TECHNICAL ADVANCE Open Access A novel model of adenine - PDF document

TECHNICALADVANCEOpenAccess Anovelmodelofadenine-induced tubulointerstitialnephropathyinmice TingJia 1,2 † ,HannesOlauson 1 † ,KarolinaLindberg 1 ,RisulAmin 1 ,KarinEdvardsson 1 ,BengtLindholm 2 , GöranAndersson 3 ,AnnikaWernerson 1 ,YvesSabbagh 4 ,SusanSchiavi 4 andTobiasELarsson 1,5* Abstract Background: Invivo modelsofuremiaareimportanttoolstostudynumerousaspectsofacuteandchronickidney dissectingtheimpactofselectedtargetgenesinrenalfailure.Adenine-basedprotocolstoinducerenalfailureare availableinrats,buthavenotbeenadaptedinmiceduetotheirreluctancetoconsumeadenine.Inthecurrent paperwedevelopedanovelmethodforinductionofrenalfailurethroughdietarydeliveryofadeninemixedina casein-baseddiet. Results: Afteraninductionphase,astablemodelofrenalimpairmentwasobtained(targeturearange80 – 100mg/dL), mimickingseveralaspectsofchronickidneydisea se-mineralandbonedisorderincludingsecondary hyperparathyroidism,boneabnormalitiesandpathologic alelevationofFGF23.Nodeathsoccurredandthelevelof uremiawasadaptablethroughadjustmentsoftheadenine content,providingsignificantadvantagescomparedto existingmodels.Inan8-weekproof-of-conceptstudy,renal histologyshowedmainlyatubulointerstitialdamagewith infiltratingleukocytes,interstitialedemaandwideningo ftheBownman'sspace.Fibrosiswaspresentinmostanimalsas definedbyhistologyandgeneexpressionchangesoffibros ismarkers.Parathyroidcellproliferationwasmarkedly letalhistologyshowedincreasedtrabecularboneandbone marrowadipositywhereasbonebiomarkers(CTXandPINP)suggestedhigherboneformation,butsurprisingly,lower boneresorptionandperturbationsinmineralmetabolism. Conclusions: Wepresentanovel,non-surgicalmethodforinductionofrenalfailureinmice.Thisisanimportant complementtoexistinguremicmodelsforpathophysiologic alstudiesinacuteandchronickidneydisease,especiallyin termsoftubulointerstitiallesions. Keywords: CKD,Chronickidneydisease,CKD-MBD,FGF-23,FGF 23,Mineralmetabolism,Experimentalrenalfailure Background Chronickidneydisease(CKD)isaglobalhealthburden [1],yeteffectivetreatmentsforitsprevention,progres- sionandassociatedcomplicationsarecurrentlylacking. Animalmodelsofrenalfailureareimportanttoolsto studypathophysiologicaleventsinkidneydiseasethat allowstranslationalstudiesaimingtoimprovemanage- mentofCKDpatients.Duetothehighavailabilityof geneticallyengineeredmousestrains,uremicmouse modelsprovidetheopportunitytoinvestigatetheimpact ofspecifictargetgenesinthesettingofrenalfailure. micearemostlydependentonsurgicalinterventions. Themostwidespreadmethodscurrentlyusedareunilat- eralureteralobstruction[2-7],whichleadstointerstitial fibrosisbyinfiltrationofmacrophagesandtubularcell deathbyapoptosisandnecrosis,and5/6nephrectomy [8,9].Thelattertechniqueusuallyinvolvestwoseparate procedures,firsttwothirdsofonekidneyisdestructed byelectrocoagulationandafterrecoveryacontra-lateral nephrectomyisperformed.The5/6nephrectomymodel hasseverallimitationsincludingasubstantialmortality ratewhennotperformedadequately,non-reversibility *Correspondence: tobias.larsson@ki.se † Equalcontributors 1 DepartmentofClinicalScience,InterventionandTechnology,Karolinska Institutet,Stockholm,Sweden 5 DepartmentofNephrology,KarolinskaUniversityHospital,Stockholm, Sweden Fulllistofauthorinformationisavailableattheendofthearticle ©2013Jiaetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedtheoriginalworkisproperlycited. Jia etal.BMCNephrology 2013, 14 :116 http://www.biomedcentral.com/1471-2369/14/116 