SARS-CoV 3CL protease cleaves its C-terminal autoprocessing site by novel subsite cooperativity

Significance

The SARS-CoV protease (3CL^pro) has “noncanonical” substrate specificity for its C-terminal autoprocessing. Phe is required at both the second position upstream of the cleavage point (P2) and the third downstream position (P3′). This finding is surprising, given that 3CL^pro reportedly requires Leu at the P2 position with no preference at the P3′ position. The conventional “consensus sequence” cannot explain this noncanonical specificity. Crystallography revealed that Phe at the P2 position changes the conformation of the substrate-binding pocket, and thereby creates the subsite for Phe at the P3′ position. This noncanonical specificity avoids the autoinhibition due to the mature C-terminal sequence of 3CL^pro, which should be serious if Leu exists at the P2 position.

Abstract

The 3C-like protease (3CL^pro) of severe acute respiratory syndrome coronavirus (SARS-CoV) cleaves 11 sites in the polyproteins, including its own N- and C-terminal autoprocessing sites, by recognizing P4–P1 and P1′. In this study, we determined the crystal structure of 3CL^pro with the C-terminal prosequence and the catalytic-site C145A mutation, in which the enzyme binds the C-terminal prosequence of another molecule. Surprisingly, Phe at the P3′ position [Phe(P3′)] is snugly accommodated in the S3′ pocket. Mutations of Phe(P3′) impaired the C-terminal autoprocessing, but did not affect N-terminal autoprocessing. This difference was ascribed to the P2 residue, Phe(P2) and Leu(P2), in the C- and N-terminal sites, as follows. The S3′ subsite is formed by Phe(P2)-induced conformational changes of 3CL^pro and the direct involvement of Phe(P2) itself. In contrast, the N-terminal prosequence with Leu(P2) does not cause such conformational changes for the S3′ subsite formation. In fact, the mutation of Phe(P2) to Leu in the C-terminal autoprocessing site abolishes the dependence on Phe(P3′). These mechanisms explain why Phe is required at the P3' position when the P2 position is occupied by Phe rather than Leu, which reveals a type of subsite cooperativity. Moreover, the peptide consisting of P4–P1 with Leu(P2) inhibits protease activity, whereas that with Phe(P2) exhibits a much smaller inhibitory effect, because Phe(P3′) is missing. Thus, this subsite cooperativity likely exists to avoid the autoinhibition of the enzyme by its mature C-terminal sequence, and to retain the efficient C-terminal autoprocessing by the use of Phe(P2).

Severe acute respiratory syndrome coronavirus (SARS-CoV) produces several functional proteins in infected human cells by cleaving them from its two overlapping “polyproteins,” pp1a (486 kDa) and pp1ab (790 kDa) (1). Papain-like protease 2 (PL2^pro) and 3C-like protease (3CL^pro, also referred to as the main protease, M^pro), are included among these polyproteins. PL2^pro cleaves three sites and 3CL^pro cleaves 11 sites in the polyproteins to generate individual functional proteins, including an RNA-dependent RNA polymerase, a helicase, a single-stranded RNA-binding protein, an exoribonuclease, an endoribonuclease, and a 2′-O-ribose methyltransferase (1). 3CL^pro is a cysteine protease that is excised from polyproteins by its own proteolytic activity (1) and forms a homodimer with one active site per subunit (2). 3CL^pro reportedly recognizes the residues from P4 to P1 on the N-terminal side and P1′ on the C-terminal side of the cleavage sites, based on the consensus sequences around the processing sites in the SARS-CoV polyproteins and the extensive mutagenesis analyses of the N-terminal autoprocessing site (3, 4).

Three types of crystal structures of SARS-CoV 3CL^pro have been reported (as reviewed in ref. 5): the wild-type active dimer (wt-dimer) (2, 6); the monomeric forms or the G11A, R298A, and S139A mutants, which cannot dimerize (7–10); and the super-active octamer (11). The 3CL^pro subunit consists of the N-terminal finger (residues 1–8), the catalytic domain (residues 8–184), and the C-terminal domain (residues 201–306) (2), and the overall domain structures are the same among all of the reported 3CL^pro structures. 3CL^pro requires dimerization for the proteolytic activity (12), as suggested by the structure of the wt-dimer (2, 6).

As for the C-terminal autoprocessing mechanism of 3CL^pro, the structure of the “C-terminal product”-bound form has been reported for the mature-type 3CL^pro C145A mutant, in which the C-terminal portion of one subunit is bound to a subunit in an adjacent asymmetric unit (13). In the present study, we crystallized the dimer of the 3CL^pro proform, containing a 10-residue C-terminal prosequence, with the C145A mutation of the catalytic cysteine residue. As expected, we found that one of these prosequences is bound, as a substrate, to the active site of a subunit from an adjacent asymmetric unit. Based on this structure and biochemical experiments, we conclude that Phe at the P3′ position is required when the P2 residue is Phe. This recognition mode appears only for the C-terminal autoprocessing of 3CL^pro in the SARS-CoV polyproteins.

Results and Discussion

The Crystal Structure of the C-Terminal Proform of SARS-CoV 3CL^pro Revealed Unexpected Binding of the P3′ Residue to the Enzyme.

The “C-terminal proform” of 3CL^pro was designed with a 10-residue C-terminal prosequence, together with a cleavage-defective active-site mutation (C145A) (Fig. 1A). We determined the crystal structure of the proform at 2.2-Å resolution (Fig. 1B) and that of the mature form at 1.7-Å resolution (Fig. S1A), as summarized in Table S1. Their overall homodimer structures are almost the same as the mature-form structures reported previously (2, 13).

