Developing aptamer probes for acute myelogenous leukemia detection and surface protein biomarker discovery

MISSING IMAGE

Material Information

Title:
Developing aptamer probes for acute myelogenous leukemia detection and surface protein biomarker discovery
Abbreviated Title:
Journal of Hematology & Oncology
Physical Description:
Mixed Material
Creator:
Mingli Yang
Guohua Jiang
Wenjing Li
Kai Qiu
Min Zhang
Christopher M Carter
Samer Z Al-Quran
Ying Li
Publisher:
Journal of Hematology & Oncology
Publication Date:

Notes

Abstract:
Background: The majority of patients with acute myelogenous leukemia (AML) still die of their disease. In order to improve survival rates in AML patients, new strategies are necessary to discover biomarkers for the detection and targeted therapy of AML. One of the advantages of the aptamer-based technology is the unique cell-based selection process, which allows us to efficiently select for cell-specific aptamers without knowing which target molecules are present on the cell surface. Methods: The NB4 AML cell line was used as the target cell population for selecting single stranded DNA aptamers. After determining the affinity of selected aptamers to leukocytes, the aptamers were used to phenotype human bone marrow leukocytes and AML cells in clinical specimens. Then a biotin-labelled aptamer was used to enrich and identify its target surface protein. Results: Three new aptamers were characterized from the selected aptamer pools (JH6, JH19, and K19). All of them can selectively recognize myeloid cells with Kd in the low nanomole range (2.77 to 12.37 nM). The target of the biotin-labelled K19 aptamer probe was identified as Siglec-5, a surface membrane protein in low abundance whose expression can serve as a biomarker of granulocytic maturation and be used to phenotype AML. More importantly, Siglec-5 expression can be used to detect low concentrations of AML cells in human bone marrow specimens, and functions as a potential target for leukemic therapy. Conclusions: We have demonstrated a pipeline approach for developing single stranded DNA aptamer probes, phenotyping AML cells in clinical specimens, and then identifying the aptamer-recognized target protein. The developed aptamer probes and identified Siglec-5 protein may potentially be used for leukemic cell detection and therapy in our future clinical practice.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
System ID:
AA00020067:00001

Full Text

PAGE 1

RESEARCHOpenAccessDevelopingaptamerprobesforacute myelogenousleukemiadetectionandsurface proteinbiomarkerdiscoveryMingliYang†,GuohuaJiang†,WenjingLi,KaiQiu,MinZhang,ChristopherMCarter,SamerZAl-QuranandYingLi*AbstractBackground: Themajorityofpatientswithacutemyelogenousleukemia(AML)stilldieoftheirdisease.Inorderto improvesurvivalratesinAMLpatients,newstrategiesarenecessarytodiscoverbiomarkersforthedetectionand targetedtherapyofAML.Oneoftheadvantagesoftheaptamer-basedtechnologyistheuniquecell-based selectionprocess,whichallowsustoefficientlyselectforcell-specificaptamerswithoutknowingwhichtarget moleculesarepresentonthecellsurface. Methods: TheNB4AMLcelllinewasusedasthetargetcellpopulationforselectingsinglestrandedDNA aptamers.Afterdeterminingtheaffinityofselectedaptamerstoleukocytes,theaptamerswereusedtophenotype humanbonemarrowleukocytesandAMLcellsinclinicalspecimens.Thenabiotin-labelledaptamerwasusedto enrichandidentifyitstargetsurfaceprotein. Results: Threenewaptamerswerecharacterizedfromtheselectedaptamerpools(JH6,JH19,andK19).Allofthem canselectivelyrecognizemyeloidcellswithKdinthelownanomolerange(2.77to12.37nM).Thetargetofthe biotin-labelledK19aptamerprobewasidentifiedasSiglec-5,asurfacemembraneproteininlowabundancewhose expressioncanserveasabiomarkerofgranulocyticmaturationandbeusedtophenotypeAML.Moreimportantly, Siglec-5expressioncanbeusedtodetectlowconcentrationsofAMLcellsinhumanbonemarrowspecimens,and functionsasapotentialtargetforleukemictherapy. Conclusions: WehavedemonstratedapipelineapproachfordevelopingsinglestrandedDNAaptamerprobes, phenotypingAMLcellsinclinicalspecimens,andthenidentifyingtheaptamer-recognizedtargetprotein.The developedaptamerprobesandidentifiedSiglec-5proteinmaypotentiallybeusedforleukemiccelldetectionand therapyinourfutureclinicalpractice. Keywords: Acutemyeloidleukemia,Aptamer,Biomarker,Cell-SELEX,Siglec-5IntroductionAcutemyelogenousleukemia(AML)isaheterogeneous groupofmalignanthematopoieticneoplasmsderived fromhematopoieticstemcellspostulatedtoarisedueto mutationsofgenesthatregulatedtheorderlyproliferation,differentiation,andmaturationofhematopoietic cells.Inthepasttwodecades,scientificadvancesutilizingmoleculartechniquesandcytogeneticdetectionhave yieldednewinsightsintothegeneticandbiologic featuresofacuteleukemia.Despitetheseadvances,the majorityofpatientswhosufferedfromAMLstilldiedof theirdisease[1-3].Withtheexceptionofasubtypeof AML,AMLM3(i.e.acutepromyelocyticleukemia, APL),wehavenotyetsucceededintranslatingour scientificdiscoveriesintomoreeffectivetreatmentsfor themajorityofAMLpatients.Whiletherapeuticintensification,improvedsupportivecare,andbonemarrowtransplantationhaveledtogradualimprovements ofoutcomeinchildrenandyoungeradultswithAML, theoverallsurvivalrateapproaches50%.Inolderindividuals(>55-60years)andinsecondaryAMLpatients, theoutlookismoredismalwithoverallsurvivalratesof *Correspondence: liying@pathology.ufl.edu†Equalcontributors UF/ShandsMedicalLaboratoryatRockyPoint,480035thDrive,Gainesville, FL32608,USA JOURNAL OF HEMATOLOGY & ONCOLOGY 2014Yangetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedtheoriginalworkisproperlycited.TheCreativeCommonsPublicDomainDedication waiver(http://creativecommons.org/publicdomain/zero/1.0/)appliestothedatamadeavailableinthisarticle,unlessotherwise stated.Yang etal.JournalofHematology&Oncology 2014, 7 :5 http://www.jhoonline.org/content/7/1/5

