Characterization of a Torovirus Main Proteinase
http://www.100md.com
病菌学杂志 2006年第8期
Virology Division, Department of Infectious Diseases and Immunology, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands
Molecular Virology Laboratory, Department of Medical Microbiology, Center of Infectious Diseases, Leiden University Medical Center, Leiden, The Netherlands
ABSTRACT
Viruses of the order Nidovirales encode huge replicase polyproteins. These are processed primarily by the chymotrypsin-like main proteinases (Mpros). So far, Mpros have been studied only for corona-, arteri-, and roniviruses. Here, we report the characterization of the Mpro of toroviruses, the fourth main Nidovirus branch. Comparative sequence analysis of polyprotein 1a of equine torovirus (EToV) strain Berne, identified a serine proteinase domain, flanked by hydrophobic regions. Heterologous expression of this domain resulted in autoprocessing at flanking cleavage sites. N-terminal sequence analysis of cleavage products tentatively identified FxxQ(S, A) as the substrate consensus sequence. EToV Mpro combines several traits of its closest relatives. It has a predicted three-domain structure, with two catalytic -barrel domains and an additional C-terminal domain of unknown function. With respect to substrate specificity, the EToV Mpro resembles its coronavirus homologue in its preference for P1-Gln, but its substrate-binding subsite, S1, more closely resembles that of arteri- and ronivirus Mpros, which prefer P1-Glu. Surprisingly, in contrast to the Mpros of corona- and roniviruses, but like that of arterivirus, the torovirus Mpro uses serine instead of cysteine as its principal nucleophile. Under the premise that the Mpros of corona- and toroviruses are more closely related to each other than to those of arteri- and roniviruses, the transition from serine- to cysteine-based proteolytic catalysis (or vice versa) must have happened more than once in the course of nidovirus evolution. In this respect, it is of interest that a mutant EToV Mpro with a Ser165Cys substitution retained partial enzymatic activity.
INTRODUCTION
Many positive-strand RNA viruses have adopted a "polyprotein" strategy to express (part of) their genome. Translation of viral RNA results in the synthesis of large multidomain protein precursors, which are proteolytically cleaved to yield a collection of intermediate and mature products. As a rule, polyproteins that are to yield secretory or membrane (glyco)proteins are processed by host-encoded proteinases within the lumen of the compartments of the exocytotic pathway (in particular the endoplasmic reticulum and the trans-Golgi network). Cytosolic processing, however, involves co- and posttranslational cleavage by viral proteinases that are present as distinct domains within the polyprotein itself (51, 72). Processing is often a prerequisite for the assembly of the viral RNA replicase/transcriptase. To add to the complexity, intermediate and mature products may have distinct functions in replication and may be required during different stages of the infection. Hence, the properties of the viral proteinases must be carefully tailored to ensure accurate and efficient cleavage in a substrate-specific and temporally well-coordinated fashion.
Many RNA virus proteinases resemble, in their mechanism of peptide hydrolysis and in their three-dimensional structure, either one of the prototypical cellular proteases papain and chymotrypsin. Papain and related proteinases are characterized by a two-domain +-fold with typically a cysteine and downstream histidine as the catalytic residues. Chymotrypsin has a distinctive two--barrel structure in which the principal nucleophile, a serine residue, is part of a catalytic His-Asp-Ser triad (20, 49). Chymotrypsin-like serine proteinase domains have been identified within the polyproteins of among others flavi-, alpha-, luteo-, astro-, and sobemoviruses (13, 17, 29, 38, 42, 53, 66).Intriguingly, however, many viruses, for instance picorna-, como-, poty-, and caliciviruses, encode proteinases with a (predicted) chymotrypsin-like fold in which, instead of serine, a cysteine acts as the main catalytic residue (1, 5, 6, 27, 28, 31, 52, 55).
The order Nidovirales comprises a group of evolutionarily related enveloped (positive) stranded RNA viruses (corona-, toro-, arteri-, and roniviruses) (10, 15, 19, 26, 56, 58, 62), which have taken the polyprotein strategy to the extreme. The 5'-most two-thirds of the nidovirus genome is occupied by two large overlapping open reading frames (ORFs), ORF1a and ORF1b, the most downstream of which is only translated upon ribosomal frameshifting (9, 15, 18, 58). These ORFs specify two huge polyproteins, pp1a and the C-terminally extended frameshift protein pp1ab. In the case of the coronaviruses, pp1ab measures up to 7,200 amino acids in length, representing the largest polyprotein known to date. Among nidoviruses, the largest degree of sequence conservation is found within the ORF1b-encoded part of pp1ab, which contains the chief replicative domains (15, 19, 30, 57), i.e., the RNA dependent RNA polymerase (RdRp) and the helicase, as well as up to three additional domains, ExoN, NendoU, and 2'-O-methyltransferase (2'-O-MT), which display sequence identity to cellular RNA-modifying enzymes (8, 37, 57). Of the latter, the ExoN and 2'-O-MT domains are lacking in the arterivirus pp1ab.
Processing of the arteriviral and coronaviral polyproteins gives rise to 12 to 13 and 15 to 16 mature products, respectively, and an as yet unknown number of functional intermediates (57, 76). The N termini of the replicase polyproteins are processed by "accessory" papain-like proteinases (31). The vast majority of cleavages, however, are carried out by a single chymotrypsin-like main proteinase (Mpro), which is located in the C-terminal half of pp1a, flanked by hydrophobic regions that are thought to be involved in targeting the replicase to intracellular membranes (30, 54, 67, 71). Recent X-ray crystallographic analysis of the structures of corona- and arterivirus Mpros confirmed the predicted chymotrypsin-like two--barrel fold (2, 4, 73). Interestingly, the corona- and arteriviral enzymes both possess an additional C-terminal domain, a feature not commonly seen in other viral and cellular chymotrypsin-like proteinases (2, 4). Also for the ronivirus Mpro, such an additional C-terminal domain was predicted (74).
One might surmise that the nidoviral Mpros, given their pivotal role in viral replication, are well conserved. Conversely, however, the primary sequences of the Mpros of corona-, roni-, and arteriviruses have diverged almost beyond recognition, and sequence identity is restricted to the regions immediately surrounding the main catalytic residues. Moreover, whereas the arteriviral Mpro is a serine proteinase with a canonical His-Asp-Ser catalytic triad (4, 64), the Mpros of the corona- and roniviruses are cysteine proteinases (28, 30, 43, 74). In the latter, the acidic catalytic residues have been lost and, apparently, these enzymes solely rely on the remaining catalytic His-Cys dyad (2, 73, 74).
Toroviruses have been poorly studied, but because of their relatively close phylogenetic relationship to coronaviruses they may in many respects represent "missing links" in comparative nidovirus studies. Here, we are the first to report the biochemical characterization of a torovirus Mpro. Upon analysis of the C-terminal half of pp1a of equine torovirus (EToV) strain Berne, we identified a chymotrypsin-like proteinase domain, which, like the Mpros of the arteri-, corona-, and roniviruses, is flanked by hydrophobic regions. When expressed in prokaryotic and mammalian cells, the proteinase domain autocatalytically released itself from its surrounding sequences. Surprisingly, the toroviral Mpro seems no more related to that of coronaviruses than to the Mpros of arteri- and roniviruses. Most saliently, in contrast to the Mpros of corona- and roniviruses, but like that of arteriviruses, the torovirus Mpro is a serine proteinase. A tentative processing scheme for the torovirus pp1a/pp1ab polyproteins is presented, and the implications of our findings for nidovirus evolution are discussed.
MATERIALS AND METHODS
Cells and virus. Equine dermis (Ederm) cells (American Type Tissue Culture Collection) and OST7-1 cells (23) were maintained as monolayers in Dulbecco's modified Eagle's medium (DMEM; BioWhitaker) supplemented with penicillin (100 IU/ml) and streptomycin (100 μg/ml) and 20% (DMEM20) or 10% (DMEM10) fetal calf serum, respectively. EToV strain Berne was propagated in Ederm cells as described previously (68). Recombinant vaccinia virus vTF7-3 (24) was obtained from B. Moss (NIH, Bethesda, MD).
Isolation of intracellular EToV RNA, cDNA synthesis, and PCR amplification. Total cytoplasmic RNA, isolated from the equivalent of 4 x 106 EToV-infected Ederm cells (68), was subjected to reverse transcription-PCR (RT-PCR) with RNase H-free Moloney murine leukemia virus reverse transcriptase (Superscript II; Invitrogen) and with the Expand long template PCR system (Roche Diagnostics GmbH), according to the instructions of the manufacturers. Oligonucleotide primers 1071 and 1072 (Table 1) were designed after sequences located at the 5' end and 3' end of EToV ORF1A, respectively (59, 61). Primers 1288 and 1325 were designed after sequences from cDNA clones p133 and p135. These clones had been identified previously as EToV specific after screening of a genomic cDNA library (61) and had been tentatively mapped in ORF1a (E. J. Snijder, unpublished). Amplicons of approximately 5.5 kb were obtained, which were gel purified, blunt ended using large fragment DNA polymerase I (LFDP) according to the instructions of the manufacturer (Invitrogen), and cloned into an EcoRV-digested vector pWSK29 (70). Sequence analysis was performed commercially (BaseClear Labservices) using the ABI PRISM BigDye Terminators v3.0 cycle sequencing kit (Applied Biosystems) on at least two independent clones in both orientations.
Preparation of EToV Mpro-specific antisera. The coding region for EToV Mpro amino acids (aa) 5 to 196 (as counted from the Mpro N terminus) was RT-PCR amplified with oligonucleotide primers 1548 and 1549 (Table 1). The resulting PCR product was digested with BamHI and EcoRV and inserted into HindIII (blunt ended)-BamHI-digested pQE9 (QIAGEN), yielding plasmid pQE-SS-Mpro(5-196). This expression vector, encoding a 22-kDa Mpro fragment fused at the N terminus to a His6 tag, was introduced in Escherichia coli M15/pREP4 cells (QIAGEN). Expression of the recombinant protein was induced by adding IPTG (isopropyl--D-thiogalactopyranoside) to a final concentration of 1 mM to bacterial cultures during exponential growth (optical density [OD] of 0.6). Bacteria were lysed 5 h after induction, and the His-tagged expression product was purified by Ni-nitrilotriacetic acid affinity chromatography using the XPRESS System version 2.0 (Invitrogen) according to the instructions of the manufacturer. The recombinant protein was eluted with a mixture of 20 mM sodium phosphate, 8 M urea, and 500 mM sodium chloride (pH 6.0) containing 500 mM imidazole. The eluate was concentrated on a Centricon concentrator (30-kDa cutoff; Amicon), dialyzed for 6 h against phosphate-buffered saline (PBS)-0.05% sodium dodecyl sulfate (SDS) and, finally, dialyzed for 16 h against PBS-0.02% SDS. New Zealand White rabbits were immunized subcutaneously with 100 μg of His-Mpro 1548/1549 fusion protein as described previously (12); the resulting antiserum was designated K158.
Antisera specific for the N terminus of EToV Mpro (K270), the C terminus of the Mpro (K272), or the N terminus of pp1a/pp1ab (K274) were prepared by repeated subcutaneous immunization of New Zealand White rabbits with 200 μg of the bovine serum albumin-coupled synthetic peptides NH2-SVFSKATSPFTLHARPPK-COOH, NH2-KPLQYFHVPSFWQPFKKQ-COOH, and NH2-MFRILKNNTRETEQHLSSSK-COOH, respectively, as described previously (63).
Construction of prokaryotic and eukaryotic EToV Mpro expression vectors. For prokaryotic expression of EToV Mpro, residues 10376 to 11472 of the EToV genome were RT-PCR amplified with oligonucleotide primers 1606 and 1607 (Table 1). A product of the anticipated size of 1,112 bp was obtained, which was gel purified, inserted into plasmid pQE9, and cloned in E. coli M15/pREP4 cells (QIAGEN) as described above. The resulting expression vector, pQE-SS-Mpro, codes for a 42-kDa polypeptide, encompassing the EToV Mpro domain and flanking sequences.
The expression construct pQE9-SS-Mpro-GST is a pQE-SS Mpro derivative, which consists of nucleotides (nt) 10376 to 11473 of EToV ORF1a and nt 258 to 935 of the glutathione S-transferase (GST) gene from plasmid pGEX2T (Amersham Pharmacia Biotech). This expression cassette was produced via PCR with oligonucleotides 2124 and 2125 and conventional cloning procedures. It codes for a 60-kDa fusion protein consisting of the 366-amino-acid-residue Mpro domain with flanking sequences linked by a proline residue to amino acid residues 1 to 226 of the GST protein.
For mammalian expression, we employed the recombinant vaccinia virus-based vTF7-3 system. To generate a suitable expression vector, the 1606/1607 PCR amplicon, digested with BamHI and EcoRV, was cloned into SacI/BamHI-digested pBluescript K/S+ (Strategene) together with a double-stranded linker, consisting of oligonucleotides 1721 and 1722. The resulting plasmid, pBS-SS-Mpro, contains residues 10376 to 11472 of the EToV genome, placed under the control of the bacteriophage T7 promoter and provided in frame with an initiation codon in an optimal context for translation initiation (40).
Site-directed mutagenesis of (putative) active-site residues was performed by splicing overlap extension-PCR mutagenesis (35) and conventional cloning techniques; mutations introduced in Mpro and the primers used are listed in Table 1.
The dual-expression construct pBS-SS-Mpro-dual, is a pBS-SS-Mpro derivative in which, downstream of the Mpro sequences, a second expression cassette is inserted, which is placed under the translational control of the encephalomyocarditis virus (EMCV) internal ribosomal entry site (IRES; taken from plasmid pTN2) (69) and which consists of nt 821 to 1128 of EToV ORF1a fused in frame to nt 10.378 to 11.472 of the EToV genome (encoding Mpro and surrounding sequences) and nt 3 to 483 of the EToV N gene. This chimeric expression cassette was produced via a series of splicing overlap extension-PCRs with oligonucleotides 1638, 1640, 1902, and 1914 through 1918 (Table 1) and conventional cloning procedures. It codes for a 66-kDa fusion protein, consisting as listed from the N terminus to C terminus of (i) residues 1 to 103 of pp1a; (ii) the 65 aa residues immediately upstream of Mpro; (iii) the 292-residue Mpro domain; (iv) the 8 residues downstream of Mpro, including the cleavage site; and (v) residues 2 to 160 of the EToV N protein. In this fusion protein, the N-terminal Mpro cleavage site is abolished (QSAG) and the Mpro proteinase is inactivated by an active site Ser-to-Ala substitution.
