1.1Forewords:powersof10Therearesomequestionstheseniorsofushaveansweredtensoftimeswhenadvisingyoungerresearchersattheirfirstexperienceswithelectrophoresisandproteomics.Oneofthemishowtoremovealbumin

fromplasma/serumandotherbiologicalfluidsbeforeanalysis.Wechallengethatthisquestionisinfactthemostappropriate.Wemaintain,first,thatallmajorproteinsinthosesamplesareatissue;then,thataproper

questionshouldbe,instead,whentoremovethemandwhendonot.Inthefollowing,wearegoingtopresentourpointofview–—whichweknowisnotsharedbythemajorityofthescientificcommunity.Mostofthe

informationwillfocusonhumanspecimensalthoughsimilarconceptsandproceduresapplytoplasma/serumfromlaboratoryandfarmanimalsaswell.

Clinicalbiochemistryhaslongrecognizedthattheconcentrationinbloodoftissuecomponentsincreasesasaresultoftissuedamage.Therecognitionoftissue-specificisoforms,eitherassequencevariantsor

as heteromultimeric assembly variants, has been the basis for discriminating tissue origin. The difference in subcellular derivation (plasma membrane, cytoplasm, mitochondria) has been associated with

nature/severity/durationofthenoxa.Thetypicalintracellularproteinscurrentlyquantitatedinbloodbytheclinicalbiochemistrylaboratoryaremedium-abundanceenzymes,assayedthroughtheircatalyticactivity,and

high-abundancestructuralproteins,identifiedthroughtheirimmunologicalreactivity.

Eitherinnovativeormoreeffectivediseasemarkersshouldbeassharplyaspossiblebothtissue-anddisease-specific.Proteomicsstudieshavedemonstratedthatsomemajorbiologicallimitationsexistinthese

directions.Thethoroughinvestigationaboutthetissuedistributionofgeneproductsindifferentorgansandcelltypescarriedoutunder ‘TheHumanProteinAtlas’project–—whichparticipates in theoverall ‘The

ReviewaArticle

Withorwithoutyou‐—Proteomicswithorwithoutmajorplasma/serumproteins

ElisabettaGianazza1a,⁎

elisabetta.gianazza@unimi.it

IngridMiller2b

LucaPalazzolo1a

ChiaraParravicini1a

IvanoEberini1a

1aDipartimentodiScienzeFarmacologicheeBiomolecolari,UniversitàdegliStudidiMilano,ViaBalzaretti9,I-20133MilanoMilan,Italy

2bInstitutfürMedizinischeBiochemie,VeterinärmedizinischeUniversitätWien,Veterinärplatz1,A-1210Wien,Austria

⁎*Correspondingauthor.

Abstract

Thefirstsectionsofthisreviewcompileanddiscussstrategiesandprotocolsformanagingplasma/serumasasourceofbiomarkersrelevanttohumandisease.Inmanysuchcases,depletionofabundant

protein(s)isacrucialpreliminarysteptotheprocedure;specificconceptualandtechnicalapproaches,however,makeitpossibletoeffectivelyusetothispurposewholeplasma/serum.Thefinalsections

focusinsteadonthecomplexityassociatedwitheachofthemajorserum/plasmaproteinsintermsofboth,multiplemolecularstructures(existenceofanumberofproteinspecies)andofmultiplemolecular

functions(behaviorasmultifunctional/multitasking/moonlightingproteins).Reviewingevidenceintheseandsomerelatedfields(regulationofthesyntheticpatternbyproteinsandnon-proteincompounds

anditsconnectionwithhealthanddisease)promptsthesuggestion/recommendationthatinformationontheabundantcomponentsofplasma/serumproteomeisroutinelyobtainedandprocessed/minedasa

valuablecontributiontothecharacterizationofanynon-physiologicalconditionandtotheunderstandingofitsmechanismsandofitsimplications/sequels.

Keywords:Serumproteins;Depletion;Enrichment;Immunodepletion;Proteinspecies

HumanProteome’endeavor‐—hasdemonstratedthatdifferencesinproteomecompositionamongcelltypesaremoreofaquantitativethanofaqualitativenature.Absolutetissuespecificityhasbeendemonstratedonlyfora

smallpercentageofproteins,themostobviousofwhichhadbeenknownfordecades.Evenmoredisturbingforitspractical,andmoreintriguingforitsbiologicalimplications,isthefindingthat,underconditionsof

cellularstress,thechangesinproteomecompositionhardlydependonthenatureofthestress.

Howdoesthecirculatingconcentrationofcurrentandperspectivediseasemarkerscomparewiththatofotherproteinsinplasma/serum?

Fig.1–—redrawnfrom[1]–—providesanoverviewonthequantitativerelationshipsamongthevariousclassesofproteinswithinplasma/serumproteome.Theoveralldynamicrangefortheirconcentrations

exceeds1010-fold.Therangeassociatedwithproteinssecretedcontinuously,ifatavaryingrate,inblood(e.g.binding/transportproteins,proteaseinhibitors,coagulationfactors,immunoglobulins)iswiderthan105

fold,thatforproteinsdiscontinuouslysecretedinblood(aslong-rangeextracellulareffectorse.g.hormonesandcytokines)isnarrowerthan102-fold.Theremaining>103-foldcorrespondstotheconcentrationrange

forproteinsnottargetedforsecretionbutleakingfromtissueseitherasintactmoleculesorasproteolyticfragments.

Proteinsleakingfromtissueswouldbecomemajorcomponentsofasampleonlyaftertheremovalofhepatocyteandplasmacellsecretionproducts,whichare10‐−3‐–10‐−6timesmoreabundant.Thesubtraction

ofalbuminaloneresultsintheremovalof50%ofthetotalproteins:whileoutstandingperse,suchanachievementisstillinadequatetomeettherequirementfortracecomponentenrichment.Asuitableprotocolusing

thedepletionapproachdemandsamuchmoreextensivecutof(all)majorproteins.Onewaytomeetthisrequirementisthroughtheuseofimmunoaffinityresins;somearemarketed,whichareabletobindhigh-or

medium-abundanceplasma/serumcomponents,resultinginthedepletionofupto99%ofthetotalproteins.

2.2Withoutmajorplasma/serumproteins2.1.2.1Howtoremovemajorplasma/serumproteins2.1.1.2.1.1Removingalbumin

Fromtheabove, removingalbuminalone fromplasma/serum,and fromallbiological fluids thatderive fromplasma/serum (e.g. urineandCSF),definitely falls shortof significantlyenriching the samples in low-abundance

proteins.Still,somespecificindicationsexistforthisprocedure.TheMrofalbuminisclosetothatofmanyotherplasma/serumcomponentssothealreadylowresolutionbymassaffordedby1DEelectrophoresis(SDS-PAGE)isfurther

limitedbythepresenceofalbumin.ThepIofalbumindifferssignificantlyfromthatofotherplasma/serumproteinsoflike(andunlike)Mr:accordingly,in2DEwithIEFunderequilibriumconditions,thespotofalbuminisquitedistinct

fromallothers.However,usingstandardprotocolswithready-madeIPGstripsforthe1stdrunandin-gelsamplereswelling,someofthealbumininitiallyloadedatpH<pIerraticallyfailstofocus.Thisresultsinahorizontalstreakin

thelowpH(anodic)region,whichoverlapsacidicproteinspots,blurringtheirpatternandconfusingtheirquantitation.Cuttingalbuminconcentrationalsocutssuchinterference.

Fig.1Achartofthecirculatingconcentrationsofdifferentclassesofproteinsfoundinplasma/serum.(Redrawnfrom[1]).

alt-text:Fig.1

Atypicalapproachtoremoveonly/mainlyalbuminfromplasma/serumisthroughdye-ligandpseudoaffinitychromatography[2,3]onimmobilizedtriazinedyes,mostoftenonCibacronBlueF3-GAimmobilizedtogiveblueresins.

Withthepresenceofbothfusedaromaticringsandsulfategroups,thismoleculerecapitulateskeyfeaturesofthemainphysiologicalalbuminligands(freefattyacidsandbilirubin),namelythecoexistenceofhydrophobicandacidic

functions.However,theresemblanceisbroad,andsimilaritymaybedetectedalsowithsuchanunrelatedmetaboliteasNAD(P).Indeed,highaffinityforthebluecolumnsissharedbyawholeclassofproteins–—the NAD/NADP-

dependentenzymes –—that interactwith the immobilizeddye through theirRossman-foldedbindingsites.Morerelevant toplasma/serumproteins, thepropertiesof the immobilized liganddocontribute toboth ionexchangeand

hydrophobic interaction. As a result, the affinity of individual serum proteins for immobilized Cibacron Blue F3-GA changes with ligand density (in different commercial products) as well as with pH and ionic strength of the

chromatographicbuffer,whetherundersteadyordynamicconditions(incubation,flow)[4,5].Moreover, theaffinityofalbumin itself forthe immobilizeddyesextensivelyvariesdependingonthespecies[6],beingveryhighforthe

humanproteinbut from lower tomuch lower formanyof itsanimalhomologs.Accordingly, it isdifficult toquantitatively removealbumin (fromhumanandevenmore fromanimal specimens)withoutcarryingalongother serum

proteins; for instance,undermostconditions, lipoproteins/apolipoproteinsareheavilydepletedaswell.Ahigh-saltbuffer is required to sharpenbindingselectivity;as such,withoutdesalting, the flow-through from theadsorption

cartridge/columnisnotdirectlysuitableforloadinginIEF.Anassessmentofdye-ligandchromatographyinconnectionwithplasmaproteomicsmaybefoundin[7].Inareverseperspectivetodepletion,ahydroxyethylthioethylester

adductedtripeptideproducedbypronasedigestionofalbuminwasusedasquantitativebiomarkerforthediagnosisofhumanexposuretosulfurmustard;sampleenrichmentinhumanserumalbuminbydye-ligandchromatographywas

instrumentaltoimprovetheperformanceofthetest[8].

Mentioningsaltsinthepreviousparagraphremindsalow-tech,old-timeapproachforseparating/removingalbuminfromotherserumproteins,i.e.ammoniumsulfateprecipitation.Albuministheonlymajorproteinsolubleat

50%saturationofthechaotropewhileitsquantitativeprecipitationcallsfor100%saturation.Asexemplifiedbyacomparisonofimmunoglobulinsolubility[9],species-specificitysetstheconditionsforsalting-outplasma/serumproteins.

Indeed,boththedepletionofhigh-abundancecomponents(fromratserum[10])andthefractionalprecipitationof(human)plasmaproteomewithammoniumsulfate(attaining10%,20%,25%,30%,35%,40%,45%,50%,and55%salt

concentrations[11])havebeenrecentlyexploredaspreliminarystepsbefore2DEanalysis.

Anotherchemicaldepletionmethodisbasedonsequentialproteinprecipitationssteps.DTT(50mM)promotestheprecipitationofproteinsrichindisulfidebonds,withtheformationofaviscousprecipitate,mainlycontaining

albumin;theadditionwithsonicationof2volofwaterand4.5volofacetonitriletothesupernatantresultsintheprecipitationofalpha-2-macroglobulin,transferrin,complement3alpha(C3),andimmunoglobulins.Thedepletedsample

inthesecondsupernatantmaybeusedinproteomicsexperiments(Fig.2)afterevaporationofthesolventtodryness[12].

Stilladifferenttypeofprecipitationprocedureisdescribedin[13]:atpH (pH(noItalics))4.2,saltremovalfromserum(human,bovine,orporcinespecimensalike)reproduciblyresultsintheprecipitationofalbuminandother

highabundanceproteins(immunoglobulins,apolipoproteinA-I,transferrin,alpha-2-HS-glycoprotein,hemopexin,vitaminD (vitaminD(noItalics))-bindingprotein,serpinA3‐–5,complementfactorB).

Forsomespecificapplications,excludingalbuminandsomeothermajorplasma/serumproteinseitherfromanalysisorfromdetectionafter2DEmaybeobtainedthroughsimpleproceduresrelyingondifferencesinMr.For

instance, inan investigationon(human)amniotic fluid focusedon low-Mrproteinsandonproteolytic fragments fromhigh-Mrcomponents,weprocessedproteins (concentratedby freeze-drying) throughtwostepsofsize-exclusion

chromatography(gelfiltrationonaSephadexG100column)beforetheelectrophoreticanalysis[14].On(rat)serum,wehavedemonstratedthefeasibilityofacontrolledoverloadingfollowedbystainingtheupperhalfofthe2nddslab

Fig.22DEcomparingthedifferentproteinprofilesofuntreatedhumanserumandofserumdepletedaccordingtothesequentialmethodin[12].Proteinlandmarksareidentifiedineachmap.

alt-text:Fig.2

withCoomassie, the lowerhalfwith silvernitrate, tocompensate for theunevenabundanceofhigherand lowerMrproteins in thesample [15].Anotherway topartitionproteinson thebasis of their size is through (centrifugal)

ultrafiltration(withMWCOat30,000),undersolventconditions(25mMNH4HCO3,pH8.2,20%(v/v)acetonitrile)thathinderprotein-proteininteractions[16].Inthereferredwork,theauthorsdidnotfurtherfractionatethecollected

proteinsbuttreatedthewholelow- MrfiltratewithtrypsinandanalyzedtheresultingpeptidesbySELDI.

Noneoftheaboveproceduresactuallyprovidesforstrictselectivitybetweenalbuminandtheotherplasma/serumcomponentsbecausethefractionationcriteriarelyonpropertiessharedtodifferentextentbymany/allproteins,

i.e.onquantitativeratherthanonqualitativedifferences.Anall-or-noneinteractionmaybeexpectedinsteadwithsuchabiospecificdevice/reagentasanimmunoaffinityresin.Someproductsarecommerciallyavailable(e.g.anagarose

slurrywithimmobilizedanti-albuminantibodies,fromSartorius),othersmaybelaboratory-preparedfromCNBr-orProteinA-Sepharosegel.Experimentalevidencehowevershowsthat,togetherwithalbumin,anumberofproteinsare

boundto,andmaybereleasedfrom,ananti-albuminresinafterplasma/serumimmunodepletion(Table1);theirensemblehasbeengiventhenameofalbuminome[17].Theuseofdifferentcommercialproductsresultsintheco-binding

ofadifferentassortmentofnon-albuminproteins[18].Albuminomecompositionhasbeenassessedalsothroughcomplementaryapproaches.Inone,humanalbuminwasimmobilizedtoasolidactivatedimmunoaffinitysupportviaa

couplingreaction.Afterincubatingtheresininaserumdilutionandextensivewashing,tightlyinteractingproteinswereelutedwithachaotropicsolutionandidentified[19].Inanother,thesupernatantfrom42%ethanol/100mMNaC

precipitation on IgG-depleted serum was processed though size exclusion chromatography and proteins co-eluting with albumin were assessed [17]. All the identified proteins are assumed to interact with albumin in vitro as a

counterpart of their interaction in vivo; accordingly, their changes have been investigated as biomarkers of liver disease (hepatitis C virus-related cirrhosis, small unifocal hepatocellular carcinomas and advanced hepatocellular

carcinomas)[20].

Table1Albuminomecomponents[17].

alt-text:Table1

Intactproteins

Apolipo proteinA-IIApolipo proteinA-IVApolipo proteinC-IIApolipo proteinC-IIICeruloplasminClusterinComplementcomponent1inhibitorHaptoglobinHemoglobin,alphachainHemoglobin,betachainHemopexinLeucine-richalpha-2-glycoproteinTransferrinTransthyretinVitaminD-bindingproteinZincalpha-2-glycoprotein

Proteinfragments

CarboxypeptidaseB2ComplementcomponentC4HornerinInter-alpha-trypsininhibitorheavychainH4KininogenParaoxonase1Peptidoglycanrecognitionprotein2Plasminogen

2.1.2.2.1.2Removingmajorplasma/serumproteinsInagreementwiththeargumentsinSection1,removingmost/allmajorproteinsfromplasma/serumisasounderapproachthanremovingalbuminalonetoincreasethesensitivityofproteomicsprotocolsandtoextenddetection

andquantitationtowardssuchminorcomponentsastissueleakageproteins.Thisstepisexpectedtoimprovetheresolutionon2DEgelsbyenablingvisualizationofproteinsthatmigratetogetherwith,orcloseto,thehigh-abundance

proteinsandbyincreasingtheloadintheleastabundantcomponents.

I

pppp

Anattempttoexploit,tothisaim,conventionalaffinityproceduresintheanalysisofratplasmacanbefoundin[21].Theauthorscomputedtheaveragehydrophobicityofthemajorproteinsinthisspecimen(N=56)andsorted

theminclassesthroughclusteranalysis.Thehigh-abundanceproteinsturnedouttobelongtothemedium-hydrophobicityclass,thereforehydrophobicinteractionchromatography(HIC)wasrationallydesignedtodepletethemfrom

plasma.Indeed,whenappliedbefore2DE,HICpermitstodetecttwiceasmanyspotsasimmunoaffinitydepletionofalbumin(above).

Atargetedinsteadofabroadapproachimpliesmolecularrecognitionofeachofthecomponentstoberemoved,whichisfeasiblethroughimmunoadsorption.Anumberofproposalshavebeenmadeforlaboratory-madeaswell

ascommercialproductsdevisedtobindavariablenumberofplasma/serumproteins(N=6‐–20),bothofhumanandrodentorigin.Themajoradvantageofsuchdevicesisofcoursetheexpectedspecificityoftheirinteraction;themajor

drawbackliesintheirextremelyhighcost.Typically,forthisapplication,IgYs[22]directedagainstplasma/serumproteinsareraisedinhensduringthelayingperiodandpurifiedfromeggs(structureinFig.3).ProductionofIgYsis

minimallyinvasiveontheimmunizedpoultry,whichcomplieswiththecurrentguidelinesonanimalwelfare.Dependingontheantigen,Igtiterdevelopmentinchickenmaycomparemoreorlessfavorablywithrabbitsormice.Chicken

IgsareclearlysuperiortorabbitIgs intermsofepitoperecognitionwhentheantigenisaproteinofmammalianorigin,whetherhumanoranimal.Suchanadvantageintermsofhigherspecificity/lowercross-reactivityandhigher

sensitivity is obviously connected with the higher phylogenetic distance between self and immunizing non-self in chicken than in mammals. Among the commercial products, it is worth mentioning those from Agilent

(https://www.agilent.com/), fromR&DSystems(https://www.rndsystems.com/)and fromSigma(http://www.sigmaaldrich.com/).They include twoclassesofdepletioncolumns:one targeting12or14of thehigh-abundanceproteins,

anotherthemiddle-abundanceproteins;thepolyclonalpolyspecificantibodiesforthelatterareobtainedthroughimmunizationwiththeflow-throughfromtheformer.Thesequentialuseofbothcolumnsisintendedtoleaveinsolution

onlylow-abundanceproteins.