andphenotypicalterationsrelatedtothesurgicalprocedureratherthanimpairedkidneyfunction[10].Themethodisalsoassociatedwithrelativelylargeinter-individualandinter-laboratoryvariationsanditsavail-abilitymaybecompromisedbylackofsurgicalexpertiseandtheappropriateoperatingfacilities.Tocircumventtheseobstacles,weaimedatestablishinganovel,non-surgicalmodelofrenalfailureinmicebyemployinganadenine-basedprotocol.Importantly,therearewell-characterizedprotocolsforadenine-inducedrenalfailureinrats,yetthistechniquehasnotbeenadaptedinmiceduetotheiraversiontoadeninefeeding.AnimalexperimentsTheexperimentswereconductedincompliancewiththeguidelinesofanimalexperimentsofKarolinskaInstitutetandthestudyprotocolwasapprovedbytheregionalethicalcommittee(StockholmSouthethicalcommittee,approvalnumberS184-10andappendixS19-13).8-week-oldC57BL/6Jmicewerehousedinstandardcageswithwoodchipbeddingandapaperrollforenrichmentatconstantambienttemperature(2122°C)andhumidity(40-50%)witha12-hourlightcycle.Allanimalshadfreeaccesstotapwaterandtheassigneddiet.Beforestudystart,allmicewereallowedacclimatizationtotheanimalfacilityconditionsandthecasein-basedchowduringa7-dayperiod.Toprovideanadenine-containingchowconsumedbythemice,adeninewasmixedwithacasein-baseddietthatbluntedthesmellandtaste.AdeninewaspurchasedfromSigmaAldrich(MO,USA)andthepowderedcasein-baseddietfromSpecialDietsServices(SDS,UK)(referencenumber824522).Otheringredientsofthedietaremaizestarch(39.3%),casein(20.0%),maltodextrin(14.0%),sucrose(9.2%),maize/cornoil(5%),cellulose(5%),vitaminmix(1.0%),DL-methionine(0.3%)andcholinebitartrate(0.2%).Totalphosphatecontentwas0.9%andtotalcalciumcontentwas0.6%.Datapresentedhereinisderivedfroman8-weekexperimentusing8-week-oldC57BL/6Jwild-typemice.Toexcludethepossibilityofanimpactofthecaseindietperseeonrenalfunction,thecontrolgroup(n=5)wasfedthesamecaseindietastheadeninegroup(n=9)butwithoutadditionofadenine.SerumandurinebiochemistriesBloodwascollectedaftercardiacpunctureatsacrificeandbytailveinincisionatintermediatetimepoints.Urinewascollectedasspoturinesamplesafterspontaneousurin-ation.Serumandurinecalcium,phosphorous,creatinineandurealevelsweremeasuredonaKonelab20XTi(ThermoScientific,Finland).Creatinineconcentrationswerevalidatedwithacolorimetricassay(BioChain,CA,USA)yieldingnearlyidenticalresults(rho=0.95and0.98forserumandurinecreatininerespectively)whencom-paredwiththeKonelabtechnique.PTHwasmeasuredbyamouseintactPTHELISAkit(Immutopics,CA,USA),FGF23withanintactFGF23ELISA(Kainos,Japan)and1,25(OH)D,CTXandPINPwithEIAkits(Immunodiag-nosticSystems,UK).Kidneysandparathyroidglandswerefixedin4%formal-dehyde,embeddedinparaffinandsectionedaccordingtostandardprocedures.Bonesweredecalcifiedinabuf-fercontaining20%formicacid.AlltissuesweresubjecttoHematoxylinandEosinstaining.KidneysectionswerealsostainedforPeriodicacid-Schiffstain(PAS)forpoly-saccharidesandmucosubstances(VWRInternational,Sweden),Trichrome(Ladewig)formuscleandcollagen(HistolabProductsAB,Sweden)andvonKossaformin-eralizedtissue(VWRInternational),accordingtothemanufacturersinstructions.Thekidneysandboneswereevaluatedinablindedfashionbyanexperiencedkidneypathologist(AW)andbonepathologist(GA),respect-ively.Immunohistochemistrywasperformedaccordingtostandardprotocolsusingarabbitmonoclonalanti-Ki67antibody(SP61:400,ThermoScientific,CA,US)andarabbitanti-humanMyeloperoxidaseantibody(A3981:100,Dako,Denmark).ProliferationindexinparathyroidglandswascalculatedasthenumberofKi67positivecellsdividedbytotalnumberofcellsinfourconsecutivesections.AlizarinredandvascularcalcificationThethoracictoabdominalpartoftheaorta(n5fromeachgroup)wasdissected,splitlongitudinallyandlaidflat.Thetissuewasincubatedwith1%alizarinredS-stainingsolution(AlfaAesar,Germany)atroomtemperaturefor10minutesandwashedthreetimeswith70%ethanolfor20minuteseach.AnalysisofrenaltranscriptsKidneyswerehomogenizedusingaTissueLyzerLT(Qiagen,Netherlands)andtotalRNAwasextractedusingE.Z.N.A.TotalRNAKitI(OmegaBio-tek,GA,US).DNAwasremovedwithE.Z.N.A.RNase-FreeDNaseSet(OmegaBio-tek).First-strandcDNAwassynthesizedusingiScriptcDNASynthesisKit(Bio-Rad,Hercules,CA,US).ForrealtimeqPCRanalysistheCFX96Real-TimePCRDetectionSystemandiQSYBRGreenSupermix(Bio-Rad)wereused.Therelativegeneexpressionwascalculatedusingthe2Cqmethodnormalizingthegeneofinterestto-actininthesamesample.etal.BMCNephrologyPage2of8http://www.biomedcentral.com/1471-2369/14/116 