Regulation of the Structurally Dynamic N-terminal Domain of Progesterone Receptor by Protein-induced Folding*

Background: The mechanism of action of the N-terminal domain (NTD) of the progesterone receptor is not well understood. Results: We show the PR NTD adopts a functional folded conformation by undergoing disorder-order transition via binding to a target protein, TBP. Conclusion: This structural reorganization of the NTD facilitates binding of co-activators required for transcriptional activation. Significance: A novel mechanism of PR-dependent transcriptional activation is defined. The N-terminal domain (NTD) of steroid receptors harbors a transcriptional activation function (AF1) that is composed of an intrinsically disordered polypeptide. We examined the interaction of the TATA-binding protein (TBP) with the NTD of the progesterone receptor (PR) and its ability to regulate AF1 activity through coupled folding and binding. As assessed by solution phase biophysical methods, the isolated NTD of PR contains a large content of random coil, and it is capable of adopting secondary α-helical structure and more stable tertiary folding either in the presence of the natural osmolyte trimethylamine-N-oxide or through a direct interaction with TBP. Hydrogen-deuterium exchange coupled with mass spectrometry confirmed the highly dynamic intrinsically disordered property of the NTD within the context of full-length PR. Deletion mapping and point mutagenesis defined a region of the NTD (amino acids 350–428) required for structural folding in response to TBP interaction. Overexpression of TBP in cells enhanced transcriptional activity mediated by the PR NTD, and deletion mutations showed that a region (amino acids 327–428), similar to that required for TBP-induced folding, was required for functional response. TBP also increased steroid receptor co-activator 1 (SRC-1) interaction with the PR NTD and cooperated with SRC-1 to stimulate NTD-dependent transcriptional activity. These data suggest that TBP can mediate structural reorganization of the NTD to facilitate the binding of co-activators required for maximal transcriptional activation.

The N-terminal domain (NTD) of steroid receptors harbors a transcriptional activation function (AF1) that is composed of an intrinsically disordered polypeptide. We examined the interaction of the TATA-binding protein (TBP) with the NTD of the progesterone receptor (PR) and its ability to regulate AF1 activity through coupled folding and binding. As assessed by solution phase biophysical methods, the isolated NTD of PR contains a large content of random coil, and it is capable of adopting secondary ␣-helical structure and more stable tertiary folding either in the presence of the natural osmolyte trimethylamine-N-oxide or through a direct interaction with TBP. Hydrogendeuterium exchange coupled with mass spectrometry confirmed the highly dynamic intrinsically disordered property of the NTD within the context of full-length PR. Deletion mapping and point mutagenesis defined a region of the NTD (amino acids 350 -428) required for structural folding in response to TBP interaction. Overexpression of TBP in cells enhanced transcriptional activity mediated by the PR NTD, and deletion mutations showed that a region (amino acids 327-428), similar to that required for TBP-induced folding, was required for functional response. TBP also increased steroid receptor co-activator 1 (SRC-1) interaction with the PR NTD and cooperated with SRC-1 to stimulate NTD-dependent transcriptional activity. These data suggest that TBP can mediate structural reorganization of the NTD to facilitate the binding of co-activators required for maximal transcriptional activation.
Steroid hormone receptors (SRs) 2 contain two transcriptional activation domains that provide interaction surfaces for transcriptional co-regulatory proteins as follows: AF1 located in the N-terminal domain (NTD) and AF2 in the C-terminal ligand binding domain (LBD). AF2 consists of a hydrophobic pocket in the LBD that, in response to a conformational change induced by the binding hormone, recognizes LXXLL motifs in co-activator proteins, including the p160 family of steroid receptor co-activators (SRC-1, -2, and -3). The well structured globular fold of the LBD has enabled high resolution x-ray crystallography structures complexed with various ligands and coregulatory peptides (1)(2)(3)(4)(5)(6)(7)(8). Because crystal structures of SRs are only available for the LBD or DBD, the current design of steroid receptor modulators is primarily based on their ability to modulate interactions of co-regulatory protein motifs with AF2. However, this and related strategies often fail to inactivate AF1, leading to potentially unwanted residual activities. Despite the fact that the NTD accounts for a major proportion of the total transcriptional activity of SRs, little is known about its structure and how co-regulatory proteins that bind to this region lead to receptor activation. This lack of information is due to the fact that the NTD consists of intrinsically disordered polypeptide (IDP) (9 -14).
IDPs lack stable secondary and/or tertiary structure and exist as dynamic conformational ensembles as opposed to a stable structure found in globular proteins (15)(16)(17)(18)(19). However, IDPs have a propensity to fold and undergo a disorder-to-order transition upon binding target proteins by a "coupled folding and binding process" that can result in stabilization of an active structural conformation (20). Intrinsically disordered regions (IDR) are functionally important, and they are most frequently found in proteins of regulatory signaling pathways, including kinases, membrane receptors, and transcription factors (21). * This work was supported, in whole or in part, by National Institutes of Health The NTDs of SRs are predicted by computational analysis to contain a high percentage of random coil, and solution phase biophysical analyses have directly confirmed the IDP characteristics of isolated NTDs of several SRs (22)(23)(24)(25)(26)(27)(28). NTDs of selected SRs have also been observed to undergo an increase in overall secondary and tertiary structure in the presence of natural osmolytes such as trimethylamine-N-oxide (TMAO) and trehalose (10,12,14,29). Osmolytes have the ability to promote local folding of proteins under conditions of cellular stress to maintain protein function and thus are a useful tool to probe the capacity of proteins to undergo a functionally relevant disorder-to-order transition (10,12,29). Co-regulatory proteins have been reported that modulate NTD-mediated transcriptional activity of different SRs; however, the structural and molecular basis for regulation has not been well defined. Only a few NTD-interacting proteins, including the general transcription factors TATA box-binding protein (TBP) and GTF2F1, have been demonstrated to have a connection between induced structural changes and function (30 -36). These protein interactions were initially thought to be a mechanism for receptor contact with basal transcriptional machinery required for activation. However, a core C-terminal domain of human TBP consisting of aa 159 -339 (TBP C ) was reported to interact with the NTDs of several SRs and to promote a more compact tertiary structure with increased ␣-helical content (14,(33)(34)(35)(36). In the case of glucocorticoid (GR) and mineralocorticoid receptors (MR), TBP C binding was also observed to be associated with an enhancement of NTD-dependent transcriptional activity (30,31,33). Coupled binding and folding events leading to disorderto-order transitions in the NTDs of the steroid receptors have been studied using isolated NTD peptides or a minimal AF1 subregion that may not represent the entire SR signaling spectrum. Recent studies indicate that SRs should be viewed as a dynamic ensemble of structures, which respond to a variety of target molecules through allosteric coupling to produce differential selection and/or activation of gene expression (37). Therefore, understanding how receptor and co-regulators combine to influence the structural dynamics of the SRs in general, and NTD/AF1 in particular, should provide greater opportunities to selectively modify SR transcriptional signaling.
