Enzymatic determination of BPA by means of tyrosinase immobilized on different carbon carriers

Page 1

A

Voru

ws

cm©

K

1

aocaCe

Pf

0d

Biosensors and Bioelectronics 23 (2007) 60–65

Enzymatic determination of BPA by means of tyrosinaseimmobilized on different carbon carriers

D.G. Mita a,b,c,∗, A. Attanasio a,c, F. Arduini d, N. Diano a,b, V. Grano a,c,U. Bencivenga b, S. Rossi b, A. Amine e, D. Moscone c,d

a Department of Experimental Medicine, Faculty of Medicine and Surgery, Second University of Naples,Via S. Maria di Costantinopoli, 16 Naples, Italy

b Institute of Genetics and Biophysics of CNR, Via Pietro Castellino, 111, 80131 Naples, Italyc Consorzio Interuniversitario Biostrutture e Biosistemi “INBB”, Viale Medaglie d’Oro, 305 Roma, Italy

d Department of Chemical Sciences and Technologies, University of Tor Vergata, Roma, Italye Faculte de Sciences et Techniques, Universite Hassan II-Mahammedia, 20650 Mohammadia, Morocco

Received 9 October 2006; received in revised form 6 February 2007; accepted 19 March 2007Available online 25 March 2007

bstract

Different tyrosinase carbon paste modified electrodes to determine bisphenol A (BPA) concentration in aqueous solutions have been constructed.ariables examined were in the carbon paste composition and in particular: (i) the immobilized enzyme amount; (ii) the carbon type (powder, singler multi-walled nanotubes); (iii) the nature of the pasting oil (mineral oil, hexadecane and dodecane). For each biosensor type the amperometricesponse was evaluated with reference to the linear range and sensitivity. Constant reference has been made to the amperometric signals obtained,nder the same experimental conditions, towards the catechol, a specific phenolic substrate for tyrosinase.

The most efficient biosensors were those constructed by using the following composition for the carbon paste: 10% of tyrosinase, 45% of singleall carbon nanotubes (SWCN) and 45% of mineral oil. This biosensor formulation displayed the following electrochemical characteristics: a

ensitivity equal to 138 �A/mM, LOD of 0.02 �M (based on three times the S/N ratio), linear range of 0.1–12 �M and response time of 6 min.

This experimental work represents a first attempt at construction of a new carbon nanotube-tyrosinase based biosensor able to determine the

oncentration of BPA, one of the most ubiquitous and hazardous endocrine disruptors which can pollute the drinking and surface water, as well asany products of the food chain.2007 Elsevier B.V. All rights reserved.

ctrod

lpirp

eywords: Tyrosinase; Carbon nanotubes; Bisphenol A; BPA; Carbon paste ele

. Introduction

Scientific and public attention has been recently focused onnew class of environmental pollutants able to mimic or antag-nize the effects of endogenous hormones. For this reason thesehemicals have been called “endocrine disruptors” (Birnbaum

nd Fenton, 2003; Sonnenschein and Soto, 1998; Jimenez, 1997;olborn et al., 1996; Colborn et al., 1993) and their adverseffects on human and wildlife have been well documented. The

∗ Corresponding author at: Institute of Genetics and Biophysics of CNR, Viaietro Castellino, 111, 80131 Naples, Italy. Tel.: +39 0816132608;ax: +39 0816132608.

E-mail address: [email protected] (D.G. Mita).

arp(ttKeq

956-5663/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.bios.2007.03.010

es; Biosensors

ist of these chemicals is extremely large and includes phthalates,lastifiers, surfactants, polychlorinated biphenyls (PCBs), diox-ns, alkyphenols (APs), bisphenol A (BPA), brominate flameetardants, polycyclic aromatic hydrocarbons (PAHs) and someesticides.

