Differential Expression and Regulation of Estrogen Receptors (ERs) in ...

2 downloads 0 Views 581KB Size Report
tuitary hormones during the reproductive cycle in a cell- and pro- moter-specific manner. ..... AtT20 corticotropes and αTSH thyrotrope precursors (not shown).
0013-7227/00/$03.00/0 Endocrinology Copyright © 2000 by The Endocrine Society

Vol. 141, No. 6 Printed in U.S.A.

Differential Expression and Regulation of Estrogen Receptors (ERs) in Rat Pituitary and Cell Lines: Estrogen Decreases ER␣ Protein and Estrogen Responsiveness* DEREK A. SCHREIHOFER, MARK H. STOLER,

AND

MARGARET A. SHUPNIK

Department of Internal Medicine, Division of Endocrinology and Metabolism, and Department of Pathology (M.H.S.), University of Virginia, Charlottesville, Virginia 22908 ABSTRACT Estrogen (E) regulates the synthesis and secretion of several pituitary hormones during the reproductive cycle in a cell- and promoter-specific manner. One mechanism underlying cell specificity is the differential expression of estrogen receptor (ER) isoforms. We used in vivo and in vitro rodent pituitary cell models to examine the expression and regulation of ER␣, ER␤, and the pituitary-specific ER␣ isoform, truncated estrogen receptor product-1 (TERP-1). In cycling female rat pituitaries, ER␤ messenger RNA (mRNA) levels fell 40% on the morning of proestrus and were suppressed by E or dihydrotestosterone in ovariectomized females. In lactotrope and gonadotrope cell lines (GH3, RC4B, L␤T2), progesterone (P) or P plus E also suppressed ER␤. TERP-1 mRNA increased 3-fold at proestrus and in

E

STROGEN (E) REGULATES the synthesis and secretion of several pituitary hormones and plays a key role in the regulation of reproductive cyclicity. Two classic pituitary targets for E action are the lactotropes, where E stimulates PRL, and the gonadotropes, where E regulates FSH and LH. There is significant gene- and cell-specific regulation of E action in the pituitary, and E has both positive and negative effects in this tissue. Such effects may be exerted on different cell types at the same time. For example, E stimulates lactotrope proliferation and PRL synthesis (1, 2) while at the same time inhibiting gonadotropin secretion in vivo (3). However, the specific molecular mechanisms underlying E action in these cells is not fully understood. Estrogen acts to modulate gene expression through estrogen receptors (ERs), which belong to a superfamily of nuclear transcription factors (4). These proteins have a conserved domain structure with a centrally located DNA-binding domain (DBD), an N-terminal ligand-independent transactivating function (AF1), and a C-terminal ligand-binding Received September 28, 1999. Address all correspondence and requests for reprints to: Margaret A. Shupnik, Ph.D., Box 800578, Department of Internal Medicine, University of Virginia Health Science Center, Charlottesville, Virginia 22908. E-mail: [email protected]. * This work was supported by a grant from the Lalor Foundation (to D.A.S.), NIH Grant RO1-DK-57082 (to M.A.S.), and NICHHD/NIH Cooperative Agreement U54-HD-28934 as part of the Specialized Cooperative Centers Program in Reproductive Research (to the core laboratories of the Center for the Study of Reproduction at the University of Virginia). Portions of this work were presented at the 81st Annual Meeting of The Endocrine Society, June 1999, San Diego, CA.

response to E treatment in vivo and in cell lines. ER␣ mRNA levels were not regulated significantly by any treatment in vivo or in cell lines. However, E suppressed ER␣ protein levels in vivo and in cell lines, and reduction of ER␣ protein levels by E or the antiestrogen ICI182,780 reduced E-stimulated transcriptional activation of the PRL promoter in GH3 cells. TERP-1 and ER␤ protein levels were low to undetectable in cell lines, but E stimulated TERP-1. Because E treatment decreases ER␤ mRNA and ER␣ protein and increases levels of TERP-1 (which can suppress ER␣/␤ activity), the dynamic steroid-induced changes in ER expression in the rat pituitary during the midcycle gondaotropin/PRL surge may provide a means for ovarian steroids to limit positive feedback. (Endocrinology 141: 2174 – 2184, 2000)

domain (LBD) with a ligand-dependent trans-activating function (AF2). Ligand-bound ERs form dimers that act on specific estrogen response elements in the promoter regions of E-regulated genes. At least three major ER isoforms are expressed in the rat pituitary including ER␣, ER␤, and truncated estrogen receptor product-1 (TERP-1) (5, 6). ER␣ and ER␤ share 95% homology in the DBD, approximately 50% homology in the LBD, and little homology in the N-terminus (7). TERP-1 contains a unique 5⬘-untranslated sequence fused to exons 5– 8 of the ER␣ (5). Translation of TERP-1 messenger RNA (mRNA) leads to a protein with most of the LBD of ER␣, but no N-terminus or DBD (5, 8). In transcription assays, the relative expression of these receptors can have profound effects on estrogen-induced trans-activation. Specifically, ER␣ has greater trans-activating ability than ER␤ (9, 10), and the binding affinity and activity of certain phytoestrogens and synthetic estrogen ligands differ for ER␣ and ER␤ (11). Although TERP-1 has no independent activity, it can stimulate ER activity at low ratios and inhibit the activity of ER␣ and ER␤ at ratios of 1:1 or greater (12, 13). Thus, the overall response of tissues and regulated genes to E is in part dependent on the relative expression of ER isoforms. The tissue-specific expression of ER mRNAs has been examined extensively in several species. Many estrogenresponsive tissues express both ER␣ and ER␤, although urogenital structures and lung express significantly more ER␤ (14). Previous studies have shown that all three ER mRNAs are present in rat pituitaries, with ER␣ sequences present at much higher levels than ER␤ (6, 15, 16). TERP-1 expression is limited to the pituitary (5, 17, 18). A recent in

