Repositorio Institucional

biosensors

Article

Electrochemical Determination of β-Lactoglobulin Employing a Polystyrene Bead-Modified Carbon Nanotube Ink

Judith Molinari 1, Laura Florez 1, Anahí Medrano 2, Leandro Monsalve 2,3 and Gabriel Ybarra 1,*

1 U.T. Nanomateriales, INTI-Procesos Superficiales, Instituto Nacional de Tecnología Industrial, Av. Gral. Paz 5445, San Martín B1650WAB, Argentina; molinari@inti.gov.ar (J.M.); lflorez@inti.gov.ar (L.F.)

2 Centro de Micro y Nanoelectrónica, Instituto Nacional de Tecnología Industrial, Av. Gral. Paz 5445, San Martín B1650WAB, Argentina; amedrano@inti.gov.ar (A.M.); monsalve@inti.gov.ar (L.M.)

3 Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Buenos Aires C1033AAJ, Argentina * Correspondence: gabriel@inti.gov.ar

Received: 27 September 2018; Accepted: 8 November 2018; Published: 15 November 2018

Abstract: In this article, we introduce the use of a carboxy-functionalized waterborne carbon nanotube ink for the fabrication of an amperometric biosensor aimed at the quantification of β-lactoglobulin. Detection of this protein from cow’s milk was performed by a sandwich immunoassay onto printed carbon nanotube electrodes. The electrodes were printed using a carbon nanotube ink modified with polystyrene beads containing a high amount of carboxylic groups for protein immobilization. This strategy showed enhanced sensing performance compared to the use of oxidative treatments for the functionalization of electrodes. These electrodes showed an excellent electrochemical behavior, and proteins could be immobilized on their surface via the carbodiimide reaction. These antibody-immobilized carbon nanotube electrodes allowed for the detection of β-lactoglobulin in sub-ppm concentrations.

Keywords: electrochemical biosensor; carbon nanotubes; food allergen; β-lactoglobulin



1. Introduction

Food allergies are a growing worldwide concern due to their impact on food safety and public health. For instance, cow’s milk contains several allergenic proteins, such as casein, β-lactoglobulin, and α-lactalbumin. Any food containing allergens should provide a warning to the consumers about their presence. Several methods for the analysis of food allergens have been developed. Commercially available methods for analyzing allergens in foods are based on mass spectroscopy, polymerase chain reaction (PCR), or immunological techniques [1]. Mass spectroscopy is a technique that requires specific, expensive, state-of-the-art equipment and highly trained personnel, and it is only used in cases where a confirmation is required. PCR is used to detect DNA sequences that codify for allergens. Therefore, it cannot be used to analyze oils, milk, or egg white, where the allergenic proteins must be detected. The immunological techniques available in the market are immunochromatography and ELISA (enzyme-linked immunosorbent assay). Immunochromatography is qualitative and it is used to carry out cleaning checks and rapid screening in production lines. ELISA remains the most used method for the detection and quantification of allergens in foods.

The growing interest of the food industry and the increasing regulations require better, faster, and cheaper quality controls. Furthermore, the need to optimize the production with continuous, on-line analysis has led the research of new analytical methods and devices such as biosensors [2,3].



Biosensors 2018, 8, 109; doi:10.3390/bios8040109



www.mdpi.com/journal/biosensors



Biosensors 2018, 8, 109



2 of 9



Electrochemical biosensors are especially attractive because the associated instrumentation to collect and process the signal is affordable and easy to miniaturize. In contrast to the optical detection employed in ELISA, which requires laboratory instrumentation, electrochemical biosensors are easier to miniaturize and integrate into small electronic circuits and portable point on care (POC) equipment. Although a considerable amount of work in the field of biosensors has been carried out in recent years, the development of electrochemical biosensors for the detection of food allergens is quite scarce (see, for instance, reviews given in [4–6]).

The use of nanomaterials provides new strategies for the development of biosensors due to their physical and chemical properties. In particular, electrochemical biosensors with increased selectivity, sensitivity, and reproducibility were obtained when carbon nanotubes (CNTs) were used for the preparation of electrodes [7,8]. CNT-based screen-printed electrodes (SPE) biosensors have been reviewed by Jaiswal et al. [9].

Several challenges concerning the construction of highly sensitive electrochemical biosensors still remain. For instance, the need for protein immobilization, either antigens or antibodies, on the electrode surface requires the presence of functionalized groups (e.g., carboxylic or amino groups) which can act as anchoring points for proteins. The surface functionalization of a carbon electrode often requires the use of an oxidative process such as oxygen plasma [10,11] or oxidative acid treatment, which may impact the conductive and electrochemical behavior of the electrodes [12]. Therefore, there is a considerable interest in the development of electrodes which can be easily prepared, present a good electrochemical response, and can be readily functionalized without the need of further oxidative treatments.

Another major concern for the development of biosensors for the quantification of food allergens is the need to enhance sensitivity and reach lower limits of detection and quantification. This could be achieved by increasing the surface concentration of immobilized proteins [3].

In this work we present a waterborne CNT ink modified with carboxylated polystyrene beads (PSB) for the preparation of electrodes which were in turn used for the construction of a biosensor for the determination of β-lactoglobulin in final rinse samples of clean-in-place systems of production lines. In particular, they are aimed to be used in the manufacture of hypoallergenic formulas obtained either from milk proteins with different degrees of hydrolysis or from non-dairy proteins such as soy. The modification with PSB follows two objectives. On the one hand, the inclusion of carboxylated PSB could avoid the need for oxidative treatments; thus, the process could be simplified and CNT degradation could be avoided. On the other hand, since the number of carboxylic functionalities increases with the number of PSB used in the preparation of the electrodes, a higher sensitivity could be achieved by increasing the surface concentration of the proteins used for biorecognition. We critically discuss how the inclusion of a functionalized component in a waterborne CNT ink affected both the electrochemical behavior of the electrodes and the performance of the amperometric biosensors fabricated employing the PSB-modified CNT ink for detection of hydrogen peroxide with horseradish peroxidase (HRP) enzymatic electrodes, as well as the detection of β-lactoglobulin with a capture (sandwich) immunosensor.