To prevent the noxious effects of the endocrine disruptors,n efficient monitoring system is required, so that immediateemediation can be activated. Analytical methods, such as highressure liquid chromatography (HPLC), gas chromatographyGC), or gas chromatography coupled with mass spectrome-ry (GC–MS), are currently employed to detect and determine

hese chemicals, in particular BPA (Chang et al., 2005; Kang andondo, 2003; Goodson et al., 2002; Roger et al., 2002; Yoshidat al., 2001; Pedersen and Lindholst, 1999). However, they areuite expensive, time-consuming, need skilled operators, require

Page 2

nd Bi

l2opo

d

oimpa1etha2

bfd(((apaso

rDttpmh1phdesbubpoadNa

ao

AbcatTbow

2

2

lahtfiapt

ow

itff

2

dacfrattfite

6cGepc

D.G. Mita et al. / Biosensors a

aborious preconcentration and extraction steps (Berkner et al.,004; Pedersen and Lindholst, 1999; Roger et al., 2002). More-ver, the apparatuses are normally located far away from theossibly polluted sites. These difficulties have been recentlyvercome with the advent of sensor and biosensor technologies.

In this paper, we propose amperometric biosensors able toetermine the concentration of BPA.

BPA is a ubiquitous substance used mainly in the productionf epoxy resins and polycarbonate plastics. Thus, BPA is presentn much food and drink packaging, as well as in lacquers coating

etal products, such as food cans or water supply pipes. Someolymers used in dental treatment also contain BPA. Estrogenicctivity of BPA has been known since 1938 (Dodds and Lawson,938). Later on (Krishan et al., 1993), BPA was found to bestrogenic in MCF-7 human breast cancer cell culture at concen-rations as lower as 2–7 ppb. Since that time, numerous studiesave been reported on the estrogenic effects of BPA on humansnd animals (Markey et al., 2005; Mueller, 2004; Ishihara et al.,003; Singleton and Khan, 2003; Brzozowski et al., 1997).

Because of the acute toxicity of BPA, some biosensors haveeen developed as well. The majority of them use antibodiesor the realization of immunosensors, and the signal trans-uction is achieved through surface plasmon resonance (SPR)Marchesini et al., 2005), total internal reflection fluorescenceTIRF) (Rodriguez-Mozaz et al., 2005), or piezoelectric effectPark et al., 2006). Very recently the SPR technique has beenpplied to the measurement of BPA through the use of transportroteins (Marchesini et al., 2006). However, all these biosensorsre quite complicated to construct and control, and require alsopecific antibodies from killed animals or particular proteinsbtained by recombinant techniques.

Finally, some amperometric enzymatic biosensors have beenealised for the detection of BPA (Andreescu and Sadik, 2004;empsey et al., 2004; Notsu et al., 2002). All of them are elec-

rochemical biosensors, because this type of biosensors presentshe advantage of intrinsic selectivity, high sensitivity, low cost,otential for miniaturization and the possibility of direct deter-ination in situ. The performance of electrochemical sensors

as been greatly enhanced by the discovery in 1991 (Iijima,991) of carbon nanotubes (NT) owing to their electro catalyticroperties in the redox behaviour of different molecules such asydrogen peroxide, NADH, ascorbic acid, uric acid, dopamine,opac (Rubianes and Rivas, 2003; Wang et al., 2003; Musameht al., 2002). Nanotubes were also utilised to develop biosen-ors for glucose, lactate, catechols, alcohols and also phenolsy Rubianes and Rivas (2005). These latter authors, in partic-lar, investigated the performance of amperometric enzymaticiosensors based on the enzyme polyphenol oxidase into NTsaste, determining phenol and catechol with an improvementf sensitivity in respect to carbon paste based biosensors. Anmperometric biosensor for determination of phenol was alsoeveloped by Zhao et al. (2005) by immobilising tyrosinase onTs modified glassy carbon. However, in none of the papers

bove reported nanotubes were used for BPA detection.The biosensors proposed in this paper have been constructed

fter carefully studying the type of enzyme (tyrosinase, laccaser peroxidase) to be entrapped in a carbon paste electrode (CPE).

ea−s

oelectronics 23 (2007) 60–65 61

lso, the type of carbonaceous material, such as graphite car-on powder (GCP), single-wall (SWNT) or multi-wall (MWNT)arbon nanotubes, and of the organic binder have been evaluatednd the best combination determined. Finally, we will discusshe reasons for the choice of tyrosinase for the experimentation.he performances of each type of biosensor proposed here wille discussed with respect to the extension of the linear rangef the electrochemical response and the sensitivity. Comparisonill be also done with other biosensors for BPA determination.