2174

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

situ hybridization study has demonstrated the presence of ER␣ in several pituitary cell types, including gonadotropes, lactotropes, and corticotropes (6). ER␤ mRNA also appears to be widely distributed in the rodent pituitary (6, 15, 16, 19). Interestingly, ER␤ is the major pituitary ER expressed in prepubertal female rats, and mRNA levels in females are almost twice those seen in males at this time (16). Similarly, under physiological conditions TERP-1 mRNA is expressed only in female rat pituitaries (8, 18). Little is known about the promoter and regulatory regions of ERs, although several alternate first exons and promoters exist for ER␣ and ER␤ (20, 21). TERP-1 mRNA is transcribed from a unique transcriptional start site, allowing divergent and cell-specific regulation of ER forms. In a recent report, ER␣ and ER␤ mRNAs in adult rat pituitaries were not altered by E (6); however, we and others have found modulation of ER␣ and TERP-1 mRNA through the estrous cycle (8, 18). Regulation of ER␤ expression across the estrous cycle has not been described. In the present study we examined the regulation of ER isoforms in the rat pituitary during the estrous cycle and in response to different steroids and hypothalamic peptides and compared these results with the regulation of ER isoforms in lactotrope and gonadotrope clonal cell lines. Our results demonstrate that there is a significant regulation of ER isoforms in pituitary cells in vivo and in vitro and that the regulation of ERs by steroids and hypothalamic peptides is cell type specific. Furthermore, E specifically suppresses ER␣ protein levels, and this suppression diminishes subsequent E responsiveness. Materials and Methods Animals and experimental protocols All animal procedures were performed in accordance with the guidelines established by the University of Virginia animal care and use committee. Male and female CD-1 rats (200 –225 g) were obtained from Charles River Laboratories, Inc. (Wilmington, MA). For estrous cycle experiments, intact female rats were maintained on a 14-h light, 10-h dark cycle. Estrous cycle stage was determined by vaginal lavage. Before inclusion in the study, animals had two consecutive normal cycles. Animals were killed by decapitation at 0900 h (AM groups) on each day of the estrous cycle, and additional animals were killed at 1700 h (PM groups) on estrus and proestrus. Total pituitary RNA was prepared from pooled tissue (five to eight rats) as previously described (8). For steroid treatments, castrate male or ovariectomized female rats (14 days postovariectomy) were injected for 3 days with oil vehicle, 17␤-estradiol (E; 20 ␮g/100 g BW), progesterone (P; 2 mg/100 g BW), both E and P, or 5␣-dihydrotestosterone (DHT; 20 ␮g/100 g BW). Ribonuclease (RNase) protection data for ER␣ and TERP-1 were previously reported for these animals (8). An additional four ovariectomized female rats were used for assessment of ER␣ and TERP protein levels after treatment with oil vehicle (n ⫽ 2) or E (n ⫽ 2) as described above. For antagonist experiments ovariectomized female rats were injected sc for 5 days with oil vehicle or 17␤-estradiol (30 ␮g) at 0930 h. On days 3–5, one E-treated group also received 250 ␮g of the dopamine agonist bromocriptine (Sigma, St. Louis, MO) in PBS, and one group received 50 ␮g of the GnRH antagonist antide (Sigma) in 20% propylene glycol at 0930 h. On day 5, animals were killed by decapitation at 1430 h. Pituitaries were removed for isolation of RNA, and trunk blood was collected for assay of LH and PRL to confirm the effectiveness of treatments. For in situ hybridization experiments castrate male and female rats were injected for 3 days with oil vehicle or E (30 ␮g). In all cases, animals were anesthetized with halothane and killed by decapitation.

2175

In situ hybridization Pituitaries were fixed in 4% paraformaldehyde and embedded in paraffin. Five-micron tissue sections were mounted on 3-aminopropylthriethoxysilane-coated glass microscope slides. Alternate sections were subjected to in situ hybridization with tritium-labeled antisense riboprobes for rat LH␤, ␤-actin, the N-terminus of rat ER␣, the C-terminus of rat ER␣, and PRL. Additional sections were processed with sense probes or were treated with RNase. Riboprobes for LH␤ (325 bp) and PRL (467 bp) were prepared from full-length coding sequence in pGem7 vector (Promega Corp., Madison, WI) and PCR2.1 vector (Invitrogen, San Diego, CA), respectively. The actin probe was prepared from a full-length clone of chicken ␤-actin (1.8 kb). Riboprobes for rat ER␣ were prepared by transcription from cloned fragments of ERs in the pGem7 vector. Probes corresponding to the N-terminus untranslated region to exon 4 (710 bp) and the C-terminus from exons 5– 8 (770 bp) were used. Probes were matched with respect to coding sequences and C plus U content. The N-terminal probe recognizes ER␣, and the C-terminal probe recognizes both ER␣ and TERP. In addition, a full-length ER␤ (11) riboprobe was prepared from coding sequence in pGem3Z. Probes were prepared using the appropriate polymerase in the presence of a mixture of 50 ␮m deoxy-CTP and 50 ␮m deoxy-UTP with a specific activity of approximately 3.4 ⫻ 107 cpm/␮g. Probes (except LH␤) were hydrolyzed to a desired length of 300 bp and hybridized at 55 C. Slides were washed in 0.1 ⫻ saline-sodium citrate at 65 C and coated with Kodak NTB-2 autoradiography emulsion (Eastman Kodak Co., Rochester, NY). Slides were counterstained with hematoxylin and eosin. Autoradiographic signals were photographed under darkfield illumination, and histology was reviewed with brightfield illumination. Hybridization signal was quantified as previously described (22). Briefly, 8-bit gray scale digital images were acquired with a Gould IP850 Image Processing Workstation (Gould Electronics, Inc., Fremont, CA) with spatial resolution of 512 ⫻ 512 pixels. Data were analyzed on a Sun-3 workstation (Sun Microsystems, Inc., Mountain View, CA) the using the ANALYZE software package (Biodynamics Research Unit, Mayo Foundation, Rochester, MN). A threshold for positive cells was selected at a half-maximum difference between background grain intensity and grain intensities over positive cells. The threshold was used to outline an area of confluent grains above each cell, and the grain intensity of this area was quantified.

Cells lines, chemicals, and treatments Mouse- and rat-derived pituitary cell lines were used. Rat somatomammotrope GH3 cells and rat lactotrope MMQ and PR1 cells were maintained in DMEM/10% newborn calf serum (NCS; Life Technologies, Inc., Gaithersburg, MD). Mouse ␣T3 pregonadotrope and L␤T2 differentiated gonadotrope cells were maintained in DMEM/10% FBS (Life Technologies, Inc.). Rat lactogonadotrope RC4B cells were maintained in 50% DMEM/50% MEM with 10% FBS, 0.02% BSA, 400 pm epidermal growth factor, and 15 mm HEPES (pH 7.6). All media contained 100 U/ml penicillin and 100 ␮g/ml streptomycin (Life Technologies, Inc.). Before treatments cells were split into culture flasks (75 mm2; Corning, Inc., Corning, NY) or 60-mm four-well culture plates (ICN Biomedicals, Inc., Costa Mesa, CA) for 1 day in maintenance medium. The following day, medium was removed, cells were washed twice with PBS, and medium was replaced with phenol red-free DMEM/5% charcoal-stripped NCS. Cells were maintained in this medium for 4 days during treatments. Cells were left untreated (NT), treated for 1 or 4 days with 10 nm E (Sigma), treated for 4 days with 10 nm DHT (Sigma), treated for 4 days with 0.1 ␮m P (Sigma), or treated for 4 days with E and P. For pituitary adenylate cyclase activating peptide treatment, 100 nm PACAP (Sigma) was added for the final 18 h of treatment; for GnRH treatment, 50 nm GnRH (Sigma) was added for the final 6 h of incubation. After treatments, cells were washed twice with PBS and collected as described below. Two separate treatment groups were prepared for each cell line. For RT-PCR analysis, each treatment group was amplified two or four times from separate RT reactions. For immunoblotting, three or four separate treatment groups were prepared and analyzed.

Transient transfections GH3 cells were transiently transfected with 2 ␮g of a reporter construct containing 2.5 kb of the estrogen-responsive rat PRL promoter

2176

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

fused to a luciferase reporter gene (gift from Dr. Richard Day), as previously described (12). Cells were plated in DMEM/5% charcoal-stripped NCS with or without 10 nm E. Untreated parallel wells were subsequently treated 18 h later with the antiestrogen ICI 182,780 (10 nm). Cells were transfected 6 h later. After 2 h in diethylaminoethyl-dextran transfection medium, GH3 cells were shocked for 2 min with 10% dimethylsulfoxide and subsequently treated with or without E for 22 h. Parallel wells were collected for protein determination by immunoblotting.

