help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hall, J. M.
Right arrow Articles by McDonnell, D. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hall, J. M.
Right arrow Articles by McDonnell, D. P.
Endocrinology Vol. 140, No. 12 5566-5578
Copyright © 1999 by The Endocrine Society

ARTICLES

The Estrogen Receptor ß-Isoform (ERß) of the Human Estrogen Receptor Modulates ER{alpha} Transcriptional Activity and Is a Key Regulator of the Cellular Response to Estrogens and Antiestrogens1

Julie M. Hall2 and Donald P. McDonnell

Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, North Carolina 27710

Address all correspondence and requests for reprints to: Dr. Donald P. McDonnell, Department of Pharmacology and Cancer Biology, Duke University Medical Center, P.O. Box 3813, Durham, North Carolina 27710. E-mail: mcdon016{at}acpub.duke.edu


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The human estrogen receptor {alpha} (ER{alpha}) and the recently identified ERß share a high degree of amino acid homology; however, there are significant differences in regions of these receptors that would be expected to influence transcriptional activity. Consequently, we compared the mechanism(s) by which these receptors regulate target gene transcription, and evaluated the cellular consequences of coexpression of both ER subtypes. Previously, it has been determined that ER{alpha} contains two distinct activation domains, ER{alpha}-AF-1 and ER{alpha}-AF-2, whose transcriptional activity is influenced by cell and promoter context. We determined that ERß, like ER{alpha}, contains a functional AF-2, however, the ERß-AF-2 domain functions independently within the receptor. Of additional significance was the finding that ERß does not contain a strong AF-1 within its amino-terminus but, rather, contains a repressor domain that when removed, increases the overall transcriptional activity of the receptor. The importance of these findings was revealed when it was determined that ERß functions as a transdominant inhibitor of ER{alpha} transcriptional activity at subsaturating hormone levels and that ERß decreases overall cellular sensitivity to estradiol. Additionally, the partial agonist activity of tamoxifen manifest through ER{alpha} in some contexts was completely abolished upon coexpression of ERß. In probing the mechanisms underlying ERß-mediated repression of ER{alpha} transcriptional activity we have determined that 1) ER{alpha} and ERß can form heterodimers within target cells; and 2) ERß interacts with target gene promoters in a ligand-independent manner. Cumulatively, these data indicate that one role of ERß is to modulate ER{alpha} transcriptional activity, and thus the relative expression level of the two isoforms will be a key determinant of cellular responses to agonists and antagonists.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE HUMAN ESTROGEN receptor (ER) belongs to the nuclear receptor superfamily of ligand-inducible transcription factors (1), whose members include the receptors for steroids, thyroid hormone, retinoic acid, vitamin D, and orphan receptors for which no ligands have yet been identified. The mechanism of action of ER is similar to that of other nuclear receptors. In the absence of hormone, the receptor is sequestered within the nuclei of target cells in a multiprotein inhibitory complex. The binding of ligand induces an activating conformational change within ER, an event that promotes homodimerization and high affinity binding to specific DNA response elements (EREs) located within the regulatory regions of target genes (2). In addition to the classic ligand-mediated activation pathway, it has been shown that ER can be activated in the absence of ligand by growth factors or other agents that elevate intracellular cAMP levels (3, 4). Although the physiological importance of the ligand-independent signaling pathways remains to be determined, it has been shown in ER knockout mice that the uterotropic responses to both 17ß-estradiol and epidermal growth factor require a functional ER (5). Thus, ER appears to be a key point of convergence of multiple signaling pathways, an observation that complicates our understanding of the pharmacology of estrogens and antiestrogens.

Until recently it was considered that a single ER was responsible for all of the biological actions of estrogens and antiestrogens. However, the recent identification of ERß (6, 7) has indicated that the cellular responses to ER ligands are far more complex. The two estrogen receptors, ER{alpha} and ERß, have similar overall structures, displaying a high degree of amino acid conservation in the central DNA-binding domain (DBD) and moderate conservation in the ligand-binding domain (LBD; C-terminus), but considerable divergence in the amino-terminus. Not surprising, therefore, ER{alpha} and ERß interact with the same DNA response elements (8) and exhibit similar, but not identical, ligand binding characteristics. Although a specific physiological role for ERß remains to be defined, its identification has provided a potential explanation for the biological actions of estrogen(s) in cells where no immunoreactive ER{alpha} could be detected. Interestingly, preliminary localization studies have revealed that there are many tissues in which both ER subtypes are coexpressed (6, 7, 9). Thus, the impact of ERß on estrogen biology is likely to occur as a consequence of 1) direct actions of ERß, where it is responsible for regulating target gene transcription; and 2) indirect activities, where ERß modulates ER{alpha} action in tissues where they are coexpressed.

Although the precise mechanism by which ER regulates transcription remains to be determined, considerable progress has been made in defining the domains within ER{alpha} required for its activity. Specifically, it has been demonstrated that the transcriptional activity of ER{alpha} is mediated by two activation functions (AFs) located in the amino-terminal (AF-1) and carboxyl-terminal (AF-2). Although both of these AFs function in a synergistic manner in most circumstances, they can also function independently in a cell- and promoter-specific manner, an activity that may explain the tissue-selective agonist activity of some ER ligands (10, 11). In this regard, it has been observed that 17ß-estradiol can function as an agonist in all environments regardless of whether AF-1 or AF-2 is the dominant activator. Not surprisingly, therefore, the pure antiestrogen ICI 182,780, which inhibits the activity of both AF-1 and AF-2, completely blocks the ability of ER{alpha} to activate transcription through classical ERE-mediated pathways. Unlike the pure antiestrogens, however, the relative agonist/antagonist activities of most other antiestrogens are determined by the cell and promoter context. For instance, compounds such as tamoxifen inhibit AF-2 activity, and consequently function as antagonists in all environments where AF-2 is required. In contexts where AF-1 is the dominant activator, on the other hand, tamoxifen manifests partial agonist activity. These observations led to the hypothesis that the tissue-selective biological activity exhibited by selective ER modulators (SERMs) such as tamoxifen, reflects their ability to differentially regulate AF-1 or AF-2. However, the identification of ER ligands such as raloxifene and GW5638, which function as estrogens in the bone and the cardiovascular system but do not appear to function as either AF-1 or AF-2 agonists, indicates that the existing models of ER pharmacology are incomplete (12, 13). Clearly, they must now be expanded to include a consideration of the impact of ERß. It is likely that the existence of ER{alpha} and ERß will be as important to ER pharmacology as the two progesterone receptor subtypes, PR-A and PR-B, are to the pharmacology of progestins and antiprogestins. We base this hypothesis on our earlier studies of PR action, where it was shown that PR-A and PR-B were not functionally identical. Specifically, we observed that both receptor forms could manifest autonomous activity in some cell contexts, whereas in others the A isoform was a weak transcriptional activator and, in fact, functioned as a transdominant inhibitor of human PR-B activity (14). The possibility that there were similarities between these two systems prompted us to explore the impact of ERß on the pharmacology of ER{alpha}. Although ER{alpha}/ERß are not derived from the same gene as are the two forms of PR, we believed that it would be useful to consider ER{alpha}/ERß as having a similar relationship as PR-A/B. The aim of this study, therefore, was to compare the transcriptional activities of ER{alpha} and ERß and to evaluate the contribution of ERß to the overall pharmacology of estrogens and antiestrogens.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Biochemicals
DNA restriction and modification enzymes were obtained from Roche Molecular Biochemicals (Indianapolis, IN), New England Biolabs, Inc. (Beverly, MA), or Promega Corp. (Madison, WI). PCR reagents were obtained from Perkin Elmer Corp. (Norwalk, CT) or Promega Corp. 17ß-Estradiol and 4-hydroxytamoxifen were purchased from Sigma Chemical Co. (St. Louis, MO). The ER antagonist ICI 182,780 was a gift from Dr. Alan Wakeling (Zeneca Pharmaceuticals, Macclesfield, UK). Raloxifene was a gift from Dr. Eric Larsen, Pfizer, Inc. (Groton, CT). GW7604 was a gift from Dr. Tim Willson (Glaxo-Wellcome, Research Triangle Park, NC). Idoxifene was a gift from Dr. Maxine Gowan (SmithKline Beecham, King of Prussia, PA). The mouse monoclonal anti-FLAG antibody was purchased from Sigma Chemical Co. Secondary antibodies, Hybond-C extra transfer membranes, and enhanced chemiluminescence reagents were purchased from Amersham Pharmacia Biotech (Arlington Heights, IL).

