A β-Turn Motif in the Steroid Hormone Receptor's ... - ACS Publications

Sep 19, 2016 - ABSTRACT: The immunophilin FKBP52 interacts with nuclear steroid hormone receptors. Studying the crystal structure of human estrogen ...
0 downloads 0 Views 5MB Size
Article pubs.acs.org/biochemistry

A β‑Turn Motif in the Steroid Hormone Receptor’s Ligand-Binding Domains Interacts with the Peptidyl-prolyl Isomerase (PPIase) Catalytic Site of the Immunophilin FKBP52 Cillian Byrne,†,‡ Morkos A. Henen,§ Mathilde Belnou,† François-Xavier Cantrelle,§ Amina Kamah,§ Haoling Qi,§ Julien Giustiniani,‡ Béatrice Chambraud,‡ Etienne-Emile Baulieu,‡ Guy Lippens,§,∥ Isabelle Landrieu,§ and Yves Jacquot*,† †

Sorbonne Universités, UPMC Univ Paris 06, Ecole Normale Supérieure, PSL Research University, CNRS UMR 7203, Laboratoire des Biomolécules, 4, place Jussieu, 75252 Paris Cedex 05, France ‡ Institut Baulieu, INSERM UMR 1195, Neuroprotection and Neuroregeneration, Université Paris-Saclay, Bât. Gregory Pincus, 80, rue du Général Leclerc, 94276 Le Kremlin Bicêtre Cedex, France § CNRS, UMR 8576, Glycobiologie Structurale et Fonctionnelle, Université des Sciences et Technologies de Lille 1, 59655 Villeneuve d’Ascq Cedex, France ∥ LISBP, Université de Toulouse, CNRS, INRA, INSA, Toulouse, France S Supporting Information *

ABSTRACT: The immunophilin FKBP52 interacts with nuclear steroid hormone receptors. Studying the crystal structure of human estrogen receptor α (hERα) and using nuclear magnetic resonance, we show here that the short V364PGF367 sequence, which is located within its ligand-binding domain and adopts a type II β-turn conformation in the protein, binds the peptidyl-prolyl isomerase (PPIase or rotamase) FK1 domain of FKBP52. Interestingly, this turn motif displays strong similarities with the FKBP52 FK1 domain-binding moiety of macrolide immunomodulators such as rapamycin and GPI-1046, an immunophilin ligand with neuroprotective characteristics. An increase in the hydrophobicity of the residue preceding the proline and cyclization of the VPGF peptide strengthen its recognition by the FK1 domain of FKBP52. Replacement of the Pro residue with a dimethylproline also enhances this interaction. Our study not only contributes to a better understanding of how the interaction between the FK1 domain of FKBP52 and steroid hormone receptors most likely works but also opens new avenues for the synthesis of FKBP52 FK1 peptide ligands appropriate for the control of hormone-dependent physiological mechanisms or of the functioning of the Tau protein. Indeed, it has been shown that FKBP52 is involved in the intraneuronal dynamics of the Tau protein.

T

C-terminal MEEVD motif of the heat shock protein Hsp90,16,17 and (iv) a putative calmodulin-binding site.18 With a ΔG# of ∼20 kcal mol−1, Xaa-Pro cis−trans isomerization is known to be the most thermodynamically favorable among dipeptide bonds.19 The PPIase enzymatic activity of FKBP52, which is supported by its FK1 domain, operates at such Xaa−Pro dipeptide bonds (where Xaa is a hydrophobic residue and preferentially a leucine20,21) to accelerate the rate at which equilibrium between the two isomers is reached.22 It has been proposed that the immunophilin FKBP52 could interact through its FK1 domain with a part of the ligandbinding domain of steroid hormone receptors.1,23−26 Whereas

he protein FKBP52 (for FK506-binding protein, 52 kDa) is an immunophilin that participates in various physiological processes, including the control of steroid hormone receptor-related transcription,1,2 reproductive functions,3,4 microtubule network dynamics,5 and trafficking of proteins.6 FKBP52 also plays a protective role in the central nervous system.7 Furthermore, it is involved in Tau protein function,8−10 in the control of amyloid β toxicity,10 and in the oligomerization of different Tau protein fragments in vitro.8,9 FKBP52 is composed of four main domains: (i) a wellconserved FK1 domain structurally related to FKBP12 (49% homology), carrying peptidyl-prolyl isomerase (PPIase) activity,11 that binds tacrolimus (FK506) and related immunomodulatory macrolides12 and interacts with steroid hormone receptors,13,14 (ii) an FK2 domain, the function of which has not yet been elucidated but which binds ATP and GTP,15 (iii) a tetratricopeptide repeat (TPR) domain that associates with the © XXXX American Chemical Society

Received: May 19, 2016 Revised: August 10, 2016

A

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 1. Chemical structures of rapamycin, GPI-1046, and the VPGF motif issued from hERα (residues 363−369 of peptide 1). Colored red is the FKBP52 FK1-binding region of rapamycin.

anhydride (15 equiv) in NMP. The completed peptides were washed successively in NMP (thrice), DCM (thrice), and MeOH (thrice) and then dried in vacuo. Cleavage from the resin with concomitant peptide deprotection was achieved using the TFA/iPr3SiH (TIPS)/H2O cocktail (95:2.5:2.5), and approximately 10 mL g−1 peptide resin, for 2 h. The suspension was filtered, and the filtrate was evaporated to dryness in vacuo. The residual material was triturated with ice-cold ether and centrifuged. The supernatant was discarded (thrice) to furnish a white solid. The crude product was dissolved in Milli-Q water and lyophilized before purification and analysis by semipreparative and analytical RP-HPLC, respectively. RP-HPLC was performed using a Waters 1525 binary pump system with a Waters 2487 dual-wavelength absorbance detector. UV detection was conducted at 220 nm. Solvent A was 0.1% TFA in water and solvent B 0.1% TFA in CH3CN. Analytical RPHPLC was performed using an ACE reversed phase 5 μm C18300 Å column (4.6 mm × 250 mm), at a flow rate of 1.0 mL min−1. Semipreparative RP-HPLC was performed using an ACE reversed phase 5 μm C18-300 Å column (250 mm × 10 mm) with a flow rate of 5.0 mL min−1 using the gradient as described above. Preparative RP-HPLC was performed using an XBridge 5 μm C18-130 Å (19 mm × 50 mm) column with a flow rate of 10 mL min−1 using the gradient as described above. For disulfide formation, cyclization via disulfide formation was conducted by using a solution of 20% DMSO in water. The peptide (∼1 mg/ mL) was stirred overnight, and the extent of the reaction was verified by analytical RP-HPLC. The solution was freeze-dried twice, and the peptide, where necessary, was purified by preparative RP-HPLC.

this interaction modifies glucocorticoid-, androgen-, and progesterone-dependent transcription, its role with respect to estrogen receptor functioning is still obscure. By using nuclear magnetic resonance (NMR) techniques, we have found that a part of the ligand-binding domain of the hormone receptors interacts with the FK1 domain of FKBP52. By focusing on the estrogen receptor α subtype (hERα), we observed that the sequence V364PGF367, which adopts, in the context of the receptor, a well-defined type II β-turn, interacts with the FK1 domain of FKBP52.23,27 Remarkably, this turn motif is structurally similar to the immunomodulatory macrolide rapamycin and the ligand GPI-1046, as shown in Figure 1. In light of these results and on the basis of X-ray diffraction of crystallographic complexes involving FKBP12, another immunophilin with a single FK1 domain, we have deciphered key chemical parameters participating in this interaction.



