Agonist activity of the 3-hydroxy metabolites of tibolone through the oestrogen receptor in the mouse N20.1 oligodendrocyte cell line and normal human astrocytes.

17beta-oestradiol (E(2)) may have a beneficial impact on the development of age-related diseases, in part through alpha and beta oestrogen receptors (ER) in glia. Tibolone, a synthetic steroid, could influence glial-mediated neuroprotection if agonist oestrogenic activity is demonstrable. We used the N20.1 mouse oligodendrocyte cell line as a glial cell model to evaluate the response of ERalpha and ERbeta through oestrogen-response element (ERE) and AP-1-driven reporters to E(2), 4-hydroxytamoxifen (4OHT) and to two tibolone metabolites, 3alpha-hydroxytibolone (3alpha-OH-Tib) and 3beta-hydroxytibolone (3beta-OH-Tib). In addition, we tested the activity of these same ligands through the endogenous ERalpha in human normal astrocytes. Because endogenous ER was not detected in the N20.1 cells, we tested the ability of exogenous ER to activate transcription in response to ligands (100 nM) using a transient cotransfection assay with an ERalpha expression vector. To test the antagonist activity of 3alpha-OH-Tib and 3beta-OH-Tib, we used them in combination with E(2) (10(-8) M), at concentrations of 10(-7) M and 10(-6) M. The human normal astrocytes were treated similarly, with the exception that no ER-encoding DNA was used. Specific ER ligand mediated activity was shown using the E(2) antagonist ICI 182 780 and the pSG5 empty vector. E(2), 3alpha-OH-Tib, and 3beta-OH-Tib stimulated ERalpha on an ERE-promoter at each concentration (P < 0.001) but not at an AP-1-driven promoter. 4OHT was an effective antagonist, but did not exhibit agonist activity on the ERE-driven promoter. 4OHT was an effective agonist through ERalpha on an AP-1-driven promoter. 3alpha-OH-Tib and 3beta-OH-Tib were not effective antagonists of E(2). Both metabolites acted through the ER because the addition of an E(2) antagonist blocked their activity. These results show that 3alpha-OH-Tib and 3beta-OH-Tib exert agonist activity, yet lack antagonist or additive activity, through the ERalpha and ERbeta on an ERE-driven but not on an AP-1-driven promoter in a glial cell model and in normal human astrocytes. This contrasts with the effects of 4OHT, which exerted little or no agonist activity, but reduced E(2)-stimulated activity through ERalpha on the ERE, in the same cells.

Activity of estradiol and selective estrogen receptor modulators in the mouse N20.1 oligodendrocyte/astrocytes cell line.

OBJECTIVE: The mechanism through which estrogen exerts its neuroprotective and anti-neurodegenerative effects in the central nervous system is poorly understood. Human glial cells are implicated in the pathogenesis of Alzheimer's disease and have both alpha and beta estrogen receptors (ER). We developed a glial cell model for ER function using the N20.1 mouse oligodendroglial cell line to evaluate the response of ERalpha and ERbeta to estradiol (E2), a raloxifene analog LY117018 (LY) and 4-hydroxytamoxifen (4OHT). DESIGN: We tested the ability of exogenous ER to activate transcription in response to ligands (100 nM) using the glial cell line N20.1 in a transient cotransfection assay with an ERalpha or ERbeta expression vector, an ERE-driven reporter and a Renilla luciferase transfection control. RESULTS: Endogenous ER was not detected in the N20.1 cells by Western immunoblotting. E2 stimulated both ERalpha and ERbeta on both ERE- and AP-1 driven promoters. The transcription stimulation by E2 in the ERalpha and ERbeta through the AP-1driven promoter, though significant, was not of the same magnitude as the stimulation of the ERalpha through the ERE-driven promoter. 4OHT and LY did not show significant transcriptional activation of either the ERalpha or ERbeta, through either the ERE or AP-1 driven promoters. LY, at a 10-fold higher concentration than E2, showed a difference in its antagonist activity on the ERbeta through the AP-1 pathway when compared with the ERE- driven promoter, demonstrating not only promoter specificity, but also receptor specificity. CONCLUSIONS: This is the first description of the activity of 4OHT and LY on estrogen receptors in glia.

Mutations of tyrosine 537 in the human estrogen receptor-alpha selectively alter the receptor's affinity for estradiol and the kinetics of the interaction.

