Gaining insight into crizotinib resistance mechanisms caused by L2026M and G2032R mutations in ROS1 via molecular dynamics simulations and free-energy calculations

J Mol Model. 2017 Apr;23(4):141. doi: 10.1007/s00894-017-3314-z. Epub 2017 Mar 30.

Abstract

ROS1 fusion kinase-highly expressed in a variety of human cancers-has emerged as an important and attractive target for anticancer drug design. Crizotinib, a well-known drug approved by the FDA as an ALK inhibitor to treat advanced NSCLC, also shows potent inhibitoy activity against ROS1. However, the development of serious resistance due to secondary mutations has been observed in clinical studies. To provide insight into the mechanisms of this drug resistance, molecular dynamics simulations and free-energy calculations were carried out for complexes of crizotinib with wild-type (WT) ROS1 as well as the mutated L2026M and G2032R forms. MD simulations indicated that the L2026M and G2032R systems are slightly less flexible than the WT system. Binding free energy calculations showed that the L2026M and G2032R mutations significantly reduce the binding affinity of crizotinib for ROS1, and that the resistance to crizotinib caused by the L2026M and G2032R mutations arises mostly from increases in entropic terms. Furthermore, calculations of per-residue binding free energies highlighted increased and decreased contributions of some residues in the L2026M and G2032R systems relative to those in the WT system. The present study therefore yielded detailed insight into the mechanisms of resistance to crizotinib caused by the L2026M and G2032R mutations, which should provide the basis for rational drug design to combat crizotinib resistance. Graphical Abstract Superposition of the average structures obtained from the last 10 ns of the molecular dynamics simulation trajectoriy for WT (green) and mutated ROS1 (cyan).

Keywords: Binding free-energy calculation; Crizotinib; Drug resistance; Molecular dynamics (MD) simulation; ROS1.