Wnt proteins regulate acetylcholine receptor clustering in muscle cells

MISSING IMAGE

Material Information

Title:
Wnt proteins regulate acetylcholine receptor clustering in muscle cells
Series Title:
Molecular Brain
Physical Description:
Mixed Material
Language:
English
Creator:
Zhang, Bin
Liang, Chuan
Bates, Ryan
Yin, Yimin
Xiong, Wen-Cheng
Mei, Lin
Publication Date:

Subjects

Subjects / Keywords:
Wnt
AChR clustering
muscle cells
synapse formation
neuromuscular junction

Notes

Abstract:
Background: The neuromuscular junction (NMJ) is a cholinergic synapse that rapidly conveys signals from motoneurons to muscle cells and exhibits a high degree of subcellular specialization characteristic of chemical synapses. NMJ formation requires agrin and its coreceptors LRP4 and MuSK. Increasing evidence indicates that Wnt signaling regulates NMJ formation in Drosophila, C. elegans and zebrafish. Results: In the study we systematically studied the effect of all 19 different Wnts in mammals on acetylcholine receptor (AChR) cluster formation. We identified five Wnts (Wnt9a, Wnt9b, Wnt10b, Wnt11, and Wnt16) that are able to stimulate AChR clustering, of which Wnt9a and Wnt11 are expressed abundantly in developing muscles. Using Wnt9a and Wnt11 as example, we demonstrated that Wnt induction of AChR clusters was dose-dependent and non-additive to that of agrin, suggesting that Wnts may act via similar pathways to induce AChR clusters. We provide evidence that Wnt9a and Wnt11 bind directly to the extracellular domain of MuSK, to induce MuSK dimerization and subsequent tyrosine phosphorylation of the kinase. In addition, Wnt-induced AChR clustering requires LRP4. Conclusions: These results identify Wnts as new players in AChR cluster formation, which act in a manner that requires both MuSK and LRP4, revealing a novel function of LRP4. Keywords: Wnt, AChR clustering, muscle cells, synapse formation, neuromuscular junction

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
Resource Identifier:
doi - 10.1186/1756-6606-5-7
System ID:
AA00010463:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EFBQR75RA_ESEQ27 INGEST_TIME 2012-04-04T14:18:49Z PACKAGE AA00010463_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

RESEARCH OpenAccessWntproteinsregulateacetylcholinereceptor clusteringinmusclecellsBinZhang1,ChuanLiang1,RyanBates1,YimingYin2,Wen-ChengXiong1andLinMei1*AbstractBackground: Theneuromuscularjunction(NMJ)isacholinergicsynapsethatrapidlyconveyssignalsfrom motoneuronstomusclecellsandexhibitsahighdegreeofsubcellularspecializationcharacteristicofchemical synapses.NMJformationrequiresagrinanditscoreceptorsLRP4andMuSK.IncreasingevidenceindicatesthatWnt signalingregulatesNMJformationinDrosophila,C.elegansandzebrafish. Results: Inthestudywesystematicallystudiedtheeffectofall19differentWntsinmammalsonacetylcholine receptor(AChR)clusterformation.WeidentifiedfiveWnts(Wnt9a,Wnt9b,Wnt10b,Wnt11,andWnt16)thatare abletostimulateAChRclustering,ofwhichWnt9aandWnt11areexpressedabundantlyindevelopingmuscles. UsingWnt9aandWnt11asexample,wedemonstratedthatWntinductionofAChRclusterswasdose-dependent andnon-additivetothatofagrin,suggestingthatWntsmayactviasimilarpathwaystoinduceAChRclusters.We provideevidencethatWnt9aandWnt11binddirectlytotheextracellulardomainofMuSK,toinduceMuSK dimerizationandsubsequenttyrosinephosphorylationofthekinase.Inaddition,Wnt-inducedAChRclustering requiresLRP4. Conclusions: TheseresultsidentifyWntsasnewplayersinAChRclusterformation,whichactinamannerthat requiresbothMuSKandLRP4,revealinganovelfunctionofLRP4. Keywords: Wnt,AChRclustering,musclecells,synapseformation,neuromuscularjunctionBackgroundTheneuromuscularjunction(NMJ)isacholinergic synapsethatexhibitsahighdegreeofsubcellularspecializationcharacteristicofchemicalsynapses[1,2].Itsformationisregulatedbyfactorsfrommotoneurons.For example,neuralagrinbindsLRP4,amemberofthe low-densitylipoproteinreceptor(LDLR)family,and subsequentlyactivatesthet yrosinekinaseMuSK[3-7], leadingtotheclusteringofAChRthroughmediator proteinsincludingcytoskeletalprotein a -actinin[2,8]. Interestingly,musclefiberprepatterningoraneural AChRclusterformationintheadvanceofinnervation requiresMuSKandLRP4butnotagrin,whereasnerveinducedAChRclustersrequireall[5,7,9].TheseobservationssuggestthatMuSKmayberegulatedbyagrinindependent,yetunidentifiedligand(s). Wntisafamilyofsecretedglycoproteinsthatplaya criticalroleindevelopment[10].Wntsignalsthrougha receptorcomplexconsistingofFrizzled(Fz)receptor andLRP5/6[11].Fzinteractstheadapterprotein dishevelled(Dvl)toactivateintracellularcanonicaland non-canonicalpathways.Recentstudiessuggestaroleof Wntinsynapseformation.InC.elegans,Wntsignaling determinesthepositionofNMJsbyinhibitingsynaptogenesis[12]whereasinDrosophila,Wntpromotesthe NMJformation[13,14].Ontheotherhand,synaptic activitymayalsoregulateW ntproteinexpression[15]. Intriguingly,theextracellulardomainofMuSKcontains acysteine-richdomain(CRD)thatishomologousto thatinFzforWntbinding[16,17].MuSKalsointeracts withDvl,whichregulatesagrin-inducedAChRclustering[18].MuSKinteractswithLRP4,acloserelativeof LRP5/6intheLDLRfamily[3,4,19].Inzebrafish, Wnt11rbindsto unplugged ,thezebrafishMuSKhomologue,toguidemotoraxons[20].Inmammalmuscle cells,agrin-inducedAChRclusteringwasenhancedby Wnt3,butreducedbyWnt3a[21,22]. *Correspondence:lmei@georgiahealth.edu1DepartmentofNeurologyandInstituteofMolecularMedicineandGenetics, GeorgiaHealthSciencesUniversity,Augusta,Georgia30912,USA FulllistofauthorinformationisavailableattheendofthearticleZhang etal MolecularBrain 2012, 5 :7 http://www.molecularbrain.com/content/5/1/7 2012Zhangetal;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommons AttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,andreproductionin anymedium,providedtheoriginalworkisproperlycited.