Fig. 1.

The crystal structure of the C-terminal proform of SARS-CoV 3CL^pro reveals unexpected binding of the P3′ residue to the enzyme. (A) The SARS-CoV 3CL^pro proform used for the 3D structure analyses. This proform contains the mutated catalytic site C145A, with the 10-aa residue C-terminal prosequence. (B) Crystal structure of the proform homodimer (green and blue ribbon models). Residues 1–310 and 1–301, respectively, can be observed in the electron density map. Ser1, His41, Ala145, and Gln306 are indicated by red sticks. (C) The C-terminal sequence (the P4, P3, P2, and P1 residues; blue stick models) and the prosequence (the P1′, P2′, P3′, and P4′ residues; pink stick models) are bound to the enzyme as a substrate in the adjacent asymmetric unit (surface model with charge). The unbiased Fo–Fc difference electron density map for omitted residues 303–310 (P4–P5′) is contoured at 3.0σ. The map was developed using PHENIX (14). The replaced active site residue (Ala145) is also indicated. (D) The 11 sequences that are cleaved by SARS-CoV 3CL^pro in the SARS-CoV polyproteins. The first and second sequences are the C- and N-terminal cleavage sequences, respectively, of 3CL^pro itself (1). (E) The consensus sequence for cleavage by 3CL^pro, analyzed by the sequence logo program (15), using the sequences shown in D and WebLogo version 2.8.2 (weblogo.berkeley.edu/) (16).

Fig. S1.

Crystal structure of the mature form of SARS-CoV 3CLpro. (A) Crystal structure of the mature form of the wild-type SARS-CoV 3CL^pro. (B) Superimposition of two subunits of SARS-CoV 3CL^pro. The structural differences between the subunits are highlighted in residues 45–51 (TAEDMLN) and the C-terminal residues 300–306 (CSGVTFQ).

Table S1.

Structure determination statistics

Statistic	Mature form (PDB ID code 2DUC)	Proform (PDB ID code 5B6O)
X-ray data
Space group	P21	P212121
Unit cell, Å
A	52.282	63.682
B	96.317	90.237
C	67.753	110.268
β,°	102.90
Resolution, Å	1.7	2.2
Wavelength, Å	1.0000	1.0000
No. of observations	357,673	252,832
No. of unique reflections*	71,412 (7,113)	32,224 (3,096)
Completeness, %	100.0 (99.5)	97.9 (96.8)
<I>/σ, I	12.0 (4.0)	24.2 (5.5)
Rsym, %†	12.5 (35.5)	7.5 (40.3)
Refinement
Rcryst, %‡	19.4	20.6
Rfree, %§	22.1	25.3
rmsd from standard stereochemistry
Bond lengths, Å	0.010	0.020
Bond angles, °	1.6	1.2
Dihedral angles, °	12.14	17.08
Ramachandran plot statistics
Favored regions, %	97.7	95.4
Allowed regions, %	2.0	4.3
Outliers, %	0.3	0.3

^*Numbers in parentheses represent values in the highest-resolution shell (wild type: 1.70–1.76 Å; C145A mutant: 2.20–2.28 Å).

^†R_sym = Σ|I_j − <I>|/ ΣI_j, where I_j is the observed integrated intensity, <I> is the average integrated intensity obtained from multiple measurements, and the summation is over all observed reflections.

^‡R_cryst = Σ||F_obs| − |F_calc||/Σ|F_obs|, where F_obs and F_calc are observed and calculated structure factor amplitudes, respectively.

^§R_free calculated with randomly selected reflections (5%).

In the crystal of the proform, the C-terminal portion of one of the subunits is bound to the active site of the subunit from an adjacent asymmetric unit (Fig. 1C and Fig. S2), as in the structure of the C-terminal product-bound form (13). The interaction of the C-terminal portion with the adjacent molecule is facilitated by the inherent flexibility of the C-terminal heptapeptide CSGVTFQ (residues 300–306; Fig. S1B), which corresponds to the P7–P1 residues. The P4, P3, P2, and P1 residues interact with the substrate recognition sites (S4–S1) (Fig. 1C and Fig. S2). Their binding modes are almost the same as those in the C-terminal product-bound structure (13) and the N-terminal substrate-bound structure (17), which clarifies the substrate preferences for the P4–P1 positions, as determined by statistical and experimental methods (1, 3, 4, 18) (Fig. 1 D and E).

Fig. S2.

Binding of the C-terminal moiety (indicated with X) of the proform of 3CL^pro with the C-terminal prosequence (pro-3CL^pro), as a substrate, to a subunit of pro-3CL^pro (indicated by A) in an adjacent asymmetric unit. This figure was prepared with LIGPLOT (37).

Our structure contains the P1′–P4′ residues of the prosequence, and thus represents the “C-terminal substrate”-bound form (Fig. 1C and Fig. S2). Statistical and experimental analyses (1, 3, 4, 18) revealed that the P1′ position has a preference for small amino acids, but no preferences have been reported for the P2′ and P3′ positions (Fig. 1 D and E). Surprisingly, we found that the side chain of the P3′ residue, Phe, is also accommodated in a specifically complementary pocket (Fig. 1C and Fig. S2). Therefore, we tested whether the P3′ position is actually recognized by the pocket, and whether it is important for autoprocessing.

The P3′ Residue of the C-Terminal Processing Position Is Recognized by the Enzyme, in Contrast to the N-Terminal Processing Site.

