Global view on the metabolism of RNA poly(A) tails in yeast ...

The evolutionarily conserved CPF complex mediates transcription termination at protein-coding genes coupling endonucleolytic cleavage to Pap1- ... Skiptomaincontent Thankyouforvisitingnature.com.YouareusingabrowserversionwithlimitedsupportforCSS.Toobtain thebestexperience,werecommendyouuseamoreuptodatebrowser(orturnoffcompatibilitymodein InternetExplorer).Inthemeantime,toensurecontinuedsupport,wearedisplayingthesitewithoutstyles andJavaScript. Advertisement nature naturecommunications articles article GlobalviewonthemetabolismofRNApoly(A)tailsinyeastSaccharomycescerevisiae DownloadPDF DownloadPDF Subjects RNARNAdecayTranscriptomics AbstractThepolyadenosinetail(poly[A]-tail)isauniversalmodificationofeukaryoticmessengerRNAs(mRNAs)andnon-codingRNAs(ncRNAs).Inbuddingyeast,Pap1-synthesizedmRNApoly(A)tailsenhanceexportandtranslation,whereasTrf4/5-mediatedpolyadenylationofncRNAsfacilitatesdegradationbytheexosome.UsingdirectRNAsequencing,wedeciphertheextentofpoly(A)taildynamicsinyeastdefectiveinallrelevantexonucleases,deadenylases,andpoly(A)polymerases.PredominantlyncRNApoly(A)tailsare20-60adenosineslong.Poly(A)tailsofnewlytranscribedmRNAsare50adenosinelongonaverage,withanupperlimitof200.ExonucleolysisbyTrf5-assistednuclearexosomeandcytoplasmicdeadenylasestrimthetailsto40adenosinesonaverage.Surprisingly,PAN2/3andCCR4-NOTdeadenylasecomplexeshavealargepoolofnon-overlappingsubstratesmainlydefinedbyexpressionlevel.Finally,wedemonstratethatmRNApoly(A)taillengthstronglyrespondstogrowthconditions,suchasheatandnutrientdeprivation. IntroductionPolyadenylationoftheRNA3′-endisaubiquitousmodificationtiedtonumerousessentialbutalsoconflictingfunctions.InthebuddingyeastSaccharomycescerevisiae,thesourceofpolyadenylationdefinesthefateofthetranscript.Pap1,asubunitofthecleavageandpolyadenylationfactor(CPF)complex,polyadenylatesnewlyproducedmRNAstopromotetheirstabilityandexporttothecytoplasm.Incontrast,Trf4orTrf5poly(A)polymerasesconstitutingsubunitsofparalogousnuclearTRAMP4orTRAMP5(Mtr4-Air1/2-Trf4/5)complexesfacilitatedegradationorprocessingofnuclearnoncodingRNAs(ncRNAs)bytheexosomecomplex.TRAMP4/5complexescanactco-orpost-transcriptionally.TheNrd1-Nab3-Sen1(NNS)systemmediatestranscriptionterminationatcrypticunstabletranscripts(CUTs)andsmallnuclear/nucleolarRNA(sn/snoRNA)lociandphysicallyrecruitsTRAMP4togetherwiththeexosomecomplextothoseRNAs1,2.Independentoftranscription,TRAMP5participatesinthecontrolofmRNAstabilityandpre-ribosomalRNA(rRNA)processing3,4,5.ConcertedpolymeraseandMtr4-mediatedhelicaseactivitiesofTRAMP4/5promotethe3′–5′exonucleolyticactivitiesoftheexosomeRrp6andDis3subunits.However,polyadenylationisnotabsolutelyrequiredforexonucleolysisasacatalyticallydeadTrf4partiallyrescuesthetrf4Δgrowthanddecayphenotypes6.ThelengthofinvivoTRAMP-dependentpoly(A)tailsisdebated.SelectedstudiesinvitroshowthattheMtr4RNAhelicaselimitsTrf4/5polyadenylationto3–5adenosines7,8.Indeedinvivo,high-throughputRNA-sequencing(RNA-seq)experimentsdetectedexclusivelysucholigoadenylatedterminalorinternaldecayintermediates3,9,which,however,mightnotberepresentativebecauseofthedetectionlimitofthesequencingmethod.Incontrast,otherinvitrostudieshaveshownthatTRAMP4/5canproducepoly(A)stretchesmanytimeslonger6,10.TheevolutionarilyconservedCPFcomplexmediatestranscriptionterminationatprotein-codinggenescouplingendonucleolyticcleavagetoPap1-mediatedpolyadenylationofthetranscript.Thebulkofyeastpoly(A)tailsisamaximumof70adenosines(As)long.However,invitrostudiesshowedthatPap1canproducemuchlongerpoly(A)tails11,12,13,raisingthequestionoftheinvivopolyadenylationlimituponmRNAdenovosynthesis.Apoly(A)tailofatleast48AsisrequiredfornuclearmRNAstabilityandefficientexportfacilitatedbythenuclearpoly(A)bindingproteinNab214,15,16.Nab2isapartoftheassembledco-andpost-transcriptionallyexport-competentmessengerribonucleoprotein(mRNP)complex,whichcontainsthemainexport-adapter,Mex67/Mtr2dimerandsupportingTHO/TREXcomplex(Thp2,Tho2,Tex1,Hpr1,Mft1)17.Nab2islaterreplacedinthecytoplasmbyPab1,whichbringstogetherthepoly(A)tailwiththecap-boundtranslationinitiationfactors18.Inthecytoplasm,themRNApoly(A)tailcanbetargetedbytwodeadenylasecomplexes,PAN2/3andCCR4-NOT19.Invitro,PAN2/3actionisenhancedbybindingtomultimerizedPab120,21,whileCCR4-NOTpreferentiallydegradesmRNAswithnoorlowPab1load22.ThosedatamightsupportthemodelpostulatingthatPAN2/3mediatesthefirstrapidstepofdeadenylation,followedbytheactionoftheCCR4-NOTcomplexresultinginmRNAdecappingand5′–3′decaybyXrn119,20,23,24.However,comprehensiveinvivotranscriptomicevidenceinyeastsupportingthismodelisstilllacking.Here,wedefinedthecontributionofallS.cerevisiaepoly(A)polymerases,theexosometogetherwithitscofactors,anddeadenylasecomplexestoshapingpoly(A)taillengthandabundanceofallRNAbiotypes.WeusedthenanoporedirectRNAsequencing(DRS)technology25tostudythepoly(A)+RNAfraction,whichimprovestheestimationoflongpolyadenosinereadscomparedwithpoly(A)test(PAT)-basedmethods26orTAIL-seq27.WefoundthatCUT,rRNA,andsnRNA3′endsweremainlypolyadenylatedbyTRAMP4/5andintheexosome-deficientbackgroundhadpoly(A)tailscomparableinlengthtomRNAs.Denovo-producedmRNApoly(A)tailswere50Aslong,withanupperlimitof200As.Atsteadystate,mRNApoly(A)tailsweredeadenylatedto40adenosinesonaverage.Surprisinglythenuclearexosome,assistedbyTRAMP5,extensivelytrimmedmRNApoly(A)tails,whileTRAMP4adenylatedmRNAsinexosome-andPap1-deficientcells.WefoundthatPAN2/3andCCR4-NOTbothmediatethefirstcytoplasmicdeadenylationphaseofmRNAs.Eachcomplexhasitsdedicatedsetofsubstratesbestdefinedbyexpressionlevel,withPAN2/3targetingabundantmRNAsandCCR4-NOTbeingmorespecialized.ResultsDRSefficientlyestimatesmRNAabundanceandpoly(A)taillengthofyeastRNAWedeterminedtherangeofpoly(A)taillengthsofyeastRNAsandtheirabundance,usingnanoporedirectRNAsequencing(DRS)ofthepoly(A)+RNAfractionextractedwitholigo-(dT)25beads.SuchRNA,efficientlydepletedofnon-polyadenylatedncRNAspecies(SupplementalFig. 1a),wasasubstrateforlibrarypreparationbasedonsplintligationwith(dT)10andreversetranscriptiontoproduceanRNA:cDNAhybrid(Fig. 1a)25.DRScountsalignedwellwithpreviouslypublisheddataderivedfromRNA-seqonpoly(A)+fractionortotalRNAafterribodepletion,aswellastilingarrays28,29(SupplementalFig. 1b),indicatingthatquantitativelyDRSiscomparabletoothergenome-widetechniques.Moreover,DRScountsstronglycorrelatedwithoneanother,whereas,aspreviouslyshown,poly(A)taillengthshadahigherdegreeofvariability30(SupplementalFig. 1c–e).Fig.1:RNApoly(A)taillengthanticorrelateswithRNAexpression.aSchematicdiagramoftheDRStechniqueusedtoestimatepoly(A)taillengths.Thepoly(A)+fractionisisolatedfromtotalRNA;thenlibrariesarepreparedaccordingtotheprotocol,withthefirststepof(dT)10adapterligation.Aftersequencingbioinformaticanalysisstepsareperformed:basecalling,mappingtotranscriptome,andestimationofthelengthofthepoly(A)tail(markedinpink)byNanopolishusingarawsignalreadout.bScatterplotsshowingintheleftpanelthecomparisonoflog2mRNAabundanceinDRSlibrariespreparedusingthepoly(A)+fractionortotalsampleandintherightpanelthemeanpoly(A)taillengthestimatesinbothlibrarytypes.cRelationshipbetweenRNAabundanceandthemeanpoly(A)taillengthofdifferentRNAbiotypes,averagedfromthreeindependentWTcontroldatasets(twofromW303andonefromBYgeneticbackgrounds,eachcomposedofaminimumoftwobiologicalrepeats).ThecutofflineshowsthelimitoftheefficientDRSpoly(A)tailestimationmethod(10 A).dRelationshipbetweenDRS-definedmeanpoly(A)taillength(leftpanel)andlog2-scaledmRNAabundance(rightpanel)relativetolog2-scaledmRNAsynthesisratefromcDTAdataset36.eRelationshipbetweenmeanmRNApoly(A)taillengthandtranscriptshalf-lifefromcDTAorRATE-seqdatasets34,36.