Establishment of monoclonal anti-human CD26 antibodies suitable for immunostaining of formalin-fixed tissue

MISSING IMAGE

Material Information

Title:
Establishment of monoclonal anti-human CD26 antibodies suitable for immunostaining of formalin-fixed tissue
Physical Description:
Mixed Material
Language:
English
Creator:
Hatano, Ryo
Yamada, Taketo
Matsuoka, Shuji
Iwata, Satoshi
Yamazaki, Hiroto
Komiya, Eriko
Okamoto, Toshihiro
Publisher:
BioMed Central (Diagnostic Pathology)
Publication Date:

Notes

Abstract:
Background: A T cell costimulatory molecule with dipeptidyl peptidase IV (DPPIV) activity in its extracellular region, CD26 is a multifunctional molecule associated with various proteins such as adenosine deaminase, caveolin-1, CXCR4, collagen, and fibronectin, while playing an important role in the regulation of inflammatory responses and tumor biology. We have focused on CD26 as a novel therapeutic target for various tumors and immune disorders, and have developed a humanized anti-CD26 monoclonal antibody (mAb), YS110, which is currently being evaluated in a phase I clinical trial for patients with CD26-expressing tumors, including malignant mesothelioma. Since detection of tumor CD26 expression is required for determining potential eligibility for YS110 therapy, the development of anti-human CD26 mAb that can clearly and reliably detect the denatured CD26 molecule in the formalin-fixed paraffin-embedded tissues is critical. Methods: To develop novel anti-CD26 mAbs capable of binding to the denatured CD26, we immunized mice with CD26 protein denatured in urea buffer. After the fusion of splenocytes and myeloma cells, the mAbs were screened for specific reactivity with human CD26 by flow cytometry, enzyme-linked immunosorbent assay, and immunohistochemistry. The binding competitiveness of novel anti-CD26 mAbs with the humanized anti-CD26 mAb YS110 was also examined. Results: We have succeeded in developing novel anti-human CD26 mAbs suitable for immunohistochemical staining of CD26 in formalin-fixed tissue sections with reliable clarity and intensity. Importantly, some of these mAbs exhibit no cross-reactivity with the humanized anti-CD26 mAb. Conclusions: These novel mAbs are potentially useful as companion diagnostic agents to analyze CD26 expression in the clinical setting while advancing future CD26-related research. Virtual slides: The virtual slides for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/ 5987140221097729 Keywords: CD26/dipeptidyl peptidase 4, Immunohistochemical staining, Companion diagnostic drug, Malignant mesothelioma, T cell costimulation
General Note:
Hatano et al. Diagnostic Pathology 2014, 9:30 http://www.diagnosticpathology.org/content/9/1/30; Pages 1-13
General Note:
doi:10.1186/1746-1596-9-30 Cite this article as: Hatano et al.: Establishment of monoclonal anti-human CD26 antibodies suitable for immunostaining of formalin-fixed tissue. Diagnostic Pathology 2014 9:30.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
© 2014 Hatano et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Resource Identifier:
System ID:
AA00021507:00001

Full Text

PAGE 1

RESEARCHOpenAccessEstablishmentofmonoclonalanti-humanCD26 antibodiessuitableforimmunostainingof formalin-fixedtissueRyoHatano1,TaketoYamada2,ShujiMatsuoka3,SatoshiIwata1,HirotoYamazaki1,ErikoKomiya1, ToshihiroOkamoto1,NamHDang4,KeiOhnuma1andChikaoMorimoto1*AbstractBackground: ATcellcostimulatorymoleculewithdipeptidylpeptidaseIV(DPPIV)activityinitsextracellularregion, CD26isamultifunctionalmoleculeassociatedwithvariousproteinssuchasadenosinedeaminase,caveolin-1, CXCR4,collagen,andfibronectin,whileplayinganimportantroleintheregulationofinflammatoryresponsesand tumorbiology.WehavefocusedonCD26asanoveltherapeutictargetforvarioustumorsandimmunedisorders, andhavedevelopedahumanizedanti-CD26monoclonalantibody(mAb),YS110,whichiscurrentlybeingevaluated inaphaseIclinicaltrialforpatientswithCD26-expressingtumors,includingmalignantmesothelioma.Since detectionoftumorCD26expressionisrequiredfordeterminingpotentialeligibilityforYS110therapy,the developmentofanti-humanCD26mAbthatcanclearlyandreliablydetectthedenaturedCD26moleculein theformalin-fixedparaffin-embeddedtissuesiscritical. Methods: Todevelopnovelanti-CD26mAbscapableofbindingtothedenaturedCD26,weimmunizedmicewith CD26proteindenaturedinureabuffer.Afterthefusionofsplenocytesandmyelomacells,themAbswerescreenedfor specificreactivitywithhumanCD26byflowcytometry,enzy me-linkedimmunosorbentassa y,andimmunohistochemistry. Thebindingcompetitivenessofnovelanti-CD26mAbswiththehumanizedanti-CD26mAbYS110wasalsoexamined. Results: Wehavesucceededindevelopingnovelanti-humanCD26mAbssuitableforimmunohistochemicalstaining ofCD26informalin-fixedtissuesectionswithreliableclarityandintensity.Importantly,someofthesemAbsexhibitno cross-reactivitywiththeh umanizedanti-CD26mAb. Conclusions: ThesenovelmAbsarepotentiallyusefulascompaniond iagnosticagentstoanalyzeCD26expressioninthe clinicalsettingwhileadvancingfutureCD26-relatedresearch. Virtualslides: Thevirtualslidesforthisarticlecanbefoundhere:http://www.diagnosticpathology.diagnomx.eu/vs/ 5987140221097729 Keywords: CD26/dipeptidylpeptidase4,Immunohistochemical staining,Companiondiagnosticdrug,Malignant mesothelioma,Tcellcostimulation *Correspondence: morimoto@ims.u-tokyo.ac.jp1DepartmentofTherapyDevelopmentandInnovationforImmuneDisorders andCancers,GraduateSchoolofMedicine,JuntendoUniversity,2-1-1, Hongo,Bunkyo-ku,Tokyo113-8421,Japan Fulllistofauthorinformationisavailableattheendofthearticle 2014Hatanoetal.;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.Hatano etal.DiagnosticPathology 2014, 9 :30 http://www.diagnosticpathology.org/content/9/1/30

