Superoxide dismutase 1 encoding mutations linked to ALS adopts a spectrum of misfolded states

MISSING IMAGE

Material Information

Title:
Superoxide dismutase 1 encoding mutations linked to ALS adopts a spectrum of misfolded states
Series Title:
Molecular Neurodegeneration
Physical Description:
Mixed Material
Language:
English
Creator:
Prudencio, Mercedes
Borchelt, David R.
Publisher:
BioMed Central
Publication Date:

Notes

Abstract:
Background: Mutations in superoxide dismutase 1 (SOD1), which are one cause of familial amyotrophic lateral sclerosis (fALS), induce misfolding and aggregation of the protein. Misfolding can be detected by the binding of antibodies raised against peptide epitopes that are normally buried in the native conformation, shifts in solubility in non-ionic detergents, and the formation of macromolecular inclusions. In the present study, we investigate the relationship between detergent-insoluble and sedimentable forms of mutant SOD1, forms of mutant SOD1 with aberrantly accessible epitopes, and mutant protein in inclusions with the goal of defining the spectrum of misfolded states that mutant SOD1 can adopt. Results: Using combined approaches in cultured cell models, we demonstrate that a substantial fraction of mutant SOD1 adopts a non-native conformation that remains soluble and freely mobile. We also show that mutant SOD1 can produce multimeric assemblies of which some are insoluble in detergent and large enough to sediment by ultracentrifugation and some are large enough to detect visually. Three conformationally restricted antibodies were found to be useful in discriminating mal-folded forms of mutant SOD1. An antibody termed C4F6 displays properties consistent with recognition of soluble, freely mobile, mal-folded mutant SOD1. An antibody termed SEDI, which recognizes C-terminal residues, detects larger inclusion structures as well as soluble misfolded entities. An antibody termed hSOD1, which recognizes aa 24-36, detects an epitope shared by soluble non-natively folded WT and mutant SOD1. This epitope becomes inaccessible in aggregates of mutant SOD1. Conclusions: Our studies demonstrate how different methods of detecting misfolding and aggregation of mutant SOD1 reveal different forms of aberrantly folded protein. Immunological and biochemical methods can be used in combination to detect soluble and insoluble misfolded forms of mutant SOD1. Our findings support the view that mutant SOD1 can adopt multiple misfolded conformations with the potential that different structural variants mediate different aspects of fALS.
General Note:
Publication of this article was funded in part by the University of Florida Open-Access publishing Fund. In addition, requestors receiving funding through the UFOAP project are expected to submit a post-review, final draft of the article to UF's institutional repository, IR@UF, (www.uflib.ufl.edu/ufir) at the time of funding. The Institutional Repository at the University of Florida (IR@UF) is the digital archive for the intellectual output of the University of Florida community, with research, news, outreach and educational materials

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
Resource Identifier:
doi - 10.1186-1750-1326-6-77
System ID:
AA00008944:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Superoxidedismutase1encodingmutations linkedtoALSadoptsaspectrumofmisfolded statesPrudencioandBorchelt PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77(17November2011)

PAGE 2

RESEARCHARTICLE OpenAccessSuperoxidedismutase1encodingmutations linkedtoALSadoptsaspectrumofmisfolded statesMercedesPrudencio1,2andDavidRBorchelt1*AbstractBackground: Mutationsinsuperoxidedismutase1(SOD1),whichareonecauseoffamilialamyotrophiclateral sclerosis(fALS),inducemisfoldingandaggregationoftheprotein.Misfoldingcanbedetectedbythebindingof antibodiesraisedagainstpeptideepitopesthatarenormallyburiedinthenativeconformation,shiftsinsolubility innon-ionicdetergents,andtheformationofmacromolecularinclusions.Inthepresentstudy,weinvestigatethe relationshipbetweendetergent-insolubleandsedimentableformsofmutantSOD1,formsofmutantSOD1with aberrantlyaccessibleepitopes,andmutantproteinininclusionswiththegoalofdefiningthespectrumof misfoldedstatesthatmutantSOD1canadopt. Results: Usingcombinedapproachesinculturedcellmodels,wedemonstratethatasubstantialfractionofmutant SOD1adoptsanon-nativeconformationthatremainssolubleandfreelymobile.WealsoshowthatmutantSOD1 canproducemultimericassembliesofwhichsomeareinsolubleindetergentandlargeenoughtosedimentby ultracentrifugationandsomearelargeenoughtodetectvisually.Threeconformationallyrestrictedantibodieswere foundtobeusefulindiscriminatingmal-foldedformsofmutantSOD1.AnantibodytermedC4F6displays propertiesconsistentwithrecognitionofsoluble,freelymobile,mal-foldedmutantSOD1.Anantibodytermed SEDI,whichrecognizesC-terminalresidues,detectslargerinclusionstructuresaswellassolublemisfoldedentities. AnantibodytermedhSOD1,whichrecognizesaa24-36,detectsanepitopesharedbysolublenon-nativelyfolded WTandmutantSOD1.ThisepitopebecomesinaccessibleinaggregatesofmutantSOD1. Conclusions: Ourstudiesdemonstratehowdifferentmethodsofdetectingmisfoldingandaggregationofmutant SOD1revealdifferentformsofaberrantlyfoldedprotein.Immunologicalandbiochemicalmethodscanbeusedin combinationtodetectsolubleandinsolublemisfoldedformsofmutantSOD1.Ourfindingssupporttheviewthat mutantSOD1canadoptmultiplemisfoldedconformationswiththepotentialthatdifferentstructuralvariants mediatedifferentaspectsoffALS.BackgroundOneconsequenceoffALSassociatedmutationsin SOD1[Swiss-Prot:P00441]thatseemstobesharedby allmutantsisthatthemutantSOD1isfarmoreprone toadoptaberrantconformationsthatresultinitsaggregation[1].Thiscommonpropertyiseasilyassessedin culturedcellmodelsinwhichmutantSOD1isoverexpressed.However,pathologicevidenceofmutantSOD1 aggregationindiseasehasalsobeenconsistently demonstratedinhumanfALSpatientsharboringmutationsinSOD1andtransgenicmousemodelsofthisdisease.Inbothcases,therearemultiplereportsofthe detectionofSOD1immunoreact iveinclusionsinspinal motorneurons[2-13].Forexample,intransgenicmouse modelsofSOD1-associatedALS,SOD1antibodyreactiveinclusionshavebeendetectedinspinalmotorneuronsofmiceexpressingtheH46R[14],G85R[4,10], andG93A[7,8,10,13]mutants;andinspinalastrocytes ofmiceexpressingtheG85R[4]mutant.However,there havebeenreportsofpoordetectionofSOD1-immunoreactiveinclusionsinspinalmotorneuronsofG37R, G85R,G93A,H46R/H48QandQuadSOD1transgenic *Correspondence:borchelt@mbi.ufl.edu1DepartmentofNeuroscience,McKnightBrainInstitute,UniversityofFlorida, Gainesville,FL32610,USA FulllistofauthorinformationisavailableattheendofthearticlePrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 2011PrudencioandBorchelt;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedtheoriginalworkisproperlycited.

PAGE 3

mice[10,15,16].Moreover,inmiceexpressingtheG37R andG93Amutantsatlevelshighenoughtocause paralysisin4to6months,themoreobviouspathology identifiedbySOD1antibodiesisavacuolarpathology [15,17].Ingeneral,intheaforementionedstudiesin whichSOD1-immunoreactiveinclusionshavebeen detected,theyhavegenerallybeenfoundtobemost abundantinend-stagemice. MisfoldedmutantSOD1canalsobedetectedbiochemically,usinganassaythatinvolvesdetergent extractionandcentrifugatio nfollowedbywesternblotting[1,18-21].IntransgenicmousemodelsofSOD1fALS,detergentinsolubleformsofmutantSOD1accumulatetohighlevelslateinthecourseofdisease, becomingreadilydetectableattheonsetofvisiblesymptoms[22].IntheG93A-Gur1model,whichreaches end-stageparalysisatabout120days,aggregatesbegin toaccumulatebetween90and105daysandthenrise dramaticallyastheanimalsapproachendstage[22]. DuringthissameintervalthelevelsofneurofilamentH inserum,whichservesasabiomarkerofaxonaldegeneration[23],alsorisedramatically.ThuswecancorrelateaggregationofmutantSOD1andaxonal degenerationintheanimalmodels.However,itremains difficulttoassigncauseandeffectasitispossiblethat theaccumulationofaggregatesservesasabiomarkerof cellulardegenerationinwhichthecellssimplylosethe abilitytopreventthemisfoldedmutantproteinfrom aggregatingbecauseofsomecombinationofdeclining chaperoneactivities,decliningproteasomeactivityto degradethemisfoldedprotein,ordecliningenergyproductiontosupporttheseactivities.Moreover,inthe beststudiedG93Amodel,itisveryclearthatnumerous pathologicabnormalitiescanbedetectedpriortothe accumulationofaggregatedmutantSOD1,indicating thatothermisfoldedformsofmutantSOD1mustmediateearlyeventsintheevolutionofdisease[22]. OnemethodthathasemergedasameanstospecificallydetectmisfoldedformsofmutantSOD1involves SOD1antibodiesraisedagainstspecificpeptide sequencesthatarenormallyburiedinthetertiarystructureofnativeWTproteinoragainstpreparationsof mutantSOD1[17,24,25].Someoftheseantibodieshave thepotentialtodetectavarietyofstructuralvariantsof SOD1,includingimmatureunfoldedprotein,misfolded solubleprotein,andinclus ionstructures.Inastudyof theG93Amousemodel,Rakhitetal[17]describedan antibodytosequencesintheC-terminusofSOD1, calledtheSEDIantibody,whichshowedminimalreactivitywithnativelyfoldedWTSOD1.Ascomparedto tissuesfrommiceexpressingWTSOD1,tissuesfrom symptomaticmiceexpressingmutantSOD1(G37R, G85R,andG93A)containedhigherlevelsofSOD1proteinsthatreactedwiththeS EDIantibody.Intissue sectionsfromtheseanimals,theSEDIantibodystained SOD1,accumulatingattheri msofvacuolarpathology intheG37RandG93Amice(morerarelyinclusionlike structureswerestained)andstructuresresembling inclusionsintheG85Rmice(vacuolarpathologylimited inthismodel).InpresymptomaticG37RandG93A models,SEDIimmunoreactiv itywasabundantatthe rimsofvacuolarpathologythatisamongthefirstidentifiablepathologicabnormality,occurringwellbeforethe onsetofsymptoms[17].Collectively,thesestudieswith conformationally-restrictedantibodiesindicatethatmisfoldedSOD1proteinsarepresentearlyintheevolution ofdisease. Inthepresentstudy,wehavesoughttousecellculturemodelstoinvestigatethecharacteristicsofthevariousformsofmisfoldedmutantSOD1proteinsandto investigatetherelationshipsbetweenmisfoldedSOD1 proteins,detergentinsolubility,andinclusions.To accomplishthisgoal,wecombinethevisualcapabilityof theYFPfusionproteinswithbiochemicalassaysofinsolubilityandwithimmunohistochemicalapproaches usingconformationallyrestrictedantibodies.OurfindingsprovidenewinsightintothespectrumconsequencesthatfALSmutationshaveonthefoldingof SOD1andhowdifferentapproachestodetectioncanbe usedtorevealaspectsofmutantSOD1misfolding.ResultsVisualizationofmisfoldedandaggregatedmutantSOD1 inculturedcellmodelsTheA4VmutationinSOD1isassociatedwithrapidly progressingdiseaseandwhenexpressedinHEK293FT cellsisamongthemostpronetoproducedetergentinsolubleaggregatesofmutantprotein[1].Todate, however,wehavenotsoughttovisualizewhetherthe accumulationofdetergent-insolubleaggregatesis accompaniedbyinclusionformationinthesecellmodels.Followingpreviouslyestablishedprotocols[1,20,21], HEK293FTcellsweretransientlytransfectedwith expressionvectorsforWTandA4VmutantSOD1,for 24hours,thenfixedandanalyzedbyimmunofluorescencestainingwiththehSOD1antibody,whichwas raisedagainstasyntheticpeptidecorrespondingtoresidues24-36andwhichspecificallyrecognizeshuman SOD1protein(hSOD1antibody)[4].Thisantibodycannotimmunoprecipitatenative lyfolded,fullymetallated, WTSOD1proteinandisthusconformationally restricted(seeSupplementalMaterialin[26]).Comparedtountransfectedcells(Figures1Aand1B),cells expressingeitherWT(Figures1Cand1D)orA4V SOD1(Figures1Eand1F)showeddiffuseimmunofluorescencewithoutobviouscytoplasmicinclusions.Within nucleiofcellstransfectedwitheitherWTormutant SOD1,therewerepunctatestructuresthatappearedtoPrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page2of18

PAGE 4

besomeconcentrationofSOD1inoraroundnuclear structures.Thesignificanceofthesestructuresisuncertainatpresent. Intheseexperiments,theabsenceofimmunoreactivity bythehSOD1antibodyintheuntransfectedhuman HEK293FTcellsmaybeduepartiallytorelativelevels ofprotein;inthetransfectedcellstheexpressedSOD1 proteinismuchmoreabundantandthusatexposures optimalforvisualizationo ftheexpressedprotein, untransfectedcellswerevirtuallyinvisible.Additionally, weexpecttheendogenousproteintobemorefully metallatedandthuslessreactivetothehSOD1antibody [26].Therefore,weprimarilyvisualizedtheoverexpressedSOD1proteinsintheseexperiments. Indetergentextractionofcellstransfectedinparallel, weobserved,aspreviouslyreported[1,16],anaccumulationofdetergent-insolublemutantprotein(Additional File1,FigureS1).However,onlyaminorityofthetotal A4VSOD1proteinthatwasexpressedinthesecellswas insolubleindetergent(AdditionalFile1,FigureS1). Thus,weexpectedthateithermostcellswouldshow diffusestainingwithonl yaminorityhavingenough expressiontoinduceinclusionformation,orthatthere mightbesmallvisibleinclusionswithinageneralbackgroundofdiffuselydistributedmutantprotein.The absenceofinclusiondetectioncouldalsobeduetoinaccessibilityoftheantibodytoitstargetepitope.Asnoted above,thehSOD1antibodycannotimmunoprecipitate nativelyfoldedWTSOD1,butshowsstrongreactivity todenaturedprotein(seeSupplementalMaterialin [26]).Therefore,weaskedwhethertreatingthecells with70%formicacidtodenatureproteinswouldreveal hSOD1immunoreactiveinclusions(AdditionalFile1, FigureS2).However,westillfailedtodetectinclusions. Todeterminewhetherwemighthavebetterdetection ofinclusionsinmousecells,usingthehumanSOD1 specifichSOD1antibody,wetransfectedmousefibroblastcultureswiththesameconstructsandimmunostainedwiththesameantibody(AdditionalFile1,Figure S3).However,weobservedthesamediffuseimmunostainingpatternthatwas observedintheHEK293FT cells. WenextexaminedHEK293FTcellsexpressingWT andmutantA4VSOD1proteinsbyimmunostaining cellswithadditionalSOD1 antibodiesthathavebeen describedasconformationally-restrictedinbinding SOD1protein.OneantibodyusedwastheSEDISOD1 antibody,whichrecognizesaminoacids143to151in thedimerinterfaceofSOD1;theseresiduesarenormallyburiedwithinthenati veprotein[17].Another antibodyusedwastheC4F6antibody,whichwasraised againstthehumanG93AvariantofSOD1andwhich hasbeendescribedtohaveverylowimmunoreactivity tonativeWTSOD1infavorofmutantSOD1[27].For WTSOD1expressingcells,SEDIimmunoreactivitywas generallymuchlower,althoughafewcellswerefound tobehighlyreactive(Figures2Aand2B).Similarly,cells expressingWTSOD1showedlimitedreactivitywiththe C4F6antibody(Figures2Cand2D).Incellsexpressing A4VhSOD1,immunoreactivitytoSEDIlargelyappeared asuniformfluorescence;similartothefewimmunoreactiveWTexpressingcells(Figures2Eand2F).Rarely,we wereabletofindwhatappearedtobeinclusionstructures(Figure2Eand2F,inset).Immunostainingofcells expressingA4VhSOD1withtheC4F6antibodygave imagessimilartothoseobservedwithSEDIantibody (Figures2Gand2H),infrequentlyfindingcellsthatcontainedintracellularinclusion-likestructures(Figures2G and2H,inset).Overall,regardlessofwhichantibody wasused,immunoreactiveinclusionstructureswererare incellsexpressinguntaggedA4VhSOD1.Asnoted above,themajorityofSOD1inculturesexpressingA4V SOD1wassolubleindetergent(seeAdditionalFile1, Figure1 HEK293FTcellsexpressingmutantSOD1proteinsdo notformvisiblecellularinclusions .A-I)Cellswereculturedon glasscoverslipspreviouslycoatedwith0.5mg/mlpoly-L-lysine untransfected(A,B),ortransfectedwithexpressionplasmidsforWT (C,D)orA4V(E,F)SOD1for24hours.Fixedcellswerestainedwith hSOD1antibodyovernight.Asecondaryfluorescent(594nm) antibodywasusedtovisualizeSOD1antibodybinding.Co-staining with4 ,6-diamidino-2-phenylindole(DAPI)wasperformedtogether withsecondaryantibodyincubations.Picturesweretakenwitha 100immersionoilobjective,scalebar=20 m.Notethatatthe sameexposure,thestainingofoverexpressedhSOD1proteinis muchbrighterthanthatofuntransfectedcells(comparefigureAto CandE).Theimagesshownarerepresentativeofatleast3 independenttransfectionexperiments. PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page3of18

PAGE 5

FigureS1).Thus,thesecellsaccumulateaformof mutantSOD1thatissolubleinnon-ionicdetergents andyetisreactivetoantibodiesthatbindepitopesnormallyburiedinthenativeconformationofWTSOD1.Relationshipbetweeninclusionsanddetergent-insoluble SOD1Tomoreeasilyvisualizewhichcellsexpressmutant SOD1andwhichcellsdevelopinclusions,wegenerated variantsofSOD1taggedwithyellowfluorescentprotein (YFP).Multiplelaboratorieshaveusedsimilar approachestovisualizeinclusionsformedbymutant SOD1inlivingcellsusingfluorescentproteinssuchas greenfluorescentprotein(GFP),YFP,redfluorescent protein(RFP),orDsRed2invariousculturedcellmodels [28-35].ThuswecreatedseveralSOD1::YFPconstructs toexpressinculturedcellsasameanstoanalyzethe relationshipsbetweendetergent-insolubleaggregates thatcanbedetectedbiochemicallyandvisiblecellular inclusions.Weusedtwocellmodelsforthesestudies; thepreviouslydescribedHEK293FTcellmodel[1,20,21] andChineseHamsterOvarycells(CHOcells).Transient transfectionsofWTSOD1taggedwithYFP(WT::YFP) inHEK293FTfor24hoursshoweduniformdistributionsoffluorescencethroug houtthecytosol(Figure 3A).SOD1::YFPfusionproteinsgeneratedfromthe A4VandG37RvariantsofSOD1(A4V::YFPandG37R:: YFP)frequentlyproducedcytosolicinclusion-likestructures(Figure3A).Notably,notallcellsproducedinclusionsandthereweremanyexamplesofcellswith diffuselydistributedYFPfluorescence.Thisoutcome maybeexplainedbythenon-homogenousexpressionof transientlytransfectedcells;wepresume,butcannot definitivelyprove,thatonlythecellswithveryhigh levelsofexpressiondevelopinclusions.Theappearance ofinclusionswasaccompaniedbytheaccumulationof significantamountsofdetergent-insolublemutantfusion proteins(Figure3B).Notably,cellsexpressingWT::YFP SOD1alsoaccumulatedsignificantamountsofdetergent-insolublefusionprot ein.However,therelative aggregationpropensityofWT::YFPSOD1proteinwas muchlowerthanthatofthemutants(Figure3C).We observedasimilarlyhighfrequencyofinclusionsincells expressingotherfALSmutantsofSOD1andYFP (G85R::YFP,D101N::YFP,S134N::YFP),withsimilar accumulationsofdetergentinsolublefusionprotein (AdditionalFile1,FigureS4). Toexaminetherelationshipbetweeninclusionformationandtheformationofdetergent-insolublestructures, wecomparedthelevelsofinsolubleuntaggedSOD1and SOD1::YFPfusionproteinsthataccumulateinourtransfectedcells(Figure4).Ascomparedtocellsexpressing WTSOD1,cellsexpressingWT::YFPSOD1fusions accumulatedsignificantlyh igherlevelsofinsoluble SOD1at24(Figures4Aand4B)and48(Figures4C and4D)hoursaftertransfection.However,despitethe accumulationofsuchlargeamountsofdetergent-insolubleSOD1wedidnotobserveinclusionstructuresin cellsexpressingWT::YFPfusions(Figures4Eand4F). Bycontrast,incellsexpressingA4V::YFPfusions,inclusionstructureswerevisiblein24hours,becomingmore abundantat48hours(Figures4Gand4H).Relativeto WT::YFPfusions,thelevelofinsolubleprotein(ratioof insolubletosoluble)incellsexpressingA4V::YFP fusionswassubstantiallyhigheratboth24and48 hours.Still,theamountofdetergent-insolubleWT::YFP Figure2 SEDIandC4F6antibodiesrecognizediffuselydistributedcytosolicmutantSOD1intransientlytransfectedHEK293FTcells CellsweretransientlytransfectedwithexpressionplasmidsforWT(A-D)orA4V(E-H)SOD1for24hoursandstainedwithSEDIorC4F6 antibodiesasexplainedinMethods.Picturesweretakenwitha40objective.Scalebars=20 m.Inset,imageofrarecellwithinclusionlike structures.Theimagesshownarerepresentativeofatleast3independenttransfectionexperiments. PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page4of18

PAGE 6

fusionproteinthataccumulatedat48hourswassimilar tocellsexpressingA4V::YFP fusionfor24hours(Figures4A-D)butonlythelatterwasabletoformnumerousvisibleinclusions(Figures4Gand4H).Weinterpret thisoutcomeasevidencethattheWT::YFPSOD1 fusionproteinsarecapableofformingdetergent-insolublestructuresthataredistinctfrominsolubleinclusions.ConformationallyrestrictedantibodiestoSOD1Withtheabilitytoeasilyvisualizewhichcellsexpress mutantSOD::YFPfusionproteinsandwhichcellsproduceinclusions,wesoughttodeterminetheefficacy withwhichdifferentconformationallyrestrictedSOD1 antibodiesdetectinclusions.Forthisstudy,weusedthe samethreeantibodiesdescribedabove(hSOD1,SEDI andC4F6).WT::YFPSOD1expressingcellsimmunostainedwithhSOD1antibod yshoweduniformcytosolic distribution(Figures5A-C),indicatingaportionofthe expressedproteindidnotachieveanativeconformation. IncellsexpressingA4V::YFPproteins,weobservedthe formationofinclusionscontainingA4V::YFPSOD1, howeverthehSOD1antibodyappearedtorecognize onlydiffuselydispersedA4V::YFPprotein(Figures5DF).AlthoughhSOD1didnotappeartodirectlydetect inclusionstructures,wecouldidentifyseveralfeaturesin whichitappearedthathSOD1antibodyrecognizedproteinthatappearedtobeconcentratedaroundthesurfaceoftheinclusion(Figure5E,arrows).Thus,we concludethatthesequenc erecognizedbythehSOD1 antibody(residues24-26)isinaccessibleinmutant SOD1thatiscompartmentalizedininclusionstructures. TheSEDIandC4F6antibodiesalsorecognizeddiffuselydistributedSOD::YFPproteinsinthecytosol, including,moreinfrequently,WT::YFPfusionproteins Figure3 MutantSOD::YFPproteinsformvisibleinclusionsincellculture .A)HEK293FTcells,onpoly-L-lysinecoatedglasscoverslips,were transfectedfor24hoursasdescribedinMethods.Cellswerefixedandobservedunderafluorescencemicroscope.Allpicturesweretakenusing a40xobjective,scalebars=50 m.Theimagesshownarerepresentativeofatleast3independenttransfectionsforeachconstruct.B) ImmunoblotofP2andS1proteinfractionsofcellsexpressingSOD1::YFPproteinsfor24hours.TheSOD1::YFPproteinmigratesatasizeof approximately50kDa(filledarrowhead),whileendogenousWTSOD1monomerrunsat16kDa(openarrowhead).C)Quantificationofthe aggregationpropensity(P2/S1)ofcellsexpressingSOD1::YFPproteins(seeMethods).Asexpected,cellstransfectedwithWT::YFPormutant SOD1::YFPaccumulatedmuchmoreinsolubleproteinthanuntransfectedcells:*p 0.05;#p 0.005.TheaggregationpropensityofA4V::YFP andG37R::YFPwerebothstatisticallygreaterthanthatofWT::YFP(#p 0.005).Thedatawereaveragedfromatleast3independentrepetitions ofeachtransfectionandimmunoblot. PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page5of18

PAGE 7

(Figures5G-I,and5M-O).IncellsexpressingA4V::YFP, theSEDIantibodywashighly reactivewithinclusionlikestructures(Figures5J-L).TheC4F6antibody showedmoderatereactivity withdiffuselydistributed WT::YFPfusionproteininasmallsubsetofcells(Figures5M-O),butshowedstrongreactivitytodiffusely distributedA4V::YFPfusionprotein(Figures5P-R).We didnotobserveimmunostainingofinclusionswiththe C4F6antibody,thustheepitoperecognizedbyC4F6 antibodyisnormallyburiedwhenthemutantSOD1:: YFPproteinsforminclusionstructureswhereastheepitopeoftheSEDIantibodyisexposed.MutantSOD1inclusionsanddetergent-insolubleSOD1 arenotfreelymobileincytosolApreviousstudybyUrushitanietalexaminedmouse N2acellsexpressingmutantSOD1-EGFPfusionproteins,findingthatdigitoninreleasedmostofthe expressedproteinbutincellsexpressingmutantSOD1EGFP,punctateinclusion-likestructuresremainedcell associated[34].Weapplie dthesametechniquetoour cellmodels.Similartodigitonin,saponinisamild detergentthatopenporesonthecellularmembranes withoutcompletelylysingthecells,allowingsolubleproteinstodiffuseoutofthecell[36].Inthisexperiment, CHOcellswereusedbecausethesecellswereless pronetodislodgeaftersaponintreatment.CellstransfectedwithavectorforA4V::YFPfusionproteinsproducedcharacteristicpunctateinclusionsthatwerenot releasedintoculturemediumbysaponin(Figures6A-D, and6M-P).Bycontrast,WT::YFPfusionproteins showedamorediffuselocalization(Figures6E-H,and 6Q,R)andthisfluorescencewascompletelyreleasedby saponintreatment(Figures6Gand6H).NotethatasignificantfractionofWT::YFPSOD1,shownaboveinFigure3,isinsolubleinnon-ionicdetergentandyet virtuallyallofthefluorescenceisreleasedfromthese cellsbysaponin.IncellsexpressingYFPalone,fluorescencewassimilarlydiffuselydistributedwithsaponin releasingmostofthefluorescenceintothemedium (Figures6I-L,and6S,T).However,somelocalizationof theYFPtonucleiwasobservedandthisnuclearprotein wasnotreleasedbysaponin(Figures6K,L,T).These findingsindicatethatthegeneraldistributionofinclusionstructuresdoesnotobviouslychangewithsaponin treatment,suggestingthestructuresareboundtoimmobileelementswithinthecytosol[36]. BasedontheimmobilityofmutantSOD1::YFPinclusions,weonceagainrevisitedthedetectionofmutant SOD1aggregatesgeneratedbyuntaggedmutantprotein intransfectedHEK293FTcellsbyaskingwhethertreatmentofcellswithsaponinwillunmaskinclusionsin cellsexpressinguntaggedA4VSOD1.Saponintreatmentefficientlyextractednearlyallofthecytosolic SOD1immunoreactivityincellsexpressingWTSOD1 (Figures7Aand7B).TheonlyremaininghSOD1immunoreactiveproteinwasconcentratedin,orpossibly around,thenucleus(Figures7Aand7B).Thislatter Figure4 Analysisofdetergentsolubilityandinclusionformation incellsco-expressingWTandmutantSOD1 .HEK293FTwere transfectedwiththeindicatedSOD1constructsfor24(A,B,E,G)or48 (C,D,F,H)hoursasnotedabovethefigure.A,C)ImmunoblotsofP2 andS1fractionsofHEK293FTcellsco-transfectedwiththeindicated SOD1constructsfor24(A)or48(C)hoursandimmunoblottedwitha SOD1antibody.TaggedSOD1::YFPproteinsmigrateonSDS-PAGEat relativemolecularmassof50kDa,whileuntaggedSOD1variants migrateatabout16kDa.B,D)Quantificationoftheaggregation propensities(P2/S1)oftaggedanduntaggedproteinsfromsinglyor doublytransfectedcellsat24(B)and48(D)hours.Statistical comparisonswerecalculatedthroughunpairedstudentt-tests,*p 0.05,#p 0.005.Thedatawereaveragedfromatleast3independent transfectionandimmunoblottingexperiments.E-H)Cellstransfected for24(E,G)or48hours(F,H)werefixedandvisualizedundera confocalmicroscope.Scalebars=20 m.Theimagesshownare representativeofatleast3independenttransfectionexperiments. PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page6of18

PAGE 8

Figure5 ImmunostainingofcellsexpressingSOD1::YFPfusionproteinswithconformationallyrestrictedSOD1antibodies .HEK293FT weretransfectedwithWT::YFPorA4V::YFPconstructs,asnotedontheFigure,andstainedwithhSOD1(A-F),SEDI(G-L)orC4F6(M-R)antibodies asexplainedinMethods.Picturesweretakenusinga60waterimmersionobjectiveofaspinningdiscconfocalmicroscope,scalebars=10 m.ArrowsindicatethelocationofSOD1::YFPinclusionsandtheabsenceofantibodystaining.Theimagesshownarerepresentativeofatleast 3independenttransfectionexperiments. PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page7of18

PAGE 9

findingisindicativeofthesmallsizeofSOD1,which wouldfacilitatepassivediffusionintothenucleus[37], andisconsistentwithpriorobservationsthatSOD1is foundinnuclearcompartments[38].Despiteremoving mostofthecytoplasmicimmunoreactivity,itwasstill difficulttoidentifycellscontainingobviousinclusionsin cellsexpressingA4VhSOD1(Figures7Cand7D).Very rarelyweobservedsmallpunctatestructuresinaperinuclearlocation(Figures7Cand7D,arrowheads). Wealsoinvestigatedtheexpressionofmutanthuman SOD1inmouseTK-Lcells,inwhichwecoulduseantibodiesspecificforhumanSOD1topotentiallymorespecificallyvisualizeinclusionsformedbymutanthuman SOD1.However,aftertreatmentwithdigitonin,we observedimmunoreactivestructuresincellstransfected mutanthSOD1thatweresimilartowhatweobservedin cellsexpressingWTSOD1(AdditionalFile1,FigureS5). ToverifythataggregatedformsofuntaggedSOD1 werenotreleasedbysaponintreatment,cellstreated withsaponinweresubjectedtodetergentextractionand centrifugationassaybeforeimmunoblottingsolubleand insolublefractions(Figure8),followingpreviously describedprocedures[1,20,21].Asexpected,saponin treatmentinducedthereleaseofasignificantamountof bothWTandmutantSOD1proteinintothecellculture media(Figures8Aand8B).InthecaseofcellstransfectedwithWTSOD1,asubstantialportionofSOD1 proteinwasreleasedintothemediawiththefraction remainingwiththecellexhibitingnearcompletesolubilityindetergent(Figures8Aand8B).Aminorfraction oftheWTSOD1wasinsolubleindetergentandthere wasnosignificantchangeinthesolubilityofWTSOD1 aftersaponintreatment(Figure8B).Incellsexpressing A4VSOD1,thelevelsofSOD1releasedbysaponin Figure6 InclusionsformedbymutantSOD1::YFPproteinsarenotreleasedbysaponintreatment .CHOcellsweretransientlytransfected withexpressionplasmidsforYFP,WT::YFP,andA4V::YFPasnotedinthefigureandthen,after24hours,treatedwithsaponin,fixedin4% paraformaldehyde,stainedwithDAPI,andimaged.A,B,E,F,I,J,M,N,Q,R,andSwerenottreatedwithsaponin.C,D,G,H,K,L,O,P,andT weretreatedwithsaponin.ScalebarsinA-Lare50 m.TheimagesinM-TaredigitallymagnifiedcellsfromA-L.Becausesaponincompletely extractedallcyotosolicfluorescencefromcellexpressingWT::YFP,nodigitalmagnificationsofthesecellsareshown.Theimagesshownare representativeofatleast3independenttransfectionexperiments.Scalebar=50 m.Imageswerecaptureonastandardfluorescence microscope. PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page8of18

PAGE 10

wereverysimilartothatofcellsexpressingWTSOD1 (Figures8Aand8B).Aswasobservedincellsexpressing WTSOD1,therewerestillrelativelyhighlevelsof detergent-soluble,cell-associated,A4VSOD1aftersaponintreatment(Figure8B) .Thisdetergent-soluble mutantproteinmaybeproteinthatdiffusedintothe nucleus.Importantly,incellsexpressingtheA4V mutant,saponintreatmentdidnotsignificantlyreduce theamountofdetergent-insolublemutantproteinassociatedwiththecells(Figure8B).Similaroutcomeswere observedincellstreatedwithdigitonin(AdditionalFile 1,FigureS6),whichsimilar tosaponin,permeabilizes cellplasmamembraneswithoutcompletelylysingcells. Thesefindingsindicatethatthemajorityoftheinsoluble mutantSOD1incellsisimmobilizedbyassociationwith somecytosolicstructure. Basedonthefindingsabove,wewereconfidentthata substantialportionofmutantSOD1aggregatesarenot extractedbysaponintreatment.Therefore,weapplied thesetechniquestoHEK293FTcellstransfectedwith expressionvectorsforuntaggedSOD1.Twenty-four hourspost-transfection,saponin-treatedcellswerefixed andimmunostainedwiththeSEDIandC4F6antibodies (Figure9).Neitherantibodydetectedcytosolicreactivity incellsexpressingWThSOD1(Figures9A-D),indicatingthatthecytosolicimmunoreactivitydetectedwith thisantibody(seeFigures1and2)isreleasedintothe mediumbysaponin.IncellsexpressingA4VhSOD1, thediffusestainingseen inuntreatedcellswithboth antibodies(seeFigure2)wasalsolost.Insteadwewere nowabletoclearlyidentifycellswithpunctatecytosolic structuresthatwerereactivetoSEDIantibody(Figures 9E-Fand9I-L).Bycontrast,theC4F6antibodydidnot detectthesepunctatestructures(Figures9Gand9H). Thesefindingswereconsistentwithwhatwasobserved incellsexpressingSOD1::YFPfusionproteinsandindicatethattheSEDIantibodyisusefulindetecting mutantSOD1ininclusionstructures.MutantSOD1aggregationandtoxicityPreviousstudiesbyMatsumotoetal[31],inaPC12cell modeloftransienttransfecti on,havedemonstratedthat cellsharboringinclusionsformedbyYFPfusionsof G85RandG93ASOD1becomepermeabletopropidium iodide(areporterofcelldeath).Wesoughttodetermine whetherourcellmodelsofaggregationshowevidenceof toxicity,usingtworeporterssystemsofcelldeath.We initiallyusedtheCHOcellsbecausewefoundthesecells tobemoresuitedtoviewingindividualcells;theywere lesspronetoclumping,exhibitedaflattermorphology, andshowedbetteradherence(meaningfewerwerelost duringvariousstepsinpreparation).TheYFPtaggedvariantsofmutantSOD1facilit atedirectvisualizationof whethercellsthatharborag gregatesareundergoing degeneration.Toassesswhethercellsexpressingmutant orWT::YFPSOD1fusionproteinsmightbedying,we usedacelldeathassaybasedonuptakeofahomodimer ofethidiumbromide.Cellsthathavelostmembrane integritytakeuptheethidiumhomodimerandfluoresce reduponbindingoftheethidiumtonucleicacids.We observedonlylowlevelsofcelldeathincellsexpressing A4V::YFPSOD1fusionprot einswithsimilarlevels occurringincellsexpressingWT::YFPSOD1fusionproteinsanduntransfectedcells(Figure10).Onlyrarelydid weobservecellsharboringA4V::YFPSOD1inclusions showinguptakeofethidiumhomodimer.Importantly,in thesestudies,cellsshowingdiffusedistributionsof mutantorWT::YFPSOD1fusionproteinsalsolacked evidenceofobviousincreasesinuptakeofthecelldeath marker.Asimilaroutcomewasobtainedinthe HEK293FTcellmodel(AdditionalFile1,FigureS7). Wealsoassessedwhethercellswithinclusionshowed markersofapoptosis,usingaTUNELassay.Forthis experimentweusedonlytheCHOcellmodelbecause theHEK293FTcellswereso poorlyadherentthrough processingstepsthatwecouldnotmakeunbiased observations.At48hourspost-transfection,wefound thatrelativelyfewCHOcellswerepositiveandonly rarelydidweobserveTUNELlabelingofcellsthat expressedA4V::YFPSOD1fu sionproteins(Additional File1,FigureS8).AsimilarlowfrequencyofTUNEL positivecellswasseenin cellsexpressingWT::YFP SOD1fusionproteinsanduntransfectedcells(AdditionalFile1,FigureS8).Overall,weobservedno Figure7 SaponintreatmentextractscytosolicSOD1 immunostaining .HEK293FTcellsweretransfectedwithvectorsfor WT(A,B)orA4VSOD1(C,D)andafter24hourstreatedwith saponinfor30minutesataconcentrationof0.01%in1PBS.The cellswerethenfixedandimmunostainedasexplainedinMethods. Imageswerecapturedonastandardfluorescencemicroscopewith animmersionoil100objective.Allbarsrepresent20 m.The imagesshownarerepresentativeofatleast3independent transfectionexperiments. PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page9of18

PAGE 11

obviousoverlapbetweencellsexpressingWTorA4V SOD1::YFPfusionproteins (whetherinclusionswere presentorabsent)andeithermarkerofapoptosisor death.Weconcludethatthesecellsshowlimitedtoxicityinresponsetotheexpressionofthefusionproteins.DiscussionInthepresentstudy,weuseacombinationofnatural SOD1mutantsandSOD1::YFPfusionproteinswith conformationally-restrictedantibodiestoexaminethe relationshipsbetweenmutantSOD1misfolding,aggregation,andinclusionbodyformation.Ourfindingsare consistentwithascenarioinwhichmutantSOD1is capableofadoptingacontinuumofstructuresthat includesolublenon-nativelyfoldedproteins,detergentinsolublemultimericcomplexes,andlargerinclusion bodies.WeobservethatmutantSOD1canadoptconformationsthatremainsolu bleinaqueousbuffersand arefreelymobileandyetarerecognizedbyconformationally-restrictedantibodiessuchastheSEDIandC4F6 antibodies.Aspreviouslydescribed,weobservethat mutantSOD1adoptsaconformationthatpromotesits assemblyintolargerstructuresthatarenotsolublein non-ionicdetergentsandsedimentbycentrifugation.In Figure8 Detergent-insolubleformsofmutantSOD1remaincellassociatedaftersaponintreatment .A)ImmunoblotsofHEK293FTcells thatweretransfectedwiththeindicatedSOD1constructsfor48hours.Priortoharvest,cellswereincubated30minutesinPBSwith0.01% saponin.Theharvestedcellpelletswereanalyzedbydetergentextractionandcentrifugationassays(S1fractions,eachlanecontains5 gof totalprotein;P2fractions,eachlanecontains20 goftotalprotein).TheamountofSOD1proteinreleasedintothecellmediawasalso evaluated(eachlaneloadedwith5 goftotalprotein).B)QuantificationofSOD1proteinfoundintheMedia,P2andS1fractionsofcells untreatedortreatedwithsaponin(Sap.),andexpressingWTorA4V.Symbolsoverbarsindicatedifferencesfromcorrespondingnon-treated control.Unpaired t -tests:*p 0.05.Thedatawereaveragedfromatleast3independenttransfectionandimmunoblottingexperiments. PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page10of18

PAGE 12

cellsexpressinguntaggedversionsofmutantSOD1, usingconformationallyrestr ictedantibodies,wefound thattheaccumulationofdetergent-insolubleprotein correlatedwiththeappearanceofsmallpunctuatestructuresthatcouldbeviewedassmallmicro-aggregates. ThesestructureswererecognizablebytheSEDIantibody,butnotbytheC4F6antibody.Moreover,although theaccumulationofdetergent-insolublemutantprotein correlatedwiththeappearanceofinclusionlikestructuresformutantSOD1,wefoundanexampleinwhich detergent-insolubilityandinclusionformationwasdissociated.LargeamountsofdetergentinsolubleWT::YFP wedetectedincellsexpressingthisSODfusionprotein butinclusionstructureswererarelyvisualized.Notably, thisdetergent-insolubleproteinwas,however,freely mobileandcouldbereleasedbysaponin.Thisfinding indicatesthatSOD1indetergent-insolubleconformationsmaynotnecessarilybeassembledintolargeinclusionstructuresandcanremainfreelymobile.Thus,we demonstratethatmutantSOD1hasthecapacitytoproduceacontinuumofstructureswitheachexhibitingfeaturesthatareselectivelyrevealeddependinguponthe detectionmethodsused.SOD1conformationandantibodyepitopeaccessibilityInthisreport,weusedthreepreviouslydescribedantibodiesthateachshowrestrictedreactivitytonative conformationsofSOD1.ForthehSOD1andSEDIantibodies,previousstudieshav edemonstratedthatthese antibodiesarenotcapableofimmunoprecipitating nativelyfoldedWTSOD1[17,26].TheC4F6antibody hasbeenshowntoreactpoorlywithnativelyfoldedWT SOD1inimmunoblotsofnon-denaturingSDS-PAGE [27].ThehSOD1antibody,whichwasraisedagainst specificpeptidesequencesofhumanSOD1(residues 24-36),recognizedsoluble, freelymobile,formsofboth WTandmutantSOD1.Inworkthatwillbepublished elsewhere,wehavefoundthatthevastmajorityofWT ormutantSOD1thatisover-expressedinthesecells doesnotcontainCu(H.LelieandDBorchelt,unpublishedobservation).Wet hussuggestthatthelackof boundCuinthenormalactivesiteproducesaconformationalchangethatexposesthehSOD1epitope(aa 24-36).However,astheproteinorganizesintoinclusions,thisepitopebecomesinaccessible,indicatingthat thisportionoftheproteinmustbeburiedwithinvery stablestructuresintheinclusions. AnantibodytermedC4F6,whichwasraisedagainst theG93AvariantofhumanSOD1,didnotrecognize inclusionstructuresformedbyeitheruntaggedorYFPtaggedproteins.TheepitopefortheC4F6antibodyis complex.Underdenaturingc onditions,thisantibody behavesasiftheepitopeincludesthespecificmutation ofGly93toAla[27].Ondenaturingimmunoblots, Figure9 TheSEDIantibodyrecognizesimmobilepunctatestructuresincellsexpressingA4VhSOD1 .HEK293FTcellsweretransiently transfectedwithexpressionplasmidsforeitherWT(A-D)orA4V(E-L)hSOD1for24hoursandstainedwiththeSEDIorC4F6antibodyas explainedinMethods.Picturesweretakenwitha40xobjective.Scalebars=10 m.TheimagesinI-LweredigitallymagnifiedfromA4VSOD1 cellsstainedwithSEDI/DAPI.Theimagesshownarerepresentativeofatleast3independenttransfectionexperiments. PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page11of18

PAGE 13

C4F6showshighspecificitytotheG93Avariantwithno reactivitytoWTorotherFALSassociatedmutationsat Gly93[27].However,undernon-denaturingconditions, theantibodybehavesasiftheepitopeincludes conformationalelementsastheantibodycanrecognize othermutantformsofSOD1,albeitmuchlessrobustly [24].TheantibodyshowsnoreactivitytonativeWT SOD1[27].Thus,althoughitislikelythatthe Figure10 CHOcellsexpressingA4V::YFPSOD1fusionproteindonotshowevidenceofincreasedmembranepermeability .Toassess whetheraggregatesofmutantSOD::YFPfusionproteinsinducecelldeath,CHOcellsweretransientlytransfectedwithexpressionvectorsfor SOD1A4V::YFP(AandB)andWT::YFP(C).After48hours,thecellswereincubatedwithdimericethidiumbromide(EtBr)asdescribedin Methods.Imageswerecapturedat40magnification(scalebar=50 m).Overall,therewerefewcellsthatshowedevidenceofincrease membranepermeabilitywithnoobviousdifferencebetweenculturestransfectedwiththe2constructsoruntransfectedcells(D).Asapositive control,untransfectedcellsweretreatedwithstaurosporine(+Str)toinducedeathandstainedwithdimericethidiumbromide(D). PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page12of18

PAGE 14

conformationalelementsrecognizedbythisantibodyare inproximitytotheprimaryepitopethatincludesGly 93,withinexon4,theepitopehasnotbeenprecisely defined. OurfindingsindicatethatasmutantSOD1assembles intoinclusions,thereisaconformationalshiftthat eithermasksordestroystheC4F6epitope.Wenote, however,thatOkamatoetal[39]reportedthatinclusionsinthespinalcordsofG93AmicewereimmunoreactivewiththeC4F6antibody.Itispossiblethat reactivitywithinclusionsintheG93Amiceisuniqueto thismodelbecausethenon-conformationalelementsof theepitopeincludetheG93Amutation[27]. Inourcellculturemodels,theSEDIandC4F6antibodiesshowedthebestdiscriminationbetweenWTand mutantSOD1inregardtothediffusecytosolicstaining. AlthoughtherewererarecellsexpressingWTSOD1 thatshowedreactivity,thefrequencywithwhichmutant cellswereimmunostainedwasmuchhigher.Thesignificanceoftheinfrequent,butrobust,stainingofcells expressingWTSOD1withtheseantibodiesisuncertain. BoscoandcolleaguesrecentlyreportedthatC4F6will reactwithoxidativelymodifiedWTSOD1[27].Thus,it ispossiblethatsometypeofoxidativeprocessinasubsetofcellstriggersthemisfoldingofWTSOD1.Itis alsopossiblethatWTSOD1canspontaneousmisfold inrarecasesofextremeover-expressionwiththemisfoldedconformationpropagatingthroughoutthecell. IncellsexpressingtheuntaggedA4VSOD1,boththe C4F6andSEDIantibodiesrobu stlydetectedadiffusely distributedformofmutantSOD1thatwasreadily releasedfromcellsbysaponin.WenotethatZetterstromandcolleagueshavedemonstratedthatspinal cordtissuesfrommiceexpressingbothWTandmutant humanSOD1containvariedamounts(higherinmutant mice)ofsolubleSOD1thataberrantlybindshydrophobicmediaandlacksnormal intramoleculardisulfide bonds[40].Thus,therearemultiplelinesofevidenceto indicatesolublemalfoldedformsofmutantproteinexist atleasttransiently.Whetherthesesolublemalfolded formsofmutantSOD1areprecursorsoftheaggregates thatformoramisfoldedsolubleentitythatisoffpathwayforaggregationisuncertain. IncomparingtheC4F6andSEDIantibodies,each haveuniqueutility.TheSEDIantibodywouldappearto beaverygoodantibodyforlabelinginclusionstructures. Althoughthisantibodyisconformationally-restrictedin reactivityasitdoesnotr ecognizenativelyfolded dimericWTSOD1[17],itisnotaconformationspecific epitope;thisantibodywillrecognizedenaturedWT SOD1andcouldpresumablyrecognizeanimmature formofSOD1inamonomericstatethatisonpathway fornativefolding.Inourhands,theC4F6antibody seemstobehavemorelikeaconformationspecific antibodyinthatitdoesnotrecognizedenaturedWT SOD1ordenaturedformsofmutantSOD1,exceptfor specificreactivitywithdenaturedG93ASOD1[27]. Whilethislaterfindingwouldindicatereactivitywitha specificmutantsequenceofSOD1(theG93Amutation andflankingsequence),thereactivityofC4F6withdiffuselydistributedcytosol icSOD1incellsexpressing A4VSOD1isconsistentwiththeantibodyrecognizinga specificmal-foldedconformationasopposedtosimply recognizinganimmatureformofmutantSOD1ora specificsequenceepitope.Thus,theC4F6antibodymay beuniquelyvaluableinidentifyingmal-foldedSOD1 proteins.Interestingly,Boscoetaldescribeddiffuse labelingofmotorneuronsofsporadicALScaseswith theC4F6antibody,suggestingthatmisfoldingofWT SOD1occursatsomepointinthecourseofsporadic disease[27]. Wenote,however,thatKarumbayaramandcolleagues usedC4F6inimmunostainsofEScell-derivedmotor neuronsexpressingWT,A4V,I113T,andG93A mutants,findingthattheantibodyspecificallyrecognizedcellsexpressingtheG93Avariant[41](Additional File1,TableS1).Inthisstudy,itappearsthatthe sequencespecificepitopeoftheC4F6antibodypredominated.ThelevelofmutantSOD1expressioninthese cellswasnotcharacterizedanditisdifficulttoexplain whytheES-motor-neuroncellmodeldidnotshowthe misfoldingofothermutantsthatwereexpressed.Immunohistochemicalanalysesoftransgenicmouse modelsandhumanALSwithconformationallyrestricted antibodiesThehSOD1antibodyhasbeenextensivelyusedin immunohistochemicalanalysesofspinalcordtissues frommiceexpressingWTandmutantSOD1withthe patternofimmunoreactivityshowingsomevariability dependinguponhowthetissueswereprepared(AdditionalFile1,TableS1).Generally,theantibodyshows cellbodyreactivitythatissimilartowhatwehave describedhereinourcellmodelintissuesfrommutant SOD1transgenicmicewhenthetissueshavebeenpreparedforimmunohistochemistrybycryoprotectionand frozensectioning,withnoantigenretrieval(Additional File1,TableS1).Immunola belingofcellbodiesin mutantmousetissuespreparedbyparaffinembedding withmicrowaveheatingforantigenretrievalhasgenerallybeenweak,buttherearealsoexamplesofrobust cellbodyimmunoreactivity(AdditionalFile1,TableS1). Theonlypublishedexampleofimmunostainingoftissuesfrommicethatover-expressWTSOD1utilizeda paraffinembeddingmethodwithmicrowaveheatingfor antigenretrieval,findingre lativeweaklabelingofcell bodieswithdiffuseneurop ilimmunoreactivity[42]. Importantly,hSOD1antibodyhasconsistentlybeenPrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page13of18

PAGE 15

foundtoshowlittleornoreactivitywithinclusion structures(AdditionalFile1,TableS1). TheC4F6andSEDIantibodieshavealsobeenusedin studiesoftissuesfromhumanALSandSOD1mouse models,thoughnotasextensivelyinthecaseofthe C4F6antibody.OkamatoreportedintenseimmunoreactivityincellbodiesoftheG93AmicewiththeC4F6 antibody,withlabelingofstructuresinterpretedas inclusions[39](AdditionalFile1,TableS1).Thedegree towhichthispatternofstainingreflectsthemutation specificcomponent(G93 A)oftheC4F6epitopeis uncertain.OurdatawouldindicatethatC4F6doesnot readilyrecognizeinclusionsformedbyotherfALSvariantsofSOD1.TheSEDIantibodyhasbeenusedinstudiesofmiceexpressingtheG37R,G85R,andG93A mutantsaswellashumanfALSandsALScases[17] (AdditionalFile1,TableS1).IntheG85Rmiceand humanfALSassociatedwithA4V,A4T,I113T,V14M, anddG27/P28mutations,SEDIantibodyshowedintense reactivitytoinclusionstructuresincellbodies.However, inmiceexpressingtheG37RandG93Amutants,SEDI antibodyprimarilyimmunostainedthemarginsof vacuolarpathologythatisprevalentintheseanimals [17].Ifinclusionswerealsopresentinthespinalcords ofthesemice,theywerenotselectivelyrecognized. Overall,thesefindingsareconsistentwithourresults andindicatetheutilityoftheSEDIantibodyindetecting mutantSOD1inclusionpathology.Itisnoteworthythat spinalcordtissuesfromsALSpatientslackingmutations inSOD1showlittleornoreactivitywiththeSEDIantibody[43].However,Forsbergandcolleaguesreported thatanSOD1antibodyraisedagainstaC-terminalpeptidethatincludedtheepitopeoftheSEDIantibody immunostainedsmallpunctateinclusionstructuresin motorneuronsofsALScasesaswellasnon-SOD1 fALScases[44].Whetherdifferencesinantibodyspecificityordifferencesintissue preparationaccountforthe differentoutcomesisunknownandthepotentialinvolvementofmisfoldedWTSOD1withinmotorneurons insALSremainsadebatedtopic.AggregationandtoxicityInthecellmodelsweuseheretoexaminemutant SOD1folding,weobserveverylowlevelsofcelldeath bythemethodsweusedtodetectdyingcells.Thisfindingindicatesthatmisfolde dandaggregatedformsof mutantSOD1arenotacutelytoxicinthesecellmodels. WefocusedeffortonmodelsinwhichSOD::YFPfusion proteinswereexpressedsothatwecoulddefinitively identifytransfectedcells.Cellsharboringinclusionsgenerallylackedevidenceofimpendingdeathandverylittle evidenceoftoxicitywasnotedincellsshowingmore diffusedistributionsofSOD::YFPfusionproteins. Although,itispossibletha tcellsthatdieddetached fromtheplatequicklyandescapeddetectionbyour methods,weareleftwithnoclearindicationoftoxicity. Severalpreviousstudieshavesoughttodetermine whetheraggregatesofmutantSOD1areassociatedwith toxicityinavarietyofneuralcellmodels.Weareaware ofonlytwostudiesthatha veusedmethodsthatwould allowforsimultaneousassessmentofaggregation(via expressionofYFPfusionproteins)andvisualreporters ofcelltoxicity.UsingaratPC12cellmodel,whichcan bedifferentiatedintoneuron-likecells,andtime-lapse analysis,Matsumotoetal[31]observedthatcellsthat formedinclusionsofSOD1::YFPfusionproteinswere muchmorelikelytodie.However,celldeathwasalso observedincellslackinginclusions.ZhangandZhu examinedtoxicityinNSC34cells,whichwerecreated byfusionofmurineneuroblastomacellsandmurine motorneurons[45],findingnoevidencefordirecttoxicitydespitethepresenceofaggregatesincellsexpressingmutantSOD1fusedtoGFP[29].Inthisstudy, NSC34cellsstablyexpressingmutantSOD::GFPfusion proteinswerereportedtobemoresensitivetooxidative stressinducedbylowlevelsofH2O2,buttherewasno reportedexaminationofwhe thercellswithinclusions werespecificallymoresensitive.Asimilarlackofdirect toxicitybymutantSOD1fusedtoeGFPwhenexpressed inNSC34cellswasreportedbyTurneretal.[33]. Therearetworeportsdescribingtheexpressionof mutantSOD1fusedtoGFPinmouseneuroblastoma N2acells.Onestudyreportedfindingacorrelation betweenexpressionofmutantsthatreadilyforminclusionsandtoxicity(MTTassay)[46].Inthisstudy, experimentalmutationsthatchangedcysteineresidues atposition6and111reducedaggregationandreduced toxicity.Inasecondstudy,Witanetal[47]failedto observedirecttoxicityduetoexpressionofmutant SOD1fusionswithEGFPdespitedetectionofinclusions athighfrequency.However,similartoZhangandZhu [29],Witanetalreportedthatcellsexpressingthe mutantfusionproteinsweremoresensitivetooxidative stressinducingagents[47].Theauthorsdidnotinvestigatewhethercellswithinclusionswerespecificallymore sensitivetooxidativestress.ThestudybyWitandiffers fromallothersinthatthefusionproteinsexpressed wereengineeredascontiguoushomodimersorcontiguousheterodimers(mutantwithWT).Overall,theoutcomesoftheseinvestigationsprovideamixedviewof theroleofmutantSOD1aggregationintoxicity,andit remainsdifficulttoexplaindivergentoutcomes. Inpreviousworkintransgenicmicetherehasbeen muchefforttodefinechangesinthephysicalnature mutantSOD1throughthecourseofdisease.InanelegantstudybyWangandcolleagues,expressionofa G85RvariantofhumanSOD1fusedtoYFPclearly demonstratedthatlargevisibleinclusionsformverylatePrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page14of18

PAGE 16

inthecourseofdisease[48].Inourpreviousstudiesof transgenicmousemodelsofALSexpressinguntagged variantsofmutantSOD1,wehavesimilarlyobserved thatthegreatestaccumulationofdetergent-insoluble aggregatesofmutantSOD1occurslateindiseaseprogression[22].IntheG93A-Gur1model,whichreaches end-stageparalysisatabout120days,aggregatesbegin toaccumulatebetween90and105daysandthenrise dramaticallyastheanimalsapproachendstage[22]. Duringthissameinterval,serumlevelsofneurofilament H,whichserveabiomarkerofaxonaldegeneration[23], similarlyrisedramatically.ThuswecancorrelateaggregationofmutantSOD1andaxonaldegenerationinthe animalmodels.Moreover,intheG93Amodel,the majorityofmotorneurondeathoccursduringtheintervalinwhichdetergent-insolubleaggregatesappear. Althoughitistemptingtolinkinclusionformationto motorneurondeath,itremainsdifficulttoassign mutantproteinaggregationascausativeofcelldeathas itispossiblethattheaccumulationofaggregatesserves asabiomarkerofcellulardegenerationinwhichthe cellssimplylosetheabilitytopreventthemisfolded mutantproteinfromaggregatingbecauseofsomecombinationofdecliningchaperoneactivities,decliningproteasomeactivitytodegradethemisfoldedprotein,or decliningenergyproductiontosupporttheseactivities. Itmaynotbepossibletosortouttheroleofaggregationintoxicityuntilwehavetoolstopreventsuch aggregatesfromoccurringinvivo[49]. Althoughwemightwishtobeabletoassigntoxicity toaspecificconformation,assembly,oraggregateof mutantSOD1,itseemslikelythatmultipleformsof misfoldedSOD1exertdelete riouseffects.Forexample, itisfairlywellestablishedthatearly,duringtheinsidiousphaseofdiseasewhenp athologicabnormalities areobservedwithlimitedsymptoms,thereislimited accumulationofthedetergent-insolubleaggregates; instead,detergentsolubleformsofmutantSOD1predominate[22].Asmentionedabove,apreviousstudyby Zetterstrometaldemonstra tedthatearlyinthecourse ofdisease,spinalcordsofmiceexpressingmutant SOD1containsolubleformsofmutantSOD1thatbind tohydrophobicaffinitychromatographymediaandlack thenormalintramoleculardisulfidelinkage[40]. Whetherthesenon-nativelyfoldedformsofmutant SOD1mediateearlyeventsintoxicityremainstobe established,theirearlyappearanceintissuesfromthe mutantmiceclearlymakesthemcandidates.OurfindingsindicatethattheC4F6antibodymightbeusefulin detectingsolublemisfoldedformsofmutantSOD1early inthecourseofdisease.However,wewouldrecommendthatsuchstudiesfocusonmutantsotherthanthe G93Amutanttoavoidmutationspecificreactivityin favorofmisfoldedconfirmationreactivity.Itwould,of course,beeasiertoexplainthediseaseifonespecific formofmutantSOD1(monomer,oligomer,aggregate, orinclusion)targetedonespecificprocessthatisunique toneuronsandofgreaterimportancetomotorneurons, buttheavailableevidencesuggestsmultiplesystems maybetargetsofmutantSOD1toxicity[50].ConclusionsThepresentstudyusesacombinationofnaturalSOD1 mutantandSOD1::YFPfusionproteinstoexaminethe relationshipsbetweendeterg entinsolublestructures, aggregates,andinclusionbodyformation.Studiesin mice,usingbothbiochemicalmethodsofdetectionand YFPtaggingmethodshaveestablishedthatlargeraggregatesformlateinthedisease,primarilyduringthetime ofrobustcelldeathandrapiddiseaseprogression [22,48].However,thesestudiesalsoestablishthatpathologicandbehavioralsymptomsoccurinmicepriorto theaccumulationoftheselargeraggregates,implying thatsmallerassembliesofmutantSOD1mustmediate someaspectoftoxicityandtriggerdisease.Ourstudies definetheutilityofdifferentapproachesindetecting mal-foldedformsofWTandmutantSOD1thatare bothsolubleandinsoluble.Goingforward,itmaybe possiblethroughtheuseofthesevarioustoolstobetter definetheroleoftheseva riousassembliesofmutant SOD1intheevolutionofpathologicandbehavioral abnormalitiesthatdefinetheALSmousemodels.MethodsSOD1cDNAexpressionplasmidsandcelllinesWTandmutanthSOD1untaggedproteinswere expressedfromplasmidsbasedonthemammalianpEFBOSexpressionvector,andhavebeenpreviously described[1,16,20,21]. YFP taggedSOD1cDNAvariants werecreatedfromawormexpressionvector(pPD30.38) thatcontainsWThSOD1fusedtoeYFP(yellowfluorescentprotein)kindlyprovidedbyDr.RickMorimoto (NorthwesternUniversity).WT::YFP,WT::RFPandYFP variantshavebeenpreviouslydescribed[35],and mutantfluorescentlytaggedSOD1variantswereconstructedfollowingsimilarprocedures. AllcellculturestudiesofSOD1aggregationused HEK293FTcells(Invitrogen,Carlsbad,CA),mouseTKLcells,orCHOcells(ATCC,Manassas,VA)asindicated,whichweremaintainedfollowingATCC recommendations.TransfectionsTransfectionofcellsforimmunocytochemistrywasperformedonglasscoverslipsthatwerepreviouslycoated with0.5mg/mlpoly-L-lysinein1phosphatebuffered salinesolution(PBS).Atotalof2 gofvectorDNAwas transfectedperwell,usingLipofectamine2000PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page15of18

PAGE 17

(Invitrogen,Carlsbad,CA).Forbiochemicalanalyses,a totalof4 gofvectorDNAwasusedtotransfectcells in60mmpoly-L-lysinecoateddishes(BDBiosciences, Bedford,MA).Eachtransfectionexperimentwas repeatedaminimumof3times.ImmunocytochemistryTransfectedcellswereeitherfixedwith4%paraformaldehydein1PBSsolutionfor15minutesorincubated for30minutesinmilddetergents(0.01%digitoninor 0.01%saponinin1PBS)priorcellfixation.Cellswere thencompletelypermeabilizedusingcold100%methanolfor5minutesfollowedbyincubationin20%normal goatserumin1PBSandimmunostainingwithhSOD1 (1:500),SEDI(1:500),orC4F 6(1:500)antibodiesovernight(in1PBSwith10%normalgoatserum).Primary antibodieswerevisualizedwithsecondaryfluorescentlylabeledantibodies(1:2000)dilutedinsamebufferasthe primaryantibody(Alexafluorgoatanti-rabbit594nm forhSOD1orSEDI,orAlexafluorgoatanti-mouse568 nmforC4F6antibody;Invitrogen,Carlsbad,CA).To visualizenuclei,cellsweretreatedwithDAPIsolution (4 ,6-diamidino-2-phenylindole,dihydrochloride,stock 14.3mMfromInvitrogen,Carlsbad,CA)atadilutionof 1:2000for10minutes.Fluorescencewasvisualized usinganOlympusIX81-DS Uspinningdiskconfocal microscopeoranepifluorescenceOlympusBX60microscopeasindicatedintheFigureLegends.DetergentextractionofSOD1aggregatesExtractionofculturedcellsinnonionicdetergents(NP40)wasperformedasdescribedpreviously[1,21,22]. Theprocedureproducestwoproteinfractionstermed S1(detergent-soluble)andP2(detergent-insoluble), whichwerethenanalyzedbyimmunoblotting(5 g totalproteinfromS1iscomparedto20 gtotalprotein fromP2).WethenquantifiedtheintensityoftheSOD1 bandsineachfractionandcalculatedthemeanratioof insolubletosolubleprotein.Datawerethenanalyzedon GraphPadPRISM5.01Software(LaJolla,CA)todeterminestatisticaldifferen ces,usingunpairedstudent t tests,andgenerategraphicrepresentations.CellDeathAssaysToassessnuclearDNAfragmentation,weusedTUNEL labelingassays(InSituCellDeathDetectionKit,Roche AppliedScience,cat#12156792910,Indianapolis,IN, USA).Briefly,CHOcellswereplatedinto12-wellplates containinglysine-coatedcoverslipsandgrownovernight. Thecellswerethentransfectedwith0.8 gDNAin2 l ofLipofectamine2000dilu ted,ultimatelyin500 lof Opti-MEM.After3hat37C,500 lofcompletemedia (Ham sF12with10%fetalbovineserumand2mMLglutamine)wasadded.Forapositivecontrol,onewell ofuntransfectedcellswastr eatedwithstaurosporine1 or2 Mfinalconcentration,(40 lor80 lofa50mM stockin1mlmedia)overnightat37C.Twenty-four hoursaftertransfection,cellswererinsed3timeswith 1PBS,fixedin4%paraformaldehydein1PBSfor20 minatRT,andthenrinsed3timesin1PBS.Before immunostaining,thecellswerepermeabilizedinpreheated10mMTri-sodiumcitrateand0.05%Tween20 (pH6.0)for10minat90C,andthenrinsed3timesin 1PBS. ForTUNELlabeling,wefollowedthemanufacturer s protocol,whichentailsmak ingareactionmixtureby mixing8 lofenzymesolutionintoa72 llabeling solutionthatiskeptinthedarkoniceuntilreadyto use.Eachcoverslipwasthenincubatedinthelabeling solutionbypipetting18 lofthesolutionmixontohalf ofthecoverslips,whichwasthencoveredwithapiece ofparafilmtokeepitfromdryingout.Theotherhalfof thecoverslipswerecovered18 loflabelingsolution withnoenzymeforanegativecontrol(alsocovered withparafilm).Allcoverslipswereincubatedinahumid chamberat37Cfor2-2.5hoursbeforerinsing3times with1PBSfollowedbyfloatingoffandremovingthe parafilm.Coverslipswerethenmountedonglassslides usingAqua-Poly/Mount(Polysciences,Inc.,Warrington, PA,USA,cat#18606)andphotographed. AsecondassayforcelldeathutilizedtheLIVE/DEAD Viability/CytotoxicityKitformammaliancells[Invitrogen(Molecularprobes),Carlsbad,CA,USA,cat# L3224].Forthisassay,cells weregrownoncoverslips andtransfectedasdescribedabove.Positivecontrol cells,treatedwithstaurosporine,werealsogeneratedas describedabove.After24or48hoursofincubation,the cellswererinsed3timeswith1PBSandthenincubatedwithComponentBofthekit(Ethidiumhomodimer-1)at5or10 Min1PBSatroomtemperature or37Cfor10to30minutes.Thecoverslipswerethen rinsedin1PBS,mountedonglassslidesandsealed usingclearnailpolishbeforephotography.Wealso foundthatafterremovalofexcessreagent,thecells couldbefixedin4%paraformaldehydeandvisualized.AdditionalmaterialAdditionalfile1:SupplementaryFiguresS1-8;SupplementaryTable 1.Suppl.Figure1.ThemajorityofSOD1intransiently-transfected HEK293FTtransfectedcellsissolubleinnon-ionicdetergent.Suppl. Figure2.Formicacidtreatmentdoesnotallowvisualizationofinclusions usingahSOD1antibody.Suppl.Figure3.TKnegativecellstransfected withSOD1constructsfor48hoursandstainedforhumanSOD1.Suppl. Figure4.HEK293FTcellsexpressingaSOD1::YFPvariantaccumulate detergent-insolubleaggregates.Suppl.Figure5.Digitonintreatmentin TKnegativecellsrevealsapunctatepatternofSOD1immunoreactivityin cellsexpressingeitherWTorA4VSOD1.Suppl.Figure6.Digitonin extractsaportionofthedetergent-solublehSOD1proteinexpressedin HEK293FTcells.SupplementalFigure7.EtBruptakeassayofcellPrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page16of18

PAGE 18

permeabilityinHEK293FTcellsexpressingmutantandWTYFPfusion proteins.Suppl.Figure8.AssessmentofcelldeathinCHOcells expressingWTandA4VSOD::YFPfusionproteins.SupplementaryTable1. Summaryofpreviousfindingswithconformationally-restrictedantibodies [4,5,15,17,26,27,39,41-43,51-56]. Acknowledgements WethankMs.HildaBrownforherhelpingeneratingmutantSOD1cDNA vectorsandtechnicalassistancewithsomeofthecellcultureexperiments. WethankDr.RichardMorimotoforprovidingaplasmidwithSOD1cDNA forthegenerationofSOD1::YFPfusionproteins.Thisworkwasfundedbya grantfromtheNationalInstitutesofNeurologicDiseaseandStroke(P01 NS049134wardtoDrs.JoanS.Valentine,P.JohnHart,D.R.Borchelt,andJ. P.Whitelegge-wethankourcolleaguesforthoughtfuldiscussions). Authordetails1DepartmentofNeuroscience,McKnightBrainInstitute,UniversityofFlorida, Gainesville,FL32610,USA.2DepartmentofNeuroscience,MayoClinic CollegeofMedicine,4500SanPabloRd,Jacksonville,FL32224,USA. Authors contributions MPandDRBdesignedresearch,MPperformedresearch,MPandDRB oversawtheeffortoftechnicalassistantsnamedbelow,analyzeddata.DRB andMPwrotethemanuscript.Allauthorsreadandapprovedthefinal manuscript. Competinginterests Theauthorsdeclarethattheyhavenocompetinginterests. Received:13June2011Accepted:17November2011 Published:17November2011 References1.PrudencioM,HartPJ,BorcheltDR,AndersenPM: Variationinaggregation propensitiesamongALS-associatedvariantsofSOD1:correlationto humandisease. HumMolGenet 2009, 18 :3217-3226. 2.ShibataN,HiranoA,KobayashiM,SiddiqueT,DengH-X,HungW-Y,KatoT, AsayamaK: Intensesuperoxidedismutase-1immunoreactivityin intracytoplasmichyalineinclusionsoffamilialamyotrophiclateral sclerosiswithposteriorcolumninvolvement. JNeuropatholExpNeurol 1996, 55 :481-490. 3.KatoS,ShimodaM,WatanabeY,NakashimaK,TakahashiK,OhamaE: Familialamyotrophiclateralsclerosiswithatwobasepairdeletionin superoxidedismutase1gene:multisystemdegenerationwith intracytoplasmichyalineinclusionsinastrocytes. JNeuropatholExpNeurol 1996, 55 :1089-1101. 4.BruijnLI,BecherMW,LeeMK,AndersonKL,JenkinsNA,CopelandNG, SisodiaSS,RothsteinJD,BorcheltDR,PriceDL, etal : ALS-linkedSOD1 mutantG85Rmediatesdamagetoastrocytesandpromotesrapidly progressivediseasewithSOD1-containinginclusions. Neuron 1997, 18 :327-338. 5.BruijnLI,HouseweartMK,KatoS,AndersonKL,AndersonSD,OhamaE, ReaumeAG,ScottRW,ClevelandDW: Aggregationandmotorneuron toxicityofanALS-linkedSOD1mutantindependentfromwild-type SOD1. Science 1998, 281 :1851-1854. 6.KatoS,HoriuchiS,LiuJ,ClevelandDW,ShibataN,NakashimaK,NagaiR, HiranoA,TakikawaM,KatoM, etal : Advancedglycationendproductmodifiedsuperoxidedismutase-1(SOD1)-positiveinclusionsare commontofamilialamyotrophiclateralsclerosispatientswithSOD1 genemutationsandtransgenicmiceexpressinghumanSOD1witha G85Rmutation. ActaNeuropathol 2000, 100 :490-505. 7.StieberA,GonatasJO,GonatasNK: Aggregatesofmutantproteinappear progressivelyindendrites,inperiaxonalprocessesofoligodendrocytes, andinneuronalandastrocyticperikaryaofmiceexpressingtheSOD1 (G93A)mutationoffamilialamyotrophiclateralsclerosis. JNeurolSci 2000, 177 :114-123. 8.StieberA,GonatasJO,GonatasNK: Aggregationofubiquitinanda mutantALS-linkedSOD1proteincorrelatewithdiseaseprogressionand fragmentationoftheGolgiapparatus. JNeurolSci 2000, 173 :53-62. 9.KatoS,Sumi-AkamaruH,FujimuraH,SakodaS,KatoM,HiranoA, TakikawaM,OhamaE: Copperchaperoneforsuperoxidedismutasecoaggregateswithsuperoxidedismutase1(SOD1)inneuronalLewybodylikehyalineinclusions:animmunohistochemicalstudyonfamilial amyotrophiclateralsclerosiswithSOD1genemutation. ActaNeuropathol 2001, 102 :233-238. 10.WatanabeM,Dykes-HobergM,CulottaVC,PriceDL,WongPC, RothsteinJD: HistologicalEvidenceofProteinAggregationinMutant SOD1TransgenicMiceandinAmyotrophicLateralSclerosisNeural Tissues. NeurobiolDis 2001, 8 :933-941. 11.OhiT,SaitaK,TakechiS,NabesimaK,TashiroH,ShiomiK,SugimotoS, AkematsuT,NakayamaT,IwakiT, etal : Clinicalfeaturesand neuropathologicalfindingsoffamilialamyotrophiclateralsclerosiswith aHis46ArgmutationinCu/Znsuperoxidedismutase. JNeurolSci 2002, 197 :73-78. 12.ArisatoT,OkuboR,ArataH,AbeK,FukadaK,SakodaS,ShimizuA,QinXH, IzumoS,OsameM, etal : Clinicalandpathologicalstudiesoffamilial amyotrophiclateralsclerosis(FALS)withSOD1H46Rmutationinlarge Japanesefamilies. Acta Neuropathol 2003, 106 :561-568. 13. SasakiS,WaritaH,MurakamiT,ShibataN,KomoriT,AbeK,KobayashiM, IwataM: Ultrastructuralstudyofaggregatesinthespinalcordof transgenicmicewithaG93AmutantSOD1gene. ActaNeuropathol 2005, 109 :247-255. 14.SasakiS,NagaiM,AokiM,KomoriT,ItoyamaY,IwataM: Motorneuron diseaseintransgenicmicewithanH46RmutantSOD1gene. J NeuropatholExpNeurol 2007, 66 :517-524. 15.WongPC,PardoCA,BorcheltDR,LeeMK,CopelandNG,JenkinsNA, SisodiaSS,ClevelandDW,PriceDL: AnadversepropertyofafamilialALSlinkedSOD1mutationcausesmotorneurondiseasecharacterizedby vacuolardegenerationofmitochondria. Neuron 1995, 14 :1105-1116. 16.WangJ,SluntH,GonzalesV,FromholtD,CoonfieldM,CopelandNG, JenkinsNA,BorcheltDR: Copper-binding-site-nullSOD1causesALSin transgenicmice:aggregatesofnon-nativeSOD1delineateacommon feature. HumMolGenet 2003, 12 :2753-2764. 17.RakhitR,RobertsonJ,VandeVC,HorneP,RuthDM,GriffinJ,ClevelandDW, CashmanNR,ChakrabarttyA: Animmunologicalepitopeselectivefor pathologicalmonomer-misfoldedSOD1inALS. NatMed 2007, 13 :754-759. 18.JonssonPA,ErnhillK,AndersenPM,BergemalmD,BrannstromT,GredalO, NilssonP,MarklundSL: Minutequantitiesofmisfoldedmutant superoxidedismutase-1causeamyotrophiclateralsclerosis. Brain 2004, 127 :73-88. 19.BassoM,MassignanT,SamengoG,CheroniC,DeBiasiS,SalmonaM, BendottiC,BonettoV: InsolublemutantSOD1ispartlyoligoubiquitinated inamyotrophiclateralsclerosismice. JBiolChem 2006, 281 :33325-33335. 20.KarchCM,BorcheltDR: Alimitedrolefordisulfidecross-linkinginthe aggregationofmutantSOD1linkedtofamilialamyotrophiclateral sclerosis. JBiolChem 2008, 283 :13528-13537. 21.PrudencioM,DurazoA,WhiteleggeJP,BorcheltDR: Modulationofmutant superoxidedismutase1aggregationbyco-expressionofwild-type enzyme. JNeurochem 2009, 108 :1009-1018. 22.KarchCM,PrudencioM,WinklerDD,HartPJ,BorcheltDR: Roleofmutant SOD1disulfideoxidationandaggregationinthepathogenesisof familialALS. ProcNatlAcadSciUSA 2009, 106 :7774-7779. 23.BoylanK,YangC,CrookJ,OverstreetK,HeckmanM,WangY,BorcheltD, ShawG: Immunoreactivityofthephosphorylatedaxonalneurofilament Hsubunit(pNF-H)inbloodofALSmodelrodentsandALSpatients: evaluationofbloodpNF-HasapotentialALSbiomarker. JNeurochem 2009, 111 :1182-1191. 24.EzziSA,UrushitaniM,JulienJP: Wild-typesuperoxidedismutaseacquires bindingandtoxicpropertiesofALS-linkedmutantformsthrough oxidation. JNeurochem 2007, 102 :170-178. 25.Gros-LouisF,SoucyG,LariviereR,JulienJP: Intracerebroventricular infusion of monoclonalantibodyoritsderivedFabfragmentagainst misfoldedformsofSOD1mutantdelaysmortalityinamousemodelof ALS. JNeurochem 2010, 113 :1188-1199.PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page17of18

PAGE 19

26.KarchCM,BorcheltDR: AnexaminationofalphaB-crystallinasamodifier ofSOD1aggregatepathologyandtoxicityinmodelsoffamilial amyotrophiclateralsclerosis. JNeurochem 2010, 113 :1092-1100. 27.BoscoDA,MorfiniG,KarabacakNM,SongY,Gros-LouisF,PasinelliP, GoolsbyH,FontaineBA,LemayN,McKenna-YasekD, etal : Wild-typeand mutantSOD1shareanaberrantconformationandacommon pathogenicpathwayinALS. NatNeurosci 2010, 13 :1396-1403. 28.CorcoranLJ,MitchisonTJ,LiuQ: Anovelactionofhistonedeacetylase inhibitorsinaproteinaggresomediseasemodel. CurrBiol 2004, 14 :488-492. 29.ZhangF,ZhuH: IntracellularconformationalalterationsofmutantSOD1 andtheimplicationsforfALS-associatedSOD1mutantinducedmotor neuroncelldeath. BiochimBiophysActa 2006, 1760 :404-414. 30.FeiE,JiaN,YanM,YingZ,SunQ,WangH,ZhangT,MaX,DingH,YaoX, etal : SUMO-1modificationincreaseshumanSOD1stabilityand aggregation. BiochemBiophysResCommun 2006, 347 :406-412. 31.MatsumotoG,StojanovicA,HolmbergCI,KimS,MorimotoRI: Structural propertiesandneuronaltoxicityofamyotrophiclateralsclerosisassociatedCu/Znsuperoxidedismutase1aggregates. JCellBiol 2005, 171 :75-85. 32.WitanH,GorlovoyP,KayaAM,Koziollek-DrechslerI,NeumannH,BehlC, ClementAM: Wild-typeCu/Znsuperoxidedismutase(SOD1)doesnot facilitate,butimpedestheformationofproteinaggregatesof amyotrophiclateralsclerosiscausingmutantSOD1. NeurobiolDis 2009, 36 :331-342. 33.TurnerBJ,AtkinJD,FargMA,ZangDW,RembachA,LopesEC,PatchJD, HillAF,CheemaSS: Impairedextracellularsecretionofmutant superoxidedismutase1associateswithneurotoxicityinfamilial amyotrophiclateralsclerosis. JNeurosci 2005, 25 :108-117. 34.UrushitaniM,EzziSA,MatsuoA,TooyamaI,JulienJP: Theendoplasmic reticulum-Golgipathwayisatargetfortranslocationandaggregationof mutantsuperoxidedismutaselinkedtoALS. FASEBJ 2008, 22 :2476-2487. 35.PrudencioM,DurazoA,WhiteleggeJP,BorcheltDR: Anexaminationof wild-typeSOD1inmodulatingthetoxicityandaggregationofALSassociatedmutantSOD1. HumMolGenet 2010, 19 :4774-4789. 36.SymonsMH,MitchisonTJ: Controlofactinpolymerizationinliveand permeabilizedfibroblasts. JCellBiol 1991, 114 :503-513. 37.CallahanJ,KopeckovP,KopecekJ: Intracellulartraffickingandsubcellular distributionofalargearrayofHPMAcopolymers. Biomacromolecules 2009, 10 :1704-1714. 38.CrapoJD,OuryT,RabouilleC,SlotJW,ChangL-Y: Copper,zincsuperoxide dismutaseisprimarilyacytosolicproteininhumancells. ProcNatlAcad SciUSA 1992, 89 :10405-10409. 39.OkamotoY,ShirakashiY,IharaM,UrushitaniM,OonoM,KawamotoY, YamashitaH,ShimohamaS,KatoS,HiranoA, etal : Colocalizationof14-3-3 proteinswithSOD1inLewybody-likehyalineinclusionsinfamilial amyotrophiclateralsclerosiscasesandtheanimalmodel. PLoSONE 2011, 6 :e20427. 40.ZetterstromP,StewartHG,BergemalmD,JonssonPA,GraffmoKS, AndersenPM,BrannstromT,OlivebergM,MarklundSL: Solublemisfolded subfractionsofmutantsuperoxidedismutase-1sareenrichedinspinal cordsthroughoutlifeinmurineALSmodels. ProcNatlAcadSciUSA 2007, 104 :14157-14162. 41.KarumbayaramS,KellyTK,PaucarAA,RoeAJ,UmbachJA,CharlesA, GoldmanSA,KornblumHI,Wiedau-PazosM: HumanembryonicstemcellderivedmotorneuronsexpressingSOD1mutantsexhibittypicalsignsof motorneurondegenerationlinkedtoALS. DisModelMech 2009, 2 :189-195. 42.WangJ,XuG,GonzalesV,CoonfieldM,FromholtD,CopelandNG, JenkinsNA,BorcheltDR: Fibrillarinclusionsandmotorneuron degenerationintransgenicmiceexpressingsuperoxidedismutase1 withadisruptedcopper-bindingsite. NeurobiolDis 2002, 10 :128-138. 43.LiuHN,SanelliT,HorneP,PioroEP,StrongMJ,RogaevaE,BilbaoJ, ZinmanL,RobertsonJ: Lackofevidenceofmonomer/misfolded superoxidedismutase-1insporadicamyotrophiclateralsclerosis. Ann Neurol 2009, 66 :75-80. 44.ForsbergK,JonssonPA,AndersenPM,BergemalmD,GraffmoKS, HultdinM,JacobssonJ,RosquistR,MarklundSL,BrannstromT: Novel antibodiesrevealinclusionscontainingnon-nativeSOD1insporadicALS patients. PLoSONE 2010, 5 :e11552. 45.CashmanNR,DurhamHD,BlusztajnJK,OdaK,TabiraT,ShawIT, DahrougeS,AntelJP: Neuroblastomaspinalcord(NSC)hybridcell linesresembledevelopingmotorneurons. DevDynamics 1992, 194 :209-221. 46.NiwaJ,YamadaS,IshigakiS,SoneJ,TakahashiM,KatsunoM,TanakaF, DoyuM,SobueG: Disulfidebondmediatesaggregation,toxicity,and ubiquitylationoffamilialamyotrophiclateralsclerosis-linkedmutant SOD1. JBiolChem 2007, 282 :28087-28095. 47.WitanH,KernA,Koziollek-DrechslerI,WadeR,BehlC,ClementAM: Heterodimerformationofwild-typeandamyotrophiclateralsclerosiscausingmutantCu/Zn-superoxidedismutaseinducestoxicity independentofproteinaggregation. HumMolGenet 2008, 17 :1373-1385. 48.WangJ,FarrGW,ZeissCJ,Rodriguez-GilDJ,WilsonJH,FurtakK, RutkowskiDT,KaufmanRJ,RuseCI,YatesJRIII, etal : Progressive aggregationdespitechaperoneassociationsofamutantSOD1-YFPin transgenicmicethatdevelopALS. ProcNatlAcadSciUSA 2009, 106 :1392-1397. 49.ChenT,BenmohamedR,ArvanitesAC,RalayRH,MorimotoRI,FerranteRJ, WattersonDM,KirschDR,SilvermanRB: Arylsulfanylpyrazolonesblock mutantSOD1-G93Aaggregation.Potentialapplicationforthetreatment ofamyotrophiclateralsclerosis. BioorgMedChem 2011, 19 :613-622. 50.IlievaH,PolymenidouM,ClevelandDW: Non-cellautonomoustoxicityin neurodegenerativedisorders:ALSandbeyond. JCellBiol 2009, 187 :761-772. 51.ClementAM,NguyenMD,RobertsEA,GarciaML,BoilleeS,RuleM, McMahonAP,DoucetteW,SiwekD,FerranteRJ, etal : Wild-type nonneuronalcellsextendsurvivalofSOD1mutantmotorneuronsin ALSmice. Science 2003, 302 :113-117. 52.WangJ,XuG,SluntHH,GonzalesV,CoonfieldM,FromholtD, CopelandNG,JenkinsNA,BorcheltDR: Coincidentthresholdsofmutant proteinforparalyticdiseaseandproteinaggregationcausedby restrictivelyexpressedsuperoxidedismutasecDNA. NeurobiolDis 2005, 20 :943-952. 53.WangJ,XuG,LiH,GonzalesV,FromholtD,KarchC,CopelandNG, JenkinsNA,BorcheltDR: Somatodendriticaccumulationofmisfolded SOD1-L126Zinmotorneuronsmediatesdegeneration:{alpha}Bcrystallinmodulatesaggregation. HumMolGenet 2005, 14 :2335-2347. 54.WangJ,XuG,BorcheltDR: Highmolecularweightcomplexesofmutant superoxidedismutase1:age-dependentandtissue-specific accumulation. NeurobiolDis 2002, 9 :139-148. 55.PardoCA,XuZ,BorcheltDR,PriceDL,SisodiaSS,ClevelandDW: Superoxidedismutaseisanabundantcomponentincellbodies, dendrites,andaxonsofmotorneuronsandinasubsetofother neurons. ProcNatlAcadSciUSA 1995, 92 :954-958. 56.BorcheltDR,WongPC,BecherMW,PardoCA,LeeMK,XuZS,ThinakaranG, JenkinsNA,CopelandNG,SisodiaSS, etal : Axonaltransportofmutant superoxidedismutase1andfocalaxonalabnormalitiesintheproximal axonsoftransgenicmice. NeurobiolDis 1998, 5 :27-35.doi:10.1186/1750-1326-6-77 Citethisarticleas: PrudencioandBorchelt: Superoxidedismutase1 encodingmutationslinkedtoALSadoptsaspectrumofmisfolded states. MolecularNeurodegeneration 2011 6 :77. 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 PrudencioandBorchelt MolecularNeurodegeneration 2011, 6 :77 http://www.molecularneurodegeneration.com/content/6/1/77 Page18of18

