|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Biological Sciences and the Walther Cancer Institute, The University of Notre Dame, Notre Dame, Indiana 46556
ABSTRACT
Studies in both mammalian and nonmammalian ovarian model systems have demonstrated that activation of the mitogen-activated protein kinase (MAPK) and protein kinase C (PKC) signaling pathways modulates steroid biosynthesis during follicle development, yet the collective evidence for facilitory versus inhibitory roles of these pathways is inconsistent. The present studies in the hen ovary describe the changing role of MAPK and PKC signaling in the regulation of steroidogenic acute regulatory protein (STAR) expression and progesterone production in undifferentiated granulosa cells collected from prehierarchal follicles prior to follicle selection versus differentiated granulosa from preovulatory follicles subsequent to selection. Treatment of undifferentiated granulosa cells with a selective epidermal growth factor receptor (EGFR) and ERBB4 receptor tyrosine kinase inhibitor (AG1478) both augments FSH receptor (Fshr) mRNA expression and initiates progesterone production. Conversely, selective inhibitors of both EGFR/ERBB4 and MAPK activity attenuate steroidogenesis in differentiated granulosa cells subsequent to follicle selection. In addition, inhibition of PKC signaling with GF109203X augments FSH-induced Fshr mRNA plus STAR protein expression and initiates progesterone synthesis in undifferentiated granulosa cells, but inhibits both gonadotropin-induced STAR expression and progesterone production in differentiated granulosa. Granulosa cells from the most recently selected (9- to 12-mm) follicle represent a stage of transition as inhibition of MAPK signaling promotes, while inhibition of PKC signaling blocks gonadotropin-induced progesterone production. Collectively, these data describe stage-of-development-related changes in cell signaling whereby the differentiation-inhibiting actions of MAPK and PKC signaling in prehierarchal follicle granulosa cells undergo a transition at the time of follicle selection to become obligatory for gonadotropin-stimulated progesterone production in differentiated granulosa from preovulatory follicles.
follicle, granulosa cells, MAP kinase, ovary, PKC, progesterone, signal transducers, STAR, steroidogenesis
The process of ovarian follicle maturation in both mammalian and avian species is coupled with a functional differentiation of the granulosa cell layer. Granulosa cells from immature follicles are generally considered undifferentiated and produce minimal amounts of steroid hormones. By comparison, granulosa cells from mature, preovulatory follicles and granulosa-luteal cells produce appreciable amounts of steroid hormones, particularly in response to gonadotropins. Considerable data exist to indicate that the process of steroidogenesis in differentiated granulosa cells is largely mediated by the actions of the cyclic adenosine monophosphate (cAMP)/protein kinase A (PKA) signaling pathway. However, recent work has demonstrated in both mammalian and avian model systems that alternative signaling pathways also mediate the production of steroids, in part through the modulation and regulation of steroidogenic acute regulatory protein (STAR). Specifically, while regulation of STAR protein and enhancement of its biological activity via phosphorylation is largely dependent upon signaling by the cAMP/PKA pathway [1–3], the mitogen activated protein kinases (MAPK) and protein kinase C (PKC) signaling pathways have also been implicated in modulating STAR protein expression and steroidogenesis. However, conflicting data describing the effects of these signaling pathways on STAR protein expression and steroidogenesis have made defining their importance difficult. Reports using granulosa cells from mammalian or nonmammalian models have suggested that acute up-regulation of STAR protein expression is attenuated by EGF-family ligands signaling through MAPK, and that alleviation from inhibitory MAPK signaling is prerequisite for STAR expression and subsequent steroid synthesis [1–5]. Similarly, it has been demonstrated that PKC signaling induced by the phorbol ester, phorbol 12-myristate 13-acetate (PMA), inhibits STAR protein expression, while attenuation of the PKC signaling pathway using pharmacological PKC inhibitors (e.g., staurosporine, GF109203X) greatly enhances gonadotropin-induced STAR protein expression and progesterone production [2, 6, 7]. These data collectively suggest an inhibitory role for the MAPK and PKC signaling pathways in granulosa cell steroidogenesis.
