negativemeizhai

ARTICLELncRNACAIFinhibitsautophagyandattenuatesmyocardialinfarctionbyblockingp53-mediatedmyocardintranscriptionCui-YunLiu1,Yu-HuiZhang2,Rui-BeiLi3,Lu-YuZhou1,TaoAn2,Rong-ChengZhang2,MeiZhai2,YanHuang2,Kao-WenYan1,Yan-HanDong1,MurugavelPonnusamy1,ChanShan1,ShengXu1,QiWang1,Yan-HuiZhang1,JianZhang2&KunWang1IncreasingevidencesuggeststhatlongnoncodingRNAs(lncRNAs)playcrucialrolesinvariousbiologicalprocesses.
However,littleisknownabouttheeffectsoflncRNAsonautophagy.
HerewereportthatalncRNA,termedcardiacautophagyinhibitoryfactor(CAIF),suppressescardiacautophagyandattenuatesmyocardialinfarctionbytargetingp53-mediatedmyocardintranscription.
MyocardinexpressionisupregulateduponH2O2andischemia/reperfusion,andknockdownofmyocardininhibitsautophagyandattenuatesmyocardialinfarction.
p53regulatescardiomyocytesautophagyandmyocardialischemia/reperfusioninjurybyregulatingmyocardinexpression.
CAIFdirectlybindstop53proteinandblocksp53-mediatedmyocardintranscription,whichresultsinthedecreaseofmyocardinexpression.
Collectively,ourdatarevealanovelCAIF-p53-myocardinaxisasacriticalreg-ulatorincardiomyocyteautophagy,whichwillbepotentialtherapeutictargetsintreatmentofdefectiveautophagy-associatedcardiovasculardiseases.
DOI:10.
1038/s41467-017-02280-yOPEN1CenterforDevelopmentalCardiology,InstituteforTranslationalMedicine,CollegeofMedicine,QingdaoUniversity,Qingdao,266021,China.
2StateKeyLaboratoryofCardiovascularDisease,HeartFailurecenter,FuwaiHospital,NationalCenterforCardiovascularDiseases,ChineseAcademyofMedicalSciences,PekingUnionMedicalCollege,Beijing,100037,China.
3SchoolofProfessionalStudies,NorthwesternUniversity,Chicago,IL60611,USA.
Cui-YunLiu,Yu-HuiZhang,Rui-BeiLiandLu-YuZhoucontributedequallytothiswork.
CorrespondenceandrequestsformaterialsshouldbeaddressedtoK.
W.
(email:wangk20150812@sina.
com)NATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunications11234567890Autophagyisanevolutionarilyconservedintracellulardegradationprocess,whichmaintainscellularhome-ostasisbyremovingdamagedproteinsandorganelleturnover.
Autophagyhasbeendemonstratedtoplayawidevarietyofphysiologicalandpathophysiologicalrolesinmanykindsofcellsandtissues.
Autophagydysregulationisassociatedwithanumberofcardiacdiseases,includingdilatedcardiomyo-pathy,ischemicheartdisease,andheartfailure1–3.
Althoughsomestudiessuggestthatautophagiccelldeathplaysapivotalroleincardiacdisease,hitherto,thereisnoeffectivetreatmentfortheautophagy-relatedheartdiseaseandheartfailure.
Thus,itisofgreatimportancetoexploreandrevealthemolecularmechanismunderlyingtheregulationofautophagy.
Under-standingthemechanismofautophagywillprovideanovelinterventionalstrategyfortreatingcardiovasculardiseasesandheartfailure.
LncRNAsareimportantclassofnoncodingRNAthatarecharacterizedbytheirlengthlongerthan200nt.
LncRNAshavebeenidentiedinmanytypesofcellsandtissues.
LncRNAsplayvitalroleinvariousnormalphysiologicalandpathologicalcon-ditionsincludingcelldifferentiation,metabolismandcancerprogression4–6.
LncRNAsregulatemoleculesbymultiplemechanisms,includingepigeneticregulation,genomicimprint-ing,RNAstability,RNAalternativespliceandmicroRNAreg-ulation7,8.
EmergingevidencessuggestthatlncRNAsareinvolvedintheregulationofcardiacdiseases9,10.
However,theinuenceoflncRNAsintheregulationofautophagyintheheartremainslargelyunknown.
P53isatumorsuppressorproteinandiswidelyknownforitsroleasatranscriptionfactor.
Itiswellknownthatp53activatesgenesthatregulatecellcyclecheckpoints,DNAdamageandrepair,andapoptosis11.
Studiesshowthatthereisanimportantrelationshipbetweenp53andautophagy.
Dele-tionorinhibitionofp53affectsautophagyinseveralcelllines,includinghuman,mouseandnematodecells12.
However,theautophagyregulatoryfunctionofp53incardiomyocytesremainslargelyunknown.
Myocardinisanuclearproteinandatranscriptionalcoactivatorofserumresponsefactorthatisspecicallyexpressedinthesmoothmuscleandcardiacmuscle13.
Myocardindysfunctionincardiomyocytestriggersapoptosis14.
Myocardinisimportantforthemaintenanceofheartfunction.
MyocardintransactivatestheBmp10andregulatescardiomyocyteproliferationandapoptosisintheembryonicheart15.
Althoughmyocardinisofgreatimportancetosomepathologyandphysiologyprocessintheheart,itisnotyetclearwhethermyocardincanregulateautophagicprogramincardiomyocyte.
Ourpresentstudyrevealsthatmyocardinpositivelymodulatestheautophagicprocessincardiomyocytesanditpromotesmyocardialinfarction.
Theknockdownofmyocardininhibitsautophagiccelldeathandattenuatesischemicinjuryinducedincreaseofmyocardialinfarctionsize.
Wedemonstratethatp53activatesmyocardintranscrip-tion.
Itupregulatesautophagyandmyocardialinfarctionbyincreasingmyocardinexpression.
Further,weobservedthatalncRNA,termedcardiacautophagyinhibitoryfactor(CAIF),directlybindstop53proteinandblocksitsbindingtothepromoterregionofmyocardin.
OurstudyindicatesthatCAIFisabletoinhibitcelldeathandattenuatemyocardialinfarctionintheheartthroughtargetingthep53/myocardindependentautophagypathway.
OurndingsofferabetterunderstandingabouttheinteractionbetweenlncRNAandproteinsinvolvedintheregulationofautophagy.
Collectively,thisstudyprovidesanovelevidencethatlncRNAmodulatescardiomyocyteautophagyandprotectsthecardiactissuefrommyocardialinfarction.
ResultsMyocardinmediatesautophagyandcelldeathincardiomyo-cyte.
Oxidativestressplaycriticalroleinthedevelopmentofcardiovascularproblemsandcardiomyocytesaremorepronetooxidantinjury.
Itiswellknownthatoxidantssuchashydrogenperoxide(H2O2)caninducesautophagyprocessincardiomyo-cytesandthedysregulationofautophagycausescardiomyocytecelldeath.
Totestwhethermyocardinparticipatesinthereg-ulationofautophagyincardiomyocytes,wetreatedcardiomyo-cyteswithH2O2toinduceautophagyandexaminedtheexpressionpatternofmyocardin.
Incardiomyocytes,H2O2exposureincreasedtheautophagyasindicatedbyaccumulationofGFP-LC3punctuatestructures(Fig.
1a),whichwasaccom-paniedbyatimedependentincreaseofmyocardinexpression(Fig.
1b,c).
TheseresultsindicatethatthelevelofmyocardinisincreasedduringH2O2-inducedautophagyincardiomyocytes.
Next,wesilencedtheexpressionofmyocardinandinvestigateditsimpactonH2O2inducedautophagy.
Theknockdownofmyocardin(SupplementaryFig.
1a)remarkablyreducedH2O2-inducedautophagicprocessasshownbyasignicantdecreaseinpunctuateaccumulationsofGFP-LC3(Fig.
1dandSupplemen-taryFig.
1b).
Inaddition,H2O2inducedincreasedofLC3-IIexpressionwassignicantlydecreasedbysilencingofmyocardinwithsiRNA(Fig.
1e).
Autophagicuxdenotesthedynamicprocessofautophagy,whichisareliableindicatorofautophagicactivity.
WethereforemeasuredtheautophagicuxusingtandemmRFP-GFP-LC3uorescenceanalysis16–18incardiomyocytes.
Theyellowpuncta,whicharethecombinationofGFPandRFPuorescence,indicateautophagosomes.
Theredpunctaindicateautolysosomes,wheregreenuorescencesofGFP-LC3punctaarequenchedduetotheacidicenvironment.
H2O2inducedautop-hagicuxincardiomyocytes,asevidencedbytheincreasednumberofRFP-LC3autolysosomes,whereastheknockdownofmyocardinefcientlyattenuatedtheH2O2-inducedincreaseintheautophagicux(SupplementaryFig.
1c).
Theautophagiccelldeathisrecognizedasalegitimatealternativeformofpro-grammedcelldeaththatoccursviathedefectiveofautophagy.
ItiswellknownthatBeclin1isakeyinitiatorofautophagicprocessthatmediatestheautophagiccelldeathincardiomyocytes19.
Inourstudy,theknockdownofBeclin1attenuatedH2O2-inducedcelldeathincardiomyocytes(SupplementaryFig.
2a).
Inaddition,autophagyinhibitor3-methyladenine(3-MA)treatmentatte-nuatedH2O2-inducedcelldeath(supplementaryFig.
2b).
TheseresultsconrmthatH2O2inducesautophagiccelldeathincar-diomyocytes.
TheknockdownofmyocardinremarkablyreducedcelldeathinducedbyH2O2(Fig.
1f).
