ebook img

Related ESCRT-III Subunit CHMP2B PDF

19 Pages·2015·5.19 MB·English
by  
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Related ESCRT-III Subunit CHMP2B

TheJournalofNeuroscience,February18,2015•35(7):3155–3173•3155 Cellular/Molecular Regulation of Postsynaptic Function by the Dementia- Related ESCRT-III Subunit CHMP2B XRomainChassefeyre,1,2*Jose´Martínez-Herna´ndez,1,2*FedericaBertaso,3,4,5NathalieBouquier,3,4,5Be´atriceBlot,1,2 MarineLaporte,1,2SandrineFraboulet,1,2YohannCoute´,6,7AnnyDevoy,8AdrianM.Isaacs,8KarinPernet-Gallay,1,2 Re´mySadoul,1,2XLaurentFagni,3,4,5andYvesGoldberg1,2,9 1InstitutNationaldelaSant´eetdelaRechercheM´edicale(INSERM),Unit´e836,F-38042Grenoble,France,2Universit´eGrenobleAlpes,GrenobleInstitut desNeurosciences(GIN),F-38042Grenoble,France,3CNRS,UMR-5203,InstitutdeG´enomiqueFonctionnelle,F-34094Montpellier,France,4Universit´es deMontpellier1&2,UMR-5203,F-34094Montpellier,France,5INSERM,Unit´e661,F-34094Montpellier,France,6INSERM,Unit´e1038,F-38054Grenoble, France,7Commissariata`l’EnergieAtomique(CEA),InstitutdeRecherchesenTechnologiesetSciencespourleVivant(iRTSV),LaboratoiredeBiologiea` GrandeEchelle,F-38054Grenoble,France,8DepartmentofNeurodegenerativeDisease,UniversityCollegeLondonInstituteofNeurology,LondonWC1N 3BG,UnitedKingdom,and9CEA,iRTSV,GroupePhysiopathologieduCytosquelette(GPC),F-38054Grenoble,France The charged multivesicular body proteins (Chmp1–7) are an evolutionarily conserved family of cytosolic proteins that transiently assemblesintohelicalpolymersthatchangethecurvatureofcellularmembranedomains.MutationsinhumanCHMP2Bcausefronto- temporaldementia,suggestingthatthisproteinmaynormallycontrolsomeneuron-specificprocess.Here,weexaminedthefunction, localization,andinteractionsofneuronalChmp2b.Theproteinwashighlyexpressedinmousebrainandcouldbereadilydetectedin neuronaldendritesandspines.DepletionofendogenousChmp2breduceddendriticbranchingofculturedhippocampalneurons,de- creasedexcitatorysynapsedensityinvitroandinvivo,andabolishedactivity-inducedspineenlargementandsynapticpotentiation.To understandthesynapticeffectsofChmp2b,wedetermineditsultrastructuraldistributionbyquantitativeimmuno-electronmicroscopy anditsbiochemicalinteractionsbycoimmunoprecipitationandmassspectrometry.Inthehippocampusinsitu,asubsetofneuronal Chmp2bwasshowntoconcentratebeneaththeperisynapticmembraneofdendriticspines.Insynaptoneurosomelysates,Chmp2bwas stablyboundtoalargecomplexcontainingothermembersoftheChmpfamily,aswellaspostsynapticscaffolds.Thesupramolecular Chmpassemblydetectedherecorrespondstoastableformoftheendosomalsortingcomplexrequiredfortransport-III(ESCRT-III),a ubiquitouscytoplasmicproteincomplexknowntoplayacentralroleinremodelingoflipidmembranes.WeconcludethatChmp2b- containingESCRT-IIIcomplexesarealsopresentatdendriticspines,wheretheyregulatesynapticplasticity.Weproposethatsynaptic ESCRT-IIIfilamentsmayfunctionasanovelelementofthesubmembranecytoskeletonofspines. Keywords: ESCRTfilaments;frontotemporaldementia;postsynapticscaffold;spinoskeleton;structuralplasticity Introduction cellular membrane remodeling during inward vesiculation of The endosomal sorting complex required for transport-III multivesicularendosomes,mitoticcellabscission,andviralbud- (ESCRT-III) is a ubiquitous protein complex that is crucial to ding(forreview,seeHenneetal.,2013;McCulloughetal.,2013). The core subunits of ESCRT-III are a family of evolutionarily conservedproteins,calledchargedmultivesicularbodyproteins ReceivedFeb.10,2014;revisedDec.23,2014;acceptedJan.6,2015. (Chmp1-7). Dominant mutations in subunit CHMP2B cause Authorcontributions:R.S.,L.F.,andY.G.designedresearch;R.C.,J.M.-H.,F.B.,N.B.,B.B.,M.L.,S.F.,Y.C.,A.D., frontotemporaldementia(FTD;Skibinskietal.,2005;Isaacset A.M.I.,K.P.-G.,andY.G.performedresearch;Y.C.andA.M.I.contributedunpublishedreagents/analytictools;R.C., J.M.-H.,F.B.,N.B.,Y.C.,K.P.-G.,R.S.,L.F.,andY.G.analyzeddata;R.C.,J.M.-H.,A.M.I.,R.S.,L.F.,andY.G.wrotethe al., 2011), raising the question whether some neuron-specific paper. processiscontrolledbythisprotein.Chmp2bandotherESCRT- ThisworkwassupportedbyagrantfromtheAgenceNationaledelaRecherche(ANR),MALZprogram, IIIsubunitsexistinthecytosolinamonomeric,autoinhibited France.R.C.andM.L.weresupportedbytheMiniste`redel’EnseignementSupe´rieuretdelaRecherche, state.Afterreleaseofintramolecularinhibition,individualsub- France.J.M.-H.andN.B.weresupportedbyANR.A.M.I.wassupportedbytheUnitedKingdomMedical ResearchCouncil.WethankY.SaoudiforhelpwithmicroscopyandN.Jaafari,A.Petiot,andW.Weissenhorn unitspolymerizeintocircularorhelicalfilaments(20–200nm forcriticallyreadingthismanuscript. wide,subjecttosubunitspecies)thatadheretolipidmembranes *R.C.andJ.M.-H.contributedequallytothiswork. andchangetheircurvature(Shimetal.,2007;Hansonetal.,2008; Theauthorsdeclarenocompetingfinancialinterests. Lataetal.,2008;Bajoreketal.,2009;Henneetal.,2012;Buchk- ThisarticleisfreelyavailableonlinethroughtheJNeurosciAuthorOpenChoiceoption. ovichetal.,2013). CorrespondenceshouldbeaddressedtoeitherYvesGoldbergorRémySadoul,,InstitutdesNeurosciences,UJF Site Sante´, Chemin Ferrini, 38042 Grenoble cedex 9, France. E-mail: [email protected] or [email protected]. R.Chassefeyre’spresentaddress:TheScrippsResearchInstitute,LaJolla,CA92037. ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense DOI:10.1523/JNEUROSCI.0586-14.2015 (http://creativecommons.org/licenses/by/3.0),whichpermitsunrestricteduse,distributionandreproductioninany Copyright©2015Chassefeyreetal. mediumprovidedthattheoriginalworkisproperlyattributed. 3156•J.Neurosci.,February18,2015•35(7):3155–3173 Chassefeyre,Martínez-Herna´ndezetal.•ESCRT-IIIFunctionatSynapses Table1.Plasmidsusedinthisstudy Plasmidname Cloningtechnique Cloningsites Feature Tags Source pIRES2-EGFP Startingvector Clontech pIRES2-EGFP-CHMP2BWT In-Fusion(Clontech) XhoI,XhoI RNAires. C-terminalFLAG Thisstudy pBI-CMV3 Startingvector Clontech pBI-CMV3-CHMP2BWT In-Fusion(Clontech) BamHI,HindIII RNAires. Thisstudy pBI-CMV3-CHMP2BR19/22/26A In-Fusion(Clontech) BamHI,HindIII RNAires. Thisstudy pBI-CMV3-CHMP2BL4D/F5D In-Fusion(Clontech) BamHI,HindIII RNAires. Thisstudy pBI-CMV3-CHMP2BL207D/L210D In-Fusion(Clontech) BamHI,HindIII RNAires. Thisstudy pSuper-mCherry Startingvector Bellyetal.,2010 pSuper-mCherryshRNAvsCHMP2B Standardligation HindIII,BglII Bellyetal.,2010 pSuper-mCherryshRNAvsGFP Standardligation HindIII,BglII Thisstudy pCDNA3.1-CHMP2BWT Topo(Invitrogen) C-terminalHA Thisstudy pCDNA3.1-CHMP2BR19/22/26A Topo(Invitrogen) C-terminalHA Thisstudy pCDNA3.1-CHMP4B Topo(Invitrogen) C-terminalFLAG Bodonetal.,2011 Thecloningtechniqueusedtoconstructeachplasmidand,ifapplicable,theassociatedkitsupplierareindicated.RNAires.IndicatestheencodedmRNAisresistanttoRNAibytheChmp2bshRNA. We have previously shown that Chmp2b preferentially po- pSuper-mCherry. cDNAs encoding wild-type (WT) or mutant ver- lymerizesundertheplasmamembraneand,whenoverexpressed, sionsoftheRNAi-resistantChmp2b(Bellyetal.,2010)wereclonedin organizesintohelicesthatdeformpatchesofthecellsurfaceinto thepBI-CMV3vector(Clontech),yieldingpBI-CMV3-Chmp2bWT, long protrusions (Bodon et al., 2011). Holomeric ESCRT-III pBI-CMV3-Chmp2bR19/22/26A, and pBI-CMV3-Chmp2bL4D/F5D. In consistsofheteropolymericChmpfilamentsassembledatsitesof thesevectors,Chmp2biscoexpressedwithfluorescentproteinZsGreen from a bidirectional promoter. cDNA encoding wild-type, RNAi- endosomalorplasmamembranedeformation,afterrecruitmentby resistantChmp2b(carryingasingleC-terminalFlagtag;Bodonetal., the upstream-acting complexes ESCRT-I and -II and/or by the 2011)wasinsertedupstreamoftheIRESsequenceofthepIRES2-EGFP Chmp4b-bindingproteinAlix.ESCRT-IIIassembliesatmembranes vector(Clontech),yieldingpIRES2-EGFP-Chmp2b(usedinFig.6B).In arethoughttobetransient,beingdissociatedafterinteractionwith thisvector,Chmp2biscoexpressedwithEGFPfromabicistronictran- the depolymerizing ATPase Vps4. This depolymerization step is script.Allconstructswereverifiedbysequencing(BeckmanCoulter). coupledtoscissionofmembranenecksmodeledbyESCRT-IIIfila- Antibodies.SeeTable2fordetailsofantibodiesanddilutions. ments(Henneetal.,2013). Animals.AnimalswerehandledandkilledinconformitywithEuro- FTD-associatedCHMP2Bmutantsarederegulatedandform peanlawandinternalregulationsofINSERM.PregnantOncinsFrance aberrantpolymericstructuresintransfectedcells(Skibinskietal., soucheA(OFA;asubstrainofSpragueDawley)rats(CharlesRiver)were 2005).StrongexpressionofsuchaCHMP2Bmutantincultured usedforprimaryneuroncultures,and2-month-oldOFAratsofboth corticalneuronsisrapidlylethal(Leeetal.,2007),andtransgenic sexeswereusedforbiochemicalanalysisofbraintissue.Theratswere miceexpressingthesamemutantthroughouttheirbraindisplay deeplyanesthetizedbyisofluraneinhalationbeforedecapitation,quickly followedbycesareansectionorbraindissection.Three-month-oldmale progressive neurodegenerative changes (Ghazi-Noori et al., C57BL/6miceand1.3-month-oldfemaleChmp2bKOmice(herecalled 2012).Theseeffectshavebeenlinkedtodefectsinautophagicand KD mice since knock-out is incomplete;Ghazi-Noori et al., 2012) or endosomaldegradationpathways(Filimonenkoetal.,2007;Lee femaleWTlittermates(onC57BL/6background)wereusedforimmu- etal.,2007;Urwinetal.,2010).However,thesephenotypesarise nohistochemistry and electron microscopy (EM) studies. Mice were fromatoxicgainratherthanlossoffunctioninthemutantallele. anesthetizedbyintraperitonealinjectionof0.1mlofsodiumpentobar- Indeed, degeneration was not seen in knock-out mice severely bital(5.6%w/v;CEVASante´Animale)andintracardiallyperfusedwith (thoughnotcompletely)deficientforChmp2b(Ghazi-Nooriet phosphate-buffered 0.9% NaCl (PBS), followed by 0.1 M phosphate- al., 2012). Furthermore, Chmp2b depletion did not affect sur- buffered4%formaldehyde,pH7.4,supplementedwith0.05%glutaral- vivalofculturedneurons(Leeetal.,2007;Bellyetal.,2010)but dehyde(Sigma).Thebrainswerecarefullyremovedandpostfixedfor4h significantlyimpairedthemorphologicalmaturationofdendritic inthesamefixative,and60-(cid:2)m-thickcoronalbrainsectionswerecut spines(Bellyetal.,2010).This,togetherwiththefactthatsynap- withaVibratome(VT1000S;Leica). togenesis correlates with Chmp2b gene transcription in some Cellcultureandtransfection.Primaryculturesofhippocampalneurons neuronalpopulations(Xuetal.,2012),ledustohypothesizethat werepreparedasdescribedpreviously(Bellyetal.,2010).Briefly,hip- pocampiweredissectedfromE18ratembryos,treatedwithtrypsin,and normalChmp2bandESCRT-IIImayfunctionatsynapses. disruptedbyseventoeightcyclesofaspirationandejectionthrougha Here,weshowthatChmp2bisrequiredformaintainingthe micropipettetip.Dissociatedhippocampalcellswereseededatadensity complexityofdendritictreesaswellasthestability,efficacy,and of1.5(cid:2)104/cm2in35mmdishesoncoverslipsprecoatedfor4hwith50 plasticityofexcitatorysynapses.Wealsoshowthattheprotein (cid:2)g/mlpoly-D-lysine(Sigma),inneuronalculturemedium(Neurobasal preferentiallylocalizesbeneaththeplasmamembraneofdendritic mediumcontainingB27supplement,penicillin/streptomycin,and0.5 spinesandisboundtostableESCRT-IIIcomplexesassociatedwith mML-glutamine;Invitrogen)supplementedwith10%heat-inactivated thepostsynapticcytoskeleton,suggestingthatESCRT-IIIpolymers horseserum(Invitrogen).Fourteen-mililimeter-wideroundcoverslips cooperatewithpostsynapticscaffoldsandactinfilamentstoshape wereusedforimmunostaining,and30-mm-widecoverslipswereused dendriticspines. for live imaging. Neurons were maintained in water-saturated 95% air/5%CO at37°C.Theseedingmediumwasreplacedafter20hwith MaterialsandMethods serum-free2neuronalculturemedium. Plasmids.SeeTable1fordetailsofallconstructs.ThepSuper-mCherry Neuronsweretransfectedusingacalciumphosphatemethodopti- vector was described by Belly et al. (2010) and expresses the inserted mizedforhighlymatureneuronalcultures(Goetzeetal.,2004;Jiangand shRNAtogetherwithmCherry.TheChmp2b-specificshRNA(Leeetal., Chen, 2006). For cotransfection experiments, for each dish, 1 (cid:2)g of 2007;Bellyetal.,2010)andtheGFP-specificshRNA[previouslyshown pSuper-mCherry-based plasmid was mixed with 2 (cid:2)g of pBI-CMV3- byAlvarezetal.(2006)tohavenoeffectonneuronalmorphologyor basedorpIRES2-EGFP-basedplasmid.Fivemicrolitersof250mMCaCl2 electrical properties] have been described and were inserted into were added to 45 (cid:2)l of water containing 3 (cid:2)g of DNA, mixed, and Chassefeyre,Martínez-Herna´ndezetal.