HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 37, 33698-33703, September 13, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/37/33698    most recent M204164200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kanelakis, K. C. Articles by Pratt, W. B. Search for Related Content PubMed PubMed Citation Articles by Kanelakis, K. C. Articles by Pratt, W. B. Social Bookmarking                      What's this?

Nucleotide Binding States of hsp70 and hsp90 during Sequential Steps in the Process of Glucocorticoid Receptor·hsp90 Heterocomplex Assembly^*

Kimon C. Kanelakis, Donna S. Shewach, and William B. Pratt

From the Department of Pharmacology, The University of Michigan Medical School, Ann Arbor, Michigan 48109

Received for publication, April 29, 2002, and in revised form, June 7, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

A minimal system of five purified proteins, hsp90, hsp70, Hop, hsp40, and p23, assembles glucocorticoid receptor (GR)·hsp90 heterocomplexes and causes the simultaneous opening of the steroid binding cleft to access by steroid. The first step in assembly is the ATP-dependent and hsp40 (YDJ-1)-dependent binding of hsp70 to the GR, which primes the receptor for subsequent ATP-dependent activation by hsp90, Hop, and p23 (Morishima, Y., Murphy, P. J. M., Li, D. P., Sanchez, E. R., and Pratt, W. B. (2000) J. Biol. Chem. 275, 18054-18060). Here we have examined the nucleotide-bound states of the two essential chaperones in each step. We show that it is the ATP-bound state of hsp70 that interacts initially with the GR. After rapid priming and washing, the primed GR·hsp70 complex rapidly binds hsp90 in the second step reaction in a nucleotide-independent manner. The rate-limiting step is the ATP-dependent opening of the steroid binding cleft after hsp90 binding. This activating step requires the N-terminal ATP-binding site of hsp90, but we cannot establish any role for a C-terminal ATP-binding site in steroid binding cleft opening. The reported specific inhibitors of the C-terminal ATP site on hsp90 inhibit the generation of steroid binding, but they have other effects in this multiprotein system that could explain the inhibition.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

In the absence of hormone, steroid receptors are recovered from cells as multiprotein heterocomplexes containing a dimer of hsp90,¹ substoichiometric amounts of hsp70,² an acidic 23-kDa protein, p23, and a tetratricopeptide repeat domain protein, such as an immunophilin or protein phosphatase 5 (for review see Refs. 1 and 2). The receptor·hsp90 heterocomplexes can be formed by incubating immunoadsorbed, hsp90-free receptors with reticulocyte lysate (3, 4). Because the ligand binding domain (LBD) of the glucocorticoid receptor (GR) must be bound to hsp90 for the receptor to have high affinity steroid binding activity (1), incubation of the hsp90-free GR with reticulocyte lysate results in generation of steroid binding activity in direct proportion to the number of GR·hsp90 heterocomplexes that are assembled (5). Thus, generation of steroid binding activity provides a functional assay for proper heterocomplex assembly. Both biochemical data (6) and data derived from GR mutants (7, 8) support a model in which a hydrophobic cleft in the LBD is opened upon formation of the GR·hsp90 heterocomplex such that it can be accessed by steroid.

The assembly of receptor·hsp90 heterocomplexes is an ATP-dependent process that is carried out by a multiprotein system that is apparently ubiquitous among cells of the animal and plant kingdoms (1). The assembly system in reticulocyte lysate has been reconstituted (9), and a mixture of five purified proteins, hsp90, hsp70, Hop, hsp40, and p23, is now used to achieve optimal receptor·hsp90 heterocomplex assembly (10, 11). Hsp90 and hsp70 are both essential for opening the steroid binding cleft, and Hop, hsp40, and p23 act as co-chaperones that increase the rate or extent of GR·hsp90 heterocomplex assembly (12). Hop (hsp70/hsp90 organizing protein) binds via different TPR domains to hsp70 and hsp90 (13), bringing these two essential chaperones together into a machinery. The machinery as it is immunoadsorbed from reticulocyte lysate also contains the hsp70 co-chaperone hsp40 (10), and it has all of the activity required for converting the GR LBD to the steroid binding state (10, 14). All of the Hop and ~30% of the hsp90 in reticulocyte lysate exist in this hsp90·Hop·hsp70·hsp40 complex (15), which has been called the foldosome (16) or the hsp90/hsp70-based chaperone machinery. Receptor-bound hsp90 must achieve the ATP-bound conformation for the receptor to be in the steroid binding state (17), and when this state is achieved, p23, which binds only to ATP-bound hsp90 (18), binds dynamically to stabilize the receptor·hsp90 heterocomplex (19).

