Heme protein-gluten films: voltammetric studies and their electrocatalytic properties

Heme protein-gluten films: voltammetric studies and their electrocatalytic properties

Analytica Chimica Acta 481 (2003) 91–99 Heme protein-gluten films: voltammetric studies and their electrocatalytic properties Hongyun Liu, Naifei Hu∗...

126KB Sizes 0 Downloads 89 Views

Analytica Chimica Acta 481 (2003) 91–99

Heme protein-gluten films: voltammetric studies and their electrocatalytic properties Hongyun Liu, Naifei Hu∗ Department of Chemistry, Beijing Normal University, Beijing 100875, China Received 29 July 2002; received in revised form 3 January 2003; accepted 8 January 2003

Abstract Direct electrochemistry and electrocatalysis of heme proteins, such as hemoglobin (Hb), myoglobin (Mb), and horseradish peroxidase (HRP), incorporated in gluten biopolymer films cast on pyrolytic graphite (PG) electrodes, were studied by voltammetry and amperometry. All the three protein-gluten films exhibited a pair of well-defined, quasi-reversible cyclic voltammetric peaks at about −0.28 V versus saturated calomel electrode (SCE) in pH 5.5 buffers, respectively, characteristic of the heme Fe(III)/Fe(II) redox couples, indicating enhanced electron transfer between the proteins and PG electrodes in a gluten film environment. The protein-gluten hydrogel films showed excellent stability. Positions of Soret absorption band of protein-gluten films suggested that the heme proteins kept their secondary structure similar to their native state in the films in the medium pH range. The heme proteins in gluten films were act as a biologic catalyst to catalyze reduction of oxygen or hydrogen peroxide. The voltammetric or amperometric responses of H2 O2 at the protein-gluten film electrodes could be used to determine the concentration of H2 O2 in solution. © 2003 Elsevier Science B.V. All rights reserved. Keywords: Direct electrochemistry; Electrocatalysis; Hemoglobin; Myoglobin; Horseradish peroxidase; Gluten

1. Introduction Research on protein-containing or enzyme-containing thin films modified on electrode surface is largely driven by their potential applicability in fabricating biosensors, biomedical devices, and enzymatic bioreactors [1,2]. Achieving direct electron exchange between redox proteins or enzymes and electrodes simplifies these devices by removing the requirement of chemical mediators, and thus has a great significance in preparing the third generation biosensors [3]. Direct electrochemistry of redox proteins can also provide a model for the mechanistic study of electron transfer between enzymes in real biological systems. ∗ Corresponding author. Fax: +86-10-6220-0567. E-mail address: [email protected] (N. Hu).

Films modified on electrodes may provide a favorable microenvironment for the proteins to directly exchange electrons with underlying electrodes, and thus afford a new opportunity for the detailed study of the enzyme electrochemistry [4–6]. Our long-term goal is to develop stable protein or enzyme films on electrodes, and realize the direct electrochemistry of immobilized redox proteins, which can be employed as the foundation of preparing new kinds of biosensors or bioreactors. In recent years, our group has studied various film systems in which incorporated redox proteins or enzymes demonstrated direct and quasi-reversible voltammetry at underlying pyrolytic graphite (PG) electrodes [7]. Successful approaches have included cast films of proteins with insoluble surfactants [8], hydrogel polymers [9–11], polyelectrolyte- or clay-surfactant composites

0003-2670/03/$ – see front matter © 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0003-2670(03)00071-0

92

H. Liu, N. Hu / Analytica Chimica Acta 481 (2003) 91–99

[12–16], and clay nanoparticle [17] films. All these films effectively enhanced the direct electron transfer between the heme proteins and electrodes compared to that on bare electrodes with the proteins in solution. In recent years, there has been an increasing interest in using natural polymers as an enzyme immobilization matrix for biosensor construction. Gluten is a protein-based biopolymer and the major by-product from the production of cereal flour. Gluten includes two contrasted fractions: polymerized proteins— glutenins, and monomeric proteins—gliadins [18]. For the glutenin component, the average molecular weight is about 3.0×105 , while for the gliadin component, the average molecular weight is estimated to be 3.6 × 104 [19]. The glutenins form disulfide-bonded polymers and contribute to gluten elasticity, while the gliadins interact by hydrogen bonding and hydrophobic interactions and provide the viscous component of gluten [20]. Gluten proteins form amorphous three-dimensional structures stabilized mainly by disulfide bridges and non-covalent interactions [21]. The gluten matrix is insoluble in water [22]. The advantages of gluten, such as its biodegradability, non-toxicity, low cost, readily availability, and especially its well-defined, three-dimensional network, made it a significant material or matrix for immobilization of biomolecules and organisms [23–26]. For example, gluten has been used as casting membranes to entrap cells containing penicillinase to fabricate a penicillin bioelectrode [23]. A procedure for entrapping cell-associated enzyme penicillin G acylase within gluten matrix has been developed [25]. We thus expected that the gluten biopolymer would also be a good film-forming material for immobilizing redox proteins or enzymes on electrode surface, and in studying the direct electrochemistry of redox proteins. In the present work, three heme proteins, such as hemoglobin (Hb), myoglobin (Mb), and horseradish peroxidase (HRP), were incorporated in gluten films modified on PG electrodes, and designated as protein-gluten films in its general form. Direct voltammetry of protein-gluten films were studied in detail. To our knowledge, this is the first report of direct electrochemistry of heme proteins incorporated in biocompatible and electrochemically inert gluten films. Enhanced, reversible electron transfer between heme proteins and underlying PG electrodes was realized in the gluten film environment. Furthermore, electro-

chemical catalytic reductions of O2 and H2 O2 were observed at protein-gluten film electrodes, showing the potential applicability of the films as biosensor. 2. Experimental 2.1. Regents Bovine erythrocyte hemoglobin (MW 66,000) was from Shanghai Lizhu Dongfeng Biotechnology Limited Company. Horse heart myoglobin (MW 17,800) was from Sigma. Lyophilized horseradish peroxidase (MW 42,100) was from Shanghai Chemical Reagent Company. Gluten from maize was from Fluka. They were all used as received. All other chemicals were reagent grade. H2 O2 was freshly prepared before being used. The supporting electrolyte was usually 0.05 M potassium dihydrogen phosphate buffers at pH 7.0 containing 0.1 M KBr. Other buffers were 0.1 M sodium acetate, 0.1 M boric acid, or 0.1 M citric acid, all containing 0.1 M KBr. The pHs of buffers were regulated with HCl or NaOH solutions. Solutions were prepared with twice distilled demineralized water. 2.2. Preparation of protein-gluten films Gluten suspension (1 mg ml−1 ) was prepared by dispersing gluten in 20% alcohol–water solvents with ultrasonication for about 30 min. Right before preparing the films, the dispersion was ultrasonicated for another 10 min. Prior to coating, basal plane pyrolytic graphite (Advanced Ceramics, geometric area 0.16 cm2 ) electrodes were abraded with metallographic sandpaper of 400 grit while flushing with water. Electrodes were sonicated in pure water for 30 s after polishing. To obtain the best cyclic voltammogram (CV) of protein-gluten films, the experimental conditions for film casting, such as the concentration of heme protein, the ratio of protein/gluten, and the total volume of protein-gluten dispersion, were optimized. Typically, 10 ␮l of the dispersion containing 7.6×10−6 M Hb and 0.5 mg ml−1 gluten was spread evenly onto a freshly abraded PG electrode with a microsyringe for preparing Hb-gluten films. Protein concentrations in similar dispersions for the other films were 9.4 × 10−6 M

H. Liu, N. Hu / Analytica Chimica Acta 481 (2003) 91–99

93

Mb and 7.7 × 10−5 M HRP. A small bottle was fit tightly over the electrode to serve as a closed evaporation chamber so that water was evaporated slowly and more uniform films were formed. The films were then dried overnight in air. The surface concentrations of electroactive proteins in the films (Γ ∗ ) were estimated by using Faraday’s law Q = nAFΓ ∗ [27], where Q is the charge value obtained by integration of CV reduction peak, n the number of electron transfer, A the electrode area, F Faraday’s constant. 2.3. Instrumentation and procedure Cyclic voltammetry, square wave voltammetry (SWV), and amperometry were performed using a CHI 660 electrochemical workstation (CH Instruments). A three-electrode cell was used, where a PG disk coated with films acted as working electrode, a platinum flake as counter electrode, and a saturated calomel electrode (SCE) as reference electrode. Voltammetries of protein-gluten films were carried out in buffers containing no heme protein. Buffers were purged with highly purified nitrogen for about 20 min before a series of experiments. A nitrogen environment was kept over solutions in the cell during experiments. In the experiment with O2 , measured volumes of air were injected through solutions via a syringe in a sealed cell, which had been previously degassed with purified nitrogen. In the experiment with H2 O2 , a few microliters of hydrogen peroxide (2.5 × 10−5 M) were injected by a microsyringe with continuously bubbling N2 during the whole procedure. All experiments were done at room temperature of 18 ± 2 ◦ C. The protein-gluten film electrodes were stored in the refrigerator at 4 ◦ C when they were not used. UV-Vis absorption spectroscopy was done with a Cintra 10e spectrometer (GBC). Sample films were prepared by casting protein-gluten dispersions on optical glass slides and then being dried in air. 3. Results and discussions 3.1. Direct cyclic voltammetric properties of protein-gluten film electrodes When protein-gluten film electrodes were placed into pH 5.5 buffers free of heme proteins, after sev-

Fig. 1. Cyclic voltammograms at 0.2 V s−1 in pH 5.5 buffers for: (a) gluten; (b) HRP-gluten; (c) Mb-gluten; and (d) Hb-gluten films.

eral CV scans, a pair of well-defined, quasi-reversible CV peaks at about −0.28 V versus SCE was observed for HRP-, Mb-, and Hb-gluten films, respectively (Fig. 1b–d). The peak pair was characteristic of heme Fe(III)/Fe(II) redox couple for the heme proteins [28–30]. Gluten films alone coated on a PG electrode showed no CV peak in the same potential range (Fig. 1a). The electrochemical parameters of the three protein-gluten films obtained from CV are listed in Table 1 for comparison. While there are some minor differences in the parameters among these three protein-gluten films, their CV behaviors are quite similar. This is understandable since all the three proteins are the same type of protein which has the same prosthetic group of heme as the electroactive center. The redox peak pairs of protein-gluten films had an approximately symmetric peak shape and nearly equal heights of reduction and oxidation peaks. The peak potential separations were only a few tens of mV. The CV reduction and oxidation peak currents for immobilized heme proteins were found to increase linearly with potential scan rates from 0.05 to 2 V s−1 . Integration of reduction peaks at different scan rates gave nearly constant charge (Q) values. All these are characteristic of quasi-reversible, diffusionless, thin-layer electrochemistry [27]. According to the Q–Γ ∗ relationship [27], the surface concentrations of electroactive proteins in the films (Γ ∗ ) were estimated to be in the range of 2.3 × 10−11 to 4.3 × 10−11 mol cm−2 for different heme proteins. Thus, the fractions of electroactive proteins among the total proteins deposited on PG electrodes were in the range of 1.4–7.3% (Table 1).

94

H. Liu, N. Hu / Analytica Chimica Acta 481 (2003) 91–99

Table 1 Electrochemical parameters of protein-gluten films from CV at scan rate of 0.2 V s−1 in pH 5.5 buffers Protein

Molecular weight

E◦ (V vs. SCE)

Ep (mV)

Total Γ of deposited protein (10−11 mol cm−2 )

Γ ∗ of electroactive protein (10−11 mol cm−2 )

Electroactive protein (%)

Hb Mb HRP

66000 17800 42100

−0.277 −0.284 −0.265

53 45 32

47.6 58.9 323

2.33 4.28 2.86

4.9 7.3 1.4

To investigate the influence of film thickness on the fraction of electroactive proteins, various amounts of Hb-gluten dispersion with the same Hb/gluten ratio were cast on PG electrodes (Table 2). Results showed that while the amount of electroactive Hb usually increased with the film thickness, the fraction of electroactive Hb declined accordingly. This indicates that only those hemoglobin molecules which are closest to the electrode surface would be able to contribute to the observed electron exchange reaction. An increase in buffer pH caused a negative shift in potentials for both cathodic and anodic CV peaks for protein-gluten films. The formal potentials (E◦ ), estimated as the midpoint of cathodic and anodic peak potentials of heme Fe(III)/Fe(II) redox couple, had a linear relationship with pH with a slope of −43.2 mV pH−1 for Hb-gluten films (pH 2.5–12), −46.2 mV pH−1 for Mb-gluten films (pH 2.5–12), and −48.4 mV pH−1 for HRP-gluten films (pH 4.5–12). All these slope values are smaller than the theoretical value of −57.6 mV pH−1 at 18 ◦ C for a single-proton coupled, reversible one-electron transfer [31,32]. The reason for this is as yet unclear. However, the linear relationship between E◦ and pH at least suggests that the electron transfer to the proteins is accompanied by proton transfer. An inflection point appeared in the plot at pH 4.5 for HRP-gluten films. At pH <4.5, E◦ values also varied linearly with pH but with a slope of −32.6 mV pH−1 .

3.2. Estimation of rate constant and other parameters To estimate the apparent heterogeneous electron transfer rate constant (ks ) and other parameters, square wave voltammetry and non-linear regression methods were applied to the protein-gluten films. The theoretical model used here was a combination of a SWV model for monomolecular adsorbates [33] with a formal potential dispersion model, as described in detail previously [34,35]. The procedure employed non-linear regression analysis for SWV forward and reverse curves with a 5-E◦ dispersion model, as in other protein film systems [9–11,34,35]. The analysis of SWV data for the protein-gluten films showed accuracy of fit of the model over a range of amplitudes and frequencies. Examples for Hb-gluten films were showed in Fig. 2. The average ks and E◦ values obtained from fitting SWV data at pH 7.0 for the three protein-gluten films are listed in Table 3 for comparison. Considering the estimation error, the ks values for the three protein-gluten films are very close. All of them are in the same magnitude of order, and close to those obtained from other heme protein film systems [7]. Good agreement of E◦ values was obtained between SWV and CV methods. The formal potentials of heme Fe(III)/Fe(II) redox couple for different protein-gluten films are very close, but different from

Table 2 Surface concentration (Γ ∗ ) and fraction of electroactive Hb for Hb-gluten films with different amounts of gluten and Hb Gluten (␮g)

Hb (␮g)

Total Γ of deposited Hb (10−11 mol cm−2 )

Γ ∗ of electroactive Hb (10−11 mol cm−2 )

Electroactive Hb (%)

1.0 2.5 5 7.5 10

1.0 2.5 5 7.5 10

9.5 24 48 71 95

1.2 2.5 2.0 2.4 2.3

12.6 10.4 4.2 3.4 2.4

Data from CV at scan rate of 0.2 V s−1 in pH 5.5 buffers.

H. Liu, N. Hu / Analytica Chimica Acta 481 (2003) 91–99

95

Table 4 Soret band position of proteins in different forms Protein

Dry protein films (nm)a

Dry protein-gluten films (nm)a

Protein in solutions (nm)

Hb Mb HRP

412 408 404

412 408 403

405b 408c 403d

a

This work. In pH 7.0 buffer [47]. c In pH 5.5 buffer [29]. d In pH 7.0 buffer [48]. b

Fig. 2. Square wave forward and reverse current voltammograms for Hb-gluten films in pH 7.0 buffer solutions at different frequencies. Points represent the experimental SWVs from which the background has been subtracted. The solid lines are the best fit obtained by non-linear regression onto the 5-E◦ dispersion model. SWV condition, pulse height 60 mV; step height 4 mV, and frequencies (Hz): (a) 125; (b) 151.5; (c) 178.5; (d) 200.

those in other films [7]. This confirms a specific effect of the film environment on E◦ of heme proteins that has been reported previously [6,35]. Film components may shift the formal potential through interaction with protein or by their influence on the electrode double-layer. 3.3. UV-Vis spectroscopy of the films

suggesting that the proteins incorporated in dry gluten films have a secondary structure similar to the native state of the proteins. The dependence of the Soret band position of protein-gluten films on pH of external solution was also tested. Taking Hb-gluten films as an example, at pH between 5.5 and 10.0, the Soret band appeared at 412 nm (Fig. 3c–f), same as that of dry Hb and Hb-gluten films, indicating that in the medium pH range, Hb basically maintains its native conformation in a gluten film environment. When pH was changed toward more acidic or basic direction, the Soret band showed blue shift accompanied by the distortion of the peak shape (Fig. 3g–i). At pH 4, for instance, the Soret band shifted to 406 nm and became much broader and

The position of the sensitive Soret absorption band of heme prosthetic group for heme proteins may provide the information about possible denaturation of the proteins [36], and UV-Vis spectroscopy was used to observe the Soret band position of protein-gluten films cast on transparent glass slides. The dry protein-gluten films showed almost the same peak positions of Soret band as those of dry protein films alone (Table 4), Table 3 Apparent heterogeneous electron transfer rate constants (ks ) and formal potentials (E◦ ) for protein-gluten films on PG electrodes in pH 7.0 buffers Films

Hb-gluten Mb-gluten HRP-gluten a

Average ks (s−1 )a

67 ± 8 88 ± 13 73 ± 5

Average E◦ (V vs. SCE) CV

SWVa

−0.334 −0.338 −0.332

−0.338 −0.343 −0.310

Average values for analysis of eight SWVs at frequencies of 100–200 Hz, amplitudes of 60–75 mV, and a step height of 4 mV.

Fig. 3. UV-Vis absorption spectra of Hb and Hb-gluten films on glass slides for: (a) dry Hb film; (b) dry Hb-gluten film; and Hb-gluten films in different pH buffers: (c) pH 5.5; (d) pH 7.0; (e) pH 9.0; (f) pH 10.0; (g) pH 4.5; (h) pH 4.0; (i) pH 11.0.

96

H. Liu, N. Hu / Analytica Chimica Acta 481 (2003) 91–99

smaller, suggesting the denaturation of Hb to a considerable extent at this pH (Fig. 3h). Mb-gluten films showed the UV-Vis spectroscopic behavior at different pH very similar to that of Hb-gluten films. Since HRP-gluten films on glassy slides were not stable in buffers, it was difficult to obtain the information of dependence of Soret band position on solution pH. 3.4. Film stability and interaction of heme proteins with gluten Long-term stability is one of the most important features required for a biosensor. The stability of the protein-gluten films in blank buffers was examined by CV under two different conditions. In the solution study, a PG electrode coated with protein-gluten films was stored in a pH 5.5 buffer all the time and CVs were run periodically. Alternatively, a protein-gluten film electrode was stored in air as its dry form for most of the storing time, and CVs were run occasionally after returning the dry electrode in the buffer solution. Hband Mb-gluten films tested with both methods showed excellent stability. After three weeks of storage, the CV peak potentials showed no changes, and the peak heights decreased less than 10% of those at the initial steady state. However, HRP-gluten films were less stable than Hb- and Mb-gluten films. After 18 days of storage, the peak potentials of HRP-gluten films showed a little negative shift with about 20% decrease of the peak heights compared with those at the initial steady state. Hb- and Mb-gluten films on glass slides were not as stable as on PG electrodes, mainly because of poorer adhesion between the films and glass surface. As mentioned above, HRP-gluten films were not stable on glass slides. The stability of protein-gluten film electrodes is better than that of protein-surfactant films [8] and similar to that of other protein-polymer [9–11,37] films. In the protein-chitosan films, Mb and Hb also demonstrate better stability than HRP [37], although the reason is not clear. To investigate the interaction between the heme protein and gluten, the blank gluten films on PG electrodes were placed into pH 5.5 buffers containing the heme proteins, and CVs were run at different soaking times. In Mb solution, a pair of CV peaks at about −0.28 V grew with immersion time, suggesting increasing amounts of Mb entering the gluten films. CVs representing films fully loaded with Mb were obtained

in about 160 h. When fully loaded Mb-gluten films were removed from the Mb solution and transferred to a buffer containing no Mb at the same pH, their CV response maintained identical to that in the Mb solution. For Hb and HRP, a similar situation was observed but with the fully loaded time of 96 and 54 h, respectively. Both cast and immersing methods showed very similar peak positions and currents for protein-gluten films at the steady state, but the former was more convenient and quantitative, and thus was used for the most of the studies in the present works. The isoelectric point of Mb, Hb, and HRP is 6.8 [38], 7.4 [39], and 8.9 [40], respectively. Thus, all the three heme proteins have positive surface charges in buffers at pH 5.5. The isoelectric point of gluten is around pH 7–8 [41] and gluten also has positive surface charges at pH 5.5. Thus, the driving force for the heme proteins to enter the gluten films should not be electrostatic attraction but mainly hydrophobic interaction between the heme proteins and gluten films. The hydrophobic interaction of heme proteins with gluten would also be mainly responsible for the fairly good stability of protein-gluten films. 3.5. Influence of water on gluten films Weighing method was used to estimate the relative amount of water absorbed by gluten and protein-gluten films. Completely dry gluten and Hb-gluten films cast on glass slides were opaque. After being soaked in water for over 24 h, both gluten and Hb-gluten films were fully swelled and saturated by water. The weighing results before and after hydration showed that water accounted for about 60–70% in fully swelled gluten and Hb-gluten films. Thus, gluten films provided a basically aqueous microenvironment for the heme proteins. It is known that water has key influence on structure and properties of gluten by hydrogen bond interaction between the hydrophilic groups of polypeptides of gluten and water molecules [42]. The addition of water to dry gluten results in the formation of gluten hydrogel and causes a more loosely cross-linked network. Regarding the fully hydrated gluten films in our work, the large water content of about 60–70% was nearly the same in the presence and absence of Hb. This more loosening structure of gluten in its hydrogel form may provide Hb with a more favorable and mainly aqueous microenvironment for

H. Liu, N. Hu / Analytica Chimica Acta 481 (2003) 91–99

transferring electrons with underlying PG electrodes. The large water content of gluten films also increases free volume in the films and thereby allows increased backbone chain segmental mobility [43], which may facilitate diffusion of the heme protein molecules in the films, and partially explain the ability of gluten films to take up Hb from its solution. Gluten is a non-crystalline protein with a considerably amorphous structure, which undergoes a glass transition from the glassy state to a liquid-like rubbery state. Addition of water results in a drop of glass transition temperature (Tg ) [43,44]. For example, Tg of gluten films with 0.8% water content measured by DSC was at about 85 ◦ C, while 22% water content in gluten films caused the Tg value down to about 20 ◦ C [43]. In our present work, thereby, Hb-gluten films fully loaded with water at room temperature were most probably at the more fluid rubbery state. This may also partly explain the enhancement of electron transfer of the heme proteins in gluten films. Dry gluten films are not electrically conductive. However, the conductivity of gluten or protein-gluten films on PG electrodes in buffers measured by the electrochemical instrument was quite good. In solution, the hydrogel gluten films with high water content may contain considerable amounts of small ions. Gluten itself is also ionized in buffers. All these may greatly enhance the conductivity of the gluten or protein-gluten films, and facilitate the electron transfer between the proteins and electrodes. 3.6. Electrocatalytic activity Electrocatalytic reduction of dioxygen by proteingluten films was examined by CV. When a certain volume of air was passed through a pH 5.5 buffer using a syringe, significant increases in reduction peak at about −0.30 V were observed for all the three protein-gluten films (Fig. 4). This increase in reduction peak was accompanied by the disappearance of oxidation peak for heme Fe(II), because heme Fe(II) had reacted with oxygen. An increase in the amount of oxygen in the solution increased the reduction peak current. For gluten films with no heme protein incorporated, the peak for direct reduction of oxygen was observed at about −0.80 V (Fig. 4b), far more negative than the potential of the catalytic peak. Thus, the heme protein in gluten films decreased the reduction

97

Fig. 4. Cyclic voltammograms at 0.2 V s−1 in 5 ml of pH 5.5 buffers for: (a) gluten films with no oxygen present; and for (b) gluten; (c) Mb-gluten; (d) Hb-gluten; (e) HRP-gluten films after 40 ml of air was injected into a sealed cell.

overpotential of oxygen by about 0.5 V. With the same volume of air injected and at the same scan rate, the three protein-gluten films showed different catalytic effect on oxygen, among which HRP-gluten films showed the largest catalytic reduction peak current. For protein-gluten films, the catalytic efficiency expressed as the ratio of reduction peak current in the presence (Ic ) and absence of oxygen (Id ), Ic /Id , decreased with scan rate, also characteristic of electrochemical catalytic reduction [45,46]. With the same scan rate, the catalytic efficiency showed the order of HRP > Hb > Mb in gluten films, the same order as observed in CV reduction peak. The catalytic activity of protein-gluten films toward hydrogen peroxide was also investigated. Taking Hb-gluten films as an example, when H2 O2 was added to a pH 5.5 buffer, compared with the system with no H2 O2 present (Fig. 5c), an obvious increase of the reduction peak at about −0.30 V was observed with the disappearance of the oxidation peak (Fig. 5d). The reduction peak current increased with the concentration of H2 O2 in solution (Fig. 5e). However, direct reduction of H2 O2 at blank gluten film electrodes was not observed (Fig. 5b). The catalytic reduction of H2 O2 at protein-gluten film electrodes was used to determine H2 O2 concentration quantitatively in solution by CV. The calibration curves for the three protein-gluten films indicate that the linear range of H2 O2 analysis is 5.0 × 10−6 to 8.0 × 10−5 M for Mb-gluten films, 2.5 × 10−6 to 9.5 × 10−5 M for Hb-gluten films, and 2.5 × 10−6 to 1.55 × 10−4 M for HRP films. An inflection

98

H. Liu, N. Hu / Analytica Chimica Acta 481 (2003) 91–99

Fig. 5. Cyclic Voltammograms at 0.2 V s−1 in pH 5.5 buffers for: (a) gluten films with no H2 O2 present; (b) gluten films with 0.03 mM H2 O2 present; (c) Hb-gluten films with no H2 O2 present; (d) Hb-gluten films with 0.03 mM H2 O2 present; and (e) Hb-gluten films with 0.06 mM H2 O2 present.

point was observed in the calibration plot for the three protein-gluten films, respectively. While HRP-gluten films showed the smallest slope for the linear calibration curve, they demonstrated the lowest detection limit and the largest linear range among the three protein-gluten films. The electrocatalytic reduction of hydrogen peroxide at protein-gluten film electrodes was also studied

by amperometry, which is one of the most widely employed techniques for biosensors. By amperometry, the current generated by the electrocatalytic reaction at electrodes is monitored at a constant potential with addition of substrates. In the present case, the potential was set at 0 V versus SCE for the protein-gluten films, and catalytic reduction currents were followed when aliquots of H2 O2 were added. At a gluten film electrode with no heme protein incorporated, no current was observed by addition of hydrogen peroxide (Fig. 6a). In contrast, with heme proteins incorporated in gluten films, stepped increase of H2 O2 concentration in buffer solutions caused the corresponding stepped growth of reduction currents (Fig. 6b–d). Among the three protein-gluten films, HRP-gluten films seemed most sensitive for detecting H2 O2 . 4. Conclusion Heme proteins incorporated in hydrated gluten biopolymer films modified on PG electrodes gave direct, stable, and nearly reversible CV responses. Effective electron transfer rates involving the heme Fe(III)/Fe(II) redox couple were much faster than those for the heme proteins in solution at bare PG electrodes. The good electrocatalytic property and stability of protein-gluten films may provide an application perspective for the films as a new type of biosensors based on direct and mediator-free electrochemistry of the heme proteins.

Acknowledgements The support of National Natural Science Foundation of China (29975003 and 20275006) is gratefully acknowledged. References

Fig. 6. Amperometric current-time curves at constant potential of 0 V in pH 5.5 buffers with injection of H2 O2 every 40 s for: (a) gluten; (b) Mb-gluten; (c) Hb-gluten; and (d) HRP-gluten films. For (a), (b), and (c), each step has increment of 0.2 mM H2 O2 ; for (d), each step has increment of 0.025 mM H2 O2 .

[1] G. Wilson, in: A. Turner, I. Karube, G. Wilson (Eds.), Biosensors, Oxford University Press, New York, 1987. [2] M.F. Chaplin, C. Bucke, Enzyme Technology, Cambridge University Press, Cambridge, UK, 1990. [3] L. Gorton, A. Lindgren, T. Larsson, F.D. Munteanu, T. Ruzgas, I. Gazaryan, Anal. Chim. Acta 400 (1999) 91. [4] F.A. Armstrong, H.A. Heering, J. Hirsy, Chem. Soc. Rev. 26 (1997) 169.

H. Liu, N. Hu / Analytica Chimica Acta 481 (2003) 91–99 [5] A. Sucheta, B.A.C. Ackrell, B. Cochran, F.A. Armstrong, Nature 356 (1992) 361. [6] J.F. Rusling, Acc. Chem. Res. 31 (1998) 363. [7] N. Hu, Pure Appl. Chem. 73 (2001) 1979. [8] J. Yang, N. Hu, Bioelectrochem. Bioenerg. 48 (1999) 117. [9] N. Hu, J.F. Rusling, Langmuir 13 (1997) 4119. [10] J. Yang, N. Hu, J.F. Rusling, J. Electroanal. Chem. 463 (1999) 53. [11] H. Sun, N. Hu, H. Ma, Electroanalysis 12 (2000) 1064. [12] H. Sun, H. Ma, N. Hu, Bioelectrochem. Bioenerg. 49 (1999) 1. [13] Y. Hu, N. Hu, Y. Zeng, Talanta 50 (2000) 1183. [14] L. Wang, N. Hu, J. Colloid Interface Sci. 236 (2001) 166. [15] H. Liu, L. Wang, N. Hu, Electrochim. Acta 47 (2002) 2515. [16] X. Chen, N. Hu, Y. Zeng, J.F. Rusling, J. Yang, Langmuir 15 (1999) 7022. [17] Y. Zhou, N. Hu, Y. Zeng, J.F. Rusling, Langmuir 18 (2002) 211. [18] C. Larre, C. Desserme, J. Barbot, J. Gueguen, J. Agric. Food Chem. 48 (2000) 5444. [19] D.D. Kasarda, J.E. Bernardin, C.C. Nimmo, in: Y. Pomeranz (Ed.), Advances in Cereal Science and Technology, vol. 1, American Association of Cereal Chemists, St. Paul, Minnesota, 1976, p. 158. [20] T.J. McMaster, M.J. Miles, L. Wannerberger, A.-C. Eliasson, P.R. Shewry, A.S. Tatham, J. Agric. Food Chem. 47 (1999) 5093. [21] V. Micard, R. Belamri, M.-H. Morel, S. Guilbert, J. Agric. Food Chem. 48 (2000) 2948. [22] N. Gontard, S. Guilbert, J.-L. Cup, J. Food Sci. 57 (1992) 190. [23] H.-P. Chao, W.-C. Lee, Biotechnol. Appl. Biochem. 32 (2000) 9. [24] C.-F. Cho, W.-C. Lee, Biosci. Bioeng. 87 (1999) 822. [25] J.-C. Long, W.-C. Lee, S.-H. Guo, Biocatal. Biotransform. 17 (2000) 431.

99

[26] J.-Y. Yu, W.-C. Lee, Ferment. Bioeng. 84 (1997) 444. [27] R.W. Murray, in: A.J. Bard (Ed.), Electroanalytical Chemistry, vol. 13, Marcel Dekker, New York, 1984, p. 191. [28] J.F. Rusling, A.-E.F. Nassar, J. Am. Chem. Soc. 115 (1993) 11891. [29] Q. Huang, Z. Lu, J.F. Rusling, Langmuir 12 (1996) 5472. [30] T. Ferri, A. Poscia, R. Santucci, Bioelectrochem. Bioenerg. 44 (1998) 177. [31] L. Meites, Polarographic Techniques, second ed., Wiley, New York, 1965, p. 278. [32] A.M. Bond, Modern Polarographic Methods in Analytical Chemistry, Marcel Dekker, New York, 1980, p. 27. [33] J.J. O’Dea, J.G. Osteryoung, Anal. Chem. 65 (1993) 3090. [34] Z. Zhang, J.F. Rusling, Biophys. Chem. 63 (1997) 133. [35] A.-E.F. Nassar, Z. Zhang, N. Hu, J.F. Rusling, T.F. Kumosinski, J. Phys. Chem. 101 (1997) 2224. [36] P. George, G.I.H. Hanania, Biochem. J. 55 (1953) 236. [37] H. Huang, N. Hu, Y. Zeng, G. Zhou, Anal. Biochem. 308 (2002) 141. [38] A. Bellelli, G. Antonini, M. Brunori, B.A. Springer, S.G. Sligar, J. Biol. Chem. 265 (1990) 18898. [39] J.B. Matthew, G.H.H. Hanania, F.R.N. Gurd, Biochemistry 18 (1979) 1919. [40] K.G. Welinder, Eur. J. Biochem. 96 (1979) 483. [41] Y.V. Wu, J. Sci. Food Agric. 81 (2001) 1223. [42] T. Mita, Agric. Biol. Chem. 54 (1990) 927. [43] N. Gontard, S. Ring, J. Agric. Food Chem. 44 (1996) 3474. [44] L. Gioia, B. Cuq, S. Guilbert, J. Mater. Res. 15 (2000) 2612. [45] C.P. Andrieux, C. Blocman, J.M. Dumas-Bouchiant, J.M. Saveant, J. Am. Chem. Soc. 101 (1979) 3431. [46] C.P. Andrieux, C. Blocman, J.M. Dumas-Bouchiant, F. M’Halla, J.M. Saveant, J. Electroanal. Chem. 19 (1980) 113. [47] H. Theorell, A. Ehrenberg, Acta Chem. Scand. 5 (1951) 823. [48] X. Chen, C. Ruan, J. Kong, J. Deng, J. Electroanal. Chem. 480 (2000) 26.