Structural mechanism underlying variations in DNA binding by the androgen receptor.

The androgen receptor (AR) is a steroid activated transcription factor which recognizes DNA motifs resembling inverted repeats of a conserved 5'-AGAACA-3'-like hexanucleotides separated by a three-nucleotide spacer from a similar, but less conserved hexanucleotide. Here, we report the structures of the human AR DNA binding domain (DBD) bound to two natural AREs (C3 and MTV) in head-to-head dimer conformations, diffracting at 2.05 Šand 2.25 Å, respectively. These structures help to explain the impact of androgen insensitivity mutations on the structure integrity, DNA binding and DBD dimerization. The binding affinity of the AR DBD to different DNA motifs were measured by the BioLayer Interferometry (BLI) and further validated by Molecular Dynamics (MD) simulations. This shows that the high binding affinity of the first DBD to the upstream 5'-AGAACA-3' motif induces the cooperative binding of the second DBD to the second hexanucleotide. Our data indicate identical interaction of the DBDs to the upstream hexanucleotides, while forming an induced closer contact of the second DBD on the non-canonical hexanucleotides. The variation in binding between the DBD monomers are the result of differences in DNA occupancy, protein-protein interactions, DNA binding affinity, and DNA binding energy profiles. We propose this has functional consequences.

Rational evolution for altering the ligand preference of estrogen receptor alpha.

Estrogen receptor α is commonly used in synthetic biology to control the activity of genome editing tools. The activating ligands, estrogens, however, interfere with various cellular processes, thereby limiting the applicability of this receptor. Altering its ligand preference to chemicals of choice solves this hurdle but requires adaptation of unspecified ligand-interacting residues. Here, we provide a solution by combining rational protein design with multi-site-directed mutagenesis and directed evolution of stably integrated variants in Saccharomyces cerevisiae. This method yielded an estrogen receptor variant, named TERRA, that lost its estrogen responsiveness and became activated by tamoxifen, an anti-estrogenic drug used for breast cancer treatment. This tamoxifen preference of TERRA was maintained in mammalian cells and mice, even when fused to Cre recombinase, expanding the mammalian synthetic biology toolbox. Not only is our platform transferable to engineer ligand preference of any steroid receptor, it can also profile drug-resistance landscapes for steroid receptor-targeted therapies.

Discovery of a novel androgen receptor antagonist, MEL-6, with stereoselective activity and optimization of its metabolic stability.

A new chemical scaffold with antagonistic activity towards the androgen receptor (AR) was identified. The parent compound, (3-Methoxy-N-[1-methyl-2-(4-phenyl-1-piperazinyl)-2-(2-thienyl)ethyl]benzamide) referred to as MEL-6, binds in the ligand binding pocket of AR and induces an antagonistic conformation of the ligand binding domain, even in presence of the antagonist-to-agonist switch mutations W741C, T877A and F876L-T877A. MEL-6 has antiproliferative effects on several AR positive prostate cancer cell lines. We further identified AR as the specific target of MEL-6 since it demonstrates little effect on other steroid receptors. In LNCaP cells it also inhibits the androgen-regulated transcriptome. These findings identify MEL-6 as a promising candidate for treatment of patients with prostate tumors that have become resistant to current clinically used AR antagonists. Analytical studies on the chemical composition of MEL-6 identified the presence of four isomers (two enantiomeric pairs), among which one isomer is responsible for the antiandrogenic activity. We therefore developed a synthetic route towards the selective preparation of the active enantiomeric pair. Various MEL-6-like analogues had improved metabolic stability while maintaining antiandrogenic activity. Metabolite identification of MEL-6 derivatives pinpointed N-dealkylation of the piperazine as the main mode for inactivation by liver enzymes. For further structural optimization, MEL-6 derivatives were purchased or synthesized having alterations on the N-phenyl group of the piperazine, the benzoyl group and additionally substituting the thiophen-2-yl ring of MEL-6 to a phenyl ring. This optimization process resulted in compound 12b with sustained AR inhibition and a 4-fold increased half-life due to the 1-(5-chloro-2-methylphenyl)-piperazine substitution, thienyl-to-phenyl substitution and chloro in para-position of the benzoyl group.

Study of novel androgen receptor V770 variant in androgen insensitivity syndrome patients reveals the transitional state of the androgen receptor ligand binding domain homodimer.