Acting separately, hsp70 and hsp90 are classically thought of as chaperones that bind to denatured regions of proteins to promote their refolding. However, there is no evidence that steroid receptors that have been stripped of hsp90 by mild salt treatment are in any way denatured prior to their reactivation by the hsp90/hsp70-based chaperone machinery. When hsp90 and hsp70 are acting together in this machinery, they act at a very focal site on the surface of the LBD (7) of the native GR to open the hydrophobic steroid binding cleft to access by hormone. In initial experiments to understand the mechanism of cleft opening and GR·hsp90 heterocomplex assembly, we have carried out two-step reactions in which the immunoadsorbed GR is first incubated with purified hsp70 and YDJ-1 (the yeast hsp40 homolog) in the presence of an ATP-regenerating system (20). This produces a GR·hsp70 complex that can be washed free of unbound hsp70 and then incubated with purified hsp90, Hop, and p23. In this first reaction the GR is "primed" to be activated by hsp90 during the second incubation. Both the initial priming step with hsp70 and the second activating step with hsp90 are ATP-dependent, and steroid binding activity is generated only during the second step (20). The two-step assembly differs somewhat from single step assembly when all five proteins are present simultaneously in the respect that the priming reaction with hsp70 is highly dependent upon the presence of YDJ-1 (21), whereas the single step assembly is less dependent upon YDJ-1 (12).

The conformation of both hsp70 (22) and hsp90 (23) is determined by their nucleotide binding state, and in both cases it is the ADP-bound conformation that possesses high affinity for binding hydrophobic substrates. To advance our understanding of the assembly process, it is important to determine what nucleotide binding state of hsp70 interacts initially with the GR and what state of hsp90 interacts with the primed GR·hsp70 complex. On the basis of indirect evidence, we have previously speculated that hsp70 binds to the GR in its ATP-bound form (20), and in this work, we demonstrate that it is the ATP-bound state of hsp70 that interacts initially with the hsp90-free GR. In the second assembly step, binding of hsp90 is rapid, and the rate-limiting step is the ATP-dependent opening of the steroid binding cleft. We do not detect any difference between the ATP-bound, the ADP-bound, and the unbound state of hsp90 in the ability to bind to the primed GR·hsp70 complex. In addition to the N-terminal ATP-binding site that acts as a conformational switch (23), hsp90 has a second ATP-binding site at the C terminus (24-27). Using inhibitors reported to be specific for one site or the other, we confirm a requirement of the N-terminal site for GR activation. However, when the GR is activated by the multiprotein machinery, we find that other effects of C-terminal inhibitors, such as inhibition of hsp70 function, do not permit determination of any requirement for ATP binding at the C-terminal site on hsp90.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Materials

[6,7-³H]Triamcinolone acetonide (38 Ci/mmol) and ¹²⁵I-conjugated goat anti-mouse and anti-rabbit IgGs were obtained from PerkinElmer Life Sciences. Untreated rabbit reticulocyte lysate was from Green Hectares (Oregon, WI). Protein A-Sepharose, novobiocin, cisplatin, and goat anti-mouse horseradish peroxidase conjugate were from Sigma. Complete-Mini protease inhibitor mixture was from Roche Molecular Biochemicals (Mannheim, Germany). The BuGR2 monoclonal IgG antibody against the GR was from Affinity Bioreagents (Golden, CO). The AC88 monoclonal IgG against hsp90 and the N27F3-4 anti-72/73-kDa hsp monoclonal IgG (anti-hsp70) were from StressGen (Victoria, BC, Canada). Escherichia coli-expressing human p23, YDJ-1, or Hop was kindly provided by Drs. David Toft (Mayo Clinic, Rochester, MN), Avrom Caplan (Mount Sinai School of Medicine), and David Smith (Mayo Clinic, Scottsdale, AZ), respectively. Hybridoma cells producing the FiGR monoclonal IgG against the GR were generously provided by Dr. Jack Bodwell (Dartmouth Medical School). Geldanamycin was obtained from the Drug Synthesis and Chemistry Branch of the Developmental Therapeutics Program, NCI, National Institutes of Health.

