localization of the 90-kda heat shock protein-binding site within

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc

Vol. 266, No. 6, Issue of February 25, pp. 3482-3490,1991 Printed in U.S.A.

Localization of the 90-kDa Heat Shock Protein-binding Site within the Hormone-binding Domain of the Glucocorticoid Receptor by Peptide Competition*

(Received for publication, August 10, 1990)

Friederich C. DalmanS, Lawrence C. Scherrerg, Larry P. Taylor$, Huda Akilg, and William B. PrattSV From the $Department of Pharmacology and thesMenta1 Health Research Institute, University of Michigan Medical School, Ann Arbor, Michigan 48109-0626

In this work, we used two approaches to localize the 90-kDa heat shock protein (hsp90)-binding site within the hormone-binding domain of the glucocorticoid receptor. In the first approach, derivatives of the glucocorticoid receptor deleted for increasing portions of the COOH terminus were translated in rabbit reticulocyte lysate, and the [3sS]methionine-labeled translation products were immunoadsorbed with the 8D3 monoclonal antibody against hsp90. The data suggest that a segment from amino acids 604 to 659 (mouse) of the receptor is required for hsp90 binding. We have recently shown that the internal deletion mutant of the mouse receptor (A574-632) binds hsp90, although the complex is somewhat unstable (Housley, P. R., San- chez, E. R., Danielsen, M., Ringold, G . M., and Pratt, W. B. (1990) J. Biol. Chem. 265,12778-12781). The two observations indicate that amino acids 574-659 are involved in forming a stable receptor-hsp90 complex and that region 632-659 is especially important. To test this hypothesis directly, we synthesized three peptides corresponding to segments in region 624-665 and three peptides spanning the highly conserved sequence a t amino acids 582-617, and we then tested the ability of the peptides to compete for the association of hsp90 with the L cell glucocorticoid receptor. In this assay, the immunopurified hsp9O-free mouse receptor is incubated with rabbit reticulocyte lysate, which directs the association of rabbit hsp90 with the mouse receptor, simultaneously converting the receptor to the steroid binding state. All three peptides spanning region 624-665 and a peptide corresponding to segment 587-606 inhibited both hsp90 association with the receptor and reconstitution of steroid binding capacity. The data from all of the approaches support a two-site model for the hsp90-binding site in which the critical contact site occurs in region 632-659, which contains a short proline-containing hydrophobic segment and adjacent dipole-plus-cysteine motif that are conserved among all of the hsp9O-binding receptors in the superfamily. A second hsp9O contact site is predicted in region 574-632, which contains the only highly conserved amino acid sequence in the receptor superfamily outside of the DNA-binding domain.

CA28010 and DK31573 (to W. B. P.) and Grant MH 422251 and by * This work was supported by National Institutes of Health Grants

a grant from the L. P. Markey Fund (to H. A.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduer- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 To whom correspondence should be addressed.

The untransformed, non-DNA-binding form of the glucocorticoid receptor exists in cytosol preparations as a heter- omolecular complex containing 1 molecule of the steroid- binding protein and 2 molecules of hsp90l (Refs. 1-9; see Ref. 10 for review). Physical conditions that cause dissociation of hsp90 from both liganded and unliganded glucocorticoid receptors (e.g. dilution, salt, increased pH) cause the receptors to transform to the DNA binding state (6, 10). Binding of glucocorticoids to the hormone-binding domain of the receptor also promotes a temperature-dependent dissociation of hsp90 and generation of the DNA binding state (6, 11, 12). The glucocorticoid receptor is apparently unique among steroid receptors in that hsp90 is required to maintain the unliganded receptor in a high affinity steroid binding conformation (13-16).

When glucocorticoid receptors are translated in rabbit reticulocyte lysate, hsp9O binds at or near the termination of receptor translation, and the newly translated receptor is in both the high affinity steroid binding conformation and the non-DNA-binding form (15). In contrast, when the receptor is translated in wheat germ extract, it is not bound to hsp90, it does not have a high affinity steroid binding conformation, and it is in the DNA-binding form (15, 17).

Smith et al. (18) have recently shown that immunoadsorbed chick progesterone receptors incubated with rabbit reticulocyte lysate become associated with rabbit hsp90. We have found (19) that the reticulocyte lysate directs the reassociation of hsp90 with the immunoadsorbed murine L cell glucocorticoid receptor. We have also found (19) that the lysate directs the temperature-dependent dissociation of unliganded glucocorticoid receptors from a prebound receptor-DNA complex. The glucocorticoid receptor is released as the hsp90-bound receptor, and this reconstitution of the receptor-hsp90 complex is accompanied by complete restitution of high affinity steroid binding activity and repression of DNA binding activity. Thus, it is now possible to both structurally and function- ally reverse receptor transformation under cell-free conditions.

Both genetic (20) and biochemical (21) studies have demonstrated that hsp90 binds to the hormone-binding domain of the glucocorticoid receptor. Danielsen et al. (22) originally speculated that a conserved sequence of -20 amino acids

’ The abbreviations and trivial names used are: hsp90,90-kDa heat shock protein; ACTH, adrenocorticotropic hormone; triamcinolone acetonide, 9a-fluoro-llp,l6a,l7a,Z~-tetrahydroxypregna-~,4-diene- 3,20-dione 16,17-acetonide; dexamethasone, 9a-fluoro-1601-methyl- Ilp,l7a,21-trihydroxypregna-1,4-diene-3,2O-dione; Hepes, 4-(2-hy- droxyethy1)-1-piperazineethanesulfonic acid SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TES, N - tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid.

3482

hsp90-binding Site 3483

(amino acids 583-602 of the mouse glucocorticoid receptor) lying within the steroid-binding domain may be the site of interaction with hsp90. This conserved region is being viewed as a potential transducing domain (20), that is, as a domain responsible for transducing the free energy involved in steroid binding into a derepression of receptor function (9, 11, 20). Recent studies of hsp90 interaction with truncated glucocorticoid receptors (23) and receptors containing internal dele- tions (24) suggest that this sequence comprises part of the hsp90-binding site.

In this work, we examine hsp90 association with truncated glucocorticoid receptors translated in rabbit reticulocyte lysate and establish a region between amino acids 604 and 659 (mouse) that is required for hsp90 binding. We have synthesized a series of peptides corresponding to portions of segment 582-617, which contains the potential transducing domain, and to portions of the region just to its COOH-terminal side (amino acids 624-665), which we predict from studies of mutant receptors to be important for hsp90 binding. We show that four of these peptides inhibit rabbit reticulocyte lysate- directed reassociation of immunoadsorbed mouse L cell receptors with hsp9O and reactivation of the receptors to the steroid-binding form. Taken together with previous indirect studies, the results of this direct study permit us to define a short region of -78 amino acids in the hormone-binding domain that forms the major binding region for hsp9O in the glucocorticoid receptor. The COOH-terminal half of this region contains a motif that is conserved among those receptors that bind hsp90 and is not present in the thyroid hormone and retinoic acid receptors, which do not form stable complexes with hsp90.

