microtubulecytoskeletoncomparedwithcellsfromyoungmice15Distinctnich - PDF document

URRENT PINION REVIEW microtubulecytoskeletoncomparedwithcellsfromyoungmice[15].Distinctnicheinteractionsarealsosuggestedbytheobservationthatprogenitorcellsfromoldmicearemobilizedmoreefficiently[16].Together,mostavailabledatasuggestintrinsicallycompromisedfunctionofindividualagedHSCs,butmaintenanceofoverallfunctionofHSCsasapopulation.Severallymphoidprecursors,suchascommonlymphoidprogenitors(CLPs)and,inparticular,pre-Bcellsaredepletedinthebonemarrowofoldmice[3].LymphopoiesisisnotonlyaffectedbyHSC-intrinsicchanges,however.Theage-associateddeclineinpre-Bcellsappearsmostlycausedbytheagedmicroenvironment[17],whereasT-celldevelopmentisseverelycurtailedbythymicinvolution[18].Lymphocytenumberismaintained,however,becausehomeostaticproliferationfillsthevoidwithantigen-experiencedcells(memoryTcells;marginalzone,B1andmemoryBcells)[2,3,19].Similartomice,thefrequencyofhumanHSCsincreaseswithage,andtheHSCcompartmentshowsamyeloidbiasatthepopulationlevelwhenassayedbyxenotransplantationinimmunodefi-cientmice[20].CONSEQUENCESOFAGINGOFHEMATOPOIETICSTEMCELLSThenegativeimpactofdonorageontheoutcomeofallogeneicbonemarrowtransplantationappearsmostlyduetohigherincidenceandseverityofgraft-versus-hostdisease,andnottoanengraftmentdefect[4].Itistemptingtospeculate,however,thattheincreasedincidenceofmyeloidmalignanciesintheelderlycanbetracedbacktothemyeloidbiasofHSCs.HSCs,theonlypersistentlineageofhemato-poieticcells,serveasarepositoryforaccumulatingDNAmutationsthatcanconferabnormalself-renewalpropertiestodownstreamprogenitorcells,leadingtoleukemia[1].Furthermore,thelowerincidenceoflymphoidmalignanciesoriginatingfromdevelopingBandTcellsintheagedcomparedwiththeyoung[21]maybecausedbyimpairedlymphopoiesis.Thefactthatthemostprevalenthematologicalmalignancyintheelderlyischroniclymphocyticleukemia(CLL)[22]isoftenover-lookedinthiscontext,however.CLLisamalignant,clonalB-cellexpansionwiththephenotypeofantigen-experiencedBcells.Althoughoriginallythoughttobedrivenbychronicstimulationbyasyetundefinedantigens[22],recentevidencesuggeststhatcellautonomoussignalingthroughtheB-cellreceptor(BCR)inresponsetointernalBCRepitopesdrivesthismalignancy[23].Never-theless,xenotransplantationstudiesinimmuno-deficientmicehaveshownthatthepropensitytogenerateoligoclonalormonoclonalB-celldevelop-mentoriginatesinHSCsofCLLpatients[24].Inthiscontext,isitalsointerestingtonotethatinmicewithatransgenicBCR,Bcellswithaphenotypeofantigen-experiencedcellsandbearingtheendogen-ousBCRincreasewithage[25].TransplantationofpurifiedHSCsfromoldmicetoyoungmicetrans-ferredthesealteredB-cellspecificities.Furthermore,transplantationofsmallnumbersofyoungHSCsortreatingrecipientswithanti-interleukin-7anti-bodies,twoapproachesthatcompromisetheB-cellregenerativecapacityofthehematopoieticsystem,alsoskewedtheB-cellpopulationtowardantigen-experiencedcellsexpressingtheendogenousBCR[26].ItisthereforeplausibletohypothesizethatagingofHSCs,andthereforetheirreducedlym-phoidpotentialandoverallfunctionalcompromise,maycontributetoaskewedB-cellrepertoireandtoapropensitytodevelopCLLoritsprecursors.MOLECULARPHENOTYPEOFAGEDHEMATOPOIETICSTEMCELLSGenome-wideexpressionstudiescomparingHSCsfromyoungandoldmiceshowedthatinagedHSCs,expressionofgenesinvolvedinmyeloiddevelop-mentwasincreasedwhereasexpressionofgenesinvolvedinlymphopoiesiswasdecreased,afindingconsistentwiththeirmyeloidbias[8].Further-more,genesetsassociatedwithinflammationwereincreasedinagedHSCs[8,27].YoungandagedHSCsdifferinepigeneticregulation.AgedHSCsshow KEYPOINTSAgingofHSCsisassociatedwithexpansionofstemcellnumber,decreasedfunctionofindividualstemcells,increasedmyeloidbias,decreasedpolarity,andincreasedDNAdamage.TheoverallfunctionoftheHSCcompartmentismaintained,however.Thesechangesarepredominantlyintrinsic.Expansionofthestemcellcompartment,increasedmyeloidbias,andadecreaseinlymphopoiesisstartearlyinlife,andmaythereforebedevelopmentallyprogrammed.Hematopoieticstemcellspossessintrinsicstemcell-protectivemechanismsthatpreservethefunctionofthestemcellcompartmentasawhole,butmaybeaccompaniedbyaccumulationofdamagedstemcells.Bothdevelopmentallyprogrammedchangesandstemcell-protectivemechanismscontributetoage-relatedchangesinHSCfunctionandinthehematopoieticsystem. Lymphoidbiologyanddiseases356Volume20Number4July2013 