Structure-Function Analysis of Squirrel Monkey FK506-Binding Protein 51, a Potent Inhibitor of Glucocorticoid Receptor Activity
http://www.100md.com
内分泌学杂志 2005年第7期
Departments of Pharmacology (W.B.D., J.G.S.) and Comparative Medicine (J.G.S.), University of South Alabama, College of Medicine, Mobile, Alabama 36688; and Department of Biochemistry and Molecular Biology (V.P., D.F.S.), Mayo Clinic Scottsdale, Scottsdale, Arizona 85259
Address all correspondence and requests for reprints to: Jonathan G. Scammell, Ph.D., Department of Pharmacology, MSB 3370, University of South Alabama, Mobile, Alabama 36688. E-mail: jscammel@jaguar1.usouthal.edu.
Abstract
FK506-binding protein 51 (FKBP51) and FKBP52 are large molecular weight immunophilins that are part of the mature glucocorticoid receptor (GR) heterocomplex. These proteins possess peptidyl-prolyl isomerase (PPIase) and tetratricopeptide repeats (TPR) domains that are important for modulation of GR activity. A naturally occurring animal model of glucocorticoid resistance, the squirrel monkey, results from the relative overexpression of FKBP51 that renders the GR in a low-affinity state. In vitro studies demonstrated that the squirrel monkey form of FKBP51 is greater than 6-fold more potent than human FKBP51 in this respect. The goals of these studies were to determine the roles of the TPR and PPIase domains in the inhibitory activity of squirrel monkey FKBP51 and to gain insight into structural features of squirrel monkey FKBP51 responsible for potent inhibition of dexamethasone-stimulated GR activity. Mutations in the TPR of squirrel monkey FKBP51 that inhibit association with heat shock protein 90 blocked GR inhibitory activity. Mutations that abrogate the PPIase activity of squirrel monkey FKBP51 had no effect on GR inhibitory activity. Chimeras of squirrel monkey and human FKBP51 were tested to identify domains responsible for their different inhibitory potencies. Amino acid differences in domains FK1 and FK2 between squirrel monkey and human FKBP51 contribute equally to the enhanced inhibitory activity of squirrel monkey FKBP51. Furthermore, squirrel monkey FKBP51 in which either FK1 or FK2 was deleted lacked GR inhibitory activity. Thus, the potent inhibitory activity of squirrel monkey FKBP51 involves both FK domains and the heat shock protein 90-binding TPR domain.
Introduction
SQUIRREL MONKEYS have high circulating levels of cortisol to compensate for the expression of glucocorticoid receptors (GR) with low activity (1). The initial observation of high cortisol levels in squirrel monkeys was made in 1970 (2), but it was 1999 before the factors responsible for lower GR activity in squirrel monkeys were identified. Using squirrel monkey B-lymphoblasts (SML) as a cell model, we initially observed that the levels of FK506-binding protein 51 (FKBP51) and FKBP52, large molecular weight FK506-binding immunophilins that compete for binding sites within the GR complex (3), were quite different from levels in human B-lymphoblasts (4). FKBP51 was found to be greater than 10-fold higher in squirrel monkey cells, whereas FKBP52 was less than half that in human cells. Similar changes in FKBP51 and FKBP52 were observed in other cell lines and in liver samples from squirrel monkeys (5, 6). That the relative overexpression of FKBP51 can cause glucocorticoid resistance was demonstrated by Denny et al. (7), who showed that expression of squirrel monkey FKBP51 dramatically increases the EC50 for dexamethasone-stimulated GR transactivation in heterologous COS-7 cells. Human FKBP51 is also inhibitory, but has less than one sixth of the activity of the squirrel monkey protein. Thus, there is convincing evidence that glucocorticoid resistance in squirrel monkeys results, at least in part, from overexpression of a potent form of FKBP51.
FKBP51 and FKBP52 exhibit sequence homology and similar structural organization. The crystal structures of at least portions of both proteins have been recently resolved (8, 9, 10). The N-terminal domain of each protein binds FK506 and has peptidyl-prolyl isomerase (PPIase) activity that converts prolyl peptide bonds within target proteins from cis- to trans-proline (11). This domain, termed FK1 for first FKBP domain, is followed by a second FKBP domain, FK2, that is structurally similar to FK1 but is catalytically inactive (8, 12). The C-terminal domains of FKBP51 and FKBP52 contain three tetratricopeptide repeats (TPR) that are highly degenerate 34-amino-acid repeats involved in protein-protein interactions. An established protein partner of the large molecular FKBPs is heat shock protein 90 (Hsp90), which mediates the interaction of FKBP51, FKBP52, and other co-chaperones with steroid hormone receptor heterocomplexes (13). However, regions outside the TPR domains can influence FKBP binding to Hsp90 (12, 14). The C termini of the FKBPs also contain a calmodulin binding site-like motif sequence (15), although calmodulin regulation of FKBP function has not been demonstrated.
The molecular details of how overexpression of FKBP51 in squirrel monkeys leads to decreased GR binding are largely unknown. Initial observations indicated that squirrel monkey SML cells exhibit an abundance of FKBP51 in Hsp90 complexes (4). We also demonstrated that incorporation of squirrel monkey FKBP51 into GR heterocomplexes is associated with a decrease in GR binding (4). These results suggested that the effect of FKBP51 is mediated through interaction with Hsp90 within the GR heterocomplex. However, direct evidence for the importance of this interaction was lacking, as were data on the role of PPIase activity in the inhibitory activity of FKBP51. Furthermore, very little was known of the structural features of squirrel monkey FKBP51 that render it a potent inhibitor of GR activity. In this study, we investigated the roles of the TPR and PPIase-like domains (FK1 and FK2) in FKBP51’s GR inhibitory activity and, by testing chimeras of squirrel monkey and human FKBP51, have identified domains that are important for their differing potencies.
Materials and Methods
Materials
Culture medium was obtained from Life Technologies (Grand Island, NY). Defined and charcoal-dextran-treated fetal bovine serum was purchased from HyClone Laboratories, Inc. (Logan, UT). Dexamethasone and protease inhibitor cocktail were purchased from Sigma Chemical Co. (St. Louis, MO). The mouse mammary tumor virus (MMTV) promoter-luciferase reporter vector was provided by Dr. R. M. Evans (The Salk Institute, La Jolla, CA). The FLAG-tagged rabbit FKBP52 expression vector (FKBP52-FLAG-pCMV5) was provided by Dr. Michael Chinkers (University of South Alabama, Mobile, AL). Antibody to Hsp90, SPA-835, was purchased from StressGen Biotechnologies Corp. (Victoria, Canada), and anti-FLAG M2 antibody was from Stratagene (La Jolla, CA). Polyclonal antibody to human FKBP51 was purchased from Affinity BioReagents, Inc. (Golden, CO).
Cell cultures
COS-7 cells were grown as monolayers in DMEM supplemented with 10% fetal bovine serum, 50 U/ml penicillin G, and 0.05 mg/ml streptomycin. Cells were grown at 37 C in a humidified atmosphere of 5% CO2/95% air. Cells were transfected using a modification of the method of Bodwell et al. (16), as described previously (7). In most experiments, medium was replaced 18 h after transfection, and cells were treated with dexamethasone (0.1–1 nM). After 24 h, cells were lysed and assayed for luciferase activity as described (17). EC50 values (defined as the concentration of ligand that produces 50% of the maximum response) were obtained from the concentration-response curves. Unless otherwise indicated, the EC50 values are the mean ± range of two independent experiments.
Cell lysates were also used to determine the levels of expressed FKBP51 and FKBP52 by Western blot. Briefly, lysates were mixed 1:1 in 2x sample buffer, separated by SDS-PAGE, and transferred to nitrocellulose. The blots were incubated at 4 C overnight with the anti-FLAG M2 antibody, anti-FKBP51 antibody, or anti-Hsp90 antibody. After washing, blots were incubated with second antibody and developed using an Immune-Star chemiluminescent kit (Bio-Rad Laboratories, Hercules, CA).
Plasmids
The construction of hGR-pcDNA1.1/Amp has been described (7). hGR-pUB6/V5-His was constructed from hGR-pcDNA1.1/Amp by excision with BamHI and XbaI and ligation into pUB6/V5-His (Invitrogen, Carlsbad, CA). The construction of expression plasmids containing either wild-type or FLAG-tagged squirrel monkey (sm51-pCI-neo, sm51FLAG-pCI-neo) or human (h51-pCI-neo, h51FLAG-pCI-neo) FKBP51 cDNA was described previously (4, 18). sm51FLAG-pCI-neo (Lys352Ala, Arg356Ala) was constructed using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) using sm51FLAG-pCI-neo as template and primers (nucleotide mutations italic): GTG CCA ATG AGG CAG GCC TGT ATG CGA GAG GGG AAG CCC (sense) and GGG CTT CCC CTC TCG CAT ACA GGC CTG CCT CAT TGG CAC (antisense), corresponding to nucleotide positions 1063–1101 in the squirrel monkey cDNA (AF140759). sm51FLAG-pCI-neo (Phe67Asp, Asp68Val) was constructed using sm51FLAG-pCI-neo as template and primers (nucleotide mutations are italic): CGG AAA ATT GGC AAA TGG AAA GAA GGA CGT CTC CAG TCA TGA TAG AAA TGA ACC G (sense) and CGG TTC ATT TCT ATC ATG ACT GGA GAC GTC CTT CTT TCC ATT TGC CAA TTT TCC G (antisense), positions 194–248 in the squirrel monkey cDNA.
Seven squirrel monkey/human FKBP51 chimeras were generated. The first, [h1–15]smFKBP51, was generated by site-directed mutagenesis using sm51-pCI-neo as template and primers (nucleotide mutations are italic) GGT GCC AAG AAC AAT GAA GAA AGC CCC ACA GCT ACT GTT GC (sense) and GCA ACA GTA GCT GTG GGG CTT TCT TCA TTG TTC TTG GCA CC (antisense), positions 36–76 in the squirrel monkey cDNA. Chimeras [h1–277]smFKBP51 and [sm1–277]hFKBP51 were produced by excising cDNAs encoding the N-terminal 277 amino acids of h51- and sm51-pCI-neo plasmids with EcoRI and BstZI from each plasmid, taking advantage of a unique BstZI site at codon 277. The resulting fragments were gel purified and cross-ligated to produce cDNAs encoding the chimeric proteins. Chimera [h26–133]smFKBP51 was generated as follows. An SpeI site was introduced by site-directed mutagenesis at codon 26 of wild-type human and squirrel monkey FKBP51 plasmids. The resulting human FKBP51 plasmid was double digested with SpeI and SacI, taking advantage of a unique SacI site at codon 133. [h133–361]smFKBP51 was generated by double digestion of wild-type human FKBP51 plasmid with SacI and PvuII, taking advantage of a unique PvuII site at codon 361. Fragments resulting from the digestions of the human FKBP51 plasmid were gel purified and ligated into the appropriately digested squirrel monkey FKBP51 plasmids.
