【病毒外文文献】2008 Structures of Two Coronavirus Main Proteases_ Implications for Substrate Binding and Antiviral Drug Design

JOURNAL OF VIROLOGY Mar 2008 p 2515 2527 Vol 82 No 5 0022 538X 08 08 00H110010 doi 10 1128 JVI 02114 07 Copyright 2008 American Society for Microbiology All Rights Reserved Structures of Two Coronavirus Main Proteases Implications for Substrate Binding and Antiviral Drug Design H17188 Xiaoyu Xue 1 2 Hongwei Yu 3 Haitao Yang 1 2 Fei Xue 1 2 Zhixin Wu 3 Wei Shen 1 2 Jun Li 1 2 Zhe Zhou 1 Yi Ding 1 Qi Zhao 1 2 Xuejun C Zhang 2 Ming Liao 3 Mark Bartlam 1 2 4 and Zihe Rao 1 2 4 Tsinghua Nankai IBP Joint Research Group for Structural Biology Tsinghua University Beijing 100084 China 1 National Laboratory of Biomacromolecules Institute of Biophysics IBP Chinese Academy of Sciences Beijing 100101 China 2 Laboratory of Avian Medicine College of Veterinary Medicine South China Agricultural University Guangzhou 510642 China 3 and College of Life Sciences Nankai University Tianjin 300071 China 4 Received 25 September 2007 Accepted 12 December 2007 Coronaviruses CoVs can infect humans and multiple species of animals causing a wide spectrum of diseases The coronavirus main protease M pro which plays a pivotal role in viral gene expression and replication through the proteolytic processing of replicase polyproteins is an attractive target for anti CoV drug design In this study the crystal structures of infectious bronchitis virus IBV M pro and a severe acute respiratory syndrome CoV SARS CoV M pro mutant H41A in complex with an N terminal autocleavage substrate were individually determined to elucidate the structural flexibility and substrate binding of M pro A monomeric form of IBV M pro was identified for the first time in CoV M pro structures A comparison of these two structures to other available M pro structures provides new insights for the design of substrate based inhibitors targeting CoV M pro s Furthermore a Michael acceptor inhibitor named N3 was cocrystallized with IBV M pro and was found to demonstrate in vitro inactivation of IBV M pro and potent antiviral activity against IBV in chicken embryos This provides a feasible animal model for designing wide spectrum inhibitors against CoV associated diseases The structure based optimization of N3 has yielded two more efficacious lead com pounds N27 and H16 with potent inhibition against SARS CoV M pro Coronaviruses CoVs are highly prevalent and severe pathogens that cause a wide range of diseases in multiple species of animals including humans 16 25 30 36 In 2003 the etiological agent responsible for the global outbreak of a life threatening atypical pneumonia that caused approximately 800 deaths worldwide was identified as the severe acute respi ratory syndrome CoV SARS CoV 7 9 14 15 24 A pro totype of the Coronaviridae family is avian infectious bronchitis virus IBV 16 30 which belongs to the genetic group III of CoV 16 and causes considerable economic losses for the poultry industry worldwide 5 13 CoVs are enveloped positive stranded RNA viruses with the largest viral RNA genomes known to date ranging from 27 to 31 kb 16 The CoV replicase gene encodes two overlapping polyproteins termed pp1a and pp1ab which mediate viral replication and transcription 3 16 29 36 The maturation of CoVs involves a highly complex cascade of proteolytic process ing events on the polyproteins to control viral gene expression and replication Most maturation cleavage events within the precursor polyprotein are mediated by the CoV main protease CoV M pro also known as 3CL protease or 3CL pro a three domain domains I to III cysteine protease with a chymotryp sin like two domain fold at the N terminus 10 18 37 The structures of CoV M pro s revealed that two CoV M pro mole cules form an active homodimer 1 2 33 35 A Cys His catalytic dyad is located in a cleft between domains I and II 1 2 35 and the N terminal residues 1 to 7 or N finger of M pro are considered to play an important role in the proteolytic activity 