Immunoprofiles and DNA Methylation of Inflammatory Marker Genes in Ulcerative Colitis-Associated Colorectal Tumorigenesis
Full text
This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Immunoprofiles and DNA Methylation of Inflammatory Marker Genes in Ulcerative Colitis-Associated Colorectal Tumorigenesis © 2021 by the authors. Licensee MDPI, Basel, Switzerland. Published version Mäki-Nevala, Satu; Ukwattage, Sanjeevi; Wirta, Erkki-Ville; Ahtiainen, Maarit; Ristimäki, Ari; Seppälä, Toni T.; Lepistö, Anna; Mecklin, Jukka-Pekka; Peltomäki, Päivi Mäki-Nevala, S., Ukwattage, S., Wirta, E.-V., Ahtiainen, M., Ristimäki, A., Seppälä, T. T., Lepistö, A., Mecklin, J.-P., & Peltomäki, P. (2021). Immunoprofiles and DNA Methylation of Inflammatory Marker Genes in Ulcerative Colitis-Associated Colorectal Tumorigenesis. Biomolecules, 11(10), Article 1440. https://doi.org/10.3390/biom11101440 2021
Biomolecules2021,11,1440.https://doi.org/10.3390/biom11101440www.mdpi.com/journal/biomolecules Article ImmunoprofilesandDNAMethylationofInflammatory MarkerGenesinUlcerativeColitis‐Associated ColorectalTumorigenesis SatuMäki‐Nevala 1, *,SanjeeviUkwattage 1 ,Erkki‐VilleWirta 2 ,MaaritAhtiainen 3 ,AriRistimäki 4,5 ,ToniT. Seppälä 5,6 ,AnnaLepistö 5,6 ,Jukka‐PekkaMecklin 3,7 andPäiviPeltomäki 1 1 DepartmentofMedicalandClinicalGenetics,UniversityofHelsinki,FI‐00014Helsinki,Finland; [email protected](S.U.);[email protected](P.P.) 2 DepartmentofGastroenterologyandAlimentaryTractSurgery,TampereUniversityHospital, FI‐33521Tampere,Finland;erkki‐[email protected] 3 DepartmentofEducationandResearch,HospitalNovaofCentralFinland,FI‐40620Jyväskylä,Finland; [email protected](M.A.);jukka‐[email protected](J.‐P.M.) 4 DepartmentofPathology,HUSLAB,HUSDiagnosticCenter,UniversityofHelsinki andHelsinkiUniversityHospital,FI‐00029Helsinki,Finland;[email protected] 5 AppliedTumorGenomicsResearchProgram,ResearchProgramsUnit,UniversityofHelsinki, FI‐00014Helsinki,Finland;[email protected](T.T.S.);[email protected](A.L.) 6 DepartmentofGastrointestinalSurgery,HelsinkiUniversityHospital,FI‐00029Helsinki,Finland 7 DepartmentofSportandHealthSciences,UniversityofJyväskylä,FI‐40014Jyväskylä,Finland *Correspondence:satu.maki‐[email protected] Abstract:Immunologicalandepigeneticchangesareinterconnectedandcontributeto tumorigenesis.Wedeterminedtheimmunoprofilesandpromotermethylationofinflammationrelatedgenesforcolitis‐associatedcolorectalcarcinomas(CA‐CRC).Theresultswerecompared withLynchsyndrome(LS)‐associatedcolorectaltumors,whicharecharacterizedbyanactive immuneenvironmentthroughinheritedmismatchrepairdefects.CA‐CRCs(n=31)were immunohistochemicallyevaluatedforimmunecellscores(ICSs)andPDCD1andCD274 expression.Seveninflammation‐associatedgenes(CD274,NTSR1,PPARG,PTGS2,PYCARD, SOCS1,andSOCS2),therepairgeneMGMT,andeightstandardmarkergenesfortheCpGIsland MethylatorPhenotype(CIMP)wereinvestigatedforpromotermethylationinCA‐CRCs,LStumors (n=29),andpairednormalmucosaebymultiplexligation‐dependentprobeamplification.Allbut oneCA‐CRCsweremicrosatellite‐stableandallLStumorsweremicrosatellite‐unstable.MostCA‐ CRCshadahighICS(55%)andapositiveCD274expressioninimmunecells(52%).NTSR1revealed frequenttumor‐specifichypermethylationinCA‐CRCandLS.WhencomparedtoLSmucosae, normalmucosaefrompatientswithCA‐CRCshowedsignificantlyhighermethylationofNTSR1 andmostCIMPmarkers.Inconclusion,CA‐CRCsshareafrequentICS high /CD274 pos expression patternwithLStumors.Elevatedmethylationinnormalmucosamayindicatefieldcancerization asafeatureofCA‐CRC‐associatedtumorigenesis. Keywords:Lynchsyndrome;ulcerativecolitis;coloncancer;immunecellscore;DNAmethylation; inflammation‐associatedgenes 1.Introduction Ulcerativecolitis(UC)togetherwithCrohn’sdiseasecompriseinflammatorybowel disease(IBD),whichisarecognizedriskfactorforcolorectalcancer(CRC)[1].The developmentofUC‐associatedCRC(CA‐CRC)isamultifactorialandcomplexprocess involvingpersistentinflammation,alterationsintheintestinalmicrobiome,imbalanceof theimmunesystem,andgeneticandepigeneticchanges[2,3].Reactiveoxygenspecies Citation:Mäki‐Nevala,S.; Ukwattage,S.;Wirta,E.‐V.; Ahtiainen,M.;Ristimäki,A.; Seppälä,T.T.;Lepistö,A.;Mecklin, J .‐P.;Peltomäki,P.Immunoprofiles andDNAMethylationof InflammatoryMarkerGenesin UlcerativeColitis‐associated ColorectalTumorigenesis. Biomolecules2021,11,1440. https://doi.org/10.3390/biom11101440 AcademicEditor:YuyanHan Received:19August2021 Accepted:27September2021 Published:30September2021 Publisher’sNote:MDPIstays neutralwithregardtojurisdictional claimsinpublishedmapsand institutionalaffiliations. Copyright:©2021bytheauthors. LicenseeMDPI,Basel,Switzerland. Thisarticleisanopenaccessarticle distributedunderthetermsand conditionsoftheCreativeCommons Attribution(CCBY)license (http://creativecommons.org/licenses /by/4.0/).
