help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2008-0304
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 Bhavnani, B. R.
Right arrow Articles by Lu, X.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bhavnani, B. R.
Right arrow Articles by Lu, X.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene
*Compound via MeSH
*Substance via MeSH
Hazardous Substances DB
*ESTRADIOL
*ESTROGENIC SUBSTANCES, CONJUGATED
*TRITIUM
Endocrinology Vol. 149, No. 10 4857-4870
Copyright © 2008 by The Endocrine Society

Structure Activity Relationships and Differential Interactions and Functional Activity of Various Equine Estrogens Mediated via Estrogen Receptors (ERs) ER{alpha} and ERβ

Bhagu R. Bhavnani, Shui-Pang Tam and XiaoFeng Lu

Department of Obstetrics and Gynecology (B.R.B.), Institute of Medical Sciences University of Toronto (B.R.B.), and The Keenan Research Center of Li Ka Shing Knowledge Institute (B.R.B., X.L.), St. Michael’s Hospital, Toronto, Ontario, Canada M5B 1W8; and Department of Pathology and Molecular Medicine (S.P.T.), Queen’s University, Kingston, Ontario, Canada K7L 3N6

Address all correspondence and requests for reprints to: Professor B. R. Bhavnani, Department of Obstetrics and Gynecology, St. Michael’s Hospital, Room 7-074-Bond Wing, 30 Bond Street, Toronto, Ontario, Canada M5B 1W8. E-mail: bhavnani{at}smh.toronto.on.ca.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The human estrogen receptors (ERs) {alpha} and β interact with 17β-estradiol (17β-E2), estrone, 17{alpha}-estradiol, and the ring B unsaturated estrogens, equilin, 17β-dihydroequilin, 17{alpha}-dihydroequilin, equilenin, 17β-dihydroequilenin, 17{alpha}-dihydroequilenin, {Delta}8-estrone, and {Delta}8, 17β-E2 with varying affinities. In comparison to 17β-E2, the relative binding affinities of most ring B unsaturated estrogens were 2- to 8-fold lower for ER{alpha} and ERβ, however, some of these unique estrogens had two to four times greater affinity for ERβ than ER{alpha}. The transcriptional activity of these estrogens in HepG2 cells transfected with ER{alpha} or ERβ, or both, and the secreted-alkaline phosphatase gene showed that all estrogens were functionally active. 17β-E2 induced the activity of secreted-alkaline phosphatase by ER{alpha} to a level higher than any other estrogen. Activity of other estrogens was 12–17% that of 17β-E2. In contrast, 17β-E2 stimulated the activity of ERβ to a 5-fold lower level than that with ER{alpha}, whereas the activity of other estrogens was 66–290% that of 17β-E2, with equilenin being the most active. The presence of both ER subtypes did not alter the functional activity of 17β-E2, although it further enhanced the activity of 17β-dihydroequilin (200%), 17β-dihydroequilenin (160%), and {Delta}8, 17β-E2 (130%). Except for 17β-E2, no correlation was observed between the functional activities and their binding affinities for ER. In conclusion, our results show that the effects of ring B unsaturated estrogens are mainly mediated via ERβ and that the presence of both ER subtypes further enhances their activity. It is now possible to develop hormone replacement therapy using selective ring B unsaturated estrogens for target tissues where ERβ is the predominant ER.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE BIOLOGICAL EFFECTS of estrogens are considered to be mediated to a major extent by two genetically distinct estrogen receptor (ER) subtypes: ER{alpha} and ERβ (1, 2). The human ER belongs to the ligand-activated nuclear receptor superfamily. Some members of this family are receptors for steroid hormones, vitamin D, retinoic acid, and thyroid hormones (3, 4). A substantial body of research shows that these two receptor subtypes are differentially expressed in various tissues, and each contributes to the overall pharmacology of estrogen (5). These ERs exist in transcriptionally inactive conformations, and binding with ligands induces an active and stable conformation that allows its dimerization (homo or hetero) (1). Different ligands appear to interact with the receptor, and each ligand receptor complex can result in a unique conformation that can further interact with or recruit cell and tissue-specific receptor-associated coactivators or corepressor proteins. Over 30 different receptor-associated protein cofactors have been identified (1). Inherent in this mechanism is the earlier concept (6) that binding affinity of a ligand for its cognate receptor was the sole determinant of biological activity of the ligand is no longer tenable (1, 7, 8, 9). Moreover, a large body of evidence now clearly indicates that the steroid hormone receptor pharmacology involves at least three related mechanisms for hormone selectivity (7). These include: 1) selectivity based on the ligand that includes the pharmacokinetics and differential metabolism in various target tissues; 2) receptor-based selectivity that considers the presence of various isoforms, subtypes, and their concentration in different tissues; and 3) the more complex affector site-based selectivity, and these have been extensively reviewed (1, 7). Conjugated equine estrogen (CEE) preparations are widely used for estrogen replacement therapy and contain sulfate esters of the classical estrogens: estrone (E1), 17β-estradiol (17β-E2), and 17{alpha}-estradiol (17{alpha}-E2); and the ring B unsaturated estrogens equilin (Eq), equilenin (Eqn), 17{alpha}-dihydroequilin (17{alpha}-Eq), 17β-dihydroequilin (17β-Eq), 17{alpha}-dihydroequilenin (17{alpha}-Eqn), 17β-dihydroequilenin (17β-Eqn), {Delta}8-E1, and {Delta}8,17β-E2 (10, 11, 12) (Fig. 1Go). The interaction of these estrogens with crude ER preparations that mostly likely contained mixtures of ER subtypes has been previously reported (13). The interaction of most of these equine estrogens with pure human ER{alpha} and ERβ has not been studied.


Figure 1
View larger version (19K):
[in this window]
[in a new window]

  		FIG. 1. Structures of equine estrogens used in our study. These estrogens, in their sulfate-conjugated forms, are components present in the drug CEE.

 
Because each of the aforementioned 11 estrogens is structurally different (Fig. 1Go), and have a wide range of binding affinities (for ER) and biological effects (13), one would expect that each of these estrogens will impart on the ERs a unique conformation (1) and so potentially have a tissue specific activity. Thus, in the current study, we compare the ligand binding interactions of the 11 equine estrogens with pure human ER{alpha} and ERβ, and their effect on the functional activity (transcriptional responses) of the human ER{alpha} and ERβ, transfected in HepG2 cells.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
Human recombinant ER{alpha} and ERβ were obtained from PanVera (Madison, WI). The human ER{alpha} and ERβ expression vectors were kindly provided by T. S. Scanlan (Department of Pharmaceutical Chemistry, University of California, San Francisco, CA). [2,4,5,7,16,17 3H(N)] 17β-E2 ([3H]17β-E2 specific activity 118 Ci/mmol) (4366.0 GBq/mmol) was from PerkinElmer Life Science (Boston, MA), and its purity was checked by HPLC on a Beckman ODS-C18 column (Beckman, Toronto, Canada) using the system: acetonitrile/water/acetic acid (35:65:1) and its radiochemical purity established as previously described (13). The unlabeled equine estrogens were authentic samples generously donated by Dr. Mike Dey (Wyeth Pharmaceuticals, Philadelphia, PA) and are the reference standards provided to U.S. Pharmacopeia. Their purity has been verified by infrared, mass, and nuclear magnetic resonance spectroscopy. Their absolute stereochemistry is identical to the ones present in the drug CEE (Premarin; Wyeth Pharmaceuticals). A number of these have been used in several clinical studies from our laboratory (10, 11, 12). The secreted-alkaline phosphatase (SEAP) reporter system was purchased from BD Biosciences Clontech (Mississauga, Ontario, Canada). FuGene 6 transfection reagent was purchased from Roche Molecular Biochemicals (Laval, Quebec, Canada), and the HepG2 cell line was obtained from American Type Culture Collection (Rockville, MD). All other biochemicals and reagents were obtained from various commercial sources. All equine estrogens were purified by recrystallizations, and their identity and purity further confirmed by infrared spectroscopy, melting points, and HPLC.

Saturation ligand binding analysis
Duplicate aliquots (1 nM) of ER{alpha} and ERβ were incubated with 0.1–3 nM [3H]17β-E2 in the presence or absence of a 200-fold excess unlabeled 17β-E2 for 16 h at 4 C in ER binding buffer [10 mM Tris (pH 7.5), 10% glycerol, 2 mM dithiothreitol, and 1 mg/ml BSA in a total volume of 100 µl].

Bound and free [3H]17β-E2 was separated using hydroxylapatite as described previously (13). The radioactivity in the bound fraction was measured in a Beckman Liquid Scintillation Spectrophotometer (LS5000TA). All aqueous counts were done in Ready Safe liquid scintillation cocktail (Beckman), and the nonaqueous counts were determined using toluene phosphor. The dissociation constant (Kd) was calculated from Scatchard plots as described previously (14, 15), and these experiments were done in quadruplicates.

