Employing Steered MD Simulations for Effective Virtual Screening: Active Pharmacophore Search by Dynamic Corrections to target MKK3-MYC Interactions

Abstract The intricate relationship between mitogen-activated protein kinase 3 (MKK3) and MYC proto-oncogene protein (MYC) activation holds deep implications for the progression of cancer, particularly in the context of triple negative breast cancer (TNBC). Despite significant progress, the challenge of discovering effective MYC-targeted drugs persists, demanding innovative approaches to control MYC-dependent malignancies. A promising avenue in this pursuit involves disrupting the protein-protein interactions (PPIs) between MKK3 and MYC. The significance of this interaction is emphasized by the activation of MYC by MKK3 in diverse cell types, presenting a novel perspective for therapeutic interventions in MYC-driven pathways. In the current study, a novel in silico strategy to screen small molecule libraries that target the MKK3-MYC interaction was conducted. Dynamic structure-based pharmacophore models were developed and utilized for screening the small molecule libraries, enabling the identification of compounds exhibiting favorable alignment with the defined pharmacophore features. Subsequently, physics-based simulations approaches were conducted on these selected hit molecules. Steered molecular dynamics (sMD) simulations were utilized to assess the correlation between the necessary forces to dissociate candidate hit ligands from the binding pocket and their corresponding average binding free energies (MM/GBSA). Comparative analysis of the average binding free energies of the identified hits obtained from the small molecule libraries represent that the identified compounds have promising predicted binding affinities compared to the reference molecule SGI-1027. Therefore, these findings may signify a crucial advancement in our ability to control MYC activation in cancer.