Mixed Ceria-Based Metal Oxides Biosensor for Operation in Oxygen

Aug 23, 2008 - Operation in Oxygen Restrictive Environments ... Department of Chemistry and Biomolecular Science, Clarkson University Potsdam, New Yor...
0 downloads 0 Views 773KB Size
Download PDF
Anal. Chem. 2008, 80, 7266–7274

Mixed Ceria-Based Metal Oxides Biosensor for Operation in Oxygen Restrictive Environments John Njagi, Cristina Ispas, and Silvana Andreescu* Department of Chemistry and Biomolecular Science, Clarkson University Potsdam, New York 13699-5810 The unique catalytic, electrochemical, and oxygen storage properties of ceria and mixed ceria/titania hybrid composites were used to fabricate a new type of electrochemical enzyme biosensor. These materials provided increased analytical performance and possibilities for operation in oxygen-free conditions of an oxidase enzyme biosensor using tyrosinase as a model example. The investigation of the enzymatic reaction in the presence and absence of oxygen was first carried out using cyclic voltammetry. The results were used to identify the role of each metal oxide in the immobilization matrix and fabricate a simple amperometric tyrosinase biosensor for the detection of phenol and dopamine. The biosensor was optimized and characterized with respect to response time, detection limit, linear concentration range, sensitivity, and kinetic parameters. The detection limit for phenol was in the nanomolar range, with a detection limit of 9.0 × 10-9 M and a sensitivity of 86 mA M-1 in the presence of oxygen and of 5.6 × 10-9 M and a sensitivity of 65 mA M-1 in the absence of oxygen. The optimized biosensor also showed selective determination of the neurotransmitter dopamine with a detection limit of 3.4 × 10-8 M and a sensitivity of 14.9 mA M-1 in the presence of oxygen and of 4.2 × 10-8 M and 14.8 mA M-1 in the absence of oxygen. This strategy shows promise for increasing the sensitivity of oxidase enzyme sensors and provides opportunities for operation in oxygen limited conditions. It can also be extended for the development of other enzyme biosensors. Recently, metal oxide semiconductors and composites of these materials have attracted considerable interest in catalysis, solid oxide fuel cells, electrochemistry and sensing devices.1 These materials have interesting catalytic and electrochemical properties due to their unique chemical and electronic configurations, mechanical stability, and biocompatibility. It is due to these properties that they were used to enhance the performance of electrochemical sensors and biosensors such as decrease in overpotential, direct electron transfer capabilities, and high sensitivity and selectivity.2,3 For example, iron oxide (Fe2O3) and * Corresponding author. Phone: 1 315 268 2394. Fax: 1 315 268 6610. E-mail: [email protected]. (1) Yuqing, M.; Jianrong, C.; Xiaohua, W. Trends Biotechnol. 2004, 22, 227– 231. (2) Khan, R.; Dhayal, M. Electrochem. Commun. In press. (3) Li, Y.-F.; Liu, Z.-M.; Liu, Y.-Y.; Yang, Y.-H.; Shen, G.-L.; Yu, R. Q. Anal. Biochem. 2006, 349, 33–40.

7266

Analytical Chemistry, Vol. 80, No. 19, October 1, 2008

manganese dioxide (MnO2) have catalytic active sites for hydrogen peroxide sensing, which is the product of many enzymatic reactions.4-6 Zinc oxide (ZnO) nanoparticles provided a favorable microenvironment for the immobilization of tyrosinase and facilitated electrostatic interactions between the enzyme and a biocompatible polymer, thus providing a simple immobilization procedure.3 Zirconia (ZrO2) was recently used for the immobilization of horseradish peroxidase (HRP) via electrodeposition and for the fabrication of a H2O2 biosensor.7 Titanium dioxide (TiO2), another semiconductor, showed excellent electrochemical activity toward H2O2, ascorbic acid, guanine, L-tyrosine, acetaminophen, and β-NADH8 and provided direct electron transfer ability for glucose oxidase (GOX), HRP,2,9-11 and hemoglobin,12 making this material a suitable candidate for the fabrication of third-generation biosensors.8,12-14 The catalytic activity was enhanced by taking advantage of the photovoltaic effect of TiO2.12,15 Most TiO2-based materials used in biosensors were prepared by sol-gel methods.8,9,16-18 Herein, we exploited for the first time the oxygen storage capacity of another semiconductor, cerium dioxide (CeO2 or ceria) and mixed CeO2/TiO2 hybrid composites for the development of a highly sensitive, simple and inexpensive “oxygen rich” oxidase enzyme sensor. CeO2 is an important n-type semiconductor with a fluoride structure that has been used in heterogeneous catalysis, solid oxide fuel cells, gas sensing,19,20 and in environmental remediation for the catalytic oxidation of different hydrocarbons.21,22 (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20)

