PolyglutamylationIsaPost-translationalModificationwitha
BroadRangeofSubstrates
□S
Receivedforpublication,July16,2007,andinrevisedform,November27,2007Published,JBCPapersinPress,November28,2007,DOI10.1074/jbc.M705813200
JuliettevanDijk
?
,JulieMiro
?
,Jean-MarcStrub
§
,BenjaminLacroix
?
,AlainvanDorsselaer
§
,BernardEdde
??
,
andCarstenJanke
?1
Fromthe
?
CentredeRecherchedeBiochimieMacromole′culaire,Universite′sMontpellier2and1,CNRS,34293Montpellier,France,
the
§
LaboratoiredeSpectrome′triedeMasseBioOrganique,CNRS,67087Strasbourg,France,andthe
?
Universite′Paris6,75252
Paris,France
Polyglutamylationisapost-translationalmodificationthat
generateslateralacidicsidechainsonproteinsbysequential
additionofglutamateaminoacids.Thismodificationwasfirst
discoveredontubulins,anditisimportantforseveralmicrotu-
bulefunctions.Besidestubulins,onlythenucleosomeassembly
proteinsNAP1andNAP2havebeenshowntobepolyglutamy-
lated.Here,usingaproteomicapproach,weidentifyalarge
numberofputativesubstratesforpolyglutamylationinHeLa
cells.Byanalyzingaselectionoftheseputativesubstrates,we
showthatseveralofthemcanserveasinvitrosubstratesfortwo
oftherecentlydiscoveredpolyglutamylases,TTLL4andTTLL5.
WefurthershowthatTTLL4isthemainpolyglutamylase
enzymepresentinHeLacellsandthatnewsubstratesofpoly-
glutamylationareindeedmodifiedbyTTLL4inacellularcon-
text.Noclearconsensuspolyglutamylationsitecouldbedefined
fromtheprimarysequenceofthehere-identifiednewsubstrates
ofpolyglutamylation.However,wedemonstratethatglutamate-
richstretchesareimportantforaproteintobecomepolyglu-
tamylated.Mostofthenewlyidentifiedsubstratesofpolyglu-
tamylationarenucleocytoplasmicshuttlingproteins,including
manychromatin-bindingproteins.Ourworkrevealsthatpoly-
glutamylationisamuchmorewidespreadpost-translational
modificationthaninitiallythoughtandthusthatitmightbea
regulatorofmanycellularprocesses.
Onefundamentalaspectofproteomiccomplexitycomes
fromthevariousprocessingeventsthatmanyproteinsundergo
followingtheirsynthesis.Post-translationalmodifications,
suchasphosphorylation,acetylation,andmethylation,are
reversiblemonomodificationsandareknowntofunctionas
switchesfortheactivityofmanyproteins.Polyglutamylationis
areversiblepolymodificationgeneratedbysequentialcovalent
attachmentofglutamicacids(upto20insomecases)toan
internalglutamateresidueofthetargetprotein(1).Thelength
oftheresultingsidechainisregulatedbythebalancebetween
theenzymesthatcatalyzeglutamylation,recentlyidentifiedas
membersofthetubulin-tyrosineligase-like(TTLL)
2
protein
family(2–4),andyetunidentifieddeglutamylaseenzymes(5).
Thus,thismodificationdoesnotonlygenerate“on”and“off”
states,butarangeofsignalsthatmightallowforgradualregu-
lationofproteinfunctions.
TheonlyknowntargetsofpolyglutamylationareH9251-and
H9252-tubulins,thestructuralunitsofmicrotubules(MTs(1,6)),
andthenucleosomeassemblyproteins,NAP1andNAP2(7).
Tubulinsaremodifiedintheiracidic,glutamate-richCtermi-
nus(1),whichisthebindingsiteformostMT-associatedpro-
teins(MAPs;reviewedinRef.8).Tubulinpolyglutamylation
wasthereforeproposedtogeneratefunctionallydivergentMTs
byregulatingtheaffinitybetweenMAPsandMTs(4,9–11).It
wasalsoshowntobeimportantforcentriolestability(12),
axonemalmotility(2,13,14),andneuriteoutgrowth(15).The
roleofNAPpolyglutamylationhasnotyetbeenaddressed,but
asfortubulins,itmayregulatetheaffinityofNAPsfortheir
bindingpartners.
Additionalpolyglutamylatedproteinsprobablyexistasthe
polyglutamylation-specificantibodyGT335(16)recognizes
proteinbandsbesidesthoseoftubulinsandNAPsonWestern
blotsofHeLacellextracts(4,7).Here,usingaproteomic
approach,weidentifynewsubstratesforpolyglutamylationand
showthattheyaremodifiedbytwopolyglutamylasesfromthe
TTLLproteinfamily.Noclear“glutamylationmotif”couldbe
definedintheirprimarysequences,butallthehere-identified
newsubstratescontainglutamate-richstretchesthataremost
likelytheacceptorsitesforthemodification.Ourstudyopens
thedoortofurtherinvestigationsoftheroleofpolyglutamyla-
tionasageneralregulatoryeventofabroadrangeofcellular
functions.
EXPERIMENTALPROCEDURES
PurificationofGT335-reactiveProteinsfromHeLaCellsby
ImmunoaffinityChromatography—10mgofGT335antibody
wascoupledtoanN-hydroxysuccinimide-activatedHi-Trap
column(1ml,GEHealthcare)accordingtothemanufacturer’s
ThisworkwassupportedinpartbytheCNRS,theUniversitiesMontpellier1
and2,AssociationdelaRecherchecontreleCancerAwardsCR504/7817
and3140(toC.J.),FrenchNationalResearchAgencyAwardJC05_42022
(toC.J.),andtheLaLiguecontreleCancer(toC.J.andB.L.).Thecostsof
publicationofthisarticleweredefrayedinpartbythepaymentofpage
charges.Thisarticlemustthereforebeherebymarked“advertisement”in
accordancewith18U.S.C.Section1734solelytoindicatethisfact.
□S
Theon-lineversionofthisarticle(availableathttp://www.jbc.org)contains
supplementalTableS1andFigs.S1andS2.
1
Towhomcorrespondenceshouldbeaddressed:CRBM,CNRS,1919Route
deMende,34293Montpellier,France.Tel.:33-4-67613335;Fax:33-4-
67521559;E-mail:carsten.janke@crbm.cnrs.fr.
