ofMarch4,2015.
Thisinformationiscurrentas
Expression
SuppressorCellsbyModulatingSTAT3
SuppressivePotentialofMyeloid-Derived
BothmiR-17-5pandmiR-20aAlleviate
ZhangandRongcunYang
MengmengWang,YuanZhang,FenghuaGuo,Zhujun
ZhiqianZhang,XiaominSu,JinyiLiu,YingyingChen,
MiaomiaoZhang,QiaofeiLiu,SipingMi,XueLiang,
http://www.jimmunol.org/content/186/8/4716
doi:10.4049/jimmunol.1002989
March2011;
2011;186:4716-4724;Prepublishedonline7JImmunol
Material
Supplementary
9.DC1.html
http://www.jimmunol.org/content/suppl/2011/03/07/jimmunol.100298
References
http://www.jimmunol.org/content/186/8/4716.full#ref-list-1
,25ofwhichyoucanaccessforfreeat:cites40articlesThisarticle
Subscriptions
http://jimmunol.org/subscriptions
isonlineat:TheJournalofImmunologyInformationaboutsubscribingto
Permissions
http://www.aai.org/ji/copyright.html
Submitcopyrightpermissionrequestsat:
EmailAlerts
http://jimmunol.org/cgi/alerts/etoc
Receivefreeemail-alertswhennewarticlescitethisarticle.Signupat:
PrintISSN:0022-1767OnlineISSN:1550-6606.
Immunologists,Inc.Allrightsreserved.
Copyright?2011byTheAmericanAssociationof
9650RockvillePike,Bethesda,MD20814-3994.
TheAmericanAssociationofImmunologists,Inc.,
ispublishedtwiceeachmonthbyTheJournalofImmunology
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
TheJournalofImmunology
BothmiR-17-5pandmiR-20aAlleviateSuppressivePotential
ofMyeloid-DerivedSuppressorCellsbyModulatingSTAT3
Expression
MiaomiaoZhang,QiaofeiLiu,SipingMi,XueLiang,ZhiqianZhang,XiaominSu,
JinyiLiu,YingyingChen,MengmengWang,YuanZhang,FenghuaGuo,ZhujunZhang,
andRongcunYang
Myeloid-derivedsuppressorcells(MDSCs)wereoneofthemajorcomponentsoftheimmunesuppressivenetwork.STAT3hasan
importantroleinregulatingthesuppressivepotentialofMDSCs.Inthisstudy,wefoundthattheexpressionofSTAT3couldbe
modulatedbybothmiR-17-5pandmiR-20a.ThetransfectionofmiR-17-5pormiR-20aremarkablyreducestheexpressionofre-
activeoxygenspeciesandtheproductionofH
2
O
2
,whichareregulatedbySTAT3.MDSCstransfectedwithmiR-17-5pormiR-20a
arelessabletosuppressAg-specificCD4andCD8Tcells.Importantly,bothmiR-17-5pandmiR-20aalleviatethesuppressive
functionofMDSCsinvivo.TheexpressionofmiR-17-5pandmiR-20aintumor-associatedMDSCswasfoundtobelowerthanin
Gr1
+
CD11b
+
cellsisolatedfromthespleensofdisease-freemice.Tumor-associatedfactordownregulatestheexpressionofboth
miR-17-5pandmiR-20a.ThemodulationofmiR-17-5pandmiR-20aexpressionmaybeimportantfortheprocessbywhich
patientswithatumorcanovercometheimmunetolerancemediatedbyMDSCs.OurresultssuggestthatmiR-17-5pandmiR-20a
couldpotentiallybeusedforimmunotherapyagainstdiseases,especiallycancer,byblockingSTAT3expression.TheJournalof
Immunology,2011,186:4716–4724.
M
yeloid-derivedsuppressorcells(MDSCs)mediateglob-
alandprofoundimmunosuppressionthatespecially
affectsTcellfunction.MDSCsarecharacterizedbythe
expressionofthecell-surfacemoleculesGr1(includesthemac-
rophageandneutrophilmarkersLy6CandLy6G)andCD11b(1)in
mice;inhumans,MDSCsaretypicallyCD11b
+
CD33
+
HLA-DR
2
(2–5).ThepopulationofCD11b
+
Gr1
+
cellsinmiceiscomposed
oftwosubsets:polymorphonuclearMDSCs,whichhaveagranu-
locyticphenotypeandexpresstheLy6Gmarker(CD11b
+
Ly6C
low
Ly6G
high
cells),andmononuclearMDSCs,whichhaveamono-
cyticphenotypeandexpresstheLy6Cmarker(CD11b
+
Ly6G
low
Ly6C
high
cells)(6).BecauseMDSCsarecriticalfortumor-asso-
ciatedimmunesuppression,therehavebeenanumberofstudies
focusingonidentifyingtherapeuticmeanstoeliminatethese
cellsortoregulatetheirfunction.Nonetheless,themechanism
bywhichtheirsuppressivefunctionisregulatedremainsun-
known.SeveralstudieshaveshownthattheeffectofMDSCs
ismediatedviatheJAK2–STAT3pathway,whichistriggered
bytumor-derivedfactors(7,8).Theconstitutiveupregulation
ofSTAT3transcriptionfactorsinducestheexpansionofMDSCs
andenhancessuppressionintumor-bearingmicewithhigh
levelsofNADPHoxidasecomponents(9).Membrane-asso-
ciatedheatshockprotein72fromtumor-derivedexosomes
mediatestheSTAT3-dependentimmunosuppressivefunctionof
mouseandhumanMDSCs(10).Thus,theinhibitionofSTAT3
activityand/orsilencingofSTAT3expressionmaybeastrategy
toovercomeMDSCfunctionandimprovetheimmuneresponse
tocancer.
MicroRNAs(miRNAs)areshort,noncodingRNAsthatnega-
tivelyregulatetargetmRNAs.Basepairinteractionsbetween
miRNAsandtargetmRNAs,oftenwithinthe39untranslatedregion
(UTR)oftarget,resultinthedegradationofthetargetmRNAsor
theinhibitionoftheirtranslation(11).ManymiRNAsclonedfrom
mousebonemarrowcells(BMCs)aredifferentiallyregulatedin
varioushematopoieticlineages(12).miRNAsplayacriticalrole
inregulatingthedevelopmentofimmunecellsandalsoinmod-
ulatinginnateandadaptiveimmuneresponses(1,13–17).
TheSTAT339UTRcontainsfourconservedmiRNAbinding
points,twoforthemiR-17/20/93.mr/106/519familymembers,
oneformiR-125/361,andanotheroneformiR-124/606.Because
STAT3hasacriticalroleinregulatingMDSC-mediatedimmu-
nosuppression,wechosetostudyeffectsofthemiR-17-5pand
miR-20aofthemiR-17/20/93.mr/106/519familyonthesup-
pressivepropertiesofMDSCs.WefoundthatmiR-17-5pand
miR-20amayalleviatethesuppressivefunctionofMDSCsby
silencingSTAT3expression.Thisresultshouldmotivatefuture
researchintoclinicalapplicationsofmiRNAmoleculesinim-
munotherapy.
DepartmentofImmunology,NankaiUniversitySchoolofMedicine,NankaiUniver-
sity,Tianjin300071,People’sRepublicofChina;andKeyLaboratoryofBioactive
Materials,MinistryofEducation,NankaiUniversity,Tianjin300071,People’sRe-
publicofChina
ReceivedforpublicationSeptember7,2010.AcceptedforpublicationFebruary8,
2011.
ThisworkwassupportedbyNationalNaturalScienceFoundationofChinaGrants
30771967,30872315,and30830096,aMinistryofScienceandTechnologygrant
(863Program,2008AA02Z129),aNationalKeyBasicResearchandDevelopment
ProgramofChinagrant(973Program,2007CB914803),andbytheNationalKey
ScientificProgram(2011CB964902).
AddresscorrespondenceandreprintrequeststoDr.RongcunYang,Departmentof
Immunology,NankaiUniversitySchoolofMedicine,NankaiUniversity,Weijin
Road,Tianjin300071,People’sRepublicofChina.E-mailaddress:yang@nankai.
edu.cn
Theonlineversionofthisarticlecontainssupplementalmaterial.
Abbreviationsusedinthisarticle:BMC,bonemarrowcell;MDSC,myeloid-derived
suppressorcell;miRNA,microRNA;qRT-PCR,quantitativereal-timePCR;ROS,
reactiveoxygenspecies;siRNA,smallinterferingRNA;UTR,untranslatedregion;
VLP,virus-likeparticle;WT,wild-type.
