ORIGINALARTICLE
HybridCytomegalovirus-U6Promoter-basedPlasmid
VectorsImproveEfficiencyofRNAInterferenceinZebrafish
JianguoSu&ZuoyanZhu&FengXiong&YapingWang
Received:24October2007/Accepted:16January2008/Publishedonline:6March2008#SpringerScience+BusinessMedia,LLC2008
AbstractShorthairpinRNA(shRNA)directedbyRNApolymeraseIII(PolIII)orPolIIpromoterwasshowntobecapableofsilencinggeneexpression,whichshouldpermitanalysesofgenefunctionsorasapotentialtherapeutictool.However,theinhibitoryeffectofshRNAremainsproblem-aticinfish.WedemonstratedthatsilencingefficiencybyshRNAproducedfromthehybridconstructcomposedoftheCMVenhancerorentireCMVpromoterplacedimmediatelyupstreamofaU6promoter.Whentestedtheexogenousgene,silencingofanenhancedgreenfluorescentprotein(EGFP)targetgenewas89.18±5.06%forCMVE-U6promotergroupand88.26±6.46%forCMV-U6promotergroup.TotestthehybridpromotersdrivingshRNAefficiencyagainstanendogenousgene,weusedshRNAagainstnotail(NTL)gene.Whenvectorizedinthezebrafish,thehybridconstructsstronglyrepressedNTLgeneexpression.TheNTLphenotypeoccupied52.09±3.06%and51.56±3.68%forCMVE-U6promoterandCMV-U6promotergroups,respectively.TheNTLgeneexpressionreduced82.17±2.96%forCMVE-U6promotergroupand83.06±2.38%forCMV-U6promotergroup.WeconcludedthattheCMVenhancerorentireCMVpromoterlocatingupstreamoftheU6-promotercouldsignificantlyimproveinhibitoryeffectinducedbytheshRNAforbothexogenousandendogenousgenescomparedwiththeCMV
J.Su:Z.Zhu(*):F.Xiong:Y.Wang
StateKeyLaboratoryofFreshwaterEcologyandBiotechnology,InstituteofHydrobiology,ChineseAcademyofSciences,Wuhan430072,Chinae-mail:zyzhu@ihb.ac.cn
J.Su
DepartmentofAquaculture,
CollegeofAnimalScienceandTechnology,NorthwestA&FUniversity,Yangling712100,China
promoterorU6promoteralone.Incontrast,thetwohybridpromoterconstructshadsimilareffectsondrivingshRNA.KeywordsHybridpromoter.CMVpromoter.U6promoter.RNAi.Zebrafish
Introduction
Inpastyears,themostexcitingdevelopmentingeneregulationwasthediscoveryofRNAinterference(RNAi),throughwhichshortinterferingRNAs(siRNAs)mediateselectivegeneinactivationbymRNAdestruction(Songetal.2004).Thishighlyefficientandspecifictechnologyopensupabroadspectrumofpotentialimplicationsforboththerapeuticsandexperimentalresearch(Dornetal.2004).Firstdescribedinplants(Napolietal.1990),ithassincebeenappliedtoawidevarietyofinvertebrateandvertebratemodels(HutvagnerandZamore2002).Inmammaliancells,theintroductionoflongdsRNAsintomammaliancellsactivatesproteinkinasePKRandRNaseL,leadingtoaninterferonresponseandhencetothenonspecificextinctionofgenesresultingincelldeath.ThisnonspecificeffectoflongdsRNAsintomammaliancellscanbebypassedbyusingsmallRNAduplexesof19–21nt,whicharesufficienttotriggerspecificRNAiinmammaliancellswithoutactivatingtheinterferonresponse(Elbashiretal.2001).
ShorthairpinRNAs(shRNAs)transcribedinvivounderthecontrolofRNApolymeraseIII(PolIII)promoterscantriggerdegradationofcorrespondingmRNAssimilartosiRNAsandinhibitspecificgeneexpression(Brummelkampetal.2002;Jacqueetal.2002).InadditiontosynthesizingmRNA,RNApolymeraseII(PolII)isresponsibleforsynthesisofmanynoncodingRNAs,includingsmall
512nuclearandnucleolarRNAs(Songetal.2004).Recently,siRNAtranscriptsexpressedfromaRNApolymeraseIIpromoter,cytomegalovirus(CMV)promoter,havebeenshowntobecapableofreducinggeneexpressioninmammaliancellsandzebrafish(Xiaetal.2002;Suetal.2007).
WedesignedshRNA-expressingconstructcontrolledbyaPolIIIU6promoter(Wang2007)andshRNA-expressingconstructcontrolledbyCMVpromoter(Suetal.2007)toinhibitanexogenoustransgenicEGFPandanendogenousnotail(NTL)expressionsinzebrafish.Whiletestingtheefficacyoftheseconstructs,wefoundtheyselectivelyinhibitedtheexpressionoftargetgenebutdidnotaffectothers.However,theefficacyofRNAiproducedbytheseconstructswasmodest,whichmightaffecttheultimateapplication.OnewaytoovercomethisproblemwastoincreasethedoseoftheshRNAbyenhancingthepromoteractivity.SomesnRNAsaresynthesizedbyPolII,whereasothersaresynthesizedbyPolIII,andtheysharesimilarenhancerelements(Barketal.1987;Carbonetal.1987;Dasetal.1988;KunkelandPederson1988;Mattajetal.1988;LoboandHernandez1989).Hence,aPolIIenhancermightbeabletoenhancethePolIIItranscription.Onepreviousstudydemonstratedthattheenhancerfromthecytomegalovirusimmediate-earlypromoter,whenplacednearthePolIIIU6promoterinaplasmidvector,couldenhancetheactivityoftheU6promoter,increaseshRNAexpression,andstrengthenthegene-silencingeffectinhumancells(Xiaetal.2003).
TheefficiencyofshRNA-carryingplasmidsintermsoftargetgeneinhibitionvarieswiththecompositionofpromoterelements,thepromoterpositionrelativetothetranscriptionalstartsite,thelocationofapromoterwithinagivenvector,andprobablythetypeoftissuesorcellstested(Ilvesetal.1996;Arendtetal.2003;Bodenetal.2003;Koper-Emdeetal.2004).Eachtissueandcondition(invivoversusinvitro)requiresoptimizationfortheuseofshRNAs-triggeredRNAi,too(Hassanietal.2007).There-fore,thechoiceofpromoterisvitalinachievingeffectivegenesilencingbyintracellularexpressionofshRNAinvariousorganisms.
Inthepresentstudy,weconstructedahybridpromoterbyplacinga465-bpfragmentoftheCMVenhancerandahybridpromoterbyappendingtheCMVenhancerandearlypromoter5′totheU6promoterandexploredwhetherthesemodifiedpromoterscouldimprovegenesilencingeffectsinzebrafish.
MaterialsandMethods
PlasmidConstructionspCMV-EGFPplasmid(referredinthefollowingaspG),carryingtheCMVpromoterand
MarBiotechnol(2008)10:511–517
enhancedgreenfluorescentprotein(EGFP)geneasreportermark,wasfromClontech(USA)asbackbone.Toensurethecorrectsequenceoftheinserts,theclonesselectedweresequencedafterinsertingeachfollowingfragment.
Toconstructthehybridpromoters,includinghybridCMVenhancerU6(CMVE-U6)promoterandhybridCMVenhancerandearlypromoterU6(CMV-U6)promoter,aCMVenhancerelement(1-465ntofhumancytomegalovirusimmediateearlypromoter)andaCMVpromoter(1-589ntofhumancytomegalovirusimmediateearlypromoter)werePCR-amplifiedfrompGwithforwardandreverseprimers.Forcloningpurposes,theforwardsequenceSCF128(Table1)wasflankedwithXhoIsiteatthe5′end,andthereversesequenceSCR94c(forCMVenhancer)orSCR94d(forCMVpromoter)(Table1)wasflankedwithHindIII,SalI,andKpnIsitesatthe5′endinorder.PCR-amplifiedproductswerepurifiedanddigestedwithXholIandHindIIIsites,andthensubclonedintopGattheXholIandHindIIIsitestoformpCMVE-CMV-EGFPorpCMV-CMV-EGFP.
TheZebrafishU6promoterwasobtainedfromzebrafishgenomicDNAwithsenseprimerSUF130carryingKpnIsiteandantisenseprimerSUR131carryingSalIsite(Table1).Theampliconwaspurified,digested,andinsertedinthepCMVE-CMV-EGFPandpCMV-CMV-EGFPat3′regionoftheinsertedCMVenhancerorCMVpromoterbetweentheKpnIandSalIsites.ThevectorswerenamedpCMVE-U6-CMV-EGFPandpCMV-U6-CMV-EGFP,respectively.
TwocomplementaryoligonucleotidesSG134aandSG135a(Table1)directedagainsttheEGFP-codingregionwithcohesiveSalIandHindIIIsiteswerechemicallysynthesizedaccordingtothesequencespreviouslydescribed(Suetal.2007),denatured,annealed,andclonedintopCMVE-U6-CMV-EGFPandpCMV-U6-CMV-EGFP.ThetwoplasmidvectorswerenamedpCMVE-U6-siEGFP-CMV-EGFP(pEUsiG)andpCMV-U6-siEGFP-CMV-EGFP(pCUsiG),respectively.pCMVE-U6-siScrambled-CMV-EGFP(pEUsiS)andpCMV-U6-siScrambled-CMV-EGFP(pCUsiS)withascrambledfragment(SG134bandSG135b)(Table1)ofsiEGFPwereusedasthecontrol.TosilencetheendogenousNTLgeneexpression,twocohesivecomplementaryoligonucleotidesSN120bandSN121b(Table1)weresynthesizedandclonedintoplasmidvectorsasdescribedabove.ThesevectorswerenamedpCMVE-U6-siNTL-CMV-EGFP(pEUsiN)andpCMV-U6-siNTL-CMV-EGFP(pCUsiN),respectively(seescheme,Fig.1).AlltheoligonucleotidesforshRNAtranscriptionwerefollowedby5thymidines(Ts)astranscriptionalterminationsignal.GeneTransferinvivoTheprotocolofgenetransferwasasdescribedpreviously(Suetal.2007).Briefly,Individual1-to2-cellembryosweremicroinjectedunderadissecting
MarBiotechnol(2008)10:511–517Table1OligonucleotidesusedinthisstudyNameSCF128SCR94cSCR94dSUF130SUR131SG134aSG135aSG134bSG135bSN120bSN121bSBAF86SBAR87SGF114SGR115SNF122SNR123
Sequence(5′–3′)
AATCGCTCGAGTAGTTATTAATAGTAATCAATTACG
AATCGAAGCTTGTCGACGGTACCCAAAACAAACTCCCATTAATCGAAGCTTGTCGACGGTACCGATCTGACGGTTCACTAAATCGGGTACCTCTTTAGCCTCCGAGAGAATCGGTCGACGAACTAGGAGCCTGGAG
TCGACGCAAGCTGACCCTGAAGTTCTTCAAGAGAGAACTTCAGGGTCAGCTTGCTTTTTAAGCTTAAAAAGCAAGCTGACCCTGAAGTTCTCTCTTGAAGAACTTCAGGGTCAGCTTGCGTCGACAGCATCTAAGGCGACCTCGTTTCAAGAGAGGTCAGTAGGTCAACCTCCATTTTTAAGCTTAAAAAACCTCCAACTGGATGACTGGAGAGAACTTTGCTCCAGCGGAATCTACGAGTCGACTGCAATGTACTCGGTCCTGTTCAAGAGACAGGACCGAGTACATTGCATTTTTAAGCTTAAAAAATGTACTCGGTCCTGTCTCTTGAACAGGACCGAGTACATTGCAGGATGATGAAATTGCCGCACTGACCAACCATGACACCCTGATGTCAAGCAGAAGAACGGCATCAAGGTAGTGGTTGTCGGGCACAGCACTGACAACCAGCAATCGAACCCGAGGAGTGAACAGG
513
SequenceinformationCMVpromoter/enhancerprimerCMVenhancerprimerCMVpromoterprimerU6promoterprimerU6promoterprimerShGFPShGFP
ShScrambledShScrambledShNTLShNTL
β-actinQ-PCRprimerβ-actinQ-PCRprimerGFPQ-PCRprimerGFPQ-PCRprimerNTLQ-PCRprimerNTLQ-PCRprimer
microscopewithpulledmicrocapillarypipettestodeliverapproximately1nlofshRNAexpressionvectorsolution(100ng/μl).Weinjectedatleast450embryosandculturedinthreedishesforeverygroup.Totallyweusedapproxi-mately4,000embryos.Theinjectedanduninjectedembry-osweresubsequentlyincubatedinsterile0.3×Danieau’ssolutionat28°C.Werefreshedthesolutionanddisposedofdeadembryostwiceperday.
PhenotypicObservationsPhenotypeswereevaluatedat48hpf(hourspostfertilization).ImagesofzebrafishembryoswererecordedusinganOlympusSZX12fluores-centmicroscopeandadigitalcamera(Liuetal.2007).FeaturesanalyzedforNTL:disruptednotochord,abnormalsomitesandreducedtail(Halpernetal.1993).EGFPexpressionwasmonitoredusingfluorescemicroscopy.Forphenotypecountandfluorescentobservation,weemployedalltheembryos.
Q-PCRAssaysTargetmRNAlevelsinembryoswerequantitativelyanalyzedusingreal-timefluorescentquanti-tativereversetranscriptionpolymerasechainreaction(Q-PCR).Zebrafishembryosinjectedanduninjectedwerecollectedat48and120hpf,andtotalRNAwasextractedfrompoolsof30–40embryospersamplewithtriplicaterepeatusingTrizol(Invitrogen).TheRNAsampleswerefurthertreatedwithRNase-freeDNaseI(Roche)toremovecontaminatedgenomicDNA,followedbyphenol/chloro-formextractionandethanolprecipitation.OnemicrogramoftotalRNAfromeachsamplewasusedforthefirst-strandcDNA,usingaSuperscriptIIIfirst-strandsynthesissystem(Invitrogen)accordingtothemanufacturer’sprotocol.FortheQ-PCRdetectionofEGFPorNTLmRNA,cDNAsampleswerediluted1:5.
Q-PCRswereperformedinanABIPrism7000SequenceDetectionSystem(AppliedBiosystems)usingthefollowingthermalcyclingprofile:95°C3min,followedby40cyclesofamplification(95°C15s,58°C15s,72°C45s),followedbydissociationcurveanalysistovalidatetheamplificationofasingleproduct.Eachreactioncontained4μlofcDNAsample,10μlofSYBRGreen
aXhoI1KpnI465SalI879HindIII937/939CMV enhancerU6 promotershRNAT5bXhoI1KpnI589SalI1003HindIII1161/1163CMV promoterU6 promotershRNAT5Fig.1SchemeofthedifferentpromoterconstructionsdrivingshRNA.Thetwopromoterconstructstestedfordriving-shEGFP,-shScrambledor-shNTLmediatedinhibitioninzebrafishwereshown.aHybridCMVenhancerandU6promoter.bHybridentireCMVpromoterupstreamofU6promoter.Ineachconstruction,theshRNAwasfollowedbyaTTTTTsequencerequiredfortheU6-transcriptionarrest
514PCRmastermix(Toyobo),1μleachofforwardandreverseprimers(2.5μM),aswellas4μlofnuclease-freewatertomakeupa20-μlreactionvolume.Allreactionswereperformedintriplicate.AllprimersweredesignedtoconformtoauniversalcyclingprogrambythePrimerExpress2.0software(AppliedBiosystems).Theendoge-noushousekeepinggeneencodingβ-actinwasusedasaninternalstandardtonormalizedifferencesintemplateamounts.TheoligonucleotidesusedforPCRamplificationofβ-actinwereasfollows:forwardprimerSBAF86andreverseprimerSBAR87(Table1).ThePCRproductsizewas135bp.FortheQ-PCRdetectionofEGFPmRNA,theprimersusedwereSGF114andSGR115(Table1).ThePCRproductoftheEGFPgenewas138bp.NTLprimersequencesweresenseSNF122andantisenseSNR123(Table1).TheNTLampliconwas126bp.
Relativeexpressionwascalculatedusingamodifiedcomparativecyclethreshold(CT)method,inwhichCTwasdefinedasthecyclenumbersatwhichfluorescencereachedasetthresholdvalue.ThedifferencesintheCTvalueofthetargetgenesfromthecorrespondinginternalcontrolβ-actingene,ΔCT(CTgene-CTactin),werecalculated.ThechangesinΔCToftheexperimentgrouptothecontrolgroup,ΔCT(ΔCTsample-ΔCTcontrol),werecomputed.Therelativeexpressionleveloftheexperimentgrouptothecontrolgroupwasdescribedbyusingtheequation2ÀΔΔCT,andthevaluewasdeterminedfora1/n-folddifferencerelativetothecontrol.Theexpressionofexperimentgrouprelativetothecontrolgroupwasmultipliedby100%tosimplifythepresentationofthedata.
StatisticalAnalysisThedataobtainedfromQ-PCRanalysiswereshownbymeans±SE.StatisticalanalysiswasdoneusingunpairedStudent’st-testforcomparisonbetweenthecontrolandtreatedgroups.Errorbarsindicatedstandarderrors.P<0.05wasconsideredstatisticallysignificant,andP<0.01wasverysignificantdifference.
Results
DesignofDifferentPromoterConstructsDrivingtheTranscriptionofshRNATwodifferenthybridpromoterconstructswereusedfortheproductionofshRNAs,includingshEGFP,shNTLandshSrambledcontrol(Fig.1).ThefirstwasaCMVE-U6hybridpromotertranscribingshRNA.Inthisconstruction,thetypeIIIRNApolymerasesynthesizedtheshRNA,andthetran-scriptionwasenhancedbytheCMVenhancer.Second,toexplorewhethertheCMVenhancerandearlypromotercouldimprovetheefficacyoftheU6-mediatedproductionofshRNAsinfish,weconstructedaCMV-U6hybrid
MarBiotechnol(2008)10:511–517
constructdrivingtheproductionoftheshRNA.Intheconstructions,thefullenhancerandpromotersequencesofCMVhavebeenintegratedupstreamoftheU6promoter.ThetranscriptionofshRNAwasterminatedby5Tsoftranscriptionalterminationsequenceatthe3′endoftheshRNA.
pEUsiGandpCUsiGTriggertheExogenousEGFPGeneKnockdownWetestedandcomparedtheefficiencyforpEUsiGandpCUsiGtotriggertheinhibitionofatandemtargetgenebyRNAiinzebrafish.AsdescribedinMethods,twoEGFPandtwocontrolshRNAexpressionvectorsdescribedinFig.1,andpGcontrolplasmid,weremicro-injectedinzebrafishembryos.Underafluorescencemicroscope,thefluorescencewasveryweakintheexperimentalgroups(Fig.2f).AsshowninFig.3,quantificationofEGFPexpressionsrevealedstronginhibi-tionefficiencyforbothvectorsat48hpf.The89.18±5.06%inhibitionofthetargetgenewasachievedwithpEUsiG,andthe88.26±6.46%reductionoftheEGFPwasobtainedwithpCUsiG,whichweresignificantlystrongerthanthecontrols(P<0.01).TherewasnosignificantdifferenceingenesilencebetweenpEUsiGandpCUsiG(P>0.05).ThecontrolpEUsiSandpCUsiSdidnotprovideinhibitionofEGFP.
pEUsiNandpCUsiNInhibittheEndogenousNTLGeneExpressionAfterhavingdemonstratedthatthepresenceofaCMVenhancer/promoterupstreamofaU6promoterstronglyincreasedtheinhibitionefficiencyofshRNAs,wefurtherexploitedthistechnologytointerfereintheendogenousNTLgeneexpressioninfish.WedesignedshNTLexpressionvectorsdirectedagainstNTLandcheckedtheirefficiencyandtheirspecificityinzebrafish.Theembryosofnotailphenotypewerecountedunderadissectionmicroscope,andoccupied52.09±3.06%at48hpfinpEUsiNgroup,51.56±3.68%inpCUsiNgroup,and1.12±0.15%inothergroups(P<0.05)(Figs.2cand4a).UsingQ-PCR,weobservedasignificantdown-regulationofNTLunderthecontrolofthehybridpromoterconstructs.Theinhibitoryefficiencywas82.17±2.96%forpEUsiNand83.06±2.38%forpCUsiNcomparedwiththeuninjectedgroup(P<0.01).ThecontrolpEUsiSandpCUsiSalsoweretestedinthesamesituation,wherenoinhibitionwasvisible(Fig.4b).
Discussion
SomescientiststriedtoexploitRNAitechnologyinfishsinceitwasdiscovered;however,untilnowtherehavestillsomeobstaclestobeovercome,suchasnonspecificityand
MarBiotechnol(2008)10:511–517515
Fig.2TheeffectofhybridCMV-U6promoterconstructsinjectionontheexpressionofEGFPandNTLinzebrafishembryosat48hpf.Upperrow(a–d)showedthesamefieldofembryosbylightmicroscopyasseenunderfluorescenceinlowerrow(e–h).Column1wasinjectedwithEGFPexpressionvector(pG).Column2wasinjectedwithshGFPconstruct(pCUsiG).Column3wasinjectedwith
shNTLconstruct(pCUsiN).Column4wasinjectedwithshScrambledconstruct(pCUsiS).ThehybridCMVE-U6promoterconstructshadthesimilarresults.shScrambledconstructandshNTLconstructdidnotinterferewithEGFPexpression.TheembryosinjectedwithshGFPorshScrambledconstructhardlyshowedthenotailphenotype
a
The percentage of no tail phenotype (%)The relative expression of NTL mRNA (%)lowefficacy.SomefindingssuggesteddsRNA-mediatedinterferencecouldinhibitthegeneexpressionindose-dependentmode(Wargeliusetal.1999;Acostaetal.2005),othersreportednonspecificeffect(Oatesetal.2000;Lietal.2000;Zhaoetal.2001;Mangosetal.2001).Theseviewsmightberesultedfromdifferentgenesequencesstudied.siRNA-basedgenesilencingwaseffective,butsiRNAduplexeswereunsteadyinvivo(Boonanuntanasarnetal.2003;Doddetal.2004;Liuetal.2005).
ThemajorityofthemammalianRNAisystemsaredrivenbyPolIIIpromotersthatexpressubiquitously(Songetal.2004).However,undercertainconditions,theactivityprovidedbyanunmodifiedPolIIIpromoterisnotsufficientinprovidingsatisfactorygene-silencingoutcomes
The relative expression of EGFP (%)120100806040200
pG
pEUsiG
pCUsiG
pEUsiS
pCUsiS
6050403020100
Blank
pEUsiN
pCUsiN
pEUsiS
pCUsiS
b
120100806040200
Blank
pEUsiN
pCUsiN
pEUsiS
pCUsiS
Fig.3pEUsiGandpCUsiGshowedsimilarinhibitoryefficiencyforexogenousgeneinzebrafish.PlasmidscontainingshEGFPorshScrambledweremicroinjectedintozebrafishasdescribedinMethods,alongwiththereportergeneEGFP.OnlypEUsiGandpCUsiGleadtoaninhibition.Datashowedmeanvalues±SE(n=3).Errorbarsindicatestandarderror
Fig.4pEUsiNandpCUsiNdemonstratedanalogoussilencingactivityforendogenousgeneinzebrafish.apEUsiNandpCUsiNweremicroinjectedasdescribedintheMethodssection,alongwiththeirrelevantpEUsiSandpCUsiSascontrols.ThenotailphenotypewasveryhighinpEUsiNandpCUsiNgroups.bZebrafishwasmicroinjectedbythedifferentplasmidconstructionsdescribedin(a),onlypEUsiNandpCUsiNprovidedsignificantinhibitionoftheendogenousNTLexpression.Means±SEareshown.Errorbarsindicatedstandarderror
516(Pauletal.2002;Bodenetal.2003;Xiaetal.2003).Infish,theshRNAunderthecontrolofPolIIIpromoters(H1andU6promoters)(Wang2007)alsoinhibitedtheexogenousorendogenousgenemodestly.Toimprovepotency,wedesignedtwohybridpromotersbyfusingtheCMVenhancerorentireCMVpromoter5′totheU6promoterandobservedimprovedinhibitionbysiRNAoftargetgeneexpression.TheU6promoterbelongstothePolIIIpromoterandislocatedonlyinthe5′-flankingregionofthegenetranscribed.Thisgene-externalfeatureresemblesthatofPolIIpromoters.WhereasthePolIIIU6promoterdirectsthesynthesisoftheU6smallnuclearRNA(snRNA),PolIIpromotersareusedfortranscriptionoftheU1toU5snRNAs.Strikingly,notonlyisthespatialorganizationoftheU6promoterelementssimilartothatoftheU2elements,theyalsosharesequencehomologiesintheenhancer(Hernandez2001).ThesefindingsfromthesnRNApromotershaveledtotheconceptthatthePolIIandIIItranscriptionmachineriessharecommonfactors.Inthecurrentstudy,wemodifiedtheU6promoterbyplacingtheenhancerorfullpromoterfromtheCMVimmediate-earlypromoterimmediatelyupstreamofaU6promoter.Thehybridpromotersdrivingthesynthesisofahairpinsequenceprovidedasignificantimprovementintermsofinhibitoryeffectforbothexogenousandendog-enousgenes.The89.18±5.06%and88.26±6.46%reduc-tionoftheexogenousEGFPgenewithhybridpromoterswerestrongerthanthe70%and32%knockdownobtainedwithpCMV-shEGFP(Suetal.2007)andpU6-shEGFP(Wang2007),respectively.shNTLproducedbythemodifiedU6promoterinhibitedendogenousNTLgeneexpressionupto82.17±2.96%and83.06±2.38%,whichwasmuchhigherthanthe54.52±2.05%suppressionwithpCMV-shEGFP(Suetal.2007)andapproximately16%reductionwithU6promoteralone(Wang2007).Boththehybridpromoterconfigurationsdirectedsynthesisofscram-bledshRNAdidnotshowanyinhibitoryactivitytowardsthetargetgenes.ThefindingsfromthecurrentstudysupportedthenotionthataCMVenhancer,eventhoroughCMVpromoter,wasabletoenhancethetranscriptionalactivityofaU6promoterinfish.
AttachingtheCMVenhancerorentirepromoterattheupstreamofU6promoterappearedhighlyeffectivegene-specificknockdown,andthereductionactivitiesoftargetgeneexpressionsweresimilarinzebrafish(Figs.3and4).Theunderlyingmechanismisasyetunknown.Itiswellrecognizedthatenhancerandpromoterrecognitionbytranscriptionfactors,repressors,andauxiliaryproteinsisacomplexprocess,involvingbothprimaryandsecondarysequencecharacteristicsofageneexpressionregulatoryelement.Thenumber,diversity,orientation,andplacementoftranscriptionfactor-bindingsiteswithinanenhancerand/
MarBiotechnol(2008)10:511–517
orapromoterarecriticalparametersthatdefinegeneexpressionlevels.
Conclusions
WehaveconstructedtwohybridpromoterscarryingtheU6promoterfusedtotheCMVenhancerorcompleteCMVpromoteranddemonstratedtheimprovedsilencingeffectsinfish.Inadditiontofunctioningasapowerfulinstrumentforbiologicalstudies,thismodifiedpromotermightbeusefulwherelimitedchoicesofshRNAsequencesprecludetheselectionofahighlyefficientRNAitargetregionandmightalsoserveasapotentiallyusefultherapeuticmeanstosilencedisease-relatedgenesandcorrectdiseasephenotypes.
AcknowledgementsTheauthorsthankMingLiformicroinjection.WealsoappreciateWeiHu,JunDai,NaWang,ShangpingChen,andotherlaboratorymembersfortechnicalassistanceandhelpfuldiscussion.ThisworkwassupportedbygrantsfromNationalNaturalScienceFoundationofChina(30740009,30540084and30428024),from973NationalBasicResearchProgramofChina(2006CB102100),fromChineseAcademyofSciences(KSCX2-YW-N-021),fromNorthwestA&FUniversityinChina(08080262,08080245and01140508),fromChinaPostdoctoralScienceFounda-tion(20070410298)andfromInstituteofHydrobiology,CAS(2007FB09).
References
AcostaJ,CarpioY,BorrotoI,GonzalezO,EstradaMP(2005)
MyostatingenesilencedbyRNAishowazebrafishgiantphenotype.JBiotechnol119:324–331
ArendtCW,TangG,ZilbersteinA(2003)Vectorsystemsforthe
deliveryofsmallinterferingRNAs:managingtheRISC.Chembiochemistry4:1129–1136
BarkC,WellerP,ZabielskiJ,JansonL,PetterssonU(1987)Adistant
enhancerelementisrequiredforpolymeraseIIItranscriptionofaU6RNAgene.Nature328:356–359
BodenD,PuschO,LeeF,TuckerL,ShankPR,RamratnamB(2003)
PromoterchoiceaffectsthepotencyofHIV-1specificRNAinterference.NucleicAcidsRes31:5033–5038
BoonanuntanasarnS,YoshizakiG,TakeuchiT(2003)Specificgene
silencingusingsmallinterferingRNAsinfishembryos.BiochemBiophysResCommun310:1089–1095
BrummelkampTR,BernardsR,AgamiR(2002)Asystemforstable
expressionofshortinterferingRNAsinmammaliancells.Science296:550–553
CarbonP,MurgoS,EbelJP,KrolA,TebbG,MattajLW(1987)A
commonoctamermotifbindingproteinisinvolvedinthetranscriptionofU6snRNAbyRNApolymeraseIIIandU2snRNAbyRNApolymeraseII.Cell51:71–79
DasG,HenningD,WrightD,ReddyR(1988)Upstreamregulatory
elementsarenecessaryandsufficientfortranscriptionofaU6RNAgenebyRNApolymeraseIII.EMBOJ7:503–512
DoddA,ChambersSP,LoveDR(2004)ShortinterferingRNA-mediatedgenetargetinginthezebrafish.FEBSLett561:89–93
MarBiotechnol(2008)10:511–517
DornG,PatelS,WotherspoonG,Hemmings-MieszczakM,BarclayJ,
NattFJ,MartinP,BevanS,FoxA(2004)siRNArelieveschronicneuropathicpain.NucleicAcidsRes32:e49
ElbashirSM,HarborthJ,LendeckelW,YalcinA,WeberK,TuschlT
(2001)Duplexesof21-nucleotideRNAsmediateRNAinter-ferenceinculturedmammaliancells.Nature411:494–498
HalpernME,HoRK,WalkerC,KimmelCB(1993)Inductionof