To further explore structure/function properties of the steroid receptor NTDs, this study uses a combination of solution biophysical and biochemical methods and mutagenesis to examine TBP C interaction with the NTD of human progesterone receptor (official gene name PGR and official protein name HUMPGRR), and its effects on the structural dynamics and transcriptional activities. Data presented show that the PR NTD undergoes folding and adopts a more stable structure upon binding TBP C and that these structural changes lead to enhanced NTD activity. A subregion of the NTD (aa 350 -428) required for TBP enhanced transcriptional activity, and residues within this region required for coupled binding and folding were identified. TBP-induced folding also enhanced SRC-1 binding with the NTD and TBP, and SRC-1 acted cooperatively to stimulate NTD-mediated transcriptional activity. These data illustrate the generality that NTDs of SRs have regions of unstructured protein that can acquire stable active structural conformations upon binding other proteins.

EXPERIMENTAL PROCEDURES
Plasmids and Antibodies-Recombinant baculovirus vectors expressing different domains of human progesterone receptor (PR) containing polyhistidine tags at the N terminus have been previously described (28). Transfer plasmids used for construction of baculoviruses were pBlueBacHis (Invitrogen) and encode the following domains and sequence regions of PR (PR-A, aa 165-933; PRB-NTD, aa 1-534; PR-A NTD, aa 165-534; hinge LBD, aa 634 -933). The C-terminal core DNA binding domain (aa 159 -339) of the human TATA-binding protein (TBP C ) cloned into a pET-21d bacterial expression vector with an N-terminal polyhistidine tag and thrombin cleavage site has been previously described (38). A GST-TBP C fusion vector for expression in bacterial cells was constructed by cloning aa 159 -339 of human TBP into pGEX-2T containing an N-terminal GST followed by an enterokinase and thrombin cleavage sites between the GST and TBPc. A series of PR NTD regions (aa 165-300, 350 -428, and 456 -555) for expression in bacterial cells were cloned into the pTYB12 vector containing an N-terminal intein tag (impact vector). Amino acid substitutions were introduced into the PR NTD fragments using the Stratagene QuikChange Lightning site-directed mutagenesis kit. Mammalian cell expression plasmids under the control of the Rous sarcoma virus promoter for full-length PR and domains of receptor have been described previously, including phPR-B and a two-domain PR-B NTD/DBD fragment (aa 1-650) (28). DNA sequencing of all plasmids was performed to verify correct sequences and point mutations.
Mouse monoclonal antibodies (mAbs) to human PR (AB52 and N559) that detect an epitope in the NTD common to PR-A and PR-B have been previously described (39,40). Antibody to SRC-1 (sc-6098) was obtained from Santa Cruz Biotechnology, Inc., and is qualified for immunoprecipitation and immunoblot assays.
Recombinant Protein Expression and Purification-PR was expressed from baculovirus vectors in Sf9 insect cells as described previously (28). Full-length PR (A or B isoform) was bound to the synthetic progestin R5020 during expression in Sf9 cells, and cells were lysed in 20 mM sodium phosphate buffer, pH 7.4, containing 350 mM NaCl, 10 mM imidazole, 10% glycerol, 15 mM ␤ mercaptoethanol, 50 mM sodium fluoride, 1 M urea, protease inhibitors (leupeptin, aprotinin, bacitracin, pepstatin A, and PMSF) and 1.2 units/ml benzonase nuclease. Whole cell extracts were submitted to differential centrifugation at 10,000 ϫ g for 30 min at 4°C; the pellet was discarded and the supernatant centrifuged at 100,000 ϫ g for 30 min at 4°C. The high speed soluble supernatant was diluted in cell lysis buffer to a concentration of 12 mg/ml and incubated in batch for 45 min at 4°C with Ni-NTA-agarose resin (Qiagen) using ϳ3 ml of resin slurry/75 ml of whole cell extract. Resins were washed with repeated cycles in cell lysis buffer containing 600 mM NaCl or 200 mM NaCl and step-eluted with increasing amounts of imidazole at 50, 200, and 300 mM. The eluted receptor at 300 mM imidazole was concentrated to ϳ2 ml with a 4-ml Amicon ultracentrifugal device with a 10,000 molecular weight cutoff and submitted to a second step purification by gel filtra-tion on an S200 FPLC column in 20 mM sodium phosphate, pH 7.4, 200 mM NaCl, 10% glycerol, 1 mM DTT, and 1 M urea.
As assessed by Coomassie Blue-stained SDS-PAGE and immunoblotting with PR-specific mAb, the purified product was Ͼ95% purity at a concentration range of 10 -30 M ( Fig.  2A). For full-length PR, the 1 M urea was important to maintain solubility at these concentrations, and it increased thermal stability. The 1 M urea did not alter hormone or DNA binding activities of the receptor indicating the receptor is in native functional conformation at this low concentration of urea (data not shown). The isolated PR NTD (either PR-A or PR-B) was expressed in Sf9 cells and purified in a similar manner as fulllength PR by Ni-NTA resins except that R5020 was not added during SF9 cell expression; 1 M urea was left out of the cell lysis and wash/elution buffers, and the second purification step was binding and elution by NaCl gradient from a Hi-Trap Q HP (GE Healthcare) column (Fig. 1A).
Fragments of the PR NTD cloned into the IMPACT vector (intein fusion) were expressed in transformed competent ER2566 cells (New England Biolabs) under induction by 0.5 mM isopropyl 1-thio-␤-D-galactopyranoside. Cells were lysed in 20 mM Tris-HCl, pH 8.0, 500 mM NaCl, 10% glycerol, 2 mg of DNase I, and protease inhibitor tablets (Roche Applied Science), and clarified supernatant was incubated with chitin beads (New England Biolabs) for 2 h at 4°C in batch mode, and beads were washed cyclically with lysis buffer containing 0.01% Triton X-100 and 1 M NaCl. Bound protein eluted by digestion with 50 mM DTT (buffer: 20 mM Tris-HCl, 100 mM NaCl, 10% glycerol) was fractionated by anion exchange Hi-Trap HQ (GE Healthcare) and NaCl gradient (50 -1000 mM NaCl) eluted purified protein was dialyzed to reduce NaCl to 100 mM. Identity of purified PR-B and NTD products was confirmed by immunoblots with mAbs to different epitopes of PR (39 -41).