MATERIALS AND METHODS Peptide Synthesis. Manual Fmoc solid phase peptide synthesis (SPPS) was performed in a fritted syringe, typically on a 0.1 mmol scale, using Rink amide resin (0.4−0.7 mmol g−1). Coupling was conducted by shaking for 1 h the requisite Fmocprotected amino acids (10 equiv), HBTU/HATU (9.5 equiv), and DIPEA (20 equiv) in NMP. For non-natural amino acids, extended couplings (2−4 h) were conducted twice using 2 equiv of amino acid and activation/coupling reagent. Fmoc deprotection was performed by using 20% piperidine in NMP (1 × 1 min, followed by 1 × 10 min). After each coupling and deprotection, the peptides were washed five times with NMP. Capping was achieved using DIPEA (2 equiv) and acetic B

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

and concentrated to 1.8 mg mL−1 using centrifugal devices (cutoff of 7 kDa, Thermo scientific). NMR Measurements. Assignment of backbone resonances of the FKBP52 FK1 domain under the conditions used in our experiments was achieved on the basis of standard pairs of threedimensional (3D) spectra acquired with 300 μM [15N,13C]FK1 domain at 600 MHz. MARS was used for assignment28 and helped by the previously deposited BMRB (Biological Magnetic Resonance Bank) entry BMRB19788.29 Stock solutions of 5 mM peptide were prepared and stored at −20 °C. All tested peptides were dissolved in the same final buffer as the FKBP52 FK1 domain. Experiments were performed with 75 μM FK1 domain. Trimethylsilyl propionate (TMSP) was used as an internal reference. Two-dimensional (2D) 1H−15N HSQC NMR measurements were taken at 293 K on an AVANCE III 900 MHz Bruker spectrometer equipped with a CP-TCI cryoprobe and a sample jet. NMR tubes (3 mm diameter) containing 200 μL of sample were used. TOPSPIN 3.1 (Bruker) was used for data processing and SPARKY30 for peak picking. Dissociation constants (KD) between the FKBP52 FK1 domain and the peptides of interest were estimated by using titration-based 2D 1H−15N HSQC experiments. Spectra were acquired over 48 scans, 2048/256 complex points, and a relaxation delay of 1 s spanning 14 and 30 ppm in the 1H and 15 N dimensions, respectively. Various protein:peptide ratios (1:1, 1:2, 1:5, 1:10, 1:12, and 1:15) were considered. CSPs of resonances for a few significantly affected residues were monitored to estimate KD values. The degree of deviation for each titration point was calculated following the equation

Characteristics of Peptides. Peptide 1: Ac-RVPGFVDNH2 (·TFA), white powder. Purity by HPLC: >95%. Semipreparative HPLC (ACE C18, 20 to 40% B over 20 min): tR = 6.16 min. Analytical HPLC (ACE C18, 5 to 60% B over 20 min): tR = 10.44 min. C38H59N11O10. MALDI-TOF [M + H]+: m/z 830.66 (calcd 830.45). Peptide 2: Ac-KALPGFRN-NH 2 (0.2TFA), white powder. Purity by HPLC: >95%. Semipreparative HPLC (ACE C18, 15 to 30% B over 20 min): tR = 9.22 min. Analytical HPLC (ACE C18, 5 to 60% B over 20 min): tR = 10.55 min. C43H70N14O10. MALDI-TOF [M + H]+: m/z 943.59 (calcd 943.55). Peptide 3: Ac-KAIPGFRN-NH 2 (2TFA), white powder. Purity by HPLC: >95%. Semipreparative HPLC (ACE C18, 15 to 60% B over 20 min): tR = 7.57 min. Analytical HPLC (ACE C18, 5 to 60% B over 20 min): tR = 10.28 min. C43H70N14O10. MALDI-TOF [M + H]+: m/z 943.59 (calcd 943.55). Peptide 4: Ac-KSLPGFRN-NH 2 (0.2TFA), white powder. Purity by HPLC: >95%. Semipreparative HPLC (ACE C18, 15 to 30% B over 20 min): tR = 9.22 min. Analytical HPLC (ACE C18, 5 to 60% B over 20 min): tR = 10.44 min. C43H70N14O11. MALDI-TOF [M + H]+: m/z 959.97 (calcd 959.54). Peptide 5: Ac-RVPGF-NH2 (·TFA), white powder. Purity by HPLC: >95%. Preparative HPLC (XBridge C18, 10 to 40% B over 10 min): tR = 4.28 min. Analytical HPLC (ACE C18, 5 to 60% B over 20 min): tR = 8.68 min. C29H45N9O6. MALDI-TOF [M + H]+: m/z 616.06 (calcd 616.36). Peptide 6: Ac-RtLPGF-NH2 (·TFA), white powder. Purity by HPLC: >95%. Preparative HPLC (ACE C18, 10 to 40% B over 10 min): tR = 6.30 min. Analytical HPLC (ACE C18, 5 to 60% B over 20 min): tR = 11.84 min. C30H47N9O6. MALDITOF [M + H]+: m/z 630.38 (calcd 630.37). Peptide 7: Accyclo(CRtLPGFC)-NH2 (·TFA), white powder. Purity by HPLC: >95%. Preparative HPLC (XBridge C18, 10 to 40% B over 10 min): tR = 5.68 min. Analytical HPLC (ACE C18, 5 to 60% B over 20 min): tR = 12.95 min. C36H55N11O8S2. MALDITOF [M + H]+: m/z 834.37 (calcd 834.38). Peptide 8: AcRLdmPGF-NH2 (·TFA), white powder. Preparative HPLC (XBridge C18, 15 to 40% B over 10 min): tR = 7.06 min. Analytical HPLC (ACE C18, 5 to 60% B over 20 min): tR = 13.75 min (>95%). C32H51N9O6. MALDI-TOF [M + H]+: m/z 658.28 (calcd 658.40). Expression and Purification of the 15N-Labeled Recombinant FKBP52 FK1 Domain. The DNA sequence encoding the Lys28−Asp141 region of the FKBP52 FK1 domain was codon-optimized for translation in the recombinant bacterial system (Genecust, Ellange, Luxembourg). The cDNA was inserted between the NdeI and XhoI restriction sites of vector pET15B, allowing its expression in Escherichia coli under the control of a T7-based promoter fused to an N-terminal His tag. The recombinant FKI pET15B plasmid was transformed into the E. coli BL21 (DE3) strain. The labeled protein was expressed in inoculated M9 medium supplemented with 15Nlabeled ammonium chloride (1 g L−1) and/or 13C-labeled glucose (2 g L−1), otherwise in LB for unlabeled protein. Bacterial growth was started at 10 °C and 50 rpm for 12 h and then switched to 37 °C and 180 rpm until the OD600 reached 0.7−0.8. Expression was then induced by 0.4 mM IPTG, and bacterial fermentation was pursued for an additional 4 h. After bacterial pellet lysis, the supernatant containing His-tagged proteins was purified on a His-Pur Ni2+-NTA affinity column (Pierce, 1 mL resin bead for 1 L of fermentation medium). The fractions containing the protein were transferred to the final buffer [100 mM NaP, 75 mM NaCl, and 1.25 mM EDTA (pH 6.5)] by desalting on G25 resin (PD10 columns, GE healthcare)