Plasmids
The mammalian expression plasmid for ER{alpha} (pRST7ER) has been described previously (11). Plasmids expressing ER{alpha} mutants (ER-AF1, ER-AF2, ER-Null) were also described previously (11). The ERß expression plasmid, pRST7ERß, was constructed as follows. A PCR 3.1 vector (Invitrogen, Carlsbad, CA) containing the human ERß coding sequence (amino acids 1–477; gift from Dr. Mark Nuttall, SmithKline Beecham) was digested with HindIII and XbaI, and ERß was ligated into pRST7ER, previously digested with HindIII and XbaI to remove the ER{alpha} complementary DNA (cDNA). An epitope-modified series of ER{alpha} and ERß expression vectors was also created. Specifically, an amino-terminus FLAG tag was added to the ER{alpha} and ERß expression vectors by PCR. The sequences of the oligonucleotides for PCR were 5'-GTGGACGTCGACATGGACTACAAAGACGACGACGACAAAATGACCATGACC-CTCCAC (forward) and 3'-GTGGAGGGATCCTCAGACTGTGGCAG-GGAAACC (reverse) for ER{alpha} and 5'-GTGGACGTCGACATGGACTACAAAGACGACGACGACAAAATGAATTACAGCATTCCC (forward) and 3'-GTGAGGTCTAGATCACTGAGACTGTGGGTT (reverse) for ERß.

The mammalian expression plasmid for the ERß mutant ERß-AF1 was constructed by site-directed mutagenesis. The pRST7ERß plasmid was used as the template, and the three point mutations were created using PCR-based oligonucleotide-directed mutagenesis, according to the manufacturer’s protocol (Stratagene, La Jolla, CA). The sequences of the oligonucleotides used for PCR were 5'-GTGGTCCCAGTGTATGCCCTGCTGCTGGCG ATGCTGGCTGCCCACGTGCTTCGCGGG (forward) and 5'-CCCGCGAAGCACGTGGGCAGCCAGCATCGCCAGCAGCAGGGCATACAC TGGGACCAC (reverse).

The plasmid ERß-AF2 was constructed as follows: an empty pRST7 vector was first created by digesting pRST7ER with HindIII and SmaI to remove the ER{alpha} cDNA, blunt ends within the vector were created with Klenow, and the vector was recircularized by ligation. The pRST7 vector was digested with XbaI, and a PCR fragment was generated from pRST7ERß (containing amino acids 90–477 of the coding sequence for ERß) and ligated into the pRST7 vector. The sequences of the oligonucleotides used for PCR were 5'-GTGAGGTCTAGAATGAAGAGGGATGCTCACTTC (forward) and 3'-GTGAGGTCTAGATCACTGAGACTGTGGGTT (reverse).

To compare the stability of the ERß mutants in transfected cells, we created a duplicate set of vectors in which an amino-terminus FLAG tag was added to the expression vectors for the ERß mutants by PCR. The sequences of the oligonucleotides used for PCR for the ERß-AF1 mutant were the same as those used to construct the FLAG-tagged wild-type ERß. The oligonucleotides used to create the FLAG-tagged ERß-AF2 mutant were 5'-GTGGACTCTAGAATGGACTACAAAGACGACGACGACAAATGCGCTGTCTG CAGCGATTAC (forward) and 3'-GTGAGGTCTAGATCACTGAGACTGTGGGTT (reverse).

The GAL4-DBD-ER N-terminus fusion constructs were cloned into the pBK-cytomegalovirus (CMV) mammalian expression vector. Construction of the pBKC-DBD plasmid has been described previously (15). pBKC-DBD-ER{alpha}-(1–182) was constructed as follows. The pBKC-DBD plasmid was digested with EcoRI and ClaI, and a PCR fragment (containing the coding sequence for the first 182 amino acids of ER{alpha}) was generated from pRST7ER and ligated into these sites. The sequences of the oligonucleotides for PCR were 5'-GTGCAGGAATTCATGACCATGACCCTCCAC (forward) and 5'-GTGCAGATCGATAGTCTCCTTGGCAGATTC (reverse). pBKC-DBD-ERß-(1–95) was constructed as follows. The pBKC-DBD plasmid was digested with EcoRI and ClaI, and a PCR fragment (containing the coding sequence for the first 95 amino acids of ERß) was generated from pRST7ERß and ligated into these sites. The sequences of the oligonucleotides for PCR were 5'-GTGCAGGAATTCATGAATTACAGCATTCCC (forward) and 5'-GTGCAGATCGATGAAGTGAGCATCCCTCTT (reverse).

pBKC-DBD-ER{alpha}-LBD(3x) was constructed as follows: the pBKC-DBD plasmid was digested with EcoRI and ClaI, and a PCR fragment (containing the coding sequence for amino acids 282–595 of ER{alpha}) was generated from pRST7-ER-AF-1 and ligated into these sites. The sequences of the oligonucleotides for PCR were 5'-GTGCAGGAATTCATGTCTGCTGGAGACATGAGA (forward) and 3'-GTGCAGATCGATGACTGTGGCAGGGAAACC (reverse).

All of the PCR-based constructs were sequenced to verify the accuracy of the amplified sequences.

Cell culture and transient transfection assays
HepG2, HeLa, and 293 cells were maintained in MEM (Life Technologies, Inc.) supplemented with 10% FCS (Life Technologies, Inc.). MCF-7 and SKBR3 cells were maintained in DMEM (Life Technologies, Inc.) supplemented with 10% FCS (Life Technologies, Inc.). Cells were plated in 24-well plates (coated with gelatin for transfections of HepG2 cells) 24 h before transfection. DNA was introduced into the cells using lipofectin (Life Technologies, Inc.). Triplicate transfections were performed using 3 µg total DNA. In standard transfections, 1500 ng reporter (C3-Luc, 3x-ERE-TATA-Luc, or 5x-GAL4-TATA-Luc), 500 ng receptor (pRST7ER, pRST7ERß, ER mutants, or GAL4 fusions), 100 ng of the pBKC-ßgal normalization vector (16), and 900 ng of the control vector pBSII-KS (Stratagene) were used. The reporter C3-Luc contains the estrogen-responsive complement 3 gene promoter, and the 3x-ERE-TATA-Luc reporter contains three copies of the vitellogenin ERE. The reporter 5x-GAL4-TATA-Luc (a gift from Dr. Xiao-Fan Wang, Duke University Medical Center) contains five palindromic copies of the GAL4 transcription factor response element cloned into pGL2-TATA-Inr (Stratagene). Cells were incubated with the DNA/lipofectin mix for 3 h, then washed with PBS and incubated with the appropriate hormone in phenol red-free medium containing 10% charcoal-stripped FCS (HyClone Laboratories, Inc., Logan, UT) for 48 h. Luciferase and ß-galactosidase assays were performed as described previously (17). All experiments were repeated a minimum of three times.

