Molecular Docking and Simulation-Based Exploration of Niclosamide as a Potential Inhibitor of the p62 ZZ Domain

Acute myeloid leukemia (AML) remains a therapeutic challenge due to complex oncogenic networks, including the often-undruggable MYC pathway. Here, we present an integrated in silico framework combining transcriptomic analysis, machine learning, and molecular dynamics (MD) simulations to explore potential therapeutic approaches targeting vault RNA1-1 (VTRNA1-1) in AML. RNA-seq profiling revealed that VTRNA1-1 depletion is associated with a profound disruption of the MYC and FOXM1 regulatory axes. To highlight compounds capable of recapitulating this transcriptomic signature, we developed a machine learning pipeline utilizing a Random Forest classifier trained on a fully compiled L1000FWD database subset. Virtual screening of approved drugs predicted the anthelmintic niclosamide as a top candidate (98.17% mimic probability). Explainable AI further rationalized this prediction by highlighting specific fragments within niclosamide’s salicylanilide core. Furthermore, a 200 ns MD simulation indicated favorable computational stability of niclosamide bound to the p62 (SQSTM1) ZZ domain. The complex showed rapid structural convergence (ligand RMSD plateauing at 1.65 nm) without dissociation, while maintaining strict receptor compactness (steady Radius of Gyration and solvent-accessible surface area) and a persistent interaction network of ~73 close atomic contacts. These findings suggest that niclosamide may function as a stable physical “lid” over the p62 ZZ domain, occluding its N-degron-binding cleft. Taken together, our computational framework highlights niclosamide as a promising candidate for AML drug repurposing, providing a hypothesis-generating foundation that warrants rigorous experimental validation.