Purification and N-terminal sequence analysis of Mpro cleavage products. EToV Mpro and Mpro-GST were expressed from plasmids pQE-SS-Mpro and pQE9-SS-Mpro-GST, respectively, in E. coli M15/pREP4 as described above. Of the 1-ml bacterial culture, the cells were collected by centrifugation at 14,000 rpm for 30 s, resuspended in 100 μl Laemmli sample buffer, and incubated for 5 min at 95°C. Proteins were separated in 15% SDS-polyacrylamide gel electrophoresis (PAGE) gels (20 μl cleared supernatant/lane), transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories), and subsequently stained with Coomassie brilliant blue R-250. The regions containing the cleavage products of interest were excised. N-terminal sequence analysis was performed by subjecting the proteins to 5 to 8 rounds of Edman degradation using standard procedures with a pulse-liquid protein sequencer (model 476A; Applied Biosystems).
Infection/transfection of mammalian cells and metabolic labeling. Monolayers of Ederm cells (1 x 106 cells), grown in 35-mm wells, were washed once with PBS containing 50 μg/ml DEAE-dextran and then infected in PBS-DEAE with EToV at a multiplicity of infection of 5 PFU/cell. At 1 h postinfection (p.i.), the inoculum was replaced by 1 ml DMEM10 and incubation at 37°C continued.
Subconfluent monolayers of OST7-1 cells (2 x 106 cells), grown in 35-mm wells, were infected with recombinant vaccinia virus vTF7-3 and, at 1 h p.i., transfected with 2 μg plasmid DNA as described previously (46).
Prior to metabolic labeling, the cells were depleted for methionine and cysteine for 30 min by replacing the tissue culture supernatant with 600 μl of prewarmed starvation medium (DMEM without L-cysteine and L-methionine [Invitrogen] supplemented with 5% FCS, 10 mM hydroxyethylpiperazine-N'-2-ethanesulfonic acid [HEPES; pH 7.4], 1 mM Glutamax). Then 100 μCi of Tran35S-label (ICN) was added and the cells were labeled for the indicated length of time at 37°C.
After labeling, the cells were washed once with ice-cold PBS and then lysed on ice in 600 μl ice-cold lysis buffer (20 mM Tris-HCl, pH 7.5, 1 mM EDTA, 100 mM NaCl, 1% Triton X-100, and 1 μg/ml aprotinin, leupeptin, and pepstatin A each). Nuclei and cell debris were pelleted by centrifugation for 5 min at 14,000 rpm and 4°C, and the supernatant was subjected to radioimmunoprecipitation assay (RIPA). To this end, the lysates were adjusted to 2% SDS, diluted 10 times with detergent solution (50 mM Tris-HCl, pH 8.0, 62.5 mM EDTA, 0.4% Na deoxycholate, 1% NP-40, and 1 μg/ml of aprotinin, leupeptin, and pepstatin A each). RIPA was performed in a final volume of 1 ml with 3 μl of antiserum K158, K270, K272, K274 (see below), or R-EToV N (14). The immune complexes were adsorbed to Pansorbin cells (Calbiochem) for 30 min at 4°C, collected by centrifugation for 1 min at 14,000 rpm, and washed three times with detergent solution containing 0.25% SDS. The pellets were then resuspended in 25 μl Laemmli sample buffer and heated for 5 min at 95°C. Proteins were separated in 15% SDS-PAGE gels, which were fixed for 30 min in 50% methanol-10% acetic acid and dried for 60 min at 80°C. Labeled proteins were visualized by fluorography using Kodak intensifying screens.
RESULTS
Comparative sequence analysis of EToV Mpro. The EToV Berne ORF1a was sequenced completely (GenBank accession no. DQ310701) and encodes a product, pp1a, 4,569 amino acids long; a ribosomal frameshift at the pseudoknot at the 3' end of ORF1a (58) would produce a C-terminally extended product, pp1ab, of 6,857 amino acids. During the preparation of our article, Draker et al. (22) reported the completion of the sequence of the bovine torovirus (BToV) strain Breda genome. These authors identified a BToV pp1a domain with sequence similarity to the adenosine diphosphate-ribose 1"-phosphatase processing enzyme family (22), a homologue of which also occurs at a similar topological position in pp1a of coronaviruses, but not in those of roni -and arteriviruses (57). Moreover, two proteinase domains were found: a cysteine proteinase within the N-terminal half of pp1a and a more C-terminally located serine proteinase (22). The pp1a polyproteins of EToV and BToV share 65% sequence identity and the three domains described by Draker and colleagures (22) are conserved in pp1a of EToV. They were in fact identified independently using standard protein-protein and PSI and PHI BLAST searches of the NCBI database with EToV pp1a sequences as a query.
The EToV serine proteinase domain, comprising residues 3278 to 3430, displayed sequence similarity with the SpIB serine protease of Staphylococcus aureus and with the pp1a serine proteinase of turkey astrovirus. Similarity was particularly evident in the areas surrounding the predicted active-site residues, with His75, Asp113, and Ser193 of S. aureus SplB (50); GenBank accession no. AF271715) aligning with His3304, Glu3347, and Ser3416 in the EToV sequence (Fig. 1A). The putative EToV proteinase domain resides within pp1a/pp1ab at a relative position similar to that of the Mpros of other nidoviruses and, likewise, is flanked by hydrophobic regions (HR-1 and -2; Fig. 2A). Based upon these observations and upon the fact that no other chymotrypsin-like proteinase motifs were evident in the EToV pp1a/pp1ab sequence, we postulated the identified domain to represent the toroviral Mpro.
Although sequence conservation between the putative torovirus Mpro and the Mpros of other nidoviruses is very weak, its sequence could be fitted to existing Mpro alignments (3, 4, 74) in the regions containing the active-site His and Cys/Ser residues. The putative EToV Mpro is predicted to adopt a two--barrel structure with His3304 and Ser3416 as the main active-site residues and, again, Glu3347 as the most likely candidate for the role of third residue within the catalytic triad (Fig. 1B). From our data, the torovirus Mpro appears to be no more closely related to the coronavirus Mpro than to its roni- and arterivirus homologues. Perhaps most surprising, the torovirus Mpro differs from the Mpros of corona- and roniviruses in that it is not a cysteine proteinase, but rather, like its arterivirus homologue, uses serine as the principal nucleophile.
Proteolytic activity of EToV Mpro. The bioinformatical analyses of the putative EToV Mpro provided a theoretical framework for subsequent experimental studies. The hydrophilic region between HR-1 and -2, which spans the predicted proteinase domain, was expressed in E. coli (Fig. 2). A construct in which the predicted active-site His residue was replaced by Gly was used as a control (Fig. 2B). IPTG-induced expression of these constructs should yield a 42-kDa primary translation product. Indeed, a protein of this size was detected in cells expressing the His-to-Ala mutant. However, in cells expressing the intact proteinase domain, two products with apparent molecular masses of 33 and 16 kDa were found instead. These results suggested that the EToV pp1a region comprising residues 3186 to 3550 indeed contains a functional proteinase domain mediating autocatalytic processing.
The 33- and 16-kDa products were purified from E. coli and N-terminally sequenced by performing 5 or 8 cycles of Edman degradation, respectively. Analysis of the 33-kDa product indicated that cleavage had occurred at the sequence 3246SNFSFQSVFSKAT3258, thus identifying Ser3252 as the N-terminal residue of EToV Mpro (Fig. 3) (Note that, throughout this article, amino acid residues flanking the scissile bond, indicated by , are given from N to C terminus in the single-letter code, where "x" indicates any residue.) The 16-kDa product proved to be a mixture of two protein species. For one of these, the N terminus was identical to that of the 33-kDa product. The other product apparently arose from proteolytic cleavage at an internal sequence, 3392SEFATQAWQTVN3403(data not shown). Indeed, cleavage of Mpro at this site would cut the proteinase effectively in half, thus giving rise to two equal-size (16.5 kDa) products. These results tentatively identified FxxQ(S, A) as the EToV Mpro substrate consensus sequence (residues are given in parentheses to indicate the variation occurring at a particular position within the cleavage site).
Analysis of the Mpro flanking sequences identified a putative additional cleavage site downstream of the proteinase domain, 3540FKKQS3544. In the full-length precursor product encoded by plasmid pQE-SS-Mpro, this sequence is located only 5 residues upstream of the C terminus. Hence, cleavage at this site could not be confirmed. As the hydrophobic region HR-2, C-terminally flanking Mpro, might interfere with efficient expression in E. coli, we did not extend the expression product with EToV sequences, but instead designed an Mpro-fusion protein, which contained the predicted cleavage site including the downstream seven residues, linked via a newly introduced proline residue to residues 1 to 226 of GST (Fig. 2A). Expression in E. coli yielded, as anticipated, two major products of 33 kDa and 26 kDa (Fig. 2B), apparently corresponding to mature Mpro and the cleaved-off GST moiety. N-terminal sequence analysis of the 26-kDa product confirmed this notion and indicated that proteolytic cleavage indeed had occurred at the predicted site 3543GlnSer3544 (data not shown).
Cleavage at the sites upstream and downstream of the Mpro domain should yield a 33-kDa protein. Accordingly, in lysates from metabolically labeled EToV-infected cells, a product of this size, comigrating exactly with the bacterial cleavage product, was detected by RIPA with Mpro-specific antisera (Fig. 4) (data not shown). The 16-kDa products, however, were not detected in EToV-infected cells.
Mutagenesis of the predicted EToV Mpro active-site residues. Further analysis of the EToV Mpro was performed by employing the vaccinia virus-based vTF7-3 mammalian expression system (24) (Fig. 5). To this end, residues 10376 to 11472 of the EToV genome were provided with an AUG codon in an optimal context for translation initiation (40) and the resulting ORF was placed under the control of the bacteriophage T7 promoter (Fig. 5A). Expression of this construct predominantly yielded the 33-kDa Mpro species. To assess the importance of the predicted catalytic residues, single-amino-acid substitutions were introduced, and their effects were studied by analyzing the processing activities of the mutant proteinases. Consistent with their proposed catalytic function, replacement of His3304 by either Gly or Arg and of Ser3416 by Ala completely abolished proteolytic activity. Interestingly, however, an Mpro mutant with a Ser3416Cys substitution retained partial activity (Fig. 5B).
EToV Mpro displays the distinctive traits of a classical serine proteinase, and hence the existence of a third acidic catalytic residue was surmised. On the basis of the alignments (Fig. 1A and B), Glu3347 was considered the most likely candidate to occupy a main chain position equivalent to that of the catalytic Asp residue in chymotrypsin. In addition, Asp3383, Glu3354, and Asp3358 were selected for mutagenesis. However, the substitution for neither of these residues by Asn or Ala affected proteolytic processing of the 42-kDa Mpro precursor (Fig. 5C). One explanation for this result would be that yet another acidic residue is part of the catalytic triad. Still, the fact that our assay relied on autocatalytic cleavages that may occur in cis posed a caveat. Mutagenesis studies of other serine proteases have shown that the acidic active-site residue is not absolutely required for peptide hydrolysis, although its substitution may cause the enzymatic activity to drop by several orders of magnitude (11, 16, 33, 41, 64, 65). Yet, in an experimental setup in which processing occurs by highly efficient cleavage in cis, even such considerable differences in enzymatic activity may go unnoticed. We therefore designed a more sensitive trans-cleavage assay, outlined in Fig. 6A, which entailed coexpression of enzymatically active Mpro and a substrate protein. Enzyme and substrate were expressed from the same transcript with translation of the latter being driven by the EMCV IRES.
The substrate was a 71-kDa chimeric protein consisting of (i) residues 1 to 103 of pp1a/pp1ab, (ii) a EToV Mpro-containing segment consisting of pp1a/pp1ab residues 3187 to 3551, and (iii) residues 2 to 160 of the EToV N protein, as listed from N to C terminus. In this fusion protein, the N-terminal Mpro cleavage site was abolished by a QSAG substitution, while the Mpro was inactivated by changing the active site Ser3416 to Ala. However, the cleavage site C terminal of Mpro was left intact and hence remained available for proteolytic cleavage in trans. This should give rise to products of 52 kDa (the pp1a-Mpro fusion protein) and of 19 kDa (the cleaved-off N moiety), which are readily distinguishable from the enzymatically active Mpro and its precursors. Processing was monitored by RIPA with antisera against the N terminus of pp1a, Mpro, and the N protein and the effect of substitutions of the selected acidic residues was assessed (Fig. 6B). A construct in which the Mpro enzyme was inactivated by a Ser3416Ala substitution was taken along as a negative control.
Mpro derivatives, in which Asp3338, Glu3354, and Asp3358 had been altered, behaved like the wild-type enzyme in the trans-cleavage assay. Substitution of Glu3347, however, reduced proteolytic activity as indicated by accumulation of the 71-kDa noncleaved substrate precursor and of the Mpro precursors of 42 and 34 kDa; the latter protein species apparently represented an intermediate generated by processing of the 42-kDa Mpro precursor at the N-terminal but not at the C-terminal cleavage site (Fig. 6B).
DISCUSSION
The main proteinases, which mediate most of the processing steps of nidovirus replicase polyproteins, have been studied so far only for corona-, arteri-, and roniviruses. Here, we report the biochemical characterization of the fourth member, the torovirus Mpro. The nidoviral Mpros all belong to a family of viral and cellular proteinases, resembling picornavirus 3C (3C-like proteinases [3CLPs]). Moreover, they are located at cognate positions within the C-terminal half of pp1a, typically flanked by hydrophobic domains at either side (74, 76). Hence, the null hypothesis is that the proteinases of extant nidoviruses have all evolved from a single ancestral proto-nidovirus 3CLP. However, if so, they have diverged to an extreme extent. As noted previously (74, 76), primary sequence diversity of the corona-, roni-, and arterivirus Mpros is so pronounced that even the most sensitive computer-aided methods do not reveal any specific relationship among them that would justify their grouping and separation from other cellular or viral 3CLPs.
Within the order Nidovirales, the toroviruses seem to be most closely related to the coronaviruses (10, 25, 58, 60). We therefore anticipated at the onset of this study that the torovirus Mpro might bridge the large evolutionary gap between the main proteinase of the coronaviruses and those of the other nidoviruses (76). Surprisingly, however, from its primary sequence, the Mpro of EToV Berne seems no more related to the coronaviral proteinase than to those of the arteri- and roniviruses (Fig. 1B). The characteristics of the torovirus Mpro in relation to those of the other nidoviruses are discussed in the paragraphs below.
Mpro-mediated processing and cleavage site specificity. Nidoviral Mpros and related proteinases in other RNA viruses are released from adjoining polyprotein domains by autocatalytic processing (21, 76). Likewise, when pp1a segments, containing the predicted EToV Mpro domain, were heterologously expressed in prokaryotic and in mammalian cells, the proteinase liberated itself by cleaving both at an N-terminal site and at a C-terminal site. This yielded a mature Mpro of 33 kDa, a protein of which size was also detected in EToV-infected cells by RIPA with Mpro-specific antisera. Interestingly, in EToV-infected cells, autocatalytic release of Mpro might be inefficient as during a 3-h continuous metabolic labeling period a considerable proportion of the proteinase seemed to remain part of a large precursor (Fig. 4). Similar observations were reported for the arteri- and coronavirus Mpros (32, 54, 63, 71). It has been suggested, that the long-lived Mpro intermediates are membrane associated via the flanking hydrophobic domains. Anchoring of proteinase precursors to intracellular membranes might well be a central phenomenon in nidovirus polyprotein processing and/or replication (39, 54, 71).
