|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL GROWTH, PHYSIOLOGY, AND REPRODUCTION |
Departments of Animal Sciences and Biochemistry/Biophysics, Oregon State University, Corvallis 97331
| Abstract |
|---|
Key Words: estradiol-17β luteinizing hormone membrane receptor oxytocin progesterone uterus
| INTRODUCTION |
|---|
From the standpoint of steroids and domestic animals, most interest has focused on the roles of progesterone (P4) and estrogen because of their effects on the reproductive tract and anterior pituitary gonadotropin secretion. Thus, for comparative purposes this review provides a brief overview of the genomic actions of estrogen and P4 in domestic animals. However, major emphasis is placed on describing the current state of knowledge regarding the nongenomic action of estrogen and P4 in domestic animals with reference to relevant research in other species.
| CHARACTERISTICS OF NUCLEAR ESTROGEN AND P4 RECEPTORS |
|---|
Two forms of the nuclear estrogen receptor (ER) are found in mammalian species. These 2 receptor subtypes are products of different genes and are designated ER
(ER
) and ERβ. Both monomeric forms of each receptor subtype are approximately 60 kDa, and both subtypes bind estradiol-17β with high affinity but differ in DNA-binding affinity. Estrogen receptor
and ERβ can be coexpressed in some target tissues but they also exhibit different tissue or cell expression patterns (Hall and McDonnell, 1999
; Heldring et al., 2007
).
The PR also exists as 2 isoforms, designated as PR-A and PR-B (Edwards, 2005
). The 2 isoforms of PR arise from a single gene as a result of alternating promoter start sites that give rise to 2 different PR mRNA. Progesterone receptor-A and PR-B have identical homologies of the LBD, DNA-binding domain, and AF-1 of the N-terminal domain; however, PR-A differs from the full-length PR-B by truncation of the most N-terminal 164 AA. The 2 isoforms of PR differ in their functional activities. Both isoforms are coexpressed in most target tissues, but the ratio of the 2 forms of PR can vary depending on cell type or physiological state. An AF-3 domain within the most N-terminal 164 AA of PR-B promotes a stronger transcriptional activity by PR-B (Li and OMalley, 2003
). It has been reported that PR-A can act as a transrepressor of PR-B and perhaps other steroid receptors such as ER (Edwards, 2005
). In PR-A and PR-B knockout mice, PR-A has been found to play a major role in mediating the actions of P4 in the uterus and ovary, whereas PR-B is more important in mammary gland development (Mulac-Jericevic et al., 2000
, 2003
).
| CHANGES IN UTERINE AND HYPOPHYSEAL NUCLEAR ER AND PR DURING VARIOUS REPRODUCTIVE STATES |
|---|
By simply examining the relationship between uterine concentrations of ER and PR and systemic levels of estrogen and P4, one might deduce that estrogen upregulates both ER and PR. Indeed, this aspect of estrogen action in the endometrium was confirmed by using ovariectomized ewes treated with estrogen or P4 alone or in combination (Koligian and Stormshak, 1977b
).
Comparatively few studies have been conducted to quantify estrogen and PR in the anterior pituitary or hypothalamus of domestic animals. Early research established the presence of androgen and ER in the hypothalamus and anterior pituitary of the ram and ewe (Pelletier and Caraty, 1981
; Glass et al., 1984
). Subsequent research revealed that estrogen and androgen receptors were present in various morphological nuclei throughout the ovine hypothalamus. However, those cells containing the greatest concentrations of ER in the hypothalamus of the ewe were located in the preoptic area, ventromedial nucleus, and arcuate nucleus (Herbison et al., 1993
; Lehman et al., 1993
). Similarly, these same sites in the hypothalamus of the ram were found to be richly endowed with androgen receptors (Herbison, 1995
). Alexander et al. (1993)
examined quantities of ER in the hypothalamus, medial amygdala, and anterior pituitary of high and low sexually performing rams. The total number of ER in the anterior pituitary was nearly 100-fold greater than the number present in hypothalamic regions. The proportion of total ER in the preoptic area of the hypothalamus that was bound by estradiol-17β was significantly greater in high-performing than in low-performing rams. In contrast, the proportion of ER that was occupied with estradiol-17β in the anterior pituitary of high-performing rams tended to be less than that in low-performing rams. These differences in ER did not appear to affect secretion of LH during sexual behavior testing because high and low sexually performing rams had similar LH profiles relative to systemic basal concentrations, number of secretory pulses, and pulse amplitude of the gonadotropin.
During the estrous cycle of the cow, quantities of estrogen and GnRH receptors in the anterior pituitary are markedly increased prior to the preovulatory surge of LH (Nett et al., 1987
). Estrogen of follicular origin apparently causes upregulation of the ER and GnRH receptors (Turzillo and Nett, 1999
). Significant increases in estradiol receptors in the hypothalamus and anterior pituitary of ewes on d 22 postpartum were found to immediately precede the onset of ovarian cyclicity (Wise et al., 1986
). In postpartum cows, receptors for both GnRH and estradiol in the anterior pituitary are lowest immediately after parturition, then increase to maximal concentrations by 2 wk postpartum (Nett et al., 1988
). Pituitary content of LH in the cow is low following parturition and increases significantly through d 30 postpartum. Exposure of the anterior pituitary to the high systemic concentrations of placental estrogen that are prevalent during the late stage of gestation may be responsible for downregulating estrogen and GnRH receptors and for synthesis of LH. Indeed, chronic treatment of ovariectomized ewes with estradiol for 3 wk caused a reduction in mRNA for the LH
and β subunits in the anterior pituitary (Nilson et al., 1983
). Collectively, these reports suggest that the estradiol exhibits a paradoxical effect on LH synthesis, with low physiological levels stimulating and chronic levels inhibiting the synthesis of this gonadotropin. On the other hand, estradiol, whether chronically administered to ovariectomized ewes or added to primary ovine pituitary cell cultures at physiological concentrations, significantly increased the synthesis and secretion of prolactin (Shupnik et al., 1979
; Vician et al., 1979
). Progesterone suppressed the estradiol-induced increase in preprolactin mRNA production and prolactin synthesis.