We previously established an analysis system for 3CL^pro autoprocessing (19) by duplicating and differentiating the participating molecules as the enzyme and the substrate (Fig. 2A). As the enzyme, the 3CL^pro moiety with the N- and C-terminal prosequences, accompanied by an S-tag (a 15-residue tag derived from RNase S, which is detected by the S protein, the other part of RNase S) and a His-tag, respectively, was synthesized by an Escherichia coli cell-free protein synthesis method (20) (Fig. 2A). Note here that the mature 3CL^pro is formed during the protein synthesis reaction (30 °C, 4 h), because both the N- and C-terminal processing sites are completely cleaved by the activity of the 3CL^pro moiety (19). The substrate molecule was differentiated from the above construct by replacing the core region of the 3CL^pro moiety with green fluorescent protein (GFP) (Fig. 2A), and was prepared by the same cell-free method. It has the S-tag, the N-terminal prosequence, the N-terminal 10 residues of 3CL^pro, GFP, the C-terminal 10 residues of 3CL^pro, the C-terminal prosequence, and the 6× His tag, in that order. This substrate molecule lacks the catalytic site, the substrate-binding site, and dimerization ability. The enzyme and the substrate were combined immediately after their synthesis, and the apparent k_cat/K_M value (i.e., cleavability) was obtained as described previously (19) (Fig. S3). Using this analysis system, we separately examined the effects of various mutations of the substrate and the enzyme on the cleavage reactions (Fig. S4).

Fig. 2.

Dual specificity of SARS-CoV 3CL^pro. (A) The enzyme and substrate molecules prepared by the E. coli cell-free protein synthesis method. Arrowheads indicate sites that are cleaved by the proteolytic activity of 3CL^pro; the circled C indicates the catalytically active Cys145. Figure reprinted from ref. 19. (B) Effect of the P3′ position (residue 309) on the cleavability (apparent k_cat/K_M) of the C-terminal processing site. Because residue 3 has no effect on C-terminal processing, the F3A and F3N mutations are negative controls. (C) Effect of the P3′ position (residue 3) on the cleavability (apparent k_cat/K_M) of the N-terminal processing site. Because residue 309 has no effect on the N-terminal processing, the F309A and F309N mutations are negative controls. (D) Effect of the P2 position on P3′ dependency. When P2 = Phe, effective cleavage occurs only when P3′ = Phe; compare FF and FA/FN. However, when P2 = Leu, effective cleavage occurs regardless of the P3′ position; no significant difference in the apparent k_cat/K_M is seen among LF, LA, and LN. (E) Effect of residue 305 on the proteolytic activity toward the N-terminal processing site. Values are mean ± SD from technical replicates (n = 3).

Fig. S3.

The processing assay method (19). (A) The enzyme and the substrate molecules, synthesized by the E. coli cell-free protein synthesis method, were mixed together, incubated for 1 h at 30 °C, and analyzed by SDS/PAGE. To estimate the activity quantitatively, a series of enzyme solutions diluted by 2-, 4-, 8-, 16-, 32-… fold was prepared before the mixing (1:1). (B) The dilution ratio at which 50% cleavage was achieved was estimated as the activity of the “enzyme” toward the N- and C-terminal processing sites in the “substrate.” The cleavability value thus obtained was converted to k_cat/K_M using the protein concentration and the standard value of the wild-type/mature enzyme. (Top) The SDS/PAGE gel stained with Coomassie brilliant blue G-250. (Bottom) The population of each molecular species, in molar ratio. The red arrow and asterisks in the top panel and the red circles in the bottom panel indicate the substrate. The blue arrow and asterisks in the top panel and the blue circles in the bottom panel indicate the N-terminally processed substrate. The purple arrow and asterisks in the top panel and the purple circles in the bottom panel indicate the substrate processed at both termini. The C-terminally processed substrate did not appear. Figure reprinted from ref. 19.

Fig. S4.

Processing assay. The enzyme and the substrate molecules (Left and Center), synthesized by the E. coli cell-free protein synthesis method, were mixed together, incubated for 1 h at 30 °C, fractionated on an SDS/PAGE gel, and stained with Coomassie brilliant blue G-250 (Right). To estimate the activity quantitatively, a series of enzyme solutions diluted by 2-, 4-, 8-, 16-, 32-… fold was prepared before the mixing (1:1). The red arrows and asterisks indicate the substrate; the blue arrows and asterisks indicate the N-terminally processed substrate, and the purple arrows and asterisks indicate the substrate processed at both termini. The C-terminally processed substrate did not appear.

The C-terminal cleavability (i.e., apparent k_cat/K_M), with the mature-form wild-type enzyme, was reduced to 1/5, by the alteration of Phe309 of the substrate (Fig. 2B, enzyme:wt, substrate:wt vs. F309A/F309N). This means that the P3′ position for C-terminal processing is actually recognized by the enzyme. In contrast, for N-terminal processing, the alteration of Phe3 of the substrate to Ala/Asn (F-to-A/N) did not affect the k_cat/K_M value (Fig. 2C, wt). Therefore, the P3′ position is recognized in C-terminal processing, in good agreement with the crystal structure of the proform with the C-terminal prosequence (Fig. 1C). In contrast, the P3′ position is not recognized in N-terminal processing, in agreement with the previous study using the N-terminal processing sequence as a substrate (4).