Western immunoblot analysis
293 cells (human embryonic kidney cells) were transfected with the expression plasmids for ER{alpha}, ERß, or the ERß mutants. Whole cell extracts were prepared as described previously (18). Fifty micrograms of whole cell extracts for each sample were run on a 10% SDS-PAGE gel and transferred to nitrocellulose. Immunoblotting was performed using a mouse monoclonal anti-FLAG antibody. Immunocomplexes were detected by ECL.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The {alpha}- and ß-forms of the human ER are functionally distinct
We used a cotransfection assay reconstituted in HepG2 (human hepatoma) or HeLa (human cervical carcinoma) cells to compare the transcriptional activities and ligand responsiveness of ER{alpha} and ERß. These cell lines were chosen for our studies because they require exogenous ER{alpha} or ERß to activate ERE-mediated transcription, enabling an assessment of the transcriptional responses of each receptor in isolation. Specifically, the ER expression vectors (pRST7ER and pRST7ERß) were transiently transfected into either HepG2 or HeLa cells together with the estrogen-responsive reporter 3x-ERE-TATA-Luc or C3-Luc. The input concentration for each expression vector used in these studies was equivalent and was that which gave a maximal response under the conditions of the assay. In addition, using a duplicate set of vectors in which we added an amino-terminus FLAG epitope, we could show by Western immunoblot analysis that the ER{alpha} and ERß cDNAs directed similar levels of ER expression (data not shown). Although we demonstrated that the presence of the FLAG tag did not affect the estradiol-mediated transcriptional activity of either receptor (data not shown), we elected to use the native receptors for our studies because we were unsure how the FLAG tag would affect more subtle functions of ER{alpha} and ERß. Using this system we examined the transcriptional responses of ER{alpha} and ERß over a range of concentrations of the different ER ligands (Fig. 1AGo). Both receptors were activated by 17ß-estradiol, although we observed that ER{alpha} is a more efficacious activator in this model system. Interestingly, all of the SERMS and pure antagonists tested displayed no agonist or inverse agonist activities on ERß on either of the promoters studied (Fig. 1AGo and data not shown). As shown previously (16), 4-hydroxytamoxifen displayed partial agonist activity on ER{alpha} in HepG2 cells on the C3-Luc reporter. In this environment, GW7604 did not exhibit agonist activity on ER{alpha}, whereas ICI 182,780, raloxifene, and idoxifene functioned as inverse agonists.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 1. The {alpha}- and ß-forms of the human ER are functionally distinct. A, HepG2 cells were transiently transfected with the ER{alpha} or ERß expression vectors and the C3-Luc reporter. Cells were induced with vehicle (nh) or increasing concentrations (ranging from 1 pM to 1 µM) of 17ß-estradiol (E2), ICI 182,780 (ICI), 4-hydroxytamoxifen (4OH-T), raloxifene (RAL), GW7604, or idoxifene (IDOX). After 48 h, transcription was quantitated by assaying for luciferase activity, and all transfections were normalized for efficiency using an internal ß-galactosidase control plasmid (pCMV-ß-gal). Each data point is the average of triplicate measurements of transcriptional activity, and the average coefficient of variation of each value is less than 10%. B, HepG2 cells and HeLa cells (C and D) were transfected with the ER{alpha} or ERß expression plasmids and the 3x-ERE-TATA-Luc or C3-Luc reporter. Cells were induced with 17ß-estradiol (E2) for 48 h, and luciferase assays were performed. Each value was normalized to the ß-galactosidase activity. Each data point is the average of triplicate determinations, and the average coefficient of variance for each value is less than 10%.

 
The initially reported form of the human ERß is 477 amino acids in length (ERß-short), and its expression has been verified in several tissues (7). However, an isoform that contains an additional 53 amino acids at the N-terminus (ERß-long) has recently been described, and this form is also detectable in vivo (19). We characterized the transcriptional activity of the two ERß isoforms to determine whether the additional sequences had any impact on the observed responsiveness to estrogens and antiestrogens. However, under the conditions of our assays, the activities of ERß-short or ERß-long in response to ER agonists and antagonists were indistinguishable (data not shown). Cumulatively, these results suggest that the responses of ER{alpha} and ERß to pharmacological agents are different, a finding that may be important in understanding the cell-selective actions of these compounds.

We observed that 17ß-estradiol was a stronger activator of ER{alpha} than ERß in HepG2 cells on the C3-Luc reporter (Fig. 1AGo). To investigate whether this observation holds in other cell and promoter contexts, we examined the effect of the agonist 17ß-estradiol on ER{alpha} and ERß transcriptional activities in HepG2 and HeLa cells on the 3x-ERE-TATA-Luc and C3-Luc reporters (Fig. 1Go, B–D). In the absence of hormone, ER{alpha} showed a significantly higher level of basal transcriptional activity than ERß in both HepG2 (Fig. 1BGo) and HeLa cells (Fig. 1Go, C and D). This effect was observed on both the simple and complex promoters. In comparing the overall efficacies of ER{alpha} and ERß, the basal activity of the former must be considered, as we have shown previously that this activity is receptor-dependent and can be completely suppressed using pure antiestrogens (16). It has been reported previously that ER{alpha} and ERß have equivalent affinities for estradiol (8); however, in our assays the EC50 for estradiol was approximately 1.5 orders of magnitude less for ER{alpha} than for ERß in both HepG2 (Fig. 1BGo) and HeLa (data not shown) cells. Furthermore, the efficacies displayed by ER{alpha} were consistently higher than those displayed by ERß under the conditions we used (Fig. 1Go, B and C). Reproducibly, we found that ERß demonstrates about 20–60% of the total activity of ER{alpha}. Similar results were observed in transient transfection assays performed in ROS (rat osteosarcoma) and SKBR3 (mammary carcinoma) cell lines (data not shown). Based on these studies, we conclude that 17ß-estradiol is a more potent and efficacious activator of ER{alpha}, and that it is likely that both receptors contribute in a unique manner to the cellular response to estrogens. Overall, our results define a major mechanistic distinction between the two ERs; ERß is strictly dependent on pure agonists for the activation of transcription from its target promoters, whereas ER{alpha} can be activated by both agonists, partial agonists (SERMS), and ligand-independent mechanisms.

The activation domains within ER{alpha} and ERß are not functionally equivalent
We and others have shown that both activation domains, AF-1 and AF-2, are required for maximal agonist-dependent and ligand-independent activation of transcription by ER{alpha} (11). Additionally, in contexts where ER{alpha}-AF-1 alone can function as an autonomous activator we were able to demonstrate that 4-hydroxytamoxifen manifests partial agonist activity. Using similar assays, reconstituted in several cell and promoter backgrounds, we were unable to detect significant ERß-mediated 4-hydroxytamoxifen agonist activity, suggesting that this receptor isoform may not possess a functional AF-1 or, alternatively, that it may have a different type of activation domain within this region.

