Wang Xiong Yu Stacy Overman Masamichi Tsuboi George J Thomas Jr and Edward H Egelman Department of Biochemistry and Molecular Genetics University of Virginia Box 800733 Charlottesville VA 229080733 USA School of Biological Sciences University of ID: 88618
Download Pdf The PPT/PDF document "OMMUNICATION The Structure of a Filament..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
noiseratiosufficientlylargethatlayer-linesmaybeextractedfromsingleparticles.Additionally,becauseuniformsymmetryandstructureareimposedonthetargetedfilament,heterogeneousregionsmaybeaveraged.Wehaveusedtheiterativehelicalrealspacereconstruction(IHRSR)methodtoreconstructfd.Datawereobtainedfromnativespecimenspreparedunderconditionsthatminimizeaggregation.Using84,310filamentseg-ments(eachoflength240Å)extractedfromfilamentsinice,wefailedtoachieveconvergencetoacommonstructurefromdifferentstartingsymmetries.Thissuggestedstructuralpolymorphism.Byusingtwodifferentinitialreconstructionsandemployingmul-tipleiterationsofthealgorithm,wewereabletoclassifythesegmentsintothreegroups,twoofwhich=25,440,=16,367)generatedreconstructionsthatconvergedindependentofthestartingsymme-try(Figure2).Thethirdgroup(=42,503)failedtoconverge,andwewereunabletosortthisgroupintomorehomogeneoussubsets.ThereconstructionsgeneratedfromareshowninFigure3and(b),respectively,anditcanbeseenthattheyaredifferent.WedeterminedtheresolutionofthesereconstructionstobeFigure3(g)).Whilethesymmetryofthesetwostructuresisnearlyindis-tinguishable,theydifferstructurally,andthesestructuraldifferencesexplainwhytheIHRSRmethoddidnotachieveconvergenceusingthewholepopulationofimages.Thetworeconstruc-tionshaveaFouriershellcorrelation(FSC)of0.5ataresolutionof1/(14Å).Thisdoesnotmeanthatthetworeconstructionsareindistinguishableat14Åresolution,butratherthatwithinthisresolutionshellthetwohaveacorrelationofonly0.5.Wehaveexcludedthepossibilitythatthetwodifferentstructuralstatesareanartifactarisingfromdifferentdefocusvaluesorsomeotheraspectofthemicroscopy,asthetwostatesarefoundonthesamemicrographs.Further,wehaveobservedthatthetwodifferentstatescanbefoundinthesamephageparticle,excludingthepossibilitythatanentirephageisentirelyinonestateortheother.Aswithallstruc-turesgeneratedfromEMimages,thereisanenan-tiomorphicambiguityinourreconstructions.Thatis,eachreconstructioncanbemirroredtogenerateasecondreconstruction,andbothenantiomorphswillproducethesamesetofprojections.Wehaveexc-ludedthepossibilitythatthetwostructuresaresim-plyenantiomorphsofeachother.Becauseofthesymmetryofthephage,thereisarotationof36.0+fromonesubunittoasubunit17.4Åaboveit,andthiscanbeexpressedasarotationof36.0°.Thatis,+34.6°isthesameas37.4°.Becauseoftheenan-tiomorphicambiguity,areconstructionwithasym-metryinvolvinga+34.6°(or37.4°)rotationcannotbedistinguishedwithourdatafromthemirrorimageofthisreconstruction,whichwillhavearotationof34.6°(or+37.4°).WehavereliedupontheX-raydiffractiondata,whichsuggestedthatthesubunitsareslewedinaright-handedmanner,tochoosetheenantiomorphsshown,whichcorrespondtoarota-tionof37.4°betweenasubunitandtheoneaboveit.Numerousstudieshavesuggestedthatthesub-unitisquitemalleable,1,14whichisconsistentwiththecapsidpolymorphismevidenthere.Infact,aminimizedfd(M13)coatproteininwhichallbutnineofthe50residuesweremutatedtoAlawasfoundtoco-assembleefficientlywiththewild-typeproteintoproduceinfectiousparticles.conformationalheterogeneityofwild-typefdwas Figure1.Electronmicrographsofnegativelystainedfd(a)usedtoobtainthepersistencelengthbyfittingthecontourlengthbetweentwopointsasafunctionofthesquareoftheirseparation,,(b)accordingtotherelationship::(2P/L)(1exp())].Thebestfit(redline)yields=1.02(±0.06)m,wheretheerrorwasdeterminedbythenon-linearcurve-fittingalgorithmintheprogramOrigin7.5(OriginLabCorpora-tion).Frompriorresultsusingbothrealdataandweexpectthatthisisanunderestimateofthetrueerror,butexpecttheerrortobelessthan0.5m.