Semiquantitative RT-PCR RNA was extracted from cells by lysis in guanidinium thiocyanate as previously described (8). Conditions for semiquantitative RT-PCR were determined separately for each pituitary cell line and rat pituitary samples because the relative abundance of ER message varied among tissues and pituitary cell lines. Total RNA was reverse transcribed in a 20-␮l mixture consisting of 5 mm MgCl2 (Perkin-Elmer Corp., Palo Alto, CA), PCR buffer II, 2 mm deoxyribonucleotides, 1 U RNase inhibitor, 2.5 ␮m random hexamers, and 2.5 U murine leukemia virus reverse transcriptase. RT reactions were incubated 10 min at room temperature, 15 min at 42 C, and 5 min at 99 C and cooled to 4 C for 5 min in a Perkin-Elmer Corp. Thermocylcer 480. For PCR reactions, MgCl2 was adjusted to 2 mm, and buffer, water, 0.2 mm primer oligonucleotides (Operon Technologies), and 2.5 U/100 ␮l Taq DNA polymerase (Sigma) were added to a final volume of 16 –100 ␮l. PCR was performed in a Stratagene Robocycler Gradient 40. General PCR conditions consisted of a single cycle of 3 min at 95 C, 3 min at annealing temperature, and 3 min at 72 C. Additional cycles were performed with 45-sec steps and a final 10-min extension step at 72 C. Optimization was performed separately for each set of primers. Annealing temperature (56 – 66 C) for each primer set was determined at 35 cycles with 0.5– 8 ␮g input RNA. Optimal cycle number was determined over a range of 15– 40 cycles. For cell lines, 640 ng input RNA fell in the linear range for all primer sets. For pituitaries 160 ng input RNA were used. Primer sequences, the expected product sizes, and PCR conditions are shown in Table 2. Primers were chosen to recognize both rat and mouse sequences and to cross intron boundaries. ER␣ was amplified over a range of 20 (␣T3) to 30 cycles (L␤T2), ER␤ was amplified over a range of 32 (L␤T2, GH3) to 37 cycles (RC4B), and TERP was amplified over a range of 33 (rat pituitary) to 36 (␣T3) or 37 cycles (all others). Both ER␣ and ER␤ primer sets were directed against Nterminal regions, and thus detect both the full-length receptors and certain C-terminal variants of ER␣ (exon 5/6 deletions) and ER␤. Previous results (5, 8) suggest that ER␣ variants in the rat pituitary are not highly regulated and represent only a small proportion of the ER␣ mRNA population. In vivo treatment samples were also assessed for regulation of the ER␤2 LBD 54-bp insertion variant (23) using PCR primers 5⬘ and 3⬘ to the region of the insert. PCR amplification over 37 cycles resulted in products of 371 bp for ER␤1 and 425 bp for ER␤2. A unique EAR1 restriction site in the insert was used to verify the identity of ER␤2. RPL19 ribosomal protein mRNA primers were included in each PCR reaction as an internal control (24). The linear range for RPL19 mRNA was also determined separately for each cell line (data not shown). For ER␣ and ER␤, RPL19 was amplified over the same cycle number as the ER of interest. However, in the case of TERP-1, RPL19 primers were added to the reaction mixture after four to seven cycles to allow sufficient initial synthesis of TERP-1 templates. After PCR, 16 ␮l of each reaction were separated on 1.5% agarose gels containing ethidium bromide (0.7 ␮g/ml). Gels were photographed and evaluated by fluoroimaging with a Molecular Dynamics, Inc. (Menlo Park, CA), Fluoroimager 515. Data were analyzed with ImageQuant software (Molecular Dynamics, Inc.).

Immunoblot analysis ER␣ and TERP-1 protein expression in cell lines was determined from 50 ␮g total protein and separated on 12% polyacrylamide-SDS gels by immunoblot analysis using a rabbit polyclonal antibody as previously described (12). The antibody (C1355) was generated against C-terminal amino acids 586 – 600 of the rat ER␣ and also detects mouse ER␣ (8). Western analysis was performed with enhanced chemiluminescence (Amersham Pharmacia Biotech, Arlington Heights, IL) using the C1355 primary antibody at 1:2,500 for 1 h, followed by a 1-h incubation with a horseradish peroxidase-conjugated donkey antirabbit IgG secondary

Endo • 2000 Vol 141 • No 6

antibody (Amersham Pharmacia Biotech) at 1:800. For rat pituitaries, 75 ␮g total protein were used, and primary antibody was used at 1:5,000. Protein loading was normalized to ␤-actin using a monoclonal primary antibody at 1:5,000 (Sigma) and a horseradish peroxidase-conjugated goat antimouse IgG secondary antibody at 1:40,000 (The Jackson Laboratory, Bar Harbor, ME). In some cases, in vitro translated proteins or proteins from transfected COS cells were included on blots as markers and positive controls for experimental conditions. Densitometry was performed with a Molecular Dynamics, Inc., Personal Densitometer SI and analyzed with ImageQuant software (Molecular Dynamics, Inc.).

Statistical analysis Data for each ER and each cell line or animal experiment were analyzed separately. After one-way ANOVA, a priori pairwise comparisons were made between treatment groups and controls using t tests. For in vivo estrous cycle experiments, all days were compared with metestrus. For the remainder of the in vivo experiments comparisons were made to the oil-treated controls. For cell lines, comparisons were made between treatments and the untreated condition. P ⬍ 0.05 was considered significant.

Results ER mRNA isoform expression and regulation in rat pituitary

In situ hybridization. Previously, we and others have shown that TERP-1 mRNA is stimulated by E treatment in whole rat pituitary (5, 6, 8, 17, 18). We used in situ hybridization to examine the regulation of ER␣, TERP, and ER␤ by E in pituitary slices. Because TERP-1 differs from ER␣ only in the first 31 bp, expression was determined using both an Nterminal (detects ER␣ only) and a C-terminal riboprobe to ER␣ (detects ER␣ and TERP-1). Other TERP mRNA species have been detected at lower abundance than TERP-1 and may contribute to the C-terminal signal; however, only TERP-1 appears to be expressed at significant levels and translated into protein (5, 8, 17, 25). Therefore, differences in regulation between the two signals can be accounted for by TERP-1 expression. In ovariectomized female rats, E treatment decreased LH␤ expression and increased PRL expression as expected (Fig. 1, A–D, and Table 1). Treatment of castrate males showed similar results in three experiments (data not shown). ER␣ expression determined with the Nterminal riboprobe was more diffuse than hormone mRNAs, but overlapped regions containing gonadotropes and lactotropes (Fig. 1E and Table 1). E treatment had no significant effect on ER␣ expression throughout the pituitary, although there was a slight decrease in hybridization signal (Fig. 1F and Table 1). However, E treatment led to a 2.9-fold increase in C-terminal message expression (Fig. 1H and Table 1). Increased signal intensity was observed throughout the slices, and dramatic alteration in specific cell populations was not observed. These data support the observation that TERP is up-regulated by E even when full-length ER␣ levels are unchanged. ER␤ mRNA was expressed at much lower levels than ER␣ and appeared to be more diffuse (data not shown). ER␤ levels were close to background and were not significantly modulated by E. In vivo studies. Pituitary expression of ER isoforms across the estrous cycle was determined by semiquantitative RT-PCR. Figure 2A shows the relative expression of ER␣, TERP, and ER␤ from 15– 40 cycles of PCR amplification in rat pituitary