To define the mechanism(s) underlying the differential activation profiles of ER{alpha} and ERß, we wanted to determine the relative contributions of the N-terminus (AF-1) and C-terminus (AF-2) activation domains to the transcriptional activity of the whole receptors. Previously, our laboratory has created mutations in ER{alpha} that abolish the activity of AF-1 or AF-2 (11). We have now constructed the corresponding mutations in ERß (Fig. 2AGo), and this has enabled us to assess the relative contributions of each AF to the transcriptional activities of ER{alpha} and ERß. To compare the activities of our mutants, HepG2 cells and HeLa cells were transiently transfected with wild-type ER{alpha} or ERß or the mutant receptor to be tested together with the C3-Luc reporter. Because of the difficulty in obtaining antibodies that can be used to measure the relative expression of ER{alpha} and ERß, we elected to perform all of our studies at input plasmid concentrations that yield the maximal activity in a given assay. Consequently, our studies do not allow us to compare ER{alpha} and ERß on a molecule:molecule basis, but, rather, permit us to compare these receptors at a functional level. We have successfully used this approach in the past to compare the transcriptional activities of a series of ER{alpha} mutants (11). The results of this analysis are shown in Fig. 2Go. As expected, ER{alpha} displays a dose-dependent increase in activity in the presence of 17ß-estradiol in both HepG2 and HeLa cells (Fig. 2BGo). In addition, as shown before (37), mutants containing AF-1 or AF-2 alone are also capable of activating transcription, although their activities are influenced by both the cell and promoter context in which they were assayed. In HepG2 cells, for instance, ER{alpha}-AF-1 is significantly more active than ER{alpha}-AF-2; thus, in this environment AF-1 appears to be the dominant activator. In HeLa cells, however, both ER{alpha}-AF-1 and ER{alpha}-AF-2 display identical activation profiles, and their combined activity is significantly less than that of the intact receptor throughout the entire range of hormone concentrations. Interestingly, in both cell contexts, ER{alpha}-AF-1 and ER{alpha}-AF-2 exhibit significantly lower ligand-independent activity compared with the intact ER{alpha}. These studies confirm our previous findings that both AF-1 and AF-2 contribute to the overall transcriptional activity of ER{alpha}, and that the relative activity of each activation domain is dependent on the cell context.



View larger version (30K):
[in this window]
[in a new window]
  		Figure 2. The activation domains within ER{alpha} and ERß are not functionally equivalent. A, The ER{alpha} activation domain mutants were created, as described previously (37). The ERß-AF1 construct was made by introducing three amino acid changes into the AF2 region of the receptor, substituting alanine for amino acids located at positions 436, 440, and 443. The ERß-AF2 construct was made by deleting the N-terminus (amino acids 1–95) of the wild-type receptor. B, HepG2 cells and HeLa cells were transiently transfected with the ER{alpha} wild-type or mutant receptors together with the C3-Luc reporter. After transfection, cells were treated with vehicle (nh) or increasing concentrations (ranging from 1 pM to 1 µM) of 17ß-estradiol (E2). After 48 h, cells were harvested and assayed for luciferase activity, and all transfections were normalized for efficiency using the internal pCMV-ß-gal control plasmid. Each data point is the average of triplicate determinations, and the average coefficient of variance for each value is less than 10%. C, HepG2 cells and HeLa cells were transiently transfected with the ERß wild-type or mutant receptor together with the C3-Luc reporter and the pCMV-ß-gal control vector. Cells were induced with 17ß-estradiol (E2) for 48 h, and luciferase assays were performed. Each value was normalized to the ß-galactosidase activity. Each data point is the average of triplicate determinations, and the average coefficient of variance for each value is less than 10%.

 
We next examined the activities of ERß and the ERß AF mutants in HepG2 and HeLa cells (Fig. 2CGo). To find suitable dose ranges of input plasmids for our studies, we created a duplicate set of vectors in which the amino-termini of the mutant receptors were modified by the addition of a FLAG tag. Using these expression vectors we were able to demonstrate that ERß and ERß-AF-1 were expressed equivalently, whereas ERß-AF-2 expression was elevated by about 50%. With this information in hand, we proceeded with the evaluation of the transcriptional activity of the untagged, native ERß mutants. Because our vectors produce equivalent amounts of receptor, we performed all of our studies at equivalent input plasmid concentrations, selecting that which gave a maximal response under the conditions of the assay for all receptors. As seen with ER{alpha}, the wild-type ERß is a ligand-dependent transcriptional activator in both cell lines. However, although ER{alpha}-AF-1 can function autonomously in some cell types, ERß-AF-1 is transcriptionally inactive in both HepG2 and HeLa cells. Overall, the most surprising finding was that removal of the amino-terminus (ERß-AF-2) enhances ERß transcriptional activity rather than decreases it as was seen with ER{alpha}. Of additional importance was the finding that ERß-AF-2 is extremely active in environments where ER{alpha}-AF-2 is essentially transcriptionally inactive. Therefore, we conclude that the C-terminus (AF-2) mediates the transcriptional activity of ERß in response to 17ß-estradiol, and that this domain functions independently of the amino-terminus. Furthermore, sequences within the N-terminus have a negative impact on the overall transcriptional activity of ERß, and their removal creates a mutant whose function is similar, but not identical, to that of ER{alpha}. It is possible that ERß does, in fact, contain a functional AF-1 domain and that it may be functional in contexts not represented by our assays. These studies clearly indicate that the activation domains of ER{alpha} and ERß are not identical and that these two receptors are different transcription factors that likely have distinct roles in estrogen biology.

The amino-terminus of ER{alpha}, but not that of ERß, functions as an autonomous activating sequence
To determine whether the amino-terminus of ERß does, in fact, possess an autonomous activation function, we evaluated the transcriptional activity of the ER{alpha} and ERß N-termini when removed from the context of their intact receptors. Specifically, the N-termini of ER{alpha} and ERß were each fused to the GAL4 transcription factor DBD (Fig. 3AGo), and the activity of each construct was compared in transient transfection studies in HepG2, HeLa, MCF-7 (human mammary carcinoma), and SKBR3 (human mammary adenocarcinoma) cells on a GAL4-responsive reporter (Fig. 3Go, B–E). The input concentration of each vector used in these studies was that which gave a maximal response under the conditions of the assay. In HepG2 cells, the GAL4-DBD-ER{alpha} construct possessed more than 18 times the activity of the GAL4 DBD alone, whereas only a 2.4-fold enhancement by the ERß construct was observed. In HeLa cells, ER{alpha} displayed a 5.5-fold increase in transcriptional activity over the control, whereas only a 1.4-fold increase was observed for ERß. Similarly, in MCF-7 and SKBR3 cells, ER{alpha} displayed 42- and 53-fold increases in activities, whereas only 3.3- and 6-fold increases were seen for ERß. These studies illustrate that while the N-terminus of ER{alpha} has a strong activation domain that functions in a cell-specific manner, the homologous region in ERß is much less active. Therefore, it is likely that the repressor function is the primary determinant of the activity of the N-terminus of ERß in the whole receptor, and that the distinct transcriptional profiles of the two ERs are mediated in part by differences in their amino-termini.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 3. The amino-terminus of ER{alpha}, but not that of ERß, functions as a strong autonomous activating sequence. A, The GAL4-DBD-ER constructs were created by inserting the amino-terminus of each receptor (amino acids 1–182 of ER{alpha} and 1–95 of ERß) downstream of the GAL4 transcription factor DBD. B–E, HepG2 cells, HeLa cells, MCF-7 cells, and SKBR3 cells were transiently transfected with the GAL4-DBD or GAL4-ER constructs together with a 5x-GAL4-TATA-Luc reporter (containing five copies of the 17-bp palindromic GAL4 transcription factor response element). Cells were harvested after 48 h and assayed for luciferase activity. All transfections were normalized for efficiency using the internal pCMV-ß-gal control plasmid. The data are presented as fold activation, where 1 represents a measure of the activity of the GAL4-DBD construct alone. Each data point is the average of triplicate determinations.