By contrast, accruing evidence suggests that both MAPK and PKC signaling are obligatory for steroidogenesis. Recent studies in both rat Leydig cells and in granulosa cells from hen preovulatory follicles demonstrate that inhibition of PKC signaling attenuates progesterone production [8–11]. Additionally, recent studies using Leydig cells, adrenocortical cells, and granulosa cells indicate that MAPK signaling is obligatory for ligand-induced STAR protein expression and steroid synthesis [12–17]. Further examination of these seemingly contradictory data suggests the possibility of stage of differentiation dependent cell-signaling requirements in the regulation of steroidogenesis [18]. Despite the morphological and functional differences between the avian and mammalian ovaries, the regulation of STAR protein expression and steroid synthesis in granulosa cells for both model systems is predicted to be dependent upon relative stage of differentiation. Specifically, stage-of-differentiation-related differences likely include changes in cross-talk among cell-signaling pathways, availability of ligand(s) or ligand receptor(s), and timing of signaling events.
By comparison to the cycle-related waves of follicle development within the mammalian ovary, the ovary from a reproductively active hen simultaneously supports follicles from all stages of development (e.g., primordial through preovulatory), and each day one follicle from a small cohort of undifferentiated (prehierarchal; 6–8 mm in diameter) follicles is selected into the preovulatory hierarchy. The most recently selected follicle is morphologically distinguishable based on size and extent of yolk incorporation (9–12 mm in diameter). While granulosa cells from prehierarchal follicles actively proliferate and maintain an undifferentiated phenotype, granulosa cells from the nongerminal disc region of preovulatory follicles are largely nonproliferative and are terminally differentiated [19]. The transition from an undifferentiated to a differentiated state is directly associated with follicle selection. Changes occurring within 24 h following follicle selection include the transition from FSH- to LH-dependence [20, 21], an increasing capacity for steroid synthesis due to increased cytochrome P450 cholesterol side chain cleavage (P450scc) enzyme activity [22], and the capacity for acute up-regulation (within 1 h) of STAR protein expression [2]. Consequently, the granulosa layer from preovulatory follicles rapidly synthesizes copious (e.g., µg) amounts of progesterone in response to LH [23, 24]. While granulosa cells from preovulatory follicles are predominantly LH-responsive, a modest responsiveness to FSH is maintained [20, 21, 25]. These inherent properties of hen granulosa cells provide an excellent model to study cell-signaling events simultaneously in granulosa cells from follicles from all stages of development.
Accordingly, these studies focused on elucidating the stage of differentiation-dependent actions of MAPK and PKC signaling in regulating STAR expression and steroid synthesis in hen granulosa cells. The three distinct stages of follicular maturation investigated included undifferentiated granulosa cells from follicles prior to follicle selection, granulosa cells from the most recently selected (9–12 mm) follicle, and differentiated cells from preovulatory follicles collected prior to the preovulatory LH surge.
Single-comb white Leghorn hens 25–35 wk of age (Creighton Bros., Warsaw, IN), laying sequences of six or more eggs on a regular basis were used in all studies described. Hens were housed individually in laying batteries, with free access to feed (Purina Layena Mash; Purina Mills, St. Louis, MO) and water, under a controlled photoperiod of 15L:9D (lights-on at midnight). The lay patterns were monitored daily. Hens were killed by cervical dislocation 16–18 h prior to a midsequence ovulation, and the ovary was removed and placed in ice-cold sterile 1% NaCl solution for immediate use. All procedures described herein were reviewed and approved by the University of Notre Dame Institutional Animal Care and Use Committee, and were performed in accordance with the Guiding Principles for the Care and Use of Laboratory Animals.
Recombinant human FSH and ovine LH were provided by the National Hormone and Pituitary Program (Torrance, CA). The cell permeable cyclic adenosine monophosphate analog, 8-bromo-cAMP (8-br-cAMP), was purchased from Sigma-Aldrich (St. Louis, MO). The selective MAPK inhibitor U0126 and the PKC inhibitor GF109203X (inhibits both conventional and novel isoforms) were purchased from BioMol (Plymouth Meeting, NJ), while the EGFR/ERBB4 receptor tyrosine kinase inhibitor, AG1478, was from Calbiochem (San Diego, CA). All hormones, growth factors, and inhibitors were utilized at optimal doses as previously described [24, 26].