Methods

Expression of Mouse GR in Sf9 Cells-- Sf9 cells were grown in SFM900 II serum-free medium (Invitrogen) supplemented with Cytomax (Kemp Biotechnology, Rockville, MD) in suspension cultures maintained at 27 °C with continuous shaking (150 rpm). Cultures were supplemented with 0.1% glucose at infection and 24 h post-infection as described by Srinivasan et al. (28). Cells were harvested, washed in Hanks' buffered saline solution, resuspended in 1.5 volumes of buffer (10 mM Hepes, pH 7.35, 1 mM EDTA, 20 mM molybdate, 1 mM phenylmethylsulfonyl fluoride) with 1 tablet of Complete-Mini protease inhibitor mixture per 3 ml of buffer, and ruptured by Dounce homogenization. The lysate was then centrifuged at 100,000 × g for 30 min, and the supernatant was collected, aliquoted, flash-frozen, and stored at 70 °C.

Glucocorticoid Receptor Heterocomplex Reconstitution-- Receptors were immunoadsorbed from 50-µl aliquots of Sf9 cytosol by rotation for 2 h at 4 °C with 14 µl of protein A-Sepharose precoupled to 7 µl of FiGR ascites suspended in 200 µl of TEG buffer (10 mM TES, pH 7.6, 50 mM NaCl, 1 mM EDTA, 10% glycerol). Prior to incubation with various mixtures of purified proteins as noted, immunoadsorbed receptors were stripped of associated hsp90 by incubating the immunopellet for an additional 2 h at 4 °C with 300 µl of 0.5 M NaCl in TEG buffer. The pellets were then washed once with 1 ml of TEG buffer followed by a second wash with 1 ml of Hepes buffer (10 mM Hepes, pH 7.35). For single step assembly of GR·hsp90 heterocomplexes, FiGR immunopellets containing GR stripped of chaperones were incubated with the five-protein assembly system (20 µg of purified hsp90, 15 µg of purified hsp70, 0.6 µg of purified human Hop, 6 µg of purified p23, 0.4 µg of purified YDJ-1) adjusted to 50 µl with HKD buffer (10 mM Hepes, pH 7.35, 100 mM KCl, 5 mM dithiothreitol), containing 20 mM sodium molybdate and 5 µl of an ATP-regenerating system (50 mM ATP, 250 mM creatine phosphate, 20 mM magnesium acetate, and 100 units/ml creatine phosphokinase). The assay mixtures were incubated for 20 min at 30 °C with suspension of the pellets by shaking the tubes every 2 min. At the end of the incubation, the pellets were washed twice with 1 ml of ice-cold TEGM buffer (TEG with 20 mM sodium molybdate) and assayed for steroid binding capacity and for receptor-associated hsp90 and hsp70. For two-step assembly, stripped GR immune pellets were mixed with hsp70, YDJ-1, and the ATP-regenerating system in a final volume of 55 µl adjusted with HKD buffer. After 5 min of incubation at 30 °C, the pellets were washed twice with TEG buffer and once with Hepes buffer and then incubated for 20 min at 30 °C with the purified heterocomplex assembly system without hsp70. At the end of the second incubation, steroid binding was assayed.

Assay of Steroid Binding Capacity-- Immune pellets to be assayed for steroid binding were incubated overnight at 4 °C in 50 µl of HEM buffer (10 mM Hepes, pH 7.35, 1 mM EDTA, 20 mM molybdate) plus 50 nM [³H]triamcinolone acetonide. Samples were then washed three times with 1 ml of TEGM buffer and counted by liquid scintillation spectrometry. The steroid binding is expressed in counts per min of [³H]triamcinolone acetonide bound/FiGR immunopellet prepared from 50 µl of Sf9 cytosol.

Western Blotting-- To assay GR and associated proteins, immune pellets were resolved on 12% SDS-polyacrylamide gels and transferred to Immobilon-P membranes. The membranes were probed with 0.25 µg/ml BuGR for GR, 1 µg/ml AC88 for hsp90, or 1 µg/ml N27F3-4 for hsp70. The immunoblots were then incubated a second time with the appropriate ¹²⁵I-conjugated or horseradish peroxidase-conjugated counterantibody to visualize the immunoreactive bands.

Protein Purification-- hsp90 and hsp70 were purified from rabbit reticulocyte lysate by sequential chromatography on DE52, hydroxyapatite, and ATP-agarose as described previously (29). Human p23 was purified from 10 ml of bacterial lysate by chromatography on DE52 followed by hydroxyapatite chromatography as described (30). For purification of YDJ-1, bacterial sonicates were cleared by centrifugation, and YDJ-1 was purified by sequential chromatography on DE52 and hydroxyapatite as described previously (10). The bacterial expression of YDJ-1 has been described (31) as has the expression of human Hop (9). Purification of human Hop was carried out in a similar manner by sequential chromatography on DE52 and hydroxyapatite. In all cases the protein-containing fractions were identified by immunoblotting, and fractions from the final purification step were pooled, concentrated by Amicon filtration, dialyzed against HKD buffer, flash-frozen, and stored at 70 °C.

