The architecture of the SARS-CoV-2 RNA genome inside virion

SARS-CoV-2 carries one of the largest RNA genomes (~30 kilobases, kb) among all RNA virus families and encodes about 29 proteins. Since its ... Skiptomaincontent Thankyouforvisitingnature.com.YouareusingabrowserversionwithlimitedsupportforCSS.Toobtain thebestexperience,werecommendyouuseamoreuptodatebrowser(orturnoffcompatibilitymodein InternetExplorer).Inthemeantime,toensurecontinuedsupport,wearedisplayingthesitewithoutstyles andJavaScript. Advertisement nature naturecommunications articles article ThearchitectureoftheSARS-CoV-2RNAgenomeinsidevirion DownloadPDF Subjects SARS-CoV-2SequencingVirusstructures AbstractSARS-CoV-2carriesthelargestsingle-strandedRNAgenomeandisthecausalpathogenoftheongoingCOVID-19pandemic.HowtheSARS-CoV-2RNAgenomeisfoldedinthevirionremainsunknown.Tofilltheknowledgegapandfacilitatestructure-baseddrugdevelopment,wedevelopavirionRNAinsituconformationsequencingtechnology,namedvRIC-seq,forprobingviralRNAgenomestructureunbiasedly.UsingvRIC-seqdata,wereconstructthetertiarystructureoftheSARS-CoV-2genomeandrevealasurprisingly“unentangledglobule”conformation.Weuncovermanylong-rangeduplexesandhigher-orderjunctions,bothofwhichareunderpurifyingselectionsandcontributetothesequentialpackageoftheSARS-CoV-2genome.Unexpectedly,theD614Gandtheothertwoaccompanyingmutationsmayremodelduplexesintomorestableforms.Lastly,thestructure-guideddesignofpotentsmallinterferingRNAscanobliteratetheSARS-CoV-2inVerocells.Overall,ourworkprovidesaframeworkforstudyingthegenomestructure,function,anddynamicsofemergingdeadlyRNAviruses. DownloadPDF IntroductionSevereacuterespiratorysyndromecoronavirus2(SARS-CoV-2)isthecausalpathogenofcoronavirusdisease2019(COVID-19)1,2,3.Asasingle-strandedandpositive-senseRNAvirus,SARS-CoV-2,togetherwithSARS-CoVandMiddleEastrespiratorysyndromecoronavirus(MERS-CoV),allbelongtotheCoronaviridaefamily4.SARS-CoV-2carriesoneofthelargestRNAgenomes(~30kilobases,kb)amongallRNAvirusfamiliesandencodesabout29proteins1,3,5,6,7.SinceitsoutbreakinlateDecember2019,SARS-CoV-2hasinfectedtensofmillionsofpeopleandcausedoveronemilliondeathsworldwide(https://covid19.who.int/).AlthoughglobaleffortsandresourceshavebeenredirectedtofightagainstSARS-CoV-2,therearenoeffectiveantiviralmedicinesavailableyet.ConsideringtheRNAnatureofSARS-CoV-2,RNA-basedtherapeuticssuchassmallinterferenceRNAs(siRNAs)orantisenseoligos(ASOs)areemergingaspotentagentstocleavetheviralRNAgenomeininfectedhostcells.BecauseRNAstructurecansignificantlyinfluencetheefficacyofsiRNAsandASOs8,9,decipheringthe3DstructureofSARS-CoV-2becomesanurgentneedpriortoRNA-baseddrugdevelopment.RNAstructuresarewidelyrecognizedascriticalmodulatorsinregulatingtranscription,translation,andreplicationsofcoronavirusandotherRNAviruses10,11,12,13,14,15,16,17.Atthisfrontier,manyeffortshavebeendevotedtostudythestructureofSARS-CoV-2.EventhoughCryo-electronmicroscopyand3DelectrontomographyarepowerfulindelineatingtheglobalarchitecturesofSARS-CoV-2,theentireRNAgenomeinsidevirionsremainsunrevealed18,19,20.Besidesthesephysicalapproaches,severalchemical-basedhigh-throughputsequencingmethods,suchasinvivoclickselective2′-hydroxylacylationandprofilingexperiment(icSHAPE)anddimethylsulfatemutationalprofilingwithsequencing(DMS-MaPseq),havebeenrecentlyappliedtoprobesingle-strandedregionsofSARS-CoV-2ininfectedcellsorinvitro21,22,23,24.Themappedsingle-strandedinformationcouldbefurtherusedasrestraintstopredictthebase-pairedregionswithin500 nt25.YetthecurrentsecondarystructuralmodelofSARS-CoV-2mightbeincompletesinceitmissedinformationoflong-rangeduplexes,whichareprevalentandvitalforcompletingthelifecyclesofpositive-strandRNAviruses26,27.Asasignificantadvance,apsoralen-basedmethodcalledcross-linkingofmatchedRNAsanddeepsequencing(COMRADES)recentlyidentifiedmanylong-rangeRNAduplexesofSARS-CoV-2insidecells,furtherhighlightingtheimportanceofRNAduplexesinmaintainingSARS-CoV-2fitness28,29.Significantlylaggingbehindthosein-cellstructuralstudies,howthe30 kbgenomeofSARS-CoV-2isorganizedandarrangedinvirionsremainsunclear.WerecentlydevelopedanRNAinsituconformationsequencingtechnology,namedRIC-seq,forunbiasedmappingofRNA-RNAspatialinteractionsinlivingcells30.RIC-sequtilizesapCp-biotintolabelproximallyinteractingchimericRNAsandhigh-throughputsequencingtoretrievetheirspatialproximityinformation.WedemonstratedthatRIC-seqcouldsuccessfullydetectshort-andlong-rangeduplexes,multiple-wayjunctions,andloop-loopcontactswithoutbasepairingpotentials.ThesemeritsmakeRIC-seqmorecompetenttodecipherthe3DstructureoftheSARS-CoV-2RNAgenome.ButSARS-CoV-2virionsaretypically80 nmindiameterandcan’tbepelleteddownashumancellsbystandardcentrifugation1.ThisfeaturemakesvirionsnotcompliantwithourcurrentRIC-seqprotocolthatincludesextensivewashingandstandardcentrifugationateveryenzymaticstep30.Toovercomethemajorchallenges,wedesignedawaytotrapSARS-CoV-2virionsonConcanavalinA(ConA)beadsthatbindspecificallytoglycoproteinspresentonthesurfaceofvirions.Byoptimizingvirioncapture,crosslinking,andenzymaticconditions,wefurtherdevelopedavirionRNAinsituconformationsequencingtechnology,namedvRIC-seq,forglobalmappingofviralRNAgenomestructuresinintactvirions.Inthisstudy,wesuccessfullyappliedthevRIC-seqtechnologytoprobetheSARS-CoV-2RNAgenomestructureinintactvirions.WereconstructedamodelofthesurprisinglycompactyetunentangledtertiarystructureoftheSARS-CoV-2RNAgenome.Atthesecondarystructurelevel,wefoundthattheoverallstructureofSARS-CoV-2ismorecompactedinvirionthanthatinthehostcell,andlong-rangeRNAduplexesareprevalentlypresent.Thecovariantanalysisrevealedthatmanylong-rangeRNA-RNAinteractionsareunderpurifyingselectionandmightcontributetothe3DpackageoftheSARS-CoV-2RNAgenome.Finally,weuncoveredseveralhighlyaccessiblesingle-strandedregionsinSARS-CoV-2forefficientviralRNAcleavageinVerocellsbyusingsiRNAs.ResultsOverviewofvRIC-seqtechnologyLiketheothercoronaviruses,theenvelopeofSARS-CoV-2containstwomajorglycoproteinsspike(S)andmembrane(M)31,32.Accordingtothisfeature,weusedmagneticbeadscoatedwithConA,aplantlectinthatcanspecificallybindglycoproteins,tocapturetheSARS-CoV-2virionspreparedfromthesupernatantsofinfectedVerocells(see“Methods”).Aftercapturingthevirionsonbeads,formaldehydewasfurtherappliedtocrosslinkthenucleocapsid(N)protein-mediatedproximalRNA-RNAinteractions,aswellasConAandthesurfaceglycoproteins(Fig. 