Pharmacophore, atom-based 3D-QSAR modeling and binding mode analysis of QcrB inhibitors as potential tuberculosis agents

Abstract QcrB is an essential subunit of cytochrome-bc1 complex and is anticipated as a therapeutic target for tuberculosis. In recent years, significant attempts have been made to develop different chemical classes of QcrB inhibitors. Among them, a series of 31 phenoxyalkylimidazoles (PABs) showed anti-tuberculosis activity with MIC 90 values ranging from 0.10µM to 20µM were used to develop a pharmacophore and 3D-QSAR model. Five featured pharmacophore model, HHRRR, which consists of two hydrophobic regions (H) and three aromatic ring features (R), was chosen as the best-fitted model based on the highest survival score and molecular occupancy. The statistically significant 3D-QSAR model constructed using the HHRRR hypothesis possessed good predictive power with an excellent correlation coefficient (R 2 = 0.9859) and cross-validation coefficient (Q 2 = 0.8593). The contour map analysis provided crucial structural insights into the activity of active compounds. Furthermore, comparative binding mode analysis of a known clinical candidate, Q203, and the most active compound through induced fit docking approach revealed that these two compounds share a similar binding mode with the Q P site of QcrB and form hydrogen bonds with the critical residue T313. Substitution of different chemical scaffolds at the R position of PABs would lead to developing potential QcrB inhibitors.