ANALYST, AUGUST 1991, VOL. 116 793 Immobilized-enzyme Electrode for Nicotinamide Adenine Dinucleotide (Reduced form) (NADH) Sensing and Application to the Kinetic Studies of NADH Dependent Dehydrogenases Hsien-Chang Chang, Akinori Ueno, Hiroshi Yamada, Tornokazu Matsue" and lsamu Uchida Department of Molecular Chemistry and Engineering, Faculty of Engineering, Tohoku University, Sendai 980, Japan Amperometric determination of nicotinamide adenine dinucleotide (reduced form) (NADH) at an immobilized-diaphorase (Dp) electrode is described. The measurement was conducted using ferrocenylmethanol as a mediator in a stirred solution at 0.20 V versusa saturated calomel electrode. A linear relationship between the steady-state current and the concentration of NADH was found over the range 0.005-0.1 25 mmol dm-3.The immobilized-Dp electrode showed outstanding stability and the current response reached a steady state within 2-3 seconds upon addition of NADH. The proposed electrode was used to follow the reactions of pig heart lactate dehydrogenase and horse liver alcohol dehydrogenase. The kinetic investigation using the immobilized-Dp electrode gave the kinetic parameters (Michaelis constants, Km values, and maximum velocities, Vm values), which were in satisfactory agreement with those determined by a conventional spectrophotometric method. Keywords: Diaphorase; immobilized-enzyme electrode; nicotinamide adenine dinucleotide sensor Recently much attention has been focused on enzyme sensors based on various electrochemical methods of detection. * Many of the enzyme sensors proposed so far are based on amperometric or potentiometric detection coupled with oxi- dases .2 Dehydrogenases requiring nicotinamide adenine di- nucleotide (reduced form, NADH; oxidized form, NAD+) have not frequently been used for enzyme sensors in spite of their range of catalytic capabilities.More than 250 NADH dependent dehydrogenases3 have so far been purified from various sources, and many attempts have been made to use dehydrogenases for selective organic syntheses.4 The major problem in the practical application of dehydrogenases is the difficulty in regeneration of the coenzyme. Although the formal potential of the NADH-NAD+ couple is -0.56V versus a saturated calomel electrode (SCE) at pH 7.0,s large overpotentials are necessary for the direct electrochemical oxidation of NADH (about 1.1 V at a glassy carbon6 and 1.3 V at a platinum electrode7).The overpotentials are mainly ascribed to the very positive formal potential for the NADH- NADH.+ couple (0.78 V versus SCE).8 Thus, it is practically impossible to reduce the overpotential if NADH*+ is formed as the intermediate. However, it has been found that quinones,g 3-P-naphthoyl- Nile Blue,lo phenazine methosulphate,11 Meldola Blue 12 and a hexacyanoferrate film on an electrode13 effectively mediate the oxidation of NADH. Generally, these approaches are limited by their low stability or slow response. Another method used to accelerate the oxidation is the use of an enzyme reaction coupled with an electron transfer mediator such as a ferrocenel4 or a ferrocyanide complex.15 Cass et al.16 found that lipoamide dehydrogenase catalyses the electro- chemical oxidation of NADH. Recently, Miki et al. 17 demon- strated that diaphorase (Dp) pasted in carbon powder exhibits catalytic activity for the oxidation of NADH. The determination of NADH using an immobilized-Dp electrode in the presence of ferrocenylmethanol (FMA) as a mediator is reported here (Fig. 1). The proposed Dp, purified from Bacillus stearothermophilus, showed excellent stability. The immobilized-Dp electrode was also used for the determi- nation of the kinetic parameters of NADH dependent de h y drogenases. * To whom correspondence should be addressed. Experimental Materials Diaphorase I (E.C. 1.6.99-; relative molecular mass, -30 000; activity, 2030 U mg-1 for NADH) (1 U = 16.67 nkat) from Bacillus stearothermophilus was obtained from Unitika (Kyoto, Japan).The enzyme was purified according to a method reported previously. 