| P4, ESTRADIOL, AND UTERINE OXYTOCIN RECEPTORS |
|---|
, which causes regression of the corpus luteum (McCracken et al., 1999
Experimental evidence also indicates that exogenous estradiol can upregulate uterine concentrations of OTR. Treatment of ovariectomized ewes with P4 (5 to 12 d) followed by 2 d of estradiol treatment caused an increase in the uterine endometrium OTR as compared with treatments with P4 only (Vallet et al., 1990
). Further, exposure of bovine endometrial explants to estradiol stimulated an increase in OTR compared with levels of receptor present in control explants (Leung and Wathes, 2000
). Whereas the reporter region of the bovine OTR gene contains at least 3 estrogen response element (ERE) half-sites, there is no firm evidence that the ligand-bound ER binds directly to these sites to stimulate OTR expression (Telgmann et al., 2003
). Similarly, the promoter region of the ovine OTR gene contains neither full ERE or ERE half-sites (Fleming et al., 2006
). Deletion and mutation analysis of the promoter region of the ovine OTR revealed that stimulation of the gene by estradiol is dependent on a guanine-cytosine-rich SP1 binding site at –104 and –64 bp.
Overall, experimental evidence favors a conclusion that estrogen upregulates and P4 downregulates uterine OTR. However, the intracellular dynamics of OTR expression in uterine target cells, as affected by exposure to estrogen and P4 during the estrous cycle, may be more complex than presently envisioned.
Pregnancy recognition in the ruminant encompasses the production of interferon-
(IFN-
) by the developing conceptus, which inhibits development of the endometrial luteolytic mechanism and pulsatile release of PGF2
(Bazer, 1992
). It has been shown experimentally that IFN-
acts to prevent increases in endometrial ER mRNA and protein, with a consequent reduction in OTR and a suppression in the uterine release of PGF2
(Spencer et al. 1995
; Spencer and Bazer, 1996
). It has been proposed that the antiluteolytic effects of IFN-
are mediated by the direct inhibition of ER
gene transcription, thereby preventing ligand-bound ER from stimulating OTR gene transcription (Fleming et al., 2006
).
| UTERINE BLOOD FLOW DURING THE ESTROUS CYCLE |
|---|
The precise mechanism by which P4 promotes vasoconstriction of uterine arteries is unknown. However, either directly or indirectly, P4 causes an increase in the concentration of uterine arterial smooth muscle
1-adrenergic receptors (Ford et al., 1984
). Activation of these adrenergic receptors by norepinephrine emanating from periarterial sympathetic nerves results in vasoconstriction of the uterine vascular bed (Greiss and Pick, 1964
).
Exposure of the uterine artery to estrogen results in a rapid increase in blood flow. Injection of estrogen directly into the uterine arteries of ovariectomized ewes increases blood flow within 30 min (Killam et al., 1973
). Similarly, in ovariectomized cows an increase in uterine blood flow occurred within 20 to 40 min after injection of 1 µg of estradiol-17β into the uterine artery (Ford and Reynolds, 1983
). The speed with which this uterine response to estrogen occurs suggests that the steroid is acting nongenomically in the vasculature.
Estrogen synthesized by porcine and bovine conceptuses is believed to act locally to promote transient increases in blood flow to the gravid uterine horn during early gestation. In sows, uterine venous blood and uterine flushings contained increased quantities of estrogen coincident with a transient increase in uterine blood flow on d 13 of gestation (Ford et al., 1982
). Similarly, in cows a transient increase in blood flow to the gravid horn occurs on d 14 to 18, and during this time the venous blood draining this horn contains increased concentrations of estradiol-17β (Ford et al., 1981
). The fact that the estrogen produced by the conceptuses of these species acts locally to cause an abrupt transient increase in uterine blood flow suggests that estrogen may be acting nongenomically, at least initially, to promote this uterine vasculature response, just as it does in the ovariectomized cow given an intraarterial injection of estradiol-17β, as described above.
| NONGENOMIC ACTIONS OF ESTROGEN |
|---|
and ERβ into Chinese hamster ovary (CHO) cells. Expression of the cDNA for each receptor resulted in production of a single transcript that gave rise to both membrane and nuclear receptors. The membrane receptor number was only 3% as great as the nuclear receptor density. Treatment of cells with estradiol-17β stimulated a phosphoinositide cascade as well as activation of adenylate cyclase, suggesting that the receptors were coupled to G
q and G
s proteins. More importantly, exposure of the transfected cells to estradiol provoked an increase in extracellular-regulated kinase (ERK) activity within 10 min, which was considered to be the consequence of the nongenomic action of the steroid (Figure 1
and Sarc homology collagen (Shc) adaptor protein upon stimulation of cells with estradiol-17β. Using a mutagenesis approach, the investigators demonstrated that Src homology (SH) 2 and phosphotyrosine binding domains of Shc and the AF-1 region of ER
are the interacting molecular sites of association. The presence of estradiol-17β rapidly induced Shc phosphorylation, presumably by a Src protein family member, with the resultant formation of an Shc-growth factor receptor-binding protein-2–son-of-sevenless (SOS, a guanine nucleotide exchange factor) complex with a consequent phosphorylation of ERK-1/-2 and activation of the transcription factor Elk-1. Estradiol-17β rapidly induced (within 20 min) formation of membrane ruffles and pseudopodia in modified MCF-7 cells. The estradiol-induced morphological changes were prevented by the antiestrogen ICI 182,780. In a subsequent study, Song et al. (2004)
localized at the plasma membrane and the mechanisms by which Shc and IGF-I receptor (IGF-IR) mediate this process. They were able to demonstrate that estradiol-17β rapidly induced formation of a ternary protein complex consisting of Shc, ER
, and IGF-IR. Knockdown of Shc with a specific small inhibitory RNA decreased the association of ER
and IGF-IR by 87%, suggesting that Shc is a crucial molecule in the formation of this ternary complex. Results of their experiments demonstrated that Shc and IGF-IR serve as key localized membrane elements that facilitate ER
-mediated rapid estrogen action.