Biomolecules2021,11,14402of18 andnitrogenintermediatesproducedbyactiveinflammatorycellspromotemutations andgeneticinstability[4],andcytokineproductionmaypromoteepigeneticchanges contributingtotumorigenesisthroughDNArepairgeneinactivationandother mechanisms[4].CRCriskinpatientswithUCisapproximatelytwo‐foldcomparedwith theaveragepopulation[3].Moreover,CA‐CRCpatientstendtobeyoungerandhave multipletumors,andmucinousandpoorlydifferentiatedcarcinomasarecommon[3]. Consistentwiththeaggressivehistopathologicalcharacteristics,Watanabeetal.found CA‐CRCtobeassociatedwithaworseprognosiscomparedwithsporadiccolorectal cancer,whichmightinpartreflectdiagnosisatlatestages[5]. Aggressivehistologicalfeatures,i.e.,mucinousandpoorlydifferentiatedtumors, alsocharacterizeLynchsyndrome(LS)‐associatedCRC[6],wherethecumulativeriskof CRCdependsonthepredisposingmutation:carriersofMLH1orMSH2germline mutationshaveover10‐foldrelativeriskscomparedtothegeneralpopulation[7].LSis themostcommonhereditaryconditionpredisposingtoCRCandiscausedbypathogenic germlinevariantsinDNAmismatchrepair(MMR)genesMLH1,MSH2,MSH6,andPMS2 [8]or,morerarely,deletionsinthe3′ endoftheEPCAMgene,leadingto hypermethylationoftheMSH2genepromoter[9].Theseheterozygousgermlinedefects decreasethelevelsoffunctionalMMRproteins,compromisingtheircriticalcancer avoidancefunctionsandresultinginhypermutatedandmicrosatellite‐unstabletumors [10,11].LS‐CRCpatientshavebettersurvivalcomparedtosporadicCRC[7,12]and tumorsarelesslikelytometastasize,possiblyduetoanactiveimmuneenvironmentof thetumors[6].Ahighrateofdenovosomaticmutationsleadstohighlevelsof neoantigens,causingabundanttumor‐infiltratinglymphocytes,whichare counterbalancedbytheincreasedexpressionofimmunecheckpointmolecules,suchas PDCD1andCD274[13]. ImmunoscoreisamethodfordescribingimmuneresponsesandisbasedonCD3+ andCD8+lymphocyteinfiltrationsinthetumorcenterandinvasivemargin.Ithasproven tobeastrongindependentprognosticmarkerinstageI–IIIcoloncancer[14].Immunecell score(ICS)wasdevelopedfollowingthosesameprinciplesandproducedsimilarresults. HighICSwasgenerallyassociatedwithafavorableoutcomeanddifferentiatedpatients withapoorvs.improvedprognosisacrosstumorstages,regardlessofMMRstatus[15,16]. Asexpected,mostMMR‐deficientCRCs,whethersporadicorLS,werecharacterizedby highICS,aprofuseamountofPDCD1‐positivelymphocytes,andahighexpressionrate ofCD274‐positiveimmunecells[15].Availableinformationofimmunecellinfiltrationin CA‐CRCislimited,andthefindingsareinpartconflicting[17–19]. Astheseobservationsimply,inflammationhasanimportantroleincancer;however, epigeneticregulationofinflammation‐associatedgenesanditsimpactonUC‐andLS‐ associatedCRCtumorigenesisarenotcompletelyunderstood.Promotermethylationof inflammation‐associatedgenes,suchasCD274,maybeaprognosticfactorinsporadic CRC[20].DNAmethylationalterationsofinflammation‐associatedgenescouldprovide aspecificmechanismforthewidespreadinflammatory/immunologicalalterationsthat accompanytumorigenesis,especiallyinassociationwiththeCpGislandmethylator phenotype(CIMP)ormicrosatelliteinstability(MSI).Inthisstudy,weaimedtoevaluate theimmunologicallandscapeofCA‐CRCsthroughICSandalteredmethylationof inflammation‐associatedgenes,usingLStumorsforcomparison(Figure1). 2.MaterialsandMethods 2.1.Material StudymaterialincludedFFPEspecimensof(paired)normalandtumorouscolonic mucosafrompatientswithCA‐CRC(normalmucosae,n=17andcarcinomas,n=31)and LS(normalmucosae,n=11,adenomas,n=14,andcarcinomas,n=15).Detailsare presentedinTable1.Samplingwasdonebeforeanycytostaticcancertreatments,either atthetimeofthediagnosisorthesurgery.AllLSadenomashadhigh‐gradedysplasiaand
Biomolecules2021,11,14403of18 theirmolecularprofiles(includingMSI,CIMP,andsomaticmutationalstatus)were comparabletoLS‐CRCs,asdetailedpreviously[21].Therefore,LSadenomasand carcinomaswerecombined,andtherespectiveresultsinthisstudyaregivenforthisjoint groupreferredtoas“LStumors”.AllLSpatientswereverifiedmutationcarriers representedinthenationwideLynchsyndromeRegistryofFinland.DNAwasextracted usingthemodifiedextractionprotocolofthephenyl–chloroformmethod[22].CA‐CRC tissuesampleswerepunchedandDNAwasextractedfromthesepunches(3×1mm), collectedfromcanceroustissueorhistologicallynormaltissueverifiedbythepathologist. Normalsampleswerederivedfromseparatetissueblocks(thewholespecimen representednormaltissuebyhistologicalevaluation,withoneexception,whichwas punchedfromatumorblock(adjacenttotumoroustissue)).ForLStissuesamples,the pathologistverifiedthehistologyandDNAwasextractedfromtheFFPEsections.Normal sampleswerederivedfromseparatetissueblocks(thewholespecimenwashistologically evaluatedasnormaltissue).Intumorsamples,theaverageoftumoroustissuewas46% (SD±13.0). Table1.Characteristicsofthepatientsamples. CharacteristicsCA‐CRCLSTumors Combined1 LSAdenomas (High‐grade Dysplasia) LSCarcinomas pValueCA‐CRC vs.LSTumors Combined1 No.ofpatients31241412NA Malesex220(64%)9(38%)5(40%)5(42%)0.060 Age,years(mean±SD)251.4(±10.7)51.2(±14.1)49.2(±15.2)53.6(±12.3)0.948 YearsofcolitisbeforeCRC diagnosis(mean±SD)223.8(±10.8)3‐ ‐ ‐ NA Genemutatedingermline2 MLH1‐ 18(75%)12(86%)8(67%)NA MSH2‐ 3(13%)1(7%)2(17%)NA MSH6‐ 3(13%)1(7%)2(17%)NA No.oftumors31291415NA MSItumors1(3%)29(100%)414(100%)415(100%)4<0.00001 CIMP(+)tumors12(39%)7(24%)2(14%)5(33%)0.274 Tumorlocation Proximal513(42%)17(59%)6(43%)11(73%)0.431 Distal15(48%)12(41%)8(57%)4(27%) NA3(10%)000 Stageofcarcinomas I13(42%)10(67%)‐ 10(67%)0.190 II6(19%)3(20%)‐ 3(20%) III9(29%)1(7%)‐ 1(7%) IV3(10%)0‐ 0 NA01(7%)‐ 1(7%) 1:LS‐associatedadenomasandcarcinomascombined.2:Calculatedperpatients.3:Informationforeightcasesnotavailable. 4:MSIinformationforoneadenomaandthreecarcinomacasesnotavailable,butimmunohistochemicalanalysissuggested MMRdefectinthesamples.5:Fromcaecumtosplenicflexure(included).NoteregardingLSsamples:Multiplesamples wereavailablefromthreeLSpatients(threecarcinomasandtwocarcinomasfromonepatienteach,andacarcinomaplus adenomafromtwopatientseach).Ifsamplingtookplaceatdifferenttimepoints(metachronousneoplasias),different ageswereincludedinthecalculationofageatdiagnosis.Abbreviations:CA‐CRC,colitis‐associatedcolorectalcancer; CIMP,CpGislandmethylatorphenotype;CRC,colorectalcancer;LS,Lynchsyndrome;MSI,microsatelliteunstable;NA, notavailableorapplicable;SD,standarddeviation.
