Structure-based screening of small-molecule interleukin-23 inhibitors inspired by monoclonal antibody interactions.

Interleukin-23 (IL-23) is a key driver of chronic inflammatory diseases, yet current therapies rely on costly monoclonal antibodies. This study aims to identify small-molecule IL-23 inhibitors using an in silico approach that mimics antibody interactions. The structure of IL-23 and the monoclonal antibody Risankizumab was reconstructed using homology modeling and deep learning. Key binding sites were characterized and used to generate 3D pharmacophore models, which guided virtual screening of compounds from DrugBank and ZINC12 databases. Top candidates were evaluated via ADMET filtering, molecular docking, molecular dynamics simulations and MM/GBSA binding free energy calculations. ZINC20572287 (r3-7) demonstrated stable binding within the IL-23p19 pocket and maintained strong hydrogen bonding over a 600 ns simulation. In contrast, no potent IL-12p40 inhibitors were identified. These findings suggest r3-7 as a promising scaffold for developing cost-effective IL-23-targeted therapeutics.