The squirrel monkey FKBP51 FK1 deletion mutant (smFKBP51FK1), in which amino acids 33–141 were deleted, was generated by first introducing unique ApaI sites into sm51FLAG-pCI-neo using primers (nucleotide mutations are italic) CCT CCA AGA AAG ACA GGG CCC TAT TAA AGA TTG TCA AAA GAG TG (sense) and CAC TCT TTT GAC AAT CTT TAA TAG GGC CCT GTC TTT CTT GGA GG (antisense) and GCT TCT TGA TTT CAA AGG AGG GGC CCT ACT TGA AGA TGG AGG C (sense) and GCC TCC ATC TTC AAG TAG GGC CCC TCC TTT GAA ATC AAG AAG C (antisense), corresponding to nucleotide positions 97–140 and 419–461 in the squirrel monkey cDNA (AF140759), respectively. The mutagenized plasmid DNA was digested with ApaI, the FK1 fragment separated by electrophoresis, and the remaining plasmid DNA was purified and religated. The squirrel monkey FKBP51 FK2 deletion mutant (smFKBP51FK2), in which amino acids 147–251 were deleted, was generated by introducing unique AscI sites into sm51FLAG-pCI-neo using primers (nucleotide mutations are italic) GGA TTT ACT TGA AGA TGG GCG CGC CAT CCG GAG AAC TAA ACG (sense) and CGT TTA GTT CTC CGG ATG GCG CGC CCA TCT TCA AGT AAA TCC (antisense) and CAC TTA AGA GCT TCG GGC GCG CCA AAG AAT CCT GGG AG (sense) and CTC CCA GGA TTC TTT GGC GCG CCC GAA GCT CTT AAG TG (antisense), corresponding to nucleotide positions 440–481 and 757–794 in the squirrel monkey cDNA (AF140759), respectively. The mutagenized plasmid DNA was digested with AscI, the FK2 fragment separated by electrophoresis, and the remaining plasmid DNA was purified and religated.
Immunoprecipitation of FLAG-tagged FKB51
Transfected COS-7 cells were washed twice with cold PBS, collected in cold HEPES-buffered saline (HEBS: 20 mM HEPES, 1 mM EDTA, 150 mM NaCl, pH 7.4), and lysed by sonication. After the addition of protease inhibitor cocktail (10 μl/ml), the cell lysates were centrifuged at 21,000 x g for 1 h at 4 C, and the soluble fraction was removed and stored on ice. Aliquots of a 50% anti-FLAG M2 affinity resin slurry in HEBS was added to each cell lysate and the mixture rotated for 2 h at 4 C. The resin was pelleted, washed three times with ice-cold HEBS, and resuspended in 2x nonreducing sample buffer. The samples were boiled and centrifuged, and the supernatants were subjected to SDS-PAGE and Western blot analysis.
Results
Role of TPR and PPIase domains
The first series of experiments was designed to determine whether the inhibitory effect of squirrel monkey FKBP51 on GR activity is mediated through interaction with Hsp90 and the GR heterocomplex. To test this, basic residues in the TPR domain, previously shown to be essential for binding of protein phosphatase 5, FKBP52, and FKBP51 to Hsp90 (14, 19, 20), were mutated in squirrel monkey FKBP51 to alanine (Lys352Ala, Arg356Ala). The effect of the resulting mutant protein on GR activation was then evaluated in cell culture. COS-7 cells were transfected with a plasmid expressing human GR, an MMTV-luciferase reporter plasmid, and either an empty vector or a vector containing FLAG-tagged squirrel monkey FKBP51 cDNA or FLAG-tagged squirrel monkey FKBP51 cDNA containing mutations in the TPR region. Treatment of cells transfected with empty vector with dexamethasone (0.3–100 nM) produced a concentration-dependent GR transactivation response with an EC50 of 1.1 ± 0.3 nM (Fig. 1A). Expression of FLAG-tagged squirrel monkey FKBP51 resulted in an expected rightward shift in the dexamethasone concentration-response curve (EC50 of 13.6 ± 1.4 nM). FLAG-tagged squirrel monkey FKBP51 has the same inhibitory activity as wild-type squirrel monkey FKBP51 (data not shown). The inhibitory effect of squirrel monkey FKBP51 was largely abrogated when mutations were made in the TPR region (EC50 of 1.8 ± 0.7 nM). We confirmed that this mutant does not coprecipitate Hsp90 (Fig. 1B, top), even though its expression was slightly higher than the wild-type protein (bottom).
FIG. 1. A, Effect of squirrel monkey FKBP51 TPR domain mutant (Lys352Ala, Arg356Ala) on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected by electroporation with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either FLAG-tagged wild-type squirrel monkey FKBP51 (sm51wt, ) or FLAG-tagged squirrel monkey FKBP51 TPR domain mutant (sm51Tm, ). Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. B, Precipitation of Hsp90 by sm51wt, but not by sm51Tm. FLAG-tagged FKBP51 proteins from soluble extracts of transfected COS-7 cells were immunoadsorbed by anti-FLAG affinity resin, washed, eluted with sample buffer, and analyzed for Hsp90 and the FLAG epitope on FKBP51 by Western blot.
The second set of experiments was designed to determine whether the inherent PPIase activity of squirrel monkey FKB51 is an important component of its inhibitory effect on GR activity. A PPIase inactive mutant of squirrel monkey FKBP51 was generated by site-directed mutagenesis of two residues in the catalytic binding pocket, Phe67Asp, Asp68Val. These residues are highly conserved among FKBPs and are required for PPIase activity (21). Mutation of these residues in human FKBP51 was shown to reduce PPIase activity by greater than 90% (12). We found that expression of the squirrel monkey PPIase mutant potently inhibited GR transactivation in COS-7 cells in a manner similar to that seen with wild-type squirrel monkey FKBP51 (Fig. 2A). Using the procedure of Janowski et al. (22), we confirmed that this double mutant of squirrel monkey FKBP51 exhibited less than 5% of the PPIase activity of the wild-type protein (data not shown). The slightly less inhibitory activity of the PPIase mutant (EC50 of 3.7 ± 0.8 nM) compared with wild-type squirrel monkey FKBP51 (EC50 of 5.8 ± 2.9 nM) was attributed to slightly lower expression of the PPIase mutant (Fig. 2B). In this and other experiments, in which polyclonal antibody to FKBP51 was used to detect expressed squirrel monkey FKBP51, a slower migrating band is sometimes apparent. This band reflects the endogenous expression of FKBP51 in COS-7 cells. Squirrel monkey FKBP51 migrates more rapidly than African green monkey or human FKBP51 on SDS gels (7).
FIG. 2. A, Effect of squirrel monkey FKBP51 PPIase mutant (Phe67Asp, Asp68Val) on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected by electroporation with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type squirrel monkey FKBP51 (sm51wt, ) or squirrel monkey FKBP51 PPIase mutant (sm51Pm, ). Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. B, Expression of sm51wt and sm51Pm in electroporated COS-7 cells. Cell lysates from each transfected set were pooled, and Western blots were performed for FKBP51.
As immunophilins compete via their TPR domains for a TPR acceptor site on Hsp90 in steroid receptor complexes (recently reviewed in Refs. 3 , 13), we investigated whether overexpression of FKBP52 might abrogate the inhibitory effect of squirrel monkey FKBP51 on GR transactivation. Dexamethasone dose-response curves were performed in COS-7 cells expressing squirrel monkey FKBP51, rabbit FKBP52, or a combination of the two immunophilins. Expression of FKBP52 alone resulted in a slight leftward shift in the dose-response curve (EC50 of 0.2 ± 0.1 nM) compared with the control curve (EC50 of 0.4 ± 0.2 nM) (Fig. 3A). More dramatically, the rightward shift in the dose-response curve induced by squirrel monkey FKBP51 expression (EC50 of 4.4 ± 0.3 nM) was blocked when FKBP52 was coexpressed (EC50 of 0.4 ± 0.1 nM). Similar expression of the FLAG-tagged FKBPs was verified by immunodetection of the FLAG epitope in cell lysates collected from each transfected set of COS-7 cells (Fig. 3B). Together, the results of these experiments suggest that inherent PPIase activity does not play a role in the inhibitory effect of squirrel monkey FKBP51 on GR activity, but interaction with Hsp90 and the GR heterocomplex through FKBP51’s TPR domain is critical.
FIG. 3. A, Overexpression of FKBP52 blocks the inhibitory effect of FKBP51 on GR transactivation in COS-7 cells. COS-7 cells were transfected by electroporation with hGR-pUB6/V5-His, MMTV-luciferase; either empty pCIneo empty vector or pCIneo containing cDNA encoding FLAG-tagged squirrel monkey FKBP51; and either empty pCMV5 vector or pCMV5 containing cDNA encoding FLAG-tagged rabbit FKBP52. Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. , control; , rabbit FKBP52 (r52); , wild-type squirrel monkey FKBP51 (sm51); , rabbit FKBP52 and wild-type squirrel monkey FKBP51 (r52+sm51). B, Expression of FLAG-tagged FKBP52 (FLAG-52) and FKBP51 (FLAG-51) in electroporated COS-7 cells. Cell lysates from each transfected set were pooled, and Western blots were performed for the FLAG epitope.
Activities of squirrel monkey/human FKBP51 chimeras
The following experiments were designed to determine the structural basis for the enhanced inhibitory activity of squirrel monkey FKBP51. Although squirrel monkey and human FKBP51 are 94% identical, squirrel monkey FKBP51 is greater than 6-fold more potent than human FKBP51 in inhibiting GR activity (7). Our initial approach was to generate point mutants of squirrel monkey or human FKBP51 and compare their activities to wild-type proteins. We found that no single amino acid change in either of the proteins significantly affected their relative potencies (data not shown), suggesting that multiple amino acid differences between squirrel monkey and human FKBP51 likely contribute to their different potencies. To address this issue, we generated chimeras of squirrel monkey and human FKBP51 and compared their inhibitory activities on dexamethasone-induced GR transactivation in COS-7 cells. Details of these chimeras are shown in Fig. 4.
FIG. 4. A, Diagram showing the relative positions of functional domains in FKBP51. The two FKBP12-like domains are indicated; FK1 is the PPIase and FK506-binding domain, whereas the related FK2 lacks PPIase and drug-binding activities. The TPR domain is the tetratricopeptide repeat that mediates interaction with Hsp90. B, Diagram of human and squirrel monkey FKBP51 chimeric proteins. The shaded boxes represent human FKBP51 sequence, whereas the open boxes represent squirrel monkey FKBP51 sequence.
Because squirrel monkey and human FKBP51 differ in five amino acids in the N-terminal region of the protein (positions 10–13 and 15), the first chimera tested [h1–15]smFKBP51 included the N-terminal 15 amino acids of human FKBP51 and the C-terminal 442 amino acids of squirrel monkey FKBP51. We found that expression of the [h1–15]smFKBP51 chimera resulted in a rightward shift in the dexamethasone concentration-response curve (EC50 of 8.1 ± 0.1 nM), similar to that seen with wild-type squirrel monkey FKBP51 (EC50 of 8.1 ± 0.4 nM) (Fig. 5A, top). The EC50 in COS-7 cells transfected with empty vector was 1.6 ± 0.3 nM. Western blots confirmed similar expression of wild-type and chimeric FKBP51 (Fig. 5A, bottom). These results suggested that the differences between human and squirrel monkey FKBP51 in the extreme N terminus do not contribute to the enhanced inhibitory activity of squirrel monkey FKBP51.