GraphPadPrism5.0(GraphPadSoftwareInc,CA,US)wasusedforstatisticalanalysis.Allvaluesareexpressedasmean±SEMunlessotherwisestated.Differencesbe-tweenadenine-treatedmiceandcontrolswerecalculatedusingtheMannWhitneynon-parametrictest.wereconsideredstatisticallysignificant.ProtocoldevelopmentWeperformedaseriesofpilotstudiestoestablishtheoptimaldegreeofrenalimpairmentatwhichcommonbiochemicalabnormalitiesassociatedwithrenaldys-functionoccurredwithoutincreasedmorbidityormor-tality.Forthispurpose,serumurealevelsrangingfrom100mg/dLwereconsideredoptimal.Urealevelsabove100mg/dLwereassociatedwithincreasedmor-bidity(weightlossandreducedphysicalactivity),whereasureaconcentrationsof70-80mg/dLledtoaweightgainandapartialreversalofbiomarkersassoci-atedwithimpairedkidneyfunctionsuchasphosphor-ousandPTH.Importantly,thecasein-dietwithoutadditionofadeninewasusedascontroldietanddidnotinfluenceserumureaandcreatininelevelsorrenalhistologycomparedtomiceonaregulardiet(datanotshown).Therequiredadeninedosetoachievethetargetureaintervalwasdeterminedto0.3%adenineduringa10-dayinductionphaseand0.15-0.20%duringamaintenancephase.Whenadheringtothisurearange,nodeathsoc-curredduringtheentirestudyorinsupportingpilotstudies.Accordingly,weproposethattheadenineconcentrationcanbemodifiedadlibitumbetween0.15-0.20%duringthemaintenancephase.Notably,wealsoevaluatedadeninede-liverybygavage,whichresultedinanapproximately30%overallmortalityandhighervariabilityinurealevelandisthereforenotrecommended.Proof-of-conceptstudyWeperformedan8-weekexperiment,atimeframegen-erallyendorsedinotheruremicmodelsofrenalfailuretostudyCKDcomplicationssuchasvascularcalcifica-tion[10].ThestudyprotocolisshowninFigure1.Thestudywasprecededbya7-dayadaptationphaseofthecaseindietwithoutadditionofadenine,andcompriseda10-dayinductionphase(day09)andamaintenancephase(day1056).Duringthemaintenancephase,theadeninecontentwasmodifiedasdescribedaboveintheinterval0.15-0.20%toachievethedesiredurealevelof100mg/dL.Ofnote,loweringtheadenineconcen-trationonday15and40,from0.2%to0.15%,resultedinarapiddeclineinureaandPTHlevels.However,long-termreversibilityandhistologicalimprovementsafteradeninewithdrawalwasnotexamined.Bodyweightandserum/urinebiochemistriesThetemporalchangesinbodyweightandmarkersofkidneyfunctionaredepictedinFigure2A.Therewasasignificantdeclineinbodyweightduringtheinductionphasebutitwasstabilizedandessentiallyunalteredduringthemaintenancephase.Inaccordancewithpre-viousreportsureaappearedtobeamoreaccuratemarkerofuremiathancreatinine,whereastheratioofcreatinine/bodyweightparalleledtheurealevels[11].Decreasedratiosbetweenurineurea/serumureaandurinecreatinine/serumcreatinineconfirmedreducedclearanceofthesemetabolites(Figure2A).Import-antly,theadenine-exposedmicedidnothaveincreasedproteinuriacomparedtocontrols.ThisislikelyexplainedbytheC57BL/6strain'sknownresistancetowardsdevelopmentofproteinuriaincombinationwiththetubulointerstitialnatureoftherenaldamageinthismodel[12].MarkersofboneandmineralmetabolismareshowninFigure2B.SimilartoCKDpatients,theadeninegroupdevelopedasignificanthyperphosphatemia,secondaryhyperparathyroidismandseverelyelevatedFGF23paralleledbyanincreasedurinaryexcretionofphos-phorous[13].Therewasnochangeinurinaryexcretionofcalcium(p=0.66forbaselineversusendpoint).Atendpoint,CTXwassignificantlydecreasedwhereastherewasaborderlinesignificantincreaseinPINPinadenine-treatedmicecomparedtomiceonacontroldiet,suggestingareducedboneresorptionbutanincreasedboneformation(Figure2B).Histologicalanalysisofkidneys,parathyroidglandsandboneareshowninFigure3A-CandinAdditionalfile1:FigureS1.Renalhistologyshowedaperitubularleukocyte