Mutation of tyrosine 537 (Y537) of the human estrogen receptor-alpha (hERalpha) produces receptors having a range of constitutive activity, which suggests that this residue modulates the conformational changes of the receptor. We investigated the effect of several mutations at this position, to phenylalanine (Y537F), to serine (Y537S), and to glutamic acid (Y537E), on the hormone-binding properties of the receptor. The affinities of the wt, the Y537F mutant, and the Y537S mutant for estradiol were similar: K(a) = 2.2 +/- 0.2, 3.9 +/- 0.5, and 2.8 +/- 0.4 nM(-1), respectively. By contrast, the affinity of the Y537E mutant for estradiol was reduced 10-fold, K(a) = 0.2 +/- 0.1 nM(-1). All proteins bound [(3)H]estradiol with a positive cooperative mechanism (n(H) = 1.7-1.9), indicating they can form dimers. The wt receptor and the Y537S and Y537E mutants exhibited biphasic dissociation kinetics, which is also indicative of dimerization. Surprisingly, the half-lives of the slow component of the wt and the Y537E mutant were indistinguishable, 118 +/- 3.4 and 122 +/- 4.5 min, respectively, even though the affinity of the Y537E mutant for hormone was reduced 10-fold. The half-life of the slow component of the Y537S mutant was reduced to 96.5 +/- 3.8 min. Molecular models were constructed and compared to identify changes in the structure that correlate with the observed effects on hormone binding. Local alterations in hydrogen bonding, the position of side chains, and the position of the peptide backbone were observed. Taken together, these results show that mutations at Y537 selectively alter the affinity and kinetics of hormone binding to the receptor, and are consistent with the idea that the estradiol-estrogen receptor interaction can follow more than one pathway.

Hydrocortisone modulates the effect of estradiol on endothelial nitric oxide synthase expression in human endothelial cells.

The interaction between hydrocortisone and estradiol on the regulation of endothelial nitric oxide synthase (eNOS) expression was investigated in human umbilical vein endothelial cells (HUVECs). Following incubation in medium containing dextran-coated-charcoal-stripped serum (DCC-stripped medium) for 4 days, incubation of HUVECs with 0.1 nM estradiol for 24 hr in the absence of hydrocortisone increased levels of eNOS mRNA measured by ribonuclease protection assay above control (0 nM estradiol). 2 microM hydrocortisone applied for 24 hr preceding and during estradiol application inhibited the estradiol-elicited increase in eNOS mRNA levels, reducing mRNA levels from 134% +/- 14% of control to 85% +/- 5% of control. Significant (ANOVA, p<0.01) reductions of estradiol-mediated increases of mRNA levels occurred over a range of hydrocortisone concentrations (10 nM, p<0.05; 2 microM, p<0.05; n=3-12). In the presence of 2 microM hydrocortisone, 10 nM estradiol significantly reduced eNOS mRNA levels to 59% +/- 3% of control. The ability of hydrocortisone to block or reverse the estradiol-mediated increase in eNOS mRNA levels may provide a link between elevated hydrocortisone levels and decreased NO production, potentially contributing to the development of hypertension and cardiovascular disease in vivo and antagonizing cardioprotective effects of estrogens.

Formation of a powerful capping motif corresponding to start of "helix 12" in agonist-bound estrogen receptor-alpha contributes to increased constitutive activity of the protein.

The human estrogen receptor alpha (hERalpha), a ligand-activated transcription factor, provides an excellent system for study of the conversion of a protein from inactive to active states. It binds to many different ligands which leads which to the expression or the suppression of genes in a cell-specific and promoter-specific manner, and a multitude of mutations have been identified that modify the activity of the receptor. Helix 12 is a key alpha-helix in the hormone-binding domain of the hERalpha that is directly involved in transcription activation. In this report, tyrosine 537 has been identified as the Ncap residue of helix 12 in the structures of the agonist-bound hERalpha ligand-binding domain. A capping motif has been identified in the sequence of the hERalpha that corresponds to the start of helix 12 in the fully active, agonist-bound conformation of the receptor. Analysis of the literature indicates that, with one exception, substitution of amino acids at position 537, which occur more frequently at the Ncap than tyrosine, is correlated with increased constitutive activity of the hERalpha. The results are consistent with the hypothesis that formation of a powerful capping motif that corresponds to the start of helix 12 in the agonist-bound receptor contributes to activation of the hERalpha. This is the first proposed example in which mutations that alter helix capping would lead to the constitutive activation of a protein. This hypothesis could therefore provide a novel mechanism through which mutations play a role in pathological processes.

