Screen-printed biosensor modified with carbon black nanoparticles for the determination of paraoxon based on the inhibition of butyrylcholinesterase

ORIGINAL PAPER

Screen-printed biosensor modified with carbon blacknanoparticles for the determination of paraoxon basedon the inhibition of butyrylcholinesterase

Fabiana Arduini & Matteo Forchielli & Aziz Amine &

Daniela Neagu & Ilaria Cacciotti & Francesca Nanni &Danila Moscone & Giuseppe Palleschi

Received: 18 June 2014 /Accepted: 16 September 2014 /Published online: 2 October 2014# Springer-Verlag Wien 2014

Abstract We have developed a screen-printed electrochemi-cal electrode (SPE) for paraoxon based on its inhibitory effecton the enzyme butyrylcholinesterase (BChE). The electrodewas first modified by drop casting with a dispersion of carbonblack nanoparticles (CBNPs) in a dimethylformamide-watermixture, and BChE was then immobilized on the surface bycross-linking. The resulting biosensor was exposed to stan-dard solutions of paraoxon, and the enzymatic hydrolysis ofbutyrylthiocholine over time was determined measuring theenzymatic product thiocholine at a working voltage of +300 mV. The enzyme inhibition is linearly related to theconcentration of paraoxon up to 30 μg L−1, and the detectionlimit is 5 μg L−1. The biosensor is stable for up to 78 days ofstorage at room temperature under dry conditions. It wasapplied to determined paraoxon in spiked waste water sam-ples. The results underpin the potential of the use of CBNPs inelectrochemical biosensors and also demonstrate that theyrepresent a viable alternative to other carbon nanomaterials

such as carbon nanotubes or graphene, and with the advantageof being very affordable.

Keywords Organophosphate . Butyrylcholinesterase .

Screen-printed electrode . Carbon black nanoparticles .

Inhibition

Introduction

Recently in the frame of an European project, a list of emerg-ing substances (Norman list) was drawn up, and the organo-phosphorus insecticides such as parathion methyl, parathionethyl are included in Norman list [http://www.norman-network.net Accessed 16 June 2014]. These insecticides, infact, are the most used due to their high insecticidal activityand relatively low persistence in the environment, thus theirdetection is an important issue in analytical chemistry. Thedetection of organophosphorus insecticides is generallycarried out using Gas or Liquid Chromatography, which arehighly sensitive and selective techniques, but require skilledpersonnel, laboratory set-up, and expensive instrumentation[1, 2]. An alternative analytical system is the use ofbiosensors, which are cost-effective, miniaturized andfriendly to use. Taking into consideration that theacetylcholinesterase (AChE) is inhibited by organophospho-rus insecticides [3], this enzyme was properly used asbiocomponent in the biosensor development; in fact, measur-ing the AChE activity before and after the biosensor exposureto environmental samples, it is possible to quantify the amountof organophosphorus insecticides present in the sample [4, 5].The biosensors which offer the best guarantee for analyticalapplications, in terms of sensibility, reproducibility, and selec-tivity, often turn out to be the electrochemical ones based ondisposable screen-printed electrodes produced by thick film

F. Arduini (*) :M. Forchielli :D. Neagu :D. Moscone :G. PalleschiDipartimento di Scienze e Tecnologie Chimiche, Università di RomaTor Vergata, Via della Ricerca Scientifica, 00133 Rome, Italye-mail: [email protected]

A. AmineFaculté de Sciences et Techniques Laboratoire Génie des Procédés etEnvironnement, Université Hassan II-Mohammedia, B.P.146 Mohammadia, Morocco

I. CacciottiUniversità degli Studi di Roma “Niccolò Cusano”, UdR INSTM, ViaDon Carlo Gnocchi 3, 00166 Rome, Italy

F. NanniDipartimento di Ingegneria dell’Impresa, Università di Roma TorVergata, UdR INSTM Roma-Tor Vergata, Via del Politecnico 1,00133 Rome, Italy

Microchim Acta (2015) 182:643–651DOI 10.1007/s00604-014-1370-y

technology [6]. In this overall scenario, an electrochemicalamperometric biosensor for organophosphates can be a bi-enzymatic one that uses acetylcholine as substrate. The twoenzymes are AChE, which hydrolyses the acetylcholine tocholine and acetic acid, and Choline Oxidase (ChOx) thatoxidises the choline to betaine with the production of H2O2.The use of ChOx is necessary in the case of amperometricbiosensors because the enzymatic products of the AChE re-action are not electroactive, and the enzymatic activity can bedetected by means of the O2 decrease quantification using aClark’s electrode [7] or by the increase of H2O2 [8]. Thealternative approach mainly used in the last years, was themonoenzymatic amperometric biosensor, in which a syntheticsubstrate is used; in fact, acetylthiocholine was adopted in-stead of the natural substrate acetylcholine. The enzymaticreaction hydrolyses the acetylthiocholine to acetic acid andthiocholine, and then the latter, being electrochemically ac-tive, can be quantified. In order to reduce the applied potentialand fouling problems during thiocholine detection, two ap-proaches can be followed: (i) the use of redox mediators suchas cobalt phthalocyanine (CoPc) [9], Prussian Blue [10],te t racyanoquinodimethane (TCNQ) [11] , cobal thexacyanoferrate [12], potassium ferricyanide [13] or (ii) theuse of nanomaterials such as carbon nanotubes [14]. Recentlyour research group demonstrated the suitability of CBNPs asuseful nanomaterial to modify SPEs in order to increase theirelectrochemical performances [15–18]; the improvementusing CBNPs was also confirmed by the Compton group thathighlighted “the potential improvement involved in the (large-ly unexplored) direct application of nano-carbon in electrodesurface modification” [19].