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

2177

FIG. 1. Expression of LH␤ (A and B), PRL (C and D), ER␣ (ER␣-N; E and F), and TERP (ER␣-C; G and H) mRNA in female rat pituitaries determined by in situ hybridization under darkfield illumination. ER␣-specific signal was determined with an N-terminal probe, and TERP signal was determined with a C-terminal probe that recognizes both ER␣ and TERP. The upper panels are from oil-treated ovariectomized rats, and the lower panels are from E-treated ovariectomized rats. All panels were photographed at ⫻20 magnification, except LH␤, which is shown at ⫻40. TABLE 1. In situ hybridization grain intensity in female rat pituitaries Treatment

LH␤

PRL

ER␣-N

ER␣-C

Oil E2

128 ⫾ 30 50 ⫾ 5

56 ⫾ 9 140 ⫾ 31

36 ⫾ 5 28 ⫾ 4

19 ⫾ 4 55 ⫾ 6

and cell lines. Figure 2B shows the expression curves for each mRNA isoform with 40 cycles normalized to 100%. In subsequent studies with rat pituitaries, ER␣ was amplified over 29 cycles, ER␤ over 36 cycles, and TERP over 33 cycles (see Table 2 and Fig. 2). ER␣ mRNA, representing full-length ER␣ and ER␣ exon 5/6 deletions, did not change significantly across the estrous cycle (Fig. 3). The exon 5/6 deletion is present at low levels (10% of full length) and is not regulated by steroids (8). In contrast, TERP-1 mRNA expression rose 3.2-fold over metestrus levels by the morning of proestrus (Fig. 3), and ER␤ levels fell 40% on the morning of proestrus compared with levels on metestrus, coincident with the peak of TERP-1 expression (Fig. 3). These ER␤ values represent both ER␤1 and ER␤2, a variant containing a 54-bp insertion in the LBD. Several factors may account for the differential regulation of ER isoforms in the pituitary across the estrous cycle, including ovarian steroids, hypothalamic peptides, and intrapituitary factors. We examined the regulation of ERs in vivo by steroids (E, P, and DHT), GnRH, and dopamine, because these factors can impact gonadotropin and PRL secretion across the estrous cycle. In agreement with our previous studies, we found that ER␣ mRNA is not highly regulated by E or P, but TERP-1 mRNA is induced dramatically by E (Fig. 4 and Ref. 8). In addition, we found that DHT significantly increased TERP-1 mRNA levels in female rats (Fig. 4A). P treatment had no effect on TERP-1 expression and did not alter the effect of E in females (Fig. 4, A and C) or males (data not shown). Treatments that increased TERP-1 mRNA in females (E, E plus P, and DHT) also significantly decreased ER␤ mRNA expression, but had no effect on ER␣ (Fig. 4A). Similar trends were observed in E-treated and E- plus Ptreated males (data not shown). ER␤ values represent both ER␤1 and ER␤2, a variant containing a 54-bp insertion in the LBD. An additional primer set was used to coamplify ER␤1 and ER␤2. Both variants were regulated in a similar manner,

and ER␤2 was expressed at slightly higher levels than ER␤1 (Fig. 4B). Immunoblots and immunohistochemical studies with available ER␤ antibodies indicate that ER␤ protein levels are very low, and regulation could not be assessed (data not shown). ER levels in the whole pituitary are a reflection of the sum of the effects of treatments on all pituitary cell types. To begin to examine the cell type-specific regulation of ER isoforms, we examined the roles of two hypothalamic factors. The actions of both GnRH acting at the gonadotropes and dopamine acting at the lactotropes can be modulated by the steroid hormone milieu (26). One potential mechanism underlying these interactions is changes in steroid receptor expression. To determine the influence of GnRH on ER message levels, ovariectomized E-treated rats were treated with the GnRH antagonist antide. The LH level in ovariectomized rats was 27.4 ⫾ 9.37 ng/ml (n ⫽ 7). LH levels were reduced by E alone to 3.57 ⫾ 0.57 ng/ml (n ⫽ 7; P ⬍ 0.01) and were not significantly further reduced by antide (2.84 ⫾ 0.49 ng/ ml; n ⫽ 6) or bromocriptine (3.15 ⫾ 0.55 ng/ml; n ⫽ 6) in the presence of E. The PRL level in ovariectomized rats was 8.18 ⫾ 3.54 ng/ml (n ⫽ 7) and was significantly increased by E treatment (20.78 ⫾ 2.55 ng/ml; P ⬍ 0.01; n ⫽ 7). Activation of dopamine receptors with bromocriptine in the presence of E led to a significant decrease in PRL levels to 8.43 ⫾ 1.76 ng/ml (n ⫽ 6) compared with E treatment alone (P ⬍ 0.01), confirming the effectiveness of this drug. Antide did not alter E-induced PRL secretion (20.98 ⫾ 4.06 ng/ml; n ⫽ 6). ER␣ mRNA levels were unaffected by any treatment, but E suppressed ER␤ mRNA and stimulated TERP-1 mRNA (Fig. 4C). Neither antide nor bromocriptine altered these responses to E. Physiological pituitary responses to E are ultimately dependent on the expression levels of ER protein. Because mRNA levels, translational efficiency, and protein turnover influence ER protein levels, we examined whether changes seen at the mRNA level were also reflected at the level of protein expression. Although ER␣ mRNA levels were not altered by steroid treatments in ovariectomized female rats, immunoblotting revealed that ER␣ protein levels were reduced 3-fold in E-treated rats (Fig. 5). In contrast, TERP protein was essentially undetectable in oil-treated animals

2178

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

Endo • 2000 Vol 141 • No 6

FIG. 2. Optimization of RT-PCR for ER␣, ER␤, and TERP-1 mRNAs in rat pituitary and pituitary cell lines. A, Photograph of ethidium-stained agarose gels. One hundred and twenty nanograms of rat pituitary total RNA or 640 ng cell line total RNA were reverse transcribed and amplified from 15– 40 cycles by PCR in 16-␮l reactions. B, Amplification curves for rat pituitary and cell lines for each ER isoform. Signal intensity at 40 cycles was normalized to 100%.

and was increased nearly 6-fold by E treatment (Fig. 5). Thus, the ratio of TERP to ER␣ in oil-treated animals was 0.3 and rose to 1.2 with E. ER mRNA isoform expression and regulation in pituitary cell lines

To further examine the cell-specific expression and regulation of ER isoforms in the pituitary, we used several rodent cell lines of lactotrope and gonadotrope origin. The relative expression of ER mRNAs was determined in E-treated cells

under conditions where TERP-1 levels were expected to be elevated. Figure 2 shows the relative expression of ER␣, ER␤, and TERP mRNAs by PCR amplification in a rat somatolactotrope cell line (GH3), rat lactotrope cell lines (PR1 and MMQ), a rat lactogonadotrope cell line (RC4B), a mouse gonadotrope precursor cell line (␣T3), and a mouse differentiated gonadotrope cell line (L␤T2). For comparison, ER expression in an E-treated male rat is shown (120 ng input RNA/lane compared with 640 ng for cell lines). All cell lines expressed high levels of ER␣, with detectable expression

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

2179

TABLE 2. PCR primer sets for semiquantitative RT-PCR Target mRNA

ER␣ ER␤ TERP RPL 19

Exon

Exon 2 Exon 4 Exon 1 Exon 2 TERP Exon 7 Exon 5 Exon 6

Primers

5⬘-GTCTGGTCCTGTGAAGGCTGCAA-3⬘ 5⬘-GCCTTCCAAGTCATCTCTCAGACG-3⬘ 5⬘-GCTGTGATGAACTACAGTGTTCCC-3⬘ 5⬘-TGGACTAGTAACAGGGCTGGCACA-3⬘ 5⬘-CCATTTCTTGAGCTTGTTGAACAG-3⬘ 5⬘-AGTGTCTGTGATCTTGTCCAGGAC-3⬘ 5⬘-CTGAAGGTCAAAGGGAATGTG-3⬘ 5⬘-GGACAGAGTCTTGATGATCTC-3⬘