N-terminal sequence analysis of mature EToV Mpro showed that autocatalytic cleavage occurred at the location 3246SNFSFQSVFS3255. The sequence 3538QPFKKQSVSN3547was identified as the C-terminal cleavage site. Unexpectedly, in prokaryotic cells, processing occurred also at a site internal to Mpro, 3392SEFATQAWQT3401, effectively cleaving the proteinase in half. It is tempting to speculate that, during EToV infection, Mpro might inactivate itself as part of a feedback autoregulatory mechanism or, otherwise, that processing at the internal site gives rise to an alternative cleavage pathway (71). However, we have no evidence that internal processing actually occurs in mammalian cells. The 16-kDa cleavage products were detected neither in the vTF7-3 expression system nor in EToV-infected cells. Hence, it cannot be excluded that internal cleavage, as observed in E. coli, is an artifact resulting from overexpression and possibly aided by protein misfolding, which might have exposed cleavage sites that are normally inaccessible. Intriguingly, however, the internal cleavage site is also conserved in BToV Mpro (22).
Catalytic center and substrate specificity. Although sequence identity between EToV Mpro and its nidovirus homologues or other viral and cellular chymotrypsin-like enzymes is limited, sequence alignments clearly implicated Ser3416 and His3304 as the main catalytic residues. Consistent with this notion, their replacement by Ala and by Gly or Arg, respectively, abrogated enzymatic activity. The Mpros of corona- and roniviruses (2, 73, 74), like the hepatitis A virus 3C proteinase (1, 7), employ a catalytic Cys-His dyad within the context of the chymotrypsin-fold. Under the assumption that the EToV Mpro is a classical serine proteinase, a third acidic catalytic residue was expected (20). Although sequence alignments of EToV Mpro with other proteinases were far from robust, particularly in the region that should contain this third residue, our analyses suggested Glu3347 as the most likely candidate. However, neither substitution mutagenesis of Glu3347 nor that of the proximal Asp3338, Glu3354, and Asp3358 noticeably affected autocatalytic release of Mpro in heterologous expression studies. Yet, in an assay that tested for processing in trans, the efficacy of Mpro cleavage of an artificial substrate was clearly reduced by substitution of Glu3347, but not by that of the other acidic residues. Although our findings do not prove that Glu3347 is included in a catalytic triad with Ser3416 and His3304, they would at least be consistent with this notion. This view might seem at odds with observations made for various other viral and cellular chymotrypsin-like proteinases, where site-directed mutagenesis of the third residue resulted in a far more dramatic loss of proteolytic activity (11, 16, 33, 41, 64, 65). Still, in the case of severe acute respiratory syndrome (SARS) coronavirus Mpro, replacement of the active-site Cys by Ser yields a functional enzyme. Although this mutant proteinase relies exclusively on a Ser-His catalytic dyad, its activity is merely 40-fold reduced as compared to that of wild-type SARS coronavirus Mpro (36). A reduction in activity of this magnitude might have only a modest effect on processing even in our trans-cleavage assay under conditions of vTF7-3-driven EToV Mpro overexpression. As the mapping of active-site residues by the combination of comparative sequence analysis and site-directed mutagenesis is inherently fraught with a measure of uncertainty, definitive answers with respect to catalytic center composition and the presence/identity of a third catalytic residue will require the elucidation of the torovirus Mpro crystal structure.
Comparison of the sequences flanking the scissile bonds of the internal and the N- and C-terminal cleavage sites tentatively identifies FxxQ(S, A) as the substrate consensus sequence of EToV Mpro. Its cleavage site specificity thus resembles that of other nidoviral Mpros, which, as is typical for 3CLPs, all require Gln or Glu at the P1 position (74, 76). The specificity for P1-(Gln, Glu) of the picorna-, arteri-, and coronavirus enzymes is determined by a strictly conserved His, which via its imidazole side chain hydrogen bonds to the carbonyl/carboxylate oxygens of the P1 side chain (1, 2, 4, 7, 44, 45, 47, 73, 74, 76). A corresponding His residue is also present in the torovirus Mpro (EToV His3430) as part of a Gly-X-His sequence (Fig. 1B), a motif well conserved in most 3CLPs (including arteri- and ronivirus Mpro), but saliently absent in the Mpros of coronaviruses; the latter have a Tyr-Met-His motif instead (2, 4, 34, 64, 74, 75). In many 3CLPs, conserved Ser/Thr residues, located five residues upstream of the catalytic Ser/Cys, are also part of the substrate-binding subsite S1 and assist in binding by forming additional hydrogen bonds with the carbonyl/carboxylate oxygen atoms of the P1 side chain (4, 44, 47, 48, 74). We propose that, in the torovirus Mpro, Thr3411 plays a similar role. This would again be a difference from the coronavirus Mpro, which lacks a corresponding Ser/Thr residue (2, 74).
The arteri- and ronivirus Mpros prefer glutamic acid at the P1 position of their substrates, whereas coronavirus Mpro requires glutamine (64, 74, 76). In arterivirus Mpros, this substrate preference is presumably conferred by a Ser residue (located three residues downstream of the S1 subsite His; S137 in EAV Mpro; see Fig. 1B), which would hydrogen bond with the carboxylate oxygen atom (that is, the one not interacting with the subsite S1 His/Thr pair) of the P1-Glu side chain (4). This Ser is not conserved in the corona- and torovirus Mpros; instead, at the corresponding positions, Glu and Leu residues, respectively, are found (Fig. 1b). The S1 specificity site of coronavirus Mpro consists of the main chain atoms of Met165, Glu166, and His172 and the side chains of Phe140 and His163, with the imidazole of the latter residue interacting with the P1-Gln side chain (2, 3, 73) (residue numbering according to SARS-CoV Mpro). Also with respect to S1 subsite composition, the torovirus main proteinase differs from that of coronaviruses. Although EToV Mpro, like its coronavirus homologues, apparently prefers P1-Gln, none of the coronavirus S1 subsite residues are conserved, except for His163. As yet, it is not clear from the EToV Mpro sequence how the P1-Gln is accommodated.
Tentative Mpro processing scheme of torovirus pp1a/pp1ab. Although each of the three established EToV Mpro cleavage sites conforms to the consensus FxxQ(S, A), we believe, on the basis of alignments of pp1a/pp1ab of coronaviruses and the toroviruses EToV and BToV, that Tyr/Met/Leu and Gly/Lys may also be tolerated at the P4 and P1' positions, respectively. Although the picture is still incomplete, the substrate specificity of the torovirus Mpro is reminiscent of and yet distinct from those of the corona-, arteri-, and roniviruses, which mainly conform to LQ(S, A), E (G, S, A), and VxHE(L, V), respectively (74, 76). In particular, a preference for a bulky hydrophobic residue at P4 and no apparent predilection for certain residues at P2 would give the EToV Mpro a unique cleavage site formula. A tentative Mpro processing scheme of EToV pp1a/1b is presented in Fig. 7, with the sequences and locations of potential cleavage sites listed in Table 2. Several points are of note. (i) All cleavage sites predicted for EToV are conserved in BToV (22). (ii) A putative cleavage site, 2870FKKQSV2875 (Table 2, site 1), identical in sequence to the one C terminal of the Mpro domain (Table 2, site 4), is found halfway in EToV pp1a at a position topologically similar to that of an accessory papain-like proteinase 2 site within coronavirus pp1a. (iii) Of the predicted sites in the ORF1b-encoded part of the replicase (Table 2, sites 10 to 14), only one, located between the RdRp and zinc finger/helicase domains, conforms to the proposed consensus FxxQ (S, A) sequence (site 10; Table 2). Future studies should determine whether the predicted sites within pp1a/pp1ab are actually cleaved by Mpro. (iv) At the very C terminus of torovirus pp1a, there is a domain that is absent from all other nidovirus pp1a proteins and that shares identity with a predicted cyclic phosphodiesterase (CPD) motif in the ns2A protein of group 2 coronaviruses and in the rotavirus VP3 guanylyl transferase (Fig. 7) (57, 60). Puzzlingly, a cleavage site C terminal of this domain is not immediately apparent, raising the question whether Mpro utilizes a noncanonical site, whether another viral or host proteinase is involved, or whether processing of pp1ab yields an RdRp with the CPD homologue attached; note that in the latter case, processing of pp1a would still produce an additional, "free" CPD.
Evolutionary implications. EToV Mpro combines several traits of its closest relatives. (i) As is typical for nidovirus Mpros, the EToV proteinase has a predicted three-domain structure, with two catalytic -barrel domains and an additional C-terminal domain of unknown function. (ii) With respect to substrate specificity, the EToV Mpro resembles its coronavirus homologue in its preference for P1-Gln. (iii) Then again, it has what appears to be a canonical substrate-binding subsite, S1, resembling that of arteri- and ronivirus Mpros, but clearly distinct from that of coronavirus Mpro. (iv) Finally, in contrast to the Mpros of corona- and roniviruses, but like that of arterivirus, the torovirus Mpro uses serine instead of cysteine as its principal nucleophile. Rather than clarifying the evolutionary history of the nidoviral Mpros, the analysis of a torovirus Mpro seems to make the picture even more complex. If anything, our findings further emphasize the unique properties of the coronavirus Mpro, which on the basis of its catalytic system and substrate-binding pocket, was already considered an outlier among viral and cellular chymotrypsin-like homologues (2, 74). It is still quite possible, however, that striking similarities between corona- and toroviral Mpros will become overt once the crystal structure of toroviral Mpro has been solved.
Corona- and toroviruses share many traits (e.g., with regard to virion composition, genome organization, and the extent of colinearity in the ORF1b-encoded region of pp1ab), which would justify their grouping together, separate from the arteri- and roniviruses (15, 19, 25, 26). However, in a recent study of Gonzalez et al. (25), entailing a systematic quantitative analysis of sequence conservation among nidoviral proteins, the expected sequence affinity between corona- and toroviruses was not evident. Thus, with the topology of the nidovirus tree undecided, it was suggested that "the nidovirus phylogeny must be verified later when the number and diversity of torovirus and ronivirus sequences will match those of arteriviruses and coronaviruses" (25). If it were to be firmly established that of the nidoviruses, corona- and toroviruses are, after all, the most closely related, and if this evolutionary relationship also holds for the Mpro (i.e., if no RNA recombination-based exchange of proteinase domains—or parts thereof—has occurred during nidoviral divergence), our observations would have one more major implication: namely that the transition from serine- to cysteine-based proteolytic catalysis (or vice versa) has happened more than once in the course of nidovirus evolution. In this respect, it is of interest that a mutant EToV Mpro with a Ser3416Cys substitution retained enzymatic activity and that, conversely, the catalytic site CysSer substitution in SARS-CoV Mpro also yields an active enzyme (36).
ACKNOWLEDGMENTS
We acknowledge the contribution of Alexander E. Gorbalenya to the bioinformatical analysis of EToV ORF1a-derived cDNA clones 133 and 135 and EToV pp1a, as well as to the design and analysis of some of the experiments. We are grateful to Peter J. M. Rottier and Jolanda D. F. de Groot-Mijnes for critical reading of the manuscript.
REFERENCES
Allaire, M., M. M. Chernaia, B. A. Malcolm, and M. N. James. 1994. Picornaviral 3C cysteine proteinases have a fold similar to chymotrypsin-like serine proteinases.Nature 369:72-76.
Anand, K., G. J. Palm, J. R. Mesters, S. G. Siddell, J. Ziebuhr, and R. Hilgenfeld. 2002. Structure of coronavirus main proteinase reveals combination of a chymotrypsin fold with an extra alpha-helical domain. EMBO J. 21:3213-3224.
Anand, K., J. Ziebuhr, P. Wadhwani, J. R. Mesters, and R. Hilgenfeld. 2003. Coronavirus main proteinase (3CLpro) structure: basis for design of anti-SARS drugs. Science 13:13.
Barrette-Ng, I. H., K. K. Ng, B. L. Mark, D. Van Aken, M. M. Cherney, C. Garen, Y. Kolodenko, A. E. Gorbalenya, E. J. Snijder, and M. N. James.2002 . Structure of arterivirus nsp4. The smallest chymotrypsin-like proteinase with an alpha/beta C-terminal extension and alternate conformations of the oxyanion hole. J. Biol. Chem. 277:39960-39966.
Bazan, J. F., and R. J. Fletterick. 1989. Detection of a trypsin-like serine protease domain in flaviviruses and pestiviruses. Virology 171:637-639.
Bazan, J. F., and R. J. Fletterick. 1988. Viral cysteine proteases are homologous to the trypsin-like family of serine proteases: structural and functional implications. Proc. Natl. Acad. Sci. USA 85:7872-7876.
Bergmann, E. M., S. C. Mosimann, M. M. Chernaia, B. A. Malcolm, and M. N. G. James.1997 . The refined crystal structure of the 3C gene product from hepatitis A virus: specific proteinase activity and RNA recognition. J. Virol. 71:2436-2448.
Bhardwaj, K., L. Guarino, and C. C. Kao. 2004. The severe acute respiratory syndrome coronavirus Nsp15 protein is an endoribonuclease that prefers manganese as a cofactor.J. Virol. 78:12218-12224.
Brierley, I., P. Digard, and S. C. Inglis. 1989. Characterization of an efficient coronavirus ribosomal frameshifting signal: requirement for an RNA pseudoknot. Cell 57:537-547.
Cavanagh, D. 1997. Nidovirales: a new order comprising Coronaviridae and Arteriviridae. Arch. Virol. 142:629-633.
Chambers, T. J., C. S. Hahn, R. Galler, and C. M. Rice. 1990. Flavivirus genome organization, expression, and replication. Annu. Rev. Microbiol. 44:649-688.
Chirnside, E. D., P. M. Francis, and J. A. Mumford. 1995. Expression cloning and antigenic analysis of the nucleocapsid protein of equine arteritis virus.Virus Res. 39:277-288.
Choi, H. K., L. Tong, W. Minor, P. Dumas, U. Boege, M. G. Rossmann, and G. Wengler. 1991. Structure of Sindbis virus core protein reveals a chymotrypsin-like serine proteinase and the organization of the virion. Nature 354:37-43.
Cornelissen, L. A. H. M. 1999. Molecular characterization of ungulate toroviruses. Ph.D. thesis. Utrecht, The Netherlands.
