Cell,Vol.108,431–434,February22,2002,Copyright2002byCellPressNuclearRNATurnoverMelissaJ.Moore1HowardHughesMedicalInstituteDepartmentofBiochemistryBrandeisUniversity415SouthStreetWaltham,Massachusetts02454Inmammaliancells,significantlymoreRNAisturnedoverinthenucleusthaninthecytoplasm.However,onlyrecentlyhavewebeguntounderstandthemecha-nismsandregulationofnuclearRNAdecay.Highereukaryotesareremarkablywastefulintheirme-tabolismofRNA.Inmammals,only50%ofthenucleo-tidesincorporatedbyRNApolymeraseI(PolI)endupinmatureribosomes.Theremainderistrimmedawayinthenucleolusbythepre-rRNAprocessingmachinery.PolIItranscriptsfairevenworse,withlessthan5%oftheirnascentphosphodiesterbondsmakingitintocytoplasmicmRNA.Typically,nineoutoftennucleo-tidesinhumanPolIItranscriptsareremovedasintronsandrecycledwithinthenucleus.TheremainingexonicsequencesaresubjecttoadditionalqualitycontrolstepswhilestillassociatedwiththenucleustoeliminatemRNAswithincompleteopenreadingframesorotherdeformities(Figure1).AlthoughnodefinitivenumbersexistthatquantifyhowmuchPolII-transcribedRNAislosttosuchqualitycontrol,ithasbeenestimatedthatonly50%oftranscriptsreachingthe3Јendofthedys-trophingene,thelongestinthehumangenome,yieldmaturenuclearmRNAs(Jacksonetal.,2000,andrefer-encestherein).PolIIItranscripts(e.g.,5SRNAandtRNAs)arealsosubjecttotrimmingandsplicingbeforeexitingthenucleus.Thus,thebulkofRNAturnoverinmammaliancellsoccursinthenucleus,notthecyto-plasm.However,incontrasttocytoplasmicmRNAde-cay,whichhasbeenstudiedextensivelyforthepastdecadeasakeyregulatorystepingeneexpression,nuclearRNAturnoverisstilllargelyunexploredterritory.Thisminireviewwillfocusonrecentprogressinthisarea.TheNuclearRNADegradationMachineryLikeintracellularproteindegradation,intracellularRNAdegradationmustbecloselyregulatedsoastopreventwholesaleeliminationofalltranscripts.TheintracellularRNAdegradationmachineryislargelyexonucleolytic,allowingRNAstoescapedecaysimplybyprotectingtheirends.The5ЈendsofPolIItranscriptsareprotectedbya7mGpppGcapstructure,whichisaddedalmostimmediatelyaftertranscriptinitiation,andthenfurtherprotectedvianuclearandcytoplasmiccapbindingpro-teins.Attheotherend,the3ЈpolyAtailissequentiallyboundbynuclearandcytoplasmicpolyAbindingpro-teins.OtherRNAs,suchasrRNAs,snRNAs,snoRNAs,andtRNAs,carryalternatecapstructures,tightly-boundproteins,and/orstrongstructuralelementsattheirends.1Correspondence:mmoore@brandeis.eduMinireviewAnRNAcanthenbetargetedfordecaybydeadenyla-tion,decapping,displacementofboundproteins,dis-ruptionofitstertiarystructure,orendonucleolyticcleav-ageataninternalsite.Onceanendisaccessible,thereexistboth3Јto5Јand5Јto3Јexonucleasesthatcandegradeitineitherdirection.ThenuclearRNAdegradationmachineryhasbeenbestcharacterizedinbuddingyeast,wherethepredomi-nantnucleardecaypathwayis3Јto5Ј(Bousquet-Anto-nellietal.,2000).Mostofthisactivityisprovidedbytheexosome(vanHoofandParker,1999;MitchellandTollervey,2000),alsoknownaffectionatelyasthe