A Glucose Oxidase Electrode Based on Electropolymerized

Apr 4, 2000 - Citation data is made available by participants in Crossref's Cited-by Linking service. For a more comprehensive list of citations to th...
1 downloads 15 Views 140KB Size
Download PDF
Anal. Chem. 2000, 72, 2177-2181

A Glucose Oxidase Electrode Based on Electropolymerized Conducting Polymer with Polyanion-Enzyme Conjugated Dopant Won Jun Sung and You Han Bae*

Department of Materials Science and Engineering, Kwangju Institute of Science and Technology, 1 Oryong-dong, Puk-gu, Kwangju 500-712, Korea

An enzyme immobilization method has been developed by electropolymerization chemistry of conducting polymer which results in a more effective and reproducible enzyme electrode. As a model system, in this study, glucose oxidase (GOD) was conjugated with a polyanion, poly(2acrylamido-2-methylpropane sulfonic acid), via a poly(ethylene oxide) spacer to improve the efficiency of enzyme immobilization into a conducting polymer. GOD was successfully conjugated with a high conjugation yield of more than 90%, and its bioactivity was preserved. The resulting polyanion-GOD conjugate was used as a dopant for the electrochemical polymerization of pyrrole. Polypyrrole was effectively deposited on a Pt wire working electrode with the polyanion-GOD conjugate. The enzyme electrode responded to glucose concentrations of up to 20 mM with a sensitivity of 40 nA/mM at an applied potential of 0.4 V within a response time of 30 s. Although the response signal decreased at the low applied potential of 0.3 V, the enzyme electrode showed sensitive response signals of about 16 nA/mM up to 20 mM in glucose concentration. Under the deoxygenated condition, reduced but clear response current signal was obtained. The results show that the current signal response of the enzyme electrode to glucose concentration may be produced by mixed mechanisms. Since Clark and Lyons1 developed the first enzyme-based oxygen electrode, biosensors have been studied and analyzed intensively.2-5 Enzymes have been the most frequently used biomolecules, due to their superior selectivity and faster catalytic reaction rates. There have been a variety of approaches taken in the development of enzyme-immobilization techniques. The physical entrapment of enzymes has been a common means of enzyme immobilization. In early enzyme-electrode models, an outer holding * Corresponding author: (Tel) +82-62-970-2361. (Fax) +82-62-970-2304. (E-mail) [email protected] (1) Clark, L. C.; Lyons, C. Ann. N.Y. Acad. Sci. 1962, 102, 29-45. (2) Gregg, B. A.; Heller, A. Anal. Chem. 1990, 62, 258-263. (3) Zambonin, C. G.; Losito, I. Anal. Chem. 1997, 69, 4113-4119. (4) Anzai, J.; Takeshita, H.; Kobayashi, Y.; Osa, T.; Hoshi, T. Anal. Chem. 1998, 70, 811-817. (5) Cosnier, S.; Innocent, C.; Allen, L.; Poitry, S.; Tsacopoulos, M. Anal. Chem. 1997, 69, 968-971. 10.1021/ac9908041 CCC: $19.00 Published on Web 04/04/2000

© 2000 American Chemical Society

polymer membrane was used to prevent the enzyme from undergoing diffusional loss.6-9 A more advanced method of physically entrapping enzymes was through matrixes made of synthetic10,11 or natural gels.12,13 In physical entrapment methods, due to the long electron-transfer pathway between enzymes and electrodes, electron mediators such as ferrocene derivatives14 and osmium complexes15 were often co-immobilized with the enzyme. Thus, a serious problem with such methods was the leakage of these harmful electron mediators when the enzyme electrode was used as an implantable biosensor.16 Moreover, although these techniques are preferable for enzyme stability, they suffer from drawbacks such as complexity of the fabrication procedure and difficulties in miniaturization. Other enzyme immobilization techniques have been investigated for more reliable enzyme attachment. Such methods include the covalent bonding of enzymes to a modified substrate by using a coupling agent such as carbodiimide17,18 and the use of a covalent cross-linking agent such as glutaraldehyde.19,20 These techniques offer the benefits of simplicity of fabrication and stable enzyme attachment. However, a covalent attachment may interfere with enzyme activity, and additional electron mediators are required due to the low sensitivity of the enzyme electrode. In fact, functionally reproducible, uniform coatings for active enzymes on an electrode have yet to be fabricated in miniaturizable size. (6) Campanella, L.; Pacifici, F.; Sammartino, M. P.; Tomassetti, M. Bioelectrochem. Bioenerg. 1998, 47, 25-38. (7) Saby, C.; Male, K. B.; Luong, J. H. T. Anal. Chem. 1997, 69, 4324-4330. (8) Csoregi, E.; Quinn, C. P.; Schmidtke, D. W.; Lindquist, S. E.; Pishko, M. V.; Ye, L.; Katakis, I.; Hubbell, J. A.; Heller, A. Anal. Chem. 1994, 66, 31313138. (9) Svorc, J.; Miertus, S.; Katrlik, J.; Stredansky, M. Anal. Chem. 1997, 69, 2086-2090. (10) Bu, H.; Mikkelsen, S. R.; English, A. M. Anal. Chem. 1998, 70, 43204325. (11) Tatsuma, T.; Saito, K.; Oyama, N. Anal. Chem. 1994, 66, 1002-1006. (12) Liu, Y.; Liu, H.; Qian, J.; Deng, J.; Yu, T. J. Chem. Technol. Biotechnol. 1995, 64, 269-276. (13) Gondo, S.; Kim, C.; Hirata, S.; Morishta, M. Biosens. Bioelectron. 1997, 12, 395-401. (14) Schmidtke, D. W.; Heller, A. Anal. Chem. 1998, 70, 2149-2155. (15) Bu, H.; English, A. M.; Mikkelsen, S. R. Anal. Chem. 1996, 68, 39513957. (16) Wilkins, E.; Atanasov, P. Med. Eng. Phys. 1996, 18, 273-288. (17) Wilkins, E.; Atanasov, P.; Muggenburg, B. A. Biosens. Bioelectron. 1995, 10, 485-494. (18) Hsiue, G. H.; Wang, C. C. Biotechnol. Bioeng. 1990, 36, 811-815. (19) Mayer, M.; Ruzicka, J. Anal. Chem. 1996, 68, 3808-3814. (20) Appleton, B.; Gibson, T. D.; Woodward, J. R. Sensors and Actuators, B 1997, 43, 65-69.

Analytical Chemistry, Vol. 72, No. 9, May 1, 2000 2177

Recently, enzyme immobilization during the electropolymerization of conductive polymers has attracted wide interest.21-23 The technique offers the advantage of uniform and reproducible immobilization of enzymes over a small area. Furthermore, electropolymerization chemistry allows one to control the thickness of the coating and, thus, the amount of enzyme immobilized, by changing the polymerization conditions.24,25 The mechanisms for enzyme immobilization by in situ electrochemical polymerization are based on physical entrapment within the polymer matrixes during electropolymerization26,27 and the use of negatively charged enzymes as a dopant for conductive polymers.28,29 However, whether the mechanism is entrapment or dopant utilization, the efficiency of enzyme immobilization is questionable.30 In the former, the statistical enzyme enclosure is restricted by its large size. In the latter, the net negative charge of the enzyme surface, mainly the carboxylic group, is too weak to act as a dopant for the conductive polymer. If the enzymes have net positive surface charges, the ionic repulsive force between matrix and enzyme makes enzyme immobilization difficult. Another disadvantage of these approaches is the potential denaturation of enzymes. In addition to the above mechanisms, a third proposal is the covalent attachment of enzymes to the functional monomer.31,32 However, the large size of the enzymes makes polymerization difficult due to steric hindrance. As a model system for more effective and reproducible enzyme electrodes, in this study, glucose oxidase (GOD) was immobilized into the conducting polymer matrix during the electropolymerization of pyrrole. To improve the efficiency of GOD immobilization into the conducting polymer, we conjugated GOD onto a strong polyanion via a poly(ethylene oxide) (PEO) spacer. The resulting polyanion-glucose oxidase conjugate was used as a dopant in the elctropolymerization procedure. The efficiency of immobilization was investigated on the basis of the electropolymerizability of the system. We then studied the properties of synthesized polypyrrole with the GOD conjugated polyanion dopant to examine its functionality as a glucose sensor. EXPERIMENTAL SECTION Reagents. 2-Acrylamido-2-methylpropane sulfonic acid (AMPS) and pyrrole were obtained from ACROS. PEG monoacrylate (MW 1 500) was donated from Kyoyang Moolsan Co. (Korea). Glucose oxidase (GOD) type II, peroxidase type IV-A, β-D-(+)-glucose, o-dianisine and Sephadex G 200-120 were purchased from Sigma. A modified Lowry protein assay reagent was obtained from (21) Dumont, J.; Fortier, G. Biotechnol. Bioeng. 1996, 49, 544-552. (22) Sangodkar, H.; Sukeerthi, S.; Srinlvasa, R. S.; Lai, R.; Contractor, A. Q. Anal. Chem. 1996, 68, 779-783. (23) Fortier, G.; Belanger, D. Biotechnol. Bioeng. 1991, 37, 854-858. (24) Shin, M. C.; Yoon, H. C.; Kim, H. S. Anal. Sci. 1996, 12, 597-604. (25) Almeida, N. F.; Beckman, E. J.; Ataai, M. M. Biotechnol. Bioeng. 1993, 42, 1037-1045. (26) Hammerle, M.; Schuhmann, W.; Schmidt, H. L. Sens. Actuators, B 1992, 6, 106-112. (27) Benedetto, G. E.; Palmisano, F.; Zambonin, P. G. Biosens. Bioelectron. 1996, 11, 1001-1008. (28) Foulds, N.; Lowe, C. R. Anal. Chem. 1988, 60, 2473-2478. (29) Kajiya, Y.; Sugai, H.; Iwakura, C.; Yoneyama, H. Anal. Chem. 1991, 63, 49-54. (30) Cosnier, S. Electroanalysis 1997, 9, 894-902. (31) Yon-Hin, B. F. Y.; Smolander, M.; Crompton, T.; Lowe, C. R. Anal. Chem. 1993, 65, 2067-2071. (32) Kojima, K.; Yamauchi, T.; Shimomura, M.; Miyauchi, S. Polymer 1998, 39, 2079-2082.

2178

Analytical Chemistry, Vol. 72, No. 9, May 1, 2000

PIERCE. Chemicals for polymer synthesis were purified by conventional purification methods. All other reagents were of reagent or higher grade. Instrumentations. Copolymer synthesis and modification for GOD immobilization were confirmed by 300 MHz FT-NMR (JEOL) and FT-IR (Perkin-Elmer 2000 series). All electrochemical experiments were carried out using a potentiostat (EG&G, model 263A). Data display and recording were supported by electrochemical analysis software (EG&G, model 270). An Ag/ AgCl (3M NaCl) reference electrode was obtained from Bioanalytical Systems, Inc., and Pt wire was used as the working electrode. The Pt wire (diameter 0.5 mm, Fisher) was insulated by a Teflon heat-shrinking tube (Cole-Parmer), and about 5 mm of one end of the Pt wire was left uncovered. A stainless steel mesh was used as the auxiliary electrode. Morphologies of grown polypyrrole were observed after gold coating with a sputter coater (SPI-MOD-ULE, SPI sppl.), by using a scanning electron microsope (SEM, JSM-5800, JEOL). Copolymer Synthesis. To convert the terminal hydroxyl groups of PEG monoacrylate macromer to carboxyl groups, dried PEG monoacrylate (50 g, 33.3 mmol), succinic anhydride (5 g, 50 mmol), triethylamine (5.56 mL, 40 mmol), and dimethylaminopyridine (DMAP) (4.9 g, 40 mmol) were reacted in dioxane (450 mL) for 24 h. The powdery product of carboxylated PEG monoacrylate was obtained by precipitation into ethyl ether. The copolymerization reaction of 2-acrylamido-2-methylpropanesulfonic acid (AMPS) and carboxylated PEG monoacrylate was initiated by benzoyl peroxide (BPO) (0.3 mol % to total monomer) at 75 °C. The feed composition of each monomer was 98 to 2 by molar ratio. All copolymer samples were dialyzed with a Spectra Por 7 membrane (MWCO 15 000, Spectrum) for 4 days to remove all unreacted monomers and low-molecular-weight compounds. Enzyme Conjugation. The carboxylic groups (0.42 mmol) of poly(AMPS-co-PEG monoacrylate) were activated by N-hydroxysuccinimide (1.2 mmol) and dicyclohexylcarbodiimide (1.2 mmol) in DMSO (80 mL) for 48 h. The activated poly(AMPS-co-PEG monoacrylate) (0.11 g) and glucose oxidase (0.5 g) were dissolved in a 0.1 M phosphate buffered solution (50 mL), and the conjugation reaction proceeded at 4 °C for 24 h. The amount of glucose oxidase was fixed to 10 mol % of the activated carboxylic acid groups in the copolymers. The resulting reaction mixture was diluted with cold deionized water and ultrafiltered (MWCO 300 000, Amicon) twice to remove any unreacted enzymes and polymers. The final product, polyanion dopants conjugated with glucose oxidase, was obtained by freezedrying, and confirmation of the conjugation was carried out via gel-filtration chromatography. Sephadex G 200-120 (Sigma) was packed in a self-assembled column (length, 80 cm; 1.5-cm i.d), and a phosphate buffer (0.1 M, pH 7) was used as an eluent (flow rate, 0.3 mL/min). The eluted solution was sampled every 10 minutes and tested by the Lowry method to determine the existence of proteins. Bioactivity of Conjugated Glucose Oxidase. To calculate the amount of glucose oxidase in the conjugate sample, the enzyme concentration was determined by the Lowry method.33,34 (33) Lowry, O. H.; Rosehrough, N. J.; Farr, A. L.; Randall, R. J. J. Biol. Chem. 1951, 193, 265-275.

The bioactivity of the conjugated glucose oxidase was determined by spectrophotometry.35 Assays were performed in a 2 mL, 0.1 M phosphate buffer (pH 7) containing 0.2 mM o-dianisidine, 20 µg of horseradish peroxidase, and 9.5 mM D-glucose. Assays were initiated by the addition of glucose oxidase (60 ng) and, after incubation at room temperature for 20 min, the reaction was quenched with the addition of 0.2 mL of 4 N H2SO4. The absorbance change from reduced o-dianisidine was measured at 400 nm by a UV-vis spectrometer (CARY 1E). The relative activity of the conjugated glucose oxidase was estimated by comparison with the absorbance change between free and conjugated glucose oxidase. Electrochemical Polymerization. Electropolymerization was performed in an undivided cell at 4 °C using a stainless mesh as an auxiliary electrode. Polypyrrole film was grown potentiostatically at 800 mV vs Ag/AgCl in a solution containing pyrrole (0.2 M) and polyanion dopants, which were conjugated with glucose oxidase, for 10 min. The amount of dopant was fixed to 0.1 wt %/solvent volume. As a control, glucose oxidase (0.4 wt %/ solvent volume) was used as a dopant for the polymerization of pyrrole (0.2 M). All other conditions were the same as those of the polyanion-glucose oxidase conjugate dopant except for the polymerization time (1 h) and potential (1.0 V). After polymerization, the electrode was rinsed several times with deionized water to remove any unreacted monomers and dopants and was stored in a 0.1 M phosphate buffer (pH 7.4) at 4 °C. Enzyme-Electrode Response. The amperometric response of the enzyme electrode was measured by chronoamperometry to determine the bioactivity of glucose oxidase incorporated in polypyrrole film. The steady-state current was measured in 8 mL of the 0.1 M phosphate buffer (pH 7.4) solution under a polarizing potential of 0.4 V vs Ag/AgCl. A glucose solution was added after the background current was stabilized, and the concentration of glucose was increased in a stepwise manner until there was no current change with the addition of analyte. All electrochemical response measurements were performed at 37 °C. To investigate the sensitivity of the enzyme electrode to oxygen, all solutions were purged with argon gas for 40 min and the argon atmosphere maintained during the experiment. RESULTS AND DISCUSSION Applying electropolymerization chemistry of conducting polymer to a biosensor, we tested a new approach, the electrochemical polymerization of pyrrole using a polyanion-glucose oxidase conjugate, to more effectively and reproducibly synthesize an enzyme electrode. Figure 1 is the schematic presentation of the approach. Synthesis of Poly(AMPS-co-carboxylated PEG Monoacrylate). Terminal hydroxyl groups of PEG monoacrylate macromers were converted to carboxylic groups, and then poly(AMPS-cocarboxylated PEG monoacrylate) was synthesized by radical polymerization as a polyanion for the conjugation backbone of the glucose oxidase. Conversion yields of carboxylation were 7080% by aqueous titration. The number-average molecular weights of copolymers were determined to be 100-160 kDa using gel permeation chromatography (Waters). Carboxylation reactions (34) Peterson, G. L. Methods Enzymol. 1983, 91, 95-121. (35) Frederick, K. R.; Tung, J.; Emerick, R. S.; Masiarz, F. R. Chamberlain, S. H.; Vasavada, A.; Rosenberg, S. J. Biol. Chem. 1990, 265, 3793-3802.

Figure 1. A new enzyme immobilization method.

Figure 2. Gel-filtration chromatogram of PAMPS-PEG-GOD conjugate. (a) PAMPS-PEO-GOD, (b) GOD.

and copolymer synthesis were confirmed by the characteristic peaks from FT-IR and 1H NMR. Carboxylated PEG monoacrylate: FT-IR 1732 cm-1 (CdO), 1644 cm-1 (CdN+), 1115 cm-1 (CH2OCH2), 1559 cm-1 (COO-); 1H NMR δ 2.4 (COCH CH COOH), δ 3.5 (OCH CH ), δ 4.1 2 2 2 2 (OCH2CH2OCO) in D2O. Poly(AMPS-co-carboxylated PEG monoacrylate): FT-IR 1732 cm-1 (CdO), 1650 cm-1 (CONH), 1458 cm-1 (CH3), and 1038 cm-1 (SdO); 1H NMR δ 1.38 (CCH3CH3), δ 3.5 (OCH2CH2) in D2O. Conjugation of Glucose Oxidase. Figure 2 shows the results of the gel-filtration chromatography of the polyanion-GOD conjugate. The two well-separated and nonoverlapping peaks of the chromatogram and the difference in width of the two peaks indicate that the conjugation reaction was successful. The conjugated sample eluted much faster than free glucose oxidase due to the conjugate’s larger size, which occurred because the GOD was covalently coupled to the poly(AMPS-co-carboxylated PEG monoacrylate), which has a hydrodynamic volume in aqueous media. In terms of width and shape, the polyanion-GOD conjugate showed a broader distribution than the free glucose oxidase due to the molecular weight distribution of the copolymer in the conjugate. Bioactivity of Glucose Oxidase in Polyanion-GOD Conjugate. The amount of glucose oxidase and its bioactivity in the conjugate were listed in Table 1. In all cases, the conjugation yield was more than 90% and there was no relationship between the pHs of the reaction medium and the conjugation yield. The bioactivity was tested in a pH 7 phosphate buffer solution and evaluated relative to the free enzyme (100%). As shown in Table 1, the bioactivity of the glucose oxidase was well preserved after conjugation, revealing the highest bioactivity when conjugated at pH 5. The bioactivity of the conjugated glucose oxidase increased as the pH of the reaction medium decreased, because Analytical Chemistry, Vol. 72, No. 9, May 1, 2000

2179

Table 1. Amount of GOD in PAMPS-PEO-GOD Conjugate and Its Relative Bioactivity oupling medium pH

GOD content in 1 mg conjugate (µg)

conjugation yield (%) (n)5)

relative bioactivity(%) (n)5)

5 6 7 free enzyme (at pH 7)

750 890 800

92 ( 5 109 ( 10 98 ( 4

136 ( 15 107 ( 8 99 ( 12 100 ( 5

Figure 3. Chronoamperogram of pyrrole polymerization with different dopant at 4 °C. (a) PAMPS-PEO-GOD, (b) GOD.

the enzyme has its optimum activity between pH 5 and 6.36 However, the lowest value was still about 99%, and this indicates that the conjugation reaction condition was preferable for preserving the enzyme’s bioactivity. Electrochemical Polymerization of Pyrrole with Polyanion Dopant Conjugated with Glucose Oxidase. The polymerization of pyrrole was performed in deionized water in order to exclude any anionic molecules which may have acted as dopants. Figure 3 illustrates the chronoamperogram during the polymerization of pyrrole in the presence of polyanion-GOD conjugate and free glucose oxidase. The chronoamperometric response shows that the GOD conjugate was effectively incorporated into the polypyrrole chain. Although the large dopant size caused high resistance in the electrochemical cell, the anodic current from pyrrole oxidation increased steeply soon after the oxidative potential (0.8 V) was applied. Then, black polymer began to precipitate onto the working electrode, and a polypyrrole film was uniformly deposited within 10 min. This result suggests that large GOD conjugates were forced to be incorporated in the polypyrrole chain by the strong ionic interaction between sulfonate ions in the conjugate and radical cations in the growing polypyrrole backbone. When the glucose oxidase surface charges were the only anionic group that could be incorporated, the anodic current from pyrrole oxidation was quite low, indicating a slow polymerization process. In fact, it was very difficult to achieve a well-deposited film under the applied potential of 0.8 V because of the high resistance in the electrochemical cell. The low mobility of the anionic groups, caused by the large size of the glucose oxidase, and the weak negative charges on the enzyme surface might be the main reasons for the high resistance. Because polypyrrole film did not grow even after 60 min given the same concentration (36) Kalisz, H. M.; Hecht, H.; Schomburg, D.; Schmid, R. D. J. Mol. Biol. 1990, 213, 207-209.

2180 Analytical Chemistry, Vol. 72, No. 9, May 1, 2000

Figure 4. Scanning electron micrographs of polypyrrole electropolymerized with (a) PAMPS-PEO-GOD conjugate, ×5000, (b) GOD, ×5000.

of dopant (0.001 g/mL) and applied potential (0.8 V) as used in the GOD conjugate dopant system, a higher concentration of dopant (0.004 g/mL) and higher polymerization potential (1.0 V) was applied. However, the anodic oxidation current was still too low and the grown film was very inhomogeneous after 1 h of polymerization. The surface morphology of the grown polypyrrole film was obtained by scanning electron micrographs. As shown in Figure 4, the surface structure of the grown polypyrrole was very rough and inhomogeneous when GOD was used as the only source of incorporatible counterions. However, although the dopant size was larger than that of GOD, the grown polypyrrole film showed a homogeneous and closely packed structure when the GOD conjugate was used as the dopant for pyrrole polymerization. Amperometric Response of the Enzyme Electrode. The chronoamperogram of glucose oxidation demonstrated that the enzyme electrode has a rapid response time and high sensitivity to glucose. Figure 5 shows the amperometric response of the enzyme electrode plotted as a function of the glucose concentration at various conditions. At the polarization potential of 0.4 V under air-saturated atmosphere, after increasing the glucose concentration, the anodic current increased and reached a steady state within 30 s (see insert in Figure 5). The sensitivity up to 20 mM glucose concentration was about 40 nA/mM, and the sensitivity to glucose addition gradually decreased at higher glucose concentrations. The influence of applied potential on the enzyme-electrode response was measured from 0.4 to 0.2 V in an air-saturated

Figure 5. Amperometric response of the enzyme electrode as a function of glucose concentration in pH 7.4 phosphate buffer (0.1 M) at 37 °C. The influence of applied potential was measured at (a) 0.4 V (air saturated), (b) 0.3 V (air saturated), (c) 0.2 V (air saturated), (d) 0.4 V (argon saturated).

atmosphere. Sensitivity to glucose concentration decreased, as the applied potential decreased, and there was no reasonable signal under 0.2 V. This decreasing sensitivity may have resulted from the lower catalytic activity of the enzyme electrode for the oxidation of hydrogen peroxide under lowered potential. When the applied potential was lowered from 0.4 to 0.3, the anodic current signal decreased, and the sensitivity was about 16 nA/ mM in the range from 0 to 20 mM. However, the enzyme-electrode response was virtually linear up to a 20mM glucose concentration. The enzyme electrode showed a very low anodic current at 0.2 V, and the linearity of the signal was also reduced to 10 mM. Among amperometric glucose sensors, the hydrogen peroxide electrode-based glucose sensor, whose signal is generated by the oxidation of hydrogen peroxide at the anode, is generally operated at a potential of 0.6 to 0.8 V vs Ag/AgCl. Although this type of glucose sensor has advantages, such as ease of fabrication and possibility of miniaturization, it suffers from interference by electrooxidizable substances in physiological fluids. However, as shown in Figure 5, the enzyme electrode responded linearly up to 20 Mm even at a potential of 0.3 V, and its sensitivity was still high enough to detect the change of a glucose concentration level of under 1 mM. It is very meaningful that the interference could be reduced by decreasing the working potential to 0.3 V without any electron mediators, which can cause a toxicity problem when leaked from the matrix. To investigate the effect of oxygen on the sensitivity, the amperometric response of the enzyme electrode to glucose

concentration in an air-saturated condition was compared with that in an argon-saturated condition. In both cases, the anodic current response to glucose showed a decreasing pattern at higher concentrations. This indicates that the oxygen supply is the ratedetermining factor for the enzyme-electrode response at the higher concentration of glucose. The amperometric response current was reduced when the test solution was deoxygenated by argon, indicating that the oxygen molecules were mainly responsible for the reoxidation of the FADH2 center of glucose oxidase. However, when in a deoxygenated condition, the enzyme electrode showed a linear anodic current response to glucose concentrations up to almost 20 mM. This indicates that the oxygen molecule played a major role but was not solely responsible for the reoxidation of FADH2 at low glucose concentrations. Although its mechanism was not clarified, direct electrooxidation of the FADH2 may have been another reason. Therefore, a possible interpretation is that the anodic signal of an enzyme electrode may be produced by mixed mechanisms: direct electrooxidation of FADH2 and reoxidation of FADH2 by oxygen molecules. CONCLUSIONS A new enzyme-immobilization method was developed for a more effective and reproducible biosensor. As a model system, glucose oxidase was successfully conjugated to the polyanion, and its bioactivity in the conjugate was preserved after conjugation. The resulting conjugate was effectively incorporated into a conducting polymer matrix, and its efficiency of incorporation was greatly improved when compared with that for conventional immobilization methods. Investigating its properties as a glucose sensor, the enzyme electrode displayed a sensitive response signal to glucose concentrations up to 20 mM with a response time of less than 30 s with a potential as low as 0.3 V. The oxygendependency test revealed that the current signal of the enzyme electrode may be produced by mixed mechanisms. The results presented in this report demonstrate that our new approach is a very effective method of synthesizing enzyme electrodes. ACKNOWLEDGMENT This study was supported by the Academic Research Fund of the Ministry of Education, Korea.

Received July 20, 1999. Accepted for publication January 24, 2000. AC9908041

Analytical Chemistry, Vol. 72, No. 9, May 1, 2000

2181