|
, 1 of the 6 X AA is Cys447, which, in the unoccupied receptor, undergoes palmitoylation by palmitoyl acyltransferase. Palmitoylation was shown to facilitate ER
translocation to the plasma membrane and association of the receptor with the membrane protein caveolin-1.
Mobilization of Ca2+ may contribute to the observed nongenomic actions of estrogens. An early report by Morley et al. (1992)
contained data indicating that exposure of pig granulosa cells to 10–6 to 10–10 M estradiol-17β caused an immediate 4- to 8-fold increase in intracellular Ca2+. This cellular response is specific of estrogens because various progestins and androgens were ineffective. The rapid estrogen-induced increase in intracellular Ca2+ was not reduced by incubating the cells in Ca2+-free medium or by pretreating with Ca2+ channel blockers. It appears then that the release of Ca2+ is from intracellular stores triggered by inositol trisphosphate generated by an ER-induced hydrolysis of membrane phosphatidylinositol 4,5-bisphosphate.
There are some who believe the nongenomic action of estrogen, at least in some tissues, is mediated by a G protein-coupled receptor referred to as GPR30 (Figure 1
). This is a bizarre protein whose role as an ER and whose location within the cell are unsettled and controversial. Human SKBR3 breast cancer cells fail to express ER
and ERβ but do contain the GPR30 protein. Filardo et al. (2000)
showed that exposure of these cells to estradiol-17β (1 nM) for 5 min resulted in phosphorylation of ERK-1/-2. Surprisingly, the estrogen antagonist, ICI 182,780 (1 µM), was as effective as estradiol-17β in causing phosphorylation of ERK-1/-2. Further, in MDA-MB-231 cells, which express some ERβ but little GPR30 protein, there is no activation of ERK in response to either estradiol or the estrogen antagonist, suggesting that the classical ER are not mediators of estrogen action in this cell type. Proof of this was provided by results of an experiment in which a GPR30 cDNA was transfected into MDA-MB-231 cells, causing overexpression of the GPR30 protein. Exposure of these transfected cells to estradiol or ICI 182,780 caused phosphorylation of ERK-1/-2, whereas estradiol-17
and P4 were ineffective. Filardo et al. (2000)
provided evidence suggesting that estradiol signaling occurred via a Gβ
subunit-dependent, pertussis toxin-sensitive pathway that ultimately results in transactivation of epidermal growth factor (EGF) receptor (EGFR) via the autocrine release of heparin-bound EGF. Binding of heparin-bound EGF to EGFR promotes receptor dimerization and autophosphorylation of Tyr residues within the cytoplasmic domain of the receptor. Specific recognition of these phosphotyrosines by the adapter proteins Grb-2, Shc, or both, and the guanine nucleotide exchange factor SOS forms the Grb-2-Shc-SOS adaptor protein complex, which links activated EGFR to MAPK via the monomeric GTPase, Ras. The Ras recruits Raf, which promotes cascade phosphorylation and activation of Mek-1 and its substrates ERK-1/-2. The response of human breast carcinoma cells to estradiol-17β in terms of phosphorylation of ERK-1/-2 is transient, rapidly returning to basal levels 30 to 60 min after initial exposure to estradiol. Filardo et al. (2002)
hypothesized that estrogens also act via GPR30 to stimulate adenylate cyclase to inhibit the EGFR to MAPK pathway. Exposure of membranes from human SKBR3 breast cancer cells and MDA-MB-231 cells (that were forced to overexpress GPR30 protein) to estradiol and ICI 182,780 responded to treatment with an increase in adenylate cyclase activity as measured by production of cyclic adenosine monophosphate (cAMP). By contrast, estradiol and ICI 182,780 failed to cause stimulation of cAMP production by membranes isolated from MDA-MB-231 cells (which contain ERβ but neglible quantities of GPR30) or promote blockade of EGF-induced activation of ERK-1/-2. Evidence is presented suggesting that estrogen-induced suppression of ERK-1/-2 activity in cancer cells expressing GPR30 may be due to protein kinase A phosphorylation of Ser residues in Raf-1. Although the reports by Filardo et al. (2000
, 2002)
suggested that GPR30 was a resident of the plasma membrane, the more recent research of Revankar et al. (2005)
indicated otherwise. To determine the cellular localization of the receptor, these latter investigators expressed GPR30 as a fusion protein with green fluorescent protein in COS-7 (monkey kidney fibroblast) cells. The receptor was expressed with the β2-adrenergic receptor, a G-protein-coupled receptor. Confocal fluorescent microscopy revealed that the β2-adrenergic receptor was localized to the plasma membrane, whereas GPR30-green fluorescent protein was localized in the endoplasmic reticulum and the Golgi apparatus. The same intracellular location for GPR30 was found for MCF-7, SKBR3, and MDA-MB-231 cells. Exposure of COS-7 cells bearing GPR30 to estradiol-17β causes a mobilization of intracellular Ca2+. Because GPR30 had been shown by Filardo et al. (2000
, 2002)
to activate the MAPK pathway via EGFR and mobilization of intracellular Ca2+ is mediated by phospholipase C (PLC)-dependent inositol-1,4,5-trisphosphate production, an experiment was conducted to determine whether either of these pathways was involved in estrogen-stimulated Ca2+ mobilization by ER
and GPR30. The GPR30-mediated Ca2+ mobilization was completely blocked by an inhibitor of EGFR, but not by an inhibitor of PLC, whereas the ER
response was completely blocked by the PLC inhibitor, but not measurably so by the EGRF inhibitor. These data suggest that ER
- and GPR30-provoked Ca2+ mobilization are mediated by distinctly different pathways.