1 2 33 35 The C terminal domain III is reported to be required for dimerization 28 Here we report the crystal structures of two CoV M pro s The first is the IBV M pro structure with a dimeric form and a unique monomeric form in one asymmetric unit The mono meric form has not been observed in any of the previously reported CoV M pro s its C terminus inserts into one of the active sites present in the dimer The second is the structure of an active site mutant H41A of SARS CoV M pro in complex with the N terminal 11 amino acid peptide as the substrate which provides insights into the substrate binding and speci ficity of the S1H11032 to S5H11032 sites in SARS CoV M pro in an unprec edented way As the CoV M pro is responsible for the maturation of itself and the subsequent maturation of the replicase polyproteins 37 it has become an attractive target for anti CoV drug design Here we also present the cocrystal structure of IBV M pro in complex with N3 a wide spectrum inhibitor that we designed previously to target CoV M pro s 34 We further demonstrate its rapid in vitro inactivation against the viral protease and potent antiviral activity toward IBV in chicken embryos This assay provides an easily accessible animal model for optimizing wide spectrum inhibitors against CoV associ ated diseases A comparison of the substrate binding sites of Corresponding author Mailing address Laboratory of Structural Biology Life Sciences Building Tsinghua University Beijing 100084 China Phone 86 10 62771493 Fax 86 10 62773145 E mail raozh These authors contributed equally H17188 Published ahead of print on 19 December 2007 2515 on April 25 2015 by EAST CAROLINA UNIV http jvi asm org Downloaded from IBV M pro and SARS CoV M pro provides further insights for the design of substrate based inhibitors targeting CoV M pro s Further modification of Michael acceptor inhibitors based on the new structural information provided here results in two improved inhibitors termed N27 and H16 with potent inhibi tion against SARS CoV M pro MATERIALS AND METHODS Protein purification and crystallization The protein expression purification and crystallization of native IBV M pro has been described previously 20 34 The crystal structure of IBV M pro could not be determined using conventional molecular replacement techniques Therefore a selenomethionyl SeMet de rivative of IBV M pro was prepared for crystallization and data collection The recombinant plasmid pGEX 4T 1 IBV M pro was used to transform the methio nine auxotrophic B834 DE3 Escherichia coli strain Novagen which was prop agated in minimal medium supplemented with 60 mg liter H110021 L SeMet The SeMet substituted IBV M pro was purified as described before and concentrated to 20 mg ml H110021 for crystallization The best crystals were obtained using streak seeding with 2 5 wt vol polyethylene glycol 4000 PEG4K 12 vol vol 2 propanol and 0 1 M sodium cacodylate pH 6 5 as the reservoir solution Crystals of IBV M pro complexed with inhibitor N3 were produced by cocrys tallization IBV M pro was incubated with an equal molar concentration of N3 for 24 h at 4 C This complex did not crystallize under conditions described above However single cubic crystals were obtained in 1 day by the hanging drop vapor diffusion method at 18 C using a reservoir solution containing 20 wt vol PEG10K and 0 1 M HEPES pH 7 5 without any seeds The coding sequence of SARS CoV M pro was cloned from the SARS CoV BJ01 strain and inserted into the BamHI and XhoI sites of pGEX 6p 1 plasmid DNA Amersham Biosciences The PCR based overlap extension method 12 was used to produce an active site knockout mutant of SARS CoV M pro with His 41 replaced by Ala H41A using pGEX 6p 1 SARS CoV M pro as a tem plate The primers were designed so that the ends of the two PCR products contained complementary sequences which allowed the two fragments to be spliced in a second PCR The four primers used for the single point mutation were the following 5H11032 CGGGATCCAGTGGTTTTAGG AAAATG 3H11032 