A relatedapproach to ‘ultradepletion’ of humanplasmawasput to the test by fractionatingplasmawithdual ion exchange columns, producingpolyclonal IgY against each fraction andusing thepurified antibodies in an

immunodepletioncolumn.Atotalof165non-redundantproteinswereidentifiedafterdepletion;ofthese,38hadnotpreviouslybeenidentifiedinnon-depletedplasma[23].

Camelidsproducefunctionalantibodiesdevoidoflightchains;antigensareboundbytheN-terminalvariabledomainoftheheavychain(seeFig.3).Thecorrespondingsingle-domainantibodyfragments(VHHsornanobodies,in

Fig.3)arefullycapableofantigenbinding,withaffinitiescomparabletoconventionalantibodies.VHHsareefficientlyproducedinSaccharomycescerevisiae.Withtheirhighstabilityandsolubility,theseaffinityreagentshaveseveral

advantages forbiotechnologicalapplications [24].Onecommercialdeviceusing themtocapture14high-abundanceproteins isproducedbyBAC/Bionity (http://www.bionity.com/en/companies/15295/bac-b-v.html).Anexampleof its

applicationmaybefoundin[25](Fig.4):removalofhigh-abundanceproteinsisefficientandenrichmentoflow-abundanceproteinsappearsreproduciblefromruntorun.

Fig.3Overviewonthestructureofimmunoglobulinsofdifferentanimalorigin.

alt-text:Fig.3

Duetothecomplexityoftheimmunologicalapproach,severalreportshavebeendevotedtotheassessmentofanumberofissues.Theefficiencyofimmunosubtractionwasfirstevaluatedfordevicesmeanttoremove6high-

abundanceproteins(albumin,IgG,IgA,transferrin,alpha-1-antitrypsin,andhaptoglobin),comparingthe2DEpatternofserumbeforeandafterremovaloftheseproteins.Onepaperratedtheyieldthroughthenumberofdetectable

spots:withsampletreatment,thecountgrewfromapproximately850toover1500,witha76%increase[26].Inanotherpaperacriterionforquantitativecomparisonwasidentifiedinthenumberofmatchedspots(applyinganimage

analysissoftwaretothe2DEpatternsofseveralreplicates);thematchestotaled197beforeand317afterdepletion,withanincreaseof61%[27](Fig.5).

Thenextissuetobeaddressedwasthecost-effectivenessofdevicesmeanttodepletefromplasma/serumanincreasingnumberofabundantproteins.Withatitleasking:Themore,thebetter?[28],apapertriestoprovidean

answertothisquestionbycomparingtheresultsobtainedafterdepleting1(albumin),6(albumin,IgG,IgA,transferrin,alpha-1-antitrypsin,andhaptoglobin,seeabove),12(thepreviousplusapoA-Iand-II,α1-acidglycoprotein,alpha-2-

macroglobulin,fibrinogen,IgM)or20proteins(thepreviousplusIgD,ceruloplasmin,apoB,complementC1q,C3,C4,plasminogen,andtransthyretin).Afteranalysisby1DEand2DEandSELDI–TOF,theoutcomeisnotunivocal.The

gainissubstantialremoving12proteinsinsteadof6,limitedremoving20proteinsinsteadof12.Asalreadymentionedforalbuminanditsalbuminome,whileadsorbedundernon-denaturingconditionstheimmunocapturedproteins

remainassociatedwithawhole rangeofpeptidesandproteins,whichare removedalongside (off-target effects) (Fig.6).A largely similar experimental designwas implemented in anotherpaper [29]. In this case, the systematic

comparison involved the immunodepletionof6,14 (the twoextraproteinsboundvsthe12aboveareC3and transthyretin), or20proteins; theanalyticalprocedures includedboth top-down (2D-DIGE)andbottom-up (LC-MS/MS)

approaches.Usingthecombination1-DE––LC-MS/MS,thenumberofZ2uniquepeptidesrosefrom159inunfractionatedserumto301followingdepletionof20proteins.Whencomparingtheresultsafterimmunoadsorptionofeither

Fig.4Relativeabundanceofproteinsidentifiedinuntreatedplasmaandintheflow-throughfractionofanimmunoadsorptioncolumntargeting14majorplasmacomponents(bold).(Redrawnfrom[25]).

alt-text:Fig.4

Fig.52DEcomparingthedifferentproteinprofilesofuntreatedhumanserumandofserumimmunodepletedinsixhigh-abundanceproteins(albumin,IgG,IgA,alpha-1-antitrypsin,transferrin,andhaptoglobin).(From[27]).

alt-text:Fig.5

14 or 20 proteins, additional spots were seen in 2DE but most of them were isoforms of already detected proteins; moreover, a greater run-to-run variability was observed in LC-MS/MS after more extensive depletion (median

%CV = 30.9vs18.2%).Theremovalofeither6or14abundantcomponentswastestedalsoontheproteinsofsuchpeculiarspecimensasthemechanicallyinducedskinsuctionblisters,whosefluidmayserveasmodelsamplefor

dermatologicaldiseases[30,31](Fig.7).Analysisofthetrypticdigestsby2D-LC(reversed-phasechromatographyathighandlowpHafteriTRAQderivatization)associatedthedepletionofthetop14componentswithalowerefficiency

onthetargetproteins,yetahigherenrichmentfactor,ahighernumberofidentifiedproteingroups,animproveddynamicrangeandahigherloadingcapacity[31].Finally,fromafurtherpaper[32],weliketoquotequantitativedataon

thebindingpropertiesoftheSigmaresinsmeanttodeplete14high-abundanceandanumberofmedium-abundanceplasma/serumcomponents:theiruseremovesatleast155proteins,38%oftheplasmaproteomeinproteinnumber

and94%ofplasmaproteininmass.Theauthorsquestionontheweaknessofsuchextremeimmunodepletion (splitlineatimmuno-depletion(seepdf))procedureinfrontofevenslightvariationsininter-assaybindingefficiency:theyargue

thattheproteinsassociatedwiththehighestpercentageimmunodepletionarealsoatthehighestriskofbeingidentifiedasfalse-positivesindifferentialproteomicstudies.

Fig.62DEofuntreatedserumvstheunboundfractionscollectedafterdepletionof1,6,12and20majorserumproteinsusingcommercialcolumns(Table1).Afixedamountofprotein(50μg)wasanalyzedinallmaps.(From[28]).

alt-text:Fig.6

Fig.7ScatterplotofLog10iTRAQratiosvsLog10averageintensitiesforallpeptide-spectrummatches.Bluecirclescorrespondtothe6mostabundantproteins(albumin,IgG,IgA,transferring,haptoglobin,alpha-1-antitrypsin),andredcirclesincludetheadditional7−–14most

Asalastline,similartoalbuminome,thetermdepletomehasbeenintroducedtorefertotheoff-targetimmunosubtractedproteins(seeabove).

2.2.2.2Howtoenrichminorplasma/serumproteinsAnoppositeperspectivetodepletionofhigh-abundanceproteinsfromplasma/serumistheenrichmentof low-abundanceonesthroughanincreaseoftheirrelativeconcentration;suchacompressionofthedynamicrangeis

expectedtoprovideadeeperproteomeprofiling.Thistechnologywasdescribedbytheinventorsas“proteinequalization”anditseffectasapproachinga"“democraticproteome"”[33].Theenrichmentprocedureisbasedonaspecific

andsaturableinteractionofproteinstoahighdiversityofbindingsitesexposedonchromatographicbeads.Thecombinatorialligandlibraryismadeupofdozensofmillionsofhexapeptidescapableofinteractingwithmost,ifnotall,

proteinsinanygivenproteome[34,35](SupplementaryFigs.1and2).BioRad(http://www.bio-rad.com/)marketsthecorrespondingcommercialkit,ProteoMiner,togetherwith4elutionreagentsmeanttosequentiallyrecoverproteins

on thebasisofdifferentproperties, ina formatcompatiblewithanalysisbySELDI.Differentialelutionundernativeconditionsmaybeachievedbychanging thepHof thebuffer [36]orbyusingamixtureofchargedaminoacids

(150mMLys,Arg,AspandGlu)[37].ElutionofallboundproteinsunderdenaturingconditionsismadepossiblebyboilingthebeadsinSDS[38,39];precipitationwithacetonitrilethenprovidesforthehighestproteinrecoveryyieldand

thebest2DEspotpattern[40].

Theconceptualbackgroundtothepeptideligandlibrarycaptureisasaturationeffect:alltheproteinspresentinthestartingsampleinconcentrationshighenoughtooverloadtheirinteractionsitesonthebeadsshouldbe

boundinequalamounts,theexcessbeingdiscardedwiththeflow-through.Afterstrippingtheboundproteinsfromtheresin,theconcentrationofthehigher-abundanceproteinsshouldbeequalizedintheelutedfraction.Conversely,the

proteinspresentinthestartingsampleinconcentrationstoolowtooverloadtheirinteractionsitesonthebeadsshouldbequantitativelybound.Accordingly,thedifferencesamongsamplesshouldbeintentionallyabolished(oratleast

reduced)forhigher-abundanceandpreservedforlower-abundanceproteins.Atestofthequantitativeperformanceofthecapturebeadsshowsanaveragevariabilityaround10%intheamountoftheisolatedproteins,independentof

theirinitialabundance.Nonormalizationeffectisinfactobservedbut,dependingontheprotein-to-beadratio,specificsetsofproteinsareeitherenrichedordepletedvsthestartingwholeproteome[43](SupplementaryFig.3).All

reportsthatevaluatetheperformanceoftheproteinenrichmentstrategy[44]orcompareittodepletionprotocols[45]acknowledgetheexpectedcompressionofthedynamicrangeofserumproteinconcentrations.

Tofurtherreducesamplecomplexity,equalizationhasbeencombinedwithadditionalchromatographicstepsonIMACandRPcolumns(tandemmonolithiccolumnswithsurface-boundiminodiaceticacidligandssaturatedwith

Zn2+,Ni2+andCu2+connectedtoareversed-phasecolumnpackedwithpolystyrenedivinylbenzenebeads)[46].AnalternativetothehexapeptideligandshasbeenproposedintheformofsinglechainvariablefragmentdisplayingM13

phagelibrary[47].Bothimmobilizationonbeadsandelutionofboundproteinsrequirespecificprecautionsnottointerferewiththestructureofthephage;thenumberofproteinsidentifiedfromhumanserumwithsuchanenrichment

procedureistwiceashighasfromtheuntreatedsample.

Theenrichmentproceduremaybeappliedtoanykindofsampleofvegetaloranimalorigin.Focusingonhumanspecimens,plasma/serumhasbeenusedmostoftenbothwhensettinguptheproceduresandwhenanalyzing

actualsamplesformarkerdiscovery.However,alsootherbodyfluidshavebeenprocessedthisway,includingcerebrospinalfluid[48],milk/milkwhey[49,50],saliva[51],seminalplasma[52],artificialperitonealdialysiseffluents[53]

synovialfluid[54].

Thepathologicalstatesmainlyinvestigatedwereneoplasias(pancreas[55],lunginnever-smokers[56]andofthenon-smallcelltype[57,58])andinfections(hepatitisB-associatedlivercirrhosis[59],HIV-1/HCVmono-andco-

infection[60],invasiveaspergillosis–—inamousemodel[61]).Twofurthertopicsofpeculiarinterestaddressedwiththistechniquewerethechangesoccurringinserumwithgestationalages(forinfantswithintrauterinegrowth

restriction[62])andthoseoccurringunderextremephysicalstress[63].

Analternativeenrichmentapproachbasedonphysicochemicalpropertiesoftheproteinsratherthanontheirbindingaffinitiesisbeingproposedin[64].Thismethod,whichtargetshydrophobicproteins,consistsofaphase

separationwithTritonX-114.Thedetergent-richphaseiscollectedafterraisingthetemperaturetoabovethecloudpoint,thendilutedatatemperaturebelowthecloudpointandconcentratedbyaTCA-acetoneprecipitationstep.Ina

comparisonwithanalysisonwholeserumandonserumprocessedthroughalbuminandIgGimmunodepletionand proteinenrichmentonhexapeptideresin, thealternativeapproachallowedtorecognizestatisticalsignificanceof

differencesbetweencontrolandtestanimals(pigsexposedtofeedsupplementationwithsubtherapeuticlevelsofoxytetracycline)forthreehydrophobicproteins,namelyapolipoproteinF,clusterinandparaoxonase.

3.3Withmajorplasma/serumproteins:Howtodealwithwholeplasma/serum3.1.3.1Strategiesinproteomics

Allwehavepresentedsofarsharesaperspective:differencesinminorcomponentsaresoughtinplasma/serumaftersimplifyingthecompositionofthesample.Removingmajorcomponentsistheprimarystrategy;further

abundantproteins(IgM,alpha-2-macroglobulin,fibrinogen,complementC3,alpha-1-acidglycoprotein,apolipoproteinA-I/II(HDL),andapolipoproteinB(LDL)).(From[31]).

alt-text:Fig.7

P

simplificationmaybeobtainedwithanyfractionationprocedurewhilethefinalresolutionintosizeableanalyticalunits isobtainedwith1D-or,moreusually,2D-procedures,eitherelectrophoreticorchromatographic.Investigations

alongthislinearenotguidedbypreviousknowledge:statisticalsignificanceofdifferencesinabundanceistheonlyselectioncriterion.Inthissense,theseinvestigationsfullycomplywiththedefinitionofproteomicsasanunbiased

qualitativeandquantitativeevaluationofallproteincomponentsinagivensample–—sampleinthiscasebeinghumanplasma/serum.

Toovercomethedifficultiesalongthistrack,moreandmoreoftentheproteomicinvestigationonhumanpathologiesisbeingmovedfromplasma/serumtotissues.Proteinswhoseconcentrationissignificantlyalteredunderthe

testvsthereferenceconditionsarethenmeasuredinplasma/serumofthesamesubjectstoverifywhetherthecomponentsof interestarepresentata leveladequateforreliablequantitation.Whenthis isthecase,thecirculating

concentrationsareevaluatedandstatisticalsignificanceofdifferencesisfinallyassessed.

However,humanspecimensotherthanfewbodyfluidsrequireinvasiveproceduresfortheirprocurement,whichareoftenneitherpracticablenorethicallyacceptable.Moreover,duetogeneticandenvironmentalinfluences,

samplingconditionsfromhumansubjectsareseldomcomparable,whichincreasesbiologicaldiversityinanytestpopulation.Forthesereasons,inmanycases,thesimpleschemeaboveisbetterextendedfromatwo-toamultistep

procedure.Proteomicanalysisatthetissueleveliscarriedoutonanimalmodelsofdiseasevsrelevantcontrols.Differentialproteinsarethensoughtinhumantissuesections,toverifywhetherinthehumanspeciesthesamedifferences

inabundancebetweentestandcontrolsamplesarepresentasintheanimalmodel.Onlyafterthischeckthequantitationonhumanplasmas/seraiseventuallycarriedoutfortheputativebiomarkers.

Whilethediscoveryphaseofthisalternativestrategyimpliestoolsandprocedurestypicalofproteomics,thevalidationphaseusuallycompletelyreliesonimmunologicalreagents.Thequalityofthelatteriscrucialinthewhole

processbutproducing/screeninghigh-specificityhigh-affinityantibodies,whetherpolyclonalormonoclonal,isnotaneasytask.

Overviewsonintegratedstrategiesforbiomarkerdiscoveryandvalidationhavebeenpublished(e.g.[65]).Alloftheseaspectsaredealtwith,throughrelevantexamples,inthefollowingsections.

3.1.1.3.1.1ReferenceimmunologicalreagentsThelimitednumberofavailablespecificantibodiesinspiteofamultitudeofeffortsintheacademicandindustrialsectorsledtheHumanProteomeOrganization(HUPO)toendorseantibody-basedproteomicsprogramsinvolving

thesystematicgenerationofantibodiesforuseinfunctionalexplorationofthehumanproteomewithahigh-throughputapproach[66,67].

ThisendeavorhasbeengiventhenameofHumanProteinAtlas(http://www.proteinatlas.org/about);itwasstartedattheRoyalInstituteofTechnology,Sweden,in2003andisfundedbythenon-profitorganizationKnutandAlice

WallenbergFoundation.Updatesontheprogramachievementsaresincepublishedonaregularbasis(e.g.[68,69]).

Thein-houseantibodyproductionprocessbeginswithabioinformaticsanalysisoftheprotein-codingpartofthegenome.Foreveryprotein,theaminoacidsequenceiscomparedtoallotherputativeprotein-codinggenesto

identify a stretchof50‐–150aminoacids thathas as lowhomologyaspossiblewith respect to all otherproteins.These sequences are then cloned fromcDNA librariesusing specificallydesignedprimers and transformed intoE.

coliEscherichiacolibacteriathatproducethecorrespondingpeptidechain(proteinepitopesignaturetagorPrEST).ThePrESTisusedforvariousapplicationsincludingimmunizationtoproduceantibodies,andforaffinitypurifyingthe

polyclonalantiseratomakethemstrictlymonospecific.Numerousqualityassuranceandvalidationstepsareperformedtoeventuallycertifytheantibodies.Thesamevalidationprotocolisappliedalsotocommercialproductsandto

antibodiespreparedbycollaboratinginstitutions.

BycopyingintheaboveparagraphsmostofthebasicinformationaboutthepipelineofantibodyproductionsummarizedintheAtlaswebsite,wewanttoemphasizethat,inorderforthequantitativeresultsfrominsolutionor

fromonmembraneimmunodetectionprocedurestobemeaningful,thequalityoftheimmunologicalreagentsmustbecertifiedthroughacomplexandstringentprocedure(typicallybyexternalbodies)andthecalibrationcurvesforthe

assaybecarefullybuilt(inhouse).SometimesthedifferentialitemcorrespondstojustoneofanumberofspeciesderivingfromasinglepolypeptidethroughdifferentialPTMprocessing(seeSection4.1).Inthesecasesthecopynumber

percellof theprotein (allspecies) isaffectedmuch less thanthatof the individualspecies;ameasureof theproteinasawhole,byELISAorby immunoblottingafter1DE,would likelymissstatisticallysignificantdifferences (an

exception,1DEafterextensiveproteolysis).Suchcasescanbeeffectivelyaddressedonlythrough2DEwithimmunoblottingandcomprehensiveimageanalysis.

3.1.2.3.1.2ReferencedataontissueproteomicsThebaselinelevelofexpressionforeachindividualproteinineachtissueisrelevantinformationwhenchangesunderpathologicalconditionsareevaluated.ThereforewegoonreportingbasicinformationabouttheProteinAtlas

project.

Proteinexpressionprofilesareproducedforeachofthecertifiedantibodiesbystainingdifferentstandardizedsetsofbiologicalsamplesfrom44differentnormalhumantissues,20differentcancertypes,46differenthumancell

linesand6hematopoieticcelltypesfrompatients.Asofspring2016,proteomeanalysishasbeencarriedoutbasedon25,039antibodies,targeting17,005uniqueproteins[69];resultsareavailableonline,inasearchablewebsite

(http://www.proteinatlas.org/about)thatalsocontainstranscriptomicsdataonthemRNAscodingforthesameproteins.InformationisdividedintoaTissueAtlas,aSubcellularAtlas,aCellLineAtlasandaCancerAtlas.

