Application of molecular docking and PSO-SVR intelligent approaches in antimalarial activity prediction of enantiomeric cycloguanil analogues.

A series of antifolate compounds, i.e. 1-(4-chlorophenyl)-6,6-dimethyl-1,3,5-triazine-2,4-diamine, or cycloguanil analogues, have shown effective inhibiting properties against Plasmodium falciparum dihydrofolate reductase (PfDHFR). In this work, the stereoselectivity of PfDHFR to the R and S enantiomer of cycloguanil analogues was obtained from molecular docking calculations and integrated into QSAR study to obtain a more accurate prediction model. Results indicate that PfDHFR can bind to cycloguanil analogues in the R and S enantiomers. Cycloguanil analogues with alkyl chain substituent prefer the R enantiomer over S because they do not experience steric hindrance with the Phe58 side chain, while cycloguanil analogues with phenol chain substituent prefer the S enantiomer over R because they do not experience steric hindrance with Leu46 and Met55 side chains. Particle swarm optimization and support vector regression were used to select relevant descriptors and generate the effective prediction model, with a high statistical significance level (r2training = 0.941; r2test = 0.884).

3D-quantitative structure-activity relationships of HEPT derivatives as HIV-1 reverse transcriptase inhibitors, based on Ab initio calculations.

Comparative molecular field analysis (CoMFA) has been applied to a large set of 1-[(2-hydroxyethoxy)methyl]-6-(phenylthio)thymine (HEPT) analogues. The starting geometry of HEPT was obtained from crystallographic data of HEPT/HIV-1 reverse transcriptase (RT) complexes. The structures of 101 HEPT derivatives were considered and fully optimized by ab initio molecular orbital calculations at the HF/3-21G level. The best CoMFA model is satisfactory in both statistical significance and predictive ability. It shows excellent, high predictive ability as r2cv = 0.858. The derived model indicates the importance of steric contributions (64.4%) as well as electrostatic interactions for the HIV-1 RT inhibition. In addition, steric and electrostatic contour maps from this analysis agree well with the experimentally observed trend that there are steric interactions between the side chain of HEPT and an aromatic ring of Tyr181. It is concluded that a moderately sized group at C5 enhances contact with Tyr181 enough to push it into a position which renders the protein nonfunctional, but a smaller group has insufficient steric requirements to do this and a larger group renders the ligand too large for the cavity. The mutation-induced resistance of reverse transcriptase is explained by this analysis. The obtained results not only lead to a better understanding of structural requirements of this set of compounds for the inhibition but also enable the suggestions for new and more potent drugs.

Conformational study of the HIV-1 reverse transcriptase inhibitor 1-[(2-hydroxyethoxy)methyl]-6-(phenylthio)thymine (HEPT).

The conformations of the HIV-1 reverse transcriptase inhibitor 1-[(2-hydroxyethoxy)methyl]-6-(phenylthio)thymine (HEPT) are calculated by semiempirical and mainly by ab initio methods in order to estimate the potential for the rotation around the carbon sulfur single bond. The results are compared to X-ray structures of HEPT associated to the HIV-1 reverse transcriptase. The NMR spectra of the compound are calculated to obtain some information about its structure in solution. The structure of HEPT in the complex is analysed to study the intermolecular interactions between the inhibitor and the surrounding protein, which determine the geometry of the inhibition complex.

Structure-activity correlation study of HIV-1 inhibitors: electronic and molecular parameters.

Quantitative structure-activity relationships (QSARs) for 40 HIV-1 inhibitors, 1-[(2-hydroxyethoxy)-methyl]-6-(phenylthio)thymine and its derivatives, were studied. Fully optimized geometries, based on the semiempirical AMl method, were used to calculate electronic and molecular properties of all compounds. In order to examine the relation between biological activities and structural properties, multiple linear regression models were employed. A suitable QSAR model was obtained, showing not only statistical significance, but also predictive ability. The significant molecular descriptors used were atomic charges of two substituted carbon atoms in the thymine ring, hydration energies and molar refractivities of the molecules. These descriptors allowed a physical explanation of electronic and molecular properties contributing to HIV-1 inhibitory potency.