Mechanisms of uterine estrogen signaling during early pregnancy in mice: an update

Abstract

Adherence of an embryo to the uterus represents the most critical step of the reproductive process. Implantation is a synchronized event between the blastocyst and uterine luminal epithelium leading to structural and functional changes for further embryonic growth and development. The milieu comprising the complex process of implantation is mediated by estrogen through diverse but interdependent signaling pathways. Mouse models have demonstrated the relevance of the expression of estrogen modulated paracrine factors to uterine receptivity and implantation window. More importantly, some factors seem to serve as molecular links between different estrogen pathways promoting cell growth, acting as molecular chaperones or amplifying estrogenic effects. Abnormal expression of these factors can lead to implantation failure and infertility. This review provides an overview of several well characterized signaling pathways that elucidates molecular cross-talk involved in the uterus during early pregnancy.

Introduction

Reproduction is a fundamental aspect of life. The World Health Organization (WHO) has recognized infertility, or the inability to reproduce, a worldwide health concern with a lifetime prevalence ranging from 6.6% to 26.4% (Boivin, et al. 2007). Although much advancement has been made using assisted reproductive technologies (ART) to achieve higher pregnancy rates by improving the selection of high-quality embryos, the implantation process is still very illusive.

The development of the preimplantation embryo and the differentiation of the uterus are distinct processes occurring simultaneously in early gestation and must be synchronized in order for successful implantation (Paria, et al. 1993; Psychoyos 1973a). It has been shown that in a mouse, implantation occurs when the developed blastocyst attaches to the luminal epithelium of the uterine endometrium on the evening of day 4 of pregnancy (Das, et al. 1994; Enders and Schlafke 1969). This attachment of the embryo to the epithelial lining promotes the disappearance of epithelium via a mechanism through entosis (cell-eat-cell) by the trophoblast cells followed by apoptosis at the site of implantation (Li, et al. 2015; Parr, et al. 1987) and subsequent stimulation of stromal cell proliferation and differentiation into secretory decidual cells, therefore forming decidualization at the blastocyst site (Huet-Hudson, et al. 1989).

These structural and functional changes occurring in the uterus promote receptivity to the invading blastocyst. This receptivity phase of the uterus is short-lived and primarily mediated by estrogen and progesterone (Paria et al. 1993; Psychoyos 1973a). The estrogenic effects in the mouse uterus are biphasic: early (phase I) responses occur within 6 hours and are characterized by water inhibition, macromolecular uptake and alteration in genes involved in vascular permeability. Late (Phase II) responses occur between 18–30 hours and are characterized by increased epithelial cell proliferation (Huet-Hudson et al. 1989). The presence of progesterone (P4) is inhibitory to estrogen-mediated epithelial proliferation, which can be detected on day 4 (D4) of gestation (Das and Martin 1973; Li, et al. 2011; Martin, et al. 1973; Pan, et al. 2006). Ovariectomized mice on the morning of D4 before preimplantation estrogen secretion exhibit delayed implantation due to blastocyst dormancy (Yoshinaga and Adams 1966). When the uterus in ovariectomized mice is exposed to progesterone alone it renders it a neutral or pre-receptive endometrium; however, receptivity for implantation is observed when exposed to estrogen (Paria et al. 1993). This demonstrates the crucial role of estrogen in the process of implantation.

The mechanisms by which estrogen transforms a progesterone-primed uterus to the receptive state, activates blastocysts and initiates implantation are not clearly delineated. The classical estrogen-signaling pathway is through nuclear estrogen receptors ERα (ESR1) and ERβ (ESR2), which act as ligand-inducible transcription factors (Beato, et al. 1995; Tsai and O’Malley 1994). However, there is increasing evidence that gene activation and cell function modulation are initiated by estrogen through a nuclear ER-independent manner. Studies with ERα null mice and also in those with wild-type mice in which both ERα and ERβ antagonist ICI 182,780 was utilized to silenced ligand-dependent ER functions, have demonstrated estrogen mediated gene expression suggesting an alternate signaling pathway (Das, et al. 2000; Das, et al. 1997; Hou, et al. 2004).