. Materials and methods

.1. Materials

Tyrosinase (E.C 1.14.18.1, 3566 Units/mg) from Mushroom,accase (E.C.1.10.3.2, 20 Units/mg) from Trametes versicolor,nd peroxidase (E.C. 1.11.1.7, 250 Units/mg) from Horseradishave been used as oxidoreductase enzymes able to give an elec-rochemical signal when used as bioelements in a biosensor. Thenal choice of the enzyme to be employed in this research wasdopted on the basis of the best electrical signal produced in theresence of BPA relative to that measured with an equal concen-ration of catechol, a specific substrate for the three enzymes.

Carbon powder (1.2 �m in size), single wall carbon nan-tubes (SWNT) and multi-wall carbon nanotubes (MWCN)ere purchased from Sigma–Aldrich.Mineral oil, hexadecane and dodecane (employed as past-

ng materials), BPA and catechol (used as substrates), and allhe other chemicals, of pure grade pro-analysis, were purchasedrom Sigma (Sigma–Aldrich, Milan, Italy) and used withouturther purification.

.2. Methods

Carbon paste was prepared by hand mixing graphite pow-er (or single or multi-wall nanotubes), enzyme and oil in anppropriate weight ratio. Physical entrapment of enzyme in theomposite electrode matrix presents the advantage of an easier,aster and cheaper electrode fabrication and at the same timeeduces the possible losses in sensitivity due to the covalentttachment. The resulting pastes were packed into the well ofhe working electrode to a depth of 1 mm. The surface exposed tohe solution was polished on a weighing paper to give a smoothnish before use. The body of the working electrode was a Teflon

ube (3 mm diameter) tightly packed with the carbon paste. Thelectrical contact was provided by a copper wire.

Amperometric measurements were performed using a VA41 amperometric detector (Methrom, Herisau; Switzerland)oupled to a chart recorder (Model L250E, Linseis, Selb,ermany). All experiments were conducted in a three electrode

lectrochemical cell with a volume of 10 mL (0.1 M phosphate,H 6.5, containing 0.1 M KCl) with the enzyme modifiedarbon paste electrode as working electrode, the Ag/AgCl

lectrode as reference electrode and the platinum wire asuxiliary electrode. The working electrode was operated at150 mV and the transient currents were allowed to decay to a

teady-state value. A magnetic stirrer and a stirring bar provided

Page 3

6 nd Bioelectronics 23 (2007) 60–65

tmEfiph

3

dHsbcasafniBbcieoC

u51plafech

w+tiw

tcoTbTcpeot

Fig. 1. Effect of different applied potentials to the response of BPA usingbvp

bwpeiict(eeeconditions used to prepare the electrodes, no protein–proteininteractions occurred. Similar results have been obtained byother authors (Erdem et al., 2000) who observed a non-linearincrease in electrochemical signal when different percentages

2 D.G. Mita et al. / Biosensors a

he convective transport in the electrolytic cell. All measure-ents were performed at room temperature (25.0 ± 0.5 ◦C).ach experimental point in the figures represents the average ofve experiments carried out under the same conditions. Whenossible (i.e. when significantly distinguishable) error barsave been introduced on the experimental points.

. Results and discussion

Being BPA a phenolic compound, it could be a substrate ofifferent enzymes, such as peroxidase, laccase and tyrosinase.owever, due to its steric hindrance, BPA is not a very good sub-

trate for these enzymes and the enzyme choice was performedy measuring the response of the different biosensors to a fixedoncentration of catechol, a reference compound known to bebetter substrate. Moreover, the more suitable type of biosen-

or had to be also selected. In fact, different kinds of biosensors,ssembled immobilising the above mentioned enzymes onto dif-erent solid matrices, such as carbon screen printed electrodes orylon membranes coupled to an oxygen sensor, while respond-ng to phenol and catechol, did not give any useful signal whenPA was used as a substrate (data not shown). Carbon pasteiosensors were the only ones able to respond to BPA. Thisould be explained in terms of extracting properties of the past-ng liquid, which is able to accumulate BPA into the bulk of thelectrode paste containing the enzyme. A similar effect has beenbserved also for other phenolic substrates, at tyrosinase basedPEs (Wang et al., 1997).

