【病毒外文文献】2009 Severe Acute Respiratory Syndrome Coronavirus nsp9 Dimerization Is Essential for Efficient Viral Growth

《【病毒外文文献】2009 Severe Acute Respiratory Syndrome Coronavirus nsp9 Dimerization Is Essential for Efficient Viral Growth》由会员分享，可在线阅读，更多相关《【病毒外文文献】2009 Severe Acute Respiratory Syndrome Coronavirus nsp9 Dimerization Is Essential for Efficient Viral Growth（12页珍藏版）》请在金锄头文库上搜索。

1、JOURNAL OFVIROLOGY, Apr. 2009, p. 30073018Vol. 83, No. 70022-538X/09/$08.00?0doi:10.1128/JVI.01505-08Copyright 2009, American Society for Microbiology. All Rights Reserved.Severe Acute Respiratory Syndrome Coronavirus nsp9 Dimerization IsEssential for Efficient Viral Growth?Zachary J. Miknis,1Eric F

2、. Donaldson,2Timothy C. Umland,1,3Ryan A. Rimmer,3Ralph S. Baric,2and L. Wayne Schultz1,3*Department of Structural Biology, State University of New York at Buffalo, Buffalo, New York 142031; Department ofMicrobiology and Immunology, University of North Carolina at Chapel-Hill, Chapel-Hill, North Car

3、olina 275992; andHauptman-Woodward Medical Research Institute, Buffalo, New York 142033Received 17 July 2008/Accepted 9 January 2009The severe acute respiratory syndrome coronavirus (SARS-CoV) devotes a significant portion of its genometo producing nonstructural proteins required for viral replicati

4、on. SARS-CoV nonstructural protein 9 (nsp9)was identified as an essential protein with RNA/DNA-binding activity, and yet its biological function within thereplication complex remains unknown. Nsp9 forms a dimer through the interaction of parallel ?-helicescontaining the protein-protein interaction m

5、otif GXXXG. In order to study the role of the nsp9 dimer in viralreproduction, residues G100 and G104 at the helix interface were targeted for mutation. Multi-angle lightscattering measurements indicated that G100E, G104E, and G104V mutants are monomeric in solution,thereby disrupting the dimer. How

6、ever, electrophoretic mobility assays revealed that the mutants bound RNAwith similar affinity. Further experiments using fluorescence anisotropy showed a 10-fold reduction in RNAbinding in the G100E and G104E mutants, whereas the G104V mutant had only a 4-fold reduction. Thestructure of G104E nsp9

7、was determined to 2.6- resolution, revealing significant changes at the dimerinterface. The nsp9 mutations were introduced into SARS-CoV using a reverse genetics approach, and theG100E and G104E mutations were found to be lethal to the virus. The G104V mutant produced highlydebilitated virus and eve

8、ntually reverted back to the wild-type protein sequence through a codon transversion.Together, these data indicate that dimerization of SARS-CoV nsp9 at the GXXXG motif is not critical for RNAbinding but is necessary for viral replication.The discovery of a novel coronavirus (CoV) as the causativeag

9、ent of severe acute respiratory syndrome (SARS; SARS-CoV) has highlighted the need for a better understanding ofCoV replication (19). After emerging in late 2002 from theGuangdong Province in China, SARS-CoV was rapidly iso-lated, and its genome sequenced to reveal a new CoV that wasphylogenetically

10、 distinct, suggesting a new classification as atype IIb CoV (15, 32, 43, 50). Genomic comparison to theclosely related murine hepatitis virus (MHV), human CoVOC43 (HCoV-OC43), and bovine CoV revealed a highly con-served genomic structure with many regions nearly identical(52). After strict quarantin

11、e controls were initiated, SARS-CoV was contained, with approximately 8,000 individuals beingclinically infected resulting in close to 800 deaths (www.who.int/csr/sars/en). Recently, the natural reservoir for SARS-CoVwas reported as the Chinese horseshoe bat, indicating that (i)the disease is still

12、circulating in animals and (ii) a future re-emergence from a zoonotic source is possible, underscoringthe importance of continued study of this virus (36, 38).CoVs devote a significant portion of their positive-sensesingle-stranded RNA (ssRNA) genome to proteins related toviral RNA replication. SARS

13、-CoV has a genome of ?29,700nucleotides, of which more than 21,000 code for the 16 non-structural proteins (nsps) (43, 50). The replicase genes openreading frame 1a (ORF1a) and ORF1b are translated intolarge polyproteins pp1a (4,300 amino acids aa), and througha ?1 ribosomal frameshift mechanism, a

14、fusion protein knownas pp1ab (7,000 aa) (58). Posttranslational processing of thepolyproteins by two distinct viral proteinases, the papain-likeproteinase and a 3C-like proteinase (3CLpro; also known asMpro), yields 16 mature nsps, many of which interact to formthe replication complex responsible fo

15、r the synthesis of nega-tive-strand template, positive-strand genomic, and all sub-genomic RNAs (sgRNA) with additional roles related tocellular processes (23, 30, 37, 52, 54, 68). The precise stoi-chiometry of the replicase complex is unknown, but yeast twohybrid screens, glutathione S-transferase

16、pull-down assays andX-ray crystallography, have revealed a number of protein-pro-tein interactions between the various nsps (27, 60, 67).In general, proteins form multimers for a variety of reasonsincluding stability, allostery, and to ensure accurate translationof genetic information (22). Translat

17、ion of large polypeptidescan result in errors in protein sequence and therefore largecomplexes are usually assembled from multiple small proteins.Viruses both violate and adhere to this rule by first producinglarge polyproteins and then processing them into individualproteins that assemble to form a

18、ctive replication complexes.For the CoVs, the protein components and stoichiometry ofthe positive- and negative-strand replication complexes remainunknown. Crystal structures of individual SARS-CoV nspshave revealed several different multimeric states for the sameprotein. For example, the structure

19、of nsp10 is reported as adimer and as a dodecamer by separate groups (29, 55), al-though genetic data suggest that the dodecamer structure is* Corresponding author. Mailing address: Department of StructuralBiology, SUNY at Buffalo, Hauptman-Woodward Institute, 700 Elli-cott St., Buffalo, NY 14203. P

20、hone: (716) 898-8640. Fax: (716) 898-8660. E-mail: schultzhwi.buffalo.edu.?Published ahead of print on 19 January 2009.3007 on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from not essential for in vitro replication (14). Biochemical studiesof nsp7 and nsp8 indicate that, in

21、dependently, they are dimericin solution but the crystal structure of the nsp7 and nsp8complex is a hexadecamer (67). For nsp9, there are four crys-tallographic structures which report a variety of dimeric inter-faces (16, 49, 57). In one case, two interfaces are present in asingle crystal form, sug

22、gesting that a tetrameric complex thatincorporates both interfaces may be possible. Although theinformation provided by these structures is significant, the bi-ological relevance to the replication complex has not beenestablished.SARS-CoV nsp9 has been shown to have RNA and DNA-binding ability throu

23、gh a variety of methods (16, 49, 57). nsp9from MHV-A59 has been shown through immunofluorescencestudies to localize with, among others, the helicase (nsp13),nucleocapsid (N protein) and 3CLpro(nsp5) (4, 5). nsp9 alsolocalizes to late endosomes at sites of replication with nsp7,nsp8, and nsp10 and is

24、 likely a member of the replicationcomplex (4). An nsp9 knockout in MHV-A59 is not viable,while fusion of a nsp9-10 polyprotein is viable but attenuatedin growth, suggesting that the mature form of nsp9 plays acritical role in viral replication (13).Several crystallographic structures of nsp9 have s

25、hown that itis composed of seven beta strands and a single alpha helix (16,49, 57). The fold of nsp9 is reported to be similar in fold todomains I and II of the 3CLproencoded within the SARS-CoVgenome (16); however, no significant sequence similarity existsbetween the two. The presumed biological di

26、mer utilizes theinteraction between the lone helices of each monomer to formthe parallel helix-helix dimer (Fig. 1A). A second dimeric formhas a beta-sheet interface stabilized by main chain atom inter-actions within the sheet regions of each monomer (Fig. 1B). Arecent crystal structure of HCoV 229E

27、 nsp9 reveals an antipa-rallel helix-helix dimer formed by a disulfide bond at Cys69(Fig. 1C). Upon generating a multiple sequence alignment(Fig. 1D) of various CoV nsp9 proteins, it was found that whilethe helix-helix interface found in Fig. 1A contained multipleconserved residues along the dimer i

28、nterface and buried?1,000 2of surface-exposed area, the sheet-sheet dimer inFig. 1B contained no conserved residues in the dimer regionand buried only ?500 2. A survey of dimer interfaces foundin protein structures suggests for a protein of ?15 kDa aburied surface area of ?1,000 2would be expected (

29、2, 28).The common protein-protein interaction motif GXXXG(20, 31) is conserved at the dimer interface, allowing the heli-ces to closely pack at the positions of G100 and G104 (Fig.1A). Therefore, we designed mutagenesis experiments to in-vestigate the stability and function of that dimer. MutantsG10

30、0E, G104E, and G104V were created to disrupt the di-meric interface and were characterized by size exclusion chro-matography (SEC), multi-angle light scattering (MALS), andcircular dichroism (CD) spectroscopy. Effects on ssRNA bind-ing were assessed by RNA electrophoretic mobility shift assay(EMSA)

31、and fluorescence anisotropy (FA). The crystal struc-ture of G104E was solved and refined to 2.6- resolution, andchanges in the helix-helix interface were observed.The development of reverse genetics approaches to studyingCoVs has also allowed us to reintroduce the G100E, G104E,and G104V mutations in

32、to SARS-CoV (Urbani strain) tostudy their effects on the virus in vivo (65). The G100E andG104E mutations were lethal to the virus, while the G104Vmutation produced a highly debilitated growth phenotype witheventual transversion of the codon at position 104 from GTG(Val) to GGG (Gly), indicating tha

33、t dimerization of nsp9 atthe GXXXG interface is required for efficient viral growth.MATERIALS AND METHODSMutant generation. SARS-CoV nsp9-pET23d(?) plasmid containing thecoding sequence of nsp9 in addition to a C-terminal His6tag was a gift fromMark Denison (Vanderbilt University). Nsp9 G100E, G104E

34、, and G104VFIG. 1. Dimer arrangements in nsp9. (A) nsp9 crystal structure(1QZ8) showing presumed biological dimer and helix-helix interface.The individual monomers are colored in blue/green and red/yellow,respectively. The positions of G100 and G104 are depicted as space-filling models for both mono