As described previously (38), TBP C was expressed by isopropyl 1-thio-␤-D-galactopyranoside (0.5 mM) induction in BL21DE3-transformed bacterial host cells and were frozen as a pellet at Ϫ80°C. Cells were lysed by two freeze-thaw cycles and then sonicated in buffer consisting of 20 mM sodium phosphate, pH 7.4, 500 mM NaCl, 10% glycerol, 1 mM MgSO 4 , 10 mM MgCl 2 , 0.03 mg/ml DNase I, 0.2 mg/ml lysozyme, and EDTAfree protease inhibitor mixture (Roche Applied Science). The lysate was centrifuged at 20,000 rpm for 30 min at 4°C, and the soluble supernatant was passed through a 0.22-m filter. The clarified lysate was bound in batch to Ni-NTA (Qiagen) affinity resin for 75 min at 4°C, and the resin was washed cyclically with lysis buffer containing high (1 M) and low NaCl (200 mM). Washed resins were suspended in thrombin cleavage buffer (20 mM sodium phosphate, pH 7.4, 200 mM NaCl, 20 mM imidazole, 10% glycerol, 15 mM ␤-mercaptoethanol, 2.5 mM CaCl 2 , and 64 units of thrombin/ml suspension) overnight at 4°C, and 1 mM PMSF was added to stop the cleavage. Eluted protein was dialyzed (20 mM sodium phosphate, pH 7.4, 200 mM NaCl, 10% glycerol, 1 mM DTT) and concentrated by Amicon ultracentrifugal device at a 3000 molecular weight cutoff (Millipore) and submitted to gel filtration by S200 FPLC (GE Healthcare). Peak S200 fractions were pooled, concentrated ϳ6-fold, diluted 1:1 in equilibration buffer minus NaCl to reduce the concentration to 100 mM NaCl, and bound to Hi-Trap SP HP cation exchange columns. Bound protein was eluted with a 100 -800 mM NaCl gradient, and the final purified TBPc protein that eluted at ϳ380 mM NaCl was concentrated by a 3000 molecular weight cutoff Amicon ultracentrifuge device. In some experiments, TBPc was purified as single step procedure leaving the N-terminal His 6 tag. The procedure was similar to that above for binding to the Ni-NTA resin except the resins after extensive washing were stepwise-eluted with increasing concentrations of imidazole, including 100, 200, and 400 mM. TBPc eluted at 400 mM imidazole was collected and concentrated as the final purified product (Fig. 4A). TBPc purification was Ͼ95% by both procedures, and the products were confirmed by immunoblotting with antibody to the C-terminal domain of TBP.
GST Pulldown and Co-immunoprecipitation Assays-A GST-TBP C fusion protein was expressed from pGEX-2T-TBP C in BL21 Gold cells, and lysates were prepared as above for Histagged TBP C . Domains of PR with N-terminal polyhistidine tags were expressed in Sf9 cells, and whole cell extracts were prepared as described above for purifications (except without imidazole, and DTT was used as reducing agent instead of BME). The total concentration of Sf9 whole cell extracts was adjusted to 7 mg/ml. Equal amounts of free GST or GST-GST-TBP C in cell lysates, as assessed by immunoblot assay with a mAb to GST, were incubated in batch in 2-ml microcentrifuge tubes for 1.5 h at 4°C with glutathione-Sepharose beads (50 l slurry) in a total volume of 250 l with binding buffer consisting of 10 mM Tris base, pH 7.4, 1 mM EDTA, 10% glycerol, 1 mM DTT, 100 mM NaCl, and 0.5% molecular grade BSA. Beads were washed with binding buffer with and without 1 M NaCl and 0.1% Nonidet P-40. Varying amounts of Sf9 whole cell extracts were incubated with the washed beads in suspension in binding buffer in a total volume of 500 l for 2 h at 4°C. The beads were washed three times in binding buffer, transferred to a clean tube, and washed twice more. Bound protein was eluted with SDS sample buffer and analyzed by immunoblot with antibodies to PR.
Hydrogen/Deuterium Exchange (HDX) and Mass Spectrometry-Solution phase amide HDX was carried out with a fully automated system as described previously (42). Briefly, 4 l of 10 M purified PR was diluted to 20 l with D 2 O-containing HDX buffer and incubated at 4°C for 10, 30, 60, 900, or 3600 s. Following an exchange, unwanted forward or back exchange was minimized, and the protein was denatured by dilution to 50 l with 0.1% (v/v) TFA in 3 M urea (held at 1°C). Samples were then passed across an immobilized pepsin column (prepared in-house (see Ref. 43)) at 50 l min Ϫ1 (0.1% v/v TFA, 15°C); the resulting peptides were trapped on a C8 trap cartridge (Hypersil Gold, Thermo Fisher). Peptides were then gradient-eluted (4% (w/v) CH 3 CN to 40% (w/v) CH 3 CN, 0.3% (w/v) formic acid over 5 min, at 2°C) across a 1 mm ϫ 50 mm C18 HPLC column (Hypersil Gold, Thermo Fisher) and electro sprayed directly into an Orbitrap mass spectrometer (LTQ Orbitrap with ETD, Thermo Fisher). Peptide ion signals were confirmed if they had a MASCOT score of 20 or greater and had no ambiguous hits using a decoy (reverse) sequence in a separate experiment using a 60-min gradient. The intensity weighted average m/z value (centroid) of each peptide's iso-topic envelope was calculated with in-house developed software and corrected for back-exchange as described previously (42,43).
Surface Plasmon Resonance (SPR) Analysis-The kinetics of TBP C binding to PR was determined by SPR on a Biacore X-100 plus instrument (GE Healthcare). The binding reaction was carried out at room temperature. Purified TBP C was immobilized to the Fc2 channel of the C1 chip as the ligand at 200 -250 response units through strong ionic interactions. The Fc1 channel was equally treated but without TBP C as the control. Multicycle kinetics procedure was employed to measure the binding. PR at different concentrations (0.03-1.0 M) was used as analyte and sequentially injected over Fc1 and Fc2 channels to measure its binding to TBP C . The sensor surface was regenerated by 0.3% SDS after each cycle of binding. The flow rate was kept constant at 30 ml/min. Data from 120 s of association and 180 s of dissociation were collected. The sensorgrams were normalized by the subtraction of Fc1 from Fc2. One medium concentration was repeated twice to monitor the reproducibility of the assay. The data were double reference subtracted and fitted for kinetics using various models depending on the actual binding reaction pattern (BIAevaluation 3.0 software). The dissociation constant (K D ) was calculated from the equation where k a is the association rate, and k d is the dissociation rate.
Circular Dichroism (CD) Spectroscopy-The far-UV CD spectra of purified recombinant PR NTD, TBP C , and PR NTD/ TBP C mixtures were recorded at 22°C on a Jasco 815 spectropolarimeter by using a 0.1-cm quartz cell, with a bandwidth of 0.5 nm and a scan step of 0.5 nm. Similar conditions were applied for recording far-UV CD spectra of purified recombinant full-length PR-A and PR-A/TBP C mixtures. All the spectra recorded were corrected for the contribution of solute concentrations. Each spectrum is a result of five spectra accumulated, averaged, and smoothed.
