Anal. Chem. 2006, 78, 4985-4989
Specific Determination of As(V) by an Acid Phosphatase-Polyphenol Oxidase Biosensor Serge Cosnier,*,† Christine Mousty,*,† Xiaoqiang Cui,‡ Xiurong Yang,‡ and Shaojun Dong‡
Laboratoire d’Electrochimie Organique et de Photochimie Redox, UMR CNRS 5630, ICMG FR CNRS 2607, Universite Joseph Fourier, Grenoble, France, and State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun, Jilin 130022, China
An original amperometric biosensor based on the simultaneous entrapment of acid phosphatase (AcP) and polyphenol oxidase (PPO) into anionic clays (layered double hydroxides) was developed for the specific detection of As(V). The functioning principle of the bienzyme electrode consisted of the successive hydrolysis of phenyl phosphate into phenol by AcP, followed by the oxidation of phenol into o-quinone by PPO. The phenyl phosphate concentration was, thus, monitored by potentiostating the biosensor at -0.2 V vs Ag/AgCl to detect amperometrically the generated quinone. The detection of As(V) was based on its inhibitory effect on AcP activity toward the hydrolysis of phenyl phosphate into phenol. The As(V) can be specifically determined in pH 6.0 acetate buffer without any interferences of As(III) or phosphate, the detection limit being 2 nM or 0.15 ppb after an incubation step for 20 min. Arsenic contamination resulting from natural geologic activity and manmade sources is a serious threat to mankind all over the world. The new drinking water standard for arsenic has been lowered to 10 µg L-1 in the United States and now constitutes the guideline value for the World Health Organization.1,2 This new awareness of the toxicity of arsenic to humans has led to a growing interest in arsenic speciation rather than in total arsenic determination. Arsenic contamination, indeed, can occur through different chemical forms exhibiting different toxicological effects. Several methods, such as inductively coupled plasma mass spectrometry (ICP-MS), hydride generation atomic absorption spectrometry, and hydride generation atomic fluorescence spectrometry, have been developed to detect arsenic.1-4 These conventional analytical techniques are precise, with detection limits within the range of 0.02-20 ppb;5 however, they are expensive and time-consuming, require formal training, and are not suitable for on-site monitoring. Moreover, most of the * Address correspondence to either author. (Cosnier) E-mail: Serge.Cosnier@ ujf-grenoble.fr. (Mousty) Fax: + 33 476 514 267. E-mail: Christine.Mousty@ ujf-grenoble.fr. † Universite Joseph Fourier. ‡ Changchun Institute of Applied Chemistry. (1) Mandal, B. K.; Suzuki, K. T. Talanta 2002, 58, 201-235. (2) Richardson, S. D.; Ternes, T. A. Anal. Chem. 2005, 77, 3807-3838. (3) Gong, Z.; Lu, X.; Ma, M.; Watt, C.; Lee, X. C. Talanta 2002, 58, 77-96. (4) Hung, D. Q.; Nekrassova, O.; Compton, R. G. Talanta 2004, 64, 269-277. (5) Karthikeyan, S.; Hirata, S. Anal. Lett. 2003, 36, 2355-2366. 10.1021/ac060064d CCC: $33.50 Published on Web 06/15/2006
© 2006 American Chemical Society
conventional methods can only detect the total arsenic content or need to convert As(V) to As(III), the concentration of As(V) being determined from the difference between the total concentration and the concentration of As(III). Only a few methods are focused on the speciation between As(III) and As(V), although in natural waters, the concentration of As(V) is much higher than that of As(III).3 Recently, the colorimetric method using the molybdenum blue complex was developed for the determination of As(V) at a detection limit of 8 or 20 µg L-1.6,7 Electrochemistry methods, mainly cathodic stripping voltammetry, anodic stripping voltammetry, and differential pulse voltammetry, have also been developed for arsenic analysis. The electrochemical methods are considered as an alternative to the other analytical techniques because of the low cost of equipment and the possibility to distinguish electrochemically between As(III) and As(V). However, these