PAGE 2

10-15%[4],typicallyattributedtoanincreaseinunfavorablecytogeneticfeatures. Currently,immunophenotypingviaimmunohistochemistryandflowcytometricanalysisplaysapivotalrolein thedetectionanddiagnosisofAML.However,thesurface biomarkerscurrentlyusedforimmunophenotypingAML areadaptedfromtheadvancementofimmunologyresearchinthelastseveraldecadesinsteadofbeingspecificallydevelopedforleukemiccelldetection.Whilewe usethesebiomarkersinourdailypracticetoclassify leukemiaintomyeloidorlymphoidlineages,theyneitheridentifythemoleculareventsunderlyingtheneoplasticprocessesnorprovideadequateinsightintothe aggressivenessorprognosisofthesediseases.Asaconsequence,manydifferentleukemicvariantsbecome groupedtogetherunderthesamenameduetothelack ofadequatebiomarkersforeffectivestratification,despitenotrepresentingthesameexactdiseasebynature orbehavior[5,6].Inaddition,whenonlyasmallnumber ofleukemiccellsarepresentitisoftendifficult,ifnot impossible,todeterminethediseasestatusduetotheir immunophenotypicsimilaritytonormalcells.Thisis oftenthecasefollowingchemotherapywhenminimal residualleukemiaispresen t.Therefore,anewstrategy usingmolecularaptamersisenvisionedtodiscoverbiomarkersandapplytheminclinicalpracticetoimprove therapeuticefficiencyandthesurvivalrateofAML patients. MolecularaptamersconsistofsinglestrandedDNAor RNAthatcanrecognizetargetproteins,peptides,and othersmallmolecules.ThroughaprocesscalledSELEX (SystematicEvolutionofLigandsbyExponentialenrichment)[7-11],DNAorRNAaptamersspecificforaknown proteinofinterestcanbeselectedfromarandompoolof oligonucleotidesequences,andthenusedasdiagnostic andtherapeuticreagents[12].Traditionally,thetargets (mostoftenproteins),inmostinstances,havetobeidentifiedfirstbeforespecificmolecularprobes,including monoclonalantibodies,canbedeveloped.However,using livecellsfromleukemiccelllines,weestablishedaunique cell-basedselectionprocess(Cell-SELEX)thatallowsfor theselectionofaptamersthatcanrecognizeliveleukemic cellsfrompatients[13].Mostimportantly,theCellSELEXmethodallowsustoselectagroupofcell-specific aptamersinarelativelyshorttimeperiodandselected aptamerscanreadilybetestedandverifiedinclinical specimens,withoutknowingwhichtargetmoleculesare presentonthecellsurface.Thusfar,whilemanyDNAor RNAaptamershavebeenselectedagainstvarioustypesof cells,afewsurfaceproteinstargetedbyindividualaptamersofinterestwereidentified[14],whichisprobably duetothetechnicalchallengeinpurificationandidentificationoflow-abundancemembraneproteins[14].The fewreportedproteinsidentifiedthroughindividual aptamerprobesincludepigpenfromtheratendothelial celllineYPEN-1[15],Tenascin-CofU251glioblastoma cells[16],immunoglobulinheavymuchaininBurkitt ’ s lymphomacells[17]andproteintyrosinekinase-7(PTK7) onCCRF-CEMT-cellacutelymphoblasticleukemic cells[18]. InordertodevelopbiomarkersforAML,weintended todesignapipelineapproachforbiomarkerdiscovery: 1)ToemploytheCell-SELEXtechniquetoselectfor DNAaptamerprobesagainstliveleukemiccells;2)To testselectedaptamersbyphenotypingnormalhuman bonemarrowcellsorleukemiccellsinclinicalspecimens;3)Toidentifytargetproteinsonleukemiccell surfaceswithmeaningfulmolecularsignaturesasdemonstratedwiththeaptamers.Inthisstudy,weselected aptamersagainstNB4AMLcells.Moreimportantly,with biotin-labelledaptamerswewereabletodemonstratethat thetargetproteinforoneofthenewaptamerswasa memberofthesialic-acid-bindingimmunoglobulin-like lectins(Siglecs).ThentheaptamerrecognizingSiglec-5 wasusedtodetectsmallnumbersofAMLcellsinhuman bonemarrowspecimens.MaterialandmethodsCellcultureandReagentsNB4andHL60humanleukemiccelllineswereobtained fromATCC(AmericanTypeCultureCollection,Manassas, Virginia)andwereculturedinRPMI1640medium (ThermoScientificHyClone,SouthLogan,Utah)supplementedwith10%fetalbovineserum(FBS)(heatinactivated,ThermoScientificHyClone,SouthLogan,Utah), andantibiotics(100units/mlpenicillin-Streptomycinfrom FisherBioReagents,Fairlawn,NJ).Beforebindingtoaptamers,cellswerewashedwithphosphate-bufferedsaline (PBS).Thebufferusedforaptamerbindingandselection waspreparedbyadding4.5g/Lglucose,5mMMgCl2, 0.1mg/mlyeasttRNA(FisherBioReagents,Fairlawn,NJ, USA)and1mg/mlBovineSerumAlbumin(BSA)(Fisher BioReagents)intoPhosphatebufferedsaline(PBS)[13]. Theusedfluorochromesincludeallophycocyanin(APC), fluoresceinisothiocyanate(FITC),phycoerythrin(PE),and peridininchlorophyllprotein(PerCP).Cell-SELEXproceduresforaptamerselectionHPLCpurifiedlibraries(Sigma-Aldrich,St.Louis, MO)containasegmentofrandomizedsequenceof 50nucleotides(nt)flankedbyPCRprimerhybridizationsites(5 -GACGCTTACTCAGGTGTGACTCG50nt-CGAAGGACGCAGATGAAGTCTC-3 ).Biotinylated PCR-primerswereusedinthePCRreactionsforthe synthesisofbiotin-labelledDNAmolecules.Afterheat denaturationat95Cfor5min,thedenaturedDNAs wereplacedoniceimmediatelyandthebiotinylated strandswereseparatedfr omthecomplementstrandsYang etal.JournalofHematology&Oncology 2014, 7 :5Page2of14 http://www.jhoonline.org/content/7/1/5

PAGE 3

bystreptavidin-coatedmagneticbeads(ThermoScientific Pierce,Pittsburgh,PA). Theselectionprocesseswereperformedsimilarlyas describedbefore[13,19].20nmolesofsynthesizedsingle strandedDNApoolweredissolvedin1mlofbinding bufferandusedforthefirstroundselection.100 … 200 pmolesofpooldissolvedin400 Lbindingbufferwere usedfortheremainingroundsofselection.TheDNA poolsweredenaturedbyheatingat95Cfor5minand placedonicefor10minbeforebinding.Thesingle strandedDNApoolwasincubatedwith1 … 2106target (NB4)cellsonicefor1hr.Afterwashing,thebound DNAswereelutedbyheatingat95Cfor5minin 400 Lof2mMTris … HClbuffer(pH8.0).Theeluted DNAswereamplifiedbyPCR(Taq-polymeraseand dNTP ’ swereobtainedfromFisher(Fairlawn,NJ))for 25 … 30cyclesof0.5minat94C,0.5minat60C,and 0.5minat72C,followedby5minat72C.The amplifiedsenseDNApoolusedforthenextround selectionwasseparatedfromantisensePCRproductsby streptavidin-coatedmagneticbeads.After8 … 20rounds ofselection,biotinylatedselectedDNApoolsorsingle strandedDNAcontrolwereboundtoNB4cellsonice for30min,washedtwicewithcoldPBSandthenincubatedwith5 LPE-streptavidin(0.5mg/ml,Becton Dickinson,NJ)onicefor30min.Afterwashedtwice withPBS,theNB4cellsweresubjecttoflowcytometry analysis.Aftertheselectedpoolshowedsignificantly higherfluorescencesignalsthantheunselectedone(see Additionalfile1:FigureS1),selectedpoolwasPCRamplifiedusingunlabeledprimers,clonedintopPCRScriptAmpSK(+)vectorwithPCR-ScriptAmpCloning Kit(AgilentTechnologies,SanDiego,CA)andtransformedintoEscherichiacoli(DH5 ),asdescribedin previousstudies[13,19].100whitecolonieswerepicked andgrewforminipreprationsofplasmidDNAwith QIAprepSpinMiniprepKit(Hilden,Germany).The DNAsequencesweredeterminedbytheDNAsequencingfacilityattheInterdisciplinaryCenterforBiotechnologyResearch,UniversityofFlorida.DNAsequences thatwerepresentinmorethantwocloneswereconsideredasaptamercandidates.Flowcytometricanalysisofaptamerbindingtotarget cellsBiotin-labelled,selectedsinglestrandedDNApoolsor individualaptamersofinterestwereincubatedwith5 105cellsin200 Lofbindingbufferwith0.1%NaN3on icefor30min.Cellswerewashedtwicewith4mlof PBSbufferandincubatedwith5 LPE-streptavidin (0.5mg/ml,BectonDickinson,NJ)for30min.Biotinlabelledunselectedlibrarywasusedasanegativecontrol. Thecellswerewashedonceandcell-boundfluorescence wasdeterminedwithaFACScanorFACSCaliburflow cytometer(BectonDickinson,NJ)bycounting20,00050,000events.TheFITC,PEandPERCPwereactivated bybluelaser(488nm)andAPCbyredlaser(635nm). Fluorescence-labelledmonoclonalantibodieswereused withaptamerstodefinelineagesofbonemarrowleukocytesandleukemiccellsinclinicalspecimens.Todeterminethebindingaffinityofselectedaptamers,all experimentsfortheaptamerbindingassaywererepeated 2 … 4times.TheGraphPadSoftware(SanDiego,CA,USA) wasusedtoanalyzethedataforobtainingtheequilibrium dissociationconstants(Kd)ofthefluorescentaptamers andthe95%confidenceinterval.ClinicalsamplepreparationandtestingAllclinicalsamplesweresubmittedforpathological evaluationtotheShandsHospitalHematopathology Laboratory,UniversityofFlorida.ThestudieswereapprovedbytheUniversityofFloridaInstitutionalReview Board.Thepresenteddataincludethirty-sixcasesof AML.Tencasesofnon-malignanthumanbonemarrow werealsousedforthestudies. Erythrocytesinallbonemarrowsamplesspecimens werelysedasdescribedbefore[20].Humanbonemarroworleukemiccellswereimmunophenotypedwith thirtyfluorochrome-conjugatedmonoclonalantibodies inourclinicalflowcytometrypanels[20],forimmunophenotypingmatureorimmaturegranulocytes,monocytes,blastsandlymphocytessothatwecandetermine howtoselectivelygatethecellpopulationofinterest. ThedataanalysiswasperformedusingFCSExpresssoftware(DeNovoSoftware,LosAngeles,CA,USA,http:// www.denovosoftware.com).Initialcellsubpopulationswere establishedusingthelevelsofCD45expressionandsidescatter(SSC)properties[21,22].Afterdefiningimmunophenotypesofleukemiccells,antibodiesforCD45, CD34,CD117,CD33,HLA-DR,CD64orCD14(Becton Dickinson,NJ)wereusedtoselectcellsofinterestto determinefluorescencelevelsofboundaptamersfor individuallygatedsubpopulations.StatisticalanalysesGraphPadSoftwarewasusedforstatisticalanalyses.The One-wayAnalysisofVariance(ANOVA)or T testwas usedtocomparefluorescencelevelsofaptamersbound onthedifferentcellpopulations.Unlessstatedotherwise,resultsweregivenasmeanstandarddeviation (SD)andthePvalueswerealsogivenforcomparisonas necessary.ProteasetreatmentforcellsNB4cells(5106)werewashedwithPBSandthen incubatedwith1mlof0.25%trypsin/0.1%EDTAin Hank ’ sbufferedsaltsolution(HBSS)(ThermoScientific HyClone,Pittsburgh,PA)at37Cfor10min.FBSwasYang etal.JournalofHematology&Oncology 2014, 7 :5Page3of14 http://www.jhoonline.org/content/7/1/5

