Abstract
In addition to its role in blood coagulation, thrombin directly stimulates protease-activated receptors (PAR) or interacts with thrombomodulin (THBD) to activate membrane-bound protein C which stimulates PAR1 and PAR4 receptors to promote downstream pleiotropic effects. Our DNA microarray, RT-PCR, and immunostaining analyses demonstrated ovarian expression of THBD, activated protein C (APC) receptor [endothelial protein C receptor (EPCR)], as well as PAR1 and PAR4 receptors in mice. After treatment of gonadotropin-primed immature mice with an ovulatory dose of human chorionic gonadotropin (hCG) (a LH surrogate), major increases in the expression of THBD, EPCR, PAR1, and PAR4 were detected in granulosa and cumulus cells of preovulatory follicles. Immunoassay analyses demonstrated sustained increases in ovarian prothrombin and APC levels after hCG stimulation. We obtained luteinizing granulosa cells from mice treated sequentially with equine CG and hCG. Treatment of these cells with thrombin or agonists for PAR1 or PAR4 decreased basal and forskolin-induced cAMP biosynthesis and suppressed hCG-stimulated progesterone production. In cultured preovulatory follicles, treatment with hirudin (a thrombin antagonist) and SCH79797 (a PAR1 antagonist) augmented hCG-stimulated progesterone biosynthesis, suggesting a suppressive role of endogenous thrombin in steroidogenesis. Furthermore, intrabursal injection with hirudin or SCH79797 led to ipsilateral increases in ovarian progesterone content. Our findings demonstrated increased ovarian expression of key components of the thrombin-APC-PAR1/4 signaling system after LH/hCG stimulation, and this signaling pathway may allow optimal luteinization of preovulatory follicles. In addition to assessing the role of thrombin and associated genes in progesterone production by the periovulatory ovary, these findings provide a model with which to study molecular mechanisms underlying thrombin-APC-PAR1/4 signaling.
Thrombin is an enzyme central in a cascade of proteolytic cleavages of the coagulation process, leading to the conversion of soluble fibrinogen to form fibrin clots. In addition, thrombin has been shown to regulate the proliferation and activation of diverse cell types mediated by several protease-activated receptors (PAR) and downstream intracellular signaling molecules (1). In addition to its direct activation of PAR, thrombin also interacts with another high-affinity receptor thrombomodulin (THBD) to promote the activation of membrane-bound protein C in the regulation of specific cellular functions. The thrombin-THBD complex activates protein C approximately 1000 times faster than thrombin alone. After thrombin-mediated cleavage of protein C, activated protein C (APC) binds to its receptor EPCR (endothelial protein C receptor) and proteolytically stimulates PAR1 and PAR4 receptors to exert intracellular functions (2, 3) mediated by the inhibitory G protein (Gi) pathway. The efficient thrombin-THBD-APC-PAR1/4 system is found in different tissues and mediates pleiotropic actions, including anticoagulant, antiinflammatory, cytoprotective, and antiapoptotic activities (4).
Although the ovulatory process is triggered by pituitary LH, diverse intraovarian paracrine systems are involved in the coordinated regulation of follicle rupture, oocyte maturation, and luteinization. LH or its surrogate hCG (human chorionic gonadotropin) have been shown to act on LH receptors in ovarian somatic cells to induce the secretion of paracrine or autocrine molecules, leading to the fine tuning of granulosa, cumulus, and oocyte functions. Treatment with LH/hCG stimulates the release of diverse local ligands (epidermal growth factor-like factors, IL-1, IGF-I, etc) to modulate the functions of follicular somatic cells and oocytes (5–7). Based on DNA microarray analyses of ovarian ligand-receptor pairs in the periovulatory ovaries, we further demonstrated the intraovarian paracrine roles of brain-derived neurotrophic factor, endothelin-1, and TNF-related weak inducer of apoptosis in the regulation of ovulatory processes (8–10). Analyzing this periovulatory DNA microarray, we identified major increases in the expression of key genes in the thrombin-THBD-APC-PAR1/4 signaling pathway. Using RT-PCR and immunostaining analyses, we investigated the LH/hCG stimulation and ovarian cell types expressing THBD, EPCR, PAR1, and PAR4 in periovulatory ovaries. We further showed increased levels of ovarian prothrombin and APC, and the ability of thrombin and agonists for PAR1 and PAR4 to decrease cAMP production and to suppress hCG-stimulated progesterone biosynthesis.
