is regulated, in part, by the oocyte-derived factor, principally bone morphogenetic protein 15. These studies also reported that removal of oocytes from cumulus cell oocyte complexes increased expression of Cyp27a1. Because FOXO1 mutants also increased the expression of Cyp27a1, it is tempting to speculate that FOXO1 itself or via its regulation of a growth factor(s) controls transcription of the Cyp27a1 gene in vivo. In addition, FSH has been shown to induce expression of Cyp51 in granulosa cells (53). Because sterols produced by this enzyme may impact meiosis, sterols, in addition to cholesterol, are likely to regulate key functions in cumulus cells and oocytes (53). Specific lipids may also be important for oocyte viability and quality. Therefore, the induction of Scd1 and Scd2 (30) (Fig. 6) that are also transcriptionally regulated by Srepf1/Srebf2 may also be important to oocyte and follicular cell functions. With the exception of the studies by Su et al. (33), Yamashita et al. (54), Ning et al. (53), and Moreau et al. (30), the expression and functional relevance of the lipid/cholesterol pathway genes in granulosa cells and cumulus cells have been largely understudied. Therefore, the potential impact of FOXO1 and oocyte factors on somatic cell functions provides important new insights into the regulation of cholesterol biosynthesis in the developing and ovulating follicles.

Our data expand the growing list of metabolic processes regulated by overexpressing or reducing FOXO1 mutants in mammalian cells (17,19,20,21,22,23). In recent studies, liver-specific expression of FOXOA3 was accomplished in transgenic mice by expressing FOXOA3 driven by the α1-antitrypsin promoter. In this model, genes regulating glucose metabolism were markedly altered leading to increased gluconeogenesis and decreased glycolysis (20). The impact of FOXO1 on hepatic glucose metabolism has been supported further by recent studies showing that disruption of Foxo1 in liver reduces glucose production, impairs cAMP-induced gluconeogenesis, and can alleviate the diabetic condition of insulin-resistant mice lacking Irs1 and Irs2 (22,23). Additionally, liver-specific expression of FOXOA3 suppressed the expression of genes regulating fatty acid metabolism and cholesterol biosynthesis. Moreover, these studies are the first to show that expression of Srebf1, but not Srebf2, was suppressed in the liver in vivo as well as in hepatocytes in culture (20). By contrast, injection of an adenoviral vector expressing FOXOA3 led to an increase in Srebf1levels in livers of infected mice (21). Although the reasons for these different responses are not clear, they may be due to the time intervals or absolute levels and duration of FOXO1 expression in each model. FOXOA3 and FOXOA3-DBD also appear to impact some genes involved in glucose metabolism in granulosa cells (Figs. 5 and 6), but these pathways may not be primary targets in these cells as they are in liver (20,47), pancreatic β-cells (19), and other cells (17).

The specific mechanisms by which FOXO1 and its mutants regulate cholesterol biosynthesis in granulosa cells appear to be mediated, in part, by their ability to impact the expression of both Srebf1 and Srebf2 genes. Because the promoters of the Srebf1 and Srebf2 genes have functional sterol response elements (SREs) and SP1 binding sites (28), it is possible that FOXO1 blocks either one or both of these sites, or to binding sites for other transcription factors that might impact expression of these genes (12). It is also possible that FOXOA3 and FOXOA3-mDBD interact directly with SREBF1 or SREBF2 to prevent their transcriptional activity. Because FOXL2 is expressed in granulosa cells and has been implicated in the regulated expression of Srebf1, Nr5a2, Cyp19, Pgc1a, and Apoa1, FOXO1 may compete with FOXL2 specific regulatory factors (55,56,57) or perhaps FOXOA3 reduces the expression of Foxl2. Lastly, because FOXO1 can interact with multiple nuclear hormone receptors (13) possibly including liver X receptor (LXR) and retinoic X receptor (RXR) (11,58), because RXRβ and LXRβ are expressed in granulosa cells but are not regulated by FOXO1 mutants in these cells (data not shown) and because these nuclear receptors are known to impact the promoter activity of Srebf1c (58) and cholesterol biosynthesis (59), it is tempting to speculate that the FOXO1 mutants might impact the functions of RXRβ and/or LXRβ in granulosa cells.