(c)Imagesinice(inset)werecollectedonaTecnai20fieldemissiongunmicroscopeat200keV.The23ÅpitchhelixofTMVwasusedtodeterminethemagnification.Defocusvaluesof1.4mfrom67micrographsofwild-typefdweredeterminedfromthecarbonfilmsupportingthephageoradjoiningholesinwhichthephageswereimaged.Thecontrasttransferfunctionwascorrectedbyphase-flippingbeforesubsequentimageprocessing.Thespacebarrepresents(a)0.9mand(c)0.4 TheStructureofaFilamentousBacteriophage reportedtobereducedintheY21Mmutant,wealsoexaminedthemutant.(Themutationoccursinaregionofthesubunitthatputativelylinksamphi-pathicandhydrophobic-helices.)WefoundthatallsegmentssampledfromY21Mcouldbesortedroughlyequallyintothesametwo(Figure3(d)and(e))relativelyhomogeneousgroups(observedforwild-typefd,butwithnomajorheterogeneouspopulation(0).Thesomewhatlowerresolution(9Å)achievedfortheY21M Figure2.ThefailureoftheIHRSRmethodtoconvergetoasinglestructureandsymmetryfromdifferentinitialvaluesisindicativeofheterogeneity.(a)Reconstructionswereinitiatedusingasolidcylinderasastartingmodelwithdifferentstartingsymmetries.(b)and(c)Theresultingreconstructionsaredifferent.Byusingthetwodifferentinitialreconstructionsandemployingmultipleiterationsofthealgorithm,wewereabletoclassifythesegmentsintothreegroups,twoofwhich((d)and(e))showconvergenceindependentofthestartingsymmetry.TheaxialrisepersubunitisalsofreetochangeintheIHRSRapproachand,usinginitialvaluesof15.8Åand17.4Å,wehavealwaysseenconvergenceto17.4Å.Onegroup((d),=25,440)generatesthereconstructionshowninFigure3(a),whilethesecondgroup((e),=16,367)generatesthereconstructionshowninFigure3(b),whichissignificantlydifferentfromthatinFigure3 Figure3.(a)and(b)Surfacesfromtwodifferentreconstructionsofwild-typefd.ThesubunitNterminusistowardthetopandonthephageexterior.In(b),severalsubunitsarenumberedtoillustratethe5-foldrotationalsymmetryofthestructure(e.g+6arerelatedbya37.4°rotationand17.4Åaxialrise).An-helicalsubunitwasbuiltintothereconstructionbystartingwiththeX-raymodel andallowingadditionalbendsbetweenresidues15and16,25and26,and35and36.The-helixshownincludesresidues648.Residues15havebeendescribedasdisorderedandarepresumedtocorrespondtothesmalladditionaldensitythatisunaccountedforbytheatomicmodel(redarrow).(c)Acutawayviewofthelumen,obtainedbyremovingthefronthalfofthereconstruction,wheresubunitC-terminalendsformaright-handedfive-starthelicalgroove.ThecoordinatesofourmodelareavailablefromtheProteinDataBank.(d)and(e)ThecorrespondingreconstructionsfromtheY21Mmutant.(f)Thecomparisonofdifferentsubunitmodelsshowsourfit(b,green),themostrecentlyrefinedX-raymodel( ,red)andtheNMRmodel( ,cyan).Thetwoviewsshownarerelatedbya90°rotationaboutthehelicalaxis(black).TheIHRSRmethodalsoallowsfortheestimationofresolutionbycomparingtrulyindependentreconstructions,eachcontainingtheentiredataset,butgeneratedfromdifferentstartingpoints,and(g)avalueof8Åwasfoundbythisapproachforthereconstructionshownin(a).Usingthemoreconventionalapproachofdividingadatasetintohalvesafteralignmenttoacommonreference,asimilarvalueof8Åwasalsofound.NearlyidenticalFouriershellcorrelation(FSC)curveswereobtainedforthereconstructionin(b),showingthatthedifferencebetweenthetwostatesisnotduetoanydifferenceinresolution. TheStructureofaFilamentousBacteriophage