Results
Preovulatory hCG treatment induced ovarian expression of different receptor genes in the thrombin-APC-PAR1/4 signaling pathway and increased ovarian APC content
Using DNA microarray and real-time RT-PCR, we analyzed the expression of genes in the thrombin-APC-PAR1/4 pathway during the preovulatory period. Based on analyses using DNA microarray (line graphs), treatment with Pergonal stimulated ovarian transcript levels for THBD, EPCR, PAR1, and PAR4 (Fig. 1, A and D−F). Furthermore, treatment with an ovulatory dose of Pregnyl containing hCG led to major increases in the expression of these four receptor genes. Although transcripts for thrombin and protein C are also expressed in the ovary throughout the same period, no apparent regulation by gonadotropins was found (Fig. 1, B and C). We focused on the effect of preovulatory hCG treatment and confirmed the DNA microarray data using real-time RT-PCR (bar graphs), showing time-dependent increases in transcript levels for THBD, EPCR, PAR1 and PAR4, but not thrombin and protein C, after hCG treatment.
Increased expression of key receptor genes in the thrombin-THBD-APC-PAR1/4 signaling pathway in preovulatory ovaries. Immature female mice were treated with eCG or Pergonal to induce follicle growth, followed at 48 h later by a single ovulatory dose of hCG or Pregnyl to induce ovulation. A–F, Transcript levels for THBD, thrombin, protein C, EPCR, PAR1, and PAR4 based on DNA microarray (left y-axis) or real-time RT-PCR (right y-axis) analyses. Line graphs represent DNA microarray data depicting the expression intensity of each transcript based on integration of hybridization signals from multiple probe sets for individual genes, whereas bar graphs depict quantitative real-time RT-PCR results. Mean ± se (n = 5–9).
We further analyzed ovarian cell types expressing the four receptor genes using real-time RT-PCR. As shown in Fig. 2, transcripts for all four genes (THBD, EPCR, PAR1, and PAR4) were found in cumulus and granulosa cells but not in the oocyte. In addition, theca cells expressed transcripts for EPCR, PAR1 and PAR4, but showed negligible THBD levels. Furthermore, treatment with hCG stimulated the expression of these genes in a time-dependent manner at 2 and 4 h after treatment. Immunohistochemical analyses (Fig. 3) further demonstrated increased expression of antigens for THBD, EPCR, PAR1, and PAR4 in the cumulus and granulosa cells of preovulatory follicles at 4 h and 8 h after hCG treatment. Similar increases were also found for EPCR, PAR1, and PAR4 in theca cells. The apparent spotty expression of these genes in granulosa cells could reflect heterogeneous protein expression in subpopulations of granulosa cells. In contrast, granulosa cells from preantral and smaller follicles showed no signals. Due to the absence of mRNAs for these genes in the oocyte, apparent antigen signals found in oocytes could represent nonspecific staining.
RT-PCR analyses of THBD (A), EPCR (B), PAR1 (C), and PAR4 (D) transcript levels in different ovarian cell types. Oocytes (OC), cumulus cells (CC), granulosa cells (GC), and theca cells (TC) were obtained from preovulatory follicles before and at 2 and 4 h after hCG treatment. Mean ± se (n = 6). *, P < 0.05 vs. 0 h controls.
Immunohistochemical localization of THBD, EPCR, PAR1, and PAR4 in preovulatory ovaries. Ovaries were obtained at 48 h after eCG treatment and at 4 and 8 h after hCG treatment of eCG-primed mice. Positive signals were increased in cumulus and granulosa cells (arrows) after hCG treatment but only background signals were found in small antral follicles (arrowheads). Samples stained with nonimmune IgG (NEG) served as negative controls. Scale bar, 200 μm.