In addition to the mature enzyme, its proforms with the N- and/or C-terminal prosequences may be involved in autoprocessing. Because this enzyme prefers Gln at the P1 position (4), the Gln(Q)-to-Asn(N) mutations were used to cause the loss of cleavage at its own N- and/or C-terminal processing site(s), which should result in the proforms. In fact, the Q-to-N mutation(s) at the P1 position(s) (position –1 and/or position 306 of the enzyme for autoprocessing of the N and C termini, respectively) abolished the autocleavage (19). These proforms (Q–1N, Q306N, and Q–1N&Q306N) actually exhibited lower activities compared with the mature form for the C- and N-terminal processing of the wild-type substrate (Fig. 2 B and C, respectively) (19). Nevertheless, the patterns of the relative activities for the wt substrate and the P3′ F-to-A/N variant substrate are quite similar between the wt and mutant enzymes (Fig. 2 B and C). Consequently, neither the N- nor the C-terminal prosequence in the enzyme affects the specificities for the N- and C-terminal P3′ positions in the substrates.

The Specificity for the P3′ Position in C-Terminal Processing Depends on the Phe Residue in the P2 Position.

Why does the specificity for the P3′ residue differ between the N- and C-terminal autoprocessing reactions? In the crystal structure (Fig. 1C), the active site of the enzyme interacts with the residues from P4 to P3′ (from Val303 to Phe309), and this region must include the structural feature that causes the P3′ specificity. Actually, Phe at the P2 position is involved in the S3′ subsite formation (Fig. 3D). Usually, the terms “pocket” and “subsite” both refer to some region in the enzyme; however, hereinafter we use the term “pocket” to indicate a hollow consisting only of enzyme residues, complementary to (a part of) the substrate, and the term “subsite” regardless of whether it consists only of parts of the enzyme or contains some part(s) of the substrate. Among the 11 3CL^pro processing sites in the SARS-CoV polyproteins, only the C-terminal autoprocessing site of 3CL^pro has Phe at the P2 position, whereas the rest have aliphatic amino acid residues (eight Leu, one Val, and one Met) (Fig. 1D). When the P2 Phe (position 305) in the C-terminal autoprocessing site of the substrate was converted to Leu, the F-to-A/N conversions at the P3′ position (position 309) reduced the apparent k_cat/K_M to a much smaller extent compared with the parent P2 Phe substrate (FF to FA/FN vs. LF to LA/LN in Fig. 2D).

Fig. 3.

Structural basis for the dual specificity. (A) Binding of the substrate peptide portion containing the C-terminal processing site. The substrate is depicted by a stick model, and the enzyme is shown as a surface presentation. The carbon atoms are colored as indicated, and the nitrogen and oxygen atoms are in blue and red, respectively. The cyan stick model represents the C-terminal portion of 3CL^pro, and the magenta stick model depicts the C-terminal prosequence. The surface presentation of the flexible heptapeptide TAEDMLN is in orange. The surface presentation of Gln189 is in green, whereas that of the rest of the enzyme (C145A) is in gray. Dotted lines represent CH–π interactions (18) between Met49 and P2 Phe and between Ala46 and P3′ Phe. Although the active Cys145 was mutated to Ala, it is shown in yellow in this figure. (B) Binding of the substrate containing the N-terminal processing site to the H41A mutant enzyme (17) (PDB ID code 2Q6G). (C and D) The positions of P2 Phe (C) and P3′ Phe (D) in the C-terminal processing site on the enzyme. (E and F) The positions of P2 Leu (E) and P3′ Phe (F) in the N-terminal processing site on the enzyme. The surfaces of the P2 and P3′ residues are depicted by space-filling models. (G and H) Schematic drawings of the interactions between the S2 pocket/ P2 position and the S3′ pocket/ P3′ position. (G) When P2 = Phe, the side chains of Gln189 and Met49 move outward, thereby providing sufficient space to accommodate P2 Phe. (H) When P2 = Leu, the side chains of Met49 and Gln189 move to narrow the S2 pocket, and thus the S3′ pocket is not formed.

The overall mechanism of this dual specificity can be explained based on the 3D structures of the enzyme. Our proform structure is that of the C-terminal substrate-bound enzyme with Phe at the P2 position (Fig. 3A). Xue et al. (17) reported the structure of the N-terminal substrate-bound enzyme with Leu at the P2 position (Fig. 3B). When P2 is Phe, in addition to the S2 pocket, the S3′ subsite is formed to accommodate the P3′ Phe (Fig. 3 A and D). The N-terminal processing sequence of 3CL^pro also has Phe at the P3′ position, but because it has Leu at the P2 position, the side chain of the P3′ Phe is not recognized (Fig. 3 B and F). In both structures, Gln at the P1 position is tightly bound in the S1 pocket, as reported previously (16, 17).

Formation of the S2 and S3′ Subsites and Their Interactions with the P2 and P3′ Residues in the C-Terminal Processing.

The structural differences in the C- and N-terminal substrate-bound structures are caused mainly by the large variations in (i) the side chain χ₁ and χ₂ angles of Gln189, (ii) the ψ angle of Asp48 in the “TAEDMLN loop” (positions 45–51), and (iii) the ψ angle of the P2′ residue (Fig. S5 A and B), as described in detail below. When Phe is in the P2 position (Fig. 3A and Fig. S5A), the side chains of Gln189 and Met49 are shifted outward, and thus there is sufficient space to accommodate the P2 Phe (Fig. 3 A and C and Fig. S5 A and C). In contrast, when Leu is in the P2 position (Fig. 3B and Fig. S5B), the side chain of Met49 is extended and lies along the side chain of the P2 Leu, and thus is tightly held by the resulting narrow S2 pocket (Fig. 3 B and E and Fig. S5 B and E).

Fig. S5.