generalhypermethylation[14].Thiswassurprising,asaginginmosttissuesisassociatedwithhypo-methylation[28].Interestingly,however,differen-tiallymethylatedregionswereenrichedingenesexpressedindownstreamlineagesandnot,orlessso,inHSCsthemselves,andincludedgenesinvolvedinerythropoiesisandlymphopoiesis.Furthermore,afractionofthesegenesaretargetsofthepolcycombrepressorcomplex(PRC)2,andexpressionofseveralPRC2componentswasslightlyreducedinagedHSCs[14,29].Downregulationofgenesinvolvedintranscriptionandchromatinregu-lation,aswellasconcomitantmisexpressionoflymphoidtranscriptssuchasIggermlinetran-scripts,wasobservedinanotherstudy[27].Afrac-tionofHSCsinagedmicealsodisplaylowerhistoneH4lysineacetylation(AcH4K16)[30],providingfurtherevidenceforchangesinepigeneticregula-tion.AgedHSCsexpresshigherlevelsoftheRhoGT-Pasecdc42,andthisincreasedexpressionwasassociatedwiththedecreasedpolarityanddecreasedAcH4K16observedinagedHSCscomparedwithyoungHSCs[30].Finally,anage-associatedincreaseinDNAdamageasmeasuredbyfocihasbeenobservedinHS

Cs,suggestingaroleforaccumulatedDNAdamage[31].THEROLEOFPROLIFERATIVEHISTORYProliferationofmostnormalcellsultimatelyleadstosenescence.AdultHSCsareremarkablyquiescent,however,thoughtheycanreversiblyenterandexitcellcycle,eveninsteadystate[32,33,34].Giventheirrelativequiescence,theexpansionofthenormallycyclingagedHSCcom-partmentisunlikelytobeexplainedbycellularsenescence.Furthermore,despiteearliersugges-tions[35],thep16INK4alocus,aprimeregulatorofsenescence,remainsepigeneticallyrepressedinagedHSCs[29].Proliferationisaccompaniedbytelomereshort-ening.Humanshaveshortertelomeresthaninbredmice,andprogressivetelomereshorteningisobservedintheperipheralblood,aproxyoftelo-mereshorteninginHSCs[36,37].Althoughgenetictelomeredysfunctioninhumanscausesdyskeratosiscongenita,adiseasewithahighincidenceofbonemarrowfailure,andalthoughacquiredsevereaplasticanemiaisassociatedwithshorttelomeres[36,37],theroleoftelomereshorteninginphysio-logicalagingofhumanHSCshasnotbeendeter-mined[38].HSCsfailinlategenerationtelomerase-deficientmice[39].However,maintenanceofmouseHSCsisnotdependentontelomeres,asincreasingtelomeraseactivityinHSCsthroughoverexpressionoftelomerasedoesnotrescueHSCsfromexhaustioninducedbyserialtransplantation[40].OthermechanismsmustthereforebeinvokedtoexplainHSCexhaustionafterserialtransplan-tation.Nevertheless,forcedproliferationofmouseHSCsbyserialtransplantation[7],transplantationofverylownumbersofHSCs[14]orrepeatedadministrationof5-fluorouracil[14]inducedchangessimilartothoseobservedinphysiologicalaging,namelylowerreconstitutioncapacityandmyeloidbias.Thesefindingsmaysuggestacontri-butionofproliferativehistorytoage-relatedchangesinHSCs.Itshouldbenoted,however,thatinserialtransplantationstudies,thechronologicalageofyoungHSCsthatunderwentthreeroundsofserialtransplantation(approximately20months)wassimilartowhatisconsidered‘aged’inC57BL/6mice[7].ItisthereforeuncleartowhatextentfunctionalchangesintheHSCcompartmentinserialtrans-plantationstudiesarecausedbychronologicalaging,orbyforcedproliferation,orbyacombi-nationofboth.Furthermore,forcedproliferationofHSCsmaydifferfromphysiologicalaging.Repeated5-fluorouraciladministration,althoughinducingafunctionalphenotypesimilartothatofagedHSCs,wasassociatedwithglobalhypomethy-lation.Ontheotherhand,physiologicalagingofHSCswasassociatedwithglobalhypermethylation,althoughafractionofdifferentiallymethylatedregionswassharedbetweenphysiologicallyagedHSCsandHSCssubjectedtoforcedproliferationion&].Thus,althoughproliferativehistorymaycon-tributetophysiologicalHSCaginginmice,thelatterisnotfullymodeledbyforcedcyclingandexpan-sionofHSCs.THEROLEOFINFLAMMATIONAgingischaracterizedbyastateoflow-levelinflam-mation,thatmaybothcauseandperpetuatetheagingprocess[41].Theaginghematopoieticsystemmayitselfbeasourceofinflammatorymediators.ThymicinvolutionleadstoadecreasedpoolofnaiveTcells.Throughperipheralexpansion,memorycellsfillthevoidintheT-cellpoolinagedindividuals[19,42].Thisaccumulationofsenescentmemorycells,byvirtueoftheirconstitutiveproductionofinflammatorycytokines,hasbeenlinkedtothegeneralstateoflow-levelinflammationandfrailtythatcharacterizesaging(‘inflammaging’)[42].Acommonfeatureofgenome-wideexpressionstudiesofagedHSCsisthepreponderanceofgenesinvolvedininflammationamongupregulatedgenes[8,27].Furthermore,recentworkhasshownthatHSCsparticipateinsystemicinflammatoryresponses[43]andresponddirectlytoinflammatorycytokines Agingofthehematopoieticsystem2013WoltersKluwerHealth|LippincottWilliams&Wilkins357 