 
ERß represses ER{alpha} transcriptional activity at subsaturating concentrations of 17ß-estradiol
Our data indicate that ER{alpha} and ERß are not functionally equivalent and that in many contexts ER{alpha} is significantly more transcriptionally active than ERß. Therefore, we considered the possibility that, similar to the effects mediated by PR-A on PR-B activity, ERß may act as a transdominant repressor or a modulator of ER{alpha} activity in contexts where ERß is less active. To test this hypothesis, we examined the effect of ERß expression on the transcriptional activity of ER{alpha} when both receptors were expressed in target cells. Specifically, the ER{alpha} expression vector was transiently transfected into HepG2 cells, either alone or in the presence of increasing amounts of the ERß expression vector together with the 3x-ERE-TATA-Luc reporter. The cells were induced with either 100 nM 17ß-estradiol, a saturating concentration at which both receptors demonstrate maximal transcriptional activity, or 100 pM 17ß-estradiol, a subsaturating concentration at which only ER{alpha} is active. As expected, ER{alpha} was a better activator of transcription than ERß when the assay was performed in the presence of 100 nM 17ß-estradiol. Interestingly, although ER{alpha} manifests significant ligand-dependent activity, the introduction of increasing amounts of ERß into the system had no effect on ER{alpha} transcriptional activity under the conditions of the assay. However, in the presence of 100 pM 17ß-estradiol, the activity of ER{alpha} was suppressed by the addition of increasing amounts of the ERß expression vector. At equivalent input vector concentrations, the estradiol-stimulated activity of the coexpressed receptors was repressed to 25% of the activity of ER{alpha} alone. These data strongly support the hypothesis that the pharmacology of estrogen(s) will differ in target cells depending on the relative expression levels of ER{alpha} and ERß.

We next compared the activities of ER{alpha}, ERß, or both receptors together over a full range of estradiol concentrations (Fig. 4BGo). Based on the observation that ERß functions as a repressor of ER{alpha} transcriptional activity at low concentrations of hormone, we predicted that the impact of ERß on ER{alpha} would differ at specific hormone concentrations and that the cellular responsiveness to estradiol would be affected by ERß expression. Interestingly, we observed that the potency of estradiol in our ER{alpha}-dependent transcription systems was right shifted by 1 log when ERß was coexpressed in the system, whereas the efficacy was unaffected. Similar results were also observed when this experiment was repeated in different cellular contexts (data not shown). From these studies, we conclude that 1) ERß is a transdominant repressor of ER{alpha} transcriptional activity at subsaturating concentrations of estradiol; and 2) ERß expression decreases the sensitivity of ER{alpha}-expressing cells to estradiol.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 4. ERß is a transdominant repressor of ER{alpha} transcriptional activity at subsaturating concentrations of 17ß-estradiol. A, HepG2 cells were transiently transfected with the 3x-ERE-TATA-Luc reporter alone (no ER), the reporter and 250 ng ERß, or the reporter and 250 ng of the ER{alpha} expression vector together with increasing concentrations of the ERß expression vector (0, 10, 50, and 250 ng). After transfection, cells were treated with vehicle (nh) or 100 nM or 100 pM 17ß-estradiol (E2). After 48 h, cells were harvested and assayed for luciferase activity, and all transfections were normalized for efficiency using the internal pCMV-ß-gal control plasmid. Each data point is the average of triplicate determinations, and the average coefficient of variance for each value is less than 10%. B, HepG2 cells were transiently transfected with either ER{alpha} or ERß expression vectors or equal quantities of both vectors together with the C3-Luc reporter and the pCMV-ß-gal control plasmid. Cells were induced with vehicle (nh) or increasing concentrations (ranging from 1 pM to 1 µM) of 17ß-estradiol (E2) for 48 h, and luciferase assays were performed. Each value was normalized to the ß-galactosidase activity. Each data point is the average of triplicate determinations, and the average coefficient of variance for each value is less than 10%.

 
Coexpression of ERß suppresses the partial agonist activity of tamoxifen through ER{alpha}
Our observations that the cellular response to estrogen is dependent on the relative expression of ER{alpha} and ERß suggested that the agonist/antagonist activities of mixed agonists such as tamoxifen may be influenced by the relative expression of the two receptors. To determine the impact of coexpression of the two receptor subtypes on tamoxifen pharmacology, the ER{alpha} and ERß expression vectors were transiently transfected into HepG2 cells, either alone or together, and the cellular response to increasing concentrations of 4-hydroxytamoxifen was measured on the C3-Luc reporter (Fig. 5Go). As seen before, tamoxifen functioned as an agonist when ER{alpha} alone was expressed in the cell and showed no agonist activity when ERß alone was expressed. In the presence of ERß, however, the ability of tamoxifen to activate ER{alpha}-mediated transcription was completely suppressed. We noticed that the basal transcription in the presence of both receptors was intermediate between ER{alpha} and ERß alone in this assay. Although we have not pursued this interesting finding further, we do know that all of the basal activity observed in the presence of ER{alpha} is ER-dependent. Thus, ERß may effectively inhibit tamoxifen-dependent ER{alpha}-mediated transcription, but not that activity mediated through the same receptor in the absence of ligand. Overall, these results suggest that the tissue-selective agonist activity of antiestrogens may be determined by the relative expression of ER{alpha} and ERß in a tissue.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 5. Coexpression of ERß suppresses the partial agonist activity of tamoxifen through ER{alpha}. HepG2 cells were transiently transfected with either the ER{alpha} or ERß expression vector or equal quantities of both vectors together with the C3-Luc reporter and the pCMV-ß-gal control plasmid. Cells were induced with vehicle (nh) or increasing concentrations (ranging from 1 pM to 1 µM) of 4-hydroxytamoxifen (4OH-T) for 48 h, and luciferase assays were performed. Each value was normalized to the ß-galactosidase activity. Each data point is the average of triplicate determinations, and the average coefficient of variance for each value is less than 10%.

 
ER{alpha} and ERß form heterodimers in vivo
We next investigated the molecular mechanism by which ERß modulates ER{alpha} transcriptional activity. It has recently been reported that coexpression of ER{alpha} and ERß results in the preferential formation of receptor heterodimers (instead of homodimers) that are capable of activating transcription from estrogen-responsive enhancers (20, 21). Our observation that classical ER{alpha}- and ERß-mediated responses are modified when the two receptors are coexpressed suggests that these alternative responses could be mediated through receptor heterodimers that may possess unique transcriptional activities. To address this issue we examined the interaction between the two receptors in a mammalian two-hybrid assay. HepG2 cells were transfected with the pBKC-DBD-ER{alpha}-LBD(3x) vector and pVP16-ERß vectors together with a GAL4-responsive reporter and treated with vehicle or 100 nM 17ß-estradiol (Fig. 6Go). The ER{alpha}-LBD(3x) vector contains three point mutations in the AF-2 domain that abolish transcriptional activity but have no effect on ER dimerization (11). We observed that ER{alpha} and ERß show a strong hormone-dependent interaction. Furthermore, our results show that ER{alpha} and ERß heterodimerize in vivo and support the recent findings that heterodimers are formed when the two subtypes are coexpressed. Although previous studies have shown that ER{alpha} and ERß heterodimerize in the absence of hormone (20, 21), these studies were carried out in vitro and may not accurately represent what occurs in the cell where ER{alpha} and ERß do not exist in isolation but, rather, interact with a variety of proteins. Overall, the results imply that the impact of each receptor on the other’s pharmacology may be mediated at least in part through the formation of heterodimers with properties that may be different from those of the homodimers.