FIG. 3. Expression of ER␣ (open bars), ER␤ (solid bars), and TERP-1 (hatched bars) mRNAs in the rat pituitary across the estrous cycle determined with semiquantitative RT-PCR. Pituitaries were collected at 0900 h on metestrus (M), diestrus (D), proestrus (P-AM), and estrus (E-AM). Additional samples were collected at 1700 h on proestrus (P-PM) and estrus (E-PM). Specific ER PCR products were quantified and corrected for coamplified RPL19. Bars represent the mean ⫾ SEM of two or four independent amplification reactions from two separate groups of rats normalized to metestrous values. Asterisks represent a significant difference from metestrus (P ⬍ 0.05). Lower panels, Representative PCR signals on ethidium bromide-containing agarose gels. Size markers are noted in the left-most lane. Specific ER signals are noted with open arrowheads, and the RPL19 control mRNA is noted with solid arrowheads.

after 20 cycles of PCR. ER␤ was relatively more abundant in cells with lactotrope characteristics (GH3, PR1, and MMQ), although it was clearly present in all cell types. An additional smaller amplification product was detected for ER␤ in ␣T3 cells, but Southern analysis with a nested probe did not hybridize to this product (data not shown). Therefore, we believe it is a nonspecific amplification. TERP-1 was most highly expressed in MMQ and RC4B cells. MMQ and PR1 cells also expressed high levels of TERP-2 mRNA, a larger TERP splice variant that contains additional intronic sequence between the 31-bp TERP-specific sequence and exon 5 of ER␣ (5). In general, cells with lactotrope characteristics expressed the highest levels of TERP, consistent with the in vivo expression of TERP in normal lactotropes (18). TERP mRNA was not expressed in two other pituitary cell lines,

Size (bp)

235 258 370 192

AtT20 corticotropes and ␣TSH thyrotrope precursors (not shown). Both steroid (E, DHT, P, and P plus E) and peptide (PACAP and GnRH) treatments were employed to examine the cell type-specific regulation of ER isoforms in GH3, RC4B, ␣T3, and L␤T2 cells. No significant changes in ER expression were observed in ␣T3 cells with any treatment, and ER␣ mRNA was not significantly altered by any treatment in cells with lactotrope characteristics (GH3 and RC4B; Fig. 6, A and B). However, P in combination with E increased ER␣ mRNA levels in the differentiated gonadotrope L␤T2 cell line, and the effect of P alone did not reach significance (P ⬍ 0.06; Fig. 6C). ER␤ mRNA levels were suppressed by P alone (L␤T2) or P plus E (GH3; Fig. 6), but this effect was not significant in RC4B cells. TERP-1 mRNA was consistently up-regulated by 1 or 4 days of E in pituitary cell lines, in agreement with previous results noted in vivo (Fig. 6). In the lactotropic cell lines (GH3 and RC4B) DHT also increased TERP-1 expression, in agreement with our in vivo data (Fig. 6, A and B). No such increase was seen in L␤T2 cells (Fig. 6C), even though these cells possess functional androgen receptors (our unpublished observations). Conversely, in cell lines with gonadotropic characteristics (RC4B and L␤T2), P increased TERP-1 expression (Fig. 6, B and C). Expression of TERP-1 was unaffected by GnRH treatment in cell lines that express GnRH receptors (L␤T2 and RC4B; data not shown), and GnRH did not affect E-induced TERP expression. In contrast, PACAP stimulated TERP-1 mRNA in GH3 and L␤T2 cells (Fig. 6, A and C). ER protein expression and regulation in pituitary cell lines

ER␣ protein was readily detected in all cell types, with differential expression among cells. Cells with the highest levels of ER␣ mRNA (␣T3 and GH3) had the highest levels of ER␣ protein. Estrogen treatment decreased ER␣ protein in all cell types between 25– 60%, and this decrease was also observed in the presence of E and P (Fig. 7A and Table 3), even though no changes were observed in mRNA expression. These results were confirmed in duplicate immunoblots from two or three separate treatment groups. TERP-1 protein levels were very low relative to ER␣ and were detectable in only some experiments in ␣T3, RC4B, and L␤T2 cells. However, E stimulated TERP-1 protein when it was detected (Fig. 7B). DHT also increased TERP-1 protein expression in RC4B, L␤T2, and ␣T3 cells. In agreement with mRNA expression data in L␤T2 cells, P alone increased TERP-1 protein, but PACAP had no significant effect. Thus, in general, stimulation of TERP-1 protein occurred in parallel with mRNA; however, ER␣ protein, but not mRNA, was significantly

2180

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

Endo • 2000 Vol 141 • No 6

FIG. 5. Expression of ER␣ and TERP-1 and ␤-actin protein in rat pituitary. Ovariectomized rats were treated for 3 days with oil vehicle (n ⫽ 2) or E (20 ␮g/100 g BW; n ⫽ 2). Seventy-five micrograms of protein were separated in each lane and detected with an antibody against the C-terminus of ER␣. Signal was detected with enhanced chemiluminescence. ER signals were corrected for loading with ␤-actin, and the ratio of TERP to ER␣ is shown below each lane.

suppressed by E treatment. ER␤ protein could not be reliably detected in cell lines using two commercially available antibodies; however, ER␤ could be detected in transfected COS-1 cells and rat ovary. Thus, ER␤ protein levels in pituitary cell lines are likely to be very low. Effect of decreased ER␣ protein on estrogenmediated transcription

We examined the possibility that reduced ER␣ protein levels would reduce E responsiveness in pituitary cells. Pretreatment of GH3 cells with 10 nm E markedly diminished E-induced transcription of a PRL promoter-containing reporter gene (Fig. 8). In addition to competing with E for ER binding, the antiestrogen ICI182,780 increases ER turnover and reduces ER levels in the nucleus (27). Pretreatment of GH3 cells with ICI182,780, reduced E responsiveness (Fig. 8). Both E and ICI182,780 pretreatments markedly reduced ER␣ protein levels (Fig. 8). After an additional 24 h of treatment with E for determination of reporter gene activity, ER␣ levels were all similarly reduced (data not shown). Discussion

FIG. 4. Effect of steroid treatment on ER␣ (open bars), ER␤ (solid bars), and TERP-1 (hatched bars) mRNA expression in rat pituitaries determined with semiquantitative RT-PCR. A, Specific ER PCR products were quantified and corrected for coamplified RPL19. Bars represent the mean ⫾ SEM of two to four independent amplification reactions from two separate groups of rats normalized to the oiltreated control values. Representative PCR signals on ethidium bromide-containing agarose gels are also shown. Size markers are noted in the left-most lane. Specific ER signals are noted with open arrowheads, and the RPL19 control mRNA is noted with solid arrowheads. B, Representative PCR signals from coamplification of ER␤1 (lower band) and ER␤2 (upper band). C, Effects of antide and bromocriptine on E-regulated ER expression in female rats. Bars are as described in A. Asterisks represent a significant (P ⬍ 0.05) difference from oiltreated controls.

