Insight into the key features for ligand binding in Y1230 mutated c-Met kinase domain by molecular dynamics simulations

Discovery of Ubiquitin-Specific Protease 7 (USP7) Inhibitors with Novel Scaffold Structures by Virtual Screening, Molecular Dynamics Simulation, and Biological Evaluation

USP7 has been regarded as a potential therapeutic target for cancer. In this study, virtual screening, molecular dynamics (MD) simulation, and biological evaluation have been applied for the discovery of novel USP7 inhibitors targeting the catalytic active site. Among the obtained compounds, compound 12 with a novel scaffold structure exhibited certain USP7 inhibitory activity (Ub-AMC assay IC₅₀ = 18.40 ± 1.75 μM, Ub-Rho assay IC₅₀ = 7.75 μM). The binding affinity between USP7_CD (USP7 catalytic domain) and this hit compound was confirmed with a K_D value of 4.46 ± 0.86 μM. Preliminary in vitro studies disclosed its antiproliferative activity on human prostate cancer cell line LNCaP with an IC₅₀ value of 15.43 ± 3.49 μM. MD simulation revealed the detailed differences of protein-ligand interactions between USP7_CD and the ligands, including the reference compound ALM4 and compound 12, providing some important information for improving the bioactivity of 12. This hit compound will serve as a promising starting point for facilitating the further discovery of novel USP7 inhibitors.

Structural insights into ligand binding features of dual FABP4/5 inhibitors by molecular dynamics simulations

The fatty acid binding protein (FABP) 4 and 5 have been considered as potential targets for the treatment of metabolic diseases. A compensatory upregulation of FABP5 due to the gene ablation of FABP4 in adipocytes indicated the importance of dual FABP4/5 inhibitors. A few compounds have been discovered as dual FABP4/5 inhibitors. However, none exhibited equivalent inhibitory activity against both FABP4 and FABP5, and almost all compounds showed weaker inhibition against FABP5. To provide a better structural understanding for the design of potent dual FABP4/5 inhibitors, molecular dynamics simulations have been performed for 100 ns to disclose the ligand binding features in FABP4 and FABP5 using Amber14, respectively. Key residues were identified by analysis of close contact, hydrogen bond occupancy, binding free energy and alanine scanning mutagenesis. In addition, induced-fit effects have been observed upon ligand binding in the process of simulations. The shifted alkyl chain of ligand in FABP4 was significantly different from that in FABP5 due to the corresponding residues (Phe58^FABP4 and Leu60^FABP5). Thus, to avoid different steric effects made by these two residues, hydrophobic groups of suitable size should be taken into account. Besides, electrostatic and steric effects with Arg107^FABP4 and Arg109^FABP5 should be paid more attention to. The results will facilitate the rational design of dual FABP4/5 inhibitors.

Probing the "fingers" domain binding pocket of Hepatitis C virus NS5B RdRp and D559G resistance mutation via molecular docking, molecular dynamics simulation and binding free energy calculations

The NS5B RdRp polymerase is a prominent enzyme for the replication of Hepatitis C virus (HCV). During the HCV replication, the template RNA binding takes place in the "fingers" sub-domain of NS5B. The "fingers" domain is a new emerging allosteric site for the HCV drug development. The inhibitors of the "fingers" sub-domain adopt a new antiviral mechanism called RNA intervention. The details of essential amino acid residues, binding mode of the ligand, and the active site intermolecular interactions of RNA intervention reflect that this mechanism is ambiguous in the experimental study. To elucidate these details, we performed molecular docking analysis of the fingers domain inhibitor quercetagetin (QGN) with NS5B polymerase. The detailed analysis of QGN-NS5B intermolecular interactions was carried out and found that QGN interacts with the binding pocket amino acid residues Ala97, Ala140, Ile160, Phe162, Gly283, Gly557, and Asp559; and also forms π⋯π stacking interaction with Phe162 and hydrogen bonding interaction with Gly283. These are found to be the essential interactions for the RNA intervention mechanism. Among the strong hydrogen bonding interactions, the QGN⋯Ala140 is a newly identified important hydrogen bonding interaction by the present work and this interaction was not resolved by the previously reported crystal structure. Since D559G mutation at the fingers domain was reported for reducing the inhibition percentage of QGN to sevenfold, we carried out molecular dynamics (MD) simulation for wild and D559G mutated complexes to study the stability of protein conformation and intermolecular interactions. At the end of 50 ns MD simulation, the π⋯π stacking interaction of Phe162 with QGN found in the wild-type complex is altered into T-shaped π stacking interaction, which reduces the inhibition strength. The origin of the D559G resistance mutation was studied using combined MD simulation, binding free energy calculations and principal component analysis. The results were compared with the wild-type complex. The mutation D559G reduces the binding affinity of the QGN molecule to the fingers domain. The free energy decomposition analysis of each residue of wild-type and mutated complexes revealed that the loss of non-polar energy contribution is the origin of the resistance. Communicated by Ramaswamy H. Sarma.

