Computer-aided drug design of capuramycin analogues as anti-tuberculosis antibiotics by 3D-QSAR and molecular docking
Yuanyuan Jin, Shuai Fan, Guangxin Lv, Haoyi Meng, Zhengyang Sun, Wei Jiang, Steven G. Van Lanen, Zhaoyong Yang · Open Chemistry · 2017
Abstract Capuramycin and a few semisynthetic derivatives have shown potential as anti-tuberculosis antibiotics.To understand their mechanism of action and structureactivity relationships a 3D-QSAR and molecular docking studies were performed. A set of 52 capuramycin derivatives for the training set and 13 for the validation set was used. A highly predictive MFA model was obtained with crossvalidated q2of 0.398, and non-cross validated partial least-squares (PLS) analysis showed a conventional r2of 0.976 and r2predof 0.839. The model has an excellent predictive ability. Combining the 3D-QSAR and molecular docking studies, a number of new capuramycin analogs with predicted improved activities were designed. Biological activity tests of one analog showed useful antibiotic activity againstMycobacterium smegmatisMC2 155 andMycobacterium tuberculosisH37Rv. Computer-aided molecular docking and 3D-QSAR can improve the design of new capuramycin antimycobacterial antibiotics.