In this work, we investigated for the first time the suitabilityof CBNPs-SPE as platform to immobilize the butyrylcholin-esterase (BChE) in order to develop a biosensor based onBChE inhibition for organophosphate detection.

Experimental

Apparatus

Cyclic voltammetry (CV) measurements were performedusing an Autolab electrochemical system (Eco Chemie,Utrecht, The Netherlands; www.ecochemie.nl) equippedwith PGSTAT-12 and GPES software (Eco Chemie, Utrecht,The Netherlands). Amperometric measurements were carriedout using a VA 641 amperometric detector (Metrohm,Herisau, Switzerland), connected to a X-t recorder (L250E,Linseis, Selb, Germany; www.linseis.com).

Micrographs of CBNPs-SPE and biosensor based CBNPs-SPE were acquired by means of a field emission gun scanningelectron microscopy (FEG-SEM, Leo Supra 35). Electro-chemical impedance spectroscopy (EIS) measurements were

carried out in the same cell with a PC-controlled Autolab. Asinusoidal voltage perturbation of 10 mV amplitude was ap-plied over the frequency range 10 kHz to 0.01 Hz, with 10measurement points per frequency decade. For the fitting ofthe data obtained by EIS, Z-views software (Scribner Associ-ates, Inc.; www.scribner.com) was used.

Reagents

Commercial CB N220 of industrial standard grade was ob-tained from Cabot Corporation (Ravenna, Italy, www.cabot-corp.com). As reported by the manufacturer, the nanoparticlesof CB N220 had a diameter comprised between 19 and 29 nmwith a surface area of 124 m2·g−1 as measured by BETmethod (N2 absorption). Butyrylcholinesterase (BChE) fromequine serum, bovine serum albumin (BSA), S-butyrylthiocholine chloride, 5,5′-dithio-bis 2-nitrobenzoic ac-id (DTNB), glutaraldehyde and paraoxon (paraoxon-ethyl),Nafion (perfluorinated ion-exchange resin, 5 %v/v solution inlower alcohols/water) were purchased from Sigma AldrichCompany (St. Louis, USA, www.sigmaaldrich.com).

Preparation of SPE

SPE was produced with a 245 DEK (Weymouth, UK; www.dek.com) screen-printing machine. Graphite based ink(Electrodag 423 SS) from Acheson (Milan, Italy; www.achesonindustries.com) was used to print the working andcounter electrodes. Silver ink (Electrodag 477 SS) was usedto print the reference electrode. As insulating ink Carboflex25.101S was used. The substrate was a flexible polyester film(Autostat HT5) obtained from Autotype Italia (Milan, Italy).The electrodes were home produced in foils of 48. Thediameter of the working electrode was 0.3 cm resulting in ageometric area of 0.07 cm2.

Carbon black nanoparticle dispersion

The dispersion of CBNPs was prepared by adding 20 mg ofCBNPs powder to 20 mL of solvent (a mixturedimethylformamide (DMF): water (1:1)), and sonicated for60 min at 59 KHz.

Preparation of CBNPs-SPE

The SPE was modified with CBNPs via drop casting, pipet-ting a small volume (6 μL) of the dispersion onto the SPEworking electrode surface in three steps of 2 μL each. Afterthat, the solvent is allowed to volatilize at RT, and a CBNPs“film” is left on the electrode surface.

644 F. Arduini et al.

Thiocholine determination

Thiocholine was enzymatically produced by BChE usingbutyrylthiocholine as substrate (because thiocholine is notcommercially available). For this purpose, 1 mL of 1 Mbutyrylthiocholine solution was prepared in phosphate buffer0.1 M (pH=8), and 100 units of BChE were added to thissolution. After 1 h, the concentration of thiocholine producedby BChE was estimated spectrophotometrically by Ellman’smethod. For this purpose, 900 μL of phosphate buffer solution(0.1M, pH=8), 100μL of 0.1MDTNB, and 5 μL thiocholinesolution (diluted 1:100 in water) were put in a spectrophoto-metric cells. The absorbance was measured, and the realconcentration was evaluated by using the Lambert–Beer lawwith the known molar extinction coefficient of TNB (ε=13,600 M−1 cm−1) [20]. After 2 h, the butyrylthiocholinehydrolysis was completed, and 1 mL solution of 1 Mthiocholine was obtained. The solution was stable for 1 dayat 4 °C.

Preparation of BChE biosensor

In the case of cross-linking method, two steps were adopted.2 μL of a glutaraldehyde solution 0.25 %v/v (diluted in water)were applied with a syringe exclusively on the working elec-trode. Then, 2 μL of a mixture of BSA, enzyme and Nafionwere placed onto the working electrode. The mixture wasobtained by mixing 25 μL of BSA (3 %w/v prepared inwater), 25 μL of Nafion (0.1 %v/v diluted in water) and25 μL of enzyme stock solution (see scheme 1).

Butyrylthiocholine determination

Butyrylthiocholine analyses were performed using an amper-ometric “drop” procedure in phosphate buffer solution(0.05 M+0.1 M KCl, pH 7.4) with an applied potential of +300 mV vs Ag/AgCl. In details, a drop (50 μL) of buffercontaining different amounts of butyrylthiocholine was placedonto the BChE biosensor in such a way that the working,counter and reference electrodes were covered. After applyingthe potential, the signal was continuously recorded and thecurrent value at the steady state was detected.

