Signaling by hypoxia-inducible factors is critical for ovulation in mice

Abstract

The steroid hormone progesterone, acting via its nuclear receptor, is a major regulator of the process of ovulation. Female mice lacking progesterone receptor (PGR) exhibit an anovulatory phenotype due to failure in follicular rupture. To identify the PGR-regulated pathways that control ovulation, we analyzed global changes in gene expression in the ovaries of wild-type and Pgr-null mice subjected to gonadotropin-induced superovulation. Our analysis uncovered several genes whose expression was reduced in the Pgr-null ovaries compared with the wild-type ovaries immediately preceding ovulation. Interestingly, these genes included three hypoxia-inducible factors (HIFs): HIF-1 alpha, HIF-2 alpha, and HIF-1 beta. These transcription factors form alphabeta-heterodimers, which regulate the transcription of specific cellular genes, thereby mediating adaptive response of the tissue to low-oxygen levels. We observed that the expression of mRNAs and proteins corresponding to HIF-1 alpha, HIF-2 alpha, and HIF-1 beta was induced in a PGR-dependent manner, specifically in the granulosa cells of the preovulatory follicles. Inhibition of the HIF transcriptional activity by echinomycin, a small-molecule inhibitor that suppresses the binding of HIF alphabeta-heterodimers to target genes, blocked ovulation by preventing the rupture of the preovulatory follicles. Echinomycin specifically inhibited the expression of genes that are known regulators of ovulation, such as a disintegrin and metalloproteinase with thrombospondin-like motifs-1 and endothelin-2. Furthermore, echinomycin reduced the expression of vascular endothelial growth factor A, a key factor controlling vascularization/angiogenesis during ovulation. Collectively, these findings unveiled a novel ovarian role for the HIF transcription factors during the ovulatory period in mice.

Figure 1

PGR regulation of HIF expression in the ovary during gonadotropin-induced superovulation. WT (n = 8) and Pgr-null (PR-null) (n = 8) mice were primed with PMSG for 48 h and then treated with hCG to induce ovulation. The ovaries were collected from these mice at 0 and 11 h after hCG administration. Ovarian tissues (from n = 4) were pooled and used to perform gene expression studies. A–C, Total RNA was extracted from these tissues, and mRNA levels of Hif1a, Epas1 (HIF-2α), and Hif1b were quantified by real-time PCR using gene-specific primers. Multiple sets of the same experiment were performed, and we used one-way ANOVA followed by Tukey’s test to analyze the difference in gene expression between different sets. Different letters indicate a statistically significant difference (P < 0.05). Ovaries (from WT and Pgr-null mice, n = 4) were used to monitor protein expression. D, Whole cell extracts were prepared from these tissues, and the protein levels of HIF-1α and HIF-2α were examined by Western blotting using specific antibodies against these factors. Protein levels of calnexin, a calcium-binding protein, were also assessed to ensure equal loading of the whole cell extracts.

Figure 2

Granulosa cell-specific induction of HIFs in the mouse ovary during ovulation. A and B, Immunohistochemical localization of HIF-1α and HIF-2α proteins. Ovaries were collected from WT and Pgr-null (PR-null) mice after sequential stimulation by PMSG and hCG (0 or 11 h after hCG). The tissues were fixed in 10% formalin, embedded in paraffin, and sectioned at 5 μm. These ovarian sections were then analyzed by immunohistochemistry as described in Materials and Methods to localize HIF-1α (A) and HIF-2α (B) proteins using rabbit polyclonal antibodies specific for each HIF-α isoform. C, Induction of Hif1a, Epas1 (HIF-2α), and Hif1b mRNAs in mouse primary granulosa cells. Pgr (+/−) and Pgr (−/−) mice were treated with PMSG for 48 h, followed by hCG treatment for 0 or 11 h. Ovaries were pooled from mice (n = 4) in each treatment group. The granulosa cells were extracted from the preovulatory follicles by needle puncture. Total RNA was isolated from these cells, and the mRNA expression levels of Hif1a, Epas1 (HIF-2α), and Hif1b were quantified by real-time PCR analysis as described in Materials and Methods. The data are representative of two independent experiments. The error bars denote the sd of three to five replicates of each real-time PCR analysis.

