Computational Studies on Tetrahydropyrimidine-2-one HIV-1 Protease Inhibitors:  Improving Three-Dimensional Quantitative Structure−Activity Relationship Comparative Molecular Field Analysis Models by Inclusion of Calculated Inhibitor- and Receptor-Based Properties

A computational chemistry study has been performed on a series of tetrahydropyrimidine-2-ones (THPs) as HIV-1 protease (HIV-1 PR) inhibitors. The present investigation focuses on the correlation of inhibitor−enzyme complexation energies (E compl), inhibitor solvation energies E solv[I], and both polar and nonpolar buried surface areas (BSAs) with the observed values of the binding affinity (pK I). Various combinations of these specific inhibitor- and receptor-based properties were also evaluated as additional descriptors to three-dimensional quantitative structure−activity relationship (3D-QSAR) models constructed using comparative molecular field analysis (CoMFA). Linear regression of the observed pK I values with E compl, E solv[I], and the BSAs yielded a strong correlation in terms of both self-consistency (r 2 ≈ 0.90) and internal predictive ability (r cv 2 > 0.50). The 3D-QSAR models obtained from CoMFA using standard partial least-squares (PLS) analysis also yielded a strong correlation between the CoMFA fields and the experimental pK i (r 2 = 0.96; r cv 2 = 0.58). Various “enhanced” 3D-QSAR models were constructed in which different combinations of the E compl, E solv[I], and BSAs were added as additional descriptors to the default steric−electrostatic CoMFA fields. Inclusion of E solv[I] in particular yielded significant improvement in the predictive ability (r cv 2 ≈ 0.80) of the resultant 3D-QSAR model.