View larger version (11K):
[in this window]
[in a new window]
  		Figure 6. ER{alpha} and ERß form heterodimers in vivo. HepG2 cells were transiently transfected with either GAL4-DBD or GAL4-ER{alpha}-LBD(3X) (in this construct AF-2 activity is removed to decrease the basal activity, but the dimerization domain is intact) together with either pVP16 or pVP16-ERß and the 5x-GAL4-TATA-Luc reporter. Cells were treated with vehicle (nh) or 100 nM 17ß-estradiol (E2) for 48 h, and luciferase assays were performed. Each luciferase value was normalized to the ß-galactosidase activity. Each data point is the average of triplicate determinations, and the average coefficient of variance for each value is less than 10%.

 
ERß interacts with target gene promoters in the absence of ligand
We have observed that ERß suppresses ER{alpha}-mediated transcription at subsaturating levels of hormone. One possible explanation for this activity is that ERß may bind its cognate response element in a constitutive manner and thus compete with ER{alpha} for access to the DNA target. To test this hypothesis, ER{alpha} and ERß were fused to the VP16 activation domain in order to bypass the need for ligand to activate the AFs within the two receptors. Thus, when tethered to DNA, the chimeras will activate transcription regardless of the nature of the bound ligand. HepG2 cells were transiently transfected with increasing concentrations of the pVP16-ER{alpha} or pVP16-ERß expression plasmids together with the 3x-ERE-TATA-Luc reporter in the presence of vehicle or 100 nM 17ß-estradiol. As shown in Fig. 7AGo, when overexpressed, VP16-ER{alpha} did have the capacity to interact with DNA in the absence of hormone. However, when VP16-ER{alpha} was limiting, hormone had a pronounced effect on DNA binding. To eliminate the possibility that the elevated transcriptional responses of VP16-ER{alpha} in the presence of agonist were due to synergy between the ER{alpha} activation functions and the VP16 activation domain in the presence of estradiol, we repeated the experiments using VP16-ER{alpha}(3x), which contains three point mutations in the ER{alpha} AF-2 domain that disrupt the ligand-dependent coactivator binding pocket and nullify the transcriptional activity of the receptor. Under these conditions we observed that hormone was required for DNA binding, even at high receptor concentrations (Fig. 7BGo). In contrast to the results obtained with VP16-ER{alpha}, we observed that the ability of VP16-ERß to interact with DNA was entirely ligand independent, regardless of the level of expression (Fig. 7CGo). These studies reveal another mechanistic difference between ER{alpha} and ERß and suggest that ERß regulates ER{alpha}-mediated transcriptional activity at the level of DNA-binding site competition.



View larger version (21K):
[in this window]
[in a new window]
  		Figure 7. ERß interacts with target gene promoters in the absence of ligand. HepG2 cells were transiently transfected with increasing concentrations of the pVP16-ER{alpha} expression vector (A), the pVP16-ER{alpha}(3x) expression vector (B), or the pVP16-ERß expression vector (C). Each expression construct includes the VP16 activation domain sequence fused 5' to the entire coding sequence for the human ER{alpha}, ER{alpha}(3x) mutant, or ERß. After transfection, cells were treated with vehicle (nh) or 100 nM 17ß-estradiol (E2) for 48 h, and luciferase assays were performed. Each value was normalized to the ß-galactosidase activity. Each data point is the average of triplicate determinations, and the average coefficient of variance for each value is less than 10%.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Differential activities of nuclear hormone receptor subtypes
The ability of two nuclear receptor subtypes to display distinct or even opposing transcriptional activities is an emerging paradigm in nuclear receptor signaling. One example is the human glucocorticoid receptor (GR), which exists in two forms ({alpha} and ß) that arise by an alternate splicing event that results in a truncation of the C-terminus in the GRß form (22). Analysis of the transcriptional properties of these two receptors indicated that they both recognize the same intracellular targets, but GRß is unable to bind agonists (22). Therefore, although GR{alpha} can enhance the transcription of glucocorticoid-responsive enhancers, GRß is not transcriptionally active and can act as a dominant negative inhibitor of GR{alpha} activity. The contrasting activities of the GRs suggest that the relative tissue distributions of the two receptors are a major determinant of the biological effects of glucocorticoid compounds. The existence of two forms of the human PR, each with different biological activity, extends the concept that receptor isoforms are an important part of steroid hormone receptor pharmacology. The two PRs are identical in sequence, with the exception that hPR-B has an additional 164 amino acids within its N-terminus originating from an alternate translation initiation event. These isoforms have been shown to bind the same ligands and regulate transcription of the same genes. However, in most cell types PR-B is a transcriptional activator, whereas PR-A acts as a transdominant repressor of PR-B. Interestingly, in the presence of an activating ligand, human PR-A can inhibit the transcriptional activities of ER, GR, androgen receptor, and mineralocorticoid receptor (15).

Our studies suggest that the human ER{alpha} and ERß provide yet another example of two nuclear receptor subtypes that demonstrate distinct transcriptional activities. As is seen with human PRs, the differential activities of the two ERs arise from functional variations in the receptor N-termini. ER{alpha} contains a constitutive AF-1 in the N-terminus that functions in a cell- and promoter-specific manner to enhance the overall transcriptional response of the receptor. However, the corresponding region of ERß lacks significant transcriptional activity and contains a repressor domain that decreases the overall transcriptional activity of the receptor. This inhibitory region functions only in the context of the intact receptor, as has been shown for the inhibitory N-terminus of PR-A (15) and for the repressor domains of the transcription factor c-Fos (23). Our studies do not eliminate the possibility that ERß does, in fact, have an AF-1, but suggest that its function is masked by the presence of an amino-terminal repressor domain. This hypothesis is supported by the fact that the N-terminus of ERß showed low levels of autonomous transcriptional activity when fused to the heterologous GAL4-DBD. Furthermore, recent studies have shown that ERß transcriptional activity can be stimulated by the mitogen-activated protein kinase pathway (23A ), and that this activity appears to require an intact ERß amino-terminus. Mutational analysis will be required to determine whether ERß contains separate activation and repression domains within its amino-terminus.

An additionally important distinction between ER{alpha} and ERß resulted from our finding that ER{alpha}-AF-1 and AF-2 act synergistically under most circumstances, whereas the AF-2 of ERß functions as an independent activation domain. It is likely, therefore, that ER{alpha} and ERß will display differences in their preferences for coactivators and corepressors in target cells. In recent years, several transcriptional coactivator proteins have been identified that interact with the hormone-binding domain of nuclear receptors and are thought to mediate the activity of AF-2. Specifically, the coactivators steroid receptor coactivator-1 (SRC-D), amplified in breast cancer (AIB1), transcriptional intermediary factor-2 (TIF-2), and glucocorticoid receptor interacting protein (GRIP-1) (murine homolog of TIF-2) and the cointegrator CBP/p300 have been shown to potentiate the activity of ER{alpha} and other nuclear receptors (24, 25, 26, 27). The C-termini of both ER{alpha} and ERß have been shown to bind the coactivators SRC-1 and GRIP-1 in an agonist-dependent manner (28, 29) (our unpublished results). Therefore, it is possible that although the cofactors that mediate transcriptional activity of the two ERs are the same, these proteins may interact with distinct regions of each receptor or differentially recruit other cellular factors when bound to one receptor vs. the other. Although sequence comparisons suggest that the N-terminus may be the primary discriminator by which coactivators can interact with the receptors, the fact that AF-2 of ER{alpha} and ERß are also functionally different suggests that the process is much more complex. It is not clear at this time whether AF-1 and/or AF-2 interact with distinct coactivators or if these domains have independent contact sites on the same coactivator. Regardless, it is clear that in the context of ER{alpha}, AF-1 is required for maximal agonist-induced transcriptional activity, as mutations in this domain have been shown to abolish tamoxifen partial agonist activity and dampen the response to estrogen (30). The absence of an efficient AF-1 in ERß clearly influences the manner in which coactivators interface with this receptor and ultimately its ligand responsiveness. Our data are also compatible with the concept that the N-terminus of ERß binds a protein that has an autonomous inhibitory activity or one that inhibits transcriptional activity by blocking the binding of coactivators to AF-1 and AF-2. If this is the case, then it is possible that in tissues in which the putative repressor protein is absent, antiestrogens could manifest partial agonist activity, and agonists would be more potent receptor activators.