Ligand competition experiments
Because most of the ring B unsaturated estrogens are not available in a radioactive form suitable for binding studies, competitive inhibition assays were used to determine their relative binding affinities (RBAs) for ER{alpha} and ERβ. In these assays, various concentrations (1–100 nM) of competitors (unlabeled equine estrogens) dissolved in ethanol and Tris assay buffer were incubated with 1 nM ER{alpha} and ERβ, and 2 nM [3H]17β-E2. Incubations were performed at 4 C for 18 h, and the bound and free steroids were separated using the hydroxylapatite method (13). Each experiment was repeated two to three times.

The relative concentration of the competing estrogen required to reduce by 50% the specific binding of [3H]17β-E2 to human ER{alpha} and ERβ was calculated (16) using the equation:

Formula
For comparative purposes, RBA of 17β-E2 for both ER{alpha} and ERβ was set at 100. The Kd for the competing estrogen (KdI; inhibitor) was determined using the Cheng-Prusoff equation:

Formula
where KdI = Kd of competing estrogen, IC50 is the concentration of inhibitor giving 50% inhibition, Lt is the concentration of [3H]17β-E2, and Kd = Kd of 17β-E2 (17).

Transfection experiments and chemiluminescent assay
HepG2 cells were cultured in MEM supplemented with 10% fetal bovine serum (FBS) as described previously (18, 19, 20). The cells were routinely maintained as monolayers in T75 flasks at 37 C. When the cells reached 80% confluence, they were subcultured in six-well plates at a cell density of 2 x 105 cells per well. When these cells reached 50% confluence, the culture medium was replaced with 2 ml phenol red free and estrogen-depleted medium supplemented with 10% charcoal-treated FBS (CTFBS) culture medium before transfection. Each of the wells then received 1 µg pERE-TA-SEAP vector in which the estrogen response element (ERE) had been inserted upstream of the TATA box and SEAP gene. As for the negative control experiments, some wells received 1 µg pSEAP-basic, which lacks eukaryotic promoter and enhancer sequences together with the empty vector of the human ER{alpha} and ERβ expression vector (1.1 µg). Transfections were performed in phenol-red free MEM supplemented with 0.5% (CTFBS) using FuGene according to the manufacturer’s protocol. For cotransfection experiments, 0.1 µg of either human ER{alpha} or ERβ or both receptor expression vectors were transfected into cells together with SEAP reporter plasmid (1.0 µg). Cotransfections were also performed with different ratios (1:1; 1:2; and 1:10) of ER{alpha} and ERβ. For all the transfection experiments, the empty vector of the human ER expression vector was used to normalize the amount of plasma DNA used in each experiment such that the amount of plasmid DNA added per well was 2.1 µg. The positive control experiment was pSEAP Basic with SV40 early promoter inserted upstream of the SEAP gene and the SV40 enhancer inserted downstream to ensure the assay operated in the linear range. After 16 h transfection, the culture media were removed, and the cells were washed twice with MEM and incubated with MEM supplemented with 0.5% CTFBS in the absence or presence of various equine estrogens (100 nM). Progesterone (100 nM) and testosterone (100 nM) were used as controls. In some experiments, various concentrations of estrogen were tested. All steroids were dissolved in ethanol, and their final concentration in the medium was 0.2%. After 24 h incubation, chemiluminescent assay was performed using 96-well microtiter plates according to the manufacturer’s protocol. Briefly, 15 µl medium from various culture conditions was mixed with 1x dilution buffer (Clontech Laboratories, Inc., Mountain View, CA), incubated for 30 min at 65 C. The samples were first cooled by placing the microplates on ice for 2–3 min, and then equilibrated to room temperature. Sixty microliters of assay buffer (supplied by Clontech Laboratories) were added to each sample and then incubated for 5 min at room temperature. Sixty microliters of 1.25 mM chemiluminescent substrate [disodium 3-(4-methoxyspiro{1,2-dioxetane-3,2'-(5'-chloro)tricyclo[3.3.1.13,7]decan}-4-yl)phenyl phosphate diluted with chemiluminescent enhancer] were added to each sample and allowed to stand for 10 min at room temperature. The SEAP chemiluminescent signals were detected using a microplate luminometer LB96V (EG&G Berthold, Groton, CT). In all transfection studies, 5 µg of an internal plasmid (pSG{Delta} LacZ) containing the Escherichia coli Lac Z gene under the control of the SV40 early promoter and enhancer was included to correct for differences in transfection and harvesting efficiency. Transfected cells were harvested as described, and β-galactosidase activities in the cell lysates were determined (21). All transfection experiments were performed in quadruplicates and each determination in triplicates. The data are presented as relative inductive efficiency (RIE), and indicate the ratio of maximal activity achieved with the test estrogen and that of 17β-E2 multiplied by 100.

Determination of ER{alpha} and ERβ protein levels
Levels of human ER{alpha} and ERβ were determined by ELISA using mouse antihuman ER {alpha} and β-antibodies (Invitrogen Canada, Inc., Burlington, Ontario, Canada) according to the manufacturer’s instructions with minor modifications. Briefly, 40 h after transfection, the cells were washed extensively with PBS and then scraped from the well in 0.5 ml solubilizing buffer [125 mM Tris-HCl (pH 8.0), containing 1% Triton X-100, 100 µM phenol-methyl sulfonyl fluoride, 1 µM leupeptin, and 1 µM pepstatin] per 1.2 x 106 cells. Cell extracts were then centrifuged at 100,000 x g for 25 min, and an aliquot of the supernatant was analyzed by ELISA. DNA content of the pellet was determined by the QuantiFluo DNA assay kit (QFDN-250; Bioassay Systems, Hayward, CA) according to the manufacturer’s protocol using calf-thymus DNA as standard. For ELISA the supernatant was diluted with 100 mM NaHCO3 (pH 8.5); various dilutions of the supernatant (1:100) were then applied to a 96-well microtiter plate and incubated with goat-antimouse horse radish peroxidase conjugated antibodies for 1 h at room temperature and then washed five times with phosphate buffered saline Tween 20 (PBST) [137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.4 mM KH2HPO4 (pH 7.4), and 0.1% Tween 20]. Wells were then incubated with goat antimouse horseradish peroxidase conjugated antibodies for 1 h at room temperature. Wells were washed with PBST and blocked a second time with 150 µl 0.1% BSA in NaHCO3 for 1 h at room temperature and then washed five times with PBST. Immunocomplexes were detected with 2',2'-azino-di-(3-ethyl-benz-thiazoline sulfonic acid) in 0.1 M citrate buffer (pH 4.2) containing 0.03% hydrogen peroxide. The green color is measured at 405 nm on a plate reader. Purified human ER{alpha} and ERβ proteins were used as standards to quantify the levels of these proteins in samples.

Statistical analysis
Statistical analysis was performed using Prism GraphPad 3.0 software (GraphPad Software Inc., San Diego, CA). For the functional assays, the results represent four independent experiments of triplicate samples. When appropriate, the data were analyzed either by one-way ANOVA with the Newman-Keuls posttest or two-way ANOVA with Bonferroni posttests.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Binding of [3H]17β-E2 with ER{alpha} and ERβ
In Fig. 2Go the results of saturation analysis with ER{alpha} and ERβ with 17β-E2 are shown. The maximal specific binding was observed at a 17β-E2 concentration of 0.15 and 0.3 nM with ER{alpha} and ERβ, respectively (Fig. 2Go, A and B). The Scatchard plots are linear and show a single class of binding sites. The Kd values calculated from these curves were 0.06 nM for ER{alpha} protein and 0.1 nM for ERβ protein. Although the 17β-E2 has nearly 2-fold lower affinity for ERβ protein in comparison to the ER{alpha} protein, the Kd values are similar to those previously reported for 17β-E2 binding to ERs in various other biological systems (5, 22, 23).


Figure 2
View larger version (20K):
[in this window]
[in a new window]

  		FIG. 2. Binding of [3H]17β-E2 with recombinant human ER{alpha} (A) and ERβ (B) in the presence or absence of excess unlabeled estradiol. Unbound radioactivity was eliminated as described in Materials and Methods, and the specific bound [3H]17β-E2 shown in the figure was calculated by subtracting nonspecific bound disintegrations per minute from total bound disintegrations per minute. Inset, Scatchard analysis of specific binding giving a Kd of 0.06 nM for ER{alpha} protein and 0.1 nM for ERβ protein.