Determination of Nucleotide-bound States of hsp90-- Purified rabbit hsp90 was incubated with 5 mM ATP, 5 mM ADP, or with HKD buffer alone for 10 min at 30 °C. Following the incubation, 17 µg of hsp90 protein was subjected to four rounds of charcoal adsorption to remove free nucleotide. Bound nucleotides were extracted with ice-cold 0.4 N perchloric acid from purified hsp90. The ATP and ADP were then separated and quantitated by strong anion-exchange, high performance liquid chromatography using a Waters (Milford, MA) gradient system controlled by Millenium software as described previously (32). Briefly, samples were loaded onto a 5 µm Partisphere 4.6 × 250 mm SAX column (Whatman, Hillsboro, OR), and nucleotides were eluted with a linear gradient of ammonium phosphate buffer ranging from 0.15 (pH 3.6) to 0.60 M (pH 3.8) at a flow rate of 2 ml/min. Retention times were 7.7 min and 23.3 min for ADP and ATP, respectively. Nucleotides were identified by their absorbance between 240 and 350 nm using a Waters photodiode array detector. ADP and ATP were quantitated by comparison of their peak areas with that of a known amount of appropriate standard at wavelengths 254 and 281. The concentration of purified hsp90 was determined by Bradford assay. All perchloric acid extracts were filtered prior to the high performance liquid chromatography step, and after correcting for the amount of nucleotide removed during filtration, the percent of the protein that could be accounted for in the nucleotide-bound form was calculated. In the samples of Fig. 3, we can say that at least 60% of the hsp90 is converted to the ATP-bound form and at least 28% to the ADP-bound form.

Binding of Different States of hsp90 to the Primed GR·hsp70 Complex-- hsp90-stripped GR immune pellets were incubated with YDJ-1 and an ATP-regenerating system in the presence (primed) or absence (unprimed) of hsp70 for 5 min at 30 °C. The immunopellets were then washed once with TEG buffer and once with Hepes buffer. Purified hsp90 was preincubated for 10 min at 30 °C with 5 mM ATP, 5 mM ADP, or HKD buffer. The washed unprimed or primed GR·hsp70 pellets were then incubated for 30 s at 30 °C with each preparation of preincubated hsp90 in the presence of Hop, YDJ-1, p23, and molybdate and in the presence of 0.01% Nonidet P-40 to reduce nonspecific binding. The pellets were then washed three times with TG buffer (8 mM TES, 25 mM NaCl, and 10% glycerol (w/v) pH 7.6) and immunoblotted for GR, hsp90, and hsp70.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

The ATP-dependent State of hsp70 Binds to GR-- We have previously shown that our purified hsp70 is predominantly (~90%) in the ATP-bound state (21). Eukaryotic DnaJ homologs, like YDJ-1, bind to hsp70 to activate its ATPase activity and convert the chaperone to the ADP-dependent conformation, which binds hydrophobic peptide substrate (33). To determine whether it is the ATP-bound or ADP-bound state of hsp70 that initially interacts with the salt-stripped, hsp90-free GR, purified hsp70 was preincubated at 30 °C with an ATP-regenerating system to maintain the ATP-bound state, or it was incubated with YDJ-1 to convert it to the ADP-bound conformation. In the experiment of Fig. 1, immune pellets with stripped GR were incubated on ice with the preincubated samples of hsp70, and the pellets were then washed. It can be seen that untreated hsp70 bound to the receptor, and the amount of binding was increased when hsp70 was preincubated with the ATP-regenerating system (EGS). Preincubation of hsp70 with YDJ-1 essentially eliminated binding to the receptor, and we conclude that the ADP-bound conformation of hsp70 does not recognize the GR.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   The ATP-bound form of hsp70 interacts initially with the GR. Purified hsp70 was preincubated for 10 min at 30 °C with HKD buffer, with YDJ-1, with an ATP-regenerating system (EGS), or with both YDJ-1 and EGS as indicated. Nonimmune (NI) or immune (I) pellets with stripped receptors were then incubated for 20 min on ice with each preparation of preincubated hsp70. The pellets were then washed three times with Hepes buffer and immunoblotted for GR and hsp70.

These data are consistent with the ATP-dependent form of hsp70 being the form that interacts initially with the receptor. The receptor is primed in the initial step by incubating with both YDJ-1 and the ATP-regenerating system (21), and as shown in Fig. 1, hsp70 that is preincubated under this condition also binds to the receptor. We have previously assayed nucleotide in the complex after the priming reaction and find that approximately one-third of the GR-bound hsp70 is bound by ADP and approximately two-thirds is bound by ATP (21). The ATPase activity of hsp70 is minimal in the absence of K⁺ (34), and we have shown that priming of the receptor by hsp70 is K⁺-dependent (21). hsp70 is typically purified as a dimer (35), and ATP hydrolysis is reported to dissociate dimers to monomers (36). Our purified hsp70 behaves as a dimer on gel electrophoresis under nondenaturing conditions (15), and we have reported the stoichiometry of the primed GR·hsp70 complex to be ~1.1 molecule of hsp70 bound per molecule of GR (20).

These observations are consistent with a priming step in which purified hsp70 in its ATP-dependent conformation binds to the GR, probably initially as a dimer. During the priming reaction, which occurs very rapidly (within 1 min at 30 °C) (21), YDJ-1-stimulated ATPase activity converts it to a GR·hsp70 complex containing one (certainly no more than two) molecule of hsp70 that can now accept hsp90 in the subsequent activating step with the hsp70-free assembly system. The initial GR·hsp70 complex shown in Fig. 1 is different from the primed complex, in that no steroid binding activity is generated when it is incubated with the hsp70-free assembly system. Because this temperature-dependent and K⁺-dependent priming step requires both YDJ-1 and sustained high levels of ATP for optimal activity, we have suggested that the GR-bound hsp70 may ratchet between ATP- and ADP-dependent conformations as it primes the receptor (21). This is consistent with the observation (21) that both ATP- and ADP-bound hsp70 are recovered in the primed complex.

State of hsp90 Interacting with the Primed GR·hsp70 Complex-- We have previously proposed that hsp90 binds to the primed GR·hsp70 complex in its ADP-bound state (20). The purified hsp90 that we use is not bound by nucleotide (21), but this form may be the same as the ADP-bound state in terms of the overall conformation of hsp90 (18). Scheibel et al. (37) have calculated that about 70% of the hsp90 in cells would be in the ATP-bound form; thus, we wanted to find out what form of hsp90 binds to the primed GR·hsp70 complex. In the experiment of Fig. 2, purified hsp90 was preincubated at 30 °C with buffer, ATP, or ADP, and the mixtures were added to unprimed GR pellets or to primed GR·hsp70 complexes. After 30 s at 30 °C, the pellets were washed and immunoblotted. All three of the hsp90 preparations bound to the primed GR·hsp70 complex. Purified hsp90 that was preincubated with nucleotide under the same conditions was assayed for bound nucleotide. As shown in Fig. 3, our purified hsp90 is in nucleotide-free form, and preincubation with ATP or ADP converts a significant portion to the ATP- or ADP-bound form.

View larger version (44K): [in this window] [in a new window]   Fig. 2.   The nucleotide state of hsp90 does not affect the initial interaction with the primed GR·hsp70 complex. Unprimed GR (U) or primed GR·hsp70 complexes (P) were incubated for 30 s at 30 °C with purified hsp90 that was preincubated with buffer, ATP, or ADP in presence of Hop, YDJ-1, and p23 as described under "Methods." The pellets were washed and immunoblotted for GR, hsp90, and hsp70.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Nucleotide binding states of hsp90. Purified hsp90 was incubated for 10 min at 30 °C with 5 mM ATP or ADP or buffer alone, and following charcoal adsorption, protein-bound nucleotide was assayed by high performance liquid chromatography as described under "Methods."

In Fig. 2, it appears that somewhat less of the ADP-preincubated sample is bound, but over several experiments we cannot detect any difference in the amount of hsp90 that binds to the primed GR·hsp70 complex when the hsp90 is added in the unbound, ADP-bound, or ATP-bound state. We have shown previously that unbound, ADP-bound, and ATP-bound preparations of hsp90 all yield the same rate of generation of steroid binding activity when added to the second step reaction mixture (21). Thus, the primed GR·hsp70 complex does not appear to bind preferentially a particular nucleotide-bound state of hsp90 when it is added under the second step conditions where Hop is present. Inasmuch as purified Hop has been reported to bind preferentially to hsp90 that is in the unbound or ADP-bound state (38), this result was unexpected.

Binding of hsp90 to the Primed GR·hsp70 Complex Is Rapid, and Cleft Opening Is Rate-limiting-- To determine what step is rate-limiting for assembly of steroid-binding GR·hsp90 complexes, we performed a time course of the second step reaction. In the experiment of Fig. 4A, unprimed receptors and primed GR·hsp70 complexes were incubated under second step conditions with hsp90, Hop, YDJ-1, and p23. At the indicated times, samples were assayed for both steroid binding and hsp90 binding. It is clear that binding of hsp90 to the primed GR·hsp70 complex is rapid, and steroid binding activity increases more slowly. As reported previously (20), hsp70 dissociates from the receptor during the second step reaction. In the procedure of Fig. 4A, the pellets are washed after the second step incubation and then incubated with steroid to assay steroid binding activity. This procedure assays the formation of stable GR·hsp90 heterocomplexes that remain intact during subsequent incubation with steroid. To be sure that we measure the rate of steroid binding cleft opening, we carried out the second step reaction at 30 °C in the presence of steroid (Fig. 4B). Under these conditions, as soon as the steroid binding cleft is opened, the steroid enters and binds, yielding a direct estimate of the rate of cleft opening. Steroid binding is generated only when the ATP regenerating system is present. Thus, cleft opening is ATP-dependent and is the rate-limiting step.

View larger version (29K): [in this window] [in a new window]   Fig. 4.   hsp90 binds rapidly to GR·hsp70 complexes, and cleft opening is rate-limiting. A, time course of second step reaction in the absence of steroid. Unprimed GR (lanes 1, 3, and 5) and primed GR·hsp70 complexes (lanes 2, 4, and 6) were incubated at 30 °C with the purified activating system (hsp90, Hop, YDJ-1, p23) in the presence of 0.01%Nonidet P-40 and an ATP-regenerating system. At the indicated times, samples were washed three times with 10 mM Hepes and incubated with [3H]triamcinolone acetonide for assay of steroid binding and immunoblotted for GR, hsp90, and hsp70. The solid squares represent steroid binding by primed GR·hsp70 complexes, and open squares represent unprimed receptors. B, time course of the second step reaction in the presence of steroid. Conditions are the same as in A except that 50 nM [3H]triamcinolone acetonide was present during the assembly reaction.

Inhibition of GR Activation with Nucleotide Site Inhibitors-- Although the second step in which the primed GR·hsp70 complex is activated by hsp90 to the steroid binding state is ATP-dependent (20), it is not known whether one ATP binding site or both sites on hsp90 are involved in opening the steroid binding cleft. In the experiment of Fig. 4, primed GR·hsp70 complexes were incubated with the hsp70-free activating system in the presence of three hsp90 nucleotide site inhibitors, geldanamycin, novobiocin, and cisplatin. All three compounds inhibited GR activation by hsp90 in the second step reaction.

The natural product inhibitors geldanamycin and radicicol bind specifically to the atypical N-terminal nucleotide-binding site on hsp90 (23, 39), and the crystal structure of the N-terminal hsp90 fragment complexed with geldanamycin or ATP has been reported (40, 41). Geldanamycin ablates glucocorticoid binding activity in intact cells (42) and prevents cell-free activation of GR to the steroid binding state by reticulocyte lysate (43) and by the purified five-protein system (12). Geldanamycin does not affect formation of a primed GR·hsp70 complex (data not shown), but it blocks receptor activation when it is present during the second step of assembly with hsp90 (Fig. 5), confirming a requirement for N-terminal nucleotide binding in GR activation.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Inhibition of hsp90-dependent activation of the primed GR by nucleotide site-specific inhibitors. Stripped GR immune pellets were incubated for 5 min at 30 °C with hsp70, YDJ-1, and an ATP-regenerating system, and the pellets were washed to remove the free proteins. The primed GR·hsp70 complexes (Pr) were then incubated for 20 min at 30 °C with the purified activating system (AS) consisting of hsp90, Hop, YDJ-1, and p23 in the presence of 10 µM geldanamycin (GA), 10 mM novobiocin (NB), or 100 µM cisplatin (CP). The pellets were washed and incubated with [3H]triamcinolone acetonide to determine steroid binding activity.

Novobiocin and some related coumarins have been shown to interact with hsp90 in vitro and to cause increased turnover in vivo of hsp90 client proteins (e.g. p60^v-src, Raf-1, p185^erbB2), much like geldanamycin and radicicol (44). By deletion analysis, the novobiocin-binding site was located to a region in the C terminus of the chaperone that also binds ATP, and ATP and novobiocin competed with each other for binding (26). Although it was thought that novobiocin was a specific inhibitor at the C-terminal ATP-binding site (26), Soti et al. (27) have recently reported that novobiocin disrupts N-terminal nucleotide binding as well. As shown in Fig. 5, 10 mM novobiocin blocks GR activation when it is present during the second assembly step.

Millimolar concentrations of novobiocin have been used to study effects on hsp90 client proteins in vivo (44) and to study effects on hsp90 directly in vitro (27). As shown in Fig. 6A, half-maximal inhibition of GR activation is achieved with 1-2 mM novobiocin when it is present during the second step of assembly. As shown in Fig. 6B, the presence of novobiocin during single-step GR·hsp90 heterocomplex assembly with the five-protein system, inhibited the binding of hsp70 to the GR as well as the binding of hsp90. Thus, novobiocin appears to inhibit GR activation by the multiprotein system by affecting hsp70 as well as hsp90, although hsp90 appears to be somewhat more sensitive in that its binding is blocked at a somewhat lower concentration.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Concentration dependence of novobiocin on GR activation in two-step and single step assembly. A, two-step assembly. Primed GR·hsp70 complexes, prepared as described in the legend to Fig. 4, were incubated for 20 min at 30 °C with the purified activating system without hsp70 in the presence of increasing concentrations of novobiocin. The pellets were then washed and incubated with [3H]triamcinolone acetonide to determine steroid binding activity (expressed as percent of control sample without novobiocin). B, single step assembly. Stripped GR immune pellets were incubated for 20 min at 30 °C with the complete five-protein assembly system and various concentrations of novobiocin. Samples were then washed and assayed for steroid binding activity and immunoblotted for GR, hsp90, and hsp70. Str, stripped GR pellet alone.

The antineoplastic drug cisplatin has been reported to inhibit the chaperone activity of hsp90 in vitro via an interaction with the C terminus of the chaperone (25). In contrast to novobiocin, which disrupted both C- and N-terminal nucleotide binding by hsp90, Soti et al. (27) found cisplatin to be a selective C-terminal nucleotide competitor. The model that has evolved from studies with cisplatin is that nucleotide binding to the N-terminal site causes a conformational change that unmasks a C-terminal binding site that is cryptic in the nucleotide-free, full-length hsp90 (27). Selective inhibition of this C-terminal site by cisplatin might thus provide a pharmacological tool for dissecting contributions of hsp90 C-terminal nucleotide binding to opening of the steroid binding cleft in the GR.

The effect of cisplatin on single step GR·hsp70 heterocomplex assembly is shown in Fig. 7. It can be seen from the immunoblot in Fig. 7A that, as the concentration of cisplatin is increased, cross-linking of GR-bound proteins occurs, as indicated by the generation of cross-linked hsp90-hsp90 dimers. At high drug concentrations, there is a marked loss of recovery of GR, hsp90 and hsp70 migrating at normal M_r for these denaturing gel conditions. Not only are GR-bound hsp90 dimers being cross-linked, but the receptor itself is being cross-linked to its accompanying chaperones, as indicated by the generation of a cross-linked GR-hsp90 species. As shown in Fig. 7B, there appears to be a close parallel between the concentration dependence for cisplatin inhibition of conversion of GR to the steroid binding state and generation of cross-linked hsp90 dimers bound to the receptor. Thus, although cisplatin is a specific inhibitor for the C-terminal nucleotide site on hsp90, the cross-linking that occurs during assembly by the multiprotein chaperone system does not permit any conclusion that ATP binding to the C-terminal site is involved in opening of the steroid binding cleft in the GR.

View larger version (26K): [in this window] [in a new window]   Fig. 7.   Concentration dependence of cisplatin on GR activation in single step assembly. Stripped GR immune pellets were incubated for 20 min at 30 °C with the complete five-protein assembly system and various concentrations of cisplatin. Samples were then washed and assayed for steroid binding and immunoblotted for GR, hsp90, and hsp70. A, immunoblot. Lane 1, stripped GR; lanes 2-7, stripped GR incubated with the assembly system without cisplatin (lane 2) and with cisplatin at 2 µM (lane 3), 5 µM (lane 4), 10 µM (lane 5), 50 µM (lane 6), and 100 µM (lane 7). The immunoblot above the GR was blotted with AC88 to show cross-linked hsp90 species. Although not shown here, this immunoblot was stripped and probed with BuGR to identify GR in the top band. B, comparison of the effect on steroid binding versus cross-linking. The immunoblot of A was scanned in a densitometer to determine the relative amounts of hsp90 monomer () and hsp90 dimer (), which are plotted relative to steroid binding activity ().

	ACKNOWLEDGEMENTS

We thank David Smith, Avrom Caplan, David Toft, and Jack Bodwell for providing reagents used in this work, and we thank Peter Csermely for his very helpful dicussion.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant DK31573 (to W. B. P.) and CA76581 (to D. S. S.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence addressed: Dept. of Pharmacology, The University of Michigan Medical School, 1301 Medical Science Research Building III, Ann Arbor, MI 48109-0632. Tel.: 734-764-5414; Fax: 734-763-4450.

Published, JBC Papers in Press, July 1, 2002, DOI 10.1074/jbc.M204164200

² In this paper, we will use the term hsp70 collectively to refer to both the heat shock-induced hsp70 and the constitutively expressed heat shock cognate hsc70.

	ABBREVIATIONS

The abbreviations used are: hsp, heat shock protein; GR, glucocorticoid receptor; LBD, ligand binding domain; Hop, hsp70/hsp90 organizing protein; TES, 2-{[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]amino}ethanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				L. Farhana, M. I. Dawson, M. Leid, L. Wang, D. D. Moore, G. Liu, Z. Xia, and J. A. Fontana Adamantyl-Substituted Retinoid-Related Molecules Bind Small Heterodimer Partner and Modulate the Sin3A Repressor Cancer Res., January 1, 2007; 67(1): 318 - 325. [Abstract] [Full Text] [PDF]

				A. C. Y. Fan, M. K. Bhangoo, and J. C. Young Hsp90 Functions in the Targeting and Outer Membrane Translocation Steps of Tom70-mediated Mitochondrial Import J. Biol. Chem., November 3, 2006; 281(44): 33313 - 33324. [Abstract] [Full Text] [PDF]

				E. H Stolte, B M L. V. van Kemenade, H. F J Savelkoul, and G. Flik Evolution of glucocorticoid receptors with different glucocorticoid sensitivity. J. Endocrinol., July 1, 2006; 190(1): 17 - 28. [Abstract] [Full Text] [PDF]

				L. A. Moraes, M. J. Paul-Clark, A. Rickman, R. J. Flower, N. J. Goulding, and M. Perretti Ligand-specific glucocorticoid receptor activation in human platelets Blood, December 15, 2005; 106(13): 4167 - 4175. [Abstract] [Full Text] [PDF]

				M. O. Leonard, C. Godson, H. R. Brady, and C. T. Taylor Potentiation of Glucocorticoid Activity in Hypoxia through Induction of the Glucocorticoid Receptor J. Immunol., February 15, 2005; 174(4): 2250 - 2257. [Abstract] [Full Text] [PDF]

				S. S. Billecke, D. I. Draganov, Y. Morishima, P. J. M. Murphy, A. Y. Dunbar, W. B. Pratt, and Y. Osawa The Role of hsp90 in Heme-dependent Activation of Apo-neuronal Nitric-oxide Synthase J. Biol. Chem., July 16, 2004; 279(29): 30252 - 30258. [Abstract] [Full Text] [PDF]

				M. C. Rosenhagen, C. Soti, U. Schmidt, G. M. Wochnik, F. U. Hartl, F. Holsboer, J. C. Young, and T. Rein The Heat Shock Protein 90-Targeting Drug Cisplatin Selectively Inhibits Steroid Receptor Activation Mol. Endocrinol., October 1, 2003; 17(10): 1991 - 2001. [Abstract] [Full Text] [PDF]

				E. Adinolfi, M. Kim, M. T. Young, F. Di Virgilio, and A. Surprenant Tyrosine Phosphorylation of HSP90 within the P2X7 Receptor Complex Negatively Regulates P2X7 Receptors J. Biol. Chem., September 26, 2003; 278(39): 37344 - 37351. [Abstract] [Full Text] [PDF]

				M. J. Paul-Clark, F. Roviezzo, R. J. Flower, G. Cirino, P. D. Soldato, I. M. Adcock, and M. Perretti Glucocorticoid Receptor Nitration Leads to Enhanced Anti-Inflammatory Effects of Novel Steroid Ligands J. Immunol., September 15, 2003; 171(6): 3245 - 3252. [Abstract] [Full Text] [PDF]

				P. J. M. Murphy, Y. Morishima, H. Chen, M. D. Galigniana, J. F. Mansfield, S. S. Simons Jr., and W. B. Pratt Visualization and Mechanism of Assembly of a Glucocorticoid Receptor{middle dot}Hsp70 Complex That Is Primed for Subsequent Hsp90-dependent Opening of the Steroid Binding Cleft J. Biol. Chem., September 12, 2003; 278(37): 34764 - 34773. [Abstract] [Full Text] [PDF]

				W. B. Pratt and D. O. Toft Regulation of Signaling Protein Function and Trafficking by the hsp90/hsp70-Based Chaperone Machinery Experimental Biology and Medicine, February 1, 2003; 228(2): 111 - 133. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/37/33698    most recent M204164200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services