Roles of ER{alpha} and ERß in determining cellular sensitivity to estrogen
One of the most important findings of this study is that the relative levels of ER{alpha} and ERß are an important determinant of cellular sensitivity to estrogens. Although ER{alpha} is the stronger transcriptional activator of the two ER isoforms, at physiological concentrations of estradiol, coexpression of ERß results in suppression of both the efficacy and the potency of hormone-stimulated responses. This suggests that it will be important to determine the extent to which the two receptors colocalize in order to more accurately predict the biological responses to ER agonists in specific target tissues. The ability of ERß to function as a transcriptional inhibitor or activator, depending on the agonist concentration, suggests that completely different patterns of gene expression may be observed at different hormone levels. In addition, the ability of ERß to switch from a transcriptional repressor to an activator as estradiol levels rise may provide cells expressing both isoforms with a mechanism to control cellular sensitivity to hormones. Such a process could explain why during the early part of the menstrual cycle, low plasma concentrations of estradiol exhibit an inhibitory effect on gonadotropin secretion, whereas when levels of hormone are elevated during the late follicular phase, the pituitary release of LH and GnRH secretion from the hypothalamus is enhanced. In light of the recent localization of ERß to rat hypothalamic neurons projecting to the pituitary (31, 32), it is possible that the balance between ER{alpha} and ERß activities in these tissues may mediate the differential sensitivities to estrogens throughout the menstrual cycle. The role of ERß in the regulation of cellular responsiveness to agonists may merit consideration in dosing regimens of estrogen-like pharmaceutical compounds, as it is likely that fluctuations in the bioavailability of receptor activating ligands may have a greater impact in tissues where ER{alpha} and ERß colocalize. This may be particularly important in ER-positive breast tumors, where it has been shown that ERß, in addition to ER{alpha}, may be expressed (33).

Our studies also suggest that the relative levels of ER{alpha} and ERß are an important determinant of the pharmacology of antiestrogens. The observation that tamoxifen is a more potent competitive antagonist of ERß (8) and does not display agonist activity on the receptor raises the possibility that there will be a better response to tamoxifen in ERß-positive tumors. In view of our finding that ERß suppresses the partial agonist activity of tamoxifen on ER{alpha}, it will be interesting to determine whether tumors expressing both subtypes show a better response to tamoxifen as well. It will also be important to determine whether ERß is down-regulated in tamoxifen-resistant tumors as an adaptive mechanism for growth.

A working model to explain the cross-talk between ER{alpha} and ERß
We have developed a working model to explain how ERß can regulate ER{alpha} transcriptional activity in cells where the receptors are coexpressed. This model is based on two fundamental observations: 1) ERß binds to target gene promoters in a ligand-independent manner; and 2) ERß can form heterodimers with ER{alpha} within cells. Thus, in the presence of low subsaturating concentrations of hormone, inactive ERß binds to its target response element and competitively blocks ER{alpha} binding. As hormone levels rise, the amount of activated ER{alpha} and ERß also rises, sufficient activated receptor is formed to compete with the unliganded, inactive ERß, and transcription can proceed. A purely competitive interaction would predict that as ERß levels rise, agonist efficacy would decrease to a level approaching that observed when ERß alone is expressed in cells. However, the observation that under conditions of hormone excess, overexpression of ERß does not decrease the efficacy of estradiol suggests that the interaction of ER{alpha} and ERß is more complex. We believe that under hormone-saturating conditions, ER{alpha} and ERß can form heterodimers and that the transcriptional activity of the heterodimer is indistinguishable from that of the ER{alpha} homodimer. Although difficult to address experimentally, it is possible that of the three potential ER complexes, the ERß homodimer has the highest affinity for corepressors and/or the lowest affinity for coactivators, and consequently, it is the least transcriptionally active. However, in the context of a heterodimer, the presence of ER{alpha} assists ERß in recruiting cofactors such as SRC-1 and GRIP-1. Thus, the resultant complex of ER{alpha}/ERß and their associated coactivators is indistinguishable from that formed by an ER{alpha} homodimer. If this latter model is found to be true, then we would predict that the major role of ERß is to modulate ER{alpha} transcriptional activity at low hormone levels.


   Acknowledgments
 
We thank Drs. Mark Nuttall and X. F. Wang for providing plasmids. We thank T. Willson for providing reagents and for sharing unpublished data.