“h/ex”inthekeydesignateshighlyexpressedmRNAs,whicharemarkedinred(forcDTA)ororange(forRATE-seq).fRelationshipbetweenmeanpoly(A)taillength(leftpanels),log2-scaledmRNAabundance(rightpanels),thelog2-scalednumberofproteinmoleculespercell37(upperpanels),thelog2-scalednumberofproteinmoleculesthatareproducedfromagiventranscriptduringitslifespan(middlepanels)andlog2-scaledrawribosomeoccupancyestimation(bottompanels;factorbandwfromSiwiakandZielenkiewicz38).Pearson’scorrelationcoefficientsareindicatedoneachscatterplot.FullsizeimageDRSprotocolstronglyrecommendspoly(A)+RNAselectionusingoligo-(dT)25beadstobothobtainsufficientamountsofRNA(50–200 ng)forlibrarysynthesisanddepletethesamplefromabundantncRNAs(rRNA),whichimpedesequencingduetostrongsecondarystructuresblockingthenanopores.However,suchanapproachcandiscriminateagainstnon-oroligoadenylatedtranscripts,whichwesoughttoquantify.Analysisofpreviouslypublishedshort-readRNA-seqdatasets31,32indicatedthatthenon-adenylatedmRNAsfractionisslimasinvitropolyadenylationpriorto(dT)-primingledtoastrongstabilizationofsn-/snoRNAandotherpoly(A)−ncRNAs,buthadlittleeffectonthelevelsofmRNAs(SupplementalFig. 1f).ToapproximatethelevelsofoligoadenylatedmRNAs,thatcouldbelostduringthepoly(A)+enrichment,wesubjectedatotalRNAsampletoDRS,whichisequivalentto(dT)-primingforRNA-seq.Thiswaspossibleasanalysisofartificialspike-insindicatedthattheDRSchemistryanddataanalysisarecompatiblewitholigoadenylatedtranscripts(SupplementalFig. 1g).Poly(A)+andtotalDRSdatasetsstronglycorrelatedwithoneanotherintermsofmRNAabundanceandpoly(A)taillength(Fig. 1b),butthetotalRNAdatasetwasenrichedinreadsbearingtailsmostlyshorterthan10As(SupplementalFig. 1h).Wefoundthathighlyexpressedtranscriptswereespeciallyenrichedwiththoseshort-tailedRNAsleadingtothedecreaseinmeanpoly(A)taillengthestimationandaslightincreaseinabundancecomparedtopoly(A)+DRSset(SupplementalFig. 1i),whichwascross-comparedandvalidatedbyreverse-transcriptionqPCR(SupplementalFig. 1j).Collectively,thisindicatesthateachmRNAhasanindividualfractionofoligoadenylatedspecies.However,weonlyfound840mRNAswithashort-tailedfractionlargeenoughtostronglyaffectthemeanpoly(A)tailandabundanceestimations(SupplementaryData 1).Consequently,weworkedwiththepoly(A)+fractionforfurtheranalysisasitprovedtobeagoodapproximationforthetotalsample.mRNApoly(A)taillengthsanticorrelatewithtranscriptabundanceandarenotlinkedwithtranscriptionortranslationratesUndersteady-state30 °Cor25 °Cgrowthconditions,poly(A)+RNAisolatedfromWTBY4741andW303yeaststrainshadmeanandmedianpoly(A)tailsof40and37 As,respectively.However,thepoly(A)taillengthdistributionvariedsignificantlybetweenindividualtranscripts(Fig. 1c).HighlyexpressedmRNAshadrelativelyshorttailswithameanof30 Asandamedianof26 As,reflectingsimilartendenciesinotherorganisms30,33.Incontrast,poly(A)taillengthsofalltranscripts,bothmRNAsandncRNAs,withlowerexpressionvariedfrom30to60 As.Suchheterogeneityofpoly(A)taillengthsincreasedgraduallywithdecreasingmRNAabundance.TopinpointthefactorbestdefiningthemeanmRNApoly(A)taillengthatsteadystate,wecross-comparedourDRSdatasetstopublisheddatadescribingkeyaspectsoftranscriptbiogenesisandfunction.mRNAsynthesisrateestimatesbestdefinedDRSmRNAabundancebutpoorlycorrelatedwithmeanpoly(A)taillengths34,35,36(Fig. 1d).Atsteadystate,themeanpoly(A)taildidnotcorrelatewithmRNAhalf-liveseither,althoughsomemRNAswithshortpoly(A)tails,includinghighlyexpressedones,weremoreabundantinthelonghalf-lifemRNAsgroup34,36(Fig. 1e).ThenumberofproteinmoleculespercellcorrelatedstronglywithmRNAabundanceandanticorrelatedwithmeanpoly(A)taillength37(Fig. 1f,toprow).Thelattercouldbeasignofco-translationalmRNAdeadenylation.However,whenconsideringtranslationalefficiency,thisinterdependencewasvirtuallylost38(Fig. 1f,middleandbottomrows).Giventhoseresults,wehypothesizethatmRNApoly(A)taillengthsarelikelytobedefinedbyexonucleolysis,atleastinpartindependentlyoftranslation.Pap1producespoly(A)tailsofmostly50 Aswiththelimitof200 As,andTrf4canpartiallycompensateforitsdepletionTodefinetheextentofexonucleolysis,wefirstsoughttodeterminethelengthofdenovo producedpoly(A)tails.Poly(A)taillengthatsteadystateisafunctionofnuclearpolyadenylationandpresumablymostlycytoplasmicdeadenylation.Therefore,tominimizetheimpactofdeadenylationonpoly(A)taillengthofnewlymademRNAs,weperturbednuclearexport.Aspreviouslyshowninmft1Δcells,exportisimpairedathightemperatures,whereasincellsdepletedforMex67,usingtheanchor-away(AA)system,exportisentirelyblocked32,39,40(Fig. 2a).Werapidlyshiftedcellsfrom25to38 °Cfor12–15 min,allowingdenovoproductionofasetofheat-stressmRNAs,whichinbothexport-blockedstrainshadpoly(A)tails50Aslongonaverage(47 Aand51 Aformft1ΔandMex67-AA,respectively,Fig. 2aandSupplementalFig. 2a).ThemaximumupperlimitofmRNApoly(A)taillengthswasroughly200AsinMex67-depletedcellsand163inmft1Δ,comparedto140inbothcontrols(Fig. 2a,bandSupplementalFig. 2a).Notethatinbothgrowthconditions(richorminimalmediausedformft1ΔandMex67-AA,respectively),denovoPap1producedpoly(A)tailsweresimilarinlengthdespiteageneralsteady-stateshorteningofpoly(A)taillengthsinminimalmedia(seealsothelastsection).Fig.2:Pap1producespoly(A)tailsofmostly50adenosineswiththelimitof200adenosines,andTrf4canpartiallycompensateforitsdepletion.aThetoppanelshowsabeeswarmplotforpoly(A)taillengthdistributionofacollectionofheat-inducedtranscriptsincontrolandMex67-depletedcells.Meansand99.5%quantilearemarkedwithsolidanddashedlines,respectively,forMex67-AAwithorwithoutrapamycinat38 °C.ThebottompanelshowslibrarysizenormalizedcountsforthosemRNAs.bEmpiricalcumulativedensityfunctionplotforpoly(A)taillengthsofheat-inducedtranscriptsincontrol(Mex67-AA-rapamycin:WT38 °C)andMex67-depletedcells(+rapamycin).Dashedlinesrepresent99.5%quantileforcontrol38 °C(green)andMex67-AA38 °C(red).cGrowthofWT,pap1-1,pap1-1rrp6Δ,andpap1-1trf4Δstrains.EqualnumbersofcellsofeachstrainwerespottedonYPDAplatesintenfoldserialdilutionsandgrownfor3daysat24or28.5 °C.dViolinplotofmeanmRNApoly(A)taillengths(top)andmRNAabundance(bottom)ofall(blueseries)andhighlyexpressedmRNAs(redseriesn = 184)forWT(n = 5079),pap1-1(n = 5161),pap1-1rrp6Δ(n = 5054),andpap1-1trf4Δ(n = 5770)strains.Fortheupperpanel,Kruskal–WallistestPvalueandHolm-correctedPvaluesfortwo-sidedDunnpairwisetestareshown.DashedgraylinesshowthemeanvaluesforWTstraintoeasevisualcomparison.Boxplotsshowmedianvalue(solidboldline)and1stand3rdquartiles,whiskersrepresent1.5 × IQR(interquartilerange).Outliersaremarkedasblackpoints.e–gScatterplotsoftherelationshipbetweenlog2foldchangeinmRNAabundance(yaxis)andabsolutechangeinmeanmRNApoly(A)tailforpap1-1(e),pap1-1rrp6Δ(f),andpap1-1trf4Δ(g)cells(xaxis),comparedwithWT.Dashedlinesmarkatwofoldchangeinexpressionlevelanda5-adenosinechangeofmeanpoly(A)taillength.hPoly(A)taillengthdistributionofTDH3andTDH1inpap1-1mutantcells(upperpanel)andrelativeabundancechangesforeachcondition(bottom).ThenumberofDRSreadsforeachconditionisshownforeachtranscript.Reddashedlinesshowthemeanforeachcondition.Pearson’scorrelationcoefficientsareindicatedoneachscatterplot.FullsizeimageToevaluatetheroleofthenuclearpoly(A)polymeraseTrf4inthebiogenesisofmRNApoly(A)tails,weinhibitedthePap1activityusingathermosensitivepap1-1mutant41.Aspreviouslyshown,deletionofthenuclearexosomesubunitRRP6inpap1-1partiallysuppressedthe28.5 °Cthermosensitivityofpap1-1cells(Fig. 2c).ItalsorestoredthelevelsofsixarbitrarilytestedtranscriptstothoseobservedinaWTstrain(SupplementalFig. 