Induction phaseMaintenance phase –71 1530405056Adenine:0%0.3%0.2%0.15%0.15%0.20%.2% t Baseline Figure1Schematicviewofthe8-weekproof-of-conceptstudyofadenine-inducedrenalfailureinmice.Thestudywasprecededbya7-dayadaptationphase,andcompriseda10-dayinductionphase(day09)andamaintenancephase(day10etal.BMCNephrologyPage3of8http://www.biomedcentral.com/1471-2369/14/116 infiltrationandinterstitial/peritubularedemareflectingthatthekidneydamageismainlytubulointerstitial.Posi-tiveimmunostainingforMyeloperoxidaseconfirmedthattheperitubularleukocytesweremainlycomprisedofneutrophilgranulocytes.(Additionalfile2:FigureS2).AnincreaseintheBowman´sspacewasseeninsomebutnotallglomeruli.Focalmicro-abscesseswerepresentinsome,butnotall,tissuespecimensexaminedintheadenine- Control groupAdenine group Time (days) Weight (grams)*************** Time (days) Time (days) BaselineEndpoint BaselineEndpoint Control groupAdenine group Time (days) Time (days) Time (days)**** Time (days) Time (days) Control g roupAdenine g roup1,25(OH)2D (pmol/L)** Control g roupAdenine g roupCTX (ng/mL)** Control g roupAdenine g PINP (ng/mL) Figure2Bodyweightandbiochemicalparametersduringthestudy.A:Bodyweightandrenalfunctionparametersduringthestudy.Bodyweight(top):Bodyweightwasreducedduringthe10-dayinductionphaseintheadenine-treatedgroupbutremainedstableduringthemaintenancephaseuntilendpoint.Markersofkidneyfunction:serumurea,serumcreatinine/bodyweight,urineurea/serumureaandurinecreatinine/serumcreatinineindicatedreducedrenalclearancerateinadenine-treatedmice.*0.05;**0.01;***0.001;****0.0001.:Temporalchangesinserumbiochemistriesofmineralmetabolism.Atendpoint,therewasasignificantincreaseinseruminorganicphosphorous,PTHandFGF23butadecreasein1,25(OH)Dlevelinadenine-treatedmice.ThebonemarkerCTXwasdecreasedandPINPborderlineincreasedintheadeninegroup,supportingareducedboneresorptionbutincreasedboneformation.*0.05;**0.01.etal.BMCNephrologyPage4of8http://www.biomedcentral.com/1471-2369/14/116 group.Asummaryofhistopathologicalfindingsinthe kidneyisprovidedinTable1. Arealdeterminationfromserialsectionedparathyroid glandsrevealednoovertglandularhypertrophy.Con- versely,proliferationrateintheparathyroidsdetermined byKi67indexwasincreased(Figure3B),andinline withtheincreasedPTHlevelinadenine-treatedmice. Sectionsoffemursshowedanextendedbonetrabeculae andincreasedbonemarrowadiposityconsistentwith theserumpatternofthebonemarkersPINPandCTX. Alizarinredstainingsdidnotshowanyovertvascular calcificationinthethoracicaortasofadenine-exposed mice(Additionalfile3:FigureS3). Geneexpressionchangesoflocalinflammatoryand fibrosismarkers Weanalyzedthetranscriptlevelsofanumberoflocally activatedgenesassociatedtorenalinflammationand fibrosis[14,15].InflammatorymarkersMmp3(GeneID: 17392),Mmp9(17395),Il7r  (16197),Ccl20(20297)and Ccl5(20304)wereallsignificantlyupregulatedinthe adenine-treatedmice(Figure4A),whereasCxcr2was unchanged(datanotshown).Correspondingmarkersof fibrosisTgfb1(21803),Col1a1(12842)andCcl2(20296) weresignificantlyupregulatedinmiceonanadenine diet(Figure4B).PrimersusedforrealtimeqPCR analysisarepresentedinAdditionalfile4:TableS1. Discussion Mousemodelsofuremiaareimportanttoolsfor invivo translationalstudiessuchasexaminingtheimpactof specificgenesintransgenicorknockoutmiceandto validatepotentialtherapeuticinterventionspriortopre- clinicaltestinginhumans.Thelimitationsofexisting surgicalmodelsofuremiainmice,includingrequire- mentsofsurgicalskills,demandofanimalfacilitiesto supportpost-operativecare,highmortalityrateandless flexibilityintermsofdynamicalterationsinurealevels, promptedustodevelopanon-surgicalmodelofrenal dysfunctionbasedondietaryintakeofadenine.This approachhassuccessfullybeenusedinratsbutnotin A B C D F E G Control group Adenine group Adenine group H&EPAS Ladewigvon Kossa H&E H Control group Adenine group Ki67 A B Control group Adenine group H&E A. B. C. Figure3 Histologicalanalysesofkidneys,parathyroidglands andbones.A :Renalhistologyshoweddepositionofsymmetric crystallinestructuresinatubularlumen(arrowA),microabscesses (B)anddilatedtubules(C).PASstainingshoweddilatedBowman ’ s space(D),atrophictubuliwithproteincasts( “ thyroidization ” )(E)and tubularatrophywiththickeningofthetubularbasementmembrane (F).Ladewigstainingrevealedamildinterstitialfibrosis(G).Extensive calcificationoftubularstructures(H)wasseenwithvonKossa staining. B :Parathyroidglandswerenothypertrophicbuthada significantlyincreasedproliferationratedeterminedbyKi67index (8.7±0.7%vs2.3±0.5%; p 0.0001). C :Inbone,therewasan increasednumberandthicknessofsubmetaphysealbonetrabeculae (arrowA)andincreasedadipocytecontentinthebonemarrow(B). Jia etal.BMCNephrology 2013, 14 :116 Page5of8 http://www.biomedcentral.com/1471-2369/14/116 micebecauseoftheirreluctancetoconsumeadenine.Accordingly,additionofadeninetoastandardmousechowelicitshighmorbidityandmortalityduetostarva-tionandmalnutritionratherthanrenalfailure.Inourmodel,thiswascircumventedbymixingtheadenineinacasein-basedchow,inwhichthecaseineffectivelyremovedtheinherentsmellandtasteofadenine.Sincethesensitivitytoadenineaswellasfoodintakemayvarybetweenvariousstrainsofmice,weproposeaprotocolwithadlibitumchangesinadenineconcentra-tionbetween0.15-0.20%duringthemaintenancephase.Theacceptedvariabilityinbloodurealevelduringthemaintenancephasemaycauseinter-individualvariationsinkidneyfunctionthatcouldaffectthephenotype.How-eversuchvariabilityisinevitableandpresentinotheruremicanimalmodelsaswell.Onthecontrary,provid-ingatargetureaintervalprovidestheopportunitytoadjustbloodurealevelsaccordingtothedesiredout-comeormechanismsofinterest.Theapparentimprovementinkidneyfunctionafterloweringtheadenineconcentration,asindicatedbyadeclineinbloodureaandPTHlevels,suggestsatleastapartialreversibilityofrenalimpairment.Howeverwedidnotexaminewhetheradeninediscontinuationtranslatedintolong-termhistologicalimprovements.Giventheseverityoftherenalhistopathologicallesionsobservedaftereightweeksofadenineexposure,weanticipatethatlong-termadeninechallengewillcausechronicrenalfailurewithlessreversibilityasobservedinadenine-induceduremiainrats[16].Regardless,thepossibilitytoshort-termmodulatekidneyfunctionprovidessignifi-cantbenefitsintermsofthepossibilitytoextendstudyprotocolsandtoanalyzeoutcomevariablesinvariousstrataofkidneyfunction.Additionaladvantagesofourmodelincludezeromortality,whichlimitsthenumberofanimalsneededforinduction,andthesmallinter-individualvariationinrenalfunctionthatcontraststhe5/6nephrectomymodel[10,17].Italsoprovidesagoodoptionforresearcherswithlimitedsurgicalcompetenceand/orrestrictionsinpost-operativecare.Somemorecommonnon-surgicaloptionstostudyuremiaincludingradiationnephropathyandadministra-tionofnephrotoxicdrugssuchasfolicacid[18], Table1Histopathologicalevaluationofkidneysfromcontrolandadenine-treatedmiceCompartmentParameterControl(n=5)(n=8)Sclerosis(present/absent)0/50/8DilatationofBowmanscapsules(present/absent)0/55/3Roundedcristalloid/amorphousstructuresintubularlumina(present/absent)0/58/0Celldebris/necroticmaterialandPMNintubularlumina(00(00)1(1Tubularatrophy(03)0(00)2(2Thyroidization(03)0(00)1(1Focaldilatationoftubuli(present/absent)0/56/2Focalcalciumdeposits(present/absent)0/55/3Fibrosis(03)0(00)1(1Inflammation(03)0(00)0(0Morphology(pathological/normal)0/50/8Theparameterstubularatrophy,celldebris/necroticmaterialandpolymorphonuclearleukocytes(PMN)intubularlumina,thyroidization,interstitialfibrosisandinterstitialinflammationaregradedasfollows:0;affecting0-5%oftherenalarea,1;6-25%,2;26-50%,and3;�50%.Dataispresentedasmedian(range).Roundedcristalloid/amorphousstructuresintubularlumina,focaldilatationandfocalcalciumdepositsintubuliaregradedaspresentorabsent.Vesselmorphologyisgradedaspathologicalornormal. B.A. Tgfb1Col1a1Ccl2 Mmp3Mmp9Il7r Ccl20Ccl5 Figure4Expressionofrenal-derivedinflammatoryandfibrosisgenes.A)InflammatorymarkersMmp3,Mmp9,Il7r,Ccl20andCcl5,andMarkeroffibrosisTgfb1,Col1a1andCcl2wereupregulatedinadeninetreatedmice.Relativegeneexpressioninthecontrolgroupissetto1.Whitebars;controlgroup,blackbars;adeninegroup.***0.001.etal.BMCNephrologyPage6of8http://www.biomedcentral.com/1471-2369/14/116 cyclosporinA[19]andcisplatin[20]shouldbemen-tioned.However,thesemodelsarenon-reversible,strain-dependentandoflimiteduseduetosystemictox-icity.Geneticmousemodelsmimickingvariousaspectsofkidneyfailurearealsoavailable,butthesearecompromisedbytheneedforbreedingtocreatecom-binedbackgroundswithothergeneticallyalteredmousestrains[20].Severaldifferenturemicmodelshavealsobeendevel-opedinrats.Advantageswithratmodelsarethatcollectedbloodandurinevolumesaresignificantlygreater,bloodsamplesatintermediatetimepointsaremoreeasilyobtainableandcertainorganssuchaspara-thyroidglandsarereadilyidentified.Anotherapparentdifferenceisthatratstoleratehigheradenineexposure,whichreportedlyproduce35timeshigherbloodureaconcentrationsthanourmodel.Thismayalsoimpactrenalhistologysinceratsinadditionaltotubulardamagegenerallyalsosufferfromextensiveglomerulardamagewhichwasnotfoundinourmodel[21].Thereareseveralbiochemicalfindingsofinterestinourmodel.Serumcalciumlevelwasunalteredintheadenine-treatedmicelikelyduetoacompensatoryriseinPTH,whichlargelymimicsthesituationinpatientswithCKDstage34.Normalserumcalciumconcentra-tionshavealsobeenreportedinotherCKDmodels[22].AnotherstrikingfindingisthecontinuousandmarkedriseinFGF23althoughotherbone-mineralmarkersremainedmoreconstantduringthemaintenancephase.ThissupportsthepresenceofrenalderivedfactorsthatregulatesFGF23synthesisinbone.Alternatively,thetubulardamagemayseverelyhamperintactFGF23deg-radationleadingtoaccumulationofcirculatingFGF23protein.Thepatternofbonemarkerssuggestinganin-creasedboneformationbutdecreasedboneresorptionissomewhatunexpectedinauremicmodelofsecondaryhyperparathyroidismandanticipatedhighboneturnoverrate.Themechanism(s)underlyingreducedboneresorp-tionareunclearbutcouldspeculativelybeduetoim-pairedosteoclastfunctionasaresultoftheexceptionallyelevatedlevelsofFGF23.Somepotentiallimitationsshouldbementioned.Wecannotexcludethepossibilityofsystemictoxicityororgan-specificdamagescausedbytheadenine.Becausetubulartoxicityofadeninemetabolitesistheunderlyingmechan-ismofadenine-inducedrenalfailure[23,24],ourmodelprimaryreflectsatubulointerstitialdiseasewhereasthemostcommoncauseofCKDinhumanisglomerularscar-ringsecondarytovasculardamage.Thus,ourmodelshouldnotberegardedasamodelofCKDpersebutratherasacomplementarymodelofrenalfailure.Wedidnotdeter-minethedietary(caloric)intake,yetbasedonpreviousexperimentsinratstheadenine-fedanimalsmayhavesomewhatloweroveralldietaryintake.Further,wedidnotperformdynamicbonehistomorphometryalthoughaden-inemodelsinratsproduceahighturnoverphenotype.Finally,possiblestraindifferencesasfoundinothermousemodels[25]warrantfurtherinvestigation.ConclusionsWepresentanovelnon-surgicalprotocolforinductionofrenalfailureinmice.Thiswillbeanimportantcomple-menttoexistingmodelsforthestudyofpathophysiologicaleventsandcomplicationsrelatedtoacuteandchronickidneydisease,specificallyintermsoftubulointerstitiallesionsandabnormalitiesinmineralmetabolism.AdditionalfilesAdditionalfile1:FigureS1.Highresolutionimagesofrenalhistology.Hematoxylinandeosinstain(upperleftpanel)showeddepositionofsymmetriccrystallinestructuresinatubularlumen,microabscessesanddilatedtubules.PASstain(upperrightpanel)showeddilatedBowmanspace,atrophictubuliwithproteincasts(thyroidization)andtubularatrophywiththickeningofthetubularbasementmembrane.Ladewigstain(lowerleftpanel)revealedamildinterstitialfibrosis.ExtensivecalcificationoftubularstructureswasseenwithvonKossastain(lowerrightpanel).Additionalfile2:FigureS2.PositiveimmunostainingforMyeloperoxidaseconfirmedthattheperitubularleukocytesweremainlycomprisedofneutrophilgranulocytes.Additionalfile3:FigureS3.AlizarinredS-stainingofrepresentativesegmentsfromthoracicaorta.Novascularcalcificationwasfoundincontroloradenine-treatedmice.Additionalfile4:TableS1.ListofprimersusedforrealtimeqPCRChronickidneydisease;FGF23:Fibroblastgrowthfactor-23;EIA:Enzyme-linkedimmunoassay;PTH:Parathyroidhormone;CTX:Carboxyl-terminalcollagencrosslinks;PINP:ProcollagentypeIN-terminuspropeptide.CompetinginterestsTheauthorsdeclarethattheyhavenocompetinginterests.TJ:Acquisitionandanalysisofdata;draftingandrevisionofmanuscript;statisticalanalysis.HO:Acquisitionandanalysisofdata;draftingandrevisionofmanuscript;statisticalanalysis.KL:Datacollection;revisionofmanuscript.RA:Datacollection;revisionofmanuscript.KE:Datacollection;revisionofmanuscript.BL:Revisionofmanuscript.GA:Acquisitionandanalysisofbonedata;revisionofmanuscript.AW:Acquisitionandanalysisofrenalhistology;revisionofmanuscript.YS:Conceptionanddesignofstudy;revisionofmanuscript.SS:Conceptionanddesignofstudy;revisionofmanuscript.TEL:Conceptionanddesignofstudy;dataanalysis;draftingandrevisionofmanuscript.Allauthorsreadandapprovedthefinalmanuscript.AcknowledgementsWewouldliketothankChristinaHammarstedt,AnneliHansson,MariaNorgårdandthestaffoftheanimalcorefacilityatAKM6,HuddingeUniversityHospitalfortheirvaluableassistance.Thestudywasinvestigator-initiatedanddrivenandsupportedbyacademicgrantsfromTheSwedishFoundationforStrategicResearch,SwedishResearchCouncil(K2012-55P-22137-01-06),SwedishKidneyFoundation,KarolinskaInstitutetandKarolinskaUniversityHospital.YSandSSareemployeesofSanofi-Aventis.TELisapart-timeemployeeofAstellas.etal.BMCNephrologyPage7of8http://www.biomedcentral.com/1471-2369/14/116 Authordetails 1 DepartmentofClinicalScience,InterventionandTechnology,Karolinska Institutet,Stockholm,Sweden. 2 RenalMedicineandBaxterNovum,Karolinska Institutet,Stockholm,Sweden. 3 DepartmentofPathology,Karolinska Institutet,Stockholm,Sweden. 4 Sanofi-GenzymeR&DCenter,Genzyme,A SanofiCompany,Framingham,USA. 5 DepartmentofNephrology,Karolinska UniversityHospital,Stockholm,Sweden. Received:21August2012Accepted:17May2013 Published:30May2013 References 1.ZhangQL,RothenbacherD: Prevalenceofchronickidneydiseasein population-basedstudies:systematicreview. BMCPublicHealth 2008, 8 :117. 2.PimentelJLJr,SundellCL,WangS,KoppJB,MonteroA,Martinez- MaldonadoM: RoleofangiotensinIIintheexpressionandregulationof transforminggrowthfactor-betainobstructivenephropathy. KidneyInt 1995, 48 (4):1233 – 1246. 3.PaulsonDF,FraleyEE: Compensatoryrenalgrowthafterunilateralureteral obstruction. KidneyInt 1973, 4 (1):22 – 27. 4.ManuchaW,OliverosL,CarrizoL,SeltzerA,VallesP: Losartanmodulation onNOSisoformsandCOX-2expressioninearlyrenalfibrogenesisin unilateralobstruction. KidneyInt 2004, 65 (6):2091 – 2107. 5.KlahrS: Urinarytractobstruction. SeminNephrol 2001, 21 (2):133 – 145. 