During the periovulatory period, ovarian prothrombin could be derived from the general circulation or due to local synthesis. We further measured ovarian prothrombin content using ELISA. As shown in Fig. 4A, treatment with hCG increased ovarian prothrombin content at 4 h after treatment, reaching a peak at 16 h later. The actions of thrombin are dramatically amplified through its activation of protein C (11). Because ovarian thrombin could be bound by the increased THBD expressed in granulosa and cumulus cells to promote the activation of protein C of ovarian or systematic origins, we further determined ovarian content of APC using an ELISA in ovarian tissue extracts. As shown in Fig. 4B, treatment with an ovulatory dose of hCG led to time-dependent increases in ovarian APC levels, reaching approximately 9-fold increases at 12 h after hCG treatment, followed by a gradual decline, indicating increases in ovarian APC after the preovulatory hCG stimulation. Furthermore, treatment with hCG increased the expression levels of THBD, EPCR and PAR4 proteins at 8 h after treatment (Fig. 4C).
Gonadotropin stimulation of levels of prothrombin and APC as well as THBD, EPCR, and PAR4 in preovulatory ovaries. Immature mice at 23 d of age were treated with eCG for 48 h, followed by an ovulatory dose of hCG. Ovaries were obtained from animals at d 23 of age (D23), at 48 h after eCG treatment (0 h), and at different times (4–24 h) after gonadotropin treatment. After protein extraction, prothrombin (A) and APC (B) levels were determined using ELISA and expressed as ng/ovary (mean ± se, n = 6–10). *, P < 0.05 vs. 0 h of hCG treatment; **, P < 0.01. C, Immature mice were injected with eCG for 48 h (0 h), followed by hCG injection for 8 h (8 h). Ovarian protein extracts were used for immunoblotting analyses. Densitometric analyses were performed to obtain ratios of receptor antigens to β-actin. From 0 h to 8 h after hCG treatment, there were 2.4 ± 0.4, 3.7 ± 0.8, and 3.3 ± 0.3-fold increases in THBD, EPCR, and PAR4 levels, respectively (n = 3).
Thrombin suppression of hCG-stimulated progesterone and cAMP production by cultured luteinizing granulosa cells
We hypothesized that ovarian APC increased after the hCG stimulation could participate in noncoagulative functions in the periovulatory ovary mediated by PAR1 and/or PAR4 receptors. We cultured luteinizing granulosa cells obtained from equine CG (eCG)-primed mice at 6 h after treatment with an ovulatory dose of hCG and measured progesterone production. As shown in Fig. 5, treatment with increasing doses of hCG led to dose-dependent increases in progesterone production whereas concomitant treatment with thrombin suppressed hCG actions in a dose-dependent manner with 0.3 or 1 μIU/ml causing significant suppression. In addition, treatment with the synthetic peptide SFLLR (a PAR1 agonist) or the peptide GYPGKF (a PAR4 agonist) also led to dose-dependent suppression of hCG-stimulated progesterone production at all doses of hCG used. Because progesterone biosynthesis in granulosa cells is mediated by the cAMP signaling pathway and thrombin receptors have been shown to inhibit cAMP production in other systems (12), we further investigated possible modulation of cAMP production by thrombin and PAR1/4 agonists in granulosa cells. We measured cAMP production by cultured granulosa cells with or without forskolin treatment to investigate potential activation of the Gi pathway. Incubation with thrombin and agonists for PAR1 and PAR4 receptors suppressed basal cAMP production (Fig. 6A). Furthermore, preincubation with these compounds for 30 min decreased forskolin-stimulated cAMP production (Fig. 6B). In contrast, treatment with thrombin or agonists for PAR1 and PAR4 did not affect granulosa cell apoptosis based on the measurement of caspase-3/7 activities (Supplemental Table 1 published on The Endocrine Society's Journals Online web site at http://mend.endojournals.org).
Thrombin suppression of hCG-stimulated progesterone production by cultured granulosa cells. Luteinizing granulosa cells were obtained from preovulatory follicles of ovaries from eCG-primed mice at 6 h after hCG treatment. For progesterone production, cells were cultured with increasing does of hCG in the presence or absence of thrombin or agonists for PAR1 or PAR4. Media progesterone content was measured by specific RIA at 48 h after culture (mean ± se, n = 3). *, P < 0.05 vs. control (C).
Thrombin suppression of cAMP production by cultured granulosa cells. Mice were injected with eCG for 48 h, followed by hCG injection for 6 h. Luteinizing granulosa cells were treated with thrombin or agonists for PAR1 and PAR4 for 15 min. To estimate Gi activity, cells were pretreated with thrombin or PAR agonists for 30 min before addition of forskolin for 15 min. A, Basal levels; B, forskolin-stimulated levels. Mean ± se (n = 6). *, P < 0.05 vs. control (C) (panel A), or forskolin (FSK) (panel B).
