Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
The Glucocorticoid Antagonist 17a-Methyl testosterone Binds to the 10 S Glucocorticoid Receptor and Blocks Agonist-Mediated Dissociation of the 10 S Oligomer to the 4 S Deoxyribonucleic Acid-Binding Subunit Bruce M. Raaka, Molly Finnerty, and Herbert H. Samuels Division of Molecular Endocrinology Departments of Medicine and Pharmacology New York University Medical Center New York, New York 10016 agonist for binding to the 10 S cytosolic receptor but does not appear to promote dissociation of the oligomer, thus inhibiting agonist-mediated nuclear actions of the glucocorticoid receptor. (Molecular Endocrinology 3: 332-341, 1989) The glucocorticoid antagonist 17a-methyltestosterone inhibits binding of the agonist [3H]triamcinolone acetonide to the glucocorticoid receptor in cytosol prepared from rat pituitary tumor GH, cells. Competitive binding studies indicate that the dissociation constant for 17a-methyltestosterone is about 1 MM. After incubation of intact GHt cells with 10 nM [3H] triamcinolone acetonide at 37 C and subsequent cell fractionation at 4 C , three glucocorticoid receptor forms are observed: cytosolic 10 S receptor, cytosolic 4 S receptor, and nuclear receptor. Concurrent incubation with 17«-methyltestosterone reduces the amount of [3H]triamcinolone acetonide bound to each of these receptor forms. Ligand-exchange assays performed at 0 C in intact cells using [3H] triamcinolone acetonide show that the exchangeable antagonist is associated predominantly with cytosolic 10 S receptor. Immunochemical analysis using monoclonal antibody BuGR2 indicates that 17a-methy I testosterone does not cause substantial accumulation of glucocorticoid receptors in GHn cell nuclei and, when present together with agonist, reduces nuclear accumulation of receptor seen with agonist alone. Results from dense amino acid labeling studies show that unlike [3H]triamcinolone acetonide, 17a-methyltestosterone does not reduce the total amount of cellular glucocorticoid receptor and does not reduce receptor half-life. These results are consistent with a model for glucocorticoid receptor transformation in which binding of agonist promotes the dissociation of an oligomeric 10 S cytosolic receptor protein to its DNA-binding 4 S subunit. The antagonist 17a-methyltestosterone competes with INTRODUCTION When cells or tissues are deprived of steroid hormones, steroid receptor proteins are generally found in the soluble "cytosolic" fraction of broken cell preparations. After exposure to steroid hormones, 60-90% of the occupied receptor is found in the particulate "nuclear" fraction. This apparent redistribution of receptor within the cell suggests that binding of hormone transforms receptor to a form capable of high affinity interactions with chromatin components in the cell nucleus, where the hormone-receptor complex acts to regulate the expression of specific genes (1, 2). Results of studies from this laboratory using rat pituitary tumor GH, cells led to the proposal of a model for glucocorticoid receptor transformation (3, 4). According to this model, hormone binds initially to an oligomeric 10 S receptor protein and shifts the apparent equilibrium between the oligomer and its hormone- and DNA-binding 4 S subunit in the direction of the subunit. The hormone-occupied subunit then appears to interact reversibly with chromatin components in the cell nucleus. Evidence from a variety of experimental systems supports the conclusion that steriod receptor transformation involves hormone-mediated dissociation of a receptor oligomer (59). An early study by Samuels and Tomkins (10) described the effect of the antagonist 17a- 0888-8809/89/0332-0341 $02.00/0 Molecular Endocrinology Copyright © 1989 by The Endocrine Society 332 333 The Glucocorticoid Antagonist 17«-Methyltestosterone methyltestosterone1 on the induction of tyrosine aminotransferase by glucocorticoid agonists in HTC cells. These authors interpreted their results in the context of a theoretically based allosteric model in which two interconvertible forms of glucocorticoid receptor exist with different abilities to induce expression of the enzyme. In the context of a model in which receptor transformation requires agonist-mediated oligomer dissociation, antagonist might act by competing with agonists for binding to the oligomeric receptor without promoting subunit dissociation. Recent direct ligandbinding studies with the high affinity glucocorticoid antagonist [3H]RU38486 document that this compound binds to the oligomeric receptor but does not promote dissociation (11, 12). In this report, we present experimental evidence from ligand-exchange assays showing that 17a-methyltestosterone, an antagonist with low affinity for the glucocorticoid receptor, also binds to the 10 S cytosolic receptor but, unlike agonists, does not appear to convert receptor to the DNA-binding form. Furthermore, we show for the first time that a glucocorticoid antagonist is able to competitively block oligomer dissociation mediated by an agonist. These results provide additional support for the conclusion that agonist-mediated oligomer dissociation plays a central role in glucocorticoid receptor transformation (3-9). RESULTS Binding of 17«-Methyltestosterone to the Glucocorticoid Receptor 17a-Methyltestosterone is an antagonist of the glucocorticoid-mediated induction of tyrosine aminotransferase in HTC cells (10) and of the stimulation of GH synthesis and mRNA accumulation in rat pituitary tumor GHT cells (13). In both cases, the antagonist has no detectable agonist activity and at sufficiently high concentrations is capable of fully blocking the action of agonists. The concentration dependence of antagonist activity along with other evidence suggested that both agonists and antagonists interact with a specific glucocorticoid receptor system (10). A competitive binding study confirms that 17a-methyltestosterone inhibits the binding of the synthetic glucocorticoid agonist [3H] triamcinolone acetonide to the glucocorticoid receptor in cytosol prepared from GHT cells (Fig. 1). Under the conditions used for this experiment, virtually all of the cytosolic receptor is known to sediment at 10 S (3, 4) and the protamine sulfate assay identifies only the 10 S receptor form (14). Thus, both steroids appear to compete for the same binding site on the 10 S receptor. The inhibition curve for 17a-methyltestosterone is shifted more than three orders of magnitude toward higher concentrations than the corresponding curve for 1 The trivial names used are: 17a-methyltestosterone, 17/3hydroxy-17«-methyl-4-androsten-3-one; triamcinolone acetonide, 9a-fluoro-11 j8,16a,17a,21 -tetrahydroxypregna-1,4diene-3,20-dione-16,17-acetonide. 