Fluorescence Emission (FE) Spectroscopy-Fluorescence emission spectra of purified recombinant PR NTD in solution were recorded in the absence or presence of varying concentrations of TMAO using a Spex FluoroMax spectrometer at excitation wavelengths of 278 or 295 nm as described (32). Measurements were taken in 1-cm rectangular cuvettes thermostated at 22°C, and all data were corrected for the contribution of the buffer. Equilibrium folding curves were fitted to the linear extrapolation model. Data were fitted to the linear extrapolation model as described (29).
Limited Proteolytic Digestion-Three sets of purified proteins (PR NTD, TBP C , and PR NTD/TBP C mixture) were digested by using trypsin (Promega). Digestions were carried out at 4°C by using a protein/enzyme mass ratio of 100:1. Reactions were terminated by adding SDS loading buffer and placing the sample tubes in boiling water. The proteolytic digestion products were resolved on SDS-PAGE followed by either Coomassie Blue R-250 staining or immunoblotting with PR NTDspecific antibodies.
Cell Culture and Reporter Gene Assays-CV-1 monkey kidney epithelial cells were grown in minimum Eagle's medium with Earle's salts supplemented with 10% fetal bovine serum as described previously (25). Cells were plated in 24-well dishes and 24 h later were transfected using Lipofectamine according the manufacturer's protocol (Invitrogen). Cells were co-transfected with 0.13 g of a SEAP reporter vector driven by a consensus glucocorticoid-response element (GRE) along with 0.05 g of pCDNA3.1-TBP or SRC-1 and either full-length human PR-B or a two-domain PR-B NTD/DBD vector (aa 1-650). Total amount of DNA added was constant by addition of an empty pRSV vector. Cells were then treated with or without the synthetic progestin R5020 (10 nM) for 24 h, and culture medium was collected and assayed for SEAP activity according to manufacturer's protocol. All treatment groups were done in three separate replicate wells, and independent experiments were repeated at least three times.

RESULTS
Intrinsic Disorder Characteristics of the PR NTD and Its Ability to Undergo a Disorder-Order Transition-FE spectroscopy and CD spectroscopy were used to examine solution structure properties of the NTD of PR. PR is expressed as two isoforms, full-length PR-B and N-terminally truncated PR-A lacking aa 1-164 of PR-B (9). The NTD of PR-A containing a His 6 tag was expressed from Sf9 cells and purified to Ͼ95% by sequential nickel affinity resins and HiTrap Q anion exchange chromatography (Fig. 1A). Both FE and CD analysis demonstrated that the NTD is largely unstructured under native conditions and acquires a substantial increase in ␣-helical content and a more compact tertiary structure in the presence of the natural osmolyte TMAO. The relative fluorescence intensity (F/F 0 at 329 nm emission) of the NTD at 278 nm excitation wavelength increased from Ͻ0.5 to 1.0 with increasing amounts of TMAO indicating more tertiary folding (Fig. 1B). A more compact structure was also shown by the increased intensity of fluorescence signal and blue shift of maximal wavelength. The apparent thermodynamic parameters of TMAO-induced folding are ⌬G ϭ Ϫ3.6 Ϯ 0.7; m ϭ 2.2 Ϯ 0.4 (Fig. 1B). The fluorescence changes are typical of those accompanying the removal of aromatic residues from polar aqueous solution into a more hydrophobic environment. Both the increase in quantum yield and the blue shift in fluorescence maximum indicate the formation of a compact structure in the presence of TMAO. The conformational transition in the PR NTD occurs in a cooperative manner, as shown by monitoring the level of fluorescence at 329 nm (upon excitation at 278) and the shift in emission maximum as a function of TMAO concentration (Fig. 1C). Fig. 1C presents the fluorescence emission spectra of PR NTD measured upon excitation at 295 nm to follow changes in the environment of Trp residues specifically. Far UV CD spectroscopy was used to examine secondary structure of the PR NTD. The CD spectrum exhibited little or no negative ellipticity at 222 nm and has a large negative maximum below 210 nm that is typical of protein without a stable ␣-helix and random coil conformation (Fig. 1D). An increase in negative ellipticity at 222 nm and a decrease Ͻ210 nm was observed in the presence of TMAO indicative of a gain in ␣-helical content and a reduction in random coil (Fig. 1D). Results in Fig. 1 are with the NTD of the PR-A isoform. Similar results were obtained with NTD of PR-B (data not shown).

HDX-MS Analysis of the Conformation Dynamics of NTD in the Intact
Receptor-As a complementary approach to examine solution structure properties of the PR NTD, we performed HDX mass spectrometry. HDX-MS measures the exchange kinetics of backbone amide hydrogens with solvent deuterium over time, with the rate of exchange reflective of the local structural environment of the protein (44,45). Protein with stable secondary and folded tertiary structure undergoes slow rates of deuterium exchange, whereas unfolded solventexposed regions exhibit rapid exchange. Digesting the protein after quenching of the D 2 O exchange reaction and sequencing peptides by ESI-MS enable assignment of exchange kinetics to specific sequence regions. Thus, HDX-MS has the added capability of localizing conformational dynamics to sequence regions within the intact protein. Full-length PR (A-isoform) bound to hormone agonist (R5020) was purified to Ͼ95% as a recombinant polyhistidine-tagged protein expressed in Sf9 insect cells by sequential nickel affinity resins and gel filtration by S200 columns (Fig. 2A). Receptor protein was confirmed to be functional with respect to near stoichiometric binding to R5020 and to a specific progesterone-response element (PRE) DNA oligonucleotide (data not shown). Using a fully automated HDX-MS platform with immobilized pepsin columns (42,43), we were able to obtain HDX with 72% peptide coverage of PR-A (Fig. 2B). Slowly exchanging portions of the protein (Fig. 2B, blue in heat maps) occurred in helices of the LBD and DBD. In contrast, the NTD exhibited much more rapid HDX kinetics with some regions undergoing Ͼ90% (red in heat map) exchange (Fig. 2B). The deuterium buildup curves of selected peptides from different domains of receptor as a function of time is shown in Fig. 2C. An overlay of the average percent deuterium uptake over the time course of the experiment onto the crystal structures of the PR LBD and DBD (46,47) is shown in Fig. 2D. Because there is no high resolution structure of the NTD and hinge, these regions are illustrated as a schematic. The overlay shows that NTD and hinge exhibit the highest rate of hydrogen-deuterium exchange, whereas structurally stable helices in the LBD and DBD exhibit dramatically less exchange. These results demonstrate the highly dynamic unstructured properties of the NTD in the context of the intact receptor.
