Follicle-stimulating hormone/cAMP regulation of aromatase gene expression requires beta-catenin

Abstract

Estrogens profoundly influence the physiology and pathology of reproductive and other tissues. Consequently, emphasis has been placed on delineating the mechanisms underlying regulation of estrogen levels. Circulating levels of estradiol in women are controlled by follicle-stimulating hormone (FSH), which regulates transcription of the aromatase gene (CYP19A1) in ovarian granulosa cells. Previous studies have focused on two downstream effectors of the FSH signal, cAMP and the orphan nuclear receptor steroidogenic factor-1 (NR5A1). In this report, we present evidence for beta-catenin (CTNNB1) as an essential transcriptional regulator of CYP19A1. FSH induction of select steroidogenic enzyme mRNAs, including Cyp19a1, is enhanced by beta-catenin. Additionally, beta-catenin is present in transcription complexes assembled on the endogenous gonad-specific CYP19A1 promoter, as evidenced by chromatin immunoprecipitation assays. Transient expression and RNAi studies demonstrate that FSH- and cAMP-dependent regulation of this promoter is sensitive to alterations in the level of beta-catenin. The stimulatory effect of beta-catenin is mediated through functional interactions with steroidogenic factor-1 that involve four acidic residues within its ligand-binding domain, mutation of which attenuates FSH/cAMP-induced Cyp19a1 mRNA accumulation. Together, these data demonstrate that beta-catenin is essential for FSH/cAMP-regulated gene expression in the ovary, identifying a central and previously unappreciated role for beta-catenin in estrogen biosynthesis, and a potential broader role in other aspects of follicular maturation.

Fig. 1.

β-catenin enhances FSH induction of endogenous Cyp19a1 and Cyp11a1 mRNAs. Primary GCs were transduced with Δ90 β-catenin or EGFP adenoviruses [3 × 10¹⁰ virus particles per milliliter (vpm)]. Samples were treated with FSH (100 ng/ml, 24 h) where indicated. (A) Representative RT-PCR analysis of Cyp19a1, Cyp11a1, Star, and Gapdh mRNAs. (B) Fold change was calculated by comparison with untreated EGFP controls (mean ± SEM, n = 4; ∗, P < 0.05). (C) Western blot analysis confirmed endogenous and Δ90 β-catenin (◀) expression (upper and lower bands, respectively); AKT was the loading control.

Fig. 2.

Residues 235–238 in the SF1 ligand-binding domain mediate a functional interaction with β-catenin. (A) SF1 schematic: DBD, DNA-binding domain; h, hinge region; LBD, ligand-binding domain (including H1, helix 1); P, Ser-203 phosphorylation site; residues 235–238, β-catenin interaction domain; pAF, proximal activation domain; and AF2, activation function 2 domain. (B) Primary GCs were transduced with adenoviruses encoding EGFP (3 × 10¹⁰ vpm), SF1 (1 × 10¹¹ vpm), or SF1(235-4A) (1 × 10¹⁰ vpm) alone, or in combination with Δ90 β-catenin (3 × 10¹⁰ vpm). RNA and protein were isolated after 48 h, and gene/protein expression was evaluated as described in Fig. 1. Cyp19a1 and Gapdh, RT-PCR analysis; WB, Western blots. (Bar Graph) Fold change was determined by comparison with untreated EGFP controls (mean ± SEM; n = 3; ∗, P < 0.05).

Fig. 3.

Cooperative induction of Cyp19a1 mRNA by FSH and SF1 requires an intact β-catenin-interaction domain. Primary GCs were transduced with EGFP, SF1, or SF1(235-4A) adenoviruses as described in Fig. 2. FSH treatment and analyses of mRNA/protein expression were described in Fig. 1. (Bar Graph) Fold change was determined by comparison with untreated EGFP controls (mean ± SEM; n = 3; ∗, P < 0.05).

Fig. 4.

β-Catenin mediates FSH/cAMP induction of CYP19A1 PII. (A) Primary GCs were transiently cotransfected with CYP19A1 (100 ng) and phRG-B Renilla luciferase reporters (10 ng) and either nonspecific (control) or Ctnnb1-specific siRNA (50 nM). Western blot analysis confirmed knockdown of endogenous β-catenin protein by Ctnnb1 siRNA as compared with mock or control siRNA-transfected GCs. AKT was the loading control. GCs were treated with 10 nM testosterone plus vehicle, 10 μM forskolin, or 100 ng/ml FSH for 24 h before luciferase assays. Bar graphs represent normalized luciferase activity, firefly/Renilla (FF/Ren); mean ± SEM; ∗, P < 0.05. (B–D) KGN cells were transiently cotransfected with CYP19A1 (200 ng) and phRG-TK Renilla reporters (20 ng), and expression vectors or siRNA, and treated with vehicle or 10 μM forskolin (24 h) before luciferase assays. (B) Nonspecific (control) or CTNNB1-specific siRNA (100 nM) cotransfected into KGN cells (n = 4). (C) Axin1 or empty vector (100 ng) cotransfected into KGN cells (n = 5). (D) Empty vector or Δ90 β-catenin (100 ng) cotransfected (n = 3).

Fig. 5.

SF1 requires β-catenin for basal and cAMP regulation of CYP19A1 PII. (A and B) Transient transfection of KGN cells with CYP19A1 and Renilla reporters was as described for Fig. 4. (A) Sf1 (10 ng) was cotransfected in combination with empty or Δ90 β-catenin vectors (100 ng). Cells were harvested 48 h after transfection for luciferase assays. Data represent fold change compared with empty vector (mean ± SEM, n = 3). (B) Empty or Sf1 vector (10 ng) was cotransfected with control or CTNNB1-specific siRNA (100 nM) (duplicate transfections). Cells were treated with vehicle or 10 μM forskolin (24 h) before luciferase assays. Data represent fold induction relative to forskolin-treated empty vector control cotransfected with control siRNA (mean ± SEM, n = 3; ∗, P < 0.05).

Fig. 6.

β-Catenin associates with the endogenous CYP19A1 PII. KGN cells were treated with vehicle or 10 μM forskolin (1 h). ChIP assays were performed with Gα_q/11, SF1, or β-catenin antibodies (n = 3). DNA was amplified by using primers against CYP19A1 PII (proximal) and an upstream (distal) region (schematic). CLS, cAMP-response-element-like sequence. Representative PCR using proximal primers is shown. Vehicle data represent three independent experiments. Forskolin results are representative of two of three experiments.