**2. Electrochemical oxidation of NAD(P)H**

Given that the nicotinamide coenzymes are electron carriers, and therefore by definition electroactive, it would appear at first sight (points 2 and 3 above) that these systems would be ideal candidates for commercial enzyme electrode devices. The electrochemical oxidation of NADH to NAD+ is however both chemically and kinetically complicated at common electrode surfaces such as gold, platinum or glassy carbon.

The electrochemical detection of NADH has generated great interest because the pyridine nucleotides NAD+ and NADP+ are ubiquitous in all living systems and are required for the reactions of more than 450 oxidoreductases. [7] Although the formal potential of NADH/ NAD+ couple in neutral pH at 25 C is estimated to be 0.56 vs. SCE [8, 9] significant over-potential is often required for the direct oxidation of NADH at bare electrodes. [10] Unlike in nature where the oxidation of NADH occurs as a 1-step hydride transfer, on bare electrodes the reaction has been shown to occur via a different and higher energy pathway which produces biologically inactive NAD. radicals as intermediates. The large amount of energy required to produce these intermediates is the origin of the large overpotential (typically 1 Volt) required at bare electrodes. As a result, the direct electrochemical oxidation of NADH has been shown to produce a mixture of products including biologically active NAD+, (NAD)2 dimers and products from the side reactions of the electrogenerated NAD. radicals.

In addition, the direct oxidation of NADH is often accompanied by electrode fouling due to the polymerisation oxidation products on the electrode surface. [11] The fouling of the electrode surface can occur by two mechanisms. First, the NAD. radicals interact directly with oxide functionalities on the surface of the electrode and second the reaction product, NAD+, adsorbs onto the electrode surface. Both of these processes are irreversible and result in the gradual blocking of the electrode during continued oxidation. It is this electrode fouling that results in the irreproducibility of the analytical signal from bare electrodes. In addition, if this method were to be used in commercial glucose sensors for SMBG, the high over potentials required would result in the response being a combination of signals from the oxidation of both NADH and common interferents, e.g. ascorbic acid, uric acid, paracetamol etc. found in blood.

322 Dehydrogenases

response to an applied voltage. The resultant current is monitored by the meter and then interpolated into an accurate measurement of glucose using on-board software algorithms

The final method of analysis used will ultimately depend on several properties of the

The answer to the first three criteria will depend largely on the system under investigation. The answer to the latter three depends on the requirements and application of the sensor under consideration. If the enzyme does not contain any redox groups, then the method of analysis will be restricted to monitoring either the release of products or the consumption of substrate by their reaction at the transducing electrode. The current produced can then be related to the concentration of analyte. Of particular relevance to this article, are the devices that incorporate nicotinamide adenine dinucleotide (NAD), flavin adenine dinucleotide (FAD)

Given that the nicotinamide coenzymes are electron carriers, and therefore by definition electroactive, it would appear at first sight (points 2 and 3 above) that these systems would be ideal candidates for commercial enzyme electrode devices. The electrochemical oxidation of NADH to NAD+ is however both chemically and kinetically complicated at common

The electrochemical detection of NADH has generated great interest because the pyridine nucleotides NAD+ and NADP+ are ubiquitous in all living systems and are required for the reactions of more than 450 oxidoreductases. [7] Although the formal potential of NADH/ NAD+ couple in neutral pH at 25 C is estimated to be 0.56 vs. SCE [8, 9] significant over-potential is often required for the direct oxidation of NADH at bare electrodes. [10] Unlike in nature where the oxidation of NADH occurs as a 1-step hydride transfer, on bare electrodes the reaction has been shown to occur via a different and higher energy pathway which produces biologically inactive NAD. radicals as intermediates. The large amount of energy required to produce these intermediates is the origin of the large overpotential (typically 1 Volt) required at bare electrodes. As a result, the direct electrochemical oxidation of NADH has been shown to produce a mixture of products including biologically active NAD+, (NAD)2 dimers and

In addition, the direct oxidation of NADH is often accompanied by electrode fouling due to the polymerisation oxidation products on the electrode surface. [11] The fouling of the electrode surface can occur by two mechanisms. First, the NAD. radicals interact directly with oxide

giving the user a concentration value in typically less than 7 seconds.

or pyrroloquinoline quinone (PQQ) dependent dehydrogenase enzymes.

enzyme. The main considerations are;

1. does the enzyme contain redox active groups

4. what is the required speed of response

3. is one of the substrates or cofactors electroactive

5. what will be the final application of the sensor

**2. Electrochemical oxidation of NAD(P)H**

electrode surfaces such as gold, platinum or glassy carbon.

products from the side reactions of the electrogenerated NAD. radicals.

2. are the products of the biochemical reaction electroactive

The problems mentioned above can be overcome by using small catalytic molecules called mediators. These molecules can transfer charge from the reduced coenzyme directly to the electrode thereby bypassing the direct oxidation of NADH. The use of this approach has three potential advantages when compared to the direct oxidation of the reduced nicotinamide. First, by judicious choice of the mediator problems associated with electrode fouling or competing reactions can be avoided (especially true if he chosen mediator has a site for hydride transfer). Virtually 100% of the oxidation product via this mechanism is biologically active NAD. Second, the rate of electron transfer between the NADH molecule and the mediator can be enhanced resulting in a more stable increased signal. The rate at which NADH is chemically oxidised will governed to a large extent by the potential difference between the formal redox potential of the two species. Finally, a mediator can be selected with a redox potential that will limit the effects of interference. This is particularly relevant in cases where the bio-analyte is present in blood at low concentrations as in the case of beta-hydroxybutyrate (0.0 – 0.3mEq/L). The electrocatalytic oxidation of NADH at a modified electrode is shown schematically in Figure 1. Of particular

**Figure 1.** Cyclic voltammograms of the o-AP modified GC electrode in phosphate buffer (a) in the absence and in the presence of increasing concentrations of NADH/mM: (b) 0.1, (c) 0.3 and (d) 0.5; and (e) 0.6 at bare GC in phosphate buffer (0.1 M, pH 7.0). Scan rate = 10 mV/s. Note, the peak oxidation for NADH at the bare electrode occurs at 0.6 Volts. When the electrode is modified, in this case with oamino phenol the oxidation of NADH is concomitant with the oxidation of the immobilised redox couple (traces b – d). *Figure and data adapted from reference 11.* 

interest are voltammograms a and e. These show the direct oxidation of NADH (e) at the glassy carbon electrode and the cyclic voltammogram (a) of the immobilised species. Addition of NADH to the solution causes an increase in 1. Note that the signal due to the oxidation of NADH now occurs at the potential of the mediator resulting in a decrease in the operating voltage of approximately 0.3 V.

Amperometric Glucose Sensors for Whole Blood Measurement Based on Dehydrogenase Enzymes 325

It is important that modified electrodes designed for the reoxidation of NADH are stable over their stated shelf-life, respond only to NADH and not to any other species present in the blood sample satisfy the following kinetic requirements; i) the reaction between the NADH molecule and the mediator is fast, ii) the transfer of charge within the mediating layer is fast and iii) the electron transfer between the reduced mediator and the electrode is

In the scientific literature, NADH oxidation at chemically modified electrodes is most commonly suggested to occur via a two-step reaction mechanism. In the first step, NADH forms a charge transfer complex with the oxidised form of the mediator bound to the electrode surface. In the second step, electron exchange takes lace and the complex breaks down producing NAD+ and a reduced mediator site. Because the electrode is polarised, the reduced mediator site is reoxidised in a non-rate limiting electron exchange to the bulk electrode material. This scheme is shown in Figure 3. The important kinetic constants are also

**Figure 3.** Two-step mechanism commonly proposed for the oxidation of NADH at chemically modified electrodes. The mediator and NADH form a charge transfer complex that dissociates to give rise to the

For this type of mechanism, the catalytically limiting current (ik) observed under controlled hydrodynamic conditions can be expressed as: ik = nFAKcatCNADH/KM where CNADH is the bulk concentration of NADH and represents the concentration of binding sites in the immobilised film. This type of model assumes that the rate of electron transfer between the mediating species in the film and the NADH is sufficiently fast so as not to be rate limiting, the NADH freely diffuses into the film whereupon it adsorbs to the catalytic site and it undergoes oxidation to NAD+. Also, the expression for *ik* is valid only for thin films where

Due to the formation of the charge-transfer complex, this reaction scheme is commonly analysed using Michaelis-Menten kinetics. From figure 3 it is possible to construct the

This is now a straight line plot. From the slope of such a plot, values of *k+2* can be calculated. By extrapolation of zero NADH concentration, i.e. the intercept, values of *KM* can be estimated. Values for *kOb*s can be obtained from Koutecky-Levitch plots under steady state

the concentration of NADH is insufficient to saturate the binding sites.[12]

**3. Kinetic modelling of NADH oxidation at chemically modified** 

also fast.

**electrodes.** 

represented.

reduced mediator and biologically active NAD+.

following kinetic argument.

Among the mediators used so far are quinones, diimines, ferrocene, thionine oxometalates, polymetallophthalocyanines, ruthenium complexes, pyrroloquinoline quinone, fluorenones, and quinonoid redox dyes such as indamines, phenazines, phenoxazines and phenothiazines.11 A generalised reaction schematic for the mediated oxidation of NADH is shown below in Figure 2.

**Figure 2.** Scheme showing the mediated oxidation of NADH. In this context, the mediator reacts with NADH in a chemical step to oxidise it (a) and is itself reduced. The mediator is then itself reoxidised at the electrode surface (step b). The chemical oxidation of NADH by this mechanism bypasses the problems associated with the direct oxidation at a bare electrode surface.

To design an NADH sensor, the mediator is normally immobilized on the electrode surface or within the electrode material. Mediators can be immobilized by chemisorption by covalent attachment directly electrode surface or by electrochemical polymerisation of the mediators at the electrode surface or, alternatively via covalently attached/physically entrapment in polymers, incorporation in carbon paste grown at electrode surface or deposited on the electrodes by drop coating. The method that is ultimately chosen to produce the modified electrode depends upon the method of mass production (for commercial sensors) and the materials used in the device. For example, in sensors that utilise screen printed carbon based electrodes (*screen-printing technology is a kind of low-cost thick film technology which allows to deposit thick films, a few to hundreds micrometers and is well suited for mass production and portable devices. Such a micro fabrication route offers high-volume production of extremely inexpensive and yet highly reproducible disposable enzyme electrodes – this will be discussed further in the text*) it is convenient to incorporate the mediator directly into the carbon ink particularly if the mediator contains delocalised aromatic rings, as found in quinine and phenoxazine dye based mediators, which form strong chemisorbed bonds with the carbon and graphite plates. The mediator loading, activity stability etc can all be investigated using conventional electrochemical techniques such as DC cyclic voltammetry. It is important that modified electrodes designed for the reoxidation of NADH are stable over their stated shelf-life, respond only to NADH and not to any other species present in the blood sample satisfy the following kinetic requirements; i) the reaction between the NADH molecule and the mediator is fast, ii) the transfer of charge within the mediating layer is fast and iii) the electron transfer between the reduced mediator and the electrode is also fast.