17Q - M E T H Y L TESTOSTERONE O"1O"1O 10-9 10-8 10-7 10"6 10~5 10"4 STEROID CONCENTRATION (M) Fig. 1. Inhibition of [3H]Triamcinolone Acetonide Binding to the Glucocorticoid Receptor by 17«-Methyltestosterone Aliquots of cytosol prepared from GHi cells were adjusted to contain 3 nM [3H]triamcinolone acetonide and the indicated concentration of either unlabeled triamcinolone acetonide or 17«-methyltestosterone and incubated at 4 C for 18 h. Binding of [3H]triamcinolone acetonide to the cytosolic glucocorticoid receptor was measured by protamine sulfate precipitation (3). The data are not corrected for nonspecific binding, which was less than 3% of total specific binding. triamcinolone acetonide. Since the dissociation constant (Kd) for triamcinolone acetonide binding to the glucocorticoid receptor from GHi cells is about 1 nM (3), this result suggests a Kd of about 1 HM for 17amethyltestosterone, in reasonable agreement with an earlier estimate of 0.4 HM for binding to the glucocorticoid receptor from HTC cells (15). Effects of 17a-Methyltestosterone on the Intracellular Distribution of Glucocorticoid Receptor Forms Direct assessment of [3H]17a-methyltestosterone binding to the glucocorticoid receptor is not feasible because of the low affinity of receptor for this steroid. Consequently a ligand-exchange assay in intact cells was used to study effects of this antagonist on the intracellular distribution of receptor forms. GHT cells were exposed to unlabeled 17a-methyltestosterone at 37 C either alone or in combination with [3H]triamcinolone acetonide. The cells were then chilled to 0 C and the culture medium was replaced several times over a 1- to 2-h period with medium without steroids. Finally, the cells were incubated at 0 C for 4-5 h with 10-15 nM [3H]triamcinolone acetonide to exchange and reoccupy receptor binding sites with the radioactive ligand. This 4- to 5-h incubation at 0 C is sufficient to achieve virtually complete occupany of glucocorticoid receptor in GHT cells not previously exposed to steroids (3). The entire exchange assay was performed at 0 C which has been shown to minimize interconversion of 10 S and 4 S cytosolic receptor forms and redistribution of receptor forms between cytosolic and nuclear compartments (3). GHT cells were incubated at 37 C for 1 h with 10 nM [3H]triamcinolone acetonide or 20 HM 17a-methyltes- MOL ENDO-1989 334 Vol 3 No. 2 amount of 4 S receptor was decreased in comparison with cells receiving agonist alone. In cells that received either no steroid or 17a-methyltestosterone alone at 37 C, the 10 S receptor predominated. The amounts of 10 S and 4 S cytosolic receptors were quantified from the gradient results shown in Fig. 2 and the amount of nuclear receptor was determined in washed nuclei from each cell culture (Table 1). In addition, Table 1 documents the efficiency of the exchange assay. Before the ligand exchange procedure in cells treated with both 10 nM [3H]triamcinolone acetonide and 20 HM 17a-methyltestosterone, occupancy of each receptor form by the labeled steroid was greatly reduced and total occupancy was only 29% of that found in control cells treated with the agonist alone. This confirms that 17a-methyltestosterone inhibits binding of the glucocorticoid agonist in intact cells. After ligand exchange, total occupancy by [3H]triamcinolone acetonide rose to 66% of control, indicating that the exchange procedure is only partially effective. If occupied with the antagonist, it is not known whether ligand exchange of 4 S or nuclear receptors is possible under the conditions used for these experiments. However, the amount of cytosolic 10 S receptor identified after exchange increased more than 6-fold from 11 to 68 fmol/100 ng DNA, while cytosolic 4 S and nuclear receptor forms increased only 1.6- and 1.3-fold, respectively. This suggests that most of the exchangeable bound 17a-methyltestosterone occupies 10 S cytosolic receptor. This conclusion is supported by the results shown in Table 1 from cells incubated with 20 ^M 17«-methyltestosterone alone at 37 C, where 113 fmol/100 ng DNA or 87% of the receptor occupied after exchange was in the cytosolic 10 S form. Similarly, 86% of total occupied receptor was in the cytosolic 10 S form in control cells that received no steroid at 37 C. Furthermore, the 113 fmol/100 HQ DNA of 10 S receptor in the cells treated with 17a-methyltestosterone at 37 C is (113/185)100 = 6 1 % of the total glucocorticoid receptor identified in tosterone or with both steroids. Based on dissociation constants of 1 nM and 1 /XM respectively, these concentrations of steroid will each result in greater than 90% occupancy of receptor binding sites at equilibrium. After ligand exchange in intact cells, cytosol was prepared and analyzed in sucrose gradients (Fig. 2). In cells treated with [3H]triamcinolone acetonide alone, 10 S and 4 S cytosolic receptors were found in roughly equal abundance. In cells treated with both steroids, the amount of 10 S receptor was increased while the 9 16 - TREATMENT AT }?' - O--O 10nM[ 3 H]TA : 0 - - 0 NO STEROID • — • 10 nV [ 3 H ] TA + 20/iM MT -O-- » - • 2 0 / i M MT - 4 - n A / \ 10 ' " • • 15 20 25 10 15 20 25 FRACTION NUMBER Fig. 2. Effect of 17a-Methylestosterone (MT) on the Sedimentation of Cytosolic Glucocorticoid Receptors Monolayer cultures of GHT cells were incubated at 37 C for 1 h in steroid-depleted medium supplemented as indicated in the figure. The cells were then chilled to 0 C by floating the culture flasks in an ice-water bath and the medium was replaced two times with ice-cold serum-free medium without steroids. After each replacement of medium, the cells were incubated for 5 min at 0 C. Finally, the medium was replaced with ice-cold steroid-depleted medium containing 10 nM [3H] triamcinolone acetonide (TA) and incubation at 0 C was continued for 5 h. Cytosolic receptor forms were then analyzed in low salt sucrose gradients as described in Material and Methods. The direction of sedimentation is from right to left in the figure and only the bottom 25 of a total 39-40 fractions are shown. Table 1. Effects of 20 MM 17a-Methyltestosterone (MeTest) on the Intracellular Distribution of Glucocorticoid Receptors and Documentation of the Ligand Exchange Assay in Intact Cells Bound [3H]Triamcinolone Acetonide (fmol/100 M g DNA) Treatment No steroid 20 HM MeTest 10nM[ 3 H]TA 10 nM [3H]TA + 20 M M MeTest 10 nM [3H]TA + 20 /XM MeTest Total (% of control) Exchange 10S Cytosolic 4S Cytosolic Nuclear Total 160(86) 113(87) 24(12) 68 (53) 11 (20) 11 (6) 7 (5) 27(14) 11 (8) 7(12) 14 (8) 10 (8) 145(74) 50 (39) 38 (68) 185 130 196 129 56 100 70a 100 66' 29' The experimental details and the ligand exchange procedures are described in the legend to Fig. 2. Cells that did not undergo the exchange procedure were incubated at 0 C for the duration of that procedure in the same medium used for the 1 -h incubation at 37 C. Cytosolic 10 S and 4 S receptor forms were quantified by summing radioactivity in appropriate fractions of sucrose gradients. Specific binding of [3H]triamcinolone acetonide (TA) to washed nuclei was determined by subtracting binding observed in cells incubated with a 500-fold excess of the unlabeled steroid. Values in parentheses indicate 10 S or 4 S cytosolic receptor or nuclear receptor as a percentage of total receptor identified with [3H]triamcinolone acetonide for each experimental condition. a Relative to "No steroid" control. 