 
RBAs of various equine estrogens for ER{alpha} and ERβ proteins
The RBAs of the remaining 10 equine estrogens (Fig. 1Go) for ER{alpha} and ERβ protein were determined by a competition binding assay using 1 nM ER{alpha} or ERβ in the presence of 2 nM [3H]17β-E2 and various concentrations (0.1–100 nM) of the unlabeled equine estrogens. The competitor binding curves obtained with each ER subtype are depicted in Fig. 3Go. The results indicate that with both receptor subtypes, the binding curves are essentially parallel and that the RBAs can be determined by calculating the amount of the nonlabeled estrogen required to reduce the [3H]17β-E2 binding by 50% (15, 16). The RBA of 17β-E2 for both receptor subtypes was arbitrarily set at 100, and the RBAs of all equine estrogens tested are given in Table 1Go, along with their KdIs.


Figure 3
View larger version (40K):
[in this window]
[in a new window]

  		FIG. 3. Dose response curves of 17β-E2 and 10 equine estrogens in the radioligand receptor binding assay using [3H]17β-E2 and recombinant human ERs {alpha} and β. Competition assays were performed as described in Materials and Methods.

 

View this table:
[in this window]
[in a new window]

  		TABLE 1. RBAs of various equine estrogens

 
In general, the RBAs of the 17β-reduced estrogens (17β-E2, 17β-Eq, 17β-Eqn, and {Delta}8,17β-E2) were 2- to 5-fold higher for both ER{alpha} and ERβ proteins in comparison to their corresponding 17{alpha}-reduced or 17-keto forms (Table 1Go). The data indicate that the binding of equine estrogens is stereospecific and structure dependent. Except for 17β-Eq, the binding affinities of all other equine estrogens tested were lower than that of 17β-E2 for both ER{alpha} and ERβ proteins. Although the binding affinities of 17β-E2 and 17β-Eq were similar for both ER{alpha} and ERβ, the binding affinities of the ring B unsaturated estrogens, with the exception of {Delta}8,17β-E2 and 17{alpha}-Eq, were two to four times greater for ERβ protein than ER{alpha} protein. Similarly, the binding affinities of E1 and 17{alpha}-E1 for ERβ were two times higher for ERβ protein than ER{alpha} protein. (Table 1Go). The order of competition with ER{alpha} was: 17β-Eq ≥ 17β-E2 > 17β-Eqn = {Delta}8,17β-E2 > 17{alpha}-Eq > E1 > 17{alpha}-Eqn ≥ {Delta}8-E1 ≥ 17{alpha}-E2 > Eqn > Eq and for ERβ 17β-Eq ≥ 17β-E2 > 17β-Eqn > {Delta}8,17β-E2 > E1 > 17{alpha}-Eqn ≥ Eq > 17{alpha}-E2 > 17{alpha}-Eq > {Delta}8-E1 > Eqn.

Functional activity of equine estrogens
The levels of ER{alpha} and ERβ after transient transfection into HepG2 cells are given in Table 2Go. These two receptor proteins were undetectable in untransfected cells. The mean levels of ER{alpha} and ERβ proteins in cells transfected with both receptor expression vectors were similar (Table 2Go). Incubation of these transfected cells with some equine estrogens showed that all were functionally active and increased SEAP activity in a dose-dependent manner as shown in Fig. 4Go. The SEAP activity increased in the presence of estrogens and, except for 17β-E2, maximum levels were observed at 1 nM and remained constant thereafter. Maximal activity with 17β-E2 was seen at 10 nM, and this difference may be due to a more rapid metabolism in Hep G2 cells in comparison to the other estrogens, particularly the ring B unsaturated estrogen (our unpublished data). In the presence of saturating doses (100 nM) of various estrogens, the relative SEAP activities observed are given in Table 3Go. 17β-E2 stimulated the activity of SEAP by ER{alpha}, to a several-fold higher level than any other estrogen tested. Progesterone and testosterone did not have significant activity. The RIEs defined as the ratio of maximal activity achieved with test estrogen and that of 17β-E2 multiplied by 100 of the other estrogens were only 12–17% that of 17β-E2. In contrast, 17β-E2 stimulated the activity of ERβ to a 5-fold lower level than that of ER{alpha}. More importantly, the RIE of other estrogens ranged from 66–290% that of 17β-E2 (Table 3Go), with Eqn (290%), {Delta}8,17β-E2 (200%), and 17β-Eqn (170%) being the three most efficacious estrogens under the conditions used. Furthermore, the functional activity of all novel ring B unsaturated estrogens was mediated via ERβ to a higher extent than via ER{alpha}. Although cotransfection studies with both ER subtypes did not alter the activity of 17β-E2 while it further enhanced the effects of 17β-Eq (200%), 17β-Eqn (160%), {Delta}8,17β-E2 (130%), and Eqn (122%). In Table 3Go, the ER{alpha} to ERβ ratios of less than one indicate greater activity via ERβ, and, thus, the major biological activity for Eqn appears to be mediated by activation of ERβ. As expected, cotransfection of ER{alpha} and ERβ did not change the lack of activation by progesterone and testosterone. To our knowledge, these are the first examples of natural steroidal estrogens that are strong activators of ERβ. The enhancement of the functional activity when both ER{alpha} and ERβ are present appears to be the sum of the two activities (Table 3Go).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Levels of ER{alpha} and ERβ in HepG2 cells after transient transfection

 

Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 4. Effect of various concentrations of equine estrogens on SEAP activity. HepG2 cells were cultured in six-well plates as described in Materials and Methods. Cells were then transfected with 1 µg pERE-TA-SEAP vector together with 0.1 µg human ER{alpha} and 0.1 µg ERβ plus 0.9 µg of the empty vector of the human ER expression vector. In all transfection experiments, 5 µg of an internal plasmid (pSG{Delta} LacZ) was used to correct for differences in transfection and harvesting efficiency. After transfection, the cells were treated with ethanol (<0.2%, control) and various concentrations of Eqn, 17β-Eqn, {Delta}8,17β-E2, and 17β-E2 as indicated in the figure. After estrogen treatment, the cells were assayed for SEAP activity as described in Materials and Methods. As a negative control, some wells were transfected with pSEAP-Basic (1.0 µg) together with the empty vector of the human ER expression vector (1.1 µg). This background activity (<5.0% of total) was then subtracted from the SEAP activity of untreated and drug-treated cells. Relative SEAP activity values represent the SEAP activity minus background, correction for transfection efficiency, and relative to drug vehicle incubation (arbitrarily set as 100%). The results represent the means of four independent experiments. Two-way ANOVA with Bonferroni posttests were used for statistical analysis. P values more than 0.05 are considered not significant (ns). For concentrations 0.01, 0.1, 1.0, 10.0, and 100.0 nM: 17β-E2 vs. Eqn, P values are not significant, not significant, less than 0.01, less than 0.001, and less than 0.001, respectively; 17β-E2 vs. 17β-Eqn, P values are not significant, less than 0.05, less than 0.001, less than 0.001, and less than 0.001, respectively; 17β-E2 vs. {Delta}8,17β-E2, P values are not significant, less than 0.05, less than 0.001, less than 0.001, and less than 0.001, respectively; Eqn vs. {Delta}8,17β-E2, P values are not significant, less than 0.01, not significant, not significant, and less than 0.01, respectively; Eqn vs. 17β-Eqn, P values are not significant, less than 0.01, not significant, not significant, and not significant, respectively; and 17β-Eqn vs. {Delta}8,17β-E2, P values are not significant for all tested concentrations.

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Relative SEAP activity of equine estrogens

 
Relationship between binding affinity and biological activity
In Figs. 5Go and 6Go, the rank order of data comparing binding affinity with functional activity of various estrogens mediated via ER{alpha} and ERβ is depicted. In general, with a number of equine estrogens, there was a lack of relationship between binding affinity and functional activity. Thus, Eqn, which has only 15 and 20% the binding affinity of 17β-E2 for ER{alpha} and ERβ, respectively (Table 1Go), exhibited the highest functional activity through ERβ (Fig. 6Go) and ranked number 3 with ER{alpha} (Table 3Go and Fig. 5Go). In contrast, 17β-Eqn had higher binding affinity compared with Eqn for both ER{alpha} and ERβ (Table 1Go), yet its functional activity was lower than that of Eqn (Fig. 6Go). Some of the other examples of this discordance between binding affinity and biological activity are depicted in Figs. 5Go and 6Go. These data clearly indicate that the earlier concept that biological activity was directly proportional to binding affinity does not apply to the pharmacology of all estrogens, however, the original concept (1, 2, 6) seems to be valid with respect to 17β-E2. This discordance between binding activity and functional activity has been observed and reviewed previously (1).


Figure 5
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 5. Lack of relationship between the rank order of binding affinity and functional activity (ER{alpha}), measured by SEAP assay, as described in Materials and Methods.

 

Figure 6
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 6. Lack of relationship between the rank order of binding affinity and functional activity (ERβ), measured by SEAP assay, as described in Materials and Methods.