35、mers. G100E and G104E are labeled inone monomer. (B) Alternate nsp9 dimer (1UW7) stabilized throughsheet regions. Each monomer is colored from the N terminus (blue) tothe C terminus (red). (C) Antiparallel helix-helix dimer of HCoV-229E nsp9 stabilized by a disulfide linkage at C69. Each monomer isc

36、olored from the N terminus (blue) to the C terminus (red). (D) Mul-tiple sequence alignment of CoV nsp9 homologs showing absoluteconservation of glycines equivalent to G100 (green) and G104 (cyan)in SARS-CoV. Images were prepared by using PyMol.3008MIKNIS ET AL.J. VIROL. on March 7, 2015 by DAHLGREN

37、 MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from mutants were generated by using a QuikChange site-directed mutagenesis kit(Qiagen) according to the manufacturers protocol using the following prim-ers: 9G100EF, 5?-AACAACCTAAATAGAGAAATGGTGCTGGGCAGTTTAGC-3?; 9G100ER, 5?-GCTAAACTGCCCAGCACCATTTCTCTATTT

38、AGGTTG-3?; 9G104EF, 5?-GAGGTATGGTGCTGGAAAGTTTAGCTGCTAC-3?; 9G104ER, 5?-GTAGCAGCTAAACTTTCCAGCACCATACCTC-3?;9G104VF,5?-GAGGTATGGTGCTGGTGAGTTTAGCTGCTAC-3?;and9G104VR, 5?-GTAGCAGCTAAACTCACCAGCACCATACCTC-3? (boldfacing indicates sites of mutation). Mutagenesis products were transformedinto DH5? cells (In

39、vitrogen, Carlsbad, CA) and plated on LB agar plates(supplemented with 100 ?g of ampicillin/ml) from which colonies were se-lected and grown in 5-ml cultures of LB supplemented with 100 ?g ofampicillin/ml. Plasmid was purified by using a Qiagen spin miniprep kitaccording to the manufacturers protoco

40、l. The presence of the mutationswithin the plasmids was confirmed through DNA sequencing on an ABI Prism3130XL genetic analyzer (Roswell Park Cancer Institute Biopolymer Facility).Expressionandpurification.PlasmidswereeachtransformedintoBL21(DE3) (Novagen) cells and screened for expression condition

41、s. Solubleprotein was obtained through the addition of IPTG (isopropyl-?-D-thiogalacto-pyranoside) to 1 mM in 1 liter of culture grown in LB (supplemented with 100?g of ampicillin/ml and grown to an optical density at 600 nm of ?0.6 at 37C),followed by shaking at 37C for 3 h. Cells were collected by

42、 centrifugation andstored at ?80C.All purification steps were carried out at 4C. Cell pellets from 1-liter cultureswere thawed, resuspended in 15 ml of buffer A (50 mM Tris pH 8.0, 300 mMNaCl, 10 mM imidazole) and 1 ml of protease inhibitor cocktail (Roche), andthen lysed by using a single pass thro

43、ugh a Microfluidizer (Microfluidics Co.) at?18,000 lb/in2. The resulting lysate was centrifuged at 35,000 rpm at 4C for 30min in a 45 Ti rotor (Beckman). Supernatant was filtered (0.22-?m pore size)and applied to a 5-ml HisTrap (GE Healthcare) column pre-equilibrated inbuffer A. Protein was washed w

44、ith buffer A followed by 12% buffer B (50 mMTris pH 8.0, 300 mM NaCl, 300 mM imidazole) and eluted from the column ina linear gradient from 12 to 100% buffer B. Fractions were analyzed by sodiumdodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and nsp9 con-taining fractions were pooled a

45、nd dialyzed against 1 liter of buffer C (10 mM TrispH 8, 150 mM NaCl, 1 mM EDTA, 5 mM dithiothreitol). Concentrated proteinwas subsequently applied to a Superdex 75 HL 16/60 column pre-equilibrated inbuffer C. Fractions were analyzed by SDS-PAGE with appropriate fractionspooled and dialyzed against

46、either 1 liter of buffer D (50 mM MES morpho-lineethanesulfonic acid; pH 5.6, 15 mM NaCl, 0.5 mM TCEP tri(2-carboxyethyl)phosphine HCl) or 1 liter of buffer C. Final samples were stored at 4C andassayed for protein concentration at 280 nm using an extinction coefficient of13,075 M?1cm?1calculated by

47、 ProtParam (63) using the coding sequence of themature nsp9 product with a C-terminal tag (LEHHHHHH).SEC and SEC-MALS. Aliquots of elution fractions from immobilized metalaffinity chromatography (IMAC) purification were applied to a Superdex 75 HR10/30 column (GE Healthcare) preequilibrated in buffe

48、r E (10 mM NaPO4pH8, 100 mM NaCl). The Superdex column was attached on an AKTA Purifiercoupled with a Wyatt Systems MiniDAWN (three detectors) and Optilab DSP(Wyatt Systems) immediately downstream of the column. Protein samples wereapplied and eluted from the column at 0.5 ml/min with simultaneous c

49、ollectionof light scattering and refractive index data, using a laser excitation wavelengthof 690 nm and light scattering detectors at 45, 90, and 135. Peak analysis wasperformed by using the ASTRA software provided with the system.CD spectroscopy. Protein samples were passed over a Superdex 75 HL 1

50、6/60column in buffer F (25 mM NaPO4pH 8, 150 mM NaCl) to remove anynonspecific aggregates. Appropriate fractions were pooled and dialyzed against1 liter of buffer G (10 mM NaPO4pH 8.0). CD spectra of nsp9 samples weregathered from 200 to 255 nm in a 1-mm-pathlength quartz cuvette at 20C (1-nmsteps,

51、50-nm/s scan speed, 4-s response time, three scans) using a Jasco J-715Spectrapolarimeter (Department of Pharmaceutical Sciences, University at Buf-falo). Secondary structure assignment was achieved through use of the K2d webserver (www.embl-heidelberg.de/?andrade/k2d) and the DICHROWEB server(www.c

52、ryst.bbk.ac.uk/cdweb/html/home.html) (1, 40, 62).For thermal denaturation experiments using CD spectroscopy, a similar pro-tocol was followed. Samples were monitored while heating from 20 to 95C usinga Peltier device. Spectra were gathered at 5 intervals, with peak monitoring at205 nm. The data were

53、 analyzed using Origin 7.0.RNA EMSA. A labeled ssRNA oligonucleotide (Integrated DNA Technolo-gies) was used for gel shift assays (Biotin-5?-CGACUCAUGGACCUUGGCAG-3?). Oligonucleotide was resuspended in RNase-free water at 1 ?M and storedat ?20C. 5 ?g of nsp9 samples (in buffer D) were mixed with 1 ?

54、M RNA, 1 ?MssDNA oligonucleotide (3CL NdeI Forward2 5?-GGTGGTCATATGAGTGGTTTTAGGAAAATGGCATT-3?), and RNase inhibitor (SUPERaseIN; Am-bion), followed by incubation for 60 min at room temperature. After incubation,the samples were cross-linked at 254 nm for 15 min. High-density Tris-borate-EDTA (TBE) l

55、oading buffer (Invitrogen) was added to the nucleic acid-proteinmixtures, which were loaded onto Novex 4 to 20% TBE polyacrylamide gels(Invitrogen) and resolved at 80 V for 2 h in 0.5? TBE buffer. The gels were thentransferred to positively charged nylon membrane (Hybond) in 0.5? TBE at 25V for 30?

56、in a semidry transfer apparatus (Bio-Rad). Membranes were cross-linked for 5 min in 2? SSC (1? SSC is 0.15 M NaCl plus 0.015 M sodium citrate)at 254 nm. The membranes were then developed using a BrightStar biodetectionkit (Ambion) according to the manufacturers protocol. Extended wash stepswere inco

57、rporated to help further decrease the nonspecific background of themembrane.Attempts were also made to compete the RNA probe off of nsp9 using anssDNA template which was added at equivalent and 10-fold-higher levels com-pared to the RNA probe. Developed membrane was exposed to Hyperfilm ECL(GE Healt

58、hcare) and analyzed for electrophoretic mobility.FA. FA measurements were performed on a Fluoromax-4 spectrophotometer(Jobin Yvon Horiba) equipped with a temperature-controlled cell and polarizingfilters. All experiments were carried out at 22C. 5?-Fluorescein-labeled RNA (5?FAM-CGACUCAUGGACCUUGGCAG

59、-3?; IDT) was used in all experiments.FAM-RNA was dissolved in buffer (20 mM Tris pH 7.2) to a final concentrationof 58 nM in a 1-ml quartz cuvette. Small aliquots (1 to 2 ?l) of nsp9 or nsp9mutants (in 10 mM MES pH 5.6) were added to the cuvette covering a proteinconcentration range of 1 to 2,000 n

60、M, followed by incubation with stirring for 5min between measurements. FA was measured by exciting at 490 nm and mon-itoring the emission at 515 nm. Slits for the excitation and emission were set to2.5 nm. The integration time was 2.0 s for the anisotropy measurements. FA datacollection was controll

61、ed by using the FA module of the Fluoressence software(version 2.1.5.0). Each anisotropy value is the average of 10 individual anisotropymeasurements. The relative standard deviation was ?2% for all measurements.The data were analyzed using nonlinear regression fitting of the data to thefollowing eq

62、uation (42) for a single site binding model using Prism software:A ? Af? ?Ab? Af?1 ? Ka?LT? ? Ka?RT? ?1 ? Ka?LT? ? Ka?RT?2? 4?LT?RT?Ka2?2Ka?RT?ITC. Dilution isothermal titration calorimetry (ITC) is a technique amenableto determining protein dimerization constants (7, 9). Purified protein sampleswer

63、e subjected to buffer exchange five times in an Amicon centrifugal concen-trator to exchange protein in appropriate buffer (20 mM MES pH 5.6, 15 mMNaCl, and 0.1 mM TCEP for wild type, G100E, and G104E; 20 mM Na TAPSN-Tris(hydroxymethyl)methyl-3-aminopropanesulfonic acid pH 8.5, 100mM MgCl2, and 0.1

64、mM TCEP for G104V). Protein samples and correspondingbuffer were filtered to 0.2 ?m (Acrodisc syringe filter, low protein binding HTTuffryn membrane; PALL Life Sciences) and thoroughly degassed.Buffer corresponding to the sample to be studied was loaded into the referenceand sample cells of the VP-I

65、TC (Microcal). Typically, 28 injections of 10 ?l ofprotein were made into the sample cell filled with buffer. Concentrations of nsp9in the injection syringe ranged from 189 ?M to 6.3 mM, and concentrationranges in the sample cell were monitored from ?0.001 mM to ?0.3 mM (nearthe published dimerizati

66、on Kd). Additional measurements on samples were madeup to a final concentration of ?1 mM within the sample cell. Samples whichshowed a saturation curve were fit to a simple dissociation model using theincluded analysis software (Origin 7) specifically designed for dilution ITC dataanalysis.Mutant de

67、sign for reverse genetics. To determine the effects of mutationsG100E, G104E, and G104V on SARS-CoV replication, the three mutations wereindividually engineered into the molecular clone of SARS-CoV, replacing thewild-type codons as follows: G100E, GGT to GAG; G104V, GGC to GTG; andG104E, GGC to GAA.