Binding Analysis of TBP C Interaction with the NTD of PR-To explore the role of a target protein to induce ordered conformation of the PR NTD under solution conditions, we examined interaction with the C-terminal domain of TBP (TBP C ). Different domains of PR were expressed in Sf9 insect cells and analyzed as cell lysates for protein interaction with TBP C by pulldown assay with a GST-TBP C fusion protein. The NTD of PR-A exhibited specific binding with immobilized TBP C -GST over that of free GST, and the binding increased with varying amounts of input PR-A NTD (Fig. 3, A and B). Full-length PR-B also bound specifically with TBP C by GST-pulldown assay with similar efficiency as the isolated PR-B NTD (Fig. 3C). The PR LBD exhibited minimal or no specific binding (data not shown) indicating TBPc binding is primarily with the NTD.  OCTOBER 18, 2013 • VOLUME 288 • NUMBER 42

JOURNAL OF BIOLOGICAL CHEMISTRY 30289
The binding kinetics of the TBP C -PR NTD interaction was quantified by SPR with a Biacore X100 Plus and using purified proteins. TBP C purified to Ͼ95% as described under "Experimental Procedures" (Fig. 4A) (38) was immobilized to C1 chip as the ligand through strong ionic interaction as described previously (33). Full-length PR-A was used as the analyte. Low density of TBP C (200 response units) was immobilized to eliminate mass transport limitation and heterogenic ligand for the kinetics assay. The regeneration was optimized to remove both TBP C and PR-A after the binding. Fresh TBP C was immobilized for each cycle; therefore, the activity of ligand was the same during the entire assay. PR-A exhibited specific binding to TBP C as indicated by a higher response and different kinetics in the assay channel (Fc2), whereas the control Fc1 channel was just the perturbation due to the analyte (data not shown). The adjusted sensorgrams show a typical dose-dependent response of binding but with complex kinetics (Fig. 4, B and  C). The spikes at the start and end of association are due to the change in buffer components, which is common in SPR assays (Fig. 4B). For clarity and to provide a better view of the sensorgrams, we removed the areas with the peaks arising due to changes in buffer conditions (Fig. 4C). The overall binding showed a moderate rate of association and a slow dissociation pattern with a calculated binding affinity (K D ) of 0.17 M (Fig.  4D). TBP C binding by SPR with isolated PR domains was only detected with NTD ( Fig. 4E) but not with LBD as analyte (data not shown).
Influence of TBP C Interaction on Structure of the PR NTD-To examine the effect of TBP C binding on structural changes in the PR NTD, we analyzed the far-UV CD spectra of TBP C , the NTD of PR-A, or an equal molar mixture of TBP C and the NTD (Fig. 5A). The experimental CD spectrum of the mixture of PR-A NTD and TBP C was significantly different from the theoretical sum of the spectra of the two proteins (Fig. 5A, lower  panel). An increased negative absorbance peak at 222 nm and a reduction of negative absorbance peaks at wavelengths below 210 nm were observed with the protein mixture indicative of more ␣-helix secondary structure and less random coil (Fig. 5A). By deconvolution of the CD spectra using the K2D2 Program (48), we calculated that the NTD alone contains 8% ␣-helix, 21% ␤-sheet, and 71% random coil, whereas TBP C alone contains 43% ␣-helix, 10% ␤-sheet, and 47% random coil configuration. In the presence of both proteins, the values are 57% ␣-helix, 5% ␤-sheet, and 38% random coil. The effect of the TBP interaction on secondary structure in the context of full-length PR-A (bound to hormone agonist) was also analyzed. Similar to results with NTD, the experimental spectrum of the protein mixture was different from the theoretical summation of the individual spectra with an increased negative absorbance peak at 222 nm (Fig. 5B). Similar results were obtained with PR-B as well (data not shown). These results indicate that binding of TBP C induces secondary structure formation in the isolated NTD and with full-length PR.
Protection of the NTD against limited proteolysis was used to explore the influence of TBP interaction on tertiary structure of the NTD and to determine whether conformational changes in the complex occur in PR NTD as opposed to TBP. Proteolytic products of the PR NTD or TBP C alone generated by limiting digestion with trypsin as compared with a mixture of PR NTD and TBP C are shown by Coomassie-stained SDS gels in Fig. 6A. The PR NTD was highly susceptible to proteolytic degradation resulting in nearly complete degradation with little or no intact NTD remaining (Fig. 6A, compare lane 1 with lane 3). TBP C by comparison is resistant to limited proteolysis (Fig. 6A, compare   lane 2 with lane 4). The major fraction of TBP C was intact after treatment with trypsin with only a small amount of proteolytic fragment detected (Fig. 6A). Limited proteolysis of the protein mixture generated multiple smaller fragments not detected by digestion with either the PR NTD or TBP C alone (Fig. 6A, lane  5). Similar results were obtained by digesting with another protease, chymotrypsin (data not shown). To confirm whether protected fragments generated from the NTD/TBP mixture were coming from the PR NTD, tryptic digests in a separate experiment were analyzed by immunoblotting with two mAbs to different epitopes within the NTD of PR (39 -41). No fragments of the NTD were detected with the mAbs when the PR NTD alone was digested (Fig. 6B). Only a small residual amount of intact NTD remained (Fig. 6B). When PR NTD and TBP C were mixed together, protected PR NTD fragments were detected, and the patterns were distinct with each mAb indicating that multiple sites within the NTD change their accessibility to trypsin in the presence of TBP C (Fig. 6B). Protection against partial proteolysis indicates that the NTD folds into a more compact conformation when complexed with TBP C in addition to the increased secondary structure detected by CD analysis.

Identification of a Subregion of PR NTD Required for Coupled Folding upon TBP Interaction-Three different sequence regions of the PR NTD (PR-A) were expressed in bacterial cells and purified as described under "Experimental Procedures."
The regions of the NTD, including aa 165-300, aa 350 -428, and aa 456 -555, were chosen (see Fig. 7) based on previous functional data indicating an inhibitory function associated with aa 165-300; a major transcriptional activity is associated with aa 350 -428 in the context of other PR domains, and aa 456 -555 were originally defined as AF1 based on activity when linked to a heterologous GAL4 DNA binding domain (49 -52). By far-UV CD analysis, each NTD fragment in native aqueous conditions contained a large content of random coil and exhibited a substantial increase in helical content in the presence of TMAO indicating a propensity to form more ordered structure (Fig. 7A). Far-UV CD spectra were collected with TBP C alone and each of the NTD fragments in the presence and absence of TBP C . The experimental summation of CD spectra of the protein mixtures was only different from the theoretical with the aa 350 -428 fragment. No differential was detected with the other NTD fragments indicating that TBP interaction and associated structural changes require a region of the PR NTD between aa 350 and 428 (Fig. 7B).