ΔCS = 0.2(Δ15N)2 + (Δ1H)2 and plotted versus peptide concentration. The following saturation equation was fitted to the data: ΔCS = Bmax

[Pt] + [Lt] + KD −

([Pt] + [Lt] + KD)2 − 4[Pt][Lt] 2[Pt]

where ΔCS is the degree of chemical shift deviation, Bmax is the maximal chemical shift deviation, [Pt] is the protein concentration, [Lt] is the titrated ligand concentration, and KD is the dissociation constant.31 ROESY experiments were conducted at 600 MHz and 298 K with 500 μM peptides 6−8 and 4096, 6144, and 4096 points in the direct dimension, respectively, and 512 points in the indirect dimension, with a 10 ppm spectral width in both dimensions. The mixing time was set to 62 ms. EXSY experiments were conducted with an exchange delay of 150 ms.



RESULTS The FK1 domain of FKBP52 (residues 28−141) was expressed as a recombinant protein in E. coli and isotopically 15N-labeled for NMR investigations. The 2D 1H−15N spectrum of the cognate FK1 domain shows a large dispersion of the signals present on the proton scale, an observation that reflects a folded domain (Figure S1A). Backbone resonance assignments under the conditions used in the experiments were obtained for all the FKI residues by using a 15N- and 13C-labeled FK1 for NMR 3D experiments. Use of TALOS32 on the backbone CA and CB chemical shifts shows that the secondary structures of the FK1 domain isolated in solution are similar to those found in the reported crystal structure of FK1 of FKBP52 (Figure S2), which is composed of six-stranded antiparallel β-sheets that wrap around a right-handed twisted helix α (Figure S1B).12,14,33,34 C

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 2. Effects of the PPIase catalytic domain of FKBP52 (0.1 mM) on the cis−trans isomerization of the Val-Pro motif of peptide 1 (1 mM, AcRVPGFVD-NH2), in the (A) absence and (B) presence of FK506. Exchange peaks related to cis−trans isomerization are shown by arrows. The black spectrum corresponds to a homonuclear TOCSY spectrum.

Looking for a Minimal FKBP52 FK1-Binding Sequence. As a first step in this study, we synthesized the R363VPGFVD369 sequence of hERα (Ac-RVPGFVD-NH2, peptide 1), which includes the short VPGF motif, and we tested its interaction with the recombinant 15N-labeled FK1 domain of FKBP52. By using 1H−15N 2D HSQC NMR experiments, we observed modest but significant chemical shift perturbations (CSPs) and changes in the intensity of some 1H−15N resonances, indicative of a weak interaction between the peptide and the protein [estimated KD value in the millimolar range (Figure S3)]. The largest CSPs were observed for two amino acids of the catalytic pocket, i.e., Ile87, as shown in Figure S4, and Trp90 (Nε−Hε resonance). Similar results were observed with the corresponding motifs of the human androgen receptor (Ac-KALPGFRNNH2, peptide 2, residues 721−728), glucocorticosteroid receptor (Ac-KAIPGFRN-NH2, peptide 3, residues 553−560), and progesterone receptor (Ac-KSLPGFRN-NH2, peptide 4, residues 734−741) (Figure S4). Interestingly, a cis−trans isomerization at the prolyl bond of the Val-Pro motif of peptide 1, which was observed in the presence of the FK1 domain, was abolished when the high-affinity macrolide FK506 was added (Figure 2). As the exchange delay of the experiment is 150 ms and an exchange peak between the cis and trans conformations of the Val-Pro bond of peptide 1 is observed on that time scale, the exchange rate (Kex) can be estimated to be on the time scale of 100 ms by using EXSY. In an attempt to determine the minimal binding motif, we tested the sequence Ac-RVPGF-NH2 (peptide 5, hERα fragment 363−367) for its ability to interact with the FK1 domain, the additional N-terminus arginine improving the solubility versus that of the highly hydrophobic Ac-VPGF-NH2 tetrapeptide. Upon incubation with the FK1 domain, this pentapeptide gave observable CSPs (Figures 3A and 4A), which

can be mapped to residues on the concave surface of this domain and are reported to also interact with FK506 (Figure 5A). Peptide Cyclization and an Increase in Hydrophobicity at the Residue Preceding the Proline Enhance the Interaction of Peptide 5 with the FKBP52 FK1 Domain. With respect to the PPIase catalytic sites, strong analogies exist between FKBP12 and FKBP52,11,14,33,35 allowing extrapolation from the crystal structure of the FKBP12−GPI-1046 complex to FKBP52 (Figure 6A). Thus, we focused our efforts on the pentapeptide sequence Ac-RVPGF-NH2 (peptide 5) and used the GPI-1046−FKBP12 complex to design chemical modifications for deciphering the driving forces required for recognition (Figure 6B). To this end, we studied the enhancement of the affinity of ligands derived from the Ac-RVPGF-NH2 sequence. First, an increase in hydrophobicity at the amino acid preceding the proline (i.e., valine) increases the surface of the protein affected by the ligand. The 1H−15N spectrum of the FK1 domain in the presence of peptide 6 (Ac-RtLPGF-NH2), where t L corresponds to tert-leucine, showed a modest increase in the number of CSPs compared to the number observed for peptide 5. Broadening of the residue Ile87 resonance was even greater than that observed with peptide 5, and with the complete disappearance of the signal of the Ile87 (Figures 3 and 4). In this regard, it should be stressed that in the course of the titration of a protein with increasing amounts of a ligand, peak broadening can be observed depending on the exchange regime between the bound and free states (and thus the KD). For an intermediate exchange rate on the NMR time scale, the resonance will experience a broadening, as observed for Ile87. The NMR time scale is given by the difference in resonance frequencies of the fully bound and free protein (Δν = νbound − νfree). D

