3D-QSAR, molecular docking, and molecular dynamics simulation of a novel thieno[3,4-d]pyrimidine inhibitor targeting human immunodeficiency virus type 1 reverse transcriptase

Human immunodeficiency virus type 1 reverse transcriptase (HIV-1 RT) is one of the most attractive drug targets for the treatment of AIDS. In this study, 67 thieno[3,4-d]pyrimidine derivatives were selected as novel HIV-1 RT inhibitors to combat viral resistance, and were subjected to 3 D-QSAR studies using CoMFA, CoMSIA, and T-CoMFA. In the 3 D-QSAR study, two methods of ligand-based alignment and pharmacophore-based alignment were used. The results showed that CoMFA (n = 8; q(2) = 0.594; r(2) = 0.974) and CoMSIA (n = 7; q(2) = 0.528; r(2) = 0.965) have good stability and predictability. The molecular docking study showed that the hydrogen bonding and van der Waals interactions of key residues such as Leu100, Lys101, Val106, Phe227 and Pro236 play an important role in ligand-receptor binding. Based on these results, 12 new thieno[3,4-d]pyrimidines were designed and their activities were predicted; the results indicated that these compounds have good predictive activity and reasonably good ADME/T profiles. MD simulation analysis of 50 ns showed that compound 23j formed four hydrogen bonds with the residues (Lys101, Lys104, Val106 and Thr318), and binds more closely to HIV-1 RT than compound 23j. Furthermore, the group at the R-1 position and the horseshoe-like conformation of these compounds are critical for the inhibitory activity and stability. These results provide useful insights for the discovery and design of a new generation of HIV-1 RT inhibitors. Communicated by Ramaswamy H. Sarma