Cowley, J. A., C. M. Dimmock, K. M. Spann, and P. J. Walker. 2000. Gill-associated virus of Penaeus monodon prawns: an invertebrate virus with ORF1a and ORF1b genes related to arteri- and coronaviruses. J. Gen. Virol. 81:1473-1484.
Craik, C. S., S. Roczniak, C. Largman, and W. J. Rutter. 1987. The catalytic role of the active site aspartic acid in serine proteases. Science 237:909-913.
Demler, S. A., and G. A. de Zoeten. 1991. The nucleotide sequence and luteovirus-like nature of RNA 1 of an aphid non-transmissible strain of pea enation mosaic virus.J. Gen. Virol. 72:1819-1834.
den Boon, J. A., E. J. Snijder, E. D. Chirnside, A. A. F. de Vries, M. C. Horzinek, and W. J. M. Spaan.1991 . Equine arteritis virus is not a togavirus but belongs to the coronaviruslike superfamily. J. Virol. 65:2910-2920.
de Vries, A. A. F., M. C. Horzinek, P. J. M. Rottier, and R. J. de Groot.1997 . The genome organization of the Nidovirales: similarities and differences between Arteri-, Toro-, and Coronaviruses.Semin. Virol. 8:33-47.
Dodson, G., and A. Wlodawer. 1998. Catalytic triads and their relatives. Trends Biochem. Sci. 23:347-352.
Dougherty, W. G., and B. L. Semler. 1993. Expression of virus-encoded proteinases: functional and structural similarities with cellular enzymes. Microbiol. Rev. 57:781-822.
Draker, R., R. L. Roper, M. Petric, and R. Tellier.2005 . The complete sequence of the bovine torovirus genome. Virus Res. 115:56-68. (First published, 30 August 2005; doi:10.1016/jvirusres.2005.07.005.)
Elroy-Stein, O., and B. Moss. 1990. Cytoplasmic expression system based on constitutive synthesis of bacteriophage T7 RNA polymerase in mammalian cells. Proc. Natl. Acad. Sci. USA 87:6743-6747.
Fuerst, T. R., E. G. Niles, F. W. Studier, and B. Moss. 1986. Eukaryotic transient-expression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase. Proc. Natl. Acad. Sci. USA 83:8122-8126.
Gonzalez, J. M., P. Gomez-Puertas, D. Cavanagh, A. E. Gorbalenya, and L. Enjuanes. 2003. A comparative sequence analysis to revise the current taxonomy of the family Coronaviridae. Arch. Virol. 148:2207-2235.
Gorbalenya, A. E. 2001. Big nidovirus genome. When count and order of domains matter. Adv. Exp. Med. Biol. 494:1-17.
Gorbalenya, A. E., V. M. Blinov, and A. P. Donchenko. 1986. Poliovirus-encoded proteinase 3C: a possible evolutionary link between cellular serine and cysteine proteinase families. FEBS Lett. 194:253-257.
Gorbalenya, A. E., A. P. Donchenko, V. M. Blinov, and E. V. Koonin. 1989. Cysteine proteases of positive strand RNA viruses and chymotrypsin-like serine proteases.A distinct protein superfamily with a common structural fold. FEBS Lett. 243:103-114.
Gorbalenya, A. E., E. V. Koonin, V. M. Blinov, and A. P. Donchenko. 1988. Sobemovirus genome appears to encode a serine protease related to cysteine proteases of picornaviruses. FEBS Lett. 236:287-290.
Gorbalenya, A. E., E. V. Koonin, A. P. Donchenko, and V. M. Blinov. 1989. Coronavirus genome: prediction of putative functional domains in the non-structural polyprotein by comparative amino acid sequence analysis. Nucleic Acids Res. 17:4847-4861.
Gorbalenya, A. E., E. V. Koonin, and M. M. Lai.1991 . Putative papain-related thiol proteases of positive-strand RNA viruses. Identification of rubi- and aphthovirus proteases and delineation of a novel conserved domain associated with proteases of rubi-, alpha- and coronaviruses. FEBS Lett. 288:201-205.
Gosert, R., A. Kanjanahaluethai, D. Egger, K. Bienz, and S. C. Baker. 2002. RNA replication of mouse hepatitis virus takes place at double-membrane vesicles. J. Virol. 76:3697-3708.
Hedstrom, L. 2002. Serine protease mechanism and specificity.Chem. Rev. 102:4501-4524.
Hegyi, A., and J. Ziebuhr. 2002. Conservation of substrate specificities among coronavirus main proteases. J. Gen. Virol. 83:595-599.
Horton, R. M., and L. R. Pease. 1991. Recombination and mutagenesis of DNA sequences using PCR, p.217 -247. In M. J. McPherson (ed.), Directed mutagenesis, a practical approach. IRL Press, Oxford, United Kingdom.
Huang, C., P. Wei, K. Fan, Y. Liu, and L. Lai. 2004. 3C-like proteinase from SARS coronavirus catalyzes substrate hydrolysis by a general base mechanism. Biochemistry 43:4568-4574.
Ivanov, K. A., T. Hertzig, M. Rozanov, S. Bayer, V. Thiel, A. E. Gorbalenya, and J. Ziebuhr. 2004. Major genetic marker of nidoviruses encodes a replicative endoribonuclease. Proc. Natl. Acad. Sci. USA 101:12694-12699.
Jiang, B., S. S. Monroe, E. V. Koonin, S. E. Stine, and R. I. Glass. 1993. RNA sequence of astrovirus: distinctive genomic organization and a putative retrovirus-like ribosomal frameshifting signal that directs the viral replicase synthesis. Proc. Natl. Acad. Sci. USA 90:10539-10543.
Kanjanahaluethai, A., and S. C. Baker. 2000. Identification of mouse hepatitis virus papain-like proteinase 2 activity.J. Virol. 74:7911-7921.
Kozak, M. 1986. Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44:283-292.
Lawson, M. A., and B. L. Semler. 1991. Poliovirus thiol proteinase 3C can utilize a serine nucleophile within the putative catalytic triad. Proc. Natl. Acad. Sci. USA 88:9919-9923.
Li, X., M. D. Ryan, and J. W. Lamb.2000 . Potato leafroll virus protein P1 contains a serine proteinase domain. J. Gen. Virol. 81:1857-1864.
Lu, Y., X. Lu, and M. R. Denison. 1995. Identification and characterization of a serine-like proteinase of the murine coronavirus MHV-A59. J. Virol. 69:3554-3559.
Matthews, D. A., P. S. Dragovich, S. E. Webber, S. A. Fuhrman, A. K. Patick, L. S. Zalman, T. F. Hendrickson, R. A. Love, T. J. Prins, J. T. Marakovits, R. Zhou, J. Tikhe, C. E. Ford, J. W. Meador, R. A. Ferre, E. L. Brown, S. L. Binford, M. A. Brothers, D. M. DeLisle, and S. T. Worland. 1999. Structure-assisted design of mechanism-based irreversible inhibitors of human rhinovirus 3C protease with potent antiviral activity against multiple rhinovirus serotypes. Proc. Natl. Acad. Sci. USA 96:11000-11007.
Matthews, D. A., W. W. Smith, R. A. Ferre, B. Condon, G. Budahazi, W. Sisson, J. E. Villafranca, C. A. Janson, H. E. McElroy, C. L. Gribskov, et al. 1994. Structure of human rhinovirus 3C protease reveals a trypsin-like polypeptide fold, RNA-binding site, and means for cleaving precursor polyprotein. Cell 77:761-771.
Mijnes, J. D. F., L. M. van der Horst, E. van Anken, M. C. Horzinek, P. J. M. Rottier, and R. J. de Groot. 1996. Biosynthesis of glycoproteins E and I of feline herpesvirus: gE-gI interaction is required for intracellular transport. J. Virol. 70:5466-5475.
Mosimann, S. C., M. M. Cherney, S. Sia, S. Plotch, and M. N. James. 1997. Refined X-ray crystallographic structure of the poliovirus 3C gene product. J. Mol. Biol. 273:1032-1047.
Nienaber, V. L., K. Breddam, and J. J. Birktoft.1993 . A glutamic acid specific serine protease utilizes a novel histidine triad in substrate binding. Biochemistry 32:11469-11475.
Paetzel, M., and R. E. Dalbey. 1997. Catalytic hydroxyl/amine dyads within serine proteases. Trends Biochem. Sci. 22:28-31.
Reed, S. B., C. A. Wesson, L. E. Liou, W. R. Trumble, P. M. Schlievert, G. A. Bohach, and K. W. Bayles. 2001. Molecular characterization of a novel Staphylococcus aureus serine protease operon. Infect. Immun. 69:1521-1527.
Rice, C. M., and J. H. Strauss. 1981. Nucleotide sequence of the 26S mRNA of sindbis virus and deduced sequence of the encoded virus structural proteins. Proc. Natl. Acad. Sci. USA 78:2062-2066.
Ryan, M. D., and M. Flint. 1997. Virus-encoded proteinases of the picornavirus super-group. J. Gen. Virol. 78:699-723.
Ryan, M. D., S. Monaghan, and M. Flint. 1998. Virus-encoded proteinases of the Flaviviridae. J. Gen. Virol. 79:947-959.
Schiller, J. J., A. Kanjanahaluethai, and S. C. Baker.1998 . Processing of the coronavirus MHV-JHM polymerase polyprotein: identification of precursors and proteolytic products spanning 400 kilodaltons of ORF1a. Virology 242:288-302.
Shanks, M., and G. P. Lomonossoff. 1990. The primary structure of the 24K protease from red clover mottle virus: implications for the mode of action of comovirus proteases.J. Gen. Virol. 71:735-738.
Siddell, S. G., J. Ziebuhr, and E. J. Snijder.2005 . Coronaviruses, toroviruses and arteriviruses, p.823 -856. In B. W. Mahy and V. ter Meulen (ed.), Topley and Wilson's microbiology and microbial infections. Virology. Hodder Arnold, London, United Kingdom.
Snijder, E. J., P. J. Bredenbeek, J. C. Dobbe, V. Thiel, J. Ziebuhr, L. L. Poon, Y. Guan, M. Rozanov, W. J. Spaan, and A. E. Gorbalenya.2003 . Unique and conserved features of genome and proteome of SARS-coronavirus, an early split-off from the coronavirus group 2 lineage. J. Mol. Biol. 331:991-1004.
Snijder, E. J., J. A. den Boon, P. J. Bredenbeek, M. C. Horzinek, R. Rijnbrand, and W. J. Spaan.1990 . The carboxyl-terminal part of the putative Berne virus polymerase is expressed by ribosomal frameshifting and contains sequence motifs which indicate that toro- and coronaviruses are evolutionarily related. Nucleic Acids Res. 18:4535-4542.
Snijder, E. J., J. A. den Boon, M. C. Horzinek, and W. J. Spaan. 1991. Characterization of defective interfering RNAs of Berne virus. J. Gen. Virol. 72:1635-1643.
Snijder, E. J., and M. C. Horzinek. 1993. Toroviruses: replication, evolution and comparison with other members of the coronavirus-like superfamily. J. Gen. Virol. 74:2305-2316.
Snijder, E. J., M. C. Horzinek, and W. J. M. Spaan. 1990. A 3'-coterminal nested set of independently transcribed mRNAs is generated during Berne virus replication. J. Virol. 64:331-338.
Snijder, E. J., S. G. Siddell, and A. E. Gorbalenya. 2005. The order Nidovirales, p.390 -404. In B. W. Mahy and V. ter Meulen (ed.), Topley and Wilson's microbiology and microbial infections. Virology. Hodder Arnold, London, United Kingdom.
Snijder, E. J., A. L. M. Wassenaar, and W. J. M. Spaan. 1994. Proteolytic processing of the replicase ORF1a protein of equine arteritis virus.J. Virol. 68:5755-5764.
Snijder, E. J., A. L. Wassenaar, L. C. van Dinten, W. J. Spaan, and A. E. Gorbalenya.1996 . The arterivirus nsp4 protease is the prototype of a novel group of chymotrypsin-like enzymes, the 3C-like serine proteases.J. Biol. Chem. 271:4864-4871.
Sprang, S., T. Standing, R. J. Fletterick, R. M. Stroud, J. Finer-Moore, N. H. Xuong, R. Hamlin, W. J. Rutter, and C. S. Craik. 1987. The three-dimensional structure of Asn102 mutant of trypsin: role of Asp102 in serine protease catalysis. Science 237:905-909.
ten Dam, E., M. Flint, and M. D. Ryan. 1999. Virus-encoded proteinases of the Togaviridae. J. Gen. Virol. 80:1879-1888.
van der Meer, Y., H. van Tol, J. K. Locker, and E. J. Snijder. 1998. ORF1a-encoded replicase subunits are involved in the membrane association of the arterivirus replication complex. J. Virol. 72:6689-6698.
Van Vliet, A. L., S. L. Smits, P. J. Rottier, and R. J. De Groot. 2002. Discontinuous and non-discontinuous subgenomic RNA transcription in a nidovirus.EMBO J. 21:6571-6580.
Vennema, H., R. Rijnbrand, L. Heijnen, M. C. Horzinek, and W. J. Spaan. 1991. Enhancement of the vaccinia virus/phage T7 RNA polymerase expression system using encephalomyocarditis virus 5'-untranslated region sequences.Gene 108:201-209.
Wang, R. F., and S. R. Kushner. 1991. Construction of versatile low-copy-number vectors for cloning, sequencing and gene expression in Escherichia coli. Gene 100:195-199.
Wassenaar, A. L. M., W. J. M. Spaan, A. E. Gorbalenya, and E. J. Snijder.1997 . Alternative proteolytic processing of the arterivirus replicase ORF1a polyprotein: evidence that NSP2 acts as a cofactor for the NSP4 serine protease. J. Virol. 71:9313-9322.
Wellink, J., and A. van Kammen. 1988. Proteases involved in the processing of viral polyproteins. Brief review. Arch. Virol. 98:1-26.
Yang, H., M. Yang, Y. Ding, Y. Liu, Z. Lou, Z. Zhou, L. Sun, L. Mo, S. Ye, H. Pang, G. F. Gao, K. Anand, M. Bartlam, R. Hilgenfeld, and Z. Rao. 2003. The crystal structures of severe acute respiratory syndrome virus main protease and its complex with an inhibitor. Proc. Natl. Acad. Sci. USA 100:13190-13195.
Ziebuhr, J., S. Bayer, J. A. Cowley, and A. E. Gorbalenya. 2003. The 3C-like proteinase of an invertebrate nidovirus links coronavirus and potyvirus homologs.J. Virol. 77:1415-1426.
Ziebuhr, J., G. Heusipp, and S. G. Siddell. 1997. Biosynthesis, purification, and characterization of the human coronavirus 229E 3C-like proteinase. J. Virol. 71:3992-3997.