2
Theabbreviationsusedare:TTLL,tubulin-tyrosineligase-like;MAP,micro-
tubule-associatedprotein;MT,microtubule;NAP,nucleosomeassembly
protein;PBS,phosphate-bufferedsaline;GST,glutathioneS-transferase;
GFP,greenfluorescentprotein.
THEJOURNALOFBIOLOGICALCHEMISTRYVOL.283,NO.7,pp.3915–3922,February15,2008
?2008byTheAmericanSocietyforBiochemistryandMolecularBiology,Inc.PrintedintheU.S.A.
FEBRUARY15,2008?VOLUME283?NUMBER7JOURNALOFBIOLOGICALCHEMISTRY3915
atInstBiophys,CAS,onOctober17,2012
www.jbc.org
Downloadedfrom
http://www.jbc.org/content/suppl/2007/11/29/M705813200.DC1.html
SupplementalMaterialcanbefoundat:
instructions.HeLacellswereextractedinPBScontaining0.1%
TritonX-100andproteaseinhibitors(aprotinin,leupeptin,and
4-(2-aminoethyl)-benzenesulfonylfluoride,eachat10H9262g/ml).
Thehighspeedsupernatant(15mgofprotein)wasloadedonto
thecolumnandafterextensivewashingwithPBS,bound
proteinswereelutedwithPBScontaining0.7MNaClandcon-
centratedto1mg/mlonanAmiconultrafiltrationdevice(4ml,
cut-off10kDa,Millipore).
WesternBlotting—ProteinsfusedtoEYFPwereimmunode-
tectedwithrabbitanti-GFPantibody(1:5,000,TorreyPines
Biolabs)andhorseradishperoxidase-conjugatedanti-rabbit
IgG(GEHealthcare)followedbydetectionwithenhancedche-
moluminescence(GEHealthcare).GT335antibodywasbiotin-
ylatedwithNHS-LC-Biotin(Pierce)accordingtothemanufac-
turer’sinstructionsandincubatedsimultaneouslywith
horseradishperoxidase-labeledstreptavidin(GEHealthcare).
Forproducingpolyclonalanti-TTLL4andanti-TTLL5anti-
bodies,bacteriallyexpressedandpurifiedfull-lengthmouse
TTLL4orfragment692–1152ofmouseTTLL5wereinjected
intorabbits.Anti-TTLLantibodieswerethenpurifiedfrom
antiseraagainsttheirrespectivetargets.
MassSpectrometricAnalysis—Toidentifytheproteincon-
tentoftheGT335affinity-purifiedfraction,theCoomassie-
stainedproteinbandswereexcisedfromthegelandsubmitted
toin-geltrypsindigestion(17).Peptideswerethenconcen-
tratedonaC18pre-column,separatedonareversed-phase
capillarycolumn(PepmapC18,75-H9262minnerdiameter,15-cm
length,LCPackings)andanalyzedbyMS/MSonaQ-TofII
massspectrometer(MicromassLtd.,Manchester,UK).Data
analysiswasperformedwithMascot(MatrixScienceLtd.,Lon-
don,UK)againsttheNCBI(TheNationalCenterforBiotech-
nologyInformation)Database.Double-andtriple-charged
peptideswereusedfortheresearchindatabase,andthepep-
tidetolerancewassetto0.2DaforMSandto0.3DaforMS/MS.
Onemissedcleavagebytrypsinwasacceptedandcarbamidom-
ethylatedcysteineandoxidizedmethionineweresetasvariable
modifications.Allentrieswereselectedforthetaxonomy.The
Mascotscorecut-offvalueforapositiveproteinhitwassetto
50,andpeptideswithascorebelow40weremanuallyinter-
pretedtovalidateordiscardtheidentification.Identifiedpro-
teinsweresubmittedtoasecondresearchincludingpolyglu-
tamylationasvariablemodification.Additionsof1–8glutamic
acidswereconsideredandanewresearchhasbeendoneeach
time.
InVitroPolyglutamylationoftheGT335Affinity-purified
Fraction—TheGT335affinity-purifiedfraction(finalprotein
concentrationof0.3mg/ml)wasincubatedfor2hat30°Cin
thepresenceof50mMTris-HCl,pH9.0,0.5mMATP,2.4mM
MgCl
2
,0.5mMdithiothreitol,and60H9262ML-[
3
H]glutamate
(45–55Ci/mmol,GEHealthcare).FollowingSDS-PAGE,gels
wereexposedtoMPfilms(GEHealthcare)for40daysat
H1100280°CafterenhancementwithAmplify(GEHealthcare).
ImmunoprecipitationExperiments—Forthedepletionof
TTLLs,theGT335affinity-purifiedfraction(10H9262goftotalpro-
tein)wasincubatedfor2hat4°Conarotaryagitatorwith
proteinGmagneticbeads(Dynabeads)thatwerecoupledto
anti-TTLL4oranti-TTLL5antibodies(2H9262g).Afterwashing
withPBS,theTTLL-depletedfractionwaspolyglutamylatedin
vitroasdescribedabove.ForimmunoprecipitationoftheEYFP
fusionsofNAPsandtheirchimeras,thehigh-speedsuperna-
tantofHeLacellextractswasincubatedundersimilarcondi-
tionswithbeadslinkedtoanti-GFPantibody.Followingexten-
sivewashing,thebeadsweredirectlyresuspendedinLaemmli
buffer.
InVitroPolyglutamylationAssaywithRecombinantProteins—
Putativesubstratesofpolyglutamylationwereclonedfrom
mousetestisorbraincDNAlibraries,expressedaspolyhisti-
dinetagfusionproteinsinbacteriaandpurifiedbynickelaffin-
itychromatography(exceptforEB1,whichwasagenerousgift
ofS.HonnappaandM.O.Steinmetz).TTLLproteinswere
expressedasGSTfusionsinbacteriaandpurifiedonglutathi-
one-Sepharoseaspreviouslydescribed(4).Becauseonly
TTLL5(CAM84325)wassolubleandactivewhenproducedin
bacteriaasitsfull-lengthversion,weproducedtruncatedver-
sionofTTLL4(TTLL4_C639,last639residuesofCAM84324)
andTTLL6(TTLL6_N705,first705residuesofCAM84326)
andashorter,naturallyoccurringversionofTTLL7(TTLL7S,
CAM84327)(3,4).Theseshorterenzymeshavebeenpreviously
demonstratedtopossesssimilaractivitiesascomparedwiththe
full-lengthTTLLs(4).Thepolyglutamylationassaywasper-
formedbyincubating1H9262gofeachpurifiedTTLLenzymewith
4H9262gofeachputativesubstratein20H9262lof50mMTris-HCl,pH
9.0,0.5mMATP,2.4mMMgCl
2
,0.5mMdithiothreitol,and7.4
H9262ML-[
3
H]glutamate(45–55Ci/mmol,GEHealthcare)for2hat
30°C.Alternatively,4H9262goftaxotere-stabilizedMTs,prepared
fromadultmousebrains(18),wereusedasasubstrate.Quan-
tificationoftheradioactivityincorporatedintothedifferent
putativesubstrateproteinswasdonebyscintillationcounting
oftheproteinbandsafterSDS-PAGEandelectrotransferonto
nitrocellulose.Eachpolyglutamylationtestwasperformedat
leastthreetimesinindependentexperiments.Theoverall
count(dpm)variedbetweeneachsetofexperiments;however,
therelativevaluesofglutamateincorporatedintodifferentsub-
stratestestedwiththesameenzymedilutionweresimilarinall
experiments.Eachfigureisonesetofagivenexperimentper-
formedatthesametimewithallsubstratespresented.Back-
groundlevels(noincorporation)arebelow100dpm.
InVivoPolyglutamylationoftheNewSubstrates—Putative
substratesofpolyglutamylationwereclonedasEYFPfusion
proteinsinmammalianexpressionvectorsandtransfectedinto
HeLacellstogetherwithanactiveoranenzymaticallyinactive
(mutationE901GintheATP-bindingsite)versionof
TTLL4_C639-EYFP.After24hoftransfection,cellswerelysed
inLaemmlibufferandsubsequentlyanalyzedbygelelectro-
phoresisfollowedbyWesternblot.InFig.5C,NAPsandtheir
chimeraswereexpressedaloneandimmunoprecipitatedwith
anti-GFPantibodypriortoanalysis.
Mutagenesis—ToexchangetheC-terminaltailsbetweenthe
differentNAPs,weintroducedtherestrictionsiteBstBIintoa
highlyconservedregionoftheNAPs.Thiswasdoneby
mutagenesisofNAPs(NAP1:T972C,NAP2:T951C,NAP4:
T1170CT1171GT1172A,introducingaL391Eaminoacid
change)usingtheQuikChangePCRmethod.Theresulting
mutantswereexpressedandwereshowntobehavelikethe
wild-typeproteinsinthepolyglutamylaseassay(datanot
shown).WethenusedBstBIincombinationwithother
NewSubstratesofPolyglutamylation
3916JOURNALOFBIOLOGICALCHEMISTRYVOLUME283?NUMBER7?FEBRUARY15,2008
atInstBiophys,CAS,onOctober17,2012
www.jbc.org
Downloadedfrom
enzymestoexchangetheNAPdomains.Theresultingchime-
raswereclonedintobacterialandmammalianexpression
vectors.
RESULTS
PurificationandIdentificationofPutativeSubstratesfor
Polyglutamylation—BecausetheGT335antibodydetectspro-
teinbandsbesidesthosecorrespondingtotheglutamylated
formsoftubulinsandNAPsonWesternblotsofHeLacell
extracts,itislikelythatthisantibodyspecificallyrecognizes
someotherpolyglutamylatedproteins.Wethereforeper-
formedalargescalepurificationofGT335-reactiveproteins
fromHeLacellextractsbychromatographyonaGT335col-
umn.Theaffinity-purifiedfractionwassubmittedtoSDS-
PAGEandstainedwithCoomassieBlue(Fig.1,lane1)ortrans-
ferredontonitrocelluloseandprobedwithGT335antibody
(Fig.1,lane2).Wevisualized40proteinbandsonthegelafter
Coomassiestaining,andabouthalfofthemweredetectedwith
GT335antibodyonWesternblot.Theremainingproteins
mighthavebeennon-specificallyretainedonthecolumn,for
examplebyinteractingwithGT335-reactiveproteins.How-
ever,itisalsopossiblethatsomeproteinsarerecognizedby
GT335antibodyonlyintheirnativeform.Wethereforeexcised
all40proteinbandsfromthegelanddeterminedtheirprotein
contentbynano-LC-MS/MSaftertrypsindigestion.Thisanal-
ysisgaverisetotheidentificationofabout170proteins(sup-
plementalTableS1).
Toidentifytruesubstratesofpolyglutamylation,themasses
obtainedbynano-LC-MS/MSwerere-analyzedtosearchfor
massesthatcouldbeattributedtopolyglutamylatedtryptic
peptides.However,wefoundnomassescorrespondingtomod-
ifiedpeptides(massincrementsofntimes129Da,themassofa
glutamylunit,withnrangingfrom1to8).Ithasbeenpreviously
reportedthatthepolyglutamylatedtrypticpeptidesoftubulins
andNAPsareundetectablewiththeclassicalmassspectrome-
trymethodusedhere(1,7).Thesameseemstobetrueforthe
othersubstratesofpolyglutamylation.Actually,becauseof
theirhighacidityand/orbranchedstructure,themodifiedpep-
tidesarelostduringthereversephasechromatographyusedto
fractionatethepeptides(7).Analternativeprocedurewouldbe
tofractionatethetrypticpeptidesbyion-exchangechromatog-
raphyfollowedbyadesaltingstep.Thismethodhasbeenused
inthecaseofNAPs(7),buttofollowthemodifiedpeptides
duringthepurificationsteps,radioactiveglutamatehadtobe
incorporatedintotheproteins.Anotherproblem,becauseof
theacidiccharacterofthemodifiedpeptides,isthattheirion-
izationinelectrosprayorMALDIisveryhardevenintheneg-
ativemode.Severalprocedureshavebeenproposedtosolve
thisproblem(19–21),butthemethodremainsfarfrom
straightforward.Takentogether,wejudgedamassspectrome-
tryapproachasnotappropriatetoidentifynewsubstratesof
polyglutamylation.
Topreselectproteinswithahighprobabilityofbeingpoly-
glutamylated,wetookadvantageofthefactthatapolyglutamy-
laseactivityco-purifieswiththeGT335-reactiveproteins(7).
IncubationoftheGT335affinity-purifiedfractionwith
[
3
H]glutamateandMgATPresultedintheradioactivelabeling
of13proteinbands(Fig.1,lane3).Becausethenewsubstrates
ofpolyglutamylationmight,astubulinsandNAPs,bemodified
withinaglutamate-richsequencewesearchedforproteinscon-
tainingsuchmotifsamongtheproteinsidentifiedwithinthe13
3
H-labeledbands.Wefoundatleastoneproteinwithagluta-
mate-richstretchineachofthesebands(Table1)andthus
FIGURE1.PurificationofGT335-reactiveproteinsfromHeLacellextracts.
