Comparison of crystal structures of human androgen receptor ligand-binding domain complexed with various agonists reveals molecular determinants responsible for binding affinity

Androgens exert their effects by binding to the highly specific androgen receptor (AR). In addition to natural potent androgens, AR binds a variety of synthetic agonist or antagonist molecules with different affinities. To identify molecular determinants responsible for this selectivity, we have determined the crystal structure of the human androgen receptor ligand-binding domain (hARLBD) in complex with two natural androgens, testosterone (Testo) and dihydrotestosterone (DHT), and with an androgenic steroid used in sport doping, tetrahydrogestrinone (THG), at 1.64, 1.90, and 1.75 A resolution, respectively. Comparison of these structures first highlights the flexibility of several residues buried in the ligand-binding pocket that can accommodate a variety of ligand structures. As expected, the ligand structure itself (dimension, presence, and position of unsaturated bonds that influence the geometry of the steroidal nucleus or the electronic properties of the neighboring atoms, etc.) determines the number of interactions it can make with the hARLBD. Indeed, THG--which possesses the highest affinity--establishes more van der Waals contacts with the receptor than the other steroids, whereas the geometry of the atoms forming electrostatic interactions at both extremities of the steroid nucleus seems mainly responsible for the higher affinity measured experimentally for DHT over Testo. Moreover, estimation of the ligand-receptor interaction energy through modeling confirms that even minor modifications in ligand structure have a great impact on the strength of these interactions. Our crystallographic data combined with those obtained by modeling will be helpful in the design of novel molecules with stronger affinity for the AR.     

Specific recognition of androgens by their nuclear receptor. A structure-function study

Androgens, like progestins, are 3-ketosteroids with structural differences restricted to the 17beta substituent in the steroid D-ring. To better understand the specific recognition of ligands by the human androgen receptor (hAR), a homology model of the ligand-binding domain (LBD) was constructed based on the progesterone receptor LBD crystal structure. Several mutants of residues potentially involved in the specific recognition of ligands in the hAR were constructed and tested for their ability to bind agonists. Their transactivation capacity in response to agonist (R1881) and antagonists (cyproterone acetate, hydroxyflutamide, and ICI 176344) was also measured. Substitution of His(874) by alanine, only marginally impairs the ligand-binding and transactivation capacity of the hAR receptor. In contrast, mutations of Thr(877) and, to a greater extent, Asn(705) perturb ligand recognition, alter transactivation efficiency, and broaden receptor specificity. Interestingly, the N705A mutant acquires progesterone receptor (PR) properties for agonist ligands but, unlike wild type AR and PR, loses the capacity to repress transactivation with nonsteroidal antagonists. Models of the hAR.LBD complexes with several ligands are presented, which suggests new directions for drug design.     

Effects of steroidal and non-steroidal antiandrogens on the androgen binding properties of the rat ventral prostate androgen receptor

Steroidal (cyproterone acetate) and non-steroidal (RU23908 and hydroxyflutamide) antiandrogens are able to block testosterone-induced increases in nuclear androgen receptor (AR) in the prostate of 1-day orchidectomized rats, but when given alone, RU23908 and hydroxyflutamide increase nuclear AR (RU23908 greater than hydroxyflutamide) in the same animal model. The increases in nuclear AR induced by antiandrogen alone or with testosterone alone are blocked by cycloheximide 1 h after administration, suggesting that androgen or antiandrogens induce de novo AR synthesis. Concomitant to nuclear AR accumulation, testosterone is able to induce depletion of cytosol and microsomal AR. Blockade of testosterone-induced depletion of microsomal AR, but not of cytosol AR, occurs in the presence of antiandrogens. Cyproterone acetate has a higher relative binding affinity (RBA) for microsomal AR and cytosol AR than RU23908 or hydroxyflutamide. This phenomenon is in good agreement with the degree of inhibition by these compounds of the association rate of androgen for the microsomal AR. This correlation between RBA and inhibition of the initial rate of hormone binding to the receptor is not found for cytosol AR. The results show that antiandrogens are not 'pure' antagonists of androgen action and they are potent agonists in the absence of testosterone. Furthermore, testosterone alone or antiandrogens per se regulate AR levels acutely by protein-synthesis dependent mechanisms of action, in rat ventral prostate.     

