Multiple Molecular Dynamics Simulations and Energy Analysis Unravel the Dynamic Properties and Binding Mechanism of Mutants HIV-1 Protease with DRV and CA-p2

Abstract

PR^S17, a variant of human immunodeficiency virus type I protease (HIV-1 PR), has 17 mutated residues showing high levels of multidrug resistance. To describe the effects of these mutated residues on the dynamic properties and the binding mechanism of PR with substrate and inhibitor, focused on six systems (two complexes of WT PR and PR^S17 with inhibitor Darunavir (DRV), two complexes of WT PR and PR^S17 with substrate analogue CA-p2, two unligand WT PR and PR^S17), we performed multiple molecular dynamics (MD) simulations combined with MM-PBSA and solvated interaction energy (SIE) methods. For both the unligand PRs and ligand-PR complexes, the results from simulations revealed 17 mutated residues alter the flap-flap distance, the distance from flap regions to catalytic sites, and the curling degree of the flap tips. These mutated residues changed the flexibility of the flap region in PR, and thus affected its binding energy with DRV and CA-p2, resulting in differences in sensitivity. Hydrophobic cavity makes an important contribution to the binding of PR and ligands. And most noticeable of all, the binding of the guanidine group in CA-p2 and Arg8' of PR^S17 is useful for increasing their binding ability. These results have important guidance for the further design of drugs against multidrug resistant PR. IMPORTANCE Developing effective anti-HIV inhibitors is the current requirement to cope with the emergence of the resistance of mutants. Compared with the experiments, MD simulations along with energy calculations help reduce the time and cost of designing new inhibitors. Based on our simulation results, we propose two factors that may help design effective inhibitors against HIV-1 PR: (i) importance of hydrophobic cavity, and (ii) introduction of polar groups similar to the guanidine group.

Keywords: HIV-1 PR; MD simulation; MM-PBSA analyses; drug resistance; solvated interaction energy analyses.

FIG 1

The structure of HIV-1 PR bound to the ligand CA-p2 (PDB ID: 6O48, A) and the ligands (B. CA-p2, C. DRV) used in this study.

FIG 2

Histogram distributions of Ile50-Ile50’ distance in unligand PR and ligand bound PR complexes.

FIG 3

Histogram distributions of Asp25-Ile50 and Asp25’-Ile50’ distance in unligand PR and ligand bound PR complex.

FIG 4

Histogram distributions for the TriCα angles (G48-G49-I50) and (G48’-G49’-I50’) in unligand PR and ligand bound PR complex.

FIG 5

Free energy decomposition analysis for ligands with the mutant PR and the WT PR; residues contributing significantly are labled.

FIG 6

The difference of the nonpolar interaction (ΔG_vdW + ΔG_SA) for two ligands with the mutant PR and the WT PR; residues contributing significantly are labled.

FIG 7

The difference of the polar interaction (ΔG_ele + ΔG_GB) for two ligands with the mutant PR and the WT PR; residues contributing significantly are labled.

FIG 8

The represented MD structures of the ligands bound WT complexes and ligands bound mutant complexes: (A) the DRV bound PR; (B) the DRV bound PR^S17; (C) the CA-P2 bound PR, (D) the CA-P2 bound PR^S17. The represented structures extracted from the MD trajectories were used. The ligands are shown as ball-and-stick and the important residues are shown as sticks.