Innovative In Silico Method for Discovering Multi-Target JAK/STAT Inhibitors as Potential Anticancer Therapies

Abstract

The JAK/STAT signaling pathway regulates key cellular processes such as immune modulation, cellular proliferation, and apoptosis. Its dysregulation is implicated in various cancers, making it an attractive target for anticancer therapies1. In this study, we present an innovative in silico strategy to identify multi-target inhibitors of JAK2, JAK3, and STAT3, key proteins in the JAK/STAT pathway, aimed at cancer treatment. We first refined the National Cancer Institute (NCI) database, consisting of 40,000 compounds, using QikProp and SwissADME to select drug-like candidates2. The Biotarget Predictor Tool (BPT), developed by our research group and available on the DRUDIT platform3, was then used in ON/OFF-target mode to predict compounds with high affinity for JAK2, JAK3, and STAT3, while minimizing off-target interactions with TNF-α and p53. Next, a two-step Docking Virtual Screening Workflow was applied, consisting of Extra Precision (XP) Docking and Induced Fit Docking (IFD), to refine the top candidates. Among them, compound 755435 (Figure 1) exhibited strong binding affinities and stable interactions with JAK2, JAK3, and STAT3, as confirmed by Molecular Dynamics Simulations and Interaction Maps with key active site residues. This multi-target approach, combined with OFF-target minimization, suggests that compound 755435 may offer a dual therapeutic advantage: enhancing efficacy in targeting multiple components of the JAK/STAT pathway while reducing the side effects associated with OFF-target interactions. Our results offer a promising foundation for the further optimization and preclinical evaluation of novel anticancer agents aimed at the JAK/STAT pathway.