The ability of FOXOA3 and its DBD mutant to block FSH-induced expression of other genes, such as Lhcgr, Cyp19, Cyp11a1, and Star, indicates that additional transcriptional processes are suppressed by FOXO1. Impaired expression of Lhcgr likely involves altered function of the critical SP1/3 binding sites on the promoter and the possible recruitment of histone deacetylases by FOXO1, although interactions of FOXO and SP1/3 have not been defined (60,61). The reduced expression of the steroidogenic genes is associated with suppression of specific transcription factors, Nr5a1, Nr5a2, Gata4 (39,62) as well as an increase in the transcriptional repressors Nr0b1(Dax1) and Nr0b2 (short heterodimer partner). Thus, in addition to reducing cholesterol biosynthesis, FOXOA3 impaired expression of genes involved in FSH-mediated differentiation. Although cAMP levels were reduced in the FOXOA3-infected cells, this was not observed in the FOXOA3-mDBD-infected cells. Therefore it is unlikely that the effects of FOXO1 mutants on genes involved in cholesterol biosynthesis and differentiation are controlled exclusively by changes in the levels of cAMP. However, some genes regulated specifically by FOXOA3 alone, and not by FOXOA3-mDBD, could be targets of both FOXO1 and cAMP. That FOXOA3-mDBD can impact genes in a manner distinct from FOXOA3 further suggests that alteration of the DBD not only reduces FOXOA3-mDBD from binding to IRE promoter elements but may also change its affinity or accessibility for other nuclear regulatory molecules. Because FOXOA3 and FOXOA3-mDBD exerted similar impacts on the cholesterol biosynthetic pathway genes suggests that, in this context, they are regulating transcription most likely via non-IRE DNA-binding mechanisms. However, their binding to other DNA regions cannot be excluded.

It is important to note that not all genes were suppressed by the FOXO1 mutants. Indeed many genes in addition to Cyp27a1 were increased by FOXOA3 and FOXOA3-mDBD, including Fgf13 that increased by FOXOA3 (3- to 7-fold) and Fgf9 that was increased 2-fold by FOXOA3 and 23- to 51-fold by FOXOA3-mDBD (Fig. 5; microarray). To what extent these growth factors impact granulosa cell functions remains to be determined. FGF13 is not a secreted form of FGF but an intracellular form that appears to be expressed and highly functional in the development of neuronal tissues, but not much is known about its intracellular functions (63). By contrast, FGF9 plays a key role in gonadal development (64,65).

In summary, these studies provide the novel information that FOXO1 has the potential to play a critical role(s) in granulosa cell function, in part by regulating FSH-induced expression of genes involved in cholesterol biosynthesis. Although reduced responsiveness to FSH may be mediated by the reduction of Fshr, this appears to be only part of the mechanism because genes within the cholesterol pathway, as well as Cyp19 and Lhcgr, are reduced more than that of Fshr and key regulatory factors such as Nr5a1, Nr5a2, and Gata4. Although the molecular mechanisms by which FOXO acts remain to be clearly dissected, many effects appear to involve the role of FOXO1 as a coregulatory factor rather than as a transcription factor. This conclusion is based on the observations that 13.3% of the genes affected by expressing FOXOA3 were also altered by the FOXOA3 DBD mutant. Therefore, these results provide additional evidence that the role of FOXO1 in granulosa cells involves pleotrophic mechanisms. Clearly metabolic homeostasis is one target of FOXO1 function in these cells.

Materials and Methods

Animals

Immature female Sprague Dawley rats and C57Bl/6 mice at 21 d of age were purchased (Harlan Inc., Indianapolis, IN), housed under a 16-h light, 8-h dark schedule in the Center for Comparative Medicine at BCM and provided food and water ad libitum. Animals were treated in accordance with the NIH Guide for Care and Use of Laboratory Animals as approved by the Animal Care and Use Committee at BCM.

Rats were injected sc for 3 d with 1.5 mg estradiol in 0.2 ml propylene glycol to stimulate granulosa cell proliferation. Ovaries were isolated and granulosa cells harvested by needle puncture as previously. The cells were cultured and infected with adenoviral vectors as described below.