6 Relative to "10 nM [3H]TA" control. 335 The Glucocorticoid Antagonist 17a-Methyltestosterone the control cells. In contrast, only (24/196)100 = 12% of total glucocorticoid receptor found in cells treated with 10 nM [3H]triamcinolone acetonide at 37 C was in the 10 S form. This suggests that 20 HM 17a-methyltestosterone binds to the 10 S receptor but does not effectively convert receptor to the cytosolic 4 S and 10 Me Test 10 15 20 nuclear forms. In addition to having more 10 S and less 4 S receptor, cells treated with both 17a-methyltestosterone and [3H]triamcinolone acetonide at 37 C have more total cytosolic receptor after ligand exchange than cells treated with the agonist alone. This suggests that 17a-methyltestosterone competes with [3H]triamcinolone acetonide for binding to the 10 S receptor form and as a consequence inhibits the dissociation of 10 S receptor and subsequent nuclear accumulation of the 4 S DNA-binding subunit which normally occurs in response to agonist binding. The competitive nature of this inhibition of agonistmediated 10 S cytosolic receptor dissociation by 17amethyltestosterone was confirmed by a dose-response experiment. GHT cells were incubated at 37 C for 1 h with 3 nM [3H]triamcinolone acetonide and various concentrations of the antagonist. After ligand exchange at 0 C in intact cells, cytosolic receptor forms were analyzed in sucrose gradients (Fig. 3). Cytosolic 10 S and 4 S receptor forms were quantified from the sucrose gradients and nuclear receptor was measured in washed nuclei (Table 2). With increasing concentrations of 17a-methyltestosterone the amount of 10 S cytosolic receptor steadily increases while cytosolic 4 S and nuclear receptor forms decrease. Effect of 17a-Methyltestosterone on Cytosolic and Nuclear Glucocorticoid Receptors Determined by Immunoblotting 25 Fraction Number Fig. 3. Sedimentation of Cytosolic Glucocorticoid Receptors from Cells Incubated with 3 PM [3H]Triamcinolone Acetonide and Various Concentrations of 17a-Methyltestosterone Monolayer cultures of GH: cells were incubated for 1 h at 37 C in steroid-depleted medium containing 3 nM [3H]triamcinolone acetonide and the indicated concentrations of 17«methyltestosterone (MeTest). After this incubation the flasks were placed in an ice-water bath and the medium was replaced two times with ice-cold serum-free medium without steroids. After each replacement of medium, the cells were incubated for 45 min at OC. Finally, the medium was replaced with steroid-depleted medium containing 15 nM [3H]triamcinolone acetonide and incubation at 0 C was continued for 4 h. Cytosolic receptor forms were then analyzed in sucrose gradients as described in Fig. 2. A reasonable explanation for the action of 17«-methyltestosterone as a glucocorticoid antagonist is that this compound binds to the glucocorticoid receptor but does not cause receptor to accumulate in the nucleus. However, two technical limitations of the ligand exchange assay in intact cells make it difficult to conclude with certainty that the antagonist does not convert a fraction of receptor to the 4 S cytosolic form or to the nuclear form. First, only about 70% of total receptor is generally identified after ligand exchange (Table 1). This failure to identify as much as 30% of the receptor population probably results wholly or partially from residual 17a-methyltestosterone competing with [3H] triamcinolone acetonide for receptor binding. In addi- Table 2. Dose dependence of the Effects of 17«-Methyltestosterone (MeTest) on the Intracellular Distribution of Glucocorticoid Receptors MeTest (MM) 0 0.33 1.0 3.3 10 20 Bound [3H]Triamcinolone Acetonide (fmol/00 M g DNA) 10S Cytosolic 4S Cytosolic Nuclear Total 24(16) 28 (20) 39 (26) 58(41) 68(51) 70 (54) 32 (22) 28 (20) 30 (20) 24(17) 21 (16) 20(15) 92 (62) 86 (60) 79 (54) 59 (42) 43 (33) 41 (31) 148 142 148 141 132 131 The experiment is described in the legend to Fig. 3. Cytosolic and nuclear receptor forms were quantified as described in the legend to Table 1. Values in parentheses indicate 10 S or 4 S cytosolic receptor or nuclear receptor as a percentage of total receptor identified with [3H]triamcinolone acetonide for each concentration of 17a-methyltestosterone. MOL ENDO-1989 336 tion, it is possible that cytosolic 4 S and nuclear receptor forms, if occupied with antagonist, either do not undergo ligand exchange or do so with considerably lower efficiency than cytosolic 10 S receptor. Second, the 4- to 6-h incubation required for intact cell ligand exchange might allow redistribution of receptor forms between nuclear and cytosolic compartments. To circumvent these limitations, an immunoblotting method was used t o determine if 17«-methyltestosterone converts a detectable amount of receptor to the nuclear form. GHi cells were incubated at 37 C for 1 h with 10 nM triamcinolone acetonide or 20 MM 17«-methyltestosterone or with both steroids. Control cells received no steroid. Cells were then lysed at 0 C and cytosolic and nuclear fractions were prepared immediately, thus minimizing the time available for possible redistribution of receptor forms between cellular compartments after cell lysis. The nuclear fraction was then extracted with salt using a procedure which solubilizes about 5 5 % of nuclear glucocorticoid receptors (14). The cytosol and nuclear extract were then incubated with monoclonal antibody BuGR2 followed by protein A-Sepharose to immunoadsorb glucocorticoid receptors. The adsorbed proteins were separated by electrophoresis and transferred to nitrocellulose. Glucocorticoid receptor bound to the nitrocellulose was detected by sequential incubations with monoclonal antibody BuGR2, rabbit anti-mouse immunoglobulin G (IgG), and 125 l-protein A followed by autoradiography (Fig. 4). The major glucocorticoid receptor band is