˘ljukic´, B.; Banks, C. E.; Compton, R. G. Nano Lett. 2006, 6,7, 1556–1558. S ˘ljukic´, B.; Compton, R. G. Electroanal. 2007, 19 (12), 1275–1280. S ˘ljukic´, B.; Banks, C. E.; Compton, R. G. Anal. Sci. 2007, Langley, C. E.; S 23, 165–170. Tong, Z.; Yuan, R.; Chai, Y.; Chen, S. J. Biotechnol. 2007, 128, 567–575. Curulli, A.; Valentini, F.; Padeletti, G.; Viticoli, M.; Caschera, D.; Palleschi, G. Sens. Actuators, B 2005, 111-112, 441–449. Vitcoli, M.; Curulli, A.; Cusma, A.; Kaciulis, S.; Nunziante, S.; Pandolfi, L.; Valentini, F.; Padeletti, G. Mater. Sci. Eng., C 2006, 26, 947–951. Zhang, Y.; He, P.; Hu, N. Electrochim. Acta 2004, 49, 1981–1988. Liu, S.; Chen, A. Langmuir 2005, 21, 8409–8413. Zhou, H.; Gan, X.; Wang, J.; Zhu, X.; Li, G. Anal. Chem. 2005, 77 (18), 6102–6104. Song, M.; Zhang, R.; Wang, X. Mater. Lett. 2006, 60, 2143–2147. Cosnier, S.; Senillou, A.; Gratzel, M.; Comte, P.; Vlachopoulos, N.; Renault, N. J.; Martelet, C. J. Electroanal. Chem. 1999, 469, 176–181. Zhou, H.; Liu, L.; Yin, K.; Liu, S.; Li, G. Electrochem. Commun. 2006, 8, 1168–1172. Chen, X.; Chen, G.; Dong, S. Analyst 2001, 126, 1728–1732. Carrara, S.; Bavastrello, V.; Ricci, D.; Stura, E.; Nicolini, C. Sens. Actuators, B 2005, 109, 221–226. Yu, J.; Liu, S.; Ju, H. Biosens. Bioelectron. 2003, 19, 509–514. Bamwenda, G. R.; Arakawa, H. J. Mol. Catal. A: Chem. 2000, 161, 105– 113. Manorama, S. V.; Izu, N.; Shin, W.; Matsubara, I.; Murayama, N. Sens. Actuators, B 2003, 89, 299. 10.1021/ac800808a CCC: $40.75  2008 American Chemical Society Published on Web 08/23/2008

CeO2 and doped CeO2 films have free radical scavenging properties, a property that has received significant attention in biology and medicine.23 Under a redox environment, cerium might interchange its oxidation state Ce3+/Ce4+ and shift between CeO2 and Ce2O3. The useful catalytic properties of ceria were attributed to its ability to release or store oxygen in its cubic structure due to a high oxygen mobility at its surface and a large diffusion coefficient of the oxygen vacancy as compared to other metal oxides.24-26 In addition to pure ceria, mixed ceria-based metal oxides have been shown to exhibit increased oxygen vacancies due to the formation of defective structures, resulting in new catalysts with enhanced physicochemical properties and a greater catalytic activity.26 As a result, mixed ceria-based oxide systems have even higher electrocatalytic activities and oxygen storage capacity. Such examples include binary and tertiary mixtures of CeO2/TiO2,26,27 CeO2/ZrO2,28 and CeO2/ZrO2/TiO2.29 These characteristics suggest that ceria and mixed ceria-titania oxides could be potentially used as “oxygen-rich” electrode materials for oxidase enzymes that could avoid or minimize problems associated with variations in the oxygen level for enzymes which are using oxygen as a cosubstrate, offering possibilities for operation in “oxygen-free” environments. Other materials that were used for this purpose include fluorocarbon pasting liquids and Kel-F poly(chlorotrifluoroethylene).30-32 In this work, we first studied the electrochemical characteristics of the CeO2/TiO2 modified glassy carbon electrode (GCE). Then, we conducted cyclic voltammetric investigations of the enzymatic oxidation of phenol in the presence and absence of molecular oxygen with and without metal oxides in the immobilization matrix. This provided insights into the electrochemical activity of CeO2 and CeO2/TiO2 and their advantages in enzymatic oxidation and electrochemical detection. Finally, on the basis of this principle, we developed and characterized an amperometric tyrosinase biosensor for the determination of the neurotransmitter dopamine. EXPERIMENTAL SECTION Reagents and Materials. Tyrosinase (EC.1.14.18.1, 3900 U mg-1 of solid), phenol (analytical grade), chitosan (from crab shells; practical grade), and titanium dioxide (nanopowder) were (21) Lin, S. S.; Chen, C. L.; Chang, D. J.; Chen, C. C. Water Res. 2002, 36, 3009–3014. (22) (a) Hamoudi, S.; Larachi, F.; Cerrella, G.; Cassanello, M. Ind. Eng. Chem. Res. 1998, 37, 3561–3566. (b) Zengxiong, C.; Wanpeng, Z.; Shaoxia, Y.; Jianbing, W. Chin. J. Catal. 2006, 27, 1073–1079. (23) (a) Tarnuzzer, R. W.; Colon, J.; Seal, S. Nano Lett. 2005, 5 (12), 2573– 2577. (b) Das, M.; Patil, S.; Bhargava, N.; Kang, J.-F.; Riedel, L. M.; Seal, S.; Hickman, J. J. Biomaterials 2007, 28, 1918–1925. (24) Dutta, P.; Pal, S.; Seehra, M. S.; Shi, Y.; Eyring, E. M.; Ernst, R. D. Chem. Mater. 2006, 18, 5144–5146. (25) Zhang, F.; Raitano, J. M.; Hanson, J. C.; Caliebe, W.; Khalid, S.; Chan, S. J. Appl. Phys. 2006, 99, 0843131–0843138. (26) Sinha, A. K.; Suzuki, K. J. Phys. Chem. B 2005, 109, 1708–1714. (27) Zhang, M.; Wang, H.; Wang, X.; Li, W. Mater. Des. 2006, 27, 489–493. (28) Oh-hori, I.; Oh-hori, N.; Itou, M.; Shin, W.; Matsubara, I.; Murayama, N. Sens. Actuators, B 2005, 108, 238–243. (29) Rajabbeigi, N.; Elyassi, B.; Khodadadi, A.; Mohajerzadeh, S.; Mortazavi, Y.; Sahimi, M. Sens. Actuators, B 2005, 108, 341–345. (30) Wang, J.; Chen, L.; Chatrathi, M. P. Anal. Chim. Acta 2000, 411, 187– 192. (31) Wang, J.; Li, S.; Mo, J.-W.; Porter, J.; Musameh, M. M.; Dasgupta, P. K. Biosens. Bioelectron. 2002, 17, 990–1003. (32) Zhao, M.; Hibbert, D. B.; Gooding, J. J. Biosens. Bioelectron. 2003, 18, 827–833.

