Chaperone BAG6 is dispensable for MHC class I antigen processing and presentation
Annegret Bitzer
a, Michael Basler
a,b, Marcus Groettrup
a,b,∗aDivisionofImmunology,DepartmentofBiology,UniversityofKonstanz,D-78457Konstanz,Germany
bBiotechnologyInstituteThurgau(BITg)attheUniversityofKonstanz,CH-8280Kreuzlingen,Switzerland
Keywords:
BAG6
Antigenprocessing Antigenpresentation MHCclassI
a b s t r a c t
AntigenprocessingfordirectpresentationonMHCclassImoleculesisamultistepprocessrequiring theconcertedactivityofseveralcellularcomplexes.Theessentialstepsatthebeginningofthispathway, namelyproteinsynthesisattheribosomeanddegradationviatheproteasome,havebeenknownforyears.
Nevertheless,thereisaconsiderablelackoffactorsidentifiedtofunctionbetweenproteinsynthesisand degradationduringantigenprocessing.Here,weanalyzedtheimpactofthechaperoneBAG6onMHCclass Icellsurfaceexpressionandpresentationofvirus-derivedpeptides.AlthoughanessentialroleofBAG6 inantigenprocessinghasbeenproposedpreviously,wefoundBAG6tobedispensableinthispathway.
Still,interactionofBAG6andthemodelantigentyrosinasewasenhancedduringproteasomeinhibition pointingtowardsaroleofBAG6inantigendegradation.Redundantchaperonepathwayspotentiallymask thecontributionofBAG6toantigenprocessingandpresentation.
1. Introduction
Antigen processing and presentation in the MHC class I restricted pathway is a process comprising severalsubsequent stepsofpolypeptideprocessingandtransportation.Generally,pro- teinsare degradedtopeptidesby theproteasome,loadedonto MHCclassImolecules,andpresentedasacomplextocytotoxic Tlymphocytes(CTLs)atthecellsurface(RockandGoldberg,1999).
Basedonthediscoveryofanintimateconnectionbetweenantigen presentationandproteinsynthesisYewdellandcolleaguespostu- latedthedefectiveribosomalproduct(DRiP)hypothesis(Yewdell etal.,1996).DRiPsarethoughttobethemainsourceofantigenand aredefinedasrapidlydegradednascentpolypeptidesduetotheir inabilitytoachieveafunctionalstate.AlthoughtheDRiPhypothesis waspostulatedalmosttwodecadesagotherearestillconsider- ablegapsintheknowledgeaboutantigenprocessingupstreamof proteasomaldegradation.Interestingly,areportbyMinamietal.
(2010)foundthechaperoneBCL2-associatedathanogene6(BAG6;
also named Scythe or BAT3) tobe essential for DRiP degrada- tionandMHCclassIcellsurfaceexpression(Minamietal.,2010).
BAG6isconservedinhighereukaryotes,ubiquitouslyexpressed, andencodedintheMHCclassIIIlocus(Banerjietal.,1990;Ozaki etal., 1999;Wangand Liew, 1994).Eventhough BAG6isclas-
∗ Correspondingauthorat:DepartmentofBiology,UniversityofKonstanz,Uni- versitaetsstrasse10,D-78457Konstanz,Germany,Fax:+497531883102.
E-mailaddress:Marcus.Groettrup@uni-konstanz.de(M.Groettrup).
sifiedasachaperone,it hasnoapparentfoldingactivity(Wang etal.,2011).TogetherwithitsbindingpartnersTRC35andUBL4A itassociateswithlonghydrophobicpatchesinclientproteinsand preventsaggregation(Leznickietal.,2013;Mariappanetal.,2010;
Minamietal.,2010;Wangetal.,2011).Therefore,BAG6hasalso beendescribedasa“holdase”whichkeepsitssubstrateinasoluble state(Wangetal.,2011).Becauseofitsbindingproperties,BAG6is especiallyinvolvedinthebiogenesisanddegradationofpolypep- tidescarryinghydrophobicdomains(Kawaharaetal.,2013;Leeand Ye,2013).
Inthisstudy,werevisitedtheroleofBAG6inMHCclassIanti- genprocessingproposedbyMinamietal.(2010).However,we foundnoinfluenceofBAG6knockdownonMHCclassIcellsurface expression.Moreover,BAG6wasnot transcriptionallyregulated afterstimulationwithIFN-␥andpresentationofvirus-derivedpep- tideswasnot alteredinBAG6knockdowncells. Together,these resultssuggestthatBAG6isdispensableorredundantforantigen processingandpresentationonMHCclassI.
2. Materialsandmethods
2.1. Mice,viruses,celllines,andcytokines
C57BL/6mice(H-2b),BALB/cmice(H-2d)andAADmice(H-2b, transgenicforanHLA-A*0201/H-2Ddchimericprotein(Newberg etal.,1996))wereoriginallypurchasedfromCharlesRiver.Mice werekeptin a specificpathogen-free facilityand usedat 8–12
Konstanzer Online-Publikations-System (KOPS) URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-0-322750
Erschienen in: Molecular Immunology ; 69 (2016). - S. 99-105 https://dx.doi.org/10.1016/j.molimm.2015.11.004
weeksofage.Animalexperimentswereapprovedbythereview boardofRegierungspräsidiumFreiburg.