positioned just below the 97 kilodalton (kDa) standard. The t w o minor bands between the major receptor band and the 66 kDa standard may correspond to receptor forms produced from alternate translation initiation sites within the receptor mRNA (16). The diffuse band seen at the bottom of each gel lane is monoclonal antibody BuGR2 from the immunoadsorption step. For control and 17a-methyltestosterone-treated cells, receptor was present in approximately equal abundance in the cytosolic fractions but was not detected in the nuclear fractions. By comparison, receptor in cells treated with triamcinolone acetonide was less abundant in the cytosolic fraction and was easily detected in the nuclear fraction. Cells treated with both steroids had increased cytosolic and decreased nuclear receptors compared with cells treated with agonist alone. These results indicate that a concentration of 17«-methyltestosterone that occupies more than 9 0 % of receptor binding sites causes no detectable accumulation of glucocorticoid receptors in the cell nucleus. Furthermore, when present together with agonist, the antagonist reduces the nuclear accumulation of receptor observed with agonist alone. Dense Amino Acid Labeling Studies A recent study from this laboratory used density labeling to document that [ 3 H]triamcinolone acetonide reduces the amount of glucocorticoid receptor in G H T Vol 3 No. 2 CYTOSOLIC NUCLEAR MT MT MT CON + TA MT CON + TA TA TA 97K — 66K — Fig. 4. Immunoblotting of Cytosolic and Nuclear Glucocorticoid Receptors GHi cells were incubated at 37 C for 1 h in steroid-depleted medium supplemented with 20 nm 17a-methylstestosterone (MT), 10 nM triamcinolone acetonide (TA), or with both steroids (MT + TA). Control cells (CON) received no steroid. Glucocorticoid receptors in cytosolic and nuclear fractions were immunoadsorbed using mouse monoclonal antibody BuGR2. After gel electrophoresis of the adsorbed protein and transfer to nitrocellulose, glucocorticoid receptor was detected by an immunoblotting procedure using sequential incubations with BuGR2, rabbit anti-mouse IgG, and 125l-protein A as described in Materials and Methods. The total immunoadsorbed receptor obtained from either 10 mg cytosolic protein or 2 mg nuclear protein was applied to each lane of the gel. Kodak XAR film was exposed at - 7 0 C for 4 h using a Cronex lightning-plus intensifying screen. Molecular weight standards were phosphorylase b (97K) and BSA (66K). cells by decreasing receptor half-life without affecting its rate of synthesis (14). The results from this study were consistent with the conclusion that binding of [3H] triamcinolone acetonide converted the 10 S receptor form to the cytosolic 4 S and nuclear receptor forms and that one or both of these latter forms was degraded at a faster rate in the cell than the unoccupied 10 S receptor (14). In view of these results, dense amino acid labeling was used to assess possible effects of 17«-methyltestosterone on the half-life of the glucocorticoid receptor. GHT cells were exposed to 20 nM 17amethyltestosterone or 10 nM [3H]triamcinolone acetonide for a total of 24 h. This period is sufficient to establish a new, lower steady state amount of receptor in the agonist-treated cells when compared to control cells cultured without steroids (14). Cells received dense amino acids during the final 10 h of this 24-h incubation. After harvesting the cells and preparing a nuclear extract, the density composition of the pooled cytosolic and nuclear receptor forms was analyzed in sucrose gradients containing 0.4 M KCI (Fig. 5). The sedimentation pattern of receptor from density labeled cells treated with 17a-methyltestosterone closely resembles that of control cells that did not receive steroid and is clearly different than the pattern in cells treated for 24 h with [3H]triamcinolone acetonide. A second density labeling experiment provided similar results. The results from the gradients shown in Fig. 5 were quantified and used to estimate values for the half-life 337 The Glucocorticoid Antagonist 17«-Methyltestosterone 10 15 20 25 15 20 25 10 15 20 25 Fraction Number Fig. 5. Sedimentation of Glucocorticoid Receptors after Dense Amino Acid Labeling in the Presence of 20 HM 17a-Methyltestosterone (MeTest) or 10 nM [3H]Triamcinolone Acetonide (TA) After incubation for 24 h in steroid-depleted medium, monolayer cultures of GH, cells were supplemented with the indicated steroid and incubation was continued for a total of 24 h. For the final 10 h of this incubation, the culture medium was replaced with dense amino acid medium also supplemented with the indicated steroid (•). Control cells were maintained in normal amino acid medium with the indicated steroid during this final 10-h period (O). Cells were then chilled to 4 C and the medium was replaced twice with 6 ml serum-free medium at 4 C. After each replacement of medium, cells were incubated for 45 min at 4 C. Finally, the cells were incubated for 30 min at 37 C in 5 ml steroid-depleted medium supplemented with 10 nM [3H]triamcinolone acetonide. Cells were harvested in lysis buffer and the lysate was adjusted to contain 0.4 M KCI and incubated at 4 C for 15 min to extract nuclear glucocorticoid receptors (14). After centrifugation for 10 min at 12,800 x g, the supernatant containing both cytosolic and extracted nuclear receptors was analyzed in sucrose gradients containing 0.4 M KCI as described in Materials and Methods. The direction of sedimentation is from right to left in the figure and only the bottom 26 of a total of 44 fractions are shown. The peaks of radioactivity centered around fractions 17-19 and 12-14 correspond to normal and dense receptors, respectively. and synthetic rate of the glucocorticoid receptor in the presence of the indicated steroids (Table 3). Consistent with previous results (14), cells incubated with [3H] triamcinolone acetonide for 24 h had only (49/92)100 = 53% of the receptor found in control cells. In contrast, cells treated with 17a-methyltestosterone had (81/ 92)100 = 88% of the control value, indicating that the antagonist caused little or no reduction in the amount of receptor. The value of 88% is a minimum estimate since ligand exchange is probably not complete, although the efficiency of exchange is higher in this experiment than in those described earlier due to the use of a 37 C incubation during exchange. The half-life estimates from the 10-h density labeling period indicate that 17a-methyltestosterone does not shorten the halflife of the glucocorticoid receptor compared with control cells incubated without steroid. In contrast, [3H]triamcinolone acetonide significantly reduces receptor