suchasinterferons[44,45],andthatdeletionofmechanismsthatdampentheeffectofinterferonsonHSCsleadstoenhancedcyclingandexhaustion[46,47].InflammationmaythereforecontributetoHSCaging.However,itisalsopossiblethattheincreasedinflammatorygeneexpressioninHSCsisareflectionoftheinflammatoryenvironmentinagedindividuals,andnotnecessarilycausallyrelatedtoage-relatedchangesinHSCs.THEROLEOFSTEMCELL-PROTECTIVEContinuousexposuretoextrinsic(suchaslow-doseenvironmentalirradiation)orintrinsic[suchasreac-tiveoxygenspecies(ROS)generatedbycellularmetabolism]stressorscanleadtostemcelldysfunc-tion,andaffecttissueintegrity.HSCsareendowedwithspecificprotectivemechanisms,however.Quiescencelikelyprotectsfromsenescence[34HSCsareexquisitelysensitivetoROS,predomi-nantlyproducedbymitochondrialrespiration[48].MechanismsthatreduceROSproductionand/orincreaseROSscavenginginHSCsincludeAtm[49],Foxotranscriptionfactors[50]andtherelianceonglycolysisforATPproduction[51].HSCshaveanincreasedcapacityofautophagyinresponsetostresscomparedwithprogenitorcells,whichundergoapoptosisinthesameconditions[52Interestingly,autophagyandROSproductionandscavengingareregulatedbymembersoftheFoxofamily[50,52],whichareinvolvedinlifespanregulationinCaenorhabditiselegans[53],suggestingthatHSCshaveadoptedorganismalmaintenancemechanismsoflowerorganisms.HSCsaremoreradioresistantthanmyeloidprogenitors,amongothers,becauseofrobustinductionofDNAdamagecheckpointsandhighexpressionofprosurvivalgenesandp21.Becauseoftheirquiescentnature,HSCsusetheerror-pronenonhomologousend-join-ing(NHEJ)pathway,whereascyclingprogenitorsusehomologousrecombinationtorepairirradia-tion-inducedDNAdouble-strandbreaks.However,evenafterHSCsareinducedtocycle,andpartiallyrewiretheirDNArepairtowardhomologousrecom-bination,theydoremainmoreradioresistantthanprogeni

tors[54].Theuseoftheerror-proneNHEJleadstopersistentDNAdamageinHSCs,andmaycausemalignancy,althoughthelatterhasnotbeenformallydemonstrated[54].ThegeneinduceduponDNAdamage,inhibitsself-renewal,anddrivesthecellstowardlymphoiddifferen-tiation.Aslymphoid-biasedHSCsappearmoresen-sitivetotheeffectof,thismechanismmaycontributetodepletionoflymphoidbiasedHSCsandrelativemaintenanceofmyeloidbiasedHSCs[55].Takentogether,mostavailableevidencesuggeststhattheHSCcompartmentismaintainedthroughattemptedrepairatthecostofretainingdamagedcellsratherthanreplacingthesebyenhancedself-renewal.Agingmightbecausedbyfailureoftheseprotectivemechanisms.However,thereislittleevi-dencethatstemcell-protectivemechanismsovertlyfailinagingHSCs.Increasedautophagyismain-tainedinagedHSCs[52].Genome-wideexpressionstudiesdidnotrevealmajorchangesingenesessen-tialformaintenanceofquiescenceinHSCs[8,27],andHSCcyclingdoesnotincreasewithage.Althoughgeneticdeletionofstemcell-specificpro-tectivemechanismsresultsinstemcelldysfunction,thesephenotypesmayrepresentfrankstemcellfailureratherthanphysiologicalaging[34].Forexample,knockoutmiceinwhichcomponentsofDNArepairmechanismsweredeletedshowmostlydefectsindownstreamprogenitorcellsinsteadystate,andHSCfunctionisonlycompromisedaftertransplantationoflimitingnumbersofHSCs[31].Myeloid/lymphoidskewingisnotobservedintheseknockoutmousemodels[56].Thus,failureofmech-anismsthatprotectHSCsfromvariousformsofstressisunlikelytoexplainage-relatedchangesinHSCs.Analternativehypothesisisthatage-relatedchangesinHSCsmayatleastinpartbearesultoftheverymechanismsaimedatmaintainingfunc-tionoftheoverallHSCcompartment.Asmentionedbefore,theHSCcompartmentisgearedtowardmaintenancethroughpreventionofdamagebyROS,andthroughattemptedrepairratherthanreplacingHSCsbyenhancedself-renewal.ThismodeofmaintenanceoftheHSCcompartmentmayhaveevolvedbecauseoftheapparentlyextremesensitivityoftheHSCcompartmenttoenhancedcyclinginsteadystate.AlthoughHSCshaveenormousself-renewalpotentialaftertrans-plantation[57],enhancedHSCcyclinginsteadystateisassociatedwithrapidexhaustioninmultiplemouseknockoutmodelsinwhichquiescenceofHSCswasdisrupted[34].TheroleofquiescenceinHSCmaintenanceisnotwellunderstood,how-ever,asinsomeknockouts,suchasCdkn2cenhancedHSCcyclingwasnotassociatedwithexhaustion[58].Ithasbeenproposedthatquies-cenceisrequiredtopreventinappropriatedifferen-tiation,andthereforelossofHSCs[34].Thus,maintaininglowlevelsofself-renewalinsteadystateandfavoringattemptsatrepairoverdisposalofdamagedHSCscouldleadtoanHSCcompartmentthatexpandswithageandmaintainsitsoverallfunction,butinwhichthefunctionofindividualHSCsiscompromised. Lymphoidbiologyanddiseases358Volume20Number4July2013 THEROLEOFDEVELOPMENTALTheprogressivemyeloidbiasoftheHSCcom-partment,whichmaybecausedbythehigherintrinsicself-renewalcapacityofmyeloid-biasedHSCs[11,12],likelyservesanevolutionarypurposeaswell.Benzetal.al.