|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Biomedical Science, Colorado State University, Fort Collins, Colorado 80523
ABSTRACT
The objectives of this study were to determine whether activation of estrogen receptor 1 (ESR1; also known as ERalpha), or estrogen receptor 2 (ESR2; also known as ERbeta), or both are required to: 1) acutely inhibit secretion of LH, 2) induce the preovulatory-like surge of LH, and 3) inhibit secretion of FSH in ovariectomized (OVX) ewes. OVX ewes (n = 6) were administered intramuscularly 25 micrograms estradiol (E2), 12 mg propylpyrazoletriol (PPT; a subtype-selective ESR1 agonist), 21 mg diaprylpropionitrile (DPN; a subtype-selective ESR2 agonist), or PPT + DPN. Like E2, administration of PPT, DPN, or combination of the two rapidly decreased (P < 0.05) secretion of LH. Each agonist induced a gradual, prolonged rise in secretion of LH after the initial inhibition, but neither agonist alone nor the combined agonists was able to induce a "normal" preovulatory-like surge of LH similar to that induced by E2. Compared with E2-treated ewes, the beginning of the increase in secretion of LH occurred earlier (P < 0.01) in DPN-treated ewes, later (P < 0.05) in PPT-treated ewes, and at a similar interval in ewes receiving the combined agonist treatment. Like E2, PPT decreased (P < 0.05) secretion of FSH, but the duration of suppression was much longer in PPT-treated ewes. DPN did not alter secretion of FSH in this study. Modulation of the number of GnRH receptors by PPT and DPN was examined in primary cultures of ovine pituitary cells. In our hands, both PPT and DPN increased the number of GnRH receptors, but the dose of DPN required to stimulate synthesis of GnRH receptors was 10 times higher than that of PPT. We conclude that in OVX ewes: 1) ESR1 and ESR2 mediate the negative feedback of E2 on secretion of LH at the level of the pituitary gland, 2) ESR1 and ESR2 do not synergize or antagonize the effects of each other; however, they do interact to synchronize the beginning of the stimulatory effect of E2 on secretion of LH, 3) ESR1 and ESR2 may mediate at least partially the positive feedback of E2 on LH secretion by increasing the number of GnRH receptors, and 4) only ESR1 appears to be involved in the negative feedback of E2 on secretion of FSH.
anterior pituitary,, estradiol,, estradiol receptor subtypes,, follicle-stimulating hormone,, luteinizing hormone
Estradiol (E2) regulates many physiologic processes in the reproductive system [1–3] through at least two main estrogen receptors (ERs), estrogen receptor 1 (ESR1) and estrogen receptor 2 (ESR2) (also known as ER
and ERß, respectively). These receptors are widely distributed in the reproductive axis [4–8]. ESR1 is the more abundant receptor in most reproductive tissues, particularly the uterus and pituitary of human and rat [5, 6, 9–13]. The ESR2 isoform has not been reported in ovine pituitaries, but it has been localized in rat gonadotropes [9]. In brain and hypothalamus of rat, mRNAs for Esr1 and Esr2 are expressed in a partially overlapping pattern [5, 6, 11, 14]. Apparently, only mRNAs for Esr2 and its protein are localized within rodent GnRH neurons [15–17]. During recent years, indirect but increasing amounts of evidence strongly support the presence of binding proteins distinct from ESR1 and ESR2 isoforms that mediate actions of E2 in the brain [11, 18, 19].