•ESCRT-IIIFunctionatSynapses J.Neurosci.,February18,2015•35(7):3155–3173•3157 Table2.Antibodiesusedinthisstudy Dilutions Antibody Species Company(reference) IF WB IHC/EM Anti-Chmp2b Rabbitpolyclonal Abcam(ab33174) 500 10,000 1000 Anti-Chmp2b Mousemonoclonal Covalab 100 2000 Anti-Chmp4b Rabbitpolyclonal Abcam(ab105767) 500 2000 Anti-Chmp4b Mousemonoclonal Covalab 100 1000 Anti-Alix Rabbitpolyclonal Covalab(ab0204) 6000 Anti-PSD95 Mousemonoclonal NeuroMab(73-028) 200 4000 Anti-GluN1 Mousemonoclonal SynapticSystems(114011) 200 4000 Anti-Synaptophysin Mousemonoclonal MerckMillipore(MAB5258) 10,000 Anti-HA Mousemonoclonal CellSignaling(2367) 200 5000 Anti-HA Rabbitpolyclonal Sigma-Aldrich(H6908) 5000 Anti-Flag Mousemonoclonal Sigma-Aldrich(F3165) 1000 10,000 Anti-Flag Rabbitpolyclonal Sigma-Aldrich(F7425) 1000 5000 Anti-Actin Mousemonoclonal Millipore(MAB1501R) 20,000 Anti-Tubulin Mousemonoclonal GiftfromA.Andrieux’slaboratory 20,000 Anti-PanMAGUK Mousemonoclonal NeuroMab(75-029) 4000 Anti-GluN2B Mousemonoclonal NeuroMab(75-097) 4000 Anti-Glu2 Mousemonoclonal Millipore(MABN71) 5000 Anti-ShankIII Rabbitpolyclonal SantaCruzBiotechnology(30193) 1000 Anti-Homer Mousemonoclonal SantaCruzBiotechnology(17842) 2000 Anti-MouseHRP Goatpolyclonal JacksonImmunoResearch(115-035-166) 20,000 Anti-RabbitHRP Goatpolyclonal JacksonImmunoResearch(115-035-044) 20,000 Anti-MouseAlexaFluor488 Goatpolyclonal Invitrogen(A-11029) 1500 Anti-MouseAlexaFluor594 Goatpolyclonal Invitrogen(A-11032) 1500 Anti-MouseCy5 Goatpolyclonal JacksonImmunoResearch(115-175-146) 1000 Anti-RabbitAlexaFluor488 Goatpolyclonal Invitrogen(A-11034) 1500 Anti-RabbitAlexaFluor594 Goatpolyclonal Invitrogen(A-11037) 1500 Anti-RabbitCy5 Goatpolyclonal JacksonImmunoResearch(111-175-144) 1000 Theantibodydilutionsweusedareindicatedforeachapplication.IF,Immunofluorescence;WB,Westernimmunoblotting;IHC,immunohistochemistry. immediatelyaddeddropwiseto50(cid:2)lof2(cid:2)N,N-Bis(2-hydroxyethyl)- a humidified chamber with primary antibodies diluted in a blocking 2-aminoethanesulfonicacid(BES)-bufferedsalinesolution,pH7.1.The solution.AfterwashinginPBS,cellswereincubatedfor1hwithsecond- mixturewasgentlystirredbytappingthetubeandbubblingaironce,left aryantibodiesconjugatedtoAlexaFluor488,AlexaFluor594,orCya- toincubatefor10min,andaddedtocoverslipsin1mloftransfection nine5(Cy5),dilutedintheblockingsolution.Coverslipswererinsedand medium[Eagle’sMEM(Invitrogen)including2.2%NaHCO andsup- mountedinMowiol. 3 plementedwithB27,0.05mMglutamine,and20mMglucose].After1hin Golgistaining.Thedendriticspinesofhippocampalneuronsinsitu theincubator,thecoverslipsweretransferredtodishescontaining2mlof werevisualizedbytheGolgiimpregnationtechnique.Forthis,weused transfection medium (without B27) preincubated for 30 min in 10% theFDRapidGolgiStainkit(FDNeuroTechnologies).Hippocampaltis- CO .Thedisheswerequicklyreturnedto5%CO andleftfor5–10min sueblocks(2.5–3mm)wereimmersedinequalvolumesofsolutionsA 2 2 beforereturningthecoverslipstotheiroriginal(conditioned)neuronal andBfor7dandimpregnatedwithsolutionCfor48hat4°C.Then,the culture medium. For immunofluorescence of exogenously expressed brainswerecutwithavibratomeinto150-(cid:2)m-thicksectionsembedded proteins,neuronswerefixed16–20haftertransfection.ForRNAiexper- inagarose,washedtwiceindouble-distilledwater,andincubatedfor10 iments,fixationorliveimagingtookplace72haftertransfection. mininamixtureofonepartofsolutionD,onepartofsolutionE,andtwo HEK-293TandHeLacellsweregrowninDMEM(Invitrogen)supple- partsofdouble-distilledwater.Sectionswerewashedtwice,dehydrated mentedwith10%FBSand2mML-glutamine(Sigma).Cellsweretrans- withincreasingethanol,andmountedwithepoxyresin(Fluka)between fectedafter24hofcultureusingjetPEI(Polyplus)forHeLacellsand twocoverslips.Theslideswereanalyzedbyinvestigatorswithoutknowl- calciumphosphateforHEKcells.Cellswerefixedorlysed36hafter edgeofgenotype.Ineachofthreeanimalspergenotype,ninesecondary transfection. ortertiarypyramidalneurondendriteswererandomlyselectedinthe Immunohistochemistry. Brain sections were incubated in 10% goat CA1stratumradiatum.Selectioncriteriawere(1)dendritesdisplayed preimmuneserum(GPi)dilutedin50mMTrisbuffer,pH7.4,containing completeGolgiimpregnation,(2)thecelltypewasclearlyidentifiable, 0.9%NaCl[Tris-bufferedsaline(TBS)],with0.2%TritonX-100,for1h. and(3)theminimumlengthofthedendriteswas50(cid:2)mfromthesoma. Sectionswerethenincubatedovernightat4°Cinaprimaryantibodyata Stacksofbright-fieldimageswith0.3(cid:2)mspacingwereacquiredwitha finalproteinconcentrationof1–2(cid:2)g/ml,whichhadbeendilutedinTBS Zeiss Axioskop 50 microscope with 63(cid:2) oil objective (NA 1.4; Plan- containing1%GPi.AfterseveralwashesinTBS,thesectionswerefurther Aprochromat)coupledtoaCCDcamera(CoolSnapES;RoperScien- incubatedfor2hinbiotinylatedgoatanti-rabbitIgG(SigmaImmuno tific)operatedbyMetaviewsoftware(MolecularDevices).Imageswere Chemicals)diluted1:100inTBScontaining1%GPi.Theywerethen analyzedwithImageJ.Thenumberofspinesofeachbasictypeperunit transferred into avidin–biotin–peroxidase complex (ABC kit; Vector dendriticlengthwasmeasured. Laboratories)diluted1:100for2hatroomtemperature.Boundperoxi- Fluorescencemicroscopy.Lightmicroscopyimageswereacquiredwith daseenzymeactivitywasrevealedusingDAB(0.05%inTB,pH7.4)as a Zeiss LSM 710 inverted confocal microscope, using a 63(cid:2) Plan- thechromogenand0.01%H O asthesubstrate.Finally,sectionswere Apochromatoil-immersionobjective(NA1.4;Zeiss)or(forShollanal- 2 2 air-driedandcoverslipped. ysis)a40(cid:2)oilobjective(NA1.3;Zeiss).Theinstrumentwascontrolled Immunofluorescentstaining.Culturedneuronswerefixedfor20minat byZensoftwareandincludesacousto-opticalfiltersforattenuatinglaser roomtemperatureinphosphate-buffered4%paraformaldehydesupple- power,scanningmirrorswithadjustablespeed,andresolutionofemitted mented with 4% sucrose. After three washes in PBS containing 0.1% lightbyadiffractiongratingandbeamguidesforsettingspectralcollec- glycine,cellswerepermeabilizedinPBScontaining0.25%TritonX-100 tionwindows.Laserlinesat488,561,and633nmwereusedforexciting for10minat4°CandrinsedwithPBS.Coverslipswereblockedfor1hin fluorescenceofEGFP,ZsGreen,orAlexaFluor488;mCherryorTexas PBScontaining5%GPiandincubatedfor1–2hatroomtemperaturein Red;andCyanine5,respectively.Formulticolorimaging,collectionwin- 3158•J.Neurosci.,February18,2015•35(7):3155–3173 Chassefeyre,Martínez-Herna´ndezetal.•ESCRT-IIIFunctionatSynapses dowswereadjustedbasedonemissionspectratominimizecross-channel r contamination. Pinhole size was usually set to generate 0.9-(cid:2)m-thick rN(cid:3)1(cid:4)r (cid:5)1r , opticalslices((cid:2)1Airyunit)atallwavelengths.Forfixedsamples,laser 1 2 powerwasadjustedto0.5–1.25mWinallchannels,andthesignalwas wherer istheradialdistancefromthemembranetotheparticleandr is 1 2 averagedoverfourscansanddigitizedonan8bitgrayscale.Intheexper- theradialdistanceoftheparticlefromthegeometriccenterofthespine. imentsshowninFigure2,BandC,Z-stackswereacquiredwith0.09(cid:2) Tocorrecttheradialdistributionfordistance-dependentincreaseinbin 0.09(cid:2)0.2(cid:2)m3voxelsizeandprocessedbyablinddeconvolutionalgo- area,particledensitiesalongtheradialcoordinateweredividedbynor- rithm(AutoQuant;MediaCybernetics)toimproveresolutionandre- malizedradialdistance,thusgivingincreasedweighttoparticledensity move shot noise. The processed images were slightly contrasted with close to the spine center. Three-dimensional reconstruction of spine ImageJforclarity. headsfromserialsectiondigitalimageswasdoneusingReconstruct ToevaluatetheprobabilityofrandomoverlapbetweenChmp2band software(JohnC.Fiala,DepartmentofBiology,BostonUniversity, postsynapticdensitity-95(PSD-95)clusters,adendriticregionwasse- MA). For comparisons between wild-type and Chmp2b-deficient lectedinawaytoexcludeasmuchaspossibleemptypixels.Pearson’s mice,acquisitionofmicroscopyimagesandmorphometricquantifi- correlationcoefficient(r)betweenChmp2bandPSD-95intensityvalues cationwereperformedwithoutknowledgeofgenotype. wascalculatedfortheactualimage,andtheprobabilityofobtainingthe PurificationofsynaptoneurosomesandPSDfractionation.Subcellular observedcoefficientwascalculatedbasedon1000randomizedimages fractionationwasperformedaccordingtoCarlinetal.(1980)withminor similartotheoneshownusingtheImageJplug-inJACOP(Bolteand modifications.Sampleswerekeptat4°Cduringallprocedures.Inbrief, Cordelie`res,2006). oneratforebrainwashomogenizedin13mlofice-coldisotoniclysis ForShollanalysis,images(1024(cid:2)1024pixels;350(cid:2)350(cid:2)m;four buffer(0.32Msucrose,4mMHEPES,pH7.4,proteaseinhibitors[EDTA- opticalsectionsinz)weretakenfromneuronscoexpressingZsGreenand freeComplete;RocheMolecular)]usingaTeflon-glasshomogenizer(15 mCherry,withinthesamefluorescencerangeforallneurons(40neurons strokes).Thehomogenatewascentrifugedat1400(cid:2)gfor10minto pertransfectioncondition,threeindependentcultures).Shollanalysis removenuclei.Thepostnuclearsupernatant(S1)wasspundownat4°C wasperformedonmaximumintensityprojectionsofmCherryfluores- for10minat13,800(cid:2)g,yieldingcrudesynaptosomes(pelletP2)and cenceusingtheShollAnalysispluginofImageJ. crudecytosol(supernatantS2).TheP2fractionwashomogenizedinlysis Forpostsynapticspinecountinginculturedneurons,transfectedneu- bufferandlayeredontodiscontinuoussucrosedensitygradients(made ronswerestainedbyimmunofluorescencewithanti-PSD-95antibody of3mleachof0.85,1.0,and1.2Msucroselayers).Thegradientswere and Cy5-conjugated secondary antibody. Triply labeled neurons centrifugedat82,500(cid:2)g(27,000rpminaBeckmanSW41rotor)for120 (mCherry-ZsGreen-Cy5)wereimagedinallthreechannelsasabove.For min,andthepurifiedsynaptoneurosomefraction(SYN)wascollectedat each neuron, 10–20 optical sections were acquired (voxel size, 0.1 (cid:2) the1.0–1.2Minterface.Thesynaptoneurosomeswereresuspendedin5 0.1(cid:2)0.2(cid:2)m3).ImageswereanalyzedwithImageJ.Postsynapticspines voloflysisbufferandspundownfor20minat150,000(cid:2)g.Thepelleted weremanuallyidentifiedinsuccessivesectionsasspine-likedendritic synaptoneurosomeswereresuspendedinTritonbuffer[50mMHEPES, protrusions(visualizedbymCherryfluorescence)carryingPSD-95clus- pH7.4,0.5%TritonX-100,proteaseinhibitors(EDTA-freeComplete; ters.Spineswerecountedalong(cid:3)540(cid:4)220(cid:2)m(mean(cid:4)SD)dendritic RocheMolecular),2mMEDTA],stirred30mininthecoldroom,and lengthperneuron.Similarresultswereobtainedwithtwoindependent centrifugedat32,000(cid:2)gfor20mintoobtainthePSDIpellet.PSDIwas cultures,allowingustopoolthedata.DualimmunostainingofmCherry- resuspendedin50mMHEPES,pH7.4.One-halfofthesuspensionwas transfectedneuronswithanti-PSD-95andanti-synapsinantibodiesver- solubilizedagainbyadding1volof1%TritonX-100andcentrifugedat ifiedthatinneuronsatthisstage,virtuallyallPSD-95clusterslocatedon 200,000(cid:2)gfor20min,yieldingpelletPSDII.Theotherhalfwassolubi- spineswerecolocalizedwiththepresynapticmarkersynapsin.Imagesof lizedbyadding1volof6%Sarkosylandcentrifugedat200,000(cid:2)gfor neuronsusedtostudyRNAieffectswereacquiredandanalyzedbyinves- 1h,yieldingPSDIII. tigatorsblindtothetransfectedconstructs. Gelfiltrationofsynaptoneurosomeextract.Size-exclusionchromatog- Electron microscopy. Free-floating brain sections were incubated in raphy (gel filtration) was used for separation of Chmp2b-containing 10%GPidilutedinTBSfor1hatroomtemperaturetoblocknonspecific molecularassembliesthathadbeensolubilizedbymilddetergentextrac- staining.Sectionswerethenincubatedovernightat4°CinTBScontain- tionfromgradient-purifiedsynaptoneurosomes.Chromatographywas ing1%GPiandaprimaryantibodyatafinalproteinconcentrationof performedat4°Cona120mlHiPrepSephacrylS-300HRcolumncou- 1–2(cid:2)g/ml.Thesectionswerethenlabeledeitherbyimmunogoldorby pledtoanAktaFPLCsystem(GEHeathcare).Thecolumnwasequili- immunoperoxidase.Forimmunoperoxidaselabeling,wefollowedthe brated,andseparationwasachievedin25mMTris-HCl,pH7.4,150mM immunohistochemistryprocedure,asdescribedabove(startingfromthe NaCl,0.5%TritonX-100,10mMNaF,10mM(cid:6)-glycerophosphateand secondaryantibodystep).Forimmunogoldlabeling,sectionswereincu- proteaseinhibitors(Roche)ataflowrateof0.5ml/min.Synaptoneuro- batedfor3hwithgoatanti-rabbitIgGcoupledto1.4nmgoldbeads somes freshly prepared from one brain were resuspended in 3 ml of (Nanoprobes) diluted 1:300 in TBS containing 1% GPi. After several deoxycholate (DOC) buffer [10 mM Tris-HCl, pH 9.0, 1% sodium washesinPBS,thesectionswerepostfixedin1%glutaraldehydediluted deoxycholate,150mMNaCl,proteaseinhibitors(Complete;RocheMo- in the same buffer for 10 min. They were washed in double-distilled lecular),andphosphataseinhibitors(P0044;Sigma-Aldrich)].Thesus- water,followedbysilverenhancementofthegoldparticleswithaHQ pension was stirred by rotation at 4°C for 30 min and centrifuged at Silverkit(Nanoprobes). 100,000(cid:2)gfor20min.Thesupernatantwasfiltered(poresize,0.2(cid:2)m; Theperoxidase-reactedsectionsandthegold–silver-labeledsections Millipore),and2.5mlwasloadedonthecolumn.Fractions(2ml)were wereprocessedforEM.Thisincludedtreatmentwithosmiumtetroxide collected,andtheelutionprofilewasanalyzedbyimmunoblottingand (1% in 0.1 M PB), block staining with uranyl acetate, dehydration bymeasurementoftotalproteinconcentrationwithBCAproteinassay throughagradedseriesofethanolsolutions,andflatembeddingonglass (Pierce).Theprofilewascalibratedbymeasuringtheelutionvolumeof slidesinDurcupan(Fluka)resin.Regionsofinterestwerecutat70nmon proteinstandardsinthesameconditions(MWGF1000;Sigma-Aldrich). anultramicrotome(ReichertUltracutE;Leica)andcollectedonone-slot Immunoprecipitation.ForimmunoprecipitationofthebrainChmp2b coppergrids.Stainingwasperformedondropsof1%aqueousuranyl complex,2mg(totalprotein)ofgradient-purifiedsynaptoneurosomes acetate,followedbyReynolds’sleadcitrate.EMimageswereacquiredin wereresuspendedin1mlofDOCbufferandincubatedfor30minat4°C. a JEOL-1200 electron microscope with a digital camera (Veleta, SIS; After centrifugation at 100,000 (cid:2) g for 20 min, the supernatant was Olympus) and analyzed with ImageJ. Measurements of gold particle collectedanddilutedwith3volofimmunoprecipitationbuffer(50mM countsalongtheradialandtangentialcoordinatesofthespinecytoplasm Tris-HCl,pH7.4,150mMNaCl,and1%TritonX-100).Thesynapto- were acquired and normalized as described by Ra´cz and Weinberg somallysatewasincubatedfor2.5hat4°Cwith15(cid:2)gofantibodythat (2008).Briefly,normalizedradialposition,r (thefractionofthedis- had been covalently coupled to superparamagnetic protein A beads N tancefromthecentertotheplasmamembrane),wascomputedaccord- (Dynabeads, 10001D; Invitrogen) by cross-linking with dimethyl- ingtothefollowingequation: pimelimidate(Sigma)accordingtoHarlowandLane(1988).Afterex- Chassefeyre,Martínez-Herna´ndezetal.•ESCRT-IIIFunctionatSynapses J.Neurosci.,February18,2015•35(7):3155–3173•3159 Figure1. Chmp2bexpressioninthemousebrain.A,Anadultmalemousewasdissected,andequalamountsoftotalproteinfromeachtissuewereanalyzedbySDS-PAGEandimmunoblotting withtheindicatedantibodies.ThearrowheadindicatestheChmp4b-specificband.(cid:7)-Tubulinwasusedastheloadingcontrol.B,Mousebrainsatincreasingdevelopmentalageswereanalyzedby immunoblottingwiththeindicatedantibodies.PSD-95wasusedasamarkerofsynaptogenesis.P,Postnatalage(days);1M,1month.C,Equalproteinamountsfromtheindicatedregionsofanadult mousebrainwereanalyzedbyimmunoblottingasabove.D,Frontal(left)orsagittal(right)sectionsfromadultmousebrainwerestainedbyimmunoperoxidasewithanti-Chmp2bantibody.Cx, Cortex;cc,corpuscallosum;Hpc,hippocampus;Hb,habenula;Ce,centralamygdala;OB,olfactorybulb;CB,cerebellum.E,Left,ImmunostainedCA1layerinafrontalsectionofthehippocampus.Note thestrongstainingofpyramidalcellbodiesandtheweakerbutdefiniteimmunoreactivityoftheneuropil.SO,Stratumoriens;Pyr,stratumpyramidale;SR,stratumradiatum.Right,Higher magnificationoftheboxedregion.Immunoreactivityisobservedinapicaldendrites(arrowheads)andinterneurons(arrows).F,SpecificityofChmp2bimmunostaining.Asacontrol,weusedbrain sectionsfromChmp2bKOmice(Ghazi-Noorietal.,2012),herecalledChmp2bKDbecauseofthepresenceofresidualChmp2bproteininthesemice(Fig.7G,H).Chmp2bimmunoreactivitywas revealedbyimmunoperoxidaseinafrontalsectionofthebrainofa5-week-oldwild-type(Wt)mouse(left)orinanequivalentsectionofasiblingKDmouse(right).NotethatdevelopmentofDAB stainingwasperformedidenticallyforthetwosections,butforalongertimethaninD,whichthereforeappearslighterthanF(Wt). tensivewashinginbuffercontaining50mMTris-HCl,pH7.4,0.25% Immunoblotting.CellsortissuelysateswereresuspendedinLaemmli sodiumdeoxycholate,and0.75%TritonX-100,togetherwithincreasing bufferandresolvedbySDS-PAGEin8–12%polyacrylamidegels.Pro- concentrationsofNaCl(150–500mM),boundproteinswereelutedby teinswereelectrotransferredontoPVDFmembranes.Membraneswere boiling in Laemmli buffer, resolved by SDS-PAGE, and blotted onto blockedfor30mininTBScontaining0.1%Tween20and5%drymilk PVDFmembranes(Millipore).Forsamplesfirstresolvedbygelfiltra- andincubatedfor1hwithprimaryantibodiesdilutedintheblocking tion,fractionsofinterestwerepooledandimmunoprecipitatedunder solution.AftersixwashesinTBS–Tween,themembraneswerefurther thesameconditions. incubatedfor30minwithsecondaryantibodiescoupledtoHRP,washed Forimmunoprecipitationoftaggedproteins,transfectedHEK-293T sixmoretimesasbefore,andincubatedwithluminescence-generating werewashedwithPBSandhomogenizedinisotoniclysisbuffer,andthe HRPsubstrate,andboundantibodieswererevealedbyluminographyon lysatewascentrifugedat1000(cid:2)gfor10min.Thepostnuclearsuper- film. natantwascentrifugedat50,000(cid:2)gfor1h,andthepelletedmem- Massspectrometryanalysis.ADOCextractfromgradient-purifiedsyn- braneswereresuspendedinDOCbuffer.Immunoprecipitationwas aptoneurosomeswasfractionatedbygelfiltration,andthehigh-molecular- performedwithanti-tagantibodyimmobilizedonproteinADyna- weight (HMW) fractions were pooled and immunoprecipitated with beadsasabove. controloranti-Chmp2bantibody.Twoother,independentlyprepared 3160•J.Neurosci.,February18,2015•35(7):3155–3173 Chassefeyre,Martínez-Herna´ndezetal.•ESCRT-IIIFunctionatSynapses Figure2. Chmp2bformsclustersindendrites.A,Culturedmousecorticalneuronswerelysedatvariousstagesofmaturation,andequalamountsoflysateproteinwereanalyzedbyimmuno- blottingwiththeindicatedantibodies.PSD-95wasusedasamarkerforsynaptogenesis.B,HippocampalneuronsweretransfectedwithemptypSuper-mCherryvector,fixedat21DIV,andstained bydualimmunofluorescencewithanti-Chmp2bandanti-PSD-95antibodies.Left,Punctatestainingofadendriticsegmentinatransfectedneuron(singleconfocalplane).Stainingofnontransfected cellsisalsovisible.Merge,Overlap(yellow)betweenChmp2bandPSD-95puncta,indendriticshaftsandspines,visualizedwithmCherry(whitecontour).Thenumberedarrowheadsindicatespines withcolocalizedpuncta.Right,Enlargedviewsofthesamespines.Scalebar,20(cid:2)m.C,Top,Dendriticsegmentdoublestainedbyanti-Chmp2bandanti-PSD-95.Bottom,Sameafterrandomization ofChmp2bstainingpattern.Thecorrelationcoefficient(r)betweenactualChmp2bandPSD-95stainingintensitiesissignificantlyhigherthanthemeanrvaluegeneratedbyrandomizedChmp2b patterns(p(cid:6)0.0001,ztest).D,SpecificityofChmp2bimmunofluorescence.HippocampalneuronsweretransfectedwithpSuper-mCherryexpressingChmp2b-specificshRNA,stainedby immunofluorescencewithanti-Chmp2b,andimagedbyconfocalmicroscopy.Left,Representativetransfectedneuron,visualizedbymCherry.Middle,Chmp2bimmunostainingofthesamefield. Arrowheadsindicatethepositionofthetransfectedcelldendritesandsoma.N,Neighboring,nontransfectedneuron.Right,Immunostaining(redhotpseudocolor)ismostlyabsentfromthe transfectedneuron(outlinedinwhite),thoughclearlydetectableinthenontransfectedcell(N).Scalebar,10(cid:2)m. synaptoneurosomeextractsweredirectlyimmunoprecipitatedwiththe ona300(cid:2)m(cid:2)5mmPepMapC18precolumnandseparatedona75 sameantibodies.Thethreepairsofimmunoprecipitatesweredigested, (cid:2)m(cid:2)250mmC18column(PepMap;Dionex).Thenano-LCmethod and the resulting peptides were analyzed by liquid chromatography- consistedofa120mingradientataflowrateof300nl/min,rangingfrom tandemmassspectrometry(LC-MS/MS)asdescribedfurtherbelow.In 5to37%acetronitrilein0.1%formicacidfor114minbeforereaching eachofthethreeexperiments,peptidesderivedfromcontrolandanti- 72%forthelast6min.MSandMS/MSdatawereacquiredusingXcalibur Chmp2bimmunoprecipitateswereanalyzedinconsecutiveLC-MS/MS (ThermoFisherScientific).Sprayvoltageandheatedcapillarywere,re- runsperformedonthesameday. spectively,setat1.4kVand200°C.Surveyfull-scanMSspectra(m/z(cid:5) Forproteindigestion,proteinsinLaemmlibufferwerestacked(2mm) 400–1600)wereacquiredintheOrbitrapwitharesolutionof60,000after on SDS-PAGE gels (4–12% NuPAGE gels; Invitrogen) before being accumulationof106ions(maximumfillingtime,500ms).The20most stainedwithCoomassieblueR-250(Bio-Rad).Thegelbandwasmanu- intenseionsfromthepreviewsurveyscandeliveredbytheOrbitrapwere allyexcisedandcutinpiecesbeforebeingwashedbysixsuccessiveincu- fragmentedbycollision-induceddissociation(collisionenergy,35%)in bations of 15 min in 25 mM NH4HCO3 and in 25 mM NH4HCO3 theLTQafteraccumulationof104ions(maximumfillingtime,100ms). containing50%(v/v)acetonitrile.Gelpieceswerethendehydratedwith Forbioinformaticanalyses,datawereprocessedautomaticallyusing 100%acetonitrileandincubatedfor45minat53°Cwith10mMDTTin MascotDaemonsoftware(version2.3.2;MatrixScience).Concomitant 25mMNH4HCO3andfor35mininthedarkwith55mMiodoacetamide searchesagainsttheUniprotdatabase(August2012version,Rattustax- in25mMNH4HCO3.Alkylationwasstoppedbyadding10mMDTTin25 onomy,42,469sequences),theclassicalcontaminantsdatabase(67se- mM NH4HCO3 and mixing for 10 min. Gel pieces were then washed quences,homemade),andthecorrespondingreverseddatabaseswere againbyincubationin25mMNH4HCO3beforedehydrationwith100% performedusingMascot(version2.4).ESI-TRAPwaschosenasthein- acetonitrile.Atotalof0.15(cid:2)gofmodifiedtrypsin(sequencinggrade; strumentandtrypsin/Pastheenzyme,andtwomissedcleavagewere Promega)in25mMNH4HCO3wasaddedtothedehydratedgelpieces allowed.Precursorandfragmentmasserrortoleranceswereset,respec- foranovernightincubationat37°C.Peptideswerethenextractedfrom tively,at10ppmand0.6Da.Peptidemodificationsallowedduringthe gelpiecesinthree15minsequentialextractionstepsin30(cid:2)lof50% searchwerecarbamidomethyl(C,fixes)acetyl(N-ter,variable),oxida- acetonitrile;30(cid:2)lof5%formicacid;and,finally,30(cid:2)lof100%aceto- tion(M,variable),anddeamidation(NQ,variable).TheIRMasoftware nitrile.Thepooledsupernatantswerethendriedundervacuum. (version 1.30.4) was used to filter the results: conservation of rank 1 Fornano-LC-MS/MSanalyses,thedriedextractedpeptideswerere- peptides,peptideidentificationFalseDiscoveryRate(cid:6)1%(ascalculated suspendedin5%acetonitrileand0.1%trifluoroaceticacidandanalyzed byusingthereversedatabasestrategy),andminimumofonespecific by on-line nano-LC-MS/MS [Ultimate 3000 (Dionex) and LTQ- peptideperidentifiedproteingroup.Thefilteredresultswereuploaded OrbitrapVelospro(ThermoFisherScientific)].Peptidesweresampled intoarelationalmassspectrometryidentificationdatabase(MSIdb),and Chassefeyre,Martínez-Herna´ndezetal.•ESCRT-IIIFunctionatSynapses J.Neurosci.,February18,2015•35(7):3155–3173•3161 Figure3. Chmp2bregulatesdendriticmorphologyinanESCRT-III-dependentmanner.A,Hippocampalneuronswerecotransfectedat10DIVwiththeindicatedmixturesofpSuper-basedand rescue(pBI-based)plasmids.pSupervectorsexpressedmCherrytogetherwiththeindicatedshRNA[GFPshRNA(shGFP);Chmp2bshRNA(shChmp2borsh2b)].RescueplasmidsexpressedZsGreen, aloneortogetherwiththeindicatedversionofRNAi-resistantChmp2b(2B*).Imagesofdoublytransfectedneuronsat14DIVwereacquiredbyconfocalmicroscopy.Depictedisthemorphologyof representativeneuronsineachcondition,asvisualizedbymCherryfluorescence(negativecontrast,maximumintensityprojections).B,Shollanalysisofdendriticarborizationinneuronstransfected asinA.Foreachplasmidmix,thegraphshowsmeannumbersofdendriticbranchesasafunctionofradialdistance(n(cid:5)40neuronspercondition).Chmp2bknockdownstronglyreducedthe distance-dependentincreaseinbranchnumber(***p(cid:5)0.00014comparedwithcontrolcurve,Kolmogorov–SmirnovtestwithBonferroni’scorrection).Expressionofwild-typeChmp2b*restored thecontrolphenotype(p(cid:5)1).Chmp4b-binding-defective(Chmp2b*R19/22/26A)andmembrane-binding-defective(Chmp2b*L4D/F5D)mutantsofChmp2bwereunabletofullyrestorethecontrol phenotype(**p(cid:5)0.004andp(cid:5)0.006,respectively).C,Hippocampalneuronsweretriplycotransfectedat10DIVwithHA-taggedChmp2bWT,Flag-taggedChmp4b,andmCherryplasmids(1(cid:2)g ofeachplasmid).Thecellswerefixed24hlaterandstainedbydualimmunofluorescencewithanti-HAandanti-Flagantibodies.Confocalmicroscopyimages(z-projections)ofdendriticfluorescence areshownforarepresentativeneuron.Chmp2bwasrecruitedtoChmp4bpuncta,someofwhichwerefoundatdendriticbranchpoints(arrowheads)oratthe(Figurelegendcontinues.) 3162•J.Neurosci.,February18,2015•35(7):3155–3173 Chassefeyre,Martínez-Herna´ndezetal.•ESCRT-IIIFunctionatSynapses ahomemadetool(hEIDI;Milbradtetal.,2014)wasusedforthecompi- tolight.mCherryfluorescencewasexcitedwith75–150(cid:2)Wnominal lation,grouping,andcomparisonoftheproteingroupsfromthediffer- laser power at 561 nm. The pinhole diameter was set to 1 Airy unit. entsamples.hEIDIprovidesmodulestohelpexploringvalidatedsearch Z-stacksof20–30opticalsections(voxelsize,0.11(cid:2)0.11(cid:2)0.2(cid:2)m3) resultsthathavebeencompiledintoaMSIdb.Theusercreatescontexts, perdendriticsegmentwereacquiredat15minintervals.Signalperpixel whichrepresentsetsofidentifications,andrunspeptidesandproteins wasaveragedovertwoscans.Afteracquiringbaselineimages,anextra- groupingalgorithmstoclusterpeptidesidentifyingafamilyofproteinsin cellularsolutioncontaining200(cid:2)Mglycine(warmedto37°Cbyin-line thesecontexts.Thegroupingalgorithmwillfirstidentifyidenticalpep- heating)wasdelivered.Stimulationwasstoppedafter4minbyperfusing tidesbasedontheirmassandsequence.Groupsofproteinsmatchingthe awarmglycine-freeextracellularsolution.Onlymodestbleachingwas samesetofpeptidesarethencreatedasisdoneintheIRMatoolboxfora observedduringtheexperiment.Cellsshowingsignsofdamage(pearling single identification. This is typically useful to consider results at the orblebbing)werediscarded. samplelevelinsteadofthereplicatelevel.Tocompareseveralcontexts, Tomeasurechangesinspinevolume,imagestacksweresubjected hEIDIalignseachproteingroupfromeachcontexttobecompared.This to blind (adaptive) deconvolution with AutoQuant as above. The alignmentisbasedonproteinandpeptidesetsimilaritymetrics. deconvolvedimages(scaledto8bit)wereusedasinputfor3Dreconstruc- TobeconsideredasspecificallyassociatedwiththeChmp2bimmu- tion of dendritic morphology by the software package NeuronStudio noprecipitate,theproteinsidentifiedbythisanalysiswerefilteredac- (http://www.mssm.edu/cnic/tools.html),whichallocatesvoxelsto spines cordingtothefollowinginclusioncriteria:twoormoreuniquepeptides throughaspecificalgorithm(Rodriguezetal.,2006,2008).Individual (specificspectralcounts);twoormoredistinctproteinfragments[spec- spineslabeledbythesoftwareandthatwereclearlyidentifiableatboth tralcounts(SC)];andratioofSCinspecifictocontrolimmunoprecipi- initialandfinaltimepointsweremanuallyselected.Thevolumeofthe tate(SC2B/SCcontrol)(cid:8)5. spinesbeforeandafterstimulationwasestimatedbycountingspinevox- Electrophysiologicalrecordingsinculturedneurons.Mature[18–21din els.Thefractionalincreaseinspinevolumewascalculatedforeachspine vitro(DIV)]hippocampalneuronsexpressingthetransfectionmarker asfollows: ZsGreenwereusedforwhole-cellpatch-clamprecordingatroomtem- perature(20–24°C).Theneuronswerecontinuouslyperfusedwithan (cid:9)V V (cid:4)V 1ex0trmacMelHluElaPrEsSo,lu1t0iomnMcognlutacionsien,g01.540(cid:2)MmMtetNroadCol,to2xminM,5C0a(cid:2)CMl2,b3icmucMuKlliCnle, V (cid:3) fVi i, methiodide,and1(cid:2)Mstrychnine,adjustedtopH7.4withNaOHandto withV thevolumeat5minbeforethestartofthestimulationandV the 330mOsmwithNa-acetate.ToinduceLTP,glycine(200(cid:2)Minextracel- volumei at45minaftertheendofthestimulation.Themedianvaluefsof lularsolution)wasperfusedfor3minontotheneurons(Luetal.,2001). (cid:9)V/Vwerecalculatedforeachneuron.Differencesbetweenculturesdid For the experiments in which NMDA current amplitudes were mea- notsignificantlycontributetotheobservedeffects(two-wayANOVA, sured,NMDA(100(cid:2)M)wasaddedtothissolutionandperfusedwitha p(cid:5)0.015forfactorplasmid,p(cid:5)0.93forfactorculture,p(cid:5)0.56for fast-exchange perfusion system. Recording pipettes were pulled from interaction). Therefore, the results from seven independent cultures borosilicate glass and had a resistance of 3–5 (cid:7)M when filled with a werepooled. solution containing (in mM) 115 CsMeSO3, 20 CsCl, 10 HEPES, 0.6 Statisticalanalysis.Thestatisticalsignificanceofdifferencesbetween EGTA,4Na2-ATP,0.4Na-GTP,10Na-phosphocreatine,and2.5MgCl2, samplemeansorsampledistributionswascalculatedasdescribedinthe adjustedtopH7.2withCsOHandto300mOsmwithCs-acetate.Cur- figurelegends.Thenormalityofsampleswassystematicallycheckedby rentswererecordedfromaholdingpotentialof(cid:8)60mVusinganAxon theShapiro-Wilkstest,andnonparametrictestswereusedwhereappro- 200Bamplifier(MolecularDevices).Cellcapacitanceandseriesresis- priate.StatisticaltestswereperformedwithR(RCoreTeam,2013).Error tancecompensationwereappliedelectronically.Signalswerefilteredat1 barsindicateSEMinallfigures. kHz,digitizedat5kHz,andrecordedusingthepClampsoftware.Min- iatureEPSCs(mEPSCs)andNMDA-evokedcurrentswereanalyzedus- Results ingClampFit(MolecularDevices). ApoolofChmp2blocalizesindendriticshaftsandspines Liveimaging.Imageacquisitionandanalysiswereperformedblindto transfected plasmids. For live imaging during chemical LTP, neurons ToaddressthephysiologicalfunctionofChmp2binneurons,we weretransferredtoaperfusionchamber(Pocon),washed,andkeptinan firstevaluatedtherelativelevelofChmp2bproteininthebrain extracellularsolution(125mMNaCl,2.5mMKCl,2mMCaCl2,1mM comparedwiththatinotherorgans.Mousetissueextractswere MgCl2,5mMHEPES,33mMglucose,50(cid:2)Mbicuculline,and0.5(cid:2)MTTX, analyzedbyWesternimmunoblottingwithananti-Chmp2ban- 1(cid:2)Mstrychnine,pH7.4)at37°Conaheatedstage.Neuronsthatsimul- tibody.Thesameblotswerealsoprobedwithanantibodyraised taneouslyexpressedGFPandmCherry(withinthesamefluorescence againstthecoreESCRT-IIIsubunitChmp4b.Forbothproteins, rangethroughoutexperiments)andhadahealthyappearancewereiden- expression was strikingly higher in the brain than in the other tifiedbyepifluorescence,anddendriticsegmentswereselectedsothat organs, except for testes (Fig. 1A). Analysis of mouse brain at 30–100spinescouldbeimagedperneuronwhileminimizingexposure variousstagesofdevelopmentindicatedthatChmp2bexpression increasedduringthefirstpostnatalweek(Fig.1B).BothChmp2b 4 andChmp4bweredetectedinextractsfromvariousregionsof (Figurelegendcontinued.) baseofspine-lookingprotrusions(asterisk).Scalebar,10(cid:2)m.D, theadultbrain,withthehighestChmp2blevelsfoundinolfac- N2AcellsweretransfectedwiththeindicatedcombinationsofRNAi-inducingplasmidsand torybulbandcerebralcortex(Fig.1C). rescueplasmids.Thecellswerelysedandanalyzedbyimmunoblottingwithanti-Chmp2ban- To define the topographical pattern of Chmp2b protein tibody.E,HeLacellswerecotransfectedwithHA-taggedChmp2bWTorChmp2bR19/22/26Ato- expression, we performed immunohistochemistry on mouse getherwithChmp4b-Flag.Cellswerestainedbydualimmunofluorescencewithanti-HAand brainsections(Fig.1D).Chmp2bimmunoreactivitywaspar- anti-Flagantibodiesandimagedbyconfocalmicroscopy.Insets,Highmagnificationofthe ticularlystronginhabenula,olfactorybulb,deeplayersofthe boxedregions.Scalebars,20(cid:2)m.ColocalizationofChmp2bandChmp4bwas80%for neocortex, amygdala, and hippocampus. This pattern gener- Chmp2bWTand12%forChmp2bR19/22/26A(quantifiedasthepercentageofChmp2b-HA- ally coincided with in situ hybridization data from the Allen positivepixelsoverlappingChmp4b-Flagpuncta).F,HEK-293TcellsweretransfectedwithFlag- Brain Atlas (http://mouse.brain-map.org/gene/show/44784). taggedChmp4bplasmid,aloneortogetherwithHA-taggedChmp2bWTorChmp2bR19/22/26A. In higher-magnificationviews,immunostainingwasevidentin Anti-HAimmunoprecipitatespreparedfromthetransfectedcellswereanalyzedbyimmuno- thesomatodendriticcompartmentofnumerousneurons(e.g.,in blottingwithanti-Flagoranti-HAantibodies.G,HEK-293TcellsweretransfectedwithFlag- taggedChmp2bWTorChmp2bL4D/F5D.Thetransfectedcellswerehomogenized,themembrane theCA1regionofthehippocampus;Fig.1E).Specificityoflabel- (P)andcytosolic(S)fractionswereseparatedbycentrifugation,andequivalentamountsofeach ingwasconfirmedbyshowingreducedstaininginthebrainof fractionwereanalyzedbyimmunoblottingwithanti-Flagantibody. Chmp2b-deficient(KD)mice(Fig.1F). Chassefeyre,Martínez-Herna´ndezetal.