methods also suffer from poor reproducibility due to the interfering effects of other electroactive metal ions, such Cu2+.4,8 Electrochemical biosensors have been increasingly developed for in-situ monitoring in environmental and health care applications due to their advantages in terms of simplicity, portability, short response time, sensitivity, and high selectivity by substrate specificity. Although a large variety of biosensors were used to analyze heavy metal ions via inhibition of biological activities,9,10 only two biosensor configurations were described for arsenic detection: a whole cell biosensor utilizing a green fluorescent protein reporter gene11 and an electrochemical biosensor based on the inhibition of acetylcholinesterase activity.12 The former displayed a response to submicrogram quantities of total arsenic; the latter, based on the monitoring of thiocholine oxidation, was faced to amperometric side-reactions due to iodide oxidation. Since As(V) is predominant in natural waters and its toxicity results from inhibition of phosphorylation activities,13 we report here a new design of arsenic biosensor based on the inhibition of (6) Lenoble, V.; Deluchat, V.; Serpaud, B.; Bollinger, J. C. Talanta 2003, 61, 267-276. (7) Dasgupta, P. K.; Huang, H.; Zhang, G.; Cobb, G. P. Talanta 2002, 58, 153164. (8) Cavicchioli, A.; La-Scalea, M. A.; Gutz, I. G. R. Electroanalysis 2004, 16, 697-710. (9) Verma, N.; Singh, M. BioMetals. 2005, 18, 121-129. (10) Sole, S.; Merkoci, A.; Alegret, S. Crit. Rev. Anal. Chem. 2003, 33, 89-126. (11) Roberto, F. F.; Barnes, J. M.; Bruhn, D. F. Talanta 2002, 58, 181-188. (12) Stoytcheva, M.; Sharkova, V.; Panayotova, M. Anal. Chim. Acta 1998, 364, 195-201. (13) Pappas, R. W. Exp.erimental Parasitol. 1991, 72, 362-367.
Analytical Chemistry, Vol. 78, No. 14, July 15, 2006 4985
Scheme 1. Schematic Representation of Enzymatic Reaction Sequences for Inhibition Detection of Arsenate
acid phosphatase (AcP). The amperometric transduction of the inhibitive effect of arsenic was based on the cooperative functioning of acid phosphatase and polyphenol oxidase (PPO) (Scheme 1). This two-enzyme configuration should overcome the electrode fouling expected from the direct detection of phenol, the AcP product, through its oxidation.14 In addition, AcP, in contrast to alkaline phosphatase, exhibited an optimum pH in the acidic range that is compatible with the optimum PPO functioning.14,15 Both enzymes were simultaneously immobilized on the electrode surface by entrapment into a clay coating. Clays, indeed, constitute a biocompatible and permeable host matrix for the development of enzyme electrodes.16 The synthetic layered double hydroxides, LDH, also referred to as anionic clays or hydrotalcite-like compounds, were chosen as the host matrix owing to their anionexchange properties.17-22 It is, thus, expected that LDH will induce the accumulation of negatively charged arsenic species in the microenvironment of the immobilized enzymes and, hence, will improve the biosensor sensitivity. The analytical characteristics of the AcP/PPO/LDH electrode have been investigated for the amperometric detection of phenyl phosphate, As(V), As(III), and inorganic phosphate. EXPERIMENTAL SECTION Materials. Acid phosphatase (EC 3.1.3.2) from potato (liquid, 478 units mL-1, 0.5 mg mL-1); polyphenol oxidase (EC 1.14.18.1) from mushroom (2590 units mg-1); sodium hydrogenarsenate heptahydrate, As(V); and phenyl phosphate disodium salt were purchased from Sigma-Aldrich. Sodium (meta) arsenite, As(III), was purchased from Fluka. The layered double hydroxide Zn3Al(OH)8Cl was synthesized by the coprecipitation method developed (14) Mousty, C.; Bergamasco, J.; Wessel, R.; Perrot, H.; Cosnier, S. Anal. Chem. 2001, 73, 2890-2897. (15) Kruzel, M.; Morawiecka, B. Acta Biochim. Pol. 1982, 29, 321-330. (16) Mousty, C. Appl. Clay Sci. 2004, 27, 159-177. (17) De Roy, A.; Forano, C.; El Malki, K.; Besse, J.-P. In Expanded Clays and Other Microporous Solids; Occelli, M. L., Robson, H. E., Eds.; Van Nostrand Reinhold: New York, 1992; pp 108-169. (18) Forano, C.; Hibino, T.; Leroux, F.; Taviot-Gueho, C. In Handbook of Clay Science; Bergaya, F., Theng, B. K. G., Lagaly, G., Eds.; Elsevier: New York, 2006. (19) Shan, D.; Mousty, C.; Cosnier, S. Anal. Chem. 2004, 76, 178-183. (20) De Melo, J. V.; Cosnier, S.; Mousty, C.; Martelet, C.; Jaffrezic-Renault, N. Anal. Chem. 2002, 74, 4037-4043. (21) Shan, D.; Cosnier, S.; Mousty, C. Anal. Chem. 2003, 75, 3872-3879. (22) Shan, D.; Cosnier, S.; Mousty, C. Anal. Lett. 2003, 36, 909-922.
