Modeling the aromatase inhibitor activity of indole-imidazole derivatives: Quantitative structure activity relationship and molecular docking

In the present study, 19 compounds of indole-imidazole derivatives were studied to obtain the structure requirements to inhibit the active sites of the aromatase enzyme. 2D quantitative structure-activity relationships (QSARs) were analyzed using the partial least squares (PLS) method. To build the QSAR model, the data set was randomly split into a training set (15 compounds) and a test set (4 compounds) for the external validation of the model. As a result, a model with three descriptors (diameter, Petitjean, Q_VSA_FPNEG) was found to be robust enough to predict the aromatase inhibitor activity of the indole-imidazole derivatives, with an R2 of 0.892 and Q2 of 0.741. A series of 57 new compounds was modeled and designed; Of these, only 18 compounds were found to have biological activity greater than that of letrozole (the reference compound). These compounds were docked to the active site of aromatase to understand their inhibitory action and their binding energy toward the aromatase enzyme. Analysis of energy of the 18 compound-aromatase complexes revealed that compound 48 has a low binding energy (strong binding affinity) to aromatase as compared to letrozole; the energy of this compound is less by 7 units than that of letrozole. This compound is enhanced by an electron-withdrawing group (COOH) at the meta position of the phenyl ring of indole.