Non-canonical Progesterone Signaling in Granulosa Cell Function

Abstract

It has been known for over three decades that progesterone (P4) suppresses follicle growth. It has been assumed that P4 acts directly on granulosa cells of developing follicles to slow their development, since P4 inhibits both mitosis and apoptosis of cultured granulosa cells. However, granulosa cells of developing follicles of mice, rats, monkeys and humans do not express the A or B form of the classic nuclear receptor for progesterone (PGR). In contrast, these granulosa cells express other progesterone binding proteins, one of which is referred to as Progesterone Receptor Membrane Component 1 (PGRMC1). PGRMC1 specifically binds P4 with high affinity and mediates P4’s anti-mitotic and anti-apoptotic action as evidenced by the lack of these P4-dependent effects in PGRMC1-depleted cells. In addition, mice in which PGRMC1 is conditionally depleted in granulosa cells show diminished follicle development. While the mechanism through which P4 activation of PGRMC1 affects granulosa cell function is not well defined, it appears that PGRMC1 controls granulosa cell function in part by regulating gene expression in T cell specific transcription factor/lymphoid enhancer factor (Tcf/Lef)-dependent manner. Clinically, altered PGRMC1 expression has been correlated with premature ovarian failure/insufficiency, polycystic ovarian syndrome and infertility. These collective studies provide strong evidence that PGRMC1 functions as a receptor for P4 in granulosa cells and that altered expression results in compromised reproductive capacity. Ongoing studies seek to define the components of the signal transduction cascade through which P4-activation of PGRMC1 results in the regulation of granulosa cell function.

Introduction and Historical Perspective

P4 is synthesized and secreted from both the follicular and luteal components of the mammalian ovary (Monniaux, et al. 1997). The corpora lutea secrete P4 at the highest rate while P4 synthesis gradually increases as the follicles develop (Roy and Greenwald 1987, 1996). Throughout the course of rat ovarian follicle development the P4 concentration within follicular fluid gradually increases from 5 μM on diestrus to 55 μM on the evening of proestrus (Fujii, et al. 1983). These elevated P4 levels play an essential role in regulating ovulation because ovulation is disrupted by treatment with progesterone receptor antagonists, such as RU486, or in mutant mice in which the progesterone receptor (PGR) is eliminated (Conneely, et al. 2002, Curry and Nothnick 1996, Lydon, et al. 1996, Pall, et al. 2000, Robker, et al. 2000, Svensson, et al. 2000). Because these P4-dependent effects are mediated to a large extent by PGR, which is expressed during the periovulatory period and not during folliculogenesis, this aspect of P4’s actions will not be emphasized in this review.

While P4 activation of PGR and its role in the periovulatory period is well accepted, P4 also influences granulosa cell function in developing follicles prior to ovulation. These actions are controversial, because granulosa cells of developing follicles do not express PGR (Natraj and Richards 1993, Park and Mayo 1991, Shao, et al. 2003). This conclusion is based on the failure to detect PGR expression in rodent granulosa cells by Northern blot, Western blot, RT-PCR, in situ hybridization and immunocytochemistry. Also, rat luteal cells do not express PGR (Park and Mayo 1991). Thus, there is overwhelming evidence that rat and mouse granulosa and luteal cells do not express PGR. In addition granulosa cells of monkeys do not express the PGR until the gonadotropin surge (Chandrasekher, et al. 1991). Finally, immunohistochemical analysis fails to detect PGR in granulosa cells of human follicles, although PGR is readily detected in luteal cells (Suzuki, et al. 1994).

In spite of the fact that granulosa cells of developing follicles do not express PGR, the rate at which follicles grow (i.e. rate of granulosa cell mitosis) is inversely associated with serum P4 levels. (Butcher, et al. 1974, Hirshfield 1984, Pedersen 1984). Further, P4 inhibits follicular development in hypophysectomized hamsters (Moore and Greenwald 1974), gonadotropin-primed hamsters (Kim and Greenwald 1987) and rabbits (Setty and Mills 1987), as well as cycling rats (Buffler and Roser 1974). That intraovarian P4 regulates follicle development is clearly shown in monkeys, in which intraovarian insertion of a P4 implant into one ovary suppresses follicle development in that ovary without altering serum gonadotropin levels, which allowed follicles to continue to grow in the contralateral non-P4 implanted control ovary (diZerega and Hodgen 1982). This is the best evidence that P4 acts within the ovary to regulate follicle growth independent of its effects on serum gonadotropin levels.