68、 This was accomplished by using the “No Seem” (66)approach whereby primers were designed utilizing type IIS restriction endonu-clease sites which allowed sticky ends to be generated with non-native nucleo-tides present. This was used to introduce the mutations into the codons describedabove, using t

69、he wild-type SARS-D fragment from the infectious clone of SARS-CoV strain Urbani (icSARS) as the backbone. The primers used were as follows:G100Emutant,G100EF(5?-AGAGAGATGGTGCTGGGCAGT-3?)andG100ER (5?-ACTGCCCAGCACCATCTCTCT-3?); G104E mutant, G104EF(5?-AGAGGTATGGTGCTGGAAAGT-3?) and G104ER (5?-ACTTTCC

70、AGCACCATACCTCT-3?); and mutant G104V, G104VF (5?-AGAGGTATGGTGCTGGTGAGT-3?) and G104VR (5?-ACTCACCAGCACCATACCTCT-3?). TheVOL. 83, 2009SARS-CoV nsp9 DIMERIZATION3009 on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from primers used to amplify the nsp9 cDNA were M13R (5?-AGGAAA

71、CAGCTATGAC-3?) and nsp9R (5?-AAGTTCCTTGAAACTGAGACG-3?). Each mutationwas engineered in this way into the SARS-D fragment and the full-length mu-tated D fragments were digested, purified, and ligated back into the originalvector. Transformation was conducted by heat shock at 42C for 2 min, andtransfo

72、rmed E. coli was plated on LB plates with appropriate selection. Plateswere grown at room temperature for 48 h, after which five colonies were pickedand amplified in 5 ml of LB broth incubated at room temperature, underappropriate selection, and with agitation. After 24 h, the plasmids were purified

73、and digested with restriction endonuclease BglI to determine whether the DNAresembled wild type by restriction screening. Plasmids that digested correctlywere amplified in 20 ml of LB broth with the appropriate antibiotic and grownfor 24 h at room temperature with agitation. The plasmids were again

74、purified,and three of each mutant were sequenced to verify that the mutation wasincorporated into the wild-type SARS-D fragment.Assembling the mutant viruses. The infectious clone of the Urbani strain ofSARS-CoV was used as the backbone for this project, and mutated SARS-Dfragments containing each o

75、f the mutations of interest were assembled into thefull-length infectious cDNA as described previously (64, 66). Briefly, icSARSfragments A through F were amplified in E. coli (TOP-10; Invitrogen), purified,and screened by restriction digestion. Large stocks of each fragment thatscreened correctly w

76、ere established and digested with the appropriate enzyme asfollows. icSARS fragments B, C, D and E were digested with BglI. icSARSfragments A and F were digested with EcoRI and NotI, respectively. icSARSfragments A and F were then dephosphorylated and digested with BglI.The digested bands were subje

77、cted to electrophoresis, excised, and gel purified(Qiagen), and ligation reactions were set up using equivalent molar ratios of eachfragment and ligase (Roche). Wild-type fragments were ligated to generate afull-length cDNA of wild-type SARS-CoV, and the SARS-D fragment bearingeach mutation was used

78、 to generate full-length cDNAs for each mutant. Theligation reactions were purified by chloroform extraction and isopropanol pre-cipitation, and nucleocapsid cDNA and full-length viral genomic cDNA wereused as templates for in vitro transcription reactions (Ambion). Nucleocapsidand full-length viral

79、 genomic transcripts were then electroporated into Verocells. Transfections were monitored for 3 days and passaged if necessary. Allrecombinant icSARS strains were propagated on Vero E6 cells in Eagle minimalessential medium (Invitrogen) supplemented with 10% fetal calf serum (Hy-Clone, Logan, UT),

80、kanamycin (0.25 ?g/ml), and gentamicin (0.05 ?g/ml) at37C in a humidified CO2incubator. All work was performed in a biologicalsafety cabinet in a biosafety level 3 laboratory containing redundant exhaust fans.Personnel were equipped with powered air-purifying respirators with high-effi-ciency partic

81、ulate air and organic vapor filters (3M, St. Paul, MN), wore Tyveksuits (DuPont, Research Triangle Park, NC), and were double gloved.Verification of viral replication. The cells were analyzed daily for cytopathiceffect (CPE), and viral subgenomic transcription was verified by reverse tran-scription-

82、PCR (RT-PCR) of leader containing transcripts of the S gene. Briefly,primers pairs were designed to amplify the leader RNA sequence and ?300-nucleotide domain into the S gene. Total RNA was harvested from transfectedcells by using TriZol reagent, and purified RNA was reverse transcribed usingrandom

83、hexamer primers and Superscript III reverse transcriptase, according tothe manufacturers protocol. S gene specific leader containing cDNA was thenamplified by PCR.Analysis of mutant G104V RNA replication. The SARS-CoV infectious clonebearing the G104V mutation was assembled in triplicate as describe

84、d above andtransfected via three independent cuvettes containing 8 ? 106Vero cells. Thetransfected cells from each cuvette were then divided into three T25 flasks, andmedium was added; therefore, three independent transfections were each di-vided into three flasks for a total of nine flasks. One fla

85、sk from each transfectionwas harvested at 12, 24, and 36 h posttransfection. As a control, icSARS was alsotransfected into cells and harvested in the same way. In each case, transfectedcells were isolated in TRIzol reagent (Invitrogen) using the manufacturersprotocol for isolation of total RNA by is

86、opropanol precipitation and eluted to 50?l in nuclease-free water. All samples were then incubated for 1 h at 37C withDNase (Applied Biosystems, Foster City, CA) according to manufacturers pro-tocol to eliminate any residual input cDNA.Total RNA from each time point and transfection was reverse tran

87、scribed tocDNA using SuperScript III (Invitrogen) with modifications to the protocol asfollows. Random hexamers (300 ng) and total RNA (5 ?l) were incubated for 10min at 70C. The remaining reagents were then added according to the manu-facturers recommendations, and the reaction was incubated at 55C

88、 for 1 h,followed by 20 min at 70C to deactivate the reverse transcriptase. Quantitativereal-time PCR was conducted using SmartCycler II (Cepheid, Sunnyvale, CA)with SYBR green (diluted to 0.25?; Cepheid) to detect subgenomic cDNA withprimers (7.5 pmol) optimized to detect from the leader sequence t

89、o the 5? endof the N gene (forward N1S, AAAGCCAACCAACCTCGATC; reverse N1A,GCGTCCTCCATTCTGGTTAT) or genomic cDNA with primers (7.5 pmol)designed to detect genomic ORF1a cDNA (forward G104Vf, AGAGGTATGGTGCTGGTGAGT; reverse nsp9r, AAGTTCCTTGAAACTGAGACG). ThecDNA from the RT reaction mixtures of each vi

90、rus was used, at a volume of 5 ?lfor each reaction, with a total reaction mixture volume of 25 ?l. Omnimix beads(Cepheid) containing all reagents except SYBR green, primer, and templatewere used to standardize the reaction conditions, and template concentrationswere normalized by concentrations of t

91、he housekeeping gene GAPDH (glycer-aldehyde-3-phosphate dehydrogenase; forward GAPDHF, CATGGGGAAGGTGAAGGTCG; reverse GAPDHR, TTGATGGTACATGACAAGGTGC). Inaddition, all products were verified by melting curve analysis.Crystallization, cryopreservation, and diffraction. Purified nsp9 G104E wascrystalliz

92、ed by the hanging-drop vapor diffusion method utilizing a Hangmansystem (41). Protein was concentrated to 5.8 mg/ml and mixed in a 1:1 ratio withcrystallization cocktail (1.7 to 1.8 M ammonium sulfate, 0.1 M citrate phosphatebuffer pH 4 to 4.3). Hexagonal rods generally appeared within 1 week and gr

93、ewto a maximum of 0.3 by 0.1 by 0.1 mm.Crystals were transferred for 30 min to 1 h into cryobuffer (2 M ammoniumsulfate, 0.1 M citrate phosphate buffer pH 4.0 to 4.2, 25% vol/vol glycerol).Cryoprotected crystals were mounted onto CryoLoops (Hampton Research),quickly plunged into liquid nitrogen, and

94、 stored in a cryodewar.Diffraction experiments were remotely conducted on beamline 11-1 of theStanford Synchrotron Radiation Laboratory (SSRL) using the Blue-Ice interfaceand Web-Ice analysis software (21, 44). The data were collected at a wavelengthof 0.979 using a 20-s exposure and 1 oscillation p

95、er frame on a MAR-325charge-coupled device detector. The data were indexed, integrated, merged, andscaled to 2.6- resolution, as implemented within the HKL2000 package (47).Phasing and refinement. Processed data was input in the CaspR (10) server formolecular replacement trials using a variety of se

96、arch models, including thepreviously solved nsp9 structures (Protein Data Bank codes 1QZ8 and 1UW7)(16, 57). The best solution was found using monomeric 1QZ8 (16) as a searchmodel and resulted in a model with four copies of nsp9 within the asymmetricunit.The nsp9 G104E model was subjected to a singl

97、e round of rigid body refine-ment as implemented within Refmac5 as part of the CCP4 package (3). Themodel was then refined using alternating cycles of manual fitting in Coot andTURBO-FRODO (8, 17) and simulated annealing in CNS 1.2 (6). Compositesimulated annealing omit maps were generated at regula

98、r intervals to guide themodeling efforts. After several rounds of rebuilding group B factors were refinedin conjunction with 50 cycles of energy minimization (CNS). In the final roundsof rebuilding and refinement, waters, ions, and cryoprotectant molecules wereadded to the model through the use of a

99、n Fo-Fcmap and a 2Fo-Fcomit mapcontoured to 3 ? and 1 ?, respectively. Model quality was monitored by usingCNS and Molprobity (11). The final structure containing four monomers of nsp9,134 waters, eight PO4, and 16 glycerols has been deposited into the Protein DataBank (PDB) with the accession code

100、3EE7. The data and refinement statisticsare listed in Table 1.Structure analysis. The individual monomers of nsp9 G104E were aligned(least-squares superposition) with one another and with monomers from 1UW7(57) and 1QZ8 (16). Dimer-only structures were also created from the G104Estructure and superi