Information generated in Esr1–/– knockout (alpha ERKO), Esr2–/– knockout (beta ERKO), and double Esr1–/–/Esr2–/– knockout (alpha/beta ERKO) mice has led to the general consensus that ESR1 is the primary receptor through which E2 mediates its negative feedback on gonadotropin secretion [1, 20–23]. The use of ER-selective ligands has confirmed the importance of ESR1 in the regulation of the hypothalamic-pituitary axis [24–31]; however, use of these ligands has also revealed the roles of ESR2 in various systems, including the hypothalamic-pituitary axis [28, 29, 32, 33]. A selective effect of E2 acting exclusively though ESR1 [25, 31, 34, 35] or ESR2 [24, 31, 33] has been demonstrated using propylpyrazoletriol (PPT, a subtype-selective ESR1 agonist), diaprylpropionitrile (DPN, a subtype-selective ESR2 agonist), and other ER-selective ligands. Moreover, some ESR1-mediated actions of E2 can be reduced [34], enhanced [34, 36], or not modified [24, 34] by activation of ESR2. The evaluation of these ER-selective ligands includes their ability to mimic, in vitro, nongenomic actions of E2 [36, 37]. The ability of PPT and DPN to cross the blood-brain barrier and elicit a biologic response has also been demonstrated. Systemic administration of PPT upregulated progesterone receptor (Pgr) mRNA [38] and PGR immunoreactivity [24] in rat hypothalamus and elicited a neuroprotective effect in mouse dopaminergic neurons [35]. Similarly, systemic administration of DPN produced an anxiolytic behavior in rats [24, 31] and bound to nuclear ER in brain within 30 min of administration [24].
Our knowledge of the actions of E2 in the hypothalamic-pituitary axis has grown exponentially during the last decade; however, the participation of the main ER subtypes in the regulation of gonadotropin secretion in ewes is practically unexplored. It has been well established that administration of E2 to ovariectomized (OVX) ewes rapidly (within minutes) decreases secretion of LH, followed several hours later by a preovulatory-like surge of LH [39–41]; in contrast, secretion of FSH does not decrease until several hours after administration of E2 [41–43]. The acute nongenomic inhibition in secretion of LH occurs at least in part by a direct action of E2 at the pituitary gland [39, 41] by a mechanism mediated via ERs [37]. It is equally well established that the preovulatory-like surge of LH is the result of the actions of E2 at the pituitary, increasing the number of GnRH receptors [39, 44, 45] and the hypothalamus, inducing a massive release of GnRH [46, 47]. Our goal was to establish the roles of ESR1 and ESR2 in this endocrine paradigm by using ER-selective ligands in OVX ewes. The specific objectives of this study were to determine whether activation of ESR1, ESR2, or both is required to: 1) acutely inhibit the secretion of LH, 2) induce the preovulatory-like surge of LH, and 3) inhibit secretion of FSH in OVX ewes. In addition, the roles of ESR1 and ESR2 in the increase of GnRH receptors number induced by E2 in primary cultures of ovine pituitary cells were evaluated.
Relative Binding Affinity of Subtype-Selective ER Agonists
To estimate the initial dose of subtype-selective ER agonists, relative binding affinities (RBAs) of these synthetic ligands were estimated. The RBAs of PPT and DPN were compared to the binding affinity of E2 in the cytosolic fraction of ovine uterus using [2,4,6,7,16,17-3H]17-ß-estradiol ([3H]-E2; 141 Ci/mmol; DuPont NEN, Boston, MA). Tissue processing and receptor assay were carried out as previously reported by Amann et al. [48]. Briefly, uterine tissue was weighed and homogenized in ice-cold TEDG buffer (10 mM Tris-HCl, 1.5 mM EDTA; 2 mM dithiothreitol, and 10% [v/v] glycerol; pH = 7.5) using a polytron homogenizer (1 g tissue in 10 ml TEDG). Homogenates were centrifuged at 30 000 x g for 60 min at 4°C, and the supernatant (cytosol) was harvested. Binding capacity of ER in the cytosol was determined using an exchange assay. Duplicate 200-µl aliquots of cytosol were incubated with equal volumes of buffer containing 4 nM [3H]-E2 with or without a serial dilution of E2, PPT, or DPN. E2 was purchased from Sigma-Aldrich Inc. (St. Louis, MO), and agonists of ERs were purchased from Tocris Cookson Inc. (Ellisville, MO). Aliquots containing cytosol, 4 nM [3H]-E2, and 100-fold excess E2 was used to estimate nonspecific binding. After incubation for 60 min at 4°C, ice-cold, dextran-coated charcoal was added. Samples were centrifuged at 3000 x g at 4°C for 15 min, and the amount of [3H]-E2 remaining in the supernatant (i.e., bound to receptor) was determined. The RBA of the subtype-specific agonists of E2 was calculated by dividing the molar concentration of E2 needed to reduce specific [3H]-E2 binding by 50% by the molar concentration of PPT or DPN that reduced specific [3H]-E2 binding by 50% (RBA = IC50 E2/IC50 test compound x 100). Binding affinity of E2 was arbitrarily assigned a value of 100% [49].
