Photoactivatable Cre recombinase 3.0 for in vivo mouse ...

Cre recombinase, which is derived from P1 bacteriophage, is the most common recombinase that has been used to catalyze directional DNA ... Skiptomaincontent Thankyouforvisitingnature.com.YouareusingabrowserversionwithlimitedsupportforCSS.Toobtain thebestexperience,werecommendyouuseamoreuptodatebrowser(orturnoffcompatibilitymodein InternetExplorer).Inthemeantime,toensurecontinuedsupport,wearedisplayingthesitewithoutstyles andJavaScript. Advertisement nature naturecommunications articles article PhotoactivatableCrerecombinase3.0forinvivomouseapplications DownloadPDF DownloadPDF Subjects GeneticengineeringGeneticmodels AbstractOptogeneticgenomeengineeringtoolsenablespatiotemporalcontrolofgeneexpressionandprovidenewinsightintobiologicalfunction.Here,wereportthenewversionofgeneticallyencodedphotoactivatable(PA)Crerecombinase,PA-Cre3.0.ToimprovePA-Cretechnology,wecomparelight-dimerizationtoolsandoptimizeformammalianexpressionusingaCAGpromoter,Magnets,and2Aself-cleavingpeptide.Topreventbackgroundrecombinationcausedbythehighsequencesimilarityinthedimerizationdomains,wemodifythecodonsformousegenetargetingandviralproduction.Overall,thesemodificationssignificantlyreducedarkleakactivityandimproveblue-lightinductiondevelopingournewversion,PA-Cre3.0.Asaresource,wehavegeneratedandvalidatedAAV-PA-Cre3.0aswellastwomouselinesthatcanconditionallyexpressPA-Cre3.0.Togetherthesenewtoolswillfacilitatefurtherbiologicalandbiomedicalresearch. IntroductionTheabilitytomanipulateDNArecombinationandgeneexpressioninaspatiotemporalspecificmannerisapowerfultechniqueingenomeengineeringstudies1.Crerecombinase,whichisderivedfromP1bacteriophage,isthemostcommonrecombinasethathasbeenusedtocatalyzedirectionalDNArecombinationbetweenloxPpairs2,3,4,5.WhilevariousinduciblesystemshavebeendevelopedbasedonCre-loxPrecombination,thetoolsoftensufferfromeitherlowefficiency(suchaswiththeCRY2-CIB1-basedsystem)orhavecomplicationssuchasthenecessityforharmfulchemicalinducerssuchastamoxifenorrapamycin6,7,8,9.WhileourpreviouslyreportedMagnets-basedPA-Cresystemimprovedonmanyoftheseshortcomings10,PA-Crestillhadamajorissuewithunintentionalrecombinationindarkconditionspriortolightstimulation.Inaddition,therearecurrentlynoinvivomousemodelsavailableforoptogenetic-basedsystems,limitingthescopeofapplicationsinbiologicalstudy.Inthisstudy,wedevelopedanimprovedversionofPA-CrecalledPA-Cre3.0,whichisbasedonthesameblue-light-dependentdimerizationsystem,Magnets.WedemonstratetheimprovedefficiencyofPA-Cre3.0anditsapplicationsinvivousingnewlygeneratedmouselinesexpressingPA-Cre3.0conditionally.Webelievethisimprovedsystemandmousemodelavailabilitycanenhancegeneticstudiesinlivingsystemstoaddressbiologicalhypothesesandunveilthemolecularandpathophysiologicalmechanismsunderlyingvariousdiseases.ResultsImprovementofPA-CresystemPreviously,wedevelopedthe1stgenerationofMagnets-basedPA-Cre,takingadvantageofblue-light-dependenthetero-dimerizationsystem,Magnets10.AlthoughtheconstructofPA-Crecouldbesuccessfullytransientlyappliedinmammaliancellsinvitroandmouseliversinvivousinghydrodynamictailvein(HTV)injectionandreporterplasmids,otherblue-light-induciblehetero-dimerizationsystemscouldbemoresuitablefordevelopingPA-Cresystems.Toaddressthisquestion,CRY2/CIB1-,iLID/SspB-,andFKF1/GI-basedPA-Creconstructswerepreparedandtestedusingluciferase(Luc)andmCherryreporters,andcomparedtotheoriginalMagnets-basedPA-Cre9,11,12,13(Fig. 1a–candSupplementaryFig. 1a,b).WefoundthattheMagnets-basedoriginalPA-CrehadthehighestCre-loxPrecombinationefficiencywithlightamongtheseconstructs(~75%comparedwithapositivecontrol,CreERT2,treatedwithtamoxifen).WhilethefoldinductionofCre-loxPrecombinationusingtheCRY2/CIB1-basedPA-Creisthebest(43.3×)amongthetests,theefficiencyofCre-loxPrecombinationwithlightwaslow(~15%)comparedwithCreERT2positivecontrol.TheFKF1/GI-basedversionalsodemonstratedaslowCre-loxPrecombinationefficiencywithbluelightastheCRY2/CIB1-basedone(calledPA-Cre2.0)11.Ontheotherhand,theiLID/SspB-basedversionhadmuchhigherleakinessindarkthantheothers.TheseresultssuggestthattheoriginalMagnets-basedPA-CreisstillpromisingforfurtherimprovementastheunintentionalCre-loxPdarkleakrecombinationislimited(SupplementaryFig. 1a).Toassessthisdarkleakinessissuefurther,wemonitoredtheLucactivity24,48,72,and96 hafterHEK293TcellsweretransfectedwiththePA-Creconstructs.TheCRY2/CIB1-basedconstruct,calledPA-Cre2.0(ref.11),wasalsotestedasabenchmarkexperiment.TheMagnets-basedPA-CreshowedanaccumulatingleakovertimewhiletheCRY2/CIB1versiondemonstratedlittletonoleakiness(SupplementaryFig. 1c).SuchleakyrecombinationindarkafterPA-CreexpressionisnotacceptableforanyinvivoapplicationsasCre-loxPrecombinationisirreversible.Toaddressthisissue,welookedtoimprovetheMagnets-basedPA-Cresystembyreducingthebackgrounddarkactivity.Fig.1:Comparisonofmultiplelight-activateddimerizationsystemsinphotoactivatableCrerecombinase.aSchematicrepresentationofphotoactivatable(PA)-Cresystemanditsreporterconstructs.SplitCre(59/60)arecomplementedalongwithnMag–pMagdimerizationuponblue-lightillumination(BLbluelight,NLSnuclearlocalizationsignal,2AP2Aself-cleavingpeptidesequence,PCMVcytomegaloviruspromoter,FlucFireflyluciferase,pApolyadenylationtranscriptional“stop”(poly-A)signalrepeatedsequence).bComparisonofPA-Crewithvariousblue-lightphotoreceptorsusingluciferase(Luc)assay.UpperdiagramshowsexperimentalprotocolusedforLucassay(blueLED,447.5 nm,8.28 W/m2,repeated20 slightand60 sdarkfor12 h).Lucassayswereconductedwithdouble-floxedinvertedFlucreporterinHEK293Tcells.Theherpessimplexvirusthymedinekinase(HSV-TK)promoter-RenillaLucplasmidwasco-transfectedasatransfectioncontroltonormalizeFireflyLucactivity.CreERT2vectorandpcDNA3.1emptyvector(Mock)treatedwith1 µMtamoxifen(Tam)addition36 hafterthetransfectionwereusedaspositiveandnegativecontrols,respectively(*P