Biochemistry: Miller-Diener et al. gradient. The activated peak eluting at 50 ..... contrast to the results by Kurl and Jacob (22) showing receptor phosphorylation ...
Proc. Natl. Acad. Sci. USA Vol. 82, pp. 4003-4007, June 1985 Biochemistry
Protein kinase activity associated with the purified rat hepatic glucocorticoid receptor (phosphorylation/steroid rebinding/ATP binding site/phospho amino acids)
ANDREA MILLER-DIENER, THOMAS J. SCHMIDT*,
AND GERALD
LITWACK
Fels Research Institute and Department of Biochemistry, Temple University School of Medicine, Philadelphia, PA 19140
Communicated by Sidney Weinhouse, February 21, 1985
ABSTRACT The M, 94,000 steroid binding component of rat hepatic glucocorticoid receptor purified 5000-fold undergoes calcium-stimulated phosphorylation in vitro by [y-32P]ATP. Exogenous histones can be phosphorylated by this preparation without calcium. Calmodulin did not stimulate phosphorylation of the glucocorticoid receptor beyond that obtained with calcium alone. Although the specific calmodulin inhibitor calmidazolium had no effect, trifluoperazine and chlorpromazine, nonspecific calmodulin inhibitors, abolished the calcium-dependent phosphorylation of receptor. EGTA blocks the effect of calcium; magnesium cannot substitute for calcium. Cyclic nucleotides (cAMP or cGMP) do not stimulate phosphorylation of the receptor in the absence of calcium. Phosphorylation of the glucocorticoid receptor is steroid dependent. Triamcinolone acetonide elicited activation and phosphorylation of receptor in the presence of calcium, whereas the antagonists progesterone, cortexolone, and .8-lapachone did not. Sodium molybdate, which blocks the thermal activation step, inhibits phosphorylation of the receptor. The activated form of the glucocorticoid receptor is required for phosphorylation to occur. The ATP analogues 8-azido-ATP or fluorosulfonylbenzoyl adenosine, inhibit phosphorylation of the Mr 94,000 component, implying the presence of an ATP binding site inherent to the receptor.
Advances in the purification of the unactivated glucocorticoid receptor to near homogeneity facilitate investigation of the physicochemical characteristics of the complex (18) and study of in vitro regulation of activation (19). In vivo phosphorylation of the glucocorticoid receptor has been demonstrated in our laboratory (20). Although we first described the kinase activity of purified receptor (21), a recent report describes the glucocorticoid receptor as possessing endogenous kinase activity (22) with a requirement for magnesium. Conditions regulating the protein kinase activity of purified glucocorticoid receptor as well as the requirement for activation of the purified complex and steroid dependence of
phosphorylation are reported.
EXPERIMENTAL PROCEDURES Materials. All chemicals were reagent grade and used without further purification. [6,7,-3H(N)]Triamcinolone acetonide (44 Ci/mmol; 1 Ci = 37 GBq), [6,7,-3H(N)]dexamethasone mesylate (40 Ci/mmol), 17a-[1,2,-3H(N)]hydroxy-11-deoxycorticosterone(40-6OCi/mmol), [1,2,-3H(N)] progesterone (90-115 Ci/mmol), and [y-32P]adenosine 5'triphosphate, tetra(triethylammonium) salt, aqueous solution (2900 Ci/mmol) were obtained from New England Nuclear. Trifluoperazine, chlorpromazine, 8-azidoadenosine 5'triphosphate, 5'-p-fluorosulfonylbenzoyl adenosine HCl, and phosphodiesterase 3':5'-cyclic nucleotide activator (calmodulin) were purchased from Sigma. Calmidazolium and histone (mixture of H1, H2A, H2B, H3, and H4) were obtained from Boehringer Mannheim. Unlabeled protein standards were purchased from Pharmacia; 14C-labeled protein standards were from New England Nuclear. Reagents for NaDodSO4/polyacrylamide gel electrophoresis were from Bio-Rad. Liquiscint and Autofluor were obtained from National Diagnostics (Somerville, NJ). All other reagents were purchased from Fisher.
