powder585qqq.com
585qqq.com 时间:2021-04-08 阅读:()
RESEARCHPAPERAnassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedproteinquantificationbymassspectrometryforeggandmilkallergensChiaraNitride1,2&JrgenNrgaard1&JoneOmar1&HendrikEmons1&María-JoséMartínezEsteso1&GavinO'Connor1,3Received:13November2018/Revised:22March2019/Accepted:28March2019/Publishedonline:28May2019#TheAuthor(s)2019AbstractTheunintentionalpresenceofeventraceamountsofcertainfoodsconstitutesamajorhazardforthosewhosufferfromfoodallergies.
Formanyfoodindustries,productandrawingredientsurveillanceformspartoftheirriskassessmentprocedures.
Thismayrequirethedetectionofmultipleallergensinawidevarietyofmatrices.
Massspectrometryoffersapossiblesolutionforthequantificationofmultipleallergensinasingleanalysis.
ThecapabilityofMStoquantifymanypeptidesfromacomplexproteindigestioniswellknown.
However,alackofmatrixcertifiedreferencematerialshasmadetheoptimisationofextractionanddigestionconditionsformultiplexedallergenquantificationdifficulttoassess.
Here,wereportasystematicstudy,usingprelim-inaryscreeningfollowedbyaDesignofExperimentsapproach,tofindtheoptimalbufferanddigestionconditionsfordetectingmilkandeggproteinmarkersinamodelprocessedfoodmatrix.
Fiveofthemostcommonlyusedbuffers,twochaotropicreagentsandtworeducingreagentswereassessedfortheoptimalextractionofmultipleproteinmarkers.
Whilethechoiceofbackgroundbufferhadlittleimpact,theuseofchaotropicandreducingreagentsshowedsignificantbenefitsfortheextractionofmostproteins.
Afullfactorialdesignexperimentwasappliedtotheparametersshowntohaveasignificantimpactonproteinrecovery.
Thesestudiessuggestthatasingleoptimalsetofextractionconditionsenablingthequantitativerecoveryofallproteinsisnoteasilyachieved.
Therefore,althoughMSiscapableofthesimultaneousquantificationofmanypeptidesinasinglerun,greaterconsiderationofproteinextractionisrequiredbeforetheseareappliedformultiplexallergenquantificationinfoodmatrices.
KeywordsFoodallergens.
Massspectrometry.
Protein.
Extraction.
QuantificationIntroductionFoodallergyiscausedbytheadversereactionoftheimmunesysteminasensitiveindividualtowardaspecificfood.
Currently,thereisnoclinicaltreatmentorcureforthosesuf-feringfromafoodallergy.
Affectedpeoplemustbecarefulaboutwhattheyeatandrelyonappropriatefoodlabellingtoavoidexposuretotheoffendingfood.
EUlegislation,namelyRegulation(EU)No1169/2011[1],requiresfoodmanufac-turerstolistallinformationregardingfoodingredientsontheproductlabelandtohighlightthepresenceofaspecificgroupoffoodingredientsthatmaytriggerfoodallergiesandintol-erances.
Theunintentionalpresenceofanallergenrepresentsafoodsafetyhazardwhichisparticularlydifficulttomanage.
Foodallergensmayenterthefoodsupplychainatanytimeduringstorage,handlingandpackagingasaconsequenceofcross-contactwithinprocessinglines.
Foodbusinessoperatorsshouldadoptriskmanagementprocedurestoprevent,ElectronicsupplementarymaterialTheonlineversionofthisarticle(https://doi.
org/10.
1007/s00216-019-01816-z)containssupplementarymaterial,whichisavailabletoauthorizedusers.
*GavinO'ConnorJRC-Allergens@ec.
europa.
eu1EuropeanCommission,Directorate-GeneralJointResearchCentre,Retieseweg111,2440Geel,Belgium2Presentaddress:SchoolofBiologicalSciences,DivisionofInfection,ImmunityandRespiratoryMedicine,ManchesterAcademicHealthScienceCentre,ManchesterInstituteofBiotechnology,TheUniversityofManchester,131PrincessStreet,Manchester,UK3Presentaddress:Fachbereich3.
2Biochemie,Physikalisch-TechnischeBundesanstalt,Bundesalle100,38116Braunschweig,GermanyAnalyticalandBioanalyticalChemistry(2019)411:3463–3475https://doi.
org/10.
1007/s00216-019-01816-zeliminate,orreducetheriskofcontaminationtoacceptablelevels,thusavoidingtheoccurrenceoffoodallergyincidentsandfoodproductrecalls.
Aquantitativeriskassessmentmeasurestheimpactofaspecifichazard[2]asacombinationofreferencedosesandexposure.
SuchanapproachiscurrentlyusedbytheVITAL(VoluntaryIncidentalTraceAllergenLabelling)riskassess-mentprocedures[3].
Thethresholdsforallergenicproteinsinfoodarebasedonclinicaldata(OralFoodChallenge)andareindicatorsoftheactionlevels,expressedasthetotalpro-teinoftheallergenicfood(mg),belowwhichonlythemostsensitiveallergicsubjectsmayreact.
Precautionaryallergenlabelling(PAL),suchasBmaycontain^statements,shouldonlybeusedafterperformingaquantitativeriskassessment,theresultsofwhichsuggestedademonstrableandsignificantriskofallergencontaminationatlevelsabovetheactionlevel[4].
Analyticalmethodshelptosupportthevalidationandverificationoftheriskassessmentoutcomes,providingquan-titativemeasurementsoftheallergenicingredientinprocessedfood[5].
Anymethodmeasuringallergenmarkersmustex-presstheirresultsinquantitiesandmeasurementunitsthatarecomparablewithothermethods.
Foruseintheriskassessmentprocedures,thisrequirestheallergencontenttobeexpressedinmilligrammesofallergenproteinperkilogrammeoffood[6].
Thefeasibilityofmassspectrometry-basedmethodsforthedetectionandquantificationofallergenshasbeenproven[7–10].
Aconsiderablenumberofstudieshavebeendedicatedtotheoptimisationofmassspectrometry-basedmethodsforthedetectionofeggandmilkproteinsusingpeptidesfromasingleBmarker^proteinfromeachfood[7–15].
Oneofthemajoradvantagesofmassspectrometryisitsabilitytosequen-tiallydetectmarkerpeptidesenablingthemultiplexedanalysisofmanyallergens.
Severalfoodmatrixeshavebeenad-dressed:biscuits[7–10,16];breakfastcereals;infantfood[11];andwine[12–15,17],usingadiverserangeofmassspectrometryplatformsandmethods.
Inparticular,atargetedmassspectrometryapproach,knownasselectivereactionmonitoring(SRM),wassuccessfullyappliedforthedetectionofmulti-allergentargets[7–13].
Thismethodhastheadvan-tageofselectivelydetectingmarkerpeptides,fromspecificallergenproteins,incomplexbackgroundmatricesandallowsthemeasurementofproteinabundancesthatdifferbyfourtofiveordersofmagnitude.
Peptidesarequantifiedbasedonmonitoringspecifiedprecursorpeptide-to-productiontransi-tions.
Forabsolutequantification,themethodofisotopedilu-tionmassspectrometry(IDMS)isoftenapplied[18].
Thisrequiresthemeasurementoftherelativeresponseofthenatu-ralpeptide,generatedbyproteolyticcleavageofthetargetprotein,anditsheavyisotopicallylabelledform,addedataknownquantitytoboththestandardsandthesamples[19].
Thismethodofmeasuringthepeakarearatiosreducessys-tematicerrorsinpeptidequantification[9–12].
However,astheinternalstandard(isotopicallylabelledpeptide)isaddedasthefreepeptide,itdoesnotredressanydeficienciesinproteinextractionorreleaseofthepeptidefromitsprecursorproteinduringdigestion[20].
Therefore,thelargestcontributortothemeasurementuncertaintyand/orbiaswhenlabelledinternalpeptidestandardsareusedisexpectedtoderivefromtheex-tractionanddigestionstepsintheanalysis.
Itisessentialtoensurethattotalextractionandcompletedigestionofthepro-teinsoccurasonlythiswillenableatrueassessmentoftheconcentrationoftheallergenpresent[16].
Tocircumventthisissue,expressionofabovine15N-αS1-caseinprotein,tocorrectforextractionandrecoveryafterbaking,wassuggestedbyNewsomeetal.
[21].
Althoughencouragingrecoveryfactorswerereported,thelabelledpro-teinwasspikedintothebiscuitsafterthebakingprocess;therefore,theunprocessedlabelledproteinmaynotpossessthepossibleheat-inducedmodificationsresultingindifferentextractionanddigestionbehavioursbeingexperiencedbythenaturalandlabelledprotein.
Despitethefactthatmostofthepublishedreports[6–14]selectthesamemarkerproteinsandproteotypicpeptides,frommilkandegg,theextractionanddigestionprotocolsdiffersignificantly,evenwhenthesamefoodmatrixisanalysed.
UnlikeinELISA[22],nosystematicinvestigationconcerningtheoptimisationofextractionanddigestionprotocolsforfoodallergenquantificationhasbeenperformedwhenusingMS.
Theaimofthestudypresentedinthispaperwastheopti-misationoftheextractionanddigestionprotocolsforthequantitativeremovalofmilkandeggproteinsfromamodelprocessedfoodmatrix.
Thequantitativetargetedproteomicpipelinewasseparatedintothreecriticalcomponentsthatcon-tributetothetruenessanduncertaintyofthefinalanswer:extraction,digestionandSRManalysis.
ThefirststepconsistsintheselectionofthemarkerpeptideandthedevelopmentoftheSRMmethod,includingproteotypicpeptidesfromfivemilkallergenicproteinsandtwoeggallergenicproteins.
WeinvestigatedtheimpactofdifferentextractionconditionsontheSRMpeptidesignalresponse,whichisthechemicalentitymeasuredinabottom-upproteomicexperiment.
Thepeptideabundancereflectsthevariationoftheperformanceoftheextractionoftheproteinsfromthefood.
Thefoodmatrixstudiedwasabiscuitincurredwiththeallergenicproteinspriortoheattreatment,therebymimickingcommerciallyavailableproducts.
AnexperimentaldesignorDesignofExperiments(DoE)approachwasusedfortheoptimisationoftheextractionanddigestionconditions.
Themainadvan-tageofDoEisthatthemaximuminformationoftheevaluatedsystemisobtainedbytheminimumnumberofexperiments.
Moreover,itallowsthestudyofmanyvariablesatthesametimeprovidinginformationontheinteractionamongthevar-iables,whichcannotbeachievedusingaonevariableatatimeapproach.
Bythissimultaneousoptimisationofthemultiplepeptideresponses,anoptimalcompromisehastobereachedbetweenthemtohaveonemethodthatsuitsall.
ANOVAon3464NitrideC.
etal.
thefinaloptimisedconditionsallowedtheassessmentoftheimpactofthreeanalyticalstages(extraction,digestionandmeasurement)ontheoverallvariability.
Tothebestofourknowledge,thisisthefirstquantitativestudyshowingtheextenttowhichpeptideresponsecanvarydependingontheextractionanddigestionconditionschosen.
MaterialsandmethodsMaterialsStandardproteinsofαS-casein(C6780),β-casein(C6905),κ-casein(C0406),β-lactoglobulinA(L7880),ovalbumin(A5503),lysozyme(L6876),[Glu1]-fibrinopeptideBhuman(Glu-Fib)(F3261-1MG),myoglobinfromequineheart(M1882),peptidestandards(angiotensinIIhuman(A9525-1MG),Met-Arg-Phe-Alaacetatesalt(M1170-1MG),bradyki-ninacetatesalt(B3259-1MG),angiotensinIhumanacetatesalthydrate(A9650-1MG)werepurchasedfromSIGMA(St.
Louis,MO,USA).
Thereagentsusedforproteinextractionanddigestionin-cludingacetone,hydrochloricacid,sodiumhydroxide,urea,guanidine-HCl,sodiumphosphatebuffer(PBS),ammoniumbicarbonate(AmBic),trizmabase,boricacid,sodiumtetraborate,triethylammoniumhydrogencarbonatebuffer(TEAB),dithiothreitol(DTT),iodoacetamide(IAA)werepur-chasedfromSIGMA(St.
Louis,MO,USA).
Trifluoroaceticacid(TFA)andmercaptoethanol(BME)werepurchasedfromMerck(Darmstadt,Germany).
SolventsincludingULC-gradeacetonitrile(ACN),ULC-gradewaterand99%formicacid(FA)werepurchasedfromBiosolve(Valkenswaard,TheNetherlands).
TrypsinGold-MassSpecGradewaspurchasedfromPromega(Madison,USA).
HyperSepC18(200mg,3mL)CartridgeswereobtainedfromThermoScientific(BiopolymersUlm,Germany).
Thefollowingheavy-isotope-labelledpeptides,wherebyallthecarbonandnitrogenatomsinthen-terminuslysineorargininewerereplacedwith99%13Cand15N,YLGYLEQLLR*(αS1),FALPQYLK*(αS2),LYAEER*(Ova),FESNFNTQATNR*(Lys),werepurchasedfromJPT(JPTPeptideTechnologiesGmbH,Germany).
Twobatchesofbiscuits,oneblankandonecontaining5000mgofproteiningredientperkilogrammeofbiscuit,werepreparedfollowingacommonrecipeforcookiesbytheInstituteofFoodResearchinLeatherhead,UK,intheframeoftheiFAAMproject.
Thebakingtemperaturewas150°Cfor9min.
MilkandeggingredientswerepurchasedfromBIOSERVICEZachGmbH(Austria).
TheSprayDriedskimmedmilkpowder(09G010)contained34.
8%w/wmilkproteinandtheeggwhitepowder(1-022161VOIB)contained71.
6%w/weggprotein.
Thebiscuitrecipealsoincludedhazelnut,peanutandwalnutflours;however,theseproteinswerenottargetedbytheanalysisoutlinedhere.
AdetaileddescriptionoftherecipeisgivenintheElectronicSupplementaryMaterial(ESM;S1).
OperatingconditionsfortheselectivereactionmonitoringbymassspectrometryFiveallergenicproteinsfrommilk(αS1casein—Bosd9,αS2casein—Bosd10,β-casein—Bosd11,κ-casein—Bosd12,β-lactoglobulin—Bodd4)andtwofromeggwhite(ovalbu-min—Gald2,lysozyme—Gald4)wereselectedasmarkersthatwouldindicatethepresenceofthefoodingredients.
Proteinsequencesweredigestedinsilicausingthepredictedcleavagesitesfortrypsin,andtheproteotypicmarkerpeptideswereselectedbasedonspecificcharacteristics.
