Switching Behavior of Layer-by-Layer Films Assembled by

Nov 12, 2009 - of Chemistry, Ningxia Medical UniVersity, Yinchuan 750004, People's Republic of China. ReceiVed: July 5, 2009; ReVised Manuscript ...
0 downloads 0 Views 647KB Size
Download PDF
J. Phys. Chem. B 2009, 113, 16021–16027

16021

pH-Sensitive “On-Off” Switching Behavior of Layer-by-Layer Films Assembled by Concanavalin A and Dextran toward Electroactive Probes and its Application in Bioelectrocatalysis Huiqin Yao†,‡ and Naifei Hu*,† Department of Chemistry, Beijing Normal UniVersity, Beijing 100875, People’s Republic of China, Department of Chemistry, Ningxia Medical UniVersity, Yinchuan 750004, People’s Republic of China ReceiVed: July 5, 2009; ReVised Manuscript ReceiVed: October 15, 2009

Concanavalin A (Con A) and dextran (Dex) were assembled into {Con A/Dex}n layer-by-layer films on electrodes by the biospecific interaction between them. At {Con A/Dex}n film electrodes, the cyclic voltammetric (CV) response of different electroactive probes, such as Fe(CN)63-, Ru(NH3)63+, ferrocenecarboxylic acid (Fc(COOH)), and ferrocenedicarboxylic acid (Fc(COOH)2), was very sensitive to the solution pH. For example, in pH 4.0 buffers, the films showed good permeability toward Fe(CN)63-, leading to a well-defined CV peak pair of Fe(CN)63- with large peak currents at about 0.17 V vs SCE. In pH 9.0 buffers, however, the CV response of Fe(CN)63- was significantly depressed or even could hardly be observed. This pH-sensitive “on-off” switching property of the films toward the probe should be attributed to the different electrostatic interaction between Fe(CN)63- and the Con A constituent in the films at different pH and could be further used to control the electrocatalytic reduction of H2O2 by horseradish peroxidase (HRP) with Fe(CN)63- as the diffusional electron transfer mediator. Fc(COOH)2 showed the similar pH-dependent “on-off” behavior at {Con A/Dex}n film electrodes, and the corresponding pH-sensitive electrocatalytic oxidation of glucose by glucose oxidase (GOD) with Fc(COOH)2 as the mediator was also realized. This work provides a new interface system that has the pH-sensitive “on-off” CV property, and the better understanding of the interactions involved in this model system may guide us to develop the novel kind of controllable biosensors based on enzymatic electrocatalysis. Introduction The study of enzyme electrochemistry has become one of the hottest research topics in bioelectrochemistry and in the development of electrochemical biosensors and bioactuators.1,2 Biocatalysis based on direct or mediated electrochemistry of various enzymes can be used to detect and determine the respective substrates, thus establishing the foundation of electrochemical biosensors and other biodevices. In this aspect, the bioelectrocatalysis that can be controlled by some external stimuli is of great significance in both fundamental and practical studies. The reversible activation and deactivation of “on-off” switching function in bioelectrocatalysis would enable their application in bioelectronic devices, and provide the basis for information storage, data processing, signal amplification, biosensor devices, and others.3,4 For example, Willner and coworkers reported that by the magnetic attraction and retraction of ferrocene-modified magnetic particles to and from the electrode surface, the electrocatalytic oxidation of glucose by glucose oxidase (GOD) could be at the “on” and “off” state, respectively.5 Recently, Katz and colleagues reported that the polymer brush functionalized with Os-complex redox units modified on electrodes exhibited a pH-sensitive structure and could be used to reversibly activate/deactivate the electrocatalytic oxidation of glucose in the presence of GOD by changing solution pH chemically or biochemically.6,7 Until now, some papers have been published by several groups in this area.3-11 * To whom correspondence should be addressed. E-mail: hunaifei@ bnu.edu.cn. Phone: (+86) 10-5880-5498. Fax: (+86) 10-5880-2075. † Beijing Normal University. ‡ Ningxia Medical University.

However, considering the complication of these works such as the need of additional magnets and specially synthesized metal complexes, it is still a challenging task to realize switchable bioelectrocatalysis with more general and convenient approaches. In recent years, a relatively new film-forming technique called layer-by-layer (LbL) assembly has aroused great interests among researchers.12-14 The LbL assembly shows distinguished advantages over other film preparation methods in the precise control of film thickness at a nanometer level according to a predesigned architecture and in its extremely simple procedure and high versatility in the assembly. The driving force of LbL assembly is originally electrostatic interaction and now has been extended to nonelectrostatic interactions including hydrogenbonding, hydrophobic interaction, and specific biological interactions.14 Among the various biospecific interactions, lectin-sugar interaction has attracted great attention recently. Concanavalin A (Con A) is the best-known member of the lectin proteins since its structure and property have been well documented.15 Its structure is shown in Scheme 1. At neutral pH, Con A exists as a tetramer, and each subunit of Con A contains a binding site to some specific sugar groups such as glucose and mannose, forming a highly specific 4:1 sugar-Con A complex.16,17 Dextran (Dex) is a glucose-based polysaccharide and contains many glucose residues on its backbone that can bind to Con A.18,19 Its structure is also shown in Scheme 1. While the biospecific affinity between Con A and Dex has been studied extensively and applied in different fields,20-22 only a few papers reported the LbL assembly of Con A and Dex.23,24 In our previous work,25 {Con A/Dex}n LbL films were assembled on electrode surface,

10.1021/jp906345s CCC: $40.75  2009 American Chemical Society Published on Web 11/12/2009