We report the discovery of the androgen receptor missense mutation V770D, that was found in two sisters suffering from complete androgen insensitivity. Experimental validation of AR V770 variants demonstrated that AR V770D was transcriptionally inactive due to the inability to dimerize and a reduced ligand binding affinity. The more conservative AR V770A variant showed a dimerization defect at low levels of DHT with a partial recovery of the transcriptional activity and of the receptor's ability to dimerize when increasing the DHT levels. With V770 located outside of the proposed LBD dimerization interface of the AR LBD homodimer crystal structure, the effects of the V770A mutation on AR dimerization were unexpected. We therefore explored whether the AR LBD dimerization interface would be better described by an alternative dimerization mode based on available human homodimeric LBD crystal structures of other nuclear receptors. Superposition of the monomeric AR LBD in the homodimeric crystal structures of GR, PR, ER, CAR, TRß, and HNF-4α showed that the GR-like LBD dimer model was energetically the most stable. Moreover, V770 was a key energy residue in the GR-like LBD dimer while it was not involved in the stabilization of the AR LBD homodimer according to the crystal structure. Additionally, the observation that 4 AIS mutations impacted the stability of the AR LBD dimer while 16 mutations affected the GR-like LBD dimer, suggested that the AR LBD dimer crystal is a snapshot of a breathing AR LBD homodimer that can transition into the GR-like LBD dimer model.

Exploiting Ligand-binding Domain Dimerization for Development of Novel Androgen Receptor Inhibitors.

Currently, all clinically used androgen receptor (AR) antagonists target the AR ligand-binding pocket and inhibit T and dihydrotestosterone (DHT) binding. Resistance to these inhibitors in prostate cancer frequently involves AR-dependent mechanisms resulting in a retained AR dependence of the tumor. More effective or alternative AR inhibitors are therefore required to limit progression in these resistant stages. Here, we applied the structural information of the ligand-binding domain (LBD) dimerization interface to screen in silico for inhibitors. A completely new binding site, the Dimerisation Inhibiting Molecules (DIM) pocket, was identified at the LBD dimerization interface. Selection of compounds that fit the DIM pocket via virtual screening identified the DIM20 family of compounds which inhibit AR transactivation and dimerization of the full-length AR as well as the isolated LBDs. Via biolayer interferometry, reversible dose-dependent binding to the LBD was confirmed. While DIM20 does not compete with 3H-DHT for binding in the LBP, it limits the maximal activity of the AR indicative of a noncompetitive binding to the LBD. DIM20 and DIM20.39 specifically inhibit proliferation of AR-positive prostate cancer cell lines, with only marginal effects on AR-negative cell lines such as HEK 293 and PC3. Moreover, combination treatment of DIM compounds with enzalutamide results in synergistic antiproliferative effects which underline the specific mechanism of action of the DIM compounds.

N/C Interactions Are Dispensable for Normal In Vivo Functioning of the Androgen Receptor in Male Mice.

The androgen receptor (AR) plays a central role in the development and maintenance of the male phenotype. The binding of androgens to the receptor induces interactions between the carboxyterminal ligand-binding domain and the highly conserved 23FQNLF27 motif in the aminoterminal domain. The role of these so-called N/C interactions in AR functioning is debated. In vitro assays show that mutating the AR in the 23FQNLF27 motif (called ARNoC) attenuates the AR transactivation of reporter genes, has no effect on ligand binding, but does affect protein-protein interactions with several AR coregulators. To test the in vivo relevance of the N/C interaction, we analyzed the consequences of the genomic introduction of the ARNoC mutation in mice. Surprisingly, the ARNoC/Y mice show a normal male development, with unaffected male anogenital distance and normal accessory sex glands, male circulating androgen levels, body composition, and fertility. The responsiveness of androgen target genes in kidney, prostate, and testes was also unaffected. We thus conclude that the N/C interactions in the AR are not essential for the development of a male phenotype under normal physiological conditions.

A disease-associated missense mutation in CYP4F3 affects the metabolism of leukotriene B4 via disruption of electron transfer.