18 The enzyme concentration was determined from the absorbance at 460 nm using ~ ~ 6 0 = 12 OOO dm3 mol-1 cm-1 .I8 Lactate dehydrogenase (LDH, (S)-lactate: NAD+ oxidoreductase, E.C. 1.1.1.27, relative molecular mass, 140 000; activity, -5000 U ml-1 for pyruvate) from pig heart was purchased from Oriental (Osaka, Japan). Alcohol dehydrogenase from horse liver (HLADH, ethanol: NAD+ oxidoreductase, E.C. 1.1.1.1, relative molecular mass, 84 0oO; activity, 27 U ml-1 for ethanol) was purchased from Boehr- inger, Mannheim, Germany. As the LDH and HLADH have not been purified, concentrations of these enzymes are expressed as U ml-1.The NADH was obtained from Sigma (St. Louis, MO, USA) and its concentration was determined based on &340 = 6220 dm3 mol-1 cm-1.19 Ferrocenylmethanol was obtained from Tokyo Kasei (Tokyo, Japan) and recrystal- lized from hexane. Sodium pyruvate and glutaraldehyde (25% solution) were purchased from Wako Chemicals (Osaka, Japan) and used as received. Cyclohexanone, 2-methyl- and 3-methylcyclohexanone were obtained from Wako and dis- tilled before use. All solutions were prepared with water purified by using a Milli-QII system (Millipore, Milford, MA, USA). Electrode FMA NADH FMA+ NAD+ Fig. 1 diaphorase (Dp) [reduced form, Dp (red); oxidized form, Dp (ox)] Scheme for electrocatalytic oxidation of NADH catalysed by794 ANALYST, AUGUST 1991, VOL.116 Instrumentation and Measurements Cyclic voltammetry and amperometry were performed with a potentiostat (Model HAB-151, Hokuto Denko, Tokyo, Japan) connected to an x-y recorder (Model WX43096, Graphtec, Tokyo, Japan) in a 4 ml solution of 0.05 mol dm-3 phosphate-NaOH buffer (pH = 7.5). The working electrode was a glassy carbon disc (3 mm in diameter) (Tokai Carbon, Tokyo, Japan, GC-20) mounted in a polytetrafluoroethylene rod. The electrode surface was polished to a mirror-like finish with 300 nm alumina. Preparation of the immobilized-Dp electrode was carried out by mixing 4 x 10-4 ml of a 1 mmol dm-3 solution of Dp with 2 x 10-4 ml of a 2% v/v glutaraldehyde solution at the electrode surface. The surface concentration of the immobilized Dp was 5.7 nmol cm-2.The electrode was kept at room temperature (25 "C) for 2 h in order to allow polymerization at the surfaces. A measurement made using a scanning electron microscope indicated that the thickness of the enzyme film was approximately 0.01 mm. The immobilized-enzyme electrode was kept in a buffer solution at 4 "C for at least 2 d before the measurements were made. The counter electrode was a platinum wire and the potentials were referenced to an SCE. All measurements were carried out using a water-jacketed cell kept at 30 "C under an atmosphere of nitrogen. The amperometric measurements for enzyme assays and kinetic studies were performed with stirring to avoid the influence of diffusion. The activity of LDH for pyruvate and that of HLADH for cyclohexanones were measured by monitoring the decrease in the concentration of NADH.The concentration of NADH was determined by the oxidation current for NADH observed at the immobilized-Dp electrode. The electrode potential was set at +0.20V versus SCE. The LDH assay solutions containing 0.1 mmol dm-3 NADH and various concentrations of pyruvate ranged from 0.06 to 0.50 mmol dm-3. The concentration of LDH was fixed at 250 U ml-1. The initial rate of the enzyme reaction was determined by the decrease in the concentration of NADH after substrate was added to the solution. In order to confirm the validity of the electrochemical method for the kinetic study described