16022

J. Phys. Chem. B, Vol. 113, No. 49, 2009

Yao and Hu

SCHEME 1: Structures of Con A and Dex

and myoglobin (Mb) in solution was loaded into the films and its electrochemistry was investigated. Stimuli-responsive films that change their properties in response to environmental stimuli have aroused considerable attention over the past decade for both fundamental and application research.3,26-28 In particular, by choosing appropriate materials and processing conditions, it is possible to create thin LbL films on electrodes that may show different electrochemical responses for electroactive probes under different pH conditions, depending on the pH-sensitive permeability of the films. Reversible “on-off” switching of the electrochemical activity can thus be achieved by varying the solution pH.29-32 For example, Crooks and co-workers assembled LbL films of amineterminated dendrimers and polymeric Gantrez containing carboxylic groups on Au-coated Si wafers. These composite films exhibited fully reversible, pH-switchable permselectivity for both cationic and anionic redox-active probe molecules.29 Advincula and colleagues reported that the PAA-BP/PAH-BP multilayer films deposited at some specific pHs showed pH-sensitive permselectivity for both cationic and anionic probes, where PAA ) poly(acrylic acid), PAH ) poly(allylamine hydrochloride), and BP ) benzophenone.31 Sun and co-workers reported that after UV irradiation the photocross-linked LbL films containing azobenzene were converted into the films which contained both free carboxylic acid and imine groups. The resultant films showed a reversible pH-switchable permeability for positively charged species.32 In the present work, Con A and Dex were chosen as the building blocks to be assembled into {Con A/Dex}n LbL films mainly because the driving force of the assembly was the specific lectin-sugar interaction between them. With the isoelectric point (pI) at about 5.0,33 Con A may carry different net surface charges at different pH, while Dex is always neutral, carrying no charge and independent of solution pH. The charge situation of {Con A/Dex}n films was thus only determined by the charge property of Con A constituent, which was positive at pH < 5.0 and negative at pH > 5.0. This character of the films is different from that of regular LbL films assembled through electrostatic interaction. For the electrostatic LbL films, both building blocks of the films have to be ionized, while for the {Con A/Dex}n films, only one component is charged. Different electroactive probes with different charges were used to investigate the pH-sensitive permeability of the films assembled on electrodes by cyclic voltammetry (CV). It was found that for anionic probes such as Fe(CN)63-, the films were permeable at pH 4.0 and “closed” at pH 9.0, but for cationic Ru(NH3)63+, the films behaved oppositely, they were at the “on” state at pH 9.0 and “off” state at pH 4.0. This pH-sensitive “on-off” switching property of {Con A/Dex}n films was further used to control the electrocatalytic reduction of H2O2 by

horseradish peroxidase (HRP) with Fe(CN)63- as the mediator and the electrocatalytic oxidation of glucose by GOD with ferrocenedicarboxylic acid (Fc(COOH)2) as the mediator. The physical chemistry basis of the pH-sensitive permeability of the films and the corresponding bioelectrocatalysis was explored and discussed. While the works on either pH-switchable bioelectrocatalysis5-7 or pH-sensitive permeability of LbL films on electrodes29-32 have been reported previously, to the best our knowledge, the study of combining them together has not been reported up to now. This work provides a simple and convenient way to construct the pH-controllable biosensors based on the LbL assembly and bioelectrocatalysis, and establishes a new “smart” model interface that has the pHsensitive “on-off” redox property. The better understanding of the essence of interactions involved in this model system at different pH may also open a general way to develop the novel kind of controllable biosensors based on enzymatic electrocatalysis. Experimental Section 1. Reagents. Chitosan (CS, the degree of deacetylation is more than 85%, MW ≈ 200 000), concanavalin A extracted from Jack beans (Con A, type V, MW ≈ 104 000), dextran (Dex, MW ≈ 200 000), 1,1′-ferrocenedicarboxylic acid (Fc(COOH)2), ferrocenecarboxylic acid (Fc(COOH)), hexaammineruthenium(III) chloride (Ru(NH3)6Cl3), horseradish peroxidase (HRP, E.C. 1.11.1.7, type II, MW ≈ 44 000, 250 000 units g-1), glucose oxidase (GOD, E.C. 1.1.3.4, type VII, MW ≈ 160 000, 192 000 units g-1), and tris(hydroxymethyl) aminomethane (Tris) were purchased from Sigma-Aldrich. Potassium ferricyanide (K3Fe(CN)6), potassium ferrocyanide (K4Fe(CN)6), and hydrogen peroxide (H2O2, 30%) were obtained from Beijing Chemical Engineering Plant. The dilute H2O2 aqueous solutions were freshly prepared before being used. All other reagents were of analytical grade. Britton-Robinson buffers at pH 3.0-9.0 contained 0.04 M acetic acid, 0.04 M phosphoric acid, 0.04 M boracic acid, and 0.1 M NaCl, and the pH was adjusted to the desired value with dilute HCl or NaOH solutions. 0.1 M Tris-HCl buffers at pH 7.4 containing 0.1 M NaCl, 1 mM MnCl2, and 1 mM CaCl2 were used to prepare Con A and Dex solutions. All solutions were prepared with water purified twice by ion exchange and subsequent distillation. 2. Film Assembly. Prior to assembly, basal plane pyrolytic graphite (PG, Advanced Ceramics, geometric area 0.16 cm2) disk electrodes were abraded on 320-grit metallographic sandpaper while flushing with water. After being ultrasonicated in water for 30 s and dried in air, the electrodes were first immersed in 1 mg mL-1 CS solutions at pH 5.0 for 30 min, forming a CS precursor layer on PG surface. The PG/CS electrodes were then alternately immersed into Con A (1 mg mL-1, pH 7.4) and Dex

Layer-by-Layer Films Assembled by Concanavalin A and Dextran

Figure 1. CVs of 0.5 mM Fe(CN)63- at 0.1 V s-1 in pH 7.4 buffers at (a) bare PG electrode, (b) PG/CS films, (c-i) PG/CS/{Con A/Dex}n films with n ) 1-7.