Blockage of endogenous thrombin actions increased progesterone production by cultured preovulatory follicles and after intrabursal injections in vivo
To demonstrate the importance of endogenous thrombin in the mediation of observed suppressive effects, we isolated preovulatory follicles from eCG-primed mice and cultured them with hCG in the presence or absence of hirudin, a peptide found in the salivary glands of medicinal leeches and the most potent natural inhibitor of thrombin (13). As shown in Fig. 7A, treatment with hCG led to major increases in progesterone production by cultured follicles at 18 h after incubation whereas cotreatment with increasing doses of hirudin further increased hCG-stimulated progesterone production in a dose-dependent manner, suggesting a suppressive effect of endogenous thrombin in these follicles. Similarly, cotreatment with SCH79797, a PAR1 antagonist (14), also led to dose-dependent increases in progesterone production induced by hCG (Fig. 7B).
Treatment with antagonists for thrombin or PAR1 augmented hCG-stimulated progesterone production in cultured preovulatory follicles and after intrabursal injections in vivo. A, Progesterone production by cultured preovulatory follicles. Preovulatory follicles from eCG-primed immature mice were cultured with hCG (0.3 μg/ml) in the presence or absence of different doses of hirudin or SCH79797 (a PAR1 antagonist) for 18 h. Media content of progesterone was measured by RIA and expressed as pg/follicle. Numbers in parentheses indicate the number of follicles analyzed. a, b, c, and d denote groups with significant differences (P < 0.05). B, Ovarian progesterone content after intrabursal injection of hirudin or the PAR1 antagonist. Immature mice primed with eCG were treated with hirudin (10 IU) (C) or PAR1 antagonist SCH79797 (1 μmol) (D) in one ovary after intrabursal injection whereas the contralateral ovary were injected with saline to serve as controls. At 48 h after injection, ovarian content of progesterone was determined. Numbers inside the parentheses represent number of animals used. *, P < 0.05 vs. saline controls.
To demonstrate the importance of the thrombin signaling system in vivo, we further performed intrabursal injection tests by administering hirudin or SCH79797 into the bursa of one ovary in eCG-primed mice. The counterlateral ovaries of the same animal served as negative controls. Ovarian content of progesterone was measured after tissue extraction and RIA. As shown in Fig. 7, C and D, treatment with hirudin and the PAR1 antagonist led to 68 and 123% increases in ovarian progesterone content, respectively. These findings confirmed the role of endogenous thrombin in the periovulatory ovaries in vivo.
Discussion
Binding of thrombin to THBD shields thrombin's procoagulant exosite I and promotes its anticoagulant and other activities by activation of protein C through the thrombin-THBD complex. This reaction is augmented by localization of protein C to cell membranes expressing EPCR (11). After thrombin activates protein C, the pleiotropic APC plays anticoagulative roles when it is in the soluble form but regulates cellular functions when it is bound to EPCR. Although one cannot rule out a role of ovarian APC and thrombin in the regulation of ovarian blood coagulation, the present study focused on the intracellular, noncoagulative actions of APC. APC bound to the cell surface performs its cytoprotective effects by acting on the effector substrate PAR1 or PAR4 (15). We demonstrated the regulation of granulosa cell progesterone and cAMP biosynthesis mediated by the intraovarian thrombin-APC-PAR1/4 signaling system (Fig. 8). The preovulatory LH/hCG surge increased the expression of THBD, EPCR, PAR1, and PAR4 in granulosa and cumulus cells of preovulatory follicles. We also detected increased ovarian content of prothrombin and APC after the preovulatory hCG stimulation. The moderate increases in ovarian prothrombin after the hCG stimulation, coupled with elevated levels of its receptor THBD, likely led to the observed large increases in ovarian APC levels. Coupled with increased expression of EPCR in granulosa and cumulus cells, ovarian APC is presumably bound by EPCR, thus facilitating its interaction with PAR1 or PAR4, both of which were also stimulated by the preovulatory LH/hCG surge. Our findings of the ability of thrombin and agonists for PAR1 and PAR4 to suppress hCG-stimulated progesterone biosynthesis represent a newly identified regulatory role for the thrombin-APC signaling system. Because the ovarian cells also express protein C, our findings suggest the involvement of endogenous APC in the ovary and are consistent with observed suppression of hCG-stimulated progesterone production at extremely low doses of thrombin, likely through its activation of protein C to allow an enzymatic amplification of the proteolytic cascade mediated by THBD and EPCR expressed in granulosa cell surface.