BACKGROUND: Cytochrome P450 4F3 (CYP4F3) is an ω-hydroxylase that oxidizes leukotriene B4 (LTB4), prostaglandins, and fatty acid epoxides. LTB4 is synthesized by leukocytes and acts as a chemoattractant for neutrophils, making it an essential component of the innate immune system. Recently, involvement of the LTB4 pathway was reported in various immunological disorders such as asthma, arthritis, and inflammatory bowel disease. We report a 26-year-old female with a complex immune phenotype, mainly marked by exhaustion, muscle weakness, and inflammation-related conditions. The molecular cause is unknown, and symptoms have been aggravating over the years. METHODS: Whole exome sequencing was performed and validated; flow cytometry and enzyme-linked immunosorbent assay were used to describe patient's phenotype. Function and impact of the mutation were investigated using molecular analysis: co-immunoprecipitation, western blot, and enzyme-linked immunosorbent assay. Capillary electrophoresis with ultraviolet detection was used to detect LTB4 and its metabolite and in silico modelling provided structural information. RESULTS: We present the first report of a patient with a heterozygous de novo missense mutation c.C1123 > G;p.L375V in CYP4F3 that severely impairs its activity by 50% (P < 0.0001), leading to reduced metabolization of the pro-inflammatory LTB4. Systemic LTB4 levels (1034.0 ± 75.9 pg/mL) are significantly increased compared with healthy subjects (305.6 ± 57.0 pg/mL, P < 0.001), and immune phenotyping shows increased total CD19+ CD27- naive B cells (25%) and decreased total CD19+ CD27+ IgD- switched memory B cells (19%). The mutant CYP4F3 protein is stable and binding with its electron donors POR and Cytb5 is unaffected (P > 0.9 for both co-immunoprecipitation with POR and Cytb5). In silico modelling of CYP4F3 in complex with POR and Cytb5 suggests that the loss of catalytic activity of the mutant CYP4F3 is explained by a disruption of an α-helix that is crucial for the electron shuffling between the electron carriers and CYP4F3. Interestingly, zileuton still inhibits ex vivo LTB4 production in patient's whole blood to 2% of control (P < 0.0001), while montelukast and fluticasone do not (99% and 114% of control, respectively). CONCLUSIONS: A point mutation in the catalytic domain of CYP4F3 is associated with high leukotriene B4 plasma levels and features of a more naive adaptive immune response. Our data provide evidence for the pathogenicity of the CYP4F3 variant as a cause for the observed clinical features in the patient. Inhibitors of the LTB4 pathway such as zileuton show promising effects in blocking LTB4 production and might be used as a future treatment strategy.

The T850D Phosphomimetic Mutation in the Androgen Receptor Ligand Binding Domain Enhances Recruitment at Activation Function 2.

Several key functions of the androgen receptor (AR) such as hormone recognition and co-regulator recruitment converge in the ligand binding domain (LBD). Loss- or gain-of-function of the AR contributes to pathologies such as the androgen insensitivity syndrome and prostate cancer. Here, we describe a gain-of-function mutation of the surface-exposed threonine at position 850, located at the amino-terminus of Helix 10 (H10) in the AR LBD. Since T850 phosphorylation was reported to affect AR function, we created the phosphomimetic mutation T850D. The AR T850D variant has a 1.5- to 2-fold increased transcriptional activity with no effect on ligand affinity. In the androgen responsive LNCaP cell line grown in medium with low androgen levels, we observed a growth advantage for cells in which the endogenous AR was replaced by AR T850D. Despite the distance to the AF2 site, the AR T850D LBD displayed an increased affinity for coactivator peptides as well as the 23FQNLF27 motif of AR itself. Molecular Dynamics simulations confirm allosteric transmission of the T850D mutation towards the AF2 site via extended hydrogen bond formation between coactivator peptide and AF2 site. This mechanistic study thus confirms the gain-of-function character of T850D and T850 phosphorylation for AR activity and reveals details of the allosteric communications within the LBD.

Antizyme Inhibitor 1 Regulates Matrikine Expression and Enhances the Metastatic Potential of Aggressive Primary Prostate Cancer.