EXPERIMENTAL PROCEDURES

Materials ~ - [~~S]Meth ion ine (1100 Ci/mmol), '251-labeled goat anti-rabbit

IgG, [3H]triamcinolone acetonide (42.5 Ci/mmol), and 13H]dexamethasone 21-mesylate (48.9 Ci/mmol) were supplied by Du Pont-New England Nuclear. AMPLIFY was supplied by Amersham Corp. Oc- tadecaneuropeptide, CD4 fragment 37-53, and ACTH fragment 1-17 were purchased from Sigma. The Riboprobe transcription system, SP6 and T7 RNA polymerases, and nuclease-treated rabbit reticulocyte lysate were obtained from Promega Biotech. The rat glucocorticoid receptor in plasmid pT3.1118 and rat receptor derivatives containing residues 407-766 (pTK976'E), 407-671 (pTK97 308), and 407-616 (pTK36'SCA) were kindly provided by Dr. Keith Yamamoto (25). Plasmid pT7N556 was constructed by Didier Picard and is composed of residues 1-556 inserted into a pSP73 vector downstream of the T7 promoter. This construct was also provided by Dr. Keith Yamamoto (University of California, San Francisco). The BuGR2 monoclonal antibody prepared against the rat glucocorticoid receptor was kindly provided by Dr. Robert Harrison I11 (University of Arkan- sas, Little Rock, AR) (26); rabbit antiserum that reacts with both hsp90 and hsp7O (27) was kindly provided by Dr. Ettore Appella (National Cancer Institute); and the 8D3 IgM monoclonal antibody against hsp9O (28) was provided by Dr. Gary Perdew (Purdue Uni- versity, West Lafayette, IN).

Methods In Vitro Transcription and Translation -Transcripts of the glu-

cocorticoid receptor and receptor derivatives 407-766, 407-671, and 407-616 were produced by SP6 RNA polymerase, and transcripts of glucocorticoid receptor derivative 1-556 were produced by T7 RNA polymerase. The transcription and translation reactions were carried out for 1 h at 30 'C in the presence of [35S]methionine exactly as described previously (15).

Immunoadsorption of in Vitro Translated Proteins -Translation products were incubated with the 8D3 anti-hsp90 antibody prebound to protein A-Sepharose adsorbed with goat anti-mouse IgM as described previously (15). Antigen-antibody complexes were washed three times with HEGD (10 mM Hepes, pH 7.4, 1 mM EDTA, 10% glycerol, and 5 mM dithiothreitol) plus 20 mM molybdate, and im-

munoadsorbed proteins were subsequently eluted off the Sepharose pellets with SDS sample buffer.

Cell Culture and Cytosol Preparation"L929 murine fibroblasts were grown in monolayer culture in Dulbecco's modified Eagle's medium supplemented with 10% calf serum a t 37 "C. Cells were harvested by scraping and washed in Earle's balanced saline. The cell pellet was suspended in 1.5 volumes of 10 mM Hepes, 1 mM EDTA, pH 7.4, and ruptured by Dounce homogenization. The homogenate was centrifuged at 100,000 X g, and the supernatant (referred to as cytosol) was used immediately or stored a t -70 "C for later use.

Immunoadsorption of L Cell Glucocorticoid Receptor-Cytosol was incubated with an equal volume of TEG (10 mM TES, pH 7.6,50 mM NaCl, 1 mM EDTA, and 10% glycerol) and 5% BuGR hybridoma fluid or an equivalent amount of nonimmune IgG for 3 h on ice. The solution was then rotated with protein A-Sepharose for 45 min a t 4 "C. The immunopellet was washed six times with 1 ml of TEGNT (TEG + 450 mM NaCl and 0.4% Triton X-100) to strip the receptor of associated proteins (including hsp90), and this was followed by two washes with 10 mM Hepes, pH 7.4.

Reassociation of Immunoadsorbed L Cell Receptor with hsp90 by Rabbit Reticulocyte Lysate-Reticulocyte lysate (usually 20 pl) was diluted with 15 g1 of Hepes and incubated with various concentrations of peptides in 15 pl of HEGD or with peptide-free HEGD in a total incubation volume of 50 pl for 30 min on ice. The lysate solution was then incubated for 30 min at 30 "C with 7 pl of the protein A- Sepharose immunopellet containing the unliganded receptor immunoadsorbed from 100 pl of L cell cytosol and dissociated from hsp90. The immunopellet was then washed three times with 1 ml of HEGD plus 20 mM molybdate, and bound proteins were eluted with SDS sample buffer for Western blot analysis. Alternatively, steroid binding capacity was assayed by suspending the washed immunopellet in 50 p1 of Hepes and incubating it with 50 nM [3H]triamcinolone acetonide with or without 50 p~ dexamethasone for 4 h on ice. The immunopellet was then washed three times with HEGD, and steroid binding was assayed by liquid scintillation spectrometry.

Gel Electrophoresis, Autoradiography, and Immunoblotting-SDS- polyacrylamide gel electrophoresis was performed on 8% slab gels as previously described (15). Following electrophoresis, proteins were stained with Coomassie Blue, the gel was incubated in AMPLIFY and dried, and autoradiography was performed a t -70 "C. Alterna- tively, for immunoblotting, proteins were transferred to Immobilon P transfer membranes, and membranes were subsequently incubated with 1% BuGR or 0.1% anti-hsp90 rabbit serum. Immunoreactive bands were then visualized by probing with a goat anti-mouse or anti- rabbit '251-labeled antibody followed by autoradiography as described previously (14).

Peptide Synthesis-Peptides were synthesized on an Applied Bio- systems Model 431 peptide synthesizer with amino acids, solvents, and reagents supplied by Applied Biosystems, Inc. The synthesis was conducted according to the manufacturer's recommended procedures using FMOC (N-(9-fluorenyl)methoxycarbonyl) as the N-terminal protecting group and the following side chain-protecting groups: N - a-FMOC-NG-(2,2,5,7,8-pentamethylchroman-6-sulfonyl)-~-arginine (Arg), 0-t-butyl (Asp, Glu, Ser, Thr, Tyr), and trityl (His, Cys). The peptides were cleaved from the resin with simultaneous removal of the side chain-protecting groups using trifluoroacetic acid with phenol, 1,2-ethanedithiol, thioanisole, and H20 as scavengers. The lyophilized crude material was purified by low pressure reverse-phase liquid chromatography using a 2.5 X 30-cm glass column packed with Waters C,, (55-105 pm) resin, and the desired fractions were pooled and lyophilized as previously described (29). Stock peptide solutions (2-5 mM) were prepared in HEG with 20 mM dithiothreitol, aliquoted, and frozen at -20 "C until use.

Six peptides were synthesized, as shown in Table I. In the few instances where there was a difference between the rat and the mouse sequences, the rat amino acid was used, except in peptide B, where the second amino acid is a proline, as in the mouse receptor, rather than a leucine, as in the rat receptor.