Prehierarchal follicles (6–8 mm; prior to follicle selection), the single, most recently selected follicle (9–12 mm), and the second (F2) plus third (F3) largest preovulatory follicles (representing a stage several days following selection) were removed from the ovary and placed into sterile ice-cold saline solution. Granulosa layers from prehierarchal and preovulatory follicles were collected and combined within their respective stage of development, and dispersed for culture as previously described [2, 27]. In some instances, an aliquot of cells was immediately frozen at –70°C (T0 control). The remaining cells were cultured at 40°C in an atmosphere of 95% air:5% CO2 in six-well polystyrene culture plates (Beckton Dickinson Labware, Franklin Lakes, NJ) with a density of approximately 1 x 106 per well in 2 ml Dulbeccos Modified Eagle Medium (DMEM) plus 2.5% FBS with 0.1 mM nonessential amino acids and 1% antibiotic-antimycotic mixture (Invitrogen, Carlsbad, CA) for 20 h. In addition, short-term cultures were conducted with F2 plus F3 follicle granulosa where 5 x 105 cells in 2 ml DMEM were incubated in 12 x 75-mm polypropylene tubes (Fisher Scientific, Pittsburgh, PA) for 4 h [26]. Where appropriate, U0126 (10 µM), AG1478 (10 µM), and GF109203X (30 µM) were preincubated with cells for 1 h prior to the addition of gonadotropins, 8-br-cAMP, or TGF
.
Immunoblot Analysis of STAR, Phosphorylated STAR, and MAPK Proteins
Analysis of STAR, phosphorylated STAR (P-STAR), MAPK, and P-MAPK protein was conducted as previously described [26, 27]. The anti-STAR serum was generously provided by Dr. D.B. Hales (University of Illinois, Chicago, IL). The rabbit anti-phosphorylated-STAR was a gift from Dr. Steven King (Baylor College of Medicine, Houston, TX), and was used at a 1:5000 dilution. The mouse monoclonal P-MAPK1/3 was from Upstate, and used at a 1:2000 dilution (Charlottesville, VA), while the rabbit anti-MAPK1 is from Santa Cruz (Santa Cruz, CA), and used at a 1:3000 dilution. All primary and secondary antibodies were diluted in 5% milk in tris-buffered saline-0.1% tween 20. The
-tubulin (1:10, 000 dilution) antiserum used for standardization of STAR and P-STAR proteins was obtained from Sigma-Aldrich. The horseradish peroxidase-conjugated anti-rabbit and anti-mouse IgG secondary antibodies were from Pierce Endogen (Rockford, IL) and used at a dilution of 1:10 000.
Northern Blot Analysis of Fshr mRNA
Levels of Fshr mRNA were evaluated by Northern blot analysis using a chicken Fshr cDNA probe as previously described [20]. While this gonadotropin receptor is reported to express multiple transcripts, only the predominant 2.5 kb transcript was quantified in the present studies. Images were visualized on phosphor screens using the Storm 840 PhosphoImager and analyzed using the ImageQuant data reduction system (Molecular Dynamics, Inc., Sunnyvale, CA). All Fshr mRNA data were standardized to 18S ribosomal (r) RNA.
Real-time PCR and quantification of EGF family ligands during follicular maturation.