Estrogen feedback at the pituitary plays a critical role in the expression and secretion of gonadotropins and PRL and is necessary for the maintenance of reproductive function. One mechanism for the varied cell-specific actions of E in the rat pituitary is the modulated expression of three different ER isoforms, ER␣, ER␤, and TERP-1. Both ER␣ and ER␤ have been reported to be transcribed from several alternate first exons and multiple promoters (20, 21, 28, 29). TERP-1 is transcribed from a promoter distinct from those for ER␣ (5, 25). Thus, these three isoforms can be differentially regulated and may result in different transcriptional capacities. In model systems, ER␣ stimulates E-regulated genes better than ER␤ (10, 30), and TERP-1 modifies the activity of full-length ERs (12). Previous studies (5, 6, 8, 17, 18) showed that E regulates ER isoforms in the rat pituitary. The present study demonstrates that pituitary TERP-1 and ER␤ mRNA levels can be physiologically regulated in vivo as well as in cell lines by steroid hormones and other peptides. Similarly, ER␣ and TERP-1 protein levels are also modulated. These results have implications for steroid feedback during the estrous cycle

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

2181

FIG. 7. Expression of ER␣ (A) and TERP-1 (B) protein in pituitary cell lines. Cells were treated as described in Fig. 6. Fifty micrograms of protein were separated in each lane and detected with an antibody against the C-terminus of ER␣. Signal was detected with enhanced chemiluminescence. ER␣ blots were exposed from 15– 60 sec, and TERP-1 blots were exposed from 3–30 min. In vitro translated TERP-1 is shown for comparison in the first lane of each blot in B.

FIG. 6. Expression of ER␣ (open bars), ER␤ (solid bars), and TERP-1 (hatched bars) mRNAs in GH3 (A), RC4B (B), and L␤T2 (C) cells determined with semiquantitative RT-PCR. Cells were treated with vehicle (NT); E for 1 or 4 days; DHT, P, or P plus E for 4 days; or vehicle for 3 days and PACAP (PAC) for 18 h. Bars represent at least three separate RT-PCRs from two separate groups of cells. ERs were coamplified with RPL19 ribosomal protein mRNA as an amplification/ loading control, and data are normalized for the NT condition for each ER.

and may provide one means for termination of the proestrous LH and PRL surge by reducing the transcriptional capacity of ER. To begin to elucidate the cell-specific expression and regulation of ERs in the pituitary, we used two in vivo treatments and examined rodent pituitary cell lines of various lineages. Previous studies have shown that ER␣ is expressed primarily in lactotropes and gonadotropes. In the normal human pituitary, ER␣ mRNA is expressed in 50% of lactotropes, 70 – 83% of gonadotropes (31), and fewer than 5% of somatotropes, thyrotropes, or corticotropes (31). In the adult female rat, ER␣ mRNA is expressed in 45% of lactotropes and 25% of gonadotropes (6). In humans, ER␤ is present in pituitary adenomas of every cell type and is the only ER form expressed in tumors that express GH exclusively (32, 33). ER␤ expression in rodent pituitaries suggests that levels are ex-

tremely low. Although we detected ER␤ mRNA by RT-PCR from whole pituitary, we did not detect significant levels of ER␤ mRNA using tritiated riboprobes. Similarly, Couse et al. (19) did not detect ER␤ mRNA in mouse pituitaries using RNase protection, although ER␤ has been detected in the intermediate lobe of the adult rat pituitary using 35S-labeled riboprobes (15). However, Wilson et al. (16) recently demonstrated that prepubertal female rats have a relative abundance of ER␤ in the pituitary, and 15% of lactotropes and 85% of FSH-positive gonadotropes express ER␤ mRNA at low levels. Similarly, Mitchner and colleagues (6) concluded that 25% of lactotropes and 15% of LH-expressing gonadotropes in the adult female pituitary contain ER␤ mRNA. Using RT-PCR we detected ER␤ mRNA in all pituitary cell lines tested, but the pattern of expression differed from that of ER␣. Cells with more lactotropic characteristics (GH3, PR1, and MMQ) expressed significantly more ER␤ than cells with gonadotrope characteristics (␣T3 and L␤T2). ER␤ protein was not detected in any of the cell lines, possibly due to both low expression levels and the relative insensitivity of the ER␤ antibodies. TERP-1 mRNA has been reported in enriched rat lactotrope populations, but is absent in enriched gonadotrope populations using Northern blot analysis (18). The present study suggests that TERP-1 may be present at some level in both lactotropes and gonadotropes, but lactotrope cell lines express more TERP-1 mRNA. ER␣ mRNA levels did not change across the estrous cycle or in response to most treatments in vivo or in cell lines, in agreement with previous studies showing only moderate to no regulation in vivo and in GH3 cells (8, 18, 34). Previous data from our laboratory demonstrated that total ER␣ (ER␣ and TERP-1) mRNA on Northern blots was stimulated by E treatment (5), and subsequent studies demonstrated that TERP-1 mRNA was differentially stimulated by E (5, 8). Although we observed a 2- to 3-fold E-induced stimulation of ER␣ by RNase protection, our RT-PCR product representing ER␣

2182

Endo • 2000 Vol 141 • No 6

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

TABLE 3. ER␣ protein expression in pituitary cell lines Treatment

GH3

RC4B

␣T3

L␤T2

NT 1 day E 4 days E DHT P P⫹E PACAP

1.0 0.35 ⫾ 0.01 0.28 ⫾ 0.05 1.06 ⫾ 0.19 1.09 ⫾ 0.23 0.51 ⫾ 0.21 1.26 ⫾ 0.32

1.0 0.49 ⫾ 0.07 0.47 ⫾ 0.15 0.69 ⫾ 0.13 0.33 ⫾ 0.03 0.21 ⫾ 0.10 0.98 ⫾ 0.39

1.0 0.51a 0.41 ⫾ 0.10 0.86 ⫾ 0.67 0.92 ⫾ 0.55 0.33 ⫾ 0.17 0.70 ⫾ 0.14

1.0 0.53 ⫾ 0.16 0.57 ⫾ 0.40 0.91 ⫾ 0.15 0.83 ⫾ 0.58 0.35 ⫾ 0.11 1.19 ⫾ 0.41

Values represent the normalized mean ⫾ SD from two to four individual experiments. a Experiment was perfomed only once.

FIG. 8. Effect of reducing ER␣ protein expression on E responsiveness in GH3 cells. Cells were pretreated with E (10 nM; 18 h) or ICI 182,780 (10 nM; 6 h) or were left untreated (NT). Cells were transiently transfected with 2 ␮g ⫺2.5 PRLluc reporter and treated with or without E (10 nM) for 22 h. Data from each pretreatment were normalized separately to its own vehicle treatment (NT). Bars represent the mean ⫾ SEM of three experiments. Lower panel, Representative immunoblot of ER␣ expression in GH3 cells after pretreatments, as described above.