FIG. 5. A, Top, Effect of the human/squirrel monkey FKBP51 N-terminal chimera, [h1–15]smFKBP51, on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type squirrel monkey FKBP51 (sm51wt, ) or a chimera of FKBP51 containing the N-terminal 15 amino acids of human FKBP51 and the C-terminal 442 amino acids of squirrel monkey FKBP51 (h15sm, ). Bottom, Expression of sm51wt and h15sm compared with control in electroporated COS-7 cells. B, Top, Effect of human/squirrel FKBP51 chimeras, [h1–277]smFKBP51 and [sm1–277]hFKBP51, on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type squirrel monkey FKBP51 (sm51wt, ), wild-type human FKBP51 (h51wt, ), or chimeras of FKBP51 containing either the N-terminal 277 amino acids of human FKBP51 and the C-terminal 180 amino acids of squirrel monkey FKBP51 (h277sm, ) or the N-terminal 277 amino acids of squirrel monkey FKBP51 and the C-terminal 180 amino acids of human FKBP51 (sm277 h, ). Bottom, Expression of h51wt, h277sm, sm51wt, and sm277h compared with control in electroporated COS-7 cells. For the Western blots shown in both A and B, cell lysates from each transfected set were pooled and analyzed for FKBP51.
We next tested the effects of swapping the N-terminal 277 amino acids of human and squirrel monkey FKBP51. These chimeras, [sm1–277]hFKBP51 and [h1–277]smFKBP51, include both FK1 and FK2 domains of either squirrel monkey or human FKBP51, respectively (Fig. 4). We found that expression of the [sm1–277]hFKBP51 chimera resulted in a rightward shift in the dexamethasone concentration-response curve (EC50 of 10.6 ± 0.1 nM) that was similar to that observed with wild-type squirrel monkey FKBP51 (EC50 of 11.4 ± 1.1 nM) (Fig. 5B, top). Expression of the [h1–277]smFKP51 chimera resulted in a less dramatic shift in the dexamethasone concentration-response curve (EC50 of 2.7 ± 0.2 nM), similar to that seen with wild-type human FKBP51 (EC50 of 2.2 ± 0.6 nM). The EC50 in COS-7 cells transfected with empty vector was 1.4 ± 0.6 nM. Western blots confirmed similar expression of wild-type and chimeric FKBP51 (Fig. 5B, bottom). These results suggested that the enhanced inhibitory potency of squirrel monkey FKBP51 results from amino acid differences in the N-terminal half of the protein. Differences between squirrel monkey and human FKBP51 in the TPR domain and C terminus of the proteins are likely irrelevant.
We next determined the contribution of the individual FK domains to the enhanced potency of squirrel monkey FKBP51. We tested the activities of chimeras of human and squirrel monkey FKBP51 in which most of the FK1 domain, [h26–133]smFKBP51, or the entire FK2 domain of squirrel monkey FKBP51, [h133–361]smFKBP51, was exchanged with the same region of human FKBP51 (see Fig. 4). Expression of [h26–133]smFKBP51 resulted in inhibition of hormone-stimulated GR activity that was greater (EC50 of 3.7 ± 0.6 nM) than that achieved with human FKBP51 (EC50 of 1.7 ± 0.2 nM) but not as striking as that observed with squirrel monkey FKBP51 (EC50 of 8.6 ± 0.6 nM) (Fig. 6A). The inhibitory potency of [h133–361]smFKBP51 was also intermediate between human and squirrel monkey FKBP51 (EC50 of 4.0 ± 0.3 nM). Western blots confirmed similar expression of the FKBP51 chimeras (Fig. 6B). These results suggest that amino acid differences between human and squirrel monkey FKBP51 in both FK domains contribute to their differing inhibitory potencies on GR transactivation.
FIG. 6. A, Effect of human/squirrel monkey FKBP51 FK1 and FK2 chimeras, [h26–133]smFKBP51 and [h133–361]smFKBP51, on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type human (h51wt, ) or squirrel monkey (sm51wt, ) FKBP51 or human/squirrel monkey FKBP51 chimeras that have human FKBP51 sequence in positions 26–133 (hFK1, ) or 133–361 (hFK2, ) within squirrel monkey FKBP51. Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. B, Expression of h51wt, sm51wt, hFK1, and hFK2 compared with control in electroporated COS-7 cells. Cell lysates from each transfected set were pooled and analyzed for FKBP51 by Western blot.
Effect of deletion of FK1 or FK2 on the activity of squirrel monkey FKBP51
The finding that amino acids in FK1 and FK2 influence the activities of the two forms of FKBP51 prompted us to investigate the effect of simply deleting these domains on the activity of squirrel monkey FKBP51. Two mutants were generated: smFKBP51FK1 in which residues 22–141 were deleted and smFKBP51FK2 in which residues 147–251 were deleted, resulting in the complete removal of either the FK1 or FK2 domain (see Fig. 4A). We found that deletion of either of the two FK domains completely eliminated the inhibitory activity of squirrel monkey FKBP51 in transfected, dexamethasone-treated COS-7 cells (Fig. 7A). The EC50 values of the FK1 and FK2 deletion mutants were 0.4 and 0.3 nM, respectively, compared with that achieved in cells transfected with empty vector or wild-type squirrel monkey FKBP51 (0.4 and 11 nM, respectively). Western blots confirmed similar protein expression from the different FKBP51 constructs (Fig. 7B). The lack of inhibitory activity of these deletion mutants did not result from loss of Hsp90 binding because Hsp90 was coprecipitated as efficiently with squirrel monkey FKBP51 lacking FK1 as with the wild-type protein (Fig. 7C).
FIG. 7. A, Effect of deletions of either FK1 or FK2 in squirrel monkey FKBP51 on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type squirrel monkey FKBP51 (sm51wt, ), or squirrel monkey FKBP51 in which either the FK1 (FK1, ) or FK2 (FK2, ) domains were deleted. Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. Each point represents the result from a single experiment. B, Expression of sm51wt, FK1, and FK2 compared with control in electroporated COS-7 cells. Cell lysates from each transfected set were pooled and analyzed for FKBP51 by Western blot. C, Precipitation of Hsp90 by sm51wt and FK1. FLAG-tagged FKBP51 proteins from soluble extracts of transfected COS-7 cells were immunoadsorbed with anti-FLAG affinity resin, washed, eluted with sample buffer, and analyzed for Hsp90 and the FLAG epitope on FKBP51 by Western blot.
Discussion
In 2003, the Smith laboratory took advantage of the null background of Saccharomyces cerevisiae yeast to directly test the roles of FKBP51 and FKBP52 in regulation of GR activity (23). When human FKBP52 was introduced into yeast expressing GR, hormone-dependent reporter activity was enhanced, reflecting increased GR ligand-binding affinity. It was found that this effect was quite specific to FKBP52 and requires both the Hsp90-binding and PPIase activities of the protein. When human FKBP51 was coexpressed, it completely blocked the stimulatory effect of FKBP52 on GR activity. However, studies to determine which domains or activities of human FKBP51 were responsible for its inhibitory activity in yeast were not performed. Here, we used African green monkey COS-7 cells, which constitutively express relatively high levels of FKBP52 and low levels of FKBP51 (4), to transiently express different mutants of squirrel monkey FKBP51, a particularly potent form of FKBP51 (7). We demonstrated that the Hsp90-binding, but not PPIase, activity of FKBP51 is necessary for its inhibitory effect on GR function.
Complete details of how FKBP52 and FKBP51 affect GR-binding activity are not known. The mature GR heterocomplex may contain either FKBP52 or FKBP51, but ligand binding appears to foster selective association of FKBP52 with the complex (24, 25). It is speculated that within the complex, the FKBP52 PPIase targets one or more prolines in the ligand-binding domain of GR and optimizes the conformation for high-affinity hormone binding (3). FKBP51 also has PPIase activity, but biochemical studies have suggested that FKBP52 and FKBP51 interact with steroid receptors differently (26, 27). The results of more recent studies have taken into account the x-ray crystal structure of FKBP51 (8) and suggest that regions of FKBP51 and FKBP52, C-terminal to the TPR domain, may assume alternative conformations that differentially affect their interaction with Hsp90 (14). Hence, the GR ligand-binding domain may be inaccessible to the active site of FKBP51, and FKBP51 acts as an inhibitor. Furthermore, the inhibitory activity of FKBP51 was quite specific because the coexpression of another TPR-containing protein, protein phosphatase 5, did not affect potentiation by FKBP52 (23). Here we demonstrate that squirrel monkey FKBP51 interaction with the receptor heterocomplex through Hsp90 is critical, but that PPIase activity, which is localized to FK1 (8, 12), is dispensable. On the other hand, we found that deletion mutants involving the FK domains abrogate FKBP51’s effects. Therefore, the inhibitory activity of FKBP51 involves both FK domains plus the Hsp90-binding TPR domain.
A role for the FK domains in the inhibitory activity of FKBP51 on GR transactivation was also implicated in a very recent report by Wochnik et al. (28), who showed expression of FKBP51/FKBP52 chimeras containing either FK1 or FK2 of human FKBP51 inhibited GR transactivation, but neither was as efficacious as wild-type FKBP51. These investigators demonstrated that Hsp90 binding, but not PPIase activity, was essential for the inhibitory effect of human FKBP51 on GR activity, consistent with our results with squirrel monkey FKBP51 on GR activity reported here and with our results with human FKBP51 on progesterone receptor transactivation shown previously (18). Wochnik et al. (28) also showed that nuclear translocation of hormone-activated GR is slowed if human FKBP51 is overexpressed. That is, FKBP51 may disrupt the normal interaction of FKBP52 with the motor protein dynein, thought to be integral to movement of GR from cytoplasm to nucleus (13). However, at physiologically relevant hormone concentrations, when compared with the effect of FKBP51 on receptor binding, the contribution of a change in the rate of nuclear translocation to the overall inhibitory action of FKBP51 on GR signaling may be quite modest. This is even more evident when describing the effect of squirrel monkey FKBP51, which we have shown is a more potent inhibitor of receptor binding than human FKBP51 (4, 7).
Indeed, another goal of this work was to gain insight into the difference in inhibitory activities of squirrel monkey and human FKBP51. FKBP51 from other New World primates has similar activity to squirrel monkey FKBP51 (5), whereas the activity of FKBP51 cloned from an Old World primate, African green monkey, was similar to that of human FKBP51 (data not shown). These results suggest that structural differences between New World and Old World FKBP51 have evolved and led to functional differences in their abilities to repress steroid receptor activity. Our examination of chimeric FKBP51 proteins indicates that the relevant amino acid changes are localized to the FK domains. If these changes are mapped onto the crystal structure of squirrel monkey FKBP51 (Fig. 8), it becomes apparent that the New World-specific amino acids in FK1 and FK2 are surface residues that likely do not impact protein conformation. Indeed, comparison of crystal structures for human and squirrel monkey FKBP51 did not suggest significant conformational differences (8). On the other hand, these amino acids might contribute to a novel protein interaction surface involving both FK domains. An intriguing candidate as a potential interaction partner for this surface is FKBP52 because FKBP51 is known to antagonize FKBP52-mediated potentiation of GR activity (23). New World FKBP51, by virtue of amino acid changes in the FK domains, could interact with FKBP52 more strongly than Old World FKBP51 and is thus a better inhibitor of FKBP52-mediated potentiation of receptor activity. An interaction between FKBP51 and FKBP52 might be favored by mutual binding to an Hsp90 dimer within the steroid receptor complex, thereby explaining the observation that an FKBP51 mutant deficient in Hsp90 binding fails to inhibit GR activity. However, in lieu of direct experimental evidence that supports an interaction with FKBP52, one must also consider the receptor itself or Hsp90 as alternative interaction partners for FKBP51. We are currently pursuing studies to identify the putative interaction partner for the FK domain region of FKBP51 and to determine how such an interaction inhibits steroid receptor function.