CopyrightC2112011byTheAmericanAssociationofImmunologists,Inc.0022-1767/11/$16.00
www.jimmunol.org/cgi/doi/10.4049/jimmunol.1002989
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
MaterialsandMethods
Mice
Four-to6-wk-oldfemaleBALB/c,C57BL/6,andnudemice(Beijing
AnimalCenter)weremaintainedinapathogen-freeanimalfacilityatleast1
wkbeforeuse.Experimentswereperformedinaccordancewithinstitutional
guidelines.Variouss.c.tumormodels,includingCT-26coloncarcinoma
(AmericanTypeCultureCollection)inBALB/cmiceandLewislung
carcinoma(AmericanTypeCultureCollection)and1D8ovariancarcinoma
(providedbyKatherineF.Roby,UniversityofTexas)inC57BL/6mice,
wereusedinthisstudy.Thenumberoftumorcellsinjecteds.c.foreach
modelwasbasedontheabilitytoformatumorwith1.5cmdiameterwithin
3–4wkofinjection.
CelllinesandMDSCs
HumanembryonickidneycelllineHEK293Tcellsandmousemonocyte/
macrophagecelllineRAW264.7werepurchasedfromtheAmericanType
CultureCollection.
MurineMDSCswereisolatedfromthespleensofCT-26tumor-bearing
miceornaivemiceusingmagneticbeadconjugatedtoGr1mAbandanLS
columnaccordingtothemanufacturer’sprotocol(MiltenyiBiotec,Ber-
gischGladbach,Germany)orsortedusingFACScanafterstainingwith
anti-Gr1andCD11bAbs.PurifiedsplenicMDSCswere.90%Gr1
+
CD11b
+
cells.
ForinvitropreparationofCD11b
+
Gr1
+
MDSCs,murineBMCswere
firstpreparedaspreviouslydescribedbyus(18).TandBlymphocytes
weredeletedusingCD4Tcell-,CD8Tcell-,andBcell-positiveselected
magneticbeads(MACS;MiltenyiBiotec)accordingtothemanufacturer’s
protocol.Unlessespeciallyindicated,CD4Tcell-,CD8Tcell-,andBcell-
freeBMCswereusedintheexperiments.Topreparethetumor-induced
CD11b
+
Gr1
+
MDSCsinvitro,5310
4
CT26,Lewis,or1D8tumorcells
werecoculturedwith2310
6
BMCsina24transwellplate,ortumor-
associatedfactorPGE
2
(2.6Amol/l;Sigma-Aldrich,St.Louis,MO)was
addedtoacultureof2310
6
BMCsinRPMI1640mediumsupplemented
with3%FCS(Invitrogen)and1%penicillinandstreptomycin(Invitrogen)
inthepresenceof10ng/mlGM-CSF(R&DSystems,Minneapolis,MN)
and10ng/mlIL-4(R&DSystems)for5dasdescribedbySinhaetal.(19).
Surfacemarkersofthecellswereanalyzedbyflowcytometry(BDBio-
sciences,SanDiego,CA)attheindicatedtime.
Flowcytometricanalysis
Cellswerecollectedinice-coldPBSandincubatedwiththefollowingAbs:
FITC-,PE-,orallophycocyanin-conjugatedanti-CD4(L3T4),anti-CD8a
(Ly-2),anti-CD19(1D8),anti-CD11b(M1/70),anti–Gr-1(RB6-8C5),anti-
Ly6G(1A8),and/oranti-Ly6C(AL21)Abs.AlloftheseAbswerepur-
chasedfromBDBiosciences.CellswerestainedandresuspendedinPBS
with1%paraformaldehydeand1%FCSandkeptat4?Cpriortoflow
cytometricanalysis(FAScan;BDBiosciences).Foreachanalysis,isotype-
matchedcontrolmAbswereusedasanegativecontrol.
miRNA,smallinterferingRNA,gene,and39UTRexpression
constructs
ThemiR-17-5pandmiR-20amimics,miR-17-5pandmiR-20ainhibitors,as
wellasmiRNAmimicsandinhibitornegativecontrolswerepurchasedfrom
Dharmacon;STAT3smallinterferingRNA(siRNA)andthesiRNAneg-
ativecontrolwerepurchasedfromSantaCruzBiotechnology.BLOCK-iT
fluorescentoligoswerepurchasedfromInvitrogen.ThepCMV-SPORT6/
STAT3completecds(cDNAcloneMGC:67973IMAGE:4500786)
containinga39UTRwithmiR-17-5pandmiR-20aseedsequenceswere
purchasedfromAmericanTypeCultureCollection.Pri–miR-17-5pand
pri–miR-20aweredirectlyclonedfrommousegenomicDNAinto
pcDNA3.1
+
expressionvectorsattheHindIIIandXhoIsitesusingthe
primers.TheresultingconstructwastermedpcDNA3.1-miR-17-5pand
pcDNA3.1-miR-20a.TheSTAT339UTRwasclonedfrommousespleen
cellgenomicDNAviaPCRusingprimerscontainingrestrictionenzyme
SacIandXbaIlinkersthatwouldallowittobeclonedintothepIS2lu-
ciferasereportervector(Addgene).TheSTAT339UTRmutant1and
mutant2weregeneratedusingfourprimersaccordingtoaprevious
method(20).MutationswereconfirmedbyplasmidDNAsequencing.The
primersusedinthisstudyarelistedinSupplementalTableI.
Transfection
FormiRNAmimics,inhibitors,siRNAs,andnegativecontrololigos,cells
weretransfectedwiththeindicatedoligonucleotides(100nM)usingthe
Lipofectaminereagent(Invitrogen)orEntranster-R(EngreenBiosystem)
accordingtothemanufacturers’instructions.Fordifferentconstructs,the
cellswereperformedusingaNucleofectionapproachaccordingtothe
manufacture’sprotocol(AmaxaBiosystems,Berlin,Germany).
RNAisolationandquantitativereal-timePCR
Fortheexpressionofgeneandpri-miRNA,quantitativereal-timePCR
(qRT-PCR)wasperformedusingtheQuantiTectSYBRGreenPCRkitwith
aspecificsetofprimersaccordingtothesuggestedmethod(Qiagen).For
maturemiR-17-5pandmiR-20a,qRT-PCRwasperformedusingastan-
dardTaqManPCRkitprotocolonanAppliedBiosystems7900HTse-
quencedetectionsystem.AmplificationofU6smallRNA(formature
miRNA;AppliedBiosystems)andGAPDHmRNA(forgeneandpri-
miRNA;AppliedBiosystems)wasperformedwitheachexperimental
sampleasanendogenouscontrol.Foldchangeswerecalculatedusingthe
ΔΔcyclethresholdmethodaccordingtothemanufacturer’sprotocol
(AppliedBiosystems).Allreactionswererunintriplicate.Theprimer
sequencesarelistedinSupplementalTableI.
Luciferasereporterassays
Cellswereplatedina48-wellplateatadensityof4310
4
cells/wellin250
mlculturemedium1dpriortotransfection;then,cellswerecotransfected
withtheindicatedexpressionplasmids,reporterplasmidpIS2withRenilla
luciferase,andcontrolplasmids(Promega)usingLipofectaminereagent
(Invitrogen)accordingtothemanufacturers’recommendationsorusing
theNucleofectortechnology(AmaxaBiosystems).Renillaandfirefly
luciferaseactivitiesweremeasured24haftertransfectionusingadual
luciferasekit(Promega).Renillaluciferaseactivitywasnormalized
accordingtotheactivityoffireflyluciferase.Theluciferaseactivity
wasmeasuredafter24h.
Westernblottinganalysis
Westernblottingwasperformedaccordingtoourpreviousprotocol.
HybridizationswithprimaryAbswerecarriedoutfor1hatroomtem-
peratureinblockingbuffer.Theprotein–Abcomplexesweredetectedusing
peroxidase-conjugatedsecondaryAbs(BoehringerMannheim)andECL
(AmershamBiosciences).Anti-mouseSTAT3andp47
phox
werepurchased
fromSantaCruzBiotechnology.
ReactiveoxygenspeciesandH
2
O
2
detection
Theoxidation-sensitivedye2,7-dichlorofluorescindiacetate(Molecular
Probes/Invitrogen)wasusedtomeasurereactiveoxygenspecies(ROS)
productionbyMDSCsaccordingtothereportedprotocol(21).Isolated
CD11b
+
Gr1
+
MDSCsweresimultaneouslyculturedwith2.5mM2,7-
dichlorofluorescindiacetateand30ng/mlPMA(Sigma-Aldrich)for30
min.Analysiswasthenconductedbyflowcytometryasdescribedabove.
