Serum protein profiling of early and advanced stage Crohn's disease

Available

online

at

www.sciencedirect.com

ScienceDirect

j o ur na l h o me p a g e:h t t p : / / w w w . e l s e v i e r . c o m / l o c a t e / e u p r o t

Serum

protein

profiling

of

early

and

advanced

stage

Crohn’s

disease

C.

Piras

_,

_A.

_Soggiu

_,

_V.

_Greco

_,

_A.

_Cassinotti

_,

_G.

_Maconi

_,

_S.

_Ardizzone

_,

A.

Amoresano

_,

_G.

_Bianchi

_Porro

_,

_L.

_Bonizzi

_,

_P.

_Roncada

a_{DIVET,Dipartimentodiscienzeveterinarieesanitàpubblica,UniversitàdegliStudidiMilano,Italy} b_{FondazioneSantaLucia–IRCCS,Rome,Italy}

c_{GastroenterologyUnit,L.SaccoUniversityHospital,Milan,Italy}

d_{DepartmentofOrganicChemistryandBiochemistry,UniversityofNaples“FedericoII”,Italy} e_{IstitutoSperimentaleItalianoL.Spallanzani,Milano,Italy}

a

r

t

i

c

l

e

i

n

f

o

Received24May2013 Receivedinrevisedform 11February2014 Accepted18February2014 Keywords: Crohn’sdisease 2Delectrophoresis Comparativeproteomics C3ccomplement Transthyretin

a

b

s

t

r

a

c

t

Crohn’sdisease(CD) represents a highlydebilitating diseaseof difficultdiagnosis and increasingincidence.SerumproteinprofilingofearlystageCrohn’sdisease(ES)CDwas investigatedinordertoimprovethecomprehensionoftheveryearlypathologic mecha-nismsandtosupportthedifficultdiagnosticprocedurescurrentlyavailable.Inflammatory proteinsandcomplement3chainC(C3c)wereover-representedduringESCD,clusterin, retinolbindingprotein,␣1-microglobulinandtransthyretinwereunder-represented.AC3c isoformwasfoundtobepresentonlyduringESCD.Bynow,lackofspecificantibodiesto detectisoformsmadeitimpossibletoperformalternativevalidation.

©2014TheAuthors.PublishedbyElsevierB.V.onbehalfofEuropeanProteomics Association(EuPA).ThisisanopenaccessarticleundertheCCBY-NC-NDlicense

(http://creativecommons.org/licenses/by-nc-nd/3.0/).

1.

Introduction

Crohn’sdisease(CD),togetherwithulcerativecolitis(UC),is themostcommonformofinflammatoryboweldisease(IBD) [1].CDisachronicinflammatoryconditionofunknown aetiol-ogythatcanaffectanyportionofthedigestivetract,butmost frequentlytheterminalileumand/or thecolon[2]. Genetic factors,an abnormalimmune responsetomicrobial infec-tionsandunbalanceinthegut-microbiotaarethoughttobe involvedindiseasepathogenesis[3,4].

∗ _{Correspondingauthorat:IstitutoSperimentaleItalianoL.Spallanzani,c/oDIVET,UniversitàdegliStudidiMilano,viaCeloria10,20133,}

Milano,Italy.Tel.:+390250318138;fax:+390250318171.

E-mailaddresses:paola.roncada@guest.unimi.it,paola.roncada@gmail.com(P.Roncada).

ThediagnosisandmanagementofIBDstillpresentsa num-ber ofchallenges for treating physicians. The presenceof intestinal inflammationisaprimarycriterion fordiagnosis and differentiation from other diseases,no definitive diag-nostictestexistsasagoldstandardforCDdiagnosis,which is madeon the basisofhistory and physicalexamination, supplementedwithobjective findingsfromlaboratory, radi-ological, endoscopicand histological studies [5]. Thelatter involve invasive procedures, which are often a burden for thepatient.Consequently,inanattempttoovercomethese

http://dx.doi.org/10.1016/j.euprot.2014.02.010

2212-9685/©2014TheAuthors.PublishedbyElsevierB.V.onbehalfofEuropeanProteomicsAssociation(EuPA).Thisisanopenaccess articleundertheCCBY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/3.0/).

problems,anumberoflaboratorymarkersforthediseasehave beenevaluated[6,7].

Thepresenceofactivegutinflammationinpatientswith IBD is associated with an acute phase reaction and the migrationofleukocytesinthegut[8,9].Thispromotesthe pro-ductionofalargenumberofproteins,detectable inserum and stools. In serum, elevated acute phase markers (e.g. erythrocyte sedimentation rate, orosomucoid (alpha-1-acid glycoprotein)andC-reactiveprotein,CRP)canbefound. How-ever, while these markers, and especially CRP, have been showntocorrelatewithdiseaseactivityandprognosisofIBD, theirspecificityisinsufficient,andtheymayreflectgeneral inflammation[10–12].Faecalcalprotectinandlactoferrinare theproductsofactivatedneutrophilcellsandtheyarereleased intothefaeces.TheyaremorespecificbiomarkersofIBD,and offerapromisingwaytoconfirmintestinalinflammation,but theycannotdifferentiate betweenIBDand other intestinal inflammationssuchasintestinalinfectionsornon-steroidal anti-inflammatorydrugenteropathy[8–12].Thus,theutilityof testingthesebiomarkersinroutineclinicalpracticeneedsto beexploredfurther,includingitscost.Serologicalstudies eval-uatingantibodiesagainstSaccharomycescerevisiae(ASCA)and perinuclearanti-neutrophilcytoplasmicantibodies(p-ANCA) mayprovideadjunctivesupportforthediagnosisofCDand itsdifferentialdiagnosiswithUC,buttheyalsosufferfroma lackofsensitivityandspecificity[13–16].

TheinvestigationofmolecularmechanismsrelatedtoCDis crucialtothedevelopmentofnewdiagnosticandtherapeutic strategies.Althoughhigherlevelsofacutephaseproteinshave alreadybeenwelldescribedinCD[17],aqualitativeanalysisof isoformsoftheseproteinsduringthediseasestate,andin par-ticularduringthevariousstagesofpathology,hasneverbeen reported.Meuwisetal.[12]suggestedthehaptoglobinsubunit (Hp␣2)asapotentialbiomarkerfortheclinicalphaseofCD, butthisstudydidnotconsiderchangesrelatedtotreatment andinthe ESCD. Moreover,therole ofproteaseinhibitors as ␣1-antitrypsin (A1AT) and ␣1-antichymotrypsin (AACT) hasbeen reportedtobeimportant,inthe pathophysiology ofthisdisease[18].Crohn’sdisease-relatedepithelialinjury andulcerationare thoughttobeduetocytokine-mediated alterationsinthebalancebetweenmatrixmetalloproteinases (MMPs) and tissueinhibitors ofMMPs (TIMPs),resulting in degradationoftheliningofthegut[19].Wethereforeaimed toinvestigatethe proteomic profileoftheES vs.AS CD in comparisontoHCsinordertoexplorethedifferential expres-sionofacutephaseproteinsorproteinisoformscharacteristic ofthepathologicalstatus,accordingtodiseasedurationand treatment.

2.

Materials

and

methods

2.1. Patients

The institutional ethics committee of Sacco Hospital of Milan approved the experimental protocol (Protocol n◦ 239/07/83/06/AP, 13/04/2007). All subjects provided written informed consent before enrollment. Serumsamples were collected from 13 healthy controls (HC), 8 early stage (ES) CD patients and 36 advanced stage (AS) CD patients for

comparative proteomicanalysis. Ofthe36 ASCD samples, 16 were excluded as they were receiving more than one anti-inflammatorydrug atthe timeofserum collection, 20 werechosentobeincludedinthisstudy.Ithasbeendecided to avoid multiple drugs treatments to do not excessively interferewiththeserumproteome.

