【病毒外文文献】2005 Study on Substrate Specificity at Subsites for Severe Acute Respiratory Syndrome Coronavirus 3CL Protease

ISSN 1672 9145 Acta Biochimica et Biophysica Sinica 2005 37 12 807 813 CN 31 1940 Q Institute of Biochemistry and Cell Biology SIBS CAS Study on Substrate Specificity at Subsites for Severe Acute Respiratory Syndrome Coronavirus 3CL Protease Yu Fei SHAN 1 and Gen Jun XU 1 2 1 Institute of Biochemistry and Cell Biology Shanghai Institutes for Biological Sciences Chinese Academy of Sciences Shanghai 200031 China 2 College of Life Science Zhejiang Sci Tech University Hangzhou 310018 China Abstract Autocleavage assay and peptide based cleavage assay were used to study the substrate specificity of 3CL protease from the severe acute respiratory syndrome coronavirus It was found that the recognition between the enzyme and its substrates involved many positions in the substrate at least including residues from P4 to P2 The deletion of either P4 or P2 residue in the substrate would decrease its cleavage efficiency dramatically In contrast to the previous suggestion that only small residues in substrate could be accommodated to the S1 subsite we have found that bulky residues such as Tyr and Trp were also acceptable In addition based on both peptide based assay and autocleavage assay Ile at the P1 position could not be hydrolyzed but the mutant L27A could hydrolyze the Ile peptide fragment It suggested that there was a stereo hindrance between the S1 subsite and the side chain of Ile in the substrate All 20 amino acids except Pro could be the residue at the P2 position in the substrate but the cleavage efficiencies were clearly different The specificity information of the enzyme is helpful for potent anti virus inhibitor design and useful for other coronavirus studies Key words SARS coronavirus 3CL protease substrate specificity autocleavage binding site inhibitor design Received August 23 2005 Accepted September 26 2005 This work was supported by a grant from the National Natural Science Foundation of China No 39930060 Corresponding author Tel 86 21 54921257 Fax 86 21 54921257 E mail gjxu DOI 10 1111 j 1745 7270 2005 00114 x The causative pathogen of severe acute respiratory syndrome SARS is a kind of positive strand RNA virus and belongs to the Coronaviridae family 1 2 In the infected cells the major part of its genome is directly expressed as large polyprotein pp1a or pp1ab precursors which undergo a series of controlled proteolytic processing to generate functional viral proteins 3 4 A 34 kDa virally encoded protease named 3CL also called the main protease plays a crucial role in the cleavage cascade The 3CL proteases or 3C like proteases also exist in other viruses many of which are infamous pathogens and me diate the viral polyprotein processing in the same way 5 The detailed information of substrate specificity is the theoretical base for inhibitor design 6 and great efforts have been made in this regard to throw light on the features of SARS coronavirus CoV 3CL protease Based on the homology and highly conserved substrate specificities 7 the cleavage sites of the 3CL protease in SARS CoV polyprotein pp1ab approximately 750 kDa were suggested to contain Q S A G N dipeptides the clea vage site is indicated by 8 The peptide based cleavage assay in vitro confirmed those 11 putative cleavage sites 9 The crystal structure resolved by Yang et al provided an unexpected binding mode between the SARS CoV 3CL protease and its substrate analog inhibitor and it was used to explain the less stringent S2 subsite specificity of the enzyme 10 However the inhibitor used in crystal diffraction did not include the residues downstream of the scissile bond It is insufficient for the design of a potent inhibitor for clinical trial We initiated this study to get more information about the substrate specificity of the SARS CoV 3CL protease especially on P1 and P2 positions In peptide based cleavage assay it was found that many residues had effects on the