Ziebuhr, J., E. J. Snijder, and A. E. Gorbalenya.2000 . Virus-encoded proteinases and proteolytic processing in the Nidovirales. J. Gen. Virol. 81:853-879.(Saskia L. Smits, Eric J. )
Molecular Virology Laboratory, Department of Medical Microbiology, Center of Infectious Diseases, Leiden University Medical Center, Leiden, The Netherlands
ABSTRACT
Viruses of the order Nidovirales encode huge replicase polyproteins. These are processed primarily by the chymotrypsin-like main proteinases (Mpros). So far, Mpros have been studied only for corona-, arteri-, and roniviruses. Here, we report the characterization of the Mpro of toroviruses, the fourth main Nidovirus branch. Comparative sequence analysis of polyprotein 1a of equine torovirus (EToV) strain Berne, identified a serine proteinase domain, flanked by hydrophobic regions. Heterologous expression of this domain resulted in autoprocessing at flanking cleavage sites. N-terminal sequence analysis of cleavage products tentatively identified FxxQ(S, A) as the substrate consensus sequence. EToV Mpro combines several traits of its closest relatives. It has a predicted three-domain structure, with two catalytic -barrel domains and an additional C-terminal domain of unknown function. With respect to substrate specificity, the EToV Mpro resembles its coronavirus homologue in its preference for P1-Gln, but its substrate-binding subsite, S1, more closely resembles that of arteri- and ronivirus Mpros, which prefer P1-Glu. Surprisingly, in contrast to the Mpros of corona- and roniviruses, but like that of arterivirus, the torovirus Mpro uses serine instead of cysteine as its principal nucleophile. Under the premise that the Mpros of corona- and toroviruses are more closely related to each other than to those of arteri- and roniviruses, the transition from serine- to cysteine-based proteolytic catalysis (or vice versa) must have happened more than once in the course of nidovirus evolution. In this respect, it is of interest that a mutant EToV Mpro with a Ser165Cys substitution retained partial enzymatic activity.
INTRODUCTION
Many positive-strand RNA viruses have adopted a "polyprotein" strategy to express (part of) their genome. Translation of viral RNA results in the synthesis of large multidomain protein precursors, which are proteolytically cleaved to yield a collection of intermediate and mature products. As a rule, polyproteins that are to yield secretory or membrane (glyco)proteins are processed by host-encoded proteinases within the lumen of the compartments of the exocytotic pathway (in particular the endoplasmic reticulum and the trans-Golgi network). Cytosolic processing, however, involves co- and posttranslational cleavage by viral proteinases that are present as distinct domains within the polyprotein itself (51, 72). Processing is often a prerequisite for the assembly of the viral RNA replicase/transcriptase. To add to the complexity, intermediate and mature products may have distinct functions in replication and may be required during different stages of the infection. Hence, the properties of the viral proteinases must be carefully tailored to ensure accurate and efficient cleavage in a substrate-specific and temporally well-coordinated fashion.
Many RNA virus proteinases resemble, in their mechanism of peptide hydrolysis and in their three-dimensional structure, either one of the prototypical cellular proteases papain and chymotrypsin. Papain and related proteinases are characterized by a two-domain +-fold with typically a cysteine and downstream histidine as the catalytic residues. Chymotrypsin has a distinctive two--barrel structure in which the principal nucleophile, a serine residue, is part of a catalytic His-Asp-Ser triad (20, 49). Chymotrypsin-like serine proteinase domains have been identified within the polyproteins of among others flavi-, alpha-, luteo-, astro-, and sobemoviruses (13, 17, 29, 38, 42, 53, 66).Intriguingly, however, many viruses, for instance picorna-, como-, poty-, and caliciviruses, encode proteinases with a (predicted) chymotrypsin-like fold in which, instead of serine, a cysteine acts as the main catalytic residue (1, 5, 6, 27, 28, 31, 52, 55).
The order Nidovirales comprises a group of evolutionarily related enveloped (positive) stranded RNA viruses (corona-, toro-, arteri-, and roniviruses) (10, 15, 19, 26, 56, 58, 62), which have taken the polyprotein strategy to the extreme. The 5'-most two-thirds of the nidovirus genome is occupied by two large overlapping open reading frames (ORFs), ORF1a and ORF1b, the most downstream of which is only translated upon ribosomal frameshifting (9, 15, 18, 58). These ORFs specify two huge polyproteins, pp1a and the C-terminally extended frameshift protein pp1ab. In the case of the coronaviruses, pp1ab measures up to 7,200 amino acids in length, representing the largest polyprotein known to date. Among nidoviruses, the largest degree of sequence conservation is found within the ORF1b-encoded part of pp1ab, which contains the chief replicative domains (15, 19, 30, 57), i.e., the RNA dependent RNA polymerase (RdRp) and the helicase, as well as up to three additional domains, ExoN, NendoU, and 2'-O-methyltransferase (2'-O-MT), which display sequence identity to cellular RNA-modifying enzymes (8, 37, 57). Of the latter, the ExoN and 2'-O-MT domains are lacking in the arterivirus pp1ab.
Processing of the arteriviral and coronaviral polyproteins gives rise to 12 to 13 and 15 to 16 mature products, respectively, and an as yet unknown number of functional intermediates (57, 76). The N termini of the replicase polyproteins are processed by "accessory" papain-like proteinases (31). The vast majority of cleavages, however, are carried out by a single chymotrypsin-like main proteinase (Mpro), which is located in the C-terminal half of pp1a, flanked by hydrophobic regions that are thought to be involved in targeting the replicase to intracellular membranes (30, 54, 67, 71). Recent X-ray crystallographic analysis of the structures of corona- and arterivirus Mpros confirmed the predicted chymotrypsin-like two--barrel fold (2, 4, 73). Interestingly, the corona- and arteriviral enzymes both possess an additional C-terminal domain, a feature not commonly seen in other viral and cellular chymotrypsin-like proteinases (2, 4). Also for the ronivirus Mpro, such an additional C-terminal domain was predicted (74).
One might surmise that the nidoviral Mpros, given their pivotal role in viral replication, are well conserved. Conversely, however, the primary sequences of the Mpros of corona-, roni-, and arteriviruses have diverged almost beyond recognition, and sequence identity is restricted to the regions immediately surrounding the main catalytic residues. Moreover, whereas the arteriviral Mpro is a serine proteinase with a canonical His-Asp-Ser catalytic triad (4, 64), the Mpros of the corona- and roniviruses are cysteine proteinases (28, 30, 43, 74). In the latter, the acidic catalytic residues have been lost and, apparently, these enzymes solely rely on the remaining catalytic His-Cys dyad (2, 73, 74).
Toroviruses have been poorly studied, but because of their relatively close phylogenetic relationship to coronaviruses they may in many respects represent "missing links" in comparative nidovirus studies. Here, we are the first to report the biochemical characterization of a torovirus Mpro. Upon analysis of the C-terminal half of pp1a of equine torovirus (EToV) strain Berne, we identified a chymotrypsin-like proteinase domain, which, like the Mpros of the arteri-, corona-, and roniviruses, is flanked by hydrophobic regions. When expressed in prokaryotic and mammalian cells, the proteinase domain autocatalytically released itself from its surrounding sequences. Surprisingly, the toroviral Mpro seems no more related to that of coronaviruses than to the Mpros of arteri- and roniviruses. Most saliently, in contrast to the Mpros of corona- and roniviruses, but like that of arteriviruses, the torovirus Mpro is a serine proteinase. A tentative processing scheme for the torovirus pp1a/pp1ab polyproteins is presented, and the implications of our findings for nidovirus evolution are discussed.
MATERIALS AND METHODS
Cells and virus. Equine dermis (Ederm) cells (American Type Tissue Culture Collection) and OST7-1 cells (23) were maintained as monolayers in Dulbecco's modified Eagle's medium (DMEM; BioWhitaker) supplemented with penicillin (100 IU/ml) and streptomycin (100 μg/ml) and 20% (DMEM20) or 10% (DMEM10) fetal calf serum, respectively. EToV strain Berne was propagated in Ederm cells as described previously (68). Recombinant vaccinia virus vTF7-3 (24) was obtained from B. Moss (NIH, Bethesda, MD).
Isolation of intracellular EToV RNA, cDNA synthesis, and PCR amplification. Total cytoplasmic RNA, isolated from the equivalent of 4 x 106 EToV-infected Ederm cells (68), was subjected to reverse transcription-PCR (RT-PCR) with RNase H-free Moloney murine leukemia virus reverse transcriptase (Superscript II; Invitrogen) and with the Expand long template PCR system (Roche Diagnostics GmbH), according to the instructions of the manufacturers. Oligonucleotide primers 1071 and 1072 (Table 1) were designed after sequences located at the 5' end and 3' end of EToV ORF1A, respectively (59, 61). Primers 1288 and 1325 were designed after sequences from cDNA clones p133 and p135. These clones had been identified previously as EToV specific after screening of a genomic cDNA library (61) and had been tentatively mapped in ORF1a (E. J. Snijder, unpublished). Amplicons of approximately 5.5 kb were obtained, which were gel purified, blunt ended using large fragment DNA polymerase I (LFDP) according to the instructions of the manufacturer (Invitrogen), and cloned into an EcoRV-digested vector pWSK29 (70). Sequence analysis was performed commercially (BaseClear Labservices) using the ABI PRISM BigDye Terminators v3.0 cycle sequencing kit (Applied Biosystems) on at least two independent clones in both orientations.
Preparation of EToV Mpro-specific antisera. The coding region for EToV Mpro amino acids (aa) 5 to 196 (as counted from the Mpro N terminus) was RT-PCR amplified with oligonucleotide primers 1548 and 1549 (Table 1). The resulting PCR product was digested with BamHI and EcoRV and inserted into HindIII (blunt ended)-BamHI-digested pQE9 (QIAGEN), yielding plasmid pQE-SS-Mpro(5-196). This expression vector, encoding a 22-kDa Mpro fragment fused at the N terminus to a His6 tag, was introduced in Escherichia coli M15/pREP4 cells (QIAGEN). Expression of the recombinant protein was induced by adding IPTG (isopropyl--D-thiogalactopyranoside) to a final concentration of 1 mM to bacterial cultures during exponential growth (optical density [OD] of 0.6). Bacteria were lysed 5 h after induction, and the His-tagged expression product was purified by Ni-nitrilotriacetic acid affinity chromatography using the XPRESS System version 2.0 (Invitrogen) according to the instructions of the manufacturer. The recombinant protein was eluted with a mixture of 20 mM sodium phosphate, 8 M urea, and 500 mM sodium chloride (pH 6.0) containing 500 mM imidazole. The eluate was concentrated on a Centricon concentrator (30-kDa cutoff; Amicon), dialyzed for 6 h against phosphate-buffered saline (PBS)-0.05% sodium dodecyl sulfate (SDS) and, finally, dialyzed for 16 h against PBS-0.02% SDS. New Zealand White rabbits were immunized subcutaneously with 100 μg of His-Mpro 1548/1549 fusion protein as described previously (12); the resulting antiserum was designated K158.
Antisera specific for the N terminus of EToV Mpro (K270), the C terminus of the Mpro (K272), or the N terminus of pp1a/pp1ab (K274) were prepared by repeated subcutaneous immunization of New Zealand White rabbits with 200 μg of the bovine serum albumin-coupled synthetic peptides NH2-SVFSKATSPFTLHARPPK-COOH, NH2-KPLQYFHVPSFWQPFKKQ-COOH, and NH2-MFRILKNNTRETEQHLSSSK-COOH, respectively, as described previously (63).
Construction of prokaryotic and eukaryotic EToV Mpro expression vectors. For prokaryotic expression of EToV Mpro, residues 10376 to 11472 of the EToV genome were RT-PCR amplified with oligonucleotide primers 1606 and 1607 (Table 1). A product of the anticipated size of 1,112 bp was obtained, which was gel purified, inserted into plasmid pQE9, and cloned in E. coli M15/pREP4 cells (QIAGEN) as described above. The resulting expression vector, pQE-SS-Mpro, codes for a 42-kDa polypeptide, encompassing the EToV Mpro domain and flanking sequences.
The expression construct pQE9-SS-Mpro-GST is a pQE-SS Mpro derivative, which consists of nucleotides (nt) 10376 to 11473 of EToV ORF1a and nt 258 to 935 of the glutathione S-transferase (GST) gene from plasmid pGEX2T (Amersham Pharmacia Biotech). This expression cassette was produced via PCR with oligonucleotides 2124 and 2125 and conventional cloning procedures. It codes for a 60-kDa fusion protein consisting of the 366-amino-acid-residue Mpro domain with flanking sequences linked by a proline residue to amino acid residues 1 to 226 of the GST protein.
For mammalian expression, we employed the recombinant vaccinia virus-based vTF7-3 system. To generate a suitable expression vector, the 1606/1607 PCR amplicon, digested with BamHI and EcoRV, was cloned into SacI/BamHI-digested pBluescript K/S+ (Strategene) together with a double-stranded linker, consisting of oligonucleotides 1721 and 1722. The resulting plasmid, pBS-SS-Mpro, contains residues 10376 to 11472 of the EToV genome, placed under the control of the bacteriophage T7 promoter and provided in frame with an initiation codon in an optimal context for translation initiation (40).
Site-directed mutagenesis of (putative) active-site residues was performed by splicing overlap extension-PCR mutagenesis (35) and conventional cloning techniques; mutations introduced in Mpro and the primers used are listed in Table 1.
The dual-expression construct pBS-SS-Mpro-dual, is a pBS-SS-Mpro derivative in which, downstream of the Mpro sequences, a second expression cassette is inserted, which is placed under the translational control of the encephalomyocarditis virus (EMCV) internal ribosomal entry site (IRES; taken from plasmid pTN2) (69) and which consists of nt 821 to 1128 of EToV ORF1a fused in frame to nt 10.378 to 11.472 of the EToV genome (encoding Mpro and surrounding sequences) and nt 3 to 483 of the EToV N gene. This chimeric expression cassette was produced via a series of splicing overlap extension-PCRs with oligonucleotides 1638, 1640, 1902, and 1914 through 1918 (Table 1) and conventional cloning procedures. It codes for a 66-kDa fusion protein, consisting as listed from the N terminus to C terminus of (i) residues 1 to 103 of pp1a; (ii) the 65 aa residues immediately upstream of Mpro; (iii) the 292-residue Mpro domain; (iv) the 8 residues downstream of Mpro, including the cleavage site; and (v) residues 2 to 160 of the EToV N protein. In this fusion protein, the N-terminal Mpro cleavage site is abolished (QSAG) and the Mpro proteinase is inactivated by an active site Ser-to-Ala substitution.