above, a spectro- photometric method was also examined. The concentration of NADH was determined by the absorbance at 340 nm using a multichannel spectroscopic system Model MCPD-110A (Ohtsuka Electric, Osaka, Japan).Results and Discussion Basic kinetic studies for the mediated electro-oxidation of NADH catalysed by Dp in solution have been reported by Matsue et aZ.18 It was found that FMA and 1-ferrocenylethanol showed smaller K , values and higher molecular activities than -0.4 -0.2 0 0.2 0.4 0.6 E N versus SCE Fig. 2 Cyclic voltammograms for 0.20 mmol dm-3 FMA on the immobilized-Dp electrode in 0.05 mol dm-3 phosphate buffer (pH 7.5). Solid line, without addition; broken line, with addition of 0.15 mmol dm-3 NADH; scan rate, 10 mV s-l; and temperature, 30 "C Co(phen)32+ (phen = 1,lO-phenanthroline) and Fe(CN)&-, indicating that these ferrocene derivatives act effectively as the mediators in accelerating the enzymic oxidation of NADH.In this paper, FMA, has been used mainly, because of its electrochemical reversibility, stability and commercial availability. Cyclic Voltammetric Behaviour of Immobilized-Dp Electrode Fig. 2 shows the cyclic voltammograms for 0.2 mmol dm-3 FMA at the immobilized-Dp electrode. In the absence of NADH, the voltammogram shows a well defined, reversible peak at about +0.20V versus SCE. The addition of 0.15 mol dm-3 NADH to the solution results in the appearance of a pre-wave at about +O.O5V and an obvious increase in the oxidation peak. The appearance of the pre-peak is caused by depletion of NADH in the vicinity of the electrode surface.20 This phenomenon is observed when catalytic reactions proceed rapidly at the electrode surface. The findings de- scribed above demonstrate that the Dp immobilized on the electrode surface effectively catalyses the oxidation of NADH by the oxidized form of FMA (Fig.1). Dependence of Current Response on the Concentration of NADH Fig. 3 shows the current response at the immobilized-Dp electrode upon the addition of NADH. The current response was extremely rapid and reached a steady-state current (&) within 2-3 seconds. The mediator rapidly shuttles between the 2.0 2 1.6 2 $ 1.2 2 (3 0.8 0.4 0 1 2 3 4 Ti me/m in Fig. 3 Current response at the immobilized-Dp electrode for successive additions of NADH (indicated by arrows) to the 0.05 mmol dm-3 phosphate buffer containing 0.2 mmol dm-3 FMA.Potential, 0.20 V versus SCE; temperature, 30 "C, [FMA], 0.02 mmol dm-3. [NADH]: A, 0; B, 12.5 x 10-6; C, 25 x 10-6; D, 50 x 10-6; E, 75 x 10-6; and F. 100 X 10-6 mol dm-3 3.5 Q 2.5 (D I s ? k 1.5 3 0 0.5 0 0.05 0.10 0.15 [NADH]/mmol dm-3 Fig. 4 Relationship between the oxidation current and concentra- tion of NADH observed at the immobilized-Dp electrode in buffer solutions containing: A, 0.05; B, 0.01; C, 0.15; and D, 0.20 mmol dm-3 FMA. Temperature, 30 "C; E = 0.2 V versus SCEANALYST, AUGUST 1991, VOL. 116 r I m u (? r - E 0.015 E E 0.010 . 0.005 795 - - - ,"' . ..' " electrode and the active centre of D p to oxidize NADH to NAD+ efficiently. The is, values increased linearly with the concentration of NADH (slope, 0.0235 mA dm3 mmol-1). The linear range depended on the concentration of FMA as shown in Fig.4. At an FMA concentration of 0.05 mmol dm-3, the is, value deviated from linearity in the range above 0.04 mmol dm-3. However, linearity was observed over a relatively wide range of concentrations (0.005-0.125 mmol dm-3) when the concentration of FMA was 0.20 mmol dm-3. The immobilized-Dp electrode showed outstand- ing stability; the variation of the is, value for repeated measurements (160 samples) over 3 months was +5%. The Dp-catalysed oxidation of NADH by FMA+ is com- posed of two basic reactions:20 2FMA+ + Dp(red) -+ 2FMA + Dp(ox) (1) Dp(ox) + NADH + Dp(red) + NAD+ + H+ (2) where Dp(red) is the reduced form of