The intraovarian thrombin-THBD-APC-PAR1/4 signaling system. Thrombin (T) from ovarian or systematic sources binds to THBD expressed in the plasma membrane of cumulus and granulosa cells in preovulatory follicles. Membrane-bound thrombin activates protein C (PC), leading to the binding of APC to EPCR expressed in the surface of the same cells. Membrane-bound APC, in turn, cleaves the extracellular domain of PAR1 or PAR4 receptors to stimulate downstream signaling, leading to the inhibition of cAMP and progesterone biosynthesis stimulated by LH/hCG mediated by the LH receptor (LHR). After the preovulatory LH increases, increased expression of THBD and EPCR in granulosa cells facilitates intraovarian thrombin and APC levels, leading to APC stimulation of increased PAR1 and PAR4 receptors in granulosa and cumulus cells. This signaling system may allow optimal luteinization of preovulatory follicles and provides a valuable model to elucidate molecular mechanisms underlying thrombin actions in noncoagulation pathways.
The present intraovarian regulatory system represents a unique hormonal modulation system mediated by the regulation of receptors but not ligands in the signaling pathway. Transcript levels for the two key ligands (thrombin and protein C) were not changes after the preovulatory LH signal. Instead, four receptors in the thrombin-APC-PAR1/4 pathway showed dramatic increases after the LH surge, leading to increase levels of APC to interact with increased levels of the downstream PAR1 and PAR4 receptors.
Because the action of APC is species specific (11) and mouse APC is not available, we focused on studies using thrombin to regulate the functions of luteinizing granulosa cells. Because these cells express thrombin, THBD, EPCR, and thrombin receptors, it is likely that THBD-bound thrombin can activate endogenous protein C bound to EPCR, leading to the APC activation of thrombin receptors (2). The importance of endogenous thrombin and APC in the preovulatory ovaries is underscored by the demonstration of increased ovarian content of APC after the preovulatory LH/hCG stimulation in vivo as well as by the observed ability of antagonists for thrombin and the PAR1 receptor to augment hCG-stimulated progesterone biosynthesis in vitro and in vivo, presumably by blocking the actions of endogenous thrombin and APC.
Both prothrombin and protein C are also produced in the liver and posttranslationally modified in a vitamin K-dependent reaction (16). Thrombin production and PAR1 receptor expression have been found in bovine ovarian follicles (17). Consistent with our detection of prothrombin in the preovulatory ovaries, bovine granulosa cells isolated from follicles of various sizes contain both prothrombin mRNA and immunologically reactive prothrombin. In addition, mRNA for γ-glutamyl carboxylase, an enzyme essential for the vitamin K-dependent posttranslational modification of prothrombin, is also expressed in granulosa cells (17). APC has neuroprotective effects in neuronal tissues and exhibits antiinflammatory effects on leukocytes and endothelial cells of blood vessels (3). APC affects endothelial cells by inhibiting inflammatory mediator release and down-regulating vascular adhesion molecules. An earlier paper reported that treatment with thrombin or a PAR1 agonist stimulated gelatinase activities and the production of IL-8 and monocyte chemoattractant protein-1 by cultured human luteinizing granulosa cells (18). Although a potential mediatory role of APC has not been analyzed, the stimulatory effects of thrombin were inhibited by thrombin antagonists, and immunocytochemical studies showed that thrombin and a PAR1 agonist induced translocation of nuclear factor κ-B to the nucleus from the cytoplasm. Of interest, earlier studies using endothelial cells also demonstrated the induction of monocyte chemoattractant protein-1 expression by APC (2). Coupled with our studies, it is becoming clear that the thrombin-APC signaling system could act through PAR1/4 receptors in luteinizing granulosa cells to regulate steroidogenesis and the secretion of gelatinase and key cytokines during the periovulatory period.
