Mapping the Structural Drivers of Insulin Recognition and Specificity Using Molecular Dynamics and Free Energy Calculations

Abstract

AbstractInsulin has been the cornerstone of diabetes treatment since its discovery over a century ago. Despite this, several aspects of insulin engagement with its target and off-target receptors remain unknown, thus limiting the use of structure-guided design of novel analogs. A recent spurt in structural information for the insulin and insulin-like growth factor 1 receptors (IR and IGF-1R), were leveraged in this study to gain a deeper, molecular-level understanding of ligand recognition at these targets. Molecular dynamics (MD) and free energy perturbation (FEP) were utilized to map the drivers of potency and specificity of several insulin variants including disease-linked mutations. The remarkable accuracy of MD-FEP in capturing functional trends (>80% of cases) across various insulin analogs underscored the method’s potential. The ability to deconvolute observations into receptor contributions, conformational perturbations and (or) solvent-network changes lays the foundations for its use in predictive design of future insulins and other therapeutic peptides.