6.DickerSE,ShirleyDG: Compensatoryhypertrophyofthecontralateral kidneyafterunilateralureteralligation. JPhysiol 1972, 220 (1):199 – 210. 7.ChevalierRL: Growthfactorsandapoptosisinneonatalureteral obstruction. JAmSocNephrol 1996, 7 (8):1098 – 1105. 8.MorrisonAB: Experimentallyinducedchronicrenalinsufficiencyinthe rat. LabInvest 1962, 11 :321 – 332. 9.ShimamuraT,MorrisonAB: Aprogressiveglomerulosclerosisoccurringin partialfive-sixthsnephrectomizedrats. AmJPathol 1975, 79 (1):95 – 106. 10.ShobeiriN,AdamsMA,HoldenRM: Vascularcalcificationinanimalmodels ofCKD:Areview. AmJNephrol 2010, 31 (6):471 – 481. 11.SongS,MeyerM,TurkTR,WildeB,FeldkampT,AssertR,WuK,KribbenA, WitzkeO: SerumcystatinCinmousemodels:areliableandprecise markerforrenalfunctionandsuperiortoserumcreatinine. NephrolDial Transplant 2009, 24 (4):1157 – 1161. 12.GurleySB,MachCL,StegbauerJ,YangJ,SnowKP,HuA,MeyerTW, CoffmanTM: Influenceofgeneticbackgroundonalbuminuriaand kidneyinjuryinIns2(+/C96Y)(Akita)mice. AmJPhysiolRenalPhysiol 2010, 298 (3):F788 – F795. 13.WolfM: Forgingforwardwith10burningquestionsonFGF23inkidney disease. JAmSocNephrol 2010, 21 (9):1427 – 1435. 14.HengerA,KretzlerM,DoranP,BonrouhiM,SchmidH,KissE,CohenCD, MaddenS,PorubskyS,GroneEF, etal : Geneexpressionfingerprintsin humantubulointerstitialinflammationandfibrosisasprognostic markersofdiseaseprogression. KidneyInt 2004, 65 (3):904 – 917. 15.KlahrS,MorrisseyJ: Obstructivenephropathyandrenalfibrosis. AmJPhysiolRenalPhysiol 2002, 283 (5):F861 – F875. 16.OkadaHKY,YawataT,UyamaH,OzonoS,MotomiyaY,HiraoY: Reversibility ofadenine-inducedrenalfailureinrats. ClinExpNephrol 1999, 3 (2):82 – 88. 17.MassyZASM,TribouilloyC,DrüekeTB: Animalmodels – whattheycantell usaboutvascularcalcificationinCKD. SeminDialysis 2007, 20 (2):110 – 112. 18.OrtegaA,RamilaD,ArduraJA,EstebanV,Ruiz-OrtegaM,BaratA,GazapoR, BoschRJ,EsbritP: Roleofparathyroidhormone-relatedproteinin tubulointerstitialapoptosisandfibrosisafterfolicacid-induced nephrotoxicity. JAmSocNephrol 2006, 17 (6):1594 – 1603. 19.KoppJB,KlotmanPE: Cellularandmolecularmechanismsofcyclosporin nephrotoxicity. JAmSocNephrol 1990, 1 (2):162 – 179. 20.YangHC,ZuoY,FogoAB: Modelsofchronickidneydisease. DrugDiscov TodayDisModels 2010, 7 (1 – 2):13 – 19. 21.YokozawaT,ZhengPD,OuraH,KoizumiF: Animalmodelofadenine- inducedchronicrenalfailureinrats. Nephron 1986, 44 (3):230 – 234. 22.DaviesMR,LundRJ,MathewS,HruskaKA: Lowturnoverosteodystrophy andvascularcalcificationareamenabletoskeletalanabolisminan animalmodelofchronickidneydiseaseandthemetabolicsyndrome. JAmSocNephrol 2005, 16 (4):917 – 928. 23.TeraiK,MizukamiK,OkadaM: Comparisonofchronicrenalfailurerats andmodificationofthepreparationprotocolasahyperphosphataemia model. Nephrology(Carlton) 2008, 13 (2):139 – 146. 24.KatsumataK,KusanoK,HirataM,TsunemiK,NaganoN,BurkeSK,FukushimaN: Sevelamerhydrochloridepreventsectopiccalcificationandrenal osteodystrophyinchronicrenalfailurerats. KidneyInt 2003, 64 (2):441 – 450. 25.LeelahavanichkulA,YanQ,HuX,EisnerC,HuangY,ChenR,MizelD,Zhou H,WrightEC,KoppJB, etal : AngiotensinIIovercomesstrain-dependent resistanceofrapidCKDprogressioninanewremnantkidneymouse model. KidneyInt 2010, 78 (11):1136 – 1153. doi:10.1186/1471-2369-14-116 Citethisarticleas: Jia etal. : Anovelmodelofadenine-induced tubulointerstitialnephropathyinmice. BMCNephrology 2013 14 :116. Submit your next manuscript to BioMed Central and take full advantage of: € Convenient online submission € Thorough peer review € No space constraints or color “gure charges € Immediate publication on acceptance € Inclusion in PubMed, CAS, Scopus and Google Scholar € Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Jia etal.BMCNephrology 2013, 14 :116 Page8of8 http://www.biomedcentral.com/1471-2369/14/116