Implantation failure and infertility is associated with aberrations in molecular pathways. The knowledge attained with the development of knockout mouse models and conditional gene deletions has advanced uterine biology immensely. This is a review of the knowledge gained from previous studies in mice attempting to delineate the mechanisms of estrogen signaling. Understanding the estrogen pathways and its mediated events during early pregnancy is critical to further advancement in ART protocols that will improve treatment of this worldwide health condition.

Role of estrogen receptors during early pregnancy

Estrogen plays a pivotal role in the observed changes of the uterus during early pregnancy. In mice, during the first two days of gestation, pre-ovulatory estrogen stimulates proliferation of the luminal and glandular epithelial cells (phase I estrogen secretion). Once the corpora lutea is formed on day 3 of gestation, progesterone secretion stimulates stromal cell proliferation, which becomes further potentiated by pre-implantation estrogen (phase II estrogen secretion) on day 4, the day of implantation (Huet-Hudson et al. 1989). This second wave of estrogen prior to implantation ceases epithelial cell proliferation and allows for differentiation to occur (Tan, et al. 1999). During the remodeling of the uterine epithelium, the epithelial cells lose polarity through down-regulation of the cell-to-cell adhesion molecules E-cadherin (Daikoku, et al. 2011; Li et al. 2015). Epithelial cells also acquire inhibition of the glycoprotein mucin 1 (MUC1) and develop protrusions along the apical surface (DeSouza, et al. 1998; Surveyor, et al. 1995). Increased endometrial capillary permeability at the location of the blastocyst is also exhibited lending to implantation and subsequent decidualization of stromal cells (Matsumoto, et al. 2002a; Psychoyos 1973b).

The classic physiologic actions of estrogen on its target organ are mediated by its binding to ER activates the receptor by promoting dimerization and then translocation to the nucleus to bind its responsive element in the DNA (Kumar and Chambon 1988). The distribution and expression of ER subtypes varies due to their tissue-specific physiologic functions in various organ systems. ERα, for example, is mainly present in mammary gland tissue, uterus, thecal cells of the ovary, bone, liver, adipose tissue, testes and epididymis of the male reproductive organs and the stroma of the prostate. ERβ is mainly found in the epithelium of the prostate, bladder, granulosa cells of the ovary, colon adipose tissue and the immune system (Dahlman-Wright, et al. 2006; Heldring, et al. 2007). Although ERα is the predominant isoform in certain tissues, both receptors have high affinity to estradiol-17β (E2) in the same estrogen response element (ERE) and they share approximately 95% and 55% homology in the DNA-binding domain and the hormone-binding domain, respectively (Kuiper, et al. 1997; Tremblay, et al. 1997). However, it has been demonstrated that the biological disruption of ERα gene causes infertility through defects in the reproductive tract and gonads of female mice while disruption of the ERβ gene by the insertion of neo-cassette into exon 3 is associated with only disruption of ovulation (Couse, et al. 2005; Eddy, et al. 1996; Krege, et al. 1998; Lubahn, et al. 1993).

The innovation of genetically induced mice has allowed for further knowledge of estrogen signaling. Studies using ERα and ERβ knockout (KO) mice have demonstrated that ERα is essential for endometrial receptivity (Buchanan, et al. 1999; Cooke, et al. 1997; Lubahn et al. 1993). Similarly, studies utilizing PR null mouse strains have demonstrated that uterine stromal cells are the mediators of progesterone inhibitory effects on estrogen induced proliferative response of the uterine epithelium (Kurita, et al. 1998). Simultaneously, Tan et al demonstrated that there is compartmentalization of uterine ERα, but extremely low to undetectable expression of ERβ, is associated with early periimplantation days of gestation (Tan et al. 1999). Early gestation (days 1 and 2) ERα mRNA is primarily localized in the luminal and glandular epithelium while localization is additionally seen in the stroma on days 3 and 4; however, by day 8 of gestation ERα exhibits down regulation in decidual cells immediately surrounding the embryo. Collectively, these studies suggest that specific regulation of ER gene expression seems to define the implantation window.

