Characterization of membrane metal threads by proteomics and analysis of a 14th c. thread from an Italian textile

Availableonlineat

ScienceDirect

www.sciencedirect.com

Original article

Characterization of membrane metal threads by proteomics and analysis of a 14th c. thread from an Italian textile

Aleksandra K. Popowich

^a,b

, Timothy P. Cleland

^b

, Caroline Solazzo

^b,∗

aDepartmentofChemistry,UniversityofAlberta,11227SaskatchewanDrive,Edmonton,AlbertaT6G2G2,Canada

bSmithsonian’sMuseumConservationInstitute,4210SilverHillRoad,Suitland,MD20746,USA

a r t i c l e i n f o

Articlehistory:

Received22January2018 Accepted9March2018 Availableonlinexxx Keywords:

Metalthreads Membrane Collagen Cyprusgold Proteomics

nanoLC-OrbitrapMS/MS

a b s t r a c t

Beginninginthe13thcentury,membranemetalthreads–madeoutofanimalskins(leather,parchment, andvellum)ormembranousmaterial(e.g.,stomach,intestine)coatedwithmetal–werethemostpopular varietyofdecorativemetalthreadsusedinEuropeantextiles.Thisworkprovidestheproteomicsground- workfortheidentificationofthespeciesandthetypeofmembraneusedinthemanufactureofa14th centurymembranegildedthread.AprotocolforsmallsampleextractionandnanoLC-OrbitrapMS/MS analysiswasfirsttestedonstandardsofpigperitoneumandcowintestinemetal-coatedwithorwithout thepresenceofaneggadhesive.Theproteomesofeachmembranewerecharacterizedandcompared byqualitativeandquantitativebioinformatics;inadditiontothepredominantcollagenproteinsineach membranetype,minortissue-specificproteins(e.g.,smoothmuscleproteinsfromintestinestandards) weredetected.Species-specificcollagenpeptides(i.e.,fromcollagenIandcollagenIII)wereconfidently identifiedtodeterminethespeciesoforigin,regardlessoftheapplicationofmetalandegg-basedadhe- sives.Likewise,thethinlayerofeggadhesivewassuccessfullycharacterizedwiththedetectionofegg white(ovalbumin,ovotransferrin,lysozyme)andeggyolk(vitellogeninI,II,III)proteins.Whenapplied tothethreadfroma14thcenturyItaliantextile,thiscomprehensivemethodologyresultedintheiden- tificationofsevencollagenIandIIIpeptidesspecifictocow,aswellasotherproteinssuggestingthatthe ancientthreadwasmadewithintestineorstomachmembranewithouttheuseofanegg-basedadhesive.

©2018ElsevierMassonSAS.Allrightsreserved.

1. Introduction

Decorativemetalthreadshavebeenextensivelyusedforthe embellishmentoftextilessinceancienttimes.Manyexamplesof metalthreadsexist inartifactsofculturalimportance,andeven earlierreferencestolavishgoldandsilvertextilescanbefoundin ancienttexts,includingadescriptionofanephodcontaininggold intheOldTestamentoftheBible(Exodus39:2–3“Theyhammered outthinsheetsofgoldandcutstrands(...)”).Thepopularityof textileswovenorembroideredwithmetalthreadspersisted,and theiruseisfrequentlyassociatedwithtextilesintendedtoportray wealthorsymbolicimportance[1].Metalthreadsweremostoften madeofgold,silver,ortheiralloys,oftengilt,withthefabrication methoddifferingbyregionandchangingovertime[2].Fivecate- goriesdescribingtheuseofmetalsintextileshavebeendefined:I.

Metalappliedwithadhesivetoalreadywovenfabrics,II.Metalwire

∗ Correspondingauthor.

E-mailaddress:SolazzoC@si.edu(C.Solazzo).

orflattenedstripsuseddirectlyinweaving,III.Metalwireorstrips woundaroundafibercore,IV.Metallicsurfaceappliedtoorganic wrappings(cellulosicorproteinaceous)woundaroundafibercore, andV.Metallicsurfaceappliedtoorganicstrips(cellulosicorpro- teinaceous)withoutafibercore[2–6].

Proteinmetalthreads(CategoriesIVandV)weremadefrom membranoustissues(e.g.,stomachorintestinalwallsofanimals), althoughskinhasalsobeenusedasasubstrate[2,7].Researchatthe endofthe19thcenturysuggestedthatgiltmembranesweremade usingtheintestinesfromslaughteredanimals[8].Cowintestine wassimilarlyusedinthemanufactureofgoldfoil,alsocalledgold- beater’sskin,duringthesametimeperiod.Metalwasappliedtothe membranewithmetalleavesorbymottlegilding,usingeitherthe naturalexudatesoftheorganicmembraneoranadditionaladhe- sive[7].Reportshavedescribed theuseof eggwhite,eggyolk, animalfat,animalandfishglues,gums,andclaysasadhesivesbutto date,noscientificinvestigationhasbeencarriedtosubstantiatethe presenceandnatureoftheadhesives[4,7,9].IncategoryIVthreads, alsoknownasgiltmembranesorCyprusgold,thegildedmembrane wascutintothinstripsandwoundaroundasilkorlinencore[2].No adhesivewasusedbetweenthegildedorganicwrappingandthe https://doi.org/10.1016/j.culher.2018.03.007

1296-2074/©2018ElsevierMassonSAS.Allrightsreserved.

fibercore;instead,thetwistofthemetalmembranethreadaround thecorewassufficienttokeepthematerialinplace.Migrationof adhesiveandwrappingmaterialintothecorefiberhashowever beenobserved[5].IncategoryVthreadsthemetalwasappliedto membranestripsandusedintextileswithoutbeingwound[4].