Although these in vivo studies are consistent with a direct effect of P4 on developing follicles, they do not reveal the cellular site of P4’s action. However, P4 affects the ability of granulosa cells of immature rats to secrete steroids in vitro (Fortune and Vincent 1983, Schreiber, et al. 1980, 1981), as well as to undergo mitosis and apoptosis in vitro (Luciano and Peluso 1995, Peluso and Pappalardo 1998). Finally, P4 inhibits mitosis (Peluso, et al. 2002) and apoptosis (Peluso, et al. 2001) of Spontaneously Immortalized Granulosa Cells (SIGCs), a cell line derived from rat granulosa cells (Stein, et al. 1991) and, like freshly isolated primary rat granulosa cells, these cells do not express PGR (Peluso, et al. 2002). These in vitro observations are difficult to reconcile with the findings that granulosa cells do not express PGR.

P4 not only influences the function of granulosa cells but also luteal cells, which are derived from granulosa cells. Specifically, inhibiting P4 synthesis results in structural changes similar to those associated with luteal regression (See review by (Stouffer, et al. 2013). These changes are prevented by supplemental P4 treatment (Stouffer, et al. 2013). However, the effect of RU486 treatment is variable; at times increasing while at other times decreasing P4 secretion. To explain these biological effects of P4 and RU486, Irving Rothchild (Rothchild 1996) put forth the concept that P4 regulates the structure and steroidogenic capacity of luteal cells. His hypothesis, developed without the aid of molecular biological and genetic approaches, further states that P4 can act through at least two different pathways but RU486 only acts through one of these pathways. While PGR is expressed by primate luteal cells (Chandrasekher, et al. 1991), PGR is not expressed rodent luteal cells (Cai and Stocco 2005, Natraj and Richards 1993, Park and Mayo 1991). These observations raise the possibility of other receptors within luteal cells that mediate some of P4’s actions.

Identification of a Novel Progesterone Binding Protein in Granulosa and Luteal Cells

How then might P4 affect granulosa and luteal cell function in the absence of PGR? Ligand binding studies done in the late 1970s (Schreiber and Erickson 1979, Schreiber and Hsueh 1979) and early 1980s (Naess 1981) reveal that immature rat ovaries, which are composed mainly of developing ovarian follicles that lack PGR, specifically bind P4. Rat luteal cells also bind P4 (Cai and Stocco 2005). Therefore, these binding studies indicate that the granulosa cells of immature rat ovaries and rat luteal cells express a P4 binding protein that is not the PGR and may function as a mediator of P4’s actions.

Interestingly, there are two different families of progestin binding proteins that could mediate P4’s actions in these cells. The first family is Progestin AdipoQ Receptors (PAQRs), initially identified in fish oocytes by Dr. Peter Thomas’s group (Zhu, et al. 2003a, Zhu, et al. 2003b). Thomas’ studies reveal that the alpha type of the membrane progestin receptor (PAQR7) is expressed in human testis and ovaries. Moreover, mRNA encoding PAQR7 is detected in human granulosa/luteal cells isolated from women undergoing in vitro fertilization, a human granulosa cell line, hGL5 cells (Peluso, et al. 2009) and rat luteal cells (Cai and Stocco 2005). Thus, a functional role for PAQRs in regulating P4’s actions within the mammalian ovary is likely.