Additionally, the analysis of the implantation window has demonstrated that the estrogen effects on the endometrium are tightly regulated. Ma et al demonstrated that lower estrogen levels tend to sustain the receptivity of the uterus; however, higher concentrations shut down this time window, although the exact mechanism is not well understood (Ma, et al. 2003). NCOA6 is a coactivator for multiple nuclear receptors and has been demonstrated that Ncoa6 KO mice fail to develop due to defects noted in the placenta and other tissues (Kuang, et al. 2002; Mahajan and Samuels 2005). Kawagoe et al demonstrated that NCOA6 regulates estrogen sensitivity and signaling affecting the uterine receptivity status (Kawagoe, et al. 2012). Using a conditional KO of Ncoa6 in mice, Kawagoe was able to demonstrate that loss of NCOA6 results in ERα accumulation in stromal cells and accumulation of steroid receptor coactivator 3 (SRC3), a potent ERα coactivator (Kawagoe et al. 2012). Therefore, the loss of NCOA6 leads to the inability to attenuate estrogen sensitivity via an accumulation of ERα and SRC3 at the implantation site rendering the uterus non-receptive with pregnancy failure.

These observations suggest a localized site of coordinated effects of estrogen on its target tissue. Since both stroma and epithelium express ERα, one would assume that estrogen-induced epithelial proliferation is controlled directly through the interaction with the specific nuclear steroid receptor. However, studies have demonstrated that target tissue estrogen-induced response is not necessarily related to its affinity or occupancy to the receptor (Das et al. 1997), since an estrogen receptor antagonist, ICI-182,780 failed to inhibit uterine estrogen responsive lactoferrin (Ltf) gene expression and water imbibition induced by certain estrogens in ERKO mice, however, this ICI-182,780 indeed suppressed the uterine Ltf expression in wild-type mice induced after E2, which indicated an estrogen signaling independent of both ERα and ERβ.

Distinct estrogen signaling pathways

Specific functions of AF-1 and AF2 domains of ERα

Binding of ER at genomic sites regulates gene expression. Different physiologic responses are initiated by estrogen binding to ER leading to receptor conformational changes that are required for transcriptional activity. Two transactivation function domains mediate transcriptional activation: activation fuction-1 (AF-1) in the N-terminal domain and activation function-2 (AF-2) in the C-terminal ligand-binding domain (LBD) (Kushner, et al. 2000; Tremblay, et al. 1999). Both AFs have unique differential gene activation through cell type-specific co-activators (Hsia, et al. 2010; Xu, et al. 1998). Previous studies demonstrated that the significance of these specific domains with regards to the functionality of ER depends on AF-1 (Merot, et al. 2004).

