* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download The Unusual Binding Properties of the Third Distinct Teleost
Survey
Document related concepts
Vesicular monoamine transporter wikipedia , lookup
Proteolysis wikipedia , lookup
Endocannabinoid system wikipedia , lookup
Two-hybrid screening wikipedia , lookup
Metalloprotein wikipedia , lookup
Genetic code wikipedia , lookup
Paracrine signalling wikipedia , lookup
G protein–coupled receptor wikipedia , lookup
Amino acid synthesis wikipedia , lookup
Biochemistry wikipedia , lookup
Biosynthesis wikipedia , lookup
Drug design wikipedia , lookup
Signal transduction wikipedia , lookup
Transcript
0013-7227/04/$15.00/0 Printed in U.S.A. Endocrinology 145(6):2968 –2977 Copyright © 2004 by The Endocrine Society doi: 10.1210/en.2003-0806 The Unusual Binding Properties of the Third Distinct Teleost Estrogen Receptor Subtype ERa Are Accompanied by Highly Conserved Amino Acid Changes in the Ligand Binding Domain M. B. HAWKINS AND P. THOMAS Zoology Department, North Carolina State University (M.B.H.), Raleigh, North Carolina 27695; and University of Texas at Austin, Marine Science Institute (P.T.), Port Aransas, Texas 78373 Three forms of estrogen receptor: ER␣, ER (ERb), and a second ER, ERa (formerly ER␥) are present in teleost fish. All ERas share amino acid changes in the ligand binding domain that may influence ligand specificity and receptor function. We compared binding specificities of the three ERs of the teleost fish, Atlantic croaker Micropogonias undulatus. Bacterially expressed Atlantic croaker (ac) ER␣, -b, and -a fusion proteins showed specific, high affinity binding to 17[3H]estradiol, with Kd values of 0.61 ⴞ 0.013, 0.40 ⴞ 0.006, and 0.38 ⴞ 0.059 nM, respectively. Rank orders of binding were: diethylstilbestrol ⬎⬎ ICI182780 > 4-hydroxytamoxifen > ICI164384 > estradiol > zearalenone > moxestrol > tamoxifen > estrone > 17␣-estradiol > estriol > 2-hydroxyestrone ⴝ genistein ⬎⬎ RU486 for acER␣; ICI182780 > diethylstilbestrol > 4-hydroxytamoxifen > estradiol > ICI164384 > E STROGENS HAVE CRITICAL roles in the development and function of many tissues, in particular those of the reproductive, nervous, and cardiovascular systems. Estrogens exert their effects via estrogen receptors (ERs) present in target tissues. Nuclear ERs are members of the steroid/ thyroid/retinoic acid superfamily of ligand-activated transcription factors (1, 2). All members of this family share a modular structure that consists of a variable trans-activation domain (A/B), a highly conserved DNA binding domain (C or DBD), a variable hinge region (D), a well-conserved ligand binding domain (E or LBD), and a variable C-terminal region (F) (1, 2). When ligand binds to the receptor, a conformational change occurs, and the receptor dimerizes. The receptor/ ligand complex then is transported to the chromatin where it interacts with nuclear cofactors and specific regulatory regions of target genes to modulate their transcription (transactivation) (2– 4). An important characteristic of estrogen-regulated proAbbreviations: ac, Atlantic croaker; DES, diethylstilbestrol; 17,20DHP, 17,20-dihydroxyprogesterone; E1, estrone; E2, 17-estradiol; 17␣E2, 17␣-estradiol; E3, estriol; ER, estrogen receptor; h, human; H, helix; IC50, competitor concentration that causes 50% displacement of [3H]estradiol; ICI164, ICI164384; ICI182, ICI182780; 11KT, 11-ketotestosterone; LBD, ligand binding domain; MOXE, moxestrol; 2OHE, 2hydroxyestrone; RBA, relative binding affinity; TAM, tamoxifen; TOH, 4-hydroxytamoxifen; wt, wild-type. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community. genistein > moxestrol > tamoxifen > zearalenone ⴝ estrone > estriol ⴝ 17␣-estradiol > 2-hydroxyestrone ⬎⬎ RU486 for acERb; and estradiol > diethylstilbestrol > 4-hydroxytamoxifen > ICI182780 > ICI 164384 > estriol > genistein > moxestrol > zearalenone > estrone > 17␣-estradiol > RU486 > tamoxifen > 2-hydroxyestrone for acERa. acERa showed higher relative binding affinities for estradiol, estriol, and RU486 and lower relative binding affinities for synthetic estrogens and antiestrogens than previously characterized ERs. Mutation of the conserved teleost substitutions (acERaPhe396) to the ER␣ or ERb counterpart shifted diethylstilbestrol and tamoxifen affinities toward those of wild-type acER␣ and acERb, supporting the hypothesis that the positions with conserved residue changes in teleost ERs are important to ER structure and function. (Endocrinology 145: 2968 –2977, 2004) cesses is that they are mediated by multiple nuclear ER subtypes. The discovery of a second ER in rats, ER, transformed the field of ER research by providing additional mechanistic explanations for the pleiotropic effects of estrogens (5). ER is widely distributed and has since been identified in other mammals, including humans, as well as in birds and fish (6 – 8). ER and the original ER (ER␣) have distinct, yet partially overlapping, distributions in estrogen target tissues and also have different ligand binding affinities and trans-activation properties (9 –12). The differences in binding affinities, trans-activation properties, and distribution are being exploited to find compounds that behave as estrogen agonists or antagonists in a tissue- and/or ER subtype-specific fashion (13–15). These selective ER modulators and subtype-specific ligands are promising tools in the treatment of estrogen-related disease and cancer without some of the deleterious side-effects. A more comprehensive understanding of the significance of structural differences among the ER subtypes to their binding affinities for estrogenic compounds is required to develop effective and specific drugs for osteoporosis, breast cancer, hormone replacement therapy, and many other conditions (16). We recently identified three distinct ER subtypes in a teleost fish, the Atlantic croaker (Micropogonias undulatus) (17). We designated two of these subtypes Atlantic croaker (ac) ER␣ and acER based on their sequence homology and phylogenetic relationships to previously identified ERs. Because the third subtype was genetically distinct from acER␣ 2968 Hawkins and Thomas • Unusual Binding Properties of ERa and acER, and others had proposed that fish may possess an ER␥ (18), we designated this new subtype acER␥ (17). The ER␥ subtype is present in other teleosts and arose from a gene duplication of ER early in the teleost lineage after the divergence of ray- and lobe-finned fishes. These three subtypes were subsequently identified in zebrafish (19). The acER