Phosphorylation/dephosphorylation reactions may be inherent to the mechanism of steroid hormone action (1-4). Treatment of crude estrogen (5) and glucocorticoid (6) receptors with phosphatase decreases hormone binding. Nonspecific phosphatase inhibitors, such as molybdate and fluoride (7, 8), prevent loss of steroid binding. The rate of activation of the glucocorticoid receptor and the appearance of a more positively charged molecular species with increased affinity for nuclei and polyanions are stimulated by addition of calf intestinal alkaline phosphatase to rat liver cytosol (9). It is difficult to determine in crude cytosols whether these effects are due to phosphorylation/dephosphorylation of the receptor itself or of some other cytosolic factor(s). However, in vivo phosphorylation of both the progesterone (10) and glucocorticoid (11) receptors has since been demonstrated. In vitro phosphorylation of several purified and partially purified receptors has now been reported. Receptors also possess enzymatic activity. For example, the purified insulin receptor expresses protein tyrosine kinase activity (12, 13). In addition, ligand-stimulated/ligand-dependent receptor phosphorylation is observed in intact cells (14, 15) as well as with partially purified insulin receptors (15, 16). The ATP binding site and insulinstimulated phosphorylation site coexist on the 83-subunit of the insulin receptor (17).
Buffers. Distilled deionized water was used to prepare all buffers. Buffer A is 10 mM 2-(N-morpholino)ethanesulfonic acid/0.1 mM dithiothreitol/10% (vol/vol) glycerol, pH 6.8 at O°C.
Purification of the Glucocorticoid Receptor. Cytosol and purified molybdate-stabilized glucocorticoid receptor were prepared as described (18). Unactivated receptor, which elutes from DEAE-cellulose at ==250 mM potassium phosphate, can be stored for future use at -70°C after elution from Sephadex G-75 with buffer A containing 10 mM Na2MoO4. Phosphorylation Incubation System. Immediately before use, DEAE-cellulose-purified unactivated glucocorticoid receptor, fresh or frozen, was eluted from Sephadex G-75 with buffer A to remove Na2MoO4 and was concentrated 5- to 10-fold by centrifugation on Centricon-30 filters. Activated receptor was prepared by heating an aliquot at 25°C for 30 min and loaded onto DEAE-cellulose followed by elution with a 5-500 mM KCl
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
*Present address: Department of Physiology and Biophysics, University of Iowa, College of Medicine, Iowa City, IA 52242.
4003
4004
Biochemistry: Miller-Diener et al.
gradient. The activated peak eluting at 50 mM KCI was used in subsequent phosphorylation experiments. Typically, 100 ,u1 of the Sephadex G-75 concentrated purified unactivated receptor or activated receptor from DEAE-cellulose was incubated with [y-32PIATP in the amounts indicated for 30 min at 250C unless stated otherwise. The reaction was terminated by addition of 50 gI of Laemmli sample buffer (23) and boiling for 5 min at 100'C or by trichloroacetic acid precipitation. NaDodSo4/Polyacrylamide Gel Electrophoresis. Samples were electrophoresed on duplicate 1.5-mm 10% NaDodS04/polyacrylamide slab gels (24) and stained with Coomassie brilliant blue R250. After destaining, one gel was processed for autoradiography with Autofluor. No intensifying screens were used. The duplicate gel was further stained with silver (25). 