Function of estrogen receptor tyrosine 537 in hormone binding, DNA binding, and transactivation.

The human estrogen receptor (hER) is a ligand-activated transcription factor which, like many other members of the nuclear receptor protein family, exhibits a dimerization-dependent transcriptional activation. Several previous reports have provided evidence of the phosphorylation of the hER at tyrosine 537 (Y537). However, the exact function of a putative phosphorylation at this site remains controversial. Using a yeast transactivation assay, and in vitro biochemical approaches, we show that phosphorylation of tyrosine 537 is not required for the hER to bind hormone, or to activate transcription. An hER tyrosine 537 to phenylalanine (Y537F) mutant retains 70-75% of the transactivation potential of wild type hER in a yeast reporter system. Furthermore, the mutated receptor exhibits wild type hormone and DNA binding affinities. However, this mutation results in a decrease in receptor stability as measured by a decrease in the extent of hormone binding over time. The most striking difference between the wild type and Y537F hER is in the estradiol binding kinetics. Whereas the off-rate for estradiol exhibits a two-state binding mechanism, the Y537F mutant hER exhibits a monophasic estradiol off-rate. On the basis of these data and other reports describing the structure and activity of Y537 mutations, as well as knowledge of the three-dimensional structure of the hER ligand binding domain, we propose an alternate model wherein Y537F mutation favors an "open" pocket conformation, affecting the estrogen binding kinetics and stability of the hormone-bound, transcriptionally active "closed" pocket conformation. Although its phosphorylation is not essential for function of the hER, Y537 is nevertheless a critical residue intricately involved with the conformational changes of the hER and its ability to activate transcription.

Effects of an environmental endocrine disruptor on fetal development, estrogen receptor(alpha) and epidermal growth factor receptor expression in the porcine male genital tract.

PURPOSE: We studied the effect of a potent reproductive tract toxin, 2,3,7,8-tetrachlorodibenzo-rho-dioxin, on fetal development and expression of estrogen receptor alpha and epidermal growth factor receptor (EGFR) in male swine. MATERIALS AND METHODS: Fetal domestic swine and miniswine were injected with 1 microg./kg. dioxin on day 50 of gestation and removed near term (114 days). Germ cell counts were performed on sections of formalin fixed testes. Estrogen receptor a protein, and messenger ribonucleic acid (mRNA) and EGFR mRNA expression were analyzed in frozen tissue using Western blotting and semiquantitative reverse transcriptase polymerase chain reaction. RESULTS: Of 15 dioxin exposed male offspring 8 (53%) had genital anomalies, including cryptorchidism in 4, epididymal detachment in 1, epididymal atresia in 1 and vasal dilatation in 3, while 3 of 17 control male swine (18%) had incompletely descended testes (p = 0.06). High intra-abdominal testes were found in 3 of 4 cryptorchid dioxin exposed but no control male swine. Mean germ cell number per tubule was 4.0+/-1.1 and 2.7+/-0.7 in control and dioxin groups, respectively (p = 0.01). Estrogen receptor a protein and mRNA were identified in fetal uterus, testis, gubernaculum and epididymis. Protein levels were 2 to 3-fold higher in dioxin exposed testis, and mRNA levels were significantly lower in gubernaculum and epididymis. EGFR mRNA expression was similar in treated and control testis and epididymis. CONCLUSIONS: Preliminary data suggest that dioxin produces cryptorchidism and wolffian duct anomalies in male swine exposed just before mid gestation. Germ cell counts and estrogen receptor alpha mRNA expression in gubernaculum and epididymis were significantly reduced, and estrogen receptor a protein expression in testis appeared to be increased by dioxin exposure. Aberrant regulation of estrogen receptor a expression by dioxin may contribute to reproductive tract anomalies in male fetuses.

Interleukin-1beta-induced nitric oxide production in rat aortic endothelial cells: inhibition by estradiol in normal and high glucose cultures.