ortholog in zebrafish is now designated ERb or ESR2b (previously ER1) to comply with official zebrafish nomenclature rules (20), whereas the acER␥ ortholog is designated ERa or ESR2a (previously ER2). The ERa designation used for zebrafish will be adopted here for the acER␥ subtype described in our previous paper (17) to standardize ER subtype nomenclature and in recognition of the fact that this subtype has not been identified subsequently in any other vertebrate classes (20). Similarly, the subtype formerly referred to as acER is renamed acERb. The ERas share a high degree of sequence similarity with other ERs in the conserved domains of the molecule. However, there are several significant amino acid changes in these domains that are shared by all of the cloned ERas (17). These amino acids may have functional significance that distinguish ERa from ER␣ and ERb. Several of these diagnostic residues are located near or within regions involved in ligand binding in mammalian ERs (21–23). Mammalian ER␣ and ER show differences in binding affinity for various estrogens and estrogenic compounds, and these differences have been attributed to specific amino acid substitutions in the LBD of ER (24). In addition, it has been proposed that species differences in the binding properties of bacterially expressed ER␣ genes are due to amino acid changes in the LBD (25). In support of this hypothesis, mutations of specific amino acids within the LBD of human, rat, and fish ER fusion proteins expressed in vitro can alter ligand binding characteristics (26 –28). The identification of three genetically distinct ERs in one species that possess naturally occurring amino acid substitutions within the LBD provides an unprecedented opportunity to investigate the role of specific amino acid substitutions in determining ER ligand binding characteristics. The binding properties of the E and F domains of acER␣, acERa, and acERb expressed in a bacterial expression system were investigated in the present study. Saturation and competition binding studies were conducted with the three recombinant proteins containing the LBDs of each receptor subtype. The relative binding affinities (RBAs) of the ER␣, ERa, and ERb subtypes to various steroid hormones, steroid hormone receptor-targeting drugs, and phytoestrogens were compared to test whether differences in their amino acid sequences are reflected in their ligand binding profiles. To further test this prediction, we examined the ligand binding specificities of acER fusion proteins that were mutated at one of these diagnostic amino acid positions (acERaPhe396). The results support the hypothesis that the amino acid substitutions have led to changes in ligand preferences and affinities. These amino acid changes may reflect distinct physiological functions for ERa and also indicate some important positions to investigate in mammalian estrogen receptors. Endocrinology, June 2004, 145(6):2968 –2977 2969 Materials and Methods Biochemicals 17-[2,4,6,7-3H]Estradiol ([3H]E2; 84 –93 Ci/mmol) was purchased from Amersham Pharmacia Biotech (Piscataway, NJ). The steroids 17,20-dihydroxyprogesterone (17,20DHP; 4-pregnen-17,20-diol-3one), 2-hydroxyestrone (2OHE; 1,3,5(10)-estratrien-2,3-diol-17-one), moxestrol (MOXE; 1,3,5(10)-estratrien-17␣-ethynyl-2,11,17-triol-11methyl ether), 11-ketotestosterone (11KT), estriol (E3), 17-estradiol (E2), estrone (E1), 17␣-estradiol (17␣E2), and cortisol were purchased from Steraloids (Newport, RI). Testosterone was obtained from SigmaAldrich Corp. (St. Louis, MO). The synthetic estrogen diethylstilbestrol (DES) was obtained from Steraloids. The antiestrogens, tamoxifen (TAM; trans-2-[4-(1,2-diphenyl-1-butenyl)phenoxy]-N,N-dimethylethylamine) and 4-hydroxytamoxifen (TOH); the antiprogestin mifepristone (RU 486; 11-(4-dimethylamino)phenyl-17-hydroxy-17-(1propynyl)estra-4,9-dien-3-one) and the fungal metabolite, zearalenone, were purchased from Sigma-Aldrich Corp. The isoflavone genistein (4⬘,5,7-trihydroxyisoflavone) was purchased from Steraloids. ICI164384 and ICI182780 (ICI164 and ICI182) were gifts from Dr. A. E. Wakeling (Zeneca Pharmaceuticals, Cheshire, UK). All other chemicals were reagent grade and purchased from general laboratory suppliers. Construction of acER fusion proteins The E and F domains [acER␣ amino acids (AAs) 211–523, acERb AAs 317– 674, and acERa AAs 287–565] of each acER cDNA were subcloned into the pET-27b⫹vector (Novagen, Madison, WI) to create fusion proteins incorporating tags for purification and detection. The fragments for subcloning were obtained by PCR of full-length cDNA clones using sequence-specific primers incorporating restriction sites. The acER constructs were transformed into NovaBlue-competent cells and sequenced in both directions to confirm their nucleotide sequence (University of Chicago Clinical Research Center DNA Sequencing Facility). The constructs were then retransformed into BL21(DE3)-competent cells (Novagen) for expression. The retention of the insert after retransformation was confirmed by restriction digestion. Expression of acER fusion proteins Host cells containing acER constructs were grown in Luria-Bertoni media (pH 7.6; 30 g/ml kanamycin, 37 C) to an approximate OD at 600 nm of 1.0. Cell cultures were cooled, and protein translation was induced with 1 mm isopropyl-b-d-thiogalactopyranoside (IPTG). After induction, cells were incubated at 25 C for 16 –20 h. Cells were harvested by centrifugation (1700 ⫻ g at 4 C for 30 min), and the cell pellet was stored at ⫺70 C. Cell pellets were weighed and then resuspended in ice-cold assay buffer [20 mm HEPES, 150 mm NaCl, 10% (wt/vol) glycerol, 1.5 mm EDTA, 6 mm monothioglycerol, and 10 mm NaMoO4] at a concentration of 3.5 ml/g pellet (⬃3.1 ⫻ 1010 cells/g). Lysozyme (10 mg/ml) was added to the resuspended cells for a final concentration of 1 mg/ml. The cell/lysozyme mixture was incubated on ice for 5 min and then sonicated (12 1-sec bursts at 30% power). Protease inhibitor cocktail set III (Calbiochem, San Diego, CA) was added (25 l/ml lysate). The crude bacterial lysate was then centrifuged (12,000 ⫻ g at 4 C for 30 min), and the supernatant was aliquoted and stored at ⫺70 C. Saturation analysis The lysate was diluted in ice-cold assay buffer, and 350 l were incubated with 50 l of varying concentrations of [3H]E2, giving final concentrations of 0.12–9.6 nm. The dilution factor (1:5 to 1:15), which resulted in saturation of binding at a final concentration of 2–3 nm [3H]E2, was determined for each lysate preparation. Nonspecific