Preparation of Media and Stock Solutions
Pituitary dissociation medium consisted of 137 mM NaCl, 5 mM KCl, and 25 mM HEPES (Sigma-Aldrich Inc.), pH 7.3, plus an enzymatic cocktail containing 1.0 mg/ml collagenase (type II), 1.0 mg/ml hyaluronidase (type V), and 0.02 mg/ml deoxyribonuclease. Enzymes were freshly prepared immediately prior to dissociation. Culture medium consisted of Dulbecco modified Eagle medium (DMEM; Sigma-Aldrich Inc.) supplemented with 10% charcoal-stripped fetal bovine serum (FBS), streptomycin sulfate (500 mg/ml), potassium penicillin G (313 mg/ml), and 2.2 g/L NaHCO3. Dissociation and culture medium as well as the enzymatic cocktail were sterilized by filtration through 0.2-µm millipore membranes (Fisher Science, Denver, CO). E2, PPT, or DPN was dissolved in ethanol on the day of treatment.
Dissociation and Incubation of Pituitary Cells
All procedures involving animals were approved by the Colorado State University Animal Care and Use Committee and complied with National Institutes of Health (NIH) guidelines. Anterior pituitary glands were collected from OVX ewes during January 2006 and February 2006. Ewes were anesthetized with sodium pentobarbital followed by exsanguination. Tissues were removed and immediately placed in ice-cold dissociation medium. Anterior pituitary tissue from OVX ewes was dispersed as described by Adams et al. [50], with the omission of trypsin digestion. Briefly, tissue was sectioned into 0.5-mm-thick slices with a Stadie-Riggs tissue slicer and washed five times with dissociation medium without enzymes. Tissue was incubated in dissociation medium containing the enzymatic cocktail at 37°C in a Dubnoff metabolic shaker for 90 min, and every 10 min the cell suspension was passed through a Pasteur pipette. After dissociation, the cell suspension was washed (400 x g; 4 min) five times with dissociation medium without enzymes, resuspended in phenol red-free DMEM containing 10% charcoal-stripped FBS, and plated at 3 x 106 cells per well in six-well tissue culture plates (Sarsteadt Inc., Newton, NC). Cells were incubated for 24 h at 37°C under an atmosphere of 95% air:5% CO2 prior to addition of treatments.
The number of GnRH receptors was determined in ovine pituitary cells with modifications as described by Gregg et al. [44]. Dispersed pituitary cells were plated in duplicate for each treatment group in six-well dishes. Cells were washed three times with ice-cold phosphate buffer. Following the final wash, cells were scraped from wells, transferred to 12 x 75-mm plastic tubes and centrifuged at 1500 x g at 4°C for 10 min. The cell pellets were resuspended in Tris buffer (10 mM tris(hydroxymethyl)aminomethase [Tris]; 1 mM CaCl2; 0.1% bovine serum albumin; pH 7.4) to a final concentration of 1 x 106 cells per 100 µl. Briefly, 100-µl aliquots of the cell suspension were transferred to 12 x 75-mm plastic tubes containing 125I-labeled [D-Ala6]GnRH (40 pM). Nonspecific binding for each cell suspension was determined with the addition of 40 nM unlabeled [D-Ala6]GnRH. After incubation at 4°C for 4 h, 3 ml Tris buffer was added, and tubes were centrifuged for 10 min at 1500 x g. The supernatants were decanted, and radioactivity in the cell pellet was determined. The binding of [D-Ala6]GnRH per treatment was expressed as percentage of binding compared to control (100%).