FIG. 8. X-ray crystallographic structure of FKBP51 FK domain region. Amino acids that are different in human and New World primate FKBP51 are highlighted with dark gray side chains in FK1 [positions 42 (Asn, humanHis, New World primate), 115 (SerAla), and 116 (AlaThr)] and FK2 [positions 167 (ThrArg), 169 (GluGln), 180 (MetVal), and 218 (TyrHis)] domains of squirrel monkey FKBP51. These are surface residues that do not significantly influence FK domain conformation, but could alter interactions with a protein partner that contacts both FK domains.
Although mutational disruption of FKBP51’s PPIase activity affected neither FKBP51 interaction with Hsp90 nor progesterone receptor (12), nor did this mutation affect inhibition in this study or in the study of Wochnik et al. (28), the PPIase inhibitor FK506 has been shown to block the inhibitory effect of FKBP51 on GR activity (4, 7). Two considerations must be taken into account when interpreting the actions of FK506. First, FK506 also blocks the potentiation of GR activity by FKBP52, and FKBP52-mediated potentiation is PPIase-dependent (23). The change in GR activity that results from adding FK506 to cells that contain both FKBP52 (an activator of GR function) and FKBP51 (an inhibitor of GR function) might vary depending on the preexisting balance of cellular FKBP levels. A second consideration is that FK506 is not fully contained within the PPIase pocket of an FKBP; in fact, immunosuppression by FK506:FKBP12 complexes depend on the protruding effector domain of FK506 rather than inhibition of FKBP12 PPIase activity (29). When bound to FK1 of FKBP51, the effector domain of FK506 could sterically interfere with protein-protein interaction involving the FK domain region that is essential for the inhibitory effect on GR activity. We and others have shown that FK506 causes the release of FKBP51 from GR heterocomplexes (4, 25, 30).
We have provided evidence here and elsewhere that the relative overexpression of a potent form of the immunophilin FKBP51 mediates at least in part glucocorticoid and progestin resistance in squirrel monkeys and other New World primates (4, 5, 7, 18). However, this is not the cause of glucocorticoid resistance in the guinea pig, a New World hystricomorph. Fuller’s laboratory showed that at least four amino acid substitutions in the ligand-binding domain contribute to the low binding affinity of the guinea pig GR (31). This group has recently suggested that these changes may inhibit FKBP52 interaction with the receptor complex, compromising the formation of a high-affinity receptor (32). Thus, although the proximal mechanisms for glucocorticoid resistance in New World hystricomorphs and primates are different, the downstream effect (i.e. loss of FKBP52) may be common. On the other hand, vitamin D resistance that is observed in many New World primates likely occurs via a separate mechanism. This form of steroid resistance results from the constitutive overexpression of a member of the heterogeneous nuclear ribonucleoprotein A family, termed vitamin D response-element (VDRE) binding protein 2 (recently reviewed in Ref. 33). VDRE binding protein 2 was isolated from vitamin D-resistant B95-8 cells from cotton-top tamarin (Saguinus oedipus) and shown to compete with the vitamin D receptor-retinoid X receptor heterodimer for binding to the vitamin D response element (33, 34). Thus, this protein does not affect receptor binding, but rather acts to squelch VDRE-directed transactivation.
In summary, the results of this study have provided insight into the structural features of squirrel monkey FKBP51 that render it a potent inhibitor of GR signaling. We demonstrated that binding to Hsp90 is required for the inhibitory effect of squirrel monkey FKBP51. We have also shown that whereas PPIase activity of squirrel monkey FKBP51 is irrelevant, the FK domains are essential. These results suggest that protein-protein interaction involving the FK domains of FKBP51 and another protein partner mediates the inhibitory activity of FKBP51 on GR activity.
References
Chrousos GP, Renquist D, Brandon D, Eil C, Pugeat M, Vigersky R, Cutler Jr GB, Loriaux DL, Lipsett MB 1982 Glucocorticoid hormone resistance during primate evolution: receptor-mediated mechanisms. Proc Natl Acad Sci USA 79:2036–2040
Brown GM, Grota LJ, Penney DP, Reichlin S 1970 Pituitary-adrenal function in the squirrel monkey. Endocrinology 86:519–529
Smith DF 2004 Tetratricopeptide repeat cochaperones in steroid receptor complexes. Cell Stress Chaperones 9:109–121
Reynolds PD, Ruan Y, Smith DF, Scammell JG 1999 Glucocorticoid resistance in the squirrel monkey is associated with overexpression of the immunophilin FKBP51. J Clin Endocrinol Metab 84:663–669
Scammell JG, Denny WB, Valentine DL, Smith DF 2001 Overexpression of the FK506-binding immunophilin FKBP51 is the common cause of glucocorticoid resistance in three New World primates. Gen Comp Endocrinol 124:152–165
Scammell JG, Tucker JA, King JA, Moore CM, Wright JL, Tuck-Muller CM 2002 A kidney epithelial cell line from a Bolivian squirrel monkey. In Vitro Cell Dev Biol Animal 38:258–261
Denny WB, Valentine DL, Reynolds PD, Smith DF, Scammell JG 2000 Squirrel monkey immunophilin FKBP51 is a potent inhibitor of glucocorticoid receptor binding. Endocrinology 141:4107–4113
Sinars CR, Cheung-Flynn J, Rimerman RA, Scammell JG, Smith DF, Clardy J 2003 Structure of the large FK506-binding protein FKBP51, an Hsp90-binding protein and a component of steroid receptor complexes. Proc Natl Acad Sci USA 100:868–873
Li P, Ding Y, Wu B, Shu C, Shen B, Rao Z 2003 Structure of the N-terminal domain of human FKBP52. Acta Crystallogr D Biol Crystallogr 59:16–22
Wu B, Li P, Shu C, Shen B, Rao Z 2003 Crystallization and preliminary crystallographic studies of the C-terminal domain of human FKBP52. Acta Crystallogr D Biol Crystallogr 59:2269–2271
Schiene-Fischer C, Yu C 2001 Receptor accessory folding helper enzymes: the functional role of peptidyl prolyl cis/trans isomerases. FEBS Lett 495:1–6
Barent RL, Nair SC, Carr DC, Ruan Y, Rimerman RA, Fulton J, Zhang Y, Smith DF 1998 Analysis of FKBP51/FKBP52 chimeras and mutants for Hsp90 binding and association with progesterone receptor complexes. Mol Endocrinol 12:342–354
Pratt WB, Galigniana MD, Harrell JM, DeFranco DB 2004 Role of hsp90 and the hsp90-binding immunophilins in signalling protein movement. Cell Signal 16:857–872
Cheung-Flynn J, Roberts PJ, Riggs DL, Smith DF 2003 C-terminal sequences outside the tetratricopeptide repeat domain of FKBP51 and FKBP52 cause differential binding to Hsp90. J Biol Chem 278:17388–17394
Callebaut I, Renoir JM, Lebeau MC, Massol N, Burny A, Baulieu EE, Mornon JP 1992 An immunophilin that binds M(r) 90,000 heat shock protein: main structural features of a mammalian p59 protein. Proc Natl Acad Sci USA 89:6270–6274
Bodwell J, Swiff F, Richardson J 1999 Long duration electroporation for achieving high level expression of glucocorticoid receptors in mammalian cell lines. J Steroid Biochem Mol Biol 68:77–82
Jones LC, Day RN, Pittler SJ, Valentine DL, Scammell JG 1996 Cell-specific expression of the rat secretogranin II promoter. Endocrinology 137:3815–3822
Hubler TR, Denny WB, Valentine DL, Cheung-Flynn J, Smith DF, Scammell JG 2003 The FK506-binding immunophilin FKBP51 is transcriptionally regulated by progestin and attenuates progestin responsiveness. Endocrinology 144:2380–2387
Russell LC, Whitt SR, Chen MS, Chinkers M 1999 Identification of conserved residues required for the binding of a tetratricopeptide repeat domain to heat shock protein 90. J Biol Chem 274:20060–20063
Ramsey AJ, Russell LC, Whitt SR, Chinkers M 2000 Overlapping sites of tetratricopeptide repeat protein binding and chaperone activity in heat shock protein 90. J Biol Chem 275:17857–17862
Fischer S, Michnick S, Karplus M 1993 A mechanism for rotamase catalysis by the FK506 binding protein (FKBP). Biochemistry 32:13830–13837
Janowski B, Wollner S, Schutkowski M, Fischer G 1997 A protease-free assay for peptidyl prolyl cis/trans isomerases using standard peptide substrates. Anal Biochem 252:299–307
Riggs DL, Roberts PJ, Chirillo SC, Cheung-Flynn J, Prapapanich V, Ratajczak T, Gaber R, Picard D, Smith DF 2003 The Hsp90-binding peptidylprolyl isomerase FKBP52 potentiates glucocorticoid signaling in vivo. EMBO J 22:1158–1167
Davies TH, Ning YM, Sanchez ER 2002 A new first step in activation of steroid receptors: hormone-induced switching of FKBP51 and FKBP52 immunophilins. J Biol Chem 277:4597–4600
Davies TH, Ning YM, Sanchez ER 2005 Differential control of glucocorticoid receptor hormone-binding function by tetratricopeptide repeat (TPR) proteins and the immunosuppressive ligand FK506. Biochemistry 44:2030–2038
Smith DF, Albers MW, Schreiber SL, Leach KL, Deibel Jr MR 1993 FKBP54, a novel FK506-binding protein in avian progesterone receptor complexes and HeLa extracts. J Biol Chem 268:24270–24273
Tai PK, Chang H, Albers MW, Schreiber SL, Toft DO, Faber LE 1993 P59 (FK506 binding protein 59) interaction with heat shock proteins is highly conserved and may involve proteins other than steroid receptors. Biochemistry 32:8842–8847
Wochnik GM, Ruegg J, Abel GA, Schmidt U, Holsboer F, Rein T 2005 FK506-binding proteins 51 and 52 differentially regulate dynein interaction and nuclear translocation of the glucocorticoid receptor in mammalian cells. J Biol Chem 280:4609–4616
Dumont FJ 2000 FK506, an immunosuppressant targeting calcineurin function. Curr Med Chem 7:731–748
Croxtall JD, Paul-Clark M, Van Hal PT 2003 Differential modulation of glucocorticoid action by FK506 in A549 cells. Biochem J 376:285–290
Keightley MC, Curtis AJ, Chu S, Fuller PJ 1998 Structural determinants of cortisol resistance in the guinea pig glucocorticoid receptor. Endocrinology 139:2479–2485
Fuller PJ, Smith BJ, Rogerson FM 2004 Cortisol resistance in the New World revisited. Trends Endocrinol Metab 15:296–299
Adams JS, Chen H, Chun R, Gacad MA, Encinas C, Ren S, Nguyen L, Wu S, Hewison M, Barsony J 2004 Response element binding proteins and intracellular vitamin D binding proteins: novel regulators of vitamin D trafficking, action and metabolism. J Steroid Biochem Mol Biol 89–90:461–465
Chen H, Hu B, Allegretto EA, Adams JS 2000 The vitamin D response element-binding protein. A novel dominant-negative regulator of vitamin D-directed transactivation. J Biol Chem 275:35557–35564(Wesley B. Denny, Viravan )
Address all correspondence and requests for reprints to: Jonathan G. Scammell, Ph.D., Department of Pharmacology, MSB 3370, University of South Alabama, Mobile, Alabama 36688. E-mail: jscammel@jaguar1.usouthal.edu.