However, although reproduction is affected in ERα null mice (Lubahn et al. 1993), several estrogen effects still persist, such as early responses to uterine edema and gene expression (Das et al. 2000; Das et al. 1997) and vascular injury response (Iafrati, et al. 1997). In the uterus of this null mouse, through alternative splicing, a chimeric small ERα protein (~55 kDa), in which 64 amino acid residues mainly for the B region can be partially deleted from the N-terminal A/B regions of ERα (Couse, et al. 1995). In addition, studies also reported uterine detection of a short form of ERα transcript, representing the deletion of a portion of exon 2 followed by insertion of a frame shift and at least two stop codons at the 5’-end of exon 3 (Couse et al. 1995), but the significance of this remains unknown. The truncated small ERα variant lacks the AF-1 domain, which Pendaries et al. demonstrated could be partially dispensable to mediate the estrogenic effects in the uterus since the variant possesses a residual estrogen-dependent transcriptional activity with an intact AF-2 region (Couse et al. 1995; Pendaries, et al. 2002). Further studies revealed the crucial role of AF-2 for estrogen mediated endometrial epithelial proliferation using antagonists and selective ER modulators (SERMs) (Arao, et al. 2011). However, further studies also revealed that AF-1 activation function is required for the E2-induced uterine epithelial proliferation, whereas it is partially dispensable for the induction of uterine edema by chronic estrogen stimulation (Abot, et al. 2013). Additionally, Kurita et al. revealed differences in the estrogen induced proliferative responses between human and mouse epithelial cells, which seems to be species specific with regards to the utilization of the AF domains within the ERα (Kurita, et al. 2005). Therefore, further investigations into these domains to evaluate the specific physiologic roles of AF-1 and AF-2 are still needed.

Interdependent regulation by uterine epithelial and stromal cells

Deletion of ERα in uterine epithelial cells leads to infertility, however, this receptor loss does not prevent estrogen induced epithelial cell proliferation (Winuthayanon, et al. 2010). In this regard, tissue recombination studies have also shown that ERα action in stromal cells mediates the estrogenic proliferation events in the epithelium via a paracrine manner (Cooke et al. 1997; Cunha, et al. 2004). In addition, Pawar et al also showed that epithelial ERα controls uterine decidualization via a paracrine mechanism through epithelial-stromal cross-talk during the early implantation (Pawar, et al. 2015). Similarly down regulation of the progesterone receptor in the uterine epithelium is depended on stromal ERα (Kurita, et al. 2000). The theory of interdependency between the endometrial epithelium and stroma proposes an intercellular cross talk through different signaling pathways (Figure 1), which can mimic the effects of the traditional ligand-receptor pathway.

Figure 1.

Leukemia inhibitory factor (LIF) signaling

LIF is a well-characterized paracrine factor produced by the glandular epithelium under estrogen stimulation that regulates implantation (Stewart, et al. 1992). LIF executes its biological function by activating its own receptor (LIFR) followed by the recruitment of glycoprotein 130 (GP130) (Taga and Kishimoto 1997). Yang et al demonstrated the expression patterns of LIFR and gp130 in the luminal epithelium on day 4 of pregnancy in mice (Yang et al 1995). LIF acts on the luminal epithelium to activate Janus kinase (JAK), a non-receptor tyrosine kinase, which mediates the phosphorylation and activation of signal transducer and activator of transcription 3 (STAT3) (Heinrich, et al. 1998; Tomida, et al. 1999). Lif null mice demonstrate normal ER and PR expression but absent expression of EGF-life growth factors such as heparin binding epidermal growth factor (Hbegf), amphiregulin (Areg), and epiregulin (Ereg) near the blastocyst on day 4 of gestation (Song, et al. 2000). Although the exact function of EGF factors is unknown, the EGF receptors are expressed on stromal cells during pregnancy suggesting a role as paracrine mediators driving stromal proliferation (Figure 1) (Song et al. 2000; Xie, et al. 2007). Furthermore, Stat3 null mice demonstrate increased epithelial expression of estrogen regulated genes Ltf and Muc1, which heighten estrogen signaling allowing for persistent proliferation in the luminal epithelium and a lack of proliferation in the stromal layer (Sun, et al. 2013), indicating an undifferentiated uterine state. Collectively these findings demonstrate that the loss of the LIF-STAT3 signaling pathway culminates in undifferentiated uterine epithelium and therefore non-receptive to the embryo implantation.