Figure 3

Echinomycin inhibits ovulation in mice. A and B, Effects of echinomycin on ovulation. After PMSG priming for 48 h, mice were treated with hCG to induce ovulation. A, In one set of experiments, mice (ranging from three to 11 for each treatment group) were injected with either vehicle (dimethylsulfoxide) or various doses of echinomycin (0.05, 0.5, 1, 5, or 10 mg/kg body weight) at 6 h after hCG. B, In another set, mice (n = 4 for each treatment group) were given either vehicle or a fixed dose of echinomycin (1 mg/kg body weight) at 3, 6, 9, 10, or 11 h after hCG, or at 24 h before hCG. The ovulated eggs were counted at 20 h after hCG, and their numbers were compared between the echinomycin-treated or untreated groups. The statistical significance of the difference between these comparisons was analyzed using one-way ANOVA and Tukey’s test. Different letters indicate a statistically significant difference (P < 0.05). C, Echinomycin-mediated inhibition of follicular rupture. Mice (n = 4 for each treatment group) were subjected to PMSG- and hCG-induced superovulation, and were treated with either vehicle (dimethylsulfoxide) or echinomycin (10 mg/kg body weight) at 6 h after hCG. The ovaries were collected at 20 h after hCG. The tissues were fixed in 10% formalin, embedded in paraffin, and sectioned. The ovarian sections were stained with hematoxylin and eosin, and visualized under a light microscope (×10). Note that the ovary from the vehicle-treated mouse harbors numerous corpora lutea (CL), whereas the one from the echinomycin-treated mouse shows many unruptured preovulatory follicles with “trapped” eggs inside (marked by arrows).

Figure 4

HIF regulation of known PGR-target genes during ovulation. WT mice were subjected to PMSG- and hCG-stimulated superovulation as described in Materials and Methods and treated with either vehicle (dimethylsulfoxide) or echinomycin (0.5, 1, or 5 mg/kg body weight) at 6 h after hCG (n = 4 for each treatment group). The ovaries were collected at 11 h after hCG, pooled, and total RNA was isolated from these tissues. A–F, Real-time PCR was performed using gene-specific primers for Adamts1, Edn2, Vegfa, Cxcr4, Hif1a, and Adam8. A–C and E, Multiple sets of the same experiment were performed, and we used one-way ANOVA followed by Tukey’s test to analyze the difference in gene expression between different sets. Different letters indicate a statistically significant difference (P < 0.05). D and F, Representative data from two independent experiments are shown. The error bars denote the sd of three to five replicates of each real-time PCR analysis.

Figure 5

Induction of HIF-target genes by chemically induced hypoxia in mouse primary granulosa cells. WT mice (n = 20 for whole cell extracts; n = 9 for total RNA) were treated with PMSG for 48 h, and the granulosa cells were extracted from the preovulatory follicles as described in Materials and Methods. These cells were treated with forskolin and PMA to mimic hCG treatment and activate PGR signaling in the presence or absence of CoCl₂, a hypoxia mimetic. Twelve hours after treatment with forskolin and PMA, the cells were harvested. A, Whole cell extracts were prepared from a portion of these cells, and the protein levels of HIF-1α, HIF-2α, HIF-1β, and calnexin were examined by Western blotting using antibodies specific for each protein. B and C, Total RNA was isolated from the granulosa cells, and the mRNA expression levels of Hif1a, Hif1b, Edn2, Vegfa, and Cxcr4 were quantified by real-time PCR analysis. For Hif1a, Edn2, and Vegfa, multiple sets of the same experiment were performed, and we used one-way ANOVA followed by Tukey’s test to analyze the difference in gene expression between different sets. Different letters indicate a statistically significant difference (P < 0.05). For Hif1b and Cxcr4, representative data from two independent experiments are shown. The error bars denote the sd of three to five replicates of each real-time PCR analysis.

Figure 6

Echinomycin regulation of hypoxia-inducible ovulatory genes. Primary granulosa cells were extracted from the ovaries of WT mice (n = 15) treated with PMSG for 48 h, and cultured in the absence or presence of forskolin and PMA as described in Materials and Methods. CoCl₂ (150 or 250 μm), and echinomycin (1, 2, or 3 nm equivalent to 1.1, 2.2, and 3.3 ng/ml medium, respectively) or vehicle (dimethylsulfoxide) were added to these cultures where indicated. After 12 h, total RNA was isolated from the granulosa cells and analyzed by real-time PCR to assess the mRNA levels of Epas1 (HIF-2α) (A), Adam8 (B), Edn2 (C), Vegfa (D), and Cxcr4 (E). Multiple sets of the same experiment were performed, and we used one-way ANOVA followed by Tukey’s test to analyze the difference in gene expression between different sets. Different letters indicate a statistically significant difference (P < 0.05).