Purification and N-terminal sequence analysis of Mpro cleavage products. EToV Mpro and Mpro-GST were expressed from plasmids pQE-SS-Mpro and pQE9-SS-Mpro-GST, respectively, in E. coli M15/pREP4 as described above. Of the 1-ml bacterial culture, the cells were collected by centrifugation at 14,000 rpm for 30 s, resuspended in 100 μl Laemmli sample buffer, and incubated for 5 min at 95°C. Proteins were separated in 15% SDS-polyacrylamide gel electrophoresis (PAGE) gels (20 μl cleared supernatant/lane), transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories), and subsequently stained with Coomassie brilliant blue R-250. The regions containing the cleavage products of interest were excised. N-terminal sequence analysis was performed by subjecting the proteins to 5 to 8 rounds of Edman degradation using standard procedures with a pulse-liquid protein sequencer (model 476A; Applied Biosystems).
Infection/transfection of mammalian cells and metabolic labeling. Monolayers of Ederm cells (1 x 106 cells), grown in 35-mm wells, were washed once with PBS containing 50 μg/ml DEAE-dextran and then infected in PBS-DEAE with EToV at a multiplicity of infection of 5 PFU/cell. At 1 h postinfection (p.i.), the inoculum was replaced by 1 ml DMEM10 and incubation at 37°C continued.
Subconfluent monolayers of OST7-1 cells (2 x 106 cells), grown in 35-mm wells, were infected with recombinant vaccinia virus vTF7-3 and, at 1 h p.i., transfected with 2 μg plasmid DNA as described previously (46).
Prior to metabolic labeling, the cells were depleted for methionine and cysteine for 30 min by replacing the tissue culture supernatant with 600 μl of prewarmed starvation medium (DMEM without L-cysteine and L-methionine [Invitrogen] supplemented with 5% FCS, 10 mM hydroxyethylpiperazine-N'-2-ethanesulfonic acid [HEPES; pH 7.4], 1 mM Glutamax). Then 100 μCi of Tran35S-label (ICN) was added and the cells were labeled for the indicated length of time at 37°C.
After labeling, the cells were washed once with ice-cold PBS and then lysed on ice in 600 μl ice-cold lysis buffer (20 mM Tris-HCl, pH 7.5, 1 mM EDTA, 100 mM NaCl, 1% Triton X-100, and 1 μg/ml aprotinin, leupeptin, and pepstatin A each). Nuclei and cell debris were pelleted by centrifugation for 5 min at 14,000 rpm and 4°C, and the supernatant was subjected to radioimmunoprecipitation assay (RIPA). To this end, the lysates were adjusted to 2% SDS, diluted 10 times with detergent solution (50 mM Tris-HCl, pH 8.0, 62.5 mM EDTA, 0.4% Na deoxycholate, 1% NP-40, and 1 μg/ml of aprotinin, leupeptin, and pepstatin A each). RIPA was performed in a final volume of 1 ml with 3 μl of antiserum K158, K270, K272, K274 (see below), or R-EToV N (14). The immune complexes were adsorbed to Pansorbin cells (Calbiochem) for 30 min at 4°C, collected by centrifugation for 1 min at 14,000 rpm, and washed three times with detergent solution containing 0.25% SDS. The pellets were then resuspended in 25 μl Laemmli sample buffer and heated for 5 min at 95°C. Proteins were separated in 15% SDS-PAGE gels, which were fixed for 30 min in 50% methanol-10% acetic acid and dried for 60 min at 80°C. Labeled proteins were visualized by fluorography using Kodak intensifying screens.
RESULTS
Comparative sequence analysis of EToV Mpro. The EToV Berne ORF1a was sequenced completely (GenBank accession no. DQ310701) and encodes a product, pp1a, 4,569 amino acids long; a ribosomal frameshift at the pseudoknot at the 3' end of ORF1a (58) would produce a C-terminally extended product, pp1ab, of 6,857 amino acids. During the preparation of our article, Draker et al. (22) reported the completion of the sequence of the bovine torovirus (BToV) strain Breda genome. These authors identified a BToV pp1a domain with sequence similarity to the adenosine diphosphate-ribose 1"-phosphatase processing enzyme family (22), a homologue of which also occurs at a similar topological position in pp1a of coronaviruses, but not in those of roni -and arteriviruses (57). Moreover, two proteinase domains were found: a cysteine proteinase within the N-terminal half of pp1a and a more C-terminally located serine proteinase (22). The pp1a polyproteins of EToV and BToV share 65% sequence identity and the three domains described by Draker and colleagures (22) are conserved in pp1a of EToV. They were in fact identified independently using standard protein-protein and PSI and PHI BLAST searches of the NCBI database with EToV pp1a sequences as a query.
The EToV serine proteinase domain, comprising residues 3278 to 3430, displayed sequence similarity with the SpIB serine protease of Staphylococcus aureus and with the pp1a serine proteinase of turkey astrovirus. Similarity was particularly evident in the areas surrounding the predicted active-site residues, with His75, Asp113, and Ser193 of S. aureus SplB (50); GenBank accession no. AF271715) aligning with His3304, Glu3347, and Ser3416 in the EToV sequence (Fig. 1A). The putative EToV proteinase domain resides within pp1a/pp1ab at a relative position similar to that of the Mpros of other nidoviruses and, likewise, is flanked by hydrophobic regions (HR-1 and -2; Fig. 2A). Based upon these observations and upon the fact that no other chymotrypsin-like proteinase motifs were evident in the EToV pp1a/pp1ab sequence, we postulated the identified domain to represent the toroviral Mpro.
Although sequence conservation between the putative torovirus Mpro and the Mpros of other nidoviruses is very weak, its sequence could be fitted to existing Mpro alignments (3, 4, 74) in the regions containing the active-site His and Cys/Ser residues. The putative EToV Mpro is predicted to adopt a two--barrel structure with His3304 and Ser3416 as the main active-site residues and, again, Glu3347 as the most likely candidate for the role of third residue within the catalytic triad (Fig. 1B). From our data, the torovirus Mpro appears to be no more closely related to the coronavirus Mpro than to its roni- and arterivirus homologues. Perhaps most surprising, the torovirus Mpro differs from the Mpros of corona- and roniviruses in that it is not a cysteine proteinase, but rather, like its arterivirus homologue, uses serine as the principal nucleophile.
Proteolytic activity of EToV Mpro. The bioinformatical analyses of the putative EToV Mpro provided a theoretical framework for subsequent experimental studies. The hydrophilic region between HR-1 and -2, which spans the predicted proteinase domain, was expressed in E. coli (Fig. 2). A construct in which the predicted active-site His residue was replaced by Gly was used as a control (Fig. 2B). IPTG-induced expression of these constructs should yield a 42-kDa primary translation product. Indeed, a protein of this size was detected in cells expressing the His-to-Ala mutant. However, in cells expressing the intact proteinase domain, two products with apparent molecular masses of 33 and 16 kDa were found instead. These results suggested that the EToV pp1a region comprising residues 3186 to 3550 indeed contains a functional proteinase domain mediating autocatalytic processing.
The 33- and 16-kDa products were purified from E. coli and N-terminally sequenced by performing 5 or 8 cycles of Edman degradation, respectively. Analysis of the 33-kDa product indicated that cleavage had occurred at the sequence 3246SNFSFQSVFSKAT3258, thus identifying Ser3252 as the N-terminal residue of EToV Mpro (Fig. 3) (Note that, throughout this article, amino acid residues flanking the scissile bond, indicated by , are given from N to C terminus in the single-letter code, where "x" indicates any residue.) The 16-kDa product proved to be a mixture of two protein species. For one of these, the N terminus was identical to that of the 33-kDa product. The other product apparently arose from proteolytic cleavage at an internal sequence, 3392SEFATQAWQTVN3403(data not shown). Indeed, cleavage of Mpro at this site would cut the proteinase effectively in half, thus giving rise to two equal-size (16.5 kDa) products. These results tentatively identified FxxQ(S, A) as the EToV Mpro substrate consensus sequence (residues are given in parentheses to indicate the variation occurring at a particular position within the cleavage site).
Analysis of the Mpro flanking sequences identified a putative additional cleavage site downstream of the proteinase domain, 3540FKKQS3544. In the full-length precursor product encoded by plasmid pQE-SS-Mpro, this sequence is located only 5 residues upstream of the C terminus. Hence, cleavage at this site could not be confirmed. As the hydrophobic region HR-2, C-terminally flanking Mpro, might interfere with efficient expression in E. coli, we did not extend the expression product with EToV sequences, but instead designed an Mpro-fusion protein, which contained the predicted cleavage site including the downstream seven residues, linked via a newly introduced proline residue to residues 1 to 226 of GST (Fig. 2A). Expression in E. coli yielded, as anticipated, two major products of 33 kDa and 26 kDa (Fig. 2B), apparently corresponding to mature Mpro and the cleaved-off GST moiety. N-terminal sequence analysis of the 26-kDa product confirmed this notion and indicated that proteolytic cleavage indeed had occurred at the predicted site 3543GlnSer3544 (data not shown).
Cleavage at the sites upstream and downstream of the Mpro domain should yield a 33-kDa protein. Accordingly, in lysates from metabolically labeled EToV-infected cells, a product of this size, comigrating exactly with the bacterial cleavage product, was detected by RIPA with Mpro-specific antisera (Fig. 4) (data not shown). The 16-kDa products, however, were not detected in EToV-infected cells.
Mutagenesis of the predicted EToV Mpro active-site residues. Further analysis of the EToV Mpro was performed by employing the vaccinia virus-based vTF7-3 mammalian expression system (24) (Fig. 5). To this end, residues 10376 to 11472 of the EToV genome were provided with an AUG codon in an optimal context for translation initiation (40) and the resulting ORF was placed under the control of the bacteriophage T7 promoter (Fig. 5A). Expression of this construct predominantly yielded the 33-kDa Mpro species. To assess the importance of the predicted catalytic residues, single-amino-acid substitutions were introduced, and their effects were studied by analyzing the processing activities of the mutant proteinases. Consistent with their proposed catalytic function, replacement of His3304 by either Gly or Arg and of Ser3416 by Ala completely abolished proteolytic activity. Interestingly, however, an Mpro mutant with a Ser3416Cys substitution retained partial activity (Fig. 5B).
EToV Mpro displays the distinctive traits of a classical serine proteinase, and hence the existence of a third acidic catalytic residue was surmised. On the basis of the alignments (Fig. 1A and B), Glu3347 was considered the most likely candidate to occupy a main chain position equivalent to that of the catalytic Asp residue in chymotrypsin. In addition, Asp3383, Glu3354, and Asp3358 were selected for mutagenesis. However, the substitution for neither of these residues by Asn or Ala affected proteolytic processing of the 42-kDa Mpro precursor (Fig. 5C). One explanation for this result would be that yet another acidic residue is part of the catalytic triad. Still, the fact that our assay relied on autocatalytic cleavages that may occur in cis posed a caveat. Mutagenesis studies of other serine proteases have shown that the acidic active-site residue is not absolutely required for peptide hydrolysis, although its substitution may cause the enzymatic activity to drop by several orders of magnitude (11, 16, 33, 41, 64, 65). Yet, in an experimental setup in which processing occurs by highly efficient cleavage in cis, even such considerable differences in enzymatic activity may go unnoticed. We therefore designed a more sensitive trans-cleavage assay, outlined in Fig. 6A, which entailed coexpression of enzymatically active Mpro and a substrate protein. Enzyme and substrate were expressed from the same transcript with translation of the latter being driven by the EMCV IRES.
The substrate was a 71-kDa chimeric protein consisting of (i) residues 1 to 103 of pp1a/pp1ab, (ii) a EToV Mpro-containing segment consisting of pp1a/pp1ab residues 3187 to 3551, and (iii) residues 2 to 160 of the EToV N protein, as listed from N to C terminus. In this fusion protein, the N-terminal Mpro cleavage site was abolished by a QSAG substitution, while the Mpro was inactivated by changing the active site Ser3416 to Ala. However, the cleavage site C terminal of Mpro was left intact and hence remained available for proteolytic cleavage in trans. This should give rise to products of 52 kDa (the pp1a-Mpro fusion protein) and of 19 kDa (the cleaved-off N moiety), which are readily distinguishable from the enzymatically active Mpro and its precursors. Processing was monitored by RIPA with antisera against the N terminus of pp1a, Mpro, and the N protein and the effect of substitutions of the selected acidic residues was assessed (Fig. 6B). A construct in which the Mpro enzyme was inactivated by a Ser3416Ala substitution was taken along as a negative control.
Mpro derivatives, in which Asp3338, Glu3354, and Asp3358 had been altered, behaved like the wild-type enzyme in the trans-cleavage assay. Substitution of Glu3347, however, reduced proteolytic activity as indicated by accumulation of the 71-kDa noncleaved substrate precursor and of the Mpro precursors of 42 and 34 kDa; the latter protein species apparently represented an intermediate generated by processing of the 42-kDa Mpro precursor at the N-terminal but not at the C-terminal cleavage site (Fig. 6B).
DISCUSSION
The main proteinases, which mediate most of the processing steps of nidovirus replicase polyproteins, have been studied so far only for corona-, arteri-, and roniviruses. Here, we report the biochemical characterization of the fourth member, the torovirus Mpro. The nidoviral Mpros all belong to a family of viral and cellular proteinases, resembling picornavirus 3C (3C-like proteinases [3CLPs]). Moreover, they are located at cognate positions within the C-terminal half of pp1a, typically flanked by hydrophobic domains at either side (74, 76). Hence, the null hypothesis is that the proteinases of extant nidoviruses have all evolved from a single ancestral proto-nidovirus 3CLP. However, if so, they have diverged to an extreme extent. As noted previously (74, 76), primary sequence diversity of the corona-, roni-, and arterivirus Mpros is so pronounced that even the most sensitive computer-aided methods do not reveal any specific relationship among them that would justify their grouping and separation from other cellular or viral 3CLPs.
Within the order Nidovirales, the toroviruses seem to be most closely related to the coronaviruses (10, 25, 58, 60). We therefore anticipated at the onset of this study that the torovirus Mpro might bridge the large evolutionary gap between the main proteinase of the coronaviruses and those of the other nidoviruses (76). Surprisingly, however, from its primary sequence, the Mpro of EToV Berne seems no more related to the coronaviral proteinase than to those of the arteri- and roniviruses (Fig. 1B). The characteristics of the torovirus Mpro in relation to those of the other nidoviruses are discussed in the paragraphs below.
Mpro-mediated processing and cleavage site specificity. Nidoviral Mpros and related proteinases in other RNA viruses are released from adjoining polyprotein domains by autocatalytic processing (21, 76). Likewise, when pp1a segments, containing the predicted EToV Mpro domain, were heterologously expressed in prokaryotic and in mammalian cells, the proteinase liberated itself by cleaving both at an N-terminal site and at a C-terminal site. This yielded a mature Mpro of 33 kDa, a protein of which size was also detected in EToV-infected cells by RIPA with Mpro-specific antisera. Interestingly, in EToV-infected cells, autocatalytic release of Mpro might be inefficient as during a 3-h continuous metabolic labeling period a considerable proportion of the proteinase seemed to remain part of a large precursor (Fig. 4). Similar observations were reported for the arteri- and coronavirus Mpros (32, 54, 63, 71). It has been suggested, that the long-lived Mpro intermediates are membrane associated via the flanking hydrophobic domains. Anchoring of proteinase precursors to intracellular membranes might well be a central phenomenon in nidovirus polyprotein processing and/or replication (39, 54, 71).