musclepioneersandfloorplateisdistinguishedbythezebrafishnotailmutation.Cell75:99–111
HassaniZ,FrancoisJ,AlfamaG,DuboisGM,ParisM,Giovannangeli
C,DemeneixBA(2007)AhybridCMV-H1constructimprovesefficiencyofPEI-deliveredshRNAinthemousebrain.NucleicAcidsRes35:e65
HernandezN(2001)SmallnuclearRNAgenes:amodelsystemto
studyfundamentalmechanismsoftranscription.JBiolChem276:26733–26736
HutvagnerG,ZamorePD(2002)RNAi:natureabhorsadouble-strand.CurrOpinGenetDev12:225–232
IlvesH,BarskeC,JunkerU,BohnleinE,VeresG(1996)
RetroviralvectorsdesignedfortargetedexpressionofRNApolymeraseIII-driventranscripts:acomparativestudy.Gene171:203–208
JacqueJM,TriquesK,StevensonM(2002)ModulationofHIV-1
replicationbyRNAinterference.Nature418:435–438
Koper-EmdeD,HerrmannL,SandrockB,BeNeckeBJ(2004)
RNAinterferencebysmallhairpinRNAssynthesizedundercontrolofthehuman7SKRNApromoter.BiolChem385:791–794
KunkelGR,PedersonT(1988)Upstreamelementsrequiredfor
efficienttranscriptionofahumanU6RNAgeneresemblethoseofU1andU2geneseventhoughadifferentpolymeraseisused.GenesDev2:196–204
LiYX,FarrellMJ,LiuR,MohantyN,KirbyML(2000)Double-strandedRNAinjectionproducesnullphenotypesinzebrafish.DevBiol217:394–405
LiuWY,WangY,SunYH,WangY,WangYP,ChenSP,ZhuZY
(2005)EfficientRNAinterferenceinzebrafishembryosusingsiRNAsynthesizedwithSP6RNApolymerase.DevGrowthDiffer47:323–331
LiuWY,WangY,QinY,WangYP,ZhuZY(2007)Site-directedgene
integrationintransgeniczebrafishmediatedbycrerecombinase
517
usingacombinationofmutantloxsites.MarBiotechnol9:420–428
LoboSM,HernandezN(1989)A7bpmutationconvertsahuman
RNApolymeraseIIsnRNApromoterintoanRNApolymeraseIIIpromoter.Cell58:55–67
MangosS,VanderbeldB,KrawetzR,SudolK,KellyGM(2001)Ran
bindingproteinRanBP1inzebrafishembryonicdevelopment.MolReprodDev59:235–248
MattajIW,DathanNA,ParryHD,CarbonP,KrolA(1988)Changing
theRNApolymerasespecificityofUsnRNAgenepromoters.Cell55:435–442
NapoliC,LemieuxC,JorgensenR(1990)Introductionofachimeric
chalconesynthasegeneintopetuniaresultsinreversibleco-suppressionofhomologousgenesintrans.PlantCell2:279–289OatesAC,BruceAE,HoRK(2000)Toomuchinterference:injection
ofdouble-strandedRNAhasnonspecificeffectsinthezebrafishembryo.DevBiol224:20–28
PaulCP,GoodPD,WinerI,EngelkeDR(2002)Effectiveexpression
ofsmallinterferingRNAinhumancells.NatBiotechnol20:505–508
SongJ,PangS,LuaY,ChiuR(2004)Poly(U)andpolyadenylation
terminationsignalsareinterchangeableforterminatingtheexpressionofshRNAfromapolIIpromoter.BiochemBiophysResCommun323:573–578
SuJ,ZhuZ,WangY,XiongF,ZouJ(2007)Thecytomegalovirus
promoter-drivenshorthairpinRNAconstructsmediateeffectiveRNAinterferenceinzebrafishinvivo.MarBiotechnolDOI10.1007/s10126-007-9059-4
WangN(2007)StudyofshortinterferingRNA(siRNA)producedin
transcriptionsystemofzebrafishembryo.Dissertationofdoctorofscience,ChineseAcademyofSciences
WargeliusA,EllingsenS,FjoseA(1999)Double-strandedRNA
inducesspecificdevelopmentaldefectsinzebrafishembryos.BiochemBiophysResCommun263:156–161
XiaH,MaoQ,PaulsonHL,DavidsonBL(2002)siRNA-mediated
genesilencinginvitroandinvivo.NatBiotechnol20:1006–1010XiaXG,ZhouH,DingH,AffarEB,ShiY,XuZ(2003)Anenhanced
U6promoterforsynthesisofshorthairpinRNA.NucleicAcidsRes31:e100
ZhaoZ,CaoY,LiM,MengA(2001)Double-strandedRNAinjection
producesnonspecificdefectsinzebrafish.DevBiol229:215–223
因篇幅问题不能全部显示，请点此查看更多更全内容