Development and application of a 2D descriptor model to predict the interaction between drugs and cyclodextrins.
cyclodextrin, beta-cyclodextrin, QSPR, FEP, inclusion complexes.
Low aqueous solubility is a problem of many molecules of pharmaceutical use and the complexation with cyclodextrins is a widely used strategy to enhance its dissolution and, consequently its bioavailability in biological environment. It is possible to predict the strength of the bonds between the host and guest molecules of the system by the phase solubility diagram method. Experimental techniques are expensive and time consuming, causing a growth in computational chemistry area. QSPR (quantitative structure-property relationship) uses descriptors derived from computational calculations as molecular parameters, being a good method to predict the binding affinity of the inclusion complex. The free-energy perturbation (FEP) is a methodology capable to calculate this affinity under the prediction of different conformations, with the ligand in many freedom degrees, with the ligands in the explicit solvent or inside the system. The main purpose of this project was to achieve a in silico methodology to enhance the experimental manipulation of the systems. To that, the constants obtained by both methodologies, QSPR and FEP were evaluated by its accuracy of prediction of the free-energy of the system. The best prediction model was QSPR-2D, having R2 equal to 0.924 to the training set e Q2 equal to 0.863 to test set. The model was achieved with eight descriptors that delineated the importance of parameters such as partition coefficient and size of the molecules to the free-energy calculations. Subsequently, the compounds in DrugBank were submitted to the calculation by the parameters of the model achieved to understand the applicability of the methodology. The next step consisted in the identification of promising drug to the phase solubility diagram essay with ƒÒ-cyclodextrin. Ciprofibrate was the chosen drug to the essay for being innovative in literature and have adequate properties to complexation with this cyclodextrin. To validate the obtained model, an experimental study will be executed to compare the free-energy values obtained in silico and the experimental ones with the compound selected.