N-terminal sequence analysis of mature EToV Mpro showed that autocatalytic cleavage occurred at the location 3246SNFSFQSVFS3255. The sequence 3538QPFKKQSVSN3547was identified as the C-terminal cleavage site. Unexpectedly, in prokaryotic cells, processing occurred also at a site internal to Mpro, 3392SEFATQAWQT3401, effectively cleaving the proteinase in half. It is tempting to speculate that, during EToV infection, Mpro might inactivate itself as part of a feedback autoregulatory mechanism or, otherwise, that processing at the internal site gives rise to an alternative cleavage pathway (71). However, we have no evidence that internal processing actually occurs in mammalian cells. The 16-kDa cleavage products were detected neither in the vTF7-3 expression system nor in EToV-infected cells. Hence, it cannot be excluded that internal cleavage, as observed in E. coli, is an artifact resulting from overexpression and possibly aided by protein misfolding, which might have exposed cleavage sites that are normally inaccessible. Intriguingly, however, the internal cleavage site is also conserved in BToV Mpro (22).
Catalytic center and substrate specificity. Although sequence identity between EToV Mpro and its nidovirus homologues or other viral and cellular chymotrypsin-like enzymes is limited, sequence alignments clearly implicated Ser3416 and His3304 as the main catalytic residues. Consistent with this notion, their replacement by Ala and by Gly or Arg, respectively, abrogated enzymatic activity. The Mpros of corona- and roniviruses (2, 73, 74), like the hepatitis A virus 3C proteinase (1, 7), employ a catalytic Cys-His dyad within the context of the chymotrypsin-fold. Under the assumption that the EToV Mpro is a classical serine proteinase, a third acidic catalytic residue was expected (20). Although sequence alignments of EToV Mpro with other proteinases were far from robust, particularly in the region that should contain this third residue, our analyses suggested Glu3347 as the most likely candidate. However, neither substitution mutagenesis of Glu3347 nor that of the proximal Asp3338, Glu3354, and Asp3358 noticeably affected autocatalytic release of Mpro in heterologous expression studies. Yet, in an assay that tested for processing in trans, the efficacy of Mpro cleavage of an artificial substrate was clearly reduced by substitution of Glu3347, but not by that of the other acidic residues. Although our findings do not prove that Glu3347 is included in a catalytic triad with Ser3416 and His3304, they would at least be consistent with this notion. This view might seem at odds with observations made for various other viral and cellular chymotrypsin-like proteinases, where site-directed mutagenesis of the third residue resulted in a far more dramatic loss of proteolytic activity (11, 16, 33, 41, 64, 65). Still, in the case of severe acute respiratory syndrome (SARS) coronavirus Mpro, replacement of the active-site Cys by Ser yields a functional enzyme. Although this mutant proteinase relies exclusively on a Ser-His catalytic dyad, its activity is merely 40-fold reduced as compared to that of wild-type SARS coronavirus Mpro (36). A reduction in activity of this magnitude might have only a modest effect on processing even in our trans-cleavage assay under conditions of vTF7-3-driven EToV Mpro overexpression. As the mapping of active-site residues by the combination of comparative sequence analysis and site-directed mutagenesis is inherently fraught with a measure of uncertainty, definitive answers with respect to catalytic center composition and the presence/identity of a third catalytic residue will require the elucidation of the torovirus Mpro crystal structure.
Comparison of the sequences flanking the scissile bonds of the internal and the N- and C-terminal cleavage sites tentatively identifies FxxQ(S, A) as the substrate consensus sequence of EToV Mpro. Its cleavage site specificity thus resembles that of other nidoviral Mpros, which, as is typical for 3CLPs, all require Gln or Glu at the P1 position (74, 76). The specificity for P1-(Gln, Glu) of the picorna-, arteri-, and coronavirus enzymes is determined by a strictly conserved His, which via its imidazole side chain hydrogen bonds to the carbonyl/carboxylate oxygens of the P1 side chain (1, 2, 4, 7, 44, 45, 47, 73, 74, 76). A corresponding His residue is also present in the torovirus Mpro (EToV His3430) as part of a Gly-X-His sequence (Fig. 1B), a motif well conserved in most 3CLPs (including arteri- and ronivirus Mpro), but saliently absent in the Mpros of coronaviruses; the latter have a Tyr-Met-His motif instead (2, 4, 34, 64, 74, 75). In many 3CLPs, conserved Ser/Thr residues, located five residues upstream of the catalytic Ser/Cys, are also part of the substrate-binding subsite S1 and assist in binding by forming additional hydrogen bonds with the carbonyl/carboxylate oxygen atoms of the P1 side chain (4, 44, 47, 48, 74). We propose that, in the torovirus Mpro, Thr3411 plays a similar role. This would again be a difference from the coronavirus Mpro, which lacks a corresponding Ser/Thr residue (2, 74).
The arteri- and ronivirus Mpros prefer glutamic acid at the P1 position of their substrates, whereas coronavirus Mpro requires glutamine (64, 74, 76). In arterivirus Mpros, this substrate preference is presumably conferred by a Ser residue (located three residues downstream of the S1 subsite His; S137 in EAV Mpro; see Fig. 1B), which would hydrogen bond with the carboxylate oxygen atom (that is, the one not interacting with the subsite S1 His/Thr pair) of the P1-Glu side chain (4). This Ser is not conserved in the corona- and torovirus Mpros; instead, at the corresponding positions, Glu and Leu residues, respectively, are found (Fig. 1b). The S1 specificity site of coronavirus Mpro consists of the main chain atoms of Met165, Glu166, and His172 and the side chains of Phe140 and His163, with the imidazole of the latter residue interacting with the P1-Gln side chain (2, 3, 73) (residue numbering according to SARS-CoV Mpro). Also with respect to S1 subsite composition, the torovirus main proteinase differs from that of coronaviruses. Although EToV Mpro, like its coronavirus homologues, apparently prefers P1-Gln, none of the coronavirus S1 subsite residues are conserved, except for His163. As yet, it is not clear from the EToV Mpro sequence how the P1-Gln is accommodated.
Tentative Mpro processing scheme of torovirus pp1a/pp1ab. Although each of the three established EToV Mpro cleavage sites conforms to the consensus FxxQ(S, A), we believe, on the basis of alignments of pp1a/pp1ab of coronaviruses and the toroviruses EToV and BToV, that Tyr/Met/Leu and Gly/Lys may also be tolerated at the P4 and P1' positions, respectively. Although the picture is still incomplete, the substrate specificity of the torovirus Mpro is reminiscent of and yet distinct from those of the corona-, arteri-, and roniviruses, which mainly conform to LQ(S, A), E (G, S, A), and VxHE(L, V), respectively (74, 76). In particular, a preference for a bulky hydrophobic residue at P4 and no apparent predilection for certain residues at P2 would give the EToV Mpro a unique cleavage site formula. A tentative Mpro processing scheme of EToV pp1a/1b is presented in Fig. 7, with the sequences and locations of potential cleavage sites listed in Table 2. Several points are of note. (i) All cleavage sites predicted for EToV are conserved in BToV (22). (ii) A putative cleavage site, 2870FKKQSV2875 (Table 2, site 1), identical in sequence to the one C terminal of the Mpro domain (Table 2, site 4), is found halfway in EToV pp1a at a position topologically similar to that of an accessory papain-like proteinase 2 site within coronavirus pp1a. (iii) Of the predicted sites in the ORF1b-encoded part of the replicase (Table 2, sites 10 to 14), only one, located between the RdRp and zinc finger/helicase domains, conforms to the proposed consensus FxxQ (S, A) sequence (site 10; Table 2). Future studies should determine whether the predicted sites within pp1a/pp1ab are actually cleaved by Mpro. (iv) At the very C terminus of torovirus pp1a, there is a domain that is absent from all other nidovirus pp1a proteins and that shares identity with a predicted cyclic phosphodiesterase (CPD) motif in the ns2A protein of group 2 coronaviruses and in the rotavirus VP3 guanylyl transferase (Fig. 7) (57, 60). Puzzlingly, a cleavage site C terminal of this domain is not immediately apparent, raising the question whether Mpro utilizes a noncanonical site, whether another viral or host proteinase is involved, or whether processing of pp1ab yields an RdRp with the CPD homologue attached; note that in the latter case, processing of pp1a would still produce an additional, "free" CPD.
Evolutionary implications. EToV Mpro combines several traits of its closest relatives. (i) As is typical for nidovirus Mpros, the EToV proteinase has a predicted three-domain structure, with two catalytic -barrel domains and an additional C-terminal domain of unknown function. (ii) With respect to substrate specificity, the EToV Mpro resembles its coronavirus homologue in its preference for P1-Gln. (iii) Then again, it has what appears to be a canonical substrate-binding subsite, S1, resembling that of arteri- and ronivirus Mpros, but clearly distinct from that of coronavirus Mpro. (iv) Finally, in contrast to the Mpros of corona- and roniviruses, but like that of arterivirus, the torovirus Mpro uses serine instead of cysteine as its principal nucleophile. Rather than clarifying the evolutionary history of the nidoviral Mpros, the analysis of a torovirus Mpro seems to make the picture even more complex. If anything, our findings further emphasize the unique properties of the coronavirus Mpro, which on the basis of its catalytic system and substrate-binding pocket, was already considered an outlier among viral and cellular chymotrypsin-like homologues (2, 74). It is still quite possible, however, that striking similarities between corona- and toroviral Mpros will become overt once the crystal structure of toroviral Mpro has been solved.
Corona- and toroviruses share many traits (e.g., with regard to virion composition, genome organization, and the extent of colinearity in the ORF1b-encoded region of pp1ab), which would justify their grouping together, separate from the arteri- and roniviruses (15, 19, 25, 26). However, in a recent study of Gonzalez et al. (25), entailing a systematic quantitative analysis of sequence conservation among nidoviral proteins, the expected sequence affinity between corona- and toroviruses was not evident. Thus, with the topology of the nidovirus tree undecided, it was suggested that "the nidovirus phylogeny must be verified later when the number and diversity of torovirus and ronivirus sequences will match those of arteriviruses and coronaviruses" (25). If it were to be firmly established that of the nidoviruses, corona- and toroviruses are, after all, the most closely related, and if this evolutionary relationship also holds for the Mpro (i.e., if no RNA recombination-based exchange of proteinase domains—or parts thereof—has occurred during nidoviral divergence), our observations would have one more major implication: namely that the transition from serine- to cysteine-based proteolytic catalysis (or vice versa) has happened more than once in the course of nidovirus evolution. In this respect, it is of interest that a mutant EToV Mpro with a Ser3416Cys substitution retained enzymatic activity and that, conversely, the catalytic site CysSer substitution in SARS-CoV Mpro also yields an active enzyme (36).
ACKNOWLEDGMENTS
We acknowledge the contribution of Alexander E. Gorbalenya to the bioinformatical analysis of EToV ORF1a-derived cDNA clones 133 and 135 and EToV pp1a, as well as to the design and analysis of some of the experiments. We are grateful to Peter J. M. Rottier and Jolanda D. F. de Groot-Mijnes for critical reading of the manuscript.
REFERENCES
Allaire, M., M. M. Chernaia, B. A. Malcolm, and M. N. James. 1994. Picornaviral 3C cysteine proteinases have a fold similar to chymotrypsin-like serine proteinases.Nature 369:72-76.
Anand, K., G. J. Palm, J. R. Mesters, S. G. Siddell, J. Ziebuhr, and R. Hilgenfeld. 2002. Structure of coronavirus main proteinase reveals combination of a chymotrypsin fold with an extra alpha-helical domain. EMBO J. 21:3213-3224.
Anand, K., J. Ziebuhr, P. Wadhwani, J. R. Mesters, and R. Hilgenfeld. 2003. Coronavirus main proteinase (3CLpro) structure: basis for design of anti-SARS drugs. Science 13:13.
Barrette-Ng, I. H., K. K. Ng, B. L. Mark, D. Van Aken, M. M. Cherney, C. Garen, Y. Kolodenko, A. E. Gorbalenya, E. J. Snijder, and M. N. James.2002 . Structure of arterivirus nsp4. The smallest chymotrypsin-like proteinase with an alpha/beta C-terminal extension and alternate conformations of the oxyanion hole. J. Biol. Chem. 277:39960-39966.
Bazan, J. F., and R. J. Fletterick. 1989. Detection of a trypsin-like serine protease domain in flaviviruses and pestiviruses. Virology 171:637-639.
Bazan, J. F., and R. J. Fletterick. 1988. Viral cysteine proteases are homologous to the trypsin-like family of serine proteases: structural and functional implications. Proc. Natl. Acad. Sci. USA 85:7872-7876.
Bergmann, E. M., S. C. Mosimann, M. M. Chernaia, B. A. Malcolm, and M. N. G. James.1997 . The refined crystal structure of the 3C gene product from hepatitis A virus: specific proteinase activity and RNA recognition. J. Virol. 71:2436-2448.
Bhardwaj, K., L. Guarino, and C. C. Kao. 2004. The severe acute respiratory syndrome coronavirus Nsp15 protein is an endoribonuclease that prefers manganese as a cofactor.J. Virol. 78:12218-12224.
Brierley, I., P. Digard, and S. C. Inglis. 1989. Characterization of an efficient coronavirus ribosomal frameshifting signal: requirement for an RNA pseudoknot. Cell 57:537-547.
Cavanagh, D. 1997. Nidovirales: a new order comprising Coronaviridae and Arteriviridae. Arch. Virol. 142:629-633.
Chambers, T. J., C. S. Hahn, R. Galler, and C. M. Rice. 1990. Flavivirus genome organization, expression, and replication. Annu. Rev. Microbiol. 44:649-688.
Chirnside, E. D., P. M. Francis, and J. A. Mumford. 1995. Expression cloning and antigenic analysis of the nucleocapsid protein of equine arteritis virus.Virus Res. 39:277-288.
Choi, H. K., L. Tong, W. Minor, P. Dumas, U. Boege, M. G. Rossmann, and G. Wengler. 1991. Structure of Sindbis virus core protein reveals a chymotrypsin-like serine proteinase and the organization of the virion. Nature 354:37-43.
Cornelissen, L. A. H. M. 1999. Molecular characterization of ungulate toroviruses. Ph.D. thesis. Utrecht, The Netherlands.
Cowley, J. A., C. M. Dimmock, K. M. Spann, and P. J. Walker. 2000. Gill-associated virus of Penaeus monodon prawns: an invertebrate virus with ORF1a and ORF1b genes related to arteri- and coronaviruses. J. Gen. Virol. 81:1473-1484.