Biomolecules2021,11,14404of18 2.2.MSIAnalysis MSIstatuswasinvestigatedusingmononucleotiderepeatmarkersBAT25and BAT26,asdescribedinMäki‐Nevalaetal.[21].ThediagnosisofMSIrequiredatleastone unstablemarker. 2.3.ImmunohistochemicalAnalysisofPDCD1andCD274,andImmuneCellScoring(ICS) CA‐CRCsampleswerestudiedforPDCD1andCD274expressionbyimmunohistochemicalanalysis,asdescribedinAhtiainenetal.[15].Also,ICSbasedonCD3andCD8 expressionwascalculatedaccordingtothepreviouslydescribedprotocol[15,16].Briefly, antibodiesusedwere:anti‐PDCD1(SP269,1:50;SpringBioscience,Pleasanton,CA,USA), anti‐CD274(E1L3N,1:100;CellSignalingTechnology,Danvers,MA,USA),anti‐CD3(LN 10,1:50;Novocastra,BuffaloGrove,IL,USA),andanti‐CD8(SP16,1:100;ThermoFisher Scientific,Waltham,MA,USA).PositivelystainedcellswereanalyzedusingQuPath[23]. Cut‐offvaluesforPDCD1,CD3,andCD8positivityweredeterminedfromalargereferenceseriesofCRCsusingreceiveroperatingcharacteristiccurvesdrawninrelationto disease‐specific5‐yearsurvival[18].Thecut‐offvaluesforPDCD1were38cells/mm2(invasivemargin)and57cells/mm2(tumorcenter).Basedonthesecut‐offvalues,thesamples weredividedintoPDCD1lowandPDCD1high.Cut‐offvaluesforICSwere:815forCD3and 384forCD8inthetumorcenterand1144forCD3and496forCD8intheinvasivefront. ExpressionsinthetumorcenterandinvasivefrontwereusedtoformulatetheICS,as specifiedinWirtaetal.[16];onepointisgivenforeachsection(tumorcenterorinvasive marginwithCD3orCD8staining)withalymphocytecountexceedingthedetermined cut‐offvaluesothatinICS4,allthesectionshadahighcellcountandinICS0,allthe sectionshadalowcellcount.SamplesweregroupedintoICSlow(scores0–2)andICShigh (scores3–4).ExpressionofCD274wasevaluatedontumorcells(TC)andtumor‐infiltrat‐ ingimmunecells(IC)throughoutthetumorcenterandtheinvasivemarginfromstained wholetissuesections.Boththepercentageofstainedimmuneandtumorcellsandthe stainingintensitywerevisuallyestimated.AsamplewasconsideredCD274posiftheproportionofCD274‐positivetumorand/ortumor‐infiltratingimmunecellswithmoderate orstrongintensityexceeded5%.ThelocationofCD274expressionontumorand/orimmunecellswasconfirmedbydetailedside‐by‐sideanalysesofCD274expressionwith correspondingHEstainedslides. 2.4.CpGIslandMethylatorPhenotype(CIMP)Analysis CommercialSALSAMS‐MLPAME042‐B2(forLSsamples)andME042‐C1(forCACRCsamples)probemixes(MRCHolland,Amsterdam,TheNetherlands)wereusedaccordingtothemanufacturer’sprotocol.TheresultswereinterpretedasdescribedinMäki‐ Nevalaetal.[21]forCA‐CRCandValoetal.[24]forLS.Briefly,eightestablishedmarker genesforCIMP(CACNA1G,CDKN2A,CRABP1,IGF2,NEUROG1,MLH1,RUNX3,and SOCS1)wereinvestigated.AsetofcorrespondingnormalsampleswereexaminedtodetermineathresholdforhypermethylationforeachprobewithastringencylevelII;these detailsaredescribedinValoetal.[24].CIMPstatuswasassignedbyathree‐levelevaluation[25].First,probe‐levelmethylationwasassessedagainstthemeanDmofnormalmucosaeplustwostandarddeviations(ifthisvaluewaslowerthanthetechnicalthreshold of0.15,thetechnicalcut‐offwasusedasathreshold).Second,agenewasconsideredhypermethylatedifatleastone‐fourthofallprobesforthatgeneweremethylated.Third, CIMP(+)statusrequiredaminimumof3/5markergenes(CACNA1G,IGF2,NEUROG1, RUNX3,andSOCS1)tobemethylatedaccordingtotheWeisenbergercriteria[26]. 2.5.O‐6‐methylguanine‐DNA‐methyltransferase(MGMT)MethylationAnalysis TheSALSAMLPAProbemixME012MGMT‐IDH1‐IDH2(MRCHolland,Amsterdam,TheNetherlands)wasappliedaccordingtothemanufacturer’sprotocol.TheprobemixincludedsixprobesfortheMGMTpromoterarea.
Biomolecules2021,11,14405of18 2.6.BisulfiteModificationandSequencing CRCcelllines(HCA7,HCT116,LIM1215,RKO,SW480,andT84)andnormalsamples(DNAfromnormalcolonicmucosa(Dr.P.Set,Amsbio,Abingdon,UnitedKingdom) andblood‐derivedhealthycontrolDNA(Promega,Madison,WI,USA)werebisulfiteconvertedusingtheEZDNAMethylation‐DirectTMKit(ZymoResearch,Orange,CA,USA) accordingtothemanufacturer’sprotocol.Methylation‐unbiasedprimers(Supplementary FigureS1andSupplementaryTableS6)wereusedtoamplifythebisulfiteconvertedDNA samples,andamplifiedproductsweresequenceddirectlyoraftercloning.NTSR1was selectedforthecloningexperiment(SupplementaryFigureS2).AmplifiedDNAsamples wereclonedintoapCRC2.1TOPOvectorusingtheTOPOTACloningSystem(Invitrogen,Carlsbad,CA,USA)andOneShotElectrocompetentE.coli(Invitrogen,Carlsbad, CA,USA).ResultingwhitecolonieswereusedforthepurificationofDNAandweresequenced,totaling14–30sequencesperamplificationproduct.ThefrequencyofmethylatedsiteswascalculatedandcomparedtoMS‐MLPAresults(SupplementaryFigureS2). 2.7.Custom‐madeMethylationSpecificMultiplexLigation‐DependentProbeAmplification (MS‐MLPA)AssayforInflammation‐RelatedGenes Theinflammation‐associatedgenepanelwasdesignedin‐house(Figure1).We startedwithalistofmorethan200inflammationand/orimmuneresponse‐associated genesinmice(Paloviita,2015,personalcommunicationandMScthesis,2016).AnextensiveliteraturereviewwasconductedinthePubMedusingthefollowingsearchterms: “gene_name”AND“methylation”AND“cancer”.inOctoberandNovember2016.Humanorthologsofthoseinflammation‐relatedgeneswhoseDNAmethylationalterations wereassociatedwithCRCandinverselyassociatedwithmRNAand/orproteinexpressionwerecharacterized.Foragenetobeconsideredfurther,itwasnecessarythatatleast oneGCGCsiteoccurredintheregionofinteresttoallowthesubsequentdesignofanMSMLPAprobe.Weselectedthemostinterestinggenes(n=10);additionally,theimmune checkpointgenesCD274andPDCD1wereincluded,althoughtothatdate,therewereno studiesreportingmethylationalterationsassociatedwithCRC.Forthese12genes,bisulfitesequencingprimersweredesignedorpreviouslypublishedoneswereusedforthe regions(SupplementaryTableS6).CpGsitesandpromoterregionswereidentifiedusing theUCSCGenomeBrowser[27].Bisulfitesequencingwasperformedasdescribedabove. MS‐MLPAprobesweredesignedifdistinctmethylationprofileswereobservedbetween variousCRCcelllinesandinnormalcontrolscomparedtoCRCcelllines. Figure1.Flowchartofthisinvestigation. Thefinalsevengeneswere:CD274(ClusterofDifferentiation274),alsoknownasthe geneforProgrammedCellDeathLigand‐1(PD‐L1);NTSR1(NeurotensinReceptor1);
Biomolecules2021,11,14406of18 PPARG(PeroxisomeProliferator‐ActivatedReceptorGamma);PTGS2(Prostaglandin‐En‐ doperoxideSynthase2),alsoknownascyclo‐oxygenase2(COX‐2);PYCARD(PYDand CARDDomainContainingProtein),SOCS1(SuppressorofCytokineSignaling1),and SOCS2(SuppressorofCytokineSignaling2).MS‐MLPAprobesweredesignedaccording totheprotocolbytheMRC‐Holland(Amsterdam,TheNetherlands)(version15)(SupplementaryFigureS1andSupplementaryTableS1).ProbeswereorderedfromtheIntegrated DNATechnologies(Coralville,IA,USA),andweretestedandoptimizedagainstbisulfite sequencingresults.Theoptimizedcustom‐madeprobemixincludingprobesfortheaforementionedsevengeneswasusedtogetherwiththeSALSAMLPAprobemixP300‐B1Reference2(MRC‐Holland,Amsterdam,TheNetherlands)andtheSALSAMLPAEK‐FAM reagentkit(MRC‐Holland,Amsterdam,TheNetherlands)accordingtothemanufacturer’sprotocol.Cut‐offvaluesforhypermethylationwerecalculatedinthesamewayas describedfortheCIMPpanel[24].Tonote,SOCS1isalsoincludedinthecommercial CIMPpanel(seeabove),butitstargetedregionisdifferent;CIMPpanelprobesarelocated 1374‐2441bpdownstreamfromthecustom‐madeprobe. 2.8.mRNAExpressionAnalysis Cancercelllines’mRNAexpressionwasassessedwithAffymetrixHumanGenome U133Plus2.0GeneChip®microarrays(Affymetrix,SantaClara,CA,USA),asdescribed [28].NormaltissueRNAwaspurchasedfromAmsbio(Abingdon,UK).Microarraydata wereanalyzedbytheAffypackageinRusingRMA(robustmulti‐arrayaverage)normalizationmethods. 2.9.StatisticalAnalysis StatisticalanalyseswereconductedusingtheSPSSsoftware,version25.0and27.0 (IBMSPSSInc.,Chicago,IL,USA).Fisher’sexacttestwasusedtostudypairwisecomparisonsofcategoricalvariables.Normaldistributionofcontinuousdatawastestedusingthe Shapiro–Wilktest.Continuousvariableswereanalyzedusingtheindependentt‐testfor normallydistributedvariablesandthenon‐parametricMann–Whitney–Utestwasappliedonnotnormallydistributedvariables.Similarly,correlationanalyseswerecalculatedeitherwithPearson’sorSpearman’scorrelationtest.Exacttwo‐sidedpvalueswere calculated.Whenapplicable,rawpvalueswerecorrectedwiththeBonferronimethodfor multiplecomparisons.Correctedpvalues<0.05wereconsideredstatisticallysignificant andarepresentedthroughoutthepaper,unlessotherwiseindicated. 3.Results 3.1.ImmunoprofilesofCA‐CRC ImmunoprofilesweredeterminedbasedonICSandimmunecheckpointproteinexpression.Thisstudycomprised31CA‐CRCs,allmicrosatellite‐stable(MSS)exceptforone (Table1).ICSsweredistributedasfollows:0in23%,1in10%,2in13%,3in16%,and4in 38%ofthetumors.Onadichotomousscale,ICSwashigh(threeorfour)in17/31(55%) CA‐CRC‐tumors(Table2).Figure2representshighandlowexpressionsofCD3,CD8, PDCD1,andCD274byIHC. MostCA‐CRCsshowedahighexpressionoftheimmunecheckpointproteinPDCD1: 18/31(58%)whenPDCD1‐positivelymphocytesinthetumorcenterwerecounted(Table 2)and21/31(68%)basedonPDCD1‐positivelymphocytesintheinvasivemargin.Halfof theCA‐CRCs(16/31,52%)expressedCD274‐positiveimmunecells,whereastumorcells wereCD274‐negative. Wecombinedthedataoftumor‐infiltratinglymphocytes,i.e.,ICSs,andtheCD274 expressionofimmunecells,assuggestedanddoneinpreviousstudies[15,29].Thesamplesweredividedintofoursubgroups:ICShigh/CD274pos(typeI;adaptiveimmuneresistance),ICSlow/CD274neg(typeII;immunologicalignorance),ICShigh/CD274neg(typeIII;
Biomolecules2021,11,14407of18 tolerance),andICSlow/CD274pos(typeIV;intrinsicinduction)accordingtoTengetal.[29]. DetailsarepresentedinTable2. Figure2.IHCstainingofimmunoproteins.Examplesof(a)CD3high,(b)CD3low,(c)CD8high, (d)CD8low,(e)PDCD1high,(f)PDCD1low,(g)CD274(IC)highand(h)CD274(IC)lowexpressions.Blackbarsintherightbottomcornersindicatethesizeof500μm. Inthisinvestigation,ICS,PDCD1,andCD274assayswereperformedonCA‐CRC alone.Wepreviously[15]immunologicallyprofiledaseparatesetofLS‐CRCs(allfrom verifiedcarriersofpathogenicgermlinemutationsofMLH1,MSH2,orMSH6),together withcohortsofsporadicCRCs,bythesamemethodusedhere,enablingdirectcomparisons(Table2).CA‐CRCdidnotsignificantlydifferfromLS‐CRCrelativetoanyimmunologicalparameterassessed.As97%ofCA‐CRCswereMSS,weselectedtheMMR‐profi‐ cient(pMMR)subgroupofsporadicCRCsforcomparison.InCA‐CRCs,CD274onimmunecellswasmoreoftenpositiveandICShigh(thedifferencereachedstatisticalsignificanceforCD274),resultinginasignificantlydifferentdistributionofICS/CD274ICtypes betweenCA‐CRCandpMMR‐CRC(Table2). Table2.PDCD1,CD274,andICSsinCA‐CRC(thisstudy),comparedwithsporadicpMMR‐CRC andLS‐CRCfromourpreviousinvestigation. Ahtiainenetal.[15]pValue Immunoprofilecharacteristics CA‐CRCpMMR‐ CRCLS‐CRCCA‐CRCvs. (n=31)(n=100)(n=48)pMMR‐ CRCLS‐CRC PDCD11 Low 13(42%)5117(35%)nsns a b c d e f g h
Biomolecules2021,11,14408of18 High18(58%)4931(65%) CD274ontumorcells Negative31(100%)9945(94%)nsns Positive013(6%) CD274onimmunecells Negative15(48%)7018(37%)0.033ns Positive16(52%)3030(63%) ICS Low(0–2)14(45%)26013(27%)nsns High(3–4)17(55%)34035(73%) ICS/CD274IC(immunological group)4 ICShigh/CD274pos(typeI)11(35%)1621(44%)0.023ns ICSlow/CD274neg(typeII)9(29%)529(19%) ICShigh/CD274neg(typeIII)6(19%)2514(29%) ICSlow/CD274pos(typeIV)5(16%)74(8%) 1:PDCD1classificationwasbasedonPDCD1‐positivelymphocytesintumorcenter(cut‐off57 cells/mm2forCA‐CRCand55cells/mm2fortheremaininggroups),2:ICSwas0for7(23%),1for3 (10%),and2for4(13%)CA‐CRCs.3:ICSwas3for5(16%)and4for12(38%)CA‐CRCs.4:CD274 expressionintumor‐infiltratingimmunecells.Abbreviations:IC,immunecell;pMMR,MMR‐ proficient;ns,non‐significant. 3.2.Immunoprofilevs.SomaticMutationalStatusandCIMP Ofthe31CA‐CRCs,27weresequencedforsomaticmutationsaspartofourprevious study[21],whichallowedustocomparemutationalstatusandimmunoprofiles.CA‐CRC tumorsbrokedownintothreesubgroupsbasedontheirstatusofMSIandhypermutability(thelatterdefinedasover10nonsynonymoussomaticmutations/Mb):grouponecomprisedhypermutatedMSItumors(n=1);grouptwo,hypermutatedMSStumors(n=9); andgroupthree,non‐hypermutatedMSStumors(n=17).WeplottedthenumberofsomaticmutationsagainstICS,andallICSgroups,exceptnumbertwo,includedtwoor threehypermutatedtumors(Figure3).Interestingly,twotumors(includingthesingleone withMSI)hadsomaticmutationnumbersaroundthe“ultramutated”range(100mutations/Mb)andbothhadlowICSvalues(0and1).Themutationrate(mutations/Mb)was notassociatedwiththeimmunologicalsubgroupdefinedbyICSandCD274expression (p=0.835).Thetenhypermutanttumorsweredistributedamongallfourimmunological subgroups,mostcommonlyICShigh/CD274pos(fourtumors)andICSlow/CD274neg(threetumors,includingthesingleMSItumor),followedbyICShigh/CD274neg(twotumors),and ICSlow/CD274pos(onetumor). OfallCA‐CRCtumors,12(39%)wereCIMP(+)and19wereCIMP(‐)(61%)(Table1). CIMPstatuswasnotassociatedwiththeimmunologicalsubgroup(p=0.486).Thelargest proportionofCIMP(+)tumors(6/12,50%)fellintotheICShigh/CD274posgroupandof CIMP(–)tumors(6/19,32%),intotheICSlow/CD274neggroup.
Biomolecules2021,11,144015of18 MLH1methylationanddeficientMMR.Ontheotherhand,studiesexistinwhichnodifferenceinMGMTmethylationbetweenUC‐relatedandnon‐inflammatorycolontumorigenesispathwayswasobserved[48,49].Inourinvestigation,MGMTmethylationdidnot significantlydifferbetweennormalmucosaandtumortissues,arguingagainstamajor roleofMGMTinUC‐relatedcolontumorigenesis.Possibledilutionwithnormalcellswas unlikelytoexplainthelackofhypermethylationintumorcellssinceweobservednocorrelationwhentumorcellpercentagewasplottedagainstmethylationdegreesofMGMT probes.AfewCA‐CRCpatientsshowedveryhighmethylationlevelsinnormalmucosa (Figure6a),possiblyservingasfielddefectsinthosepatients.Interestingly,thecase markedwithasolidarrowinFigure6ahadasomaticKRASG>Asubstitution(Gly12Asp mutation,ref.[21])intumortissue,supportingthehypothesisofSvrceketal.[47]. OurCA‐CRCandLSsampleswereformalin‐fixed,paraffin‐embedded(FFPE),and thechosenapproachesweretailoredtoarchivalsamples.Limitationsofthisstudyinclude relativelysmallsamplesizesandarelianceontargetedmethylationdata.Additionally, thecorrelationbetweenmethylationandexpressioncouldbeaddressedbycelllinestudiesaloneinsteadofpatientspecimens.Furthermore,manyofourCA‐CRCpatientshad receivedimmunosuppressivemedication,andstratificationbytreatmenthistorywasnot possible.OurfindingsneedtobeconfirmedinlargercohortsofCA‐CRCpatients,preferablytakingtreatmenthistoriesintoaccount. Inconclusion,ourstudyrevealed,first,ICShigh/CD274posasthemostfrequentimmunologicalsubcategoryinCA‐CRCs,therebyresemblingLS‐CRCs,andsecond,frequent epigeneticfielddefectsinnon‐neoplasticmucosaofCA‐CRCpatients.Fromtheclinical pointofview,theimmunologicalsubtypemayberelevantforimmunotherapy[35].EpigeneticfielddefectsmayprovideearlybiomarkersofcarcinogenesisandidentifyUCpatientswhomightbenefitfrommoreintensivesurveillance[50]. SupplementaryMaterials:Thefollowingareavailableonlineatwww.mdpi.com/2218‐ 273X/11/10/1440/s1,FigureS1:IllustrationofCpGislandsandinvestigatedgenomicregionsofinflammation‐associatedgenesincludedintheMS‐MLPApanel,FigureS2:NTSR1cloningexperimentcomparedtoMS‐MLPA,FigureS3:Correlationbetweenexpressionandmethylationofinflammation‐associatedgenesincelllines,TableS1:MS‐MLPAprobes,TableS2:Cut‐offmethylation dosageratiosforhypermethylation,TableS3:Averagemethylationdosageratiosandtheirstandard deviationsofeachCIMPmarkers,TableS4:AveragemethylationdosageratiosandstandarddeviationsofeachMGMTprobe,TableS5:Methylationdosageratiosofinflammation‐associatedgenes incancercelllines,TableS6:Bisulfitesequencingprimers. AuthorContributions:Conceptualization,S.M.‐N.andP.P.;methodology,S.M.‐N.andP.P.;formal analysis,S.M.‐N.,S.U.andE.‐V.W.;investigation,S.M.‐N.,S.U.andM.A.;resources,J.‐P.M.,T.T.S., A.L.andA.R.;datacuration,S.M.‐N.;writing—originaldraftpreparation,S.M.‐N.;writing—review andediting,P.P.,E.‐V.W.,J.‐P.M.,T.T.S.;visualization,S.M.‐N.;supervision,P.P.;projectadministration,P.P.andS.M.‐N.;fundingacquisition,P.P.J.‐P.M.,T.T.S.,S.M.‐N.andA.R.Allauthors havereadandagreedtothepublishedversionofthemanuscript. Funding:ThisstudywasfundedbyJaneandAatosErkkoFoundation(toP.P.,J.‐P.M.andT.T.S.); theAcademyofFinland(grantnumbers330606toP.P.and331284toS.M.‐N.);theFinnishCancer Foundation(toP.P.,J.‐P.M.,T.T.S.andA.R.);FinnishMedicalFoundation(toT.T.S.);EmilAaltonen Foundation(toT.T.S.);theSigridJuseliusFoundation(toP.P.,T.T.S.andA.R.)andtheHiLIFEFellows2017–2020(toP.P.). InstitutionalReviewBoardStatement:TheInstitutionalReviewBoardsoftheHelsinkiUniversity CentralHospital(466/E6/01,2October2001andamendment17December2008),JyväskyläCentral Hospital(Dnro10U/2011,3May2011)andtheCentralFinlandHealthCareDistrict(K‐Sshp Dnro4/2011)approvedthisresearch.TheNationalAuthorityforMedicolegalAffairs(Dnro 1272/04/044/07)andtheNationalSupervisoryAuthorityforWelfareandHealth(Valvira/Dnro 10741/06.01.03.01/2015,14January2016)approvedthecollectionofarchivalspecimens.Thestudy wasconductedaccordingtotheguidelinesoftheDeclarationofHelsinki. InformedConsentStatement:Inacquisitionofsamplesandpatientinformation,theguidelinesof ourethicsapprovalswerefollowed,includinginformedconsentprocedureswhenapplicable.