and the ER␣ exon 5/6 deletion variant mRNA was not changed. We believe that in part this is due to a larger dynamic range for the RNase protection. Thus, the change in TERP-1 mRNA expression observed in the present study is also smaller than that observed previously (5, 8). This is probably due to better detection of low levels of expression with RT-PCR. Interestingly, other investigators using RTPCR found a similar lack of effect of E on ER␣ (34). Immunoblotting demonstrated that full-length ER␣ protein levels could be highly regulated in vivo in the pituitary, even in the face of relatively steady state mRNA levels. Other studies have shown small reductions in E binding in the pituitary around proestrus, suggesting that total ER protein levels are reduced (35–37). The decrease in ER␣ protein levels provides one mechanism by which E-induced positive feedback may be limited, and such posttranscriptional changes in ER␣ may be an important mechanism for ER regulation in both the pituitary and other tissues (38). In the present study ER␣ protein was readily suppressed by E in all pituitary cell lines tested, including PR1 cells (data not shown). Dissociation of mRNA and protein expression may provide an important mechanism for cell-specific actions of E. Recent findings by Alarid et al. (38) show that in PR1 pituitary cells the reduction

of ER␣ in response to E occurs via proteosomal degradation. A similar dissociation of mRNA and protein expression has been demonstrated for the androgen receptor (39) and thyroid hormone receptor (40, 41). Functionally, decreases in ER protein levels can lead to decreases in E responsiveness, as demonstrated by transfection studies in the present study. Estrogen pretreatment elevated absolute luciferase expression before E treatment, even though ER␣ levels were reduced. It is unlikely that ERs were saturated, because E treatment still led to a 2.5-fold increase in PRLluc expression. However, ICI182,780 pretreatment led to similar reductions in ER␣ protein levels without increasing basal reporter activity. The subsequent response to E was the same as that with E pretreatment, suggesting that the reduction in E responsiveness results from a reduction in ER expression rather than an elevation in basal ER activity. Because TERP protein was undetectable in GH3 cells, it is unlikely that E-induced TERP expression contributed substantially to the reduced E responsiveness of the PRLluc reporter in pretreated cells because TERP levels must exceed those of full-length ERs to inhibit responsiveness (12, 13). ER␤ mRNA levels were lowest on the morning of proestrus, when in vivo steroid levels are highest. Changes in steroid levels probably mediate this effect in females because treatment of ovariectomized animals with E, E plus P, or DHT suppressed ER␤ mRNA expression. This is the first demonstration of steroid down-regulation of ER␤ mRNA in the pituitary; however, Mitchner et al. (6) observed a similar, but nonsignificant, effect in E-treated rats. ER␤ mRNA down-regulation is also intriguing in light of recent evidence that ER␤ expression in the female rat pituitary decreases after puberty, again suggesting an effect of increased circulating steroids (16). During the estrous cycle, this suppression of ER␤ mRNA could be associated with a decrease in protein levels around the time of the proestrous surge and, like ER␣ suppression, would lead to a decrease in E-mediated transcription. In contrast to our in vivo data and a published report showing up-regulation of ER␤ by 24 h of E treatment in GH3 cells (34), E and DHT had no effect on ER␤ mRNA expression in pituitary cell lines. E and DHT could affect the pituitary indirectly by altering the release of hypothalamic factors, but we saw no direct evidence of regulation by GnRH in cell lines or by GnRH or dopamine in vivo. Alternatively, the effects on ER␤ in whole pituitary might be due to direct or indirect effects on cell types other than lactotropes or gonadotropes. During the estrous cycle, the fall in ER␤ precedes increases in circulating P, and P alone did not alter ER␤

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

mRNA in rats, GH3 cells, or RC4B cells in the absence of E. Although this may be due to a requirement for E induction of PR, the observation that P alone decreases ER␤ in L␤T2 cells suggests that P might have a role distinct from that of E. Two additional lines of evidence support a role for P in ER␤ regulation. A recent report demonstrated that P reduces ER␤ mRNA levels in T-47D breast cancer cells, and an inverse correlation between ER␤ mRNA levels and PR expression was observed in breast tumor biopsies (42). In contrast to ER␣ and ER␤, TERP-1 mRNA began to rise on diestrus and peaked on the morning of proestrus, when in vivo steroid levels are highest, as previously demonstrated (8, 18). Estrogen consistently stimulates TERP-1 mRNA expression in vivo (5, 6, 8, 17, 18), in GH3 cells (34), and in pituitary cell lines in the present study. The E-induced increase in TERP-1 mRNA levels is recapitulated in the expression of TERP-1 protein in ovariectomized rats treated with E, in which TERP-1 protein levels equal or exceed those of the full-length ER␣, and in our pituitary cell lines, where TERP-1 protein can be detected. At the ratios observed in vivo, TERP-1 inhibits ER-mediated transcription (12), and the transcriptional activity of remaining ER␣ and ER␤ at proestrus would be further reduced. Although the results from no one cell line completely describe events in the intact rat pituitary gland, several regulatory pathways appear to be identical between the pituitary and lactotrope and gonadotrope cell lines. These include the decrease in ER␣ protein with E treatment, the decrease in ER␤ mRNA with E plus P treatment, and the increase in TERP-1 mRNA and protein with E. TERP-1 has biphasic effects on ER␣ and ER␤ activity, stimulating ER at low ratios and inhibiting activity at a ratio of 1:1. Thus, after proestrus, ER␣ protein and potentially ER␤ protein will be decreased, and TERP-1 protein will approach or exceed a ratio of 1:1 with full-length receptors. Under these conditions, TERP can further suppress ER-mediated transcription (12, 13). We hypothesize that increased TERP expression in the face of decreased full-length ER levels effectively suppresses E actions in the pituitary and may play a role in terminating the proestrous surge.

7. 8. 9. 10. 11. 12. 13. 14.

15. 16. 17. 18. 19.

20. 21. 22.

23. 24.

Acknowledgments

25.

The authors thank the Markey Center for Cell Signaling at the University of Virginia for use of the Fluoroimager and Robocycler, the core laboratories of the Center for the Study of Reproduction at the University of Virginia, and Drs. Jack Gorski and Tae-Yon Chun for PR1 cell extracts.

26.

References 1. Lieberman ME, Maurer RA, Gorski J 1978 Estrogen control of prolactin synthesis in vitro. Proc Natl Acad Sci USA 75:5946 –5949 2. Lieberman ME, Maurer RA, Claude P, Wiklund J, Wertz N, Gorski J 1981 Regulation of pituitary growth and prolactin gene expression by estrogen. Adv Exp Med Biol 138:151–163 3. Nedvidkova J, Haluzik M, Pacak K, Schreiber V, Starka L 1998 Estrogenic effect of estradiol-sulfamate on the male rat anterior pituitary. J Steroid Biochem Mol Biol 67:359 –362 4. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schutz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, et al 1995 The nuclear receptor superfamily: the second decade. Cell 83:835– 839 5. Friend KE, Ang LW, Shupnik MA 1995 Estrogen regulates the expression of several different estrogen receptor mRNA isoforms in rat pituitary. Proc Natl Acad Sci USA 92:4367– 4371 6. Mitchner NA, Garlick C, Ben-Jonathan N 1998 Cellular distribution and gene

27. 28.

29. 30. 31.

32.

