Comprehensive in silico characterization of nonsynonymous SNPs in the human ezrin (EZR) gene and their role in disease pathogenesis

Abstract

Ezrin (EZR) is a crucial linker between the actin cytoskeleton and the plasma membrane. It interacts with proteins involved in cancer-related signaling pathways. To assess the impact of nonsynonymous single nucleotide polymorphisms (nsSNPs) on EZR structure and function, we employed bioinformatics tools (SIFT, PolyPhen-2, PROVEAN, PhD-SNP, SNPs&GO, SuSPect, and FATHMM) and identified deleterious variants. Stability analyses using MUpro, mCSM, I-Mutant 2.0, and DynaMut2 revealed six destabilizing nsSNPs (F240S, H288D, I248T, L59Q, L125S, and L225P). Structural modeling using HOPE, MutPred2, AlphaFold, Swiss-Model, and protein-protein docking using HADDOCK 2.4 assessed the impact on the EZR-EBP50 complex. Binding free energy calculations, salt bridge analysis, and interface residue mapping further confirmed that the L225P, F240S, and I248T mutations significantly impaired EZR-EBP50 interaction, potentially disrupting key signaling pathways. Molecular dynamics simulations indicated that mutant EZR proteins exhibited reduced stability, flexibility, and hydrogen bonding. This first comprehensive in silico analysis of EZR highlights pathogenic nsSNPs that may contribute to disease progression. These findings provide a foundation for experimental validation and may inform targeted therapies for EZR-related pathologies.