PAGE 2

IntroductionCD26isa110-kDatypeIImembrane-boundglycoproteinwithdipeptidylpeptidaseIV(DPPIV)activityinits extracellulardomain[1-3].CD26iscomposedof766 aminoacids(AAs),andisanchoredtothelipidbilayer byasinglehydrophobicsegmentatresidues7 – 28.The cytoplasmictailofCD26iscomposedofonly6amino acidresiduesattheN-terminalend(AA1 – 6)withouta commonsignalingmotifstructure.Thepredominant partofCD26consistsoftheextracellulardomain(AA 29 – 766)dividedintothreeregions,aglycosylatedregion,acysteine-richregionandaC-terminalDPPIV catalyticregion[4,5].DPPIVbelongstotheserineproteasefamily,abletocleavedipeptidesfrompolypeptides withN-terminalpenultimateprolineoralanine,and regulatestheactivitiesofanumberofcytokinesandchemokines[3].CD26isamultifunctionalmoleculeassociatedwithvariousproteinssuchasadenosinedeaminase (ADA),caveolin-1,CXCR4,collagen,andfibronectin, andisexpressedonvariouscelltypes,includingepithelialcells(kidneyproximaltubules,bileduct,prostate andintestine),endothelialcellsaswellasTlymphocytes [4-6].ThefunctionofCD26isdependentoncelltypes andthemicroenvironment,whichinfluenceitsmultiple biologicalroles[4-7]. InadditiontobeingamarkerofTcellactivation, CD26isassociatedwithTcellsignaltransductionprocessesasacostimulatorymolecule[4].WhileCD26expressionisincreasedfollowingactivationofrestingT cells,CD4+CD26highTcellsrespondmaximallytorecall antigenssuchastetanustoxoid[8].Moreover,crosslinkingofCD26andCD3withsolid-phaseimmobilized monoclonalantibodies(mAbs)caninduceTcellcostimulationandIL-2productionbyCD26+Tcells[4].Furthermore,highCD26cellsurfaceexpressioninCD4+T cellsiscorrelatedwiththeproductionofTH1-typecytokinesandhighmigratoryactivity[4].TakingintoaccountthedatathateffectorTcellsininflamedlesions expresshighlevelsofCD26,itisconceivablethat CD4+CD26+Tcellsplayanimportantroleintheinflammatoryprocess[5,9,10].Wehaverecentlyfoundthat cytotoxicactivityofCD8+Tcellsisalsoregulatedvia CD26-mediatedcostimulation[11].Morerecently,we haveshownthathumanizedanti-CD26mAbappearsto beapromisingnoveltherapyfortheclinicalcontrol ofgraft-versus-hostdisease(GVHD)inaxenogeneic GVHDmurinemodel[12].CD26isalsoexpressedon varioustumorssuchasmalignantmesothelioma,renal carcinoma,coloncancer,hepatocellularcarcinoma,lung cancer,prostatecancer,gastrointestinalstromaltumor (GIST),thyroidcancer,T-lymphoblasticlymphomaand T-acutelymphoblasticleukemia[13].Wehaveshown thatadministrationofanti-CD26mAbresultedinboth invitroandinvivoinhibitionoftumorcellgrowth, migrationandinvasion,andenhancedsurvivalofmouse xenograftmodelsinoculatedwithT-lymphoma,renal cellcarcinomaormalignantmesothelioma[14-16]. Basedonthesefindings,wehavefocusedonCD26asa noveltherapeutictargetforvarioustumorsandimmune disorders,andhavedevelopedahumanizedanti-CD26 mAb,YS110,whichisbeinginvestigatedcurrentlyina phaseIclinicaltrialforpatientswithCD26-expressing tumors,includingmalignantmesothelioma[17]. Thedevelopmentofcompaniondiagnosticagentsto beusedinconjunctionwiththeappropriatetherapeutic modalitiesisessentialtomaximizetherapeuticeffectivenesswhileminimizingassociatedtoxicities.Detection oftumorCD26expressioniscriticaltodetermining potentialeligibilityfortreatmentwithhumanizedantiCD26mAb,anditisalsoimportanttodetermine whetherCD26expressionontumorsorlymphocytesis affectedbyanti-CD26mAbtherapy.ImmunohistochemicalstainingofCD26withthemanyanti-CD26mAbs previouslydevelopedinourlaboratory[18]didnot revealananti-CD26mAbthatcanclearlydetectthe denaturedCD26moleculeinformalin-fixedparaffinembeddedtissues.Meanwhile,testingofseveralcommerciallyavailableanti-CD26mAbsdesignatedasresearch reagentsforimmunohistochemicalstaining,andamAb purchasedfromMBLindicatedthatthesemAbscould stainthedenaturedCD26informalin-fixedtissues,but notwithsufficientclarity.Ontheotherhand,ourtesting ofcommerciallyavailableanti-CD26polyclonalantibodies (pAbs),andapAbpurchasedfromR&DSystemsshowed thatthesereagentsexhibitedreliablestainingpatternand intensity[19].However,thedisadvantageofpAbsisthe potentiallot-to-lotvariabilityinstainingpatternandintensity,whichmakespAbsnottheidealreagentsfordiagnostictestingofpatientspecimensintheclinicalsetting, whereconsistencyanduniformityarerequired. Inthepresentstudy,byimmunizingmicewithCD26 proteindenaturedinureabuffer,wehavesucceededin developingnovelanti-humanCD26mAbsthatcanbe usedascompaniondiagnosticreagentssuitableforimmunohistochemicalstainingofCD26informalin-fixed tissuesectionswithreliableclarityandintensity.In addition,sincesomeofthesemAbsdisplaynocrossreactivitywiththetherapeutichumanizedanti-CD26 mAb,theymaybesuitableforassaysanalyzingCD26 expressionduringorfollowingtreatmentwiththehumanizedanti-CD26mAb.MaterialsandmethodsAnimalsFemaleBALB/cmicewerepurchasedfromCLEAJapan (Tokyo,Japan)andhousedinaspecificpathogen-free facilityinmicro-isolatorcages.AnimalexperimentswereHatano etal.DiagnosticPathology 2014, 9 :30 Page2of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 3

conductedfollowingprotocolsapprovedbytheAnimal CareandUseCommitteeatJuntendoUniversity.AntibodiesTodeterminetheepitopeofthenewlydevelopedmouse anti-humanCD26mAbs,murineanti-humanCD26 mAbs(clone4G8,1F7,14D10,5F8,16D4Bor9C11) whichhavebeenalreadydevelopedinourlaboratory wereused[18].Wehavepreviouslyshownthatthese mAbsaredividedinto5separategroupsbytheirepitopes,4G8recognizingthe1-247thAAsregionof CD26,1F7and14D10recognizingthe248-358thAAs regionofCD26,5F8recognizingthe359-449th(closer tothe359th)AAsregionofCD26,16D4Brecognizing the450-577thAAsregionofCD26,and9C11recognizingthe359-653th(butdifferentfrom5F8or16D4B) AAsregionofCD26.Thehumanizedanti-CD26mAb (YS110)wasgeneratedbyutilizingthecomplementarity determiningregionsofthemurineanti-humanCD26 mAb14D10[18],andgenerouslyprovidedbyY'sTherapeutics(Tokyo,Japan).Tocomparethestainingpattern andintensityofhumanCD26onformalin-fixedtissue sections,weusedtwocommercialanti-humanCD26 AbsavailableforCD26detectionbyimmunohistochemistry.Oneistheculturesupernatantformofamouse anti-humanCD26mAb(clone44 – 4)purchasedfrom MBL(Nagoya,Japan),andtheotherisapurifiedgoat anti-humanCD26pAbpurchasedfromR&DSystems (Minneapolis,MN).HumanpolyclonalIgG(venilon-I) waspurchasedfromAlfresaCorporation(Tokyo,Japan), andmouseIgG1isotypecontrol(cloneMG1-45)was purchasedfromBioLegend(SanDiego,CA).YS110, controlhumanIgG,4G8,1F7,5F8,16D4B,9C11,purifiedclone18,clone19,andmouseIgG1isotypecontrol werelabeledusinganAlexaFluor647MonoclonalAntibodyLabelingKit(MolecularProbes,Eugene,OR)accordingtothemanufacturer'sinstructions.cDNAconstructsandtransfectionAsdescribedpreviously[18],C-terminaldeletionmutantsofhumanCD26cDNAconstructsweregenerated byusingNcoIrestrictionenzymesitestodeletedomain representingthe740-766thAAsintheCterminus, usingNheItodeletefromthe578thAA,usingBspEI todeletefromthe450thAA,usingStuItodeletefrom the359thAA,andusingPstItodeletefromthe248th AA.ThesecDNAswereligatedin-frameintopcDL-SR expressionvector[20].Thegreenfluorescenceprotein (GFP)-expressingvectorpEB6-CAG-GFPwasakindgift fromDr.YoshihiroMiwa(TsukubaUniversity,Tsukuba, Japan)[21].EachCD26deletionconstructinpcDL-SR wasco-transfectedwithpEB6-CAG-GFPintoCOS-7 cellsusingLipofectamine2000reagent(Invitrogen, Carlsbad,CA).After24hoursoftransfection,cellswere harvested,followedbystainingwithAlexaFluor647labeled4G8,1F7,5F8,YS110,clone18orclone19,and thenanalyzedbyflowcytometry.PreparationofimmunogenSolubleCD26(sCD26)wasproducedaccordingtothe methoddescribedpreviously[22].Briefly,theexpression vectorRcSR -26d3-9,whichcontainsadeletionofthe codingsequenceforaminoacids3 – 9ofCD26,was transfectedintoadihydrofolatereductasedeficitChinese hamsterovary(CHO)cellline,DXB-11byelectroporation,togetherwithpMT-2providingthedihydrofolate reductasegene.ThetransfectedCHOcellswereculturedinserum-freeCHO-S-SFMIImedium(Invitrogen) supplementedwith1 Mmethotrexate(NacalaiTesque, Kyoto,Japan).Theculturesupernatantwascollected andsubjectedtoaffinitychromatographyonADASepharoseaccordingtothemethoddescribedpreviously [23].PurifiedsCD26wasdenaturedin8Mureabuffer supplementedwith20mMHEPESand50mMdithiothreitol(DTT)bygentlerotationfor8hoursatRT.Developmentofhybridomasandmonoclonalanti-human CD26antibodiesDenaturedsCD26wasdialyzedinPBS,and100 gofproteinper50 lofPBSwasemulsifiedwith50 lofadjuvant, TiterMaxGold(TiterMaxUSA,Norcross,GA).A6-wkoldfemaleBALB/cmousewasimmunizeds.c.with100 l oftheemulsionseventimeseverytwoweeksandfinally injectedi.v.withhalfvolumeoftheemulsion.Threedays afterthefinalimmunization,thespleenwasremovedand 100106spleencellswerefusedwith100106P3U1 myelomacellsbyusingpolyethyleneglycol4000(Merck, Darmstadt,Germany)andwereculturedinRPMI1640 supplementedwith10%fetalbovineserum(FBS,Japan Bioserum,Fukuyama,Japan),5%BriClone(NICB,Dublin, Ireland)andHAT(Invitrogen)in96-wellflat-bottom plates(Costar,CorningIncorporated,Corning,NY).Hybridomasupernatantswerescreenedforselectivereactivity withhumanCD26byusingflowcytometryandenzymelinkedimmunosorbentassay(ELISA).Thesupernatants whichcandetecthumanCD26bybothflowcytometry andELISAwerefinallyscreenedforimmunostainingof formalin-fixedparaffin-embeddedhumantissuesections. Thehybridomaswereclonedbylimitingdilutionandculturemediumwasexchangedforserum-freeGITmedium (WakoPureChemicals,Osaka,Japan).MonoclonalantibodieswerepurifiedfromthesupernatantsusingProtein AIgGPurificationKit(Pierce,Rockford,IL)accordingto themanufacturer ’ sinstructions.FlowcytometryACD26-negativeJurkatTcellline(Jurkatparent)anda stableJurkatTcelllinetransfectedwithhumanCD26Hatano etal.DiagnosticPathology 2014, 9 :30 Page3of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 4