Experiment 1. In a preliminary study, administration of PPT or DPN to OVX ewes at a dose based on their RBA did not induce a clear inhibition in secretion of gonadotropins; therefore, for this study the dose of PPT and DPN used was five times more than the estimate based on their RBA. Between October 2004 and January 2005, 24 mature Western-range ewes that had been OVX for at least 2 mo were randomly distributed to receive an intramuscular injection of 25 µg E2 (n = 6), 12 mg PPT (n = 6), 21.1 mg DPN (n = 6), or both PPT plus DPN (n = 6). The compounds were freshly dissolved in ethanol the day of treatment and emulsified in 3 ml mineral oil. All ewes were fitted with an indwelling jugular cannula (Angiocath; Becton Dickinson, Sandy, UT) in the right external jugular vein to withdraw blood samples. Blood samples were collected every 15 min from 4 h before to 4 h after treatment. Additional blood samples were collected at 1-h intervals from 5 to 28 h after treatments. Blood was allowed to clot for 1 h at room temperature and then was stored at 4°C overnight. Serum was separated by centrifugation and stored at –20°C for subsequent quantification of LH [51] and FSH [52] by radioimmunoassay. The reference standards for LH and FSH were NIH-oLH-S24 and NIH-oFSH-S12, respectively. Triplicate standard curves were included in each assay, and samples were analyzed in duplicate at 100 and 50 µl sample per tube for LH and FSH, respectively. Intraassay coefficients of variation for LH and FSH were 5.3% and 6.6%, respectively. Interassay coefficients of variation for LH and FSH were 11% and 11.6%, respectively. The minimum detectable doses of LH and FSH averaged 34 pg and 1.56 ng, respectively.
The following parameters were examined: 1) basal LH, defined as the lowest hormone concentration between pulses of LH, 2) mean LH, defined as the average concentration of LH, 3) number of pulses of LH (a pulse of LH was defined as a concentration of LH equal to or higher than the basal concentration of LH detected within an individual during the 4 h before treatment period, plus two standard deviations, followed by at least one descending hormone concentration, equal to or below mean LH [note: number of pulses was detected by visual counting and using computerized algorithms [53] with essentially the same results]), and 4) interval from administration of treatment to the increase in secretion of LH. Basal and mean LH as well as pulses of LH were determined during the 4 h before treatment and the 4 h after treatment.
Experiment 2. To determine the ER subtype involved in the mediation of the increase in the number of GnRH receptors induced by E2, dispersed pituitary cells (3 x 106 cells/well) were treated with E2 or agonists specific for ESR1 or ESR2. Cells were treated with various concentrations of E2 (0.4 pM to 40 nM), PPT (0.1 nM to 10 µM), or DPN (0.1 nM to 10 µM). The number of GnRH receptors was determined 12 h after addition of treatments. To examine the time course of the increase in the number of GnRH receptors after stimulation with E2, PPT, or DPN, cells were treated with the minimal stimulating concentrations of these compounds determined from the dose-response experiment. Therefore, cells were treated with E2 (0.4 nM), PPT (10 nM), DPN (1 µM), or PPT (10 nM) + DPN (1 µM) for 0, 4, 8, 12, 16, 20, or 24 h. Each treatment was conducted in duplicate with four pituitaries. After incubation, number of GnRH receptors was determined by radioreceptor assay.
Data were subjected to ANOVA using the general linear model of SAS [54]. Serum concentrations of LH were analyzed in two different ways. First, serum concentrations of LH and FSH were analyzed using an unstructured covariance model. In this analysis, samples were averaged at 2-h and 1-h intervals for LH and FSH, respectively. Second, basal LH, mean LH, and number of pulses of LH were subjected to repeated-measures analysis. Number of pulses of LH was subjected to Arc Sine transformation. Sources of variation included in the model were treatment (E2, PPT, DPN, and PPT + DPN), period (4 h), and the interaction of treatment by period. Ewe nested in treatment was used as the error term for treatment effect. Interval from administration of treatment to the increase in secretion of LH was analyzed in a completely randomized design. The number of GnRH receptors in cells in each treatment group was expressed as a percentage of mean counts per minute of 125I-labeled [D-Ala6]GnRH bound to untreated cells (control = 100%). Binding was evaluated by ANOVA using the MIXED model of SAS [54]. Time course data for the number of GnRH receptors was subjected to Log10 transformation. When differences among treatment means were detected, they were separated using least significant difference adjusted by the Tukey procedure.
