Structure-based virtual screening of influenza virus RNA polymerase inhibitors from natural compounds: Molecular dynamics simulation and MM-GBSA calculation

Comput Biol Chem. 2020 Apr:85:107241. doi: 10.1016/j.compbiolchem.2020.107241. Epub 2020 Feb 26.

Abstract

The resistances of matrix protein 2 (M2) protein inhibitors and neuraminidase inhibitors for influenza virus have attracted much attention and there is an urgent need for new drug. The antiviral drugs that selectively act on RNA polymerase are less prone to resistance and possess fewer side effects on the patient. Therefore, there is increased interest in screening compounds that can inhibit influenza virus RNA polymerase. Three natural compounds were found by using molecular docking-based virtual screening, which could bind tightly within the polymerase acidic protein-polymerase basic protein 1 (PA-PB1) subunit of influenza virus polymerase. Firstly, their drug likeness properties were evaluated, which showed that the hepatotoxicity values of all the three compounds indicating they had less or no hepatotoxicity, and did not have the plasma protein biding (PPB) ability, the three compounds needed to be modified in some aspects, like bulky molecular size. The stability of the complexes of PA-hits was validated through molecular dynamics (MD) simulation, revealing compound 2 could form more stable complex with PA subunit. The torsional conformations of each rotatable bond of the ligands in PA subunit were also monitored, to investigate variation in the ligand properties during the simulation, compound 3 had fewer rotatable bonds, indicating that the molecule had stronger rigidity. The bar charts of protein-ligand contacts and contacts over the course of trajectory showed that four key residues (Glu623, Lys643, Asn703 and Trp706) of PA subunit that participated in hydrogen-bond, water bridge and hydrophobic interactions with the hit compounds. Finally, the binding free energy and contributed energies were calculated by using MM-GBSA method. Out of the three compounds, compound 1 showed the lowest total binding free energy. Among all the interactions, the contribution of the covalent binding and the van der Waals energy were more than other items, compound 1 formed more stable hydrogen bonds with the residues of PA subunit binding pocket. This study smoothed the path for the development of novel lead compounds with improved binding properties, high drug likeness, and low toxicity to humans for the treatment of influenza, which provided a good basis for further research on novel and effective influenza virus PA-PB1 interaction inhibitors.

Keywords: Drug likeness; Influenza virus; Molecular dynamics simulations; Protein-protein interaction inhibitor; Virtual screening.

MeSH terms

  • Antiviral Agents / chemistry
  • Antiviral Agents / pharmacology*
  • Biological Products / chemistry
  • Biological Products / pharmacology*
  • Computational Biology
  • DNA-Directed RNA Polymerases / antagonists & inhibitors*
  • DNA-Directed RNA Polymerases / metabolism
  • Drug Evaluation, Preclinical
  • Enzyme Inhibitors / chemistry
  • Enzyme Inhibitors / pharmacology*
  • Humans
  • Influenza A virus / drug effects*
  • Influenza A virus / enzymology
  • Microbial Sensitivity Tests
  • Molecular Docking Simulation
  • Molecular Dynamics Simulation
  • Molecular Structure
  • Thermodynamics*

Substances

  • Antiviral Agents
  • Biological Products
  • Enzyme Inhibitors
  • DNA-Directed RNA Polymerases