Label-free method for anti-glucopeptide antibody detection in Multiple Sclerosis

Label-free

method

for

anti-glucopeptide

antibody

detection

in

Multiple

Sclerosis

Feliciana

Real-Fernández

a,b

,

Giada

Rossi

a,b

,

Francesco

Lolli

a,c

,

Anna

Maria

Papini

a,d

,

Paolo

Rovero

a,b,

*

a_{InterdepartmentalLaboratoryofPeptideandProteinChemistryandBiology–PeptLab,Italy}1

b

DepartmentofNeurosciences,Psychology,DrugResearchandChildHealth,SectionofPharmaceutical SciencesandNutraceutics,UniversityofFlorence,ViaUgoSchiff6,I-50019SestoFiorentino,Italy

c

DepartmentofBiomedical,ExperimentalandClinicalSciences,UniversityofFlorence,VialeMorgagni50, I-50134Firenze,Italy

d

DepartmentofChemistry“UgoSchiff”,ViadellaLastruccia13,I-50019SestoFiorentino,Italy GRAPHICAL ABSTRACT

ABSTRACT

Surfaceplasmonresonancetechniqueisparticularlyinterestinginimmunologybecauseithasthepotentialto visualizelabel-freeantigen-antibodyinteractionsinreal-time,thusenablingantibodydetectionandmonitoring. HereinwereleasetheguidelinesforthecorrectuseofamethodtodetectspecificantibodiesdirectlyinMultiple Sclerosispatients’serausingaglucopeptide-basedlabel-freebiosensor.Theprotocoldescribes thestrategy employedfortheimmobilizationofglucopeptideantigenontoagoldsensorchipandtheevaluationofthe specificbindingofserumantibodiestotheimmobilizedantigen.

 Label-freemethodfortherealtimescreeningofdisease-specificantibodieswithinafewminutes;  Thedescribedprotocolemployssmallquantitiesofglucopeptideantigenandbloodserumsamplessaving

method-cost;

*Correspondingauthorat:UniversitàdiFirenze,PoloScientifico,DipartimentodiScienzeFarmaceutiche,ViaUgoSchiff,6 I-50019SestoFiorentino,Firenze,Italy.Tel.:+390554573724;fax:+390554573584.

E-mailaddress:rovero@unifi.it(P.Rovero).

1

www.peptlab.eu

http://dx.doi.org/10.1016/j.mex.2015.03.004

2215-0161/ã2015TheAuthors.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http:// creativecommons.org/licenses/by/4.0/).

MethodsX2(2015)141–144

ContentslistsavailableatScienceDirect

MethodsX

 Stabilityoftheimmobilizedglucopeptideantigenguaranteestheregenerationofthesurfaceallowingre-use theimmunosensorwithhighautomatedthroughput.

TheantibodiesdetectedusingthedescribedmethodologycanbeevaluatedasbiomarkersofMultipleSclerosis. TheSPRdetectionsystemisabletocharacterizeantibodiessignificantlydifferentfromthoseevaluatedinthe classicalenzyme-linkedimmunosorbentassays(ELISA).

ã2015TheAuthors.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http:// creativecommons.org/licenses/by/4.0/).

ARTICLE INFO

Keywords:Antibodydetection,Surfaceplasmonresonance,Serodiagnosis,Label-freeimmunoassay

Articlehistory:Received3February2015;Accepted6March2015

Methoddetails

One of the major challenges in Multiple Sclerosis diagnosis is the set-up of simple immunodiagnosticmethods.Infact,thegoldstandardforthediagnosisandprognosisofthedisease is,uptonow,theuseofmagneticresonanceimagingmarkersandcerebrospinalfluidanalysis.Surface plasmonresonance(SPR)techniquehasbeensuccessfully usedtomeasurethebindingofalarge numberofbiomolecularinteractionsincludingthoseofantibodieswithcognateantigens[1].The methodforanti-glucopeptideantibodydetectioninMultipleSclerosisdescribedhereinenables label-freespecificantibodydetectiondirectlyinpatients’sera,usingapreviouslydescribedglucopeptide antigen, termed CSF114(Glc) [2]. A direct comparison of antibody profiles in Multiple Sclerosis patients’serabymeansofenzyme-linkedimmunosorbentassay(ELISA)andSPR-basedbiosensor evidencedthat,fromadiagnosticpointofview,resultsshouldbeindependentlyevaluated[3]. Glucopeptideantigenimmobilization:selectionoftheimmobilizationbuffer

TheglucopeptideCSF114(Glc)waspreparedbymicrowave-assistedsolidphasepeptidesynthesis andfurthercharacterizedbymassspectrometryandanalyticalHPLCasdescribedelsewhere[4].A stocksolutionofCSF114(Glc)waspreparedinpurewater(1

m

g/

m

L)andstoredat+4C.Immediately priortoimmobilizationprocedure,peptidestocksolutionwasdilutedintheimmobilizationbufferto afinalconcentrationof10

m

g/mL.