Expression of inducible nitric oxide synthase (iNOS) and the resultant increased nitric oxide (NO) production are associated with septic shock, atherosclerosis, and cytokine-induced vascular injury. Estrogen is known to impact vascular injury and vascular tone, in part through regulation of NO production. In the current study, we examined the effect of physiological concentrations of estradiol on interleukin-1beta (IL-1beta)-induced NO production in rat aortic endothelial cells (RAECs). 17Beta-estradiol significantly decreased IL-1beta-induced iNOS protein levels and reduced NO production in RAECs. High glucose (25 mM) elevated the increase in IL-1beta-induced iNOS protein and NO production. Nevertheless, estradiol still inhibited IL-1beta-induced iNOS and NO production even in the presence of high glucose. These data suggest that estradiol may exert its beneficial effects in part by inhibiting induction of endothelial iNOS, a possible mechanism for the protective effect of estradiol against diabetes-associated cardiovascular complications.

Cardiovascular disease in the diabetic woman.

The incidence of cardiovascular disease (CVD) with age is increasing in the United States, and elderly women constitute a disproportional component of the aging population. Elderly women also have a relatively high incidence of diabetes, which contributes to this relatively high CVD risk. Although CVD is less common in premenopausal women than in men, this difference begins to disappear after the onset of menopause, presumably related to decreased levels of female sex hormones (estrogen and/or progesterone). Diabetes mellitus removes the normal premenopausal gender-related differences in the prevalence of CVD by mechanisms that are not clearly defined, including metabolic and hemodynamic factors associated with diabetes. Dyslipidemia in diabetes mellitus consists of low high density lipoprotein cholesterol, elevated triglyceride levels, and a small, dense, more atherogenic low density lipoprotein particle (i.e. oxidized). Dyslipidemia interacts with associated hemodynamic (i.e. hypertension) and metabolic abnormalities (i.e. increased platelet aggregation and plasminogen activator inhibitor-1 levels) to promote CVD risks in diabetic women. Recent controlled trials underscore the critical importance of aggressively treating CVD risk factors, especially dyslipidemia, in women with diabetes.

Clinical review 91: Female sex hormones and cardiovascular disease in women.

Cardiovascular disease is the leading cause of mortality in women, a fact that is underappreciated by women and physicians. Clinical and experimental data underscore the cardioprotective effects of female sex hormones, particularly estrogen. Indeed, the loss of female sex hormones after menopause contributes to the striking increase in the incidence of cardiovascular morbidity and mortality after menopause. Estrogen replacement therapy improved lipoprotein profiles in the postmenopausal women, but this accounts for less than half of the cardioprotective effects of estrogen replacement therapy. Addition of progestins to estrogen therapy in women appears not to significantly attenuate the cardioprotective effects of estrogen replacement therapy despite experimental data suggesting otherwise. This review addresses potential mechanisms, other than influences on lipoproteins, by which estrogen and progesterone exert their cardiovascular protective effects. Particular emphasis is directed to genomic and nongenomic effects of estrogen and progesterone that are exerted directly on cardiovascular tissue.

Hinge and amino-terminal sequences contribute to solution dimerization of human progesterone receptor.

We and others have shown previously that progesterone receptors (PR) form homodimers in solution in the absence of DNA and that dimers are the preferential form of receptor that binds with high affinity to target DNA. To determine the sequence regions involved in solution homodimerization, wild type PR and truncated PR proteins were expressed in an insect baculovirus system. The expression constructs included the ligand-binding domain [LBD, amino acids (aa) 688-933], the LBD plus hinge (hLBD, aa 634-933), the hLBD plus the DNA-binding domain (DhLBD, aa 538-933), and the full- length A and B isoforms of PR. PR-PR interactions were detected by three methods, coimmunoprecipitation of the PR fragments with full-length PR-A, pull-down of PR-polypeptides with polyhistidine-tagged versions of the same polypeptides immobilized to metal affinity columns and cooperative ligand-binding assays (Hill coefficient, n(H) > 1 indicating PR-PR interaction). By all three assays, the LBD alone was not sufficient to mediate protein-protein interaction. However, the LBD did exhibit other properties ascribed to this domain, including binding to steroids with a relatively good affinity and specificity, ligand-induced conformational changes at the carboxyl terminus tail and binding of heat shock protein 90 and its dissociation in response to hormone. Thus, failure of the expressed LBD to mediate dimerization does not appear to be due to an extensively misfolded or unstable polypeptide. The minimal carboxyl-terminal fragment capable of mediating PR-PR interaction was the hLBD construct. However, by immobilized metal affinity chromatography assay, self-association of PR-A was 3.5-fold more efficient than that of either the DhLBD or hLBD constructs. An expressed amino-terminal domain (aa 165-535) lacking the DNA-binding domain, hinge, and LBD was found to physically associate with PR-A or with another amino-terminal fragment lacking the LBD, but retaining the DNA-binding domain. These results provide evidence for direct amino-terminal interactions in the more efficient PR-PR interaction exhibited by wild-type PR-A, as compared with DhLBD and hLBD constructs. The overall results of this paper are consistent with the conclusion that the carboxyl-terminal LBD is not sufficient for mediating PR dimerization and that multiple regions, including the hinge and amino-terminal sequences, contribute either directly or indirectly to homodimerization of PR.