Conformational differences between the C-terminal substrate sequence-bound form and the N-terminal substrate sequence-bound form. (A and B) The C-terminal substrate (A) and the N-terminal substrate (B) are indicated by stick models. The P1, P2, P3, P4, and P5 residues are shown in blue, and the P1′, P2′, and P3′ residues are in pink. Important local movements of the side chains of Gln189 (green) and Met49 (orange) and rotations about the ψ angles of Asp48 (magenta) and P2′ (pink) are indicated with arrows. Dotted lines in A represent CH–π interactions between Met49 and P2 Phe and between Ala46 and P3′ Phe. (C–F) Stick representations corresponding to Fig. 3 C–F. (G and H) The ϕ and ψ angles of each residue in the flexible heptapeptide TAEDMLN (orange), in the C-terminal substrate sequence-bound form (A) and the N-terminal substrate sequence-bound form (B). (I and J) Anchoring of the main chain of the substrate sequence (blue and pink) by the enzyme moiety (gray) in the C-terminal substrate sequence-bound form (A) and the N-terminal substrate sequence-bound form (B). Dashed lines indicate hydrogen bonds, and the associated figures indicate their distances in Angstroms.

The detailed mechanisms for the formation of the S2 and S3′ subsites are as follows. The Gln189 side chain movement is local (green in Fig. S5 A and B). In contrast, the Met49 side chain movement is coupled with large movements of the TAEDMLN residues (orange in Fig. S5 A and B), resulting in the formation of a short 3₁₀ helix. The TAEDMLN loop is considered inherently flexible and tends to form a 3₁₀ helix, with such conformational differences observed among previously reported crystal structures (2, 5–12). Even in the present crystal structure of the mature form without any substrate peptide, this TAEDMLN loop region forms a short 3₁₀ helix in one subunit (yellow in Fig. S1B), but is not well ordered in the other subunit (red in Fig. S1B). This conversion is induced mainly by the rotation about a single bond, i.e., the change in the ψ angle of Asp48 (Fig. S5 G and H). This results in formation of the S2 subsite for the P2 Phe (Fig. 3C and Fig. S5C). The S2 subsite for the P2 Phe consists of His41, Met49–Tyr54, His164, Met165, Asp187–Gln189, and the P3′ residue of the substrate (Fig. 3C and Fig. S5C). The binding of the P2 Phe to the S2 subsite results in formation of the S3′ pocket, consisting of Thr25, His41, Cys44–Ala46, Met49, Phe (P2), and Gly (P1′) (Fig. 3D and Fig. S5D). Of note, two residues of the peptide substrate, Phe (P2) and Gly (P1′), are components of the S3′ subsite, where the location of Gly (P1′) is fixed by Gln (P1) binding. In this manner, the P2 Phe and the P3′ Phe are recognized and bound to the enzyme. In both positions, CH–π interactions (21) occur between the CH₃ group of Met49 and the phenyl group of Phe (P2) and between the CH₃ group of Ala46 and the phenyl group of Phe (P3′) (Fig. 3A and Fig. S5A).

A comparison of Fig. 3A and Fig. 3B reveals that the conformations of the Phe (P2) and Leu (P2) substrates, respectively, are quite different; the C-terminal portion of the Phe (P2) substrate is bent, allowing the P3′ Phe to bind to the S3′ pocket. In fact, the major conformational difference of the Phe (P2) substrate is simply the rotation about the ψ angle of the P2′ residue (Fig. S5 A and I), compared with the Leu (P2) substrate (Fig. S5 B and J).

Dual Specificity.

As described above, SARS-CoV 3CL^pro recognizes the N- and C-terminal autoprocessing sites with different manners and specificities. On the basis of this dual specificity of SARS-CoV 3CL^pro, along with previous reports on the N-terminal autoprocessing site (3, 4), the different roles of Phe at the P3′ position between the N- and C-terminal processing sites are discussed in SI Text. The dual specificity cannot be expressed as a single “consensus sequence,” which is based on the assumption of the “AND” linkage of the recognition of each substrate residue (Pi) by the corresponding subsite (Si) (i = …4, 3, 2, 1, 1′, 2′, 3′, 4′…). In contrast, this cooperativity between subsites S2 and S3′ of SARS-CoV 3CL^pro can be expressed as another logical operation, “IMP” (implication), rather than as “AND,” as described in detail in Table 1 and SI Text.

Table 1.

The logical operation between the P2 and P3′ positions

A (P2 = Phe?)	B (P3′ = Phe?)	A IMP B (Cleavage?)
True	True	True
True	False	False
False	True	True
False	False	True

A: Amino acid residue in P2 position. True: Phe; false: Leu (and presumably Met, Cys).

B: Amino acid residue in P3′ position. True: Phe; false: Asn, Ala (presumably any amino acid).

A IMP B: efficient cleavage. True: yes; false: no.

Strategy of the Virus.