LCMV-WE was originally obtained from F. Lehmann-Grube (Hamburg, Germany) and propagated on the fibroblast line L929.VacciniavirusWesternReservestrain(VV-WR)wasorig- inallyobtainedfromH. Hengartner (UniversityHospitalZürich, Switzerland).Recombinantvacciniavirus(rVV)expressingtyrosi- nase(rVV-Tyr)wasakindgiftfromV.H.Engelhard(Universityof Virginia,VA,USA).Bothvacciniavirusstrainswerepropagatedon BSC-40cells.
B8-Db(H-2d)isamurinefibroblastcelllinestablytransfected withH-2Db (Basler et al., 2004). B8-Db cells were cultured in IMDM. Primary murine embryonic fibroblasts from AAD mice, HeLacells, HEK293cells, andMelJuSocells weremaintainedin DMEM.ThehumanlungfibroblastcelllineIMRSwasmaintained inMEMincluding15%FCS.MelJuSocellsandIMRScellswerea kindgiftfromN.Koch(UniversityofBonn,Germany).Allmedia werepurchasedfromInvitrogen-LifeTechnologiesandcontained GlutaMAX,10%FCSand100U/mlpenicillin/streptomycin.
Recombinanthumanandmurineinterferon-␥(IFN-␥)waspur- chasedfromPeprotech(Hamburg,Germany)andusedat200U/ml fortwodays.
2.2. Transfections
Knockdown of BAG6 was performed through simultaneous transfectionoffourdifferentsiRNAstargetingBAG6.At80%con- fluence,cellsweretransfectedwithmouseorhumanBAG6siRNA (ON-TARGETplus SMART pool siRNA, Thermo Scientific) using DharmaFECT1transfectionreagent(ThermoScientific)according to the manufacturer’s instruction. After 48h, cells were trans- fectedasecondtimetoachievemaximumefficiency.Controlcells weretransfectedwitha mixof fournon-targetingsiRNAs(ON- TARGETplusNon-TargetingpoolsiRNA,ThermoScientific).
For transient overexpression of BAG6, HeLa cells were transfected with a pCMV6-Entry expression plasmid encoding BAG6-Myc-DDK(OriGeneTechnologies).Controlcellsweretrans- fected with empty vector. Transfection was performed at 80%
confluence using FuGENE HD transfection reagent (Promega) according to the manufacturer’s instructions. Cells were fur- theranalyzed24hposttransfection.Transientoverexpressionof tyrosinase(Tyr) inHEK293 cells wasachievedlikewise using a pCMV6-Entryexpressionconstructencodingtyrosinase-Myc-Flag andTrans-IT-LT1transfectionreagent(MirusBioLLC.)accordingto themanufacturer’sinstructions.
2.3. Flowcytometry
MHCclassIsurfaceexpressionwasanalyzedbyflowcytom- etryusingfluorescently-labeledantibodies.Cellswereincubated togetherwithantibodiesdiluted1:150inPBS,2%(v/v)FCS,2mM EDTAfor20minat4◦C.Cellswerewashedandacquiredwiththe useoftheAccuriC6flowcytometersystem(BDBiosciences).Data wereanalyzedwithFlowJosoftware(TreeStar).Anti-H-2DbFITC (cloneKH95),anti-H-2DdPE(clone43-2-12)andanti-HLA-ABCAPC (cloneG46-2.6)antibodieswereobtainedfromBDBiosciences.
2.4. Real-timeRT-PCR
Total RNA was extracted using NucleoSpin RNA II kit (Macherey-Nagel) according tothe manufacturer’s instructions.
ExtractedRNAwasreverse-transcribed usingtheReverseTran- scription System (Promega) according to the manufacturer’s instruction.Quantitative real-time RT-PCRwas performedwith the LightCycler Instrument II (Roche) and LightCycler Fast- Start DNA Master SYBR Green I kit (Roche). Gene expression
was calculated relative to control cells and normalized to HPRT(hypoxanthineguaninephosphoribosyltransferase)expres- sionusingREST-384Osoftwareversion2 (GeneQuantification).
Sense and antisense primers used for amplification were as follows:BAG6(human)TACAATAACAATCACGAGGGCC,GGTGGT- GTAGTGAGACATAGG;TRC53(human)CGTGACCTTTGTGTCCAGAG, TTAAACAGAGAAACTGTAGCACGG;HPRT(human)TGGACAGGACT- GAACGTCTTG, CCAGCAGGTCAGCAAAGAATTTA; LMP7 (human) AATGCAGGCTGTACTATCTGCG,TGCAGCAGGTCACTGACATCTG.
2.5. SDSPAGEandwesternblot
CellswerelysedinRIPAbuffer(150mMNaCl,50mMTrispH8, 1%(v/v)TritonX-100,0.5%(v/v)sodiumdesoxycholate,0.1%(w/v) SDS)includingproteaseinhibitors(cOmpleteEDTA-free, Roche) for30minonice.Lysateswerecentrifugedat20,000×gand4◦C for15min.SupernatantsweremixedwithSDSsamplebufferand boiledfor5minat95◦C.ProteinswereseparatedbySDS-PAGEand blotted onto nitrocellulose membrane (Whatman). Membranes were blockedfor 1hin Roti-Block solution (Roth) followed by overnightincubation in primary antibodiesat 4◦C. Membranes werewashedandincubatedfor2hwithappropriateperoxidase- conjugatedsecondaryantibodies(Dako).Membraneswerewashed andproteinswerevisualizedwithenhancedchemiluminescence.
Primaryantibodiesused:Rabbit anti-BAG6(kindgift fromR.S.
Hegde,Cambridge,UK),mouseanti-␣-tubulin(cloneAA13,Sigma) andmouseanti-FLAG(cloneM2,Sigma).
2.6. Generationofepitope-specificcytotoxicTlymphocyte(CTL) lines
LCMV-specificCTLlinesweregeneratedfromfemaleC57BL/6 orBALB/cmiceinfectedi.v.with200PFULCMV-WE.Fourweeks postinfectionsplenocytes from infected micewerecultured in IMDM 10% FCS, P/S, supplemented with 40U/ml IL-2, 100M 2-mercaptoethanoland10−6Mpeptide.Cytokine-supplemented mediumwasaddedeveryotherdayfor8–14days.BeforeCTLs wereusedinpresentationassays,deadcellswereremovedbyFicoll densitycentrifugation.
Tyr-specific CTL-lines were generated from male AAD mice infectedi.p.with2×106PFUrVV-Tyr.Splenocytesfrommemory micewereculturedforthreeweeksasdescribedabove.Toincrease percentageofspecificTcells,CTLswererestimulatedtwicewith peptide-loaded,irradiated(20Gy)AADsplenocytesonday8and 16.
2.7. Syntheticpeptides
ThesyntheticpeptidesGP276-286(SGVENPGGYCL),NP118-126 (RPQASGYVM)andNP396-404(FQPQNGQFI)wereobtainedfromP.
Henklein(Charité,Berlin,Germany).ThesyntheticpeptideTyr369- 377(YMDGTMSQV)wasobtainedfromSigma.
2.8. Antigenpresentationassay
B8-DbcellsorAADmouseembryonicfibroblasts(MEFs)were transfectedwithBAG6siRNAasdescribedabove.Transfectedcells wereharvestedandinfectedwithLCMVorrVVwithamultiplic- ityofinfection(MOI)of10for 2h,4h, or5h. Serialdilutionof infectedcellswasperformedtoachievedifferenteffectortostimu- latorratios(E/S).Epitope-specificCTLlineswereaddedtoinfected cellsandincubatedinpresenceof10g/mlbrefeldinAfor3hat 37◦C. Activationof CTLswas determinedbyintracellular IFN-␥ stainingandflowcytometry.Allsampleswerepreparedasdupli- cates.
2.9. IntracellularIFN-staining
IntracellularIFN-␥stainingwasperformedasdescribedprevi- ously(Basleretal.,2004)exceptthatanti-CD8aAPC(clone53-6.7, eBioscience)(1:150)wasusedtostaincytotoxicTcells.
2.10. Immunoprecipitation
Cellstransientlytransfected withanexpressionconstructfor tyrosinase-Myc-Flagormocktransfectedcellswereincubatedwith 10MMG132(Sigma)orDMSO(1:1000)for5hat37◦C.Cellswere thenharvestedandlysedinlysisbuffer(20mMTrispH7.8,50mM NaCl,10mMMgCl2,1%(v/v)NP40)includingproteaseinhibitors (cOmpleteEDTA-free,Roche)for30minonice.Lysateswerecen- trifugedat20,000×gand4◦Cfor15min.Supernatantswereloaded ontoproteinGaffinitygel(Sigma)togetherwithanti-Mycantibody (clone9E10,Sigma)andincubatedovernightat4◦Conarotator.
Theaffinitygelwaswashedthreetimesinhighsaltbuffer(50mM TrispH8,650mMNaCl,5mMEDTA,0.5%(v/v)TritonX-100)and threetimesinlowsaltbuffer(50mMTrispH8,150mMNaCl,5mM EDTA,0.5%(v/v)TritonX-100).Then,precipitatedproteinswere
heatedinSDSsamplebufferfor5minat95◦Candsubsequently analyzedbySDS-PAGEandwesternblotasdescribedabove.
2.11. Statisticalanalysis
Theunpairedtwo-tailedStudent’sttestwasappliedforstatis- ticalanalysisusingGraphPadPrismsoftware.
3. Results
3.1. MHCclassIsurfaceexpressionisindependentofBAG6
IfDRiPsresultfromincompletecaptureofnascentpolypeptides bychaperones,deletionofchaperonesshouldincreaseDRiPrates.
AhigherproportionofDRiPsshouldinturnincreaseMHCclassI surfaceexpressionduetoincreasedpeptidesupply.Ontheother hand,deletionofchaperonescouldleadtoinefficientshuttlingand degradationofantigen.ToanalyzetheinfluenceofBAG6onMHC classIsurfaceexpression,themurinefibroblastcelllineB8-Db(H- 2d+H-2Db)andHeLacells weresubjectedtoBAG6knockdown.