2b).Wehypothesizedthatpartialexosomeinactivationinpap1-1setthespacefornon-canonicalpoly(A)polymerasestoadenylatemRNAsandrestoretheirfunctionality.Indeed,thelossofTRF4exacerbatedthegrowthdefectofpap1-1cellsanddidnotsupportmRNAstability41(Fig. 2candSupplementalFig. 2b).Wethussequencedpoly(A)+RNAfromrelevantstrainsandusedareverse-transcriptionquantitativepolymerasechainreaction(qPCR)-derivedcoefficient(RT-qPCR)tonormalizetheDRSdatasets.Thiswasrequiredbecausewenoticedthatamarkeddecreaseinpoly(A)+RNAyieldfrompap1-1andpap1-1trf4ΔcellsresultedinthelossofadifferenceinmRNAabundancescomparedtoWTcellswhenonlylibrarysizenormalizationwasemployed(SupplementalFig. 2c,d).Inthepap1-1mutant,mRNApoly(A)tailshorteningwasproportionaltothepoly(A)taillengthincontrolcells(SupplementalFig. 2e),suggestingthatthemutationimpairspolymeraseprocessivity.mRNApoly(A)tailsweresignificantlyshortenedby6.86,3.73,and7.66Asonaverageinthepap1-1,pap1-1rrp6Δ,andpap1-1trf4Δbackgrounds,respectively(Fig. 2d,blueseries).Thus,lossofRRP6ledtoapartialrescueofgrowth,mRNAabundance,andpoly(A)taillengthinpap1-1background.Sincenosucheffectwasseenintrf4Δ,weconcludethatthemechanismofrescueislikelymostlyrelatedtoTrf4/5-mediatedpolyadenylation,whichisonlyvisibleinexosome-deficientbackgrounds42.Previousstudiessuggestedalinkbetweendeficientadenylationanddecayofthetranscript41.Thoughthiswasconfirmedinouranalysisbytheabsolutelevelofpoly(A)+RNA,especiallyforhighlyexpressedmRNAs,wefoundnoglobalcorrelationbetweenabsolutechangesinmeanpoly(A)taillengthandfoldchangesinRNAabundanceforthepap1-1,pap1-1rrp6Δ,andpap1-1trf4Δbackgrounds(Fig. 2e–g,forsingle-geneexamplesseeFig. 2h,andSupplementalFig. 2f).ThelackofsuchcorrelationindicatesthatsomemRNAsareadenylatedcorrectlyinpap1-1,andthesteady-statemRNAabundance/poly(A)taillengthpatternresultsfromothernuclearorcytoplasmiccompensatorymechanisms.ncRNAspossesslongpoly(A)tails,andTRAMP5assistsRrp6inmRNApoly(A)tailtrimmingToevaluatethedirectimpactofthenuclearexosomeandTRAMP4/5onpoly(A)tailbiogenesisandthestabilityofdifferentRNAbiotypes,weproducedDRSdatasetsforWT,trf4Δ,trf5Δ,andrrp6Δstrainsgrownat30 °C.WefirstinspectednuclearncRNAs,whicharethepredominantexosometargets.Asexpected,thesumofreadsmappingtoCUT,sn-/snoRNA,andrRNAlociwassignificantlyincreasedintherrp6ΔdatasetcomparedwithWT(Fig. 3a,bottompanel).Increasedadenylationofusuallynon-polyadenylatedncRNAs,suchas25SrRNA,SCR1,andLSR1,waspreviouslyreported28,43andvalidatedbyRT-qPCR(SupplementalFig. 3a).Importantly,foldstabilizationofCUTsinourDRSdatasetswascomparabletopreviousreports28(SupplementalFig. 3b),suggestingthatdespitetheoligo-(dT)enrichment,wedetectedthebulkofthosencRNAs.EventhoughTrf4/5areexpectedtooligoadenylatencRNAs7,8,tooursurprise,poly(A)tailsofCUTs,sn-/snoRNAs,andrRNAswereonaverage45,54,and52Aslonginrrp6Δcells.SuchlongtailswerealsodetectedintheWTstrain(Fig. 3a,toppanel).ObservedncRNAadenylationwaspartiallydependentonTrf4,manifestedbybothmeanpoly(A)taillengthandabundanceinthetrf4Δandrrp6Δtrf4Δdatasets6(Fig. 3a).WevalidatedtheabundanceandadenylationpatternofsnR13inthepoly(A)+fractionbyNorthernblotanalysis(SupplementalFig. 3c,comparelefttorightpanel).ConsistentwiththeroleofTRAMP5inrRNAmaturation5,weobservedadecreaseintheabundanceandpoly(A)taillengthofmanyrRNAprecursorsintrf5Δcells(Fig. 3a).ThisrecapitulatedtheknownfunctionsoftheexosomeandTRAMP4inncRNAdegradationandstronglysuggestedthatTRAMP4/5producepoly(A)tailscomparableinlengthtomRNAs.Fig.3:mRNApoly(A)tailtrimmingbytheexosomeisstimulatedbyTrf5.aThetoppanelshowsaviolinplotofmeanpoly(A)taillengthsofnuclearncRNAsdetectedinWTandexosomemutantcells.ThenumberofRNAsdetectedn = [WT,trf5∆,rrp6∆,trf4∆,rrp6∆trf4∆]wasforCUTs = [240,129,519,421];sn/snoRNAs = [47,40,63,53,56]andrRNAs = [15,12,23,14,23].ThebottompanelshowsthesumofallDRScountsthatmaptothosencRNAloci,normalizedtothesumofcountsthatweremappedtoprotein-codingloci.Fortheupperpanel,Kruskal–WallistestPvalueandHolm-correctedPvaluesfortwo-sidedDunnpairwisetestareshown.DashedgraylinesshowthemeanvaluesforWTstraintoeasevisualcomparison.Boxplotsshowmedianvalue(solidboldline)and1stand3ndquartiles,whiskersrepresent1.5 × IQR(interquartilerange).Outliersaremarkedasblackpoints.bViolinplotofthemeanmRNApoly(A)taillengthdistributionandlog2-scaledmRNAabundanceinthetopandbottompanels,respectively,forWT(n = 5269),trf5Δ(n = 5009),rrp6Δ(n = 4951),trf4Δ(n = 5258),andrrp6Δtrf4Δ(n = 4928)strains(highlyexpressedmRNAsn = 184).Fortheupperpanel,Kruskal–WallistestPvalueandHolm-correctedPvaluesfortwo-sidedDunnpairwisetestareshown.DashedgraylinesshowthemeanvaluesforWTstraintoeasevisualcomparison.Boxplotsshowmedianvalue(solidboldline)and1stand3ndquartiles,whiskersrepresent1.5 × IQR(interquartilerange).Outliersaremarkedasblackpoints.cRelationshipbetweenabsolutechangeinmeanmRNApoly(A)taillengthandfoldchangeinmRNAabundancefor(fromlefttoright)rrp6Δ,trf4Δ,andtrf5ΔcomparedwithWTand(rightmost)rrp6Δtrf4Δcellscomparedwithrrp6Δ.mRNAswithpoly(A)tailsthatwerelongerbyatleasttenadenosinesinrrp6Δarehighlightedingreenintherrp6Δandrrp6Δtrf4Δplots.Dashedlinesmarkatwofoldchangeinexpressionand5-adenosinechangeofmeanpoly(A)taillength.Pearson’scorrelationcoefficientsareindicated.dTherightmostpanelshowstheNorthernblotanalysisofTDH3andRPL18BmRNA3′-UTRsthatwerecleavedoffviaRNaseHdigestionandrunonahigh-resolutionacrylamidegelforWT,rrp6Δ,trf4Δ,andtrf5Δcells.ThemiddlepanelshowstheNorthernsignaldistribution,averagedfromtwobiologicalreplicates.Theleftmostpanelshowsthepoly(A)taillengthdistributionintheDRSdatasets.FullsizeimageThedeletionofRRP6oritscofactorsTRF4orTRF5didnotaffecttheabundanceofprotein-codinggenes.However,poly(A)tailswereelongatedbyanaverageof3.4and4 Asinrrp6Δandtrf5Δcells,respectively,remainingonaverageunalteredintrf4Δcells(meanchange = 0.5adenosines;Fig. 3b,c,SupplementalFig. 3d,e).MostaffectedmRNAswereexpressedatlowormoderatelevelsandhadmedium-longpoly(A)tailinWTcells(SupplementalFig. 3f,g).ThissuggestsaroleofTRAMP5inmRNApoly(A)tailtrimmingbytheexosome.However,therelativelymodestcorrelationofpoly(A)tailchangesinrrp6Δandtrf5Δcells(SupplementalFig. 3e)mightindicatethatTRAMP5canalsoactasacofactoroftheotherexosomenucleaseDis3.BasedonthetotalRNAfractionandNorthernblotanalysis,wevalidatedchangesinthepoly(A)taillengthofRPL18BmRNA,whichhadelongatedpoly(A)tailsinalldeletionstrains,andTDH3mRNA,whichremainedlargelyunaffected(Fig. 3d).Finally,weinvestigatedwhethersomemRNApoly(A)tailelongationinrrp6ΔcellscanbeattributedtoTRAMP4-mediatedadenylationassessingpoly(A)taillengthinadoublerrp6Δtrf4Δstraincomparedwithrrp6Δ(Fig. 3candSupplementalFig. 3h–j).ManymRNAs,whichhadpoly(A)tailsthatwereatleast10Aslongerintherrp6Δstrain(greendots),tendedtobeshortenedinthedoublerrp6Δtrf4Δstrain,indicatingthatTRAMP4indeedpolyadenylatedthem.However,poly(A)tailsofmRNAswerestillgloballyelongatedinrrp6Δtrf4Δstrainbyanaverageof2.3AscomparedtoWT(Fig. 3b).Thissuggeststhatpoly(A)tailelongationinrrp6ΔisnotonlyrelatedtoactivationofTrf4-polymeraseactivity.Overall,theseresultsindicatethatthelengthofpoly(A)tailsofsomemRNAsisregulatedbyRrp6,mainlyresultingfromtheTRAMP5-enhancedexonucleolytictrimmingofpoly(A)tails.However,adenylationofafewmRNAsmaybeperformedbyTRAMP4.Dis3andRrp6bothmediateextensivemRNApoly(A)tailtrimminganddegradationRrp6,incontrasttoDis3,confersonlyoneexosomenucleolyticactivitythatisnotessential.Todeterminethefullextentofexosomeactivity,wedepletedDis3orDis3andRrp6simultaneouslyfromthenucleusfor40 minat30 °CusingtheAAsystem39,which,asexpected,stabilizedtheNEL025cCUT(SupplementalFig. 