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 3. Superimposed 1H−15N HSQC spectrum of the FKBP52 FK1 domain (75 μM) in the absence (blue) or presence (overlaid in red) of the peptides at 750 μM: (A) Ac-RVPGF-NH2 (peptide 5), (B) Ac-RtLPGF-NH2 (peptide 6), (C) Ac-cyclo(CRtLPGFC)-NH2 (peptide 7), and (D) AcRL-dmP-GF-NH2 (peptide 8). Insets are close-ups showing the resonance of residue Ile87. The second peak corresponds to an unassigned resonance of a lateral HN chain.

and 7 did not display a significant population of the cis conformation (Figure 8A,B). In contrast and on the basis of the fact that 5,5-disubstituted prolines lock in a cis conformation, the 5,5-dimethylproline (dmP or dmPro)-containing peptide AcRL-dmP-GF-NH2 (peptide 8) ROESY spectrum showed a cross-peak between the Hα proton of the proline and the Hα proton of the preceding leucine, exclusively, confirming the presence of a population with a cis conformation at the LeudmPro motif [100% of the population was cis (Figure 8C)].39 1 H−15N HSQC titration experiments revealed once again the total disappearance of the signal corresponding to Ile87 (Figures 3D and 4D) and an occupancy of almost all of the surface of the PPIase site (Figure 5D). An average KD value of 178 ± 58 μM was calculated on the basis of the observed gradual CSP along the titration for six 1H−15N resonances (i.e., Asp68, Asp72, Arg73, Lys74, Lys88, and Asp91), showing large CSPs. The lowest KD value in the series is 100 μM, corresponding to the resonance of Arg73 (Figure 7C). The increase in the observed CSP values, comparable to that of peptide 7, can be correlated with a significant enhancement of the affinity relative to that of linear peptide 5.

The surface of the FK1 domain involved in the interaction was similar to the surface of peptide 5 (Figure 5B vs Figure 5A), and the interaction remained weak, as we were not able to saturate the protein with a 10-fold excess of peptide 6 (Figure 7A). A KD in the low millimolar to high micromolar range was determined. Remarkably, the cyclized version of this compound, i.e., Accyclo(CRtLPGFC)-NH2 (peptide 7), where the cyclization was induced by a disulfide bridge between two cysteines, displayed even more significant CSPs, with disappearance of the IlE87 (Figures 3C and 4C). These perturbations were mapped on the FK1 surface and showed a strong occupancy of the catalytic pocket (Figure 5C). By considering the gradual CSP along a titration of the 1H−15N resonances of residues Tyr57, Ile92, and Glu131, an average KD value of 340 ± 135 μM was estimated (Figure 7B). The lowest estimated KD value, of 150 μM, was obtained by fitting the CSP of Ile92 resonances with the saturation function. The Importance of cis Isomerization Is Significant for the Interaction with the FKBP52 FK1 Domain. The conformation of the ligands within the PPIase (FK1) site of FKBP proteins shows an Xaa−Pro cis conformation, suggesting a type VI β-turn,36−38 which could allow the stabilization of Hbonds with residues Ile87 and Tyr113 of the FK1 domain of FKBP52. Thus, we were interested in exploring the effects of an Xaa−Pro cis locked conformation on binding affinity. Peptides 6



DISCUSSION We have used 2D HSQC NMR spectroscopy to study the association of the FKBP52 immunophilin FK1 domain with the E

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 4. CSPs in the 1H−15N HSQC spectrum reported along the FK1 sequence, in the presence of the peptide sequences at 750 μM: (A) AcRVPGF-NH2 (peptide 5), (B) Ac-RtLPGF-NH2 (peptide 6), (C) Ac-cyclo(CRtLPGFC)-NH2 (peptide 7), and (D) Ac-RL-dmP-GF-NH2 (peptide 8). Red stars mark the broadened residue Ile87. The red threshold is defined as the average CSP for peptide 8.

the 1H−15N correlations of the amide functions of the protein backbone and some side chains of residues of FK1. Accordingly, deviation of chemical shifts (CSDs) or resonance intensity changes, which result from modifications of the environment around well-identified amino acids, would reflect not only interaction with the ligands but also which amino acids interact. 1 H−15N HSQC experiments showed that peptide 1 induces changes in the resonances of residue Ile87 and of the HεNε of the Trp90 side chain, two amino acids of the FK1 domain of FKBP52 that are known to line the bottom of the enzymatic site and interact with macrolide immunomodulators such as FK506 or rapamycin.40−42 Whereas Ile87 forms a hydrogen bond with a carbonyl of the pipecoyl-α-keto amide group of FK506, Trp90 lines the bottom of the enzymatic site.40,41 It is of note that these two residues contribute to the hydrophobic character of the PPIase pocket.34 Thus, an interaction of peptide 1 within the PPIase pocket of the protein is highly likely. The inhibition of the PPIase activity on the Val-Pro dipeptide motif of peptide 1 by FK506 confirmed this interaction. The analysis of the hERα 363−369 sequence shows that the V364PGF367 tetrapeptide motif is structurally similar to the FK1binding motif of FK506, from which the neuroprotective molecule GPI-1046 [3-(3-pyridyl)-1-propyl-(2S)-1-(3,3-dimethyl-1,2-dioxopentyl)-2-pyrrolidinecarboxylate] was designed (Figure 1).43 Accordingly, the proline motif of GPI1046 mimics the pipecolinyl ring of macrolide immunomodulators, known to bind with the PPIase catalytic site.40,41 GPI1046 displays a neurotrophic profile and associates with the FK1 domain of some FKBP proteins, including FKBP12 and FKBP52, which present strong similarities in their enzymatic site.34,43−45 The prolyl motif of GPI-1046 is engulfed within a FKBP12 FK1 hydrophobic pocket that is mainly lined with hydrophobic residues, i.e., Tyr26 {Tyr57}a (strand β3), Phe46

Figure 5. Surface representation of the regions of the FK1 domain of FKBP52 interacting with the peptide sequences (A) Ac-RVPGF-NH2 (peptide 5), (B) Ac-RtLPGF-NH2 (peptide 6), (C) Ac-cyclo(CRtLPGFC)-NH2 (peptide 7), and (D) Ac-RL-dmP-GF-NH2 (peptide 8). Residues with CSP above the threshold (red line in Figure 4) are colored red. The ribbon representation of FK1 is transparent (PDB entry 4LAV12).

synthesized peptides 1 and 5−8, which are derived from the type II β-turn short primary sequence RVPGF of the ligandbinding domain of the hERα (residues 363−367). Briefly, this method allows the observation of resonances corresponding to F

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 6. (A) Alignment of the FKBP52 FK1 domain sequence (top line) with the FKBP12 FK1 domain sequence (middle line) and the resulting consensus sequence (bottom line). Sequence alignments were performed using Multalin.56 The arrows show the key residues involved in the stabilization of the complex formed between FKBPs and their ligands such as FK506 and GPI-1046. The red arrows show the residues forming the hydrophobic pocket in which the prolyl residue of GPI-1046 is engulfed. The black arrows show the residues forming the pocket in which the isopentyl moiety of GPI-1046 is engulfed. (B) Complex of FKBP12 with GPI-1046 (PDB entry 1F40,45 drawing made using VMD57). The protein FKBP12 is drawn as a Connolly surface.

