Transactivation of the epidermal growth factor receptor is involved in the lutropin receptor-mediated down-regulation of ovarian aromatase expression in vivo

Abstract

Ovarian follicular development and differentiation is characterized by dramatic changes in aromatase (Cyp19a1) expression. In preovulatory follicles, activation of the FSH receptor increases aromatase expression until the surge of LH decreases it. Here we provide in vivo evidence that down-regulation of Cyp19a1 by the LH surge requires efficient signaling through the epidermal growth factor receptor (EGFR). The human chorionic gonadotropin (hCG)-induced down-regulation of Cyp19a1 expression in the two different mouse models with inactivating mutations of the EGFR (wa2 and velvet) is impaired but not abolished. The hCG-induced phosphorylation of ovarian ERK1/2, expression of C/EBPbeta, and the phosphorylation of Connexin43 (two downstream targets of ERK1/2 action) are also decreased in these two mouse models. In contrast, disruption of EGFR signaling does not have any affect on the hCG-induced phosphorylation of cAMP response element-binding protein or AKT. This study provides the first in vivo evidence linking the LH receptor, the EGFR, and ERK1/2 as sequential components of a pathway that regulates ovarian Cyp19a1 expression.

Figure 1

PMSG-induced expression of the Lhcgr- and hCG-induced phosphorylation of the EGFR in Egfr^+/velvet mice. A, Real-time PCR shows expression of the Lhcgr in nonstimulated ovaries or ovaries from mice that were injected with PMSG 44 h before collection. Each bar represents the mean ± sem (n = 6 mice) of the relative expression for Lhcgr mRNA normalized to Gapdh mRNA. B, Western blot shows phosphorylation of the EGFR on Tyr¹⁰⁶⁸ in ovarian lysates of PMSG-primed mice before or 2 h after injection of hCG. A representative blot is shown on the top. The bar graph shows the level for EGFR phosphorylation after normalization for a loading control (AKT in this case). Each bar represents mean ± sem from six mice. *, Significant differences (P < 0.05) between wt and Egfr^+/velvet mice.

Figure 2

The hCG-induced phosphorylation of CREB and AKT is not impaired in Egfr^+/velvet or Egfr^wa2/wa2 mice. The representative Western blots show the phosphorylation of CREB (A and B) or AKT (C and D) in ovarian lysates of Egfr^+/velvet, Egfr^wa2/wa2, and their wt controls. All mice were injected with PMSG and analyzed either 44 h later or 2 h after an injection of hCG as indicated. The bar graphs show the relative levels of the indicated phosphoproteins after normalization for the loading controls as shown. Each bar represents the mean ± sem from six mice.

Figure 3

The hCG-induced phosphorylation of ERK1/2 is reduced in Egfr^+/velvet or Egfr^wa2/wa2 mice. A and B, The representative Western blots show the phosphorylation of ERK1/2 in ovarian lysates of Egfr^+/velvet, Egfr^wa2/wa2, and their wt controls. All mice were injected with PMSG and analyzed either 44 h later or 2 h after an injection of hCG as indicated. The bar graphs show the relative levels for pERK1/2 after normalization for a loading control (GAPDH). Each bar represents the mean ± sem from six mice. *, Significant differences (P < 0.05) between wt and EGFR mutant mice. C, Immunohistochemical staining for pERK1/2 (blue) in the ovaries of a wt animal shows that the hCG-induced phosphorylation of ERK1/2 occurs mostly in granulosa cells.

Figure 4

The induction of C/EBPβ expression and the phosphorylation of Cx43 are reduced in Egfr^+/velvet or Egfr^wa2/wa2 mice. The representative Western blots show the expression of C/EBPβ (A and B) or the phosphorylation of Cx43 (C and D) in ovarian lysates of Egfr^+/velvet, Egfr^wa2/wa2, and their wt controls. All mice were injected with PMSG and analyzed either 44 h later or 2 h after an injection of hCG as indicated. The bar graphs show the relative levels of the C/EBPβ or pCx43 after normalization for the loading controls as shown. Each bar represents the mean ± sem from six mice. *, Significant differences (P < 0.05) between wt and mutant mice.

Figure 5

The hCG-dependent down-regulation of ovarian Cyp19a1 is compromised in Egfr^+/velvet (A) or Egfr^wa2/wa2 (B) mice. Real-time PCR shows the expression of Cyp19a1 in the ovaries of Egfr^+/velvet, Egfr^wa2/wa2, and their wt controls. Cyp19a1 expression was measured in nonstimulated ovaries 44 h after an injection of PMSG and 24 h after an injection of hCG to the PMSG-primed mice. Each bar represents the mean ± sem of the ratio of Cyp19a1 to Gapdh from six mice. *, Significant differences (P < 0.05) between wt and EGFR mutant mice that received the same hormonal treatment.

Figure 6

The gonadotropin-dependent regulation of the expression of ovarian Cyp11a1 and the Fshr are compromised in Egfr^+/velvet or Egfr^wa2/wa2 mice. Real-time PCR shows the expression of the Fshr (A) and Cyp11a1 (B) Cyp11a1 in the ovaries of Egfr^+/velvet, Egfr^wa2/wa2, and their wt controls. Gene expression was measured in nonstimulated ovaries 44 h after an injection of PMSG and 24 h after an injection of hCG to the PMSG-primed mice. Each bar represents the mean ± sem of the ratio of Fshr to Gapdh (A) or Cyp11a1 to Gapdh (B) from six mice. *, Significant differences (P < 0.05) between wt and EGFR mutant mice that received the same hormonal treatment.