 
Coexpression of various amounts of ER{alpha} and ERβ and its effect on the biological activity of some equine estrogens
Earlier studies had shown that ER{alpha} and ERβ are not functionally equivalent, with ER{alpha} being significantly more transcriptionally active than ERβ. Moreover, data also suggest that not only does ERβ modulate ER{alpha} transcriptional activity, but at subsaturating levels of 17β-E2, ERβ functions as an inhibitor of ER{alpha} transcriptional activity (22). To define further the mechanism(s) underlying the differential activation of ER{alpha} and ERβ by various equine estrogens and to investigate the effect of different amounts of ER{alpha} and ERβ on the pharmacology of estrogens, we transiently transfected HepG2 cells with different levels of ER{alpha} and ERβ. Specifically in one set of cells, we kept the level of ER{alpha} expression vector constant and increased the level of ERβ expression vector; in the second set of cells, we kept the level of ERβ constant and increased the level of ER{alpha}, together with ERE-TA-SEAP reporter. Under these conditions the transfected HepG2 cells were treated with 100 nM 17β-E2, or with some of the ring B unsaturated estrogens (17β-Eq, 17β-Eqn, {Delta}8,17β-E2, and Eqn). The results indicate that at the equivalent concentration (1:1) of ER{alpha} and ERβ, 17β-E2 stimulated the transcriptional activity, however, increasing the ERβ concentration 2 or 10-fold (1:2; 1:10) inhibited the ER{alpha} activity progressively to levels that were 32 and 34% lower than those at ER{alpha}: ERβ was (1:1) (Fig. 7Go). In contrast, a 2- to 10-fold increase in ER{alpha} vector concentration did not influence the transcriptional activity of ERβ by 17β-E2. On the contrary, there appears to be a further 15–25% increase in the transcriptional activity compared with when equal amounts of the ER{alpha} to ERβ ratio were present (1:1) (Fig. 7Go). With the novel ring B unsaturated estrogen 17β-Eq, an estrogen that has similar binding affinities as 17β-E2 for both ER isotypes (Table 1Go), a completely opposite pattern of activity was observed (Fig. 8Go). Thus, increasing the vector concentration of ERβ to give ER{alpha} to ERβ ratios of 1:1, 1:2, and 1:10 resulted in increasing levels of transcriptional activity of ER{alpha} induced by 17β-Eq, and in contrast, increasing the vector concentration of ER{alpha} to give ER{alpha} to ERβ ratios of 2:1 and 10:1 resulted in suppression of the transcriptional activity of ERβ to baseline levels (Fig. 7Go). Similar results were also observed with other ring B unsaturated estrogens such as 17β-Eqn, Eqn, and {Delta}8,17β-E2 (Figs. 9–11GoGoGo). The data further support the concept (1, 7) that the structure of the estrogen rather than the binding affinity plays a key role in the overall pharmacology of ER{alpha} and ERβ. These data further support the hypothesis that the biological activity of various estrogens depends on the relative levels of both ER{alpha} and ERβ in the estrogen target cell. Although the suppression of transcriptional activity of ER{alpha} by increasing concentrations of ERβ in the presence of 17β-E2 has been noted previously, the observations that in the presence of ring B unsaturated estrogens, the activity of ERβ is inhibited by increasing levels of ER{alpha} is to our knowledge a first such observation. Similarly, the enhancement of ER{alpha} transcriptional activity by ring B unsaturated estrogens in the presence of increasing levels of ERβ is novel and has not been reported previously. Interestingly, we observed in some of these cotransfection experiments that cells transfected with ERE plus ERβ displayed some activity in the absence of estrogen (Figs. 7–11GoGoGoGoGo); the significance or mechanism involved is not apparent at present.


Figure 7
View larger version (27K):
[in this window]
[in a new window]

  		FIG. 7. Effects of 17β-E2 on ERE-reporter gene cotransfected with different ratios of ER{alpha} and ERβ in HepG2 cells as described in the legend to Fig. 4Go. Cotransfection experiments were detailed in the legend to Fig. 4Go except different ratios (1:1; 1:2; 1:10) of ER{alpha} and ERβ were used as indicated in the figure. Relative SEAPs were calculated as described under Fig. 4Go and expressed as percentage (%) of control (drug vehicle treated cells). P values are: A vs. B, D, G, and H, P < 0.001; A vs. E, P < 0.01; A vs. E and F are not significant; B vs. D, G and H are not significant; C vs. B, D, G, and H, P < 0.001; C vs. E, P < 0.01; C vs. F, not significant; D vs. G and H, not significant; E vs. B and D, not significant; E vs. G and H, P < 0.01; F vs. B, D, G, and H, P < 0.001; F vs. E and H, P < 0.05; and G vs. H, not significant.

 

Figure 8
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 8. Effects of 17β-Eq on ERE-reporter gene cotransfected with different ratios of ER{alpha} and ERβ in HepG2 cells as described in the legend to Fig. 4Go. Cotransfection experiments were detailed in the legend to Fig. 4Go except different ratios (1:1; 1:2; 1:10) of ER{alpha} and ERβ were used as indicated in the figure. Relative SEAPs were calculated as described under Fig. 4Go and expressed as percentage (%) of control (drug vehicle treated cells). P values are: A vs. D, E, and F, P < 0.001; A vs. G, P < 0.01; A vs. B, C, and H, not significant; B vs. D–F, P < 0.001; B vs. G, P < 0.05; B vs. G, not significant; C vs. D–F, P < 0.001; C vs. G, P < 0.05; D vs. E, not significant; D vs. F, P < 0.01; E vs. F, P < 0.01; G vs. D and E, not significant; G vs. F, P < 0.001; H vs. D–F, P < 0.001; H vs. G, P < 0.01, and H vs. B and C, not significant.

 

Figure 9
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 9. Effects of 17β-Eqn on ERE-reporter gene cotransfected with different ratios of ER{alpha} and ERβ in HepG2 cells as described in the legend to Fig. 4Go. Cotransfection experiments were detailed in the legend to Fig. 4Go except different ratios (1:1; 1:2; 1:10) of ER{alpha} and ERβ were used as indicated in the figure. Relative SEAPs were calculated as described under Fig. 4Go and expressed as percentage (%) of control (drug vehicle treated cells). P values are: A vs. D–F, P < 0.001; A vs. C and G, P < 0.01; A vs. B and H, not significant; B vs. D–F, P < 0.001; B vs. C and G, P < 0.05; B vs. H, not significant; C vs. E and F, P < 0.001; C vs. D, P < 0.01; D vs. E and F, not significant; E vs. F, not significant; G vs. D, not significant, G vs. D, P < 0.01; G vs. E and F, P < 0.001; H vs. D–F, P < 0.001 and H vs. C and G, P < 0.05.

 

Figure 10
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 10. Effects of Eqn on ERE-reporter gene cotransfected with different ratios of ER{alpha} and ERβ in HepG2 cells as described in the legend to Fig. 4Go. Cotransfection experiments were detailed in the legend to Fig. 4Go except different ratios (1:1; 1:2; 1:10) of ER{alpha} and ERβ were used as indicated in the figure. Relative SEAPs were calculated as described under Fig. 4Go and expressed as percentage (%) of control (drug vehicle treated cells). P values are: A vs. D–F, P < 0.001; A vs. C, P < 0.01; A vs. G, P < 0.05; A vs. B and H, not significant; B vs. D–F, P < 0.001; B vs. C, P < 0.01; B vs. G, P < 0.05; C vs. D, not significant; C vs. E, P < 0.05; C vs. F, P < 0.001; D vs. E, not significant; D vs. F, P < 0.01; E vs. F, not significant; G vs. C and D, not significant; G vs. E, P < 0.01; G vs. F, P < 0.001; H vs. D–F, P < 0.001; H vs. C, P < 0.01; and H vs. G, P < 0.05.

 

Figure 11
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 11. Effects of {Delta}8,17β-E2 on ERE-reporter gene cotransfected with different ratios of ER{alpha} and ERβ in HepG2 cells as described in the legend to Fig. 4Go. Cotransfection experiments were detailed in the legend to Fig. 4Go except different ratios (1:1; 1:2; 1:10) of ER{alpha} and ERβ were used as indicated in the figure. Relative SEAPs were calculated as described under Fig. 4Go and expressed as percentage (%) of control (drug vehicle treated cells). P values are: A vs. D–F, P < 0.001; A vs. C and G, P < 0.01; A vs. B and H, not significant; B vs. D–F, P < 0.001; B vs. C and G, P < 0.01; B vs. H, not significant; C vs. E and F, P < 0.001; C vs. D, not significant; D vs. E, P < 0.05; D vs. F, P < 0.001; E vs. F, P < 0.05; G vs. E and F, P < 0.001; G vs. C and D, not significant; H vs. D–F, P < 0.001; and H vs. C and G, P < 0.05.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The 17β-E2 binding affinities of ER{alpha} are nearly 2-fold higher than for ERβ in our assays in HepG2 cells transfected with recombinant human ERs. We chose to perform our transfection studies in the HepG2 cell line because some of our previous work dealing with the effect of estrogens on apolipoprotein synthesis, secretion, and regulation was done in this cell line (18, 19, 20). These ER transfected cells respond to estrogens, and this cell line has also been used by others for similar transcriptional assays (24). Because in this cell line the constitutive activation function 1 (AF-1) appears to be the dominant activator, whereas in some cell models, both AF-1 and activation function 2 (AF-2) are involved in the pharmacology of ER{alpha} and ERβ (24), it would be of interest to perform similar studies in other cell types.