Thrombin is a serine protease that cleaves fibrinogen to form fibrin monomers and uniquely cleaves cell surface receptors, known as PAR. Four members of this seven-transmembrane domain, G protein-coupled receptor family have been PAR1–4 (12). Protease cleavage of these receptors creates a neo-NH2 terminus, which acts as a tethered ligand that binds to the seven-transmembrane segment of the PAR. The tethered neo-NH2 terminus activates the receptor, independent of thrombin or trypsin binding. Free peptides, as short as six amino acids, can mimic the neo-NH2 terminus and activate PAR. We demonstrated that PAR1 and PAR4 are expressed in ovarian granulosa and cumulus cells and used thrombin receptor-activating peptides containing the initial amino acid sequence SFLLR for PAR1 and the peptide GYPGKF for PAR4 to suppress cAMP production and progesterone biosynthesis by cultured granulosa cells. Because these peptides have no protease activity and thus, in contrast to thrombin, are incapable of cleaving fibrinogen. Our use of these peptide agonists for PAR1 and PAR4 receptors further demonstrated fibrin-independent, receptor-mediated roles of thrombin in the ovary. The observed activation of the inhibitory G protein mediated by PAR1 and PAR4 receptors is consistent with studies using other thrombin target cells (19, 20).
In preovulatory ovaries, diverse paracrine signaling systems have been identified. Preovulatory LH stimulates granulosa cells to produce brain-derived growth factor (8), epidermal growth factor-like factors (5), vascular endothelial growth factor (21), prepronociceptin (22), endothelin-1 (10), and other hormones. These intraovarian factors, in turn, stimulate diverse ovarian physiological processes, including oocyte maturation, cumulus cell expansion, first polar body extrusion, luteinization, and early embryo development. In addition to the regulation of progesterone biosynthesis by the intraovarian thrombin signaling system reported here, an earlier study indicated that the a TGF superfamily member (TNF-related weak inducer of apoptosis) of theca cell origin is also induced by the preovulatory LH surge and capable of suppressing LH/hCG-stimulated progesterone synthesis by luteinizing granulosa cells (9). These findings are consistent with the notion that redundant and overlapping intraovarian mechanisms are involved in the prevention of excessive luteinization. Although disruption of EPCR gene in mice caused placental thrombosis and early embryonic lethality (23), no infertility phenotypes for PAR1- or PAR4-null mice were found despite their survival to adulthood (24, 25), suggesting the presence of redundant pathways. In addition to understanding the regulation of ovarian functions during the preovulatory period, the present finding provides a unique model with which to elucidate the molecular mechanisms underlying the THBD-APC-PAR1/4 signaling system mediated through the regulation of key receptors. It is becoming clear that this signaling system may provide fine control of thrombin and APC actions in different tissues, leading to pleiotropic control of cellular processes.
Materials and Methods
Animals and reagents
Immature female CD-1 mice from Charles River Laboratories (Wilmington, MA) or CLEA Japan (Tokyo, Japan) were used for all experiments except the DNA microarray analyses. Immature mice were treated with 5 IU of eCG (Calbiochem, La Jolla, CA), followed by 5 IU hCG (Sigma, St. Louis, MO) 48 h later to simulate follicle maturation and ovulation, respectively. Animal care followed National Institutes of Health guidelines and was approved by Animal Research Committees at Akita University School of Medicine and Stanford University School of Medicine. Thrombin and its antagonist huridin were obtained from Sigma; PAR1 agonist (synthetic peptide SFLLR) and PAR4 agonist (synthetic peptide GYPGKF) were from Bachem (Torrance, CA); and the PAR1 antagonist SCH79797 was from Tocris Bioscience (Ellisville, MO).
DNA microarray and real-time RT-PCR analysis
For the original DNA microarray studies (8), B6D2F1 mice were injected at 21 d of age with Humegon (7.5 U per animal; Organon, Oss, The Netherlands) containing both FSH and LH activities to stimulate follicular growth. Forty eight hours later, some animals were treated ip with Pregnyl (5 U per animal) containing LH activity to induce ovulation. Ovaries were dissected from animals killed bihourly after Humegon treatment (three mice per group) and hourly after Pregnyl treatment for RNA extraction (TRIzol; Invitrogen, Carlsbad, CA). RNA samples were hybridized to the Affymetrix mouse MGU74v2 arrays A, B, and C as described earlier (8).