Subsequently, surfaceexpressionof MHC classImolecules was
Fig.1. KnockdownoroverexpressionofBAG6hasnoinfluenceonMHCclassIsurfaceexpression.(AandC)B8-Dbcells(A)andHeLacells(C)weretransfectedwithBAG6 siRNAorcontrolsiRNAandsurfaceexpressionofMHCclassImoleculesH-2DbandH-2Dd(A)orHLA-A,B,C(C)wasanalyzedbyflowcytometry.(BandD)Westernblotanalysis ofwholecelllysatespreparedfromB8-Dbcells(B)orHeLacells(D)transfectedwithBAG6siRNAorcontrolsiRNA.(E)HeLacellsweretransfectedwithaBAG6expression constructoranemptyplasmid(control)andsurfaceexpressionofMHCclassImoleculesHLA-A,B,Cwasanalyzedbyflowcytometry.(F)Westernblotanalysisofwholecell lysatespreparedfromHeLacellstransfectedwithaBAG6expressionconstructoranemptycontrolplasmid.Tubulinwasusedasloadingcontrol.Forallexperimentsone outoftwoindependentexperimentsisshown.
Fig.2.BAG6isnottranscriptionallyregulatedafterIFN-␥stimulation.HeLacells(A),IMRScells(B)orMelJuSocells(C)werestimulatedwithIFN-␥for24hor48horwere leftunstimulatedandmRNAlevelsofBAG6andLMP7weredeterminedbyreal-timeRT-PCR.Expressionlevelswerecalculatedrelativetounstimulatedcellsandnormalized tohypoxanthineguaninephosphoribosyltransferase(HPRT)expression.Bargraphsshowmeans±SDfromthreeindependentexperiments.ns(notsignificant)P>0.05,
**P<0.01,***P<0.001(unpairedtwo-tailedStudent’sttest).
determinedbyflowcytometry.Nodifferenceinsurfaceexpression ofthetransfectedMHCclassImoleculeH-2Dbandtheendogenous moleculeH-2Dd could beobserved between B8-Db cells trans- fectedwithBAG6siRNAandcellstransfectedwithcontrolsiRNA (Fig.1A).SuccessfulknockdownleadingtoablationoftheBAG6 proteinbelow thedetectionlimit ofwesternanalysiswascon- firmedbySDS-PAGEandimmunoblotting(Fig.1B).Thesameresult wasobtainedwithHeLacellsinwhichtotalMHCclassIsurface expression(HLA-A,B,C)didnotdifferbetweenBAG6knockdown andcontrolcells(Fig.1C).VerysuccessfulknockdowninHeLacells wasagainconfirmedbySDS-PAGEandimmunoblotting(Fig.1D).
Torule out limiting effects of BAG6onMHC class I surface expression,overexpressionwasperformedinHeLacells.Cellswere transientlytransfectedwithanexpressionconstructencodingfor BAG6andsubjectedtoflowcytometricanalysis24hposttransfec- tion(Fig.1E).NoinfluenceonHLA-A,B,Csurfaceexpressioncould bedetectedalthoughlysatesoftransfectedcellssubjectedtoSDS- PAGEandimmunoblottingrevealedveryhighexpressionlevelsof BAG6comparedtocontrolcells(Fig.1F).
Takentogether,wecouldnotobserveaninfluenceofBAG6on MHCclassIsurfaceexpression.LossofBAG6mightbeeffectively compensatedbyothercytosolicchaperonesinthesettingapplied here.Ontheotherhand,BAG6isnotalimitingfactorinantigen
processingsinceitsoverexpressionhadnoeffectonMHCclassI surfaceexpression.
3.2. BAG6isnottranscriptionallyregulatedafterinterferon gamma(IFN-)stimulation
StimulationofcellswiththecytokineIFN-␥inducesorenhances transcriptionofseveralgeneswhoseproductsdelayviralreplica- tionandenhancepeptidepresentationonMHCclassI(Platanias, 2005;Senand Lengyel,1992).Like many ofthesegenes, BAG6 isencodedin theMHC locusand cytokineregulatedexpression controlcouldbepossible(Banerjietal.,1990).Yet,stimulationof HeLacellswithIFN-␥for24hor48hdidnotinduceupregulation ofBAG6mRNA(Fig.2A,leftside).Incontrast,expressionofthe immunoproteasomesubunitLMP7,a well-known IFN-regulated gene, wasstrongly upregulated afterstimulation (Fig. 2A,right side).AlthoughKamperetal.foundexpressionofthebag6geneto beupregulatedafterstimulationofthemelanomacelllineMelJuSo andthelungfibroblastcelllineIMRSwithIFN-␥wecouldneither confirmthisfindingforHeLacells,norfortheothertwocelllines (Fig.2BandC)(Kamperetal.,2012).Overall,ourresultsdonot pointtowardsacentralroleofBAG6inatypeIIinterferonregulated antiviralimmuneresponse.
Fig.3.BAG6isdispensableforpresentationofvirus-derivedepitopes.(AandB)CellsweretransfectedwithBAG6siRNAorcontrolsiRNAandin(A)infectedwithlymphocytic choriomeningitisvirus(LCMV),(MOI=10).Epitope-specificCTLswereaddedtoinfectedcellsindifferenteffectortostimulatorratios(E/S)for3handthepercentageof IFN-␥+ofCD8+cellswasdeterminedbyflowcytometryasameasureforepitopepresentation.Valuesrepresentmean±SDofduplicatedeterminations.(A)B8-Dbcells infectedwithLCMVfor2,4or5hwereanalyzedforpresentationofNP396-404,GP276-286orNP118-126.(B)TransfectedMEFsfromAADmicewerestimulatedwithIFN-␥
for2days,infectedfor5hwithrVV-TyrorVV-WRandthenanalyzedforthepresentationofTyr369-377(D).(C)WesternblotanalysisofwholecelllysatespreparedfromAAD MEFstransfectedwithBAG6siRNAorcontrolsiRNAandstimulatedwithIFN-␥for2days.Tubulinwasusedasloadingcontrol.(D)HEK293cellsweretransientlytransfected
3.3. PresentationofLCMV-derivedepitopesisindependentof BAG6
Toaddresswhethergenerationofvirus-derivedpeptideepi- topesisdependentonBAG6,weinvestigatedantigenpresentation inlymphocyticchoriomeningitisvirus(LCMV)-infectedcellsafter BAG6knockdown.B8-DbcellsweretransfectedwithBAG6siRNA andsubsequentlyinfectedwithLCMV-WEfor2,4or5h.Peptides presentedonMHCclassIweredetectedwithpeptide-specificCTL linesspecificfortheepitopesnucleoprotein(NP)118–126(H-2Ld), NP396-404 (H-2Db)and glycoprotein(GP) 276–286 (H-2Db)by stainingfortheCTLsurfacemarkerCD8andintracellularIFN-␥.
CTLsproduceIFN-␥uponstimulationoftheirreceptorwiththe appropriatepeptide-MHCcomplex.Thus,theproportionof IFN-
␥producingCTLsofallCD8bearingCTLsreflectstheamountof peptidespresentedbyinfectedcells.GP276-286representsanepi- topederivedfromthesecretorypathway,whereasNP118-126and NP396-404arederivedfromacytosolicprotein.Comparedtocon- trolcells,BAG6knockdowncellsarenotconfinedintheirabilityto presentLCMV-derivedepitopes(Fig.3A).
3.4. Presentationofretrotranslocatedtyrosinase-derivedepitope isnotaffectedbyBAG6knockdown
BAG6maintainsretrotranslocatedER-associatedproteindegra- dation(ERAD) substratesin a soluble state and thereby allows proteasomaldegradationofthesesubstrates(ClaessenandPloegh, 2011;Wangetal.,2011).Tostudythepresentationofanepitope derivedfromaretrotranslocatedproteinwechosethetyrosinase epitopeTyr369-377(D).Tyrosinaseisatransmembraneglycopro- teinproducedinmelanocytesandmelanomacellsthatistargeted totheER(Jimbowetal.,2000;Petrescuetal.,2000).Misfolded tyrosinaseisexportedbacktothecytosolwhereitisdeglycosylated leadingtotheconversionofN371toD371throughdeamination (Skipperetal.,1996).TheHLA-A*0201-restrictedepitopeTyr369- 377(D)spansacrossthismodifiedaminoacidandallowsanalysis ofanepitopederived exclusivelyfromaretrotranslocatedanti- gen.TostudypresentationofTyr369-377(D),wemadeuseofthe transgenicmousestrainAAD(H-2b,HLA-A2.1/H2-Dd)expressinga hybridMHCclassImoleculeconsistingofthealpha-1andalpha- 2domainofhumanHLA-A*0201andthealpha-3transmembrane andcytoplasmicdomainofmouseH-2Dd(Newbergetal.,1996).
AADmouseembryonic fibroblasts(MEFs)weretransfectedwith BAG6siRNA and stimulatedwith IFN-␥ to induce MHC class I expression.Cellswerethen infected withrecombinantvaccinia virusexpressingtyrosinase(rVV-Tyr)for5h.Tyr369-377(D)-MHC complexesweredetectedwithspecificCTLlinesgeneratedfrom AAD mice and analyzed by staining for CD8 and intracellular IFN-␥. BAG6knockdowndidnot alter Tyr369-377(D) presenta- tion (Fig. 3B) although no detectable BAG6 protein remained afterknockdown (Fig.3C). While BAG6seemed to bedispens- ableforTyr369-377(D)presentation,endogenousBAG6couldbe pulleddownwithtransientlyexpressedtyrosinaseinHEK293cells (Fig.3D).Thisinteractionwasevenenhancedwhenproteasomal degradationwasblockedbyMG132treatmentindicatingaroleof BAG6intyrosinasedegradation.
Overall,wefoundnoevidenceforanessentialroleofBAG6in generatingLCMV-ortyrosinase-derived ligandsfor thepresen- tationonMHCclassI.However,wecannotformallyexcludethe possibilitythatotherepitopesmightdependonBAG6foreffective processingandpresentation.
4. Discussion
Antigenprocessingandpresentationhavecrucial impactson theefficiencyandoutcomeofanimmuneresponse.Apossibilityto rapidlydetecttranslationofforeignproteinswithinacellistheuse ofDRiPsassubstratesforantigenprocessing.Asaconsequence,the
“immunopeptidome”surveyedbyCTLsmirrorstranslationrates ratherthanproteinconcentrations(Caronetal.,2011;Fortieretal., 2008;Khanetal.,2001).Bourdetskyandcolleaguescouldindeed showthatpeptidespresentedonMHCclassIshiftedfasterfrom lighttoheavyisotopeformsthantheirsourceproteinswhencells wereswitchedtomediacontainingaminoacidswithheavyiso- topes(Bourdetskyetal.,2014).Thisraisesthequestionwhether DRiPsandDRiP-derivedpeptidesaresimpleby-productsofpro- teinsynthesisorifcellshaveevolvedmechanismsregulatingDRiP formationand/orprocessing.Yewdell andNicchitta(2006)sug- gested the existenceof “immunoribosomes” specialized onthe production of substratesfor the MHC class Irestricted antigen processingpathway(YewdellandNicchitta,2006).Suchimmunori- bosomesmayhaveadifferentsubunitcompositionorlocalizeina differentintracellularcompartmentthanconventionalribosomes.
Althoughintellectuallyattractive,thishypothesishasnotyetbeen confirmedexperimentally.Instead,ribosomesaremorelikelyto functionasaplatformforassociatedfactorsactinguponthenascent chainandinfluencing itsfate(Krameret al.,2009).Whilemore andmorefactorsarebeingidentifiedtomediateco-translational andpost-translationalproteinqualitycontrollittleisknownabout thecontributionof thesefactors toantigen processing.Specifi- callydesigned torecognize non-nativepolypeptides chaperones are amongst the candidates ideally suited for DRiP processing.
Chaperonesmediateanimportantlayerofqualitycontrolandcan facilitatedegradationof non-nativepolypeptidesthrough inter- actionwithfactorsoftheubiquitin-proteasomesystem(Hohfeld etal.,2001;Rodrigo-BrenniandHegde,2012).
The chaperone BAG6is involved in post-translational inser- tion of tail-anchored proteins into the ERmembrane (Leznicki etal., 2010;Mariappan et al.,2010).While assistingin protein biosynthesis ontheone hand,BAG6alsomediates degradation ofmislocalizednascentchains(Hessaetal.,2011)andERADsub- strates(ClaessenandPloegh,2011;Claessenetal.,2014;Payapilly andHigh,2014;Wangetal.,2011).Moreover,Rodrigo-Brennietal.
(2014)recentlydescribedRNF126asanE3ligaseinteractingwith BAG6andspecificallyubiquitylatingchaperoneboundclientpro- teins(Rodrigo-Brennietal.,2014).ThefactthatBAG6hasbeen describedasa“holdase”conferringnoevidentfoldingactivityonto itssubstratesmakesthis chaperoneevenmoreinteresting with regard toproteinquality controland degradation (Wang etal., 2011).
Here,weaddressedwhetherBAG6caninfluenceantigenpro- cessingand presentationintheMHCclassIrestrictedpathway.
Knockdownof BAG6had noinfluence onMHC classI cellsur- face expression indicating that loss of BAG6 can be effectively counterbalancedbyothercytosolicchaperonesunderphysiological conditions.ThisresultisincontrasttoapreviousreportbyMinami etal.(2010)whofoundreducedMHCclassIcellsurfaceexpression inBAG6knockdowncells(Minamietal.,2010)butinagreement withamorerecentstudybyKochandcolleagueswhodidnotfind suchaneffect(Kamperetal.,2012).Wehavecurrentlynoexplana- tionforthisdiscrepancy,especially,sincewecouldnotdetectany residualBAG6expressionaftertransfectionwithBAG6siRNA.Pre- sentationofthreeLCMV-derivedepitopeswasentirelyunaffected
withanexpressionconstructfortyrosinase(Tyr)-Myc-Flagoremptyplasmid(control)andtreatedwith10MMG132orsolventDMSOfor5h.Celllysatesweresubjected toimmunoprecipitation(IP)withanti-Mycantibodyfollowedbywesternblot(WB)analysiswithBAG6orFlagspecificantibodies.Loadindicatesproteinexpressionlevels intotalproteinlysates.Forallexperimentsoneoutoftwoindependentexperimentsisshown.
althoughoneoftheepitopesanalyzed(GP276-286)derivesfrom anER-targetedglycoprotein.Hence,degradationoftheGP276-286 precursorcouldbehandledbytheERADpathwayorthemislocal- izedproteindisposalpathwayinbothofwhichBAG6isinvolved.
Strikingly,thetyrosinase-derived epitopeTyr369-377(D), which strictly requires retrotranslocation from the ER in order to be presented,wasalsonotaffectedbyBAG6knockdown.Together, theseresultsindicate thatBAG6isdispensableforantigenpro- cessingandpresentationonMHCclassI.Enhancedinteractionof BAG6and tyrosinaseunderproteasome inhibitionontheother handstillindicatesthatBAG6couldbeapotentialplayerinanti- genprocessing.Theredundancyincellularchaperonenetworks could,however,maskthecontributionofBAG6tothispathway.
BAG6togetherwithitsE3 ligaseRNF126indeed resemblesthe HSP70/HSP90chaperonesystem,whichcooperateswiththeE3lig- aseCHIPtopromotesubstratedegradation(Connelletal.,2001;
Rodrigo-Brennietal.,2014).ThefindingthatBAG6substratescould stillbedegradedin theabsenceofBAG6and/orRNF126,albeit moreslowlyandlessefficiently,furtherindicatestheexistenceof oneorseveralredundantpathways(Rodrigo-Brennietal.,2014).
Moreover,BAG6andHSP70seemtocompeteforsubstratebind- ingasBAG6couldpreventHSP70-mediatedrefoldingofdenatured luciferaseinaninvitroassay(Wangetal.,2011).AlthoughBAG6 preferablybindstolongerhydrophobicpatches(Mariappanetal., 2010)andHSP70/HSP90toshorterhydrophobicpatches(Jackson, 2013;Rudigeretal.,1997)considerableoverlapinbothsystems islikely.Definitionofthepotentialproportionofantigenthatis handledbyBAG6wouldthereforebepossibleifBAG6-mediated degradationcouldbeinhibitedaftersubstratebindingtothechap- erone.Thereby,antigencouldbetrappedonBAG6anddegradation viaadifferentroutewouldbeblocked.
Compartmentalizedantigenprocessingwasfirstsuggestedby Levetal.(2010)(Levetal.,2010)andcouldbeanotherexplanation forourfindings.Inthiscase,BAG6mightbeinvolvedindegradation ofbulkproteinswhilebeingdispensableforepitopegeneration.
Takentogether,weproposethatBAG6isdispensableforantigen processingandpresentationonMHCclassIbutitmightbecome essentialunderconditionsoflimitedchaperoneavailability.
Conflictofinterest
Theauthorsdeclarenoconflictsofinterest.
Acknowledgments
WethankRamanujanS.Hegde,VictorH.EngelhardandNorbert Kochforsharingantibodies,virusstrainsandcelllineswithus.
References
Banerji,J.,Sands,J.,Strominger,J.L.,Spies,T.,1990.Agenepairfromthehuman majorhistocompatibilitycomplexencodeslargeproline-richproteinswith multiplerepeatedmotifsandasingleubiquitin-likedomain.Proc.Natl.Acad.
Sci.U.S.A.87,2374–2378.
Basler,M.,Youhnovski,N.,vandenBroek,M.,Przybylski,M.,Groettrup,M.,2004.
Immunoproteasomesdown-regulatepresentationofasubdominantTcell epitopefromlymphocyticchoriomeningitisvirus.J.Immunol.173,3925–3934.
Bourdetsky,D.,Schmelzer,C.E.,Admon,A.,2014.Thenatureandextentof contributionsbydefectiveribosomeproductstotheHLApeptidome.Proc.
Natl.Acad.Sci.U.S.A.111,1591–1599.
Caron,E.,Vincent,K.,Fortier,M.H.,Laverdure,J.P.,Bramoulle,A.,Hardy,M.P., Voisin,G.,Roux,P.P.,Lemieux,S.,Thibault,P.,Perreault,C.,2011.TheMHCI immunopeptidomeconveystothecellsurfaceanintegrativeviewofcellular regulation.Mol.Syst.Biol.7,533.
Claessen,J.H.,Ploegh,H.L.,2011.BAT3guidesmisfoldedglycoproteinsoutofthe endoplasmicreticulum.PLoSOne6,e28542.
Claessen,J.H.,Sanyal,S.,Ploegh,H.L.,2014.ThechaperoneBAG6captures dislocatedglycoproteinsinthecytosol.PLoSOne9,e90204.
Connell,P.,Ballinger,C.A.,Jiang,J.,Wu,Y.,Thompson,L.J.,Hohfeld,J.,Patterson,C., 2001.Theco-chaperoneCHIPregulatesproteintriagedecisionsmediatedby heat-shockproteins.Nat.CellBiol.3,93–96.
Fortier,M.H.,Caron,E.,Hardy,M.P.,Voisin,G.,Lemieux,S.,Perreault,C.,Thibault, P.,2008.TheMHCclassIpeptiderepertoireismoldedbythetranscriptome.J.
Exp.Med.205,595–610.
Hessa,T.,Sharma,A.,Mariappan,M.,Eshleman,H.D.,Gutierrez,E.,Hegde,R.S., 2011.Proteintargetinganddegradationarecoupledforeliminationof mislocalizedproteins.Nature475,394–397.
Hohfeld,J.,Cyr,D.M.,Patterson,C.,2001.Fromthecradletothegrave:molecular chaperonesthatmaychoosebetweenfoldinganddegradation.EMBORep.2, 885–890.
Jackson,S.E.,2013.Hsp90:structureandfunction.Top.Curr.Chem.328,155–240.
Jimbow,K.,Park,J.S.,Kato,F.,Hirosaki,K.,Toyofuku,K.,Hua,C.,Yamashita,T., 2000.Assembly,target-signalingandintracellulartransportoftyrosinasegene familyproteinsintheinitialstageofmelanosomebiogenesis.PigmentCellRes.
13,222–229.
Kamper,N.,Franken,S.,Temme,S.,Koch,S.,Bieber,T.,Koch,N.,2012.
Gamma-Interferon-regulatedchaperonegovernshumanlymphocyteantigen classIIexpression.FASEBJ.26,104–116.
Kawahara,H.,Minami,R.,Yokota,N.,2013.BAG6/BAT3:emergingrolesinquality controlfornascentpolypeptides.J.Biochem.153,147–160.
Khan,S.,deGiuli,R.,Schmidtke,G.,Bruns,M.,Buchmeier,M.,vandenBroek,M., Groettrup,M.,2001.Cuttingedge:neosynthesisisrequiredforthe
presentationofaTcellepitopefromalong-livedviralprotein.J.Immunol.167, 4801–4804.
Kramer,G.,Boehringer,D.,Ban,N.,Bukau,B.,2009.Theribosomeasaplatformfor co-translationalprocessing,foldingandtargetingofnewlysynthesized proteins.Nat.Struct.Mol.Biol.16,589–597.