TheproductionofH
2
O
2
wasdetectedusingtheAmplexredhydrogen
peroxide/peroxidaseassaykit(Invitrogen)asrecommendedbytheman-
ufacturer.Inbrief,1310
4
cellswereresuspendedinPBS.Afteraddition
ofPMA(30ng/ml)for30min,theabsorbanceat560nmwasmeasured
usingamicroplateplatereader(Bio-Rad)at37?C.Absorbanceresults
werenormalizedtoastandardcurvegeneratedbyserialdilutionsof20
mMH
2
O
2
.
InvitroMDSCsuppressionassay
Tomeasurethesuppressionbytumor-associatedCD11b
+
Gr1
+
MDSCsin
theAg-specificresponse,MDSCs(5310
5
/wellunlessotherwisestated)
thathadbeentransfectedwithmiR-17-5pandmiR-20amimics,miR-17-
5pandmiR-20ainhibitors,andSTAT3siRNAoranegativecontrolfor
miRNAmimics,inhibitorandsiRNAwereaddedintoaCD4orCD8Tcell
(1310
6
)cultureofsplenocytesduringhumanpapillomavirustype16
virus-likeparticle([VLP]25mg/ml)stimulation,andthesupernatants
werecollectedaftercoculturefor48h.Inthisexperiment,transfected
MDSCsservedasthesuppressors,isolatedCD4andCD8Tcellsfrom
splenocytesofmiceimmunizedwithhumanVLPsservedastheres-
ponders,andVLPswereusedasstimulators.Noneofnegativecontrols
wasfoundtoaffecttheexpressionofSTAT3(SupplementalFig.1).VLP-
specificCD4orCD8Tcellsweregeneratedusingpublishedmethods(18)
andisolatedusingpositiveCD4andCD8selectivebeadsandanLScol-
umnfromMitenyiBiotec,usingthemanufacturer’sprotocol.
InvivoMDSCsuppressiveexperiments
ForinvivoMDSCsuppressiveexperiments,MDSCstransfectedwithmiR-
17-5pandmiR-20amimics,miR-17-5pandmiR-20ainhibitors,andSTAT3
siRNAornegativecontrolswereinjectedintothetumorsofmiceondays+1
and+7(2310
6
pretreatedMDSCs/mouse,respectively)aftertheinocu-
TheJournalofImmunology4717
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
lationofCT-26tumorcells.TheliveMDSCstransfectedwithwith
BLOCK-iTfluorescentoligoscanbemeasuredinvivoevenafterinjection
for5d(SupplementalFig.2).Invaccinatedmice,naivemicewerefirst
injectedinthefoodpadswith1310
6
dendriticcellsandwarmed(10min
at42?C)andirradiated(8000rad)withCT-26cells;1wklatertheywere
injecteds.cwith1310
6
liveCT-26cellsandMDSCs.
ELISA
CommercialsandwichELISAkitswereusedtomeasurethelevelsofIFN-g
(PierceEndogen).TheODsofeachofthesamplesweremeasuredat450
nmusingaSpectraMax190ELISAplatereader.Cytokinelevelswere
quantifiedfromthreetitrationsusingstandardcurvesandareexpressedin
picogramspermilliliter.
Statisticalanalysis
Forstatisticalanalysis,weusedtheStudentttest;a95%confidencelimit
wasdelineatedassignificantandwasdefinedasp,0.05.
Results
STAT3isregulatedbymiR-17-5pandmiR-20a
UsingTargetScan5.1(http://www.targetscan.org)combinedwith
PicTar(http://pictar.bio.nyu.edu/)andmiRanda(http://www.microrna.
org/)analyses,wefoundthatthe39UTRoftheSTAT3genehastwo
highlyconservedseedsequences(position156–162andposition
403–409inmice)forthemiR-17-5pandmiR-20ainthemiR-17–92
cluster(Fig.1A).WeinvestigatedtheeffectofmiR-17-5pand
miR-20aontheexpressionofSTAT3.Becausethereexisttwoseed
sequencesintheSTAT339UTR,wefirsttestedthecontribution
ofeachseedsequenceonregulatingtheexpressionofSTAT3.
ThemouseSTAT339UTRcontainingposition156–162(STAT3
39UTR1)andcontainingposition403–409(STAT339UTR2)or
mutatedversions(STAT339UTR1MutorSTAT339UTR2Mut)
(Fig.1B)wereclonedintoapIS2reportervectordownstream
FIGURE1.BindingofmiR-17-5pandmiR-20awithSTAT339UTR.A,SchematicrepresentationofthepredictedSTAT339UTRindicatingthebinding
sitesformiR-17-5pandmiR-20a,anddesignedmutatedversionoftheSTAT339UTR.B,Constructedluciferasereportervectors.Luciferasereporter
vectorswereconstructedinpLS2-REPORT-STAT339UTRwiththefollowing:1):seedsequence1(156–162,STAT339UTR1);2)seedsequence2(403–
409,STAT339UTR2);3)mutatedseedsequence1(STAT339UTR1Mut);4)mutatedseedsequence2(STAT339UTR2Mut);5)seedsequence1andseed
sequence2[STAT339UTR(1+2)];6)mutatedseedsequence1andseedsequence2(STAT339UTR1Mut+2);7)seedsequence1andmutatedsequence2
(STAT339UTR1+2Mut).TheSTAT339UTRwasclonedandmutatedbyPCR-basedmutagenesisaccordingtotheprotocolinMaterialsandMethods.C,
BothmiR-17-5pandmiR-20adirectlyinteractwiththeSTAT339UTRposition156–162(39UTR156–162)andposition403–409(39UTR403–409)orboth
positions(39UTR156–162and403–409)toinhibitreportergeneexpression.293Tcellswerecotransfectedwiththeluciferasereportervectorsand
pcDNA3.1-pri-miR17-5p(miR-17-5p),pcDNA3.1-pri-miR-20a(miR-20),orcontrolstructures(miRctr)into293Tcells.Theluciferaseactivitywas
measuredafter24haccordingtotheprotocoldescribedinMaterialsandMethods.D,Relativeluciferaseactivityin293TcellscotransfectedwithSTAT3
39UTR(1+2)andpcDNA3.1-pri-miR17-5p(miR-17-5p)orpcDNA3.1-pri-miR-20a(miR-20a)andwithincreasingamountsofmiR-17-5pinhibitorsor
miR-20ainhibitors.ControlcellswerecotransfectedwithmiRNAinhibitornegativecontrols(Ctr.).Theluciferaseactivitywasmeasuredafter24h
accordingtotheprotocoldescribedinMaterialsandMethods.Datapresentedaremeans6SDfromtriplicatetests.Theluciferaseactivityin293Tcells
cotransfectedwithpLS2-REPORT-STAT339UTRandmiRNAcontrolstructure(miRctr)wasarbitrarilysetto1.p,0.05,p,0.01.R.A,relative
activity.
4718miR-17-5pANDmiR-20aREGULATEMDSCs
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
fromRenillaluciferaseandthenusedtoassesswhethermiR-17-5p
andmiR-20acouldrepressluciferasegeneexpressionin293Tcells.
AsshowninFig.1C,luciferaseexpressionincellstransfectedwith
STAT339UTR1andSTAT339UTR2isrepressedbymiR-17-5p
andmiR-20a;themutationofbothposition157–162andposition
404–409ofthe39UTRseedmatchsequencesrelievestheinhibi-
tionbymiR-17-5pandmiR-20a.Furthermore,bothmiR-17-5p
andmiR-20ainhibitorsalsoremarkablyblockmiR-17-5p–and
miR-20a–induceddownregulationofluciferaseactivityincells
transfectedwithSTAT339UTR1andSTAT339UTR2(Fig.1D),
suggestingthateachseedsequenceofSTAT339UTRplaysanim-
portantroleinregulatingtheexpressionofSTAT3.Tofurther
confirmthecontributionofeachseedsequence,STAT339UTR
containingposition156–162andposition403–409(STAT3
39UTR1+2),STAT339UTR1mutantwithwild-type(WT)seed
sequence403–409(STAT339UTR1Mut+2)andSTAT339UTR2
mutantwithWTseedsequence157–162(STAT339UTR1+2Mut)
(Fig.1B)werealsoclonedintoapIS2reportervector.Themutation
ofeitherposition157–162orposition404–409intheSTAT3
39UTR1+2doesnotsignificantlyaffecttheluciferaseexpressionin
thepresenceofmiR-17-5pandmiR-20a(Fig.1C),furtherin-
dicatingthatbothseedsequencesregulatetheexpressionofSTAT3.
WenextinvestigatedwhetherbindingofmiR-17-5pandmiR-20a
toSTAT339UTRcouldregulateSTAT3expression.Themonocyte/
macrophagecelllineRAW264.7wastransfectedwithmiR-17-5por
miR-20amimics.Transfectionwasconfirmedusingfluorescently
labeledmiRNAsimilarprobes(Fig.2A),andqRT-PCR(Fig.2C)and
STAT3expressionweremeasuredatboththemRNAandprotein
levels.WefoundthatmiR17-5pormiR-20aaffectsSTAT3expres-
siononlyattheproteinlevel.AsshowninFig.2,thelevelsofSTAT3
proteininRAW264.7cellstransfectedwithmiR17-5pormiR-20a
mimicswereindeedlowerthaninnegativecontroltransfected
cells,whereasthetranscriptionlevelofSTAT3wasnotaffected.
Asapositivecontrol,STAT3siRNAnotonlydownregulatesthe
transcriptionbutalsoproteinlevelsofSTAT3.Notably,thesup-
pressionbymiR-17-5pandmiR-20awasfoundtobemeasurably
different(Fig.2),eventhoughbothbelongtothesamecluster.This
resultsupportstheideathatmiR-17-5pandmiR-20a,although
membersofthesamemiR-17-92polycistronicmiRNAcluster,may
varyinthedegreeofbindingaffinitytotheirtargetmRNAs(22).
FIGURE2.ExpressionofSTAT3isregulatedbymiR-17-5pandmiR-20a.A,RAW264.7cellstransfectedwithBLOCK-iTfluorescentoligos(1,2)or
withunlabeledoligos(3,4).1and3,RAW264.7cellsunderlightmicroscope;2and4,RAW264.7cellsunderafluorescencemicroscopeaftertransfection
for24h.Scalebars,20mm.B,BothmiR-17-5pandmiR-20areducethetranscriptionofgp91andp47butnotSTAT3(1)anddecreasetheexpressionof
STAT3andp47(2).RAW264.7cellstransfectedwithmiR-17-5pmimics(miR-17-5p),miR-20amimics(miR-20a),STAT3siRNA(siRNA)orapooled
negativecontrol(Ctr.)containingmiRNAmimicsandsiRNAnegativecontrololigoswereculturedfor24handthemRNAandproteinlevelsofSTAT3,
gp91,andp47ineachtransfectedcultureweremeasuredusingqRT-PCRandWesternblotting(gp91Absdonotworkwell;onlyp47isshown).C,Relative
levelsofSTAT3,gp91,andp47transcriptionandexpressioninRAW264.7cellstransfectedwithincreasingamountsofmiR-17-5pormiR-20amimics.
miR-17-5pandmiR-20aare,respectively,levelsofmaturemiR-17-5pandmiR-20ainRAW264.7cellstransfectedwithmiR-17-5p,miR-20amimics,or
miRNAmimicsnegativecontrol(Ctr.);STAT3,gp91,andp47are,respectively,transcriptionallevelsofSTAT3,gp91,andp47inRAW264.7cells
transfectedwithmiR-17-5p,miR-20amimics,ormiRNAmimicsnegativecontrol(Ctr.).RAW264.7cellsweretransfectedwithdifferentamountsofmiR-
17-5pormiR-20amimics.LevelsofmaturemiR-17-5pandmiR-20aandtranscriptionallevelsofSTAT3,gp91,andp47weredetectedusingqRT-PCR
aftertransfectionfor24h.D,ExpressionlevelsofSTAT3inRAW264.7cellstransfectedwithincreasingamountsofmiR-17-5pormiR-20amimics.
RAW264.7cellsweretransfectedwithdifferentamountsofmiR-17-5pormiR-20amimics,andexpressionlevelsofSTAT3proteinweredetectedusing
Westernblottingaftertransfectionfor24h.
TheJournalofImmunology4719
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
BecauseSTAT3mediatesthefunctionofMDSCsbyregulating
NADPHoxidaseexpression(21),wealsoinvestigatedtheeffectof
miR-17-5pandmiR-20aontheSTAT3-associatedNADPHoxi-
dase.AsshowninFig.2,miR-17-5pandmiR-20aalsoreduced
thetranscriptionalandproteinlevelsofp47
phox
andgp91
phox
,two
subunitsofNADPHoxidase.Furthermore,thisdownregulationis
dose-dependent(Fig.2C).Thus,ourresultsdemonstratethatboth
miR-17-5pandmiR-20atargettheSTAT339UTRtoreducethe
expressionofSTAT3.
BothmiR-17-5pandmiR-20alowerthesuppressivepotential
ofMDSCsinvitro
BecausethelevelsofROSandH
2
O
2
aredirectlyregulatedbythe
STAT3–NADPHpathway(21),wefirsttestedtheeffectofmiR-17-
5pandmiR-20aontheexpressionofROSandproductionofH
2
O
2
inRAW264.7monocyte/macrophagecells.ROSlevelsweremea-
suredusingtheoxidation-sensitivefluorescentdye2,7-dichloro-
fluorescindiacetate.AsshowninSupplementalFig.3,RAW264.7
cellstransfectedwithmiR-17-5pandmiR-20amimicshavelower
levelsofROSexpressionandH
2
O
2
production.Asapositive
control,STAT3-targetedsiRNAsalsoinhibittheexpressionof
ROSandtheproductionofH
2
O
2
(SupplementalFig.3).
Inpreviouswork,we(18)andothers(23–25)haveshownthat
MDSCsfromtumor-bearingmicesignificantlysuppressTcells,
whereasMDSCsfromnaivemicedonotexpressSTAT3anddo
nothaveasignificantimmunosuppressiveactivity(10).Oneofthe
maincharacteristicsofMDSCsfromtumor-bearingmiceisthe
highproductionofROS(23–25),whichisamainfactorinMDSC-
mediatedTcellsuppression(23,25–28).Thus,weinvestigated
whethermiR-17-5pandmiR-20acouldaffecttheexpressionof
ROSandtheproductionofH
2
O
2
inMDSCs.MDSCscouldbe
successfullytransfectedwithmiR-17-5pandmiR-20amimicsor
miR-17-5pandmiR-20ainhibitors(Fig.3A,3B).Transfectionwas
alsoconfirmedusingfluorescentlylabeledmiRNAsimilarprobes
withhighertransfectionefficiency(92%)(SupplementalFig.4).As
showninFig.3Aand3B,thetransfectionofmiR-17-5pandmiR-
20amimicsreducestheexpressionofROSandtheproductionof
H
2
O
2
intheisolatedtumor-associatedMDSCsinthepresenceof
PMA.Furthermore,theeffectofmiR-17-5pandmiR-20aonthe
expressionofROSandproductionofH
2
O
2
canbeblockedby
inhibitorsofmiR-17-5pormiR-20a(Fig.3).However,intheab-
senceofPMAtheexpressionofROSandtheproductionofH
2
O
2
wasnotremarkablyalteredbytransfectionofmiR-17-5pandmiR-
20amimics(SupplementalFig.5).
FIGURE3.BothmiR-17-5pandmiR-20adecreaseMDSCsuppressioninvitro.AandB,miR-17-5pandmiR-20areducestheexpressionofROSandthe
productionofH
2
O
2
intumor-associatedMDSCs.Gr1
+
CD11b
+
MDSCswereisolatedfromspleensofCT26tumor-bearingmice(sixmice),transfected,and
stimulatedwithPMA(30ng/ml)for30min.ROSlevelsweremeasuredinGr1
+
CD11b
+
cellsbylabelingcellswiththeoxidation-sensitivedye2,7-
dichlorofluorescindiacetate.H
2
O
2
productionwasdetectedasdescribedinMaterialsandMethods.MDSCsweretransfected,respectively,withmiR-17-5p
mimics,miR-20amimics,miR-17-5pinhibitor,miR-20ainhibitor,ornegativecontrol(Ctr.).Geomean,geometricmean.CandD,miR-17-5pandmiR-20a
reducethesuppressivepotentialoftumor-associatedMDSCs.miR-17-5pandmiR-20ablocktheeffectoftumorassociatedMDSCsonVLP-specificCD4
andCD8Tlymphocytes.ThemiR17-5pandmiR-20ainhibitors,incontrast,promotethesuppressionoftumor-associatedMDSCsonVLP-specificCD4
andCD8Tlymphocytes.VLP-specificTlymphocytesweregeneratedandcoculturedwithMDSCsinresponsetoVLPsaccordingtotheprotocoldescribed
inMaterialsandMethods.CD4/CD8inC,VLP-specificCD4TcellsorCD8Tcells;AginC,VLP;suppressorinCandD,MDSCs.Cr1inC,
untransfectedMDSCs;Cr2inCandCtr.inD,negativecontroltransfectedMDSCs;M1inCandmiR-17-5pinD,MDSCstransfectedwithmiR-17-5p
mimics;M2inCandmiR-20ainD,MDSCstransfectedwithmiR-20amimics;SiinCandSTAT3siRNAinD,MDSCstransfectedwithSTAT3siRNA;I1
inCandmiR-17-5p.inhinD,MDSCstransfectedwithmiR-17-5pinhibitor;I2inCandmiR-20a.inhinD,MDSCstransfectedwithmiR-20ainhibitor.
InhibitionpercentageindicatestheconcentrationofIFN-ginthesupernatantsofacocultureofTcellsplusAgwithMDSCsdividedbytheconcentrationof
IFN-ginsupernatantsofacocultureofTcellsplusAgwithoutMDSCs.p,0.05,p,0.01.Thisisonerepresentativeofthreeexperiments.
4720miR-17-5pANDmiR-20aREGULATEMDSCs
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
WenexttestedtheeffectofmiR-17-5pandmiR-20aonMDSC-
mediatedsuppressioninAg-specificTcells.Consistentwithpre-
viousdatafromourlaboratory(18)andothers(23,25,27–29),
isolatedCD11b
+
Gr1
+
MDSCsfromspleensofmicewithCT-26
tumorswerefoundtoremarkablesuppressCD4andCD8Tcell
activation(Fig.3C).Importantly,bothmiR-17-5pandmiR-20a
remarkablyreducetheabilityofMDSCstosuppressAg-specific
CD4andCD8Tcells.Furthermore,theeffectofmiR-17-5pand
miR-20aonMDSCscanbeblockedbymiR-17-5pandmiR-20a
inhibitorsbutnotbynegativecontrols.AsshowninFig.3Cand
3D,theproductionofIFN-gbyAg-specificCD4andCD8Tcells
uponstimulationwithaspecificAgishigherincocultureofAg-
specificCD4orCD8TcellswithAginthepresenceofMDSCs
transfectedwithmiR-17-5pandmiR-20amimicsthaninthepres-
enceofcontrol-transfectedMDSCs(p,0.01).MDSCstransfected
withmiR-17-5pandmiR-20ainhibitorshaveasignificantlyhigher
abilitytosuppressIFN-gproductionbyAg-specificCD4andCD8
TcellsthancontroltransfectedMDSCs(p,0.05).Asapositive
control,STAT3-targetedsiRNAsalsoreduceMDSCsuppressionof
Ag-specificCD4andCD8Tcells.Thisresulthasalsobeenfurther
confirmedinMDSCsfromspleensofmicebearingLewislung
cancerand1D8ovariantumors(18).Thus,ourresultsdemonstrate
thatmiR-17-5pandmiR-20ainhibitthesuppressionoftumor-
associatedMDSCsonAg-specificCD4andCD8Tcells.
BothmiR-17-5pandmiR-20aalleviateMDSCsuppression
invivo
We(18)haveshownpreviouslythattheinfusionoftumor-
associatedMDSCspromotestumorgrowth,whereasaninfusion
ofGr1
+
CD11b
+
cellsfromthespleensofdisease-freemicedoes
notremarkablyaffectthetumorgrowthascomparedwithnon-
infusedmice.TheadoptivetransferofMDSCsfromtumor-
bearingmicecanabolishtheeffectofatumorvaccineonlung
metastasesinaSTAT3-dependentmanner(10).Toobservethe
effectofmiRNA-modifiedMDSCsonthetumorgrowth,weex-
aminedtheeffectthatMDSCswouldhaveontumorgrowthin
threedifferentmodels:vaccinatedWTmice,unvaccinatedWT
mice,andnudemice.Tumorgrowthwasfoundtobesignificantly
slowerinvaccinatedorunvaccinatedWTmiceinfusedwith
MDSCstransfectedwithmiR-17-5pormiR-20amimicsthanin
miceinfusedwithcontroltransfectedMDSCs(Fig.4,Supple-
mentalFig.6);conversely,tumorsinvaccinatedorunvaccinated
WTmiceinfusedwithMDSCstransfectedwithmiR-17-5pand
miR-20ainhibitorswerefoundtogrowfasterthanthoseinvac-
cinatedorunvaccinatedWTmiceinfusedwithcontroltransfected
MDSCs(Fig.4,SupplementalFig.6).Asapositivecontrol,a
STAT3-targetedsiRNAwasalsofoundtoreduceMDSCsup-
pressionintumor-bearingmice.Tumorgrowthwasfoundtobe
slowerinmiceinfusedwithMDSCstransfectedwithaSTAT3-
targetedsiRNAthaninthoseinfusedwithcontroltransfected
MDSCs(Fig.4A).Thus,miR-17-5pandmiR-20acanaffectMDSC
immunosuppressioninvivo.TheseeffectsaredependentonTcells,
asthesameresultswerenotobservedinnudemice(Supplemental
Fig.6)(10).
Tumor-associatedfactorsregulatetheexpressionofmiR-17-5p
andmiR-20a
Wenextinvestigatedwhethertumor-associatedfactorscanaffect
theexpressionofmiR-17-5pandmiR-20a.AsshowninFig.5A,
Gr1
+
CD11b
+
cellsisolatedfromthespleensofCT-26tumor-
bearingmicehavelowerlevelsofexpressionofbothpri-and
maturemiR-17-5pandmiR-20athandoGr1
+
CD11b
+
cellsfrom
thespleensofdisease-freemice(p,0.05).Theexpressionofpri-
andmaturemiR-17-5pandmiR-20awerealsofoundtobedown-
regulatedinbothmononuclearMDSC(CD11b
+
Ly6G
low
Ly6C
high
)
andpolymorphonuclearMDSC(CD11b
+
Ly6C
low
Ly6G
high
)sub-
setscomparedwiththoseisolatedfromdisease-freemice(Fig.5A,
p,0.05).SimilarresultswereobtainedforisolatedGr1
+
CD11b
+
MDSCs,mononuclearMDSCs,orpolymorphonuclearMDSCs
fromthespleensofmicebearingLewisand1D8tumors.These
cellsalsoexhibitlowerlevelsofmaturemiR-17-5pandmiR-20a
thandocellsfromdisease-freemice.
MDSCscanbegeneratedfromBMCsinvitrouponexposureto
tumor-associatedfactors(SupplementalFig.7)(25,30).Thus,we
observedtheexpressionofmiR-17-5pandmiR-20ainthepresence
oftumorsupernatantsatdifferenttimepoints.AsshowninFig.5B,
tumor-associatedfactorscanupregulatetheexpressionofmiR-17-
5pandmiR-20a,buttheexpressionofmiR-17-5pandmiR-20ais
downregulatedafter6hexposuretotumorsupernatants.Conversely,
theexpressionofSTAT3wasfoundtobeupregulatedafteratempo-
rarydownregulation.BMCsincubatedwithcontrolmediumhave
higherlevelsofmiR-17-5pandmiR-20abutlowerlevelsofSTAT3
thandoBMCsincubatedwithtumorsupernatants(Fig.5B).Because
CD11b
+
Gr1
+
MDSCscanbeinducedbytumorortumor-associated
factors(19),thisimpliesthathigherlevelsofSTAT3intumor-induced
MDSCscanbefromtumor-mediateddownregulationofmiR-17-5p
and20a.Thus,ourresultsdemonstratethattumor-associatedfactor
canregulatetheexpressionofmiR-17-5pandmiR-20a.
BothmiR-17-5pandmiR-20adecreasesuppressivepotentialof
CD11b
+
Ly6G
+
butnotCD11b
+
Ly6C
+
MDSCs
CD11b
+
Ly6G
low
Ly6C
high
monocyticMDSCsmediatesuppression
throughNOandarginase,whichdependsontheC/EBPbtran-
FIGURE4.BothmiR-17-5pandmiR-20aalleviatethesuppressionby
MDSCsinvivo.A,miR-17-5p,miR-20a,orSTAT3siRNAalleviatesthe
MDSCsuppression.B,miR-20aandmiR-17-5pinhibitorsexacerbate
MDSCsuppression.CD11b
+
Gr1
+
MDSCswereisolatedfromspleensof
micebearingCT-26coloncancer.MDSCswere,respectively,transfected
bymiR-17-5pmimics(miR-17-5p,A),miR-20amimics(miR-20a,A),and
STAT3siRNA(STAT3siRNA,A)ormiR-17-5pinhibitors(miR-17-5p
inh.,B)andmiR-20ainhibitors(miR-20ainh.,B)aswellascontrololigos
(Ctr.,A,B).InfusionwithdifferentlymodifiedMDSCswasperformed
accordingtotheprotocoldescribedinMaterialsandMethods.Tumors(six
micepergroup)weremeasuredwithacaliperandweighedbyweighing
scales.Errorbarsrepresentmean6SD.p,0.05,p,0.01.
TheJournalofImmunology4721
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
scriptionfactor(31),whereasCD11b
+
Ly6G
high
Ly6C
low
granulo-
cyticMDSC-mediatedsuppressionisviaROSandH
2
O
2
,which
dependsontheSTAT3transcriptionfactor(32).Becausetumor-
associatedfactorsregulatetheexpressionofmiR-17-5pandmiR-
20ainbothCD11b
+
Ly6G
+
andCD11b
+
Ly6C
+
subsetsandaffect
thelevelofSTAT3,wenextdetectedtheeffectofmiR-17-5pand
miR-20aonsuppressionoftheCD11b
+
Ly6G
+
andCD11b
+
Ly6C
+
MDSCs.AsshowninFig.6Aand6B,thetransfectionofmiR-17-
5pandmiR-20amimicsreducedtheexpressionofROSandthe
productionofH
2
O
2
intheisolatedtumor-associatedCD11b
+
Ly6G
+
butnotCD11b
+
Ly6C
+
MDSCs.TheroleofmiR-17-5pand
miR-20aintheexpressionofROSandproductionofH
2
O
2
by
CD11b
+
Ly6G
+
canbeblockedbymiR-17-5pormiR-20ainhib-
itors.Importantly,CD11b
+
Ly6G
+
MDSC-mediatedsuppression
onAg-specificTcellscouldbealleviatedbybothmiR-17-5pand
miR-20amimics,aswellasbyexacerbatedbymiR-17-5pand
miR-20ainhibitors.However,miR-17-5pandmiR-20aortheir
inhibitorsdidnotremarkablyaffecttheCD11b
+
Ly6C
+
MDSC-
mediatedsuppression(Fig.6C,6D).Thus,althoughmiR-17-5p
andmiR-20aalsoaffecttheexpressionofSTAT3inCD11b
+
Ly6C
+
MDSCs,STAT3-mediatedsuppressionisregulatedbymiR-
17-5pandmiR-20aonlyinCD11b
+
Ly6G
+
cells.
Discussion
Inthisstudy,wefoundthatmiR-17-5pandmiR-20acantargetthe
STAT339UTRtoinhibittheexpressionofSTAT3.Wealso
demonstratethattheexpressionofmiR-17-5pandmiR-20ais
regulatedbytumor-associatedfactors.Ourresultssuggestanovel
mechanismfortheregulationofMDSCsuppression.Whilecells
differentiateintoMDSCsinatumormicroenvironment,tumor-
associatedfactorsdownregulatetheexpressionofmiR-17-5pand
miR-20aandpromotetheSTAT3-associatedsuppressivefunction
ofMDSCs,whichcontainhigherlevelsofSTAT3andcomponents
oftheNADPHoxidase.OurresultsalsosuggestthatmiR-17-5p
andmiR-20acouldpotentiallybeusedastargetsinimmuno-
therapystrategiestoinhibitSTAT3expression.
OveractiveSTAT3isalsoassociatedwithanincreasedpro-
liferationandsurvivalofmyeloidprogenitors,possiblythrough
theupregulationofitstargetgenes,suchasBcl-x
L
,cyclinD1,c-
myc,andsurvivin(7),orviaS100A8andS100A9proteins(7).
Importantly,theconstitutiveupregulationofSTAT3transcription
factorspromotesthesuppressivefunctionintumor-bearingmice
(9).Indeed,theablationofSTAT3usingconditionalknockout
miceorselectiveinhibitorsdramaticallyreducestheexpansionof
MDSCsandimprovesTcellresponsesintumor-bearingmice(7,
23,29).Theinnateandadaptiveimmuneresponsesareregulated
bySTAT3signalingintumorcells(33).Inhibitionofdendriticcell
differentiationandaccumulationofMDSCsincancerisregulated
bySTAT3-mediatedS100A9protein(34).Thus,STAT3hasbeen
suggestedasatargetofimmunotherapyagainstdiseases.Indeed,
ourresultsdemonstratethattumorgrowthinmiceinfusedwith
MDSCstransfectedwithmiR-17-5pormiR-20aisslowerthanin
miceinfusedbycontroltransfectedMDSCs.TheselectiveSTAT3
inhibitorJSI-124(35)alsodownregulatesSTAT3activityin
MDSCsanddramaticallyreducestheirpresenceintumor-bearing
mice(7,29).Thetreatmentoftumor-bearingmicewiththeSTAT3
inhibitorJSI-124substantiallyenhancestheeffectofcancerim-
munotherapyandanti-tumorimmuneresponses(2,7,29,36).
FIGURE5.Tumor-associatedfactorsregulatetheexpressionofmiR-17-5pandmir-20a.A,Expressionofpri-miR-17-5p,pri-miR-20a,STAT3,and
GAPDHintumor-associatedCD11b
+
Gr1
+
MDSCsandexpressionofmaturemiR-17-5p,maturemiR-20a,andSTAT3inCD11b
+
Gr1
+
(Ly6G
+
Ly6C
+
)
MDSCs(P1),CD11b
+
Ly6G
+
MDSCs(P2),andCD11b
+
Ly6C
+
MDSCs(P3)isolatedfromspleensofCT-26tumor-bearingmice(CT26),Lewistumor-
bearingmice(Lewis),and1D8tumor-bearingmice(1D8).IsolatedMDSCsfromspleensofCT-26coloncancer-bearingmice,Lewislungcancer-bearing
mice,and1D8ovariancancer-bearingmicewerefoundtohavealowerexpressionofmiR-17-5pandmiR-20a.p,0.05.B,Tumor-associatedfactors
regulatedthetranscriptionofmiR-17-5p,miR-20a,andSTAT3inBMCs.BMCswerecoculturedwithCT-26tumorcellsinatranswellplate.Thelevelsof
maturemiR-17-5pandmaturemiR-20aandSTAT3mRNAweredetectedusingqRT-PCRattheindicatedtimesaccordingtotheprotocoldescribedin
MaterialsandMethods.C,Tumor-associatedfactorsregulatetheexpressionofSTAT3andp47
phox
inBMCs.BMCswerecoculturedwithCT-26tumor
cellsinatranswellplate.ThelevelsofSTAT3proteinwereassayedattheindicatedtimeusingWesternblotting.
4722miR-17-5pANDmiR-20aREGULATEMDSCs
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
SilencingofSTAT3signalingcanalsoberealizedinvitroand
invivousingantisenseoligonucleotides(33),dominant-negative
STAT3(37),orphosphotyrosylpeptides(39).STAT3activitycan
alsoberegulatedbyS100A9protein(34).Comparedtothese
strategies,aremarkableadvantageofusingmiRNAapproachesis
thattherapeuticbenefitcanberealizedbycorrectingmiRNA
deficienciesbyeitherantagonizingorrestoringmiRNAfunction
(39).AnotheradvantageofusingmiRNAisitsabilitytocon-
currentlytargetmultipleeffectorsofpathways(40).
TherearedifferentsubsetsofMDSCsintumor-bearingmice,
thatis,CD11b
+
Ly6G
low
Ly6C
high
monocyticMDSCsandCD11b
+
Ly6G
high
Ly6C
low
granulocyticMDSCs(32).GranulocyticMDSCs
expressahighlevelofROSandverylittleNO,whereasmonocytic
MDSCshaveverylittleROSbutahighlevelofNO(32).CD11b
+
Ly6G
low
Ly6C
high
monocyticMDSCsmediatesuppressionthrough
NOandarginase,whichdependsontheC/EBPbtranscription
factor(31),whereasCD11b
+
Ly6G
high
Ly6C
low
granulocyticMDSC
suppressionismediatedviaROSandH
2
O
2
,whichdependsonthe
STAT3transcriptionfactor(32).Indeed,ourresultsshowthatonly
granulocyticMDSC-mediatedsuppressionisregulatedbymiR-17-
5pandmiR-20abytargetingSTAT3.miR-17-5pandmiR-20ado
notremarkablyaffectthemonocyticMDSC-mediatedsuppression,
althoughthemiR-17-5pandmiR-20aalsodownregulatetheex-
pressionofSTAT3inthesecells.
Usingbioinformatics,Westernblottinganalysis,andreporter
assays,multiplemiR-17-92targetshavebeenidentified.Eachof
theseisproposedtocontributetoaspecificfunctionalreadoutof
miR-17-92(22).WehavedemonstratedthatmiR-17-5pandmiR-
20ainthemiR-17-92cluster,whichishighlyconservedbetween
miceandhumans,canblockthesuppressivefunctionofMDSCs
byloweringSTAT3expression.Notably,thetype2skewingtumor
environmentalsodownregulatesmiR-17-92expressioninTcells,
therebydiminishingthepersistenceoftumor-specificTcellsand
tumorcontrol(41).FunctionaldiversityinthemiR-17-92cluster
mayresultfromthedifferenttargetmRNAssubjectedtomiR-17-
92post-transcriptionalsilencingindifferentcelltypesand/or
differentdevelopmentalorphysiologicalcontexts.
Disclosures
Theauthorshavenofinancialconflictsofinterest.
FIGURE6.BothmiR-17-5pandmiR-20adecreaseCD11b
+
Ly6G
+
butnotCD11b
+
Ly6C
+
suppressioninvitro.AandB,miR-17-5pandmiR-20areduce
theexpressionofROSandtheproductionofH
2
O
2
intumor-associatedCD11b
+
Ly6G
+
MDSCsbutnotinCD11b
+
Ly6C
+
MDSCs.CD11b
+
Ly6G
+
MDSCs
andCD11b
+
Ly6C
+
MDSCsweresortedfromspleensofCT-26tumor-bearingmice(sixmice),transfected,andthenstimulatedwithPMA(30ng/ml)for30
min.ROSlevelsweremeasuredinCD11b
+
Ly6G
+
andCD11b
+
Ly6C
+
MDSCsbylabelingcellswiththeoxidation-sensitivedye2,7-dichlorofluorescin
diacetate.H
2
O
2
productionwasdetectedasdescribedinMaterialsandMethods.CD11b
+
Ly6G
+
andCD11b
+
Ly6C
+
MDSCsweretransfected,respectively,
withmiR-17-5pmimics,miR-20amimics,miR-17-5pinhibitor,miR-20ainhibitor,ornegativecontrol.Geomean,geometricmean.CandD,miR-17-5p
andmiR-20areducethesuppressivepotentialoftumor-associatedCD11b
+
Ly6G
+
butnotCD11b
+
Ly6C
+
MDSCs.miR-17-5pandmiR-20ablocktheeffect
oftumor-associatedCD11b
+
Ly6G
+
butnotCD11b
+
Ly6C
+
MDSCsonVLP-specificCD4(C)andCD8Tlymphocytes(D).ThemiR17-5pandmiR-20a
inhibitors,incontrast,promotethesuppressionoftumor-associatedCD11b
+
Ly6G
+
butnotCD11b
+
Ly6C
+
MDSCsonVLP-specificCD4(C)andCD8T
lymphocytes(D).VLP-specificTlymphocytesweregeneratedandcoculturedwithCD11b
+
Ly6G
+
orCD11b
+
Ly6C
+
MDSCsinresponsetoVLPs
accordingtotheprotocoldescribedinMaterialsandMethods.CD4inCandCD8inD,VLP-specificCD4Tcells(C)orCD8Tcells(D);AginCandD,
VLP;suppressorinCandD,CD11b
+
Ly6G
+
orCD11b
+
Ly6C
+
MDSCs;Cr1inCandD,untransfectedCD11b
+
Ly6G
+
orCD11b
+
Ly6C
+
;Cr2inCandD,
negativecontroltransfectedCD11b
+
Ly6G
+
orCD11b
+
Ly6C
+
MDSCs;M1inCandD,CD11b
+
Ly6G
+
orCD11b
+
Ly6C
+
MDSCstransfectedwithmiR-17-
5pmimics;M2inCandD,CD11b
+
Ly6G
+
orCD11b
+
Ly6C
+
MDSCstransfectedwithmiR-20amimics;I1inCandD,CD11b
+
Ly6G
+
orCD11b
+
Ly6C
+
MDSCstransfectedwithmiR-17-5pinhibitor;I2inCandD,CD11b
+
Ly6G
+
orCD11b
+
Ly6C
+
MDSCstransfectedwithmiR-20ainhibitor.p,0.01.
TheJournalofImmunology4723
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
References
1.Ostrand-Rosenberg,S.,andP.Sinha.2009.Myeloid-derivedsuppressorcells:
linkinginflammationandcancer.J.Immunol.182:4499–4506.
2.Fujita,M.,X.Zhu,K.Sasaki,R.Ueda,K.L.Low,I.F.Pollack,andH.Okada.
2008.InhibitionofSTAT3promotestheefficacyofadoptivetransfertherapy
usingtype-1CTLsbymodulationoftheimmunologicalmicroenvironmentin
amurineintracranialglioma.J.Immunol.180:2089–2098.
3.Mirza,N.,M.Fishman,I.Fricke,M.Dunn,A.M.Neuger,T.J.Frost,
R.M.Lush,S.Antonia,andD.I.Gabrilovich.2006.All-trans-retinoicacid
improvesdifferentiationofmyeloidcellsandimmuneresponseincancer
patients.CancerRes.66:9299–9307.
4.Srivastava,M.K.,J.J.Bosch,J.A.Thompson,B.R.Ksander,M.J.Edelman,
andS.Ostrand-Rosenberg.2008.Lungcancerpatients’CD4
+
Tcellsareacti-
vatedinvitrobyMHCIIcell-basedvaccinesdespitethepresenceofmyeloid-
derivedsuppressorcells.CancerImmunol.Immunother.57:1493–1504.
5.Zea,A.H.,P.C.Rodriguez,M.B.Atkins,C.Hernandez,S.Signoretti,
J.Zabaleta,D.McDermott,D.Quiceno,A.Youmans,A.O’Neill,etal.2005.
Arginase-producingmyeloidsuppressorcellsinrenalcellcarcinomapatients:
amechanismoftumorevasion.CancerRes.65:3044–3048.
6.Movahedi,K.,M.Guilliams,J.VandenBossche,R.VandenBergh,
C.Gysemans,A.Beschin,P.DeBaetselier,andJ.A.VanGinderachter.2008.
Identificationofdiscretetumor-inducedmyeloid-derivedsuppressorcellsub-
populationswithdistinctTcell-suppressiveactivity.Blood111:4233–4244.
7.Nefedova,Y.,andD.I.Gabrilovich.2007.TargetingofJak/STATpathwayin
antigenpresentingcellsincancer.Curr.CancerDrugTargets7:71–77.
8.Kortylewski,M.,M.Kujawski,T.Wang,S.Wei,S.Zhang,S.Pilon-Thomas,
G.Niu,H.Kay,J.Mule,W.G.Kerr,etal.2005.InhibitingStat3signalinginthe
hematopoieticsystemelicitsmulticomponentantitumorimmunity.Nat.Med.11:
1314–1321.
9.Nefedova,Y.,S.Nagaraj,A.Rosenbauer,C.Muro-Cacho,S.M.Sebti,and
D.I.Gabrilovich.2005.Regulationofdendriticcelldifferentiationandantitu-
morimmuneresponseincancerbypharmacologic-selectiveinhibitionofthe
Janus-activatedkinase2/signaltransducersandactivatorsoftranscription3
pathway.CancerRes.65:9525–9535.
10.Chalmin,F.,S.Ladoire,G.Mignot,J.Vincent,M.Bruchard,J.P.Remy-Martin,
W.Boireau,A.Rouleau,B.Simon,D.Lanneau,etal.2010.Membrane-
associatedHsp72fromtumor-derivedexosomesmediatesSTAT3-dependent
immunosuppressivefunctionofmouseandhumanmyeloid-derivedsuppressor
cells.J.Clin.Invest.120:457–471.
11.Lai,E.C.2002.MicroRNAsarecomplementaryto39UTRsequencemotifsthat
mediatenegativepost-transcriptionalregulation.Nat.Genet.30:363–364.
12.Chen,C.Z.,andH.F.Lodish.2005.MicroRNAsasregulatorsofmammalian
hematopoiesis.Semin.Immunol.17:155–164.
13.Tsitsiou,E.,andM.A.Lindsay.2009.microRNAsandtheimmuneresponse.
Curr.Opin.Pharmacol.9:514–520.
14.Bazzoni,F.,M.Rossato,M.Fabbri,D.Gaudiosi,M.Mirolo,L.Mori,
N.Tamassia,A.Mantovani,M.A.Cassatella,andM.Locati.2009.Induction
andregulatoryfunctionofmiR-9inhumanmonocytesandneutrophilsexposed
toproinflammatorysignals.Proc.Natl.Acad.Sci.USA106:5282–5287.
15.Taganov,K.D.,M.P.Boldin,K.J.Chang,andD.Baltimore.2006.NF-kB-
dependentinductionofmicroRNAmiR-146,aninhibitortargetedtosignaling
proteinsofinnateimmuneresponses.Proc.Natl.Acad.Sci.USA103:12481–12486.
16.Tili,E.,J.J.Michaille,A.Cimino,S.Costinean,C.D.Dumitru,B.Adair,
M.Fabbri,H.Alder,C.G.Liu,G.A.Calin,andC.M.Croce.2007.Modulation
ofmiR-155andmiR-125blevelsfollowinglipopolysaccharide/TNF-astimula-
tionandtheirpossiblerolesinregulatingtheresponsetoendotoxinshock.J.
Immunol.179:5082–5089.
17.Ceppi,M.,P.M.Pereira,I.Dunand-Sauthier,E.Barras,W.Reith,M.A.Santos,
andP.Pierre.2009.MicroRNA-155modulatestheinterleukin-1signaling
pathwayinactivatedhumanmonocyte-deriveddendriticcells.Proc.Natl.Acad.
Sci.USA106:2735–2740.
18.Yang,R.,Z.Cai,Y.Zhang,W.H.t.Yutzy,K.F.Roby,andR.B.Roden.2006.
CD80inimmunesuppressionbymouseovariancarcinoma-associatedGr-1
+
CD11b
+
myeloidcells.CancerRes.66:6807–6815.
19.Sinha,P.,V.K.Clements,A.M.Fulton,andS.Ostrand-Rosenberg.2007.
ProstaglandinE
2
promotestumorprogressionbyinducingmyeloid-derived
suppressorcells.CancerRes.67:4507–4513.
20.Zhang,Z.,B.Zeng,Z.Zhang,G.Jiao,H.Li,Z.Jing,J.Ouyang,X.Yuan,
L.Chai,Y.Che,etal.2009.Suppressorofcytokinesignaling3promotesbone
marrowcellstodifferentiateintoCD8
+
Tlymphocytesinlungtissueviaup-
regulatingNotch1expression.CancerRes.69:1578–1586.
21.Corzo,C.A.,M.J.Cotter,P.Cheng,F.Cheng,S.Kusmartsev,E.Sotomayor,
T.Padhya,T.V.McCaffrey,J.C.McCaffrey,andD.I.Gabrilovich.2009.
Mechanismregulatingreactiveoxygenspeciesintumor-inducedmyeloid-
derivedsuppressorcells.J.Immunol.182:5693–5701.
22.Olive,V.,I.Jiang,andL.He.2010.mir-17-92,aclusterofmiRNAsinthemidst
ofthecancernetwork.Int.J.Biochem.CellBiol.42:1348–1354.
23.Kusmartsev,S.,andD.I.Gabrilovich.2003.Inhibitionofmyeloidcelldifferen-
tiationincancer:theroleofreactiveoxygenspecies.J.Leukoc.Biol.74:186–196.
24.Kusmartsev,S.,Y.Nefedova,D.Yoder,andD.I.Gabrilovich.2004.Antigen-
specificinhibitionofCD8
+
Tcellresponsebyimmaturemyeloidcellsincancer
ismediatedbyreactiveoxygenspecies.J.Immunol.172:989–999.
25.Sinha,P.,V.K.Clements,andS.Ostrand-Rosenberg.2005.Reductionof
myeloid-derivedsuppressorcellsandinductionofM1macrophagesfacilitatethe
rejectionofestablishedmetastaticdisease.J.Immunol.174:636–645.
26.Nagaraj,S.,K.Gupta,V.Pisarev,L.Kinarsky,S.Sherman,L.Kang,D.L.Herber,
J.Schneck,andD.I.Gabrilovich.2007.Alteredrecognitionofantigenis
amechanismofCD8
+
Tcelltoleranceincancer.Nat.Med.13:828–835.
27.Markiewski,M.M.,R.A.DeAngelis,F.Benencia,S.K.Ricklin-
Lichtsteiner,A.Koutoulaki,C.Gerard,G.Coukos,andJ.D.Lambris.2008.
Modulationoftheantitumorimmuneresponsebycomplement.Nat.Immu-
nol.9:1225–1235.
28.Ando,T.,K.Mimura,C.C.Johansson,M.G.Hanson,D.Mougiakakos,
C.Larsson,T.MartinsdaPalma,D.Sakurai,H.Norell,M.Li,etal.2008.
TransductionwiththeantioxidantenzymecatalaseprotectshumanTcells
againstoxidativestress.J.Immunol.181:8382–8390.
29.Nefedova,Y.,M.Fishman,S.Sherman,X.Wang,A.A.Beg,and
D.I.Gabrilovich.2007.Mechanismofall-transretinoicacideffectontumor-
associatedmyeloid-derivedsuppressorcells.CancerRes.67:11021–11028.
30.Liu,Q.,M.Zhang,X.Jiang,Z.Zhang,L.Dai,S.Min,X.Wu,Q.He,J.Liu,
Y.Zhang,etal.2011.miR-223suppressesdifferentiationoftumor-induced
CD11b+Gr1+myeloidderivedsuppressorcellsfrombonemarrowcells.Int.J.
CancerDOI:10.1002/ijc.25921.
31.Marigo,I.,E.Bosio,S.Solito,C.Mesa,A.Fernandez,L.Dolcetti,S.Ugel,
N.Sonda,S.Bicciato,E.Falisi,etal.2010.Tumor-inducedtoleranceandim-
munesuppressiondependontheC/EBPbetatranscriptionfactor.Immunity32:
790–802.
32.Youn,J.I.,S.Nagaraj,M.Collazo,andD.I.Gabrilovich.2008.Subsetsof
myeloid-derivedsuppressorcellsintumor-bearingmice.J.Immunol.181:5791–
5802.
33.Wang,T.,G.Niu,M.Kortylewski,L.Burdelya,K.Shain,S.Zhang,
R.Bhattacharya,D.Gabrilovich,R.Heller,D.Coppola,etal.2004.Regulation
oftheinnateandadaptiveimmuneresponsesbyStat-3signalingintumorcells.
Nat.Med.10:48–54.
34.Cheng,P.,C.A.Corzo,N.Luetteke,B.Yu,S.Nagaraj,M.M.Bui,M.Ortiz,
W.Nacken,C.Sorg,T.Vogl,etal.2008.Inhibitionofdendriticcelldifferen-
tiationandaccumulationofmyeloid-derivedsuppressorcellsincancerisreg-
ulatedbyS100A9protein.J.Exp.Med.205:2235–2249.
35.Blaskovich,M.A.,J.Sun,A.Cantor,J.Turkson,R.Jove,andS.M.Sebti.2003.
DiscoveryofJSI-124(cucurbitacinI),aselectiveJanuskinase/signaltransducer
andactivatoroftranscription3signalingpathwayinhibitorwithpotentantitumor
activityagainsthumanandmurinecancercellsinmice.CancerRes.63:1270–
1279.
36.Molavi,O.,Z.Ma,S.Hamdy,R.Lai,A.Lavasanifar,andJ.Samuel.2008.
SynergisticantitumoreffectsofCpGoligodeoxynucleotideandSTAT3in-
hibitoryagentJSI-124inamousemelanomatumormodel.Immunol.CellBiol.
86:506–514.
37.Niu,G.,R.Heller,R.Catlett-Falcone,D.Coppola,M.Jaroszeski,W.Dalton,
R.Jove,andH.Yu.1999.Genetherapywithdominant-negativeStat3suppresses
growthofthemurinemelanomaB16tumorinvivo.CancerRes.59:5059–5063.
38.Turkson,J.,D.Ryan,J.S.Kim,Y.Zhang,Z.Chen,E.Haura,A.Laudano,
S.Sebti,A.D.Hamilton,andR.Jove.2001.Phosphotyrosylpeptidesblock
Stat3-mediatedDNAbindingactivity,generegulation,andcelltransformation.
J.Biol.Chem.276:45443–45455.
39.Bader,A.G.,D.Brown,andM.Winkler.2010.ThepromiseofmicroRNA
replacementtherapy.CancerRes.70:7027–7030.
40.Garzon,R.,G.Marcucci,andC.M.Croce.2010.TargetingmicroRNAsincancer:
rationale,strategiesandchallenges.Nat.Rev.DrugDiscov.9:775–789.
41.Sasaki,K.,G.Kohanbash,A.Hoji,R.Ueda,H.A.McDonald,T.A.Reinhart,
J.Martinson,M.T.Lotze,F.M.Marincola,E.Wang,etal.2010.miR-17-92
expressionindifferentiatedTcells:implicationsforcancerimmunotherapy.J.
Transl.Med.8:17.
4724miR-17-5pANDmiR-20aREGULATEMDSCs
byguestonMarch4,2015
http://www.jimmunol.org/
Downloadedfrom
|
|