Expression and Localization of Progesterone Membrane Receptor Component 1

In addition to PAQRs, a second purported mediator of P4’s action is Progesterone Receptor Membrane Component 1 (PGRMC1). PGRMC1 is highly expressed in granulosa cells of developing follicles and luteal cells. Specifically, PGRMC1 is detected in virtually all granulosa cells of rat antral follicles (Peluso, et al. 2006). While present at the plasma membrane of every granulosa cell, it is also detected in the nucleus of a limited number of granulosa cells of small antral follicles (Figure 1A). Within 48 h of gonadotropin treatment, PGRMC1 localization changes to almost exclusively at or near select regions of the plasma membrane with few cells showing nuclear staining (Figure 1B). After hCG-induction of ovulation and luteinization, PGRMC1 expression increases with 100% of the luteal cells expressing high levels of PGRMC1 localized diffusely throughout the cytoplasm and nucleus (Peluso, et al. 2006).

Figure 1.

The expression of PGRMC1 in granulosa cells of antral (A) and preovulatory follicles (B) as assessed by immunohistochemistry. The antral follicle was obtained from an immature rat, while the preovulatory follicle was from an immature rat 48 h after PMSG injection. PGRMC1 was detected as a brown stain. The thecal layers at the base of each image show robust PGRMC1 expression. Data from Peluso et al (Peluso, et al. 2006).

It is important to appreciate that in the cytoplasm, PGRMC1 is mainly thought to localize to the endoplasmic reticulum (Falkenstein, et al. 1999, Losel, et al. 2005, Meyer, et al. 1996). Since steroids can enter the cell, it is not necessary for PGRMC1 to be at the extracellular surface of the plasma membrane to transduce P4’s action. However, BSA-conjugated P4 mimics P4’s actions, suggesting that at least a fraction of P4’s receptor is localized to the extracellular surface of the plasma membrane (Peluso and Pappalardo 1999). That some PGRMC1 localizes to the extracellular surface of the plasma membrane is conclusively demonstrated because PGRMC1 is detected among the plasma membrane proteins after biotinylation of these proteins (Peluso, et al. 2006).

PGRMC1 as a P4 Binding Protein

The expression studies imply that PGRMC1 could mediate P4’s actions in granulosa cells of developing follicles. If so, then the first event in PGRMC1’s mechanism is to bind P4. Interestingly, there are reports that PGRMC1 does not bind P4. These reports used bacterially-expressed PGRMC1 fusion proteins (reviewed in (Cahill 2007)) and not PGRMC1 fusion proteins expressed in mammalian cells. As a result, the bacterially expressed PGRMC1 may be unable to bind P4, since it may not be properly folded. In contrast, partially purified PGRMC1-fusion protein isolated from either SIGCs (Peluso, et al. 2008) or human granulosa/luteal cells (hGL5 cells) (Peluso, et al. 2009) specifically binds P4 with high affinity (EC₅₀ ≈ 10 nM) (Figure 2A). In addition the synthetic progestin, R5020, also binds PGRMC1 with the same affinity as P4 (Figure 2A). Further, siRNA treatment specifically depletes Pgrmc1 mRNA and proteins levels and reduces the capacity of SIGCs to bind P4 (Peluso, et al. 2008) (Figure 2B). Collectively, these studies demonstrate that PGRMC1 binds P4 and that is required for P4 binding protein in SIGCs. However as it name implies, PGRMC1 likely binds P4 as part of a complex with one member of the complex being membrane progestin receptor alpha (PAQR7)(Thomas, et al. 2014), which is also expressed by SIGCs, rat ovarian cells (Cai and Stocco 2005) and human granulosa/luteal cells (Peluso, et al. 2009). These ligand-binding studies together with the expression data, are supportive of PGRMC1 being a mediator of P4’s action in both granulosa and luteal cells. However, this must be demonstrated by genetic manipulation of PGRMC1 levels.

Figure 2.

The capacity of partially purified PGRMC1-GFP to bind P4 and R5020 (A) and the effect of PGRMC1 siRNA treatment on Pgrmc1 mRNA levels and specific ³H-P4 binding to SIGCs (B). Data in panel A taken from Peluso et al (Peluso, et al. 2009) and the data shown in panel B are unpublished observations (J Peluso, unpublished observations) that confirm our published data (Peluso, et al. 2008). The effect of progesterone (P4) on serum-induced SIGC mitosis is shown in panel C. Data from panel C from Peluso et al (Peluso 2013).