reconstructionsisattributedtothesmallernumberofsegmentssampled.Asinglecontinuous-helicalsubunitdoesnotsatisfactorilyfitintothereconstructionofthatinFigure3(a),consistentwithsuggesteddiscontinuityintheIncontrast,acontinuous(residues648)flankedbyterminalresidues(15,49and50)inunspecifiedconformationsreadilyfitstheFigure3(b)reconstructioninaccordwithreporteddisorderoftheNterminus.Figure3(b)modelisconsistentalsowiththeexperimentalobservationthatTyr21andTyr24areburiedatahydrophobicintersubunitinterface.Althoughatomicdetailsofthecapsidsubunitarenotresolved,experimentallydeterminedconstraintsforside-chains,includingtheorientationsofTyr21,Tyr24andTrp26,availableforfuturetestingofhigherresolutionreconstructions.Thepresentresultsprovideanovelviewofpolymorphisminthenativestateofthefdvirion. Figure3legendonpreviouspage TheStructureofaFilamentousBacteriophage Weobserveanaxialrisepersubunitof17.4Åincryoelectronmicroscopyimagesoffullyhydratedfd(usingthe23Åpitchhelixoftobaccomosaicvirus(TMV)asamagnificationstandard).PreviousX-raydiffractionstudiesofdriedfibersreportedariseof16.0Å.Theshrinkageinfibersusedfordiffractionisconsistentwithobservationsthatboththeaxialriseandtheinterfilamentspacingchangeasafunctionofrelativehumidityinfdandinotherfilamentousphages.Figure3(f)comparesthecapsidsubunitstructuresproposedonthebasisofX-rayfiberdiffraction(red,PDBcode 2C0WandNMR(cyan,PDBcode withourcontinuous-helixmodel(green).Neitherthe northe modelfitsintoeitherreconstruction(Figure4).SinceboththeNMRandX-rayfiberdiffractionmodelswerebasedupontheY21Mmutant,itdoesnotseempossiblethatthosetwomodelsmightfitthedatafromthewild-typefdthatwefailedtoanalyzeduetoapparentheterogeneity(alloftheY21MsegmentswereclassifiedintooneofthetwostructuralstatesshowninFigure3(d)and(e)).Theaveragesubunit-helixtiltangle()inourmodel(Figure3(b))is21°,whichisclosetotherange=16±4)foundinorientedfibers.Thesmall Figure4.Cryoelectronmicroscopyreconstructionsofthediscontinuous-helixstructureshowninFigure3viewedfrom(a)theexteriorsurfaceand(c)thelumen,illustratingtheinabilityofatomicmodelsfromX-rayfiberdiffraction( ,red)andNMRspectroscopy( ,cyan)tofitthedensitydistributionaccurately.(b)and(d)Correspondingviewsforthecontinuous-helixreconstructionthediscontinuous-helixstructureshowninFigure3Arrowsindicatewheretheexistingmodelsarenotcontainedwithinthedensityenvelope. TheStructureofaFilamentousBacteriophage W.(2004).Structureoftheacrosomalbundle.Nature,10424.VanLoock,M.S.,Yu,X.,Yang,S.,Lai,A.L.,Low,C.,Campbell,M.J.&Egelman,E.H.(2003).ATP-mediatedconformationalchangesintheRecAfila-Structure(Camb.),18725.Wu,Y.,He,Y.,Moya,I.A.,Qian,X.&Luo,Y.(2004).CrystalstructureofarchaealrecombinaseRADA:asnapshotofitsextendedconformation.Mol.Cell26.Conway,A.B.,Lynch,T.W.,Zhang,Y.,Fortin,G.S.,Fung,C.W.,Symington,L.S.&Rice,P.A.(2004).CrystalstructureofaRad51filament.NatureStruct.Mol.Biol.,79127.Wang,Y.A.,Yu,X.,Yip,C.K.,Strynadka,N.C.&Egelman,E.H.(2006).StructuralpolymorphisminbacterialEspAfilamentsrevealedbycryo-EMandanimprovedapproachtohelicalreconstruction.ture,inthepress.28.Rivetti,C.,Guthold,M.&Bustamante,C.(1996).ScanningforcemicroscopyofDNAdepositedontomica:equilibrationversuskinetictrappingstudiedbystatisticalpolymerchainanalysis.J.Mol.Biol.29.Orlova,A.&Egelman,E.H.(1993).Aconformationalchangeintheactinsubunitcanchangetheflexibilityoftheactinfilament.J.Mol.Biol.,33430.Yang,S.,Yu,X.,Galkin,V.E.&Egelman,E.H.(2003).Issuesofresolutionandpolymorphisminsingle-particlereconstruction.J.Struct.Biol.31.Galkin,V.E.,Orlova,A.,Fattoum,A.,Walsh,M.P.&Egelman,E.H.(2006).TheCH-domainofcalponindoesnotdeterminethemodesofcalponinbindingtoJ.Mol.Biol.,47832.Trachtenberg,S.,Galkin,V.E.&Egelman,E.H.(2005).RefiningthestructureoftheHalobacteriumsalinarumflagellarfilamentusingtheiterativehelicalrealspacereconstructionmethod:insightsintopolymorphism.J.Mol.Biol.,665EditedbyW.Baumeister(Received10April2006;receivedinrevisedform26May2006;accepted12June2006)Availableonline30June2006 TheStructureofaFilamentousBacteriophage