What is the advantage of this “alternative” specificity of SARS-CoV 3CL^pro? The P1 and P2 positions are considered the major recognition sites of 3CL^pro (Fig. 4A, Left) (1, 4, 18). In this study, we found a substrate recognition mode unique to the C-terminal autoprocessing of 3CL^pro (Fig. 4A, Right), in which the P3′ Phe is also recognized, owing to the P2 Phe (Fig. 4A). In the canonical specificity, this protease recognizes residues on the N-terminal side of the cleavage site (P1–P4), so that the C-terminal portion of the N-terminal-side product retains the main recognition sites. Therefore, the N-terminal side products may compete with the substrates, and better substrates would become stronger inhibitors after cleavage. There are 11 3CL^pro cleavage sites in the SARS-CoV polyproteins (1). The 3CL^pro cleavage products could become inhibitors of 3CL^pro, but except for the mature 3CL^pro, they can escape from the enzyme, and thus their inhibitory activities are inconsequential (Fig. 4B). If the C-terminal portion of the mature 3CL^pro had Leu at the P2 position, then the strong self-inhibitory activity would be a serious problem. Consequently, it seems reasonable that the alternative recognition pattern (P2 = Phe/P3′ = Phe) is used for the C-terminal processing of 3CL^pro (Fig. 4B), to minimize the inhibitory activity. After the cleavage, the P3′ Phe residue is separated from the P2 Phe residue, and thus the C-terminal portion of the enzyme no longer has sufficient binding affinity for the active site of the enzyme. This is the only site that uses this recognition pattern (P2 = Phe/P3′ = Phe) in the SARS-CoV polyproteins. This unique property is advantageous because it provides sufficient autoprocessing activities from the polyproteins while minimizing the inhibitory activities of the autoprocessed products.

Fig. 4.

Strategy for reducing the inhibitory effect of the C-terminal portion of the mature protease. (A) Two types of substrate recognition by SARS-CoV 3CL^pro. (B) Strategy of the virus for reducing the inhibitory effect of the C-terminal portion of its mature protease.

Competitive Inhibition of Proteolytic Activity by the Autoprocessed C-Terminal Region.

Indeed, the mature enzyme with the Phe305-to-Leu alteration (the P2 position of the C-terminal processing site) exhibited reduced cleavage activities for both the C terminus (Phe305/Phe309) and N terminus (Leu-2/Phe3) [F305L vs. wt in Fig. 2D (FF) and Fig. 2E, respectively]. This reduced activity likely occurs because the F305L mutant has a stronger inhibitory sequence (P2 = Leu) than the wild-type enzyme (P2 = Phe). In contrast to the mature form, the proforms with the C-terminal prosequence show no difference between P2 = Leu and P2 = Phe in C- and N-terminal proteolytic activities [F305L&Q306N vs. Q306N in Fig. 2D (FF) and Fig. 2E, respectively]. Thus, for the C-terminally unprocessed enzymes, there is no difference between the two systems, P2 = Leu and P2 = Phe/P3′ = Phe, with regard to both the cleavability (apparent k_cat/K_M) as a substrate (LF vs. FF in Fig. 2D) and the inhibitory effect against the proenzyme. After cleavage at this site, the mature P2 = Phe enzyme (wt) shows higher proteolytic activity than the mature P2 = Leu enzyme (F305L) (Fig. 2E), indicating that the C-terminal portion of the P2 = Phe (wt) enzyme has a lower inhibitory effect compared with the C-terminal portion of the P2 = Leu (F305L) variant (Fig. 4B).

We directly measured the inhibitory effect of the tetrapeptide derived from the C-terminal portion of the enzyme in a peptidase assay using an 11-aa residue fluorogenic peptide (Table 2). The tetrapeptide Val-Thr-Leu-Gln (P2 = Leu) had an 11-fold stronger inhibitory effect (K_i = 11.5 ± 6.0 μM) than the tetrapeptide Val-Thr-Phe-Gln (wt; P2 = Phe) (K_i = 126 ± 42 μM). This result provides further evidence of the reduced inhibitory effect of the C-terminal portion of the mature 3CL^pro by the presence of Phe instead of Leu at position P2. This difference might be crucially important in the maturation process of the polyprotein in the cell. Two polyprotein molecules produce one 3CL protease dimer, which cleaves 11 positions for one polyprotein (1). In this stoichiometry, the stronger total K_i value of the two inhibitory moieties at the two C-termini of the protease dimer compared with the K_M value for the substrates must matter, and this is the case if this moiety has the P2 = Leu sequence (K_i = 11.5 μM ± 6.0 μM vs. K_M = 39 ± 5.4 μM; Table 2). In contrast, the C-terminal moiety of the wild-type enzyme has lower affinity for the binding site than for the substrate (K_i = 126 ± 42 μM > K_M = 39 ± 5.4 μM), which ensures the reduced inhibitory effect of this region.

Table 2.

Kinetic parameters obtained using peptide substrates and peptide inhibitors

Parameter	Value
KM (NMA-TSAVLQSGFRK(DNP)-NH2)	39 ± 5.4 μM
kcat (NMA-TSAVLQSGFRK(DNP)-NH2)	0.37 ± 0.040 s−1
Ki (VTFQ)	126 ± 42 μM
Ki (VTLQ)	11.5 ± 6.0 μM
Ki (VTAQ)	18.4 ± 9.3 μM

As shown in Fig. 2D, the difference in K_i value causes an approximate twofold increase in the apparent k_cat/K_M (wt vs. F305L) even at low enzyme concentrations (i.e., low [I]) in this experiment (Figs. S3 and S4). The difference in the inhibitory effect between Phe and Leu at the P2 position of the C terminus on the apparent k_cat/K_M (cleavability) must be greater in the cells. When the polyproteins are expressed in mammalian cells, the 3CL^pro moiety (or NSP5) exists between two membrane proteins, NSP4 and NSP6. The 3CL^pro moiety and the C and N termini of NSP4 and NSP6, respectively, are on the cytoplasmic side of the endoplasmic reticulum (22, 23). Moreover, 3CL^pro forms a dimer. Therefore, the local concentration of the C-terminal inhibitory sequence of 3CL^pro on the endoplasmic reticulum likely would be high.