Indian hedgehog (IHH) signaling

IHH, a member of the hedgehog gene family, is a progesterone regulated factor produced in the epithelium and controls stromal function via paracrine mechanisms (Figure 1) (Matsumoto, et al. 2002b; Takamoto, et al. 2002). Using a conditional knockout mouse, Ihh^d/d, studies have demonstrated that in the absence of Ihh a uterine non-receptive state is achieved secondary to failure of stromal cell proliferation and vascularization along with increased estrogen signaling during the periimplantation phase (Franco, et al. 2010; Lee, et al. 2006). The lack of stromal cell proliferation is in part due to Ihh’s regulation of the EGFR in the stromal compartment, which allows the stroma to be activated by the EGF factors produced by the epithelium secondary to estrogen stimulation (Franco et al. 2010). These observations suggest that the hedgehog signaling cascade plays a crucial role in the events occurring just prior to decidualization.

Chicken ovalbumin upstream promoter-transcription factor II (COUP-TFII) signaling

Previously studies have shown that epithelial IHH stimulates COUP-TFII [also known as nuclear receptor subfamily 2, group F, member 2 (Nr2f2)], a stromal factor that mediates decidualization (Lee, et al. 2010; Lee et al. 2006; Takamoto et al. 2002). Using PR-Cre to cause conditional ablation of endometrial COUP-TFII in mice demonstrate a defect linked to decreased expression of bone morphogenic protein 2, factor produced by the stroma in response to progesterone stimulation (Kurihara, et al. 2007). This aberration results in failure to undergo structural changes involved in decidualization. Additionally, COUP-TFII -deficient mice show an increase in epithelial ERα expression and increased estrogen activity resulting in Ltf and Muc1 expression (Kurihara et al. 2007). Furthermore, studies have shown that the loss of epithelial ERα activity by COUP-TFII is critical for successful progression of embryo implantation and decidualization (Lee et al. 2010). Overall, these studies conclude that COUP-TFII plays a major role in epithelial remodeling and differentiation through controlling ERα activity to support the initiation of embryo implantation.

Fibroblastic growth factor (FGF)/Insulin-like growth factor (IGF) signaling

Stromal factors that regulate epithelial function have also been identified in the intercellular communication pathways, which play a critical role in the implantation window. Specifically, fibroblast growth factors (FGFs) and insulin-like growth factor-1 (IGF1) have been proposed for stromal epithelial communication in a variety of tissues. The FGF family is a group of stromal ERα-induced paracrine factors that act on the epithelium to activate ERK1/2 signaling cascades that stimulate epithelial proliferation (Figure 1) (Li et al. 2011). In this regard, based on uterine co-culture experiments, evidence suggest that estrogen mediated epithelial proliferation may involve stroma-derived factors FGF10 and BMP8a (Chung, et al. 2015). With the FGF10 receptor, FGFR2, primarily detected in the epithelial cells in both the co-culture system and the adult ovariectomized uteri, collectively these results suggest that FGF10/FGFR2 signaling may be specifically involved in the stroma-epithelial cross-talk during early pregnancy. However, Filant et al. demonstrated that conditional ablation of FGFR2 after birth results in abnormal basal cell appearance and stratification in the luminal epithelium, as well as, subfertility that progressed to infertility (Filant, et al. 2014). These results show the critical importance of FGFR2 in postnatal uterine development of LE and female fertility; however, further studies are needed to delineate the molecular mechanism resulting in the observed phenomenon in FGFR2 null mice that lead to complete infertility in multiparous FGFR2 mutant mice. Similarly IGF1, following estrogen stimulation, is abundantly detected in the uterus with IGF1R identified in the epithelium (Kapur, et al. 1992; Murphy and Ghahary 1990). A lack of IGF1 expression is observed in ERKO mice stimulated with estrogen, validating these previous findings (Hewitt, et al. 2010). The fact that IGF1R and IGF1 are abundantly expressed in the uterine epithelium, suggests that IGF1 may be a paracrine mediator involved in the epithelial proliferation during early pregnancy. It is hypothesized that IGF1 stimulates activation of PI3/AKT pathway in the epithelium, which phosphorylates and inactivates glycogen synthase kinase 3 beta (GSK3β), allowing for epithelial proliferation (Zhu and Pollard 2007). When analyzing the role IGF1 in IGF1 knockout mice, Sato et al. demonstrated that uterine growth is supported by systemic IGF1in the absence of local IGF1 production (Sato, et al. 2002). This suggests that local IGF1 is not a direct mediator to estrogen effects in the uterus but rather systemic IGF1 may be the key factor for growth.