Because of equivocal data regarding the cellular location of GPR30, Filardo et al. (2007)
conducted a series of experiments to prove that GPR30 is present in the plasma membrane. For these experiments, HEK-29 cells transfected with a hemagglutin-epitope tag incorporated at the amino terminus of human GPR30 and the SKBR3 breast cancer cells were utilized. Using a variety of techniques, including confocal microscopy, western blotting of subcellular fractions, estrogen binding, and estradiol activation of G-protein in cell compartments, these investigators provided strong evidence that GPR30 is found in the plasma membrane. However, it should be noted that these cancer cells were, in some experiments, exposed to exceedingly high concentrations of estradiol.
Rapid Estrogen-Induced Responses by the Pituitary
Administration of estradiol-17β to ovariectomized ewes resulted in an almost immediate reduction in systemic concentration of LH, followed in approximately 12 h by an ovulatory-type surge of the gonadotropin (Nett et al., 1984
). The almost instantaneous response to estrogen suggested an inhibitory effect on the hypothalamus or hypophysis and precluded it being the result of the traditional effect of the steroid on gene expression. Recognizing that this response must be due to an extranuclear cellular site of action, Arreguin-Arevalo and Nett (2005)
began to explore the mechanism by which estradiol-17β was able to evoke such a rapid response by the pituitary. Using a primary culture of ovine pituitary cells, these investigators showed that preincubation of cells for 15 min with estradiol-17β or estradiol-17β conjugated to BSA or a peptide (PEP) prevented GnRH-induced secretion of LH (Figure 2
). This was a specific response to estradiol-17β because pretreatment of cells with P4, testosterone, hydrocortisone, or estradiol-17
failed to affect secretion of LH by themselves, and did not suppress GnRH-induced secretion of LH. Further, when pituitary cells were coin-cubated with these steroids plus estradiol-17β, they did not block the ability of estradiol-17β to inhibit GnRH-induced release of LH. Treatment of cells with tamoxifen, hydroxytamoxifen, or ICI 182,780 blocked inhibition by estradiol-17β (E2) and the estradiol-17β conjugates (E2-BSA, E2-PEP). When cells were treated with selective ER agonists, the ER
-agonist (propylprazoletriol), but not the ERβ agonist (diarylpropionitrile), decreased the GnRH-induced secretion of LH. The fact that E2-BSA and E2-PEP mimicked the action of estradiol-17β suggests that this effect was mediated by an ER associated with the plasma membrane.
|
|
|
Rapid Estrogen Signaling in the Vascular System
As mentioned in an earlier section of this review, ovarian steroids have been shown to exert a local effect on the uterine arteries of domestic ruminants. In particular, the presence of estrogen was found to markedly increase arterial dilation and blood flow. Simoncini et al. (2000)
reported that estrogen activated the phosphatidylinositol-3-kinase (PI3K)-Akt/protein kinase B pathway (Figure 1
), with consequent activation of endothelial nitric oxide synthase (eNOS). Using human vascular endothelial cells, these investigators demonstrated that ER
binds in a ligand-dependent manner to the p85
regulatory subunit of PI3K. Stimulation with estrogen increases ER
-associated PI3K activity. This kinase catalyzes the phosphorylation of phosphatidylinositol-4,5-bisphosphate to form phosphatidylinositol-3,4,5-trisphosphate. The phosphatidylinositol-3,4,5-trisphosphate associates with 3-phosphoinositide-dependent protein kinase-1, a protein kinase that phosphorylates Akt/protein kinase B and leads to enhanced activity of eNOS with a resultant increase in NO, which causes relaxation of vascular smooth muscle. Estrogen receptor β did not interact with p85
or recruit PI3K activity after estrogen stimulation. To determine whether estradiol-17β-stimulated eNOS activation was mediated by Akt, the investigators transiently transfected bovine aortic endothelial cells with adeno-viruses containing constitutively active and dominant negative Akt mutants. Transfection of these endothelial cells with the constitutively active Akt resulted in a substantial increase in eNOS activity, whereas overexpression of the dominant negative Akt completely abolished estradiol-17β-stimulated eNOS activity. The short duration over which endothelial cells responded to estradiol-17β suggested that recruitment and activation of PI3K was by ligand-bound ER
but independent of gene transcription. In contrast to these observations, Chen et al. (2004)
reported that estrogen-induced activation of eNOS in ovine uterine arterial endothelial cells (UAEC) occurred by a different pathway. According to these investigators, exposure of UAEC to estradiol increased eNOS activity and NO production within 2 min as a consequence of the estradiol binding to ER
in the plasma membrane. Exposure of UAEC to estradiol resulted in rapid phosphorylation of ERK-1/-2, which was maximal within 5 min after treatment of cells with the estrogen. Treatment of cells with the ER antagonist ICI 182,780 or PD98059, an inhibitor of the MAPK pathway, abolished estradiol-17β-induced phosphorylation of ERK-1/-2 and eNOS activity. However, exposure of UAEC to the PI3K inhibitor LY294004 failed to prevent estradiol-induced production of NO by these cells. The investigators concluded that rapid activation of eNOS to produce NO in ovine UAEC by estrogen is mediated, at least in part, through the MAPK pathway with activation of ERK, and not via activation of the PI3K pathway. The production of NO would result in rapid uterine vasodilation by estrogen, a phenomenon shown to occur in vivo.
