Superscript4444kkkcom

Diabetologia(2006)49:332–342DOI10.
1007/s00125-005-0087-1ORIGINALARTICLEE.
Oetjen.
A.
Lechleiter.
R.
Blume.
D.
Nihalani.
L.
Holzman.
W.
KnepelInhibitionofmembranedepolarisation-inducedtranscriptionalactivityofcyclicAMPresponseelementbindingprotein(CREB)bythedual-leucine-zipper-bearingkinaseinapancreaticisletbetacelllineReceived:28July2005/Accepted:7September2005/Publishedonline:21December2005#Springer-Verlag2005AbstractAims/hypothesis:Theactivationofthetran-scriptionfactorcyclicAMPresponseelementbindingpro-tein(CREB)byproteinkinaseAisinhibitedbythehumanorthologueofthemitogen-activatedproteinkinase,dual-leucine-zipper-bearingkinase(DLK)interatocarcinomacells.
However,pancreaticbetacellsareelectricallyex-citableandamajorpathwayregulatingCREBinthesecellsismembranedepolarisation,leadingtocalciuminfluxandactivationofthecalcium/calmodulin-dependentproteinphosphatasecalcineurin.
Therefore,theeffectofDLKonCREBactivityinducedbymembranedepolarisationwasinvestigatedinthebetacelllineHIT.
Materialsandmethods:Reportergeneassaysandbiochemicaltechniqueswereused.
Results:RT-PCR,WesternblotanalysisandimmunohistochemistrydemonstratedtheexpressionofDLKinHITcellsandprimarymouseislets.
Intransienttransfectionexperiments,DLKinhibitedbothGAL4–CREBactivityinducedbymembranedepolarisation,andtranscriptiondirectedbytheCREBbindingsite,thecyclicAMPresponseelement.
Furthermore,DLKinhibitedthetranscriptionalactivityconferredbytheCREBcoactiva-tor,CREBbindingprotein,bothunderbasalconditionsandaftermembranedepolarisation.
DLKwasalsoeffec-tiveinresponsetoglucose,themostpotentphysiologicalstimulusandknowntocausemembranedepolarisationofbetacells.
InhibitionofcalcineurinenhancedDLKactiv-ity,whereasoverexpressionofcalcineurinreducedtheinhibitionbyDLKoftranscriptiondirectedbycyclicAMPresponseelementaftermembranedepolarisation.
Conclusions/interpretation:Theseresultsdemonstrateacalcineurin-sensitiveinhibitionbyDLKofCREBactivityaftermembranedepolarisationinpancreaticisletbetacells.
Thisinhibitionmay,atleastpartially,bemediatedatthecoactivatorlevel.
TheresultsthussuggestthatDLKplaysaroleintheregulationofbetacellfunction,includinginsulingenetranscriptionandbetacellapoptosis.
KeywordsBetacells.
Calcineurin.
CyclicAMPresponseelementbindingprotein.
Dual-leucine-zipper-bearingkinase.
MembranedepolarisationAbbreviationsCBP:CREBbindingprotein.
CRE:cyclicAMPresponseelement.
CREB:cyclicAMPresponseelementbindingprotein.
DLK:dual-leucine-zipper-bearingkinase.
ERK:extracellularsignal-regulatedkinase.
GST:glutathioneS-transferase.
JIP:c-JunN-terminalkinaseinteractingprotein.
JNK:c-JunN-terminalkinase.
MAPK:mitogen-activatedproteinkinase.
MAPKKK:mitogen-activatedproteinkinasekinasekinase.
ZPK:zipperproteinkinaseIntroductionPancreaticisletbetacellssynthesiseandsecretethepeptidehormoneinsulinandarethusessentialforglucosehomeo-stasis,withbetacelldysfunctionleadingtodiabetesmellitus.
AmongthemultiplefactorsrequiredfornormalbetacellfunctionisthetranscriptionfactorcyclicAMPresponseelementbindingprotein(CREB)[1–5].
CREBisubiquitouslyexpressedandcouplesdiversesignallingpathwaystogeneexpression[6,7].
ThroughitsbasicregionleucinezipperdomainCREBbindsasadimertogenesthatcarryCREBbindingsiteswiththeconsensuscoreoctamermotif5'-TGACGTCA-3'(cyclicAMPre-sponseelement[CRE]).
ThetranscriptionalactivityofCREBismarkedlyinducedbyvarioussignallingpathways,suchasthoseactivatedbycyclicAMP,calcium,extracel-E.
Oetjen.
A.
Lechleiter.
R.
Blume.
W.
Knepel(*)MolecularPharmacology,UniversityofGttingen,Robert-KochStr.
40,37099Gttingen,Germanye-mail:wknepel@med.
uni-goettingen.
deTel.
:+49-551-395787Fax:+49-551-395699D.
Nihalani.
L.
HolzmanDivisionofNephrology,DepartmentofInternalMedicine,UniversityofMichiganMedicalSchool,AnnArbor,MI,USAlularsignal-regulatedkinase(ERK1/2)andp38mitogen-activatedproteinkinase(MAPK)throughthephosphory-lationofCREBonSer119(inCREB-327,correspondingtoSer133inCREB-341)[6,7].
ThisphosphorylationallowsCREBtointeractwithitscoactivatorCREBbindingprotein(CBP)[7,8],whichthenmediatesCREBtranscriptionalactivitytothegeneraltranscriptionmachinery[7,8].
CREBhasbeenshowntoplayapivotalroleinmanydifferentphysiologicalanddevelopmentalfunctions,suchaslearn-ingandmemory[7],drugaddiction[9],Tcelldevelopmentandactivation[7],growthfactor-dependentcellsurvival[7]andneurodegeneration[10].
Itisalsoinvolvedintheregulationofglucosehomeostasis.
CREBparticipatesinglucosehomeostasisbystimulatingglucoseproductionintheliver[11]andactivatingglucagongenetranscriptioninpancreaticisletalphacells[12].
InpancreaticisletbetacellsCREBmaintainsbetacellfunctionbybindingtoandactivatingtheinsulingene[1–4,13],aswellasbyinhibitingapoptosisandpromotingbetacellsurvival[5].
IntheNIH3T3andinahumanteratocarcinomacellline,proteinkinaseA-inducedCREBactivitywasinhibitedbythemitogen-activatedproteinkinasekinasekinase(MAPKKK)zipperproteinkinase(ZPK),possiblybyadirectinteractionbetweenCREBandZPK[14].
However,theeffectofZPKonCREBactivityinpancreaticisletbetacellswasunknown.
Furthermore,betacellsareelectricallyexcitable,andamajorpathwaythatregulatesCREBinthesecellsisclosureofATP-dependentpotassiumchannelsandmembranedepolarisation,leadingtocalciuminfluxthroughvoltage-dependentL-typecalciumchannelsandactivationofthecalcium/calmodulin-dependentproteinphosphatasecalcineurin[3,4,12,15–17].
Therefore,inthepresentstudytheeffectofdual-leucine-zipper-bearingkinase(DLK),themouseorthologueofZPK[18],onCREBactivityinducedbymembranedepolarisationwasinvestigatedinthebetacelllineHIT.
MaterialsandmethodsPlasmidconstructsTheplasmidsp4xSomCRELuc[2],5xGal4E1BLuc[3],andtheexpressionvectorsforDLKandDLKmt[19],calcineurinsubunitsAandB[20],GAL4-CBP,containingtheC-terminalaminoacidsofCBPfrom1,678to2,441(kindgiftofR.
H.
Goodman,OregonHealthandSciencesUniversity,Portland,OR,USA)[8]havebeendescribedbefore.
TheexpressionvectorforglutathioneS-transferase(GST)-c-Jun(aminoacids1–135)waskindlyprovidedbyM.
Kracht,MedicalSchoolHanover,Hanover,Germany[21].
ForGAL4–CREBafragmentencodingfulllengthCREB-327wasamplifiedbyPCRusing5'-ATATAGGATCCGTATGACCATGGAATCTGGA-3'and5'-ACGCGGAGCTCTTAATCTGATTTGTGGCAGT-3'asup-anddownstreamprimers,respectively.
TheplasmidRSV-CREB[22]servedastemplate.
ThePCRproductwasclonedintotheBamHIandSacIsitesofpSG424[23].
Allconstructswereconfirmedbysequencing.
CellcultureandtransfectionofDNAHIT-T15cells[12]weregrowninRPMI1640mediumsupplementedwith10%FCS,5%horseserum,penicillin(100U/ml)andstreptomycin(100μg/ml).
CellsweretransfectedbytheDEAE-dextranmethod[12]withindicatorplasmidandexpressionvectorasindicated.
CotransfectionswerecarriedoutwithaconstantamountofDNA,whichwasmaintainedbyadditionofBluescript(Stratagene,LaJolla,CA,USA).
Tocheckfortransfectionefficiency1μgcytomegalovirus-greenfluorescentprotein(plasmidCMV-GFPtpz)per6-cmdishwascotransfected.
Cellextractswereprepared48haftertransfection.
CellswerestimulatedbymembranedepolarisationwithKCl(40mmol/l)6hbeforeharvest.
Intheexperimentsusinghighglucosestimulation,thecellswereincubatedaftertransfectioninmediumcontaining0.
4mmolglucose/l;forstimulationwithhighglucose,theglucoseconcentrationwasraisedto20mmol/l24hbeforeharvest.
Theactivitiesofluciferaseandgreenfluorescentproteinweredeterminedasdescribedpreviously[12,24].
WesternblotanalysisHITcellswereharvestedinlysisbuffer(Tris10mmol/l,pH7.
4,TritonX-1001%,NonidetP-400.
5%,NaCl150mmol/l,sodiumfluoride20mmol/l,sodiumorthovana-date0.
2mmol/l,EGTA1.
0mmol/l,EDTA1.
0mmol/l,phenylmethylsulfonylfluoride0.
2mmol/l).
Cellswereincubatedfor10minatroomtemperaturein100μllysisbufferper6-cmdish,scrapedfromthedishandpassedtentimesthrougha26-gaugeneedle.
Lysateswerethencentrifugedat16,000gfor20minatroomtemperature.
Laemmligelloadingbuffer(Tris-HCl62.
5mmol/l,pH6.
8,SDS2%,glycerol10%,β-mercaptoethanol5%,bromphe-nolblue0.
5%)wasaddedtothesupernatant,thesampleswereboiled,subjectedtoSDS-PAGE(5.
5%gel)andtransferredtoanitrocellulosemembrane.
Themembranewasincubatedin5%fat-freedriedmilkinTRIS-bufferedsalineTween(Tris-HCl25mmol/l,pH7.
4,NaCl137mmol/l,KCl5mmol/l,CaCl20.
7mmol/l,MgCl20.
1mmol/l,Tween200.
2%)for2hatroomtemperatureandthenwithfreshTris-bufferedsalineTweensupplemen-tedwithantibodyagainstDLK[18]ina1:2,000dilutionovernightat4°C.
Beforeandafterincubationwiththesecondaryantibodythemembranewaswashedfourtimesfor10min,eachtimeinTris-bufferedsalineTween.
Antibody-antigencomplexwasdetectedwithECLreagents(AmershamBiosciencesUK,LittleChalfont,Bucks,UK).
RT-PCRTotalRNAwasextractedfromHITcellsandfromisolatedisletsofNMRImice[4]usingacommercialkit(RNeasy,Qiagen,Hilden,Germany).
Forfirst-strandcDNAsynthe-sis,randomhexamerprimers(AmershamBiosciences)wereused.
TheRTenzymewasobtainedfromInvitrogen333(SuperscriptIIreversetranscriptase;Carlsbad,CA,USA).
ForPCRamplificationthefollowingprimerswereused:upstreamprimer,5'-CCCAAGC·TTGGGAAGTCCCCTTTGAGG-3';anddownstreamprimer,5'-GGCGAGCTC·CAAAATCTGAGATCTTCACC-3',thedotsmarkingthebeginningofDLKcodingsequence(sizeoftheexpectedproduct:432bp).
PCRwithouttheRTstepservedascontrolforDNAcontamination;PCRusingtheDLKencodingexpressionvectorastemplateservedascontrolforthesize.
ImmunohistochemistryFrozen5-μmsectionsfrommousepancreaswerecutinacryostatat20°Candmounted.
Sectionswerefixedbychloroform-acetone,driedatroomtemperature,andrehydratedbyplacinginawetchamberfor30minandbyincubationwithPBSforafurther30min.
Allsubsequentstepswerecarriedoutinawetchamber.
Sectionswereincubatedovernightat4°CwithanantibodyagainstDLK[18](1:50dilution),washedfourtimeswithPBSandFig.
1DLKisexpressedintheinsulin-producingpancreaticisletbetacelllineHIT(a–c)andinprimarymouseislets(d–e).
aSketchofDLKindicatingtherelativepositionoftheprimersandthesizeoftheexpectedfragment.
bRT-PCRofDLKtranscripts.
AnRT-PCRproductoftheexpectedsizewasobtainedasshownbyagarosegelelectrophoresis(lane3).
Lane1,PCRusingtheexpressionvectorforDLKastemplate;lane2,PCRreactionwithtotalRNAbutomissionoftheRTenzyme;lane4,molecularsizemarkers.
cWesternblot.
WholeHITcellextractwassubjectedtoimmunoblotting.
Lanes1and2,abandcorrespondingtothesizeofDLK;lane3,2μgoftheexpressionvectorforDLKweretransientlytransfectedintoHITcells,wholecellextractwasprepared48haftertransfection,andasmallportionwassubjectedtoimmunoblottingtoserveasacontrol.
dRT-PCRofDLKtranscriptsfromisolatedmouseislets(lane4).
Lane1,PCRusingtheexpressionvectorforDLKastemplate;lane2,PCRreactionwithouttemplate;lane3,PCRreactionwithtotalRNAbutomissionoftheRTenzyme.
eImmunohistochemistryofmousepancreas.
Primarymouseisletsarestainedbyananti-DLKantibody.
Notethatcellsthroughouttheisletsarestained,indicatingtheexpressionofDLKininsulin-producingbetacells.
Nostainingwasseenwhenthefirstantibodywasomitted(specificitycontrol)(notshown).
Magnification:*400334incubatedwithatetramethylrhodamineisothiocyanate-labelledanti-rabbitantibodyfor15minatroomtempera-ture.
AfterwashingfourtimeswithPBSthesectionsweredehydratedbyincreasingethanolconcentrations.
DLKwasvisualisedbyaLeicamicroscopeDM5000DusingLeicaFW400software(Leica,Solms,Germany).
InvitrokinaseassayCellsweretreated48haftertransfectionwithcyclosporinA(5μmol/l)ortacrolimus(167nmol/l)for30minandharvestedtopreparewhole-cellextract.
Briefly,disheswerewashedoncewithice-coldPBS,cellswerescrapedintoEppendorftubesandwashedwithice-coldPBS.
Thesupernatantfractionwasdiscardedandthepelletwasresuspendedin50μlspeciallysisbuffer(Tris10mmol/l,pH7.
05,sodiumpyrophosphate30mmol/l,NaCl50mmol/l,TritonX-1001%,Na3VO42mmol/l,NaF50mmol/l,β-glycerophosphate20mmol/l,supplementedwithphenylmethylsuphonylfluoride0.
5mmol/l,leupeptin0.
5μg/ml,pepstatin0.
5μg/ml,ocadaicacid400nmol/l).
Cellswereincubatedonicefor20minandcentrifuged(10,000g,15min,4°C).
Theproteincontentinthesupernatantfrac-tionwasdeterminedaccordingtoBradford[25].
Lysateswerestoredat80°C.
Theinvitrokinaseassayusing30μgproteinandGST-c-Junassubstratewasperformedasdescribedpreviously[21].
TheamountofphosphorylatedGST-c-JunwasdeterminedbyaPhosphorImagerBasReader1500(FujiPhotoFilm,Tokyo,Japan)usingTINA2.
0software(Raytest,Straubenhardt,Germany).
Fig.
2InhibitionbyDLKofCREB-directedtranscriptionaftermembranedepolarisation.
Theplasmid5*Gal4E1BLuc(2μg)wascotransfectedwithanexpressionvectorforaGAL4–CREBfusionprotein(2μg)andwith0.
2,0.
6,2and6μgofanexpressionvectorforwild-typeDLK.
CellswerestimulatedwithKCl(40mmol/l)6hbeforeharvest.
LuciferaseactivityisexpressedrelativetothemeanvalueineachexperimentoftheactivitymeasuredintherespectivecontrolswithoutKCltreatment.
Valuesaremeans±SEMofthreeindependentexperiments,eachdoneinduplicateFig.
3EffectofDLKandaDLKmutantonCREB-directedtranscriptionaftermembranedepolarisation(a)oraftercyclicAMPtreatment(b).
Theplasmid5*Gal4E1BLuc(2μg)wascotransfectedwithGAL4–CREB(2μg)and2μgofBluescript(minussign)or2μgoftheexpressionvectorforwild-typeDLK(DLKwild-type)orDLKK185A(DLKmutant)asindicated.
CellsweretreatedwithKCl(40mmol/l;blackbars)(a)ortheadenylatecyclaseactivatorforskolin(10μmol/l;blackbars)(b)for6hbeforeharvest.
LuciferaseactivityisexpressedrelativetothemeanvalueineachexperimentoftheactivitymeasuredintherespectivecontrolswithoutKClorforskolintreatment.
Valuesaremeans±SEMofthreeindependentexperimentseachdoneinduplicate.
*p<0.
05vscontrol335MaterialsLuciferinwaspurchasedfromPJKIndustrievertretungen(Kleinblittersdorf,Germany);glutathioneagarosebeadsandforskolinwerepurchasedfromSigma(Taufkirchen,Germany).
CyclosporinAwasprovidedbyNovartisPharma(Basel,Switzerland);tacrolimus(FK506)byFujisawa(Osaka,Japan).
AstocksolutionofcyclosporinA(10mg/ml)waspreparedinethanolwith20%Tween80andfurtherdilutedinRPMI.
ForskolinwasdissolvedinDMSO,andtacrolimusinethanol.
Controlsreceivedthesolventonly.
StatisticalanalysisAllresultsareexpressedasmeans±SEM.
StatisticalsignificancewascalculatedwiththeWilcoxontestforpairedsamplesfornon-parametricdistribution.
Avalueofp<0.
05wasconsideredsignificant.
ResultsExpressionofDLKinthepancreaticisletbetacelllineHITandinprimarymouseisletsDLKhasbeenreportedtobeexpressedinatissue-specificfashionwithbyfarthehighestlevelsinthebrainandlowerlevelsinthekidneyandlung[18].
Itsexpressioninpancreaticisletshasnotyetbeenexamined.
UsingthepancreaticisletbetacelllineHIT,abandoftheexpectedsizewasdetectedbyRT-PCRandWesternblotanalysis(Fig.
1a),indicatingthatDLKisexpressedinthispancreaticisletbetacellline.
Inaddition,theexpression"Fig.
4InhibitionbyDLKofCRE-directedtranscription.
aEffectofDLKonCRE-directedtranscriptionaftermembranedepolarisation.
Theplasmidp4xSomCRELuc(2μg)wastransientlycotransfectedwith0.
2,0.
6,2and6μgofanexpressionvectorforwild-typeDLK.
CellswerestimulatedwithKCl(40mmol/l)for6hbeforeharvest.
Luciferaseactivityisexpressedrelativetothemeanvalueineachexperimentofthedepolarisation-inducedactivitymeasuredinthecontrols(noDLK).
Valuesaremeans±SEMofthreeindependentexperiments,eachdoneinduplicate.
*p<0.
05vscontrol.
bEffectofDLKanditsmutantonCRE-directedtranscriptionaftermembranedepolarisation.
Theplasmidp4xSomCRELuc(2μg)wastransientlycotransfectedwith0.
2μgofBluescript(minussign)or0.
2μgofanexpressionvectorforwild-typeDLK(DLKwild-type)ormutantDLK(DLKmutant).
KCl-stimulatedcells,blackbars.
LuciferaseactivityisexpressedrelativetothemeanvalueineachexperimentoftheactivitymeasuredintherespectivecontrolswithoutKCltreatment.
Valuesaremeans±SEMoffourindependentexperiments,eachdoneinduplicate.
*p<0.
05vscontrol.
cEffectofDLKanditsmutantonCRE-directedtranscriptionaftercyclicAMPtreatment.
Plasmidcotransfection,aspanel(b)withtheexceptionthat2μgofplasmidwereused.
Cellswerestimulatedwiththeadenylatecyclaseactivatorforskolin(10μmol/l;blackbars)for6hbeforeharvest.
Luciferaseactivityisexpressedrelativetothemeanvalueineachexperimentoftheactivitymeasuredintherespectivecontrolswithoutforskolintreatment.
Valuesaremeans±SEMoffourindependentexperiments,eachdoneinduplicate.
*p<0.
05vscontrol336ofDLKinprimarymouseisletswasdemonstratedbyRT-PCRandimmunohistochemistry(Fig.
1b).
InhibitionbyDLKofmembranedepolarisation-inducedCREBactivityinHITbetacellsToexaminetheeffectofDLKonmembranedepolarisa-tion-inducedCREBtranscriptionalactivityinthebetacellline,theGAL4systemwasused.
AnexpressionvectorencodingCREBfusedtotheDNAbindingdomainoftheyeasttranscriptionfactorGAL4wastransfectedintoHITcellstogetherwithaluciferasereportergenethatisdirectedbyfivecopiesoftheGAL4DNA-bindingsiteplacedinfrontofaminimalviralE1Bpromoter(Fig.
2,upperpanel).
High-potassium-inducedmembranedepolarisationstimu-latedCREBactivityaboutfour-fold(Fig.
2).
Thecotrans-fectionofincreasingamountsofanexpressionvectorforDLKcausedaninhibitionofmembranedepolarisation-inducedCREBactivity(Fig.
2).
Themaximuminhibitionwasabout45%(Fig.
2).
DLKhadnoeffectonbasalactivity(resultsnotshown).
TotestwhetherthekinaseactivityofDLKisrequiredfortheeffectofDLK,aDLKkinase-deadmutantwasemployedinwhichtheexchangeofLys185toAlawithintheATP-bindingdomainabolishesDLKkinaseactivity[19,26].
AsshowninFig.
3a,membranedepolarisation-inducedCREBactivitywasinhibitedbywild-typeDLKbutnotbytheDLKmutant,indicatingthattheeffectofDLKisspecificanddependsonitskinaseactivity.
CREBactivityinducedbytheadenylatecyclaseactivatorforskolinwasalsoreducedbycotrans-fectionoftheDLKexpressionvector(Fig.
3b),consistentwiththepreviousstudyinNIH3T3andNT2teratocarci-nomacells[14].
However,incontrasttostimulationbymembranedepolarisation(Fig.
3a),forskolin-inducedCREBactivitywasalsoreducedbytheDLKkinase-deadmutant(Fig.
3b).
Takentogether,thedatademonstratethatDLKinhibitedCREBactivityinducedbymembranedepolarisationinthepancreaticisletbetacellline.
InhibitionbyDLKofCRE-directedtranscriptionafterstimulationbymembranedepolarisationinHITbetacellsAlthoughotherCRE-bindingtranscriptionfactors,suchasATF-2,areknown[6],CREBhasbeenshownpreviouslytobethenuclearprotein,inextractsfromthebetacelllineHIT,thatbindstotheCRE;ithasalsobeenshowntoconfercyclicAMPandmembranedepolarisationresponsivenesstotheCREinthesebetacells[1–4].
ToexaminewhetherCREBthusconfersDLKresponsivenesstotheCREinpancreaticisletbetacells,aCRE-luciferasereportergenewasused.
"Fig.
5DLKinhibitsthetranscriptionalactivityoftheCREBcoactivatorCBPunderbasalconditionsandaftermembranedepolarisation.
aTheplasmid5*Gal4E1BLuc(2μg)wascotrans-fectedwithanexpressionvectorforGAL4–CBPC-terminus(2μg)andwith0.
2,0.
6,2and6μgoftheexpressionvectorforwild-typeDLK.
Cellswereleftuntreated(closedcircles)ortreatedwithKCl(40mmol/l;opencircles)6hbeforeharvest.
LuciferaseactivityisexpressedrelativetothemeanvalueineachexperimentoftheactivitymeasuredintherespectivecontrolswithoutKClandwithoutDLK.
Valuesaremeans±SEMofthreeindependentexperimentseachdoneinduplicate.
bEffectofDLKanditsmutantonthetranscriptionalactivityofCBP.
Theplasmid5xGal4E1BLuc(2μg)wascotransfectedwithanexpressionvectorforGAL4–CBP(2μg)andwith0.
6μgofBluescript(minussign)or0.
6μgoftheexpressionvectorforwild-typeDLK(DLKwild-type)orDLKK185A(DLKmutant)asindicated.
Luciferaseactivityisexpressedrelativetothemeanvalueineachexperimentoftheactivitymeasuredintherespectivecontrols(noKCl,noDLK).
Valuesaremeans±SEMoffourindependentexperimentseachdoneinduplicate.
*p<0.
05vscontrol337MembranedepolarisationstimulatedCRE-directedtran-scription4.
2±0.
6-fold(n=6)inthebetacelllineHIT.
ThecotransfectionofincreasingamountsoftheDLKexpressionvectormarkedlyinhibitedmembranedepolarisation-inducedCREtranscriptionalactivity(Fig.
4a),whereasthekinase-deadDLKmutantwasinactive(Fig.
4b).
Whenagreateramountoftheexpressionvectors(2μgperdishcomparedwith0.
2μginFig.
4b)wasused,DLKinhibitedde-polarisation-inducedCREactivityby86±4%(p<0.
05),whereasthekinase-deadDLKmutantproducednoeffect(n=8each).
Incontrast,forskolin-inducedCREactivitywasinhibitedbybothDLKandmutantDLK(Fig.
4c),similartoforskolin-inducedCREBactivity(Fig.
3b).
Theseresultsdemonstratethatafterstimulationbymembranedepolarisa-tion,CREBaswellasCREtranscriptionalactivityisinhibitedbyDLKkinaseactivityinthebetacelllineHIT.
InhibitionbyDLKofthetranscriptionalactivityofCBPinHITbetacellsCREBtranscriptionalactivityisgenerallythoughttobetransmittedtothegeneraltranscriptionmachinerythroughtheCREBcoactivatorCBP,whichisrecruitedbyCREBafterphosphorylationatSer119(inCREB-327)[7,8].
CBPisalargemodularprotein,anditsC-terminusisknowntoexhibitstrongtranscriptionalactivityandtoexertregula-toryfunctions[8,27].
ToexaminewhetherDLKmayhaveaneffectonCBP,aGAL4–CBPC-terminus(aminoacids1,678–2,441)fusionproteinwasusedtogetherwiththeGAL4–luciferasereportergene(Fig.
5a,top).
MembranedepolarisationstimulatedGAL4–CBPactivity3.
5-fold(Fig.
5a).
CotransfectionofincreasingamountsofDLKproducedamarkedinhibitionofGAL4–CBPtranscrip-tionalactivitybothunderbasalconditionsandafterstimulationbymembranedepolarisation(Fig.
5a).
Thekinase-deadDLKmutantwasinactive(Fig.
5b).
ThesedatademonstratethatDLKkinaseactivityinhibitsCBPactivityandtherebysuggestthatDLKmay,atleastinpart,inhibitdepolarisation-inducedCREBtranscriptionalactivityinpancreaticisletbetacellsatthecoactivatorlevel.
InhibitionbyDLKofCRE/CREBandCBPtranscriptionalactivityinresponsetoglucoseGlucoseisthemostpotentphysiologicalstimulusforpan-creaticisletbetacellsandisknowntocausemembranedepolarisation[15,28].
Useofhighglucoseasamorephys-iologicalstimulusformembranedepolarisationenhancedCRE-andCBP-dependenttranscriptionabout2.
3-fold(Fig.
6).
CotransfectionofDLKinhibitedthisglucose-inducedtranscriptionalactivity(Fig.
6).
ThesedatasuggestthatDLKinhibitsCRE-mediatedgenetranscriptiontoasimilardegreewhenmembranedepolarisationiseffectedbymorephysiologicalapproaches.
CalcineurinsensitivityoftheinhibitionbyDLKinHITbetacellsInaggregatingneuronal-glialcultures,theinhibitionbycyclosporinAofthecalcium/calmodulin-dependentproteinphosphatasecalcineurinpreventedamembranedepolarisa-tion-inducedshiftintheelectrophoreticmobilityofDLK[19],suggestingthatthephosphorylationstateofDLKmayFig.
6DLKinhibitsCREandCBPtranscriptionalactivityinducedbyglucose.
aCRE-mediatedtranscription.
Theplasmidp4xSom-CRELuc(2μg)wastransientlycotransfectedwith2μgofBluescript(minussign)or2μgofanexpressionvectorforwild-typeDLK(DLK).
Cellswerestimulatedbyhighglucose(20mmol/l;blackbars)asindicated.
Luciferaseactivityisexpressedrelativetothemeanvalueineachexperimentoftheactivitymeasuredintherespectivecontrols(lowglucose).
Valuesaremeans±SEM,n=6.
*p<0.
05vshighglucosewithoutDLK.
bCBPtranscriptionalactivity.
Theplasmid5*Gal4E1BLuc(2μg)wascotransfectedwithanexpressionvectorforGAL4–CBP(2μg)andwith2μgofBluescript(minussign)or2μgoftheexpressionvectorforwild-typeDLK(DLK).
Cellswerestimulatedbyhighglucose(20mmol/l;blackbars)asindicated.
Luciferaseactivityisexpressedrelativetothemeanvalueineachexperimentoftheactivitymeasuredintherespectivecontrols(lowglucose).
Valuesaremeans±SEM,n=6.
*p<0.
05vshighglucosewithoutDLK338beregulatedbycalcineurin.
InordertoinvestigatewhethertheDLKkinaseactivityisregulatedbycalcineurininthepancreaticisletbetacelllineHIT,anexvivoassaywasperformed.
AsaMAPKKK,DLKisknowntoactivatethec-JunN-terminalkinase(JNK),whoseactivitycanbemeasuredincelllysates[21].
HITcellsweretransfectedwiththeDLKexpressionvectorortheemptyvectorandJNKactivitywasmeasuredincelllysatesusingGST-c-Junassubstrate.
AsshowninFig.
7a,treatmentofHITcellswithtwochemicallydistinctinhibitorsofcalcineurin,cyclosporinAandtacrolimus,enhancedJNKactivityonlyinDLK-transfectedcells,suggestingthatinhibitionofcalcineurinincreasesDLKactivityinthisbetacellline.
ThelackofeffectofcyclosporinAandtacrolimusincellstransfectedwiththeemptyvector(Fig.
7a)maybeduetothesensitivityoftheassayusedthatisinsufficienttodetectendogenousDLKactivity.
InviewoftheevidencethatDLKkinaseactivityisregulatedbycalcineurininHITcells(Fig.
7a),thecalcineurinsensitivityoftheinhibitionbyDLKofCRE/CREB-directedtranscriptionwasexaminednext.
AsshowninFig.
7b,theinhibitionbyDLKofmembranedepolarisa-tion-inducedCREtranscriptionalactivitywasdiminishedafteroverexpressionofcalcineurin(p<0.
05)(n=6).
ThesedatafurthersupporttheviewthatDLKkinaseactivityandthustheinhibitionbyDLKofdepolarisation-inducedCRE/CREBtranscriptionalactivityiscalcineurin-sensitive.
DiscussionFeedingtriggersaseriesofhormonalandnutrientcuesthatactonpancreaticisletbetacellstopromoteinsulinrelease,insulingenetranscriptionandbetacellsurvival[15,28].
Elevationsinbloodglucoseprovidethemostpotentstimulus.
Uponitsuptakeintobetacells,glucoseoxidationinthepancreaticbetacellinducesclosureofKATPchannels,membranedepolarisation,calciumentryviavoltage-sensitiveL-typecalciumchannels,andconsequentphosphorylationandthusactivationofthetranscriptionfactorCREB[5,15,29],whichisessentialforbetacellfunction[1–5].
Twomechanismsforlinkingmembranedepolarisation-inducedelevationsofintracellularcalciumtotheactivationofCREBhavebeenproposed,mostlybasedonstudiesinneuronalsystems.
First,calciumitselfmayenterthenucleusandactivatenuclearCaMkinaseIVandleadtophosphorylationofCREBonSer119.
Thismayholdtrueshortlyafterdepolarisation[30].
Alternatively,calciummayactlocallynearthemembraneandclosetothemouthoftheL-typevoltage-dependentcalciumchannel.
Amodelhasbeenproposedinwhichcalciumbindscalmod-ulintetheredtothechannel,causingittoassociatewiththeIQmotifofthechannel;thisassociationthenleadstotheactivationoftheRas-mitogen-activatedextracellularsig-nal-regulatedkinasekinase-ERK1/2MAPKsignallingpathwayandsustainedCREBSer119phosphorylationFig.
7DLKactivityiscalcineurin-sensitive.
aThecalcineurininhibitorscyclosporinA(CsA)andtacrolimus(FK506)enhanceDLK-dependentc-Junphosphorylation.
HITcellsweretransientlytransfectedwithanexpressionvectorforDLKortheemptyvector(6μgeach).
After48hcellsweretreatedwithCsA(5μmol/l)orFK506(167nmol/l)for30minasindicated.
Whole-cellextractswerepreparedandaninvitrokinaseassaywasperformedusingGST-c-Jun(aminoacids1–135)assubstrate.
Phosphorylationofc-Junisexpressedrelativetothephosphorylationofc-Junintheabsenceoftreatment.
Valuesaremeans±SEMofsixindependentexperiments.
*p<0.
05vscontrol.
bOverexpressionofcalcineurin(CN)attenuatedtheinhibitoryeffectofDLKonCRE-directedtranscriptionaftermembranedepolarisation.
Theplasmidp4*Som-CRELuc(1μg)wastransientlycotransfectedwith2μgofanex-pressionvectorforwild-typeDLK(DLK),or2μg,each,oftheexpressionvectorsfortheCNsubunitsAandBor4μgofemptyvector(minussign)orDLKplusCNasindicated.
CellsweretreatedwithKCl(40mmol/l)for6hbeforeharvest.
LuciferaseactivityisexpressedrelativetothemeanvalueineachexperimentoftherespectivecontroltreatedwithKCl(noDLK).
Valuesaremeans±SEMofthreeindependentexperimentseachdoneinduplicate.
*p<0.
05,comparinginhibitionbyDLKinabsenceandpresenceofCN339[30,31].
Inadditiontothesekinases,thecalcium/calmodulin-dependentproteinphosphatasecalcineurinisalsorequiredforCREBactivationbymembranedepolar-isation[3,16,17].
Thepresentstudynowdemonstratesthat,incontrasttothestimulationbyERK1/2,anothermemberoftheMAPKsignallingpathways,DLK,cannegativelyregulatemembranedepolarisation-inducedCREBactivityinabetacellline.