Wnt signaling

The biological effect of estrogen can also be associated with Wnt signaling pathways. Wnt is a family of genes that encode a large group of glycoproteins that have a critical role in embryonic development and are also involved in tumorigenesis (Smalley and Dale 1999). The canonical Wnt signaling pathway, which involves regulation of β-catenin, has been the most widely studied. The activation of Wnt signaling stabilizes intracellular β-catenin by antagonizing kinase activity of GSK3β. In the absence of Wnt signaling, GSK3β forms a multimolecular complex with axin (a bridging molecule), adenomatous polyposis coli and β-catenin, leading to phosphorylation and then subsequent degradation via ubiquitination pathway of β-catenin. When activated, β-catenin translocates to the nucleus and forms a complex with downstream effectors such as lymphoid enhancer factor (Lef)/T cell factor (Tcf) family which stimulate transcription of Wnt Target genes. These target genes are involved in cellular organization during embryonic development, proliferation and differentiation as well as cell-to-cell communication and cell fate specification (Smalley and Dale 1999).

Previously, studies have shown that Wnt4 expression is upregulated at the site of embryo implantation during decidualization (Daikoku, et al. 2004). Further studies revealed that Wnt4 plays a key role in implantation and decidualization (Franco, et al. 2011) and this action is mediated downstream of progesterone via β-catenin signaling pathway in uterine stromal activity with proliferation and differentiation (Li, et al. 2013; Rider, et al. 2006).

We previously demonstrated the presence of an ER-independent pathway of estrogen stimulation via Wnt pathway (Hou et al. 2004). After exposing ERα KO (ERKO) mice with estrogen, prompt stabilization and localization of β-catenin in the nucleus of uterine epithelial cells was observed. This finding was confirmed with the injection of adenovirus-driven expression of SFRP2, a Wnt antagonist suppressed rapidly by estrogen during the early phase in the uterus in an ER-independent manner, since (Das et al. 2000), demonstrating down regulation of β-catenin and halting epithelial cell growth without affecting early estrogen effects (Hou et al. 2004). Similarly, studies have also shown that Wnt/β-catenin downstream effectors Lef1 and Tcf3 are upregulated in an estrogen independent manner (Ray, et al. 2008). Through immunofluorescence studies, Lef1/Tcf3 localization was confirmed in the epithelial cells after estrogen exposure and interestingly found to be interacting with ERα in a time-dependent manner (Ray et al. 2008). Furthermore, evidence was provided for an ERα and Tcf3/Lef1 complex occupying a certain DNA region of estrogen responsive gene promoters, suggesting a non-classical induction mechanism of the Wnt/β-catenin pathway that is necessary in the estrogen-dependent gene regulation.

GPR30 signaling

GPR30 (also known as GPER1), a G-protein coupled receptor, has been implicated in early non-genomic signaling mediated by E2. In mouse uterus, GPR30 localizes primarily in the uterine epithelial cells (Gao, et al. 2011). Studies from GPR30 knock-out mice appear to imply that GPR30’s role in uterine biology is minimal for estrogenic growth regulation (Martensson, et al. 2009; Otto, et al. 2009; Wang, et al. 2008). In contrast, utilizing selective activation of GPR30 by G-1, studies have shown that GPR30 is involved in regulating early signaling events, including the inhibition of ERK1/2 and ERα (Ser118) phosphorylation signals in the uterine stromal compartment, suggesting a paracrine signaling is involved (Figure 1) (Gao et al. 2011). However, it should be noted that this study was unable to exclude the possibility through the off-target effects of G-1. Moreover, further studies should be considered to show that Gper1 null mice are insensitive to G-1 in the above uterine effects. Overall, studies show that GPR30 can act as a negative regulator of ERα-dependent uterine growth in response to E2.