{Phe77} (strand β4b), and Val55 {Val86}, Ile56 {Ile87}, Trp59 {Trp90}, and Phe99 {Phe130} (strand β6).45 Likewise, the residues interacting with the isopentyl moiety of GPI-1046, i.e., Phe36 {Phe67} (strand β4a), Asp37 {Asp68} (strand β4a), and Tyr82 {Tyr113}, His87 {Ser118}, Ile90 {Lys121}, and Ile91 {Ile122}, are hydrophobic. Lastly, the pyridylprolyl of GPI-1046 binds within a site composed of residues Val55 {Val86}, Ile56 {Ile87}, Tyr82 {Tyr113}, and His87 {Ser118}.34,45 Importantly, apart from His87 and Ile90, the amino acids forming the catalytic site are remarkably well conserved between FKBP12 and FKBP52. Thus, we hypothesized that the cyclization of peptide ligands could improve binding by converging toward macrolide immunomodulators (e.g., tacrolimus and rapamycin), while retaining peptidic character. In the same context, the analysis of the binding of GPI-1046 within the FK1 domain of FKBP12 suggests that an increase in hydrophobicity at the residue preceding the proline could potentiate FK1−ligand contacts. Accordingly, the increase in hydrophobicity at the residue preceding the proline by replacement of the valine by a tertleucine (peptide 6) increased the size of the surface of interaction. It is of note that such a chemical modification generates sequences more reminiscent of the peptides corresponding to human progesterone, glucocorticosteroid, and androgen receptor (peptides 2−4, respectively), which share at the same position a leucine or an isoleucine. The differences of hydrophobicity in the residue preceding the proline may explain differences in potentiation between hERα and other steroid hormone receptors.46 Accordingly, the cyclized peptide with a tert-leucine (peptide 7) showed a KD value lower than those of compounds 1, 5, and 6. The interaction of peptide-derived ligands with the FK1 domain of FKBP52 usually occurs through a cis conformation at the proline/residue i − 1 peptide bond, as shown for FKBP12 or FKBP35.47−49 Thus, we were interested in exploring the effects of the control of cis−trans isomerization on peptide interactions. To this end, we used a 5,5-dimethyl-L-proline (dmP or dmPro), which is known to induce a cis conformation, and incorporated it into peptide 8. Significant changes were observed, when compared to the corresponding linear trans peptides 5 and 6.

Figure 7. Close-ups of 1H−15N HSQC spectra of 75 μM FKBP52 FK1 domain with increasing concentrations of the peptide sequences: (A) Ac-RtLPGF-NH2 (peptide 6), (B) Ac-cyclo(CRtLPGFC)-NH2 (peptide 7), and (C) Ac-RL-dmP-GF-NH2 (peptide 8). Peptide ratios are 1:0 (grey), 1:1 (pink), 1:2 (blue), 1:5 (violet), 1:10 (red), and 1:12 (green). Data from panels B and C can be fitted with a saturation curve to derive a KD.

G

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 8. Details of ROESY spectra of the peptide sequences (A) Ac-RtLPGF-NH2 (peptide 6), (B) Ac-cyclo(CRtLPGFC)-NH2 (peptide 7), and (C) Ac-RL-dmP-GF-NH2 (peptide 8). In panel C, the NOE cross-peak between the Hα Pro and Hα Leu residues is typical of a cis conformation for this bond.



The recorded 2D 1H−15N HSQC NMR spectrum showed CSPs of resonances corresponding to residues in the Ser69−Lys74 loop of the FKBP52 FK1 domain (Figure 4D). Interestingly, these residues are components of a flexible bulge important for ligand recognition and enzymatic activity in the center of strand β3 (a part of the 40s loop) (Figure 9).1,2,14,23,29 A drop of

CONCLUSION By using NMR techniques on peptides derived from different steroid hormone receptors, with a special emphasis on hERα for which the role of FKBP52 is still obscure, we have shown that a type II β-turn sequence located in their ligand-binding domain could participate in the recruitment of FKBP52. With respect to hERα, this turn motif (residues 364−367) is a part of a larger interaction platform that includes the region delimited by residues 295−311, which is in charge of the recruitment of calmodulin,51,52 Hsp70,53 and the receptor itself.54 Our study suggests that the FKBP52 PPIase-binding site could target the ligand-binding domain of steroid hormone receptors to participate in the formation of multiprotein complexes.11,13,14,55 Whether our observations, which are restricted to a short peptide sequence of the human glucocorticoid, progesterone, and androgen receptors, could be extrapolated to the whole protein will be the object of future investigations. Such investigations could be helpful in exploring the physiological role of FKBP52 when it is bound to hERα. Indeed, even if this interaction has been shown to exist for hERα,24−26 it fails to modify transcription.46 Remarkably, the hERα V364PGF367 motif is structurally similar to the FKBP ligand GPI-1046. Furthermore, this sequence seems to interact with a bulge (Ser69−Lys74) located in the FK1 domain of FKBP52 and is a substrate of its PPIase domain. The increase in hydrophobicity of the residue preceding the proline, accompanied by the cyclization of the VPGF motif or, independently, substitution with a dmP, induced an increase in the size of the surface of contact between the two partners, and an enhancement in affinity. On the basis of these encouraging results, we are currently developing new FKBP52 peptide ligands with submicromolar KD values.

Figure 9. FK1 domain of FKBP52 (PDB entry 1Q1C14). The FK1 domain is shown as a ribbon (gray). The bulge (Ser69−Lys74) is colored red (ribbon). Residues constituting the PPIase pocket (annotated in red) are shown as green sticks and spheres to outline the surface.

approximately 1 order of magnitude in the KD was recorded. Thus, an Xaa−Pro cis conformation appears to be key for an increased level of peptide−FK1 recognition. The optimal binding of the cis conformation could be explained, at least in part, on the basis of the orientation and distance between the two carbonyl groups of the Xaa−Pro amide bonds that could form hydrogen bonds with the CO group of Ile87 and phenolic OH Tyr113 of FK1 (Figure S5), two residues that are particularly well conserved among immunophilins, as stressed in studies performed by others.12,29,47,48,50 However, we cannot exclude an appropriate orientation of the side chains or a contribution of the two methyls at position 5 of the proline. Peptide 7, although in a mainly trans conformation in terms of the Xaa−Pro bond, seems to display an affinity for the FK1 domain better than that of its linear counterpart (peptide 6). In fact, it may present an optimal orientation of the pharmacophores, in a manner similar to that of the Xaa−Pro cis conformation of peptide 8, and independent of the Xaa−Pro cis−trans equilibrium.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.6b00506. 1 H−15N 2D HSQC spectra of the [15N]FK1 domain of FKBP52 (Figure S1), TALOS secondary structure prediction of FKBP52 FK1 residues (Figure S2), data fit of the CSPs (ΔCS, in parts per million) of residue Ile87 of FKBP52 as a function of increasing peptide 1 concentration (Figure S3), structures of the estrogen, androgen, glucocorticoid, and progesterone receptors (Figure S4), enlarged regions of the 1H−15N HSQC spectrum of FK1 with the free FK1 domain (blue) and in the presence of a 10-fold excess of peptides 1−4 (red), and differences in the distance between the two carbonyl H