101、mposed on the analogous dimers from 1QZ8 (16) or 1UW7(57). In all cases, a C?root mean square deviation (RMSD) was calculated byusing either PyMol (12) or Coot. In the case of features such as the symmetryrelated helix-helix dimer formed by G104E, manual inspection (Coot) was madeof the helices comp

102、ared to 1QZ8 (16). Further examination of the helix-helixdimerization interface was accomplished through ProMotif, as found within thePDBSum webserver (26, 3335).RESULTSExpression and purification. Variants of nsp9 were gener-ated with the purpose of disrupting the helix-helix dimer foundin the crys

103、tal structure of nsp9 as solved by Egloff et al. (16) toassess its biological relevance. Mutations at G100 and G104were created and verified through DNA sequencing. Wild-typensp9, along with the G100E, G104E, and G104V mutants, werefound to induce to high levels of soluble expression uponinduction w

104、ith IPTG.3010MIKNIS ET AL.J. VIROL. on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from Wild-type, G100E, G104E, and G104V forms of nsp9 werepurified using IMAC (HisTrap HP; GE Healthcare), whichyielded protein of 85% purity, aliquots of which were used forsize estimation b

105、y SEC and SEC-MALS. Further purificationto obtain stocks for crystallization, EMSA, and FA analyseswas obtained by applying protein from the IMAC column to apreparative gel filtration column (Superdex 75 HL 16/60; GEHealthcare). The final protein yield was of ?98% purity asjudged by SDS-PAGE stained

106、 with silver (data not shown).Due to the presence of a noncleavable C-terminal His6tag andlinker, the pure nsp9 monomer had an expected molecularmass of 13.4 kDa. The structure of the G104E mutant inaddition to the wild-type nsp9 structures found in the PDBsuggested that the location of the C-termin

107、al His6tag does notaffect the stability of the helix-helix dimer.Size estimation using SEC and SEC-MALS. For analysis ofthe mutants oligomeric state, aliquots of peak fractions fromIMAC were immediately applied to an analytical gel filtrationcolumn. The elution from the Superdex column was directed

108、toa Wyatt Systems Mini-DAWN coupled with an Optilab DSP toallow for simultaneous collection of SEC data in conjunctionwith MALS data.While SEC is commonly used to estimate the molecularweight of a sample and thus the oligomerization state, themolecular weight estimates are highly dependent on the co

109、n-formation of a protein and any interactions between the pro-tein and the chromatographic matrix. In effect, a protein witha nonglobular structure or one that interacts with the SECmatrix by another property other than size (e.g., electrostatic,hydrophobic, etc.) will elute at a volume correspondin

110、g to aprotein with a much larger size. Size estimates made by MALSare independent of nonideal SEC behavior and have beenshown to give a more accurate size determination for a varietyof proteins than is possible using SEC alone (18). The additionof the SEC to the system has the added benefit of effec

111、tivelyseparating all species within a sample prior to their analysis byMALS, leading to data which is more amenable to size assign-ment and has an expected error of only ?5% (18).Nsp9 wild type eluted as a single peak and light scatteringestimated a molecular size of 25.5 kDa, corresponding to apart

112、icle size very close to the predicted value of 26.8 kDa for adimer. In contrast, the SEC-MALS data on the nsp9 mutantsgave molecular size estimates in good agreement with theprotein being monomeric (Table 2). Molecular size estimateswere also made using elution data from SEC alone by calcu-lating a

113、calibration curve using well-characterized proteins,also shown in Table 2. Sutton et al. reported Kdvalues of 0.16to 0.46 mM for the formation of the dimer (57). However, ourgel filtration experiments on wild-type nsp9 using an analyticalcolumn indicate that dimerization occurs at protein concentra-

114、tions between 40 and 105 ?M. Similar experiments showedG100E, G104E, and G104V mutants were monomeric at orabove concentrations where the wild-type protein was dimeric.The transition of nsp9 wild type from dimer to monomercould not be detected using ITC. The concentration rangerequired to capture th

115、e dissociation from dimer to monomerwas too low to generate any substantial change in enthalpyabove the background. The G100E and G104E samples be-haved in ITC experiments as monomers and showed no signsof a monomer-dimer equilibrium. A weak dimerization con-stant of 0.871 mM was determined for G104

116、V.Secondary structure analysis. Analysis of nsp9 samplesthrough the use of CD spectroscopy showed that the overallTABLE 1. X-ray data and refinement statisticsGeneral category andspecific parameterValueaCrystal dataSpace group.P31a ().91.836b ().91.836c () .84.217?, ?, ? ().90, 90, 120Mol/AUb.4Vmc.3

117、.81% Solvent.67.76Data collectionResolution range () .79.5662.6 (2.692.60)Wavelength ().0.9792No. of observations.235,950No, of unique reflections.24,437I/?I.34.01 (2.75)Redundancy.3.7 (2.8)Data completeness.98.8Rmerged.5.3 (39.4)RefinementNo. of reflections used.24,148Resolution range () .79.5662.6

118、Free set size (%).9.80No. of atoms (protein).3,462No. of atoms (nonprotein).252Rworke.0.2138Rfreef.0.2694Mean B value (2).70.66RMSDBond length ().0.006402Bond angle ().1.69421Ramachandran plotFavored (%) .91.8Allowed (%).7.3Disallowed (%) .0.9aNumbers in parentheses are for the highest resolution sh

119、ell.bMolecules per asymmetric unit.cMatthews coefficient.dRmerge?hkl?i?Ihkl,i? ?Ihkl?hkl,i?Ihkl?, where Ihkl,iis the intensity of an individualmeasurement of the reflection with Miller indices h, k, and l, and ?Ihkl? is themean intensity of that reflection.eRwork?hkl?Fobshkl? ? ?Fcalchkl?Fobshkl?, w

120、here ?Fobshkl? and ?Fcalchkl? are the ob-served and calculated structure amplitudes.fRfreeis Rworkcalculated with the free set data only, which were omitted fromrefinement.TABLE 2. nsp9 sizes estimated by SEC-MALS and SECProteinSize (kDa)MonomerSuperdex 75SEC-MALSnsp9 wild type13.426.425.5nsp9 G100E

121、13.518.515nsp9 G104E13.517.816.3nsp9 G104V13.517.2613.64VOL. 83, 2009SARS-CoV nsp9 DIMERIZATION3011 on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from secondary structure composition was maintained in all sam-ples. As shown in Table 3, the secondary structure distributioni

122、n the wild-type, G100E, G104E, and G104V proteins wasestimated from the CD data. Given that structure assignmentfor all samples was nearly identical to percentages predictedfrom the crystal structure, it was determined that the disrup-tion of dimerization seen in the SEC and SEC-MALS data wasnot due

123、 to protein misfolding or the C-terminal His tag. In par-ticular, the amount of ?-helix remained similar to the wildtype, indicating that the mutations had not affected the helicalstructure of the dimerization interface. In addition, the crystalstructure of the G104E mutant (discussed below) showed

124、thatsecondary structure was not disrupted by the mutation. Mea-surement of the melting temperature of the samples using CDspectroscopy revealed that wild-type, G100E, and G104V pro-teins have similar melting temperatures (58.6 to 62.5C), whileG104E protein had a melting temperature that was signifi-

125、cantly higher (75C). While it is unclear why G104E is morestable, it does indicate that the individual mutations do notdestabilize nsp9.Gel shift assays. In order to determine whether the muta-tions and subsequent disruption of dimerization altered theability of nsp9 to bind to RNA, we performed RNA

126、 EMSAsusing standard techniques. To maximize the visualization ofRNA-protein complexes, samples were UV cross-linked priorto separation of complexes from free RNA through electro-phoresis (24, 25). RNA EMSA experiments revealed that wild-type nsp9 exhibited clear binding with the RNA templates (Fig.

127、2). Bovine serum albumin controls did not bind RNA (data notshown). Importantly, the G100E, G104E, and G104V mutantswere also able to bind ssRNA (Fig. 2). To date, the RNA-binding surface for nsp9 has not been determined; however,the ability of the mutants to bind suggests that the dimerinterface is

128、 not critical for RNA binding. Changes in mobilitywere seen between the wild type and the mutants, but thiscould be a result of the change in protein charge or size of thecomplexes due to the monomeric or dimeric state of the pro-tein. Unlabeled ssDNA probe at a 10-fold excess was unable tocompete w

129、ith the ssRNA for binding and resulted in betterresolution of the nsp9/RNA bands (Fig. 3).FA. While EMSA allowed for the visualization of the inter-actions between the RNA and nsp9, it did not allow for quan-titative measurements of the binding affinities and comparisonsto known data. Consequently,

130、RNA-binding measurements weremade using a fluorescently labeled 20-mer of ssRNA on afluorometer equipped for measurement of FA data (Fig. 4).Wild type bound to ssRNA with a Kdof 55 nM compared toKds of 250, 620, and 320 nM for G104V, G104E, and G100E,respectively.Reverse genetics. The three mutants

131、were designed for andincorporated into the wild-type SARS-D fragment of icSARS.The three mutant SARS-D fragments were assembled, ampli-fied, sequence verified, and placed into the infectious clone ofSARS-CoV. The full-length cDNAs were then transcribed andTABLE 3. nsp9 structural features as measure

132、d byCD spectroscopySample%Mean Tm? SD(C)SheetcontentHelixRandomnsp9 wild type3995162.5 ? 1.6nsp9 G100E4684660.5 ? 3.8nsp9 G104E3995175 ? 0.7nsp9 G104V4494758.6 ? 2.3Expected value41652NAaaNA, not applicable.FIG. 2. Wild-type nsp9 along with mutants bind ssRNA. Lane 1,nsp9 wild type/RNA (WT); lane 2,

133、 G100E/RNA; lane 3, G104E/RNA;lane 4, G104V/RNA; lane 5, ssRNA. The positions of free probe andprotein/RNA shifts are indicated. Protein-RNA complexes were re-solved on a 4 to 20% TBE gel and observed after detection of thebiotinylated ssRNA.FIG. 3. ssDNA does not prevent a ssRNA from binding to nsp

134、9.Wild-type nsp9 was loaded into lanes 1, 4, and 7. nsp9 G100E is foundin lanes 2, 5, and 8. G104E nsp9 is loaded into lanes 3, 6, and 9. Lanes1 to 3 contain only RNA and protein, lanes 4 to 6 contain equivalentamounts of ssRNA and ssDNA, and lanes 7 to 9 contain an 10-foldexcess of ssDNA over ssRNA

135、. Lane 10 is an RNA-only loading con-trol.3012MIKNIS ET AL.J. VIROL. on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from transfected into Vero cells and monitored for CPE. In allcases, detection of CPE was not evident. To determinewhether low-level RNA synthesis occurred, R