Takentogether,theseresultssuggestthatmyocardinmediatesoxidativestressinducedacti-vationofautophagiccelldeathincardiomyocytes.
Next,weexaminedtheroleofmyocardininthepathogenesisofmyocardialinfarctionusingexperimentalanimalmodel.
InmicewithmyocardialI/Rinjury,theexpressionlevelofmyocardinwassignicantlyincreasedcomparedtocontrolmice(SupplementaryFig.
2c).
ThesilencingofexpressionofmyocardinbyadenoviraldependentdeliveryofsiRNAinvivo(SupplementaryFig.
2d)attenuatedautophagy(SupplementaryFig.
3a–candFig.
1g).
Inaddition,invivoadministrationofmyocardinsiRNAsignicantlyreducedmyocardialinfarctionsize(Fig.
1h)intheheartwithI/Rinjury.
Toconrmthatmyocardinmediatesautophagiccelldeath,weadministeredBeclin1-siRNAalongwithadenovirusexpressingmyocardinandinducedmyocardialI/Rinmice.
TheoverexpressionofmyocardinaugmentedI/R-inducedmyocardialinfarctionsizes,whichwasinhibitedbyknockdownofBeclin1(supplementaryFig.
3d),whichindicatesthatmyocardinactivatesautophagy-dependentcelldeathincardiomyocytesduringI/Rinjury.
ThesilencingofmyocardinsignicantlyprotectedthemyocardialfunctionfromARTICLENATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-y2NATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunicationsI/Rinjury(SupplementaryFig.
3e,f).
Collectively,thesedatasuggestthatmyocardincontributestotheinductionofautophagicprocessandcardiomyocytecelldeathintheheartduringI/Rinjury.
Myocardinactsasasmoothmuscleandcardiacmuscle-specictranscriptionalactivatorthattransactivatestheexpressionofmanygenes13.
Tounderstandthemechanismsofautophagyregulationbymyocardinintheheart,weexaminedwhethermyocardincontrolstheexpressionofkeymolecules(Atg5,Atg7,Atg10,Beclin1,etc)involvedintheexecutionofautophagicprocess.
Theoverexpressionofmyocardininculturedcardio-myocytesremarkablyincreasedthelevelofBeclin1mRNA,butitdidnotaffecttheexpressionofotherautophagy-relatedgenessuchasAtg5,Atg7,Atg10andAtg12(SupplementaryFig.
4a–e).
TheseresultsindicatethatBeclin1mightbeapotentialtargetofmyocardininthecascadeofautophagy.
abcdfeghControl0624Time(h)%CellwithGFP-LC3vacuolesH2O2H2O2H2O2H2O2H2O2+Mycd-scH2O2+Mycd-siRNAH2O2H2O201020304050P<0.
001P=0.
002P<0.
001%CellwithGFP-LC3vacuolesMycd-siRNAMycd-sc++––––––––––––––+++01020304050P<0.
001P<0.
001P<0.
001Celldeath(%)Mycd-siRNAMycd-sc+++++01020304050P<0.
001P<0.
001P<0.
001+++++I/RTotalareaoccupiedbyAV(%)+ShamMycd-siRNAMycd-sc0510152025P<0.
001P<0.
001P<0.
001+++++I/R+ShamMycd-siRNAMycd-scINF/LV(%)010203040P<0.
001P<0.
001P<0.
001Time(h)MyocardinlevelsP<0.
001P=0.
001P<0.
001012345Time(h)MyocardinActin36kDa95ActinLC3-ILC3-II+++++Mycd-siRNAMycd-sckDa173612062412062412H2O2–––––––––––––––––––––––––––Fig.
1Myocardinparticipatesintheregulationofautophagy.
aH2O2inducesautophagosomeformation.
CardiomyocyteswereinfectedwithGFP-LC3andthentreatedwithH2O2attheindicatedtime.
TheextentofautophagywasassessedbyanalyzingstainingpatternsofGFP-LC3,andquantitationofautophagywasshown.
n=3.
b,cH2O2inducesanincreaseinmyocardinlevels.
CardiomyocytesweretreatedwithH2O2atindicatedtime.
MyocardinmRNAlevels(b)andproteinlevels(c)wereanalyzed.
n=4.
dKnockdownofmyocardininhibitspunctateaccumulationsofGFP-LC3inducedbyH2O2.
CardiomyocyteswereinfectedwithadenoviralmyocardinsiRNA(Mycd-siRNA)oritsscrambleform(Mycd-sc),theninfectedwithGFP-LC3.
Twenty-fourhoursafterinfectioncellsweretreatedwithH2O2.
RepresentativephotosofGFP-LC3cellswereshownintheleftpanel(Bar=20μm)andthepercentageofcellswithGFP-LC3punctawasquantiedintherightpanel.
n=4.
eRepresentativeimmunoblotforconversionofLC3-ItoLC3-II.
CardiomyocyteswereinfectedwithadenoviralMycd-siRNAorMycd-sc.
Twenty-fourhoursafterinfectioncellsweretreatedwithH2O2.
ThepositionsofLC3-IandLC3-IIareindicated.
n=3.
fKnockdownofmyocardinreducescelldeath.
Cellsweretreatedasdescribedin(e).
Quantitationofcelldeathwasshown.
n=4.
gMyocardinknockdownsuppressesautophagyinvivo.
MicewereinjectedwithMycd-siRNAorMycd-scasdescribedinmethodssection,andthenweresubjectedto45minischemiaand3hreperfusion(I/R).
Quanticationofautophagicvacuolesinthearea-atriskwasshown.
n=6.
hKnockdownofmyocardinreducesmyocardialinfarctionuponI/Rinvivo.
Miceweretreatedasdescribedin(g).
Theupperpanelsarerepresentativephotosofmidventricularmyocardialslices(Bar=2mm).
Thelowpanelshowsinfarctsizes.
LVleftventricle,INFinfarctarea.
n=6miceNATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-yARTICLENATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunications3p53activatesmyocardintranscriptionandexpression.
Todeterminetheupstreamactivatorofmyocardinexpressionduringautophagy,weanalyzedthepromoterregionofmousemyocardinandfoundthatmyocardinpossessespotentialbindingsiteofp53(Fig.
2a),whichraisesthepossibilitythatmyocardincouldbeadirecttargetofp53.
Toexaminethis,weconstructedluciferasevectorsconsistingofwild-typeormutantmyocardinandtrans-fectedcells.
Theluciferaseassayshowedthatp53stimulatedthewild-typemyocardinpromoteractivityasindicatedbyincreasedluciferaseactivity,whilemutationsinthep53-bindingsiteinhibitedtheinteractionofp53withmyocardinpromoterasindicatedbysignicantlyreducedlevelofluciferaseactivity(Fig.
2b).
Toconrmthisresult,weenhancedorsilencedtheexpressionofp53usingadenoviralvectororp53specicsiRNArespectivelyandassessedtheexpressionofmyocardininculturedcardiomyocytes.
Theenforcedexpressionofp53(SupplementaryFig.
5a)signicantlyincreasedtheexpressionofmyocardin(Fig.
2c),whereasp53knockdown(SupplementaryFig.
5b)exhibitedasignicantdecreaseintheexpressionlevelofmyocardin(Fig.
2d).
TheChIPassayrevealedthatp53isboundtothemyocardinpromoterunderthenormalphysiologicalconditionincardiomyocytes,andH2O2treatmentledtoanincreaseintheassociationofp53withmyocardinpromoter(Fig.
2e).
TheCHIPassayintheheartwithmyocardialI/Rinjuryshowedthatthebindingofp53tothemyocardinpromoterinthearea-at-riskwashighlyincreasedcomparedtothatintheremoteregionsofmyocardiumwithIRinjury(SupplementaryFig.
5c,d).
Theseresultsconrmthatmyocardinisapotentialtranscrip-tionaltargetofp53.
Byusingluciferasereportersystem,weobservedthatH2O2inducedtheelevationofmyocardinpromoteractivityincardiomyocytesandknockdownofp53attenuatedtheincreaseofmyocardinpromoteractivity(Fig.
2f).
Thesedataindicatethatp53cantranscriptionallyactivatemyocardinexpression.
Knockdownofp53inhibitsautophagyinvitroandinvivo.
Next,wetestedthefunctionalroleofp53inautophagy.
Incul-turedcardiomyocytes,knockdownofp53effectivelyreducedabcdWildtype(wt)p53BSGGGCCTGACCMyocardinMyocardinMutant(–1801~–1792)Luciferaseactivity(foldinduction)++++β-galp53pGL4.
17Mycd-wtMycd-mut051015P=0.
002P<0.
001++++++–––––––––––––––P<0.
001p53-scp53-siRNA++Myocardinlevels0.
00.
51.
01.
5P=0.
007ActinMyocardinp53-scp53-siRNA++36kDa95β-galMyocardinlevelsp53++012345P<0.
001p53Myocardinβ-galActin++36kDa95–––––fH2O2Luciferaseactivity(foldinduction)p53-sc++pGL4.
17Mycd-wtp53-siRNA++++++++P=0.
001P<0.
001P<0.
001051015–––––––––––––––eCHIP:p53MNegativeInput6012+++++p53H2O2(h)250bp100bp12––––––––––––––––––––––Fig.
2Myocardinisatranscriptionaltargetofp53.
aMousemyocardinpromoterregioncontainsapotentialp53-bindingsite.
bp53promotesmyocardinpromoteractivity.
Cardiomyocytesweretreatedwiththeadenoviralβ-galorp53,theconstructsoftheemptyvector(pGL-4.
17),thewild-typepromoter(Mycd-wt)orthepromoterwithmutationsinthebindingsite(Mycd-mut)respectively.
Luciferaseactivitywasassayed.
n=4.
cp53inducestheincreaseofmyocardinexpressionlevels.