RBAs for PPT and DPN were 1.52% and 0.52%, respectively, compared with E2 (Fig. 1). Changes in serum concentrations of LH in ewes treated with E2 or the ER-selective agonists are shown in Figure 2. The secretory profile of LH after treatment with E2 was characterized by a decrease in LH during the next 4 h after treatment (P < 0.05), followed by a preovulatory-like surge of LH (Fig. 2A). Similar to the effect of E2 (Fig. 2A), treatment with PPT (Fig. 2B), DPN (Fig. 2C), and the combination of the two (Fig. 2D) rapidly decreased secretion of LH (P < 0.05). The inhibitory effect of E2 and the ER-selective agonists on secretion of LH involved a decrease (P < 0.05) in basal LH (Fig. 3A) and mean LH (Fig. 3B), as well as the number of pulses of LH (Fig. 5A). The secretory profiles of LH in two representative ewes per group from 4 h before until 4 h after treatment are depicted in Figure 4. Pulses of LH were abolished within 15 min after treatment with either E2 or the ER-selective agonists. When PPT was present (Fig. 2, B and D), the duration of the inhibition of secretion of LH was 2 h longer than with E2. Distinct from E2 (Fig. 2A), neither agonist alone (Fig. 2, B and C) nor their combination (Fig. 2D) was able to induce a preovulatory-like surge of LH. However, each agonist induced a gradual, prolonged rise in secretion of LH after the initial inhibition. Compared with E2-treated ewes (Figs. 2A and 5B), the beginning of the increase in LH secretion occurred earlier (P < 0.01) in DPN-treated ewes (Figs. 2C and 5B), later (P < 0.05) in PPT-treated ewes (Figs. 2B and 5B), and at a similar interval in PPT + DPN-treated ewes (Figs. 2D and 5B). Regarding secretion of FSH, PPT alone (Fig. 6B) or in combination with DPN (Fig. 6D) decreased (P < 0.05) secretion of FSH similar to the decrease observed after E2 administration (Fig. 6A); but, compared with E2-treated ewes, serum concentrations of FSH were suppressed much longer. In contrast, administration of DPN alone did not decrease secretion of FSH (Fig. 6C).
|
|
|
|
|
|
For experiment 2, the minimal doses of E2, PPT, and DPN needed to increase the number of GnRH receptors were 40 pM E2 (Fig. 7A), 10 nM PPT (Fig. 7B), and 0.1 µM DPN (Fig. 7C). Treatment with E2 (Fig. 8A), PPT (Fig. 8B), or DPN (Fig. 8C) induced an increase in GnRH analog binding above controls at 8 h.
|
|
To estimate the initial dose of ER-selective ligands to use in OVX ewes, RBAs of PPT and DPN were determined in ovine uterine cytosol. Uterus, like pituitary gland, contains a high ratio of ESR1 to ESR2 [5, 9–13]. To date, there have been no reports of the RBAs of these agonists in tissues; however, RBAs for both PPT [55, 56] and DPN [57, 58] have been determined in cell lines transfected with expression vectors encoding either human ESR1 or ESR2. In those studies, the RBAs of PPT for ESR1 and DPN for ESR2 were
20-fold higher than those estimated in our study. It is possible that differences in RBA were influenced by the source of ERs (natural versus synthetic proteins) or may reflect species differences (ovine versus human). As in other reports [55–59], we observed a lower (3-fold) RBA of DPN compared with PPT, which justifies the higher dose (two to seven times) of DPN than PPT used in most studies [24, 26, 27, 35, 60].
In this and other studies [39–41], administration of E2 to OVX ewes induced a rapid decrease in secretion of LH, followed by a preovulatory-like surge of LH several hours later. At the dose used in this study, both agonists mimicked the acute inhibitory effect of E2 on secretion of LH. That is, both agonists caused a rapid (within minutes) decrease in secretion of LH, which remained at basal levels for several hours. An acute-nongenomic action of PPT and DPN has also been observed in vitro [36]. We previously reported that E2 acutely suppressed the GnRH-induced release of LH in primary cultures of ovine pituitary cells [37]. In that study, PPT at the maximum dose tested, but not DPN, acutely suppressed secretion of LH [37]. In the current study, DPN was given at a higher dose than in the previous in vitro study. Therefore, it is likely that had a higher dosage of DPN been used in that in vitro study [37], it also would have suppressed GnRH-induced release of LH. From studies using OVX rats chronically treated with E2 [25, 27, 28] and those using Esr1–/– and Esr2–/– mice [1, 20–23], it has been concluded that ESR1 is the main ER involved in the mediation of the inhibitory feedback of E2 on secretion of LH. As mentioned above, the acute and transitory inhibition of secretion of LH induced by E2 occurs, at least in part, at the pituitary gland [37]. It seems that the pituitary is not the primary target for long-term negative feedback effect of E2 [61], particularly if high doses of E2 are used [45, 61–64] or during the period of amplified sensitivity to the negative feedback of E2, such as seasonal anestrus in ewes [65, 66].
