An explorative study on potent Gram-negative specific LpxC inhibitors: CoMFA, CoMSIA, HQSAR and molecular docking

Pathogenic Gram-negative bacteria are responsible for nearly half of the serious human infections. Hologram quantitative structure-activity relationships (HQSAR), comparative molecular field analysis (CoMFA), and comparative molecular similarity index analysis (CoMSIA) were implemented on a group of 32 of potent Gram-negative LpxC inhibitors. The most effective HQSAR model was obtained using atoms, bonds, donor, and acceptor as fragment distinction. The cross-validated correlation coefficient (q²), non-cross-validated correlation coefficient (r²), and predictive correlation coefficient (r²_Pred) for test set of HQSAR model were 0.937, 0.993, and 0.892, respectively. The generated models were found to be statistically significant as the CoMFA model had (r² = 0.967, q² = 0.804, r²_Pred = 0.827); the CoMSIA model had (r² = 0.963, q² = 0.752, r²_Pred = 0.857). Molecular docking was employed to validate the results of the HQSAR, CoMFA, and CoMSIA models. Based on the obtained information, six new LpxC inhibitors have been designed.