Prediction of Novel Inhibitors of the Main Protease (M-pro) of SARS-CoV-2 through Consensus Docking and Drug Reposition

Abstract

Since the outbreak of the COVID-19 pandemic in December 2019 and its rapid spread worldwide, the scientific community has been under pressure to react and make progress in the development of an effective treatment against the virus responsible for the disease. Here, we implement an original virtual screening (VS) protocol for repositioning approved drugs in order to predict which of them could inhibit the main protease of the virus (M-pro), a key target for antiviral drugs given its essential role in the virus' replication. Two different libraries of approved drugs were docked against the structure of M-pro using Glide, FRED and AutoDock Vina, and only the equivalent high affinity binding modes predicted simultaneously by the three docking programs were considered to correspond to bioactive poses. In this way, we took advantage of the three sampling algorithms to generate hypothetic binding modes without relying on a single scoring function to rank the results. Seven possible SARS-CoV-2 M-pro inhibitors were predicted using this approach: Perampanel, Carprofen, Celecoxib, Alprazolam, Trovafloxacin, Sarafloxacin and ethyl biscoumacetate. Carprofen and Celecoxib have been selected by the COVID Moonshot initiative for in vitro testing; they show 3.97 and 11.90% M-pro inhibition at 50 µM, respectively.

Keywords: 2019-nCov; 3CL-pro; COVID-19; M-pro; SARS coronavirus; SARS-CoV-2; chymotrypsin-like protease.

Figure 1

Overview of the main structural features of the SARS-CoV-2 M-pro. Panel (A) shows a general overview of the M-pro homodimeric structure (PDBid 6W63) and the relative location of the missense mutations shown in Table 1. The domain structure for one of the two protomers is also shown (with domain I in red, domain II in blue and domain III in green). Panel (B) shows the most important residues from the different subsites in the context of the binding site. Thus, the residues from the S₃, S₂, S₁ and S₁’ subsites are shown in yellow, cyan, red and green, respectively. Other important residues at the binding site that have not been assigned by Tang et al. to any of these subsites are shown in magenta. This figure was obtained with the help of the Maestro program [24].

Figure 2

Superposition of the binding site residues of SARS-CoV-2 M-pro. Residues corresponding to the free enzyme (PDBid 6M03), M-pro bound to a noncovalent inhibitor (PDBid 6W63) and M-pro bound to a covalent inhibitor (PDBid 6LU7) are shown in red, blue and green, respectively. This figure was obtained with the help of the Maestro program [24].

Figure 3

Three-dimensional view of the binding site in experimental complexes between M-pro and reversible and irreversible inhibitors. Panels (A–D) correspond to complexes with 13a (PDBid 6Y7M), 13b (PDBid 6Y2F), N3 (PDBid 6LU7) and X77 (PDBid 6W63), respectively. This figure was obtained with the help of the Maestro program [24].

Figure 4

General scheme of the workflow used for approved-drug reposition. VS screening steps are shown against a yellow background to distinguish them from preparation or results postprocessing steps.

Figure 5

Panels (A–G) show results for Perampanel, Carprofen, Celecoxib, Alprazolam, Trovafloxacin, Sarafloxacin and Ethyl biscoumacetate; respectively. For each predicted M-pro inhibitor, the figure shows: (A) the superposition of the three equivalent docked poses (if more than one was found for the same hit, the one that presented the highest mean docking score was chosen); and (B) the corresponding Glide docked pose after the MM-GBSA minimization. Docking score values obtained with Glide, FRED and AutoDock Vina are shown in the same color as the equivalent pose obtained with the corresponding program. The ∆G_bind energy value obtained after the MM-GBSA minimization of the Glide pose is also shown for each predicted inhibitor. This figure was obtained with the help of the Maestro program [24].

Figure 5

Panels (A–G) show results for Perampanel, Carprofen, Celecoxib, Alprazolam, Trovafloxacin, Sarafloxacin and Ethyl biscoumacetate; respectively. For each predicted M-pro inhibitor, the figure shows: (A) the superposition of the three equivalent docked poses (if more than one was found for the same hit, the one that presented the highest mean docking score was chosen); and (B) the corresponding Glide docked pose after the MM-GBSA minimization. Docking score values obtained with Glide, FRED and AutoDock Vina are shown in the same color as the equivalent pose obtained with the corresponding program. The ∆G_bind energy value obtained after the MM-GBSA minimization of the Glide pose is also shown for each predicted inhibitor. This figure was obtained with the help of the Maestro program [24].

Figure 5

Panels (A–G) show results for Perampanel, Carprofen, Celecoxib, Alprazolam, Trovafloxacin, Sarafloxacin and Ethyl biscoumacetate; respectively. For each predicted M-pro inhibitor, the figure shows: (A) the superposition of the three equivalent docked poses (if more than one was found for the same hit, the one that presented the highest mean docking score was chosen); and (B) the corresponding Glide docked pose after the MM-GBSA minimization. Docking score values obtained with Glide, FRED and AutoDock Vina are shown in the same color as the equivalent pose obtained with the corresponding program. The ∆G_bind energy value obtained after the MM-GBSA minimization of the Glide pose is also shown for each predicted inhibitor. This figure was obtained with the help of the Maestro program [24].

Figure 5

Panels (A–G) show results for Perampanel, Carprofen, Celecoxib, Alprazolam, Trovafloxacin, Sarafloxacin and Ethyl biscoumacetate; respectively. For each predicted M-pro inhibitor, the figure shows: (A) the superposition of the three equivalent docked poses (if more than one was found for the same hit, the one that presented the highest mean docking score was chosen); and (B) the corresponding Glide docked pose after the MM-GBSA minimization. Docking score values obtained with Glide, FRED and AutoDock Vina are shown in the same color as the equivalent pose obtained with the corresponding program. The ∆G_bind energy value obtained after the MM-GBSA minimization of the Glide pose is also shown for each predicted inhibitor. This figure was obtained with the help of the Maestro program [24].

Figure 6

Histograms corresponding to the docking scores of four reference libraries containing predicted M-pro inhibitors. Panels (A–D) show the results for OTAVA-ML-SARS, OTAVA-SARS-CoV-2, COVID-Moonshot and DD-top-1000, respectively. Score thresholds for selecting the top 30% ligands with the highest affinity are shown for each histogram.

Figure 7

General scheme of the workflow used for defining the docking score thresholds to be used to identify high affinity poses for M-pro during the VS.