Mice were injected sc with 4.0 IU eCG to stimulate follicular development followed 48 h later with 4.0 IU hCG to stimulate ovulation and luteinization. Ovaries were isolated from nontreated mice (control) as well as mice at selected time intervals after eCG (48 h) and hCG (4, 8, 16, 24, and 48 h). Granulosa cells were isolated from ovaries at all time intervals except at 24 and 48 h after hCG when luteinization was complete. Total RNA was prepared from granulosa cells or whole ovaries by the Qiagen Kit according to the manufacturer’s specifications (QIAGEN, Chatsworth, CA).

Granulosa cell culture

Granulosa cells from estradiol-treated rats and eCG-primed mice (24 h) were plated in defined medium (serum-free DMEM-F12) in six-well or 12-well cultures dishes that were precoated with 1% serum. The cells were infected with adenoviral vectors expressing GFP (control), constitutively active FOXO1 (FOXOA3; FKHR:AAA) or a DNA-binding mutant (FOXOA3-mDBD, FKHR;HRAAA) that were obtained from Dr. William Sellers (Harvard University) and have been described previously (9,13). The cells were infected 5 or 6 h after plating with 10 MOIs of each adenoviral vector for 4 h or at MOI values as specified in the figure legends. The cells were then washed and cultured in the presence or absence of hormones [FSH (100 ng/ml) + testosterone (10 ng/ml)] for 3, 6, 12, or 24 h as indicated in each experiment. Either total RNA or whole-cell lysates were prepared in duplicate at each time interval. RNA was isolated by the Qiagen Kit as indicated above. Protein extracts for Western blots were prepared in 10% boiling SDS buffer as described previously (52). In addition, some cells were plated on serum-coated coverslips for microscopic and immunofluorescent analyses. These cells were fixed in 4% paraformaldehyde at room temperature for 30 min and stored in sterile PBS until used.

Western blots

To determine the expression levels of endogenous and mutant FOXO proteins in the cultured cells, 20 μl of protein extracts were loaded per well of a 12% acrylamide gel. Proteins were transferred to Imobilin filters by semidry transfer and then processed by routine procedures. Immunoreactive bands were visualized by blotting with anti-FOXO1 antibody (1:1000; Cell Signaling Technology, Beverly, MA) followed by secondary antibody and enhanced chemiluminescence detection protocols as previously described (52) Immunoreactive actin or AKT was used as a loading control.

Immunofluorescent analyses

Rat granulosa cells were plated on serum-coated coverslips and infected with adenoviral vectors as described in the figure legends. The infections efficiency was determined by visualizing the GFP-tagged proteins in the control vector and the FOXOA3 and FOXOA3-mDBD vectors. Apoptosis was evaluated by immunostaining for cleaved caspase 3 as reported previously (66). Cell morphology was determined by differential interference contrast imaging using a ZeissAxioplan2 microscope (Carl Zeiss, Thornwood, NY) in the Integrated Microscopy Core at BCM.

RIAs

cAMP was measured by RIA as described previously (51). Progesterone was analyzed by the Ligand Assay Core Laboratory at the University of Virginia, Specialized Cooperative Centers Program in Reproduction and Infertility Research.

Microarray analyses and real-time RT-PCR

To determine genes regulated by expressing FOXOA3 or its DBD mutant, total RNA was prepared in duplicate from control and adenoviral infected rat granulosa cells at 12 and 24 h after treatment with FSH/T or vehicle (PBS) alone. These intervals were chosen because rat granulosa cell differentiation is induced progressively by FSH/T at these times (Fig. 4) (34). RNA was submitted to the Microarray Core Facility at BCM, analyzed by a Bioanalyser 2100 (Agilent Technologies, Wilmington, DE) before microarray hybridization. Regulated genes were identified using Rat Genome 230.2 Arrays (Affymetrix, Santa Clara, CA). The Affymetrix Gene Chip data were analyzed by specific programs, including Bioconductor Software as previously described (25) as well as NIH-DAVID (26). Selected genes the expression of which was markedly altered by the mutant FOXO factors compared with controls were verified by real-time RT-PCR using specific primers and normalized to L19. RT-PCR was performed using the Roto-Gene 3000 Thermocycler (Corbett Research, Sydney, Australia).