cleavage efficiency and the deletion of residues 808 Acta Biochim Biophys Sin Vol 37 No 12 Institute of Biochemistry and Cell Biology SIBS CAS downstream of the P1 position or upstream of the P3 position in the substrate would decrease its hydrolysis rate dramatically It suggested P2 and P4 residues may play important roles in the recognition between substrate and enzyme A number of replacements at P1 P2 and P4 positions in substrates were made and examined by autocleavage assay It was found that all residues except Pro were acceptable at the P2 position in substrate and bulky residues could also be allowed at the P1 position We also found that Ile could not be the residue at the P1 position in substrate but it could be hydrolyzed by the enzyme mutant L27A Leu27 of the enzyme was replaced by Ala As for the P4 position all eight residues tested were acceptable Materials and Methods Cloning and expression of SARS CoV 3CL protease The gene was amplified by polymerase chain reaction PCR using forward primer NP2 and reverse primer CP2 All the primer sequences used in this study are listed in Table 1 The PCR product was purified and digested with NcoI and XhoI then inserted into NcoI XhoI cut plasmid pET 28a The constructed plasmid pET28a 3CLsm encodes the SARS CoV 3CL protease in which the first amino acid Ser was replaced by Met pET28a 3CLsm was transformed into Escherichia coli BL21 DE3 and a single selected colony was grown in one liter of Luria Broth LB medium containing kanamycin 50 g ml When the absorbance of culture at 600 nm reached 0 6 it was induced with 10 M isopropyl D thiogalactopyranoside IPTG for 6 h Cells were harvested and the enzyme was purified with a metal affinity column as described previously 11 Finally the protease was made in buffer A 20 mM Tris HCl 200 mM NaCl 1 mM EDTA 1 mM dithiothreitol DTT 10 V V glycerol pH 7 5 and kept at 80 C as the stock solution 2 mg ml A 280 2 5 Synthetic peptide substrates A series of peptides which were derived from the cleavage site of the SARS CoV 3CL protease at its N terminal were prepared by the solid phase method The synthetic pep tides were purified through a Beckman ultrasphere C 18 reverse phase high performance liquid chromatography Table 1 Oligonucleotide sequences of primers for amplification or site directed mutagenesis The oligonucleotide sequences indicating mutated codons are underlined Pri1 five primers with different codons shown by NNN the codons used for each mutant were CGC Arg TGG Trp ATT Ile TTT Phe TAC Tyr Pri2 11 primers with different codons shown by NNN the codons used for each mutant were TTT Phe TGG Trp CAC His CCA Pro TAC Tyr TGT Cys GAC Asp ATG Met CAA Gln ACT Thr AGG Arg Pri4 eight primers with different codons shown by NNN the codons used for each mutant were TAC Tyr AGG Arg CCA Pro ATG Met TGT Cys CTG Leu ATT Ile GGT Gly Primer NP1 NP2 CP2 Pri1 Pri2 Pri4 P27A P27A P48A P48A P48K P48K Sequence 5 3 ATGGCTAGCTCAATCACTTCTGCTGTTCTGCAGAGTGGTTT TAGGAAAATGGCA TATACCATGGGTTTTAGGAAAATGGCATTC GGTGCTCGAGTTGGAAGGTAACACCAGAGCATTGTC ATGGCTAGCTCAATCACTTCTGCTGTTCTGCAGNNNGGTTT TAGGAAAATGGCA ATGGCTAGCTCAATCACTTCTGCTGTTCTGCAGAGTNNNTT TAGGAAAATGGCA ATGGCTAGCTCAATCACTTCTNNNGTTCTGCAGAGTGGTTT TAGGAAAATGGCA TGTGGAACTACAACTGCTAATGGATTGTGGTTG CAACCACAATCCATTAGCAGTTGTAGTTCCACA ATTTGCACAGCAGAAGCTATGCTTAATCCTAAC GTTAGGATTAAGCATAGCTTCTGCTGTGCAAAT ATTTGCACAGCAGAAAAGATGCTTAATCCTAAC GTTAGGATTAAGCATCTTTTCTGCTGTGCAAAT Polarity Forward Forward Reverse Forward Forward Forward Forward Reverse Forward Reverse Forward Reverse Remark For N terminal amplification For N terminal amplification For C terminal amplification For mutations at P1 position For mutations at P2 position For mutations at P4 position For mutating Leu27 to Ala For mutating Asp48 to Ala For mutating Asp48 to Lys Dec 2005 Yu Fei SHAN et al Substrate Specificity at Subsites for Severe Acute Respiratory Syndrome Coronavirus 3CL Protease 809 http www abbs info RP HPLC column 4 6 mm 250 mm The identity and homogeneity of peptides were confirmed by