(1 mg mL-1, pH 7.4) aqueous solutions for 30 min each, with intermediate water rinsing for about 10 s and air stream drying until the desired number of bilayers (n) was obtained, forming {Con A/Dex}n LbL films on the PG/CS surface. 3. Apparatus and Procedures. A CHI 660A or 621B electrochemical workstation (CH Instruments) was used for electrochemical measurements. A typical three-electrode cell was used with a saturated calomel electrode (SCE) as the reference, a platinum foil as the counter, and a PG electrode with films as the working electrode. The solution containing electroactive probes was purged with high-purity nitrogen at least for 10 min before electrochemical measurements. The nitrogen atmosphere was then kept for the entire experiment. Electrochemical impedance spectroscopy (EIS) was performed in 1:1 K4Fe(CN)6:K3Fe(CN)6 mixture solutions with total concentration of 5 mM, and a sinusoidal potential modulation with amplitude of (5 mV and frequency from 105 to 0.1 Hz was superimposed on the formal potential of Fe(CN)64-/3- redox couple at 0.17 V vs SCE. All experiments were performed at ambient temperature of 20 ( 2 °C. Results and Discussion 1. Assembly of {Con A /Dex}n LbL Films. With its pKa at about 6.5,34 CS carries positive charges at pH 5.0 and could be adsorbed on negatively charged PG surface.35 With its pI value at about 5.0,33 Con A carries net negative surface charges at pH 7.4 and thus could be adsorbed on oppositely charged CS surface by electrostatic attraction. Dex is a neutral polysaccharide carrying no charge, but it has a strong biospecific interaction with Con A20-24 and could be adsorbed on the Con A surface, forming a Con A/Dex bilayer. By repeating this cycle, the assembly of {Con A/Dex}n LbL films was thus realized on PG/ CS surface mainly by the biospecific binding between Con A and Dex. In our previous work,25 the assembly of {Con A/Dex}n multilayer films was confirmed by EIS and quartz crystal microbalance (QCM). In the present work, CV with Fe(CN)63as the electroactive probe was further used to monitor the assembly of {Con A/Dex}n films on PG/CS surface (Figure 1). For bare PG electrode and PG/CS films, Fe(CN)63- in solution at pH 7.4 displayed a well-defined and nearly reversible CV oxidation-reduction peak pair at about 0.17 V. However, when a Con A/Dex bilayer was adsorbed on PG/CS, the CV response of probe was severely suppressed, indicating that a barrier is formed on the electrode surface and hinders the probe from exchanging electrons with PG electrodes. The reduction peak current (Ipc) of Fe(CN)63- decreased with the number of bilayers (n) of {Con A/Dex}n films, accompanied by the increase of peak separation (∆Ep) (See Figure S1 of Supporting Information), suggesting that the multilayer films are successfully fabricated on PG/CS surface.

J. Phys. Chem. B, Vol. 113, No. 49, 2009 16023

Figure 2. Influence of solution pH on (a) CV reduction peak current (Ipc) and (b) peak separation (∆Ep) of 1 mM K3Fe(CN)6 at 0.1 V s-1 for {Con A/Dex}4 films.

Figure 3. (A) CVs of 1 mM K3Fe(CN)6 at 0.1 V s-1 for {Con A/Dex}4 films in buffers at pH (a) 4.0 and (b) 9.0. (B) Dependence of CV reduction peak current (Ipc) of K3Fe(CN)6 at 0.1 V s-1 on solution pH switched between pH 4.0 and 9.0 for the same {Con A/Dex}4 films.

2. pH-Sensitive Switching Behavior of {Con A/Dex}4 Films toward Fe(CN)63-. For {Con A/Dex}4 films with n ) 4, the pH of Fe(CN)63- solutions had great influence on CV behaviors of Fe(CN)63- (see Figure S2 of Supporting Information). While Fe(CN)63- showed relatively large CV reduction peak current (Ipc) and small peak separation (∆Ep) in the pH range of 2.0-5.0, the Ipc decreased drastically with increase of pH from 5.0 to 9.0, accompanied by the increase of ∆Ep (Figure 2). Particularly, when pH g 9.0, the CV signal of Fe(CN)63- almost disappeared. This pH-sensitive CV behavior of Fe(CN)63- should not be attributed to the property of the probe itself since the CV behavior of Fe(CN)63- at bare PG electrodes was essentially pH-independent (see Figure S3 of Supporting Information). It thus must be related to the interaction between the probe and the films. These results inspired us to use the {Con A/Dex}4 films to control the electrochemical “on-off” property of the probe by switching solution pH. The “on-off” switching behavior of {Con A/Dex}4 films toward Fe(CN)63- was examined at two typical pHs. At pH 4.0, the films were at the “on” state and the probe showed a nearly reversible CV peak pair with quite large peak currents; at pH 9.0, the films were at the “off” state and the CV response of Fe(CN)63- even could hardly be observed (Figure 3A). This “on-off” switching property was quite reversible. When the {Con A/Dex}4 film electrodes switched in solutions containing Fe(CN)63- between pH 4.0 and 9.0, the corresponding CV responses were repeatedly cycled between the “on” and “off” states (Figure 3B). This “on-off” behavior could be repeated at least for 10 cycles, and only a very little decrease of Ipc at pH 4.0 was observed with the increase of cycles, while the reason for the decrease is not clear yet. In {Con A/Dex}n films, the Dex component is a neutral polymer, and its structure and property is pH-independent. However, the Con A constituent is pH-sensitive, and its property may be different at different pH. Thus, the pH-dependent “on-off” property of {Con A/Dex}n films toward Fe(CN)63should be ascribed to the Con A component. The pI value of Con A is at about 5.0.33 At pH 4.0, the Con A in the films carries net positive surface charges and would have a strong