1a).Next,thevirionswerepermeabilizedandtreatedwithmicrococcalnuclease(MNase)todigestsingle-anddouble-strandedRNAsprotrudedfromthenearbyproteincomplexes.Subsequently,alltheproximalRNAswere3′end-labeledwithpCp-biotinandligatedtogetherbyT4RNAligase.Lastly,theresultingchimericRNAsmarkedwithpCpatthejuncturewereenrichedandconvertedintolibrariesforpaired-endsequencing(about260 bp,Fig. 1aandSupplementaryFig. 1a).Fig.1:OverviewandevaluationofvRIC-seqtechnology.aSchemeofvRIC-seqtechnology.ConcanavalinA(ConA)beadswereusedtocapturethevirionfordiverseenzymetreatmentsinsubsequentsteps.bTheproportionsofchimericreadsmappedtoSARS-CoV-2.cScatterplotsshowingthecorrelationbetweentwobiologicalreplicatesforthenumberofchimericreads(interactionstrength).R,Pearsoncorrelationcoefficient.dCircosplotshowingthedistributionofchimericreadsalongtheSARS-CoV-2genome.Theinnerredcirclestandsforthefractionsofadenineoruracilwithin100 ntwindows,andtheouterbluecircleshowsthecoverageofchimericreads.FSEframe-shiftingelement.e,fvRIC-seqconfirmedknowncoronavirusRNAstructuresinthe5′UTR(1–480 nt,e)and3′UTR(29,546–29,870 nt,f)oftheSARS-CoV-2RNAgenome.Connectionscoresshownindifferentcolorswereusedforassessingthebase-pairingprobability.Thedashedlinesillustratedthepseudoknot.FullsizeimageWeobtained~24.6millionuniquereadsforeachreplicateand~3.4millionchimericreadsresultingfromdifferentRNAfragments.UsingtheRIC-seqdataanalysisworkflowestablishedearlier30,wefoundthat97.3%ofthechimericreadscouldbealignedtotheSARS-CoV-2genome,and2.3%weremappedtohostRNA(Fig. 1b).ThetraceamountsofhostRNAreadsmightbederivedfromdetachedVerocellsatthevirioncollectionstage.Notably,~0.4%ofthechimericreadswerevirus-to-hostRNAinteractions(Fig. 1b),reflectingtherandomligationratesbetweenvirionsandVerocellsorrepresentinginteractionsbetweenviralandhostRNAsthathappenedinsidethedetachedVerocells.Besides,thevirus-to-hostinteractionsalsoraisedthepossibilitythatsomehostRNAsmightbepackagedintothevirion,justliketRNAintheHIVvirionsforitsreplication33.However,suchapossibilitywasexcludedbecauseweobservedarandomvirus-to-hostRNAinteractionpatternacrossallthegreenmonkeychromosomes(SupplementaryFig. 1b).ThepCp-biotinlabelingandselectionweresuccessfulbecause~85%oftheadditionalnucleotidesatchimericjunctionswerecytosine(SupplementaryFig. 1c).vRIC-seqishighlyreproducibleonchimericreadscoverage(R = 0.999)andinteractionstrength(R = 0.995)ofpairwisesites(Fig. 1candSupplementaryFig. 1d).Moreover,wenotedthat99.7%oftheSARS-CoV-2genomewascoveredatleast2500×bychimericreads(SupplementaryFig. 1e),andtheremaining0.3%coveredby<2500×wasmainlylocatedatthestem-loop1of5′UTRandthepoly(A)regionatthe3′terminal.Next,wealignedthepairwiseinteractingRNAfragmentstotheSARS-CoV-2genome,andsuchanalysisrevealedmanyregionsthatarepreferablycutbyMNaseandsubsequentlylabeledwithpCp-biotin(Fig. 1d).Wealsonoticedthatthefirst3500nucleotides(nt)ofORF1bhadarelativelowervRIC-seqcoverage,andnucleotidecontentdidn’tcontributetothisdifference(Fig. 1d).RecapitulateknownstructuresoftheSARS-CoV-2genomeWefirstdeterminedtovalidatetheviralRNAspatialinteractionsrevealedbyvRIC-seq.Forthispurpose,wededucedseveralconservedRNAduplexesincoronavirususingvRIC-seqdataandcompareditwithrecentlyproposedSARS-CoV-2secondarystructuremodels13,21,22,23,24,34,35,36.Asexpected,vRIC-seqfaithfullyrecapitulatedallofthestem-loopsinthe5′UTR(1–395 nt)ofSARS-CoV-2exceptstem-loop1(SL1),whichis~40 ntinlengthandcan’tbepurifiedbyAMPureXPbeads(Fig. 1e).Inaddition,the3′UTRofSARS-CoV-2containsseveralconservedstructuralelementsknowntofunctionallyimpactviralRNAsynthesisandtranslation,includingthestem-loopII-likemotif(s2m),hyper-variableregion(HVR),andmutuallyexclusivebulgedstem-loop(BSL)orpseudoknot(PK)36.WefoundthatvRIC-seqsuccessfullycapturedthecanonicals2mandHVRstructures(Fig. 1f).However,incontrasttotheprevioustheoreticalmodel36,ourdatapreferentiallysupportedadoublehairpinconformationforBSLandP2stem,ratherthanthePKconformation(Fig. 1f).TheseresultsdemonstratethatvRIC-seqcanfaithfullyprobeRNAspatialinteractionsinthevirion,andsupportthatthisproximityinformationcanbeusedforstructuremodeling.TopologicalorganizationoftheSARS-CoV-2genomeAftervalidatingthevRIC-seqdata,wenextinvestigatedthefeaturesofSARS-CoV-2genomeorganizationinthevirion.Tothisend,wefirstcalculatedthespanningdistanceofpairwiseinteractingRNAfragments.Approximately90.6%ofthepairwiseinteractionshappenedwithin100 nt,whereas~6.3%oftheinteractionsspannedmorethan600 nt,andthreesharppeakswith810,1360,and2090 ntseparately,wereobserved(Fig. 2a).Ofnote,thoselong-distanceinteractionswerenotcausedbythediscontinuoustranscriptionofSARS-CoV-2duringnegative-strandsynthesis6,becauseweobservedaclearenrichmentofpCpatthechimericjunctions(seeredlines,SupplementaryFig. 