SensorchipCM5(GEHealthcare,Uppsala,Sweden)wasinsertedintotheSPRdetector(Biacore T100,GEHealthcare).TherunningbufferHBS-EP+10(0.1MHEPES,1.5MNaCl,30mMEDTAand 0.5%v/vSurfactantP20;yieldedpH7.4whendiluted)wasdilutedandflowedoverthesensorchip channels.Allexperimentswereconductedat+25C.

TheimmobilizationbufferwaspreviouslyselectedusingthepHscoutingprotocol,inwhichthe peptide antigen, solved in different buffers, wasflowedovertheinactivesensorchipfor120s at aflowrate of10

m

L/min.Theregenerationofthechipsurfacewasperformedwithapulseof0.1MNaOHfor30sata flowrateof10

m

L/minaftereachsolutioninjected.TheimmobilizationbufferswereusedatpHbetween 3.5andtheisoelectricpointoftheantigeninordertoachievetheelectrostaticpre-concentrationof glucopeptideinthedextranmatrix ofCM5chip(pre-concentrationisfavoredbylowionicstrengthinthe buffer).Thebestimmobilizationbufferwasselectedinjectingtheglucopeptidein10mMcarbonate bufferpH9.6,PBSbufferpH7.2,10mM,1mMand0.1mMacetatebufferatpH4.5,5.5and6.0.Buffersthat give irregular sensorgrams or signals with irregular slopes, probably due to ligand aggregation/ precipitationorchipsaturation,werediscarded.Thebuffer0.1mMsodiumacetatepH5.5presentedthe highestsensorgramslopeandforthisreasonwasselectedastheoptimalimmobilizationbuffer. Glucopeptideantigenimmobilization

The flow cell of the sensor chip surface was activated by injecting a 0.4M 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide(EDC)and0.1MN-hydroxysuccinimide(NHS)mixture(50:50), 142 F.Real-Fernándezetal./MethodsX2(2015)141–144

preparedimmediatelybeforeuse,ataflowrateof10

m

L/minduring420s.TheglucopeptideCSF114 (Glc)wassubsequentlyinjectedat10

m

L/minataconcentrationof10

m

g/mLinthepreviouslyselected immobilizationbuffer0.1mMsodiumacetatepH5.5,usingtheaimofimmobilizationprocedureto raiseafinalimmobilizationlevelof800100resonanceunits(RU).Unreactedsuccinimidegroupson sensorchipsurfacewereblockedbyinjecting60s-pulsesof1MethanolamineatpH8.5at10

m

L/min until complete deactivation. One channel withoutimmobilized ligandwas usedas reference,to removethenon-specificsignaldependingoninteractionsbetweenmoleculespresentinthebiological samplesandgoldonsensorchipsurface.Atthispurposeanotherdifferent_flowcellofthesensorchip wasactivatedandimmediatelyblockedwithethanolamine.

Monitoringglucopeptideantigen-antibodiesinteraction:protocoloptimization

Human serumsampleswerethawed tillambienttemperatureand then diluted1:100and/or 1:50inrunningbuffer.Toestablishareproduciblemethodforautoantibodydetection,dilutedserum samplesofarepresentativehighpositivepatientandahealthycontrolwereinjectedintriplicateat flowrateof30

m

L/minovertheimmobilizedglucopeptideatdifferentcontacttimes(range60–240s). Dissociationwasmonitoredfor60sbyinjectingtherunningbuffersuddenlyaftersamplesataflow rateof30

m

L/min.Interactionswererecordedasseparatesensorgramsandmeasurementsregistered 15saftertheendofeachsampleinjection.Responsesweremeasuredinresonanceunits(RU)asthe differencebetweenreferenceandactivechannel.

Theselectedoptimalconditionsincludethe1:50seradilutioninjectedfor240s,whichpresented increasedsignaldifferencesbetweenpositiveandnegativesamplesmaintaininglow non-specific interactionsinthereferencechannel.Aftereachsampleinjection,surfacewasregeneratedwithtwo 60s pulses of a solution100mMNaOH allowingthecomplete removal of speci_fically and non-specifically bounded biological material from the surface. Following this protocol all further experimentswereperformednotoverandabove100measurementsperchannel.

Theanalyticalvariabilityoftheassaywascheckedscreeningonecharacteristichighpositiveserum atdilution1:50testingsuccessivelythesamesample(15runs)orindifferentexperiments(12runs performedonceaweek).Thewithin-assayandbetween-assaycoefficientsofvariation(SE/mean) werebelow5%andthenconsideredasalowsignalaverage.Thispositivesample,andanothersample selected as negative control, can be used as control each 15 samples measurement during the screeningofapoolofsera,verifyingthestabilityoftheprobeuponalargenumberofcycles.Foreach measurementasamplevolumeof150

m

Lwasemployed,thussmallamountof3

m

Lofpatientserum wasrequiredforeachassay.