A new generation of progesterone receptor modulators

Progesterone receptor (PR) modulators have evolved both structurally and mechanistically over the past half-century. Classical steroidal PR agonists continue to play an important role in women's health such as in oral contraception and post-menopausal hormone therapy whereas steroid-based PR antagonists and selective PR modulators are being evaluated clinically in a wide range of gynecologic conditions. This review will focus primarily on the newer generation of PR modulators derived from structurally unique chemical scaffolds. For example, tanaproget (TNPR) is described as a non-steroidal PR agonist with high affinity and selectivity for the PR that is significantly more potent than many of its steroidal counterparts in a variety of pre-clinical efficacy models. Similarly, we present numerous examples of unique non-steroidal PR antagonists in various stages of characterization and development. A basic understanding of the structural determinants for high affinity binding of these new PR modulators to the PR ligand-binding domain (LBD) is also discussed. Finally, as the biology of the PR becomes further defined, we speculate on the future development of novel PR modulators.     

Recombinant expression and purification of human androgen receptor in a baculovirus system

A full-length human androgen receptor (hAR) cDNA was used to produce recombinant baculovirus. Spodoptera frugiperda (Sf9) cells infected with this virus expressed protein with an N-terminal hexahistidine tag (His(6)-hAR) in soluble and insoluble forms. The soluble cytosolic His(6)-hAR demonstrated similar association and dissociation half-times for mibolerone, similar binding affinity for mibolerone, and similar steroid specificity as bona fide AR. Under native conditions, the soluble cytosolic His(6)-hAR was purified to apparent homogeneity in the presence of dihydrotestosterone, using metal ion affinity chromatography. The insoluble pellet fraction was solubilized with strong denaturant 6 M guanidine HCl, and His(6)-hAR was purified from it in the presence of 6 M guanidine HCl. Both the solubilized crude pellet fraction and the solubilized/purified His(6)-hAR could be renatured to bind mibolerone. The baculovirus system will therefore provide an efficient means for producing hAR for ligand-binding assays, as well as purifying hAR for detailed molecular analyses.     

Glutamic acid 709 substitutions highlight the importance of the interaction between androgen receptor helices H3 and H12 for androgen and antiandrogen actions

The mutation of a single amino acid in the ligand binding domain of the human androgen receptor (AR) can induce functional abnormalities; for example, in androgen binding or interactions with coregulators. We report here on the structure/function analysis of the ARE709K substitution that is associated with partial androgen insensitivity syndrome. We introduced several mutations at position 709 and tested the consequences of these changes on AR structure and activity in the presence of androgen and antiandrogens. Our results demonstrate that a strong interaction between helix H12 and residue 709 in H3 is required to obtain a fully functional AR. We show that glutamic acid 709 can be replaced by a bulky tyrosine residue without significant effect on the activation by agonists. In contrast, smaller or linear residues that are unable to maintain a tight interaction with H12 induce a substantial loss of androgen-induced AR activity. We also show that the agonist activity of partial antiandrogens is dependent on the side-chain residue at position 709. Strikingly, the ARE709Y substitution causes the conversion of cyproterone acetate into a pure antiandrogen and bicalutamide into a partial agonist. Together, our structural and functional data reveal the key role of glutamic acid 709 in androgenic and antiandrogenic activities.     

Preliminary investigations into triazole derived androgen receptor antagonists

A range of 1,4-substituted-1,2,3-N-phenyltriazoles were synthesized and evaluated as non-steroidal androgen receptor (AR) antagonists. The motivation for this study was to replace the N-phenyl amide portion of small molecule antiandrogens with a 1,2,3-triazole and determine effects, if any, on biological activity. The synthetic methodology presented herein is robust, high yielding and extremely rapid. Using this methodology a series of 17 N-aryl triazoles were synthesized from commercially available starting materials in less than 3 h. After preliminary biological screening at 20 and 40 μM, the most promising three compounds were found to display IC₅₀ values of 40–50 μM against androgen dependent (LNCaP) cells and serve as a starting point for further structure–activity investigations. All compounds in this work were the focus of an in silico study to dock the compounds into the human androgen receptor ligand binding domain (hARLBD) and compare their predicted binding affinity with known antiandrogens. A comparison of receptor–ligand interactions for the wild type and T877A mutant AR revealed two novel polar interactions. One with Q738 of the wild type site and the second with the mutated A877 residue.     