diaphorase and Dp(ox) is the oxidized form. When the concentration of NADH is small, the over-all reaction is controlled by reaction (2) and the is, value increases with the concentration of NADH.As the concentration of NADH increases, the relative contribution of reaction (1) to the over-all reaction becomes important. The above explains the deviation of the is, values from linearity at high concentra- tions of NADH. The relative contribution of reactions (1) and (2) to the over-all reaction is also governed by the concentra- tion of FMA+. The linear range of the is, versus the concentration of NADH plot becomes wider with increasing concentration of FMA. ,' Determination of LDH and HLADH Activity The determination of LDH activity for pyruvate was carried out in the presence of a specific amount of LDH (20-fold dilution of the enzyme as purchased), 0.2 mmol dm-3 FMA, 0.1 mmol dm-3 NADH and various concentration of pyru- vate.The initial concentration of NADH was large compared with the K , value of LDH for NADH (described later). Therefore, the over-all enzyme reaction should be controlled by the reaction between LDH and pyruvate. The decrease in the concentration of NADH along with the enzyme catalysed t c C 2 3 0 LDH NaP Time - Fig. 5 Time dependence of the current responses at immobilized-Dp electrode. Sodium pyruvate (Nap) A, 0.06, B. 0.16 and C, 0.50 mmol dm-3, was added to the buffer solution containing 0.20 mmol dm-3 FMA, 0.10 mmol dm-3 NADH and LDH (20-fold dilution of the original bottle). Temperature, 30 "C Table 1 HLADH ass;y tor cqclohexanone derivatives. All values in U m l 1 Method Substrate* Electrochemical Spectrophotometric C yclohexanone 3.680 4.820 2-Methylcyclohexanone 0.083 0.072 3-Methylcyclohexanone 1.960 1.970 * All substrates at a concentration of 0.1 mmol dm--3.reduction of pyruvate was monitored from the oxidation current for NADH at the immobilized-Dp electrode. Fig. 5 shows the decay in the oxidation current immediately after the addition of pyruvate. From the initial velocity (vg) of the decay in the concentration of NADH, the activity of the original LDH suspension €or pyruvate was calculated to be about 5000 U ml-1, which is in good agreement with the values deter- mined by the spectrophotometric method. The HLADH activities for three cyclohexanone derivatives (cyclohexanone, 2-methyl- and 3-methylcyclohexanone) were also determined by the proposed electrochemical methods.The results are summarized in Table 1. The values determined by the present procedure are in good agreement with those obtained by the spectrophotometric method. This table also shows large differences in activity among the substrates. These differences show the specificity of HLADH for cyclohexanone derivatives .zl Kinetic Study of LDH The K , and V, values for the oxidation of pyruvate catalysed by LDH were also determined by the proposed electrochem- ical method. First, the dependence of v0 on the initial concentration of NADH in the presence of an excess of pyruvate (0.5 mmol dm-3) (Fig. 6) was investigated. The value of vo was found to be constant when the concentration of NADH was >0.075 mmol dm-3. At such concentrations of NADH, the binding sites of Dp for NADH are saturated and thus the enzyme reaction is controlled by the reaction between Dp and pyruvate.Therefore, the kinetic studies of LDH for pyruvate were carried out in the presence of 0.1 mmol dm-3 NADH, which was in the region of saturation. The kinetic measurement for LDH was carried out at +0.20 V versus SCE. The immobilized-Dp electrode was first immersed in a solution containing FMA, NADH and LDH. The electrode was held for 1 min, then various amounts of > 0 . 