The modified carbon paste electrodes were constructed bysing, in the appropriate amounts, 10% in weight of enzyme,0% of carbon powder (graphite) and 40% of mineral oil. When0 �M BPA or catechol in 0.1 M phosphate buffer, 0.05 M KCl,H 6.5, T = 25 ◦C, were used as samples we obtained the fol-owing ratios between the electrical signals measured with BPAnd those determined with catechol: 50% for tyrosinase, 3.5%or peroxidase and 2.6% for laccase. These ratios indicated thenzyme tyrosinase as the best enzyme for determining BPA con-entrations and, as a consequence, all the experiments reportedenceforward have been carried out with tyrosinase.

The more suitable potential to be applied to the biosensoras selected varying it from −250 mV versus Ag/AgCl up to200 mV. Fig. 1 shows the results obtained: the current started

o increase at potentials lower that 100 mV, and reached a max-mum at −150 mV (versus Ag/AgCl). Then, this latter potentialas chosen as working potential in the whole work.The next step was to determine how the response of the

yrosinase-modified carbon paste electrode was affected by itsomposition, in particular how the analytical signal dependedn the amount of the immobilized enzyme (Petit et al., 1995).o this aim three different working electrodes were preparedy employing three tyrosinase concentrations: 2.5, 5 and 10%.o allow direct comparison between the results, the differentarbon pastes were prepared by maintaining the ratio carbon

owder/mineral oil constant at 1.25. This means that the firstlectrode type was constructed by mixing 2.5% of enzyme, 54%f carbon powder and 43% of mineral oil; the second electrodeype was constructed by mixing 5.0% of enzyme, 53% of car-

Ft0c(

iosensors with 10% tyrosinase in GCP. Potential range −250 mV; +200 mVs Ag/AgCl. Phosphate buffer 0.05 mol L−1 + KCl 0.1 mol L−1, pH 6.5; tem-erature = 25 ◦C. Response for BPA solution 2 × 10−5 M.

on powder and 42% of mineral oil; the third electrode typeas constructed by mixing 10.0% of enzyme, 50% of carbonowder and 40% of mineral oil. With each of these workinglectrodes we have studied the electrochemical response to BPAn the concentration range from 3 to 20 �M. The results of thisnvestigation are displayed in Fig. 2 where the electrochemi-al currents are reported as a function of BPA concentration forhree different enzyme percentages. Data in Fig. 2 show that:i) each electrode type was working in its linear range; (ii) atach BPA concentration the electrochemical signals are not lin-ar function of enzyme concentration, but increase faster thannzyme concentration. This behaviour indicates that under the

ig. 2. Biosensors electrochemical signals as a function of BPA concentra-ion. Applied potential = −150 mV vs Ag/AgCl. Phosphate buffer 0.05 M + KCl.1 M, pH 6.5, temperature = 25 ◦C. All the values are average of tripli-ate measurements. Symbols: (©) = 2.5% tyrosinase; (�) = 5.0% tyrosinase;�) = 10.0% tyrosinase.