Mass spectrometric characterization of the human androgen receptor ligand-binding domain expressed in Escherichia coli

The ligand-binding domain (LBD) of the human androgen receptor (hAR LBD), encompassing amino acids (AAs) 647-919, was expressed in Escherichia coli with an N-terminal polyhistidine tag (His(10)-hAR LBD) from a pET-16b vector. The overexpressed protein was initially insoluble in inclusion bodies, and was subsequently solubilized in 8 M guanidine hydrochloride (GdnHCl). The solubilized His(10)-hAR LBD was purified to apparent homogeneity by metal ion affinity chromatography in the presence of 6 M GdnHCl. The isolated protein migrated as a single band in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with an apparent molecular mass of 33-34 kDa, as expected from the plasmid construct. Immunoblot analysis with C-terminal antibodies raised against a peptide corresponding to the last 19 AAs (AAs 901-919) of hAR revealed that the purified protein contained an immunoreactive epitope present within the AR and was of the appropriate size. Further characterization, using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI/TOF-MS), showed a single protein species of average mass 34 580 Da, confirming the size and purity of the purified His(10)-hAR LBD. Detailed tryptic peptide mapping analysis, using MALDI/TOF-MS, identified a total of eight peptides with a 30% coverage of the LBD, including the last tryptic peptide in the hAR sequence. These data confirm that the purified protein was the intact hAR LBD. AA sequencing of these tryptic peptides, using an HPLC-coupled electrospray ionization ion trap mass spectrometer (LC/ESI-ITMS and MS/MS), unambiguously confirmed that the peptides were from the hAR LBD. The purified His(10)-hAR LBD in 6 M GdnHCl could be renatured as determined by ligand-binding activity, with a similar equilibrium dissociation constant (K(d)) for [(3)H]-mibolerone and a similar steroid specificity to the AR isolated from rat ventral prostate.     

Switching androgen receptor antagonists to agonists by modifying C-ring substituents on piperidino[3,2-g]quinolinone

New nonsteroidal human androgen receptor (hAR) agonists were developed from an hAR antagonist pharmacophore, 2(1H)-piperidino[3,2-g]quinolinone. (+/-)-trans-7,8-Diethyl-4-trifluoromethyl-2(H)-piperidino-[3,2-g]quinoli none was synthesized and demonstrated potent hAR agonist activity (EC50=3 nM) in the cell-based cotransfection assay and high binding affinity (Ki=16 nM) in the competitive receptor binding assay.     

Structural evidence for ligand specificity in the binding domain of the human androgen receptor. Implications for pathogenic gene mutations

The crystal structures of the human androgen receptor (hAR) and human progesterone receptor ligand-binding domains in complex with the same ligand metribolone (R1881) have been determined. Both three-dimensional structures show the typical nuclear receptor fold. The change of two residues in the ligand-binding pocket between the human progesterone receptor and hAR is most likely the source for the specificity of R1881 to the hAR. The structural implications of the 14 known mutations in the ligand-binding pocket of the hAR ligand-binding domains associated with either prostate cancer or the partial or complete androgen receptor insensitivity syndrome were analyzed. The effects of most of these mutants could be explained on the basis of the crystal structure.     

Structural basis for androgen receptor agonists and antagonists: interaction of SPEED 98-listed chemicals and related compounds with the androgen receptor based on an in vitro reporter gene assay and 3D-QSAR

The androgen receptor (AR) activity of listed chemicals, so called SPEED 98, by the Ministry of the Environment, Japan, and structurally related chemicals was characterized using MDA-kb2 human breast cancer cells stably expressing an androgen-responsive luciferase reporter gene, MMTV-luc. Since our results suggested that chemicals with diverse chemical structures were capable of disrupting the endocrine systems mediated by AR, a comparative molecular field analysis (CoMFA) model was developed to analyze the structural requirements necessary to disrupt AR function. A significant CoMFA model with r(2)=0.825 and q(2)=0.332 was developed for AR antagonist activity of 35 pure antagonists excluding procymidone. On the other hand, a good CoMFA model with r(2)=0.983 and q(2)=0.555 was obtained for antagonist activity of 13 chemicals with both agonist and antagonist activities. The steric and electrostatic properties were sufficient to describe the structural requirements for AR antagonist activity. In addition, the structural difference of AR agonists and antagonists was explained based on CoMFA results and the AR-LBD crystal structure. As several ERalpha agonists such as diethylstilbestrol (DES) acted as AR antagonists, the surface area of the AR ligand-binding domain (LBD) was compared with that of the ERalpha-LBD based on their reported crystal structures to analyze how those ligands interact with LBDs. The surface area of AR-LBD was shown to be smaller than that of ERalpha-LBD and therefore compounds with both estrogenic and antiandrogenic activities can fit well into the ERalpha-LBD but may protrude from the AR-LBD. It is likely that this subtle difference of the surface areas of the LBDs determines whether an ERalpha agonist acts as an AR antagonist or an agonist.     