2. Experimental Section

2.1. Reagents

N-hydroxysuccinimide (NHS), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), o-phenylenediamine dihydrochloride (OPD), Type I horseradish peroxidase (HRP), β-casein and β-lactoglobulin from bovine milk, and rabbit polyclonal antibody anti-(bovine β-lactoglobulin) were purchased from Sigma-Aldrich (St. Louis, MO, USA), and rabbit polyclonal antibody anti-(bovine β-lactoglobulin)-HRP conjugate was purchased from Abcam (Cambridge, UK). Different buffers were used: phosphate buffer saline (PBS) of pH 7.4 (0.1 M NaH2PO4 (Mallinckrodt) and 0.15 M NaCl (Biopack)); PBS, 0.5 M NaCl, and 0.2% polysorbate 20 (Biopack); 0.1 M phosphate buffer of pH 7.0



Biosensors 2018, 8, 109



3 of 9



(KH2PO4 (Merck)); blocking buffer (0.01% polysorbate 20 (Biopack) and 1% gelatin (Merck) in 0.1 M buffer phosphate of pH 7.0); rinsing buffer (0.05% polysorbate 20 (Biopack) in 0.1 M buffer phosphate of pH 7.0); and measurement buffer (PBS 0.1 M KCl, 4 mM hydroquinone, and 1.5 mM hydrogen peroxide (Biopack)).

2.2. Optimization of Experimental Parameters and Specificity of the Immunoassay

The concentration of β-lactoglobulin used for the immobilization, the dilutions of the primary and secondary antibodies, the used buffers, the incubation time, and detection parameters were optimized by ELISA based on a protocol developed by Vitkova et al. [13]. The optimized values found for ELISA were tested on the biosensor, and the dilutions of the primary and secondary antibodies were adjusted again to obtain the optimal electrochemical signal.

The polystyrene microplates were filled with 1 and 10 µg mL−1 anti-(bovine β-lactoglobulin) rabbit polyclonal antibody in 0.1 M carbonate/bicarbonate buffer of pH 9.0 (50 µL per well) and incubated for 1 h at room temperature. The coated plates were washed three times with 0.5 M PBS buffer containing 0.5 M NaCl and 0.2% polysorbate 20 to remove unbound antigens. Each well was then filled with 50 µL of different concentrations (0, 1, 10, 25, 50, and 100 µg mL−1) of β-lactoglobulin from bovine milk in PBS containing 1% gelatin for 2 h at 37 ◦C and was washed three times with washing buffer, and the mixture was incubated for 1 h at 37 ◦C 50 µL with 0.1 and 1 µg mL−1 anti-(bovine β-lactoglobulin)-HRP conjugated rabbit polyclonal antibody in washing buffer and washed three times with washing buffer. Aliquots (50 µL per well) of 1 mg mL−1 OPD with 50 mM citrate/phosphate buffer of pH 5.0 containing 0.3 mg mL−1 H2O2 were incubated during 15 min at room temperature in darkness. The reaction was stopped with 50 µL per well of 2 M H2SO4. Absorbance was measured at 492 nm.

The tested values of concentration of β-lactoglobulin used in the immobilization on polystyrene wells were 0, 1, 10, 25, 50, and 100 µg mL−1. The used primary antibody dilutions were 1 and 10 µg mL−1, while the used dilutions of the secondary antibody were 0.1 and 1 µg mL−1. The best results for ELISA were obtained with primary antibody dilutions of 1 and 10 µg mL−1 for the immobilization on polystyrene and with dilutions of 0.1 µg mL−1 for the secondary antibodies. When the optimized values found for ELISA were used in the electrochemical biosensor, it was found that a higher primary antibody concentration was necessary, so a value of 100 µg mL−1 was used.

Since the method proposed in this work is meant to be specific for the detection of β-lactoglobulin in rinse samples obtained after cleaning-in-place of production lines, proteins that could contaminate the surface of the equipment and products were selected and tested for cross reaction. The specificity of the immunoassay was tested against bovine serum albumin, β-casein, and a soy milk extract. Samples of β-casein and bovine serum albumin of different concentrations (0.1, 1, 10 and 100 µg mL−1) were prepared in PBS containing 1% gelatin. A soy milk formula (Nutrilon ProExpert, Nutricia Bagó, Batch No. PTL170342/00145 C2, expiration date 5 May 2019) with a weight content of 13% of soy proteins was extracted and diluted according to Ridascreen Fast b-lactoglobulin ELISA kit. It was found that bovine serum albumin and the soy milk extract did not exhibit any significant cross reaction. On the other hand, a cross reactivity of 1% to β-casein was found.

2.3. Preparation of the Electrodes

Thick film carbon electrodes were printed onto acrylic substrates by screen printing technology as described in [14] (Figure 1a), and the working electrodes were then further coated by drop casting using 0.5 µL with a waterborne CNT ink composed by 2.5% multiwalled carbon nanotubes (Nanocyl, Sambreville, Belgium), acrylic-styrene resin Joncryl® 617, and polyvinylpyrrolidone (PVP K30). This ink, to which 2% carboxylated polystyrene microspheres (Ø 0.9 µm, Sigma Aldrich) was added, was prepared according to our previously described procedure [15]. The electrodes were integrated in an electrochemical cell constructed with poly(methyl-methacrylate) (Figure 1b). The geometrical area of the electrodes was 0.8 mm2. Oxygen plasma treatment was used in the preparation of CNT



Biosensors 2018, 8, 109



4 of 9



ink enzByiomsenesoersle20c1t8r, o8,dxeFsOR(OPEPETR -RCEVNIETW). In this case, the CNT ink electrodes were treated with4 oafn9 oxygen



plasma to promote the formation of carboxylic groups. A Diener plasma polymerization equipment



was



usepdolwymitehrizaantioonxyeqgueinpmpreenst swuarseuosfed1

and a treatment time of 15 s.



wmitbhaar,naoxsyegt etnemprpesesruarteuorfe1omf 5b0ar◦, aCs,eatntedmapetrreatautrme eofn5t0ti°mC,e



of



15



s.



a



c b

Figure F1ig. ure(a)1.S(car)eeSncr-eperni-nptreindtedeleecletcrtoroddeess;; (bb)) eelleecctrtorcohcehmeimcailcacelllcs;el(lcs); e(lce)ctreolcehcetmroicahlemplaictfaolrmplatform comprechoemnpdrienhgenedleincgtreolcehctermochiceaml iccealllsceallnsdanedleecltercotnroinciicnisntsrturummeennttaattiioonn ccoonnnneectcetdedtotoa alalpatpoptovpiavUiaSUB SB port.

port.

