Quantum Descriptors for Predicting and Understanding the Structure-Activity Relationships of Covalent Warheads

Predictive modeling and understanding chemical warhead reactivities have the potential to accelerate targeted covalent drug discovery. Here we used density functional theory calculations to reexamine the ground-state quantum descriptors and important structure-activity (SAR) relationships of thiol-Michael additions. The calculated reaction profiles of the singly- and doubly-activated olefins including several model warheads related to afatinib revealed strong correlations between the reaction barriers and electronic properties related to the carbanion intermediate. Among them the change in the Cb charge upon carbanion formation is the strongest reactivity predictor, while the least expensive reactant-only properties, the electrophilicity index and the carbon charge also show strong rank correlations, suggesting their utility as quantum descriptors for building machine learning models. We also investigated the reactivity enhancement effect of the  beta-dimethylaminomethyl (DMAM) substitution which led to several FDA-approved covalent drugs. Our data revealed that the  -DMAM substitution is protonated at neutral pH and it lowers the reaction barrier through inducing the charge accumulation at the C  upon carbanion formation; by contrast, the inductive effect of the beta-trimethylaminomethyl substitution is diminished due to steric hindrance. The data also suggested that the beta-DMAM substitution downshifts the thiol pKa and thereby activates EGFR’s front pocket cysteine C797 for nucleophilic attack. Taken together, our study contributes novel insights for large-scale modeling and understanding the SARs of covalent warheads.