2183

regulation of estrogen receptors ␣ and ␤ in the rat pituitary gland. Endocrinology 139:3976 –3983 Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson J-Å 1996 Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 93:5925–5930 Friend KE, Resnick EM, Ang LW, Shupnik MA 1997 Specific modulation of estrogen receptor mRNA isoforms in rat pituitary throughout the estrous cycle and in response to steroid hormones. Mol Cell Endocrinol 131:147–155 McInerney EM, Tsai MJ, O’Malley BW, Katzenellenbogen BS 1996 Analysis of estrogen receptor transcriptional enhancement by a nuclear hormone receptor coactivator. Proc Natl Acad Sci USA 93:10069 –10073 Pettersson K, Grandien K, Kuiper GGJM, Gustafsson J-Å 1997 Mouse estrogen receptor beta forms estrogen response element-binding heterodimers with estrogen receptor ␣. Mol Endocrinol 11:1486 –1496 Kuiper GGJM, Lemmen JG, Carlsson B, Corton JC, Safe SH, vanderSaag PT, vanderBurg B, Gustafsson J-Å 1998 Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor ␤. Endocrinology 139:4252– 4263 Schreihofer DA, Resnick EM, Soh AY, Shupnik MA 1999 Transcriptional regulation by a naturally occurring truncated rat estrogen receptor (ER), truncated ER product-1 (TERP-1). Mol Endocrinol 13:320 –329 Resnick EM, Schreihofer DA, Periasamy A, Shupnik MA 2000 Truncated estrogen receptor product-1 suppresses estrogen receptor transactivation by dimerization with estrogen receptors ␣ and ␤. J Biol Chem 275:7158 –7166 Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors ␣ and ␤. Endocrinology 138:863– 870 Shughrue PJ, Lane MV, Scrimo PJ, Merchenthaler I 1998 Comparative distribution of estrogen receptor-␣ (ER-␣) and B (ER-␤) mRNA in the rat pituitary, gonad, and reproductive tract. Steroids 63:498 –504 Wilson ME, Richard H Price J, Handa RJ 1998 Estrogen receptor-␤ messenger ribonucleic acid expression in the pituitary gland. Endocrinology 139:5151– 5156 Shupnik MA, Gordon MS, Chin WW 1989 Tissue-specific regulation of rat estrogen receptor mRNAs. Mol Endocrinol 3:660 – 665 Demay F, Tiffoche C, Thieulant M-L 1996 Sex- and cell-specific expression of an estrogen receptor isoform in the pituitary gland. Neuroendocrinology 63:522–529 Couse JF, Lindzey J, Grandien K, Gustafsson J-k, Korach KS 1997 Tissue distribution and quantitative analysis of estrogen receptor-␣ (ER␣) and estrogen receptor-␤ (ER␤) messenger ribonucleic acid in the wild-type and ER␣knockout mouse. Endocrinology 138:4613– 4621 Grandien K, Berkenstam A, Gustafsson JA 1997 The estrogen receptor gene: promoter organization and expression. J Biochem Cell Biol 29:1343– 69 Koh-Shoda T, Hirata H, Hoshi K, The multiple untranslated first exons of the human estrogen receptor ␤. 81st Annual Meeting of The Endocrine Society, San Diego, CA, 1999, p 490 Goth MI, Charles E, Lyons J, Ellwood ME, Barrett JR, Thorner MO 1996 Chronic estrogen treatment in male rats reveals mammosomatotropes and allows inhibition of prolactin secretion by somatostatin. Endocrinology 137:274 –280 Petersen DN, Tkalcevic GT, Koza-Taylor PH, Turi TG, Brown TA 1998 Identification of estrogen receptor ␤2, a functional variant of estrogen receptor ␤ expressed in normal rat tissues. Endocrinology 139:1082–1092 Camp TA, Rahal JO, Mayo KE 1991 Cellular localization and hormonal regulation of follicle-stimulating hormone and luteinizing hormone receptor messenger RNAs in the rat ovary. Mol Endocrinol 5:1405–17 Tiffoche C, Vaillant C, Thieulant ML, Novel promoter regulating variant of the estrogen receptor transiently expressed in rat pituitary. 81st Annual Meeting of The Endocrine Society, San Diego, CA, 1999, p 523 Colin IM, Jameson JL 1998 Estradiol sensitization of rat pituitary cells to gonadotropin-releasing hormone: involvement of protein kinase C- and calcium-dependent signaling pathways. Endocrinology 139:3796 –3802 Dauvois S, White R, Parker MG 1993 The antiestrogen ICI 182780 disrupts estrogen receptor nucleocytoplasmic shuttling. J Cell Sci 106:1377–1388 Kato J, Hirata S, Koh T, Yamada-Mouri N, Hoshi K, Okinaga S 1998 The multiple untranslated first exons and promoters system of the oestrogen receptor gene in the brain and peripheral tissues of the rat and monkey and the developing rat cerebral cortex. J Steroid Biochem Mol Biol 65:281–293 Iwamoto H, Hirata S, Koh-Shoda T, Hoshi K, The multiple untranslated first exons of the rat estrogen receptor beta gene. 81st Annual Meeting of The Endocrine Society, San Diego, CA, 1999, p 490 McInerney EM, Weis KE, Sun J, Mosselman S, Katzenellenbogen BS 1998 Transcription activation by the human estrogen receptor subtype ␤ (ER␤) studied with ER␤ and ER␣ receptor chimeras. Endocrinology 139:4513– 4522 Friend KE, Chiou YK, Lopes MB, Laws Jr ER, Hughes KM, Shupnik MA 1994 Estrogen receptor expression in human pituitary: correlation with immunohistochemistry in normal tissue, and immunohistochemistry and morphology in macroadenomas. J Clin Endocrinol Metab 78:1497–1504 Shupnik MA, Pitt LK, Soh AY, Anderson A, Lopes MB, Laws Jr ER 1998 Selective expression of estrogen receptor ␣ and ␤ isoforms in human pituitary tumors. J Clin Endocrinol Metab 83:3965–3972

2184

ESTROGEN RECEPTORS IN RAT PITUITARY AND CELL LINES

33. Chaidarun SS, Swearingen B, Alexander JM 1998 Differential expression of estrogen receptor-beta (ER beta) in human pituitary tumors: functional interactions with ER alpha and a tumor-specific splice variant. J Clin Endocrinol Metab 83:3308 –3315 34. Mitchner NA, Garlick C, Steinmetz RW, Ben-Jonathan N 1999 Differential Regulation and action of estrogen receptors ␣ and ␤ in GH3 cells. Endocrinology 140:2651–2658 35. Thrower Sn, Lim L 1980 A comparison of the relationships between progestin receptors and oestrogen receptors in neural and non-neural target tissues of the rat during the oestrous cycle. Biochem J 190:691– 695 36. Bu¨chi KA, Keller PJ 1980 Variation of hypophyseal estradiol receptors during rat estrous cycle. J Steroid Biochem 12:347–349 37. Diekman MA, Anderson LL 1983 Quantification of receptors for estradiol-17␤ and progesterone in the pituitary and hypothalamus of gilts during the estrous cycle and early pregnancy. Biol Reprod 29:946 –952

Endo • 2000 Vol 141 • No 6

38. Alarid ET, Bakopoilos N, Solodin N 1999 Proteosome-mediated proteolysis of estrogen receptor: a novel component in autologous down-regulation. Mol Endocrinol 13:1522–1534 39. Yeap BB, Krueger RG, Leedman PJ 1999 Differential posttranscriptional regulation of androgen receptor gene expression by androgen in prostate and breast cancer cells. Endocrinology 140:3282–3291 40. Bigler J, Eisenman RN 1988 c-erbA encodes multiple proteins in chicken erythroid cells. Mol Cell Biol 8:4155– 4161 41. Strait KA, Schwartz HL, Perez-Castillo A, Oppenheimer JH 1990 Relationship of c-erbA mRNA content to tissue triiodothyronine nuclear binding capacity and function in developing and adult rats. J Biol Chem 265:10514 –10521 42. Dotzlaw H, Leygue E, Watson PH, Murphy LC 1999 Estrogen receptor-␤ messenger RNA expression in human breast tumor biopsies: relationship to steroid receptor status and regulation by progestins. Cancer Res 59:529 –532