PAGE 4

thenaddedtoquenchtheprotease.Afterwashingwith PBS,thetreatedcellswereusedforaptamer-binding assaysasdescribedearlier.Enrichmentandidentificationoftheaptamer-bound targetproteinAtotalof,8108NB4cellsintheactivegrowingphase wereharvested,andusedastargetcellsforaptamerK19 bindingfollowedbyenrichmentoftheaptamer-bound targetprotein.TheNB4cellswerepre-incubatedwith 8mlofRPMImediacontaining1mgofheat-denatured HerringSpermDNA(Promega)at4Cfor15minto blockpotentialnonspecificbindingoftheaptamerto thecells.Thecellswerethenincubatedinthebinding bufferwithorwithoutbiotin-labelledaptamerK19(at thefinalconcentrationof300nM)andthebindingwas performedwithoutanyaptamerswasusedasanegative control.Todeterminethespecificityofaptamerbinding, anadditionalnegativecontrolwasmadebypre-incubating thecellswith300nMoftheunlabeledK19aptamerfor 1hrpriortothebindingofthebiotin-labelledaptamer. Afterbinding,thecellswerewashedthreetimeswith PBStoremovetheunboundaptamer.Asmallaliquotof eachcellsample(5105cells)wastaken,andanalysed byflowcytometrywithPE-streptavidintomonitorthe aptamerbinding. Theaptamer-boundorcontrolcellswerethenlysedin 10mloflysisbuffercontaining10mMHEPESpH7.4, 150mMNaCl,1%TritonX-100and1mMEDTAplus HaltTMproteaseinhibitorcocktail(ThermoScientific Pierce,Pittsburgh,PA)onicefor15min.Aftercentrifugationat14000gfor15min,thesupernatantwasincubatedwith1mg(100 l)ofmagneticstreptavidinbeads at4Cfor30mintocapturetheprotein-aptamercomplexes.Thebeadswithboundaptamer-proteincomplexeswerethencollectedonanEasySepmagnetstand (StemcellTechnologies,Vancouver,BC,Canada)and washedfivetimeswith15mlofthelysisbuffer.The enrichedproteinswereheatedforelutionandseparated by11%SDS-polyacrylamidegelelectrophoresis(SDSPAGE).Thegelswerethensilver-stainedwiththePierce SilverStainKit(ThermoScientificPierce,Rockford,IL). Theaptamer-specificproteinbandswereexcisedand trypsin-digestedinsitu[23]andanalysedbyQSTAR LC-MS/MSandaMASCOTdatabasesearchatthe InterdisciplinaryCenterforBiotechnologyResearch MassSpectrometryCoreFacility,UniversityofFlorida.Studiesofaptamer-antibodycompetitionFluorescein-conjugatedmousemonoclonalanti-human Siglec-5(Clone194128,R&DSystems,Minneapolis, MN,USA)andbiotin-labelledorunlabeledK19aptamerswereusedinthecompetitionstudies.Competition experimentswerecarriedoutintwoways:1)NB4cells (2105)wereincubatedwith300nMoftheunlabeled K19aptameroracontrolaptamerin100 Lofbinding bufferat4Cfor45min.AfterwashingwithPBStoremovetheunboundaptamers,cellswereincubatedwith 5 g/mlfluorescein-conjugatedanti-Siglec-5antibodyor controlIgG1antibodyin50 LofPBSwith0.5%BSAat 4Cfor45min.Afterwashingoffoftheunboundantibodies,thecellswereanalysedbyflowcytometry.2)The NB4cellswereincubatedwiththeanti-Siglec-5orthe controlantibodyandthenwiththebiotin-labelled aptamerK19orcontrolaptamers.AfterPBSwashing, PE-streptavidinwasaddedfollowedbyflowcytometric analysisasdescribedearlier.Non-RadioactiveCellProliferationAssayCellTiter96Non-RadioactiveCellProliferationAssay Kit(Promega,Wisconsin)wasusedtodetermineviable cellnumbersafterNB4cellswereincubatedwithvarious amountsofaptamer-streptavidin-saporincomplexesor mixturesofaptamerandunlabeledsaporin.Afterincubationfor72hours,theassayisperformedbyaddinga premixed,optimizedDyeSolutiontoculturewellsofa 96-wellplate.In4-hourincubation,livingcellsconvert thetetrazoliumcomponentoftheDyeSolutionintoa formazanproduct.TheSolubilization/StopSolutionthen wasaddedtotheculturewellstosolubilisetheformazan product,andtheabsorbanceat570nmisrecorded usinga96-wellplatereader.ResultsUsingCell-SELEXforselectionofaptamersboundtoNB4 cellsCulturedAMLNB4andHL60celllineshavebeenused foraptamerselection,andaptamersselectedagainst HL60cellscanrecognizemonocyticcells[19].Because ofpreviousunsuccessfulattemptstoselectaptamers againstNB4cells,wefocusedontheviabilityoftheculturedcellsusedforaptamerselection.Throughcareful optimization,wecriticallyimprovedthecellculture conditionsnecessarytomaintainNB4cellsintheactive proliferationphase.Theuseofcellsintheactiveproliferationphaseimprovedthecellsviabilityduringthe aptamerselectionprocedures,whichinturnreduced nonspecificaptamerbindingcausedbydeadcellfragmentsordebris.Asaresult,wewereabletoselecta panelofaptamersforNB4cells.Inaddition,wesignificantlyreducedthetimeperiodneededforCell-SELEX, andwereabletoobtaintheaptamerswithapproximatelyeightroundsofselection.10aptamercandidates wereobtainedthroughsequencing100individualclones andwechosethreerepresentativeaptamers(JH6,JH19 andK19)(Table1)forfurtherstudiesbecausethethree newaptamersshowedmuchbetterrecognitiontoNB4 cellsthantoHL60cells,andtheboundaptamersYang etal.JournalofHematology&Oncology 2014, 7 :5Page4of14 http://www.jhoonline.org/content/7/1/5

PAGE 5

exhibitedupto8to22foldincreasesinfluorescenceintensitycomparedtotheDNAlibrarycontrol (Figure1a).Wethendeter minedtheaffinityofthe threeaptamerstoNB4cells.Allthethreeaptamers havehighaffinityforNB4cellswithcalculatedKdof 2.77nMforJH6,7.57nMforJH19and12.37nMfor K19(Figure1bandTable1). Theselectedaptamerscandifferentiallyrecognize myeloidcellsinnormalhumanbonemarrowspecimens Becauseallthreeaptamerswereselectedagainstthe AMLNB4cellline,wetestedwhethertheselectedaptamers(JH6,JH19andK19)haveanabilitytorecognize differenttypesofleukocytesinhumanbonemarrow specimens.Whilenobindingonlymphocyteswasseen, Table1Sequencesandbindingaffinityofselectedaptamers AptamersKd(nM)Confidenceinterval(nM)(95%)Aptamersequences* JH6 2.770.69-4.855 -GTACGCCGCAAGACGAGTTGTGTATAAGCCGGC-3 JH19 7.574.13-11.025 -AGGTGTGACTCGATCTGTGGGGGTTGGGGGGTGGTTTTTCGGAA-3 K19 12.377.79-16.945 -AAGGGGTTGGGTGGGTTTATACAAATTAATTAATATTGTATGGTATATTT-3 *Allaptamershavetwoshortsequences(5 GACGCTTACTCAGGTGTGACTCG3 and5 CGAAGGACGCAGATGAAGTCTC3 )attachedtothe5 and3 ends,respectively, oftheaptamersequencesshownabove.Theshortsequencesat5 and3 endswereusedforPCRamplification. Figure1 AptamerrecognitionofculturedNB4andHL60leukemiccells.(a) .ComparisonofaptamerrecognitionofculturedNB4andHL60 leukemiccells.Individuallysynthesizedbiotin-labelledaptamersandPE-streptavidinwereanalyzedwithflowcytometryinordertocomparethei r abilitytorecognizeNB4andHL60cells.Single-strandedlibraryDNAwasusedasanegativecontrol.Thebindingofselectedaptamerswithcells isillustratedasthefollowing:negativecontrol(black);JH6(green);JH19(blue);K19(red).Thefinalconcentrationoftheseaptamersinbinding bufferwas150nM. (b) .DeterminationoftheaptamerbindingaffinitiestoNB4cells.Thebiotin-labeledaptamersandPE-labeledstreptavidinwere usedforthebindingassays.Thebackgroundbindingwasmeasuredbyusingunselectedsingle-strandedlibraryDNA.Thefluorescenceintensity geometricmeansofboundaptamerswasdeterminedbyflowcytometry.Theequilibriumdissociationconstants(Kd)ofthefluorescentligands wereobtainedbyfittingthedependenceofspecificbindingfluorescenceintensityontheconcentrationoftheligandstotheEquation Y=Bmax*X/(Kd+X)usingtheGraphPadSoftware(SanDiego,CA,USA)asdescribedinpreviousstudies[13]. Yang etal.JournalofHematology&Oncology 2014, 7 :5Page5of14 http://www.jhoonline.org/content/7/1/5

PAGE 6

allthethreeaptamersshowedhighlevelsofbinding (6to17foldsoffluorescenceintensityoverbackground) onmatureandimmaturegranulocytesandmonocytes (Figure2).Theresultssuggestthatthethreeaptamers mayrecognizemyeloid-specificsurfacemarkers.The boundaptamerK19hadhigherfluorescenceintensityon granulocytes,monocytes,andNB4cellsthanbound aptamersJH6andJH19(Figures1and2).Inaddition,all threeaptamershadlow,butstatisticallysignificant,levels ofbindingonCD34(+)earlyhematopoieticprecursors (Figure2). Theselectedaptamerscandifferentiallyrecognize leukemiccellsfromAMLnon-M3andAMLM3cases BecausethethreeaptamersrecognizedmaturinggranulocytesandmonocytesbetterthanCD34(+)earlyprogenitors,weseparatedAMLclinicalspecimensintothree groups:1)AMLnon-M3CD34(+);2)AMLnon-M3 CD34( );and3)AMLM3.WethendeterminedifaptamersJH6,JH19,andK19coulddifferentiallyrecognize anygroupsofAMLcases.Whiletheseaptamersshowed lowlevelsofreactivityonnormalCD34(+)progenitors,all threeaptamerscanrecognizebothCD34(+)andCD34( ) cellsofAMLnon-M3caseswiththemedianvaluesof fluorescenceintensitybeing~8to30foldhigherthan thoseofbackgroundbinding(Figure3).However,the levelsofthethreeaptamersboundonAMLnon-M3cases variedsignificantly,andtherewasnostatisticalsignificanceinaptamerbindinglevelsbetweenthenormal CD34(+)cellsandleukemiccellsfromAMLnon-M3 cases.Critically,allthreeaptamershadmuchlowerlevels ofbindingonleukemiccellsofAMLM3casesthannormalCD34(+)earlyprogenitorsorleukemiccellsofAML non-M3cases.Thesedifferenceswerestatisticallysignificant(Figure3). Usingbiotin-labelledK19aptamerstoenrichandidentify itstargetprotein Inordertodetermineifthetargetsoftheaptamersmay representsurfaceproteinsormoietiesassociatedwith surfacemembraneproteins,wetreatedNB4cellswith trypsinbeforebindingtheaptamersoncells.Asshown inFigure4a,thebindingsitesofaptamersJH6,JH19 andK19,asindicatedbythefluorescenceintensityof boundaptamers,werepartiallyoralmostcompletely abolishedby10minoftrypsindigestion.Theseresults suggestthatthetargetmoleculesrecognizedbythese aptamersmaybedirectlyorindirectlyrelatedtosurface proteinsanchoredonthecellmembrane. SinceaptamerK19boundNB4cellsdemonstrate relativelyhigherfluorescentintensity,suggestingmore abundantaptamerK19bindingsitesascomparedtothe cellsboundwithaptamersJH6andJH19,andthree aptamersshowedsimilarbindingpatternswhenapplied tobonemarrowCD34(+)cells,granulocytesandmonocytes,wefocusedonidentificationoftheproteintarget associatedwiththebindingsiteofaptamerK19.Flow cytometricanalysisisaverysensitivetechnology,andwe estimatedthattherewereonlyafewhundredaptamer K19bindingsitesonindividualNB4cellswhenwe comparedthefluorescenceintensityofK19tothoseof PE-beads(QuantiBRITEPE,BectonDickinson),which aredesignedtoestimatethenumberofboundantibody moleculespercell. ToverifythespecificbindingofaptamerK19during targetproteinenrichment,weusedanegativecontrol,in whichunlabeledaptamerK19wasusedtoblockthe bindingofbiotin-labelledaptamerK19toNB4cells. Flowcytometricanalysisofsmallaliquotsoftheaptamerboundcellsamples,whichweremadetoenrichtarget proteins,demonstratedthattheunlabeledaptamercan Figure2 RecognitionofnormalbonemarrowleukocytesbyaptamersJH6,JH19,andK19. Thefluorescenceintensityofboundaptamers orsingle-strandednegativecontrolDNAonnormalhumanbonemarrowcells,includinglymphocytes,granulocytes,monocytes,andCD34+cells wasdeterminedbyflowcytometry.Fluorescenceintensityisshownasmeanstandarddeviation,and “ ” “ ** ” ,and “ *** ” representthe P values of<0.05,<0.01,and<0.001,respectively. Yang etal.JournalofHematology&Oncology 2014, 7 :5Page6of14 http://www.jhoonline.org/content/7/1/5

