Ovarian expression and regulation of the stromelysins during the periovulatory period in the human and the rat

Abstract

The matrix metalloproteinases (MMPs) are postulated to facilitate follicular rupture. In the present study, expression of the stromelysins (MMP3, MMP10, MMP11) was analyzed in the periovulatory human and rat ovary. Human granulosa and theca cells were collected from the dominant follicle at various times after human chorionic gonadotropin (hCG). Intact rat ovaries, granulosa cells, and residual tissue (tissue remaining after granulosa cell collection) were isolated from equine CG (eCG)-hCG-primed animals. Mmp10 mRNA was highly induced in human granulosa and theca cells and intact rat ovaries, granulosa cells, and residual tissue. Localization of MMP10 to granulosa and theca cells in both human and rat ovarian follicles was confirmed by immunohistochemistry. Mmp3 mRNA was unchanged in human cells and rat granulosa cells, but increased in intact rat ovaries and residual tissue. Mmp11 mRNA decreased following hCG treatment in human granulosa and theca cells as well as rat granulosa cells. Regulation of Mmp10 in cultured rat granulosa cells revealed that the EGF inhibitor AG1478 and the progesterone receptor antagonist RU486 suppressed the induction of Mmp10 mRNA, whereas the prostaglandin inhibitor NS398 had no effect. Studies on the Mmp10 promoter demonstrated that forskolin plus PMA stimulated promoter activity, which was dependent upon a proximal AP1 site. In conclusion, there are divergent patterns of stromelysin expression associated with ovulation, with a marked induction of Mmp10 mRNA and a decrease in Mmp11 mRNA, yet a species-dependent pattern on Mmp3 mRNA expression. The induction of Mmp10 expression suggests an important role for this MMP in the follicular changes associated with ovulation and subsequent luteinization.

FIG. 1.

A–C) Real-time PCR analysis of the mRNA profiles for the stromelysin class of MMPs in intact rat ovaries, granulosa cells collected in vivo, and residual tissue collected in vivo. Relative levels of mRNA for each stromelysin member were normalized to Rpl32 for each sample. Values represent the mean ± SEM (n = 5 animals/time point for intact ovary, n = 3–5 animals/time point for granulosa cells in vivo, and n = 4 animals/time point for residual tissue in vivo). Bars with no common superscripts (number, intact ovary; lowercase letter, granulosa cells; uppercase letter, residual tissue) among the respective MMP member in each panel are significantly different (P < 0.05).

FIG. 2.

Analysis of MMP10 (A) and MMP11 (B) mRNA in human granulosa and theca cells isolated from the dominant follicle in women undergoing elective surgery throughout the periovulatory period after hCG treatment. Relative levels of MMP10 and MMP11 mRNA were normalized to GAPDH for each sample. Values represent the mean ± SEM (n = 5 patients/time point). Bars with no common superscripts in each panel are significantly different (P < 0.05).

FIG. 3.

Real-time PCR analysis of stromelysin mRNA expression in rat granulosa cells cultured in vitro for Mmp3 (A), Mmp10 (B), and Mmp11 (C). Cells were cultured in the absence (Control) or presence of hCG for varying lengths of time. Relative levels of mRNA for each stromelysin member were normalized to Rpl32 for each sample. Values represent the mean ± SEM (n = 3–4 animals/time point). Bars with no common superscripts in each panel are significantly different (P < 0.05).

FIG. 4.

Real-time PCR analysis of stromelysin mRNA expression in rat theca-interstitial cells cultured in vitro for Mmp3 (A), Mmp10 (B), and Mmp11 (C). Cells were collected from eCG-primed immature rats and cultured in the absence (Control) or presence of FSK + PMA for varying lengths of time. Relative levels of mRNA for each stromelysin member were normalized to Rpl32 for each sample. Values represent the mean ± SEM (n = 3–4 animals/time point). Bars with no common superscripts in each panel are significantly different (P < 0.05).

FIG. 5.

Hormonal regulation of the hCG-induced Mmp10 mRNA expression in rat granulosa cells following 6 h of culture. Cells were pretreated in media alone (Control), with the progesterone receptor antagonist RU486 (1 μM), the prostaglandin-endoperoxide synthase 2 inhibitor NS-398 (1 μM), or the EGFR tyrosine kinase selective inhibitor AG1478 (1 μM) for 30 min. Cells were then cultured in the absence (Control) or presence of hCG (1 IU) plus each of the inhibitors of various hormonal signaling pathways for 6 h. Relative levels of mRNA for Mmp10 were normalized to the Rpl32 gene in each sample following real-time PCR and expressed as a fold change relative to the untreated 0-h control (mean ± SEM; n = 3–6 independent culture experiments). Values with no common superscripts in each panel are significantly different (P < 0.05).

FIG. 6.

Regulation of the Mmp10 promoter by FSK + PMA in rat granulosa cells. Granulosa cells were isolated from gonadotropin-primed immature rats (48 h post-eCG), transfected with various rat Mmp10 promoter constructs for 24 h, and subsequently cultured with FSK + PMA for 6 h. Relative levels of the promoter activity where normalized to the luciferase:renilla ratio, and data are presented as the relative luciferase activity. Bars with no common superscripts within a promoter construct are significantly different from their respective control (P < 0.05). Inset: data are presented as the fold change in activity relative to the corresponding control. Bars with no common superscripts are significantly different (P < 0.05). Values represent the mean ± SEM (n = 3–4).

FIG. 7.

Protein localization of MMP10 by immunohistochemical staining in human preovulatory (A), early ovulatory (B), and late ovulatory (C) dominant follicles, and rat ovary (E) at 8 h after hCG stimulation. Representative images of human follicles (A–D) and rat ovaries (E and F) were collected at defined times after hCG stimulation and processed for immunohistochemistry. MMP10 reaction product was visualized with diaminobenzidine producing a brown precipitate. Sections were then counterstained with hematoxylin. Negative control sections, with no primary antibody, are illustrated for a human follicle collected at the late ovulatory period (D) and a rat ovary collected at 8 h after hCG (F). GC, granulosa cells; TA, tunica albuginea; TC, theca cells. Original magnification ×20 (A–C, F), ×30 (D), ×15 (E).