Page 4

D.G. Mita et al. / Biosensors and Bioelectronics 23 (2007) 60–65 63

Table 1Analytical characteristic of different BPA biosensors (average response ± S.D., n = 5)

Biosensors type Linear range (�M) Sensitivity (nA/�M) Detection limit (�M) SensitivityBPA/sensitivityCatechol (%)

Graphite carbon powder (GCP) 0.1–15 68 ± 4 0.1 50SWCP 0.1–12 138 ± 9 0.02 71M

oestfep

mbtb

iIrbtoreoh

cowo4tsicpTdwttrriwsBtsi

ttwiBw

boii2ebwoTeiTtresults obtained show that no significant differences in the valuesof electrochemical currents or sensitivities were observed eitherfor BPA or for catechol when mineral oil or dodecane were used,while a marked difference was seen with hexadecane. These

WCP 1–16 92 ± 9

f horseradish peroxidase were entrapped in a carbon pastelectrode. The same non-linear behaviour is observed when thelopes of the straight lines in Fig. 2 (i.e. the electrode sensi-ivities: 10.02 nA �M−1 for 2.5% tyrosinase, 22.82 nA �M−1

or 5% tyrosinase and 64.48 nA �M−1 for 10% tyrosinase) arexamined as a function of enzyme concentration in the carbonaste.

The above results appear to indicate that it is possible toodulate both the electrochemical response of a carbon paste

iosensor to BPA and its relative sensitivity on the basis ofhe amount of immobilized tyrosinase, and consequently on theasis of the carbon paste composition.

It is known that the structure of the carbonaceous materialnfluences the electrochemical response (Valentini et al., 2003).n light of these findings, the influence of different carbon mate-ials on the response to BPA using the tyrosinase carbon pasteiosensor has been evaluated. In addition to the graphite powder,herefore, single-wall (SW) or multi-wall (MW) carbon nan-tubes were used as carbonaceous materials. SW nanotubesesult in a mixture of single carbon hollow fibres of differ-nt length and thickness, while MW nanotubes appear, whenbserved by means of a microscope, as a finite number of carbonollow fibres embedded one inside the other.

In this case, we have also changed the composition of thearbon paste owing to the different structural nature of the nan-tubes relative to the carbon powder. The carbon paste preparedith the nanotubes had the following composition in percentf weight: 10% tyrosinase, 45% nanotubes (SW or MW) and5% mineral oil. The amount of oil was increased because ofhe higher surface area of nanotubes compared to the graphiteuperficial area (Valentini et al., 2003). In Table 1, the character-stics of these sensors are compared with those of the graphitearbon powder biosensor, noting that the latter had a differentercentage composition of carbon paste. Inspecting the data inable 1, the carbon paste biosensor realized with graphite pow-er shows the widest linear range, but the two biosensors realisedith CNTs are characterised by a higher sensitivity. Moreover,

he SWNT biosensor appears more interesting since it offershe highest sensitivity together with the lowest detection limitesulting from a low current background and a high signal/noiseatio. Real amperometric recordings of both sensors are shownn Fig. 3, where it can be noted as the same response is obtainedith half BPA concentration in the case of the SWNT biosen-

or. Data in Table 1 show a higher ratio of sensitivity towards

PA and catechol with the NT carbon paste biosensors. Taken

ogether, the data in Table 1, suggest the use of the SWNT biosen-ors as the best strategy to determine low BPA concentrationsn aqueous solutions. In addition, these results seem to confirm

FSap1

1 82

he ability of CNTs to strongly interact with BPA, probably dueo the hexagonal arrays of carbon atoms in the graphene sheets,hich can interact with the benzene rings of bisphenol. MWNTs,

n fact, have been used as solid phase extraction adsorbent forPA, 4-n-nonylphenol and 4-tert-octylphenol, in assays coupledith HPLC analysis (Cai et al., 2003).With the aim of improving the performance of NT-based

iosensor, we have also studied the dependence of its responsen the nature of the oil used in preparing the carbon paste, sincet is well known that the sensitivity of a carbon paste biosensors affected by the density of the oil employed (Chough et al.,002). As mentioned before, tyrosinase carbon paste modifiedlectrodes have been found to be sensitive to the effect of organicinders (Rogers et al., 2001; Wang et al., 1997). For this study,e have used the following composition for carbon paste: 5.0%f tyrosinase, 52.0% in single wall nanotubes and 43.0% in oil.he oils tested were: mineral oil, hexadecane or dodecane. Thelectrochemical responses of the SWNT biosensors have beennvestigated towards BPA and catechol as reference compound.he results of this investigation are reported in Table 2, where

he main characteristics of these biosensors are summarized. The

ig. 3. Original recordings obtained using biosensors with 10% tyrosinase inWCP (a) and GCP (b). Applied potential = −150 mV vs Ag/AgCl. Standardmounts of BPA were added in a 10 mL phosphate buffer 0.05 M + KCl 0.1 M,H 6.5 to get a final concentration of 0.5 × 10−6, 1 × 10−6, 2 × 10−6 M (a) and× 10−6, 2 × 10−6, 4 × 10−6 M (b).

Page 5

64 D.G. Mita et al. / Biosensors and Bioelectronics 23 (2007) 60–65

Table 2Influence of the organic binder on the analytic characteristics of the BPA biosensor (average response ± S.D., n = 5)

Oil type IBPA/ICatechol (%) SensitivityBPA (nA/�M) SensitivityCatechol (nA/�M) Response time (min)

Mineral oil 55 65 ± 4 117 ± 12 5 ± 1Hexadecane 73 99 ± 5 136 ± 11 30 ± 7D

rarittfpDtochtmubrl

ssA0MirfinL2ficw

Bpctarsl

dTws

a9

4

cfic(s(h

tSpt5et

oowscbwBcdlca2rii3cbte

odecane 59 68 ± 7

esults suggest that the partitioning of the substrate between thequeous and the oil phases plays a significant role in the observedesults. More specifically, the alterations of biosensor responsen the presence of different oils are likely due to changes inhe partitioning of substrate molecule in different oils. Parti-ioning effects for substrate in other enzyme electrodes, such asor tyrosinase carbon paste electrodes constructed using variousasting liquids, have also been reported (Chough et al., 2002).ata in Table 2 show that the observed advantages (greater elec-

rochemical currents and sensitivity) obtained in the presencef hexadecane disappear when the response times are taken inonsideration. However, the biosensors prepared with the twoydrocarbons suffered from some stability problems more thanhe mineral oil one (data not shown), indicating that the latter

ight prove more useful, especially if the biosensor has to besed in a repetitive or continuous way. In any case the choiceetween the different electrodes types depends on the targetequired by the operator, i.e. if it prefers high sensitivity andong time for the measure or low sensitivity and fast measure.

Our biosensor based on tyrosinase immobilized in SWCPhows a lower detection limit of 0.02 �M (S/N = 3) and higherensitivity (138 ± 11 �A/mM) than the biosensor reported byndreescu et al. based on tyrosinase immobilized in GCP (LOD.15 �M (S/N = 3) and sensitivity (20.91 ± 0.99 �A/mM)).oreover, our biosensor shows better analytical performances

n terms of sensitivity and detection limit also in respect to whateported by Notsu et al. (2002) and Dempsey et al. (2004). In therst case a boron doped diamond electrode modified with tyrosi-ase was developed to detect BPA in flow injection analysis withOD of 1 �M and a wide linear range 1–100 �M (Notsu et al.,002). In the work of Dempsey et al. phenothiazine polymericlms were adopted to immobilized tyrosinase and its electro-hemical mediator property permitted the BPA measurementith the sensitivity of 0.4 �A/mM and LOD 23 �M (S/N = 3).The lower detection limit and the higher selectivity toward

PA in respect to catechol shown by the SWNT biosensorrepared in the present work, is probably due to favourableombination of the extracting proprieties of the oil present inhe carbon paste, of the electrochemical properties of nanotubesnd of the ability of CNTs to strongly interact with BPA aseported by Cai et al., 2003. To our best knowledge, this biosen-or detects BPA at lowest concentration reported as far in theiterature.

Before concluding, some observations about the repro-

ucibility and stability of the SWNT biosensors are in order.he intraelectrode repeatability never exceeded the 3% (n = 5),hile the interelectrode reproducibility was around 11%. The

torage stability was also evaluated, keeping the biosensors dry

mpet

115 ± 7 7 ± 1

t 4 ◦C, when not in use. After 1 month, the activity was still0% of the original.

. Conclusions

From all the results reported above it is possible to con-lude that: (i) among tyrosinase, laccase and peroxidase, therst enzyme is the most useful for the determination of BPAoncentration by means of modified carbon paste electrodes;ii) the analytical signal is function of the carbon paste compo-ition: concentration of entrapped enzyme, type of the carbonGCP, NWCT, or MWCT), nature of the pasting oil (mineral oil,exadecane or dodecane).

Under our experimental conditions the biosensors showinghe most interesting characteristics were those prepared usingWNT and dodecane oil or mineral oil to construct the carbonaste. These biosensor types, indeed, showed the following elec-rochemical characteristics: sensitivity equal to 138 �A/mM andmin of response time. Moreover, these biosensors showed anxcellent detection limit, equal to 0.02 �M, and good biosensoro biosensor reproducibility (11%) as well as great stability.

Their small size and the other properties of single wall nan-tubes, when coupled with other oils can lead to the developmentf novel sensors facilitating rapid and on-site monitoring of BPA,ith significant implications for health and safety. The high sen-

itivity and reproducibility, together with the simplicity and lowost, make the tyrosinase-modified electrode very attractive. Weelieve that an analytical method such as the one reported hereill have widespread applications as it provides a direct assay forPA. Independently from the interference with other phenolicompounds, our SWNC biosensor appears not yet suitable forirect application to polluted surface waters since our detectionimit, 2 × 10−8 M, is just one order higher than the BPA averageoncentration (0.5 �g/L) estimated by some authors (Cousins etl., 2002) which analysed the BPA concentrations measured by1 European and 13 United States Researches in streams andivers in Japan, Europe and United States. Of course, 0.5 �g/Ls an average measure. Indeed, Staples et al. (2000) measuredn upstream waters a BPA concentration equal to 8 �g/L, i.e..5 × 10−8 M, a value 57% higher than our detection limit. Theonclusion is that it should be feasible to use our biosensor inioremediation systems, for example with a modest preconcen-ration of the sample, for instance using NTs also as solid phasextraction adsorbent such as for chromatographic measurements

entioned before. Moreover, it is possible to ameliorate the

erformance of our biosensor, for example, by increasing thenzyme concentration in the carbon paste. Studies in this direc-ion are being carried out in our laboratories.

Page 6

nd Bi

A

tAls“(

R

ABBB

CCC

C

C

C

DDE

G

II

JKK

M

M

M

MM

NP

PP

R

R

R

RRSSS

V

W

D.G. Mita et al. / Biosensors a

cknowledgments

We are grateful to the Regione Campania and CRdC in Indus-rial Biotechnologies for the study grant awarded to Dr. Angelinattanasio. We would also like to thank the “Assessorato per

’Ambiente” of the Regione Campania for providing financialupport for this research project. Our thanks also go to theIstituto Superiore per la Prevenzione e Sicurezza sul Lavoro”ISPESL) which partly funded this research.

eferences

ndreescu, S., Sadik, O.A., 2004. Anal. Chem. 76, 552–560.erkner, S., Streck, G., Herrmann, R., 2004. Chemosphere 54, 575–584.irnbaum, L.S., Fenton, S.E., 2003. Environ. Health Perspect. 111, 389–394.rzozowski, A.M., Pike, A.C., Dauter, Z., Hubbard, R.E., Bonn, T., Engstrom,

O., Ohman, L., Greene, G.L., Gustafsson, J.A., Carlquist, M., 1997. Nature389, 753–758.

ai, Y., Jiang, G., Liu, J., Zhou, Q., 2003. Anal. Chem. 75, 2517–2521.hang, C.M., Chou, C.C., Lee, M.R., 2005. Anal. Chim. Acta 539, 41–47.hough, S.Y., Mulchandani, A., Mulchandani, P., Chem, W., Wang, J., Rogers,

A.R., 2002. Electroanalysis 14, 273–276.olborn, T., Dumenoski, D., Uyers, J.P., 1996. Our Stolen Future. Book AD,

New York.olborn, T., von Saal, F.S., Soto, A.M., 1993. Environ. Health Perspect. 101,

378–384.ousins, J.T., Staples, C.A., Klecka, G.M., Mackai, D., 2002. Hum. Ecol. Risk

Assess. 8, 1107–1135.empsey, E., Diamond, D., Collier, A., 2004. Biosens. Bioelectron. 20, 367–377.odds, E.C., Lawson, W., 1938. Proc. R. Soc. Lords B 125, 222–232.rdem, A., Pabuccuoglu, A., Meric, B., Kerman, K., Ozsoz, M., 2000. Turk. J.

Med. Sci. 30, 349–354.

oodson, A., Summerfield, W., Cooper, I., 2002. Food Addit. Contam. 19,

796–802.ijima, S., 1991. Nature 354, 56–58.shihara, A., Sawatsubashi, S., Yamauchi, K., 2003. Mol. Cell. Endocrinol. 199,

105–117.

WY

Z

oelectronics 23 (2007) 60–65 65

imenez, B., 1997. Trends Anal. Chem. 16, 596–604.ang, J.H., Kondo, F., 2003. J. Food Prot. 66, 1439–1443.rishan, K.A., Starhis, P., Permuth, S.F., Tokes, L., Feldman, D., 1993.

Endocrinology 132, 2279–2286.archesini, G.R., Meulenberg, E., Haasnoot, W., Irth, H., 2005. Anal. Chim.

Acta 528, 37–45.archesini, G.R., Meulenberg, E., Haasnoot, W., Mizuguchi, M., Irth, H., 2006.

Anal. Chem. 78, 1107–1114.arkey, C.M., Wadia, P.R., Rubin, B.S., Sonnenschein, C., Soto, A.M., 2005.

Biol. Reprod. 72, 1344–1351.ueller, S.O., 2004. Anal. Bioanal. Chem. 378, 582–587.usameh, M., Wang, J., Merkoci, A., Lin, Y., 2002. Electrochem. Commun. 4,

743–746.otsu, H., Tatsuma, T., Fujishima, A., 2002. J. Electroanal. Chem. 523, 86–92.ark, J.W., Kurosawa, S., Aizawa, H., Goda, Y., Takai, M., Ishihara, K., 2006.

Analyst 131, 155–162.edersen, S.N., Lindholst, C., 1999. J. Chromatogr. A 864, 17–24.etit, C., Gonzalez-Cortes, A., Kauffmann, J.M., 1995. Talanta 47, 1783–

1789.odriguez-Mozaz, S., Lopez de Alda, M., Barcelo, D., 2005. Water Res. 39,

5071–5079.oger, J., Hassan, S., Emmanuel, S., Thierry, D., Katja, D., 2002. J. Chromatogr.

A 974, 143–159.ogers, K.R., Becker, J.Y., Cembrano, J., Chough, S.H., 2001. Talanta 54,

1059–1065.ubianes, M.D., Rivas, G.A., 2003. Electrochem. Commun. 5, 689–694.ubianes, M.D., Rivas, G.A., 2005. Electroanalysis 17, 73–78.ingleton, D.W., Khan, S.A., 2003. Front. Biosci. 8, 110–118.onnenschein, C., Soto, A.M., 1998. J. Steroid Biochem. Mol. Biol. 65, 143–150.taples, C.A., Dorn, P.B., Klecka, G.M., Branson, D.R., O’Block, S.T., Harris,

L.R., 2000. Chemosphere 40, 521–525.alentini, F., Amine, A., Orlanducci, S., Terranova, M.L., Palleschi, G., 2003.

Anal. Chem. 75, 5413–5421.ang, J., Lu, F., Kane, S.A., Choi, Y.K., Smyth, M.R., Rogers, K., 1997. Elec-

troanalysis 9, 1102–1106.ang, J., Musameh, M., Lin, Y., 2003. J. Am. Chem. Soc. 125, 2408–2409.

oshida, T., Horie, M., Hoshino, Y., Nakazawa, H., 2001. Food Addit. Contam.18, 69–75.

hao, Q., Guan, L., Gu, Z., Zhuang, Q., 2005. Electroanalysis 17, 85–88.