Drug Design Of HIV Integrase Inhibitors And Drug Resistance Mechanism Of HIV Integrase Mutants

Acquired immunodeficiency syndrome (AIDS) is caused by human immunodeficiency virus (HIV). This disease results in heavily damage of human immune system, then makes people lost the resist to all kinds of pathogens, and finally lead to death. Nowadays, AIDS remains an irremediable disease. An essential step in HIV replication is the integration of the viral cDNA into host chromosomal DNA by integrase (IN), IN has no sequence homologue in the human host, therefore, it is considered as a pivotal and validated drug target. Recently, new drug development and inhibitor modification targeting IN become the focus of the medical research. Besides, drug resistance against most of the anti-HIV drugs has appeared. The drug resistant mutations in HIV are major impediment to successful highly active antiretroviral therapy (HAART) and new drug design. Understanding the mechanism of HIV drug resistance in terms of structure and function will be useful for the rational inhibitor modification and drug design targeting IN, whereas, the mechanism of drug resistance caused by diketoacids (DKAs), the most potent IN inhibitors is still unknown.A pharmacophore model could be used for database searching and new structural entities revealing. There are several classes of DKAs structures have been reported, simultaneously, the catalytic domain structure of IN with 5-CITEP determined by X-ray crystallography is available. Hence, both ligand based and receptor based pharmacophore models are developed in this study, no similar studies have been reported. Firstly, we developed a three-dimensional pharmacophore model for HIV IN from DKAs inhibitors, inhibitor conformations mapped into the pharmacophore model were superimposed with their docking conformations. Corresponding positions between the pharmacophore model and IN residues were thus obtained. The pharmacophore model was refined according to whether the pharmacophore features were compatible with residues around them. An optimal pharmacophore model was generated and consisted of 1 hydrophobic feature, 3 hydrogen pair features and 1 hydrogen-bond donor feature. This pharmacophore model based on DKAs had high reliability with a goodness of hit (GH) score of 0.56, a high percentage yield of actives (Y) of 63.6.1% and a lower false positive rate (FP) of 0.41%. The pharmacophore model had high reliability. Additionally, a receptor-based pharmacophore model was generated according to the binding mode between IN and 6 classes of DKAs. We attempted to refine the obtained pharmacophore model according to the corresponding residues of IN around pharmacophore features. Finally, the key interactions and the common pharmacophore features were obtained by comparing the pharmacophore models based on ligands and acceptor. The final pharmacophore model can contribute to the discovery and design of new IN inhibitors.A QSAR (Quantitative structure-activity relationship) analysis has been done to gain the relationship between structure and biological activity of drug molecules. The model obtained from QSAR analysis can be used to predict the binding mode and the biological activity of new inhibitors, and the model can also be used to guide the new drug design and lead compound optimization. There are some studies on QSAR analysis of IN inhibitors, whereas, we found no QSAR model of IN DKAs inhibitors was built. DKAs are the most potential IN inhibitors, QSAR studies on DKAs will benefit the understanding of interaction between DKAs and IN, and will benefit the modification of IN inhibitors. In this paper, QSAR analysis combined with pharmacophore model construction and molecular docking have been done against IN DKAs inhibitors. Firstly, a pharmacophore model was built. The structure alignment for all the inhibitors were performed based on the pharmacophore; for a comparison, the common substructure-based alignment was performed in the structure alignment of inhibitors. As a result, a optimal CoMSIA model gains a conventional correlation coefficient r2 of 0.955 and a leave-one-out cross-validated coefficient q2 of 0.665, and the model also can predict the activities of the test DKAs well (r2 = 0.559). In order to validate the rationality of the optimal CoMSIA model, the CoMSIA contours were mapped into the binding mode of IN and DKAs, and the rationality of the substitutions suggested by QSAR model were investigated according to the IN residues around these substitutions. Reasonably modification rules were suggested based on this QSAR study.Drug resistance against most of the anti-HIV drugs has appeared. The drug resistant mutations in HIV are major impediment to successful highly active antiretroviral therapy (HAART) and new drug design. In order to make drugs take full effects, understanding the mechanism of HIV drug resistance in terms of structure and function became a attractive issue. Previous studies on drug resistant IN mutants are all from comparing 1 drug-resistant IN strain with wild-type IN, whether the mechanism mutually existed for multiple drug-resistant strains remains unclear. In order to understand the drug resistance mechanism mutually existed for multiple drug-resistant strains caused by the most potent IN inhibitors of DKAs, 3 S-1360-resistant HIV strains were selected and molecular docking and molecular dynamics simulations were performed to illustrate the mechanism of drug resistance against DKAs inhibitors. The results showed that: after Thr66 was mutated to Ile66 in the three mutants, the long side chain of Ile66 occupied the center of IN active pocket, then prevented the inhibitor S-1360 from moving into depth of the active pocket, thus, the binding site of S-1360 is close to loop 3 region whereas far from the residues K156 and K159 which are crucial for virus DNA binding; Additionally, it was found from secondary structural analyses that the length of loop 3 and helix 1 regions are longer in the three mutant complexes in comparison to the wild type IN complex. Also, there are hydrogen bonds between S-1360 and E152 in helix 1 region, the hydrogen bond depressed the flexibility of the region, whereas, in the three mutant complexes, there is no hydrogen bond between S-1360 and the residues in helix 1 and loop 3 regions of IN, therefore, the regions show higher flexibility in the three mutant complexes. All above results contribute to drug resistance. Altogether, the drug resistance lies in the different binding sites of the inhibitor and the conformation changes of loop 3 and helix 1 regions. The mechanism mutually existed for multiple drug-resistant strains will be useful for improving the inhibition potency of DKA inhibitors against drug-resistant IN, and rational inhibitor modification and design targeting IN.