Integrated computational-experimental discovery of clathrin inhibitors: from virtual screening to mechanistic validation

Abstract

Virtual screening has emerged as a powerful strategy to identify novel clathrin inhibitor compounds, targeting clathrin-mediated endocytosis (CME). In this study, a multi-step computational approach was employed, integrating molecular docking, prime/MM-GBSA simulations, molecular dynamics (MD) simulations, alanine scanning mutagenesis, quantum mechanics/molecular mechanics (QM/MM) calculations, dynamic cross-correlation matrix (DCCM) analysis and principal component analysis (PCA). A diverse chemical library was screened against the clathrin terminal domain, a critical hub for protein–protein interactions in vesicle formation. Top-ranking compounds with lower binding energy were prioritized to assess their binding affinity with the clathrin N-terminal domain (NTD). Experimental validation of selected hits revealed two compounds (19 and 20) exhibiting better binding affinities to the clathrin NTD with K_D values of 1.36 × 10⁻⁵ and 8.22 × 10⁻⁶ M. The two compounds demonstrated minimal cytotoxicity and inhibitory activities on CME. This work underscores the efficacy of virtual screening in discovering clathrin inhibitors and provides a foundation for developing therapeutics to modulate CME-related pathologies.