136、T-PCR was con-ducted to assay for leader containing transcripts using primersdesigned to detect subgenomic S, but no bands were detected,suggesting either that no sgRNA synthesis was occurring orthat transcript levels were below the levels of detectability.Mutants G100E and G104E were introduced int

137、o the infec-tious clone three times, and in all three cases no recombinantmutant virus was detected or rescued. However, recombinantwild-type icSARS prepared in parallel did produce CPE andsgRNA synthesis. For mutant G104V, the experiment was re-peated four times. The first two times there was no ev

138、idence ofviral replication by cytopathology or by RT-PCR. The thirdtime, however, CPE was evident, and subgenomic S was de-tected by RT-PCR. In addition, the recombinant virus wassuccessfully passaged in Vero cells and also produced wildtype-sized plaques. Total viral RNA was harvested and theregion

139、 flanking the mutation site was amplified by RT-PCR,electroporated on a 0.8% agarose gel, gel purified, and se-quenced. Interestingly, sequence analysis revealed that theG104V mutation had reverted to wild type-SARS-CoV at thatposition. The mutated codon introduced was GTG which en-codes a Val, and

140、this reverted to GGG which encodes a Gly.Interestingly, the wild-type virus had a Gly at that position, butthe codon was GGC. Since two changes were originally intro-duced and only one change was sufficient to revert it to awild-type sequence at that site, the other mutation was stillpresent, indica

141、ting that it was indeed a revertant and not anaccidental contamination with wild type-SARS D fragment.This suggests that the G104V mutant likely replicates veryinefficiently, which allowed the revertant to evolve. The G104Vmutant was generated and tested a fourth time, with resultsidentical to those

142、 of the first two trials, demonstrating therarity of this event.Crystallization and structure determination. Crystals ofnsp9 G104E grew in a hexagonal, rod-like morphology anddiffracted to a maximum of 2.6- resolution on beamline 11-1of SSRL. No significant disruptions in the secondary structurewere

143、 seen, with nsp9 G104E exhibiting seven beta strands andone alpha helix as previously observed in 1UW7 (57) and1QZ8 (16). As determined by molecular replacement (Fig.5A), four monomers of nsp9 G104E are present within theasymmetric unit. In four regions residues could not be modeleddue to the lack o

144、f 2Fo-Fcomit density (chain A residues 36 and37 and residues 60 and 61, chain C residues 37 and 38, andchain D residue 37). This arrangement forms a helix-like bun-dle of the individual monomers through two separate motifs,the previously observed sheet-sheet dimer interface and a newloop-sheet inter

145、face, with each monomer oriented roughly per-pendicular to each other along the helix axis. The helix-helixdimer interface is generated through crystallographic symme-try (Fig. 5B); however, the interface is structurally distortedand will be discussed below. The helical arrangement of nsp9G104E mono

146、mers forms a continuous strand parallel to the baxis of each unit cell throughout the crystal by crystallographicsymmetry, which continually exposes a charged surface alongthe edge of the helix (Fig. 5C). At the center of the asymmetricunit is the previously observed sheet-sheet dimer interfacewhich

147、 has a C?RMSD of 1.92 compared to the sheet-sheetdimer reported in 1UW7 (57). The G104E sheet-sheet dimerintroduces an additional six hydrogen bonds and 11 non-bonded interactions compared to 1UW7 (57). A loop-sheetinterface is present at either end of the central sheet-sheetdimer and results in the

148、 generation of the helical motif in theasymmetric unit. This interface is formed through the associ-ation of the loop between beta strands 6 and 7, with strand 6 ofan opposing monomer. Residues Asp78-C and Asp78-B makehydrogen bonds with residues Asp47-D/Lsy86-D and Asp47-A/Lsy86-A, respectively. Th

149、e two loop-sheet interactionspresent in the tetramer each bury roughly 800 2of surfacearea. Moderate sequence conservation is found in these resi-dues.In the G104E mutant structure, the helix-helix dimer ispresent as in both 1QZ8 (16) and 1UW7 (57) and is thought tobe the biologically relevant dimer

150、, as seen in Fig. 5B. Thehelix-helix interface is disturbed but not disrupted completely,with the presence of Glu instead of Gly at the interface causingthe C?positions at position 104 to move to 6.56 apart,compared to 4.76 in 1QZ8 (16) and 5.4 in 1UW7 (57), alsoshown in a superposition in Fig. 5B.

151、The distances betweenother atoms in the helices are essentially maintained, com-pared to 1QZ8 (16) and 1UW7 (57), with an overall RMSDover the helices of 0.66 . In addition, the helical crossingangles are markedly different for G104E (?50.9) compared to1QZ8 (16) and 1UW7 (57) (?41.1 and ?38.7, respe

152、ctively).The helix-helix interface is held together by hydrogen bondsthat form between either the OE1 or OE2 atoms of the Gluside chain and the main-chain nitrogen of Leu 9 of the oppos-ing strand (Fig. 5D). A number of phosphate ions were foundwithin the structure coordinated by surface charged res

153、iduesthought to be involved in nucleotide binding. Specific overlapsupon superposition of 1QZ8 (16) with our structure occur atFIG. 4. FA measurement of RNA binding by wild-type (WT) (f),G100E (), G104E (?), or G104V () nsp9. Each measurement is anaverage of 10 individual measurements. The data were

154、 fitted to theequation given in Materials and Methods to generate the curves de-picted in the graph and derive binding equilibrium Kdvalues.VOL. 83, 2009SARS-CoV nsp9 DIMERIZATION3013 on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from sulfate 693 of 1QZ8 (16) with glycerol

155、s 3 and 8 and phosphates1 and 2 of G104E. If the phosphates are taken as representa-tive binding areas for backbone phosphates of target nucleicacid, then the nearby residues are potential sites to target formutagenesis examining RNA or DNA binding. The overallsymmetry motifs generated by crystallog

156、raphic symmetry in1QZ8 (16) and 1UW7 (57) are not replicated in our structuredue to the presence of the loop-sheet interaction and the 31screw-axis.DISCUSSIONSARS-CoV, like other CoVs, produces a number of smallhighly conserved RNA-binding proteins essential to viral rep-lication. We have used a com

157、bination of structure guided mu-tagenesis and reverse genetics to evaluate the role of nsp9 inreplication. In the present study, G100 and G104 of a GXXXGdimerization motif were mutated to determine whether thensp9 homodimer could be disrupted at the conserved helix-helix interface and whether the ho

158、modimer was relevant tovirus replication. nsp9 mutants were analyzed in vitro by SEC-MALS, CD spectroscopy, EMSA, and FA. The crystal struc-ture of G104E nsp9 was solved, and disruption of the helix-helix interface was observed. In addition, each nsp9 mutantwas introduced into the virus using revers

159、e genetics, and the invivo effects were monitored. The biochemical data stronglysuggest that mutation of either glycine results in the produc-tion of monomeric nsp9 in vitro and nonviable virus in vivo.nsp9 mutations disrupt dimerization. An investigation of anonredundant PDB set revealed that the G

160、XXXG motif beingexamined in nsp9 was well represented among soluble proteins(31) but has not been extensively studied for its contributionsto homo- and heterodimerization of soluble proteins contain-ing the motif. The motif was also commonly found in trans-membrane helices, where it was postulated t

161、o allow for closeapproach of adjacent helices for subsequent dimerization. Inseveral examples, mutation of either Gly in the motif lead todecreased or complete loss of dimerization of transmembranehelices (45, 51, 56). The SEC-MALS and SEC analyses inconjunction with EMSA data showed that G100E, G10

162、4E, andG104V mutants were monomeric in solution, while wild-typensp9 was dimeric, indicating that the GXXXG motif was im-portant for maintaining the nsp9 dimer. This also indicates thatthe helix-helix dimer and not the sheet-sheet dimer is themajor dimer in solution.FIG. 5. Tetrameric arrangement of

163、 G104E crystals. (A) G104E structure exhibiting sheet-sheet dimer (center of structure, yellow and bluemonomers) and loop-sheet interface (flanking sheet-sheet dimer on either side, red/yellow or blue/green monomers) present within the asymmetricunit containing four monomers of nsp9. (B) Helix-helix

164、 dimer in the G104E structure, similar to the helix-helix dimer in 1QZ8. Alignment of thetwo structures is depicted with 1QZ8 in red and G104E in green. Significant deviations in both crossing angles and buried surface area arise asa result of the G104E mutation. (C) Electrostatic surface of helix-l

165、ike arrangement of G104E monomers along the b axis of a unit cell. Colors:blue, negative surface; white, neutral surface; red, positive surface. (D) Interactions stabilizing the helix-helix dimer. Primary contacts are formedbetween G104E and L9 of a symmetry related monomer. Distances between atoms

166、are in angstroms. The L9? and G104E? designations indicateresidues are a from symmetry-related monomer. Images were prepared by using PyMol.3014MIKNIS ET AL.J. VIROL. on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from RNA binding by nsp9. The results from the EMSA of wild-

167、type nsp9 and the G100E, G104E and G104V mutants showthat each protein binds RNA, indicating the dimeric state ofthe protein is not critical for RNA binding. nsp9 apparentlyutilizes RNA as the preferred ligand. An unlabeled DNAoligonucleotide could not compete with labeled RNA duringan EMSA, which c

168、onfirms reports by Egloff et al., who sug-gested that nsp9 preferred RNA over DNA (16).In order to distinguish quantitatively between wild-type andmutant RNA binding affinities, FA measurements were used.FA measurements using an ssRNA 20-mer give Kdvalues of 55nM for wild type and a range from 250 t

169、o 650 nM for themutants, which is comparable to the values obtained by Egloffet al. in which a larger fragment of RNA (560 bases) wasreported to have a binding constant of 400 nM, using an assaymonitoring the change in tryptophan fluorescence (16). Theequilibrium binding constants presented here wer

170、e determinedat a lower ionic strength (15 mM NaCl) and may be lower dueto coulombic effects. The G100 and G104 mutants have Kdvalues higher than that of the wild type ranging from 250 to 650nM. Disruption of the nsp9 dimer does not abrogate binding ofRNA but does result in a 5- to 12-fold decrease i

171、n affinity.The relatively low affinity binding constants may reflect that,as part of the replication complex, nsp9 may not have a specificbinding sequence but may act in conjunction with nsp7 andnsp8 as a processivity factor. Among the SARS-CoV nsps,nsp9 has already been identified to have binding i

172、n low micro-molar affinity for ssRNA and ssDNA (16, 57), nsp10 was re-ported to have high micromolar affinity for dsRNA (29), whilensp8 and the nsp7-8 hexadecamer were also shown to binddouble-stranded RNA and double-stranded DNA with mod-erate efficiency (67). The interaction of these RNA-bindingpr

173、oteins with relatively weak binding affinities suggests that thepresence of a complete replication complex may be necessaryfor high-affinity binding. nsp9 has been shown in several casesto interact with these proteins through several different meth-ods (57). A specific target RNA sequence for nsp9 w

174、ithin theSARS-CoV genome has not been identified, although interac-tions with a stem-loop in the 3? region of the genome wererecently reported (69).Local conditions could also affect binding of nucleic acid tothese proteins. CoVs use large membrane-associated replicasecomplexes found within viral ge

175、nerated double-membrane ves-icles to achieve viral replication (53); thus, it can be assumedthat the local concentration of RNA around nsp9 would bemarkedly higher than nanomolar levels. Alternatively, multipleweak binding sites in nsp9 may contribute to the formation ofthe viral replication complex

176、 without altering the individualnsp9 proteins. In addition, the presence of the replicationcomplex may serve to introduce protein-RNA interaction sitesthat are not available for the individual nsps.Dimerization is critical for viral function. Several experi-ments suggest that the nsp9 dimer is requi

177、red for viral growth.In vitro, the SEC-MALS data show that the dimerization ofnsp9 was successfully disrupted with the G100E, G104E, andG104V mutations. In vivo, the G100E and G104E mutations innsp9 were lethal to efficient virus replication in cell culture,based on detection for subgenomic mRNAs. I

178、n contrast, theG104V mutation displayed a single example of delayed growthkinetics with reversion to an alternate codon for wild-type virusproduction.Analysis of the genomic RNA and subgenomic mRNA usingreal-time PCR suggested that, while subgenomic synthesis wasnot detected at any time, genomic RNA

179、 increased slightly atearly time points before decreasing at later times during infec-tion. This increase was not significantly different than back-ground (data not shown), but this observation leaves open thepossibility that genomic plus-strand RNA synthesis occurred ata reduced rate in the G104V m

180、utant. van Marle et al. con-structed a mutant incorporating a single point mutation in ansp that only made genome length molecules but not mRNA inthe related equine arteritis virus (59). Low-level genomic rep-lication similar to that observed in equine arteritis virus mayhave provided the opportunit

181、y for the reversion to occur in theG104V mutation. An additional possibility would be that anerror in the T7 transcription step could have introduced therevertant nucleotide in the G104V mutant; however, the errorrate for the T7 RNA polymerase is one error per 20,000 nu-cleotides and, when this erro

182、r rate is factored in with a genomesize of 29,751 nucleotides, the probability of an error occurringin that specific nucleotide would be ?1/500,000,000.Functionally, it would seem that during infection nsp9 mustform a dimer to properly bind and orient RNA for subsequentuse by the replicase machinery

183、 despite the ability of the mu-tants to bind RNA in vitro. It is unlikely that the nsp9 dimermutations have an effect on processing of the ORF1a andORF1ab polyproteins. The MHV nsp9-10 fusion studied byDeming et al. (13) was shown to be functionally competent forviral replication, indicating that ns

184、p9-10 did not need to beprocessed for replication to occur. There is some evidenceusing light scattering that an nsp9-10 construct is dimeric insolution (Z. J. Miknis and L. W. Schultz, unpublished results).Further modeling indicates that the C termini of nsp9 can lineup with the N termini of nsp10

185、and still maintain the dimerstructures of both proteins. Therefore, the nsp9-10 fusion maybe functionally identical to its wild-type processed counter-parts.Interestingly, the G104V mutation was still able to producerevertant viruses, despite SEC and SEC-MALS data indicatingthat it is monomeric. ITC

186、 data suggest that G104V can form adimer, but ?8-fold more weakly than the wild type, as esti-mated by SEC. Dimerization has been shown in the case of theRNA-binding protein NS1 of influenza virus to be necessaryfor RNA binding (61), as well as for the She2P protein fromyeast (46). The hepatitis A v

187、irus 3C protease has affinity forviral RNA that is increased from millimolar to micromolarlevels upon dimerization of the enzyme (48). G104V may beable to form a weakly associated helix-helix dimer that is stillable to function at a reduced level within the virus, allowingsufficient virus replicatio

188、n to evolve a revertant strain.While the nsp9 dimerization mutants are able to bind toRNA effectively, the mutations could also disrupt interactionsthat allow them to form hetero-oligomers with other SARSproteins. In rabies virus, the nucleoprotein shows no sequencespecificity for RNA; however, when

189、 it is incorporated into theribonucleoprotein complex that is ultimately responsible forRNA transcription, the specificity is gained for genomic RNA(39). A similar mechanism may exist in SARS, where nsp9 canVOL. 83, 2009SARS-CoV nsp9 DIMERIZATION3015 on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:

190、/jvi.asm.org/Downloaded from switch from nonspecific binding to specific binding upon for-mation of the replicase complex.Structural analysis. The crystal structure of G104E showsthe clear formation of three interaction motifs defined by ahelix-helix dimer, sheet-sheet dimer, and a loop-sheet inter-

191、face. This is in contrast to SEC and SEC-MALS, where noG104E dimers or tetramers were seen in solution, even inexperiments performed at protein concentrations used in thecrystallizations. The concentration of nsp9 G104E in the crys-tal may be high enough to favor dimer, or there may be enoughenergy

192、gained from formation of the crystal lattice and othercrystal contacts to force dimerization. Hydrogen bonds formbetween the OE1 or OE2 atoms of the G104E side chainacross the dimer interface with the main-chain nitrogen of Leu9. These hydrogen bonds serve to allow the formation of thedimer in the c

193、rystal, the equivalent of which are not accessiblein the G104V and G100E mutants. The structure does showsignificant disruption of the dimer interface, including a widerseparation between the two halves of the dimer and a steeperhelix crossing angle. Together, these differences may beenough to disru

194、pt the formation of a dimer in solution.The G104E mutant showed a higher Tmthan did the wildtype, the G100E mutant, or the G104V mutant as measured byCD spectroscopy. It is unclear why the G104E mutant displaysa Tm?10C higher than that for the other mutants and thewild type. The G104E was also the o

195、nly sample that producedcrystals which may be a reflection of its greater thermal stabil-ity compared to the wild type and the G100E or G104V mu-tant.A recent report has described an antiparallel helix-helixdimer of nsp9 from HCoV-229E that is held together by adisulfide bridge (49). For the experim

196、ents described here, re-ducing agents were maintained in wild-type and mutant nsp9samples in all experiments. However, wild-type nsp9 samplesshowed no evidence of disulfide bond formation and could bedissociated into monomers under oxidizing conditions in solu-tion, as monitored by high-resolution S

197、EC. This may representa significant difference between the function of nsp9 in SARS-CoV and HCoV-229E.As suggested by Sutton et al. (57), it is likely that the sheet-sheet dimer formed in the center of the tetramer is notbiologically relevant. The new loop-sheet interface and thehelix-helix dimer bu

198、ry roughly equivalent surface areas (?8002) and make approximately the same number of intermolec-ular contacts. The new loop-sheet interface does have contactsfrom residues D78, D47, and K86 that are moderately con-served among various CoVs. The sheet-sheet dimer has a sig-nificantly smaller buried

199、surface area but has more H-bondsand nonbonded contacts for stabilization.Interestingly, the formation of the sheet-sheet dimer flankedby the loop-sheet interface within the tetramer results in theformation through crystallographic symmetry of a continuoushelical arrangement of nsp9 along the b axis

200、 of the crystal. Thisarrangement of tetramers exposes a near-continuous patch ofpositively charged surface wrapping around the overall helicalstructure, suggesting the potential for binding of long stretchesof RNA. Although this tetramer was not observed in solution,it does suggest that nsp9 is capa

201、ble of forming polymeric struc-tures that might act as scaffolds for binding genomic RNA.Summary. Crystallography is an essential tool for the studyof protein structure and function. However, the biological im-plications of structures are often not clear. In the case of theSARS-CoV nsp9 RNA-binding

202、protein, several different dimerinterfaces have been proposed based on interactions seen inthe crystal. We have shown that the conserved helix-helixdimer interface containing a GXXXG protein-protein interac-tion motif is biologically relevant to SARS-CoV replication.Disruption of this interface by s

203、ite-directed mutagenesis of theglycine residues in the GXXXG motif resulted in monomericforms of nsp9 in solution. Subsequent introduction of muta-tions into SARS-CoV by reverse genetics showed that theformation of the nsp9 dimer was necessary for viral viability. Asingle mutant G104V reverted to wi

204、ld-type nsp9, indicatingsignificant evolutionary and functional pressure on the dimer-ization interface. The RNA-binding affinity of monomeric nsp9mutants was reduced but not eliminated, indicating that thedimer retains a slight advantage over the monomer in RNAbinding. The inability of the nsp9 mon

205、omers to function in vivomay not reflect an RNA-binding ability but rather the correctpositioning of RNA in the replication complex requiring aproperly dimerized nsp9. This view is supported by the crystalstructure of G104E nsp9 in which the helix-helix dimer inter-face is significantly disrupted fr

206、om wild-type nsp9. Future ex-periments combining crystallography and in vitro biophysicalanalysis of protein-protein and protein-RNA interactions willbe necessary to unravel the complicated structure of the CoVreplication complex. Collectively, our data add to a growingbody of literature that implic

207、ates nsp9 as a key ingredient thatintimately engages other proteins in the replicase complex tomediate efficient virus transcription and replication.ACKNOWLEDGMENTSWe thank Wendy Franke, Carleen Pope, and Peggy Cegielski fortechnical assistance and Mark Denison (Vanderbilt University) for thegift of

208、 the original nsp9 plasmid.Portions of this research were carried out at the SSRL, a nationaluser facility operated by Stanford University on behalf of the U.S.Department of Energy, Office of Basic Energy Sciences. The SSRLStructural Molecular Biology Program is supported by the Departmentof Energy,

209、 Office of Biological and Environmental Research, and bythe National Institutes of Health, National Center for Research Re-sources, Biomedical Technology Program, and the National Instituteof General Medical Sciences. The work described here was funded inpart by NIAID grant R21AI061057 (to L.W.S. an

210、d T.C.U.), the JohnR. Oishei Foundation (to L.W.S. and T.C.U.), the William G.McGowan Charitable Fund (to L.W.S. and T.C.U.), and the StateUniversity of New York at Buffalo (Z.J.M.). This study was also sup-ported by NIH grants AI23946 and AI059136 to R.S.B.REFERENCES1. Andrade, M. A., P. Chacon, J.

211、 J. Merelo, and F. Moran. 1993. Evaluation ofsecondary structure of proteins from UV circular dichroism spectra using anunsupervised learning neural network. Protein Eng. 6:383390.2. Bahadur, R. P., P. Chakrabarti, F. Rodier, and J. Janin. 2003. Dissectingsubunit interfaces in homodimeric proteins.