This mechanism appears to exist only in the SARS-CoV 3CL protease. However, there may be different types of mechanisms for reducing the inhibitory effects of the C-termini of the mature autoprocessing proteases in other viruses.

SI Text

Dual Specificities of the P3′ Residue Depending on the P2 Residue.

The structures and mechanism described in the main text explain the previously reported biochemical results (4) indicating that Leu, Met, and Phe are preferred for the P2 position over β-branched residues, such as Ile and Val (4). It must be noted that the substrates in the previous report had sequences derived from the N-terminal processing site (4), so Phe was present at the P3′ position. In this context, both substrates with the P2 Leu and the P2 Met likely are bound to the enzyme, as in the structure presented in Fig. 3B, whereas the substrate with the P2 Phe should bind to the enzyme, as in the structure shown in Fig. 3A.

When the enzyme assumes the conformation with Leu at the P2 position (Fig. 3B), the S2 pocket snugly interacts with the entire side chain, from the β methylene (-CH₂-) group to the two δ methyl groups (Fig. 3 E and H). In contrast, in the enzyme conformation with Phe at the P2 position, the presence of the phenyl group prevents the enzyme from snugly interacting by itself with the entire P2 side chain. Thus, the second phenyl group of Phe at the P3′ position hydrophobically participates in the interaction (Fig. 3 C and G). The S3′ subsite specific to the P3′ Phe is created only when P2 is Phe (Fig. 3C), by using the P2 Phe as part of the S3′ subsite (Fig. 3D, blue). These mechanisms elegantly explain why Phe is required at the P3′ position when the P2 position is occupied by Phe, rather than by Leu or Met.

It seems peculiar that the P3′ residue of the N-terminal processing site of the enzyme is the same Phe as in the C-terminal processing site, although this residue is not recognized for cleavage. Only two processing sites in the polyproteins of SARS-CoV have a Phe residue at the P3′ position, and these are the N- and C-terminal autoprocessing sites of the protease. During 3CL^pro maturation, one monomer can perform the N-terminal processing of another precursor by forming a dimer in a different mode from that of the normal 3CL^pro dimer (34). Furthermore, dimerized precursors can process the N termini of other precursor dimers or monomers (19). The present results suggest that in these N-terminal processing processes, the Phe residue at the P3′ position of the processing site is not required, because of the presence of Leu at the P2 position. In contrast, Phe3 is required for the structure of the mature enzyme after the cleavage; the Phe3 phenyl group of the mature enzyme, derived from the P3′ residue in the profrom, is buried in the protomer and supports the N finger (N-terminal residues 1–7), which plays an important role in dimerization and active site formation (2).

Subsite Cooperativity and the Logical Operation.

The cooperativity between two subsites (i.e., subsite cooperativity) revealed in the present study was not detected by conventional statistical methods, such as the consensus sequence analysis represented by the logo plot (Fig. 1E). Some apparent subsite cooperativities in other proteases have been described phenomenologically (35, 36). In contrast, the present study has identified the molecular mechanism on the basis of the 3D structure of the enzyme and substrates. We attempted to describe this subsite cooperativity between the S2 subsite and the P3′ subsite in the SARS-CoV 3CL protease using a logical operation. On the one hand, it cannot be expressed by the “AND” operation assuming the one-to-one correspondence between residue (Pi) by subsite (Si) (i = … 4, 3, 2, 1, 1′, 2′, 3′, 4′ …). On the other hand, the unique subsite cooperativity can be expressed by another simple logical operation, “IMP” (implication, “⇒”), as P2(Phe) IMP P3′ (Phe) (Table 1). This “IMP” operator returns FALSE if and only if its first operand (the P2 position) is TRUE and its second operand (the P3′ position) is FALSE, whereas in all other cases it returns TRUE. In this expression, for the P2 position (the first operand), TRUE means Phe, and FALSE means Leu; for the P3′ position, TRUE means Phe, and FALSE means Ala or Asn; for the return value, TRUE means “effective cleavage,” and FALSE means “ineffective cleavage.” Although these TRUE and FALSE assignments are based on our experiments (Fig. 2), the FALSE value may be expanded by considering the structures of the S2 and S3′ pockets (Fig. 3 C–F) and the previously reported substrate specificity data (4). For the P2 position, FALSE means an aliphatic residue without the β-branch, whereas for the P3′ position, FALSE means any residue other than Phe.

Methods

Cell-Free Syntheses of SARS-CoV 3CL^pro Species and Their Substrates, and Assays of the Proteolytic Activities (Cleavabilities).

The mature and proforms of SARS-CoV 3CL^pro (the enzyme in Fig. 2A) and their substrates (the substrate in Fig. 2A) were synthesized by the E. coli cell-free protein synthesis method (19, 20). The proteolytic activities (trans-processing assays) were measured as reported previously (19) and as described briefly in SI Text and Fig. S3.

Purification of the Mature 3CL^pro.

The wild-type 3CL^pro was synthesized and automatically processed in the E. coli cell-free protein synthesis system, and was purified by successive chromatography steps on columns of Econo-Pack High Q (two tandemly connected 5-mL cartridges; Bio-Rad), Mono P 5/50 (GE Healthcare), and HiPrep 16/60 Sephacryl S-300 HR (GE Healthcare). From the 9-mL cell-free synthesis solution, approximately 5.8 mg of the purified 3CL^pro was obtained.

Preparation of the Proform of 3CL^pro with the C-Terminal Prosequence.