In our earlier studies, we had used luciferase assays, however, in the present study, we used SEAP as a transcription reporter molecule to monitor the activity of promoters and enhancers. The use of this enzyme has a number of advantages (25, 26); it allows one to determine the expression of SEAP reporter gene using simple, sensitive, nonradioactive assays of secreted phosphatase activity using disodium 3-(4-methoxyspiro{1,2-dioxetane-3,2'-(5'-chloro)tricyclo[3.3.1.13,7]decan}-4-yl)phenyl phosphate as the chemiluminescent substrate. The chemiluminescent assay can detect as little as 10–13 g SEAP protein, making it one of the most sensitive enzymatic reporters available. The assay is linear over a 104-fold range of enzyme concentrations. SEAP reporter encodes a truncated form of the placental enzyme that lacks the membrane-anchoring domain, thus allowing the protein to be secreted efficiently from transfected cells. Changes in levels of SEAP activity detected in culture medium have been directly proportional to changes in intracellular concentrations of SEAP mRNA and protein (24, 25). Furthermore, preparations of cell lysates are not required for assay, and the kinetics of gene expression can be studied simply by repeated collection of the culture medium from the same cultures. Because SEAP is extremely heat labile, endogenous alkaline phosphatase activity can be eliminated by pretreatment of samples at 65 C for 30 min. An important added advantage is that transfected cells are not disturbed by measurement of SEAP activity in the medium, so a single set of cultures can be used for both the SEAP assay, and further analyses such as RNA and protein determinations.

In the present study, the Kd values for the binding of 17β-E2 to ER{alpha} and ERβ were 0.06 and 0.1 nM, respectively. These findings are in keeping with previously reported Kds for rat ER{alpha} and ERβ (23). The RBAs of the 10 remaining equine estrogens were determined by competition assays, and the results indicated that 17β-Eq had higher RBA than 17β-E2. This observation is similar to our earlier findings that this equine estrogen had higher affinity for both rat uterine and human endometrial ER (mixture of receptor subtypes) (13). The data further indicated that the binding affinities of equine estrogens are stereospecific in that the 17β-reduced estrogens, 17β-E2, 17β-Eq, 17β-Eqn, and {Delta}8,17β-E2 had binding affinities that were 2- to 5-fold higher for both ER{alpha} and ERβ compared with their corresponding 17{alpha}-reduced or 17-ketoforms. Interestingly, RBAs of some novel ring B unsaturated estrogens were two to four times greater for ERβ protein than ER{alpha} protein (Table 1Go). Although no unique physiological/endogenous steroidal estrogen ligand for ERβ has been identified to date (5, 9, 23, 27), Gustafsson and colleagues (9) have demonstrated an androgen 5{alpha} androstane, 3β,17β-diol (3β-diol), a metabolite of 5{alpha}-dihydrotestosterone formed in the prostate, has estrogenic activity that is mediated via ERβ. Whether 3β-diol is the unique endogenous ligand for ERβ remains to be established. However, 3β-diol binds to both ER{alpha} and ERβ, albeit with slightly higher affinity with the latter (23, 29, 30, 31, 32, 33). Because 3β-diol also binds to ER{alpha}, the specificity of its activity is most likely due to its site of formation (prostate), and this has been recently reviewed (9, 33). A number of these and other (34) studies with 3β-diol suggest that ERβ ligands may be useful in the inhibition of prostatic epithelial cell proliferation. Our data indicate that some natural estrogens such as the ring B unsaturated equine estrogens of the type present in the drug CEE have the characteristics that can be useful as selective ERβ ligands. Although previous studies have reported that some synthetic compounds such as 4-OH-tamoxifen, dienestrol (4,4'-diethylidene-ethylene-diphenol), and ICI-164384 have RBAs that are higher for the ERβ than ER{alpha} (23), the present data to our knowledge, are first examples of natural steroidal estrogens that have higher binding affinities for human ERβ rather than ER{alpha}. However, there are some natural nonsteroidal compounds such as genistein and coumestrol that have had higher binding affinities for ERβ (23).

A number of studies (35, 36, 37, 38, 39, 40) have investigated the structural requirements for various diverse compounds to bind with ERs. These studies show the importance of the overall steroidal ring structure with the presence of phenolic and 17β-hydroxyl functional groups. The presence of a phenolic ring was also essential with nonsteroidal compounds, however, a few exceptions have been reported (37). Thus, antiestrogens such as tamoxifen, clomiphene, nafoxidine, and toremifene lack a phenolic ring yet are ER binders. Interestingly, 4-hydroxy tamoxifen, the active metabolite of tamoxifen, does indeed have the important phenolic hydroxyl functional group and is a stronger ER binder.

As can be seen in Fig. 1Go, that all of equine estrogens have: 1) a phenolic function (OH) at C-3 position of the aromatic A ring, 2) a carbonyl or alcohol function at C-17 position, and 3) a relatively planer and rigid steroid ring structure. Simple space-filling Stuart models of some of these estrogens are depicted in Fig. 12Go. These structures were made using CS ChemDraw Ultra and CS Chem3D Pro (Cambridge, MA). In all of these structures, the C18 methyl group is projecting above the plane of the paper. This type of simple modeling indicates differences in the orientation of the phenolic hydroxyl functional group, which appears to play an important role in the final structural conformation of ER{alpha} ligand binding domain (LBD) and ERβ LBD bound to these estrogens (39, 40, 41). Note the difference in the orientation of the phenolic hydroxyl proton among 17β-E2, Eqn, 17β-Eq, 17β-Eqn, {Delta}8,17β-E2, and Eq. Most estrogens which express their functional activity via interaction with ERβ; this phenolic proton (hydrogen) is oriented in essentially the same plane as the C18 methyl group, i.e. above the plane of the paper. This orientation of the hydrogen may influence the hydrogen bonding of the phenolic hydroxyl group of these estrogens with specific amino acids in the ligand binding pocket of the ER (39, 40, 41). Because ER{alpha} and ERβ are the members of the nuclear receptor superfamily, they display the classical characteristic of transcription factors such as a DNA-binding domain, and a C-terminal LBD made up of 12 {alpha}-helical structures (40, 41, 42). ER{alpha} and ERβ are made up of 595 and 530 amino acids, respectively, and both contain two activation functions, AF-1 in the N terminus and AF-2 in the C terminus within the LBD. Although the AF-1 and AF-2 of ER{alpha} are functional and required in most cell types, however, in target tissues such as the uterus, AF-1 is sufficient for 17β-E2 action. AF-1 is relatively weaker than AF-2, and in ERβ, AF-1 is nonfunctional or absent (22, 43). The hormone-dependent AF-2 is strongly activated when ERs are bound to agonists such as 17β-E2 and diethylstilbestrol but is inactivated when bound to antagonists such as tamoxifen and ICI 182, 780 (22, 43, 44).


Figure 12
View larger version (40K):
[in this window]
[in a new window]

  		FIG. 12. Space-filling Stuart models of some equine estrogens. Carbon atoms are black, oxygen atoms are red, and the OH group is represented by red plus blue. The numbers in parentheses represent the relative SEAP activities (see Table 3Go for details). Note the difference in the orientation of the proton of the phenolic OH group. When this proton is pointing up in the direction of the C18-methyl group, the transcriptional activation occurs to a greater extent via the ERβ.

 
The crystal structures of various ER ligand complexes indicate that the ERs can accommodate a number of ligands that contain an aromatic ring with a number of different hydrophobic groups (40). The three-dimensional x-ray structures also indicate that estrogen bound to LBD allows the formation of a hydrophobic cleft consisting of helices 3, 5, and 12 on the LBD, which in ER agonist structure functions as a docking site for a number of coactivators (steroid receptor coactivator), GRIPI, and other p160 coactivators needed for the ligand-dependent AF-2 and gene activation. Antiestrogens such as tamoxifen, due to their bulky structure, change the position of helix 12 and block the entry of coactivators into the hydrophobic cleft (40, 41, 43). Because all ring B unsaturated equine estrogens were transcriptionally active, it would appear that they, like the classical estrogens, are permissive to the formation of the hydrophobic cleft on the surface of the LBD and the subsequent docking of the coactivators, i.e. they allow the formation of an "agonist conformation" that results in the activation of transcription (45).

In general, both ER{alpha} and ERβ interact with 17β-E2, E1, estriol, and 17{alpha}-E2 with similar affinity (23); the transcriptional activity of ERβ when bound to the most potent endogenous estrogen 17β-E2 is only 20–60% of the activity of ER{alpha} (22). Moreover, when the two receptors are coexpressed, it appears that receptor heterodimers (ER{alpha}/ERβ) can be formed, and more importantly ERβ can modulate the transcriptional activity of ER{alpha}. Thus, at a low concentration of 17β-E2, ERβ is a dominant repressor of ER{alpha}’s activity, and the differential activities of the two ERs and ER ligands also depend on the specific ligand-induced conformation of ER{alpha} and ERβ (22).

Although when both ER{alpha} and ERβ are present, ERβ appears to inhibit activation of ER{alpha} by 17β-E2 (22, 43), our data support this only when activation of ER{alpha} is induced by 17β-E2, and further indicate that even when the concentration of 17β-E2 is high (100 nM), increasing concentrations of ERβ (1:2) and (1:10) decreases the transcriptional activity of ER{alpha} by 33–66%, respectively. In contrast, increasing the concentration of ER{alpha} (2:1 and 10:1) results in an increase in transcriptional activity of ER{alpha} by 15–25%, respectively. Thus, our results extend the previous observations (22) by indicating that ERβ is not only a dominant repressor of ER{alpha}’s activity at low concentrations of 17β-E2, but even at relatively higher concentration of 17β-E2 coupled with higher ERβ concentration, represses ER{alpha}’s activity. The combined results indicate that the transcriptional activity of ER{alpha} is not only dependent on the concentration of 17β-E2 but also on concentrations of ER{alpha} and ERβ in the specific cell. Previous observations (43) that in most ERE context, ERβ tends to be a weaker activator than ER{alpha}. This appears to be valid only when the classical estrogen 17β-E2 is used, whereas in contrast with some novel ring B unsaturated estrogens, the reverse effect was observed.

In contrast to the classical estrogen 17β-E2, in some of the ring B unsaturated estrogens such as 17β-Eq, 17β-Eqn, and {Delta}8,17β-E2, a completely different pattern of activity was observed. Thus, increasing concentrations of ERβ compared with ER{alpha} resulted in increased transcriptional activity, whereas increasing concentrations of ER{alpha} resulted in inhibition of ERβ’s transcriptional activity to baseline levels. The stimulatory effect of increasing ERβ levels was most impressive when Eqn was the ligand. Thus, when the ratio of ER{alpha} to ERβ was 1:10, the increase in the transcriptional activity was nearly 2-fold compared with the ER{alpha} to ERβ ratio of 1:1. To our knowledge, these are first examples of improvement of transcriptional activity of ERβ in the presence of ER{alpha} by natural but novel equine estrogens (Figs. 8–11GoGoGoGo). Together with earlier findings, our results show that the main determinants of the transcriptional activity of ER{alpha} and ERβ are the individual concentrations of these two transcriptional factors in a target cell and the structure of the estrogen ligand. Furthermore, ERβ can, depending on the structure of the estrogen, act as a dominant transcriptional activator in some cell types.

Thus, our study and previous observations (1, 22) suggest that varying concentrations of ER{alpha} and ERβ may have a profound effect on transcriptional activity. This may be of importance, e.g. in the etiology of breast cancer, in which it has been demonstrated that approximately 70% of breast tumors express ER, and most tumor cells coexpress both ER{alpha} and ERβ, although with considerable variable expression level (46, 47). Therefore, different types of ER dimers can be formed in such tumors (ER{alpha} and ERβ homodimers or ER{alpha}/ERβ heterodimers). The effect of these dimers on ERE-dependent signaling pathways is still not completely understood. Some studies have shown that ERβ opposes ER{alpha} on reporter constructs and offsets physiological effects of ER{alpha} on cell proliferation (22, 48, 49). It has been proposed that ERβ can act as a negative dominant of ER{alpha} (50, 51). Finally, the same ligand could exert opposite activities on the same promoter depending on the ER isotype expressed (52). Together, the present study is in good agreement with previous investigations that indicated the ratio ER{alpha} to ERβ plays an important role in modulating the transcriptional mechanism of a target gene in the presence of various ligands. Because most of the equine estrogens appear to operate through ERβ, their interactions may have a negative effect on the signal pathway of 17β-E2 and ER{alpha}. Here, we focused on the activity of the ER isotype from HepG2 cells toward its capacity to activate transcription at ERE using SEAP as a reporter gene upon exposure to various equine estrogens. Because the pERE-TA-SEAP promoter does not contain other potential estrogen-responsive consensus sequences such as stimulatory protein 1 or activator protein-1 (AP-1) response elements, it is highly unlikely that the results obtained for our studies are due to the other estrogen-responsive consensus elements. Previously, we have demonstrated that Eqn induced human apolipoprotein A (apoA)-I promoter activity by nearly 3-fold, and it is operating through the apoA-I electrophile/antioxidant response element (20). This response element has some resemblance to the consensus AP-1 response element. We have determined whether Fos-Jun binding to the apoA-I electrophile/antioxidant response element motif occurs (20). By performing EMSA, no DNA-protein complex was supershifted by the specific antibodies against c-fos or C-Jun (Zhang, X., J.-J. Jiao, B. R. Bhavnani, and S.-P. Tam, unpublished data). This suggests that the alternative pathway of ER action via AP-1 sites is not responsible for Eqn action. However, we could not exclude the possibility that equine estrogens may interact with stimulatory protein 1 or AP-1 sites present on the promoter of other target genes.

The high degree of functional activity of Eqn and other ring B unsaturated estrogens mediated via ERβ suggests that the structure of ERβ LBD bound to these equine estrogens can stabilize or induce a unique conformation of helix 12 that allows stimulation of AF-2 activity by greater recognition of coactivators leading to a greater transcriptional activity. This may explain that even though the binding affinity of Eqn for ERβ is extremely low (Table 1Go), its functional activity mediated via ERβ is over 2-fold higher than that of 17β-E2 (Table 3Go). These observations are novel and firmly establish that binding affinities are not the main determinants of biological activity, as demonstrated previously by others with 17β-E2 (for review, see Ref. 1).

While our work was in the final stages of submission, Greene and colleagues (53) published the crystal structure of a synthetic CEE analog 17β-methyl-17{alpha}-dihydroequilenin (NCI 122), complexed with ER{alpha} LBD and the coactivator GRIP-1. The RBA data indicated that this synthetic ring B unsaturated estrogen, along with 17β-Eqn and 17{alpha}-Eqn, had a greater β selectivity than 17β-E2, in keeping with our observations. Moreover, the crystallographic data of the NCI(122) complexed with ER{alpha} and GRIP supported the lower potency of this ring B unsaturated estrogen. Whether the natural ring B unsaturated estrogens in which the bulky 17β-methyl group is absent but present in NCI(122) will give a similar crystal structure with ER{alpha}, remains to be investigated. The need to perform similar crystallographic studies with ring B unsaturated estrogen complexed with ERβ would be of importance.

A number of nonsteroidal synthetic selective ER{alpha} and ERβ agonists have been described and recently reviewed (1, 23, 54, 55). To date, the specificity of these compounds has been tested in vivo only in animal models and in in vitro transcriptional assays. These data show a fairly high degree of selectivity, and some of these compounds have higher potencies than 17β-E2 (55). The data further indicated an important role for ERβ in the ovary, cardiovascular system, and the brain. It would be of interest to compare the potencies of these nonsteroidal ERβ selective agonists with some of the natural steroidal ring B unsaturated estrogens such as Eqn, 17β-Eq, 17β-Eqn, and {Delta}8,17β-E2 described in the present study.

Although ER{alpha} is the dominant receptor in the adult uterus, ERβ is known to be expressed in high levels in the brain (56, 57), peripheral nervous system (58), prostate (31), testis (59), ovary (60), and vascular endothelium (61). As in other estrogen target tissues such as the uterus and breast, 17β-E2 is the endogenous estrogen required for the normal function of the tissues, however, in the ERβ rich tissues, 17β-E2 action appears to be mediated essentially by ERβ and not ER{alpha} (28). A number of studies (reviewed in Ref. 9) have suggested the usefulness of ERβ agonists in the treatment and management of prostate cancer, autoimmune diseases, colon cancer, malignancies of the immune system, and neurodegeneration. The ring B unsaturated estrogens such as Eqn, 17β-Eqn, and {Delta}8,17β-E2, which appear to express their biological function by stimulating the transcriptional activity of mainly ERβ, may be of use either individually or in various combinations in the prevention or therapeutic management of some of the aforementioned disorders, particularly neurodegenerative disorders such as Alzheimer’s disease.

