EuropeanJournalofPharmacology736(2014)86–94ContentslistsavailableatScienceDirectEuropeanJournalofPharmacologyjournalhomepage:www.elsevier.com/locate/ejpharPulmonary,gastrointestinalandurogenitalpharmacologyAstragalosideIVamelioratesdiabeticnephropathyinvolvingprotectionofpodocytesinstreptozotocininduceddiabeticratsJianguoChena,n,1,YifangChena,1,YunlingLuoa,DingkunGuib,JianhuaHuangc,DongyuanHea,naDepartmentofNephrology,ZheJiangHospital,HangZhou,PRChinaDepartmentofNephrology,ShanghaiSixthPeople'sHospital,Shanghai,PRChinacInstituteofIntegratedChineseandWesternMedicine,HuashanHospital,FudanUniversity,Shanghai,PRChinabarticleinfoArticlehistory:Received2February2014Receivedinrevisedform23April2014Accepted24April2014Availableonline6May2014Keywords:AstragalosideIVDiabeticnephropathyPodocyteIntegrin-linkedkinaseIntegrinα3β1abstractPodocytelossanddysfunctionplaykeyroleduringthedevelopmentofdiabeticnephropathy(DN).TheaimofthisstudywastoobservetheprotectiveeffectsofastragalosideIVonpodocyteindiabeticratsandexploreitsmechanismspreliminary.HealthymaleSprague-Dawley(SD)ratswererandomizedintonormalcontrolgroup,diabeticnephropathygroupanddiabeticnephropathywithAS-IVtreatmentgroup.DNwasinducedbyintraperitonealinjectionofstreptozotocin(STZ).AS-IVtreatmentstarted2weeksbeforeSTZinjectionandlasted14weeks.24hUrinaryproteinsweremeasured4,8and12weeksafterSTZinjection.Bodyweight,bloodglucose,bloodureanitrogen(BUN),creatinine(Cr),alanineaminotransferase(ALT)andaspartateaminotransferase(AST)weremeasured12weeksafterSTZinjection.Renalpathology,podocytemorphologicalchanges,podocytedensity,proteinandmRNAexpressionofintegrinα3,integrinβ1andintegrin-linkedkinase(ILK)weredetectedbyhistopathology,electronmicroscopy,immunohis-tochemistry,westernblotandreal-timePCR,respectively.Hyperglycemia,proteinuria,mesangialexpan-sionandpodocyteloss,increasedproteinexpressionofILKanddecreasedproteinexpressionofintegrinα3andintegrinβ1weredetectedindiabeticrats.AS-IVtreatmentamelioratedpodocyteloss,renalhistopathologyandpodocytefootprocesseffacement,decreasedproteinuria,partiallyrestoredproteinexpressionofintegrinα3,integrinβ1andILK.ThesefindingssuggestedthatAS-IVmayprotectpodocyteandamelioratediabeticnephropathybyinhibitingtheexpressionofILKandrestoringtheexpressionofintegrinα3β1indiabeticrats.&2014ElsevierB.V.Allrightsreserved.1.IntroductionChronickidneydisease(CKD)hasbeenrecognizedasamajorpublichealthproblemoftheworldwhichdevelopstoendstagerenaldisease(ESRD)irrespectiveofunderlyingcauses(Khwajaetal.,2007;Zhangetal.,2008).Diabeticnephropathyisaseriouscomplicationofdiabetes.TheprevalenceofdiabeticnephropathyincreasedstrikinglyinrecentdecadesandcurrentlydiabeticnephropathyhasbecometheleadingcauseoftheESRD(Andersenetal.,1983;Stephensetal.,1990).Duringtheprogres-sionofdiabeticnephropathypodocytesinjuriesplaythecentralroleinthedeteriorationofrenalfunction(Marshall,2007).Innormalconditionspodocytesanchorattheoutersideoftheglomerularbasementmembrane(GBM)byintegrinandestablishtheglomerularfiltrationbarriertogetherwithGBMandglomerularnCorrespondingauthors.E-mailaddresses:doctor_chen@126.com(J.Chen),hdy331@qq.com(D.He).1Theseauthorscontributedequallytothiswork.capillaryendothelialcells(DrumondandDeen,1994).Duringtheprocessofglomerularinjury,thefootprocessesofpodocyteretractandbroadenandthenpodocytemaydetachfromtheGBM,becausepodocytesarehighlydifferentiatedcellswithalmostnocapabilitytoundergocelldivision(Pagtalunanetal.,1997).TheconsequenceofpodocytedetachmentistheirreversiblereductionofpodocytenumberandthefailuretocovertheoutersideoftheGBMcompletelybyremainingpodocytes.GlomerularparietalepithelialcellsmayadheretothenakedareasofGBMandleadtosegmentalglomerularsclerosis(Krizetal.,1995,1998).Integrin-linkedkinaseisanintracellularserine/threoninekinasethatinteractswithmanyintegrins(Hanniganetal.,1996).ILKhasbeenreportedtobeinvolvedintheregulationofcellularactivitiesandsignalingpathwaysincludingcelladhesion(Dedharetal.,1999;Wu,1999).PreviousstudiesshowedthatILKdysre-gulationwasinvolvedinpodocyteinjuryofDN(Guoetal.,2001).Astragalusmembranaceus(Fisch)Bgeisawidelyusedherbforthetreatmentofcardiovasculardiseases,kidneydiseasesanddiabetesinChinesetraditionalmedicineforcenturies(Aietal.,2008;RiosandWaterman,1997).AS-IVisoneofthemainactiveingredientshttp://dx.doi.org/10.1016/j.ejphar.2014.04.0370014-2999/&2014ElsevierB.V.Allrightsreserved.J.Chenetal./EuropeanJournalofPharmacology736(2014)86–9487ofAstragalusmembranaceus(Fisch)Bgeandhasbeenreportedtohavemanypharmacologicalactivities.3