Indetail,thetissuemicroarraysforhigh-throughputimmunohistochemistryconsistsofmultiple1-mmdiametercoresfromformalin-fixedparaffin-embeddedtissuespecimens,arrangedinasingleparaffinblock.Eachblockis

cutinto200‐–250sectionstobeusedforseparateimmunohistochemicalstainingexperiments;processingseveralarraysundersimilarconditionssizablyreducesintra-experimentalvariation.Thestainedslidesarethenscannedand

pathologistsannotateeachcoreimagebyscoringtheintensityofimmunoreactivitywithitscellularlocalizationandbycountingthefractionofimmunostainedcells.

3.1.3.3.1.3MultistepexperimentalplansFromthefirstproposalstothemostrecentexperiences,increasinglevelsofcomplexitymaybefoundindifferentexperimentalplans,shiftingbetweentissuesandbodyfluidsandbetweenhumanandmodelanimalspecimens.

Abreakthroughinthisareawastheuseofxenografts,i.e.immune-incompetentnude(mutationinFoxn1gene)orSCID(mutationinPrkdcgene)orRAG-2/γ(c)KO(H-2dRAG2−/−IL-2Rγ−/−,orRAG2−/−γc(gammacommon)−/−

micetransplantedwithhumantumorcells.Proteinssecreted(orshed)bythetumorcanbeidentifiedinmousecirculationandtheirhumanderivationconfirmed.Thexenograftsetupprovidesamuchmorefavorableratiooftherelevant

humantumor-derivedbiomarkersvstheirrelevantmouseplasma/serumproteinbackground.Examplesofthisapproachincludethefollowingreports,inwhicheitherestablishedcelllinesorpathologicalmaterialderivedfromindividual

cancerpatientsaregrafted.

Asfortheestablishedlines,PC339cellsprovideamodelofandrogen-independenthumanprostatecancer;tumor-derivedhumannm23/nucleoside-diphosphatekinaseandsixhumanenzymesinvolvedinglycolysiswereidentified

inmicebearingthexenograft [70].Ahepatocellularcarcinomamodelmaybeobtainedby injectionofHCCLM3cells; the levelsofcirculatingcytokeratin19andof its fragmentCYFRA21-1were foundassociatedwithmetastasis

potential[71].Threecelllines(liposarcoma[SW872],osteosarcoma[KHOS‐

24OS], and mammary adeno-carcinoma [MDA-MB-436]) exist in two, non-angiogenic (microscopic, dormant tumors) or angiogenic (rapidly growing tumors) phenotypes; analysis of the corresponding xenografts suggested that

platelet-associatedplateletfactor-4,butnotitsplasmacounterpart,mayrepresentapotentialbiomarkerofearlytumorpresence[72].GraftedMKN45gastriccancercellswerefoundtoreleaseinmouseplasmahumanapoA-I;itslevel

waslowerinmicewithlargethaninthosewithsmalltumors[73].

Asanexampleofbiopsy-derivedcelllines,specimensfrombothhumanoralsquamouscellcarcinomasandadjacentcontroltissueweretransplantedorthotopicallyintomousetongues;severalproteinswerefoundtodifferin

concentrationbetweencontrolandcancer-bearingmice;amongthese,EGFRlevelsinverselycorrelatedwiththeinvasivephenotype[74,75].

The first type of advance from the above schemewas the validation in human sera of the findings in the xenograftmousemodels. For instance, transplant ofNPC-TW02 cells, derived froma keratinizing nasopharyngeal

carcinoma,resultedinelevatedcirculatinglevelsofperoxiredoxin2andcarbonicanhydrase2;similarresultscamefromtheanalysisofplasmafromnasopharyngealcarcinomapatients[76].Ninety-sevenproteinswereidentifiedinthe

culturesupernatantoftheovariancancercelllineSKOV-3.Afterthecellswerexenograftedintotheperitonealcavitiesofnudemice,threeoftheseproteinsweredetectedinanimalsera;oneofthem,14‐–3-3zeta,wasidentifiedasa

candidatebiomarkerthroughanELISA-basedscreeningofclinicalbloodsamples.Theaverageserumlevelsof14‐–3-3zetawerehigherinpatientswithepithelialovariancancerthanwithbenigngynecologicaldiseasesandcorrelated

withclinicalparametersofdiseaseseverity[77].Usingadifferentovariancancercellline,TOV-112D,morethan200>200humanproteinswereidentifiedinthexenograftserum;threeofthem(chlorideintracellularchannel1,cathepsin

D,andperoxiredoxin6)werethenfoundelevatedinserafromovariancarcinomapatients[78].

Asimilarshiftinplansappliestoadifferenttypeofanimalmodels,thegeneticallyengineeredmice.

Bothseraand tissuesampleswerecompared fromPdx1-CreKrasG12DInk4a/Arflox/loxmiceasmodelsofpancreaticcancerand fromKrasG12DInk4a/Arf lox/lox andPdx1-Cre Ink4a/Arflox/lox control animals. A panel of five proteins

selectedonthebasisoftheirincreasedlevelatanearlystageoftumordevelopmentinthemouse(neutrophilgelatinase-associatedlipocalin,lithostathine-1-alpha,regeneratingislet-derivedprotein3-alpha,metalloproteinaseinhibitor1,

andinsulin-likegrowthfactor-bindingprotein4)wastestedinablindedstudyon26subjectswithincreasedcancerrisk(fromthelargecohortof>18,.000individualsenrolledintheCaroteneandRetinolEfficacyTrial,CARET).The

paneldiscriminatedpancreaticcancercasesfrommatchedcontrolsinbloodspecimensobtainedbetween7and13 monthspriortothedevelopmentofsymptomsandclinicaldiagnosisofpancreaticcancer[79].Plasmadatafroma

comparisonbetweenAdenoCre-infectedK-ras/PtenmiceasanovariancancermodelandAdeno-emptyinjectedcontrolswereevaluatedagainstevidencefromthreehumanovariancancercelllines(OVCAR3,CAOV3,andES2)andfrom

cellsintheascitesfluidofanovariancancerpatient.Asetofproteins(insulin-likegrowthfactor-bindingprotein2,metalloproteinaseinhibitor1,retinoicacidreceptorresponderprotein2,monocytedifferentiationantigenCD14,and

granulin)weremeasuredathigherlevelsinpathologicalthanincontroltissuesandbodyfluids,bothhumanandmurine.Thedifferencefromcontrolswassignificantfromtheearlystagesofthedisease[80].

Anexamplenotrelatedwithcancer involvesatransgenicmousemodelwithcardiac-specificoverexpressionofactivatedcalcineurin(CnA),whichresults inseverecardiachypertrophy.Fourproteins(myosinheavychain7,

insulin-likegrowthfactor-bindingprotein7,annexinA2,anddesmin)werefoundsignificantlydifferentintheCnAheartscomparedtocontrolsandoverexpressedinproteomicandtranscriptomicdatasetofheartfailure.Immunological

quantitationinmouseandhumanserademonstratedthatallfourproteinsincreasedbetweentwofoldand150-foldinheartfailure[81].

,

3.2.3.2ProceduresinproteomicsBothoptimalandmaximumloadinaseparationprocedureisafunctionofthesize(length,surface,orvolume)oftheseparationmediumduringitslimitingstep.In2DE,thevolumeofthe1stdIPGstripcontrolsthetotalamount

ofproteinstobeprocessedpertest.Usingstandardready-madecommerciallyavailablestrips,theamountofproteins/thevolumeofwhole(andaswellasofdepleted)plasma/serumtoberunisalsostandard.

Usingstandardready-madecommerciallyavailableIPGstripsisthemostcommonprocedureinIEFforpracticalreasonsthatincludeboththeeaseofthecurrentsimplifiedprotocolsandthereproducibilityofthesubstratefor

theseparationprocedure,whichentailsthereproducibilityoftheresults.Theseobviousadvantagesshouldhoweverbeweighedagainstafewdisadvantages,lackofflexibilitybeingthemostevidentandthemoststringent.

FromtheearlyhistoryofIPGs[82]weareaccustomedtopolymerizeourowngels,inallkindofpHgradient,andtocutstripsfromthem,ofallsizesandshapes[15].Thismakesitveryeasytoloadverylargevolumesofwhole

serum;theonlyprecaution,alreadyrecalledin2.1.1,istoapplythesamplenearthecathode,sincerunningthroughregionswithpHlowerthanpImostoftenresultsinsubstantialalbuminstreaking,asfirstdemonstratedbyusbackin

1985[83].Conversely,exposuretoacidicpHofmuchofthealbumincontentofasamplecannotbeavoidedwhenproteinsareloadedthroughthein-gelrehydrationprocedurethatiscustomarywiththecommercialIPGstrips.

Even foranimals the sizeof amouse –—few tens of grams inweight –—tens ofmicroliters ofwhole plasma/serumaremost often available for testing. This allows for each sample to be analyzed undermore than one

experimentalcondition.Using2DE,obviousvariations to thebasicprocedureaimat focusingononeormoreareas in thepI/Mrplaneviazooming-in i.e.bynarrowing the rangeofproteinchargeandmass tobecoveredathigh

resolution.However,alessfrequentlytakenbuthighlyeffectivealternativeistoruneachsamplebothunderreducingandundernon-reducingconditions.Aswehavedemonstratedforanumberofspecies,humansandanimalsaswell,

thischangeinexperimentalprotocolsignificantlymodifiesthemigrationpositionnotonlyofdisulfide-associatedmulti-chainassemblies,whichsplit intotheirconstituentsubunits,butalsoofdisulfide-bridgedsingle-chainproteins,

whoseunfoldinginthepresenceofSDSisrestrictedbythepresenceofcovalentbonds[84].

Thefollowingobservations, takenfrom[84],summarizeour findingsabout themajorserumproteins in fourspecies–—Homosapiens(Fig.8),Rattusnorvegicus,Musmusculus, Bos taurus – — and stress advantages and

disadvantagesofalternativeprocedures.Undernon-reducingconditionsalbuminofallspeciesmigratestoamuchlowerMrthanunderreducingconditions:55vs68kDa,consistentwiththepresenceof17S-Sbonds.Eitherbecauseof

betterrecoveryorhigheraffinityforCBBstain,thespotfornon-reducedalbuminislargeranddarker.Polymericformsontopofadiffuseverticalstreakingarealsoobserved.Unreducedconditionsallowforthebestresolutionbetween

albuminandhemopexin inhuman,mouseandratserumsamples.Albumin fragmentsarepresent indifferentamount invariousspecies.Where theyareprominent (ratandmouseserum)detection ispossibleonlyafter reduction

–—whichimpliesthepresenceinratandmousecirculationofasubstantialamountofcleavedbutundissociatedalbumin.

TransferrinalsocontainsalargenumberofS-Sbridges(19)anditsMrshiftsfrom71to77kDabetweennon-reducingandreducingconditions.Bovineserumcontainstwodifferentmolecularformsoftransferrin,whichdiffer

bythepresenceorabsenceofaC-terminalpeptideofabout6kDa[85].Theresolutionbetweenthetwoisminimalundernon-reducingconditionsbut increasesonreduction.Thechargemicroheterogeneityofnativetransferrinis

connectedtobothcovalentpost-transcriptionalmodifications(glycosylation,implyingupto8sialicacidresidues)andnon-covalentinteractions(bindingofupto2Fe++ions).Thesusceptibilitytoureadenaturationvariesdependingon

ironsaturationandredoxstatus.Thediferricproteinisunaffectedbyurea,andinthemonoferrictransferrinonlytheiron-freedomainisassumedtounfold[86];inovotransferrin,themoleculecontaining4S-Sbridgesisstillableto

bindiron,andthecorrespondingcomplexdoesn'tunfoldin8Murea,whereasthe3S-Sand2S-Sproteinsexistonlyasapoforms[87].Accordingly,moreproteinsspeciesareresolvedbeforethanafterS-Sreduction.

Theamountofimmunoglobulinsvariesamongspecies,beingmuchhigherinhumansandBovidaethaninRodentia.Asexpected,withoutsamplereductionIgsformhorizontalstreaksathighMrwhereasafterreductionheavy

andlightchainsareresolvedatlowerMr.

Fig.82DEpatternofserumfromacontrolsubjectrunundernonreducing(left)vsreducingconditions(right).Ourunpublishedresults.

alt-text:Fig.8

Baselinelevelsofhaptoglobinalsovaryamongspecies,beinghigherinhumansandmice.Intheseserumsamples,alphaandbetachainsareresolvedaftersamplereductionbutnointacthaptoglobinisobservedinthe2DEmap,

probablyasaresultofaggregation.Incontrast,inratserumadistinctiffaintseriesofhighMrhaptoglobinspotsisseenundernon-reducingconditions.Possiblyduetotheirlowabundanceonceseparated,bothalphaandbetachains

arehardlydetectable in rat serumafter sample reduction. In rats andbovines,noticeable levels canonlybedetected in caseof inflammation [88,89]. In addition,haptoglobinappears indifferentphenotypesandpolymorphisms,

dependingonthespecies[90].

ApolipoproteinA-IIisahomodimerinhumansandamonomericproteinintheotherspecies.Accordingly,theMrofapoA-IIinhumanserumistwiceashighundernon-reducingthanunderreducingconditions.

Themostcrowdedareainserum2DEpatternsistheonewhereseveral‘alpha-globulins’migrate–—pIca.4.5‐–6andMrca.50‐–70kDa.InhumanandratsamplestheintrinsicdifferencesinpIandMrbetweenadjacentspot

rowsunderreducedconditionsaresufficientfordiscriminationofallcomponentswithminimaloverlaps.Full-rangeIPGandSDS-PAGEareadequateforquantitationofallserumproteinsandzoomingisanaid,notarequirement,for

resolution in the ‘alpha-globulin’ areaof these species.Withmouse serum the reducedproteinsmigrate in closerproximity, and zoomingbecomesmore valuable. In reducedbovine serumanumberof proteins concentrate in an

extremelynarrowMrrangewhileformingacontinuumofchargedistributionwithfewsharplydefinedspots:evenwithspecificPAAgradientsforfinersizeresolutionthediscriminationamongproteinsisthusverypoor.

Forallserumsamplesthesituationintermsofoverallresolutionoverthespecified‘alpha-globulin’areaisatleastasgoodundernon-reducingasunderreducingconditions–—andinsomecasesitismorefavorabletorunthe

proteinsunreducedorpartiallyreduced.Thisisespeciallythecasewithbovineserum:foritsanalysisitseemsappropriatetouseatwo-stepstrategy,namelyarununderreducingconditionsonawiderange2DEgelfollowedbyarun

undernon-reducingconditionsonanarrow-range2DEgel(e.g.pH4‐–6IPG,SDS-PAGEon6‐–12%PAA).Thesecondstepisadequatefortheevaluationofthe‘alpha-globulins’,thefirststepfortheevaluationoftheotherproteins(low

Mr,oralkalinepI).

4.4Focusonmajorplasma/serumproteinsMuchofwhatwehavedescribedsofarhasasanexplicitorimplicitrationalethequestforoptimalprocedurestodetectandquantifyminorserumcomponents,mostofwhichderivedfromperipheraltissuesas

aresultofpathologicalevents.Inthisperspective,majorserumproteinsareregardedasaworthlessburden.Inthefollowingweproposeadifferent,oratleastcomplementary,view.

Thepointwewanttomakeisthatinformationprovidedbyabundantproteinsshouldnotbeoverlooked,mainlywhendealingwithapoorlycharacterizednaturallyoccurringconditionorwithanewlydevised

experimentalmodel.Ideally,inouropinion,thesurveyonwholeplasma/serumshouldalwaysbethefirststepofamultistageinvestigation,tryingtoaddress,oneafteranother,qualitativeandquantitativefeaturesof

differentsetsoftestproteinsatincreasinglevelsofresolution.

4.1.4.1ProteinspeciesEachproteinmay exist in variousmolecular forms as the compound result of several types of PTM [91].Changes in the relative abundance of these protein species reflect the varying activity of the enzymes involved in

alternativePTMprocessing(tworecentreviewarticlesonthistopicasappliedtoe.g.neurodegenerationandautoimmunediseaseare:[92,93]).Theactivitiesofthemodifyingenzymesareinturnaffectedbyanumberofphysiological

andpathologicalfactors:datafromtheassessmentofindividualproteinspeciesmaythusprovidecluestotheattendantconditions.

4.1.1.4.1.1ProteolysisAlbuminispresentinplasma/seruminformsthatdifferbysize–—proteolyticfragmentsontheoneside,polymersontheother.Theformeraredetectableonlyunderreducingconditions,thelatteronlyundernon-reducing

conditions(Section3.2).Innormalhumanplasma,albuminfragmentshaveMraround45,28and19kDaandmakeuplessthan2<2%oftotalalbumin.Theoxidantenvironmentoftheextracellularmilieupreventsfragmentdissociation

tooccurinvivo:indeed,theratiofragmentsvsintactalbuminisnothigherinurinethaninplasma/serumsamplesfromthesamesubject.

Chymase is a chymotrypsin-type serine protease stored in the intracellular granules of mast cells until liberated by degranulating stimuli. Proteolysis of its substrates is part of the inflammatory response: processing of

angiotensinItoangiotensinIIsupportsbloodpressure;degradationofinflammatorycytokinesandotherbioactivepeptideslimitsthespreadofthereaction;activationofprocollagenasetocollagenasecontributestoextracellularmatrix

remodelingandtoresolutionoftissueinjury[94,95].Chymaserecognizes inalbuminthetargetsequenceRETYtocleave itatY84andproducea larger,ca.59 kDa,andasmaller,ca.10 kDa, fragment [96].However, as the two

fragmentsdonotdissociate(becauseoftheC75‐

S‐

S‐

C91 covalent bond) nicking by chymase has no influence on the properties of albumin, including cation binding (Cu2 + or Zn2 +) at a site near the cleavage point [96]. Accordingly, the in vitro demonstration of anticoagulant

propertiesforS-carboxymethylatedCNBr-fragmentsofalbuminisunlikelytohaveanyinvivocounterpart,andnotonlybecauseoftheunphysiologicaltypeofcleavage[97].

A peculiar type of nicking is involved in the antiprotease mechanism of serpins, as exemplified by alpha-1-antitrypsin. A protease, most often elastase from granulocytes, docks with the exposed reactive loop of alpha-1-

antitrypsin,becauseitssequenceandconformationresembleamodelsubstrate.Aftercleavageatthe‘bait’peptidebond(M358-S359),thereactivelooprapidlyinsertswithinanunderlyingbeta-pleatedsheetmotifasanextra,central

beta-strand.Intheprocess,thecatalyticmachineryofelastaseisdistortedandtheenzymecannotcompletethecatalyticcycleandregeneratetothefree,activeformfromthereactionintermediate(inwhichelastaseacylatesS359).

The thermodynamic changes associated with the conformational rearrangement of alpha-1-antitrypsin from a metastable (S, for stressed) to a stable (R, for relaxed) conformation are highly favorable, and allow the coupled,

thermodynamicallyunfavorable,distortionandtrappingofelastasetotakeplace[98,99].Theoccurrenceofcirculatingelastase–alpha-1-antitrypsincomplexeshasproventobeamostsensitive indicatorofgranulocyteactivation in

sepsisandtrauma[100,101].Assuchithasforinstancebeenevaluatedasamarkerofunwantedsideeffectsofexogenoussurfactantreplacementforimmaturepretermneonateswithrespiratorydistresssyndrome[102].Baselinelevels

ofthecomplexesarearound0.1μg/mLtoincreaseafewfoldsduringgranulocyteactivation[103];theirquantitationismadebyELISA.

Thyroxine-bindingglobulin(TBG)isanon-inhibitoryserpin.Cleavagebyelastaseresultsinthelossofa4-to5-kDaC-terminalfragmentandamajorconformationalchangethatweakenstheaffinityforT4[104].Theproteolytic

fragmentsareobservedinseraofsepticpatients[104]butalsoduringtherapiddecreaseofTBGfollowingcardiopulmonarybypass[105].Thefallinconcentrationtoameanlevelof60%vsthepreoperativecontrolat12hourshseems

thusconnectedwiththelocalinflammatoryresponsetothesurgicalprocedureandresultsintheaccumulationoffreeT4attheinflamedsites[105].

WehavedescribedourselveshowapolipoproteinA-IinHDLisatargetofthematrixmetalloproteinasesreleasedfromactivatedmacrophages.Thefragments,of26,22,14and9kDainsize,resultfrombothN-andC-terminal

cleavageanddissociatefromboundlipids.HDL3C(smallpre-betaHDL)aremoreextensivelydegradedthantheotherHDLsubclasses,withadverseeffectsonreversecholesteroltransportinwhichthesmallestlipoproteinparticlesare

specifically involved [106].Wecoulddetect invivoproteolytic fragmentsofapolipoproteinA-I in somepatientswithacutemyocardial infarction, likelyasa resultof thesameactivationevents that in thesesubjects lead toplaque

fissuration[107].Moreover,weobservedsuchfragmentsalsoafterthrombolysiswith(recombinant)tissue-typeplasminogenactivator.Thistherapywasanearlybreakthroughforthetreatmentofcoronaryocclusion.Oncepercutaneous

coronaryinterventionbecameafeasiblealternative,thrombolysiswassometimesassociatedtosupposedlyfacilitateit.Unexpectedly,however,thebenefitoffacilitatedangioplastyturnedouttobereducedincomparisonwiththenon-

facilitatedintervention.DegradationofcirculatingHDLmayprovideapotentialmechanismforthisoutcome.

Bikuninisa25kDaproteinaseinhibitorpresentinhumanplasmacovalentlylinked,throughaglycosaminoglycanchain,tooneortwohomologousheavychains,forminghighmolecularweightproteinaseinhibitorscalledpre-

alpha-andinter-alpha-inhibitor.Duringinflammation,bikuninsynthesisisdown-regulated,stillitsexcretioninurineincreases.Thisapparentinconsistencyisconnectedwithneutrophilactivationandthecleavageoftheinhibitorfrom

theC-terminusoftheheavychainsbythereleasedproteases[108].

C-reactiveproteinorCRP,themainpositiveacutephasereactantproteininhumans,displaysseveralfunctionsassociatedwithhostdefense:itpromotesagglutination,bacterialcapsularswelling,phagocytosisandcomplement

fixationthroughitscalcium-dependentbindingtophosphorylcholine;itcanscavengenuclearmaterialreleasedfromdamagedcirculatingcellsinteractingwithDNAandhistones.CRPispartofacomplexhomeostaticmechanismthat

governstheacutephasereaction,regulatingtheimmuneresponseandtheactivityofmatrix-degradingenzymes.Itisnoteworthythatdifferentpropertiesaredisplayedbythewholepentamericassemblyaswellasbysomeproteolytic

fragments.Forinstance,whilenativeCRPdoesnotinducephagocyticleukocytestochemotaxortoproducesuperoxide,treatmentofpurifiedCRPwithhumanneutrophil-derivedacidproteasesproducessubstanceswithpotenteffects

onleukocytefunction.Asalikelyexplanation,thesequenceofCRPcontainspeptidescloselyresemblingtheimmunomodulatortuftsin[109].IncontrasttotheinabilityofintactCRPtoinhibiteitherhumanleukocyteelastaseorhuman

leukocytecathepsinG,bothassociatedwithchronicinflammatorytissuedamage,thepeptide62‐–76fromtheinnerdisulfideloopofhumanCRPinhibitstheseenzymesatconcentrationsfarlowerthantheacute-phaseconcentrationof

the protein [110]. CRP peptide 201‐–206 induces L-selectin shedding from human neutrophils and inhibits L-selectin-mediated neutrophil adhesion to TNFalpha-activated human coronary artery endothelial cells under non-static

conditions.Italsoattenuatesshear-inducedup-regulationofplateletP-selectinexpression,plateletcaptureofneutrophils,andsubsequenthomotypicneutrophiladhesioninhumanwholeblood[111].

4.1.2.4.1.2PolymerizationThissectionismuchshorterthanthepreviousonebutitstartswithalbuminaswell.Circulatinghomo-andhetero-polymericcomplexesofalbuminshowMraround210,168,147,132,and110kDaandrepresentbetween0.3

and2.8%oftotalalbumin[112].Inurine,theirpresenceisanobviousmarkerofimpairmentinultrafiltrationselectivity,forinstanceinassociationwithlupusglomerulonephritis[113].ThesurfaceantigenofhepatitisBvirus(HBsAg)

actsasareceptorforpolymerizedhumanserumalbumin;itslevelsarepredictiveoftheoutcomeofthedisease[114,115].

Wehavementionedabovethemainstepsoftheinhibitionoperatedbyalpha-1-antitrypsinonelastase.Pathologicalmutations,includingGlu342Lys(Z),subvertitspeculiarconformationtogenerateapolymerogenicstatefor

whichatleastthreemodelshavebeenproposed.Theclassicalor‘loop-sheet’modelinvolvesanintermolecularlinkageviathereactiveloopasasinglestrand,whereas,inthebeta-hairpinmodel,theintermolecularlinkageconsistsof

twobeta-strandsand, inthetriple-strandmodel,of thethreeC-terminalbeta-strandsofalpha-1-antitrypsin [99,116,117].Polymerizationoccursmainlyat the timeofalpha-1-antitrypsinsynthesis,with the formationofPAS-positive

inclusionsinsidethehepatocytes;theresultingpathologicallylowconcentrationofthecirculatingprotein(belowtheprotectivethresholdatca.600mg/L)isunabletopreventproteolyticdamageatthesitesofinflammationinlungs.

However,polymersarealsofoundinthecirculationofZZsubjects,bothwithandwithoutchronicobstructivepulmonarydisease(medianconcentrationaround40mg/L[118]).Polymersmaybespecificallydetectedwithamonoclonal

antibodythatrecognizesaconformation-dependentneoepitopeonbothpolymerizedandelastase-complexedmolecularformsofalpha-1-antitrypsin[119].

4.1.3.4.1.3DifferentialglycosylationInrecentyearsmassspectrometry, incombinationwithliquidchromatographyorcapillaryelectrophoresis,canreliablyassessproteinglycosylation,throughtheanalysisofeitherglycopeptides(afterproteolysis)orglycans

(afterPNGaseForchemicaltreatment).Theevidenceprovidedbytheseapproachesissharpeningourunderstandingofglycobiology:asexemplifiedbytransferrin[120,121],aproteinglycosylationpatternmayvarydependingonthe

tissueinwhichitissynthesizedorthebiologicalfluidinwhichitisaccumulated,andisextensivelyaffectedbydifferentphysiologicalandpathologicalstates.

Mucheffort for in-depth investigationof thesephenomena is concentratedonpathological tissues.Anumberof syndromesarebeingstudied in this respect, includingmultiple sclerosis [122],Alzheimerdisease [123] and

atherosclerosis [124]. Main focus, however, is on cancer, as glycans are involved in such crucial aspects of the disease as cell signaling and communication, tumor cell dissociation and invasion, cell-matrix interactions, tumor

angiogenesis,immunemodulationandmetastasisformation[125–129].

Mostproteins inplasma/serumareglycoproteins; for eachof themanumberofmicroheterogeneous species (glycoforms)maybe recognized.Systematic changes inglycosylationofplasma/serumproteinsareobserved in

connectionwithcancer,asexemplifiedbyovariancarcinoma[130].CancercellsproduceinflammatorycytokinesthatinfluenceglycosylationinhepatocyteswithanincreaseinsialylLewisX.Sialylationofacutephaseproteinsincreases

theirhalf-lifeandconfersanti-apoptoticpropertiesthusfavoringthesurvivalofcancercells.Dataonglycosylation,e.g. thecompoundinfluenceofbranchingandfucosylation,maybetakenasmarkerofcarcinomaprogressionand

prognosis[131].AreportonexperimentalmousecancersdetectedanincreaseininternalNeu5Gcalpha2‐–6sialylationontheGlcNAcoftheNeu5Gc2-3Gal1-3GlcNActerminalsequenceasacommonfeaturebutfoundthesubstitutionof

Neu5GcbyNeu5Actobeinducedbycolonandnotbybreasttumor.Amongtheabundantserumcomponents,transferrinwasfoundtochangesignificantlyinterminalglycosylationpatternbutnotinoverallexpressionlevel[132].

Glycosylationofplasma/serumproteinsisalsoaffectedinalltypesofdiseasesfeaturinganinflammatorycomponent[133].alpha-1-acidglycoprotein(AGP)isthemostextensivelyglycosylatedofplasma/serumproteins(over

50% by weight) and is a positive acute phase reactant, its concentration increasing between 5- and 50-fold in response to noxa. Within the calycin superfamily, lipocalin family, it belongs to the subfamily of immunocalins. The

immunomodulatory andbindingproperties of alpha-1-acid glycoprotein strongly dependon its carbohydrate composition: strikingly, different glycoformsbring about opposite immunomodulatory effectswhereas a pool ofN-linked

oligosaccharidechainsattachedtoasolublepolyacrylamidematrixdisplayasimilaractivity to itsparentAGP in thebiological tests [134].Theproteinglycosylationpattern ismodifiedbydisease: inacute inflammation,arelative

increaseofAGPglycoformswithbiantennaryunitsisobserved(atypeIglycosylationchange).Conversely,insomechronicinflammatorystatesthereisarelativedecreaseofAGPglycoformswithbiantennaryheteroglycans(atypeII

glycosylationchange).Thissametypeofglycosylationchangesisalsoseeninpregnancy,duringestrogenadministrationoruponliverdamage[135].Forallthesereasons,AGPisregardedasamodelandextensivelyinvestigated[136].

alpha-1-antitrypsinbothinhibitsserineproteasesandisinvolvedinimmunomodulationthrougheffectsonneutrophilchemotaxis,immunecomplexsignalingandapoptosis;itappearsthattheglycansofalpha-1-antitrypsinare

moreimportantforthelatterfunctionsratherthanfortheformer[137].

Acompletelydifferentareaofglycobiologytowhichtheanalysisofplasma/serumglycoproteinsisrelevantisthelaboratorydiagnosisofcongenitaldisordersofglycosylation.FordisordersofN-glycosylation,causalmutations

occurinatleast12differentgenes,whichencodemonosaccharidetransferasesoftheendoplasmicreticulum.Transferrinisoftentakenasthereferencematerialfortheanalysisofglycosylationanomalies:bymassresolution,theunder-

glycosylationoftransferrinischaracterizedasthetotalabsenceofoneorbothN-linkedoligosaccharides[138].

Despitegreatadvancements,thecurrentproceduresfordirectglycanassessmentoutlinedinthefirst linesofthissectionarestillcomplexandlengthy.However,duetotheinfluenceofglycosylationonbothpIandMrofa

protein,thequantitationofallthespecieswith2DEexperiments,underconditionsallowingtheirresolution,isastraightforwardandefficientapproachtodifferentiatevariousphysiologicalandpathologicalconditions.

4.2.4.2InflammationmarkersInflammationissuchabroadandgeneralphenomenonthatitsoccurrencecannotbetakenasdiagnosticofanyspecificcondition.However,itsdetailedcourseisfinelytunedbytheinterplayofseveralfactorsandsomefeatures

varyfromonecasetoanother.

Thesecretionofplasma/serumproteinsbytheliverfollowstwomainpatterns.ProteinssuchasCRP,serumamyloidAandalpha-1-acidglycoprotein,whichrequirethesynergisticactionofIL6andIL1formaximalinduction,are

saidtype1acutephasereactants(APRs).Proteinssuchasfibrinogen,haptoglobin,andalpha-2-macroglobulin,whichonlydependonIL6forup-regulation,aresaidtype2APR;oftentheirsynthesisissuppressedratherthanenhanced

byIL1.SomecytokinesareselectiveininducingoneorfewAPRs;cytokinesandothermediatorsinteractinadditive,synergistic,co-operative,andantagonisticways.Forinstance,asseeninculturedhepatomacells(HepG2),fourof

them–—IL1,IL6,TNF-alphaandTGF-beta–—arerequiredtocoordinatelyinducethesynthesisofalpha-1-antichymotrypsinwhilerepressingthatofalbuminandalpha-fetoprotein.Asaresultoftheabove,individualinflammatory

proteinsdifferintime-course(rapid-vsslow-reactingAPRs)aswellasindirectionandextentofchangeintheircirculatingconcentration(majorpositivevsmoderatepositivevsnegativeAPRs).

Theacutephaseproteinsfunctionasmediatorsandinhibitorsofinflammation,regulateimmuneresponses,actastransportproteinsforproductsgeneratedduringtheinflammatoryprocess,and/orplayanactiveroleintissue

repairandtissueremodeling.At leastsomeacutephaseproteinsmightconstitutean induciblesystemof factorsprotectingagainstcelldeathbyapoptosis: for instance,alpha-1-acidglycoproteinandalpha-1-antitrypsin inhibit the

activationofcaspase-3andcaspase-7tobecomethemaineffectors(executioners)ofapoptosis[139].

SomeAPRsbehave likecytokines:CRPactivatesmacrophages,migration inhibition factor (MIF)preventschemotaxis.TheAPRswithanti-proteolyticactivityblock themigrationofcells into the lumenofbloodvesselsand

preventtheestablishmentofageneralizedsystemicinflammation.

Theabovesummary(takenfromtheCytokines&cellsonlinepathfinderencyclopedia(COPE)websiteatwww.copewithcytokines.org)remindsthecomplexityoftheissue.Becausetheyvaryduringthecourseofaninflammatory

condition,groupanalysisofAPRs(serumAPRprofile)isnodoubtmoremeaningfulthanmeasuringanysingleprotein[140];APRprofilesshouldinvolveatleastonemajor(CRPorserumamyloidA,SAA),onemoderate(haptoglobin,

alpha-1-acid-glycoprotein,orceruloplasmin),andonenegativeAPR.Calculationofanindexfromvaluesofrapid-andslow-reactingpositiveandnegativeAPRshasbeenrepeatedlyproposedinveterinarymedicine,becauseitappearsto

increasestatisticalsensitivityandspecificityfordetectingnon-healthysubjects[141].

Examplesofdisease-specific associations for someAPRsareprovidedby two reports. Inoculation intonudemiceofhumancancer cell linesderived fromstomach,nasopharyngeal, colon, oral andbrain cancers results in

increasedlevelsofseveralmouseAPRs,includinghaptoglobin.Conversely,serumamyloidA(SAA),isfoundonlyinmicebearingtumorsinducedbythestomachcancercellline(SC-M1)[142].Inageneticallymodifiedmousemodel

(gp130F/F) that develops a gastric adenoma phenotype, the most acidic isoforms of alpha-1-antitrypsin are down-regulated and the most alkaline are up-regulated in tumourtumor-bearing relative to the tumourtumor-free cohorts

[143,144]–—thislattercase(Fig.9)beingconnectedaswelltotheissueofdifferentialglycosylationinSection4.1.3.

Thehighlightedintricacyofcross-influencesshouldsuggesttherelevanceofadetailedinvestigationoftheactualpatternofAPRs:inindividualclinicalcasesitcouldsharpenthediagnosticpowerofmultifactorialinformation,in

individualanimalmodelsofdiseaseitcouldhelpunderstandingthemechanismsofdisease.AsignificantfactorforthelatteristhenumberandvarietyofactivitieseachoftheAPRmaybeinvolvedin,assketchedintheaboveand

specifiedinthefollowing.

4.3.4.3Moonlightingproteins?

Fig.9DIGEofserumfromagastriccancermousemodel,highlightingdifferentiallyregulatedalpha-1-antitrypsinisoforms;box-and-whiskerplotsrepresentingtheaveragespotvolumeassociatedwithalpha-1-antitrypsininmouseserum,showingup-

regulation(toppanel,acidspots)anddown-regulation(bottompanel,basicspots)intumourtumor-bearingrelativetothetumourtumor-freecohort.(From[144]).

alt-text:Fig.9

Thegrowingevidencethatasinglepolypeptidechain–—asingleproteinor,better,asingleproteinspeciesor,whenrelevant,asingleproteindomain–—cansometimesperformmultiplephysiologicallyrelevantbiochemicalor

biophysicalfunctionsisattractingattentiononmoonlighting(multifunctionalormultitasking)proteins[145].Theabilitytomoonlightisoneofthewaysusedbycellstocarryoutmanycomplexprocesseswithalimitednumberofgenes.

Ancestralmoonlightingproteinsoriginallyperformedasinglefunction,usuallyenzymaticcatalysis,but,throughevolution,theyhavethenacquiredsecondarynon-enzymaticrolese.g.asreceptors,ionchannelsorchaperones.Some

moonlightingproteinscanperformbothfunctionssimultaneouslybutothersswitchrolesdependingonvariousfactorsincludingthecelltypeinwhichtheyareexpressed,theircellularlocation,theoligomerizationstatusandthebinding

of different ligands at different sites. Databases collecting and organizing information about the few hundreds of characterized moonlighting proteins are MultitaskProtDB at http://wallace.uab.es/multitask [146] and MoonProt

http://www.moonlightingproteins.org[147].Firstattemptshavestartedtowardsthecomputationalcharacterizationofmoonlightingproteins[148]andtheiridentificationviabioinformaticstools[149].

Noplasma/serumproteinislistedintheabovedatabases,whichonlyincludecellularproteincomponents.Antibodieshavebeendiscussedinthisperspectivetoconcludethat,whiletheydomoonlightasaclass(astheybindto

multiple antigen ligands), they do not as individual proteins (since the multiple functions they perform are encoded in different domains) [150]. Ceruloplasmin, on the contrary, in a review article has been acknowledged as a

moonlightingprotein[151].Thisplasma/serumglycoproteinisinvolvedincoppertransportandstoragebybinding6‐–7atomsofmetalpermolecule.Thereisaquestion,though,onhowthevariousenzymaticactivitiestheproteinisable

toperform(asferroxidase,ascorbateoxidase,nitricoxideoxidase)aresharedbetweenthevariousmetalbindingsites.Theseactivitiesmayeventuallyresultinavarietyofeffects:degradationoforganicsubstratesanddefenseagainst

oxidantstressandoverallanti-oxidant (splitword:oxi-dant(seepdf))activityor,conversely,underpeculiarcircumstances,pro-oxidantactivity.

Withouteverbeingmentionedassuch,someplasma/serumproteinsactuallyconformtothedefinitionofmoonlighters.Forinstance,ithaslongbeenknownthatsharplydistinctpartsofthemoleculeofalpha-1-acidglycoprotein

arerelevanteitherinligandbinding(theinnercavity,orcalyx)orimmunomodulation(theglycidicchains,seeinSection4.1.3).Otherproteinsarestillbecomingassociatedwithpreviouslyunrecognizedfunctions.Duetoitsscattered

nature,relevantliteratureisnoteasilysearchedforbykeywords;thetwoexamplesinthefollowingparagraphsshouldbetakenasanecdotalratherthanexhaustiveonrecentadvances.

Acute-phaseserumamyloidAisoforms(SAA1,SAA2,andSAA3)aresecretedduringtheacutephaseofinflammation.Theyareknowntohaveseveralroles,includingthetransportofcholesteroltotheliverforsecretionintothe

bile,therecruitmentofimmunecellstoinflammatorysites,andtheinductionofenzymesthatdegradeextracellularmatrix.Ithasrecentlybeendemonstratedthathuman(andmouse)SAAsalsobindretinolwithnanomolaraffinity;the

crystalstructureofthecomplexisthatofatetramerwithahydrophobicpocket.Duringinfection,SAAsareup-regulatedandtheirbindingretinolinvivocontributestolimitthebacterialburdenintissues[152].Hematopoietictissue

retainspoolsofstemcellsandprovidesamicroenvironmentforprogenitorcellrenewalthroughoutlife.Inadults,hematopoiesisisdependentonthechemokinereceptorCXCR4anditsligandCXCL12.Anadditionalroleisplayedby

elastasethat,whenboundtoleukocytesurface,actsasareceptorforalpha-1-antitrypsin.Bindingofantiproteasetoproteaseformsamotogeniccomplex[153].InthebloodofHIV-1patients,alpha-1-antitrypsin (splitword:anti-trypsinfor

line-break(seepdf))isboundandinactivatedbyanti-HIV-1gp120antibodies;thepresenceofsuchimmunecomplexescorrelateswiththedecreaseinnumberofCD4+lymphocytesinthesesubjects[154].

Inamoregeneralperspective,chaperone-assistedproteinfoldingisoneoftheprocessesthatconcurtoproteostasis,i.e.themaintenanceofadequatelevelsofindividualproteinsintheircorrectfoldingstate(Fig.10).Wehave

mentioned that intracellular chaperoning is one of the additional functions moonlighting proteins are often performing. Recent research has recognized among plasma/serum proteins a growing number of abundant extracellular

chaperonesconstitutivelysecretedinbodyfluids,whichactasbothsensorsanddisposalmediatorsformisfoldedproteins[155].Thechaperonesselectivelybindtoexposedhydrophobicregionsonmisfoldedproteinstopreventthem

fromaggregatingintheformoftoxicinsolubledeposits(suchasthoseseenintypeIIdiabetesandAlzheimer’'sandpriondisease).Thesamechaperonesarealsoimplicatedinclearingfrombodyfluidsthesoluble,stabilizedmisfolded

proteinsviareceptor-mediatedendocytosisforsubsequentlysosomaldegradation[156].

Clusterin, haptoglobin and alpha-2-macroglobulin are the plasma/serum components most directly associated with extracellular chaperone function but individual reports have investigated such property for a number of

abundantproteins(apoEisoformsdifferentfromtheε4allele,fibrinogen,alpha-1-acidglycoproteinandlipocalin-typeprostaglandinDsynthase).Albuminisconsiderablylessefficientatpreventingproteinaggregationthanrecognized

chaperonesbut,givenitsabundance,itsactivitymaybephysiologicallyrelevant(e.g.topreventamyloidformationbyAbetaandbytheVal30Metmutantoftransthyretin).

4.4.4.4Cross-regulationExtremelyreducedlevelsofindividualproteinsortheircompleteabsencefromthecirculationasaresultofmutationsinthegenescodingforsomemajorplasma/serumcomponentsareobservedasrarerecessivedisorders.The

pathologicalconsequenceseachoftheconditionsentailsvaryinpenetranceandseverity.Natureandrangeoftheeffectshelpdefinethefunctionoftheindividualproteinaswellastheredundancyofeffectorsthatmayvicariateforthe

missingcomponent.Moreover,changes in the levelsofothercomponentsofplasma/serumproteomeassessmodeandstrengthof thecross-regulationamongproteinconcentrations. Inadditionto theanalysisof thespontaneously

occurringnullphenotypesinthehumanpopulation,theseaspectsofproteostasismightbestudiedinsuchanimalmodelsastheknock-out(KO)mice.

In linewiththemarginal interestplasma/serumproteinsarecurrentlyreceiving,however,wecouldfindintheliteratureasatisfactorydescriptionof justonegeneticdisease,analbuminemia,andof justonegeneticmodel,

knocking-outofapolipoproteinA-I;onlythelatterwasstudiedcomprehensivelywithup-to-dateproteomicstools.Nocomparabledatahavebeenpublishede.g.regardingthediseasethatinvolvesthesecondmostabundantplasma/serum

protein‐—atransferrinemia,characterizedbyabnormalsynthesisoftransferrinandleadingtoironoverloadandmicrocytichypochromicanemia.Norarereportsavailableregardingalpha-1-antitrypsindeficiency,connectedwiththeZ

(Glu342Lys)ortheSmutants(Glu288Val),inheritedintheZZandSZgenotypes,muchmoreoftenthanwithanull‐–nullsetup.Themostcommonmanifestationofthedisorderisemphysema,whichbecomesevidentbythethirdtofourth

decade;alesscommonmanifestationisliverdisease,whichoccursinchildrenandadults,andmayresultincirrhosisandliverfailure.Insomepopulations,thefrequencyoftheZorSgenesisespeciallyhigh,whichhasbeententatively

connectedwithheterozygoteadvantageagainstlunginfections.

Analbuminemic individuals manifest mild edema, hypotension and fatigue. The most common biochemical finding is hyperlipidemia, with a significant increase in the total and LDL cholesterol concentrations, but normal

concentrationsofHDLcholesterolandtriglycerides.Invivostudieswithcontinuousinfusionoflabeledaminoacids(stableisotope)inanalbuminemicsubjectsdetectedincreasedproductionofIDL,LDLandVLDLanddelayedclearance

ofVLDL(apoB).Becausealbuminincreaseslipoproteinlipaseactivitybybindingfreefattyacids,whicharestronginhibitorsofthisenzyme,hypoalbuminemiacouldbeakeyfactorinthelowVLDLclearance.Ahigherratioofcholesterol

toTGinapoB-containingparticlesandanincreaseintheproportionofcholesterolesterrelativetoproteincouldexplainthedefectivecatabolism[157].Thefindingsaresimilaracrossspeciese.g.inanalbuminemicrats[158].Overall,

totalplasma/serumproteinlevelismaintainedwithinreferencevaluesthroughthecompensatoryincreaseinglobulinconcentration.Differentialregulationisobserved;theliverincreasesatthesametimesynthesisandsecretionofboth

positive and negative APRs (of transferrin, hemopexin, ceruloplasmin and fibrinogen and to a lesser extent of alpha-1-antitrypsin, alpha-1-antich ymotrypsin, alpha-2-macroglobulin, not of alpha-1-acid glycoprotein) [159]

Immunocompetentcellsalsoincreasethesynthesisofimmunoglobulins(IgG,IgAandIgM).HypercholesterolemiainanalbuminemicratsisassociatedwithincreasedlevelbutunmodifiedactivityofhepaticHMG‐

CoA reductase, which points to a post-translational regulation of this enzyme and favors an extrahepatic origin of hypercholesterolemia in these animals. Cholesterol 7-alpha-hydroxylase is up-regulated as a compensatory

responsetohypercholesterolemia[160].

Aneffectonlipidmetabolismsimilartothatinanalbuminemiaisobservedinpatientswithnephroticsyndrome,inwhomtheconcentrationofcirculatingalbuminisloweredbyincreasedlossinsteadofbyreducedsynthesis.Ata

difference fromanalbuminemia,nephroticsyndromecausesseveredeficiencies inLDLreceptor,HDLreceptor,andLCAT.Thehepaticresponse inconditionscharacterizedbyreducedplasmacolloidosmoticpressure (hypo-oncotic

states)involvesincreasedlevelsofmRNAsencodingagroupofsecretedproteins,includingthenegativeAPPsalbumin,transferrinandapoA-I,andthepositiveAPRfibrinogen.LevelsofmRNAsencodingnegativeAPRsandfibrinogen

correlatewithoneanother,suggestingthattheyarecoordinatelycontrolled.ThetranscriptionfactorsEGRF-1andHNF-4seemtobeinvolvedinthisregulation[161].

Animal models of primary hypertriglyceridemia (apoCIII transgenic mice) and hypercholesterolemia (LDL receptor knockout mice) show higher susceptibility to Ca2 +-induced inner mitochondrial membrane permeability

transition(MPT),aprocessthatcanleadtocelldeathas it is followedbymitochondrialenergyfailureandthereleaseofpro-apoptoticfactors.Bothmenandmicewithanalbuminemia,whopresenthighlevelsoftriglyceridesand

Fig.10Majorelementsofextracellularproteostasis.Proteinsundergorigorousqualitycontrolbeforetheyaresecreted,generallyinanativelyfoldedstate.Onceintheextracellularenvironment,theyencounteravarietyofstressorsthatcancausethemto

partiallyunfoldandpopulatemisfoldedstates.Misfoldedproteinscanaggregateintosolubleoligomersandsubsequentlyintoinsolublefibrillaroramorphousaggregates.Extracellularchaperonesformstablecomplexeswithmisfoldedproteinspecies,

includingmisfoldedmonomersandoligomers.Thesecomplexesmaintainmisfoldedproteinsinsolutionandfacilitatetheirclearancefromextracellularfluidsviareceptor-mediatedendocytosisandsubsequentdegradationinlysosomes.Insomecases,

misfolded,modified,oraggregatedproteinscanalsobeclearedviareceptor-mediatedendocytosiswithouttheinvolvementofextracellularchaperones;largeinsolubleaggregatesmustbephagocytosed.Furthermore,extracellularproteases,suchasplasmin,

maybeactivatedbyproteinaggregatesandsubsequentlydegradethem.(From[155]).

alt-text:Fig.10

i

cholesterol,alsopresentabnormalmitochondrialfunctions.ThesearchforputativemediatorsofMPTsensitizationinmicerevealeda30%increaseintheexpressionoftheapoptoticregulatorcyclophilinDtogetherwitha20%decrease

inmitochondrialnitrosothiolcontent:nearly80%oftotalplasmareducedthiolgroupsandnitrosothiolsresideinalbuminmolecules[162].

Ourinvestigationonanimalscomparedmice,inwhichmurineapoA-Ihadbeenknockedout,toderivedstrainsmadetransgenicforhumanapoA-I,humanapoA-II,orboth[163](Fig.11).TheabsoluteabundanceofhumanapoA-I

wasnotinfluencedbythepresenceapoA-II,converselythepresenceoftheapoA-ItransgeneresultedinastatisticallysignificantincreaseinexpressionoftheapoA-IItransgene.Severalproteinsnotinvolvedinlipidmetabolismwere

alsoinfluencedbythepresenceofhumanapolipoproteintransgenes,withapoA-IexertingamuchgreatereffectthanapoA-II.Alltheproteinswhoselevelsaremodulatedbytheexpressionofhumanapolipoproteintransgenes‐—alpha-

1-acidglycoprotein,alpha-1-macroglobulin,esterase,haptoglobinandcontrapsin‐—havecytokine-responsiveelementsintheirpromoterregionandareknowntobeinvolvedinacutephasereactions.TheinhibitoryeffectofHDLon

cytokinesecretion[164–166]couldbetheexplanationforourobservationsontheeffectofexpressionofapolipoproteintransgenes.

4.5.4.5ImmunoglobulinsImmunoglobulinsformalargeandheterogeneousclassofplasma/serumproteins;becauseoftheiroriginfromdifferentiatedBcellsoftheadaptiveimmunesystem(insteadofliver)andbecauseoftheirroleasantibodiesthey

deserveacommentontheirowninadditiontounsystematicmentioninprevioussectionsofthisreview.

Gamma,kappaandlambda-typechainshavemorealkalinepIsthanmostothermajorplasma/serumproteins;theirremovalisrelevantforreducingoverallproteinloadmorethanforpreventingspotcomigration.Theaffinityof

manyIgclassesforproteinAorGhasbeenrepeatedlyexploitedfortheirsubtraction;somecommercialproductsfeaturemixesofresin-immobilizedproteinAorGwitharesin-immobilizedtr iazinedye(e.g.fromBioRad)oraresin-

immobilizedanti-albuminantibody (e.g. fromBiosystemsorSartorius) for theremovalof the twomostabundantproteins/proteinclasses inplasma/serumnamelyalbuminand Igs. Immunoglobulin precipitationat50%ammonium

sulfatesaturationisaclassicalprocedureintheproductionofreagentantibodies,atbothlaboratoryandindustryscale[167].

Proteomicprocedureshavebeensystematicallyappliedtothefieldofimmunologyinordertorecognize‐—inbacteria(e.g.Neisseriameningitidis[168]orChlamydiatrachomatis[169]),fungi(e.g.Aspergillusfumigatus[170]),

(pluricellular)parasites(e.g.Amblyommaamericanum[171]),orcancercells(e.g.prostatecancer[172])–—againstwhichnon-selfproteinsthehostorganismraisesthemostintensereaction.Thisisachievedbyseparatingby2DEthe

proteomeofthepathogensorsomerelevantsubproteomee.g.membraneproteins,andbyusingserumfromaffectedpatientsasasourceofprimaryantibodiesinimmunoblotting.Proteinsinrepeatedlydetectedspotsareidentified

throughMSprocedures;acknowledgedimmunodominantantigensaretakenastargetsofpreventiveand/ortherapeuticvaccines.Thiswayofproceedinghasbeengiventhenameofimmunoproteomicsandfocusesonthecomplementof

allidentifiedantigensinapathogen[173].Alternativeexperimentalapproachestoelectrophoresisimplytheuseofproteinarraysofthepathogen.

Thesameapproachisabletorecognizestransversalandlongitudinalvariationsintheimmunoproteome.Thelatterofcoursemosteasilyapplytosituationsinwhichapathogenexistsindifferentformswithtimesuchasthe

alterationofimmunoproteomeprofileofEchinococcusgranulosushydatidfluidwithprogressionofcysticechinococcosis[174].Theformerallowforthediscriminationofmoreandlesssevereoutcomeofagivencondition.Forinstance,

Fig.11VariationinproteinexpressioninmicetransgenicforhumanapolipoproteinA-I(apoAI),forhumanapolipoproteinA-II(apoAII)andforboth(apoAI/II),andintheirgeneticbackground(miceKOformurineapolipoproteinA-I,KO).(From[163]).

alt-text:Fig.11

g

y

s

theevolutionoftheinfectionandthereplicativestateofthepathogenappeartoaffecttheimmunoproteomeofMycobacteriumtuberculosis,shiftingtargetfrommembrane-associatedtoextracellularproteinsofthebacillusduringthe

activephaseofthedisease.Overall,theimmuneresponsecorrelateswithbacillaryburdenbuttargetpreferencesvaryamongpatients;diagnosisandmonitoringtreatmentoutcomemaybenefitfro mtheuseoftheidentifiedbiomarkers

[175].Finally,Helicobacterpyloriinfectionmayprogressintoulcerorintocancer,apredisposingconditiontothelatterbeingatrophicbodygastritis.DifferentinvestigationshavebeenabletoassociateHelicobacterimmunoproteome

withthedifferentconditions.Threeproteinshavebeenidentifiedasduodenalulcer-related[176];serafromatrophicbodygastritisandgastriccancerpatientsdifferentiallyrecognize17Helicobacterspots[177].

Throughdifferencesintheimmunetargetingtheinvestigationsaboveexemplifyontheinteractionbetweenselfandnon-selfduringinfection/infestation.Tissotandcolleagueshaveevaluatedtheapplicationoftypicalproteomic

proceduressuchas2DEtothediagnosticsoflymphoidtissuecancers.AclassificationofthemonoclonalIgGsofmultiplemyelomasbasedonlyontheirelectrophoreticpropertiesturnedoutnottobepossiblebecauseoftheextreme

diversityamongthepathologicalheavyandlightchains[178].Tothispurpose,theoverallperformanceofhigh-resolutionelectrophoresiswaslowerthanthatofzonalelectrophoresiswithimmunofixation.Howevertheoppositeistruein

selectedcases,forinstancewiththecharacterizationofrareIgDmyeloma[179]orwiththeclassificationofcryoglobulins[180].

Theroleofimmunoglobulinglycosylationhaslongbeenstudied[181,182].Asreviewedin[183],currentinvestigationareasinclude:theglycosylationsitesofIgE,IgM,IgD,IgE,IgA,andIgG;howglycanscanencode"“self"

identitybyfunctioningaseitherdangerassociatedmolecularpatternsorself-associatedmolecularpatterns(SAMPs);theroleofglycansasmarkersofproteinintegrityandage;howtheglycocalyxcandictatethemigrationpatternof

immunecells;andhowthecombinationofFcN-glycansandIgisotypedictatetheeffectorfunctionofimmunoglobulins.Recently,ithasbeendemonstratedthatasingleglycanonIgEisindispensableforinitiationofanaphylaxis[184]

In immunoglobulin A nephropathy, galactose-deficient immunoglobulin A (non-Italics)1 is recognized by unique autoantibodies, resulting in the formation of pathogenic immune complexes that ultimately deposit in the glomerular

mesangiumandinducerenalinjury.NewapproachesusingMShaveprovidedinsightatpathologicalIgA1glycosylationasconnectedwithabnormalexpressionandactivityofseveralkeyglycosyltransferase[185].

5.5ConclusionsKnowledge of the properties of proteins andunderstanding of theirworking is steadily growing; both shared features and individual traits come anewon focus. The interest is currentlymore on cellular

componentsthanonplasma/serumproteome,yetrecentadvancements–—dutifullyrecordedinthisreview–—havebeenachievedalsooncirculatingproteins.Wemaintainthatnotoverlookinginformationprovided

byquali-andquantitativechangesinmajorserum/plasmaproteinsmayeffectivelycontributetoourawarenessofthecross-influencesbetweenvariouslevelsofintegrationintheorganism.