4a,greenseries).Accordingly,ourDRSdatasetrevealedhigherlevelsofCUTsandsnRNAandrRNAprecursorsinDIS3-AAandDIS3-/RRP6-AAcells.Inthedouble-depletedstrain,CUTshadpoly(A)tailsof56AsandsnRNAsandrRNAstailsof63As(Fig. 4a).AtsomesnRNAloci,transcriptionterminationcanbemediatedbybothCPFandNNS44.Thus,theselongpoly(A)tailsmaybeattributabletoPap1activity.However,wefoundthatpoly(A)tailsofnuclearncRNAswere~37 Asinthepap1-1rrp6Δstrain(Fig. 4a),confirmingthatTRAMPscanproducepoly(A)tailsthatarecomparableinlengthstomRNAs.Fig.4:Dis3andRrp6mediateextensivemRNApoly(A)tailtrimming.aViolinplotofmeanpoly(A)taillengthsofCUTs,snRNAs,andrRNAsdetectedinWT,DIS3-AA,DIS3-/RRP6-AA,andpap1-1rrp6Δcells(top),andthesumofallDRScountsthatmaptothosencRNAloci,normalizedtothesumofcountsmappedtoprotein-codingloci(bottom).ThenumberofRNAsdetectedn = [WT,Dis3-AA,Dis3-/Rrp6-AA,pap1-1rrp6∆]wasforCUTs = [143,436,579,434],sn/snoRNAs = [36,52,59,59],andrRNAs = [17,23,24,23].Fortheupperpanel,Kruskal–WallistestPvalueandHolm-correctedPvaluesfortwo-sidedDunnpairwisetestareshown.DashedgraylinesshowthemeanvaluesforWTstraintoeasevisualcomparison.Boxplotsshowmedianvalue(solidboldline)and1stand3rdquartiles,whiskersrepresent1.5 × IQR(interquartilerange).Outliersaremarkedasblackpoints.bMeanmRNApoly(A)taillengths(toppanel),andlog2mRNAabundance(bottom)forWT(n = 5079),DIS3-AA(n = 5035),DIS3/RRP6-AA(n = 5109)andski2Δ(n = 5081)strains.DataforhighlyexpressedmRNAsareshowninred.Fortheupperpanel,Kruskal–WallistestPvalueandHolm-correctedPvaluesfortwo-sidedDunnpairwisetestareshown.DashedgraylinesshowthemeanvaluesforWTstraintoeasevisualcomparison.Boxplotsshowmedianvalue(solidboldline)and1stand3rdquartiles,whiskersrepresent1.5 × IQR(interquartilerange).Outliersaremarkedasblackpoints.cRelationshipbetweenabsolutechangeinmeanpoly(A)taillengthbetweenDIS3-andDIS3-/RRP6-AAcellscomparedwithWT.Bluelinesdesignatethemeanpoly(A)taillengthchangeineachstrain.Dashedlinesmarka10-adenosinechangeofmeanpoly(A)taillength.dRelationshipbetweenabsolutechangeinmeanmRNApoly(A)taillengthandlog2fold-changemRNAabundanceforDIS3-AA,DIS3-/RRP6-AA,andski2Δcells,respectively,comparedwithWT.Dashedlinesshowatwofoldchangeinexpressionanda5-adenosineschangeofpoly(A)taillength.eViolinplotsofthepoly(A)taillengthdistributionofDRSreadsforTDH3andRPL18BmRNAsforWT,DIS3-AA,andDIS3-/RRP6-AA.ThenumberofDRSreadsforeachconditionisshownforeachtranscript.Reddashedlinesshowthemeanforeachcondition.Pearson’scorrelationcoefficientsareindicatedoneachscatterplot.FullsizeimageThemeanmRNApoly(A)taillengthincreasedfrom40AsinWTto47and56AsinDIS3-AAandDIS3-/RRP6-AAcells,respectively(Fig. 4b,c).ThevastmajorityofmoderatelytominimallyexpressedmRNAswereaffected,regardlessoftheirpoly(A)taillengthintheWTbackground.Incontrast,poly(A)taillengthofhighlyexpressedmRNAswaslessaffected(Fig. 4b,dandSupplementalFig. 4b–e).Accordingly,thepoly(A)taillengthofTDH3mRNAwasmostlyunchangedinDIS3-AAandDIS3-/RRP6-AAcells,similarlytotherrp6Δbackground.Whilealldepletionsmarkedlyincreasedthepoly(A)taillengthofRPL18BmRNA(Fig. 4eandotherexamplesSupplementalFig. 4f),thatchangeoccurredwithoutmajoralterationsoftheefficiencyofpoly(A)+RNAextraction(SupplementalFig. 4a,right).AlargeproportionofmRNAswasupregulatedinDis3-andDis3/Rrp6-depletedcells,butagaintranscriptstabilizationdidnotcorrelatewithpoly(A)taillengthchange(Fig. 4d).Sincethelossofacytoplasmicexosomecofactor,Ski2helicase,didnotleadtosubstantialmeanpoly(A)taillengthchangesbutonlytominoralterationsinmRNAstability(Fig. 4b,dandSupplementalFig. 4g,h),weconcludethatDis3andRrp6regulatemRNAstabilityandpoly(A)taillength,mainlyinthenucleus.PAN2/3andCCR4-NOTcytoplasmicdeadenylasestargetbothdistinctandoverlappingsetsoftranscriptsCytoplasmicmRNAdecayisregulatedbyCCR4-NOTandPAN2/3deadenylasecomplexes.Wesequencedpoly(A)+RNAfromstrainsdeletedforthecatalyticsubunitsofeachcomplex,ccr4Δandpan2Δ.AsubstantialelongationofmRNApoly(A)tailswasobservedinpan2Δandccr4Δbackgroundsbyanaverageof7.17and6.66As,respectively,withnomajorchangeinexpressionlevels(Fig. 5aandSupplementalFig. 5a,b).Surprisingly,meanpoly(A)tailchangesinbothstrainswerepoorlycorrelated,indicatingthatPAN2/3andCCR4-NOTactivitiesaredirectedtowardsdifferentsetsofsubstrates(Fig. 5b).Accordingly,Pan2appearedtotargetmRNAsofhighandmediumabundancepreferentially,anditsimpactonpoly(A)taillengthdecreasedtogetherwithtranscriptexpressionlevels.TheoppositetrendwasobservedforCcr4(Fig. 5candSupplementalFig. 5a,b).GiventhathighlyexpressedmRNAsconsistedof~60%ofallmRNAmoleculesinthelibrary(SupplementalFig. 5c),atsteady-statePAN2/3appearstobethemaincellulardeadenylasewiththelowesttranscriptspecificity,whereasCcr4ismoretarget-specific.Tofurthervalidateourgeneralconclusions,weinspectedpolyadenylationprofilesofselectedmRNAs.AspreviouslyshownbyLM-PAT,thepoly(A)taillengthsofthreewell-expressedmRNAs,involvedinseptinorganization(CDC42,APQ12,andSHS1),dependedonCCR4-NOTand/orPAN2/345(SupplementalFig. 5d).InourDRSdataset,thehighlyexpressedTDH3,RPS28B,andRPP1BmRNAswereexclusivelytargetsofPan2(SupplementalFig. 5e),whileHTA146,ALD5,andUTP22mRNAswereCcr4-specific(SupplementalFig. 5fandforotherexamples,seeSupplementalFig. 5g).Fig.5:PAN2/3andCCR4-NOTcytoplasmicdeadenylasecomplexescantargetdistinctsetsoftranscripts.aRelationshipbetweenlog2foldchangeinmRNAabundanceandabsolutechangeinmeanpoly(A)taillengthforpan2Δ(top)andccr4Δ(bottom)strainsrelativetoWT.Dashedlinesshowatwofoldchangeinexpressionanda5-adenosinechangeofmeanpoly(A)taillength.bRelationshipbetweenabsolutechangeinpoly(A)taillengthinccr4ΔcellsrelativetoWTcomparedwithanabsolutechangeinpoly(A)taillengthrelativetoWTforpan2Δ.Bluelinesdesignatethemeanpoly(A)taillengthchangeineachstrain.Dashedlinesmarka10-adenosinechangeofmeanpoly(A)taillength.cTheviolinplotinthetoppanelshowsthedistributionofmeanpoly(A)taillengthsofdetectedmRNAs,binnedbyexpressionlevelinWT.Thebottompanelshowslog2-scaledmRNAabundance.ThesumofcountsfromeachmRNAcategoryisindicated.StatisticalsignificanceforeachexpressiongroupwascalculatedusingtheKruskal–WallistestPvalueandHolm-correctedPvaluesforthetwo-sidedDunnpairwisetest.Boxplotsshowmedianvalue(solidboldline)and1stand3rdquartiles,whiskersrepresent1.5 × IQR(interquartilerange).Outliersaremarkedasblackpoints.dPiechartshowinginredthefractionofcellularmRNAsthatweredeadenylatedbyonlytheexosome(1643),onlyCCR4-NOTand/orPAN2/3(625),orallthreecomplexes(1588).Outofatotalof6230annotatedmRNAs,4665werereproduciblydetected.TranscriptsthatwereunaffectedorcouldnotbedetectedbyDRSarehighlightedingray.FullsizeimageFinally,weinspectedthenucleasesensitivityofallmRNApoly(A)tailsandfoundthat61.8%ofallannotatedmRNAsand82.6%ofmRNAsthatcouldbedetectedbyDRSweresubstratestoeithernuclearand/orcytoplasmicdeadenylation(Fig. 5d).TheseresultscollectivelyindicatethattheregulationofmRNApoly(A)taillengthbyexonucleolysisisamultistep,substrate-specificprocessthatoccursinthenuclearaswellascytoplasmiccompartments.Transcriptpoly(A)taillengthstronglydependsongrowthconditionsGrowthconditionsalterboththeratesofmRNAtranscriptionanddecay47.Wecomparedpoly(A)taillengthprofilesofcellsgrowninrichmediaatasteadystateat25and30 °Cwithcellssubjectedtoheatstressat38 °Cfor12 minor37 °Cfor1 h(Fig. 6aandSupplementalFig. 6a–c).At30 °Cand25 °C,mRNAabundanceandpoly(A)taillengthwerecomparable.Incontrast,12 minofheatstressdecreasedtheabundanceandshortenedpoly(A)tailsofhighlyexpressedmRNAs.Curiously,1 hofincubationat37 °CincreasedthemRNAabundanceofhighlyandmoderatelyexpressedmRNAsandgloballyelongatedpoly(A)tailscomparedwith25–30 °Cconditions.Toexaminesomefunctionalgroups,weselectedmRNAsthatwereeitherup-ordownregulatedbyheatstressmorethanfivefold(SupplementaryData 1,Fig. 6bandSupplementalFig. 