212、Proteins 53:708719.3. Bailey, S. 1994. The Ccp4 Suite: programs for protein crystallography. ActaCrystallogr. D Biol. Crystallogr. 50:760763.4. Bost, A. G., R. H. Carnahan, X. T. Lu, and M. R. Denison. 2000. Fourproteins processed from the replicase gene polyprotein of mouse hepatitisvirus colocaliz

213、e in the cell periphery and adjacent to sites of virion assembly.J. Virol. 74:33793387.5. Brockway, S. M., C. T. Clay, X. T. Lu, and M. R. Denison. 2003. Character-ization of the expression, intracellular localization, and replication complexassociation of the putative mouse hepatitis virus RNA-depe

214、ndent RNApolymerase. J. Virol. 77:1051510527.6. Brunger, A. T., P. D. Adams, G. M. Clore, W. L. DeLano, P. Gros, R. W.3016MIKNIS ET AL.J. VIROL. on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from Grosse-Kunstleve, J. S. Jiang, J. Kuszewski, M. Nilges, N. S. Pannu, R. J.Rea

215、d, L. M. Rice, T. Simonson, and G. L. Warren. 1998. Crystallography andNMR system: a new software suite for macromolecular structure determi-nation. Acta Crystallogr. D Biol. Crystallogr. 54:905921.7. Burrows, S. D., M. L. Doyle, K. P. Murphy, S. G. Franklin, J. R. White, I.Brooks, D. E. McNulty, M.

216、 O. Scott, J. R. Knutson, D. Porter, P. R. Young,and P. Hensley. 1994. Determination of the monomer-dimer equilibrium ofinterleukin-8 reveals it is a monomer at physiological concentrations. Bio-chemistry 33:1274112745.8. Cambillau, C., et al. 1997. TURBO-FRODO. http:/www.afmb.univ-mrs.fr/-TURBO-.9.

217、 Chen, S., L. L. Chen, J. Z. Tan, J. Chen, L. Du, T. Sun, J. H. Shen, K. X.Chen, H. L. Jiang, and X. Shen. 2005. Severe acute respiratory syndromecoronavirus 3C-like proteinase n terminus is indispensable for proteolyticactivity but not for enzyme dimerization: biochemical and thermodynamicinvestiga

218、tion in conjunction with molecular dynamics simulations. J. Biol.Chem. 280:164173.10. Claude, J. B., K. Suhre, C. Notredame, J. M. Claverie, and C. Abergel. 2004.CaspR: a web server for automated molecular replacement using homologymodeling. Nucleic Acids Res. 32:W606W609.11. Davis, I. W., A. Leaver

219、-Fay, V. B. Chen, J. N. Block, G. J. Kapral, X. Wang,L. W. Murray, W. B. Arendall III, J. Snoeyink, J. S. Richardson, and D. C.Richardson. 2007. MolProbity: all-atom contacts and structure validation forproteins and nucleic acids. Nucleic Acids Res. 35:W375W383.12. DeLano, W. L. 2002. The PyMOL mole

220、cular graphics system. DeLano Sci-entific, San Carlos, CA.13. Deming, D. J., R. L. Graham, M. R. Denison, and R. S. Baric. 2006. MHV-A59 ORF1a replicase protein nsp7-nsp10 processing in replication. Adv.Exp. Med. Biol. 581:101104.14. Donaldson, E. F., A. C. Sims, R. L. Graham, M. R. Denison, and R.

221、S. Baric.2007. Murine hepatitis virus replicase protein nsp10 is a critical regulator ofviral RNA synthesis. J. Virol. 81:63566368.15. Drosten, C., S. Gunther, W. Preiser, S. van der Werf, H. R. Brodt, S. Becker,H. Rabenau, M. Panning, L. Kolesnikova, R. A. M. Fouchier, A. Berger,A. M. Burguiere, J.

222、 Cinatl, M. Eickmann, N. Escriou, K. Grywna, S.Kramme, J. C. Manuguerra, S. Muller, V. Rickerts, M. Sturmer, S. Vieth,H. D. Klenk, A. D. M. E. Osterhaus, H. Schmitz, and H. W. Doerr. 2003.Identification of a novel coronavirus in patients with severe acute respiratorysyndrome. N. Engl. J. Med. 348:19

223、671976.16. Egloff, M. P., F. Ferron, V. Campanacci, S. Longhi, C. Rancurel, H. Dutar-tre, E. J. Snijder, A. E. Gorbalenya, C. Cambillau, and B. Canard. 2004. Thesevere acute respiratory syndrome-coronavirus replicative protein nsp9 is asingle-stranded RNA-binding subunit unique in the RNA virus worl

224、d. Proc.Natl. Acad. Sci. USA 101:37923796.17. Emsley, P., and K. Cowtan. 2004. Coot: model-building tools for moleculargraphics. Acta Crystallogr. D Biol. Crystallogr. 60:21262132.18. Folta-Stogniew, E., and K. R. Williams. 1999. Determination of molecularmasses of proteins in solution: implementati

225、on of an HPLC size exclusionchromatography and laser light scattering service in a core laboratory. J. Bi-omol. Tech. 10:5163.19. Fouchier, R. A. M., T. Kuiken, M. Schutten, G. van Amerongen, J. vanDoornum, B. G. van den Hoogen, M. Peiris, W. Lim, K. Stohr, andA. D. M. E. Osterhaus. 2003. Aetiology:

226、 Kochs postulates fulfilled for SARSvirus. Nature 423:240241.20. Gimpelev, M., L. R. Forrest, D. Murray, and B. Honig. 2004. Helical packingpatterns in membrane and soluble proteins. Biophys. J. 87:40754086.21. Gonzalez, A., P. Moorhead, S. McPhillips, and N. Sauter. 2005. New tools tointegrate data

227、 analysis and data collection at SSRL: Web-Ice. Acta Crystal-logr. A Found. Crystallogr. 61:C486.22. Goodsell, D. S., and A. J. Olson. 2000. Structural symmetry and proteinfunction. Annu. Rev. Biophys. Biomol. Struct. 29:105153.23. Graham, R. L., J. S. Sparks, L. D. Eckerle, A. C. Sims, and M. R. De

228、nison.2008. SARS coronavirus replicase proteins in pathogenesis. Virus Res. 133:88100.24. Greenberg, J. R. 1979. Ultraviolet light-induced cross-linking of messengerRNA to proteins. Nucleic Acids Res. 6:715732.25. Gu, H. D., and D. R. Schoenberg. 2003. U2AF modulates poly(A) lengthcontrol by the pol

229、y(A)-limiting element. Nucleic Acids Res. 31:62646271.26. Hutchinson, E. G., and J. M. Thornton. 1996. PROMOTIF: a program toidentify and analyze structural motifs in proteins. Protein Sci. 5:212220.27. Imbert, I., E. J. Snijder, M. Dimitrova, J. C. Guillemot, P. Lecine, and B.Canard. 2008. The SARS

230、-Coronavirus PLnc domain of nsp3 as a replication/transcription scaffolding protein. Virus Res. 133:136148.28. Jones, S., and J. M. Thornton. 1995. Protein-protein interactions: a review ofprotein dimer structures. Prog. Biophys. Mol. Biol. 63:3165.29. Joseph, J. S., K. S. Saikatendu, V. Subramanian

231、, B. W. Neuman, A. Brooun,M. Griffith, K. Moy, M. K. Yadav, J. Velasquez, M. J. Buchmeier, R. C.Stevens, and P. Kuhn. 2006. Crystal structure of nonstructural protein 10from the severe acute respiratory syndrome coronavirus reveals a novel foldwith two zinc-binding motifs. J. Virol. 80:78947901.30.

232、Kamitani, W., K. Narayanan, C. Huang, K. Lokugamage, T. Ikegami, N. Ito,H. Kubo, and S. Makino. 2006. Severe acute respiratory syndrome corona-virus nsp1 protein suppresses host gene expression by promoting hostmRNA degradation. Proc. Natl. Acad. Sci. USA 103:1288512890.31. Kleiger, G., R. Grothe, P

233、. Mallick, and D. Eisenberg. 2002. GXXXG andAXXXA: common alpha-helical interaction motifs in proteins, particularlyin extremophiles. Biochemistry 41:59905997.32. Ksiazek, T. G., D. Erdman, C. S. Goldsmith, S. R. Zaki, T. Peret, S. Emery,S. X. Tong, C. Urbani, J. A. Comer, W. Lim, P. E. Rollin, S. F

234、. Dowell, A. E.Ling, C. D. Humphrey, W. J. Shieh, J. Guarner, C. D. Paddock, P. Rota, B.Fields, J. DeRisi, J. Y. Yang, N. Cox, J. M. Hughes, J. W. LeDuc, W. J.Bellini, L. J. Anderson, and S. W. Grp. 2003. A novel coronavirus associatedwith severe acute respiratory syndrome. N. Engl. J. Med. 348:1953

235、1966.33. Laskowski, R. A. 2001. PDBsum: summaries and analyses of PDB structures.Nucleic Acids Res. 29:221222.34. Laskowski, R. A., V. V. Chistyakov, and J. M. Thornton. 2005. PDBsummore: new summaries and analyses of the known 3D structures of proteinsand nucleic acids. Nucleic Acids Res. 33:D266D2

236、68.35. Laskowski, R. A., E. G. Hutchinson, A. D. Michie, A. C. Wallace, M. L.Jones, and J. M. Thornton. 1997. PDBsum: a Web-based database of sum-maries and analyses of all PDB structures. Trends Biochem. Sci. 22:488490.36. Lau, S. K., P. C. Woo, K. S. Li, Y. Huang, H. W. Tsoi, B. H. Wong, S. S.Wong

237、, S. Y. Leung, K. H. Chan, and K. Y. Yuen. 2005. Severe acute respi-ratory syndrome coronavirus-like virus in Chinese horseshoe bats. Proc. Natl.Acad. Sci. USA 102:1404014045.37. Law, A. H., D. C. Lee, B. K. Cheung, H. C. Yim, and A. S. Lau. 2007. Rolefor nonstructural protein 1 of severe acute resp

238、iratory syndrome coronavirusin chemokine dysregulation. J. Virol. 81:416422.38. Li, W., Z. Shi, M. Yu, W. Ren, C. Smith, J. H. Epstein, H. Wang, G. Crameri,Z. Hu, H. Zhang, J. Zhang, J. McEachern, H. Field, P. Daszak, B. T. Eaton,S. Zhang, and L. F. Wang. 2005. Bats are natural reservoirs of SARS-li

239、kecoronaviruses. Science 310:676679.39. Liu, P., J. Yang, X. Wu, and Z. F. Fu. 2004. Interactions amongst rabies virusnucleoprotein, phosphoprotein and genomic RNA in virus-infected andtransfected cells. J. Gen. Virol. 85:37253734.40. Lobley, A., L. Whitmore, and B. A. Wallace. 2002. DICHROWEB: an i

240、nter-active website for the analysis of protein secondary structure from circulardichroism spectra. Bioinformatics 18:211212.41. Luft, J. R., and G. T. Detitta. 1992. Hangman: a macromolecular hanging-drop vapor-diffusion technique. J. Appl. Crystallogr. 25:324325.42. Lundblad, J. R., M. Laurance, a

241、nd R. H. Goodman. 1996. Fluorescencepolarization analysis of protein-DNA and protein-protein interactions. Mol.Endocrinol. 10:607612.43. Marra, M. A., S. J. M. Jones, C. R. Astell, R. A. Holt, A. Brooks-Wilson,Y. S. N. Butterfield, J. Khattra, J. K. Asano, S. A. Barber, S. Y. Chan, A.Cloutier, S. M.