Binding of [3H]progesterone to the human progesterone receptor: differences between individual and mixed isoforms.

The human progesterone receptor (hPR) exists as two isoforms, hPR-A and hPR-B, which differ only in that hPR-A lacks 164 amino acids present at the amino-terminus of hPR-B. In this study we have separately expressed hPR-A and hPR-B and asked whether the progesterone-binding mechanisms are the same or different for the two forms of hPR and for their mixture. We investigated 1) the cooperativity of binding [3H]progesterone to the receptor, as measured by the Hill coefficient (nH); and 2) the dissociation rate of [3H]progesterone from the receptor. To compare the effects of dimerization, these ligand-binding properties were measured over a range of receptor concentrations. Binding of [3H]progesterone to hPR-A was positively cooperative at all concentrations used; the limiting value for the Hill coefficient was 1.47 +/- 0.11 at high receptor concentrations (5-19 nM) and 1.31 +/- 0.06 at low receptor concentrations (1-4 nM). Similarly, little change was observed in the dissociation rate constant over the same concentration range; the values at high and low concentrations were 4.59 +/- 0.15 and 3.03 +/- 0.25 x 10(-3) min-1, respectively. By contrast, the hPR-B concentration had a marked effect on positive cooperative binding and the dissociation rate of progesterone. At high hPR-B concentrations (3-5 nM), the limiting Hill coefficient was 1.49 +/- 0.11, which is indicative of moderately strong positive cooperativity, whereas at lower hPR-B concentrations (1-3 nM), the Hill coefficient was reduced to 1.1, which is essentially noncooperative. The [3H]progesterone dissociation rate was 4.52 +/- 0.44 x 10(-3) min-1 at the higher concentrations of hPR-B and was increased to 1.6 +/- 0.11 x 10(-3) min-1 at the lower concentrations. Thus, over the same concentration range where hPR-A exhibited no significant change in positive cooperativity or the dissociation rate, these progesterone-binding properties were highly dependent on the concentration of hPR-B. When hPR-A and hPR-B were mixed, positive cooperative binding and the dissociation rate were more similar to hPR-B than to hPR-A, in that both binding parameters were dependent on the concentration of receptor. However, the hPR-AB mixture differed from hPR-B alone in that the mixture required a greater receptor concentration (7-10 vs. 3-5 nM) to exhibit positive cooperativity and the increased dissociation rate. These results show, first, that each hPR isoform displays different [3H]progesterone-binding properties, which are most prominent at low concentrations of receptor, and second, that one isoform can influence the other. As the two receptor forms differ only at the N-terminus, yet positive cooperativity and changes in the dissociation rate constant are indicative of conformational changes affecting hormone binding, these results also strongly suggest that the N-terminus may directly or indirectly interact with the C-terminal ligand-binding domain.

A-ring nitro- and amino-substituted estradiol analogs produce a negative cooperative or noncooperative [3H]estradiol-estrogen receptor binding mechanism.