In conclusion, we show that the rank order of binding affinity and transactivation efficacy (functional activity) of various estrogen components of CEE are not directly related, and more importantly, are different for ER{alpha} and ERβ. A number of ring B unsaturated estrogens display considerable selectivity for ERβ, and their functional activity appears to be exerted through this receptor subtype. Depending on whether the estrogen is the classical estrogen 17β-E2 or one of the ring B unsaturated estrogens such as Eqn, ERβ can act as a dominant activator or a dominant repressor of ER{alpha}. Recent findings (2008) from the Women’s Health Initiative Estrogen Alone Trial showed that treatment of hysterectomized postmenopausal women with CEE alone for over 7 yr not only did not increase invasive breast cancer, but more importantly, may have reduced its occurrence in these women (28). Whether the differential transcriptional activities of the various estrogen components of CEE that are mediated through ER{alpha} or ERβ as described in our study played a role in the findings of the Women’s Health Initiative Estrogen Alone Trial remains to be investigated.


   Acknowledgments
 
We thank Francine Bhavnani for her excellence in the preparation of this manuscript.


   Footnotes
 
This work was supported by the Medical Research Council of Canada Grants MT-11329 (to B.R.B.) and MT 11223 (to S.P.T.), and by a basic research grant from Wyeth Pharmaceuticals (Philadelphia, PA) (to B.R.B.).

Disclosure Statement: S.P.T. and X.L. have nothing to declare. B.R.B. has received honorariums and basic research grants from Wyeth Pharmaceuticals, Women’s Health Division (Philadelphia, PA).

First Published Online July 3, 2008

Abbreviations: AF-1, Activation function 1; AF-2, activation function 2; AP-1, activator protein-1; apoA, apolipoprotein A; CEE, conjugated equine estrogen; CTFBS, charcoal-treated fetal bovine serum; 17{alpha}-Eqn, 17{alpha}-dihydroequilenin; 17β-Eqn, 17β-dihydroequilenin; 17{alpha}-Eq, 17{alpha}-dihydroequilin; 17β-Eq, 17β-dihydroequilin; 5{alpha}-androstane-3β, 17β-diol(3β-diol); Eq, equilin; Eqn, equilenin; 17{alpha}-E2, 17{alpha}-estradiol; 17β-E2, 17β-estradiol; ER, estrogen receptor; ERE, estrogen response element; E1, estrone; FBS, fetal bovine serum; LBD, ligand binding domain; PBST, phosphate buffered saline Tween 20; RBA, relative binding affinity; RIE, relative inductive efficiency; SEAP, secreted-alkaline phosphatase.

Received March 4, 2008.