Homology-modeled ligand-binding domains of medaka estrogen receptors and androgen receptors: A model system for the study of reproduction

Estrogen and androgen and their receptors play critical roles in physiological processes such as sexual differentiation and development. Using the available structural models for the human estrogen receptors alpha and beta and androgen receptor as templates, we designed in silico agonist and antagonist models of medaka estrogen receptor (meER) alpha, beta-1, and beta-2, and androgen receptor (meAR) alpha and beta. Using these models, we studied (1) the structural relationship between the ligand-binding domains (LBDs) of ERs and ARs of human and medaka, and (2) whether medaka ER and AR can be potential models for studying the ligand-binding activities of various agonists and antagonists of these receptors by docking analysis. A high level of conservation was observed between the sequences of the ligand-binding domains of meERα and huERα, meERβ1 and huERβ, meERβ2, and huERβ with 62.8%, 66.4%, and 65.1% identity, respectively. The sequence conservation between meARα and huAR, meARβ, and huAR was found with 70.1% and 61.0% of identity, respectively. Thirty-three selected endocrine disrupting chemicals (EDCs), including both agonists and antagonists, were docked into the LBD of ER and AR, and the corresponding docking score for medaka models and human templates were calculated. In order to confirm the conservation of the overall geometry and the binding pocket, the backbone root mean square deviation (RMSD) for Cα atoms was derived from the structure superposition of all 10 medaka homology models to the six human templates. Our results suggested conformational conservation between the ERs and ARs of medaka and human, Thus, medaka could be highly useful as a model system for studies involving estrogen and androgen interaction with their receptors.     

4-(Anilino)pyrrole-2-carboxamides: Novel non-steroidal/non-anilide type androgen antagonists effective upon human prostate tumor LNCaP cells with mutated nuclear androgen receptor

Various 4-(anilino)pyrrole-2-carboxamides were designed and synthesized as novel androgen receptor (AR) antagonists without steroidal or anilide structure, based on our strategy for developing full antagonists of nuclear receptors. Introduction of a bulky N-alkyl group, such as a cyclohexylmethyl or benzyl group, increased the binding affinity for wild-type AR and the potency for growth inhibition of androgen-dependent SC-3 cells. Among the compounds obtained, N-[4-[(benzyl)(4-nitrophenyl)amino]-1-methylpyrrole-2-carbonyl]pyrrolidine (22) is as potent an AR antagonist as the typical anilide-type AR antagonists hydroxyflutamide and bicalutamide. Further, compound 22 had potent binding affinity for T877A mutated AR, and dose-dependently inhibited the testosterone-induced production of prostate-specific antigen in LNCaP cells bearing T877A AR.     

Synthesis of novel steroidal agonists,partial agonists,and antagonists for the glucocorticoid receptor

Adverse effects of glucocorticoids could be limited by developing new compounds that selectively modulate anti-inflammatory activity of the glucocorticoid receptor (GR). We have synthesized a novel series of steroidal GR ligands, including potent agonists, partial agonists and antagonists with a wide range of effects on inhibiting secretion of interleukin-6. Some of these new ligands were designed to directly impact conformational stability of helix-12, in the GR ligand-binding domain (LBD). These compounds modulated GR activity and glucocorticoid-induced gene expression in a manner that was inversely correlated to the degree of inflammatory response. In contrast, compounds designed to directly modulate LBD epitopes outside helix-12, led to dissociated levels of GR-mediated gene expression and inflammatory response. Therefore, these new series of compounds and their derivatives will be useful to dissect the ligand-dependent features of GR signaling specificity.