ThehighspeedsupernatantofHeLacellextractswassubmittedtoGT335
affinitychromatography.Theboundfractionwasrunona10%SDS-polyac-
rylamidegelandstainedwithCoomassieBlue(lane1),ortransferredonto
nitrocellulosefordetectionwithGT335antibody(WB,lane2).Theincorpora-
tionof[
3
H]glutamatebytheco-purifiedpolyglutamylaseactivitywasvisual-
izedbyautoradiography(lane3).Theproteinbands(whosecontentislisted
inSupplementalTableS1)arenumbered,andtheirnumberisinaboxforthe
onesthatappearradioactivelylabeledfollowinginvitropolyglutamylation.
Molecularmassmarkersareindicated.
NewSubstratesofPolyglutamylation
FEBRUARY15,2008?VOLUME283?NUMBER7JOURNALOFBIOLOGICALCHEMISTRY3917
atInstBiophys,CAS,onOctober17,2012
www.jbc.org
Downloadedfrom
expectthattheseproteinsareverylikelytargetsof
polyglutamylation.
TTLL4andTTLL5ModifyNewSubstratesinVitro—Tocon-
firmwhethertheverylikelytargetsofpolyglutamylation,
preselectedinTable1,canindeedbepolyglutamylated,a
selectionofthemwasproducedinbacteriaandpurifiedfor
analysisinaninvitropolyglutamylationassay.Notallpro-
teinsweretestedbecausesomecouldnotbeproducedin
bacteriaduetotheirlargesize(nucleoporin,PELP1,Myb-
bindingprotein1A)andotherssharedahighhomologywith
someofthetestedproteins(PHAPIIisashorterversionof
SETthatdiffersonlyinfewaminoacidsattheNterminus
andANP32Eshares57%identitywithANP32A(22)).Totest
theabilityoftheselectedproteinstobepolyglutamylated
invitro,weneedasourceofpolyglutamylaseactivity.Pre-
sently,theonlyknownmammalianpolyglutamylaseen-
zymesaremembersoftheTTLLproteinfamily(4).Following
overexpressionofthedifferentTTLLsinHeLacells,only
TTLL4andTTLL5increasedtheGT335reactivityofnon-tu-
bulinsubstrates(4).Thus,thesetwoTTLLsarestrongcandi-
datesforbeingtheenzymesthatpolyglutamylatenon-tubulin
substrates.WethereforeusedrecombinantTTLL4_C639(a
truncatedversionofTTLL4,see“ExperimentalProcedures”)
andTTLL5forinvitropolyglutamylationassays.Toinvestigate
thespecificityofdifferentenzymesforthenewsubstrates,we
alsotestedtwootherpolyglutamylases,TTLL6_N705and
TTLL7S.Thetestedproteins,NAP1,NAP2,NF45,SET,B23,
ANP32A,andANP32B(Table1,proteinsinbold)areall
stronglymodifiedbyTTLL4_C639andtoalesserextentby
TTLL5(Fig.2A).Thetwootherenzymes,TTLL6_N705and
TTLL7S,whicharebothhighlyactiveonMTs,didnotmodify
thenewsubstrates(Fig.2A).Thus,invitro,non-tubulinsub-
stratesarespecificallypolyglutamylatedbyTTLL4andTTLL5.
Moreover,becauseallpreselectedproteinstestedaremodified,
itislikelythatthepresenceofglutamate-richsequencesis
indeed,asproposedearlier,asignatureofpolyglutamylated
proteins.
Amongtheproteinbandsthatwerenotlabeledwith[
3
H]glu-
tamatebythepolyglutamylaseactivityco-purifiedonthe
GT335column,weidentifiedseveralotherproteinscontaining
glutamate-richstretches(supplementalTableS1,proteinsin
bold).Theseproteinsmightbepolyglutamylatedeithertoa
lowerextentorbyadifferentenzyme,whichwasnotco-puri-
fiedontheGT335-column.Toinvestigatethispossibility,we
repeatedtheinvitropolyglutamylationassayswiththesameset
ofTTLLenzymeson6oftheseproteins(GRP78,NASP,RAN-
GAP,RNP-K,NCT,andEB1;Fig.2B).Asbefore,TTLL4_C639
wasthemostefficientenzyme.Itincorporated[
3
H]glutamateinto
RNP-KandEB1,toalesserextentintoRANGAPandNCT,but
poorlyintoGRP78andNASP.NotethatTTLL6_N705and
TTLL7S,whichwereinactiveonthepreviouslytestednon-tubulin
substrates,slightlymodifiedEB1.Thismightbeduetotheclose
structuralhomologybetweenEB1andtubulins(23).Overall,the
proteinsmodifiedinthistestincorporatedmuchlessglutamate
thanthepreviouslyidentifiedsubstrates,whichmightexplainwhy
theywerenotseenintheautoradiography(Fig.1,lane3).How-
ever,theyhavetobeconsideredaspotentialsubstratesofpolyglu-
tamylationandtheirmodificationmightbeimportantinvivo.
TTLL4IstheMajorPolyglutamylaseCo-purifiedwiththe
NewSubstrates—BecausetheGT335affinity-purifiedfraction
possessesapolyglutamylaseactivityabletomodifythenew
substrates(Fig.1,lane3),wewantedtodetermineifitcontains
eitherTTLL4orTTLL5,thetwoenzymesthatmodifythesub-
stratesinvitro.Wethereforeraisedpolyclonalantibodiesagainst
theseTTLLs.Theepitopesofboth,anti-TTLL4andanti-TTLL5
antibodiesarelocatedoutsidethedomainsthatareconserved
betweenthedifferentTTLLs(supplementalFig.S1).Accordingly,
whenweprobedextractsofHeLacellsoverexpressingthediffer-
entfull-lengthTTLLpolyglutamylases,anti-TTLL4,andanti-
TTLL5antibodiesspecificallyrecognizedtheirtargetsandexhib-
itednocross-reactivitywithanyotherTTLLpolyglutamylase(Fig.
3A).WethenusedtheseantibodiestoprobetheGT335-affinity
purifiedfraction,butnoneoftheantibodiesdetectedaproteinof
theexpectedsizeonWesternblot(notshown),mostprobably
becauseoftheabsenceorlowconcentrationoftheenzymes.
Becausebothantibodieshavebeendemonstratedtoprecipitate
theirtargetTTLLproteins(notshown),weusedthemtodeplete
theGT335affinity-purifiedfraction.Followingincubationofthe
depletedfractionswith[
3
H]glutamateandMgATP,weobserveda
strongdecreaseofglutamateincorporationwhenTTLL4,butnot
whenTTLL5wasdepleted(Fig.3B).ThusTTLL4isthemajor
enzymeinvolvedinthepolyglutamylationofthenewsubstrates
identifiedinHeLacells.
TTLL4PolyglutamylatestheNewSubstratesinHeLaCells—
Toshowthattheproteinsthatarepolyglutamylatedinvitrocan
alsobemodifiedbyTTLL4underinvivoconditions,we
expressedthemasEYFP-taggedversionsinHeLacells.In
untransfectedcells(notshown)orcellsoverexpressinganinac-
tive(ATPase-dead)TTLL4_C639(Fig.4,firstlane),onlypro-
teinsinthe50-kDaregion(tubulins,NAPs)anda180-kDapro-
teinaredetectedwithGT335antibody.Thisindicatesthatthe
endogenouspolyglutamylaseactivityofHeLacellsmightbetoo
lowtomodifymostsubstratesabovedetectionlevelsofGT335
TABLE1
Preselectionofverylikelytargetsofpolyglutamylation
Amongtheproteinsidentifiedwithinthe13proteinbandsthatwerelabeledwith
H20851
3
HH20852glutamate(Fig.1andSupplementalTableS1),weselectedthosethatpossess
longglutamate-richstretcheswithintheirprimarysequenceandthatthusarevery
likelytargetsofpolyglutamylation.Thebandnumberscorrespondtothoseindi-
catedinFig.1.Invitropolyglutamylationtestswereperformedontheproteinsin
boldandallofthemdidincorporateglutamate.
BandProteinAccessionMasspI
1Nucleoporingi8573683579935.86
5Proline-,glutamicacid-,
leucine-richprotein1(PELP1)
gi244153831195484.30
Myb-bindingprotein1Agi711538251487629.34
19H9251-Tubulingi32015497614.95
20Nucleosomeassembly
protein1-like4(NAP2)
gi5174613427974.60
21H9252-Tubulingi18088719496404.75
22Nucleosomeassembly
protein1-like1(NAP1)
gi4758756453464.36
26NF45proteingi532313446698.26
27ProteinSETgi46397790334694.23
28PHAPIIgi403009320844.12
29B23nucleophosmingi825671309194.71
32Acidicnuclearphosphoprotein
32family,memberE(ANP32E)
gi13569879306743.77
34Acidicnuclearphosphoprotein
32family,memberA(ANP32A)
gi5453880285683.99
35Acidicnuclearphosphoprotein
32family,memberB(ANP32B)
gi1498227222634.19
NewSubstratesofPolyglutamylation
3918JOURNALOFBIOLOGICALCHEMISTRYVOLUME283?NUMBER7?FEBRUARY15,2008
atInstBiophys,CAS,onOctober17,2012
www.jbc.org
Downloadedfrom
inWesternblot.Accordingly,intheabsenceofexogenouspoly-
glutamylaseactivity,onlysubstratesthatwerestronglymodi-
fiedinvitro(NAP1,NAP2,SET,andtheANP32proteins)could
bedetectedbyGT335antibodyfollowingtheiroverexpression
(Fig.4,lowerpanel,“D”).Wethenexpressedtheselectedpro-
teinstogetherwithanactiveversionofTTLL4_C639(Fig.4,
“A”).AsignificantincreaseinGT335reactivitywasobservedon
arangeofendogenousproteins(Fig.4,comparefirstandsecond
lane),demonstratingthatGT335specificallydetectspolyglu-
tamylationonalargerangeofproteins.Moreover,theGT335
reactivityofNAP1,NAP2,SET,andtheANP32proteins
increasedfurther,andotherputativesubstrates(RANGAPand
nucleolin)alsobecamelabeledwithGT335antibody(Fig.4,
lowerpanel,“A”).Theexpressionlevelsofthesubstrates,as
visualizedwithanti-GFPantibody,aresimilarincellsco-trans-
fectedwithATPase-deadoractiveTTLL4_C639(Fig.4,upper
panel,“D”and“A”).Thus,increaseofGT335detectionlevels
clearlycorrelateswithincreasedpolyglutamylationofthe
respectiveproteinsbytheactiveTTLL4_C639.Moreover,a
shiftinmigrationwasobservedformostofthemodifiedpro-
teins,suggestingthatthechargeand/orsizeoftheseproteins
havebeenchanged(Fig.4,upperpanel,compare“D”with“A”
forthemodifiedproteins).
RequirementsforPolyglutamylation—Weidentifiedanum-
berofproteinsthataresubjectedtopolyglutamylation.All
theseproteinspossessglutamate-richstretcheswithintheir
primarysequence,butnoobviouspolyglutamylationconsensus
sitecouldbedefined.Tobetterunderstandthedeterminantsof
polyglutamylation,weanalyzedtheprimarysequencesofsev-
eralmembersoftheNAPfamily(Fig.5AandsupplementalFig.
S2A)andtheirabilitytobemodifiedinvitrobyTTLL4_C639
(Fig.5B)orinHeLacellsbyendogenouspolyglutamylases(Fig.
5C).NAP1andNAP2are,incontrasttoNAP3andNAP4,
stronglypolyglutamylatedinvitroandinvivo(Fig.5).NAP1
waspreviouslyshowntopossessamajorpolyglutamylationsite
initsCterminusandaminorsiteinitsNterminus(7).Both
FIGURE2.InvitropolyglutamylationofselectedputativesubstrateswithdifferentTTLLenzymes.Proteinsselectedamongthebandslabeledbythe
co-purifiedpolyglutamylaseactivity(AandTable1)andproteinsselectedamongtheotherbandsaccordingtothepresenceofglutamate-richsequences(B;
proteinsinboldinsupplementalTableS1)wereexpressedandpurifiedfrombacteria.Theywerethensubjectedtoaninvitropolyglutamylationassaywith
recombinantTTLL4_C639,TTLL5,TTLL6_N705,andTTLL7S.TheactivityofeachTTLLwasverifiedbyusingMTsasasubstrate.Theamountof[
3
H]glutamate
incorporatedinthedifferentsubstrateswasdeterminedbyscintillationcounting.
NewSubstratesofPolyglutamylation
FEBRUARY15,2008?VOLUME283?NUMBER7JOURNALOFBIOLOGICALCHEMISTRY3919
atInstBiophys,CAS,onOctober17,2012
www.jbc.org
Downloadedfrom
sitesconsistofglutamate-richstretchesthatarenotwellcon-
servedamongthemembersoftheNAPfamily(supplemental
Fig.S2AandFig.5A).NAP2divergesfromNAP1intheN-ter-
minalsite.Becausethisisaminorpolyglutamylationsite,NAP2
isstillstronglymodified.Onthecontrary,thesequenceof
NAP4divergesinthemajorC-terminalsiteandthusNAP4is
onlylowlymodified.NAP3divergesinbothsitesandaccord-
ingly,itisnotmodified.Thus,wehaveshownacorrelation
betweenthepresenceofthetwopreviouslyidentifiedpolyglu-
tamylationsitesandtheextentofmodification.Inadditionto
thesetwosites,allNAPspossessacentraldomainwhichisrich
inglutamates(supplementalFig.S2A).Thisdomaindoesnot
seemtobetargetedbythepolyglutamylation,asNAP3isnot
modifiedalthoughithasthemostextendedglutamate-rich
centraldomain(Fig.5A).Thefactthatthisdomainisnotmod-
ified,althoughitislocatedattheoutersurfaceoftheprotein
accordingtothestructureofNAP1(24),suggeststhatthepres-
enceofaglutamate-richsequenceisnottheonlyprerequisite
foritspolyglutamylation.
WethenproducedsixchimerasoftheNAP1,2,and4pro-
teinsbyswappingtheirC-terminaltails(Fig.5A)andtested
theirabilitytobemodifiedbyTTLL4_C639invitro(Fig.5B)or
byendogenouspolyglutamylaseactivityinHeLacells(Fig.5C).
Chimerasthatpossessthenon-acidicC-terminaltailofNAP4
(NAP1_4andNAP2_4),werepoorlypolyglutamylatedbothin
vitroandinvivo,ingoodagreementwiththelackofthemajor
polyglutamylationsite.Ontheotherhand,chimerasthatbear
thehighlyacidictailofNAP2(NAP1_2andNAP4_2)are,as
NAP2,stronglymodified.Concerningthechimerasthathavethe
highlyacidictailofNAP1(NAP2_1andNAP4_1),NAP2_1is
modifiedtothesameextentasNAP1butNAP4_1islessmodified
FIGURE3.Identificationofthepolyglutamylaseactivityco-purifiedwiththe
GT335-reactiveproteins.Thespecificityofpolyclonalanti-TTLL4andanti-
TTLL5wasdemonstratedonextractsfromHeLacellsoverexpressingthediffer-
entTTLLpolyglutamylasesinfusionwithEYFP(A).Theseantibodieswerethen
usedtoimmunodepletetheGT335affinity-purifiedfraction(B).Asacontrol,
depletionwasalsoperformedwithrabbitIgG.Thedepletedfractionswereincu-
batedwith[
3
H]glutamatefor2h,runonanSDS-PAGEandstainedwithCoomas-
sieBlue.[
3
H]Glutamateincorporationwasvisualizedbyautoradiography.Only
thedepletionwithanti-TTLL4antibodyreducedthepolyglutamylaseactivityin
theGT335affinity-purifiedfraction.Molecularmassmarkersareindicated.
FIGURE4.Invivopolyglutamylationofselectedputativesubstrates.Thepreviouslytestedsubstratesofpolyglutamylation(Fig.2),aswellasnucleolin
whichwasnottestedinvitrobecauseitwashardtoproduceinbacteria,wereco-transfectedintoHeLacellswithanactiveorinactive(ATPase-dead)version
ofTTLL4_C639infusionwithEYFP.After24hoftransfection,thecellswerelysedinLaemmlibuffer,runonSDS-PAGE,transferredontonitrocelluloseand
probedwithanti-GFPantibodytochecktheexpressionleveloftheoverexpressedproteins(upperpanel)andwithGT335antibodytovisualizetheextentof
polyglutamylationofthesubstrates(lowerpanel).Orangearrowspointtothedifferentoverexpressedsubstratesvisualizedwithanti-GFPantibody.Inthelower
panel,greenarrowsindicatethesubstratesthathavebeenpolyglutamylatedinvivo,whileredarrowspointtoproteinsthatwerenotmodified.Ingeneral,we
consideredasubstrateasbeingmodified,when:(i)abandofthesizeoftheprotein(ascomparedwiththeanti-GFPsignal)isrecognizedbyGT335,and(ii)its
GT335reactivityincreasesuponco-expressionofactiveTTLL4_C639,and(iii)thisbanddoesnotcorrespondtoendogenousGT335-reactiveproteins(compare
withcellsnotexpressingthegivensubstrate).Notethatcellsco-overexpressingB23andactiveTTLL4_C639weredying,explainingthedecreasedoverall
GT335reactivityinthecorrespondinglane.Molecularmassmarkersareindicated.
NewSubstratesofPolyglutamylation
3920JOURNALOFBIOLOGICALCHEMISTRYVOLUME283?NUMBER7?FEBRUARY15,2008
atInstBiophys,CAS,onOctober17,2012
www.jbc.org
Downloadedfrom
thanonewouldexpectaccordingtothepresenceofthetailof
NAP1.Thisagainsuggeststhatother,yetunknowndeterminants
ofpolyglutamylationcontributetothemodification.
Tofurtherdemonstratethatthepresenceofaglutamate-rich
stretchisnottheonlyprerequisiteforaproteintobecome
polyglutamylated,weanalyzeddifferentmembersofthe
nucleophosmin/nucleoplasminfamily.NucleophosminB23
hastwoputativeacidicpolyglutamylationsites(supplemental
Fig.S2B).NPN2,anoocyte-specifichomologueofB23,ismore
acidicinthefirstsitebutlessinthesecond(supplementalFig.
S2B).Although,takentogether,thetotalnumberofglutamate
residuesinthetwopotentialpolyglutamylationsitesissimilar,
NPN2ismodifiedtoalowerextentthannucleophosminB23by
TTLL4_C639(Fig.6).Thissuggeststhatthesecondsitecon-
tributesmoretothedegreeofpolyglutamylationthanthefirst
one.Consistently,NPN3,whichcompletelylacksthesecond
site,ispoorlymodifiedbyTTLL4_C639(Fig.6).Again,these
resultsrevealthatnotallglutamate-richstretchespresentin
proteinsareequallysubjectedtopolyglutamylation.
WefurtheranalyzedthehighlyhomologousproteinsEB1and
EB3.Althoughbothproteinscarryaglutamate-richCterminus
(supplementalFig.S2C),onlyEB1
wasmodifiedbyTTLL4_C639in
vitro(Fig.6).OtherTTLLenzymes
thatareactiveonEB1(Fig.2B)were
inactiveonEB3aswell(notshown).
Finally,wetestedtheprotein
HMG-B1,whichhasnotbeeniden-
tifiedasapotentialsubstrate,but
hasanextendedglutamate-richC
terminus(supplementalFig.S2D).
HMG-B1wasalmostnotmodified
intheinvitroassaywithTTLL4_
C639(Fig.6).
Takentogether,theseobserva-
tionsshowthatthepresenceofglu-
tamate-richsequencescanbeapre-
requisite,butisnotsufficientby
itselfforaproteintobecomeatar-
getofpolyglutamylation,suggesting
thatotherfeaturesarerequired.
DISCUSSION
Polyglutamylationwasinitially
discoveredonbrainH9251-tubulinand
subsequentlyalsoobservedonH9252-tu-
bulin(1,5,6).Itwasconsideredasa
tubulin-specificmodificationuntil
itwasdemonstratedtotargetalso
NAP1andNAP2(7).Inthepresent
study,weshowthatpolyglutamyla-
tionisanevenmoregeneralpost-
translationalmodification.
NewSubstratesforPolyglutamyl-
ation—Usingtheaffinityofpolyglu-
tamylatedproteinsfortheGT335
antibody,wepurifiedmanypoten-
tialpolyglutamylatedproteins(sup-
plementalTableS1).Wedemonstratedthatseveralofthese
proteinsareindeedsubjectedtopolyglutamylationastheyare
modifiedbyTTLL4invitroand/orinacellularcontext.Aswe
couldnottestalltheputativesubstrates,anumberofother
targetsofpolyglutamylationmightbepresentamongthepro-
teinslistedinsupplementalTableS1.Moreover,thislistisby
nomeansexhaustiveasmanyofthesubstratescanbelowabun-
danceproteins.Itisalsopossiblethatnotallpolyglutamylated
proteinsinteractwiththeGT335antibody.Thisantibodyhas
beenraisedagainstasyntheticpeptide(EGEGEEEG)modified
bytheadditionoftwoglutamylunitsonthethirdglutamate
(16)anditmight,therefore,bebiasedtowardproteinsthatare
polyglutamylatedwithinglutamate-richsequences.
SpecificityofthePolyglutamylationReaction—Allthehere-
identifiedsubstratesofpolyglutamylationaremodifiedwithin
glutamate-richstretches,suggestingthatanacidicenviron-
mentattheacceptorsiteisgenerallynecessaryforpolyglu-
tamylationtooccur.Nevertheless,wedemonstratethatone
cannotsimplyconcludethataproteinpossessingglutamate-
richsequencesissubjectedtopolyglutamylation.Substraterec-
ognitionbypolyglutamylasesisprobablyacomplexmecha-
FIGURE5.InvitroandinvivomodificationofNAPsandchimeras.ThesequencesoftheNAPproteins(see
supplementalFig.S2A)andofdifferentchimerasinwhichtheC-terminaltailsofNAP1,NAP2andNAP4
wereexchangedareschematized(A).PreviouslyidentifiedN-terminalandC-terminalpolyglutamylationsites
areshowninredaccordingtotheirsize(domain1and3insupplementalFig.S2A).NAPsandtheirchimeras
wereexpressedinbacteria,purifiedandsubjectedtoinvitropolyglutamylationusingrecombinant
TTLL4_C639asasourceofenzyme(B).Theamountof[
3
H]glutamateincorporatedinthedifferentsubstrates
wasdeterminedbyscintillationcounting.NAPsandchimeraswerealsoclonedinfusionwithEYFPinmam-
malianexpressionvectorsandtransfectedintoHeLacells.Following24hoftransfection,theywerefirst
immunoprecipitatedandthenprobedwitheitheranti-GFPantibodytoseetheexpressionlevel,orwithGT335
antibodytodetecttheirpolyglutamylationbyendogenousenzymaticactivity(C).
NewSubstratesofPolyglutamylation
FEBRUARY15,2008?VOLUME283?NUMBER7JOURNALOFBIOLOGICALCHEMISTRY3921
atInstBiophys,CAS,onOctober17,2012
www.jbc.org
Downloadedfrom
nism.Solvingthecrystalstructureofthepolyglutamylase
enzymesandthusknowingthepropertiesofthesurfaceresi-
duesoftheenzymethatareincontactwiththesubstrate,will
certainlyfacilitateunderstandingoftheprerequisitesforthe
polyglutamylationreaction.Takentogether,wesuggestthat
proteinswithglutamate-richsequencescanbegenerallycon-
sideredaspotentialsubstratesforpolyglutamylation,although
thisneedstobetestedforeachsinglecase.
PossibleNewRolesofPolyglutamylation—Polyglutamylation
generatesacidicsidechainsofvariablelengthsonproteins.
Suchamodificationmightactasafine-regulatorofmanycel-
lularfunctions,forexamplebycontrollingtheaffinityofthe
here-identifiedtargetproteinsfortheirbindingpartners,as
earlierproposedfortubulins(reviewedinRef.25).Manyofthe
newsubstratesofpolyglutamylation(ANP32proteins,SET,
nucleophosminB23,nucleolin,andRNP-K),aswellastheNAP
proteins,shuttlebetweenthecytosolandthenucleus(22,26,
27).BecausenucleoporinandRANGAP,proteinsinvolvedin
nucleocytoplasmictransport,arealsomodified,polyglutamyla-
tioncouldplayanimportantregulatoryroleinnucleocytoplas-
micshuttling.Moreover,mostoftheseshuttlingproteinsare
histonechaperones(NAP1,NAP2,B23,SET/PHAPII,nucleo-
lin)and/orregulatorsoftranscription(ANP32proteins,SET/
PHAPII,NF45,PELP1,RNP-K),suggestingthatpolyglutamy-
lationmightalsoregulateaccesstochromatin.Finally,EB1and
RANGAPareregulatorsofMTdynamics,suggestingthatnot
onlythepolyglutamylationoftubulin,butalsoofitsassociated
proteinscouldbeimportantforthedynamicsandfunctionsof
MTs.Incyclingcells,themaximumyieldoftubulinpolyglu-
tamylationisachievedduringmitosis(18).Itisthustemptingto
speculatethat,bytargetingsimultaneouslyMTsandchroma-
tin-bindingproteins,polyglutamylationallowsforacross-talk
betweenspindledynamicsandchangesinchromatinstructure
accompanyingmitosis.