RESULTS

Amino Acids 616-671 in Rat Glucocorticoid Receptor Are Required for hsp90 Binding-To localize a minimal hsp90- binding region within the hormone-binding domain of the glucocorticoid receptor, derivatives of the rat receptor deleted for increasing portions of the COOH terminus were translated in rabbit reticulocyte lysate and immunoadsorbed with the

3484 hp90-binding Site TABLE I

Synthesized peptides

Peptide Residues

Mouse Rat Sequence

A B C D E F 652-665 664-677 MLFVSSELQRLQVS

582-598 594-610 KWAKAIPGFRNLHLDDQ 587-606 599-618 IPGFRNLHLDDQMTLLQYSW 597-617 609-629 DWMTLLQYSWMFLMAFALGWR 624-665 636-677 GNLLCFAPDLIINEQRMSLPLMYPQCKHMLFVSSELQRLQVS 637-665 649-677 EQRMSLPLMYPQCKHMLFVSSELQRLQVS

FIG. 1. Association of hsp90 with truncated rat glucocorticoid receptors translated in vitro. The wild-type rat glucocorticoid receptor and the truncated mutants indicated at the top were translated in rabbit reticulocyte lysate, and aliquots of lysate containing the [:''S]methionine-labeled receptor were immunoadsorbed with the 8D3 monoclonal antibody against hsp90 or with nonimmune IgM as described under "Methods." The immunoadsorbed proteins were resolved by SDS-PAGE and autoradiography. Panels A-E represent the receptor forms indicated by the let- ters A-E next to the diagrams at the top. Lanes 1, total translation products be- fore immunoadsorption; hnes 2, material immunoadsorbed by the 8D3 monoclonal antibody; lanes 3, material immunoadsorbed by nonimmune IgM.

DNA Hormone ,95 I

A I

B

C

766

671

D m 4 "

E '6

A 1 2 3

6 C D E 1 2 3 1 2 3 1 2 3 1 2 3

97-

06- n 8 x 45- s

8D3 monoclonal antibody directed against hsp90 (Fig. 1). The autoradiograph in Fig. lA shows that the wild-type receptor is immunospecifically absorbed by the 8D3 monoclonal antibody. As reported previously (15, 17), the 8D3 monoclonal antibody does not interact with the receptor itself, and im- munospecific absorption of the receptor indicates association with hsp90. Receptor derivatives terminating at amino acids 766 and 671 are also immunospecifically adsorbed (Fig. 1, B and C ) , whereas those terminating a t amino acids 616 and 556 are not (Fig. 1, D and E ) . These results suggest that a region between amino acids 616 and 671 of the rat receptor is required for hsp9O binding, and they confirm previous reports (3,20,21) that the amino-terminal half of the receptor is not involved in hsp90 binding.

Reconstitution of Receptor-hsp9O Complex by Reticulocyte Lysate-To test the ability of peptides to compete for the association of hsp90 with the glucocorticoid receptor, we have modified a recently described system (18, 19) for reconstitution of the receptor-hsp90 complex in which the immunoadsorbed, hsp90-free receptor is incubated with rabbit reticulocyte lysate. In the modified reconstitution system, the immunoadsorbed receptor bound to a protein A-Sepharose pellet is stripped of hsp90 by washing with salt and detergent, and then the immunopellet is incubated with reticulocyte lysate. As shown in Fig. 2 ( l a n e 1 ), the immunoadsorbed, hsp9O-free L cell glucocorticoid receptor does not bind steroid; but after incubation with rabbit reticulocyte lysate ( l a n e 3), the recep-

tor is associated with hsp90, and it now binds steroid. A control immunopellet without the mouse receptor does not contain hsp90 after incubation with reticulocyte lysate (lune 5), showing that reconstitution is occurring between receptor derived from mouse L cell cytosol and hsp90 derived from rabbit reticulocyte lysate.

As shown in Fig. 3, the amount of hsp90 that associates with the protein A-Sepharose immunopellet in a standard assay containing 20 p1 of reticulocyte lysate depends upon the amount of L cell cytosol that was immunoadsorbed with the BuGR anti-receptor antibody. In subsequent experiments, each aliquot of the protein A-Sepharose pellet contains the receptor immunoadsorbed from 100 pl of L cell cytosol for assay of receptor-associated hsp90 and the receptor immunoadsorbed from 50 pl of cytosol for assay of steroid binding.

The extent of reactivation of steroid binding capacity is decreased by dilution of lysate (Table 11). Dilution with HEGD in which the peptides were dissolved inhibits reactivation more than dilution in Hepes (Table 11), and in our standard assay, we have employed the condition where 20 pl of lysate is diluted with 15 pl of Hepes and 15 p1 of HEGD. In the presence of lysate alone, 50-6075 of the immunoadsorbed receptors are reactivated to the steroid binding state; and under the standard assay conditions we used to assay for peptide competition (20 pl of lysate, 15 p1 of Hepes, 15 p1 of HEGD), this value is decreased by about one-third (Table 11).

Synthetic Peptides Inhibit Reticulocyte Lysate-mediated As-

hsp90-binding Site 3485 1 2 3 4 5

hsp 90 I - [3H]DM bound to GR 0

FIG. 2. Rabbit reticulocyte lysate-directed reassociation of hsp90 with mouse glucocorticoid receptor (GR) and reactivation of steroid binding activity. Aliquots (1200 pl) of L cell cytosol were immunoadsorbed to protein A-Sepharose with 5% BuGR or with nonimmune IgG, and receptor-associated proteins were removed by washing with TEGNT followed by Hepes as described under "Methods." The immunopellets were resuspended in 50 pl of reticulocyte lysate, incubated for 30 min a t 30 "C, and washed four times with HEGD plus 20 mM molybdate. The final wash suspension was divided into aliquots, such that of the original 1200 pl of immunoadsorbed L cell cytosol, 50 and 450 pl were applied to SDS-PAGE and immunoblotted for detection of the receptor and hsp90, respectively. Of the remaining immunoadsorbed pellet from 700 pl of cytosol, 500 p1 of cytosol equivalents were incubated overnight with ['H]dexamethasone 21-mesylate (["]DM), and the receptor was resolved by SDS-PAGE and autoradiography. The remaining pellet from 100 pl of cytosol was incubated with ['H]triamcinolone acetonide in the presence or absence of competing nonradioactive dexamethasone to determine specific steroid binding capacity. Lane I , immunoadsorbed receptor washed with TEGNT to remove hsp90 and without subsequent treatment; lanes 2 and 3, receptor immunoadsorbed from L cell cytosol with nonimmune IgG and BuGR, respectively, and incubated with reticulocyte lysate; lanes 4 and 5, protein A-Sepharose pellets with adsorbed nonimmune IgG and BuGR, respectively, that were not exposed to L cell cytosol were incubated with rabbit reticulocyte lysate. The specific binding of ['Hltriamci- nolone acetonide for each sample was: lane 1, 856 cpm; lane 2, 100 cpm; lane 3,14,200 cpm; lane 4,230 cpm; and lune 5,409 cpm.