16024

J. Phys. Chem. B, Vol. 113, No. 49, 2009

Yao and Hu

Figure 4. (A) CVs of 1 mM K3Fe(CN)6 at 0.1 V s-1 for {Con A/Dex}1 films with n ) 1 in pH (a) 4.0 and (b) 9.0 buffers. (B) Influence of the number of bilayers (n) for {Con A/Dex}n films on CV reduction peak current (Ipc) of 1 mM K3Fe(CN)6 in buffers at pH 4.0 (O) and 9.0 (b) at 0.1 V s-1. 3-

electrostatic attraction with negatively charged Fe(CN)6 in solution, thus making the probe to go through the films more easily and leading to the quite large CV response of Fe(CN)63-. In contrast, at pH 9.0, the Con A in the films carries net negative surface charges and would have a strong electrostatic repulsion with Fe(CN)63-. This may hinder the probe from entering the films and exchanging electrons with underlying electrodes, thus resulting in the very small CV signal of Fe(CN)63-. The pHsensitive behavior of {Con A/Dex}n films toward Fe(CN)63- is therefore most probably attributed to the electrostatic interaction between the Con A component of the films and the probe. Further studies showed that the outermost layer of the films had little influence on the CV behavior of Fe(CN)63-. For example, at pH 4.0, Fe(CN)63- demonstrated the identical CVs for {Con A/Dex}3/Con A and {Con A/Dex}4 films with quite large peak heights, while at pH 9.0, both films were “closed” for the probe (see Figure S4 of Supporting Information). These results suggest that the interpenetration between neighboring Con A and Dex layers may happen in the film assembly, which is a common phenomenon in LbL assembly.12 These results also confirm that the Con A instead of Dex component in the films plays a key role in interaction with the probe, since both {Con A/Dex}3/Con A and {Con A/Dex}4 films have the same amount of Con A but the different amount of Dex. EIS was also performed to investigate the pH-dependent “on-off” behavior of {Con A/Dex}4 films with Fe(CN)3-/4as the redox probe at its formal potential of 0.17 V (see Figure S5 of Supporting Information). At pH 4.0, the EIS response in the form of Nyquist diagram showed a Warburg line with no obvious semicircle in high-frequency domain. At pH 9.0, however, a large semicircle was observed. The diameter of semicircle usually equals the charge transfer resistance (Rct) of the probe in electron transfer.36 For the film systems, Rct mainly reflects the restricted diffusion of the probe through the film phase37,38 and can be estimated by using the Randles equivalent circuit as the model.39 The considerably larger Rct value of the probe at pH 9.0 than that at 4.0 was consistent with the CV results (Figure 3A). Moreover, the pH-sensitive “on-off” EIS response was reversible and could be repeated for several cycles between pH 4.0 and 9.0 (see Figure S5 of Supporting Information). 3. Influence of Film Thickness. The number of bilayers (n) or the thickness of {Con A/Dex}n films showed considerable influence on the “on-off” property of the films toward Fe(CN)63-. For {Con A/Dex}1 films with n ) 1, while the CV response of Fe(CN)63- at pH 9.0 was smaller than that at pH 4.0, it could still be clearly observed (Figure 4A), suggesting that the very thin films could not be completely “closed” at pH 9.0 toward the probe. This is understandable since one Con A/Dex bilayer contains very small amounts of negatively charged Con A at pH 9.0, which limits its repulsion with

Figure 5. CVs of 1 mM Ru(NH3)6Cl3 at 0.1 V s-1 for {Con A/Dex}4 films in buffers at pH (a) 9.0 and (b) 4.0.

Fe(CN)63-. Moreover, the Con A/Dex bilayer may not completely cover the electrode surface; some underlying CS layer with positive charges may be exposed and tend to attract Fe(CN)63-. With the increase of the number of bilayers (n) for the films, the CV reduction peak current of Fe(CN)63- (Ipc) at pH 4.0 showed a slow decreasing trend (Figure 4B), implying that the permeability of the films toward the probe becomes a little poorer when the films become thicker. However, in pH 9.0 solutions, the Ipc of the probe decreased dramatically with n from 1 to 4, and the CV response was nearly disappeared when n g 4. This should be mainly ascribed to the larger amounts of negatively charged Con A component in the thicker films and the corresponding stronger electrostatic repulsion between the films and the probe. Considering that the {Con A/Dex}4 films with n ) 4 demonstrated the most pronounced difference in Ipc values between pH 4.0 and 9.0, the films with n ) 4 were usually used in the present work. 4. pH-Sensitive On-Off Behavior of {Con A/Dex}4 Films toward Ru(NH3)63+. To further support our conjecture that the electrostatic interaction between the films and the probe plays a key role in the pH-sensitive switching behavior, positively charged Ru(NH3)63+ was used to replace Fe(CN)63- as the electroactive probe to test the film permeability at different pH. As seen from Figure 5, in pH 9.0 solutions, Ru(NH3)63+ demonstrated a pair of quite large CV reversible peaks at about -0.2 V. However, at pH 4.0, this peak pair was greatly depressed. That is, the {Con A/Dex}4 films were at the “on” state at pH 9.0 and at the “off” state at pH 4.0 for Ru(NH3)63+, totally contrary to that for Fe(CN)63- (Figure 3). This is because Ru(NH3)63+ carries the opposite charges in comparison with Fe(CN)63-. At pH 9.0, negatively charged Con A in the films tended to attract positively charged Ru(NH3)63+, leading to the well-defined CV signal. At pH 4.0, however, positively charged Con A in the films would electrostatically repulse Ru(NH3)63+, resulting in the great decrease of the CV signal. Considering that Fe(CN)63- and Ru(NH3)63+ have similar size (6.0 Å for Fe(CN)63-, 6.2 Å for Ru(NH3)63+),31 these results confirm again that the microstructure and porosity of the {Con A/Dex}n films are pH-independent, and the charge property of probe and its interaction with the films play a central role in its permeation through the films, especially when the charge density of the probe is large. In control experiments with bare PG electrodes, CVs of Ru(NH3)63+ in solutions at pH 9.0 showed no substantial difference with that at pH 4.0 (see Figure S6 of Supporting Information). 5. pH-Controlled Bioelectrocatalysis with K3Fe(CN)6 as Mediator. The pH-responsive switching property of {Con A/Dex}4 films toward Fe(CN)63- could be used to control or modulate electrocatalytic reduction of H2O2 by HRP with Fe(CN)63- as the mediator. When the {Con A/Dex}4 film electrodes were placed in pH 4.0 buffers containing Fe(CN)63-, H2O2, and HRP, a large catalytic reduction wave centered at

Layer-by-Layer Films Assembled by Concanavalin A and Dextran

Figure 6. (A) CVs of {Con A/Dex}4 films at 0.01 V s-1 in solutions containing 1 mM K3Fe(CN)6, 0.5 mM H2O2, and 0.5 mg mL-1 HRP at pH (a) 4.0 and (b) 9.0. (B) CVs of {Con A/Dex}4 films at 0.01 V s-1 in pH 4.0 buffers containing 1 mM K3Fe(CN)6 and 0.5 mg mL-1 HRP with (a) 0, (b) 0.3, and (c) 0.5 mM H2O2.

about 0.15 V was observed (Figure 6A). The mechanism of the bioelectrocatalysis can be expressed by the following equations40-42

H2O2 + 2H+ + HRP(red) f 2H2O + HRP(ox)

(1) 3HRP(ox) + Fe(CN)46 f HRP(red) + Fe(CN)6

(2) Fe(CN)3h Fe(CN)46 + e 6 at electrode

(3)

This mechanism was supported by a series of CV experiments (Figure 6B). In pH 4.0 buffers, Fe(CN)63- showed a reversible CV peak pair at about 0.17 V for {Con A/Dex}4 films. When both H2O2 and HRP were added in the Fe(CN)63- solution, a large increase in CV reduction peak for Fe(CN)63- was observed, accompanied by the disappearance of the oxidation peak of Fe(CN)64-. The reduction wave increased linearly with the concentration of H2O2 in solution in the range of 0.05-0.60 mM and then tended to level off when the concentration of H2O2 became larger (see Figure S7 of Supporting Information), suggesting a progressive enzyme inactivation in the presence of high concentration of substrate.43 All these are characteristic of electrochemical reduction of H2O2 catalyzed by HRP enzyme and mediated by Fe(CN)63-/4- redox couple. However, when the {Con A/Dex}4 films were placed in pH 9.0 buffers containing the same amount of Fe(CN)63-, H2O2, and HRP, the electrocatalytic response became quite small and even could hardly be observed (Figure 6A). This is because the films are at the “off” state toward Fe(CN)63- at this pH, and the mediator can not diffuse through the films and exchange electron with the electrode, finally resulting in the interruption of the electrocatalytic cycle. Therefore, the electrocatalytic reduction of H2O2 by HRP with Fe(CN)63- as the mediator could be controlled or switched by the permeability of {Con A/Dex}4 films toward Fe(CN)63- at different pH. The CV reduction peak current ratio, Ipc4/Ipc9, could be amplified by the bioelectrocatalysis, where Ipc4 and Ipc9 were CV reduction peak/wave current at pH 4.0 and 9.0, respectively. The Ipc4/Ipc9 ratio was only about 12 in Fe(CN)63- solutions containing no HRP and H2O2 but increased to about 32 in the presence of HRP and H2O2 (Figure 6A), indicating a significant magnification through electrocatalysis by both enzyme and mediator reactions. In contrast, for bare PG electrodes, while Fe(CN)63- showed the almost identical CV responses at pH 4.0 and 9.0 (see Figure S3 of Supporting Information), the electrocatalytic responses of Fe(CN)63- in the presence of H2O2 and HRP were different at these two pHs with

J. Phys. Chem. B, Vol. 113, No. 49, 2009 16025

Figure 7. (A) CVs of 0.5 mM Fc(COOH) at 0.1 V s-1 for {Con A/Dex}4 films in buffers at pH (a) 4.0 and (b) 9.0. (B) CVs of 0.5 mM Fc(COOH)2 at 0.1 V s-1 for {Con A/Dex}4 film electrodes in buffers at pH (a) 4.0 and (b) 9.0.