Statistics

The data for real-time RT-PCR assays are represented as means ± sd. Data were analyzed by using GraphPad Prism Programs (ANOVA or t test; GraphPad Prism, San Diego, CA) to determine significance. Values were considered significantly different if P ≤ 0.05 or P ≤ 0.01.

Acknowledgments

We thank the Ligand Assay Core Laboratory at the University of Virginia for analyzing progesterone in media samples of the cultured cells and the Integrated Microscopy Core Laboratory at BCM for their expertise and equipment.

Footnotes

Current addresses are: Department of Biochemistry (I.H.G.) and Department of Physiology (I.G.-R.), University of Las Palmas, Las Palmas Grand Canaria, Spain.

The Ligand Assay Core Laboratory (University of Virginia) and Integrated Microscopy Core Laboratory at BCM are each funded by an NIH-SCCPIR Grant.

Disclosure Summary: The authors have nothing to disclose.

First Published Online February 5, 2009

Abbreviations: CG, Choriogonadotropin; FGF, fibroblast growth factor; FSH/T, FSH/ testosterone; GFP, green fluorescent protein; IREs, insulin response elements; LXR, liver X receptor; mDBD, mutant DNA-binding domain; MOI, multiplicity of infection; RXR, retinoic X receptor; SREBF, sterol element-binding transcription factor.

