Proteomic serum profile in menstrual-related and post menopause migraine

(1)

Contents lists available atScienceDirect

Journal

of

Pharmaceutical

and

Biomedical

Analysis

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

Proteomic

serum

proﬁle

in

menstrual-related

and

post

menopause

migraine

Elisa

Bellei

a

_,

_Cecilia

_Rustichelli

b,∗

_,

_Stefania

_Bergamini

a

_,

_Emanuela

_Monari

a

_,

_Carlo

_Baraldi

c

,

Flavia

Lo

Castro

d

_,

_Aldo

_Tomasi

a

_,

_Anna

_Ferrari

c

a_Department_of_Surgery,_Medicine,_Dentistry_and_{Morphological}_Sciences_with_Transplant_Surgery,_Oncology_and_Regenerative_Medicine_Relevance,

UniversityofModenaandReggioEmilia,viadelPozzo,71,41124Modena,Italy

b_Department_of_Life_Sciences,_University_of_Modena_and_Reggio_Emilia,_via_G._Campi,_103,₄₁₁₂₅_Modena,_Italy

c_Department_of_Biomedical,_Metabolic_and_Neural_Sciences,_University_of_Modena_and_Reggio_Emilia,_Via_del_Pozzo,_71,₄₁₁₂₄_Modena,_Italy d_School_of_Pharmacology_and_Clinical_Toxicology,_University_of_Modena_and_Reggio_Emilia,_Via_del_Pozzo,_71,₄₁₁₂₄_Modena,_Italy

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received10October2019

Receivedinrevisedform6February2020 Accepted9February2020

Availableonline21February2020 Keywords:

Serumprotein

Menstrual-relatedmigraine Postmenopausemigraine Proteomics

2-DE LC–MS/MS

a

b

s

t

r

a

c

t

Theaimofthis pilotstudywas toanalyzetheserumproteomic proﬁleofwomensufferingfrom

menstrual-relatedmigraine(MMgroup,n=15)andmigraineinpost-menopause(PMgroup,n=15)

incomparisonwithnon-headachecontrolfemales(Cgroup,n=15).Serumsamplesweresubjected

totwo-dimensionalgelelectrophoresis(2-DE)followedbymassspectrometry(MS)analysisfor

pro-teinidentiﬁcation.Basedon2D-gelmapsandPDQuest2-Dsoftware,13differentiallyexpressedspots,

corresponding to12 uniqueproteinsidentiﬁedbyLiquid Chromatography-Electrospray

Ionization-Quadrupole-TimeofFlight/tandemmassspectrometry(LC-ESI-QToF-MS/MS),weredetectedintheMM

andPMgroupsvsCgroup.Fiveinﬂammatoryandregulatoryofvascularintegrityproteins(prothrombin,

serumamyloidP-component,IgkappachainCregion,apolipoproteinA-I,serumamyloidA-4protein)

werefoundderegulatedinbothMMandPMgroups comparedtoCgroup;MMgroup showedthe

upregulationofotherinﬂammatoryproteinfragments(inter-alpha-trypsininhibitorheavychainH4

andcomplementC4-A)comparedtoCgroup;PMgroup,incomparisonwithCgroup,displayeda

note-worthyupregulationoftransthyretinandotherderegulatedproteins(tetranectin,alpha-1-antitrypsin,

haptoglobin,apolipoproteinA-IV)playingaroleinanti-inﬂammatoryandreparativeprocesses.In

conclu-sion,proteomicanalysiswasabletorevealdifferencesinproteinexpressionbetweenmigrainesufferers

andnon-headachewomen;asinotherneurologicaldiseasescharacterizedbyneuroinﬂammation,the

serumproteomeofmigrainewomenpresentsanabundanceofproteinsindicativeofcellulardamage,

oxidativestressandinflammation.Thisrelevantinflammatorystatus,ifconfirmedinlargerseries,could

representatargetformenstrual-relatedmigrainetreatment.

license(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Abbreviations:MM,menstrual-relatedmigraine;PM,postmenopausemigraine; THRB,prothrombin;A1AT,alpha-1-antitrypsin;APOA4,apolipoproteinA-IV;SAMP, serumamyloidP-component;IGKC,IgkappachainCregion;APOA1,apolipoprotein A-I;SAA4,serumamyloidA-4protein;TETN,tetranectin;HPT,haptoglobin;TTHY, transthyretin;ITIH4,inter-alpha-trypsininhibitorheavychainH4;CO4A, comple-mentC4-A;2-DE,two-dimensionalgelelectrophoresis;LC-ESI-QToF-MS/MS,liquid chromatography-electrosprayionization-quadrupole-timeofﬂight/tandemmass spectrometry;LC-ESI-QO-MS/MS,liquidchromatography/electrospray ionization-quadrupole-orbitrap/tandemmassspectrometry;CNS,centralnervoussystem;CSF, cerebrospinalﬂuid.

∗ Correspondingauthor.

E-mailaddress:cecilia.rustichelli@unimore.it(C.Rustichelli).

1. Introduction

More than 70 % of women suffering from migraine have menstrual-relatedattacks,i.e.fromtwodaysbeforetothreedays aftertheonsetofmenstrualﬂow.Thethirdeditionofthe Interna-tionalClassiﬁcationofHeadacheDisorders(ICHD-3)[1]includes menstrualmigraineintheappendix,dividingitinto“pure men-strualmigraine”,whentheattacksareexclusivelyrelatedtothe menstrualcycle(inatleastthreeconsecutivemenstrualcycles), and in “menstrual-related migraine”, when the attacks occur both duringandoutsidethemenstrualperiod.Theevidenceon menstrualmigraineisstilllimited.Thedecreaseinestrogen con-centrationafterbeingexposedtohighlevelsisconsideredthemain trigger[2],howeverthemechanismsthatmediatethisactionare

https://doi.org/10.1016/j.jpba.2020.113165

(2)

notknown.Menstrual-relatedattacksarelonger,moredisabling, moreresistanttothetreatment,andtendtoreappearmorethan thosethatarenotmenstrual-related.Unfortunately,thetreatment ofmenstrualmigraineisthesameasthatfor“regularmigraine” andofteninadequate[3].