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry



(3) Hong, J., Kim, S. T., Tranguch, S., Smith, D. F., and Dey, S. K. (2007) Deficiency of co-chaperone immunophilin FKBP52 compromises sperm fertilizing capacity. Reproduction 133, 395−403. (4) Tranguch, S., Wang, H., Daikoku, T., Xie, H., Smith, D. F., and Dey, S. K. (2007) FKBP52 deficiency-conferred uterine progesterone resistance is genetic background and pregnancy stage specific. J. Clin. Invest. 117, 1824−1834. (5) Chambraud, B., Belabes, H., Fontaine-Lenoir, V., Fellous, A., and Baulieu, E. E. (2007) The immunophilin FKBP52 specifically binds to tubulin and prevents microtubule formation. FASEB J. 21, 2787−97. (6) Galigniana, M. D., Radanyi, C., Renoir, J. M., Housley, P. R., and Pratt, W. B. (2001) Evidence that the peptidylprolyl isomerase domain of the hsp90-binding immunophilin FKBP52 is involved in both dynein interaction and glucocorticoid receptor movement to the nucleus. J. Biol. Chem. 276, 14884−14889. (7) Gold, B. G. (1999) FK506 and the role of the immunophilin FKBP-52 in nerve regeneration. Drug Metab. Rev. 31, 649−663. (8) Chambraud, B., Sardin, E., Giustiniani, J., Dounane, O., Schumacher, M., Goedert, M., and Baulieu, E. E. (2010) A role for FKBP52 in Tau protein function. Proc. Natl. Acad. Sci. U. S. A. 107, 2658−2663. (9) Giustiniani, J., Guillemeau, K., Dounane, O., Sardin, E., Huvent, I., Schmitt, A., Hamdane, M., Buée, L., Landrieu, I., Lippens, G., Baulieu, E. E., and Chambraud, B. (2015) The FK506-binding protein FKBP52 in vitro induces aggregation of truncated Tau forms with prion-like behavior. FASEB J. 29, 3171−3181. (10) Blair, L. J., Baker, J. D., Sabbagh, J. J., and Dickey, C. A. (2015) The emerging role of peptidyl-prolyl isomerase chaperones in tau oligomerization, amyloid processing, and Alzheimer’s disease. J. Neurochem. 133, 1−13. (11) Peattie, D. A., Harding, M. W., Fleming, M. A., DeCenzo, M. T., Lippke, J. A., Livingston, D. J., and Benasutti, M. (1992) Expression and characterization of human FKBP52, an immunophilin that associates with the 90-kDa heat chock protein and is a component of steroid receptor complexes. Proc. Natl. Acad. Sci. U. S. A. 89, 10974−10978. (12) Bracher, A., Kozany, C., Haehle, A., Wild, P., Zacharias, M., and Hausch, F. (2013) Crystal structures of the free and ligand-bound FK1FK2 domain segment of FKBP52 reveal a flexible inter-domain hinge. J. Mol. Biol. 425, 4134−4144. (13) Sinars, C. R., Cheung-Flynn, J., Rimerman, R. A., Scammell, J. G., Smith, D. F., and Clardy, J. (2003) Structure of the large FK506binding protein FKBP51, an Hsp90-binding protein and a component of steroid receptor complexes. Proc. Natl. Acad. Sci. U. S. A. 100, 868− 873. (14) Wu, B., Li, P., Liu, Y., Lou, Z., Ding, Y., Shu, C., Ye, S., Bartlam, M., Shen, B., and Rao, Z. (2004) 3D structure of human FK506-binding protein 52: implications for the assembly of the glucocorticoid receptor/Hsp90/immunophilin heterocomplex. Proc. Natl. Acad. Sci. U. S. A. 101, 8348−8353. (15) Le Bihan, S., Renoir, J. M., Radanyi, C., Chambraud, B., Joulin, V., Catelli, M. G., and Baulieu, E. E. (1993) The mammalian heat shock protein binding immunophilin (p59/HBI) is an ATP and GTP binding protein. Biochem. Biophys. Res. Commun. 195, 600−607. (16) Callebaut, I., Renoir, J. M., Lebeau, M. C., Massol, N., Burny, A., Baulieu, E. E., and Mornon, J. P. (1992) An immunophilin that binds M(r) 90,000 heat shock protein: main structural features of a mammalian p59 protein. Proc. Natl. Acad. Sci. U. S. A. 89, 6270−6274. (17) Cheung-Flynn, J., Roberts, P. J., Riggs, D. L., and Smith, D. F. (2003) C-terminal sequences outside the tetratricopeptide repeat domain of FKBP51 and FKBP52 cause differential binding to Hsp90. J. Biol. Chem. 278, 17388−17394. (18) Massol, N., Lebeau, M. C., Renoir, J. M., Faber, L. E., and Baulieu, E. E. (1992) Rabbit FKBP59-heat shock protein Binding Immunophilin (HBI) is a Calmodulin Binding Protein. Biochem. Biophys. Res. Commun. 187, 1330−1335. (19) Eyles, S. J. (2001) Proline not the only culprit? Nat. Struct. Biol. 8, 380−381. (20) Albers, M. W., Walsh, C. T., and Schreiber, S. L. (1990) Substrate specificity for the human rotamase FKBP: a view of FK506 and

groups that form the H-bond with specific FKBP52 residues of different ligands (Figure S5) (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Telephone: +33-(0)1-44-2744-44. Author Contributions

C.B. and M.A.H. contributed equally to this work. Funding

M.A.H. was financed by 14-3-3 Stabs EU Project 286418 (FP7, Marie Curie Actions). The Région Nord, the Institut Pasteur of Lille, European Union (FEDER), the French Research Ministry, and the Université des Sciences et Technologies de Lille 1 funded the NMR facilities. We are grateful to the University Pierre & Marie Curie (Paris 6), ED406, the Centre National de la Recherche Scientifique (CNRS), the Ecole Normale Supérieure (ENS), and the Institut Baulieu for their financial support. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We acknowledge the support from the TGE RMN THC (FR3050, France), the French Laboratory of Excellency Distalz, and the ANR TAF.