Molecular links between the phase I and phase II estrogenic responses in the uterus

Early (phase I) and late (phase II) estrogenic responses in the uterus have been recognized for more than 70 years, yet mechanisms involved in their regulation remain controversial. One concept is that an early events(s), occurring within the first 6 h, prepares the uterus for later (18–30 h) increase in DNA synthesis, cell proliferation and protein synthesis. An alternate view is that the late growth phase is a result of the continuous presence of a stimulus. Discussion of either concept usually makes the assumption that all of the responses are dependent upon ligand interaction with one of the two estrogen receptor isoforms (ERα and ERβ). However, we and others have shown that ERα null mice (ERKO) or wild-type mice in which ER functions are silenced by ER antagonist ICI 182,780 manifest expression of several early genes in response to 4-hydroxyextradiol-17β or a xenoestrogen (kepone), as well as induction of early responses such as water imbibition and macromolecular uptake by 4-hydroxyextradiol-17β (Das, et al. 1998; Das et al. 2000; Das et al. 1997; Hewitt, et al. 2003; Hou et al. 2004; Ray, et al. 2006; Watanabe, et al. 2003). Furthermore, studies have also shown that ICI was able to suppress expression of Ltf, a well characterized estrogen-responsive uterine gene, in the wild-type mice after E2, indicating the effectiveness of ICI in this study (Das et al. 1998; Das et al. 1997). Utilizing the same effective dose of ICI (Das et al. 1998; Das et al. 1997), we have identified two such ER-independent uterine genes Bip and Sik-SP that are regulated by E2 in ERKO mice (Das et al. 2000). The bimodal nature of estrogen effects coupled with phase I ER independent estrogenic responses and phase II, mostly ER dependent responses has ignited interest in understanding the pathways linking these two phases.

Role of Bip

Bip, also known as grp78 encoded by Hspa5, is a member of the heat shock protein HSP70 chaperone family and it is induced by estrogen in an ER independent manner as a phase I response (Das et al. 2000; Ray et al. 2006). It is a protein that resides in the endoplasmic reticulum (Figure 2), where assembly of newly synthesized peptides occurs, and is abundantly present during cell proliferation and differentiation particularly at the site of embryo implantation during decidualization (Simmons and Kennedy 2000). As a chaperone molecule, the role of BIP is for functional maturation of steroid hormone receptors. In the mouse uterus it mediates estrogen dependent responses through molecular association with ERα (Ray et al. 2006). Studies have demonstrated through an in vivo and in vitro mouse model that suppression of Bip antagonizes ERα mediated gene transcription and compromises estrogen-dependent phase II growth response (uterine epithelial cell proliferation) with sustained phase I responses (water accumulation and macromolecular uptake). Most interesting, is the lack of growth response in the presence of ERKO state even if Bip is upregulated (Ray, et al. 2007). Although this study analyzed xenoestrogen and Bip, it demonstrates the close relationship between Bip and ERα in regulation of uterine growth. Together, studies suggest that the functional activation of ERα via Bip plays a role in coordinating phase I responses with those of phase II for regulated growth and differentiation via estrogen signaling in the mouse uterus.

Figure 2.

Some organochlorine compounds, such as polychlorinated biphenyls, are highly persistent organic pollutants in many industrial nations. These compounds have gained attention recently secondary to their potential for adverse effects on health and reproduction. The reproductive toxicity is thought to be due to their estrogen-like properties, hence they are categorized as xenoestrogens. The ability to bind to ERα, allows for mimicking effect on target organ function, although the mechanisms are not well defined (Das et al. 1997). There are however, significant differences in coactivator recruitment and transcriptional activation in tissues exposed to xenoestrogens corresponding to distinct biological effects causing endocrine disruption. Furthermore, these compounds are effective at very low doses comparable to their level of exposure making them very potent estrogens (Ray et al. 2007).