Secondary structure predictions (PSIPRED) of the 350 -428-aa region of the NTD revealed two short ␣-helices that could potentially be binding sites for TBP or that stabilize upon interaction with TBP (Fig. 7C). To disrupt the potential ␣-helices, we introduced alanine substitutions for each of the glutamic acid residues between aa 389 and 392 and a proline for valine 406 (Fig. 7C). Each of these mutants within the context of the 350 -428-aa NTD fragment as a purified protein were analyzed by CD in the presence and absence of TBP C . Two of the mutations, E390A and V406P, abrogated secondary structural changes as a result of the TBP C interaction (Fig. 7D). Other mutations (E389A, E391A, and E392A) had no effect (data not shown). These data taken together suggest that a region of the FIGURE 3. TBPc binding with the PR NTD by pulldown assay. Full-length PR-B or isolated NTDs were expressed in the baculovirus insect cell system and were prepared as whole cell lysates and incubated with a GST-TBPc fusion protein immobilized to glutathione-Sepharose resins. Free GST immobilized to resins or blank beads was used as controls for nonspecific binding. After washing, proteins were eluted from the beads and analyzed by immunoblotting with antibodies to epitopes of the PR NTD. OCTOBER 18, 2013 • VOLUME 288 • NUMBER 42

JOURNAL OF BIOLOGICAL CHEMISTRY 30291
NTD with the potential to form two short helices is required for coupled folding and binding with TBP C .
Effect of TBP Interaction on NTD/AF1-dependent Transcriptional Activity-CV-1 cells were co-transfected with either an expression vector containing a two-domain PR-B NTD-DBD construct or full-length PR-B in the presence and absence of an expression plasmid for TBP. Cells were also transfected with a GRE/PRE-dependent SEAP reporter gene, and reporter gene activity was measured as described previously (33). As anticipated, the two-domain PR-B NTD-DBD construct exhibited constitutive transcriptional activity in the absence of hormone (28). Co-transfection with TBP increased the transcriptional activity mediated by PR-B NTD-DBD (Fig. 8A). Similar results were observed in HeLa cells and with other reporter genes (luciferase) driven by natural PREs of murine mammary tumor virus or a synthetic consensus PRE linked to the tk promoter (Fig. 8C). Thus functional response was not restricted to a single cell line or reporter gene. Enhancement of PRE-tk-Luc by TBP is of note because the tk promoter lacks a TATA box (53), suggesting that TBP stimulation of PR NTD-mediated transcription is not due to simply facilitating formation of the RNA polymerase initiation complex. Because this PR construct lacks the LBD and AF2, these data indicate that TBP interaction can stimulate NTD-mediated activity independent of AF2 (Fig. 8A). Internal deletion of regions of the NTD between aa 475 and 534 had no effect on the constitutive activity of the two-domain PR-B NTD-DBD or stimulation by TBP; however, the stimulatory effect of TBP was lost by deletion of aa 323-427 (Fig. 8A). Full-length PR-B requires hormone for transactivation, and cotransfected TBP also enhanced progestin (R5020)-dependent transcriptional activity of PR-B (Fig. 8B). Deletion mutation between aa 475 and 534 within the context of full-length PR-B had no effect on TBP enhancement of progestin (R5020)-dependent PR-mediated transactivation. However, deletion of aa 323-427 reduced, but did not eliminate, the stimulatory effect of TBP (Fig. 8B). The partial reduction in response to TBP by deletion of aa 323-427 in the full-length PR-B, as opposed to the nearly complete loss observed with the two-domain PR NTD-DBD protein, indicates that TBP can regulate the NTD independently but has additional effects on AF2 or other regions of the intact receptor. The observed requirement of overlapping sequences within the NTD (aa 323-427 and 350 -428) for functional response to TBP (Fig. 8), and for folding in the presence of TBP C in vitro (Fig. 7), indicates that transcriptional activity is associated with folding of the NTD-TBP C complex.
TBP Facilitates Binding and Activity of Steroid Receptor Coactivator 1 (SRC-1) with the PR NTD-The p160 family of SRCs has been reported to contain protein interaction surfaces in regions other than the LXXLL motifs of the nuclear receptor box that can interact with and enhance NTD-dependent transcription (54 -59). A co-immunoprecipitation assay was used to examine whether TBP-induced folding influences SRC-1 interaction with the PR NTD. Purified isolated PR NTD was mixed with HeLa cell nuclear extracts as a source of cellular SRC-1. Samples were immunoprecipitated with a specific antibody to SRC-1, and antibody-isolated proteins were analyzed by immunoblot with a PR-specific antibody that recognizes an epitope in the NTD. Results show an interaction of SRC-1 with the PR NTD that was greatly enhanced by addition of TBP C FIGURE 5. Increased secondary structure and folding conferred by TBP interaction with the PR NTD. A, upper panels, far-UV CD spectrum of isolated PR-A NTD alone, TBP alone, and an equal molar mixture of PR A NTD ϩ TBP C . Lower panels, theoretical sum of the individual CD spectra of PR NTD ϩ TBP C (red) and experimentally determined sum (black). B, upper panel, CD spectrum of full-length PR-A (bound with R5020), TBP C , and equal molar mixture of PR-A ϩ TBP C . Lower panel, theoretical sum (red) and experimentally determined sum (black) of the CD spectra of full-length PR-A ϩ TBP C .  OCTOBER 18, 2013 • VOLUME 288 • NUMBER 42 (Fig. 9A). To determine whether TBP influences the ability of SRC-1 to enhance NTD-dependent transcriptional activity, CV-1 cells expressing the constitutively active two-domain PR-B NTD-DBD protein were co-transfected with an SRC-1 expression plasmid in the absence and presence of TBP. As shown in Fig. 9B, SRC-1 or TBP alone increased transcriptional activity mediated by PR NTD DBD, although both proteins together acted cooperatively giving a greater stimulation than either alone. These data collectively support the conclusion that TBP may act as a scaffold to create a structural reorganization of the NTD required for optimal binding and function of transcriptional co-activators such as SRC-1.

DISCUSSION
Large segments of IDP are predicted to exist with high frequency in activation domains of transcription factors, including NRs, and the highest predicted level of disorder is in the NTD across receptors and species (14,60). IDRs are increasingly appreciated as functionally important for molecular recognition of protein-protein interactions involved in cellular signaling. IDRs display conformational flexibility capable of adapting to the structure of different proteins and thus can potentially expand the range of interacting proteins in the cell as compared with globular and more structurally stable motifs. Conformational flexibility of IDRs also provides a large interaction surface with high specificity and low affinity binding properties that are ideally suited for transient reversible interactions involved in signal transduction and transcriptional regulation.