Abstract
FK506-binding protein 51 (FKBP51) and FKBP52 are large molecular weight immunophilins that are part of the mature glucocorticoid receptor (GR) heterocomplex. These proteins possess peptidyl-prolyl isomerase (PPIase) and tetratricopeptide repeats (TPR) domains that are important for modulation of GR activity. A naturally occurring animal model of glucocorticoid resistance, the squirrel monkey, results from the relative overexpression of FKBP51 that renders the GR in a low-affinity state. In vitro studies demonstrated that the squirrel monkey form of FKBP51 is greater than 6-fold more potent than human FKBP51 in this respect. The goals of these studies were to determine the roles of the TPR and PPIase domains in the inhibitory activity of squirrel monkey FKBP51 and to gain insight into structural features of squirrel monkey FKBP51 responsible for potent inhibition of dexamethasone-stimulated GR activity. Mutations in the TPR of squirrel monkey FKBP51 that inhibit association with heat shock protein 90 blocked GR inhibitory activity. Mutations that abrogate the PPIase activity of squirrel monkey FKBP51 had no effect on GR inhibitory activity. Chimeras of squirrel monkey and human FKBP51 were tested to identify domains responsible for their different inhibitory potencies. Amino acid differences in domains FK1 and FK2 between squirrel monkey and human FKBP51 contribute equally to the enhanced inhibitory activity of squirrel monkey FKBP51. Furthermore, squirrel monkey FKBP51 in which either FK1 or FK2 was deleted lacked GR inhibitory activity. Thus, the potent inhibitory activity of squirrel monkey FKBP51 involves both FK domains and the heat shock protein 90-binding TPR domain.
Introduction
SQUIRREL MONKEYS have high circulating levels of cortisol to compensate for the expression of glucocorticoid receptors (GR) with low activity (1). The initial observation of high cortisol levels in squirrel monkeys was made in 1970 (2), but it was 1999 before the factors responsible for lower GR activity in squirrel monkeys were identified. Using squirrel monkey B-lymphoblasts (SML) as a cell model, we initially observed that the levels of FK506-binding protein 51 (FKBP51) and FKBP52, large molecular weight FK506-binding immunophilins that compete for binding sites within the GR complex (3), were quite different from levels in human B-lymphoblasts (4). FKBP51 was found to be greater than 10-fold higher in squirrel monkey cells, whereas FKBP52 was less than half that in human cells. Similar changes in FKBP51 and FKBP52 were observed in other cell lines and in liver samples from squirrel monkeys (5, 6). That the relative overexpression of FKBP51 can cause glucocorticoid resistance was demonstrated by Denny et al. (7), who showed that expression of squirrel monkey FKBP51 dramatically increases the EC50 for dexamethasone-stimulated GR transactivation in heterologous COS-7 cells. Human FKBP51 is also inhibitory, but has less than one sixth of the activity of the squirrel monkey protein. Thus, there is convincing evidence that glucocorticoid resistance in squirrel monkeys results, at least in part, from overexpression of a potent form of FKBP51.
FKBP51 and FKBP52 exhibit sequence homology and similar structural organization. The crystal structures of at least portions of both proteins have been recently resolved (8, 9, 10). The N-terminal domain of each protein binds FK506 and has peptidyl-prolyl isomerase (PPIase) activity that converts prolyl peptide bonds within target proteins from cis- to trans-proline (11). This domain, termed FK1 for first FKBP domain, is followed by a second FKBP domain, FK2, that is structurally similar to FK1 but is catalytically inactive (8, 12). The C-terminal domains of FKBP51 and FKBP52 contain three tetratricopeptide repeats (TPR) that are highly degenerate 34-amino-acid repeats involved in protein-protein interactions. An established protein partner of the large molecular FKBPs is heat shock protein 90 (Hsp90), which mediates the interaction of FKBP51, FKBP52, and other co-chaperones with steroid hormone receptor heterocomplexes (13). However, regions outside the TPR domains can influence FKBP binding to Hsp90 (12, 14). The C termini of the FKBPs also contain a calmodulin binding site-like motif sequence (15), although calmodulin regulation of FKBP function has not been demonstrated.
The molecular details of how overexpression of FKBP51 in squirrel monkeys leads to decreased GR binding are largely unknown. Initial observations indicated that squirrel monkey SML cells exhibit an abundance of FKBP51 in Hsp90 complexes (4). We also demonstrated that incorporation of squirrel monkey FKBP51 into GR heterocomplexes is associated with a decrease in GR binding (4). These results suggested that the effect of FKBP51 is mediated through interaction with Hsp90 within the GR heterocomplex. However, direct evidence for the importance of this interaction was lacking, as were data on the role of PPIase activity in the inhibitory activity of FKBP51. Furthermore, very little was known of the structural features of squirrel monkey FKBP51 that render it a potent inhibitor of GR activity. In this study, we investigated the roles of the TPR and PPIase-like domains (FK1 and FK2) in FKBP51’s GR inhibitory activity and, by testing chimeras of squirrel monkey and human FKBP51, have identified domains that are important for their differing potencies.
Materials and Methods
Materials
Culture medium was obtained from Life Technologies (Grand Island, NY). Defined and charcoal-dextran-treated fetal bovine serum was purchased from HyClone Laboratories, Inc. (Logan, UT). Dexamethasone and protease inhibitor cocktail were purchased from Sigma Chemical Co. (St. Louis, MO). The mouse mammary tumor virus (MMTV) promoter-luciferase reporter vector was provided by Dr. R. M. Evans (The Salk Institute, La Jolla, CA). The FLAG-tagged rabbit FKBP52 expression vector (FKBP52-FLAG-pCMV5) was provided by Dr. Michael Chinkers (University of South Alabama, Mobile, AL). Antibody to Hsp90, SPA-835, was purchased from StressGen Biotechnologies Corp. (Victoria, Canada), and anti-FLAG M2 antibody was from Stratagene (La Jolla, CA). Polyclonal antibody to human FKBP51 was purchased from Affinity BioReagents, Inc. (Golden, CO).
Cell cultures
COS-7 cells were grown as monolayers in DMEM supplemented with 10% fetal bovine serum, 50 U/ml penicillin G, and 0.05 mg/ml streptomycin. Cells were grown at 37 C in a humidified atmosphere of 5% CO2/95% air. Cells were transfected using a modification of the method of Bodwell et al. (16), as described previously (7). In most experiments, medium was replaced 18 h after transfection, and cells were treated with dexamethasone (0.1–1 nM). After 24 h, cells were lysed and assayed for luciferase activity as described (17). EC50 values (defined as the concentration of ligand that produces 50% of the maximum response) were obtained from the concentration-response curves. Unless otherwise indicated, the EC50 values are the mean ± range of two independent experiments.
Cell lysates were also used to determine the levels of expressed FKBP51 and FKBP52 by Western blot. Briefly, lysates were mixed 1:1 in 2x sample buffer, separated by SDS-PAGE, and transferred to nitrocellulose. The blots were incubated at 4 C overnight with the anti-FLAG M2 antibody, anti-FKBP51 antibody, or anti-Hsp90 antibody. After washing, blots were incubated with second antibody and developed using an Immune-Star chemiluminescent kit (Bio-Rad Laboratories, Hercules, CA).
Plasmids
The construction of hGR-pcDNA1.1/Amp has been described (7). hGR-pUB6/V5-His was constructed from hGR-pcDNA1.1/Amp by excision with BamHI and XbaI and ligation into pUB6/V5-His (Invitrogen, Carlsbad, CA). The construction of expression plasmids containing either wild-type or FLAG-tagged squirrel monkey (sm51-pCI-neo, sm51FLAG-pCI-neo) or human (h51-pCI-neo, h51FLAG-pCI-neo) FKBP51 cDNA was described previously (4, 18). sm51FLAG-pCI-neo (Lys352Ala, Arg356Ala) was constructed using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) using sm51FLAG-pCI-neo as template and primers (nucleotide mutations italic): GTG CCA ATG AGG CAG GCC TGT ATG CGA GAG GGG AAG CCC (sense) and GGG CTT CCC CTC TCG CAT ACA GGC CTG CCT CAT TGG CAC (antisense), corresponding to nucleotide positions 1063–1101 in the squirrel monkey cDNA (AF140759). sm51FLAG-pCI-neo (Phe67Asp, Asp68Val) was constructed using sm51FLAG-pCI-neo as template and primers (nucleotide mutations are italic): CGG AAA ATT GGC AAA TGG AAA GAA GGA CGT CTC CAG TCA TGA TAG AAA TGA ACC G (sense) and CGG TTC ATT TCT ATC ATG ACT GGA GAC GTC CTT CTT TCC ATT TGC CAA TTT TCC G (antisense), positions 194–248 in the squirrel monkey cDNA.
Seven squirrel monkey/human FKBP51 chimeras were generated. The first, [h1–15]smFKBP51, was generated by site-directed mutagenesis using sm51-pCI-neo as template and primers (nucleotide mutations are italic) GGT GCC AAG AAC AAT GAA GAA AGC CCC ACA GCT ACT GTT GC (sense) and GCA ACA GTA GCT GTG GGG CTT TCT TCA TTG TTC TTG GCA CC (antisense), positions 36–76 in the squirrel monkey cDNA. Chimeras [h1–277]smFKBP51 and [sm1–277]hFKBP51 were produced by excising cDNAs encoding the N-terminal 277 amino acids of h51- and sm51-pCI-neo plasmids with EcoRI and BstZI from each plasmid, taking advantage of a unique BstZI site at codon 277. The resulting fragments were gel purified and cross-ligated to produce cDNAs encoding the chimeric proteins. Chimera [h26–133]smFKBP51 was generated as follows. An SpeI site was introduced by site-directed mutagenesis at codon 26 of wild-type human and squirrel monkey FKBP51 plasmids. The resulting human FKBP51 plasmid was double digested with SpeI and SacI, taking advantage of a unique SacI site at codon 133. [h133–361]smFKBP51 was generated by double digestion of wild-type human FKBP51 plasmid with SacI and PvuII, taking advantage of a unique PvuII site at codon 361. Fragments resulting from the digestions of the human FKBP51 plasmid were gel purified and ligated into the appropriately digested squirrel monkey FKBP51 plasmids.