Migraineis acomplex condition,which involvesa combina-tionofgenetic,environmentalandlifestylefactors[4].Proteomics studiestheexpressionandfunctionofproteins,togetherwiththe pathophysiologicalprocessesinwhichtheyareinvolved,helpingto deﬁnemorepreciselythepathogenesisofdiseases.Therefore,the proteomicapproachcouldbeappropriatetostudyamultifactorial diseasesuchasmigraine.Indeed,whilethegenomeisconstant, theproteomeiscontinuouslymodulatedbygenome-environment interactions[5].

Theaimofthispilotstudywastoanalysetheserumproteome of women suffering from menstrual-related migraine and post menopausemigraineincomparisonwithnon-headachewomen, tosearchandidentifydifferentiallyexpressedproteinsaspotential biomarkersofmenstrualmigraine,whichcouldbeinformativeon thebiologicalpathwaysthatcontributetothiscomplexdisorder. 2. Materialandmethods

2.1. Chemicalandreagents

Protease Inhibitor Cocktail, ProteoPrep® _{Immunoafﬁnity} albumin and IgG depletion Kit, Urea, Thiourea, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS), Dithiotreitol (DTT), BSA, Iodoacetamide, Tetram-ethylethylenediamine (TEMED), Tris, Ammonium Persulfate, SDS, Potassium Tetrathionate,Potassium Acetate, Glycerol, and Silver Nitrate were purchased from Merck KGaA. Ampholytes pH 3–10, Ready IPG StripTM _pH _3–10, _{Acrylamide/bis} _solution 29:1,Tris-Glycine-SDSbuffer,pH8.3,andPrecisionPlusProteinTM StandardsAllBluewerefromBio-RadLaboratories. Acetonitrile (ACN),FormicAcid(FA),AceticAcidandEthanolwereofLC–MS purity grade (Carlo Erba Reagents). The ultra-pure water was puriﬁedbyaMilli-QPlus185systemfromMillipore(MerckKGaA). 2.2. Subjectsandprocedures

Three groups participated in the study (Table 1): MM group,composedof15womendiagnosedwithmenstrual-related migrainewithoutauraaccordingtothediagnosticcriteriaofthe ICHD-3, appendix, A1.1.2 [1], aged between 18 and 45 years, whowerenot takinghormonalcontraceptives;PMgroup, com-prising15 women in menopause(spontaneous amenorrhea for atleast 12 months,not inducedby a medical condition), diag-nosedwithmigrainewithoutaura[1]andwhohadsufferedfrom menstrual-relatedmigrainebeforemenopause,agednomorethan 65years,whowerenottakinghormonereplacementtherapy;C group,15non-headachewomen(controls),agedbetween18and 45,whowerenottakinghormonalcontraceptives.Theinclusion criteria comprised theabsence of major acute medical or psy-chiatriccomorbidities,normalliverandkidneyfunctionand,for theCgroup,nomorethan3daysoftension-typeheadacheper month and no diagnosisof any other type of primary or sec-ondaryheadache,accordingtothediagnosticcriteriaofICHD-3 [1].Womenwhoweretakingmigraineprophylaxisandwhowere unabletounderstandthepurposesofthestudywereexcluded.The migrainewomenwereenrolledamongthepatientsconsecutively afferent,fortheﬁrsttime,attheHeadacheCenteroftheUniversity HospitalofModena;non-headachewomenwerepatients’friends. Allwomenprovidedtheirwrittenconsenttoparticipationinthe study,thatwasconductedinaccordancewiththeethicalprinciples

Table1

Characteristicsofthestudiedsubjectgroups(n=15ineachgroup).

Variable MMgroup N(%) PMgroup N(%) Cgroup N(%) Meanage±SD(years) 34.8±7.1 57.0±5.0a _29.2_±_7.2

Agerange(years) 21÷45 51÷65 22÷45 BMI±SD 23.1±2.5 24.1±4.1 23.3±4.2 Smoker 2(13) 1(7) 3(20) Alcoholconsumption 5(33) 7(47) 4(27) Coffeeconsumption 13(87) 10(67) 6(40) Varieddiet 14(93) 12(80) 14(93) Vegetarian/vegandiet 1(7) 2(13) 0(0) Yearsofmigraine±SD 16.8±8.5 33.1±13.5b ₀₍₀₎

Migrainedays/3months±SD 25.7±17.2 27.3±19.2 0(0) Analgesicintake/3months±SD 27.6±17.2 28.4±17.4 0(0)

Acutemigrainetreatment

Triptans 9(60) 8(53) 0(0) NSAIDs 8(53) 9(60) 0(0) Analgesiccombination 2(13) 2(13) 0(0) Comorbidities Hypertension 1(7) 4(27) 1(7) Allergicrhinitis 3(20) 3(20) 4(27) Hypercholosterolemia 1(7) 3(20) 0(0) Thyroiditis 1(7) 4(27) 1(7) Concomitanttreatment Antihypertensiveagents 1(7) 4(27) 1(7) Antihistamines 3(20) 3(20) 4(27) Statins 1(7) 3(20) 0(0) Levothyroxine 1(7) 4(27) 1(7)

a_PM_vs_MM_group,_P_<_0.01;_PM_vs_C_group,_P_<_0.001;_one-way_ANOVA_test. b_PM_vs_MM_group,_P_<_0.001;_one-way_ANOVA_test.

oftheHelsinkiDeclaration,lastedition(2013)andapprovedbythe EthicsCommitteeoftheProvinceofModena(protocoln.86/16).

Inwomenoffertileage,bloodforproteomicanalysiswastaken betweentheseventhandtenthday,startingfromtheﬁrstdayof menstruationand,inmigrainewomen,atleasttwodaysafterthe lastmigraineattack.Foreachwoman,duringamedical examina-tion,aformwascompiledforthecollectionofpersonaldata,life habits,clinicalhistory, andforthepatientsalsothehistory and characteristicsofmigraine.