Membranemetalthreadswerefirstusedinthe11thcenturyby LevantinetradersinCyprus[1].Beginninginthe13thcentury,gilt membranesbecamethemostcommonvarietyofmetalthreads usedinEuropeantextiles,especially thosefromItalyandSpain, aswellasWesternAsiabecausetheywereflexible,lightweight, andinexpensive[7].Substantialfocushasbeengiven toidenti- fyingthemetalsusedinmetalthreads,asstudiesshowthatthe compositionofthemetalandgold/silverratiosaresuggestiveof geographicoriginofthemetalthread[6,7,10].Recently,a range ofelectronmicroscopytechniquesandmicro-Ramanspectroscopy have beenused tocharacterize cross-sectionsof metal threads revealingdetailsoftheproductiontechnologythroughstudyof the3Dtexture[11].Conversely,littleattentionhasbeengivento identifyingthemembranetypeoranimalspeciesusedinorganic metalthreads,althoughtheelucidationofthisinformationwould providevaluableinsightintotheorigin,productiontechnique,and useofmembranemetalthreadsinthemedievalperiod[1].Identi- ficationofthemembraneportionofthesubstrateposesdifficulty, especially whena textile isin a deteriorated state.Using mor- phology onthe microscopic scale, differences between leather, membranousmaterial,parchment,and vellumhave beenmade [1,2];however,theappearanceofthesubstratecanbedrastically alteredfromitsoriginalstatebecauseofdecomposition,embrit- tlement,wear,andtreatmentdamage,makingvisualidentification indefiniteandoftentimesimpossible.Incontrasttomorphological identification,DNAamplificationandmolecularbiologytechniques havebeenused toidentifytheanimal species inorganicmetal threadsfoundintextilesfromthe11–15thcenturies[9].Unfor- tunately,thisstudycouldonlynarrowdowntheidentificationto afewcandidatespecies.Here,wehaveadoptedabottom-uppro- teomicsapproachtoanalyzea14thc.membranethread:theentire membranesampleischaracterizedthroughasingleextractionand digestionofthewholeextract,includingthemembraneproteins and other binding proteinsif present. The complex mixture is separatedbyliquidchromatographybeforeanalysisbyanOrbit- rapVelosmassspectrometerandthedataobtainedaresearched againsta publicdatabase.In theabsence ofhistorical informa- tiononthemembrane’sanimalorigin,anyspeciescanbetargeted throughaproteomicsdatabasesearch.

2. Researchaims

Toresolvetheidentificationissueoftheorganicsubstrateof ametalthreadseveralcenturiesold,aproteomicsapproachwas devised.Proteomicshasbeenusedsuccessfullytoidentifybinders [12]and collagen-basedsubstratessuchasparchment[13],but hasneverbeenusedtocharacterizemembranemetalthreads.The aimofthis projectistoshowthata complexorganicsubstrate, composedofamembranebaseandproteinbinders,canbecom- prehensivelycharacterized,includingthetypeandspeciesofthe membraneandpresenceofproteinaceousadhesives.Totestthe applicabilityofproteomicstoancientsamples,twoseriesofstan- dards(Table1)werepreparedattheUniversityofAppliedArts Vienna(Austria)fromtheabdominalmembraneofpigandfrom intestineofcow,thensubjectedtodifferenttreatments(heatfixa- tion,eggwhite,oreggyolkusedasadhesivesforthemetalcoating).

Pigand cowwerechosenasspeciesfor theircommonality and easyavailability,andasprobablesourceobtainedinthepastfrom butcheredanimals.Theproteomeofeachmembranewasdeter- minedbyextractingproteinsinsamplesoflessthanonemilligram,

Table1

Referencestandards.

Reference Membrane Adhesive Fixation Metalleaftype Piguntreated Peritoneum None None None Piggilt Peritoneum None None Giltsilver Piggiltheat Peritoneum None Heat Giltsilver Piggilteggwhite Peritoneum Eggwhite None Giltsilver Piggilteggyolk Peritoneum Eggyolk None Giltsilver

Cowuntreated Gut None None None

Cowsilver Gut None None Silver

Cowsilverheat Gut None Heat Silver

Cowsilvereggwhite Gut Eggwhite None Silver

Fig.1.14thcenturyItaliantextile(#21714-1)fromtheHistoryMuseuminGraz, Austria,©UniversalmuseumJoanneum,MuseumfürGeschichte,Kulturhistorische Sammlung/InstituteofConservation,UniversityofAppliedArtsVienna/Elisabeth Delvai2017.Thearrowpointstowherethethreadwassampled.Inthebottomright, theimageofthethreadwasacquiredwithHIROXKH-87003Ddigitalmicroscope (Hirox-USA,Inc.,NJ),courtesyofThomasLam(Smithsonian’sMuseumConservation Institute).Software:PowerPoint2013.

thusrevealingtissue-specificproteins,whilepeptidesspecificto themembranespeciesand theeggadhesivewerecharacterized inuntreatedandtreatedsamples.Finally,thedevelopedprotocol wasappliedtoamembranemetalthreadfroma14thcenturyItal- iantextile(Fig.1),thoughttobemadefromananimal’sinternal organandnotskin.

3. Materialsandmethods 3.1. Standards

Thegiltmembranereferencestandardswerepreparedatthe Universität für angewandte Kunst (University of Applied Arts Vienna,Austria)basedonwhatisknownoftraditionalprotocols [14,15].Pigperitoneum,whichisthemembranesackcontaining theabdominalorgans,wasobtainedfromtheVeterinaryUniver- sityofVienna(Vienna,Austria).Cattlegutwasobtainedfromalocal butcher.Thefatwasscrapedoffthemembraneandthemembrane washedandrinsed,thenstretchedandpinned.Giltsilverleaves orsilverleaveswereimmediatelyappliedonthedampmembrane usingeggyolk,eggwhite,ornoadhesive(Fig.S1).Thestandards wereallowedtodryovernight.Selectstandardsweresubjectedto heatfixationwithaheatedspatulaatapproximately80^◦C.Refer- encestandardsarelistedinTable1.

3.2. Historicalsample

Theoriginalsample(#21714-1)comesfromthecollectionofthe HistoryMuseumattheUniversalmuseumJoanneum(Graz,Aus- tria).Thetextile,asilkfabricwithmetalfiberswovenintoit,was producedinItalyinthe14thcentury,depictsdogsorcatsalongside birds.Thesamplewastakenfromalargepieceoffabricthatiswell preserved.Scanningelectronmicroscopyshowedthatthemem- branewascoatedwithgiltsilver,andwoundaroundbastfibers [16],althoughmostofthemetalhaswornawayfromthesampled thread,asseeninFig.1.

3.3. Reagents

Iodoacetamide(IAM) and tris(hydroxymethyl)aminomethane hydrochloride (TRIS HCl) were obtained from Sigma–Aldrich (St. Louis, MO); guanidine hydrochloride (GuHCl), methanol (MeOH), acetonitrile (ACN), and Optima grade formic acid from Fisher Scientific (Fair Lawn, NJ); ammonium bicarbonate (ABC) from VWR International LLC (West Chester, PA); tris(2- carboxyethyl)phosphine(TCEP)fromThermoScientific(Rockford, IL).SequencegrademodifiedtrypsinfromPromega(Madison,WI).

QuantitationofproteinwasdeterminedwiththePierceBCAProtein AssayKitfromThermoScientific(Rockford,IL).

3.4. Samplepreparation

Standardswerecutwitharazorinsquaresoflessthan1mg andweresolubilizedin600␮Lof4MGuHCland50mMTrisHClat pH8,usingfivecyclesof30shomogenizationwithaBeadRuptor Elite(OmniInternational)at7m/s,with5mindwelltimeat−20^◦C betweeneachcycleandfollowingthefinalcycle.Proteinextrac- tionwasallowedtocontinue byshakingthesamplesovernight atroomtemperature.Allsampleswerecentrifugedat10,000rpm for20minandtheproteinconcentrationofthesupernatantwas measuredbythePierceBCAProteinAssay.Thesupernatantwas reduced with500mMTCEP for a final concentrationof 50mM TCEPandpHadjustedto8.0.A250␮Laliquotwasalkylatedwith 40mMIAMandvortexedfor45mininthedark.Thiswasfollowed bybufferexchangewith100mMABCusing3000MWCOAmicon Ultra-0.5mLcentrifugalfilters.Samplesweredigestedwith0.5␮g oftrypsin(trypsin:proteinratio1:100),overnightat37^◦C.Forpep- tidecleanup,thesampleswereloadedonaC18stagetipmadefrom anEmporeSPEExtractionDisk(3M)[17].Thetipswerewashed with0.1%FAsolution.Peptidemixtureswerethenelutedin2× 20␮Lof80:20(v/v)acetonitrile:0.1%FA.Allsampleswerethen drieddowntoafinalvolumeof10␮L.

Solubleproteinswereextractedfrom100␮gofanancient14th centurythreadsample,withthefibercorealreadyremoved,using theabovedescribedprotocol,withtheentire600␮L extraction solutionbeingreduced andalkylated.In additiontothesoluble extracts,theinsolublefractionthatwaspelletedbycentrifugation wasretainedanddilutedin100mMABC,afterwhichbothsoluble andinsolublefractionsunderwentthesamplepreparationstepsas above.

3.5. ProteinanalysisbynanoLC-OrbitrapMS/MS

Thedesaltedsamplesweredilutedand0.5–2␮goftotalprotein wasinjectedandanalyzedbyLC-MS/MS.Thepeptideswerefirst loadedontoanin-housepackedThermoBioBasicC18precolumn (30mm×75␮mi.d.)afterwhichtheywereseparatedonanin- housepackedanalyticalcolumn(210mm×75␮mi.d.)madeofthe samestationaryphase,usingaThermoScientificDionexUltimate 3000UHPLCsystemwiththefollowinggradient:2%B0–8min,55%

B98min,90%B100–103min,2%B104–120min,wherebufferA

is0.1%FAinH₂OandbufferBis0.1%FAinACN.TheUHPLCwas directlycoupledtoaThermoScientificLTQVelosDualPressure LinearIonTrapmassspectrometerwhichanalyzedthepeptidesin positivemodeusingthefollowingparameters:MS160,000resolu- tion,100msacquisitiontime,1×10⁶automaticgaincontrol(AGC), MS215,000resolution,250msacquisitiontime,5×10⁵AGC,top 8,30normalizedcollisionenergy(NCE)higher-energycollisional dissociation(HCD).

3.6. Bioinformaticsanalysis

PEAKS8.0(Bioinformatics SolutionsInc.)wasusedtosearch the RAW data for matches against the UniProt database (www.uniprot.org) of publicly available sequences. The mam- malian database was imported on June 15th, 2017, while a customizeddatabasewascreatedcontainingthecompleteUniprot databasesofSusscrofaandBostaurustowhichwereaddedGallus galluseggproteins(Timestamp14thJuly2017).Searcheswerecar- riedoutusingtrypsinasenzyme,threemissedcleavages,peptide masstolerance(PMS)of10ppm, fragmentmasserrortolerance (MS/MS)of0.02Da,carbamidomethylationasafixedmodification, andthefollowingvariablemodificationsasconfiguredinPEAKS:

deamidation(NQ),hydroxylation/oxidation(RYFPNKD),and oxi- dation(M).PEAKSPTMandSPIDERwerebothenabledtoidentify unspecificPTMs.ResultswerefilteredusinganFDRoflessthan orequalto1%forpeptidespectrummatches,a proteinscoreof

−10lgP≥20,and1uniquepeptide.

QuantitativebioinformaticsanalysiswasperformedusingPro- teomeDiscoverer2.2(ThermoFisher).Searcheswerecarriedoutby MSAmanda2.0,usingthesamesearchparametersusedinPEAKS.

ProteinquantificationwasperformedinProteomeDiscoverer2.2, basedonsummedabundanceofuniqueplusrazorpeptides,with datanormalizedtocollagenIalpha1(COL1A1).

Collagen peptides from pig and cow standards identified as “unique” to a protein group by PEAKS 8.0 software, when searching against the Mammalia protein database, were fur- ther investigated using the Basic Local Alignment Search Tool (BLAST)availableonlinefrombothUniprot(http://www.uniprot.

org/blast searched against UniProtKB protein database) and the National Center for Biotechnology Information NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi searched against NCBI Mammaliadatabase).

4. Resultsanddiscussion 4.1. Membranecharacterization

Theproteomesofperitoneumandintestinewereexaminedto evaluatetheirdifferentialcomposition,andwerefurtheranalyzed usingquantitativebioinformaticstosearchforproteinmarkersthat couldpotentiallyspecifythemembranetype.Theperitoneumisa continuousmembranethatlinestheabdominalwall(parietalperi- toneum)andcoverstheabdominalorgans(visceralperitoneum);

itiscomposedofathinlayerofmesothelium,whichismadeof a monolayerof squamousepithelial cells onabasement mem- brane,supportedbyalayerofconnectivetissue[18].Thebasement membrane is composed of a specialized layer of extracellular matrix protein,withtype IVcollagenbeing themostabundant protein[19–22].Otherpredominantproteinconstituentsinbase- ment membrane include laminin, nidogen (NID), and perlecan (HSPG)[18,23,24].Althoughintestinaltissuealsocontainsabase- mentmembrane,incontrasttoperitoneum,intestinehasalayer ofmuscletissue,specificallysmoothmuscletissue.Smoothmus- cleproteinsincludecalponin-1(CNN1)andmyosin-11(MYH11) [19,25–27].

Analysisoftheperitoneum(TableS1)andintestine(TableS2) standardsshowedthattherewasminimaldifferencebetweenthe collagenproteins, otherextracellularproteins, bloodand serum proteins,andcellularproteincontentofthesetwomembranes(see fullresultsinSI filesS1andS2). Asexpected,multiplesmooth muscleproteinswerefoundintheintestinesamplesonly,which correspondtothemuscularlayerpresentinintestine.Basement membranespecificproteinswere,interestingly,not detectedin peritoneumsamples,butverylowlevelsofNID1,HSPG,andcol- lagenIValpha1(COL4A1)wereobservedintheintestinesamples (TableS2).Theseproteinsaredifficulttoextractbecauseoftheir lowsolubilities[22,28].ThecollagenIVnetworkforinstancecon- tainsdisulfidebonds makingit highly insolubleand extraction withreducingagents,suchasTCEPusedhere,isrequiredtobreak disulfidebondsandcansomewhatimproveextractability[29–31].

However,here the top layerof the membrane was scraped to removefatduringpreparationandmostlikelythemesotheliallayer wasremovedalongsidethefat.Thebasementmembraneisalso muchthinnerthantheconnectiveormusclelayers,solowerabun- danceoftheseproteinsisexpected.

Theabundancesofcollagenandmuscleproteins, normalized relativetocollagenIalpha1(COL1A1),areshowninFig.2usingthe genenamestoallowforinterspeciescomparison.CollagenIIIalpha 1(COL3A1),andtoalesserextentcollagenIalpha2(COL1A2),were foundinhigherabundanceintheintestinethantheperitoneum.

Ofallthemuscleproteinsidentifiedintheintestine,actin,desmin,

Fig.2.Normalizedrelativeabundanceofcollagenandmuscleproteins(indicated bytheirgenecodes)inperitoneum,intestine,anda14thcenturymembranethread.

Correspondingproteinnames:COL1A1,collagenIalpha1;COL1A2,collagenIalpha 2;COL3A1,collagenIIIalpha1;COL4A1,collagenIValpha1;ACTA2,actin,aortic smoothmuscle;CNN1,calponin-1;DES,desmin;MYL6,myosinlightpolypeptide 6;MYL9,myosinregulatorylightpolypeptide9;MYH11,myosin11;TAGLN,trans- gelin;TPM1,tropomyosinalpha1chain;TPM2,tropomyosinbetachain.Software:

Excel2013andPowerPoint2013.

andmyosin-11werefoundwiththehighestabundancesinallcow standardsandassuchrepresentusefulmarkersformembranedif- ferentiation.

4.2. Speciesidentification

The long-term survival of collagen proteins makes them excellent markers for species identificationin ancient artifacts.

Previously,collagenpeptidemarkershavebeenidentifiedbypep- tide massfingerprinting forspecies identificationof bones[32]

andskinsubstrates[13,33].Intheanalysisofarcheologicalanimal glue,oftentheseproteinsaretheonlyproteinsdetectedinsuffi- cientabundancetoallowspeciesidentification[34–36].Inartworks whereanimalgluewascommonlyusedasbinderinpaintingsand frescoes,collagenpeptidesspecifictobovinehavepreviouslybeen identifiedinan18thcenturygiltsample[37].

AsshowninFig.2,COL1A1andCOL1A2werethemostabundant collagenchainsfoundintheintestineandabdominalmembranes, followedbyCOL3A1.Severalmorecollagenproteins(typesIV,V, VI,andXIV)wereidentifiedfromtheLC–MS/MSanalysisofthe untreated(adhesiveandmetalfree)pigandcowstandards(Tables S1andS2),butwithmuchlowercoverages.Havingfurtherestab- lishedtheabundanceofcollagenIandIIIchainsinthemembrane standards,speciesidentificationwasfocusedontheseproteins:the speciesmarkersidentifiedinthestandardmembraneswerealso usedasa proxytodeterminewhetherthecollagenchains,and byconsequencespeciesidentification,wouldbeaffectedbythe manufactureprocesses.

Distinctivemarkersforpigandcowwereidentifiedbysearching againsttheUniprotMammaliaproteindatabaseintheuntreated pig(SIfileS1)andcow(SIfileS2)membranesamples,andbyselect- ingpeptidesuniquelyfoundintheproteinsandspeciesofinterest (COL1A1,COL1A2,andCOL3A1ofSusscrofaandBostaurus).The selectedpeptideswerefurtheridentifiedusingBLAST(Uniprotand NCBI)asbeingspecifictotheSusscrofaspecies(Table2forpigsam- ples)orthebovidclade(Table3forcowsamples)thatincludesthe Bos,Bubalus,andBisongenera.Thereareindeednopeptidespe- cifictotheBostaurusspeciesinthetargetedproteins.Allcollagen peptidesareshowninSIFileS3.

NoCOL1A1specificpeptideswerefoundineitherpigorcow:

unsurprisingly,cowCOL1A1shares97%ofitssequencewiththat of pig, for instance. COL1A2 and COL3A1 have greater degrees of sequence variationbetween animal species.Six COL1A2 and nineCOL3A1peptideswereidentifiedintheuntreatedpigsample (correspondingtorespectivelyfiveandsixdifferentsequenceseg- ments,Table2),andsixCOL1A2andthreeCOL3A1peptideswere identifiedintheuntreatedcowsample(correspondingtorespec- tivelyfourandone differentsequencesegments,Table3),after manualinspectionofthedatainPeaks.

With15and9peptidesidentifiedinthecollagenchainsforpig andcowrespectively,speciesmarkersweresuccessfullyidentified inbothmembrane types.Theadditionofthegiltorsilver layer didnotsignificantlyaffectthedetectionofthemarkerpeptides andnodegradationontheproteinchainswasobservedbecauseof themetal.Metaladductsonpeptidesweresearchedforbutnone werefound;themetalseemstowearoffeasily(aphenomenonalso observedonancienttextiles)andwascompletelyfilteredaftersol- ubilizationoftheproteins.Becauseofthepresenceoftheeggyolk binder,someofthecollagenpeptidesinthepigwerenotdetected (fivemarkerswerenotidentifiedatallinthepigsamplewithegg yolk).Asimilarresultwasobservedfortheeggwhiteinboththe pigandcow,buttoalesserextent.

Overall,therewasnodramaticlossofcollagenpeptidesafter treatment,suggestingthatspeciesidentificationshouldbepossible inancientsamplesusingcollagenpeptidesonly.

Table2

SpecificcollagenpeptidesidentifiedfortheSusscrofa(pig)standards,withpeaksscoresgivenin−10lgP.

Protein Peptidesequence Position Standards

Untreated Gilt Giltheat Gilteggwhite Gilteggyolk GNDGSVGPVGPAGPIGSAGPPGFPGAPGPK 235–264

+2Hpro 53.33 29.74 62.93 34.58

+3Hpro 65.91 45.91 68.18 26.13

GPTGPAGVR 424–432 32.71 31.52 33.19 31.13 33.97

COL1A2:A0A1S7J1Y9PIG GPTGDPGKNGEKGHAGLAGAR 499–519

+1Hpro+1deam 52.20 52.83

+2Hpro+1deam 54.85 44.77 58.02 29.46 48.44

IGPPGPSGISGPPGPPGPAGK 795–815

Susscrofadomesticus +3Hpro 55.28 61.94 57.17 60.78 58.03

+4Hpro 32.49 34.73 33.84 29.58 28.41

+5Hpro 29.41 32.28 21.26

GenBank:BAX02569.1 GYPGNPGPAGAAGAPGPQGAVGPAGK 949–974

+2Hpro 63.31 74.20 65.44 70.27 76.47

+3Hpro 65.50 76.37 67.61 82.88 75.54

PGPAGAAGAPGPQGAVGPAGK 954–974

+1Hpro 57.89 48.23 51.82 59.98

+2Hpro 61.39 67.05 60.37 63.22 61.06

GEVGPAGSPGPSGSPGQR 352–369

+2Hpro 51.76 57.04 54.05

GVAGEPGRDGVPGGPGLR 520–537

+2Hpro 40.98 42.81 39.54 39.81 40.43

+3Hpro 40.82 42.65 43.86 40.22 43.83

GDSGAPGERGPPGAVGPSGPR 676–696

+2Hpro 42.37 51.10 50.62 41.45 47.46

GPPGAVGPSGPR 685–696

+1Hpro 39.00 39.76 37.97 41.53

+2Hpro 25.83 41.45

COL3A1:F1RYI8PIG GAPGEKGEGGPPGIAGQPGGTGPPGPPGPQGVK 829–861

+4Hpro 32.24 20.28 40.95

Susscrofa +5Hpro 17.36 32.64 18.25

+6Hpro 27.57 19.91 31.68

RefSeq:NP001230226.1 GEGGPPGIAGQPGGTGPPGPPGPQGVKGER 835–864

+4Hpro 35.04 33.69

+5Hpro 25.78

GEGGPPGIAGQPGGTGPPGPPGPQGVK 835–861

+3Hpro 62.51 64.77 64.68

+4Hpro 34.83 33.35 39.03 26.88

GSPGPQGPPGAPGPGGISGITGAR 934–957

+2Hpro 65.66 63.52 56.79 68.32

+3Hpro 58.40 65.56 60.57 39.16 56.83

NGDRGETGPAGPAGAPGPAGSR 1062–1083

+1Hpro+1deam 46.17 29.06 55.25 15.67 56.69

Total#peptides 25 22 25 21 16

4.3. Characterizationoftheeggadhesive

Eggwhiteiscomposedoftheproteinsovalbumin(50%),ovo- tranferrin(12–13%)andlysozyme(3%)[38,39],whileeggyolk’s primaryproteincomponentsaretheproteinsvitellogenins(I,II,and III),withvitellogeninIIbeingthemostabundant[40].Theresultsof bothpigandcowstandardscontainingeggwhiteoryolkadhesive (TableS3)identifiedthecorrespondingeggwhiteproteinsandegg yolkproteins.Overall,proteomicanalysisofthestandardsdemon- stratedthatthepresenceofvitellogeninsinasampleindicatesthe useofeggyolk,whereasthepresenceofovalbumin,ovotransfer- rin,andlysozyme,withoutanyvitellogenins,indicatestheuseof eggwhitealone.Ovomucoid(11%oftotaleggwhiteprotein)was alsodetectedwith61%coverageand16peptidesinthecowintes- tineeggwhitestandard,butwithonly3%coverageand1peptidein thepigperitoneumeggwhitestandard.Inaddition,standardswith eggyolkadhesivecontainedsmallquantitiesofeggwhiteproteins (SIFilesS1andS2andTableS3forquantitativedata),mostlikely resultingfromincompleteseparationduringthepreparationofthe adhesive.Thecapabilitiesofquantitativebioinformaticsanalysisto distinguishbetweenyolkandwholeeggbasedonthelevelsofegg whiteproteinshasyettobeexamined.Inadditiontodetecting thesecharacteristiceggproteins,theeggcontainingstandardsdid

notexperienceanyreductionincollagenortissueproteinsignal (Fig.2).

4.4. Analysisofa14thcenturymembranemetalthread

Theproteomicsmethodologywasappliedtotheanalysisofa membranemetalthreadfroma14thcenturyItaliantextileand theresultssearchedagainsttheentireMammaliaproteindatabase (SIFileS4).Severalproteinsspecifictosmoothmuscletissuewere detected in the ancient sample, including MYH11 and desmin (TablesS4andS5).Thenormalizedrelativeabundancesofcolla- genandmuscleproteinsareshowninFig.2.Overall,thepresence ofmultipleproteinsfoundonlyinmuscletissuesuggestthatthe membranemaybefromintestineorstomach,bothofwhichhave amuscularlayerandareknowntohavebeencommonlyusedin medievalgiltmembranes[2,7].

COL1A1,COL1A2,andCOLIIIA1weredetectedwiththehigh- estnumberofpeptidesfromBostaurus(45,42,and38peptides, respectively).In spiteof anaverageloss inprotein coverageof 16% on the main three collagen chains, most bovine markers identifiedinthereferencesampleswereobservedintheancient sample:sevenout ofnine of thesepeptides(e.g.,COL1A2 pep- tideGAP^*GAIGAP^*GPANGDR[Fig.3])weredetectedinthesoluble

Table3

SpecificcollagenpeptidesidentifiedfortheBosTaurus(cow)standards,withpeaksscoresgivenin−10lgP.

Protein Peptidesequence Position Standards 14thc.

threadSol.

14thc.

threadIns.

Untreated Silver Silverheat Silvereggwhite

GAPGAIGAPGPAGANGDR^a,b,c 674–691

+2Hpro+1deam 54.48 58.18 55.47 53.06 44.55 16.37

GAPGAIGAPGPAGANGDRGEAGPAGPAGPAGPR^a,b,c 674–706

+2Hpro+1deam 63.42 62.63 58.48 68.55 33.75 72.00

COL1A2:

CO1A2BOVIN

SGETGASGPPGFVGEK^a,b,c 829–844

+1Hpro 48.91 46.16 45.48

BosTaurus GYPGNAGPVGAAGAPGPQGPVGPVGK^a,b,c 947–972

+2Hpro 78.21 78.88 72.41 80.92 59.44 47.41

RefSeq:

NP776945.1

+2Hpro+1deam 51.76

AGPVGAAGAPGPQGPVGPVGK^a,b,d 952–972

+1Hpro 61.47 43.40 48.78

IGQPGAVGPAGIR^a,b,c,d 1066–1078 39.13

+1deam 36.14 33.59

+1Hpro 42.92 42.94 38.72 41.03 40.43 37.86

+1Hpro+1deam 32.48 34.05

+2Hpro 32.15

COL3A1:

Q08E14BOVIN

GAPGEKGEGGPPGAAGPAGGSGPAGPPGPQGVK^a,b,c,d,e 828–860

+3Hpro 80.88 75.98 45.82 68.08 61.29

+4Hpro 38.30 36.80

BosTaurus GEGGPPGAAGPAGGSGPAGPPGPQGVK^a,b,c,d,e 834–860

+2Hpro 86.72 84.30 59.22 48.45 39.79

RefSeq:

NP001070299.1

GEGGPPGAAGPAGGSGPAGPPGPQGVKGER^a,b,c,d,e 834–863

+3Hpro 79.56 81.63 46.83 64.53 39.64

Total#peptides 10 8 11 8 10 7

Sol.:solublefraction;Ins.:insolublefraction.

aBostaurus.

b Bosmutus.

c Bosindicus.

d Bubalusbubalis.

eBisonbisonbison.

fractionandfiveintheinsolublefraction,asshowninTable3.Pep- tideIGQPGAVGPAGIR,frequentlyusedtoidentifybovinebypeptide massfingerprintingbecauseofitstwocharacteristicpeaksatm/z 1192(nohydroxyproline)andm/z1208(onehydroxyproline),was observedinthehistoricsamplewithdeamidationontheglutamine residueQ.Thismodification,commonlyfoundinancientproteins, wasnotobservedinthestandardsamples.Finally,noeggproteins weredetected,suggestingthatthemembranewaslikelyprepared withouttheuseofanegg-basedadhesive.

5. Conclusions

Theanalysis of a 14thcentury thread demonstrates for the firsttimethecapabilitiesofproteomicsfortheanalysisofancient membranemetalthreads,despitesmallsamplesizesandcomplex matrices.

Inaddition,thestudywassuccessfulin:

1-Findingproteinmarkerstodifferentiatethetypeofmembrane madefromaninternalorgan.Thepossibilitythatanimalskins (leather,parchment)wereusedformetalthreadswillbefur- therexplored,andinparticularthesusceptibilityofeachtype ofmembranetodegradation.

2-Identifyingcollagenspeciesmarkersindependentlyofthepres- enceofbindersandmetal.Whilebovinewasclearlyidentified intheancientsample,thereis,toourknowledge,littleinfor- mationregardingthespeciesusedinmembranethreads.The methodsoffabricationandspeciesusedarelikelytovaryby geographicoriginorforotherreasonssuchasstylistic,qual- ityofmaterials,cost,etc.Knowledgeofthespeciescouldhave

importantimplicationsfordeterminingtheprovenanceofthe threads.

3-Identifyingegg-basedadhesivesalongsidethemembranepro- teins. The complete absence of egg peptides in the ancient sample indicatesthat sucha binder was likely not used. A collagen-basedglue,whoseproteinswouldbeundistinguish- able from the membrane collagens, would require careful separation during sample preparation. Although beyondthe scopeofthisstudy,thepresentworkwillneedtobebroadened totheanalysisofdifferentcombinationsofmembrane+protein binderstobestcharacterizemembranemetalthreadsinancient textiles.

Theproteomicsstudyofmembranemetalthreadsrepresentsa newapplicationtothefieldofculturalheritage.Theexpertiseand carenecessarytomaketextileswiththesethreads,aswellasthe highcostofthematerialsused(e.g.,gold,silk)makethemsome ofthemostvaluableitemsofancienttextiles.Theimplementation ofproteomicstodeterminethechoiceofmembranetype,animal speciesanduseofproteinbinder,inconjunctionwiththeanalysis ofthemetallayer,willresultinamorethoroughunderstandingof theoriginandtechnologyofthesemetalthreads.

Funding

AleksandraK.PopowichwassupportedbyanAlbertaSmith- sonianInternshipProgramScholarshipfromtheGovernmentof Alberta.

Fig.3.SpectrumoftheuniquecowCOL1A2peptide(GAP*GAIGAP*GPANGDR)fromthe14thcenturymembranemetalthread.Hydroxylatedprolineisdenotedby*and deamidatedasparagineisunderlined.Software:PowerPoint2013.

Disclosureofinterest

Theauthorsdeclarethattheyhavenocompetinginterest.

Acknowledgments

WeparticularlythankElisabethDelvaiwhoprovidedsamples duringcompletionofhermasterdegreeattheInstituteofConserva- tion,UniversityofAppliedArtsVienna,aswellastheMuseumfür Geschichte,UniversalmuseumJoanneum.Theauthorsalsothank Dr.BernhardPichler,headofArchaeometryDepartment,University ofAppliedArtsVienna,Dr.RobertKoestler,directoroftheMuseum ConservationInstituteandMaryBallard,textileconservatoratMCI fortheirhelpincoordinatingtheprojectandvaluablediscussions onmetalthreads.FinallyDr.ThomasLamandAsherG.Newsome, MCI,arethankedfortheirtechnicalsupport.

AppendixA. Supplementarydata

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

j.culher.2018.03.007.

References

[1]M.Járó,GoldembroideryandfabricsinEurope:XI–XIVcenturies,GoldBull.23 (1990)40–57.

[2]N.Indictor,TheUseofMetalinHistoricTextiles(PersonalNotes),NewYork, 1987.

[3]N.Indictor,R.J.Koestler, Theidentification andcharacterization ofmetal wrappingsinhistorictextilesusingmicroscopyandenergydispersiveX-ray spectrometry:problemsassociatedwithidentificationandcharacterization, Scan.ElectronMicrosc.2(1986)491–497.

[4]N.Indictor,M.Ballard,Theeffectsofagingontextilesthatcontainmetal:impli- cationsforanalyses,in:InternationalRestorerSeminarVeszprem,Hungary, 1989,pp.67–75.

[5]N.Indictor,R.J.Koestler,M.Wypyski,A.E.Wardwell,Metalthreadsmadeof proteinaceoussubstratesexaminedbyscanningelectronmicroscopy:energy dispersiveX-rayspectrometry,Stud.Conserv.34(1989)171–182.

[6]N.Indictor,R.J.Koestler,C.Blair,A.E.Wardwell,Theevaluation ofmetal wrappingsfromMedievaltextilesusingscanningelectronmicroscopy–energy dispersiveX-rayspectrometry,Text.Hist.19(1988)3–22.

[7]M.Braun-Ronsdorf,GoldandsilverfabricsfromMedievaltomoderntimes, CIBARev.(1961)2–16.

[8]F.Bock,GoldstickereienundWebereieninalterundneuerZeitunddasdazu verwandteGoldgespinnst,G.P.J.Bieling,Nürnberg,1884.

[9]D. DeReyer, A.Y.Jeantet, S.Pilbout,A.Anglo,M.Monnerot,Leslamelles des fils métalliques organiques dans les textiles médiévaux: approche méthodologiquedeleuroriginebiologique,Stud.Conserv.47(2002)122–133.

[10]E.Hoke,I.Petrascheck-Heim,Microprobeanalysisofgildedsilverthreadsfrom Mediaevaltextiles,Stud.Conserv.22(1977)49–62.

[11]T.G.Weiszburg,K.Gherdán,K.Ratter,N.Zajzon,Z.Bend ˝o,G.Radnóczi,Á.Takács, T.Váczi,G.Varga,G.Szakmány,Medievalgildingtechnologyofhistoricalmetal threadsrevealedbyelectronopticalandmicro-ramanspectroscopicstudyof focusedionbeam-milledcrosssections,Anal.Chem.(2017).

[12]C.Tokarski,E.Martin,C.Rolando,C.CileCren-Olivé,Identificationofproteins inrenaissancepaintingsbyproteomics,Anal.Chem.78(2006)1494–1502.

[13]S.Fiddyment,B.Holsinger,C.Ruzzier,A.Devine,A.Binois,U.Albarella,R.

Fischer,E.Nichols,A.Curtis,E.Cheese,M.D.Teasdale,C.Checkley-Scott,S.J.

Milner,K.M. Rudy,E.J. Johnson,J.Vnouˇcek,M.Garrison,S.McGrory,D.G.

Bradley,M.J.Collins,Animaloriginof13th-centuryuterinevellumrevealed usingnoninvasivepeptidefingerprinting,Proc.Natl.Acad.Sci.U.S.A.112(2015) 15066–15071.

[14]M.Jaro,A.Toth,E.Gondar,Determinationofthemanufacturingtechniqueof a10thcenturymetalthread,in:K.Grimstad(Ed.),ICOMCommitteeforCon- servation,9thTriennialMeeting,ICOMCommitteeforConservation,Dresden, GermanDemocraticRepublic,1990,pp.299–301.

[15]M.Járó,A.Tóth,ScientificidentificationofEuropeanmetalthreadmanufactur- ingtechniquesofthe17–19thcenturies,Endeavour15(1991)175–184.

[16]E.Delvai,MedievalfabricfragmentsfromthecollectionoftheUniversalmu- seumJoanneuminGraz,Austria,TechnologicalAnalysisandConservation, InstituteofConservation,UniversityofAppliedArtsVienna,Vienna,2017.

[17]J.Rappsilber,M.Mann,Y.Ishihama,Protocolformicro-purification,enrich- ment,pre-fractionationandstorageofpeptidesforproteomicsusingStageTips, Nat.Protoc.2(2007)1896–1906.

[18]V.S.LeBleu,B.MacDonald,R.Kalluri,Structureandfunctionofbasementmem- branes,Exp.Biol.Med.232(2007)1121–1129.

[19]M.J.Randles,M.J.Humphries,R.Lennon,Proteomicdefinitionsofbasement membranecompositioninhealth anddisease,MatrixBiol.57–58(2017) 12–28.

[20]E.Poschl,U.Schlotzer-Schrehardt,B.Brachvogel,K.Saito,Y.Ninomiya,U.

Mayer,CollagenIVisessentialforbasementmembranestabilitybutdispens- ableforinitiationofitsassemblyduringearlydevelopment,Development131 (2004)1619–1628.

[21]R.Timpl,H.Wiedemann,V.VanDelden,H.Furthmayr,K.Kuhn,Anetwork modelfortheorganizationoftypeIVcollagenmoleculesinbasementmem- branes,Eur.J.Biochem.120(1981)203–211.

[22]K.Kuhn,Basementmembrane(TypeIV) collagen,MatrixBiol.14(1994) 439–445.

[23]J.W.Fox,U.Mayer,R.Nischt,M.Aumailley,D.Reinhardt,H.Wiedemann,K.

Mann,R.Timpl,T.Krieg,J.Engel,M.-L.Chu,Recombinantnidogenconsistsof threeglobulardomainsandmediatesbindingoflaminintocollagentypeIV, EMBOJ.10(1991)3137–3146.

[24]H.Sakai,S.Takeoka,H.Yokohama,Y.Seino,H.Nishide,E.Tsuchida,Purification ofconcentratedhemoglobinusingorganicsolventandheattreatment,Protein Expr.Purif.4(1993)563–569.

[25]S.J.Winder,M.P.Walsh,Calponin:thinfilament-linkedregulationofsmooth musclecontraction,Cell.Signal.5(1993)677–686.

[26]G.J.Babu,D.M.Warshaw,M.Periasamy,Smoothmusclemyosinheavychain isoformsandtheirroleinmusclephysiology,Microsc.Res.Technol.50(2000) 532–540.

[27]S.J.Winder,B.G.Allen, O.Clement-Chomienne,M.P.Walsh,Regulationof smoothmuscleactin–myosininteractionandforcebycalponin,Acta.Physiol.

Scand.164(1998)415–426.

[28]R.Timpl,J.C.Brown,Supramolecularassemblyofbasementmembranes,Bioes- says18(1996)123–132.

[29]R.Timpl,G.R.Martin,P.Bruckner,G.Wick,H.Wiedemann,Natureofthecol- lagenousproteininatumorbasemementmembrane,Eur.J.Biochcm.84(1978) 43–52.

[30]H.K.Kleinman,M.L.McGarvey,L.A.Liotta,P.GehronRobey,K.Tryggvason, G.R.Martin,IsolationandcharacterizationoftypeIVprocollagen,laminin,and heparansulfateproteoglycanfromtheEHSsarcoma,Biochemistry21(1982) 6188–6193.

[31]M.W.Karakashian,P.Dehm,T.S.Gramling,E.C.Leroy,Precursor-sizecompo- nentsarethebasiccollagenoussubunitsofmurinetumorbasementmembrane, Coll.Relat.Res.2(1982)3–17.

[32]M.Buckley,M.Collins,J.Thomas-Oates,J.C.Wilson,Speciesidentificationby analysisofbonecollagenusingmatrix-assistedlaserdesorption/ionisation time-of-flightmassspectrometry,RapidCommun.MassSpectrom.23(2009) 3843–3854.

[33]D.P.Kirby,M.Buckley,E.Promise,S.A.Trauger,T.R.Holdcraft,Identificationof collagen-basedmaterialsinculturalheritage,Analyst138(2013)4849–4858.

[34]N.Bleicher,C.Kelstrup,J.V.Olsen,E.Cappellini,Molecularevidenceofuseof hidegluein4thmillenniumBCEurope,J.Archaeol.Sci.63(2015)65–71.

[35]H.Rao,Y.Yang,I.Abuduresule,W.Li,X.Hu,C.Wang,Proteomicidentificationof adhesiveonabonesculpture-inlaidwoodenartifactfromtheXiaoheCemetery, Xinjiang,China,J.Archaeol.Sci.53(2015)148–155.

[36]C. Solazzo,B.Courel,J. Connan,B.E. vanDongen,H.Barden,K.Penkman, S. Taylor, B. Demarchi, P. Adam, P. Schaeffer, A. Nissenbaum, O. Bar- Yosef,M.Buckley,IdentificationoftheEarliestCollagen-andPlant-Based Coatings fromNeolithicArtefacts, vol.6,NahalHemarCave,Israel,2016, pp.31053.

[37]S. Dallongeville, M.Koperska, N.Garnier,G. Reille-Taillefert,C. Rolando, C. Tokarski, Identification of animalgluespecies inartworks usingpro- teomics:applicationtoa18thcenturygiltsample,Anal.Chem.83(2011) 9431–9437.

[38]H.Sugino,T.Nitoda,L.R.Juneja,HenEggs:BasicandAppliedScience,CRCPress, BocaRaton,FL,USA,1997.

[39]J.Wu,A.Acero-Lopez,Ovotransferrin:structure,bioactivities,andpreparation, FoodRes.Int.46(2012)480–487.

[40]K.Mann,M.Mann,Thechickeneggyolkplasmaandgranuleproteomes,Pro- teomics8(2008)178–191.