binding was determined for each concentration of [3H]E2 by adding DES to duplicate tubes for a final concentration of 10 m. Four microliters of ethanol (volume used for adding DES and other compounds) were added to total binding tubes; this amount of ethanol (total, 0.5%) did not influence binding. After overnight incubation at 4 C, free [3H]E2 was removed from bound by incubation with an equal volume of dextrancoated charcoal (0.1% dextran and 0.5% charcoal) for 10 min at 4 C, followed by centrifugation (3400 ⫻ g at 4 C for 10 min). The supernatant 2970 Endocrinology, June 2004, 145(6):2968 –2977 Hawkins and Thomas • Unusual Binding Properties of ERa was poured into scintillation vials, and 5 ml CytoScint (ICN, Costa Mesa, CA) scintillation cocktail were added. Total bound [3H]E2 was measured in a liquid scintillation counter (Beckman Coulter, Fullerton, CA). Specific binding was determined by subtracting nonspecific binding from total binding. The equilibrium Kd and binding capacity were calculated by nonlinear regression analysis using a one-site binding equation (PRISM software, GraphPad, Inc., San Diego, CA). Scatchard plots were used to linearize the specific binding data (29). Competition analysis Assays were performed essentially as described for the saturation binding experiments. All compounds were diluted in ethanol. Four microliters of each compound dilution were added to each tube before adding lysate and saturating amounts of [3H]E2 (2–3 nm). Total binding and nonspecific binding were determined by adding either ethanol or DES to additional duplicate tubes. All assays were repeated at least once (except for 11KT and 17,20DHP, which did not bind), and duplicate or triplicate tubes for each competitor concentration were run in all assays. The IC50 values were calculated using nonlinear regression curves for single site competitive binding analysis. IC50 is the competitor concentration that causes 50% displacement of [3H]E2. The data are expressed as percent total specific binding of [3H]E2 vs. log of the competitor concentration. The RBA for each competitor was calculated as the ratio of the IC50 for E2 to that of the competitor. The RBAs for the mutant constructs were calculated as the ratio of the IC50 for each compound for the acERa fusion protein to that of the mutant ER fusion protein. All analyses were performed using PRISM software (GraphPad, Inc.). Sequence alignments The LBDs of all ERs given in Table 2 were aligned to acERa using the NCBI database, BLAST. The alignments were compared with the CLUSTALX 1.5 alignment published previously (17) to identify the amino acids corresponding to each diagnostic position. Teleost ERs that possess 10 of 11 ERa diagnostic amino acids in the LBD are considered to be ERa in our discussion (17). Site-directed mutagenesis The acERa expression construct was mutated at Phe396 (TTY) to Ile (ATH) as in the ERs and to Met (ATG) as in the ER␣s. Mutations were performed at Bio S&T, Inc. (Montreal, Canada), and each mutation was confirmed by sequence analysis. The mutated constructs were transformed into BL21(DE3)-competent cells and then prepared and assayed identically to the wild-type constructs. Results Saturation analysis Saturation analysis of [3H]E2 binding to the acER␣,a,b fusion proteins demonstrated there was high affinity and saturable binding to all three ER subtypes (Fig. 1). Nonlinear regression analysis gave Kd values of 0.61 ⫾ 0.013, 0.38 ⫾ 0.059, and 0.40 ⫾ 0.006 nm for acER␣, acERa, and acERb, respectively (n ⫽ 2 or 3). Transformation of the data using Scatchard analysis was linear, indicating a single class of binding sites (Fig. 1). Ligand specificity The IC50 values and RBAs for each compound are given in Table 1. Curves used to calculate IC50 values are shown in Figs. 2-6. In general, the three acER fusion proteins bound FIG. 1. Saturation binding of [3H]E2 to acER fusion proteins: A, acER␣; B, acERa; and C, acERb. Crude bacterial extracts were incubated in the presence or absence of 10 M DES for 18 h at 4 C. Unbound radioligand was removed as described. Shown are specific binding data from a representative experiment (mean ⫾ SE for each concentration; n ⫽ 2 or 3). Specific bound radioligand was calculated by subtracting nonspecific bound counts from total bound counts. Kd values were determined from the curve shown, which represents the fit of the data by nonlinear regression analysis to a one-site binding equation. Inset, Scatchard plot transformation of the data. Hawkins and Thomas • Unusual Binding Properties of ERa Endocrinology, June 2004, 145(6):2968 –2977 2971 TABLE 1. IC50 and RBA of various competitors for acER␣, acERa, and acERb fusion proteins Compound E2 DES E1 Estriol 17␣E2 2OHE Moxestrol TAM TOH ICI164 ICI182 Genistein Zearalenone RU 486 Testosterone 11KT Cortisol 17,20DHP acER␣ acERa acERb IC50 (nM) RBA (%) IC50 (nM) RBA (%) IC50 (nM) RBA (%) 4.9 .1 48.9 126.5 51.2 197.2 10.2 19.3 1.9 3.5 .7 201.1 5.1 wb nb nb nb nb 100 4898 10 3.9 9.6 2.5 48 25.4 262 141 706 2.4 97 wb nb nb nb nb 2.7 2.8 93.1 27.1 127.1 1114 32.3 257.6 4.1 13.4 7.5 29.8 58.5 218.3 nb nb nb nb 100 96 2.9 9.8 2.1 .2 8.3 1.0 65 20 36 9 4.6 1.2 nb nb nb nb 2.7 .8 76.7 155.2 166.3 897.4 23.6 55.8 1.8 5.5 .8 14.8 72.8 wb nb nb nb nb 100 315 3.5 1.7 1.6 .3 11 4.8 144 49 324 18 3.6 wb nb nb nb nb Each value is the mean of at least two competition assays. nb, No significant displacement of 2–3 nM [3H]E2. wb, Less than 50% displacement of [3H]E2 at the highest concentration tested. FIG. 2. Competitive binding of major natural and synthetic estrogens to acER fusion proteins: A, acER␣; B, acERa; and C, acERb. Crude bacterial extracts were incubated with saturating amounts (2–3 nM) of [3H]E2 and increasing concentrations of competitors. Unbound radioligand was removed as described. Shown are specific binding data from a representative experiment (mean ⫾ SE for each concentration; n ⫽ 2 or 3). Specific bound radioligand was calculated by subtracting nonspecific bound counts from total bound counts. IC50 values were determined by nonlinear regression analysis using an equation for competition for a single binding site. The results for the competitors E2, E1, E3, and DES are shown. steroids and other compounds as expected for ERs, but there were some notable distinctions in their binding profiles. Physiological and synthetic estrogens All three acERs bound E2 (all compound abbreviations are defined in Materials and Methods) similarly with IC50 values of 4.9 nm for acER␣ and 2.7 nm for acERa and acERb (Fig. 2). DES has a 50-fold greater (relative) affinity than [3H]E2 for acER␣ and a 3-fold greater affinity for acERb. However, the DES IC50 for acERa is 2.8 nm, a value equal to or slightly lower than that of E2 for acERa (Table 1). The estrogen metabolite E3 had a 2.5- to 5.8-fold greater affinity for acERa than for acER␣ and acERb (Fig. 2 and Table 1). In addition, the rank of E3 in the order of ligand preferences was much higher for acERa. E3 was the 6th best competing ligand for acERa but the 11th compound for acER␣ and b. E1 had a 10-fold lower affinity than E2 for acER␣ and a 30-fold lower affinity for acERa and acERb (Fig. 2 and Table 1). acER␣ also had 2- to 5-fold higher affinities than acERa and acERb for 17␣E2, 2OHE, and MOXE, with IC50 values of 51.2, 197.2, and 10.2 nm, respectively (Fig. 3 and Table 1). Antiestrogens TOH, the hydroxylated metabolite of TAM, bound acER␣ and acERb with a greater affinity than E2, with IC50 values near 1.8 nm (Fig. 4 and Table 1). TOH had the third highest affinity overall for acERa, with an IC50 of 4.1 nm and an RBA 65% that of E2. The pure ER antagonist ICI164 had a 1.4-fold greater affinity for acER␣ than E2, with an IC50 of 3.5 nm (Fig. 4). The relative affinities of ICI164 for acERa and acERb were both lower than that for E2 with IC50 values of 13.45 and 5.5 nm, respectively. acER␣ had a 7-fold and acERb had a 3-fold greater relative affinity to the antiestrogen ICI182 than to E2. In contrast, the affinity of ICI182 for acERa was one third of that for E2, with an IC50 of 7.5 nm (Table 1). 2972 Endocrinology, June 2004, 145(6):2968 –2977 Hawkins and Thomas • Unusual Binding Properties of ERa FIG. 3. Competitive binding of estrogens to acER fusion proteins: A, acER␣; B, acERa; and C, acERb. Assays were conducted as described in Fig. 2. The results for the competitors E2, 2OHE, MOXE, and 17␣ E2 are shown. FIG. 4. Competitive binding of antiestrogens to acER fusion proteins: A, acER␣; B, acERa; and C, acERb. Assays were conducted as described in Fig. 2. The results for the competitors E2, TAM, TOH, ICI164, and ICI182 are shown. FIG. 5. Competitive binding of phytoestrogens to acER fusion proteins: A, acER␣; B, acERa; and C, acERb. Assays were conducted as described in Fig. 2. The results for the competitors E2, genistein (GEN), and zearalenone (ZEAR) are shown. TAM competed for binding to acER␣ and acERb much better than to acERa (Fig. 4). TAM was the second worst binder to acERa (13th out of 14), whereas it was the 8th best competitor for acER␣ and acERb. TAM bound to acER␣ best, with a 5-fold higher affinity than that to acERb and a 25-fold higher affinity than that to acERa, with an IC50 of 19.3 nm (Table 1). Naturally occurring estrogenic compounds Zearalenone had a more than 20-fold greater affinity for acER␣ than acERb and acERa, with IC50 values of 5.1, 72.8, and 58.5 nm, respectively (Fig. 5 and Table 1). In contrast, the affinity of genistein for acER␣ was the lowest of all of the compounds with measurable binding tested, whereas the affinities for acERa and acERb were 3- to 6-fold greater, with IC50 values of 29.8 and14.8 nm, respectively. Nonestrogenic ligands Interestingly, the antiprogestin RU 486 bound to acERa with a higher affinity than that of TAM, with an IC50 of 218.2 nm (Fig. 6 and Table 1). RU 486 was unable to displace 50% of the [3H]E2 with acER␣ or acERb, although there was some displacement at a concentration of 100 m competitor (Fig. 6). The C19 steroids (testosterone and 11KT) and the C21 steroids (cortisol and 17,20DHP) were unable to displace 50% of the [3H]E2 binding at the highest concentration tested (10 m). Hawkins and Thomas • Unusual Binding Properties of ERa Endocrinology, June 2004, 145(6):2968 –2977 2973 FIG. 6. Competitive binding of nonestrogenic ligands to acER fusion proteins: A, acER␣; B, acERa; and C, acERb. Assays were conducted as described in Fig. 2. The results for the competitors E2, mifepristone (RU 486), cortisol (CORT), 17,20DHP (17,20), 11KT, and testosterone (T) are shown. Site-directed mutagenesis Mutation of acERaPhe396 to the acERb residue Ile (acERaPhe-Ile) decreased the IC50 of DES from 3.8 to 1.2 nm (Fig. 8A). The RBA of DES for acERaPhe-Ile was 316% of that of acERa-wild-type (wt). The IC50 of TAM for acERaPhe-Ile decreased from 310 to 260 nm, and the RBA was 119% that of acERa-wt (Fig. 8B). Mutation of acERaPhe396 to the acER␣ residue Met (acERaPhe-Met) decreased the IC50 of DES from 3.8 to 2.5 nm (Fig. 8A). The RBA of DES for acERaPhe-Met was 152% that of acERa-wt. The IC50 of TAM for acERaPhe-Met decreased from 310 to 220 nm, and the RBA was 141% that of acERa-wt (Fig. 8B). Discussion As expected for ERs, there were general similarities in the ligand binding profiles among the three croaker ER fusion proteins and to other ERs. For example, E2, DES, TOH, ICI164, and ICI182 were the top five competitors for all three croaker receptors. These findings are consistent with those previously reported for ERs from other species (9, 11, 25, 30). However, the ligand preferences and RBAs of the acERs for these compounds showed notable differences. Specifically, acERa binding differed markedly from that of acER␣, acERb, and previously analyzed ERs. For acER␣ and acERb, the IC50 values for DES, ICI182, and TOH were 1.5to 50-fold lower (i.e. higher relative affinity) than the IC50 values for the natural ligand, E2. This is in general agreement with previous findings for other ERs. In contrast, for acERa the IC50 values for these compounds were 2.5- to 28-fold higher than those for E2. A particularly striking difference between acERa and other ERs is acERa’s lower affinity for TOH (1.5-fold higher IC50 than E2). TOH consistently shows a higher RBA than the endogenous ligand, E2, for other ERs in mammals, birds, lizards, and fish (9, 11, 25, 30). This is the first demonstration of an ER with a lower affinity to TOH than to E2. The RBA of the antiestrogen ICI164 for acERa is only 20% that of E2. This is in contrast to previous findings for ERs of other species and for croaker ER␣ and ERb, where ICI164 ranks either above or just below E2 in binding ability. In addition, the antiestrogen ICI182 has a 10- to 20-fold lower affinity for acERa than it does for acER␣ and acERb. Both of these compounds have a side-chain at the 7 position on the E2 skeleton, and perhaps this substitution is less tolerated by acERa. E3 was the sixth best competitor for acERa overall and was considerably better at competing than E1, which is in contrast to findings for other nuclear ERs, where E3 ranks below E1 and is usually