HRP was immobilized according to a previously described procedure [15]. Rabbit polyclonal

HRP was immobilized according to a previously described procedure [15]. Rabbit polyclonal

antibodainetsibaonditeis-(abnotvi-(ibnoevβin-elaβc-ltaocgtologlboubulilnin))wweerree iimmmmoobibliizleizdeodntoontthoe tehleecterloedcetrsoudrfeacseubryfamceeabnys omf eans of the carbthoedciaimrboiddieimreidaectrieoancti[o1n6][1b6]etbweteweenenththeepprriimmaarryyaamminiongoroguropsuopfsthoef athnteibaondtyibaonddythaencadrbthoxeycl arboxyl groupsglroocuaptseldocoantedponlypsotylyrsetynreenpeaprtairctilcelsesimimmmerrsseeddininthtehceacrbaornbonnannoatunboetiunbkewiinthk1w00itμhL1o0f00.µ1LMof 0.1 M EDC anEdDC10anµdL10ofμ2L5omf 2M5 mNMHNSHfoSrfo3r03m0 mini.n. AAfftteerr wwaashshinign,g5,05μ0LµoLf aof10a0 1μ0g0mµLg−1manLti−-(1boavnintie-(bovine β-lactoβg-lloabctuoglilnob)uslionl)ustoioluntioinnin0.01.1MM pphhoospsphahtaetbeufbfeurffoefrpHof7.p0Hwa7s.i0ncwubaasteidnfcourb2ahteadt 3f7o°rC.2Ahfteart 37 ◦C. After wwFaianssahhlilinyn,ggt,,httehheeeleelecetlcretocrdoterdosedws weesreerweweianrscehueibdnacateugdabioanvtaeenrdndiogrevhatedarytn4tiog°Chbetinuastae4wd.◦eCt chinamabwerewt icthhathme bbleorckwinitghbuthffeerb. locking buffer. Finally, the electrodes were washed again and ready to be used.

2.4. Electrochemical Measurements

2.4. Electrochemical Measurements

Solutions of β-lactoglobulin in PBS of different concentrations ranging from 0.02 to 10 ppm

Sowluetrieoinnscuobfaβte-dlaacntdogadlodbeudltioneianchPeBleSctorfoddeifffoerr2enhtacto3n7 c°Cen. Wtrastihoendselreacntrgoidnegs wfreorme in0c.u02battoed1w0 ipthpm were incubat0e.1dμagnmdLa−1dadnetid-(btoovienaecβh- lealcetocgtrloobduelinfo)-rHR2Phcoantju3g7at◦eCd.aW ntiabsohdeydinePleBcStirnodtheessawmeerecoinndcituiobnast.ed with 0.1 µg mElLec−tr1ocahnetmi-i(cbaol vmienaesuβre-mlaecnttosgwloerbeuclairnri)e-dHoRuPt acto2n5j°uCgiante50dμaLnotifbaoPdByS ibnufPfeBrSofinpHth7e, 0s.a1mMeKcColn, ditions. Electrocu4hsmeedMmfoihcryadtlhrmeoqaeumainspuoenrroeemm(reeetdrnioctxsmmweeaedsriueartecomar)re,rnainetsdd(F1o.i5ugutmraeMt12Hc5)2.O◦PC2o.tAeinnpti5oa0rltsaµwbLleeropefomtaeenPatBsiouSsrtebadtuNafnfaednroroepffoecprr[He1d77]t,wo0ai.ns1 M KCl, 4 mM hthyedtreoxqt uaginaionnstea(rAegd|oAxgCml|e0d.1iaMtoKr)C, larnedfer1e.n5cemeMlectHro2dOe.2F. oAr tphoe ratmabpleeropmoetternictmioesatsautrNemaennotps,othce[17] was used fowr otrhkeinagmelpecetrodmeeptortiecnmtiaelawsausrseemt aetn−t0s.2(8F0igVu, arned1tch)e. rPeosuteltnintgiaclusrwrenetrewmasereacsourrdeed aatn6d0 sr.eTfehrisred to in the textpoatgenatiinasl tvaaluAe gis|nAeggaCtilv|e0e.1noMughKtCo lprroedfeurceentchee reeldeuctcrtioodneo.f Fhyodrrothqeuinaomnpe etoro1m,4 ebternizcomqueinaosnuerements, the woruknidnegr deilfefucstiroond-ceopntorotellnedtiacolnwdiatisonsest[1a8t].−0.280 V, and the resulting current was recorded at 60 s.

This po3te. Rnteisaullvtsalue is negative enough to produce the reduction of hydroquinone to 1,4 benzoquinone under diffusion-controlled conditions [18].