Craik, C. S., S. Roczniak, C. Largman, and W. J. Rutter. 1987. The catalytic role of the active site aspartic acid in serine proteases. Science 237:909-913.
Demler, S. A., and G. A. de Zoeten. 1991. The nucleotide sequence and luteovirus-like nature of RNA 1 of an aphid non-transmissible strain of pea enation mosaic virus.J. Gen. Virol. 72:1819-1834.
den Boon, J. A., E. J. Snijder, E. D. Chirnside, A. A. F. de Vries, M. C. Horzinek, and W. J. M. Spaan.1991 . Equine arteritis virus is not a togavirus but belongs to the coronaviruslike superfamily. J. Virol. 65:2910-2920.
de Vries, A. A. F., M. C. Horzinek, P. J. M. Rottier, and R. J. de Groot.1997 . The genome organization of the Nidovirales: similarities and differences between Arteri-, Toro-, and Coronaviruses.Semin. Virol. 8:33-47.
Dodson, G., and A. Wlodawer. 1998. Catalytic triads and their relatives. Trends Biochem. Sci. 23:347-352.
Dougherty, W. G., and B. L. Semler. 1993. Expression of virus-encoded proteinases: functional and structural similarities with cellular enzymes. Microbiol. Rev. 57:781-822.
Draker, R., R. L. Roper, M. Petric, and R. Tellier.2005 . The complete sequence of the bovine torovirus genome. Virus Res. 115:56-68. (First published, 30 August 2005; doi:10.1016/jvirusres.2005.07.005.)
Elroy-Stein, O., and B. Moss. 1990. Cytoplasmic expression system based on constitutive synthesis of bacteriophage T7 RNA polymerase in mammalian cells. Proc. Natl. Acad. Sci. USA 87:6743-6747.
Fuerst, T. R., E. G. Niles, F. W. Studier, and B. Moss. 1986. Eukaryotic transient-expression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase. Proc. Natl. Acad. Sci. USA 83:8122-8126.
Gonzalez, J. M., P. Gomez-Puertas, D. Cavanagh, A. E. Gorbalenya, and L. Enjuanes. 2003. A comparative sequence analysis to revise the current taxonomy of the family Coronaviridae. Arch. Virol. 148:2207-2235.
Gorbalenya, A. E. 2001. Big nidovirus genome. When count and order of domains matter. Adv. Exp. Med. Biol. 494:1-17.
Gorbalenya, A. E., V. M. Blinov, and A. P. Donchenko. 1986. Poliovirus-encoded proteinase 3C: a possible evolutionary link between cellular serine and cysteine proteinase families. FEBS Lett. 194:253-257.
Gorbalenya, A. E., A. P. Donchenko, V. M. Blinov, and E. V. Koonin. 1989. Cysteine proteases of positive strand RNA viruses and chymotrypsin-like serine proteases.A distinct protein superfamily with a common structural fold. FEBS Lett. 243:103-114.
Gorbalenya, A. E., E. V. Koonin, V. M. Blinov, and A. P. Donchenko. 1988. Sobemovirus genome appears to encode a serine protease related to cysteine proteases of picornaviruses. FEBS Lett. 236:287-290.
Gorbalenya, A. E., E. V. Koonin, A. P. Donchenko, and V. M. Blinov. 1989. Coronavirus genome: prediction of putative functional domains in the non-structural polyprotein by comparative amino acid sequence analysis. Nucleic Acids Res. 17:4847-4861.
Gorbalenya, A. E., E. V. Koonin, and M. M. Lai.1991 . Putative papain-related thiol proteases of positive-strand RNA viruses. Identification of rubi- and aphthovirus proteases and delineation of a novel conserved domain associated with proteases of rubi-, alpha- and coronaviruses. FEBS Lett. 288:201-205.
Gosert, R., A. Kanjanahaluethai, D. Egger, K. Bienz, and S. C. Baker. 2002. RNA replication of mouse hepatitis virus takes place at double-membrane vesicles. J. Virol. 76:3697-3708.
Hedstrom, L. 2002. Serine protease mechanism and specificity.Chem. Rev. 102:4501-4524.
Hegyi, A., and J. Ziebuhr. 2002. Conservation of substrate specificities among coronavirus main proteases. J. Gen. Virol. 83:595-599.
Horton, R. M., and L. R. Pease. 1991. Recombination and mutagenesis of DNA sequences using PCR, p.217 -247. In M. J. McPherson (ed.), Directed mutagenesis, a practical approach. IRL Press, Oxford, United Kingdom.
Huang, C., P. Wei, K. Fan, Y. Liu, and L. Lai. 2004. 3C-like proteinase from SARS coronavirus catalyzes substrate hydrolysis by a general base mechanism. Biochemistry 43:4568-4574.
Ivanov, K. A., T. Hertzig, M. Rozanov, S. Bayer, V. Thiel, A. E. Gorbalenya, and J. Ziebuhr. 2004. Major genetic marker of nidoviruses encodes a replicative endoribonuclease. Proc. Natl. Acad. Sci. USA 101:12694-12699.
Jiang, B., S. S. Monroe, E. V. Koonin, S. E. Stine, and R. I. Glass. 1993. RNA sequence of astrovirus: distinctive genomic organization and a putative retrovirus-like ribosomal frameshifting signal that directs the viral replicase synthesis. Proc. Natl. Acad. Sci. USA 90:10539-10543.
Kanjanahaluethai, A., and S. C. Baker. 2000. Identification of mouse hepatitis virus papain-like proteinase 2 activity.J. Virol. 74:7911-7921.
Kozak, M. 1986. Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44:283-292.
Lawson, M. A., and B. L. Semler. 1991. Poliovirus thiol proteinase 3C can utilize a serine nucleophile within the putative catalytic triad. Proc. Natl. Acad. Sci. USA 88:9919-9923.
Li, X., M. D. Ryan, and J. W. Lamb.2000 . Potato leafroll virus protein P1 contains a serine proteinase domain. J. Gen. Virol. 81:1857-1864.
Lu, Y., X. Lu, and M. R. Denison. 1995. Identification and characterization of a serine-like proteinase of the murine coronavirus MHV-A59. J. Virol. 69:3554-3559.
Matthews, D. A., P. S. Dragovich, S. E. Webber, S. A. Fuhrman, A. K. Patick, L. S. Zalman, T. F. Hendrickson, R. A. Love, T. J. Prins, J. T. Marakovits, R. Zhou, J. Tikhe, C. E. Ford, J. W. Meador, R. A. Ferre, E. L. Brown, S. L. Binford, M. A. Brothers, D. M. DeLisle, and S. T. Worland. 1999. Structure-assisted design of mechanism-based irreversible inhibitors of human rhinovirus 3C protease with potent antiviral activity against multiple rhinovirus serotypes. Proc. Natl. Acad. Sci. USA 96:11000-11007.
Matthews, D. A., W. W. Smith, R. A. Ferre, B. Condon, G. Budahazi, W. Sisson, J. E. Villafranca, C. A. Janson, H. E. McElroy, C. L. Gribskov, et al. 1994. Structure of human rhinovirus 3C protease reveals a trypsin-like polypeptide fold, RNA-binding site, and means for cleaving precursor polyprotein. Cell 77:761-771.
Mijnes, J. D. F., L. M. van der Horst, E. van Anken, M. C. Horzinek, P. J. M. Rottier, and R. J. de Groot. 1996. Biosynthesis of glycoproteins E and I of feline herpesvirus: gE-gI interaction is required for intracellular transport. J. Virol. 70:5466-5475.
Mosimann, S. C., M. M. Cherney, S. Sia, S. Plotch, and M. N. James. 1997. Refined X-ray crystallographic structure of the poliovirus 3C gene product. J. Mol. Biol. 273:1032-1047.
Nienaber, V. L., K. Breddam, and J. J. Birktoft.1993 . A glutamic acid specific serine protease utilizes a novel histidine triad in substrate binding. Biochemistry 32:11469-11475.
Paetzel, M., and R. E. Dalbey. 1997. Catalytic hydroxyl/amine dyads within serine proteases. Trends Biochem. Sci. 22:28-31.
Reed, S. B., C. A. Wesson, L. E. Liou, W. R. Trumble, P. M. Schlievert, G. A. Bohach, and K. W. Bayles. 2001. Molecular characterization of a novel Staphylococcus aureus serine protease operon. Infect. Immun. 69:1521-1527.
Rice, C. M., and J. H. Strauss. 1981. Nucleotide sequence of the 26S mRNA of sindbis virus and deduced sequence of the encoded virus structural proteins. Proc. Natl. Acad. Sci. USA 78:2062-2066.
Ryan, M. D., and M. Flint. 1997. Virus-encoded proteinases of the picornavirus super-group. J. Gen. Virol. 78:699-723.
Ryan, M. D., S. Monaghan, and M. Flint. 1998. Virus-encoded proteinases of the Flaviviridae. J. Gen. Virol. 79:947-959.
Schiller, J. J., A. Kanjanahaluethai, and S. C. Baker.1998 . Processing of the coronavirus MHV-JHM polymerase polyprotein: identification of precursors and proteolytic products spanning 400 kilodaltons of ORF1a. Virology 242:288-302.
Shanks, M., and G. P. Lomonossoff. 1990. The primary structure of the 24K protease from red clover mottle virus: implications for the mode of action of comovirus proteases.J. Gen. Virol. 71:735-738.
Siddell, S. G., J. Ziebuhr, and E. J. Snijder.2005 . Coronaviruses, toroviruses and arteriviruses, p.823 -856. In B. W. Mahy and V. ter Meulen (ed.), Topley and Wilson's microbiology and microbial infections. Virology. Hodder Arnold, London, United Kingdom.
Snijder, E. J., P. J. Bredenbeek, J. C. Dobbe, V. Thiel, J. Ziebuhr, L. L. Poon, Y. Guan, M. Rozanov, W. J. Spaan, and A. E. Gorbalenya.2003 . Unique and conserved features of genome and proteome of SARS-coronavirus, an early split-off from the coronavirus group 2 lineage. J. Mol. Biol. 331:991-1004.
Snijder, E. J., J. A. den Boon, P. J. Bredenbeek, M. C. Horzinek, R. Rijnbrand, and W. J. Spaan.1990 . The carboxyl-terminal part of the putative Berne virus polymerase is expressed by ribosomal frameshifting and contains sequence motifs which indicate that toro- and coronaviruses are evolutionarily related. Nucleic Acids Res. 18:4535-4542.
Snijder, E. J., J. A. den Boon, M. C. Horzinek, and W. J. Spaan. 1991. Characterization of defective interfering RNAs of Berne virus. J. Gen. Virol. 72:1635-1643.
Snijder, E. J., and M. C. Horzinek. 1993. Toroviruses: replication, evolution and comparison with other members of the coronavirus-like superfamily. J. Gen. Virol. 74:2305-2316.
Snijder, E. J., M. C. Horzinek, and W. J. M. Spaan. 1990. A 3'-coterminal nested set of independently transcribed mRNAs is generated during Berne virus replication. J. Virol. 64:331-338.
Snijder, E. J., S. G. Siddell, and A. E. Gorbalenya. 2005. The order Nidovirales, p.390 -404. In B. W. Mahy and V. ter Meulen (ed.), Topley and Wilson's microbiology and microbial infections. Virology. Hodder Arnold, London, United Kingdom.
Snijder, E. J., A. L. M. Wassenaar, and W. J. M. Spaan. 1994. Proteolytic processing of the replicase ORF1a protein of equine arteritis virus.J. Virol. 68:5755-5764.
Snijder, E. J., A. L. Wassenaar, L. C. van Dinten, W. J. Spaan, and A. E. Gorbalenya.1996 . The arterivirus nsp4 protease is the prototype of a novel group of chymotrypsin-like enzymes, the 3C-like serine proteases.J. Biol. Chem. 271:4864-4871.
Sprang, S., T. Standing, R. J. Fletterick, R. M. Stroud, J. Finer-Moore, N. H. Xuong, R. Hamlin, W. J. Rutter, and C. S. Craik. 1987. The three-dimensional structure of Asn102 mutant of trypsin: role of Asp102 in serine protease catalysis. Science 237:905-909.
ten Dam, E., M. Flint, and M. D. Ryan. 1999. Virus-encoded proteinases of the Togaviridae. J. Gen. Virol. 80:1879-1888.
van der Meer, Y., H. van Tol, J. K. Locker, and E. J. Snijder. 1998. ORF1a-encoded replicase subunits are involved in the membrane association of the arterivirus replication complex. J. Virol. 72:6689-6698.
Van Vliet, A. L., S. L. Smits, P. J. Rottier, and R. J. De Groot. 2002. Discontinuous and non-discontinuous subgenomic RNA transcription in a nidovirus.EMBO J. 21:6571-6580.
Vennema, H., R. Rijnbrand, L. Heijnen, M. C. Horzinek, and W. J. Spaan. 1991. Enhancement of the vaccinia virus/phage T7 RNA polymerase expression system using encephalomyocarditis virus 5'-untranslated region sequences.Gene 108:201-209.
Wang, R. F., and S. R. Kushner. 1991. Construction of versatile low-copy-number vectors for cloning, sequencing and gene expression in Escherichia coli. Gene 100:195-199.
Wassenaar, A. L. M., W. J. M. Spaan, A. E. Gorbalenya, and E. J. Snijder.1997 . Alternative proteolytic processing of the arterivirus replicase ORF1a polyprotein: evidence that NSP2 acts as a cofactor for the NSP4 serine protease. J. Virol. 71:9313-9322.
Wellink, J., and A. van Kammen. 1988. Proteases involved in the processing of viral polyproteins. Brief review. Arch. Virol. 98:1-26.
Yang, H., M. Yang, Y. Ding, Y. Liu, Z. Lou, Z. Zhou, L. Sun, L. Mo, S. Ye, H. Pang, G. F. Gao, K. Anand, M. Bartlam, R. Hilgenfeld, and Z. Rao. 2003. The crystal structures of severe acute respiratory syndrome virus main protease and its complex with an inhibitor. Proc. Natl. Acad. Sci. USA 100:13190-13195.
Ziebuhr, J., S. Bayer, J. A. Cowley, and A. E. Gorbalenya. 2003. The 3C-like proteinase of an invertebrate nidovirus links coronavirus and potyvirus homologs.J. Virol. 77:1415-1426.
Ziebuhr, J., G. Heusipp, and S. G. Siddell. 1997. Biosynthesis, purification, and characterization of the human coronavirus 229E 3C-like proteinase. J. Virol. 71:3992-3997.
Ziebuhr, J., E. J. Snijder, and A. E. Gorbalenya.2000 . Virus-encoded proteinases and proteolytic processing in the Nidovirales. J. Gen. Virol. 81:853-879.(Saskia L. Smits, Eric J. )