| NONGENOMIC ACTIONS OF PROGESTINS |
|---|
An SH3-interacting domain motif located in the N-terminus between AA residues 421 to 428 (near the AF-1 domain) was discovered by Boonyaratanakornkit et al. (2001)
in both nPR-B and nPR-A. The SH3-interacting domains are polyproline sequence motifs characteristic of a class II consensus PEP for Src kinase-like SH3 domains. Polyproline sequences form a left-handed helix that recognizes the hydrophobic pocket of SH3 domain-containing proteins. In coimmunoprecipitation assays, the SH3 domain of nPR interacted with a wide variety of Src Tyr kinase family members, including Src Tyr kinase (Edwards et al., 2003
). The SH3 domain of Src recognizes the polyproline sequence, and this interaction participates in the activation of the Tyrkinase activity of Src. The Src Tyr kinase activates the MAPK cascade, which ultimately results in the phosphorylation or activation of several transcription factors, including c-fos and c-jun. Activation of the MAPK cascade rapidly induces transcription of cyclin proteins, which then cause the cell to enter into mitosis. Boonyaratanakornkit et al. (2007)
, using breast cancer epithelial cells, determined that nPR-B was the isoform that preferentially activated Src and MAPK, and was important for inducing expression of the cyclin D1 gene. Mutated nPR-B (mutated in the proline-X-X-proline motif) was not able to induce cell-cycle progression or expression of cyclin D1. Although these responses seem to be genomic, because they occur within a few hours via activation of kinase cascades, they are considered to be nongenomic functions of the steroid receptor because the receptor itself does not interact with the DNA. These experiments were confirmed by the research of Faivre et al. (2005)
, who observed an induction of phosphorylated ERK-1/-2 by treatment with progestin R5020 within 5 to 10 min, and activated MAPK was required for progestin to stimulate cyclin D1 and cdk 2 expression in MCF-7 and T47D breast cancer cells, both of which have high expression of nPR. Subsequent research by this group provided evidence for a biphasic response to P4 treatment, with both P4-dependent upregulation of the EGF/Wnt-1 pathway in addition to the MAPK pathway needed to induce P4-mediated proliferation of progestin-responsive breast cancer cells (Faivre and Lange, 2007
).
The PI3K pathway may be induced by an isoform of nPR to mediate oocyte maturation and germinal vesicle breakdown in frogs (Xenopus laevis). The Xenopus PR is related to the mammalian PR-B isoform and is termed XPR-1. The XPR-1 has been observed localized to the plasma membrane of oocytes, and treatment with P4 can activate PI3K within 30 min (Bagowski et al., 2001
). However, there is evidence that P4-mediated induction of the G-protein subunits β
actually induce Xenopus oocyte maturation via Gβ
stimulation of adenylate cyclase (Guzmán et al., 2005
). A recent study implicated classical steroid hormone receptors for both P4 and testosterone in signaling through Gβ
in Xenopus by pharmacological methods, that is, by using the specific ligands R5020 and R881 (Evaul et al., 2007
). However, another study by Josefsberg et al. (2007)
demonstrated that a recently discovered membrane PR β (mPRβ) ortholog, XmPRβ, mediates oocyte maturation in Xenopus. These researchers used microinjection of a neutralizing antibody to XmPRβ into the oocyte and could block the P4-induced maturation. There was also an induction of MAPK within 10 min after transfected (XmPRβ) CHO cells were exposed to 30 nM P4. The effect of XmPRβ on Gβ
signaling was not investigated. Thus, the exact protein that mediates rapid P4-initiated events in Xenopus oocytes remains to be determined. It may be that all of the above signaling cascades are necessary, or this may represent a redundancy in this important system.
A novel membrane-associated PR has been identified that is functionally and structurally distinct from genomic nPR: a G-protein-coupled receptor first characterized in spotted seatrout ovaries (Thomas et al., 2002
). In the spotted seatrout, a progestin (17,20β,21-trihydroxy-4-pregnen-3-one) rapidly mediates the induction of oocyte maturation (Thomas et al., 2002
). This action occurs at the cell surface, is not blocked by inhibiting transcription, and is induced rapidly within a few minutes of progestin administration, suggesting these actions are not mediated by an nPR. Zhu et al. (2003b)
described this protein: a 352-AA protein with a molecular weight of 40 kDa with 7 transmembrane domains characteristic of a G-protein-coupled receptor. Homologous genes for this mPR were found to exist in humans, mice, and swine via a BLAST search (Zhu et al., 2003a
). Similar proteins have been identified by other researchers in human myometrium (Fernandes et al., 2005
) and in ovine hypothalamus, pituitary, uterus, ovary, and corpus luteum (Ashley et al., 2006
). There are some conflicting data that suggest these proteins cannot mediate plasma membrane events (Ashley et al., 2006
; Krietsch et al., 2006
).