PAGE 7

completelyabolishthebindingofbiotinylatedones,indicativeofthebindingspecificityofaptamerK19(Figure4b). Theprotein-aptamercomplexeswereextractedwith thebuffercontaining1%TritonX-100,capturedusing streptavidin-coatedmagneticbeads,andseparatedby SDS-PAGE.Wethenappliedsilver-stainforproteindetection(Figure4c).Comparedwiththenegativecontrols (lane1,withoutaptameraddedandlane2,blockedby unlabeledaptamerK19),severalapparentK19-specific proteinbandswereshowninlanes3and4.Thesebands wereexcisedforfurthertrypsintreatment,andanalysed bymassspectrometry(MS).ItisnoteworthythatSDSPAGEanalyseswererununderbothreducingandnonreducingconditions,andthesmearbandat130 … 140kDa obtainedunderthenon-reducingconditionwasreproduciblydetected(Figure4c,arrow). TheMSdataofpeptideswereusedtosearchthe MASCOTdatabaseinordertoidentifypossibleprotein candidates.Thecandidateproteinhits(~20)include manyRNAorDNAbindingproteins,intracellularly localizedsolubleproteinssuchaslysozymeC,andcontaminatedkeratins.TheonlycellsurfaceproteinidentifiedonthelistwasSialicacid-bindingIg-likelectin5 (Siglec-5).TheMSdatashowed5uniquepeptidesidentifiedasfragmentsofSiglec-5.ThesequencesofidentifiedpeptidesaremarkedontheSiglec-5sequenceas showninFigure5.Siglec-5existsasadisulfide-linked dimerof~140kDa[24,25],whichisinagreementwith thesizeoftheK19-bound130 … 140kDaproteinband identifiedonSDS-PAGEunderthenon-reducingcondition(Figure4c,arrow). AptamerK19andanti-Siglec-5antibodycancompete againsteachotherforthebindingsitesontheNB4cells ToconfirmthatSiglec-5istheproteintargetofthe aptamerK19,wecarriedoutthecompetitionexperiment usingafluorescein-conjugatedanti-humanSiglec-5antibody.AsshowninFigure6aandc,theaptamerK19 andtheSiglec-5antibodycancompeteagainsteach Figure3 ComparisonofaptamerrecognitionofAMLleukemic cellsandnon-malignantCD34(+)cells. TheAMLcaseswere separatedintothreegroups:1)CD34(+)AMLnon-M3;2)CD34( ) AMLnon-M3;and3)AMLM3.Thefluorescencelevelsofbound aptamersorsingle-strandednegativecontrolDNAweredetermined byflowcytometry.Thefluorescenceintensitylevelsofbound aptamers(foldsoverbackground)werecalculated( a ,JH6; b ,JH19 and c ,K19).Individualvaluesforeachaptamerboundoneachcase areshownasindividualsymbols,andmeanstandarddeviationof individualgroupsarealsoshown.The P valuesaregivenas “ ” “ ** ” ,and “ *** ” representingthe P valuesof<0.05,<0.01,and <0.001,respectively. Yang etal.JournalofHematology&Oncology 2014, 7 :5Page7of14 http://www.jhoonline.org/content/7/1/5

PAGE 8

otherforthebindingsitesontheNB4cells.Incontrast, thecontrolaptamerE10,whichcanalsobindtoNB4 cells(unpublishedresults),doesnotdisplayanycompetitionwiththeSiglec-5antibody(Figure6bandd),and thereactivityofaptamerK19towardNB4cellswasnot affectedbyisotypecontrolantibodies(Figure6a).Thus, weconfirmedthatSiglec-5isthetargetedproteinrecognizedbyaptamerK19,andthatthebindingsiteofaptamerK19ontheSiglec-5proteinmaybestericallyclose totheepitopeboundbytheSiglec-5antibody. Figure4 Usingbiotin-labeledK19aptamertoenrichitstargetprotein.(a) .Effectoftrypsinpre-treatmentonbindingofaptamerstoNB4 cells.NB4cellsweretreatedwithtrypsinat37Cfor10min.Thebindingofselectedaptamerswithtrypsin-treatedNB4cellsisillustrated:JH6(a1) ; JH19(a2);andK19(a3).Thefinalconcentrationoftheseaptamersinbindingbufferwas300nM.Thebackgroundbindingwasmeasuredbyusing biotinylatedsingle-strandednegativecontrolDNA. (b) .Bindingcompetitionofbiotinylated-K19toNB4cellsbytheunlabeledK19aptamers.NB4 cellswerepre-incubatedwiththeunlabeledK19aptamers(300nM)for1hrpriortobindingofthebiotin-labeledaptamers.Thefluorescence intensityofboundaptamersisshownbythehistograms(blue,bindingofbiotinylatedK19;red,bindingofbiotinylatedK19afterblockingby non-labeledK19).Thebackgroundbindingwasmeasuredbyusingbiotinylatedsingle-strandednegativecontrolDNA. (c) .Silver-stained polyacrylamidegelelectrophoresis(SDS-PAGE)separatingtheproteinscapturedbyaptamerK19.NB4cellswerepre-incubatedwithorwithout theunlabeledaptamerK19for1hrpriortobindingofthebiotin-labeledaptamers.TheNB4cellswerethenlysed.Theprotein-aptamerDNA complex(s)werecapturedbythemagneticstreptavidinbeads,andwerethenseparatedbySDS-PAGEfollowedbysilverstainingfordetectionof characteristicproteinbands.LaneM,molecularmarkers;Lane1,proteinscapturedwithstreptavidinbeads(noaptamer);Lane2,proteinscaptured bybiotin-aptamerK19afterblockingbyunlabeledaptamerK19;Lanes3and4,proteincapturedwithbiotinylatedaptamerK19. Yang etal.JournalofHematology&Oncology 2014, 7 :5Page8of14 http://www.jhoonline.org/content/7/1/5

PAGE 9

Siglec-5canbeusedasabiomarkerforgranulocytic maturationandAMLcelldetectionaswellasbeusedas apotentialtargetforleukemiccellgrowthinhibition Siglec-5wasreportedtobeexpressedongranulocytes [26],butitsexpressionduringgranulocyticormonocytic maturationhasnotbeenwellcharacterized.SinceaptamerK19recognizedmaturinggranulocytesmuchbetter thanCD34(+)earlyprogenitorsinnormalhumanbone marrow(Figure2),wefurtherdeterminedwhetherits bindingsites(i.e.Siglec-5proteinlevels)ongranulocytes varyduringgranulocyticmaturation.Byflowcytometric analysis,weseparatedmaturinggranulocytesormonocytesintothreesubsets:earlystage,immediatestage, andmaturedstage,accordingtotheexpressionlevelsof CD13andCD11bforgranulocytesandCD64andCD14 formonocytes(Figure7,leftpanel)[27-29].Wethen determinedthefluorescencelevelsofaptamerK19 boundoneachsubset.Comparedwiththenegativecontrol,thefluorescenceintensityofboundaptamerK19 ongranulocytesgraduallyincreasedduringgranulocytic maturation(Figure7,rightpanel),indicatingprogressiveup-regulationofSiglec-5levelsduringgranulocytic maturation.However,persistentlyhighlevelsofSiglec-5 expressionwereobservedonbothCD64(+)/CD14( ) immatureandCD64(+)/CD14(+)maturemonocytes. BecauseSiglec-5isoverexpressedinasubsetofAML cells,weselectedanAMLcasewithrelativelyhigh levelsofSiglec-5expression,andspikedsmallnumbers oftheAMLcellsintoanormalhumanbonemarrow specimen.Then,basedonthedifferentialexpressionlevels ofSiglec-5onnormalCD34(+)cellsandCD34(+) leukemiccells,weusedaptamerK19toaidinthedetectionofAMLcellsmixedintoanormalbonemarrowspecimen(Figure8).Additionally,todemonstrate Siglec-5canbeapotentialbiomarkerfortargeted therapy,wetestedbiotinylatedSiglec-5aptamerK19 andsaporin-cross-linkedtostreptavidin(SA-SAP)for inhibitingNB4cellproliferationinvitro.Compared withunlabeledsaporin(SAP)(Figure9a)orthebiotinylatedsinglestrandedDNAcontrol(Figure9b),the Siglec-5aptamerK19canenhancethetoxicityofSASAPtoNB4cellswithanestimatedIC50of25to50 nM.TheenhancedtoxiceffectofbiotinylatedK19aptamercanbeblockedbynon-labelledaptamerK19, indicatingthattheenhancedcelltoxicityismediated throughthespecificbindingtosurfaceSiglec-5proteins (Figure9c). Discussion Themolecularcharacteristicsofleukemiccells,especially attheproteomiclevel,arecriticalforunderstanding leukemiapathogenesisanddesigningtargetedtherapy.In thelastseveraldecades,proteomicanalysishasbeenperformedtoadvancethediscoveryofdiseasedcell-specific proteinbiomarkers,butsofaronlyafewAMLbiomarkers havebeenintroducedintoclinicalpracticeforAML detectionandtherapy.Currently,westilllackeffective biomarkersforAMLdiagnosisandtargetedtherapy. Thus,ourintentinthisstudyistodevelopnew Figure5 ThesequencesofSiglec-5proteinandpeptidesrecoveredbymassspectrometricanalysisofaptamerK19-enrichedproteins. TheIg-likeV-typedomainandtwoIg-likeC2-typedomainsofSiglec-5areunderlinedwhilethesignalpeptide,thetransmembraneregion,and theimmunoreceptortyrosine-basedinhibitory(ITIM)motifaremarkedwithdottedunderlines.Thepeptidesidentifiedbymassspectrometryare highlightedbybordering. Yang etal.JournalofHematology&Oncology 2014, 7 :5Page9of14 http://www.jhoonline.org/content/7/1/5

PAGE 10

molecularprobesthattargetsurfaceproteinbiomarkers onAMLcells. Membraneproteinsfunctionasadhesion-anchors, receptors,transportersandenzymesthatplayrolesin variousphysiologicalprocesses,andtheirdysregulation maycontributetothepathogenesisofmanydisease processes,includingAML.Itisestimatedthat20 … 35%of themammaliangenomeencodesmembraneproteins [30].However,membraneproteinsremainmostlyunderrepresentedintheproteomicanalysisduetotechnical challenges.Proteomicstudies,usingmassspectrometrybasedtechnology,aimatidentifyingindividualproteinsso thatanassay,mostoftenantibody-based,canbedeveloped foraspecificprotein.Bycontrast,theCell-SELEXapproachproducesagroupofcell-specificaptamersthatcan beverifiedinclinicalspecimenswithoutpurifiedproteinsorevenknowledgeoftheirproteintargets.The selectedaptamerscaneasilybelabelledforflowcytometryorimageanalysisofcellsinclinicalspecimens.If oneoftheaptamerprobesdetectsasurfacemarkerof interest,thespecificaptamerprobecanbeusedtoenrichandpurifythetargetprotein,suchasaptamerK19 anditstargetSiglec-5.Itshouldbenotedthatourstudiesdonotseektocomparethecapabilityofaptamers withantibodies.Numeroushighqualitymonoclonalantibodieshavebeenproducedforisolatedproteins.However, itisdifficulttomakeantibodieswhenwedonotknowthe targetproteinsontumorcells. WeusedHL-60andNB4humanleukemiccelllines forourexperiments,andthetwocelllinesareclosely related.Whilebothcandifferentiateunderchemical induction,onlytheNB4cellsoriginatedfromAMLM3, andcarrythet(15;17)chromosometranslocation[31-33]. Byusingthesetwoleukemiccelllines,wecanaddress Figure6 CompetitionbindingassaysofaptamerK19andanti-Siglec-5antibody. Forblockingaptamerbindingwithantibodies,NB4cells werepre-incubatedwithanti-Siglec-5orisotypecontrolantibodiesbeforebindingtobiotin-labelledaptamersK19 (a) orE10 (b) .Thefluorescence intensitiesofboundaptamersweredeterminedbyflowcytometry.SimilarapproacheswereemployedwithaptamersK19 (c) orE10 (d) usedtoblock bindingoftheFITC-anti-Siglec-5antibodyonNB4cells.TheFITC-isotypecontrol(IgG1)wasusedasanegativecontrolforspecificantibodybindin g. Yang etal.JournalofHematology&Oncology 2014, 7 :5Page10of14 http://www.jhoonline.org/content/7/1/5

PAGE 11

threequestions:A).Isitpossibletoselectsinglestranded DNAaptamersthatarecapableofdetectingdifferences insurfaceproteinexpressionbetweentwocloselyrelatedleukemiccelllines(HL60andNB4)?B).Canthese selectedaptamersbefurtherusedonclinicalspecimens forphenotypingAMLandidentifyingnewbiomarkers? C).Canthenewlyidentifiedbiomarkerbeusedtoaid thedetectionofAMLcellsinhumanbonemarrow specimens? Asaresult,weusedNB4leukemiccellstoselectand characterizethreenewDNAaptamers(JH6,JH19,and K19),whichhavemorebindingsitesonNB4cellsthan onHL60cells.ThisisincontrasttotheaptamerKH1C12 previouslyselectedfromHL60cells,whichselectively recognizedHL60cells[19].Geneexpressionprofiling studiesshowedthatNB4andHL60celllineshadthemost closelyrelatedprofilesofmRNAexpression[33].Thus, ourresultswithaptamersselectedagainstNB4andthose previouslyselectedagainstHL60cellsindicatethatitis possibletoselectaptamerscapableofdetectingdifferences insurfaceproteinexpressionbetweentwocloselyrelated leukemiccelllines. Figure7 UsingSiglec-5asabiomarkerforgranulocyticdifferentmaturationinhumanbonemarrow. Leukocytesfromnormalbone marrowaspirateswereincubatedwithaptamerK19orthesingle-strandednegativecontrolDNA. (a) .Siglec-5expressionofmaturinggranulocytes. Thegranulocytepopulationwasfirstidentifiedusinglevelsofside-scatteredlight(SSC)incombinationwiththefluorescenceintensityofCD45. Then, maturinggranulocyteswereseparatedintothreesubsetsaccordingtotheexpressionlevelsofCD13andCD11b(leftpanel):earlystage(yellow), immediatestage(purple),andmaturedstage(green).Fluorescenceintensity(PE)ofthegranulocytesubsetsboundwithsinglestrandedDNAcontrol (middlepanel)oraptamerK19/Siglec-5(rightpanel)isshowninrelationtofluorescencelevelsofCD11b(APC). (b) .Siglec-5expressionofmaturing monocytes.Thematureandimmaturemonocyteswerefirstidentifiedusinglevelsofside-scatteredlight(SSC)incombinationwiththefluorescence intensityofCD64.Then,maturingmonocyteswereseparatedintothreesubsetsaccordingtotheexpressionlevelsofCD64andCD14(leftpanel):early stage(yellow),immediatestage(teal)andmaturedstage(fuchsia).Fluorescenceintensity(PE)ofthemonocytesubsetsboundwithsingle-strande d negativecontrolDNA(middlepanel)oraptamerK19/Siglec-5(rightpanel)isshowninrelationtofluorescencelevelsofCD14(APC). Yang etal.JournalofHematology&Oncology 2014, 7 :5Page11of14 http://www.jhoonline.org/content/7/1/5

PAGE 12

Thefluorescenceintensitylevelsofboundaptamerson leukemiccells,comparedwiththoseonCD34(+)progenitorsinnormalbonemarrowspecimens,varysignificantly amongdifferentAMLcases(Figure3).Thefindingsare likelyreflectiveoftheheterogeneityofdiseaseintheAML group.Whiletheoverallfluorescencelevelsofbound aptamersbetweentheCD34(+)normalprogenitorsand AMLnon-M3groupsarenotstatisticallysignificant, subsetsofAMLnon-M3casesmayoverexpressoneor moresurfacebiomarkersthatcanberecognizedby aptamers.TheheterogeneityofAMLrequiresgreateffortandresourcesinordertodevelopusefulbiomarkersforitsdetectionandtreatment,butwecan alsousetheheterogeneityofbiomarkerexpressionfor diagnosisortargetedtherapy.IfanAMLcaseoverexpressesoneormoresurfacebiomarkersthatcan berecognizedbyaptamers(e.g.Siglec-5inFigure8), theaptamerprobesmaybecomeusefultoolsfordetectingtheminimalresidualdiseaseofAMLafter chemotherapy. DespitetheirselectionfromNB4cellsderivedfroma caseofAMLM3andtheirabilitytorecognizematuring granulocytesandmonocyteswell(Figure2),allthree aptamersshowrelativelylowerlevelsofbindingtoAML M3(Figure3).Significantdown-regulationofnormal myeloidmarkersmayoccuronAMLM3cellsinclinical specimens,andonewell-knownexampleisCD15.Like CD15,theaberrantlydown-regulatedexpressionofaptamertargetproteins,includingSiglec-5inAMLM3, makesitpossibleforustousethemasbiomarkersto differentiatebetweenAMLnon-M3andAMLM3cases inclinicalpractice.Conversely,ifAMLcellsinaclinical specimenshowhighlevelsofreactivitytothethreeaptamers,thecaseisunlikelytorepresentAMLM3. Thehuman,CD33-related,sialicacidbinding, immunoglobulin-likelectins(CD33rSiglecs)comprisea Figure8 UsingSiglec-5asabiomarkertodetectlowlevelsofAMLcellsinhumanbonemarrowspecimen. ArepresentativeAMLbone marrow(leftpanelsin a .and b .)andnormalbonemarrow(NORBM)(rightpanelsin a .and b .)caseareshown.Inaddition,theAMLspecimenis dilutedwithnormalbonemarrowcells(middlepanelsin a .and b .),andthefinalAMLcellfractionwasapproximately1.6%.CD34(+)cellsinthree samplesaregatedselectivelyaccordingtothelevelsofCD34andSSC (a.) ,andtheexpressionofSiglec-5onCD34(+)cellsisdisplayedin (b.) TheCD34(+)leukemiccells(reddotsin b .)havehigherlevelsofSiglec-5thannormalCD34(+)cellsinnormalbonemarrow(bluedotsin b ). Yang etal.JournalofHematology&Oncology 2014, 7 :5Page12of14 http://www.jhoonline.org/content/7/1/5

PAGE 13

familyofreceptors(includingSiglec-5)thataredifferentiallyexpressedonleukocytes.TheaptamerK19, whichrecognizesSiglec-5,canrecognizegranulocytes andmonocytes,withnosignificantbindingtobone marrowlymphocytes(Figure2).Inaddition,with aptamerK19wedemonstratedrelativelylowlevelsof Siglec-5onCD34(+)progenitorcellsinnormalbone marrow(Figure2),andup-regulatedSiglec-5duringthe granulocyticmaturation(Figure7).Theseresultsare consistentwithpreviouslyreportedresultsthatSiglec-5 expressionwasup-regulatedlaterthanCD33during invitro myeloiddifferentiationofCD34(+)cellspurified fromcordblood[26]. CD33hasbeenusedasoneofthecommonmarkers ofAML.TheantibodyagainstCD33,gemtuzumab (Mylotarg ™ ),hasbeentestedforthetreatmentofAML, andwasreportedtobeeffectiveatinducingremissions inabout25-30%ofrelapsedAMLpatientsdespiteits reversibletoxicityonnormalmyeloidcells[34].Dueto therestrictedexpressionofCD33ondifferentleukocyte celltypes,othermembersoftheCD33rSiglecsfamily, includingSiglec-5,havealsobeenexploredastargetsfor cell-directedtherapiesofAML[35].Itwasshownthat anti-Siglec-5antibodylinkedwithsaporintoxininduced cellkillinginU937humanleukemiccells[36].OurstudiesshowthattheSiglec-5aptamerK19cancompete withanti-Siglec-5antibodyforbindingtoNB4cells (Figure6),andthroughSiglec-5proteinsitcanalsomediateuptakeofSaporintoinhibitNB4cellproliferation invitro.Thepublishedresultsofimmunotoxinstudies wereperformedwithantibodiesdirectlyconjugatedto toxin([37,38]),anditshouldbenotedthatSaporinis linkedtostreptavidininourstudiesratherthandirectly conjugatedtoaptamerK19.Therefore,theefficiencyof aptamer-mediatedup-takeofsaporinmightbelow.In thesestudies,ourintentistodemonstratethepotential ofSiglect-5anditsaptamers,anditisnecessarytoperformfurtheroptimizationofaptamersandaptamer-toxin conjugatesinordertodeterminewhetherSiglec-5andits aptamercantrulybeusedasabiomarkerfordetection andtargetedtherapyofAML. Insummary,inthisreportedstudy,wehavedemonstratedapipelineapproachforbiomarkerdiscovery.We firstemployedtheCell-SELEXtechniquetoselectDNA aptamersthatcanbeusedasmolecularprobestophenotypenormalhematopoieticcellsandAMLcells.Wethen usedoneoftheaptamerstoenrichandidentifyitstarget proteinonthesurfaceofleukemiccells.Finally,wedemonstratedthattheidentifiedbiomarker(Siglec-5)canaid inthedetectionofAMLcellsatlowconcentrations,and canpotentiallymediatetargetedtherapyofAMLcells. Thisstrategydevelopedwithleukemiccellsshouldbe applicabletoothertypesofcancertofacilitatebiomarker discoveryandtargetedcancertherapy. Additionalfile Additionalfile1:FigureS1. Flowcytometryassayformonitoring enrichmentofthespecificaptamerpoolagainstNB4leukemiacells.After 10roundsofselectionprocesses,thephycoerythrin(PE)labeledaptamer poolshowedsignificantincreasesinfluorescenceintensityontargetNB4 cells,butitproducedminimalchangeinfluorescenceintensityonHL60 cells.TheseresultsindicatethattheaptamersrecognizingtargetNB4 cellswereenrichedpreferentially. Competinginterests Theauthorsdeclarenocompetingfinancialinterests. Figure9 UsingSiglec-5aptamerK19toenhancethesaporintoxicityforinhibitingNB4cellproliferationinvitro. Variousconcentrations ofbiotinylated-aptamerK19(Biotin-K19)orbiotinylatedsingle-strandednegativecontrolDNA(Biotin-SSDNA)andstreptavidin-saporin(SA-SA P)orfree saporin(SAP)weremixed,andwerethenincubatedwithNB4cells (aandb) .Thenon-biotinylatedaptamerK19(K19)wasusedtoblockthebinding ofbiotin-K19toNB4cells (c) .Theviablecellswerethenmeasuredin72hours.Theexperimentswereperformedintriplicates,andthedataisshown asfractionsofthecontrolgroups.The P valuesaregivenas “ ” or “ ** ” representingthe P valuesof<0.05or<0.01,respectively. Yang etal.JournalofHematology&Oncology 2014, 7 :5Page13of14 http://www.jhoonline.org/content/7/1/5

PAGE 14

Authors ’ contributions MYandGJ,performedresearchandanalyzedthedataandwrotethepaper; WL,KQandMZ,performedresearchwork;SAandCC,performedleukemia dataanalysisandwrotethepaper;YL,designedresearch,analyzedthedata, andwrotethepaper.Allauthorsreadandapprovedthefinalmanuscript. Funding ThisworkwassupportedbytheNationalInstitutesofHealth[CA129311toY.L.]. Received:2November2013Accepted:24December2013 Published:9January2014 References1.AppelbaumFR,GundackerH,HeadDR,SlovakML,WillmanCL,GodwinJE, AndersonJE,PetersdorfSH: Ageandacutemyeloidleukemia. Blood 2006, 107: 3481 – 3485. 2.DohnerH,EsteyEH,AmadoriS,AppelbaumFR,BuchnerT,BurnettAK, DombretH,FenauxP, etal : Diagnosisandmanagementofacutemyeloid leukemiainadults:recommendationsfromaninternationalexpert panel,onbehalfoftheEuropeanLeukemiaNet. Blood 2010, 115: 453 – 474. 3.O ’ DonnellMR,AbboudCN,AltmanJ,AppelbaumFR,ArberDA,AttarE, BorateU,CoutreSE, etal : Acutemyeloidleukemia. JNatlComprCanc Netw 2012, 10: 984 – 1021. 4.SekeresMA: Treatmentofolderadultswithacutemyeloidleukemia:state oftheartandcurrentperspectives. Haematologica 2008, 93: 1769 – 1772. 5.ZaidiSZ,OwaidahT,AlSF,AhmedSY,ChaudhriN,AljurfM: Thechallenge ofriskstratificationinacutemyeloidleukemiawithnormalkaryotype. HematolOncolStemCellTher 2008, 1: 141 – 158. 6.GregoryTK,WaldD,ChenY,VermaatJM,XiongY,TseW: Molecular prognosticmarkersforadultacutemyeloidleukemiawithnormal cytogenetics. JHematolOncol 2009, 2: 23. 7.GoldL,JanjicN,JarvisT,SchneiderD,WalkerJJ,WilcoxSK,ZichiD: AptamersandtheRNAworld,pastandpresent. ColdSpringHarbPerspect Biol 2012, 4: 1 – 9. 8.BarbasAS,MiJ,ClaryBM,WhiteRR: Aptamerapplicationsfortargeted cancertherapy. FutureOncol 2010, 6: 1117 – 1126. 9.CerchiaL,GiangrandePH,McNamaraJO,deF,V: Cell-specificaptamersfor targetedtherapies. MethodsMolBiol 2009, 535: 59 – 78. 10.EllingtonAD,ConradR: Aptamersaspotentialnucleicacid pharmaceuticals. BiotechnolAnnuRev 1995, 1: 185 – 214. 11.BrodyEN,WillisMC,SmithJD,JayasenaS,ZichiD,GoldL: Theuseof aptamersinlargearraysformoleculardiagnostics. MolDiagn1999, 4: 381 – 388. 12.BunkaDH,PlatonovaO,StockleyPG: Developmentofaptamer therapeutics. CurrOpinPharmacol 2010, 10: 557 – 562. 13.ShangguanD,LiY,TangZ,CaoZC,ChenHW,MallikaratchyP,SefahK, YangCJ, etal : Aptamersevolvedfromlivecellsaseffectivemolecular probesforcancerstudy. ProcNatlAcadSciUSA 2006, 103: 11838 – 11843. 14.DuaP,KimS,LeeDK: Nucleicacidaptamerstargetingcell-surface proteins. Methods 2011, 54: 215 – 225. 15.BlankM,WeinschenkT,PriemerM,SchluesenerH: Systematicevolutionof aDNAaptamerbindingtoratbraintumormicrovessels.selective targetingofendothelialregulatoryproteinpigpen. JBiolChem 2001, 276: 16464 – 16468. 16.DanielsDA,ChenH,HickeBJ,SwiderekKM,GoldL: Atenascin-Captamer identifiedbytumorcellSELEX:systematicevolutionofligandsby exponentialenrichment. ProcNatlAcadSciUSA 2003, 100: 15416 – 15421. 17.MallikaratchyP,TangZ,KwameS,MengL,ShangguanD,TanW: Aptamer directlyevolvedfromlivecellsrecognizesmembranebound immunoglobinheavymuchaininBurkitt  slymphomacells. MolCell Proteomics 2007, 6: 2230 – 2238. 18.ShangguanD,CaoZ,MengL,MallikaratchyP,SefahK,WangH,LiY,TanW: Cell-specificaptamerprobesformembraneproteinelucidationincancer cells. JProteomeRes 2008, 7: 2133 – 2139. 19.SefahK,TangZW,ShangguanDH,ChenH,Lopez-ColonD,LiY,ParekhP, MartinJ, etal : Molecularrecognitionofacutemyeloidleukemiausing aptamers. Leukemia 2009, 23: 235 – 244. 20.JiangG,ZhangM,YueB,YangM,CarterC,Al-QuranSZ,LiB,LiY: PTK7:Anew biomarkerforimmunophenotypiccharacterizationofmaturingTcellsand Tcellacutelymphoblasticleukemia. LeukRes 2012, 36: 1347 – 1353. 21.BorowitzMJ,GuentherKL,ShultsKE,StelzerGT: Immunophenotypingof acuteleukemiabyflowcytometricanalysis.UseofCD45andright-anglelightscattertogateonleukemicblastsinthree-coloranalysis. AmJClin Pathol 1993, 100: 534 – 540. 22.WoodBL: Ten-colorimmunophenotypingofhematopoieticcells. CurrProtocCytom 2005, Chapter6: 6.21.1 – 6.21.11. 23.GranvoglB,PloscherM,EichackerLA: Samplepreparationbyin-gel digestionformassspectrometry-basedproteomics. AnalBioanalChem 2007, 389: 991 – 1002. 24.CornishAL,FreemanS,ForbesG,NiJ,ZhangM,CepedaM,GentzR, AugustusM, etal : Characterizationofsiglec-5,anovelglycoprotein expressedonmyeloidcellsrelatedtoCD33. Blood 1998, 92: 2123 – 2132. 25.CrockerPR,McMillanSJ,RichardsHE: CD33-relatedsiglecsaspotential modulatorsofinflammatoryresponses. AnnNYAcadSci 2012, 1253: 102 – 111. 26.VirgoP,Denning-KendallPA,Erickson-MillerCL,SinghaS,EvelyR,HowsJM, FreemanSD: IdentificationoftheCD33-relatedSiglecreceptor,Siglec-5 (CD170),asausefulmarkerinbothnormalmyelopoiesisandacute myeloidleukaemias. BrJHaematol 2003, 123: 420 – 430. 27.LokenMR,WellsDA: Theroleofflowcytometryinmyelodysplastic syndromes. JNatlComprCancNetw 2008, 6: 935 – 941. 28.Stetler-StevensonM,ArthurDC,JabbourN,XieXY,MolldremJ,BarrettAJ, VenzonD,RickME: Diagnosticutilityofflowcytometric immunophenotypinginmyelodysplasticsyndrome. Blood 2001, 98: 979 – 987. 29.WoodBL: Flowcytometricdiagnosisofmyelodysplasiaand myeloproliferativedisorders. JBiolRegulHomeostAgents 2004, 18: 141 – 145. 30.SavasJN,SteinBD,WuCC,YatesJRIII: Massspectrometryaccelerates membraneproteinanalysis. TrendsBiochemSci 2011, 36: 388 – 396. 31.PassMB,BorregaardN,CowlandJB: Derangementoftranscriptionfactor profilesduringinvitrodifferentiationofHL60andNB4cells. LeukRes 2007, 31: 827– 837. 32.BalleriniP,BesanconF,CayreYE: [Effectoftranslocationt(15;17)onthe geneexpressionregulationofmyeloblastinduringalltransretinoicacid inducedmyeloiddifferentiationinhumanleukemiccells]. CRSeances SocBiolFil 1995, 189: 521 – 530. 33.LeupinN,KuhnA,HugliB,GrobTJ,JaggiR,ToblerA,DelorenziM,FeyMF: Geneexpressionprofilingrevealsconsistentdifferencesbetweenclinical samplesofhumanleukaemiasandtheirmodelcelllines. BrJHaematol 2006, 135: 520 – 523. 34.WalterRB,AppelbaumFR,EsteyEH,BernsteinID: Acutemyeloidleukemia stemcellsandCD33-targetedimmunotherapy. Blood 2012, 119: 6198 – 6208. 35.O ’ ReillyMK,PaulsonJC: Siglecsastargetsfortherapyinimmune-cellmediateddisease. TrendsPharmacolSci 2009, 30: 240 – 248. 36.NguyenDH,BallED,VarkiA: Myeloidprecursorsandacutemyeloid leukemiacellsexpressmultipleCD33-relatedSiglecs. ExpHematol 2006, 34: 728 – 735. 37.ChuTC,MarksJWIII,LaveryLA,FaulknerS,RosenblumMG,EllingtonAD, LevyM: Aptamer:toxinconjugatesthatspecificallytargetprostatetumor cells. CancerRes 2006, 66: 5989 – 5992. 38.ZhangY,HongH,CaiW: Tumor-targeteddrugdeliverywithaptamers. CurrMedChem 2011, 18: 4185 – 4194.doi:10.1186/1756-8722-7-5 Citethisarticleas: Yang etal. : Developingaptamerprobesforacute myelogenousleukemiadetectionandsurfaceproteinbiomarker discovery. JournalofHematology&Oncology 2014 7 :5.Yang etal.JournalofHematology&Oncology 2014, 7 :5Page14of14 http://www.jhoonline.org/content/7/1/5



PAGE 1

Supplemental data: Figure S1. Flow cytometry assay for monitoring enrichment of the specific aptamer pool against NB4 leukemia cells. After 10 rounds of selection processes, the phycoerythrin (PE) labeled aptamer pool show ed significant increases in fluorescence intensity on target NB4 cells, but it produced minimal change in fluorescence intensity on HL60 cells. These results indicate that the aptamers recognizing target NB4 cells were enriched preferentially.