2.3. Samplecollectionandpreparation

Venous blood (10 mL) was collected in the morning from migrainepatientsandnon-headachecontrolsatfastandallowedto clotatroomtemperaturefor1h;serumwassubsequentlyobtained bycentrifugationat2000xgfor10minat+4◦C.Topreserve pro-teins,serumwasaddedwithaproteaseinhibitorcocktailpriorto bedividedintoaliquotsandstoredat₋₈₀◦Cuntilproteomic anal-ysis.For thispurpose, thecomplexity ofserumwasreducedby theselectivedepletionofthetwomostabundantserumproteins usingtheProteoPrep®_{Immunoafﬁnity}_albumin_and_IgG_depletion Kit,accordingtothemanufacturer’sprotocol.Foreachgroupof patients,aswellasforthecontrolgroup,serumsampleswere com-binedtoform5pools/group.Totalproteincontentofeachpool wasmeasuredspectrophotometricallyat␭=595nm,usingBSAas standardforthecalibrationcurve.

2.4. Two-dimensionalgelelectrophoresis(2-DE)andimage analysis

Theobtainedserumsamplesweresubjectedto2-DEanalysis. For theﬁrst-dimensionseparation,a total of100␮g ofprotein wasdilutedwitharehydrationbuffercomposedof6Murea,2 Mthiourea,4%CHAPS,25mMDTT,0.2%ampholytespH3–10,and thenloadedonto17cm, immobilizedpHgradientstrips(Ready IPGStripTM_), _pH_range_3-10._IEF _was_conducted_at ₊₂₀◦_C, _ﬁrst performinganactivelyrehydrationofthestripsat50Vfor12h,

(3)

thenincreasingthevoltageat250Vfor15 min,rampingupto 10.000V for3 h, and finallyfocusing toreach75.000V-hours. TheIEFwasfollowedbytheequilibrationofthestripsin1%DTT and2.5%Iodoacetamideinequilibrationbuffer(6Murea,50mM Tris−HClpH8.8,30%glycerol,2%SDS).Thesecond-dimension sep-arationwasperformedbySDS-PAGEon8–16%polyacrylamide gradientgel(Acrylamide/bissolution29:1,1.5MTris−HCl,pH8.8, 10%SDS,1%TEMED,10%ammoniumpersulfate),using1X Tris-Glycine-SDS,pH8.3asrunningbufferduringtheelectrophoretic run.Finally,proteinswerestainedwithasilver nitratestaining protocol,compatiblewithMSdetection.Briefly,thegelswere incu-batedinafixingbuffersolution(40%ethanol/10%aceticacid)for4 h,thentheywerefirstwashedin40%ethanolandsubsequentlyin ultra-puredeionizedwater.Afterbeingsensitizedbytheenhancer solution(0.3%potassiumtetrathionate,0.5Mpotassiumacetate, 30%ethanol), thegelswerecoveredwitha 0.2 %silver nitrate solution,incubatedfor75mininthedarkandthendeveloped, ter-minatingtheprocesswiththeadditionoftheblockersolution(4% Tris,2%aceticacid).

Eachgelimagewasacquiredbyacalibrateddensitometer,series GS-800(Bio-RadLaboratories,Hercules,CA,USA),andanalyzedby thePDQuest2-Dimageanalysissoftware,version7.3.1(Bio-Rad Laboratories,Hercules,CA,USA). Thissoftwaredetects thespot stainintensity,asopticaldensity(OD),andthespotarea, provid-ingaﬁnalvalueexpressedas“spotvolume”,thatistheproductof ODvalueperarea(ODxmm2).The2Dgelimageswerecompared todetectdifferentiallyexpressedproteinspotsamongthe differ-entgroups.Thefold-changesofspotabundancewereachievedas ratiobetweenthevaluesofspotvolumemeasuredinbothpatients’ groupsvscontrols.

2.5. MSproteinidentiﬁcation

Proteinswereextractedfromthespotbyan“in-gel”trypsin digestionprotocol,fullyreportedinSupplementaryMaterial. After-ward,peptides wereundoubtedly identifiedusingtwodifferent LC–MS/MSprocedures.Inthefirstapproach,analyseswere per-formed using an Agilent 1200 series LC systemcoupled to an Agilent6520Q-TOFmassspectrometer(LC-ESI-QTof-MS/MS) (Agi-lent,Waldbronn,Germany).Thepeptidemixturesweredissolved in5%aqueousFAandloaded(4.0␮L)ontoareversed-phase40nL enrichmentChip(G4240-62001,Agilent)withamobilephaseof water-ACN(97:3,v/v;+0.1FA)ataflowof3.0␮L/min.The precol-umnwasthenswitchedin-linewiththenanoflowpumpandthe peptideswereanalyzedonaZORBAX300SB-C18(43mm×75␮m ID;5␮mparticlesize,Agilent)analyticalcolumnwithamobile phase of(A)water-ACN (97:3, v/v;+0.1FA)and (B)water-ACN (5:95,v/v;+0.1FA)usingthefollowinggradientprogram:0to0.5 min,isocraticat3%(B);0.5–17min,lineargradientfrom3to25 %(B);17–20min,lineargradientfrom25to35%(B);20–21min, lineargradientfrom35to95%(B);21–25min,isocraticat95% (B);25–26min,lineargradientfrom95to3%(B).Theequilibration periodbetweeneachrunwas13minlong;totalruntime:39min. Theflowratewassetat400nL/min.TheESIsourceoperatedin positiveionizationmode;thedryinggas(nitrogen)flowand tem-peraturewere6L/minand350◦C,respectively,thecapillaryvoltage wassetat1950Vandthefragmentorvoltagewas160V.MSand MS/MSspectrawereacquiredinAutoMS/MSmodesetwiththe followingparameters:MSscan,m/z300–1700umaat2spectra/s andMS/MSscan,m/z50–1700umaat3spectra/s.Datawere col-lectedinacentroidmodewithaprecursorisolationwindowset at4uma.The5mostintensemulti-chargedionswereselectedfor MS2_{nitrogen-promoted}_{collision-induced}_dissociation_with_active exclusiononwith1repeatandreleaseafter0.17min.Precursors withchargestate+1wereexcludedandthepreferredchargestate was:2,3,>3,unknown;thepropercollisionenergywascalculated

asfollows:3.6×(m/z)/100+2.TheAgilentMassHunter Worksta-tionAcquisition softwareversion B.05.00 (B2043) wasusedfor instrumentcontrol,dataacquisition,qualitativeandquantitative dataanalysis.Theinstrumentwascalibratedwithastandardmix prioreachanalysisbatchtoensuremassaccuraciesatorbelowthe 3ppm.

Replicated spots were cut from different gels and further analyzedbyLC-ESI-QO-MS/MSusinganUHPLC-MSQExactiveTM (ThermoFisherScientific, Reinach,Switzerland)toconfirm pro-teinidentification;theoptimizedexperimentalconditionsandthe resultsarereportedinSupplementaryMaterial.Protein identifica-tionswereobtainedbyMASCOTsearchengine(version2.4)against theUniProtknowledgebasedatabase(UniProtKB).

2.6. Dataandstatisticalanalysis

Allcollecteddata,madeanonymous,wereenteredintoaspecific database.Descriptiveanalysisofallvariableswasconducted. Statis-ticalanalysiswascarriedoutbyStataIC13software.Thecontinuous variables normallydistributedwereexpressedasmean ± stan-darddeviation(SD)andthedichotomousvariablesaspercentages. Thecomparison betweenmeanswasdoneby one-wayANOVA, whileFisher’sexacttestwasusedforbinaryvariables.Differences wereconsideredsignificantwithaPvalue<0.05.Regardingthe fold-changesofspotabundance,values>1.5wereconsideredas significant.

3. Resultsanddiscussion

Theevaluationoftheproteinprofilesofhumanblood,using pro-teomicmethodologies,providesawindowonthestateofillnessor healthofanindividual.Accordingly,theresultsofourstudyshowed thattheserumproteomeofmigrainepatients(MMandPMgroups), duringthe attack-free interval,wasdifferent fromthat of non-headachewomen(Cgroup).Basedonproteinsresolutionby2-DE (Fig.1)andPDQuest2-Dimageanalysissoftware,13proteinspots weredifferentiallyexpressedinMMandPMgroupscomparedto C group.These proteinswereidentified byLC-ESI-QToF-MS/MS analysesas12uniqueproteins,sincetwospotsresulteddifferent isoformsofthesameprotein(Table2).Specifically,thefirstand secondcolumnsofTable2reportedtheproteinentrynameand theprimaryaccessionnumber,respectively,bothderivedfromthe UniProtknowledgedatabase,whilethethirdcolumnshowedthe proteinfullnames.MSdataareillustratedinthelast4columns:the highestscoresobtainedwithMASCOTsearchengine,thenumber oftotalpeptideandthesignificantpeptidesmatchingthe identi-fiedproteins,thenumberoftotalandsignificantsequences,andthe percentageofsequencedaminoacidsforeachidentifiedprotein.

Inparticular,prothrombin(THRB),serumamyloidP-component (SAMP),IgkappachainCregion(IGKC),apolipoproteinA-I(APOA1), serumamyloidA-4protein(SAA4)resulteddifferentiallyexpressed inbothMMandPMgroupscomparedtoCgroup;2protein frag-ments,inter-alpha-trypsininhibitorheavychainH4 (ITIH4)and complement C4-A (CO4A)were differentiallyexpressedonly in MM group vsC group; other 5 proteins, transthyretin (TTHY), tetranectin(2spots,TETN1_and_TETN2_),_{alpha-1-antitrypsin}_(A1AT), haptoglobin(HPT)andapolipoproteinA-IV(APOA4)were differen-tiallyexpressedonlyinPMgroupvsCgroup.Moreover,asecond setofreplicatedproteinspotswasfurtheranalyzedby LC-ESI-QO-MS/MSSystem,conﬁrming allproteinspreviously identiﬁedby LC-ESI-QToF-MS/MSanalysis(resultsandmethodarepresentedin SupplementaryMaterial).

BymeansoftheresearchcarriedoutbyUniProtandTheHuman Protein Atlas databases, most (58 %) of theidentiﬁed proteins wereinﬂammatoryandbinding/transportproteins,whilethe

(4)

oth-Fig.1.Representative2-DEgelsobtainedfromserumofnon-headachecontrolfemales(Cgroup),menstrually-relatedmigraine(MMgroup)andpost-menopausalmigraine patients(PMgroup).Thedifferentiallyexpressedproteinspotsdetectedinmigraineursvscontrolsarehighlightedbyacircle;abbreviationsofproteinnamesderivefrom theUniProtdatabase(proteinentryname)andcorrespondtothosereportedinTable2.

2-DE:17cmIPGstrip,pH3–10,inﬁrst-dimensionseparation(IEF)and8–16%polyacrylamidegradientgelforsecond-dimensionseparation(SDS-PAGE).Proteinspots evidencedbySilver-staining.Molecularweightmarker(kDa):PrecisionPlusProteinTM_Standards_All_Blue.

erswereinvolved inlipidmetabolism,immunityand inhibition functions(Fig.2).

3.1. ProteinsdifferentiallyexpressedinbothMMandPMgroups vsCgroup

Thefold-changesofproteinexpressionaregraphicallyshown inFig.3(therawdataareprovidedasSupplementaryMaterial). Fiveproteins,deregulatedinbothMMandPMgroupsvscontrols suggestedaparticularlinkbetweentheseproteinsandmigraine withoutaura.Speciﬁcally,3proteinswereup-regulatedinmigraine vsnon-headachewomen:THRB(MM+2.19;PM+4.06),SAMP(MM +3.88;PM+2.19)andIGKC(MM+1.86;PM+1.76),while2 pro-teins,APOA1(MM−16.8;PM−2.01)andSAA4(MM−2.05;PM −4.36),werefounddown-regulated.THRB, aserineprotease, is fundamental,ascoagulationfactorII,inbloodhomeostasis reg-ulatinginﬂammatory response, cell proliferation,apoptosis and maintainingvascularintegrity[6].THRBwouldbealsoinvolved

in neuroinflammation[7]. Its blood levels increase in migraine patientsand thistendencytohypercoagulabilitywasassociated withtheriskofstrokeinmigrainewithaura.Therefore,ourresults supportedthe hypothesis that THRB couldrepresent a specific markerofmigraine[8].Anotherup-regulatedinflammatory pro-teinwasSAMP.Itisaconstituentofamyloiddeposit,whichbelongs tothepentraxinesuperfamily,acutephaseproteins,inducedby IL-6 thatmediates inflammationand innate humoralimmunity [9].SAMPlevelswerehigherinMMthaninPMgroup.Probably, inflammatoryprocesseschangewithdiseaseprogressionbecause migrainepatientswithaheadachehistorylastingmorethanfive yearspresentlowerserumlevelsofpentraxin-3thanpatientswith arelativelyrecentdiagnosis[10].Thethirdup-regulatedprotein wasIGKC,apromisingbiomarkerinthediagnosisand monitor-ingofthetherapeuticresponseofseveralsolidtumours[11].Thus, asintumours,aninflammatory andimmuneresponsecouldbe alsoactivatedinmigraine.Amongtheunder-expressedproteins therewasAPOA1,themainpolypeptideofhumanplasma

(5)

high-Table2

DifferentiallyexpressedserumproteinsidentiﬁedbyLC-ESI-QToF-MS/MSanalysis.

Entrynamea _Acc._numberb _Protein_name _Scorec _Pep/pep._sign.d _Seq/seq._sign.e _Cov.f_(%)

ProteinsdifferentiallyexpressedinbothMMandPMgroupsvsCgroup

THRB P00734 Prothrombin 154 12/8 7/4 10

SAMP P02743 SerumamyloidP-component 87 7/6 5/5 21

IGKC P01834 IgkappachainCregion 655 40/33 7/6 86

APOA1 P02647 ApolipoproteinA-I 493 78/35 24/16 63

SAA4 P01834 SerumamyloidA-4protein 53 37/7 7/4 47 ProteinsdifferentiallyexpressedonlyinMMgroupvsCgroup

ITIH4 Q14624 Inter-alpha-trypsininhibitor

heavychainH4(fragment)

61 9/5 5/3 6

CO4A P0C0L4 ComplementC4-A(fragment) 177 20/13 7/5 4 ProteinsdifferentiallyexpressedonlyinPMgroupvsCgroup

TTHY P02766 Transthyretin 105 25/8 7/5 42

TETN1 _P05452 _Tetranectin ₇₈ _5/2 _4/1 ₂₁

TETN2 _P05452 _Tetranectin ₄₃ _4/3 _2/1 ₁₁

A1AT P01009 Alpha-1-antitrypsin 171 25/10 11/6 31

HPT P00738 Haptoglobin 170 40/11 9/6 20

APOA4 P06727 ApolipoproteinA-IV 1083 139/72 39/28 79

a_Entry_name_from_UniProt_knowledge_database₍_{www.uniprot.org}_). b_Primary_accession_number_from_UniProt_database.

c _The_highest_score_with_MASCOT_search_engine.

d _Number_of_total_{peptides/significant}_peptides_matching_the_identified_protein. e_Number_of_total_{sequences/significant}_sequences.

f _Sequence_coverage:_percentage_of_sequenced_amino_acids_for_each_identiﬁed_protein.

Fig.2.Mainfunctionsoftheidentiﬁedproteins.Inﬂammation:Haptoglobin(HPT),Inter-alpha-trypsininhibitorheavychainH4(ITIH4),Prothrombin(THRB),Serumamyloid A-4protein(SAA4);Binding/transport:SerumamyloidP-component(SAMP),Transthyretin(TTHY),Tetranectin(TETN);Immunity:IgkappachainCregion(IGKC), Comple-mentC4-A(CO4A);Lipidmetabolism:ApolipoproteinA-I(APOA1),ApolipoproteinA-IV(APOA4);Inhibition:Alpha-1-antitrypsin(A1AT).Proteinfunctionswereretrieved bytheUniProtdatabase.

density lipoprotein (HDL).It is a negative acute phase protein, which expression is reduced more than 25 % during theacute phaseresponse.Itexhibitsalsoanti-apoptotic,anti-inﬂammatory andantioxidantfunctions[12].Theotherunder-expressedprotein wasSAA4.Itbelongstotheserumamyloidfamilyandisaminor acute-phasereactant inhumans [13]. Invitro studiessuggest a chemo-attractive,protectivefunctioninthebrain.Thedeclinein itsplasmaconcentrationshasbeenhypothesizedasamarkerin theearlydiagnosisofAlzheimer’sdisease[14].

Essentially, ourresults showed that a subclinical inflamma-torystatewaspresentin migrainewomen, supportingtherole ofinflammationinthepathogenesisofmigraine.Inparticular,it isassumedthattheactivationofthetrigeminalsensorypathways leadstoaninflammatoryresponseinvolvingimmune,vascularand neuroncellswithreleaseofneuropeptidesandpro-inflammatory

neurotransmitters,capableofinducingastateofsterile inﬂamma-tionintheintracranialmeningesand,consequently,sensitization of trigeminalmeningeal nociceptors. Neuroinﬂammation in the trigeminovascularsystem,althoughitdoesnotcompletelyexplain theonsetoftheattack,couldbecrucialinpromotingthe progres-sionofepisodicmigraineintoachronicform[15].

3.2. ProteinsdifferentiallyexpressedonlyinMMgroupvsCgroup Alwaystakingthefold-changeasaparameter(Fig.3),2protein fragmentsresulted up-regulatedexclusivelyinMM group com-paredtocontrols,namelyITIH4(+1.89)andCO4A(+2.04).ITIH4 belongs tointer-alpha trypsininhibitorsand isa type II acute-phaseproteininvolvedininﬂammatoryresponses;itsbloodlevels risefrom1.4to3timesinacute-phaseprocesses[16].Moreover,it

(6)

Fig.3.Representationofthefold-changedatashowingtheup-regulatedanddown-regulatedproteinsinthemenstrually-relatedmigraine(MM)andpost-menopausal migraine(PM)groupscomparedtocontrol(C)group.Thefold-changesofspotabundancewerecalculatedasratiobetweenthevaluesofspotvolumedetectedbythe PDQuest2-Dimageanalysissoftware.

hasbeenassociatedtoneuroinflammationindepression[17].CO4A isamediatoroflocalinflammationwithantimicrobialproperties thatcontributestopropagatingtheinflammatoryresponseviathe classicalcomplementpathway[18].Menstrual-relatedmigraineis included,inthelastICHD-3classification[1],onlyintheappendix, pendingfurtherscientificevidencetocharacterizeit[19].Itcanbe assumedthatthehighinflammatorystatefoundinMMwomen couldbeatthebasisoftheseverityofthemenstrualattacks. Inter-estingly,theproteinfragmentsareverypromisingbiomarkers.In fact,theirgenerationfromthespecificprocessesofadiseasecould reducetheoverlapbetweendiagnosticgroups[20].

3.3. ProteinsdifferentiallyexpressedonlyinPMgroupvsCgroup Lastly,4proteinswerefoundup-regulatedonlyinPMgroup vs C group (Fig. 3): TTHY (+12.2), TETN (+1.94, spot 1; +2.52, spot2),A1AT(+2.16)andHPT(+2.33),whileAPOA4was down-regulated(−2.61).Themostrelevantaspectwastheup-regulation of TTHY,a carrier proteinwith theprimary function of retinol andthyroidhormonestransport. TTHYpromotesthe inflamma-toryresponseandinthecentralnervoussystem(CNS)isassociated withcognitivefunctions,memory,andemotioncontrolthrough theregenerationofdamagedneurons[21].Inanimalmodelsof oli-goemia,atypicalconditionofshockandmigraine,TTHYincreases in cerebrospinalfluid (CSF), as an indicatorof oxidative stress. Overexpressionofthisproteinisreportedindifferentconditions ofchronicpainwithoxidative stress,where TTHYcouldensure neuroprotection[22].TwoTETNisoformsweremoreexpressedin PMgroup.ThisproteinbelongstotypeClectins.Itisan indica-torofproteolysis,fibrinolysisandtissueremodelling.Infact,TETN concentrationishighintheCSFafteranepilepticattack[23]. Fur-thermore,TETNisapotentialbiomarkerforParkinson’sdiseaseand coronaryarterydisease[24,25].A1ATandHPT,bothpositiveacute phaseproteins,werealsoupregulated.SerumA1ATconcentration increasesinresponsetoinflammationortissuetrauma.By inhibit-ingtheproductionofproinflammatorycytokines,A1ATlimitsthe damageininflamedtissues[26].HPTconcentrationsubstantially increasesduringacuteinflammationinresponsetointerleukin-6. Itinhibitsprostaglandinsynthesis,angiogenesisandexerts anti-inflammatory,antioxidant,protectiveforendotheliumandweak antibacterialactivity[27].IncreasedplasmaHPTconcentrationhas beenfoundinpatientswithParkinson’sdisease,Guillain-Barré syn-drome,multiplesclerosisandidiopathicintracranialhypertension [28].Finally,APOA4wasdownregulatedinPMgroup.Thisprotein playsacrucialroleinlipidabsorption;moreover,itisinvolvedin

mitigatingtheexcessiveinflammatoryactivationattheendothelial levelandshowsanti-atheroscleroticproperties[29].Thus,inthe PMgroup,proteinswithanti-inflammatoryfunctions,involvedin protectiveandreparativeprocesseswerealsoabundant.Therefore, itcanbehypothesizedthatthelonghistoryofmigraineinPMgroup hadtriggeredanoppositeresponse,defensiveagainst inflamma-tion.Thisphenomenon,inadditiontothecessationofhormonal fluctuations,couldcontributetothereductionoftheseverityand prevalenceofmigraineaftermenopause[19].

Notably,manyofthedifferentiallyexpressedproteinsinthe MM and PMgroups (Table2) had a dualrole, as regulators of vesselintegrityandinflammation.Migraine,especiallywithaura, hasbeenassociatedwithanunfavourableriskprofileforischemic vascularevents.Theexactbiologicalmechanismofthisincreased riskisunknown[30].Fromourproteomicanalysis,the deregu-lationofproteinswithadoubleinvolvement,intheinflammatory responseandintheendothelialactivation,couldpreciselyunderlie theincreasedriskofischemiceventsandsupportthelinkbetween migraineandvasculardiseases.

3.4. Limitationsandstrengths

Themainlimitationsofourpilotstudywerethelowsample sizeandthecross-sectionaldesignthatdidnotallowestablishing whetherthederegulatedproteinpattern wasthecauseor con-sequence of migraine.Inflammation couldbea normal process inmenstrualcycleandmenopause(moredetailsin Supplemen-taryMaterial).TheserumproteomicprofilefoundinthePMgroup couldbeaffectedbythehigherageof thesepatients(Table1). However,therewerenosignificantdifferencesbetweenthePM andMMgroupsduetocomorbidities,migrainedaysandanalgesic consumption. PMwomenonly sufferedfroma longermigraine history. Moreover, we had noelements to define whetherthe serum proteome couldreflect theenvironment in CNS; never-theless,thesameproteinshavebeenfoundinotherneurological diseases,characterized,likemigraine,byneuroinflammationsuch asParkinson’sdisease,multiplesclerosis,Alzheimer’sdiseaseand chronicpain.Our studywasconductedrigorously:the differen-tiallyexpressedproteinswere firstidentified and subsequently verifiedandvalidatedbytwo differentmassspectrometers,the LC-ESI-QTof-MS/MSandLC-ESI-QO-MS/MSsystems,respectively. Allproteinsidentifiedinthefirstanalysissessionwereconfirmed bythesecondanalysis.Tothebestofourknowledge,thisisthefirst studyprovidinginsightintothedifferentialexpressionofproteins inwomensufferingfrommenstrual-relatedandpostmenopausal

(7)

migrainewithrespecttonon-headachewomen;so,todate,we havenopossibilityofcomparison.Therefore,ourresultsneedtobe conﬁrmedregardingtheirdiagnosticvalueonlargersamples,also usingmoresensitiveanalyticalmethodsinnextinvestigations. 4. Conclusions

Insummary,theresultsofourpilotstudyshowedthat:1.the proteomicapproachwasabletodistinguishbetweenmigraineand non-headachestatus;2.migrainewomen,inbothfertileandpost menopauseage,presentedagreatabundanceofinflammatoryand regulatoryofvascularintegrityproteins;3.MMgroupshoweda greatdown-regulationofAPOA1(−16.8),anegativeacutephase protein,intimatelylinkedwithinflammationandoverexpression ofinflammatoryproteinfragments(ITIH4andCO4A)comparedto theCgroup;thisabundantinflammatoryproteomemightexplain theseverityofthemenstrualattacksandpoorresponsetoacute treatments;4.PMgroup,olderthanMM,displayedanoteworthy up-regulationofTTHY(+12.2)andanumberofproteins differen-tiallyexpressed,higherthanCgroup;theseproteinsplayedalsoan importantroleinanti-inflammatoryandreparativeprocesses, sug-gestingthatthelonghistoryofmigrainecouldinduceaprotective response.Finally,therelevantinflammatorystatushighlightedby proteomicanalysis,ifconfirmedinlargerseries,couldrepresenta targetformenstrual-relatedmigrainetreatment.

Creditauthorstatement

Allauthorsparticipatedin studydesign, performedresearch, data analysis, and drafting of manuscript content. All authors reviewedandapprovedthemanuscript.

Funding

ThisstudywascarriedoutbyresearchfundsfromtheUniversity ofModenaandReggioEmilia(Italy)attributedtoDr.AnnaFerrari (FAR2016).

DeclarationofCompetingInterest

Theauthorsdeclarethattheyhavenoknowncompeting ﬁnan-cialinterestsorpersonalrelationshipsthatcouldhaveappearedto inﬂuencetheworkreportedinthispaper.

Acknowledgements

Theauthorsthankthe“CentroInterdipartimentaleGrandi Stru-menti”(CIGS)oftheUniversityofModenaandReggioEmiliafor technicalassistance.LC–MS/MSinstrumentsatCIGSweregranted bytheFoundation“FondazioneCassadiRisparmiodiModena– FCRM.”

AppendixA. Supplementarydata

Supplementarymaterial relatedto thisarticle canbe found, in theonline version, at doi:https://doi.org/10.1016/j.jpba.2020. 113165.

References

[1]HeadacheClassiﬁcationCommitteeoftheInternationalHeadacheSociety (IHS),TheinternationalclassiﬁcationofheadachedisordersCephalalgia,38, 3rdedition,2018,pp.1–211,http://dx.doi.org/10.1177/0333102417738202. [2]B.W.Somerville,Estrogen-withdrawalmigraine:I.Durationofexposure

requiredandattemptedprophylaxisbypremenstrualestrogen administration,Neurology25(1975)239–244,http://dx.doi.org/10.1212/ WNL.25.3.239.

[3]E.A.MacGregor,T.W.Victor,X.Hu,Q.Xiang,R.A.Puenpatom,W.Chen,J.C. Campbell,Characteristicsofmenstrualvsnonmenstrualmigraine:aposthoc, within-womananalysisoftheusual-carephaseofanonrandomized menstrualmigraineclinicaltrial,Headache50(2010)528–538,http://dx.doi. org/10.1111/j.1526-4610.2010.01625.x.

[4]M.Kowalska,M.Prendecki,W.Kozubski,M.Lianeri,J.Dorszewska,Molecular factorsinmigraine,Oncotarget7(2016)50708–50718,http://dx.doi.org/10. 18632/oncotarget.9367.

[5]C.Vogel,E.M.Marcotte,Insightsintotheregulationofproteinabundance fromproteomicandtranscriptomicanalyses,Nat.Rev.Genet.13(2012) 227–232,http://dx.doi.org/10.1038/nrg3185.

[6]H.Krenzlin,V.Lorenz,B.Alessandri,Theinvolvementofthrombininthe pathogenesisofglioblastoma,J.Neurosci.Res.95(2017)2080–2085,http:// dx.doi.org/10.1002/jnr.24049.

[7]K.Göbel,P.Kraft,S.Pankratz,C.C.Gross,C.Korsukewitz,R.Kwiecien,R. Mesters,B.E.Kehrel,H.Wiendl,C.Kleinschnitz,S.G.Meuth,Prothrombinand factorXareelevatedinmultiplesclerosispatients,Ann.Neurol.80(2016) 946–951,http://dx.doi.org/10.1002/ana.24807.

[8]G.E.Tietjen,S.A.Collins,Hypercoagulabilityandmigraine,Headache58 (2018)173–183,http://dx.doi.org/10.1111/head.13044.

[9]B.Bottazzi,A.Inforzato,M.Messa,M.Barbagallo,E.Magrini,C.Garlanda,A. Mantovani,ThepentraxinsPTX3andSAPininnateimmunity,regulationof inﬂammationandtissueremodelling,J.Hepatol.64(2016)1416–1427,http:// dx.doi.org/10.1016/j.jhep.2016.02.029.

[10]M.Ceylan,O.F.Bayraktutan,S.Becel,Ö.Atis,A.Yalcin,D.Kotan,Serumlevels ofpentraxin-3andotherinﬂammatorybiomarkersinmigraine:association withmigrainecharacteristics,Cephalalgia36(2016)518–525,http://dx.doi. org/10.1177/0333102415598757.

[11]J.P.Pandey,E.Kistner-Grifﬁn,L.Black,A.M.Namboodiri,M.Iwasaki,Y. Kasuga,G.S.Hamada,S.Tsugane,IGKCandFc␥Rgenotypesandhumoral immunitytoHER2inbreastcancer,Immunobiology219(2014)113–117,

http://dx.doi.org/10.1016/j.imbio.2013.08.005.

[12]P.Sirniö,J.P.Väyrynen,K.Klintrup,J.Mäkelä,M.J.Mäkinen,T.J.Karttunen,A. Tuomisto,DecreasedserumapolipoproteinA1levelsareassociatedwithpoor survivalandsystemicinﬂammatoryresponseincolorectalcancer,Sci.Rep.7 (2017)5374,http://dx.doi.org/10.1038/s41598-017-05415-9.

[13]C.L.Chen,T.S.Lin,C.H.Tsai,C.C.Wu,T.Chung,K.Y.Chien,M.Wu,Y.S.Chang, J.-S.Yu,Y.-T.Chen,Identiﬁcationofpotentialbladdercancermarkersinurine byabundant-proteindepletioncoupledwithquantitativeproteomics,J. Proteomics85(2013)28–43,http://dx.doi.org/10.1016/j.jprot.2013.04.024. [14]L.Ijsselstijn,J.M.Papma,L.J.M.Dekker,W.Calame,C.Stingl,P.J.Koudstaal,

N.D.Prins,P.A.E.SillevisSmitt,T.M.Luider,Serumproteomicsinamnestic mildcognitiveimpairment,Proteomics13(2013)2526–2533,http://dx.doi. org/10.1002/pmic.201200190.

[15]L.Edvinsson,K.A.Haanes,K.Warfvinge,Doesinﬂammationhavearolein migraine?Nat.Rev.Neurol.15(2019)483–490,http://dx.doi.org/10.1038/ s41582-019-0216-y.

[16]L.Zhuo,V.C.Hascall,K.Kimata,Inter-␣-trypsininhibitor,acovalent protein-glycosaminoglycan-proteincomplex,J.Biol.Chem.279(2004) 38079–38082,http://dx.doi.org/10.1074/jbc.R300039200.

[17]Q.Wang,X.Su,X.Jiang,X.Dong,Y.Fan,J.Zhang,C.Yu,W.Gao,S.Shi,J.Jiang, W.Jiang,T.Wei,iTRAQtechnology-basedidentiﬁcationofhumanperipheral serumproteinsassociatedwithdepression,Neuroscience330(2016) 291–325,http://dx.doi.org/10.1016/j.neuroscience.2016.05.055.

[18]T.E.Hugli,Biochemistryandbiologyofanaphylatoxins,Complement3(1986) 111–127.

[19]K.G.Vetvik,E.A.MacGregor,Sexdifferencesintheepidemiology,clinical features,andpathophysiologyofmigraine,LancetNeurol.16(2017)76–87,

http://dx.doi.org/10.1016/S1474-4422(16)30293-9.

[20]Y.R.Yang,S.L.Liu,Z.Y.Qin,F.J.Liu,Y.J.Qin,S.M.Bai,Z.Y.Chen,Comparative proteomicsanalysisofcerebrospinalﬂuidofpatientswithGuillain–Barré syndrome,Cell.Mol.Neurobiol.28(2008)737–744,http://dx.doi.org/10. 1007/s10571-007-9257-7.

[21]J.Brouillette,R.Quirion,Transthyretin:akeygeneinvolvedinthe maintenanceofmemorycapacitiesduringaging,Neurobiol.Aging29(2008) 1721–1732,http://dx.doi.org/10.1016/j.neurobiolaging.2007.04.007. [22]M.Sharma,S.Khan,S.Rahman,L.R.Singh,Theextracellularprotein,

transthyretinisanoxidativestressbiomarker,Front.Physiol.10(2019)5,

http://dx.doi.org/10.3389/fphys.2019.00005.

[23]L.Wang,Y.Pan,D.Chen,Z.Xiao,Z.Xi,F.Xiao,X.Wang,Tetranectinisa potentialbiomarkerincerebrospinalﬂuidandserumofpatientswith epilepsy,Clin.Chim.Acta411(2010)581–583,http://dx.doi.org/10.1016/j. cca.2010.01.022.

[24]E.S.Wang,H.B.Yao,Y.H.Chen,G.Wang,W.W.Gao,Y.R.Sun,J.G.Guo,J.W.Hu, C.C.Jiang,J.Hu,ProteomicanalysisofthecerebrospinalﬂuidofParkinson’s diseasepatientspre-andpost-deepbrainstimulation,Cell.Physiol.Biochem. 31(2013)625–637,http://dx.doi.org/10.1159/000350082.

[25]Y.Chen,H.Han,X.Yan,F.Ding,X.Su,H.Wang,Q.Chen,L.Lu,R.Zhang,W.Jin, Tetranectinasapotentialbiomarkerforstablecoronaryarterydisease,Sci. Rep.5(2015)17632,http://dx.doi.org/10.1038/srep17632.

[26]O.Repetto,L.Mussolin,C.Elia,L.Martina,M.Bianchi,S.Buffardi,A.Sala,R. Burnelli,M.Mascarin,V.DeRe,Proteomicidentiﬁcationofplasmabiomarkers inchildrenandadolescentswithrecurrentHodgkinlymphoma,J.Cancer9 (2018)4650–4658,http://dx.doi.org/10.7150/jca.27560.

(8)

[27]C.B.F.Andersen,K.Stødkilde,K.L.Sæderup,A.Kuhlee,S.Raunser,J.H. Graversen,S.K.Moestrup,Haptoglobin,Antioxid.RedoxSignal.26(2017) 814–831,http://dx.doi.org/10.1089/ars.2016.6793.

[28]J.Hannerz,K.Ericson,D.Greitz,P.H.B.Skejo,G.Edman,Orbitalphlebography inidiopathicintracranialhypertensionandchronictension-typeheadache, ActaRadiol.ShortRep.2(2013),204798161349886,http://dx.doi.org/10. 1177/2047981613498861.

[29]J.Peng,X.Li,ApolipoproteinA-IVmayinhibitinﬂammationviaregulating NLRP3/IL-1pathtoanti-atherosclerosis,Int.J.Cardiol.278(2019)223,http:// dx.doi.org/10.1016/j.ijcard.2018.11.103.

[30]T.Kurth,P.Ridker,J.Buring,Migraineandbiomarkersofcardiovascular diseaseinwomen,Cephalalgia28(2008)49–56,http://dx.doi.org/10.1111/j. 1468-2982.2007.01467.x.