References

  1. Hsin H, Kenyon C 1999 Signals from the reproductive system regulate the lifespan of C. elegans. Nature 399:362–366 [DOI] [PubMed] [Google Scholar]
  2. Guarente L, Kenyon C 2000 Genetic pathways that regulate ageing in model organisms. Nature 408:255–262 [DOI] [PubMed] [Google Scholar]
  3. Hosaka T, Biggs 3rd WH, Tieu D, Boyer AD, Varki NM, Cavenee WK, Arden KC 2004 Disruption of forkhead transcription factor (FOXO) family members in mice reveals their functional diversification. Proc Natl Acad Sci USA 101:2975–2980 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Huang H, Tindall DJ 2007 Dynamic FoxO transcription factors. J Cell Sci 120:2479–2487 [DOI] [PubMed] [Google Scholar]
  5. Castrillon DH, Miao L, Kollipara R, Horner JW, DePinho RA 2003 Suppression of ovarian follicle activation in mice by the transcription factor Foxo3a. Science 301:215–218 [DOI] [PubMed] [Google Scholar]
  6. Tothova Z, Kollipara R, Huntley BJ, Lee BH, Castrillon DH, Cullen DE, McDowell EP, Lazo-Kallanian S, Williams IR, Sears C, Armstrong SA, Passegué E, DePinho RA, Gilliland DG 2007 FoxOs are critical mediators of hematopoietic stem cell resistance to physiological oxidative stress. Cell 128:325–339 [DOI] [PubMed] [Google Scholar]
  7. Paik JH, Kollipara R, Chu G, Ji H, Xiao Y, Ding Z, Miao L, Tothova Z, Horner JW, Carrasco DR, Jiang S, Gilliland DG, Chin L, Wong WH, DH C, DePinho RA 2007 FoxOs are lineage-restricted redundant tumor suppressors and regulate endothelial cell homeostasis. Cell 128:309–323 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Liu L, Rajareddy S, Reddy P, Du C, Jagarlamudi K, Shen Y, Gunnarsson D, Selstam G, Boman K, Liu K 2007 Infertility caused by retardation of follicular develoment in mice with oocyte-specific expression of Foxo3a. Development 134:199–209 [DOI] [PubMed] [Google Scholar]
  9. Ramaswamy S, Nakamura N, Sansal I, Bergeron L, Sellers WR 2002 A novel mechanism of gene regulation and tumor suppression by transcription factor FKHR. Cancer Cell 2:81–91 [DOI] [PubMed] [Google Scholar]
  10. Park Y, Maizels ET, Feiger ZJ, Alam H, Peters CA, Woodruff TK, Unterman TG, Lee EJ, Jameson JL, Hunzicker-Dunn M 2005 Induction of cyclin D2 in rat granulosa cells requires FSH-dependent relief from FOXO1 repression coupled with positive signals from Smad. J Biol Chem 280:9135–9148 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. van der Vos KE, Coffer PJ 2008 FOXO-binding partners: it takes two to tango. Oncogene 27:2289–2299 [DOI] [PubMed] [Google Scholar]
  12. Gomis RR, Alarcón C, He W, Wang Q, Seoane J, Lash A, Massague J 2006 A FoxO-Smad synexpression group in human keratinocytes. Proc Natl Acad Sci USA 103:12747–12752 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Rudd MD, Gonzalez-Robayna I, Hernandez-Gonzalez I, Weigel NL, Bingman 3rd W, Richards JS 2007 Constitutively active FOXO1 and a DNA binding domain mutant exhibit distinct co-regulatory functions to enhance progesterone receptor A activity. J Mol Endocrinol 38:673–690 [DOI] [PubMed] [Google Scholar]
  14. Zhao HH, Herrera RF, Coronado-Heinsohn E, Yang MC, Ludes-Meyers JH, Seybold-Tilson KJ, Nawaz Z, Yee D, Barr FG, Diab SG, Brown PH, Fuqua SA, Osborne CK 2001 Forkhead homologue in rhabdomyosarcoma functions as a bifunctional nuclear hormone receptor-interacting protein with coactivator and corepressor functions. J Biol Chem 276:27907–27912 [DOI] [PubMed] [Google Scholar]
  15. Hang H, Muddiman DC, Tindall DJ 2004 Androgens negatively regualte forkhead transcription factor FKHR (FOXO1) through a proteolytic mechanism in prostate cancer cells. J Biol Chem 279:13866–13877 [DOI] [PubMed] [Google Scholar]
  16. Reddy P, Liu L, Adhikari D, Jagarlamudi K, Rajareddy S, Shen Y, Du C, Tang W, Hämäläinen T, Peng SL, Lan Z-J, Cooney AJ, Huhtaniemi I, Liu K 2008 Oocyte-specific deletion of Pten causes premature activation of the primordial follicle pool. Science 319:611–613 [DOI] [PubMed] [Google Scholar]