halflife. The values for synthetic rates shown in Table 3 indicate only small differences for the three experimental conditions. The calculated value for 17a-methyltestosterone-treated cells underestimates the synthetic rate since it has not been corrected for the efficiency of ligand exchange. The observation that the antagonist 17a-methyltestosterone does not reduce receptor halflife is consistent with the conclusion that only agonists promote dissociation of the 10 S receptor and that the 4 S receptor form is degraded in the intact cell at a faster rate than the oligomer (14). DISCUSSION The interactions of the antagonist 17a-methyltestosterone with the glucocorticoid receptor differ from those of the agonist triamcinolone acetonide. First, the two steroids appear to compete for binding to the same site (or sites) in the 10 S cytosolic receptor, but the affinity for the antagonist is at least 1000-fold lower than that for agonist (Fig. 1). Second, based on exchange assay results in intact GH, cells, the antagonist does not appear to promote dissociation of the 10 S oligomer to the 4 S subunit and, by competing with agonist for the ligand binding site, inhibits oligomer dissociation mediated by the binding of agonist (Figs. 2 and 3). Third, based on both exchange assay results and immunochemical detection, the antagonist does not cause substantial accumulation of glucocorticoid receptors in the nuclear fraction and inhibits nuclear accumulation of receptor promoted by the agonist (Tables 1 and 2, Fig. 4). Finally, while the agonist significantly shortens glucocorticoid receptor half-life, the antagonist does not (Fig. 5, Table 3). These observations are consistent with a model (3, 4) for receptor transformation in which the unoccupied receptor exists in intact cells primarily as a 10 S oligomeric protein that is found in the soluble (cytosolic) fraction of broken cell preparations. When the steroid binding site is occupied with an agonist, the distribution between the oligomer and its 4 S hormone- and DNAbinding subunit shifts toward the subunit. Experimental evidence supporting this model came originally from dense amino acid labeling studies and from studies of the dose-dependent effects of a glucocorticoid agonist on the intracellular amounts of 10 S and 4 S cytosolic receptors and nuclear receptors in GH, cells (3). Furthermore, sodium molybdate stabilizes oligomeric steroid receptor forms in vitro (17) and was found to inhibit the agonist-mediated nuclear accumulation of gluco- MOL ENDO-1989 338 Vol 3 No. 2 Table 3. Effects of 17a-Methyltestosterone (MeTest) or [3 H]Triamcinolone Acetonide (TA) on the Half-Life and Synthetic Rate of the Glucocorticoid Receptor Condition No steroid 20 MM MeTest 10nM[ 3 H]TA Total Receptor (fmol/100MgDNA) Dense Receptor (% of total) t* (h) (fmol/h-IOOMgDNA) 92 81 34 31 16.9 18.8 3.8 3.0 49 48 10.6 3.2 Amounts of total receptor and dense receptor were calculated by summing the radioactivity in appropriate fractions of the sucrose have been found to be in good agreement with more thorough time course studies (14). The receptor synthetic rate (ks) was density receptor remaining in cells after a time (t) of incubation with dense amino acids, R,=o is the total amount of dense and normal receptor found in the gradients, and K, = 0.693/tV2 (38). Since the estimation of t,/2 is dependent only on the relative amounts of dense and normal receptor forms found in the gradients, measurements based on single times of dense amino acid labeling have been found to be in good agreement with more thorough time course studies (14). The receptor synthetic rate (ks) was estimated from the formula ks = kdRDt/1 -e~kdt, where RD,, is the amount of dense receptor found in the gradients after a time (t) of incubation of cells with dense amino acids (38). corticoid receptors in intact GH, cells by a mechanism which appears to involve stabilization of intracellular receptor oligomers (4). The results presented here suggest that the glucocorticoid antagonist 17a-methyltestosterone binds to the oligomeric 10 S receptor but does not appear to promote dissociation of the 4 S DNA-binding subunit from the oligomer. Consequently the antagonist does not cause substantial accumulation of receptor in the cell nucleus. By competitively inhibiting the binding of agonist to the 10 S oligomeric receptor, the antagonist effectively inhibits the subunit dissociation mediated by agonist binding that is essential for the nuclear accumulation and biological actions of the glucocorticoid receptor. The ability of the antagonist to block the actions of an agonist therefore depends on both the concentrations and the relative binding affinities of the two ligands. While making direct binding studies using [3H]17amethyltestosterone impractical, the low affinity of the antagonist for the glucocorticoid receptor was essential for the ligand exchange assay at 0 C used in this study. Even at this low temperature, the antagonist dissociates from the receptor at a rate sufficiently rapid to allow significant exchange with the high affinity agonist [3H]triamcinolone acetonide. However, at 0 C the agonist does not promote the subunit dissociation and nuclear accumulation that normally occurs rapidly at 37 C (3, 4). For this reason, it is assumed throughout this study that the relative abundance and intracellular location of receptor forms does not change during the ligand exchange procedure at OC. The absence of immunochemically detectable receptor in the nucleus after rapid lysis and fractionation of cells treated at 37 C with 17a-methyltestosterone (Fig. 4) appears to confirm the conclusion from ligand exchange studies that the antagonist does not cause significant nuclear accumulation of receptor in intact cells. The conclusion that 17a-methyltestosterone and other glucocorticoid antagonists which bind reversibly to the receptor do not cause nuclear accumulation of receptor is consistent with results from several direct binding studies using 3H-labeled steroids. First, proges- terone was found to be an effective glucocorticoid antagonist in HTC cells (10). [3H]Progesterone did not lower the amount of cytosolic receptor in HTC cells (18), and did not cause nuclear accumulation of receptor in mouse mammary tumor cells (19). Since [^progesterone elicits high levels of nonspecific binding in both cytosolic and nuclear fractions from GHT cells (Raaka, B. M. and H. H. Samuels, unpublished results), it was not possible to use this steroid in the present study to determine effects of an antagonist on 10 S and 4 S receptor forms by direct binding assays. Second, two groups concluded independently that [3H]cortexolone (11-deoxycortisol) did cause nuclear accumulation of glucocorticoid receptors in rat thymus cells (20, 21). Because of the low affinity of receptor for this compound (15) and high nonspecific binding (20), these studies did not provide quantitative data concerning the fraction of total cellular glucocorticoid receptor that accumulated in nuclei, and the available evidence suggests that the fraction was small. Since cortexolone was shown to be a partial agonist in HTC cells (10), it is reasonable to expect a fractional nuclear accumulation of receptor in proportion to the agonist activity of this steroid. Finally, the synthetic glucocorticoid antagonist [3H]RU38486 has high affinity for the glucocorticoid receptor and causes little or no nuclear accumulation of receptor in rat thymus cells (22) or hepatoma cell lines (23). Although several reports (24, 25) have indicated that [3H]RU38486 accumulates in nuclei, it has not clearly been shown that the labeled steroid occupied glucocorticoid receptors. Recent studies have shown that [3H]RU38486 does not promote dissociation of glucocorticoid receptor oligomers under conditions in which oligomer dissociation is promoted by agonist [3H]triamcinolone acetonide (11,12). The subunit structure and intracellular location of the 10 S oligomeric glucocorticoid receptor are not known with certainty. Recent studies have indicated that oligomeric forms of steroid receptors contain a common 90 kDa heat shock protein that does not bind steroid (26-28) and which dissociates from the hormone-binding subunit upon ligand-mediated transformation (9, 29). Evidence supporting a trimeric structure consisting The Glucocorticoid Antagonist 17a-Methyltestosterone of a single hormone-binding subunit and two 90 kDa subunits has been presented for both glucocorticoid (30) and progesterone (26) receptors. Controversy has arisen concerning the intracellular location of untransformed steroid receptors. While immunochemical (31) and rapid cell enucleation (32) studies have suggested that the unliganded estrogen receptor may be located in the cell nucleus, other immunochemical studies have supported a cytoplasmic location for the unliganded glucocorticoid receptor (33, 34). The possible nuclear localization of certain unliganded steroid receptors does not negate the concept of ligand-mediated receptor transformation but indicates that this process may occur in the nucleus rather than in the cytoplasm. While direct evidence of the intracellular existence of steroid receptor oligomers has not been obtained to date, evidence from a variety of experimental systems (3-9) supports the conclusion that transformation of steroid receptors in situ involves the hormone-mediated dissociation of a receptor oligomer to its hormone- and DNA-binding subunit. In particular, a recent study has shown that certain glucocorticoid receptor deletion mutants which act as constitutive activators of transcription (35, 36) are unable to form oligomeric receptor forms (37). This suggests that intracellular steroid receptor oligomers may serve to mask the DNA binding and transcription-enhancing domains of the glucocorticoid receptor. By competing with agonists for binding to an intracellular receptor oligomer, glucocorticoid antagonists would effectively block the agonist-mediated oligomer dissociation required to unmask functional domains of the receptor. 339 resulting particulate fraction, referred to as washed nuclei, was incubated with 0.25 ml of ethanol for 30 min at 37 C to extract [3H]triamcinolone acetonide. Specific binding to receptor proteins was determined by comparing radioactivity in nuclear fractions derived from cells treated with the labeled steroid and with or without a 500-fold excess of the unlabeled steroid. Radioactivity in ethanol extracts was determined by scintillation counting at 43% efficiency. The DNA content of the nuclear fraction was determined by the method of Burton (39). Sucrose Gradient Centrifugation For resolution of 10 S and 4 S cytosolic receptor forms, discontinuous gradients were made of 1.3 ml each of 30%, 20%, and 10% sucrose solutions (wt/vol) prepared in buffer, pH 7.4 at 25 C, containing 20 M Tris-HCI, 20 M sodium molybdate, and 2 mM dithiothreitol. After centrifugation for 16 h at 58,000 rpm and 4 C in an SW 60 Ti rotor, 39 or 40 fractions were collected from each gradient. Using these conditions, the molybdate-stabilized 10 S receptor oligomer does not dissociate during sedimentation (3). Peak positions for reference proteins /?-amylase, 9.4 S (40), and BSA, 4.4 S (41), were fractions 8-9 and 19-20, respectively. Throughout the text the term "4 S" is used to describe a hormone- and DNAbinding form of the glucocorticoid receptor that sediments at about 3.5 S under high salt conditions and in the region of 4 to 6 S under low salt conditions (3). The variable sedimentation rate under low salt conditions probably results from reversible, concentration-dependent interactions of the receptor protein with itself or other proteins during sedimentation (3). For dense amino acid labeling studies, discontinuous gradients consisted of 30%, 22.5%, and 15% sucrose solutions (wt/vol) prepared in 95% D2O, 5% H2O and containing 20 mM Tris-HCI, pH 7.4 at 25 C, 2 mM dithiothreitol, and 0.4 M KCI (3, 14). Centrifugation was for 45 h at 5 C and 58,000 rpm in the SW 60 Ti rotor. Immunoblotting MATERIALS AND METHODS Cell Culture Conditions GHi cells were grown in monolayer cultures at 37 C in 75-cm2 plastic flasks as described previously (3). Before each experiment, cells were cultured for 24 h in Ham's F-10 medium supplemented to 10% (vol/vol) with calf serum that had been treated with AG 1X-8 anion exchange resin and activated charcoal to remove steroid and thyroid hormones (13). This medium is designated steroid-depleted medium. Culture medium containing dense amino acids was prepared as described previously (38). Cell Harvesting and Analysis of Receptor Forms After removing the culture medium, cells were washed three times with isotonic saline at 4 C and drained thoroughly. The washed cells were scraped off the culture flasks at 4 C in 0.25 ml buffer, pH 7.4 at 25 C, composed of 20 miui Tris-HCI, 5 mM sodium molybdate, 2 mM dithiothreitol, and 0.02% Triton X100 (lysis buffer). After mixing for 5 s at the maximum setting of a Vortex Genie, the soluble and paniculate fractions were separated by centrifugation for 10 min at 12,800 x g in an Eppendorf model 5412 microcentrifuge. The supernatant from this centrifugation is referred to as cytosol. For analysis and quantitation of glucocorticoid receptors from cells previously incubated with [3H]triamcinolone acetonide, cytosol