Real time PCR for Egf, amphiregulin, and Btc was performed using primer pairs and conditions as described previously [26]. Briefly, RNA was isolated using TRIzol Reagent (Invitrogen, Carlsbad, CA) for random primed cDNA synthesis using the Promega Reverse Transcription System (Promega, Madison, WI) according to the protocol provide with the kit. Real-time PCR reactions were performed using the ABgene Absoulte QPCR Sybr Green mix (ABgene, Rochester, NY) on the ABI 7700 Thermocycler (Applied Biosystems, Foster City, CA) using the following parameters: 2 min at 94°C followed by 40 cycles each of 15 sec at 95°C, 1 min at 60°C, 1 min at 72°C, with each reaction run in triplicate. The Ct value was determined for each reaction through use of the Sequence Detection software (v.1.6.3, Applied Biosystems). Quantification was performed using the 
Ct method [28]. Briefly, the target Ct was determined for each sample then normalized to the 18s rRNA Ct from the same sample (18s rRNA Ct subtracted from the target Ct yields the
Ct). These values were then compared with control levels using the 2–
Ct method, and expressed as fold-difference compared with ovarian stromal tissue as an arbitrary reference.
Progesterone in media samples was quantified by radioimmunoassay as previously described [9]. Data are expressed as a mean fold-difference compared with an appropriate control for the combined replicate experiments.
Experiments were independently replicated a minimum of three times unless otherwise stated. Data among replicate experiments were standardized by expressing individual experiments as a fold-difference versus either T0 controls (for Fshr data), or versus cultured control cells (progesterone). Transformed data from replicated experiments were analyzed by one-way ANOVA without including values from the control group (arbitrarily set to 1.0), and the Fisher's protected least significant difference multiple range test. Alternatively, in cases where a specific comparison was between two groups, original data from the replicate experiments were analyzed using a Student t-test.
Stage-of-Differentiation Dependent Effects of Inhibiting MAPK Signaling on Progesterone Production
The cAMP agonist, 8-br-cAMP (1 mM), induced a modest fold-increase of progesterone in undifferentiated granulosa cells from prehierarchal follicles following a 20 h culture (corresponds to an increase from an overall mean of 0.2 ng/ml from cultured control cells to 0.5 ng/ml in 8-br-cAMP-treated cells). This longer culture period was chosen because it was previously determined that no significant progesterone production is induced from undifferentiated granulosa cells by 8-br-cAMP after short-term culture [23]. The fold-increase in progesterone production induced by 8-br-cAMP was further increased when pretreated and cultured in the presence of the MAPK inhibitor, U0126 (Fig. 1A; P < 0.01). The induction of progesterone by 8-br-cAMP correlates with an increase in STAR protein expression and phosphorylation (Fig. 1A, top). Similarly, granulosa cells from the single, most recently selected (9–12 mm) follicle showed enhanced progesterone production in response to FSH (100 ng/ml) when cultured with U0126 (Fig. 1B; P < 0.003; mean progesterone levels in FSH-treated cells amounted to 3.7 ng/ml), as well as an increase in STAR protein expression (Fig. 1B, top).
|
By comparison, short-term (4 h) challenge of differentiated granulosa cells from preovulatory follicles (F2 plus F3) with FSH, and particularly LH (100 ng/ml) or 8-br-cAMP, enhanced progesterone production compared with control cultured cells (mean control progesterone levels were 17.6 ng/ml, compared with 36.2 ng/ml, 242.8 ng/ml, and 245.6 ng/ml for the FSH, LH, and 8-br-cAMP treatments, respectively). However, unlike undifferentiated granulosa cells from prehierarchal follicles, pre-treatment with U0126 greatly attenuated both gonadotropin- and 8-br-cAMP-induced progesterone synthesis (Fig. 2A; P < 0.001). The disparity in response between undifferentiated and differentiated granulosa cells was not due to length of culture, as a 20-h treatment of F2 plus F3 follicle granulosa cells with either gonadotropin or 8-br-cAMP in the presence of U0126 also greatly attenuated agonist-induced progesterone production (Fig. 2B). Unexpectedly, the attenuation of progesterone synthesis by U0126 was not reflected by STAR protein expression or phosphorylation, as treatment with U0126 for 4 (Fig. 2A, top) or 20 h (data not shown) enhanced both FSH- and LH-induced STAR protein expression and phosphorylation (Fig. 2A, top).