forward A 5H11032 CCGCTCGAGTCATTGGAAGGTAACACCAGA 3H11032 reverse A 5H11032 A ATGACCGCTCTTGGACAGTATACTGT 3H11032 forward B and 5H11032 CCAAGAG CGGTCATTTGCACAGCAGAA 3H11032 reverse B Specifically in the first PCR two sets of primers forward A reverse B and forward B reverse A were used to generate the templates for the second PCR The two primers forward A reverse A were used in the second PCR and then the PCR products were inserted into the BamHI and XhoI sites of the pGEX 6p 1 plasmid The resulting plasmids containing the H41A mutation were verified by sequencing and then transformed into E coli BL21 DE3 cells The protein expression and purification of the SARS CoV M pro were described previously 35 The crystallization of SARS CoV M pro H41A was the same as that for the wild type protease 33 35 An 11 amino acid peptidyl substrate of the sequence TSAVLQSGFRK was dis solved at a 20 mM concentration in 7 5 wt vol PEG6K 6 vol vol dimeth ylsulfoxide DMSO and 0 1 M morpholineethanesulfonic acid Mes pH 6 0 A3 H9262l aliquot of this solution was added to the crystallization drop 3 H9262l and the crystals were soaked for 8 days before data collection Diffraction data collection A total of four data sets were collected Table 1 Data for the SeMet IBV M pro derivative were collected to a 2 8 resolution at the peak wavelength for the maximum fH11033 at 100 K using a Structural Biology Center 2 000 by 2 000 charge coupled display detector on beamline BL19 ID of the Advanced Photon Source Argonne National Laboratory The cryoprotectant solution contained 20 vol vol glycerol 2 wt vol PEG4K 9 6 vol vol 2 propanol and 0 08 M sodium cacodylate pH 6 5 Another data set for the native IBV M pro was collected to a 2 35 resolution at 100 K on beamline BL 5A at Photon Factory KEK Japan using an ADSC Q315 e coupled display detector Data for the IBV M pro N3 complex and SARS CoV M pro H41A mu tant peptidyl substrate complex were collected at 100 K in house with a Rigaku CuKH9251 rotating anode X ray generator MM007 at 40 kV and 20 mA 1 5418 and using a Rigaku R AXIS IVH11001H11001 image plate detector The IBV M pro complex crystal was used directly in data collection without a cryoprotectant The cryo protectant solution for the SARS CoV M pro mutant complex contained 30 PEG400 and 0 1 M Mes pH 6 0 All data integrations and scaling were per formed using HKL2000 23 The Matthews coefficient of the new IBV M pro crystal form suggested the existence of three protein molecules per asymmetric unit with an estimated solvent content of 54 Structure solution refinement and analysis The IBV M pro structure was solved by the single wavelength anomalous dispersion method 11 using the diffraction data set collected at the peak wavelength for selenium The analysis of the selenium positions performed with the program SHELXD 27 located all 12 expected selenium sites four in each protein molecule Phasing and density modifications subsequently were performed with SOLVE 32 and RESOLVE 31 The resulting electron density maps were of sufficient quality for chain tracing Molecular replacement performed with CNS 4 was employed for tracing the typical homodimer named molecules A and B into the electron density map using the crystal structure of human CoV 229E HCoV 229E M pro as a starting model Protein Data Bank code 1P9S The third M pro molecule named molecule C was clearly identified in the electron density map and its tracing was facilitated using the noncrystallographic symmetry of the selenium positions Cycles of manual adjustment to the model with Coot 8 and subse quent refinement using REFMAC 21 led to a final model with a crystallo graphic R factor R cryst of 22 7 and a free R factor R free of 25 9 at 2 35 resolution The IBV M pro N3 complex structure was determined by the molecular re placement method implemented in CNS using the homodimer molecules A and B from the above described native IBV M pro structure as the search model Manual adjustments to the model were made with the program O 22 