We have investigated the relation between ligand structure and binding mechanism between the calf uterine estrogen receptor. A series of structurally altered estradiol analogs was used in which either an amino- or a nitro group had been added to the 2 or 4 position on the phenolic A-ring. The binding affinity of both amino analogs and the 4-nitro analog for the estrogen receptor was reduced relative to that of estradiol, as measured by competitive binding assay; the values were between 0.008% and 8% of estradiol's affinity. The slope of the displacement curve for the 4-nitro analog was also significantly different from that of estradiol (p < 0.05), indicating that the binding mechanism of these two ligands was different. The affinity of the 2-nitroestradiol ligand for the receptor was too low to be measured. The binding mechanism was then further investigated by measuring the Hill coefficient of [3H]estradiol binding in the presence of the analog. The presence of a nitro group on C4 eliminated the positive cooperativity of the [3H]estradiol-estrogen receptor interaction; the Hill coefficient of [3H]estradiol binding in the presence of the analog was 0.99 compared with 1.7 for [3H]estradiol alone. Most interestingly, the presence of an amino group on either C2 or C4 brought about a switch from a positive to a negative cooperative binding interaction; the Hill coefficients of [3H]estradiol binding in the presence of the analogs were between 0.6 and 0.7. These results provide additional support for an induced-fit mechanism of ligand-estrogen receptor interactions.

Ligand-mediated modulation of estrogen receptor conformation by estradiol analogs.

The studies presented here show how changing the structure of the ligand can affect the conformation of the receptor. Five different estradiol analogs have been tested for binding to the calf uterine estrogen receptor. In three of the analogs the phenolic hydroxyl group had been moved from the 3 to the 1, 2, or 4 position on the A-ring (1-hydroxyestratrien-17 beta-ol, 2-hydroxyestratrien-17 beta-ol, or 4-hydroxyestratrien-17 beta-ol). In the remaining two analogs either the A- or the D-ring hydroxyl group had been removed altogether (estratrien-17 beta-ol or 3-hydroxyestratriene). Competition binding assay showed that the relative binding affinity for the estrogen receptor had been weakened by all changes in the structure of the ligand. Furthermore, the ligands in which either the 3 beta- or the 17 beta-hydroxyl group was missing produced nonparallel slopes in the linear portions of the displacement curves compared to that of estradiol; the ligands in which the phenolic hydroxyl had simply been moved around the A-ring, however, did not. These observations implied that the receptor binding mechanism used by the monohydroxyl ligands was different from that of estradiol. Saturation binding analysis showed that while the presence of any of the dihydroxyl ligands or that of estratrien-17 beta-ol decreased the positive cooperativity of the [3H]estradiol-estrogen receptor interaction, the presence of the 3-hydroxyestratriene ligand increased it. These results suggest that both the binding mechanism and the affinity of the ligand for the receptor are exquisitely sensitive to the structure of the ligand.

A metal-linked gapped zipper model is proposed for the hsp90-glucocorticoid receptor interaction.

In the presence of certain metals, regions of the hormone binding domain of the glucocorticoid receptor (GR) are capable of binding the 90 kDa heat shock protein (hsp90). Using secondary structure prediction methods in correlation with the experimental data, we propose a model which predicts the presence of two widely spaced leucine zipper-like heptads on either side of a central subdomain. The heptads could interact hydrophobically with similar regions on the hsp90 homodimer, bringing putative metal binding residues on each protein close enough to establish a shared metal bridge. The central subdomain between heptads is suggested to contain regions involved in metal binding, steroid binding, and conformational mobility. The hypothetical model that we are proposing therefore addresses the nature of the structural link between hsp90 binding, hormone binding, and conformational changes in the receptor.

Dimerization of the RU486-bound calf uterine progesterone receptor.

Dimerization of the RU486-bound progesterone receptor was studied by measuring the Hill coefficient of RU486 binding in calf uterine cytosol at receptor concentrations between 0.3 and 15 nM. The limiting value of the Hill coefficient at high receptor concentrations was 1.38 +/- 0.01. The limiting value of the Hill coefficient at low receptor concentrations was 1.05 +/- 0.03. The dimerization constant, defined as the concentration of receptor at which the Hill coefficient was midway between the limiting values, was 2.6 +/- 0.1 nM. In contrast, the dimerization constant of the progesterone-bound receptor, which was measured using the same approach, is 7 nM [Skafar, D. F. Biochemistry 30 (1991) 6148-6154]. The results presented here support and quantify the observation that the RU486-bound progesterone receptor will dimerize at lower receptor concentrations than the progesterone-bound receptor.

Differences in the binding mechanism of RU486 and progesterone to the progesterone receptor.