Novel mPR isoforms have been shown to decrease cAMP levels in Atlantic croaker (Micropogonias undulates) oocytes, and germinal vesicle breakdown (GVBD; a marker of oocyte maturation) was blocked by inhibition of the PI3K/Akt pathway (Pace and Thomas, 2005
). Although MAPK activation was observed in croaker oocytes, blockage of MAPK activation did not inhibit GVBD, suggesting that signaling via PI3K/Akt is the primary mechanism by which GVBD occurs in response to P4 in croaker oocytes.
Signaling through the ovine mPR in transfected CHO cells demonstrated rapid (within 1 min) increases in intracellular calcium (Ashley et al., 2006
). When cellular localization was studied, the ovine mPR localized to the endoplasmic reticulum. Because the increase in intracellular calcium was via liberation of cellular calcium stores, the mPR could function in this manner to stimulate rapid steroid responses. It is notable, however, that some researchers claim the mPR is localized to the plasma membrane and mediates plasma membrane-associated responses (Thomas, 2003
; Zhu et al., 2003b
; Pace and Thomas, 2005
). Other studies in transfected cells in addition to that by Ashley et al. (2006)
have visualized the mPR localized to either the endoplasmic reticulum (human mPR
, β, and
; Krietsch et al., 2006
) or to an intracellular tubular network (human mPR
isoform only; Fernandes et al., 2005
). There may be species differences in the cellular localization (and signaling pathways) of these proteins. Interestingly, Krietsch et al. (2006)
duplicated the materials and methods of Zhu et al. (2003b)
and were unable to stimulate either cAMP production or MAPK activity with transfected human mPR and both pufferfish and spotted seatrout mPR sequences (seatrout mPR sequence was provided by Zhu et al. 2003b
). Because the seatrout mPR sequence was amplified by PCR before inserting into the vector, some small mutations may have been induced that affected the protein sequence of the transfected seatrout mPR.
Membrane P4-binding proteins have been identified in spermatozoa, bovine follicular and luteal cell membranes, and murine granulosa cells by using the c-terminal PR antibody c-262 (Peluso et al., 2001
; Bramley et al., 2002
; Luconi et al., 2002
). There is some debate regarding whether the P4-binding protein in spermatozoa could be nPRC (Luconi et al., 2002
).
Peluso et al. (2001
, 2003)
later characterized the P4-binding protein in the membranes of murine granulosa cells. Progesterone prevents these cells from entering into an apoptotic pathway. Because these cells in the rat ovary do not express nPR before the preovulatory surge of gonadotropin, it is unlikely the binding protein is nPR. Peluso et al. (2004)
cloned this P4-binding protein and found it to consist of 407 AA, similar to a homolog expressed in the mouse lung. After a domain analysis, this protein was termed plasminogen activator inhibitor mRNA binding protein-1 (PAIRBP-1) for its homology to other proteins (Peluso et al., 2006
). The analysis provided no evidence of a transmembrane domain, but did identify several hyaluronic acid binding sites. Further investigation of PAIRBP-1 by Peluso et al. (2006)
led to the identification of a membrane-bound partner, PR membrane component-1 (PGRMC-1), that is required for PAIRBP-1 to exert the effects mediated by P4 on these cells. The PGRMC-1 is a relatively small protein (28 kDa) and possesses a short N-terminal extracellular domain, a single transmembrane domain, and a cytoplasmic domain that contains several potential SH2 and SH3 domains. The PGRMC-1 protein is homologous to the IL-6 receptor, a member of a growth factor receptor superfamily that includes the cytokines, growth hormone, and prolactin. Progesterone receptor membrane component-1 has been observed as a dimer (56 kDa), but can form aggregates up to 200 kDa. When spontaneously immortalized granulosa cells are treated with an antibody that blocks the N-terminal extracellular domain, treatment with P4 will not result in a decrease in the total number of apoptotic spontaneously immortalized granulosa cells (Peluso et al., 2006
).
Rapid Progestin Inhibition of the Hypothalamus
The first evidence of a nongenomic action of P4 in the brain was reported in the 1940s by Hans Selye (1942)
, who observed an anesthetic effect almost immediately after administration of a bolus of P4 to rats. Researchers in 1959 later observed depressed electroencephalogram activity in female rabbits following the postcoitis rise in P4 (Sawyer and Kawakami, 1959
).
Later research demonstrated in the hypothalamus that allopregnanolone (3
,5
-tetrahydroprogesterone) enhances the inhibitory effect of gamma-aminobutyric acid (GABA) on GABA type A (GABAA) receptor-expressing neurons, possibly through P4 secreted from neighboring glial cells (Rønnekleiv and Kelly, 2005
).
Evidence suggests that P4 is rapidly metabolized in the brain, and the metabolites (i.e., allopregnanolone) actually mediate the rapid effects of P4 in the brain. However, there is also evidence for a direct effect of P4 in the hypothalamus. Hamsters primed for 2 h with P4, upon receiving a bolus injection of P4 (or P4 conjugated to BSA) directly into the midbrain ventral tegmental area rapidly exhibit lordosis behavior (within 10 min; DeBold and Frye, 1994
). Progesterone administered concurrently with estradiol-17β in the hypothalamus/ preoptic area of the ewe can rapidly inhibit the estradiol-induced preovulatory surge of GnRH and LH (Richter et al., 2005
). These actions are presumed to be mediated by the nPR, because of the observations of Skinner et al. (1998)
that, upon pretreatment with RU 486 (an nPR antagonist), P4 can no longer inhibit estradiol-induced GnRH secretion. In ovariectomized estradiol-primed ewes, removal of a P4 implant will rapidly increase the frequency of pulsatile GnRH release (within 175 min), and these effects are seen only in estradiol-primed females. Further investigation by Richter et al. (2005)
revealed that the inhibition by P4 was most likely at the level of estradiol signaling, and not at the level of GnRH release, because treatment with P4 after the onset of the estradiol-induced surge did not inhibit GnRH release.