ThediagnosisofCDwasbasedonpublishedinternational criteria[5].ESandASCDweredefinedaccordingwiththe dura-tionofthedisease,aspreviouslydescribed[20–22],following thesecriteria: Early=first attackofCD inapatientwithno previoushistoryofanygastrointestinalsymptomsorsurgery; Advanced=CD in a patient with at least five years history fromthetimeofinitialdiagnosisandwithpersistentclinical activity requiringimmunosuppressors,immunomodulators, steroidsorsurgery.ESCDpatientshadneverreceived corti-costeroids, antimetabolitesorbiologicaltherapyand serum wascollectedwithin3monthsfromthediagnoses(median1 month).Weexcludedanypatientwhohadanimmediateneed for surgery,severe comorbidity, documentedchronic infec-tion,apositivestoolcultureforpathogens,oramalignancy.AS CDpatientsweretreatedorwithsystemicsteroids,including oralbudesonideorprednisone,orwithmonoclonalantibodies againstTNF␣(i.e.infliximabandcertolizumabpegol),orwith oralimmunosuppressors(i.e.azathioprineormethotrexate).

TableS1hasbeen addedassupportinginformationand summarizesthebaselinedemographicandclinical character-isticsoftheenrolledpatients.

Supplementary material related to this article can be found, in the online version, at doi:10.1016/j.euprot.

2014.02.010.

2.2. Samples

Asingleblood samplewastakenfrom each HC/CDsubject after overnight fasting, and after 10–15min of rest. Blood clotting wasachieved bystanding tubes vertically atroom temperature(22◦C)for60min.Afterbloodclotting,samples werecentrifugedat1500×gfor10minat4◦Candthe super-natants(serum)werestoredin1mLaliquotsat-80◦Cuntil use.

2.3. 2Delectrophoresis

Threeexperimentalreplicateswereperformedforeach sam-pletominimizetechnicalgel-to-gelvariation.Thegelswere stainedwithcoomassieG250,forevaluationoflinearprotein expressionandformassspectrometricanalysis.Immobiline Drystrips(pH3–10,length18cm,GE-Healthcare)were rehy-drated with 350l buffer containing 8M urea, 4% CHAPS, 65mM DTT, 1% ampholine and 0.002% bromophenol blue

[23,24] for14h atroom temperature. Serumsampleswere

defrosted and diluted in a buffer containing 8M urea,4% CHAPS,65mMDTT,1%ampholinepH3.5–10and0.002% bro-mophenolblue.100gofproteinwereloadedbycathodiccup loading.IsoelectricfocusingwasperformedusingtheEttan IPGphorIIIIEFSystem(GEHealthcare)withatotalof140kVh. Stripswereequilibratedinasolutioncontaining6Murea,30% glycerol,2%SDSand50mMTris–HCl(pH8.8),withtheaddition of1%w/vDTTinthefirststep,and2.5%w/viodoacetamide inthesecondstep.

For the second dimension, proteins were separated by SDS-PAGE on 10% polyacrylamide gels using Ettan Daltsix (GE-Healthcare)accordingtothefollowingprocedure:30min ata constant current of 12mA followed by 24mA per gel untilthebromophenolbluefrontreachedthebottomofthe gel.Gels were then stained withCoomassieG250 or silver stainingcompatiblewithmassspectrometry[25].Ultra nar-rowcustomizedIPGstrips(pH4–5.5,length13cm[26,27])were rehydratedwith350lrehydrationbufferand1mgofsample proteinwasloadedbycathodiccuploading.Forthesecond dimension,proteinswereseparatedbySDS-PAGEon10% poly-acrylamidegelsusingEttanDaltsix(GE-Healthcare)according tothefollowingprocedure:30minataconstantcurrent of 12mAfollowedby24mApergeluntilthebromophenolblue frontreachedthebottomofthegel.Gelswerethenstained withCoomassieG250orMSmodifiedsilverstaining[25].Ultra narrowcustomizedIPGstrips(pH4–5.5,length13cm[26,27]) wererehydratedwith350lrehydrationbufferand 1mgof sampleproteinwasloadedbycathodiccuploading.

For the second dimension, proteins were separated by 8–16%gradient–sodiumdodecylsulphate–polyacrylamidegel electrophoresisusinganEttanDaltsix(GE-Healthcare).Gels werestainedusingCoomassieG250.

2.4. Imageanalysis

AllimageswereacquiredusingaPHAROSFXlaserscanner (Bio-Rad)at100mresolution.Gelimageswereimportedinto ProgenesisSameSpots(v4.5;NonlinearDynamics,Newcastle, UK)foranalysis[28].

All imported images were processed with Progenesis SameSpotsto checkimage quality (saturation,dimension). Thealignedimageswerethenautomaticallyanalyzedusing the2Danalysismoduleforspotdetection,background sub-traction,normalization,andspotmatching,andallspotswere manuallyreviewedandvalidatedtoensureproperdetection andmatching.

2.5. Statisticalanalysis

Statistical analysis was performed using the Progenesis Stats moduleon the log-normalized volumes forall spots. Mann–Whitneytestandone-wayANOVAwereusedto con-firmthepvaluebetweendifferentgroups,p-valuesunder0.05 wereconsideredstatisticallysignificant.FDR(falsediscovery rate)andpoweranalysiswerealsocalculated,qvalues<0.05 andpowervalues>0.8respectivelywereconsideredtobe sig-nificant.Multivariatestatisticalanalysiswascarriedoutusing theProgenesisStatsmoduletoperformprincipalcomponents analysis(PCA)[29].

2.6. Proteinidentificationbymassspectrometry 2.6.1. Insitudigestion

Protein spots of interest were excised from the Colloidal Coomassie-stainedpreparative gel.Theexcisedspots were first washed with acetonitrile (ACN) and then with 0.1M ammonium bicarbonate. Protein samples were reduced by incubation in10mM dithiothreitol for45minat56◦C.The cysteine residues were alkylated by incubation in 55mM

iodoacetamidefor30minatroomtemperatureinthedark. Thegelparticleswerethenwashedwithammonium bicar-bonate andACN.Enzymaticdigestionwascarried outwith trypsin(12.5ng/l)in50mMammoniumbicarbonatepH8.5 at4◦Cfor1h.Thebufferedsolutionwasthenremovedanda freshaliquotoftheenzyme/buffersolutionwasaddedfor16h at37◦C.Aminimumreactionvolume,sufficienttoachievethe completerehydrationofthegelwasused.Peptideswerethen extracted,washingthegelparticleswith50mMammonium bicarbonateand0.1%trifluoroaceticacid(TFA)in50%ACNat roomtemperature,thenlyophilizedandresuspendedin30l of0.1%formicacid/3%ACNforMSanalysis.

2.6.2. MALDI-TOFMassspectrometry

Positive ReflectronMALDI-TOFmass spectrawere recorded on a Voyager DE STR instrument (Applied Biosystems, Framingham,MA).TheMALDImatrix waspreparedby dis-solving10mgof˛-cyano-4-hydroxycinnamicacid in1mlof ACN/water(90:10v/v).Typically,1lofmatrixwasappliedto themetallicsampleplateand1lofeachsamplewasthen added.Accelerationandreflectorvoltagesweresetupas fol-lows:targetvoltageat20kV,firstgridat95%oftargetvoltage, delayedextractionat600nstoobtainthebestsignal-to-noise ratios and thebest possibleisotopicresolutionwith multi-point externalcalibrationusingpeptide mixturepurchased fromAppliedBiosystems.Eachspectrumrepresentsthesum of1500laserpulsesfromrandomlychosenspotspersample position.Rawdatawereanalyzedusingthecomputersoftware providedbythemanufacturersandarereportedas monoiso-topicmasses.