30

25 . 8

" -

0 100 200 300 400

-OR WJOtcytosoc) FIG. 3. Extent of hsp90 binding depends on amount of im-

munoadsorbed receptor. The aliquots of L cell cytosol indicated on the abscissa were immunoadsorbed to protein A-Sepharose with BuGR or nonimmune IgG (lane N I ) ; receptor-associated proteins were removed by washing with TEGNT followed by Hepes; and the pellets were incubated for 30 min a t 30 "C with 50 pl of a reticulocyte lysate/HEGD/Hepes mixture as described under "Methods." After washing the pellets, bound hsp9O was assayed by SDS-PAGE followed by immunoblotting and probing with anti-hsp90 serum followed by "'I-labeled counterantibody. The inset shows an autoradiograph of the "'I-labeled hsp90 bands on the immunoblot, and the main figure is a plot of the '*'I radioactivity present in each band.

sociation of hsp90 and Reactivation of Steroid Binding Capac- ity-Fig. 4 shows the effect of increasing concentrations of synthetic peptides A-F on reticulocyte lysate-mediated reconstitution of steroid binding capacity. In Fig. 4A, the stoichio- metric ratio of peptide to hsp9O is lower than in Fig. 4B. In Fig. 4A, each sample contained 21 pl of reticulocyte lysate in a total reactivation volume of 30 p1, with peptides A, C, and F being inactive, peptides D and E being the most potent, and peptide B showing inhibition only at the highest concentra-

TABLE I1 Effect of dilution on reconstitution of steroid binding

capacity by reticulocyte lysate Immunoadsorbed, hsp90-free receptors were incubated for 30 min

at 30 "C with 20 pl of Hepes alone or with 20 pl of rabbit reticulocyte lysate and additions as noted. At the end of the incubation, the immunopellets were assayed for their capacity to bind ['Hltriamci- nolone acetonide in a specific manner.

Specific steroid binding

cpm % Incubation conditions

Reticulocyte lysate alone 19,300 100 Lysate plus 30 pl Hepes 14,300 74 Lysate plus 15 pl Hepes and 15 pl HEGD 12,900 67 Lysate plus 30 pl HEGD 9,800 50 Hepes alone 507 3

O r nl 1m

1

J lm,

10 1m l m ,

PEPTIDE C O N c ~ ~ (IrM)

FIG. 4. Inhibition of reconstitution of steroid binding capacity by synthetic peptides. Immunoadsorbed, hsp90-free receptors were incubated for 30 min a t 30 "C in A with 20 p1 of reticulocyte lysate in a total reaction volume of 30 pl or in B with 20 pl of reticulocyte lysate in the standard 50-p1 reactivation mixture. In both cases, the reactivation mixture was preincubated with the indicated concentrations of synthetic peptides A-F prior to addition to the immunoadsorbed receptor. Following the incubation, the immunopellets were assayed for their capacity to bind ['HH]triamcinolone acetonide in a specific manner. The symbols for each of the peptides are indicated in B.

tion. To increase the molar ratio between the peptides and hsp90, which is present in the lysate at -2 pM (15), the reaction in Fig. 4B, was diluted to the standard 50-pl system described above, whereas the final concentration of peptides remained constant. This resulted in increased activity of peptides B, D, and E and the appearance of a modest inhibition by peptide F, with peptides A and C being inactive. In contrast to the other peptides, peptide C does not dissolve well; and for this reason, no further work was performed with it. I t should be noted that the peptides inhibit reconstitution of steroid binding capacity, but they do not affect the binding capacity of receptors that are already in the high affinity steroid binding conformation (data not shown). Three unre- lated peptides (octadecaneuropeptide, CD4 fragment 37-53, and ACTH fragment 1-17) were tested at 500 p~ and found to be inactive.

Fig. 5 shows the effect of the synthetic peptides on association of hsp90. Lanes I represent hsp90 bound to the immune (i.e. BuGR) pellets after incubation with reticulocyte lysate,

hsp90-binding Site

' I N I " I NI" I NI " I Nl" I NI '

FIG. 5. Inhibit ion of hsp9O binding by synthetic peptides. Immunoadsorbed, hsp90-free receptors were incubated with reticulocyte lysate and the indicated synthetic peptides (final concentration of 500 PM) as described in the legend to Fig. 4. Following their incubation, the immunopellets were washed, and pellet-associated hsp90 was assayed by immunoblotting with anti-hsp90 serum followed by "'I-labeled counterantibody. Shown is an autoradiograph of the "'I-labeled hsp90 bands on the immunoblot. Each synthetic peptide is indicated by the letter over each bracket, and None repre- sents the control without peptide. Lunes I , hsp 90 bound to the BuGR immunopellet; lanes NI, hsp90 bound to the nonimmune pellet.

I N o n e A B D E F

PEPTIDE FIG. 6. Relationship between peptide-mediated inhibition

of hsp9O binding to receptor and inhibition of reconsti tution of steroid binding capacity. Receptor-associated hsp90 was determined by cutting out the '251-labeled bands shown in the immunoblot of Fig. 5, assaying the radioactivity in each band, and subtracting the nonimmune from the immune value. Specific binding of ["HJtriam- cinolone acetonide was assayed in a portion of the same immunopellet used for assay of hsp90.

and lanes NI represent hsp9O bound to the nonimmune IgG pellet. The presence of the synthetic peptide is noted by the letter above each bracket. In all cases except for peptide E, there is a difference between lanes I and NI, representing receptor-associated hsp90. In all of our experiments, we have found that peptide E causes a marked increase in hsp90 associated with the nonimmune pellet as shown here. I t seems that binding of peptide E to hsp90 somehow makes it stick nonspecifically to protein A-Sepharose.

Fig. 6 shows the relationship between the amount of receptor-associated hsp90 (shaded bars) and steroid binding capacity (open bars) in the same immunopellet after incubation with reticulocyte lysate in the presence of 500 p~ synthetic peptides. The hsp9O values were derived by counting the '"I- labeled bands shown in the immunoblot of Fig. 5 and subtracting the nonimmune values from the immune values. There is a clear correlation between the ability of each peptide to inhibit hsp90 association and its ability to inhibit reconstitution of steroid binding capacity. Figs. 7 and 8 show that there is a rough correlation between the concentration dependence for inhibition of both hsp90 association and reconstitution of steroid binding capacity on the part of peptides D and B, respectively. The increase in nonspecific hsp90 binding caused by peptide E precludes a similar comparison with that peptide.

M P u) \

O\O 1' 0 ' 10 loo

. . . ' o lo00

P E r m D E f x x K m m A r n ~

FIG. 7. Concentration dependence of peptide D for inhibition of hsp90 association and reconsti tution of steroid binding capacity. Immunoadsorbed, hsp90-free receptors were incubated with reticulocyte lysate in the presence of various concentrations of peptide D, and receptor-associated hsp90 (0) and specific binding of [''H]triamcinolone acetonide (0) were assayed. The dashed line rep- resents the level of receptor-associated hsp90 and specific steroid binding capacity in control samples reactivated without peptide. An autoradiograph of the "'I-labeled immunoblot used to derive the hsp90 values is at the top.

16 -

10 -

0 a

2

- 1

1 1m 1000

pEpTDEalNc€mumoIM) FIG. 8. Concentration dependence of peptide B fo r inhibi-

tion of hsp9O association and reconsti tution of steroid binding capacity. The experiment performed was as described in the legend to Fig. 7, except that various concentrations of peptide B were present. 0 and 0, receptor-associated hsp90 and specific binding of ['HI triamcinolone acetonide, respectively, in the presence of various concentrations of peptide B; 0, specific steroid binding in samples reactivated in the presence of 1 mM peptide A.

DISCUSSION

In our previous report (15) examining the association of hsp9O with the in vitro translated wild-type rat glucocorticoid receptor, we were unable to demonstrate association of the heat shock protein with the incompletely translated receptor. A t that time, it was not clear whether the COOH terminus of the receptor was required for hsp90 binding or whether hsp90 bound to the receptor only at the end of translation. It is clear from the data of Fig. 1 that the COOH terminus is not required for hsp9O binding.