Scanning electron microscopy (SEM) images of electrodes prepared with PSB-modified CNT

3. Resuilntkss are shown in Figure 2a,b. PSB can be clearly seen in the surface, partially covered by

interconnected CNTs. A cross section obtained by a focused ion beam (Figure 2c) showed that PSBs

Scawnenreinegveenlelyctdroisntrimbuitcerdosicnotphye (fSilEmMs )animd athgaets tohfeeilnekcterfofdiceiesnptlyrepcoaarted wthiethsuPrSfaBc-emoof dthifie eSdPEC,NT inks are shoawssnuriinngFaigguoroed2eale,bct.riPcaSlBcocnatnacbt,ewcitlheaartlhyicskeneenssionf tahroeusnudr1fa0cμem, p. artially covered by interconnected CNTs. A crTohsesesleeccttrioocnheombitcaalinbeedhabviyoraoffotchuesperdinitoedn ebleecatmrod(eFsigwuarseex2pcl)osrehdowwiethd hthydartoPquSiBnsonwe ebrye evenly

cyclic voltammetry. Figure 3 presents the cyclic voltammograms obtained for an uncoated screen

distribuptreindteidnctahrbeofin lemlesctraondde (tShPaEt),thane SiPnEk ceofafitecdiewntitlhy tchoeaCteNdT tihnke wsuitrhfoauctePoSBf st,haendSPanE,SPaEsscuoraitnedg a good electricwalitchonthteacPtS, Bw-mithodaifitehdicCkNneTssinokf. aItrocaunndbe1s0eeµnmt.hat the cyclic voltammogram obtained for the



uncoated SPE presented the biggest peak potential difference, which is associated with quasi-reversible kinetics. When the CNT ink without PSBs was used, the peak potential difference was much smaller and the peak current was significantly higher. This fact has been reported before and is related to the electrocatalytic properties of CNTs toward the oxido-reduction of hydroquinone/bezoquinone [15]. With the inclusion of carboxylated PSB, only a slight decrease in Biosensorst2h0e18c,u8r,r1e0n9t was observed, indicating that PSB did not affect significantly the electrochemical behavior of the CNT electrode.



5 of 9



Biosensors 2018, 8, x FOR PEER REVIEW



5 of 9



uncoated SPE presented the biggest peak potential difference, which is associated with quasi-reversible kinetics. When the CNT ink without PSBs was used, the peak potential difference was much smaller and the peak current was significantly higher. This fact has been reported before and is related to the electrocatalytic properties of CNTs toward the oxido-reduction of hydroquinone/bezoquinone [15]. With the inclusion of carboxylated PSB, only a slight decrease in tahe current was observed, indicating that PSB did not affect significantly the electrochemicabl behavior of the CNT electrode.



c

mFiaggaunreifi2mFc.iaagtSguioncreainfnis2cn.aotSifinco5agnn0sn0eoi0lfne×5gc0te0(ralo0e)×ncta(rainom)ndaani7mgd1ea7,2sg13e,o2s53f×o5ef×l(e(ebblce))tc,,rtaarononddddeeassacpprcorrrisionsntssetseedcsdtweiocwintthiiootCbhntNaoCiTbnNetiadnTikbniywendfkiothcbwuypsioetfdlhoycispoutynosrelbeydnesaetimoybner(eac)bdn.seeaabmt ea(cdb)s. at

The electrochemical behavior of the printed electrodes was explored with hydroquinone by cyclic voltammetry. Figure 3 presents the cyclic voltammograms obtained for an uncoated screen printed carbon electrode (SPE), an SPE coated with the CNT ink without PSBs, and an SPE coated with the PSB-modified CNT ink. It can be seen that the cyclic voltammogram obtained for the uncoated SPE presented the biggest peak potential difference, which is associated with quasi-reversible kinetics. When the CNT ink without PSBs was used, the peak potential difference was much smaller and the peak current was significantly higher. This fact has been reported before and is related to the electrocatalytic properties of Cc NTs toward the oxido-reduction of hydroquinone/bezoquinone [15]. With the inclusion of carboxylated PSB, only a slight decrease in the current was observed, indicating

Figure 2. Scanning electron images of electrodes printed with CNT ink with polystyrene beads at

that PSB didmnagontifaicfafteiocntssoifg5n0i0fi0×ca(an) talnydt7h1e,23e5l×e(cbt)r,oancdheamcroicssaslebcteiohnaovbitoairneodf btyhefocCuNsedTioenlebcetarmod(ce)..

Figure 3. Cyclic voltammograms for hydroquinone at a carbon screen printed electrode (orange), a CNT-ink-coated electrode (black), and a PSB-CNT-coated electrode (green) of 4 mM hydroquinone in a 0.1 M phosphate buffer solution of pH 7.4 at a scan rate of 0.05 V s−1.



Figure 3F.iguCryec3l.icCyvcolilctavmoltmamomgroagmrasmfsofrorhhyyddrrooqquuininoonne eataat caarcbaornbsocnreesncrpereinntepdreinletcetrdodeele(ocrtarongdee), (aorange), a CNT-iCnNk-Tc-oinakte-cdoaetleedcterloecdtreo(dbela(bclka)c,ka),nadnda aPSPBSB-C-CNNTT--ccooaatteeddeelelectcrtordoed(eg(rgeerne)enof) 4ofm4MmhMydhroyqdurionoqnueinone in a 0.1 M ipnhao0s.1phMaptehobsupfhfaetresboulfufetriosnoluotfiopnHof7p.4Ha7t.4aastcaasncarnartaeteofof00.0.055VV ss−−1.1.



Biosensors 2018, 8, 109



6 of 9



In oBirodsenesrortso20a18s,s8e, xsFsOtRhPeEpERerRfEoVrIEmWance of PSB-modified CNT ink as a material for the pro6dofu9ction of enzymatic electrodes, HRP was immobilized onto the electrodes employing the carbodiimide reaction

between theIncaorrdbeorxtyolaiscsegsrsotuheppseirnfotrhmeabnceeaodfsPaSnB-dmtohdeifiaemd CinNoTgirnokuaps sa minattehreiael nfozrythmeep.roTdhuecstieonenozf ymatic electrodeenszywmearteicteelsetcetdrodfoesr, tHheRPquwaanstiifimcmatoiboinlizoedf hoyndtorotgheenelpecetrroodxeids eemanpdloyainlignethaer dcaerpboednidimeindcee of the measureerdenazccyutmiorrnaetibncettewwleeictethnrottdhheeesccoawrnbecroeexnyttlericsattgeidroonufoporsf Hitnh2etOheq2ubianenattdhifseica0ant–ido1n.t5hemofaMmhyirndaornogggreeo,nusphpsoeiwrnoixtnhidgeetehnaaznytdmthaee. iTlmihneemsaerobilized enzymedseepxehnidbeintecedoaf thhiegmh ecaastuarleydticcuarrcetnivt iwtyit.hTthheesceonrceesnutlrtastiwoneorfeHco2Om2 pinatrheed0–to1.5HmRMP erannzgyem, shaotiwc ienlgectrodes preparedthawt itthhe CimNmTobinilkizeedleecntrzoydmeess ienxhwibhitiecdh aoxhyigghecnatpalaytsimc aactwivaitsy.uTshedeseforrestuhletsgweenreercaotimopnaorefdcatorboxylic