  17. Gross DN, van den Heuvel AP, Birnbaum MJ 2008 The role of FoxO in the regulation of metabolism. Oncogene 27:2320–2336 [DOI] [PubMed] [Google Scholar]
  18. Kim JJ, Buzzio OL, Li S, Lu Z 2005 Role of FOXO1A in the regulation of insulin-like growth factor binding protein-1 in human endometrial cells; interaction with progesterone receptor. Biol Reprod 73:833–839 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Buteau J, Shlien A, Foisy S, Accili D 2007 Metabolic diapause in pancreatic b-cells expressing a gain-of-function mutant of the forkhead protein Foxo1. J Biol Chem 282:287–293 [DOI] [PubMed] [Google Scholar]
  20. Zhang W, Patil S, Chauhan B, Guo S, Powell DR, Le J, Klotsas A, Matika R, Xiao X, Franks R, Heindenreich KA, Sajan MP, Farese RV, Stolz DB, Tso P, Koo S-H, Montminy M, Unterman TG 2006 FoxO1 regulates multiple metabolic pathways in the liver. J Biol Chem 281:10105–10117 [DOI] [PubMed] [Google Scholar]
  21. Matsumoto M, Han S, Kitamura T, Accili D 2006 Dual role of transcription factor FoxO1 in controlling hepatic insulin sensitivity and lipid metabolism. J Clin Invest 116:2464–2472 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Matsumoto M, Pocai A, Rossetti L, DePinho RA, Accili D 2007 Impaired regulation of hepatic glucose production in mice lacking forkhead transcription factor Foxo1 in liver. Cell Metab 6:208–216 [DOI] [PubMed] [Google Scholar]
  23. Dong XC, Copps KD, Guo S, Li Y, Kollopara R, DePinho RA, White MF 2008 Inactvation of hepatic Foxo1 by insulin signaling is required for adaptive nutrient homeostasis and endocrine growth regulation. Cell Metab 8:65–76 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Richards JS, Sharma SC, Falender AE, Lo YH 2002 Expression of FKHR, FKHRL1 and AFX genes in the rodent ovary: evidence for regulation by IGF-1, estrogen and the gonadotropins. Mol Endocrinol 16:580–599 [DOI] [PubMed] [Google Scholar]
  25. Hernandez-Gonzalez I, Gonzalez-Robayna IJ, Shimada M, Wayne CM, Ochsner SA, White L, Richards JS 2006 Gene expression profiles of cumulus cell oocyte complexes (COCs) during ovulation reveal cumulus cells express neuronal and immune-related genes: does this expand their role in the ovulation process? Mol Endocrinol 20:1300–1321 [DOI] [PubMed] [Google Scholar]
  26. Dennis G J, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA 2003 DAVID: database for annotation, visualization, and integrated discovery. Genome Biol 4:3 [PubMed] [Google Scholar]
  27. Sakakura Y, Shimano H, Sone H, Takahashi A, Inoue N, Toyoshima H, Suzuki S, Yamada N, Inoue K 2001 Sterol regulatory element binding proteins induce an entire pathway of cholesterol synthesis. Biochem Biophys Res Commun 286:176–183 [DOI] [PubMed] [Google Scholar]
  28. Weber LW, Boli M, Stampfl A 2004 Maintaining cholesterol biosynthesis: sterol regulatory element binding proteins. World J Gastroenterol 10:3081–3087 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kumadaki S, Matsuzaka T, Kato T, Yahaqi N, Yamamoto T, Okada S, Kobayashi K, Takahashi A, Yatoh S, Suzuki H, Yamada N, Shimano H 2008 Mouse Elovl-6 promoter is an SREBP target. Biochem Biophys Res Commun 368:261–266 [DOI] [PubMed] [Google Scholar]
  30. Moreau C, Froment P, Tosca L, Moreau V, Dupont J 2006 Expression and regulation of the SCD2 desaturase in the rat ovary. Biol Reprod 74:75–87 [DOI] [PubMed] [Google Scholar]
  31. Sampath H, Ntambi JM 2006 Stearoyl-coenzyme A desaturase 1, sterol regulatory element binding protein-1c and peroxisome proliferator-activated receptor-α: independent and interactive roles in the regulation of lipid metabolism. Curr Opin Clin Nutr Metab Care 9:84–88 [DOI] [PubMed] [Google Scholar]
  32. Guequen Y, Ferrari L, Souidi M, Batt AM, Lutton C, Siest G, Visvikis S 2007 Compared effect of immunosuppressive drugs cyclosporin A and rapamycin on cholesterol homeostasis key enzyme CYP27A1 and HMG-CoA reductase. Basic Clin Pharmacol Toxicol 100:392–397 [DOI] [PubMed] [Google Scholar]