|
Active EGFR Signaling Is Required for Progesterone Production in Differentiated but Not Undifferentiated Granulosa Cells
To determine whether the actions of MAPK signaling on granulosa cell progesterone production are mediated specifically through the activation of EGFR/ERBB4 receptor(s), undifferentiated granulosa cells from prehierarchal follicles were pretreated without or with the selective EGFR/ERBB4 inhibitor, AG1478, and then challenged with FSH or 8-br-cAMP. Similar to U0126 treatment, AG1478 enhanced FSH- and 8-br-cAMP-induced progesterone production (Fig. 3A; P < 0.03) and STAR protein expression and phosphorylation (Fig. 3A, top). Additionally, treatment with AG1478 in combination with FSH or LH significantly increased progesterone production in granulosa cells from the 9- to 12-mm follicle when compared to the gonadotropin treatment alone (Fig. 3B; FSH, P < 0.03; LH, P < 0.02). The stimulatory effect of AG1478 on FSH- and LH-induced progesterone production is associated with an increase in STAR protein expression (Fig. 3B, top). By contrast, co-treatment with AG1478 largely prevented FSH-, LH-, and 8-br-cAMP-induced progesterone production in differentiated granulosa cells after both short-term (Fig. 4A; P < 0.01) and longer-term (Fig. 4B; P < 0.001) culture. However, FSH-, LH-, and 8-br-cAMP-induced STAR protein expression and phosphorylation are enhanced when treated with AG1478 for 4 (Fig. 4A, top) or 20 h (data not shown).
|
|
Active PKC Signaling in Granulosa Cells Is Required for Agonist-Induced Progesterone Production Subsequent to Follicle Selection and Differentially Affects Gonadotropin-Induced STAR Protein Expression and Phosphorylation in a Stage-Dependent Fashion
Culture of undifferentiated granulosa cells with the PKC inhibitor, GF109203X, significantly enhanced FSH-induced progesterone production when compared with undifferentiated cells cultured with FSH alone (Fig. 5A; P < 0.05), and enhanced FSH-induced STAR protein and phosphorylation (Fig. 5A, top). By contrast, FSH- and LH-induced progesterone production was dramatically attenuated by GF109203X in granulosa cells collected from the most recently selected (9–12 mm) follicle (Fig. 5B; FSH, P < 0.001; LH, P < 0.003). The attenuation of agonist-induced progesterone synthesis correlated with the abrogation of STAR protein expression (Fig. 5B, top). Moreover, treatment of differentiated granulosa cells from F2 plus F3 follicles with GF109203X blocked FSH-, LH-, and 8-br-cAMP-induced progesterone production following both short-term (Fig. 6A) and longer-term (Fig. 6B) culture (P < 0.001), as well as STAR protein expression and phosphorylation at 4 (Fig. 6A, top) and 20 h (data not shown).
|
|
Activated MAPK and EGF Family Ligand mRNA Expression Decline as Follicles Mature
Analysis of the level of P-MAPK in granulosa cells from different stages of folliculogenesis indicates that levels of activated MAPK remain high following follicle selection into the preovulatory hierarchy, and subsequently decline (Fig. 7A). An increase in P-MAPK is observed in granulosa cells from the largest, F1, preovulatory follicle as compared to granulosa cells from the second largest, F2, follicle. Likewise, mRNA expression of the EGF family members, Egf, amphiregulin, and Btc significantly decline during follicle development (Fig. 7B). While Egf mRNA expression remains elevated at the 6- to 8-mm and 9- to 12-mm stages, it subsequently declines significantly by the F3 stage, and declines further by the F2 and F1 preovulatory stage (Fig. 7B, top). Furthermore, data indicate that mRNA encoding the related family members, amphiregulin and Btc, decline in a similar fashion following selection into the preovulatory hierarchy (Fig. 7B middle and bottom, respectively).