-0-Beta-D-xylopyranosyl-6-0-beta-D-glucopyranosylcycloastra-genolisthechemicalnameofAS-IV.AS-IVcouldsynergizewithferulicacidtoinhibitrenaltubulointerstitialfibrosisinratswithobstructivenephropathy(Mengetal.,2011)andreduceischemicacutekidneyinjuryinrats(Tanetal.,2013).OurpreviousstudyshowedthatAS-IVcouldimprovehighglucose-inducedcelladhesiondysfunctionincul-turedmousepodocytes(Chenetal.,2008).InthisstudywewillobservetheeffectofAS-IVonexpressionofα3β1integrinandILKandinvestigatetheircorrelationswithdiabeticpodocyteinjuryinvivo.2.Materialsandmethods2.1.DrugpreparationAstragalosideIV(C41H68O14,molecularweight¼784,CASno.84687-43-4)purchasedfromXi'anSobeoPharmaceuticalTechnol-ogyCompany,Limited(Xi'an,China)wassuspendedin1%carbox-ymethylcellulose(CMC)solutionandwasgiventothediabeticratsbyoralgavagewith3differentdosagesasdescribedbefore(Guietal.,2012).2.2.AnimalstudyandexperimentaldesignHealthymaleSprague-Dawley(SD)ratsweighing180–200gwerepurchasedfromExperimentalAnimalCenter,ZhejiangUni-versity,Zhejiangprovince,China.Theywerehousedinaroomwithair-conditionedtemperatureat2371Candalternating12hcyclesoflightanddark.Animalswerefedwithstandarddietandfreetowater.Theratswererandomlydividedintofivegroups(n¼8/eachgroup):(1)normalcontrolrats(NC),(2)diabeticrats(DN),(3)diabeticratstreatedwithlowdoseofAS-IVat2.5mg/kg/d(DNþAL),(4)diabeticratstreatedwithmoderatedoseofAS-IVat5mg/kg/d(DNþAM)and(5)diabeticratstreatedwithhighdoseofAS-IVat10mg/kg/d(DNþAH).Diabetesmellituswasinducedbysingleintraperitonealinjectionofstreptozotocin(STZ)dilutedwith0.1Mcitratebuffer(pH4.5).TheSTZdosagewas65mg/kg.Normalcontrolratswereintraperitonealinjectedwithequalvolumeofvehicle.FortyeighthoursafterSTZinjection,tailveinbloodglucosewasmeasured.Ratswithbloodglucosebeyond300mg/dlwereconsideredasdiabeticrats.AS-IVtreatmentstarted2weeksbeforeSTZinjectionandlasted14weeks.RatsofAS-IVtreatmentgroupsweregivenAS-IVsuspendedin1%CMCbyoralgavageoncedaily.NormalcontrolratsweregiventhesamevolumeofCMC.Attheendof4,8and12weeksafterSTZinjection,24hurineofeachgroupwerecollected,centrifugedat800gfor10minat251CandstoredatÀ801C.Urinaryproteinwasassayedbypyrogallolredcolorimetricassaykitaccordingtotheinstruc-tionsofthemanufacturer(NanjingJianchengBioengineeringInstitute,Nanjing,China).Attheendof12thweekafterSTZinjectionratswereweighedandthensacrificed.Bloodsampleswerecollectedfromtheabdominalaorta.Bloodglucose,bloodureanitrogen(BUN),creatinine(Cr),alanineaminotransferase(ALT)andaspartateaminotransferase(AST)weremeasuredbyautomaticbiochemistryanalyzer(HitachiModel7600,Japan).Thekidneyswerecollectedandcutintopiecesforhistopathology,electronmicroscopy,immunohistochemistry,westernblotandreal-timePCRanalysis.Alltheworkwasperformedaccordingtothe“GuidefortheCareandUseofLaboratoryAnimals”publishedbytheZhejiangUniversityandwasapprovedbytheAnimalEthicsCommitteeofZhejiangUniversity,Zhejiangprovince,China.2.3.HistologyThekidneyswerefixedwith10%neutralbufferedformalinandembeddedinparaffin,cutinto4μmsectionsforhematoxylinandeosin,PeriodicAcid-SchiffandMassonstaining.MesangialmatrixexpansionintheglomeruliwasevaluatedinPAS-stainedsectionsusingImage-ProPlus4.5(Soetal.,2013).ThemeanpercentareaofPAS-stainedglomeruliwascalculatedfor20randomlyselectedfieldsofeachkidneysection.Theglomerulosclerosisineachglomeruluswasscoredsemi-quantitativelyasfollows:0,nosclerosis;1,sclerosisino25%ofglomerulus;2,sclerosisin25–50%ofglomerulus;and3,sclerosisin450%ofglomerulus(Fujiharaetal.,2000).Toevaluateinterstitialfibrosis,20fieldsforeachsectionwereassessedonMasson-stainedsections.Semi-quantitativeanalysisineachfieldwasassessedasfollows:0,nofibrosis;1,fibrosislessthan10%ofareas;2,fibrosis10%toapproximately25%ofareas;3,fibrosis25%toapproximately50%ofareas;and4,fibrosismorethan50%ofareas.Theaveragesofinterstitialfibrosisscoreswerecalculatedfromthetotalevaluatedinterstitiallesionsineachsection.Thepathologicchangeswereassessedbyarenalpathologistwhowasblindedtothisstudy.2.4.ImmunohistochemistryWilm'sTumor1protein(WT1)isacharacteristicproteinofpodocyteandisnecessaryforpodocytematuration.InscientificresearchesWT1iscommonlyusedasmolecularmarkerofpodocyte(MichaudandKennedy,2007;Suetal.,2010).InthisstudyweusedWT1fortheanalysisofpodocytesdensity(podo-cytenumbersperglomerulus)byimmunohistochemistryunder4μmparaffin-embeddedsections.Rabbitanti-WT1antibody(SantaCruz,USA)wasusedasthefirstantibody.Thecolorwasvisualizedbydiaminobenzidineandcounterstainedwithhema-toxylin.TheWT1signalwasquantifiedbylightmicroscopewithimageanalyzer.PodocytedensitywasexpressedasapercentageofWT1immunostainedareaoccupiedbytotalglomerulararea.Twentyconsecutiveglomerularsectionsandanaverageof20glomeruliperratwereobserved.2.5.ElectronmicroscopystudiesElectronmicroscopicsectionswereperformedbyroutineprocedures.Renalcortexwascutintopiecesonice,fixedwith2.5%glutaraldehydedissolvedin0.1Msodiumcacodylate(pH7.4)at41Covernightandwashedinthesamebuffer.Thetissuefragmentswerepostfixedin1%cacodylate-bufferedOsO4for2h,dehydrated,andembeddedinEpon.Ultrathinsectionswerestainedwithuranylacetateandleadcitrateandexaminedbyelectronmicroscopy.ThenumberofpodocytefootprocessespresentineachmicrographwasdividedbythetotallengthofGBMregionsineachimagetodeterminetheaveragedensityofpodocytefootprocesses.Theelectronmicroscopephotoswereevaluatedinablindfashion.2.6.WesternblottingKidneycortexwashomogenizedinlysisbufferonicewithahomogenizer.Thesupernatantswerecollectedaftercentrifugingat10,000rpmfor5minat41C.Proteinconcentrationofthesupernatantswasmeasuredbythebicinchoninicacid(BCA)proteinassay(Pierce,Rockford,IL,USA).Thewholetissuelysatesweremixedwithequalamountof2ÂSDSloadingbuffer(125mmol/lTris–HCl,4%SDS,20%glycerol,100mmol/ldithio-threitol,and0.2%bromphenolblue).Sampleswereseparatedby10%sodiumdodecylsulfate(SDS)/polyacrylamidegelelectrophor-esisandelectro-transferredtoapolyvinylidenedifluoride(PVDF)88J.Chenetal./EuropeanJournalofPharmacology736(2014)86–94membrane(ImmobilonP,Millipore,Bedford,Mass,USA).Non-specificbindingsiteswereblockedwith2%non-fatmilkTris-bufferedsalineTween(TBSand0.1%Tween20)atroomtempera-turefor1h.Thenmembraneswereincubatedwithvariousprimaryantibodiesovernightat41C.Afterwashing3timeswithTBST,membraneswereincubatedwithhorseradishperoxidase-conjugatedsecondaryantibodiesfor1hatroomtemperature.ProteinbandswerevisualizedbyECLPlus(Amersham,ArlingtonHeights,IL,USA).OpticaldensityofthebandswasmeasuredbyaBio-Radgelimagingsystem.2.7.Real-timequantitativePCRTotalRNAofkidneycortexwasextractedbytheTrizolproce-dureaccordingtotheinstructionsofthemanufacturer(Invitrogen,Carlsbad,CA,USA).RNAwasreversetranscribedusingtheSuper-ScriptRTkit(InvitrogenCarlsbad,USA).Real-timequantitativePCRwasperformedbytheABIPRISM7900SequenceDetectionSystem(AppliedBiosystems).Theprimerconcentrationsweredeterminedbyanalyzingtheoptimalconcentrationsofeachprimer.ThesequencesofprimerpairsusedforPCRswereasfollows:ILKsense50ATTTCGTTGTGGACCAGAGC30andanti-sense50TGCGTGCCGAGGTATGAG30;integrinα3,sense50CCCCTCGC-TTTGTACGGTTA30andanti-sense50TGTCCCTGTCAGCCTCCACT30;integrinβ1sense50GTTCCATGCGTAGCGACAA30andanti-sense50TTCTCCCTGCTTTCCACTTTAG30.Toconfirmtheamplificationspe-cificity,PCRproductsofeachprimerpairweresubjectedtoameltingcurveanalysis.Eachreactionwasrepeated3timesandratioresultswerecalculatedbythe2ÀΔΔCTmethoddescribedpreviously(LivakandSchmittgen,2001).ThemRNAlevelswerenormalizedtoGAPDH.2.8.StatisticalanalysisStatisticswereconductedbysigmaplot12.5software.Alldatawereexpressedasmean7standarddeviation(S.D.)andanalyzedbyone-wayortwo-wayANOVA.MultiplecomparisonbetweenthegroupswasperformedusingtheS–N–Kmethod.Po0.05wasconsideredstatisticallysignificant.amelioratedproteinuriaofdiabeticratssignificantlyinadosedependentway(Fig.1).AsshowninFig.2,thediabeticratsexhibitedfocalmesangialmatrixexpansion,partialglomerulo-sclerosisandinterstitialfibrosisat12weeksafterSTZinjection.AS-IVtreatmentamelioratedmesangialexpansion,glomerulo-sclerosisandinterstitialfibrosiscomparedwiththeuntreateddiabeticrats.Markedpodocytefootprocesseffacementwasobservedbyelectronmicroscopyindiabeticrats12weeksafterSTZinjection,whereastheratspretreatedwithAS-IVshowedimprovementinpodocytefootprocesseffacement.AS-IVamelio-ratedpodocytefootprocesseffacementofdiabeticratsinadose-dependentway(Fig.3).3.2.AS-IVdecreasedpodocytelossinSTZ-induceddiabeticratsThepodocytedensitywasobtainedbycountingthenumberofWT-1-positivecellsofeachglomerular.InnormalcontrolgroupWT-1positivecellsarebroadlylocatedalongtheglomerularstructure.InDNkidneys,thedensityofWT-1positivecellsdecreased,whichmeantthatthepodocytenumberperglomerulardecreased.AS-IVtreatmentcandosedependentlyincreasethepodocytedensityinSTZ-induceddiabeticrats(Fig.4).3.3.AS-IVregulatedtheproteinandmRNAexpressionofintegrinα3andβ1subunitsandintegrin-linkedkinaseinSTZ-induceddiabeticratsIntegrinisanimportantadhesionproteinwhichplaysakeyroleincelladhesiontoextracellularmatrix(ECM)andsignaltransduction(AlbeldaandBuck,1990;Hynes,1992;JulianoandHaskill,1993).PodocyteisattachedtoGBMbyintegrinα3β1(Chenetal.,2000).AsshowninFig.5,