Paraoxon determination

The inhibitory effect of paraoxon on BChE biosensor wasevaluated by determining the decrease in the current obtainedfor the oxidation of thiocholine that was produced by theenzyme. To do this, the response toward the substrate wasanalyzed as described above, after the BChE biosensor wasincubated in the insecticide solution for a certain period (in-cubation time) and then rinsed three times with distilled water.After that, the response toward the substrate was registeredand the degree of inhibition was calculated as a relative decayof the biosensor response (Equation 1).

I% ¼ I0−Iið Þ=I0½ � � 100 ð1Þ

where Io and Ii represent the biosensor response before andafter the incubation procedure, respectively

Sample collection and measurement

Two samples of waste water were supplied by Tover Italia,Rome (paint industry) and collected in different days (i.e.samples A and C). One sample of waste water was suppliedby BASF The Chemical Company (Rome, Italy) (i.e. sampleB). All the samples were tested before and after spiking withparaoxon. The samples supplied by Tover Italia were filteredbefore the analysis using Acrodisk syringe filter 37 mm andglass membrane 1μm filter. All samples were diluted 1:2 (v/v)with phosphate buffer 0.1 M+KCl 0.2 M for successiveelectrochemical analyses.

Results and discussion

Electrochemical properties of CBNPs towards thiocholine

In a previous paper, we demonstrated the ability of CBNPsusing commercial available SPE modified with CBNPs dis-persion (1 mg ·mL−1 prepared in acetonitri le) toelectrocatalyze the oxidation of several thiols such as cysteine,cysteamine, glutathione and thiocholine [16]. The high sensi-tivity reached for thiol detection was used to measure mercuryions. In this paper we have tested CBNPs-SPE prepared usingthe dispersion of CBNPs in DMF: H2O 1:1 (v/v) withthiocholine, the cholinesterase enzymatic product. The cyclicvoltammograms are shown in Fig. 1. We have observed anincrease of the oxidation current and a decrease of the oxida-tive potential; a broad peak was in fact observed at +300 mV,while in the case of the bare SPE it was placed at +700 mV.Thus, also in this case, we have demonstrated the ability ofCBNPs to electrocatalyze the thiocholine detection, and these

Scheme 1 Schematic preparation of the BChE-CBNPs-SPE biosensorand its typical response

Biosensor based on butyrylcholinesterase inhibition and for paraoxon determination 645

electrocatalytic properties were also observed when varyingthe CBNPs-SPE preparation procedure using different SPEsand dispersions, demonstrating the robustness of the sensorproduction.

The valuable electrochemical properties obtained usingCBNPs toward thiocholine oxidation are better when com-pared with other carbon nanomaterials. In the case of CNTs, infact, the applied potential is low as in the case of CBNPs, butthe detection limit is higher than the one obtained usingCBNPs [21, 22]. This behaviour is probably ascribed to thebetter signal/noise ratio in the case of CBNPs with respect toCNTs, as we previously observed in the case of otherelectroactive compounds [18]. In the case of graphene, thehigh applied potential required to detect thiocholine was re-ported in literature [23–25]. In fact, in the case of porous-reduced graphene oxide modified AChE biosensor, 0.75 Vasapplied potential was necessary to detect thiocholine [23]; asimilar potential was also found for AChE biosensor based onCdS–decorated graphene nanocomposite (0.68 V) [24] or 3-carboxyphenylboronic acid/reduced graphene oxide–goldnanocomposites (0.7 V) [25]. The reason for the very goodelectrochemical performances of CBNPs can be ascribed totheir high number of defect sites [15], which lead to detect thethiocholine at low applied potential, with high sensitivitywithout fouling problem. Furthermore, these results suggestthat the CBNPs used in this work are competitive not onlywith CNTs [18] but also with graphene.

Taking into consideration the results obtained usingCBNPs-SPE in the enzymatic product thiocholine detection,this sensor was then used to immobilize the butyrylcholines-terase enzyme in order to produce a novel BChE biosensor fororganophosphate detection.

BChE biosensor

Very often in literature we found biosensors characterised by avery low detection limit but coupled with a satisfactory shelf

stability, obtained however only at low temperature, a charac-teristic that hampers a possible commercialisation of the bio-sensor [14, 26–31]. In order to construct a biosensorcharacterised by both high operating and shelf stability atroom temperature, we have recently performed a study inwhich several different immobilisations procedures were in-vestigated, demonstrating that the cross-linked BChE ischaracterised by high shelf stability using the SPE modifiedwith electrochemical mediator Prussian Blue (PB-SPE)[32–34]. In this work, we tested the immobilisation of BChEby means of glutaraldehyde, Nafion and BSA on CBNPs-modified SPEs. Glutaraldehyde was necessary to link theenzyme, and Nafion was useful for the adherence of theenzymatic membrane on the CBPNs. Furthermore, the addi-tion of the BSA in the enzymatic membrane was necessary toimprove the enzyme shelf-life. To achieve satisfactory analyt-ical performances, it is important that i) few enzymatic unitsare immobilised on the surface of the electrode, because in thecase of irreversible inhibition, as for BChE inhibition byorganophosphate insecticides, the degree of inhibition in-creases at enzyme units decrease, ii) however, it is also im-portant to have a sufficient number of immobilised units, inorder to have a thiocholine production adequate to be quanti-fied. In the Fig. 2 the behaviour when increasing the enzy-matic units is showed. As expected, we have observed theincrease of current due to the increase of enzymaticthiocholine production and the decrease of degree of inhibi-tion due to the fact that it is an irreversible inhibition type;thus, 26 mU were chosen as a compromise (Fig. 2a). More-over, keeping in mind that it is an irreversible inhibition, theconcentration of substrate selected for inhibition measure-ments should be suitable to reach the Vmax; thus the biosen-sor response in function of enzymatic substrate amount wasevaluated. Once optimized, the biosensor was then testedtowards butyrylthiocholine, the enzymatic substrate that wasanalysed in the range comprised between 1×10−5 M and 1×10−2 M, observing a Michaelis Menten behaviour with aKMapp = (2.2 ± 0.3) mM (Fig. 2b) . The minimumbutyrylthiocholine concentration that gives Vmax was 5 mM,and it was chosen for the insecticide determination; in addi-tion, the repeatability of the biosensor was tested at this

