甲基化DNA測序的試劑配制

DataSupplementTitle:CirculatingmethylatedSEPT9DNAinPlasmaisaBiomarkerforColorectalCancer.TheodeVos1,ReimoTetzner2,FabianModel1,3,GunterWeiss2,MatthiasSchuster2,JurgenDistler2,KathrynV.Steiger1,RobertGrutzmann5,ChristianPilarsky5,JensK.Habermann6,PhillipR.Fleshner7,BentonM.Oubre8,RobertDay1,AndrewZ.Sledziewski1,CatherineLofton-Day11)DetailedmSEPT9AssayProtocol:DNAExtraction:DNAwasextractedfrom5mLofbloodplasmausingamodifiedviralDNA/RNAextractionkit(chemagenAG,BaesweilerGermany).Plasmasampleswerethawedatroomtemperatureandextractedfollowingthedirectionsofthekitwithmodifications.Sampleswerelysedandtreatedwithproteaseat56°Cfor10minina50mLFalcontube.100^Lofmagneticparticlesand15mLofbindingbufferwerethenadded,andbindingwasperformedfor60minatroomtemperatureonarotator.Magneticparticleswerecapturedfor4min,thesupernatantdiscardedandthepelletwasresuspendedin3mLofwashbuffer.1.5mLofparticlesolutionweretransferredtoa2mLSafeLock,thebeadscapturedandthesupernatantdiscarded.Thiswasrepeatedtocompletethe3mLtransfer.Tubeswerebrieflycentrifugedandtheresidualwashbufferwasremovedbypipettingafterbeadseparation.Thetubeswerethenplacedina56°Cdryblockfor5min,100ofelutionbufferwasadded,thetubesincubatedat65°Cwithshakingonathermomixerfor15min,theparticlesseparatedonamagneticstandandtheelutedDNAtransferredtoa0.5mLSafeLocktube(Eppendorf).A5^laliquotoftheDNAsamplewastransferredto45ofelutionbufferforthemeasurementofgenomicDNA.BisulfiteConversion:Thesampleinputforbisulfitetreatmentwas95-100ofextractedDNAinelutionbuffer.Thebisulfitereagents(for25reactions)werepreparedasfollows.Bisulfitesolution:Sodiumbisulfite(4.71gm)andsodiumsulfite(1.13gm)weredissolvedin10mLofddH2Oinafalcontube,byvigorousshakingandheatingto50°Cifrequired,andthepHadjustedto5.4-5.5with0.2MNaOHasnecessary.DME-radicalscavengersolution:188mgof6-hydroxy-2,5,7,8-tetramethyl-chroman-2-carboxylicacidwasdissolvedin1.5mLdiethyleneglycoldimethylether(DME),vortexingtoensureanuniformsolution.BisulfiteReaction:190^Lofbisulfitesolutionand30^lDME-radicalscavengersolutionwereaddedtothe95-100DNAsamplein0.5mLSafeLocktubes.ThetubeswereincubatedinaEppendorfMastercycler(Eppendorf)accordingtothefollowingprotocol:5min99°C,25min50°C,5min99°C,1h25min50°C,5min99°C,4h55min50°C,hold20°C.Thisprotocolallowedovernightbisulfiteconversion.BisulfitePurification:Followingbisulfiteconversion,DNAwaspurifiedusingacustomizedkitfromchemagenAG.Thebisulfitereaction(320^L)wastransferredtoa2mLSafeLocktube,and1^LofpolyA(500ng/^L)and1.5mLofbindingbufferwereadded.10^Lchemagenmagneticparticleswereaddedandthesamplewasmixedbyvortexing.Thesampleswereincubatedatroomtemperatureonathermalmixeratarotationof1000rpmfor60min.Magneticparticleswereseparatedonamagneticstand,theliquiddiscarded,thetubesbrieflycentrifugedandtheresidualliquidremovedfollowingmagneticseparation.TheparticleswerewashedtwicewithwashbufferIIfromthekit,andoncewith70%ethanol.Followingtheethanolwash,thetubeswerecentrifugedagain,andtheresidualliquidremovedfollowingmagneticseparation.Theparticlesweredriedbyplacingopentubesina55°Cheatblock,and55虬elutionbuffer(10mMTrispH7.2)added.Sampleswereincubatedat55°Cfor15minonathermalmixerwithrotationsetto1000rpm,placedonamagneticseparatorandtheeluatecontainingDNAtransferredtoanewtube.A5^Laliquotofbis-DNAwasaddedto45^Lofelutionbufferforthemeasurementofa10folddilutedsample.Thispurificationmethodleavesoutthedesulfonationstepfollowingbisulfiteconversion,tomaketheDNAamenabletocarryoverpreventionbyUNGasetreatmentasdescribepreviously.