2a,b).Fig.2:GlobalviewofSARS-CoV-2genomeorganization.aSpanningdistanceofpairwiseinteractingRNAs.P1,P2,andP3markthreepeakscorrespondingtochimericinteractionsspanning810,1360,and2090nucleotides.bRNAinteractionmapoftheSARS-CoV-2genome.TheblacktrianglesrepresentRNAtopologicaldomains.Theframe-shiftingelement(FSE),thetranscription-regulatorysequenceintheleader(TRS-L),andthebody(TRS-B)aremarkedasblacklines.cTheglobalconfigurationoftheSARS-CoV-2RNAgenomeinvirions,modeledbytheminiMDSsoftware.The30 kbRNAgenomeofSARS-CoV-2ispresentedasarope,andeachcodingregionandUTRaremarkedwithdifferentcolors.ThevRIC-seqdetectedRNAcontactfrequencieswereusedforthemodeling.Thesolidredlinesrepresentchimericsignalsthatsupportthelocalinteractions,whereasthedashedredlinesdepictlong-rangeinteractions.Thesamepicturerotatedin180degreesisshownatthebottom.FullsizeimageFollowingourpreviousapproach,wedividedtheSARS-CoV-2genomeinto10-ntwindowsandconstructedanRNAinteractionmap(Fig. 2b).Usingthemap,weidentified254clusteredinteractionspositionedperpendiculartothediagonal,suggestingthewidespreadoccurrenceoflocalduplexesintheSARS-CoV-2genome(Fig. 2b).Surprisingly,wealsonoticed77long-rangeinteractionsthatweresequentiallydistributedandcoveredalmosttheentiregenome(Fig. 2b).Similartotheorganizationofhumanprimarytranscripts30,weobserved49RNAtopologicaldomainswithamedianlengthof630 ntintheSARS-CoV-2genome(Fig. 2b,SupplementaryFig. 2c,andSupplementaryData 1).Thelargestdomainislocatedbeforethe3′UTRandcontainssequencesencodingORF7a,ORF7b,ORF8,andtheNprotein(Fig. 2b).LikeRIC-seqtechnology30,vRIC-seqcanalsocapturebase-pairedRNAduplexesandprotein-mediatedindirectRNAcontacts.Toexaminethebase-pairingprobabilitiesfortheobservedlong-rangeinteractionswithin2.5 kb,wecalculatedtheminimumfreeenergy(MFE)forpairwiseinteractingfragmentsthatspannedover400 nt.WeobservedsignificantlylowerMFEvaluesthanrandomlyshuffledsequenceswiththesamenucleotidecontent(P = 1.62e−10,SupplementaryFig. 2d),indicatingthatthoselong-rangeRNA–RNAinteractionsmaybedirectlybase-paired.Notably,wealsouncovered62pairwiseinteractingRNAfragmentsthatcouldspanover2.5 kb(SupplementaryFig. 2eandSupplementaryData 2).69.4%(43/62)ofthoselong-rangeinteractionsweresupportedbymoreCOMRADES29readsthantherandomlyselectedgenomic-span-matchedcontrols(SupplementaryFig. 2f).However,thoselong-rangeduplexesseemednotthepreferredconformationinvirionsbecausetheindividualRNAfragmentsshowed12-foldstrongerinteractionswiththeirlocalpartnersthandistalpartners(SupplementaryFig. 2g).Forexample,long-rangeinteractionsbetween1180–1190 ntand29,343–29,354 nt,bothshowedstrongerlocalinteractions(SupplementaryFig. 2h–j).ThesedatasuggestthatsomealternativetopologyoftheSARS-CoV-2genomemightbepresentinthevirion.3DglobuleconfigurationoftheSARS-CoV-2genomeinvirionsPreviousmicroscopystudiesrevealedasphericalshapewithameandiameterof~80 nmforCoVandSARS-CoV-2viralparticles1,37,38.Toexplorehowthe30 kbRNAgenomeofSARS-CoV-2isfoldedinthetinyvirion,weutilizedthecontactfrequenciesdataofdifferentRNAfragmentsasconstraintstomodeltheglobalconformationoftheSARS-CoV-2genome.Pursuingthis,weadoptedawidelyusedminiMDS(multidimensionalscaling)approachtoinferthe3DstructureofSARS-CoV-239.Ourmodelingrevealeda3DglobuleconformationoftheRNAgenome(Fig. 2candSupplementaryVideo 1).Notably,thethread-likegenomewasunentangled,appearingtohaveamildlyhelicalconformation.Moreover,differentsegmentsoftheviralgenomeseemedtooccupyseparateterritories(seedifferentcolors,Fig. 2c),forminganorganizationapparentlysimilartoknownstructuresofmammaliangenomes40.Ofnote,this“globule”andknot-freeconfigurationmightreflectthearrangementoftheNprotein,whichhasbeenshowntobindtheCoVgenomeandinteractedwithMproteinviaitsC-terminaldomainintheinteriorofthelipidmembraneofvirions18.Wenexttriedtousetheinsilicofoldedorin-cellstructuremodeltosimulatethe3DconfigurationoftheSARS-CoV-2genome28,36.Wefoundthattheinsilicomodelcontainsonly3196base-pairswithin120 ntandcovered~28%oftheSARS-CoV-2genome36.Thelackoflong-rangeRNA-RNAinteractionspreventsusfromsimulatingthe3DconfigurationoftheSARS-CoV-2genomeusingtheinsilicomodel.ForviralRNAduplexesrevealedbyCOMRADES28,76.9%showedmultiplealternativeRNA-RNAinteractionsinVerocells(SupplementaryFig. 2k).Bycontrast,vRIC-seqrevealedthat88.9%oftheSARS-CoV-2RNAduplexesinvirionsshowedonedominantinteraction(SupplementaryFig. 2k).Moreover,theShannonentropy(higherShannonentropyrepresentsmorealternativeinteractions)foreach10-ntgenomicwindowwas6.08inhostcellsbyCOMRADESand2.35invirionbyvRIC-seq(SupplementaryFig. 2l).ThehigherstructuralflexibilityandheterogeneityofSARS-CoV-2genomicRNAsinsidecellsblockedusforfurthersimulation.TheseresultscollectivelyhighlightthevaluesofvRIC-seqdatainmodelingviralRNAgenomestructureinsidethevirions.SecondarystructuremodeloftheSARS-CoV-2genomeBasedonthe3DRNAinteractionmap,wefurtherdevelopedanadaptivestrategytomodelthesecondarystructureoftheentireSARS-CoV-2genome.Briefly,wefirstpredictedlocalduplexes,whichwerepositionedinourSARS-CoV-2RNAinteractionmapasperpendicularsignalstothediagonal,andthenusedtheseduplexesasconstraintstopredictlong-rangeduplexes(SupplementaryFig. 3a).Toevaluatetheperformanceofthisstrategy,wefirstpredictedthesecondarystructureof28SrRNAusingourpreviouslypublishedRIC-seqdatainHeLacells30.Thestructuralmodelachievedasensitivityof83.0%andapositivepredictivevalue(PPV)of78.3%,andbothcriteriaweresignificantlyhigherthanstructuresmerelybasedontheminimalfreeenergyvaluesprovidedbyseveralcomputationaltools(SupplementaryFig. 3b).Importantly,ouralgorithmshowedhigherPPVandsensitivityifRIC-seqdetectedduplexesspanningover600 ntwerecounted(SupplementaryFig. 