  33. Su YQ, Sugiura K, Wigglesworth K, O'Brien MJ, Affourtit JP, Pangas SA, Matzuk MM, Eppig JJ 2008 Oocyte regulation of metabolic cooperativity between mouse cumulus cells and oocytes: BMP15 and GDF9 control cholesterol biosynthesis in cumulus cells. Development 135:111–121 [DOI] [PubMed] [Google Scholar]
  34. Fitzpatrick SL, Richards JS 1991 Regulation of cytochrome P450 aromatase messenger ribonucleic acid and activity by steroids and gonadotropins in rat granulosa cells. Endocrinology 129:1452–1462 [DOI] [PubMed] [Google Scholar]
  35. Orly J 2000 Molecular events defining follicular developments and steroidogenesis in the ovary. Totowa, NJ: Humana Press, Inc. [Google Scholar]
  36. Goldring NB, Durica JM, Lifka J, Hedin L, Ratoosh SL, Miller WL, Orly J, Richards JS 1987 Hormonal regulation of cholesterol side chain cleavage P-450 (P-450scc) mRNA in rat ovarian follicles and corpora lutea. Endocrinology 120:1942–1950 [DOI] [PubMed] [Google Scholar]
  37. Orly J 1989 Orchestrated expression of steroidogenic side-chain cleavage cytochrome P450 during follicular development in the rat ovary. J Reprod Fertil 27:155–162 [PubMed] [Google Scholar]
  38. Ikeda Y, Luo X, Abbud R, Nilson JH, Parker KL 1995 The nuclear receptor steroidogenic factor-1 is essential for the formation of the ventromedial hypothalamic nucleus. Mol Endocrinol 9:478–486 [DOI] [PubMed] [Google Scholar]
  39. Tremblay JJ, Viger Rs 2001 GATA factors differentially activate multiple gonadal promoters through conserved GATA regulatory elements. Endocrinology 142:977–986 [DOI] [PubMed] [Google Scholar]
  40. Duggavathi R, Volle DH, Mataki C, Antal MC, Messeddeq N, Auwerx J, Murphy BD, Schoonjans K 2008 Liver receptor homolog 1 is essential for ovulation. Genes Dev 22:1871–1876 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Tremblay JJ, Viger RS 2001 Nuclear receptor DAX-1 represses the transcriptional cooperation between GATA-4 and SF-1 in Sertoli cells. Biol Reprod 64:1191–1199 [DOI] [PubMed] [Google Scholar]
  42. Volle DH, Duggavathi R, Magnier BC, Houten SM, Cummins CL, Lobaccaro JM, Verhoeven G, Schoonians K, Auwerx J 2007 The small heterdimer partner is a gonadal gatekeeper of sexual maturation in male mice. Genes Dev 21:303–315 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Richards JS 1994 Hormonal control of gene expression in the ovary. Endocr Rev 15:725–751 [DOI] [PubMed] [Google Scholar]
  44. Nowak K, Killmer K, Gessner C, Lutz W 2007 E2F-1 regulates expression of FOXO1 and FOXO2. Biochim Biophys Acta 1769:244–252 [DOI] [PubMed] [Google Scholar]
  45. Fan HY, Liu Z, Cahill N, Richards JS 2008 Targeted disruption of Pten in ovarian granulosa cells enhances ovulation and extends the life span of luteal cells. Mol Endocrinol 22:2128–2140 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Cunningham MA, Zhu Q, Hammond JM 2004 FoxO1a can alter cell cycle progression by regulating the nuclear localization of p27kip in granulosa cells. Mol Endocrinol 18:1756–1767 [DOI] [PubMed] [Google Scholar]
  47. Accili D, Arden KC 2004 FoxOs at the crossroads of cellular metabolism, differentiation and transformation. Cell 117:421–426 [DOI] [PubMed] [Google Scholar]
  48. Puigserver P, Rhee J, Donovan J, Walkey CJ, Yoon JC, Oriente F, Kitamura Y, Altomonte J, Dong H, Accili D, Spiegelman BM 2003 Insulin-regulated hepatic gluconeogenesis through FOXO1-PGC1 α interactions. Nature 423:550–555 [DOI] [PubMed] [Google Scholar]
  49. Motola DL, Cummins CL, Rottiers V, Sharma KK, Li T, Li Y, Suino-Powell K, Xu HE, Auchus RJ, Antebi A, Mangelsdorf DJ 2006 Identification of ligands for DAF12 that govern dauer formation and reproduction in C. elegans. Cell 124:1209–1223 [DOI] [PubMed] [Google Scholar]
  50. Rottiers V, Antebi A 2006 Control of Caenorhabditis elegans life history by nuclear receptor signal transduction. Exp Gerontol 41:904–909 [DOI] [PubMed] [Google Scholar]