functionHaRliPtienszpyrmeavtioc uelsecttorotdheesipmrempaorbediliwzaitthioCnNoTfinHkRePle.cAtrondienscirnewasheicihnocxuyrgreennptldasemnasiwtyasoufsaerdofuonr d 100% was foutnhedgwenheerantitohneofPcSaBrb-moxoyldicififuendctCioNnaTlitiineskpwreavisouussetodth(Feiigmumreob4i)li.zaEtinoznyomf HatRicP.eAlenctinrocrdeeassewinere also tested ocEnuntrzoryeSmnPtaEdtiecwnesliietthcytrotohfdeaersaodwundedrieti1ao0ln0so%otfewstatwesdofoolunantyodeSrwPshEoewfnCitthhNetThPeSiBnad-kmd(oit2di0oifniµeomdf CtwtNhoiTclakiny)ekarsnwdoafsiCnuNsteThdiis(nFkciga(u2s0ereμn4mo). further increasethinickc)uarnrdenint tdheisncsaistye nwoafusrftohuernidnc.rease in current density was found.



Figure 4F.iguPreerf4o. rPmerafonrcmeanocfedoifffdeirfefenrtenHt RHPRPenenzzyymmaaticc eelleecctrtordoedse. sB.arBs asrhsowshs othwesvtahlueesvaolfuceusrroenftcurrent densitydmenesaitsyumreedasuarted60ats6a0tsaant aanppaplipeldiedpoptoetennttiiaall ooff −−0.02.8208V0 Vin ianPaBSPbBuSffebrucfofenrtaicnoinngta4inminMg 4 mM hydroquphiylandsormonqae-utiirnneoattnheedeinCabNthsTee-nainbckse-epnorcfienHtoe2fdOHe22lOe(c2gt(rrgoerdyeey)s)a(anOndPdTww-CiittNhhTtth)h,eeaandadddicdtaiiortbniooonnf S1oP.f5E1m.c5MoamtHeMd2Ow2Hi(tr2heOdo)2nfe(orraenodxd)yftgwoernooxygen plasma-ltaryeearsteodf PCBNS-mT-oidnikfi-epdrCinNteTdinekl(ePcStBro-CdNeTs)(.OPT-CNT), and carbon SPE coated with one and two layers of PBS-modified CNT ink (PSB-CNT).

The PSB-modified CNT ink printed electrodes were then used for the production of a portable

TheimPmSuBn-omseondsoifir efodr CthNe Tquiannktifpicraintiotendoef lβe-clatrcotodgelosbwulienr,eotnheeonf uthseedmfaoinr mthielkparloledrguecntiicopnrootfeiansp. ortable

immunoTshenmsoetrhofodrwtahsebaqsuedanotnifiacsaatniodwniochf iβm-mlaucntoagslsoabyu, ilninw, hoincheaollfertgheensminasinolumtioilnkwaellrercgaepntuicredproteins. The metbhyodanwtibaosdbieasseidmomnobailsizaenddwonictoh itmhemeulencotraosdseays,uirnfawcehiacnhdaliltsergperenssenincesowluastiorenvweaelerde cwaipthtured by antibodiHeRs Pim-comnojubgialtiezdedaonntitbootdhiees.eleTchterodmeesausurrfeadce ealnecdtriotcshpermeisceanl cesigwnaasl riesvetahluesd wreiltahtedHRtPo-ctohnejugated antibodiβHe-Rsla.PcT-tcohogenlojmubguealaitnesdurcaeondnticbeeonldetcryat.rtioocnheimn icthael sisganmaplleis tshoulustiroenlsa,tedduteo tthoe tβh-elacetnozgylmoabtuiclinacctoivnictyenotrfation in the sample sAolsuchtieomnast,icdureeprteosetnhteateionnzyomf tahteicsaancdtiwviicthy iomfmHuRnPoa-scsoanyjuisgsahtoewdnanintitbhoediny.set of Figure 5.

A sAcfhteerminactuicbarteiopnrewseitnhtastaimonpleosf cthonetasiannindgwdiicfhferiemntmcuonncoeanstrsaatiyonis sohf oβw-lanctiongltohbeuliinnsaentdofthFeigure 5.

After inHcRuPb-acotinojungawteidthantsiabmodpy,lePsBScbounftfearincointgaindinifgfe1r.5enmtMcHon2Oc2eanntdra4timoMnshyodfroβq-uliancotnoeg, lwohbiuchlinwaas nd the uHsRePd-caosducnaosejlueprldeegecndaatdestodeoaxd.nTrmeathhdneeeotdvxicbaioamlonutcdeoeedyrnoi,,atfrwPttaohBtraie,Soswcnbupaourlsfrafeβpfcne-letlaradcoccebtiodotnangiilntcnoaoebcidnonuntliaiantntacg,6cta0t1swws.s5ihaitttmohhawtMnthnheaiepHnepleF2elliOicegetdcru2otrpradeoone5dstd.eTean4nshtideamalencoMxudpfrre−cher0uniy.m2trd–8ret0reniomnVtqateluw–pctiauoinsmrionvfnoteesuseccn,woudwuerlrtvhdoeeicshwweares collectedbe. fTithteedvtoalauepoowfetrhleawcu, arnrednat loimbtiat ionfeddetaectt6io0n sofat0.a17n3appppmliewdaspeosttiemnatiteadl ofrfo−m0t.h2e8s0taVndwarads found to depednedvioantiotnhoef cthoenbcleannktr.aUtsioinng tohfisβe-xlpaecrtiomgelnotbalusleitnu,pa, sit swhaoswponssiinbleFtiogudirsecr5im. iTnhatee eβx-lpacetroigmloebnultianl points could becofintcteendtrtaotioanps orawngeirnlgafwro,manthdeasulibmpiptmofledveetletoct1io0nppomf .0.173 ppm was estimated from the standard

deviation of the blank. Using this experimental setup, it was possible to discriminate β-lactoglobulin

concentrations ranging from the sub ppm level to 10 ppm.