3c).Together,thesedatademonstratetheaccuracyofouradaptivestrategyindeducingRNAstructures.Havingvalidatedthepredictionalgorithm,wenextappliedittoreconstructthesecondarystructureofthewholeSARS-CoV-2genome(Fig. 3).OurstructuralmodelwashighlyfavoredbythevRIC-seqdata,asbase-pairedregionsshowedstrongerinteractionstrengththansize-matchedunpairedcontrolsequences(P 0.01)wereclassifiedasinter-domain.Lastly,adjacentdomainsweremergedifbothdidnotcontaininteractionsbetweentheir5′and3′boundary.MFEanalysisoftheduplexesrevealedbyvRIC-seqToexplorewhetherthepixelswithastrongvRIC-seqsignalintheRNAinteractionmapcouldformlong-rangeduplexes,wefirstdividedtheSARS-CoV-2genomeintonon-overlapping10-ntwindows.Thepairwise10-ntwindowswithaconnectionscore>0.01wereusedfordownstreamanalyses30.Next,weclusteredpairwise10-ntwindowsadjoiningoroverlappingatbothendsasoneinteraction.ThelowesthybridfreeenergywasthencomputedforthepossiblehybridsformedbetweenthesepairwiseRNAstretchesusingthebifoldfunctionintheRNAstructuresuite(v6.2)withdefaultparameters60.Lastly,artificialsequenceswiththesamenucleotidecontentasrealinteractionsweregeneratedtentimes.Thelowesthybridfreeenergyforthoseshuffledsequenceswasalsocalculated.3DstructuralsimulationofSARS-CoV-2genomeBasedontheRNAinteractionmap,weusedtheminiMDSprogramtomodelthespatialconformationoftheSARS-CoV-2genomeusingthefollowingparameters39:minimds.py-l10-m0.01-p0.01.ThespatialcoordinatesreportedbyminiMDSweresmoothedwiththeLOWESS(locallyweightedscatterplotsmoothing)algorithmandthenvisualized62.Weaddedagriddedandgrayspheretothemodeled3DstructureoftheSARS-CoV-2genomeforeasyvisualization.PairwiseinteractionscapturedbyvRIC-seqinlocalordistalregionswereshownassolidordashedredlines,respectively.Moreover,wealsoprovidedamovieillustratingthe3DmodeloftheSARS-CoV-2genomeinvirion(SupplementaryVideo 1).SARS-CoV-2RNAsecondarystructuremodelingThesecondarystructureofSARS-CoV-2genomicRNAwasconstructedinsilicobasedsolelyonthevRIC-seqdatabyanadaptivelyoptimizedalgorithmwedevelopedinthisstudy.WefirstsplittheSARS-CoV-2genomeintoshortersegmentaldomainsbymaximizingtheratiobetweenintra-domainandinter-domain’svRIC-seqsignals.Notably,thedomainssmallerthan4 kbwillnotbefurthersplittoavoidthepotentiallossoflong-rangeduplexesoverthedomains’boundaries.Likeapreviouslydescribedapproach15,wedeterminedthesecondarystructureforeachdomainindependently.Tothisend,wesystematicallyscreenedpairwise5-ntwindowswithconnectionscoreshigherthan0.03,andthewindowsadjoiningoroverlappingatbothendswerefurtherclusteredashigh-confidenceinteractions.Foreachinteractionspannedregionwithinadomain,weusedtheFoldprogramintheRNAstructuresoftwaresuite(v6.2)toperformstructureprediction60.Themaximumdistancebetweenanytwopairedpositionswasallowedwithin2500 nt.FromthestructuralcandidatesreportedbytheFoldprogram,weselectedtheonethatmatchedbestwithvRIC-seqdataandforceditasaconstraintinthesubsequentprediction.Ofnote,wegeneratedduplexesforshortlocalinteractionsfirstandthenusedthemasrestraintstoperformpredictionforlong-rangeinteractionsspannedregions.Moreover,interactionshavingstrongervRIC-seqsignalswereprocessedwithpriority.Finally,byrestrainingduplexesgeneratedintheformerstage,wefoldedeachdomain’sentiresequence,includingregionsnotcoveredbythehigh-confidenceinteractions.ThestructureagreedbestwithvRIC-seqsignalswereselected.ThefinalsecondarystructuremodeloftheviralgenomeRNAinSARS-CoV-2wasvisualizedbytheVARNAprogram(v3-93)63andtheIntegrativeGenomicsViewer(IGV)visualizationtool(v.2.3.92)64.EvaluatetheaccuracyofthealgorithmToevaluatetheadaptivealgorithm’saccuracy,wefirstusedourpreviouslypublishedrRNA+RIC-seqdatainHeLacellstopredictthesecondarystructureof28SrRNA30.WealsousedfourwidelyusedcomputationalalgorithmstopredictthesecondarystructurebasedontheMFE,includingRNAstructure(v6.2)60,Mfold(v3.6)65,RNAfold66,andLinearFold67.Allparametersweresetasdefaultexceptforthemaximumpairingdistancewithin1600 ntwasallowed.Weusedtwocriteriatoevaluateapredictedsecondarystructure’saccuracy:sensitivityandPPV.Sensitivitywasdefinedasthefractionofbasepairsthatwerecorrectlypredicted.PPVwasthefractionofbasepairsinthepredictedstructurethatwerecorrect.Weusedarelaxedstructurecomparisonmode.Abasepairi/jwasconsideredcorrectlypredictedifanyofthefollowingpairsexistinthereferencestructure:i ± 5/j ± 5andviceversa.Theannotatedstructuralmodelof28SrRNAwasdownloadedfromtheRNAcentraldatabase(https://rnacentral.org/rna/URS000075EC78/9606)68.ComparisonwithCOMRADESTheCOMRADESdatageneratedwithSARS-CoV-2infectedVeroE6cellsweredownloadedfromtheGeneExpressionOmnibusdatabaseundertheaccessionnumberGSE15466228.WeanalyzedtheCOMRADESdatawiththeidenticalpipelineforvRIC-seqdata.WeusedthePPVtocomparetheinteractionscapturedbyvRIC-seqwithCOMRADESidentifiedduplexesinthehostcells.Here,PPVwasdefinedasthepercentageofvRIC-seqinteractionssupportedbyatleasttwouniquechimericreadsfromtheCOMRADESdataset.TheShannonentropyforeach10-ntgenomicwindowwascalculatedaspreviouslyreported29.Identificationofco-variantbasepairsAtotalof429non-redundantcoronavirusgenomesweredownloadedfromtheViPRdatabase69andalignedbytheMAFFTprogram(v7.471)70withdefaultparameterstoidentifyco-variantbasepairsinSARS-CoV-2.BasedonthemultiplesequencealignmentresultsandtheSARS-CoV-2secondarystructurederivedfromvRIC-seq,weemployedthecmbuildandcmcalibrateprogramintheInfernalpackage(v1.1.3)71tobuildthecovariancemodelandusedthecmserachinthesamepackagetosearchhomologs.HomologswithE-valuehigherthan0.01wereremoved,andthealignmentofleftsequenceswassubjectedtotheR-scapeprogram(v1.5.4)72toexplorecovarianceintheSARS-CoV-2.Intotal,406co-varianteventswereidentified(SupplementaryData 3).Thealignmentfileandcovariancemodelareavailableathttps://github.com/caochch/RIC2Structure.Co-variantbasepairsamongdifferentSARS-CoV-2strainswereinferredfrommultiplesequencealignmentof200,621non-redundantandhigh-qualitySARS-CoV-2genomesequencesdownloadedfromtheGlobalInitiativeforSharingAllInfluenzaData(GISAID)database73(date:7Jan2021).Abasepairwasclassifiedasaco-varianteventwhenbothnucleotidesweredifferentfromthereferenceSARS-CoV-2sequence(EPI_ISL_402125)butstillbase-paired.Intotal,98co-varianteventswereidentified(SupplementaryData 