6a–d).Weobservedthatupregulationwasaccompaniedbypoly(A)tailselongation,whereasdownregulatedtranscriptshad,onaverage,shortertails.Atasteadystateof30 °C,similarpoly(A)lengthshiftswereobserved,althoughexpressionlevelswereunchangedwhencomparedwith25 °C.Finally,downregulatedmRNAsseemtostronglyrelyonexonucleolysisfortheirregulationofabundancesince88%ofthemweretargetsofeithertheexosomeordeadenylases,comparedto34%inthecaseofupregulatedones(Fig. 6c).Fig.6:Transcriptpoly(A)taillengthdependsongrowthconditions.aViolinplotsinthetoppanelshowthemeanpoly(A)taillengthdistributionacrossmRNAsfromwild-typecellsthatweregrownatdifferenttemperatures,binnedbyexpressionlevelinWT.ThebottompanelshowsthefractionofmRNAsDRScountsforeachexpressionbininagivencondition.Boxplotsshowmedianvalue(solidboldline)and1stand3rdquartiles,whiskersrepresent1.5 × IQR(interquartilerange).Outliersaremarkedasblackpoints.bSameas(a)butformRNAsthatwereatleastfivefoldupregulated(orangeseries)ordownregulated(greenseries)followingheatshock.Boxplotsshowmedianvalue(solidboldline)and1stand3rdquartiles,whiskersrepresent1.5 × IQR(interquartilerange).Outliersaremarkedasblackpoints.cPiechartsshowingthenucleasesensitivityofheat-regulatedmRNAs,whicharelistedinSupplementaryData 1.dRelationshipbetweenmRNAabundanceandmeanpoly(A)taillengthforcellsthatweregrowninrichmedia(black,sameasFig. 1a)andMM-URA(blue)at25 °Casexplainedinthemaintext.FullsizeimageToevaluatetheimpactofnutrientcontentonmRNAandpoly(A)tailmetabolism,wecomparedDRSdatasetsfromcellsgrowninarichmediumcontainingglucoseasthecarbonsourcewiththosefromaminimalmediumthatcontainedraffinoseandgalactose(Fig. 6d).At25 °Cinminimalmedium,mRNAs,regardlessofabundance,generallyharboredshortpoly(A)tails,sotheanti-correlationbetweenthelengthsofpoly(A)tailsandexpressionlevelsevidentinrichmediawasvirtuallylost(Fig. 6d,compareblackandblueseriesandSupplementalFig. 6e).ThiseffectwasnotattributabletoadecreaseinPap1activitybecausepoly(A)tailsofnewlyproducedmRNAshadacomparablesize,regardlessofgrowthconditions(Fig. 2a,bandSupplementalFig. 2a).HighlyexpressedmRNAsconstitutedasmallerfractionofallmRNAcountsincellsgrownintheminimalmediumcomparedtotherichmedium(SupplementalFig. 6e).TheabundanceofthesemRNAsfurtherdecreasedafterheatshockinaminimalmedium,albeitwithoutastrongimpactonpoly(A)taillength.Collectively,growthconditions,especiallythenutrientsource,hadamajorimpactonmRNAexpressionandpoly(A)taillengthprofiles.DiscussionOurcomprehensiveDRSanalysisunveiledmechanismsofpost-transcriptionalmRNAandncRNApoly(A)tailbiogenesis,whichwasregulatedinresponsetogrowthconditions,anddemonstratedthatnuclearandcytoplasmicnucleasesplayanimportantroleinthisprocess(Fig. 7).Fig.7:mRNApoly(A)taillengthisextensivelyregulatedpost-transcriptionallyinboththenucleusandthecytoplasm.Fordetails,see“Discussion”.FullsizeimageThebiogenesisofncRNApoly(A)tailsremainedunclearbecausebothpolyadenylationanddegradationactinparallel.ThewidelyacceptedmodelpostulatedthatTRAMP4/5-mediatedoligoadenylationenhancesdecaybyrecruitingtheexosome/Rrp6complex,whereastheadditionofalongpoly(A)tailbyPap1promotesstabilityandexport.Thismodeldoesnotfullyalignwithourresults,whichshowedthatnuclearncRNAshavepoly(A)tailslengthscomparabletomRNAs,eveninPap1-impairedcells.TheTRAMP4/5,exosome,andNNScomplexesinteractbothfunctionallyandphysically1,48butdonotformalargestableassembly.Theactionofeachofthesecomplexesshouldinsteadbeconsideredinaprobabilisticmanner.TRAMP4/5recruitmentcanthusincreasethelikelihoodofdecaybutdoesnotassureit.Moreover,althoughTRAMPisnotaprocessiveenzyme7,8,itcanstillberecruitedforasecondorthirdroundofadenylationor,asshownbyourresults,targetPap1-adenylatedsubstrates.Furthermore,invitrostudiesshowedthatNNSRNA-bindingactivitystimulatesTRAMP4adenylationbystabilizingthecomplexonthesubstrate1,supportingthesynthesisoflongpoly(A)tails,atleastonNNStargets.Theadditionofalongpoly(A)tailcanberequiredtoinitiatedecayonhighlystructuredsubstrates,suchasrRNAprecursorsbecausethecoreexosomechannelleadingtotheDis3activesiteaccommodatesa25–30ntlongRNA49.Short,3–5-adenosinelong,poly(A)tailsoftendetectedbyRNA-seqatncRNAs3,9ratherrepresentatraceofongoingdecay.However,eachfull-lengthncRNAcanundergoaphaseofmultipleTRAMP-adenylationcyclesbeforedecayisinitiated.Therelativeamountofthesehighlypolyadenylatedspeciesisafunctionofthetimebetweentranscriptionterminationanddecayinitiation.ThenuclearbiogenesisofmRNApoly(A)tailsisseeminglynoncontroversial.Cleavageandpolyadenylationarecoupledandthoughttocooperatetightlywithnuclearexport17.Thus,mRNAsappeartoescapenucleardegradationandbesubjectonlytocytoplasmicdecay.Thisviewstemsfromtheprovocativegene-gatinghypothesisthatwasproposedbyBlobelin198550andproventrueonlyforsomemRNAsupregulatedinresponsetogrowthconditions51.Single-moleculetransportstudiesshowedthat,regardlessofgenelocalization,mRNAscandiffusefreelywithintheinterchromatinspaceforlongperiodsbeforeexport17,52.Dependingonthenuclearlocalizationofgenes,somemRNAscanthushavelongernucleardwellperiodsthatallowampletimetobecometargetsoftheexosomeanditsTRAMP4andTRAMP5cofactors.WefoundthatmRNAswithlowandmediumexpressionwerethemostprominentexosome/TRAMP4/5targets.WehypothesizethattheexportofthosemRNAsislessefficientthanhighlyexpressedtranscripts.Indeed,gene-gatedHSP104expressionisonlyregulatedbyTRAMP4/exosome-mediateddecayinexport-impairedmft1Δcells40.Wealsospeculatethatexosome-mediatedexonucleolysisandTRAMP4/5re-adenylationcouldinfluencetheexportrateofsomemRNAs,forwhichastrictlengthofthepoly(A)tailisrequired14(Fig. 7).Ourdatarevealedthatatsteadystate,boththeCCR4-NOTandPAN2/3cytoplasmicdeadenylasecomplexeshavealargenon-overlappingsetoftargetsandactredundantlyonsharedsubstratesrepresentingonly10.9%ofdetectedmRNAs.ThisrefinesthebiphasicmRNAdeadenylationmodelthatpostulatesthatinitialdeadenylationismediatedbyPAN2/3followedbyCCR4-NOT20,24.Ourfindingsfavorthenotionthatintheinitialphase,bothcomplexesactindependentlyaccordingtotheirsubstratespecificityandthatinitialdeadenylationisunlikelytobecoordinatedbetweenthesetwocomplexes.PAN2/3substraterecruitmentisnonspecificbecauseitreliesonpolyadenosineandPab1binding20,21,53.HighlyexpressedmRNAsconstitute~60%ofallprotein-codingmolecules,andthesearethemostcommonPAN2/3targets.Incontrast,theCCR4-NOTcomplex,inadditiontoPab1binding22,hasbeenshownindifferentorganismstobetargetedtoitssubstratesviaadaptorproteinsthatcanalsoacceleratedeadenylation54,55,56,thusmakingthisdeadenylasecomplexmorelikelytargetedtomRNAsoflowexpressionasevidencedbyourdatasets.CCR4-NOTalsodirectlystimulatesdecapping57,suggestingthatrecruitmentofthiscomplexismoretightlylinkedtosubsequentdecay.However,thisaspectcannotbefullyaddressedwithourdataduetoconstraintsindetectingnon-oroligoadenylatedRNAs.MethodsYeaststraincultureTheyeaststrainsthatwereusedinthisstudyarelistedinSupplementalTable 1.TheyeaststrainsweregrowninYPDAmedia,withtheexceptionofMEX67-AAstrainsthatcontainedURA-plasmids,whichweregrowninminimalmediathatlackeduracilwith2%raffinoseand2%galactose.Allofthecultureswereharvestedinthelogarithmicgrowthphase.Toperformheatshock,theW303WT,mft1Δ,andMEX67-AAstrainswerepre-grownat25 °C.Thecultureswerecomplementedwithanequalvolumeofmediathatwaspreheatedto51 °Candincubatedat38 °Cfor12or15 minformft1ΔandMEX67-AAseries,respectively.Afterward,thecellswereinactivatedbyaddinganequalvolumeof96%ethanolthatwascooledto−20 °C.PreculturesfortheW303pap1-1serieswerepreparedat25 °C.Thecellswerethendilutedandshiftedto30 °Cfor20 h.Thedeadenylasemutantstrainsccr4Δ,pop2Δ,andpan2ΔmutanttogetherwithaWTstrain,weregrowncontinuouslyat30 °C.Therrp6Δ,trf4Δ,trf5Δ,andrrp6Δtrf4ΔexosomemutantstogetherwithWTstrainsweregrownat30 °C.TheDIS3-AA,DIS3-/RRP6-AA,andski2Δweregrownat30 °Candtheanchor-awaystrainsweresupplementedwithrapamycin(CaymanChemicalscat.no.13346)toafinalconcentrationof1 μg/mlfor40 minbeforeharvest.AWTW303strainwaspre-grownat25 °Candthentransferredfor1 hofgrowthinawaterbathwithshakingat37 °C.RNApreparation,reversetranscription,andqPCRanalysisRNAfromyeastcellswasextractedusingthehotacidphenolmethod.Cellpelletsfrom10to30 mlculturesatOD0.3to0.6wereresuspendedin400–600 μlofTESbuffer(10 mMTris[pH7.5],5 mMethylenediaminetetraaceticacid,and1%sodiumdodecylsulfate[SDS])andcomplementedwithanequalvolumeofacidphenol(catalogno.P4682,Sigma-Aldrich).Thesampleswereincubatedat65 °Cwithvortexingintworoundsfor40and25 min.Betweenandafterincubations,thesampleswerecentrifugedat20,000 × gfor10 minat4 °C.Theresultingsupernatantwasvortexedwithanequalvolumeofchloroform(catalogno.112344306,Chempur)andcentrifugedagain.TheRNAsfromthesupernatantwereprecipitatedusing96%ethanolanda65 mMfinalconcentrationofLiCl.YeastpolyadenylatedRNAforDRSlibraryconstructionandreverse-transcriptionqPCRanalysiswasextractedfrom35to75 μgtotalRNAusingtheThermoFisherdynabeadsoligo-(dT)25kit(catalogno.61002or61005)accordingtothemanufacturer’sprotocol.Inbrief,35 μgofRNAwasresuspendedin50 μlofwaterandmixedwith50 μlofBindingBuffer(20 mMTris-HCl,pH7.5,1.0 MLiCl,2 mMEDTA).Thissamplewasdenaturedfor2 minat65 °Candthenputonicefor2 min.Thenthesamplewasmixedwith50 μlofbeadspre-washedinBindingBuffer(slurrybeadsvolumeperonesamplewas100 μl).Thesampleswereincubated15 minatroomtemperatureandthenwashedtwotimeswith100 μlofWashBuffer(10 mMTris-HCl,pH7.5,0.15 MLiCl,1 mMEDTA10 mMTris-HCl,pH7.5).Bufferwasremovedfromdynabeads,whichwerethenresuspendedin10-12 μlofwaterandeluted2 minat80 °C.ThetotalRNAfractionwastreatedwithDNAse(Turbo-DNase,Ambion)accordingtothemanufacturer’sprotocol.Totalandpoly(A)+fractionswereusedforreversetranscriptionwith(dT)18andrandomhexamersusingSuperScriptIII(Invitrogen,cat.no.18080093)asspecifiedinthemanufacturer’sprotocol.cDNAwasdiluted10–20timesandusedforqPCRanalysiswiththePlatinumSYBRGreenqPCRSuper-Mix-UDGkit(Invitrogen,cat.no.11733-046)andLightCycler480(Roche).ThelistofprimersthatwereusedforqPCRanalysisisspecifiedinSupplementalTable 2.TherelativecDNAconcentrationwascalculatedwiththe2ndderivativemaximummethodandnormalizedforquantitiesanddilutionsthatwereusedineachexperimentseparately.Eachreactionwascarriedwithtwoorthreebiologicalreplicates,eachmeasuredinduplicate.AnalysisofqPCRresultsisdepositedatMendeleydata(https://doi.org/10.17632/v5vm3dmm8y.1).Preparationofstandardswithpredefinedpoly(A)lengthsTemplatesforspike-inswerepreparedintwoconsecutivePCRreactions.First,an827-bpfragmentofRenillaluciferasefrompRL5BoxplasmidwasamplifiedusingRLucF1andRLucR1primers.ThepurifiedampliconwasusedasatemplateinthesecondPCRreactionwithRLuc_T7_F2primercontainingtheT7promotersequence,andbackwardprimerRLuc_Ax_R2introducingpoly(A)tailofadefinedlength(from10to90As).TheresultingPCRproductwasanalyzedandpurifiedbygelelectrophoresis.Invitrotranscriptionreactionwasperformedat37 °Cfor1.5 hina50 µlreactionvolumecontaining:600pmolsT7template,10 µlof5×transcriptionbuffer(200 mMTris-HCl,30 mMMgCl2,10 mMspermidine,50 mMNaCl),5 µlofrNTPsmix(20 mMeach),5 µlof100 mMDTT,0.5 µlof1%TritonX-100,80Uribonucleaseinhibitor,100UT7RNApolymerase.Then,theDNAtemplatewasdigestedbyadding4UofTURBODNase(Ambion)forthenext15 min.Thereactionwasstoppedwith2.5 µlof0.5 MEDTApH8.0andfollowingphenol/chloroformextractionandethanolprecipitation.Thequalityofspike-insRNAwasvisuallyassessedbydenaturingelectrophoresis.Spike-inswerepurifiedonRNApurificationbeads(KapaPurebeads)andsubjectedtonanoporesequencing.NorthernblotanalysisToanalyzesnR13snoRNA,5 µgofthetotalRNAor100 ngpoly(A)+RNAwasseparatedbyelectrophoresisin6%denaturingpolyacrylamide–8 Mureagelin1×TBE,followedbyRNAimmobilizationonaHybondN + nylonmembrane(Amersham,cat.no.RPN82B)bywetelectrotransferin0.5×TBE.RNAwasfixedbyultravioletcross-linking.HybridizationwasperformedinPerfectHybPlushybridizationbuffer(Sigma-Aldrich,cat.no.H7033).TheblotwashandledaccordingtostandardproceduresandprobedwithaDNAoligonucleotidethatwaslabeledatthe5′-endwithT4PNK(NewEnglandBiolabs,cat.no.M0201L)and[γ-32P]adenosinetriphosphate(HartmannAnalytic)at42 °Covernight.Afterhybridization,themembranewaswashedtwicefor30 minwith2×SSCand0.1%SDSat42 °CandthenexposedtoaPhosphorImagerscreen(FujiFilm),whichwasscannedusingaFLA7000scanner(FujiFilm).ToanalyzeTDH3andRPL18BmRNAs,20 µgofthetotalRNAor100 ngpoly(A)+RNAwaspre-annealedto100 ngDNAoligonucleotide,directedtotherespectivetranscript3′-UTRandsubsequentlydigestedwith5URNaseH(NewEnglandBiolabs,cat.no.M0297S)inabufferthatcontained20 mMTris-HCl(pH7.5),100 mMKCl,10 mMMgCl2,5%sucrose(w/v),and10 mMDTT.Followingdigestion,RNAwasextractedwithphenol:chloroformandchloroformaloneandprecipitatedwith99%ethanolinthepresenceof0.3 Msodiumacetate(pH5.2)andGlycoBlue(ThermoFisherScientific,cat.no.AM9515)coprecipitate.PelletswereresuspendedindeionizedH2O,andthefurtherstepswereanalogoustosnR13snoRNA.Betweenhybridizations,theprobeswerestrippedoffthemembranesat65 °Cbyboilingwith0.1%SDS.NorthernblotsignalswerequantifiedusingImageJfromtwobiologicalrepeats.ForbettercomparisonwithDRSdatasets,theoverallsignalwasnormalizedtotheintensityoftheregioncorrespondingtothemainbandonthenorthernblot.NanoporesequencingDirectRNAsequencingwasperformedasdescribedbyBilskaetal.58.Briefly,RNAlibrarieswerepreparedfrom500 nghuman,murine,A.thaliana,orC.eleganscap-enrichedmRNAmixedwith50–200 ngoligo-(dT)25-enrichedmRNAfromSaccharomycescerevisiaeyeastwithaDirectRNASequencingKit(catalogno.SQK-RNA002,OxfordNanoporeTechnologies)accordingtothemanufacturer’sinstructions.SequencingwasperformedusingR9.4flowcellsonaMinIONdevice(ONT).RawdatawerebasecalledusingGuppy(ONT).Rawsequencingdata(fast5files)weredepositedattheEuropeanNucleotideArchive(ENA,(foralistofaccessionnumbers,seeSupplementalTable 3).Thefulllistofsequencingruns,withappropriatesummaries,isshowninSupplementaryData 2.BioinformaticanalysisPreparationofyeastgenomeannotationCustomannotationoftheyeastgenomewasusedtomapthesequencingreads(filename:SacCer3_custom_annotation_ORFs_ncRNA_CUTs_SUTs_XUTs.bed).TheannotationwaspreparedbymergingpublisheddatasetsfromDavidetal.59;filename:Steinmetz_annotations_sacCer3_latin_sorted.bed),vanDijketal.60;filename:van_Dijk_2011_XUTs_V64.bed),andsgd_otherfromtheUniversityofCaliforniaSantaCruz(UCSC)GenomeBrowser.ORF-TfeaturesthatwerenotlistedinDavidetal.59butarespecifiedasORF-TfeaturesintheUCSCGenomeBrowser(TableBrowserandtracksgdGenes)wereaddedtothecustomannotation.Importantly,thesemissingORF-TfeatureswerenotspecifiedasnoncodingintheDavidetal.59annotation.XUTfeaturesfromvanDijketal.60andCUTandSUTinformationfromDavidetal.59weremergedintothecustomfile,andoverlappingfeatureIDsweremergedusingthebedtoolsmergefunction.Thesgd_othertrackfromtheUCSCGenomeBrowserwasusedtofindothernoncodingfeatures.Thenoncodingfeaturesthatwereshorterthan50 bpwerefilteredawaybecausethetechnologydoesnotallowthemappingofreadsofthislength.LTR,transposon,tRNAs,andreplicationoriginswerealsodiscardedfromtheannotation.Thisprocedureextracted93additionalfeatures.Theresultingcustomannotationcontained9105features,including2870noncodingonesandisavailableatMendeleydata(https://doi.org/10.17632/v5vm3dmm8y.1).PubliclyavailabledatasetsThedatasetcontainingcalculatedquantitativemeasuresoftranslationfor4621yeastgeneswasobtainedfromSupplementalTable S1fromSiwiakandZielenkiewicz,201038.Expressionvaluesoftranscriptsfrompoly(A)+ortotalRNAfraction,describedinPresnyaketal.29,weredownloadedfromGEO(GSE57385,files:GSE57385_PolyA_FPKM.txt.gzandGSE57385_RiboZero_FPKM.txt.gz).Tilingarraydata28wereobtainedfromDomenicoLibriandarepubliclyavailableat(E-MTAB-1246).ProteinabundanceestimatespercellwereobtainedfromSupplementaryInformationTable 4fromHoetal.37.Poly(A)lengthmeasurementswiththePALseqmethod30weredownloadedfromGEO(GSE52809,file:GSE52809_Cerevisiae_total.txt.gz).RATE-seqdatawereobtainedfromSupplementalTable S5fromNeyomotinetal.34,withthekindassistancefromM.Schmid.DataregardingRNAPolIIdensitywereobtainedfromSupplementalTable S1fromPelechanoetal.35(column:RNApolIIdensity(perkb)),withthekindassistancefromM.Schmid.cDTAdatafromSunetal.36wereobtainedfromArrayExpress:E-MTAB-760,withthekindassistancefromM.Schmid.Short-read(dT)-primedRNA-sequencingdatawereobtainedfromdatadepositedbySchmidetal.31andTudeketal.32withaccessionnumbersGEO:GSE108477andGEO:GSE108550.SamplesusedweretheaverageofonereplicateofMex67-AAandNab2inputRNA-rapamycinwithorwithoutinvitropolyadenylationandribodepletion.OnlysignalaroundthecleavagesitewasusedandnormalizedtoS.pombespike-inasdescribedintheoriginalpublications.Poly(A)lengthsdeterminationYeast-originatingDRSreadswereseparatedfromtheothersamplesbymappingtotherespectivereferencetranscriptomesusingMinimap22.1761andfilteringoutreadsthatwereunmappedtotheyeastreference.Theobtainedreadsweremappedtoreferencetranscriptsequences(describedabove)usingMinimap2.1761withoptions-k14-axmap-ont–secondary = noandprocessedwithsamtools1.9tofilteroutsupplementaryalignmentsandreadsmappingtoreversestrand(samtoolsview-b-F2320).Thepoly(A)taillengthsforeachreadwereestimatedusingtheNanopolish0.13.2polyafunction25.Insubsequentanalyses,onlylengthestimateswiththeQCtagthatwasreportedbyNanopolishasPASSwereconsidered.Sincethereplicatesstronglycorrelatedwithoneanotherrepeatsweresummed.Tableswiththenumberofcounts,mean,median,andgeometricmeanpoly(A)taillengthsaredepositedatMendeleydata(https://doi.org/10.17632/v5vm3dmm8y.1).Thesameprocedureformappingandpoly(A)lengthestimationwasappliedtopredefinedpoly(A)standards,butreadsweremappedtoRenillaluciferasesequenceinstead.DataanalysisandvisualizationAnalysesofchangesinmeanpoly(A)taillengthandRNAabundancewereperformedusingRandvisualizedusingtheggplot2package62,63.Boxplots(withinviolinplotsrepresentingRNApoly(A)lengthsdistributionorexpressionlevel)showmedianvalue(solidboldline)and1stand3rdquartiles,whiskersrepresent1.5 × IQR(interquartilerange).Outliers(datapointsthatdonotfallwithin1.5 × IQR)aremarkedasblackpoints.Transcriptsthatweredetectedby≤5readswereremovedfromthedatasetsbysettingtheirvalueto0.01forcountnumberandpoly(A)taillengthalike,toallowtheirinclusioninthecomparisons.Countsforeachtranscriptwerenormalizedtolibrarysizeorinthecaseofthepap1-1serieslibrarysizeandqPCR-derivedcoefficient.Forthepurposeofgraphicaldisplay,librarysizenormalizedcountsweremultipliedby1,600,000,whichistheapproximatemaximumnumberofS.cerevisiaemappedreadsthatwereobtainedfromthelargestDRSdataset.BinningmRNAsbytheexpressionlevelThebinningofmRNAbyexpressionlevelwasperformedonmRNAsdetectedineachsetofstrainsbysettingthefollowingcutoffs:(1)highlyexpressedmRNAs:log2(librarynormalizedcounts) ≥ −10,(2)moderatelytohighlyexpressedmRNAs:−12 3′digestionofthetranscript.GenesDev.8,855–866(1994).CAS  PubMed  Article  GoogleScholar  Yamashita,A.etal.Concertedactionofpoly(A)nucleasesanddecappingenzymeinmammalianmRNAturnover.Nat.Struct.Mol.Biol.12,1054–1063(2005).CAS  PubMed  Article  GoogleScholar  Workman,R.E.etal.NanoporenativeRNAsequencingofahumanpoly(A)transcriptome.Nat.Methods1–9.https://doi.org/10.1038/s41592-019-0617-2(2019).Jalkanen,A.L.,Coleman,S.J.&Wilusz,J.DeterminantsandimplicationsofmRNApoly(A)tailsize—doesthisproteinmakemytaillookbig?Semin.CellDev.Biol.34,24–32(2014).CAS  PubMed  Article  GoogleScholar  Chang,H.,Lim,J.,Ha,M.&Kim,V.N.TAIL-seq:genome-widedeterminationofpoly(A)taillengthand3′endmodifications.Mol.Cell53,1044–1052(2014).CAS  PubMed  Article  GoogleScholar  Gudipati,R.K.etal.ExtensivedegradationofRNAprecursorsbytheexosomeinwild-typecells.Mol.Cell48,409–421(2012).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Presnyak,V.etal.CodonoptimalityisamajordeterminantofmRNAstability.Cell160,1111–1124(2015).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Subtelny,A.O.,Eichhorn,S.W.,Chen,G.R.,Sive,H.&Bartel,D.P.Poly(A)-tailprofilingrevealsanembryonicswitchintranslationalcontrol.Nature508,66–71(2014).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  Schmid,M.,Tudek,A.&Jensen,T.H.Simultaneousmeasurementoftranscriptionalandpost-transcriptionalparametersby3′endRNA-Seq.CellRep.24,2468–2478.e4(2018).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Tudek,A.etal.AnuclearexportblocktriggersthedecayofnewlysynthesizedpolyadenylatedRNA.CellRep.24,2457–2467.e7(2018).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Lima,S.A.etal.Shortpoly(A)tailsareaconservedfeatureofhighlyexpressedgenes.Nat.Struct.Mol.Biol.24,1057(2017).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Neymotin,B.,Athanasiadou,R.&Gresham,D.DeterminationofinvivoRNAkineticsusingRATE-seq.RNA20,1645–1652(2014).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Pelechano,V.,Chávez,S.&Pérez-Ortín,J.E.Acompletesetofnascenttranscriptionratesforyeastgenes.PLOSONE5,e15442(2010).ADS  PubMed  PubMedCentral  Article  CAS  GoogleScholar  Sun,M.etal.Comparativedynamictranscriptomeanalysis(cDTA)revealsmutualfeedbackbetweenmRNAsynthesisanddegradation.GenomeRes.22,1350–1359(2012).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Ho,B.,Baryshnikova,A.&Brown,G.W.UnificationofproteinabundancedatasetsyieldsaquantitativeSaccharomycescerevisiaeproteome.CellSyst.6,192–205.e3(2018).CAS  PubMed  Article  GoogleScholar  Siwiak,M.&Zielenkiewicz,P.Acomprehensive,quantitative,andgenome-widemodeloftranslation.PLOSComput.Biol.6,e1000865(2010).ADS  PubMed  PubMedCentral  Article  CAS  GoogleScholar  Haruki,H.,Nishikawa,J.&Laemmli,U.K.Theanchor-awaytechnique:rapid,conditionalestablishmentofyeastmutantphenotypes.Mol.Cell31,925–932(2008).CAS  PubMed  Article  GoogleScholar  Rougemaille,M.etal.DissectingmechanismsofnuclearmRNAsurveillanceinTHO/sub2complexmutants.EMBOJ.26,2317–2326(2007).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Burkard,K.T.D.&Butler,J.S.Anuclear3′-5′exonucleaseinvolvedinmRNAdegradationinteractswithPoly(A)polymeraseandthehnRNAproteinNpl3p.Mol.Cell.Biol.20,604–616(2000).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Carneiro,M.O.etal.Pacificbiosciencessequencingtechnologyforgenotypingandvariationdiscoveryinhumandata.BMCGenomics13,375(2012).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Kuai,L.,Fang,F.,Butler,J.S.&Sherman,F.PolyadenylationofrRNAinSaccharomycescerevisiae.Proc.NatlAcad.Sci.USA101,8581–8586(2004).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  Grzechnik,P.&Kufel,J.PolyadenylationlinkedtotranscriptionterminationdirectstheprocessingofsnoRNAprecursorsinyeast.Mol.Cell32,247–258(2008).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Traven,A.etal.TheCcr4-Pop2-NOTmRNAdeadenylasecontributestoseptinorganizationinSaccharomycescerevisiae.Genetics182,955–966(2009).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Balagopal,V.,Bolisetty,M.,Husaini,N.A.,Coller,J.&Graveley,B.R.Ccr4andPop2controlpoly(A)taillengthinSaccharomycescerevisiae.PreprintatbioRxivhttps://doi.org/10.1101/140202(2017).Castells-Roca,L.etal.HeatshockresponseinyeastinvolveschangesinbothtranscriptionratesandmRNAstabilities.PLoSONE6,e17272(2011).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  Fasken,M.B.,Laribee,R.N.