The fact that receptor fragments 407-766 and 407-671 bind hsp90 but that fragment 407-616 is not coimmunoadsorbed with hsp9O by the 8D3 monoclonal antibody is of interest in view of the observation of Picard and Yamamoto (30) that truncated receptors terminating at amino acids 768 and 671 are localized predominantly in the cytoplasm of COS7 cells

hsp90-binding Site

DNA Hormone

3487

m H Minimal hsp90 binding segment

Housley et al (1990) f l f i L < Unstable deletion

Simons et al(l989) 5*" Core steroid binding sequence

FIG. 9. Localization of hsp90-binding region within hormone-binding domain of glucocorticoid

published by Danielsen et al. (22). The potential hsp90-binding region is indicated by the open rectangle within receptor. All numbers represent amino acid assignments for the mouse receptor according to the primary sequence

the stippled hormone-binding domain. From the work of Howard et al. (23), the amino-terminal border should not be less than residue 556, and we propose that the COOH-terminal border should be in the region of residues 659- 661. The minimal hsp90-binding segment (residues 604-659) reflects the data of Fig. 1 in this paper. Housley et al. (24) have reported that the receptor with an internal deletion of amino acids 574-632 binds hsp90, but that the complex appears to be unstable. Simons et al. (33) have derived a receptor core fragment (residues 525-661) by proteolysis that has steroid binding activity and is bound to hsp90.

in the absence of hormone, whereas a receptor terminating at amino acid 615 is nuclear. This is consistent with the specu- lation (30) that hsp90 may play a role in determining the localization of the unliganded glucocorticoid receptor in those cell types where the unliganded receptor is localized predominantly to the cytoplasm. It is clear, however, that in some cell types, the unliganded glucocorticoid receptor is both associated with hsp90 and localized to the cell nucleus (31, 32). Thus, other factors must be involved in determining the cellular localization of the unliganded receptor.

The approach used in the experiment of Fig. 1 has some limitations with respect to interpretation of the data. First of all, it must be realized that reticulocyte lysate contains -2 PM hsp90 and that the ratio of hsp90 to the translated receptor is about 3001 (15). We have shown previously (15) that nearly all of the full-length receptor can be immunoadsorbed after translation i n uitro if a high enough concentration of the 8D3 monoclonal antibody is added to the sample. However, to conserve antibody in this work, we have immunoadsorbed only a portion of hsp90 and thus only a portion of the receptor. As one can see by comparing A-C in Fig. 1, roughly the same proportion of the full-length translation product and trunca- tions 407-766 and 407-671 are immunoadsorbed (lane 2 ) when compared to the total amount of translation product in each sample (lane 1 ). Second, loss of the ability to bind hsp90 on the part of truncation 407-616 could reflect a gross structural change in the molecule rather than deletion of the site. The fact that the truncated receptors shown in Fig. 1 all bind glucocorticoid response elements (25) argues against such a gross structural change. The loss of region 616-671 could also produce a more subtle effect. For example, segment 407-616 could provide enough of a contact region for hsp90 to yield a relatively stable receptor-hsp90 complex that can be detected with some anti-hsp90 antibodies as reported by Howard et al. (23); but for some structural reason, the epitope for the 8D3 monoclonal antibody used in this work may be blocked. Be- cause of these reservations regarding the deletion approach of Fig. 1, we used the peptide competition method to further define a core hsp90-binding region.

Fig. 9 presents a schematic summary of observations that help to define an hsp90-binding region within the hormone- binding domain of the glucocorticoid receptor. The data of Fig. 1 suggest that the segment from amino acids 616 to 671 of the rat receptor (amino acids 604-659 of the mouse recep-

tor) is required for a stable receptor-hsp90 complex. We have recently shown (24) that a mutant mouse glucocorticoid receptor containing an internal deletion between amino acids 574 and 632 is associated with hsp90, although the association appears to be rather unstable in the absence of molybdate. These observations suggest that amino acids 574-659 are involved in hsp9O binding, and it is likely that a feature determined by residues 632-659 is especially important.

Although these experiments do not eliminate any contri- bution by segment 659-783 to the binding of hsp90 by the full-length receptor, this COOH-terminal region of the hormone-binding domain is clearly not required for formation or maintenance of a stable receptor-hsp90 complex. This analysis is also consistent with the observation of Simons et al. (33), who demonstrated that a 16-kDa fragment produced by proteolytic cleavage of the untransformed rat glucocorticoid receptor can bind dexamethasone in a high affinity manner. This 16-kDa fragment contains amino acids 537-673 of the rat receptor, which correspond to residues 525-661 of the mouse receptor (Fig. 9). This core steroid-binding fragment is stabilized by molybdate (33), and it is bound to a heat shock protein that is almost certainly hsp90.'

Thus, the results from three experimental approaches in- volving 1) in vitro translation of receptor fragments (Fig. 1 and Ref. 23), 2) transfection of receptors with internal dele- tions (24), and 3) proteolytic cleavage of the wild-type receptor (33) all contribute to an assignment of the region shown in Fig. 9 as the hsp90-binding site. It is of considerable impor- tance that the region binding hsp90 also forms a major part of the hormone-binding site. As noted in Fig. 9, Met-610 and Cys-644 lie within the predicted hsp90-binding region. Met- 610 becomes covalently bound to the A-ring of triamcinolone acetonide upon UV irradiation of the hormone-receptor complex (34), and Cys-644 becomes covalently bound to dexamethasone 21-mesylate in the region of the D-ring (35, 36).

To test the hsp90-binding site model of Fig. 9, we synthesized peptides A-F shown in Fig. 10. Peptides A-C correspond to regions of the highly conserved segment of the hormone- binding domain that is eliminated in the internal deletion mutant (A574-632) that binds hsp90 but forms a less stable complex than the wild-type receptor (24). Peptides D-F correspond to regions of segment 632-659, which appears to be a critical component of the hsp90-binding site. Four of these

S. S. Simons, Jr., personal communication.

3488 hsp90-binding Site MOUSE GLUCOCORTICOID RECEPTOR

DNA HORMONE

V/A I g@g~,;;. :>,; ;:; '_i ,' 5 , 'i:j <

,,,,. *:I&+- ?<,%,<, ?I ._. ., b;,,",, . , ,., , ,,

565 CONSERVATION OF HYDROPHOBIC AND CHARGED RESIDUES

1 .c an-VI 00 0 +YeO+Un. ..--..e.... e..+. m . r - m - + . w 0 . 5 ~ . + n w . . + a . ..-... so. ..a+. 0." 0O . . . an+.. . . -... .

665

1 ..rm . 0 . .

0 .

