RESEARCHARTICLEAnewglucocerebrosidasedeficientneuronalcellmodelprovid - PDF document

RESEARCHARTICLEAnewglucocerebrosidase-deficientneuronalcellmodelprovidesatooltoprobepathophysiologyandtherapeuticsforGaucherWendyWestbroek,MatthewNguyen,MarinaSiebert1,2,TaylorLindstrom,RobertA.Burnett,ElmaAflakiOliveJung,RafaelTamargo,JorgeL.Rodriguez-Gil,WalterAcosta,AnHendrix,BahaftaBehreNahidTayebi,HidejiFujiwara,RohiniSidhu,BenoitRenvoise,EdwardI.Ginns,AmaliaDutra,EvgeniaPakCaroleCramer,DanielS.Ory,WilliamJ.PavanandEllenSidranskyABSTRACTGlucocerebrosidaseisalysosomalhydrolaseinvolvedinthebreakdownofglucosylceramide.Gaucherdisease,arecessivelysosomalstoragedisorder,iscausedbymutationsinthegene.Dysfunctionalglucocerebrosidaseleadstoaccumulationofglucosylceramideandglycosylsphingosineinvariouscelltypesandorgans.MutationsinarealsoacommongeneticriskfactorforParkinsondiseaseandrelatedsynucleinopathies.Inrecentyears,researchonthepathophysiologyofGaucherdisease,themolecularlinkbetweenGaucherandParkinsondisease,andnoveltherapeutics,haveacceleratedtheneedforrelevantc

ellmodels Received4January2016;Accepted12May2016 SectiononMolecularNeurogenetics,MedicalGeneticsBranch,NationalHumanGenomeResearchInstitute,NationalInstitutesofHealth,Bethesda,MD20892, E.S.,0000-0002-3019-8500ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense(http://creativecommons.org/licenses/by/3.0),whichpermitsunrestricteduse,distributionandreproductioninanymediumprovidedthattheoriginalworkisproperlyattributed. ©2016.PublishedbyTheCompanyofBiologistsLtdDiseaseModels&Mechanisms(2016)9,769-778doi:10.1242/dmm.024588 DiseaseModels&Mechanisms 2011),orover-expressionofplasmidscontainingmutantGBA1(Cullenetal.,2011).Althoughthesemodelshaveprovenuseful,exogenousmanipulationofGCaseorexpressionoftencreatesunwantedoff-targeteffects.PrimaryneuronalculturesfromonemousemodelwereusedtoprobemitochondrialfunctioninGD(OsellameandDuchen,2013;Osellameetal.,2013).Recently,thedevelopmentofinducedpluripotentstemcell(iPSC)linesfromGDpatientsandca

rriershasgainedpopularity,providingtheopportunitytodevelopcellculturesofpreviouslyinaccessiblediseasedhumanneurons(Tiscorniaetal.,2013;Woodardetal.,2014;Sunetal.,2015;Schondorfetal.,2014;Awadetal.,2015).ThemaindisadvantagesofbothprimaryrodentneuronalculturesandiPSC-generatedneuronsarelowcellcultureyieldandthelabor-intensivenessofestablishmentandmaintenance.WehypothesizedthatimmortalizedGDneuronsderivedfromaGDmousemodelcouldprovideahigh-yield,easy-to-maintainalternativeforinvestigationsofthecellularmechanismsinvolvedinGD.Suchimmortalizedneuronscouldalsohaveutilityfortheevaluationofnoveltherapeuticsandthevalidationofdifferentreagentsandantibodies.ImmortalizationofprimarycellsisaccomplishedbyexogenousintroductionofimmortalizinggenessuchastheSV40largeTantigen(SV40-T),whichincreaseslifespanandinducesunlimitedproliferationbyinactivationofthecell-cyclesuppressorspRb,SEN6andp53(Ozeretal.,1996;Tevethiaetal.,1998;ManfrediandPrives,1994;Ozer,2000;Jhaetal.,1998).Neuronsaret

erminallydifferentiatedpost-mitoticcells,whichmakesgenedeliveryviatraditionaltransfectionmethodsdifficult.Lentiviralexpressionvectorshavetheabilitytotransduceproliferatingandnon-proliferatingcells,andhavebeenusedforinfectionofprimaryrodentneuronalcultures(Lewisetal.,1992;Weinbergetal.,1991;Zhangetal.,2006;DingandKilpatrick,2013;Eleftheriadouetal.,2014;Lietal.,2012).Inthisstudy,wereportthesuccessfulSV40-T-mediatedimmortalizationofmousecorticalneuronsderivedfromapreviouslyestablishedmousemodeldeficientinmurineglucocerebrosidase(Tybulewiczetal.,1992).RESULTSTheEF1promotordrivesexpressioninculturedmousecorticalcellsSeveralindependentstudiesestablishedthatpromotordeterminationforoptimalgeneexpressioninaspecificcelltypeisbeneficial(Dayetal.,2009;Tsuchiyaetal.,2002).Therefore,wetestedapanelofeightdifferentpromotersfusedtoenhancedgreenfluorescentprotein(eGFP)fortheirexpressioncapacityinC57BL/6primarymouseneuronalcultures(Table1).Brainsfrom17EC57BL/6embryoswereharvesteda

ndneuronalcultureswereestablished.Six-day-oldprimaryembryoniccorticalneuronalculturesweretransducedwithlentiviruscontainingrecombinantgenesencodingmPol2,Grp78,FerH,CAG,CMV13,PGK,EF1orTRE-TightfusedtoeGFP.WeidentifiedtheEF1promotorasanoptimalpromoterforprimarymousecorticalcellswithrobusteGFPexpressionfivedaysafterinfectionatmultiplicityofinfection(MOI)40(Fig.1A).Atthispoint,theinfectedprimarymousecorticalcellculturesconsistofamixedpopulationofneuronsandglialcells.OurfindingswereinagreementwithpreviousstudieswhereEF1wasidentifiedasasuitablepromoterforexpressioninratcorticalcellculturesandmouseneuralprecursorcells(Tsuchiyaetal.,2002;Zengetal.,2003).EF1-drivenexpressionofSV40-TimmortalizesprimarycorticalmousecellsExogenousintroductionoftheimmortalizinggeneknowntoinduceproliferationandincreaselife-spanbyinactivationofkeycell-cyclesuppressors(Ozeretal.,1996;Tevethiaetal.,1998;ManfrediandPrives,1994;Ozer,2000;Jhaetal.,1998).Culturesofprimaryembryoniccorticalcellswerees

tablishedfromapreviouslydescribedmousemodelrepresentativeoftype2GD(Tybulewiczetal.,1992).Six-day-oldculturesofcorticalcellswereinfectedwithEF1-SV40-TlentivirusatMOI40.Fivedaysafterinfection,SV40-Texpressingcellswereselectedbytreatingwithpuromycinforfourweeks.Afterselection,theimmortalizedcorticalcellcultureswerepassagedandexpressionofSV40-Twasconfirmedinproteinlysatesofimmortalizedculturesofeachgenotypebywesternblotting(Fig.1B).LaserscanningconfocalmicroscopyoncellsstainedwithanuclearDAPIstain(Fig.1D,E)andananti-SV40Tantibody(Fig.1F,G)revealedlocalizationofSV40-Tinthenucleus(Fig.1H,I)inimmortalizedneuronsofbothgenotypes.ToinvestigatewhetherSV40-Texpressioninducesinvivotumorgrowth,weperformedintraperitonealinjectionwith10cellsofeachgenotypeinSwissnu/numiceandmonitoredtumorformation.Allmicedisplayedvisibletumorsfourweeksafterinjection(Fig.1C).EstablishmentofCD24-positiveimmortalizedneuronsAspreviouslymentioned,culturesestablishedfrommousecortexcontainamixedcellpop

ulationconsistingofneuronsandglialcells.Weperformedimmuno-cytochemistryontheestablishedSV40-Timmortalizedcultureswiththeneuronalmarkermicrotubule-associatedprotein-2(MAP-2)andtheastroglialmarkerglialfibrillaryacidicprotein(GFAP).Theimmortalizedculturesofeachgenotypewerepositiveforbothmicrotubule-associatedprotein-2(MAP-2)andglialfibrillaryacidicprotein(GFAP)(Fig.2A,B,E,F).Interestingly,GFAP-positivecellswereonlysporadicallydetectedincultures(Fig.2F).WenextselectedcellsthatwerenegativefortheneuralstemandprecursorcellmarkerCD29andpositiveforthedifferentiatedneuronmarkerCD24(Pruszaketal.,2007).ImmortalizedcorticalcellculturesofbothgenotypesweresubjectedtopositiveCD24selectionandnegativeCD29selectionusingfluorescence-activatedcellsorting(FACS)(Fig.2C,G).AfterFACS,thevastmajorityofcellsstainedpositiveforMAP-2,whereasGFAP-positivecellswereabsent(Fig.2D,H).MultiplestudieshaveshownthatSV40-Timmortalizationofcellsinducesaberrantkaryotypes(BloomfieldandDuesberg,2015;Stone