To confirm DNA microarray data, ovaries were dissected from immature CD-1 mice at 23 d of age before or at 48 h after 5 IU eCG treatment. Some eCG-primed animals were further treated with an ovulatory dose (5 IU) of hCG. For isolation of different ovarian cell types, largest follicles were punctured to obtain oocytes, granulosa cells, cumulus cells, and theca shells. Total RNA was extracted using an RNeasy Micro Kit (QIAGEN Sciences, Valencia, CA), and samples (2 μg) were reverse transcribed for subsequent PCR analysis. Real-time PCR was performed in 25 μl final volume containing 2 μl of the reverse transcriptase reaction product, 0.5 μm primers, 0.2 μm fluorescently labeled probe (3′,5′-carboxy tetramethylrhodamine; 5′, 6′-carboxy fluorescein), and the PCR reagent mixtures (QuantiTect Probe PCR Kit, QIAGEN Sciences). Standard curves were generated by serial dilution of each plasmid DNA. Primer pairs and fluorescent probes used are listed in Supplemental Table 2. Real-time PCR was performed using iQ SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA), and histone H2A levels were used for copy number normalization. The assays were performed on a Smart Cycler TD System (Cepheid, Sunnyvale, CA) with an initial enzyme activation step of 15 min at 95 C, followed by 45 cycles of two-step PCR (94 C, 15 sec; 60 C, 60 sec). Data are presented as relative expression and normalized based on histone H2a levels. Results represent mean ± se of fold changes of normalized expression.
Immunohistochemistry
To localize ovarian cell types expressing different genes, ovaries were obtained from eCG-primed mice before and at 4 and 8 h after hCG injection. After fixation with 20% formalin neutral buffer solution for 16 h at 4 C, tissues were embedded in paraffin and sectioned at 5-μm intervals before deparaffinization and dehydration. Endogenous peroxidase activities were quenched with 1% periodic acid for 30 min. After blocking with 5% normal serum (DAKO Corp., Carpinteria, CA) for 30 min, slides were incubated with THBD antibody (R&D systems, Minneapolis, MN; 25 μg/ml), EPCR antibody (Novus Biologicals, Littleton, CO; 20 μg/ml), PAR-1 antibody (GenWay Biotech, San Diego, CA; 15 μg/ml), or PAR-4 antibody (GenWay Biotech; 5 μg/ml) overnight at 4C and washed three times in Tris-buffered saline (TBS). Slides were then incubated with biotinylated secondary antibodies (DAKO) for 30 min. After three washes, bound antibodies were visualized using a Histostain SP kit (Zymed Laboratories, South San Francisco, CA). For negative controls, the primary antibody was replaced by nonimmune IgG (DAKO).
Measurement of ovarian prothrombin and APC content as well as immunoblotting for THBD, EPCR, and PAR4
Ovaries were obtained from immature (d 23) mice and those at 48 h after treatment with 5 IU eCG. Some eCG-primed animals were injected with a single injection of hCG (5 IU) for different time intervals. Ovaries (four to five per group) were collected and homogenized in M-PER mammalian protein extraction reagent (Thermo Scientific, Rockford, IL) containing proteinase inhibitor (Thermo Scientific), incubated for 15 min on ice, and then centrifuged at 11,000 × g at 4 C for 15 min. Extracted proteins were used to measure prothrombin and APC levels using ELISA kits for prothrombin (Assaypro, St. Charles, MO) and mouse APC (USCN Life Science, Wuhan, China), respectively, according to manufacturer's instructions. For the determination of ovarian content of THBD, EPCR, and PAR4, ovaries from mice at 48 h after eCG treatment and 8 h after hCG treatment were extracted as described earlier before loading on 10% Bis-Tris gels (Invitrogen) for electrophoresis at 120 V for 2 h. Subsequently, proteins in gels were electrotransferred onto polyvinylidene difluoride membranes for immunoblotting analyses using primary antibodies against THBD (Abcam, San Francisco, CA; 1:2000 dilution), EPCR (R&D systems; 1:500 dilution), PAR4 (Santa Cruz Biotechnology, Inc.; Santa Cruz; CA; 1:200 dilution), and β-actin (Abcam; 1:10000 dilution). After washing, membranes were incubated with secondary antibodies conjugated to horseradish peroxidase before signal development. After stripping, the same membranes were used for β-actin immunostaining as loading controls.