570 580 590 600 610 620 630 640 650 660

cu Y I ~ ~ T L ~ ~ L ~ R Q V I ~ ~ ~ A ~ I P G ~ R N L H L D D P M : s R ......................................................... E ""~"""~""~"~"~-"""""~~"""""""""""~~~""""""""""""""~~~~""""~

YR LLS---R-A-K-M-QV-----VL---K--~-E--I--I-----C.SS---S----~~NS~~-~-~---VF--EK-~QSA--EL-QG-"Q--LQ~---L~ AR LLSS--E--E--LVHV--- - - -L-- - - - - -V-- - -AVI-- - - -G--V--M--- -FT-S~-~-- - - -VF--~- -H~S~--S . -V~-~"L-Q-F~WL-~~

ER M-GL-TN-AD-ELVHMIN"-RV"-VD-~-H--VH--ECA-LEILMIG-V---MEHPVK -L---N-LLDRNPGKCVEGMEIFD--~TSSRF-HMNL P R LL-S--Q--E--LLSV---S-SL-------T"-I"f--I-----S--V-G-------S-Q~-y------L-----~SS~~SL-LT-W~-pQ-~~--~

VDR MLPH-AD-VSYSIQKVIGF"M"---D-TSE--IV--KSSAIEVI~RSNESFTMDD MSWTCGNQDYmRVSDVTKAGHSLEL-EPLIRFQVGL TBR FSHFTKIITPAITRV-DF--RL-M-CE-PCE--II"KGCC-EI-SL~V-Y DPESET-TLNGEMAVIRGQLKNGGLGWSDAIFDLGMS-SSFNm I U R LWDKFSE-STKCI-KT-EF--QL---TT-TIA--IT--KAA CLDILI-RICTRYTPEQDTMT-SDG-TL-RTQHHNAGFGPLTDLVFAFANQ-LP-EHD

SYNTHETIC PEPTIDES DERIVED FROM GR SEQUENCE

D A- B-

c- E-

F-

FIG. 10. Conservation of hydrophobic and charged residues within hsp90-binding region of steroid receptors. The sequence between residues 564 and 665 of the mouse glucocorticoid receptor (GR M) (22) is aligned with the comparable sequences of the rat (GR R ) (37) and human (GR H) (38) glucocorticoid receptors as well as with sequences of the human mineralocorticoid ( M R ) (39), androgen (AR) (40,41), progesterone ( P R ) (42), estrogen ( E R ) (43, 44), vitamin D (VDR) (45), thyroid hormone (THR) (46), and retinoic acid (RAR) (47, 48) receptors. Sequences were aligned to the glucocorticoid receptor using the Bestfit comparison program from the sequence analysis software package of the Genetics Computer Group of the University of Wisconsin using the full- length receptor sequence in all cases except for the vitamin D receptor, where 120 residues spanning the sequence were used. The solid dots above the sequences refer to hydrophobic residues and plus and minus signs to charged residues. The top line indicates conservation of a charged or hydrophobic residue from the glucocorticoid receptor through the progesterone receptor (GR-PR) in the series of sequences presented below; two symbols indicate conservation through the estrogen receptor (GR-ER), and three symbols represent conservation through all of the receptors (GR-RAR). The vertical arrows indicate charged residues that are conserved through all of the known hsp9O-binding receptors. The short bracket demarcates a 4-amino acid sequence (residues 636-639) that contains dipole residues from the glucocorticoid receptor through the vitamin D receptor. The negatively and positively charged amino acids are separated by one amino acid, except for the estrogen receptor, where they lie as an adjacent (Asp-Arg) pair. The solid bars show the positions of synthetic peptides A-F.

peptides were active at inhibiting reconstitution of L cell receptor steroid binding capacity by rabbit reticulocyte lysate, with peptides D and E being the most potent and peptide B having substantial activity (Fig. 4). The same relationship is maintained for inhibition of hsp90 association (Figs. 5 and 6), although as noted, the effect of peptide E on hsp90 association is difficult to evaluate because the peptide enhances nonspecific hsp9O binding (Fig. 5). Peptide F is a weak inhib- itor of both reactivation of steroid binding capacity and association of hsp90. That peptides B, D, and E inhibit reconstitution of the receptor-hsp90 complex is consistent with the proposed hsp90-binding region outlined by the rectangle lying within the hormone-binding domain in the model of Fig. 9.

At this time, we envision two principal sites of hsp90 contact lying within the hsp90-binding region of the receptor shown in Fig. 9. One site of contact lies in the amino-terminal half of the hsp90-binding region in the area of the peptide B sequence. A second contact region lies in the COOH-terminal half in the area defined by the sequence of peptide D. The sites presumably lie close to each other on the surface of the receptor and may indeed form a continuous binding surface with which hsp90 interacts. Binding of peptide with either portion of the hsp9O-binding region as defined from primary sequence produces enough steric hindrance to compete for hsp90 binding in the reconstitution assay. The easiest way to explain the deletion data of Fig. 1 is that loss of segment 604-

649 (mouse) reduces the affinity of hsp90 binding enough so that we do not recover hsp90 bound to fragment 407-616 (rat) using our method of isolation in Fig. 1. With the longer fragment 407-671 (rat), enough additional area of contact is provided to yield a high affinity interaction that is readily assayed in our immunoadsorption procedure.

Fig. 10 presents a comparison of sequences (22, 37-48) for this region of several receptors, where the conservation of charged residues and hydrophobic residues (solid dots) is indicated, with one symbol meaning conservation from the glucocorticoid through the progesterone receptor, two symbols conservation through the estrogen receptor, and three symbols conservation through all receptors.

Of the receptors shown in Fig. 10, the unliganded glucocorticoid, mineralocorticoid, androgen, progesterone, and estrogen receptors all form stable complexes with hsp90 (see Ref. 10 for references), whereas the thyroid hormone receptor (17) and the retinoic acid receptor3 do not form complexes that are detectable using the same method we have employed in the experiments of Fig. 1. As we have noted (17), this failure to bind hsp90 may explain why in the intact cells the unliganded thyroid hormone (49-51) and retinoic acid (52) receptors become tightly associated with the nucleus without re-

' F. C. Dalman, L. J. Sturzenbecker, A. A. Levin, D. A. Lucas, G. H. Perdew, M. Petkovich, P. Chambon, J. F. Grippo, and W. B. Pratt, manuscript submitted for publication.

hsp90-binding Site 3489

maining in a loosely bound “docking” state until the binding of hormone triggers a progression to high affinity sites. Un- fortunately, there is some controversy as to whether the unliganded vitamin D receptor binds tightly only in the nucleus like the thyroid hormone receptor or whether it enters into a “docking” complex from which it can undergo ligand- dependent translocation like the glucocorticoid receptor (53- 55). It is also not known whether the vitamin D receptor can enter into a stable complex with hsp90 or not.

In considering the sequences shown in Fig. 10, it is clear that the highest conservation through all of the receptors is in the segment from Trp-583 to Leu-602 (mouse glucocorticoid receptor), which was previously suggested as a potential site of interaction with hsp90 (20,22). However, as this region is conserved in both the thyroid hormone and retinoic acid receptors, which do not form stable complexes with hsp90, and as a receptor containing a deletion of this region (A574- 632) binds hsp90 (24), a model of the receptor-hsp90 complex based upon interaction of the heat shock protein with only this highly conserved portion of the sequence cannot be correct. However, the observations that the complex formed between hsp90 and mutant receptor A574-632 is less stable (24) and that peptide B can inhibit reassociation of hsp90 (Figs. 5, 6, and 8) both suggest that the amino-terminal half of the sequence presented in Fig. 10 contains a contact site for hsp90, as is also suggested by the data of Howard et al. (23). The vertical arrows in Fig. 10 point to charged residues that are conserved through all of the receptors that are known to bind hsp90. This region contains an interesting periodicity of charged residues with intervening conserved regions of hydrophobicity. The charged residues should lie on the surface of the protein, which would be consistent with a potential role in protein-protein interaction.