Supplementarydatatothisarticlecanbefoundonlineathttp://dx.doi.org/10.1016/j.jprot.2016.04.002.

AcknowledgmentsSomeoftheearlyachievementsofEGandIMhavetodowithalbuminandmuchoftheworkinproteomicstheysharewithIEdealswithbiologicalfluids.Onthisbackgroundagreatdealofthetopicscovered

bythisreviewarelinkedtoourday-by-dayexperiencemorethantoanassessmentoftheliterature.

Writing our previous review had Mahler symphonies as soundtrack. This time the most often played piece was Händel’'s Messiah – — the number of encore compensating the gloom of the missed live

performance.But,sinceourinterestformusicisallbutnarrow,thetitleofthereviewpositivelycomesfromamuchmorerecentcomposition.

BibliographyReferences(Generalcomment:onseveralpages(inthepdfview)thetwotextcolumnsarenotofequallength,e.g.pages2,4,8,12,13...,notalwaysforobviousreasons.)[1]NLN.L.AndersonandNGN.G.Anderson,Thehumanplasmaproteome:history,character,anddiagnosticprospects,Mol.Cell.Proteomics1,2002,845–867.

[2]S.Subramanian,Dye-ligandaffinitychromatography:theinteractionofCibacronBblueF3GAwithproteinsandenzymes,CRCCrit.Rev.Bioeng.16,1984,169–205.

[3]AA.DenizliandEE.Piskin,Dye-ligandaffinitysystems,J.Biochem.Biophys.Methods49,2001,391–416.

[4]EE.GianazzaandPP.Arnaud,Ageneralmethodforthefractionationofplasmaproteins:Pseudo-ligandaffinitychromatographyonimmobilizedCibacronBblueF3-G,Biochem.J.201,1982,129–136.

[5]EE.GianazzaandPP.Arnaud,ChromatographyofplasmaproteinsonimmobilizedCibacronBblueF3-GA.Mechanismofthemolecularinteraction,Biochem.J.203,1982,637–641.

[6]II.MillerandMM.Gemeiner,AnelectrophoreticstudyoninteractionsofalbuminsofdifferentspecieswithimmobilizedCibacronBblueF3GA,Electrophoresis19,1998,2506–2514.

[7]FF.DiGirolamoandPGP.G.Righetti,Plasmaproteomicsforbiomarkerdiscovery:astudyinblue,Electrophoresis32,2011,3638–3644.

r

[8]TMT.M.Andacht,BGB.G.Pantazides,BSB.S.Crow,AA.Fidder,DD.Noort,JDJ.D.Thomas,etal.,Anenhancedthroughputmethodforquantificationofsulfurmustardadductstohumanserumalbuminviaisotope

dilutiontandemmassspectrometry,J.Anal.Toxicol.38,2014,8–15.

[9]GAG.A.Hebert,PLP.L.PelhamandBB.Pittman,Determinationoftheoptimalammoniumsulfateconcentrationforthefractionationofrabbit,sheep,horse,andgoatantisera,Appl.Microbiol.25,1973,26–36.

[10]AA.MahnandMM.Ismail,Depletionofhighlyabundantproteinsinbloodplasmabyammoniumsulfateprecipitationfor2D-PAGEanalysis,J.Chromatogr.BAnal.Technol.Biomed.LifeSci.879,2011,3645–3648.

[11]SS.Saha,SS.Halder,DD.Bhattacharya,DD.BanerjeeandAA.Chakrabarti,FractionalPprecipitationofPplasmaPproteomebyAammoniumSsulfate:CcaseSstudiesinLeukemiaandThalassemia,J.Proteomics

Bioinform.5,2012,163–171.

[12]CC.Fernandez-Costa,VV.Calamia,PP.Fernandez-Puente,JLJ.L.Capelo-Martinez,CC.Ruiz-RomeroandFJF.J.Blanco,Sequentialdepletionofhumanserumforthesearchofosteoarthritisbiomarkers,Proteome

Sci.10,2012,55.

[13]AKA.K.Henning,DD.Albrecht,KK.Riedel,TCT.C.MettenleiterandAA.Karger,Analternativemethodforserumproteindepletion/enrichmentbyprecipitationatmildlyacidicpHvaluesandlowionicstrength,

Proteomics15,2015,1935–1940.

[14]EE.Gianazza,RR.Wait,SS.Begum,II.Eberini,MM.Campagnoli,SS.Labò,etal.,Mappingthe5‐–50kDafractionofhumanamnioticfluidproteinsby2DEandESI-MS,ProteomicsClin.Appl.1,2007,167–175.

[15]EE.Gianazza,AA.GueriniRocco,AA.MarchettoandLL.Vergani,ImmobilizedpHgradientswithelectrodicplateaus(andotherunusualproceduresfor2DE),Electrophoresis28,2007,2953–2956.

[16]RSR.S.Tirumalai,KCK.C.Chan,DAD.A.Prieto,HJH.J.Issaq,TPT.P.ConradsandTDT.D.Veenstra,Characterizationofthelowmolecularweighthumanserumproteome,Mol.Cell.Proteomics2,2003,1096–1103.

[17]RLR.L.Gundry,QQ.Fu,CAC.A.Jelinek,JEJ.E.VanEykandRJR.J.Cotter,Investigationofanalbumin-enrichedfractionofhumanserumanditsalbuminome,ProteomicsClin.Appl.1,2007,73–88.

[18]RJR.J.Holewinski,ZZ.Jin,MJM.J.Powell,MDM.D.MaustandJEJ.E.VanEyk,Afastandreproduciblemethodforalbuminisolationanddepletionfromserumandcerebrospinalfluid,Proteomics13,2013,743–750.

[19]DD.Scumaci,MM.Gaspari,MM.Saccomanno,GG.Argiro,BB.Quaresima,CMC.M.Faniello,etal.,Assessmentofanadhocprocedureforisolationandcharacterizationofhumanalbuminome,Anal.

Biochem.418,2011,161–163.

[20]CMC.M.Camaggi,EE.Zavatto,LL.Gramantieri,VV.Camaggi,EE.Strocchi,RR.Righini,etal.,Serumalbumin-boundproteomicsignatureforearlydetectionandstagingofhepatocarcinoma:samplevariabilityand

dataclassification,Clin.Chem.Lab.Med.48,2010,1319–1326.

[21]AA.Mahn,AA.Reyes,MM.Zamorano,WW.CifuentesandMM.Ismail,Depletionofhighlyabundantproteinsinbloodplasmabyhydrophobicinteractionchromatographyforproteomicanalysis,J.Chromatogr.B

Anal.Technol.Biomed.LifeSci.878,2010,1038–1044.

[22]RR.Schade,EGE.G.Calzado,RR.Sarmiento,PAP.A.Chacana,JJ.Porankiewicz-AsplundandHRH.R.Terzolo,Chickeneggyolkantibodies(IgY-technology):areviewofprogressinproductionanduseinresearchand

humanandveterinarymedicine,Altern.Lab.Anim.Altern.Lab.Anim33,2005,129–154.

[23]SHS.H.Tan,AA.Mohamedali,AA.KapurandMSM.S.Baker,Ultradepletionofhumanplasmausingchickenantibodies:aproofofconceptstudy,J.ProteomeRes.12,2013,2399–2413.

[24]MMM.M.HarmsenandHJH.J.DeHaard,Properties,production,andapplicationsofcamelidsingle-domainantibodyfragments,Appl.Microbiol.Biotechnol.77,2007,13–22.

[25]MM.Kullolli,JJ.Warren,MM.ArampatzidouandSJS.J.Pitteri,Performanceevaluationofaffinityligandsfordepletionofabundantplasmaproteins,J.Chromatogr.BAnal.Technol.Biomed.LifeSci.939,2013,10–16.

[26]BAB.A.Chromy,ADA.D.Gonzales,JJ.Perkins,MWM.W.Choi,MHM.H.Corzett,BCB.C.Chang,etal.,Proteomicanalysisofhumanserumbytwo-dimensionaldifferentialgelelectrophoresisafterdepletionofhigh-

abundantproteins,J.ProteomeRes.3,2004,1120–1127.

[27]KK.Bjorhall,TT.MiliotisandPP.Davidsson,Comparisonofdifferentdepletionstrategiesforimprovedresolutioninproteomicanalysisofhumanserumsamples,Proteomics5,2005,307–317.

[28]SS.Roche,LL.Tiers,MM.Provansal,MM.Seveno,MTM.T.Piva,PP.Jouin,etal.,Depletionofone,six,twelveortwentymajorbloodproteinsbeforeproteomicanalysis:themorethebetter?,J.Proteome72,2009,

945–951.

[29]MPM.P.Smith,SLS.L.Wood,AA.Zougman,JTJ.T.Ho,JJ.Peng,DD.Jackson,etal.,Asystematicanalysisoftheeffectsofincreasingdegreesofserumimmunodepletionintermsofdepthofcoverageandotherkey

aspectsintop-downandbottom-upproteomicanalyses,Proteomics11,2011,2222–2235.

[30]NN.Macdonald,MM.Cumberbatch,MM.Singh,JGJ.G.Moggs,GG.Orphanides,RJR.J.Dearman,etal.,Proteomicanalysisofsuctionblisterfluidisolatedfromhumanskin,Clin.Exp.Dermatol.31,2006,445–448.

[31]ACA.C.Muller,FPF.P.Breitwieser,HH.Fischer,CC.Schuster,OO.Brandt,JJ.Colinge,etal.,Acomparativeproteomicstudyofhumanskinsuctionblisterfluidfromhealthyindividualsusingimmunodepletionand

iTRAQlabeling,J.ProteomeRes.11,2012,3715–3727.

[32]BBB.B.Patel,CAC.A.Barrero,AA.Braverman,PDP.D.Kim,KAK.A.Jones,DED.E.Chen,etal.,Assessmentoftwoimmunodepletionmethods:off-targeteffectsandvariationsinimmunodepletionefficiencymay

confoundplasmaproteomics,J.ProteomeRes.11,2012,5947–5958.

[33]PGP.G.Righetti,EE.Boschetti,LL.LomasandAA.Citterio,ProteinEequalizerTtechnology::thequestfora"“democraticproteome"”,Proteomics6,2006,3980–3992.

[34]PGP.G.Righetti,EE.BoschettiandGG.Candiano,MarkTwain:howtofathomthedepthofyourpetproteome,J.Proteome75,2012,4783–4791.

[35]PGP.G.Righetti,GG.Candiano,AA.CitterioandEE.Boschetti,Combinatorialpeptideligandlibrariesasa"“TrojanHhorse"”indeepdiscoveryproteomics,Anal.Chem.87,2015,293–305.

[36]EE.Fasoli,AA.Farinazzo,CJC.J.Sun,AVA.V.Kravchuk,LL.Guerrier,FF.Fortis,etal.,Interactionamongproteinsandpeptidelibrariesinproteomeanalysis:pHinvolvementforalargercaptureofspecies,J.

Proteome73,2010,733–742.

[37]GG.Candiano,LL.Santucci,MM.Bruschi,AA.Petretto,CC.D'Ambrosio,AA.Scaloni,etal.,"“Cheek-to-cheek"”urinaryproteomeprofilingviacombinatorialpeptideligandlibraries:Anovel,unexpectedelution

system,J.Proteome75,2012,796–805.

[38]GG.Candiano,VV.Dimuccio,MM.Bruschi,LL.Santucci,RR.Gusmano,EE.Boschetti,etal.,Combinatorialpeptideligandlibrariesforurineproteomeanalysis:investigationofdifferentelutionsystems,

Electrophoresis30,2009,2405–2411.

[39]FF.DiGirolamo,EE.Boschetti,MCM.C.Chung,FF.GuadagniandPGP.G.Righetti,"“Proteomineering"”ornot?Thedebateonbiomarkerdiscoveryinseracontinues,J.Proteome74,2011,589–594.

[40]EE.Bellei,EE.Monari,SS.Bergamini,TT.OzbenandAA.Tomasi,Optimizingproteinrecoveryyieldfromserumsamplestreatedwithbeadstechnology,Electrophoresis32,2011,1414–1421.

[41]LL.Sennels,MM.Salek,LL.Lomas,EE.Boschetti,PGP.G.RighettiandJJ.Rappsilber,Proteomicanalysisofhumanbloodserumusingpeptidelibrarybeads,J.ProteomeRes.6,2007,4055–4062.

[42]CC.Simo,AA.Bachi,AA.Cattaneo,LL.Guerrier,FF.Fortis,EE.Boschetti,etal.,Performanceofcombinatorialpeptidelibrariesincapturingthelow-abundanceproteomeofredbloodcells.1.Behaviorofmono-to

hexapeptides,Anal.Chem.80,2008,3547–3556.

[43]FF.DiGirolamo,PGP.G.Righetti,MM.Soste,YY.FengandPP.Picotti,Reproducibilityofcombinatorialpeptideligandlibrariesforproteomecaptureevaluatedbyselectedreactionmonitoring,J.Proteome89,2013,

215–226.

[44]CC.Sihlbom,II.Kanmert,HH.BahrandPP.Davidsson,EvaluationofthecombinationofbeadtechnologywithSELDI-TOF-MSand2-DDIGEfordetectionofplasmaproteins,J.ProteomeRes.7,2008,4191–4198.

[45]CC.Fernandez,HMH.M.Santos,CC.Ruiz-Romero,FJF.J.BlancoandJLJ.L.Capelo-Martinez,Acomparisonofdepletionversusequalizationforreducinghigh-abundanceproteinsinhumanserum,

Electrophoresis32,2011,2966–2974.

[46]SS.SelvarajuandZZ.ElRassi,Reductionofproteinconcentrationrangedifferencefollowedbymulticolumnfractionationpriorto2-DEandLC-MS/MSprofilingofserumproteins,Electrophoresis32,2011,

674–685.

[47]GG.Zhu,PP.Zhao,NN.Deng,DD.Tao,LL.Sun,ZZ.Liang,etal.,SinglechainvariablefragmentdisplayingM13phagelibraryfunctionalizedmagneticmicrosphere-basedproteinequalizerforhumanserumprotein

analysis,Anal.Chem.84,2012,7633–7637.

[48]EE.Mouton-Barbosa,FF.Roux-Dalvai,DD.Bouyssie,FF.Berger,EE.Schmidt,PGP.G.Righetti,etal.,In-depthexplorationofcerebrospinalfluidbycombiningpeptideligandlibrarytreatmentandlabel-freeprotein

quantification,Mol.Cell.Proteomics9,2010,1006–1021.

[49]YY.Liao,RR.Alvarado,BB.PhinneyandBB.Lonnerdal,Proteomiccharacterizationofhumanmilkwheyproteinsduringatwelve-monthlactationperiod,J.ProteomeRes.10,2011,1746–1754.

[50]CEC.E.Molinari,YSY.S.Casadio,BTB.T.Hartmann,AA.Livk,SS.Bringans,PGP.G.Arthur,etal.,Proteomemappingofhumanskimmilkproteinsintermandpretermmilk,J.ProteomeRes.11,2012,1696–1714.

[51]PP.Jagtap,SS.Bandhakavi,LL.Higgins,TT.McGowan,RR.Sa,MDM.D.Stone,etal.,WorkflowforanalysisofhighmassaccuracysalivarydatasetusingMaxQuantandProteinPilotsearchalgorithm,

Proteomics12,2012,1726–1730.

[52]ADA.D.Rolland,RR.Lavigne,CC.Dauly,PP.Calvel,CC.Kervarrec,TT.Freour,etal.,Identificationofgenitaltractmarkersinthehumanseminalplasmausinganintegrativegenomicsapproach,Hum.

Reprod.28,2013,199–209.

[53]AMA.M.Lichtenauer,RR.Herzog,SS.Tarantino,CC.AufrichtandKK.Kratochwill,EqualizertechnologyfollowedbyDIGE-basedproteomicsfordetectionofcellularproteinsinartificialperitonealdialysis

effluents,Electrophoresis35,2014,1387–1394.

[54]MJM.J.Peffers,BB.McDermott,PDP.D.CleggandCMC.M.Riggs,ComprehensivepProteinpProfilingofsSynovialfFluidinoOsteoarthritisfFollowingpProteineEqualization,Osteoarthr.Cartil.23Vol.23,2015,

OsteoarthritisResearchSociety,1204–1213.

[55]FF.Fakelman,KK.Felix,MWM.W.BuchlerandJJ.Werner,Newpre-analyticalapproachforthedeepproteomeanalysisofserafrompancreatitisandpancreascancerpatients,Arch.Physiol.Biochem.116,2010,

208–217.

[56]JSJ.S.Au,WCW.C.Cho,TTT.T.Yip,CC.Yip,HH.Zhu,WWW.W.Leung,etal.,Deepproteomeprofilingofserafromnever-smokedlungcancerpatients,Biochem.Pharmacol.61,2007,570–577.

[57]EE.Monari,CC.Casali,AA.Cuoghi,JJ.Nesci,EE.Bellei,SS.Bergamini,etal.,Enrichedseraproteinprofilingfordetectionofnon-smallcelllungcancerbiomarkers,ProteinSci.9,2011,55.

[58]EE.Milan,CC.Lazzari,SS.Anand,II.Floriani,VV.Torri,CC.Sorlini,etal.,SAA1isover-expressedinplasmaofnonsmallcelllungcancerpatientswithpooroutcomeaftertreatmentwithepidermalgrowthfactor

receptortyrosine-kinaseinhibitors,J.Proteome76,2012,91–101,SpecNo..

[59]CC.Marrocco,SS.Rinalducci,AA.Mohamadkhani,GMG.M.D'AmiciandLL.Zolla,Plasmagelsolinprotein:acandidatebiomarkerforhepatitisB-associatedlivercirrhosisidentifiedbyproteomicapproach,Blood

Transfus.8Suppl.3,2010,s105–s112.

[60]VV.Shetty,PP.Jain,ZZ.Nickens,GG.Sinnathamby,AA.MehtaandRR.Philip,InvestigationofplasmabiomarkersinHIV-1/HCVmono-andcoinfectedindividualsbymultiplexiTRAQquantitativeproteomics,

OMICS15,2011,705–717.

[61]AA.Fekkar,CC.Pionneau,JYJ.Y.Brossas,CC.Marinach-Patrice,GG.Snounou,MM.Brock,etal.,DIGEenablesthedetectionofaputativeserumbiomarkeroffungalorigininamousemodelofinvasive

aspergillosis,J.Proteome75,2012,2536–2549.

[62]MDM.D.Ruis-Gonzalez,MDM.D.Canete,JLJ.L.Gomez-Chaparro,NN.Abril,RR.CaneteandJJ.Lopez-Barea,Alterationsofproteinexpressioninserumofinfantswithintrauterinegrowthrestrictionanddifferent

gestationalages,J.Proteome119,2015,169–182.

[63]EE.Balfoussia,KK.Skenderi,MM.Tsironi,AKA.K.Anagnostopoulos,NN.Parthimos,KK.Vougas,etal.,Aproteomicstudyofplasmaproteinchangesunderextremephysicalstress,J.Proteome98,2014,1–14.

[64]LL.Soler,II.Miller,KK.Hummel,EE.Razzazi-Fazeli,FF.Jessen,DD.Escribano,etal.,Growthpromotioninpigsbyoxytetracyclinecoincideswithdownregulationofseruminflammatoryparametersandof

hibernation-associatedproteinHP-27,Electrophoresis2016.

[65]APA.P.Drabovich,EE.Martinez-MorilloandEPE.P.Diamandis,Towardanintegratedpipelineforproteinbiomarkerdevelopment,Biochim.Biophys.Acta2015,1854,677–686.

[66]S.Hanash,HUPOinitiativesrelevanttoclinicalproteomics,Mol.Cell.Proteomics3,2004,298–301.

[67]MM.Uhlen,EE.Bjorling,CC.Agaton,CAC.A.Szigyarto,BB.Amini,EE.Andersen,etal.,Ahumanproteinatlasfornormalandcancertissuesbasedonantibodyproteomics,Mol.Cell.Proteomics4,2005,1920–1932

[68]MM.Uhlen,PP.Oksvold,LL.Fagerberg,EE.Lundberg,KK.Jonasson,MM.Forsberg,etal.,Towardsaknowledge-basedHhumanPproteinAatlas,Nat.Biotechnol.28,2010,1248–1250.

[69]MM.Uhlen,LL.Fagerberg,BMB.M.Hallstrom,CC.Lindskog,PP.Oksvold,AA.Mardinoglu,etal.,Proteomics.Tissue-basedmapofthehumanproteome,Science347,2015,1260419.

[70]GJG.J.vandenBemd,JJ.Krijgsveld,TMT.M.Luider,ALA.L.vanRijswijk,JAJ.A.DemmersandGG.Jenster,Massspectrometricidentificationofhumanprostatecancer-derivedproteinsinserumofxenograft-bearing

mice,Mol.Cell.Proteomics5,2006,1830–1839.

[71]YY.Li,ZYZ.Y.Tang,BB.Tian,SLS.L.Ye,LXL.X.Qin,QQ.Xue,etal.,SerumCYFRA21-1levelreflectshepatocellularcarcinomametastasis:studyinnudemicemodelandclinicalpatients,J.CancerRes.Clin.

Oncol.132,2006,515–520.

[72]DD.Cervi,TTT.T.Yip,NN.Bhattacharya,VNV.N.Podust,JJ.Peterson,AA.Abou-Slaybi,etal.,Platelet-associatedPF-4asabiomarkerofearlytumorgrowth,Blood111,2008,1201–1207.

[73]PKP.K.Chong,HH.Lee,JJ.Zhou,SCS.C.Liu,MCM.C.Loh,JBJ.B.So,etal.,ReducedplasmaAPOA1levelisassociatedwithgastrictumorgrowthinMKN45mousexenograftmodel,J.Proteome73,2010,1632–1640.

[74]RR.Balys,MM.Alaoui-Jamali,MM.Hier,MM.Black,GG.Domanowski,LL.Rochon,etal.,Clinicallyrelevantoralcancermodelforserumproteomiceavesdroppingonthetumourmicroenvironment,J.

Otolaryngol.35,2006,157–166.

[75]KK.Bijian,AMA.M.Mlynarek,RLR.L.Balys,SS.Jie,YY.Xu,MPM.P.Hier,etal.,Serumproteomicapproachfortheidentificationofserumbiomarkerscontributedbyoralsquamouscellcarcinomaandhosttissue

microenvironment,J.ProteomeRes.8,2009,2173–2185.

[76]CCC.C.Wu,PHP.H.Peng,YTY.T.Chang,YSY.S.Huang,KPK.P.Chang,SPS.P.Hao,etal.,IdentificationofpotentialserummarkersfornasopharyngealcarcinomafromaxenograftedmousemodelusingCy-dye

labelingcombinedwiththree-dimensionalfractionation,Proteomics8,2008,3605–3620.

[77]YY.He,XX.Wu,XX.Liu,GG.YanandCC.Xu,LC-MS/MSanalysisofovariancancermetastasis-relatedproteinsusinganudemousemodel:14-3-3zetaasacandidatebiomarker,J.ProteomeRes.9,2010,

6180–6190.

[78]HYH.Y.Tang,LAL.A.Beer,TT.Chang-Wong,RR.Hammond,PP.Gimotty,GG.Coukos,etal.,Axenograftmousemodelcoupledwithin-depthplasmaproteomeanalysisfacilitatesidentificationofnovelserum

biomarkersforhumanovariancancer,J.ProteomeRes.11,2012,678–691.

[79]VMV.M.Faca,KSK.S.Song,HH.Wang,QQ.Zhang,ALA.L.Krasnoselsky,LFL.F.Newcomb,etal.,Amousetohumansearchforplasmaproteomechangesassociatedwithpancreatictumordevelopment,PLoS

Med.5,2008,e123.

[80]SJS.J.Pitteri,LL.JeBailey,VMV.M.Faca,JDJ.D.Thorpe,MAM.A.Silva,RCR.C.Ireton,etal.,Integratedproteomicanalysisofhumancancercellsandplasmafromtumorbearingmiceforovariancancerbiomarker

discovery,PLoSONEPLoSOne4,2009,e7916.

[81]HYH.Y.Tang,LAL.A.Beer,JLJ.L.Tanyi,RR.Zhang,QQ.LiuandDWD.W.Speicher,Proteinisoform-specificvalidationdefinesmultiplechlorideintracellularchannelandtropomyosinisoformsasserological

biomarkersofovariancancer,J.Proteome89,2013,165–178.

[82]BB.Bjellqvist,KK.Ek,PGP.G.Righetti,EE.Gianazza,AA.Görg,RR.Westermeier,etal.,IsoelectricfocusinginimmobilizedpHgradients:Principles,methodologyandsomeapplications,J.Biochem.Biophys.

Methods6,1982,317–339.

[83]EE.Gianazza,SS.Astrua-Testori,PP.GiaconandPGP.G.Righetti,Animprovedprotocolfor2DmapsofserumproteinswithimmobilizedpHgradientsinthefirstdimension,Electrophoresis6,1985,332–339.

[84]RR.Wait,SS.Begum,DD.Brambilla,AMA.M.Carabelli,FF.Conserva,AA.RoccoGuerini,etal.,Redoxoptionsintwo-dimensionalelectrophoresis,AminoAcids28,2005,239–272.

[85]SS.Tsuji,HH.Kato,YY.Matsuoka,TT.Fukushima,II.Nanjoh,TT.Amano,etal.,Phylogeneticalandontogeneticalstudiesonthemolecularweightheterogeneityofbovineserumtransferrin,Biochem.

Genet.22,1984,1127–1143.

[86]RWR.W.EvansandJJ.Williams,Theelectrophoresisoftransferrinsinurea/polyacrylamidegels,Biochem.J.189,1980,541–546.

[87]HH.Yamashita,TT.NakatsukaandMM.Hirose,Structuralandfunctionalcharacteristicsofpartiallydisulfide-reducedintermediatesofovotransferrinNlobe:Cystinelocalizationbyindirectend-labelingapproach

andimplicationsforthereductionpathway,J.Biol.Chem.270,1995,29806–29812.

[88]II.Miller,PP.Haynes,MM.Gemeiner,RR.Aebersold,CC.Manzoni,MRM.R.Lovati,etal.,Proteinsofratserum:II.Influenceofsomebiologicalparametersonthe2-DEpattern,Electrophoresis19,1998,

1493–1500.

[89]RR.Wait,II.Miller,II.Eberini,FF.Cairoli,CC.Veronesi,MM.Battocchio,etal.,Strategiesforproteomicswithincompletelycharacterisedgenomes:TtheproteomeofBostaurusserum,Electrophoresis23,2002,

3418–3427.

[90]FF.Meier,BB.Seidel,GG.Geserick,PP.LutherandDD.Patzelt,HaptoglobintypisierungvonSerumprobenausgewählterSäugetieremittelsStärkegelelektrophorese,MhVet-Med.35,1980,617–620.

[91]PRP.R.Jungblut,HGH.G.Holzhutter,RR.ApweilerandHH.Schluter,Thespeciationoftheproteome,Chem.Cent.J.2,2008,16.

[92]AA.Mastrangelo,TT.Colasanti,CC.Barbati,AA.Pecani,DD.Sabatinelli,MM.Pendolino,etal.,TheRroleofPosttranslationalPproteinMmodificationsinRheumatologicalDdiseases:FfocusonRheumatoidArthritis,

J.Immunol.Res.2015,2015,712490.

[93]RJR.J.Ren,EBE.B.Dammer,GG.Wang,NTN.T.SeyfriedandAIA.I.Levey,Proteomicsofproteinpost-translationalmodificationsimplicatedinneurodegeneration,Transl.Neurosci.3,2014,23.

[94]HH.Fukami,HH.OkunishiandMM.Miyazaki,Chymase:itspathophysiologicalrolesandinhibitors,Curr.Pharm.Des.4,1998,439–453.

[95]GHG.H.Caughey,Mastcellproteasesasprotectiveandinflammatorymediators,Adv.Exp.Med.Biol.716,2011,212–234.

[96]WWW.W.Raymond,SWS.W.Ruggles,CSC.S.CraikandGHG.H.Caughey,Albuminisasubstrateofhumanchymase.Predictionbycombinatorialpeptidescreeninganddevelopmentofaselectiveinhibitorbasedon

thealbumincleavagesite,J.Biol.Chem.278,2003,34517–34524.

[97]DKD.K.Galanakis,Anticoagulantalbuminfragmentsthatbindtofibrinogen/fibrin:possibleimplications,Semin.Thromb.Hemost.18,1992,44–52.

[98]JAJ.A.Huntington,RJR.J.ReadandRWR.W.Carrell,Structureofaserpin-proteasecomplexshowsinhibitionbydeformation,Nature407,2000,923–926.

[99]BB.Gooptu,JAJ.A.DickensandDAD.A.Lomas,Themolecularandcellularpathologyofalpha(1)-antitrypsindeficiency,TrendsMol.Med.20,2014,116–127.

[100]CPC.P.Speer,AA.NinjoandMM.Gahr,Elastase-alpha1-proteinaseinhibitorinearlydiagnosisofneonatalsepticemia,J.Pediatr.108,1986,987–990.

[101]KK.Harm,KPK.P.BartfeldandTT.Mathew,Plasmaconcentrationsofgranulocyticelastase-alpha1-proteinaseinhibitorcomplexinpatientswithsevereheadinjury,multipletraumaorcerebralbleeding,Clin.

Biochem.22,1989,149–153.

[102]FKF.K.Tegtmeyer,JJ.Moller,AA.Richter,BB.WilkenandTT.Fischer,Plasmaconcentrationofelastase-alpha1-proteinaseinhibitorcomplexinsurfactant-treatedpretermneonateswithrespiratorydistress

syndrome,Eur.Respir.J.7,1994,260–264.

[103]NN.Ishikawa,HH.Kamitsuji,TT.Murakami,AA.NakayamaandYY.Umeki,Plasmalevelsofgranulocyteelastase-alpha1-proteinaseinhibitorcomplexinchildrenwithhemolyticuremicsyndromecausedby

verotoxin-producingEscherichiacoli,Pediatr.Int.42,2000,637–641.

[104]BB.Jirasakuldech,GCG.C.Schussler,MGM.G.Yap,HH.Drew,AA.JosephsonandJJ.Michl,Acharacteristicserpincleavageproductofthyroxine-bindingglobulinappearsinsepsissera,J.Clin.Endocrinol.

Metab.85,2000,3996–3999.

[105]BB.Afandi,GCG.C.Schussler,AHA.H.Arafeh,AA.Boutros,MGM.G.YapandAA.Finkelstein,Selectiveconsumptionofthyroxine-bindingglobulinduringcardiacbypasssurgery,Metabolism49,2000,270–274.

[106]II.Eberini,LL.Calabresi,RR.Wait,GG.Tedeschi,AA.Pirillo,LL.Puglisi,etal.,MacrophagemetalloproteinasesdegradeHDL-associatedapoA-IatboththeN-andtheC-terminus,Biochem.J.362,2002,627–634.

[107]II.Eberini,EE.Gianazza,LL.Breghi,SS.Klugmann,LL.Calabresi,MM.Gomaraschi,etal.,ApolipoproteinA-Ibreakdownisinducedbythrombolysisincoronarypatients,Ann.Med.39,2007,306–311.

[108]CC.Mizon,FF.Piva,VV.Queyrel,MM.Balduyck,EE.HachullaandJJ.Mizon,Urinarybikunindeterminationprovidesinsightintoproteinase/proteinaseinhibitorimbalanceinpatientswithinflammatorydiseases,

Clin.ChemistryLab.Med.40,2002,579–586.

[109]FAF.A.Robey,KK.Ohura,SS.Futaki,NN.Fujii,HH.Yajima,NN.Goldman,etal.,ProteolysisofhumanC-reactiveproteinproducespeptideswithpotentimmunomodulatingactivity,J.Biol.Chem.262,1987,

7053–7057.

[110]EJE.J.YavinandMM.Fridkin,PeptidesderivedfromhumanC-reactiveproteininhibittheenzymaticactivitiesofhumanleukocyteelastaseandcathepsinG:useofoverlappingpeptidesequencestoidentifya

uniqueinhibitor,J.Pept.Res.51,1998,282–289.

[111]DD.ElKebir,YY.Zhang,LAL.A.Potempa,YY.Wu,AA.FournierandJGJ.G.Filep,C-reactiveprotein-derivedpeptide201-206inhibitsneutrophiladhesiontoendothelialcellsandplateletsthroughCD32,J.Leukoc.

Biol.90,2011,1167–1175.

[112]BB.Kshirsagar,BB.WilsonandRCR.C.Wiggins,Polymericcomplexesandfragmentsofalbumininnormalhumanplasma,Clin.Chim.Acta143,1984,265–273.

[113]CC.Bazzi,CC.Petrini,VV.Rizza,EE.Sabadini,GG.Arrigo,AA.Beltrame,etal.,SDS-PAGEpatternsandpolymericalbumininproteinuriaoflupusglomerulonephritis,Clin.Nephrol.43,1995,96–103.

[114]AA.Alberti,PP.PontissoandGG.Realdi,Virusreceptorsforpolymerizedhumanalbumin:aprognosticmarkerinHBeAg-positivechronichepatitistypeB?,J.Med.Virol.10,1982,141–146.

[115]II.Mora,JCJ.C.Porres,FJF.J.Bartolome,CC.HernandezGuio,JJ.GutiezandVV.Carreno,ReceptorsforpolymerizedhumanserumalbuminandotherhepatitisBvirusmarkersduringacutehepatitisB‐‐–predictive

valueoftheoutcomeofthedisease,Hepato-Gastroenterology33,1986,250–254.

[116]DAD.A.LomasandRR.Mahadeva,Alpha1-antitrypsinpolymerizationandtheserpinopathies:pathobiologyandprospectsfortherapy,J.Clin.Invest.110,2002,1585–1590.

[117]RWR.W.CarrellandDAD.A.Lomas,Alpha1-antitrypsindeficiency‐‐—amodelforconformationaldiseases,N.Engl.J.Med.346,2002,45–53.

[118]DLD.L.DeMeo,MM.Banos,J.Perez,L.Tan,AFA.F.Barker,EE.Campbell,etal.,Circulatingpolymersofalpha-1-antitrypsininPiZZsubjects,Am.J.Respir.Crit.CareMed.185,2012,A4361.

[119]SS.Janciauskiene,RR.Dominaitiene,NHN.H.Sternby,EE.PiitulainenandSS.Eriksson,DetectionofcirculatingandendothelialcellpolymersofZandwildtypealpha1-antitrypsinbyamonoclonalantibody,J.

Biol.Chem.277,2002,26540–26546.

[120]GG.deJongandHGH.G.vanEijk,Microheterogeneityofhumanserumtransferrin:AbiologicalphenomenonstudiedbyisoelectricfocusinginimmobilizedpHgradients,Electrophoresis9,1988,589–598.

[121]GG.deJong,WLW.L.vanNoort,RAR.A.Feelders,CMC.M.deJeu-JasparsandHGH.G.vanEijk,Adaptationoftransferrinproteinandglycansynthesis,Clin.Chim.Acta212,1992,27–45.

[122]AA.Grigorian,HH.Mkhikian,CFC.F.Li,BLB.L.Newton,RWR.W.ZhouandMM.Demetriou,PathogenesisofmultiplesclerosisviaenvironmentalandgeneticdysregulationofN-glycosylation,Semin.

Immunopathol.34,2012,415–424.

[123]SS.Schedin-Weiss,BB.WinbladandLOL.O.Tjernberg,TheroleofproteinglycosylationinAlzheimerdisease,FEBSJ.281,2014,46–62.

[124]QQ.PuandCC.Yu,Glycosyltransferases,glycosylationandatherosclerosis,Glycoconj.J.31,2014,605–611.

[125]MNM.N.Christiansen,JJ.Chik,LL.Lee,MM.Anugraham,JLJ.L.AbrahamsandNHN.H.Packer,Cellsurfaceproteinglycosylationincancer,Proteomics14,2014,525–546.

[126]ZZ.MaandKK.Vosseller,CancermetabolismandelevatedO-GlcNAcinoncogenicsignaling,J.Biol.Chem.289,2014,34457–34465.

[127]RRR.R.DrakeandLL.Ball,GlycosylationandCancer,AdvancesinCancerResearch2015,Elsevier.

[128]SSS.S.PinhoandCAC.A.Reis,Glycosylationincancer:mechanismsandclinicalimplications,Nat.Rev.Cancer15,2015,540–555.

[129]SRS.R.Stowell,TT.JuandRDR.D.Cummings,Proteinglycosylationincancer,Annu.Rev.Pathol.10,2015,473–510.

[130]RR.Saldova,MRM.R.Wormald,RAR.A.DwekandPMP.M.Rudd,Glycosylationchangesonserumglycoproteinsinovariancancermaycontributetodiseasepathogenesis,Dis.Markers25,2008,219–232.

[131]SS.Hashimoto,TT.Asao,JJ.Takahashi,YY.Yagihashi,TT.Nishimura,ARA.R.Saniabadi,etal.,Alpha1-acidglycoproteinfucosylationasamarkerofcarcinomaprogressionandprognosis,Cancer101,2004,

2825–2836.

[132]SYS.Y.Lin,YYY.Y.Chen,YYY.Y.Fan,CWC.W.Lin,STS.T.Chen,AHA.H.Wang,etal.,Precisemappingofincreasedsialylationpatternandtheexpressionofacutephaseproteinsaccompanyingmurinetumor

progressioninBALB/cmousebyintegratedseraproteomicsandglycomics,J.ProteomeRes.7,2008,3293–3303.

[133]OO.GornikandGG.Lauc,Glycosylationofserumproteinsininflammatorydiseases,Dis.Markers25,2008,267–278.

[134]SDS.D.ShiyanandNVN.V.Bovin,Carbohydratecompositionandimmunomodulatoryactivityofdifferentglycoformsofalpha1-acidglycoprotein,Glycoconj.J.14,1997,631–638.

[135]AA.MackiewiczandKK.Mackiewicz,Glycoformsofserumalpha1-acidglycoproteinasmarkersofinflammationandcancer,Glycoconj.J.12,1995,241–247.

[136]FF.CecilianiandVV.Pocacqua,Theacutephaseproteinalpha1-acidglycoprotein:amodelforalteredglycosylationduringdiseases,Curr.ProteinPept.Sci.8,2007,91–108.

[137]CC.McCarthy,RR.Saldova,MRM.R.Wormald,PMP.M.Rudd,NGN.G.McElvaneyandEPE.P.Reeves,Theroleandimportanceofglycosylationofacutephaseproteinswithfocusonalpha-1antitrypsininacuteand

chronicinflammatoryconditions,J.ProteomeRes.13,2014,3131–3143.

[138]EE.MarklovaandZZ.Albahri,Screeninganddiagnosisofcongenitaldisordersofglycosylation,Clin.Chim.Acta385,2007,6–20.

[139]WW.VanMolle,GG.Denecker,II.Rodriguez,PP.Brouckaert,PP.VandenabeeleandCC.Libert,Activationofcaspasesinlethalexperimentalhepatitisandpreventionbyacutephaseproteins,J.

Immunol.163,1999,5235–5241.

[140]PDP.D.Eckersall,Acutephaseproteinsasmarkersofinflammatorylesions,Comp.Haematol.Int.5,1995,93–97.

[141]EE.Gruys,MJMM.J.M.Toussaint,NN.Upragarin,AMA.M.VanEderen,AA.Adewuyi,DD.Candiani,etal.,Acutephasereactants,challengeinthefuture,In:5thInternationalColloquiumonAnimalAcutePhase

Proteins.Dublin,Ireland,2005,60.

[142]HFH.F.Juan,JHJ.H.Chen,WTW.T.Hsu,SCS.C.Huang,STS.T.Chen,JJ.Yi-ChungLin,etal.,Identificationoftumor-associatedplasmabiomarkersusingproteomictechniques:frommousetohuman,

Proteomics4,2004,2766–2775.

[143]MAM.A.Penno,MM.Klingler-Hoffmann,JAJ.A.Brazzatti,AA.Boussioutas,TT.Putoczki,MM.Ernst,etal.,2D-DIGEanalysisofserafromtransgenicmousemodelsrevealsnovelcandidateproteinbiomarkersfor

humangastriccancer,J.Proteome77,2012,40–58.

[144]JMJ.M.Humphries,MAM.A.Penno,FF.Weiland,MM.Klingler-Hoffmann,AA.Zuber,AA.Boussioutas,etal.,Identificationandvalidationofnovelcandidateproteinbiomarkersforthedetectionofhumangastric

cancer,Biochim.Biophys.Acta2014,1844,1051–1058.

[145]CJC.J.Jeffery,Anintroductiontoproteinmoonlighting,Biochem.Soc.Trans.42,2014,1679–1683.

[146]SS.Hernandez,GG.Ferragut,II.Amela,JJ.Perez-Pons,JJ.Pinol,AA.Mozo-Villarias,etal.,MultitaskProtDB:adatabaseofmultitaskingproteins,NucleicAcidsRes.42,2014,D517–D520.

[147]MM.Mani,CC.Chen,VV.Amblee,HH.Liu,TT.Mathur,GG.Zwicke,etal.,MoonProt:adatabaseforproteinsthatareknowntomoonlight,NucleicAcidsRes.43,2015,D277–D282.

[148]IKI.K.KhanandDD.Kihara,Computationalcharacterizationofmoonlightingproteins,Biochem.Soc.Trans.42,2014,1780–1785.

[149]SS.Hernandez,AA.Calvo,GG.Ferragut,LL.Franco,AA.Hermoso,II.Amela,etal.,Canbioinformaticshelpintheidentificationofmoonlightingproteins?,Biochem.Soc.Trans.42,2014,1692–1697.

[150]ACA.C.Martin,Structuralbiologyofmoonlighting:lessonsfromantibodies,Biochem.Soc.Trans.42,2014,1704–1708.

[151]PP.BielliandLL.Calabrese,Structuretofunctionrelationshipsinceruloplasmin:a'‘moonlighting'’protein,Cell.Mol.LifeSci.59,2002,1413–1427.

[152]MGM.G.Derebe,CMC.M.Zlatkov,SS.Gattu,KAK.A.Ruhn,SS.Vaishnava,GEG.E.Diehl,etal.,SerumamyloidAisaretinolbindingproteinthattransportsretinolduringbacterialinfection,eLife3,2014,e03206.

[153]TT.LapidotandII.Petit,Currentunderstandingofstemcellmobilization:therolesofchemokines,proteolyticenzymes,adhesionmolecules,cytokines,andstromalcells,Exp.Hematol.30,2002,973–981.

[154]CLC.L.Bristow,MAM.A.Babayeva,MM.LaBrunda,MPM.P.MullenandRR.Winston,alpha1ProteinaseinhibitorregulatesCD4+lymphocytelevelsandisratelimitinginHIV-1disease,PLoSOne7,2012,e31383.

[155]ARA.R.Wyatt,JJJ.J.Yerbury,HH.EcroydandMRM.R.Wilson,Extracellularchaperonesandproteostasis,Annu.Rev.Biochem.82,2013,295–322.

[156]ARA.R.Wyatt,JJJ.J.Yerbury,RAR.A.DabbsandMRM.R.Wilson,Rolesofextracellularchaperonesinamyloidosis,J.Mol.Biol.421,2012,499–516.

[157]CC.Maugeais,SS.Braschi,KK.Ouguerram,PP.Maugeais,PP.Mahot,BB.Jacotot,etal.,Lipoproteinkineticsinpatientswithanalbuminemia.Evidencefortheroleofserumalbuminincontrollinglipoprotein

metabolism,Arterioscler.Thromb.Vasc.Biol.17,1997,1369–1375.

[158]SS.Nagase,KK.ShimamuneandSS.Shumiya,Albumin-deficientratmutant,Science205,1979,590–591.

[159]KK.Sugiyama,TT.EmoriandSS.Nagase,Synthesisandsecretionofplasmaproteinsbyisolatedhepatocytesofanalbuminemicrats,J.Biochem.92,1982,775–779.

[160]KK.LiangandNDN.D.Vaziri,HMG-CoAreductase,cholesterol7alpha-hydroxylase,LCAT,ACAT,LDLreceptor,andSRB-1inhereditaryanalbuminemia,KidneyInt.64,2003,192–198.

[161]JJ.Kang,MM.Holland,HH.JonesandGAG.A.Kaysen,Coordinateaugmentationinexpressionofgenesencodingtranscriptionfactorsandliversecretoryproteinsinhypo-oncoticstates,KidneyInt.56,1999,

452–460.

[162]TRT.R.Figueira,RFR.F.Castilho,AA.Saito,HCH.C.OliveiraandAEA.E.Vercesi,ThehighersusceptibilityofcongenitalanalbuminemicratstoCa2+‐−inducedmitochondrialpermeabilitytransitionisassociated

withtheincreasedexpressionofcyclophilinDandnitrosothioldepletion,Mol.Genet.Metab.104,2011,521–528.

[163]RR.Wait,GG.Chiesa,CC.Parolini,II.Miller,SS.Begum,DD.Brambilla,etal.,Referencemapsofmouseserumacute-phaseproteins:CchangeswithLPS-inducedinflammationandapolipoproteinA-IandA-II

transgenes,Proteomics5,2005,4245–4253.

[164]CJC.J.FurlanetoandAA.Campa,AnovelfunctionofserumamyloidA:Apotentstimulusforthereleaseoftumornecrosisfactor-α,interleukin-1β,andinterleukin-8byhumanbloodneutrophil,Biochem.Biophys.

Res.Commun.268,2000,405–408.

[165]NN.Hyka,JMJ.M.Dayer,CC.Modoux,TT.Kohno,CKC.K.Edwards,III,PP.Roux-Lombard,etal.,ApolipoproteinA-Iinhibitstheproductionofinterleukin-1βandtumornecrosisfactor-αbyblockingcontact-

mediatedactivationofmonocytesbyTlymphocytes,Blood97,2001,2381–2389.

[166]CJC.J.Furlaneto,FPF.P.Ribeiro,EE.Hatanaka,GG.Souza,MAM.A.CassatellaandAA.Campa,ApolipoproteinsA-IandA-IIdownregulateneutrophilfunctions,Lipids37,2002,925–928.

[167]BALB.A.L.HurnandSMS.M.Chantler,Productionofreagentantibodies,MethodsEnzymol.70,1980,104–141.

[168]NN.Tsolakos,CC.Brookes,SS.Taylor,AA.Gorringe,CMC.M.Tang,IMI.M.Feavers,etal.,IdentificationofvaccineantigensusingintegratedproteomicanalysesofsurfaceimmunogensfromserogroupB

Neisseriameningitidis,J.Proteome101,2014,63–76.

[169]KPK.P.Karunakaran,HH.Yu,XX.Jiang,QQ.Chan,KMK.M.Moon,LJL.J.Foster,etal.,OutermembraneproteinspreferentiallyloadMHCclassIIpeptides:implicationsforaChlamydiatrachomatisTcellvaccine,

Vaccine33,2015,2159–2166.

[170]EDE.D.Virginio,PHP.H.Kubitschek-Barreira,MVM.V.Batista,MRM.R.Schirmer,EE.Abdelhay,MAM.A.Shikanai-Yasuda,etal.,ImmunoproteomeofAspergillusfumigatususingseraofpatientswithinvasive

aspergillosis,Int.J.Mol.Sci.15,2014,14505–14530.

[171]ZMZ.M.Radulovic,TKT.K.Kim,LML.M.Porter,SHS.H.Sze,LL.LewisandAA.Mulenga,A24-48hfedAmblyommaamericanumticksalivaimmuno-proteome,BMCGenomics15,2014,518.

[172]LL.Alsoe,JEJ.E.Stacy,AA.Fossa,SS.Funderud,OHO.H.BrekkeandGG.Gaudernack,Identificationofprostatecancerantigensbyautomatedhigh-throughputfilterimmunoscreening,J.Immunol.

Methods330,2008,12–23.

[173]AA.KrahandPRP.R.Jungblut,Immunoproteomics,MethodsMol.Med.94,2004,19–32.

[174]CSC.S.Ahn,XX.Han,YAY.A.Bae,XX.Ma,JTJ.T.Kim,HH.Cai,etal.,AlterationofimmunoproteomeprofileofEchinococcusgranulosushydatidfluidwithprogressionofcysticechinococcosis,Parasites

Vectors8,2015,10.

[175]SS.Kunnath-Velayudhan,HH.Salamon,HYH.Y.Wang,ALA.L.Davidow,DMD.M.Molina,VTV.T.Huynh,etal.,DynamicantibodyresponsestotheMycobacteriumtuberculosisproteome,Proc.Natl.Acad.Sci.U.S.

A.107,2010,14703–14708.

[176]YFY.F.Lin,CYC.Y.Chen,MHM.H.Tsai,MSM.S.Wu,YCY.C.Wang,EYE.Y.Chuang,etal.,Duodenalulcer-relatedantigensfromHelicobacterpylori:immunoproteomeandproteinmicroarrayapproaches,Mol.Cell.

Proteomics6,2007,1018–1026.

[177]EE.Lahner,GG.Bernardini,SS.Possenti,GG.Renzone,AA.Scaloni,AA.Santucci,etal.,ImmunoproteomicsofHelicobacterpyloriinfectioninpatientswithatrophicbodygastritis,apredisposingconditionfor

gastriccancer,Int.J.Med.Microbiol.301,2011,125–132.

[178]DHD.H.Vu,PP.SchneiderandJDJ.D.Tissot,ElectrophoreticcharacteristicsofmonoclonalimmunoglobulinGofdifferentsubclasses,J.Chromatogr.BAnal.Technol.Biomed.LifeSci.771,2002,355–368.

[179]JJ.Stulik,MM.TichyandHH.Kovarova,Two-dimensionalgelelectrophoresisoffourserumsamplesfrompatientswithIgDmyeloma,Clin.Chim.Acta218,1993,149–158.

[180]JDJ.D.Tissot,FF.Invernizzi,FF.Schifferli,FF.SpertiniandPP.Schneider,Two-dimensionalelectrophoreticanalysisofcryoproteins:Areportof335samples,Electrophoresis20,1999,606–613.

[181]H.Krotkiewski,Carbohydratemoietyofimmunoglobulinsinhealthandpathology,ActaBiochim.Pol.46,1999,341–350.

[182]PMP.M.Rudd,TT.Elliott,PP.Cresswell,IAI.A.WilsonandRAR.A.Dwek,Glycosylationandtheimmunesystem,Science291,2001,2370–2376.

[183]EE.Maverakis,KK.Kim,MM.Shimoda,MEM.E.Gershwin,FF.Patel,RR.Wilken,etal.,GlycansintheimmunesystemandTtheAalteredGglycantheoryofautoimmunity:acriticalreview,J.Autoimmun.57,2015,

1–13.

[184]KTK.T.Shade,BB.Platzer,NN.Washburn,VV.Mani,YCY.C.Bartsch,MM.Conroy,etal.,AsingleglycanonIgEisindispensableforinitiationofanaphylaxis,J.Exp.Med.212,2015,457–467.

[185]CC.Reily,HH.Ueda,ZQZ.Q.Huang,JJ.Mestecky,BAB.A.Julian,CDC.D.Willey,etal.,Cellularsignalingandproductionofgalactose-deficientIgA1inIgAnephropathy,anautoimmunedisease,Int.J.Immunol.

Res.2014,2014,197548.

An opposite perspective to depletion of high-abundance proteins from plasma/serum is the enrichment of low-abundance ones through an increase of their relative concentration; such acompression of the dynamic range is expected to provide a deeper proteome profiling. This technology was described by the inventors as “protein equalization” and its effect as approaching a"“democraticproteome"” [33]. The enrichment procedure is based on a specific and saturable interaction of proteins to a high diversity of binding sites exposed on chromatographic beads. Thecombinatorialligandlibraryismadeupofdozensofmillionsofhexapeptidescapableofinteractingwithmost,ifnotall,proteinsinanygivenproteome[34,35](SupplementaryFigs.1and2).BioRad(http://www.bio-rad.com/)marketsthecorrespondingcommercialkit,ProteoMiner,togetherwith4elutionreagentsmeanttosequentiallyrecoverproteinsonthebasisofdifferentproperties,inaformatcompatiblewithanalysisbySELDI.DifferentialelutionundernativeconditionsmaybeachievedbychangingthepHofthebuffer[36]orbyusingamixtureofchargedaminoacids(150mMLys,Arg,AspandGlu)[37].ElutionofallboundproteinsunderdenaturingconditionsismadepossiblebyboilingthebeadsinSDS[38,39];precipitationwithacetonitrilethenprovidesforthehighestproteinrecoveryyieldandthebest2DEspotpattern[40].

Theconceptualbackgroundtothepeptideligandlibrarycaptureisasaturationeffect:alltheproteinspresentinthestartingsampleinconcentrationshighenoughtooverloadtheirinteractionsitesonthebeadsshouldbeboundinequalamounts,theexcessbeingdiscardedwiththeflow-through.Afterstrippingtheboundproteinsfromtheresin,theconcentrationofthehigher-abundanceproteins should be equalized in the eluted fraction. Conversely, the proteins present in the starting sample in concentrations too low to overload their interaction sites on the beads should bequantitativelybound.Accordingly,thedifferencesamongsamplesshouldbeintentionallyabolished(oratleastreduced)forhigher-abundanceandpreservedforlower-abundanceproteins.Atestofthequantitativeperformanceofthecapturebeadsshowsanaveragevariabilityaround10%intheamountoftheisolatedproteins,independentoftheirinitialabundance.Nonormalizationeffectisinfact

observedbut,dependingontheprotein-to-beadratio,specificsetsofproteinsareeitherenrichedordepletedvsthestartingwholeproteome[43](SupplementaryFig.3).Allreportsthatevaluatetheperformanceoftheproteinenrichmentstrategy[44]orcompareittodepletionprotocols[45]acknowledgetheexpectedcompressionofthedynamicrangeofserumproteinconcentrations.

Thefollowingarethesupplementarydatarelatedtothisarticle.

SupplementaryFig.12DEmapanalysisofcontrolserum(upperleftpanel)vsthreeligandlibrarybeadeluates.ThefirstelutionwasaccomplishedbyusingthreecolumnvolumesofTUCsolution(2Mthiourea,7Murea,2%CHAPS)followedbythe

secondelutionusingthreevolumesofanaqueoussolutionof9Murea,pH3.5(pHadjustedwithcitricacid).Athirdelutionwasthenperformedwithasolutioncomposedofacetonitrile,isopropanol,trifluoroaceticacid,andwaterinthefollowing

volumetricproportions:16.6,33.3,0.5,and49.5%.(From[41]).

alt-text:SupplementaryFig.1

SupplementaryFig.22DEofeluatesfrom0-mer(beadswithoutpeptides)to6-merasasequenceofcolumnscomparedtotheinitialRBClysateandtotheeluatefromahexapeptidecolumndirectlyloadedwiththeRBClysate.Thetotalnumberofspots

countedpermapismarked.(From[42]).

alt-text:SupplementaryFig.2

SupplementaryFig.3Numberofproteinsidentifiedinsamplestreatedwithcombinatorialpeptideligandlibrary(CPLL)beadsvstheuntreatedextract,asafunctionofcopies/cell:allproteins(leftpanel)andunsharedproteins(rightpanel);datafor

yeastcellproteins(S.cerevisiaestrainS288C,BY4741).(Redrawnfrom[43].)

alt-text:SupplementaryFig.3

QueriesandAnswersQuery:

Yourarticle is registeredasaregular itemand isbeingprocessed for inclusion inaregular issueof the journal. If this isNOTcorrectandyourarticlebelongs toaSpecial Issue/Collectionpleasecontact

a.anwerbasha@elsevier.comimmediatelypriortoreturningyourcorrections.

Answer:Yes

Query:

Theauthornameshavebeentaggedasgivennamesandsurnames(surnamesarehighlightedintealcolor).Pleaseconfirmiftheyhavebeenidentifiedcorrectly.

Answer:Yes

Highlights

• Thereviewpresentsanddiscussesprotocolsfordepletionofabundantserumproteins.

• Italsopresentsapproachesfortheeffectiveanalysisofwhole,untreatedserum.

• Nextitelaboratesonmultiplemolecularforms/multiplefunctionsofserumproteins.

• Itthenrecommendsthatinformationonabundantserumproteinsberoutinelyobtained.

Graphicalabstract

alt-text:Image1