mass spectroscopy and RP HPLC The sequences of peptides are shown in Ta ble 2 Plasmids of proenzyme for autocleavage assay To construct the plasmids encoding proenzyme for autocleavage assay a pair of primers NP1 and CP2 was used The PCR product was digested with NheI and XhoI and then inserted into NheI XhoI cut plasmid pET 28a The constructed plasmid pET28a 3CLsc encodes a proenzyme with a 31 mer leading peptide N 31 tag at its N terminal and a His 6 tag at C terminal There is an autocleavage site between N 31 tag and the first amino acid Ser of the protease If autocleavage occurs N 31 tag would be cut off and the plasmid will produce a band with an apparent molecular weight of 35 kDa on sodium dodecyl sulfate polyacrylamide gel electrophoresis SDS PAGE Otherwise the molecular weight of the proenzyme would be approximately 39 kDa Site directed mutagenesis of the enzyme Site directed mutagenesis at P4 P1 and P2 positions on N 31 tag was performed by a recombination PCR method the position of amino acids in substrates named from N to C terminal as follows P4 P3 P2 P1 P1 P2 P3 P4 13 Briefly the gene of the SARS 3C like protease was amplified by PCR using forward primers containing mutational codons and a reverse primer CP2 The PCR products were purified and digested with NheI and XhoI then inserted into NheI XhoI cut plasmid pET 28a Plas mids of mutants were transformed into BL21 DE3 Single colonies were selected to grow in LB medium and induced with 10 M IPTG for 4 8 h when A 600 got to 0 6 The cells were harvested and analyzed by 12 SDS PAGE Mutagenesis of Leu27 and Asp48 was performed by three step PCR in a similar way Results Expression and purification of SARS CoV 3CL pro tease As the first N terminal amino acid of the enzyme is Ser three strategies were used to express SARS CoV 3CL protease in E coli The first strategy was replacing the first Ser with Met the second was adding an additional Met prior to Ser and the third was introducing a leading sequence to the N terminal which was cut off by pro teases after expression It was found that the specific activities of all the resulting enzymes were compatible as reported previously We had successfully over expressed the SARS CoV 3CL protease in E coli by mutating the Table 2 Relative severe acute respiratory syndrome coronavirus 3CL protease cleavage efficiencies of synthetic peptides with different sequences Peptide Sequence Cleavage efficiency Sub1 T S I T S A V L Q S G F R K M A 4 78 Sub2 T S A V L Q S G F R K M A 4 52 Sub3 S A V L Q S G F R K M A 4 49 Sub4 A V L Q S G F R K M A 3 61 Sub5 V L Q S G F R K M A UD Sub0 T S A V L Q S G F 1 00 Sub6 T S A V L Q S G 0 42 Sub7 T S A V L Q S UD Sub8 T S A V L Q S Y F 0 17 Sub9 T S A V L Q S W F 0 22 Sub10 T S A V L Q S P F UC Sub11 T S A V L Q I G F UC Sub12 T S A V L Q W G F 0 05 UD the cleavage efficiency for the peptides was too low to be determined in competitive cleavage reaction UC the peptides were not cleaved under the condi tions provided in Materials and Methods Peptide based cleavage assay in vitro Cleavage assays were performed in buffer B 20 mM Tris HCl pH 7 35 200 mM NaCl 1 mM DTT and 1 mM EDTA containing 200 M of each peptide and 1 M of enzyme The reaction mixtures were incubated at 20 C for 0 16 h and quenched by trifluoroacetic acid 1 final concentration The products were analyzed by RP HPLC using a 0 75 linear gradient of acetonitrile containing 0 1 trifluoroacetic acid The peaks were collected and identified by mass spectroscopy The cleavage efficiency of different peptides was compared by relative specificity constant k cat K m rel according to Cordingley et al 12 Briefly 200 M of each peptide was incubated in buffer B with 1 M enzyme in the presence of 200 M peptide Sub0 for competitive cleavage reaction The peak areas of products remained linear at the used time intervals generally less than 10 substrate hydrolyzed and the peak areas were integrated to calculate k cat K m rel 810 Acta Biochim Biophys Sin Vol 37 No 12 Institute of