  51. Zeleznik AJ, Saxena D, Little-Ihrig L 2003 Protein kinase B is obligatory for follicle-stimulating hormone-induced granulosa cell differentiation. Endocrinology 144:3985–3994 [DOI] [PubMed] [Google Scholar]
  52. Gonzalez-Robayna IJ, Falender AE, Ochsner S, Firestone GL, Richards JS 2000 FSH stimulates phosphorylation and activation of protein kinase B (PKB/Akt) and serum and glucocorticoid-induced kinase (Sgk): evidence for A-kinase independent signaling in granulosa cells. Mol Endocrinol 14:1283–1300 [DOI] [PubMed] [Google Scholar]
  53. Ning G, Ouyang H, Wang S, Chen X, Xu B, Yang J, Zhang H, Zhang M, Xia G 2008 CREB up-regulated CYP51 expression involved in FSH-induced mouse oocyte maturation. Mol Endocrinol 22:1682–1694 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Yamashita Y, Nishibori M, Terada T, Isobe N, Shimada M 2005 Gonadtropin-induced δ 14-reductase and δ7 reductase gene expression in cumulus cells during meiotic resumption of porcine oocytes. Endocrinology 146:186–194 [DOI] [PubMed] [Google Scholar]
  55. Moumné L, Batista F, Benayoun BA, Nallathambi J, Fellous M, Sundaresan P, Veitia RA 2008 The mutations and potential targets of the forkhead transcription factor FOXL2. Mol Cell Endocrinol 282:2–11 [DOI] [PubMed] [Google Scholar]
  56. Uda M, Ottolenghi C, Crisponi L, Garcia JE, Deiana M, Kimber W, Forabosco A, Cao A, Schlessinger D, Pilia G 2004 Foxl2 disruption causes mouse ovarian failure by pervasive blockage of follicle development. Hum Mol Genet 13:1171–1181 [DOI] [PubMed] [Google Scholar]
  57. Ottolenghi C, Pelosi E, Tran J, Colombino M, Douglass E, Nedorezov T, Cao A, Forabosco A, Schelssinger D 2007 Loss of Wnt4 and Foxl2 lead to female-to-male sex reversal extending to germ cells. Hum Mol Genet 16:2795–2804 [DOI] [PubMed] [Google Scholar]
  58. Kamei Y, Miura S, Suganami T, Akaike F, Kanai S, Sugita S, Katsumata A, Aburatani H, Unterman TG, Ezaki O, Ogawa Y 2008 Regulation of SREBP1c gene expression in skeletal muscle: role of retinoid X receptor/liver X receptor and forkhead-O1 transcription factor. Endocrinology 149:2293–2305 [DOI] [PubMed] [Google Scholar]
  59. Wang Y, Rogers PM, Su C, Varga G, Stayrook KR, Burris TP 2008 Regulation of cholesterologenesis by oxysterol receptor, LXRα. J Biol Chem 283:26332–26339 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Villagra A, Ulloa N, Zhang X, Yuan Z, Sotomayor E, Seto E 2007 Histone deacetylase 3 down-regulates cholesterol synthesis through repression of lanosterol synthase gene expression. J Biol Chem 282:35457–35470 [DOI] [PubMed] [Google Scholar]
  61. Doetzlhofer A, Rotheneder H, Lagger G, Koranda M, Kurten V, Brosch G, Wintersberger E, Seiser G 1999 Histone deacetylase 1 can repress transcription by binding to Sp1. Mol Cell Biol 19:5504–5511 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Cottom J, Salvador LM, Maizels ET, Reierstad S, Park Y, Carr DW, Davare MA, Hell JW, Palmer SS, Dent P, Kawakatsu H, Ogata M, Hunzicker-Dunn M 2003 Follicle-stimulating hormone activates extracellular signal-regulated kinase but not extracellular signal-regulated kinase kinase through a 100-kDa phosphotyrosine phosphatase. J Biol Chem 278:7167–7179 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Itoh N, Ornitz DM 2008 Functional evolutionary history of the mouse Fgf gene family. Dev Dynam 237:18–27 [DOI] [PubMed] [Google Scholar]
  64. Drummond AE, Tellbach M, Dyson M, Findlay JK 2007 Fibroblast growth factor-9, a local regulator of ovarian function. Endocrinology 148:3711–3721 [DOI] [PubMed] [Google Scholar]
  65. DiNapoli L, Batchvarov J, Capel B 2006 FGF9 promotes survival of germ cells in the fetal testis. Development 133:1519–1527 [DOI] [PubMed] [Google Scholar]
  66. Fan HY, Shimada M, Liu Z, Cahill N, Noma N, Wu Y, Gossen J, Richards JS 2008 Selective expression of KrasG12D in granulosa cells of the mouse ovary causes defects in follicular development and ovulation. Development 135:2127–2137 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Molecular Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES