Horseradish peroxidase (EC 1.11.1.7, 300 U/mg) was purchased from Aladdin Reagent Co. Ltd (Beijing, China). Bovine serum albumin (BSA) was obtained from NewProbe Chemicals Co. (Beijing, China). Hydrogen peroxide (30%, w/v solution) was purchased from Yingdaxigui Reagent Company (Tianjin, China) and the concentration of the more diluted hydrogen peroxide solutions prepared from 30% hydrogen peroxide was determined by titration with potassium permanganate. Phenylhydrazine was obtained from Sinopharm Chemical Reagent Co. Ltd (Beijing, China). Phosphate buffer solution (PBS, pH 7.0) contained 3.8 M KH₂PO₄ and 6.2 M Na₂HPO₄. All other reagents were of analytical grade.

Prior to modification, the glassy carbon electrode surface was polished with 0.05 mm alumina paste on a micro-cloth and subsequently ultrasonically cleaned thoroughly with acetone, NaOH (1:1), HNO₃ (1:1), and doubly distilled water and then dried at room temperature. A mixed solution of 2.0 U/mL HRP and 0.5% (w/w) BSA was prepared by dissolving HRP and BSA in 0.2 mL PBS (50 mM, pH 7.0). A volume of 3 mL of the mixed solution was then added dropwise on the surface of the glassy carbon electrode. The surface area of the electrode is 0.28 cm². The HRP/BSA electrode was cross-linked by placing the electrode in a closed vessel contained 25% glutaraldehyde and water vapor for 20 min and dried at room temperature for 1 h and stored at 4 °C until use.

Cyclic voltammetry and amperometric experiments were performed with a LabChem 10M electrochemical workstation (Tianjin, China) and a conventional three-electrode system. The modified glassy carbon electrode served as working electrode, a platinum wire as counter electrode, and a saturated calomel electrode (SCE) as reference electrode. A 0.10 M pH 7.0 phosphate buffer solution (PBS) was used as supporting electrolyte. A magnetic stirrer (approximately 400 rpm) was employed during the amperometric measurements. All experiments were carried out at room temperature.

The electrochemical properties of the electrode modified were characterized by cyclic voltammetry. Electrochemical experiments were performed in a conventional electrochemical cell containing a three electrode arrangement and the potential swept from −0.4 to 0.2 V (vs. SCE) at a sweeping rate of 0.1 V/s. The cyclic voltammetry experiment was performed in 5 mL pH 7.0 PBS. All experiments were conducted at 25 °C.

Electrochemical experiments were undertaken in the electrochemical cell containing 25 mL of the supporting electrolyte. Amperometric measurements were carried out in a stirred solution (approx. 400 rpm). The electrodes were immersed into a stirred phosphate buffer solution (pH 6.5) containing 1.0 mM hydroquinone as the electron mediator and an initial baseline current was recorded. 0.1 mM H₂O₂ solution was added to get an increased steady-state current record. Subsequently, solution of increasing phenylhydrazine concentration was added with a micropipette in succession and the current decrease was recorded. All experiments were conducted at 25 °C.

Pearson correlation was a measurement of the strength and direction of a linear relationship between two groups of data. Statistical significance is determined for a p-value <0.05 for all tests. Strong correlation is determined for an absolute r value >0.7; weak correlation > 0.5. The bi-variate correlation analysis was carried out by the R statistical package (version 2.13.0; http://www.r-project.org/) [37]. The linear regression analysis was performed by the online Free Statistics and Forecasting Software (version 1.1.23-r6; http://www.wessa.net/slr.wasp).

The following reaction sequence can be proposed for the case of mediated electrocatalytic reduction of H₂O₂ at HRP electrode [38,39]:

where HRP is horseradish peroxidase. Fe³⁺ is the HRP cofactor. Compound (I) (oxidation state +5) and Compound(II) (oxidation state +4) are oxidized intermediates. QH₂ and Q represent hydroquinone and its oxidized form (benzoquinone), respectively. In these reactions, a two-electron oxidation of the ferriheme moiety of HRP enzyme is caused by H₂O₂. The resulting intermediate Compound (I) consists of oxyferryl iron and a porphyrin π cation radical. Compound (I) is reduced to Compound (II) and then it was further reduced to the original form of HRP (Fe³⁺) by the mediator QH₂. Q is finally reduced back to QH₂ by a rapid reaction involving the acceptance of two electrons from the electrode.

The inhibition of phenylhydrazine to HRP is an anticompetitive inhibition [39]. When the inhibitor (phenylhydrazine) is added into the electrolyte solution, it can combine with Compound (I) causing the decrease of the current i. The possible mechanism of inhibition could be represented as follows: (2) HRP ( Fe 3 + ) + H 2 O 2 ⟶ k 1 Compound ( I ) ⟶ k 2 … + I ⥮ k 1 Compound ( I ) ‐ Iwhere compound(I) - I represents complex of compound(I) with inhibitor.

After adding substrate or inhibitor dropwise, the current rapidly returns toward equilibrium (steady state) with increasing time. For the pre-steady-state part of enzymatic reaction, the current obeys exponential decay equation, i.e.,

where i is the faradaic current at time t. i_∞ is the steady-state current. τ is relaxation time, being equal to the time constant of an exponential return of a faradaic current to equilibrium after adding substrate or inhibitor, which reflects the rate at which excited current return to steady-state current.

Suppose that a steady-state current (i_∞) is the current when the biosensor system approaches to equilibrium as t → ∞:

where i(t) is the current at time t. In practice, the current at the response time t_R is assumed as i_∞. t_R is defined as the time when the absolute current slope falls below a given small value (ε < 0.0001) [27], i.e.,

A reduction current (i_∞) observed on electrode is a combination of the mass transport-limited current (i_∞,L) and the reaction-rate-limited current (i_∞,K) [40], accords with the Koutecky-Levich equation, i.e.,

For a rotating disk electrode, the mass-transport-limited current (i_∞,L) depends on the angular rotation velocity (ω) of the electrode and the bulk concentration ([H₂O₂]) of H₂O₂, i.e., i_∞,L is determined by the Levich equation [22,41]:

where n is the number of electrons transferred to the enzyme in one catalytic cycle (n = 2). F is the Faraday constant. A is the area of the electrode surface. D is the diffusion coefficient of H₂O₂. ν is the kinematic viscosity of water. For a given biosensor that is rotating at a constant angular velocity, i_∞,L is proportional to [H₂O₂], Equation (7) reduces to:

where a₁ is constant (a₁ = 0.62nFAD^2/3ν^−1/6ω^1/2).

When the transport of H₂O₂ is high enough to keep its concentration at the electrode surface equal to that in the bulk solution, the reaction-rate-limited current i_∞,K is governed by the ping-pong kinetic scheme, being expressed by [24,39]:

Where Γ is the surface concentration of HRP enzyme. [QH₂] is the concentration of hydroquinone. [I] is the concentration of phenylhydrazine inhibitor. Assuming that a₂ = 1/nFAΓk₁, b = 1/nFAΓk_I, and c₁ = (k₂ + k₃)/(nFAΓk₂k₃[QH₂])+1/nFAΓ, Equation (9) can be also reduced to:

Substituting Equations (8) and (10) into Equation (6), the Koutecky-Levich equation can be reduced to:

where a₃ = a₂ + 1/a₁.

Combining b[I] term with constant c₁ in Equation (11) in a constant phenylhydrazine concentration, we obtain c₂ = c₁ + b[I]. The dependence of the steady-state current on the substrate concentration is written as:

Combining a₃/[H₂O₂] term with constant c₁ in Equation (11) in a constant H₂O₂ concentration, we obtain c₃ = c₁ + a₃/[H₂O₂]. The dependence of the steady-state current on the inhibitor concentration is written as:

Horseradish peroxidase (HRP) and bovine serum albumin (BSA) are immobilized on the surface of a glassy carbon (GC) electrode. Hydroquinone/quinone mediator facilitates the electron transfer between protein and the electrode. Experiments are performed with H₂O₂ to ascertain if HRP is indeed immobilized on the electrode.

Figure 1(a) depicts the cyclic voltammetry curves of the proposed biosensor in the presence and absence of H₂O₂. A redox couple is observed in the absence of H₂O₂. However, an enhancement of the peak current is observed in the presence of H₂O₂. On the other hand, it has been reported that phenylhydrazine is able to inhibit or abrogate HRP activity by competing with the normal hydroquinone [42,43]. The inhibition effect of phenylhydrazine is measured by using the proposed biosensor.

Upon 19 successive additions of 0.1 mM H₂O₂ substrate to the electrolyte solution, i increases with [H₂O₂]. Subsequently, upon 11 successive additions of 0.5 mM phenylhydrazine inhibitor to the electrolyte solution, i decreases with increasing [I] see Figure 1(b).

The steep increased (or decreased) in magnitude of the current is observed when H₂O₂ (or phenylhydrazine) is added into the electrolyte solution, and then the current returns to equilibrium (steady state i_∞) following an exponential decay with a certain recovery speed (relaxation time τ). The values of i_∞ and τ are best computed by non-linear regression of data fitted to Equation (2). These values at different substrate concentration are listed in Table S1 (Appendix); they at different inhibitor concentration in Table S2 (Appendix).

From Figure 2(a), the relation between i_∞ and [H₂O₂] is not linear. The calibration curve is valid only under low concentration of substrate conditions. More specifically, when [H₂O₂] < 0.4 mM, the relation is very close to a straight line (calibration curve): i_∞ = −7.86(±0.074) × [H₂O₂] − 0.025(±0.018) with r = −0.999; p < 0.0001. This linear correlation is a prerequisite for applying the electrochemical biosensors. Thus electrocatalysis reduction to H₂O₂ with HRP electrode can detect trace levels of H₂O₂. In fact, our result shows the steady-state current follows the Michaelis-Menten kinetic model within a broader concentration range. Kinetic constants are determined by fitting the initial rate data to Lineweaver-Burk plot Figure 2(b). Thus, the current is regarded as a relative enzyme reaction rate. This methodology could monitor in real-time subtle changes in enzyme activity.

The pre-steady-state time course of the i → i_∞ process is fitted to Equation (3) for a single exponential decay, in which τ is the apparent relaxation time for output current (i.e., the reciprocal of the apparent first-order rate constant). It is found, however, that τ is not a simple function of the substrate concentration. As shown in Figure 3, τ is low in both the low- and high-concentration regions; a relatively larger τ occurs in at a moderate concentration ([H₂O₂] = 0.8–1.4 mM). This suggests that, within this concentration range, the rate at which the equilibrium is approached is slowed down.

If the kinetic control becomes operative, there is a possible explanation: a “shoulder” appears on the curve of t_0.5versus[H₂O₂] in the two-dimensional-in-space biosensor model, as has been proposed in reference [27]. Where t_0.5 is the half time (t_0.5 = 0.693τ). Baronas et al.[27] showed that the shoulder occurs at [H₂O₂] near to the Michaelis constant K_m. At [H₂O₂] ≪ K_m (τ = 1/k₁K_m) the reaction kinetics for [H₂O₂] is a zero order throughout the biosensor, whereas for [H₂O₂] ≫ K_m (τ = 1/k₁[H₂O₂]) the kinetics is a first order throughout. At intermediate values of [H₂O₂] the kinetics undergoes a transition from zero order to first order. This phenomenon is also manifested in inhibition assays, so it is not obvious whether agreement between our data and the shoulder of t_0.5 is not a coincidence. Inhibition kinetics also experiences a transition from zero order to first order, but it is not proof.

Phenylhydrazine can inhibit HRP activity [42,43]. In our biosensor, i_∞ decreased steadily as [I] is increased. However, the Lineweaver-Burk plot for the substrate dependence shows that the reaction does not follow Michaelis-Menten kinetics, as the curve connecting the experimental points is clearly curved Figure 4(a). However, there is a more simple linear relationship between i_∞ and [I]. The dependence of the rate of i_∞ change on [I] is illustrated in Figure 4(b). As shown in the figure, the regression line is discontinuous at inhibitor concentration of 3–4 mM. The [I]-i_∞ relation is broken into two individual lines. When [I] < 3 mM, decrease of i_∞ along a regression line is given by the equation: i_∞ = −1.92(±0.075) × [I] − 9.19(±0.03) with r = −0.999; p < 0.0001. When [I] > 4 mM, decline of i_∞ along another line is given by the equation: i_∞ = −1.41(±0.024) × [I] − 8.54(±0.1) with r = −0.999; p < 0.0001.

A similar phenomenon occurs in the relation between [I] and τ as well. In Figure 5, τ increases exponentially as the inhibitor concentration increases. However, in the concentration of 3–4 mM, the points determined experimentally is lower than τ value extrapolated from low concentration and thus the curve displays a drop.