Mechanism of the maturation process of SARS-CoV 3CL protease

Abstract

Severe acute respiratory syndrome (SARS) is an emerging infectious disease caused by a novel human coronavirus. Viral maturation requires a main protease (3CL(pro)) to cleave the virus-encoded polyproteins. We report here that the 3CL(pro) containing additional N- and/or C-terminal segments of the polyprotein sequences undergoes autoprocessing and yields the mature protease in vitro. The dimeric three-dimensional structure of the C145A mutant protease shows that the active site of one protomer binds with the C-terminal six amino acids of the protomer from another asymmetric unit, mimicking the product-bound form and suggesting a possible mechanism for maturation. The P1 pocket of the active site binds the Gln side chain specifically, and the P2 and P4 sites are clustered together to accommodate large hydrophobic side chains. The tagged C145A mutant protein served as a substrate for the wild-type protease, and the N terminus was first digested (55-fold faster) at the Gln(-1)-Ser1 site followed by the C-terminal cleavage at the Gln306-Gly307 site. Analytical ultracentrifuge of the quaternary structures of the tagged and mature proteases reveals the remarkably tighter dimer formation for the mature enzyme (K(d) = 0.35 nm) than for the mutant (C145A) containing 10 extra N-terminal (K(d) = 17.2 nM) or C-terminal amino acids (K(d) = 5.6 nM). The data indicate that immature 3CL(pro) can form dimer enabling it to undergo autoprocessing to yield the mature enzyme, which further serves as a seed for facilitated maturation. Taken together, this study provides insights into the maturation process of the SARS 3CL(pro) from the polyprotein and design of new structure-based inhibitors.

Fig. 1

Crystal structures of the wild-type and C145 mutant 3CL^pro. In A and B, the overall three-dimensional structures of wild-type and C145A mutant 3CL^pro are shown as ribbons (protomer A is green and protomer B is blue), and the “product” in the active site cleft between domain I and II is shown in yellow. In C, the dimer structure is composed of protomer A (shown with a solid tube in green) and protomer B (shown with charge potentials). The C terminus of protomer B′ (shown in cyan) in another asymmetric unit is intercalated into the active site of protomer B. D, an enlarged view of C near the active site, showing the C-terminal amino acids of protomer B′ as well as the N-terminal amino acids of protomer A in the neighborhood of the active site of protomer B.

Fig. 2

SDS-PAGE analysis of the maturation of SARS-CoV recombinant proteases. In A, the six constructs of the recombinant protease are listed, including the wild-type or C145A mutant enzyme containing N- and/or C-terminal 10 additional amino acids and the N-terminal Trx tag. In B, MW represents the molecular weight markers. Lanes 1 and 3 represent 300 mm imidazole eluant from Ni-NTA column of constructs 1 and 3, respectively, where the protease disappeared, because it underwent autoprocessing and lost the tags. Lanes 2, 4, 5, and 6 represent the protease of constructs 2, 4, 5, and 6, respectively, eluted by 300 mm imidazole from the Ni-NTA column. They either lack the N-terminal cleavage sequence or contain C145A mutant, so they were retained in the Ni-NTA column before elution.

Fig. 3

Facilitated processing of Trx-10aa-C145A-10aa-GST by the active 3CL protease.A, the inactive C145A (to prevent autoactivation) containing N- and C-terminal tags was prepared as a substrate for active 3CL^pro. The substrate protein (lane S) was treated with 1/10 of wild-type 3CL^pro (lane E), and the products were monitored with SDS-PAGE analysis after specified periods of incubation time shown on the top of the figure. According to the SDS-PAGE data, the N-terminal tag was digested followed by the C-terminal tag cleavage, as indicated by the product fragments on the right panel. B, the time course of formation of Trx tag (▴) and GST tag (□) from facilitated processing.

Fig. 4

AUC experiments of wild-type SARS 3CL^pro. Shown here is an example of using AUC to measure the K_d of the wild-type protease dimer-monomer equilibrium. The A₂₈₀ absorbance of the protein as a function of radius and time was recorded to calculate the sedimentation coefficient and the K_d value A, circles represent the experimental data, and the lines are the computer-generated results from fitting the data to the Lamm equation with the SedFit program. B, fitting residuals plotted as a function of radial position. C, grayscale of the residual bitmap. The randomly distributed residuals and bitmap show the quality of the data fitting. D, continuous sedimentation coefficient of wild-type 3CL^pro derived from the data shown in A.

Fig. 5

Molecular interactions of the active site residues of protomer B with the C-terminal residues of protomer B′.A, stereo view of the electron density map of the C-terminal region (red stick) of protomer B′ bound in the S pockets (cyan stick) of protomer B. B, details of the molecular interactions between the active site S1-S6 pockets of protomer B and the C-terminal residues of protomer B′. H-bonds are shown as green broken lines.

Fig. 6

Superposition of 3CL^pro active sites and inhibitors.A, superimposition of the active site of five 3CL^pro protease structures: cyan and blue, protomer A and B of C145A; light green and gold, protomer A and B of the wild type, respectively; crimson, HCoV 229E 3CL^pro (1P9S); and pink, TGEV 3CL^pro (1P9U). B, superimposition of the six C-terminal residues of SARS 3CL^pro (SGVTFQ) (cyan), the inhibitor of TGEV 3CL^pro (1P9U, pink), the inhibitor of TGEV 3CL^pro (VNSTLQ) at the active site of SARS 3CL^pro (1UK4, gold), and the inhibitor of Rhinovirus 3CL^pro, AG7088 at the active site (1CQQ, green).

Fig. 7

Proposed scheme of SARS 3CL^pro maturation. Two polyproteins are shown in pink and cyan, each with three domains (I, II, and III). The maturation processing is composed of Step 1: polyprotein (cyan) approaches a second polyprotein (pink) and inserts its N terminus into the active site to be cleaved; Step 2: the N terminus of the uncleaved polyprotein (pink) then inserts its N terminus into the active site for processing; Step 3: after N-terminal processing, the polyprotein with the N terminus flips over to its new position from the active site to form a premature dimer; Step 4: the C terminus of the partially digested polyprotein in the premature dimer is inserted into the active site of another immature dimer to be cleaved and finally the mature dimer is formed.