Insights into pathological mutations in insulin-like growth factor I through in silico screening and molecular dynamics simulation

Abstract

Insulin-like growth factor I (IGF-I) is an anabolic growth hormone indispensable for cell growth, proliferation, differentiation, and other metabolic processes. Three missense mutations in IGF-I have been identified to be disease-related, while more mutations are waiting for phenotype annotation. However, there is no previous work regarding effective and accurate identification of pathological mutations of IGF-I, neither regarding the effects of mutations on the protein structure and dynamics. In this study, we first predicted potential deleterious mutations present in IGF-I using 16 in silico tools. Then, these mutations were further evaluated through multiple bioinformatics methods including conservation analysis, physicochemical characterization, and molecular dynamics simulation. After rigorous screening, five mutations (T4M, V17M, V44M, R50W, and M59R) were finally selected, of which two have been previously reported to be deleterious. These mutations locate at conserved regions and change the residue size locally. In the conventional simulations, the mutations destabilized the overall IGF-I structure by destroying two important hydrogen bonds within the key region of "C-neck." This finding was further confirmed by the thermal unfolding simulations and the free-energy calculations, where the mutants were associated with faster and greater loss of helix and lower energy barriers in comparison with the wild-type protein. The rigorous phenotype prediction and comprehensive structural analysis of missense mutations will not only pave the way of screening for harmful mutations in IGF-I but also provide new prospects for the rational design of IGF-I analogues and tailored medicine.

Keywords: Free-energy calculation; Hydrogen bond network; Phenotype prediction; Structural analysis; Structural stability.