ABBREVIATIONS ΔG , activation energy; BMRB, Biological Magnetic Resonance Bank; CSP, chemical shift perturbation; DCM, dichloromethane; DIPEA, diisopropylethylamine; dmPro or dmP, dimethylproline; DMSO, dimethyl sulfoxide; EDTA, ethylenediaminetetraacetic acid; hERα, human estrogen receptor α; EXSY, exchange spectroscopy; FKBP52, FK-506-binding protein of 52 kDa; Fmoc, fluorenylmethoxycarbonyl; HATU, N-[(dimethylamino)-1H-1,2,3-triazolo[4,5-b]pyridin-1-ylmethylene]-N-methylmethanaminium; HBTU, N,N,N′,N′-tetramethyl-O-(1H-benzotriazol-1-yl)uronium hexafluorophosphate; HSQC, heteronuclear single-quantum coherence; IPTG, isopropyl β-D-1-thiogalactopyranoside; MALDI-TOF, matrix-assisted laser desorption ionization time of flight; MeOH, methanol; NMP, 1-methyl-2-pyrrolidinone; PDB, Protein Data Bank; PPIase, peptidyl-prolyl isomerase; ROESY, rotating-frame nuclear Overhauser effect; RP-HPLC, reverse phase highperformance liquid chromatography; SPPS, solid phase peptide synthesis; tLeu (tL), tert-leucine; TFA, trifluoroacetic acid; TIPS, triisopropylsilane (iPr3SiH); TMSP, trimethylsilylpropionic acid; TOCSY, total correlation spectroscopy; TPR, tetratricopeptide repeat.

■ ■

#

ADDITIONAL NOTE The numbers in braces correspond to the residue numbers in FKBP52. a

REFERENCES

(1) Riggs, D. L., Cox, M. B., Tardif, H. L., Hessling, M., Buchner, J., and Smith, D. F. (2007) Noncatalytic role of the FKBP52 peptidylprolyl isomerase domain in the regulation of steroid hormone signaling. Mol. Cell. Biol. 27, 8658−8669. (2) Sivils, J. C., Storer, C. L., Galigniana, M. D., and Cox, M. B. (2011) Regulation of steroid hormone receptor function by the 52-kDa FK506-binding protein (FKBP52). Curr. Opin. Pharmacol. 11, 314− 319. I

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

VI β-turn mimic via ring-closing metathesis. J. Org. Chem. 68, 5006− 5008. (38) Che, Y., and Marshall, G. R. (2006) Impact of cis-proline analogs on peptide conformation. Biopolymers 81, 392−406. (39) An, S. S. A., Lester, C. C., Peng, J. L., Li, Y. J., Rothwarf, D. M., Welker, E., Thannhauser, T. W., Zhang, L. S., Tam, J. P., and Scheraga, H. A. (1999) Retention of the cis proline conformation in tripeptide fragments of bovine pancreatic ribonuclease A containing a non-natural proline analogue 5,5-dimethylproline. J. Am. Chem. Soc. 121, 11558− 11566. (40) Michnick, S. W., Rosen, M. K., Wandless, T. J., Karplus, M., and Schreiber, S. L. (1991) Solution structure of FKBP, a rotamase enzyme and receptor for FK506 and rapamycin. Science 252, 836−839. (41) Van Duyne, G., Standaert, R. F., Karplus, P. A., Schreiber, S. L., and Clardy, J. (1991) Atomic structure of FKBP-FK506, an immunophilin-immunosuppressant complex. Science 252, 839−842. (42) Siekierka, J. J., Staruch, M. J., Hung, S. H., and Sigal, N. H. (1989) FK-506, a potent novel immunosuppressive agent, binds to a cytosolic protein which is distinct from the cyclosporin A-binding protein, cyclophillin. J. Immunol. 143, 1580−1583. (43) Steiner, J. P., Hamilton, G. S., Ross, D. T., Valentine, H. L., Guo, H., Connolly, M. A., Liang, S., Ramsey, C., Li, J. H., Huang, W., Howorth, P., Soni, R., Fuller, M., Sauer, H., Nowotnik, A. C., and Suzdak, P. D. (1997) Neurotrophic immunophilin ligands stimulate structural and functional recovery in neurodegenerative animal models. Proc. Natl. Acad. Sci. U. S. A. 94, 2019−2024. (44) Blackburn, E. A., and Walkinshaw, M. D. (2011) Targeting FKBP isoforms with small-molecule ligands. Curr. Opin. Pharmacol. 11, 365− 371. (45) Sich, C., Improta, S., Cowley, D. J., Guenet, C., Merly, J. P., Teufel, M., and Saudek, V. (2000) Solution structure of a neurotrophic ligand bound to FKBP12 and its effects on protein dynamics. Eur. J. Biochem. 267, 5342−5354. (46) Riggs, D. L., Roberts, P. J., Chirillo, S. C., Cheung-Flynn, J., Prapapanich, V., Ratajczak, T., Gaber, R., Picard, D., and Smith, D. F. (2003) The Hsp90-binding peptidylprolyl isomerase FKBP52 potentiates glucocorticoid signaling in vivo. EMBO J. 22, 1158−1167. (47) Bizzarri, M., Tenori, E., Martina, M. R., Marsili, S., Caminati, G., Menichetti, S., and Procacci, P. (2011) New perspective on how and why immunophilin FK506-related ligands. J. Phys. Chem. Lett. 2, 2834− 2839. (48) Gopalakrishnan, R., Kozany, C., Wang, Y., Schneider, S., Hoogeland, B., Bracher, A., and Hausch, F. (2012) Exploration of pipecolate sulfonamides as binders of the FK506-binding proteins 51 and 52. J. Med. Chem. 55, 4123−4131. (49) Alag, R., Balakrishna, A. M., Rajan, S., Qureshi, I. A., Shin, J., Lescar, J., Grüber, G., and Yoon, H. S. (2013) Structural insights into substrate binding by PvFKBP35, a peptidylprolyl cis-trans isomerase from the human malarial Plasmodium vivax. Eukaryotic Cell 12, 627− 634. (50) Martina, M. R., Tenori, E., Bizzarri, M., Menichetti, S., Caminati, G., and Procacci, P. (2013) The precise chemical-physical nature of the pharmacore in FK506 binding protein inhibition: ElteX, a new class of nanomolar FKBP12 ligands. J. Med. Chem. 56, 1041−1051. (51) Gallo, D., Jacquemotte, F., Cleeren, A., Laïos, I., Hadiy, S., Rowlands, M. G., Caille, D., Nonclercq, D., Laurent, G., Jacquot, Y., and Leclercq, G. (2007) Calmodulin-independent, agonistic properties of a peptide containing the calmodulin binding site of estrogen receptor alpha. Mol. Cell. Endocrinol. 268, 37−49. (52) Carlier, L., Byrne, C., Miclet, E., Bourgoin-Voillard, S., Nicaise, D., Tabet, J. C., Desmadril, M., Leclercq, G., Lequin, O., and Jacquot, Y. (2012) Biophysical studies of the interaction between calmodulin and the R287-T311 region of human estrogen receptor a reveals an atypical binding process. Biochem. Biophys. Res. Commun. 419, 356−361. (53) Gallo, D., Haddad, I., Duvillier, H., Jacquemotte, F., Laïos, I., Laurent, G., Jacquot, Y., Vinh, J., and Leclercq, G. (2008) Trophic effect in MCF-7 cells of ERα17p, a peptide corresponding to a platform regulatory motif of the estrogen α  Underlying mechanisms. J. Steroid Biochem. Mol. Biol. 109, 138−149.