PAGE 2

Thereare19differentWntsinhumanandmice. WhetherandwhichWntissufficienttostimulateAChR clusteringintheabsenceofagrinremainsunknown. Here,westudiedtheeffectsof19WntsonAChRclusteringinmusclecellsandidentifiedfiveWnts(Wnt9a, Wnt9b,Wnt10b,Wnt11,andWnt16)thatareableto stimulateAChRclustering,independentofagrin. ExpressionanalysisindicatedthatWnt9aandWnt11 areabundantlyexpressedindevelopingmuscles.Using thesetwoWntsasexample,weinvestigatedmechanisms bywhichWntsregulateAChRclustering.ResultsindicatethatWntsplayanimportantroleinAChRclustering,likelybydirectbindingtoMuSKandinamanner dependentonLRP4.ResultsWntsinduceAChRclusteringinmusclecellsTosystematicallyinvest igateWntfunctioninAChR clusteringinmammaliancells,wetransfectedHEK293 cellswithplasmidsencoding19Wnts(eitherFlag-or HA-tagged)thathavebeenidentifiedinhumanand mice.Conditionedmediawerecollected48hafter transfection.Westernblottingbyanti-Flagand-HA antibodiesrecognizedtheexpressionofrespectiveWnts (datanotshown).ActivityofrecombinantWntswas verifiedbyluciferaseactivityinHEK293cellstransfected withTop-Flashreporter(datanotshown).Tostudythe effectofWntsonAChRclustering,C2C12myotubes werestimulatedwithconditionedmediacontaininga particularWnt.Ascontrol,theywerealsotreatedin parallelexperimentswithconditionedmediaofnontransfectedHEK293cellsoragrin.Sixteenhoursafter treatment,myotubeswerefixedandstainedwithR-BTX torevealAChRclusters[3].AsshowninFigure1A, agrinstimulationincreasesthenumberofAChRclusters inC2C12myotubes,asreportedpreviously[3,18].Intriguingly,5ofthe19Wnts(Wnt9a,Wnt9b,Wnt10b, Wnt11andWnt16)wereabletoinduceAChRclusters intheabsenceofagrin(Figure1A).Thiseffectappeared tobespecificasconditionedmediumfromnon-transfectedHEK293cellshadnoeffectonAChRclustering. TheseresultssuggestedthatWntproteinsaresufficient toalterAChRclusteringinculturedmusclecells.TreatmentofC2C12myotubesbyWnt9a,Wnt9b,Wnt10b, Wnt11orWnt16hadnoeffectonproteinlevelsof AChRorMuSKwithin16hofexperiments(datanot shown),suggestingthattheypromoteAChRclustering withoutincreasingthelevelsofAChRorMuSK proteins. TodeterminewhichWnt(s)maybeinvolvedinregulationofNMJformationinvivo,weexaminedmRNA levelsof19WntsindevelopingmusclesbyqRT-PCR. FourWntswereexpressedatlevelssignificantlyhigher thantherestofthegroups:Wnt2,Wnt4,Wnt9aand Wnt11(Figure1C).Inthisstudy,wewillfocuson Wnt9aandWnt11becausefirstthattheyaremost abundantWntsthatstimulateAChRclusteringaslevels ofWnt9b,Wnt10b,andWnt16werebarelydetectable. Second,theyareexpressedathigherlevelsatperiods mostrelevanttoNMJformation/maturation.Non-additiveeffectofWntandagrinonAChRclusteringWenextdeterminediftheinductionofAChRclusters byWnt9aandWnt11wasdose-dependent.RecombinantWnt9aandWnt11werepurifiedfromtherespectiveconditionedmediabyaffinitychromatographyusing Flag-M2AffinityGel.AsshowninFigure2A,treatment withincreasingconcentrationsofWnt9aandWnt11led toelevatednumbersofAChRclustersinmyotubes, indicatingthattheireffectsw ereconcentration-dependent.Theeffectsweresaturable,withEC50(halfmaximaleffectiveconcentration)valuesaround0.5nM, indicatingthatWntsactbyactivatinghigh-affinity receptors.ThemaximalresponseofWnts,however,was about50%ofthatforagrin,suggestingthatWntsmay notbeasefficientasagrinininitiatingpathwaysleading toAChRclustering. ToinvestigatemechanismsbywhichWntsstimulate AChRclustering,weinvestigatedpossiblefunctional interactionbetweenWntandagrin-initiatedpathways.If WntsandagrinstimulateAChRclustersbydifferent mechanisms,theireffectsshouldbeadditive-thesum ofagrin-andWnt-inducedresponses.Alternatively, theireffectcouldbesynergisticorinhibitory.Theseare importantquestionsbecaus eanswerscouldshedlight onpotentialmechanismsofWnt-inducedAChRclustering.Tothisend,myotubesweretreatedwithincreasing concentrationsofagrintogetherwithorwithoutWnt9a orWnt11at1nM,concentrationstoelicitmaximal response.Thedose-responsecurveofagrininthe absenceofWntssuperimposedwiththoseinthepresenceofWnt(Figure2B),indicatingnochangeofEC50orthemaximalresponseofagrinbyWnt.Theseresults didnotsupportadditive,synergistic,orantagonist effectsbetweenagrinandWnts.Instead,thesepharmacologicalstudiessuggestthatWntsinduceAChRclustersviasimilarmechanismsofagrin.DependenceofWnt-inducedAChRclusteringonMuSKMuSKiscriticalforagrin-inducedAChRclustering[5] aswellasaneuralAChRclustering(orprepatterning) priortothearrivalofmotoneuronaxonterminals [23,24].InZebrafish,Wnt11rwasshowntointeract withtheMuSKhomologue unplugged andinduceAChR prepattern[20].HavingdemonstratedthatWntsand agrinmayinduceAChRclustersviasimilarmechanisms, wedeterminedifWnt-inducedAChRclusteringrequires interactionwithMuSK.First,wedeterminedifWnt9aZhang etal MolecularBrain 2012, 5 :7 http://www.molecularbrain.com/content/5/1/7 Page2of8

PAGE 3

C 0 10000 20000 30000 P0 P15 P30 AdultRelative quantityWnt1 Wnt16 Wnt11 Wnt10b Wnt10a Wnt9b Wnt9a Wnt8b Wnt8a Wnt7b Wnt7a Wnt6 Wnt5b Wnt5a Wnt4 Wnt3a Wnt3 Wnt2b Wnt2BAChR clusters / mm 0 4 8 12 16 Wnt1 Wnt16 Wnt11 Wnt10b Wnt10a Wnt9b Wnt9a Wnt8b Wnt8a Wnt7b Wnt7a Wnt6 Wnt5b Wnt5a Wnt4 Wnt3a Wnt3 Wnt2b Wnt2 Control Agrin A Control Wnt3 Wnt3a Wnt9a Agrin Wnt11 Wnt2b Wnt2 Wnt5a Wnt5b Wnt7a Wnt8a Wnt8b Wnt10a Wnt9b Wnt10b Wnt16 Wnt1 Wnt4 Wnt6 Wnt7b Figure1 WntproteinsinduceAChRclustersinmusclecells A ,C2myotubeswerestimulatedwithconditionedmediacontainingWnt proteinsorcontrolmediumfor16h.AChRclusterswerevisualizedbyR-BTXstainingandindicatedbyarrows. B .Quantificationdataof A .AChR clustersgreaterthan4 minlengthwerequantified.*,p<0.01,Student sttest. C ,WntmRNAsareexpressedinmuscle.TotalRNAswere extractedfromskeletalmusclesofP0,P15,P30andadultmiceandusedforqRT-PCR.Relativeexpressionwasshowninhistograms. Zhang etal MolecularBrain 2012, 5 :7 http://www.molecularbrain.com/content/5/1/7 Page3of8

PAGE 4

andWnt11bindstoMuSKdirectly.Flag-taggedWnt9a andWnt11wereimmobilizedonbeadsandincubated withMuSKect-Myc,asecretedformofMuSK sextracellulardomainfusedwithMyc.AsshowninFigure3A, MuSKect-MycwasassociatedwithWnt9aandWnt11. Incontrast,ascontrol,Flag-Wnt7a,whichdidnotstimulateAChRclustering(Figure1),didnotinteractwith MuSKect-Myc(Figure3A).Theseresultsindicatedirect interactionbetweenthetwoWntsandMuSKandsuggestthatWntmayinduceAChRclusteringthrough interactionwithMuSK. Next,wedeterminedwhichdomaininMuSKisnecessaryforinteractionwithWnts.MuSKhasaCRDthatis homologoustotheCRDinFz [16,17],whichisknown tobindtoWnts[25].TheCRDinzebrafish unplugged isnecessaryforinteractionwithWnt11r[20].TodeterminewhethertheCRDofmammalianMuSKisalso necessaryforbindingtoWnts,wegeneratedect CRD, amutantMuSKectwithouttheCRD.AsshowninFigure3B,deletionoftheCRDsignificantlyattenuatedthe bindingactivity,indicatinganecessaryroleofthis domain.However,thebindingactivitywasnotabolished,suggestingpossibleinvolvementofotherdomains inbindingtoWnt.Asexpected,theCRDaloneissufficienttointeractwithWnt(Figure3B).Theseresults suggestthatWntinteractionbymammalianMuSK involvesCRDaswellasotherdomains. TodeterminewhetherMuSKisrequiredforWntinducedAChRclustering,weexaminedWntfunctionin MuSK-/-musclecells.Thesecellswerederivedfrom MuSK-/-miceandareunabletoformAChRclustersin responsetoagrin[26](Figure3C,non-transfected).We generatetwoexpressionco nstructsinpIRES-GFP: pMuSK-GFPandpMuSK CRD-GFP,whichexpress FlaggedMuSKandMuSK CRD,respectively,underthe controlofaCMVpromoterandGFPunderthecontrol ofanIRES(internalribosomeentrysite).Expressionof wild-typeMuSKinMuSK-/-myotubesenabledthemto formAChRclustersinresponsetoagrin.Inthese experiments,GFPwasco-expressedandAChRclusters werequantifiedonlyinGFP-positivemyotubes.Remarkably,Wnt9aandWnt11wereunabletoinduceAChR clustersinMuSK-/-myotubes(Figure3C),indicating therequirementofMuSKinWnt-inducedAChRclustering.Thisphenotypewasrescuedbyexpressionof wild-typeMuSKinMuSK-/-myotubes(Figure3C),in supportofacriticalroleofMuSK.Moreover,WntinducedAChRclustersweresignificantlyfewerin MuSK CRD-transfectedMuSK-/ -myotubes,compared tothoseexpressingwild-typeMuSK.Thedeletionof CRDhadnoeffectonagrin-inducedclusters,indicating thattherequirementoftheCRDwasspecificforWnt. TogethertheseobservationssuggestthatWntsinduce AChRclustersbydirectlyinteractingwithMuSK.WntinducesMuSKphosphorylationanddimerizationToinvestigatemechanismsbywhichWntsregulate MuSK,wefirstdeterminedifWntstimulationleadsto anincreaseinMuSKphosphorylationinmusclecells. C2C12myotubeswerestimulatedbyagrin,Wnt9aor Wnt11for1h.CellswerelysedandMuSKwasisolated byimmunoprecipitationandanalyzedfortyrosinephosphorylationwith4G10,atyros inephosphorylation-specificantibody[3].AsshowninFigure3D,treatmentof A B 0.001 0.01 0.1 1 10 0 5 10 15 20 Agrin Wnt9a Wnt11 Protein concentration (nM)AChR clusters / mm 0.001 0.01 0.1 1 10 0 5 10 15 20 Agrin Agrin+Wnt9a Agrin+Wnt11 Protein concentration (nM)AChR cluster / mm Figure2 CharacterizationofWnteffectonAChRclustering .Myotubeswerestimulatedwithincreasingconcentrationsofagrinalone(A)or togetherwith1nMWnt9aorWnt11(B).AChRclusterswereassayedasinFigure1. Zhang etal MolecularBrain 2012, 5 :7 http://www.molecularbrain.com/content/5/1/7 Page4of8

PAGE 5

Wnt9aorWnt11enhancedthelevelsofMuSKtyrosine phosphorylation. Incontrast,theph osphorylationwas notincreasedinmusclecellstreatedwithWnt7a,which didnotstimulatedAChRcl usters(Figure1A).The levelsofMuSKphosphorylationbythetwoWntswere apparentlylowerthanthatbyagrin,incorrelationwith theirefficaciesofAChRclu sterinduction,suggesting thatMuSKphosphorylationmaybeamechanismby whichWntsregulateAChRclusters. Inresponsetoagrinstimulation,MuSKisthoughtto formhomodimerstoactivatetheintracellularkinase domain[2].However,evidenceislackingthatagrin indeedincreasesMuSKdimerization.Toaddressthis issueandtodeterminewhetherWntsfacilitateMuSK dimerization,wetransfectedC2C12myoblastswithtwo vectors:Flag-MuSKandMuSK-Myc,whichexpress Flag-tagged(N-terminal)andMyc-tagged(C-terminal) MuSK,respectively.Resultingmyotubeswerestimulated A IP: Flag IB: Myc IB: Flag Input IB: Flag IB: Myc B IP: Flag Input Relative ratio Ig1-3 CRD MuSKectMyc FlagWnt 0 1 2 Ig1-3 CRD MuSKectMyc Ig1-3 CRD CNontransfected AgrinWnt9aWnt11Control MuSK 0 4 8 12 AgrinWnt9aWnt11Control Non-transfected MuSK MuSKCRD AChR clusters / mmMuSKCRD IP: MuSK IB: -actin DIB: 4G10 IB: MuSK EIP: Flag Lysates IB: Myc IB: Flag IB: Myc IB: Flag Relative ratio 0 1 2Control Agrin Wnt9a Wnt11 LysatesWnt7a Wnt9a Wnt11 Control Control ectCRD CRD ectIB: Myc IB: Flag IB: Flag IB: Myc Control ectCRD CRD ect Control Wnt11 Wnt9a Wnt7a Agrin Control Wnt11 Wnt9a Agrin Figure3 WnteffectonAChRclusteringrequiresinteractionwithMuSK A ,Wnt9aandWnt11,butnotWnt7a,bindMuSKinvitro.Flag-Wnts immobilizedbeadswereincubatedwithMuSKect-Myc.Boundproteinswereisolated,resolvedbySDS-PAGEandblottedwithantibodiesagainst MycandFlag. B ,Wnt11bindsCRDofMuSK.Flag-Wnt11immobilizedbeadswereincubatedwithMuSKect,MuSKect CRDorMuSKCRD. InteractionswereassayedasinA.*,p<0.05,n=3. C ,Wnt-inducedAChRclusteringisrescuedbyMuSK,butnotMuSK CRD,inMuSK-/-cells. MuSK-/-cellsweretransfectedwithoutorwithrespectiveconstructsandresultingmyotubes(identifiedbyGFPencodedbytheexpression construct)wereassayedforAChRclusterformationinresponsetoagrinorWntsasinFigure1.Arrows,AChRclusters.*,p<0.01,Student st test. D ,Wnt9aandWnt11induceMuSKtyrosinephosphorylation.C2C12myotubesweretreatedbyagrinorWntsfor1h.MuSKwaspurifiedby immunoprecipitationandprobedwiththe4G10antibody. E ,Wnt9aandWnt11induceMuSKdimerization.Flag-MuSKandMuSK-Mycwere transfectedintoC2C12myoblastsandresultingmyotubesweretreatedbyWnts,agrinorcontrolmediafor1h.Flag-MuSKwasprecipitatedby FlagantibodyandassociatedMuSK-Mycwasdeterminedbyanti-Mycantibody.BandintensityofimmunoblotwasanalyzedbyImageJsoftware. *,p<0.01,one-wayANOVAwithStudent sttest. Zhang etal MolecularBrain 2012, 5 :7 http://www.molecularbrain.com/content/5/1/7 Page5of8