4).Co-variantbasepairswereplottedusingtheR-chieprogram74.SNPdensityandentropyscorecalculationSNPannotationforSARS-CoV-2wasdownloadedfromChinaNationalCenterforBioinformation(https://bigd.big.ac.cn/ncov,date:3Feb2021)75.Intotal,thisdatasetcontained9414synonymousSNPs.Eachnucleotide’sentropyscorewascalculatedaccordingtothemultiplesequencealignmentof200,621SARS-CoV-2genomesequences.Thefollowingformulaaspreviouslydescribedwasused:S = −100 × Sum(Pi × log2Pi),inwhichPiisthefrequencyoftheithallele69.Gaps‘−’ateitherendofasequenceinthealignmentwereremoved,andambiguousresidues‘N’wereexcludedfromentropycalculation.siRNAdesignBasedonthesingle-strandedregionsrevealedbyvRIC-seqandthesiRNAdesigningprinciplesdescribedearlier76,werandomlyselectedsixsiRNAsforexperimentalvalidationofSARS-CoV-2silencinginVerocells.Toexplorewhetherthein-virionSARS-CoV-2structurecouldguidesiRNAdesign,wecollected28siRNAsdesignedbyinsilicomodeling36,30siRNAsbytheViennaRNAwebserver(http://rna.tbi.univie.ac.at/cgi-bin/RNAxs/RNAxs.cgi),aswellasallthe331siRNAsdesignedfromthelinearsequenceofSARS-CoV-2.Usingthesecollections,wecomparedthenumberofsingle-strandednucleotidesatsiRNA-targetedregionsininfectedcells.WefoundthatthesixpotentsiRNAstargetregionsalsotendtobesingle-strandedinthehostcells.Comparisonwiththein-cellstructureThenormalizedSHAPE-MaPreactivitiesoneachnucleotideoftheSARS-CoV-2RNAintheinfectedVeroE6cellsweredownloadedfromAnnaPyle’slaboratorywiththelinkofhttps://github.com/pylelab/SARS-CoV-2_SHAPE_MaP_structure24.ToevaluatethestructuraldynamicsalongtheSARS-CoV-2genome,wecountedthepercentageofsharedbasepairswithin1 kbslidingwindowsforevery10 ntalongtheentireRNAgenome.Thecenterofslidingwindowsmovedfromthefirstnucleotidetothe29901stnucleotide.Whenthecenterofaslidingwindowissmallerthan501orlargerthan29401,the5′or3′endoftheslidingwindowexceedstheboundaryoftheviralgenome.Therefore,wetruncatedtheseslidingwindowsattheendpointoftheviralgenomeandcountedthepercentageofsharedbasepairswithinthetruncatedwindows.Programmed−1frameshiftassayThe−1PRFsequencefromSARS-CoV-2wasinsertedbetweenthecodingsequencesofrenillaluciferase(Rluc)andfireflyluciferase(Fluc)togenerateapDual-SARS-CoV-2plasmid(−1,WT).Forthein-framecontrolreporterpDual-SARS-CoV-2(0,Ctrl),anadditionalcytosinewasinsertedimmediatelyafterthesilentmutatedslipperysequencesuchthatFlucandRlucwereinthesamereadingframe(denotedas“0”)41.The33 ntsequence(GGTTATGGCTGTAGTTGTGATCAACTCCGCGAA)thatcouldformaduplexwiththeStem1regionofFSE-PKweredeletedbyPCRusingKODhot-starthighfidelityDNApolymerase(MerckMillipore,71086)withprimersshowninSupplementaryTable 1.FrameshiftingefficiencywasmeasuredbytransienttransfectionoftheDual-LuciferaseReporterplasmidsinto293Tcells(ATCC,CRL-3216)usingLipofectamine2000aspreviouslydescribed77.293Tcellswereseededina6-wellplateatadensityof30%.Onthefollowingday,0.2 μgofeachplasmidwereindividuallytransfectedintocellsandallowedforfurtherincubationof24 hina5%CO2incubator.LuciferaseactivitywasmeasuredusingaDual-LuciferaseReporterAssayKitbyfollowingthemanufacturer’sinstructions(Promega,E1910).WequantifiedtheframeshiftingefficiencybynormalizingWTandDelvaluestotheirin-framecontrolsaspreviouslydescribed78.ReportingsummaryFurtherinformationonresearchdesignisavailableinthe NatureResearchReportingSummarylinkedtothisarticle. Dataavailability Thedatasupportingthefindingsofthisstudyareavailablefromthecorrespondingauthorsuponreasonablerequest.vRIC-seqdatahavebeendepositedintheGeneExpressionOmnibus(GEO)databaseunderaccessionnumberGSE155733. Sourcedataareprovidedwiththispaper. Codeavailability HomemadescriptsforreconstructingthesecondarystructureofSARS-CoV-2genomeandforsimulatingglobalconfigurationcouldbefoundathttps://github.com/caochch/RIC2Structure. References1.Zhu,N.etal.AnovelcoronavirusfrompatientswithpneumoniainChina,2019.N.Eng.J.Med.382,727–733(2020).2.Ren,L.L.etal.Identificationofanovelcoronaviruscausingseverepneumoniainhuman:adescriptivestudy.Chin.Med.J.133,1015–1024(2020).PubMed  Article  PubMedCentral  GoogleScholar  3.Wu,F.etal.AnewcoronavirusassociatedwithhumanrespiratorydiseaseinChina.Nature579,265–269(2020).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  4.deWit,E.,vanDoremalen,N.,Falzarano,D.&Munster,V.J.SARSandMERS:recentinsightsintoemergingcoronaviruses.Nat.Rev.Microbiol.14,523–534(2016).PubMed  PubMedCentral  Article  CAS  GoogleScholar  5.Gordon,D.E.etal.ASARS-CoV-2proteininteractionmaprevealstargetsfordrugrepurposing.Nature583,459–468(2020).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  6.Kim,D.etal.TheArchitectureofSARS-CoV-2Transcriptome.Cell181,914–921.e910(2020).CAS  PubMed  PubMedCentral  Article  GoogleScholar  7.Zhou,P.etal.Apneumoniaoutbreakassociatedwithanewcoronavirusofprobablebatorigin.Nature579,270–273(2020).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  8.Patzel,V.etal.DesignofsiRNAsproducingunstructuredguide-RNAsresultsinimprovedRNAinterferenceefficiency.Nat.Biotechnol.23,1440–1444(2005).CAS  PubMed  Article  PubMedCentral  GoogleScholar  9.Vickers,T.A.,Wyatt,J.R.&Freier,S.M.EffectsofRNAsecondarystructureoncellularantisenseactivity.NucleicAcidsRes.28,1340–1347(2000).CAS  PubMed  PubMedCentral  Article  GoogleScholar  10.Hsue,B.&Masters,P.S.Abulgedstem-loopstructureinthe3′untranslatedregionofthegenomeofthecoronavirusmousehepatitisvirusisessentialforreplication.J.Virol.71,7567–7578(1997).CAS  PubMed  PubMedCentral  Article  GoogleScholar  11.Madhugiri,R.etal.Structuralandfunctionalconservationofcis-actingRNAelementsincoronavirus5′-terminalgenomeregions.Virology517,44–55(2018).CAS  PubMed  Article  PubMedCentral  GoogleScholar  12.Plant,E.P.etal.Athree-stemmedmRNApseudoknotintheSARScoronavirusframeshiftsignal.PLoSBiol.3,e172(2005).PubMed  Article  CAS  PubMedCentral  GoogleScholar  13.Yang,D.