E vs Ag/AgCl (V)0,0 0,2 0,4 0,6 0,8 1,0

i(A)

0

2e-6

4e-6

6e-6

8e-6

Fig. 1 CVs in phosphate buffer 0.05M+KCl 0.1M, pH=7.4, in absenceof 1 mM thiocholine using bare (continuous lines) and CBNPs-SPE(dotted lines) and in presence of 1 mM thiocholine using bare (dashedlines) and CBNPs-SPE (dashed-dotted lines)

Enzymatic units (mU)10 20 30 40 50 60In

hib

itio

n d

egre

e (%

)

0

20

40

60

80

0123456

[butyrylthiocholine] M0,000 0,004 0,008 0,012

0,0

0,5

1,0

1,5

2,0

2,5

Fig. 2 a Biosensor response towards 5 mM butyrylthiocholine and20 μg L−1 paraoxon (as inhibition degree) in function of enzymatic units.b Calibration plot of butyrylthiocholine. Applied potential: +300 mV vsAg/AgCl, 0.05 M phosphate buffer+0.1 M KCl, pH 7.4

646 F. Arduini et al.

butyrylthiocholine level. The value obtained for six successiveanalyses was 1.56±0.06μAwith a RSD% equal to 3.9 %. Thesatisfactory repeatability of the biosensor is not only a goodanalytical property, but also shows the possibility of using thesame biosensor several times without fouling problems that,instead, occur in the case of bare SPE. This behaviour is due tothe presence of CBNPs.

BChE biosensor characterization by means of electrochemicalimpedance spectroscopy and scanning electron microscopy

Electrochemical Impedance Spectroscopy can provide usefulinformation on the impedance changes of the electrode surfaceduring the fabrication process of biosensors, by measuring thevalue of electron transfer resistance (Rct). The Rct, estimatedaccording to the diameter of the semicircle present at the highfrequency region, represents, in fact, the difficulty of electrontransfer of ferro/ferricyanide redox probe between the solutionand the electrode, giving information about the change of theelectrode surface. The electrochemical impedance spectrosco-py was performed with CBNPs-SPE in absence and presenceof the enzymatic membrane (BChE-CBNPs-SPE) at opencircuit potential (OCP). Fitting of spectra was done using theequivalent electrical circuit showed in Fig. 3 (inset) whichcomprises the electrolyte resistance, Re (around 150 Ω), inseries with a parallel combination of Rct (interfacial chargetransfer resistance), Zw (diffusion of the analytes in solutionand corresponding to Warburg impedance straight line of thecurves) and CPE (Constant Phase Element). Fig. 3 shows theNyquist plots for CBNPs-SPE and BChE-CBNPs-SPE. TheRct for BChE-CBNPs-SPE was much higher (2842±89Ω)than the CBNPs-SPE (229±4Ω) confirming the depositionof the enzymatic layer that hampers the electron-transfer of theelectrochemical probe (ferro/ferricyanide). The constant phaseelement determination, CPE, was necessary due to the non-homogeneous surface of the working electrode and it ismodeled as a non-ideal capacitor of capacitance C androughness/non-uniformity factor α. The α resulted 0.70 forthe CBNPs-SPE and 0.58 for BChE-CBNPs-SPE,

demonstrating the increase of the electrode surface roughnessin the presence of the enzymatic membrane. These experi-mental evidences were supported and confirmed by the SEManalysis that was performed in order to investigate the mor-phological characteristics of the bare SPE and the CBNPs-SPE before and after the enzyme immobilization (BChE-CBNPs-SPE). In fact, it is well known that the morphologyof the substrate plays a pivotal role in the immobilization ofthe enzyme and strongly influences the biosensorperformance.

In Fig. 4 low and high magnification micrographs of bareSPE, CBNPs-SPE and BChE-CBNPs-SPE are compared. Inall cases, the low magnification images testified the obtain-ment of a complete and uniform deposition on the workingelectrode (Fig. 4a, c and e).