Database searches with the measured monoiso-topic peptide masses were performed against the Swiss Prot database using the peptide search MASCOT

(http://www.matrixscience.com). Search parameters were

typicallysetto70ppmmaximaltolerance,onemissed cleav-age sitefortrypticpeptides allowed,andthe modifications acceptedwerecarboamidomethylationwithiodoacetamideof cysteinesandpossibleartefactualoxidationofmethionines. 2.6.3. NanoLCmassspectrometry

Trypticpeptidesderivedfromeachsamplewereanalyzedby nLC–MS/MSanalysisusinga4000Q-Trap(AppliedBiosystems) coupled to an 1100 nano HPLC system (Agilent Technolo-gies).ThesampleswereloadedonanAgilentreversed-phase pre-columncartridge(Zorbax300SB-C18, 5×0.3mm,5m) at10l/min(Asolvent0.1%formicacid,loadingtime5min). Peptides were separated on an Agilent reversed-phase col-umn(Zorbax300SB-C18,150mm×75m,3.5m),ataflow rateof0.3l/minwitha0–65%lineargradientin60min(A solvent 0.1% formic acid,2% ACN in MQ water; Bsolvent 0.1%formicacid,2%MQ waterinACN). Nanospraysource wasusedat2.5kVwithliquidcoupling,withadeclustering potentialof20V,usinganuncoatedsilicatipfrom NewObjec-tives(O.D.150m,I.D.20m,T.D.10m).Datawereacquired ininformation-dependentacquisition(IDA)mode,inwhich afullscanmassspectrumwasfollowedbyMS/MSofthe5 mostabundantions(2seach).Inparticular,spectra acquisi-tionofMS/MSanalysiswasbasedonasurveyEnhancedMS Scan(EMS)from400m/zto1400m/zat4000amu/s.Thisscan mode wasfollowedbyanEnhancedResolutionexperiment

(ER)forthefivemostintenseionsandthenMS2spectra(EPI) wereacquiredusingthebestcollisionenergycalculatedonthe basesofm/zvaluesandchargestate(rollingcollisionenergy) from100m/zto1400m/zat4000amu/s.Datawereacquired andprocessedusingAnalystsoftware(AppliedBiosystems).

LCMS Data were analyzedusing Analystsoftware (ver-sion 1.4.1) and MS/MS centroid peak lists were generated using the MASCOT.dll script (version 1.6b9). The threshold ofMS/MScentroid peakswastaken tobe0.1%ofthe base peak.MS/MSspectrahavinglessthan10peakswererejected. MS/MSspectraweresearchedagainsttheSwissProtdatabase using the licensed version of Mascot 2.1 (Matrix Science), afterconvertingthe acquiredMS/MSspectrainto the mas-cotgeneric fileformat. Thequerywasperformed withthe maximaltoleranceforparentmassesof200ppmanda max-imaltoleranceforfragmentsof0.2Da,searchingpeptideson 2+and 3+chargestate. Atthemosttwo miss-cleavagefor trypticpeptideswereallowed,andthemodificationsaccepted werecarboamidomethylationofcysteinesasfixed modifica-tionsand oxidationofmethioninesand pyro-gluN-termQ asvariable post-translational modifications. Thetaxonomy wasrestrictedtoHomosapiensSpectrawithaMASCOTscore <25wererejected.However,spectraldataweremanually val-idatedandcontainedsufficientinformationtoassignpeptide sequence.

2.6.4. Westernblotanalysis

Proteinextractfrom sera samples, 7g and5g/lanewere loaded on a SDS-PAGE system 10%, transfer blotting was achievedbya semi dry apparatus(NOVABLOT) onlow flu-orescencePVDF.Membraneswereincubated for2husinga commercialmonoclonalantibody(NB100-64339,Novus Bio-logicals)specificagainstC3ccomplementandsuccessivelyfor 1hwithasecondaryantibody(A3562,SIGMA).

3.

Results

All samples were analyzed using 3–10 pH IPG strips and obtaineddatahavebeenconsidered asmaindata. Further-moreultranarrow4–5.5[30]pH2Delectrophoresishavebeen employedtoconfirmobtaineddataandtobetterannotatethe supposedspecificproteinisoforms.

3.1. HCvs.ESCD

WhencomparingESCDandHC,imageanalysishighlighted 10 proteins whose expression was significantly altered in thediseasestate,theseproteinsweresuccessfullyidentified

byMS (Fig.1a and b, Table1). Inparticular, ˛1-antitrypsin

(A1AT,36.7% higherin ESCD), ˛1-antichymotrypsin(AACT, 50.2%higherinESCD),andcomplement3fragmentC(C3c, 270% higher in ES CD) were up-represented, haptoglobin (HPT,32%higherinESCD)expressionshowedatendencyto up-regulation(p=0.08)(Fig.1a).

Ontheotherhand,RETBP,CLUS,AMBP,andTTHYwere sig-nificantlyunder-representedinESCDcomparedwithHC,with levels49.8%,51%,74%and56.6%lower(p≤0.05),respectively

(Fig.1b). T able 1 – Differ entiall y repr esented pr oteins dur ing ES CD identified by MS. ↑ = up-r egula tion in ES Cr ohn’ s disease pa tients; ↓ = do wn-r egula tion in ES Cr ohn’ s disease pa tients. Spot nu mber Unipr ot accession n u mber Pr otein name ES CD vs. HC AS CD vs. ES CD AS CD vs. HC Mascot scor e Sequence co v e ra g e (%) Theor etical Mr/pI Ratio of means p -V alue Ratio of means p -V alue Ratio of means p -V alue 1 P01009 ␣ 1-Antitr ypsin 1.367686 0.00054 0.706468 4.00812E − 05 0.966226 0.61547942 255 73 60.59/4.70 2 P01011 ␣ 1-Antic h ymotr ypsin 1.497752 0.003195 0.81714 0.174199 1.223873 0.139132 373 87 54.93/5.00 3 P00738 Hapto globin 1.314226 0.121067 1.255693 0.176734 1.650265 0.000394 203 63 40.42/5.36 4 P01024 Complement 3 fr a gment C 3.760332 9.46E − 09 0.442775 2.47E − 07 1.664982 0.00608 263 58 40.91/4.84 5 P02753 Retinol binding pr otein 4 0.5027 0.000762 2.195367 0.000813 1.10361 0.447741 131 71 20.10/5.23 6 P10909 Clusterin 0.489887 7.9E − 08 2.229412 0.002048 1.092159 0.513289 163 87 37.21/4.85 7 P02760 Alpha-1-micr o globulin 0.260326 1.9E − 06 4.221332 0.005435 1.098921 0.658607 218 65 38.9/5.95 8 P02766 T ransth ytr etin 0.438517 1.87E − 06 1.866248 5.25E − 05 0.818381 0.021019 263 91 35.39/5.52 9 P02766 T ransth ytr etin isoform pI 4.7 HC = 0 – AS = 0 – HC = 0 – 170 80 35.39/4.72 10 P01024 Complement 3 fr a gment C isoform pI 4.9 HC = 0 – 0.404369 0.027986 HC = 0 – 255 57 38.80/4.95

Fig.1–(a)Up-representedserumproteinsinESCD(n=8)andHCs(n=13).A1AT(␣1-antitrypsin):HCs(2.96±0.54)vs.CD (4.05±0.65);AACT(␣1-antichymotrypsin):HCs(0.55±0.15)vs.CD(0.83±0.22);HPT(haptoglobin):HCs(4.30±1.03)vs.CD (5.67±0.86);C3c(complement3fragmentc):HCs(0.17±0.09)vs.CD(0.63±0.13)(*p≤0.05).(b)Down-representedserum proteinsinESCD(n=8)andHCs(n=13):RETBP(retinolbindingprotein):HCs(0.12±0.03)vs.CD(0.06±0.03);CLUS (clusterin):HCs(0.35±0.06)vs.CD(0.17±0.03);AMBP(␣1-microglobulin):HCs(0.067±0.016)vs.CD(0.017±0.012);TTHY (transthytretin):HCs(0.78±0.17)vs.CD(0.34±0.1).*p<0.05forMannWhitneytestcarriedoutforeachsamplingdatapoint betweenHCandCDproteins.

Furthermore,twoproteinsapparentlyabsentinHC sam-ples were identified in ES CD samples: one isoform of transthyretinatpI4.7andoneisoformfragmentof comple-mentC3catpI4.9(Fig.2).Bothisoformswerefoundin3–10and 4–5.5pHrangeexperiments,aswellastheyweresuccessfully identifiedbyMALDITOFandnLC–MS/MSanalysis.

3.2. ESCDvs.ASCD

AsshowninFig.3a,theproteinsover-representedduringthe ESofCD didnotmaintainhigherrelativelevelsduringthe ASofthe disease,and theirexpressionwas similarto HC. Notableexceptions tothis,however,were C3ccomplement fragmentand,toagreaterextent,haptoglobin(C3cp=0.024; HPTp=0.01).Proteinswhichwereunder-representedduring theESofCD,werealsoobservedtoreturntonormal(HC) lev-elsinASCD(Fig.3b).Notably,nodifferenceswereobserved within the AS CD patient group according to the type of anti-inflammatorytreatment(datanotshown).

3.3. PCAanalysisofallgroups

Toprovideevidenceforthedifferencesbetweenserum sam-ples,multivariateunsupervisedPCAanalysiswasperformed.

Fig.4representsthebiplotofthefirsttwoprincipal compo-nents(PC).

Thebiplotshowedaneffectiveseparationofsamplesin twoclasses:ESCDsamples(pink,negativescoresofPC1),and HCs(cyan,positivescoresofPC1).Spotsthatcontributeto dif-ferencesbetweenthetwogroupsarepresentasloadingsin the PCAplot(spotnumbers,Fig.4)andconfirmtheresults obtainedbynon-parametricunivariateanalysis.

Fig.5showsthebiplotwhichalsoincludessamplesfrom CDAS.ThegraphhighlightsthesimilaritybetweentheASCD (violet) andHCs(cyan)thatwould beincluded inthe same groupusingthisanalysis.ThisconfirmsthatinASCDprotein expressionreturnssimilartocontrolvalues.

AvalidationoftheC3fragmentprofilehasbeenattempted usingcommercialmonoclonalantibodyNB100-64339,Novus Biologicals.Suchavalidationhasbeenperformedintheearly stage Crohn’s disease samplesas considered more signifi-cant(Fig.6).Unfortunatelythespecificfragmentisoformof C3complementhighlightedinthe2DEwasnotdetectedby thisantibodywhichisrecognizingaproteinbandat115kDa supposedlyassociatedwiththeintactC3protein.Thisband showsanoveralltrendsimilartoofC3ccomplementisoform, detectedbythe2DE, howevertheseresults,donotshowa statisticallysignificantdifferentialexpressionbutonlya sug-gestiveprofile.

Fig.2–Detailsof2-Dmap(pI4.7–5.6;kDa45–30).1=complement3fragmentc;2=transthyretin;3=transthyretinisoformpI 4.7;4=complement3fragmentcisoformpI4.9.

Fig.3–Expressionprofilingofup-representedproteinsduringASofCrohn’sdisease:onlyHPTandC3ccomplementwere up-representedcomparedtoHCs(HPTp=0.01;C3cp=0.024)(*p≤0.05).

Fig.4–BiplotofthefirsttwocomponentsoftheprincipalcomponentanalysisofHCsandESCD.Ovalsindicatetheareas wherethedatapointsofthetwogroupsaredistributedpinkspots:ESCD,cyanspots:HCs.

4.

Discussion

The differential protein display was evaluated in ES CD patientsvs.HCandinASCDpatientsvs.ESCD.

4.1. HCvs.ESCD

Imageand multivariate (PCA) analysisof2Dmaps showed anoverexpressionof˛1-antitrypsin,˛1-antichimotrypsinand complement factor 3 chain c during ES CD (Fig. 1a) in comparisontoHC, haptoglobin showed aover-represented trend. On the other hand, clusterin, retinol binding pro-tein, ˛1-microglobulin and transthyretin (Fig. 1b) were under-represented. These are mostly inflammation-related proteinsthathavealreadybeendescribedduringCrohn’s dis-ease. However, this is an innovative study because of the presence of a rare group of Crohn’s disease patients that have been investigated during the ES of this disease. The analysisoftheserumproteomeofthisparticularpoolof sam-plesrevealed,asdescribedbelow,thepresenceofdifferential expressionofthedescribedproteinsandtheirisoforms.

4.2. ESCDvsASCD

Obtainedresultshighlightedthepresenceofimportant dif-ferences between theproteomic profiles ofHC and ESCD, however,the relativelevels ofthese proteinscameback to

levels similar to HC in AS CD patients (Fig. 3). Different immuneactivationpatternshavebeenpreviouslydescribed forT-cellresponsesinastudyinchildren,wherevariations in mucosal T-cell immune regulationduring the courseof humanIBDwereobserved[20].Inparticular,attheonsetof CD,mucosalTcellsappeartomountatypicalTh1response thatresemblesanacuteinfectiousprocess,butthisresponse islostwithprogressiontoASCD,suggestingthatpatientswith theinitialmanifestationsofIBDmayrepresentanideal pop-ulationinwhichimmunomodulationhasoptimaltherapeutic efficacy.

Moreover,theoverexpressionofsomeproteinsintheESCD mayreflecttheirpotentialroleindiseasepathogenesis accord-ingtotheparticularitiesoftheirfunctionsintheinflammatory cascade, or simplybeanaspecificmarkerofinflammatory activation.

Differentially represented proteins and isoforms dur-ing ES and AS CD in comparison to HC: ˛1-antitrypsin and ˛1-antichymotrypsin are protease inhibitors and posi-tiveacute-phaseproteinswhoseplasmaconcentrationsare increased asa resultof inflammation.During acute phase response,overexpressionofthesetwoproteinsiscrucialin theirreversibleinhibitionofproteaseactivity.Bothproteins are synthesizedprimarilyintheliver,butisoforms ofA1AT havebeenpreviouslyshowntobepresentalsoingutPaneth cells[31],andvariableassociationshavebeenfoundbetween A1ATlevelsorpolymorphismsandCD,withconflictingresults

[32–34].Notably,anexcessoftissutalproteaseinhibitorslike

Fig.5–BiplotofthefirsttwocomponentsoftheprincipalcomponentanalysisofHCsandES/ASCD.Ovalsindicatetheareas wherethedatapointsofthetwogroupsaredistributedpinkspots:ESCDsamples.Violetspots:ASCD,cyanspots:HCs.

Fig.6–ImmunoblottingoftheC3c.(a)Onedimensional electrophoresis(coomassiestaining).(b)Onedimensional immunoblottingwith7␮gperlaneproteinload.(c)One dimensionalimmunoblottingwith5␮gperlaneprotein load.

A1AThasbeendescribed [31]and thiswasassociatedwith adisturbanceinthenormalprocessingofantimicrobial pep-tides,likedefensins,whichthenallowsincreasedexposureof theintestinalmucosatoluminalmicro-organisms[35], trig-geringanexcessiveimmuneactivationinIBD.Ontheother hand,the downregulationofthese antiproteases,that was observedinASCD,maysuggestnewmechanismsofdisease controlbythevariousanti-inflammatorytreatmentsusedor changesinthemechanismsofimmuneactivationduringthe advancedphasesofthedisease.Asdescribedbefore,recent findingshighlightedthatanimpairedmetabolismofproteases andproteasesinhibitorscouldbelinkedtothe etiopathogen-esisofthisdisease[19].

4.3. ComplementC3c

Like other acute phase proteins, complement C3c is over-represented. This data confirms results from previ-ous studies [36]. Complement-derived proteolytic products

mediatefunctionswhichcontributetopathogen disruption andelimination[37].However,theinappropriateactivationof the complementsystemhasbeen involvedinpathogenetic mechanisms ofimmunological and inflammatorydiseases, including IBD, as shown in several studies [17,38–49]. The centralrole ofcomplement component3(C3)isduetothe complexity of the functional regulation of its fragments, includingthe majorfragment,C3b,whichanchorsthe con-vertaseassemblyeffectingactivationoffragmentsC3andC5. C3c isafragmentderivedfrom the proteolysisofC3b; C3c over-representationwasfoundnotonlyduringtheESCDbut also during AS disease,irrespective ofvarious treatments, suggestsadysregulatedcomplementmetabolismthatcould berelatedtodiseasepathogenesisandtothemaintenanceof gutinflammation.Recentmetaproteomicstudiessuggestthe possibilitytoevaluatethedifferencesbetweenthenormaland thepathologicalmicrobiota[50].Furthermore,imageanalysis revealedaC3cisoformatpI4[50].Furthermore,image anal-ysisrevealedaC3cisoformatpI4.9apparentlymissinginHC samples,furtherhighlightingthespecificroleofthe comple-mentsysteminCDimmuneactivationaswellastheneedto betterexploretheroleofthisisoform.Aputativeexplanation oftheroleofthecomplementsysteminotherautoimmune pathologieshavealreadybeenproposed[30]buttheprovided explanation doesnotincludethedescription oftherole or the provenienceof C3c fragment. However the differential representationofsuchaC3isoform maybeinteresting,we couldnotindependentlyvalidatesuchanevidencebytheuse ofanimmunochemicalmethodwithcommercialantibodies. OnlyatrendprofileoftheintactComplementC3proteincould bedetected.Suchapartialresultmaybedue toa number of factors affecting the superimposition of 2DE data with classical Westernblot detection. Theproblemofvalidation ofspecificisoformsofproteinsthataresignificantlydifferent in2Delectrophoresisisaquestionregardingtheavailability of epitope qualified and validated reagents. Also in this case,therearenocommercialantibodiesavailableagainsta specificisoformsand thecommercialproductsaredirected toaspecificundisclosedsequenceoftheentireproteins.

4.4. Clusterin

Thepresentstudyshowsalsothatclusterin,retinolbinding protein, ˛1-microglobulin and transthyretin are underrep-resented during ES CD. An over-expression of clusterin is reportedinawide varietyofmodelsofstress anddisease, includingwithdrawalofgrowthfactorsandexposuretotoxic agents [51]. Moreover clusterin expression [51]. Moreover clusterin expression increases in humans usually during intense oxidativestress, althoughthis phenomenon isnot observedinrheumatoidarthritis[52].Thesefindingssuggest apossibleroleforclusterininprotectingcellsfromoxidative stressduringpathologicalstatusandinflammatoryresponse. Accordingly,stronglyenhancedlevelsofclusterinwere pre-viously described in the ileum of CD patients, correlating with disease activity [53]. Increasedlevel ofclusterin dur-ing ES CD is therefore expected toprotect tissues against oxidativestressdamagecausedbyinflammation[54,55].The under-representation ofclusterininESCD observedinour study,isthereforeanunexpectedresult,andwarrantsfurther

investigationtoexploretheetiopathogenicimplications;for example,tostudyifimpairedprotectionfromoxidativestress, couldberesponsiblefortissueinjuryandimpairedmucosal barrier. The result may also suggest potential analogies betweenCDandrheumathoidarthritispathogenesis[56].

4.5. ˛1-Microglobulin

˛1-Microglobulin is animmunosuppressive plasma protein synthetizedbytheliver.Thedown-regulationobservedinthis study hasneverbeenreported beforeanddeserves further investigation as it may reflect a loss ofimmunoregulatory functionwhichcouldinfluencethebalancebetweenpro-and anti-inflammatorypathwaysofimmuneresponse.Howeverit hasbeeninterestinglydocumentedaconsiderablyincreased excretionof˛1-microglobulininpatientsaffectedbyCrohn’s disease.Itsincreasedexcretionissohightoinducescientists tospeculate˛1-microglobulininurineasputativebiomarker fortheevaluationofIBDactivityindex[57].

Retinolbindingprotein(RBP)issynthesizedprimarilyinthe liver, and requires retinol binding to trigger its secretion

[58,59]. In vivo,RBP binds the largerprotein, transthyretin,

whileinvitroonetetramer ofTTRbinds twomoleculesof retinolbindingprotein.However,theconcentrationofRBPin plasmaislimitingand thecomplexisolated fromserum is composedofbothTTRandRBP(inaonetoonestoichiometry) witharesultantmolecularmassofabout80kDa.The bind-ingofRBPtoTTRpreventsextensivelossofthelowmolecular weightRBPthroughglomerularfiltrationandmayalsorestrict freepartitioningofRBPintotheintercellularspaceoutsidethe vascularsystem[58,60,61].Retinolbindingproteinisavitamin Acarrier;itsdecreaseinCDhasbeenpreviouslydescribed[62] asasign ofintestinalmalabsorption.Themostinteresting characteristicofRBPisduetoitscommonpositive correla-tionwithsystemicinflammationthatisthephenomenonthat characterizesCD[63].Butsurprisingly,obtainedresults, con-firmeditsdown-regulationduringESCDinspiteofwhatwas expected.HoweverarecentstudybyKeichoandcolleagues foundRBPdecreasedinpatientswithtuberculosis[64].This experimentalevidence suggests again a putative microbial implicationinthepathogenesisofCDparticularlyif consider-ingthatMycobacteriumaviumsubsp.paratuberculosishasbeen oftendescribedastheputativecauseofCD[65]andcouldbe foundincowmilkforhumanconsumption[66].Moreover par-ticularattentionneedstobepaidtomicrobialpathogensdue totherecentevidencesinthefieldofmultidrugresistance[67].

4.6. Transthyretin

TTR is mainly synthesized by the liver and then secreted into the circulation. It is normally considered a negative acute-phaseproteinaswellas anutritionalindex[68]; our results match favourably with previous studies showing decreased levels of TTR in active CD [18], as well as its increase after effective treatment with infliximab [69]. Its concentrationwasfoundtobedecreasedalsoinrheumatoid arthritisconfirmingthealreadydescribedanalogiesbetween this two pathologies [70]. In ES CD it has been found an isoformatpI4.7.ThepresenceofthisisoformonlyinESCD couldbelinkedtoaputativePTMofthisproteinlinkedtothe

pathological status. Theputative role of TTRasbiomarker hasbeenalreadydescribedforotherpathologies[71].

5.

Conclusion

Ourcollectedevidenceshighlightedadifferentialprotein dis-playofseveralserumproteinsinESCDpatients.Inaddition itwasfoundatransthyretinisoform(pI4.7)andC3cfragment isoform(pI4.9)whichhasnotbeendescribedbeforeintype ofclinicalsamples.Theseisoformscouldhelptoexplainthe molecularchangesobservedduringtheclinicalphaseofCD and in particularthe role of C3ccomplement in the com-plement cascadeactivation andinthe diseaseprogression. TheC3cisoform isofparticularinterest asitwasstill rep-resented in78% ofASCD patientsafteranti-inflammatory therapies.Thisobservationmayindicatethatotherfactors(for example,microbialcomponents)couldstillplayanimportant roleintheadvancedphasesofthedisease.Inparticularitis importanttoquotethatmanyotherproteomicstudieshave investigatetheacutephaseproteinsinotherdiseasessimilar toCDasrheumatoidarthritis[72].Noneofthisstudies per-formedonserumdescribedanaugmentedexpressionofC3c complementasitwasfoundinthisstudy[70].Anotherstudy conductedbyCylwikandcolleaguesdescribestheanalysisof allserumacute-phaseproteinsduringthecourseof rheuma-toidarthritishighlightingthatC3complementconcentration wasnotinfluencedbythediseaseactivity[73].Moreovereven if C3 complement has been described as autoantigen dur-ing IBD[74], theover-representationofC3chasneverbeen described duringthis kindofpathologies.Therelevanceof C3complementandisoformsasputativebiomarkerforother pathologieshasalsobeendescribedhighlightingtherelevance offurtherstudiesonthis topic[75–77]. Allthis experimen-tal evidences could suggest that C3c proteolytic isoforms expressioninCrohn’sdiseasecould bedisease specificand encouragesfurtherinvestigationinwiderpopulations[78]in particularinthelightofthedifficultiesactuallyencounteredin thediagnosesofthispathology.Althoughthedetectionofthe describedisoformscouldrepresentadifficulttasksincethe currentlyavailablecommercialantibodiesare not recogniz-ingthisisoforms.Thedevelopmentofnewgenerationaffinity bindersisoformspecificwillrepresentakeymilestoneforthe construction ofdiagnostic testsfortheearlyphasesofthis pathology.

r

e

f

e

r

e

n

c

e

s

[1] BaumgartDC,SandbornWJ.Inflammatoryboweldisease:

clinicalaspectsandestablishedandevolvingtherapies.The

Lancet2007;369(9573):1641–57.

[2] BaumgartDC,CardingSR.Inflammatoryboweldisease:

causeandimmunobiology.TheLancet

2007;369(9573):1627–40.

[3] PresleyLL,YeJ,LiX,LeBlancJ,ZhangZ,RueggerPM,etal.

Host–microberelationshipsininflammatoryboweldisease

detectedbybacterialandmetaproteomicanalysisofthe

mucosal–luminalinterface.InflammatoryBowelDiseases

2012;18(3):409–17.

[4] EricksonAR,CantarelBL,LamendellaR,DarziY,Mongodin

revealshumanhost–microbiotasignaturesofCrohn’s

disease.PLoSONE2012;7(11):e49138.

[5] TravisS,StangeE,LémannM,ÖreslandT,ChowersY,Forbes

A,etal.Europeanevidencebasedconsensusonthe

diagnosisandmanagementofCrohn’sdisease:current

management.Gut2006;55(Suppl.1):i16–35.

[6] IskandarHN,CiorbaMA.Biomarkersininflammatorybowel

disease:currentpracticesandrecentadvances.

TranslationalResearch2012;159(4):313–25.

[7] LewisJD.Theutilityofbiomarkersinthediagnosisand

therapyofinflammatoryboweldisease.Gastroenterology

2011;140(6):1817–26,e2.

[8] TibbleJA,SigthorssonG,FosterR,ForgacsI,BjarnasonI.Use

ofsurrogatemarkersofinflammationandRomecriteriato

distinguishorganicfromnonorganicintestinaldisease.

Gastroenterology2002;123(2):450–60.

[9] FagerholMK.Surrogatemarkersofintestinalinflammation

arepredictiveofrelapseinpatientswithinflammatory

boweldisease.Gastroenterology2000;119(1):15–22.

[10] TibbleJ,BjarnasonI.Non-invasiveinvestigationof

inflammatoryboweldisease.WorldJournalof

Gastroenterology2001;7(4):460–5.

[11] VermeireS,VanAsscheG,RutgeertsP.C-reactiveproteinas

amarkerforinflammatoryboweldisease.Inflammatory

BowelDiseases2004;10(5):661–5.

[12] MeuwisMA,FilletM,GeurtsP,DeSenyD,LutteriL,Chapelle

JP,etal.Biomarkerdiscoveryforinflammatorybowel

disease:usingproteomicserumprofiling.Biochemical

Pharmacology2007;73(9):1422–33.

[13] VermeireS,JoossensS,PeetersM,MonsuurF,MarienG,

BossuytX,etal.ComparativestudyofASCA

(Anti-Saccharomycescerevisiaantibody)assaysin

inflammatoryboweldisease.Gastroenterology

2001;120(4):827–33.

[14] JoossensS,DapernoM,ShumsZ,VanSteenK,GoekenJA,

TrapaniC,etal.Interassayandinterobservervariabilityin

thedetectionofanti-neutrophilcytoplasmicantibodiesin

patientswithulcerativecolitis.ClinicalChemistry

2004;50(8):1422–5.

[15] JoossensS,ReinischW,VermeireS,SendidB,PoulainD,

PeetersM,etal.Thevalueofserologicmarkersin

indeterminatecolitis:aprospectivefollow-upstudy.

Gastroenterology2002;122(5):1242–7.

[16] KleblFH,BatailleF,BerteaCR,HerfarthH,HofstädterF,

SchölmerichJ,etal.Associationofperinuclear

antineutrophilcytoplasmicantibodiesand

anti-Saccharomycescerevisiaeantibodieswithvienna

classificationsubtypesofCrohn’sdisease.Inflammatory

BowelDiseases2003;9(5):302–7.

[17] BeneL,FüstG,FeketeB,KovácsÁ,HorváthL,ProhászkaZ,

etal.HighnormalserumlevelsofC3andC1inhibitor:two

acute-phaseproteinsbelongingtothecomplementsystem,

occurmorefrequentlyinpatientswithCrohn’sdiseasethan

ulcerativecolitis.DigestiveDiseasesandSciences

2003;48(6):1186–92.

[18] WeekeB,JarnumS.Serumconcentrationof19serum

proteinsinCrohn’sdiseaseandulcerativecolitis.Gut

1971;12(4):297–302.

[19] MonteleoneI,FedericiM,SarraM,FranzeE,CasagrandeV,

ZorziF,etal.Tissueinhibitorofmetalloproteinase-3

regulatesinflammationinhumanandmouseintestine.

Gastroenterology2014;143(5):1277–87,e1-4.

[20] KugathasanS,SaubermannLJ,SmithL,KouD,ItohJ,Binion

DG,etal.MucosalT-cellimmunoregulationvariesinearly

andlateinflammatoryboweldisease.Gut

2007;56(12):1696–705.

[21] BaertF,MoortgatL,VanAsscheG,CaenepeelP,VergauweP,

DeVosM,etal.Mucosalhealingpredictssustainedclinical

remissioninpatientswithearly-stageCrohn’sdisease.

Gastroenterology2010;138(2):463–8.

[22] D‘HaensG,BaertF,VanAsscheG,CaenepeelP,VergauweP,

TuynmanH,etal.Earlycombinedimmunosuppressionor

conventionalmanagementinpatientswithnewly

diagnosedCrohn’sdisease:anopenrandomisedtrial.The

Lancet2008;371(9613):660–7.

[23] GörgA,WeissW,DunnMJ.Currenttwo-dimensional

electrophoresistechnologyforproteomics.Proteomics

2004;4(12):3665–85.

[24] GaviraghiA,DeriuF,SoggiuA,GalliA,BonacinaC,BonizziL,

etal.Proteomicstoinvestigatefertilityinbulls.Veterinary

ResearchCommunications2010;34:33–6.

[25] YanJX,WaitR,BerkelmanT,HarryRA,WestbrookJA,

WheelerCH,etal.Amodifiedsilverstainingprotocolfor

visualizationofproteinscompatiblewithmatrix-assisted

laserdesorption/ionizationandelectrospray

ionization-massspectrometry.Electrophoresis

2000;21(17):3666–72.

[26] D‘auriaE,AgostoniC,GiovanniniM,RivaE,ZetterströmR,

FortinR,etal.Proteomicevaluationofmilkfromdifferent

mammalianspeciesasasubstituteforbreastmilk.Acta

Paediatrica2005;94(12):1708–13.

[27] RoncadaP,DeriuF,GaviraghiA,MartinoP,BonizziL.

ProteomicstudyofantibioticresistanceinEscherichiacoli

strains.VeterinaryResearchCommunications

2009;33:157–60.

[28] TurkR,PirasC,KovacicM,SamardzijaM,AhmedH,De

CanioM,etal.Proteomicsofinflammatoryandoxidative

stressresponseincowswithsubclinicalandclinical

mastitis.JournalofProteomics2012;75(14):4412–28.

[29] BortolatoM,GodarSC,MelisM,SoggiuA,RoncadaP,CasuA,

etal.NMDARsmediatetheroleofmonoamineoxidaseAin

pathologicalaggression.ScienceSignalling2012;

32(25):8574.

[30] ScambiC,LaVerdeV,DeFranceschiL,BarausseG,PoliF,

BenedettiF,etal.Comparativeproteomicanalysisofserum

frompatientswithsystemicsclerosisandsclerodermatous

GVHD.EvidenceofdefectivefunctionoffactorH.PLoSONE

2010;5(8):e12162.

[31] MolmentiE,PerlmutterD,RubinD.Cell-specificexpression

ofalpha1-antitrypsininhumanintestinalepithelium.

JournalofClinicalInvestigation1993;92(4):2022.

[32] FolwacznyC,UrbanS,SchröderM,HofmannB,NoehlN,

KönigA,etal.a1-Antitrypsinallelesandphenotypesin

patientswithinflammatoryboweldisease.Scandinavian

JournalofGastroenterology1998;33(1):78–81.

[33] TaddeiC,AudrainMAP,ReumauxD,SesboüeR,TestaA,

GalmicheJP,etal.Alpha1-antitrypsinphenotypesand

anti-neutrophilcytoplasmicauto-antibodiesin

inflammatoryboweldisease.EuropeanJournalof

Gastroenterology&Hepatology1999;11(11):1293.

[34] KotlowskiR,BernsteinCN,SilverbergMS,KrauseDO.

Population-basedcase–controlstudyofalpha1-antitrypsin

andSLC11A1inCrohn’sdiseaseandulcerativecolitis.

InflammatoryBowelDiseases2008;14(8):1112–7.

[35] ElphickD,LiddellS,MahidaYR.Impairedluminalprocessing

ofhumandefensin-5inCrohn’sdisease:persistenceina

complexwithchymotrypsinogenandtrypsin.TheAmerican

JournalofPathology2008;172(3):702–13.

[36] HalstensenT,MollnesT,BrandtzaegP.Persistent

complementactivationinsubmucosalbloodvesselsof

activeinflammatoryboweldisease:immunohistochemical

evidence.Gastroenterology1989;97(1):10.

[37] VolanakisJE,FrankMM.Thehumancomplementsystemin

healthanddisease.InformaHealthcare;1998.

[38] ScheininT,BöhlingT,HalmeL,KontiainenS,BjørgeL,Meri

epitheliuminulcerativecolitisandCrohn’sdisease.Human

Pathology1999;30(12):1427–30.

[39] NielsenH,BinderV,DaughartyH,SvehagSE.Circulating

immunecomplexesinulcerativecolitis.I.Correlationto

diseaseactivity.ClinicalandExperimentalImmunology

1978;31(1):72.

[40] PetersenNE,ElmgreenJ,TeisnerB,SvehagSE.Activationof

classicalpathwaycomplementinchronicinflammation.

ActaMedicaScandinavica1988;223(6):557–60.

[41] UekiT,MizunoM,UesuT,KisoT,NasuJ,InabaT,etal.

Distributionofactivatedcomplement:C3b,anditsdegraded

fragments,iC3b/C3dg,inthecolonicmucosaofulcerative

colitis(UC).Clinical&ExperimentalImmunology

1996;104(2):286–92.

[42] HalstensenTS,DasKM,BrandtzaegP.Epithelialdepositsof

immunoglobulinG1andactivatedcomplementcolocalise

withtheM(r)40kDputativeautoantigeninulcerative

colitis.Gut1993;34(5):650–7.

[43] HalstensenT,MollnesT,GarredP,FausaO,BrandtzaegP.

Surfaceepitheliumrelatedactivationofcomplementdiffers

inCrohn’sdiseaseandulcerativecolitis.Gut

1992;33(7):902–8.

[44] ErikZN,HenningG,GunnarB,OleTU.Complement

activationcapacityinplasmabeforeandduringhigh-dose

prednisolonetreatmentandtaperinginexacerbationsof

Crohn’sdiseaseandulcerativecolitis.BMCGastroenterology

2005;5:31.

[45] Zimmermann-NielsenE,AgnholtJ,Thorlacius-UssingO,

DahlerupJ,BaatrupG.Complementactivationinplasma

beforeandafterinfliximabtreatmentinCrohndisease.

ScandinavianJournalofGastroenterology

2003;38(10):1050–4.

[46] LauferJ,OrenR,GoldbergI,HorwitzA,KopolovicJ,Chowers

Y,etal.CellularlocalizationofcomplementC3andC4

transcriptsinintestinalspecimensfrompatientswith

Crohn’sdisease.Clinical&ExperimentalImmunology

2000;120(1):30–7.

[47] AhrenstedtÖ,KnutsonL,NilssonB,Nilsson-EkdahlK,

OdlindB,HällgrenR.Enhancedlocalproductionof

complementcomponentsinthesmallintestinesofpatients

withCrohn’sdisease.NewEnglandJournalofMedicine

1990;322(19):1345–9.

[48] SimonsenT,ElmgreenJ.Defectivemodulationof

complementinCrohn’sdisease:C3bopsonizationof

zymosanbythealternativepathway.ScandinavianJournal

ofGastroenterology1985;20(7):883–6.

[49] HodgsonH,PotterB,JewellD.C3metabolisminulcerative

colitisandCrohn’sdisease.ClinicalandExperimental

Immunology1977;28(3):490.

[50] DelChiericoF,VernocchiP,BonizziL,CarsettiR,Castellazzi

AM,DallapiccolaB.,etal.Early-lifegutmicrobiotaunder

physiologicalandpathologicalconditions:thecentralroleof

combinedmeta-omics-basedapproaches.Journalof

Proteomics2012;75(15):4580–7.

[51] JenneDE,TschoppJ.Clusterin:theintriguingguisesofa

widelyexpressedglycoprotein.TrendsinBiochemical

Sciences1992;17(4):154–9.

[52] TrougakosIP,GonosES.Regulationof

clusterin/apolipoproteinJ,afunctionalhomologuetothe

smallheatshockproteins,byoxidativestressinageingand

age-relateddiseases.FreeRadicalResearch

2006;40(12):1324–34.

[53] GasslerN,AutschbachF,HeuschenG,WitzgallR,OttoHF,

ObermullerN.ExpressionofclusterininCrohn’sdiseaseof

theterminalileum.HistolHistopathol2001;16(3):

755–62.

[54] TrougakosIP,GonosES.Oxidativestressinmalignant

progression:theroleofclusterin,asensitivecellular

biosensoroffreeradicals.AdvancesinCancerResearch

2009;104:171–210.

[55] AntonelouMH,KriebardisAG,StamoulisKE,TrougakosIP,

PapassideriIS.ApolipoproteinJ/clusterininhuman

erythrocytesisinvolvedinthemolecularprocessofdefected

materialdisposalduringvesiculation.PLoSONE

2011;6(10):e26033.

[56] DevauchelleV,BrebanM,DougadosM,FournierC,Chiocchia

G.Clusterin,apotentialpathophysiologicalactorin

rheumatoidarthritis.ArthritisResearch&Therapy

2004;6(Suppl.3):68.

[57] DericiU,TuncerC,Ebinc¸FA,MutluayR,YakaryilmazF,

KulaksizogluS.Doestheurinaryexcretionof

˛1-microglobulinandalbuminpredictclinicaldisease

activityinulcerativecolitis?AdvancesinTherapy

2008;25(12):1342–52.

[58] NewcomerME,OngDE.Plasmaretinolbindingprotein:

structureandfunctionoftheprototypiclipocalin.

BiochimicaetBiophysicaActa2000;1482(1–2):57.

[59] RonneH,OcklindC,WimanK,RaskL,ObrinkB,PetersonPA.

Ligand-dependentregulationofintracellularprotein

transport:effectofvitaminaonthesecretionofthe

retinol-bindingprotein.TheJournalofCellBiology

1983;96(3):907–10.

[60] KanaiM,RazA,GoodmanDWS.Retinol-bindingprotein:the

transportproteinforvitaminAinhumanplasma.Journalof

ClinicalInvestigation1968;47(9):2025.

[61] SoggiuA,PirasC,BonizziL,HusseinHA,PisanuS,Roncada

P.Adiscovery-phaseurineproteomicsinvestigationintype

1diabetes.ActaDiabetol2012;49(6):453–64.

[62] MainA,MillsP,RussellR,Bronte-StewartJ,NelsonLM,

McLellandA.VitaminAdeficiencyinCrohn’sdisease.Gut

1983;24(12):1169–75.

[63] BarazzoniR,ZanettiM,SemolicA,PirulliA,CattinMR,Biolo

G,etal.Highplasmaretinolbindingprotein4(RBP4)is

associatedwithsystemicinflammationindependentlyof

lowRBP4adiposeexpressionandisnormalizedby

transplantationinnonobese:nondiabeticpatientswith

chronickidneydisease.ClinicalEndocrinology

2011;75(1):56–63.

[64] KeichoN,MatsushitaI,TanakaT,ShimboT,LeHangNT,

SakuradaS,etal.Circulatinglevelsofadiponectin,leptin,

fetuin-Aandretinol-bindingproteininpatientswith

tuberculosis:markersofmetabolismandinflammation.

PLoSONE2012;7(6):e38703.

[65] OlsenI,TollefsenS,AagaardC,ReitanLJ,BannantineJP,

AndersenP,etal.IsolationofMycobacteriumavium

subspeciesparatuberculosisreactiveCD4Tcellsfrom

intestinalbiopsiesofCrohn’sdiseasepatients.PLoSONE

2009;4(5):e5641.

[66] RoncadaP,PirasC,SoggiuA,TurkR,UrbaniA,BonizziL.

Farmanimalmilkproteomics.JournalofProteomics

2012;75(14):4259–74.

[67] PirasC,SoggiuA,BonizziL,GaviraghiA,DeriuF,DeMartino

L,etal.Comparativeproteomicstoevaluatemultidrug

resistanceinEscherichiacoli.MolecularBioSystems2011.

[68] ReimundJM,ArondelY,EscalinG,FinckG,BaumannR,

DuclosB.Immuneactivationandnutritionalstatusinadult

Crohn’sdiseasepatients.DigestiveandLiverDisease

2005;37(6):424–31.

[69] Kup ˇcováV,Tureck ´yL,DetkovaZ,PrikazskaM,KeleovaA.

Changesinacutephaseproteinsafteranti-tumornecrosis

factorantibody(infliximab)treatmentinpatientswith

Crohnsdisease.PhysiologicalResearch2003;52:

89–93.

[70] DohertyNS,LittmanBH,ReillyK,SwindellAC,BussJM,

AndersonNL.Analysisofchangesinacute-phaseplasma

rheumatoidarthritisusingtwo-dimensionalgel

electrophoresis.Electrophoresis1998;19(2):355–63.

[71] VascottoC,SalzanoAM,D’AmbrosioC,FruscalzoA,

MarchesoniD,CarladiLoretoO,etal.Oxidizedtransthyretin

inamnioticfluidasanearlymarkerofpreeclampsia.Journal

ofProteomeResearch2007;6(1):160–70.

[72] RhodesJ,CollinsP.Lessonsforinflammatoryboweldisease

fromrheumatology.DigestiveandLiverDisease

2006;38(3):157–62.

[73] CylwikB,ChrostekL,Gindzienska-SieskiewiczE,

SierakowskiS,SzmitkowskiM.Relationshipbetweenserum

acute-phaseproteinsandhighdiseaseactivityinpatients

withrheumatoidarthritis.AdvancesinMedicalSciences

2010;55(1):80–5.

[74] LundgrenBA,RorsmanF,Portela-GomesGM,GrimeliusL,

EkdahlKN,NilssonB,etal.IdentificationofcomplementC3

asanautoantigenininflammatoryboweldisease.European

JournalofGastroenterology&Hepatology2010;22(4):429.

[75] KimPY,TanO,DiakiwS,CarterD,SekeryeEO,WasingerVC,

etal.IdentificationofplasmaComplementC3asapotential

biomarkerforneuroblastomausingaquantitative

proteomicapproach.JournalofProteomics2014;96:

1–12.

[76] GoldknopfIL,ShetaEA,BrysonJ,FolsomB,WilsonC,DutyJ,

etal.ComplementC3candrelatedproteinbiomarkersin

amyotrophiclateralsclerosisandParkinson’sdisease.

BiochemicalandBiophysicalResearchCommunications

2006;342(4):1034–9.

[77] HanasJS,HockerJR,CheungJY,LarabeeJL,LernerMR,

LightfootSA,etal.Biomarkeridentificationinhuman

pancreaticcancersera.Pancreas2008;36(1):61–9.

[78] UrbaniA,ModestiA,TimperioAM,BiniL,CastagnolaM,

FasanoM,etal.Integrativeproteomics:perspectivein

complexsysteminterpretation.MolecularBioSystems