A Hierarchical Computational Screen Targeting A Homology-Modeled Androgen Receptor Ligand Binding Domain Antagonist Conformer Identifies Candidate Small-Molecule Inhibitors for Prostate Cancer

Androgen receptor (AR) dysregulation initiates transcriptional programs that promote prostate cancer (PCa) development and progression. Consequently, modulation of AR activity through androgen deprivation therapy and small-molecule inhibitors, such as bicalutamide (BIC) and enzalutamide (ENZ), remains as the mainstay treatment for PCa. However, resistance to these therapies ultimately develops, leading to poor patient prognosis and limited treatment options. The advent of sophisticated methods for simulating protein-ligand interactions allows the rapid identification of novel small-molecule inhibitors from large chemical libraries. The majority of available anti-AR drugs block the binding of testosterone (TES) or dihydrotestosterone (DHT) to the ligand binding domain (LBD) to prevent AR activation. However, the lack of wild-type AR LBD crystal structures in antagonist-bound conformation poses limitations in the structure-based identification of AR inhibitors and potentially lowers the hit rate during validation. To address these gaps, we performed a hierarchical computational screen of lead-like compounds targeting an antagonist model of the AR LBD generated through homology modelling. Using the crystal structure of the progesterone receptor LBD (55.04% protein sequence identity) bound to an inhibitor (PDB: 4OAR), we generated a homology model of the wild-type AR LBD antagonist conformer with GMQE and QMeanDisco Global scores of 0.72 and 0.76 ± 0.05, respectively. Molecular docking of 64,833 lead-like compounds from the ZINC22 chemical database identified 467 hits, which were subsequently filtered based on drug-likeness, PAINS hits, clogP, and hydrogen bond donors and acceptors of 200 known AR binders. A total of 16 compounds passed the screening. Conformational analysis of 9 selected complex structures through 100-ns atomistic molecular dynamics simulations showed that all remain relatively stable compared to DHT with C-alpha RMSD values less than 5.0 Å. Furthermore, MM/GBSA analysis showed comparable binding free energies of the compounds (-23.21 to -44.61 kcal/mol) to both DHT (-40.31 kcal/mol) and ENZ (-22.77 kcal/mol) and favorable interactions with amino acids known to be critical for androgen binding. Finally, principal component analysis of the AR LBD C-alpha trajectories showed that two compounds, SS2 and SS4, induce similar receptor dynamics as ENZ, highlighting their potential as AR inhibitors and warranting further in vitro and in vivo investigations.

*Unless otherwise noted, all abstracts presented at ENDO must not be released to the press or the public until the date and time of presentation. For oral presentations, the abstracts are embargoed until the session begins. The Endocrine Society reserves the right to lift the embargo on specific abstracts that are selected for promotion prior to or during ENDO.*