Ipc4 > Ipc9 (see Figure S8 of Supporting Information). This may be related to the different turnover rate of the enzyme at different pHs. According to the kinetics of the oxidation of ferrocyanide by H2O2 catalyzed by HRP in aqueous solutions, the apparent rate constant is pH-dependent and shows larger value at lower pH.44 However, the Ipc4/Ipc9 ratio for bare PG was only 1.7, much smaller than that of 32 for the {Con A/Dex}4 films, suggesting that the “on-off” switching function of the films in bioelectrocatalysis should be mainly attributed to the pH-sensitive property of the films but not to the probe itself or the related reactions. 6. pH-Controlled Bioelectrocatalysis with Fc(COOH)2 as Mediator. The sensing of glucose is of great significance in several applications including the diagnosis and control of diabetes in clinic and wastewater treatment in food industry. Among various approaches in determination of glucose, electrochemical methods using GOD enzyme have aroused great interests.45-47 The electrocatalytic oxidation of glucose by GOD usually needs an electroactive mediator to shuttle electrons between GOD and electrodes. If the pH-sensitive permeability of {Con A/Dex}4 films can be applied to the mediator, the electrocatalytic oxidation of glucose by GOD would be controlled or tuned by solution pH. However, Fe(CN)63- cannot be used as the efficient mediator in the electrocatalysis of glucose by GOD, probably because the formal potential of Fe(CN)63-/4- cannot match that of the GOD redox couple. Thus, in the present work, two ferrocene derivatives, Fc(COOH) and Fc(COOH)2, were selected as the candidate mediator. In general, ferrocene and its derivatives are well-known electroactive mediators in electrocatalytic oxidation of glucose by GOD in the absence of oxygen.48-50 In particular, both Fc(COOH) and Fc(COOH)2 are water-soluble, and their carboxylic groups would take different ionized forms at different pH and may be suitable for testing the pH-dependent permeability of the films. CV results demonstrated that {Con A/Dex}4 films also showed pH-sensitive “on-off” switching function toward Fc(COOH) and Fc(COOH)2, and the films were generally at the “on” state at pH 4.0 and at the “off” state at pH 9.0 (Figure 7), similar to the situation of Fe(CN)63-. With their pKa values at about 6.8,51,52 both Fc(COOH) and Fc(COOH)2 are negatively charged at pH 9.0 and carry no charge at pH 4.0. Thus, the charge situation of the ferrocene compounds is similar to that of Fe(CN)63-, which leads to their similar pH-dependent “on-off” behavior for the films. In control experiments, both ferrocene probes showed nearly reversible CV responses at bare PG electrodes, and no substantial difference in CV signals was observed in buffers at pH 4.0 and 9.0 (see Figure S9 of Supporting Information). In pH 4.0 solutions, Fc(COOH) and Fc(COOH)2 showed almost the same CV peak heights at {Con A/Dex}4 film electrodes. However, at pH 9.0, the CV peak heights of

16026

J. Phys. Chem. B, Vol. 113, No. 49, 2009

Yao and Hu

Figure 8. (A) CVs of {Con A/Dex}4 films at 0.005 V s-1 in pH 4.0 buffers containing 0.5 mM Fc(COOH)2 and 1.0 mg mL-1 GOD with (a) 0, (b) 1.0, (c) 3.0, and (d) 4.0 mM glucose. (B) CVs of {Con A/Dex}4 films at 0.005 V s-1 in solutions containing 0.5 mM Fc(COOH)2, 4.0 mM glucose, and 1.0 mg mL-1 GOD at pH (a) 4.0 and (b) 9.0.

Fc(COOH) and Fc(COOH)2 were obviously different, and those of the former were larger than those of the latter (Figure 7). The possible explanation is that Fc(COOH) carries less negative charges than Fc(COOH)2 at pH 9.0, which may result in the weaker electrostatic repulsion between the films and Fc(COOH). That is, {Con A/Dex}4 films cannot be completely “closed” toward Fc(COOH) at pH 9.0. Considering that Fc(COOH)2 showed better “on-off” property than Fc(COOH) at different pH for the films, Fc(COOH)2 was selected as the electron mediator for the following experiments so that the better pHsensitive bioelectrocatalysis of glucose by GOD could be realized. The pH-dependent “on-off” switching behavior of {Con A/Dex}4 films toward Fc(COOH)2 was reversible and could be repeated at least for 10 cycles (see Figure S10 of Supporting Information). In pH 4.0 solutions containing no oxygen, the obvious increase of CV oxidation peak current of Fc(COOH)2 at about 0.45 V was observed in the presence of glucose and GOD for {Con A/Dex}4 films, accompanied by the decrease or even disappearance of the reduction peak (Figure 8A). The oxidation peak/wave current (Ipa) increased initially with the concentration of glucose, and then tended to level off (Figure S11 of Supporting Information). When the concentration of glucose was high enough, the CV response changed from the peak-like shape to the wavelike one (Figure 8A). All these are characteristic of electrocatalysis, and the mechanism can be expressed by the following equations53,54

GOD(FAD) + glucose f GOD(FADH2) + gluconolactone (4) GOD(FADH2) + 2Fc(COOH)2 ox f GOD(FAD) + 2Fc(COOH)2 red (5) Fc(COOH)2 red-e- h Fc(COOH)2 oxat electrode

(6)

where GOD(FAD) and GOD(FADH2) represent oxidized and reduced forms of glucose oxidase, respectively. In pH 9.0 buffers, however, the bioelectrocatalysis was at the “off” state, and almost no CV response was observed in the same Fc(COOH)2 + GOD + glucose solution for the films (Figure 8B). This is mainly because Fc(COOH)2 is difficult to go through the films and then exchange electron with underlying electrodes at this pH, leading to the interruption of the

electrocatalytic cycle. The electrocatalytic oxidation of glucose by GOD could thus be controlled by the permeability of {Con A/Dex}4 films toward Fc(COOH)2 at different pH. The CV oxidation peak/wave current ratio, Ipa4/Ipa9, could also be amplified by the bioelectrocatalysis from 3 to 12 (Figure 8B), where Ipa4 and Ipa9 were CV oxidation peak/wave current of Fc(COOH)2 at pH 4.0 and 9.0, respectively. In control experiments, the same Fc(COOH)2 + GOD + glucose system was tested by CV at bare PG electrodes, and the electrocatalysis of glucose by GOD was clearly observed at both pH 4.0 and 9.0 (see Figure S12 of Supporting Information). While the CV catalytic response at pH 4.0 was a little bit larger than that at pH 9.0 for bare PG electrodes, the Ipa4/Ipa9 ratio was only 1.1, much smaller than 12 for the {Con A/Dex}4 films. This suggests again that the pH-dependent CV “on-off” property of the mediator and the corresponding bioelectrocatalysis mainly originate from the interaction between the films and the probe. In addition, the pH-sensitive “on-off” bioelectrocatalysis of the system could be repeated at least for several cycles by switching the films in solutions between pH 4.0 and 9.0, although the Ipa at pH 4.0 showed a slow decreasing trend with the number of cycles (see Figure S13 of Supporting Information). The {Con A/Dex}n films could be disintegrated by being exposed to the concentrated glucose in solution because of the competition of glucose with Dex in interaction with Con A.25 However, the maximum concentration of glucose used in the present work was 7 mM, much smaller than that for inducing the disintegration of the films (100 mM).25 Under the condition of relatively low concentration of glucose, {Con A/Dex}4 films were quite stable and could be used to determine glucose in the bioelectrocatalytic system. Conclusion {Con A/Dex}n LbL films assembled on PG electrodes demonstrate pH-sensitive “on-off” switching property toward electroactive probes. In pH 4.0 buffers, the probes of Fe(CN)63-, Fc(COOH), and Fc(COOH)2 are at the “on” state for the films, showing nearly reversible CV responses with quite large peak heights. At pH 9.0, these probes are at the “off” state, and the CV responses are greatly depressed. On the contrary, the positively charged Ru(NH3)63+ shows quite large CV response at pH 9.0 and very small signal at pH 4.0 for the films. The pH-dependent permeability of the {Con A/Dex}n films toward different probes can be explained by the electrostatic interaction between the films and the probe. Since Dex is a neutral polysaccharide, the charge situation of Con A component at

Layer-by-Layer Films Assembled by Concanavalin A and Dextran different pH determines the charge property of the whole films. This unique charge property of {Con A/Dex}n films cannot be observed in the electrostatic LbL films and is related to the unique biospecific interaction between Con A and Dex. Maybe the most exciting result obtained in this study is the realization of pH-controlled bioelectrocatalysis. The different permeability of {Con A/Dex}n films toward Fe(CN)63- at different pH can be used to control the electrocatalytic reduction of H2O2 by HRP, and the pH-sensitive “on-off” function of the films toward Fc(COOH)2 can be used to modulate the electrocatalytic oxidation of glucose by GOD. Herein, Fe(CN)63- and Fc(COOH)2 not only act as the electroactive probe for the films but also act as the electron transfer mediator in bioelectrocatalysis. This novel system may open a general and convenient way to establish a “smart” interface that can sensitively respond the pH stimulation from environments, and may also guide us to develop the novel kind of controllable and tunable electrochemical biosensors or bioreactors based on enzymatic electrocatalysis. Acknowledgment. The financial support from the National Natural Science Foundation of China (NSFC 20975015 and 20775009) is acknowledged. Supporting Information Available: Thirteen figures showing the dependence of Ipc and ∆Ep of Fe(CN)63- on the assembly step of {Con A/Dex}n films, CVs of Fe(CN)63- for {Con A/Dex}4 films at different pH, CVs of Fe(CN)63-, Ru(NH3)63+, Fc(COOH)2, and Fc(COOH) at bare PG electrodes, CVs of Fe(CN)63- for {Con A/Dex}3/Con A and {Con A/Dex}4 films at different pH, EIS responses of Fe(CN)63-/4- for {Con A/Dex}4 films at different pH, the dependence of Rct on solution pH switched between pH 4.0 and 9.0, dependence of Ipc on concentration of H2O2 for {Con A/Dex}4 films in pH 4.0 solutions containing Fe(CN)63- + HRP + H2O2, CVs of Fe(CN)63- in the presence of HRP and H2O2 at bare PG electrodes at different pH, dependence of Ipa of Fc(COOH)2 on solution pH switched between pH 4.0 and 9.0 for {Con A/Dex}4 films, dependence of Ipa on concentration of glucose for {Con A/Dex}4 films in pH 4.0 solutions containing Fc(COOH)2 + GOD + glucose, CVs of Fc(COOH)2 in the presence of GOD and glucose at bare PG electrodes at different pH, and dependence of Ipa for Fc(COOH)2 + GOD + glucose on solution pH switched between pH 4.0 and 9.0 for {Con A/Dex}4 films. This information is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Katz, E.; Shipway, A. N.; Willner, I. Encyclopedia of Electrochemistry; Wilson, G. S., Ed.; Wiley-VCH: Weinheim, 2002; Vol. 9, Chapter 17, pp 559-626. (2) Armstrong, F. A.; Wilson, G. S. Electrochim. Acta 2000, 45, 2623. (3) Willner, I. Acc. Chem. Res. 1997, 30, 347. (4) Willner, I.; Katz, E. Angew. Chem., Int. Ed. 2000, 39, 1180. (5) Hirsch, R.; Katz, E.; Willner, I. J. Am. Chem. Soc. 2000, 122, 12053. (6) Tam, T. K.; Ornatska, M.; Pita, M.; Minko, S.; Katz, E. J. Phys. Chem. C 2008, 112, 8438. (7) Tam, T. K.; Zhou, J.; Pita, M.; Ornatska, M.; Minko, S.; Katz, E. J. Am. Chem. Soc. 2008, 130, 10888. (8) Lee, J.; Lee, D.; Oh, E.; Kim, J.; Kim, Y.-P.; Jin, S.; Kim, H.-S.; Hwang, Y.; Kwak, J. H.; Park, J.-G.; Shin, C.-H.; Kim, J.; Hyeon, T. Angew. Chem., Int. Ed. 2005, 44, 7427.