Biochemistry and Cell Biology SIBS CAS first amino acid Ser to Met The purification procedure was described previously by using a metal nitrilotriacetic acid column followed by a gel filtration column Sephacryl S 200 GE Healthcare USA 9 11 14 The product showed a single band on SDS PAGE After determining the concentration the enzyme was subsequently employed in peptide based cleavage assays Peptide based cleavage assay for core part of sub strates In order to obtain information about how the length of substrate affects the enzyme catalysis a group of peptides were synthesized These peptides of different lengths were derived from the sequence TSITSAVLQSGFRKMA one of the cleavage sites for the SARS CoV 3CL protease which was previously proved to be an efficient substrate 11 The capability of various peptides to compete with a 9 mer peptide Sub0 P6 P3 for cleavage by the 3CL protease was estimated by RP HPLC The k cat K m rel for each peptide compared to the reference substrate Sub0 is shown in Fig 1 for Sub0 P6 P3 The 8 mer peptide Sub6 P6 P2 had a k cat K m rel value comparable to that of Sub0 It suggested that residues of P3 P7 taken as a whole might contribute some energy to the binding between the enzyme and its substrate as the rate of hydrolysis enhanced The cleavage rates for Sub7 7 mer P6 P1 and Sub5 10 mer P3 P7 were too low to be determined in the competitive cleavage assay It meant that the deletion of residue P4 or P2 would make it difficult for a substrate to be hydrolyzed by the SARS CoV 3CL protease suggesting that residues P4 and P2 are two critical positions in the interaction between the enzyme and its substrate P1 has been reported to play an important role in the recognition of the enzyme and the substrate But systematic investigation of the P1 position has not been reported in published work Requirements of P1 P2 and P4 positions in substrate It was proposed that the P1 position is one of the important determinants in substrate specificity for 3CL 3C protease 3 5 7 15 P1 positions of 11 known cleavage sites in the SARS CoV 3CL protease are all occupied by amino acid residues with small side chains Ser Ala Gly and Asn Asn at the P1 position is regarded as a noncanonical residue 8 Consequently it was suggested in published reports that the SARS CoV 3CL protease similar to other 3CL proteases had a relatively small S1 subsite required for P1 residue binding that may exclude bulky amino acid residues In a previous study we successfully constructed an autocleavage assay by introducing an N 31 tag to the N terminal of the SARS CoV 3CL protease Fig 2 B We found that the expression product of a mutant proenzyme of C145S Cys145 of the enzyme was replaced by Ser appeared in two bands One was 38 5 kDa the other was 34 6 kDa Fig 2 A lane 8 It was ascribed to an incomplete proteolysis of the mutant Therefore the mutant served as a marker for estimating molecular weights The assay was simple sensitive and reliable so that it was employed to study the requirements of substrate for the SARS CoV 3CL protease To verify the suggestion above P1 Ser in N 31 tag was mutated to a number of bulky amino acids Tyr Ile Phe Trp and Lys Interestingly all these mutants showed a 35 kDa band on SDS PAGE except the P1 I mutant SARS CoV 3CL protease flanked an N 31 tag in which the P1 Ser was replaced by Ile the other mutants are designated similarly Fig 3 It implied that most of the mutant proen zymes effectively performed autocleavage That is to say the S1 subsite could accommodate bulky amino acids According to the peptide based cleavage assay the P2 residue was more important than those of P3 to P7 But Fig 1 Peptide based competitive cleavage assay In buffer B 200 M of each peptide was incubated with 1 M enzyme in the presence of 200 M peptide Sub0 for competitive cleavage reaction at 20 C Products were analyzed by RP HPLC The relative cleavage efficiency was the capability of various peptides to compete with a 9 mer peptide Sub0 P6 P3 for cleavage by the 3CL