The squirrel monkey FKBP51 FK1 deletion mutant (smFKBP51FK1), in which amino acids 33–141 were deleted, was generated by first introducing unique ApaI sites into sm51FLAG-pCI-neo using primers (nucleotide mutations are italic) CCT CCA AGA AAG ACA GGG CCC TAT TAA AGA TTG TCA AAA GAG TG (sense) and CAC TCT TTT GAC AAT CTT TAA TAG GGC CCT GTC TTT CTT GGA GG (antisense) and GCT TCT TGA TTT CAA AGG AGG GGC CCT ACT TGA AGA TGG AGG C (sense) and GCC TCC ATC TTC AAG TAG GGC CCC TCC TTT GAA ATC AAG AAG C (antisense), corresponding to nucleotide positions 97–140 and 419–461 in the squirrel monkey cDNA (AF140759), respectively. The mutagenized plasmid DNA was digested with ApaI, the FK1 fragment separated by electrophoresis, and the remaining plasmid DNA was purified and religated. The squirrel monkey FKBP51 FK2 deletion mutant (smFKBP51FK2), in which amino acids 147–251 were deleted, was generated by introducing unique AscI sites into sm51FLAG-pCI-neo using primers (nucleotide mutations are italic) GGA TTT ACT TGA AGA TGG GCG CGC CAT CCG GAG AAC TAA ACG (sense) and CGT TTA GTT CTC CGG ATG GCG CGC CCA TCT TCA AGT AAA TCC (antisense) and CAC TTA AGA GCT TCG GGC GCG CCA AAG AAT CCT GGG AG (sense) and CTC CCA GGA TTC TTT GGC GCG CCC GAA GCT CTT AAG TG (antisense), corresponding to nucleotide positions 440–481 and 757–794 in the squirrel monkey cDNA (AF140759), respectively. The mutagenized plasmid DNA was digested with AscI, the FK2 fragment separated by electrophoresis, and the remaining plasmid DNA was purified and religated.
Immunoprecipitation of FLAG-tagged FKB51
Transfected COS-7 cells were washed twice with cold PBS, collected in cold HEPES-buffered saline (HEBS: 20 mM HEPES, 1 mM EDTA, 150 mM NaCl, pH 7.4), and lysed by sonication. After the addition of protease inhibitor cocktail (10 μl/ml), the cell lysates were centrifuged at 21,000 x g for 1 h at 4 C, and the soluble fraction was removed and stored on ice. Aliquots of a 50% anti-FLAG M2 affinity resin slurry in HEBS was added to each cell lysate and the mixture rotated for 2 h at 4 C. The resin was pelleted, washed three times with ice-cold HEBS, and resuspended in 2x nonreducing sample buffer. The samples were boiled and centrifuged, and the supernatants were subjected to SDS-PAGE and Western blot analysis.
Results
Role of TPR and PPIase domains
The first series of experiments was designed to determine whether the inhibitory effect of squirrel monkey FKBP51 on GR activity is mediated through interaction with Hsp90 and the GR heterocomplex. To test this, basic residues in the TPR domain, previously shown to be essential for binding of protein phosphatase 5, FKBP52, and FKBP51 to Hsp90 (14, 19, 20), were mutated in squirrel monkey FKBP51 to alanine (Lys352Ala, Arg356Ala). The effect of the resulting mutant protein on GR activation was then evaluated in cell culture. COS-7 cells were transfected with a plasmid expressing human GR, an MMTV-luciferase reporter plasmid, and either an empty vector or a vector containing FLAG-tagged squirrel monkey FKBP51 cDNA or FLAG-tagged squirrel monkey FKBP51 cDNA containing mutations in the TPR region. Treatment of cells transfected with empty vector with dexamethasone (0.3–100 nM) produced a concentration-dependent GR transactivation response with an EC50 of 1.1 ± 0.3 nM (Fig. 1A). Expression of FLAG-tagged squirrel monkey FKBP51 resulted in an expected rightward shift in the dexamethasone concentration-response curve (EC50 of 13.6 ± 1.4 nM). FLAG-tagged squirrel monkey FKBP51 has the same inhibitory activity as wild-type squirrel monkey FKBP51 (data not shown). The inhibitory effect of squirrel monkey FKBP51 was largely abrogated when mutations were made in the TPR region (EC50 of 1.8 ± 0.7 nM). We confirmed that this mutant does not coprecipitate Hsp90 (Fig. 1B, top), even though its expression was slightly higher than the wild-type protein (bottom).
FIG. 1. A, Effect of squirrel monkey FKBP51 TPR domain mutant (Lys352Ala, Arg356Ala) on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected by electroporation with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either FLAG-tagged wild-type squirrel monkey FKBP51 (sm51wt, ) or FLAG-tagged squirrel monkey FKBP51 TPR domain mutant (sm51Tm, ). Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. B, Precipitation of Hsp90 by sm51wt, but not by sm51Tm. FLAG-tagged FKBP51 proteins from soluble extracts of transfected COS-7 cells were immunoadsorbed by anti-FLAG affinity resin, washed, eluted with sample buffer, and analyzed for Hsp90 and the FLAG epitope on FKBP51 by Western blot.
The second set of experiments was designed to determine whether the inherent PPIase activity of squirrel monkey FKB51 is an important component of its inhibitory effect on GR activity. A PPIase inactive mutant of squirrel monkey FKBP51 was generated by site-directed mutagenesis of two residues in the catalytic binding pocket, Phe67Asp, Asp68Val. These residues are highly conserved among FKBPs and are required for PPIase activity (21). Mutation of these residues in human FKBP51 was shown to reduce PPIase activity by greater than 90% (12). We found that expression of the squirrel monkey PPIase mutant potently inhibited GR transactivation in COS-7 cells in a manner similar to that seen with wild-type squirrel monkey FKBP51 (Fig. 2A). Using the procedure of Janowski et al. (22), we confirmed that this double mutant of squirrel monkey FKBP51 exhibited less than 5% of the PPIase activity of the wild-type protein (data not shown). The slightly less inhibitory activity of the PPIase mutant (EC50 of 3.7 ± 0.8 nM) compared with wild-type squirrel monkey FKBP51 (EC50 of 5.8 ± 2.9 nM) was attributed to slightly lower expression of the PPIase mutant (Fig. 2B). In this and other experiments, in which polyclonal antibody to FKBP51 was used to detect expressed squirrel monkey FKBP51, a slower migrating band is sometimes apparent. This band reflects the endogenous expression of FKBP51 in COS-7 cells. Squirrel monkey FKBP51 migrates more rapidly than African green monkey or human FKBP51 on SDS gels (7).
FIG. 2. A, Effect of squirrel monkey FKBP51 PPIase mutant (Phe67Asp, Asp68Val) on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected by electroporation with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type squirrel monkey FKBP51 (sm51wt, ) or squirrel monkey FKBP51 PPIase mutant (sm51Pm, ). Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. B, Expression of sm51wt and sm51Pm in electroporated COS-7 cells. Cell lysates from each transfected set were pooled, and Western blots were performed for FKBP51.
As immunophilins compete via their TPR domains for a TPR acceptor site on Hsp90 in steroid receptor complexes (recently reviewed in Refs. 3 , 13), we investigated whether overexpression of FKBP52 might abrogate the inhibitory effect of squirrel monkey FKBP51 on GR transactivation. Dexamethasone dose-response curves were performed in COS-7 cells expressing squirrel monkey FKBP51, rabbit FKBP52, or a combination of the two immunophilins. Expression of FKBP52 alone resulted in a slight leftward shift in the dose-response curve (EC50 of 0.2 ± 0.1 nM) compared with the control curve (EC50 of 0.4 ± 0.2 nM) (Fig. 3A). More dramatically, the rightward shift in the dose-response curve induced by squirrel monkey FKBP51 expression (EC50 of 4.4 ± 0.3 nM) was blocked when FKBP52 was coexpressed (EC50 of 0.4 ± 0.1 nM). Similar expression of the FLAG-tagged FKBPs was verified by immunodetection of the FLAG epitope in cell lysates collected from each transfected set of COS-7 cells (Fig. 3B). Together, the results of these experiments suggest that inherent PPIase activity does not play a role in the inhibitory effect of squirrel monkey FKBP51 on GR activity, but interaction with Hsp90 and the GR heterocomplex through FKBP51’s TPR domain is critical.
FIG. 3. A, Overexpression of FKBP52 blocks the inhibitory effect of FKBP51 on GR transactivation in COS-7 cells. COS-7 cells were transfected by electroporation with hGR-pUB6/V5-His, MMTV-luciferase; either empty pCIneo empty vector or pCIneo containing cDNA encoding FLAG-tagged squirrel monkey FKBP51; and either empty pCMV5 vector or pCMV5 containing cDNA encoding FLAG-tagged rabbit FKBP52. Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. , control; , rabbit FKBP52 (r52); , wild-type squirrel monkey FKBP51 (sm51); , rabbit FKBP52 and wild-type squirrel monkey FKBP51 (r52+sm51). B, Expression of FLAG-tagged FKBP52 (FLAG-52) and FKBP51 (FLAG-51) in electroporated COS-7 cells. Cell lysates from each transfected set were pooled, and Western blots were performed for the FLAG epitope.
Activities of squirrel monkey/human FKBP51 chimeras
The following experiments were designed to determine the structural basis for the enhanced inhibitory activity of squirrel monkey FKBP51. Although squirrel monkey and human FKBP51 are 94% identical, squirrel monkey FKBP51 is greater than 6-fold more potent than human FKBP51 in inhibiting GR activity (7). Our initial approach was to generate point mutants of squirrel monkey or human FKBP51 and compare their activities to wild-type proteins. We found that no single amino acid change in either of the proteins significantly affected their relative potencies (data not shown), suggesting that multiple amino acid differences between squirrel monkey and human FKBP51 likely contribute to their different potencies. To address this issue, we generated chimeras of squirrel monkey and human FKBP51 and compared their inhibitory activities on dexamethasone-induced GR transactivation in COS-7 cells. Details of these chimeras are shown in Fig. 4.
FIG. 4. A, Diagram showing the relative positions of functional domains in FKBP51. The two FKBP12-like domains are indicated; FK1 is the PPIase and FK506-binding domain, whereas the related FK2 lacks PPIase and drug-binding activities. The TPR domain is the tetratricopeptide repeat that mediates interaction with Hsp90. B, Diagram of human and squirrel monkey FKBP51 chimeric proteins. The shaded boxes represent human FKBP51 sequence, whereas the open boxes represent squirrel monkey FKBP51 sequence.
Because squirrel monkey and human FKBP51 differ in five amino acids in the N-terminal region of the protein (positions 10–13 and 15), the first chimera tested [h1–15]smFKBP51 included the N-terminal 15 amino acids of human FKBP51 and the C-terminal 442 amino acids of squirrel monkey FKBP51. We found that expression of the [h1–15]smFKBP51 chimera resulted in a rightward shift in the dexamethasone concentration-response curve (EC50 of 8.1 ± 0.1 nM), similar to that seen with wild-type squirrel monkey FKBP51 (EC50 of 8.1 ± 0.4 nM) (Fig. 5A, top). The EC50 in COS-7 cells transfected with empty vector was 1.6 ± 0.3 nM. Western blots confirmed similar expression of wild-type and chimeric FKBP51 (Fig. 5A, bottom). These results suggested that the differences between human and squirrel monkey FKBP51 in the extreme N terminus do not contribute to the enhanced inhibitory activity of squirrel monkey FKBP51.
FIG. 5. A, Top, Effect of the human/squirrel monkey FKBP51 N-terminal chimera, [h1–15]smFKBP51, on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type squirrel monkey FKBP51 (sm51wt, ) or a chimera of FKBP51 containing the N-terminal 15 amino acids of human FKBP51 and the C-terminal 442 amino acids of squirrel monkey FKBP51 (h15sm, ). Bottom, Expression of sm51wt and h15sm compared with control in electroporated COS-7 cells. B, Top, Effect of human/squirrel FKBP51 chimeras, [h1–277]smFKBP51 and [sm1–277]hFKBP51, on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type squirrel monkey FKBP51 (sm51wt, ), wild-type human FKBP51 (h51wt, ), or chimeras of FKBP51 containing either the N-terminal 277 amino acids of human FKBP51 and the C-terminal 180 amino acids of squirrel monkey FKBP51 (h277sm, ) or the N-terminal 277 amino acids of squirrel monkey FKBP51 and the C-terminal 180 amino acids of human FKBP51 (sm277 h, ). Bottom, Expression of h51wt, h277sm, sm51wt, and sm277h compared with control in electroporated COS-7 cells. For the Western blots shown in both A and B, cell lysates from each transfected set were pooled and analyzed for FKBP51.