Biological Actions Mediated by P4-PGRMC1 Signaling in Granulosa Cells

As previously indicated, P4 effects granulosa cell mitosis, apoptosis and steroid synthesis in cells that do not express PGR. Given these actions, it is possible that PGRMC1 is involved in each of these diverse aspects of granulosa cell biology. This concept was tested using both SIGCs and/or hGL5 cells as outlined in the following paragraphs.

P4 and PGRMC1 as Regulators of Mitosis

P4 attenuates mitogen-induced proliferation of rat granulosa cells isolated from both immature and preovulatory rat follicles (Peluso, et al. 2006), human granulosa/luteal cells obtained from women undergoing ovulation induction for infertility treatment (Chaffkin, et al. 1992) and SIGCs (Peluso, et al. 2002). Further, P4 does so in a dose-dependent manner (10-1000 nM). Furthermore, treatment with PGRMC1 siRNA attenuates P4’s ability to suppress the percentage of SIGCs incorporating BrdU and the percentage of cells in metaphase (J Peluso, unpublished observations). Finally, PGRMC1 siRNA treatment ablates P4’s ability to suppress the number of cells present after 22 h of culture (Figure 2C) (Peluso 2013), while forced expression of PGRMC1 blocks entry into the cell cycle (J Peluso, unpublished observation). Taken together, these studies support the concept that P4-PGRMC1 signaling is involved in regulating the rate of granulosa cell proliferation.

PGRMC1’s ability to influence cell cycle progression is complex in that it appears to play specific roles at different stages of the cell cycle. For example, PGRMC1 regulates the transition from G_o to G₁ stage of the cell cycle and also prolongs the duration of metaphase through its ability to interact with the mitotic spindle (Lodde and Peluso 2011). These observations imply that PGRMC1 has different and specific modes of action that allow for its involvement in regulating the diverse signaling pathway that control different stages of the cell cycle.

P4-PGRMC1 Regulates Apoptosis

Over the same dose range that inhibits mitosis, P4 also suppresses the rate at which rat granulosa cells (Peluso, et al. 2005), rat luteal cells (Peluso, et al. 2005), human granulosa/luteal cells (Engmann, et al. 2006) and SIGCs (Peluso, et al. 2004) undergo apoptosis (Figure 3A). It is important to appreciate that P4’s anti-apoptotic effects can only be detected in mature rat luteal cells if endogenous P4 synthesis is inhibited with aminoglutethamide cells (Peluso, et al. 2005). This is because the effect of supplemental P4 treatment are not observed in the presence of the large amount of P4 that is secreted by luteal cells (Peluso, et al. 2006). As with mitosis, siRNA knockdown of PGMRC1 prevents P4 from inhibiting SIGCs from undergoing apoptosis (Peluso, et al. 2008) (Figure 3B). In addition, forced expression of PGRMC1 increases the ability of P4 to suppress apoptosis (Peluso, et al. 2008). These studies implicate P4-PGRMC1 signaling as a regulator of apoptosis.

Figure 3.

The inhibitory effect of progesterone (P4) on apoptosis of human granulosa/luteal cells, rat granulosa cells and rat luteal cells from PMSG-hCG primed immature rats 4 days after hCG treatment (A). Note that to detect the effects of P4 in luteal cells, endogenous P4 secretion was blocked with aminoglutethamide. The effect of P4 on the rate of spontaneously immortalized granulosa cells (SIGCs) apoptosis is shown in B. The anti-apoptotic action of P4 is ablated after PGRMC1 siRNA treatment. The * indicates a value that is significantly different from control (p < 0.05). In panel A all values under the bracket marked with an * are different from their respective control values. Panel A from data redrawn from the following publications: (Engmann, et al. 2006, Peluso, et al. 2009, Peluso, et al. 2005). Data for panel B is from Peluso et al (Peluso, et al. 2008).