•ESCRT-IIIFunctionatSynapses J.Neurosci.,February18,2015•35(7):3155–3173•3163 Figure4. Chmp2bisinvolvedinthemaintenanceofexcitatorysynapsesinculturedneurons.A,Hippocampalneuronsweretransfectedat16DIVwithpSuper-mCherryencodingcontrol(GFP) shRNA.Theneuronswerefixedat18DIV,stainedbydualimmunofluorescencewithanti-PSD-95andanti-synapsinIantibodies,anddendriteswereimagedbyconfocalmicroscopy.Postsynaptic spines(arrowheads)wereidentifiedasdendriticprotrusions(top)containingPSD-95puncta(middle).NearlyallofthespinescolocalizedwiththepresynapticmarkerSynapsin(bottom).mCherry fluorescencewasusedasamasktofacilitatevisualizationofpunctabelongingtothetransfectedneuron.Scalebar,15(cid:2)m.B,Neuronsweretransfectedat16DIVwiththesameplasmid combinationsasinFigure3A.Imagesofdendriticmorphologywereacquired,andpostsynapticspineswereidentifiedasinAandcounted.Arepresentativefieldisshownforeachcondition.Scale bar,5(cid:2)m.C,MeandensityofpostsynapticspinesalongthedendritesofneuronstransfectedasinB.Comparedwiththecontrol(GFP)shRNA,theChmp2b-depletingshRNAinducedalargedrop inspinedensity(***p(cid:5)2.4(cid:2)10(cid:8)11,two-sidedWilcoxontestwithHolmcorrection).Coexpressingwild-typeChmp2b*completelyrestoreddensity(p(cid:5)1vscontrol,***p(cid:5)6.3(cid:2)10(cid:8)14vs Chmp2bshRNAalone).CoexpressingtheChmp4bbinding-defectivemutantChmp2b*R19/22/26Adidnotrescuespinedensityatall(***p(cid:5)3.8(cid:2)10(cid:8)6vsChmp2b*WT,p(cid:5)0.14vsChmp2bshRNA alone);neitherdidthemembranebinding-defectivemutantChmp2b*L4D/F5D(***p(cid:5)2.9(cid:2)10(cid:8)7vsChmp2bWT,p(cid:5)1vsChmp2bshRNAalone).Neuronspercondition:control,n(cid:5)34;Chmp2b shRNA,n(cid:5)33;Chmp2b*WT,n(cid:5)30;Chmp2b*R19/22/26A,n(cid:5)16;Chmp2b*L4D/F5D,n(cid:5)14.D–F,Hippocampalneuronsweretransfectedat18DIVwiththeindicatedplasmids.Transfected neuronswereidentifiedbyfluorescenceat21DIV,andtheirminiatureEPSCsweremeasuredbypatch-clamprecordingsinwhole-cellconfiguration.D,Sampletracesfromneuronstransfectedas indicated.E,HistogramofmeanmEPSCamplitudesinneuronstransfectedasindicated(n(cid:5)5cells,(cid:10)100minispercellineachcondition).TheamplitudeofmEPSCswasmarkedlyreducedin Chmp2b-deficientneurons(**p(cid:5)0.005,two-sidedttest).F,HistogramofmeanmEPSCfrequenciesinneuronstransfectedasindicated.FrequencieswerelowerinChmp2b-deprivedneurons (*p(cid:5)0.012,two-sidedttest).G,Neuronstransfectedasabovewerestimulatedat21DIVbybath-appliedNMDA(control,n(cid:5)6cells;Chmp2bshRNA,n(cid:5)5cells).ThedensityofNMDA-evoked currents(calculatedascurrentnormalizedbycapacitance)wasstronglydecreasedinChmp2b-deprivedneurons(*p(cid:5)0.031,two-sidedttest). 3164•J.Neurosci.,February18,2015•35(7):3155–3173 Chassefeyre,Martínez-Herna´ndezetal.•ESCRT-IIIFunctionatSynapses Figure5. Chmp2bisinvolvedinthemaintenanceofexcitatorysynapsesinvivo.A,Brainsectionsfrom5-week-oldChmp2bknockdownmice(KD)orsiblingcontrols(Wildtype)werestainedby theGolgiimpregnationtechnique.Staineddendriticsegmentswerevisualizedbybright-fieldmicroscopyofthestratumradiatum.Shownarerepresentativeimagesofdendriticshaftsandspines inbothgenotypes.B,Spineswereclassifiedintothin,stubby,ormushroomtypes,andtheirdensityperunitdendriticlengthwasmeasuredforeachtype(n(cid:5)4animals,36dendritespergenotype). Left,HistogramshowingthemeandensityofeachspinetypeinKDandWTmice.ThetotalspinedensitywassignificantlylowerinKDmicethanincontrols(*p(cid:5)0.039,two-sidedttest).Right, Histogramshowingthemeanproportionofthethreespinetypes.ThepercentageofmushroomspineswassignificantlyreducedintheKDmice(*p(cid:5)0.02).C,Representativeelectronmicrographs oftheCA1stratumradiatumfromWtandKDsiblings.Presynapticprofilesarehighlightedinblue,andpostsynapticspines(s)areinbrown.D,Quantificationofsynapsedensityintheelectron micrographs.WT:n(cid:5)3animals,851axo-spinoussynapsesfrom30micrographs;KD:n(cid:5)3animals,627axo-spinoussynapsesfrom30micrographs.Thedensityofpostsynapticspinesperunit surfacewassignificantlylowerintheKD(***p(cid:5)0.0008). Chmp2bwasalsoreadilydetectableinprimaryculturesofrat ogy. To assess the effect of Chmp2b depletion on dendrite neurons. Consistent with the developmental profile shown in branching,neuronsweretransfectedat10DIVandimagedat14 Figure 1B, immunoblotting analysis of cultured hippocampal DIV by confocal microscopy. As shown in Figure 3, A and B, (Belly et al., 2010) or cortical (Fig. 2A) neurons showed that dendriticbranchingwasreducedinChmp2b-depletedneurons endogenousChmp2bexpressionsteadilyincreasedduringneuronal comparedwithneuronstransfectedwithcontrolshRNA.Coex- maturation,reachingaplateauafter9DIV,thusoverlappingthe pression of an RNAi-resistant version of the Chmp2b mRNA periodofsynapseformation.Asalreadyobservedinnon-neuronal togetherwiththeshRNAalmostfullyrescueddendriticcomplex- cells,immunofluorescentstainingofmaturehippocampalneurons ity.ThesedataindicatethatChmp2bisrequiredformaintaining revealed Chmp2b immunoreactivity throughout the entire cyto- theintegrityofdendritictrees. plasm.However,focusingondendrites, we noted that Chmp2b Innon-neuronalsettings,Chmp2proteinsareknowntoin- staining formed numerous small clusters, many of which were teractwithChmp4forformationofbiologicallyactiveESCRT-III found at the base, neck, and head of dendritic spines. Within (Morita et al., 2011). Consistent with such an interaction, we spine heads, Chmp2b clusters were often juxtaposed to PSDs observedthatwhencoexpressedinneuronsat10DIV,epitope- labeled with anti-PSD-95 antibody (Fig. 2B). The overlap of taggedChmp2bandChmp4bcolocalizedintopunctadistributed Chmp2b with PSD-95 puncta was significantly more frequent along the dendrites (Fig. 3C). Chmp2b/Chmp4b puncta were thanexpectedbyrandomsuperposition(Fig.2C).Thespecificity repeatedly found at dendritic branch points or at the base of of synaptic staining with the anti-Chmp2b antibody was con- dendritic protrusions. We then tested whether the effect of firmedbyextinctionwithawellvalidatedshRNAconstruct(Belly Chmp2bondendritebranchingdependedonitsabilitytobindto etal.,2010;Fig.2D). endogenous Chmp4b. For this, we designed a Chmp2b point mutant deficient for binding to Chmp4b. Morita et al. (2011) Chmp2bregulatesthestabilityofdendriticbranches showedthatbindingofChmp4btothecloseChmp2bhomolog Aftertheseobservations,weusedculturedhippocampalneurons Chmp2a depends on three arginine residues in the Chmp2a N toexaminewhetherChmp2bplayedaroleinthemaintenanceof terminus. Based on sequence alignment with Chmp2a, we re- dendritesandsynapses.Forthis,neuronsweretransfectedwith placedarginineresidues19,22,and26ofChmp2bbyalanine.We thesameChmp2b-depletingshRNAasabove.Controltransfec- demonstratedthat,unlikeChmp2bWT,Chmp2bR19/22/26Afailed tionswereperformedwithaGFP-specificshRNAwithnoeffect tocolocalizewithChmp4binthecellcytoplasmandtocoimmu- on neurons (Alvarez et al., 2006). The shRNA vectors also ex- noprecipitate with Chmp4b in cell lysates (Fig. 3D–F). Thus, pressedmCherry,allowingvisualizationofthecellularmorphol- Chmp2bR19/22/26AisunabletointeractwithChmp4b.

Description:
containing ESCRT-III complexes are also present at dendritic spines, where they regulate .. acetonitrile; 30 μl of 5% formic acid; and, finally, 30 μl of 100% aceto- nitrile. Ultrastructural localization of Chmp2b in dendritic spines.
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.