Molecular drivers of metastasis in patients with high-risk localized prostate cancer are poorly understood. Therefore, we aim to study molecular drivers of metastatic progression in patients with high-risk prostate cancer. A retrospective matched case-control study of two clinico-pathologically identical groups of patients with high-risk prostate cancer was undertaken. One group developed metastatic recurrence (n = 19) while the other did not (n = 25). The primary index tumor was identified by a uro-pathologist, followed by DNA and RNA extraction for somatic copy-number aberration (SCNA) analysis and whole-transcriptome gene expression analysis. In vitro and in vivo studies included cell line manipulation and xenograft models.The integrative CNA and gene expression analyses identified an increase in Antizyme Inhibitor 1 (AZIN1) gene expression within a focal amplification of 8q22.3, which was associated with metastatic recurrence of patients with high-risk prostate cancer in four independent cohorts. The effects of AZIN1 knockdown were evaluated, due to its therapeutic potential. AZIN1 knockdown effected proliferation and metastatic potential of prostate cancer cells and xenograft models. RNA sequencing after AZIN1 knockdown in prostate cancer cells revealed upregulation of genes coding for collagen subunits. The observed effect on cell migration after AZIN1 knockdown was mimicked when exposing prostate cancer cells to bio-active molecules deriving from COL4A1 and COL4A2. Our integrated CNA and gene expression analysis of primary high-risk prostate cancer identified the AZIN1 gene as a novel driver of metastatic progression, by altering collagen subunit expression. Future research should further investigate its therapeutic potential in preventing metastatic recurrence. IMPLICATIONS: AZIN1 was identified as driver of metastatic progression in high-risk prostate cancer through matrikine regulation.

Small-molecule profiling for steroid receptor activity using a universal steroid receptor reporter assay.

A critical step in the development of novel drug candidates for the treatment of steroid related diseases is ensuring the absence of crosstalk with steroid receptors (SRs). Establishing this SR cross-reactivity profile requires multiple reporter assays as each SR associates with its unique enhancer region, a labor intensive and time-consuming approach. To overcome this need for multi-reporter assays, we established a steroid receptor inducible luciferase reporter assay (SRi-Luc) that allows side-by-side examination of agonistic and antagonistic properties of small-molecules on all steroid receptors. This state-of-the-art SRi-Luc consists of a unique alteration of four distinct keto-steroid- and estrogen response elements. As proof of principle, the SRi-Luc assay was used to profile a set of novel designed steroidal 1,2,3-triazoles. These triazolized steroidal compounds were developed via our in-house triazolization methodology, in which an enolizable ketone is converted into a triazolo-fused or -linked analog by treatment with a primary amine or ammonium salt in the presence of 4-nitrophenyl azide. From these designed steroidal 1,2,3-triazoles, six successfully reduced androgen receptor activity by 40 %. Although opted as antiandrogens, their cross-reactivity with other SRs was apparent in our SRi-Luc assay and rendered them unsuited for further antagonist development and clinical use. Overall, the SRi-Luc overcomes the need of multi-reporter assays for the profiling of small-molecules on all SRs. This not only reduces the risk of introducing biases, it as well accelerates early-stage drug discovery when designing particular SR selective (ant)agonists or characterizing off-target effects of lead molecules acting on any drug target.

The androgen receptor depends on ligand-binding domain dimerization for transcriptional activation.

Whereas dimerization of the DNA-binding domain of the androgen receptor (AR) plays an evident role in recognizing bipartite response elements, the contribution of the dimerization of the ligand-binding domain (LBD) to the correct functioning of the AR remains unclear. Here, we describe a mouse model with disrupted dimerization of the AR LBD (ARLmon/Y ). The disruptive effect of the mutation is demonstrated by the feminized phenotype, absence of male accessory sex glands, and strongly affected spermatogenesis, despite high circulating levels of testosterone. Testosterone replacement studies in orchidectomized mice demonstrate that androgen-regulated transcriptomes in ARLmon/Y mice are completely lost. The mutated AR still translocates to the nucleus and binds chromatin, but does not bind to specific AR binding sites. In vitro studies reveal that the mutation in the LBD dimer interface also affects other AR functions such as DNA binding, ligand binding, and co-regulator binding. In conclusion, LBD dimerization is crucial for the development of AR-dependent tissues through its role in transcriptional regulation in vivo. Our findings identify AR LBD dimerization as a possible target for AR inhibition.

Structure and stability of the designer protein WRAP-T and its permutants.