We next tested the effects of swapping the N-terminal 277 amino acids of human and squirrel monkey FKBP51. These chimeras, [sm1–277]hFKBP51 and [h1–277]smFKBP51, include both FK1 and FK2 domains of either squirrel monkey or human FKBP51, respectively (Fig. 4). We found that expression of the [sm1–277]hFKBP51 chimera resulted in a rightward shift in the dexamethasone concentration-response curve (EC50 of 10.6 ± 0.1 nM) that was similar to that observed with wild-type squirrel monkey FKBP51 (EC50 of 11.4 ± 1.1 nM) (Fig. 5B, top). Expression of the [h1–277]smFKP51 chimera resulted in a less dramatic shift in the dexamethasone concentration-response curve (EC50 of 2.7 ± 0.2 nM), similar to that seen with wild-type human FKBP51 (EC50 of 2.2 ± 0.6 nM). The EC50 in COS-7 cells transfected with empty vector was 1.4 ± 0.6 nM. Western blots confirmed similar expression of wild-type and chimeric FKBP51 (Fig. 5B, bottom). These results suggested that the enhanced inhibitory potency of squirrel monkey FKBP51 results from amino acid differences in the N-terminal half of the protein. Differences between squirrel monkey and human FKBP51 in the TPR domain and C terminus of the proteins are likely irrelevant.
We next determined the contribution of the individual FK domains to the enhanced potency of squirrel monkey FKBP51. We tested the activities of chimeras of human and squirrel monkey FKBP51 in which most of the FK1 domain, [h26–133]smFKBP51, or the entire FK2 domain of squirrel monkey FKBP51, [h133–361]smFKBP51, was exchanged with the same region of human FKBP51 (see Fig. 4). Expression of [h26–133]smFKBP51 resulted in inhibition of hormone-stimulated GR activity that was greater (EC50 of 3.7 ± 0.6 nM) than that achieved with human FKBP51 (EC50 of 1.7 ± 0.2 nM) but not as striking as that observed with squirrel monkey FKBP51 (EC50 of 8.6 ± 0.6 nM) (Fig. 6A). The inhibitory potency of [h133–361]smFKBP51 was also intermediate between human and squirrel monkey FKBP51 (EC50 of 4.0 ± 0.3 nM). Western blots confirmed similar expression of the FKBP51 chimeras (Fig. 6B). These results suggest that amino acid differences between human and squirrel monkey FKBP51 in both FK domains contribute to their differing inhibitory potencies on GR transactivation.
FIG. 6. A, Effect of human/squirrel monkey FKBP51 FK1 and FK2 chimeras, [h26–133]smFKBP51 and [h133–361]smFKBP51, on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type human (h51wt, ) or squirrel monkey (sm51wt, ) FKBP51 or human/squirrel monkey FKBP51 chimeras that have human FKBP51 sequence in positions 26–133 (hFK1, ) or 133–361 (hFK2, ) within squirrel monkey FKBP51. Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. B, Expression of h51wt, sm51wt, hFK1, and hFK2 compared with control in electroporated COS-7 cells. Cell lysates from each transfected set were pooled and analyzed for FKBP51 by Western blot.
Effect of deletion of FK1 or FK2 on the activity of squirrel monkey FKBP51
The finding that amino acids in FK1 and FK2 influence the activities of the two forms of FKBP51 prompted us to investigate the effect of simply deleting these domains on the activity of squirrel monkey FKBP51. Two mutants were generated: smFKBP51FK1 in which residues 22–141 were deleted and smFKBP51FK2 in which residues 147–251 were deleted, resulting in the complete removal of either the FK1 or FK2 domain (see Fig. 4A). We found that deletion of either of the two FK domains completely eliminated the inhibitory activity of squirrel monkey FKBP51 in transfected, dexamethasone-treated COS-7 cells (Fig. 7A). The EC50 values of the FK1 and FK2 deletion mutants were 0.4 and 0.3 nM, respectively, compared with that achieved in cells transfected with empty vector or wild-type squirrel monkey FKBP51 (0.4 and 11 nM, respectively). Western blots confirmed similar protein expression from the different FKBP51 constructs (Fig. 7B). The lack of inhibitory activity of these deletion mutants did not result from loss of Hsp90 binding because Hsp90 was coprecipitated as efficiently with squirrel monkey FKBP51 lacking FK1 as with the wild-type protein (Fig. 7C).
FIG. 7. A, Effect of deletions of either FK1 or FK2 in squirrel monkey FKBP51 on dexamethasone-stimulated GR transactivation in COS-7 cells. COS-7 cells were transfected with hGR-pcDNA1.1/Amp, MMTV-luciferase, and either pCIneo empty vector (control, ) or pCIneo containing cDNA encoding either wild-type squirrel monkey FKBP51 (sm51wt, ), or squirrel monkey FKBP51 in which either the FK1 (FK1, ) or FK2 (FK2, ) domains were deleted. Transfected cells were treated with dexamethasone (Dex) for 24 h, and cell lysates were collected for assay of luciferase activity. Each point represents the result from a single experiment. B, Expression of sm51wt, FK1, and FK2 compared with control in electroporated COS-7 cells. Cell lysates from each transfected set were pooled and analyzed for FKBP51 by Western blot. C, Precipitation of Hsp90 by sm51wt and FK1. FLAG-tagged FKBP51 proteins from soluble extracts of transfected COS-7 cells were immunoadsorbed with anti-FLAG affinity resin, washed, eluted with sample buffer, and analyzed for Hsp90 and the FLAG epitope on FKBP51 by Western blot.
Discussion
In 2003, the Smith laboratory took advantage of the null background of Saccharomyces cerevisiae yeast to directly test the roles of FKBP51 and FKBP52 in regulation of GR activity (23). When human FKBP52 was introduced into yeast expressing GR, hormone-dependent reporter activity was enhanced, reflecting increased GR ligand-binding affinity. It was found that this effect was quite specific to FKBP52 and requires both the Hsp90-binding and PPIase activities of the protein. When human FKBP51 was coexpressed, it completely blocked the stimulatory effect of FKBP52 on GR activity. However, studies to determine which domains or activities of human FKBP51 were responsible for its inhibitory activity in yeast were not performed. Here, we used African green monkey COS-7 cells, which constitutively express relatively high levels of FKBP52 and low levels of FKBP51 (4), to transiently express different mutants of squirrel monkey FKBP51, a particularly potent form of FKBP51 (7). We demonstrated that the Hsp90-binding, but not PPIase, activity of FKBP51 is necessary for its inhibitory effect on GR function.
Complete details of how FKBP52 and FKBP51 affect GR-binding activity are not known. The mature GR heterocomplex may contain either FKBP52 or FKBP51, but ligand binding appears to foster selective association of FKBP52 with the complex (24, 25). It is speculated that within the complex, the FKBP52 PPIase targets one or more prolines in the ligand-binding domain of GR and optimizes the conformation for high-affinity hormone binding (3). FKBP51 also has PPIase activity, but biochemical studies have suggested that FKBP52 and FKBP51 interact with steroid receptors differently (26, 27). The results of more recent studies have taken into account the x-ray crystal structure of FKBP51 (8) and suggest that regions of FKBP51 and FKBP52, C-terminal to the TPR domain, may assume alternative conformations that differentially affect their interaction with Hsp90 (14). Hence, the GR ligand-binding domain may be inaccessible to the active site of FKBP51, and FKBP51 acts as an inhibitor. Furthermore, the inhibitory activity of FKBP51 was quite specific because the coexpression of another TPR-containing protein, protein phosphatase 5, did not affect potentiation by FKBP52 (23). Here we demonstrate that squirrel monkey FKBP51 interaction with the receptor heterocomplex through Hsp90 is critical, but that PPIase activity, which is localized to FK1 (8, 12), is dispensable. On the other hand, we found that deletion mutants involving the FK domains abrogate FKBP51’s effects. Therefore, the inhibitory activity of FKBP51 involves both FK domains plus the Hsp90-binding TPR domain.
A role for the FK domains in the inhibitory activity of FKBP51 on GR transactivation was also implicated in a very recent report by Wochnik et al. (28), who showed expression of FKBP51/FKBP52 chimeras containing either FK1 or FK2 of human FKBP51 inhibited GR transactivation, but neither was as efficacious as wild-type FKBP51. These investigators demonstrated that Hsp90 binding, but not PPIase activity, was essential for the inhibitory effect of human FKBP51 on GR activity, consistent with our results with squirrel monkey FKBP51 on GR activity reported here and with our results with human FKBP51 on progesterone receptor transactivation shown previously (18). Wochnik et al. (28) also showed that nuclear translocation of hormone-activated GR is slowed if human FKBP51 is overexpressed. That is, FKBP51 may disrupt the normal interaction of FKBP52 with the motor protein dynein, thought to be integral to movement of GR from cytoplasm to nucleus (13). However, at physiologically relevant hormone concentrations, when compared with the effect of FKBP51 on receptor binding, the contribution of a change in the rate of nuclear translocation to the overall inhibitory action of FKBP51 on GR signaling may be quite modest. This is even more evident when describing the effect of squirrel monkey FKBP51, which we have shown is a more potent inhibitor of receptor binding than human FKBP51 (4, 7).
Indeed, another goal of this work was to gain insight into the difference in inhibitory activities of squirrel monkey and human FKBP51. FKBP51 from other New World primates has similar activity to squirrel monkey FKBP51 (5), whereas the activity of FKBP51 cloned from an Old World primate, African green monkey, was similar to that of human FKBP51 (data not shown). These results suggest that structural differences between New World and Old World FKBP51 have evolved and led to functional differences in their abilities to repress steroid receptor activity. Our examination of chimeric FKBP51 proteins indicates that the relevant amino acid changes are localized to the FK domains. If these changes are mapped onto the crystal structure of squirrel monkey FKBP51 (Fig. 8), it becomes apparent that the New World-specific amino acids in FK1 and FK2 are surface residues that likely do not impact protein conformation. Indeed, comparison of crystal structures for human and squirrel monkey FKBP51 did not suggest significant conformational differences (8). On the other hand, these amino acids might contribute to a novel protein interaction surface involving both FK domains. An intriguing candidate as a potential interaction partner for this surface is FKBP52 because FKBP51 is known to antagonize FKBP52-mediated potentiation of GR activity (23). New World FKBP51, by virtue of amino acid changes in the FK domains, could interact with FKBP52 more strongly than Old World FKBP51 and is thus a better inhibitor of FKBP52-mediated potentiation of receptor activity. An interaction between FKBP51 and FKBP52 might be favored by mutual binding to an Hsp90 dimer within the steroid receptor complex, thereby explaining the observation that an FKBP51 mutant deficient in Hsp90 binding fails to inhibit GR activity. However, in lieu of direct experimental evidence that supports an interaction with FKBP52, one must also consider the receptor itself or Hsp90 as alternative interaction partners for FKBP51. We are currently pursuing studies to identify the putative interaction partner for the FK domain region of FKBP51 and to determine how such an interaction inhibits steroid receptor function.