purchased from Sigma (St. Louis MO) and used as received. Ceria particles were prepared according to a procedure described in the literature.33 The synthesis and the structural and morphological characterization of these particles were reported previously.33 Sodium phosphate monobasic and methanol were from Fisher Scientific. Sodium phosphate (dibasic, anhydrous) and acetic acid (glacial) were from J. T. Baker (Phillipsburg, NJ). All experiments involving tyrosinase were carried out in 0.1 mol L-1 phosphate buffer solution (PB) at an optimized pH of 6.5.34,35 Distilled, deionized water (Millipore, Direct-Q system) with a resistivity of 18.2 MΩ was used for the preparation of the buffer solutions. Instrumentation. Cyclic voltammetry and amperometric experiments were performed with an Epsilon potentiostat (Bioanalytical Systems Inc., West Lafayette, IN). All experiments were carried out using a conventional electrochemical cell equipped with a Ag/AgCl/3 M NaCl (BAS MF-2052, RE-5B) as the reference electrode, a platinum wire (BAS MW-1032) as the counter electrode, and a modified glassy carbon electrode (GCE) (BAS MF-2012 with a diameter of 3.0 mm) as the working electrode. All the potentials were referred to the Ag/AgCl reference electrode. Procedures. Biosensor Principle and Experimental Design. Taking advantage of the properties of ceria and mixed CeO2/TiO2 oxides, we fabricated novel “oxygen rich” biosensors based on these materials. To demonstrate this concept, we selected a model oxidase enzyme, polyphenol oxidase, or tyrosinase. Tyrosinase is a well-known copper-protein which utilizes molecular oxygen to catalyze the oxidation of phenolic compounds in a two step process: ortho-hydroxylation of monophenols to ortho-diphenols, and a rapid, two-electron oxidation of ortho-diphenols into orthoquinones.36 We describe a very simple and direct one-step procedure for immobilization of tyrosinase in a metal oxide hybrid composite onto a glassy carbon electrode (GCE). For this purpose, we used a positively charged natural biopolymer, chitosan (Chit) as a binder. The biosensor principle is schematically presented in Figure 1, showing the ceria and titania oxides embedded within the Chit film and the enzymatic reaction. Chit is widely available and possesses several other desirable properties such as excellent film-forming ability, chemical inertness, biocompatibility, good adhesion, high mechanical strength, and hydrophilicity. Chit has been intensively used for the immobilization of enzymes, and numerous chitosan-based-enzyme sensors have been reported in the literature.37-4142-49 (33) Andreescu, D.; Matijevic´, E.; Goia, D. V. Colloids Surf., A: Physicochem. Eng. Aspects 2006, 291, 93–100. (34) Njagi, J.; Andreescu, S. Biosens. Bioelectron. 2007, 23 (2), 168–175. (35) Andreescu, S.; Avramescu, A.; Bala, C.; Magearu, V.; Marty, J.-L. Anal. Bioanal. Chem. 2002, 374, 39–45. (36) Solomon, E. I.; Sundaram, U. M.; Machonkin, T. E. Chem. Rev. 1996, 96 (7), 2563–2606. (37) Yang, X.; Dai, T.; Wei, M.; Lu, Y. Polymer 2006, 47, 4596–4602. (38) Yang, M.; Yang, Y.; Liu, B.; Shen, G.; Yu, R. Sens. Actuators, B 2004, 101, 269–276. (39) Miao, Y.; Chia, L. S.; Goh, N. K.; Tan, S. N. Electroanalysis 2001, 13, 347– 349. (40) Meshali, M. M.; Gabr, K. E. Int. J. Pharm. 1993, 89, 177–181. (41) Krajewska, B. Enzyme Microb. Technol. 2004, 35, 126–139. (42) Kumar, M. N. V. R. A. React. Funct. Polym. 2000, 46 (1), 1–27. (43) Taqieddin, E.; Amiji, M. Biomaterials 2004, 25, 1937–1945. (44) Vasquez-Duhalt, R.; Tinoco, R.; D’Antonio, P.; Topoleski, L. D. T.; Payne, G. F. Bioconjugate Chem. 2001, 12, 301–306.

Analytical Chemistry, Vol. 80, No. 19, October 1, 2008

7267

solution and shaken vigorously for 2 min. The composite consisting of TiO2 and CeO2 is denoted as CeO2/TiO2/Chit. A volume of 5 µL of this aqueous dispersion was casted onto the surface of a clean GCE and allowed to dry for 40 min to fabricate the CeO2/ TiO2/Chit-GCE electrode. Biosensors involving TiO2, CeO2, and Chit were made using a similar procedure. These sensors were denoted as follows: Chit-GCE for the control electrode in the absence of metal oxides, CeO2/Chit-GCE in the absence of TiO2, and TiO2/Chit-GCE in the absence of CeO2. The electrodes were thoroughly rinsed with distilled water and stored in PB when not in use.

Figure 1. Schematic diagram illustrating the design and operation principle of the mixed ceria-based metal oxide tyrosinase sensor.