and subsequent refinement was performed in CNS Data quality and refinement statistics are summarized in Table 1 The structure of the mutant protein SARS CoV M pro H41A in complex with its N terminal peptide substrate was determined by the molecular replacement method using a SARS CoV M pro monomer Protein Data Bank code 1UK2 35 as a search model In the complex structure there are two M pro molecules named A and B per asymmetric unit and it forms a symmetrical homodimer An 11 mer peptide was identified in molecule A and an 8 mer peptide in mol ecule B from the initial difference electron density maps The validation of all final models was carried out with PROCHECK 17 In vitro inhibition assays Proteolytic activity assays of IBV M pro have been described previously 33 34 The fluorogenic substrate of SARS CoV M pro MCA AVLQSGFR Lys Dnp Lys NH2 H1102295 purity GL Biochem Shanghai Ltd Shanghai China was used to assess the activity of IBV M pro The excita tion and emission wavelengths of the fluorogenic substrate were 320 and 405 nm respectively The assay was performed in a buffer of 50 mM Tris HCl pH 7 3 and 1 mM EDTA at 30 C and kinetic parameters were determined by following our previous work 34 In ovo inhibition Titers of the IBV M41 viruses were established as follows The virus was serially 10 fold diluted in phosphate buffered saline PBS and then inoculated into the allantoic cavity of 10 day old specific pathogen free SPF chicken embryos six embryos per dilution and 0 1 ml virus dilution per embryo The embryos were incubated at 37 C and were inspected daily Eight days after inoculation the eggs were opened and examined to check for typical lesions including crispature and dwarfism in embryos yolk sac shrinking an increase in allantoic fluid and lithate deposits on the midkidney of embryos that might signify IBV infection Six embryos inoculated with PBS were used as negative controls and another six uninoculated embryos were used as blank controls The dilution that could cause 50 of embryos to be infected by IBV was calculated using the method described by Reed and Muench 26 and determined as the virus titer 50 egg infectious dose EID 50 To assess whether N3 could be used as an anti IBV preventive agent or a curative agent two groups of in ovo inhibition experiments were performed For the curative group a series of doses of N3 0 02 to 0 64 H9262mol was injected into the allantoic cavity of 10 day old SPF chicken embryos 3 h for eight embryos repeated per dose of N3 or 6 h for six embryos repeated per dose of N3 after inoculation by a 100 EID 50 titer of IBV M41 virus For the preventive group N3 was preinjected into the embryos 3 h for eight embryos repeated per dose of N3 or 6 h six embryos repeated per dose of N3 prior to the inoculation by a 100 EID 50 titer of virus Eight days after inoculation the eggs were opened to check if the embryos were infected by IBV The inhibitor dose that could protect 50 of embryos from IBV infection was calculated using the method described by Reed and Muench 26 and expressed as the 50 protective dose PD 50 Meanwhile a preliminary toxicity assay was performed to assess any potential adverse effects of N3 on the development of chicken embryos The highest dose of N3 0 64 H9262mol dissolved in DMSO was injected into 16 embryos Sixteen embryos inoculated with DMSO were used as negative controls while another 16 uninoculated embryos were used as blank controls Eight days after inoculation half of the eggs were opened and examined for pathological changes to the organs of the embryos The remainder of the eggs were continuously incubated at 37 C until the chickens were hatched All in ovo experiments were performed in a biosafety level 2 bioprotective laboratory 2516 XUE ET AL J VIROL on April 25 2015 by EAST CAROLINA UNIV http jvi asm org Downloaded from Accession codes Coordinates and structure factors for IBV M pro IBV M pro in complex with inhibitor N3 and the SARS CoV M pro H41A mutant in complex with an N terminal substrate