“DeathStarofRNA.”DiscoveredintheTollerveylab,theexo-someisacomplexoftentoelevenproteins,mostofwhichhave3Јto5Јexonucleaseactivityinvitroorarepresumedexonucleasesbasedonsequencehomology.AccessoryfactorsincludeRNAhelicases,aputativeGTPase,andproteinsthattargettheexosometopartic-ularRNAsequences.Initiallycharacterizedasthespe-ciesresponsiblefor3Јendmaturationof5.8SrRNA,nuclearexosomesarealsorequiredfor3ЈprocessingofsnRNAsandsnoRNAsandforthedegradationofpre-rRNAspacerregions.Alloftheexosomalsubunitsareconservedinhumans,wherethecomplexissynony-mouswiththePM-Sclparticle,atargetofautoimmuneantibodies(vanHoofandParker,1999;MitchellandTollervey,2000,andreferencestherein).Exosomesalsofunctioninthecytoplasm,wheretheyarerequiredfor3Јto5ЈdecayofdeadenylatedmRNAs(vanHoofandParker,1999).Cytoplasmicexosomesdifferfromtheirnuclearcounterpartsinthattheylackonesubunit,thenuclearproteinRrp6p.Whilethetencoresubunitsareallessentialforgrowth,deletionofRrp6ponlycausestemperature-sensitivelethality.CellslackingRrp6pactivityaccumulatestablenuclearRNAswithincompletelyprocessed3Јends.Rrp6phasalsobeenimplicatedintheturnoverofnuclearpre-mRNAsandmRNAs,andtheretentionnearthetranscriptionsiteofmRNAscontainingaberrant3Јends(BurkardandButler,2000;Hillerenetal.,2001;andseebelow).Thereisalsoevidencefora5Јto3Јexonucleolyticdecaypathwayinthenucleus.S.cerevisiaecontainstwoknown5Јto3Јexonucleases,Rat1pandXrn1p,bothofwhichareconservedthroughouteukaryotes.Rat1pispredominantlynuclearandisessential.Itisknowntobeinvolvedinpre-rRNAprocessingandsnoRNAmaturation,andmayalsofunctioninmRNAmaturation,astemperature-sensitiverat1-1cellsrapidlyaccumulatenuclearpolyAϩRNAaftershifttorestrictivetemperature.TheXrn1penzymeprovidesthemajorde-caypathwayfordeadenylatedandnonsense-containingmRNAsinthecytoplasm,althoughitisnotanessentialprotein.Whentheirsubcellularlocalizationsarealtered,bothRat1pandXrn1parefunctionallyinterchangeable,andevenundernormalconditions,Xrn1pappearstooverlapRat1p’sfunctionsinthenucleus(forreferences,seeGeerlingsetal.,2000;Xueetal.,2000).AlthoughRat1pandtheexosomeconstitutethemajornucleardecaymachineryknowntodate,becauseourknowl-Cell432Figure1.NuclearRoutesofProcessingandTurnoverforRNAPolIITranscriptsNuclearprocessingofPolIItranscriptsincludescapping,excisionofintrons,andadditionofapolyAtail.ThenuclearRNAdegradationmachineryconsistslargelyofexonucleases,thebestcharacterizedofwhicharetheexo-someandRat1p.Inadditiontodegradingex-cisedintrons,nuclearexonucleasesappeartoassistintheeliminationofinefficientlypro-cessedpre-mRNAsandmalformedmRNAs.edgeislimited,therecouldwellexistotherimportantdegradativeenzymesthatawaitcharacterization.NuclearDiscardPathwaysforInefficientlyandImproperlyProcessedTranscriptsInS.cerevisiae,mutationsinpre-mRNAsplicingfactorscandramaticallyreducesplicedmRNAlevelswithoutconcomitantlyincreasingthelevelsofunsplicedpre-mRNAs.Thisimpliestheexistenceofanefficientpre-mRNAdiscardpathway.Recently,theTollerveylabdemonstratedthatsuchapathwaydoesoperateinS.cerevisiae(Bousquet-Antonellietal.,2000).Theyfoundthatwhenexosomemutations(exoϪ)werecom-binedwithastrongsplicingmutation(prp2-1),pre-mRNAlevelsrose20-to50-foldoverthelevelsobservedwiththeprp2-1mutationalone.Surprisingly,splicedmRNAlevelsalsoincreasedsignificantlyintheexoϪprp2-1cellsoverexoϩprp2-1cells.Theseandotherdatasuggestthattheexosomenormallydegradespre-mRNAsthatarenotefficientlyprocessed,butwhenthesepre-mRNAsarenotdegraded,theycanbeslowlysplicedintheprp2-1background.Consistentwiththenotionthatthereisdynamiccompetitionbetweensplic-ingofnascenttranscriptsandtheirdegradationbytheexosome,disruptionofexosomeactivityaloneinanotherwisewild-typebackgroundincreasedthesteady-statelevelsofbothpre-mRNAsandsplicedmRNAs.However,thesamemutationshadlittleeffectonthesteady-statelevelofanunsplicedmRNA,suggestingthatintronlessmRNAsmaynotbeefficientsubstratesfornuclear3Јto5Јdecayinbuddingyeast.Inanotherrecentstudy,theexosomewaslinkedtoaquality-controlsystemformonitoringpropermRNA3Ј-endformation.InS.cerevisiae,bothhypo-andhyper-adenylatedmRNAsareretainedatornearthesiteoftranscription.Similarly,inefficientlyprocessedtranscriptsinmammaliancellsaccumulatearoundthetranscriptionsite(Hillerenetal.,2001;andreferencestherein).Thissuggeststheexistenceofsurveillancesystemsthatin-hibittranscriptionsitereleaseofRNAsthathavenotbeenproperlyprocessed.Remarkably,arecentstudyinS.cerevisiaerevealedthatmutationofthreedifferentexosomesubunits,includingthenuclear-specificsub-unitRrp6,alleviatedthetranscriptionsiteaccumulationofhypo-andhyperadenylatedmRNAs(Hillerenetal.,2001).Thus,inadditiontoitsexonucleolyticactivities,anotherfunctionoftheexosomemaybetoretainatthesiteoftranscriptionmRNAsthatlackappropriatelengthpolyAtails,orareotherwisemalformed.Consistentwithsucharole,Rrp6interactsgeneticallyandphysicallywithboththepolyApolymeraseandanucleocytoplas-micshuttlingprotein,Npl3p,involvedinmRNAexport(BurkardandButler,2000).FurtherexperimentswillberequiredtodeterminewhetherthedegradativeandmRNAretentionfunctionsoftheexosomeareequallydependentonintactexonucleaseactivesitesorarefunctionallyseparable.Also,itwillbeofgreatinteresttoinvestigatesuchquality-controlpathwaysinmammaliancells,wheretheproliferationofalternativesplicingandalternativepolyadenylationarelikelyaccompaniedbysignificantlyincreasedfrequenciesofprocessingerrors.Nonsense-MediatedmRNADecayAmoreextensivelystudiedquality-controlsystemthanthosedescribedaboveisthenonsense-mediatedmRNAdecay(NMD)pathway(MaquatandCarmichael,2001;Wiluszetal.,2001).ThisisameansforeliminatingmRNAswithincompleteopenreadingframes,therebypreventingsynthesisofpotentiallydeleterious,trun-catedproteins.SuchmRNAscanarisebygeneticmuta-tion(e.g.,frameshiftandnonsensemutations),throughtranscriptionalandprocessingerrors,orviathenaturalgenerearrangementprocessrequiredforlymphocytedevelopment.StudiesinbuddingyeastareconsistentwithNMDbeingapurelycytoplasmicprocess.Support-ingthisideaareobservationsthatNMDinvolvesthenormaltranslationapparatus,whichuponrecognitionofaprematureterminationcodon,triggersmRNAde-cappingwithoutdeadenylationandthen5Јto3Јdegra-dationviathecytoplasmicXrn1exonucleasepathway.Ontheotherhand,experimentsinmammaliancellshaveclearlyimplicatedthenucleusinNMD.Earlyexperi-mentsrelyingonbiochemicalseparationofthenuclearandcytoplasmiccompartmentsrevealedthatmanymammalianmRNAscontainingtruncatedopenreadingframesarefoundatlowerlevelsthanthecorrespondingMinireview433wild-typemRNAsinbothnuclearandcytoplasmicfrac-tions.ApopularhypothesistoexplainthisphenomenonisthatmammalianmRNAsaresubjecttoreadingframerecognitionbycytoplasmicribosomescoincidentwithnuclearexport.Inthisway,aprematurelytruncatedopenreadingframecouldbedetectedandtheoffendingmRNAdegradedbeforeitiseverreleasedfromthenu-clearporecomplex.However,somerecentstudieshaverekindledanoldideathattheremayactuallybereadingframerecognitionandeventranslationwithinthenu-cleusitself.First,subcellularlocalizationstudieshaveshownthatnotonlyisUpf3p,aproteinrequiredforNMD,predominantlynuclearinmammaliancells,significantamountsofthecytoplasmictranslationfactorseIF4EandeIF4Garealsofoundinthenucleus(forreferences,seeIborraetal.,2001;Ishigakietal.,2001;Muhlemannetal.,2001).Thenuclearfractionofthelatterfactors(ف10%oftotal)issimilartothefractionofproteinsyn-thesisrecentlyattributedtothenucleiofpermeablizedmammaliancells(Iborraetal.,2001).Thenuclearassoci-ationofnewlymadeproteinsobservedbyIborraetal.isdependentonongoingtranscription,andRNA/proteincolabelingtechniquesrevealedthatnewlymadenuclearproteinscolocalizewithnewlymadeRNAs.Iftranslationdoesoccurinthenucleus,towhatextentdoesitinvolvepredominantlynuclearfactors?ApartialanswertothiscamefromtheMaquatlab,whichshowedthatthefirst,or“pioneering,”roundoftranslationandthisleadstoNMDoccurswhilemammalianmRNAsarestillboundbythenuclearcapbindingcomplexandnuclearpolyAbindingprotein(Ishigakietal.,2001).Finally,theMooreandWilkinsonlabsreportedthatdisruptionofmRNAreadingframecanleadtoaccumulationofunsplicedpre-mRNAsatornearthetranscriptionsiteofthemutantallele(Muhlemannetal.,2001).Thisresultopensthetantalizingpossibilitythat,asinbacteria,readingframerecognitioninmammaliancellsmightevenoccurcotran-scriptionally.Asofthiswriting,itremainstobedeter-minedwhetherthepre-mRNAsthataccumulateupondisruptionofreadingframerepresentmoleculesinlineformRNAsynthesisandthenNMD,ormoleculesthathavesomehowbeentargetedfordestructionpriortosplicing.Also,itiscurrentlyunknownwhatrelationship,ifany,existsbetweenthesetranscriptionsiteaccumula-tionsandthoseobservedinyeastformRNAswithaber-rant3Јends(Hillerenetal.,2001;seeabove).IfsomeormostNMDdoesoccurinsidethemamma-liannucleus,itwillbeofgreatinteresttodeterminewhichexonucleasepathwayisinvolved.DoesNMDofnuclearmRNAsinvolvedeadenylation-independentde-cappingandthen5Јto3Јdecay,asdoescytoplasmicNMDinS.cerevisiae?Ifso,whatisthedecappingen-zymeandwhich5Јto3Јexonucleaseortholog(Rat1p,Xrn1p,oranovelenzyme)isinvolved?Or,giventhattheexosomeisthemoreactiveexonucleaseinthenu-cleusofbuddingyeast(Bousquet-Antonellietal.,2000),perhapsmammalianNMDisapredominantly3Јto5Јdegradationpathway.Recentworkshowingthattheexosomeisattractedtonaturallyshort-livedmRNAsbyproteinsboundtoinstabilityelements(seebelow),mightsuggestaparallelmechanismforexosomeattractiontomRNAsdestinedforNMD.IntronDegradationEventhoughintronsrepresentupto90%ofthenucleo-tidesincorporatedbyRNAPolIIinthenucleiofhighereukaryotes,untilrecentlyverylittleattentionhasbeenpaidtothemetabolismofthesesequencesoncetheyhavebeensplicedout.Althoughafewintronshaveregu-latoryrolesorencodesmallnucleolarRNAs(snoRNAs),themajorityarethoughttohavenospecificfunction.Aftersplicing,theymustbedisassembledfromthespliceosomeanddegraded.Duetotheirlariatstructure,arequisitestepinintrondegradationiscleavageofthe2Ј-5Јphosphodiesterbondatthebranchsite.Thisreactioniscarriedoutbyanevolutionarilyconserveddebranchingenzyme,deletionofwhichconfersaseveregrowthdefectinSchizosaccharomycespombe(Nametal.,1997).WhilelackofthisenzymeisoflittlephenotypicconsequenceinS.cerevisiae,whichhasrelativelyfewintronscomparedtoS.pombeandhighereukaryotes,debranching-deficientS.cerevisiaecellsdoaccumulatecircularintronsmissingthelariattail(ChapmanandBoeke,1991).Thissuggeststhat,likeothernuclearRNAs,themaindegradationpathwayfordebranchedintronsisanexonucleolyticone.However,whichexo-nucleasesareresponsibleisnotyetknown.Recently,theWilkinsonlabshowedthatthehalf-livesoffourmammalianintronsaftersplicingweresurpris-inglylong,rangingfrom6to29min.Suchhalf-livesarecomparabletonaturallyunstablemRNAssuchasc-mycandc-fos(Clementetal.,2001;andreferencestherein).Giventhatthelongestintroninthatstudywas2.4kb,andthehalf-livesoftheintronstheyexaminedcorrelatedlooselywithlength,itmaynowbeofinteresttodeter-minewhetherlongerintronshaveevenlongerhalf-livesandwhetherintronsthataresignificantlymorestablethanthemeanhaveanyregulatoryconsequencesforgeneexpression.NuclearStabilizationofUnstablemRNAsManymRNAsarenaturallyshort-lived.Thisisanimpor-tantcontrolmechanismfortemporallylimitingthesyn-thesisoftheencodedproteins(e.g.,cytokinesandgrowthfactors).SuchmRNAsoftencontaininstabilityelementsintheir3ЈUTRs.Onewell-studiedinstabilityelementistheARE,orAU-richelement(BrennanandSteitz,2001;GuhaniyogiandBrewer,2001).Recently,itwasshownthatthiselementdestabilizesanmRNAbyattractingexosomestoit(Chenetal.,2001;Mukherjeeetal.,2002).Exosomesarerecruitedthroughdirectinter-actionwiththeARE(Mukherjeeetal.,2002),aswellasthroughadaptorproteinsthatbridgetheAREandexosome(Chenetal.,2001).Giventhatexosomespuri-fiedfrombothnuclearandcytoplasmiccompartmentsstimulatedegradationofARE-containingmRNAs(Chenetal.,2001),onewondershowARE-containingmRNAsareeverabletorunthegauntletofexosomesinthenucleusandescapeintacttothecytoplasm.TheanswerliesintheexistenceofrepressorproteinsthatbindAREsequencesinthenucleusandpreventefficientexosomeacquisitioninthatcompartment.OnesuchproteinisHuR,amemberoftheELAVfamilyofRNAbindingpro-teins.HuRisapredominantlynuclearshuttlingprotein,overexpressionofwhichstabilizesARE-containingmRNAs.Inadditiontofendingoffnuclearexosomes,HuRalsohelpsshepherdARE-containingmRNAsoutofthenu-cleusviaanalternatemRNAexportpathway.OnceinCell434thecytoplasm,HuRremainsboundlongenoughfortheARE-containingmRNAstobetakenupintopolysomes(BrennanandSteitz,2001).Presumably,theeventualdissociationofHuRanditsreturntothenucleusthenallowsassociationofcytoplasmicexosomeswithARE-containingmRNAs,resultinginrapiddecay.Itiscur-rentlyunknownwhetherHuRdissociationisapassiveprocessdependentsolelyonitsdissociationconstantandlowerconcentrationofHuRinthecytoplasm,orwhetherthetranslationmachineryassistsinHuRre-movaltoactivelylimitthenumberofproteinmoleculestranslatedfromARE-containingRNAs.Insummary,althoughourunderstandingofthema-chineryandregulationofRNAturnoverinthenucleusisrapidlyincreasing,itisclearthatmanymorerevelationsawait.Amongthoserevelationswillundoubtedlybemul-tiplesurprises.StudyofNMDinhighereukaryotesisalreadypushingourperceptionsastotheextenttowhichRNAprocessingandtranslationarephysicallycompartmentalizedineukaryoticcells.TheelucidationofothernucleardecaypathwaysmaywellrevealalevelofinterplaybetweentheRNAprocessinganddecaymachineriesthatwecanonlynowjustimagine.SelectedReadingBousquet-Antonelli,C.,Presutti,C.,andTollervey,D.(2000).Cell102,765–775.Brennan,C.M.,andSteitz,J.A.(2001).Cell.Mol.LifeSci.58,266–277.Burkard,K.T.,andButler,J.S.(2000).Mol.Cell.Biol.20,604–616.Chapman,K.B.,andBoeke,J.D.(1991).Cell65,483–492.Chen,C.Y.,Gherzi,R.,Ong,S.E.,Chan,E.L.,Raijmakers,R.,Pruijn,G.J.,Stoecklin,G.,Moroni,C.,Mann,M.,andKarin,M.(2001).Cell107,451–4.Clement,J.Q.,Maiti,S.,andWilkinson,M.F.(2001).J.Biol.Chem.276,16919–16930.Geerlings,T.H.,Vos,J.C.,andRaue,H.A.(2000).RNA6,1698–1703.Guhaniyogi,J.,andBrewer,G.(2001).Gene265,11–23.Hilleren,P.,McCarthy,T.,Rosbash,M.,Parker,R.,andJensen,T.H.(2001).Nature413,538–542.Iborra,F.J.,Jackson,D.A.,andCook,P.R.(2001).Science293,1139–1142.Ishigaki,Y.,Li,X.,Serin,G.,andMaquat,L.E.(2001).Cell106,607–617.Jackson,D.A.,Pombo,A.,andIborra,F.(2000).FASEBJ.14,242–254.Maquat,L.E.,andCarmichael,G.G.(2001).Cell104,173–176.Mitchell,P.,andTollervey,D.(2000).Nat.Struct.Biol.7,843–846.Muhlemann,O.,Mock-Casagrande,C.S.,Wang,J.,Li,S.,Custodio,N.,Carmo-Fonseca,M.,Wilkinson,M.F.,andMoore,M.J.(2001).Mol.Cell8,33–43.Mukherjee,D.,Gao,M.,O’Connor,J.P.,Raijmakers,R.,Pruijn,G.,Lutz,C.S.,andWilusz,J.(2002).EMBOJ.21,165–174.Nam,K.,Lee,G.,Trambley,J.,Devine,S.E.,andBoeke,J.D.(1997).Mol.Cell.Biol.17,809–818.vanHoof,A.,andParker,R.(1999).Cell99,347–350.Wilusz,C.J.,Wang,W.,andPeltz,S.W.(2001).GenesDev.15,2781–2785.Xue,Y.,Bai,X.,Lee,I.,Kallstrom,G.,Ho,J.,Brown,J.,Stevens,A.,andJohnson,A.W.(2000).Mol.Cell.Biol.20,4006–4015.