As discussed above, region 632-659 appears to be critical for the formation of a stable receptor-hsp90 complex. This is roughly the region spanned by peptides D (amino acids 624- 665) and E (amino acids 637-665), which are inhibitors in the reticulocyte reconstitution assay. There is a 9-amino acid region of hydrophobicity (7 of 9 residues) from Leu-627 to Ile-635 conserved for all of the receptors binding hsp90 (i .e. the glucocorticoid receptor through the estrogen receptor). This hydrophobic sequence contains a central proline residue that is also conserved only in the known hsp90-binding receptors. Immediately to the COOH-terminal side of this hydrophobic region (amino acids 627-635), there is a dipole that is conserved from the glucocorticoid through the progesterone receptor and that is also present in the vitamin D receptor. The dipole motif consists of a negatively charged amino acid separated by 1 residue from a positively charged amino acid. The estrogen receptor also has a dipole (Asp-Arg) within the region that is bracketed in Fig. 10, but it differs slightly in that the two charged moieties are not separated by an intervening amino acid.

In the glucocorticoid receptor, 2 cysteines are located 5 (Cys-644) and 10 (Cys-649) amino acids to the COOH-terminal side of the positively charged residue in the dipole. Although the other receptors that are known to form stable complexes with hsp90 do not possess a vicinal thiol pair, they all have 1 cysteine residue located either 5 residues (estrogen receptor) or 10 residues (mineralocorticoid, androgen, and progesterone receptors) to the COOH-terminal side of the positively charged amino acid in the dipole. The presence of a dipole, if it is spatially close to the cysteine, could be expected to make the cysteine sulfhydryl more acidic, and that may be an important determinant to be considered in

the role of this conserved motif in a possible protein-protein interaction.

The thyroid hormone and retinoic acid receptors do not form a stable complex with hsp90, and they do not contain the conserved hydrophobic region (amino acids 627-635) or the adjacent dipole-plus-cysteine motif. This leads us to focus on these two features within region 632-659 as being of particular interest as potential determinants of stable association with hsp90, and the latter feature in particular is an important site to study through site-specific mutagenesis.

Given that the core steroid-binding segment of Simons et al. (33) is bound to a heat shock protein that is almost certainly hsp90 and is stabilized by molybdate and that mutant receptor A574-632 binds hsp90 and is stabilized by molybdate (24), segment 632-659 is likely to be the region involved in transition metal oxyanion stabilization of the complex. It should be mentioned here that although it is quite clear that molybdate physically stabilizes the receptor-hsp90 complex (e.g. Ref. 9) and that it can directly interact with cysteine sulfhydryl groups located in the hormone-binding domain (56), it has not yet been demonstrated that the stabilizing effect of molybdate involves a direct interaction with the receptor. It is still theoretically possible that molybdate could interact entirely with hsp90 to maintain it in a conformation that has high affinity for the receptor. If so, the site on the receptor that interacts with such a hypothetical metal-stabilized conformation of hsp90 is still likely to be segment 632-659.

In addition to the likely involvement of region 632-659 in hsp90 binding and molybdate stabilization, it is clear that the cysteines in the dipole-plus-cysteine motif located in this region play a critical role in determining steroid binding to the glucocorticoid receptor (57-59), and it is likely that con- formational changes occurring at this site are critical for determining agonist versus antagonist activity of dexamethasone mesylate (60, 61). It will be especially important to develop a three-dimensional concept of the conformation of the entire region defined by the primary sequences in Fig. 10. Despite the uncertainties inherent to computer modeling, this approach may provide useful information about this limited region until an x-ray crystallographic analysis of the hormone-binding domain becomes available.

Acknowledgments-We wish to thank Keith Yamamoto for providing the plasmids containing cDNAs for the rat glucocorticoid receptor; Gary Perdew and Robert Harrison I11 for providing the 8D3 and BuGR monoclonal antibodies against hsp9O and the glucocorticoid receptor, respectively; and Ettore Appella for providing the rabbit antiserum against hsp90 that was used for Western blotting. We also thank Dave Toft, Stoney Simons, and Clark Distelhorst for sharing the results of experiments prior to publication.

REFERENCES

1. Sanchez, E. R., Toft, D. O., Schlesinger, M. J., and Pratt, W. B.

2. Housley, P. R., Sanchez, E. R., Westphal, H. M., Beato, M., and Pratt, W. B. (1985) J. Biol. Chern. 260, 13810-13817

3. Gehring, U., and Arndt, H. (1985) FEBS Lett. 179, 138-142 4. Okret, S., Wikstrom, A. C., and Gustafsson, J. A. (1985) Biochern-

istry 24,6581-6586 5. Mendel, D. B., Bodwell, J. E., Gametchu, B., Harrison, R. W.,

and Munck, A. (1986) J. Biol. Chem. 261, 3758-3763 6. Sanchez, E. R., Meshinchi, S., Tienrungroj, W., Schlesinger, M.,

Toft, D. O., and Pratt, W. B. (1987) J. Biol. Chern. 262,6986- 6991

7. Denis, M., Wikstrom, A. C., and Gustafsson, J. A. (1987) J. Biol. Chern. 262 , 11803-11806

8. Mendel, D. B., and Orti, E. (1988) J. Bwl. Chern. 263,6695-6702 9. Bresnick, E. H., Dalman, F. C., and Pratt, W. 3. (1990) Biochem-

(1985) J. Biol. Chem. 260, 12398-12401

istry 29,520-527

3490 hp90-binding Site

25. 26.

27.

28. 29.

30. 31.

32.

33.

34.

35.

36.

37.

10. Pratt, W. B. (1987) J. Cell. Biochem. 3 5 , 51-68 11. Meshinchi, S., Sanchez, E. R., Martell, K. J., and Pratt, W. B.

(1990) J. Biol. Chem. 265,4863-4870 12. Denis, M., Poellinger, L., Wikstrom, A. C., and Gustafsson, J. A.

(1988) Nature 333,686-688 13. Bresnick, E. H., Sanchez, E. R., and Pratt, W. B. (1988) J. Steroid

Biochem. 30 , 267-269 14. Bresnick, E. H., Dalman, F. C., Sanchez, E. R., and Pratt, W. B.