Accepted for publication June 23, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. McDonnell DP 2004 The molecular determinants of estrogen receptor pharmacology. Maturitas 48(Suppl 1):S7–S12
  2. Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA 1996 Cloning of a novel receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 93:5925–5930[Abstract/Free Full Text]
  3. Evans RM 1988 The steroid and thyroid hormone receptor superfamily: transcriptional regulators of development and physiology. Science 240:889–895[Abstract/Free Full Text]
  4. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schütz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, Evans RM 1995 The nuclear receptor superfamily: the second decade. Cell 83:835–839[CrossRef][Medline]
  5. Nilsson S, Gustafsson JA 2002 Biological role of estrogen and estrogen receptors. Crit Rev Biochem Mol Biol 37:1–28[CrossRef][Medline]
  6. Clark JH, Peck Jr EJ 1979 Female sex steroids: receptors and function. New York: Springer-Verlag
  7. Katzenellenbogen JA, O'Malley BW, Katzenellenbogen BS 1996 Tripartite steroid hormone receptor pharmacology: interaction with multiple effector sites as a basis for the cell- and promoter-specific action of these hormones. Mol Endocrinol 10:119–131[Free Full Text]
  8. Bhavnani BR 2000 Pharmacology of estrogens: basic aspects. Menopause Review 5:9–22
  9. Koehler KF, Helguero LA, Haldosén LA, Warner M, Gustafsson JA 2005 Reflections on the discovery and significance of estrogen receptor β. Endocr Rev 26:465–478[Abstract/Free Full Text]
  10. Bhavnani BR 1988 The saga of the ring B unsaturated equine estrogens. Endocr Rev 9:396–416[Abstract/Free Full Text]
  11. Bhavnani BR 1998 Pharmacokinetics and pharmacodynamics of conjugated equine estrogens: chemistry and metabolism. Proc Soc Exp Biol Med 217:6–16[CrossRef][Medline]
  12. Bhavnani BR, Cecutti A, Gerulath A 1998 Pharmacokinetics and pharmacodynamics of a novel estrogen {Delta}8-estrone in postmenopausal women and men. J Steroid Biochem Mol Biol 67:119–131[CrossRef][Medline]
  13. Bhavnani BR, Woolever CA 1991 Interaction of ring B unsaturated estrogens with estrogen receptors of human endometrium and rat uterus. Steroids 56:201–210[CrossRef][Medline]
  14. Scatchard G 1949 The attraction of proteins for small molecules and ions. Ann NY Acad Sci 51:660–672[CrossRef]
  15. Pan CC, Woolever CA, and Bhavnani BR 1985 Transport of equine estrogens: binding of conjugated and unconjugated equine estrogens with human serum proteins. J Clin Endocrinol Metab 61:499–507[Abstract/Free Full Text]
  16. Korenman SG 1969 Comparative binding affinity of estrogens and its relation to estrogenic potency. Steroids 13:163–177[CrossRef][Medline]
  17. Hulme EC, Birdsall NJM 1992 Strategy and tactics in receptor-binding studies. In: Hulme EC, ed. Receptor-ligand interactions. A practical approach. Chap 4. Oxford, UK: Oxford University Press; 63–176
  18. Tam SP, Archer TK, Deeley RG 1985 Effects of estrogen on apolipoprotein secretion by the human hepatocarcinoma cell line, HepG2. J Biol Chem 260:1670–1675[Abstract/Free Full Text]
  19. Tam SP, Archer TK, Deeley RG 1986 Biphasic effects of estrogen on apolipoprotein synthesis in human hepatoma cells: mechanism of antagonism by testosterone. Proc Natl Acad Sci USA 83:3111–3115[Abstract/Free Full Text]
  20. Zhang X, Jiao JJ, Bhavnani BR, Tam SP 2001 Regulation of human apolipoprotein A1-gene expressed by equine estrogens. J Lipid Res 42:1789–1800[Abstract/Free Full Text]
  21. Zhang X, Chen ZQ, Wang Z, Mohan W, and Tam S-P 1996 Protein-DNA interactions at a drug-responsive element of the human apolipoprotein A-I gene. J Biol Chem 271:27152–27160[Abstract/Free Full Text]
  22. Hall JM, McDonnell DP 1999 The estrogen receptor β-isoform (ERβ) of the human estrogen receptor modulates ER{alpha} transcriptional activity and is a key regulator of the cellular response to estrogens and antiestrogens. Endocrinology 140:5566–5578[Abstract/Free Full Text]
  23. 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 {alpha} and β. Endocrinology 138:863–870[Abstract/Free Full Text]
  24. Norris JD, Fan D, Stallcup MR, McDonnell 1998 Enhancement of estrogen receptor transcriptional activity by the coactivator GRIP-1 highlights the role of activation function 2 in determining estrogen receptor pharmacology. J Biochem 273:6679–6688
  25. Berger J, Hauber J, Hauber R, Geiger R, Cullen BR 1988 Secreted placental alkaline phosphatase: a powerful new quantitative indicator of gene expression in eukaryotic cells. Gene 66:1–10[CrossRef][Medline]
  26. Cullen BR, Malim MH 1992 Secreted placental alkaline phosphatase as a eukaryotic reporter gene. Methods Enzymol 216:362–368[Medline]
  27. Gustafsson JA 1998 Therapeutic potential of selective estrogen receptor modulators. Curr Opin Chem Biol 2:508–511[CrossRef][Medline]
  28. Stefanick M, Anderson G, Margotis KL, Hendrix DO, Rodabough RJ, Pasket ED, Lane DS, Hubbell FA, Assaf AR, Sarto GE, Schenken RS, Yasmeen S, Lessin L, Chlebowski RT, WHI Investigators 2008 Effects of conjugated of conjugated equine estrogens on breast cancer and mammography screening in postmenopausal women with hysterectomy. JAMA 295:1647–1657
  29. Kuiper GG, Lemmen JG, Carlsson B, Corton JC, Saje SH, VanderSaag PT, van der Burg B, Gustafsson JA 1998 Interaction of estrogen chemicals and phytoestrogens with estrogen receptor β. Endocrinology 139:4252–4263[Abstract/Free Full Text]
  30. Kuiper GG, Shughrue PJ, Merchenthaler I, Gustafsson JA 1998 The estrogen receptor β subtype: a novel mediator of estrogen action in neuroendocrine system. Front Neuroendocrinol 19:253–286[CrossRef][Medline]
  31. Weihua Z, Makela S, Andersson LC, Salmi S, Sagi S, Webster JI, Jenson EV, Nilsson S, Warner M, Gustafsson JA 2001 A role for estrogen receptor β in the regulation of growth of the ventral prostate. Proc Natl Acad Sci USA 98:6330–6335[Abstract/Free Full Text]
  32. Weihua Z, Lathe R, Warner M, Gustafsson JA 2002 An endocrine pathway in the prostate, ERβ, AR,5{alpha}-androstane-3β-17β-diol, and C4PBI, regulates prostate growth. Proc Natl Acad Sci USA 99:13589–13594[Abstract/Free Full Text]
  33. Prins GS, Korach KS 2008 The role of estrogens and estrogen receptors in normal prostate growth and disease. Steroids 73:233–244[CrossRef][Medline]
  34. Imamov O, Morani A, Shim GJ, Omoto Y, Thulin-Anderson C, Warner M, Gustaffson JA 2004 Estrogen receptor β regulates epithelial cellular differentiation in the mouse ventral prostate. Proc Natl Acad Sci USA 101:9375–9380[Abstract/Free Full Text]
  35. Anstead GM, Carlson KE, Katzenellenbogen JA 1997 The estradiol pharmacophore: ligand structure-estrogen receptor binding affinity relationships and a model for the receptor binding site. Steroids 62:268–303[CrossRef][Medline]
  36. Tabira Y, Nakai M, Asai D, Yakabe Y, Tahara Y, Shinmyozu T, Noguchi M, Takatsuki M, Shimohigashi Y 1999 Structural requirements of para-alkylphenols to bind to estrogen receptor. Eur J Biochem 262:240–245[Medline]
  37. Blair RM, Fang H, Branham WS, Hass BS, Dial SL, Moland CL, Tong W, Shi L, Perkins R, Sheehan DM 2000 The estrogen receptor relative binding affinities of 188 natural and xenochemicals: structural diversity of ligands. Toxicol Sci 54:138–153[Abstract/Free Full Text]
  38. Oostenbrink BC, Pitera JW, van Lipzig MM, Meerman JH, van Gunsteren WF 2000 Simulations of the estrogen receptor ligand-binding domain: affinity of natural ligands and xenoestrogens. J Med Chem [Erratum (2001) 44:1124] 43:4594–4605[CrossRef]
  39. Katzenellenbogen JA, Muthyala R, Katzenellenbogen BS 2003 The nature of the ligand-binding pocket of estrogen receptor {alpha} and β: the search for subtype-selective ligands and implications for the prediction of estrogenic activity. Pure Appl Chem 75:2397–2403[CrossRef]
  40. Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389:753–758[CrossRef][Medline]
  41. Shiau AK, Barstad D, Radek JT, Meyers MJ, Nettles KW, Katzenellenbogen BS, Katzenellenbogen JA, Agard DA, Greene GL 2002 Structural characterization of a subtype-selective ligand reveals a novel mode of estrogen receptor antagonism. Nat Struct Biol 9:359–364[Medline]
  42. Shiau AK, Barstad D, Loria PM, Cheng L, Kushner PJ, Agard DA, Greene GL 1998 The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen. Cell 95:927–937[CrossRef][Medline]
  43. Kushner PJ, Webb P, Uht RM, Liu M-M, Price Jr RH 2003 Estrogen receptor action through target genes with classical and alternative response elements. Pure Appl Chem 75:1757–1769[CrossRef]
  44. McDonnell DP, Clemm DL, Hermann T, Goldman ME, Pike JW 1995 Analysis of estrogen receptor function in vitro reveals three distinct classes of antiestrogens. Mol Endocrinol 9:659–669[Abstract/Free Full Text]
  45. Kimbrel EA, McDonnell DP 2003 Function and mode of action of nuclear receptors: estrogen, progesterone, and vitamin D. Pure Appl Chem 75:1671–1683[CrossRef]
  46. Dotzlaw H, Leygne E, Watson PH, Murphy LC 1997 Expression of estrogen receptor-β in human breast tumors. J Clin Endocrinol Metab 82:2371–2374[Abstract/Free Full Text]
  47. Fuqua SA, Schiff R, Parra I, Moore I, Moore SK, Osborne CK, Clark GM, Allred DC 2003 Estrogen receptor β protein in human breast cancer: correlation with clinical tumor parameters. Cancer Res 53:2434–2439
  48. Lazennec G, Bresson D, Lucas A, Chauveau C, Vignon F 2001 ER β inhibits proliferation and invasion of breast cancer cells. Endocrinology 142:4120–4130[Abstract/Free Full Text]
  49. Strom A, Hartman J, Foster JS, Kietz S, Wimalasena J, Gustafsson JA 2004 Estrogen receptor β inhibits 17β-estradiol-stimulated proliferation of breast cancer cell line T47D. Proc Natl Acad Sci USA 101:1566–1571[Abstract/Free Full Text]
  50. Paruthiyil S, Parmar H, Kerekatte V, Cunha GR, Firestone GL, Leitman DC Estrogen receptor β inhibits human breast cancer cell proliferation and tumor formation by causing a G2 cell cycle arrest. Cancer Res 64:423–428
  51. Laennec G 2006 Estrogen receptor β, a possible tumor suppressor involved in ovarian carcinogenesis. Cancer Lett 231:151–157[CrossRef][Medline]
  52. Gougelet A, Mueller SD, Korach KS, Renoir JM 2007 Oestrogen receptors pathways to oestrogen responsive elements: transcriptional function-1 acts as the keystone of oestrogen receptor (ER) β-mediated transcriptional repression of ER{alpha}. J Steroid Biochem Mol Biol 104:110–122[CrossRef][Medline]
  53. Hsieh RW, Rajan SR, Sharma SK, Greene GL 2008 Molecular characterization of a B-ring unsaturated estrogen: implications for conjugated equine estrogen components of Premarin. Steroids 73:59–68[CrossRef][Medline]
  54. Veeneman GH 2005 Non-steroidal subtype selective estrogens. Curr Med Chem 12:1077–1136[CrossRef][Medline]
  55. Harris HA 2007 Estrogen receptor-β: recent lessons from in vivo studies. Mol Endocrinol 21:1–13[Abstract/Free Full Text]
  56. Shughrue PJ, Scrimo PJ, Merchenthaler I 2000 Estrogen binding and estrogen receptor characterization (ER{alpha} and ERβ) in the cholinergic neurons of the rat basal forebrain. Neuroscience 96:41–49[CrossRef][Medline]
  57. Mitra SW, Hoskin E, Yudkovitz J, Pear I, Wilkinson HA, Hayashi S, Pfaff DW, Ogawa S, Rohrer SP, Schaeffer JM, McEwen BS, Alves SE 2003 Immunolocalization of estrogen receptor β in the mouse brain: comparison with estrogen receptor {alpha}. Endocrinology [Erratum (2003) 144:2844] 144:2055–2067
  58. Bennett HL, Gustafsson JÅ, Keast JR 2003 Estrogen receptor expression in lumbosacral dorsal root ganglion cells innervating the female rat urinary bladder. Auton Neurosci 105:90–100[CrossRef][Medline]
  59. Makinen S, Makela S, Weihua Z, Warner M, Rosenlund B, Salmi S, Hovatta O, Gustafsson JÅ 2001 Localization of oestrogen receptors {alpha} and β in human testis. Mol Hum Reprod 7:497–503[Abstract/Free Full Text]
  60. Cheng G, Weihua Z, Makinen S, Makela S, Saji S, Warner M, Gustafsson JÅ 2002 A role for the androgen receptor in follicular atresia of estrogen receptor β knockout mouse ovary. Biol Reprod 66:77–84[Abstract/Free Full Text]
  61. Lindner V, Kim SK, Karas RH, Kuiper GG, Gustafsson JÅ 1998 Increased expression of estrogen receptor-β mRNA in male blood vessels after vascular injury. Circ Res 83:224–229[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 ReproductionHome page
H. H Ortega, N. R Salvetti, and V. Padmanabhan
Developmental programming: prenatal androgen excess disrupts ovarian steroid receptor balance
Reproduction, May 1, 2009; 137(5): 865 - 877.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 Bhavnani, B. R.
Right arrow Articles by Lu, X.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bhavnani, B. R.
Right arrow Articles by Lu, X.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene
*Compound via MeSH
*Substance via MeSH
Hazardous Substances DB
*ESTRADIOL
*ESTROGENIC SUBSTANCES, CONJUGATED
*TRITIUM

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals