<%BANNER%>

UFIR



The Tyrphostin Agent AG490 Prevents and Reverses Type 1 Diabetes in NOD Mice
http://www.plosone.org/home.action;jsessionid=B6C4E0B5634110D75DD3A0B8CF7AF9BD ( Publisher's URL )
CITATION PDF VIEWER
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/IR00001304/00001
 Material Information
Title: The Tyrphostin Agent AG490 Prevents and Reverses Type 1 Diabetes in NOD Mice
Series Title: 73. Davoodi-Semiromi A, Wasserfall CH, Xia CQ, Cooper-DeHoff RM, Wabitsch M, Clare-Salzler M, Atkinson M. The Tyrophostin agent AG-490 Prevents and Reverses Type 1 Diabetes in NOD Mice. PLoS One. 2012;7(5):e36079
Physical Description: Journal Article
Creator: Dehoff, Rhonda
Davoodi-Semiromi, Abdoreza
Atkinson, Mark
Clare-Salzler, Michael
Wabitsch, Martin
Xia, Chang Qing
Wasserfall, Clive H.
Publisher: PLoS One
Place of Publication: Cambridge, United Kingdon
Publication Date: May 14, 2012
 Notes
Abstract: Background Recent studies in the NOD (non-obese diabetic) mouse model of type 1 diabetes (T1D) support the notion that tyrosine kinase inhibitors have the potential for modulating disease development. However, the therapeutic effects of AG490 on the development of T1D are unknown. Materials and Methods Female NOD mice were treated with AG490 (i.p, 1 mg/mouse) or DMSO starting at either 4 or 8 week of age, for five consecutive week, then once per week for 5 additional week. Analyses for the development and/or reversal of diabetes, insulitis, adoptive transfer, and other mechanistic studies were performed. Results AG490 significantly inhibited the development of T1D (p = 0.02, p = 0.005; at two different time points). Monotherapy of newly diagnosed diabetic NOD mice with AG490 markedly resulted in disease remission in treated animals (n = 23) in comparision to the absolute inability (0%; 0/10, p = 0.003, Log-rank test) of DMSO and sustained eugluycemia was maintained for several months following drug withdrawal. Interestingly, adoptive transfer of splenocytes from AG490 treated NOD mice failed to transfer diabetes to recipient NOD.Scid mice. CD4 T-cells as well as bone marrow derived dendritic cells (BMDCs) from AG490 treated mice, showed higher expression of Foxp3 (p<0.004) and lower expression of co-stimulatory molecules, respectively. Screening of the mouse immune response gene arrary indicates that expression of costimulaotry molecule Ctla4 was upregulated in CD4+ T-cell in NOD mice treated with AG490, suggesting that AG490 is not a negative regulator of the immune system. Conclusion The use of such agents, given their extensive safety profiles, provides a strong foundation for their translation to humans with or at increased risk for the disease.
Acquisition: Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by Rhonda Dehoff.
Publication Status: Published
Funding: This study was supported by grants from the Juvenile Diabetes Research Foundation International (Innovative grant # 5-2006-937), and National Institutes of Health R21A176394-01 (The American Recovery And Reinvestment Act) awarded to ADS. Publication of this article was funded in part by the University of Florida Open-Access Publishing Fund. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
 Record Information
Source Institution: University of Florida Institutional Repository
Holding Location: University of Florida
Rights Management:
This item is licensed with the Creative Commons Attribution License. This license lets others distribute, remix, tweak, and build upon this work, even commercially, as long as they credit the author for the original creation.
System ID: IR00001304:00001

Downloads

This item is only available as the following downloads:

PLOSone_2012 ( PDF )


Full Text

PAGE 1

TheTyrphostinAgentAG490PreventsandReverses Type1DiabetesinNODMiceAbdorezaDavoodi-Semiromi1* ,CliveH.Wasserfall2,ChangQingXia2,RhondaM.Cooper-DeHoff1, MartinWabitsch3,MichaelClare-Salzler2,MarkAtkinson2*1 DepartmentofPharmacotherapy&TranslationalResearch,CollegeofPharmacy,UniversityofFlorida,Gainesville,Florida,UnitedStatesofAmeri ca, 2 Departmentof Pathology,ImmunologyandLaboratoryMedicine,UniversityofFlorida,Gainesville,Florida,UnitedStatesofAmerica, 3 DivisionofPediatricEndocrinology,Diabetesand ObesityUnit,DepartmentofPediatricsandAdolescentMedicine,UniversityofUlm,Eythstr,Ulm,GermanyAbstractBackground:RecentstudiesintheNOD(non-obesediabetic)mousemodeloftype1diabetes(T1D)supportthenotionthat tyrosinekinaseinhibitorshavethepotentialformodulatingdiseasedevelopment.However,thetherapeuticeffectsof AG490onthedevelopmentofT1Dareunknown.MaterialsandMethods:FemaleNODmiceweretreatedwithAG490(i.p,1mg/mouse)orDMSOstartingateither4or8 weekofage,forfiveconsecutiveweek,thenonceperweekfor5additionalweek.Analysesforthedevelopmentand/or reversalofdiabetes,insulitis,adoptivetransfer,andothermechanisticstudieswereperformed.Results:AG490significantlyinhibitedthedevelopmentofT1D(p=0.02,p=0.005;attwodifferenttimepoints). MonotherapyofnewlydiagnoseddiabeticNODmicewithAG490markedlyresultedindiseaseremissionintreatedanimals (n=23)incomparisiontotheabsoluteinability(0%;0/10,p=0.003,Log-ranktest)ofDMSOandsustainedeugluycemiawas maintainedforseveralmonthsfollowingdrugwithdrawal.Interestingly,adoptivetransferofsplenocytesfromAG490 treatedNODmicefailedtotransferdiabetestorecipientNOD. Scid mice.CD4T-cellsaswellasbonemarrowderived dendriticcells(BMDCs)fromAG490treatedmice,showedhigherexpressionofFoxp3(p 0.004)andlowerexpressionofcostimulatorymolecules,respectively.Screeningofthemouseimmuneresponsegenearraryindicatesthatexpressionof costimulaotrymoleculeCtla4wasupregulatedinCD4 + T-cellinNODmicetreatedwithAG490,suggestingthatAG490is notanegativeregulatoroftheimmunesystem.Conclusion:Theuseofsuchagents,giventheirextensivesafetyprofiles,providesastrongfoundationfortheirtranslation tohumanswithoratincreasedriskforthedisease.Citation: Davoodi-SemiromiA,WasserfallCH,XiaCQ,Cooper-DeHoffRM,WabitschM,etal.(2012)TheTyrphostinAgentAG490PreventsandReversesType1 DiabetesinNODMice.PLoSONE7(5):e36079.doi:10.1371/journal.pone.0036079 Editor: PaoloFiorina,Children’sHospitalBoston/HarvardMedicalSchool,UnitedStatesofAmerica Received January9,2012; Accepted March26,2012; Published May14,2012 Copyright: 2012Davoodi-Semiromietal.Thisisanopen-accessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense,which permitsunrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited. Funding: ThisstudywassupportedbygrantsfromtheJuvenileDiabetesResearchFoundationInternational(Innovativegrant # 5-2006-937),andNational InstitutesofHealthR21A176394-01(TheAmericanRecoveryAndReinvestmentAct)awardedtoADS.Publicationofthisarticlewasfundedinpartbythe Universityof FloridaOpen-AccessPublishingFund.Thefundershadnoroleinstudydesign,datacollectionandanalysis,decisiontopublish,orpreparationofth emanuscript. CompetingInterests: Theauthorshavedeclaredthatnocompetinginterestsexist. *E-mail:dsemiromi@cop.ufl.edu(ADS);atkinson@ufl.edu(MA)IntroductionTyrosinephosphorylationinhibitors,termedtyrphostins,area familyoflowmolecularweightmoleculeshavingtheabilityto inhibittyrosinephosphorylationthroughproteintyrosinekinases (PTKs).Thefamilyhasatleast31members,withAG490,also knownastyrphostinB42,beingone[1].AG490hasbeenwidely studiedincancerandautoimmunediseases,butnotsubjectto investigationinthesettingofT1D.Tyrosinekinaseinhibitors (TKIs)havebeenbroadlyusedinmanycancerclinicaltrials[2–4], T1D[5,6]inmousemodelofmultiplesclerosis[7]andin intestinalinflmmationisreported[8].Oraladministrationof Sunitinib(Sutent)andImatinib(Gleevec),twoTKIs,havebeen showntopreventandreverseT1DintheNODmouse[5].In severalpublicationssuccessfulremissionofT1Dhasbeenreported [9–16];however,whenitcomestomonotherapyapproachlonglastingremissionofdiabetesisstillchalllenging[17]. TheJak-Statsignalingpathwayplaysacriticalroleinmediating inflammatoryimmuneresponsetoavarietyofsignals[18–21]. Currently,severalJak-Statinhibitorshavebeendevelopedandare underclinicalinvestigation[4,22–25].AG490blockstheactivationoftheJakandSTATfamilyofmolecules[1].AG490hasbeen reportedtobemorepotentasaJak2inhibitor(4.3-fold)thanJak3 [26];however,ithasnoimpactonmembraneassociatedkinases suchasLyn,Btk,Syk,SrcandZap70[27,28]suggestingthatitis notageneralkinaseinhibitor[26,27,29–31].Additionally,AG490 blockstumorgrowthandinhibits/delaystheonsetofautoimmune diseasessuchasexperimentalautoimmuneencephalomyelitis (EAE)[32],inhibitsantigen-specificT-cellinfiltrationinEAE mice[32–34],andsignificantlyextendsratheartallograftsurvival [35].AG490alsoblocksIL-12mediatedTcellproliferation, inhibitsphosphorylationofStat3,decreasesIFNc production mediatedbyIL-12[32],andinhibitsdifferentiationofantigenPLoSONE|www.plosone.org1May2012|Volume7|Issue5|e36079

PAGE 2

specificTh1cells invivo [36]-allsuggestingthatthisdrugpossess immunomodulatingproperties. Previously,wereportedseveraldefectsintheJak-Stat5signaling pathwayinNODmiceincludingautoactivationofStat5,germlinemutationinStat5BandimpairedDNAbindingofStat5Bin NODmice[37–39].GiventheinfluenceofAG490ontheJak-Stat pathway,wesoughttoblocktheJak-Statsignalingpathwayand investigatetheeffectsofthiscompoundonearly(4weeks)andlate modelofT1Ddevelopment(8weeks),insulitisandinitiationof remissionofnewlydiagnosedandestablisheddiabeticNODmice.Results AG490preventsdevelopmentofautoimmunediabetes inNODPrediabeticfemaleNODmiceweretreatedwithAG490(i.p, 1mg/mouse)andcontrolmicereceivedthesamevolumeof vehicle(DMSO,solventofAG490,andsterilePBS)threetimes/ weekfor5consecutiveweekandonceaweekforfivemoreweek startingat4(preinsulitis)or8weekofage(establishedinsulitis). OurdatasuggeststhatAG490preventsdevelopmentofautoimmunediabetesinNODmicewhenappliedat4or8weekofage (p=0.02andp=0.005,respectivelyatweek30)(figure1A&B). Bloodglucoseandbodyweights(figure1CandD)weremonitored inbothgroupsasearlyas6weekofageandcontinuedforthe entireperiodofinvestigation.Therewasnostatisticallysignificant differencesintermsofbodyweightsbetweenAG490andDMSO treatedmicesuggestingadministrationofAG490atthegivendose issafeinmice.Additionally,wealsodidnotobserveanyadverse effectofAG490treatedmicebygrossexaminationsuggestingthat efficacyandsafetyofmultipleinjectionsofAG490.Thisdata suggeststhatAG490isanefficientandsafedruginpreventionof diabetesinNODmice.AG490inducesremissionofautoimmunediabetesin NODmiceWesoughttoinvestigatewhetherAG490wouldreverse pathogenesisofT1DinnewlydiagnoseddiabeticNODmice. NewlydignosedandestablisheddiabeticNODmiceweretreated withAG490(i.p,1mg/mouse;noinsulinwasgiventomiceatany time)andcontrolgroupwastreatedwithvehicleandbloodglucose wasmeasured.OurdataindicatethatAG490ledsignificantlyto remissionofdiabeticmice(p=0.003,Log-ranktest)whileDMSO treatedcontrolmiceledtoremissionof0%(n=10)(figure2A& B).Onceeuglycemiawasestablished(TableS1),treatmentwas ceasedandnormaglycemialastsforseveralweeksandsustainable euglycemiawasmaintainedafterdrugwithdrawalinmajorityof thecuredmice.ThesedatasuggestthatNODmicedidnot developresistancetoAG490treatmentandadditionallysuggests thatAG490hasadurableeffectonrestorationofbloodglucose metabolism/regulation.However,diabeticNODmicetreated withDMSOorleftuntreatedexpiredwithin2–4weeks(figure2B). AverageofbloodglucoseinthecuredNODmiceatthedisease onsetwas314.43mg/dl(Max=374;Min=277mg/dl)versus 384.58mg/dl(Max=478,Min=335mg/dl)(p 0.0014)inmice thatAG490failedtorestoreeuglycemiawhileanaverageblood glucoseincureNODmicewas169.82 6 23.21mg/dlversus 555.13 6 63.79(p=1.15E-11)inuncuredNODmice.Thisdata suggestthatAG490wasmoreeffectiveindiabeticmicewithblood glucoselevel 314.43mg/dl.Acombinationtherapy,e.g AG490 + insulin,mayimproveefficacyofAG490treatmentin micewithbloodglucoselevel 314.43mg/dl.AG490treatmentdidnotcauseclinicalsignsoftoxicity andbehaviorTreatment-relatedmortalityinthecourseofi.p(intraperitoneal) injectionofAG490andthereafterdidnotobserve.Clinicalsignsof toxicityrelatedtotreatmentsuchaslossofbodyweight,strange behaviorsuchasaggressiveness,hypermobility,hunchback,and othersignssuchasdiscolorationofthefurorfurloss,urineand stooldiscolorationdidnotobserved.Treatmentdidnotchange apetiteandfood/waterconsumptionasjudgedbybodyweight gain/lossasmeasured1–2timesperweekandcontinued throughtoutthestudyandnodifferencesbetweenAG490and DMSOtreatedmicewereobserved(seefigure1CandD)owning tothegoodgeneralconditionofanimls.Laboratoryexamination suchashematology,clinicalchemistryandurineanalysisdidnot perform.AG490inducesremissionofdiabetesinSTZinduced diabeticmiceTofutherexplorewhetherAG490canreversediabetesinSTZinduceddiabeticmice,BLAB/cmiceweretreatedwithSTZand thenweretreatedoneweekaterthelastSTZinjectioneitherwith AG490orDMSO(n=5/group).Ourdataindicatethatthemean bloodglucoseinDMSOtreatedmicewassignificantlyhigherthan AG490treatedmiceafter2-weeksoftreatment(DMSO: 498.6 6 29.3mg/dl;AG490:240.4 6 95.2mg/dl;p 0.004) (figure2C&D).ThisdatafurthersuggeststhatAG490isan effectiveandsafedrugtoreverseonsetofdiabetesandfurtheris suggestingthattheJak-Statsignalingpathwaymayhaveadditional role(s)inpathogenesisofdiabetesandonglucosemetabolismthan currentlyappreciated.AdoptivetransferofAG490treatedsplenocytesdidnot transferdiabetesinNOD. Scid miceInordertounderstandhowAG490reversedandblockedthe onsetofautoimmunediabetes invivo ,wetreatedprediabeticNOD mice(4-weekofage)withAG490orwithDMSOfor5week.Mice weresacrificiedatweek10andsplenocytecellsuspensionswere preparedwithHistopaquegradientseparation.Thedeadcellsand reticulocyteswereremovedbytheHistopaquegradientseparation,therefore,deadcellswereexcludedwithoutmagnetic purificationandwithoutactivationofT-cellsandcellviability wasgreaterthan99%.Fivemillionfreshelypreparedofpooled splenocytesofNODmicetreatedwithAG490(n=5)orDMSO (n=5)weretransferredbyi.pinjectionintorecipientNOD. Scid miceandbloodglucoseandbodyweightwascloselymonitored. Ourdataasshowninfigure3suggeststhatpreventionof autoimmunediabetesbyAG490couldbeduetoinactivationof atuoreactiveT-cellsand/orinduction/generationofregulatoryTcellorotherregulatorypopulation.AdministrationofAG490doesnotdiminishleukocyte infiltrationwithinthepancreasInanotherseparateexperiment,NODmiceat4weekofage weretreatedwithAG490orwithDMSO(n=5/gorup),three timesperweekfor5consecutiveweek.Miceweresacrificedat week10andmicroscopicsectionswerepreparedandstainedwith H&Eandscoringofinsulitiswasperformedasdescribedin materialsandmethods(figure4A).Therewasnostatistically significancedifferencesregardinginsulitisbetweenthetwogroups. TheC-peptideconcentrationwassignificantlydifferentinAG490 treatedversusDMSOtreatedmice(4weekgroup);however,no significantdifferencesinserumC-peptideconcentrationofAG490 orDMSOtreatedNODmice(8weekgroup)wasfound(figure4B).AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org2May2012|Volume7|Issue5|e36079

PAGE 3

AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org3May2012|Volume7|Issue5|e36079

PAGE 4

TheC-peptideconcentrationofAG490treatedNOD(8week group)wassignificatlyhigherthantheC-peptidelevelofmice whentreatmentwasinitiatedat4weeksofage.Inaseparate experiment,femaleNODmiceweretreatedeitherwithAG490or DMSOfor5weeksandthenweresacrificiedatweek10.Total RNAextractedfromenrichedinfiltratingleukocytesfromFigure1.AG490preventsdevelopmentofautoimmunediabetes. PrediabeticfemaleNODmiceat4(UT:n=11;DMSO:n=13;andAG490 treatedmice:n=15mice/group)(A),orat8weekofage,(n=10pergroup)(B)weretreatedwithAG490orDMSO(i.p)for5consecutiveweek,three timeperweekfollowedbyonceaweekfor5moreweek[Kaplan-Meiertest(*p=0.02,1A)and(**p=0.005,1B,atweek30)].Ourdataalsoindicate thatsignificantdifferences(p 0.01)wasobservedbetweentheAG490andDMSOtreatedmiceforupweek52;whentreatmentwasstartedatweek 4.BodyweightofNODmicetreatedwithAG490(C)orDMSO(D)inmicewhentreatmentinitiatedfrom4weeksofageisshown.[Thisfigureis intendedtobeincoloronlineandblackandwhiteinprint.] doi:10.1371/journal.pone.0036079.g001 Figure2.AG490inducesremissionofdiabetesinNODandinSTZ-induceddiabeticmice. NewlydiagnosedandestablisheddiabeticNOD miceweretreatedwithAG490(i.p,1mg/mouse)withoutinsulinsupplement(A)(n=23)orwithDMSO(B)(n=10).Bloodglucosewasmeasuredby ACCU-CHECK2–3timesweekly.SustainableeuglycemiawasobservedforseveralmonthsafterdrugwithdrawalwhileDMSOtreatedmiceexpired within2–4weekafterdiseaseonset.TheaveragebloodglucoseatthediseaseonsetinthecuredNODmicewas314.43mg/d(Max=374; Min=277mg/dl)versus384.58mg/dl(Max=478,Min=335mg/dl)inmicethatAG490failedtorestoreeuglycemia(p 0.0014).Anaverageblood glucoseincuredNODmicetreatedwithAG490was169.82mg/dl 6 23.21,n=7versus555.13 6 63.79,n=17inuncuredNODmice(p=1.15E-11).(C) DiabeteswasinducedbySTZinjectioninmaleBALB/cmiceandoneweekafterthelastinjectiondiabeticmiceweretreatedwitheitherAG490or DMSO(D)(n=5/group).Thebloodglucoselevelwassignificantlylower(p 0.004)inAG490treatedmicewhencomparedwithDMSOtreatedmice. [Thisfigureisintendedtobeincoloronlineandblackandwhiteinprint.] doi:10.1371/journal.pone.0036079.g002 AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org4May2012|Volume7|Issue5|e36079

PAGE 5

pancreaseofAG490andDMSOtreatedmice(n=5mice/group) wassubjectedtoTaqManReal-timeRT-PCR.LevelofFoxp3 andTGFb 1expressionwereverylowbutdetectableinAG940 treatedNODmicebutwefailedtodetectexpressionofFoxp3in threeindependentRT-PCRinDMSOtreatedNODmice (figure4C).Thisdatasuggeststhatalthoughadministrationof AG490wasunabletoblockinfiltrationoftheleukocytesinthe isletsorimprovetheC-peptideconcentration;however,itchanged compositionoftheinfiltratedleukocytesintheislets.AdministrationofAG490doesnotblockexpressionof Jak-1andStat-1inleukocytesinfiltratingtheisletsExpressionoftheJakandStatfamilymembersintheisletsof NODmicetreatedwithAG490orDMSOwerescreenedby immunofluorescenhistochemistrytechnique.ExpressionofJak3, Stat3andStat5wasverylow/undetectableinAG490treated mice.ExpressionofJak2wasundetectableoritwasveryloweven inDMSOtreatedNODmice.AG490hadnoeffecton constitutiveexpressionoftheJak1andStat1(figureS1).Higher expressionofJak1andStat1isofimportancebecauseitisknown thatIFN c usesJAK-1foractivationofStat1[19,21,40]. Additionally,sinceadminsitrationofAG490mightinhibit expressionofotherkinasesthanmembersoftheJAKfamily,we screenedexpressionofEGFR,c-Kit,PDGFRandc-Ableby westernblotinsplenocytesofNODmicetreatedwithAG490or DMSO(n=5).Inseveralindependentwesternblots,alldetected bandsdidnotmatchwithexpectedmolecularweightsofthose proteinsuggestingthatexpressionofthesekinaseswasverylow andAG490likelydoesnothaveanyimpactonexpressionofthese kinases.AdministrationofAG490significantlyincreases expressionofFoxp3TofurtherunderstandastowhetherAG490preventedand reverseddiabetesintreatedNODmice,wemeasuredexpression ofFoxp3andotherregulatorymarkersbyflowcytometryanalysis indifferentorgansoftheimmunesysteminNODmicethatwere treatedwitheitherAG490orDMSO.Asshowninfigure5A–D significanthigherexpressionofCD25 + Foxp3 + T-cells(p 0.004) isevidentinallorganstestedinAG490treatedmice.Furthermore,AG490treatmentalsoincreasedexpressionofFoxp3in CD25 + T-cellsinperipheralbloodofthesamegroupsofNOD (Fig.5E).AG490isnotanegativeregulatoroftheimmunesystem; screeningofthemouseimmuneresponsegenesarrayInaseparateexperiment,pooledtotalRNAextractedfrom purifiedCD4 + T-cellsfromspleenoftheAG490orDMSO treatedNODmice(treated3x/weekfor5weeksandeuthanizedat week10,n=5/group)wereusedtoscreenthemouseimmune responsegenesarray(AppliedBiosystems)byreal-timeTaqMan GeneExpressionRT-PCR.Asshowninfigure5F,AG490didnot negativelyregulateoverallgeneexpressionofthemouseimmune responsegenes(seeTableS2).Interestingly, invivo treatmentof NODmicewithAG490increased(2.15fold)expressionofcostimulatory Ctla4 inCD4 + T-cell.The Ctla4 geneisadiabetes candidategeneforT1Danditsassociationiswelldocumentedin theliterature[41–44].Globalgeneexpressionscreeningfrom differentcellularcomponentoftheimmunesystemofNODmice treatedwithAG490iswarrantedtobetterunderstandmechanism underlyingmodulatoryfunctionofAG490andthatmayleadto identifyitstargetgene(s)/pathway(s).Invivo and invitro administrationofAG490modulates phenotypeandfunctionofdendriticcellsBonemarrowofAG490andDMSOtreatedNODmice(4 weeks/3x/week,treatedfor5weeks)wereharvestedatweek10 andyeildedthesamenumberofcellsineachgroup( p =not significant).Wedidnot,however,noticeanyremarkable differenceinmorphology,suchasthesize,shapeandclonoy formation,betweenthetwogroupsinthecourseofdifferentiation incellculture.Asshowninfigure6AandB,BMDCs(bone marrowderiveddendriticcells)ofmicetreatedbyAG490had markdlylowerexpressionofco-stimualtorymoelculesCD80and CD86andadhesionmoleculeCD62L.Additionally,ourreal-time TaqManGeneexpressionAsssays(Assays-on-Demand)(Life technologies,AppliedBiosystems)indicatesthatAG490significantlyincreasedexpressionofregulatorycytokineTgfb 1in matureandimmatureDC(dendriticcell)whenitwascompared withtheshamtreatedcontrolNODmice(figure6C,n=5/group, p=0.02).Furthermore,BMDCstreatedwithAG490(overnight, 20mM)orDMSOsuppressedproliferationofpurifiedresponding CD4 + CD25 2 T-cells invitro (figure6D).DiscussionPresenceofproinflammatorycytokinesduetoongoinginflammationinNODmice,leadstoautoactivationofdifferentmembers oftheJakandStatfamilyfollowedbyactivationofdownstream eventsasithasbeenreportedbyus[37,45]andothers[46]. Recently,ithasbeenshownthatatranscriptomicanalysisof glomerularandtubulointerstitialtissuesfromdiabeticsubjectswith earlyandprogressivediabeticnephropathyrevealedinvolvement ofmultiplemembersoftheJak/Statsignalingpathway[47]. Interestingly,higherexpressionofthe Stat5B inprogressive diabeticnephropathypatientshasbeenshown[47]. Tyrosinekinaseinhibitorshavebeenreportedtodecrease insulinresistance[48]andincreaseinsulinsecretion invitro [49,50]. Inourstudy,AG490decreasedhyperglycemiainchemically induceddiabeticmiceandinnaturallydevelopeddiabeticNOD mice.Inchemicallyinducedstudy,sincethetreatmentwith AG490wasstartedoneweekafterdiabetesonset,beta-cell Figure3.InvivoAG490treatedsplenocytesdidnottransfer diabetestoNOD.Scidmice. PrediabeticNODmice(4-weeksofage) weretreatedwithAG490orDMSOforfiveconsecutiveweek(n=5per group),threetimesperweekandmiceweresacrificedoneweekafter thelastinjectionatweek10.Splenocytecellsuspensionofpooledmice immediatelypreparedanddeadcellsandredbloodcellswereremoved byHistopacquegradientseparationmethodandfreshlyisolated splenocytescontaining 99%viablecellsweretransferred(5 6 106, i.p)intoNOD. Scid mice.Bloodglucoseinrecipientmicewasmonitored atleastonceaweekasdescribedearlier(*:p=0.02,Kaplan-Meiertest). doi:10.1371/journal.pone.0036079.g003 AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org5May2012|Volume7|Issue5|e36079

PAGE 6

destructionwasmoreorlesscompletedatthisstage;therefore, improvementinbloodglucoselevelistheresultofincreased insulinsensitivitybyAG490.Rapidrecoveryfromdiabetesin responsetoAG490therapy,inefficiencyofAG490ontheCpeptidelevels,thelackofanyAG490effectonleukocyteislet infiltration-areallinagreementwithimprovedinsulinsensitivity inperipheraltargettissues.Ontheotherhand,Real-timeRTPCRdataconsistentlyindicatesthatexpressionofFoxp3in infiltratinglymphocytesinthepancreasofAG490treatedNOD micewasdetectablewhileitwasundetectableinDMSOtreated mice(seefigure4C).ThisdatasuggeststhatAG490treatment changedcompositionofinfiltratingleukocyteoftheisletinNOD mice. Ourpreviousreportsregardingautoactivationandimpaired functionoftheJak-Statpathway[37,45]andourdatainthis study-allmaysuggestthattheJak-Statsignalingpathwaymaybe Figure4.AG490doesnotblockinfiltrationofleukocytesintheislets. A-PrediabeticNODmiceat4weeksofageweretreatedwithAG490 orDMSO(n=5pergroup)for5consecutiveweeks(3 6 perweek)andmiceweresacrificedoneweekafterthelastinjectionatweek10.Insulitiswas scoredbytwodifferentinvestigatorsandanaverageoftwodifferentscoresisshown(p=n.s).Atleast150isletspergroupintwointerrupted sections,100mmapart,wasscored.B-TheC-peptidelevelwasmeasuredasdescribedinmaterialsandmethodsandnosignificantdifferenceswere found.C-FiveprediabeticNODmiceat4-weeksofageweretreatedwithAG490orDMSO(n=5pergroup)asdescribedabovefor5consecutive weeksandthensacrificedatweek10,C-real-timeRT-PCRwasperformedonenrichedinfiltratedlymphocytefromthepancreasof5mice/groupand expressionlevelofFoxp3andTGFb 1wasmeasuredinthreeindependentRT-PCRbyTaqManGeneexpression(LifeTechnologies,Applied Biosystems,Assays-on-Demand)(*=expressionwasverylowand/orundetectable).[Figure4Aisintendedtobeincoloronlineandblackandwhite inprint.] doi:10.1371/journal.pone.0036079.g004 AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org6May2012|Volume7|Issue5|e36079

PAGE 7

AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org7May2012|Volume7|Issue5|e36079

PAGE 8

anexcellenttargetforimmunotherapyofT1Dandmayalsobe effectiveintypeIIdiabetes,althoughthisremainstobeseen[51]. Autoactivationofkinases/transcriptionfactorsduetoongoing autoimmunitywillgenerateinflammatorycytokineswhichsubsequentlywillleadtofurtherinflammation.TheJak-Statsignaling pathwayisthemostimportantsignalingpathwayfortransduction ofsignalsfromthecellsurfacetothenucleus[21,52–54].This signalingpathwayisinvolvedinthegenerationofpro-inflammatorycytokinesandisboundtohavemanymoreimportantrolesin thepathogenesisofT1Dthancurrentlyisappreciated.Ourstudy uncoveredseveralnovelfeaturesofAG490suchaspreventionof diabetes,initiationofremissioninnewlydiagnosedNODmice andinpharmacologicalinduceddiabeticmice,increasingnumber ofFoxp3regulatoryT-cellsandprobablygenerationofregulatory DC.Itisimportanttomentionthatsignificantdifferences (p 0.01)inincidenceofdiabeteswasobservedbetweenthe AG490andDMSOtreatedmiceforupto52weekwhen treatmentwasstartedatweek4.Thisdatamaysuggestthat appropriateblockingoftheJak-Statsignalingpathwaycouldbring aboutadurableandlong-lastingtoleranceandfurtheris supportingthenotionofsafetyandefficacyofAG490in preventionofautoimmunediabetesinNODmouse. OurdatasuggestthatadministrationofAG490blocks expressionofseveralmembersoftheJakandStatfamiliesin infiltratedleukocytesintheisletsofNODmice.LackoftheStat3 expressionisfatalperinataly[55],however,noneofthetreated miceinthisstudydiedbecauseofAG490treatmentsuggesting thattreatmentbyAG490wasnotlifethreatening.Ourmicroarray real-timeTaqMangeneexpressiondatasuggeststhatAG490may exertitsregulatoryeffectviaupregulationofnegativecostimulatorymolecule, Ctla4 ,inCD4 + T-cells;however,additional screeningofdifferentcellularcomponentoftheimmunesystemis warrantedtoidentifyittargetgene(s)/pathway(s)andbetter understanditsmolecularmechanism.Ourdataindicatethatbody massinAG490andtheshamtreatedcontrolmicewasnot significantlydifferentfromeachothersuggestingthatAG490 treatmentdidnotcauseobesity[56]. Inthisstudy,wedemonstratethatAG490initiatesremissionof diabetesinnewlydiagnosedandestablisheddiabeticmice,no insulinwasinjectedatanytime,andsustainedeuglycemiawas evidentforseveralmonthsfollowingdrugwithdrawal.Ithasbeen shownthatAG490increasedglucoseuptake(102 6 9)inL6 myotubes[48]underhighglucoseconcentrationincomparison withvehicle-treatedcells[57].AG490decreasedfibronectin expression;higherexpressionoffibronectinisassociatedwith nephropathyindiabeticsubjects,underhighglucoseconditionin glomerularmesangialcell;therefore,AG490potentiallycouldbe aneffectivetreatmentforpreventionofdiabeticnephropathy[58]. Adoptivetransferofsplenocytesisolatedfrominvivotreatedmice withAG490failedtotransferdiabetestoimmunodeficientNOD micesupportingthenotionofinactivationofautoreactiveT-cellor inductionofregulatoryT-cellbyAG490.DCplaysacentralrole ininductionandmaintenanceofthecentralandperipheral tolerance[59,60]anditsdevelopmentisimpairedinNODmice [61].Asshowninthisstudy,DCtreatedwithAG490possesses regulatoryfunctionandsuppressproliferationofrespondingTcells invitro .Recently,Louvetetalhavereportedsuccessful reversalofdiabetesbyoraladministrationoftyrosinekinase inhibitors[5],however,diabeteswastransferredfromdrugand theshamtreatedcontrolmiceintoimmunedeficienthostswiththe samekinetics[5]suggestingthatinductionoftolerancewas achievedbyothermeansthangenerationoftheregulatoryT-cells. Insummary,thisstudyprovidesevidenceregardingapplication ofAG490inprevention/reversalofautoimmuneT1DinNOD mouse.AdministrationofAG490 invivo wasverywelltolerated andtreatedNODmicewithAG490showednosignsoftoxicity andtheyappearedhealthierastheyhadalongerlifespanand theyweresignificantlydiabetes-freewhencomparedtothesham treatedoruntouchedcontrolNODmice.Additionally,thisstudy providesevidenceregardingthecrucialroleoftheJak-Stat signalingpathwayindevelopmentofautoimmunediabetesand regulationofbloodglucose.Therefore,theJak-Statsignaling pathwaymaybeanexcellentpathwaytobetargetedfor immunotherapyand/ortoreversediabetes.Theuseofsuch compoundslikeAG490,giventheirextensivesafetyprofiles,offers asolidfoundationfortheirtranslationtodiabeticsubjectswithor atincreasedriskforthedisease.MaterialsandMethods EthicsstatementAllmiceweremaintainedandhandledinaccordancewiththe UniversityofFloridaInstitutionalAnimalCareandUseCommittee(IACUC),whospecificallyapprovedthisstudy.MiceFemaleNOD/LtJ,NOD. Scid ,andBALB/cmicewere purchasedfromJacksonLaboratories(BarHarbor,ME)and housedinventilatedcagesinaspecificpathogenfreefacility.CompoundAG490(EMDChemicals,Gibbstown,NJ),5mg,initiallywas dissolvedinsterileDMSO(EMDChemicals)andbroughttofinal volumebysterilePBSfollowedbyseveralpipettingtobringthe compoundintosolution.Onevialofcompoundcontaining5mg ofAG490wasinjectedinto5mice(1mg/mouse)viathei.proute. Thecontrolgroupswerereceivedthesamevolumeofthevehicle underthesameregimensandconditions.AdministrationofAG490Micewereinjectedeitherat4or8weeksofage,threetimesper weekfor5consecutiveweeksfollowedby5moreinjections,oncea weekfor5additionalweeks.Bodymassandbloodglucoselevels weremeasuredatthebeginningofthestudy(baseline)andFigure5.AG490upregulatesexpressionofFoxp3regulatoryCD25 + T-cellsinvivo. A-PrediabeticNODmice(4week)weretreatedeither withAG490orDMSOthreetimesperweekfor5consecutiveweeksandtheyweresacrificedoneweekafterthelastinjectionatweek10.Data representspoolcellsuspensionsof5NODmicepergroupandisgatedonliveCD4 + T-cells.AG490treatmentsignificantlyincreasesnumberofFoxp3 regulatoryT-cells(p 0.004)(A–D).E-CD25 + Foxp3 + gatedpopulationonCD4 + T-cellofseverallymphnodesinAG490andtheshamtreatedcontrol NODmice.Barsrepresentstandarddeviationofthemean.F-WholebloodofNODmouse(poolof5mice/group)treatedasdescribedabovewas staineddirectlywithantibodiesindicatedasdescribedinmaterialsandmethods.DataofgatedCD4 + T-cellsareshown.G-TotalRNAfrompurified CD4 + T-cellsofsplenocytesofNODmice(10weeksold)treatedeitherwithAG490orDMSOfrom5weeksweresubjectedtoreal-timeRT-PCRand expressionoftheimmuneresponsegeneswasscreenedasdescribedinmaterialsandmethods.Datashownisthemeanoffoldchangeexpressionof 18genesaftertheywerenormalizedbasedonexpressionoftwohousekeepinggenesandthencomparedwiththeexpressionofthesamegenein DMSOtreatedcontrolNODmice.Spl=spleen,PLN=pancreaticlymphnode,MLN=mesentericlymphnode,PP=Peyer’spatches,PB=peripheral blood. doi:10.1371/journal.pone.0036079.g005 AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org8May2012|Volume7|Issue5|e36079

PAGE 9

monitoredatleasttwiceaweekstartingatweek6andcontinued thereafter.Bloodglucoseconcentrationsweredeterminedusing ACCU-CHEKComfortCurve(Roche,Indianapolis),anddiabetesdiagnosedifvaluesweregreaterthan250mg/dlintwo consecutivetestswith24hoursapart. Figure6.AG490modulatesphenotypeandfunctionofDC. A-DendriticcellsweregeneratedfrombonemarrowofNODmicetreatedeither withAG490orDMSOasdescribedinmaterialsandmethodsandstaineddirectlywithCD11c,MHCII,CD80andCD86.DatashownrepresentsCD11c gatedpopulationofpoolfreshpurifiedimmatureCD11cofAG490andDMSOtreatedmice.B-HistogramsrepresentexpressionofCD11c,CD86, CD80andCD62LongatedCD11cpopulationofmicetreatedwithAG490orDMSO invivo .C-BonemarrowcellswereisolatedfromNODmice treatedwithAG490/DMSOfor5weeksandthendifferentiatedintoDC invitro asdescribed.TotalRNA(200ng)ofDCwasconvertedintocDNAand thenexpressionofTgfb 1wasmeasuredbyReal-TimeTaqManGeneexpressionassays(AppliedBiosystems).Datarepresentsthemean(foldchange) oftworeal-timeRT-PCRexperimentsandthebarsrepresentstandarddeviationofthemean.FoldchangeofTgfb 1expressionisshown(n=5mice/ group,*p=0.02).D-BMDCsweregeneratedfromBALB/cmicetreated invitro with20mMAG490orDMSOfor12hrsandthenco-culturedwith constantnumbersofCD4 + CD25 2 T-cells(1 6 105)withsolubleanti-CD3(2.5ng/ml)for72hrs.Incorporationof3H-thymidinewasmeasuredinthe last16hrofcellculture.Experimentwasperformedintriplicateformatandrepeatedatleasttwice.Dataanalysiswasperformedusingthestudent t testandpvalue # 0.05consideredsignificant.Barsrepresentdeviationofthemean(*p=0.002,**p=0.0002;***p=0.001).[FiguresA&Bare intendedtobeincoloronlineandblackandwhiteinprint]. doi:10.1371/journal.pone.0036079.g006 AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org9May2012|Volume7|Issue5|e36079

PAGE 10

TolerabilityandclinicalsignsoftoxicityClinicalsignsoftoxicityrelatedtotreatmentsuchasweight gain/loss,abnormalbehavior,discolorationofurine,stoolandfur monitoredatleast3timesperweekandcontinuedthroughoutthe study.Bodyweightwasrecordedbeforeinitiationofthestudyand measured1–2timesweeklythereafter.Inductionofdiabetesbyi.pinjectionofstreptozotocin (STZ)STZ(Sigma-Aldrich,St.Louis,MO)wasfreshlypreparedin citratebuffer(pH=4.5)justbeforeinjectionandfinalvolumewas adjustedbyaddingsterilePBS.BALB/cmice(n=10)were received5doseofSTZ(50mg/kg)byi.pinjectionin5-consecitive days.Diabeticmiceweretreatedoneweekpostdiabetesonsetwith eitherAG490orDMSO(control).PurificationofCD4 + ,CD4 + CD25 + / 2 T-cellsanddendritic cellsPurificationofCD4 + CD25 + T-cellswasperformedfrom pooledsplenocytesusingaCD4 + CD25 + isolationkit(Miltenyi Biotec,Auburn,AL)andautoMACSmagneticseparator,accordingtoinstructionsprovidedbythemanufacturer.Purityofpurified CD4 + CD25 + andCD4 + CD25 2 T-cellsweregreaterthan85– 90%and95%,respectively,asjudgedbyflowcytometryanalysis. SplenicCD4 + T-cellsanddendriticcells(fromspleenandbone marrowderivedDCs)wereenrichedwithCD4 + andCD11c MicroBeads,respectively(MiltenyiBiotech)usingautoMACS instrumentaccordingtomanufacturer’sinstructions.Generationofbonemarrow-deriveddendriticcellsDendriticcells(DC)weregeneratedfrombonemarrowin 5days.Inbrief,micewereeuthanizedandfemurandtibiae removedandcleanedfromothertissues.Bothendsofeachbone werecutandmarrowflushedwithRPMI1640containing10% serum.Redbloodcellswerelysedwithanammoniumchloride lysissolution(Sigma-Aldrich),andcellswereplatedat1 6 106cells/ ml,2ml/wellin12-wellplate(Falcon;BDBiosciences)inRPMI 1640medium,supplementedwith1%penicillinandstreptomycin, 10%heat-inactivatedandfilteredFCScontainingrmGM-CSF (500U/mlR&Dsystems)andIL-4(1000U/ml,BDBiosciences). Onday2,non-adherentandlooselyadherentcellswereharvested byreplacingone-halfofthecell-containingsupernatantfromeach wellandreplacedwith1mloffreshmediumcontainingcytokines. Onday5,DCmaturationproceededbyaddingofLPSandthose cellsthatdidnotreceiveLPSwereconsideredasimmatureDC. Dailymicroscopicinspectionofcellculturebymicroscopic examinationwasperformeduntiltheendoftheDCculture.T-cellculture,Proliferation/SuppressionassaysTheAG490/DMSOtreatedbonemarrowderivedDC (BMDC)wereculturedatdifferentratios,withconstantnumber ofCD4 + CD25 2 T-cells(1 6 105),in96-wellroundbottomplates containingsolubleanti-CD3(2.5mg/ml)for72hourrs.Proliferationofrespondingpopulationsweremeasuredinthelast 16hoursofcellculturebymeasuring3H-thymidineincorporation intotheDNAusingScintillationCounter(Beckman).TaqManmouseImmuneResponseArrayandReal-Time TaqManGeneexpressionanalysesTotalRNAwasextractedfromCD4anddendriticcellsusing RNAeasykit(QIAGEN).CD4 + T-cellswaspurifiedfrom5 individualmiceusingmouseCD4 + T-cellisolationkit(Miltenyi Biotech)followedbyautoMACSseparationaccordingtothe manufacturer’sinstructions.ToscreentheTaqmanArrayMouse ImmuneResponse,200ngoftotalRNAwasconvertedtocDNA usingHighCapacitycDNAReverseTranscriptionkit(Applied Biosystems).ThecDNAfrom5micewerepooledtogether(one microgramoftotalRNA)andthenusedtoscreen96-immune responserelatedgenesincluding4housekeepinggenes.Inthecase ofReal-TimeTaqManGeneexpression(Assay-On-Demand), 100ngoftotalRNAwasusedforreversetranscriptionfollowedby PCRamplificationusing7300AppliedBiosystemsPCRmachine. AllprimersandprobeswerepurchasedfromAppliedBiosystems (Assays-on-Demand).AdoptivetransferFemaleNODmice(4weeksold)weretreatedwithAG490or DMSO(n=5pergroup)for5consecutiveweek,withthree injectionsperweek.Miceweresacrificedoneweekafterthelast injection(i.e.,week10).Cellsuspensionsofsplenocytesfrompool micewereprepared,withdeadcellsandredbloodcellsremovedby Histopaque(Sigma)gradientseparation.Cellviabilitywas 99%as judgedbytrypanblueexclusion.5 6 106splenocyteswere transferredbyi.pinjectionintofemaleNOD. Scid recipientmice.FlowcytometryanalysisCellsurfacestainingwasperformedusinganti-mouseCD4 (L3T4;BDPharmingen,SanDiego,CA),CD25(3C7;BD Pharmingen),CD11c(HL3;BDPharmingen),CD80(16-10A1; eBioscience,SanDiego,CA),biotin-conjugatedMHCII(M5/ 114.15.2;eBioscience)andanti-hTGF b -Biotin(BAF240;R&D Systems)andbiotinlabeledantichickenIgY(BAF010;R&D System)forisotypecontrol.Purifiedratanti-mouseCD16/CD32 (2.4G2;BDPharmingen)wasusedtoblockFcreceptorinmyeloid celllineages.Intra-cellularstainingofFoxp3(FJK,16S),IL-10 (JES5-16E3),IFN c (XMG1.2)wasperformedusingFoxp3 intracellularstainingkit(eBioscience),accordingtoinstructions providedbythemanufacturer.Viablecellsweregatedanda minimumof30,000eventswereacquiredusingaFACScaliber (BDBiosciences).DatawasanalyzedusingFCSExpress(v3) software(DeNovoSoftware,LosAngelesCA).Histologyandimmunofluorescencehistochemistry (IFHC)Pancreatawerefixedin4%PFA(paraformaldehyde)and microscopicsectionspreparedwith100mmintervalfromparaffin embeddedtissue,andstainedwithH&E,atthePathologyCore LaboratoriesatUniversityofFlorida.Insulitisscoringwas performedinablindedfashion.Anumericcodewasgivento eachslideandthenscoredbytwodifferentinvestigators.Athree pointscoringmethodwasusedwhere0wasgiventopre-insulitis orintactislets,1wasgiventoisletswith 50%infiltrationand2 whereinfiltrationwas 50%.IFHCwasperformedon4mm paraffinsectionsofpancreasfromtheNODmice.Theconcentrationofprimaryantibodieswasbasedonrecommendationofthe manufacturers.Stat1p91(SC-591),Stat3(SC-7179),Stat5B(SC835),IFN c (SC-8308),insulin(SC-7838),Jak1(SC-277),JAk2(SC1656),Jak3(SC-6932),Tyk2(SC-136265),werefromSantaCruz (SantaCruz,CA)andwerediluted1:50andTNFa-PEconjugated Ab(Pharmingen)wasdiluted1:100.Secondaryfluorescence conjugatedantibodieswithAlexaFluor488(green)(1:200)or AlexaFluor594(red)(1:500)(Invitrogen,Carlsbad,CA)against thehostprimaryantibodywereusedforsignaldetection.Anti-fade solution(Invitrogen)containingDAPIwasusedforcounter stainingandmounting.AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org10May2012|Volume7|Issue5|e36079

PAGE 11

WesternBlotTotalproteinwasextractedfromspleenofNOD micetreatedeitherwithDMSOorAG490usingRIPAlysisand extractionbuffercontainingHaltProteaseandphosphatase inhibitorcocktail(ThermoScientific).Thefollowingprimary antibodieswereusedinthisstudy:anti-mouseCD140b (PDGFRb),anti-mouseCD117(c-kit)fromeBioscience,and anti-mouseAbeandEGFR(BDBioscience),anti-phosphoPPAR c antibody(Millipore,Billerica),antiPPAR c antibody(R&D Systems),andHRPconjugatedantibodies(Sigma)withECL detectionkit(Millipore)wereusedforsignaldetection.StatisticalanalysisAllanalysesforstatisticalsignificancewereperformedusing Prism5.0software(GraphPadSoftwareInc,LaJolla,CA).The t testsusedforcomparisonofthemeanfortwogroupsandstatistical comparisonamongdifferentgroupswereperformedusingonewayanalysisofvariance(ANOVA)withTukey’sand/orDunnett’s postcorrectiontestformultiplecomparisons,and/orfor comparisonwithacontrolgroup,respectively.TheReal-Time data(RQvalues,foldchange)wascalculatedusingintegratedSDS software,V1.4.0(AppliedBiosystems)andasdescribedelsewhere [62].The D Ctforeachtreatedsampleswascomparedwiththe mean D Ctofcontrolsamples(unstimulated)usingtherelative quantitation22 ( DD Ct)methodasdescribedelsewhere[62]. Invivo datawasanalyzedusingKaplan-Meiertest.A P valuelessthan 0.05wasconsideredsignificant.SupportingInformationFigureS1TheeffectsofAG490onmembersoftheJakStatinthepancreas. PrediabeticNODmice(4week)were treatedeitherwithAG490orDMSOthreetimesperweekfor5 consecutiveweeksandweresacrificedoneweekafterthelast injectionatweek10.Pancreatawerefixedandimmunofluorescencestainingwasperformedonmicroscopicslidesasdescribedin materialsandmethods.Atleast30isletspermarkerpergroup wereanalyzed. (DOCX)TableS1AG490reversesdiabetesinNODmice. ReversalofdiabeteswasperformedbyadministrationofAG490 viai.proute.RecentlydiagnoseddiabeticNODmice(n=23)were underAG490treatmentforaperiodof21weeks.Theageof diseaseonset,bloodglucoselevelatdiseaseonsetandthemeanof bloodglucoselevelaftertreatmentandalsototalnumberof AG490injectionsareshown.BoldcasesrepresentNODmousein whichAG490waseffectiveandsuccessfullyestablishedeuglycemia(n=7).*=Representsthemeanofthebloodglucoseleveland standarddeviationofthemeaninNODmicefromtwoweekspost diabetesonsettilltheendofthetherapyisshown(week21). (DOCX)TableS2ThemouseimmuneresponsegenestoAG490 treatment. FemaleNODmiceweretreatedwithAG490or DMSO3x/weekfor5consecutiveweeksandthentheywere sacrificedoneweekafterthelastinjectionatweek10.TotalRNAof purifiedCD4 + T-cellsofsplenocytesoftreatedmiceandcontrols werepooledtogetherandthensubjectedtoreal-timeRTPCRusing themouseimmuneresponsegenesarray(AppliedBiosystems) followingmanufacturer’sinstructions.Dataanalyzingwasperformed usingintegratedsoftwarebasedonexpressionoftwohousekeeping genesandthencalibratedbasedonthecontrolmice(DMSOtreated mice).ShownrepresentsfoldchangegeneexpressioninAG490 treatedNODmicewhencomparedwiththeshamtreatedmice. (DOCX)AuthorContributionsConceivedanddesignedtheexperiments:ADSCHWMCSMA. Performedtheexperiments:ADSCHWCQX.Analyzedthedata:ADS CHWMCSMA.Contributedreagents/materials/analysistools:RMC MWMCSMA.Wrotethepaper:ADSCHWMA.Editedtherevised manuscript:RMC.Providedfundingforthestudy:ADS.References1.GoncalvesS,Fernandez-SanchezR,Sanchez-NinoMD,TejedorA,NeriaF, etal.(2010)Tyrphostinsaspotentialtherapeuticagentsforacutekidneyinjury. CurrMedChem17:974–986. 2.BaselgaJ(2006)Targetingtyrosinekinasesincancer:thesecondwave.Science 312:1175–1178. 3.HagerkvistR,SandlerS,MokhtariD,WelshN(2007)Ameliorationofdiabetes byimatinibmesylate(Gleevec):roleofbeta-cellNF-kappaBactivationandantiapoptoticpreconditioning.FASEBJ21:618–628. 4.SantosFP,VerstovsekS(2011)JAK2inhibitors:what’sthetruetherapeutic potential?BloodRev25:53–63. 5.LouvetC,SzotGL,LangJ,LeeMR,MartinierN,etal.(2008)Tyrosinekinase inhibitorsreversetype1diabetesinnonobesediabeticmice.ProcNatlAcad SciUSA105:18895–18900. 6.MokhtariD,WelshN(2010)Potentialutilityofsmalltyrosinekinaseinhibitors inthetreatmentofdiabetes.ClinSci(Lond)118:241–247. 7.CrespoO,KangSC,DanemanR,LindstromTM,HoPP,etal.(2011)Tyrosine KinaseInhibitorsAmeliorateAutoimmuneEncephalomyelitisinaMouse ModelofMultipleSclerosis.JClinImmunol.10.1007/s10875-011-9579-6[doi]. 8.SidhuM,CotonerCA,GulengB,ArihiroS,ChangS,etal.(2011)Small moleculetyrosinekinaseinhibitorsforthetreatmentofintestinalinflammation. InflammBowelDis.10.1002/ibd.21646[doi]. 9.MakiT,IchikawaT,BlancoR,PorterJ(1992)Long-termabrogationof autoimmunediabetesinnonobesediabeticmicebyimmunotherapywithantilymphocyteserum.ProcNatlAcadSciUSA89:3434–3438. 10.MakiT,GottschalkR,OgawaN,MonacoAP(2005)Preventionandcureof autoimmunediabetesinnonobesediabeticmicebycontinuousadministrationof FTY720.Transplantation79:1051–1055. 11.OgawaN,ListJF,HabenerJF,MakiT(2004)CureofovertdiabetesinNOD micebytransienttreatmentwithanti-lymphocyteserumandexendin-4. Diabetes53:1700–1705. 12.ParkerMJ,XueS,AlexanderJJ,WasserfallCH,Campbell-ThompsonML, etal.(2009)Immunedepletionwithcellularmobilizationimpartsimmunoregulationandreversesautoimmunediabetesinnonobesediabeticmice.Diabetes 58:2277–2284. 13.SherryNA,ChenW,KushnerJA,GlandtM,TangQ,etal.(2007)Exendin-4 improvesreversalofdiabetesinNODmicetreatedwithanti-CD3monoclonal antibodybyenhancingrecoveryofbeta-cells.Endocrinology148:5136–5144. 14.SimonG,ParkerM,RamiyaV,WasserfallC,HuangY,etal.(2008)Murine antithymocyteglobulintherapyaltersdiseaseprogressioninNODmicebya time-dependentinductionofimmunoregulation.Diabetes57:405–414. 15.Suarez-PinzonWL,PowerRF,YanY,WasserfallC,AtkinsonM,etal.(2008) Combinationtherapywithglucagon-likepeptide-1andgastrinrestores normoglycemiaindiabeticNODmice.Diabetes57:3281–3288. 16.SuriA,CalderonB,EsparzaTJ,FrederickK,BittnerP,etal.(2006) Immunologicalreversalofautoimmunediabeteswithouthematopoietic replacementofbetacells.Science311:1778–1780.311/5768/1778. 17.Couzin-FrankelJ(2011)Clinicalstudies.Tryingtoresettheclockontype1 diabetes.Science333:819–821.333/6044/819. 18.DarnellJE,Jr.,KerrIM,StarkGR(1994)Jak-STATpathwaysand transcriptionalactivationinresponsetoIFNsandotherextracellularsignaling proteins.Science264:1415–1421. 19.DarnellJE,Jr.(1997)STATsandgeneregulation.Science277:1630–1635. 20.IhleJN,WitthuhnBA,QuelleFW,YamamotoK,ThierfelderWE,etal.(1994) Signalingbythecytokinereceptorsuperfamily:JAKsandSTATs.Trends BiochemSci19:222–227. 21.IhleJN(1996)Januskinasesincytokinesignalling.PhilosTransRSoc LondBBiolSci351:159–166. 22.PardananiA(2008)JAK2inhibitortherapyinmyeloproliferativedisorders: rationale,preclinicalstudiesandongoingclinicaltrials.Leukemia22:23–30. 2404948. 23.SantosFP,KantarjianHM,JainN,ManshouriT,ThomasDA,etal.(2010) Phase2studyofCEP-701,anorallyavailableJAK2inhibitor,inpatientswith primaryorpost-polycythemiavera/essentialthrombocythemiamyelofibrosis. Blood115:1131–1136.blood-2009-10-246363.AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org11May2012|Volume7|Issue5|e36079

PAGE 12

24.vanGE,WeimarW,GastonR,BrennanD,MendezR,etal.(2008)Phase1 dose-escalationstudyofCP-690550instablerenalallograftrecipients: preliminaryfindingsofsafety,tolerability,effectsonlymphocytesubsetsand pharmacokinetics.AmJTransplant8:1711–1718. 25.VerstovsekS,KantarjianH,MesaRA,PardananiAD,Cortes-FrancoJ,etal. (2010)SafetyandefficacyofINCB018424,aJAK1andJAK2inhibitor,in myelofibrosis.NEnglJMed363:1117–1127. 26.LuoC,LaajaP(2004)InhibitorsofJAKs/STATsandthekinases:apossible newclusterofdrugs.DrugDiscovToday9:268–275. 27.KirkenRA,ErwinRA,TaubD,MurphyWJ,BehbodF,etal.(1999) TyrphostinAG-490inhibitscytokine-mediatedJAK3/STAT5a/bsignal transductionandcellularproliferationofantigen-activatedhumanTcells. JLeukocBiol65:891–899. 28.MeydanN,GrunbergerT,DadiH,ShaharM,ArpaiaE,etal.(1996)Inhibition ofacutelymphoblasticleukaemiabyaJak-2inhibitor.Nature379:645–648. 29.O’SheaJJ(1997)Jaks,STATs,cytokinesignaltransduction,andimmunoregulation:arewethereyet?Immunity7:1–11. 30.O’SheaJJ,PesuM,BorieDC,ChangelianPS(2004)Anewmodalityfor immunosuppression:targetingtheJAK/STATpathway.NatRevDrugDiscov 3:555–564. 31.WangLH,KirkenRA,ErwinRA,YuCR,FarrarWL(1999)JAK3,STAT,and MAPKsignalingpathwaysasnovelmoleculartargetsforthetyrphostinAG-490 regulationofIL-2-mediatedTcellresponse.JImmunol162:3897–3904. 32.BrightJJ,DuC,SriramS(1999)TyrphostinB42inhibitsIL-12-inducedtyrosine phosphorylationandactivationofJanuskinase-2andpreventsexperimental allergicencephalomyelitis.JImmunol162:6255–6262. 33.ConstantinG,BrockeS,IziksonA,LaudannaC,ButcherEC(1998)Tyrphostin AG490,atyrosinekinaseinhibitor,blocksactivelyinducedexperimental autoimmuneencephalomyelitis.EurJImmunol28:3523–3529. 34.ConstantinG,LaudannaC,BrockeS,ButcherEC(1999)Inhibitionof experimentalautoimmuneencephalomyelitisbyatyrosinekinaseinhibitor. JImmunol162:1144–1149. 35.BehbodF,Erwin-CohenRA,WangME,TrawickBW,QuX,etal.(2001) ConcomitantinhibitionofJanuskinase3andcalcineurin-dependentsignaling pathwayssynergisticallyprolongsthesurvivalofratheartallografts.JImmunol 166:3724–3732. 36.ChangelianPS,FlanaganME,BallDJ,KentCR,MagnusonKS,etal.(2003) PreventionoforganallograftrejectionbyaspecificJanuskinase3inhibitor. Science302:875–878. 37.Davoodi-SemiromiA,LalorayaM,KumarGP,PurohitS,JhaRK,etal.(2004) AmutantStat5bwithweakerDNAbindingaffinitydefinesakeydefective pathwayinnonobesediabeticmice.JBiolChem279:11553–11561. 38.Davoodi-SemiromiA,McDuffieM,LitherlandS,Clare-SalzlerM(2007) TruncatedpStat5BisassociatedwiththeIdd4locusinNODmice.Biochem BiophysResCommun. 39.Davoodi-SemiromiA,CheikhiA,XiaC,LitherlandS,Clare-SalzlerM(2007) ModulationofCD4 + Foxp3 2 T-cellswithaJAK-STAT5kinaseinhibitor. JImmunol178:S237–S23c. 40.GilmourKC,PineR,ReichNC(1995)Interleukin2activatesSTAT5 transcriptionfactor(mammaryglandfactor)andspecificgeneexpressioninT lymphocytes.ProcNatlAcadSciUSA92:10772–10776. 41.PlagnolV,HowsonJM,SmythDJ,WalkerN,HaflerJP,etal.(2011)Genomewideassociationanalysisofautoantibodypositivityintype1diabetescases. PLoSGenet7:e1002216.10.1371/journal.pgen.1002216[doi]. 42.HowsonJM,RosingerS,SmythDJ,BoehmBO,ToddJA(2011)Genetic analysisofadult-onsetautoimmunediabetes.Diabetes60:2645–2653.db110364[pii];10.2337/db11-0364[doi]. 43.HowsonJM,WalkerNM,SmythDJ,ToddJA(2009)Analysisof19genesfor associationwithtypeIdiabetesintheTypeIDiabetesGeneticsConsortium families.GenesImmun10Suppl1:S74–S84.gene200996[pii];10.1038/ gene.2009.96[doi]. 44.BarrettJC,ClaytonDG,ConcannonP,AkolkarB,CooperJD,etal.(2009) Genome-wideassociationstudyandmeta-analysisfindthatover40lociaffect riskoftype1diabetes.NatGenet41:703–707. 45.SeydelF,GarriganE,StutevossB,BelkinN,MakadiaB,etal.(2008)GM-CSF inducesSTAT5bindingatepigeneticregulatorysiteswithintheCsf2promoter ofnon-obesediabetic(NOD)mousemyeloidcells.JAutoimmun31:377–384. 46.Rumore-MatonB,ElfJ,BelkinN,StutevossB,SeydelF,etal.(2008)M-CSF andGM-CSFregulationofSTAT5activationandDNAbindinginmyeloidcell differentiationisdisruptedinnonobesediabeticmice.ClinDevImmunol2008: 769795. 47.BerthierCC,ZhangH,SchinM,HengerA,NelsonRG,etal.(2009)Enhanced expressionofJanuskinase-signaltransducerandactivatoroftranscription pathwaymembersinhumandiabeticnephropathy.Diabetes58:469–477. 48.ThironeAC,JebaileyL,BilanPJ,KlipA(2006)OppositeEffectofJAK2on Insulin-DependentActivationofMitogen-ActivatedProteinKinasesandAktin MuscleCells:PossibleTargettoAmeliorateInsulinResistance.Diabetes55: 942–951. 49.RezendeLF,VieiraAS,NegroA,LangoneF,BoscheroAC(2009)Ciliary neurotrophicfactor(CNTF)signalsthroughSTAT3-SOCS3pathwayand protectsratpancreaticisletsfromcytokine-inducedapoptosis.Cytokine46: 65–71. 50.ZhangF,ZhangQ,TengholmA,SjoholmA(2006)InvolvementofJAK2and Srckinasetyrosinephosphorylationinhumangrowthhormone-stimulated increasesincytosolicfreeCa2 + andinsulinsecretion.AmJPhysiolCellPhysiol 291:C466–C475. 51.O’SheaJJ(2004)TargetingtheJak/STATpathwayforimmunosuppression. AnnRheumDis63Suppl2:ii67–ii71. 52.GhoreschiK,LaurenceA,O’SheaJJ(2009)Januskinasesinimmunecell signaling.ImmunolRev228:273–287. 53.ImadaK,LeonardWJ(2000)TheJak-STATpathway.MolImmunol37:1–11. 54.O’SheaJJ,ParkH,PesuM,BorieD,ChangelianP(2005)Newstrategiesfor immunosuppression:interferingwithcytokinesbytargetingtheJak/Stat pathway.CurrOpinRheumatol17:305–311. 55.OhtaniT,IshiharaK,AtsumiT,NishidaK,KanekoY,etal.(2000)Dissection ofsignalingcascadesthroughgp130invivo:reciprocalrolesforS.Immunity12: 95–105. 56.UdyGB,TowersRP,SnellRG,WilkinsRJ,ParkSH,etal.(1997)Requirement ofSTAT5bforsexualdimorphismofbodygrowthratesandlivergene expression.ProcNatlAcadSciUSA94:7239–7244. 57.BatesSH,GardinerJV,JonesRB,BloomSR,BaileyCJ(2002)Acute stimulationofglucoseuptakebyleptininl6musclecells.HormMetabRes34: 111–115. 58.WangX,ShawS,AmiriF,EatonDC,MarreroMB(2002)Inhibitionofthe Jak/STATsignalingpathwaypreventsthehighglucose-inducedincreaseintgfbetaandfibronectinsynthesisinmesangialcells.Diabetes51:3505–3509. 59.SteinmanRM(2003)Someinterfacesofdendriticcellbiology.APMIS111: 675–697. 60.SteinmanRM,HawigerD,NussenzweigMC(2003)Tolerogenicdendriticcells. AnnuRevImmunol21:685–711.61.KishimotoH,SprentJ(2001)AdefectincentraltoleranceinNODmice.Nat Immunol2:1025–1031. 62.Davoodi-SemiromiA,Copper-DeHoffR(2011)NotetoReaderson:Longterm treatmentwithACEinhibitorenalaprildecreasesbodyweightgainand increaseslifespaninrats.BiochemicalPharmacologyInpress2012, doi:10.1016/j.bcp.2011.12.033.AG490PreventsandReversesDiabetes PLoSONE|www.plosone.org12May2012|Volume7|Issue5|e36079