[Formula: see text]-Propeller proteins are common natural disc-like pseudo-symmetric proteins that contain multiple repeats ('blades') each consisting of a 4-stranded anti-parallel [Formula: see text]-sheet. So far, 4- to 12-bladed [Formula: see text]-propellers have been discovered in nature showing large functional and sequential variation. Using computational design approaches, we created perfectly symmetric [Formula: see text]-propellers out of natural pseudo-symmetric templates. These proteins are useful tools to study protein evolution of this very diverse fold. While the 7-bladed architecture is the most common, no symmetric 7-bladed monomer has been created and characterized so far. Here we describe such a engineered protein, based on a highly symmetric natural template, and test the effects of circular permutation on its stability. Geometrical analysis of this protein and other artificial symmetrical proteins reveals no systematic constraint that could be used to help in engineering of this fold, and suggests sequence constraints unique to each [Formula: see text]-propeller sub-family.

Androgen and glucocorticoid receptor direct distinct transcriptional programs by receptor-specific and shared DNA binding sites.

The glucocorticoid (GR) and androgen (AR) receptors execute unique functions in vivo, yet have nearly identical DNA binding specificities. To identify mechanisms that facilitate functional diversification among these transcription factor paralogs, we studied them in an equivalent cellular context. Analysis of chromatin and sequence suggest that divergent binding, and corresponding gene regulation, are driven by different abilities of AR and GR to interact with relatively inaccessible chromatin. Divergent genomic binding patterns can also be the result of subtle differences in DNA binding preference between AR and GR. Furthermore, the sequence composition of large regions (>10 kb) surrounding selectively occupied binding sites differs significantly, indicating a role for the sequence environment in guiding AR and GR to distinct binding sites. The comparison of binding sites that are shared shows that the specificity paradox can also be resolved by differences in the events that occur downstream of receptor binding. Specifically, shared binding sites display receptor-specific enhancer activity, cofactor recruitment and changes in histone modifications. Genomic deletion of shared binding sites demonstrates their contribution to directing receptor-specific gene regulation. Together, these data suggest that differences in genomic occupancy as well as divergence in the events that occur downstream of receptor binding direct functional diversification among transcription factor paralogs.

A Mechanistic High-Content Analysis Assay Using a Chimeric Androgen Receptor That Rapidly Characterizes Androgenic Chemicals.

Human health is at risk from environmental exposures to a wide range of chemical toxicants and endocrine-disrupting chemicals (EDCs). As part of understanding this risk, the U.S. Environmental Protection Agency (EPA) has been pursuing new high-throughput in vitro assays and computational models to characterize EDCs. EPA models have incorporated our high-content analysis-based green fluorescent protein estrogen receptor (GFP-ER): PRL-HeLa assay, which allows direct visualization of ER binding to DNA regulatory elements. Here, we characterize a modified functional assay based on the stable expression of a chimeric androgen receptor (ARER), wherein a region containing the native AR DNA-binding domain (DBD) was replaced with the ERα DBD (amino acids 183-254). We demonstrate that the AR agonist dihydrotestosterone induces GFP-ARER nuclear translocation, PRL promoter binding, and transcriptional activity at physiologically relevant concentrations (<1 nM). In contrast, the AR antagonist bicalutamide induces only nuclear translocation of the GFP-ARER receptor (at µM concentrations). Estradiol also fails to induce visible chromatin binding, indicating androgen specificity. In a screen of reference chemicals from the EPA and the Agency for Toxic Substances and Disease Registry, the GFP-ARER cell model identified and mechanistically grouped activity by known (anti-)androgens based on the ability to induce nuclear translocation and/or chromatin binding. Finally, the cell model was used to identify potential (anti-)androgens in environmental samples in collaboration with the Houston Ship Channel/Galveston Bay Texas A&M University EPA Superfund Research Program. Based on these data, the chromatin-binding, in vitro assay-based GFP-ARER model represents a selective tool for rapidly identifying androgenic activity associated with drugs, chemicals, and environmental samples.

Drivers of AR indifferent anti-androgen resistance in prostate cancer cells.

Inhibition of the androgen receptor (AR) by second-generation anti-androgens is a standard treatment for metastatic castration resistant prostate cancer (mCRPC), but it inevitably leads to the development of resistance. Since the introduction of highly efficient AR signalling inhibitors, approximately 20% of mCRPC patients develop disease with AR independent resistance mechanisms. In this study, we generated two anti-androgen and castration resistant prostate cancer cell models that do not rely on AR activity for growth despite robust AR expression (AR indifferent). They are thus resistant against all modern AR signalling inhibitors. Both cell lines display cross-resistance against the chemotherapeutic drug docetaxel due to MCL1 upregulation but remain sensitive to the PARP inhibitor olaparib and the pan-BCL inhibitor obatoclax. RNA-seq analysis of the anti-androgen resistant cell lines identified hyper-activation of the E2F cell-cycle master regulator as driver of AR indifferent growth, which was caused by deregulation of cyclin D/E, E2F1, RB1, and increased Myc activity. Importantly, mCRPC tissue samples with low AR activity displayed the same alterations and increased E2F activity. In conclusion, we describe two cellular models that faithfully mimic the acquisition of a treatment induced AR independent phenotype that is cross-resistant against chemotherapy and driven by E2F hyper-activation.

