QSAR Modeling of Antimycobacterial Activity and Activity Against Other Bacteria of 3‐Formyl Rifamycin SV Derivatives

Abstract

Rifamycins form a class of antibiotics with a specific potency as drug against tuberculosis via inhibition of the DNA-dependent RNA polymerase. In the present study, literature data on the antibacterial potency against Mycobacterium tuberculosis and other bacteria of 53 derivatives of 3-formyl rifamycin SV are subjected to a QSAR analysis using AM1-based quantum chemical descriptors and the recently introduced dynamic approach that allows explicit consideration of the conformational space of properly generated 3D structures. Data pre-treatment includes normalization of the minimum inhibition concentration (MIC) values of ordinary bacteria using the well-known antituberculosis drug rifampicine as reference compound (RIA), and averaging over the different strains as motivated by a mathematical analysis of the two-step inhibition process. For both this generalized potency against ordinary bacteria and the antimycobacterial activity, QSAR modelling yields 3-variable regression equations with acceptable statistics for screening purposes (r2 around 0.60), which however differ partly in the stepwise-selected molecular descriptors and the respective conformers. While both types of activity are increased by increasing HOMO energy reflecting an increased electron donor capability and tendency to undergo hydroquinone-semiquinone-quinone oxidation, the models differ in the best 2nd and 3rd local quantum chemical descriptors, pointing to partially conflicting electronic structure requirements for optimal antimycobacterial activity and generalized activity against other bacteria. The discussion includes a detailed mechanistic analysis of the underlying bioreactivity aspects.