protease Results of competition assay showed that peptides Sub1 P9 P7 Sub2 P6 P7 Sub3 P5 P7 and Sub4 P4 P7 had similar k cat K m rel values It indicated that resi dues P5 P9 contributed little to the cleavage reaction The k cat K m rel value for Sub2 P6 P7 is about five times that Dec 2005 Yu Fei SHAN et al Substrate Specificity at Subsites for Severe Acute Respiratory Syndrome Coronavirus 3CL Protease 811 http www abbs info few reports supported the importance of the P2 position to the substrate specificity for 3CL proteases and related 3C proteases In viral protein pp1ab nine amino acids Gly Lys Ile Asn Ala Val Glu Leu and Ser were found at P2 positions near the cleavage sites So we replaced P2 Gly with the other 11 residues Cys Asp Phe His Met Pro Gln Arg Thr Trp and Tyr It was found that P2 could accept a wide range of amino acid residues except Pro as shown by autocleavage assay in which P2 P resulted in only one band with an apparent molecular weight of 39 kDa Fig 2 A lane 6 whereas the others yielded a band of 35 kDa Fig 2 A C It clearly indicated that only Pro could not be allowed at the P2 position The P4 position was also suggested as critical to the substrate specificity of 3CL proteases because of its limited variety in that only four kinds of residues Ala Val Pro and Thr had been found in the cleavage sites for the SARS CoV 3CL protease 9 In this study eight mutants Tyr Arg Met Pro Cys Leu Ile and Gly were constructed and tested by autocleavage assay Results showed that all the mutants effectively performed autocleavage as shown by producing an apparent molecular weight of 35 kDa on SDS PAGE data not shown The region of the N terminal in coronavirus 3CL proteases is essential 3 16 The deletion of the N terminal residues 1 5 in the SARS CoV 3CL protease eliminated most of its activity unpublished data In regard to the fact that the amino acids at P1 and P2 positions in N 31 tag were also the first two N terminal residues in the enzyme modification of any position may affect the enzyme activity It was also reported previously that the trans cleavage assay might have a different result from the cis cleavage assay 3 11 Both of them may consequently lead to an artificial conclusion To exclude the possibilities a parallel experiment of peptide based cleavage assay should be conducted Five synthetic 9 mer peptides with a single point substitution at P2 in the substrate Table 2 were afforded for trans cleavage assay Peptide Sub2 served as the positive control Fig 4 A B SARS CoV 3CL protease hydrolyzed Sub8 Tyr at P2 and Sub9 Trp at P2 at a relatively low rate Fig 1 The hydrolysis rate for Sub12 Trp at P1 was even lower and undetectable for Sub10 Pro at P2 even with a higher concentration of enzyme 10 M and a prolonged incubation time 48 h employed Fig 4 C D Site directed mutagenesis on Leu27 and Asp48 Based on the fact that the SARS CoV 3CL protease could not hydrolyze the substrate with Ile at the P1 position it is reasonable to suggest a repulsion interaction may exist in Ile substrate According to the suggestion of Anand et al the residues around S1 in the SARS CoV 3CL protease are comprised of Leu27 His41 and Asp48 8 In our previous study we found the mutant of H41A was inactive so that the mutation of His41 was not under consideration Fig 3 Autocleavage assay of P2 position mutants Total cell lysates of isopropyl D thiogalactopyranoside induced E coli samples were separated by 12 sodium dodecyl sulphate polyacrylamide gel electrophore sis and stained with Coomassie brilliant blue 1 protein marker 2 a mutant C 145 S has two expression products as an apparent molecular weight positive control 3 7 mutants of P1 Y P1 F P1 I P1 W P1 K 8 lysate of E coli cell containing plasmid pET 28a as the negative control Fig 2 Autocleavage assay of P1 position mutants Total cell lysates of isopropyl D thiogalactopyranoside induced E coli samples were separated by 12 sodium dodecyl sulphate polyacrylamide gel electrophoresis and stained with Coomassie brilliant b