among the poorer competitors tested. An exception is the channel catfish ERb, where E3 has a higher affinity than E1 (11). However, this receptor does not appear to be an ERa, as it contains only one of the amino acids (acERaV317) in the LBD diagnostic of ERas (Table 2) (17). E3 is produced by the ovary of teleost fish (31). In humans, E3 is present at high levels in the placenta (32) and is produced by adipose tissue in men and postmenopausal women (33). A distinct binding preference for an endogenous ligand such as E3 may reflect a novel function for the ERa subtype. Interestingly, mifepristone (RU 486) competed well with [3H]E2 for binding to acERa, but caused only slight displacement of [3H]E2 from the other two acERs at the highest concentration tested. This is the first evidence that RU 486 has relatively high binding activity for an ER and raises the possibility that this compound has estrogenic or antiestrogenic actions via binding to ERs. RU 486 is a 19-nortestosterone derivative and is a potent antiprogestin (34). RU 486 also has weak estrogenic activity in human breast cancer cells (35) and rat uterine myocytes (36). It is not known whether RU 486 exerts these effects in mammals via classical ER pathways. Alternatively, these effects may be indirect via aromatase inhibition (37). It seems important to investigate the potential estrogenic actions of RU 486 further because of its clinical uses in humans (34). Both acERa and acERb arose from a duplication of an ER gene early in the teleost lineage and consequently share a higher degree of amino acid identity with each other than they do to acER␣ (17). It is therefore not surprising that for some compounds acERa has similar binding affinities to acERb and mammalian ERs. For instance, there is less than a 2-fold difference in RBA between acERa and acERb for E2, E1, 17␣E2, 2OHE, MOXE, genistein, and zearalenone. acERa, acERb, and human ER (hER) have RBAs for genistein greater than those for zearalenone, a feature opposite that of ER␣s (30). In addition, rat ER has a 7-fold 2974 Hawkins and Thomas • Unusual Binding Properties of ERa Endocrinology, June 2004, 145(6):2968 –2977 TABLE 2. Amino acids in vertebrate ERs corresponding to positions diagnostic for teleost ER␣ and a subtypes (13) Human ER␣ amino acid acERa[␣*]amino acid L349 L324 [M166*] M421 F396 Y526 H495 C530 M499 M342 V317 L409 S384 D411 S386 K416 S391 T460 L435 S470 E439 Q502 T471 ERas M. undulatus (ERa) S. aurata (ER) P. olivaceous (ER) O. latipes (ER) O. niloticus (ERII) C. carpio (ER) D. rerio (ERa,2) O. mykiss (ER) C. auratus (ER1) L L L L L L L L L F F F F F F L F F H H H H H H H H H M M M M M M M M M V I V V V I V V V S S S S S S S S S S S S S S S S S S S S S S S S S S S L L L L L L L L L E E E E D E E E E T T T T T T T T T ERbs M. undulatus (ER) A. japonica (ER) I. punctatus (ER) D. rerio (ERb,1) C. auratus (ER2) L L L L L I L I I I S S S S S R R R R R M M V M M K K K K K N N N N N Q S N N T T T M S S E E E E E L A A I I ERs S. acanthias (ER) Tetrapod ERs (n ⫽ 10) L L I I Y L C C M M I V Ia D D Q K P P E E Db,c T V I I D D T3 ER␣s M. undulatus ER␣ Fish ER␣s (n ⫽ 13) M M M M Y Y C C M M I1,2,3 Tetrapod ER␣s (n ⫽ 9) L M Fi Lii Y C M Vi L Iii D D D E1,2 N4 K Mi Rii S S T Siii G G S1,2,5,6 N3 S Td,e,f,g Mb,c R R R Qiv Sequences were aligned to acER␣ and acERa using BLAST and compared to alignments published previously (13). The corresponding hER␣ amino acid, and acERa or ER␣* amino acid are given at the top of each column. Diagnostic positions previously investigated in mammalian ERs and discussed in the text are underlined. Amino acid numbering for human ER positions is the same as in Ref. 18. The amino acid positions for ERs that differ from the majority of their group have subscripts. Accessions for subscripts: a, C. jaccus, CAA70546.2 (ER); b, cow, AF110402(ER); c, human, AF215937(ER); d, chicken BAA88667(ER); e, C. japonica AAC36463.2 (ER); f, S. vulgaris AF113513(ER); g, O. aries AF177936 (ER); 1, O. niloticus AAD00245(ER␣); 2, C. gariepinus CAC37560(ER␣); 3, I. punctatus AF253505 (ER␣); 4, O. mykiss A37197 (ER) 5, C. auratus AAL12298(ER␣); 6, D. rerio AAK16740(ER␣); i, C. uniparens BAB79437 (ER); ii, X. laevis 625330 (ER); iii, C. japonica AF442965(ER␣); iv, human AAA52399 (ER␣). Accessions for teleost ERs designated ERa: M. undulatus, AF298182 (ERa, ER␥); S. aurata, AF136980(ER); P. olivaceous, BAB85623(ER); O. niloticus, AAD00246 (ERII); O. latipes, BAB79705(ER); C. carpio, BAB91218(ER); D. rerio, NM_180966 (ER2,a); O. mykiss, CAC06714(ER); C. auratus, AF061269(ER1). Accessions for teleost ERbs: M. undulatus (ERb, ER) AF298181; A. japonica, BAA19851(ER); I. punctatus, AF185568(ER); D. rerio, AJ414566, AF349413.3 (ER1, ERb); C. auratus, AF177465(ER2). ERs: S. acanthias, AAK57823(ER); S. scrofa, AF267736; rat, CAA05631; mouse, 008537. Teleost ER␣s: M. undulatus, AF298183; P. major, BAA22517; S. aurata, AF136979; P. olivaceous, BAB85622; O. aureus, CAA63774; O. latipes, BAA86925; H. trimaculatus, AF326201; S. salar, CAA61999. Tetrapod ER␣s: C. crocodilus, BAB79436; chicken, CAA27433; T. guttata, AAB81108; rat, CAA43411; mouse, AAA37580. higher RBA for genistein than rat ER␣, but zearalenone was not investigated (9). The binding profile of acER␣ is most similar to that of rainbow trout ER␣ (25). The compounds DES, TOH, and ICI164 are all better competitors than E2 for rainbow trout and croaker ER␣. In addition, zearalenone has an RBA approaching that of E2 (96% for croaker and 82% for trout) for both fish ER␣s. TAM and E1 have RBAs between 25% and 10% for both ERs, whereas E3 and genistein have RBAs less than 10%. Twenty-six amino acids within parts of helix 3 (H3), H6, H8, H11, and H12 line the mammalian ER ligand binding pocket and/or interact with bound E2 (22) (Fig. 7). These residues are highly conserved across vertebrate ERs, including those of Atlantic croaker. However, some of the amino acids lining the binding pocket and adjacent amino acids are changed in croaker ERs (Table 2 and Fig. 7). Four residue changes in the croaker ER subtypes are at positions surrounding the hER␣ pocket. These changes are conserved in other fish ER subtypes, suggesting an important role for these positions in determining species- and subtype-specific binding characteristics (Table 2). For example, acER␣Met166 is equivalent to hER␣Leu349 that interacts with the A ring of E2 in the hER␣ pocket (Fig. 7). This Leu to Met change is found in all 14 fish ER␣s identified to date (Table 2). A second residue in the euteleost ERs at the equivalent position to