cDNA(Jurkat-CD26WT)describedpreviously[24]were usedforscreeningofhybridomas.Cellswerewashedin PBScontaining1%FBS,1mMEDTAand0.1%sodium azide,andincubatedwith100 lofhybridomasupernatantor20 g/mlofpurifiedmouseanti-humanCD26 mAbfor25minat4C,andsubsequentlystainedwith PE-conjugatedgoatanti-mouseIgpAb(BDBiosciences, SanJose,CA)for25minat4C.AcquisitionwasperformedusingFACSCalibur(BDBiosciences)anddata wereanalyzedwithFlowJosoftware(TreeStar,Ashland, OR).Forcross-blockingstudiesofhumanizedantiCD26mAb(YS110),cellswerepretreatedwithunlabeledYS110orcontrolhumanIgG(50 g/ml,respectively)for25minat4C,andsubsequentlyincubated with100 lofhybridomasupernatantor20 g/mlof purifiedmouseanti-humanCD26mAbfor25minat 4C,andfinallystainedwithPE-conjugatedgoatantimouseIgpAbfor25minat4C.Forcross-blocking studiesofmurineanti-CD26mAbs,cellswerepretreated withunlabeled4G8,1F7,5F8,16D4B,9C11ormouse IgG1isotypecontrol(50 g/ml,respectively)for25min at4C,andsubsequentlystainedwithAlexaFluor647labeledclone18orclone19orPE-conjugatedgoatantimouseIgpAbfor25minat4C.ELISAThe96-wellimmunoplates(NUNC,Roskilde,Denmark) werecoatedwithnativesCD26ordenaturedsCD26describedaboveincarbonatebicarbonatebuffer(200ng/ well,respectively)orbufferaloneasanegativecontrol at4Covernight.Eachwelloftheplatewasblockedwith 3%bovineserumalbumin(BSA,Sigma,St.Louis,MO) inPBSfor1houratRT,andthenincubatedwith3-fold dilutedhybridomasupernatantsor5 g/mlofpurified mouseanti-humanCD26mAborgoatanti-human CD26pAbfor1houratRT,andsubsequentlyincubated withhorseradishperoxidase(HRP)-conjugatedgoatantimouseIgpAb(BDBiosciences)orHRP-conjugated donkeyanti-goatIgGAb(SantaCruzBiotechnology, SantaCruz,CA)for1houratRT.Tetramethylbenzidine (TMB)PeroxidaseSubstrate(KPL,Gaithersburg,MD) wasfinallyaddedtoeachwellandthereactionwas stoppedby2NH2SO4.Theabsorbanceat450nm/ 570nmwasmeasuredinaMicroplateReader(Bio-Rad, Hercules,CA)anddatawereanalyzedwithMicroplate Manager6software(Bio-Rad).TissuespecimensandimmunohistochemicalstainingFormalin-fixedparaffin-embeddedtissuespecimensof malignantmesotheliomaandnormalliver,kidneyand prostatewereusedforpositivecontrolsintheimmunohistochemicalexamination.Theuseofhumansample fromautopsycaseswithhepatocellularcarcinoma,renal cellcarcinoma,prostateadenocarcinoma,colonadenocarcinomaandlungadenocarcinomawasgenerously permittedbythebereavedfamilies.ThisstudywasapprovedbytheOkayamaRosaiHospitalethicalreview boardandtheKeioUniversitySchoolofMedicineethicalreviewboard,andthepurposeofthestudywas explainedtoallpatientsandtheirwritteninformedconsentwasobtained.Allstudiesonhumansubjectswere carriedoutaccordingtotheprinciplessetoutin theDeclarationofHelsinki.Formalin-fixedparaffinembeddedtissuespecimenswerecutinto4-6 m sectionsanddeparaffinized.Antigenretrievalwasperformedby1)autoclavingin10mMcitratebuffer (pH6.0)for20minat120C,2)0.05%trypsinfor 15minat37C,3)0.02%proteinaseKfor10minat 37C,or4)boilingin10mMcitratebuffer(pH6.0) for10minat100C,andthesectionsweretreated with0.3%H2O2inmethanolfor10minatRTtoinactivateendogenousperoxidase,thentreatedwith2.5% horseserum(VectorLaboratories,Burlingame,CA)for 10minatRTtoblocknon-specificbindingofthesecondaryhorseantibody.Thesectionsweretreatedwith 100 lofhybridomasupernatantsorpurifiedmouse anti-humanCD26mAborgoatanti-humanCD26pAb for2hoursatRT,andsubsequentlytreatedwithHRPconjugatedhorseanti-mouseIgpAborHRP-conjugated horseanti-goatIgGpAb(VectorLaboratories)for30min atRT.Thereactionwasvisualizedwith3,3 ’ -diaminobenzidine(DAB)(DojindoLaboratories,Kumamoto,Japan), andthetissuesectionswerecounterstainedfornucleus withhematoxylin.Toconfirmthebindingspecificityof AbstohumanCD26,theanti-humanCD26Ab(100 g/ ml)wasgentlyrotatedwith500 g/mlofsCD26at4C overnight,andaftercentrifugation,thesupernatantwas usedinsteadoftheprimaryanti-humanCD26Ab.ExpressionpatternofCD26wasevaluatedandverifiedindependentlybytwopathologists.Theopticalmicroscopeimages weretakenusingAxioScope.A1microscope(CarlZeiss, Oberkochen,Germany).ResultsScreeningofhybridomacellsTodevelopanovelanti-CD26mAbcapableofbinding tothedenaturedCD26,weimmunizedmicewithCD26 proteindenaturedinureabuffer.Todeterminethedenaturingcondition,weincubatedCD26proteinin8M ureabufferatRTfor30min,3hoursor12hours,and analyzedthebindingofanti-CD26mAb(clone5F8)or anti-CD26pAb(R&DSystems)totheureatreated CD26proteinbyELISAasdescribedinMaterialsand Methods.ThisanalysisshowedthedecreaseintheabsorbancewhenCD26proteinwasincubatedfor30min inureabuffer,withadditionaldecreaseinabsorbanceat 3hoursofincubation,whiletherewasbarelynoticeable differencebetween3hoursofincubationand12hoursHatano etal.DiagnosticPathology 2014, 9 :30 Page4of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 5

ofincubation(datanotshown).Thesedatasuggestthat mostoftheCD26proteinsweredenaturedwhenincubatedin8Mureabufferformorethan3hours,andwe usedthisurea-treatedCD26proteinasanimmunogen. AfterthefusionofsplenocytesandP3U1myeloma cells,theculturesupernatantwascollectedandscreened forselectivereactivitywithhumanCD26.Forthefirst screeningofhybridomacells,weusedanendogenous CD26-deficitJurkatcellline(Jurkatparent)andastable Jurkatcelllinetransfectedwithfull-lengthhumanCD26 (Jurkat-CD26WT),andthebindingtohumanCD26was analyzedbyflowcytometry.AsshowninFigure1A,we obtainedanumberofhybridomassecretingantibodies, someofwhichcouldstainJurkat-CD26WTwithbright intensityandotherscouldstainwithintermediateor dullintensity(redlines)whileJurkatparentcellsshowed nostainingwithallofthesesupernatants(bluelines). Thesedataindicatethatthisscreeningmethodexcludes BMBL(44-4)MFI 144 MFI 303CD26mAb(5F8)CD26Relative Cell Numberclone 1clone 5clone 11 clone 16clone 18clone 19MFI:66 MFI:59 MFI:34MFI:181MFI:114 MFI:60 ARelative Cell NumberMFI 13-25 MFI 25-100 MFI 100-200CD26 MFI >200 Figure1 Flowcytometryanalysiswithnovelanti-CD26mAbs.A. Jurkat-CD26WTcells(redlines)orJurkatparentcells(bluelines)were incubatedwiththehybridomasupernatant,andsubsequentlystainedwithPE-labeledanti-mouseIgpAb,andanalyzedbyflowcytometry. B. Jurkat-CD26WTcellswereincubatedwiththehybridomasupernatant(clone1,5,11,16,18or19)orpurifiedmouseanti-CD26mAb(5F8) orcommercialmouseanti-CD26mAb(MBL,clone44 – 4),andsubsequentlystainedwithPE-labeledanti-mouseIgpAb,andanalyzedbyflow cytometry.Thegrayareasineachhistogramshowthedatainvolvingtheisotypecontrol.Themeanfluorescenceintensity(MFI)ofeachstaining isshown.Datashownarerepeatedtwice (A) andfivetimes (B) withsimilarresults. Hatano etal.DiagnosticPathology 2014, 9 :30 Page5of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 6

thepossibilityofnon-specificbindingtootherproteins besideCD26.Therepresentativehistogramsofthese novelanti-CD26mAbsavailableforimmunohistochemicalstainingwereshowninFigure1B. Thepositivesupernatantswerethenscreenedby ELISAforreactivitywithnativeordenatured(urea treated)sCD26protein.Toexcludethepossibilityof non-specificbindingtoBSAusedforblocking,we preparedthewellscoatedwithbufferalone(without sCD26),subsequentlyblockedwithBSAandincubated withhybridomasupernatants.Theabsorbanceofthe wellsat450nmwassubtractedfromtheabsorbanceof thewellscoatedwithnativeordenaturedsCD26.The clonewasjudgedtobepositiveiftheabsorbancetothe nativesCD26washigherthan0.1.Theabsorbanceto thenativeordenaturedsCD26wasquitedifferentfrom clonetoclone,andtherepresentativeabsorbanceof novelanti-CD26mAbsavailableforimmunohistochemicalstainingwasshowninFigure2.WhensCD26was denaturedinureabuffer,theabsorbanceof5F8,which cannotdetectdenaturedCD26informalin-fixedtissues, wasapparentlydecreased,whiletheabsorbanceof commercialmAb(purchasedfromMBL)orpAb(purchasedfromR&DSystems)wascomparativelymaintained(Figure2).Althoughthedecreaseofabsorbance tothedenaturedsCD26wasalsoobservedwiththe novelanti-CD26mAbs,particularlywithclone1,clone 11andclone16,theabsolutevalueofabsorbancetothe denaturedsCD26wasmuchhigherthanthatof5F8,exceptforclone11(Figure2).Asaresultofthescreening, 31clonesthatsecretedanti-humanCD26mAbswere evaluatedforbothflowcytometryandELISA.Immunohistochemicalstainingwithnovelanti-CD26 mAbsTodeterminewhetherthenewlydevelopedanti-CD26 mAbsweresuitableforimmunohistochemicalstaining ofCD26informalin-fixedtissuesections,surgically resectedtissuespecimensofnormalliver,kidney,prostate,andmalignantmesotheliomawereimmunostained withthesemAbs,withcommercialanti-CD26mAb (purchasedfromMBL)andanti-CD26pAb(purchased fromR&DSystems)beingusedascontrols.Although weexaminedseveralantigenretrievalconditions,tissue specimensstainedwithanti-CD26mAbpurchased fromMBLexhibitedonlyaslightlypositivereaction withweakstainingintensity,revealingthismAbtobe inappropriateforthedetectionofCD26expressionin formalin-fixedclinicalsamples(Figure3A-i).Incontrast, tissuespecimensstainedwithanti-CD26pAbpurchased fromR&DSystemsexhibitedaclearstainingpatternof CD26,namelythesurfacemembraneofbilecanaliculi, thebrushborderofrenalproximaltubularepithelial cellsandprostateepithelialcellswerespecificallystained withlowbackground(Figure3A-ii).Wehavepreviously shownthatCD26wasalsohighlyexpressedinvarious pathologictypesofmalignantmesothelioma,including localizedmalignantmesothelioma,well-differentiated papillarymalignantmesothelioma,anddiffusemalignant mesothelioma[16],andthespecificstainingofmalignantmeshotheliomacellswasalsoobservedwiththe useoftheanti-CD26pAb(Figure3A-ii).Aftertesting thehybridomasupernatantsfromthe31clonesdescribedaboveforimmunohistochemicalstaining,we finallyobtained6clones(clone1,5,11,16,18or19) capableofstainingCD26informalin-fixedtissueswith muchstrongerintensitythanthemAbpurchasedfrom MBL.AsshowninFigure3A,tissuespecimensstained withtworepresentativeclones(clone18or19)exhibited reliablestainingpatternandintensitycomparabletothe pAbpurchasedfromR&DSystems(panelsiiiandiv), whilenoapparentstainingofCD26wasobservedinthe specimensstainedwithclone3(judgedtobenegative forimmunostaining)(panelv).Representativeresultsof immunostainingwiththeother4clones(clone1,5,11 or16)wereshowninAdditionalfile1:FigureS1. Wenextexaminedimmunohistochemicalstaining withpurifiednovelmAbsinsteadofthehybridoma culturesupernatants.TodeterminetheoptimalAb concentrationforimmunostaining,weevaluatedthe anti-humanCD26Absinconcentrationsrangingfrom 1 g/mlto100 g/ml.AsshowninFigure3B,staining ofmalignantmesotheliomacellswashardlyobserved with1 g/mlofclone18,clone19mAborpAbpurchasedfromR&DSystems,whilethestainingintensity wasenhancedinadose-dependentmannerupto 100 g/mlofthesethreeAbs(panelsi,ii,iii). 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6Absorbance at 450 nm15111618195F8MBLR&D Native sCD26 Denatured sCD26 Figure2 ELISAanalysiswithnovelanti-CD26mAbs. Nontreated nativesolubleCD26(sCD26)orureatreateddenaturedsCD26was incubatedwiththehybridomasupernatant(clone1,5,11,16,18or19) orpurifiedmouseanti-CD26mAb(5F8)orcommercialmouseantiCD26mAb(MBL,clone44 – 4)orpurifiedgoatanti-CD26pAb(R&D Systems).Theabsorbanceat450nm/570nmwasmeasured,anddata areshownasmeanS.E.fromthreeindependentexperiments. Hatano etal.DiagnosticPathology 2014, 9 :30 Page6of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 7

(i) MBL (ii) R&D (iii) clone 18 (iv) clone 19 (v) clone 3 Anti-CD26 Ab :LiverKidneyProstateMesotheliomacase1 case2A(i) R&D (ii) clone 18 (iii) clone 19BAnti-CD26 Ab : 100 g/ml 1 g/ml 10 g/ml100 g/ml sCD26 pretreatC(i) R&D (ii) clone 18 (iii) clone 19 Anti-CD26 Ab : Lung adenocarcinoma Prostate adenocarcinoma Hepatocellular carcinoma Renal cell carcinoma Colon adenocarcinoma Figure3 (Seelegendonnextpage.) Hatano etal.DiagnosticPathology 2014, 9 :30 Page7of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 8

Meanwhile,stainingoftissueswithevenhigherAbconcentrationsresultedinsimilarintensityascompared withthosestainedwith100 g/mloftheAbs(datanot shown).Inaddition,toconfirmthebindingspecificityof theseAbstohumanCD26,thesectionsweretreated withpurifiedanti-humanCD26Abpreincubatedwith sCD26.AsshowninFigure3B,thebindingoftheseAbs wascompletelyinhibitedbysCD26(panelsi,ii,iii). TheseresultsindicatethatthenewlydevelopedmAbs specificallybindtohumanCD26,and100 g/mlseems tobeanoptimalconcentrationoftheseAbsforimmunohistochemicalstaining. Wefurtherexaminedimmunohistochemicalstaining ofCD26-expressingtumortissuesotherthanmalignant mesothelioma(hepatocellularcarcinoma,renalcellcarcinoma,prostateadenocarcinoma,colonadenocarcinoma,andlungadenocarcinoma)withthepurifiedmAb ofclone18or19.AsshowninFigure3C,eachtumor tissuestainedwithclone18or19(panelsiiandiii)exhibitedclarityandintensitysimilartothelevelsobserved withtheanti-CD26pAbpurchasedfromR&DSystems (paneli).ResultsfromtheimmunostainingstudiesindicatethatCD26canbedetectedbothonthecellsurface aswellascytoplasmofthesecarcinomatissues.Cross-blockingstudieswithhumanizedanti-CD26mAbInadditiontodetectingCD26expressionontumorcells orlymphocytespriortothetherapeuticadministration ofhumanizedanti-CD26mAb,itisalsoimportantto evaluatewhetheranti-CD26mAbtherapyaffectsCD26 expressiononrelevanttissues.Forthispurpose,wenext examinedthebindingcompetitivenessofthe6novel anti-CD26mAbswiththehumanizedanti-CD26mAb YS110.Jurkat-CD26WTwaspretreatedwithunlabeled YS110orcontrolhumanIgGfor25min,subsequently incubatedwithhybridomasupernatants,andstained withPE-labeledanti-mouseIgpAb.AsshowninFigure4 (representativehistogramsareshowninAdditionalfile 1:FigureS2),bindingofYS110or1F7toCD26was completelyblockedbyYS110whilethebindingof5F8 toCD26washardlyaffected,indicatingthatYS110was sufficientlyboundtoCD26.Althoughbindingofclone 1,11,16or19toCD26washardlyaffectedbyYS110 pretreatment,bindingofclone5waspartiallyinhibited, andbindingofclone18wascompletelyinhibitedby YS110(Figure4andAdditionalfile1:FigureS2).Taken together,thesedatasuggestthatclone19wascapableof detectingdenaturedCD26informalin-fixedtissuesectionswiththemostreliablestainingpatternandintensity,exhibitednocross-reactivitywithYS110,andwas suitableforanalysisofCD26expressiononclinicalsamplesfollowingtheadministrationofYS110.Epitopemappingofnovelanti-CD26mAbsTodefinetheCD26epitoperecognizedbyclone18and 19,weconductedcross-blockingstudiesusingantiCD26mAbswithepitopesthathadbeenextensively characterizedpreviouslyasdescribedinMaterialsand Methods[18].Toconfirmthebindingofanti-CD26 mAbstoCD26,Jurkat-CD26WTwasincubatedwithunlabeled4G8,1F7,5F8,16D4B,9C11ormouseIgG1isotypecontrolfor25min,andsubsequentlystainedwith PE-labeledanti-mouseIgpAb.AsshowninFigure5-i, eachanti-CD26mAbwassufficientlyboundtoCD26 (Seefigureonpreviouspage.) Figure3 Representativeresultsofimmunostainingwithnovelanti-CD26mAbs.A. Thetissuespecimensofliver,kidney,prostateortwo casesofmalignantmesotheliomawerestainedwith100 lofcommercialmouseanti-humanCD26mAbsupernatant(MBL,clone44 – 4)(i),or 10 g/mlofpurifiedgoatanti-humanCD26pAb(R&DSystems)(ii),ornewlydevelopedhybridomasupernatant(clone18(iii),clone19(iv)or clone3(v)). B .Malignantmesotheliomatissuespecimenswerestainedwithcommercialgoatanti-humanCD26pAb(R&DSystems)(i),orpurified novelmouseanti-humanCD26mAbs(clone18(ii)orclone19(iii))attheindicatedconcentrationsofAbsinthepresenceorabsenceofsCD26. C. The tissuespecimensofhepatocellularcarcinoma,renalcellcarcinoma,prostateadenocarcinoma,colonadenocarcinomaorlungadenocarcinomawere stainedwith100 g/mlofcommercialgoatanti-humanCD26pAb(R&DSystems)(i),orpurifiedmouseanti-humanCD26mAbs(clone18(ii)orclone 19(iii)).Allspecimenswerecounterstainedwithhematoxylin(originalmagnification,200X). %MFI to control IgG blockingStainin g antibody after YS110 blockin g 02040 60 80 10015111618191F75F8YS110 Figure4 Analysisofcrossreactivityofnovelanti-CD26mAbs withhumanizedanti-CD26mAb. Jurkat-CD26WTcellswere pretreatedwithunlabeledhumanizedanti-CD26mAb(YS110)or humancontrolIgG,andthentreatedwiththehybridomasupernatant (clone1,5,11,16,18or19)orpurifiedmouseanti-CD26mAb(1F7or 5F8),andsubsequentlystainedwithPE-labeledanti-mouseIgpAb.For stainingwithhumanizedanti-CD26mAb,cellswerestainedwithAlexa Fluor647-labeledYS110afterpretreatmentwithunlabeledYS110. Datawereanalyzedbyflowcytometry,andthepercentageofmean fluorescenceintensity(MFI)afterYS110blockingtoMFIaftercontrol IgGblockingisshown.Datashownarerepeatedtwicewith similarresults. Hatano etal.DiagnosticPathology 2014, 9 :30 Page8of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 9

(bluelines),whiletherewasnobindingoftheisotype control(redline).ModulationofcellsurfaceCD26into thecytoplasmfollowingtreatmentwiththeseanti-CD26 mAbsdidnotoccurundertheseexperimentalconditions.Similarly,Jurkat-CD26WTwaspretreatedwith unlabeledanti-CD26mAbs,andsubsequentlystained withAlexaFluor647-labeledclone18or19.Asshown inFigure5,bindingofclone18or19toCD26wascompletelyinhibitedby1F7(panelsii)or4G8(panelsiii),respectively,withnoeffectbytheotheranti-CD26mAbs. Theseresultssuggestthattheepitopedefinedbyclone 18mightbeidenticalto1F7,locatingbetweenthe248358thAAsregionofCD26,whiletheepitopedefinedby clone19mightbeidenticalto4G8,locatingbetweenthe 1-247thAAsregionofCD26. Forcross-blockingstudiesinvolvingtheother4novel anti-CD26mAbs,Jurkat-CD26WTwasincubatedwith unlabeledclone1,5,11,16,18,19ormouseIgG1isotypecontrolfor25min,andsubsequentlystainedwith AlexaFluor647-labeled4G8,YS110,5F8,16D4Bor 9C11.AsshowninAdditionalfile1:FigureS3,binding of4G8toCD26wascompletelyblockedbyclone19, andbindingofYS110toCD26wascompletelyinhibited byclone18andpartiallyinhibitedbyclone5,consistent withtheresultsshowninFigures4and5.Clone1 blockedcompletelythebindingof9C11andpartiallythe bindingof16D4BtoCD26,whileclone16completely inhibitedthebindingofboth9C11and16D4BtoCD26 (Additionalfile1:FigureS3).Ontheotherhand,clone 11inhibitedthebindingof5F8toCD26completely (Additionalfile1:FigureS3).Theseresultsstronglysuggestthatthenovelanti-CD26mAbshaveawiderange ofepitopesandcanbebroadlydividedinto4separate groups;theepitopeofclone19beingsimilarto4G8,the epitopesofclone5and18beingsimilarto1F7and YS110,theepitopeofclone11beingsimilarto5F8,and theepitopesofclone1and16beingsimilarto9C11 (clone16isalsosimilarto16D4B). Tofurtherconfirmtheepitopeinvolvedinbindingof clone18and19tohumanCD26,wetestedtheabilityof thesetwomAbstobindtoCD26deletionmutantsbyflow cytometry[18].Wefir sttestedthebindingofthepreviously Contl. IgG4G81F75F816D4B9C11Anti-Mouse Ig-PEBlocking clone 18-AlexaFluor647 clone 19-AlexaFluor647 (i) (ii) (iii) 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 200 100 80 60 40 20 0 Figure5 Blockingexperimentofnovelanti-CD26mAbbindingtoCD26. Jurkat-CD26WTcellswerepretreatedwithunlabeledmouse anti-CD26mAbs(4G8,1F7,5F8,16D4B,or9C11)(bluelines)ormouseIgG1isotypecontrol(Contl.IgG)(redlines),andsubsequentlystained withPE-labeledanti-mouseIgpAb (i) orAlexaFluor647-labeledanti-CD26mAbs(clone18 (ii) orclone19 (iii) ),andanalyzedbyflowcytometry. TherepresentativehistogramsofCD26expressionareshown,andthegrayareasineachhistogramshowthedatainvolvingtheisotypecontrol. Datashownarerepeatedtwicewithsimilarresults. Hatano etal.DiagnosticPathology 2014, 9 :30 Page9of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 10