Biomolecules2021,11,144016of18 DataAvailabilityStatement:ThemRNAexpressionprofilingdataispubliclyavailableattheGEO (accessionnumber:GSE58058).Theoriginalmethylationdatasetsusedand/oranalyzedduringthe currentstudyareavailablefromthecorrespondingauthoronreasonablerequest. Acknowledgments:WethankSatuValoforcoordinatingUCsamplecollectionandsharingher expertiseinMS‐MLPApaneldesign,SailaSaarinenfortechnicalsupport,PauliinaPaloviitafor Master’sthesisdataonmurinegenesassociatedwithinflammationandimmuneresponses,and NooraPorkkaforassistanceinpreparingdisplayitemsforthispaper.WealsothanktheCorefacilitySequencingUnitatFIMMTechnologyCentresupportedbyUniversityofHelsinkiandBiocenter Finlandforthesequencingfacilities. ConflictsofInterest:Theauthorsdeclarenoconflictofinterest.Thefundershadnoroleinthe designofthestudy;inthecollection,analyses,orinterpretationofdata;inthewritingofthemanuscript,orinthedecisiontopublishtheresults. References 1. Terzić,J.;Grivennikov,S.;Karin,E.;Karin,M.Inflammationandcoloncancer.Gastroenterology2010,138,2101–2114.e5, doi:10.1053/j.gastro.2010.01.058. 2. AlBakir,I.;Curtius,K.;Graham,T.A.FromColitistoCancer:AnEvolutionaryTrajectoryThatMergesMathsandBiology. Front.Immunol.2018,9,2368,doi:10.3389/fimmu.2018.02368. 3. Rogler,G.Chroniculcerativecolitisandcolorectalcancer.CancerLett.2014,345,235–241,doi:10.1016/j.canlet.2013.07.032. 4. Hartnett,L.;Egan,L.J.Inflammation,DNAmethylationandcolitis‐associatedcancer.Carcinogenesis2012,33,723–731, doi:10.1093/carcin/bgs006. 5. Watanabe,T.;Konishi,T.;Kishimoto,J.;Kotake,K.;Muto,T.;Sugihara,K.;theJapaneseSocietyforCanceroftheColonand Rectum.Ulcerativecolitis‐associatedcolorectalcancershowsapoorersurvivalthansporadiccolorectalcancer:Anationwide Japanesestudy.Inflamm.BowelDis.2011,17,802–808,doi:10.1002/ibd.21365. 6. Drescher,K.M.;Sharma,P.;Lynch,H.T.Currenthypothesesonhowmicrosatelliteinstabilityleadstoenhancedsurvivalof LynchSyndromepatients.Clin.Dev.Immunol.2010,2010,170432,doi:10.1155/2010/170432. 7. Møller,P.;Seppala,T.T.;Bernstein,I.;Holinski‐Feder,E.;Sala,P.;GarethEvans,D.;Lindblom,A.;Macrae,F.;Blanco,I.;Sijmons, R.H.;etal.Cancerriskandsurvivalinpath_MMRcarriersbygeneandgenderupto75yearsofage:AreportfromtheProspectiveLynchSyndromeDatabase.Gut2018,67,1306–1316,doi:10.1136/gutjnl‐2017‐314057. 8. Thompson,B.A.;Spurdle,A.B.;Plazzer,J.P.;Greenblatt,M.S.;Akagi,K.;Al‐Mulla,F.;Bapat,B.;Bernstein,I.;Capella,G.;den Dunnen,J.T.;etal.Applicationofa5‐tieredschemeforstandardizedclassificationof2,360uniquemismatchrepairgenevariantsintheInSiGHTlocus‐specificdatabase.Nat.Genet.2014,46,107–115,doi:10.1038/ng.2854. 9. Ligtenberg,M.J.;Kuiper,R.P.;Chan,T.L.;Goossens,M.;Hebeda,K.M.;Voorendt,M.;Lee,T.Y.;Bodmer,D.;Hoenselaar,E.; Hendriks‐Cornelissen,S.J.;etal.HeritablesomaticmethylationandinactivationofMSH2infamilieswithLynchsyndromedue todeletionofthe3′exonsofTACSTD1.Nat.Genet.2009,41,112–117,doi:10.1038/ng.283. 10. Cejka,P.;Stojic,L.;Mojas,N.;Russell,A.M.;Heinimann,K.;Cannavo,E.;diPietro,M.;Marra,G.;Jiricny,J.Methylation‐induced G(2)/MarrestrequiresafullcomplementofthemismatchrepairproteinhMLH1.EMBOJ.2003,22,2245–2254,doi:10.1093/emboj/cdg216. 11. Wang,L.;Tsutsumi,S.;Kawaguchi,T.;Nagasaki,K.;Tatsuno,K.;Yamamoto,S.;Sang,F.;Sonoda,K.;Sugawara,M.;Saiura,A.; etal.Whole‐exomesequencingofhumanpancreaticcancersandcharacterizationofgenomicinstabilitycausedbyMLH1haploinsufficiencyandcompletedeficiency.GenomeRes.2012,22,208–219,doi:10.1101/gr.123109.111. 12. Dominguez‐Valentin,M.;Seppala,T.T.;Sampson,J.R.;Macrae,F.;Winship,I.;Evans,D.G.;Scott,R.J.;Burn,J.;Moslein,G.; Bernstein,I.;etal.SurvivalbycoloncancerstageandscreeningintervalinLynchsyndrome:AprospectiveLynchsyndrome databasereport.Hered.CancerClin.Pract.2019,17,28,doi:10.1186/s13053‐019‐0127‐3. 13. Biller,L.H.;Syngal,S.;Yurgelun,M.B.RecentadvancesinLynchsyndrome.Fam.Cancer2019,18,211–219,doi:10.1007/s10689‐ 018‐00117‐1. 14. Pages,F.;Mlecnik,B.;Marliot,F.;Bindea,G.;Ou,F.S.;Bifulco,C.;Lugli,A.;Zlobec,I.;Rau,T.T.;Berger,M.D.;etal.International validationoftheconsensusImmunoscorefortheclassificationofcoloncancer:Aprognosticandaccuracystudy.Lancet2018, 391,2128–2139,doi:10.1016/S0140‐6736(18)30789‐X. 15. Ahtiainen,M.;Wirta,E.V.;Kuopio,T.;Seppala,T.;Rantala,J.;Mecklin,J.P.;Bohm,J.CombinedprognosticvalueofCD274(PD‐ L1)/PDCDI(PD‐1)expressionandimmunecellinfiltrationincolorectalcanceraspermismatchrepairstatus.Mod.Pathol.2019, 32,866–883,doi:10.1038/s41379‐019‐0219‐7. 16. Wirta,E.V.;Seppala,T.;Friman,M.;Vayrynen,J.;Ahtiainen,M.;Kautiainen,H.;Kuopio,T.;Kellokumpu,I.;Mecklin,J.P.; Bohm,J.Immunoscoreinmismatchrepair‐proficientand‐deficientcoloncancer.J.Pathol.Clin.Res.2017,3,203–213, doi:10.1002/cjp2.71. 17. Michael‐Robinson,J.M.;Pandeya,N.;Walsh,M.D.;Biemer‐Huttmann,A.E.;Eri,R.D.;Buttenshaw,R.L.;Lincoln,D.;Clouston, A.D.;Jass,J.R.;Radford‐Smith,G.L.Characterizationoftumour‐infiltratinglymphocytesandapoptosisincolitis‐associated neoplasia:Comparisonwithsporadiccolorectalcancer.J.Pathol.2006,208,381–387,doi:10.1002/path.1895.
Biomolecules2021,11,144017of18 18. Rajamaki,K.;Taira,A.;Katainen,R.;Valimaki,N.;Kuosmanen,A.;Plaketti,R.M.;Seppala,T.T.;Ahtiainen,M.;Wirta,E.V.; Vartiainen,E.;etal.Geneticandepigeneticcharacteristicsofinflammatoryboweldiseaseassociatedcolorectalcancer.Gastroenterology2021,doi:10.1053/j.gastro.2021.04.042. 19. Soh,J.S.;Jo,S.I.;Lee,H.;Do,E.J.;Hwang,S.W.;Park,S.H.;Ye,B.D.;Byeon,J.S.;Yang,S.K.;Kim,J.H.;etal.Immunoprofilingof Colitis‐associatedandSporadicColorectalCanceranditsClinicalSignificance.Sci.Rep.2019,9,6833,doi:10.1038/s41598‐019‐ 42986‐1. 20. Goltz,D.;Gevensleben,H.;Dietrich,J.;Dietrich,D.PD‐L1(CD274)promotermethylationpredictssurvivalincolorectalcancer patients.Oncoimmunology2017,6,e1257454,doi:10.1080/2162402x.2016.1257454. 21. Maki‐Nevala,S.;Ukwattage,S.;Olkinuora,A.;Almusa,H.;Ahtiainen,M.;Ristimaki,A.;Seppala,T.;Lepisto,A.;Mecklin,J.P.; Peltomaki,P.Somaticmutationprofilesasmolecularclassifiersofulcerativecolitis‐associatedcolorectalcancer.Int.J.Cancer 2021,148,2997–3007,doi:10.1002/ijc.33492. 22. Isola,J.;DeVries,S.;Chu,L.;Ghazvini,S.;Waldman,F.AnalysisofchangesinDNAsequencecopynumberbycomparative genomichybridizationinarchivalparaffin‐embeddedtumorsamples.Am.J.Pathol.1994,145,1301–1308. 23. Bankhead,P.;Loughrey,M.B.;Fernandez,J.A.;Dombrowski,Y.;McArt,D.G.;Dunne,P.D.;McQuaid,S.;Gray,R.T.;Murray, L.J.;Coleman,H.G.;etal.QuPath:Opensourcesoftwarefordigitalpathologyimageanalysis.Sci.Rep.2017,7,16878, doi:10.1038/s41598‐017‐17204‐5. 24. Valo,S.;Kaur,S.;Ristimaki,A.;Renkonen‐Sinisalo,L.;Jarvinen,H.;Mecklin,J.P.;Nystrom,M.;Peltomaki,P.DNAhypermethylationappearsearlyandshowsincreasedfrequencywithdysplasiainLynchsyndrome‐associatedcolorectaladenomasand carcinomas.Clin.Epigenetics2015,7,71,doi:10.1186/s13148‐015‐0102‐4. 25. Berg,M.;Hagland,H.R.;Soreide,K.ComparisonofCpGislandmethylatorphenotype(CIMP)frequencyincoloncancerusing differentprobe‐ andgene‐specificscoringalternativesonrecommendedmulti‐genepanels.PLoSONE2014,9,e86657, doi:10.1371/journal.pone.0086657. 26. Weisenberger,D.J.;Siegmund,K.D.;Campan,M.;Young,J.;Long,T.I.;Faasse,M.A.;Kang,G.H.;Widschwendter,M.;Weener, D.;Buchanan,D.;etal.CpGislandmethylatorphenotypeunderliessporadicmicrosatelliteinstabilityandistightlyassociated withBRAFmutationincolorectalcancer.Nat.Genet.2006,38,787–793,doi:10.1038/ng1834. 27. UCSCGenomeBrowser.Availableonline:http://genome‐euro.ucsc.edu/cgi‐bin/hgGateway(accessedon18May2018). 28. Niskakoski,A.;Kaur,S.;Staff,S.;Renkonen‐Sinisalo,L.;Lassus,H.;Jarvinen,H.J.;Mecklin,J.P.;Butzow,R.;Peltomaki,P.EpigeneticanalysisofsporadicandLynch‐associatedovariancancersrevealshistology‐specificpatternsofDNAmethylation.Epigenetics2014,9,1577–1587,doi:10.4161/15592294.2014.983374. 29. Teng,M.W.;Ngiow,S.F.;Ribas,A.;Smyth,M.J.ClassifyingCancersBasedonT‐cellInfiltrationandPD‐L1.CancerRes.2015,75, 2139–2145,doi:10.1158/0008‐5472.CAN‐15‐0255. 30. Ahlquist,T.;Lind,G.E.;Costa,V.L.;Meling,G.I.;Vatn,M.;Hoff,G.S.;Rognum,T.O.;Skotheim,R.I.;Thiis‐Evensen,E.;Lothe, R.A.Genemethylationprofilesofnormalmucosa,andbenignandmalignantcolorectaltumorsidentifyearlyonsetmarkers. Mol.Cancer2008,7,94,doi:10.1186/1476‐4598‐7‐94. 31. Qian,X.C.;Brent,T.P.Methylationhotspotsinthe5′flankingregiondenotesilencingoftheO6‐methylguanine‐DNAmethyltransferasegene.CancerRes.1997,57,3672–3677. 32. Däbritz,J.;Menheniott,T.R.Linkingimmunity,epigenetics,andcancerininflammatoryboweldisease.Inflamm.BowelDis. 2014,20,1638–1654,doi:10.1097/MIB.0000000000000063. 33. Yasmin,R.;Siraj,S.;Hassan,A.;Khan,A.R.;Abbasi,R.;Ahmad,N.Epigeneticregulationofinflammatorycytokinesandassociatedgenesinhumanmalignancies.Mediat.Inflamm.2015,2015,201703,doi:10.1155/2015/201703. 34. Yan,L.;Meizhi,H.;Yaoyao,Z.;Chen,Y.;Shuyi,W.;Xiaohui,B.;Odong,C.;Lang,X.ThePrognosticandClinicopathological RolesofPD‐L1ExpressioninColorectalCancer:ASystematicReviewandMeta‐Analysis.Front.Pharmacol.2019,28,139, doi:10.3389/fphar.2019.00139. 35. Picard,E.;Verschoor,C.P.;Ma,G.W.;Pawelec,G.RelationshipsbetweenImmuneLandscapes,GeneticSubtypesandResponses toImmunotherapyinColorectalCancer.Front.Immunol.2020,11,369,doi:10.3389/fimmu.2020.00369. 36. McGrail,D.J.;Pilie,P.G.;Rashid,N.U.;Voorwerk,L.;Slagter,M.;Kok,M.;Jonasch,E.;Khasraw,M.;Heimberger,A.B.;Lim,B.; etal.Hightumormutationburdenfailstopredictimmunecheckpointblockaderesponseacrossallcancertypes.Ann.Oncol. 2021,32,661–672,doi:10.1016/j.annonc.2021.02.006. 37. Gevensleben,H.;Holmes,E.E.;Goltz,D.;Dietrich,J.;Sailer,V.;Ellinger,J.;Dietrich,D.;Kristiansen,G.PD‐L1promotermethylationisaprognosticbiomarkerforbiochemicalrecurrence‐freesurvivalinprostatecancerpatientsfollowingradicalprostatectomy.Oncotarget2016,7,79943–79955,doi:10.18632/oncotarget.13161. 38. Franzen,A.;Vogt,T.J.;Muller,T.;Dietrich,J.;Schrock,A.;Golletz,C.;Brossart,P.;Bootz,F.;Landsberg,J.;Kristiansen,G.;etal. PD‐L1(CD274)andPD‐L2(PDCD1LG2)promotermethylationisassociatedwithHPVinfectionandtranscriptionalrepression inheadandnecksquamouscellcarcinomas.Oncotarget2018,9,641–650,doi:10.18632/oncotarget.23080. 39. Marwitz,S.;Scheufele,S.;Perner,S.;Reck,M.;Ammerpohl,O.;Goldmann,T.Epigeneticmodificationsoftheimmune‐check‐ pointgenesCTLA4andPDCD1innon‐smallcelllungcancerresultsinincreasedexpression.Clin.Epigenetics2017,9,51, doi:10.1186/s13148‐017‐0354‐2. 40. Kamimae,S.;Yamamoto,E.;Kai,M.;Niinuma,T.;Yamano,H.O.;Nojima,M.;Yoshikawa,K.;Kimura,T.;Takagi,R.;Harada, E.;etal.EpigeneticsilencingofNTSR1isassociatedwithlateralandnoninvasivegrowthofcolorectaltumors.Oncotarget2015, 6,29975–29990,doi:10.18632/oncotarget.5034.
Biomolecules2021,11,144018of18 41. Wu,Z.;Martinez‐Fong,D.;Tredaniel,J.;Forgez,P.Neurotensinanditshighaffinityreceptor1asapotentialpharmacological targetincancertherapy.Front.Endocrinol.2012,3,184,doi:10.3389/fendo.2012.00184. 42. Ouyang,Q.;Zhou,J.;Yang,W.;Cui,H.;Xu,M.;Yi,L.Oncogenicroleofneurotensinandneurotensinreceptorsinvarious cancers.Clin.Exp.Pharmacol.Physiol.2017,44,841–846,doi:10.1111/1440‐1681.12787. 43. Kakiuchi,N.;Yoshida,K.;Uchino,M.;Kihara,T.;Akaki,K.;Inoue,Y.;Kawada,K.;Nagayama,S.;Yokoyama,A.;Yamamoto, S.;etal.FrequentmutationsthatconvergeontheNFKBIZpathwayinulcerativecolitis.Nature2020,577,260–265, doi:10.1038/s41586‐019‐1856‐1. 44. Hahn,M.A.;Li,A.X.;Wu,X.;Yang,R.;Drew,D.A.;Rosenberg,D.W.;Pfeifer,G.P.Lossofthepolycombmarkfrombivalent promotersleadstoactivationofcancer‐promotinggenesincolorectaltumors.CancerRes.2014,74,3617–3629,doi:10.1158/0008‐ 5472.CAN‐13‐3147. 45. Kang,K.;Bae,J.H.;Han,K.;Kim,E.S.;Kim,T.O.;Yi,J.M.AGenome‐WideMethylationApproachIdentifiesaNewHypermethylatedGenePanelinUlcerativeColitis.Int.J.Mol.Sci.2016,17,1291,doi:10.3390/ijms17081291. 46. Issa,J.P.;Ahuja,N.;Toyota,M.;Bronner,M.P.;Brentnall,T.A.Acceleratedage‐relatedCpGislandmethylationinulcerative colitis.CancerRes.2001,61,3573–3577. 47. Svrcek,M.;Buhard,O.;Colas,C.;Coulet,F.;Dumont,S.;Massaoudi,I.;Lamri,A.;Hamelin,R.;Cosnes,J.;Oliveira,C.;etal. MethylationtoleranceduetoanO6‐methylguanineDNAmethyltransferase(MGMT)fielddefectinthecolonicmucosa:An initiatingstepinthedevelopmentofmismatchrepair‐deficientcolorectalcancers.Gut2010,59,1516–1526, doi:10.1136/gut.2009.194787. 48. Scarpa,M.;Scarpa,M.;Castagliuolo,I.;Erroi,F.;Kotsafti,A.;Basato,S.;Brun,P.;D’Inca,R.;Rugge,M.;Angriman,I.;etal. Aberrantgenemethylationinnon‐neoplasticmucosaasapredictivemarkerofulcerativecolitis‐associatedCRC.Oncotarget 2016,7,10322–10331,doi:10.18632/oncotarget.7188. 49. Konishi,K.;Shen,L.;Wang,S.;Meltzer,S.J.;Harpaz,N.;Issa,J.P.RareCpGislandmethylatorphenotypeinulcerativecolitisassociatedneoplasias.Gastroenterology2007,132,1254–1260,doi:10.1053/j.gastro.2007.01.035. 50. Zeng,Z.;Mukherjee,A.;Zhang,H.FromGeneticstoEpigenetics,RolesofEpigeneticsinInflammatoryBowelDisease.Front. Genet.2019,10,1017,doi:10.3389/fgene.2019.01017.