hER␣Met421 is changed to acERaPhe396 and acERbIle426. All but one of the nine identified euteleost ERas possess a Phe(TTY) at this position (Table 2). The Met to Ile change (ATG-ATH) in croaker ERb is shared by all but one euteleost ERbs as well as all tetrapod ERs, including hER. These mutations have therefore been highly conserved for at least 200 million yr, strongly suggesting a functional significance. The two exceptions in each of the  clades Hawkins and Thomas • Unusual Binding Properties of ERa FIG. 7. Schematic of the crystal structure of the human ER␣ ligand binding pocket (18) showing amino acid changes in croaker ERs. Amino acids that make direct hydrogen bonds are blue. Residues that line the hER␣ pocket or interact with E2 have arcs. Amino acids that are known to alter ER-ligand interactions and are changed in teleost ERs (Table 2) are red (18 –20, 24). (Anguilla japonica and Danio rerio) have a Leu (CTN or TTR) in the equivalent position (Table 2). Two additional changes, ER␣Tyr526-acERaHis495 and hER␣Cys530-acERaMet499, are conserved in all nine fish ERs that we designate as as in Table 2. The hER␣Cys530 position is unchanged in tetrapod s, but is changed to Arg in euteleost bs. This teleost-specific change differentiates ERb from the tetrapod ER as well as from the third teleost subtype, ERa. The finding that amino acid changes in this position alone can distinguish all of the vertebrate ER subtypes identified to date suggests that it is a key residue in the evolution of ER subtypes and thus may have an important role in the development of the pleiotropic actions of estrogens. Crystallographic and mutagenesis studies of mammalian ERs and mutagenesis studies of fish ERs indicate that these four residue changes may account for the differences in ligand binding profiles we observed for the croaker ERs. For example, the 17␣-hydroxyl of E2 hydrogen-bonds with hER␣His524 (equivalent to acERaHis493) (22). The additional His at hER␣Tyr526 (acERaHis495) may help stabilize the extra hydroxyl group of E3, resulting in the increased relative affinity of this endogenous steroid for acERa. Previous studies have demonstrated that the replacement of hER␣Tyr526 (acERaHis495) with Ala raises the IC50 of E2-induced transcriptional activity 4-fold (24), suggesting that this position is important to E2 binding. In this study genistein had a higher affinity for acERa and acERb than for acER␣. This is analogous to findings for hERs, where hER␣Met421 switches to hERIle373. This residue is on the ␣ face of the cavity where the O4 of genistein’s flavone ring lies (24). The same change occurs from acER␣ to acERb (acER␣Met238 to acERbIle426), whereas acERa changes to Phe396. It has been proposed (20) that the change in the slightly polar Met for the less polar Ile allows for more polar substituents at the distal end of the cavity, resulting in Endocrinology, June 2004, 145(6):2968 –2977 2975 the higher affinity of hER for genistein. As the Phe in acERa is also nonpolar, it may account for the increased affinity of genistein for acERa as well. hER␣ Met421 (acERaPhe396) contacts DES and TOH on their A⬘ and B rings, respectively, but does not contact E2 (23). In addition, the region of the pocket that includes this residue changes conformation depending on the ligand. acERa has at least a 10-fold lower affinity for DES, TAM, and TOH compared with acER␣ and acERb, suggesting that the mutation in the ERas to the larger and less polar Phe may be responsible for its unusual binding affinities for these compounds. The studies we present here provide further evidence of the functional importance of this position to ligand recognition. We show that mutation of acERaPhe396 to the acERb residue Ile increases the RBA of DES to 316% of that of acERa (Fig. 8A). This increase accounts for 90% of the difference seen in binding affinity between acERa and acERb (Table 1). These data strongly suggest that this position is critical to the interaction of ERs with DES. Alternatively, the change of acERaPhe396 to the acER␣ amino acid Met increased the RBA of DES to just 150% of acERa, only 5% of the 13-fold difference seen between the wt receptors. The change from Phe to Met also seems to slightly increase the affinity of the ER mutant construct for TAM, but this increase is just 10% of the difference seen between acERa and acER␣ (Fig. 8B). It is likely that a Met in this position interacts with additional ␣-specific amino acids to produce the higher affinities that vertebrate ER␣s have for DES and TAM. The conserved ERa residue acERaMet499 might also alter ligand binding characteristics for DES, TAM, and TOH. Affinity labeling studies suggested that hER␣Cys530 (acERaMet499) might be involved in binding of the antiestrogen, TAM aziridine (38). However, site-directed mutagenesis and carboxymethylation studies contradict this finding (28, 39). hER␣Cys530 is located in H11 adjacent to residues known to be critical to ligand binding of the D ring of E2 (21, 28). Even if this residue is not directly on the face of the cavity, Met (acERa) is larger and slightly more polar than Cys (hER␣) and does not form disulfide bonds. Therefore, this substitution may indirectly alter the size and shape of the cavity such that compounds with diphenolic structures such as DES, TAM, and TOH do not fit as well into the ERa pocket. acER␣ binds E2 with about a 2-fold weaker affinity than acERa or acERb. Zebrafish ER␣ also has nearly a 2-fold weaker affinity for E2 than ERa, but unlike croaker, ERb also has a 2-fold weaker affinity for E2 (19). In the rainbow trout ER␣ (and in all other euteleost ER␣s; Table 2), there is a conservative substitution of hER␣Leu349 to a Met(acER␣Met166). Reciprocal mutagenesis at this position causes a temperature-dependent 2-fold lowering of Kd for the rainbow trout ER␣ (26). This substitution may therefore be responsible for the lower affinity of E2 for acER␣. In contrast, acER␣ binds 17␣E2, E1, and 2OHE better than acERa and acERb, albeit binding is low for all three compounds compared with that for E2. It is possible that acER␣Met166 may play a role in these differences as well, but relative affinities to other estrogens in the reciprocal mutagenesis system were not evaluated (26). 