Using a QuikChange Mutagenesis Kit (Stratagene), the plasmid encoding wild-type 3CL^pro was modified at three points: the active site Cys145 was changed to Ala, the S-tag portion and the 10-aa extension at the N terminus were removed so that the first translated methionine was followed directly by the first Ser residue of 3CL^pro, and then the His-tag was removed. The resulting plasmid was introduced into E. coli strain BL21(DE3), and protein expression was induced by isopropyl β-d-1-thiogalactopyranoside. After ultrasonic disruption of the cells, the proform of 3CL^pro with the C-terminal extension was purified in the same manner as the mature form of 3CL^pro. The first translated methionine residue was completely removed in the E. coli cells, as determined by N-terminal amino acid sequencing and MALDI-TOF mass spectrometry, as described previously (19). From 200 mL of the E. coli culture, approximately 15.9 mg of the purified protein was obtained.

Crystallization and Data Collection.

Well-diffracting crystals of the mature form of SARS-CoV 3CL^pro were obtained in drops composed of 2 µL of the protein (11 mg/mL in 10 mM Tris⋅HCl pH 7.5 buffer containing 0.1 mM EDTA and 1 mM DTT) and 2 µL of reservoir solution, by the hanging drop vapor diffusion technique at 20 °C. The reservoir solution contained 0.1 M MES buffer (pH 5.8), 4% (wt/vol) PEG 6000, 3% (vol/vol) DMSO, and 1 mM DTT. Large, single crystals (0.3 mm in the longest dimension) appeared within 2 wk. The crystals were flash-cooled in a final cryoprotectant solution composed of the mother liquor and 25% (vol/vol) glycerol. Diffraction data were collected on an ADSC Quantum210 detector at beamline AR-NW12 in the Photon Factory (Tsukuba, Japan). The data were indexed, integrated, and scaled using HKL2000 (24). The crystal belongs to the space group P2₁, with unit cell dimensions a = 52.3 Å, b = 96.3 Å, and c = 67.8 Å, and diffraction data were obtained up to 1.7-Å resolution.

The crystal of the proform of 3CL^pro with the mutated active site C145A and the C-terminal prosequence was grown in drops composed of 2 µL of protein solution (10 mg/mL in 10 mM Tris⋅HCl pH 7.5 buffer, containing 0.1 mM EDTA and 1 mM DTT) and 2 µL of reservoir solution [0.1 M sodium chloride, 0.1 M HEPES-Na buffer pH 7.3, and 12% (wt/vol) PEG 4000] by the same method. Large, single crystals measuring 0.5 mm in the longest dimension appeared within 1 wk. Diffraction data were collected on an ADSC Quantum210 detector at beamline BL-5A in the Photon Factory. The data were indexed, integrated, and scaled using HKL2000 (24). This crystal belongs to the space group P2₁2₁2₁, with unit cell dimensions a = 63.7 Å, b = 90.2 Å, and c = 110.3 Å, and diffraction data were obtained up to 2.2-Å resolution.

Structure Determination and Refinement.

General processing of the scaled data was performed with the programs in the CCP4 suite (25). The phases were determined by the molecular replacement method with the Molrep program in the suite. As search models, the reported structure (PDB ID code 1UJ1) (2) was used for the mature form (PDB ID code 2DUC), and the mature-form structure was then used for the proform (PDB ID code 5B6O). The models were rebuilt manually using O (26), and were refined with CNS (27) and PHENIX (14). The structures were refined to an R-factor of 19.4% (R_free = 22.1%) at 1.7-Å resolution for SARS-3CL^pro-wild type (PDB ID code 2DUC) and to an R-factor of 21.8% (R_free = 25.9%) at 2.2-Å resolution for 3CL^pro-C145A-10aa (PDB ID code 5B6O). Structural alignments were accomplished with the DEJAVU (28), LSQMAN (29), and LSQKAB (30). The protein secondary structure was defined by the DSSP algorithm (31). The quality of the model was inspected with the PROCHECK program (32). Graphic figures were created with Pymol (33).

Kinetic Parameters Using Peptides.

The fluorogenic peptide NMA-TSAVLQSGFRK(DNP)-NH₂, which has the amino acid sequence of the N-terminal processing site of SARS-CoV 3CL^pro, was used as a substrate. Various concentrations of this substrate (6.25, 12.5, 25, or 50 μM) were cleaved with 25 nM 3CL^pro at 30 °C in 0.2 mL of a solution consisting of 20 mM Tris⋅HCl pH 7.5, 200 mM NaCl, 1 mM EDTA, 1 mM DTT, and 1% DMSO. Cleavage of the peptide bond between the Gln (Q) and Ser (S) residues was monitored with a Tecan Infinite F200 fluorescence microplate reader, with excitation at 380 nm and emission at 465 nm. From the double-reciprocal plot of the substrate concentration vs. the initial velocity of the cleavage reaction, the K_M and k_cat values were calculated.

The inhibition assay was performed with the same system, using the tetrapeptides VTFQ (ValThrPheGln), VTLQ (ValThrLeuGln), and VTAQ (ValThrAlaGln) at 0, 2.0, 2.2, 2.4, 2.6, and 2.8 mM concentrations, in 0.2 mL of a solution consisting of 1.5 μM (i.e., 1/26 of the K_M value) NMA-TSAVLQSGFRK(DNP)-NH₂, 20 mM Tris⋅HCl pH 7.5, 200 mM NaCl, 1 mM EDTA, 1 mM DTT, and 8% (vol/vol) DMSO. The ratio of the rates with/without an inhibitor tetrapeptide, v₀/v_i, was measured. From the slope of the plot of (v₀/v_i) − 1 against [I]₀, the 1/K_i(app) value was obtained.