Theappliedcriteriaforpeptidemarkerselectionincludedthefollowing:theuniquenessofthepeptideforthemarkerprotein,deter-minedusingtheMS-Homologyblasttoolofproteinprospec-tor(http://prospector.
ucsf.
edu/prospector/mshome.
htm),searchingtheSwissprotandtheNCBInrdatabases;alackofpeptidescontainingcysteineandnaturalpost-translationalmodifications(PTMs);theresultingtrypticpeptidescontain-ingbetween5and12aminoacidresidues[23].
AXevoTQ-Striplequadrupole(QqQ)massspectrometer(Waters,Manchester,UK)equippedwithaTrizaicionsourcewasusedtoacquiretheSRMdataforthisstudy.
Standardproteinsweredigestedwithtrypsin(50mmolL1Ambic,pH8;1:50,E:S)andthereleasedpeptidesweredirectlyin-fusedintotheQqQusingasyringepumpat1.
5μLmin1.
Theoptimisedsourceparametersusedwereasfollows:capillaryvoltage3.
5kV;conevoltage25V;sourcetemperature100°C;conegasflow50Lh1;collisiongasflow0.
11mLmin1.
Nitrogen(99.
998%purity,L'AirliquidBelgie,Liège,Belgium)andargon(99.
999%purity,L'AirliquidBelgie,Liège,Belgium)wereusedastheconeandcollisiongases,respectively.
TheQqQwasfirstoperatedinascanningmodetodefinetheoptimumconditionsfortheprecursorionsignals.
Onlychargestatesbetween+2and+4wereconsideredwhenoptimisingthesourceconditionsasthesewereresponsibleforthemostintensepeptidesignalsfromthedigestedmilkpep-tides.
Subsequently,themassspectrometerwasoperatedinproductionscanningmode,wherethetopy-andb-ionswererankedinincreasingorderofintensityandthetop6wereselectedforfurthercollisionenergy(CE)rampingoptimisa-tion.
Sevendifferentcollisioncellvoltages,between8and22Vinincrementsof2V,werestudied.
ThemassspectrometerwascoupledtoananoAcquity-UPLCsystem(Waters,Manchester,UK).
Thechromato-graphicseparationofthepeptideswasperformedonaC18reversed-phaseionKeysystem(BEHC18,130,1.
7μm,150μm*50mm).
Thesolventsusedwere0.
1%FAinwater(solventA)and0.
1%FAinACN(solventB).
ThesampleAnassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3465injectionvolumewas2μLusingafullloopinjection.
Adualpumpreverseflushtrappingconfigurationwasapplied.
Thesamplewasloadedonatrappingcolumn(Symmetry300C18,5μm,300μm*50mm)andanisocraticflowof1%Bover2minwithaflowrateof10μlmin1wasapplied.
Thetrap-pingcolumnwasthenbackflushedinlinewiththeionKeyataflowof2μLmin1utilisingthefollowinggradient:1–5%Bover0.
5min,5–33%Bover15.
5min,33–99%Bover2min,99%Bfor3.
5min,99–1%Bin1minandfinally1%Bfor4.
5mintoallowthecolumntore-equilibrate.
Twotothreetransitionsforeachpeptide,preferablywheretheselectedm/zoftheproductwasgreaterthanthatoftheprecursor,wereincludedinthefinalSRMprotocol.
Thedwelltimewasfixedat50millisecondsforeachmeasuredtransi-tion.
TheSRMtransitionswereseparatedintodifferenttimedeventsresultinginaminimumof12pointsperchromato-graphicpeakforeachtransitionmonitored.
Thedataacquisitionandthedataelaborationwereper-formedusingMassLynxandTargetLynxsoftwarefromWaters.
Thechromatographicpeakswereintegratedautomat-ically,andapeakarealistwasgeneratedforeachpeptidefromeachsample.
Foreachnaturalandisotopicallylabelledpep-tidepair,thepeakarearationatural/labelledwascalculated.
Forpeptideswherenoisotopicallylabelledanalogpeptidewasavailable,auniversalisotopicstandardpeptidewasused.
TheuniversalinternalstandardpeptidechosenwasFALPQYLK*fromαS2-caseinasitelutedinthemiddleofthegradient.
Asystemsuitabilitytest(SST),includinganMS1scanandaproductionscan,wasperformedbeforeeverysetofanalysistoverifythatthemethodperformswithinitsspecifications(retentiontime,m/z).
ProteinquantificationTheRCDC(reducingagentanddetergentcompatible)pro-teinassay(BioRad)wasusedtodeterminethetotalproteinconcentrationinthesampleextracts.
Blanksamples,carriedthroughallstagesofthemanipulation,wereusedasblanksorzeroreferences.
Screening:extractionanddigestionFivebufferingsystems—ammoniumbicarbonate(AMBIC)[9,10],tris-buffer(TRIS)[6–8],phosphatebuffersaline(PBS)[24],triethylammoniumbicarbonatebuffer(TEAB)[12],boratebuffersaline(BBS)[25]—twochaotropicagents[26,27]—ureaandguanidine—andtworeducingagents—dithiothreitol(DTT)andmercaptoethanol(BME)—werescreened.
SeparatebuffersolutionsofAMBIC,TRIS,PBS,TEABandBBSwerepreparedusingMilliQwater(18MΩcm1),at50mmolL1concentrationandpH8.
0.
Thechaotropicagents,ureaandguanidine,wereaddedtoseparatealiquotsofthebuffersolutions,resultinginafinalconcentrationof5molL1ofeach.
ThereducingagentswereaddedtoseparatealiquotsoftheureaAMBICandureaBBSbuffersolutionssothatDTTwas50mmolL1andtheBMEwas2%(v/v).
Agraphicalrepresentationofthescreeningap-proachbuffersolutionsisprovidedinFig.
S1(seeESM).
Atotalof35differentextractioncombinationswereinvestigatedintriplicate.
Onegramofthegroundbiscuitwassolubilisedintheex-tractionbuffer(1:15,w:w),mixedvigorouslyfor1minandsonicatedinacoldbathfor5min.
Theextractionwascarriedoutfor6hat22°C.
Thesolutionwascentrifugedat9500rpmfor45minat22°Candthesupernatantwascollected.
Theextractedproteinswereprecipitatedwithfourvolumesof80%coldacetoneovernightat27°C.
Thesolutionswerecentri-fugedat9500rpmfor30minat4°Candtheacetonewasdiscarded.
Theprecipitatewaswashedtwicewith3mLofcoldacetoneandprecipitatedat27°Cfor1heachtime.
Theacetonewasevaporatedusinganitrogenpurgeatroomtemperature;theprecipitateswereresuspendedin2.
5mLof10mmolL1DTT,3molL1urea,50mmolL1AmBicatpH8.
2.
Twohundredfiftymicrolitresoftheproteinextractweredilutedwith25mmolL1AmBictoyieldafinalcon-centrationof1molL1urea,whichisbelowtheacceptedthresholdfortrypsindigestion.
Thehorsemyoglobin(SIGMAM1882)wasusedasaqualitycontrolofthetrypsindigestionacrosstheextractionbuffersstudied.
Onehundredmicrolitresofstocksolution(14.
2nmol)wereaddedtoeachextractandtheproteinswerereduced(10mmolL1DTT,1hat37°C)andalkylated(50mmolL1IAA,0.
5hat22°Cinthedark).
Trypsinwasaddedatanenzymetosubstrateratioof1:50andthedigestionwascarriedoutfor16hat37°C,whileapplyingaconstantgentleshaking.
Thedigestionwasstoppedbyaddingformicacidtoav/vof0.
1%.
Experimentaldesigns:extractionanddigestionThedevelopedmethodusedtheDesignofExperimentstoorganiseandsuggesttheexperimentsinconjunctionwitharesponsesurfacemethodologyforthefurtheranalysisoftheresultsobtained.
TheUnscramblerXsoftware(v10.
3,CAMO,Trondheim,Norway)wasusedforDoEandfurtherdatatreatment.
Inbothoptimisations,extractionanddigestion,thesameprocedurewasfollowed,wherebyapreliminaryfullfactorialdesign(FFD)wasperformedtoidentifywhichofthestudiedparametersorvariableshadasignificanteffectontheextrac-tionanddigestionprocesses.
TheFFDwasfollowedbyacentralcompositedesign(CCD),inwhichthestatisticallysignificantparametersfromtheFFDwereoptimised.
TheoptimisedvariablesfortheextractionanddigestioncanbeobservedinTables1and2,respectively.
3466NitrideC.
etal.
PreparationofpeptidesforSRManalysesTrifluroaceticacidwasaddedtothedigeststoafinalv/vof0.
1%.
Isotopicallylabelledpeptideswereaddedtothesolu-tionatmassfractionscorrespondingto50mgkg1ofthetotalproteincontentofthetargetallergenicfoodingredient.
Allthedigestsweredilutedtentimestoyieldafinalconcentrationoforganicsolventbelow1%toenableoptimalisolationontheSPEcartridge.
PeptideswereisolatedonHyperSepC18Cartridges(200mg)(ThermoScientific,Biopolymers,Ulm,Germany).
Thecolumnswereactivatedwithmethanol(5col-umnvolumes(CV))andequilibratedwith0.
1%TFA,5%ACNinwater(5CV).
Afterloadingthesample,thecolumnwaswashedwith0.
1%FA,5%ACNinwater(10CV);elutedin0.
1%FA,60%ACNinwater(10CV).
Peptideswerecon-centratedusinganitrogenevaporatoratambienttemperatureandthendilutedwith0.
1%FAinMilliQwateruptoafinalvolumeof1mL.
Sincetheisotopicallylabelledpeptideshaveidenticalchemicalandphysicalbehaviourtotheallergenmarkerpeptides,anypossiblelossofapeptidemarkerduetotheanalyticalprocedurespostdigestionwillbeaccountedfor,astheratiosofnaturaltolabelledpeptidepeakareasweremonitored.
Allsolutionswerespikedwithastandardpeptide,Glu-1-FibrinopeptideB(Glu-Fib),ataconcentrationof50ngmL1beforethemassspectrometryanalysis.
ThiswasusedpurelytoassessthestabilityoftheMSplatformovertheanalysistimescale.
RateofdigestionProteinswereextractedanddigestedusingtheoptimisedpro-tocols.
Thepeptidereleaseratewasevaluated,bymonitoringthenaturaltolabelledpeptideratiosusingtheSRMmethod,overarangeof36h,duringwhich14samplepointswerecollectedat0,0.
5,1,2,3,4,6,8,10,12,22,26,30and34h.
Twoenzyme-to-substrateratioswereconsidered:1:50and1:100,wherethetrypsinwasaddedonlyattimezero,thebeginningoftheincubation.
Themultipleadditionoftheen-zymewasalsostudied.
1:50wasaddedattimezeroandafter22hofdigestion—toafinal1:25(E:S)ratio;1:100wasaddedattimezero,after6,12and22hofdigestion—toafinal1:25(E:S)ratio.
NesteddesignTherepeatabilityofthefinalmethodwasassessedusingatwo-factorfullynestedexperimentdesign.
Threeseparateex-tractsofthe5000mgofthetotalingredientprotein/kggroundbiscuitswereperformedonthesameday.
Eachextractwasdigestedintriplicate.
Fouradditionsoftheenzyme(1:100,E:S)wereperformedevery3hoverthedigestiontimeof12h.
Threereplicatemeasurementswereperformedperdigestedsample.
ResultsanddiscussionThespecificpurposeofthisstudywastoinvestigatetheinflu-enceoftheextractionanddigestionconditionsontheSRMsignalresponseforpeptidesselectedassuitablemarkersforthequantificationofBtotal^eggandmilkprotein.
Thefinalaimwasthedevelopmentofamultiplexedquantitativemeth-odofanalysis.
Theextractionanddigestionprotocolswereoptimisedforbakedbiscuitsandtherepeatabilityoftheex-tractionanddigestionwerestudiedviaatwo-waynestedANOVAstudydesign.
TheSRMmethoddevelopmentconsistedoffourmainsteps:(1)selectionofallergenicmarkerproteinsfromeggandmilk;(2)selectionofproteotypicpeptidesbyinsilicodigestionoftheproteins;(3)methoddevelopmentbyhydro-lysingthestandardproteinswithtrypsin;(4)methodrefine-mentbyanalysinghydrolysedcrudeproteinsextractedfromthebiscuits.
Theproteinmarkersforcow'smilk(Bosdomesticus)wereasfollows:α-S1-casein,α-S2-casein,β-casein,κ-caseinandβ-lactoglobulin.
Theproteinmarkersforhen'seggwhite(Gallusgallus)wereasfollows:ovalbuminandlysozyme.
Table2ExperimentalfactorsandevaluatedlevelsfortheFFDandCCDofthedigestionExperimentalfactorsFFDCCD0+0+++Time(h)258911.
2513.
515.
7518AmBic(mM)25507506.
2512.
518.
7525DTT(mM)25507506.
2512.
518.
7525ACN(%*)0510DMSO(%*)051006.
2512.
518.
7525*FinalvolumeTable1ExperimentalfactorsandevaluatedlevelsfortheFFDandCCDoftheextractionExperimentalfactorsFFDCCD0+0+++Temperature(°C)722370271214pH6.
57.
58.
5Time(h)369AmBic(mM)255075DTT(mM)255075Urea(M)3692.
83.
75.
87.
98.
7Anassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3467Twenty-sevenpeptideswereselectedascandidateproteotypicpeptidesaccordingtoconventionalcriteria[21].
Thespecificityanduniquenessofthepeptideswereestablishedbycomparingthepeptidesequencesagainsttheonlineproteindatabases.
Onlyfourandtwopeptideswerefoundtobeuniqueforcow'smilkandhen'segg,respectively(ESMTableS1).
However,whilemostofthecandidatepep-tideswereuniqueformilkandegg,theywerenotspecies-specific.
Twoofthetargetpeptidesequencesfrommilkpro-teinsandonefromeggproteinswerefoundinothernon-milk-andegg-derivedproteinsbutwerenotexcludedatthisstageofthescreening.
Theoptimisationofthechromatographyandthemassspectrometryoperatingconditionswasperformedusingtrypsinisedstandardproteins.
ThefinalSRMcomprisedthem/zoftheprecursor(Q1)andthem/zoftwoorthreeproductiontransitions(Q3)(ESMFig.
S2).
TheoptimalinstrumentalconditionsarelistedinTableS2(seeESM).
Thecrudeextract(6molL1urea,50mmolL1Ambic,pH8)ofthebiscuitproteinswashydrolysed(1:50,E:S)andthereleasedpeptideswereanalysedbySRM.
Theselectedpeptideswererefinedbasedonthesignalintensityandspec-ificityofeachtransition(interference).
ThepeptidesAMKPWIQPKandITVDDK-αS2;TPEVDDEALEKandIDALNENK-βLg;EAMAPKandEMPFPK-βCNwerenotincludedinthefinalmethodbecausethesignalintensitywasextremelylowandthechromatographicresolutionwaspoor.
Theinfluenceofdifferentbuffercompositionandcondi-tionswasstudiedinordertomaximiseproteinextraction.
Preliminaryscreeningoftheextractionconditionswasper-formedusingfivebuffers,twochaotropicagents(guanidinehydrochlorideandurea)andtworeductants(DTTandBME).
Theextractionbufferswereselectedaccordingtothosecom-monlyreportedintheliterature.
TheextractionperformancewasevaluatedattheproteinlevelbyRCDC.
Changingthebuffersdidnotrevealanymajorimpactonproteinextractability(ESMFig.
S3,panela).
Byaddingthechaotropicagentalone,theextractabilityincreasedbyafactorof7,whileincombinationwiththereductant,itfurtherincreasedbyafactorof3(ESMFig.
S3,panelaandc).
Ureawasthebestperformingchaotropicagentandincreasedtheyieldofextractionsignificantlybydenaturingproteins.
TheRCDCassayonlyprovidesinformationrelatingtothetotalproteinconcentration.
Itdoesnotdeliverdetailedinfor-mationaboutspecificproteins.
TheinformationatthepeptidelevelwasacquiredbySRManalysisofthetrypsinisedextract.
Aninfluenceoftheextractionbufferontrypsindigestionef-ficiencycannotbeexcluded,evenaftertheprecipitationandclean-upoftheproteins.
TheAmBic,theTEABandtheBBSweretheextractionbuffersthatshowedthegreatestcompati-bilitywithtrypsindigestion.
Thecombineduseofachaotropicandreducingagentincreasedthepeakareaofthepeptidesfromβ-lactoglobulinbyafactorof10,forlysozymebyafactorof13andforovalbuminbyafactorof4(ESMFigs.
S4toS7).
Theobservedincreasedependedonthesidechaininteractions(i.
e.
disulfidebridges)andthelocationofthepeptideintheproteinsequence.
Similarincreasesinsignal,whenusingthereducingreagents,werenotobservedforthealphaandbetacaseinproteins(ESMFig.
S6).
Thereductantsbreaktheglutennetworkhelpingthereleaseoftheproteinsandalsobreakthedisulfidebondsmakingtheproteinmoreaccessibletotheenzyme.
Therefore,itisnotsurprisingthatthereductantsarenotsignificantlyimprovingextractionanddigestionforthealphaandbetacaseinssincetheydonotcontaincysteinedisulfidebridges.
Incontrast,theκcaseindoescontainadisulfidebondandexperienceanincreaseinextractionontheinclusionofthereducingreagents.
Theper-formancebetweenthetworeducingagentsstudiedwasnotsignificantlydifferentformostpeptidesusingattest(P=95%)(ESMFigs.
S6-S7).
Therefore,DDTwasselectedforthefurtherstudiesasthiswasmorecompatiblewithcurrentworkingpractices.
Thescreeningevidencesuggestedthattherewaslittleob-servabledifferencebetweenthebackgroundbufferstudiedwhenlookingatthetotalproteincontent(ESMFig.
S3);theuseofchaotropicagentswasbeneficial,withurea,resultinginagreaterextractionefficiencythanguanidineformostpro-teins;theuseofareducingreagentwasbeneficialespeciallyforproteinswithdisulphidebridges,withlittleobservabledifferencebetweenDDTandBMR.
Basedonthisscreeningevidence,AmBic,ureaandDTTwerechosenasthebestperformingcomponentsoftheextractionbuffer.
Moreover,thesecomponentsareESI-MS-compatible.
Consequently,thesampledigestsolutioncanbedirectlyinjectedintotheLCMSforSRManalysis,minimisingthesamplehandlingwhichcanleadtopeptideloss.
Asmentionedpreviously,atwo-stageDoEwasusedtooptimisetheextractionanddigestionprotocolswiththeaimofcreatingasinglemulti-allergenmethod.
AFFDwasemployedtostudythemajorinfluencingvariablesoftheex-tractionanddigestionsteps.
Subsequently,aCCDwasusedtooptimisethestatisticallysignificantvariablesofeachofthesteps.
Followingtheoptimisationoftheexperimentalconditionsforscreening(amountofsample,choiceofextractionbuffer,etc.
),theoptimisationoftheextractionandthedigestionstepswereachievedinsequentialexperiments.
Inthefirststep,theextractionwasoptimisedbyfixingthedigestionparameters,i.
e.
agenericdigestionmethodwasusedwhilevaryingtheextractionconditions.
Inthesecondstep,thedigestionwasoptimisedusingthefixedoptimalextractionconditionsobservedpreviously.
IntheFFDoftheextraction,sixvariables(pH,temperature,time,AmBicconcentration,ureaconcentrationandDTTconcentration)werestudiedattwolevels.
ThevariablesstudiedandtheirrespectivelevelsaresummarisedinTable1.
Inaddition,acentralpointwas3468NitrideC.
etal.
includedinthedesignatwhichallfactorswerekeptatthemeanvalueofthetwolevels.
Thiswasusedtoestimatetheanalyticalprecisionoftheexperiments.
Thirty-twoexperi-mentswereperformedinarandomisedorder.
Thetotalpro-teinextractionefficiencywasdeterminedpriortotrypsindigestionusingtheRCDC,whiletheprotein-specificin-formationwasobtainedbymonitoringthepeakarearatiosoftheindividualpeptidesafterdigestion.
Generally,thetotalproteinyieldofextractionistheonlyresponsethatisoptimisedinproteinextractionexperiments,butinthispar-ticularcase,wherethegoalistoquantifypeptidesasasur-rogatefortheintactprotein,thestudyoftheindividualpep-tidepeakareasbecomesessential.
Accordingtoboththemultilinearregression(MLR)andthepartialleastsquare(PLS)models,constructedusingUnscramblersoftware,onlythetemperatureandtheureaconcentrationwerestatisticallysignificant(plevel0.
05).
Thecon-centrationofDTT(mmolL1)didnotsignificantlyaffectthepeakarea;therefore,itwassetat5mmolL1fortheoptimisedFig.
3Profileofmarkerpeptidesreleasedunderoptimiseddigestionconditionsover35hofdigestion.
Theadditionoftrypsinwasstaticandmultiple.
Staticadditionoftrypsinwasperformedonlyattime0hattwoE:Sratios—1:50and1:100.
Multipleadditionsoftrypsinwereperformedatequidistantintervalsoftimetoafinal1:25(E:S)ratio—1:50(E:S)ratiowasaddedattime0andtime22h;1:100(E:S)ratiowasaddedattime0,6,12and22h.
Errorbarsrepresentthestandarddeviationof3replicatemeasurements.
aαS1CN-YLGYLEQLLR.
bαS2CN-FALPQYLK.
cOva-LYAEER.
dLys-FESNFNTQATNRAnassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3471Table3Overallrelativevariancefortherepeatanalysisofmilk-andegg-derivedpeptides.
Contributionsexpressedas%relativefromextraction,digestionandLCMSpeakarearatiomeasurementsProteinPeptideSourceofvariabilitySSMSTotalCV%Variance%oftotalαS1YLGYLEQLLR*Extraction0.
1460.
073514.
0Digestion0.
2800.
04767.
8Ratiomeasurement0.
0690.
00418.
2FFVAPFPEVFGKExtraction2.
29666E+121.
14833E+1210.
30.
8Digestion6.
6312E+121.
1052E+1241.
5Ratiomeasurement6.
30122E+123.
50068E+1157.
7αS2VIPYVRExtraction65,911,263,31532,955,631,6586.
42.
0Digestion1.
22099E+1120,349,773,8380.
0Ratiomeasurement9.
14193E+111.
01577E+1198.
0FALPQYLK*Extraction0.
5460.
2735.
60.
0Digestion1.
6940.
28275.
4Ratiomeasurement0.
3560.
04024.
6NAVPITPTLNRExtraction6.
12179E+113.
0609E+11450.
4Digestion9.
24282E+111.
54047E+110.
0Ratiomeasurement6.
32202E+137.
02446E+1299.
6ALNEINQFYQKExtraction21,969,405,62810,984,702,81410.
90.
0Digestion88,508,367,12914,751,394,52242.
0Ratiomeasurement83,726,462,7914,651,470,15558.
0βCNVLPVPQKExtraction3.
62137E+131.
81068E+134017.
7Digestion6.
57953E+131.
09659E+1381.
4Ratiomeasurement6.
92243E+1138,457,965,1450.
9AVPYPQRExtraction6.
2232E+123.
1116E+1237.
20.
0Digestion2.
13986E+133.
56643E+1292.
8Ratiomeasurement1.
61913E+1289,951,794,8437.
2κCNYIPIQYVLSRExtraction1.
14481E+125.
72405E+116.
42.
8Digestion2.
99726E+124.
99543E+1138.
5Ratiomeasurement3.
03071E+121.
68373E+1158.
7βLGGLDIQKExtraction9,374,383,9914,687,191,99623.
10.
0Digestion29,866,046,8984,977,674,48379.
0Ratiomeasurement5,269,055,283585,450,587.
121.
0IPAVFKExtraction76,749,448,70338,374,724,3528.
74.
5Digestion1.
81446E+1130,241,083,29027.
4Ratiomeasurement2.
46673E+1113,704,048,15168.
1ALPMHIRExtraction4,533,019,3262,266,509,6639.
60.
0Digestion34,287,299,4115,714,549,9022.
0Ratiomeasurement97,025,982,7795,390,332,37798.
0OvaLYAEER*Extraction0.
0398860190.
0199430097.
37.
0Digestion0.
0913919290.
01523198855.
0Ratiomeasurement0.
0513012310.
00285006838.
0GLWEKExtraction3.
13527E+111.
56763E+1111.
80.
1Digestion9.
37364E+111.
56227E+1170.
8Ratiomeasurement3.
38734E+1118,818,572,80629.
1HIATNAVLFFGRExtraction4.
15205E+112.
07602E+118.
936.
7Digestion2.
9878E+1149,796,694,36320.
4Ratiomeasurement3.
68934E+1120,496,340,99842.
9LysGTDVQAWIRExtraction53,321,662,87926,660,831,4407.
433.
6Digestion49,068,398,6418,178,066,44033.
8Ratiomeasurement35,821,898,5721,990,105,47632.
63472NitrideC.
etal.
method.
TheresponsesurfacesdemonstratingtheeffectsofthesignificantfactorsonthepeakarearatioareshowninFig.
2andESMFig.
S11.
Forallthepeptidesstudied,amax-imumwasobservedafter18hofdigestion;thus,thetimeneededtocarryouttheextractionwasfixedat18h.
Insum-mary,theoptimisedconditionsusedforthedigestionwerefixedat25mmolL1AmBic,5mmolL1DTT,10%DMSOwithadigestiontimeof18h.
Apeptideactingasareliablequantifiermustbefullyre-leasedfromitsconstituentprotein.
Thepeptidehastobeinastableforminthefinaldigestionsolutionsuchthatitsmolaramountisrepresentativeofthemolesofproteinpresentintheextract.
Anydegradationorunreleasedpeptidewillresultintheincorrectdeterminationoftheconcentrationofproteininitiallypresentinthesample.
Toinvestigatetherateofpro-teindigestion,atimecourseexperimentwasperformed.
Severalaliquotsoftheextractweredigested,andateachtimepointofthedigestion,analiquotwastakenforanalysis.
TheeffectsofenzymeconcentrationanddigestiontimeforfourpeptidesfromthedifferenteggandmilkproteinsareshowninFig.
3andESMFig.
S12.
Itcanbeclearlyobservedthatforpeptidesderivedfromtheeggproteins,lysozymeandovalbu-min,adigestiontimeinexcessof16hwasrequiredtoreachaplateauforthepeptideconcentration(Fig.
3andESMFig.
S12).
Thisplateaueffectcanbeindicativeofthetotaldiges-tionoftheproteinpresent.
Hence,theamount-of-substanceconcentrationofthepeptidewouldrepresenttheconcentrationoftheinitialprotein.
Theeffectofdigestiontimeonthemea-suredmilkpeptidesismarkedlydifferentfromthoseofegg(Fig.
3andESMFig.
S12).
TheαS1caseinpeptidesarecompletelyreleasedinarelativelyshortdigestiontimeandcanbeobservedtodecayafter16h.
Thismayindicatethedegradationand/orthechemicalmodificationofthepeptideonceitisreleasedfromtheparentprotein.
Thetwoβ-caseinpeptidesandoneβ-lactoglobulinpeptidewereadverselyaffectedbybothenzymeconcentrationandtime(ESMFig.
S12,panelf,i,j).
Formostproteins,adigestiontimeof12–20hwasoptimal,butagain,acarefulselectionofthepeptidesthatarestableandfullyreleasedfromtheirconstituentproteinisessentialforaccuratequantification.
Theoveralleffectandtherelativeuncertaintycontributionfromtheextractionanddigestionstepswereinvestigatedusingasimpletwo-factorfullnesteddesign.
Forthis,threeseparateproteinextractswereeachdigestedthreetimesandeachdigestwasanalysedintriplicate.
ThedigestswereanalysedinarandomorderontheLCMSandstatisticalout-lierswereremovedbeforeperformingtheANOVA.
Anad-vantageofperformingsuchanexperimentisthattheindivid-ualcontributionsoftheextraction,digestionandLCMSanal-ysistotheoverallmeasurementvariabilitycanbequantified.
Thisisinformativewhendesigningthemethodvalidationplanasthefactorsinfluencingtheoveralluncertaintycouldbebettercontrolledand/orrepeatedtoassessoverallprecision.
Theoverallprecisionofthemeasurementandtheindividualcontributionsfromtherepeatabilityofextraction,digestionandmeasurementsforthemilkandeggwhitepeptidesarereportedinTable3.
Asexpected,themeasurementprecisionfortheβ-caseinandtheβ-lactoglobulinpeptides,whichshowedrapiddegradationoverthedigestiontime,waspoor.
Wherepresent,theadditionofisotope-labelledpeptideinter-nalstandardsimprovedtheprecisionofpeptidemeasurementsgreatly.
Theresultsobtainedfromthesestudiessuggestthat,fortherelativelyhigh-levelallergencontaminationcookiestudied,measurementprecisionsaslowas5%relativestandarddevi-ation(RSD)couldbeachieved.
WhereRSDsincrease,nor-mallyduetothelackofanisotopicinternalstandardoralowmeasuredsignalintensity,digestionwasthegreatestsourceofvariationinthemajorityofcases.
ThestudysuggeststhatwhileMSiscapableofthemultiplexeddetectionofmanyTable3(continued)ProteinPeptideSourceofvariabilitySSMSTotalCV%Variance%oftotalFESNFNTQATNR*Extraction0.
0062004140.
0031002076.
50.
0Digestion0.
0229790510.
00382984272.
5Ratiomeasurement0.
0077242250.
00042912427.
5HGLDNYRExtraction39,193,790.
6319,596,895.
3228.
40.
0Digestion856,334,672.
1142,722,445.
386.
4Ratiomeasurement127,696,469.
57,094,248.
30613.
6MyoExtraction4.
88984E+122.
44492E+125.
324.
8Digestion4.
70281E+127.
83802E+1115.
1Ratiomeasurement8.
04677E+124.
47043E+1160.
1Glu-FibExtraction1.
2567E+126.
28352E+11149.
4Digestion1.
43487E+122.
39144E+110.
0Ratiomeasurement3.
42278E+122.
0134E+1190.
6*Peptidesforwhichanisotope-labelledpeptidestandardwasavailableAnassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3473peptides,thatauniversalbufferwhichextractsallproteinsequallyisunlikely.
Therefore,carefulselectionofquantitativemarkerproteinsandrobustpeptideswillberequired.
ConclusionsTheabilityofmodernLCMStosensitivelyandselectivelydetectthepresenceofallergenpeptideshasresultedinvariousmultiplexedmethodsforthesimultaneousdetectionoffoodallergens.
However,whenquantifyingallergens,allpartsoftheanalyticalprocedureneedtobeconsidered,notjustthefinalMSmethod.
WhileMScouldquantifymanypeptidesinonerun,theabilitytocompletelyextractanddigestthetargetmarkerproteinsisessentialifaccurateresultsaretobeobtain-ed.
Thenumberofallergensthatcouldbeaccommodatedreasonablyinamultiplexedassaydependsonthefollowing:thephysicochemicalpropertiesandstructureoftheproteins;thecomplexityofthefoodmatrixanditsintermolecularinter-actions;andthefoodprocessingconditions(doughformation,baking,extrusion,andcooking/boiling).
Theoptimisationofproteinextractionanddigestionisfun-damentaltoprovidereliablequantificationifproteotypicpep-tidesaretobeusedfordeterminingtheamountofthealler-genicfoodproteinpresent.
Currentmethodscommonlyuseisotopicallylabelledpeptidesasinternalstandard.
Thisap-proachisusuallyreducingthemeasurementbiasandimprov-ingtheprecisionofmeasurementsatthepeptidelevel.
However,differencesinextractionareoftenoverlooked.
ThenumberofmultiplexedMSmethodscurrentlybeingreported,whereasinglecommonbufferisusedfortheextractionofmultipleproteinsfrommanydifferentallergenicfoodsources,suggeststhatMShasalottoofferintheareaoffoodallergenquantification.
However,failuretofullyaddressallaspectsoftheanalyticalproceduremayresultinunreportedbiasbythecurrentlyusedapproaches.
Theevidencepresentedinthisstudyshowsthatthequantitativeremovalofproteinsfromheat-treatedmatricesisstillamajorchallengeformultiplexedmethods.
Whilstmultiplexeddetectionispossible,theaccu-ratequantificationofthetotalfoodallergenproteinmassfrac-tionwillnotbefeasiblewithoutauniversalbufferthatextractsanddissolvesthewidevarietyofproteinsderivedfromthedifferentallergenicfoodsources.
Inthisstudy,thedevelopedmethod,specificallydesignedfortheabsolutequantificationofeggandmilkallergensinbiscuits,requiredacompromisedextractionanddigestionpro-tocol.
Nouniversalconditionswerefoundthatmaximisedtheresponseofallthetargetpeptidesfrommilkandegg.
Therefore,notallconstituentproteinscanbeconsideredgoodquantitativemarkersandshouldsimplybeusedasqualitativemarkers.
Knowingthedigestionkineticsdeliveredevidencethatpeptideswererepresentativeoftheproteincontent,there-byprovidingconfidenceinthedeterminedconcentrationforeachprotein.
Withtheinclusionoflabelledpeptideinternalstandards,theprecisionoftheMSpeakarearatiomeasure-mentshadverylittleimpactontheoverallvariabilityoftheanalyticalresult.
Therefore,tightercontrolofthedigestionandextractionproceduresisessentialforachievingimprovedmeasurementuncertaintiesinthefuture.
AcknowledgmentsThisworkwasundertakenaspartoftheEUiFAAMproject:IntegratedApproachestoFoodAllergenandAllergyRiskManagement(GrantAgreementNo.
312147),whichalsoprovidedthecookiematerialanalysedinthisstudy.
CompliancewithethicalstandardsConflictofinterestTheauthorsdeclarethattheyhavenoconflictsofinterest.
OpenAccessThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.
0InternationalLicense(http://creativecommons.
org/licenses/by/4.
0/),whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCommonslicense,andindicateifchangesweremade.
References1.
Regulation(EU)No1169/2011oftheEuropeanParliamentandoftheCouncilof25October2011ontheprovisionoffoodinforma-tiontoconsumers,amendingRegulations(EC)No1924/2006and(EC)No1925/2006oftheEuropeanParliamentandoftheCouncil,an.
In:OJL304,2011.
2.
RemingtonBC,BaumertJL,BlomMW,HoubenGF,TaylorSL,KruizingaAG.
QuantitativeriskassessmentofUKfoodproductscross-contaminatedwithallergens.
ClinTranslAllergy.
2015;5:O10.
https://doi.
org/10.
1186/2045-7022-5-S3-O10.
3.
AllenKJ,RemingtonBC,BaumertJL,CrevelRWR,HoubenGF,Brooke-TaylorS,etal.
Allergenreferencedosesforprecautionarylabeling(VITAL2.
0):clinicalimplications.
JAllergyClinImmunol.
2014;133:156–64.
https://doi.
org/10.
1016/J.
JACI.
2013.
06.
042.
4.
Diaz-AmigoC,PoppingB.
Foodallergens:aregulatory/labellingoverviewincludingtheVITALapproach.
EncyclFoodChem.
2019:615–21.
https://doi.
org/10.
1016/B978-0-08-100596-5.
21798-8.
5.
WalkerMJ,BurnsDT,ElliottCT,GowlandMH,MillsENC.
IsfoodallergenanalysisflawedHealthandsupplychainrisksandaproposedframeworktoaddressurgentanalyticalneeds.
Analyst.
2016;141:24–35.
https://doi.
org/10.
1039/C5AN01457C.
6.
O'ConnorG,HaponiukM,UberthFJointDGSANTandDGJRCWorkshop-HarmonisationofapproachesforinformingEUallergenlabellinglegislation,JRC108259.
7.
HeickJ,FischerM,PppingB.
Firstscreeningmethodforthesimultaneousdetectionofsevenallergensbyliquidchromatogra-phymassspectrometry.
JChromatogrA.
2011;1218:938–43.
https://doi.
org/10.
1016/j.
chroma.
2010.
12.
067.
8.
MonaciL,PilolliR,DeAngelisE,GodulaM,ViscontiA.
Multi-allergendetectioninfoodbymicrohigh-performanceliquidchro-matographycoupledtoadualcelllineariontrapmassspectrometry.
JChromatogrA.
2014;1358:136–44.
https://doi.
org/10.
1016/j.
chroma.
2014.
06.
092.
3474NitrideC.
etal.
9.
ParkerCH,KhudaSE,PereiraM,RossMM,FuTJ,FanX,etal.
Multi-allergenquantitationandtheimpactofthermaltreatmentinindustry-processedbakedgoodsbyELISAandliquidchromatography-tandemmassspectrometry.
JAgricFoodChem.
2015;63:10669–80.
https://doi.
org/10.
1021/acs.
jafc.
5b04287.
10.
CristinaL,ElenaA,DavideC,MarziaG,LuciaD,CristianoG,etal.
Validationofamassspectrometry-basedmethodformilktracesdetectioninbakedfood.
FoodChem.
2016;199:119–27.
https://doi.
org/10.
1016/j.
foodchem.
2015.
11.
130.
11.
LutterP,ParisodV,WeymuthH.
Developmentandvalidationofamethodforthequantificationofmilkproteinsinfoodproductsbasedonliquidchromatographywithmassspectrometricdetection.
JAOACInt.
2011;94:1043–59.
12.
MonaciL,LositoI,DeAngelisE,PilolliR,ViscontiA.
Multi-allergenquantificationoffining-relatedeggandmilkproteinsinwhitewinesbyhigh-resolutionmassspectrometry.
RapidCommunMassSpectrom.
2013;27:2009–18.
https://doi.
org/10.
1002/rcm.
6662.
13.
CryarA,PritchardC,BurkittW,WalkerM,O'ConnorG,BurnsDT,etal.
Towardsabsolutequantificationofallergenicproteinsinfood-lysozymeinwineasamodelsystemformetrologicallytraceablemassspectrometricmethodsandcertifiedreferencemate-rials.
JAOACInt.
2013;96:1350–61.
https://doi.
org/10.
5740/jaoacint.
12-438.
14.
TolinS,PasiniG,CurioniA,ArrigoniG,MasiA,MainenteF,etal.
Massspectrometrydetectionofeggproteinsinredwinestreatedwitheggwhite.
FoodControl.
2012;23:87–94.
https://doi.
org/10.
1016/j.
foodcont.
2011.
06.
016.
15.
TolinS,PasiniG,SimonatoB,MainenteF,ArrigoniG.
AnalysisofcommercialwinesbyLC-MS/MSrevealsthepresenceofresidualmilkandeggwhiteallergens.
FoodControl.
2012;28:321–6.
https://doi.
org/10.
1016/j.
foodcont.
2012.
05.
015.
16.
GrovesK,CryarA,WalkerM,QuagliaM.
Assessmentofrecoveryofmilkproteinallergensfromprocessedfoodformassspectrome-tryquantification.
JAOACInt.
2018;101:152–61.
https://doi.
org/10.
5740/jaoacint.
17-0214.
17.
LositoI,IntronaB,MonaciL,MinellaS,PalmisanoF.
DevelopmentofamethodforthequantificationofcaseinatetracesinItaliancommercialwhitewinesbasedonliquidchromatography–electrosprayionization–iontrap–massspectrom-etry.
JAgricFoodChem.
2013;61:12436–44.
https://doi.
org/10.
1021/jf4034909.
18.
ArseneCG,OhlendorfR,BurkittW,PritchardC,HenrionA,O'ConnorG,etal.
Proteinquantificationbyisotopedilutionmassspectrometryofproteolyticfragments:cleavagerateandaccuracy.
AnalChem.
2008;80.
https://doi.
org/10.
1021/ac7024738.
19.
BurkittWI,PritchardC,ArseneC,HenrionA,BunkD,O'ConnorG.
TowardSystèmeInternationald'Unité-traceableproteinquanti-fication:fromaminoacidstoproteins.
AnalBiochem.
2008;376.
https://doi.
org/10.
1016/j.
ab.
2008.
02.
010.
20.
PritchardC,GrovesKJ,BiesenbruchS,O'ConnorG,AshcroftAE,ArseneC,etal.
Quantificationofhumangrowthhormoneinserumwithalabeledproteinasaninternalstandard:essentialconsider-ations.
AnalChem.
2014;86:6525–32.
https://doi.
org/10.
1021/ac501032q.
21.
NewsomeGA,SchollPF.
QuantificationofallergenicbovineMilkαS1-caseininbakedgoodsusinganintact15N-labeledproteininternalstandard.
JAgricFoodChem.
2013;61:5659–68.
https://doi.
org/10.
1021/jf3015238.
22.
SteinhoffM,FischerM,Paschke-KratzinA.
Comparisonofextrac-tionconditionsformilkandhen'seggallergens.
FoodAdditContamPartA.
2011;28:373–83.
https://doi.
org/10.
1080/19440049.
2010.
545957.
23.
JohnsonPEP,BessantC,ValeriaC,GiosafattoL,BaumgartnerS,AldickT,etal.
Currentperspectivesandrecommendationsforthedevelopmentofmassspectrometrymethodsforthedeterminationofallergensinfoods.
JAOACInt.
2011;94:1026–33.
24.
GomaaA,BoyeJI.
Impactofthermalprocessingtimeandcookiesizeonthedetectionofcasein,egg,glutenandsoyallergensinfood.
FoodResInt.
2013;52:483–9.
https://doi.
org/10.
1016/j.
foodres.
2013.
01.
019.
25.
ShubhadaS,SundaramPV.
Borateion-assistedstabilizationofbeta-galactosidasefromAspergillusoryzaebypolyhydroxycom-poundsinwater-miscibleorganicsolvents.
EnzymMicrobTechnol.
1993;15:881–6.
26.
GreeneRF,PaceCN.
Ureaandguanidinehydrochloridedenatur-ationofribonuclease,lysozyme,α-chymotrypsin,andβ-Lactoglobulin.
JBiolChem.
1974;249:5388–93.
27.
HerveyWJ,StraderMB,HurstGB.
Comparisonofdigestionpro-tocolsformicrogramquantitiesofenrichedproteinsamples.
JProteomeRes.
2007;6:3054–61.
https://doi.
org/10.
1021/pr070159b.
28.
KolliparaL,ZahediRP.
Proteincarbamylation:invivomodifica-tionorinvitroartefactProteomics.
2013;13:941–4.
https://doi.
org/10.
1002/pmic.
201200452.
29.
SunS,ZhouJ-Y,YangW,ZhangH.
Inhibitionofproteincarbamylationinureasolutionusingammonium-containingbuffers.
AnalBiochem.
2014;446:76–81.
https://doi.
org/10.
1016/j.
ab.
2013.
10.
024.
30.
StraderMB,TabbDL,HerveyWJ,PanC,HurstGB.
Efficientandspecifictrypsindigestionofmicrogramtonanogramquantitiesofproteinsinorganicaqueoussolventsystems.
AnalChem.
2006;78:125–34.
https://doi.
org/10.
1021/ac051348l.
Publisher'snoteSpringerNatureremainsneutralwithregardtojurisdic-tionalclaimsinpublishedmapsandinstitutionalaffiliations.
Anassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3475
Formanyfoodindustries,productandrawingredientsurveillanceformspartoftheirriskassessmentprocedures.
Thismayrequirethedetectionofmultipleallergensinawidevarietyofmatrices.
Massspectrometryoffersapossiblesolutionforthequantificationofmultipleallergensinasingleanalysis.
ThecapabilityofMStoquantifymanypeptidesfromacomplexproteindigestioniswellknown.
However,alackofmatrixcertifiedreferencematerialshasmadetheoptimisationofextractionanddigestionconditionsformultiplexedallergenquantificationdifficulttoassess.
Here,wereportasystematicstudy,usingprelim-inaryscreeningfollowedbyaDesignofExperimentsapproach,tofindtheoptimalbufferanddigestionconditionsfordetectingmilkandeggproteinmarkersinamodelprocessedfoodmatrix.
Fiveofthemostcommonlyusedbuffers,twochaotropicreagentsandtworeducingreagentswereassessedfortheoptimalextractionofmultipleproteinmarkers.