PGRMC1 and Steroidogenesis

P4 can also enhance its own synthesis (Rothchild 1996) and there is evidence to suggest that this action is mediated by PGRMC1. Specifically, this concept is based on the observation that PGRMC1 interacts with proteins Sterol regulatory element binding protein Cleavage-Activating Protein (SCAP) and Insulin Induced Gene 1 (INSIG1) that regulate cholesterol metabolism (Suchanek, et al. 2005). Briefly, SCAP and INSIG1 control cholesterol metabolism by regulating the transcriptional action of Sterol Regulatory Element Binding Protein, which induces the expression of Steroidogenic Acute Regulatory Protein (StAR) (Yang, et al. 2002). Progesterone induces the expression of StAR in MA-10 testicular interstitial cells (Schwarzenbach, et al. 2003) and promotes cholesterol and P4 synthesis in rat and human granulosa/luteal cells (Rung, et al. 2005). Thus, the possibility exists that P4-PGRMC1 interaction regulates a complex series of protein-protein interactions that promotes cholesterol and P4 biosynthesis. This concept is consistent with the observations that an antibody to PGRMC1 attenuates and forced PGRMC1 expression enhances adrenal steroidogenesis (Min, et al. 2004).

The synthetic progestin, R5020, stimulates P4 secretion from cultured rat granulosa cells (Fanjul, et al. 1983, Ruiz de Galarreta, et al. 1985), luteal cells (Rothchild 1996), MA-10 cells (Schwarzenbach, et al. 2003) and hGL5 cells (Peluso, et al. 2009). R5020’s ability to stimulate P4 secretion is observed at concentrations equal to or greater than 16 μM. While 16 μM may appear to be a pharmacological dose, it is within the physiological range, since women undergoing either natural or gonadotropin-stimulated cycles have P4 follicular fluid levels between 15 to 60 μM (Enien, et al. 1995, Kamel, et al. 1994, Tarlatzis, et al. 1993). However, human PGRMC1 binds R5020 with high affinity (EC₅₀ of ≈ 10 nM; Figure 2A; (Peluso, et al. 2009). Given this binding characteristic, all of the R5020 binding sites within human PGRMC1 would be occupied by R5020 in the nM range. Therefore, it is unlikely that the ability of R5020 to acutely promote P4 secretion is mediated through its ability to interact with PGRMC1. Further, PGRMC1-deplete hGL5 cells respond to R5020 by increasing their rate of P4 secretion (Peluso, et al. 2009). Thus, these pharmacological and genetic-based observations make it clear that in hGL5 cells, PGRMC1 does not mediate P4’s ability to acutely stimulate its own secretion. The mechanism through which R5020 promotes steroidogenesis in hGL5 cells is not known but could involve one of the other progestin binding proteins that are expressed by both human granulosa cells and hGL5 cells (i.e. PGR and PAQR family members) (Peluso, et al. 2009).

PGRMC1 and Follicle Growth In Vivo

PGRMC1’s role in regulating granulosa cell mitosis and apoptosis is supported by examining follicle development in a conditional knockout mouse in which PGRMC1 is conditionally deleted from granulosa cells using anti-Mullerian hormone type II receptor cre recombinase transgenic mouse (Amhr2-cre). This analysis shows that ovaries from immature (22-25 day old) Pgrmc1 conditional knockout (cKO; Amhr2^cre/+;Pgrmc1^fl/fl) mice have fewer antral follicles compared to either the control (i.e. Amhr2^+/+;Pgrmc1^fl/fl) or heterozygous (Amhr2^cre/+;Pgrmc1^fl/+) female mice (Figure 4). Although the ovaries of heterozygous mice have the same number of antral follicles as control mice, they have a higher percentage of atretic antral follicles (Figure 4). The reason for this is unclear but a reduced level of PGRMC1 likely results in an increase in granulosa cell apoptosis, which would account for the increase in atretic follicles in the heterozygous mice. Importantly, the heterozygous mouse mimics women with premature ovarian failure, who have ≈ 50% reduction in the level of PGRMC1 (Mansouri, et al. 2008). Taken together, these findings are consistent with PGRMC1’s role in maintaining the viability of granulosa cells and thus the process of folliculogenesis.

