The catalytic domain of protein kinases exists in equilibrium of active and inactive states, in which in the aspartate-phenylalanine-glycine (DFG) motif (DFG-in/-out states) can undergo significant conformational changes. Limited structural data of the inactive, DFG-out conformations of kinases hampers the rational drug discovery and development of drugs targeting these conformations. We developed DFGmodel, a method that takes the sequence or the structure of a kinase in active conformation to generate models in DFG-out conformations. Models generated by DFGmodel accurately distinct known inhibitors from likely non-binders, suggesting that they are potentially useful for structure-based discovery of unique kinase modulators with optimal binding profiles. Finally, we applied DFGmodel to identify novel and potent inhibitors for a single kinase target (i.e. Dyrk1A), as well as to develop unique multi-kinase inhibitors which were confirmed in a Drosophila cancer model. Our computational approach provides structural framework for chemists and biologists to characterize kinases in the inactive states and explore new chemical spaces with structure-based drug design.

DFGmodel: Modeling protein kinases in inactive conformations and its applications in drug discovery

253rd National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.