Granulosa cell cultures and analyses of progesterone and cAMP production
Immature mice were primed with eCG for 48 h, followed by hCG treatment. At 6 h after hCG injection, luteinizing granulosa cells were punctured from preovulatory follicles with ovarian debris and small follicles were removed. Cells were collected after centrifugation at 500 × g for 10 min. After repeated washing, granulosa cells were resuspended in McCoy's 5a media (Invitrogen) supplemented with 10−7m androstenedione, 2 mml-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin. After treatment of cultured granulosa cells with increasing doses of hCG with or without thrombin and agonists for PAR1 and PAR4 receptors, media cAMP levels (basal level, 15 min of incubation; stimulated levels, cells pretreated for 30 min followed by 10 μm forskolin for 15 min) and progesterone content (48 h of culture) were determined using specific immunoassays as described elsewhere (26).
Caspase-3/7 activity assay
For apoptosis studies, luteinizing granulosa cells from hCG-treated animals were cultured in 96-well plates (10,000 cells per well) with medium described above, and caspase-3/7 activity was determined using a luminescent assay (Caspase-Glo 3/7 Assay; Promega Corp., Madison, WI). Caspase-Glo 3/7 reagent (100 μl) was added to the culture plate, and cells in culture medium were incubated at room temperature for 1 h before measurement of luminescence of each sample in a plate-reading luminometer (Bio-Rad).
Preovulatory follicle cultures and intrabursal injections
To evaluate the effects of antagonists for thrombin and PAR1 on progesterone production, preovulatory follicles were isolated from ovaries of immature mice at 48 h after eCG treatment. Follicles (10–20 per 4 ml round-bottom vial; BD Falcon, Bedford, MA) were cultured with hCG (300 ng/ml) in the presence or absence of different antagonists in Leibovitz's L-15 medium (Invitrogen). The vials were flushed at the start of the culture with O2/N2 (at a 1:1 ratio), sealed, and cultured at 37 C for 18 h. After culture, media content of progesterone was determined.
To investigate the actions of endogenous thrombin signaling in vivo, intrabursal injections of thrombin antagonist hirudin and the PAR1 antagonist SCH79797 were performed. At 48 h after treating immature mice with 5 IU eCG, animals were lightly anesthetized before injections of hirudin (10 IU/10 μl) or SCH79797 (1 μmol/10 μl) through the fat pad into ovarian bursa by using a 30-gauge needle. Successful injections were confirmed based on bursa swelling. Saline was injected into the contralateral ovary as control. After intrabursal injections, animals were treated ip with 5 IU hCG, and ovaries were dissected 48 h later for progesterone measurement. Ovarian progesterone content was determined as described elsewhere (9). Briefly, ovaries were homogenized in water and steroids were extracted with diethyl ether. The mixture was snap frozen in dry ice/acetone to remove the aqueous phase. The organic phase was collected and the extraction was repeated. After extracts were pooled and evaporated to dryness under vacuum, steroids were dissolved in an assay buffer and used for RIA.
Statistical analysis
Results are presented as mean ± se of three or more independent determinations. Statistical significance was determined by using the ANOVA test followed by Fisher's protected least significant difference with P < 0.05 being statistically significant.
Acknowledgments
This work was supported by funds from the National Institute of Child Health and Human Development (U54 HD068158 as part of the Specialized Cooperative Centers Program in Reproduction and Infertility Research); K.K. was supported by a Grant-In-Aid for Scientific Research (Grant-In-Aid for Young Scientists B:21791539 and Grant-in-Aid for Scientific Research on Priority Areas, THE GERMLINE: Its Developmental Cycle and Epigenome Network: 23013004) and research funds from the Terumo Life Science Foundation, the Kanae Foundation for the Promotion of Medical Science, the Novartis Foundation for the Promotion of Science, and the Yamaguchi Endocrine Research Foundation.
Disclosure Summary: The authors have nothing to disclose.
Abbreviations
- APC
Activated protein C
- eCG
equine chorionic gonadotropin
- EPCR
endothelial protein C receptor
- Gi
inhibitory G protein
- hCG
human chorionic gonadotropin
- PAR
protease-activated receptor
- THBD
thrombomodulin.