Lee,J.G.,Ye,Y.,2013.Bag6/Bat3/Scythe:anovelchaperoneactivitywithdiverse regulatoryfunctionsinproteinbiogenesisanddegradation.Bioessays35, 377–385.
Lev,A.,Princiotta,M.F.,Zanker,D.,Takeda,K.,Gibbs,J.S.,Kumagai,C.,Waffarn,E., Dolan,B.P.,Burgevin,A.,VanEndert,P.,Chen,W.,Bennink,J.R.,Yewdell,J.W., 2010.CompartmentalizedMHCclassIantigenprocessingenhances
immunosurveillancebycircumventingthelawofmassaction.Proc.Natl.Acad.
Sci.U.S.A.107,6964–6969.
Leznicki,P.,Clancy,A.,Schwappach,B.,High,S.,2010.Bat3promotesthe membraneintegrationoftail-anchoredproteins.J.CellSci.123,2170–2178.
Leznicki,P.,Roebuck,Q.P.,Wunderley,L.,Clancy,A.,Krysztofinska,E.M.,Isaacson, R.L.,Warwicker,J.,Schwappach,B.,High,S.,2013.TheassociationofBAG6 withSGTAandtail-anchoredproteins.PLoSOne8,e59590.
Mariappan,M.,Li,X.,Stefanovic,S.,Sharma,A.,Mateja,A.,Keenan,R.J.,Hegde,R.S., 2010.Aribosome-associatingfactorchaperonestail-anchoredmembrane proteins.Nature466,1120–1124.
Minami,R.,Hayakawa,A.,Kagawa,H.,Yanagi,Y.,Yokosawa,H.,Kawahara,H., 2010.BAG-6isessentialforselectiveeliminationofdefectiveproteasomal substrates.J.CellBiol.190,637–650.
Newberg,M.H.,Smith,D.H.,Haertel,S.B.,Vining,D.R.,Lacy,E.,Engelhard,V.H., 1996.ImportanceofMHCclass1alpha2andalpha3domainsinthe recognitionofselfandnon-selfMHCmolecules.J.Immunol.156,2473–2480.
Ozaki,T.,Hanaoka,E.,Naka,M.,Nakagawara,A.,Sakiyama,S.,1999.Cloningand characterizationofratBAT3cDNA.DNACellBiol.18,503–512.
Payapilly,A.,High,S.,2014.BAG6regulatesthequalitycontrolofapolytopicERAD substrate.J.CellSci.127,2898–2909.
Petrescu,S.M.,Branza-Nichita,N.,Negroiu,G.,Petrescu,A.J.,Dwek,R.A.,2000.
Tyrosinaseandglycoproteinfolding:rolesofchaperonesthatrecognize glycans.Biochemistry39,5229–5237.
Platanias,L.C.,2005.Mechanismsoftype-I-andtype-II-interferon-mediated signalling.Nat.Rev.Immunol.5,375–386.
Rock,K.L.,Goldberg,A.L.,1999.Degradationofcellproteinsandthegenerationof MHCclassI-presentedpeptides.Annu.Rev.Immunol.17,739–779.
Rodrigo-Brenni,M.C.,Gutierrez,E.,Hegde,R.S.,2014.Cytosolicqualitycontrolof mislocalizedproteinsrequiresRNF126recruitmenttoBag6.Mol.Cell55, 227–237.
Rodrigo-Brenni,M.C.,Hegde,R.S.,2012.Designprinciplesofprotein biosynthesis-coupledqualitycontrol.Dev.Cell23,896–907.
Rudiger,S.,Buchberger,A.,Bukau,B.,1997.InteractionofHsp70chaperoneswith substrates.Nat.Struct.Biol.4,342–349.
Sen,G.C.,Lengyel,P.,1992.Theinterferonsystem.Abird’seyeviewofits biochemistry.J.Biol.Chem.267,5017–5020.
Skipper,J.C.,Hendrickson,R.C.,Gulden,P.H.,Brichard,V.,VanPel,A.,Chen,Y., Shabanowitz,J.,Wolfel,T.,Slingluff,C.L.,Boon,T.,Hunt,D.F.,Engelhard,V.H., 1996.AnHLA-A2-restrictedtyrosinaseantigenonmelanomacellsresultsfrom posttranslationalmodificationandsuggestsanovelpathwayforprocessingof membraneproteins.J.Exp.Med.183,527–534.
Wang,Q.,Liu,Y.,Soetandyo,N.,Baek,R.,Hegde,R.S.,2011.Aubiquitin
ligase-associatedchaperoneholdasemaintainspolypeptidesinsolublestates forproteasomedegradation.Mol.Cell42,758–770.
Wang,R.,Liew,C.C.,1994.ThehumanBAT3orthologinrodentsispredominantly anddevelopmentallyexpressedintestis.Mol.Cell.Biochem.136,49–57.
Yewdell,J.W.,Anton,L.C.,Bennink,J.R.,1996.Defectiveribosomalproducts (DRiPs):amajorsourceofantigenicpeptidesforMHCclassImolecules?J.
Immunol.157,1823–1826.
Yewdell,J.W.,Nicchitta,C.V.,2006.TheDRiPhypothesisdecennial:support, controversy,refinementandextension.TrendsImmunol.27,368–373.