developedanti-humanCD26mAbs,4G8,1F7,or5F8to confirmtheexpressionpatternofCD26deletionmutants onCOS-7cells.AsshowninAdditionalfile1:FigureS4, 4G8recognizedfull-lengthCD26andall5CD26deletion mutantswhile1F7or5F8losttheabilitytorecognizethe CD26moleculewithdeletionfromthe248thAAorfrom the359thAA,respectively,ind icatingthattheexpression patternsofCD26deletionmutantswereidenticaltothose reportedpreviously[18].Wethenanalyzedthebindingof YS110,clone18andclone19totheCD26deletionmutants.AsshowninFigure6(representativehistogramsare showninAdditionalfile1:FigureS4),bothYS110and clone18recognizedfulllengthCD26,the1-739thAAsregionofCD26,the1-577thAAsregionofCD26,the1449thAAsregionofCD26andthe1-358thAAsregionof CD26,butlosttheabilitytorecognizetheCD26molecule withdeletionfromthe248thAA,suggestingthatthesequenceofthe248-358thAAsregiononCD26mightbe importantforbindingofYS110andclone18.Ontheother hand,clone19recognizedfull-lengthCD26andall5CD26 deletionmutants,suggestingthattheepitopedefinedby clone19mightbelocatedbetweenthe1-247thAAregion (Figure6andAdditionalfile1:FigureS4).YS110,clone18 andclone19didnotbindtoCOS-7cellstransfectedwith vectoralone(mock)(Figure6andAdditionalfile1:Figure S4).Takentogether,resultsfromthecross-blockingstudies andthoseinvolvingCD26deletionmutantsstronglysuggestthattheepitopedefinedbyclone19maybelocated betweenthe1-247thAAsregion,andtheepitopedefined byclone18betweenthe248 -358thAAsregion,being almostidenticaltoYS110.DiscussionAlthoughanti-humanCD26mAbswhichwehavedevelopedpreviouslyorcommerciallyavailablemAbscannot clearlydetectdenaturedCD26informalin-fixedparaffinembeddedtissues,theanti-humanCD26pAbpurchased fromR&DSystemsisabletostainCD26withreliableclarityandintensity.However,itisofconcernthatthestainingpatternandintensitymaydifferamongdifferentlots oftheanti-CD26pAb.Sincetreatmentwithtargeted therapeuticagentsdependsontheabilitytoreliablydetect theappropriatetargetsonclinicalsamples,uniformityof thediagnosticreagentsiscritical,suggestingthatpAbs thatareusedasresearchreagentsarenotappropriatefor diagnosticusesintheclinicalsetting.Inthepresentstudy, wedescribethesuccessfuldevelopmentofnovelantihumanCD26mAbsbyimmunizingmicewithCD26proteindenaturedinureabufferthatcanpotentiallybeused asdiagnosticreagentsclinically. Inanattempttoimprovediagnosticaccuracy,markers usedforimmunohistochemistryhavebeenstudied,such asgalectin-3,HBME-1andCK-19fordiagnosisof benignandmalignantthyroidlesions[25,26],andFAPandCalponinfordiagnosingwhetherductalcarcinoma insituhasmicroinvasion[27].CD26ishighlyexpressed onthesurfaceofmalignantmesotheliomacellsespeciallytumorsoftheepitheloidandbiphasictypes,but notonbenignmesothelialtissues[16,17].IthasbeenrecentlyreportedthattheexpressionlevelofCD26in prostatecancertissuesishigherthanthatofnormal prostatictissuesandincreasedwithprostatecancerstage advancement,andCD26expressioniscorrelatedwith prostatespecificantigen,suggestingthatCD26maybea goodmarkerforprostatecancerdiagnosis[28].Furthermore,theoverallsurvivalofpatientswithCD26-positive GISTsisworsethanthatofpatientswithCD26-negative GISTs,suggestingthatCD26appearstobeareliable biomarkerofmalignantGISTofthestomach[29].These observationsstronglysuggestthatimmunohistochemical 0 10 20 30 40 50 60 70 8090YS110clone18clone19 1-766th AAs 1-739th AAs 1-577th AAs 1-449th AAs 1-358th AAs Mock%positive cells1-247th AAs Figure6 StainingforCD26expressiononCOS-7cellstransfectedwithCD26deletionmutantsbynovelanti-CD26mAbs. cDNAof deletedCD26wascotransfectedwithGFP-expressingplasmidtoCOS-7cells.After24h,thetransfectedcellswerestainedwithAlexaFluor647labeledanti-CD26mAbs(YS110,clone18orclone19)orisotypecontrol,andanalyzedbyflowcytometry.FollowinggatingforGFPpositivecells amongallacquiredcells,thepercentageofCD26positivecellswasanalyzed.Datashownarerepeatedtwicewithsimilarresults. Hatano etal.DiagnosticPathology 2014, 9 :30 Page10of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 11

stainingofCD26informalin-fixedtumortissuesisimportantfordiagnosisandprognosisofvarioustumors. Sinceseveralanti-humanCD26mAbssuchasTa1, 1F7,5F8and14D10thatwerealreadydevelopedinour laboratorybyimmunizingmicewithaCD26positive humanTcellline(EL156)oraPHA-stimulated Aotus trivirgatus Tcelllineoramurinepre-BhumanCD26 transfectant(300 – 19)cannotclearlydetectCD26in formalin-fixedtissues[1,8,18],itwasourhypothesisthat utilizinghumanCD26proteinbutnothumanCD26 positivecellsasanimmunogenwouldbeimportantfor thedevelopmentofmAbscapableofrecognizingthedenaturedCD26molecule.SimilartoCD26,onlypAbs couldreacttothedenaturedHLAclassImoleculesin formalin-fixedparaffin-embeddedtissues.Torigoeetal. recentlysucceededindevelopinganovelanti-panHLA classImAbsuitableforimmunohistochemicalstaining offixedtissuesbyimmunizingarecombinantHLA-A proteindenaturedinureabuffer[30].Theexactrole playedbyureatreatmentoftheCD26proteinin expandingtherepertoireoftheobtainedanti-CD26 mAsisnotyetclear,sincewehavenotexaminedforpotentialdifferencesinthecharacteristicsofmAbsobtainedafterimmunizingmicewithurea-treatedsCD26 proteinornon-treatednativesCD26proteininthis study.However,asshowninFigures2and3,tissue specimensstainedwithanti-CD26mAbpurchasedfrom MBLexhibitedonlyapartiallypositivereactionwith weakstainingintensity,whilethismAbshowedhigher absorbancetotheureatreatedsCD26proteinthanthe absorbanceobtainedfromthenovelanti-CD26mAbs capableofstainingCD26withstrongintensityinfixed tissues.Thesedatastronglysuggestthatthestructureof CD26denaturedbythemethodofantigenretrievalafter formalin-fixationisquitedifferentfromthatofCD26denaturedbyureabuffer,andalsosuggestthatanti-CD26 mAbssuitableforimmunohistochemistrymaybeobtainedmoreefficientlybyimmunizingmicewithCD26 proteindenaturedbymethodsotherthanureatreatment, suchastreatmentwithguanidinehydrochlorideorsodium dodecylsulfate(SDS),orwithproteasessuchastrypsinor proteinaseK,orbyboiling.Furtherstudiesareneededto clarifytheissueinvolvingpretreatmentoftheimmunogensandthecharacteristicsofmAbsobtainedafter immunization. Inthepresentstudy,wehavesucceededindeveloping novelanti-CD26mAbswithawiderangeofepitopes (Figures5,6andAdditionalfile1:FigureS3).Sincemost ofthesenovelmAbscompletelyinhibitedthebindingof theanti-CD26mAbs(4G8,1F7,5F8,16D4Bor9C11) developedpreviouslybyourgroup,theepitopesdefined bythesenovelmAbsareexpectedtobesimilartothose recognizedbytheearliermAbs.However,thesenovel anti-CD26mAbsarecapableofdetectingdenatured CD26infixedtissueswithstrongintensity,unlikethe previouslydevelopedmAbs.Similarly,whileclone18 andYS110recognizethesimilarepitopeonCD26 (Figure4,Additionalfile1:FigureS2andAdditionalfile 1:FigureS3),onlyclone18canstainCD26clearlyinfixed tissueswithstrongintensity,suggestingthatslightdifferencesintherecognizedepitopescandeterminewhether mAbbindingtoitsdenaturedantigencanoccur. Cross-blockingstudiesshowedthat,incontrastto clone5or18,thebindingofclone1,11,16or19to CD26washardlyaffectedbythehumanizedanti-CD26 mAbYS110(Figure4,Additionalfile1:FigureS2and Additionalfile1:FigureS3),suggestingthatthese4 novelanti-CD26mAbsaresuitableforanalyzingCD26 expressioninclinicalsamplesfollowingYS110therapy. PotentialusesofthesenovelmAbsintheclinicalsetting maybetodetectCD26expressioninformalin-fixedtissues,oroncirculatingcellsinbloodsamples,orsCD26 levelsinbodilyfluidsthroughsuchmethodsasimmunohistochemistry,flowcytometry,orELISA.Furthermore, thesenovelmAbsarepotentiallyusefulforanalyzing CD26expressioninfixedtissuesoronthesurfaceof lymphocytesortumorsduringorfollowingtheadministrationofhumanizedanti-CD26mAbinanimaldisease modelsthatinvolveinoculatedhumanlymphocytesor tumors[12,16],andareexpectedtocontributetofuture CD26-relatedresearcheffort. Sinceweintendtoutilizethesenovelanti-CD26mAbsas companiondiagnosticagentsintheclinicalsetting,our currenteffortisfocusedonimprovingimmunohistochemicalstainingmethodsbyexaminingsuchissuesastheconditionofantigenretrievalorblocking,ortheoptimal concentrationoftheprimar yantibody(anti-CD26mAb) thatcanmaximizestainingintensitywhileloweringbackgroundstaining.Furthermore,wealsoidentifiedtheamino acidsequenceofthevariableregioninboththeheavychain andlightchainofclone19(datanotshown),andwillaim torefinetheabilityofthismAbtobindtoCD26through geneticengineeringtechniques. Inconclusion,wehavesucceededindevelopingnovel anti-humanCD26mAbssuitableforimmunohistochemicalstainingofCD26informalin-fixedtissuesections withreliableclarityandintensity.Furthermore,since someofthesemAbsexhibitnocross-reactivitywiththe therapeutichumanizedanti-CD26mAb,theyarepotentiallyusefulascompaniondiagnosticagentsinthe clinicalsettingwhileadvancingfutureCD26-related research.AdditionalfileAdditionalfile1:FigureS1. Representativeresultsofimmunostaining withnovelanti-CD26mAbs.Thetissuespecimensofliver,kidney, prostateortwocasesofmalignantmesotheliomawerestainedwithHatano etal.DiagnosticPathology 2014, 9 :30 Page11of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 12

thehybridomasupernatant(clone1,5,11or16),counterstainedwith hematoxylin(originalmagnification,200X). FigureS2. Analysisof crossreactivityofnovelanti-CD26mAbswithhumanizedanti-CD26 mAb.Jurkat-CD26WTcellswerepretreatedwithunlabeledhumanized anti-CD26mAb(YS110)(bluelines)orhumancontrolIgG(redlines),and thentreatedwiththehybridomasupernatant(clone1,5,11,16,18or19) orpurifiedmouseanti-CD26mAb(1F7or5F8),andsubsequentlystained withPE-labeledanti-mouseIgpAb,orstainedwithAlexaFluor647labeledYS110.Datawereanalyzedbyflowcytometry,andthe representativehistogramsareshown.Thegrayareasineachhistogram showthedataofisotypecontrol. FigureS3. Blockingexperimentof novelanti-CD26mAbbindingtoCD26.Jurkat-CD26WTcellswere pretreatedwiththehybridomasupernatant(clone1,5,11,16,18or19) (bluelines)ormouseIgG1isotypecontrol(Contl.IgG)(redlines),and subsequentlystainedwithAlexaFluor647-labeledanti-CD26mAbsor PE-labeledanti-mouseIgpAb,andanalyzedbyflowcytometry.The representativehistogramsareshown,andthegrayareasineach histogramshowthedataofisotypecontrol.Datashownarerepeated twicewithsimilarresults. FigureS4. StainingforCD26expressionon COS-7cellstransfectedwithCD26deletionmutantsbynovelanti-CD26 mAbs.cDNAofdeletedCD26wascotransfectedwithGFP-expressing plasmidtoCOS-7cells.After24h,thetransfectedcellswerestainedwith AlexaFluor647-labeledanti-CD26mAbsorisotypecontrol,andanalyzed byflowcytometry.TherepresentativehistogramsofAlexaFluor647were obtainedbygatingforGFPpositivecellsamongallacquiredcells,and thegrayareasineachhistogramshowthedataofisotypecontrol. Competinginterests Theauthorsdeclarenocompetingfinancialinterestsassociatedwiththis manuscript. Authors ’ contributions RHandCMdesignedandcoordinatedthestudy.RH,TY,SMandEK conductedtheexperiments.CM,TYandKOsupervisedpartofthe experiments.Allauthorscontributedtotheinterpretationsandconclusions presented.RHandCMwrotethemanuscript,andNHD,SIandHY participatedineditingit.Allauthorsreadandapprovedthefinalmanuscript. Acknowledgements TheauthorsthankMs.HirokoMadokoroforexcellentassistancewith immunohistochemicalstainingofCD26,andalsothankMs.HarunaOtsuka andMs.AyaMiwafortheassistancewithpreparationofsolubleCD26and theexperimentofepitopemapping.ThisworkwassupportedbyGrant-in-Aid ofTheMinistryofEducation,Science,Sports(KOandCM)andCulture,Ministry ofHealth,Labour,andWelfare,Japan(CM). Authordetails1DepartmentofTherapyDevelopmentandInnovationforImmuneDisorders andCancers,GraduateSchoolofMedicine,JuntendoUniversity,2-1-1, Hongo,Bunkyo-ku,Tokyo113-8421,Japan.2DepartmentofPathology,Keio UniversitySchoolofMedicine,35Shinanomachi,Shinjuku-ku,Tokyo 160-8582,Japan.3DepartmentofPathology&Oncology,JuntendoUniversity SchoolofMedicine,2-1-1,Hongo,Bunkyo-ku,Tokyo113-8421,Japan.4DivisionofHematology/Oncology,UniversityofFlorida,1600SWArcher Road-Box100278,RoomMSBM410A,Gainesville,FL32610,USA. Received:3October2013Accepted:24January2014 Published:6February2014 References1.FoxDA,HusseyRE,FitzgeraldKA,AcutoO,PooleC,PalleyL,DaleyJF, SchlossmanSF,ReinherzEL: Ta1,anovel105KDhumanTcellactivation antigendefinedbyamonoclonalantibody. JImmunol 1984, 133: 1250 – 6. 2.NanusDM,EngelsteinD,GastlGA,GluckL,VidalMJ,MorrisonM,FinstadCL, BanderNH,AlbinoAP: Molecularcloningofthehumankidney differentiationantigengp160:humanaminopeptidaseA. ProcNatlAcad SciUSA 1993, 90: 7069 – 73. 3.TanakaT,CameriniD,SeedB,TorimotoY,DangNH,KameokaJ,Dahlberg HN,SchlossmanSF,MorimotoC: Cloningandfunctionalexpressionofthe TcellactivationantigenCD26. JImmunol 1992, 149: 481 – 6. 4.MorimotoC,SchlossmanSF: ThestructureandfunctionofCD26inthe T-cellimmuneresponse. ImmunolRev 1998, 161: 55 – 70. 5.OhnumaK,DangNH,MorimotoC: Revisitinganoldacquaintance:CD26 anditsmolecularmechanismsinTcellfunction. TrendsImmunol 2008, 29: 295 – 301. 6.DeMeesterI,KoromS,VanDammeJ,ScharpeS: CD26,letitcutorcutit down. ImmunolToday 1999, 20: 367 – 75. 7.vonBoninA,HuhnJ,FleischerB: Dipeptidyl-peptidaseIV/CD26onTcells: analysisofanalternativeT-cellactivationpathway. ImmunolRev 1998, 161: 43 – 53. 8.MorimotoC,TorimotoY,LevinsonG,RuddCE,SchrieberM,DangNH, LetvinNL,SchlossmanSF: 1F7,anovelcellsurfacemolecule,involvedin helperfunctionofCD4cells. JImmunol 1989, 143: 3430 – 9. 9.MasuyamaJ,YoshioT,SuzukiK,KitagawaS,IwamotoM,KamimuraT, HirataD,TakedaA,KanoS,MinotaS: Characterizationofthe4C8antigen involvedintransendothelialmigrationofCD26(hi)Tcellsaftertight adhesiontohumanumbilicalveinendothelialcellmonolayers. JExpMed 1999, 189: 979 – 90. 10.OhnumaK,InoueH,UchiyamaM,YamochiT,HosonoO,DangNH, MorimotoC: T-cellactivationviaCD26andcaveolin-1inrheumatoid synovium. ModRheumatol 2006, 16: 3 – 13. 11.HatanoR,OhnumaK,YamamotoJ,DangNH,MorimotoC: CD26-mediated co-stimulationinhumanCD8(+)Tcellsprovokeseffectorfunctionviapro-inflammatorycytokineproduction. Immunology 2013, 138: 165 – 72. 12.HatanoR,OhnumaK,YamamotoJ,DangNH,YamadaT,MorimotoC: Preventionofacutegraft-versus-hostdiseasebyhumanizedanti-CD26 monoclonalantibody. BrJHaematol 2013, 162: 263 – 77. 13.HavrePA,AbeM,UrasakiY,OhnumaK,MorimotoC,DangNH: Theroleof CD26/dipeptidylpeptidaseIVincancer. FrontBiosci 2008, 13: 1634 – 45. 14.HoL,AytacU,StephensLC,OhnumaK,MillsGB,McKeeKS,NeumannC, LaPushinR,CabanillasF,AbbruzzeseJL, etal : Invitroandinvivo antitumoreffectoftheanti-CD26monoclonalantibody1F7onhuman CD30+anaplasticlargecellT-celllymphomaKarpas299. ClinCancerRes 2001, 7: 2031 – 40. 15.InamotoT,YamochiT,OhnumaK,IwataS,KinaS,InamotoS,TachibanaM, KatsuokaY,DangNH,MorimotoC: Anti-CD26monoclonalantibody-mediated G1-SarrestofhumanrenalclearcellcarcinomaCaki-2isassociatedwith retinoblastomasubstratedephosphorylation,cyclin-dependentkinase2 reduction,p27(kip1)enhancement, anddisruptionofbindingtothe extracellularmatrix. ClinCancerRes 2006, 12: 3470 – 7. 16.InamotoT,YamadaT,OhnumaK,KinaS,TakahashiN,YamochiT,InamotoS, KatsuokaY,HosonoO,TanakaH, etal : Humanizedanti-CD26monoclonal antibodyasatreatmentformalignantmesotheliomatumors. ClinCancer Res 2007, 13: 4191 – 200. 17.AoeK,AmatyaVJ,FujimotoN,OhnumaK,HosonoO,HirakiA,FujiiM, YamadaT,DangNH,TakeshimaY,InaiK,KishimotoT,MorimotoC: CD26 overexpressionisassociatedwithprolongedsurvivalandenhanced chemosensitivityinmalignantpleuralmesothelioma. ClinCancerRes 2012, 18: 1447 – 56. 18.DongRP,TachibanaK,HegenM,ScharpeS,ChoD,SchlossmanSF, MorimotoC: Correlationoftheepitopesdefinedbyanti-CD26mAbsand CD26function. MolImmunol 1998, 35: 13 – 21. 19.YamadaK,HayashiM,MadokoroH,NishidaH,DuW,OhnumaK,Sakamoto M,MorimotoC,YamadaT: NuclearlocalizationofCD26inducedbya humanizedmonoclonalantibodyinhibitstumorcellgrowthby modulatingofPOLR2Atranscription. PLoSOne 2013, 8: e62304. 20.TakebeY,SeikiM,FujisawaJ,HoyP,YokotaK,AraiK,YoshidaM,AraiN: SR alphapromoter:anefficientandversatilemammaliancDNAexpression systemcomposedofthesimianvirus40earlypromoterandtheR-U5 segmentofhumanT-cellleukemiavirustype1longterminalrepeat. MolCellBiol 1988, 8: 466– 72. 21.TanakaJ,MiwaY,MiyoshiK,UenoA,InoueH: ConstructionofEpstein-Barr virus-basedexpressionvectorcontainingmini-oriP. BiochemBiophysRes Commun 1999, 264: 938 – 43. 22.TanakaT,Duke-CohanJS,KameokaJ,YaronA,LeeI,SchlossmanSF,Morimoto C: Enhancementofantigen-inducedT-cellproliferationbysolubleCD26/ dipeptidylpeptidaseIV. ProcNatlAcadSciUSA 1994, 91: 3082 – 6. 23.IkushimaH,MunakataY,IshiiT,IwataS,TerashimaM,TanakaH,Schlossman SF,MorimotoC: InternalizationofCD26bymannose6-phosphate/insulin-likegrowthfactorIIreceptorcontributestoTcellactivation. ProcNatl AcadSciUSA 2000, 97: 8439 – 44.Hatano etal.DiagnosticPathology 2014, 9 :30 Page12of13 http://www.diagnosticpathology.org/content/9/1/30

PAGE 13

24.TanakaT,KameokaJ,YaronA,SchlossmanSF,MorimotoC: The costimulatoryactivityoftheCD26antigenrequiresdipeptidylpeptidase IVenzymaticactivity. ProcNatlAcadSciUSA 1993, 90: 4586 – 90. 25.SalehHA,JinB,BarnwellJ,AlzohailiO: Utilityofimmunohistochemical markersindifferentiatingbenignfrommalignantfollicular-derived thyroidnodules. DiagnPathol 2010, 5: 9. 26.deMatosLL,DelGiglioAB,MatsubayashiCO,deLimaFarahM,DelGiglio A,daSilvaPinhalMA: ExpressionofCK-19,galectin-3andHBME-1inthe differentiationofthyroidlesions:systematicreviewanddiagnostic meta-analysis. DiagnPathol 2012, 7: 97. 27.HuaX,YuL,HuangX,LiaoZ,XianQ: Expressionandroleoffibroblast activationprotein-alphainmicroinvasivebreastcarcinoma. DiagnPathol 2011, 6: 111. 28.LuZ,QiL,BoXJ,LiuGD,WangJM,LiG: ExpressionofCD26andCXCR4in prostatecarcinomaanditsrelationshipwithclinicalparameters. JRes MedSci 2013, 18: 647 – 52. 29.YamaguchiU,NakayamaR,HondaK,IchikawaH,HasegawaT,ShitashigeM, OnoM,ShojiA,SakumaT,KuwabaraH: Distinctgeneexpression-defined classesofgastrointestinalstromaltumor. JClinOncol 2008, 26: 4100 – 8. 30.TorigoeT,AsanumaH,NakazawaE,TamuraY,HirohashiY,YamamotoE, KanasekiT,HasegawaT,SatoN: Establishmentofamonoclonalanti-pan HLAclassIantibodysuitableforimmunostainingofformalin-fixedtissue: unusuallyhighfrequencyofdown-regulationinbreastcancertissues. PatholInt 2012, 62: 303 – 8.doi:10.1186/1746-1596-9-30 Citethisarticleas: Hatano etal. : Establishmentofmon oclonalanti-human CD26antibodiessuitableforimmunost ainingofformalin -fixedtissue. DiagnosticPathology 2014 9 :30. Submit your next manuscript to BioMed Central and take full advantage of: € Convenient online submission € Thorough peer review € No space constraints or color “gure charges € Immediate publication on acceptance € Inclusion in PubMed, CAS, Scopus and Google Scholar € Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Hatano etal.DiagnosticPathology 2014, 9 :30 Page13of13 http://www.diagnosticpathology.org/content/9/1/30



PAGE 1

Fig. S1 Anti CD26 Ab : Liver Kidney Prostate Mesothelioma case1 case2 clone 1 clone 5 clone 11 clone 16

PAGE 2

Contl. IgG pretreat YS110 pretreat Isotype control clone 1 clone 5 clone 11 clone 16 clone 18 clone 19 1F7 5F8 YS110 CD26 Fig. S2

PAGE 3

Fig. S3 Anti Ms Ig 4G8 YS110 5F8 16D4B Staining : AlexaFluor647 PE Blocking : Contl. IgG clone 1 clone 5 clone 11 clone 16 clone 19 clone 18 9C11

PAGE 4

Anti Human CD26 mAb Isotype control Fig. S4 YS110 clone 18 clone 19 5F8 CD26(1 766th AAs) CD26(1 739th AAs) CD26(1 577th AAs) CD26(1 449th AAs) CD26(1 358th AAs) CD26(1 247th AAs) Mock 4G8 1F7 AlexaFluor647