&Leibowitz,J.L.Thestructureandfunctionsofcoronavirusgenomic3′and5′ends.VirusRes.206,120–133(2015).CAS  PubMed  PubMedCentral  Article  GoogleScholar  14.Watts,J.M.etal.ArchitectureandsecondarystructureofanentireHIV-1RNAgenome.Nature460,711–716(2009).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  15.Li,P.etal.IntegrativeanalysisofZikavirusgenomeRNAstructurerevealscriticaldeterminantsofviralinfectivity.CellHostMicrobe24,875–886.e875(2018).CAS  PubMed  Article  PubMedCentral  GoogleScholar  16.Huber,R.G.etal.StructuremappingofdengueandZikavirusesrevealsfunctionallong-rangeinteractions.Nat.Commun.10,1408(2019).ADS  PubMed  Article  CAS  PubMedCentral  GoogleScholar  17.Dethoff,E.A.etal.PervasivetertiarystructureinthedenguevirusRNAgenome.Proc.NatlAcad.Sci.USA115,11513–11518(2018).CAS  PubMed  Article  PubMedCentral  GoogleScholar  18.Barcena,M.etal.Cryo-electrontomographyofmousehepatitisvirus:insightsintothestructureofthecoronavirion.Proc.NatlAcad.Sci.USA106,582–587(2009).ADS  CAS  PubMed  Article  PubMedCentral  GoogleScholar  19.Ke,Z.etal.StructuresanddistributionsofSARS-CoV-2spikeproteinsonintactvirions.Nature488,498–502(2020).20.Yao,H.etal.MoleculararchitectureoftheSARS-CoV-2virus.Cell183,730–738.e713(2020).CAS  PubMed  PubMedCentral  Article  GoogleScholar  21.Sun,L.etal.InvivostructuralcharacterizationoftheSARS-CoV-2RNAgenomeidentifieshostproteinsvulnerabletorepurposeddrugs.Cell184,1865–1883.e20(2021).22.Manfredonia,I.etal.Genome-widemappingofSARS-CoV-2RNAstructuresidentifiestherapeutically-relevantelements. NucleisAcidsRes.48,12436–12452(2020).23.Lan,T.C.T.etal.StructureofthefullSARS-CoV-2RNAgenomeininfectedcells.Preprintathttps://www.biorxiv.org/content/10.1101/2020.06.29.178343v1(2020).24.Huston,N.C.etal.ComprehensiveinvivosecondarystructureoftheSARS-CoV-2genomerevealsnovelregulatorymotifsandmechanisms.Mol.Cell81,584–598.e5(2021).25.Kwok,C.K.,Tang,Y.,Assmann,S.M.&Bevilacqua,P.C.TheRNAstructurome:transcriptome-widestructureprobingwithnext-generationsequencing.TrendsbiochemicalSci.40,221–232(2015).CAS  Article  GoogleScholar  26.Nicholson,B.L.&White,K.A.Functionallong-rangeRNA–RNAinteractionsinpositive-strandRNAviruses.Nat.Rev.Microbiol.12,493–504(2014).CAS  PubMed  PubMedCentral  Article  GoogleScholar  27.Lange,S.J.etal.Globalorlocal?PredictingsecondarystructureandaccessibilityinmRNAs.NucleicAcidsRes.40,5215–5226(2012).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  28.Ziv,O.etal.Theshort-andlong-rangeRNA-RNAInteractomeofSARS-CoV-2.Mol.Cell80,1067–1077.e5(2020).29.Ziv,O.etal.COMRADESdeterminesinvivoRNAstructuresandinteractions.Nat.methods15,785–788(2018).CAS  PubMed  PubMedCentral  Article  GoogleScholar  30.Cai,Z.etal.RIC-seqforglobalinsituprofilingofRNA–RNAspatialinteractions.Nature582,432–437(2020).31.Watanabe,Y.,Allen,J.D.,Wrapp,D.,McLellan,J.S.&Crispin,M.Site-specificglycananalysisoftheSARS-CoV-2spike.Science369,330–333(2020).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  32.Oostra,M.,deHaan,C.A.,deGroot,R.J.&Rottier,P.J.Glycosylationofthesevereacuterespiratorysyndromecoronavirustriple-spanningmembraneproteins3aandM.J.Virol.80,2326–2336(2006).CAS  PubMed  PubMedCentral  Article  GoogleScholar  33.Wakefield,J.K.,Wolf,A.G.&Morrow,C.D.Humanimmunodeficiencyvirustype1canusedifferenttRNAsasprimersforreversetranscriptionbutselectivelymaintainsaprimerbindingsitecomplementarytotRNA(3Lys).J.Virol.69,6021–6029(1995).CAS  PubMed  PubMedCentral  Article  GoogleScholar  34.SandersW,etal.ComparativeanalysisofcoronavirusgenomicRNAstructurerevealsconservationinSARS-likecoronaviruses.Preprintathttps://www.biorxiv.org/content/10.1101/2020.06.15.153197v1.full.pdf(2020).35.Tavares,R.d.C.A.,Mahadeshwar,G.,Wan,H.,Huston,N.C.&Pyle,A.M.TheglobalandlocaldistributionofRNAstructurethroughouttheSARS-CoV-2genome.J.Virol.95,e02190–20(2021).36.Rangan,R.etal.RNAgenomeconservationandsecondarystructureinSARS-CoV-2andSARS-relatedviruses:afirstlook.RNA26,937–959(2020).CAS  PubMed  PubMedCentral  Article  GoogleScholar  37.Goldsmith,C.S.etal.UltrastructuralcharacterizationofSARScoronavirus.Emerg.Infect.Dis.10,320(2004).PubMed  PubMedCentral  Article  GoogleScholar  38.Gui,M.etal.Electronmicroscopystudiesofthecoronavirusribonucleoproteincomplex.ProteinCell8,219–224(2017).PubMed  Article  PubMedCentral  GoogleScholar  39.Rieber,L.&Mahony,S.miniMDS:3Dstructuralinferencefromhigh-resolutionHi-Cdata.Bioinformatics33,i261–i266(2017).CAS  PubMed  Article  PubMedCentral  GoogleScholar  40.Lieberman-Aiden,E.etal.Comprehensivemappingoflong-rangeinteractionsrevealsfoldingprinciplesofthehumangenome.Science326,289–293(2009).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  41.Kelly,J.A.etal.Structuralandfunctionalconservationoftheprogrammed−1ribosomalframeshiftsignalofSARScoronavirus2(SARS-CoV-2).J.Biol.Chem.295,10741–10748(2020).CAS  PubMed  Article  PubMedCentral  GoogleScholar  42.Zhang,K.etal.Cryo-electronmicroscopyandexploratoryantisensetargetingofthe28-kDaframeshiftstimulationelementfromtheSARS-CoV-2RNAgenome.Preprintathttps://www.biorxiv.org/content/10.1101/2020.07.18.209270v1(2020).43.Stammler,S.N.,Cao,S.,Chen,S.J.&Giedroc,D.P.AconservedRNApseudoknotinaputativemolecularswitchdomainofthe3′-untranslatedregionofcoronavirusesisonlymarginallystable.RNA17,1747–1759(2011).CAS  PubMed  PubMedCentral  Article  GoogleScholar  44.Korber,B.etal.TrackingchangesinSARS-CoV-2spike:evidencethatD614GincreasesinfectivityoftheCOVID-19virus.Cell182,e819(2020).812-827.Article  CAS  GoogleScholar  45.Plante,J.A.etal.SpikemutationD614GaltersSARS-CoV-2fitness.Nature592,116–121(2020).46.Qureshi,A.,Tantray,V.G.,Kirmani,A.R.