Calibrationcurvewithpurifiedantibodies

Immunoaffinitypurifiedanti-CSF114(Glc)antibodies[5] fromonehighpositive controlserum wereinjectedatfivedifferentconcentrations(from1.25

m

g/mLto20

m

g/mL)ontheflowchannels. Therunningbufferwasthenflushedfor60sandchipsurfacewasregeneratedwithtwoinjectionsof 100mMNaOHduring60seachone.Allthestepswereperformedataflowrateof30

m

L/min.The resultingresonancesignalwasplottedagainstantibodiesconcentrationtogivethecalibrationcurve reportedinFig.1.Theobtainedresultsshowedadirectcorrespondencebetweentheinstrumental signalandtheantibodiesconcentrationinsidethesample,indicatingthatthehereinproposedSPR basedmethodcanbesuccessfullyappliedtodetectandtoquantifypolyclonalantibodiesinhuman serumspecificallyinteractingwiththeimmobilizedglucopeptideantigen.

SPRbindingstudieswithhumansera

ThespecificantibodyreactivityofMultipleSclerosispatients’seraagainsttheimmobilizedCSF114 (Glc)antigenwasevaluatedaccordingtotheabovedescribedoptimizedprotocol.Eachsamplewas testedintriplicateataflowrateof30

m

L/min.Humanserawerediluted1:50inrunningbufferand wereinjectedduring240sfollowedby60sofbufferflow.Aftereachmeasurement,chipsurfacewas regeneratedinjectingtwopulsesofa100mMNaOHsolutionduring60s.Interactionofsampleswith F.Real-Fernándezetal./MethodsX2(2015)141–144 143

sensorchipflowcellsweremonitoredasseparatesensorgramsandmeasurementsweretaken15s aftertheendofeachinjection.Theantiglucosylatedpeptideresponses,measuredinRUandresulting fromthesignaldifferencebetweenactiveandreferencechannel,allowedasignificantquantification ofspecificantibodiespresentinpatients'seraascomparedtonormalblooddonors.

SPR-basedanalytequantificationishardtoperformwheninjectingacomplexbiologicalmatrixas sample.Thepresentexperimentsdemonstratedthat,thankstoanappropriatecovalent immobiliza-tionstrategyandtocarefullyoptimizedrunparameters,SPR-basedimmunoassayscanbesuccessfully usedtospecificallydetectantibodiesofprovendiseaserelevance,directlyinpatients’serawithoutthe needofanypreviouspuri_ficationstep.

Acknowledgements

MethodsXthanksthereviewersofthisarticle(RichardB.M.SchasfoortandNathanJ.Wittenberg) fortakingthetimetoprovidevaluablefeedback.

References

[1]A.M.Lokate,J.B.Beusink,G.A.Besselink,G.J.Pruijn,R.B.Schasfoort,Biomolecularinteractionmonitoringofautoantibodies byscanningsurfaceplasmonresonancemicroarrayimaging,J.Am.Chem.Soc.129(2007)14013–14018.

[2]F.Lolli,B.Mulinacci,A.Carotenuto,B.Bonetti,G.Sabatino,B.Mazzanti,A.M.D’Ursi,E.Novellino,M.Pazzagli,L.Lovato,M.C. Alcaro,E.Peroni,M.C.Pozo-Carrero,F.Nuti,L.Battistini,G.Borsellino,M.Chelli,P.Rovero,A.M.Papini,AnN-glucosylated peptidedetectingdisease-specificautoantibodies:biomarkersofmultiplesclerosis,Proc.Natl.Acad.Sci.U.S.A.102(2005) 10273–10278.

[3]F.Real-Fernández,I.Passalacqua,E.Peroni,M.Chelli,F.Lolli,A.M.Papini,P.Rovero,Glycopeptide-basedantibodydetection inmultiplesclerosisbysurfaceplasmonresonance,Sensors12(2012)5596–5607.

[4]F.Rizzolo,G.Sabatino,M.Chelli,P.Rovero,A.M.Papini,Aconvenientmicrowave-enhancedsolid-phasesynthesisofdifficult peptidesequences:casestudyofgramicidinAandCSF114(Glc),Int.J.Pept.Res.Ther.13(2007)203–208.

[5]F.Lolli,B.Mazzanti,M.Pazzagli,E.Peroni,M.C.Alcaro,G.Sabatino,R.Lanzillo,V.BresciaMorra,L.Santoro,C.Gasperini,S. Galgani,M.D’Elios,V.Zipoli,S.Sotgiu,M.Pugliatti,P.Rovero,M.Chelli,A.M.Papini,TheglycopeptideCSF114(Glc)detects serumantibodiesinmultiplesclerosis,J.Neuroimmunol.167(2005)131–137.

y = 27.624x R = 0.99937 0 100 200 300 400 500 600 0 5 10 15 20 25 Response (RU) Antibody [µg/mL]

Fig.1.Measuredvaluesregisteredatthedifferentconcentrationsofpurifiedanti-CSF114(Glc)antibodies. 144 F.Real-Fernándezetal./MethodsX2(2015)141–144