The NTD accounts for a major proportion of the total transcriptional activity of all SRs, and in the case of androgen receptor and GR, it is the predominant activation domain over that of AF2 in the C-terminal LBD (8, 14, 61). The NTD is also impor- tant for cell and target gene-specific activities of SRs and in mediating the relative agonist and antagonist activities of small molecule ligand-selective receptor modulators (62). Despite the functional importance of the NTD, it has been less extensively studied than the C-terminal AF2. Although the NTD is the least conserved domain in SRs both at the amino acid sequence level and in length, it has been proposed that NTDs share common structural and mechanistic features as IDRs capable of folding into active stable conformations upon binding partner proteins or DNA (10 -14).
Biophysical analyses have directly demonstrated the IDP nature of the NTDs of GR, androgen receptor, estrogen receptor, and mineralocorticoid receptors (MRs). Folding of fulllength NTDs or NTD subregions of these receptors has been reported to be induced by organic osmolytes or through binding another protein (29 -36). With GR and MR, mutational analyses correlating induced folding in vitro with functional responses in cell transfection experiments have further shown a functional association between induced folding and transcriptional activity mediated by the NTDs (30 -31, 33). Using multiple solution phase biophysical methods, we further explored this hypothesis for SRs by demonstrating the intrinsic disorder properties of the NTD of PR and its ability to undergo induced folding either in the presence of the natural osmolyte TMAO or upon binding TBP. A strong association of TBP induced folding with transcriptional activity mediated by the NTD of PR was also observed.
HDX coupled with mass spectrometry has emerged as a sensitive approach for characterization of conformational dynamics of proteins in solution, and the technology has the advantage that it is not limited by the size of protein and can localize differential conformational dynamics to peptide sequences (42)(43)(44)(45). Because previous studies have examined IDP characteristics with isolated SR NTDs only, we used HDX-MS to examine conformational mobility of the NTD in the context of intact PR. HDX-MS analysis of full-length PR (A isoform) revealed that peptides derived from the NTD exhibited very rapid exchange relative to much slower kinetics of H/D exchange in peptides derived from the LBD and DBD (Fig. 2). The rapid H/D exchange rates of these data indicate that the dynamic conformation is a property of the NTD in the context of full-length PR. This is an important question because physical and allosteric interactions between domains of SRs have been reported, and conceptually interactions with other domains could stabilize the structure of the NTD. HDX-MS analysis of another full-length NR has been reported for the FIGURE 8. TBP enhances transcriptional activity of PR NTD and requires a region between aa 323 and 427 for maximal functional response. A, CV-1 cells were cotransfected with a constitutively active PR-B NTD-DBD two-domain polypeptide along with a pGRE/PRE-SEAP reporter gene in the absence and presence of a plasmid that expresses full-length TBP. Activity of the PR-B NTD/DBD construct was significantly increased by overexpression of TBP. Deletion mutation of the NTD between aa 323 and 427 abolished the stimulatory effect of TBP, whereas deletion of sequences between aa 475 and 534 had no effect. B, CV-1 cells were co-transfected as above except with full-length PR-B, and cells were treated with vehicle (EtOH) or the agonist R5020 (10 nM) for 24 h prior to assay of reporter gene activity. Overexpression of TBP significantly increased R5020-dependent PR-B-mediated activation of the reporter gene. With intact PR-B, deletion mutation within the NTD between aa 323 and 427 reduced the stimulatory effect of TBP, whereas deletion of aa 475-534 had no effect. SEAP reporter gene activity data were calculated as mean relative luminesence units (RLU) (ϮS.E.) from three replicate cell plating treatment groups. This is a single representative of three independent experiments. C, TBP enhancement of PR NTD-mediated transcriptional activity with other reporter genes in HeLa cells. HeLa cells were transfected with a two-domain PR-B NTD-DBD construct in the absence and presence of increasing amounts of TBP and a murine mammary tumor virus-luciferase or a PRE 2-tk-luciferase reporter as described previously (52). Firefly luciferase activity was measured and normalized to co-transfected constitutively active Renilla luciferase.
PPAR␥/retinoid X receptor heterodimer. Similar to results in this study, the NTD of PPAR␥ had a high rate of amide H/D exchange relative to the DBD and LBD (63). Interestingly, in the first high resolution x-ray structure of an intact NR with PPAR␥/retinoid X receptor heterodimers complexed to DNA, the NTD was disordered precluding structural information and restricting structure to the DBD, hinge, and LBD. The NTD of PPAR␥ is short (ϳ100 aa) as compared with steroid receptors that are up to 600 aa. This study is the first report of HDX-MS analysis of an intact steroid receptor with a long NTD (417 aa for PR-A NTD). A method for differential HDX-MS analysis to examine the effects of protein binding on conformational dynamics of intrinsically disordered protein regions was recently reported that should be applicable for further characterization of the intrinsically disordered NTDs of SRs (64).
To examine protein induced folding of the PR NTD, we chose to study interaction with TBP. Because TBP is the most commonly analyzed binding partner to affect folding of SR NTDs, it provided the best opportunity to further validate the hypothesis of a shared structural and functional mechanism for NTDs of all SRs. As determined by SPR, TBP binds the PR NTD with a low micromolar affinity (0.17 M), and binding kinetics showed a slower rate of dissociation than association (Fig. 4). The affinity and kinetics are similar to that reported for TBP C binding with NTDs of GR, MR, and ER␣ (31,33,36). The onand off-rates are consistent with a two-step model that involves a fast initial binding event and a slower dissociation once the complex is formed due to structural rearrangements that form a more stable conformation (31,33). As assessed by CD analysis, the NTD of PR exhibited an increase in ␣-helical content and a decrease in random coil upon complexing with TBP C (Fig. 5). When complexed with TBP C , the NTD also adopted a more folded tertiary structure as indicated by protection against proteolysis under limiting conditions (Fig. 6). Although the results of the CD analysis cannot rule out changes in secondary structure of TBP C in the complex, it appears that the major structural alteration is with the NTD. The CD spectrum of TBP C alone shows it is a much more ordered stable structure in the absence of an interacting protein partner than the NTD (Fig. 5). TBP C alone was also stable against limited proteolysis versus the NTD that was essentially completely digested under the same conditions (Fig. 6). These data collectively support the conclusion that changes in secondary and tertiary structure in the protein complex occur in the PR NTD.