Knowing the critical role that Bip plays in regulation of estrogen-dependent ERα mediated gene transcription and growth, the xenoestrogen mediated effects with regards to upregulation of Bip under certain conditions could be potentially harmful in respect to enhanced uterine estrogenicity. Specifically, the xenoestrogen kepone can induce sustainable levels of uterine Bip without involving ER, which in turn regulates the kepone-dependent ERα mediated gene expression (Ray et al. 2007). Furthermore, with the notion that stress can regulate Bip expression and the ability of uterine growth via stress induced estrogen response in mice, studies have demonstrated endogenous Bip via stress-related signals contributes to uterine estrogenicity for kepone (Ray et al. 2007). Thus the combination of a variety of signals in the body, such as stress, and xenoestrogens can act as plausible risk factor enhancing estrogenicity and therefore major health concerns.

Role of Sik-SP

The nucleolus is the nuclear subdomain that primarily carries out assembly of ribosomal subunits in eukaryotic cells. A recent study has uncovered an unexpected role for uterine estrogen signaling which involves a nucleolar protein Sik-similar protein (Sik-SP, also known as Nop58/Nop5/Nol5) (Chung, et al. 2012). Studies have shown that the expression of uterine Sik-SP is tightly regulated by E2 in an ER-independent manner, but is still required for the control of ERα-mediated late uterine functions (Figure 2) (Chung et al. 2012; Das et al. 2000). Specifically, using both the in vivo and in vitro co-culture approaches, studies have shown that E2-induced Sik-SP directly interacts with ERα to mediate ERα-dependent gene regulation and is necessary to coordinate the biphasic responses in the uterus for its appropriate growth under the direction of E2. Overall, this finding of ERα-independent early Sik-SP contributing to ERα-regulated events adds new insights to our understanding of nucleolar involvement in uterine estrogen signaling.

Taken together, these studies provide evidence of non-classical pathways that mediate estrogen actions in a time dependent fashion, possibly shedding a light on how the biphasic: phase I and phase II estrogenic responses are molecularly linked to mediate uterine cell proliferation.

ER-independent genes associate with embryo implantation

To understand the functional significance of estrogen-induced ER-independent early uterine genes, studies were undertaken to determine whether E2-administration in the delayed implantation model in mice enhances the expression of Bip and Sik-SP at the site of implantation. Indeed, results demonstrated that these genes are specifically up-regulated in the sub-luminal stromal cells at the site of the implanting embryo following activation with E2; however, the delayed stage of the uterus does not show any expression at the site of embryo (Chung et al. 2012; Reese, et al. 2001). Furthermore, this induced expression is consistent with the status of expression in normal implantation sites on D5 for Bip (Figure 3) and Sik-SP (Chung et al. 2012). Taken together, the studies have shown that these ER-independent genes are physiologically important during the onset of embryo implantation under the direction of E2.

Figure 3.

Conclusions

This review has served as an update of the literature describing the molecules involved in estrogen signaling in the mouse uterus during early pregnancy. We have discussed the signaling pathways that are ER dependent and ER independent as well as, the molecular links that shed light into the complexity of the bimodal estrogen actions occurring in early pregnancy. Dysregulation of the cross-talk between these pathways can lead to implantation failure through the inability to obtain a receptive uterine epithelium. Environmental toxins can mimic estrogen pathways, however the mediated effects differ from normal through the enhanced estrogenicity of the uterus creating a non-receptive uterine epithelium. Continued research into the mechanisms involved in estrogen signaling will expand our understanding of this delicate and time sensitive event. Understanding the molecular interactions will provide the knowledge needed to improve current treatments of infertility through the exploration of new ideas, techniques and technology.