(1)PCRamplificationofsulfonatedDNArequiresanextendedactivationtimetoallowdesulfonationtooccurpriortoamplification.PCRAnalysis:TheoligonucleotidesequencesandassayinformationareoutlinedintheTablebelow.SupplementalTable1.Oligonucleotidesequences,concentrationsandcyclingconditionsfortherealtimePCRassaysusedinthisstudy.PCRForwardPrimerReversePrimerBlockerProbeCyclingConditionsSEPT9ASSAYCONCAAATAATCCCATCCAACTA0.3MMGATT-DS-GTTGTTTATTAGTTATTATGT0.3MMGTTATTATGTTGGATrnGTGGTTAATGTGTAG-C31.0MMFAM-TTAACCGCGAAATCCGAC-BHQ10.1MMActivation-95°C30min.55Cycles:95OC10sec(4.4oC/sec),56°C30sec(2.2oC/sec)cooling-40°C5sec(2.2°C/sec).CFF1(genomicassay)ASSAYCONCTAAGAGTAATAATGGATGGATGATG0.63MMCCTCCCATCTCCCTTCC0.63MMN/A6FAM-ATGGATGAAGAAAGAAAGGATGAGT-BHQ-10.2mMActivation-95°C15min.50Cycles:95°C10sec(4.4oC/sec),58°C60sec(2.2oC/sec)cooling-40°C5sec(2.2°C/sec).Pactin(bisulfiteassay)ASSAYCONCGTGATGGAGGAGGIIIAGTAAGTT0.9MMCCAATAAAACCTACTCCTCCCTTAA0.9MMN/AFAM-ACCACCACCCAACACACAATAACAAACACA-BHQ1a0.1mMActivation-95°C30min.50Cycles:95°C10sec(4.4oC/sec),57°C30sec(2.2°C/sec),72°C10sec(4.4°C/sec)cooling-40°C5sec(2.2°C/sec).FortheSEPT9PCR,thereverseprimercontainedanabasicd-spacerbaseinthe5thposition.TheblockerwasterminatedwithaC3spaceratthe3’endtopreventextensSfiPT9TherealtimeprobeusedtheFAM/BHQ-1fluorophorsdyes.TheCFF1PCRwasidenticaltopreviousstudies.Theprimersforthep-actinassaywereshortenedcomparedwithpreviousstudies,buttheprobewasidenticaltopreviousstudies(2,3).OligonucleotidequalitywasdeterminedinhousebyMALDI-TOFpriortouse,andthelimitofdetectionforeachPCRwasevaluatedpriortouseinstudies.TheQuantitectMultiplexmastermixfromQiagenwasusedata1xfinalconcentrationinallassays.ThetotalPCRreactionvolumewas25譏using96wellreactionplatesonaRocheLC480realtimethermalcycler.FortheSEPT9and0-actinPCRreactions,theactivationtemperaturewasextendedtoallowthedesulfonationreactiontoproceedpriortoamplification.RealtimePCRanalysis:TotalgenomicDNAwasmeasuredbyrealtimePCRanalysisofthedilutedgenomicaliquotusingtheCFF1reaction.TotalbisulfiteDNAwasmeasuredfroma10ULaliquotofthebisulfiteDNAeluatewiththep-actinreaction.SEPT9methylationwasmeasuredintriplicateusing10uLofthebisDNAeluateperreplicate,andinasinglemeasureusing10uLofthedilutedbisulfiteDNAeluate.PCRAssayPrequalification:Priortothestudy,the90%limitofdetection(LoD)wasdeterminedfortheSEPT9PCRbyanalysisofbisulfitetreatedmethylated(Sss1treated)DNAdilutedinabackgroundofbisulfitetreated50ngperipheralbloodleukocyte(PBL)DNA(RocheAppliedScience).AsillustratedinDataSupplementFigure1,thedilutionserieswasfrom100pgto3.125pgin2foldsteps.Wetested12replicatesforfourdifferentPBLbackgrounds,foratotalof48replicatesateachconcentration.The90%LoDwasdefinedasthelowestconcentrationofspikedmethylatedDNAinabackgroundof50nghumangenomicDNAforwhichthemeasurementvalueshadanareaundertheROCcurve(AUC)of0.9comparedwithmeasurementswithoutspikedmethylatedDNA,andwasestimatedtobe9.4pgforthe^SEPT9assay.Forp-actin,theLoDof9.7pgwasdeterminedwithadilutionseriesofbisulfitetreatedmethylated(Sss1treated)DNA,andforthegenomicCFF1PCRtheLODof4.3pgwasmeasuredwithadilutionseriesofgenomicPBLDNA.2)CharacterizationofAssayperformancewithmodelsamples:Modelsamples:Forworkflowdevelopmenttwotypesofmodelsamplewereproduced:1)purifiedmethylatedDNA(CpGenome,Millipore)wasspikedintoplasmanegativeformethylatedSEPT9;2)plasmapositiveformethylatedSEPT9wasspikedintoplasmanegativefortheSEPT9biomarker.TodeterminetheperformanceofDNAextractionfromplasma,totalgenomicDNArecoverywasmeasuredusingthegenomicrealtimePCRassayCFF1.AnexampleexperimentfromtheassaydevelopmentprocessusingspikedplasmasamplesisshowninFigure2.TherecoveryoftotalgenomicDNAisshownforeightindividualsamples,inwhichwemeasuredaverageDNArecoveryof2.93土1.36ng/mLofinputplasmafromasetofsurrogateplasmasamplesinwhichlowlevelsofplasmacontainingmethylatedSEPT9werespikedintoaSEPT9negativeplasmabackground(Figure2a).Basedontheseandadditionalstudies(notshown)wedemonstratedequivalentrecoveryofgenomicDNAtothatmeasuredwiththeresearchassay.Forbisulfiteconversionandpurification,theprotocolwastestedusingsurrogatesamplesasdescribedforgenomicDNAextractionabove,andasillustratedinthesampleexperimentinFigure2a,therecoveryofbis-DNAmeasuredusingtherealtimep-actinPCRreactionwas1.6土0.63ng/mLofstartingplasma,ayieldintherangeof55%ofthetotalgenomicDNA.TotesttherecoveryofmethylatedtargetDNAinthe^SEPT9assay,methylatedSEPT9positiveplasmawasspikedinadilutionseriesintoabackgroundofmethylatedSEPT9negativeplasma(Figure2b)withthetargetconcentrationofSEPT9biomarkeratlessthat10pg/mLinthe8-folddilutionsamples.Foreachdilution,thePCRpositiveratefortheoriginalresearchassayandthenew^SEPT9assaywasmeasuredin8independentspikedsamples.ThedetectionratesdifferedmarginallybetweenthetwoassaysatthehigherconcentrationsoftheSEPT9biomarker,andwereidentical(50%)atthegreatestdilution(Figure2b).Basedontheseresultsandnumerousadditionalexperiments(datanotshown),forsurrogatesamplesweconsideredthenewassayessentiallyequivalenttothepreviouslydescribedresearchassay,andproceededtovalidatetheassaywithclinicalsamplesinatrainingandteststudy.StudyQualityControlPositivecontrolsforDNAextractionwere25ng/mLCpGenomemethylatedDNAdilutedin5mg/mLbovineserumalbumin(BSA),whilenegativeextractioncontrolswereBSAwithoutspikedDNA.Positivecontrolsforbisulfiteprocessingwerecomposedof10ngfullymethylatedCpGenomeDNAspikedinto90ngofhumangenomicDNApreparedfrombuffycoatcells(RocheAppliedSciences)ina100^Lvolumeofelutionbuffer.NegativebisulfiteconversioncontrolswerecomposedofelutionbufferaloneSupplementalFigure1.TheLoDcalibrationcurveforthemethylatedSEPT9realtimeassayusedinthetrainingandteststudies.TheconcentrationofspikedmethylatedDNAinpicogramsisindicatedacrossthetopofthegraph.ThemeasuredconcentrationofthemethylatedspikeisindicatedontheY-axis.Thedottedbluelineindicatesperfectidentitybetweenexpectedandobserved,theblacklineconnectstheobservedmedianvaluesandthedashedredlinerepresentstheregressionoftheobservedmedianvalues.Thedottedblacklinesshowthe90%confidenceintervalaroundthemedianvalue.SupplementalFigure2.Performanceofthe^SEPT9assayonmodelDNAsamples.SupplementalFigure2a.ConcentrationoftotalgenomicDNA(graybars)basedontheCFF1PCRassay,andbis-DNA(hatchedbars)basedonthep-actinPCRassay,calculatedpermLofinputplasmaforeightindependentsamplepools.SampleswereproducedbyspikingplasmapositiveformethylatedSEPT9intomethylatedSEPT9negativeplasmaSupplementalFigure2b.SEPT9positiverateinpercentagemeasuredfortheSEPT9FRETbasedresearchassay(graybars)andthenew^SEPT9assay(hatchedbars).PlasmasampleswerepreparedinadilutionseriesofmethylatedSEPT9positiveplasmaspikedintoabackgroundofmethylatedSEPT9negativeplasm