&Ahangar,A.G.AreviewoncurrentstatusofantiviralsiRNA.Rev.Med.Virol.28,e1976(2018).PubMed  PubMedCentral  Article  CAS  GoogleScholar  47.Masters,P.S.CoronavirusgenomicRNApackaging.Virology537,198–207(2019).CAS  PubMed  PubMedCentral  Article  GoogleScholar  48.Simon,L.M.etal.InvivoanalysisofinfluenzaAmRNAsecondarystructuresidentifiescriticalregulatorymotifs.NucleicAcidsRes.47,7003–7017(2019).CAS  PubMed  PubMedCentral  Article  GoogleScholar  49.Tomezsko,P.J.etal.DeterminationofRNAstructuraldiversityanditsroleinHIV-1RNAsplicing.Nature582,438–442(2020).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  50.Schmidt,N.etal.TheSARS-CoV-2RNA-proteininteractomeininfectedhumancells.Nat.Microbiol.6,339–353(2020).51.Flynn,R.A.etal.DiscoveryandfunctionalinterrogationofSARS-CoV-2RNA-hostproteininteractions.Cell184,1–18(2021).52.Boerneke,M.A.,Ehrhardt,J.E.&Weeks,K.M.PhysicalandfunctionalanalysisofviralRNAgenomesbySHAPE.Annu.Rev.Virol.6,93–117(2019).CAS  PubMed  PubMedCentral  Article  GoogleScholar  53.Kang,S.etal.CrystalstructureofSARS-CoV-2nucleocapsidproteinRNAbindingdomainrevealspotentialuniquedrugtargetingsites.ActaPharm.Sin.B10,1228–1238(2020).CAS  PubMed  PubMedCentral  Article  GoogleScholar  54.Gao,Q.etal.DevelopmentofaninactivatedvaccinecandidateforSARS-CoV-2.Science369,77–81(2020).55.Cameron,V.&Uhlenbeck,O.C.3′-PhosphataseactivityinT4polynucleotidekinase.Biochemistry16,5120–5126(1977).CAS  PubMed  Article  PubMedCentral  GoogleScholar  56.Dobin,A.etal.STAR:ultrafastuniversalRNA-seqaligner.Bioinformatics29,15–21(2013).CAS  Article  GoogleScholar  57.Krzywinski,M.etal.Circos:aninformationaestheticforcomparativegenomics.GenomeRes.19,1639–1645(2009).CAS  PubMed  PubMedCentral  Article  GoogleScholar  58.Durand,N.C.etal.Juicerprovidesaone-clicksystemforanalyzingloop-resolutionHi-Cexperiments.CellSyst.3,95–98(2016).CAS  PubMed  PubMedCentral  Article  GoogleScholar  59.Knight,P.A.&Ruiz,D.Afastalgorithmformatrixbalancing.IMAJ.Numer.Anal.33,1029–1047(2013).MathSciNet  MATH  Article  GoogleScholar  60.Reuter,J.S.&Mathews,D.H.RNAstructure:softwareforRNAsecondarystructurepredictionandanalysis.BMCBioinforma.11,1–9(2010).Article  CAS  GoogleScholar  61.Popenda,M.etal.Automated3DstructurecompositionforlargeRNAs.NucleicAcidsRes.40,e112(2012).CAS  PubMed  PubMedCentral  Article  GoogleScholar  62.Cleveland,W.S.Robustlocallyweightedregressionandsmoothingscatterplots.J.Am.Stat.Assoc.74,829–836(1979).MathSciNet  MATH  Article  GoogleScholar  63.Darty,K.,Denise,A.&Ponty,Y.VARNA:interactivedrawingandeditingoftheRNAsecondarystructure.Bioinformatics25,1974(2009).CAS  PubMed  Article  PubMedCentral  GoogleScholar  64.Robinson,J.T.etal.Integrativegenomicsviewer.Nat.Biotechnol.29,24–26(2011).CAS  PubMed  PubMedCentral  Article  GoogleScholar  65.Zuker,M.Mfoldwebserverfornucleicacidfoldingandhybridizationprediction.NucleicAcidsRes.31,3406–3415(2003).CAS  PubMed  PubMedCentral  Article  GoogleScholar  66.Lorenz,R.etal.ViennaRNAPackage2.0.AlgorithmsMol.Biol.6,26(2011).PubMed  PubMedCentral  Article  GoogleScholar  67.Huang,L.etal.LinearFold:linear-timeapproximateRNAfoldingby5′-to-3′dynamicprogrammingandbeamsearch.Bioinformatics35,i295–i304(2019).CAS  PubMed  Article  PubMedCentral  GoogleScholar  68.Consortium,T.R.RNAcentral:acomprehensivedatabaseofnon-codingRNAsequences.NucleicAcidsRes.45,D128–D134(2016). GoogleScholar  69.Pickett,B.E.etal.ViPR:anopenbioinformaticsdatabaseandanalysisresourceforvirologyresearch.NucleicAcidsRes.40,D593–D598(2012).CAS  PubMed  Article  PubMedCentral  GoogleScholar  70.Katoh,K.&Standley,D.M.MAFFTmultiplesequencealignmentsoftwareversion7:improvementsinperformanceandusability.Mol.Biol.Evol.30,772–780(2013).CAS  Article  GoogleScholar  71.Nawrocki,E.P.&Eddy,S.R.Infernal1.1:100-foldfasterRNAhomologysearches.Bioinformatics29,2933–2935(2013).CAS  PubMed  Article  PubMedCentral  GoogleScholar  72.Rivas,E.,Clements,J.&Eddy,S.R.AstatisticaltestforconservedRNAstructureshowslackofevidenceforstructureinlncRNAs.Nat.Methods14,45–48(2017).CAS  PubMed  Article  PubMedCentral  GoogleScholar  73.Elbe,S.&Buckland‐Merrett,G.Data,diseaseanddiplomacy:GISAID’sinnovativecontributiontoglobalhealth.Glob.Chall.1,33–46(2017).PubMed  PubMedCentral  Article  GoogleScholar  74.Lai,D.,Proctor,J.R.,Zhu,J.Y.A.&Meyer,I.M.R-chie:awebserverandRpackageforvisualizingRNAsecondarystructures.NucleicAcidsRes.40,e95–e95(2012).CAS  PubMed  PubMedCentral  Article  GoogleScholar  75.Zhao,W.-M.etal.The2019novelcoronavirusresource.Yichuan=Hereditas42,212–221(2020).PubMed  PubMedCentral  GoogleScholar  76.Lagana,A.,Shasha,D.&Croce,C.M.SyntheticRNAsforgeneregulation:designprinciplesandcomputationaltools.Front.Bioeng.Biotechnol.2,65(2014).PubMed  Article  PubMedCentral  GoogleScholar  77.Chen,J.etal.TheRNA-bindingproteinROD1/PTBP3cotranscriptionallydefinesAID-loadingsitestomediateantibodyclassswitchinmammaliangenomes.CellRes.28,981–995(2018).CAS  PubMed  PubMedCentral  Article  GoogleScholar  78.Wang,X.etal.RegulationofHIV-1Gag-Polexpressionbyshiftless,aninhibitorofprogrammed−1ribosomalframeshifting.Cell176,625–635.e614(2019).CAS  PubMed  Article  PubMedCentral  GoogleScholar  