Cardiomyocyteswereinfectedwithadenoviralβ-galorp53.
MyocardinexpressionwasanalyzedbyqRT-PCR(leftpanel)andimmunoblot(rightpanel).
n=4.
dKnockdownofp53reducesthemyocardinexpression.
Cardiomyocyteswereinfectedwithadenoviralp53-siRNAorp53-sc.
MyocardinlevelswereanalyzedbyqRT-PCR(leftpanel)andimmunoblot(rightpanel).
n=4.
eChIPanalysisofp53bindingtothepromoterofmyocardin.
CardiomyocytesweretreatedwithH2O2atindicatedtime.
CHIPwasperformedwithp53orβ-actin(negative)antibody.
fKnockdownofp53inhibitstheincreaseofmyocardinpromoteractivityinducedbyH2O2.
Cardiomyocytesweretreatedwiththeadenoviralp53-siRNAorp53-sc,theconstructsoftheemptyvector(pGL-4.
17),orthewild-typepromoter(wt),andthenweretreatedwithH2O2.
Luciferaseactivitywasassayed.
n=4ARTICLENATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-y4NATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunicationsH2O2inducedincreaseofmyocardinexpression(Fig.
3a)andaccumulationofGFP-LC3-IIpunctate(Fig.
3b).
Theresultsofelectronmicroscopicanalysisshowedthatp53knockdownsig-nicantlyreducedH2O2-inducedaccumulationofautophagicvesiclesinculturedcardiomyocytes(Fig.
3c,d).
Inaddition,theknockdownofp53attenuatedH2O2inducedupregulationofLC3-IIexpression(Fig.
3e)andcelldeath(Fig.
3f).
Ontheotherhand,overexpressionofp53augmentedH2O2-inducedcelldeathincardiomyocytes,whereas3-MAtreatmentremarkablyinhib-itedtheeffectofp53onH2O2-inducedcelldeath(SupplementaryFig.
5e),whichindicatesthatp53inducesautophagiccelldeath.
InanimalmodelofI/Rinjury,thedeliveryofp53-siRNA(Sup-plementaryFig.
5f)attenuatedtheaccumulationofautophago-somes(Fig.
3g)inresponsetoI/Rinjury.
Theaccumulationofautophagosomesismainlyduetoeithertheoveractivationofautophagicuxortheblockageofdownstreamofautophagicvacuoleprocessing.
WethusmeasuredtheautophagicuxinthepresenceoflysosomalinhibitorbalomycinA1(BafA1),whichinhibitsfusionofautophagosomeandlysosome.
TreatmentofI/R-injuredmicewithBafA1resultedinasignicantincreaseinLC3-IIlevelscomparedtountreatedcontrols,whichindicatesthatI/Renhancedcardiacautophagicux.
Theadministrationofp53siRNAsignicantlydecreasedtheconversionofLC3-ItoLC3-IIinI/RinjuryeveninthepresenceofBafA1(Supple-mentaryFig.
5g).
Theseresultsdemonstratethatp53isinvolvedinactivationofcardiacautophagicuxintheheartduringischemicinjuryandknockdownofp53canattenuatethisprocess.
Inaddition,p53-siRNAsignicantblockedtheI/Rinjuryinducedincreaseinthesizeofmyocardialinfarction(Fig.
3h).
Takentogether,thesedatasuggestthatp53contributestotheactivationofautophagysignalandcardiomyocytecelldeathintheheart.
aceControlNNNNb%CellwithGFP-LC3vacuolesp53-siRNAp53-sc+++++P<0.
001P=0.
003P=0.
00601020304050dAutophagicvacuoles/cellp53-siRNAp53-sc+++++05101520P<0.
001P<0.
001P<0.
001Celldeath(%)p53-siRNAp53-sc+++++f01020304050P<0.
001P<0.
001P<0.
001g+++++I/R+Shamp53-siRNAp53-scTotalareaoccupiedbyAV(%)P<0.
001P<0.
001P<0.
0010510152025hINF/LV(%)010203040P<0.
001P<0.
001P<0.
001MyocardinActinp53-siRNAp53-sc––––––––––––––+++++H2O2H2O2H2O2H2O2H2O2H2O2H2O2+p53-scH2O2+p53-siRNA36kDa95ActinLC3-ILC3-IIp53-siRNAp53-sc+++++kDa1736–––––––––––––––––––––––––––––––+++++I/R+Shamp53-siRNAp53-sc––––––––––Fig.
3Knockdownofp53inhibitsautophagyinvitroandinvivo.
aKnockdownofp53suppressesH2O2-inducedmyocardinexpression.
Cardiomyocyteswereinfectedwithadenoviralp53-siRNAorp53-sc.
Twenty-fourhoursafterinfection,cellsweretreatedwithH2O2.
Myocardinlevelswereanalyzedbyimmunoblot.
bKnockdownofp53inhibitspunctateaccumulationsofGFP-LC3.
CardiomyocyteswereinfectedwithadenoviralGFP-LC3,p53-siRNAandorp53-sc.
Twenty-fourhoursafterinfectioncellsweretreatedwithH2O2.
ThepercentageofcellswithGFP-LC3punctawasquantied.
n=3.
c,dp53knockdownreducesautophagicvacuoles.
Cardiomyocytesweretreatedasdescribedin(a).
RepresentativeEMimageswereshown(c).
Bar=2M.
Thearrowsdepictautophagosomes,andthenucleusisdenotedbyN.
Quanticationofautophagicvacuoleswasshownin(d).
n=5.
eRepresentativeimmunoblotforconversionofLC3-ItoLC3-II.
Cardiomyocytesweretreatedasdescribedin(a).
ThepositionsofLC3-IandLC3-IIareindicated.
fp53knockdownreducesH2O2-inducedcelldeath.
Cardiomyocytesweretreatedasdescribedin(a).
Quantitationofcelldeathisshown.
n=4.
gKnockdownofp53reducesautophagyinvivo.
Micewereinjectedwithadenoviralp53-siRNAorp53-scasdescribedinmethods,andthenweresubjectedto45minischemiaand3hreperfusion(I/R).
Quanticationofautophagicvacuolesinthearea-atriskwasshown.
n=7.
hp53knockdownreducesI/R-inducedmyocardialinfarction.
Miceweretreatedasdescribedin(g).
Theinfarctsizeswerequantied.
LVleftventricle,INFinfarctarea.
n=8NATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-yARTICLENATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunications5CAIFblocksp53-mediatedmyocardintranscription.
Toinvestigatetheunderlyingmechanismofp53-mediatedregulationofmyocardinandautophagyincardiomyocytes,wemadeattempttoidentifyp53-interactinglncRNA,whichhasbeenreportedtoplayimportantrolesinnormalphysiologyaswellasinmanycardiacdiseases.
First,wescreenedtheexpressionofsomelncRNAs,whicharehighlyexpressedintheheart,usinglncRNAarraymethodperformedbyFantomproject(Supple-mentaryTable1).
WedetectedtheexpressionlevelsoflncRNAsincardiomyocytesusingqRT-PCR.
AmongthoselncRNAs,velncRNAsweresubstantiallydecreaseduponH2O2treatment(SupplementaryTable1).
ToidentifywhichlncRNAisinvolvedintheregulationofp53,weperformedRNAimmunoprecipita-tion(RIP)experiments.
Theresultsshowedthatp53antibodysignicantlyenrichedwithAK020546,whichwenamedasCAIF,inculturedprimarycardiomyocytes(Fig.
4a)aswellasinvivo(SupplementaryFig.
6a).
ThisresultindicatesthatthereisastronginteractionbetweenCAIFandthep53protein.
Tovalidatetheirinteraction,wegeneratedabiotinylatedprobeofCAIFandutilizedforRNApull-downassay.
TheresultsshowedthatCAIFprobestronglyboundwithp53incardiomyocytes(Fig.
4b)andinthemicehearttissue(SupplementaryFig.
6b),whichconrmsthatCAIFisabletodirectlybindwithp53.
Toexaminethep53-bindingregionofCAIF,weperformeddeletion-mappingexperimentswithtruncatedCAIF.
TheRNApull-downexperi-mentswithtruncatedCAIFshowedthatthe209ntregionatthe3′-endofCAIFisrequiredforthespecicinteractionwithp53(SupplementaryFig.
6c).
AstheCAIFdirectlyinteractswithp53,wenextexaminedwhetherCAIFisinvolvedintheregulationofp53-dependentexpressionofmyocardin.
WedetectedthatCAIFisdistributedindifferentcelltypesincludingbroblastsandendothelialcellsinabcdf250bp100bpp53CHIP:MgeMyocardinlevels++CAIF-siRNACAIF-sc–––––––––––––––––––––––––––––––––––––––012345P<0.
001p53+CAIFβ-gal+++++Myocardinlevels012345P=0.
99P<0.
001P<0.
001P<0.
001Luciferaseactivity(foldinduction)CAIF-scpGL4.
17Mycd-wtCAIF-siRNA++++++051015P<0.
001P<0.
001h0AK003290AK149029AK006774AK020546AK076876204060lgGp53*BoundRNA(%toinput)CAIF-siRNARandomprobeCAIF-DNAprobeInputControlCAIF-scCAIF-siRNANegativeCAIF-scInput++++++Actinp53ActinPulldownBio-CAIFBio-NCIIPPp53MyocardinActin++CAIF-siRNACAIF-sc36kDa9536kDa55Fig.
4CAIFbindstop53andblocksp53-mediatedmyocardintranscription.
aTheqRT-PCRshowinglncRNAenrichmentbyananti-p53antibody.
CardiomyocytesweresubjectedtoRIPassayusingananti-p53antibodyorIgG.