J. Phys. Chem. B, Vol. 113, No. 49, 2009 16027 (9) Lion-Dagan, M.; Katz, E.; Willner, I. J. Am. Chem. Soc. 1994, 116, 7913. (10) Riskin, M.; Basnar, B.; Huang, Y.; Willner, I. AdV. Mater. 2007, 19, 2691. (11) Bartlett, P. N.; Birkin, P. R.; Wang, J. H. Anal. Chem. 1998, 70, 3685. (12) Decher, G. Science 1997, 277, 1232. (13) Lvov, Y.; Mo¨hwald, H. Protein Architecture: Interfacing Molecular Assemblies and Immobilization Biotechnology; Marcel Dekker: New York, 2000. (14) Decher, G.; Schneloff, J. B. Multilayer Thin Films: Sequential Assembly of Nanocomposite Materials; Wiley-VCH: Weinheim, 2003. (15) Becker, J. W.; Reeke, G. N.; Wang, J. L.; Cunningham, B. A.; Edelman, G. M. J. Biol. Chem. 1975, 250, 1513. (16) Welch, K. T.; Turner, T. A.; Preast, C. E. Bioorg. Med. Chem. Lett. 2008, 18, 6573. (17) Anzai, J. I.; Kobayashi, Y. Langmuir 2000, 16, 2851. (18) Doyle, R. J.; Woodside, E. E.; Fishel, C. W. Biochem. J. 1968, 106, 35. (19) Ballerstadt, R.; Schultz, J. S. Sens. Actuators B 1998, 46, 50. (20) Oh, J. K.; Siegwart, D. J.; Matyjaszewski, K. Biomacromolecules 2007, 8, 3326. (21) Chinnayelka, S.; McShane, M. J. Biomacromolecules 2004, 5, 1657. (22) Zhang, J.; Roll, D.; Geddes, C. D.; Lakowicz, J. R. J. Phys. Chem. B 2004, 108, 12210. (23) Chinnayelka, S.; McShane, M. J. J. Fluorescence 2004, 14, 585. (24) Yang, X.; Li, Z.; Wang, Q.; Wang, K. Acta Chim. Sinica 2007, 65, 1185. (25) Yao, H.; Guo, X.; Hu, N. Electrochim. Acta 2009, 54, 7330. (26) Pita, M.; Katz, E. Electroanalysis 2009, 21, 252. (27) Zhou, J.; Liu, J.; Wang, G.; Lu, X.; Wen, Z.; Li, J. AdV. Funct. Mater. 2007, 17, 3377. (28) Kang, M. S.; Martin, C. R. Langmuir 2001, 17, 2753. (29) Liu, Y.; Zhao, M.; Bergbreiter, D. E.; Crooks, R. M. J. Am. Chem. Soc. 1997, 119, 8720. (30) Zhao, M.; Liu, Y.; Crooks, R. M.; Bergbreiter, D. E. J. Am. Chem. Soc. 1999, 121, 923. (31) Park, M.-K.; Deng, S.; Advincula, R. C. J. Am. Chem. Soc. 2004, 126, 13723. (32) Kang, E.; Liu, X.; Sun, J.; Shen, J. Langmuir 2006, 22, 7894. (33) Lvov, Y.; Ariga, K.; Ichinose, I.; Kunitake, T. Thin Solid Films 1996, 284/285, 797. (34) Denuziere, A.; Ferrier, D.; Domard, A. Carbohydr. Polym. 1996, 29, 317. (35) Hill, H. A. O. Pure Appl. Chem. 1987, 59, 743. (36) Katz, E.; Willner, I. Electroanalysis 2003, 15, 913. (37) Sabatani, E.; Rubinstein, I. J. Phys. Chem. 1987, 91, 6663. (38) Sabatani, E.; Cohen-Boulakia, J.; Bruening, M.; Rubinstein, I. Langmuir 1993, 9, 2974. (39) Randles, J. E. B. Discuss. Faraday Soc. 1947, 1, 11. (40) Pan, S.; Arnold, M. A. Anal. Chim. Acta 1993, 283, 663. (41) Li, W.; Yuan, R.; Chai, Y.; Zhou, L.; Chen, S.; Li, N. J. Biochem. Biophys. Methods 2008, 70, 830. (42) Lu, B.; Smyth, M. R.; Quinn, J.; Bogan, D.; O’Kennedy, R. Electroanalysis 1996, 8, 619. (43) Adeiran, S. A.; Lambeir, A. M. Eur. J. Biochem. 1989, 186, 571. (44) Hasinoff, B. B.; Dunford, H. B. Biochemistry 1970, 9, 4930. (45) Wang, J. Electroanalysis 2001, 13, 983. (46) Pumera, M.; Sanchez, S.; Ichinose, I.; Tang, J. Sens. Actuators B 2007, 123, 1195. (47) Wang, J. Talanta 2008, 75, 636. (48) Cass, A. E. G.; Davis, G.; Francis, G. D.; Hill, H. A. O.; Aston, W. J.; Higgins, I. J.; Plotkin, E. O.; Scott, L. D. L.; Turner, A. P. F. Anal. Chem. 1984, 56, 667. (49) Badia, A.; Carlini, R.; Fernandez, A.; Battaglini, F.; Mikkelsen, S. R.; English, A. M. J. Am. Chem. Soc. 1993, 115, 7053. (50) Bu, H.-z.; Mikkelsen, S. R.; English, A. M. Anal. Chem. 1995, 67, 4071. (51) Benkeser, R.; Goggin, D.; Schroll, G. J. Am. Chem. Soc. 1954, 76, 4025. (52) Carraher, C. E., Jr. J. Inorg. Organomet. Polym. Mater. 2005, 15, 121. (53) Noci, S. D.; Frasconi, M.; Favero, G.; Tosi, M.; Ferri, T.; Mazzei, F. Electroanalysis 2008, 20, 163. (54) Murthy, A. S. N.; Sharma, J. Anal. Chim. Acta 1998, 363, 215.

JP906345S