Acknowledgments—WethankS.HonnappaandM.O.Steinmetz
(PaulScherrerInstitut,Villigen,Switzerland)forprovidingessential
reagents,A.Bouaziz,J.-M.Donnay,J.C.Mazur,C.Rouveyrol,andT.
Zobel(CRBM,Montpellier,France)fortechnicalassistance,andK.
Rogowski,C.Bonne-Andrea,A.Burgess(CRBM,Montpellier,France),
andM.Magiera(IGMM,Montpellier,France)forcriticalreadingof
themanuscript.
REFERENCES
1.Edde,B.,Rossier,J.,LeCaer,J.P.,Desbruyeres,E.,Gros,F.,andDenoulet,
P.(1990)Science247,83–85
2.Janke,C.,Rogowski,K.,Wloga,D.,Regnard,C.,Kajava,A.V.,Strub,J.M.,
Temurak,N.,vanDijk,J.,Boucher,D.,vanDorsselaer,A.,Suryavanshi,S.,
Gaertig,J.,andEdde,B.(2005)Science308,1758–1762
3.Ikegami,K.,Mukai,M.,Tsuchida,J.,Heier,R.L.,Macgregor,G.R.,andSetou,
M.(2006)J.Biol.Chem.281,30707–30716
4.vanDijk,J.,Rogowski,K.,Miro,J.,Lacroix,B.,Edde,B.,andJanke,C.
(2007)Mol.Cell26,437–448
5.Audebert,S.,Desbruyeres,E.,Gruszczynski,C.,Koulakoff,A.,Gros,F.,
Denoulet,P.,andEdde,B.(1993)Mol.Biol.Cell4,615–626
6.Rudiger,M.,Plessman,U.,Kloppel,K.D.,Wehland,J.,andWeber,K.
(1992)FEBSLett.308,101–105
7.Regnard,C.,Desbruyeres,E.,Huet,J.C.,Beauvallet,C.,Pernollet,J.C.,and
Edde,B.(2000)J.Biol.Chem.275,15969–15976
8.Nogales,E.(2000)Annu.Rev.Biochem.69,277–302
9.Boucher,D.,Larcher,J.C.,Gros,F.,andDenoulet,P.(1994)Biochemistry
33,12471–12477
10.Larcher,J.C.,Boucher,D.,Lazereg,S.,Gros,F.,andDenoulet,P.(1996)
J.Biol.Chem.271,22117–22124
11.Bonnet,C.,Boucher,D.,Lazereg,S.,Pedrotti,B.,Islam,K.,Denoulet,P.,
andLarcher,J.C.(2001)J.Biol.Chem.276,12839–12848
12.Bobinnec,Y.,Khodjakov,A.,Mir,L.M.,Rieder,C.L.,Edde,B.,and
Bornens,M.(1998)J.CellBiol.143,1575–1589
13.Gagnon,C.,White,D.,Cosson,J.,Huitorel,P.,Edde,B.,Desbruyeres,E.,
Paturle-Lafanechere,L.,Multigner,L.,Job,D.,andCibert,C.(1996)J.Cell
Sci.109,1545–1553
14.Million,K.,Larcher,J.,Laoukili,J.,Bourguignon,D.,Marano,F.,andTour-
nier,F.(1999)J.CellSci.112,4357–4366
15.Ikegami,K.,Heier,R.L.,Taruishi,M.,Takagi,H.,Mukai,M.,Shimma,S.,
Taira,S.,Hatanaka,K.,Morone,N.,Yao,I.,Campbell,P.K.,Yuasa,S.,
Janke,C.,Macgregor,G.R.,andSetou,M.(2007)Proc.Natl.Acad.Sci.
U.S.A.104,3213–3218
16.Wolff,A.,deNechaud,B.,Chillet,D.,Mazarguil,H.,Desbruyeres,E.,
Audebert,S.,Edde,B.,Gros,F.,andDenoulet,P.(1992)Eur.J.CellBiol.59,
425–432
17.Fraering,P.C.,Ye,W.,Strub,J.M.,Dolios,G.,LaVoie,M.J.,Ostaszewski,
B.L.,vanDorsselaer,A.,Wang,R.,Selkoe,D.J.,andWolfe,M.S.(2004)
Biochemistry43,9774–9789
18.Regnard,C.,Desbruyeres,E.,Denoulet,P.,andEdde,B.(1999)J.CellSci.
112,4281–4289
19.Juhasz,P.,andBiemann,K.(1994)Proc.Natl.Acad.Sci.U.S.A.91,
4333–4337
20.Kjellstrom,S.,andJensen,O.N.(2004)Anal.Chem.76,5109–5117
21.Wilson,C.L.,Monteith,W.B.,Danell,A.S.,andBurns,C.S.(2006)J.Pept.
Sci.12,721–725
22.Santa-Coloma,T.A.(2003)Cerebellum2,310–320
23.Honnappa,S.,John,C.M.,Kostrewa,D.,Winkler,F.K.,andSteinmetz,
M.O.(2005)EMBOJ.24,261–269
24.Park,Y.J.,andLuger,K.(2006)Proc.Natl.Acad.Sci.U.S.A.103,1248–1253
25.Verhey,K.J.,andGaertig,J.(2007)CellCycle6,2152–2160
26.Borer,R.A.,Lehner,C.F.,Eppenberger,H.M.,andNigg,E.A.(1989)Cell
56,379–390
27.Michael,W.M.,Eder,P.S.,andDreyfuss,G.(1997)EMBOJ.16,
3587–3598
FIGURE6.Invitropolyglutamylationofproteinsrelatedtothenewsub-
strates.Invitropolyglutamylationassayswereperformedusingrecombinant
TTLL4_C639onproteinsfromthenucleophosmin/nucleoplasminandEB
familiesaswellasonHMG-B1(seesequencesinsupplementalFig.S2A).The
amountof[
3
H]glutamateincorporatedinthedifferentsubstrateswasdeter-
minedbyscintillationcounting.
NewSubstratesofPolyglutamylation
3922JOURNALOFBIOLOGICALCHEMISTRYVOLUME283?NUMBER7?FEBRUARY15,2008
atInstBiophys,CAS,onOctober17,2012
www.jbc.org
Downloadedfrom
|
|