4986
Analytical Chemistry, Vol. 78, No. 14, July 15, 2006
by De Roy.17 All other chemical reagents were of analytical grade. Water was doubly distilled in a quartz apparatus. Apparatus. The amperometric measurement was performed with an Autolab PGSTAT 100. All the experiments were performed with a conventional thermostatic three-electrode cell (10 mL) at 30 °C. A Pt wire and a Ag/AgCl electrode (saturated KCl solution) were used as counter and reference electrodes, and both were placed in a separate compartment containing the buffer solution. A glassy carbon electrode (diameter 5 mm) was used as the working electrode and polished before use with a 1-µm diamond paste. Enzyme Electrode Preparation. The clay colloidal suspension was prepared by dispersing LDH (2 mg/mL) in deionized and decarbonated water overnight under stirring conditions. PPO was dissolved in water at a concentration of 2 mg/mL. AcP solution was directly used without any operation after being received from Sigma. A defined amount of the mixture of PPO, AcP, and LDH solution (for example, containing 15 µg of PPO, 5 µg of AcP, and 30 µg of LDH) was spread on the surface of the glassy carbon electrode, and the electrode was stored at 4 °C in the refrigerator for drying overnight. The resulting electrode was then placed in saturated glutaraldehyde vapor for 15 min to induce a cross-linking of the entrapped proteins. Finally, the PPO/AcP/ LDH biocoating was immersed into 0.1 M acetate buffer (pH 6.0, 10 mM MgCl2) for 40 min under stirring (500 rpm). Acetate buffer was used for all the analysis procedures. Since the magnesium ions stimulated the AcP activity,13 MgCl2 (10 mM) was added to the buffer solutions. The amperometric detection of phenyl phosphate was carried out by holding the bienzyme electrode at -0.2 V in 0.1 M acetate buffer containing MgCl2 (10 mM) thermostated at 30 ( 0.1 °C. RESULTS AND DISCUSSION Amperometric Responses of AcP/LDH and PPO/AcP/ LDH Electrodes to Phenyl Phosphate. A monoenzyme electrode was first designed by entrapment of AcP into the LDH matrix. This biosensor was aimed at the detection of phenyl phosphate via the oxidation of phenol at +0.9V, resulting from the enzymatic hydrolysis of phenyl phosphate. Unfortunately, this kind of biosensor suffered from poor stability, as evidenced by a rapid decrease in the steady-state current response that was observed after each phenyl phosphate injection (Figure 1A). This phenomenon was ascribed to fouling of the electrode by electroinactive polymers resulting from the polymerization of electrochemically oxidized phenol.14,23,24 Consequently, the AcP/LDH electrode was totally unsuitable for the determination of As(V) through its inhibitory effect on the AcP activity. To circumvent this detrimental passivation, PPO was associated with AcP to convert the oxidative transduction step into a reductive one. The functioning principle of the bienzyme electrode is based on the hydrolysis of phenyl phosphate to phenol by AcP while PPO catalyzes the oxidation of phenol to o-quinone. The amperometric detection of phenyl phosphate (PP) was, thus, monitored by holding the biosensor at -0.2 V to detect the generated o-quinone (Scheme 1). In addition, an amplification of the biosensor response is expected due to the electrochemical recycling of the PPO (23) Rosen, I.; Rishpon, J. J. Electroanal. Chem. 1989, 258, 27-39. (24) Wilson, M. S.; Rauh, R. D. Biosens. Bioelectron. 2004, 20, 276-283.