|
Inhibition of EGFR and PKC Signaling Promotes Fshr mRNA Expression in Undifferentiated Granulosa Cells
To determine if the effects of AG1478 and GF109203X to enhance STAR protein levels and progesterone production in undifferentiated granulosa cells can be attributed, at least in part, by the up-regulation of FSHR expression, the relative abundance of Fshr mRNA was evaluated using Northern blot analysis. Levels of Fshr mRNA in undifferentiated granulosa cells increased when cultured with AG1478 for 20 h as compared to the level in control cultured cells, while the presence of both AG1478 and FSH further increased Fshr mRNA expression as compared to those treated with AG1478 alone (Fig. 8A, P < 0.002). Similarly, treatment of undifferentiated granulosa cells for 20 h in the presence of GF109203X increased Fshr mRNA when compared with control cells, and co-treatment with FSH resulted in a further increase in Fshr mRNA expression (Fig. 8B; P < 0.03).
|
The novel findings discussed herein demonstrate a functional transition from inhibitory to obligatory MAPK and PKC signaling during follicle development with respect to the promotion of STAR protein expression and progesterone production. Moreover, the two pathways demonstrate a temporal difference in the stage at which this transition occurs, plus unique actions on STAR protein expression. The data presented are consistent with the proposal that in undifferentiated granulosa cells from prehierarchal follicles, activation of either EGFR receptor-mediated MAPK [2] or PKC [26] signaling tonically suppresses FSH-induced STAR protein expression and progesterone production. By comparison, following selection into the preovulatory hierarchy, granulosa cells undergo functional differentiation, and both the MAPK and PKC signaling pathways become obligatory for gonadotropin- (primarily LH-) induced progesterone production. Accordingly, these data describe a role for MAPK and PKC activation initially in the suppression of granulosa cell differentiation prior to follicle selection, and subsequently for the fully potentiated production of progesterone during the preovulatory stages of follicle development (Fig. 9).
|
It has previously been reported in undifferentiated granulosa cells from hen prehierarchal follicles that the MAPK or PKC signaling pathway, activated via EGF family ligands, inhibits premature differentiation, as demonstrated by their ability to inhibit FSH-induced LH receptor (LHR) expression and progesterone production [24, 29, 30]. Associated with this tonic suppression of differentiation is the finding that basal levels of activated MAPK, as well as the mRNA for EGF family members Egf, amphiregulin, and Btc, are greater in undifferentiated granulosa cells from prehierarchal follicles compared to those in the later stages of development in the preovulatory hierarchy (Fig. 7, A and B). Results herein support these findings in that inhibition of MAPK signaling (using U0126), or inhibition of PKC signaling (using GF109203X), facilitates FSH- and 8-br-cAMP-induced progesterone production in undifferentiated granulosa cells (Figs. 1A and 5A). Similarly, inhibition of EGFR receptor activation, using AG1478, enhances FSH- and 8-br-cAMP-induced progesterone production, as well as the expression and phosphorylation of STAR protein (Fig. 3A). Together, these data establish that MAPK signaling is maintained through active EGFR and/or ERBB4 receptor tyrosine kinase signaling [26] to prevent premature differentiation. This suppression of differentiation is achieved in part through inhibition of FSHR expression, as blocking either EGFR receptor activation or PKC signaling results in an up-regulation of Fshr mRNA expression (Fig. 8A). While inhibitory effects of both MAPK and PKC activity are induced by EGF family ligands and EGFR signaling in undifferentiated granulosa cells [26], each pathway can independently inhibit STAR expression and progesterone production, as previously demonstrated by the ability of exogenously administered TGF
to reverse the potentiating effects of U0126 or GF109203X [24].
These inhibitory actions of MAPK and PKC signaling on progesterone production in undifferentiated hen granulosa cells are similar to those reported in mammalian models. For instance, in immortalized rat granulosa cells transfected with a plasmid containing the rat FSHR (rFSHR-17 cells), FSH-induced STAR protein expression and progesterone production were enhanced when cells were cultured in the presence of U0126 [5]. In a human granulosa tumor cell line (KGN; previously characterized as expressing an undifferentiated phenotype), forskolin-induced STAR protein expression was enhanced by U0126, while U0126 blocked the inhibitory actions of MAPK signaling on progesterone production [4]. It is also noted that both EGF and PMA treatments are reported to reverse the phenotypic remodeling that occurs in porcine granulosa cells following FSH-induced differentiation, in vitro [31, 32]. Taken together, these results provide evidence that the MAPK and PKC signaling pathways represent a conserved mechanism by which premature granulosa cell differentiation can be inhibited.