Mapping experiments revealed that a central region of the NTD between aa 350 and 428 appeared to be sufficient for induced secondary structure in the presence of TBP C , whereas other regions are not involved (Fig. 7, A and B). This region of the NTD (aa 350 -428) contains two predicted short ␣-helices, and point mutations in selected residues abrogated TBP-coupled folding and binding (Fig. 7, C and D). These data suggest that short helical element(s) embedded within the larger disordered NTD may constitute a recognition site for TBP that can induce folding upon binding. Short local structural elements or pre-formed molecular recognition features have been identified within intrinsic disordered regions that may be required for assembly of protein complexes through disorder-to-order transitions (21). The central region of the NTD sufficient for TBPinduced folding was also required for functional response to TBP within intact cells. With the constitutively active two-domain PR NTD-DBD construct, deletion of aa 327-427 completely abolished the stimulatory effect of TBP on reporter gene activity in cell transfection experiments (Fig. 8A). Deletion of aa 475-534 that overlap the C-terminal region of the NTD that did not undergo TBP-induced folding had no effect on TBP enhancement of NTD-dependent activity (Fig. 8A). In the context of full-length PR-B, deletion of the central region of the NTD (aa 323-427) partially reduced the stimulatory effect of TBP, although deletion of aa 475 to 534 had no effect (Fig. 8B). Because full-length PR-B requires hormone for activity suggests that TBP may also affect AF2 in the LBD through cooperative interactions between the NTD/AF1 and AF2 in the LBD in the context of intact receptors. Nonetheless, these correlations with functional activity in cells suggest that TBP-coupled folding and binding is important for modulating the transcriptional activity of the PR NTD.
The C-terminal region of the NTD (aa 456 -555) was defined some time ago as a core AF1 based on functional activity of FIGURE 9. TBP facilitates steroid receptor co-activator-1 (SRC-1) interaction with the NTD. A, SRC-1 in HeLa nuclear extracts was incubated with the isolated NTD of PR-A, and samples were immunoprecipitated with an antibody specific to SRC-1. The immunoprecipitates were analyzed by immunoblotting with a PR antibody that detects an epitope in the NTD. Left panel shows immunoblots in duplicates and right panel, densitometric analysis of NTD bands from multiple experiments. B, CV-1 cells were co-transfected with a constitutively active two-domain PR-B NTD/DBD construct and the pGRE/ PRE-SEAP reporter gene vector in the absence and presence of plasmids that overexpresses SRC-1, TBP, or both SRC-1 and TBP together. At 48 h after cotransfection, SEAP reported gene activity was measured, and data were calculated as mean relative luminesence units (RLU) (ϮS.E.) from three replicate cell plating treatment groups. This is a single representative of three independent experiments.
overlapping NTD sequence regions linked to the GAL4 DNA binding domain and assayed for transactivation of a GAL4reporter gene (51). We subsequently showed, however, that internal deletion of aa 456 -555 in the context of full-length PR-B or a two-domain DBD-NTD PR polypeptide had no effect, respectively, on progesterone-dependent or constitutive transcriptional activity toward a PRE controlled target gene (28,52). Additionally, this region of the NTD linked to PR's own DBD exhibited only 1/10th the constitutive activity of the entire NTD (28). These data collectively indicate that aa 456 -555 does not constitute a minimal core AF1 as initially believed. The reason for the apparent discrepancy of results is likely due to interdomain allosteric interactions in the intact receptor that are not mimicked with an NTD region fused to a heterologous DBD as reported previously (23-25, 28, 65-66). Therefore, our data do not support the existence of a minimal core AF1 located between aa 456 and 555, and we prefer to use the term AF1 more generally as any NTD-dependent transcriptional activity in the context of receptors own the DNA binding domain.
The modular nature of the NTD with the potential to fold into multiple distinct active conformations is well exemplified by studies of the MR. Three regions of the MR NTD capable of mediating transactivation of reporter genes were identified, including two unstructured modules that undergo folding by osmolytes and binding of co-regulatory proteins such as CBP, SRCs, RIP140, and SMRT and a third region of pre-formed stable secondary structure (␤-sheet) that binds TBP without folding (31). Previous studies of truncation and deletion mutations of NTD of PR also illustrated that distinct regions of the NTD were required for co-activator enhancement of PR activity when bound to hormone agonist or antagonist suggesting the NTD has the flexibility to utilize distinct regions for different activities (52). These data, taken together with the highly flexible characteristics of IDPs, suggests that AF1 is not a discrete minimal sequence or motif in the NTD but consists of multiple regions capable of folding into different active stable conformations dependent on the binding partner protein.
The PR-A and PR-B isoforms exhibit distinct functional activities, and because they differ only by an additional (164 aa) N-terminal sequence in PR-B indicates that the NTD accounts for activity differences. In general, PR-B is a stronger transcriptional activator than PR-A, and distinct functional roles of the PR isoforms in different target tissues in vivo have been demonstrated from studies of PR-A and PR-B selective knock-out mice. Distinct structural conformations of the NTD have been implicated to be a contributing factor for the different activities of PR-A and PR-B (9). A separate transcription activation function (termed AF3) was mapped to sequences in the unique N-terminal extension of PR-B that has the potential to exert allosteric effects on the NTD or AF2 in the LBD (27,67). In this study, no differences were detected with isolated NTDs from PR-A or PR-B with respect to osmolyte or TBP-induced protein folding. Both NTDs exhibited similar characteristics as IDPs with the propensity to undergo a disorder-to-order transition. Because PR-B is a stronger activator, functional assays in this study were done only with PR-B or a two-domain PR-B NTD-DBD receptor construct. Thus, the experimental approach in this study was not able to resolve structural conformation differences in the NTD of PR-A and PR-B.
Because TBP is not a classical transcriptional co-activator, its physiological role in mediating the activity of NTD/AF1 is not clear. In previous studies, conditional folding of the NTD of GR and MR induced by osmolytes or TBP was shown to facilitate binding of co-activators such as SRC-1, -2, and -3, CBP and co-repressors SMRT and RIP140 (31,33). Conditional folding by TBP also enhanced the ability of SRC-1 to stimulate GR or MR NTD-dependent transcriptional activity. In this study, TBP C facilitated binding of SRC-1 with the PR NTD, and in cell transfection assays, TBP and SRC-1 cooperated to stimulate PR NTD-mediated transcriptional activity (Fig. 9). These data collectively suggest that TBP functions to reorganize or stabilize a structure in the NTD that provides a molecular recognition for binding SRCs and assembly of a co-regulator protein complex. How SRCs contact the NTD under these conditions is not known. In addition to LXXLL motifs that mediate binding to AF2 within the LBD of SRs, other interaction surfaces of SRCs have been documented to interact with the NTD of SRs, including a C-terminal AD2 region that binds to the NTD of the androgen receptor and an N-terminal fragment of SRC-2 (or TIF2) that binds the NTDs of GR or PR to modulate co-repressor interactions (54 -58). The N-terminal fragment of TIF2 (termed TIF2.0) was recently shown to bind the GR NTD/AF1 with an affinity similar to that of TBPc and was able to induce folding of the GR NTD/AF1 indicating that classical co-activators are also capable of directly altering structure of SR NTD (59). Whether TIF2.0 and TBP work together to affect structure and activity of the NTD of GR was not examined and will be important to know (59). The role of co-activators in modifying structure and activity of the NTD provides another dimension to our understanding of how steroid receptors are regulated as a transcription factor.