At doses 2 to 200 times higher than those used in this study [24, 26, 27, 31, 34, 35, 60], systemic administration of PPT and DPN to rodents differentially regulates gene expression, hormone secretion, and animal behavior. For example, DPN failed to increase PGR in the hypothalamus, but it decreased anxiety-related behavior; in contrast, PPT upregulated PGR and increased anxiety-related behavior in mice [24]. Similarly, chronic administration of PPT, but not DPN, to OVX rats at a dose more than 40 times higher than that used in our study increased PGR in rat pituitaries [27]. Despite the selective gene activation induced by PPT and DPN in rodents, a cross-activation of ERS1 by DPN has been reported in cells transfected with an expression vector for Ers1 [57, 67]. In these studies, DPN activated ERS1 when given at concentrations 10-fold [67] to 170-fold [57] higher than the minimal concentration needed to activate ESR2, and the extent of cross-activation was dependent on the particular gene being studied [67]. To date, there have been no reports regarding the extent of selectivity of these ER agonists in the sheep, and a potential cross-activation of ESR1 by DPN cannot be ruled out; however, based on the following information, it seems unlikely that changes in secretion of LH induced by DPN are the result of cross-activation of ESR1. First, systemic administration of PPT and DPN in rodents [24, 26, 27, 31, 34, 35, 60] at doses higher than those used in our study specifically activated only the appropriate ER isoforms. Second, since administration of either ER agonist at a dose five times lower than that employed in this study (preliminary data not shown) did not induce a statistically significant decrease in secretion of LH, it seems that the ER agonist doses used are close to the minimal effective dose required to decrease secretion of LH and, even so, the dose of PPT was highly effective in decreasing secretion of FSH. Thus, unless the extent of cross-activation of ESR1 by DPN is much higher in sheep than in rodents, it is unlikely that the changes induced by DPN in these studies occurred due to activation of ESR1.
Neither PPT nor DPN alone or combined was able to induce a preovulatory-like surge of LH. However, each agonist induced a gradual, prolonged rise in secretion of LH after the initial negative feedback. It is likely that the induction of a preovulatory surge of LH, which depends on a coordinated series of events between the pituitary and the hypothalamus, may require a greater dose of ER agonists. It has been shown that the effective dose of ER-selective ligand is dependent on the gene being targeted [34, 38]. Although the ability of PPT and DPN to reach the central nervous system and modulate gene transcription and animal behavior has been demonstrated in rodents [24, 31, 38, 60], a rapid clearance rate of ER agonists in sheep potentially could prevent full expression of the massive release of LH. This may be relevant, since ewes exposed to E2 for short intervals did not have surges of GnRH [68]. However, in our study, the length of the inhibition of secretion of FSH, as well as the length of the acute inhibition of secretion of LH induced by PPT, was longer than that induced by E2, suggesting that the half-life of PPT may not be a limiting factor. To our knowledge, the pharmacokinetics of these ER synthetic ligands has not been reported.