In contrast to the differentiation-promoting effects of U0126 in undifferentiated granulosa cells from prehierarchal follicles, selective inhibition of MAPK phosphorylation in differentiated granulosa cells from preovulatory follicles attenuates gonadotropin- and 8-br-cAMP-induced progesterone synthesis (Fig. 2, A and B). Moreover, inhibition of EGFR receptor signaling using AG1478 also blocks gonadotropin- and 8-br-cAMP-induced progesterone synthesis in granulosa cells from preovulatory follicles (Fig. 4, A and B). Collectively these data indicate that not only is MAPK signaling required for progesterone production in preovulatory follicles, but also that such signaling is likely mediated through EGFR-mediated tyrosine kinase activation. Similar results have been reported in granulosa cells from rat early antral follicles, where inhibition of MAPK signaling using U0126 attenuates FSH-induced progesterone production [15, 16]. Although these findings are in conflict with those reported in undifferentiated porcine granulosa cells [32], they are consistent with the recent finding that EGFR receptor signaling is required for LH-induced progesterone production in granulosa from preovulatory follicles [33]. Specifically, LH-induced EGFR receptor activation is likely mediated by the acute up-regulation of amphiregulin, epiregulin, or betacellulin in the mouse [34, 35], and amphiregulin, epidermal growth factor, or betacellulin in the hen [26].
Unlike the stage-dependent effects on progesterone production, the stimulatory effects following inhibition of MAPK signaling on STAR protein expression and phosphorylation in preovulatory follicle granulosa cells are similar to those seen in prehierarchal follicle granulosa cells. In both undifferentiated and differentiated granulosa cells, gonadotropin-induced STAR protein expression is enhanced by inhibition of EGFR receptor tyrosine kinase or MAPK signaling (Fig. 3, A and B, and Fig. 4A) [24]. These data indicate that the inhibitory effects of MAPK signaling on the expression of STAR are maintained throughout follicle differentiation, despite the requirement for active EGFR/MAPK signaling for progesterone production in preovulatory follicles (Fig. 4, A and B). In addition to the requirement for STAR protein, P450scc enzyme activity is also indispensable for progesterone production [36]. However, treatment of hen preovulatory follicle granulosa cells with the membrane-permeable substrate, 22-OH-cholesterol, in combination with either U0126 or AG1478 plus LH does not attenuate progesterone production (data not shown), indicating that P450scc enzyme activity is unaffected by the inhibition of MAPK signaling. Though the up-regulation and enhanced phosphorylation of STAR protein in the absence of progesterone production is seemingly contradictory, recent evidence from rat Leydig cells indicates that active MAPK signaling is required to maintain mitochondrial electrochemical potential [12]. There may also be additional targets for MAPK signaling in the progesterone synthesis pathway upstream of p450scc enzyme, such as those involved in cholesterol uptake and synthesis. For example, it has been demonstrated in luteal cells from mammalian follicles that the selective cholesterol uptake receptor, scavenger receptor class B type I (SR-BI), is regulated by luteinizing hormone receptor signaling [37]. It has further been shown using in rat Leydig cells that SR-BI is up-regulated in response to treatment with EGF ligands, and that both FSH and EGF ligands up-regulate SR-BI in mouse granulosa cells [35, 38]. However, the chicken ortholog to mammalian SR-BI has yet to be reported.
On the other hand, PKC signaling is demonstrated to be obligatory for STAR protein expression (Fig. 6A) and progesterone production (Fig. 6, A and B) in differentiated granulosa cells from preovulatory follicles. While the role of PKC signaling in inhibition of granulosa cell steroidogenesis has been well documented [2, 39, 40], there have been reports that PKC signaling is essential in the processes of ovulation and subsequent luteinization. For instance, in granulosa cells from rat preovulatory follicles, it has been demonstrated that treatment with LH alone is not sufficient to induce luteinization, yet when treated in combination with the PKC activator, phorbol myristate acetate (PMA), a stable luteal phenotype, is observed. Furthermore, inhibition of PKC signaling using calphostin-C attenuates LH-induced progesterone production [41, 42]. Similarly, in Japanese quail, inhibition of PKC signaling blocks gonadotropin-induced progesterone production in preovulatory follicle granulosa cells [11]. It has recently been demonstrated in chicken granulosa cells that treatment with LH results in the activation of PKC [26]. Collectively, these data suggest that LH mediates its effects on steroidogenesis, at least in part, through activation of the PKC signaling pathway.