IP-enrichedRNAwasthenanalyzedbyqRT-PCR.
n=4.
bWesternblotshowingp53proteininteractionwithCAIF.
RNApull-downassaywasperformedincardiomyocytesusingbiotin-labeledCAIFprobe(Bio-CAIF)andnegativecontrolprobe(Bio-NC).
c,dKnockdownofCAIFincreasesmyocardinexpression.
CardiomyocyteswereinfectedwithadenovirusCAIF-siRNAorCAIF-sc.
MyocardinmRNAlevelswereanalyzedbyqRT-PCR(c).
n=5.
Myocardinproteinlevelswereanalyzedbywesternblot(d).
eCAIFinhibitsp53-mediatedmyocardinexpression.
Cardiomyocyteswereinfectedwithadenovirusp53,CAIForβ-gal.
Forty-eighthoursafterinfectionmyocardinlevelswereanalyzedbyqRT-PCR.
n=4.
fKnockdownofCAIFdecreasesitsinteractionwithp53.
CardiomyocyteswereinfectedwithadenovirusCAIF-siRNAorCAIF-sc.
Forty-eighthoursafterinfectionRNApull-downassaywasperformedandp53proteinwasdetectedbywesternblot.
gCAIFknockdownenhancesthebindingofp53tothepromoterofmyocardin.
CardiomyocyteswereinfectedwithadenovirusCAIF-siRNAorCAIF-sc.
Forty-eighthoursafterinfectionCHIPwasperformedwithp53antibody.
hKnockdownofCAIFcausestheincreaseofmyocardintranscriptionalactivity.
CardiomyocytesweretreatedwiththeadenoviralCAIF-siRNAorCAIF-sc,theconstructsoftheemptyvector(pGL-4.
17)andthewild-typemyocardinpromoter(Mycd-wt).
Luciferaseactivitywasassayed.
n=4ARTICLENATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-y6NATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunicationstheheart,whichrevealsthattheexpressionofCAIFisnotspecictocardiacmyocytes(SupplementaryFig.
7a).
Inculturedprimarycardiomyocytes,theknockdownofCAIF(SupplementaryFig.
7b)signicantlyincreasedthelevelsofmyocardinmRNAandprotein(Fig.
4c,d),whileoverexpressionofCAIF(SupplementaryFig.
7c)signicantlydecreasedtheexpressionofmyocardin(Supplemen-taryFig.
7d,e).
Similarly,thesilencingofCAIFwithsiRNA(SupplementaryFig.
7f)remarkablyenhancedthemyocardinlevel(SupplementaryFig.
7g).
Inaddition,ourdatashowedthattheoverexpressionofCAIFcounteractedthepromotiveeffectofp53onmyocardinexpression(Fig.
4e).
Next,wesoughttoexplorethemechanismbywhichCAIFinhibitsp53-mediatedmyocardinexpression.
Forthisexperiment,wesynthesizedbiotinylatedDNAprobecomplementarytoCAIFandRNApull-downassaywascarriedoutwithbiotinylatedDNAprobe.
Then,thewesternblotanalysisforp53wasdone.
OurresultsshowedthatCAIFknockdowncausedasignicantdecreaseinthebindingofp53(Fig.
4f).
TheCHIPassayresultsdemonstratethatCAIFknockdownenhancedthebindingofp53tothepromoterofmyocardin(Fig.
4g)anditsignicantlyincreasedthemyocardintranscriptionalactivity(Fig.
4h).
WealsofoundthatthebindingofCAIFtop53isremarkablydecreasedintheheartwithI/Rinjury(SupplementaryFig.
7h),whichmightleadtoincreasedassociationofp53withmyocardinpromoterasshowninSupplementaryFig.
5d.
Takentogether,ourdatasuggestthatCAIFbindstop53invivo,andthisinteractionblocksp53-mediatedmyocardintranscription.
CAIFblocksautophagicsignalincardiomyocytesandinvivo.
WenextinvestigatedthefunctionofCAIFduringautophagiccelldeathincardiomyocytes.
IncardiomyocytesexposedtoH2O2,thelevelofCAIFwassignicantlydecreasedalongwithincreasedlevelofautophagy(Fig.
5a).
TheoverexpressionofCAIFsig-nicantlyinhibitedH2O2-inducedautophagy(Fig.
5b–dandSupplementaryFig.
8a).
TodeterminethefunctionalrelevanceoftheinteractionbetweenCAIFandp53andtheinductionofautophagy,weobservedtheabilityofdifferenttruncatedversionsofCAIFtoregulateH2O2-inducedautophagy.
Ourresultsfoundthatthedeletionof209ntregionatthe3'endofCAIFfailedtoinhibitautophagyinducedbyH2O2(SupplementaryFig.
8b).
Thisresultindicatesthatthe209ntfragmentatthe3′-endofCAIFaloneissufcienttoinhibitautophagy.
Inaddition,theover-expressionofCAIFremarkablyinhibitedH2O2-inducedcelldeathincardiomyocytes(Fig.
5d,lowerpanel).
Incontrast,theknockdownofCAIFsignicantlyincreasedH2O2-inducedcelldeath,whichwasinhibitedbytreatmentwith3-MA(Supple-mentaryFig.
8c).
ThisdatasuggestthattheexpressionofCAIFisrequiredtoblockautophagiccelldeathincardiomyocytes.
InmicemodelofI/Rinjury,theexpressionofCAIFwassignicantlydecreased(Fig.
5e).
TheadministrationofCAIF(SupplementaryFig.
9a)attenuatedautophagy(Fig.
5fandSupplementaryFig.
9b)andinfarctionsizeinducedbyI/Rinjuryinmice(Fig.
5g).
TheechocardiographyresultsindicatethatCAIFadministrationimprovedI/Rinjuryinduceddysregulationintheheartfunction(SupplementaryFig.
9c,d).
Together,thesedatademonstratethattheexpressionofCAIFisessentialtoinhibitautophagyinducedcardiomyocytecelldeathandcardiacdysfunctioncausedbyischemicinjury.
CAIFregulatesautophagythroughcontrollingp53andmyo-cardin.
Next,weperformedtheexperimentstoconrmwhethermyocardinisadownstreamtargetofCAIFduringautophagy.
Wedetectedmyocardinexpressionandtheresultsshowedthatknockdownofp53(Fig.
6a)reducesthepromotingeffectofCAIFknockdownonexpressionofmyocardin(Fig.
6b).
TheknockdownofCAIFsignicantlyenhancedH2O2-inducedmyo-cardinexpression,butco-transfectionofp53siRNAsignicantlyattenuatedtheeffectofCAIFknockdownonmyocardinexpres-sion(Fig.
6c,upperpanel).
WealsofoundthatknockdownofCAIFaggravatedH2O2-inducedautophagyandcelldeathincardiomyocytes.
However,thesilencingofp53expressionatte-nuatedtheeffectsofCAIFknockdownonautophagy(Fig.
6c,lowpanel)andcelldeath(Fig.
6d).
Together,thesedatasuggestthatCAIFexertsitsinhibitoryeffectonautophagyandcelldeaththroughregulatingp53activityandmyocardinexpression.
DiscussionAutophagyisgenerallyregardedasanadaptiveresponsetostressthathelpstomaintaintheintracellularhomeostasisbyremovingdamagedproteinsandintracellularorganelles.
Theautophagycanprotectcellsfromdifferentkindsofinjuries20,21,22.
However,excessiveordefectiveautophagyisdetrimentaltocellsbycausinglargescaleaccumulationofautophagicproductssuchasautop-hagosomesaswellasdegradationofvitalproteinsandorganellesinvolvedinsurvival.
Currentstudiessuggestthatautophagyiseitherprotectiveordetrimentaltothecardiactissuedependsuponthedegreeofinductionanddurationoftheinjury.
Underischemiccondition,theaccelerationofautophagyduringreper-fusioncancausedamagetocardiomyocytesanditenhancescardiaccelldeath23.
Itiswellknownthatoxidativestressandactivationofautophagyaremajorplayersinthecardiomyocytecelldeathandcardiacinjuryassociatedheartdysfunction24.
Inthepresentstudy,wedemonstratethatoxidativestressinducedincreaseofp53promotestheexpressionofmyocardin,whichisinvolvedintheactivationofautophagicprocess.
Theover-activationofautophagybyp53-myocardinsignalingleadstocardiomyocytecelldeathandaccelerationofischemicinjury.
Mostinterestingly,wefoundthatacardiacspecicexpressionofanlncRNA,namedasCAIF,caninhibitp53-dependentexpres-sionofmyocardinandautophagyinductionincardiomyocytes.
However,thedepletionofCAIFexpressionincardiomyocytesduringI/Rinjuryleadstoautophagiccelldeathandcardiacdysfunction.
Thisistherststudytoillustratethatp53-myocardindependentactivationofautophagyisinvolvedincardiomyocytecelldeathandCAIFplayindispensibleroleingoverningthismolecularaxistocontrolautophagicprocess.
LncRNAsarecriticalregulatorofvariouscellularprocesses.
Theyplayanimportantroleinbiologicalfunctionsandthealterationsinexpressioniswellconnectedtodevelopmentandprogressionofvariousdiseases25–28.
SeverallncRNAshavebeenreportedtoparticipateintheregulationofautophagy29,30.
However,veryfewstudiesdemonstratetheregulatoryfunctionsoflncRNAsinautophagyintheheart.
Inthepresentstudy,wehaveidentiedanewlncRNA(CAIF),whichhasthecapabilitytoregulateautophagyincardiomyocytes.
OurresultsfoundthatenforcedexpressionofCAIFinhibitsautophagiccelldeathandattenuatesmyocardialinfarctionsizesintheI/Rheart.
OurresultsrevealthatCAIFcouldbeapotentialtherapeutictooltopreventthedefectiveautophagy-mediatedlossofcardiacmyocytesaswellasforthetreatmentofmyocardialinfarctionandheartfailure.
LncRNAsregulatestheexpressionandactivityoftargetmoleculesthroughdifferentmechanismsincludingepigeneticmodication31–33,proteindegradation34,RNAstability,transla-tionandsplicingandtrafcking8.
LncRNAsalsoparticipateintheregulationofRNAtranscription.
Forexample,thelncRNAXistmediatedtranscriptionalsilencingbyrecruitingpolycombrepressivecomplex2(PRC2)acrosstheXchromosome35.
TheH19RNArecruitsMBD1,interactswithmethyltransferases,inducesrepressivehistonemodicationinthetargetgenesandnallycontrolstargetgenesexpression36.
ThelncRNAHOTAIRNATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-yARTICLENATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunications7mediatestheinteractionbetweenSnailandenhancerofzestehomolog2(EZH2),andrepressesdownstreamtargets37.
Like-wise,thenewlyidentiedlncRNA,CAIF,regulatestheexpressionofmyocardin.
However,theregulatorymechanismofgeneexpressionisdifferingfromotherlncRNAs.
Surprisingly,lncRNACAIFdirectlyinteractswithp53,atranscriptionfactorandblocksitsbindingtopromoterregionofmyocardin.
Ourndingindi-catesthatlncRNAcaninterferegeneexpressionbyformingcomplexwithtranscriptionfactorsandinhibitingtheiractivity.
Inaddition,ourstudysuggeststhatlncRNAscouldhavediversemechanismforregulatinggeneexpressionbeyonditswelldenedepigeneticregulatoryfunctions.
Thus,furthercharacterizationoflncRNAsignalingpathwaysandregulatorymechanismsoflncRNAwillprovideabetterunderstandingaboutlncRNA-mediatedgeneregulation.
Currently,itisunknownaboutthefactorsassociatedwiththesuppressionoflevelofCAIFunderoxidativestressconditionaswellasinI/Rinjury.
ThedepletionofCAIFunderI/Rconditioncouldbeeitherduetoinhibitionofitsexpressionorincreaseddegradation.
Furtherdetailedstudiesarewarrantedtoaddressthisphenomenon.
Somepreviousstudiesshowthatp53suppressesautophagyandinhibitionofp53degradationpreventstheactivationofagcebdArea-at-riskRemoteareaCAIFlevels0.
00.
51.
01.
5600I/R(min)Sham60**fNNNNControlControl0624Time(h)H2O2H2O2H2O2H2O2CAIFlevels0.
00.
51.
01.
5P=0.
001P=0.
05P<0.
001β-gal%CellwithGFP-LC3vacuolesCAIF+–––––––++++P<0.
001P<0.
001P<0.
00101020304050Autophagicvacuoles/cellβ-galCAIF+––––––––––––––++++0510152025P<0.
001P<0.
001P<0.
001Celldeath(%)β-galCAIF+++++01020304050P<0.
001P<0.
001P<0.
001TotalareaoccupiedbyAV(%)++––––––––––+++I/R+Shamβ-galCAIF++––––––––––+++I/R+Shamβ-galCAIF0510152025P<0.
001P<0.
001P<0.
001INF/LV(%)010203040P<0.
001P<0.
001P<0.
001ActinLC3-ILC3-IIkDa173612H2O2H2O2H2O2+β-galH2O2+CAIFH2O2+CAIFH2O2+β-gal30Fig.
5CAIFinhibitsautophagyincardiomyocytesandinvivo.
aH2O2inducesadecreaseinCAIFlevels.
CardiomyocytesweretreatedwithH2O2atindicatedtime.
CAIFwereanalyzedbyqRT-PCR.
n=4.
bCAIFinhibitspunctateaccumulationsofGFP-LC3.
CardiomyocyteswereinfectedwithadenoviralGFP-LC3,CAIForβ-gal.
Twenty-fourhoursafterinfectioncellsweretreatedwithH2O2.
RepresentativeimagesshowGFP-LC3staining(leftpanel),Bar=20μm.
ThepercentageofcellswithGFP-LC3punctawasquantiedin(rightpanel).
n=5.
cCAIFreducesautophagicvacuolesincardiomyocytes.
CellswereinfectedwithadenoviralCAIForβ-gal.
Twenty-fourhoursafterinfectioncellsweretreatedwithH2O2.
RepresentativeEMimageswereshown(leftpanel).
Bar=2M.
Thearrowsdepictautophagosomes,andthenucleusisdenotedbyN.
Quanticationofautophagicvacuoleswasshowninrightpanel.
n=4.
dCAIFinhibitstheconversionofLC3-ItoLC3-IIandreducescelldeathinresponsetoH2O2.
Cardiomyocytesweretreatedasdescribedin(c).
RepresentativeimmunoblotforconversionofLC3-ItoLC3-IIwasshown(upperpanel).
Quantitationofcelldeathwasshown(lowpanel).
n=4.
eMiceweresubjectedtoischemiaatindicatedtimeand3hreperfusion(I/R).
Area-at-riskandtheremotezonewerepreparedforqRT-PCRanalysisofCAIFlevels.
n=6.
*p<0.
05vs.
sham.
fCAIFreducedautophagyuponI/Rinjury.
MicewereinjectedwithCAIForβ-galasdescribedinmethods,thenweresubjectedto45minischemiaand3hreperfusion(I/R).
Quanticationofautophagicvacuolesinthearea-atriskwasshown.
n=6.
gCAIFreducesmyocardialinfarctionuponI/Rinvivo.
Miceweretreatedasdescribedin(f).
Theinfarctsizeswerequantied.
LVleftventricle,INFinfarctarea.
n=7ARTICLENATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-y8NATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunicationsautophagyinHCT116coloncancercellsandHeLacell12.
Otherstudiessuggestthatp53isabletoinduceautophagy38–40.
Inconsistentwiththesereports,ourdatashowthatp53promotesautophagy.
Ourresultsdemonstratethattheinhibitionofp53withsiRNAattenuatesautophagiccelldeathandmyocardialinfarctionsizesintheheartwithI/Rinjury,whileoverexpressionofp53enhancedH2O2-inducedautophagiccelldeathandinhi-bitionofautophagywith3-MAblockedthisresponseincardio-myocytes.
Thesendingsconrmtherelationshipbetweenp53andautophagyininductionofcelldeathanditisofgreatimportancetounderstandmanyphysiologicalandpathologicalprocesses.
Asthedysregulationofp53activityandautophagyplaycrucialroleintheprogressionofcardiacinjury,ourstudystrengthenthenotionthatp53mediatedsustainedactivationofautophagyisdetrimentaltocellsandtissuefunction.
p53isasequence-specicDNA-bindingtranscriptionfactorandmanygenescanberegulatedbyp53attranscriptionallevel.
Ourcurrentresultsdemonstratethatmyocardinfunctionsasadownstreamtargetofp53inregulatingautophagy.
p53promotesmyocardinexpressionandregulatescardiomyocyteautophagy.
Severalpreviousstudiessuggestthatp53canregulatemyocardin.
p53regulatessmoothmusclecellsdifferentiationthroughtar-getingmyocardin41.
InVSMC,thetreatmentofantioxidantresveratrolactivatesp53signalinganddownregulatedthetran-scriptionofmyocardin42.
Giventhefactthattheactivityandresponseoftranscriptionfactorsvarydependsontheirupstreamactivatorsanditalsodiffersinvariouscelltypes.
Inthiscontext,itispossiblethatp53activatesmyocardinexpressionincardio-myocytesbutithasinhibitoryfunctionindifferentiatingsmoothmusclecells.
Afurtherdetailedstudywillberequiredtodelineatethisvariationofp53-dependentexpressionofmyocardinindif-ferentcelltypes.
Itiswellknownthatmyocardinhasimportantfunctionsintheheartandcontributestothecardiacgrowth,chambermaturation,andembryonicsurvival.
Myocardinisinvolvedincellprolifera-tion,migration,andmyogenesisbyinteractingwithserumresponsefactor(SRF)13,43.
Itactsasasmoothmuscleandcardiacmuscle-specictranscriptionalactivatorandtransactivatesthecadb+++++++++++%CellwithGFP-LC3vacuolesH2O2H2O2p53-siRNAp53-scCAIF-siRNACAIF-sc020406080P<0.
001P=0.
003P=0.
004Celldeath(%)+++++++++++p53-siRNAp53-scCAIF-siRNACAIF-sc020406080P<0.
001P=0.
003P=0.
004Myocardinlevels+++++p53-siRNAp53-scCAIF-siRNACAIF-sc012345P=0.
99P<0.
001P<0.
001MyocardinActin36kDa95MyocardinActin36kDa95p53Actin++––––––––––––+––––––––––––––––++++p53-siRNAp53-scCAIF-siRNACAIF-sc36kDa55––––––––––––––––––––––––––––––––––––––Fig.
6CAIFexertsitsautophagiceffectthroughp53andmyocardin.
a,bKnockdownofp53reducesthepromotingeffectofCAIFknockdownonmyocardinexpression.
CardiomyocyteswereinfectedwithadenoviralCAIF-siRNAorCAIF-sc,p53-siRNAorp53-sc.
p53expressionlevelswereanalyzedbyimmunoblot(a).
Myocardinexpressionlevelswereanalyzedbyimmunoblot(b,upperpanel)andbyqRT-PCR(b,lowpanel).
n=5.
c,dKnockdownofp53attenuatesthepromotingeffectofCAIFknockdownonautophagyandcelldeathinducedbyH2O2.
CardiomyocyteswereinfectedwithadenoviralCAIF-siRNAorCAIF-sc,p53-siRNAorp53-sc,andthenexposedtoH2O2.
Myocardinlevelswereanalyzedbyimmunoblot(c,upperpanel).
AutophagywasassessedbyGFP-LC3staining.
ThepercentageofcellswithGFP-LC3punctawasquantiedin(c,lowpanel).
n=5.
Quantitationofcelldeathisshownin(d).
n=5NATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-yARTICLENATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunications9expressionofatrialnatriureticfactor(ANF),myosinlightchain(MLC)-2V,andα-MHCgenes,aswellastheNkx2.
5gene13.
Ourstudyforthersttimedemonstratesthefunctionofmyocardininautophagyinduction.
OurdatashowthatmyocardinparticipatesintheregulationofcardiomyocytesautophagyuponH2O2orI/Rtreatment.
Inaddition,wealsodemonstratethatmyocardininducesanincreasedexpressionofBeclin1,whichisrequiredfortheexecutionofautophagy,indicatingthatthereisacross-talkbetweenmyocardinandBeclin1.
WewillattempttofurtherclarifytherelationshipbetweenmyocardinandBeclin1intheautophagicmachineryinfuturestudies.
MethodsTreatmentandcultureofcardiomyocytes.
Weisolatedcardiomyocytesfrom1-to2-days-oldmice.
First,heartswerewashedandmincedinHEPES-bufferedsalinesolutionafterdissectedfromthemice.
Tissueswerethendispersedinaseriesofincubationsat37°CinHEPES-bufferedsalinesolutioncontaining1.
2mg/mlpancreatinand0.
14mg/mlcollagenase.
Lysateswerecollectedandcentrifugedat200*gfor5min.
Aftercentrifugation,precipitatedcellswerere-suspendedinDulbecco'smodiedEaglemedium/F-12(GIBCO)supplementedwith5%heat-inactivatedhorseserum,ascorbate(0.
1mM),insulin-transferring-sodiumselenitemediasupplement(Sigma,St.
Louis,MO),penicillin(100U/ml),streptomycin(100μg/ml),andbromodeoxyuridine(0.
1mM).
Thecellswerepre-platedat37°Cfor1h.
Dissociatedcellswerethendilutedto1*106cells/mlandplatedin10μg/mllaminin-coateddifferentculturedishes.
Incelldeathassay,weused200MH2O2totreatisolatedcardiomyocytesinvitro.
CellswerethenstainedbyTrypanBlue.
AndwecountedthenumbersofTrypanBlue-positiveandTrypanBlue-negativecellsusinghemocytometeranddeterminedtherateofcelldeath.
AutophagicuxwasanalyzedincardiomyocytesinfectedwithadenovirusexpressingtandemRFP-GFP-LC3andthentreatedasindicated.
ThecellswerewashedwithPBS,xedwith4%paraformaldehydeandvisualizedusingaZeissLSM-510confocallaser-scanningmicroscope.
LC3greenandreddotswerequantiedusingtheImageJsoftware.
Aminimumof50cellswerescoredforeachcondition.
Anobserverblindedtosampleidentityassessedtheautophagicux.
TherawdataformeasurementofautophagicuxareshowninSupplementaryDataset1.
Adenoviralconstructionsandinfection.
WeusemousecDNAasthetemplatetoobtainthefulllengthofCAIFbyPCR.
Theforwardprimerwas5′-TCGTGAATTTTGTCAGTTTTGTGATATCC-3′;thereverseprimerwas5′-AGGTAAGTTTAACTGGTCAGGAAATAAAC-3′.
MousemyocardincDNAwaspurchasedfromOrigene.
WeusetheAdeno-Xexpressionsystem(Clontech)toconstructadenovirusesharboringtheCAIF,myocardin,andβ-galactosidase(β-gal),respectively.
ThemouseCAIF-siRNAtargetsequenceis5′-GGGACGCTTGTGCAAACTT-3′.
Ascrambleformwasusedasacontrol,5′-GTATCGCGAGTAGTCAGTC-3′.
ThemousemyocardinsiRNAtargetsequenceis5′-GCACACGAAGGCCATTCTT-3′.
Ascrambleformwasusedasacontrol,5′-CTCAGACTGCAGTACGACT-3′.
Beclin1siRNAtargetsequenceis5′-GATCCTGGACCGGGTCACC-3′;thescramblesequenceis5′-CCTAGGCTCACGTGACGCG-3′.
WeusethepSi-lenceradeno1.
0-CMVSystem(Ambion)togenerateadenovirusescarryingCAIF-siRNA,myocardinsiRNA,Beclin1siRNAandtheirscrambleformsaccordingtotheKit'sinstructions.
AndHEK293cellswereusedtoamplifyallconstructs.
Immunoblot.
Immunoblotwasperformedaccordingtothefollowingprocedure.
Cellswerelysedinthelysisbufferfor1hat4°C.
Componentsofthelysisbufferare20mmol/LTrispH7.
5,2mmol/LEDTA,3mmol/LEGTA,2mmol/Ldithiothreitol(DTT),250mmol/Lsucrose,0.
1mmol/Lphenylmethylsulfonyluoride,1%TritonX-100,andaproteaseinhibitorcocktailisalsoincluded.
Sampleswereloadedintothewellsof10or12%SDS-PAGE,andtransferredtonitrocellulosemembranesafterelectrophoresis.
Blotswereprobedusingthepri-maryantibodies.
Theanti-Myocardinantibody(1:500,Abcam,ab203614),anti-p53antibody(1:500,Abcam,ab131442),anti-Actinantibody(1:2000,Abcam,ab6276),andanti-LC3antibody(1:500,Abcam,ab192890)wereusedinthisstudy.
Afterincubatedwiththeprimaryantibody,wewashthemembraneusingPBSforfourtimesandaddedthehorseradishperoxidase-conjugatedsecondaryantibodies.
Antigen–antibodycomplexeswerevisualizedbyenhancedchemiluminescence.
UncroppedblotsareshowninSupplementaryFigs.
10and11.
Quantitativereal-timePCR(qRT-PCR).
CAIFwasquantiedusingSYBRGreenreal-timePCR.
ThesequencesofCAIFprimerswereforward:5′-CTTCACTCCTGCAAATGTGTT-3′;reverse:5′-TTATAGTGGGATGGGCAGTTT-3′.
qRT-PCRformyocardinwasperformedaswedescribed44.
Thesequencesofmyocardinprimerswereforward:5′-TCAGTCTTA-CAGTTACGGCTT-3′;reverse:5′-GCACTATCCAATTGTTTTCTCG-3′.
ThesequencesofATG5primerswereforward:5′-GCGGCCTTTCATCCAGAAGC-3′;reverse:5′-CGACTGCGGAAGGACAGACT-3′.
ThesequencesofATG7primerswereforward:5′-GCCAGGACACCCTGTGAACT-3′;reverse:5′-CCAAGG-CAGCGTTGATGACC-3′.
ThesequencesofATG10primerswereforward:5′-CAGCGGTGGCAGAAGTGATT-3′;reverse:5′-TGTTCCTGCTGGGTGATGGT-3′.
ThesequencesofATG12primerswereforward:5′-GACGTTA-GAGTGTGCCATTA-3′;reverse:5′-GTAAATGCCTTCAGTCATCC-3′.
ThesequencesofBeclin1primerswereforward:5′-GCACCATGCAGGTGAGCTTC-3′;reverse:5′-TTTCGCCTGGGCTGTGGTAA-3′.
Theresultswerestandardizedtocontrolvaluesofglyceraldehyde-3-phosphatedehydrogenase(GAPDH).
GAPDHforwardprimer:5′-TGTGTCCGTCGTGGATCTGA-3′;reverse:5′-CCTGCTTCACCACCTTCTTGA-3′.
WeusedagarosegelelectrophoresistoconrmthespecicityofthePCRamplication.
Therelativeexpressionofdif-ferentsetsofgeneswasquantiedtoGAPDHmRNA.
RNA-bindingproteinimmunoprecipitation(RIP)assay.
WeperformedRIPusingaMagnaRIPRNA-BindingProteinImmunoprecipitationKit(Millipore)accordingtothemanufactuer'sprotocal.
Briey,cardiomyocyteswereharvestedbyaddingRIPlysisbufferandincubatedwithproteinbeadsandp53antibodycomplexovernightat4°C.
Afterwashingoffunboundmaterials,RNAsbindingtop53wereelutedandquantied.
WeusedqRT-PCRtoexaminecertainRNAsco-immunoprecipitatedwiththep53antibody.
RNApull-downassay.
WeusedmpliScribeT7-FlashBiotin-RNATran-scriptionKit(Epicentre,Madison,WI,USA)totranscribeandpurifyBiotin-labeledCAIFprobeinvitro.
Thebiotin-labeledprobewasincubatedwithcellsproteinextractfor2h.
Incubatingthemixturewithstreptavidinagarosebeads(Invitrogen)atroomtemperature(RT)for1h.
Afterstringentwashingwiththewash/bindingbuffer,theretrievedproteinwasanalyzedbywesternblot.
Pull-downassaywithbiotinylatedDNAprobe.
WedissolvedthebiotinylatedDNAprobecomplementarytoCAIFin500lofwash/bindingbuffer(0.
5MNaCl,20mMTris-HCl,pH7.
5,and1mMEDTA).
Streptavidin-coatedmagneticbeads(Sigma)wereincubatedwiththeprobesat25°Cfor2htogenerateprobe-coatedmagneticbeads.
Cardiomyocytelysateswereincubatedwithprobe-coatedbeads.
Afterwashingwiththewash/bindingbuffer,theproteincomplexesboundtothebeadswereeluted,extractedandloadedontoSDS-PAGEforwesternblotanalysis.
CAIFpull-downprobe,5′-GCTACTGCACACTCAATTCTGGGAGACC-3′;andrandompull-downprobe,5′-TGATGTCTAGCGCTTGGGCTTTG-3′.
Chromatinimmunoprecipitation(ChIP)assay.
ChIPassaywasperformedasfollows.
Inbrief,cellswerewashedwithPBSandxedfor10minwith1%for-maldehydeatroomtemperature.
Thecross-linkingreactionwasquenchedusing0.
1Mglycineandtreatedfor5min.
CellswerewashedtwicewithPBSandlysedfor1hat4°Cinalysisbuffer.
Thecelllysatesweresonicatedintochromatinfrag-mentswithanaveragelengthof500to800bp.
ForChIPs,thesampleswerepreclearedwithProtein-Aagarose(Roche)for1hat4°Conarockingplatform.
Afterthat,5μgspecicantibodieswereaddedandrockedforovernightat4°C.
WeusedsalmonspermDNAblockedProtein-Aagarosetocaptureimmunoprecipi-tates.
TheQIAquickSpinKit(Qiagen)wasusedtopurifyDNAfragments.
Thepuriedwasusedasatemplateandampliedwiththefollowingprimersets.
Fortheanalysisofp53bindingtothepromoterregionofmyocardin,theoligonu-cleotideswereforward:5′-TTTCACAGAGTTTCCTCCATG-3′;reverse:5′-TCCCAGCTCATCAAAGAAGA-3′.
TheheartsamplesfrommiceweresubjectedtoCHIPassayusingEpiQuikTissueChromatinImmunoprecipitation(ChIP)Kitaccordingtomanufacturer'sinstructions.
Constructionofmousemyocardinpromoter.
ThemyocardinpromoterwasampliedfrommousegenomicDNAbyPCR.
Theforwardprimerwas5′-TTTCACAGAGTTTCCTCCATG-3′.
Thereverseprimerwas5′-ACCTTCCTTTTTCCCATTCTC-3′.
ThemyocardinpromoterluciferasereporterplasmidpGL4.
17-MycdwasconstructedbyligatingthemyocardinpromoterregionintothepGL4.
17vector(Promega).
Site-directedmutagenesisintheputa-tivep53-bindingsitewasperformedusingtheQuikChangeIIXLSite-DirectedMutagenesisKit(Stratagene).
Wesequencedtheconstructtomakesurethatonlythedesiredmutationshadbeenintroduced.
Luciferaseactivityassay.
Dual-LuciferaseReporterAssaySystem(Promega)wasusedtoperformluciferaseactivityassayaccordingtothemanufacturer'sinstruc-tions.
TheconstructedpGL4.
17-MycdorpGL4.
17-Mycd–mutweretransfectedintocells(150ng/well)usingLipofectamine2000(Invitrogen).
Then,thosecellswereinfectedwithindicatedadenovirus.
At48hafterinfection,luciferaseactivitywasmeasured.
Transmissionelectronmicroscopy.
Conventionalelectronmicroscopywasper-formedasfollows.
Cellswerexedwith2.
5%glutaraldehydeandthenpostxedwith1%osmiumtetraoxide.
SamplesweredehydratedinanethanolseriesandembeddedinEmbed812resin.
Theultrathinsectionswerecutusingadiamondknife.
Then,thesectionsweremountedoncoppergridsanddouble-stainedwithARTICLENATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-y10NATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunicationsuranylacetateandleadcitrate.
Thenumberofautophagicvacuoleswasdeterminedforaminimumof100cells.
Heartultrastructuralanalysiswasalsoperformed.
TheobservationandmicrophotographsofthesampleswerecarriedoutusingaFEITecnaispirittransmissionelectronmicroscope.
TheinvestigatorwhoassessedtheautophagicvesiclewasblindedtotheEMsampleidentity.
Echocardiographicassessment.
After1weekoftheshamorI/Rsurgery,trans-thoracicechocardiographywasusedonmicetoassesscardiacstructureandfunction.
Echocardiographicparametersweremeasured.
WeuseM-modeecho-cardiographyinmeasuringLeftventricularejectionfraction(LVEF)andLeftventricularend-systolicvolume(LVESV).
Measurementsweretakenfrommorethanthreebeatsandaveraged.
RawdataformeasurementofcardiacfunctionareshowninSupplementaryDataset1.
Afterinvivoechocardiographystudyofcardiacfunction,themiceheartswerecollected,weightedandxedforhistologicalexamination.
Animalexperiments.
ExperimentswereperformedinmaleadultC57BL/6mice(10weeksold),whichwereobtainedfromInstituteofLaboratoryAnimalScienceofChineseAcademyofMedicalSciences(Beijing,China).
Allexperimentscom-pliedwiththeguidingprinciplesforthecareanduseoflaboratoryanimalsinQingdaoUniversityandwereapprovedbytheCommitteeforAnimalExperi-mentation.
Animalsusedinexperimentshadsimilarbodyweight.
Animalsusedforstatisticalanalysishadnormalphysiologicalindexduringsurgicalexperiments.
Computer-generatedrandomnumberswereassignedtoeachmouse.
Weallocatedmicetoeachgroupbychance.
Weperformedintracoronarydeliveryofadenoviruses.
Micewereanesthetizedandintubated,ventilatedwithaHX-300Sanimalventilator.
Theheartwasexposedthroughasmallleftanteriorthoracotomy.
Afterremovalofthepericardialsac,2*1011moiadenovirusesofCAIF,2*1010moiadenovirusesofmyocardin-siRNA,2*1010moiadenovirusesofp53-siRNAor2*1011moiadenovirusesofBeclin1-siRNAwereinjected,respectively,throughacatheterfromtheapexoftheheartleftventricleintotheaorticrootwhiletheaortaandpulmonaryarterieswerecross-clampedfor20s.
Thechestwasthenclosedafterremovalofairandblood.
Then,micewereallowedtorecover.
Fivedaysaftertheinjectionofadenoviruses,themiceweresubjectedtoI/Rsurgery.
Toexaminetheeffectsof3-MA,themicewereinjectedwith3-MA1hbeforeI/Rsurgery.
ExperimentalstrategiesforI/Rinjurymodelareasfollows.
CardiacI/Rinmicewasinducedby45minischemia,followedby3hor1weekreperfusion.
After3hofreperfusion,weinjectedevansbluedye(1mlofa2.
0%solution;Sigma-Aldrich)intojugularveinintotheheart.
Theischemiczoneandthenonischemiczonewasdelineatedbythedye.
Theheartwasrapidlyexcisedandsectionedseriallyinto1mm-thicksections.
Then,sliceswereincubatedin1.
0%2,3,5-triphenyltetrazoliumchloride(Sigma-Aldrich)for15minat37°Ctodeterminetheinfarctarea.
Thestainingwasstoppedbyice-coldsterilesalineandsliceswerexedin10%neutralbufferedformaldehydeforanother15min.
Thesliceswereweighedandbothsidesofeachsliceweredigitallyphotographed.
Theareasofinfarction(INF)andnonischemicleftventricle(LV)weremeasuredanddeterminedusingcomputerizedplanimetry(NIHImage1.
57)byanobserverblindedtothesampleidentity.
RawdataformeasurementofinfarctionsizeareshowninSupplementaryDataset1.
Toassessinductionofautophagyinvivo,adenovirusRFP-GFP-LC3wasadministeredintotheinfarctborderzoneofmiceafterI/R.
Micewerekilled7dayslaterandGFPandRFPsignalswereanalyzedonfrozensectionsbyconfocalmicroscopy.
LC3greenandreddotswerequantiedusingtheImageJsoftware.
RawdataformeasurementofautophagyareshowninSupplementaryDataset1.
Autophagicuxwasassessedbywesternblot.
BalomycinA1(BafA1)wasadministeredtomicebyintraperitonealinjection(1.
5mg/kg)2hbeforetheywereeuthanized.
Tolimitcircadianvariabilityofautophagy,micewereeuthanizedatthesametimeofday.
Statisticalanalysis.
Eachcellularexperimentalgroupwasrepeatedforatleastthreetimes.
Andeachanimalgroupwasrepeatedforatleastvetimes.
Thedataareexpressedasthemean±SDcalculatedbyGraphPadPrismorSPSS.
Formultiplecomparisons,one-wayanalysisofvariance(ANOVA)followedbyTukeyposthoctestwasperformed.
TheresultswereconsideredstatisticallysignicantwhenP<0.
05.
AllstatisticalanalyseswereperformedwithGraphPadPrismVersion6(GraphPadSoftwareInc.
,SanDiego,CA,USA)andSPSSpackage(SPSSInc.
,Chicago,IL,USA).
Dataavailability.
Allrelevantdataareprovidedinthisarticleanditssupple-mentaryinformationles,orareavailablefromthecorrespondingauthorsuponreasonablerequest.
Received:25December2016Accepted:14November2017References1.
Yamamoto,S.
,Sawada,K.
,Shimomura,H.
,Kawamura,K.
&James,T.
N.
Onthenatureofcelldeathduringremodelingofhypertrophiedhumanmyocardium.
J.
Mol.
CellCardiol.
32,161–175.
(2000).
2.
Tanaka,Y.
etal.
AccumulationofautophagicvacuolesandcardiomyopathyinLAMP-2-decientmice.
Nature406,902–906.
(2000).
3.
Maejima,Y.
etal.
Mst1inhibitsautophagybypromotingtheinteractionbetweenBeclin1andBcl-2.
Nat.
Med.
19,1478–1488.
(2013).
4.
Guttman,M.
etal.
LincRNAsactinthecircuitrycontrollingpluripotencyanddifferentiation.
Nature477,295–300.
(2011).