Table 1. Amperometric Performance of PPO/AcP/LDH Bioelectrodes to Phenyl Phosphate Determination serials
PPO (µg)
AcP (µg)
LDH (µg)
sensitivity (A M-1 cm-2)
linear range (M)
R2 (n)
a b c d
15 15 30 30
5 10 5 5
30 30 30 45
1.186 ( 0.0048 0.480 ( 0.0056 0.820 ( 0.013 0.855 ( 0.011
1.7 × 10-8 to 5.7 × 10-5 6.0 × 10-7 to 7.9 × 10-5 4.1 × 10-6 to 7.8 × 10-5 5.7 × 10-6 to 4.1 × 10-5
0.9994 (37) 0.9989 (29) 0.9978 (12) 0.9984 (11)
Figure 2. Influence of pH values on the amperometric response of the PPO/AcP/LDH bioelectrode to 2 µM phenyl phosphate. (Eapp ) -0.2 V, 0.1 M acetate buffer, 10 mM MgCl2).
Figure 1. (A) Current response as a function of time at AcP/LDH electrode, with successive additions of phenyl phosphate: (a) 4 µM and (b and c) 10 µM (Eapp ) 0.9 V). (B) Current response as a function of time at the PPO/AcP/LDH electrode, with successive additions of phenyl phosphate: (a) 2 µM and (b and c) 10 µM (Eapp ) -0.2 V). (C) Calibration curve for phenyl phosphate at the PPO/AcP/LDH electrode. (D) Curve C expanded in µM range. (Eapp) -0.2 V, in 0.1 M acetate buffer pH 6.0, 10 mM MgCl2).
substrate (catechol).25-27 The steady-state current response of the bienzyme electrode to PP injections was stable, indicating the absence of any electrode fouling (Figure 1B). Moreover, no inhibition effect on the PPO activity that is due to As(V) has been detected. Optimization of the Bienzyme Electrode Configuration. The biosensor elaboration was based on the simultaneous immobilization of PPO and AcP in LDH, and the optimum composition was investigated through the efficiency of the amperometric detection of PP (Table 1). According to our previous works on LDH/enzyme biosensors,19,21,22 the LDH amount was fixed at 30 µg for the construction of the biosensor configuration. If the PPO (25) Cosnier, S.; Gondran, C.; Watelet, J.; de Giovani, W.; Furriel, R. P. M.; Leone, F. A. Anal. Chem. 1998, 70, 3952-3956. (26) Cosnier, S.; Innocent, C. Bioelectrochem. Bioenerg. 1993, 31, 147-160. (27) Besombes, J. L.; Cosnier, S.; Labbe, P. Talanta 1996, 43, 1615-1619.