DownloadreferencesAcknowledgementsWethankDr.GuangxiaGaoforsharingthepDual-SARS-CoV-2(0and−1)plasmids.ThisworkwassupportedbytheStrategicPriorityProgramofCAS(XDB37000000),NationalKeyR&DProgram(2017YFA0504400),andtheNSFC(91740201,91940306,and81921003)toY.X.,bytheNSFC(31900465)toC.C.,bytheNSFC(31970611)toJ.C.,bytheNationalKeyR&DProgram(2020YFA0707500,2018YFA0900801)andtheStrategicPriorityResearchProgram(XDB29010000)toX.W.,bytheNationalKeyR&DProgram(2020YFA0707600),NationalMajorSciences&TechnologyProjectforControlandPreventionofMajorInfectiousDiseasesinChina(2018ZX10301401),andChineseAcademyofMedicalSciences(CAMS)InnovationFundforMedicalSciences(2016-I2M-1-014)toJ.W.AuthorinformationAuthornotesTheseauthorscontributedequally:ChangchangCao,ZhaokuiCai,XiaXiao.Theseauthorsjointlysupervisedthiswork:JianweiWang,YuanchaoXue.AffiliationsKeyLaboratoryofRNABiology,InstituteofBiophysics,ChineseAcademyofSciences,Beijing,ChinaChangchangCao, ZhaokuiCai, JuanChen, NaijingHu & YuanchaoXueUniversityofChineseAcademyofSciences,Beijing,ChinaZhaokuiCai, NaijingHu & YuanchaoXueNationalHealthCommissionofthePeople’sRepublicofChinaKeyLaboratoryofSystemsBiologyofPathogensandChristopheMérieuxLaboratory,InstituteofPathogenBiology,ChineseAcademyofMedicalSciences&PekingUnionMedicalCollege,Beijing,ChinaXiaXiao, JianRao & JianweiWangCASKeyLaboratoryofInfectionandImmunity,InstituteofBiophysics,ChineseAcademyofSciences,Beijing,ChinaMinnanYang, XiaoruiXing, YongleWang & XiangxiWangSchoolofLifeSciences,HenanNormalUniversity,Xinxiang,ChinaManmanLiStateKeyLaboratoryofStemCellandReproductiveBiology,InstituteofZoology,ChineseAcademyofSciences,Beijing,ChinaBingZhouInstituteforStemCellandRegeneration,ChineseAcademyofSciences,Beijing,ChinaBingZhouKeyLaboratoryofRespiratoryDiseasePathogenomics,ChineseAcademyofMedicalSciencesandPekingUnionMedicalCollege,Beijing,ChinaJianweiWangAuthorsChangchangCaoViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarZhaokuiCaiViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarXiaXiaoViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarJianRaoViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarJuanChenViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarNaijingHuViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarMinnanYangViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarXiaoruiXingViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarYongleWangViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarManmanLiViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarBingZhouViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarXiangxiWangViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarJianweiWangViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarYuanchaoXueViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarContributionsY.X.andJ.W.initiatedandplannedthestudy;C.C.performedthebioinformaticsanalysiswiththehelpofN.H.andB.Z.;Z.C.developedvRIC-seqtechnology,createdthedeep-sequencinglibrary,andperformedmostexperiments;X.X.andJ.R.performedSARS-CoV-2infectionandqPCRquantificationexperimentsundertheguidanceofJ.W.;M.Y.andX.X.preparedvirionsundertheguidanceofX.W.;Y.W.,M.L.,andJ.C.constructedDual-LuciferaseReportersandmeasuredtheluciferaseactivity.Y.X.,C.C.,andJ.W.wrotethemanuscript.CorrespondingauthorsCorrespondenceto JianweiWangorYuanchaoXue.Ethicsdeclarations Competinginterests Theauthorsdeclarenocompetinginterests. AdditionalinformationPeerreviewinformationNatureCommunicationsthankstheanonymousreviewersfortheircontributionstothepeerreviewofthiswork.Publisher’snoteSpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpublishedmapsandinstitutionalaffiliations.SupplementaryinformationSupplementary InformationSupplementary Data1Supplementary Data2Supplementary Data3Supplementary Data4Supplementary Data5SupplementaryMovie1DescriptionofAdditionalSupplementaryFilesReportingSummarySourcedata SourceDataRightsandpermissions OpenAccessThisarticleislicensedunderaCreativeCommonsAttribution4.0InternationalLicense,whichpermitsuse,sharing,adaptation,distributionandreproductioninanymediumorformat,aslongasyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCommonslicense,andindicateifchangesweremade.Theimagesorotherthirdpartymaterialinthisarticleareincludedinthearticle’sCreativeCommonslicense,unlessindicatedotherwiseinacreditlinetothematerial.Ifmaterialisnotincludedinthearticle’sCreativeCommonslicenseandyourintendeduseisnotpermittedbystatutoryregulationorexceedsthepermitteduse,youwillneedtoobtainpermissiondirectlyfromthecopyrightholder.Toviewacopyofthislicense,visithttp://creativecommons.org/licenses/by/4.0/. ReprintsandPermissionsAboutthisarticleCitethisarticleCao,C.,Cai,Z.,Xiao,X.etal.ThearchitectureoftheSARS-CoV-2RNAgenomeinsidevirion. NatCommun12,3917(2021).https://doi.org/10.1038/s41467-021-22785-xDownloadcitationReceived:05December2020Accepted:30March2021Published:24June2021DOI:https://doi.org/10.1038/s41467-021-22785-xSharethisarticleAnyoneyousharethefollowinglinkwithwillbeabletoreadthiscontent:GetshareablelinkSorry,ashareablelinkisnotcurrentlyavailableforthisarticle.Copytoclipboard ProvidedbytheSpringerNatureSharedItcontent-sharinginitiative Furtherreading InvivostructureanddynamicsoftheSARS-CoV-2RNAgenome YanZhang KunHuang ZhihuZhao NatureCommunications(2021) CommentsBysubmittingacommentyouagreetoabidebyourTermsandCommunityGuidelines.Ifyoufindsomethingabusiveorthatdoesnotcomplywithourtermsorguidelinespleaseflagitasinappropriate. DownloadPDF Advertisement Explorecontent Researcharticles Reviews&Analysis News&Comment Videos Collections Subjects FollowusonFacebook FollowusonTwitter Signupforalerts RSSfeed Aboutthejournal Aims&Scope Editors JournalInformation Openaccess Articleprocessingcharges EditorialValuesStatement JournalImpact Editors'Highlights Contact Top50Articles Editorialpolicies Publishwithus Forauthors ForReviewers Submitmanuscript Search Searcharticlesbysubject,keywordorauthor Showresultsfrom Alljournals Thisjournal Search Advancedsearch Quicklinks Explorearticlesbysubject Findajob Guidetoauthors Editorialpolicies