Whilethechoiceofbackgroundbufferhadlittleimpact,theuseofchaotropicandreducingreagentsshowedsignificantbenefitsfortheextractionofmostproteins.
Afullfactorialdesignexperimentwasappliedtotheparametersshowntohaveasignificantimpactonproteinrecovery.
Thesestudiessuggestthatasingleoptimalsetofextractionconditionsenablingthequantitativerecoveryofallproteinsisnoteasilyachieved.
Therefore,althoughMSiscapableofthesimultaneousquantificationofmanypeptidesinasinglerun,greaterconsiderationofproteinextractionisrequiredbeforetheseareappliedformultiplexallergenquantificationinfoodmatrices.
KeywordsFoodallergens.
Massspectrometry.
Protein.
Extraction.
QuantificationIntroductionFoodallergyiscausedbytheadversereactionoftheimmunesysteminasensitiveindividualtowardaspecificfood.
Currently,thereisnoclinicaltreatmentorcureforthosesuf-feringfromafoodallergy.
Affectedpeoplemustbecarefulaboutwhattheyeatandrelyonappropriatefoodlabellingtoavoidexposuretotheoffendingfood.
EUlegislation,namelyRegulation(EU)No1169/2011[1],requiresfoodmanufac-turerstolistallinformationregardingfoodingredientsontheproductlabelandtohighlightthepresenceofaspecificgroupoffoodingredientsthatmaytriggerfoodallergiesandintol-erances.
Theunintentionalpresenceofanallergenrepresentsafoodsafetyhazardwhichisparticularlydifficulttomanage.
Foodallergensmayenterthefoodsupplychainatanytimeduringstorage,handlingandpackagingasaconsequenceofcross-contactwithinprocessinglines.
Foodbusinessoperatorsshouldadoptriskmanagementprocedurestoprevent,ElectronicsupplementarymaterialTheonlineversionofthisarticle(https://doi.
org/10.
1007/s00216-019-01816-z)containssupplementarymaterial,whichisavailabletoauthorizedusers.
*GavinO'ConnorJRC-Allergens@ec.
europa.
eu1EuropeanCommission,Directorate-GeneralJointResearchCentre,Retieseweg111,2440Geel,Belgium2Presentaddress:SchoolofBiologicalSciences,DivisionofInfection,ImmunityandRespiratoryMedicine,ManchesterAcademicHealthScienceCentre,ManchesterInstituteofBiotechnology,TheUniversityofManchester,131PrincessStreet,Manchester,UK3Presentaddress:Fachbereich3.
2Biochemie,Physikalisch-TechnischeBundesanstalt,Bundesalle100,38116Braunschweig,GermanyAnalyticalandBioanalyticalChemistry(2019)411:3463–3475https://doi.
org/10.
1007/s00216-019-01816-zeliminate,orreducetheriskofcontaminationtoacceptablelevels,thusavoidingtheoccurrenceoffoodallergyincidentsandfoodproductrecalls.
Aquantitativeriskassessmentmeasurestheimpactofaspecifichazard[2]asacombinationofreferencedosesandexposure.
SuchanapproachiscurrentlyusedbytheVITAL(VoluntaryIncidentalTraceAllergenLabelling)riskassess-mentprocedures[3].
Thethresholdsforallergenicproteinsinfoodarebasedonclinicaldata(OralFoodChallenge)andareindicatorsoftheactionlevels,expressedasthetotalpro-teinoftheallergenicfood(mg),belowwhichonlythemostsensitiveallergicsubjectsmayreact.
Precautionaryallergenlabelling(PAL),suchasBmaycontain^statements,shouldonlybeusedafterperformingaquantitativeriskassessment,theresultsofwhichsuggestedademonstrableandsignificantriskofallergencontaminationatlevelsabovetheactionlevel[4].
Analyticalmethodshelptosupportthevalidationandverificationoftheriskassessmentoutcomes,providingquan-titativemeasurementsoftheallergenicingredientinprocessedfood[5].
Anymethodmeasuringallergenmarkersmustex-presstheirresultsinquantitiesandmeasurementunitsthatarecomparablewithothermethods.
Foruseintheriskassessmentprocedures,thisrequirestheallergencontenttobeexpressedinmilligrammesofallergenproteinperkilogrammeoffood[6].
Thefeasibilityofmassspectrometry-basedmethodsforthedetectionandquantificationofallergenshasbeenproven[7–10].
Aconsiderablenumberofstudieshavebeendedicatedtotheoptimisationofmassspectrometry-basedmethodsforthedetectionofeggandmilkproteinsusingpeptidesfromasingleBmarker^proteinfromeachfood[7–15].
Oneofthemajoradvantagesofmassspectrometryisitsabilitytosequen-tiallydetectmarkerpeptidesenablingthemultiplexedanalysisofmanyallergens.
Severalfoodmatrixeshavebeenad-dressed:biscuits[7–10,16];breakfastcereals;infantfood[11];andwine[12–15,17],usingadiverserangeofmassspectrometryplatformsandmethods.
Inparticular,atargetedmassspectrometryapproach,knownasselectivereactionmonitoring(SRM),wassuccessfullyappliedforthedetectionofmulti-allergentargets[7–13].
Thismethodhastheadvan-tageofselectivelydetectingmarkerpeptides,fromspecificallergenproteins,incomplexbackgroundmatricesandallowsthemeasurementofproteinabundancesthatdifferbyfourtofiveordersofmagnitude.
Peptidesarequantifiedbasedonmonitoringspecifiedprecursorpeptide-to-productiontransi-tions.
Forabsolutequantification,themethodofisotopedilu-tionmassspectrometry(IDMS)isoftenapplied[18].
Thisrequiresthemeasurementoftherelativeresponseofthenatu-ralpeptide,generatedbyproteolyticcleavageofthetargetprotein,anditsheavyisotopicallylabelledform,addedataknownquantitytoboththestandardsandthesamples[19].
Thismethodofmeasuringthepeakarearatiosreducessys-tematicerrorsinpeptidequantification[9–12].
However,astheinternalstandard(isotopicallylabelledpeptide)isaddedasthefreepeptide,itdoesnotredressanydeficienciesinproteinextractionorreleaseofthepeptidefromitsprecursorproteinduringdigestion[20].
Therefore,thelargestcontributortothemeasurementuncertaintyand/orbiaswhenlabelledinternalpeptidestandardsareusedisexpectedtoderivefromtheex-tractionanddigestionstepsintheanalysis.
Itisessentialtoensurethattotalextractionandcompletedigestionofthepro-teinsoccurasonlythiswillenableatrueassessmentoftheconcentrationoftheallergenpresent[16].
Tocircumventthisissue,expressionofabovine15N-αS1-caseinprotein,tocorrectforextractionandrecoveryafterbaking,wassuggestedbyNewsomeetal.
[21].
Althoughencouragingrecoveryfactorswerereported,thelabelledpro-teinwasspikedintothebiscuitsafterthebakingprocess;therefore,theunprocessedlabelledproteinmaynotpossessthepossibleheat-inducedmodificationsresultingindifferentextractionanddigestionbehavioursbeingexperiencedbythenaturalandlabelledprotein.
Despitethefactthatmostofthepublishedreports[6–14]selectthesamemarkerproteinsandproteotypicpeptides,frommilkandegg,theextractionanddigestionprotocolsdiffersignificantly,evenwhenthesamefoodmatrixisanalysed.
UnlikeinELISA[22],nosystematicinvestigationconcerningtheoptimisationofextractionanddigestionprotocolsforfoodallergenquantificationhasbeenperformedwhenusingMS.
Theaimofthestudypresentedinthispaperwastheopti-misationoftheextractionanddigestionprotocolsforthequantitativeremovalofmilkandeggproteinsfromamodelprocessedfoodmatrix.
Thequantitativetargetedproteomicpipelinewasseparatedintothreecriticalcomponentsthatcon-tributetothetruenessanduncertaintyofthefinalanswer:extraction,digestionandSRManalysis.
ThefirststepconsistsintheselectionofthemarkerpeptideandthedevelopmentoftheSRMmethod,includingproteotypicpeptidesfromfivemilkallergenicproteinsandtwoeggallergenicproteins.
WeinvestigatedtheimpactofdifferentextractionconditionsontheSRMpeptidesignalresponse,whichisthechemicalentitymeasuredinabottom-upproteomicexperiment.
Thepeptideabundancereflectsthevariationoftheperformanceoftheextractionoftheproteinsfromthefood.
Thefoodmatrixstudiedwasabiscuitincurredwiththeallergenicproteinspriortoheattreatment,therebymimickingcommerciallyavailableproducts.
AnexperimentaldesignorDesignofExperiments(DoE)approachwasusedfortheoptimisationoftheextractionanddigestionconditions.
Themainadvan-tageofDoEisthatthemaximuminformationoftheevaluatedsystemisobtainedbytheminimumnumberofexperiments.
Moreover,itallowsthestudyofmanyvariablesatthesametimeprovidinginformationontheinteractionamongthevar-iables,whichcannotbeachievedusingaonevariableatatimeapproach.
Bythissimultaneousoptimisationofthemultiplepeptideresponses,anoptimalcompromisehastobereachedbetweenthemtohaveonemethodthatsuitsall.
ANOVAon3464NitrideC.
etal.
thefinaloptimisedconditionsallowedtheassessmentoftheimpactofthreeanalyticalstages(extraction,digestionandmeasurement)ontheoverallvariability.
Tothebestofourknowledge,thisisthefirstquantitativestudyshowingtheextenttowhichpeptideresponsecanvarydependingontheextractionanddigestionconditionschosen.
MaterialsandmethodsMaterialsStandardproteinsofαS-casein(C6780),β-casein(C6905),κ-casein(C0406),β-lactoglobulinA(L7880),ovalbumin(A5503),lysozyme(L6876),[Glu1]-fibrinopeptideBhuman(Glu-Fib)(F3261-1MG),myoglobinfromequineheart(M1882),peptidestandards(angiotensinIIhuman(A9525-1MG),Met-Arg-Phe-Alaacetatesalt(M1170-1MG),bradyki-ninacetatesalt(B3259-1MG),angiotensinIhumanacetatesalthydrate(A9650-1MG)werepurchasedfromSIGMA(St.
Louis,MO,USA).
Thereagentsusedforproteinextractionanddigestionin-cludingacetone,hydrochloricacid,sodiumhydroxide,urea,guanidine-HCl,sodiumphosphatebuffer(PBS),ammoniumbicarbonate(AmBic),trizmabase,boricacid,sodiumtetraborate,triethylammoniumhydrogencarbonatebuffer(TEAB),dithiothreitol(DTT),iodoacetamide(IAA)werepur-chasedfromSIGMA(St.
Louis,MO,USA).
Trifluoroaceticacid(TFA)andmercaptoethanol(BME)werepurchasedfromMerck(Darmstadt,Germany).
SolventsincludingULC-gradeacetonitrile(ACN),ULC-gradewaterand99%formicacid(FA)werepurchasedfromBiosolve(Valkenswaard,TheNetherlands).
TrypsinGold-MassSpecGradewaspurchasedfromPromega(Madison,USA).
HyperSepC18(200mg,3mL)CartridgeswereobtainedfromThermoScientific(BiopolymersUlm,Germany).
Thefollowingheavy-isotope-labelledpeptides,wherebyallthecarbonandnitrogenatomsinthen-terminuslysineorargininewerereplacedwith99%13Cand15N,YLGYLEQLLR*(αS1),FALPQYLK*(αS2),LYAEER*(Ova),FESNFNTQATNR*(Lys),werepurchasedfromJPT(JPTPeptideTechnologiesGmbH,Germany).
Twobatchesofbiscuits,oneblankandonecontaining5000mgofproteiningredientperkilogrammeofbiscuit,werepreparedfollowingacommonrecipeforcookiesbytheInstituteofFoodResearchinLeatherhead,UK,intheframeoftheiFAAMproject.
Thebakingtemperaturewas150°Cfor9min.
MilkandeggingredientswerepurchasedfromBIOSERVICEZachGmbH(Austria).
TheSprayDriedskimmedmilkpowder(09G010)contained34.
8%w/wmilkproteinandtheeggwhitepowder(1-022161VOIB)contained71.
6%w/weggprotein.
Thebiscuitrecipealsoincludedhazelnut,peanutandwalnutflours;however,theseproteinswerenottargetedbytheanalysisoutlinedhere.
AdetaileddescriptionoftherecipeisgivenintheElectronicSupplementaryMaterial(ESM;S1).
OperatingconditionsfortheselectivereactionmonitoringbymassspectrometryFiveallergenicproteinsfrommilk(αS1casein—Bosd9,αS2casein—Bosd10,β-casein—Bosd11,κ-casein—Bosd12,β-lactoglobulin—Bodd4)andtwofromeggwhite(ovalbu-min—Gald2,lysozyme—Gald4)wereselectedasmarkersthatwouldindicatethepresenceofthefoodingredients.
Proteinsequencesweredigestedinsilicausingthepredictedcleavagesitesfortrypsin,andtheproteotypicmarkerpeptideswereselectedbasedonspecificcharacteristics.
Theappliedcriteriaforpeptidemarkerselectionincludedthefollowing:theuniquenessofthepeptideforthemarkerprotein,deter-minedusingtheMS-Homologyblasttoolofproteinprospec-tor(http://prospector.
ucsf.
edu/prospector/mshome.
htm),searchingtheSwissprotandtheNCBInrdatabases;alackofpeptidescontainingcysteineandnaturalpost-translationalmodifications(PTMs);theresultingtrypticpeptidescontain-ingbetween5and12aminoacidresidues[23].
AXevoTQ-Striplequadrupole(QqQ)massspectrometer(Waters,Manchester,UK)equippedwithaTrizaicionsourcewasusedtoacquiretheSRMdataforthisstudy.
Standardproteinsweredigestedwithtrypsin(50mmolL1Ambic,pH8;1:50,E:S)andthereleasedpeptidesweredirectlyin-fusedintotheQqQusingasyringepumpat1.
5μLmin1.
Theoptimisedsourceparametersusedwereasfollows:capillaryvoltage3.
5kV;conevoltage25V;sourcetemperature100°C;conegasflow50Lh1;collisiongasflow0.
11mLmin1.
Nitrogen(99.
998%purity,L'AirliquidBelgie,Liège,Belgium)andargon(99.
999%purity,L'AirliquidBelgie,Liège,Belgium)wereusedastheconeandcollisiongases,respectively.
TheQqQwasfirstoperatedinascanningmodetodefinetheoptimumconditionsfortheprecursorionsignals.
Onlychargestatesbetween+2and+4wereconsideredwhenoptimisingthesourceconditionsasthesewereresponsibleforthemostintensepeptidesignalsfromthedigestedmilkpep-tides.
Subsequently,themassspectrometerwasoperatedinproductionscanningmode,wherethetopy-andb-ionswererankedinincreasingorderofintensityandthetop6wereselectedforfurthercollisionenergy(CE)rampingoptimisa-tion.
Sevendifferentcollisioncellvoltages,between8and22Vinincrementsof2V,werestudied.
ThemassspectrometerwascoupledtoananoAcquity-UPLCsystem(Waters,Manchester,UK).
Thechromato-graphicseparationofthepeptideswasperformedonaC18reversed-phaseionKeysystem(BEHC18,130,1.
7μm,150μm*50mm).
Thesolventsusedwere0.
1%FAinwater(solventA)and0.
1%FAinACN(solventB).
ThesampleAnassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3465injectionvolumewas2μLusingafullloopinjection.
Adualpumpreverseflushtrappingconfigurationwasapplied.
Thesamplewasloadedonatrappingcolumn(Symmetry300C18,5μm,300μm*50mm)andanisocraticflowof1%Bover2minwithaflowrateof10μlmin1wasapplied.
Thetrap-pingcolumnwasthenbackflushedinlinewiththeionKeyataflowof2μLmin1utilisingthefollowinggradient:1–5%Bover0.
5min,5–33%Bover15.
5min,33–99%Bover2min,99%Bfor3.
5min,99–1%Bin1minandfinally1%Bfor4.
5mintoallowthecolumntore-equilibrate.
Twotothreetransitionsforeachpeptide,preferablywheretheselectedm/zoftheproductwasgreaterthanthatoftheprecursor,wereincludedinthefinalSRMprotocol.
Thedwelltimewasfixedat50millisecondsforeachmeasuredtransi-tion.
TheSRMtransitionswereseparatedintodifferenttimedeventsresultinginaminimumof12pointsperchromato-graphicpeakforeachtransitionmonitored.
Thedataacquisitionandthedataelaborationwereper-formedusingMassLynxandTargetLynxsoftwarefromWaters.
Thechromatographicpeakswereintegratedautomat-ically,andapeakarealistwasgeneratedforeachpeptidefromeachsample.
Foreachnaturalandisotopicallylabelledpep-tidepair,thepeakarearationatural/labelledwascalculated.
Forpeptideswherenoisotopicallylabelledanalogpeptidewasavailable,auniversalisotopicstandardpeptidewasused.
TheuniversalinternalstandardpeptidechosenwasFALPQYLK*fromαS2-caseinasitelutedinthemiddleofthegradient.
Asystemsuitabilitytest(SST),includinganMS1scanandaproductionscan,wasperformedbeforeeverysetofanalysistoverifythatthemethodperformswithinitsspecifications(retentiontime,m/z).
ProteinquantificationTheRCDC(reducingagentanddetergentcompatible)pro-teinassay(BioRad)wasusedtodeterminethetotalproteinconcentrationinthesampleextracts.
Blanksamples,carriedthroughallstagesofthemanipulation,wereusedasblanksorzeroreferences.
Screening:extractionanddigestionFivebufferingsystems—ammoniumbicarbonate(AMBIC)[9,10],tris-buffer(TRIS)[6–8],phosphatebuffersaline(PBS)[24],triethylammoniumbicarbonatebuffer(TEAB)[12],boratebuffersaline(BBS)[25]—twochaotropicagents[26,27]—ureaandguanidine—andtworeducingagents—dithiothreitol(DTT)andmercaptoethanol(BME)—werescreened.
SeparatebuffersolutionsofAMBIC,TRIS,PBS,TEABandBBSwerepreparedusingMilliQwater(18MΩcm1),at50mmolL1concentrationandpH8.
0.
Thechaotropicagents,ureaandguanidine,wereaddedtoseparatealiquotsofthebuffersolutions,resultinginafinalconcentrationof5molL1ofeach.
ThereducingagentswereaddedtoseparatealiquotsoftheureaAMBICandureaBBSbuffersolutionssothatDTTwas50mmolL1andtheBMEwas2%(v/v).
Agraphicalrepresentationofthescreeningap-proachbuffersolutionsisprovidedinFig.
S1(seeESM).
Atotalof35differentextractioncombinationswereinvestigatedintriplicate.
Onegramofthegroundbiscuitwassolubilisedintheex-tractionbuffer(1:15,w:w),mixedvigorouslyfor1minandsonicatedinacoldbathfor5min.
Theextractionwascarriedoutfor6hat22°C.
Thesolutionwascentrifugedat9500rpmfor45minat22°Candthesupernatantwascollected.
Theextractedproteinswereprecipitatedwithfourvolumesof80%coldacetoneovernightat27°C.
Thesolutionswerecentri-fugedat9500rpmfor30minat4°Candtheacetonewasdiscarded.
Theprecipitatewaswashedtwicewith3mLofcoldacetoneandprecipitatedat27°Cfor1heachtime.
Theacetonewasevaporatedusinganitrogenpurgeatroomtemperature;theprecipitateswereresuspendedin2.
5mLof10mmolL1DTT,3molL1urea,50mmolL1AmBicatpH8.
2.
Twohundredfiftymicrolitresoftheproteinextractweredilutedwith25mmolL1AmBictoyieldafinalcon-centrationof1molL1urea,whichisbelowtheacceptedthresholdfortrypsindigestion.
Thehorsemyoglobin(SIGMAM1882)wasusedasaqualitycontrolofthetrypsindigestionacrosstheextractionbuffersstudied.
Onehundredmicrolitresofstocksolution(14.
2nmol)wereaddedtoeachextractandtheproteinswerereduced(10mmolL1DTT,1hat37°C)andalkylated(50mmolL1IAA,0.
5hat22°Cinthedark).
Trypsinwasaddedatanenzymetosubstrateratioof1:50andthedigestionwascarriedoutfor16hat37°C,whileapplyingaconstantgentleshaking.
Thedigestionwasstoppedbyaddingformicacidtoav/vof0.
1%.
Experimentaldesigns:extractionanddigestionThedevelopedmethodusedtheDesignofExperimentstoorganiseandsuggesttheexperimentsinconjunctionwitharesponsesurfacemethodologyforthefurtheranalysisoftheresultsobtained.
TheUnscramblerXsoftware(v10.
3,CAMO,Trondheim,Norway)wasusedforDoEandfurtherdatatreatment.
Inbothoptimisations,extractionanddigestion,thesameprocedurewasfollowed,wherebyapreliminaryfullfactorialdesign(FFD)wasperformedtoidentifywhichofthestudiedparametersorvariableshadasignificanteffectontheextrac-tionanddigestionprocesses.
TheFFDwasfollowedbyacentralcompositedesign(CCD),inwhichthestatisticallysignificantparametersfromtheFFDwereoptimised.
TheoptimisedvariablesfortheextractionanddigestioncanbeobservedinTables1and2,respectively.
3466NitrideC.
etal.
PreparationofpeptidesforSRManalysesTrifluroaceticacidwasaddedtothedigeststoafinalv/vof0.
1%.
Isotopicallylabelledpeptideswereaddedtothesolu-tionatmassfractionscorrespondingto50mgkg1ofthetotalproteincontentofthetargetallergenicfoodingredient.
Allthedigestsweredilutedtentimestoyieldafinalconcentrationoforganicsolventbelow1%toenableoptimalisolationontheSPEcartridge.
PeptideswereisolatedonHyperSepC18Cartridges(200mg)(ThermoScientific,Biopolymers,Ulm,Germany).
Thecolumnswereactivatedwithmethanol(5col-umnvolumes(CV))andequilibratedwith0.
1%TFA,5%ACNinwater(5CV).
Afterloadingthesample,thecolumnwaswashedwith0.
1%FA,5%ACNinwater(10CV);elutedin0.
1%FA,60%ACNinwater(10CV).
Peptideswerecon-centratedusinganitrogenevaporatoratambienttemperatureandthendilutedwith0.
1%FAinMilliQwateruptoafinalvolumeof1mL.
Sincetheisotopicallylabelledpeptideshaveidenticalchemicalandphysicalbehaviourtotheallergenmarkerpeptides,anypossiblelossofapeptidemarkerduetotheanalyticalprocedurespostdigestionwillbeaccountedfor,astheratiosofnaturaltolabelledpeptidepeakareasweremonitored.
Allsolutionswerespikedwithastandardpeptide,Glu-1-FibrinopeptideB(Glu-Fib),ataconcentrationof50ngmL1beforethemassspectrometryanalysis.
ThiswasusedpurelytoassessthestabilityoftheMSplatformovertheanalysistimescale.
RateofdigestionProteinswereextractedanddigestedusingtheoptimisedpro-tocols.
Thepeptidereleaseratewasevaluated,bymonitoringthenaturaltolabelledpeptideratiosusingtheSRMmethod,overarangeof36h,duringwhich14samplepointswerecollectedat0,0.
5,1,2,3,4,6,8,10,12,22,26,30and34h.
Twoenzyme-to-substrateratioswereconsidered:1:50and1:100,wherethetrypsinwasaddedonlyattimezero,thebeginningoftheincubation.
Themultipleadditionoftheen-zymewasalsostudied.
1:50wasaddedattimezeroandafter22hofdigestion—toafinal1:25(E:S)ratio;1:100wasaddedattimezero,after6,12and22hofdigestion—toafinal1:25(E:S)ratio.
NesteddesignTherepeatabilityofthefinalmethodwasassessedusingatwo-factorfullynestedexperimentdesign.
Threeseparateex-tractsofthe5000mgofthetotalingredientprotein/kggroundbiscuitswereperformedonthesameday.
Eachextractwasdigestedintriplicate.
Fouradditionsoftheenzyme(1:100,E:S)wereperformedevery3hoverthedigestiontimeof12h.
Threereplicatemeasurementswereperformedperdigestedsample.
ResultsanddiscussionThespecificpurposeofthisstudywastoinvestigatetheinflu-enceoftheextractionanddigestionconditionsontheSRMsignalresponseforpeptidesselectedassuitablemarkersforthequantificationofBtotal^eggandmilkprotein.
Thefinalaimwasthedevelopmentofamultiplexedquantitativemeth-odofanalysis.
Theextractionanddigestionprotocolswereoptimisedforbakedbiscuitsandtherepeatabilityoftheex-tractionanddigestionwerestudiedviaatwo-waynestedANOVAstudydesign.
TheSRMmethoddevelopmentconsistedoffourmainsteps:(1)selectionofallergenicmarkerproteinsfromeggandmilk;(2)selectionofproteotypicpeptidesbyinsilicodigestionoftheproteins;(3)methoddevelopmentbyhydro-lysingthestandardproteinswithtrypsin;(4)methodrefine-mentbyanalysinghydrolysedcrudeproteinsextractedfromthebiscuits.
Theproteinmarkersforcow'smilk(Bosdomesticus)wereasfollows:α-S1-casein,α-S2-casein,β-casein,κ-caseinandβ-lactoglobulin.
Theproteinmarkersforhen'seggwhite(Gallusgallus)wereasfollows:ovalbuminandlysozyme.
Table2ExperimentalfactorsandevaluatedlevelsfortheFFDandCCDofthedigestionExperimentalfactorsFFDCCD0+0+++Time(h)258911.
2513.
515.
7518AmBic(mM)25507506.
2512.
518.
7525DTT(mM)25507506.
2512.
518.
7525ACN(%*)0510DMSO(%*)051006.
2512.
518.
7525*FinalvolumeTable1ExperimentalfactorsandevaluatedlevelsfortheFFDandCCDoftheextractionExperimentalfactorsFFDCCD0+0+++Temperature(°C)722370271214pH6.
57.
58.
5Time(h)369AmBic(mM)255075DTT(mM)255075Urea(M)3692.
83.
75.
87.
98.
7Anassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3467Twenty-sevenpeptideswereselectedascandidateproteotypicpeptidesaccordingtoconventionalcriteria[21].
Thespecificityanduniquenessofthepeptideswereestablishedbycomparingthepeptidesequencesagainsttheonlineproteindatabases.
Onlyfourandtwopeptideswerefoundtobeuniqueforcow'smilkandhen'segg,respectively(ESMTableS1).
However,whilemostofthecandidatepep-tideswereuniqueformilkandegg,theywerenotspecies-specific.
Twoofthetargetpeptidesequencesfrommilkpro-teinsandonefromeggproteinswerefoundinothernon-milk-andegg-derivedproteinsbutwerenotexcludedatthisstageofthescreening.
Theoptimisationofthechromatographyandthemassspectrometryoperatingconditionswasperformedusingtrypsinisedstandardproteins.
ThefinalSRMcomprisedthem/zoftheprecursor(Q1)andthem/zoftwoorthreeproductiontransitions(Q3)(ESMFig.
S2).
TheoptimalinstrumentalconditionsarelistedinTableS2(seeESM).
Thecrudeextract(6molL1urea,50mmolL1Ambic,pH8)ofthebiscuitproteinswashydrolysed(1:50,E:S)andthereleasedpeptideswereanalysedbySRM.
Theselectedpeptideswererefinedbasedonthesignalintensityandspec-ificityofeachtransition(interference).
ThepeptidesAMKPWIQPKandITVDDK-αS2;TPEVDDEALEKandIDALNENK-βLg;EAMAPKandEMPFPK-βCNwerenotincludedinthefinalmethodbecausethesignalintensitywasextremelylowandthechromatographicresolutionwaspoor.
Theinfluenceofdifferentbuffercompositionandcondi-tionswasstudiedinordertomaximiseproteinextraction.
Preliminaryscreeningoftheextractionconditionswasper-formedusingfivebuffers,twochaotropicagents(guanidinehydrochlorideandurea)andtworeductants(DTTandBME).
Theextractionbufferswereselectedaccordingtothosecom-monlyreportedintheliterature.
TheextractionperformancewasevaluatedattheproteinlevelbyRCDC.
Changingthebuffersdidnotrevealanymajorimpactonproteinextractability(ESMFig.
S3,panela).
Byaddingthechaotropicagentalone,theextractabilityincreasedbyafactorof7,whileincombinationwiththereductant,itfurtherincreasedbyafactorof3(ESMFig.
S3,panelaandc).
Ureawasthebestperformingchaotropicagentandincreasedtheyieldofextractionsignificantlybydenaturingproteins.
TheRCDCassayonlyprovidesinformationrelatingtothetotalproteinconcentration.
Itdoesnotdeliverdetailedinfor-mationaboutspecificproteins.
TheinformationatthepeptidelevelwasacquiredbySRManalysisofthetrypsinisedextract.
Aninfluenceoftheextractionbufferontrypsindigestionef-ficiencycannotbeexcluded,evenaftertheprecipitationandclean-upoftheproteins.
TheAmBic,theTEABandtheBBSweretheextractionbuffersthatshowedthegreatestcompati-bilitywithtrypsindigestion.
Thecombineduseofachaotropicandreducingagentincreasedthepeakareaofthepeptidesfromβ-lactoglobulinbyafactorof10,forlysozymebyafactorof13andforovalbuminbyafactorof4(ESMFigs.
S4toS7).
Theobservedincreasedependedonthesidechaininteractions(i.
e.
disulfidebridges)andthelocationofthepeptideintheproteinsequence.
Similarincreasesinsignal,whenusingthereducingreagents,werenotobservedforthealphaandbetacaseinproteins(ESMFig.
S6).
Thereductantsbreaktheglutennetworkhelpingthereleaseoftheproteinsandalsobreakthedisulfidebondsmakingtheproteinmoreaccessibletotheenzyme.
Therefore,itisnotsurprisingthatthereductantsarenotsignificantlyimprovingextractionanddigestionforthealphaandbetacaseinssincetheydonotcontaincysteinedisulfidebridges.
Incontrast,theκcaseindoescontainadisulfidebondandexperienceanincreaseinextractionontheinclusionofthereducingreagents.
Theper-formancebetweenthetworeducingagentsstudiedwasnotsignificantlydifferentformostpeptidesusingattest(P=95%)(ESMFigs.
S6-S7).
Therefore,DDTwasselectedforthefurtherstudiesasthiswasmorecompatiblewithcurrentworkingpractices.
Thescreeningevidencesuggestedthattherewaslittleob-servabledifferencebetweenthebackgroundbufferstudiedwhenlookingatthetotalproteincontent(ESMFig.
S3);theuseofchaotropicagentswasbeneficial,withurea,resultinginagreaterextractionefficiencythanguanidineformostpro-teins;theuseofareducingreagentwasbeneficialespeciallyforproteinswithdisulphidebridges,withlittleobservabledifferencebetweenDDTandBMR.
Basedonthisscreeningevidence,AmBic,ureaandDTTwerechosenasthebestperformingcomponentsoftheextractionbuffer.
Moreover,thesecomponentsareESI-MS-compatible.
Consequently,thesampledigestsolutioncanbedirectlyinjectedintotheLCMSforSRManalysis,minimisingthesamplehandlingwhichcanleadtopeptideloss.
Asmentionedpreviously,atwo-stageDoEwasusedtooptimisetheextractionanddigestionprotocolswiththeaimofcreatingasinglemulti-allergenmethod.
AFFDwasemployedtostudythemajorinfluencingvariablesoftheex-tractionanddigestionsteps.
Subsequently,aCCDwasusedtooptimisethestatisticallysignificantvariablesofeachofthesteps.
Followingtheoptimisationoftheexperimentalconditionsforscreening(amountofsample,choiceofextractionbuffer,etc.
),theoptimisationoftheextractionandthedigestionstepswereachievedinsequentialexperiments.
Inthefirststep,theextractionwasoptimisedbyfixingthedigestionparameters,i.
e.
agenericdigestionmethodwasusedwhilevaryingtheextractionconditions.
Inthesecondstep,thedigestionwasoptimisedusingthefixedoptimalextractionconditionsobservedpreviously.
IntheFFDoftheextraction,sixvariables(pH,temperature,time,AmBicconcentration,ureaconcentrationandDTTconcentration)werestudiedattwolevels.
ThevariablesstudiedandtheirrespectivelevelsaresummarisedinTable1.
Inaddition,acentralpointwas3468NitrideC.
etal.
includedinthedesignatwhichallfactorswerekeptatthemeanvalueofthetwolevels.
Thiswasusedtoestimatetheanalyticalprecisionoftheexperiments.
Thirty-twoexperi-mentswereperformedinarandomisedorder.
Thetotalpro-teinextractionefficiencywasdeterminedpriortotrypsindigestionusingtheRCDC,whiletheprotein-specificin-formationwasobtainedbymonitoringthepeakarearatiosoftheindividualpeptidesafterdigestion.
Generally,thetotalproteinyieldofextractionistheonlyresponsethatisoptimisedinproteinextractionexperiments,butinthispar-ticularcase,wherethegoalistoquantifypeptidesasasur-rogatefortheintactprotein,thestudyoftheindividualpep-tidepeakareasbecomesessential.
Accordingtoboththemultilinearregression(MLR)andthepartialleastsquare(PLS)models,constructedusingUnscramblersoftware,onlythetemperatureandtheureaconcentrationwerestatisticallysignificant(plevel0.
05).
Thecon-centrationofDTT(mmolL1)didnotsignificantlyaffectthepeakarea;therefore,itwassetat5mmolL1fortheoptimisedFig.
3Profileofmarkerpeptidesreleasedunderoptimiseddigestionconditionsover35hofdigestion.
Theadditionoftrypsinwasstaticandmultiple.
Staticadditionoftrypsinwasperformedonlyattime0hattwoE:Sratios—1:50and1:100.
Multipleadditionsoftrypsinwereperformedatequidistantintervalsoftimetoafinal1:25(E:S)ratio—1:50(E:S)ratiowasaddedattime0andtime22h;1:100(E:S)ratiowasaddedattime0,6,12and22h.
Errorbarsrepresentthestandarddeviationof3replicatemeasurements.
aαS1CN-YLGYLEQLLR.
bαS2CN-FALPQYLK.
cOva-LYAEER.
dLys-FESNFNTQATNRAnassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3471Table3Overallrelativevariancefortherepeatanalysisofmilk-andegg-derivedpeptides.
Contributionsexpressedas%relativefromextraction,digestionandLCMSpeakarearatiomeasurementsProteinPeptideSourceofvariabilitySSMSTotalCV%Variance%oftotalαS1YLGYLEQLLR*Extraction0.
1460.
073514.
0Digestion0.
2800.
04767.
8Ratiomeasurement0.
0690.
00418.
2FFVAPFPEVFGKExtraction2.
29666E+121.
14833E+1210.
30.
8Digestion6.
6312E+121.
1052E+1241.
5Ratiomeasurement6.
30122E+123.
50068E+1157.
7αS2VIPYVRExtraction65,911,263,31532,955,631,6586.
42.
0Digestion1.
22099E+1120,349,773,8380.
0Ratiomeasurement9.
14193E+111.
01577E+1198.
0FALPQYLK*Extraction0.
5460.
2735.
60.
0Digestion1.
6940.
28275.
4Ratiomeasurement0.
3560.
04024.
6NAVPITPTLNRExtraction6.
12179E+113.
0609E+11450.
4Digestion9.
24282E+111.
54047E+110.
0Ratiomeasurement6.
32202E+137.
02446E+1299.
6ALNEINQFYQKExtraction21,969,405,62810,984,702,81410.
90.
0Digestion88,508,367,12914,751,394,52242.
0Ratiomeasurement83,726,462,7914,651,470,15558.
0βCNVLPVPQKExtraction3.
62137E+131.
81068E+134017.
7Digestion6.
57953E+131.
09659E+1381.
4Ratiomeasurement6.
92243E+1138,457,965,1450.
9AVPYPQRExtraction6.
2232E+123.
1116E+1237.
20.
0Digestion2.
13986E+133.
56643E+1292.
8Ratiomeasurement1.
61913E+1289,951,794,8437.
2κCNYIPIQYVLSRExtraction1.
14481E+125.
72405E+116.
42.
8Digestion2.
99726E+124.
99543E+1138.
5Ratiomeasurement3.
03071E+121.
68373E+1158.
7βLGGLDIQKExtraction9,374,383,9914,687,191,99623.
10.
0Digestion29,866,046,8984,977,674,48379.
0Ratiomeasurement5,269,055,283585,450,587.
121.
0IPAVFKExtraction76,749,448,70338,374,724,3528.
74.
5Digestion1.
81446E+1130,241,083,29027.
4Ratiomeasurement2.
46673E+1113,704,048,15168.
1ALPMHIRExtraction4,533,019,3262,266,509,6639.
60.
0Digestion34,287,299,4115,714,549,9022.
0Ratiomeasurement97,025,982,7795,390,332,37798.
0OvaLYAEER*Extraction0.
0398860190.
0199430097.
37.
0Digestion0.
0913919290.
01523198855.
0Ratiomeasurement0.
0513012310.
00285006838.
0GLWEKExtraction3.
13527E+111.
56763E+1111.
80.
1Digestion9.
37364E+111.
56227E+1170.
8Ratiomeasurement3.
38734E+1118,818,572,80629.
1HIATNAVLFFGRExtraction4.
15205E+112.
07602E+118.
936.
7Digestion2.
9878E+1149,796,694,36320.
4Ratiomeasurement3.
68934E+1120,496,340,99842.
9LysGTDVQAWIRExtraction53,321,662,87926,660,831,4407.
433.
6Digestion49,068,398,6418,178,066,44033.
8Ratiomeasurement35,821,898,5721,990,105,47632.
63472NitrideC.
etal.
method.
TheresponsesurfacesdemonstratingtheeffectsofthesignificantfactorsonthepeakarearatioareshowninFig.
2andESMFig.
S11.
Forallthepeptidesstudied,amax-imumwasobservedafter18hofdigestion;thus,thetimeneededtocarryouttheextractionwasfixedat18h.
Insum-mary,theoptimisedconditionsusedforthedigestionwerefixedat25mmolL1AmBic,5mmolL1DTT,10%DMSOwithadigestiontimeof18h.
Apeptideactingasareliablequantifiermustbefullyre-leasedfromitsconstituentprotein.
Thepeptidehastobeinastableforminthefinaldigestionsolutionsuchthatitsmolaramountisrepresentativeofthemolesofproteinpresentintheextract.
Anydegradationorunreleasedpeptidewillresultintheincorrectdeterminationoftheconcentrationofproteininitiallypresentinthesample.
Toinvestigatetherateofpro-teindigestion,atimecourseexperimentwasperformed.
Severalaliquotsoftheextractweredigested,andateachtimepointofthedigestion,analiquotwastakenforanalysis.
TheeffectsofenzymeconcentrationanddigestiontimeforfourpeptidesfromthedifferenteggandmilkproteinsareshowninFig.
3andESMFig.
S12.
Itcanbeclearlyobservedthatforpeptidesderivedfromtheeggproteins,lysozymeandovalbu-min,adigestiontimeinexcessof16hwasrequiredtoreachaplateauforthepeptideconcentration(Fig.
3andESMFig.
S12).
Thisplateaueffectcanbeindicativeofthetotaldiges-tionoftheproteinpresent.
Hence,theamount-of-substanceconcentrationofthepeptidewouldrepresenttheconcentrationoftheinitialprotein.
Theeffectofdigestiontimeonthemea-suredmilkpeptidesismarkedlydifferentfromthoseofegg(Fig.
3andESMFig.
S12).
TheαS1caseinpeptidesarecompletelyreleasedinarelativelyshortdigestiontimeandcanbeobservedtodecayafter16h.
Thismayindicatethedegradationand/orthechemicalmodificationofthepeptideonceitisreleasedfromtheparentprotein.
Thetwoβ-caseinpeptidesandoneβ-lactoglobulinpeptidewereadverselyaffectedbybothenzymeconcentrationandtime(ESMFig.
S12,panelf,i,j).
Formostproteins,adigestiontimeof12–20hwasoptimal,butagain,acarefulselectionofthepeptidesthatarestableandfullyreleasedfromtheirconstituentproteinisessentialforaccuratequantification.
Theoveralleffectandtherelativeuncertaintycontributionfromtheextractionanddigestionstepswereinvestigatedusingasimpletwo-factorfullnesteddesign.
Forthis,threeseparateproteinextractswereeachdigestedthreetimesandeachdigestwasanalysedintriplicate.
ThedigestswereanalysedinarandomorderontheLCMSandstatisticalout-lierswereremovedbeforeperformingtheANOVA.
Anad-vantageofperformingsuchanexperimentisthattheindivid-ualcontributionsoftheextraction,digestionandLCMSanal-ysistotheoverallmeasurementvariabilitycanbequantified.
Thisisinformativewhendesigningthemethodvalidationplanasthefactorsinfluencingtheoveralluncertaintycouldbebettercontrolledand/orrepeatedtoassessoverallprecision.
Theoverallprecisionofthemeasurementandtheindividualcontributionsfromtherepeatabilityofextraction,digestionandmeasurementsforthemilkandeggwhitepeptidesarereportedinTable3.
Asexpected,themeasurementprecisionfortheβ-caseinandtheβ-lactoglobulinpeptides,whichshowedrapiddegradationoverthedigestiontime,waspoor.
Wherepresent,theadditionofisotope-labelledpeptideinter-nalstandardsimprovedtheprecisionofpeptidemeasurementsgreatly.
Theresultsobtainedfromthesestudiessuggestthat,fortherelativelyhigh-levelallergencontaminationcookiestudied,measurementprecisionsaslowas5%relativestandarddevi-ation(RSD)couldbeachieved.
WhereRSDsincrease,nor-mallyduetothelackofanisotopicinternalstandardoralowmeasuredsignalintensity,digestionwasthegreatestsourceofvariationinthemajorityofcases.
ThestudysuggeststhatwhileMSiscapableofthemultiplexeddetectionofmanyTable3(continued)ProteinPeptideSourceofvariabilitySSMSTotalCV%Variance%oftotalFESNFNTQATNR*Extraction0.
0062004140.
0031002076.
50.
0Digestion0.
0229790510.
00382984272.
5Ratiomeasurement0.
0077242250.
00042912427.
5HGLDNYRExtraction39,193,790.
6319,596,895.
3228.
40.
0Digestion856,334,672.
1142,722,445.
386.
4Ratiomeasurement127,696,469.
57,094,248.
30613.
6MyoExtraction4.
88984E+122.
44492E+125.
324.
8Digestion4.
70281E+127.
83802E+1115.
1Ratiomeasurement8.
04677E+124.
47043E+1160.
1Glu-FibExtraction1.
2567E+126.
28352E+11149.
4Digestion1.
43487E+122.
39144E+110.
0Ratiomeasurement3.
42278E+122.
0134E+1190.
6*Peptidesforwhichanisotope-labelledpeptidestandardwasavailableAnassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3473peptides,thatauniversalbufferwhichextractsallproteinsequallyisunlikely.
Therefore,carefulselectionofquantitativemarkerproteinsandrobustpeptideswillberequired.
ConclusionsTheabilityofmodernLCMStosensitivelyandselectivelydetectthepresenceofallergenpeptideshasresultedinvariousmultiplexedmethodsforthesimultaneousdetectionoffoodallergens.
However,whenquantifyingallergens,allpartsoftheanalyticalprocedureneedtobeconsidered,notjustthefinalMSmethod.
WhileMScouldquantifymanypeptidesinonerun,theabilitytocompletelyextractanddigestthetargetmarkerproteinsisessentialifaccurateresultsaretobeobtain-ed.
Thenumberofallergensthatcouldbeaccommodatedreasonablyinamultiplexedassaydependsonthefollowing:thephysicochemicalpropertiesandstructureoftheproteins;thecomplexityofthefoodmatrixanditsintermolecularinter-actions;andthefoodprocessingconditions(doughformation,baking,extrusion,andcooking/boiling).
Theoptimisationofproteinextractionanddigestionisfun-damentaltoprovidereliablequantificationifproteotypicpep-tidesaretobeusedfordeterminingtheamountofthealler-genicfoodproteinpresent.
Currentmethodscommonlyuseisotopicallylabelledpeptidesasinternalstandard.
Thisap-proachisusuallyreducingthemeasurementbiasandimprov-ingtheprecisionofmeasurementsatthepeptidelevel.
However,differencesinextractionareoftenoverlooked.
ThenumberofmultiplexedMSmethodscurrentlybeingreported,whereasinglecommonbufferisusedfortheextractionofmultipleproteinsfrommanydifferentallergenicfoodsources,suggeststhatMShasalottoofferintheareaoffoodallergenquantification.
However,failuretofullyaddressallaspectsoftheanalyticalproceduremayresultinunreportedbiasbythecurrentlyusedapproaches.
Theevidencepresentedinthisstudyshowsthatthequantitativeremovalofproteinsfromheat-treatedmatricesisstillamajorchallengeformultiplexedmethods.
Whilstmultiplexeddetectionispossible,theaccu-ratequantificationofthetotalfoodallergenproteinmassfrac-tionwillnotbefeasiblewithoutauniversalbufferthatextractsanddissolvesthewidevarietyofproteinsderivedfromthedifferentallergenicfoodsources.
Inthisstudy,thedevelopedmethod,specificallydesignedfortheabsolutequantificationofeggandmilkallergensinbiscuits,requiredacompromisedextractionanddigestionpro-tocol.
Nouniversalconditionswerefoundthatmaximisedtheresponseofallthetargetpeptidesfrommilkandegg.
Therefore,notallconstituentproteinscanbeconsideredgoodquantitativemarkersandshouldsimplybeusedasqualitativemarkers.
Knowingthedigestionkineticsdeliveredevidencethatpeptideswererepresentativeoftheproteincontent,there-byprovidingconfidenceinthedeterminedconcentrationforeachprotein.
Withtheinclusionoflabelledpeptideinternalstandards,theprecisionoftheMSpeakarearatiomeasure-mentshadverylittleimpactontheoverallvariabilityoftheanalyticalresult.
Therefore,tightercontrolofthedigestionandextractionproceduresisessentialforachievingimprovedmeasurementuncertaintiesinthefuture.
AcknowledgmentsThisworkwasundertakenaspartoftheEUiFAAMproject:IntegratedApproachestoFoodAllergenandAllergyRiskManagement(GrantAgreementNo.
312147),whichalsoprovidedthecookiematerialanalysedinthisstudy.
CompliancewithethicalstandardsConflictofinterestTheauthorsdeclarethattheyhavenoconflictsofinterest.
OpenAccessThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.
0InternationalLicense(http://creativecommons.
org/licenses/by/4.
0/),whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCommonslicense,andindicateifchangesweremade.
References1.
Regulation(EU)No1169/2011oftheEuropeanParliamentandoftheCouncilof25October2011ontheprovisionoffoodinforma-tiontoconsumers,amendingRegulations(EC)No1924/2006and(EC)No1925/2006oftheEuropeanParliamentandoftheCouncil,an.
In:OJL304,2011.
2.
RemingtonBC,BaumertJL,BlomMW,HoubenGF,TaylorSL,KruizingaAG.
QuantitativeriskassessmentofUKfoodproductscross-contaminatedwithallergens.
ClinTranslAllergy.
2015;5:O10.
https://doi.
org/10.
1186/2045-7022-5-S3-O10.
3.
AllenKJ,RemingtonBC,BaumertJL,CrevelRWR,HoubenGF,Brooke-TaylorS,etal.
Allergenreferencedosesforprecautionarylabeling(VITAL2.
0):clinicalimplications.
JAllergyClinImmunol.
2014;133:156–64.
https://doi.
org/10.
1016/J.
JACI.
2013.
06.
042.
4.
Diaz-AmigoC,PoppingB.
Foodallergens:aregulatory/labellingoverviewincludingtheVITALapproach.
EncyclFoodChem.
2019:615–21.
https://doi.
org/10.
1016/B978-0-08-100596-5.
21798-8.
5.
WalkerMJ,BurnsDT,ElliottCT,GowlandMH,MillsENC.
IsfoodallergenanalysisflawedHealthandsupplychainrisksandaproposedframeworktoaddressurgentanalyticalneeds.
Analyst.
2016;141:24–35.
https://doi.
org/10.
1039/C5AN01457C.
6.
O'ConnorG,HaponiukM,UberthFJointDGSANTandDGJRCWorkshop-HarmonisationofapproachesforinformingEUallergenlabellinglegislation,JRC108259.
7.
HeickJ,FischerM,PppingB.
Firstscreeningmethodforthesimultaneousdetectionofsevenallergensbyliquidchromatogra-phymassspectrometry.
JChromatogrA.
2011;1218:938–43.
https://doi.
org/10.
1016/j.
chroma.
2010.
12.
067.
8.
MonaciL,PilolliR,DeAngelisE,GodulaM,ViscontiA.
Multi-allergendetectioninfoodbymicrohigh-performanceliquidchro-matographycoupledtoadualcelllineariontrapmassspectrometry.
JChromatogrA.
2014;1358:136–44.
https://doi.
org/10.
1016/j.
chroma.
2014.
06.
092.
3474NitrideC.
etal.
9.
ParkerCH,KhudaSE,PereiraM,RossMM,FuTJ,FanX,etal.
Multi-allergenquantitationandtheimpactofthermaltreatmentinindustry-processedbakedgoodsbyELISAandliquidchromatography-tandemmassspectrometry.
JAgricFoodChem.
2015;63:10669–80.
https://doi.
org/10.
1021/acs.
jafc.
5b04287.
10.
CristinaL,ElenaA,DavideC,MarziaG,LuciaD,CristianoG,etal.
Validationofamassspectrometry-basedmethodformilktracesdetectioninbakedfood.
FoodChem.
2016;199:119–27.
https://doi.
org/10.
1016/j.
foodchem.
2015.
11.
130.
11.
LutterP,ParisodV,WeymuthH.
Developmentandvalidationofamethodforthequantificationofmilkproteinsinfoodproductsbasedonliquidchromatographywithmassspectrometricdetection.
JAOACInt.
2011;94:1043–59.
12.
MonaciL,LositoI,DeAngelisE,PilolliR,ViscontiA.
Multi-allergenquantificationoffining-relatedeggandmilkproteinsinwhitewinesbyhigh-resolutionmassspectrometry.
RapidCommunMassSpectrom.
2013;27:2009–18.
https://doi.
org/10.
1002/rcm.
6662.
13.
CryarA,PritchardC,BurkittW,WalkerM,O'ConnorG,BurnsDT,etal.
Towardsabsolutequantificationofallergenicproteinsinfood-lysozymeinwineasamodelsystemformetrologicallytraceablemassspectrometricmethodsandcertifiedreferencemate-rials.
JAOACInt.
2013;96:1350–61.
https://doi.
org/10.
5740/jaoacint.
12-438.
14.
TolinS,PasiniG,CurioniA,ArrigoniG,MasiA,MainenteF,etal.
Massspectrometrydetectionofeggproteinsinredwinestreatedwitheggwhite.
FoodControl.
2012;23:87–94.
https://doi.
org/10.
1016/j.
foodcont.
2011.
06.
016.
15.
TolinS,PasiniG,SimonatoB,MainenteF,ArrigoniG.
AnalysisofcommercialwinesbyLC-MS/MSrevealsthepresenceofresidualmilkandeggwhiteallergens.
FoodControl.
2012;28:321–6.
https://doi.
org/10.
1016/j.
foodcont.
2012.
05.
015.
16.
GrovesK,CryarA,WalkerM,QuagliaM.
Assessmentofrecoveryofmilkproteinallergensfromprocessedfoodformassspectrome-tryquantification.
JAOACInt.
2018;101:152–61.
https://doi.
org/10.
5740/jaoacint.
17-0214.
17.
LositoI,IntronaB,MonaciL,MinellaS,PalmisanoF.
DevelopmentofamethodforthequantificationofcaseinatetracesinItaliancommercialwhitewinesbasedonliquidchromatography–electrosprayionization–iontrap–massspectrom-etry.
JAgricFoodChem.
2013;61:12436–44.
https://doi.
org/10.
1021/jf4034909.
18.
ArseneCG,OhlendorfR,BurkittW,PritchardC,HenrionA,O'ConnorG,etal.
Proteinquantificationbyisotopedilutionmassspectrometryofproteolyticfragments:cleavagerateandaccuracy.
AnalChem.
2008;80.
https://doi.
org/10.
1021/ac7024738.
19.
BurkittWI,PritchardC,ArseneC,HenrionA,BunkD,O'ConnorG.
TowardSystèmeInternationald'Unité-traceableproteinquanti-fication:fromaminoacidstoproteins.
AnalBiochem.
2008;376.
https://doi.
org/10.
1016/j.
ab.
2008.
02.
010.
20.
PritchardC,GrovesKJ,BiesenbruchS,O'ConnorG,AshcroftAE,ArseneC,etal.
Quantificationofhumangrowthhormoneinserumwithalabeledproteinasaninternalstandard:essentialconsider-ations.
AnalChem.
2014;86:6525–32.
https://doi.
org/10.
1021/ac501032q.
21.
NewsomeGA,SchollPF.
QuantificationofallergenicbovineMilkαS1-caseininbakedgoodsusinganintact15N-labeledproteininternalstandard.
JAgricFoodChem.
2013;61:5659–68.
https://doi.
org/10.
1021/jf3015238.
22.
SteinhoffM,FischerM,Paschke-KratzinA.
Comparisonofextrac-tionconditionsformilkandhen'seggallergens.
FoodAdditContamPartA.
2011;28:373–83.
https://doi.
org/10.
1080/19440049.
2010.
545957.
23.
JohnsonPEP,BessantC,ValeriaC,GiosafattoL,BaumgartnerS,AldickT,etal.
Currentperspectivesandrecommendationsforthedevelopmentofmassspectrometrymethodsforthedeterminationofallergensinfoods.
JAOACInt.
2011;94:1026–33.
24.
GomaaA,BoyeJI.
Impactofthermalprocessingtimeandcookiesizeonthedetectionofcasein,egg,glutenandsoyallergensinfood.
FoodResInt.
2013;52:483–9.
https://doi.
org/10.
1016/j.
foodres.
2013.
01.
019.
25.
ShubhadaS,SundaramPV.
Borateion-assistedstabilizationofbeta-galactosidasefromAspergillusoryzaebypolyhydroxycom-poundsinwater-miscibleorganicsolvents.
EnzymMicrobTechnol.
1993;15:881–6.
26.
GreeneRF,PaceCN.
Ureaandguanidinehydrochloridedenatur-ationofribonuclease,lysozyme,α-chymotrypsin,andβ-Lactoglobulin.
JBiolChem.
1974;249:5388–93.
27.
HerveyWJ,StraderMB,HurstGB.
Comparisonofdigestionpro-tocolsformicrogramquantitiesofenrichedproteinsamples.
JProteomeRes.
2007;6:3054–61.
https://doi.
org/10.
1021/pr070159b.
28.
KolliparaL,ZahediRP.
Proteincarbamylation:invivomodifica-tionorinvitroartefactProteomics.
2013;13:941–4.
https://doi.
org/10.
1002/pmic.
201200452.
29.
SunS,ZhouJ-Y,YangW,ZhangH.
Inhibitionofproteincarbamylationinureasolutionusingammonium-containingbuffers.
AnalBiochem.
2014;446:76–81.
https://doi.
org/10.
1016/j.
ab.
2013.
10.
024.
30.
StraderMB,TabbDL,HerveyWJ,PanC,HurstGB.
Efficientandspecifictrypsindigestionofmicrogramtonanogramquantitiesofproteinsinorganicaqueoussolventsystems.
AnalChem.
2006;78:125–34.
https://doi.
org/10.
1021/ac051348l.
Publisher'snoteSpringerNatureremainsneutralwithregardtojurisdic-tionalclaimsinpublishedmapsandinstitutionalaffiliations.
Anassessmentoftheimpactofextractionanddigestionprotocolsonmultiplexedtargetedprotein.
.
.
3475
- powder585qqq.com相关文档
- Ambient585qqq.com
- Electrically-isolated
- 基金585qqq.com
- Mutual585qqq.com
- ws-secureconversation-1.3-os
- recommended585qqq.com
raksmart：香港机房服务器实测评数据分享，告诉你raksmart服务器怎么样
raksmart作为一家老牌美国机房总是被很多人问到raksmart香港服务器怎么样、raksmart好不好？其实，这也好理解。香港服务器离大陆最近、理论上是不需要备案的服务器里面速度最快的，被过多关注也就在情理之中了。本着为大家趟雷就是本站的光荣这一理念，拿了一台raksmart的香港独立服务器，简单做个测评，分享下实测的数据，仅供参考！官方网站：https://www.raksmart.com...
PacificRack：洛杉矶KVM月付1.5美元起,1G内存套餐年付12美元起
PacificRack在本月发布了几款特价产品，其中最低款支持月付仅1.5美元，基于KVM架构，洛杉矶机房，PR-M系列。PacificRack简称PR，QN机房旗下站点，主要提供低价VPS主机产品，基于KVM架构，数据中心为自营洛杉矶机房，现在只有PR-M一个系列，分为了2个类别：常规(Elastic Compute Service)和多IP产品(Multi IP Server)。下面列出几款秒...
NameSilo域名优惠码活动
NameSilo是通过之前的感恩节优惠活动中认识到这家注册商的，于是今天早上花了点时间专门了解了NameSilo优惠码和商家的详细信息。该商家只销售域名，他们家的域名销售价格还是中规中矩的，没有像godaddy域名标价和使用优惠之后的价格悬殊很大，而且其特色就是该域名平台提供免费的域名停放、免费隐私保护等功能。namesilo新注册域名价格列表，NameSilo官方网站：www.namesilo....
585qqq.com为你推荐
-
易烊千玺弟弟创魔方世界纪录王俊凯和王源两人和易烊千玺的弟弟玩过吗www.bbb336.comwww.zzfyx.com大家感觉这个网站咋样，给俺看看呀。多提意见哦。哈哈。125xx.com115xx.com是什么意思yinrentangweichentang万艾可正品的作用真的不错吗www.mfav.orgwww.osta.org.cn国家职业资格证书全国联网查询,为什么随便输入什么都可以查，都要验证码关键词分析怎么样分析关键词？铂金血痕“斑斑的血痕”是什么意思？网页源代码网页源代码是什么，具体讲一下？干支论坛干支计时的干支计时蚕食嫩妻养成记的小说