&Corbett,A.H.Nab3facilitatesthefunctionoftheTRAMPcomplexinRNAprocessingviarecruitmentofRrp6independentofNrd1.PLoSGenet.11,e1005044(2015).PubMed  PubMedCentral  Article  CAS  GoogleScholar  Makino,D.L.,Baumgärtner,M.&Conti,E.CrystalstructureofanRNA-bound11-subuniteukaryoticexosomecomplex.Nature495,70–75(2013).ADS  CAS  PubMed  Article  GoogleScholar  Blobel,G.Genegating:ahypothesis.Proc.NatlAcad.Sci.USA82,8527–8529(1985).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  Burns,L.T.&Wente,S.R.Fromhypothesistomechanism:uncoveringnuclearporecomplexlinkstogeneexpression.Mol.Cell.Biol.34,2114–2120(2014).PubMed  PubMedCentral  Article  CAS  GoogleScholar  Vargas,D.Y.,Raj,A.,Marras,S.A.E.,Kramer,F.R.&Tyagi,S.MechanismofmRNAtransportinthenucleus.Proc.NatlAcad.Sci.USA102,17008–17013(2005).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  Wolf,J.&Passmore,L.A.mRNAdeadenylationbyPan2–Pan3.Biochem.Soc.Trans.42,184–187(2014).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Miller,J.E.&Reese,J.C.Ccr4-Notcomplex:thecontrolfreakofeukaryoticcells.Crit.Rev.Biochem.Mol.Biol.47,315–333(2012).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Stowell,J.A.W.etal.ReconstitutionoftargeteddeadenylationbytheCcr4-NotcomplexandtheYTHdomainproteinMmi1.CellRep.17,1978–1989(2016).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Webster,M.W.,Stowell,J.A.&Passmore,L.A.RNA-bindingproteinsdistinguishbetweensimilarsequencemotifstopromotetargeteddeadenylationbyCcr4-Not.eLife8,e40670(2019).PubMed  PubMedCentral  Article  GoogleScholar  Alhusaini,N.&Coller,J.ThedeadenylasecomponentsNot2p,Not3p,andNot5ppromotemRNAdecapping.RNA22,709–721(2016).CAS  PubMed  PubMedCentral  Article  GoogleScholar  Bilska,A.etal.Immunoglobulinexpressionandthehumoralimmuneresponseisregulatedbythenon-canonicalpoly(A)polymeraseTENT5C.Nat.Commun.11,2032(2020).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  David,L.etal.Ahigh-resolutionmapoftranscriptionintheyeastgenome.Proc.NatlAcad.Sci.USA103,5320–5325(2006).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  vanDijk,E.L.etal.XUTsareaclassofXrn1-sensitiveantisenseregulatorynon-codingRNAinyeast.Nature475,114–117(2011).PubMed  Article  CAS  GoogleScholar  Li,H.Minimap2:pairwisealignmentfornucleotidesequences.Bioinforma.Oxf.Engl.34,3094–3100(2018).CAS  Article  GoogleScholar  RDevelopmentCoreTeam.R:ALanguageandEnvironmentforStatisticalComputing.http://www.R-project.org(RFoundationforStatisticalComputing,2016).Wickham,H.ggplot2:ElegantGraphicsforDataAnalysis(Springer-VerlagNewYork,2016).DownloadreferencesAcknowledgementsWethankJoannaKufelforprovidingsomeoftheyeaststrainsandcommentsonthemanuscript,DominicoLibriforadviceconcerningGudipatietal.28datasets,PawełGrzechnikforcriticalreadingandManfredSchmidfortechnicaladvice.ThisworkwasmainlysupportedbytheTEAM/2016-1/3FoundationforPolishSciencegrant(toA.D.).A.D.isalsoarecipientoftheERAChair’sposition,fundedbytheEU(agreementno.810425).S.M.’sworkwassupportedbyaPolishNationalScienceCentregrant(no.2020/38/E/NZ2/00372toS.M.).WorkintheR.TomeckigroupwassupportedbyaPolishNationalScienceCentregrant(no.2017/26/E/NZ1/00724toR.T.).FundingforthisworkintheT.H.J.laboratorywasprovidedbytheIndependentResearchFundDenmarkandaFEBSLongTermFellowshiptogetherwithanEMBOLong-TermFellowship(ALTF328-2019)toM.T.AuthorinformationAuthornotesTheseauthorscontributedequally:AgnieszkaTudek,PawełS.Krawczyk.AuthorsandAffiliationsInstituteofBiochemistryandBiophysics,Warsaw,PolandAgnieszkaTudek, PawełS.Krawczyk, RafałTomecki & AndrzejDziembowskiLaboratoryofRNABiology,InternationalInstituteofMolecularandCellBiology,Warsaw,PolandPawełS.Krawczyk, SewerynMroczek, KatarzynaMatylla-Kulińska & AndrzejDziembowskiInstituteofGeneticsandBiotechnology,FacultyofBiology,UniversityofWarsaw,Warsaw,PolandSewerynMroczek, RafałTomecki, KatarzynaMatylla-Kulińska & AndrzejDziembowskiDepartmentofMolecularBiologyandGenetics,AarhusUniversity,AarhusC,DenmarkMattiTurtola & TorbenHeickJensenAuthorsAgnieszkaTudekViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarPawełS.KrawczykViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarSewerynMroczekViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarRafałTomeckiViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarMattiTurtolaViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarKatarzynaMatylla-KulińskaViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarTorbenHeickJensenViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarAndrzejDziembowskiViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarContributionsA.T.designedandperformedtheexperiments,analyzedthedata,andwrotethemanuscript.P.K.processedtheDRSdata,performedessentialbioinformaticandstatisticalanalyses,andeditedthemanuscript.S.M.implementedDRSandpreparedallsequencinglibraries.R.T.performedtheNorthernblotanalyses,editedthemanuscript,andprovidedfunding.M.T.preparedMEX67-AAsamplesandeditedthemanuscript.KMKpreparedDRSspike-ins.T.H.J.editedthemanuscriptandprovidedfunding.A.D.conceptualizedandcoordinatedtheproject,providedfunding,andeditedthemanuscript.CorrespondingauthorsCorrespondenceto AgnieszkaTudekorAndrzejDziembowski.Ethicsdeclarations Competinginterests Theauthorsdeclarenocompetinginterests. AdditionalinformationPeerreviewinformationNatureCommunicationsthanksAlainJacquierandother,anonymous,reviewersfortheircontributionstothepeerreviewofthiswork.Peerreviewreportsareavailable.Publisher’snoteSpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpublishedmapsandinstitutionalaffiliations.Supplementaryinformation ManuscriptsupplementaryinformationPeerReviewFileSupplementaryData1SupplementaryData2DescriptionofadditionalsupplementaryfilesReportingSummaryRightsandpermissions 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/. ReprintsandPermissionsAboutthisarticleCitethisarticleTudek,A.,Krawczyk,P.S.,Mroczek,S.etal.GlobalviewonthemetabolismofRNApoly(A)tailsinyeastSaccharomycescerevisiae. NatCommun12,4951(2021).https://doi.org/10.1038/s41467-021-25251-wDownloadcitationReceived:18January2021Accepted:29July2021Published:16August2021DOI:https://doi.org/10.1038/s41467-021-25251-wSharethisarticleAnyoneyousharethefollowinglinkwithwillbeabletoreadthiscontent:GetshareablelinkSorry,ashareablelinkisnotcurrentlyavailableforthisarticle.Copytoclipboard ProvidedbytheSpringerNatureSharedItcontent-sharinginitiative Furtherreading Poly(a)selectionintroducesbiasandunduenoiseindirectRNA-sequencing MarcusJ.Viscardi JoshuaA.Arribere BMCGenomics(2022) CommentsBysubmittingacommentyouagreetoabidebyourTermsandCommunityGuidelines.Ifyoufindsomethingabusiveorthatdoesnotcomplywithourtermsorguidelinespleaseflagitasinappropriate. DownloadPDF AssociatedContent Focus Genetics,genomicsandepigenetics Advertisement Explorecontent Researcharticles Reviews&Analysis News&Comment Videos Collections Subjects FollowusonFacebook FollowusonTwitter Signupforalerts RSSfeed Aboutthejournal Aims&Scope Editors JournalInformation Openaccess EditorialValuesStatement JournalMetrics Editors'Highlights Contact Editorialpolicies TopArticles Publishwithus Forauthors ForReviewers Submitmanuscript Search Searcharticlesbysubject,keywordorauthor Showresultsfrom Alljournals Thisjournal Search Advancedsearch Quicklinks Explorearticlesbysubject Findajob Guidetoauthors Editorialpolicies Closebanner Close SignupfortheNatureBriefingnewsletter—whatmattersinscience,freetoyourinboxdaily. Emailaddress Signup IagreemyinformationwillbeprocessedinaccordancewiththeNatureandSpringerNatureLimitedPrivacyPolicy. Closebanner Close Getthemostimportantsciencestoriesoftheday,freeinyourinbox. SignupforNatureBriefing