was applied directly to sucrose gradients as described below. The particulate fraction from the 12,800 x g centrifugation was washed three times by resuspending in 1 ml of lysis buffer followed by centrifugation as before. The Cytosol and washed nuclei were prepared from GHi cells as described above. A soluble nuclear extract was prepared by resuspending the washed nuclei for 30 min at 4 C in 0.5 ml lysis buffer containing 0.4 M KCI followed by centrifugation at 12,800 x g for 10 min. The procedure generally solubilizes about 55% of receptor-bound [3H]triamcinolone acetonide from GHT cell nuclei (14). The cytosol and nuclear extract were adjusted to final concentrations of 200 mM KCI and 10 nM triamcinolone acetonide. For immunoprecipitation, 0.01 ml ascites fluid containing monoclonal antibody BuGR2 (42) was added to 1 ml adjusted cytosol or nuclear extract containing approximately 10 or 2 mg protein, respectively. After incubation at 0 C for 16 h, 0.1 ml (wt/vol) suspension of protein Alinked Sepharose CL-4B in lysis buffer was added and the resulting suspension was incubated at 4 C for 3 h with gentle mixing by rotation. The protein A-linked Sepharose was then sedimented by centrifugation at 12,800 x g for 10 min and washed twice at 4 C using 1 ml lysis buffer. After extensive washing at room temperature first using 50 mM Tris-HCI, pH 7.5, 5 mM EDTA, 0.75 M NaCI, and 0.05% NP-40 followed by 50 mM Tris-HCI, pH 7.5, 5 mM EDTA, and 0.2 M NaCI, the protein A-linked Sepharose was resuspended in 0.075 ml 10 mM sodium phosphate, pH 7.0, 1 % 2-mercaptoethanol, and 1 % sodium dodecyl sulfate and incubated at 37 C for 30 min to release adsorbed proteins. Immunoadsorbed proteins were analyzed by electrophoresis in a 7.5% polyacrylamide gel in the presence of sodium dodecyl sulfate (43). Proteins were then transferred by electroelution at 0.75 A for 3 h at room temperature. Transfer buffer was 25 mM Tris base, 192 mM glycine, and 20% (vol/ vol) methanol (44). After transfer, the nitrocellulose filter was incubated for 15 min in an aqueous solution containing 10% (vol/vol) acetic acid and 10% (vol/vol) isopropanol. Unless Vol 3 No. 2 MOL ENDO-1989 340 otherwise noted, this and all subsequent incubations were performed at room temperature with gentle shaking on a rotating platform. After rinsing briefly with water, the filter was incubated for 10 min with buffer consisting of 20 mM Tris-HCI, pH 7.5, and 200 mM NaCI (TBS) followed by a 60-min incubation in milk solution consisting of 20 mM Tris-HCI, pH 7.5,1 mM EDTA, 100 mM NaCI, 0.005% Antifoam A, and 5% (wt/ vol) Carnation nonfat dry milk (45). Ascites fluid containing BuGR2 was then added to the milk solution at a dilution of 1:1000 and incubation was continued for 16 h at 4 C. After incubation with BuGR2, the filter was washed for 15 min with TBS containing 0.05% Tween-20 and then for 15 min in TBS. After a 60-min incubation in milk solution, purified rabbit anti-mouse IgG was added at a 1:2000 dilution and incubation was continued for 2 h. After 15-min washes first in TBS containing 1 M NaCI and 0.05% Tween-20 and then without Tween-20, the filter was incubated for 2 h in 50 ml of milk solution containing 125l-Protein A at 150,000 cpm/ml. After three 15-min washes in milk solution containing 1 M NaCI and 0.05% Tween-20, the filter was rinsed briefly in water prior to autoradiography. Chemicals and Media [6,7-3H]Triamcinolone acetonide (44 Ci/mmol) was obtained from New England Nuclear (Boston, MA). 125l-Protein A (30 mCi/mg) was from Amersham (Arlington Heights, IL) and rabbit anti-mouse IgG was from Zymed Laboratories (South San Francisco, CA). Steroids, protein A-Sepharose CL-4B, protamine sulfate, sodium molybdate, D2O, and other chemicals were obtained from Sigma (St. Louis, MO). Cell culture media and serum were supplied by Grand Island Biological Co. (Grand Island, NY). Amino acids enriched in the dense isotopes 2 H, 13 C, and 15N to 98%, 80%, and 90% atom, respectively, was obtained from Merck, Montreal, Qubec, Canada. Monoclonal antibody BuGR2 which recognizes the hormone-binding subunit of glucocorticoid receptors (42) was generously provided by Dr. Robert W. Harrison (University of Arkansas Medical School, Little Rock, AR). Acknowledgments Received April 27,1988. Revision received October 31,1988. Accepted November 11,1988. Address requests for reprints to: Dr. Bruce M. Raaka, Department of Medicine, UH Room 451, New York University Medical Center, 550 First Avenue, New York, New York 10016. This research was supported by Grant AM-21566 from the NIH and Grant BC-550 from the American Cancer Society. REFERENCES 1. O'Malley BW, Schwartz RJ, Schrader WT 1976 A review of regulation of gene expression by steroid hormone receptors. J Steroid Biochem 7:1151 -1159 2. Chandler VL, Maler BA, Yamamoto KR 1983 DNA sequences bound specifically by glucocorticoid receptor in vitro render a heterologous promoter hormone responsive in vivo. Cell 33:489-499 3. Raaka BM, Samuels HH 1983 The glucocorticoid receptor in GH, cells. Evidence from dense amino acid labeling and whole cell studies for an equilibrium model explaining the influence of hormone on the intracellular distribution of receptor. J Biol Chem 258:417-425 4. Raaka BM, Finnerty M, Sun E, Samuels HH 1985 Effects of molybdate on steroid receptors in intact GH-, cells. Evidence for dissociation of an intracellular 10 S receptor oligomer prior to nuclear accumulation. J Biol Chem 260:14009-14015 5. Holbrook NJ, Bodwell JE, Jeffries M, Munck A 1983 Characterization of nonactivated and activated glucocorticoid-receptor complexes from intact rat thymus cells. J Biol Chem 258:6477-6485 6. Vedeckis WV 1983 Subunit dissociation as a possible mechanism of glucocorticoid receptor activation. Biochemistry 22:1983-1989 7. Sherman MR, Moran MC, Tuazon FB, Stevens YW 1983 Structure, dissociation, and proteolysis of mammalian steroid receptors. Multiplicity of glucocorticoid receptor forms and proteolytic enzymes in rat liver and kidney cytosols. J Biol Chem 258:10366-10377 8. Muller RE, Traish AM, Wotiz HH 1983 Estrogen receptor activation precedes transformation. J Biol Chem 258:9227-9236 9. Sanchez ER, Meshinchi S, Tienrungroj W, Schlesinger MJ, Toft DO, Pratt WB 1987 Relationship of the 90-kDa Murine heat shock protein to the untransformed and the transformed states of the L cell glucocorticoid receptor. J Biol Chem 262:6986-6991 10. Samuels HH, Tomkins GM 1970 Relation of steroid structure to enzyme induction in hepatoma tissue culture cells. J Mol Biol 52:57-74 11. Sablonniere B, Danze PM, Formstecher P, Lefebvre P, Dautrevaux M 1986 Physcial characterization of the activated and