Crystallographic structure of wild-type SARS-CoV-2 main protease acyl-enzyme intermediate with physiological C-terminal autoprocessing site

Abstract

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), the pathogen that causes the disease COVID-19, produces replicase polyproteins 1a and 1ab that contain, respectively, 11 or 16 nonstructural proteins (nsp). Nsp5 is the main protease (M^pro) responsible for cleavage at eleven positions along these polyproteins, including at its own N- and C-terminal boundaries, representing essential processing events for subsequent viral assembly and maturation. We have determined X-ray crystallographic structures of this cysteine protease in its wild-type free active site state at 1.8 Å resolution, in its acyl-enzyme intermediate state with the native C-terminal autocleavage sequence at 1.95 Å resolution and in its product bound state at 2.0 Å resolution by employing an active site mutation (C145A). We characterize the stereochemical features of the acyl-enzyme intermediate including critical hydrogen bonding distances underlying catalysis in the Cys/His dyad and oxyanion hole. We also identify a highly ordered water molecule in a position compatible for a role as the deacylating nucleophile in the catalytic mechanism and characterize the binding groove conformational changes and dimerization interface that occur upon formation of the acyl-enzyme. Collectively, these crystallographic snapshots provide valuable mechanistic and structural insights for future antiviral therapeutic development including revised molecular docking strategies based on M^pro inhibition.

Fig. 1. Wild-type SARS-CoV-2 M^pro acyl-enzyme intermediate structure at 1.95 Å resolution.

a Overview of M^pro dimer. Each protomer colored spectrally (N-terminus blue to C-terminus red). A transparent molecular surface is shown around each protomer (chain A—orange, chain B—blue). b M^pro structure determined here shown in molecular surface colored as in (a). A symmetry-related chain in the crystal lattice (B′, white) directs its C-terminal six residues into the substrate binding groove of chain B (Ser301–Gln306 shown in CPK space filling representation). c Substrate binding groove (blue surface) of chain B with covalently bound C-terminal P1–P6 residues of B′. The N-terminus of chain A (Ser1, the so-called N-finger, orange surface) provides structural support for the S1 pocket of chain B. The side chains of the catalytic residues Cys145, His41, and residues that make direct hydrogen bonds to substrate are shown. d 2mF_o-DF_c electron density contoured at 1.0σ around the chain B′ C-terminus clearly reveals the thioester bond. Electron density for W_cat adjacent to the thioester carbonyl carbon shown in green, also contoured at 1.0σ. A simulated annealing OMIT map for the bound substrate is shown in Supplementary Fig. 5a.

Fig. 2. Comparison of wild-type acyl-enzyme intermediate and substrate-free M^pro structures.

a Superposition of the substrate-free (black) and acyl-enzyme (blue) forms reveals changes in the substrate binding groove width. The main chain atoms for bound B′ substrate are shown as transparent van der Waals spheres. b Molecular surface of wild-type M^pro with three-ordered water molecules (cyan spheres). Superposition of the acyl-enzyme structure shows these waters are coincident with oxygen atom positions and will be displaced upon substrate binding. c Analysis of the wild-type acyl-enzyme active site reveals a potential deacylating water (catalytic/nucleophilic–W_cat) approaching the Re-face of the thioester. Ball and stick diagram depicting the geometry and atomic interactions of the thioester linkage between the Sγ of Cys145 and main chain carbonyl carbon of substrate Gln306. The trigonal planar nature of the thioester group, defined by atoms Cα, C, and O of Gln306, and Sγ of Cys145 is shown as is the χ₁ dihedral angle (defined by atoms N, Cα, Cβ, and Sγ). The oxyanion hole hydrogen bond distances and angles are also labeled. Proposed deacylating water (W_cat) shown as a cyan sphere. α_BD is the Bürgi–Dunitz angle (W_cat-C=O) and d_a the attack distance.

Fig. 3. C145A SARS-CoV-2 M^pro product complex at 2.0 Å resolution.

a 2mF_o-DF_c electron density (contoured at 1.0σ) in chain B of the C145A mutant shows presence of the bound C-terminal product of symmetry-related molecule B′. Also see Supplementary Fig. 5b. b 2mF_o-DF_c electron density (1.0σ) of empty protomer, chain A, of the same C145A mutant structure shows presence of a highly ordered water molecule hydrogen bonded to Nε2 of His41, consistent with a general base role of the latter and coincident in position with the W_cat weakly observed in the acyl-enzyme complex as in c and Fig. 1c. c Superposition of the product chain A (empty binding site; magenta) and chain B (product bound; green) with the acyl-enzyme (chain B; blue).

Fig. 4. Modeling of the SARS-CoV-2 M^pro enzyme–substrate complex.

a CPK molecular surface of SARS-CoV-1 C145A catalytic mutant ES complex (PDB 5B60), including C-terminal cleavage site P6–P4′ (P1′–P3′ with green carbons). b CPK molecular surface for the SARS-CoV-2 M^pro acyl-enzyme active site. The additional residues P1′–P4′ (magenta carbons) are modeled based on (a). Sequence alignment for all M^pro processing sites shown in Supplementary Fig. 1b. Note the identical sequence preceding the scissile bond between SARS-CoV-1 and -2 M^pro, but divergence in P1′–P3′ (N-terminus of the subsequent nsp6). Despite these differences, the S1′–S3′ pockets observed in the SARS-CoV-2 M^pro acyl-enzyme active site are similar to that in (a), i.e., already preformed in the absence of P1′–P3′ (modeled here), and apparently not dependent on the binding of P2′. It is also evident from this panel that the P1′–P3′ side chains are not sterically matched to the S1′–S3′ pockets, perhaps an advantage in protein maturation.

Fig. 5. M^pro inhibitor binding in relation to the enzyme–substrate complex model.

The surface in each panel is that of chain B of the acyl-enzyme structure. The C-terminal autocleavage site enzyme–substrate complex model for SARS-CoV-2 (see Fig. 4b) is shown in black lines. The protomer B active site binding pockets (S1, S2, S4, S2′, and S3′) and bound B′ substrate residues (italics) are labeled in panel a. b–h Superposed drugs are shown in colored cpk representation with published names provided for each. PDB accession codes: 11a—6LZE, 11b—6M0K, 13b—6Y2G, telaprevir—7C7P, x0072—5R7Y, x0434—5R83, and x1392—5RFT. Drawings for each inhibitor can be found in Supplementary Fig. 8.

Fig. 6. Captured alternate SARS-CoV-2 M^pro C-terminal conformations can inform drug discovery.

a Superposition of SARS-CoV-2 M^pro acyl-enzyme intermediate protomers determined here with chain A and chain B in orange and blue, respectively. The alternate C-terminal orientations—labeled Cter (A) and (B)—observed reveal a druggable pocket at the dimerization interface. Arrows connect to corresponding C-terminal orientation in (b) and (c). b The C-terminus of chain A (orange VdW representations) is packed at and stabilizes the dimerization interface (blue and oranges surfaces), an interaction typical of the mature dimer. c In the acyl-enzyme and product complexes, chain B redirects its C-terminus ~180° (blue VdW representations) as also shown in (a), allowing capture within the active site cleft of a neighboring dimer in the crystal, with the extended peptide binding groove at the dimerization site now exposed (delineated by black ellipse). A recent structure-based fragment screen found several small molecules bound within this region including compound x1187 (magenta spheres; PDB 5RFA).