DLKisaserine/threoninekinaseandbelongstothemixed-lineagekinasesthatfunctionasMAPKKK[18,32].
Assuch,DLKphosphorylatesandactivatesthedual-specificityMAPKKMKK7andMKK4,whichinturnphosphorylateandactivatetheMAPKJNK[26,33,34].
C-terminalofthekinasedomainDLKhastwoleucinezippermotifsthatmediateproteindimerisationbyformingcoiled-coilstructures[18].
However,withinthecellandunderbasalconditions,DLKisassociatedwiththescaffoldproteinsJNKinteractingprotein(JIP)[34–36]andPlentyofSH3[36],whichassembleandfacilitatetheactivationoftheDLK-dependentJNKmodule.
JIPmaintainsDLKinamonomeric,unphosphorylated,inactivestate[35,36].
ItisassumedthatsignalspromotingthedissociationofDLKfromJIPleadtohomodimerisationofDLKviaitsleucinezipperdomains,followedbyitsautophosphorylationandactivation[35–37].
ThustheoverexpressionofDLKalsoresultsinitsactivation[34].
Thepresentstudydemon-stratesthatDLKisexpressedintheinsulin-producingpancreaticisletbetacelllineHITandinprimarymouseisletsincludingbetacells.
Itshowsfurthermorethat,afteroverexpressionandthusactivation,DLKinhibitsmem-branedepolarisation-inducedCREBtranscriptionalactiv-itythroughitskinaseactivityinthebetacellline.
TheinhibitionbyDLKisconferredtotheCREbyCREB(presentstudy),whichhasbeenshownpreviouslytobethenuclearproteininHITbetacellsthatbindstotheCREandmediatesmembranedepolarisationresponsivenessoftheCRE[1–4].
InNIH3T3andNTera-2teratocarcinomacelllinestheoverexpressionofthehumanorthologueofDLK,ZPK,hasbeenshowntoinhibitCREBactivityafterstimulationbyproteinkinaseA[14],whichisconfirmedbythepresentstudyinHITbetacells.
However,incontrasttostimulationbymembranedepolarisation,proteinkinaseA-inducedCREBactivitywasinhibitedalsobythekinase-deadmutantofDLK(presentstudy),suggestingthatthemechanismofinhibitionbyDLKmaydifferdependingonthestimulusused.
ZPKisabletointeractdirectlywithCREBandtophosphorylateaCREBpeptidethatcontainstheCREBleucinezipperdomain[14].
Althoughthefunctionalconsequences,ifany,oftheseeffectsofDLKareunknown,theyraisethepossibilitythatproteinkinaseAmightallowDLK–CREBinteraction,whichmayinterferewithCREBtransactivationindependentlyoftheDLKkinaseactivity.
OurresultssuggestanothersiteofactionforDLKafterstimulationofCREBbymembranedepolarisa-tion,namelytheCREBcoactivatorCBP.
InadditiontothephosphorylationonSer119,CREBtranscriptionalactivityisregulatedbymembranedepolar-isation-inducedsignallingpathwaysatadditionalstepsofCREB-mediatedtranscriptionalactivation[7,27].
ItwasthoughtinitiallythattheCREBcoactivatorCBPconfersconstitutivetranscriptionalactivitytopromoter-boundCREB.
Meanwhile,itbecameclearthatmembranedepolarisation-inducedcalciuminfluxregulatesnotonlyCREBphosphorylationatSer119andthustherecruitmentofCBP[7,29],butalsotargetsCBPdirectly.
ConsistentwithpreviousstudiesinthemousepituitarycelllineAtT20[27],incorticalneurons[38],inhippocampalneurons[39]andinHITbetacells[24],thepresentstudyconfirmsthatmembranedepolarisationstimulatesCBPtranscriptionalactivity.
ItdemonstratesfurthermorethatDLKthroughitskinaseactivitymarkedlyinhibitsCBPtranscriptionalactivitybothunderbasalconditionsandafterstimulationbymembranedepolarisation.
SeveralkinaseshavebeenshowntoactivateCBP/p300,includingcalcium/calmodu-lin-dependentkinaseIV,MAPKKK1andERK1/2[27,39–41].
Asshowninthepresentstudy,DLKisthefirstexampleofakinasethatinhibitsCBPtranscriptionalactivity.
Takentogether,ourfindingssuggestthatDLKinhibitsmembranedepolarisation-inducedCREBactivitybyacting,atleastinpart,ontheCREBcoactivatorCBP.
ThisinhibitionbyDLKappearstobesubjectedtoregulationbycalcineurinphosphatase.
Calcineurinisacalcium/calmodulin-dependentserine/threonineproteinphosphatasethatisactivatedbymem-branedepolarisation-inducedcalciuminfluxandisin-hibitedspecificallybytheimmunosuppressivedrugscyclosporinAandtacrolimus[42].
Inaggregatingneuro-nal-glialcultures,depolarisationofplasmamembraneledtodephosphorylationofDLKthatwasblockedbycyclosporinA,suggestingthatthephosphorylationstateofDLKisregulatedbymembranedepolarisationviacalcineurin[19].
ConsistentwiththeviewthatlackofautophosphorylationrendersDLKenzymicallyinactive[34],DLKkinaseactivitywasshowninthepresentstudytobeenhancedbycyclosporinAandtacrolimus,whichareknowntoinhibitcalcineurinphosphataseactivitywithhighpotencyalsoinpancreaticisletbetacells[4,16,17,29].
Ontheotherhand,overexpressionofcalcineurindecreasedtheinhibitionbyDLKofmembranedepolarisation-inducedCRE/CREBtranscriptionalactivity.
Whentakentogether,thesedatasuggestthatcalcineurinregulatesthephosphor-ylationstateandthekinaseactivityofDLKandtherebycounteractstheinhibitionbyDLKofdepolarisation-inducedCREBactivityattheleveloftheCREBcoactivatorCBP.
InadditiontotheregulationofDLKactivity,othermechanismsexistthroughwhichcalcineurincaninfluenceJNK[43–45].
Interestingly,ithasbeenshownpreviouslythatcyclo-sporinAandtacrolimusinhibitCRE-directedtranscriptionafterstimulationbymembranedepolarisation-inducedcalciuminfluxinbetacellswithoutdecreasingthephos-phorylationofCREBonSer119[3,4,16,17,29].
SeverallinesofevidencesuggestthatthisinhibitionbycyclosporinAandtacrolimusismediatedviacalcineurin[3,4,16,17,29]and,atleastinpart[46],takesplaceattheleveloftheCREBcoactivatorCBP[24].
Inviewofthefindingsofthepresentstudythat(1)DLKkinaseactivityiscalcineurin-340sensitive,andthat(2)DLKinhibitsdepolarisation-inducedCREBandCBPtranscriptionalactivity,thereisapossibilitythat,inadditiontotheTransducerofregulatedCREBactivity[46],DLKmaybeanothercalcineurinsubstrateinvolvedinthestimulationofCREBtranscrip-tionalactivitybymembranedepolarisationanditsinhibi-tionbytheimmunosuppressivedrugscyclosporinAandtacrolimus.
However,thepart,ifany,ofDLKintheseresponsesisunknownandremainstobedemonstrated.
SeverallinesofevidencesuggestthattheregulationbyDLKofdepolarisation-inducedCREB/CREactivitymayplayacriticalroleinbetacellfunction.
Amajorstimulusthatcausesmembranedepolarisationinpancreaticisletbetacellsisglucose,andthepresentstudyshowsthatDLKinhibitsalsoglucose-inducedCREB/CREtranscriptionalactivity.
Glucoseincreasescalcineurinphosphataseactivityandinsulingenetranscriptioninnormalbetacells[3,4].
ThehumaninsulingenecarriesfourcopiesoftheCREandisregulatedbyCREB[3,4].
Inadditiontotheactivationofinsulingenetranscription,CREBisessentialforbetacellsurvival,sincemiceexpressingadominant-negativeCREBmutantdevelopdiabetesmellitussecondarytobetacellapoptosis[5].
WhereasactivationoftheMAPKERK1/2hasbeenlinkedtocellsurvival,DLKisgenerallylinkedtoinductionofapoptosis[32,47].
StudiesinneuronalsystemsindicatethatDLKmightbeanimportantkinasethatmediatesgrowth-factor-deprivation-inducedneuronalapoptosis[32,47].
DLKmayfunctionsimilarlyintheelectricallyexcitablepancreaticisletbetacells.
DLKisthoughttobeactivatedbycellularstressandinflammatorycytokines[32,35].
Reactiveoxygenspeciesthatareformedundertheseconditions[48]arelikelytoinhibitglucose-inducedcalcineurinphosphataseactivityinbetacells[42,49],thusfurtherpromotingDLKkinaseactivity.
ItistemptingtospeculatethatDLKkinaseactivitymaythencompromisebetacellfunctionthroughinhibitionofCBP/CREBactivityandaconsequentdecreaseininsulinsynthesisandinductionofbetacellapoptosis.
Consistentwiththisview,thedownregulationormutationoftheDLK-bindingscaffoldproteinJIP1hasbeenshowntoleadtodecreasedinsulingeneexpression,highersusceptibilitytoapoptoticstimulianddiabetesmellitus[50,51].
AcknowledgementsWethankourcolleaguesF.
Vetterlein(Depart-mentofAnaesthesiologicalResearch,UniversityofGttingen)andP.
Hülper(Children'sHospital,PaediatricsI,UniversityofGttingen)fortheirvaluableandgenerousadviceandhelpinperformingtheimmunohistochemicalanalysis.
ThisworkwassupportedbyagrantfromtheDeutscheForschungsgemeinschaftSFB402(A3).
References1.
EggersA,SiemannG,BlumeR,KnepelW(1998)Gene-specifictranscriptionalactivityoftheinsulincAMP-responsiveelementisconferredbyNF-YincombinationwithcAMPresponseelement-bindingprotein.
JBiolChem273:18499–185082.
OetjenE,DiedrichT,EggersA,EckertB,KnepelW(1994)DistinctpropertiesofthecAMP-responsiveelementoftheratinsulinIgene.
JBiolChem269:27036–270443.
OetjenE,GrapentinD,BlumeRetal(2003)RegulationofhumaninsulingenetranscriptionbytheimmunosuppressivedrugscyclosporinAandtacrolimusatconcentrationsthatinhibitcalcineurinactivityandinvolvingthetranscriptionfactorCREB.
NaunynSchmiedebergsArchPharmacol367:227–2364.
OetjenE,BaunD,BeimescheSetal(2003)InhibitionofhumaninsulingenetranscriptionbytheimmunosuppressivedrugscyclosporinAandtacrolimusinprimary,matureisletsoftransgenicmice.
MolPharmacol63:1289–12955.
JhalaUS,CanettieriG,ScreatonRAetal(2003)cAMPpromotespancreaticβ-cellsurvivalviaCREB-mediatedin-ductionofIRS2.
GenesDev17:1575–15806.
HabenerJF,MillerCP,VallejoM(1995)cAMP-dependentregulationofgenetranscriptionbycAMPresponseelement-bindingproteinandcAMPresponseelementmodulator.
VitamHorm51:1–577.
MayrB,MontminyM(2001)Transcriptionalregulationbythephosphorylation-dependentfactorCREB.
NatRev/MolCellBiol2:599–6098.
KwokRPS,LundbladJR,ChriviaJCetal(1994)NuclearproteinCBPisacoactivatorforthetranscriptionfactorCREB.
Nature370:223–2269.
CarlezonWAJr,ThomeJ,OlsonVGetal(1998)RegulationofcocainerewardbyCREB.
Science282:2272–227510.
DawsonTM,GintyDD(2002)CREBfamilytranscriptionfactorsinhibitneuronalsuicide.
NatMed8:450–45111.
HerzigS,LongF,JhalaUSetal(2001)CREBregulateshepaticgluconeogenesisthroughcoactivatorPGC-1.
Nature413:179–18312.
SchwaningerM,LuxG,BlumeR,OetjenE,HidakaH,KnepelW(1993)MembranedepolarizationandcalciuminfluxinduceglucagongenetranscriptioninpancreaticisletcellsthroughthecyclicAMP-responsiveelement.
JBiolChem268:5168–517713.
DalleS,LonguetC,CostesSetal(2004)GlucagonpromotescAMP-responseelement-bindingproteinphosphorylationviaactivationofERK1/2inMIN6celllineandisolatedisletsofLangerhans.
JBiolChem279:20345–2035514.
ReddyUR,BasuA,BannermannP,IkegakiN,ReddyCD,PleasureD(1999)ZPKinhibitsPKAinducedtranscriptionalactivationbyCREBandblocksretinoicacidinducedneuronaldifferentiation.
Oncogene18:4474–444415.
DochertyK,ClarkAR(1994)Nutrientregulationofinsulingeneexpression.
FASEBJ8:20–2716.
SchwaningerM,BlumeR,OetjenE,KnepelW(1993)TheimmunosuppressivedrugscyclosporinAandFK506inhibitcalcineurinphosphataseactivityandgenetranscriptionmediatedthroughthecAMP-responsiveelementinanon-immunecellline.
NaunynSchmiedebergsArchPharmacol348:541–54517.
SchwaningerM,BlumeR,OetjenE,LuxG,KnepelW(1993)InhibitionofcAMP-responsiveelement-mediatedgenetran-scriptionbycyclosporinAandFK506aftermembranedepolarization.
JBiolChem268:23111–2311518.
HolzmanLB,MerrittSE,FanG(1994)Identification,molec-ularcloning,andcharacterizationofdualleucinezipperbearingkinase.
JBiolChem269:30808–3081719.
MataM,MerrittSE,FanG,YuGG,HolzmanLB(1996)Characterizationofdualleucinezipper-bearingkinase,amixedlineagekinasepresentinsynapticterminalswhosephosphor-ylationstateisregulatedbymembranedepolarizationviacalcineurin.
JBiolChem271:16888–1689620.
KincaidRL,GiriPR,HiguchiSetal(1990)Cloningandcharacterizationofmolecularisoformsofthecatalyticsubunitofcalcineurinusingnonisotopicmethods.
JBiolChem265:11312–1131921.
KrauseA,HoltmannH,EickemeierSetal(1998)Stress-activatedproteinkinase/JunN-terminalkinaseisrequiredforinterleukin(IL)-1-inducedIL-6andIL-8geneexpressioninthehumanepidermalcarcinomacelllineKB.
JBiolChem273:23681–2368934122.
WaltonKM,RehfussRP,ChriviaJC,LochnerJE,GoodmanRH(1992)Adominantrepressorofcyclicadenosine3',5'-monophosphate(cAMP)-regulatedenhancer-bindingproteinactivityinhibitsthecAMP-mediatedinductionofthesomato-statinpromoterinvivo.
MolEndocrinol6:647–65523.
SadowskiI,PtashneM(1989)AvectorforexpressingGAL4(1-147)fusionsinmammaliancells.
NuclAcidsRes17:753924.
OetjenE,ThomsKM,LauferYetal(2005)Theimmunosup-pressivedrugscyclosporinAandtacrolimusinhibitmembranedepolarization-inducedCREBtranscriptionalactivityattheco-activatorlevel.
BritJPharmacol144:982–99325.
BradfordMM(1976)Arapidandsensitivemethodforthequantitationofmicrogramquantitiesofproteinutilizingtheprincipleofprotein–dyebinding.
AnalBiochem72:248–25426.
MerrittSE,MataM,NihalaniD,ZhuC,HuX,HolzmanLB(1999)ThemixedlineagekinaseDLKutilisesMKK7andnotMKK4assubstrate.
JBiolChem274:10195–1020227.
ChawlaS,HardinghamGE,QuinnDR,BadingH(1998)CBP:asignal-regulatedtranscriptionalcoactivatorcontrolledbynuclearcalciumandCaMkinaseIV.
Science281:1505–150928.
OhnedaK,EeH,GermanM(2000)Regulationofinsulingenetranscription.
SeminCellDevBiol11:227–23329.
SchwaningerM,BlumeR,KrügerM,LuxG,OetjenE,KnepelW(1995)InvolvementoftheCa2+-dependentphosphatasecalcineuriningenetranscriptionthatisstimulatedbycAMPthroughcAMPresponseelements.
JBiolChem270:8860–886630.
DolmetschRE,PajvaniU,FifeK,SpottsJM,GreenbergME(2001)SignalingtothenucleusbyanL-typecalciumchannel–calmodulincomplexthroughtheMAPkinasepathway.
Science294:333–33931.
GomezE,PritchardC,HerbertTP(2002)cAMP-dependentproteinkinaseandCa2+influxthroughL-typevoltage-gatedcalciumchannelsmediateRaf-independentactivationofextra-cellularregulatedkinaseinresponsetoglucagon-likepeptide-1inpancreaticβ-cells.
JBiolChem277:48146–4815132.
GalloKA,JohnsonGL(2002)Mixed-lineagekinasecontrolofJNKandp38MAPKpathways.
NatRev/MolCellBiol3:663–67233.
FanG,MerritSE,KortenjannM,ShawPE,HolzmanLB(1996)Dualleucinezipper-bearingkinase(DLK)activatesp46SAPKandp38mapkbutnotERK2.
JBiolChem271:24788–2479334.
NihalaniD,MerrittS,HolzmanLB(2000)Identificationofstructuralandfunctionaldomainsinmixedlineagekinasedualleucinezipper-bearingkinaserequiredforcomplexformationandstress-activatedproteinkinaseactivation.
JBiolChem275:7273–727935.
NihalaniD,MeyerD,HolmanLB(2001)Mixedlinagekinase-dependentJNKactivationisgovernedbyinteractionsofscaffoldproteinJIPwithMAPKmodulecomponents.
EMBOJ20:3447–345836.
NihalaniD,WongHN,HolzmanLB(2003)RecruitmentofJNKtoJIPandJNK-dependentJIP1phosphorylationregulatesJNKmoduledynamicsandactivation.
JBiolChem278:28694–2870237.
XuZ,KukekovNV,GreeneLA(2003)POSHactsasascaffoldforamultiproteincomplexthatmediatesJNKactivationinapoptosis.
EMBOJ22:252–26138.
HuSC,ChriviaJ,GoshA(1999)RegulationofCBP-mediatedtranscriptionbyneuronalcalciumsignaling.
Neuron22:799–80839.
ImpeyS,FongAL,WangYetal(2002)PhosphorylationofCBPmediatestranscriptionalactivationbyneuralactivityandCaMKinaseIV.
Neuron34:235–24440.
SeeRH,CalvoD,ShiYetal(2001)Stimulationofp300-mediatedtranscriptionbythekinaseMEKK1.
JBiolChem276:16310–1631741.
JanknechtR,NordheimA(1996)Regulationofthec-fospromoterbytheternarycomplexfactorSap-1aanditsco-activatorCBP.
Oncogene12:1961–196942.
YakelJL(1997)Calcineurinregulationofsynapticfunction—fromionchannelstotransmitterreleaseandgenetranscription.
TrendsPharmacolSci18:124–13443.
MatsudaS,MoriguchiT,KoyasuS,NishidaE(1998)Tlymphocyteactivationsignalsforinterleukin-2productioninvolveactivationofMKK6-p38andMKK7-SAPK/JNKsig-nallingpathwayssensitivetocyclosporinA.
JBiolChem273:12378–1238244.
MuratA,PellieuxC,BrunnerHR,PedrazziniT(2000)Calcineurinblockadepreventscardiacmitogen-activatedpro-teinkinaseactivationandhypertrophyinrenovascularhyper-tension.
JBiolChem275:40867–4087345.
DeWindtLJ,LimHW,HaqS,ForceT,MolkentinJD(2000)CalcineurinpromotesproteinkinaseCandc-JunNH2-terminalkinaseactivationintheheart.
JBiolChem275:13571–1357946.
ScreatonRA,ConkrightMD,KatohYetal(2004)TheCREBcoactivatorTORC2functionsasacalcium-andcAMP-sensitivecoincidencedetector.
Cell119:61–7447.
ChangL,KarinM(2001)MammalianMAPkinasessignallingcascades.
Nature410:37–4048.
EvansJL,GoldfineID,MadduxBA,GrodskyGM(2002)Oxidativestressandstress-activatedsignalingpathways:aunifyinghypothesisoftype2diabetes.
EndocrRev23:599–62249.
NamgaladzeD,HoferHW,UllrichV(2002)RedoxcontrolofcalcineurinbytargetingthebinuclearFe(2+)–Zn(2+)centerattheenzymeactivesite.
JBiolChem277:5962–596950.
WaeberG,DelplanqueJ,BonnyCetal(2000)ThegeneMAPK8IP1,encodingislet-brain-1,isacandidatefortype2diabetes.
NatGenet24:291–29551.
LingZ,VandeCasteeleM,DongJetal(2003)VariationsinIB1/JIP1expressionregulatesusceptibilityofbeta-cellstocytokine-inducedapoptosisirrespectiveofc-JunNH2-terminalkinasesignaling.
Diabetes52:2497–2502342