amount remained constant while the AcP amount was doubled, the biosensor performance, namely, its sensitivity and linear range, decreased instead of increased (Table 1a and b). This indicates that the PPO activity is the limiting factor of the bienzyme system.14 Therefore, the deposited amount of PPO was increased up to 30 µg while that of AcP remained at 5 µg (Table 1c). Surprisingly, a slight decrease in PP sensitivity was observed, the linear range values being similar. This may be ascribed to an excess of proteins (35 µg), as compared to the deposited mass of LDH (30 µg), since we previously reported a decrease in the biosensor performance for an enzyme/LDH ratio above 1.21 To eliminate a possible enzyme release into the solution during the coating formation, the LDH amount was increased to 45 µg, and the deposited amount of PPO and AcP was maintained at 30 and 5 µg. Only a small increase in sensitivity was recorded (Table 1d). This may indicate that the beneficial effect brought about by a better enzyme entrapment was counterbalanced by the increase in coating thickness and, hence, the increase in the diffusion constraints for PP and phenol. The same phenomenon was observed in our previous work for cyanide detection using a PPO/ LDH electrode.19 Therefore, the composition of PPO/AcP/LDH (15, 5, and 30 µg, respectively) was used for further experiments. Since both PPO and AcP are active in the acidic range,14,15 the effect of pH on the biosensor functioning was investigated over the range 4.5-6.5 in 0.1 M acetate buffer containing MgCl2 (10 mM). Figure 2 shows the amperometric response of the bienzyme electrode to phenyl phosphate (2 µM). An optimum biosensor response was achieved at pH 6, which is in good agreement with the optimum value found for PPO/LDH bioelectrode.19,21 Previous reports indicated that the pH has a weak effect on the activity of AcP immobilized either on composite beads of chitosan and activated clay28 or in Nafion film29 in the pH range 4-6. Therefore, (28) Chang, M. Y.; Juang, R. S. Proc. Biochem. 2004, 39, 1087-1091. (29) Calvo-Marzal, P.; Rosatto, S. S.; Granjeiro, P. A.; Aoyama, H.; Kubota, L. T. Anal. Chim. Acta 2001, 441, 207-214.
Analytical Chemistry, Vol. 78, No. 14, July 15, 2006
4987
Table 2. Effect of the Concentration of Phenyl Phosphate on the As(V) Determination Linear range
phenyl phosphate concn (µM) 20 10 5 2 2 (20 min)
M 10-6
10-5
2.7 × to 2.2 × 6.9 × 10-8 to 1.5 × 10-6 6.9 × 10-8 to 1.4 × 10-6 6.9 × 10-8 to 7.9 × 10-6 8.9 × 10-9 to 7.9 × 10-8
LOD ppb
R2 (n)
RA (%)
nM
ppb
202-1648 5-112 5-105 5-592 0.7-6
0.9889 (5) 0.9933 (6) 0.9973 (5) 0.9966 (6) 0.9846 (4)
80 88 87 90 71
760 64 69 30 2
57 5 5 2 0.15
the optimum pH value in this range depends mainly on the PPO activity. Figure 1C and D present the calibration curve for phenyl phosphate at the bienzyme electrode under the optimum experimental conditions. The curve was linear with PP concentration up to 60 µM and curved at higher concentrations, a pseudoplateau being reached above 200 µM, reflecting saturation of the bienzymatic system. The reproducibility of the current response of the biosensors was also examined. For a given biosensor, the relative standard deviation (RSD) was 6.7% for 13 injections of 2 µM PP, and for six different electrodes, the RSD was 9.0%. Inhibitory Detection of As(V). The amperometric detection of As(V) at PPO/AcP/LDH electrodes was investigated by the following procedure on the basis of an inhibition process. A defined concentration of PP was added to the electrolyte solution, and the current response was recorded at -0.2 V. After stabilization of the steady-state current response, As(V) was added, inducing immediately a decrease in the biosensor response. The As(V) detection was quantified as an inhibition percentage (In, %) corresponding to the ratio of the current decrease (I0 - I) versus the original current I0 (no inhibition) in the steady state. The limit of detection (LOD) was determined as a signal-to-noise ratio of 3. Figure 3A presents the influence of the substrate concentration on the resulting calibration curves for the inhibitive detection of As(V). As expected, it appears that the inhibition percentage increases for a defined As(V) concentration when the substrate concentration changes from 20 to 2 µM, the concentration range corresponding to the linear part of the PP calibration curve. This phenomenon is in agreement with the competitive inhibition process reported for the free enzyme.13 The detection limit and linear range (LR) of the biosensor are also improved when the substrate concentration decreases until 2 µM (Table 2). However, the decrease in PP concentrations led to a smaller direct biosensor response to enzymically generated quinone, concentrations lower than 2 µM being not exploitable for observation of an inhibition process. Therefore, for the inhibitive detection of As(V), 2 µM of PP was chosen as the best compromise between sensitivity of the inhibition process and enough current intensity to quantify a range of inhibitor concentrations. Under these experimental conditions, the in-situ detection of As(V) provided a 2 ppb LOD and a LR of 5-592 ppb. It should be noted that this LOD value is close to those recently reported with the ICP-MS method (0.5 ppb)30 or is more sensitive than the molybdene blue method (20 ppb)6 and the potentiometric flow method (30.6 ppb).31 (30) Coelho, N. M. M.; Coelho, L. M.; Lima, E. S.; Pastor, A.; de la Guardia, M. Talanta 2005, 66, 818-822. (31) Rodriguez, J. A.; Barrado, E.; Vega, M.; Lima, J. L. F. C. Electroanalysis 2005, 17, 504-511.
4988 Analytical Chemistry, Vol. 78, No. 14, July 15, 2006
Figure 3. (A) Inhibition calibration curves for the detection of arsenate with different concentrations of phenyl phosphate: (a) 2, (b) 5, (c) 10, (d) 20, and (e) 2 µM using the incubation method (Eapp) -0.2 V, in 0.1 M acetate buffer pH 6.0, 10 mM MgCl2). (B) Logarithm plot of curve e; insert is current response on 2 µM phenyl phosphate injection (f) without arsenate and (g) after 20 min of incubation in 2 nM arsenate.
After the detection of As(V), the biosensor was washed carefully with electrolyte, and the amperometric response toward PP was again measured to determine the recovered activity (RA) in the absence of As(V). The RA (%) is I0′/I0, where I0 and I0′ correspond to the biosensor responses to PP before and after As(V) determination. In the low As(V) linear concentration range, the As(V) inhibition was completely reversible, with RA ) 9499%, whereas the RA decreased to ∼80% after the detection of higher concentrations up to 2 mM. This illustrates the reversible character of the inhibition process by As(V) (Table 2). Moreover, the decrease in RA at higher concentrations may reflect the postulated accumulation of As(V) anion within the biocoating due to the anion exchange properties of LDH.19 To exploit the potentialities of LDH for the preconcentration process of the anionic inhibitor, the effect of incubation time on the inhibition percentage was studied by using two As(V)
Figure 4. Effect of incubation time on biosensor inhibition with (a) 2 × 10-4 and (b) 1 × 10-5 M arsenate. Same experimental conditions as given in Figure 3.
Figure 5. Dixon plots for PPO/AcP/LDH electrodes in the presence of phenyl phosphate: (a) 2, (b) 5, (c) 10, and (d) 20 µM. Same experimental conditions as given in Figure 3.
concentrations (2 × 10-4 and 1 × 10-5 M). It appears that the inhibition percentage increases with incubation time and reaches a pseudoplateau after 20 min (Figure 4). Consequently, an incubation time of 20 min was chosen for inhibition detection of As(V) under the optimum experimental conditions previously obtained (Figure 3A, curve e). Although the LR was reduced, a marked enhancement in LOD value was observed; namely, 0.15 ppb instead 2 ppb (Figure 3B, Table 2). In addition, the RA decrease to 71% corroborates the presence of specific electrostatic interactions between positively charged the LDH layers and the As(V) anion. It should be noted that the main advantage brought by the preconcentration step leads to a loss of the biosensor reusability. Kinetic Study of the Inhibitory Effect of As(V). An apparent inhibition binding constant, Ki, was determined from the Dixon plot for different PP concentrations (Figure 5). The plots of the reciprocal of the current response (1/i) versus the inhibitor concentration with various substrate concentrations did not intersect at the x-axis, suggesting that inhibition was effectively competitive.13 The Ki was then calculated from the intersection of the lines as 1 mM, and this value falls within the range of inhibitor concentration used in the kinetic experiment. There is no reference about the kinetics of inhibition by As(V) for the free or the immobilized AcP using PP as substrate. Nevertheless, a Ki value (3.45 mM) for As(V) was determined for the inhibition of the p-nitrophenyl phosphate (PNPP) hydrolysis by the free AcP.13 (32) Su, Y.; Mascini, M. Anal. Lett. 1995, 28, 1359-1378. (33) Katsu, T.; Kayamoto, T. Anal. Chim. Acta 1992, 265, 1-4.
Taking in account that PNPP is a better substrate than PP for AcP and, hence, displays more affinity for AcP, the Ki value of As(V) should be smaller for PP than for PNPP. This hypothesis is in good agreement with the experimental Ki value (1 mM), indicating, thus, that the LDH host matrix does not change the enzymatic mechanism. Specificity of the Biosensor. In addition to the detection limit of the determination, the speciation between the two oxidation states As(III) and As(V) is very important in view of their different biological activities.4 Indeed, it appears fundamental to determine individual species of arsenic rather than total arsenic contents.2 In this work, the response of the biosensor to As(III) was also investigated under the same conditions used for the As(V) detection. No appreciable change in the steady-state current response of the biosensor was observed when the As(III) concentration was increased up to 1 × 10-4 M. It should be noted that the absence of inhibition was already described for a 200:1 inhibitor-to-substrate ratio for free enzyme.13 This implies that As(V) can be specifically detected by the bienzyme electrode without any interference of As(III) in the concentration range 8.9 × 10-9 to 1 × 10-4 M. Phosphate, which is a well-known water pollutant, is another competitive inhibitor of alkaline and acid phosphatase.13,25 As a consequence, the effect of phosphate on the amperometric response of the bienzyme electrode was examined under the optimum conditions used for As(V) determination: PP (2 µM) in 0.1 M acetate buffer (pH 6.0, 10 mM MgCl2). It appears that the biosensor response is weakly affected by the presence of phosphate, a linear decrease of the response being observed between 0.17 and 3.2 mM (R2 ) 0.995, n ) 11), the LOD being only 0.17 mM. The latter is drastically less sensitive than those previously reported for phosphate detection by alkaline phosphatase biosensors (2,25 4,32 and 50 µM33). This implies that our biosensor can be used to detect As(V) without any interference of phosphate at levels commonly encountered in drinking water, groundwater, and surface water. CONCLUSION We demonstrated herein, for the first time, the original association of AcP and PPO for the specific amperometric determination of As(V) without electrode fouling. In addition, we have illustrated, in connection with this bienzyme system, the attractive potentialities offered by a LDH coating for the preconcentration of anionic As(V). It is expected that this strategy of nanomolar As(V) detection, coupled with conventional electrochemical methods, will be useful for the accurate determination of As(V) and As(III). ACKNOWLEDGMENT The authors gratefully acknowledge the financial support for this work from the Program de Recherches Avance´es de Coope´rations Franco-Chinoises (PRA BT03-02, “Biosensors in extreme environment: preparation, characterization and applications”: S. Cosnier and E. Wang) and from the Research Ministry of France (ACI. Program Nanohybrides Enzymes-LDH, 2003-NR0005). Received for review January 10, 2006. Accepted May 10, 2006. AC060064D Analytical Chemistry, Vol. 78, No. 14, July 15, 2006
4989