The binding mechanism of the antagonist RU486 to the progesterone receptor was compared with that of the agonists progesterone and R5020. Both progesterone and RU486 bound to the receptor with a Hill coefficient of 1.2, indicating the binding of each ligand is positive cooperative. However, when each ligand was used to compete with [3H]progesterone for binding to the receptor at receptor concentrations near 8 nM, at which the receptor is likely a dimer, the competition curve for RU486 was significantly steeper than the curves for progesterone and R5020 (p less than 0.001). This indicated that a difference in the binding mechanism of RU486 and progesterone can be detected when both ligands are present. In contrast, at receptor concentrations near 1 nM, at which the receptor is likely a monomer, the competition curves for all three ligands were indistinguishable (p = 0.915). These results indicate that RU486 and agonists have different binding mechanisms for the receptor and further suggest that this difference may be related to site-site interactions within the receptor.

Differential DNA binding by calf uterine estrogen and progesterone receptors results from differences in oligomeric states.

The studies presented here provided evidence that the calf uterine estrogen and progesterone receptors exhibit different DNA-binding properties in vitro as a result of having different dimerization constants. The affinity of the estrogen and progesterone receptors for DNA was measured by using isocratic elution from DNA-Sepharose. The hormone-free estrogen receptor had a 10-fold higher affinity for DNA than did the hormone-free progesterone receptor when measured at receptor concentrations of 6-12 nM and 180 mM KCl. No effect on DNA binding by binding progesterone to its receptor was detected. This contrasts with the increased affinity for DNA and increased number of ions released upon DNA binding exhibited by the hormone-bound estrogen receptor. Between 2 and 3 ions were released when the progesterone receptor and the diluted estrogen receptor bound DNA. These observations suggested the progesterone receptor was in the monomeric state, whereas the estrogen receptor was in the dimeric state at receptor concentrations of 6-12 nM. When the dimerization constant of the progesterone receptor was measured, the value of approximately 7 nM obtained was 20-fold higher than the value of 0.3 nM reported for the estrogen receptor. This makes it likely the two receptors exist in different forms at the same concentration in vitro. It is also suggested the predominant form of the estrogen and progesterone receptors in vivo could differ.

Affinity and conformation of the estrogen receptor when bound to single-stranded vs. double-stranded nonspecific DNA.

The affinity of the hormone-bound estrogen receptor for single-stranded and double-stranded DNA was compared using isocratic elution chromatography. The receptor bound single-stranded DNA with a two-fold higher affinity than double-stranded DNA (17.9 x 10(4) M-1 vs. 9.1 x 10(4) M-1) at 0.2 M KCl. The same number of ions were released when the receptor bound either single-stranded or double-stranded DNA (11.8 vs. 10.6, respectively). These results indicate the hormone-bound estrogen receptor has no strong preference for single-stranded vs. double-stranded nonspecific DNA, and has a similar conformation when bound to either form of DNA at physiological salt concentrations.

Modulation of the estrogen receptor's affinity for DNA by estradiol.

The binding constant for estrogen receptor-DNA interaction when measured in the presence and absence of estradiol revealed a distinct difference dependent upon whether the receptor was hormone-bound or hormone-free. The binding constant of estrogen receptor-DNA interaction was determined by analysis of the exponential elution profile of the estrogen receptor from DNA-Sepharose columns using Tris buffer at a constant salt concentration. The binding constant of the hormone-bound estrogen receptor for DNA in Tris buffer, pH 7.4, containing 0.2 M KCl was 10.1 +/- 0.8 X 10(6) M-1, 5-fold higher than the value for the hormone-free estrogen receptor. Analysis of the number of ionic bonds between the estrogen receptor and DNA indicates that the hormone-free receptor establishes eight salt bridges, while the hormone-bound estrogen receptor establishes 10-13. The affinity of the hormone-bound estrogen receptor for DNA in Tris buffer at pH 7.4 in 0.2 M KCl is 10-fold greater than at pH 8.0, suggesting that ionic bonding between the receptor and DNA may involve histidine residues of the receptor. The concentration-dependence of the hormone-bound receptor's affinity for DNA emphasizes the receptor's associative state as an influence on the receptor's DNA binding characteristics. Our results demonstrate that estradiol modifies the conformation of the estrogen receptor to a state having an increased affinity for DNA.