retal.,1991;Tooulietal.,2002),aphenomenonalsofrequentlyobservedinwidelyusedcelllinessuchasHeLaandHEK293(Landryetal.,2013;StepanenkoandDmitrenko,2015).ChromosomeanalysisontheCD24-positiveimmortalizedcorticalneuronsrevealedaberrantheterogeneous Table1.PromotorpanelPromoterPromoterdescriptionReporterCMVHumancytomegalovirusimmediateearlypromoter/enhancereGFPFerHHumanferritinheavychainpromoter,SV40eGFPmPol2MurineRNApolymeraseIIpromotereGFPGrp78Hamsterglucose-responseprotein78promoter,CMVenhancereGFPCAGChicken-actinpromoter,CMVenhancereGFPPGKHumanphosphoglyceratekinasepromotereGFPHumanelongationfactor1promotereGFPTRE-TightModifiedCMVpromoterinduciblewitheGFP RESEARCHARTICLEDiseaseModels&Mechanisms(2016)9,769-778doi:10.1242/dmm.024588 DiseaseModels&Mechanisms karyotypes,whichwasexpectedastheculturesareheterogeneousandnotclonal.Chromosomalabnormalitiesincludenumericalabnormalities(+Y,+5,+14,+16)andmorphologicalabnormalitiesliketranslocation5;19(Fig.3A,B).immortalizedneu

ronshavedeficientGCaseenzymaticactivityandshowsubstratestorageThepreviouslydescribedneonatal-lethalmousemodeloftype2GDischaracterizedbysevereGCaseenzymedeficiencyandaccumulationofGlcCerinthebrain(Tybulewiczetal.,1992).WeanalyzedtheCD24-positiveimmortalizedmouseneuronsandfoundthatthesecharacteristicswerepreserved.immortalizedneuronsshowedaseveredeficiencyinenzymeactivity(3.40±0.36%relativeGCaseactivity,mean±s.e.m.)comparedwithimmortalizedneurons(93.44±12.26%relativeGCaseactivity),whichwashighlysignificant((Fig.3C).GCaseproteincouldnotbedetectedinlysateswiththepreviouslydescribedGCase-specificinhibodyMDW933 Fig.1.ExpressionanalysisandinvivotumorformationofSV40-Timmortalizedmousecorticalcells.(A)EGFPexpressioninprimaryC57BL/6mousecorticalcellculturesfivedaysafterinfectionwithEF1-eGFPlentivirusatMOI40.(B)Westernblotanalysisofproteinlysatesof(lane1)and(lane2)immortalizedcorticalcellswithantibodiesagainstSV40-Tand-actin(proteinloadingcontrol).SV40-Tproteinexpressionwa

sdetectedinbothimmortalizedcells.(C)Invivointraperitonealtumorformationafterinjectionof10cellsofimmortalizedcellsinfiveSwissnu/numicepergenotype.Fourweeksafterinjection,miceinjectedwithcellsdevelopedvisiblesolidtumors(leftpanel)whereasmiceinjectedwithcellsdevelopedsolidbloodytumors(rightpanel).ImmortalizedmousecorticalcellswerestainedforDAPI(D,E),anti-SV40-T(F,G),andtheimagemergedwithbright-fieldforvisualizationofneuronmorphology(H,I).DAPIandSV40-Tco-localizeinthenucleus(H,I).Scalebar:20 Fig.2.EstablishmentofCD24-positiveneuroncultures.Expressionof(A,E)MAP-2and(B,F)GFAPinSV40-TimmortalizedcellsbeforeFACS.BothMAP-2andGFAPareexpressedincultures.GFAPisonlysporadicallyexpressedincultures.(C,G)FACSofSV40-TimmortalizedcellswithantibodiesagainstCD24andCD29surfacemarkers.(D,H)ExpressionofMAP-2andGFAPinSV40-TneuronalculturesafterFACS.NoGFAP-positivecellsaredetected.Scalebar:20 RESEARCHARTICLEDiseaseModels&Mechanisms(2016)9,769-778doi:10.1242/dmm.024588 DiseaseModels&Mech