Molecular dynamics simulation and binding free energy studies of novel leads belonging to the benzofuran class inhibitors of Mycobacterium tuberculosis Polyketide Synthase 13

In this work, the binding mechanism of new Polyketide Synthase 13 (Pks13) inhibitors has been studied through molecular dynamics simulation and free energy calculations. The drug Tam1 and its analogs, belonging to the benzofuran class, were submitted to 100 ns simulations, and according to the results obtained for root mean square deviation, all the simulations converged from approximately 30 ns. For the analysis of backbone flotation, the root mean square fluctuations were plotted for the Cα atoms; analysis revealed that the greatest fluctuation occurred in the residues that are part of the protein lid domain. The binding free energy value (ΔG_bind) obtained for the Tam16 lead molecule was of -51.43 kcal/mol. When comparing this result with the ΔG_bind values for the remaining analogs, the drug Tam16 was found to be the highest ranked: this result is in agreement with the experimental results obtained by Aggarwal and collaborators, where it was verified that the IC₅₀ for Tam16 is the smallest necessary to inhibit the Pks13 (IC₅₀ = 0.19 μM). The energy decomposition analysis suggested that the residues which most interact with inhibitors are: Ser1636, Tyr1637, Asn1640, Ala1667, Phe1670, and Tyr1674, from which the greatest energy contribution to Phe1670 was particularly notable. For the lead molecule Tam16, a hydrogen bond with the hydroxyl of the phenol not observed in the other analogs induced a more stable molecular structure. Aggarwal and colleagues reported this hydrogen bonding as being responsible for the stability of the molecule, optimizing its physic-chemical, toxicological, and pharmacokinetic properties.

Identification of the hot spot residues for pyridine derivative inhibitor CCT251455 and ATP substrate binding on monopolar spindle 1 (MPS1) kinase by molecular dynamic simulation

Protein kinase monopolar spindle 1 plays an important role in spindle assembly checkpoint at the onset of mitosis. Over expression of MPS1 correlated with a wide range of human tumors makes it an attractive target for finding an effective and specific inhibitor. In this work, we performed molecular dynamics simulations of protein MPS1 itself as well as protein bound systems with the inhibitor and natural substrate based on crystal structures. The reported orally bioavailable 1 h-pyrrolo [3,2-c] pyridine inhibitors of MPS1 maintained stable binding in the catalytic site, while natural substrate ATP could not stay. Comparative study of stability and flexibility of three systems reveals position shifting of β-sheet region within the catalytic site, which indicates inhibition mechanism was through stabilizing the β-sheet region. Binding free energies calculated with MM-GB/PBSA method shows different binding affinity for inhibitor and ATP. Finally, interactions between protein and inhibitor during molecular dynamic simulations were measured and counted. Residue Gly605 and Leu654 were suggested as important hot spots for stable binding of inhibitor by molecular dynamic simulation. Our results reveal an important position shifting within catalytic site for non-inhibited proteins. Together with hot spots found by molecular dynamic simulation, the results provide important information of inhibition mechanism and will be referenced for designing novel inhibitors.