Expression of OT neurons in the supraoptic nucleus and the paraventicular nucleus of the hypothalamus are under control of GABA via GABAA receptors in these neurons (Brussaard et al., 2000
). Treatment of OT neurons with GABA decreases the rate of firing, seen as an increase in the amount of time required to resensitize Cl– channels. During late pregnancy in female rats, metabolites of P4 have been shown to inhibit firing of OT neurons by preventing protein kinase C-induced modulation of GABAA receptors. During parturition, GABAA receptors are less sensitive to P4 metabolites, and counteraction of protein kinase C no longer occurs. This is accompanied by an increase in firing activity of OT neurons at the time of parturition (Brussaard et al., 2000
).
In some reports, the short time between exposure to P4 and the detected hypothalamic response suggest that the effects of P4 may be mediated by membrane-associated steroid binding proteins. Results of research involving the use of PR antagonists (see Skinner et al., 1998
) revealed that the membrane-associated binding protein may in fact be the nPR. The true identity of the hypothalamic membrane progestin binding protein(s) remains to be determined.
Rapid Progestin Inhibition of Uterine OTR Activation
In studies of the bovine endometrium, a relatively high molar concentration of P4 (10–5 M) was able to inhibit the OT-stimulated release of PGF2
from dissociated endometrial epithelial cells in the presence of actinomycin D (which blocks RNA synthesis from DNA; Bogacki et al., 2002
). In these cells, P4 blocked both the binding of OT to its receptor and OT-stimulated Ca2+ increase. Various steroids were then investigated (progestins and testosterone at 10–5 M; Duras et al., 2005
) for their ability to inhibit OT signaling. All steroids investigated inhibited OT-induced PGF2
secretion and intracellular Ca2+ response. The observed cellular responses were attributed to a nongenomic action of steroids. But the authors did acknowledge the possibility that such responses might actually represent artificial, nonspecific effects of the steroids because of a long (4-h) incubation period and a comparatively high concentration of steroids utilized.
This premise was tested by Burger et al. (1999)
, who investigated the effects of progestins at a range of doses between 10 to 200 µM on human OTR transfected into various cell lines. These researchers reported that their data demonstrated that the effects of P4 on OT binding and OTR function were nonspecific steroid effects, that is, a steroid overloading of the plasma membrane, which then interferes with the ability of the OTR to interact with signaling proteins. Wenz and Barrantes (2003)
, using artificial membrane bilayers, tested the effects of various steroids, including P4, on the integrity of lipid domains. Their data indicated that the lower the hydrophobicity of the steroid (hydrophobicity determined by the group bound to carbon 17), the more lipid domain-disrupting activity the steroid displayed at higher molar concentrations. Progesterone and all naturally occurring metabolites of P4 that possess low hydrophobicity groups attached to carbon 17 demonstrated domain-disrupting activity in the artificial lipid domains, but only at high molar concentrations (10µ M P4).
In ovine endometrial plasma membranes, a lower, more physiological concentration of P4 (16 nM) can specifically inhibit the binding of OT to the OTR (Figure 5
; Dunlap and Stormshak, 2004
). The fact that RU 486 (antagonist) can block the inhibiting effect of P4 on binding of OT suggests the possibility that membrane-localized nPR may somehow be associated with the OTR. Inhibition of OT binding by P4 may reflect an induced allosteric alteration in the structural conformation of the OTR monomer or dimer, which is directly or indirectly relieved or prevented by an antagonist to the nPR such as RU 486. Similarly, inhibition of OT binding by P4 and progestins at physiological levels (below 1 µ M) of P4 has been reported from studies of the murine and the human OTR (Grazzini et al., 1998
). Specific and saturable membrane binding of progestins has been characterized in the ovine endometrium, and the binding sites for OT and progestin appear to be closely associated in this tissue (Dunlap and Stormshak, 2004
). In the ewe, the inhibition of OT binding to membrane preparations by low levels of P4 (2 to 16 nM) is dose dependent (Figure 6
; Bishop and Stormshak, 2006
). As stated above, there is evidence suggesting that P4 inhibits OT binding either by competing with OT for its receptor site or by binding to a closely associated protein that alters OTR conformation. Bishop and Stormshak (2006)
demonstrated that preexposure of ovine endometrial explants to P4 (8 nM) rapidly inhibited OT-induced phosphoinositide hydrolysis (within 30 min; Figure 7
) and OT-induced PGF2
production (within 2 h). Thus, there is evidence in the ruminant endometrium for a physiologically relevant nongenomic action of P4 to interfere with the ability of OT to signal through the OTR. The data generated from the bovine model (Bogacki et al., 2002
; Duras et al., 2005
) may represent an artificial, in vitro effect of steroid as suggested by Burger et al. (1999)
. However, the effects of lower concentrations of P4 in the ovine endometrium (Dunlap and Stormshak, 2004
; Bishop and Stormshak, 2006
) may represent a more specific effect because progestins display lipid-domain effects at high molar concentrations only (Wenz and Barrantes, 2003
). The ovine endometrium also displays specific and saturable binding to P4 by the radioligand binding assay (Dunlap and Stormshak, 2004
). The ovine mPR has been observed in the uterus of the ewe (Ashley et al., 2006
), but whether this protein is involved in the inhibition of the OTR remains to be investigated. The fact that this protein appears to be localized to the endoplasmic reticulum makes a direct protein-protein interaction between the mPR and the OTR unlikely, but the mPR may mediate other downstream signaling events that interfere with OTR function.
|
|
|
-hydroxyprogesterone and 5β-hydroxyprogesterone displayed 10- to 20-fold lower affinity for the binding site (ruling out a GABAA receptor involvement; Rae et al., 1998b
Treatment of diestrus-II rat uteri with P4 (40 µg/mL) and metabolites of P4 (3β-hydroxy-5β-pregnan-20-one, 5β-pregnan-3,20-dione, and 3
-hydroxy-5
-pregnan-20-one) will inhibit uterine contractions within 2 min (Putnam et al.,1991
). The effects of P4 only are blocked by treatment with RU 486, but a GABAA receptor antagonist (picrotoxin) will block the inhibitory effects of the metabolites. Picrotoxin does not affect the ability of P4 to inhibit contractions. This suggests that during gestation in the rat, P4 metabolites may act through a GABAA receptor system to inhibit uterine contractions (Putnam et al., 1991
).