theproteinexpressionofintegrinα3andintegrinβ1subunitsdecreasedinSTZinduceddiabeticrats,whiletheILKproteinexpressionincreased.AS-IVtreatmentdose-dependentlyincreasedtheintegrinα3andintegrinβ1subunitsproteinexpressionanddecreasedILKproteinexpression.ThesefindingssuggestedthatthetherapeuticeffectofAS-IVwasassociatedwithregulationoftheintegrinα3β1andILKproteinexpression.WealsoobservedthemRNAchangesinILK,integrinα3andintegrinβ1subunitsofeachgroup.ThemRNAexpressionofILK,integrinα3andintegrinβ1wasnotconsistentwiththeirproteinexpression(Fig.6).TheILKmRNAexpressionhadnosignificantdifferencebetweennormalcontrol,DNandDNwithAS-IVtreatmentgroups.Comparedwithnormalcontrolgroupintegrinα3mRNAexpressiondecreasedsignificantlyinDNgroupattheendof12weeksandAS-IVtreatmentdidnotincreasetheintegrinα3mRNAexpressionsignificantly.Integrinβ1mRNAexpressiondown-regulatedsignificantlyinDNratscomparedwith3.Results3.1.AS-IVsignificantlyamelioratedproteinuria,podocytefootprocesseffacementandrenalhistopathologyinSTZ-induceddiabeticratsComparedwithnormalcontrolgroupsevereproteinuriawasobservedinSTZ-induceddiabeticrats(Po0.05).AS-IVtreatmentFig.1.AS-IVdose-dependentlyamelioratedproteinuriaofdiabeticrats.EffectsofAS-IVonproteinuriaindiabeticratsat4weeks(A),8weeks(B)and12weeks(C)afterSTZinjection.NC,normalcontrolrats;DN,STZ-induceddiabeticrats;DNþAL,DNratstreatedwithlowdoseAS-IV(2.5mg/kg/d);DNþAM,DNratstreatedwithmediumdoseAS-IV(5mg/kg/d);andDNþAH,DNratstreatedwithhighdoseAS-IV(10mg/kg/d).Resultswereexpressedasthemeans7S.D.nPo0.05vs.NC.♯Po0.05vs.DN.J.Chenetal./EuropeanJournalofPharmacology736(2014)86–94Fig.2.Representativehistologicalchangesinkidneysofeachgroupofrats.(A–E)Stainedwithhematoxylinandeosin.(F,G)StainedwithPAS.(K–O)StainedwithMasson.(P)Mesangialmatrixexpansionanalysisofeachgroupofrats.(Q)Semiquantitativeanalysisofglomerulosclerosisofeachgroupofrats.(R)Semiquantitativeanalysisofinterstitialfibrosisofeachgroupofrats.AS-IVamelioratedfocalmesangialmatrixexpansion,glomerulosclerosisandinterstitialfibrosisinSTZinduceddiabeticratsindosedependentway.AS-IVtreatmentwasstarted2weeksbeforeSTZinjectionandlasted14weeks.NC,normalcontrolrats;DN,STZ-induceddiabeticrats;DNþAL,DNratstreatedwithlowdoseAS-IV(2.5mg/kg/d);DNþAM,DNratstreatedwithmediumdoseAS-IV(5mg/kg/d);andDNþAH,DNratstreatedwithhighdoseAS-IV(10mg/kg/d).Resultswereexpressedasthemeans7S.D.nPo0.05vs.NC.♯Po0.05vs.DN.normalcontrolrats.AfterAS-IVtreatmentintegrinβ1mRNAexpressiondidhavesomechanges,butthechangeswerenotdosedependentandnosignificantdifferenceswereobtainedcomparedwithDNgroup.Nosignificantdifferenceswereobtainedamongthreedifferentdosetreatmentgroupseither.4.DiscussionDiabeticnephropathy(DN)isanimportantcomplicationoftype1ortype2diabetes,whichischaracterizedbycontinuousalbuminuria,decreasedglomerularfiltrationrateandenhancedcardiovascularmorbidityandmortality(Maueretal.,1981).From1980to2006,thepercentageofdiabeticnephropathyinpatientsstartingdialysishasincreasedfromnear0%to45%.AlthoughcomparedwiththeprecedingyeartheincidenceofESRDcausedbydiabetesdeclinedin2008;diabeticnephropathyremainstheleadingcauseofESRDintheUnitedStatesandotherdevelopedcountries(Collinsetal.,2012;Ritzetal.,1999).Thepathologicalcharacteristicsofdiabeticnephropathyareglomerularandtubularbasementmembranethickening,mesangialandinterstitialexpan-sionwithincreasedextracellularmatrix(Maueretal.,1981).Inourstudy,diabeticratsdevelopedapparentproteinuria.The24hurinaryproteinincreasedsignificantly4weeksafterSTZinjectioncomparedwithnormalcontrolrats.Attheendof8weeksand12weeksafterSTZinjection,24hurinaryproteinincreasedfurther.AS-IVtreatmentamelioratedurinary24hurinaryproteininadosedependentway.Theproteinurialevelparalleledwithrenalpatho-logicalchanges.At12weeksafterSTZinjection,thediabeticratsshowedfocalmesangialmatrixexpansion,partialglomerulo-sclerosisandinterstitialfibrosiscomparedwithnormalcontrol3.4.EffectsofAS-IVonbodyweightandmetabolicparametersinbloodBodyweight,bloodglucose(BG),bloodureanitrogen(BUN),serumcreatinine(SCr),alanineaminotransferase(ALT)andaspar-tateaminotransferase(AST)weremeasuredandcomparedbetweeneachgroupattheendof12weeksafterSTZinjection(Table1).Comparedwithnormalcontrolratsbodyweightofdiabeticratsdecreasedwhilethebloodglucoseincreased.Thedifferenceswerestatisticallysignificant(Po0.05).Butnodiffer-encesofBGandbodyweightweredetectedbetweenAS-IVtreatedanduntreateddiabeticrats.Andtherewerenodifferenceofbloodureanitrogen,serumcreatinine(SCr)andaminotransferase(ASTandALT)betweenAS-IVtreatedanduntreateddiabeticrats,probablyindicatingthattheprotectiveeffectsofAS-IVhadnothingtodowithglucoseregulationandAS-IVmayhavenorenalandhepatictoxicity.90J.Chenetal./EuropeanJournalofPharmacology736(2014)86–94Fig.3.Representativeelectronphotomicrographsfromeachgroupofrats.Kidneyofnormalcontrolrats(A),DNrats(B)andDNratstreatedwiththreedosesAS-IV(C–E).(F)Quantitativeanalysisofdensityofpodocytefootprocesses.AS-IVamelioratedpodocytefootprocesseffacementdose-dependently.NC,normalcontrolrats;DN,STZ-induceddiabeticrats;DNþAL,DNratstreatedwithlowdoseAS-IV(2.5mg/kg/d);DNþAM,DNratstreatedwithmediumdoseAS-IV(5mg/kg/d);andDNþAH,DNratstreatedwithhighdoseAS-IV(10mg/kg/d).Resultswereexpressedasthemeans7S.D.nPo0.05vs.NC.♯Po0.05vs.DN.ratsandpretreatmentwithAS-IVamelioratedthesepathologicalchangescomparedwiththeuntreateddiabeticrats.Podocyte,alsocalledglomerularvisceralepithelialcell,consistsofthreemajorcomponents:themaincellbody,cytoplasmicprocessesandfootprocessesthatdirectlycontactwiththeGBM.Podocyteisthemostimportantpartoftheglomerularfiltrationbarrierandthetargetcellintheglomerulusdiseases(DrumondandDeen,1994).Podocyteinjuriesplaythecentralroleinalmostalltheproteinuriadiseasesincludingdiabeticnephropathy(GrekaandMundel,2012).Indiseasesassociatedwithproteinuria,thefootprocessesofpodocyteeffaced.AsshowninFig.3,podocytefootprocesseffacementdevelopedindiabeticratsandAS-IVtreatmentreducedpodocytefootprocesseffacementindiabeticrats.Ifinjuriescontinueorstrengthen,detachmentfromGBMorpodocyteapoptosismayoccur(Haraetal.,1995,1998;Kerjaschkietal.,1986).Apoptosiswasonceconsideredtobethekeymechanismofpodocytelossduringproteinurianephropathy(Riedletal.,2011;Sohnetal.,2010;Xuetal.,2012).ButstudiesshowedthatdetachmentofpodocytefromGBMmaybemoreimportant.Lemleyetal.foundthatthepodocytesexcretedintotheurinearealive(Lemley,2008).Petermannetal.confirmedthatthepodocytesexcretedintotheurinehavetheadherentcapacity(Petermannetal.,2003).Podocyteistheterminaldifferentiatedcellswithlittlecapabilitytoundergocelldivision.PodocytedetachmentfromGBMandapoptosisleadtoirreversiblepodocytedecrease,filtrationbarrierinjuriesandproteinuria.Afterpodo-cyteslossparietalepithelialcellscouldadherewiththenakedareaofGBM,whicheventuallyleadtoglomerulosclerosis(Komai-KomaJ.Chenetal./EuropeanJournalofPharmacology736(2014)86–9491Fig.4.PodocytesdetectedbyimmunohistochemistrystainingofWT-1expressionintheglomeruliattheendof12weeksafterSTZinjection.WT-1wasstainedbrown.(A)NC,normalcontrolrats;(B)DN,STZ-induceddiabeticrats;(C)DNþAL,DNratstreatedwithlowdoseAS-IV(2.5mg/kg/d);(D)DNþAM,DNratstreatedwithmediumdoseAS-IV(5mg/kg/d);and(E)DNþAH,DNratstreatedwithhighdoseAS-IV(10mg/kg/d).(F)Semi-quantitativeanalysisofpodocytedensityofeachgroupofrats.Semiquantitativeresultswereexpressedasthemeans7S.D.nPo0.05vs.NC.♯Po0.05vs.DN.Fig.5.AS-IVregulatedtheproteinexpressionofILK,integrinα3andintegrinβ1.(A)TheproteinexpressionofILK,integrinα3andintegrinβ1examinedbyWesternBlot.Semi-quantitativeanalysisof(B)integrinα3,(C)integrinβ1,and(D)ILK.NC,normalcontrolrats;DN,STZinduceddiabeticrats;DNþAL,DNratstreatedwithlowdoseAS-IV(2.5mg/kg/d);DNþAM,DNratstreatedwithmediumdoseAS-IV(5mg/kg/d);andDNþAH,DNratstreatedwithhighdoseAS-IV(10mg/kg/d).Resultsareexpressedarethemeans7S.D.nPo0.05vs.NC.♯Po0.05vs.DN.etal.,2004).Previousstudiesshowedthatpodocytedetachmentcouldbedetected1monthafterhyperglycemia,whichsuggestthatpodocytelossmaybeanearlyphenomenonindiabetes(RegoliandBendayan,1997).Podocytedensity(number/volume)reductionhasbeenreportedinbothtypes1and2diabetes(Pagtalunanetal.,1997;WhiteandBilous,2004;Whiteetal.,92J.Chenetal./EuropeanJournalofPharmacology736(2014)86–94Fig.6.EffectsofAS-IVonmRNAexpressionofILK,integrinα3andintegrinβ1indiabeticratsat12weeksafterSTZinjection.ThemRNAexpressionof(A)ILK,(B)integrinα3,and(C)integrinβ1.NC,normalcontrolrats;DN,STZ-induceddiabeticrats;DNþAL,DNratstreatedwithlowdoseAS-IV(2.5mg/kg/d);DNþAM,DNratstreatedwithmediumdoseAS-IV(5mg/kg/d);andDNþAH,DNratstreatedwithhighdoseAS-IV(10mg/kg/d).Resultsareexpressedasthemeans7S.D.nPo0.05vs.NC.Table1Physicalandplasmabiochemicalparameters.NCBW(g)BG(mmol/L)BUN(mmol/L)SCr(μmol/L)ALT(U/L)AST(U/L)534.2735.88.070.76.6271.2622.78714.0560.575.572.276.3DN235.2722.7a36.378.7a11.6376.0625.2678.37.473.176.876.2DNþAL252.1734.3a36.576.9a10.974.324.5572.6267.875.973.174.8DNþAM243.0729.6a37.174.8a10.773.222.3072.0760.872.973.573.6DNþAH249.4732.4a32.474.1a7.7871.4724.7578.2265.474.375.773.7BW,bodyweight;BG,bloodglucose;BUN,bloodureanitrogen;SCr,serumcreatinine;ALT,alanineaminotransferase.AS-IVtreatmentwasstarted2weeksbeforeSTZinjectionandlasted14weeks.NC,normalcontrolrats;DN,STZ-induceddiabeticrats;DNþAL,DNratstreatedwithlowdoseAS-IV(2.5mg/kg/d);DNþAM,DNratstreatedwithmediumdoseAS-IV(5mg/kg/d);DNþAH,DNratstreatedwithhighdoseAS-IV(10mg/kg/d).Resultsareexpressedasthemeans7S.D.aPo0.05vs.NCgroup.2002).ThereforepreventingpodocytedetachmentfromGBMmaybeapromisingtargetforDNtreatment.CurrentlytherearenointerventionsspecificallypreventingpodocytedetachmentinDN.Ourresultswereconsistentwithpreviousstudies.Byimmunohis-tochemistrystainingofpodocytespecificbio-markWT1,wefoundpodocytenumberdecreasedindiabeticratscomparedwithnormalcontrolrats.AS-IVtreatmentcouldreducepodocytelossindiabeticratsinducedbySTZinjection.DecreasingofcelladhesionandcytoskeletonchangesarenecessaryforthedetachmentofpodocytefromGBM.Integrinandintegrin-linkedkinase(ILK)arereportedtobeinvolvedintheseprocesses(WuandDedhar,2001).Integrinsareafamilyofcellsurfaceproteinswhichmediatecelltocellandcelltoextra-cellularmatrix(ECM)adhesion.Integrinheterodimersconsistedofαandβsubunits;eachαsubunitcanassociatewithseveralβsubunitsandviceversa.Someintegrinheterodimersalsohavemorethanoneligand.Tillnow14αand8βsubunitshavebeenidentified(Hynes,1992;LivakandSchmittgen,2001).Integrinservesascellanchorandsignaltransmitter(Ruoslahti,1991).IntegrinintegratescytoskeletonwithECM(Horwitzetal.,1986)andplaysakeyroleincellaggregation(Phillipsetal.,1988),immuneresponse(Springer,1990),tissuerepairing(Todaetal.,1987)andtumorinvasion(RuoslahtiandGiancotti,19).Podo-cytesconnectwithGBMbyintegrinα3β1.Decreasedexpressionofα3β1integrinonpodocyteshasbeenreportedinpatientsofprimaryFSGS,diabetes,andSTZ-induceddiabeticrats(Chenetal.,2000).Reductionofintegrinα3β1onpodocytecouldbeinducedbyhighglucoseandwasaccompaniedbythedecreasedadhesionofpodocytetoGBMcomponent(Kitsiouetal.,2003).Theseresultssuggestthatintegrinα3β1reductionleadstothepodocytedetachmentanddysfunctionoftheglomerularfiltrationbarrierinDN;restorationofintegrinα3β1expressionmayreducethepodocytelossandproteinuria.Thishypothesisisconsistentwithourresults.Thediabeticratshaddecreasedproteinexpressionofintegrinα3andintegrinβ1,lownumericaldensityofpodocyteandapparentproteinuriacomparedwithnormalcontrolgroup.AS-IVtreatmentrestoredintegrinα3andintegrinβ1expressions,increasednumericaldensityofpodocyteandfinallydecreasedproteinuria.ILKisaserine/threoninekinasewhichcaninteractwithintegrinβ1,integrinβ3(Hanniganetal.,1996)andcytoskeleton-associatedproteinssuchasPINCH(Tuetal.,1999),CH-ILKBP(Tuetal.,2001)andpaxillin(NikolopoulosandTurner,2001).ILKhasbeenreportedtobeinvolvedintheregulationofintegrin-mediatedprocessesincludingcelladhesion,cell-shapechangesanddepositionofextracellularmatrix(Dedharetal.,1999;Wu,1999).TheproteinexpressionofILKincreasedstrikinglyindiabeticpatients’kidneys,especiallyinmesangialcellsandepithelialcellsofglomeruli(Guoetal.,2001).IncreasedILKlevelhasalsobeenreportedindiabeticrats(DAIetal.,2012).UnderhighglucoseenvironmentILKactivationandup-regulationleadtothereducedintegrinα3β1expressionandimpairedcell–matrixadhesionofpodocyte(Chenetal.,2008).ILKinhibitioncouldrestoreintegrinα3β1expressionandcell–matrixadhesionofpodocyte(Teixeiraetal.,2005).Inourstudytheproteinexpres-sionofILKincreasedinDNrats.AS-IVtherapyinhibitedtheup-regulationofILK.WealsoobservedthemRNAexpressionofILK,integrinα3andintegrinβ1byreal-timePCR.ThemRNAexpres-sionsofILK,integrinα3andintegrinβ1werenotcorrelatedwiththeirproteins.TheseresultssuggestthatthemechanismsofAS-IVregulatingtheproteinexpressionofintegrinα3β1andILKinDNratsarenotsimplybyup-regulatingtheexpressionofmRNA.Itmayinvolvesomepost-transcriptionalregulationandmoreexperimentsareneededinthefuture.Finallyweexaminedthebloodmetabolicparametersofeachgrouprats.Therewerenostatisticaldifferencesofbloodglucose,BUN,Cr,ASTandALTbetweendiabeticratswithorwithoutAS-IVtreatment.SotheprotectiveeffectsofAS-IVonpodocytewereJ.Chenetal./EuropeanJournalofPharmacology736(2014)86–9493probablyindependentofbloodglucoselevelandAS-IVmayhavenorenalandhepatictoxicityinrats.5.ConclusionsInsummary,AS-IVimprovescell–matrixadhesionofpodocytestoGBM,amelioratesrenalhistopathology,podocytefootprocesseffacementandproteinuria.Thiseffectmaybeinvolvedinintegrinα3β1up-regulationandILKinhibition.SourcesoffundingThisstudywassupportedbyagrantfromtheHealthResearchFunds,MinistryofHealth,PRChina(WKJ2009-2-017)andagrantfromtheNaturalScienceFoundationofZhejiangProvince(Approvalno.LY12H05001).ReferencesAi,P.,Yong,G.,Dingkun,G.,Qiuyu,Z.,Kaiyuan,Z.,Shanyan,L.,2008.AqueousextractofAstragaliRadixinduceshumannatriuresisthroughenhancementofrenalresponsetoatrialnatriureticpeptide.J.Ethnopharmacol.116,413–421.Albelda,S.M.,Buck,C.A.,1990.Integrinsandothercelladhesionmolecules.FASEBJ.4,2868–2880.Andersen,A.,Christiansen,J.S.,Andersen,J.,Kreiner,S.,Deckert,T.,1983.Diabeticnephropathyintype1(insulin-dependent)diabetes:anepidemiologicalstudy.Diabetologia25,496–501.Chen,H.-C.,Chen,C.-A.,Guh,J.-Y.,Chang,J.-M.,Shin,S.-J.,Lai,Y.-H.,2000.Alteringexpressionofα3β1integrinonpodocytesofhumanandratswithdiabetes.LifeSci.67,2345–2353.Chen,J.,Gui,D.,Chen,Y.,Mou,L.,Liu,Y.,Huang,J.,2008.AstragalosideIVimproveshighglucose-inducedpodocyteadhesiondysfunctionviaα3β1integrinupre-gulationandintegrin-linkedkinaseinhibition.Biochem.Pharmacol.76,796–804.Collins,A.J.,Foley,R.N.,Chavers,B.,Gilbertson,D.,Herzog,C.,Johansen,K.,Kasiske,B.,Kutner,N.,Liu,J.,St.Peter,W.,2012.UnitedStatesRenalDataSystem2011annualdatareport:atlasofchronickidneydisease&end-stagerenaldiseaseintheUnitedStates.Am.J.KidneyDis.:Off.J.Natl.KidneyFound.59(A7),e1.Dai,H.-Y.,Zheng,M.,Tang,R.-N.,Ma,K.-L.,Ni,J.,Liu,B.-C.,2012.Inhibitionofintegrin-linkedkinasebyangiotensinIIreceptorantagonist,irbesartanattenu-atespodocyteinjuryindiabeticrats.Chin.Med.J.125,888–3.Dedhar,S.,Williams,B.,Hannigan,G.,1999.Integrin-linkedkinase(ILK):aregulatorofintegrinandgrowth-factorsignalling.TrendsCellBiol.9,319–323.Drumond,M.C.,Deen,W.M.,1994.Structuraldeterminantsofglomerularhydraulicpermeability.Am.J.Physiol.-RenalPhysiol.266,F1–F12.Fujihara,C.K.,Noronha,I.D.L.,Malheiros,D.M.A.C.,Antunes,G.R.,deOliveira,I.B.,Zatz,R.,2000.Combinedmycophenolatemofetilandlosartantherapyarrestsestablishedinjuryintheremnantkidney.J.Am.Soc.Nephrol.11,283–290.Greka,A.,Mundel,P.,2012.Cellbiologyandpathologyofpodocytes.Annu.Rev.Physiol.74,299.Gui,D.,Guo,Y.,Wang,F.,Liu,W.,Chen,J.,Chen,Y.,Huang,J.,Wang,N.,2012.AstragalosideIV,anovelantioxidant,preventsglucose-inducedpodocyteapoptosisinvitroandinvivo.PLoSOne7,e39824.Guo,L.,Sanders,P.W.,Woods,A.,Wu,C.,2001.Thedistributionandregulationofintegrin-linkedkinaseinnormalanddiabetickidneys.Am.J.Pathol.159,1735–1742.Hannigan,G.,Leung-Hagesteijn,C.,Fitz-Gibbon,L.,Coppolino,M.,Dedhar,S.,Hannigan,G.,Dedhar,S.,Hannigan,G.,1996.Regulationofcelladhesionandanchorage-dependentgrowthbyanew.Nature379,91–96.Hara,M.,Yamamoto,T.,Yanagihara,T.,Takada,T.,Itoh,M.,Adachi,Y.,Yoshizumi,A.,Kawasaki,K.,Kihara,I.,1995.Urinaryexcretionofpodocalyxinindicatesglomerularepithelialcellinjuriesinglomerulonephritis.Nephron69,397–403.Hara,M.,Yanagihara,T.,Itoh,M.,Matsuno,M.,Kihara,I.,1998.Immunohistochem-icalandurinarymarkersofpodocyteinjury.Pediatr.Nephrol.12,43–48.Horwitz,A.,Duggan,K.,Buck,C.,Beckerle,M.C.,Burridge,K.,1986.Interactionofplasmamembranefibronectinreceptorwithtalin—atransmembranelinkage.Nature320,531–533.Hynes,R.O.,1992.Integrins:versatility,modulation,andsignalingincelladhesion.Cell69,11–25.Juliano,R.,Haskill,S.,1993.Signaltransductionfromtheextracellularmatrix.J.CellBiol.120,577–585.Kerjaschki,D.,Poczewski,H.,Dekan,G.,Horvat,R.,Balzar,E.,Kraft,N.,Atkins,R.,1986.Identificationofamajorsialoproteinintheglycocalyxofhumanvisceralglomerularepithelialcells.J.Clin.Investig.78,1142.Khwaja,A.,ElKossi,M.,Floege,J.,ElNahas,M.,2007.ThemanagementofCKD:alookintothefuture.KidneyInt.72,1316–1323.Kitsiou,P.V.,Tzinia,A.K.,Stetler-Stevenson,W.G.,Michael,A.F.,Fan,W.-W.,Zhou,B.,Tsilibary,E.C.,2003.Glucose-inducedchangesinintegrinsandmatrix-relatedfunctionsinculturedhumanglomerularepithelialcells.Am.J.Physiol.-RenalPhysiol.284,F671–F679.Komai-Koma,M.,Jones,L.,Ogg,G.S.,Xu,D.,Liew,F.Y.,2004.TLR2isexpressedonactivatedTcellsasacostimulatoryreceptor.Proc.Natl.Acad.Sci.USA101,3029–3034.Kriz,W.,Elger,M.,Mundel,P.,Lemley,K.V.,1995.Structure-stabilizingforcesintheglomerulartuft.J.Am.Soc.Nephrol.5,1731–1739.Kriz,W.,Gretz,N.,Lemley,K.V.,1998.Progressionofglomerulardiseases:isthepodocytetheculprit?KidneyInt.54,687–697.Lemley,K.V.,2008.Diabetesandchronickidneydisease:lessonsfromthePimaIndians.Pediatr.Nephrol.23,1933–1940.Livak,K.J.,Schmittgen,T.D.,2001.Analysisofrelativegeneexpressiondatausingreal-timequantitativePCRandthe2ÀΔΔCTmethodMethods25,402–408.Marshall,S.M.,2007.Thepodocyte:apotentialtherapeutictargetindiabeticnephropathy?Curr.Pharm.Des.13,2713–2720.Mauer,S.M.,Steffes,M.W.,Brown,D.M.,1981.Thekidneyindiabetes.Am.J.Med.70,603–612.Meng,L.,Tang,J.,Wang,Y.,Zhao,J.,Shang,M.,Zhang,M.,Liu,S.,Qu,L.,Cai,S.,Li,X.,2011.AstragalosideIVsynergizeswithferulicacidtoinhibitrenaltubulointer-stitialfibrosisinratswithobstructivenephropathy.Br.J.Pharmacol.162,1805–1818.Michaud,J.,Kennedy,C.,2007.Thepodocyteinhealthanddisease:insightsfromthemouse.Clin.Sci.112,325–335.Nikolopoulos,S.N.,Turner,C.E.,2001.Integrin-linkedkinase(ILK)bindingtopaxillinLD1motifregulatesILKlocalizationtofocaladhesions.J.Biol.Chem.276,23499–23505.Pagtalunan,M.E.,Miller,P.L.,Jumping-Eagle,S.,Nelson,R.G.,Myers,B.D.,Rennke,H.G.,Coplon,N.S.,Sun,L.,Meyer,T.W.,1997.PodocytelossandprogressiveglomerularinjuryintypeIIdiabetes.J.Clin.Investig.99,342.Petermann,A.T.,Krofft,R.,Blonski,M.,Hiromura,K.,Vaughn,M.,Pichler,R.,Griffin,S.,Wada,T.,Pippin,J.,Durvasula,R.,2003.Podocytesthatdetachinexperi-mentalmembranousnephropathyareviable1.KidneyInt.,1222–1231.Phillips,D.R.,Charo,I.F.,Parise,L.V.,Fitzgerald,L.,1988.Theplateletmembraneglycoproteinlib–Iliacomplex.Blood71,831–843.Regoli,M.,Bendayan,M.,1997.Alterationsintheexpressionoftheα3β1integrinincertainmembranedomainsoftheglomerularepithelialcells(podocytes)indiabetesmellitus.Diabetologia40,15–22.Riedl,E.,Pfister,F.,Braunagel,M.,Brinkkötter,P.,Sternik,P.,Deinzer,M.,Bakker,S.J.,Henning,R.H.,vandenBorn,J.,Krämer,B.K.,2011.CarnosinepreventsapoptosisofglomerularcellsandpodocytelossinSTZdiabeticrats.Cell.Physiol.Biochem.28,279–288.Rios,J.,Waterman,P.,1997.AreviewofthepharmacologyandtoxicologyofAstragalus.Phytother.Res.11,411–418.Ritz,E.,Rychlík,I.,Locatelli,F.,Halimi,S.,1999.End-stagerenalfailureintype2diabetes:amedicalcatastropheofworldwidedimensions.Am.J.KidneyDis.34,795–808.Ruoslahti,E.,1991.Integrins.J.Clin.Investig.87,1.Ruoslahti,E.,Giancotti,F.,19.Integrinsandtumorcelldissemination.CancerCells1.ColdSpringHarbor,NY,pp.119–126(19).So,B.-I.,Song,Y.-S.,Fang,C.-H.,Park,J.-Y.,Lee,Y.,Shin,J.H.,Kim,H.,Kim,K.-S.,2013.G-CSFpreventsprogressionofdiabeticnephropathyinrat.PLoSOne8,e77048.Sohn,E.,Kim,J.,Kim,C.-S.,Kim,Y.S.,Jang,D.S.,Kim,J.S.,2010.ExtractoftheaerialpartsofAsterkoraiensisreduceddevelopmentofdiabeticnephropathyviaanti-apoptosisofpodocytesinstreptozotocin-induceddiabeticrats.Biochem.Biophys.Res.Commun.391,733–738.Springer,T.A.,1990.Adhesionreceptorsoftheimmunesystem.Nature346,425–434.Stephens,G.W.,Gillaspy,J.A.,Clyne,D.,Mejia,A.,Pollak,V.,1990.Racialdifferencesintheincidenceofend-stagerenaldiseaseintypesIandIIdiabetesmellitus.Am.J.KidneyDis.:Off.J.Natl.KidneyFound.15,562.Su,J.,Li,S.-J.,Chen,Z.-H.,Zeng,C.-H.,Zhou,H.,Li,L.-S.,Liu,Z.-H.,2010.Evaluationofpodocytelesioninpatientswithdiabeticnephropathy:Wilms'tumor-1proteinusedasapodocytemarker.DiabetesRes.ClinPract.87,167–175.Tan,S.,Wang,G.,Guo,Y.,Gui,D.,Wang,N.,2013.Preventiveeffectsofanaturalanti-inflammatoryagent,astragalosideIV,onischemicacutekidneyinjuryinrats.Evid.-BasedComplementaryAltern.Med.,2013.Teixeira,V.D.P.C.,Blattner,S.M.,Li,M.,Anders,H.-J.,Cohen,C.D.,Edenhofer,I.,Calvaresi,N.,Merkle,M.,Rastaldi,M.P.,Kretzler,M.,2005.Functionalcon-sequencesofintegrin-linkedkinaseactivationinpodocytedamage.KidneyInt.67,514–523.Toda,K.,Tuan,T.-L.,Brown,P.J.,Grinnell,F.,1987.Fibronectinreceptorsofhumankeratinocytesandtheirexpressionduringcellculture.J.CellBiol.105,3097–3104.Tu,Y.,Huang,Y.,Zhang,Y.,Hua,Y.,Wu,C.,2001.Anewfocaladhesionproteinthatinteractswithintegrin-linkedkinaseandregulatescelladhesionandspreading.J.CellBiol.153,585–598.Tu,Y.,Li,F.,Goicoechea,S.,Wu,C.,1999.TheLIM-onlyproteinPINCHdirectlyinteractswithintegrin-linkedkinaseandisrecruitedtointegrin-richsitesinspreadingcells.Mol.Cell.Biol.19,2425–2434.White,K.E.,Bilous,R.W.,2004.Structuralalterationstothepodocytearerelatedtoproteinuriaintype2diabeticpatients.Nephrol.Dial.Transplant.19,1437–1440.White,K.E.,Bilous,R.W.,Marshall,S.M.,ElNahas,M.,Remuzzi,G.,Piras,G.,DeCosmo,S.,Viberti,G.,2002.Podocytenumberinnormotensivetype1diabeticpatientswithalbuminuria.Diabetes51,3083–30.94J.Chenetal./EuropeanJournalofPharmacology736(2014)86–94Wu,C.,1999.Integrin-linkedkinaseandPINCH:partnersinregulationofcell–extracellularmatrixinteractionandsignaltransduction.J.CellSci.112,4485–44.Wu,C.,Dedhar,S.,2001.Integrin-linkedkinase(ILK)anditsinteractorsanewparadigmforthecouplingofextracellularmatrixtoactincytoskeletonandsignalingcomplexes.J.CellBiol.155,505–510.Xu,J.,Li,Z.,Xu,P.,Yang,Z.,2012.Protectiveeffectsofleukemiainhibitoryfactoragainstoxidativestressduringhighglucose-inducedapoptosisinpodocytes.CellStressChaperones17,485–493.Zhang,L.,Zhang,P.,Wang,F.,Zuo,L.,Zhou,Y.,Shi,Y.,Li,G.,Jiao,S.,Liu,Z.,Liang,W.,2008.PrevalenceandfactorsassociatedwithCKD:apopulationstudyfromBeijing.Am.J.KidneyDis.51,373–384.