PAGE 6

without(control)orwithagrin,Wnt9a,orWnt11.FlagMuSKwasprecipitatedfromthelysatesbyanti-Flag antibodyandprobedforMuSK-Mycwithanti-Mycantibody.AsshowninFigure3E,stimulationwithWntsas wellasagrinincreasedtheamountofMuSK-Myc,comparedtocontrol,suggestingdimerizationofMuSK.This resultisinsupportofthemodelthatWntsinteract withMuSKandcauseitsdimerizationandactivation.InvolvementofLRP4inWnt-inducedAChRclusteringLRP4interactsnotonlywithagrin,butalsowithMuSK, andiscriticalforagrin-inducedAChRclusteringand foraneuralAChRclusterformationandmusclefiber prepatterning[3,4,9].W eexaminedwhetherLRP4is involvedinWnt-inducedAC hRclusteringinmuscle cells.Tothisend,weculturedprimarymusclecells fromLRP4mittnullmice.Thismutationwasinducedby ENU,andanearlystopcodonattheC-terminaltothe LDLRtypeAdomainswasintroduced,whichshould deletemostoftheprotein[9].LRP4mittnullmicedonot formNMJs.AsshowninFigure4A,agrinstimulated AChRclustersinprimarymyotubesfromwild-type mice,butwasunabletodosoinmyotubesofLRP4mittlittermates,inagreementwit hpreviousreport[4,9]. Wnt9aandWnt11wereabletoelicitAChRclustersin wild-typeprimarymyotubes,demonstratingthatthis effectisnotlimitedtoclonalC2C12cells.Interestingly, however,theeffectwasabolishedinmyotubesof LRP4mittnullmice,indicatingthatWnt-inducedAChR clusteringrequiresLRP4,inadditiontoMuSK. Next,wedeterminedifWnt9aandWnt11bindto LRP4usingasimilarmethodtostudytheWnt-MuSK interaction.Flag-taggedWnt9aandWnt11wereimmobilizedonbeadsandincubatedwithLRP4N-Myc,a Myc-taggedrecombinantproteincontainingtheextracellulardomainofLRP4.AsshowninFigure4B, LRP4N-MycwasassociatedwithWnt9aandWnt11.In contrast,ascontrol,Flag-Wnt7a,whichdidnotstimulateAChRclustering(Figure1A),didnotinteractwith LRP4N-Myc.Theseresultsin dicatedirectinteraction betweenLRP4andWnt9aorWnt11.DiscussionInthestudywesystematicallystudiedtheeffectofall19 differentWntsonAChRclusterformationinmuscle cells.Wnt9a,Wnt9b,Wnt10b,Wnt11andWnt16were abletoinduceAChRclusteringindependentofagrin.In developingmuscles,Wnt9aandWnt11areexpressed relativelyabundantly.UsingWnt9aandWnt11asexample,wedemonstratedthatWntinductionofAChRclusterswasdose-dependentandnon-additivetothatof agrin,suggestingthatWntsmayactviasimilarpathways toinduceAChRclusters.Wnt9aandWnt11function requiresbothMuSKandLRP4,sinceWnt-induced AChRclusteringwasabolishedinMuSK-/-orLRP4 mutantmusclecells.TheseresultsidentifyWntsasnew playersinAChRclusterformation. WntregulationofNMJformationislikelytobecomplex.InadditiontopromotingAChRclusteringinthe absenceofagrin,Wnt3couldpotentiatewhereasWnt3a inhibitsagrin-inducedrece ptorclustering[21,22].It remainsunknownwhetherLRP4servesasreceptorfor Wntstoactivatecanonicalpathwaysandhowthese pathwaysinteractwithpathwaysthatareinitiatedbythe interactionofagrinandWntswiththeLRP4-MuSK complex.Finally,Wntsignalingisimplicatedinsynapse formationintheCNS.Forexample,Wnt-7afromgranulecellsinducesaxonandgrowthconeremodelingin mossyfibers[27].HowWntregulatesCNSsynaptogenesisremainsunclear.LRP4isexpressedabundantlyin thePSD(postsynapticdensity)fractionofthebrain[28]. ItwouldbeofinteresttodeterminewhetherWntregulationofCNSsynapseformationrequiresLRP4. B A IP: Flag Input IB: Myc IB: Flag IB: Myc IB: Flag 0 4 8 12 16 LRP4 AChR clusters / mmMitt Mitt Mitt Mitt Wt Wt Wt Wt Wnt9a Control Agrin Wnt11 Wnt9aWnt11 Mitt Wt AgrinControl Control Wnt11 Wnt9a Wnt7a Figure4 LRP4isrequiredforWnt-inducedAChRclusters A Wnt9aandWnt11failtostimulateAChRclusteringinLRP4mittmusclecells.Primarymusclecellswereculturedfromwtandmt miceandassayedforAChRclustersasinFigure1.Histogramshows quantitativeanalysisofAChRclusters. B ,Wntsbindtoectodomain ofLRP4.Flag-taggedWnt7a,Wnt9aorWnt11wereimmobilized withbeadsandincubatedwithLRP4N-Myc.Interactionswere determinedbyprecipitationandWesternblotusinganti-Myc antibody. Zhang etal MolecularBrain 2012, 5 :7 http://www.molecularbrain.com/content/5/1/7 Page6of8

PAGE 7

ConclusionsThisstudyprovidesevidencethatWnt9a,Wnt9b, Wnt10b,Wnt11andWnt16areabletostimulateAChR clustersinmusclecells.ThiseffectrequiresbothMuSK andLRP4.Wnt9andWnt11appeartoactbyinducing MuSKdimerization.TheseresultssuggestthatWnts mayplayaroleinNMJformation.MethodsReagentsandantibodiesTaqDNApolymerase,T4DNAligase,andrestriction enzymeswerepurchasedfromPromega.Horseradishperoxidaseconjugatedgoatanti-mouseandgoatanti-rabbit antibodiesandenhancedchemifluoresent(ECL)reagents forWesternblottingwerefromAmersham.Rhodamineconjugated a -bungarotoxin(R-BTX)wasfromInvitrogen. OligonucleotidesweresynthesizedbyOperonBiotechnologies.Unlessotherwisespecified,allchemicalswerefrom Sigma-Aldrich.AntibodieswerepurchasedfromSigma (FlagM2,F3165);UpstateBiotechnology(4G10,05-1050); Novus( b -actin,NB600-501).Rabbitanti-MuSKantibody wasdescribedpreviously[18].Rabbitanti-LRP4antibody wasdescribedpreviously[3].ConstructsAgrin,MuSKandLRP4constructsweredescribedpreviously[3,18].ExpressionconstructsofWnt1,Wnt4, Wnt6andWnt7bweregenerouslyprovidedbyDr.XiHe, whichwerefusedwithanHAtag.TogenerateFlag-Wnt constructs,WntcDNAs(ex ceptWntsdescribedabove) weregeneratedbyPCRandsubclonedintopFlag-CMV1 downstreamofanartificialsignalpeptidesequenceanda Flagepitope.TemplatesforPCRwerepurchasedfrom OpenBiosystems(catalognumberinparentheses):Wnt2 (4162686),Wnt2b(8734025),Wnt3(5726751),Wnt3a (40007188),Wnt5a(3487288),Wnt5b(6438917),Wnt7a (6415801),Wnt8a(40129440),Wnt8b(40056929),Wnt9a (30435371),Wnt9b(5588904),Wnt10a(4921327), Wnt10b(7868324),Wnt11(40129997)andWnt16 (40105502).Oligonucleotidesequencesforoverexpression constructsareomittedduetospacelimit,butareavailable uponrequest.TheauthenticityofallconstructswasverifiedbyDNAsequencing.CellcultureandtransfectionHEK293cellsandmouseC2C12musclecellswere maintainedasdescribedpre viously[3,29].Theywere transfectedwithPEI(polyethylenimine,Sigma,408727), asdescribedpreviously[30]withmodification.RecombinantproteinproductionandpurificationAgrinwasgeneratedandpreparedaspreviously described[18,31].ToproduceWntrecombinant proteins,HEK293cellsweretransfectedwithplasmids encodingFlag-andHA-taggedWnts.Twenty-four hoursaftertransfection,cellswereswitchedtoDulbecco sModifiedEagleMediumsupplementedwith0.05% offetalbovineserum,andconditionedmediawerecollectedandcleanedbycentrifugation(1000rpm,10min atroomtemperature).Controlconditionedmediumwas collectedfromnon-transfec tedHEK293cellsinparallel experiments.Flag-taggedWnt7a,Wnt9aandWnt11 werepurifiedbyaffinitychromatographyusinganti-Flag M2AffinityGel(Sigma,A2220)permanufacturer s instruction.AChRclusterassaysAChRclustersinC2C12myotubesweremeasuredas describedpreviouslywithmodification[3,29,31].QuantitativerealtimePCRTotalRNAofmousemusclewereextractedusingTrizol reagent(Invitrogen)followingmanufacturer sinstruction andtranscribedintocDNAtemplates.Theabundanceof WntmRNAwasdeterminedbyquantitativerealtime PCR(qRT-PCR)usingappropriateprimersandSQBR Greenasindicator[32].GADPHwasusedasinternal control.OligonucleotidesequencesusedforqRT-PCR wereomittedduetospacelimit,butareavailableupon request.SolutionbindingassaysandimmunoblottingThesolutionbindingassa ysandimmunoblottingwere performedaspreviouslydescribed[3].StatisticalanalysisDataofmultiplegroupswereanalyzedbyANOVA. Two-tailedStudent sttestwasusedtocomparedata betweentwogroups.Differenceswereconsideredsignificantat P <0.05.Valuesanderrorbarsinfiguresdenote meanSD.ListofabbreviationsNMJ:neuromuscularjunction;AChR:acetylcholine receptor;LDLR:low-densitylipoproteinreceptor;Fz: Frizzled;Dvl:dishevelled;CRD:cysteine-richdomain; IRES:internalribosomeentrysite;PSD:postsynaptic densityAcknowledgements TheauthorswishtothankDr.XiHeforWntconstructs.Wearegratefulto membersoftheMeiandXionglaboratoriesfordiscussion.Thisworkwas supportbygrantsfromNIH(L.M.andW.C.X)andMDA(L.M.). Authordetails1DepartmentofNeurologyandInstituteofMolecularMedicineandGenetics, GeorgiaHealthSciencesUniversity,Augusta,Georgia30912,USA.2Schoolof LifeSciences,PekingUniversity,Beijing100871,China.Zhang etal MolecularBrain 2012, 5 :7 http://www.molecularbrain.com/content/5/1/7 Page7of8

PAGE 8

Authors contributions BZ,WCX,andLMconceivedtheidea,designedexperiments,andwrotethe paper.BZexecutedexperimentsandanalyzeddata.CLassistedinAChR clustersassaysandYYassistedingeneratingWntconstructs.RBassistedin writingthepaper.Allauthorsreadandapprovedthemanuscript. Competinginterests Theauthorsdeclarethattheyhavenocompetinginterests. Received:4January2012Accepted:6February2012 Published:6February2012 References1.SanesJR,LichtmanJW: Developmentofthevetebrateneuromuscular junction. AnnualReviewofNeuroscience 1999, 22 :389-442. 2.WuH,XiongWC,MeiL: Tobuildasynapse:signalingpathwaysin neuromuscularjunctionassembly. Development 2010, 137 :1017-1033. 3.ZhangB,LuoS,WangQ,SuzukiT,XiongWC,MeiL: LRP4servesasa coreceptorofagrin. Neuron 2008, 60 :285-297. 4.KimN,StieglerAL,CameronTO,HallockPT,GomezAM,HuangJH, HubbardSR,DustinML,BurdenSJ: Lrp4isareceptorforAgrinandforms acomplexwithMuSK. Cell 2008, 135 :334-342. 5.DeChiaraTM,BowenDC,ValenzuelaDM,SimmonsMV,PoueymirouWT, ThomasS,KinetzE,ComptonDL,RojasE,ParkJS,SmithC,DiStefanoPS, GlassDJ,BurdenSJ,YancopoulosGD: ThereceptortyrosinekinaseMuSK isrequiredforneuromuscularjunctionformationinvivo. Cell 1996, 85 :501-512. 6.McMahanUJ: Theagrinhypothesis.[Review]. ColdSpringHarbSymp QuantBiol 1990, 55 :407-418. 7.GautamM,NoakesPG,MoscosoL,RuppF,SchellerRH,MerlieJP,SanesJR: Defectiveneuromuscularsynaptogenesisinagrin-deficientmutantmice. Cell 1996, 85 :525-535. 8.DobbinsGC,LuoS,YangZ,XiongWC,MeiL: alpha-Actinininteractswith rapsyninagrin-stimulatedAChRclustering. MolBrain 2008, 1 :18. 9.WeatherbeeSD,AndersonKV,NiswanderLA: LDL-receptor-relatedprotein 4iscrucialforformationoftheneuromuscularjunction. Development 2006, 133 :4993-5000. 10.vanAmerongenR,NusseR: TowardsanintegratedviewofWntsignaling indevelopment. Development 2009, 136 :3205-3214. 11.HeX,SemenovM,TamaiK,ZengX: LDLreceptor-relatedproteins5and 6inWnt/beta-cateninsignaling:arrowspointtheway. Development 2004, 131 :1663-1677. 12.KlassenMP,ShenK: Wntsignalingpositionsneuromuscularconnectivity byinhibitingsynapseformationinC.elegans. Cell 2007, 130 :704-716. 13.PackardM,KooES,GorczycaM,SharpeJ,CumberledgeS,BudnikV: The DrosophilaWnt,wingless,providesanessentialsignalforpre-and postsynapticdifferentiation. Cell 2002, 111 :319-330. 14.MathewD,AtamanB,ChenJ,ZhangY,CumberledgeS,BudnikV: Wingless signalingatsynapsesisthroughcleavageandnuclearimportof receptorDFrizzled2. Science 2005, 310 :1344-1347. 15.LiY,LiB,WanX,ZhangW,ZhongL,TangSJ: NMDAReceptorActivation StimulatesTranscription-IndependentRapidWnt5aProteinSynthesisvia theMAPKSignalingPathway. MolBrain 2012, 5 :1. 16.GlassDJ,BowenDC,StittTN,RadziejewskiC,BrunoJ,RyanTE,GiesDR, ShahS,MattssonK,BurdenSJ,DiStefanoPS,ValenzuelaDM,DeChiaraTM, YancopoulosGD: AgrinactsviaaMuSKreceptorcomplex. Cell 1996, 85 :513-523. 17.ValenzuelaDM,StittTN,DiStefanoPS,RojasE,MattssonK,ComptonDL, NunezL,ParkJS,StarkJL,GiesDR,ThomasS,LeBeauMM,FernaldAA, CopelandNG,JenkinsNA,BurdenSJ,GlassDJ,YancopoulosGD: Receptor tyrosinekinasespecificfortheskeletalmusclelineage:expressionin embryonicmuscle,attheneuromuscularjunction,andafterinjury. Neuron(USA) 1995, 15 :573-584. 