It has been reported that P4 may bind to an mPR in bovine T lymphocytes to prevent proliferation of stimulated cells via inhibition of the
subunit of the IL-2 receptor (Cannon and Pate, 2004
). More recently, Ndiaye et al. (2007)
demonstrated the presence of mPR
, β, and
and PCRMC-1 but not nPR, in T lymphocytes.
Spermatozoa respond to stimulation by P4 and they lack a transcriptionally active nucleus. These cells lack both nPR-A and nPR-B, but P4 treatment can stimulate calcium influx, Tyr phosphorylation of proteins, efflux of chloride, and increases in cAMP (Luconi et al., 2002
). In the travels of spermatozoa to the oocyte, they do come into contact with increased concentrations of P4 within the cumulus oophorus (1 to 10 µg/mL of P4 has been measured in cells surrounding human oocytes; Luconi et al., 2002
). Progesterone stimulates capacitation of human spermatozoa and induces the acrosome reaction in caprine spermatozoa (Somanath et al., 2000
). Possible candidates for a PR have been identified by using the c-terminal PR antibody c-262, as mentioned before (Luconi et al., 2002
), and the 56- to 57-kDa protein may be similar to the one identified by Peluso et al. (2001)
as being present in the ovary. However, characterization of this protein has yet to be performed. There is some discussion about whether this putative PR could be another isoform of the genomic PR, the PR-C form (Luconi et al., 2002
).
| BIOLOGICAL SIGNIFICANCE |
|---|
Responses to the nongenomic action of steroids may also be so subtle as to be nondetectable. As an example, there is evidence that steroids may act nongenomically to sensitize intracellular systems for the purpose of enhancing their subsequent genomic transcriptional efficiency. This possibility is supported by the data of Vasudevan et al. (2001)
, who subjected SK-NE-BE(2)C neuroblastoma cells transfected with human (h) ER
to a 2-pulse exposure of estrogen. Cells were first exposed for 20 min to 10–9 M estradiol-17β-BSA, followed after 4 h by exposure to a second pulse of 10–9 M estradiol-17β. This treatment sequence significantly increased transcriptional expression of hER
as compared with controls or cells having been exposed to only estradiol-17β (E2). A reverse-pulse sequence in which cells were first exposed to E2 and then E2-BSA did not potentiate transcription, demonstrating that E2-initiated membrane events are essential for later transcriptional effects. Of course, it is possible that the apparent beneficial nongenomic effect of estradiol-17β is attributed to free E2 having been leached off the E2-BSA, and thus entering the cell to directly initiate the transcription processes 4 h prior to exposure of cells to a larger concentration of E2. The result of this latter scenario could be an enhanced transcription of the hER
.
In conclusion, the literature is replete with a preponderance of reports describing the nongenomic actions of estrogens and progestins in transfected or cancer cells. However, although these studies have been instrumental in promoting interest in the field, there is a need for further research to document the importance of the nongenomic action of steroids to the biology of the living animal.
| Footnotes |
|---|
1 Corresponding author: Fred.Stormshak{at}oregonstate.edu
Received for publication August 2, 2007. Accepted for publication October 24, 2007.
| LITERATURE CITED |
|---|
secretion in the ovine endometrium. Endocrinology 147:937–942.
from bovine endometrial tissue. Biol. Reprod. 67:184–188.
and E2 secretion from bovine endometrial cells. Prostaglandins Other Lipid Mediat. 76:105–116.[Medline]
-mediated signaling in Xenopus laevis oocytes by signaling through classical steroid receptors. Mol. Endocrinol. 21:186–196.
and periarterial sympathetic nerve stimulation. Biol. Reprod. 15:58–65.[Abstract]
stimulated adenylyl cyclase is involved in Xenopus laevis oocyte maturation. J. Cell. Physiol. 202:223–229.[CrossRef][Medline]
transcriptional activity and is a key regulator of the cellular response to estrogens and antiestrogens. Endocrinology 140:5566–5578.
, β, and
) localize to the endoplasmic reticulum and are not activated by progesterone. Mol. Endocrinol. 20:3146–3164.
subunit and luteinizing hormone β genes in the ovine anterior pituitary. J. Biol. Chem. 258:12087–12090.
-DHT and 17β-estradiol receptors in the ram hypothalamus. J. Steroid Biochem. 14:603–611.[CrossRef][Medline]
and ERβ expressed in Chinese hamster ovary cells. Mol. Endocrinol. 13:309–319.
to the plasma membrane. Proc. Natl. Acad. Sci. USA 101:2075–2081.
-Shc association and Shc pathway activation. Mol. Endocrinol. 16:116–127.This article has been cited by other articles:
![]() |
K. Y. Chung and B. J. Johnson Application of cellular mechanisms to growth and development of food producing animals J Anim Sci, April 1, 2008; 86(14_suppl): E226 - E235. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |