Differential Effects of the hsp70-binding Protein BAG-1 on Glucocorticoid Receptor Folding by the hsp90-based Chaperone Machinery*

The heat shock protein hsp70/hsc70 is a required component of a five-protein (hsp90, hsp70, Hop, hsp40, and p23) minimal chaperone system reconstituted from reticulocyte lysate that forms glucocorticoid receptor (GR)·hsp90 heterocomplexes. BAG-1 is a cofactor that binds to the ATPase domain of hsp70/hsc70 and that modulates its chaperone activity. Inasmuch as BAG-1 has been found in association with several members of the steroid receptor family, we have examined the effect of BAG-1 on GR folding and GR·hsp90 heterocomplex assembly. BAG-1 was present in reticulocyte lysate at a BAG-1:hsp70/hsc70 molar ratio of ∼0.03, and its elimination by immunoadsorption did not affect GR folding and GR·hsp90 heterocomplex assembly. At low BAG-1:hsp70/hsc70 ratios, BAG-1 promoted the release of Hop from the hsp90-based chaperone system without inhibiting GR·hsp90 heterocomplex assembly. However, at molar ratios approaching stoichiometry with hsp70, BAG-1 produced a concentration-dependent inhibition of GR folding to the steroid-binding form with corresponding inhibition of GR·hsp90 heterocomplex assembly by the minimal five-protein chaperone system. Also, there was decreased steroid-binding activity in cells that were transiently or stably transfected with BAG-1. These observations suggest that, at physiological concentrations, BAG-1 modulates assembly by promoting Hop release from the assembly complex; but, at concentrations closer to those in transfected cells and some transformed cell lines, hsp70 is continuously bound by BAG-1, and heterocomplex assembly is blocked.

A number of signaling proteins, including several members of the nuclear receptor family, the dioxin receptor, nitric-oxide synthase, and several protein kinases, exist in cytosolic complexes with the ubiquitous and abundant heat shock protein (hsp) 1 hsp90 (for review, see Refs. 1 and 2). These signaling protein⅐hsp90 heterocomplexes can be formed under cell-free conditions by incubating the immunoadsorbed proteins with reticulocyte lysate (3,4). A minimal heterocomplex assembly system has been reconstituted (5)(6)(7)(8)(9); and five proteins, including hsp90, hsp70, 2 Hop (60-kDa hsp organizer protein), hsp40 and p23, participate in the ATP/Mg 2ϩ -dependent and K ϩ -dependent assembly process (for review of heterocomplex assembly, see Refs. 10 and 11). Hip and BAG-1 (Bcl-2-associated gene product-1) are co-chaperones of hsp70 that are potential participants in or regulators of this multiprotein hsp90-based chaperone system. BAG-1 is an hsp70-binding protein that associates with the ATPase domain of the molecular chaperone with a K D of ϳ1-10 nM (12). BAG-1 was originally cloned from a mouse library as a Bcl-2-binding protein and shown to have anti-apoptotic activity (13). Subsequently, BAG-1 was shown to form complexes with some signaling proteins, including receptors for hepatocyte growth factor and platelet-derived growth factor (14), the serine/threonine protein kinase Raf-1 (15), and the retinoic acid receptor (16). An isoform of the human BAG-1 protein called RAP46 (46-kDa receptor-associated protein; BAG-1M) was cloned and shown to bind to members of the nuclear receptor family, including the glucocorticoid receptor (GR) (17). The ability of BAG-1 (18,20) and RAP46 (BAG-1M) (19) to interact with hsp70 suggests an explanation for their association with the various signaling proteins.
It is now clear that there is a conserved family of BAG-1related proteins (12), and three forms of BAG-1 itself exist as a result of alternative initiation of translation within a common mRNA (21). The originally identified BAG-1 protein is the shortest, migrating at ϳ32 kDa. Human ϳ46-kDa RAP46 is the same as BAG-1M, and we will refer to it as BAG-1M in this work. A form migrating at ϳ58 kDa is called BAG-1L (22). The effect of BAG-1 on steroid receptor function appears to depend upon the isoform that is expressed. For example, BAG-1M (RAP46) was found to be a negative regulator of GR activity (23), whereas BAG-1L, but not BAG-1 or BAG-1M, forms complexes with the androgen receptor and enhances androgen receptor-mediated transactivation of a reporter gene (22).
hsp70 is a protein chaperone possessing an amino-terminal domain that binds the nucleotides ADP and ATP and a carboxyl-terminal domain that binds peptide, with the binding of ADP versus ATP being coupled to its peptide-binding activity (for review, see Ref. 24). hsp70 possesses weak ATPase activity and an intrinsic ADP-ATP exchange activity (25). BAG-1, BAG-1M, and BAG-1L have been reported to bind to the hsp70 ATPase domain (19,26,27). BAG-1 accelerates the release of ADP from hsp70 (20,28), acting much like bacterial GrpE in the DnaK * This work was supported by National Institutes of Health Grants DK31573 (to W. B. P.) and CA67329 (to J. C. R.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶  30). However, unlike GrpE, which enhances the folding activity of DnaK, BAG-1 forms a ternary complex with hsp70 and a non-native substrate, uncoupling the ATP-dependent release of the hsp70-associated substrate (31).
Because hsp70 is absolutely required for formation of heterocomplexes between the GR and hsp90 and for changing the folding state of the GR hormone-binding domain to form the high affinity steroid-binding conformation (32), we have asked in this work how BAG-1 affects GR⅐hsp90 heterocomplex assembly and GR folding by the multiprotein hsp90-based chaperone system. BAG-1 itself was the dominant BAG-1 isoform in rabbit reticulocyte lysate, where it was present at a low concentration with a BAG-1:hsp70 molar ratio of ϳ0.03. Elimination of BAG-1 from lysate had little or no effect on its ability to assemble GR⅐hsp90 heterocomplexes and to generate steroidbinding activity. Addition of low concentrations of purified recombinant BAG-1 to a five-protein minimal assembly system consisting of purified hsp90, hsp70, Hop, YDJ-1, and p23 promoted release of Hop from assembly complexes without inhibiting GR⅐hsp90 heterocomplex assembly. At concentrations approaching stoichiometry with hsp70, however, BAG-1 produced a concentration-dependent inhibition of GR folding to the steroid-binding form, with corresponding inhibition of GR⅐hsp90 heterocomplex assembly. Overexpression of BAG-1 resulted in decreased glucocorticoid-binding activity. Thus, at the low levels that exist in reticulocyte lysate, BAG-1 appears to modify heterocomplex assembly by promoting Hop release from the hsp90-based chaperone system; but at higher levels that may be closer to those achieved in transfected cells and some types of transformed cells (27), it has the potential to regulate GR folding in a negative manner. [6, H]Triamcinolone acetonide (42.8 Ci/mmol) and 125 I-conjugated goat anti-mouse and anti-rabbit IgGs were obtained from NEN Life Science Products. Untreated rabbit reticulocyte lysate was from Green Hectares (Oregon, WI). Charcoal-stripped bovine calf serum, protein A-Sepharose, and goat anti-mouse horseradish peroxidase conjugates were from Sigma, and donkey anti-rabbit IgG was from Pierce. Trans-Fast reagent was from Promega (Madison, WI), and Complete-Mini protease inhibitor mixture was from Roche Molecular Biochemicals (Mannheim, Germany). The BuGR2 monoclonal IgG against the GR and the 2G6 monoclonal anti-p48 IgG against Hip were from Affinity Bioreagents (Golden, CO). The AC88 monoclonal IgG against hsp90, the N27F3-4 anti-72/73-kDa hsp monoclonal IgG (anti-hsp70), and the anti-hsp40 rabbit polyclonal antibody were from Stressgen Biotech Corp. (Victoria, British Columbia, Canada). The JJ3 monoclonal IgG against p23 and Escherichia coli cells expressing human p23 were gifts from Dr. David Toft (Mayo Clinic). E. coli cells expressing YDJ-1 were a gift from Dr. Avrom Caplan (Mount Sinai School of Medicine). The DS14F5 monoclonal IgG against Hop and E. coli cells expressing Hop were kindly provided by Dr. David F. Smith (University of Nebraska Medical School). The BAG-1 (C-16) affinity-purified rabbit polyclonal antibody was from Santa Cruz Biotechnologies (Santa Cruz, CA). Hybridoma cells producing the FiGR monoclonal IgG against the GR were generously provided by Dr. Jack Bodwell (Dartmouth Medical School). The baculovirus for mouse GR was kindly provided by Dr. Edwin Sanchez (Medical College of Ohio). Construction of the pcDNA3-hu-BAG-1 plasmid for expression of human BAG-1 has been described previously (33), as was the GM-701 human fibroblast cell line stably expressing FLAG-BAG-1 (26). The cDNA for expressing GFP-GR, described previously (34), was provided by Dr. Paul Housley (University of South Carolina School of Medicine).

Methods
Cell Culture and Transfection-Monkey kidney COS-7 cells were grown in DMEM supplemented with 10% fetal bovine serum. When cells were confluent, they were rinsed three times with serum-free DMEM and incubated for an additional hour in DMEM with 5% serum. Cells were then transfected with 2 g/ml GFP-GR cDNA and 3 g/ml pcDNA3 vector or pcDNA3-hu-BAG-1. The DNAs were preincubated for 10 min at room temperature with 6 l of TransFast reagent/g of DNA in 200 l of DMEM, and the mixtures were added to the cell cultures. After 1 h at 37°C, the medium was replaced with DMEM containing 10% fetal bovine serum, and the cells were cultured for 24 h. The medium was then replaced by 10% charcoal-stripped bovine calf serum for an additional 24 h of incubation. The cells were washed three times with Earle's balanced saline and suspended in 1 volume of buffer containing 10 mM Hepes, pH 7.35, 1 mM EDTA, 20 mM sodium molybdate, 20 mM sodium vanadate, and 1 tablet of Complete-Mini protease inhibitor mixture/3 ml of buffer. Cells were ruptured by Dounce homogenization and centrifuged for 1 h at 100,000 ϫ g to prepare the cytosol.
GM-701 human fibroblasts (the parental line and a subline stably transfected with FLAG-tagged murine BAG-1) were grown in DMEM supplemented with 10% fetal bovine serum. When cells were ϳ80% confluent, they were harvested, and the cytosol was prepared as described for COS-7 cells.
Expression of Mouse GR in Sf9 Cells-Sf9 cells were grown in SFM900 II serum-free medium (Life Technologies, Inc.) supplemented with Cytomax (Kemp Biotechnology, Rockville, MD) in suspension cultures maintained at 27°C with continuous shaking (150 rpm). Cultures were infected in log phase of growth with recombinant baculovirus at a multiplicity of infection of 3.0. Cultures were supplemented with 1% glucose at infection and 24 h post-infection as described by Srinivasan et al. (35). Cells were harvested, washed in Hanks' balanced saline solution, resuspended in 1.5 volumes of buffer (10 mM Hepes, pH 7.5, 1 mM EDTA, 20 mM molybdate, and 1 mM phenylmethylsulfonyl fluoride), and ruptured by Dounce homogenization. The lysate was then centrifuged at 100,000 ϫ g for 30 min, and the supernatant was collected, aliquoted, flash-frozen, and stored at Ϫ70°C.
Immunoadsorption of GR and Hop-Receptors were immunoadsorbed from 50-l aliquots of Sf9 cytosol by rotation for 2 h at 4°C with 14 l of protein A-Sepharose precoupled to 7 l of FiGR ascites suspended in 300 l of TEG buffer (10 mM TES, pH 7.6, 50 mM NaCl, 4 mM EDTA, and 10% glycerol). Prior to incubation with reticulocyte lysate or various mixtures of purified proteins as noted, immunoadsorbed receptors were stripped of associated hsp90 by incubating the immune pellet for an additional 2 h at 4°C with 1 ml of 0.5 M KCl in TEG buffer. The pellets were then washed once with 1 ml of TEG buffer, followed by a second wash with 1 ml of 10 mM Hepes, pH 7.4. For immunoadsorption of Hop, 300-l aliquots of reticulocyte lysate were immunoadsorbed to 5 l of protein A-Sepharose prebound with DS14F5 antibody against Hop or with nonimmune mouse IgG. The samples were rotated at 4°C for 2 h, and immune pellets were washed three times with 1 ml of TEGM buffer (TEG buffer with 20 mM sodium molybdate).
Glucocorticoid Receptor Heterocomplex Reconstitution-FiGR immune pellets containing GRs stripped of hsp90 were incubated with 50 l of rabbit reticulocyte lysate, with combinations of lysate DE52 fractions A-C, or with various mixtures of proteins (20 g of purified hsp90, 20 g of purified hsp70, 2 l of lysate from bacteria expressing Hop, 6 g of purified p23, and 2 l of lysate from bacteria expressing YDJ-1) and adjusted to 50 l with HKD buffer (10 mM Hepes, 100 mM KCl, and 5 mM dithiothreitol, pH 7.35) containing 20 mM sodium molybdate and 5 l of an ATP-regenerating system (50 mM ATP, 250 mM creatine phosphate, 20 mM MgOAc, and 100 units/ml creatine phosphokinase). The assay mixtures were incubated for 30 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 and assayed for steroid-binding capacity and, in some experiments, for receptor-associated proteins.
Assay of Steroid-binding Capacity-Immune pellets to be assayed for steroid binding were incubated overnight in 50 l 10 mM Hepes, pH 7.5, 1 mM EDTA, 20 mM molybdate plus 50 nM [ 3 H]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 as counts/min of [ 3 H]triamcinolone acetonide bound per FiGR immune pellet prepared from 50 l of Sf9 cytosol.
For assay of steroid binding in the cytosol from COS-7 cells or GM-701 cells, 100 l of cytosol was incubated overnight with 50 nM [ 3 H]triamcinolone acetonide (Ϯ1000-fold excess of radioinert dexamethasone). Bound steroid was separated from free steroid by adding 1.5 volumes of dextran-coated charcoal suspension (1% (w/v) charcoal and 0.2% (w/v) dextran in 10 mM Hepes and 1 mM EDTA, pH 7.35). The radioactivity in the supernatant was assayed, and the specific binding was normalized for the cytosol protein concentration.
Western Blotting-To assay the GR and associated proteins or Hop and associated proteins, immune pellets were resolved on 10% SDSpolyacrylamide gels and transferred to Immobilon-P membranes. The membranes were probed with 2 g/ml BuGR2 for GR, 1 g/ml AC88 for hsp90, 1 g/ml N27F3-4 for hsp70, 0.1% DS14F5 mouse ascites for Hop, and 0.5% anti-hsp40 antibody or 0.1% JJ3 mouse ascites for p23. The immunoblots were then incubated a second time with the appropriate 125 I-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, hydroxylapatite, and ATP-agarose as described previously (32). Human p23 (36) was purified from 10 ml of bacterial lysate by chromatography on DE52 as described (8), followed by hydroxylapatite chromatography. Fractions containing p23 were identified by immunoblotting, pooled, concentrated by Amicon filtration to ϳ1.5 ml, dialyzed against HKD buffer, aliquoted, and stored at Ϫ70°C. The bacterial expression of YDJ-1 has been described previously (37,38), as has the expression of human hsp organizer protein (Hop) (5). In this work, bacterial lysate expressing YDJ-1 or Hop was added to the heterocomplex assembly system without purification of the recombinant proteins. Murine BAG-1 was expressed as a GST fusion protein in E. coli and purified by adsorption to glutathione-agarose, followed by thrombin cleavage and ion-exchange and Superdex-75 gel filtration chromatography as described previously (20). Purified BAG-1 (3 mg/ml) was stored at Ϫ70°C in 20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.1% ␤-mercaptoethanol, and 1 mM EDTA. After unfreezing, the solution was cleared of insoluble protein by centrifugation at 100,000 ϫ g.

BAG-1 Exists in Multiprotein hsp
Heterocomplexes-Hop is a 60-kDa protein that interacts with hsp70 and hsp90 (39) via separate binding sites (40), acting as an adaptor protein (41) in the hsp90-based chaperone machinery. We have shown previously that immunoadsorption of Hop from reticulocyte lysate is accompanied by co-immunoadsorption of hsp90, hsp70, and hsp40 as an hsp90⅐Hop⅐hsp70⅐hsp40 heterocomplex (8). This complex immunoadsorbed from reticulocyte lysate or prepared from purified proteins is able to convert the GR to the high affinity steroid-binding conformation (6,8). Formation of a heterocomplex between hsp90 and the hormone-binding domain of the GR is required to produce the proper folding state for steroid binding (42,43).
In Fig. 1, we investigated whether either BAG-1 or Hip is present in the multiprotein GR⅐hsp90 heterocomplex assembly machine that we call a foldosome (44). As shown in Fig. 1, immunoadsorption of rabbit reticulocyte lysate with a monoclonal antibody against Hop yielded co-adsorption of hsp90 and hsp70 as well as the hsp70-associated proteins hsp40, Hip, and BAG-1. hsp70 possesses separate non-overlapping binding sites for Hop, hsp40, and Hip (18). However, Hip and BAG-1 compete for binding to the ATPase domain of hsp70 (28), and they must exist in separate hsp90⅐Hop⅐hsp70⅐hsp40 heterocomplexes. It should be noted that, to visualize Hip and BAG-1, the autoradiograms for these proteins required a much longer exposure time than for the major proteins in Fig. 1. Thus, it seems that only a fraction of the multiprotein hsp heterocom-plexes isolated from reticulocyte lysate contain either Hip or BAG-1.
Fractionation of Reticulocyte Lysate-In Fig. 2, the rabbit reticulocyte lysate proteins were fractionated by chromatography on DE52. The dominant BAG-1 isoform is ϳ32-kDa BAG-1, which eluted early in the salt gradient with another slower migrating immunoreactive species that we refer to as BAG-1M ( Fig. 2A). A trace amount of an immunoreactive species migrated as expected for BAG-1L, but eluted at higher salt concentrations, indicating that it is more acidic than BAG-1 and BAG-1M. Reticulocyte lysate contained very little BAG-1 relative to both Hip and the established components of the hsp90based chaperone system: hsp90, hsp70, Hop, hsp40, and p23 ( Fig. 2A). The reticulocyte lysate proteins were combined into three DE52 subfractions designated A, B, and C (Fig. 2B).
In Fig. 3A, immune pellets containing GRs stripped of hsp90 were incubated with DE52 subfractions A-C in various combinations. Maximum generation of steroid-binding activity required all three fractions of reticulocyte lysate. Because of a low level of cross-contamination of components between fractions, some steroid-binding activity was generated when just two subfractions were present (Fig. 3A). As shown in Fig. 3B, substitution of fraction B with purified hsp90 and of fraction C with purified p23 yielded a reconstituted system that was entirely dependent upon the presence of lysate fraction A.
Removal of BAG-1 from Lysate Fraction A-To determine if BAG-1 in reticulocyte lysate was affecting (either negatively or positively) GR folding, BAG-1 was eliminated from lysate fraction A by immunoadsorption. As shown in Fig. 4A, adsorption of fraction A with protein A-Sepharose prebound with anti-BAG-1 antibody (I) eliminated BAG-1 from fraction A. Substitution of fraction A with either nonimmune IgG-adsorbed (A(NI)) or anti-BAG-1 antibody-adsorbed (A(I)) fraction A yielded comparable levels of GR steroid-binding activity and resulted in assembly of similar GR⅐hsp90 heterocomplexes (Fig.  4B). Thus, we suggest that, at the levels that are present in reticulocyte lysate, BAG-1 does not affect the ability of the GR to be folded to the steroid-binding state by the hsp90-based chaperone system.

Effects of a High Concentration of BAG-1 on GR Folding and Heterocomplex
Assembly-To determine the effect of BAG-1 on GR folding at concentrations approaching stoichiometry with hsp70, we performed experiments in which stripped GR immune pellets were incubated with a minimal folding system (8) consisting of purified hsp90, hsp70, and p23 and lysate from bacteria expressing Hop and YDJ-1 in the presence of increas- FIG. 1. BAG-1 is a component of the foldosome complex in reticulocyte lysate. Rabbit reticulocyte lysate was incubated for 30 min at 30°C in the presence of an ATP-regenerating system. Samples were immunoadsorbed with nonimmune IgG or with DS14F5 antibody against Hop; the immune pellets were washed three times with 1 ml of Hepes; and pellet-associated proteins were assayed by immunoblotting. ing concentrations of purified recombinant murine BAG-1 (Fig.  5). The concentration of BAG-1 is presented as a molar ratio with respect to hsp70. BAG-1 produced ϳ50% inhibition of GR folding when present at a molar ratio of 0.4. The inhibition of folding was prevented by heat inactivation of the BAG-1 protein at 100°C (data not shown).
Geldanamycin is a benzoquinone ansamycin antibiotic that binds to the nucleotide-binding site of hsp90 and specifically blocks hsp90 function (45)(46)(47). When steroid receptor⅐hsp90 heterocomplexes are assembled in reticulocyte lysate in the presence of geldanamycin, they cannot bind steroid. In this case, the complexes still contain the adaptor protein Hop in addition to hsp90 and hsp70 (48). Fig. 6 presents the composition of GR⅐hsp90 heterocomplexes assembled by the purified minimal folding system in the absence of inhibitor (lane 2) or in the presence of 10 M geldanamycin (lane 4) or in the presence of BAG-1 (lane 6) at a BAG-1:hsp70 molar ratio of 1.2. In contrast to complexes assembled in the presence of geldanamycin, which had an increased amount of Hop, Hop protein was eliminated from GR⅐hsp90 heterocomplexes assembled in the presence of BAG-1. By comparing lanes 5 and 6, it can be seen that the added BAG-1 was present in the GR⅐hsp90 heterocomplex. As shown in Fig. 7, incubation of preassembled complexes with BAG-1 did not affect either steroid-binding activity or the composition of the GR⅐hsp90 heterocomplex.
Effect of Low Concentrations of BAG-1-Although in vitro studies of BAG-1 effects on protein folding assays have used BAG-1 (or BAG-1M) at concentrations that are stoichiometric or slightly higher with respect to hsp70/hsc70 (19,28,31), the concentration of BAG-1 in reticulocyte lysate is very low. By comparing the intensity of bands developed with 125 I-labeled counterantibody in immunoblots containing aliquots of rabbit reticulocyte lysate and standard amounts of purified hsp70 and BAG-1, we estimated the BAG-1:hsp70 molar ratio in reticulocyte lysate to be ϳ0.03 (data not shown). Even accounting for all three of the BAG species shown in Fig. 2, this ratio could at most double to ϳ0.06. Fig. 8 presents the effects of three concentrations of BAG-1 on GR heterocomplex assembly by the purified minimal assembly system. When a low concentration of BAG-1 was present such that the BAG-1:hsp70 molar ratio was 0.07 (Fig. 8A, lane 4), no inhibition of GR folding could be observed as assayed by steroid-binding activity (Fig. 8B). How-ever, even at this low concentration of BAG-1, Hop was released from the GR⅐hsp90 heterocomplex (Fig. 8A). As the concentration of BAG-1 was increased, GR folding and GR⅐hsp90 heterocomplex assembly were inhibited in a concentration-dependent fashion. By excising and counting the 125 Ilabeled bands for GR and hsp90 in several experiments, we have determined that the decrease in steroid-binding activity at each BAG-1 concentration was accompanied by a comparable decrease in the hsp90:GR ratio (Fig. 8B).

BAG-1 Overexpression Decreases Steroid-binding Activity in Vivo-
To determine if overexpression of BAG-1 would affect GR steroid-binding activity in cells, COS-7 cells were cotransfected with cDNAs expressing a fusion protein containing the mouse GR and a cDNA expressing BAG-1. As shown in Fig. 9A, transfection with BAG-1 reduced steroid-binding activity from the transfected GR by ϳ65% (lane 4) with respect to the vector control (lane 3). We also examined the effect of BAG-1 on the endogenous GR in GM-701 fibroblasts stably transfected with BAG-1 (26). As shown in Fig. 9B, steroid-binding activity in BAG-1-overexpressing cells (lane 2) was ϳ50% of that in the parent cell line (lane 1). On the original autoradiogram, a thin band of endogenous BAG-1 was visible both in untransfected COS-7 cells and in the parental GM-701 cell line that was not visible in the Western blot photographs of Fig. 9, A (lanes 1-3)

DISCUSSION
Previous studies have established that a system containing hsp90, hsp70, Hop, YDJ-1, and p23 is sufficient to assemble stable steroid receptor⅐hsp90 heterocomplexes (5)(6)(7)(8)(9). Although the hsp70-binding proteins BAG-1 and Hip are not required for assembly, they may nevertheless play a role in modulating the assembly process. Here, we have examined the role of BAG-1 in GR folding in reticulocyte lysate. Upon immunoadsorption of Hop from reticulocyte lysate, only a small fraction of the hsp90⅐Hop⅐hsp70⅐hsp40 foldosome complexes contained BAG-1 (Fig. 1). This suggests that any activity of BAG-1 in GR⅐hsp90 heterocomplex assembly by reticulocyte lysate would likely involve a dynamic association with multiple receptor⅐hsp90⅐Hop⅐hsp70 complexes during the assembly process.
We prepared a partially reconstituted GR⅐hsp90 heterocomplex assembly system in which hsp70 and its co-chaperones hsp40, BAG-1, and Hip were present at the same stoichiometry relative to each other that existed in whole reticulocyte lysate (Figs. 2 and 3). Selective elimination of BAG-1 by its immunoadsorption from this system had little or no effect on the ability of the system to convert the GR to the steroid-binding form and no apparent effect on the composition of the GR⅐hsp90 heterocomplex (Fig. 4).
We have previously noted that GR⅐hsp90 heterocomplexes assembled by reticulocyte lysate contain little or sometimes no Hop, whereas those assembled by the minimal reconstituted system of purified hsp90, hsp70, Hop, YDJ-1, and p23 contain Hop (6). This led to the notion that reticulocyte lysate may contain an activity that facilitates the exit of Hop from the receptor heterocomplex during its assembly (6). It is interesting that addition of BAG-1 to the purified minimal assembly system at a BAG-1:hsp70 molar ratio of 0.07, which is in the range of the ratio existing in reticulocyte lysate, resulted in a GR⅐hsp90 heterocomplex that contained very little Hop relative to the BAG-1-free system (Fig. 8A, cf. lanes 2 and 4). It is possible that BAG-1 at concentrations that are substoichiometric with respect to hsp70 can facilitate the release of Hop from receptor-bound hsp70 in the GR⅐hsp90 heterocomplex without affecting GR folding. Promoting the release of Hop would facilitate the formation of new hsp90⅐Hop⅐hsp70 complexes and thus facilitate the hsp90-based chaperone cycle.
Johnson et al. (49) have shown that Hop binds preferentially to the ADP-bound form of hsp70. Inasmuch as BAG-1 reportedly binds to the ATPase domain of hsp70 (19,26) and promotes the release of ADP (20, 28), it may alter the conformation of hsp70 to reduce its affinity for Hop, thus promoting Hop dissociation during GR⅐hsp90 heterocomplex assembly. Hip is FIG. 5. Purified recombinant BAG-1 inhibits GR folding by a purified GR⅐hsp90 heterocomplex assembly system. Stripped GR immune pellets were incubated for 30 min at 30°C with 20 g of purified rabbit hsp90, 20 g of purified rabbit hsp70, 2 l of bacterial lysate containing Hop, 2 l of bacterial lysate containing YDJ-1, 8 g of purified human p23, and the indicated molar ratios of purified BAG-1 in the presence of 100 mM KCl and an ATP-regenerating system. At the end of the incubation, steroid-binding activity was assayed. BAG-1 that was heat-treated (100°C for 5 min) did not inhibit GR activation (data not shown).
FIG. 6. Composition of GR⅐hsp90 heterocomplexes formed by the minimal assembly system in the presence of geldanamycin or BAG-1. Stripped GR immune pellets were incubated for 20 min at 30°C with the minimal assembly system (hsp90, hsp70, Hop, YDJ-1, and p23) in the presence of 20 mM sodium molybdate. The GR, hsp90, hsp70, Hop, and BAG-1 in the washed immune pellets were assayed by SDS-PAGE and Western blotting, and a portion of each immune pellet was incubated with [ 3 H]triamcinolone acetonide to determine steroidbinding activity. Lanes 1 and 2, nonimmune and immune pellets, respectively, reconstituted with the minimal assembly system without inhibitor; lanes 3 and 4, nonimmune and immune pellets, respectively, reconstituted in the presence of 10 M geldanamycin; lanes 5 and 6, nonimmune and immune pellets, respectively, reconstituted in the presence of BAG-1 at a molar ratio of 1.2:1 with respect to hsp70. FIG. 7. BAG-1 does not affect preassembled GR⅐hsp90 heterocomplexes. Stripped GR immune pellets were incubated for 10 min at 30°C with the minimal assembly system. The immune pellets were then washed once with TEGM buffer and once with 10 mM Hepes, pH 7.4; and the washed pellets were suspended in HKD buffer containing the ATP-regenerating system and 20 mM molybdate. The samples were incubated for an additional 10 min at 0 or 30°C, and steroid-binding activity and the GR and its associated proteins were assayed. Lanes 1 and 2, nonimmune and immune reconstituted pellets, respectively, incubated at 0°C; lanes 3 and 4, nonimmune and immune reconstituted pellets, respectively, incubated at 30°C; lanes 5 and 6, reconstituted pellets incubated at 30°C with BAG-1 at a BAG-1:hsp70 molar ratio of 1.2. The steroid-binding activities (black bars) are means Ϯ S.E. from three experiments expressed as a percent of the 0°C control. thought to stabilize the ADP-bound conformation of hsp70 (28) and may have the opposite effect compared with BAG-1. In the purified minimal assembly system where no Hip is present, it may be easier to observe an effect of BAG-1 at low molar ratios with respect to hsp70 than in a more complex assembly system, such as DE52 fraction A of reticulocyte lysate (Figs. 3B and 4), where Hip is also present.
Other studies of BAG-1 effects on hsp70 chaperone activity in vitro have been performed at BAG-1:hsp70 molar ratios of 1 or more (28,31). These ratios are much higher than in reticulocyte lysate (0.03-0.06), but probably approach ratios existing in transfected cells overexpressing BAG-1 and in some types of tumors (27). At high BAG-1:hsp70 molar ratios, we found that BAG-1 inhibited the generation of steroid-binding activity (Figs. 5, 6, and 8) and the assembly of GR⅐hsp90 heterocomplexes ( Figs. 6 and 8). A similar situation is seen with geldanamycin, which binds to the nucleotide-binding site of hsp90 and blocks its conversion to the ATP-dependent conformation (46). hsp90 in the GR⅐hsp90 heterocomplex must be in the ATPbound conformation to be capable of hormone binding (50). Because this conversion is prevented, GR⅐hsp90 heterocomplexes generated in the presence of geldanamycin do not bind steroid. Also, Toft and co-workers (49,50) have shown that ADP-bound hsp90 has a high affinity for Hop and that binding of ATP markedly reduces Hop-binding affinity. Thus, the GR⅐hsp90 heterocomplex assembled in the presence of geldanamycin contains abundant amounts of Hop and does not bind steroid (Fig. 6, lane 4). In contrast to the situation with geldanamycin, these heterocomplexes assembled in the presence of BAG-1 are free of Hop ( Fig. 6 and 8), suggesting that both hsp70 and hsp90 in the assembled heterocomplexes are no longer in the ADP-bound conformation that possesses high affinity for Hop.
In summary, at the low levels at which BAG-1 exists in reticulocyte lysate, BAG-1 appears to modulate the GR⅐hsp90 assembly process by promoting Hop release from the assembly complex. When it is present at high levels in the cell-free assembly system or when it is overexpressed (Fig. 9), BAG-1 affects GR folding in a negative manner to yield decreased steroid-binding activity.