Alternatively, the lack of a normal preovulatory-like surge of LH after administration of PPT and/or DPN may suggest that at the level of the hypothalamus, E2-binding proteins distinct from the classic ERs may mediate the massive release of GnRH induced by E2. Indirect but increasing evidence indicates that E2 is able to activate many cellular responses in the brain by a mechanism independent of the classic ESR1 and ESR2 [11, 18, 19]. We propose that the initial and rapid decrease in secretion of LH induced by E2 is mediated by both ESR1 and ESR2 desensitizing the pituitary gland to GnRH. The gradual rise in secretion of LH at the time of the expected preovulatory surge may be explained by a direct activation of ESR1 and ESR2 in the pituitary gland, thus enhancing its sensitivity to GnRH, and in this way overcoming the negative feedback of E2. The gradual prolonged rise in secretion of LH observed in ewes treated with ER-selective agonists is compatible with an absence of a massive release of GnRH but increased sensitivity of the pituitary gland to basal secretion of GnRH. The ability of PPT and DPN to modulate secretion of GnRH has not been reported to date. In agreement with this interpretation, we found that similar to E2 [39, 44, 45], PPT and DPN increased the number of GnRH receptors in primary cultures of ovine pituitary cells. The minimal effective dose of DPN required to increase the number of GnRH receptors was 10-fold higher than that for PPT. As mentioned above, a cross-activation of ESR1 by DPN has been observed in vitro studies when given at concentrations 10-fold [67] to 170-fold [57] higher than the minimal concentration needed to activate ESR2. Therefore, the potential cross-activation of ESR1 by DPN in the induction of number of GnRH receptors cannot be ruled out. The ability of PPT and DPN to modulate the number of GnRH receptors in vivo remains to be corroborated.
Compared with E2-treated ewes, the onset of the increase in secretion of LH occurred earlier in DPN-treated ewes, later in PPT-treated ewes, and at a similar interval in ewes treated with PPT + DPN. Thus, simultaneous activation of ESR1 and ESR2 may be necessary to synchronize the beginning of the positive feedback of E2 on secretion of LH. This leads us to hypothesize that changes at the beginning of the increase in secretion of LH induced by the ER agonists could be explained by the timing of the increase in the number of GnRH receptors in the pituitary gland; however, the increase in the number of GnRH receptors occurred at the same time, regardless of whether the cells were treated with E2, PPT, or DPN. These data may suggest that activation of both ESR1 and ESR2 are required to synchronize the beginning of the positive feedback of E2 on secretion of LH by mechanisms other than the synthesis of GnRH receptors. Since ESR2 has been detected in GnRH neurons [15–17], it is plausible that the early rise in secretion of LH induced by DPN results from a discrete release of GnRH, and that activation of ESR1 may delay the secretory response of the gonadotropes to GnRH or the release of GnRH by acting through interneurons containing ESR1 in the hypothalamus [69].
A negative effect of E2 on secretion of FSH in ewes has been reported [41, 43], and the evidence indicates that the pituitary gland [43, 70–72] and the hypothalamus [23] are the targets for this negative feedback. Distinct from the acute (within minutes) inhibition on secretion of LH, where both ER subtypes appear to be involved, the delayed (after 3 h) inhibitory effect on secretion of FSH induced by E2 seems to occur exclusively through ESR1. Inhibition of FSH secretion by PPT but not by DPN has been reported in OVX rats chronically treated with these drugs [27], and a preeminent role of ESR1 as the mediator of inhibition of FSH secretion by E2 has been documented in Esr1–/– mice [20, 23]. An alternative explanation for the lack of effect of DPN on secretion of FSH is that an enhanced release of GnRH from the hypothalamus that is mediated by ESR2 could overcome a putative inhibition in secretion of FSH in the pituitary gland by cross-activation of ESR1, resulting in no changes in secretion of FSH. Either interpretation, however, supports the conclusion that only ESR1 appears to be involved in the negative feedback of E2 on secretion of FSH.
We conclude that in OVX ewes: 1) ESR1 and ESR2 mediate the negative feedback of E2 on secretion of LH at the level of the pituitary gland, 2) ER subtypes do not synergize or antagonize the effects of each other; however, they do interact to synchronize the beginning of the stimulatory effect of E2 on secretion of LH, 3) ESR1 and ESR2 may at least partially mediate the positive feedback of E2 on LH secretion by increasing the number of GnRH receptors, and 4) only ESR1 appears to be involved in the negative feedback of E2 on secretion of FSH.
FOOTNOTES
1Supported by National Research Initiative Competitive Grant no. 2005-35203-15376 from the United States Department of Agriculture Cooperative State Research, Education, and Extension Service. ![]()
Correspondence: 2Terry M. Nett, Animal Reproduction and Biotechnology Laboratory, Department of Biomedical Science, 3801 West Rampart Rd., Fort Collins, CO 80523-1683. FAX: 970 491 3557; e-mail: Terry.Nett{at}colostate.edu
Received: 9 January 2007.
First decision: 7 February 2007.
Accepted: 9 April 2007.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |