Immobilization of flavan-3-ols onto sensor chips to study their interactions with proteins and pectins by SPR

(1)

HAL Id: hal-02635232

https://hal.inrae.fr/hal-02635232

Submitted on 27 May 2020

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

their interactions with proteins and pectins by SPR

Aude Watrelot, Dong Tien Tran, Thierry Buffeteau, Denis Deﬀieux, Carine Le Bourvellec, Stéphane Quideau, Catherine Renard

To cite this version:

Aude Watrelot, Dong Tien Tran, Thierry Buffeteau, Denis Deﬀieux, Carine Le Bourvellec, et al..

Immobilization of flavan-3-ols onto sensor chips to study their interactions with proteins and pectins by SPR. Applied Surface Science, Elsevier, 2016, 371, pp.512-518. �10.1016/j.apsusc.2016.03.002�.

�hal-02635232�

(2)

Version postprint M anus cr it d ’a ut e ur / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt

Immobilization of ﬂavan-3-ols onto sensor chips to study their interactions with proteins and pectins by SPR

Aude A. Watrelot

^a,b,∗

, Dong Tien Tran

^c,d

, Thierry Buffeteau

^c

, Denis Defﬁeux

^c,d

, Carine Le Bourvellec

^a,b

, Stéphane Quideau

^c,d

, Catherine M.G.C. Renard

^a,b

aINRA,UMR408SécuritéetQualitédesProduitsd’OrigineVégétale,DomaineStPaul,SiteAgroparc,84914Avignon,France

bUniversitéd’Avignon,UMR408SécuritéetQualitédesProduitsd’OrigineVégétale,F-84000Avignon,France

cUniversitédeBordeaux,InstitutdesSciencesMoléculaires(UMR-CNRS5255),351coursdelaLibération,33405Talence,France

dInstitutEuropéendeChimieetBiologie(IECB),2rueRobertEscarpit,33607Pessac,France

Keywords:

Polyphenol Polysaccharide Procyanidin Bovineserumalbumin Poly-l-proline

Surfaceplasmonresonance

a b s t r a c t

Interactionsbetweenplantpolyphenolsandbiomacromoleculessuchasproteinsandpectinshavebeen studiedbyseveralmethodsinsolution(e.g.isothermaltitrationcalorimetry,dynamiclightscattering, nuclearmagneticresonanceandspectrophotometry).Herein,theseinteractionswereinvestigatedin realtimebySurfacePlasmonResonance(SPR)analysisafterimmobilizationofflavan-3-olsontoasensor chipsurface.(−)-epicatechin,(+)-catechinandflavan-3-ololigomerswithanaveragedegreeofpoly- merizationof2and8werechemicallymodifiedusingN-(2-(tritylthio)ethyl)propiolamideinorderto introduceaspacerunitontothecatecholicBring.Modifiedflavan-3-olswerethenimmobilizedontoa carboxymethylateddextransurface(CM5).Immobilizationwasvalidatedandfurtherverifiedbyevaluat- ingflavan-3-olinteractionwithbovineserumalbumin(BSA),poly-l-prolineorcommercialpectins.BSA wasfoundtohaveastrongerassociationwithmonomericflavan-3-olsthanoligomers.SPRanalysisof selectedflavan-3-olsimmobilizedontoCM5sensorchipsshowedastrongerassociationforcitruspectins thanapplepectins,regardlessofflavan-3-oldegreeofpolymerization.

1. Introduction

Flavan-3-olsbelongtoalargeanddiversefamilyofplantsec- ondarymetabolites,thepolyphenolicﬂavonoids.Thisclassofplant phenolicshas widespreadinterest to foodprocessors and is of particularinterestfortheirpotentialhealthbeneﬁts.Flavan-3-ol oligomersand polymers,calledproanthocyanidins,formtheso- called“condensedtannins”.Tannins have longbeenrecognized for their capacity to interact with and precipitate proteins or polysaccharides.Thisabilitytointeractwithmacromoleculeshas

Abbreviations: SPR,surfaceplasmonresonance;CM5,carboxymethyldextran;

BSA,bovineserumalbumin;PLP,ploy-l-proline;DP,degreeofpolymerization;ITC, isothermaltitrationcalorimetry;EPI,(−)epicatechin;CAT,(+)-catechin;DP2,flavan- 3-olsdimer;DP8,flavan-3-olsoligomerwithadegreeofpolymerizationof8;STrt, linkerwithtritylgroup;SH,linkerwiththiolgroup;ATR,attenuatedtotalreflection;

RU,resonanceunits.

∗ Corresponding authorat:INRA, UMR408Sécurité etQualité desProduits d’OrigineVégétale,DomaineStPaul,SiteAgroparc,84914AvignonCedex9,France.

E-mailaddresses:aude.watrelot@avignon.inra.fr,watrelotaude@yahoo.fr (A.A.Watrelot).

implicationsin foodchemistry (e.g.redwine ﬁningand astrin- gency) [1]. Insight into the mechanism of interaction between tanninsandmacromoleculeshasapplicationsinfoodandbeverage productionandpotentiallyinhumanhealth.

Flavan-3-olsareabundantinbeverages(e.g.wine, tea,cider) and fruits(e.g.apple, grape, nuts)[2]. Theterm“procyanidins”

specifically designates oligo/polymers of (+)-catechin and (−)- epicatechin.Procyanidinsarecharacterizedbythetypeoflinkages betweenthoseflavan-3-olbuildingblocksandthenumberofsuch constitutiveunitsordegreeofpolymerization(DP)[3].Procyani- dinsfromappleareaninterestingmodelduetotheirstructural homogeneitycombinedwiththeirvariationinsize.Theyarecom- posedof(−)-epicatechinunitsthat arelinkedtogetherthrough theirC4and C8 positions witha DPof up to50 [4], where in someinstances(+)-catechincanbefoundasaterminalunit.Such flavonoidoligo/polymers areabletobindtobiomacromolecules suchasproteins[2]orpolysaccharides,includingnotablypectins [5].

Pectinsareplantpolysaccharides,whicharemainlyfoundin themiddlelamellaandprimarycellwall.Theirbackbonechains are characterized by partially methyl esteriﬁed and acetylated http://dx.doi.org/10.1016/j.apsusc.2016.03.002

(3)

Version postprint M anus cr it d ’a ut e ur / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt

Journal homepage : www.elsevier.com/locate/apsusc

galacturonicacidunits(homogalacturonan),andrhamnoseunitsto whicharelinkedneutralsugarsidechains(rhamnogalacturonans) [6].

Interactions between polyphenols and biomacromolecules, mostlyproteins, have been studiedin solutionby variousana- lyticaltechniquessuchasisothermaltitration calorimetry(ITC), dynamic light scattering or nuclear magnetic resonance spectroscopy,forunderstandingthestrengthandequilibrium/energy parameters of interactions [5,7–12]. Only a few studies have beenperformedbyimmobilizingoneofthetwocompoundson a sensor surface for determining the affinity strength by Sur- facePlasmonResonance(SPR).SPRisanoptical-basedreal-time detection methodwhich allows for theanalysis of biomolecu- lar recognitionat surfacesbetweenan immobilized ligandand ananalyte.Forexample,salivaryproteinshavebeenimmobilized ontoadextranSPRsensorchipforcharacterizingtheinteraction betweenproteinand␤-pentagalloylglucose,acommonprecursor ofgallotannins[13–15].Conversely,aspolyphenolsareknownto interactinanonspecificmannerwithproteins,otherstudieshave chosentoimmobilize polyphenols suchaschemicallymodified vescalineandbiotinylatedvescalaginontomodifieddextranand streptavidin-coatedsensorchips,respectively,forexaminingtheir interactionwithproteinssuchastopoisomeraseII␣andfilamen- tousactin[16].Theliteratureis veryscarceconcerningtheSPR analysisofproteinandpolysaccharideinteractionwithpolyphe- nols. Joergensen et al. [17] have studied associations between pectinsandproteinimmobilized ontoa goldsensorchip,while Seoetal.[18]haveworkedoninteractionsbetweencarbohydrates immobilizedontoa goldsurface andlectins.Moreover,Hayashi et al. [19] have shown higher affinity between ␤-cyclodextrin immobilizedontoacarboxymethylateddextransensorchipand (−)-epigallocatechin-3-O-gallateand (−)-epicatechin-3-O-gallate thannon-galloylated(−)-epigallocatechinand(−)-epicatechin.In ourstudy,theflavan-3-olmonomer/oligomerswereimmobilized ontocarboxymethylated dextransensor chipsin ordertomini- mizenon-specificinteractionsbetweenpolyphenolsandpectins orproteins,forthefirsttime.

Theaimofthisstudywas:first,todefinetheappropriatechemi- calmethodfortheimmobilizationofflavan-3-olmonomers,dimers andoligomersontoa SPRsensorchip;second,toanalyzequal- itativelytheparameters of interactions ofthese flavonoidsand proteinsorpectinsbySPRandthird,tocomparewithinteractions alreadyobservedinsolution.

2. Materialandmethods 2.1. Chemicals

Unlessspeciﬁed,allchemicalsusedforchemicalmodiﬁcation and(+)-catechin,(−)-epicatechin,appleandcitruspectins,bovine serum albumin(BSA) and poly-l-proline (PLP) werepurchased fromSigma-Aldrich(Germany).Propiolicacidwaspurchasedfrom FlukaChemica(Germany).Sodiumhydroxide,magnesiumsulfate, sodiumbicarbonateand hydrochloricacidwerepurchasedfrom VWR(France).tert-butanol waspurchasedfrom AcrosOrganics (Geel,Belgium),CDCl3,DMSO-d6 andacetone-d6 used forNMR analysiswerepurchasedfromEurisotop(France).

2.2. Procyanidinpreparationandcharacterization

Procyanidins of DP2 and DP8were extracted from‘Kermer- rien’and‘Goldendelicious’appleparenchyma,respectively,and were characterized as described by Watrelot et al. (Table 1) [20]. TheDP2 contained 91% of (−)-epicatechin and 9% of (+)- catechin.Thisfractioncontained>90%polyphenolsandincluded

thedimersB2((−)-epicatechin-(4␤→8)-(−)-epicatechin,85%)and B1((−)-epicatechin-(4␤→8)-(+)-catechin,15%).TheDP8fraction wasdeterminedtobe83%polyphenolicandwascomposedof(−)- epicatechin(86%) and(+)-catechin (14%) onlyas terminalunits [20].

2.3. Pectincharacterization

Commerciallyavailable appleand citrus pectins were characterized by their galacturonic acid and neutral sugar content determined by gas chromatography as well as their partition coefﬁcient evaluatedby high performance size exclusion chromatography, as described by Watrelot et al. [5]. The pectin compositionsareshowninTable2.Bothpectinsdifferedinneutral sugarscontent.Citruspectinswerecharacterizedbyahigherrham- nose(2.6%w/w)andgalactose(18.3%w/w)contentthaninapple pectins(1.6%and11.6%w/w,respectively),whileglucosecontent washigherinapplepectins(8.3%w/w)thanincitruspectins(4.4%

w/w).

2.4. Modiﬁcationofﬂavan-3-ols

2-[(Triphenylmethyl)thio]ethanamine(CASnumber:1095-85- 8)waspreparedbyanadaptionoftheprocedurereportedbyLiu etal.[21]Toastirredsolutionof2-mercaptoethylaminehydrochlo- ride(1.1eq.)inanhydrousdichloromethane(4mL)at0^◦Cunder argonatmospherewasaddeddropwisetrifluoroaceticacid(TFA, 1.6mL),followedbychlorotriphenylmethane(1eq.).Thereaction mixturewasstirredfor2hat0^◦C,thenconcentratedanddiluted withCHCl₃(5mL).Thesolutionwasstirredvigorously,mixedwith 3MNaOH(8mL)atroomtemperaturefor1h.Theresultingsuspen- sionwasthenextractedwithCHCl3(3×60mL),andthecombined organiclayers were washed withbrine (3×30mL), dried over MgSO₄,filteredandevaporated todryness toaffordthedesired productasayellowishsolidinquantitativeyieldwithoutanypurifi- cation(spectroscopicdatainsupportinginformation).

N-(2-[(triphenylmethylthio]ethyl)propiolamidewasprepared by an adaptation of the procedure reported by Shiu et al.

[22] To a stirred solution of propiolic acid (4 eq.) and 4- (dimethylamino)pyridine (DMAP, 0.4eq.) in anhydrousCH₂Cl₂ (3mL)at0^◦CunderargonatmospherewasaddeddropwiseN,N’- diisopropylcarbodiimide(DIC,5eq.).Totheresultingsolutionwas addedasolutionof2-[(triphenylmethyl)thio]ethanamine(1eq.) inanhydrousCH₂Cl₂(2mL).Theresultingsuspensionwasallowed towarm upslowly and wasthen stirred at roomtemperature for20hunderargon.Afterremovingthewhiteprecipitatebyﬁl- tration and diluting with CHCl₃ (90mL), the resulting solution waswashedwith0.5MHCl(10mL),aqueoussaturatedNaHCO3

(10mL)andbrine(20mL),driedoverMgSO4,ﬁlteredandevapo- rated.Theresiduewaspuriﬁedbycolumnchromatography,eluting withcyclohexane/EtOAc(8:2;v/v)tofurnishthedesiredproduct asayellowishsolidin43%yield(spectroscopicdatainsupporting information).

Accordingtotheproceduredescribedby Arizaet al.[23],to astirredsolutionof(+)-catechin, (−)-epicatechin,dimer DP2or oligomers DP8 (1 eq.) in a mixture of anhydrous DMF (1mL) andtert-butanol(1mL)wereaddedN-(2-(tritylthio)ethyl)propi- olamide(1.1eq.formonomeranddimerand3eq.foroligomers) andN,N-dimethylaminopyridine(1.5eq.formonomeranddimer and4.5eq.foroligomers).Thereactionmixtureofmonomerswas heatedat40^◦Cunderanargonatmospherefor24hand52hfor dimer;whilereactionmixtureofoligomerswasheatedat60^◦C underanargonatmospherefor5days.Thesolventsandvolatiles werethenevaporatedat50^◦C.Thecrudemixtureofmonomers was puriﬁed by column chromatography, eluting with CH2Cl2

toremove impurities,and then withCH₂Cl₂/acetone (8:2;v/v),

(4)

Version postprint M anus cr it d ’a ut e ur / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt

Table1

CharacterizationofprocyanidinsDP2andDP8(mg/gofdrymatter)(datafromWatrelotetal.[5]).

DPn Procyanidins Dihydrochalcones Hydroxycinnamicacids

Extension Terminal PLZ XPL CQA PCQ

EPI EPI CAT

DP2 2 464 333 79 0 0 8 0

DP8 8 562 64 21 1 0 25 1

SD 0.5 43.4 1.1 0.4 0.1 nd 7.1 nd

DPn:numberaveragedegreeofpolymerization;EPI:(−)-epicatechin;CAT:catechin;PLZ:phloridzin;XPL:phloretinxyloglucoside;CQA:caffeoylquinincacid;PCQ:para- coumaroylquinicacid;SD:pooledstandarddeviation.

Table2

Characterizationofcommercialpectinsfromappleandcitrus:sugarcomposition(mg/g)andpartitioncoefﬁcient(Kav)(datafromWatrelotetal.,[5]).

Rha Fuc Ara Xyl Man Gal Glc GalA MeOH(DM%) Kav

Applepectins 12 – 20 8 1 88 63 564 74(73) 0.02

Citruspectins 20 1 21 4 – 138 33 535 77(79) 0.04

SD 1.4 0.5 0.4 0.2 0.4 1.5 0.8 10.3 1.8(0.02) –

Rha:rhamnose;Fuc:fucose;Ara:arabinose;Xyl:xylose;Gal:galactose;Man:mannose;Glc:glucose;GalA:galacturonicacid;meOH:methanol;DM:degreeofmethylation;

SD:pooledstandarddeviation.

tofurnishadiastereoisomericmixtureoftheexpectedproducts, namedCAT-STrtand EPI-STrtin48%and 53%yieldrespectively (spectroscopicdata insupporting information).The crude mix- turesofdimerandoligomerweretrituratedinCH2Cl2toobtain theexpectedproductsnamedDP2-STrt(55%yield)andDP8-STrt (spectroscopicdatainsupportinginformation).

Detritylationwascarriedoutbyanadaptionoftheprocedure reportedbyBasitetal.[24].Tosolutionsoftheflavan-3-ols-STrt (1eq.)inanhydrousCH₂Cl₂(1mL)wereaddedtrifluoroaceticacid (5eq.)andtriethylsilane(5eq.),andtheresultingsolutionswere stirredfor4hatroomtemperature.Thesolventandvolatileswere thenevaporatedandthecrudeproductsweresuccessivelytritu- ratedinCHCl₃(2mL)andpetroleumether(10×1mL)andfurther driedundervacuumtofurnishtheexpectedcompoundsasyellow- ishplastics,namedCAT-SH(92%yield),EPI-SH(75%yield),B2-SH (78%yield)andDP8-SH(spectroscopicdatainsupportinginformation).ThesecompoundsreactedpositivelywithEllman’sreagent (5,5-dithio-bis-(2-nitrobenzoicacid)),thusconfirmingthepres- enceoffreethiols.AllmodificationstepsweremonitoredbyNMR spectroscopy(300MHz)usingsamplesinacetone-d6orDMSO-d6.

2.5. Surfaceplasmonresonanceexperiments

Allsurface plasmon resonance (SPR) experiments wereper- formedusingaBiacore^TM3000biosensor(Biacore,GEHealthcare, Uppsala,Sweden).Thesensorpossesseda microfluidiccartridge thatprovidesfourseparatechannelsfor differentassayssimul- taneously.Allmeasurementswereperformedat23^◦CinHBS-EP runningbufferfromGEHealthcare(0.01MHEPESpH7.4,0.15M NaCl,3mMEDTA,0.005%surfactantP20).1nMflavan-3-ol-SHwas dissolvedinHBS-EPbufferwith5%v/vaceticacidandthenfiltered (0.45␮m)beforeinjection.

2.5.1. ImmobilizationontoCM5surface

Carboxymethyldextransensorchips(CMD200m,Xantec,Dues- seldorf,Germany)wereusedforimmobilizationofﬂavan-3-ol-SH accordingtothethiolcouplingprotocolfromGEHealthcare.The baselinewasstabilizedfor300swithHBS-EPbufferat5␮L/min.

Then10␮LofEDC(0.2M)/NHS(0.05M)wereinjected,followed by20␮Lof80mM2-(2-pyridinyldithio)-ethanamine(PDEA)pre- paredina0.1MboratebufferpH8.5.Twotimesof10␮Lof1nM ﬂavan-3-ol-SHwasinjected.Then,20␮Lofamixtureof50mMcys- teinand1Msodiumchloridepreparedina0.1Msodiumacetate

bufferpH4.3wasinjected.Thereferencetrackwaspreparedunder thesameconditionsasabovebutwithoutﬂavan-3-ol-SH.

2.5.2. InteractionanalysisbySPR

Apple andcitrus pectins,BSA andPLP weresolubilizedat3, 30and300nMinHBS-EPbufferwith5%v/vaceticacid.Interac- tionanalyseswerecarriedoutataﬂowrateof20␮L/min.After interactionanalysiswithpectins orproteins, sensorchips were regeneratedbyinjectionofSDS0.1%w/vsolution.

2.6. ATRexperiments

TheATRspectraofCAT,CAT-SH,DP8,andDP8-SHcompounds wererecordedwithaThermoNicoletNexus670FTIRspectrome- terequippedwithaliquidnitrogencoolednarrow-bandmercury cadmiumtelluride(MCT)detectorusingaSilver-Gate(germanium crystal)ATRaccessory(Specac).Theelectricﬁeldoftheinfrared beamwaspolarizedeitherperpendicular(s-polarized)orparallel (p-polarized)totheplaneofincidencewithaBaF2wiregridpolar- izer(Specac).Eachspectrumwasobtainedafterevaporationofa drop(20␮L)ofasolution(2mgin100␮LofEtOH),ataresolution of4cm⁻¹,byco-adding500scans.

2.6.1. DeterminationoftheopticalconstantsofCAT,CAT-SH, DP8,andDP8-SHcompounds

The optical constants (refractive index n(¯

v

) and extinction coefﬁcient k(¯

v

) of CAT, CAT-SH, DP8, and DP8-SH compounds havebeendeterminedfrompolarizedattenuatedtotalreﬂectance (ATR)spectra,usingtheinterdependenceofn(¯

v

)andk(¯

v

)bythe Kramers-Kronig relations. Dignam et al. have shown how the Kramers-KronigrelationscanbeappliedtopolarizedATRspectra [25]. The in-plane optical constants (nxy and kxy) were calcu- latedfromthes-polarizedATRspectrum,whereastheout-of-plane opticalconstants(nzandkz)wereobtainedfromthep-polarized ATRspectrumandthebeforedeterminednxyandkxy.Then,the isotropicopticalconstantsofCAT,CAT-SH,DP8,andDP8-SHcom- poundshavebeencalculatedfromthein-planeandout-of-plane opticalconstantsusingtherelations:n_iso=(2n_xy+nz)/3andk_iso= (2k_xy+k_z)/3.

(5)

Version postprint M anus cr it d ’a ut e ur / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt

Journal homepage : www.elsevier.com/locate/apsusc

3. Results

3.1. Flavan-3-olderivatization

Inordertoimmobilizeflavan-3-olmonomers/oligomersonto sensor chips,a linkerbearinga protectedthiolgroup atoneof itsextremitieswassynthesized.Theotherextremityofthelinker wasboundtotheflavan-3-olcatecholicB-ringthroughaVilarrasa reaction[23,26].Inordertooptimizethemobilityofflavan-3-ols immobilizedontosurfaces,flavan-3-olcatecholsweremodifiedby usingonlyoneequivalentofthelinkerperequivalentofflavan-3-ol.

Modiﬁedﬂavan-3-olswerethendetritylatedtoreleasetheirthiol group(s)(Scheme1).Thesereactionsperformedonmonomersand dimersweremonitoredby¹Hand¹³CNMRanalyses,whereasonly

1HNMRanalysiswasusedtofollowtheappearanceanddisappear- anceoftritylgroups(between7.5and7ppm)onDP8oligomers.

Mid-infraredextinctioncoefﬁcientk(¯

v

)spectraofamonomer(CAT) andflavan-3-ololigomers(DP8)werealsodeterminedfromATR spectrabeforeandafterderivatizationinordertoconfirmthepres- enceofthiolatelinkers(Fig.1).Thecovalentlinkageofflavan-3-ols tothelinkerviaanamidebondwasconfirmedbythepresenceof amideI(␯_{C O})andamideII(␦NH+_{C N})modesofthelinkerat 1655and1500cm⁻¹,respectivelywereobservedforbothderiva- tizedmolecules(CAT-SHandDP8-SH)whereasthesebandswere not present in the initialCAT and DP8 spectra. The 1630 and 1470cm⁻¹ bands,assignedtotheC Cstretchingmodesofthe Aringofflavan-3-ols,werenotobservableontheDP8spectrum, likelybecausetheintensityofthesetwomodesoccursfromthe unlinkedrings.