Validation of the Decipher Test for Predicting Distant Metastatic Recurrence in Men with High-risk Nonmetastatic Prostate Cancer 10 Years After Surgery.

BACKGROUND: Decipher is a genomic classifier designed to predict the development of distant metastases after surgical treatment of prostate cancer (PC). Its long-term prognostic role in a high-risk PC population has not been investigated previously. OBJECTIVE: To determine the prognostic role of the Decipher genomic classifier in two high-risk PC case-control studies. DESIGN, SETTING, AND PARTICIPANTS: Patients who developed distant metastases after surgery for high-risk, nonmetastatic PC in a European tertiary referral center from 1991 to 2011 were matched to patients not developing distant metastases (n=54). A validation study (n=298) was performed using a similar US case-control cohort. Formalin-fixed, paraffin-embedded tissue blocks from the index PC lesion were used for RNA extraction and gene expression analysis. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: The outcome investigated was the development of distant metastasis within 10-yr follow-up. Multivariable logistic regression analysis was performed, with statistical significance considered at p<0.05. RESULTS AND LIMITATIONS: In both the European and US case-control studies, the median Decipher scores were higher in the population that developed metastases. In the multivariable analysis, each 10% increase in Decipher score translated to an increase in the risk of distant metastases within 10-yr follow-up, with an odds ratio of 1.53 (95% confidence interval [CI] 1.06-2.22; p=0.025) and 1.58 (95% CI 1.31-1.92; p<0.001) for the European and US cohorts, respectively. Median follow-up for the European cohort was 12yr (interquartile range 8-12). The study limitation is the small size of the European cohort. CONCLUSIONS: Our study validates Decipher as a predictor for metastatic recurrence even in patients with high-risk, nonmetastatic PC within 10-yr follow-up. PATIENT SUMMARY: Decipher is a test based on gene expression profiles in primary tumors in prostate cancer. It has already been proven to predict cancer recurrence after surgery, but this has not yet been shown for patients with high-risk prostate cancer. This is the first study confirming that Decipher predicts a patient's risk of developing cancer recurrence after surgery for high-risk prostate cancer.

Synthesis, biological evaluation and molecular modeling of a novel series of fused 1,2,3-triazoles as potential anti-coronavirus agents.

Synthesis and biological evaluation of a novel library of fused 1,2,3-triazole derivatives are described. The in-house developed multicomponent reaction based on commercially available starting materials was applied and broad biological screening against various viruses was performed, showing promising antiviral properties for compounds 14d, 14n, 14q, 18f and 18i against human coronavirus 229E. Further in silico studies identified the key molecular interactions between those compounds and the 3-chymotrypsin-like protease, which is essential to the intracellular replication of the virus, supporting the hypothesis that the protease is the target molecule of the potential antiviral derivatives.

Structure of the homodimeric androgen receptor ligand-binding domain.

The androgen receptor (AR) plays a crucial role in normal physiology, development and metabolism as well as in the aetiology and treatment of diverse pathologies such as androgen insensitivity syndromes (AIS), male infertility and prostate cancer (PCa). Here we show that dimerization of AR ligand-binding domain (LBD) is induced by receptor agonists but not by antagonists. The 2.15-Å crystal structure of homodimeric, agonist- and coactivator peptide-bound AR-LBD unveils a 1,000-Å2 large dimerization surface, which harbours over 40 previously unexplained AIS- and PCa-associated point mutations. An AIS mutation in the self-association interface (P767A) disrupts dimer formation in vivo, and has a detrimental effect on the transactivating properties of full-length AR, despite retained hormone-binding capacity. The conservation of essential residues suggests that the unveiled dimerization mechanism might be shared by other nuclear receptors. Our work defines AR-LBD homodimerization as an essential step in the proper functioning of this important transcription factor.