Electrode Preparation and Electrochemical Studies. Before modification, the GCE working electrode was polished with 0.3 µm alumina powder and sonicated for at least 10 min in distilled water. Then, the electrode was rinsed alternatively with methanol and water and dried under a nitrogen stream prior to modification with the metal oxide hybrid composite. Cyclic voltammetry experiments were carried out in unstirred solutions in the presence or absence of oxygen at a scan rate of 100 mV s-1. For measurements in deoxygentaed solutions, oxygen removal was accomplished by purging a stream of nitrogen for at least 40 min before the test. For the amperometric tests in deoxygenated solutions, a nitrogen atmosphere was maintained over the solution during the experiments. Determination of phenol and dopamine was carried out using amperometry in stirred solutions with the three electrode system. All amperometric measurements were carried out at a fixed potential vs Ag/AgCl in PB (0.1 M) at pH 6.5 in a standard BAS electrochemical cell (4 mL). The reproducibility of the biosensors was evaluated by measuring the response of six identical electrodes prepared under the same experimental conditions. The operational stability was assessed by repetitive measurement using the same amount of substrate with a final concentration of 3 × 10-6 M. Between measurements the cell was rinsed with PB. All experiments were carried out at room temperature. Preparation of the Metal Oxide Hybrid Composite and Biosensor Fabrication. 1.0% chitosan (w/v) was prepared by dissolving 1 g of Chit flakes in 0.05 mol L-1 acetic acid and stirred for 2 h at room temperature. The solution was filtered through a 5-10 µm filter paper from Fischer Scientific. TiO2/CeO2 dispersions were prepared by adding 1 mg of TiO2 nanopowder and 1 mg of CeO2 in 1 mL of distilled water. The mixture was sonicated for 1 h. Then, 8 µL of the white dispersion was mixed with 4 µL of 200 U/µL tyrosinase (prepared in 0.1 M PB pH 6.5) and 8 µL of Chit (45) Wei, X.; Cruz, J.; Gorski, W. Anal. Chem. 2002, 74, 5039–5046. (46) Fernandes, R.; Wu, L.-Q.; Chen, T.; Yi, H.; Rubloff, G. W.; Ghodssi, R.; Bentley, W. E.; Payne, G. F. Langmuir 2003, 19, 4058–4062. (47) Kumar, G.; Smith, P. J.; Payne, G. F. Biotechnol. Bioeng. 1999, 63 (2), 154–165. (48) Yi, H.; Wu, L.-Q.; Bentley, W. E.; Ghodssi, R.; Rubloff, G. W.; Culver, J. N.; Payne, G. F. Biomacromolecules 2005, 6 (6), 2881–2894. (49) Wu, L.-Q.; Payne, G. F. Trends Biotechnol. 2004, 22 (11), 593–599.

7268

Analytical Chemistry, Vol. 80, No. 19, October 1, 2008

RESULTS AND DISCUSSION The biosensor described in this work was first optimized and characterized with respect to the role, effect, and content/ratio of metal oxides and presence/absence of oxygen in the reaction medium. For these optimization studies, we used phenol, a tyrosinase substrate. Then, the optimized biosensor was applied for the detection of the neurotransmitter dopamine. Cyclic Voltammetric Studies. In order to evaluate the role of metal oxides in the immobilization matrix, cyclic voltammetric (CV) experiments were first carried out only with the Chit modified GCE electrode. Figure 2A shows the CV responses of the tyrosinase-based Chit biosensor in oxygenated and deoxygenated solutions in PB with and without 2 × 10-5 M phenol. In oxygenated solutions, an increase in the reduction current is observed starting from 100 mV with a maximum current at ∼-100 mV vs Ag/AgCl. This increase is due to the reduction of quinone, generated in the tyrosinase catalyzed reaction in the presence of molecular oxygen, to catechol (Figure 1). In oxygen-free conditions, the CV shows a small shoulder at ∼0 mV, indicating that the enzymatic reaction takes place to a limited extend. In contrast, under the same experimental conditions, in deoxygenated solutions, the addition of CeO2/TiO2 in the Chit layer results in a welldefined voltammetric peak with a maximum current at ∼0 V (Figure 2B). These results clearly show that the presence of CeO2/TiO2 promotes the enzymatic reaction even in the absence of oxygen. We speculate that the CeO2/TiO2 provides sufficient oxygen for the enzymatic reaction to occur due to its high oxygen storage capacity.26,27 To identify the role of each metal oxide in the matrix in the absence of oxygen, the CV responses of the biosensors fabricated with each of the oxides were compared (Figure 2C). It can be clearly seen that both CeO2 and TiO2 facilitate the enzymatic reaction in deoxygenated conditions. However, the reduction current is significantly amplified when the mixed TiO2/CeO2 was used. This behavior can be attributed to the higher oxygen storage capacity of the mixed ceria-based metal oxides26-29 providing a rich internal source of oxygen for the enzymatic reaction. Moreover, the TiO2 is also known to enhance the electrocatalytic activity,13-16 further enhancing the electrochemical signal. The reduction peak for the TiO2/Chit enzyme electrode is shifted toward more positive values with a maximum current at ∼70 mV, while for both the CeO2/Chit and CeO2/TiO2/Chit the maximum peak appears at ∼0 V. Amperometric Measurements in the Presence and Absence of Oxygen. Biosensor Calibration, Sensitivity, Detection Limit, and Response Time. The four different biosensor configurations were compared for assessing the role of the CeO2 and CeO2/

Figure 3. Calibration curves of CeO2/TiO2/Chit, CeO2/Chit, TiO2/ Chit, and Chit biosensors in oxygenated solutions.

Figure 2. Cyclic voltammograms of the tyrosinase modified GCE electrode based on (A) Chit and (B) CeO2/TiO2/Chit in the absence (N2, O2) and presence (N2/phenol, O2/phenol) of 2.0 × 10-5 M phenol in O2 saturated (O2, O2/phenol) and N2 saturated (N2, N2/phenol) 0.1 M PB. (C) Voltammograms of the tyrosinase-based Chit (chit), Chit/ TiO2 (TiO2), Chit/CeO2 (CeO2), and Chit/CeO2/TiO2 (CeO2/TiO2) GCE electrode in the presence of 2.0 × 10-5 M phenol in N2 saturated 0.1 M PB. The pH was 6.5, and the scan rate was 100 mV s-1.

TiO2 as an internal oxygen supply for the tyrosinase catalyzed reaction. Figure 3 shows the amperometric responses to successive additions of 2 × 10-6 M phenol and the corresponding calibration curves of the four biosensors in oxygenated conditions. The analytical characteristics of these sensors are summarized in Table 1. The detection limits provided in Table 1 were calculated according to 3Sb/m criteria where m is the slope of the linear calibration plot and Sb is the standard deviation (n ) 3) of the

amperometric signals obtained with a phenol sample corresponding to the lower detectable concentration. Incorporating CeO2/TiO2 in the immobilization matrix increased both the sensitivity and linear range of the biosensor by 2-fold while the detection limit was lowered with almost 2 orders of magnitude as compared to the control Chit electrode. The linear range for the CeO2/TiO2/Chit was (1.0 × 10-8)-(5.0 × 10-5) M as compared to (1.5 × 10-7)-(2.0 × 10-5) M for the Chit GCE. The CeO2/TiO2 biosensor was 18 times more sensitive than that reported by Chen et al. using a TiO2/grafting copolymer of poly(vinyl alcohol) with 4-vinylpyridine GCE.16 When only CeO2 or TiO2 was used instead of the binary oxide mixture, the biosensor was more sensitive than the Chit control. The sensitivity of the four biosensors varied in the following order: CeO2/TiO2/ Chit > CeO2/Chit > TiO2/Chit > Chit. It is clear from these results that the use of the mixed metal oxides in the immobilization matrix provided enhanced performance including higher sensitivity, lower detection limit, and extended linear range. These could be linked to their superior electrocatalytic activity and oxygen storage capabilities. Operation in Oxygen-Free Environment. Figure 4 compares the amperometric response to successive additions of phenol in the presence and absence of oxygen using the CeO2/TiO2/ Chit and the control Chit electrodes. In the absence of oxygen, the Chit sensor showed almost negligible responses to the substrate additions. In the same conditions, in deoxygenated solutions, the CeO2/TiO2/Chit biosensor shows well defined amperometric responses. The magnitude of the current and the sensitivity of the sensor were slightly lower than in the presence of oxygen but higher than that observed with the Chit control electrode (Table 1). There were no major differences between the detection limit values for the CeO2/TiO2 and CeO2based biosensors in the presence and absence of oxygen, indicating that both provide comparable capabilities for detecting low substrate concentrations in the nanomolar range. The sensitivity of the four biosensors follows the same order as in the presence of oxygen (Table 1). The response time quantified as t90%, for all four biosensor configurations, was very fast (less than 2 s (±0.015) for n ) 3 sensors × 3 determinations), indicating that their response is mainly determined by enzyme kinetics and not by diffusion Analytical Chemistry, Vol. 80, No. 19, October 1, 2008

7269

app Table 1. Analytical Performance and KM Constant of Tyrosinase Biosensors for Phenol Based on Different CeO2 and TiO2 Metal Oxides in Oxygenated and Deoxygenated Buffer Solutions at pH 6.5a

N2 saturated

O2 saturated biosensor configuration TiO2/CeO2/Chit CeO2/Chit TiO2/Chit Chit a

detection limit (M) 9.0(±0.1) 7.8(±0.5) 1.0(±0.2) 1.5(±0.3)

× × × ×

-9

10 10-9 10-8 10-8

sensitivity (mA M-1)

app KM (µM)

detection limit (M)

86.21 ± 1.80 57.19 ± 1.08 55.53 ± 1.47 37.41 ± 1.72

65.84 ± 2.41 103.17 ± 2.48 145.96 ± 10.30 234.36 ± 8.96

5.6(±0.1) × 10 9.0(±0.3) × 10-9 1.2(±0.4) × 10-8 no response -9

sensitivity (mA M-1)

app KM (µM)

64.73 ± 0.91 43.77 ± 1.56 37.47 ± 5.56

6.63 ± 1.74 8.12 ± 1.02 8.72 ± 0.50

Average values ± SD obtained with n ) 3 biosensors.

approach could be used to evaluate the responsiveness of these biosensors. For a two-substrate enzyme, such as the case of tyrosinase, the KM is more complex since it incorporates both substrate (phenol) and cosubstrate (O2) dependency. Experimenapp tally, the apparent Michaelis constant, KM was determined by measuring the initial reaction rates with phenol substrate in PB at pH 6.5 at room temperature from triplicate measurements, under the assumption that the reaction rate is proportional to the reduction current. Measurements in the absence of oxygen were carried out by continuously purging nitrogen during the experiapp ment. The apparent KM constant was then calculated from the corresponding Lineweaver-Burk representation.56,57 app

-7

Figure 4. Amperometric responses to successive 1 × 10 M phenol additions in the absence (N2) and presence (O2) of oxygen using the tyrosinase-based CeO2/TiO2/Chit and Chit biosensors. Experimental parameters: 0 V applied potential, phosphate buffer at pH 6.5. The oxygen was removed by purging the solution with N2 for at least 40 min and maintaining a N2 flow over the solution during measurements.

through the hybrid composite matrix. This is due to the highly permeable polyelectrolyte layer facilitating rapid mass transfer of the substrate to the catalytic active centers. The response time of these sensors is much lower than that reported for tyrosinase immobilized in other polymeric films such as polypyrrole (∼30-60 s)14,50-54 or in a carbon paste (∼6 min)51,55 for which the response was strongly affected by diffusion limitations. Kinetic Parameters. To obtain quantitative information on the effect of the metal oxide microenvironment on enzyme activity, catalytic efficiency and substrate affinity, the apparent Michaelis app constant of immobilized enzyme, KM was determined using the Michaelis-Menten approach in oxygenated and deoxygenated conditions. Since the substrate diffusion seems not to be a limiting factor (the response time was only 2 s), a simple Michaelis-Menten (50) Bo ¨yu ¨ kbayram, A. E.; Kıralp, S.; Toppare, L.; Yac˘, Y. Bioelectrochemistry 2006, 69, 164–171. (51) Mailley, P.; Cummings, E. A.; Mailley, S.; Cosnier, S.; Eggins, B. R.; McAdams, E. Bioelectrochemistry 2004, 63, 291–296. (52) Cosnier, S. Biosens. Bioelectron. 1999, 14, 443–456. (53) Marin-Zamora, M. E.; Rojas-Melgarejo, F.; Garcia-Canovas, F.; Garcia-Ruiz, P. A. J. Biotechnol. 2007, 131, 388–396. (54) Rijairavanich, P.; Aoki, K.; Chen, J.; Surareungchai, W.; Somasundrum, M. Electroanalysis 2004, 16 (8), 605–611. (55) Mita, D. G.; Attanasio, A.; Arduini, F.; Diano, N.; Grano, V.; Bencivenga, U.; Rossi, S.; Aminee, A.; Mosconec, D. Biosens. Bioelectron. 2007, 23, 60–65.

7270

Analytical Chemistry, Vol. 80, No. 19, October 1, 2008

KM 1 1 1 ) + Is Imax Imax C

(1)

where Is is the steady state current obtained after adding substrate, C is the substrate concentration, and Imax is the maximum current in substrate saturation conditions. app The KM for the immobilized tyrosinase in the CeO2/TiO2/ Chit composite in oxygenated buffer solutions (6.58(±0.24) × 10-5 M, n ) 3) decreased as compared to that of the control Chit electrode (2.34(±0.09) × 10-4 M). These are lower than the published value for the free enzyme using phenol as a substrate (7.0 × 10-4 M).58 This could indicate a higher catalytic efficiency and affinity for substrate of the immobilized tyrosinase in the presence of CeO2/TiO2 and could be related to a higher concentration of cosubstrate on the mixed metal oxides close to the enzyme as compared to the concentration in bulk solution. Lower app KM values for immobilized enzymes are quite unusual, although this trend has also been observed for other tyrosinase catalyzed reactions with the enzyme immobilized in other matrixes.53,58,59 Our calculated value is close to that reported by Rijiravanich et al.54 who immobilized tyrosinase onto a Pt disk using polystyrene latex particles (6.4 × 10-5 M) and by Liu et al.58 who utilized a carbon paste modified with colloidal gold (5.36(±0.3) × 10-5 M). A similar decrease has also been reported by Zhou et al.15 for a tyrosinase-based GCE electrode with the enzyme immobilized in app a TiO2/agarose matrix (1.04 × 10-5 M). Changes in KM and differences between kinetic constants of immobilized and free enzymes have been related to a variety of factors including substrate binding efficiency, conformational changes, diffusion (56) Nelson, D. L.; Cox, M. M. Lehninger Principles of Biochemistry, 4th ed.; W. H. Freeman and Company: New York, 2005; pp 202-212. (57) Li, J.; Tan, S. N.; Ge, H. Anal. Chim. Acta 1996, 335, 137–145. (58) Liu, S.; Yu, J.; Ju, H. J. Electroanal. Chem. 2003, 540, 61–67. (59) Du, Y.; Luo, X.-L.; Xu, J.; Chen, H. Bioelectrochemistry 2006, 71, 44.

Figure 5. Effect of CeO2/TiO2 mass ratio (w/w) on the amperometric response of the tyrosinase biosensor (mean values ( SD for n ) 3 biosensors).

barriers to substrate/enzyme binding, nature of the immobilization material generating a different microenvironment of the enzyme (hydroplilicity, hydrophobicity, etc.), as well as different substrate concentrations in the immobilization layer.51-53 For oxidase enzymes, as with tyrosinase in this study, changes in the kinetic parameters could also be explained by changes in oxygen app demand.60 The variation in the KM observed in our work could be explained by the kinetic model described by Rijiravanich et al.54 and by Marin-Zamora et al.,53 who attributed this behavior to an increase in the local substrate concentration. app Surprisingly, in deoxygentated buffer solutions, the KM values, calculated from the linear range, were 1 order of magnitude lower than those obtained in the presence of oxygen for all four biosensor configurations. In these conditions, the enzyme reaction is using the internal oxygen supply retained onto the CeO2/TiO2 located on the electrode surface in the close proximity of the enzyme, thus increasing the possibility for substrate binding. Optimization of the Metal Oxide Content. The performance of the biosensors was optimized with respect to metal oxide loading in terms of mass ratios of TiO2/CeO2 that gave the higher amperometric response to 3 × 10-6 M phenol. Biosensors containing mass ratios of 0:8, 2:6, 3:5, 4:4, 5:3, 6:2, and 8:0 (w/w) CeO2/TiO2 were fabricated and tested for this purpose while maintaining the Chit amount in the immobilization matrix constant. The results are presented in Figure 5. The higher amperometric response was obtained with a 1:1 TiO2/CeO2 ratio. Increasing the amount of CeO2 or TiO2, or their use as single metal in the chitosan matrix, does not result in higher current intensity. Biosensor Stability and Reproducibility. The stability and reproducibility of the various biosensors were evaluated by measuring the amperometric response to consecutive additions of 3 × 10-6 M phenol, with washing the cell between tests. Figure 6 illustrates the operational stability of all four biosensors tested. No significant differences in terms of operational stability of the different configurations were observed, with all retaining a residual response of 71-75% after 50 continuous assays. All biosensors were stable for ∼15 continuous assays (with less than 5% variation), which is better than that reported previously with other phenol biosensors.50 This extended stability can be attributed to (60) McMahon, C. P.; Rocchitta, G.; Serra, P. A.; Kirwan, S. M.; Lowrr, J. P.; O’Neill, R. D. Anal. Chem. 2006, 78, 2352–2359.

Figure 6. Operational stability of the tyrosinase modified CeO2/TiO2/ Chit, CeO2/Chit, TiO2/Chit, and Chit biosensors. Experimental parameters: 1 × 10-6 M phenol in 0.1 M PB at pH 6.5.

Figure 7. Cyclic voltammograms of tyrosinase-based CeO2/TiO2/ Chit and Chit in the presence and absence of 2.0 × 10-5 M dopamine in oxygenated PBS, 0.1 M (pH 6.5) at a scan rate of 100 mV/s.

the biocompatibility of the chitosan.41 The observed decrease in activity can be due to the adsorption (or polymerization) of quinone onto the electrode surface. The reproducibility of the TiO2/CeO2/Chit enzyme sensor was evaluated for six identical electrodes prepared independently on different days following the same experimental protocol. The average steady-state current for additions of 3 × 10-6 M phenol was found to be 1.71 µA with a relative standard deviation of 1.25%. The long-term stability of the biosensor was evaluated by measuring the sensitivity of the biosensors to phenol every 2 days. The sensors were washed between measurements and stored at 4 °C in PB solution at pH 6.5. Results showed that the enzyme electrode preserved the same apparent activity for the first 7 days and retained 75% of its initial activity after 2 weeks storage. Similar stability data were reported in the literature for other tyrosinase biosensors.61 The oxygen storage capacity of the TiO2/CeO2 sensor was tested under continuous operation conditions for 4 h in a N2 saturated buffer solution. The biosensor can operate continuously for about 50 min, with a signal retention of ∼94%. After 250 min, the sensor response Analytical Chemistry, Vol. 80, No. 19, October 1, 2008

7271

Figure 8. Effect of operating potential of the tyrosinase-based CeO2/ TiO2/Chit and Chit electrodes in oxygenated buffer solutions to additions of 5.0 × 10-5 M (standard deviation from n ) 3 measurements).

was 58% of the initial value, while the biosensor in the absence of TiO2/CeO2 showed no response. Application for Detection of the Neurotransmitter Dopamine. These materials could be particularly important and useful in fabricating biosensors for in vivo monitoring. Relevant to this application we present here the utilization of the optimized biosensor for the determination of dopamine (DA), a major neurotransmitter in the brain which is involved in cognition, locomotion, and motivation,62 and for which measurements in in vivo under low-oxygen concentrations are required. Traditionally, DA is determined directly by electrochemical means, mainly with fast scan voltammetry.61-64 However, because of the relatively high applied potential (∼500 mV),65,66 the use of a direct electrochemical detection has three major limitations: (i) low sensitivity, (ii) electrode poisoning, and (iii) difficulties in differentiating between individual neurotransmitters and other coexisting species (e.g., ascorbic acid) due to the overlapping signals of oxidation potentials. These can be avoided by using an enzymatic approach with enzyme recycling using tyrosinase. Conventionally, tyrosinase is used to catalyze phenols and diphenols. More recently, other phenolic compounds such as polyphenol isoflavonoids and environmental phenolic estrogens have been reported as tyrosinase substrates and detected with tyrosinase biosensors.67 Following the same principle, DA, which is a catechol-like phenolic compound, has also been determined.68-70 However, the sensitivity of tyrosinase biosensors toward dopamine (61) Valentini, F.; Palleschi, P.; Morales, E. L.; Orlanducci, S.; Tamburri, E.; Terranova, M. L. Electroanalysis 2007, 19 (8), 859–869. (62) Robinson, D. L.; Venton, B. J.; Heien, M. L. A. V.; Wightman, R. M. Clin. Chem. 2003, 94, 1763–1773. (63) Anastassiou, C. A.; Patel, B. A.; Arundell, M.; Yeoman, M. S.; Parker, K. H.; O’Hare, D. Anal. Chem. 2006, 78 (19), 6990–6998. (64) Venton, B. J.; Wightman, R. M. Anal. Chem. 2003, 1, 414–421. (65) Cao, X.; Luo, L.; Ding, Y.; Zou, X.; Bian, R. Sens. Actuators, B 2008, 129, 941–946. (66) Zou, X.; Luo, L.; Ding, Y.; Wu, Q. Electroanalysis 2007, 19 (17), 1840– 1844. (67) Andreescu, S.; Sadik, O. A. Anal. Chem. 2004, 76, 552–560. (68) Zhou, Y. L.; Tian, H.; Zhi, J. F. Biosens. Bioelectron. 2007, 22, 822–828. (69) Tsai, Y.-C.; Chiu, C. C. Sens. Actuators, B 2007, 125, 10–16. (70) Liu, A.; Honma, I.; Zhou, H. Biosens. Bioelectron. 2005, 21, 809–816.

7272

Analytical Chemistry, Vol. 80, No. 19, October 1, 2008

Figure 9. (A) Amperometric response of tyrosinase-based CeO2/ TiO2/Chit and Chit biosensors to successive additions of 2.50 × 10-4 M AA and 1.0 × 10-6 M DA. (B) Corresponding linear calibration curves for DA (n ) 3) in oxygenated solutions. (C) Amperometric response of CeO2/TiO2/Chit and Chit to successive additions of 1 × 10-6 M DA in oxygenated and deoxygenated PB solutions.

is significantly reduced due to the presence of an ethylamine substituent on the aryl ring, which affects the enzyme reaction kinetic. This decrease is characteristic for all diphenolic substrates with a bulky substituent on the aromatic ring61 due to the fact that these substrates need to undergo rearrangement at the copper site of the enzyme for simple electron transfer.36 As a result, strategies for increasing sensitivity of tyrosinase sensors are

app Table 2. Analytical Performance and KM Constant of Tyrosinase Based CeO2/TiO2/Chit and Chit Biosensors for Dopamine in Oxygenated and Deoxygenated Buffer Solutions at pH 6.5a

O2 saturated59

a

N2 saturated

biosensor configuration

detection limit (mol L-1)

sensitivity (mA M-1)

detection limit (mol L-1)

sensitivity (mA M-1)

TiO2/CeO2/Chit Chit

3.4(±0.1) × 10-8 1.9(±0.20) × 10-6

14.91 ± 0.50 9.71 ± 0.09

4.20(±0.10) × 10-8 no response

14.79 ± 0.10

Average values ± SD obtained with n ) 3 biosensors.

necessary in order to determine DA in useful concentrations ranges using this approach. The biosensor described in this work could provide this opportunity. In addition, because of the low operation potential, in the negative potential range, this sensor is expected to selectively determine DA in the presence of common interferences such as ascorbic acid, while also avoiding electrode fouling. Figure 7 shows the CV responses of the tyrosinase-based Chit and CeO2/TiO2/Chit biosensors with and without 2.0 × 10-5 M DA. No visible peak was observed with the Chit electrode when DA was added in the reaction cell. This CV was almost superimposed to that of Chit in the absence of DA. This observation is in agreement with data reported by Liu et al.70 However, when CeO2/ TiO2 was included in the immobilization matrix, a well-defined redox peak was observed, with a reduction potential with a maximum at ∼-150 mV. This peak increased proportionally with the concentration of DA in the reaction medium (results not shown) suggesting that tyrosinase catalyzes the oxidation of DA and that the CeO2/TiO2 facilitates the reduction of the enzymatically generated quinone. Figure 8 shows the effect of the operating potential on the amperometric responses of the CeO2/TiO2/Chit and Chit biosensors for injections of 5.0 × 10-5 M DA. This parameter is important to ensuring good sensor sensitivity. In addition, low operating potentials can reduce interferences from electroactive species. The applied potentials were tested in the range of -300 to 50 mV vs Ag/AgCl reference electrode where the reduction of the enzymatically generated quinone occurs. Results showed that in both the presence and absence of oxygen, applied potentials lower than -150 mV ensured higher intensity of current. In the absence of oxygen, the same trend was observed (results not shown). The effect of ascorbic acid (AA), the major interfering species in biological media, was also tested. Figure 9A shows the typical amperometric response of the CeO2/TiO2/Chit and Chit electrodes to the injection of 2.5 × 10-4 M AA and 6.5 × 10-6 M DA at an operating potential of -150 mV. No amperometric responses were obtained when AA was added to the reaction cell, even at high concentrations. When DA was present, well defined responses were observed. Figure 9B shows the corresponding calibration curves in oxygenated conditions. With the optimized CeO2/TiO2/Chit biosensor, in oxygenated conditions, the detection limit for dopamine was 3.4(±0.1) × 10-8 M with a sensitivity of 14.9(±0.5) mA M-1, while for the Chit electrode these parameters were 1.9(±0.2) × 10-6 M and 9.71(±0.09) mA M-1 respectively (Table 2). The sensitivity of the CeO2/TiO2/Chit is similar to that reported with a tyrosinase electrode fabricated using multiwalled carbon nanotubes (12 mA M-1).68 In the absence of oxygen, the sensitivity of the CeO2/TiO2/Chit electrode was 14.8(±0.1) mA

M-1 with a detection limit of 4.2(±0.1) × 10-8 M. No response was observed for 2.5 × 10-4 M AA. This sensitivity is comparatively higher than that of the control Chit electrode. The kinetic parameters show a similar trend as obtained with phenol as a substrate, with the corresponding difference due to the fact that tyrosinase has a lower affinity for DA. The app calculated KM for dopamine (7.99 × 10-4 M) was 1 order of magnitude higher compared to that of phenol (0.66 × 10-4 M). This difference can be attributed to diffusion limitations because dopamine is a relatively large molecule as compared to phenol.67 We further evaluated the selectivity of the biosensor toward serotonin (ST), another phenolic neurotransmitter present in the brain, together with DA.62 Under the same experimental conditions, the biosensor showed no response to ST. We attributed this effect to the bulkier ST molecule. This information could be potentially useful to design an enzyme biosensor to selectively discriminate between the two neurotransmitters, ST and DA. Further improvements of this system are to adapt this design to a microelectrode format and expand the same principle to other oxidase enzymes. CONCLUSION We demonstrated that mixed TiO2/CeO2 hybrid composites provide enhanced analytical characteristics to tyrosinase biosensors including high sensitivity and possibilities for operation in “oxygen-free” conditions. The optimized biosensor allowed detection of phenol in the nanomolar concentration range and of the neurotransmitter dopamine in the submicromolar concentrations with high selectivity with respect to ascorbic acid in both air-saturated and deoxygenated solutions. Moreover, the developed biosensors showed good reproducibility, high sensitivity, very short response time, and a wide linear concentration range. Application of these materials could be extended to other oxidase enzyme sensors. Of particular interest would be the development of enzyme sensors for the detection of other neurotransmitters such as glutamate. The enhanced analytical characteristics were attributed to the oxygen transport and storage properties and to the electrocatalytic activity of CeO2, which are enhanced by the presence of TiO2. With respect to DA, there are several distinct advantages in using this biosensor as compared to direct electrochemical oxidation. These include enhanced sensitivity and detection limit, reduced electrode fouling, and lower applied potential, which reduces the risk of interferences, as demonstrated in this study with ascorbic acid. Moreover, the materials proposed in this study are biocompatible and nontoxic and could be easily incorporated in microsensors for in vivo analysis. Recent progress in material science allowing development of various types and structures of CeO2, CeO2/TiO2, and Analytical Chemistry, Vol. 80, No. 19, October 1, 2008

7273

other metal oxides is expected to further enhance the characteristics of these biosensors. Some examples that could be used for this purpose include nano-CeO2 coating TiO2,27 nanotubes and nanowires,69,71 and mesoporous CeO2/TiO2.26 These could find applications in implantable sensors for in vivo monitoring of various oxidase enzyme substrates and in biofuel cells.

ACKNOWLEDGMENT This work was supported in part by NSF Grants 0804506 and 0727861 and Grant USDA-PUF 8000019748-01.

(71) Swamy, B. E. K.; Venton, B. J. Analyst 2007, 132, 876–884.

AC800808A

7274

Analytical Chemistry, Vol. 80, No. 19, October 1, 2008

Received for review April 22, 2008. Accepted July 21, 2008.