&]showedthatalthoughinfetalliverbalanced()HSCspredominate,thefre-quencyofmyeloid-biased()HSCsincreasesasdevelopmentproceeds.CLPsderivedfromweredefectiveandexpressedlowerlevelsofB-line-age-specificgenes,aphenotypesimilartothatofagedCLPs[57].Theadaptiveimmunesystemdevelopsearlyinlifetoestablishantigen-specificresponsestothemultitudeofantigensencounteredduringextrauterinelife,whilepreventingresponsesagainstself.ThecostofthisprocessisiterativerecombinationofTandB-cellreceptorlociaccom-paniedbymassiveproliferationanddeletionofthemajorityofthecells,whichcarriestheriskofmalig-nancy.Indeed,acutelymphoblasticleukemiaispredominantlyadiseaseofchildren[21].Oncetheimmunerepertoireisestablished,cessationofrecombinationandestablishmentofvigorousandrapidmemoryresponseslikelybenefitsfitness.Thy-micinvolutionshouldprobablybeviewedinthiscontextaswell.SupportingthisnotionisthefactthatnotonlychangesinthetypesofBandTcellsproduced[59],butalsothedecreaseinB-celldevel-opment[59],theemergenceof-HSCclones[57 YoungAdultAged FIGURE1.Schematicoverviewoffunctionalchangesinthehematopoieticsystemthroughoutdevelopmentandaging.HSC,hematopoieticstemcell. Agingofthehematopoieticsystem2013WoltersKluwerHealth|LippincottWilliams&Wilkins359 PuiCH,RobisonLL,LookAT.Acutelymphoblasticleukaemia.Lancet2008;371:1030–1043.ChiorazziN,RaiKR,FerrariniM.Chroniclymphocyticleukemia.NEnglJMed2005;352:804–815.Duhren-vonMindenM,UbelhartR,SchneiderD,etal.Chroniclymphocyticleukaemiaisdrivenbyantigen-independentcell-autonomoussignalling.Nat-ure2012;489:309–312.ThisstudyshowsthatCLL,whichwasbelievedtodrivenbyasofyetunknownspecicantigens,maybecausedbyaninternalepitopeoftheBCR,andthereforebecellautonomous.KikushigeY,IshikawaF,MiyamotoT,etal.Self-renewinghematopoieticstemcellistheprimarytargetinpathogenesisofhumanchroniclymphocyticleukemia.Cancercell2011;20:246–259.JohnsonSA,RozzoSJ,CambierJC.Aging-dependentexclusionofantigen-inexperiencedcellsfromtheperipheralBcellrepertoire.JImmunol2002;168:5014–5023.GuerrettazLM,JohnsonSA,CambierJC.AcquiredhematopoieticstemcelldefectsdetermineB-cellrepertoirechangesassociatedwithaging.ProcNatlAcadSciUSA2008;105:11898–11902.ChambersSM,ShawCA,GatzaC,etal.Aginghematopoieticstemcellsdeclineinfunctionandexhibitepigeneticdysregulation.PLoSBiol2007;GonzaloS.Epigeneticalterationsinaging.JApplPhysiol2010;109:586–AttemaJL,PronkCJ,NorddahlGL,etal.Hematopoieticstemcellageingisuncoupledfrom

p16INK4A-mediatedsenescence.Oncogene2009;28:2238–2243.FlorianMC,DorrK,NiebelA,etal.Cdc42activityregulateshematopoieticstemcellagingandrejuvenation.CellStemCell2012;10:520–530.ThisstudyshowsthatthatinhibitionofCdc42correctscellpolarity,aswellHSCfunctionandlocalization.Importantly,inhibitionofCdc42alsorejuvenatestheAcH4K16stainingpatternofHSCs,suggestiveofreversibleepigeneticregulationofaging.RossiDJ,BryderD,SeitaJ,etal.DecienciesinDNAdamagerepairlimitthefunctionofhaematopoieticstemcellswithage.Nature2007;447:725–729.TakizawaH,RegoesRR,BoddupalliCS,etal.Dynamicvariationincyclingofhematopoieticstemcellsinsteadystateandinammation.JExpMed2011;208:273–284.WilsonA,LaurentiE,OserG,etal.Hematopoieticstemcellsreversiblyswitchfromdormancytoself-renewalduringhomeostasisandrepair.Cell2008;135:1118–1129.RossiL,LinKK,BolesNC,etal.Lessismore:unveilingthefunctionalcoreofhematopoieticstemcellsthroughknockoutmice.CellStemCell2012;11:302–317.OutstandingsystematicoverviewofmouseknockoutsinwhichHSCfunctionisJanzenV,ForkertR,FlemingHE,etal.Stem-cellageingmodiedbythecyclin-dependentkinaseinhibitorp16INK4a.Nature2006;443:421–426.CaladoRT,YoungNS.Telomerediseases.NEnglJMed2009;361:2353–SavageSA,BertuchAA.Thegeneticsandclinicalmanifestationsoftelomerebiologydisorders.GenetMed2010;12:753–764.LansdorpPM.Roleoftelomeraseinhematopoieticstemcells.AnnNYAcadSci2005;1044:220–227.WongKK,MaserRS,BachooRM,etal.TelomeredysfunctionandAtmdeciencycompromisesorganhomeostasisandacceleratesageing.Nature2003;421:643–648.AllsoppRC,MorinGB,HornerJW,etal.EffectofTERTover-expressiononthelong-termtransplantationcapacityofhematopoieticstemcells.NatMed2003;9:369–371.FranceschiC,BonafeM.Centenariansasamodelforhealthyaging.BiochemSocTrans2003;31:457–461.FranceschiC,BonafeM,ValensinS.Humanimmunosenescence:thepre-vailingofinnateimmunity,thefailingofclonotypicimmunity,andthellingofimmunologicalspace.Vaccine2000;18:1717–1720.KingKY,GoodellMA.InammatorymodulationofHSCs:viewingtheHSCasafoundationfortheimmuneresponse.NatRevImmunol2011;11:685–692.EssersMA,OffnerS,Blanco-BoseWE,etal.IFNalphaactivatesdormanthaematopoieticstemcellsinvivo.Nature2009;458:904–908.BaldridgeMT,KingKY,BolesNC,etal.QuiescenthaematopoieticstemcellsareactivatedbyIFN-gammainresponsetochronicinfection.Nature2010;465:793–797.FengCG,WeksbergDC,TaylorGA,etal.Thep47GTPaseLrg-47(Irgm1)linkshostdefenseandhematopoieticstemcellproliferation.CellStemCell2008;2:83–89.KingKY,BaldridgeMT,WeksbergDC,etal.Irgm1protectshematopoieticstemcellsbynegativeregulationofIFNsignaling.Blood2011;118:1525–NakaK,HiraoA.Maintenanceofgenomicintegrityinhematopoieticstemcells.IntJHematol2011;93:434–439.ItoK,HiraoA,AraiF,etal.RegulationofoxidativestressbyATMisrequiredforself-renewalofhaematopoieticstemcells.Nature2004;431:997–1002.TothovaZ,KolliparaR,HuntlyBJ,etal.FoxOsarecriticalmediatorsofhematopoieticstemcellresistancetophysiologicoxidativestress.Cell2007;128:325–339.SimsekT,KocabasF,ZhengJ,etal.Thedistinctmetabolicproleofhematopoieticstemcellsreectstheirlocationinahypoxicniche.CellStemCell2010;7:380–390.WarrMR,BinnewiesM,FlachJ,etal.FOXO3Adirectsaprotectiveautophagyprograminhaematopoieticstemcells.Nature2013;494:323–327.ThisstudydemonstratesspecicincreaseinthepropensityofHSCstoperformautophagyinconditionsinwhichprogenitorscellsundergoapoptosis.Further-more,agingdoesnotappeartoaffectthisprocess.KenyonC.ApathwaythatlinksreproductivestatustolifespaninCaeno-rhabditiselegans.AnnNYAcadSci2010;1204:156–162.MohrinM,BourkeE,AlexanderD,etal.Hematopoieticstemcellquiescencepromoteserror-proneDNArepairandmutagenesis.CellStemCell2010;7:174–185.WangJ,SunQ,MoritaY,etal.Adifferentiationcheckpointlimitshemato-poieticstemcellself-renewalinresponsetoDNAdamage.Cell2012;148:1001–1014.RossiDJ,SeitaJ,CzechowiczA,etal.HematopoieticstemcellquiescenceattenuatesDNAdamageresponseandpermitsDNAdamageaccumulationduringaging.CellCycle2007;6:2371–2376.BenzC,CopleyMR,KentDG,etal.Hematopoieticstemcellsubtypesexpanddifferentiallyduringdevelopmentanddisplaydistinctlymphopoieticpro-grams.CellStemCell2012;10:273–283.RigorousanalysisoftheclonalcompositionoftheHSCcompartmentduringdevelopmentandaging,showingthattheemergenceofmyeloidbiasedclonesstartsinthefetalbonemarrow.YuanY,ShenH,FranklinDS,etal.Invivoself-renewingdivisionsofhaematopoieticstemcellsareincreasedintheabsenceoftheearlyG1-phaseinhibitor,p18INK4C.NatCellBiol2004;6:436–442.Montecino-RodriguezE,DorshkindK.Evolvingpatternsoflymphopoiesisfromembryogenesisthroughsenescence.Immunity2006;24:659–662.Dominguez-GerpeL,Rey-MendezM.Evolutionofthethymussizeinresponsetophysiologicalandrandomeventsthroughoutlife.MicroscResTech2003;62:464–476.SteinmannGG.Changesinthehumanthymusduringaging.Currenttopicsinpathology.CurrTopPathol1986;75:43–88.DingL,MorrisonSJ.Haematopoieticstemcellsandearlylymphoidprogeni-torsoccupydistinctbonemarrowniches.Nature2013;495:231–235.See[63GreenbaumA,HsuYM,DayRB,etal.CXCL12inearlymesenchymalprogenitorsisrequiredforhaematopoieticstem-cellmaintenance.Nature2013;495:227–230.Thisstudyand[62]bothdemonstratethatearlylymphoidprogenitors,com-mittedB-cellprecursorsandHSCsoccupydistinctnichesinthebonemarrow.Importantly,whereasHSCsoccupyaperivascularniche,earlylymphoidprogeni-torsdependontheosteoblasticniche.PollinaEA,BrunetA.Epigeneticregulationofagin

gstemcells.Oncogene2011;30:3105–3126.OkitaK,YamanakaS.Inducedpluripotentstemcells:opportunitiesandchallenges.PhilosTransRSocLondBBiolSci2011;366:2198–2207.WahlestedtM,NorddahlGL,StenG,etal.Anepigeneticcomponentofhematopoieticstemcellagingamenabletoreprogrammingintoayoungstate.Blood2013.[Epubaheadofprint]WilliamsPD,DayT.Antagonisticpleiotropy,mortalitysourceinteractions,andtheevolutionarytheoryofsenescence.Evolution2003;57:1478–1488. Agingofthehematopoieticsystem2013WoltersKluwerHealth|LippincottWilliams&Wilkins361 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. microtubulecytoskeletoncomparedwithcellsfrom youngmice[15].Distinctnicheinteractionsarealso suggestedbytheobservationthatprogenitorcells fromoldmicearemobilizedmoreefficiently[16]. Together,mostavailabledatasuggestintrinsically compromisedfunctionofindividualagedHSCs, butmaintenanceofoverallfunctionofHSCsasa population.Severallymphoidprecursors,suchas commonlymphoidprogenitors(CLPs)and,in particular,pre-Bcellsaredepletedinthebone marrowofoldmice[3].Lymphopoiesisisnotonly affectedbyHSC-intrinsicchanges,however.The age-associateddeclineinpre-Bcellsappearsmostly causedbytheagedmicroenvironment[17],whereas T-celldevelopmentisseverelycurtailedbythymic involution[18].Lymphocytenumberismaintained, however,becausehomeostaticproliferationfills thevoidwithantigen-experiencedcells(memory Tcells;marginalzone,B1andmemoryBcells) [2,3,19]. Similartomice,thefrequencyofhumanHSCs increaseswithage,andtheHSCcompartment showsamyeloidbiasatthepopulationlevelwhen assayedbyxenotransplantationinimmunodefi- cientmice[20]. CONSEQUENCESOFAGINGOF HEMATOPOIETICSTEMCELLS Thenegativeimpactofdonorageontheoutcomeof allogeneicbonemarrowtransplantationappears mostlyduetohigherincidenceandseverityof graft-versus-hostdisease,andnottoanengraftment defect[4].Itistemptingtospeculate,however,that theincreasedincidenceofmyeloidmalignanciesin theelderlycanbetracedbacktothemyeloidbiasof HSCs.HSCs,theonlypersistentlineageofhemato- poieticcells,serveasarepositoryforaccumulating DNAmutationsthatcanconferabnormalself- renewalpropertiestodownstreamprogenitorcells, leadingtoleukemia[1].Furthermore,thelower incidenceoflymphoidmalignanciesoriginating fromdevelopingBandTcellsintheagedcompared withtheyoung[21]maybecausedbyimpaired lymphopoiesis.Thefactthatthemostprevalent hematologicalmalignancyintheelderlyischronic lymphocyticleukemia(CLL)[22]isoftenover- lookedinthiscontext,however.CLLisamalignant, clonalB-cellexpansionwiththephenotypeof antigen-experiencedBcells.Althoughoriginally thoughttobedrivenbychronicstimulationby asyetundefinedantigens[22],recentevidence suggeststhatcellautonomoussignalingthrough theB-cellreceptor(BCR)inresponsetointernal BCRepitopesdrivesthismalignancy[23 & ].Never- theless,xenotransplantationstudiesinimmuno- deficientmicehaveshownthatthepropensityto generateoligoclonalormonoclonalB-celldevelop- mentoriginatesinHSCsofCLLpatients[24].Inthis context,isitalsointerestingtonotethatinmice withatransgenicBCR,Bcellswithaphenotypeof antigen-experiencedcellsandbearingtheendogen- ousBCRincreasewithage[25].Transplantationof purifiedHSCsfromoldmicetoyoungmicetrans- ferredthesealteredB-cellspecificities.Furthermore, transplantationofsmallnumbersofyoungHSCs ortreatingrecipientswithanti-interleukin-7anti- bodies,twoapproachesthatcompromisetheB-cell regenerativecapacityofthehematopoieticsystem, alsoskewedtheB-cellpopulationtowardantigen- experiencedcellsexpressingtheendogenousBCR [26].Itisthereforeplausibletohypothesizethat agingofHSCs,andthereforetheirreducedlym- phoidpotentialandoverallfunctionalcompromise, maycontributetoaskewedB-cellrepertoireandtoa propensitytodevelopCLLoritsprecursors. MOLECULARPHENOTYPEOFAGED HEMATOPOIETICSTEMCELLS Genome-wideexpressionstudiescomparingHSCs fromyoungandoldmiceshowedthatinagedHSCs, expressionofgenesinvolvedinmyeloiddevelop- mentwasincreasedwhereasexpressionofgenes involvedinlymphopoiesiswasdecreased,afinding consistentwiththeirmyeloidbias[8].Further- more,genesetsassociatedwithinflammationwere increasedinagedHSCs[8,27].YoungandagedHSCs differinepigeneticregulation.AgedHSCsshow KEYPOINTS
AgingofHSCsisassociatedwithexpansionofstem cellnumber,decreasedfunctionofindividualstemcells, increasedmyeloidbias,decreasedpolarity,and increasedDNAdamage.Theoverallfunctionofthe HSCcompartmentismaintained,however.
Thesechangesarepredominantlyintrinsic.
Expansionofthestemcellcompartment,increased myeloidbias,andadecreaseinlymphopoiesisstart earlyinlife,andmaythereforebe developmentallyprogrammed.
Hematopoieticstemcellspossessintrinsicstemcell- protectivemechanismsthatpreservethefunctionofthe stemcellcompartmentasawhole,butmaybe accompaniedbyaccumulationofdamagedstemcells.
Bothdevelopmentallyprogrammedchangesandstem cell-protectivemechanismscontributetoage-related changesinHSCfunctionandinthe hematopoieticsystem. Lymphoidbiologyanddiseases 356 www.co-hematology.com Volume20
Number4
July2013 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. suchasinterferons[44,45],andthatdeletionof mechanismsthatdampentheeffectofinterferons onHSCsleadstoenhancedcyclingandexhaustion [46,47].Inflammationmaythe

reforecontributeto HSCaging.However,itisalsopossiblethatthe increasedinflammatorygeneexpressioninHSCs isareflectionoftheinflammatoryenvironment inagedindividuals,andnotnecessarilycausally relatedtoage-relatedchangesinHSCs. THEROLEOFSTEMCELL-PROTECTIVE MECHANISMS Continuousexposuretoextrinsic(suchaslow-dose environmentalirradiation)orintrinsic[suchasreac- tiveoxygenspecies(ROS)generatedbycellular metabolism]stressorscanleadtostemcelldysfunc- tion,andaffecttissueintegrity.HSCsareendowed withspecificprotectivemechanisms,however. Quiescencelikelyprotectsfromsenescence[34 & ]. HSCsareexquisitelysensitivetoROS,predomi- nantlyproducedbymitochondrialrespiration [48].MechanismsthatreduceROSproduction and/orincreaseROSscavenginginHSCsinclude Atm[49],Foxotranscriptionfactors[50]andthe relianceonglycolysisforATPproduction[51].HSCs haveanincreasedcapacityofautophagyinresponse tostresscomparedwithprogenitorcells,which undergoapoptosisinthesameconditions[52 && ]. Interestingly,autophagyandROSproductionand scavengingareregulatedbymembersoftheFoxo family[50,52 && ],whichareinvolvedinlifespan regulationin Caenorhabditiselegans [53],suggesting thatHSCshaveadoptedorganismalmaintenance mechanismsoflowerorganisms.HSCsaremore radioresistantthanmyeloidprogenitors,among others,becauseofrobustinductionofDNAdamage checkpointsandhighexpressionofprosurvival genesandp21.Becauseoftheirquiescentnature, HSCsusetheerror-pronenonhomologousend-join- ing(NHEJ)pathway,whereascyclingprogenitors usehomologousrecombinationtorepairirradia- tion-inducedDNAdouble-strandbreaks.However, evenafterHSCsareinducedtocycle,andpartially rewiretheirDNArepairtowardhomologousrecom- bination,theydoremainmoreradioresistantthan progenitors[54].Theuseoftheerror-proneNHEJ leadstopersistentDNAdamageinHSCs,andmay causemalignancy,althoughthelatterhasnotbeen formallydemonstrated[54].Thegene Batf is induceduponDNAdamage,inhibitsself-renewal, anddrivesthecellstowardlymphoiddifferen- tiation.Aslymphoid-biasedHSCsappearmoresen- sitivetotheeffectof Batf ,thismechanismmay contributetodepletionoflymphoidbiasedHSCs andrelativemaintenanceofmyeloidbiasedHSCs [55].Takentogether,mostavailableevidence suggeststhattheHSCcompartmentismaintained throughattemptedrepairatthecostofretaining damagedcellsratherthanreplacingtheseby enhancedself-renewal. Agingmightbecausedbyfailureofthese protectivemechanisms.However,thereislittleevi- dencethatstemcell-protectivemechanismsovertly failinagingHSCs.Increasedautophagyismain- tainedinagedHSCs[52 && ].Genome-wideexpression studiesdidnotrevealmajorchangesingenesessen- tialformaintenanceofquiescenceinHSCs[8,27], andHSCcyclingdoesnotincreasewithage. Althoughgeneticdeletionofstemcell-specificpro- tectivemechanismsresultsinstemcelldysfunction, thesephenotypesmayrepresentfrankstemcell failureratherthanphysiologicalaging[34 & ].For example,knockoutmiceinwhichcomponentsof DNArepairmechanismsweredeletedshowmostly defectsindownstreamprogenitorcellsinsteady state,andHSCfunctionisonlycompromisedafter transplantationoflimitingnumbersofHSCs[31]. Myeloid/lymphoidskewingisnotobservedinthese knockoutmousemodels[56].Thus,failureofmech- anismsthatprotectHSCsfromvariousformsof stressisunlikelytoexplainage-relatedchanges inHSCs. Analternativehypothesisisthatage-related changesinHSCsmayatleastinpartbearesultof theverymechanismsaimedatmaintainingfunc- tionoftheoverallHSCcompartment.Asmentioned before,theHSCcompartmentisgearedtoward maintenancethroughpreventionofdamageby ROS,andthroughattemptedrepairratherthan replacingHSCsbyenhancedself-renewal.This modeofmaintenanceoftheHSCcompartment mayhaveevolvedbecauseoftheapparently extremesensitivityoftheHSCcompartmentto enhancedcyclinginsteadystate.AlthoughHSCs haveenormousself-renewalpotentialaftertrans- plantation[57 & ],enhancedHSCcyclinginsteady stateisassociatedwithrapidexhaustioninmultiple mouseknockoutmodelsinwhichquiescenceof HSCswasdisrupted[34 & ].Theroleofquiescence inHSCmaintenanceisnotwellunderstood,how- ever,asinsomeknockouts,suchasCdkn2c  /  mice, enhancedHSCcyclingwasnotassociatedwith exhaustion[58].Ithasbeenproposedthatquies- cenceisrequiredtopreventinappropriatedifferen- tiation,andthereforelossofHSCs[34 & ].Thus, maintaininglowlevelsofself-renewalinsteadystate andfavoringattemptsatrepairoverdisposalof damagedHSCscouldleadtoanHSCcompartment thatexpandswithageandmaintainsitsoverall function,butinwhichthefunctionofindividual HSCsiscompromised. Lymphoidbiologyanddiseases 358 www.co-hematology.com Volume20
Number4
July2013 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. andthymicinvolution[59–61]startveryearlyin life,andarethereforelikelyaresultofdevelopmen- talprograms,andnotofaging perse . IthasrecentlybeenshownthatmostHSCs resideintheperivascularbonemarrowniche, whereasveryearlylymphoidprogenitorsresidein theosteoblasticniche[62 && ,63 && ].Itisnotknown whether a -HSCsand b -HSCsoccupydistinctniches, butitispossiblethatthelocalizationofagedearly progenitorsfartherawayfromtheosteoblastsisa reflectionofashifttowardmyelopoiesis[15]. Furthermore,althoughtheargumenthasbeenmade thatepigeneticchangescontributetoagingofstem cells[14 & ,27,64],epigeneticregulationiscriticalfor development.Thegenerationofind

ucedpluripo- tentstatecells(iPSCs)fromsomaticcellsinvolves near-completeepigeneticrewiringofthecells towardadevelopmental‘ground’state[65].The recentdemonstrationthatHSCsinmicederived fromiPSCsgeneratedfromagedHSCsarefunction- ally‘young’suggeststhatagingofHSCscarriesa largeepigeneticcomponentandisreversible.The argumentcouldbemade,aswasacknowledgedby theauthors,thatthedatamaybeareflectionofthe selectionofasubsetofHSCsthatarenotorareless affectedbyage-associateddamage[66].However, Florian etal. [30 && ]showedthatpharmacological inhibitionoftheRhoGTPaseCdc42allowedrejuve- nationofHSCfunctionintermsofpolarity,long- termrepopulationcapacityandlocalizationcloser totheosteoblasticniche.ThefactthattheAcH4K16 stainingpatternalsoappeared‘younger’suggestsa reversalofepigeneticchanges,whichcouldatleast inpartbedevelopmentallyprogrammed. CONCLUSION Althoughitispossiblethatproliferativehistory contributestotheagingofHSCs,partofthefunc- tionalphenotypeoftheagedhematopoieticsystem maybetheresultofintrinsicandextrinsicdevel- opmentalandstemcell-protectivemechanisms aimedatmaximizingfitnessduringreproductive life(summarizedinFig.1).Hematopoieticaging mightthereforebeanexampleofantagonistic pleiotropy,whichpositsthattraitsthatconfer fitnesspriortoandduringthereproductiveage canbedetrimentalafterreproductiveageandcon- tributetoaging[67].Hematopoieticregulationis gearedtowardmaintainingafunctionalHSCcom- partmentwithoptimalhomeostaticresponsesto stressandtowardgeneratinganadaptiveimmune systemearlyinlifewhileavoidinglymphoid malignanciesduringreproductiveage.Thesepro- cessesmaydecreasefitnesslaterinlife,however, assubfunctional;damagedHSCsaccumulateand establishmentofresponsestoneoantigens(tumor- associatedorinfectious)becomesproblematic becauseofthereducedgenerationofnaiveTand Bcells. Acknowledgements Partoftheauthor’sworkonagingwassupportedbyNIH grantRO1AG029262. Conflictsofinterest Therearenoconflictsofinterest. REFERENCESANDRECOMMENDED READING Papersofparticularinterest,publishedwithintheannualperiodofreview,have beenhighlightedas: & ofspecialinterest && ofoutstandinginterest AdditionalreferencesrelatedtothistopiccanalsobefoundintheCurrent WorldLiteraturesectioninthisissue(pp.397–398). 1. RossiDJ,JamiesonCH,WeissmanIL.Stemscellsandthepathwaystoaging andcancer.Cell2008;132:681–696. 2. DorshkindK,Montecino-RodriguezE,SignerRA.Theageingimmunesystem: isitevertoooldtobecomeyoungagain?NatRevImmunol2009;9:57–62. 3. LintonPJ,DorshkindK.Age-relatedchangesinlymphocytedevelopmentand function.NatImmunol2004;5:133–139. 4. KollmanC,HoweCW,AnasettiC, etal. Donorcharacteristicsasriskfactors inrecipientsaftertransplantationofbonemarrowfromunrelateddonors:the effectofdonorage.Blood2001;98:2043–2051. 5. HarrisonDE,AstleCM,StoneM.Numbersandfunctionsoftransplantable primitiveimmunohematopoieticstemcells.Effectsofage.JImmunol1989; 142:3833–3840. 6. MorrisonSJ,WandyczAM,AkashiK, etal. Theagingofhematopoieticstem cells.NatMed1996;2:1011–1016. 7. DykstraB,OlthofS,SchreuderJ, etal. Clonalanalysisrevealsmultiple functionaldefectsofagedmurinehematopoieticstemcells.JExpMed 2011;208:2691–2703. 8. RossiDJ,BryderD,ZahnJM, etal. Cellintrinsicalterationsunderliehema- topoieticstemcellaging.ProcNatlAcadSciUSA2005;102:9194–9199. 9. HarrisonDE.Long-termerythropoieticrepopulatingabilityofold,young,and fetalstemcells.JExpMed1983;157:1496–1504. 10. SudoK,EmaH,MoritaY,NakauchiH.Age-associatedcharacteristicsof murinehematopoieticstemcells.JExpMed2000;192:1273–1280. 11. BeermanI,BhattacharyaD,ZandiS, etal. Functionallydistincthematopoietic stemcellsmodulatehematopoieticlineagepotentialduringagingbya mechanismofclonalexpansion.ProcNatlAcadSciUSA2010;107: 5465–5470. 12. ChoRH,SieburgHB,Muller-SieburgCE.Anewmechanismfortheagingof hematopoieticstemcells:agingchangestheclonalcompositionofthestem cellcompartmentbutnotindividualstemcells.Blood2008;111:5553– 5561. 13. LiangY,VanZantG,SzilvassySJ.Effectsofagingonthehomingand engraftmentofmurinehematopoieticstemandprogenitorcells.Blood2005; 106:1479–1487. 14. & BeermanI,BockC,GarrisonBS, etal. Proliferation-dependentalterationsof theDNAmethylationlandscapeunderliehematopoieticstemcellaging.Cell StemCell2013;12:413–425. ExtensivecharacterizationofdifferentialmethylationinyoungandagedHSCs,as wellasHSCsthathaveundergoneforcedproliferation. 15. KohlerA,SchmithorstV,FilippiMD, etal. Alteredcellulardynamicsand endosteallocationofagedearlyhematopoieticprogenitorcellsrevealedby time-lapseintravitalimaginginlongbones.Blood2009;114:290–298. 16. XingZ,RyanMA,DariaD, etal. Increasedhematopoieticstemcellmobiliza- tioninagedmice.Blood2006;108:2190–2197. 17. LabrieJE3rd,SahAP,AllmanDM, etal. Bonemarrowmicroenvironmental changesunderliereducedRAG-mediatedrecombinationandBcellgenera- tioninagedmice.JExpMed2004;200:411–423. 18. ChinnIK,BlackburnCC,ManleyNR,SempowskiGD.Changesinprimary lymphoidorganswithaging.SeminImmunol2012;24:309–320. 19. SprentJ,SurhCD.NormalTcellhomeostasis:theconversionofnaivecells intomemory-phenotypecells.NatImmunol2011;12:478–484. 20. PangWW,PriceEA,SahooD, etal. Humanbonemarrowhematopoieticstem cellsareincreasedinfrequencyandmyeloid-biasedwithage.ProcNatlAcad SciUSA2011;108:20012–20017. Lymphoidbiologyanddiseases 360 www.co-hematology.com Volume20
Number4
Jul