242、 Coughlin, D. Freeman, N. Girn, O. L. Griffin, S. R. Leach, M.Mayo, H. McDonald, S. B. Montgomery, P. K. Pandoh, A. S. Petrescu, A. G.Robertson, J. E. Schein, A. Siddiqui, D. E. Smailus, J. E. Stott, G. S. Yang,F. Plummer, A. Andonov, H. Artsob, N. Bastien, K. Bernard, T. F. Booth, D.Bowness, M. Czu

243、b, M. Drebot, L. Fernando, R. Flick, M. Garbutt, M. Gray,A. Grolla, S. Jones, H. Feldmann, A. Meyers, A. Kabani, Y. Li, S. Normand,U. Stroher, G. A. Tipples, S. Tyler, R. Vogrig, D. Ward, B. Watson, R. C.Brunham, M. Krajden, M. Petric, D. M. Skowronski, C. Upton, and R. L.Roper. 2003. The genome seq

244、uence of the SARS-associated coronavirus.Science 300:13991404.44. McPhillips, T. M., S. E. McPhillips, H. J. Chiu, A. E. Cohen, A. M. Deacon,P. J. Ellis, E. Garman, A. Gonzalez, N. K. Sauter, R. P. Phizackerley, S. M.Soltis, and P. Kuhn. 2002. Blu-Ice and the distributed control system: soft-ware fo

245、r data acquisition and instrument control at macromolecular crystal-lography beamlines. J. Synchrotron Radiat. 9:401406.45. Munter, L. M., P. Voigt, A. Harmeier, D. Kaden, K. E. Gottschalk, C. Weise,R. Pipkorn, M. Schaefer, D. Langosch, and G. Multhaup. 2007. GxxxGmotifs within the amyloid precursor

246、 protein transmembrane sequence arecritical for the etiology of A?42. EMBO J. 26:17021712.46. Niessing, D., S. Huttelmaier, D. Zenklusen, R. H. Singer, and S. K. Burley.2004. She2p is a novel RNA binding protein with a basic helical hairpinmotif. Cell 119:491502.47. Otwinowski, Z., and W. Minor. 199

247、7. Processing of X-ray diffraction datacollected in oscillation mode. Macromol. Crystallogr. A 276:307326.48. Peters, H., Y. Y. Kusov, S. Meyer, A. J. Benie, E. Bauml, M. Wolff, C.Rademacher, T. Peters, and V. Gauss-Muller. 2005. Hepatitis A virus pro-teinase 3C binding to viral RNA: correlation wit

248、h substrate binding andenzyme dimerization. Biochem. J. 385:363370.49. Ponnusamy, R., R. Moll, T. Weimar, J. R. Mesters, and R. Hilgenfeld. 2008.Variable oligomerization modes in coronavirus nonstructural protein 9. J.Mol. Biol. 383:10811096.50. Rota, P. A., M. S. Oberste, S. S. Monroe, W. A. Nix, R

249、. Campagnoli, J. P.Icenogle, S. Penaranda, B. Bankamp, K. Maher, M. H. Chen, S. X. Tong, A.Tamin, L. Lowe, M. Frace, J. L. DeRisi, Q. Chen, D. Wang, D. D. Erdman,T. C. T. Peret, C. Burns, T. G. Ksiazek, P. E. Rollin, A. Sanchez, S. Liffick,B. Holloway, J. Limor, K. McCaustland, M. Olsen-Rasmussen, R

250、. Fouchier,S. Gunther, A. D. M. E. Osterhaus, C. Drosten, M. A. Pallansch, L. J.Anderson, and W. J. Bellini. 2003. Characterization of a novel coronavirusassociated with severe acute respiratory syndrome. Science 300:13941399.VOL. 83, 2009SARS-CoV nsp9 DIMERIZATION3017 on March 7, 2015 by DAHLGREN M

251、EDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from 51. Schneider, D., and D. M. Engelman. 2004. Motifs of two small residues canassist but are not sufficient to mediate transmembrane helix interactions. J.Mol. Biol. 343:799804.52. Snijder, E. J., P. J. Bredenbeek, J. C. Dobbe, V. Thiel, J. Ziebuhr, L.

252、L. M.Poon, Y. Guan, M. Rozanov, W. J. M. Spaan, and A. E. Gorbalenya. 2003.Unique and conserved features of genome and proteome of SARS-corona-virus, an early split-off from the coronavirus group 2 lineage. J. Mol. Biol.331:9911004.53. Snijder, E. J., Y. van der Meer, J. Zevenhoven-Dobbe, J. J. Onde

253、rwater, J.van der Meulen, H. K. Koerten, and A. M. Mommaas. 2006. Ultrastructureand origin of membrane vesicles associated with the severe acute respiratorysyndrome coronavirus replication complex. J. Virol. 80:59275940.54. Sperry, S. M., L. Kazi, R. L. Graham, R. S. Baric, S. R. Weiss, and M. R.Den

254、ison. 2005. Single-amino-acid substitutions in open reading frame(ORF) 1b-nsp14 and ORF 2a proteins of the coronavirus mouse hepatitisvirus are attenuating in mice. J. Virol. 79:33913400.55. Su, D., Z. Lou, F. Sun, Y. Zhai, H. Yang, R. Zhang, A. Joachimiak, X. C.Zhang, M. Bartlam, and Z. Rao. 2006.

255、Dodecamer structure of severe acuterespiratory syndrome coronavirus nonstructural protein nsp10. J. Virol. 80:79027908.56. Sulistijo, E. S., and K. R. MacKenzie. 2006. Sequence dependence of BNIP3transmembrane domain dimerization implicates side chain hydrogen bond-ing and a tandem GxxxG motif in sp

256、ecific helix-helix interactions. J. Mol.Biol. 364:974990.57. Sutton, G., E. Fry, L. Carter, S. Sainsbury, T. Walter, J. Nettleship, N.Berrow, R. Owens, R. Gilbert, A. Davidson, S. Siddell, L. L. M. Poon, J.Diprose, D. Alderton, M. Walsh, J. M. Grimes, and D. I. Stuart. 2004. Thensp9 replicase protei

257、n of SARS-coronavirus, structure and functional in-sights. Structure 12:341353.58. Thiel, V., K. A. Ivanov, A. Putics, T. Hertzig, B. Schelle, S. Bayer, B. Weiss-brich, E. J. Snijder, H. Rabenau, H. W. Doerr, A. E. Gorbalenya, and J.Ziebuhr. 2003. Mechanisms and enzymes involved in SARS coronavirusg

258、enome expression. J. Gen. Virol. 84:23052315.59. van Marle, G., L. C. van Dinten, W. J. Spaan, W. Luytjes, and E. J. Snijder.1999. Characterization of an equine arteritis virus replicase mutant defectivein subgenomic mRNA synthesis. J. Virol. 73:52745281.60. von Brunn, A., C. Teepe, J. C. Simpson, R

259、. Pepperkok, C. C. Friedel, R.Zimmer, R. Roberts, R. Baric, and J. Haas. 2007. Analysis of intraviralprotein-protein interactions of the SARS coronavirus ORFeome. PLoSONE 2:e459.61. Wang, W., K. Riedel, P. Lynch, C. Y. Chien, G. T. Montelione, and R. M.Krug. 1999. RNA binding by the novel helical do

260、main of the influenza virusNS1 protein requires its dimer structure and a small number of specific basicamino acids. RNA 5:195205.62. Whitmore, L., and B. A. Wallace. 2004. DICHROWEB, an online server forprotein secondary structure analyses from circular dichroism spectroscopicdata. Nucleic Acids Re

261、s. 32:W668W673.63. Wilkins, M. R., E. Gasteiger, A. Bairoch, J. C. Sanchez, K. L. Williams, R. D.Appel, and D. F. Hochstrasser. 1999. Protein identification and analysis toolsin the ExPASy server. Methods Mol. Biol. 112:531552.64. Yount, B., K. M. Curtis, and R. S. Baric. 2000. Strategy for systemat

262、icassembly of large RNA and DNA genomes: transmissible gastroenteritisvirus model. J. Virol. 74:1060010611.65. Yount, B., K. M. Curtis, E. A. Fritz, L. E. Hensley, P. B. Jahrling, E. Prentice,M. R. Denison, T. W. Geisbert, and R. S. Baric. 2003. Reverse genetics witha full-length infectious cDNA of

263、severe acute respiratory syndrome corona-virus. Proc. Natl. Acad. Sci. USA 100:1299513000.66. Yount, B., M. R. Denison, S. R. Weiss, and R. S. Baric. 2002. Systematicassembly of a full-length infectious cDNA of mouse hepatitis virus strainA59. J. Virol. 76:1106511078.67. Zhai, Y., F. Sun, X. Li, H.

264、Pang, X. Xu, M. Bartlam, and Z. Rao. 2005.Insights into SARS-CoV transcription and replication from the structure ofthe nsp7-nsp8 hexadecamer. Nat. Struct. Mol. Biol. 12:980986.68. Ziebuhr, J., E. J. Snijder, and A. E. Gorbalenya. 2000. Virus-encoded pro-teinases and proteolytic processing in the Ni

265、dovirales. J. General Virology81:853879.69. Zust, R., T. B. Miller, S. J. Goebel, V. Thiel, and P. S. Masters. 2008. Geneticinteractions between an essential 3? cis-acting RNA pseudoknot, replicasegene products, and the extreme 3? end of the mouse coronavirus genome.J. Virol. 82:12141228.3018MIKNIS ET AL.J. VIROL. on March 7, 2015 by DAHLGREN MEDICAL LIBRARYhttp:/jvi.asm.org/Downloaded from