Comparative structural analysis of α-glucosidase inhibitors on difference species: a computational study.

Structural feature analysis of chlorogenic acid derivatives made up of varying lengths of alkyl groups as α-glucosidases inhibitors were performed by QSAR techniques. The statistically significant models derived from the study were validated by leave one out, Y-randomization and test set methods. The predictive capacity of the models was assessed by its validation parameters such as crossvalidated correlation coefficients (Q(2)), predictive residual analysis and other correlation parameters. The results obtained from the study show that the models were constructed with vsurf like properties (vsurf_ID4, vsurf_ID7 and vsurf_CW8), partial charge (Q_VSA_FNEG) and conformation dependent charged (dipoleX) descriptors. The integy moments of hydrophobicity descriptors (ID4 and ID7) are contributed for the inhibitory activity of the α-glucosidases enzymes of both the species. The vsurf_ID7 descriptor has contributed significantly (negatively) for the inhibitory activity prediction of α-glucosidases enzymes of S. cerevisiae. The partial negative charge on the surface of the molecules is detrimental for the activity, which reveals that the active site of the enzymes may have negatively charged groups. The pharmacophore analysis results also confirm the presence of hydrophilic properties on the vdW surface of the molecules. These results explain that the active sites of α-glucosidase enzymes of both the species have the same environment for the interaction. The alkyl side chain on the molecules is important for the pharmacokinetic behavior of the molecules and reduces the interaction energy of the molecules with the water. Hence, these results will be useful for designing novel molecules with multiple activities.