Inhibitors were docked with active site of 3CLpro protein. Panel-I: ligand interaction map. Panel-II: interaction of inhibitors specific amino acids of 3CLpro at active site. Panel-III: lipophilic cavity of active site with drugs interacting with specific amino acids. Boceprevir (a), partapravir (b), tipranavir (c), ombitasvir (d), ivermectin (e), and micafungin (f) are arranged in columns for comparison. Drugs are represented in green color with ball and stick model. Arrows indicate the C–H, N–H, and C–O bonds between drugs and with Cys145, His41, and Glu166 residues since they are essential for the enzymatic activity of 3CLpro enzyme. However, we also observed the drugs interacting with neighboring amino acid residues.
100 nanosecond (ns) Molecular Dynamics simulations for micafungin and ivermectin
To investigate the stability of these docking poses, 100 ns molecular dynamics (MD) simulation studies were performed for two compounds, ivermectin and micafungin. Figure 7a panel I–III shows the MD simulations of micafungin with the monomer form of 3CLpro. The MD simulation data suggests that micafungin was stable and remained bound in the active site pocket throughout the 100ns simulated time (compare Fig. 7a I–III). The analysis of protein-ligand interaction fingerprints between the monomeric 3CLpro enzyme and micafungin (Fig. 7a, panel-IV) shows that micafungin has a predominant interaction with Glu166. It is possible that the micafungin remained bound in the pocket of the monomeric form of 3CLpro for the entire length of the 100-ns trajectory via a hydrogen bonding with Glu166. As noted earlier, interaction with Glu166 suggests interference with the dimerization of the 3CLpro in SARS-CoV-2, which is required for its activity28 thus explaining the inhibitory activity of micafungin against 3CLpro enzyme. Supplementary Movie 1 presents the real-time interaction of micafungin with the active site of monomeric 3CLpro indicating the stability of micafungin in the catalytic pocket of monomer.
MD simulation studies were carried as described under Methods section. Interaction of micafungin (a), and ivermectin (b) with monomeric form of 3 CLpro enzyme. c Interaction of ivermectin with active site of 3CLpro homodimer. Panel I–III represents the interaction of ligand at different time points in nano second (ns). Panel-IV represents the ligand-binding fingerprint of micafungin and ivermectin with specific amino acids of 3CLpro enzymes.
The MD simulations for ivermectin are shown in Fig. 7b (panel II–III), where ivermectin diffuses out of the catalytic pocket of 3CLpro monomer after 85 ns. It is evident from the protein-ligand fingerprint map that ivermectin interacts with both Cys145 and His41 of the 3CLpro monomer for about 14 ns (Fig. 7b, panel-IV). Later, ivermectin loses its interaction with His41 and does not show interaction with any amino acids of interest (Cys145, His41, and Glu166) and eventually diffuses out of the pocket at 85 ns (Fig. 7b, panel-II). Supplementary Movie 2 shows the instability of the ivermectin in the catalytic pocket of the monomeric form of 3CLpro. Since the homodimer is the active form of 3CLpro enzyme28, we hypothesized that the homodimeric form of 3CLpro is required to stabilize ivermectin in the catalytic pocket and hence is responsible for the inhibitory activity of ivermectin. To test our hypothesis, MD simulations of ivermectin with the homodimer form of 3CLpro was performed. Interestingly, we observed that ivermectin remained bound in the catalytic pocket of the homodimer (compare Fig. 7c, panel I–III) throughout the period of the simulation. The detailed analysis of the homodimer 3CLpro-ivermectin fingerprint region (Fig. 7c, panel-IV) shows that ivermectin interacts with both Cys145 and His41 for 2 ns, then with His41. After 85 ns, ivermectin contacts with Ser1 of the neighboring monomer, suggesting that this amino acid residue assists in the stabilization of ivermectin in the catalytic binding pocket (Fig. 7c, panel IV). Supplementary Movie 3 exhibits the real-time MD simulation interaction and the stability of ivermectin in the catalytic pocket with homodimer of 3CLpro from 0–100ns.
Further, Supplementary Fig. 3 shows the binding affinity (S-score) over the course of the MD simulation for ivermectin with the monomer and homodimer form of 3CLpro, and micafungin with monomer form of 3CLpro. We observed that the S-score for micafungin was stable over the period of computation whereas, ivermectin with monomer form of 3CLpro fluctuated from −9.64 to −2.2 kcal/mol. As shown in the Supplementary Fig. 3, upon leaving the active site ~85 ns ivermectin exhibited an increase in S-score after (~−2.0 kcal/mol). This is in stark contrast to the S-score of ivermectin in complex with the homodimer form of 3CLpro, which remained stable throughout the simulation with an average of −5.64 kcal/mol. Taken together, this computational model provides a framework for the possible interaction between these inhibitors and 3CLpro. However, the structural interaction of these drugs with SARS-CoV-2 3CLpro needs to be validated by X-ray crystallographic studies.
COVID-19 is a disease caused by the SARS-CoV-2 and is a major threat to public health globally because of the high rates of infection and mortality. Morbidity and mortality continue to rise due to the lack of a specific vaccine and drugs that prevent COVID-19 disease progression. There is an urgent need to identify and test potential therapeutics for this disease. One approach that may lead to a more rapid increase in treatment options is to repurpose currently approved FDA drugs for their ability to prevent or reduce the spread of virus and severity of COVID-19 pathogenesis. Many of the FDA approved drugs are being repurposed in clinics to treat COVID-19. However, the effectiveness of these drugs and their specific targets for preventing or reducing the severity of symptoms of COVID-19 has not yet been completely established2,29,30,31. Therefore, several laboratories are identifying specific drugs for many targets of SARS-CoV-2. Herein, we have investigated 47 FDA approved drugs that inhibit the SARS-COV-2 3CLpro enzymatic activity, the main enzyme for viral replication and the preferred drug target for COVID- 1932. We used MOE computational studies for the initial screening to select the drugs that have high affinity for 3CLpro and further functional inhibitory activity of the 47 selected drugs was confirmed using in vitro enzymatic assay. As noted in the previous studies24,25, our data suggests that inhibitory effects of drugs predicted from the computational screening as defined by the S-score do not agree with our experimental in vitro studies. Thus, additional in vitro screening for all the drugs is warranted.
credited to joanna ho