Biosensors 2018, 8, 109

Biosensors 2018, 8, x FOR PEER REVIEW



7 of 9



7 of 9



Figure F5i.gEulreec5t.roElcehctermocihceaml idcaeltedremterimnaintaiotinonooffββ--llaaccttoogglolobbuluinli.nD.eDpeenpdeenndceeonfcteheofcuthrreenctumrreeansutrmedeaatsured at



60 s on 6th0esβo-nlacthtoegβlo-lbacutloignlocbounlicnenctornacteinotnraatitoannaatpapnlieapdpplioedtenptoitaelnotifal−o0f.2−800.2V80. EVr.roErrrboarrbsawrserweecraelculated



calculated from the standard deviation of three independent experiments. The experimental points



afropmowthewcremlesar−t2weaanf:intdydtea=dxrdt9too1da.te2hpvxeoi0awc.0to9eino5rc,nleawnwothfr:aeytthri=eorn9ey1eo.c2ifonxβr0d.r0-9lee5a,spcwpteoonhgnedlordeebsnuyttlcioenoxrterhpxeeespprrcioeumsnrsderesendnttoistn.tdhTpeephncmesuiert(rxyRep2neet=xrdp0iem.r9ne7ses7inst)ye.tadIenlxisppnerote:µisnSsAcethsdcewmimne−aμrt2Aeicafinttdedx



to to



the conrceepnrtersaetnitoantioonf oβf-tlhaectcoapgtluorbeuimlinmeuxnpoaresssasyedforinthpepemlect(rRo2ch=em0.i9ca7l7d).etIenrsmeitn:aSticohneomf βa-tliaccrtoegplroebsuelinntation of



the cap(tbuluree ipmenmtaugonno);aRsseadyanfodrOtxhsetaenldecftorrohcyhdermoqiucianlodneetaenrdm1i,4nbaetinoznoqoufinβo-nlea,crteosgpleoctbivuelliyn. (blue pentagon);



Red and Ox stand for hydroquinone and 1,4 benzoquinone, respectively. 4. Discussion



4. DiscussioAns shown in Figure 3, the waterborne CNT ink used in this work presented electrocatalytic



AsprsohpoewrtinesitnowFairgdusrtehe3o,xtihdoe-rwedautcetriobnoronf ehyCdrNoqTuiinnokneu/1s,e4dbeinnzotqhuisinwonoer, kwhpicrhesisenoftteedn uesleedctarsoacatalytic propertreieleedscotrxtoomcwheeadmridaictsoarlthibnieohsoyexndisrdoorgose-,nreepldeecurtorcoxtdiidoeens emonfzuyshtmybadetircfoubqniuocstiienononsanolreisz/e[1d1,14t]o. bHpeornowzveoivdqeeur,aiinnncohonoredri,enrwgtohsiitbceeshufiosserdothfinteen used as a redimomx ombileizdaitaiotonrofinbiohmyodlercougleesn. Apserreopxoirdteed beenfzoyrem, oaxtiidcatbivieotsreenatsmoernsts[1su1]c.h aHs cohwemeivcealr,oxinidaotirodner to be used inanedlepcltarsomcahetrmeaitcmaelnbt iuosseednfsoorrtsh,eegleencetraotdioensomf cuasrbtobxeylifcufnucntcitoionnaalliztiesdotnothperoCvNiTdecaannicmhpoariring sites



for the tihmeimr coobnidluizcatitvieonproopfebrtiioems. Tohleeciuntlreosd. uActsiorneopfoartlaerdgebneufomrbee,rooxfisdtrautcivtueratlredaeftemctesnmtasysaudcvherasselcyhemical oxidatioafnfecatntdhepirlaesxmcelaletnrteaeltemctreonnt turasnesdpofrotratnhde mgeecnhearnaictiaol nproofpecratirebso[x1y2]l.icOfnunthcetiootnhaerlithiaensdo, nthtehe CNT can imptporaeipsraerntthciaeeliorbflPcoSockBnisndoguncottfhivteheseuiprrfreaolceepcteorfrottaihceetsive. leeTcathrreoead.ienWmthrioegndhtuSaPclstEiosoancfofeaoctfetdtahweliairtrhegleethctenrouCcmNheTbmeiinrckaolwfbeisththraouvuicottruPdSruBaesl defects may adwveerreseulyseadf,fethcte thpeeairk epxocteelnletinalt einlectthreonvotrltaanmsmpoogrrtaamnsdsmhoewchedantihcaatl pthreopoexridtioe-sre[d1u2c]t.ioOnnotfhe other hand, thhyedrporqeusiennoncee/boefnPzoSqBusinoonnethweassucrafrarcieedoofuttheexeclleucsitvreoldyebmy iCgNhtT,alwsoithaofufetctantyhesiirgneleocftrtohcehemical behaviodristdinucetivtoe psiagrntaial l dbuloe cktoingunodfertlhyeinigr eclearcbtoronacStPivEe. aAreasi.mWilahr ebnehSaPvEiosr cowaatsedobwseirtvhedthefoCr NT ink



withoutPPSBS-BCsNwT-ecroeatuesdeSdP,Etheelecpteroadkeps,ootnelnytaiaffleicntedthbeyvaoslltiagmhtmdeocgreraasme isnschuorwrenetddtehnastityth, we hoixcihdmo-aryedbeuction of hydroqautitnriobnutee/dbteonazsoliqguhtilnyosnmeawllearsCcNaTrraicetdivoe uarteeaxdculue stoivtehleyinbcyluCsiNonTo,fwPiStBhso. ut any sign of the distinctive signal dcounefiTgtohueruartneiosduneltrcsloyonbifntiargminceedadrwbthoiethnhtShyPepEoHt.hRAePsiessnizmthyamiltaatrthicbeeelihencactlrvuoisdoioersnwaonafdscthaorebbisomexrmyv-ufeundnocsfteoionrnsoaPrlSizinBed-aCsPNaSnBTd-wcisoicahated SPE electrodcoens,voennileyntarfofeuctetetdo abvyoiadstlhieghpotsdt-efucrnecatisoenainlizcautirornebnyt odxeindasititvye, wtrehaitcmhenmtsa.yThbues,atthtreibpuretpeadrattoioan slightly smalleroCf NenTzyamctaitvice ealreectarodduese tcooutlhdebiencsliumspiolifnieodf wPSitBhosu. t affecting the electrochemical properties of



ThCeNreTs.uMltsoroebotvaeirn, ewdhewnitchomthpearHedRtPo eonxyzgyemn aptliacsmelaecttrreoatdeedsCaNnTd itnhke, iamn minucrneaosseeninsothreinsigansaalndwich configuirnatteinosnitycoonffiarbmouetd1t0h0e%hywpaosthobesseisrvtehdatwthiteh itnhceluinsciolunsioofncaorf bPoSxBy-infutnhcetiConNaTlizinekd uPsSeBd ifsoar cthoenvenient route topraevpoairdatitohneopfoesntz-yfumnactitcioenleactlrizodateiso. nThbeyseorxeisdualttsivsuegtgreesatttmhaetnthtse. iTnkhucosa, ttihnge wpritehptahreaatidodnitioofneonfzymatic electrodaPceShsBisecvaoeluldol.wdHebodewaesnvimienrc,prieltiafiisseeadinppwthairetehanmot utohtuanattfofoneflcyitmianmgfroabtchitlieiozneedloeHfcttRhrPoecaohnuedtmmthoiusctsallaapyheirgrohpoefertrshteienesisnitkoivfcitoCyaNtcionTugsld.wMbaesoreover, when coacmtivpeafroerdprtooteoixnyimgemnobpillaizsamtioant.rOetahteerdwCiseN, tTheincukr,raennt idnecnrseitayswe oinultdhheasvieginnacrleiansteednlisniteyarolyfwabitohut 100% was obstehrevtehdickwnietshs tohf ethienfcilumsoiornthoefnPuSmBbienr othf leayCeNrsTofinPkSBu-CseNdTfoinrkt.he preparation of enzymatic electrodes.



These resultRsesguargdginesgttthheaqt uthanetiifnickatcioonatoinf gβ-wlacithogtlhobeualidnd, itthieonreosuflPtsSoBbstaailnleodwuesdinagnthinecPreSaBs-me oinditfhieedamount of immoCbNiTlizeeldectHroRdPesasnhdowthtuhsataiht iigshpeorsssiebnlesittoivciotyvacloenutlldy bimemacohbiileizveedan. tHi-βo-wlaectvoegrl,oibtuilsinapanptaibroednitesthat only a fractiownithofthtehperoopuetrmoroisent tlaatyioenraonfdtihnesuinffkiciceonat atimnoguwntaass atocteinvheanfocerspenrositteivinityi.mAmlimobitiolifzdaettieocnti.onOotfherwise,



the current density would have increased linearly with the thickness of the film or the number of



layers of PSB-CNT ink.



Regarding the quantification of β-lactoglobulin, the results obtained using the PSB-modified



CNT electrodes show that it is possible to covalently immobilize anti-β-lactoglobulin antibodies



with the proper orientation and in sufficient amount as to enhance sensitivity. A limit of detection



Biosensors 2018, 8, 109



8 of 9



of approximately 0.2 ppm was estimated. This value is low enough to consider this method as an alternative to those commercially established for milk allergen analysis in clean-in-place practices. A maximum value of 2 ppm of β-lactoglobulin has been proposed as a threshold in final rinse waters in clean-in-place practices [19]. Values for the limit of detection in ELISA typically lie in the 1–5 ppm concentration range [20], and limits of detection lower than 0.2 ppm are unusual in commercially available kits [21]. Recently, electrochemical biosensors developed for food allergen detection with limits of detection ranging from values as low as a few ppbs to ppms have been reported [2,19,20].

In sum, the hypothesis that the combination of a carboxy-functionalized microparticles and CNTs might provide the anchoring sites needed for protein immobilization while preserving mostly unaffected the electrochemical and conductive properties of CNTs has been proved correct. It has also been shown that a higher sensitivity can be achieved in enzyme electrodes by the use of a component (PBS) which increases the amount of immobilized proteins.

5. Conclusions

A waterborne PSB-modified CNT ink was developed and used for printing electrodes which exhibited enhanced electrochemical properties when compared with carbon SPE and plasma treated carbon SPE electrodes. Polystyrene beads containing a high amount of carboxylic groups permitted the direct protein immobilization via the carbodiimide reaction without the need of oxidative treatments and with a negligible adverse impact on the electrochemical performance of carbon nanotubes. Using this modified carbon nanotube ink, a sandwich immunoassay was developed for the determination of β-lactoglobulin, an allergenic protein from cow’s milk, with a detection limit low enough for acceptance as an inexpensive and portable screening test in the food industry.

Author Contributions: Conceptualization: J.M. and G.Y.; methodology: J.M, L.F., A.M. and LM.; writing—original draft preparation: J.M. and G.Y; writing—review & editing: L.M. and G.Y.; supervision: G.Y.

Funding: ANPCyT: PICT 2013-0427, PICT 2014-3748, PICT 2014-3750, and PICT 2017-2787.

Acknowledgments: We thank INTI and ANPCyT (PICT 2013-0427, PICT 2014-3748, PICT 2014-3750, and PICT 2017-2787) for financial support, Paulina Lloret and Lionel Veiga for SEM images, Gustavo Giménez and Luciano Patrone for the focused ion beam experiments.

Conflicts of Interest: The authors declare no conflict of interest.

References

1. Sharma, G.M.; Khuda, S.E.; Parker, C.H.; Eischeid, A.C.; Pereira, M. Detection of Allergen Markers in Food: Analytical Methods. In Food Safety: Innovative Analytical Tools for Safety Assessment; Scrivener Publishing LLC: Beverly, MA, USA, 2017; pp. 65–121. ISBN 9781119160588.

2. Moina, C.; Ybarra, G. Fundamentals and applications of immunosensors. In Advances in Immunoassay Technology; InTech: Zagreb, Croatia, 2012.

3. Felix, F.S.; Angnes, L. Electrochemical immunosensors—A powerful tool for analytical applications. Biosens. Bioelectron. 2018, 102, 470–478. [CrossRef] [PubMed]

4. Pilolli, R.; Monaci, L.; Visconti, A. Advances in biosensor development based on integrating nanotechnology and appiled to food-allergen management. Trends Anal. Chem. 2013, 47, 12–26. [CrossRef]

5. Vasilescu, A.; Nunes, G.; Hayat, A.; Latif, U.; Marty, J. Electrochemical affinity biosensors based on disposable screen-printed electrodes for detection of food allergens. Sensors 2016, 16, 1863. [CrossRef] [PubMed]

6. Neethirajan, S.; Weng, X.; Tah, A.; Cordero, J.O.; Ragavan, K.V. Nano-biosensor platforms for detecting food allergens—New trends. Sens. Bio-Sens. Res. 2018, 18, 13–30. [CrossRef]

7. Syazana, A.; Minhaz, U. Electrochemical immunosensors and their recent nanomaterial-based signal amplification strategies: A review. RSC Adv. 2016, 6, 24995–25014.

8. Tiwari, J.N.; Vij, V.; Kemp, K.C.; Kim, K.S. Engineered Carbon-Nanomaterial-Based Electrochemical Sensors for Biomolecules. ACS Nano 2016, 10, 46–80. [CrossRef] [PubMed]

9. Jaiswal, N.; Tiwari, I. Recent build outs in electroanalytical biosensors based on carbon-nanomaterial modified screen printed electrode platforms. Anal. Methods 2017, 9, 3895–3907. [CrossRef]



Biosensors 2018, 8, 109



9 of 9



10. De Las Heras, E.; Ybarra, G.; Bracera, I.; Corengia, P. Surface modification by plasma-based processes. In Functional Properties of Biological Surfaces: Characterization and Technological Applications; Favret, E., Fuentes, N.O., Eds.; World Scientific Publishing Co.: Hackensack, NJ, USA, 2009.

11. Molinari, J.; Moina, C.; Ybarra, G. Electrochemical Immunosensor for the determination of β-Casein. J. Electrochem. Sci. Eng. 2015, 5, 9–16. [CrossRef]

12. Gong, B.; Ikematsu, A.; Waki, K. Impacts of structure defects and carboxyl and carbonyl functional groups on the work function of multiwalled carbon nanotubes. Carbon 2017, 114, 526–532. [CrossRef]

13. Vitkova, M.; Rauch, P.; Fukal, L. Optimisation of indirect competitive ELISA of α-, β-, and κ-caseins for the recognition of thermal and proteolytic treatment of milk and milk products. Czech J. Food Sci. 2002, 20, 53–62. [CrossRef]

14. Cortina, M.E.; Melli, L.J.; Roberti, M.; Mass, M.; Longinotti, G.; Tropea, S.; Lloret, P.; Rey Serantes, D.A.; Salomón, F.; Lloret, M.; et al. Electrochemical magnetic microbeads-based biosensor for point-of-care serodiagnosis of infectious diseases. Biosens. Bioelectron. 2016, 80, 24–33. [CrossRef] [PubMed]

15. Garate, O.; Veiga, L.; Medrano, A.V.; Longinotti, G.; Ybarra, G.; Monsalve, L.N. Waterborne carbon nanotube ink for the preparation of electrodes with applications in electrocatalysis and enzymatic biosensing. Mat. Res. Bull. 2018, 106, 137–143. [CrossRef]

16. Kurzer, F.; Douraghi-Zadeh, K. Advances in the chemistry of carbodiimides. Chemical Reviews 1967, 67, 107–152. [CrossRef] [PubMed]

17. Salomón, F.; Tropea, S.; Brengi, D.; Hernández, A.; Alamón, D.; Parra, M.; Longinotti, G.; Ybarra, G.; Lloret, P.; Mass, M.; et al. Smartphone-controlled platform for point-of-care diagnosis of infectious diseases. In Proceedings of the IEEE 9th Ibero-American Congress on Sensors, Bogotá, Colombia, 15–18 October 2014; pp. 1–4.

18. Lloret, P.; Longinotti, G.; Ybarra, G.; Socolovsky, L.; Moina, C. Synthesis, characterization and biofunctionalization of magnetic gold nanostructured particles. Mater. Res. Bull. 2013, 48, 3671–3676. [CrossRef]

19. Ashley, J.; D’Aurelio, R.; Piekarska, M.; Temblay, J.; Pleasants, M.; Trinh, L.; Rodgers, T.L.; Tothill, I.E. Development of a β-Lactoglobulin Sensor Based on SPR for Milk Allergens Detection. Biosensors 2018, 8, 32. [CrossRef] [PubMed]

20. Gómez-Arribas, L.N.; Benito-Peña, E.; Hurtado-Sánchez, M. Biosensing Based on Nanoparticles for Food Allergens Detection. Sensors 2018, 18, 1087. [CrossRef] [PubMed]

21. FDA Homepage Appendix 1: Evaluation of Commercially Available Analytical Methods for Proteins from Major Food Allergen. Available online: https://www.fda.gov/Food/GuidanceRegulation/ GuidanceDocumentsRegulatoryInformation/Allergens/ucm106577.htm (accessed on 27 September 2018).

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).