   Footnotes
 
1 This work was supported by NIH Grant DK-48807 (to D.P.M.). Back

2 Supported by a Predoctoral Fellowship from the U.S. Army Medical Research Acquisition Activity. Back

Received April 5, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Evans RM 1988 The steroid and thyroid receptor superfamily. Science 240:889–895[Abstract/Free Full Text]
  2. Tsai M, O’Malley BW 1994 Molecular mechanisms of action of steroid/thyroid receptor superfamily members. Annu Rev Biochem 63:451–486[CrossRef][Medline]
  3. Aronica SM, Katzenellenbogen BS 1993 Stimulation of estrogen receptor-mediated transcription and alteration in the phosphorylation state of the rat uterine estrogen receptor by estrogen, cyclic adenosine monophosphate, and insulin-like growth factor-I. Mol Endocrinol 7:743–752[Abstract/Free Full Text]
  4. Ignar-Trowbridge DM, Teng CT, Ross KA, Parker MG, Korach KS, McLachlan JA 1993 Peptide growth factors elicit estrogen receptor-dependent transcriptional activation at an estrogen-responsive element. Mol Endocrinol 7:992–998[Abstract]
  5. Curtis SW, Washburn T, Sewall C, DiAugustine R, Lindzey J, Couse JF, Korach KS 1996 Physiological coupling of growth factor and steroid receptor signaling pathways: estrogen receptor knockout mice lack estrogen-like responses to epidermal growth factor. Proc Natl Acad Sci USA 93:12626–12630[Abstract/Free Full Text]
  6. Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson J-A 1996 Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 93:5925–5930[Abstract/Free Full Text]
  7. Mosselman S, Polman J, Dijkema R 1996 ERß: identification and characterization of a novel human estrogen receptor. FEBS Lett 392:49–53[CrossRef][Medline]
  8. Kuiper GGJM, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson J-A 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors {alpha} and ß. Endocrinology 138:863–870[Abstract/Free Full Text]
  9. Couse JF, Lindsey J, Grandien K, Gustafsson J-A, Korach KS 1997 Tissue distribution and quantitative analysis of estrogen receptor-{alpha} (ER{alpha}) and estrogen receptor-ß (ERß) messenger ribonucleic acid in the wild-type and ER-{alpha} knockout mouse. Endocrinology 138:4613–4621[Abstract/Free Full Text]
  10. Berry M, Metzger D, Chambon P 1990 Role of the activating domains of the oestrogen receptor in the cell-type and promoter-context dependent agonistic activity of the anti-oestrogen 4-hydroxytamoxifen. EMBO J 9:2811–2818[Medline]
  11. Tzuckerman MT, Esty A, Santiso-Mere D, Danielian P, Parker MG, Stein RB, Pike JW, McDonnell DP 1994 Human estrogen receptor transactivational capacity is determined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions. Mol Endocrinol 8:21–30[Abstract]
  12. Willson TM, Norris JD, Wagner BL, Asplin I, Baer P, Brown HR, Jones SA, Henke B, Sauls H, Wolfe S, Morris DC, McDonnell, DP 1997 Dissection of the molecular mechanism of action of GW5638, a novel estrogen receptor ligand, provides insights into the role of estrogen receptor in bone. Endocrinology 138:3901–3911[Abstract/Free Full Text]
  13. McDonnell DP, Dana SL, Hoener PA, Lieberman BA, Imhof MO, Stein RB 1995 Cellular mechanisms which distinguish between hormone- and antihormone-activated estrogen receptor. Ann NY Acad Sci 761:121–137[Abstract]
  14. Vegeto E, Shahbaz MM, Wen, DX, Goldman ME, O’Malley BW, McDonnell DM 1993 Human progesterone receptor A form is a cell- and promoter-specific repressor of human progesterone receptor B function. Mol Endocrinol 7:1244–1255[Abstract]
  15. Giangrande PH, Pollio G, McDonnell DP 1997 Mapping and characterization of the functional domains responsible for the differential activity of the A and B isoforms of the human progesterone receptor. J Biol Chem 52:32889–32900
  16. Norris JD, Fan D, Wagner BL, McDonnell DP 1996 Identification of the sequences within the human C3 promoter required for estrogen responsiveness provides insight into the mechanism of tamoxifen mixed agonist activity. Mol Endocrinol 10:1605–1616[Abstract]
  17. Norris JD, Fan D, Aleman C, Marks JR, Futreal PA, Wiseman RW, Iglehart JD, Deininger PL, McDonnell DP 1995 Identification of a new subclass of Alu DNA repeats which can function as estrogen receptor-dependent transcriptional enhancers. J Biol Chem 270:22777–22782[Abstract/Free Full Text]
  18. Schreiber E, Matthias P, Muller MM, Schaffner, W 1998 Identification of a novel lymphoid specific octamer binding protein (OTF-2B) by proteolytic clipping bandshift assay (PCBA). EMBO J 7:4221–4229[Medline]
  19. Moore JT, McKee DD, Moore LB, Jones SA, Su J-L, Slentz-Kesler K, Horne EL, Kliewer SA, Lehmann JM, Willson TM 1998 Cloning, ligand binding and functional activity of human estrogen receptor ß isoforms. Biochem Biophys Res Commun 247:75–78[CrossRef][Medline]
  20. Cowley SM, Hoarse S, Mosselman S, Parker MG 1997 Estrogen receptors {alpha} and ß form heterodimers on DNA. J Biol Chem 272:19858–19862[Abstract/Free Full Text]
  21. Pettersson K, Grandien K, Kuiper GGJM, Gustafsson J-A 1997 Mouse estrogen receptor ß forms estrogen response element-binding heterodimers with estrogen receptor {alpha}. Mol Endocrinol 11:1486–1496[Abstract/Free Full Text]
  22. Oakley RH, Sar M, Cidlowski JA 1996 The human glucocorticoid receptor ß isoform. Expression, biochemical properties, and putative function. J Biol Chem 271:9550–9559[Abstract/Free Full Text]
  23. Brown HJ, Sutherland JA, Cook A, Bannister AJ, Kouzarides T 1995 An inhibitor domain in c-Fos regulates activation domains containing the HOB1 motif. EMBO J 14:124–131[Medline]
  24. Tremblay A, Tremblay GB, Labrie F, Giguere V 1999 Ligand-independent recruitment of SRC-1 to estrogen receptor beta through phosphorylation of activation function AF-1. Mol Cell 3:513–519[CrossRef][Medline]
  25. Horwitz KB, Jackson TA, Bain DL, Richer JK, Takimoto GS, Tung L 1996 Nuclear receptor coactivators and corepressors. Mol Endocrinol 10:1167–1177[Abstract]
  26. Onate SA, Tsai S-Y, Tsai M-J, O’Malley BW 1995 Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science 270:1354–1357[Abstract/Free Full Text]
  27. Eng FCS, Barsalou A, Akutsu N, Mercier I, Zechel C, Mader S, White JH 1998 Different classes of coactivators recognize distinct but overlapping binding sites on the estrogen receptor ligand binding domain. J Biol Chem 273:28371–24377[Abstract/Free Full Text]
  28. Shibata H, Spencer TE, Onate SA, Jenster G, Tsai S-Y, Tsai M-J, O’Malley BW 1997 Role of co-activators and co-repressors in the mechanism of steroid/thyroid receptor action. Recent Prog Horm Res 52:141–164
  29. Hong H, Kohli K, Trivedi A, Johnson DL, Stallcup MR 1996 GRIP1, a novel mouse protein that serves as a transcriptional coactivator in yeast for the hormone binding domains of steroid receptors. Proc Natl Acad Sci USA 93:4948–4952[Abstract/Free Full Text]
  30. Tremblay GB, Tremblay A, Copeland NG, Gilbert DJ, Jenkins NA, Labrie F, Giguere V 1997 Cloning, chromosomal localization, and functional analysis of the murine estrogen receptor ß. Mol Endocrinol 11:353–365[Abstract/Free Full Text]
  31. McInerney EM, Katzenellenbogen BS 1996 Different regions in activation function-1 of the human estrogen receptor required for antiestrogen- and estradiol-dependent transcription activation. J Biol Chem 271:24172–24178[Abstract/Free Full Text]
  32. Alves SE, Lopez V, McEwen BS, Weiland NG 1998 Differential colocalization of estrogen receptor ß (ERß) with oxytocin and vasopressin in the paraventricular and supraoptic nuclei of the female rat brain: an immunocytochemical study. Proc Natl Acad Sci USA 95:3281–3286[Abstract/Free Full Text]
  33. Shughrue PJ, Komm B, Merchenthaler I 1996 The distribution of estrogen receptor-ß mRNA in the rat hypothalamus. Steroids 61:678–681[CrossRef][Medline]
  34. Dotzlaw H, Leygue E, Watson PH, Murphy LC 1996 Expression of estrogen receptor-ß in human breast tumors. J Clin Endocrinol Metab 82:2371–2374[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 EndocrinologyHome page
K. L. Gonzales, M. J. Tetel, and C. K. Wagner
Estrogen Receptor (ER) {beta} Modulates ER{alpha} Responses to Estrogens in the Developing Rat Ventromedial Nucleus of the Hypothalamus
Endocrinology, September 1, 2008; 149(9): 4615 - 4621.
[Abstract] [Full Text] [PDF]

Home page
 Hum ReprodHome page
X. Di, L. Yu, A.B. Moore, L. Castro, X. Zheng, T. Hermon, and D. Dixon
A low concentration of genistein induces estrogen receptor-alpha and insulin-like growth factor-I receptor interactions and proliferation in uterine leiomyoma cells
Hum. Reprod., August 1,&nb