Structural and Evolutionary Analysis Indicate That the SARS-CoV-2 Mpro Is a Challenging Target for Small-Molecule Inhibitor Design

Abstract

The novel coronavirus whose outbreak took place in December 2019 continues to spread at a rapid rate worldwide. In the absence of an effective vaccine, inhibitor repurposing or de novo drug design may offer a longer-term strategy to combat this and future infections due to similar viruses. Here, we report on detailed classical and mixed-solvent molecular dynamics simulations of the main protease (Mpro) enriched by evolutionary and stability analysis of the protein. The results were compared with those for a highly similar severe acute respiratory syndrome (SARS) Mpro protein. In spite of a high level of sequence similarity, the active sites in both proteins showed major differences in both shape and size, indicating that repurposing SARS drugs for COVID-19 may be futile. Furthermore, analysis of the binding site's conformational changes during the simulation time indicated its flexibility and plasticity, which dashes hopes for rapid and reliable drug design. Conversely, structural stability of the protein with respect to flexible loop mutations indicated that the virus' mutability will pose a further challenge to the rational design of small-molecule inhibitors. However, few residues contribute significantly to the protein stability and thus can be considered as key anchoring residues for Mpro inhibitor design.

Keywords: COVID-19; SARS-CoV; SARS-CoV-2; coronavirus; drug design; evolutionary analysis; ligand tracking approach; molecular dynamics simulations; small-molecule inhibitors.

Figure 1

The differences between the severe acute respiratory syndrome coronavirus main protease (SARS-CoV Mpro) and SARS-CoV-2 Mpro structures. (A) The overall structure of both SARS-CoV and SARS-CoV-2 Mpros with differing amino acids are marked as black (SARS-CoV Mpro) and blue (SARS-CoV-2 Mpro). (B) Close-up of the active site cavity and bound N3 inhibitor into SARS-CoV (black sticks) and SARS-CoV-2 (blue sticks) Mpros. The catalytic water molecule that resembles the position of the third member of the catalytic triad adopted from the cysteine proteases is shown for both SARS-CoV (black sphere) and SARS-CoV-2 (blue sphere) Mpros. The active site residues are shown as red sticks and the proteins’ structures are shown in surface representation. The differing residues in position 46 located near the entrance to the active site are marked with an asterisk (*) on the (A) and as blue and black lines on the (B) panel.

Figure 2

The differences between the maximal accessible volume of the binding cavities calculated during molecular dynamics (MD) simulations of both apo structures of Mpros (SARS-CoV and SARS-CoV-2) and structures with co-crystallised N3 inhibitor (SARS-CoV^N3 and SARS-CoV-2^N3) used as different starting points for 10 replicas of 50 ns per structure. The position of the blue sphere (hot-spot with highest density) in each structure reflects the position of the catalytic water molecule.

Figure 3

Flexibility of loops surrounding the entrance to the binding cavity of (A) SARS-CoV-2 Mpro, (B) SARS-CoV Mpro, (C) SARS-CoV Mpro^N3, and (D) SARS-CoV Mpro^N3. For the picture clarity, only residues creating loops were shown. Upper row: RMSF data. The active site residues are shown as red sticks, and the A46S replacement between SARS-CoV and SARS-CoV-2 main proteases is shown as light blue sticks. The width and colour of the shown residues reflect the level of loop flexibility. The wider and darker residues are more flexible. Lower row: the results of normal mode analysis as a superposition of active site surroundings; structures are coloured white—initial conformation, black—final conformation, gray—transient conformation.

Figure 4

Localisation of the local hot-spots identified in the binding site cavities in SARS-CoV-2 and SARS-CoV main proteases. Hot-spots of individual cosolvents are represented by spheres, and their size reflects the hot-spot density. The colour coding is as follows: purple—urea, green—dimethylsulfoxide, yellow—methanol, orange—acetonitrile, pink—phenol, red—benzene. The active site residues are shown as red sticks, and the proteins’ structures are shown in cartoon representation; loop 44–52 is grey. The proteins’ structures come from the MD simulation snapshots (first frame of the production stage).

Figure 5

Localisation of the evolutionary-correlated residues of Mpros (black sticks). The correlated mutation analysis (CMA) analysis provided four groups of evolutionary-correlated residues. The SARS-CoV-2 Mpro structure is presented as a cartoon, the active site residues are shown as red sticks, the unique residues of the SARS-CoV-2 Mpro as blue sticks, and the asterisk (*) indicates the residue belonging to the evolutionary-correlated residues, unique for SARS-CoV Mpro. The loop C44-P52 is coloured black and the F185-T201 loop is orange. Please note that within one of the correlated groups (upper left), the residues from C44-P52 loop are correlated with Q189 from the linker loop and with residue from III domain.

Figure 6

Potential mutability of SARS-CoV-2 Mpro. (A) Structure of SARS-CoV-2 Mpro with the most energetically favourable potential mutations of amino acids marked as green surface. Positions of amino acids that differ from the ones in SARS-CoV Mpro structure marked as blue sticks. Catalytic dyad marked as red. (B) The catalytic site of SARS-CoV-2 Mpro is shown as surface with the most energetically favourable potential mutations shown as green, neutral as white, and unfavourable as red. The C44-P52 loop is shown as black mesh.