FIG. 8. X-ray crystallographic structure of FKBP51 FK domain region. Amino acids that are different in human and New World primate FKBP51 are highlighted with dark gray side chains in FK1 [positions 42 (Asn, humanHis, New World primate), 115 (SerAla), and 116 (AlaThr)] and FK2 [positions 167 (ThrArg), 169 (GluGln), 180 (MetVal), and 218 (TyrHis)] domains of squirrel monkey FKBP51. These are surface residues that do not significantly influence FK domain conformation, but could alter interactions with a protein partner that contacts both FK domains.
Although mutational disruption of FKBP51’s PPIase activity affected neither FKBP51 interaction with Hsp90 nor progesterone receptor (12), nor did this mutation affect inhibition in this study or in the study of Wochnik et al. (28), the PPIase inhibitor FK506 has been shown to block the inhibitory effect of FKBP51 on GR activity (4, 7). Two considerations must be taken into account when interpreting the actions of FK506. First, FK506 also blocks the potentiation of GR activity by FKBP52, and FKBP52-mediated potentiation is PPIase-dependent (23). The change in GR activity that results from adding FK506 to cells that contain both FKBP52 (an activator of GR function) and FKBP51 (an inhibitor of GR function) might vary depending on the preexisting balance of cellular FKBP levels. A second consideration is that FK506 is not fully contained within the PPIase pocket of an FKBP; in fact, immunosuppression by FK506:FKBP12 complexes depend on the protruding effector domain of FK506 rather than inhibition of FKBP12 PPIase activity (29). When bound to FK1 of FKBP51, the effector domain of FK506 could sterically interfere with protein-protein interaction involving the FK domain region that is essential for the inhibitory effect on GR activity. We and others have shown that FK506 causes the release of FKBP51 from GR heterocomplexes (4, 25, 30).
We have provided evidence here and elsewhere that the relative overexpression of a potent form of the immunophilin FKBP51 mediates at least in part glucocorticoid and progestin resistance in squirrel monkeys and other New World primates (4, 5, 7, 18). However, this is not the cause of glucocorticoid resistance in the guinea pig, a New World hystricomorph. Fuller’s laboratory showed that at least four amino acid substitutions in the ligand-binding domain contribute to the low binding affinity of the guinea pig GR (31). This group has recently suggested that these changes may inhibit FKBP52 interaction with the receptor complex, compromising the formation of a high-affinity receptor (32). Thus, although the proximal mechanisms for glucocorticoid resistance in New World hystricomorphs and primates are different, the downstream effect (i.e. loss of FKBP52) may be common. On the other hand, vitamin D resistance that is observed in many New World primates likely occurs via a separate mechanism. This form of steroid resistance results from the constitutive overexpression of a member of the heterogeneous nuclear ribonucleoprotein A family, termed vitamin D response-element (VDRE) binding protein 2 (recently reviewed in Ref. 33). VDRE binding protein 2 was isolated from vitamin D-resistant B95-8 cells from cotton-top tamarin (Saguinus oedipus) and shown to compete with the vitamin D receptor-retinoid X receptor heterodimer for binding to the vitamin D response element (33, 34). Thus, this protein does not affect receptor binding, but rather acts to squelch VDRE-directed transactivation.
In summary, the results of this study have provided insight into the structural features of squirrel monkey FKBP51 that render it a potent inhibitor of GR signaling. We demonstrated that binding to Hsp90 is required for the inhibitory effect of squirrel monkey FKBP51. We have also shown that whereas PPIase activity of squirrel monkey FKBP51 is irrelevant, the FK domains are essential. These results suggest that protein-protein interaction involving the FK domains of FKBP51 and another protein partner mediates the inhibitory activity of FKBP51 on GR activity.
References
Chrousos GP, Renquist D, Brandon D, Eil C, Pugeat M, Vigersky R, Cutler Jr GB, Loriaux DL, Lipsett MB 1982 Glucocorticoid hormone resistance during primate evolution: receptor-mediated mechanisms. Proc Natl Acad Sci USA 79:2036–2040
Brown GM, Grota LJ, Penney DP, Reichlin S 1970 Pituitary-adrenal function in the squirrel monkey. Endocrinology 86:519–529
Smith DF 2004 Tetratricopeptide repeat cochaperones in steroid receptor complexes. Cell Stress Chaperones 9:109–121
Reynolds PD, Ruan Y, Smith DF, Scammell JG 1999 Glucocorticoid resistance in the squirrel monkey is associated with overexpression of the immunophilin FKBP51. J Clin Endocrinol Metab 84:663–669
Scammell JG, Denny WB, Valentine DL, Smith DF 2001 Overexpression of the FK506-binding immunophilin FKBP51 is the common cause of glucocorticoid resistance in three New World primates. Gen Comp Endocrinol 124:152–165
Scammell JG, Tucker JA, King JA, Moore CM, Wright JL, Tuck-Muller CM 2002 A kidney epithelial cell line from a Bolivian squirrel monkey. In Vitro Cell Dev Biol Animal 38:258–261
Denny WB, Valentine DL, Reynolds PD, Smith DF, Scammell JG 2000 Squirrel monkey immunophilin FKBP51 is a potent inhibitor of glucocorticoid receptor binding. Endocrinology 141:4107–4113
Sinars CR, Cheung-Flynn J, Rimerman RA, Scammell JG, Smith DF, Clardy J 2003 Structure of the large FK506-binding protein FKBP51, an Hsp90-binding protein and a component of steroid receptor complexes. Proc Natl Acad Sci USA 100:868–873
Li P, Ding Y, Wu B, Shu C, Shen B, Rao Z 2003 Structure of the N-terminal domain of human FKBP52. Acta Crystallogr D Biol Crystallogr 59:16–22
Wu B, Li P, Shu C, Shen B, Rao Z 2003 Crystallization and preliminary crystallographic studies of the C-terminal domain of human FKBP52. Acta Crystallogr D Biol Crystallogr 59:2269–2271
Schiene-Fischer C, Yu C 2001 Receptor accessory folding helper enzymes: the functional role of peptidyl prolyl cis/trans isomerases. FEBS Lett 495:1–6
Barent RL, Nair SC, Carr DC, Ruan Y, Rimerman RA, Fulton J, Zhang Y, Smith DF 1998 Analysis of FKBP51/FKBP52 chimeras and mutants for Hsp90 binding and association with progesterone receptor complexes. Mol Endocrinol 12:342–354
Pratt WB, Galigniana MD, Harrell JM, DeFranco DB 2004 Role of hsp90 and the hsp90-binding immunophilins in signalling protein movement. Cell Signal 16:857–872
Cheung-Flynn J, Roberts PJ, Riggs DL, Smith DF 2003 C-terminal sequences outside the tetratricopeptide repeat domain of FKBP51 and FKBP52 cause differential binding to Hsp90. J Biol Chem 278:17388–17394
Callebaut I, Renoir JM, Lebeau MC, Massol N, Burny A, Baulieu EE, Mornon JP 1992 An immunophilin that binds M(r) 90,000 heat shock protein: main structural features of a mammalian p59 protein. Proc Natl Acad Sci USA 89:6270–6274
Bodwell J, Swiff F, Richardson J 1999 Long duration electroporation for achieving high level expression of glucocorticoid receptors in mammalian cell lines. J Steroid Biochem Mol Biol 68:77–82
Jones LC, Day RN, Pittler SJ, Valentine DL, Scammell JG 1996 Cell-specific expression of the rat secretogranin II promoter. Endocrinology 137:3815–3822
Hubler TR, Denny WB, Valentine DL, Cheung-Flynn J, Smith DF, Scammell JG 2003 The FK506-binding immunophilin FKBP51 is transcriptionally regulated by progestin and attenuates progestin responsiveness. Endocrinology 144:2380–2387
Russell LC, Whitt SR, Chen MS, Chinkers M 1999 Identification of conserved residues required for the binding of a tetratricopeptide repeat domain to heat shock protein 90. J Biol Chem 274:20060–20063
Ramsey AJ, Russell LC, Whitt SR, Chinkers M 2000 Overlapping sites of tetratricopeptide repeat protein binding and chaperone activity in heat shock protein 90. J Biol Chem 275:17857–17862
Fischer S, Michnick S, Karplus M 1993 A mechanism for rotamase catalysis by the FK506 binding protein (FKBP). Biochemistry 32:13830–13837
Janowski B, Wollner S, Schutkowski M, Fischer G 1997 A protease-free assay for peptidyl prolyl cis/trans isomerases using standard peptide substrates. Anal Biochem 252:299–307
Riggs DL, Roberts PJ, Chirillo SC, Cheung-Flynn J, Prapapanich V, Ratajczak T, Gaber R, Picard D, Smith DF 2003 The Hsp90-binding peptidylprolyl isomerase FKBP52 potentiates glucocorticoid signaling in vivo. EMBO J 22:1158–1167
Davies TH, Ning YM, Sanchez ER 2002 A new first step in activation of steroid receptors: hormone-induced switching of FKBP51 and FKBP52 immunophilins. J Biol Chem 277:4597–4600
Davies TH, Ning YM, Sanchez ER 2005 Differential control of glucocorticoid receptor hormone-binding function by tetratricopeptide repeat (TPR) proteins and the immunosuppressive ligand FK506. Biochemistry 44:2030–2038
Smith DF, Albers MW, Schreiber SL, Leach KL, Deibel Jr MR 1993 FKBP54, a novel FK506-binding protein in avian progesterone receptor complexes and HeLa extracts. J Biol Chem 268:24270–24273
Tai PK, Chang H, Albers MW, Schreiber SL, Toft DO, Faber LE 1993 P59 (FK506 binding protein 59) interaction with heat shock proteins is highly conserved and may involve proteins other than steroid receptors. Biochemistry 32:8842–8847
Wochnik GM, Ruegg J, Abel GA, Schmidt U, Holsboer F, Rein T 2005 FK506-binding proteins 51 and 52 differentially regulate dynein interaction and nuclear translocation of the glucocorticoid receptor in mammalian cells. J Biol Chem 280:4609–4616
Dumont FJ 2000 FK506, an immunosuppressant targeting calcineurin function. Curr Med Chem 7:731–748
Croxtall JD, Paul-Clark M, Van Hal PT 2003 Differential modulation of glucocorticoid action by FK506 in A549 cells. Biochem J 376:285–290
Keightley MC, Curtis AJ, Chu S, Fuller PJ 1998 Structural determinants of cortisol resistance in the guinea pig glucocorticoid receptor. Endocrinology 139:2479–2485
Fuller PJ, Smith BJ, Rogerson FM 2004 Cortisol resistance in the New World revisited. Trends Endocrinol Metab 15:296–299
Adams JS, Chen H, Chun R, Gacad MA, Encinas C, Ren S, Nguyen L, Wu S, Hewison M, Barsony J 2004 Response element binding proteins and intracellular vitamin D binding proteins: novel regulators of vitamin D trafficking, action and metabolism. J Steroid Biochem Mol Biol 89–90:461–465
Chen H, Hu B, Allegretto EA, Adams JS 2000 The vitamin D response element-binding protein. A novel dominant-negative regulator of vitamin D-directed transactivation. J Biol Chem 275:35557–35564(Wesley B. Denny, Viravan )