PAGE 20

Prudencio and Borchelt, 2011 Suppl. Fig. 1. W T A4V 0.0 1.01008 2.01008 3.01008 4.01008P2 S1 Band intensity levels of SOD1 protein (A.U.)# # Suppl. Fig. 1. The majority of SOD1 in transiently-trans fected HEK293FT transfected cells is soluble in nonionic detergent. Band intensities were calculated from western blots of P2 (black bars) and S1 (white bars) fractions, and adjusted for the amount of total protein loaded on the SDS-PAGE gels (5 g for S1 and 20 g for P2). The data represented here derives from at leas t five independent transfec tion experiments. Significant differences exist between the levels of SOD1 P2 vs S1 fractions for each SOD1 protein expressed; unpaired student t -test: #p 0.005. The data were averaged from 3 independent transfection experiments. A possible explanation for the inabil ity to visualize inclusions in tr ansiently-transfected cells may be related to the amount of dete rgent-soluble SOD1 protein vs. the amount of detergent-insoluble aggregated SOD1 protein that is present. In tr ansient 24 hour transfections, the levels of detergent-soluble SOD1 protein in HEK293FT cells expressing mutant A4V SOD1 is about 4 times higher than the levels of detergent-insoluble aggregated proteins. Thus, it seems r easonable to explain our inability to observe inclusions may be due to a masking effect in which soluble protein obscures visualization of inclusions.

PAGE 21

Prudencio and Borchelt, 2011 Suppl. Fig. 2. WT hSOD1 hSOD1/DAPI A4V A C B D Suppl. Fig. 2. Formic acid treatment does not allow vi sualization of inclusions using a hSOD1 antibody. HEK293FT cells were transfected with either WT (A -B) or A4V (C-D) hSOD1 for 24 hours. Staining with hSOD1 antibody was performed as described in Methods, but here cells were treated w ith 70% formic acid for 10 minutes prior to fixation of the ce lls. Images were captured with a 40 x objective. Scale bars = 20 m. The images shown are representative of at least 3 independent transfection and immunostaining experiments.

PAGE 22

Prudencio and Borchelt, 2011 Suppl. Fig. 3. WTA4V UT A BC Suppl. Fig. 3. TK negative cells transf ected with SOD1 constructs for 48 hours and stained for human SOD1 as explained in Methods. Pictures were taken with a 40x objective. UT denotes untransfected cells. A possible explanation for our inability to detect muta nt SOD1 inclusions reside s in the possibility that the amount of endogenous WT SOD1, produced by th e HEK293FT cells, may increase the overall hSOD1 staining and interfere with the visual ization of protein inclus ions. Thus, to eliminate the background generated by endogenous WT SOD1 protein, we perf ormed transient transfections in a mouse fibroblast cell line (TKL cells) and used antibodies specific to human SOD1 to detect transfected protein. In mouse TKL cells, we do not detect detergent-insoluble SOD1 aggregates biochemically until 48 hours post-transfection (data not shown). Therefore, we immunostained fixed cells at 48 hrs, finding WT and A4V SOD1 proteins showed a similar diffuse pattern of fluorescence. The images s hown are representative of at least 3 independent transfection and immuno staining experiments.

PAGE 23

Prudencio and Borchelt, 2011 Suppl.Fig. 4. D101N::YFP S134N::YFP G85R::YFP G8 5 R: : Y FP D 1 01N:: Y FP S 1 34N::YFP 0 5 10 15Relative aggregation propensity of SOD1::YFP proteins (A.U.) S1 P2 *#AB C Suppl. Fig. 4. HEK293FT cells expressing a SOD1::YFP vari ant accumulate detergent-insoluble aggregates. A) Fluorescence images of SOD1::YFP c onstructs expressed for 24 hours. B) Immunoblot of P2 and S1 protein fractions of cells expressing SOD1::Y FP proteins for 24 hours. The SOD1::Y FP protein migrates at a relative molecular mass of 50 kDa (filled arro whead), while endogenous WT SOD1 monomer migrates at 16 kDa (open arrowhead). C) Quantification of the aggregation propensity (P2/S1) of ce lls expressing SOD1::YFP proteins. The data were averaged from at least 3 independent tr ansfection experiments for each mutant. Unpaired student t -tests were performed to compare the level of dete rgent-insoluble mutant SOD1::YFP with that of WT::YFP SOD1 fusion protein (see Fig. 1 for immunoblot of WT::YFP): p 0.05; #p 0.005.

PAGE 24

Prudencio and Borchelt, 2011 Suppl. Fig. 5. SOD1 SOD1/DAPI A C B D A4V WT Suppl. Fig. 5. Digitonin treatment in TK negative cells re veals a punctate pa ttern of SOD1 immunoreactivity in cells expressing either WT or A4V SOD1. To test whether membrane pe rmeabilization of transfected TKL cells would release soluble S OD1 and reveal inclusions, we treated cells with digitonin (similar in properties to saponin) at 48 hours post-transfection. In these cells, we observed hSOD1 to immunostains small punctate structures in cells expres sing both WT and A4V mutant SOD1. No obvious difference in the frequency or intensity of these structures as not ed. A-D) TKL cells were transf ected for 48 hours, treated with 0.01% digitonin in 1x PBS, and stained for human SOD1. Transfection, detergent treat ment and staining were performed as described for HEK293FT cells in Materials and Methods. Im ages corresponding to WT (A-B) and A4V (C-D) were taken using an immersion oil 100x objectiv e. The images shown are re presentative of at least 3 independent repetitions of the experiment.

PAGE 25

Prudencio and Borchelt, 2011 Suppl. Fig. 6. WT W T + Dig. A4V A 4V + Dig 0 2.0106 4.0106 5.0107 1.0108 1.5108Media P2 S1 Band intensity levels of SOD1 protein (A.U.) S1 P2 Digitonin mediaA*#*B Suppl. Fig. 6. Digitonin extracts a po rtion of the detergentsoluble hSOD1 protein expressed in HEK293FT cells. Digitonin released WT SOD1 into medium with a corresponding decr ease in the amount of WT SOD1 in the S1 fraction of digitonin treated ce lls. In cells expressing A4V SOD1, dig itonin released mutant protein into the media, but the level of mutant pr otein in the S1 and P2 fractions of treated cells was not statistically different from that of untreated ce lls. A) Immunoblot of HEK293FT cells transfected with th e indicated hSOD1 constructs for 48 hours. Prior harvest, cells were incubated 30 minutes in 1x PBS with 0.01% digitonin (Dig.). Then the harvested cell pellets were analyzed by deterg ent extraction and centrifuga tion assays. The amount of hSOD1 protein released into the cell media was also evaluated. B) Qu antification of SOD1 protein found in the Media, P2 and S1 fractions of cells Symbols over bars indica te statistical differences from the corresponding untreated control. The data were averaged from 3 i ndependent transfection experi ments for each construct and condition. Unpaired t -tests: p 0.05; #p 0.005.

PAGE 26

Prudencio and Borchelt, 2011 Suppl. Figure 7 A4V::YFP WT::YFP YFPSOD1 EtBrMerge Supplemental Figure 7. EtBr uptake assay of cell permea bility in HEK293FT cells expressing mutant and WT YFP fusion proteins. Cells were transfected for 24 hours, a nd then incubated with the EtBr cell permeability reagent (see main text). Images were captured on live cells on an inverted fluorescence microscope at 40X.

PAGE 27

Prudencio and Borchelt, 2011 Suppl. Fig. 8. YFPTUNELMergeA4V::YFP A4V::YFP WT::YFP UTf A B C D Suppl. Fig. 8. Assessment of cell death in CHO cells ex pressing WT and A4V SOD::YFP fusion proteins. At 48 hours post-transfection, we found that relatively few cells were TUNEL positive and only rarely did we observe labeling of cells that ex pressed A4V:YFP SOD1 fu sion proteins (Rows A and B). A similar low frequency of TUNEL positive cells were seen in cells expressing WT::YFP SOD1 fusion proteins (Row C) and untransfected cells (Row D). As a positive control, cells treated with staurosporin e demonstrated robust TUNEL labeling (Row D, inset far right panel).

PAGE 28

Prudencio and Borchelt, 2011 Supplementary Table 1. Summary of findings with conformationally-r estricted antibodies. Antibody Model Technique Findings Reference hSOD1 G93A Frozen, 4% paraformaldehyde, No antigen retrieval Intense labeling of motor neuron cell bodies and neuritis in gray matter. [51] hSOD1 G37R, L126Z Frozen, 4% paraformaldehyde, No antigen retrieval Intense labeling of motor neuron cell bodies and neuritis in gray matter. [26] hSOD1 H46R/H48Q, WT Paraffin, 4% paraformaldehyde, antigen retrieval microwave heat In both mutant and WT mice, diffuse labeling of neuropil, no obvious labeling of cell bodies. Weak labeling of inclusionlike structures in mutant mice. [42] hSOD1 PrP vector G37R mice Paraffin, 4% paraformaldehyde, antigen retrieval microwave heat In young presymptomatic mice, intense labeling of motor neuron cell bodies, neuritis of neuropil. [52] hSOD1 L126Z Paraffin, 4% paraformaldehyde, antigen retrieval microwave heat; Frozen, 4% paraformaldehyde, Progressive accumulation of immunoreactivity in cell bodies and neuritis of neuropil. [53] C4F6 Human sALS Paraffin, fixation unknown Intense labeling of motor neuron cell bodies, some cases of neuropil labeling. [27] C4F6 hESCderived motor neurons In situ preparation Cells expressing G93A labeled – showing punctate structures. No labeling of cells expressing WT, A4V, or I113T mutants. [41]

PAGE 29

Prudencio and Borchelt, 2011 C4F6 G93A mouse Frozen, 4% paraformaldehyde, No antigen retrieval described Intense labeling of cell bodies and neuropil. Some structures reported to resemble inclusions. [39] SEDI G37R G85R, G93A mice, Human A4V fALS Paraffin, 4% paraformaldehyde, Peroxide/citrate pretreatment, no heat Mice Intense immunoreactivity at margins of vacuolar pathology in G37R and G93A mice. Reactivity with inclusion structures in G85R mice. Overall reactivity with cell bodies limited in all cases. Humans – Examples of reactivity in cell body, round inclusions in neuropil. [17] SEDI Human SOD1-fALS, nonSOD1fALS, sALS Paraffin, formalin, Antigen retrieval Citrate with heat Intense immunoreactivity with inclusion structures of motor neurons from SOD1linked fALS cases (A4V, I113T, A4T, deleted G27/P28, V14M). No reactivity in ALS patients lacking SOD1 mutations. [43] m/h SOD1 G37R Paraffin, 4% paraformaldehyde, antigen retrieval microwave heat Diffuse neuropil labeling, intensification around vacuoles, some cell body labeling, rare inclusion-like structures. [15,54] m/h SOD1 NonTg mouse Frozen, 4% paraformaldehyde, No antigen retrieval Intense labeling of cell bodies, neurites. [55]

PAGE 30

Prudencio and Borchelt, 2011 m/h Sod1 G37R, WT Paraffin, 4% paraformaldehyde, antigen retrieval microwave heat In G37R mice – intense labeling at margins of vacuolar pathology. White matter tracts labeled. Limited labeling of grey matter neuropil or cell bodies. In WT over-expressing miceGeneral neuropil labeling, intensification of white-matter tracts. Limited intensification of cell bodies as compared to non-transgenic animals [56] m/h SOD1 G85R Paraffin, 4% paraformaldehyde, antigen retrieval – microwave heat Intense reactivity with inclusions in motor neuron cell bodies. [4] m/h SOD1 G85R and G93A mice; human SOD1-fALS –L126 truncation Paraffin, 4% paraformaldehyde, antigen retrieval – microwave heat Mice – intense reactivity with inclusions in motor neuron cell bodies and neuropil. Human – increase reactivity with neuropil inclusions [5]