1 5 1 T O I I I I I 1 1 [NADH]/mmol dm-3 0 0.05 0.1 0.15 0.2 0.25 Fig. 6 Relationship between initial rate (vo) and the concentration of NADH measured by spectrophotometric method. Measurements were carried out in a phosphate buffer solution (2 ml) containing 0.5 mmol dm-3 pyruvate and LDH (20-fold dilution of the original bottle).Temperature, 30 "C 0.020 ~ Fig. 7 Double reciprocal plot for kinetic studies of LDH observed by 0, the immobilized-Dp electrode and A, the spectrophotometric method. Concentrations of FMA and NADH were 0.20 and 0.10 mmol dm-3, respectively. Temperature, 30 "C796 ANALYST, AUGUST 1991, VOL. 116 pyruvate were injected into the solution. The rate of enzyme activity of LDH on pyruvate was monitored by the decrease in the concentration of NADH by using the immobilized-Dp electrode. The vo value at the various concentrations of pyruvate can be determined from the slope of the initial decay curve. The Lineweaver-Burk plot for vo and concentration of pyruvate is linear as shown in Fig.7. The K, and V , values determined from the slope and intercept were 0.097 and 0.161 mmol dm-3 min-1, respectively. These values are in good agreement with those obtained from the ordinary spectro- scopic method. In conclusion, the above results demonstrate that the immobilized-Dp electrode has excellent capability of sensing NADH in a solution. The Dp immobilized at the electrode surface showed a good stability and no obvious decrease in the activity was found for at least 3 months. Optimization of the immobilization conditions (e.g., the ratio of Dp to glutar- aldehyde etc.) would improve the sensitivity of the immobi- lized-Dp electrode. A variety of chemicals can be detected by using immobilized-enzyme electrodes co-immobilized with other NADH dependent enzymes.The electrochemical be- haviour of the immobilized-enzyme electrodes co-immobi- lized with an L-amino acid dehydrogenase is now being investigated for the detection of a specific L-amino acid.22 The present system can also be widely applied to assays and kinetic studies of various types of NADH dependent enzyme reactions. References 1 Johnson, D. C., Ryan, M. C., and Wilson, G. S., Anal. Chem., 1986.58.33R. 2 Yokoyama, K., Tamiya, E., and Karube, I., J. Electroanal. Chem., 1989,273, 107. 3 You, K.-s., CRC Crit. Rev. Biochem., 1985, 17,313. 4 Chenault, H. K., and Whitesides, G. M., Appl. Biochem. Biotech., 1987, 14, 147. 5 Lehninger, A. L., in Principles of Biochemistry, North Pub- lisher, New York., 1982, ch. 17. 6 Moiroux, J., and Elving, P. J., Anal. Chem., 1978, 50, 1056. 7 Jaegfeldt, H., J. Electroanal. Chem., 1980, 110, 295. 8 Matsue, T., Suda, M., Uchida, I., Kato, T., Akiba, U., and Osa, T., J. Electroanal. Chem., 1987, 234, 163 and references cited therein. 9 Tse, D. C.-S., and Kuwana, T., Anal. Chem., 1978,50, 1315. 10 Schelter-Graf, A., Schmidt, H. L., and Huck, H., Anal. Chim. Acta, 1984, 163,299. 11 Torstensson, A., and Gorton, L., J. Electroanal. Chem., 1981, 130, 199. 12 Gorton, L., J. Chem. Soc., Faraday Trans. I , 1986, 82, 1245. 13 Yon Hin, B. F. Y., and Lowe, C. R., Anal. Chern., 1987, 59, 2111. 14 Green, M. J., and Hill, H. A. O., J. Chem. SOC., Faraday Trans. 1, 1986, 82, 1237. 15 Yao, T., and Wasa, T., Anal. Chirn. Acta, 1985, 175,301. 16 Cass, A. E. G., Davis, G., Green, M. J., and Hill, H. A. O., J. Electroanal. Chem., 1985, 190, 117. 17 Miki, K., Ikeda, T., Todoriki, S., and Senda, M., Anal. Sci., 1989, 5 , 269. 18 Matsue, T., Yamada, H., Chang, H.-c., Uchida, I., Nagata, K., and Tomita, K., Biochim. Biophys. Acta, 1990, 1038, 29. 19 Winer, A. D., J. Biol. Chem., 1964, 239,3598. 20 Matsue, T., Yamada, H., Chang, H.-c., and Uchida, I., Bioelectrochern. Bioenerg., 1990, 24, 347. 21 Dulton, H., and Branden, C.-I., Bioorg. Chem., 1981, 10, 1. 22 Chang, H.-c., Yamada, H., Ueno, A., Matsue, T., and Uchida, I . , Denki Kagaku, 1990,58, 1211. Paper 0/03562I Received August 6th, 1990 Accepted January 28th, I991