(1989) J. Bid. Chem. 264,4992-4997 15. Dalman, F. C., Bresnick, E. H., Patel, P. D., Perdew, G. H.,

Watson, S. J., Jr., and Pratt, W. B. (1989) J. Biol. Chem. 264 ,

16. Nemoto, T., Ohara-Nemoto, Y., Denis, M., and Gustafsson, J. A. (1990) Biochemistry 29, 1880-1886

17. Dalman, F. C., Koenig, R. J., Perdew, G. H., Massa, E., and Pratt, W. B. (1990) J. Bwl. Chem. 265,3615-3618

18. Smith, D. F., Schowalter, D. B., Kost, S. L., and Toft, D. 0. (1990) Mol. Endocrinol. 4 , 1704-1711

19. Scherrer, L. C., Dalman, F. C., Massa, E., Meshinchi, S., and Pratt, W. B. (1990) J. Biol. Chem. 2 6 5 , 21397-21400

20. Pratt, W. B., Jolly, D. J., Pratt, D. V., Hollenberg, S. M., Giguere, V., Cadepond, F. M., Schweizer-Groyer, G., Catelli, M. G., Evans, R. M., and Baulieu, E. E. (1988) J. Biol. Chem. 263,

21. Denis, M., Gustafsson, J. A., and Wikstrom, A. C . (1988) J. Biol.

22. Danielsen, M., Northrop, J. P., and Ringold, G. M. (1986) EMBO

23. Howard, K. J., Holley, S. J., Yamamoto, K. R., and Distelhorst, C. W. (1990) J. Biol. Chem. 2 6 5 , 11928-11935

24. Housley, P. R., Sanchez, E. R., Danielsen, M., Ringold, G. M., and Pratt, W. B. (1990) J. Biol. Chem. 265,12778-12781

. Rusconi, S., and Yamamoto, K. R. (1987) EMBO J. 6,1309-1315

19815-19821

267-293

Chem. 263 , 18520-18523

J. 5, 2513-2522

Gametchu, B., and Harrison, R. W. (1984) Endocrinology 114,

Ehrhart, J. C., Duthu, A,, Ullrich, S., Appella, E., and May, P.

Perdew, G. H. (1988) J. Biol. Chem. 263, 13802-13805 Watson, S. J., Khachaturian, H., Taylor, L., Fischell, W., Gold-

stein, A., and Akil, H. (1983) Proc. Natl. Acad. Sci. U. s. A.

Picard, D., and Yamamoto, K. R. (1987) EMBO J. 6 , 3333-3340 Sanchez, E. R., Hirst, M., Scherrer, L. C., Tang, H. Y., Welsh,

M. J., Harmon, J. M., Simons, S. S., Ringold, G. M., and Pratt, W. B. (1990) J. Biol. Chem. 265,20123-20130

Gasc, J. M., Delahaye, F., and Baulieu, E. E. (1989) Exp. Cell. Res. 1 8 1,492-504

Simons, S. S., Jr., Sistare, F. D., and Chakraborti, P. K. (1989) J. Biol. Chem. 264,14493-14497

Carlstedt-Duke, J., Stromstedt, P. E., Persson, B., Cederlund, E., Gustafsson, J. A., and Jornvall, H. (1988) J. Biol. Chem. 263,

Simons, S. S., Pumphrey, J. G., Rudikoff, S., and Eisen, H. J.

Smith, L. I., Bodwell, J. E., Mendel, D. B., Ciardelli, T., North,

Miesfeld, R., Rusconi, S., Godowski, P. J., Maler, B. A., Okret,

274-279

(1988) Oncogene 3, 595-603

80, 891-894

6842-6846

(1987) J. Biol. Chem. 262,9676-9680

W. G., and Munck, A. (1988) Biochemistry 27,3747-3753

S., Wikstrom, A. C., Gustafsson, J. A., and Yamamoto, K. R. (1986) Cell 46,389-399

38. Hollenberg, S. M., Weinberger, C., Ong, E. S., Cerelli, G., Oro, A., Lebo, R., Thompson, E. B., Rosenfeld, M. G., and Evans, R. M. (1985) Nature 318 , 635-641

39. Arriza, J. L., Weinberger, C., Cerelli, G., Glaser, T. M., Handelin, B. L., Housman, D. E., and Evans, R. M. (1987) Science 237 , 268-275

40. Lubahn, D. B., Joseph, D. R., Sullivan, P. M., Willard, H. F., French, F. S., and Wilson, E. M. (1988) Science 240,327-330

41. Chang, C. S., Kokontis, J., and Liao, S. T. (1988) Science 240 ,

42. Misrahi, M., Atger, M., d‘Auriol, L., Loosfelt, H., Meriel, C., Fridlansky, F., Guiochon-Mantel, A., Galibert, F., and Milgrom, E. (1987) Biochem. Bwphys. Res. Commun. 143, 740-748

43. Green, S., Walter, P., Kumar, V., Krust, A., Bornert, J. M., Argos, P., and Chambon, P. (1986) Nature 320 , 134-139

44. Greene, G. L., Gilna, P., Waterfield, M., Baker, A., Hort, Y., and Shine, J. (1986) Science 231 , 1150-1154

45. Baker, A. R., McDonnell, D. P., Hughes, M., Crisp, T. M., Mangelsdorf, D. J., Haussler, M. R., Pike, J. W., Shine, J., and O’Malley, B. W. (1988) Proc. Natl. Acad. Sci. U. S. A. 8 5 ,

46. Weinberger, C., Thompson, C. C., Ong, E. S., Lebo, R., Gruol, D.

47. Giguere, V., Ong, E. S., Segui, P., and Evans, R. M. (1987) Nature

48. Petkovich, M., Brand, N. J., Krust, A., and Chambon, P. (1987) Nature 330,444-450

49. Samuels, H. H., Tsai, J. S., Casanova, J., and Stanley, F. (1974) J. Clin. Invest. 54, 853-865

50. Casanova, J., Horowitz, Z. D., Copp, R. P., McIntyre, W. R., Pascual, A. and Samuels, H. H. (1984) J. Biol. Chem. 259, 12084-12091

51. Spindler, B. J., MacLeod, K. M., Ring, J., and Baxter, J. D. (1975) J. Bid. Chem. 250,4113-4119

52. Nervi, C., Grippo, J. F., Sherman, M. I., George, M. D., and Jetten, A. M. (1989) Proc. Natl. Acad. Sci. U. S. A. 8 6 , 5854- 5858

324-326

3294-3298

J., and Evans, R. M. (1986) Nature 324 , 641-646

330,624-629

53. Walters, M. R. (1985) Endocr. Rev. 6, 512-543 54. Walters, S. N., Reinhart, T. A., Dominick, M. A., Horst, R. L.,

and Littledike, E. T. (1986) Arch. Biochem. Biophys. 246,366- 373

55. Haussler, M. R., Mangelsdorf, D. J., Komm, B. S., Terpening, C. M., Yamaoka, T., Allegretto,, E. A., Baker, A. R., Shine, J., McDonnell, D. P., Hughes, M., Weigel, N. L., O’Malley, B. W., and Pike, J . W. (1988) Recent Prog. Horm. Res. 44,263-305

56. Meshinchi, S., Matit, G., Hutchison, K. A., and Pratt, W. B. (1990) J. Biol. Chem. 2 6 5 , 11643-11649

57. Miller, N. R., and Simons, S. S., Jr. (1988) J. Biol. Chem. 263,

58. Simons, S. S., Chakraborti, P. K., and Cavanaugh, A. H. (1990)

59. Lopez, S., Miyashita, Y., and Simons, S. S. (1990) J. Bid. Chem.

60. Sistare, F. D., Hager, G . L., and Simons, S. S. (1987) Mol.

61. Miller, P. A., and Simons, S. S. (1988) Endocrinology 122,2990-

15217-15225

J. Biol. Chem. 265 , 1938-1945

265, 16039-16042

Endocrinol. 1,648-658

2998

localization of the 90-kda heat shock protein-binding site within

Documents