Figure 4.

The total number of antral follicles within the ovaries of controls (+/+), PGRMC1 heterozygous (+/−) and PGRMC1 conditional knockout (−/−) mice. Values are shown as means and SE with n of 6-8/treatment group. * indicates a value is different from control (p < 0.05) as demonstrated by an ANOVA followed by Dunnett’s Multiple Comparison Test. The percentage of atretic antral follicles is shown in the base of each column. The percentage of atretic antral follicles is greater in the ovaries of PGRMC1 +/− mice compared to the percentage in either the controls or PGRMC1 −/− mice as assessed by Fisher Exact/Chi-Square tests (p< 0.05). (Peluso 2013).

P4-PGRMC1 Signal Transduction Cascade

That PGRMC1 mediates P4’s ability to regulate mitosis and apoptosis is clearly illustrated through the use of siRNA-based knockdown and forced expression experiments. Unfortunately, the mechanism through which P4 activation of PGRMC1 regulates both mitosis and apoptosis of granulosa still remains to be determined. Specifically, how does P4 binding to PGRMC1 either at the plasma membrane or within the cytoplasm activate a signal transduction cascade that affects the rate at which granulosa cells undergo mitosis and/or apoptosis? There are a few studies indicating that P4 increases protein kinase G (PKG) activity (Peluso, et al. 2007, Peluso and Pappalardo 2004) and suppresses MEK/ERK activity (Peluso, et al. 2003). Since activators of PKG inhibit and MEK/ERK activity induces apoptosis (Peluso, et al. 2003), the ability of P4 to regulate these two kinase pathways is probably a component of the anti-apoptotic P4-PGRMC1 signaling pathway. Unfortunately, siRNA-based studies have not been conducted, and therefore it remains to be determined whether or not P4 activation of PGRMC1 is directly linked to P4’s effects on these kinase cascades.

Expression of Genes in Ovarian Cells Regulated by PGRMC1

What is clear is that P4 activation of PGRMC1 alters gene expression, specifically genes that are involved in regulating mitosis and/or apoptosis. For example, P4 promotes SIGC survival by suppressing Bad and increasing Bcl2a1d expression, thereby changing the Bcl21d/Bad ratio to favor cell survival (Peluso, et al. 2010). In human granulosa cells (hGL5 cells), PGRMC1 suppresses mRNA levels that encode activators of apoptosis such as caspase-3 (Peluso, et al. 2012a). These PGRMC1-dependent actions would make the cells less likely to undergo apoptosis and promote mitosis (Peluso, et al. 2012a).

P4-PGRMC1 as a Regulator of Tcf/Lef Activity

While PGRMC1 knockdown studies are important in that they reveal a role for P4-PGRMC1 signaling in the regulation of gene expression, these types of studies provide little information about the mechanism through which P4-activated PGRMC1 alters gene expression. As indicated, PGRMC1 localizes to the nuclei of many granulosa cells in developing ovarian follicles. Similarly, PGRMC1 is also present within the nuclei of SIGCs that are proliferating but not in the nuclei of mitotically arrested SIGCs (Peluso, et al. 2012b). The mechanism that regulates the cellular localization is unknown but PGRMC1’s nuclear localization suggests that nuclear PGRMC1 may be directly involved in regulating gene expression. Furthermore, the most likely mechanism through which PGRMC1 influences gene expression is by either increasing or decreasing the activity of transcription factors. Recent studies indicate that in SIGCs P4 significantly and faithfully suppresses Tcf/Lef activity and P4’s ability to suppress Tcf/Lef activity is dependent on the presence of PGRMC1 (Figure 5A) (Peluso, et al. 2012b). These observations provide some insight into how P4-PGRMC1 signaling suppresses mitosis, because the promoter region of genes involved in the initiation of mitosis, such as c-myc, possess Tcf/Lef consensus binding elements. (Dang and Lewis 1997, Dang, et al. 2006).

Figure 5.