5.
Klattenhoff,C.
A.
etal.
Braveheart,alongnoncodingRNArequiredforcardiovascularlineagecommitment.
Cell152,570–583.
(2013).
6.
Yildirim,E.
etal.
XistRNAisapotentsuppressorofhematologiccancerinmice.
Cell152,727–742.
(2013).
7.
Gartler,S.
M.
&Riggs,A.
D.
MammalianX-chromosomeinactivation.
Annu.
Rev.
Genet.
17,155–190.
(1983).
8.
Wang,K.
C.
&Chang,H.
Y.
MolecularmechanismsoflongnoncodingRNAs.
Mol.
Cell43,904–914.
(2011).
9.
Ishii,N.
etal.
Identicationofanovelnon-codingRNA,MIAT,thatconfersriskofmyocardialinfarction.
J.
Hum.
Genet.
51,1087–1099.
(2006).
10.
Viereck,J.
etal.
LongnoncodingRNAChastpromotescardiacremodeling.
Sci.
Transl.
Med.
8,326ra322.
(2016).
11.
Foster,S.
S.
,De,S.
,Johnson,L.
K.
,Petrini,J.
H.
&Stracker,T.
H.
Cellcycle-andDNArepairpathway-speciceffectsofapoptosisontumorsuppression.
Proc.
Natl.
Acad.
Sci.
USA109,9953–9958.
(2012).
12.
TasdemirE.
etal.
Regulationofautophagybycytoplasmicp53.
Nat.
CellBiol.
10,676-687(2008).
13.
Wang,D.
etal.
Activationofcardiacgeneexpressionbymyocardin,atranscriptionalcofactorforserumresponsefactor.
Cell105,851–862.
(2001).
14.
Huang,J.
etal.
Myocardinisrequiredforcardiomyocytesurvivalandmaintenanceofheartfunction.
Proc.
NatlAcad.
Sci.
USA106,18734–18739.
(2009).
15.
Huang,J.
etal.
MyocardinregulatesBMP10expressionandisrequiredforheartdevelopment.
J.
Clin.
Invest.
122,3678–3691.
(2012).
16.
RuoziG.
etal.
AAV-mediatedinvivofunctionalselectionoftissue-protectivefactorsagainstischaemia.
Nat.
Commun.
6,7388(2015).
17.
Li,D.
L.
etal.
Doxorubicinblockscardiomyocyteautophagicuxbyinhibitinglysosomeacidication.
Circulation133,1668–1687.
(2016).
18.
Hariharan,N.
etal.
DeacetylationofFoxObySirt1playsanessentialroleinmediatingstarvation-inducedautophagyincardiacmyocytes.
Circ.
Res.
107,1470–1482.
(2010).
19.
Ma,X.
etal.
Impairedautophagosomeclearancecontributestocardiomyocytedeathinischemia/reperfusioninjury.
Circulation125,3170–3181.
(2012).
20.
Gupta,M.
K.
etal.
UBC9-mediatedsumoylationfavorablyimpactscardiacfunctionincompromisedhearts.
Circ.
Res.
118,1894–1905.
(2016).
21.
Pattison,J.
S.
,Osinska,H.
&Robbins,J.
Atg7inducesbasalautophagyandrescuesautophagicdeciencyinCryABR120Gcardiomyocytes.
Circ.
Res.
109,151–160.
(2011).
22.
Gottlieb,R.
A.
&Mentzer,R.
M.
Autophagyduringcardiacstress:Joysandfrustrationsofautophagy.
Annu.
Rev.
Physiol.
72,45–59.
(2010).
23.
Matsui,Y.
etal.
Distinctrolesofautophagyintheheartduringischemiaandreperfusion:RolesofAMP-activatedproteinkinaseandBeclin1inmediatingautophagy.
Circ.
Res.
100,914–922.
(2007).
24.
Nakai,A.
etal.
Theroleofautophagyincardiomyocytesinthebasalstateandinresponsetohemodynamicstress.
Nat.
Med.
13,619–624.
(2007).
25.
LiuX.
etal.
LncRNANBR2engagesametaboliccheckpointbyregulatingampkunderenergystress.
Nat.
CellBiol.
18,431–442(2016).
26.
Brown,C.
J.
etal.
ThehumanXISTgene:Analysisofa17kbinactiveX-specicRNAthatcontainsconservedrepeatsandishighlylocalizedwithinthenucleus.
Cell71,527–542.
(1992).
27.
Rinn,J.
L.
etal.
FunctionaldemarcationofactiveandsilentchromatindomainsinhumanHOXlocibynoncodingRNAs.
Cell129,1311–1323.
(2007).
28.
Wang,K.
C.
etal.
AlongnoncodingRNAmaintainsactivechromatintocoordinatehomeoticgeneexpression.
Nature472,120–124.
(2011).
29.
Pawar,K.
,Hanisch,C.
,PalmaVera,S.
E.
,Einspanier,R.
&Sharbati,S.
DownregulatedlncRNAMEG3eliminatesmycobacteriainmacrophagesviaautophagy.
Sci.
Rep.
6,19416(2016).
30.
Deng,X.
etal.
PM2.
5exposure-inducedautophagyismediatedbylncRNAloc146880whichalsopromotesthemigrationandinvasionoflungcancercells.
Biochim.
Biophys.
Acta1861,112–125(2016).
31.
Tsai,M.
C.
etal.
LongnoncodingRNAasmodularscaffoldofhistonemodicationcomplexes.
Science329,689–693.
(2010).
32.
Sado,T.
,Hoki,Y.
&Sasaki,H.
TsixsilencesXistthroughmodicationofchromatinstructure.
Dev.
Cell9,159–165.
(2005).
33.
DiRuscio,A.
etal.
DNMT1-interactingrnasblockgene-specicDNAmethylation.
Nature503,371–376.
(2013).
34.
Li,J.
etal.
LongnoncodingRNANRONcontributestoHIV-1latencybyspecicallyinducingtatproteindegradation.
Nat.
Commun.
7,11730(2016).
NATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-yARTICLENATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunications1135.
Zhao,J.
,Sun,B.
K.
,Erwin,J.
A.
,Song,J.
J.
&Lee,J.
T.
PolycombproteinstargetedbyashortrepeatRNAtothemouseXchromosome.
Science322,750–756.
(2008).
36.
Monnier,P.
etal.
H19lncRNAcontrolsgeneexpressionoftheimprintedgenenetworkbyrecruitingMBD1.
Proc.
NatlAcad.
Sci.
USA110,20693–20698.
(2013).
37.
BattistelliC.
etal.
TheSnailrepressorrecruitsEZH2tospecicgenomicsitesthroughtheenrollmentofthelncRNAHOTAIRinepithelial-to-mesenchymaltransition.
Oncogene36,342–355(2016).
38.
Crighton,D.
etal.
DRAM,ap53-inducedmodulatorofautophagy,iscriticalforapoptosis.
Cell126,121–134.
(2006).
39.
Feng,Z.
etal.
TheregulationofAMPKbeta1,TSC2,andPTENexpressionbyp53:Stress,cellandtissuespecicity,andtheroleofthesegeneproductsinmodulatingtheIGF-1-AKT-mTORpathways.
CancerRes.
67,3043–3053.
(2007).
40.
Budanov,A.
V.
&Karin,M.
p53targetgenesSestrin1andSestrin2connectgenotoxicstressandmTORsignaling.
Cell134,451–460.
(2008).
41.
Molchadsky,A.
etal.
p53playsaroleinmesenchymaldifferentiationprograms,inacellfatedependentmanner.
PLoSONE3,e3707.
(2008).
42.
Zheng,J.
P.
etal.
Resveratrolinducesp53andsuppressesmyocardin-mediatedvascularsmoothmusclecelldifferentiation.
Toxicol.
Lett.
199,115–122.
(2010).
43.
Wang,Z.
,Wang,D.
Z.
,Pipes,G.
C.
&Olson,E.
N.
Myocardinisamasterregulatorofsmoothmusclegeneexpression.
Proc.
NatlAcad.
Sci.
USA100,7129–7134.
(2003).
44.
Tan,W.
Q.
,Wang,K.
,Lv,D.
Y.
&Li,P.
F.
Foxo3ainhibitscardiomyocytehypertrophythroughtransactivatingcatalase.
J.
Biol.
Chem.
283,29730–29739.
(2008).
AcknowledgementsThisworkwassupportedbytheNationalNaturalScienceFoundationofChina(81522005,81470522,31430041,and81230005),andtheTaishanScholarProgramofShandongProvince.
AuthorcontributionsK.
W.
Designedresearch.
C-Y.
L.
,Y-H.
Z.
,R-B.
L.
,L-Y.
Z.
,T.
A.
,R-C.
Z.
,M.
Z.
,Y.
H.
,K-W.
Y.
,Y-H.
D.
andM.
P.
:Performedcellularexperiments.
C.
S.
,S.
X.
,Q.
W.
,Y-H.
Z.
:Conductedanimalexperiments.
K.
W.
andJ.
Z.
:Wrotethemanuscript.
AdditionalinformationSupplementaryInformationaccompaniesthispaperathttps://doi.
org/10.
1038/s41467-017-02280-y.
Competinginterests:Theauthorsdeclarenocompetingnancialinterests.
Reprintsandpermissioninformationisavailableonlineathttp://npg.
nature.
com/reprintsandpermissions/Publisher'snote:SpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpublishedmapsandinstitutionalafliations.
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/.
TheAuthor(s)2017ARTICLENATURECOMMUNICATIONS|DOI:10.
1038/s41467-017-02280-y12NATURECOMMUNICATIONS|(2018)9:29|DOI:10.
1038/s41467-017-02280-y|www.
nature.
com/naturecommunications