have been deposited in the Protein Data Bank under accession numbers 2Q6D 2Q6F and 2Q6G respectively RESULTS Overall structure of native IBV M pro The IBV M pro crystal structure at a 2 35 resolution shows three M pro molecules named A B and C per asymmetric unit Fig 1A which is unique among all CoV M pro structures reported to date While molecules A and B form a typical catalytically active and sym metrical homodimer molecule C is not involved in such a dimer Instead its C terminus inserts into the substrate binding site of molecule A Fig 1A Molecules A and B are quite similar with an RMSD root mean square deviation of 1 1 for all equivalent CH9251 atoms while molecule C is less similar to either A or B having a mean RMSD of 2 5 for the CH9251 atoms of residues 6 to 183 Each IBV M pro molecule is comprised of three domains I to III Fig 1B Domains I and II i e residues 3 to 99 and 100 to 182 respectively have a chymotrypsin like two H9252 barrel fold in common with the M pro structures of transmissible gas troenteritis CoV TGEV HCoV 229E and SARS CoV 1 2 35 Domain III residues 199 to 307 of IBV M pro consists of five H9251 helices that adopt a globular structure apparently unique to CoV M pro Domains II and III are connected by a loop of residues 183 to 198 which exhibits two distinct conformations in the three M pro molecules In molecules A and B it assumes a fairly extended conformation in molecule C however resi dues 186 to 190 form a short helix Fig 1E The substrate binding sites are located in the deep cleft between domains I and II with the catalytic dyad formed by His 41 and Cys 143 at the center of this cleft Each subunit contains one substrate binding site contributed mainly from itself Nevertheless the two monomers swap their N termini to stabilize the S1 pocket in the IBV M pro dimer similar swapping was also observed in the M pro structures of TGEV HCoV 229E and SARS CoV 1 2 19 33 35 This arrangement may explain the require TABLE 1 Data collection and refinement statistics Parameter Data set for Se Met IBV M pro Native IBV M pro IBV M pro N3 SARS CoV M pro H41A substrate Data collection statistics Wavelength A 0 9795 1 0000 1 5418 1 5418 Resolution A 50 2 8 2 91 2 80 b 50 2 35 2 43 2 35 50 2 00 2 07 2 00 50 2 40 2 49 2 40 Space group P6 1 22 P6 1 22 P1 P2 1 Cell parameters a A 118 2 118 9 53 2 52 0 b A 118 2 118 9 54 5 95 8 c A 267 7 270 9 66 7 67 7 H9251 90 0 90 0 111 1 90 0 H9252 90 0 90 0 104 3 102 9 H9253 120 0 120 0 91 3 90 0 Total reflection 713 639 339 766 165 955 82 777 Unique reflection 56 512 47 480 42 883 25 190 Completeness 100 0 100 0 98 9 99 8 94 2 82 6 99 8 99 9 Redundancy 12 6 8 6 7 2 7 3 3 9 3 3 3 3 3 3 R merge a 0 170 0 715 0 054 0 358 0 041 0 225 0 106 0 474 Sigma cutoff 0 0 0 0 I H9268 I 16 6 2 5 39 8 5 3 30 4 5 1 11 8 2 5 Refinement statistics Resolution range A 50 2 35 50 2 00 30 2 50 R work c 22 7 21 6 19 9 R free 25 9 24 2 26 7 RMSD from ideal geometry Bonds A 0 009 0 011 0 007 Angles 1 62 1 75 1 39 Average B factor A 2 Main chain 50 3 40 4 29 7 Solvent 56 4 49 9 42 1 Ramachandran plot d Favored 85 7 91 6 84 3 Allowed 14 0 8 4 14 4 Generously allowed 0 3 0 0 0 7 Disallowed 0 0 0 0 0 6 a R merge H11005H9018I i H11002H11021IH11022 H9018I where I i is the intensity of an individual reflection and H11021IH11022 is the average intensity of that reflection b R work H11005H9018F p H11002F c H9018F p where F c is the calculated and F p is the observed structure factor amplitude c Ramachandran plots were generated by using the program PROCHECK d Numbers in parentheses correspond to the highest resolution shell VOL 82 2008 STRUCTURES OF SUBSTRATE BOUND IBV AND SARS CoV M pro 2517 on April 25 2015 by EAST CAROLINA UNIV http jvi asm org Downloaded from 2518 XUE ET AL J VIROL on April 25 2015 by EAST CAROLINA UNIV http jvi asm org Downloaded from ment of dimerization for the full activity of the M pro proteins 1 2 19 33 35 According to a structure based sequence alignment Fig 2 there is one deletion and two insertions in IBV