Finally, it is generally recognized that the 9- to 12-mm stage of development in the hen represents the earliest identifiable stage following follicle selection. For instance, during follicle development granulosa cells from 9- to 12-mm follicles demonstrate the first detectable increase in P450scc [22] and Lhr mRNA expression [21], enhanced cAMP responsiveness to FSH [23] and increased resistance to apoptosis [43]. In the present studies, granulosa cells from 9- to 12-mm follicles respond to inhibition of EGFR receptor and MAPK signaling in a fashion similar to undifferentiated granulosa cells from prehierarchal follicles. Specifically, inhibition of both MAPK and EGFR receptor signaling enhances gonadotropin-induced STAR protein expression as well as progesterone production (Figs. 1A and 3A). These data indicate that the transition from inhibitory to obligatory MAPK signaling has yet to occur at this stage of development. By comparison, the requirement of PKC signaling in these processes occurs at, or shortly following, follicle selection into the preovulatory hierarchy, as inhibition of PKC signaling in granulosa cells from the most recently selected 9- to 12-mm follicle dramatically attenuates both STAR protein expression and progesterone production (Fig. 5B). This is in marked contrast to undifferentiated granulosa cells from prehierarchal follicles, in which inhibition of PKC increases FSH-induced STAR protein expression and progesterone production (Fig. 5A).
In summary, the results presented herein demonstrate stage-of-differentiation-dependent actions for MAPK and PKC signaling in granulosa cells with respect to the regulation of STAR expression and progesterone production. In undifferentiated granulosa cells from prehierarchal follicles MAPK and PKC signaling inhibit premature differentiation prior to follicle selection. In contrast, both MAPK and PKC are obligatory for progesterone production in terminally differentiated granulosa cells from preovulatory follicles (Fig. 9). While inhibition of MAPK signaling facilitates both gonadotropin-induced STAR expression and phosphorylation in granulosa cells at each stage of follicle development, inhibition of PKC signaling selectively blocks STAR expression in actively differentiating (9- to 12-mm follicle) and differentiated (preovulatory follicle) granulosa cells following follicle selection. As progesterone synthesis can be induced, in vitro, via cAMP signaling at all stages of follicle development, we propose that the cellular cross-talk through which cAMP promotes progesterone production in granulosa cells is altered at or immediately subsequent to selection of a follicle into the preovulatory hierarchy, and that this cross-talk involves changes in MAPK and PKC signaling.
ACKNOWLEDGMENTS
We thank Dr. Steven King (Baylor College of Medicine, Houston, TX) for the gift of the rabbit anti-phosphorylated-STAR antibody, and Dr. D.B. Hales (University of Illinois, Chicago, IL) for the anti-STAR serum.
FOOTNOTES
1Supported by National Science Foundation grant IOB-0445949 (to A.L.J.). ![]()
Correspondence: 2A.L. Johnson, Department of Biological Sciences, P.O. Box 369, The University of Notre Dame, Notre Dame, IN 46556. FAX: 574 631 7413; e-mail: johnson.128{at}nd.edu
Received: 7 December 2006.
First decision: 2 January 2007.
Accepted: 29 March 2007.
REFERENCES

Ct method. Methods 2001; 25:402–408[CrossRef][Medline]This article has been cited by other articles:
![]() |
A. L. Johnson, M. J. Haugen, and D. C. Woods Role for Inhibitor of Differentiation/Deoxyribonucleic Acid-Binding (Id) Proteins in Granulosa Cell Differentiation Endocrinology, June 1, 2008; 149(6): 3187 - 3195. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |