CHEMICAL & BIOSENSORS
Effect of
L-Aspartate Concentration on the
Response of the Amperometric
L-Glutamate
Sensor for the Measurement of
L-Glutamate
and Aspartate Aminotransferase Activity
in Serum
Ku-Shang Chang
Department of Biochemical Science and Technology & Graduate Institute of Microbiology and Biochemistry, National Taiwan University,
Taipei, Taiwan and Department of Food Science, Yuanpei University, Hsinchu, Taiwan, R.O.C.
Chien-Yuan Chen
Department of Biochemical Science and Technology & Graduate Institute of Microbiology and Biochemistry, National Taiwan University,
Taipei, Taiwan, R.O.C.
Abstract: L-glutamate oxidase was immobilized in a photo-cross-linkable polymer membrane on a palladium strip electrode for the amperometric measurement of aspartate aminotransferas eactivity. The sample, serum for example, was injected into a bufferedL-aspartate and a-ketoglutarate solution. L-aspartate is the essential substrate and can transfer toL-glutamate via the aspartate aminotransferase catalyzing reaction. Aspartate aminotransferase activity can be measured by determining the increasing rate ofL-glutamate. Under the optimal condition, the current increasing rate was proportional to the aspartate aminotransferase activity of the sample in the range of 8 – 200 U/L. The data are in good correlation (R2¼ 0.998) with data from
Received 25 July 2006; accepted 15 September 2006
The authors would like to express their appreciation to the National Science Council of Taiwan for financial support of this research under Contract No. NSC-92-2113-M-002-050.
Address correspondence to Chien-Yuan Chen, Department of Biochemical Science and Technology & Graduate Institute of Microbiology and Biochemistry, National Taiwan University, Taipei, Taiwan, R.O.C. E-mail: chenyuan@ntu.edu.tw; tommy. first@msa.hinet.net
Copyright # Taylor & Francis Group, LLC ISSN 0003-2719 print/1532-236X online DOI: 10.1080/00032710601017821
a commercial aspartate aminotransferase assay kit. Good reproducibility (relative standard deviation ¼ 3.03%, n ¼ 8) was obtained from a sample with 50 U/L aspartate aminotransferase activity. The sensor is expectable to be applied in a clinical point-of-care diagnosis.
Keywords: Biosensor, aspartate aminotransferase, L-glutamate oxidase, polyvinyl alcohol bearing a styrylpyridinium residue, amperometric sensor
INTRODUCTION
Elevated serum levels of aspartate aminotransferase (AST) indicate liver or heart function disorder (Rietz and Guilbault 1975a). When the liver is damaged, the enzyme will be released into the bloodstream, where it is measured to obtain clinical diagnostic information.
Most of the clinical laboratories employ spectrophotometric or fluori-metric methods (Karmen 1955; Reitman and Frankel 1957; Rietz and Guilbault 1975b; Clark 1987) for determining AST activity. These methods are based on the measurement of absorbance decrease of NADH, which is the coenzyme involved in the reaction of malate dehydrogenase (MDH). The MDH-catalyzing reaction is summarized as follows (Clark 1987).
L-aspartate þaketoglutarate AST oxaloacetate þL-glutamate ð1Þ oxaloacetate þ NADH þ HþMDH! malate þ NADþ ð2Þ However, these methods rely on complicated and expensive instruments and thus are not suitable for point-of-care applications. Alternatively, ampero-metric sensors appear very attractive because they can provide specific and repetitive assays by miniaturizable low-cost devices. A carbon strip electrode, prepared with a screen-printing technique, was often employed for constructing an amperometric biosensor. The following reaction, using
L-glutamate oxidase (L-GLOx) is an alternative to the amperometric sensor to measure AST activity.
L-glutamate þ O2 ! LGLOx
a-ketoglutarate þ NH3þH2O2 ð3Þ
The reference range for AST activity in normal adults is between 30 and 40 U/L only. The determination of transaminase activity on an amperometric sensor usually requires a rather long time (Wollenberger et al. 1989; Compagnone et al. 1992; Cooper et al. 1991), owing to the low AST activity in the serum sample (Hart et al. 1999). For achieving a rapid and highly sensitive measurement, the following approachs are taken into consideration. First, hydrogen peroxide liberated from enzymatic reactions in the vicinity of the electrode surface must be oxidized immediately.
In one of our recent studies, palladium metal was deposited on the electrode (Chang et al. 2003). The deposition of palladium on the electrode surface provides a very porous structure consisting of aggregates of fine palladium particles. This results in a significant increase in the electrode surface area, which leads to an increase in surface activity for the measurement of the hydrogen peroxide liberated from enzymatic reactions (White et al. 1994; Newman et al. 1995; Chi and Dong 1993; Johnston et al. 1995). Second, the substrate must diffuse quickly through the thin layer to undergo the enzymatic reaction. Third, substrate and oxygen must diffuse quickly through the enzyme membrane layer to undergo the enzymatic reaction (Mizutani et al. 1998). The use of an electrochemical mediator is useful to shuttle electrons from the active site of the enzyme or biomolecule to the electrode surface. However, the electroactive species employed in such a system would be easily diffused away from the electrode surface and thus rendered unsuitable for continued use. A photo-cross-linkable polymer polyvinyl alcohol bearing a styrylpyridinium residue (PVA-SbQ) has also been reported to be suitable for fabrication of a thin enzyme membrane (about 1 mm thick) (Sohn et al. 1997). The enzyme can be immobilized in the matrix with high surface density (Jaffrezic-Renault et al. 1999). The authors presented a highly sensitive L-glutamate sensor for measuring the activity of alanine aminotransferase (ALT) in serum (Chang et al. 2003). The sensor is composed of immobilized L-GLOx in a photo-cross-linkable
polymer, PVA-SbQ, membrane on a palladium-deposited screen-printed carbon electrode. PVA-SbQ is a polyvinyl alcohol bearing a styrylpyridinium residue as its photosensitive group. The cross-linking of the polymer is performed using florescent light or UV radiation to produce cyclodimerization of the styrylpyridinium groups (Ichimura and Watanabe 1980). PVA-SbQ has various advantages when applied as an immobilizing agent for proteins (Ichimura 1984; Newman et al. 1995; Rouillon et al. 1994; Perdomo et al. 1999). Its cross-linking reaction is induced in the light with a wavelength longer than 320 nm: 1) there occurs no photo-induced deactivation to an enzyme because such a light is not absorbed by proteins and 2) chemical denature is prevented because no covalent bonds form between the enzyme and the matrix (Ichimura 1984). The measurement of AST activities in the serum was more difficult than the measurement of ALT activities because the substrate for AST catalytic reaction, L-aspartate, may cause a cross-reaction withL-GLOx and generate a cathodic signal in the sensor, as has
been reported (Kusakabe et al. 1983; Zilkha et al. 1995; Janasek and Spohn 1999). In preliminary tests, the L-glutamate sensor did not work when the
reagent contained high concentrations of substrate. In this research, an optimized study was performed for the measurement of AST activity in serum. The response curves of low AST concentration and high AST concen-tration were compared at the same substrate dose of reagent. The detection time, reproducibility, and sensitivity of the sensor for the measurement of AST activity in serum were also determined.
EXPERIMENTAL SECTION Reagents
Bovine serum albumin (BSA), monosodiumL-glutamate (MSG),L-aspartate,
a-ketoglutarate (a-KG), glutaraldehyde, and AST (EC 2.6.1.1, from porcine heart) were purchased from Sigma Chemical Co. (St. Louis, MO, USA). PVA-SbQ (degree of polymerization 1700, degree of saponification 87%) was obtained from Toyo Gosei Kogyo Chemical (Tokyo, Japan). The enzyme
L-GLOx (EC 1.4.3.11) was produced from Streptomyces sp. NTU 3304 in our
laboratory (Chen and Su 1991). All other reagents were of analytical grade.
Electrode Modification
A commercial glucose test strip (Boehringer-Mannheim Roche Diagnostics Corp., Indianapolis, USA) consisting of two palladium strip electrodes was used. The palladium connecting strips of the commercial electrode were used as electrodes by cutting off the enzyme loaded pads. The prepared palladium strips were then modified and connected to a laboratory-built poten-tiostat as the work and counter electrodes.
Enzyme Immobilization
The enzyme was immobilized by a combination process of PVA-SbQ photo-cross-linking and glutaraldehyde exposure. The enzyme solution with activity of 70 U/mL was prepared by mixing 50 mg PVA-SbQ, 10 mg BSA, and 140 mg ofL-GLOx with activity of 70 U/mL. Then 1 mL of the
mixture was applied onto the active area of the electrode. The sensor was then placed in a dark sealed box with saturated glutaraldehyde vapor. The box was kept at 48C for 12 hours to complete the cross-linking reaction among the protein molecules by glutaraldehyde. The sensor was then exposed to florescent light for 25 min to complete the photo-cross-linking reaction among PVA-SbQ molecules.
Apparatus and Procedure
Amperometric measurements were performed using a laboratory-built potentiostat. Input and output signals from the potentiostat were coupled to a PC (Pentium 166 MHz) using a peripheral interface card (AT-MIO-16E, National Instruments, Austin, TX, USA). The interface card consisted of a 16-channel analog-to-digital (A/D) converter (12 bit) and a 2-channel digital-to-analog (D/A) converter (12 bit). Voltage output, data display, and signal recording were programmed using the LabVIEW 5.1 software
package (National Instruments, Austin, TX, USA). All measurements were taken with a three-electrode system, using an Ag/AgCl electrode as the reference and the modified palladium strips as the working and the counter electrodes.
The amperometric measurement was performed at 0.4 V vs. Ag/AgCl in a thermostat water-jacketed cylindrical cell. Phosphate buffer (PBS, pH 7.0) was used as the working solution. All experiments were carried out at 358C. For the amperometric measurement of MSG, 100 mM of MSG solution were injected into a 9.9-mL working solution, using a microsyringe, when a steady state of the operation system had been reached. The baseline current and the response current following the injection of MSG were displayed and simultaneously logged by the computer until a steady state was achieved. Magnetic stirring was used to ensure the homogeneity of the solution. The difference between the baseline and the steady-state current was used to calculate the concentration of MSG.
For the determination of AST activities in serum, a buffered substrate solution was prepared by dissolving 1 mM a-KG and 25 mM L-aspartate in 100 mM phosphate buffer (pH 7.0). The electrodes at working potential of 0.4 V vs. Ag/AgCl were settled in the reaction cell with a given volume of the buffered substrate solution. Samples were injected into the reaction cell when a baseline current had been reached. The response current was recorded continuously, and the current-to-time curve was recorded. The standard sample solutions for which the AST activity had been determined by the Sigma AST assay kit were used to establish the calibration graph.
RESULTS AND DISCUSSION Fabrication of theL-Glutamate Sensor
The L-glutamate produced in the AST catalytic reaction is converted into
a-KG and hydrogen peroxide by L-GLOx. The hydrogen peroxide was
then oxidized on the palladium electrode by the polarizing potential applied to the electrode. The response current obtained from the oxidation of hydrogen peroxide corresponds to the concentration of hydrogen peroxide, L-glutamate, and the AST activity within the working capacity
ofL-GLOx.
The AST activity is as low as 30 U/L in human serum (Mizutani et al. 1998). A thin membrane of immobilized enzyme with satisfactory diffusion characteristics is desirable for achieving a rapid and highly sensitive measurement of AST activity. PVA-SbQ has been reported to be suitable for the fabrication of an immobilized enzyme membrane as thin as 1 mm with good diffusion characteristics (Ichimura 1984). It was chosen as the enzyme- immobilizing agent in this study and further combined with
glutaraldehyde to fabricate the immobilized enzyme membrane. The reaction between glutaraldehyde and a primary amino group by the formation of the Schiff-base bond has been investigated and applied in covalently binding amino groups on an enzyme molecule, L-GLOx for example (Chang et al. 2003; Chen 1987). Protein molecules with no catalytic activity, such as BSA, are usually co-cross-linked with the enzyme molecules to increase the size of cross-linked molecules. The authors’ recent work showed that the optimal composition for the enzyme immobilization was 25% PVA-SbQ and 2.5% BSA, by weight (Chang et al. 2003). The immobilized enzyme membrane on the active area of the electrode was constructed according to the procedure previously described. The optimal period of exposure to glutar-aldehyde vapor in the dark sealed box was 12 hours (Table 1). Another study was conducted to examine the effect of the applied volume of the enzyme solution on the response of the sensor. The results showed that a solution of 1 mL enzyme yielded maximal response and 0.5 mL resulted in a rapid response. The between-electrode reproducibility of the sensor, however, was poor when the applied volume was less than 1 mL because the enzyme membrane was unable to distribute evenly on the electrode. The applied volume of the enzyme solution was thus decided as 1 mL in the following experiments.
The stability and reproducibility of the response obtained with the same electrode was satisfactory. The within-electrode relative standard deviation was 2.6% for the response to 10 mM MSG over 100 tests. The sensor was stable for 5 mo in dry condition under room temperature. The immobilized enzymes can retain their functional characteristics to a large extent for several months upon repetition of wetting and drying (Chang et al. 2003). The satisfactory stability can be attributed to the method of combining photo-cross-linking of PVA-SbQ with a chemical-cross-linking agent (e.g., glutaraldehyde), to immobilize the enzyme effectively, and to the presence of a great number of 2OH groups in the PVA-SbQ matrix that offer a hydrophilic environment around the enzyme molecules.
Table 1. Effect of the exposure time in saturated glutaraldehyde vapor on the response characteristics of the sensor
Exposure time (h) Responsea(nA) SD (nA) RSD (%)
0 1215 246 20.24
6 1113 182 16.32
12 1088 92 8.46
a
Each item of the data is the mean value of three measurements. Measurements were carried out with 100 mM MSG as the substrate in 100 mM sodium phosphate buffer (pH 7.0). The polarizing potential applied to the sensor was 0.4 V vs. Ag/AgCl.
Effect of pH and Temperature on the Response of the Sensor
As described in our previous paper (Chang et al. 2003), theL-GLOx electrode exhibited an anodic response to hydrogen peroxide starting at þ0.2 V, reaching its diffusion current limit at about þ0.4 V. This result is in agreement with the findings (Chi and Dong 1993: Newman et al. 1995) in similar experiments. In order to minimize the interference from electroactive species (e.g., ascorbic acid and uric acid), an operation potential of þ0.4 V was selected for the following tests.
The effect of pH on the response of the sensor was examined in the pH range from 5.0 to 9.5. The maximal sensitivity was observed in buffer solutions with pH 7.0. The fact that optimal pH for immobilized L-GLOx
was close to that of the soluble L-GLOx revealed that the immobilizing
process did not change the catalytic behavior of L-GLOx due to the pH
variation (Chen 1987). The positively charged PVA-SbQ matrices were reported to change the optimal pH value for some immobilized enzymes (Ichimura and Watanabe 1980). However, it had only a negligible influence on the optimal pH value ofL-GLOx in the present work.
The effect of temperature on the sensitivity of the sensor was also studied in the range from 15 to 808C (Fig. 1). The maximum sensitivity was observed at 658C. It is 158C higher than the optimal temperature for solubleL-GLOx, indicating that the immobilization process improved the thermal stability of the enzyme (Chen 1987). The decrease in response when the temperature was higher than 658C is considered as thermal inactivation of L-GLOx.
Although the response of the sensor was enhanced at higher temperatures, the response of the sensor became unstable at a temperature higher than
Figure 1. Effect of temperature on the response of the sensor. Measurements were carried out with 100 mM MSG as the substrate in 100 mM sodium phosphate buffer (pH 7.0). The polarizing potential applied to the sensor was 0.4 V vs. Ag/AgCl.
508C (data not shown). The temperature of 378C was chosen for the present study under the consideration of human body temperature.
Determination of MSG Concentration
Two calibration curves for the determination of MSG were established by the steady-state method and the kinetic method, respectively. The sensor exhibited a linear calibration range, from 50 nM to 100 mM, with a slope of 12.18 nA/mM and a correlation coefficient of 0.997 (n ¼ 10) based on the steady-state method. The detection limit of MSG was 50 nM (S/N ¼ 3). The calibration curve obtained from the kinetic method showed dynamic linear ranges from 50 nM to 75 mM, with a correlation coefficient of 0.994.
Optimization of Substrate Composition
a-KG andL-aspartate are necessary substrates for the measurement of AST
activity. Responses of the enzyme-catalyzed reactions have been reported to be proportional to the substrate concentration (Mizutani et al. 1998). However, as shown in Fig. 2, the sensitivity of theL-glutamate sensor was
severely inhibited bya-KG andL-aspartate.
In order to achieve a better sensitivity for the measurement of AST activity using ourL-glutamate sensor, the optimal concentration ofa-KG andL -aspara-tate was investigated. Table 2 shows the effect ofL-aspartate anda-KG concen-tration on the response of the sensor in the measurement of AST activity. A standard AST solution with 30 U/L activity (the normal range in human
Figure 2. Effect of (O)L-aspartate and (V) a-ketoglutatate concentrations on the response of the sensor to 100 mM MSG. The sensor was operated in 100 mM sodium phosphate buffer (pH 7.0) at 358C. The polarizing potential applied to the sensor was 0.4 V vs. Ag/AgCl.
blood is 25– 30 U/L) was injected into the working solution. The concentrations ofa- KG andL-aspartate in the working solution were varied in the range from 0– 4 mM and 15– 50 mM, respectively. When the concentration ofa-KG was more than 2 mM, the response current of the sensor became very unstable and the increase rate of the response current could not be determined. The optimal substrate composition for the measurement of AST activity was decided as 1 mMa-KG and 25 mML-aspartatem as shown in Table 2.
Measurements of AST Activity
Comparison of the response curves obtained from previous study for the deter-mination of ALT activity (Chang et al. 2003) with the response curves for the determination of AST activity (Fig. 3) revealed that the latter obtained a higher background current.L-Aspartate causes a cross-reaction toL-GLOx, yielding a cathodic signal to the sensor (Kusakabe et al. 1983; Zilkha et al. 1995; Janasek and Spohn 1999), which produces high background current. As shown in Fig, 3 line B, steady-state background current could be obtained when the sensor was settled in the reaction cell filled with working solution. The current profiles for standard solution of 50 U/L ALT and AST were as shown in Fig, 3.L-Glutamate is a product of the AST catalytic
reaction. The activity of AST can be determined from the increase of the response current of the L-glutamate sensor. Electrochemically active
sub-stances in serum, such as ascorbic acid and uric acid, may cause severe inter-ference on an electrochemical sensor system. In this study, the responses from electrochemically active species caused an elevation of the baseline. They created no interference on a dynamical AST determination when the dynamic data are taken after the steady state of an elevated baseline has been reached. As shown in Fig. 3, the current increased with extension of reaction time after standard ALT was injected, indicating that the hydrogen peroxide released in the reaction was produced by theL-GLOx immobilized Table 2. Effect ofL-aspartate and a-ketoglutarate concentration on the response of the sensor for 30 U/L AST
Current increase rate(nA/min) a-ketoglutarate concentration (mM) L-aspartate concentration (mM) 0 0.5 1 2 4 15 0 5.74 3.46 —a ndb 25 0 6.97 8.80 —a ndb 50 0 3.10 3.89 5.27 —a
aSignal could not be detected because of noise distortion. b
in the PVA-SbQ membrane. Interestingly, the response current decreased immediately after the sample solution was injected. The response curve was different from the response curves obtained for the determination of 50 U/L ALT activity. The decrease of response current is caused by the dilution of
L-aspartate and the consequent reduction of the background current when
the sample solution was added. Two minutes elapsed before the response
Figure 3. Response curves of the sensor for (A) ALT and (B) AST standard solution with 50 U/L concentrations, which were added to the working solution at the time point indicated. The sensor was operated in 100 mM sodium phosphate buffer (pH 7.0) containing 1 mM a-ketoglutarate, 25 mM L-aspartate (for AST) and/or 100 mML-alanine (for ALT) at 358C. The polarizing potential applied to the sensor was 0.4 V vs. Ag/AgCl.
Figure 4. Calibration graphs of the sensor for the determination of AST activity. The sensor was operated in 100 mM sodium phosphate buffer (pH 7.0) containing 1 mM a-ketoglutarate and 25 mML-aspartate at 358C. The polarizing potential applied to the sensor was 0.4 V vs. Ag/AgCl.
current began to increase linearly with the extension of reaction time. The increasing rate of the response current was related to AST activity. The increasing rate of the response current was accordingly measured from 2 min after the injection of sample solution. The slopes (nA/min) obtained from various AST standard solutions were plotted against the AST activity, determined by the AST assay kits over the range 8 – 200 U/L ( y ¼ 0.2588 20.7178, R2¼ 0.998, n ¼ 10), to yield a calibration graph (Fig. 4). The within-electrode reproducibility of the sensor showed a relative standard deviation of 3.03% (Table 3) for AST with 50 U/L activity in eight measurements. However, the reproducibility of the sensor showed a relative standard deviation of more than 10% for AST with 8 U/L activity. This result demonstrates that the present method was poor in repro-ducibility for the measurement of low ALT concentration. Since the
L-GLOx in this experiment could catalyze the L-aspartate, there was high
background current to interfere with the amperometric reaction. The sensi-tivity and reproducibility of the L-GLOx biosensor can be enhanced by modifying the characteristic of L-GLOx to reduce the background reaction betweenL-GLOx andL-aspartate.
CONCLUSION
A simple and convenient AST measuring device for diagnostic and point-of-care purposes is required. AL-glutamate sensor, based on a
palladium-deposited screen-printed carbon electrode and an immobilized L-GLOx membrane, was developed for the determination of AST activity in serum. The time required for measurement of AST activity was 4 min. The linear range of the sensor was 8 – 200 U/L AST activity (R2¼ 0.998). The reprodu-cibility of the sensor was relative standard deviation of 3.03% (n ¼ 8). The present L-glutamate sensor is simpler and less expensive than commercial devices for measuring AST. However, a cross- reaction between the
L-GLOx andL-asparate, and thus a high background current, was observed. Table 3. Response of the sensor during the test of
within electrode reproducibility of 50 U/L AST
AST response (nA/min)
Meana 12.58
SD 0.385
RSD (%) 3.03
a
Data are the mean values of eight measurements. Measurements were carried out with 1 mM a-ketogluta-rate and 25 mM L-aspartate as the substrate in 0.1 M sodium phosphate buffer (pH 7.0). The polarizing potential applied to the sensor was 0.4 V vs. Ag/AgCl.
The substrate composition in the buffered working solution has been optimized to maximize the sensitivity of the sensor. Further study concerning a device for sequentially or simultaneously measuring AST and ALT activities is currently being undertaken in our laboratory.
REFERENCES
Chang, K.S., Hsu, W.L., Chen, H.Y., Chang, C.K., and Chen, C.Y. 2003. Determi-nation of glutamate pyruvate transaminase activity in clinical specimens using a biosensor composed of immobilizedL-glutamate oxidase in a photo-crosslinkable polymer membrane on a palladium-deposited screen-printed carbon electrode. Anal. Chim. Acta, 481: 199 – 207.
Chen, C.Y. 1987. Doctoral Thesis, Graduate Institute of Agriculture Chemistry, National Taiwan University, Taipei, Taiwanp. 226.
Chen, C.Y. and Su, Y.C. 1991. Amperometric L-glutamate sensor using a novel L-glutamate oxidase from Streptomyces platensis NTU 3304. Anal. Chim. Acta, 243: 9 – 15.
Chi, Q. and Dong, S. 1993. Flow-injection analysis of glucose at an amperometric glucose sensor based on electrochemical deposition of palladium and glucose oxidase on a glassy carbon electrode. Anal. Chim. Acta, 278: 17 – 23.
Clark, L.C. 1987. Biosensors, Fundamentals and Applications. Turner, A.P.F., Karube, I. and Wilson, G..S. eds.; Oxford Science Publications: New York, p. 11. Compagnone, D., Federici, G., Massouud, R., Santoro, L., Anichini, M., and
Palleschi, G. 1992. Analysis for transaminases in serum with an amperometric glutamate electrode. Clin. Chem., 38: 2306 – 2310.
Cooper, J.M., McNeil, C.J., and Spoors, J.A. 1991. Amperometric enzyme electrode for the determination of aspartate aminotransferase and alanine aminotransferase in serum. Anal. Chim. Acta, 245: 57 – 62.
Hart, J.P., Abass, A.K., Cowell, D.C., and Chappell, A. 1999. Development of a dispo-sable amperometric NH biosensor based on a chemically modified screen-printed carbon electrode coated with glutamate dehydrogenase, 2-oxoglutarate, and NADH. Electroanalysis, 11: 406 – 411.
Ichimura, K. 1984. A convenient photochemical method to immobilize enzyme. J. Polym. Sci., Polym. Chem. Ed., 22: 2817 – 2828.
Ichimura, K. and Watanabe, S. 1980. Immobilization of enzymes with use of photosen-sitive polymers having the stabazolium group. J. Polym. Sci., Polym. Chem. Ed., 18: 891 – 902.
Jaffrezic-Renault, N., Senillou, A., Martelet, C., Wan, K., and Chovelon, J.M. 1999. ISFET microsensors for the detection of pollutants in liquid media. Sens. Act., B59: 154 – 164.
Janasek, D. and Spohn, U. 1999. Chemiluminometric flow injection analysis procedures for the enzymatic determination of L-alanine, a-ketoglutarate and L-glutamate. Biosens. Bioelectron., 14: 123 – 129.
Johnston, D.A., Cardosi, M.F., and Vaughan, D.H. 1995. The electrochemistry of hydrogen peroxide on evaporated gold/palladium composite electrodes. Electroana-lysis, 7: 520 – 525.
Karmen, A. 1955. A note on the spectrophotometric assay of glutamic-oxaloacetic transaminase in human blood serum. J. Clin. Invest., 34: 131 – 133.
Kusakabe, H., Midorikawa, Y., Fujishima, T., Kuninaka, A., and Yoshino, H. 1983. Purification and properties of a new enzyme,L-glutamate oxiodase, from Strepto-myces sp. X-119-6 grown on wheat bran. Agric. Biol. Chem., 47: 1323 – 1328. Mizutani, F., Sato, Y., Sawaguchi, T., Yabuki, S., and Iijima, S. 1998. Rapid
measure-ment of transaminase activities using an amperometric-glutamate sensing electrode based on a glutamate oxidase – polyion complex-bilayer membrane. Sens. Act., B52: 23 – 29.
Newman, J.D., White, S.F., Tothill, I.E., and Turner, A.P.F. 1995. Catalytic materials, membranes and fabrication technologies suitable for the construction of ampero-metric biosensors. Anal. Chem., 67: 4594 – 4599.
Perdomo, J., Sundermeier, C., Hinkers, H., Morell, O.M., Seifert, W., and Knoll, M. 1999. Containment sensors for determination ofL-lactate and glucose. Biosen. Bio-electron., 14: 27 – 32.
Reitman, S. and Frankel, S. 1957. A colorimetric method for the determination of serum glutamaic oxalacetic and glutamic pyruvic transaminases. Am. J. Clin. Pathol., 28: 56 – 63.
Rietz, B. and Guilbault, G.G. 1975a. Fluorimetric assay of serum glutamate oxaloace-tate transaminase, glutamate pyruvate transaminase and a-hydroxybutyrate dehy-drogenase by solution. Anal. Chim. Acta, 77: 191 – 198.
Rietz, B. and Guilbault, G.G. 1975b. Use of reagent tablets for assay of three enzymes. Clin. Chem., 21: 1544 – 1546.
Rouillon, R., Tocabens, M., and Marty, J.L. 1994. Stabilization of chloroplasts by entrapment in polyvinylalcohol bearing styrylpyridinium groups. Anal. Lett., 27: 2239 – 2248.
Sohn, B.K., Cho, B.W., Kim, C.S., and Kwon, D.H. 1997. ISFET glucose and sucrose sensors by using platinum electrode and photo-crosslinkable polymers. Sens. Act., B41: 7 – 11.
White, S.F., Turner, A.P.F., Schmid, R.D., Bilitewski, U., and Bradley, J. 1994. Inves-tigations of platinized and rhodinized carbon electrodes for use in glucose sensors. Electroanalysis, 6: 625 – 632.
Wollenberger, U., Scheller, F.W., Boehmer, A., Passarge, M., and Mueller, H.G. 1989. A specific enzyme electrode forL-glutamate-development and application. Biosens. Bioelectron., 4: 381 – 391.
Zilkha, E., Obrenovitch, T.P., Koshy, A., Kusakabe, H., and Bennetto, B.P. 1995. Extracellular glutamate: On-line monitoring using microdialysis coupled to enzyme-amperometric analysis. J. Neurosci. Meth., 60: 1 – 9.