2976 Endocrinology, June 2004, 145(6):2968 –2977 FIG. 8. IC50 values for DES (A) and TAM (B) in competitive binding assays with acERa-wt and acERa-mutant fusion proteins. The amino acid Phe396 of the acERa-wt fusion protein was mutated to Met (F to M) or Ile (F to I) and used in competitive binding assays as described in Figs. 2– 6. Data are averaged from two independent experiments (each performed in triplicate), with error bars representing the SEM. The RBA for each competitor was calculated as the ratio of the IC50 for each compound with the wt construct to that of the mutant construct. hER␣Glu353 hydrogen-bonds with the 3-hydroxyl group on the A ring of estrogens and is critical to the discrimination of 3-hydroxyl estrogens from the C19 and C21 3-ketosteroids (40). This Glu is 100% conserved among ERs, including the acERs. It is therefore surprising that acERa binds RU 486, given that this compound possesses a keto group at position 3 as in C19 and C21 steroids. The 17-hydroxyl orientation in the  position and the lack of the C19 methyl group in RU 486 must allow for some binding of this compound to ERs. It would be interesting to know whether this affinity of RU 486 for acERa is universal to all ERas, and consequently which, if any of the conserved amino acids are responsible for this difference. Hawkins and Thomas • Unusual Binding Properties of ERa This study sheds light on the molecular nature of ligand interactions with ERs, but does not address how these compounds, once bound to the croaker ERs, modulate transcription of estrogen-responsive genes. Compounds can be either agonists or antagonists of ER actions depending on the conformational changes they induce in the receptor upon binding. These conformational changes alter the ability of ERs to interact with cell-specific regulatory proteins and subsequently activate gene transcription (16). For instance, TAM binds to both mammalian ER␣ and ER, but only activates ER␣ (41). In addition, TAM and TOH do not activate any of the zebrafish ERs, which is in contrast to findings for mammalian ERs (19, 42, 43). It is not known whether these differences seen in zebrafish are due to species differences in ER trans-activation or ligand binding or due to the population of coactivators present (or absent) in the transfected cells. However, it does seem likely that, like zebrafish, croaker ERs will exhibit some trans-activation properties different from those of mammalian ERs, because zebrafish and croaker share many of the teleost-specific amino acid substitutions in regulatory regions of the molecule (17). The three acERs provide a natural starting point for uncovering key amino acid positions involved in receptor function. The acERs represent three clades of ERs that diverged more than 150 million yr ago. These groups, in particular the ERas, possess distinct amino acid substitutions that arose after their divergence and were then nearly all retained in the members of the clade (Table 2). This strong degree of conservation suggests that these positions are critical to receptor function. Evidence from receptor ligand studies of mammalian ERs allowed us to identify four conserved amino acid substitutions in the acERs that might be involved in the different binding profiles we observed in this study. Mutation of one of these substitutions, acERaPhe396 to the corresponding ER␣ (Met) or ERb (Ile) amino acid, shifted the binding affinities for DES and TAM toward those for the corresponding wt acER␣ and acERb. This is the first demonstration of a direct role for this position in ligand discrimination and supports the hypothesis that the amino acid changes at this position have been highly conserved because of its functional significance. Other conserved substitutions within the fish ERa clade point to at least seven additional amino acids that could be important (Table 2). These positions were largely overlooked in earlier mammalian studies because of the sequence conservation between mammalian ER␣ and ER subtypes and a lack of direct ligand interactions. The availability of a native receptor model in which residue changes have evolved together to create a functional protein with novel binding properties could provide useful information on the roles of these positions in mammalian systems. More site-directed mutagenesis studies are needed to determine the role that each of these positions plays in ligand binding. This approach may ultimately lead to the development of more ER subtype-specific agonists and antagonists and could also focus studies on trans-activation, receptor dimerization, and cofactor recruitment. Acknowledgments We thank Dr. Yong Zhu for assistance with establishing the bacterial expression system and Dr. John Godwin for laboratory support. Hawkins and Thomas • Unusual Binding Properties of ERa Received June 27, 2003. Accepted February 17, 2004. Address all correspondence and requests for reprints to: Dr. Mary Beth Hawkins, Department of Zoology, North Carolina State University, Box 7617, Raleigh, North Carolina 27695. E-mail: beth_hawkins@ ncsu.edu. This work was supported by University of Texas Marine Science Institute, the Harry Page Marine Science Fellowship, the Houston Livestock Show and Rodeo Natural Sciences Fellowship and by Texas Sea Grant R/ES-92 and EPA STAR Grant R827399 (to P.T.). References 1. Tsai MJ, O’Malley BW 1994 Molecular mechanisms of action of steroid/ thyroid receptor superfamily members. Annu Rev Biochem 63:451– 486 2. Hewitt SC, Korach KS 2002 Estrogen receptors: structure, mechanisms and function. Rev Endocr Met Dis 3:193–200 3. Beato M, Sanchez-Pacheco A 1996 Interaction of steroid hormone receptors with the transcription initiation complex. Endocr Rev 17:587– 609 4. Torchia J, Glass C, Rosenfeld MG 1998 Co-activators and co-repressors in the integration of transcriptional responses. Curr Opin Cell Biol 10:373–383 5. Kuiper G, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA 1996 Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 93:5925–5930 6. Mosselman S, Polman J, Dijkema R 1996 ER: identification and characterization of a novel human estrogen receptor. FEBS Lett 392:49 –53 7. Foidart A, Lakaye B, Grisar T, Ball GF, Balthazart J 1999 Estrogen receptor- in quail: cloning, tissue expression and neuroanatomical distribution. J Neurobiol 40:327–342 8. Todo T, Adachi S, Yamauchi K 1996 Molecular cloning and characterization of Japanese eel estrogen receptor cDNA. Mol Cell Endocrinol 119:37– 45 9. Kuiper G, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors ␣ and . Endocrinology 138: 863– 870 10. Barkhem T, Carlsson B, Nilsson Y, Enmark E, Gustafsson JA, Nilsson S 1998 Differential response of estrogen receptor ␣ and estrogen receptor  to partial estrogen agonists/antagonists. Mol Pharmacol 54:105–112 11. Xia ZF, Gale WL, Chang XT, Langenau D, Patino R, Maule AG, Densmore