The effect of P4 and PGRMC1 on Tcf/Lef promoter in SIGCs as assessed by a luciferase reporter assay (A). In panel B, the ability of P4 to suppress Tcf/Lef activity in the presence or absence of PGRMC1-Flag protein was assessed by a luciferase reporter. In all graphs, * indicates a value that is significantly different than control (p < 0.05), ** indicates a value significantly less that cells transfected with an expression vector encoding PGRMC1-Flag (Peluso, et al. 2012b).

The PGRMC1 siRNA study illustrates that P4’s ability to inhibit Tcf/Lef activity is dependent on PGRMC1, but it does not demonstrate that nuclear PGRMC1 interacts with the Tcf/Lef site. One way to demonstrate that nuclear PGRMC1 interacts with the Tcf/Lef site is to use electrophoretic mobility shift assay (EMSA). Nuclear extract from SIGCs not treated with P4 contains proteins that bind to the Tcf/Lef probe. Moreover, nuclear extract from P4-treated SIGCs reduces the amount of Tcf/Lef probe bound to nuclear protein by ≈ 50% (J Peluso, unpublished observation). Unfortunately, the PGRMC1 antibody is not suitable for use in a super-shift EMSA, thus PGRMC1 could not be shown to be one of the proteins that binds Tcf/Lef response elements.

This obstacle can be overcome by transfecting cells with an expression construct that encodes a PGRMC1-Flag fusion protein. Importantly, the presence of PGRMC1-Flag increases Tcf/Lef activity by nearly 2-fold compared to controls and the PGRMC1-Flag induced increase in Tcf/Lef activity is suppressed by P4 (Figure 5B). After transfection with PGRMC1-Flag construct, a super-shift EMSA using an anti-Flag antibody reveals that PGRMC1-Flag is among the proteins that bind to the Tcf/Lef response element and P4 reduces the amount of PGRMC1 that is bound to this transcription factor site. Taken together, these studies demonstrate that P4 inhibits Tcf/Lef activity in part by suppressing the ability of nuclear PGRMC1 to interact with this transcription factor response element (Peluso, et al. 2012b).

Expression and Function of Other PGRMC Family Members

Although progress is being made to define PGRMC1’s role and mechanism of action in ovarian function, there are other members of the PGRMC family, including PGRMC2 and neudesin (NENF). As can be seen in Figure 6, granulosa cells also express these other two PGRMC family members. Among these PGRMC family members, the relative level of expression is considerably different with PGRMC1 having the lowest level of expression, although Pgrmc1 mRNA levels in granulosa cells are higher compared to their levels in whole ovary. The higher Pgrmc1 mRNA levels in granulosa cells is consistent with its known function in granulosa cells. Interestingly, PGRMC2 is expressed at a 5 to 10 fold greater level than PGRMC1, while NENF is expressed at nearly 100 times that of PGRMC1. This pattern of PGRMC family expression is also observed in mRNA levels obtained from the entire immature mouse ovary, as well as SIGCs (J Pru & J Peluso unpublished observation).

Figure 6.

Expression of PGRMC family members in the immature mouse ovary and granulosa cells as assessed by real-time PCR. Note that the level of expression for the PGRMC family member is dramatically different with the lowest expression for PGRMC1, followed by PGRMC2 and then NENF, which has the highest level of expression (J Pru, unpublished observations).

Given the structural similarity between PGRMCs (Kimura, et al. 2013), it is likely that PGRMC2 influences granulosa cell function. There are several ways that PGRMC2 could influence granulosa cell function including it 1) functioning as an antagonist to PGRMC1, 2) synergizing with PGRMC1 or 3) regulating a redundant pathway. Each of these possibilities merits experimental investigation. The likely key to understanding PGRMC2’s function will be to determine the proteins to which it binds. It is probable that PGRMC2 will bind a unique set of intracellular proteins as well as some of the same proteins that bind PGRMC1. To date, the only published study on the function of PGRMC2 involves its role in promoting the migrations of ovarian cancer cells (Wendler and Wehling 2013).

