Identification of FDA-approved drugs against SARS-CoV-2 RNA-dependent RNA polymerase (RdRp) through computational virtual screening

Abstract

The SARS-CoV-2 coronavirus is responsible for the COVID-19 outbreak, which overwhelmed millions of people worldwide; hence, there is an urgency to identify appropriate antiviral drugs. This study focuses on screening compounds that inhibit RNA-dependent RNA-polymerase (RdRp) essential for RNA synthesis required for replication of positive-strand RNA viruses. Computational screening against RdRp using Food and Drug Administration (FDA)-approved drugs identified ten prominent compounds with binding energies of more than - 10.00 kcal/mol, each a potential inhibitor of RdRp. These compounds' binding energy is comparable to known RdRp inhibitors remdesivir (IC50 = 10.09 μM, SI = 4.96) and molnupiravir (EC50 = 0.67 - 2.66 µM) and 0.32-2.03 µM). Remdesivir and molnupiravir have been tested in clinical trial and remain authorized for emergency use in the treatment of COVID-19. In docking simulations, selected compounds are bound to the substrate-binding pocket of RdRp and showed hydrophobic and hydrogen bond interaction. For molecular dynamics simulation, capmatinib, pralsetinib, ponatinib, and tedizolid phosphate were selected from the initial ten candidate compounds. MD simulation indicated that these compounds are stable at 50-ns MD simulation when bound to RdRp protein. The screen hit compounds, remdesivir, molnupiravir, and GS-441524, are bound in the substrate binding pocket with good binding-free energy. As a consequence, capmatinib, pralsetinib, ponatinib, and tedizolid phosphate are potential new inhibitors of RdRp protein with potential of limiting COVID-19 infection by blocking RNA synthesis.

Supplementary information: The online version contains supplementary material available at 10.1007/s11224-022-02072-1.

Keywords: Docking; FDA; Molecular simulation; RNA-dependent RNA polymerase; SARS-CoV-2; Virtual screening.

Fig. 1

Structural model of SARS-CoV-2 RNA-dependent RNA polymerase. A RdRp protein structure in the right-handed form with three domains highlighted: finger domain (violet), palm domain (blue), and thumb domain (pink). The N-terminal region of RdRp is also shown (cyan). B Space fill model highlighting the RNA template binding tunnel/active site presence between three domains. Colouring scheme as in A

Fig. 2

Remdesivir and GS-441524 binding to SARS-CoV-2 RdRP. Central image: Remdesivir (yellow) and its primary metabolite GS-441524 (green) bind to the mouth of the binding tunnel formed at the junction of the finger, palm, and thumb domains (colour scheme as in Fig. 1). Left and right images: Remdesivir and GS-441524 (white) form hydrophobic interactions and hydrogen bonds with residues of RdRp (cyan)

Fig. 3

Binding of virtual screen hit compounds to SARS-CoV-2 RdRp. Ensemble binding of the ten identified hit compounds from the virtual screen is shown docked to the binding tunnel as for the reference compounds in Fig. 2. RdRp colour scheme as in Fig. 1. There is a high degree of overlap in the putative binding sites for the compounds

Fig. 4

Modelled interactions between SARS-CoV-2 RdRp and risperidone, rimegepant, and irinotecan. Protein–ligand interactions formed between the screen hit compounds (grey) by hydrophobic interaction (grey dotted line), salt bridges (dotted yellow line), and H-bonds (solid yellow line) with numbered residues (cyan) in the binding tunnel of RdRp are shown

Fig. 5

Modelled interactions between SARS-CoV-2 RdRp and indocyanine green, alectinib, and pralsetinib. Protein–ligand interactions formed between the screen hit compounds (grey) by hydrophobic interaction (grey dotted line), salt bridges (dotted yellow line), and H-bonds (solid yellow line) with numbered residues (cyan) in the binding tunnel of RdRp are shown

Fig. 6

Modelled interactions between SARS-CoV-2 RdRp and ponatinib, capmatinib, lonafarnib, and tedizolid phosphate. Protein–ligand interactions formed between the screen hit compounds (grey) by hydrophobic interaction (grey dotted line), salt bridges (dotted yellow line), and H-bonds (solid yellow line) with numbered residues (cyan) in the binding tunnel of RdRp are shown

Fig. 7

Clustering of putative inhibitors of SARS-CoV-2 RdRp. ChemBioServer clustering found four clusters based on hierarchical clustering. The first cluster shows five compounds, including the reference compound remdesivir

Fig. 8

Molecular dynamic simulation of RdRp-ligand complexes. RMSD of the Cα backbone for selected protein–ligand complex compounds for 50-ns MD simulation

Fig. 9

Trajectory of hydrogen bond interactions between RdRp and selected hit compounds during MD simulation. A Total number of hydrogen bonds formed between RdRp-ligand complex. B Total number of hydrogen bond interactions between protein and solvent