3.2. CM5immobilizationandinteractionanalysis

Toobtainagoodflexibilityandsubsequentbindingsiteavail- abilityofimmobilizedflavanols,flavan-3-ols-SHwereimmobilized ontocarboxymethyldextran(CM5) sensorchipsusingaroutine

couplingmethodfromGEHealthcareLifeSciences (Ligandthiol coupling,LaboratoryguidelinesBR-2001-21AB).Forthismethod, reactivedisulﬁdegroupswereintroducedintothedextranmatrix then,thethiolcouplingoccurredthroughathiol-disulﬁdeexchange withthethiolgroupsontheligand.EPI-SH,DP2-SHandDP8-SH wereimmobilizedontoCM5sensorchipwiththefollowingres- onanceunits(312RU,417RUand305RUrespectively),thenthe excessreactivegroupsweredeactivatedwitha50mMcysteine- containing0.1Msodiumacetatesolution.

Themodifiedsensorchipswerefirsttestedwithaprotein(BSA) andapolypeptide(PLP)wellknowntointeractinsolutionwith theseflavan-3-ols[9,11].InteractionsbetweenBSAandflavan-3- olsdependedonBSAconcentrations.At3nMBSA,theassociation step washigher with(−)-epicatechin(85 RU) thanwithdimer (34RU)oroligomerDP8(17RU)(Fig.2).ForBSAconcentration higherthan30nM,theassociationstepshowedtwostagesthatis anincreaseofassociationafterthebeginningofinjectionofBSA followedbyadecreaseofSPRsignalafter1minbeforetheendof injection(datanotshown).Withpoly-l-prolinenointeractionwas detectedat3nMofpeptide,whilesimilarlevelswereobtainedfor allflavan-3-olsat30nMofpeptide(Table3).

Onlyweakresonanceunitswereobtainedforapplepectinswith either EPIor DP8 (around15 RU),a slightly higher level being reached for DP2(22 RU) whatever theconcentrations of apple pectins(Fig.3andTable3).Higherassociationwasfoundwithcit- ruspectinupto120RUforallﬂavan-3-olsatthehighestpectin concentration(Table3).

4. Discussion

Thebindingaffinitydifferencesbetweenflavan-3-olderivatives immobilizedontoCM5surfaceandbiomacromoleculesmightbe explainedeitherbythebiomacromoleculesorflavan-3-olstruc- tureandconformation.Ontheonehand,BSAiscomposedofthree domains and its tertiary structure (likean equilateral triangle)

1. Ph3CCl, 30%TFA/CH2Cl2, 0 ^oC, 2 h 2. NaOH 3M, r.t., 1 h

DIC, 4-DMAP/CH2Cl2

H2N STrt O

OH

O NH

STrt

Ar, 0 ^oC to r.t., 20 h

Ar, 40 ^oC, 24 h

4-DMAP, DMF/t-BuOH (1:1) Ar, r.t., 4 h

TFA, Et3SiH/CH2Cl2

(43%) (quantitative)

R1 = OH, R2 = H: 48%

R1 = H, R2 = OH: 57%

O

OH HO

OH OH

R2

H2N SH

O

OH HO

OH O

O O

HN STrt O

OH HO

R2

O

O O

HN STrt

O OH

OH

OH HO

OH

O

OH HO

OH O

O O

HN SH

O OH

OH

OH HO

OH R1

R1 = OH, R2 = H: 92%

R1 = H, R2 = OH: 75%

R1

O

OH HO

R2

O

O O

HN SH

R1

Ar, 40 ^oC, 52 h 4-DMAP, DMF/t-BuOH (1:1)

O

OH HO

OH O

OH OH

OH HO

OH OH OH

(55%) (78%)

TFA, Et3SiH/CH2Cl2

Ar, r.t., 4 h

Ar, 60 ^oC, 5 d 4-DMAP, DMF/t-BuOH (1:1)

DP8 DP8-STrt TFA, Et3SiH/CH2Cl2

Ar, r.t., 4 h DP8-SH DP2

DP2-SH

CAT-STrt or EPI-STrt CAT-SH or EPI-SH

CAT or EPI

DP2-STrt

Scheme1. ReactionschemeofchemicalsynthesisofN-(2-[(Triphenylmethylthio]ethyl)propiolamidelinkerandﬂavan-3-ols.

(6)

Version postprint M anus cr it d ’a ut e ur / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt

Fig.1.Extinctioncoefficientspectraof(+)-catechin(CAT),(+)-catechinmodified bylinkerwiththiolgroup(CAT-SH),flavan-3-ololigomer(DP8)andflavan-3-ol oligomermodifiedbylinkerwiththiolgroup(DP8-SH).Assignmentofthemajor bandsinthe1700–1300cm⁻¹spectralrange.

is similartoHumanSerum Albumin. DomainsIIand IIIforma positivelychargedhydrophobiccavitywhichallowsforassocia- tionswithligands[27].Ontheotherhand,proline-richproteins (PRPs),morepreciselyPLP,formanopenextendedstructurewhich presentsamaximumsurfaceareaforinteractionperresidue[28].

Therefore,weexpectedthatthesmallerpolypeptidewouldbind more,asthesterichindranceshouldbelimited(smallermolecule).

Theoppositewasobserved,perhapsduetolimitedﬂexibility(i.e.

nopossibilitytoadjustitsconformationtotheboundflavan-3-ols) [29]orduetoinvolvementoftheB-ringoftheflavan-3-olshere modifiedand/orboundtothesensorchip,intheinteraction.

Differentresponselevelswereobservedbetweenﬂavan-3-ols andthetwopectins.Applepectinsshowedhigherresonanceunits withthedimerDP2whilecitruspectinsinteractedwithmoreinten- sitybutindifferentlytothemonomer,dimerandoligomer.Apple and citrus pectins differ by theirneutral sugar composition, in particularbyhigherrhamnoseandgalactoseincitruspectins,i.e.

presumablymoretypeIrhamnogalacturonans.Thehighrhamnose contentinapplepectinscouldinduceanincreaseofthemolecule ﬂexibilityandleadtoachangeofconformation[30].Apartfrom rhamnose,neutralsugarstendtoadoptvariousconformationsin solution.Watrelotetal.[31]showedthat␤–1→4galactanchains bindmorestronglytoprocyanidinsthanarabinanchains,andit wasdemonstratedthatpectinswithshortneutralsugarsidechains

Table3

Resonance units corresponding to interactions between ﬂavan-3-ols (−)- epicatechin(EPI),dimer(DP2)andoligomer(DP8)andanalytes(appleandcitrus pectins,poly-L-prolineandbovineserumalbumin)at3,30and300nMinHBS-EP bufferat25^◦C.

Analytes/Ligands EPI(312RU) DP2(417RU) DP8(305RU) Applepectins

3nM 16 22 14

30nM 16 21 14

300nM 11 15 10

Citruspectins

3nM 61 60 61

30nM 118 119 120

300nM 112 121 119

Bovineserumalbumin

3nM 85 34 17

Poly-l-proline

3nM – – –

30nM 58 64 58

300nM 61 58 65

associatewithﬂavan-3-olpolymers.Thisinteractionwouldbedue toacooperativeinteractionbetweenpectinsandprocyanidins.

Polyphenolswereimmobilized bythecatecholgroup,which istheB-ringofﬂavan-3-ols.Consequently,forthemonomer(i.e.

(−)-epicatechin),theB-ringwasnotavailabletobindtobiomacro- molecules.Therefore,sensorchipspreparedherecouldonlydetect interactions due eithertohydrophobicinteractions betweenA- orpyranC-ringandmethylgroupsofpectinsoraminoacidsof proteins, or tohydrogenbondsbetweenhydroxyl groups ofA- and/orC-ringandhydroxylgroupsofpectinsorproteins,orboth.

The dimer DP2 wasimmobilized by one B-ring,which left the otherfreetointeract.Therefore,thehigherassociationofapple pectinswithDP2couldbeduetobothparameters(conformation ofapplepectinsandahigheravailabilityofbindingsitesofdimer).

Regardingﬂavan-3-ololigomers,thenumberofcatecholgroups immobilizedwasunknownandtheconformationisfullydifferent fromthatofmonomeranddimer,notably withincreasedrigid- ity[32],whileorientationwiththerespecttothesurfaceremains unknown.Associations,bothwithpectins orproteins,werenot correlatedtothenumberofhydroxylgroupsofpolyphenolsimmo- bilizedcontrarytotheworkofJean-Gillesetal.[33]butratherin theconformationadoptedbypolyphenols.

Associations between the same polyphenols and macro- molecules (pectins and proteins) were recently measured in solution byIsothermal TitrationCalorimetry (ITC) [5,20].ITC is baseduponheatexchangeandmeasurestheenergyreleasedupon (ornecessaryfor)ligand-macromoleculeinteractions.Weakener- giesofinteractionswerefoundbetweenpectinsandﬂavan-3-ol monomers and dimers, so that method was not an appropri- atemethodformeasuringinteractionsparametersbetweenthose twomolecules[5].Incontrast,SPRmeasurementsshowedthat, dependingonpectinconcentrations,bothpectinstendtointeract morewiththedimerthanwiththemonomer.Therefore,provided orgeneratedheatbyinteractionsbetweenpectinsandﬂavan-3-ols monomerordimermightbetoolowfordetection,whiletheinjec- tionofpectinsonasurfaceof(−)-epicatechinordimerleadsto formationoflayersofpectinswhichinducedachangeofrefracted lightandalsoresonanceunits.

Associations between model proteins or polypeptides and flavan-3-olswerealreadymeasuredbyITC[7,10–12,20].ITCevi- dencedhigheraffinityofflavan-3-olforBSAthanforpoly-l-proline asfoundherebySPR.Thisaffinityincreasedwiththeincreaseofthe degreeofpolymerizationofflavan-3-ols,aspreviouslyobserved [34].However,inthecaseofimmobilizedflavan-3-ols(usingSPR), affinitywithBSAdecreasedwiththeincreaseoftheDP.Moreover, degreeofpolymerizationdidnotinfluenceRUlevelsobservedwith poly-l-proline.Asmentionedforanalysiswithpectins,impactof

(7)

Version postprint M anus cr it d ’a ut e ur / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt

Journal homepage : www.elsevier.com/locate/apsusc

Fig.2.Sensorgramofinteractiondoublecorrectedbetweenbovineserumalbumin(BSA)at3nMand(−)-epicatechin(89RU,redline),dimerDP2(72RU,greenline)and oligomerDP8(81RU,blueline),inHBS-EPbufferat25^◦C.(Forinterpretationofthereferencestocolourinthisﬁgurelegend,thereaderisreferredtothewebversionofthis article.)

Fig.3.Sensorgramofinteractiondoublecorrectedbetweenapplepectinsat3,30and300nMand(−)-epicatechin(redline),dimerDP2(greenline)andoligomerDP8(blue line)inHBS-EPbufferat25^◦C.(Forinterpretationofthereferencestocolourinthisﬁgurelegend,thereaderisreferredtothewebversionofthisarticle.)

immobilizationonﬂavan-3-olsconformationprobablymodiﬁed theinteractionsatthemolecularlevel.

5. Conclusions

Inthiswork,flavan-3-olmonomer,dimerandoligomerwere chemicallymodified through theirB-ring andfor thefirst time grafted onto CM5 surface with success. The immobilization of flavan-3-olderivativesontoaCM5sensorchips,confirmedusing SPR,allowedcomparisonofinteractionsofthesameflavan-3-ol- bearingsensorchipswithvariedmacromolecules,whichbrought newinformationoninteractioncharacteristics.Forthefirsttime moleculesofproteicandpolysaccharidicnatureweredirectlycom- paredinthesamesystem:though,asexpected,higheraffinities wereobtainedwithBSA,citruspectininparticularboundremark- ablytothesensorchips.Therefore(i)interactionsofflavan-3-ols withpolysaccharidesshouldnotbeneglected,particularlyinplant foodsystems,and(ii)furtherworkontheimpactoffinestructure

ofpolysaccharidesontheinteractionswithtanninsisstillneeded.

SPR analysisshould beusedin addition toITC tocompare the behaviorofmacromoleculeswhenoneisimmobilizedonasolid surfaceorbothareinsolution.Thismethodisusefulforabetter understandingoftheeffectofthemacromoleculeconformationon associations.Becauseinthis studytheflavan-3-olswereimmo- bilizedontoasurfacethroughB-ringsand affinitieswerelower thanobtainedinsolution,theresultssuggestthattheflavan-3- olB-ringisinvolvedinmacromolecularinteractions.Thismethod gavenewinformationabouttheinteractionmechanismsoccurring betweenmacromolecules(condensedtannin,protein,polysaccharide),whichareinvolvedinastringencymouthfeelperceptionand colloidsformation.

Acknowledgments

TheauthorsthankDr.CarmeloDiPrimofortheuseofhissur- face plasmon resonance equipment, and Dr. James A. Kennedy

(8)

Version postprint M anus cr it d ’a ut e ur / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt M anus cr it d ’a ut eu r / A ut hor m anus cr ipt

formanuscriptreview.Theresearchleadingtotheseresultshas receivedfundingfromtheEuropeanCommunity’sSeventhFrame- workProgramme(FP7/2007-2013)underthegrantagreementNo.

FP7-222654-DREAM.

AppendixA. Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound,in theonlineversion, athttp://dx.doi.org/10.1016/j.apsusc.2016.03.

002.

References

[1]J.M.McRae,J.A.Kennedy,Wineandgrapetannininteractionswithsalivary proteinsandtheirimpactonastringency:areviewofcurrentresearch, Molecules16(3)(2011)2348–2364.

[2]S.Quideau,D.Defﬁeux,C.Douat-Casassus,L.Pouységu,Plantpolyphenols:

chemicalproperties,biologicalactivities,andsynthesis,Angew.Chem.Int.Ed.

50(3)(2011)586–621.

[3]R.W.Hemingway,J.J.Karchesy,ChemistryandSigniﬁcanceofCondensed Tannins,PlenumPress,NewYork,1989,pp.553p.

[4]P.Sanoner,S.Guyot,N.Marnet,D.Molle,J.-F.Drilleau,Polyphenolproﬁlesof frenchciderapplevarieties(MalusDomesticaSp.),J.Agric.FoodChem.47(12) (1999)4847–4853.

[5]A.A.Watrelot,C.LeBourvellec,A.Imberty,C.M.Renard,Interactionsbetween pecticcompoundsandprocyanidinsareinﬂuencedbymethylationdegree andchainlength,Biomacromolecules14(2013)709–718.

[6]B.L.Ridley,M.A.O’Neill,D.Mohnen,Pectins:structure,biosynthesis,and oligogalacturonide-relatedsignaling,Phytochemistry57(6)(2001)929–967.

[7]C.Poncet-Legrand,A.Edelmann,J.-L.Putaux,D.Cartalade,P.Sarni-Manchado, A.Vernhet,Poly(l-Proline)interactionswithFlavan-3-Olsunits:inﬂuenceof themolecularstructureandthepolyphenol/proteinratio,FoodHydrocoll.20 (5)(2006)687–697.

[8]C.Poncet-Legrand,C.Gautier,V.Cheynier,A.Imberty,Interactionsbetween Flavan-3-OlsandPoly(l-Proline)studiedbyisothermaltitrationcalorimetry:

effectofthetanninstructure,J.Agric.FoodChem.55(22)(2007)9235–9240.

[9]C.Pascal,C.Poncet-Legrand,A.Imberty,C.Gautier,P.Sarni-Manchado,V.

Cheynier,A.Vernhet,Interactionsbetweenanonglycosylatedhuman proline-richproteinandﬂavan-3-olsareaffectedbyproteinconcentration andpolyphenol/proteinratio,J.Agric.FoodChem.55(12)(2007)4895–4901.

[10]R.A.Frazier,A.Papadopoulou,I.Mueller-Harvey,D.Kissoon,R.J.Green, Probingprotein-tannininteractionsbyisothermaltitrationmicrocalorimetry, J.Agric.FoodChem.51(18)(2003)5189–5195.

[11]R.A.Frazier,A.Papadopoulou,R.J.Green,Isothermaltitrationcalorimetry studyofepicatechinbindingtoserumalbumin,J.Pharm.Biomed.Anal.41(5) (2006)1602–1605.

[12]R.A.Frazier,E.R.Deaville,R.J.Green,E.Stringano,I.Willoughby,J.Plant,I.

Mueller-Harvey,Interactionsofteatanninsandcondensedtanninswith proteins,J.Pharm.Biomed.Anal.51(2)(2010)490–495.

[13]G.Gyémánt,Á.Zajácz,B.Bécsi,C.Ragunath,N.Ramasubbu,F.Erd"odi,G.Batta, L.Kandra,Evidenceforpentagalloylglucosebindingtohumansalivary

␣-amylasethrougharomaticaminoacidresidues,Biochim.Biophys.Acta BBA—ProteinsProteom.1794(2)(2009)291–296.

[14]J.R.L.Guerreiro,V.DeFreitas,D.S.Sutherland,M.G.F.Sales,SPRbasedstudies forpentagalloylglucosebindingto␣-amylase,ProcediaEng.47(2012) 498–501.

[15]A.Kiss,B.Bécsi,B.Kolozsvári,I.Komáromi,K.E.Kövér,F.Erd"odi, Epigallocatechin-3-gallateandpenta-O-galloyl-B-d-glucoseinhibitprotein phosphatase-1,FEBSJ.280(2)(2013)612–626.

[16]S.Quideau,C.Douat-Casassus,D.M.DelannoyLópez,C.DiPrimo,S.Chassaing, R.Jacquet,F.Saltel,E.Genot,Bindingofﬁlamentousactinandwindinginto ﬁbrillaraggregatesbythepolyphenolicc-glucosidicellagitanninvescalagin, Angew.Chem.Int.Ed.50(22)(2011)5099–5104.

[17]L.Joergensen,B.Klösgen,A.C.Simonsen,J.Borch,E.Hagesaether,New insightsintothemucoadhesionofpectinsbyAFMroughnessparametersin combinationwithSPR,Int.J.Pharm.411(1–2)(2011)162–168.

[18]J.H.Seo,K.Adachi,B.K.Lee,D.G.Kang,Y.K.Kim,K.R.Kim,H.Y.Lee,T.Kawai, H.J.Cha,Facileandrapiddirectgoldsurfaceimmobilizationwithcontrolled orientationforcarbohydrates,BioconjugateChem.18(6)(2007)2197–2201.

[19]N.Hayashi,R.Chen,M.Hiraoka,T.Ujihara,H.Ikezaki,B-cyclodextrin/surface plasmonresonancedetectionsystemforsensingbitter-astringenttaste intensityofgreenteacatechins,J.Agric.FoodChem.58(14)(2010) 8351–8356.

[20]A.A.Watrelot,C.M.G.C.Renard,C.LeBourvellec,Comparisonof

microcalorimetryandhazeformationtoquantifytheassociationofb-type procyanidinstopoly-l-prolineandbovineserumalbumin,LWT—FoodSci.

Technol.63(1)(2015)376–382.

[21]R.Liu,R.Liew,J.Zhou,B.Xing,Asimpleandspeciﬁcassayforreal-time colorimetricvisualizationof␤-lactamaseactivitybyusinggoldnanoparticles, Angew.Chem.Int.Ed.46(46)(2007)8799–8803.

[22]H.-Y.Shiu,T.-C.Chan,C.-M.Ho,Y.Liu,M.-K.Wong,C.-M.Che, Electron-deﬁcientalkynesascleavablereagentsforthemodiﬁcationof cysteine-containingpeptidesinaqueousmedium,Chem.Weinh.Bergstr.Ger.

15(15)(2009)3839–3850.

[23]X.Ariza,O.Pineda,J.Vilarrasa,G.W.ShippsJr.,Y.Ma,X.Dai,Bocdeneand mocdenederivativesofcatecholsandcatecholamines,Org.Lett.3(9)(2001) 1399–1401.

[24]H.Basit,A.VanderHeyden,C.Gondran,B.Nysten,P.Dumy,P.Labbé, Tetheredbilayerlipidmembranesonmixedself-assembledmonolayersofa novelanchoringthiol:impactoftheanchoringthioldensityonbilayer formation,LangmuirACSJ.Surf.Colloids27(23)(2011)14317–14328.

[25]M.J.Dignam,Fouriertransformpolarizationspectroscopy,Appl.Spectrosc.

Rev.24(1–2)(1988)99–135.

[26]C.Chalumeau,D.Defﬁeux,S.Chaignepain,S.Quideau,Developmentofan afﬁnity-basedproteomicstrategyfortheelucidationofproanthocyanidin biosynthesis,ChemBioChem12(8)(2011)1193–1197.

[27]S.Sugio,A.Kashima,S.Mochizuki,M.Noda,K.Kobayashi,Crystalstructureof humanserumalbuminat2.5Åresolution,ProteinEng.12(6)(1999)439–446.

[28]M.P.Williamson,Thestructureandfunctionofproline-richregionsin proteins,Biochem.J.297(2)(1994)249–260.

[29]T.Richard,X.Vitrac,J.M.Merillon,J.P.Monti,Roleofpeptideprimary sequenceinpolyphenol–proteinrecognition:anexamplewithneurotensin, Biochim.Biophys.Acta1726(3)(2005)238–243.

[30]C.M.G.C.Renard,M.-J.Crépeau,J.-F.Thibault,Structureoftherepeatingunits intherhamnogalacturonicbackboneofapple:beetandcitruspectins, Carbohydr.Res.275(1)(1995)155–165.

[31]A.A.Watrelot,C.LeBourvellec,A.Imberty,C.M.Renard,Neutralsugarside chainsofpectinslimitinteractionswithprocyanidins,Carbohydr.Polym.99 (2014)527–536.

[32]D.Zanchi,P.V.Konarev,C.Tribet,A.Baron,D.I.Svergun,S.Guyot,Rigidity, conformation,andsolvationofnativeandoxidizedtanninmacromoleculesin water-ethanolsolution,J.Chem.Phys.130(24)(2009).

[33]D.Jean-Gilles,L.Li,V.G.Vaidyanathan,R.King,B.Cho,D.R.Worthen, ChichesterIII,C.O.:seeram,N.P.inhibitoryeffectsofpolyphenolpunicalagin ontype-iicollagendegradationinvitroandinﬂammationinvivo,Chem.Biol.

Interact.205(2)(2013)90–99.

[34]R.L.Kilmister,P.Faulkner,M.O.Downey,S.J.Darby,R.J.Falconer,The complexityofcondensedtanninbindingtobovineserumalbumin—an isothermaltitrationcalorimetrystudy,FoodChem.190(2016)173–178.