SARS-CoV-2 polyprotein substrate regulates the stepwise Mpro cleavage reaction

Abstract

The processing of the Coronavirus polyproteins pp1a and pp1ab by the main protease M^pro to produce mature proteins is a crucial event in virus replication and a promising target for antiviral drug development. M^pro cleaves polyproteins in a defined order, but how M^pro and/or the polyproteins determine the order of cleavage remains enigmatic due to a lack of structural information about polyprotein-bound M^pro. Here, we present the cryo-EM structures of SARS-CoV-2 M^pro in an apo form and in complex with the nsp7-10 region of the pp1a polyprotein. The complex structure shows that M^pro interacts with only the recognition site residues between nsp9 and nsp10, without any association with the rest of the polyprotein. Comparison between the apo form and polyprotein-bound structures of M^pro highlights the flexible nature of the active site region of M^pro, which allows it to accommodate ten recognition sites found in the polyprotein. These observations suggest that the role of M^pro in selecting a preferred cleavage site is limited and underscores the roles of the structure, conformation, and/or dynamics of the polyproteins in determining the sequence of polyprotein cleavage by M^pro.

Keywords: 3CL main protease (M(pro)); SARS CoV-2; cryogenic electron microscopy (cryo-EM); polyprotein; proteolytic processing.

Figure 1

nsp7-10 polyprotein processing by M^pro.A, preparation of the nsp7-10 polyprotein. Different oligomers of nsp7-10 polyprotein (peaks I, II, and III) were separated by size-exclusion chromatography in the presence of 1 M NaCl in the purification buffer (left). Proteins in fraction III were analyzed by SDS‒PAGE (right). The molecular weight of the nsp7-10 polyprotein is 58.2 kDa. B, limited nsp7-10 polyprotein proteolysis by M^pro. Substrate (nsp7-10), products (nsp7, nsp8, nsp9, and nsp10), intermediates (nsp7-8, nsp7-9), and M^pro were separated by SDS‒PAGE and labeled. The time after mixing M^pro with nsp7-10 (1:1 ratio) is indicated above the lanes. C, limited nsp7-10 proteolysis assay under substrate excess conditions (M^pro:nsp7-10 = 1:4). D, schematic illustration of stepwise nsp7-10 polyprotein cleavage by M^pro. Nonstructural proteins within nsp7-10 and their molecular weights are indicated. M^pro recognition sites found between nsps are depicted as red lines, and the cleavage order is indicated by red arrows. M^pro, main protease.

Figure 2

Cryo-EM structure of the nsp7-10 polyprotein bound M^pro-C145A.A, cryo-EM density map of the M^pro-C145A and nsp7-10 complex. Density maps corresponding to each protomer of the M^pro dimer, and the recognition site are indicated by color. Three domains of M^pro and the P1, P2, and P3' positions of the recognition site are indicated. B, a magnified view of the active center cleft of M^pro (boxed area of protomer A in panel A). The cryo-EM density map is partially transparent, and amino acid residues contacting the nsp9/10 region (<4 Å) are depicted as stick models and labeled. The density map and the model of nsp9/10 are omitted for clarity. C, a magnified view of the nsp9/10 and M^pro interaction. Structures of the nsp9/10 and M^pro residues participating in the nsp9/10 interaction are depicted as stick models with a partially transparent surface model of the M^pro and polyprotein complex. Locations of the nsp9 and nsp10 connecting to the nsp9/10 recognition site (stick model) are indicated. Orientation is the same as B. D, low path filtered cryo-EM density map of the M^pro and polyprotein complex (gray transparent overlayed with M^pro and nsp9/10 recognition site) shows the densities of the nsp7-10 polyprotein outside from the nsp9/10 region (dashed ovals). These densities are located above the active center of M^pro without any contact with M^pro. M^pro, main protease.

Figure 3

Cryo-EM structure of wild-type M^pro.A, cryo-EM density map of M^pro. Density maps corresponding to each protomer of the M^pro dimer are colored and indicated. Three domains and the active center of M^pro are indicated. B, left, comparison of the M^pro structures in the apo form (gray and block) and in the nsp7-10 complex (green, cyan, and yellow); Right, a magnified view of the active center of the M^pro (boxed area of protomer A in the right panel). The flexible P2 (pink) and P5 loops (blue) around the substrate binding cleft move toward the nsp9/10 recognition site in the M^pro and polyprotein complex (indicated by pink and blue arrows). C, comparison of the B-factor distributions in the apo form (top) and in the nsp7-10 complex (bottom) of M^pro. Cartoon representation of the models with gradients of color (blue, white to red) and thickness (narrow to wide) reflecting the scale of the B factors (low to high). Domains and loops of the apo-form M^pro showing a higher B-factor compared with the M^pro and nsp7/10 complex are indicated by black dashed ovals. M^pro, main protease.

Figure 4

Two distinct models explaining the stepwise polyprotein processing by M^pro.A, polyprotein-driven model. B, M^pro-driven model. The different preprocessed nsps are colored separately and indicated and both the protomers of M^pro are colored cyan and light green with the active site region circled in red.