14C protein standards were phosphorylase b (Mr, 97,000), albumin (Mr, 69,000), ovalbumin (Mr, 46,000), carbonic anhydrase (Mr, 30,000), lactoglobulin (Mr, 18,360), and cytochrome c (Mr, 12,300). DNA Cellulose Binding Assay. Percentage activation of the purified receptor was determined by binding to DNA cellulose (26). Hydroxylapatite Assay. Specific steroid binding in cytosol or in purified samples was measured by the hydroxylapatite technique (27). Radioactivity was measured by using an Intertechnique LS 30 liquid scintillation spectrometer (35% efficiency for 3H). Phospho Amino Acid Analysis. Approximately 25 pmol of purified activated receptor (4 ml) was incubated with 0.4 nmol of [_y-32P]ATP and 100 ,uM CaCl2 for 30 min at 25°C. The reaction was terminated by addition of 100% trichloroacetic acid (final concentration, 10%). The pellet was washed four times with 2 ml of 10% trichloroacetic acid and twice with 2 ml of ether/acetone (1:1). The dried protein mixture was mixed with 0.5 ml of 4 M HCl and the solution was degassed. The sample was hydrolyzed in sealed tubes at 110°C for 3 hr, evaporated to dryness, and reconstituted in 100 ,ul of HPLC buffer (10 mM KP/12.5% methanol, final pH 3.0). Phospho amino acids were separated (28) on a Partisil-10 SAX exchange column (Whatman). Sample volume injected was 75 ,u1. The column was run at room temperature (flow rate, 1 ml/min). Phosphothreonine, phosphotyrosine, and phosphoserine (Sigma) were run as standards. Phospho amino acids were detected by ninhydrin (29). One-milliliter fractions were collected; 0.5-ml aliquots were assayed for 32p; 0.3-ml aliquots were treated with ninhydrin to detect amino acids. RESULTS Effect of Exogenous Reagents on Kinase Activity of the Purified Glucocorticoid Receptor. Since several receptors of polypeptide hormones exhibit endogenous kinase activity (12, 30) and since in vivo phosphorylation of the glucocorticoid receptor has recently been demonstrated by workers in our laboratory (20), we investigated kinase activity of the highly purified receptor. A summary of the effects of exogenous reagents on kinase activity is presented in Table 1 and Fig. 1. Effect of ATP Analogues on Kinase Activity. The demonstration of an ATP binding site was necessary to classify the receptor as a kinase. Highly purified activated receptor was preincubated with fluorosulfonylbenzoyl adenosine or photoaffinity-labeled with 8-azido-ATP and then incubated with calcium and [y-32P]ATP. The receptor failed to become phosphorylated, indicating that an ATP binding site was blocked. Subsequently, the receptor was incubated with [14C]fluorosulfonylbenzoyl adenosine. The labeled receptor was eluted from DEAE-cellulose at 50 mM KCI with coincidence of 3H and "'C labels (data not shown). Calculations from preliminary data suggest 3-5 binding sites for ["'Cifluorosulfonylbenzoyl adenosine per steroid binding entity (Mr,
94,000).
Proc. Natl. Acad. Sci. USA 82
(1985)
Table 1. Summary of effects of exogenous reagents on kinase activity associated with purified rat hepatic glucocorticoid receptor
Phosphorylation of Mr 94,000 component
Condition
[3H]TA*GR
GR (stripped) GR (stripped)/Ca2+ GR (boiled) [3H]TA-GR/Ca2+ (100 ,uM-10 mM) [3H]TA*GR/EGTA [3H]TA-GR/Ca27EGTA [3H]TA-GR/Mg2+ (100 ,uM-10 mM) [3H]TA-GR/calmodulin [3H]TA-GR/calmodulin/Ca2+ [3H]TA-GR/trifluoperazine (20 AM)
+ -
+
[3HITA-GR/trifluoperazine/Ca2+ [3H]TA'GR/chlorpromazine (20 ,zM) [3H]TA-GR/chlorpromazine/Ca2+ [3H]TA-GR/calmidazolium (10 AM) [3H]TA GR/calmidazolium/Ca2+ + [3H]TA GR/cAMP (1-10 mM) [3H]TA-GR/cAMP/Ca2+ + [3H]TA-GR/cGMP (1-10 mM) [3H]TA-GR/cGMP/Ca2+ + [3H]TA-GR/Ca27Na2MoO4 (10 mM) + [3HITAGR/Ca27PLP (10 mM) [3H]TA*GR/PLP TA, triamcinolone acetonide; GR, glucocorticoid receptor; PLP, pyridoxal 5'-phosphate.
Steroid Dependency of Kinase Activity. Unactivated [6,73H~triamcinolone acetonide-purified glucocorticoid receptor was treated with Dextran-coated charcoal to strip the receptor of bound steroid. Complete removal of steroid required two 10-sec incubations with Dextran-coated charcoal. Stripped receptor did not incorporate 32p when incubated in the presence or absence of calcium. The receptor could rebind steroid if incubated with steroid (100 nM) for 2 hr at 0C-40C. Once the purified unactivated glucocorticoid receptor was rebound to steroid and activated at 250C for 30 min, kinase activity returned. Extent of stripping and rebinding was determined by the hydroxylapatite assay. Stripped receptor consistently retained