non-activated forms of the glucocorticoid-receptor complex bound to the steroid antagonist [3H]RU 486. J Steroid Biochem 25:605-614 12. Groyer A, Schweizer-Groyer G, Cadepond F, Manlier M, Baulieu E-E 1987 Antiglucocorticosteroid effects suggest why steroid hormone is required for receptors to bind DNA in vivo but not in vitro. Nature 328:624-626 13. Samuels HH, Stanley F, Shapiro LE 1979 Control of growth hormone synthesis in cultured GH, cells by 3,5,3'triiodo-L-thyronine and glucocorticoid agonists and antagonists: studies on the independent and synergistic regulation of the growth hormone response. Biochemistry 18:715-721 14. Mclntyre WR, Samuels HH 1985 Triamcinolone acetonide regulates glucocorticoid-receptor levels by decreasing the half-life of the activated nuclear-receptor form. J Biol Chem 260:418-427 15. Rousseau GG, Schmit J-P 1977 Structure-activity relationships for glucocorticoids-1: Determination of receptor binding and biological activity. J Steroid Biochem 8:911 919 16. Miesfeld R, Rusconi S, Godowski PJ, Maler BA, Okret S, Wikstrom A-C, Gustafsson J-A, Yamamoto KR 1986 Genetic complementation of a glucocorticoid receptor deficiency by expression of cloned receptor cDNA. Cell 46:389-399 17. Dahmer MK, Housley PR, Pratt WB 1984 Effects of molybdate and endogenous inhibitors on steroid-receptor inactivation, transformation, and translocation. Annu Rev Physiol 46:67-81 18. Rousseau GG, Baxter JD, Higgins SJ, Tompkins GM 1973 Steroid-induced nuclear binding of glucocorticoid receptors in intact hepatoma cells. J Mol Biol 79:539-554 19. Shyamala G 1975 Glucocorticoid receptors in mouse mammary tumors: specific binding to nuclear components. Biochemistry 14:437-444 20. Turnell RW, Kaiser N, Milholland RJ, Rosen F 1974 Glucocorticoid receptors in rat thymocytes. Interactions with the antiglucocorticoid cortexolone and mechanism of its action. J Biol Chem 249:1133-1138 21. Wira CR, Munck A 1974 Glucocorticoid-receptor complexes in rat thymus cells. "Cytoplasmic"-nuclear transformations. J Biol Chem 249:5328-5336 22. Moguilewsky M, Philibert D 1984 RU 38486: potent antiglucocorticoid activity correlated with strong binding to the cytosolic glucocorticoid receptor followed by an impaired activation. J Steroid Biochem 20:271-276 23. Chasserot-Golaz S, Beck G 1984 An approach to the The Glucocorticoid Antagonist 17«-Methyltestosterone 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. mechanism of the potent antiglucocorticoid: 17beta-hydr oxy-11 beta-4-dimethylaminophenyl-17alpha-propynylestra-4,9-dien-3-one. J Steroid Biochem 21:585-591 Bourgeois S, Pfahl M, Baulieu E-E 1984 DNA binding properties of glucocorticosteroid receptors bound to the steroid antagonist RU-486. EMBO J 3:751-755 Johnson GS, Ralhan R 1986 Glucocorticoid agonists as well as antagonists are effective inducers of mouse mammary tumor virus RNA in mouse mammary tumor cells treated with inhibitors of ADP-ribosylation. J Cell Physiol 129:36-42 Renoir JM, Buchou T, Mester J, Radanyi C, Baulieu E-E 1984 Oligomeric structure of molybdate-stabilized, nontransformed 8S progesterone receptor from chicken oviduct cytosol. Biochemistry 23:6016-6023 Joab I, Radanyi C, Renoir M, Bochou T, Catelli MG, Binart N, Mester J, Baulieu E-E 1984 Common non-hormone binding component in nontransformed chick oviduct receptors for four steroid hormones. Nature 308:850-853 Riehl RM, Sullivan WP, Vroman BT, Bauer VJ, Pearson GR, Toft DO 1985 Immunological evidence that the nonhormone binding component of avian steroid receptors exists in a wide range of tissues and species. Biochemistry 24:6586-6591 Mendel DB, Bodwell JE, Gametchu B, Harrison RW, Munck A 1986 Molybdate-stabilized nonactivated glucocorticoid-receptor complexes contain a 90-kDa non-steroid-binding phosphoprotein that is lost on activation. J BiolChem 261:3758-3763 Denis M, Wikstrom A-C, Gustafsson J-A1987 Molybdatestabilized nonactivated glucocorticoid receptor contains a dimer of Mr 90,000 non-hormone-binding protein. J Biol Chem 262:11803-11806 King WJ, Greene GL 1984 Monoclonal antibodies localize oestrogen receptor in the nuclei of target cells. Nature 307:745-747 Welshons WV, Lieberman ME, Gorski J 1984 Nuclear localization of unoccupied oestrogen receptors. Nature 307:747-749 Antakly T, Eisen HJ 1984 Immunocytochemical localization of glucocorticoid receptor in target cells. Endocrinology 115:1984-1989 Fuxe K, Wikstrom A-C, Okret S, Agnati LF, Harfstrand A, Yu Z-Y, Granholm L, Zoli M, Vale W, Gustafsson J-A 1985 Mapping of glucocorticoid receptor immunoreactive neurons in the rat tel- and di-encephalon using a monoclonal antibody against rat liver glucocorticoid receptor. 341 Endocrinology 117:1803-1812 35. Godowski PJ, Rusconi S, Miesfeld R, Yamamoto KR1987 Glucocorticoid receptor mutants that are constitutive activators of transcriptional enhancement. Nature 325:365368 36. Hollenberg SM, Giguere V, Segui P, Evans RM 1987 Colocalization of DNA-binding and transcriptional activation functions in the human glucocorticoid receptor. Cell 49:39-46 37. Pratt WB, Jolly DJ, Pratt DV, Hollenberg SM, Giguere V, Cadepond FM, Schweizer- Groyer G, Catelli M-G, Evans RM, Baulieu E-E 1988 A region of the steroid binding domain determines formation of the non-DNA-binding, 9 S glucocorticoid receptor complex. J Biol Chem 263:267273 38. Raaka BM, Samuels HH 1981 Regulation of thyroid hormone nuclear receptor levels in GH, cells by 3,5,3'-triiodoL-thyronine. Use of dense amino acid labeling to determine the influence of hormone on the receptor half-life and the rate of appearance of newly synthesized receptor. J Biol Chem 256:6883-6889 39. Burton K 1956 A study of the conditions and mechanism of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem J 62:315-323 40. Niu E-M, Neal RM, Pierce VK, Sherman MR 1981 Structural similarity of molybdate-stabilized steroid receptors in human breast tumors, uteri and leukocytes. J Steroid Biochem 15:1-10 41. Castellino FJ, Barker B 1968 Examination of the dissociation of multichain proteins in guanidine hydrochloride by membrane osmometry. Biochemistry 7:2207-2217 42. Hendry WJ, Danzo BJ, Harrison RW 1987 Analysis of the disruptive action of an epididymal protease and the stabilizing influence of molybdate on nondenatured and denatured glucocorticoid receptor. Endocrinology 120:629639 43. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685 44. Towbin H, Staehelin T, Gordon J 1979 Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76:4350-4354 45. Johnson DA, Gautsch JW, Sportsman JR, Elder JH 1984 Improved technique utilizing nonfat dry milk for analysis of proteins and nucleic acids transferred to nitrocellulose. Gene Anal Techn 1:3-8