LD 2000 Phylogenetic sequence analysis, recombinant expression, and tissue distribution of a channel catfish estrogen receptor . Gen Comp Endocrinol 118:139 –149 12. Xia ZF, Patino R, Gale WL, Maule AG, Densmore LD 1999 Cloning, in vitro expression, and novel phylogenetic classification of a channel catfish estrogen receptor. Gen Comp Endocrinol 113:360 –368 13. Lonard DM, Smith CL 2002 Molecular perspectives on selective estrogen receptor modulators (SERMs): progress in understanding their tissue-specific agonist and antagonist actions. Steroids 67:15–24 14. Harris HA, Katzenbellenbogen JA, Katzenbellenbogen BS 2002 Characterization of the biological roles of the estrogen receptors, ER␣ and ER, in estrogen target tissues in vivo through the use of an ER␣-selective ligand. Endocrinology 143:4172– 4177 15. Harris HA, Albert LM, Leathbury Y, Malamas S, Mewshaw RE, Miller CP, Kharode YP, Marzolf J, Komm BS, Winneker C, Frail DE, Henderson RA, Zhu Y, Keith Jr JC 2003 Evaluation of an estrogen receptor- agonist in animal models of human disease. Endocrinology 144:4241– 4249 16. McDonnell DP 2003 Mining the complexities of the estrogen signaling pathways for novel therapies. Endocrinology 144:4237– 4240 17. Hawkins MB, Thornton JW, Crews D, Skipper JK, Dotte A, Thomas P 2000 Identification of a third distinct estrogen receptor and reclassification of estrogen receptors in teleosts. Proc Natl Acad Sci USA 97:10751–10756 18. Enmark E, Pelto-Huikko M, Grandien K, Lagercrantz S, Lagercrantz J, Fried G, Nordenskjöld M, Gustafsson JA 1997 Human estrogen receptor -gene structure, chromosomal localization, and expression pattern. J Clin Endocrinol Metab 82:4258 – 4265 19. Menuet A, Pellegrini E, Anglade I, Blaise O, Laudet V, Kah O, Pakdel F 2002 Molecular characterization of three estrogen receptor forms in zebrafish: binding characteristics, transactivation properties, and tissue distributions. Biol Reprod 66:1881–1892 20. Sprague J, Doerry E, Douglas S, Westerfield M 2001 The Zebrafish Information Network (ZFIN): a resource for genetic, genomic and developmental research. Nucleic Acids Res 29:87–90 21. Ekena K, Weis KE, Katzenellenbogen JA, Katzenellenbogen BS 1997 Dif- Endocrinology, June 2004, 145(6):2968 –2977 2977 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. ferent residues of the human estrogen receptor are involved in the recognition of structurally diverse estrogens and antiestrogens. J Biol Chem 272:5069 –5075 Brzozowski AM, Pike ACW, 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 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 Pike ACW, Brzozowski AM, Hubbard RE, Bonn T, Thorsell AG, Engstrom O, Ljunggren J, Gustafsson JA, Carlquist M 1999 Structure of the ligandbinding domain of oestrogen receptor  in the presence of a partial agonist and a full antagonist. EMBO J 18:4608 – 4618 Matthews J, Celius T, Halgren R, Zacharewski T 2000 Differential estrogen receptor binding of estrogenic substances: a species comparison. J Steroid Biochem Mol Biol 74:223–234 Matthews JB, Clemons JH, Zacharewski TR 2001 Reciprocal mutagenesis between human ␣(L349, M528) and rainbow trout (M317, I496) estrogen receptor residues demonstrates their importance in ligand binding and gene expression at different temperatures. Mol Cell Endocrinol 183:127–139 Danielian PS, White R, Hoare SA, Fawell SE, Parker MG 1993 Identification of residues in the estrogen-receptor that confer differential sensitivity to estrogen and hydroxytamoxifen. Mol Endocrinol 7:232–240 Ekena K, Weis KE, Katzenellenbogen JA, Katzenellenbogen BS 1996 Identification of amino acids in the hormone binding domain of the human estrogen receptor important in estrogen binding. J Biol Chem 271:20053–20059 Scatchard G 1949 The attractions of proteins for small molecules and ions. Ann NY Acad Sci 51:660 – 672 Kuiper G, Lemmen JG, Carlsson B, Corton JC, Safe SH, van der Saag PT, van der Burg P, Gustafsson JA 1998 Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor . Endocrinology 139:4252– 4263 Ponthier JL, Shackleton CHL, Trant JM 1998 Seasonal changes in the production of two novel and abundant ovarian steroids in the channel catfish (Ictalurus punctatus). Gen Comp Endocrinol 111:141–155 Peter M, Dorr HG, Sippell WG 1994 Changes in the concentrations of dehydroepiandrosterone-sulfate and estriol in maternal plasma during pregnancy: a longitudinal-study in healthy women throughout gestation and at term. Horm Res 42:278 –281 Grodin JM, Siiteri PK, Macdonal. Pc 1973 Source of estrogen production in postmenopausal women. J Clin Endocrinol Metab 36:207–214. Baulieu EE 1989 Contragestion and other clinical-applications of RU-486, an antiprogesterone at the receptor. Science 245:1351–1357 Jeng MH, Langanfahey SM, Jordan VC 1993 Estrogenic actions of RU486 in hormone-responsive MCF-7 human breast-cancer cells. Endocrinology 132: 2622–2630 Dibbs KI, Sadovsky Y, Li XJ, Koide SS, Adler S, Fuchs AR 1995 Estrogenic activity of RU-486 (mifepristone) in rat uterus and cultured uterine myocytes. Am J Obstet Gynecol 173:134 –140 Schmidt M, Loffler G 1997 RU486 is a potent inhibitor of aromatase induction in human breast adipose tissue stromal cells. J Steroid Biochem Mol Biol 60:197–204 Harlow KW, Smith DN, Katzenellenbogen JA, Greene GL, Katzenellenbogen BS 1989 Identification of cysteine-530 as the covalent attachment site of an affinity-labeling estrogen (ketononestrol aziridine) and antiestrogen (tamoxifen aziridine) in the human estrogen-receptor. J Biol Chem 264:17476 –17485 Hegy GB, Shackleton CHL, Carlquist M, Bonn T, Engstrom O, Sjoholm P, Witkowska HE 1996 Carboxymethylation of the human estrogen receptor ligand-binding domain-estradiol complex: HPLC/ESMS peptide mapping shows that cysteine 447 does not react with iodoacetic acid. Steroids 61:367–373 Ekena K, Katzenellenbogen JA, Katzenellenbogen BS 1998 Determinants of ligand specificity of estrogen receptor-␣: estrogen versus androgen discrimination. J Biol Chem 273:693– 699 Watanabe T, Inoue S, Ogawa S, Ishii Y, Hiroi H, Ikeda K, Orimo A, Muramatsu M 1997 Agonistic effect of tamoxifen is dependent on cell type, ERE-promoter context, and estrogen receptor subtype: functional difference between estrogen receptors ␣ and . Biochem Biophys Res Commun 236:140 – 145 Bardet PL, Horard B, Robinson-Rechavi M, Laudet V, Vanacker JM 2002 Characterization of oestrogen receptors in zebrafish (Danio rerio). J Mol Endocrinol 28:153–163 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 . Mol Endocrinol 11:353–365 Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.