18.LuoZ,WangQ,ZhouJ,WangJ,LiuM,HeX,Wynshaw-BorisA,XiongW, LuB,MeiL: RegulationofAChRClusteringbyDishevelledInteracting withMuSKandPAK1. Neuron 2002, 35 :489-505. 19.HerzJ,BockHH: LIPOPROTEINRECEPTORSINTHENERVOUSSYSTEM. AnnualReviewofBiochemistry 2002, 71 :405-434. 20.JingL,LefebvreJL,GordonLR,GranatoM: Wntsignalsorganizesynaptic prepatternandaxonguidancethroughthezebrafishunplugged/MuSK receptor. Neuron 2009, 61 :721-733. 21.HenriquezJP,WebbA,BenceM,BildsoeH,SahoresM,HughesSM, SalinasPC: WntsignalingpromotesAChRaggregationatthe neuromuscularsynapseincollaborationwithagrin. ProcNatlAcadSci USA 2008, 105 :18812-18817. 22.WangJ,RuanNJ,QianL,LeiWL,ChenF,LuoZG: Wnt/beta-catenin signalingsuppressesRapsynexpressionandinhibitsacetylcholine receptorclusteringattheneuromuscularjunction. JBiolChem 2008, 283 :21668-21675. 23.LinW,BurgessRW,DominguezB,PfaffSL,SanesJR,LeeKF: Distinctroles ofnerveandmuscleinpostsynapticdifferentiationofthe neuromuscularsynapse. Nature 2001, 410 :1057-1064. 24.YangX,ArberS,WilliamC,LiL,TanabeY,JessellTM,BirchmeierC, BurdenSJ: Patterningofmuscleacetylcholinereceptorgeneexpression intheabsenceofmotorinnervation. Neuron 2001, 30 :399-410. 25.BhanotP,BrinkM,SamosCH,HsiehJC,WangY,MackeJP,AndrewD, NathansJ,NusseR: AnewmemberofthefrizzledfamilyfromDrosophila functionsasaWinglessreceptor. Nature 1996, 382 :225-230. 26.ZhouH,GlassDJ,YancopoulosGD,SanesJR: DistinctdomainsofMuSK mediateitsabilitiestoinduceandtoassociatewithpostsynaptic specializations. JCellBiol 1999, 146 :1133-1146. 27.HallAC,LucasFR,SalinasPC: Axonalremodelingandsynaptic differentiationinthecerebellumisregulatedbyWNT-7asignaling. Cell 2000, 100 :525-535. 28.TianQB,SuzukiT,YamauchiT,SakagamiH,YoshimuraY,MiyazawaS, NakayamaK,SaitohF,ZhangJP,LuY,KondoH,EndoS: InteractionofLDL receptor-relatedprotein4(LRP4)withpostsynapticscaffoldproteinsvia itsC-terminalPDZdomain-bindingmotif,anditsregulationbyCa/ calmodulin-dependentproteinkinaseII. EurJNeurosci 2006, 23 :2864-2876. 29.LuoS,ZhangB,DongXP,TaoY,TingA,ZhouZ,MeixiongJ,LuoJ, ChiuFC,XiongWC,MeiL: HSP90betaregulatesrapsynturnoverand subsequentAChRclusterformationandmaintenance. Neuron 2008, 60 :97-110. 30.BoussifO,Lezoualc hF,ZantaMA,MergnyMD,SchermanD,DemeneixB, BehrJP: Aversatilevectorforgeneandoligonucleotidetransferinto cellsincultureandinvivo:polyethylenimine. ProcNatlAcadSciUSA 1995, 92 :7297-7301. 31.ZhangB,LuoS,DongXP,ZhangX,LiuC,LuoZ,XiongWC,MeiL: Betacateninregulatesacetylcholinereceptorclusteringinmusclecells throughinteractionwithrapsyn. JNeurosci 2007, 27 :3968-3973. 32.LiuX,BatesR,YinDM,ShenC,WangF,SuN,KirovSA,LuoY,WangJZ, XiongWC,MeiL: SpecificregulationofNRG1isoformexpressionby neuronalactivity. JNeurosci 2011, 31 :8491-8501.doi:10.1186/1756-6606-5-7 Citethisarticleas: Zhang etal .: Wntproteinsregulateacetylcholine receptorclusteringinmusclecells. MolecularBrain 2012 5 :7. Submit your next manuscript to BioMed Central and take full advantage of: Convenient online submission Thorough peer review No space constraints or color gure charges Immediate publication on acceptance Inclusion in PubMed, CAS, Scopus and Google Scholar Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Zhang etal MolecularBrain 2012, 5 :7 http://www.molecularbrain.com/content/5/1/7 Page8of8