Considering the high magnification micrographs, the bareSPE showed a webbed surface with irregularly shaped andrandomly orientated micrometer-sized flakes of graphitebound together with an inert polymeric binder and coveredof small particles assigned to the cross-linking agents in theoriginal ink (Fig. 4a and b). On the other hand, the SEMmicrographs of the CBNPs-modified working electrode con-firmed the homogeneous and uniform CBNPs deposition andrevealed a rough and sponge-like structure, characterized bythe presence of numerous and diffuse cauliflower aggregatesof CBNPs (Fig. 4c and d). From the high magnificationmicrographs (Fig. 4d, inset) it is clear that the CBNPs film

Fig. 3 Complex plane impedance plots at an open circuit potential forCBNPs-SPE a and BChE-CBNPs-SPE b using a 10mM ferricyanide and10 mM ferrocyanide solution in 0.1 M KCl. Inset: Randles circuit

Fig. 4 SEM micrographs of bare SPE a-b, CBNPs-SPE c-d and BChE-CBNPs-SPE e-f (insets: high magnification SEM micrographs)

Biosensor based on butyrylcholinesterase inhibition and for paraoxon determination 647

Tab

le1

Different

electrochemicalbiosensorsforparaoxon

detection

Type

ofbiosensor

Linearrange

LOD

Electrochem

icaldetection

mode

Incubatio

ntim

eStorage

stability

References

Temperature

Tim

e

Biosensorsbasedon

enzymeinhibitio

n(Paraoxonistheenzymeinhibitor)

GC/GO-N

TA-His6-taggedAChE

10−9–10−

5M

6.510

−10M

Amperometry

at+0.2V

10min

4°C

97%

oftheenzymeactiv

ityover

4weekstoragein

0.01

MPB

S[22]

SPE/Ni-NiO

nanoparticles

His6-taggedAChE

10−1

3–10−

8M

10−1

2M

Amperometry

at+0.1V

20min

4°C

–[23]

SPE/SWCNTs-CoP

c-AChE

5–50

μgL−1

3μgL−1

(≅110

−8M)

Chrono-am

perometry

at+0.05

V15

min

4°C

Atleast3months

[24]

SPE/Cyst-Glut-AChE

5–20

μgL−1

2μgL−1

(≅710

−9M)

Amperometry

at+0.3Vusing

ferricyanide

insolutio

n15

min

4°C

Around60

%ftheinitialactiv

ityafter

1weekin

buffer

[13]

SPE/M

WCNTs-A

ChE

Upto

6.9×

10−9

M510

−10M

Amperometry

at+0.2V

30min

4°C

The

response

was

stablefor7days

[14]

PtSPE

/AChE

-gelatine

2.5–10

μgL−1

2.5μgL−1

(≅910

−9M)

Amperometry

at+0.41

V15

min

––

[25]

SPE/CBNPs-G

lut-Nf-BSA-BChE

5–30

μgL−1

5μgL−1

(≅210

−8M)

Amperometry

at+0.3V

20min

RT

Atleast80

days

indrycondition

Thiswork

Substratebiosensors

(Paraoxonistheenzymesubstrate)

GC/M

C-CBNPs-OPH

Upto

8×10

−6M

1.210

−7M

Amperometry

at+0.9V

–4°C

–[32]

GC/M

C-O

PH-bacteria

0.05–25×10

−6M

910

−9M

Amperometry

at+0.84

V–

4°C

Duringone-month

testthecurrentsignals

still

remained70

%of

initialresponse

[33]

GCglasscarbon

electrode,GOgraphene

oxide,NTA

nitrilo

aceticacid,A

ChE

acetylcholinesterase,P

BSphosphatebuffer

solutio

n,SP

Escreen-printed

electrode,SW

CNTs

singlewallcarbonnanotubes,

CoP

cCobalt-Phtalocyanine,C

ystcysteam

ine,Glutglutaraldehyde,MWCNTs

multiw

allcarbon-nanotubes,PtSPESP

Ewith

platinum

asworking

electrode,CBNPscarbon

blacknanoparticles,NfN

afion,

BSA

,allb

uminebovine

serum,B

ChE

,butylcholinesterase,M

Ccarbon

mesoporous,OPHorganophosphorus

hydrolase

648 F. Arduini et al.

completely and uniformly covered the SPE surface, being notpossible, in fact, to observe the presence of the graphiteplatelets.

Significant morphology differences of the CBNPs-SPEworking electrode before and after enzyme immobilizationwere detected, revealing the BChE-CBNPs-SPE a superficialuniform film (Fig. e and f), testifying the occurred enzymeimmobilization.

Paraoxon determination

Our goal was the insecticide detection in waste water sample,thus in a complex matrix. In order to fine tune an analyticalsystem capable to work well in waste water samples, withoutsophisticated sample pre-treatment and interference problem,the “medium exchange method” proposed by us in an ourprevious workwas adopted [34]. This method consists in threesteps; briefly, in the first step the enzymatic activity wasmeasured in buffer solution in presence of the only enzymaticsubstrate. After, in the second step, the biosensor was put incontact with the sample contaminated with insecticides for aselected time, followed by the several times rinsing of thebiosensor with distilled water. In the last step, the enzymaticresidual activity was finally determined in a new buffer aliquotin presence of the only enzymatic substrate. In this way, it waspossible to avoid electrochemical interferences such as ascor-bic acid, phenolic compounds, etc., since the enzymatic activ-ity was always quantified in phosphate buffer in absence ofany electroactive interfering species. Furthermore, washingthe biosensor with distillate water after the inhibition step,only irreversible inhibitors (e.g. organophosphate) able to linkthe enzyme by covalent bonk can be detected. In fact, theother types of inhibitor that could be present in waste watersamples such as Cd2+, Cu2+, Fe3+ (reversible inhibitors) wereavoided.

In order to obtain a sensitive measurement, the incubationtime was optimized. The degree of inhibition increases withthe incubation time, due to the fact that the cholinesteraseinhibition by organophosphate is an irreversible inhibition.In our case we chose 20 min for the incubation time ascompromise between a sensitive measurement and no tediousanalysis time (data not shown). Under optimized parameters,the biosensor was challenged with paraoxon, obtaining alinear range up to 30μg L−1 with a calibration curve describedby the following equation: y=(2.1±0.1)×+(3.6±1.8), R2=0.971. 5 μg L−1 was the detection limit, calculated as theamount of analyte that gives 10 % of inhibition. The devel-oped biosensor allows the paraoxon detection at ppb levelsand at low applied potential when compared with biosensorsdeveloped using carbon black and/or mesoporous carbon andorganophosphorus hydrolase enzyme [35, 36] (Table 1). The-se results demonstrate for the first time the possibility to applythe CBNPs in biosensors based on butyrylcholinesterase

inhibition. The use of CBNPs allows i) to use low appliedpotential, ii) to easily prepare a stable dispersion for mass-produced modified sensors (e.g. using BioDot automatizedlow volume dispensing equipment, www.biodot.com) and iii)to employ a cost effective nanomaterial. Furthermore, thisminiaturized sensor can be adaptable for integration inmicrofluidic platform [33]. The biosensor here proposed canbe competitive with the ones integrated in the microfluidicplatforms reported in literature [37, 38] because it employs aneasy drop casting CBNPs modification procedure. On thecontrary, the use of electrochemical mediators, such asCoPc, requires necessarily its incorporation in the ink beforethe SPE printing.

Paraoxon determination in waste water samples

In order to check the suitability of the developed biosensor inreal samples, firstly it was applied for sensing in drinkingwater samples collected in our Department. In this case noinhibition was observed, demonstrating the absence of insec-ticide at 10 μg L−1 level. In order to evaluate the accuracy ofthe biosensor, the sample was fortified with 50 μg L−1 ofparaoxon and diluted 1:2 (v/v)in phosphate buffer, obtaininga recovery value of (96±2) % demonstrating the accuracy ofthis biosensor in a drinking water matrix. In order to challengethe biosensor in the waste water samples, three different

Fig. 5 Photograph of the analysed samples

Days

1 8 15 22 36 50 78

Deg

ree

of

inh

ibit

ion

(%

)

0102030405060

Days

1 8 15 22 36 50 78

Res

idu

al a

ctiv

ity

(%)

0

20

40

60

80

100 a b

(a) (b)

Fig. 6 Storage stability evaluated by the residual activity a and by thedegree of inhibition b using paraoxon 20 μg L−1. Applied potential: +300 mV vs Ag/AgCl, 0.05 M phosphate buffer+0.1 M KCl, pH 7.4,5 mM butyrylthiocholine as substrate, 20 min as incubation time (n=3)

Biosensor based on butyrylcholinesterase inhibition and for paraoxon determination 649

samples were analysed (Fig. 5); two of them were collectedfrom a paint industry (sample A and C) and the other one(sample B) from a catalyst industry. In each case no inhibitionwas observed, thus the samples were fortified with 50 μg L−1

of paraoxon, a level lower than the legal limit of 100 μg L−1

for surface waste waters [39]. For the samples A, B and C therecovery values were (88±6) %, (96±9) % and (76±9) %,respectively, confirming the satisfactory accuracy of the bio-sensor and the capability to work even in a more complexmatrix.

Storage stability of biosensor and enzymatic substrate

The storage stability is a key point for the commercializationof biosensors, probably the reason of the relevant gap betweenthe research sector and the market. In order to test the practi-cability of the developed biosensor, the storage stability wastested. When the biosensor was not in use, it was stored at RTin dry conditions. The CBNPs-SPE sensor is stable at RT indry conditions for at least 100 days, as demonstrated in ourprevious work [17]. In the case of biosensors, usually the lowstorage stability is mainly due to the biocomponent; however,by using the enzymatic membrane here described, the shelflife at RT in dry conditions is rather satisfactory (Fig. 6a),highlighting that the storage condition at 4 °C is not necessary,as in many papers present in the Literature [14, 26–31], and ashighlighted in the Table 1. Until now, we have observed thesame degree of inhibition for biosensors stored at RT in dryconditions within 78 days, with RSD% not higher of 6 % forall the analyses of inhibition performed (Fig. 6b). Also wewant to underline that the inhibition measurements shown inFig. 6b, did not belong to the same electrode, but to severalbiosensors stored at RT and tested at different times. Eachelectrode is used for just one inhibition measurement. Indeed,after each inhibition experiment, the biosensor lost about 50%of its original activity. After few experiments done with thesame electrode, the activity falls to zero, thus a single use ofsuch biosensor is highly recommended.

Concerning the substrate, it is better to maintain at 4 °C thebutyrylthiocholine solution rather than the butyrylthiocholinepowder, because this latter is highly hygroscopic. At 4 °C it ispossible to maintain a 5 mM solution of the enzymatic sub-strate in working buffer for no more than 2 weeks; on thecontrary, the powder should be stocked in Eppendorf vials inweighting amounts after flowing nitrogen for 1 min, to attain astorage stability for at least of 6 months at RT [32].

Conclusions

We have demonstrated that the use of BChE cross-linked onSPE modified with carbon black nanoparticles showed a

detection limit of 5 μg L−1 with good storage stability at RTin dry conditions, making the so assembled biosensor com-petitive with the ones reported in Literature. The use of carbonblack nanoparticles avoids the fouling problem during thethiocholine electrochemical detection, thanks to the high num-ber of defect sites present in carbon black nanoparticles. Theproposed biosensor may be very useful as alarm system in thecase of neurotoxic agents (i.e. organophosphate, carbamate).Furthermore, the disposability of this biosensor is extremelyuseful in the case of organophosphorous insecticides quanti-fication since the reactivation of the enzyme after its irrevers-ible inhibition requires incubation time with specificreactivator.

The developed biosensor based on carbon black nanopar-ticles and BChE inhibition allowed in situ paraoxon detectioncharacterised by low detection limit and/or absence of elec-trochemical interferences due to the use of low applied poten-tial and the “medium exchange method” even in a complexmatrix like the waste water samples.

Acknowledgments This work was supported by National Industria2015 (MI01_00223) ACQUA-SENSE project and Marie Curie FP7-PEOPLE-2011-IRSES, 294901 “Peptide Nanosensors”. The authorsthank Prof. F. Cataldo (Actinium Chemical Research srl) for the CBNPssamples, Tover Italia s.r.l. (Rome) and BASF Italia DivisioneCatalizzatori (Rome) for the waste water samples.

References

1. Bergh C, Torgrip R, Ostman C (2010) Simultaneous selective detec-tion of organophosphate and phthalate esters using gas chromatogra-phy with positive ion chemical ionization tandem mass spectrometryand its application to indoor air and dust. Rapid Commun MassSpectrom 24:2859–2867

2. Freitas Ventura F, Oliveira J, Pedreira Filho WDR, Gerardo RibeiroM (2012) GC-MS quantification of organophosphorous pesticidesextracted from XAD-2 sorbent tube and foam patch matrices. AnalMethods 4:3666–3673

3. Rosenfeld CA, Sultatos LG (2006) Concentration-dependent kineticsof acetylcholinesterase inhibition by the organophosphate paraoxon.Toxicol Sci 90:460–469

4. Andreescu S, Marty JL (2006) Twenty years research in cholinester-ase biosensors: From basic research to practical applications. BiomolEng 23:1–15

5. Arduini F, Amine A, Moscone D, Palleschi G (2010) Biosensorsbased on cholinesterase inhibition for insecticides, nerve agents andaflatoxin B1 detection (review). Microchim Acta 170:193–214

6. Taleat Z, Khoshroo A, Mazloum-Ardakani M (2014) Screen-printedelectrodes for biosensing: a review (2008–2013) Microchim Acta, inpress, doi: 10.1007/s00604-014-1181-1

7. Fennouh S, Casimiri V, Burstein C (1997) Increased paraoxon detec-tion with solvents using acetylcholinesterase inactivation measuredwith a choline oxidase biosensor. Biosens Bioelectron 12:97–104

8. Cremisini C, Di Sario S, Mela J, Pilloton R, Palleschi G (1995)Evaluation of the use of free and immobilised acetylcholinesterasefor paraoxon detection with an amperometric choline oxidase basedbiosensor. Anal Chim Acta 311:273–280

650 F. Arduini et al.

9. Hart JP, Hartley IC (1994) Voltammetric and amperometric studies ofthiocholine at screen-printed carbon electrode chemically modifiedwith cobalt phthalocyanine: studies towards a pesticide sensor.Analyst 119:259–263

10. Ricci F, Arduini F, Amine A, Moscone D, Palleschi G (2004)Characterisation of Prussian blue modified screen-printed electrodesfor thiol detection. J Electroanal Chem 563:229–237

11. Montesinos T, Pérez-Munguia S, Valdez F, Marty JL (2001)Disposable cholinesterase biosensor for the detection of pesticidesin water miscible organic solvents. Anal Chim Acta 431:231–237

12. Arduini F, Cassisi A, Amine A, Ricci F, Moscone D, Palleschi G(2009) Electrocatalytic oxidation of thiocholine at chemically modi-fied cobalt hexacyanoferrate screen-printed electrodes. J ElectroanalChem 626:66–74

13. Arduini F, Guidone S, Amine A, Palleschi G, Moscone D (2013)Acetylcholinesterase biosensor based on self-assembled monolayer-modified gold-screen printed electrodes for organophosphorus insec-ticide detection. Sens Actuat B 179:201–208

14. Joshi KA, Tang J, Haddon R, Wang J, Chen W, Mulchandani A(2005) A disposable biosensor for organophosphorus nerve agentsbased on carbon nanotubes modified thick film strip electrode.Electroanal 17:54–58

15. Arduini F, Amine A, Majorani C, Di Giorgio F, De Felicis D, CataldoC, Moscone D, Palleschi G (2010) High performance electrochemi-cal sensor based on modified screen-printed electrodes with cost-effective dispersion of nanostructured carbon black. ElectrochemCommun 12:346–350

16. Arduini F, Majorani C, Amine A, Moscone D, Palleschi G (2011)Hg2+ detection by measuring thiol groups with a highly sensitivescreen-printed electrodemodified with a nanostructured carbon blackfilm. Electrochim Acta 56:4209–4215

17. Arduini F, Di Nardo F, Amine A, Micheli L, Palleschi G, Moscone D(2012) Carbon black-modified screen-printed electrodes as electro-analytical tools. Electroanal 24:743–751

18. Suprun E, Arduini F, Moscone D, Palleschi G, Shumyantseva VV,Archakov AI (2012) Direct electrochemistry of Heme proteins onelectrodes modified with didodecyldimethyl ammonium bromideand carbon black. Electroanal 24:1923–1931

19. Lo TWB,Aldousa L, ComptonRG (2012) The use of nano-carbon asan alternative to multi-walled carbon nanotubes in modified elec-trodes for adsorptive stripping voltammetry. Sens Actuat B 162:361–368

20. Ellman GL (1958) A colorimetric method for determining low con-centrations of mercaptans. Arch Biochem Biophys 74:443–445

21. Liu G, Riechers SL, Mellen MC, Lin Y (2005) Sensitive electro-chemical detection of enzymatically generated thiocholine at carbonnanotube modified glassy carbon electrode. Electrochem Commun7(11):1163–1169

22. Liu G, Lin Y (2006) Biosensor based on self-assembling acetylcho-linesterase on carbon nanotubes for flow injection/amperometricdetection of organophosphate pesticides and nerve agents. AnalChem 78(3):835–843

23. Li Y, Bai Y, Han G, Li M (2013) Porous-reduced graphene oxide forfabricating an amperometric acetylcholinesterase biosensor. SensActuat B 185:706–712

24. Wang K, Liu Q, Dai L, Yan J, Ju C, Qiu B, Wu X (2011) A highlysensitive and rapid organophosphate biosensor based on enhance-ment of CdS–decorated graphene nanocomposite. Anal Chim Acta695:84–88

25. Liu T, Su H, Qu X, Ju P, Cui L, Ai S (2011) Acetylcholinesterasebiosensor based on 3-carboxyphenylboronic acid/reduced grapheneoxide–gold nanocomposites modified electrode for amperometricdetection of organophosphorus and carbamate pesticides. SensActuat B 160:1255–126

26. Du D, Chen S, Song D, Li H, Chen X (2008) Development ofacetylcholinesterase biosensor based on CdTe quantum dots/goldnanoparticles modified chitosan microspheres interface. BiosensBioelectron 24:475–479

27. Zhang H, Li ZF, Snyder A, Xie J, Stanciu LA (2014) Functionalizedgraphene oxide for the fabrication of paraoxon biosensors. AnalChim Acta 827:86–94

28. Ganesana M, Istarnboulie G, Marty JL, Noguer T, Andreescu S(2011) Site-specific immobilization of a (His) 6-tagged acetylcholin-esterase on nickel nanoparticles for highly sensitive toxicity biosen-sors. Biosens Bioelectron 30:43–48

29. Ivanov AN, Younusov RR, Evtugyn GA, Arduini F, Moscone D,Palleschi G (2011) Acetylcholinesterase biosensor based on single-walled carbon nanotubes-Co phtalocyanine for organophosphoruspesticides detection. Talanta 85:216–221

30. Pohanka M, Jun D, Kuca K (2008) Amperometric biosensor for realtime assay of organophosphates. Sensors 8:5303–5312

31. Zamfir LG, Rotariu L, Bala C (2011) A novel, sensitive, reusable andlow potential acetylcholinesterase biosensor for chlorpyrifos basedon 1-butyl-3-methylimidazolium tetrafluoroborate/multiwalled car-bon nanotubes gel. Biosens Bioelectron 26:3692–3695

32. Arduini F, Palleschi G (2012) Disposable Electrochemical BiosensorBased on Cholinesterase Inhibition with Improved Shelf-Life andWorking Stability for Nerve Agent Detection. In: Nikolelis DP (ed)NATO science for peace and security series A: chemistry and biologyportable chemical sensors weapons against bioterrorism. Springer,The Netherlands, pp 261–278

33. Arduini F, Neagu D, Dall’Oglio S, Moscone D, Palleschi G (2012)Towards a portable prototype based on electrochemical cholinester-ase biosensor to be assembled to soldier overall for nerve agentdetection. Electroanalysis 24:581–590

34. Arduini F, Ricci F, Tuta CS, Moscone D, Amine A, Palleschi G(2006) Detection of carbamic and organophosphorus pesticides inwater samples using a cholinesterase biosensor based on PrussianBlue-modified screen-printed electrode. Anal Chim Acta 580:155–162

35. Lee JH, Park JY, Min K, Cha HJ, Choi SS, Yoo YJ (2010) A novelorganophosphorus hydrolase-based biosensor using mesoporous car-bons and carbon black for the detection of organophosphate nerveagents. Biosens Bioelectron 25:1566–1570

36. Tang X, Zhang T, Liang B, Han D, Zeng L, Zheng C, Li T, Wei M,Liu A (2014) A. Sensitive electrochemical microbial biosensor for p-nitrophenylorganophosphates based on electrode modified with cellsurface-displayed organophosphorus hydrolase and orderedmesopore carbons. Biosens Bioelectron 60:137–142

37. Crew A, Lonsdale D, Byrd N, Pittson R, Hart JP (2011) A screen-printed, amperometric biosensor array incorporated into a novelautomated system for the simultaneous determination of organophos-phate pesticides. Biosens Bioelectron 26:2847–2851

38. Tan HY, Loke WK, Nguyen NT, Tan SN, Tay NB, Wang W, Ng SH(2014) Lab-on-a-chip for rapid electrochemical detection of nerveagent Sarin. Biomed Microdevices 16(2):269–275

39. DL 152/2006, 3 April 2006 “Norme in materia ambientale”, http://www.gazzettaufficiale.it/eli/id/2006/04/14/006G0171/sg

Biosensor based on butyrylcholinesterase inhibition and for paraoxon determination 651