rapamycin as leucine-(twisted amide)-proline mimics. J. Org. Chem. 55, 4984−4986. (21) Kamah, A., Cantrelle, F. X., Huvent, I., Giustiniani, J., Guillemeau, K., Byrne, C., Jacquot, Y., Landrieu, I., Baulieu, E. E., Smet, C., Chambraud, B., and Lippens, G. (2016) Isomerization and oligomerization of truncated and mutated Tau forms by FKBP52 are independent processes. J. Mol. Biol. 428, 1080−1090. (22) Lu, K. P., Finn, G., Lee, T. H., and Nicholson, L. K. (2007) Prolyl cis-trans isomerization as a molecular timer. Nat. Chem. Biol. 3, 619− 629. (23) De Leon, J. T., Iwai, A., Feau, C., Garcia, Y., Balsiger, H. A., Storer, C. L., Suro, R. M., Garza, K. M., Lee, S., Sang Kim, Y., Chen, Y., Ning, Y. M., Riggs, D. L., Fletterick, R. J., Guy, R. K., Trepel, J. B., Neckers, L. M., and Cox, M. B. (2011) Targeting the regulation of androgen receptor signaling by the heat shock protein 90 cochaperone FKBP52 in prostate cancer cells. Proc. Natl. Acad. Sci. U. S. A. 108, 11878−11883. (24) Tai, P. K., Maeda, Y., Nakao, K., Wakim, N. G., Duhring, J. L., and Faber, L. E. (1986) A 59-kilodalton protein associated with progestin, estrogen, androgen, and glucocorticoid receptors. Biochemistry 25, 5269−5275. (25) Gougelet, A., Bouclier, C., Marsaud, V., Maillard, S., Mueller, S. O., Korach, K. S., and Renoir, J. M. (2005) Estrogen receptor alpha and beta subtype expression and transactivation capacity are differentially affected by receptor-, hsp90- and immunophilin-ligands in human breast cancer cells. J. Steroid Biochem. Mol. Biol. 94, 71−81. (26) Sanchez, E. R. (2012) Chaperoning steroidal physiology: Lessons from mouse genetic models of Hsp90 and its cochaperones. Biochim. Biophys. Acta, Mol. Cell Res. 1823, 722−729. (27) Jacquot, Y., Gallo, D., and Leclercq, G. (2007) Estrogen receptor alpha  Identification by a modeling approach of a potential polyproline II recognizing domain within the AF-2 region of the receptor that would play a role of prime importance in its mechanism of action. J. Steroid Biochem. Mol. Biol. 104, 1−10. (28) Jung, Y. S., and Zweckstetter, M. (2004) Mars - robust automatic backbone assignment of proteins. J. Biomol. NMR 30, 11−23. (29) Mustafi, S. M., LeMaster, D. M., and Hernández, G. (2014) Differential conformational dynamics in the closely homologous FK506-binding domains of FKBP51 and FKBP52. Biochem. J. 461, 115−123. (30) Goddard, T. D., and Kneller, D. G. (2002) Sparky 3, University of California, San Francisco. (31) Montaville, P., Coudevylle, N., Radhakrishnan, A., Leonov, A., Zweckstetter, M., and Becker, S. (2008) The PIP2 binding mode of the C2 domains of rabphilin-3A. Protein Sci. 17, 1025−1034. (32) Cornilescu, G., Delaglio, F., and Bax, A. (1999) Protein backbone angle restraints for searching a database for chemical shift and sequence homology. J. Biomol. NMR 13, 289−302. (33) Craescu, C. T., Rouvière, N., Popescu, A., Cerpolini, E., Lebeau, M. C., Baulieu, E. E., and Mispelter, J. (1996) Three-dimensional structure of the immunophilin-like domain of FKBP59 in solution. Biochemistry 35, 11045−11052. (34) Li, P., Ding, Y., Wu, B., Shu, C., Shen, B., and Rao, Z. (2003) Structure of the N-terminal domain of human FKBP52. Acta Crystallogr., Sect. D: Biol. Crystallogr. 59, 16−22. (35) Ruan, B., Pong, K., Jow, F., Bowlby, M., Crozier, R. A., Liu, D., Liang, S., Chen, Y., Mercado, M. L., Feng, X., Bennett, F., von Schack, D., McDonald, L., Zaleska, M. M., Wood, A., Reinhart, P. H., Magolda, R. L., Skotnicki, J., Pangalos, M. N., Koehn, F. E., Carter, G. T., AbouGharbia, M., and Graziani, E. I. (2008) Binding of rapamycin analogs to calcium channels and FKBP52 contributes to their neuroprotective activities. Proc. Natl. Acad. Sci. U. S. A. 105, 33−38. (36) Gramberg, D., Weber, C., Beeli, R., Inglis, J., Bruns, C., and Robinson, J. A. (1995) Synthesis of a type-VI beta-turn mimetic and its incorporation into a high affinity somatostatin receptor ligand. Helv. Chim. Acta 78, 1588−1606. (37) Boruah, A., Rao, I. N., Nandy, J. P., Kumar, S. K., Kunwar, A. C., and Iqbal, J. (2003) Synthesis of a novel cis-proline-derived cyclic type J

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry (54) Gallo, D., Jacquot, Y., Cleeren, A., Jacquemotte, F., Laïos, F., Laurent, G., and Leclercq, G. (2007) Molecular basis of agonistic activity ERα17p, a synthetic peptide corresponding to a sequence located at the N-terminal part of the estrogen receptor ligand binding domain. Lett. Drug Design Discovery 4, 346−355. (55) Storer, C. L., Dickey, C. A., Galigniana, M. D., Rein, T., and Cox, M. B. (2011) FKBP51 and FKBP52 in signaling and disease. Trends Endocrinol. Metab. 22, 481−490. (56) Corpet, F. (1988) Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res. 16, 10881−10890. (57) Humphrey, W., Dalke, A., and Schulten, K. (1996) VMD: visual molecular dynamics. J. Mol. Graphics 14, 33−38.

K

DOI: 10.1021/acs.biochem.6b00506 Biochemistry XXXX, XXX, XXX−XXX