Like PGRMC2, NENF is highly expressed in granulosa cells and SIGCs and shares sequence homology to PGRMC1 (Kimura, et al. 2013). Since NENF is secreted by neural cells (Kimura, et al. 2008, Kimura, et al. 2005), it is hypothesized that NENF binds P4 thereby reducing the amount of P4 available to interact with PGRMC1. However, there are reports indicating the NENF does not bind P4, but in these studies P4 binding was assessed using dot blots and bacterially-expressed NENF (Kimura, et al. 2008, Kimura, et al. 2005)]. Given that this approach is relatively insensitive and it is unclear if NENF bacterial protein is folded to allow for P4 binding, the capacity of NENF to bind P4 still remains to be determined.

There is another way for NENF to affect granulosa cell function. In SIGCs, P4 suppresses MAP kinase activity (Peluso, et al. 2003), which likely accounts in part for both P4’s anti-mitotic and anti-apoptotic actions (Peluso, et al. 2003). Interestingly, NENF has neurotrophic actions and these actions are mediated through the activation of MAP kinase (Kimura, et al. 2008, Kimura, et al. 2005). Thus, the balance between P4 and NENF could influence the overall activity of MAP kinase and thereby the fate of granulosa cells. However, a receptor for NENF has not been identified (Kimura, et al. 2013) and this is required before any mechanism for NENF’s action can be completely developed.

Clinical Correlations

Although it has only been known that PGRMC1 is expressed in the mammalian granulosa cells since 2006 (Peluso, et al. 2006), there are now three publications suggesting that it plays a clinically relevant role in regulating follicular function in women. The first report is from Dr. Dahl’s group in Sweden in which he identified a mother and daughter with an X; autosome translocation (Mansouri, et al. 2008). This translocation results in a reduced expression of PGRMC1 by 50%. The same publication describes a missense mutation in PGRMC1 that is observed in 1 of 67 women with premature ovarian failure/insufficiency. This mutation at amino acid 165 reduces the ability of PGRMC1 to transduce P4’s anti-apoptotic action by 50% and also attenuates its ability to interact with several cytochrome P450 enzymes (Mansouri, et al. 2008). In their second publication Dahl’s group reports that PGRMC1 levels are relatively constant throughout the menstrual cycle but are lower in postmenopausal women and women with premature ovarian failure and polycystic ovary syndrome (Schuster, et al. 2010). Finally, an examination of the granulosa/luteal cells obtained from women undergoing ovulation induction as part of their infertility treatment reveal that women with elevated levels of Pgrmc1 mRNA levels have 30% fewer follicles and few oocytes retrieved (Elassar, et al. 2012). The reason for the association between higher levels of Pgrmc1 mRNA is not completely known but may be related to the finding that there is an increase in the methylation of the PGRMC1 promoter (Elassar, et al. 2012).

Like PGRMC1, PGRMC2 appears to be associated with altered follicle development in women. This is based on the finding that young women with diminished ovarian reserve (i.e. fewer ovarian follicles) have elevated Pgrmc2 mRNA levels compared to normal young women (Skiadas, et al. 2012). To date there are no studies on ovarian function and NENF. Taken together these correlative clinical studies provide new insight by demonstrating that genetic changes or altered expression of PGRMC family members affects follicle growth in women. These findings are consistent with actions of PGRMC family members observed studies on rodent granulosa cells both in vitro and in vivo.

Summary and Future Research Perspectives

P4’s biological actions on developing ovarian follicles have been known for several decades, but the receptors that mediate these effects remain a point of contention, since the granulosa cells of these follicles do not express PGR. The application of biochemical and genetic approaches reveal that PGRMC1 and possibly other PGRMC family members play important roles in this aspect of ovarian follicle function. Recent studies implicate PGRMC1 as a regulator of granulosa cell apoptosis and mitosis. Moreover, PGRMC1 functions to regulate gene transcription, in part through regulating Tcf/Lef transcription factor activity. Thus, the basic elements of the P4-PGRMC1 signal transduction pathway have been identified in granulosa cells. Future research must now be directed toward determining the molecular steps that link P4-activated PGRMC1 to Tcf/Lef-dependent transcription and the subsequent regulation of mitosis, apoptosis and steroidogenesis.